From 3a52b65795f7a81f6f0ae48d96e235388fc86b87 Mon Sep 17 00:00:00 2001
From: Lorenzo Ampil <lorenzojulioampil@gmail.com>
Date: Mon, 21 Oct 2019 12:55:51 +0800
Subject: [PATCH 001/505] Add special tokens to documentation for bert examples
 to resolve issue: #1561

---
 transformers/modeling_bert.py | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/transformers/modeling_bert.py b/transformers/modeling_bert.py
index 8c92241fa2..9a9cd31b4b 100644
--- a/transformers/modeling_bert.py
+++ b/transformers/modeling_bert.py
@@ -557,7 +557,7 @@ class BertModel(BertPreTrainedModel):
 
         tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
         model = BertModel.from_pretrained('bert-base-uncased')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
@@ -667,7 +667,7 @@ class BertForPreTraining(BertPreTrainedModel):
 
         tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
         model = BertForPreTraining.from_pretrained('bert-base-uncased')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids)
         prediction_scores, seq_relationship_scores = outputs[:2]
 
@@ -739,7 +739,7 @@ class BertForMaskedLM(BertPreTrainedModel):
 
         tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
         model = BertForMaskedLM.from_pretrained('bert-base-uncased')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids, masked_lm_labels=input_ids)
         loss, prediction_scores = outputs[:2]
 
@@ -808,7 +808,7 @@ class BertForNextSentencePrediction(BertPreTrainedModel):
 
         tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
         model = BertForNextSentencePrediction.from_pretrained('bert-base-uncased')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids)
         seq_relationship_scores = outputs[0]
 
@@ -871,7 +871,7 @@ class BertForSequenceClassification(BertPreTrainedModel):
 
         tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
         model = BertForSequenceClassification.from_pretrained('bert-base-uncased')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         labels = torch.tensor([1]).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids, labels=labels)
         loss, logits = outputs[:2]
@@ -945,7 +945,7 @@ class BertForMultipleChoice(BertPreTrainedModel):
         tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
         model = BertForMultipleChoice.from_pretrained('bert-base-uncased')
         choices = ["Hello, my dog is cute", "Hello, my cat is amazing"]
-        input_ids = torch.tensor([tokenizer.encode(s) for s in choices]).unsqueeze(0)  # Batch size 1, 2 choices
+        input_ids = torch.tensor([tokenizer.encode(s, add_special_tokens=True) for s in choices]).unsqueeze(0)  # Batch size 1, 2 choices
         labels = torch.tensor(1).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids, labels=labels)
         loss, classification_scores = outputs[:2]
@@ -1017,7 +1017,7 @@ class BertForTokenClassification(BertPreTrainedModel):
 
         tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
         model = BertForTokenClassification.from_pretrained('bert-base-uncased')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         labels = torch.tensor([1] * input_ids.size(1)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids, labels=labels)
         loss, scores = outputs[:2]

From 6e011690a980c3c3f69fdfc3af8705859250cc6b Mon Sep 17 00:00:00 2001
From: Lorenzo Ampil <lorenzojulioampil@gmail.com>
Date: Sun, 27 Oct 2019 13:59:14 +0800
Subject: [PATCH 002/505] Add special tokens to documentation for the rest of
 pytorch model examples #1561

---
 transformers/modeling_ctrl.py       |  4 ++--
 transformers/modeling_distilbert.py |  8 ++++----
 transformers/modeling_gpt2.py       |  4 ++--
 transformers/modeling_openai.py     |  4 ++--
 transformers/modeling_roberta.py    |  6 +++---
 transformers/modeling_transfo_xl.py |  4 ++--
 transformers/modeling_xlm.py        | 10 +++++-----
 transformers/modeling_xlnet.py      | 10 +++++-----
 8 files changed, 25 insertions(+), 25 deletions(-)

diff --git a/transformers/modeling_ctrl.py b/transformers/modeling_ctrl.py
index 55e64d318b..b31755e13d 100644
--- a/transformers/modeling_ctrl.py
+++ b/transformers/modeling_ctrl.py
@@ -261,7 +261,7 @@ class CTRLModel(CTRLPreTrainedModel):
 
         tokenizer = CTRLTokenizer.from_pretrained('ctrl')
         model = CTRLModel.from_pretrained('ctrl')
-        input_ids = torch.tensor(tokenizer.encode("Links Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Links Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
@@ -438,7 +438,7 @@ class CTRLLMHeadModel(CTRLPreTrainedModel):
         tokenizer = CTRLTokenizer.from_pretrained('ctrl')
         model = CTRLLMHeadModel.from_pretrained('ctrl')
 
-        input_ids = torch.tensor(tokenizer.encode("Links Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Links Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids, labels=input_ids)
         loss, logits = outputs[:2]
 
diff --git a/transformers/modeling_distilbert.py b/transformers/modeling_distilbert.py
index d3b4ccff5d..990d76e378 100644
--- a/transformers/modeling_distilbert.py
+++ b/transformers/modeling_distilbert.py
@@ -411,7 +411,7 @@ class DistilBertModel(DistilBertPreTrainedModel):
 
         tokenizer = DistilBertTokenizer.from_pretrained('distilbert-base-uncased')
         model = DistilBertModel.from_pretrained('distilbert-base-uncased')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
@@ -495,7 +495,7 @@ class DistilBertForMaskedLM(DistilBertPreTrainedModel):
 
         tokenizer = DistilBertTokenizer.from_pretrained('distilbert-base-uncased')
         model = DistilBertForMaskedLM.from_pretrained('distilbert-base-uncased')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids, masked_lm_labels=input_ids)
         loss, prediction_scores = outputs[:2]
 
@@ -569,7 +569,7 @@ class DistilBertForSequenceClassification(DistilBertPreTrainedModel):
 
         tokenizer = DistilBertTokenizer.from_pretrained('distilbert-base-uncased')
         model = DistilBertForSequenceClassification.from_pretrained('distilbert-base-uncased')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         labels = torch.tensor([1]).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids, labels=labels)
         loss, logits = outputs[:2]
@@ -643,7 +643,7 @@ class DistilBertForQuestionAnswering(DistilBertPreTrainedModel):
 
         tokenizer = DistilBertTokenizer.from_pretrained('distilbert-base-uncased')
         model = DistilBertForQuestionAnswering.from_pretrained('distilbert-base-uncased')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         start_positions = torch.tensor([1])
         end_positions = torch.tensor([3])
         outputs = model(input_ids, start_positions=start_positions, end_positions=end_positions)
diff --git a/transformers/modeling_gpt2.py b/transformers/modeling_gpt2.py
index 0b5b83aa75..87878abb4e 100644
--- a/transformers/modeling_gpt2.py
+++ b/transformers/modeling_gpt2.py
@@ -338,7 +338,7 @@ class GPT2Model(GPT2PreTrainedModel):
 
         tokenizer = GPT2Tokenizer.from_pretrained('gpt2')
         model = GPT2Model.from_pretrained('gpt2')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
@@ -503,7 +503,7 @@ class GPT2LMHeadModel(GPT2PreTrainedModel):
         tokenizer = GPT2Tokenizer.from_pretrained('gpt2')
         model = GPT2LMHeadModel.from_pretrained('gpt2')
 
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids, labels=input_ids)
         loss, logits = outputs[:2]
 
diff --git a/transformers/modeling_openai.py b/transformers/modeling_openai.py
index 52f3b7db72..c6b13dee4e 100644
--- a/transformers/modeling_openai.py
+++ b/transformers/modeling_openai.py
@@ -343,7 +343,7 @@ class OpenAIGPTModel(OpenAIGPTPreTrainedModel):
 
         tokenizer = OpenAIGPTTokenizer.from_pretrained('openai-gpt')
         model = OpenAIGPTModel.from_pretrained('openai-gpt')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
@@ -478,7 +478,7 @@ class OpenAIGPTLMHeadModel(OpenAIGPTPreTrainedModel):
 
         tokenizer = OpenAIGPTTokenizer.from_pretrained('openai-gpt')
         model = OpenAIGPTLMHeadModel.from_pretrained('openai-gpt')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids, labels=input_ids)
         loss, logits = outputs[:2]
 
diff --git a/transformers/modeling_roberta.py b/transformers/modeling_roberta.py
index eb340dc7fb..cbd2e0106d 100644
--- a/transformers/modeling_roberta.py
+++ b/transformers/modeling_roberta.py
@@ -154,7 +154,7 @@ class RobertaModel(BertModel):
 
         tokenizer = RobertaTokenizer.from_pretrained('roberta-base')
         model = RobertaModel.from_pretrained('roberta-base')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
@@ -209,7 +209,7 @@ class RobertaForMaskedLM(BertPreTrainedModel):
 
         tokenizer = RobertaTokenizer.from_pretrained('roberta-base')
         model = RobertaForMaskedLM.from_pretrained('roberta-base')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids, masked_lm_labels=input_ids)
         loss, prediction_scores = outputs[:2]
 
@@ -303,7 +303,7 @@ class RobertaForSequenceClassification(BertPreTrainedModel):
 
         tokenizer = RobertaTokenizer.from_pretrained('roberta-base')
         model = RobertaForSequenceClassification.from_pretrained('roberta-base')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         labels = torch.tensor([1]).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids, labels=labels)
         loss, logits = outputs[:2]
diff --git a/transformers/modeling_transfo_xl.py b/transformers/modeling_transfo_xl.py
index 6d430e1804..ad1c7bdea4 100644
--- a/transformers/modeling_transfo_xl.py
+++ b/transformers/modeling_transfo_xl.py
@@ -578,7 +578,7 @@ class TransfoXLModel(TransfoXLPreTrainedModel):
 
         tokenizer = TransfoXLTokenizer.from_pretrained('transfo-xl-wt103')
         model = TransfoXLModel.from_pretrained('transfo-xl-wt103')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states, mems = outputs[:2]
 
@@ -808,7 +808,7 @@ class TransfoXLLMHeadModel(TransfoXLPreTrainedModel):
 
         tokenizer = TransfoXLTokenizer.from_pretrained('transfo-xl-wt103')
         model = TransfoXLLMHeadModel.from_pretrained('transfo-xl-wt103')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids)
         prediction_scores, mems = outputs[:2]
 
diff --git a/transformers/modeling_xlm.py b/transformers/modeling_xlm.py
index b29e721556..a7c8f4e941 100644
--- a/transformers/modeling_xlm.py
+++ b/transformers/modeling_xlm.py
@@ -332,7 +332,7 @@ class XLMModel(XLMPreTrainedModel):
 
         tokenizer = XLMTokenizer.from_pretrained('xlm-mlm-en-2048')
         model = XLMModel.from_pretrained('xlm-mlm-en-2048')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
@@ -607,7 +607,7 @@ class XLMWithLMHeadModel(XLMPreTrainedModel):
 
         tokenizer = XLMTokenizer.from_pretrained('xlm-mlm-en-2048')
         model = XLMWithLMHeadModel.from_pretrained('xlm-mlm-en-2048')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
@@ -671,7 +671,7 @@ class XLMForSequenceClassification(XLMPreTrainedModel):
 
         tokenizer = XLMTokenizer.from_pretrained('xlm-mlm-en-2048')
         model = XLMForSequenceClassification.from_pretrained('xlm-mlm-en-2048')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         labels = torch.tensor([1]).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids, labels=labels)
         loss, logits = outputs[:2]
@@ -754,7 +754,7 @@ class XLMForQuestionAnsweringSimple(XLMPreTrainedModel):
 
         tokenizer = XLMTokenizer.from_pretrained('xlm-mlm-en-2048')
         model = XLMForQuestionAnsweringSimple.from_pretrained('xlm-mlm-en-2048')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         start_positions = torch.tensor([1])
         end_positions = torch.tensor([3])
         outputs = model(input_ids, start_positions=start_positions, end_positions=end_positions)
@@ -849,7 +849,7 @@ class XLMForQuestionAnswering(XLMPreTrainedModel):
 
         tokenizer = XLMTokenizer.from_pretrained('xlm-mlm-en-2048')
         model = XLMForQuestionAnswering.from_pretrained('xlm-mlm-en-2048')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         start_positions = torch.tensor([1])
         end_positions = torch.tensor([3])
         outputs = model(input_ids, start_positions=start_positions, end_positions=end_positions)
diff --git a/transformers/modeling_xlnet.py b/transformers/modeling_xlnet.py
index e191ebadd0..fab405fd2b 100644
--- a/transformers/modeling_xlnet.py
+++ b/transformers/modeling_xlnet.py
@@ -584,7 +584,7 @@ class XLNetModel(XLNetPreTrainedModel):
 
         tokenizer = XLNetTokenizer.from_pretrained('xlnet-large-cased')
         model = XLNetModel.from_pretrained('xlnet-large-cased')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
@@ -900,7 +900,7 @@ class XLNetLMHeadModel(XLNetPreTrainedModel):
         tokenizer = XLNetTokenizer.from_pretrained('xlnet-large-cased')
         model = XLNetLMHeadModel.from_pretrained('xlnet-large-cased')
         # We show how to setup inputs to predict a next token using a bi-directional context.
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is very <mask>")).unsqueeze(0)  # We will predict the masked token
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is very <mask>", add_special_tokens=True)).unsqueeze(0)  # We will predict the masked token
         perm_mask = torch.zeros((1, input_ids.shape[1], input_ids.shape[1]), dtype=torch.float)
         perm_mask[:, :, -1] = 1.0  # Previous tokens don't see last token
         target_mapping = torch.zeros((1, 1, input_ids.shape[1]), dtype=torch.float)  # Shape [1, 1, seq_length] => let's predict one token
@@ -983,7 +983,7 @@ class XLNetForSequenceClassification(XLNetPreTrainedModel):
 
         tokenizer = XLNetTokenizer.from_pretrained('xlnet-large-cased')
         model = XLNetForSequenceClassification.from_pretrained('xlnet-large-cased')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         labels = torch.tensor([1]).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids, labels=labels)
         loss, logits = outputs[:2]
@@ -1163,7 +1163,7 @@ class XLNetForQuestionAnsweringSimple(XLNetPreTrainedModel):
 
         tokenizer = XLMTokenizer.from_pretrained('xlm-mlm-en-2048')
         model = XLMForQuestionAnswering.from_pretrained('xlnet-large-cased')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         start_positions = torch.tensor([1])
         end_positions = torch.tensor([3])
         outputs = model(input_ids, start_positions=start_positions, end_positions=end_positions)
@@ -1276,7 +1276,7 @@ class XLNetForQuestionAnswering(XLNetPreTrainedModel):
 
         tokenizer =  XLNetTokenizer.from_pretrained('xlnet-large-cased')
         model = XLMForQuestionAnswering.from_pretrained('xlnet-large-cased')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         start_positions = torch.tensor([1])
         end_positions = torch.tensor([3])
         outputs = model(input_ids, start_positions=start_positions, end_positions=end_positions)

From ec276d6abad7eae800f1a1a039ddc78fde406009 Mon Sep 17 00:00:00 2001
From: Lorenzo Ampil <lorenzojulioampil@gmail.com>
Date: Sun, 27 Oct 2019 14:00:40 +0800
Subject: [PATCH 003/505] Add special tokens to documentation for the
 tensorflow model examples #1561

---
 transformers/modeling_tf_bert.py       | 14 +++++++-------
 transformers/modeling_tf_ctrl.py       |  4 ++--
 transformers/modeling_tf_distilbert.py |  8 ++++----
 transformers/modeling_tf_gpt2.py       |  4 ++--
 transformers/modeling_tf_openai.py     |  4 ++--
 transformers/modeling_tf_roberta.py    |  6 +++---
 transformers/modeling_tf_transfo_xl.py |  4 ++--
 transformers/modeling_tf_xlm.py        |  8 ++++----
 transformers/modeling_tf_xlnet.py      | 10 +++++-----
 9 files changed, 31 insertions(+), 31 deletions(-)

diff --git a/transformers/modeling_tf_bert.py b/transformers/modeling_tf_bert.py
index afe9b2946b..d2d3c7be37 100644
--- a/transformers/modeling_tf_bert.py
+++ b/transformers/modeling_tf_bert.py
@@ -647,7 +647,7 @@ class TFBertModel(TFBertPreTrainedModel):
 
         tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
         model = TFBertModel.from_pretrained('bert-base-uncased')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
@@ -686,7 +686,7 @@ class TFBertForPreTraining(TFBertPreTrainedModel):
 
         tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
         model = TFBertForPreTraining.from_pretrained('bert-base-uncased')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         prediction_scores, seq_relationship_scores = outputs[:2]
 
@@ -732,7 +732,7 @@ class TFBertForMaskedLM(TFBertPreTrainedModel):
 
         tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
         model = TFBertForMaskedLM.from_pretrained('bert-base-uncased')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         prediction_scores = outputs[0]
 
@@ -776,7 +776,7 @@ class TFBertForNextSentencePrediction(TFBertPreTrainedModel):
 
         tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
         model = TFBertForNextSentencePrediction.from_pretrained('bert-base-uncased')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         seq_relationship_scores = outputs[0]
 
@@ -821,7 +821,7 @@ class TFBertForSequenceClassification(TFBertPreTrainedModel):
 
         tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
         model = TFBertForSequenceClassification.from_pretrained('bert-base-uncased')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         logits = outputs[0]
 
@@ -952,7 +952,7 @@ class TFBertForTokenClassification(TFBertPreTrainedModel):
 
         tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
         model = TFBertForTokenClassification.from_pretrained('bert-base-uncased')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         scores = outputs[0]
 
@@ -1005,7 +1005,7 @@ class TFBertForQuestionAnswering(TFBertPreTrainedModel):
 
         tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
         model = TFBertForQuestionAnswering.from_pretrained('bert-base-uncased')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         start_scores, end_scores = outputs[:2]
 
diff --git a/transformers/modeling_tf_ctrl.py b/transformers/modeling_tf_ctrl.py
index c8d181548b..66766a066e 100644
--- a/transformers/modeling_tf_ctrl.py
+++ b/transformers/modeling_tf_ctrl.py
@@ -402,7 +402,7 @@ class TFCTRLModel(TFCTRLPreTrainedModel):
 
         tokenizer = CTRLTokenizer.from_pretrained('ctrl')
         model = TFCTRLModel.from_pretrained('ctrl')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
@@ -465,7 +465,7 @@ class TFCTRLLMHeadModel(TFCTRLPreTrainedModel):
         tokenizer = CTRLTokenizer.from_pretrained('ctrl')
         model = TFCTRLLMHeadModel.from_pretrained('ctrl')
 
-        input_ids = torch.tensor(tokenizer.encode("Links Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Links Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids, labels=input_ids)
         loss, logits = outputs[:2]
 
diff --git a/transformers/modeling_tf_distilbert.py b/transformers/modeling_tf_distilbert.py
index 188394816e..c2d0f73999 100644
--- a/transformers/modeling_tf_distilbert.py
+++ b/transformers/modeling_tf_distilbert.py
@@ -532,7 +532,7 @@ class TFDistilBertModel(TFDistilBertPreTrainedModel):
 
         tokenizer = DistilBertTokenizer.from_pretrained('distilbert-base-uncased')
         model = TFDistilBertModel.from_pretrained('distilbert-base-uncased')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
@@ -590,7 +590,7 @@ class TFDistilBertForMaskedLM(TFDistilBertPreTrainedModel):
 
         tokenizer = DistilBertTokenizer.from_pretrained('distilbert-base-uncased')
         model = TFDistilBertForMaskedLM.from_pretrained('distilbert-base-uncased')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         prediction_scores = outputs[0]
 
@@ -645,7 +645,7 @@ class TFDistilBertForSequenceClassification(TFDistilBertPreTrainedModel):
 
         tokenizer = DistilBertTokenizer.from_pretrained('distilbert-base-uncased')
         model = TFDistilBertForSequenceClassification.from_pretrained('distilbert-base-uncased')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         logits = outputs[0]
 
@@ -702,7 +702,7 @@ class TFDistilBertForQuestionAnswering(TFDistilBertPreTrainedModel):
 
         tokenizer = DistilBertTokenizer.from_pretrained('distilbert-base-uncased')
         model = TFDistilBertForQuestionAnswering.from_pretrained('distilbert-base-uncased')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         start_scores, end_scores = outputs[:2]
 
diff --git a/transformers/modeling_tf_gpt2.py b/transformers/modeling_tf_gpt2.py
index 4188b273ba..24f2857b80 100644
--- a/transformers/modeling_tf_gpt2.py
+++ b/transformers/modeling_tf_gpt2.py
@@ -436,7 +436,7 @@ class TFGPT2Model(TFGPT2PreTrainedModel):
 
         tokenizer = GPT2Tokenizer.from_pretrained('gpt2')
         model = TFGPT2Model.from_pretrained('gpt2')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
@@ -477,7 +477,7 @@ class TFGPT2LMHeadModel(TFGPT2PreTrainedModel):
         tokenizer = GPT2Tokenizer.from_pretrained('gpt2')
         model = TFGPT2LMHeadModel.from_pretrained('gpt2')
 
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         logits = outputs[0]
 
diff --git a/transformers/modeling_tf_openai.py b/transformers/modeling_tf_openai.py
index 747c5171fd..08034b2d2e 100644
--- a/transformers/modeling_tf_openai.py
+++ b/transformers/modeling_tf_openai.py
@@ -413,7 +413,7 @@ class TFOpenAIGPTModel(TFOpenAIGPTPreTrainedModel):
 
         tokenizer = OpenAIGPTTokenizer.from_pretrained('openai-gpt')
         model = TFOpenAIGPTModel.from_pretrained('openai-gpt')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
@@ -449,7 +449,7 @@ class TFOpenAIGPTLMHeadModel(TFOpenAIGPTPreTrainedModel):
 
         tokenizer = OpenAIGPTTokenizer.from_pretrained('openai-gpt')
         model = TFOpenAIGPTLMHeadModel.from_pretrained('openai-gpt')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         logits = outputs[0]
 
diff --git a/transformers/modeling_tf_roberta.py b/transformers/modeling_tf_roberta.py
index 244c83f2b3..dcf00c3add 100644
--- a/transformers/modeling_tf_roberta.py
+++ b/transformers/modeling_tf_roberta.py
@@ -204,7 +204,7 @@ class TFRobertaModel(TFRobertaPreTrainedModel):
 
         tokenizer = RobertaTokenizer.from_pretrained('roberta-base')
         model = TFRobertaModel.from_pretrained('roberta-base')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
@@ -281,7 +281,7 @@ class TFRobertaForMaskedLM(TFRobertaPreTrainedModel):
 
         tokenizer = RobertaTokenizer.from_pretrained('roberta-base')
         model = TFRobertaForMaskedLM.from_pretrained('roberta-base')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids, masked_lm_labels=input_ids)
         prediction_scores = outputs[0]
 
@@ -349,7 +349,7 @@ class TFRobertaForSequenceClassification(TFRobertaPreTrainedModel):
 
         tokenizer = RoertaTokenizer.from_pretrained('roberta-base')
         model = TFRobertaForSequenceClassification.from_pretrained('roberta-base')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         labels = tf.constant([1])[None, :]  # Batch size 1
         outputs = model(input_ids)
         logits = outputs[0]
diff --git a/transformers/modeling_tf_transfo_xl.py b/transformers/modeling_tf_transfo_xl.py
index a3e403ce06..87863163f0 100644
--- a/transformers/modeling_tf_transfo_xl.py
+++ b/transformers/modeling_tf_transfo_xl.py
@@ -654,7 +654,7 @@ class TFTransfoXLModel(TFTransfoXLPreTrainedModel):
 
         tokenizer = TransfoXLTokenizer.from_pretrained('transfo-xl-wt103')
         model = TFTransfoXLModel.from_pretrained('transfo-xl-wt103')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states, mems = outputs[:2]
 
@@ -696,7 +696,7 @@ class TFTransfoXLLMHeadModel(TFTransfoXLPreTrainedModel):
 
         tokenizer = TransfoXLTokenizer.from_pretrained('transfo-xl-wt103')
         model = TFTransfoXLLMHeadModel.from_pretrained('transfo-xl-wt103')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         prediction_scores, mems = outputs[:2]
 
diff --git a/transformers/modeling_tf_xlm.py b/transformers/modeling_tf_xlm.py
index 84de1517ee..7b7305f22e 100644
--- a/transformers/modeling_tf_xlm.py
+++ b/transformers/modeling_tf_xlm.py
@@ -550,7 +550,7 @@ class TFXLMModel(TFXLMPreTrainedModel):
 
         tokenizer = XLMTokenizer.from_pretrained('xlm-mlm-en-2048')
         model = TFXLMModel.from_pretrained('xlm-mlm-en-2048')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
@@ -623,7 +623,7 @@ class TFXLMWithLMHeadModel(TFXLMPreTrainedModel):
 
         tokenizer = XLMTokenizer.from_pretrained('xlm-mlm-en-2048')
         model = TFXLMWithLMHeadModel.from_pretrained('xlm-mlm-en-2048')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
@@ -667,7 +667,7 @@ class TFXLMForSequenceClassification(TFXLMPreTrainedModel):
 
         tokenizer = XLMTokenizer.from_pretrained('xlm-mlm-en-2048')
         model = TFXLMForSequenceClassification.from_pretrained('xlm-mlm-en-2048')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         labels = tf.constant([1])[None, :]  # Batch size 1
         outputs = model(input_ids)
         logits = outputs[0]
@@ -715,7 +715,7 @@ class TFXLMForQuestionAnsweringSimple(TFXLMPreTrainedModel):
 
         tokenizer = XLMTokenizer.from_pretrained('xlm-mlm-en-2048')
         model = TFXLMForQuestionAnsweringSimple.from_pretrained('xlm-mlm-en-2048')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         start_scores, end_scores = outputs[:2]
 
diff --git a/transformers/modeling_tf_xlnet.py b/transformers/modeling_tf_xlnet.py
index 8a25be78c1..d2029db485 100644
--- a/transformers/modeling_tf_xlnet.py
+++ b/transformers/modeling_tf_xlnet.py
@@ -791,7 +791,7 @@ class TFXLNetModel(TFXLNetPreTrainedModel):
 
         tokenizer = XLNetTokenizer.from_pretrained('xlnet-large-cased')
         model = TFXLNetModel.from_pretrained('xlnet-large-cased')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
@@ -835,7 +835,7 @@ class TFXLNetLMHeadModel(TFXLNetPreTrainedModel):
         model = TFXLNetLMHeadModel.from_pretrained('xlnet-large-cased')
 
         # We show how to setup inputs to predict a next token using a bi-directional context.
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is very <mask>"))[None, :]  # We will predict the masked token
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is very <mask>", add_special_tokens=True))[None, :]  # We will predict the masked token
         perm_mask = tf.zeros((1, input_ids.shape[1], input_ids.shape[1]))
         perm_mask[:, :, -1] = 1.0  # Previous tokens don't see last token
         target_mapping = tf.zeros((1, 1, input_ids.shape[1]))  # Shape [1, 1, seq_length] => let's predict one token
@@ -888,7 +888,7 @@ class TFXLNetForSequenceClassification(TFXLNetPreTrainedModel):
 
         tokenizer = XLNetTokenizer.from_pretrained('xlnet-large-cased')
         model = TFXLNetForSequenceClassification.from_pretrained('xlnet-large-cased')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         logits = outputs[0]
 
@@ -946,7 +946,7 @@ class TFXLNetForQuestionAnsweringSimple(TFXLNetPreTrainedModel):
 
         tokenizer = XLNetTokenizer.from_pretrained('xlnet-base-cased')
         model = TFXLNetForQuestionAnsweringSimple.from_pretrained('xlnet-base-cased')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
         outputs = model(input_ids)
         start_scores, end_scores = outputs[:2]
 
@@ -1010,7 +1010,7 @@ class TFXLNetForQuestionAnsweringSimple(TFXLNetPreTrainedModel):
 
 #         tokenizer = XLMTokenizer.from_pretrained('xlm-mlm-en-2048')
 #         model = XLMForQuestionAnswering.from_pretrained('xlnet-large-cased')
-#         input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+#         input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
 #         start_positions = tf.constant([1])
 #         end_positions = tf.constant([3])
 #         outputs = model(input_ids, start_positions=start_positions, end_positions=end_positions)

From d36680df546a9d4a20c58c0eab2b14ee054436ca Mon Sep 17 00:00:00 2001
From: Lorenzo Ampil <lorenzojulioampil@gmail.com>
Date: Sun, 27 Oct 2019 14:51:36 +0800
Subject: [PATCH 004/505] Rever changes to TF distilbert due to failed test:
 TFDistilBertModelTest.test_pt_tf_model_equivalence

---
 transformers/modeling_tf_distilbert.py | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/transformers/modeling_tf_distilbert.py b/transformers/modeling_tf_distilbert.py
index c2d0f73999..188394816e 100644
--- a/transformers/modeling_tf_distilbert.py
+++ b/transformers/modeling_tf_distilbert.py
@@ -532,7 +532,7 @@ class TFDistilBertModel(TFDistilBertPreTrainedModel):
 
         tokenizer = DistilBertTokenizer.from_pretrained('distilbert-base-uncased')
         model = TFDistilBertModel.from_pretrained('distilbert-base-uncased')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
@@ -590,7 +590,7 @@ class TFDistilBertForMaskedLM(TFDistilBertPreTrainedModel):
 
         tokenizer = DistilBertTokenizer.from_pretrained('distilbert-base-uncased')
         model = TFDistilBertForMaskedLM.from_pretrained('distilbert-base-uncased')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
         outputs = model(input_ids)
         prediction_scores = outputs[0]
 
@@ -645,7 +645,7 @@ class TFDistilBertForSequenceClassification(TFDistilBertPreTrainedModel):
 
         tokenizer = DistilBertTokenizer.from_pretrained('distilbert-base-uncased')
         model = TFDistilBertForSequenceClassification.from_pretrained('distilbert-base-uncased')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
         outputs = model(input_ids)
         logits = outputs[0]
 
@@ -702,7 +702,7 @@ class TFDistilBertForQuestionAnswering(TFDistilBertPreTrainedModel):
 
         tokenizer = DistilBertTokenizer.from_pretrained('distilbert-base-uncased')
         model = TFDistilBertForQuestionAnswering.from_pretrained('distilbert-base-uncased')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
         outputs = model(input_ids)
         start_scores, end_scores = outputs[:2]
 

From 0e4cc050d63058969c53cce295dea99e90483c01 Mon Sep 17 00:00:00 2001
From: Sergey Mironov <mironov.sergey@huawei.com>
Date: Thu, 24 Oct 2019 18:15:55 +0300
Subject: [PATCH 005/505] Add support for resumable downloads for HTTP
 protocol.

---
 transformers/configuration_auto.py  |  3 ++
 transformers/configuration_utils.py |  7 ++++-
 transformers/file_utils.py          | 47 ++++++++++++++++++++++-------
 transformers/modeling_auto.py       |  8 +++++
 transformers/modeling_tf_auto.py    | 12 ++++++++
 transformers/modeling_tf_utils.py   |  8 ++++-
 transformers/modeling_utils.py      |  8 ++++-
 transformers/tokenization_auto.py   |  3 ++
 transformers/tokenization_utils.py  |  6 +++-
 9 files changed, 87 insertions(+), 15 deletions(-)

diff --git a/transformers/configuration_auto.py b/transformers/configuration_auto.py
index edd21a670c..0e8eb03d03 100644
--- a/transformers/configuration_auto.py
+++ b/transformers/configuration_auto.py
@@ -92,6 +92,9 @@ class AutoConfig(object):
             force_download: (`optional`) boolean, default False:
                 Force to (re-)download the model weights and configuration files and override the cached versions if they exists.
 
+            resume_download: (`optional`) boolean, default False:
+                Do not delete incompletely recieved file. Attempt to resume the download if such a file exists.
+
             proxies: (`optional`) dict, default None:
                 A dictionary of proxy servers to use by protocol or endpoint, e.g.: {'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.
                 The proxies are used on each request.
diff --git a/transformers/configuration_utils.py b/transformers/configuration_utils.py
index cfa6502bcd..0f9d609b20 100644
--- a/transformers/configuration_utils.py
+++ b/transformers/configuration_utils.py
@@ -93,6 +93,9 @@ class PretrainedConfig(object):
             force_download: (`optional`) boolean, default False:
                 Force to (re-)download the model weights and configuration files and override the cached versions if they exists.
 
+            resume_download: (`optional`) boolean, default False:
+                Do not delete incompletely recieved file. Attempt to resume the download if such a file exists.
+
             proxies: (`optional`) dict, default None:
                 A dictionary of proxy servers to use by protocol or endpoint, e.g.: {'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.
                 The proxies are used on each request.
@@ -119,6 +122,7 @@ class PretrainedConfig(object):
         """
         cache_dir = kwargs.pop('cache_dir', None)
         force_download = kwargs.pop('force_download', False)
+        resume_download = kwargs.pop('resume_download', False)
         proxies = kwargs.pop('proxies', None)
         return_unused_kwargs = kwargs.pop('return_unused_kwargs', False)
 
@@ -130,7 +134,8 @@ class PretrainedConfig(object):
             config_file = pretrained_model_name_or_path
         # redirect to the cache, if necessary
         try:
-            resolved_config_file = cached_path(config_file, cache_dir=cache_dir, force_download=force_download, proxies=proxies)
+            resolved_config_file = cached_path(config_file, cache_dir=cache_dir, force_download=force_download,
+                                               proxies=proxies, resume_download=resume_download)
         except EnvironmentError:
             if pretrained_model_name_or_path in cls.pretrained_config_archive_map:
                 msg = "Couldn't reach server at '{}' to download pretrained model configuration file.".format(
diff --git a/transformers/file_utils.py b/transformers/file_utils.py
index 27875212ff..24abd60781 100644
--- a/transformers/file_utils.py
+++ b/transformers/file_utils.py
@@ -22,6 +22,7 @@ from botocore.config import Config
 from botocore.exceptions import ClientError
 import requests
 from tqdm import tqdm
+from contextlib import contextmanager
 
 logger = logging.getLogger(__name__)  # pylint: disable=invalid-name
 
@@ -152,7 +153,7 @@ def filename_to_url(filename, cache_dir=None):
     return url, etag
 
 
-def cached_path(url_or_filename, cache_dir=None, force_download=False, proxies=None):
+def cached_path(url_or_filename, cache_dir=None, force_download=False, proxies=None, resume_download=False):
     """
     Given something that might be a URL (or might be a local path),
     determine which. If it's a URL, download the file and cache it, and
@@ -161,6 +162,7 @@ def cached_path(url_or_filename, cache_dir=None, force_download=False, proxies=N
     Args:
         cache_dir: specify a cache directory to save the file to (overwrite the default cache dir).
         force_download: if True, re-dowload the file even if it's already cached in the cache dir.
+        resume_download: if True, resume the download if incompletly recieved file is found.
     """
     if cache_dir is None:
         cache_dir = TRANSFORMERS_CACHE
@@ -173,7 +175,9 @@ def cached_path(url_or_filename, cache_dir=None, force_download=False, proxies=N
 
     if parsed.scheme in ('http', 'https', 's3'):
         # URL, so get it from the cache (downloading if necessary)
-        return get_from_cache(url_or_filename, cache_dir=cache_dir, force_download=force_download, proxies=proxies)
+        return get_from_cache(url_or_filename, cache_dir=cache_dir,
+            force_download=force_download, proxies=proxies,
+            resume_download=resume_download)
     elif os.path.exists(url_or_filename):
         # File, and it exists.
         return url_or_filename
@@ -234,19 +238,22 @@ def s3_get(url, temp_file, proxies=None):
     s3_resource.Bucket(bucket_name).download_fileobj(s3_path, temp_file)
 
 
-def http_get(url, temp_file, proxies=None):
-    req = requests.get(url, stream=True, proxies=proxies)
-    content_length = req.headers.get('Content-Length')
-    total = int(content_length) if content_length is not None else None
-    progress = tqdm(unit="B", total=total)
-    for chunk in req.iter_content(chunk_size=1024):
+def http_get(url, temp_file, proxies=None, resume_size=0):
+    headers={'Range':'bytes=%d-'%(resume_size,)} if resume_size > 0 else None
+    response = requests.get(url, stream=True, proxies=proxies, headers=headers)
+    if response.status_code == 416:  # Range not satisfiable
+        return
+    content_length = response.headers.get('Content-Length')
+    total = resume_size + int(content_length) if content_length is not None else None
+    progress = tqdm(unit="B", total=total, initial=resume_size)
+    for chunk in response.iter_content(chunk_size=1024):
         if chunk: # filter out keep-alive new chunks
             progress.update(len(chunk))
             temp_file.write(chunk)
     progress.close()
 
 
-def get_from_cache(url, cache_dir=None, force_download=False, proxies=None, etag_timeout=10):
+def get_from_cache(url, cache_dir=None, force_download=False, proxies=None, etag_timeout=10, resume_download=False):
     """
     Given a URL, look for the corresponding dataset in the local cache.
     If it's not there, download it. Then return the path to the cached file.
@@ -289,17 +296,35 @@ def get_from_cache(url, cache_dir=None, force_download=False, proxies=None, etag
         if matching_files:
             cache_path = os.path.join(cache_dir, matching_files[-1])
 
+    if resume_download:
+        incomplete_path = cache_path + '.incomplete'
+        @contextmanager
+        def _resumable_file_manager():
+            with open(incomplete_path,'a+b') as f:
+                yield f
+            os.remove(incomplete_path)
+        temp_file_manager = _resumable_file_manager
+        if os.path.exists(incomplete_path):
+            resume_size = os.stat(incomplete_path).st_size
+        else:
+            resume_size = 0
+    else:
+        temp_file_manager = tempfile.NamedTemporaryFile
+        resume_size = 0
+
     if not os.path.exists(cache_path) or force_download:
         # Download to temporary file, then copy to cache dir once finished.
         # Otherwise you get corrupt cache entries if the download gets interrupted.
-        with tempfile.NamedTemporaryFile() as temp_file:
+        with temp_file_manager() as temp_file:
             logger.info("%s not found in cache or force_download set to True, downloading to %s", url, temp_file.name)
 
             # GET file object
             if url.startswith("s3://"):
+                if resume_download:
+                    logger.warn('Warning: resumable downloads are not implemented for "s3://" urls')
                 s3_get(url, temp_file, proxies=proxies)
             else:
-                http_get(url, temp_file, proxies=proxies)
+                http_get(url, temp_file, proxies=proxies, resume_size=resume_size)
 
             # we are copying the file before closing it, so flush to avoid truncation
             temp_file.flush()
diff --git a/transformers/modeling_auto.py b/transformers/modeling_auto.py
index d98110d4bd..ea5a539c8c 100644
--- a/transformers/modeling_auto.py
+++ b/transformers/modeling_auto.py
@@ -112,6 +112,9 @@ class AutoModel(object):
             force_download: (`optional`) boolean, default False:
                 Force to (re-)download the model weights and configuration files and override the cached versions if they exists.
 
+            resume_download: (`optional`) boolean, default False:
+                Do not delete incompletely recieved file. Attempt to resume the download if such a file exists.
+
             proxies: (`optional`) dict, default None:
                 A dictionary of proxy servers to use by protocol or endpoint, e.g.: {'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.
                 The proxies are used on each request.
@@ -237,6 +240,8 @@ class AutoModelWithLMHead(object):
 
             force_download: (`optional`) boolean, default False:
                 Force to (re-)download the model weights and configuration files and override the cached versions if they exists.
+            resume_download: (`optional`) boolean, default False:
+                Do not delete incompletely recieved file. Attempt to resume the download if such a file exists.
 
             proxies: (`optional`) dict, default None:
                 A dictionary of proxy servers to use by protocol or endpoint, e.g.: {'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.
@@ -357,6 +362,9 @@ class AutoModelForSequenceClassification(object):
             force_download: (`optional`) boolean, default False:
                 Force to (re-)download the model weights and configuration files and override the cached versions if they exists.
 
+            resume_download: (`optional`) boolean, default False:
+                Do not delete incompletely recieved file. Attempt to resume the download if such a file exists.
+
             proxies: (`optional`) dict, default None:
                 A dictionary of proxy servers to use by protocol or endpoint, e.g.: {'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.
                 The proxies are used on each request.
diff --git a/transformers/modeling_tf_auto.py b/transformers/modeling_tf_auto.py
index df0ad6e401..cfe19ead2a 100644
--- a/transformers/modeling_tf_auto.py
+++ b/transformers/modeling_tf_auto.py
@@ -109,6 +109,9 @@ class TFAutoModel(object):
             force_download: (`optional`) boolean, default False:
                 Force to (re-)download the model weights and configuration files and override the cached versions if they exists.
 
+            resume_download: (`optional`) boolean, default False:
+                Do not delete incompletely recieved file. Attempt to resume the download if such a file exists.
+
             proxies: (`optional`) dict, default None:
                 A dictionary of proxy servers to use by protocol or endpoint, e.g.: {'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.
                 The proxies are used on each request.
@@ -237,6 +240,9 @@ class TFAutoModelWithLMHead(object):
             force_download: (`optional`) boolean, default False:
                 Force to (re-)download the model weights and configuration files and override the cached versions if they exists.
 
+            resume_download: (`optional`) boolean, default False:
+                Do not delete incompletely recieved file. Attempt to resume the download if such a file exists.
+
             proxies: (`optional`) dict, default None:
                 A dictionary of proxy servers to use by protocol or endpoint, e.g.: {'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.
                 The proxies are used on each request.
@@ -360,6 +366,9 @@ class TFAutoModelForSequenceClassification(object):
             force_download: (`optional`) boolean, default False:
                 Force to (re-)download the model weights and configuration files and override the cached versions if they exists.
 
+            resume_download: (`optional`) boolean, default False:
+                Do not delete incompletely recieved file. Attempt to resume the download if such a file exists.
+
             proxies: (`optional`) dict, default None:
                 A dictionary of proxy servers to use by protocol or endpoint, e.g.: {'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.
                 The proxies are used on each request.
@@ -472,6 +481,9 @@ class TFAutoModelForQuestionAnswering(object):
             force_download: (`optional`) boolean, default False:
                 Force to (re-)download the model weights and configuration files and override the cached versions if they exists.
 
+            resume_download: (`optional`) boolean, default False:
+                Do not delete incompletely recieved file. Attempt to resume the download if such a file exists.
+
             proxies: (`optional`) dict, default None:
                 A dictionary of proxy servers to use by protocol or endpoint, e.g.: {'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.
                 The proxies are used on each request.
diff --git a/transformers/modeling_tf_utils.py b/transformers/modeling_tf_utils.py
index a96e2765fd..aed2a25643 100644
--- a/transformers/modeling_tf_utils.py
+++ b/transformers/modeling_tf_utils.py
@@ -176,6 +176,9 @@ class TFPreTrainedModel(tf.keras.Model):
             force_download: (`optional`) boolean, default False:
                 Force to (re-)download the model weights and configuration files and override the cached versions if they exists.
 
+            resume_download: (`optional`) boolean, default False:
+                Do not delete incompletely recieved file. Attempt to resume the download if such a file exists.
+
             proxies: (`optional`) dict, default None:
                 A dictionary of proxy servers to use by protocol or endpoint, e.g.: {'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.
                 The proxies are used on each request.
@@ -201,6 +204,7 @@ class TFPreTrainedModel(tf.keras.Model):
         cache_dir = kwargs.pop('cache_dir', None)
         from_pt = kwargs.pop('from_pt', False)
         force_download = kwargs.pop('force_download', False)
+        resume_download = kwargs.pop('resume_download', False)
         proxies = kwargs.pop('proxies', None)
 
         # Load config
@@ -209,6 +213,7 @@ class TFPreTrainedModel(tf.keras.Model):
                 pretrained_model_name_or_path, *model_args,
                 cache_dir=cache_dir, return_unused_kwargs=True,
                 force_download=force_download,
+                resume_download=resume_download,
                 **kwargs
             )
         else:
@@ -236,7 +241,8 @@ class TFPreTrainedModel(tf.keras.Model):
 
             # redirect to the cache, if necessary
             try:
-                resolved_archive_file = cached_path(archive_file, cache_dir=cache_dir, force_download=force_download, proxies=proxies)
+                resolved_archive_file = cached_path(archive_file, cache_dir=cache_dir, force_download=force_download,
+                                                    resume_download=resume_download, proxies=proxies)
             except EnvironmentError as e:
                 if pretrained_model_name_or_path in cls.pretrained_model_archive_map:
                     logger.error(
diff --git a/transformers/modeling_utils.py b/transformers/modeling_utils.py
index d082137d5d..db3ab93f8f 100644
--- a/transformers/modeling_utils.py
+++ b/transformers/modeling_utils.py
@@ -246,6 +246,9 @@ class PreTrainedModel(nn.Module):
             force_download: (`optional`) boolean, default False:
                 Force to (re-)download the model weights and configuration files and override the cached versions if they exists.
 
+            resume_download: (`optional`) boolean, default False:
+                Do not delete incompletely recieved file. Attempt to resume the download if such a file exists.
+
             proxies: (`optional`) dict, default None:
                 A dictionary of proxy servers to use by protocol or endpoint, e.g.: {'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.
                 The proxies are used on each request.
@@ -275,6 +278,7 @@ class PreTrainedModel(nn.Module):
         cache_dir = kwargs.pop('cache_dir', None)
         from_tf = kwargs.pop('from_tf', False)
         force_download = kwargs.pop('force_download', False)
+        resume_download = kwargs.pop('resume_download', False)
         proxies = kwargs.pop('proxies', None)
         output_loading_info = kwargs.pop('output_loading_info', False)
 
@@ -284,6 +288,7 @@ class PreTrainedModel(nn.Module):
                 pretrained_model_name_or_path, *model_args,
                 cache_dir=cache_dir, return_unused_kwargs=True,
                 force_download=force_download,
+                resume_download=resume_download,
                 **kwargs
             )
         else:
@@ -315,7 +320,8 @@ class PreTrainedModel(nn.Module):
 
             # redirect to the cache, if necessary
             try:
-                resolved_archive_file = cached_path(archive_file, cache_dir=cache_dir, force_download=force_download, proxies=proxies)
+                resolved_archive_file = cached_path(archive_file, cache_dir=cache_dir, force_download=force_download,
+                                                    proxies=proxies, resume_download=resume_download)
             except EnvironmentError:
                 if pretrained_model_name_or_path in cls.pretrained_model_archive_map:
                     msg = "Couldn't reach server at '{}' to download pretrained weights.".format(
diff --git a/transformers/tokenization_auto.py b/transformers/tokenization_auto.py
index ec056de17f..455f503a9d 100644
--- a/transformers/tokenization_auto.py
+++ b/transformers/tokenization_auto.py
@@ -87,6 +87,9 @@ class AutoTokenizer(object):
             force_download: (`optional`) boolean, default False:
                 Force to (re-)download the vocabulary files and override the cached versions if they exists.
 
+            resume_download: (`optional`) boolean, default False:
+                Do not delete incompletely recieved file. Attempt to resume the download if such a file exists.
+
             proxies: (`optional`) dict, default None:
                 A dictionary of proxy servers to use by protocol or endpoint, e.g.: {'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.
                 The proxies are used on each request.
diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index 5e5be872ef..b602ebf20a 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -251,6 +251,9 @@ class PreTrainedTokenizer(object):
             force_download: (`optional`) boolean, default False:
                 Force to (re-)download the vocabulary files and override the cached versions if they exists.
 
+            resume_download: (`optional`) boolean, default False:
+                Do not delete incompletely recieved file. Attempt to resume the download if such a file exists.
+
             proxies: (`optional`) dict, default None:
                 A dictionary of proxy servers to use by protocol or endpoint, e.g.: {'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.
                 The proxies are used on each request.
@@ -286,6 +289,7 @@ class PreTrainedTokenizer(object):
     def _from_pretrained(cls, pretrained_model_name_or_path, *init_inputs, **kwargs):
         cache_dir = kwargs.pop('cache_dir', None)
         force_download = kwargs.pop('force_download', False)
+        resume_download = kwargs.pop('resume_download', False)
         proxies = kwargs.pop('proxies', None)
 
         s3_models = list(cls.max_model_input_sizes.keys())
@@ -352,7 +356,7 @@ class PreTrainedTokenizer(object):
                 if file_path is None:
                     resolved_vocab_files[file_id] = None
                 else:
-                    resolved_vocab_files[file_id] = cached_path(file_path, cache_dir=cache_dir, force_download=force_download, proxies=proxies)
+                    resolved_vocab_files[file_id] = cached_path(file_path, cache_dir=cache_dir, force_download=force_download, proxies=proxies, resume_download=resume_download)
         except EnvironmentError:
             if pretrained_model_name_or_path in s3_models:
                 msg = "Couldn't reach server at '{}' to download vocabulary files."

From 8d6b9d717c676560d291a7d2913b44e1cf226cd5 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Mon, 4 Nov 2019 17:07:51 +0100
Subject: [PATCH 006/505] fix #1532 and encode_plus

---
 .../tests/tokenization_tests_commons.py       | 21 ++++--
 transformers/tokenization_utils.py            | 74 ++++++++++++++++---
 2 files changed, 77 insertions(+), 18 deletions(-)

diff --git a/transformers/tests/tokenization_tests_commons.py b/transformers/tests/tokenization_tests_commons.py
index a921696b77..cb137e8395 100644
--- a/transformers/tests/tokenization_tests_commons.py
+++ b/transformers/tests/tokenization_tests_commons.py
@@ -223,7 +223,11 @@ class CommonTestCases:
             sequence = tokenizer.encode(seq_0, add_special_tokens=False)
             num_added_tokens = tokenizer.num_added_tokens()
             total_length = len(sequence) + num_added_tokens
-            information = tokenizer.encode_plus(seq_0, max_length=total_length - 2, add_special_tokens=True, stride=stride)
+            information = tokenizer.encode_plus(seq_0,
+                                                max_length=total_length - 2,
+                                                add_special_tokens=True,
+                                                stride=stride,
+                                                return_overflowing_tokens=True)
 
             truncated_sequence = information["input_ids"]
             overflowing_tokens = information["overflowing_tokens"]
@@ -250,10 +254,12 @@ class CommonTestCases:
             )
 
             information = tokenizer.encode_plus(seq_0, seq_1, max_length=len(sequence) - 2, add_special_tokens=True,
-                                                stride=stride, truncation_strategy='only_second')
+                                                stride=stride, truncation_strategy='only_second',
+                                                return_overflowing_tokens=True)
             information_first_truncated = tokenizer.encode_plus(seq_0, seq_1, max_length=len(sequence) - 2,
                                                                 add_special_tokens=True, stride=stride,
-                                                                truncation_strategy='only_first')
+                                                                truncation_strategy='only_first',
+                                                                return_overflowing_tokens=True)
 
             truncated_sequence = information["input_ids"]
             overflowing_tokens = information["overflowing_tokens"]
@@ -285,7 +291,7 @@ class CommonTestCases:
 
             # Testing single inputs
             encoded_sequence = tokenizer.encode(sequence_0, add_special_tokens=False)
-            encoded_sequence_dict = tokenizer.encode_plus(sequence_0, add_special_tokens=True)
+            encoded_sequence_dict = tokenizer.encode_plus(sequence_0, add_special_tokens=True, return_special_tokens_mask=True)
             encoded_sequence_w_special = encoded_sequence_dict["input_ids"]
             special_tokens_mask = encoded_sequence_dict["special_tokens_mask"]
             self.assertEqual(len(special_tokens_mask), len(encoded_sequence_w_special))
@@ -297,7 +303,8 @@ class CommonTestCases:
             # Testing inputs pairs
             encoded_sequence = tokenizer.encode(sequence_0, add_special_tokens=False) + tokenizer.encode(sequence_1,
                                                                                                          add_special_tokens=False)
-            encoded_sequence_dict = tokenizer.encode_plus(sequence_0, sequence_1, add_special_tokens=True)
+            encoded_sequence_dict = tokenizer.encode_plus(sequence_0, sequence_1, add_special_tokens=True,
+                                                          return_special_tokens_mask=True)
             encoded_sequence_w_special = encoded_sequence_dict["input_ids"]
             special_tokens_mask = encoded_sequence_dict["special_tokens_mask"]
             self.assertEqual(len(special_tokens_mask), len(encoded_sequence_w_special))
@@ -309,7 +316,9 @@ class CommonTestCases:
             # Testing with already existing special tokens
             if tokenizer.cls_token_id == tokenizer.unk_token_id and tokenizer.cls_token_id == tokenizer.unk_token_id:
                 tokenizer.add_special_tokens({'cls_token': '</s>', 'sep_token': '<s>'})
-            encoded_sequence_dict = tokenizer.encode_plus(sequence_0, add_special_tokens=True)
+            encoded_sequence_dict = tokenizer.encode_plus(sequence_0,
+                                                          add_special_tokens=True,
+                                                          return_special_tokens_mask=True)
             encoded_sequence_w_special = encoded_sequence_dict["input_ids"]
             special_tokens_mask_orig = encoded_sequence_dict["special_tokens_mask"]
             special_tokens_mask = tokenizer.get_special_tokens_mask(encoded_sequence_w_special, already_has_special_tokens=True)
diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index ac765165e2..f860755775 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -744,6 +744,9 @@ class PreTrainedTokenizer(object):
                     stride=0,
                     truncation_strategy='longest_first',
                     return_tensors=None,
+                    return_token_type_ids=True,
+                    return_overflowing_tokens=False,
+                    return_special_tokens_mask=False,
                     **kwargs):
         """
         Returns a dictionary containing the encoded sequence or sequence pair and additional informations:
@@ -770,7 +773,30 @@ class PreTrainedTokenizer(object):
                 - 'do_not_truncate': Does not truncate (raise an error if the input sequence is longer than max_length)
             return_tensors: (optional) can be set to 'tf' or 'pt' to return respectively TensorFlow tf.constant
                 or PyTorch torch.Tensor instead of a list of python integers.
+            return_token_type_ids: (optional) Set to False to avoid returning token_type_ids (default True).
+            return_overflowing_tokens: (optional) Set to True to return overflowing token information (default False).
+            return_special_tokens_mask: (optional) Set to True to return special tokens mask information (default False).
             **kwargs: passed to the `self.tokenize()` method
+
+        Return:
+            A Dictionary of shape::
+
+                {
+                    input_ids: list[int],
+                    token_type_ids: list[int] if return_token_type_ids is True (default)
+                    overflowing_tokens: list[int] if a ``max_length`` is specified and return_overflowing_tokens is True
+                    num_truncated_tokens: int if a ``max_length`` is specified and return_overflowing_tokens is True
+                    special_tokens_mask: list[int] if ``add_special_tokens`` if set to ``True`` and return_special_tokens_mask is True
+                }
+
+            With the fields:
+                ``input_ids``: list of token ids to be fed to a model
+                ``token_type_ids``: list of token type ids to be fed to a model
+
+                ``overflowing_tokens``: list of overflowing tokens if a max length is specified.
+                ``num_truncated_tokens``: number of overflowing tokens a ``max_length`` is specified
+                ``special_tokens_mask``: if adding special tokens, this is a list of [0, 1], with 0 specifying special added
+                tokens and 1 specifying sequence tokens.
         """
 
         def get_input_ids(text):
@@ -792,10 +818,17 @@ class PreTrainedTokenizer(object):
                                       add_special_tokens=add_special_tokens,
                                       stride=stride,
                                       truncation_strategy=truncation_strategy,
-                                      return_tensors=return_tensors)
+                                      return_tensors=return_tensors,
+                                      return_token_type_ids=return_token_type_ids,
+                                      return_overflowing_tokens=return_overflowing_tokens,
+                                      return_special_tokens_mask=return_special_tokens_mask)
 
     def prepare_for_model(self, ids, pair_ids=None, max_length=None, add_special_tokens=True, stride=0,
-                          truncation_strategy='longest_first', return_tensors=None):
+                          truncation_strategy='longest_first',
+                          return_tensors=None,
+                          return_token_type_ids=True,
+                          return_overflowing_tokens=False,
+                          return_special_tokens_mask=False):
         """
         Prepares a sequence of input id, or a pair of sequences of inputs ids so that it can be used by the model.
         It adds special tokens, truncates
@@ -820,21 +853,27 @@ class PreTrainedTokenizer(object):
                 - 'do_not_truncate': Does not truncate (raise an error if the input sequence is longer than max_length)
             return_tensors: (optional) can be set to 'tf' or 'pt' to return respectively TensorFlow tf.constant
                 or PyTorch torch.Tensor instead of a list of python integers.
+            return_token_type_ids: (optional) Set to False to avoid returning token_type_ids (default True).
+            return_overflowing_tokens: (optional) Set to True to return overflowing token information (default False).
+            return_special_tokens_mask: (optional) Set to True to return special tokens mask information (default False).
 
         Return:
             A Dictionary of shape::
 
                 {
                     input_ids: list[int],
-                    overflowing_tokens: list[int] if a ``max_length`` is specified, else None
-                    special_tokens_mask: list[int] if ``add_special_tokens`` if set to ``True``
+                    token_type_ids: list[int] if return_token_type_ids is True (default)
+                    overflowing_tokens: list[int] if a ``max_length`` is specified and return_overflowing_tokens is True
+                    num_truncated_tokens: int if a ``max_length`` is specified and return_overflowing_tokens is True
+                    special_tokens_mask: list[int] if ``add_special_tokens`` if set to ``True`` and return_special_tokens_mask is True
                 }
 
             With the fields:
-                ``input_ids``: list of tokens to be fed to a model
+                ``input_ids``: list of token ids to be fed to a model
+                ``token_type_ids``: list of token type ids to be fed to a model
 
                 ``overflowing_tokens``: list of overflowing tokens if a max length is specified.
-
+                ``num_truncated_tokens``: number of overflowing tokens a ``max_length`` is specified
                 ``special_tokens_mask``: if adding special tokens, this is a list of [0, 1], with 0 specifying special added
                 tokens and 1 specifying sequence tokens.
         """
@@ -843,23 +882,31 @@ class PreTrainedTokenizer(object):
         len_pair_ids = len(pair_ids) if pair else 0
 
         encoded_inputs = {}
+
+        # Handle max sequence length
         total_len = len_ids + len_pair_ids + (self.num_added_tokens(pair=pair) if add_special_tokens else 0)
         if max_length and total_len > max_length:
             ids, pair_ids, overflowing_tokens = self.truncate_sequences(ids, pair_ids=pair_ids,
                                                                         num_tokens_to_remove=total_len-max_length,
                                                                         truncation_strategy=truncation_strategy,
                                                                         stride=stride)
-            encoded_inputs["overflowing_tokens"] = overflowing_tokens
-            encoded_inputs["num_truncated_tokens"] = total_len - max_length
+            if return_overflowing_tokens:
+                encoded_inputs["overflowing_tokens"] = overflowing_tokens
+                encoded_inputs["num_truncated_tokens"] = total_len - max_length
 
+        # Handle special_tokens
         if add_special_tokens:
             sequence = self.build_inputs_with_special_tokens(ids, pair_ids)
             token_type_ids = self.create_token_type_ids_from_sequences(ids, pair_ids)
-            encoded_inputs["special_tokens_mask"] = self.get_special_tokens_mask(ids, pair_ids)
+            special_tokens_mask = self.get_special_tokens_mask(ids, pair_ids)
         else:
             sequence = ids + pair_ids if pair else ids
             token_type_ids = [0] * len(ids) + ([1] * len(pair_ids) if pair else [])
+            special_tokens_mask = [0] * (len(ids) + (len(pair_ids) if pair else 0))
+        if return_special_tokens_mask:
+            encoded_inputs["special_tokens_mask"] = self.get_special_tokens_mask(ids, pair_ids)
 
+        # Prepare inputs as tensors if asked
         if return_tensors == 'tf' and is_tf_available():
             sequence = tf.constant([sequence])
             token_type_ids = tf.constant([token_type_ids])
@@ -870,12 +917,15 @@ class PreTrainedTokenizer(object):
             logger.warning("Unable to convert output to tensors format {}, PyTorch or TensorFlow is not available.".format(return_tensors))
 
         encoded_inputs["input_ids"] = sequence
-        encoded_inputs["token_type_ids"] = token_type_ids
+        if return_token_type_ids:
+            encoded_inputs["token_type_ids"] = token_type_ids
 
         if max_length and len(encoded_inputs["input_ids"]) > max_length:
             encoded_inputs["input_ids"] = encoded_inputs["input_ids"][:max_length]
-            encoded_inputs["token_type_ids"] = encoded_inputs["token_type_ids"][:max_length]
-            encoded_inputs["special_tokens_mask"] = encoded_inputs["special_tokens_mask"][:max_length]
+            if return_token_type_ids:
+                encoded_inputs["token_type_ids"] = encoded_inputs["token_type_ids"][:max_length]
+            if return_special_tokens_mask:
+                encoded_inputs["special_tokens_mask"] = encoded_inputs["special_tokens_mask"][:max_length]
 
         return encoded_inputs
 

From 8df7dfd2a723465b0cca9f5e808a75e074482d02 Mon Sep 17 00:00:00 2001
From: Filip Povolny <filip.povolny@ondewo.com>
Date: Tue, 5 Nov 2019 11:09:16 +0100
Subject: [PATCH 007/505] Make dummy inputs a local variable in
 TFPreTrainedModel.

---
 transformers/modeling_tf_utils.py | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/transformers/modeling_tf_utils.py b/transformers/modeling_tf_utils.py
index a96e2765fd..110a590f55 100644
--- a/transformers/modeling_tf_utils.py
+++ b/transformers/modeling_tf_utils.py
@@ -51,7 +51,6 @@ class TFPreTrainedModel(tf.keras.Model):
     config_class = None
     pretrained_model_archive_map = {}
     base_model_prefix = ""
-    dummy_inputs = tf.constant(DUMMY_INPUTS)  # dummy inputs to build the network
 
     def __init__(self, config, *inputs, **kwargs):
         super(TFPreTrainedModel, self).__init__(*inputs, **kwargs)
@@ -266,14 +265,15 @@ class TFPreTrainedModel(tf.keras.Model):
             # Load from a PyTorch checkpoint
             return load_pytorch_checkpoint_in_tf2_model(model, resolved_archive_file)
 
-        ret = model(model.dummy_inputs, training=False)  # build the network with dummy inputs
+        dummy_inputs = tf.constant(DUMMY_INPUTS)  # dummy inputs to build the network
+        ret = model(dummy_inputs, training=False)  # build the network with dummy inputs
 
         assert os.path.isfile(resolved_archive_file), "Error retrieving file {}".format(resolved_archive_file)
         # 'by_name' allow us to do transfer learning by skipping/adding layers
         # see https://github.com/tensorflow/tensorflow/blob/00fad90125b18b80fe054de1055770cfb8fe4ba3/tensorflow/python/keras/engine/network.py#L1339-L1357
         model.load_weights(resolved_archive_file, by_name=True)
 
-        ret = model(model.dummy_inputs, training=False)  # Make sure restore ops are run
+        ret = model(dummy_inputs, training=False)  # Make sure restore ops are run
 
         return model
 

From dfb61caf77a02a735af0bf430f6d6082b7d01cfd Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Tue, 5 Nov 2019 11:25:13 +0100
Subject: [PATCH 008/505] fix #1692

---
 transformers/modeling_tf_xlnet.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/modeling_tf_xlnet.py b/transformers/modeling_tf_xlnet.py
index 8a25be78c1..e83674e214 100644
--- a/transformers/modeling_tf_xlnet.py
+++ b/transformers/modeling_tf_xlnet.py
@@ -539,7 +539,7 @@ class TFXLNetMainLayer(tf.keras.layers.Layer):
         assert input_mask is None or attention_mask is None, "You can only use one of input_mask (uses 1 for padding) " \
             "or attention_mask (uses 0 for padding, added for compatbility with BERT). Please choose one."
         if input_mask is None and attention_mask is not None:
-            input_mask = 1.0 - attention_mask
+            input_mask = 1.0 - tf.cast(attention_mask, dtype=dtype_float)
         if input_mask is not None and perm_mask is not None:
             data_mask = input_mask[None] + perm_mask
         elif input_mask is not None and perm_mask is None:

From 124409d0754f5b84ff47daea51dede1d3b37a8dd Mon Sep 17 00:00:00 2001
From: Filip Povolny <filip.povolny@ondewo.com>
Date: Tue, 5 Nov 2019 11:48:45 +0100
Subject: [PATCH 009/505] Make dummy inputs a property of TFPreTrainedModel.

---
 transformers/modeling_tf_utils.py | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

diff --git a/transformers/modeling_tf_utils.py b/transformers/modeling_tf_utils.py
index 110a590f55..33cfdc503d 100644
--- a/transformers/modeling_tf_utils.py
+++ b/transformers/modeling_tf_utils.py
@@ -52,6 +52,15 @@ class TFPreTrainedModel(tf.keras.Model):
     pretrained_model_archive_map = {}
     base_model_prefix = ""
 
+    @property
+    def dummy_inputs(self):
+        """ Dummy inputs to build the network.
+
+        Returns:
+            tf.Tensor with dummy inputs
+        """
+        return tf.constant(DUMMY_INPUTS)
+
     def __init__(self, config, *inputs, **kwargs):
         super(TFPreTrainedModel, self).__init__(*inputs, **kwargs)
         if not isinstance(config, PretrainedConfig):
@@ -265,15 +274,14 @@ class TFPreTrainedModel(tf.keras.Model):
             # Load from a PyTorch checkpoint
             return load_pytorch_checkpoint_in_tf2_model(model, resolved_archive_file)
 
-        dummy_inputs = tf.constant(DUMMY_INPUTS)  # dummy inputs to build the network
-        ret = model(dummy_inputs, training=False)  # build the network with dummy inputs
+        ret = model(model.dummy_inputs, training=False)  # build the network with dummy inputs
 
         assert os.path.isfile(resolved_archive_file), "Error retrieving file {}".format(resolved_archive_file)
         # 'by_name' allow us to do transfer learning by skipping/adding layers
         # see https://github.com/tensorflow/tensorflow/blob/00fad90125b18b80fe054de1055770cfb8fe4ba3/tensorflow/python/keras/engine/network.py#L1339-L1357
         model.load_weights(resolved_archive_file, by_name=True)
 
-        ret = model(dummy_inputs, training=False)  # Make sure restore ops are run
+        ret = model(model.dummy_inputs, training=False)  # Make sure restore ops are run
 
         return model
 

From 60a5babd57dd80f855df859abf006ee4488ff639 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Tue, 5 Nov 2019 12:01:23 +0100
Subject: [PATCH 010/505] adding files

---
 transformers/configuration_t5.py              | 130 +++++
 ...rt_t5_original_tf_checkpoint_to_pytorch.py |  65 +++
 transformers/modeling_t5.py                   | 373 +++++++++++++
 transformers/modeling_tf_t5.py                | 496 ++++++++++++++++++
 transformers/tokenization_t5.py               | 214 ++++++++
 5 files changed, 1278 insertions(+)
 create mode 100644 transformers/configuration_t5.py
 create mode 100755 transformers/convert_t5_original_tf_checkpoint_to_pytorch.py
 create mode 100644 transformers/modeling_t5.py
 create mode 100644 transformers/modeling_tf_t5.py
 create mode 100644 transformers/tokenization_t5.py

diff --git a/transformers/configuration_t5.py b/transformers/configuration_t5.py
new file mode 100644
index 0000000000..a37a5b2157
--- /dev/null
+++ b/transformers/configuration_t5.py
@@ -0,0 +1,130 @@
+# coding=utf-8
+# Copyright 2010, The T5 Authors and HuggingFace Inc.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" T5 model configuration """
+
+from __future__ import absolute_import, division, print_function, unicode_literals
+
+import json
+import logging
+import sys
+import six
+from io import open
+
+from .configuration_utils import PretrainedConfig
+
+logger = logging.getLogger(__name__)
+
+T5_PRETRAINED_CONFIG_ARCHIVE_MAP = {
+    't5-base-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-base-uncased-config.json",
+    't5-large-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-large-uncased-config.json",
+}
+
+
+class T5Config(PretrainedConfig):
+    r"""
+        :class:`~transformers.T5Config` is the configuration class to store the configuration of a
+        `T5Model`.
+
+
+        Arguments:
+            vocab_size_or_config_json_file: Vocabulary size of `inputs_ids` in `T5Model`.
+            hidden_size: Size of the encoder layers and the pooler layer.
+            num_hidden_layers: Number of hidden layers in the Transformer encoder.
+            num_attention_heads: Number of attention heads for each attention layer in
+                the Transformer encoder.
+            intermediate_size: The size of the "intermediate" (i.e., feed-forward)
+                layer in the Transformer encoder.
+            hidden_act: The non-linear activation function (function or string) in the
+                encoder and pooler. If string, "gelu", "relu", "swish" and "gelu_new" are supported.
+            hidden_dropout_prob: The dropout probabilitiy for all fully connected
+                layers in the embeddings, encoder, and pooler.
+            attention_probs_dropout_prob: The dropout ratio for the attention
+                probabilities.
+            max_position_embeddings: The maximum sequence length that this model might
+                ever be used with. Typically set this to something large just in case
+                (e.g., 512 or 1024 or 2048).
+            type_vocab_size: The vocabulary size of the `token_type_ids` passed into
+                `T5Model`.
+            initializer_range: The sttdev of the truncated_normal_initializer for
+                initializing all weight matrices.
+            layer_norm_eps: The epsilon used by LayerNorm.
+    """
+    pretrained_config_archive_map = T5_PRETRAINED_CONFIG_ARCHIVE_MAP
+
+    def __init__(self,
+                 vocab_size_or_config_json_file=50257,
+                 n_positions=1024,
+                 n_ctx=1024,
+                 n_embd=768,
+                 n_layer=12,
+                 n_head=12,
+                 resid_pdrop=0.1,
+                 embd_pdrop=0.1,
+                 attn_pdrop=0.1,
+                 layer_norm_epsilon=1e-5,
+                 initializer_range=0.02,
+
+                 num_labels=1,
+                 summary_type='cls_index',
+                 summary_use_proj=True,
+                 summary_activation=None,
+                 summary_proj_to_labels=True,
+                 summary_first_dropout=0.1,
+                 **kwargs):
+        super(T5Config, self).__init__(**kwargs)
+        self.vocab_size = vocab_size_or_config_json_file if isinstance(vocab_size_or_config_json_file, six.string_types) else -1
+        self.n_ctx = n_ctx
+        self.n_positions = n_positions
+        self.n_embd = n_embd
+        self.n_layer = n_layer
+        self.n_head = n_head
+        self.resid_pdrop = resid_pdrop
+        self.embd_pdrop = embd_pdrop
+        self.attn_pdrop = attn_pdrop
+        self.layer_norm_epsilon = layer_norm_epsilon
+        self.initializer_range = initializer_range
+
+        self.num_labels = num_labels
+        self.summary_type = summary_type
+        self.summary_use_proj = summary_use_proj
+        self.summary_activation = summary_activation
+        self.summary_first_dropout = summary_first_dropout
+        self.summary_proj_to_labels = summary_proj_to_labels
+        if isinstance(vocab_size_or_config_json_file, six.string_types):
+            with open(vocab_size_or_config_json_file, "r", encoding="utf-8") as reader:
+                json_config = json.loads(reader.read())
+            for key, value in json_config.items():
+                self.__dict__[key] = value
+        elif not isinstance(vocab_size_or_config_json_file, int):
+            raise ValueError(
+                "First argument must be either a vocabulary size (int)"
+                "or the path to a pretrained model config file (str)"
+            )
+
+    @property
+    def max_position_embeddings(self):
+        return self.n_positions
+
+    @property
+    def hidden_size(self):
+        return self.n_embd
+
+    @property
+    def num_attention_heads(self):
+        return self.n_head
+
+    @property
+    def num_hidden_layers(self):
+        return self.n_layer
diff --git a/transformers/convert_t5_original_tf_checkpoint_to_pytorch.py b/transformers/convert_t5_original_tf_checkpoint_to_pytorch.py
new file mode 100755
index 0000000000..608027ebac
--- /dev/null
+++ b/transformers/convert_t5_original_tf_checkpoint_to_pytorch.py
@@ -0,0 +1,65 @@
+# coding=utf-8
+# Copyright 2018 The T5 authors and HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Convert T5 checkpoint."""
+
+from __future__ import absolute_import
+from __future__ import division
+from __future__ import print_function
+
+import argparse
+import torch
+
+from transformers import T5Config, T5ForPreTraining, load_tf_weights_in_t5
+
+import logging
+logging.basicConfig(level=logging.INFO)
+
+def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, t5_config_file, pytorch_dump_path):
+    # Initialise PyTorch model
+    config = T5Config.from_json_file(t5_config_file)
+    print("Building PyTorch model from configuration: {}".format(str(config)))
+    model = T5ForPreTraining(config)
+
+    # Load weights from tf checkpoint
+    load_tf_weights_in_t5(model, config, tf_checkpoint_path)
+
+    # Save pytorch-model
+    print("Save PyTorch model to {}".format(pytorch_dump_path))
+    torch.save(model.state_dict(), pytorch_dump_path)
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    ## Required parameters
+    parser.add_argument("--tf_checkpoint_path",
+                        default = None,
+                        type = str,
+                        required = True,
+                        help = "Path to the TensorFlow checkpoint path.")
+    parser.add_argument("--t5_config_file",
+                        default = None,
+                        type = str,
+                        required = True,
+                        help = "The config json file corresponding to the pre-trained T5 model. \n"
+                            "This specifies the model architecture.")
+    parser.add_argument("--pytorch_dump_path",
+                        default = None,
+                        type = str,
+                        required = True,
+                        help = "Path to the output PyTorch model.")
+    args = parser.parse_args()
+    convert_tf_checkpoint_to_pytorch(args.tf_checkpoint_path,
+                                     args.t5_config_file,
+                                     args.pytorch_dump_path)
diff --git a/transformers/modeling_t5.py b/transformers/modeling_t5.py
new file mode 100644
index 0000000000..fa3c22f24b
--- /dev/null
+++ b/transformers/modeling_t5.py
@@ -0,0 +1,373 @@
+# coding=utf-8
+# Copyright 2018 T5 Authors and HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" PyTorch T5 model. """
+
+from __future__ import absolute_import, division, print_function, unicode_literals
+
+import json
+import logging
+import math
+import os
+import sys
+from io import open
+
+import torch
+from torch import nn
+from torch.nn import CrossEntropyLoss, MSELoss
+
+from .modeling_utils import PreTrainedModel, prune_linear_layer
+from .configuration_t5 import T5Config
+from .file_utils import add_start_docstrings
+
+logger = logging.getLogger(__name__)
+
+####################################################
+# This dict contrains shortcut names and associated url
+# for the pretrained weights provided with the models
+####################################################
+T5_PRETRAINED_MODEL_ARCHIVE_MAP = {
+    't5-base-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-base-uncased-pytorch_model.bin",
+    't5-large-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-large-uncased-pytorch_model.bin",
+}
+
+####################################################
+# This is a conversion method from TF 1.0 to PyTorch
+# More details: https://medium.com/huggingface/from-tensorflow-to-pytorch-265f40ef2a28
+####################################################
+def load_tf_weights_in_t5(model, config, tf_checkpoint_path):
+    """ Load tf checkpoints in a pytorch model.
+    """
+    try:
+        import re
+        import numpy as np
+        import tensorflow as tf
+    except ImportError:
+        logger.error("Loading a TensorFlow model in PyTorch, requires TensorFlow to be installed. Please see "
+            "https://www.tensorflow.org/install/ for installation instructions.")
+        raise
+    tf_path = os.path.abspath(tf_checkpoint_path)
+    logger.info("Converting TensorFlow checkpoint from {}".format(tf_path))
+    # Load weights from TF model
+    init_vars = tf.train.list_variables(tf_path)
+    names = []
+    arrays = []
+    for name, shape in init_vars:
+        logger.info("Loading TF weight {} with shape {}".format(name, shape))
+        array = tf.train.load_variable(tf_path, name)
+        names.append(name)
+        arrays.append(array)
+
+    for name, array in zip(names, arrays):
+        name = name.split('/')
+        # adam_v and adam_m are variables used in AdamWeightDecayOptimizer to calculated m and v
+        # which are not required for using pretrained model
+        if any(n in ["adam_v", "adam_m", "global_step"] for n in name):
+            logger.info("Skipping {}".format("/".join(name)))
+            continue
+        pointer = model
+        for m_name in name:
+            if re.fullmatch(r'[A-Za-z]+_\d+', m_name):
+                l = re.split(r'_(\d+)', m_name)
+            else:
+                l = [m_name]
+            if l[0] == 'kernel' or l[0] == 'gamma':
+                pointer = getattr(pointer, 'weight')
+            elif l[0] == 'output_bias' or l[0] == 'beta':
+                pointer = getattr(pointer, 'bias')
+            elif l[0] == 'output_weights':
+                pointer = getattr(pointer, 'weight')
+            elif l[0] == 'squad':
+                pointer = getattr(pointer, 'classifier')
+            else:
+                try:
+                    pointer = getattr(pointer, l[0])
+                except AttributeError:
+                    logger.info("Skipping {}".format("/".join(name)))
+                    continue
+            if len(l) >= 2:
+                num = int(l[1])
+                pointer = pointer[num]
+        if m_name[-11:] == '_embeddings':
+            pointer = getattr(pointer, 'weight')
+        elif m_name == 'kernel':
+            array = np.transpose(array)
+        try:
+            assert pointer.shape == array.shape
+        except AssertionError as e:
+            e.args += (pointer.shape, array.shape)
+            raise
+        logger.info("Initialize PyTorch weight {}".format(name))
+        pointer.data = torch.from_numpy(array)
+    return model
+
+
+####################################################
+# PyTorch Models are constructed by sub-classing
+# - torch.nn.Module for the layers and
+# - PreTrainedModel for the models (it-self a sub-class of torch.nn.Module)
+####################################################
+
+class T5Layer(nn.Module):
+    def __init__(self, config):
+        super(T5Layer, self).__init__()
+        self.attention = T5Attention(config)
+        self.intermediate = T5Intermediate(config)
+        self.output = T5Output(config)
+
+    def forward(self, hidden_states, attention_mask=None, head_mask=None):
+        attention_outputs = self.attention(hidden_states, attention_mask, head_mask)
+        attention_output = attention_outputs[0]
+        intermediate_output = self.intermediate(attention_output)
+        layer_output = self.output(intermediate_output, attention_output)
+        outputs = (layer_output,) + attention_outputs[1:]  # add attentions if we output them
+        return outputs
+
+
+
+class T5PreTrainedModel(PreTrainedModel):
+    """ An abstract class to handle weights initialization and
+        a simple interface for dowloading and loading pretrained models.
+    """
+    config_class = T5Config
+    pretrained_model_archive_map = T5_PRETRAINED_MODEL_ARCHIVE_MAP
+    load_tf_weights = load_tf_weights_in_t5
+    base_model_prefix = "transformer"
+
+    def _init_weights(self, module):
+        """ Initialize the weights """
+        if isinstance(module, (nn.Linear, nn.Embedding)):
+            # Slightly different from the TF version which uses truncated_normal for initialization
+            # cf https://github.com/pytorch/pytorch/pull/5617
+            module.weight.data.normal_(mean=0.0, std=self.config.initializer_range)
+        elif isinstance(module, nn.LayerNorm):
+            module.bias.data.zero_()
+            module.weight.data.fill_(1.0)
+        if isinstance(module, nn.Linear) and module.bias is not None:
+            module.bias.data.zero_()
+
+
+T5_START_DOCSTRING = r"""    The T5 model was proposed in
+    `Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer`_
+    by Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, Peter J. Liu.
+    It's an encoder decoder pre-trained transformer.
+
+    This model is a PyTorch `torch.nn.Module`_ sub-class. Use it as a regular PyTorch Module and
+    refer to the PyTorch documentation for all matter related to general usage and behavior.
+
+    .. _`Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer`:
+        https://arxiv.org/abs/1910.10683
+
+    .. _`torch.nn.Module`:
+        https://pytorch.org/docs/stable/nn.html#module
+
+    Parameters:
+        config (:class:`~transformers.T5Config`): Model configuration class with all the parameters of the model. 
+            Initializing with a config file does not load the weights associated with the model, only the configuration.
+            Check out the :meth:`~transformers.PreTrainedModel.from_pretrained` method to load the model weights.
+"""
+
+T5_INPUTS_DOCSTRING = r"""
+    Inputs:
+        **input_ids**: ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
+            Indices of input sequence tokens in the vocabulary.
+            To match pre-training, T5 input sequence should be formatted with [CLS] and [SEP] tokens as follows:
+
+            (a) For sequence pairs:
+
+                ``tokens:         [CLS] is this jack ##son ##ville ? [SEP] no it is not . [SEP]``
+
+            (b) For single sequences:
+
+                ``tokens:         [CLS] the dog is hairy . [SEP]``
+
+            T5 is a model with relative position embeddings so you should be able to pad the inputs on
+            the right or the left.
+
+            Indices can be obtained using :class:`transformers.T5Tokenizer`.
+            See :func:`transformers.PreTrainedTokenizer.encode` and
+            :func:`transformers.PreTrainedTokenizer.convert_tokens_to_ids` for details.
+        **attention_mask**: (`optional`) ``torch.FloatTensor`` of shape ``(batch_size, sequence_length)``:
+            Mask to avoid performing attention on padding token indices.
+            Mask values selected in ``[0, 1]``:
+            ``1`` for tokens that are NOT MASKED, ``0`` for MASKED tokens.
+        **position_ids**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
+            Indices of positions of each input sequence tokens in the position embeddings.
+            Selected in the range ``[0, config.max_position_embeddings - 1]``.
+        **head_mask**: (`optional`) ``torch.FloatTensor`` of shape ``(num_heads,)`` or ``(num_layers, num_heads)``:
+            Mask to nullify selected heads of the self-attention modules.
+            Mask values selected in ``[0, 1]``:
+            ``1`` indicates the head is **not masked**, ``0`` indicates the head is **masked**.
+"""
+
+@add_start_docstrings("The bare single stack (encoder or decoder) of a T5 Model transformer outputting raw hidden-states"
+                      "without any specific head on top.",
+                      T5_START_DOCSTRING, T5_INPUTS_DOCSTRING)
+class T5Model(T5PreTrainedModel):
+    r"""
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **last_hidden_state**: ``torch.FloatTensor`` of shape ``(batch_size, sequence_length, hidden_size)``
+            Sequence of hidden-states at the output of the last layer of the model.
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``torch.FloatTensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        tokenizer = T5Tokenizer.from_pretrained('t5-base-uncased')
+        model = T5Model.from_pretrained('t5-base-uncased')
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        outputs = model(input_ids)
+        last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
+
+    """
+    def __init__(self, config):
+        super(T5Model, self).__init__(config)
+
+        self.embeddings = T5Embeddings(config)
+        self.encoder = T5Encoder(config)
+        self.pooler = T5Pooler(config)
+
+        self.init_weights()
+
+    @property
+    def get_input_embeddings(self):
+        return self.embeddings.word_embeddings
+
+    def set_input_embeddings(self, new_embeddings):
+        self.embeddings.word_embeddings = new_embeddings
+
+    def _prune_heads(self, heads_to_prune):
+        """ Prunes heads of the model.
+            heads_to_prune: dict of {layer_num: list of heads to prune in this layer}
+            See base class PreTrainedModel
+        """
+        for layer, heads in heads_to_prune.items():
+            self.encoder.layer[layer].attention.prune_heads(heads)
+
+    def forward(self, input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None):
+        if attention_mask is None:
+            attention_mask = torch.ones_like(input_ids)
+        if token_type_ids is None:
+            token_type_ids = torch.zeros_like(input_ids)
+
+        # We create a 3D attention mask from a 2D tensor mask.
+        # Sizes are [batch_size, 1, 1, to_seq_length]
+        # So we can broadcast to [batch_size, num_heads, from_seq_length, to_seq_length]
+        # this attention mask is more simple than the triangular masking of causal attention
+        # used in OpenAI GPT, we just need to prepare the broadcast dimension here.
+        extended_attention_mask = attention_mask.unsqueeze(1).unsqueeze(2)
+
+        # Since attention_mask is 1.0 for positions we want to attend and 0.0 for
+        # masked positions, this operation will create a tensor which is 0.0 for
+        # positions we want to attend and -10000.0 for masked positions.
+        # Since we are adding it to the raw scores before the softmax, this is
+        # effectively the same as removing these entirely.
+        extended_attention_mask = extended_attention_mask.to(dtype=next(self.parameters()).dtype) # fp16 compatibility
+        extended_attention_mask = (1.0 - extended_attention_mask) * -10000.0
+
+        # Prepare head mask if needed
+        # 1.0 in head_mask indicate we keep the head
+        # attention_probs has shape bsz x n_heads x N x N
+        # input head_mask has shape [num_heads] or [num_hidden_layers x num_heads]
+        # and head_mask is converted to shape [num_hidden_layers x batch x num_heads x seq_length x seq_length]
+        if head_mask is not None:
+            if head_mask.dim() == 1:
+                head_mask = head_mask.unsqueeze(0).unsqueeze(0).unsqueeze(-1).unsqueeze(-1)
+                head_mask = head_mask.expand(self.config.num_hidden_layers, -1, -1, -1, -1)
+            elif head_mask.dim() == 2:
+                head_mask = head_mask.unsqueeze(1).unsqueeze(-1).unsqueeze(-1)  # We can specify head_mask for each layer
+            head_mask = head_mask.to(dtype=next(self.parameters()).dtype) # switch to fload if need + fp16 compatibility
+        else:
+            head_mask = [None] * self.config.num_hidden_layers
+
+        ##################################
+        # Replace this with your model code
+        embedding_output = self.embeddings(input_ids, position_ids=position_ids, token_type_ids=token_type_ids)
+        encoder_outputs = self.encoder(embedding_output, extended_attention_mask, head_mask=head_mask)
+        sequence_output = encoder_outputs[0]
+        outputs = (sequence_output,) + encoder_outputs[1:]  # add hidden_states and attentions if they are here
+
+        return outputs  # sequence_output, (hidden_states), (attentions)
+
+
+@add_start_docstrings("""T5 Model with a `language modeling` head on top. """,
+    T5_START_DOCSTRING, T5_INPUTS_DOCSTRING)
+class T5WithLMHead(T5PreTrainedModel):
+    r"""
+        **lm_labels**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
+            Labels for computing the masked language modeling loss.
+            Indices should be in ``[-1, 0, ..., config.vocab_size]`` (see ``input_ids`` docstring)
+            Tokens with indices set to ``-1`` are ignored (masked), the loss is only computed for the tokens with labels
+            in ``[0, ..., config.vocab_size]``
+
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **loss**: (`optional`, returned when ``lm_labels`` is provided) ``torch.FloatTensor`` of shape ``(1,)``:
+            Masked language modeling loss.
+        **prediction_scores**: ``torch.FloatTensor`` of shape ``(batch_size, sequence_length, config.vocab_size)``
+            Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``torch.FloatTensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        tokenizer = T5Tokenizer.from_pretrained('t5-base-uncased')
+        model = T5ForMaskedLM.from_pretrained('t5-base-uncased')
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        outputs = model(input_ids, lm_labels=input_ids)
+        loss, prediction_scores = outputs[:2]
+
+    """
+    def __init__(self, config):
+        super(T5ForMaskedLM, self).__init__(config)
+
+        self.transformer = T5Model(config)
+        self.lm_head = nn.Linear(config.n_embd, config.vocab_size)
+
+        self.init_weights()
+
+    def get_output_embeddings(self):
+        return self.lm_head
+
+    def forward(self, input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None,
+                lm_labels=None):
+
+        outputs = self.transformer(input_ids,
+                            attention_mask=attention_mask,
+                            token_type_ids=token_type_ids,
+                            position_ids=position_ids, 
+                            head_mask=head_mask)
+
+        sequence_output = outputs[0]
+        lm_logits = self.cls(sequence_output)
+
+        outputs = (lm_logits,) + outputs[2:]  # Add hidden states and attention if they are here
+        if lm_labels is not None:
+            shift_logits = lm_logits[..., :-1, :].contiguous()
+            shift_labels = lm_labels[..., 1:].contiguous()
+            loss_fct = CrossEntropyLoss(ignore_index=-1)
+            loss = loss_fct(shift_logits.view(-1, shift_logits.size(-1)),
+                            shift_labels.view(-1))
+            outputs = (loss,) + outputs
+
+        return outputs  # (lm_loss), lm_logits, (hidden_states), (attentions)
diff --git a/transformers/modeling_tf_t5.py b/transformers/modeling_tf_t5.py
new file mode 100644
index 0000000000..deb453846c
--- /dev/null
+++ b/transformers/modeling_tf_t5.py
@@ -0,0 +1,496 @@
+# coding=utf-8
+# Copyright 2018 T5 Authors and The HuggingFace Inc. team.
+# Copyright (c) 2018, NVIDIA CORPORATION.  All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" TF 2.0 T5 model. """
+
+from __future__ import absolute_import, division, print_function, unicode_literals
+
+import json
+import logging
+import math
+import os
+import sys
+from io import open
+
+import numpy as np
+import tensorflow as tf
+
+from .configuration_t5 import T5Config
+from .modeling_tf_utils import TFPreTrainedModel, get_initializer
+from .file_utils import add_start_docstrings
+
+logger = logging.getLogger(__name__)
+
+####################################################
+# This dict contrains shortcut names and associated url
+# for the pretrained weights provided with the models
+####################################################
+TF_T5_PRETRAINED_MODEL_ARCHIVE_MAP = {
+    't5-base-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-base-uncased-tf_model.h5",
+    't5-large-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-large-uncased-tf_model.h5",
+}
+
+####################################################
+# TF 2.0 Models are constructed using Keras imperative API by sub-classing
+# - tf.keras.layers.Layer for the layers and
+# - TFPreTrainedModel for the models (it-self a sub-class of tf.keras.Model)
+####################################################
+
+####################################################
+# Here is an example of typical layer in a TF 2.0 model of the library
+# The classes are usually identical to the PyTorch ones and prefixed with 'TF'.
+#
+# Note that class __init__ parameters includes **kwargs (send to 'super').
+# This let us have a control on class scope and variable names:
+# More precisely, we set the names of the class attributes (lower level layers) to
+# to the equivalent attributes names in the PyTorch model so we can have equivalent
+# class and scope structure between PyTorch and TF 2.0 models and easily load one in the other.
+#
+# See the conversion methods in modeling_tf_pytorch_utils.py for more details
+####################################################
+class TFT5Layer(tf.keras.layers.Layer):
+    def __init__(self, config, **kwargs):
+        super(TFT5Layer, self).__init__(**kwargs)
+        self.attention = TFT5Attention(config, name='attention')
+        self.intermediate = TFT5Intermediate(config, name='intermediate')
+        self.transformer_output = TFT5Output(config, name='output')
+
+    def call(self, inputs, training=False):
+        hidden_states, attention_mask, head_mask = inputs
+
+        attention_outputs = self.attention([hidden_states, attention_mask, head_mask], training=training)
+        attention_output = attention_outputs[0]
+        intermediate_output = self.intermediate(attention_output)
+        layer_output = self.transformer_output([intermediate_output, attention_output], training=training)
+        outputs = (layer_output,) + attention_outputs[1:]  # add attentions if we output them
+        return outputs
+
+
+####################################################
+# The full model without a specific pretrained or finetuning head is
+# provided as a tf.keras.layers.Layer usually called "TFT5MainLayer"
+####################################################
+class TFT5MainLayer(tf.keras.layers.Layer):
+    def __init__(self, config, **kwargs):
+        super(TFT5MainLayer, self).__init__(**kwargs)
+
+    def _resize_token_embeddings(self, new_num_tokens):
+        raise NotImplementedError  # Not implemented yet in the library fr TF 2.0 models
+
+    def _prune_heads(self, heads_to_prune):
+        raise NotImplementedError  # Not implemented yet in the library fr TF 2.0 models
+
+    def call(self, inputs, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, training=False):
+        # We allow three types of multi-inputs:
+        # - traditional keyword arguments in the call method
+        # - all the arguments provided as a dict in the first positional argument of call
+        # - all the arguments provided as a list/tuple (ordered) in the first positional argument of call
+        # The last two options are useful to use the tf.keras fit() method.
+
+        if isinstance(inputs, (tuple, list)):
+            input_ids = inputs[0]
+            attention_mask = inputs[1] if len(inputs) > 1 else attention_mask
+            token_type_ids = inputs[2] if len(inputs) > 2 else token_type_ids
+            position_ids = inputs[3] if len(inputs) > 3 else position_ids
+            head_mask = inputs[4] if len(inputs) > 4 else head_mask
+            assert len(inputs) <= 5, "Too many inputs."
+        elif isinstance(inputs, dict):
+            input_ids = inputs.get('input_ids')
+            attention_mask = inputs.get('attention_mask', attention_mask)
+            token_type_ids = inputs.get('token_type_ids', token_type_ids)
+            position_ids = inputs.get('position_ids', position_ids)
+            head_mask = inputs.get('head_mask', head_mask)
+            assert len(inputs) <= 5, "Too many inputs."
+        else:
+            input_ids = inputs
+
+        if attention_mask is None:
+            attention_mask = tf.fill(tf.shape(input_ids), 1)
+        if token_type_ids is None:
+            token_type_ids = tf.fill(tf.shape(input_ids), 0)
+
+        # We create a 3D attention mask from a 2D tensor mask.
+        # Sizes are [batch_size, 1, 1, to_seq_length]
+        # So we can broadcast to [batch_size, num_heads, from_seq_length, to_seq_length]
+        # this attention mask is more simple than the triangular masking of causal attention
+        # used in OpenAI GPT, we just need to prepare the broadcast dimension here.
+        extended_attention_mask = attention_mask[:, tf.newaxis, tf.newaxis, :]
+
+        # Since attention_mask is 1.0 for positions we want to attend and 0.0 for
+        # masked positions, this operation will create a tensor which is 0.0 for
+        # positions we want to attend and -10000.0 for masked positions.
+        # Since we are adding it to the raw scores before the softmax, this is
+        # effectively the same as removing these entirely.
+
+        extended_attention_mask = tf.cast(extended_attention_mask, tf.float32)
+        extended_attention_mask = (1.0 - extended_attention_mask) * -10000.0
+
+        # Prepare head mask if needed
+        # 1.0 in head_mask indicate we keep the head
+        # attention_probs has shape bsz x n_heads x N x N
+        # input head_mask has shape [num_heads] or [num_hidden_layers x num_heads]
+        # and head_mask is converted to shape [num_hidden_layers x batch x num_heads x seq_length x seq_length]
+        if not head_mask is None:
+            raise NotImplementedError
+        else:
+            head_mask = [None] * self.num_hidden_layers
+            # head_mask = tf.constant([0] * self.num_hidden_layers)
+
+        ##################################
+        # Replace this with your model code
+        embedding_output = self.embeddings(input_ids, position_ids=position_ids, token_type_ids=token_type_ids)
+        encoder_outputs = self.encoder([embedding_output, extended_attention_mask, head_mask], training=training)
+        sequence_output = encoder_outputs[0]
+        outputs = (sequence_output,) + encoder_outputs[1:]  # add hidden_states and attentions if they are here
+
+        return outputs  # sequence_output, (hidden_states), (attentions)
+
+
+####################################################
+# TFT5PreTrainedModel is a sub-class of tf.keras.Model
+# which take care of loading and saving pretrained weights
+# and various common utilities.
+# Here you just need to specify a few (self-explanatory)
+# pointers for your model.
+####################################################
+class TFT5PreTrainedModel(TFPreTrainedModel):
+    """ An abstract class to handle weights initialization and
+        a simple interface for dowloading and loading pretrained models.
+    """
+    config_class = T5Config
+    pretrained_model_archive_map = TF_T5_PRETRAINED_MODEL_ARCHIVE_MAP
+    base_model_prefix = "transformer"
+
+
+T5_START_DOCSTRING = r"""    The XXX model was proposed in
+    `XXX: Pre-training of Deep Bidirectional Transformers for Language Understanding`_
+    by Jacob Devlin, Ming-Wei Chang, Kenton Lee and Kristina Toutanova. It's a bidirectional transformer
+    pre-trained using a combination of masked language modeling objective and next sentence prediction
+    on a large corpus comprising the Toronto Book Corpus and Wikipedia.
+
+    This model is a tf.keras.Model `tf.keras.Model`_ sub-class. Use it as a regular TF 2.0 Keras Model and
+    refer to the TF 2.0 documentation for all matter related to general usage and behavior.
+
+    .. _`XXX: Pre-training of Deep Bidirectional Transformers for Language Understanding`:
+        https://arxiv.org/abs/1810.04805
+
+    .. _`tf.keras.Model`:
+        https://www.tensorflow.org/versions/r2.0/api_docs/python/tf/keras/Model
+
+    Note on the model inputs:
+        TF 2.0 models accepts two formats as inputs:
+
+            - having all inputs as keyword arguments (like PyTorch models), or
+            - having all inputs as a list, tuple or dict in the first positional arguments.
+
+        This second option is usefull when using `tf.keras.Model.fit()` method which currently requires having all the tensors in the first argument of the model call function: `model(inputs)`.
+
+        If you choose this second option, there are three possibilities you can use to gather all the input Tensors in the first positional argument :
+
+        - a single Tensor with input_ids only and nothing else: `model(inputs_ids)
+        - a list of varying length with one or several input Tensors IN THE ORDER given in the docstring:
+            `model([input_ids, attention_mask])` or `model([input_ids, attention_mask, token_type_ids])`
+        - a dictionary with one or several input Tensors associaed to the input names given in the docstring:
+            `model({'input_ids': input_ids, 'token_type_ids': token_type_ids})`
+
+    Parameters:
+        config (:class:`~transformers.XxxConfig`): Model configuration class with all the parameters of the model. 
+            Initializing with a config file does not load the weights associated with the model, only the configuration.
+            Check out the :meth:`~transformers.PreTrainedModel.from_pretrained` method to load the model weights.
+"""
+
+XXX_INPUTS_DOCSTRING = r"""
+    Inputs:
+        **input_ids**: ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, sequence_length)``:
+            Indices of input sequence tokens in the vocabulary.
+            To match pre-training, XXX input sequence should be formatted with [CLS] and [SEP] tokens as follows:
+
+            (a) For sequence pairs:
+
+                ``tokens:         [CLS] is this jack ##son ##ville ? [SEP] no it is not . [SEP]``
+                
+                ``token_type_ids:   0   0  0    0    0     0       0   0   1  1  1  1   1   1``
+
+            (b) For single sequences:
+
+                ``tokens:         [CLS] the dog is hairy . [SEP]``
+                
+                ``token_type_ids:   0   0   0   0  0     0   0``
+
+            Xxx is a model with absolute position embeddings so it's usually advised to pad the inputs on
+            the right rather than the left.
+
+            Indices can be obtained using :class:`transformers.XxxTokenizer`.
+            See :func:`transformers.PreTrainedTokenizer.encode` and
+            :func:`transformers.PreTrainedTokenizer.convert_tokens_to_ids` for details.
+        **attention_mask**: (`optional`) ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, sequence_length)``:
+            Mask to avoid performing attention on padding token indices.
+            Mask values selected in ``[0, 1]``:
+            ``1`` for tokens that are NOT MASKED, ``0`` for MASKED tokens.
+        **token_type_ids**: (`optional`) ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, sequence_length)``:
+            Segment token indices to indicate first and second portions of the inputs.
+            Indices are selected in ``[0, 1]``: ``0`` corresponds to a `sentence A` token, ``1``
+            corresponds to a `sentence B` token
+            (see `XXX: Pre-training of Deep Bidirectional Transformers for Language Understanding`_ for more details).
+        **position_ids**: (`optional`) ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, sequence_length)``:
+            Indices of positions of each input sequence tokens in the position embeddings.
+            Selected in the range ``[0, config.max_position_embeddings - 1]``.
+        **head_mask**: (`optional`) ``Numpy array`` or ``tf.Tensor`` of shape ``(num_heads,)`` or ``(num_layers, num_heads)``:
+            Mask to nullify selected heads of the self-attention modules.
+            Mask values selected in ``[0, 1]``:
+            ``1`` indicates the head is **not masked**, ``0`` indicates the head is **masked**.
+"""
+
+@add_start_docstrings("The bare Xxx Model transformer outputing raw hidden-states without any specific head on top.",
+                      XXX_START_DOCSTRING, XXX_INPUTS_DOCSTRING)
+class TFXxxModel(TFXxxPreTrainedModel):
+    r"""
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **last_hidden_state**: ``tf.Tensor`` of shape ``(batch_size, sequence_length, hidden_size)``
+            Sequence of hidden-states at the output of the last layer of the model.
+        **pooler_output**: ``tf.Tensor`` of shape ``(batch_size, hidden_size)``
+            Last layer hidden-state of the first token of the sequence (classification token)
+            further processed by a Linear layer and a Tanh activation function. The Linear
+            layer weights are trained from the next sentence prediction (classification)
+            objective during Xxx pretraining. This output is usually *not* a good summary
+            of the semantic content of the input, you're often better with averaging or pooling
+            the sequence of hidden-states for the whole input sequence.
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``tf.Tensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``tf.Tensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        import tensorflow as tf
+        from transformers import XxxTokenizer, TFXxxModel
+
+        tokenizer = XxxTokenizer.from_pretrained('xxx-base-uncased')
+        model = TFXxxModel.from_pretrained('xxx-base-uncased')
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        outputs = model(input_ids)
+        last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
+
+    """
+    def __init__(self, config, *inputs, **kwargs):
+        super(TFXxxModel, self).__init__(config, *inputs, **kwargs)
+        self.transformer = TFXxxMainLayer(config, name='transformer')
+
+    def call(self, inputs, **kwargs):
+        outputs = self.transformer(inputs, **kwargs)
+        return outputs
+
+
+@add_start_docstrings("""Xxx Model with a `language modeling` head on top. """,
+    XXX_START_DOCSTRING, XXX_INPUTS_DOCSTRING)
+class TFXxxForMaskedLM(TFXxxPreTrainedModel):
+    r"""
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **prediction_scores**: ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, sequence_length, config.vocab_size)``
+            Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``Numpy array`` or ``tf.Tensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``Numpy array`` or ``tf.Tensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        import tensorflow as tf
+        from transformers import XxxTokenizer, TFXxxForMaskedLM
+
+        tokenizer = XxxTokenizer.from_pretrained('xxx-base-uncased')
+        model = TFXxxForMaskedLM.from_pretrained('xxx-base-uncased')
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        outputs = model(input_ids)
+        prediction_scores = outputs[0]
+
+    """
+    def __init__(self, config, *inputs, **kwargs):
+        super(TFXxxForMaskedLM, self).__init__(config, *inputs, **kwargs)
+
+        self.transformer = TFXxxMainLayer(config, name='transformer')
+        self.mlm = TFXxxMLMHead(config, self.transformer.embeddings, name='mlm')
+
+    def call(self, inputs, **kwargs):
+        outputs = self.transformer(inputs, **kwargs)
+
+        sequence_output = outputs[0]
+        prediction_scores = self.mlm(sequence_output, training=kwargs.get('training', False))
+
+        outputs = (prediction_scores,) + outputs[2:]  # Add hidden states and attention if they are here
+
+        return outputs  # prediction_scores, (hidden_states), (attentions)
+
+
+@add_start_docstrings("""Xxx Model transformer with a sequence classification/regression head on top (a linear layer on top of
+    the pooled output) e.g. for GLUE tasks. """,
+    XXX_START_DOCSTRING, XXX_INPUTS_DOCSTRING)
+class TFXxxForSequenceClassification(TFXxxPreTrainedModel):
+    r"""
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **logits**: ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, config.num_labels)``
+            Classification (or regression if config.num_labels==1) scores (before SoftMax).
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``Numpy array`` or ``tf.Tensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``Numpy array`` or ``tf.Tensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        import tensorflow as tf
+        from transformers import XxxTokenizer, TFXxxForSequenceClassification
+
+        tokenizer = XxxTokenizer.from_pretrained('xxx-base-uncased')
+        model = TFXxxForSequenceClassification.from_pretrained('xxx-base-uncased')
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        outputs = model(input_ids)
+        logits = outputs[0]
+
+    """
+    def __init__(self, config, *inputs, **kwargs):
+        super(TFXxxForSequenceClassification, self).__init__(config, *inputs, **kwargs)
+        self.num_labels = config.num_labels
+
+        self.transformer = TFXxxMainLayer(config, name='transformer')
+        self.dropout = tf.keras.layers.Dropout(config.hidden_dropout_prob)
+        self.classifier = tf.keras.layers.Dense(config.num_labels,
+                                                kernel_initializer=get_initializer(config.initializer_range),
+                                                name='classifier')
+
+    def call(self, inputs, **kwargs):
+        outputs = self.transformer(inputs, **kwargs)
+
+        pooled_output = outputs[1]
+
+        pooled_output = self.dropout(pooled_output, training=kwargs.get('training', False))
+        logits = self.classifier(pooled_output)
+
+        outputs = (logits,) + outputs[2:]  # add hidden states and attention if they are here
+
+        return outputs  # logits, (hidden_states), (attentions)
+
+
+@add_start_docstrings("""Xxx Model with a token classification head on top (a linear layer on top of
+    the hidden-states output) e.g. for Named-Entity-Recognition (NER) tasks. """,
+    XXX_START_DOCSTRING, XXX_INPUTS_DOCSTRING)
+class TFXxxForTokenClassification(TFXxxPreTrainedModel):
+    r"""
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **scores**: ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, sequence_length, config.num_labels)``
+            Classification scores (before SoftMax).
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``Numpy array`` or ``tf.Tensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``Numpy array`` or ``tf.Tensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        import tensorflow as tf
+        from transformers import XxxTokenizer, TFXxxForTokenClassification
+
+        tokenizer = XxxTokenizer.from_pretrained('xxx-base-uncased')
+        model = TFXxxForTokenClassification.from_pretrained('xxx-base-uncased')
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        outputs = model(input_ids)
+        scores = outputs[0]
+
+    """
+    def __init__(self, config, *inputs, **kwargs):
+        super(TFXxxForTokenClassification, self).__init__(config, *inputs, **kwargs)
+        self.num_labels = config.num_labels
+
+        self.transformer = TFXxxMainLayer(config, name='transformer')
+        self.dropout = tf.keras.layers.Dropout(config.hidden_dropout_prob)
+        self.classifier = tf.keras.layers.Dense(config.num_labels,
+                                                kernel_initializer=get_initializer(config.initializer_range),
+                                                name='classifier')
+
+    def call(self, inputs, **kwargs):
+        outputs = self.transformer(inputs, **kwargs)
+
+        sequence_output = outputs[0]
+
+        sequence_output = self.dropout(sequence_output, training=kwargs.get('training', False))
+        logits = self.classifier(sequence_output)
+
+        outputs = (logits,) + outputs[2:]  # add hidden states and attention if they are here
+
+        return outputs  # scores, (hidden_states), (attentions)
+
+
+@add_start_docstrings("""Xxx Model with a span classification head on top for extractive question-answering tasks like SQuAD (a linear layers on top of
+    the hidden-states output to compute `span start logits` and `span end logits`). """,
+    XXX_START_DOCSTRING, XXX_INPUTS_DOCSTRING)
+class TFXxxForQuestionAnswering(TFXxxPreTrainedModel):
+    r"""
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **start_scores**: ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, sequence_length,)``
+            Span-start scores (before SoftMax).
+        **end_scores**: ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, sequence_length,)``
+            Span-end scores (before SoftMax).
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``Numpy array`` or ``tf.Tensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``Numpy array`` or ``tf.Tensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        import tensorflow as tf
+        from transformers import XxxTokenizer, TFXxxForQuestionAnswering
+
+        tokenizer = XxxTokenizer.from_pretrained('xxx-base-uncased')
+        model = TFXxxForQuestionAnswering.from_pretrained('xxx-base-uncased')
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        outputs = model(input_ids)
+        start_scores, end_scores = outputs[:2]
+
+    """
+    def __init__(self, config, *inputs, **kwargs):
+        super(TFXxxForQuestionAnswering, self).__init__(config, *inputs, **kwargs)
+        self.num_labels = config.num_labels
+
+        self.transformer = TFXxxMainLayer(config, name='transformer')
+        self.qa_outputs = tf.keras.layers.Dense(config.num_labels,
+                                                kernel_initializer=get_initializer(config.initializer_range),
+                                                name='qa_outputs')
+
+    def call(self, inputs, **kwargs):
+        outputs = self.transformer(inputs, **kwargs)
+
+        sequence_output = outputs[0]
+
+        logits = self.qa_outputs(sequence_output)
+        start_logits, end_logits = tf.split(logits, 2, axis=-1)
+        start_logits = tf.squeeze(start_logits, axis=-1)
+        end_logits = tf.squeeze(end_logits, axis=-1)
+
+        outputs = (start_logits, end_logits,) + outputs[2:]
+
+        return outputs  # start_logits, end_logits, (hidden_states), (attentions)
diff --git a/transformers/tokenization_t5.py b/transformers/tokenization_t5.py
new file mode 100644
index 0000000000..3f8f4bf556
--- /dev/null
+++ b/transformers/tokenization_t5.py
@@ -0,0 +1,214 @@
+# coding=utf-8
+# Copyright 2018 T5 Authors and HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" Tokenization class for model T5."""
+
+from __future__ import absolute_import, division, print_function, unicode_literals
+
+import collections
+import logging
+import os
+import unicodedata
+from io import open
+
+from .tokenization_utils import PreTrainedTokenizer
+
+logger = logging.getLogger(__name__)
+
+####################################################
+# Mapping from the keyword arguments names of Tokenizer `__init__`
+# to file names for serializing Tokenizer instances
+####################################################
+VOCAB_FILES_NAMES = {'vocab_file': 'vocab.txt'}
+
+####################################################
+# Mapping from the keyword arguments names of Tokenizer `__init__`
+# to pretrained vocabulary URL for all the model shortcut names.
+####################################################
+PRETRAINED_VOCAB_FILES_MAP = {
+    'vocab_file':
+    {
+        't5-base-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-base-uncased-vocab.txt",
+        't5-large-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-large-uncased-vocab.txt",
+    }
+}
+
+####################################################
+# Mapping from model shortcut names to max length of inputs
+####################################################
+PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
+    't5-base-uncased': 512,
+    't5-large-uncased': 512,
+}
+
+####################################################
+# Mapping from model shortcut names to a dictionary of additional
+# keyword arguments for Tokenizer `__init__`.
+# To be used for checkpoint specific configurations.
+####################################################
+PRETRAINED_INIT_CONFIGURATION = {
+    't5-base-uncased': {'do_lower_case': True},
+    't5-large-uncased': {'do_lower_case': True},
+}
+
+
+def load_vocab(vocab_file):
+    """Loads a vocabulary file into a dictionary."""
+    vocab = collections.OrderedDict()
+    with open(vocab_file, "r", encoding="utf-8") as reader:
+        tokens = reader.readlines()
+    for index, token in enumerate(tokens):
+        token = token.rstrip('\n')
+        vocab[token] = index
+    return vocab
+
+
+class T5Tokenizer(PreTrainedTokenizer):
+    r"""
+    Constructs a T5Tokenizer.
+    :class:`~transformers.T5Tokenizer` runs end-to-end tokenization: punctuation splitting + wordpiece
+
+    Args:
+        vocab_file: Path to a one-wordpiece-per-line vocabulary file
+        do_lower_case: Whether to lower case the input. Only has an effect when do_wordpiece_only=False
+    """
+
+    vocab_files_names = VOCAB_FILES_NAMES
+    pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+    pretrained_init_configuration = PRETRAINED_INIT_CONFIGURATION
+    max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
+
+    def __init__(self, vocab_file, do_lower_case=True,
+                 unk_token="[UNK]", sep_token="[SEP]", pad_token="[PAD]", cls_token="[CLS]",
+                 mask_token="[MASK]", **kwargs):
+        """Constructs a T5Tokenizer.
+
+        Args:
+            **vocab_file**: Path to a one-wordpiece-per-line vocabulary file
+            **do_lower_case**: (`optional`) boolean (default True)
+                Whether to lower case the input
+                Only has an effect when do_basic_tokenize=True
+        """
+        super(T5Tokenizer, self).__init__(unk_token=unk_token, sep_token=sep_token,
+                                           pad_token=pad_token, cls_token=cls_token,
+                                           mask_token=mask_token, **kwargs)
+        self.max_len_single_sentence = self.max_len - 2  # take into account special tokens
+        self.max_len_sentences_pair = self.max_len - 3  # take into account special tokens
+
+        if not os.path.isfile(vocab_file):
+            raise ValueError(
+                "Can't find a vocabulary file at path '{}'. To load the vocabulary from a Google pretrained "
+                "model use `tokenizer = T5Tokenizer.from_pretrained(PRETRAINED_MODEL_NAME)`".format(vocab_file))
+        self.vocab = load_vocab(vocab_file)
+
+    @property
+    def vocab_size(self):
+        return len(self.vocab)
+
+    def _tokenize(self, text):
+        """ Take as input a string and return a list of strings (tokens) for words/sub-words
+        """
+        split_tokens = []
+        if self.do_basic_tokenize:
+            for token in self.basic_tokenizer.tokenize(text, never_split=self.all_special_tokens):
+                for sub_token in self.wordpiece_tokenizer.tokenize(token):
+                    split_tokens.append(sub_token)
+        else:
+            split_tokens = self.wordpiece_tokenizer.tokenize(text)
+        return split_tokens
+
+    def _convert_token_to_id(self, token):
+        """ Converts a token (str/unicode) in an id using the vocab. """
+        return self.vocab.get(token, self.vocab.get(self.unk_token))
+
+    def _convert_id_to_token(self, index):
+        """Converts an index (integer) in a token (string/unicode) using the vocab."""
+        return self.ids_to_tokens.get(index, self.unk_token)
+
+    def convert_tokens_to_string(self, tokens):
+        """ Converts a sequence of tokens (string) in a single string. """
+        out_string = ' '.join(tokens).replace(' ##', '').strip()
+        return out_string
+
+    def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
+        """
+        Build model inputs from a sequence or a pair of sequence for sequence classification tasks
+        by concatenating and adding special tokens.
+        A BERT sequence has the following format:
+            single sequence: [CLS] X [SEP]
+            pair of sequences: [CLS] A [SEP] B [SEP]
+        """
+        if token_ids_1 is None:
+            return [self.cls_token_id] + token_ids_0 + [self.sep_token_id]
+        cls = [self.cls_token_id]
+        sep = [self.sep_token_id]
+        return cls + token_ids_0 + sep + token_ids_1 + sep
+
+    def get_special_tokens_mask(self, token_ids_0, token_ids_1=None, already_has_special_tokens=False):
+        """
+        Retrieves sequence ids from a token list that has no special tokens added. This method is called when adding
+        special tokens using the tokenizer ``prepare_for_model`` or ``encode_plus`` methods.
+
+        Args:
+            token_ids_0: list of ids (must not contain special tokens)
+            token_ids_1: Optional list of ids (must not contain special tokens), necessary when fetching sequence ids
+                for sequence pairs
+            already_has_special_tokens: (default False) Set to True if the token list is already formated with
+                special tokens for the model
+
+        Returns:
+            A list of integers in the range [0, 1]: 0 for a special token, 1 for a sequence token.
+        """
+
+        if already_has_special_tokens:
+            if token_ids_1 is not None:
+                raise ValueError("You should not supply a second sequence if the provided sequence of "
+                                 "ids is already formated with special tokens for the model.")
+            return list(map(lambda x: 1 if x in [self.sep_token_id, self.cls_token_id] else 0, token_ids_0))
+
+        if token_ids_1 is not None:
+            return [1] + ([0] * len(token_ids_0)) + [1] + ([0] * len(token_ids_1)) + [1]
+        return [1] + ([0] * len(token_ids_0)) + [1]
+
+    def create_token_type_ids_from_sequences(self, token_ids_0, token_ids_1=None):
+        """
+        Creates a mask from the two sequences passed to be used in a sequence-pair classification task.
+        A BERT sequence pair mask has the following format:
+        0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1
+        | first sequence    | second sequence
+
+        if token_ids_1 is None, only returns the first portion of the mask (0's).
+        """
+        sep = [self.sep_token_id]
+        cls = [self.cls_token_id]
+        if token_ids_1 is None:
+            return len(cls + token_ids_0 + sep) * [0]
+        return len(cls + token_ids_0 + sep) * [0] + len(token_ids_1 + sep) * [1]
+
+    def save_vocabulary(self, vocab_path):
+        """Save the tokenizer vocabulary to a directory or file."""
+        index = 0
+        if os.path.isdir(vocab_path):
+            vocab_file = os.path.join(vocab_path, VOCAB_FILES_NAMES['vocab_file'])
+        else:
+            vocab_file = vocab_path
+        with open(vocab_file, "w", encoding="utf-8") as writer:
+            for token, token_index in sorted(self.vocab.items(), key=lambda kv: kv[1]):
+                if index != token_index:
+                    logger.warning("Saving vocabulary to {}: vocabulary indices are not consecutive."
+                                   " Please check that the vocabulary is not corrupted!".format(vocab_file))
+                    index = token_index
+                writer.write(token + u'\n')
+                index += 1
+        return (vocab_file,)

From 568c0ffb7ef73555567f8bd467cf80c2b1e6ac13 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Tue, 5 Nov 2019 16:40:29 +0100
Subject: [PATCH 011/505] adding T5 model

---
 transformers/modeling_encoder_decoder.py |   4 +-
 transformers/modeling_t5.py              | 471 ++++++++++++++++++++---
 2 files changed, 412 insertions(+), 63 deletions(-)

diff --git a/transformers/modeling_encoder_decoder.py b/transformers/modeling_encoder_decoder.py
index a884abd0a2..713cf5252e 100644
--- a/transformers/modeling_encoder_decoder.py
+++ b/transformers/modeling_encoder_decoder.py
@@ -217,9 +217,7 @@ class PreTrainedEncoderDecoder(nn.Module):
         encoder_hidden_states = kwargs_encoder.pop("hidden_states", None)
         if encoder_hidden_states is None:
             encoder_outputs = self.encoder(encoder_input_ids, **kwargs_encoder)
-            encoder_hidden_states = encoder_outputs[
-                0
-            ]  # output the last layer hidden state
+            encoder_hidden_states = encoder_outputs[0]
         else:
             encoder_outputs = ()
 
diff --git a/transformers/modeling_t5.py b/transformers/modeling_t5.py
index fa3c22f24b..d93e96211d 100644
--- a/transformers/modeling_t5.py
+++ b/transformers/modeling_t5.py
@@ -1,5 +1,5 @@
 # coding=utf-8
-# Copyright 2018 T5 Authors and HuggingFace Inc. team.
+# Copyright 2018 Mesh TensorFlow authors, T5 Authors and HuggingFace Inc. team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -20,11 +20,14 @@ import json
 import logging
 import math
 import os
+import math
 import sys
+import itertools
 from io import open
 
 import torch
 from torch import nn
+import torch.nn.functional as F
 from torch.nn import CrossEntropyLoss, MSELoss
 
 from .modeling_utils import PreTrainedModel, prune_linear_layer
@@ -119,31 +122,389 @@ def load_tf_weights_in_t5(model, config, tf_checkpoint_path):
 # - PreTrainedModel for the models (it-self a sub-class of torch.nn.Module)
 ####################################################
 
-class T5Layer(nn.Module):
+class T5DenseReluDense(nn.Module):
     def __init__(self, config):
-        super(T5Layer, self).__init__()
-        self.attention = T5Attention(config)
-        self.intermediate = T5Intermediate(config)
-        self.output = T5Output(config)
+        super(T5DenseReluDense, self).__init__()
+        self.wi = nn.Linear(config.d_model, config.d_ff, bias=False)
+        self.wo = nn.Linear(config.d_ff, config.d_model, bias=False)
+        self.dropout = nn.Dropout(config.dropout)
 
-    def forward(self, hidden_states, attention_mask=None, head_mask=None):
-        attention_outputs = self.attention(hidden_states, attention_mask, head_mask)
-        attention_output = attention_outputs[0]
-        intermediate_output = self.intermediate(attention_output)
-        layer_output = self.output(intermediate_output, attention_output)
-        outputs = (layer_output,) + attention_outputs[1:]  # add attentions if we output them
+    def forward(self, hidden_states):
+        h = self.wi(hidden_states)
+        h = F.relu(h)
+        h = self.dropout(h)
+        h = self.wo(h)
+        return h
+
+
+class T5LayerFF(nn.Module):
+    def __init__(self, config):
+        super(T5LayerFF, self).__init__()
+        self.DenseReluDense = T5DenseReluDense(config)
+        self.layer_norm = nn.LayerNorm(config.layer_norm_epsilon)
+        self.dropout = nn.Dropout(config.dropout)
+
+    def forward(self, hidden_states):
+        norm_x = self.layer_norm(hidden_states)
+        y = self.DenseReluDense(norm_x)
+        layer_output = hidden_states + self.dropout(y)
+        return layer_output
+
+
+class T5Attention(nn.Module):
+    NEW_ID = itertools.count()
+
+    def __init__(self, config):
+        super(T5Attention, self).__init__()
+        self.layer_id = next(T5Attention.NEW_ID)
+
+        self.output_attentions = config.output_attentions
+        self.relative_attention_num_buckets = config.relative_attention_num_buckets
+        self.dim = config.d_model
+        self.n_heads = config.num_heads
+        self.dropout = config.dropout_rate
+        assert self.dim % self.n_heads == 0
+
+        self.q = nn.Linear(self.dim, self.dim, bias=False)
+        self.k = nn.Linear(self.dim, self.dim, bias=False)
+        self.v = nn.Linear(self.dim, self.dim, bias=False)
+        self.o = nn.Linear(self.dim, self.dim, bias=False)
+
+        self.relative_attention_bias = nn.Embedding(self.relative_attention_num_buckets, self.n_heads)
+        self.pruned_heads = set()
+
+    def prune_heads(self, heads):
+        attention_head_size = self.dim // self.n_heads
+        if len(heads) == 0:
+            return
+        mask = torch.ones(self.n_heads, attention_head_size)
+        heads = set(heads) - self.pruned_heads
+        for head in heads:
+            head -= sum(1 if h < head else 0 for h in self.pruned_heads)
+            mask[head] = 0
+        mask = mask.view(-1).contiguous().eq(1)
+        index = torch.arange(len(mask))[mask].long()
+        # Prune linear layers
+        self.q = prune_linear_layer(self.q, index)
+        self.k = prune_linear_layer(self.k, index)
+        self.v = prune_linear_layer(self.v, index)
+        self.o = prune_linear_layer(self.o, index, dim=1)
+        # Update hyper params
+        self.n_heads = self.n_heads - len(heads)
+        self.dim = attention_head_size * self.n_heads
+        self.pruned_heads = self.pruned_heads.union(heads)
+
+    @staticmethod
+    def _relative_position_bucket(relative_position,
+                                  bidirectional=True,
+                                  num_buckets=32,
+                                  max_distance=128):
+        """
+        Adapted from Mesh Tensorflow:
+        https://github.com/tensorflow/mesh/blob/0cb87fe07da627bf0b7e60475d59f95ed6b5be3d/mesh_tensorflow/transformer/transformer_layers.py#L593
+
+        Translate relative position to a bucket number for relative attention.
+        The relative position is defined as memory_position - query_position, i.e.
+        the distance in tokens from the attending position to the attended-to
+        position.  If bidirectional=False, then positive relative positions are
+        invalid.
+        We use smaller buckets for small absolute relative_position and larger buckets
+        for larger absolute relative_positions.  All relative positions >=max_distance
+        map to the same bucket.  All relative positions <=-max_distance map to the
+        same bucket.  This should allow for more graceful generalization to longer
+        sequences than the model has been trained on.
+        Args:
+            relative_position: an int32 Tensor
+            bidirectional: a boolean - whether the attention is bidirectional
+            num_buckets: an integer
+            max_distance: an integer
+        Returns:
+            a Tensor with the same shape as relative_position, containing int32
+            values in the range [0, num_buckets)
+        """
+        ret = 0
+        n = -relative_position
+        if bidirectional:
+            num_buckets //= 2
+            ret += (n < 0).to(torch.long) * num_buckets  # mtf.to_int32(mtf.less(n, 0)) * num_buckets
+            n = torch.abs(n)
+        else:
+            n = torch.max(n, 0)
+        # now n is in the range [0, inf)
+
+        # half of the buckets are for exact increments in positions
+        max_exact = num_buckets // 2
+        is_small = (n < max_exact)
+
+        # The other half of the buckets are for logarithmically bigger bins in positions up to max_distance
+        val_if_large = max_exact + (
+            torch.log(n.float() / max_exact)
+            / math.log(max_distance / max_exact) * (num_buckets - max_exact)).to(torch.long)
+        val_if_large = torch.min(val_if_large, num_buckets - 1)
+
+        ret += torch.where(is_small, n, val_if_large)
+        return ret
+
+    def compute_bias(self, qlen, klen):
+        """ Compute binned relative position bias """
+        context_position = torch.arange(qlen, dtype=torch.long)[:, None]
+        memory_position = torch.arange(klen, dtype=torch.long)[None, :]
+        relative_position = memory_position - context_position  # shape (qlen, klen)
+        rp_bucket = self._relative_position_bucket(relative_position,
+                                                   bidirectional=not self.is_decoder,
+                                                   num_buckets=self.relative_attention_num_buckets)
+        values = self.relative_attention_bias(rp_bucket)  # shape (qlen, klen, num_heads)
+        values = values.permute([2, 0, 1]).unsqueeze(0) # shape (1, num_heads, qlen, klen)
+        return values
+
+    def forward(self, input, mask, kv=None, position_bias=None, cache=None, head_mask=None):
+        """
+        Self-attention (if kv is None) or attention over source sentence (provided by kv).
+        """
+        # Input is (bs, qlen, dim)
+        # Mask is (bs, klen) (non-causal) or (bs, klen, klen)
+        bs, qlen, dim = input.size()
+        if kv is None:
+            klen = qlen if cache is None else cache['slen'] + qlen
+        else:
+            klen = kv.size(1)
+        # assert dim == self.dim, 'Dimensions do not match: %s input vs %s configured' % (dim, self.dim)
+        n_heads = self.n_heads
+        dim_per_head = self.dim // n_heads
+        mask_reshape = (bs, 1, qlen, klen) if mask.dim() == 3 else (bs, 1, 1, klen)
+
+        def shape(x):
+            """  projection """
+            return x.view(bs, -1, self.n_heads, dim_per_head).transpose(1, 2)
+
+        def unshape(x):
+            """  compute context """
+            return x.transpose(1, 2).contiguous().view(bs, -1, self.n_heads * dim_per_head)
+
+        q = shape(self.q(input))                                          # (bs, n_heads, qlen, dim_per_head)
+        if kv is None:
+            k = shape(self.k(input))                                      # (bs, n_heads, qlen, dim_per_head)
+            v = shape(self.v(input))                                      # (bs, n_heads, qlen, dim_per_head)
+        elif cache is None or self.layer_id not in cache:
+            k = v = kv
+            k = shape(self.k(k))                                          # (bs, n_heads, qlen, dim_per_head)
+            v = shape(self.v(v))                                          # (bs, n_heads, qlen, dim_per_head)
+
+        if cache is not None:
+            if self.layer_id in cache:
+                if kv is None:
+                    k_, v_ = cache[self.layer_id]
+                    k = torch.cat([k_, k], dim=2)                             # (bs, n_heads, klen, dim_per_head)
+                    v = torch.cat([v_, v], dim=2)                             # (bs, n_heads, klen, dim_per_head)
+                else:
+                    k, v = cache[self.layer_id]
+            cache[self.layer_id] = (k, v)
+
+        # q = q / math.sqrt(dim_per_head)                                     # No scaling in T5
+        scores = torch.matmul(q, k.transpose(2, 3))                           # (bs, n_heads, qlen, klen)
+
+        if position_bias is None:
+            position_bias = self.compute_bias(qlen, klen)
+        scores += position_bias
+
+        mask = (mask == 0).view(mask_reshape).expand_as(scores)               # (bs, n_heads, qlen, klen)
+        scores.masked_fill_(mask, -float('inf'))                              # (bs, n_heads, qlen, klen)
+
+        weights = F.softmax(scores.float(), dim=-1).type_as(scores)           # (bs, n_heads, qlen, klen)
+        weights = F.dropout(weights, p=self.dropout, training=self.training)  # (bs, n_heads, qlen, klen)
+
+        # Mask heads if we want to
+        if head_mask is not None:
+            weights = weights * head_mask
+
+        context = torch.matmul(weights, v)                                    # (bs, n_heads, qlen, dim_per_head)
+        context = unshape(context)                                            # (bs, qlen, dim)
+
+        context = self.o(context)
+
+        outputs = (context,)
+        if self.output_attentions:
+            outputs = outputs + (weights,)
         return outputs
 
 
+class T5LayerSelfAttention(nn.Module):
+    def __init__(self, config):
+        super(T5LayerSelfAttention, self).__init__()
+        self.SelfAttention = T5Attention(config)
+        self.layer_norm = nn.LayerNorm(config.layer_norm_epsilon)
+        self.dropout = nn.Dropout(config.dropout)
 
-class T5PreTrainedModel(PreTrainedModel):
+    def forward(self, hidden_states, attention_mask=None, head_mask=None):
+        norm_x = self.layer_norm(hidden_states)
+        attention_output = self.SelfAttention(norm_x,
+                                              attention_mask=attention_mask,
+                                              head_mask=head_mask)
+        y = attention_output[0]
+        layer_output = hidden_states + self.dropout(y)
+        outputs = (layer_output,) + attention_output[1:]  # add attentions if we output them
+        return outputs
+
+
+class T5LayerCrossAttention(nn.Module):
+    def __init__(self, config):
+        super(T5LayerCrossAttention, self).__init__()
+        self.EncDecAttention = T5Attention(config)
+        self.layer_norm = nn.LayerNorm(config.layer_norm_epsilon)
+        self.dropout = nn.Dropout(config.dropout)
+
+    def forward(self, hidden_states, kv, attention_mask=None, head_mask=None):
+        norm_x = self.layer_norm(hidden_states)
+        attention_output = self.EncDecAttention(norm_x,
+                                                kv=kv,
+                                                attention_mask=attention_mask,
+                                                head_mask=head_mask)
+        y = attention_output[0]
+        layer_output = hidden_states + self.dropout(y)
+        outputs = (layer_output,) + attention_output[1:]  # add attentions if we output them
+        return outputs
+
+
+class T5Block(nn.Module):
+    def __init__(self, config):
+        super(T5Block, self).__init__()
+        self.is_decoder = config.is_decoder
+        self.layer_000 = T5LayerSelfAttention(config)
+        if self.is_decoder:
+            self.layer_001 = T5LayerCrossAttention(config)
+            self.layer_002 = T5LayerFF(config)
+        else:
+            self.layer_001 = T5LayerFF(config)
+
+    def forward(self, hidden_states, attention_mask=None,
+                encoder_hidden_states=None, encoder_attention_mask=None, head_mask=None):
+        self_attention_outputs = self.layer_000(hidden_states,
+                                                attention_mask=attention_mask,
+                                                head_mask=head_mask)
+        hidden_states = self_attention_outputs[0]
+        outputs = self_attention_outputs[1:]
+
+        if self.is_decoder:
+            cross_attention_outputs = self.layer_001(hidden_states,
+                                                     kv=encoder_hidden_states,
+                                                     attention_mask=encoder_attention_mask,
+                                                     head_mask=head_mask)
+            hidden_states = cross_attention_outputs[0]
+            outputs = cross_attention_outputs[1:] + outputs
+            hidden_states = self.layer_002(hidden_states)
+        else:
+            hidden_states = self.layer_001(hidden_states)
+
+        outputs = (hidden_states,) + outputs  # add attentions if we output them
+        return outputs
+
+
+class T5Stack(nn.Module):
+    def __init__(self, config):
+        super(T5Stack, self).__init__()
+        self.blocks = nn.ModuleList([T5Block(config) for _ in range(config.num_layers)])
+        self.final_layer_norm = nn.LayerNorm(config.layer_norm_epsilon)
+        self.dropout = nn.Dropout(config.dropout)
+
+    def forward(self,
+                hidden_states,
+                attention_mask=None,
+                encoder_hidden_states=None,
+                encoder_attention_mask=None,
+                head_mask=None):
+
+        if attention_mask is None:
+            attention_mask = torch.ones_like(input_ids)
+
+        # We can provide a self-attention mask of dimensions [batch_size, from_seq_length, to_seq_length]
+        # ourselves in which case we just need to make it broadcastable to all heads.
+        if attention_mask.dim() == 3:
+            extended_attention_mask = attention_mask[:, None, :, :]
+
+        # Provided a padding mask of dimensions [batch_size, seq_length]
+        # - if the model is a decoder, apply a causal mask in addition to the padding mask
+        # - if the model is an encoder, make the mask broadcastable to [batch_size, num_heads, seq_length, seq_length]
+        if attention_mask.dim() == 2:
+            if self.config.is_decoder:
+                batch_size, seq_length = input_ids.size()
+                seq_ids = torch.arange(seq_length, device=input_ids.device)
+                causal_mask = seq_ids[None, None, :].repeat(batch_size, seq_length, 1) <= seq_ids[None, :, None]
+                extended_attention_mask = causal_mask[:, None, :, :] * attention_mask[:, None, None, :]
+            else:
+                extended_attention_mask = attention_mask[:, None, None, :]
+
+        # Since attention_mask is 1.0 for positions we want to attend and 0.0 for
+        # masked positions, this operation will create a tensor which is 0.0 for
+        # positions we want to attend and -10000.0 for masked positions.
+        # Since we are adding it to the raw scores before the softmax, this is
+        # effectively the same as removing these entirely.
+        extended_attention_mask = extended_attention_mask.to(dtype=next(self.parameters()).dtype)  # fp16 compatibility
+        extended_attention_mask = (1.0 - extended_attention_mask) * -10000.0
+
+        # If a 2D ou 3D attention mask is provided for the cross-attention
+        # we need to make broadcastabe to [batch_size, num_heads, seq_length, seq_length]
+        if encoder_attention_mask.dim() == 3:
+            encoder_extended_attention_mask = encoder_attention_mask[:, None, :, :]
+        if encoder_attention_mask.dim() == 2:
+            encoder_extended_attention_mask = encoder_attention_mask[:, None, None, :]
+
+        encoder_extended_attention_mask = encoder_extended_attention_mask.to(dtype=next(self.parameters()).dtype)  # fp16 compatibility
+        encoder_extended_attention_mask = (1.0 - encoder_extended_attention_mask) * -10000.0
+
+        # Prepare head mask if needed
+        # 1.0 in head_mask indicate we keep the head
+        # attention_probs has shape bsz x n_heads x N x N
+        # input head_mask has shape [num_heads] or [num_hidden_layers x num_heads]
+        # and head_mask is converted to shape [num_hidden_layers x batch x num_heads x seq_length x seq_length]
+        if head_mask is not None:
+            if head_mask.dim() == 1:
+                head_mask = head_mask.unsqueeze(0).unsqueeze(0).unsqueeze(-1).unsqueeze(-1)
+                head_mask = head_mask.expand(self.config.num_hidden_layers, -1, -1, -1, -1)
+            elif head_mask.dim() == 2:
+                head_mask = head_mask.unsqueeze(1).unsqueeze(-1).unsqueeze(-1)  # We can specify head_mask for each layer
+            head_mask = head_mask.to(dtype=next(self.parameters()).dtype) # switch to fload if need + fp16 compatibility
+        else:
+            head_mask = [None] * self.config.num_hidden_layers
+
+        all_hidden_states = ()
+        all_attentions = ()
+        position_bias = None
+        for i, layer_module in enumerate(self.layer):
+            if self.output_hidden_states:
+                all_hidden_states = all_hidden_states + (hidden_states,)
+
+            layer_outputs = layer_module(hidden_states,
+                                         attention_mask=extended_attention_mask,
+                                         encoder_hidden_states=encoder_hidden_states,
+                                         encoder_attention_mask=encoder_extended_attention_mask,
+                                         head_mask=head_mask[i])
+            hidden_states = layer_outputs[0]
+
+            if self.output_attentions:
+                all_attentions = all_attentions + (layer_outputs[1],)
+
+        hidden_states = self.final_layer_norm(hidden_states)
+        layer_output = self.dropout(hidden_states)
+
+        # Add last layer
+        if self.output_hidden_states:
+            all_hidden_states = all_hidden_states + (hidden_states,)
+
+        outputs = (hidden_states,)
+        if self.output_hidden_states:
+            outputs = outputs + (all_hidden_states,)
+        if self.output_attentions:
+            outputs = outputs + (all_attentions,)
+        return outputs  # last-layer hidden state, (all hidden states), (all attentions)
+
+
+class T5PreTrainedModel(PreTrainedEncoderDecoder):
     """ An abstract class to handle weights initialization and
         a simple interface for dowloading and loading pretrained models.
     """
     config_class = T5Config
     pretrained_model_archive_map = T5_PRETRAINED_MODEL_ARCHIVE_MAP
     load_tf_weights = load_tf_weights_in_t5
-    base_model_prefix = "transformer"
 
     def _init_weights(self, module):
         """ Initialize the weights """
@@ -238,19 +599,23 @@ class T5Model(T5PreTrainedModel):
     """
     def __init__(self, config):
         super(T5Model, self).__init__(config)
+        self.shared = nn.Embeddings(config.vocab_size, config.d_model)
 
-        self.embeddings = T5Embeddings(config)
-        self.encoder = T5Encoder(config)
-        self.pooler = T5Pooler(config)
+        encoder_config = copy.deepcopy(config)
+        self.encoder = T5Stack(encoder_config)
+
+        decoder_config = copy.deepcopy(config)
+        decoder_config.is_decoder = True
+        self.decoder = T5Stack(decoder_config)
 
         self.init_weights()
 
     @property
     def get_input_embeddings(self):
-        return self.embeddings.word_embeddings
+        return self.shared
 
     def set_input_embeddings(self, new_embeddings):
-        self.embeddings.word_embeddings = new_embeddings
+        self.shared = new_embeddings
 
     def _prune_heads(self, heads_to_prune):
         """ Prunes heads of the model.
@@ -260,50 +625,36 @@ class T5Model(T5PreTrainedModel):
         for layer, heads in heads_to_prune.items():
             self.encoder.layer[layer].attention.prune_heads(heads)
 
-    def forward(self, input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None):
-        if attention_mask is None:
-            attention_mask = torch.ones_like(input_ids)
-        if token_type_ids is None:
-            token_type_ids = torch.zeros_like(input_ids)
+    def forward(self, encoder_input_ids, decoder_input_ids, **kwargs):
+        # keyword arguments come in 3 flavors: encoder-specific (prefixed by
+        # `encoder_`), decoder-specific (prefixed by `decoder_`) and those
+        # that apply to the model as whole.
+        # We let the specific kwargs override the common ones in case of conflict.
+        kwargs_common = dict((k, v) for k, v in kwargs.items()
+                             if not k.startswith("encoder_") and not k.startswith("decoder_"))
+        kwargs_decoder = kwargs_common.copy()
+        kwargs_encoder = kwargs_common.copy()
+        kwargs_encoder.update(dict((k[len("encoder_") :], v) for k, v in kwargs.items() if k.startswith("encoder_")))
+        kwargs_decoder.update(dict((k[len("decoder_") :], v) for k, v in kwargs.items() if k.startswith("decoder_")))
 
-        # We create a 3D attention mask from a 2D tensor mask.
-        # Sizes are [batch_size, 1, 1, to_seq_length]
-        # So we can broadcast to [batch_size, num_heads, from_seq_length, to_seq_length]
-        # this attention mask is more simple than the triangular masking of causal attention
-        # used in OpenAI GPT, we just need to prepare the broadcast dimension here.
-        extended_attention_mask = attention_mask.unsqueeze(1).unsqueeze(2)
-
-        # Since attention_mask is 1.0 for positions we want to attend and 0.0 for
-        # masked positions, this operation will create a tensor which is 0.0 for
-        # positions we want to attend and -10000.0 for masked positions.
-        # Since we are adding it to the raw scores before the softmax, this is
-        # effectively the same as removing these entirely.
-        extended_attention_mask = extended_attention_mask.to(dtype=next(self.parameters()).dtype) # fp16 compatibility
-        extended_attention_mask = (1.0 - extended_attention_mask) * -10000.0
-
-        # Prepare head mask if needed
-        # 1.0 in head_mask indicate we keep the head
-        # attention_probs has shape bsz x n_heads x N x N
-        # input head_mask has shape [num_heads] or [num_hidden_layers x num_heads]
-        # and head_mask is converted to shape [num_hidden_layers x batch x num_heads x seq_length x seq_length]
-        if head_mask is not None:
-            if head_mask.dim() == 1:
-                head_mask = head_mask.unsqueeze(0).unsqueeze(0).unsqueeze(-1).unsqueeze(-1)
-                head_mask = head_mask.expand(self.config.num_hidden_layers, -1, -1, -1, -1)
-            elif head_mask.dim() == 2:
-                head_mask = head_mask.unsqueeze(1).unsqueeze(-1).unsqueeze(-1)  # We can specify head_mask for each layer
-            head_mask = head_mask.to(dtype=next(self.parameters()).dtype) # switch to fload if need + fp16 compatibility
+        # Encode if needed (training, first prediction pass)
+        encoder_hidden_states = kwargs_encoder.pop("hidden_states", None)
+        if encoder_hidden_states is None:
+            encoder_inputs_ids = kwargs_encoder.pop("input_ids")
+            hidden_states = self.shared(encoder_inputs_ids)  # Convert inputs in embeddings
+            encoder_outputs = self.encoder(hidden_states, **kwargs_encoder)
+            encoder_hidden_states = encoder_outputs[0]
         else:
-            head_mask = [None] * self.config.num_hidden_layers
+            encoder_outputs = ()
 
-        ##################################
-        # Replace this with your model code
-        embedding_output = self.embeddings(input_ids, position_ids=position_ids, token_type_ids=token_type_ids)
-        encoder_outputs = self.encoder(embedding_output, extended_attention_mask, head_mask=head_mask)
-        sequence_output = encoder_outputs[0]
-        outputs = (sequence_output,) + encoder_outputs[1:]  # add hidden_states and attentions if they are here
+        # Decode
+        decoder_inputs_ids = kwargs_decoder.pop("input_ids")
+        hidden_states = self.shared(decoder_inputs_ids)  # Convert inputs in embeddings
+        kwargs_decoder["encoder_hidden_states"] = encoder_hidden_states
+        kwargs_decoder["encoder_attention_mask"] = kwargs_encoder.get("attention_mask", None)
+        decoder_outputs = self.decoder(hidden_states, **kwargs_decoder)
 
-        return outputs  # sequence_output, (hidden_states), (attentions)
+        return decoder_outputs + encoder_outputs
 
 
 @add_start_docstrings("""T5 Model with a `language modeling` head on top. """,
@@ -342,7 +693,7 @@ class T5WithLMHead(T5PreTrainedModel):
         super(T5ForMaskedLM, self).__init__(config)
 
         self.transformer = T5Model(config)
-        self.lm_head = nn.Linear(config.n_embd, config.vocab_size)
+        self.lm_head = nn.Linear(config.d_model, config.vocab_size)
 
         self.init_weights()
 

From 88e5bef58f34dca87f28ab489fdecbeaaef8b316 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Tue, 5 Nov 2019 17:02:52 +0100
Subject: [PATCH 012/505] share position biases

---
 transformers/modeling_t5.py | 65 +++++++++++++++++++++++--------------
 1 file changed, 40 insertions(+), 25 deletions(-)

diff --git a/transformers/modeling_t5.py b/transformers/modeling_t5.py
index d93e96211d..e1a1d019ff 100644
--- a/transformers/modeling_t5.py
+++ b/transformers/modeling_t5.py
@@ -154,9 +154,10 @@ class T5LayerFF(nn.Module):
 class T5Attention(nn.Module):
     NEW_ID = itertools.count()
 
-    def __init__(self, config):
+    def __init__(self, config, has_relative_attention_bias=False):
         super(T5Attention, self).__init__()
         self.layer_id = next(T5Attention.NEW_ID)
+        self.has_relative_attention_bias = has_relative_attention_bias
 
         self.output_attentions = config.output_attentions
         self.relative_attention_num_buckets = config.relative_attention_num_buckets
@@ -170,7 +171,8 @@ class T5Attention(nn.Module):
         self.v = nn.Linear(self.dim, self.dim, bias=False)
         self.o = nn.Linear(self.dim, self.dim, bias=False)
 
-        self.relative_attention_bias = nn.Embedding(self.relative_attention_num_buckets, self.n_heads)
+        if self.has_relative_attention_bias:
+            self.relative_attention_bias = nn.Embedding(self.relative_attention_num_buckets, self.n_heads)
         self.pruned_heads = set()
 
     def prune_heads(self, heads):
@@ -304,6 +306,8 @@ class T5Attention(nn.Module):
         scores = torch.matmul(q, k.transpose(2, 3))                           # (bs, n_heads, qlen, klen)
 
         if position_bias is None:
+            if not self.has_relative_attention_bias:
+                raise ValueError("No position_bias provided and no weights to compute position_bias")
             position_bias = self.compute_bias(qlen, klen)
         scores += position_bias
 
@@ -325,20 +329,23 @@ class T5Attention(nn.Module):
         outputs = (context,)
         if self.output_attentions:
             outputs = outputs + (weights,)
+        if self.has_relative_attention_bias:
+            outputs = outputs + (position_bias,)
         return outputs
 
 
 class T5LayerSelfAttention(nn.Module):
-    def __init__(self, config):
+    def __init__(self, config, has_relative_attention_bias=False):
         super(T5LayerSelfAttention, self).__init__()
-        self.SelfAttention = T5Attention(config)
+        self.SelfAttention = T5Attention(config, has_relative_attention_bias=has_relative_attention_bias)
         self.layer_norm = nn.LayerNorm(config.layer_norm_epsilon)
         self.dropout = nn.Dropout(config.dropout)
 
-    def forward(self, hidden_states, attention_mask=None, head_mask=None):
+    def forward(self, hidden_states, attention_mask=None, position_bias=None, head_mask=None):
         norm_x = self.layer_norm(hidden_states)
         attention_output = self.SelfAttention(norm_x,
                                               attention_mask=attention_mask,
+                                              position_bias=position_bias,
                                               head_mask=head_mask)
         y = attention_output[0]
         layer_output = hidden_states + self.dropout(y)
@@ -347,17 +354,18 @@ class T5LayerSelfAttention(nn.Module):
 
 
 class T5LayerCrossAttention(nn.Module):
-    def __init__(self, config):
+    def __init__(self, config, has_relative_attention_bias=False):
         super(T5LayerCrossAttention, self).__init__()
-        self.EncDecAttention = T5Attention(config)
+        self.EncDecAttention = T5Attention(config, has_relative_attention_bias=has_relative_attention_bias)
         self.layer_norm = nn.LayerNorm(config.layer_norm_epsilon)
         self.dropout = nn.Dropout(config.dropout)
 
-    def forward(self, hidden_states, kv, attention_mask=None, head_mask=None):
+    def forward(self, hidden_states, kv, attention_mask=None, position_bias=None, head_mask=None):
         norm_x = self.layer_norm(hidden_states)
         attention_output = self.EncDecAttention(norm_x,
                                                 kv=kv,
                                                 attention_mask=attention_mask,
+                                                position_bias=position_bias,
                                                 head_mask=head_mask)
         y = attention_output[0]
         layer_output = hidden_states + self.dropout(y)
@@ -366,20 +374,22 @@ class T5LayerCrossAttention(nn.Module):
 
 
 class T5Block(nn.Module):
-    def __init__(self, config):
+    def __init__(self, config, has_relative_attention_bias=False):
         super(T5Block, self).__init__()
         self.is_decoder = config.is_decoder
-        self.layer_000 = T5LayerSelfAttention(config)
+        self.layer_000 = T5LayerSelfAttention(config, has_relative_attention_bias=has_relative_attention_bias)
         if self.is_decoder:
-            self.layer_001 = T5LayerCrossAttention(config)
+            self.layer_001 = T5LayerCrossAttention(config, has_relative_attention_bias=has_relative_attention_bias)
             self.layer_002 = T5LayerFF(config)
         else:
             self.layer_001 = T5LayerFF(config)
 
-    def forward(self, hidden_states, attention_mask=None,
-                encoder_hidden_states=None, encoder_attention_mask=None, head_mask=None):
+    def forward(self, hidden_states, attention_mask=None, position_bias=None,
+                encoder_hidden_states=None, encoder_attention_mask=None, encoder_decoder_position_bias=None,
+                head_mask=None):
         self_attention_outputs = self.layer_000(hidden_states,
                                                 attention_mask=attention_mask,
+                                                position_bias=position_bias,
                                                 head_mask=head_mask)
         hidden_states = self_attention_outputs[0]
         outputs = self_attention_outputs[1:]
@@ -388,6 +398,7 @@ class T5Block(nn.Module):
             cross_attention_outputs = self.layer_001(hidden_states,
                                                      kv=encoder_hidden_states,
                                                      attention_mask=encoder_attention_mask,
+                                                     position_bias=encoder_decoder_position_bias,
                                                      head_mask=head_mask)
             hidden_states = cross_attention_outputs[0]
             outputs = cross_attention_outputs[1:] + outputs
@@ -402,7 +413,8 @@ class T5Block(nn.Module):
 class T5Stack(nn.Module):
     def __init__(self, config):
         super(T5Stack, self).__init__()
-        self.blocks = nn.ModuleList([T5Block(config) for _ in range(config.num_layers)])
+        self.blocks = nn.ModuleList([T5Block(config, has_relative_attention_bias=bool(i == 0))
+                                     for i in range(config.num_layers)])
         self.final_layer_norm = nn.LayerNorm(config.layer_norm_epsilon)
         self.dropout = nn.Dropout(config.dropout)
 
@@ -413,8 +425,12 @@ class T5Stack(nn.Module):
                 encoder_attention_mask=None,
                 head_mask=None):
 
+        batch_size, seq_length = hidden_states.shape[0], hidden_states.shape[1]
+        encoder_seq_length = encoder_hidden_states.shape[1] if encoder_hidden_states is not None else 0
         if attention_mask is None:
-            attention_mask = torch.ones_like(input_ids)
+            attention_mask = torch.ones(batch_size, seq_length).to(hidden_states.device)
+        if encoder_attention_mask is None:
+            encoder_attention_mask = torch.ones(batch_size, encoder_seq_length).to(hidden_states.device)
 
         # We can provide a self-attention mask of dimensions [batch_size, from_seq_length, to_seq_length]
         # ourselves in which case we just need to make it broadcastable to all heads.
@@ -426,8 +442,7 @@ class T5Stack(nn.Module):
         # - if the model is an encoder, make the mask broadcastable to [batch_size, num_heads, seq_length, seq_length]
         if attention_mask.dim() == 2:
             if self.config.is_decoder:
-                batch_size, seq_length = input_ids.size()
-                seq_ids = torch.arange(seq_length, device=input_ids.device)
+                seq_ids = torch.arange(seq_length, device=hidden_states.device)
                 causal_mask = seq_ids[None, None, :].repeat(batch_size, seq_length, 1) <= seq_ids[None, :, None]
                 extended_attention_mask = causal_mask[:, None, :, :] * attention_mask[:, None, None, :]
             else:
@@ -469,16 +484,22 @@ class T5Stack(nn.Module):
         all_hidden_states = ()
         all_attentions = ()
         position_bias = None
+        encoder_decoder_position_bias = None
         for i, layer_module in enumerate(self.layer):
             if self.output_hidden_states:
                 all_hidden_states = all_hidden_states + (hidden_states,)
 
             layer_outputs = layer_module(hidden_states,
                                          attention_mask=extended_attention_mask,
+                                         position_bias=position_bias,
                                          encoder_hidden_states=encoder_hidden_states,
                                          encoder_attention_mask=encoder_extended_attention_mask,
+                                         encoder_decoder_position_bias=encoder_decoder_position_bias,
                                          head_mask=head_mask[i])
             hidden_states = layer_outputs[0]
+            if i == 0:
+                position_bias = layer_outputs[2] if len(layer_outputs) > 3 else None
+                encoder_decoder_position_bias = layer_outputs[4] if len(layer_outputs) > 5 else None
 
             if self.output_attentions:
                 all_attentions = all_attentions + (layer_outputs[1],)
@@ -700,14 +721,8 @@ class T5WithLMHead(T5PreTrainedModel):
     def get_output_embeddings(self):
         return self.lm_head
 
-    def forward(self, input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None,
-                lm_labels=None):
-
-        outputs = self.transformer(input_ids,
-                            attention_mask=attention_mask,
-                            token_type_ids=token_type_ids,
-                            position_ids=position_ids, 
-                            head_mask=head_mask)
+    def forward(self, encoder_input_ids, decoder_input_ids, **kwargs):
+        outputs = self.transformer(encoder_input_ids, decoder_input_ids, **kwargs)
 
         sequence_output = outputs[0]
         lm_logits = self.cls(sequence_output)

From 151e4ab4e786b9b4b702205b5077ea2dfe67b4dd Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Tue, 5 Nov 2019 16:26:51 +0000
Subject: [PATCH 013/505] Fix CTRL past

---
 transformers/modeling_ctrl.py | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/transformers/modeling_ctrl.py b/transformers/modeling_ctrl.py
index 1873040a8e..589a065a11 100644
--- a/transformers/modeling_ctrl.py
+++ b/transformers/modeling_ctrl.py
@@ -63,7 +63,8 @@ def scaled_dot_product_attention(q, k, v, mask, attention_mask=None, head_mask=N
     scaled_attention_logits = matmul_qk / np.sqrt(dk)
 
     if mask is not None:
-        scaled_attention_logits += (mask * -1e4)
+        nd, ns = scaled_attention_logits.size(-2), scaled_attention_logits.size(-1)
+        scaled_attention_logits += (mask[ns-nd:ns, :ns] * -1e4)
 
     if attention_mask is not None:
         # Apply the attention mask
@@ -357,7 +358,7 @@ class CTRLModel(CTRLPreTrainedModel):
         inputs_embeds = self.w(input_ids)
         # inputs_embeds = embedded.unsqueeze(0) if len(input_ids.shape)<2 else embedded
         seq_len = input_ids.shape[-1]
-        mask = torch.triu(torch.ones(seq_len, seq_len), 1).to(inputs_embeds.device)
+        mask = torch.triu(torch.ones(seq_len + past_length, seq_len + past_length), 1).to(inputs_embeds.device)
 
         inputs_embeds *= np.sqrt(self.d_model_size)
 

From 3835e1e651ebeeddaa8dd8cb5f4d30912ec5ec6d Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Wed, 6 Nov 2019 11:52:29 +0100
Subject: [PATCH 014/505] adding tokenizer

---
 transformers/tokenization_t5.py | 188 +++++++++-----------------------
 1 file changed, 51 insertions(+), 137 deletions(-)

diff --git a/transformers/tokenization_t5.py b/transformers/tokenization_t5.py
index 3f8f4bf556..cff6a41baf 100644
--- a/transformers/tokenization_t5.py
+++ b/transformers/tokenization_t5.py
@@ -16,16 +16,15 @@
 
 from __future__ import absolute_import, division, print_function, unicode_literals
 
-import collections
 import logging
 import os
-import unicodedata
-from io import open
 
 from .tokenization_utils import PreTrainedTokenizer
 
 logger = logging.getLogger(__name__)
 
+SPIECE_UNDERLINE = u'▁'
+
 ####################################################
 # Mapping from the keyword arguments names of Tokenizer `__init__`
 # to file names for serializing Tokenizer instances
@@ -39,8 +38,7 @@ VOCAB_FILES_NAMES = {'vocab_file': 'vocab.txt'}
 PRETRAINED_VOCAB_FILES_MAP = {
     'vocab_file':
     {
-        't5-base-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-base-uncased-vocab.txt",
-        't5-large-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-large-uncased-vocab.txt",
+        't5': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-spiece.model",
     }
 }
 
@@ -48,167 +46,83 @@ PRETRAINED_VOCAB_FILES_MAP = {
 # Mapping from model shortcut names to max length of inputs
 ####################################################
 PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
-    't5-base-uncased': 512,
-    't5-large-uncased': 512,
+    't5': 512,
 }
 
-####################################################
-# Mapping from model shortcut names to a dictionary of additional
-# keyword arguments for Tokenizer `__init__`.
-# To be used for checkpoint specific configurations.
-####################################################
-PRETRAINED_INIT_CONFIGURATION = {
-    't5-base-uncased': {'do_lower_case': True},
-    't5-large-uncased': {'do_lower_case': True},
-}
-
-
-def load_vocab(vocab_file):
-    """Loads a vocabulary file into a dictionary."""
-    vocab = collections.OrderedDict()
-    with open(vocab_file, "r", encoding="utf-8") as reader:
-        tokens = reader.readlines()
-    for index, token in enumerate(tokens):
-        token = token.rstrip('\n')
-        vocab[token] = index
-    return vocab
-
-
 class T5Tokenizer(PreTrainedTokenizer):
-    r"""
-    Constructs a T5Tokenizer.
-    :class:`~transformers.T5Tokenizer` runs end-to-end tokenization: punctuation splitting + wordpiece
-
-    Args:
-        vocab_file: Path to a one-wordpiece-per-line vocabulary file
-        do_lower_case: Whether to lower case the input. Only has an effect when do_wordpiece_only=False
     """
+        SentencePiece based tokenizer. Peculiarities:
 
+            - requires `SentencePiece <https://github.com/google/sentencepiece>`_
+    """
     vocab_files_names = VOCAB_FILES_NAMES
     pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
-    pretrained_init_configuration = PRETRAINED_INIT_CONFIGURATION
     max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
 
-    def __init__(self, vocab_file, do_lower_case=True,
-                 unk_token="[UNK]", sep_token="[SEP]", pad_token="[PAD]", cls_token="[CLS]",
-                 mask_token="[MASK]", **kwargs):
-        """Constructs a T5Tokenizer.
+    def __init__(self, vocab_file, eos_token="</s>", unk_token="<unk>",
+                 pad_token="<pad>", **kwargs):
+        super(T5Tokenizer, self).__init__(eos_token=eos_token, unk_token=unk_token,
+                                          pad_token=pad_token, **kwargs)
 
-        Args:
-            **vocab_file**: Path to a one-wordpiece-per-line vocabulary file
-            **do_lower_case**: (`optional`) boolean (default True)
-                Whether to lower case the input
-                Only has an effect when do_basic_tokenize=True
-        """
-        super(T5Tokenizer, self).__init__(unk_token=unk_token, sep_token=sep_token,
-                                           pad_token=pad_token, cls_token=cls_token,
-                                           mask_token=mask_token, **kwargs)
-        self.max_len_single_sentence = self.max_len - 2  # take into account special tokens
-        self.max_len_sentences_pair = self.max_len - 3  # take into account special tokens
+        try:
+            import sentencepiece as spm
+        except ImportError:
+            logger.warning("You need to install SentencePiece to use T5Tokenizer:"
+                           "https://github.com/google/sentencepiece"
+                           "pip install sentencepiece")
 
-        if not os.path.isfile(vocab_file):
-            raise ValueError(
-                "Can't find a vocabulary file at path '{}'. To load the vocabulary from a Google pretrained "
-                "model use `tokenizer = T5Tokenizer.from_pretrained(PRETRAINED_MODEL_NAME)`".format(vocab_file))
-        self.vocab = load_vocab(vocab_file)
+        self.vocab_file = vocab_file
+
+        self.sp_model = spm.SentencePieceProcessor()
+        self.sp_model.Load(vocab_file)
 
     @property
     def vocab_size(self):
-        return len(self.vocab)
+        return self.sp_model.get_piece_size()
+
+    def __getstate__(self):
+        state = self.__dict__.copy()
+        state["sp_model"] = None
+        return state
+
+    def __setstate__(self, d):
+        self.__dict__ = d
+        try:
+            import sentencepiece as spm
+        except ImportError:
+            logger.warning("You need to install SentencePiece to use XLNetTokenizer: https://github.com/google/sentencepiece"
+                           "pip install sentencepiece")
+        self.sp_model = spm.SentencePieceProcessor()
+        self.sp_model.Load(self.vocab_file)
 
     def _tokenize(self, text):
         """ Take as input a string and return a list of strings (tokens) for words/sub-words
         """
-        split_tokens = []
-        if self.do_basic_tokenize:
-            for token in self.basic_tokenizer.tokenize(text, never_split=self.all_special_tokens):
-                for sub_token in self.wordpiece_tokenizer.tokenize(token):
-                    split_tokens.append(sub_token)
-        else:
-            split_tokens = self.wordpiece_tokenizer.tokenize(text)
-        return split_tokens
+        return self.sp_model.EncodeAsPieces(text)
 
     def _convert_token_to_id(self, token):
         """ Converts a token (str/unicode) in an id using the vocab. """
-        return self.vocab.get(token, self.vocab.get(self.unk_token))
+        return self.sp_model.piece_to_id(token)
 
     def _convert_id_to_token(self, index):
         """Converts an index (integer) in a token (string/unicode) using the vocab."""
-        return self.ids_to_tokens.get(index, self.unk_token)
+        return self.sp_model.id_to_piece(index)
 
     def convert_tokens_to_string(self, tokens):
         """ Converts a sequence of tokens (string) in a single string. """
-        out_string = ' '.join(tokens).replace(' ##', '').strip()
+        out_string = self.sp_model.decode_pieces(tokens)
         return out_string
 
-    def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
+    def save_vocabulary(self, save_directory):
+        """ Save the sentencepiece vocabulary (copy original file) and special tokens file
+            to a directory.
         """
-        Build model inputs from a sequence or a pair of sequence for sequence classification tasks
-        by concatenating and adding special tokens.
-        A BERT sequence has the following format:
-            single sequence: [CLS] X [SEP]
-            pair of sequences: [CLS] A [SEP] B [SEP]
-        """
-        if token_ids_1 is None:
-            return [self.cls_token_id] + token_ids_0 + [self.sep_token_id]
-        cls = [self.cls_token_id]
-        sep = [self.sep_token_id]
-        return cls + token_ids_0 + sep + token_ids_1 + sep
+        if not os.path.isdir(save_directory):
+            logger.error("Vocabulary path ({}) should be a directory".format(save_directory))
+            return
+        out_vocab_file = os.path.join(save_directory, VOCAB_FILES_NAMES['vocab_file'])
 
-    def get_special_tokens_mask(self, token_ids_0, token_ids_1=None, already_has_special_tokens=False):
-        """
-        Retrieves sequence ids from a token list that has no special tokens added. This method is called when adding
-        special tokens using the tokenizer ``prepare_for_model`` or ``encode_plus`` methods.
+        if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file):
+            copyfile(self.vocab_file, out_vocab_file)
 
-        Args:
-            token_ids_0: list of ids (must not contain special tokens)
-            token_ids_1: Optional list of ids (must not contain special tokens), necessary when fetching sequence ids
-                for sequence pairs
-            already_has_special_tokens: (default False) Set to True if the token list is already formated with
-                special tokens for the model
-
-        Returns:
-            A list of integers in the range [0, 1]: 0 for a special token, 1 for a sequence token.
-        """
-
-        if already_has_special_tokens:
-            if token_ids_1 is not None:
-                raise ValueError("You should not supply a second sequence if the provided sequence of "
-                                 "ids is already formated with special tokens for the model.")
-            return list(map(lambda x: 1 if x in [self.sep_token_id, self.cls_token_id] else 0, token_ids_0))
-
-        if token_ids_1 is not None:
-            return [1] + ([0] * len(token_ids_0)) + [1] + ([0] * len(token_ids_1)) + [1]
-        return [1] + ([0] * len(token_ids_0)) + [1]
-
-    def create_token_type_ids_from_sequences(self, token_ids_0, token_ids_1=None):
-        """
-        Creates a mask from the two sequences passed to be used in a sequence-pair classification task.
-        A BERT sequence pair mask has the following format:
-        0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1
-        | first sequence    | second sequence
-
-        if token_ids_1 is None, only returns the first portion of the mask (0's).
-        """
-        sep = [self.sep_token_id]
-        cls = [self.cls_token_id]
-        if token_ids_1 is None:
-            return len(cls + token_ids_0 + sep) * [0]
-        return len(cls + token_ids_0 + sep) * [0] + len(token_ids_1 + sep) * [1]
-
-    def save_vocabulary(self, vocab_path):
-        """Save the tokenizer vocabulary to a directory or file."""
-        index = 0
-        if os.path.isdir(vocab_path):
-            vocab_file = os.path.join(vocab_path, VOCAB_FILES_NAMES['vocab_file'])
-        else:
-            vocab_file = vocab_path
-        with open(vocab_file, "w", encoding="utf-8") as writer:
-            for token, token_index in sorted(self.vocab.items(), key=lambda kv: kv[1]):
-                if index != token_index:
-                    logger.warning("Saving vocabulary to {}: vocabulary indices are not consecutive."
-                                   " Please check that the vocabulary is not corrupted!".format(vocab_file))
-                    index = token_index
-                writer.write(token + u'\n')
-                index += 1
-        return (vocab_file,)
+        return (out_vocab_file,)

From 73f2c342f53f2ff02124da23ba029d80c386e7ce Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Wed, 6 Nov 2019 11:52:39 +0100
Subject: [PATCH 015/505] fixing template

---
 templates/adding_a_new_model/configuration_xxx.py | 2 +-
 templates/adding_a_new_model/modeling_xxx.py      | 1 -
 2 files changed, 1 insertion(+), 2 deletions(-)

diff --git a/templates/adding_a_new_model/configuration_xxx.py b/templates/adding_a_new_model/configuration_xxx.py
index b1614e71af..14c1c2c79e 100644
--- a/templates/adding_a_new_model/configuration_xxx.py
+++ b/templates/adding_a_new_model/configuration_xxx.py
@@ -84,7 +84,7 @@ class XxxConfig(PretrainedConfig):
                  summary_first_dropout=0.1,
                  **kwargs):
         super(XxxConfig, self).__init__(**kwargs)
-        self.vocab_size = vocab_size_or_config_json_file if isinstance(vocab_size_or_config_json_file, six.string_types) else -1
+        self.vocab_size = vocab_size_or_config_json_file if isinstance(vocab_size_or_config_json_file, int) else -1
         self.n_ctx = n_ctx
         self.n_positions = n_positions
         self.n_embd = n_embd
diff --git a/templates/adding_a_new_model/modeling_xxx.py b/templates/adding_a_new_model/modeling_xxx.py
index ff64f13f40..ee705e753c 100644
--- a/templates/adding_a_new_model/modeling_xxx.py
+++ b/templates/adding_a_new_model/modeling_xxx.py
@@ -280,7 +280,6 @@ class XxxModel(XxxPreTrainedModel):
 
         self.init_weights()
 
-    @property
     def get_input_embeddings(self):
         return self.embeddings.word_embeddings
 

From 076a207935bfcc38416cd0baa887d3e025ebef28 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Wed, 6 Nov 2019 11:52:50 +0100
Subject: [PATCH 016/505] adding tests and updating model

---
 transformers/__init__.py                   |  11 +-
 transformers/configuration_t5.py           |  53 +++---
 transformers/modeling_t5.py                | 151 ++++++++--------
 transformers/tests/modeling_common_test.py |  32 ++--
 transformers/tests/modeling_t5_test.py     | 176 +++++++++++++++++++
 transformers/tests/modeling_tf_t5_test.py  | 190 +++++++++++++++++++++
 transformers/tests/tokenization_t5_test.py |  77 +++++++++
 7 files changed, 571 insertions(+), 119 deletions(-)
 create mode 100644 transformers/tests/modeling_t5_test.py
 create mode 100644 transformers/tests/modeling_tf_t5_test.py
 create mode 100644 transformers/tests/tokenization_t5_test.py

diff --git a/transformers/__init__.py b/transformers/__init__.py
index 53f3c39dc7..bf896276d6 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -42,6 +42,7 @@ from .tokenization_xlnet import XLNetTokenizer, SPIECE_UNDERLINE
 from .tokenization_xlm import XLMTokenizer
 from .tokenization_roberta import RobertaTokenizer
 from .tokenization_distilbert import DistilBertTokenizer
+from .tokenization_t5 import T5Tokenizer
 
 # Configurations
 from .configuration_utils import PretrainedConfig
@@ -52,10 +53,10 @@ from .configuration_transfo_xl import TransfoXLConfig, TRANSFO_XL_PRETRAINED_CON
 from .configuration_gpt2 import GPT2Config, GPT2_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_ctrl import CTRLConfig, CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_xlnet import XLNetConfig, XLNET_PRETRAINED_CONFIG_ARCHIVE_MAP
-from .configuration_ctrl import CTRLConfig, CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_xlm import XLMConfig, XLM_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_roberta import RobertaConfig, ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_distilbert import DistilBertConfig, DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_t5 import T5Config, T5_PRETRAINED_CONFIG_ARCHIVE_MAP
 
 # Modeling
 if is_torch_available():
@@ -69,10 +70,10 @@ if is_torch_available():
                                 BertForTokenClassification, BertForQuestionAnswering,
                                 load_tf_weights_in_bert, BERT_PRETRAINED_MODEL_ARCHIVE_MAP)
     from .modeling_openai import (OpenAIGPTPreTrainedModel, OpenAIGPTModel,
-                                OpenAIGPTLMHeadModel, OpenAIGPTDoubleHeadsModel,
-                                load_tf_weights_in_openai_gpt, OPENAI_GPT_PRETRAINED_MODEL_ARCHIVE_MAP)
+                                  OpenAIGPTLMHeadModel, OpenAIGPTDoubleHeadsModel,
+                                  load_tf_weights_in_openai_gpt, OPENAI_GPT_PRETRAINED_MODEL_ARCHIVE_MAP)
     from .modeling_transfo_xl import (TransfoXLPreTrainedModel, TransfoXLModel, TransfoXLLMHeadModel,
-                                    load_tf_weights_in_transfo_xl, TRANSFO_XL_PRETRAINED_MODEL_ARCHIVE_MAP)
+                                      load_tf_weights_in_transfo_xl, TRANSFO_XL_PRETRAINED_MODEL_ARCHIVE_MAP)
     from .modeling_gpt2 import (GPT2PreTrainedModel, GPT2Model,
                                 GPT2LMHeadModel, GPT2DoubleHeadsModel,
                                 load_tf_weights_in_gpt2, GPT2_PRETRAINED_MODEL_ARCHIVE_MAP)
@@ -95,6 +96,8 @@ if is_torch_available():
                                 DistilBertForSequenceClassification, DistilBertForQuestionAnswering,
                                 DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
     from .modeling_encoder_decoder import PreTrainedEncoderDecoder, Model2Model
+    from .modeling_t5 import (T5PreTrainedModel, T5Model, T5WithLMHeadModel,
+                              T5_PRETRAINED_MODEL_ARCHIVE_MAP)
 
     # Optimization
     from .optimization import (AdamW, ConstantLRSchedule, WarmupConstantSchedule, WarmupCosineSchedule,
diff --git a/transformers/configuration_t5.py b/transformers/configuration_t5.py
index a37a5b2157..9db918e59f 100644
--- a/transformers/configuration_t5.py
+++ b/transformers/configuration_t5.py
@@ -64,44 +64,29 @@ class T5Config(PretrainedConfig):
     pretrained_config_archive_map = T5_PRETRAINED_CONFIG_ARCHIVE_MAP
 
     def __init__(self,
-                 vocab_size_or_config_json_file=50257,
-                 n_positions=1024,
-                 n_ctx=1024,
-                 n_embd=768,
-                 n_layer=12,
-                 n_head=12,
-                 resid_pdrop=0.1,
-                 embd_pdrop=0.1,
-                 attn_pdrop=0.1,
-                 layer_norm_epsilon=1e-5,
+                 vocab_size_or_config_json_file=32128,
+                 n_positions=512,
+                 d_model=512,
+                 d_ff=2048,
+                 num_layers=12,
+                 num_heads=12,
+                 relative_attention_num_buckets=32,
+                 dropout_rate=0.1,
+                 layer_norm_epsilon=1e-6,
                  initializer_range=0.02,
-
-                 num_labels=1,
-                 summary_type='cls_index',
-                 summary_use_proj=True,
-                 summary_activation=None,
-                 summary_proj_to_labels=True,
-                 summary_first_dropout=0.1,
                  **kwargs):
         super(T5Config, self).__init__(**kwargs)
-        self.vocab_size = vocab_size_or_config_json_file if isinstance(vocab_size_or_config_json_file, six.string_types) else -1
-        self.n_ctx = n_ctx
+        self.vocab_size = vocab_size_or_config_json_file if isinstance(vocab_size_or_config_json_file, int) else -1
         self.n_positions = n_positions
-        self.n_embd = n_embd
-        self.n_layer = n_layer
-        self.n_head = n_head
-        self.resid_pdrop = resid_pdrop
-        self.embd_pdrop = embd_pdrop
-        self.attn_pdrop = attn_pdrop
+        self.d_model = d_model
+        self.d_ff = d_ff
+        self.num_layers = num_layers
+        self.num_heads = num_heads
+        self.relative_attention_num_buckets = relative_attention_num_buckets
+        self.dropout_rate = dropout_rate
         self.layer_norm_epsilon = layer_norm_epsilon
         self.initializer_range = initializer_range
 
-        self.num_labels = num_labels
-        self.summary_type = summary_type
-        self.summary_use_proj = summary_use_proj
-        self.summary_activation = summary_activation
-        self.summary_first_dropout = summary_first_dropout
-        self.summary_proj_to_labels = summary_proj_to_labels
         if isinstance(vocab_size_or_config_json_file, six.string_types):
             with open(vocab_size_or_config_json_file, "r", encoding="utf-8") as reader:
                 json_config = json.loads(reader.read())
@@ -119,12 +104,12 @@ class T5Config(PretrainedConfig):
 
     @property
     def hidden_size(self):
-        return self.n_embd
+        return self.d_model
 
     @property
     def num_attention_heads(self):
-        return self.n_head
+        return self.num_heads
 
     @property
     def num_hidden_layers(self):
-        return self.n_layer
+        return self.num_layers
diff --git a/transformers/modeling_t5.py b/transformers/modeling_t5.py
index e1a1d019ff..ce443cf882 100644
--- a/transformers/modeling_t5.py
+++ b/transformers/modeling_t5.py
@@ -20,8 +20,8 @@ import json
 import logging
 import math
 import os
-import math
 import sys
+import copy
 import itertools
 from io import open
 
@@ -30,7 +30,7 @@ from torch import nn
 import torch.nn.functional as F
 from torch.nn import CrossEntropyLoss, MSELoss
 
-from .modeling_utils import PreTrainedModel, prune_linear_layer
+from .modeling_utils import PreTrainedModel
 from .configuration_t5 import T5Config
 from .file_utils import add_start_docstrings
 
@@ -127,7 +127,7 @@ class T5DenseReluDense(nn.Module):
         super(T5DenseReluDense, self).__init__()
         self.wi = nn.Linear(config.d_model, config.d_ff, bias=False)
         self.wo = nn.Linear(config.d_ff, config.d_model, bias=False)
-        self.dropout = nn.Dropout(config.dropout)
+        self.dropout = nn.Dropout(config.dropout_rate)
 
     def forward(self, hidden_states):
         h = self.wi(hidden_states)
@@ -141,8 +141,8 @@ class T5LayerFF(nn.Module):
     def __init__(self, config):
         super(T5LayerFF, self).__init__()
         self.DenseReluDense = T5DenseReluDense(config)
-        self.layer_norm = nn.LayerNorm(config.layer_norm_epsilon)
-        self.dropout = nn.Dropout(config.dropout)
+        self.layer_norm = nn.LayerNorm(config.d_model, eps=config.layer_norm_epsilon)
+        self.dropout = nn.Dropout(config.dropout_rate)
 
     def forward(self, hidden_states):
         norm_x = self.layer_norm(hidden_states)
@@ -157,6 +157,7 @@ class T5Attention(nn.Module):
     def __init__(self, config, has_relative_attention_bias=False):
         super(T5Attention, self).__init__()
         self.layer_id = next(T5Attention.NEW_ID)
+        self.is_decoder = config.is_decoder
         self.has_relative_attention_bias = has_relative_attention_bias
 
         self.output_attentions = config.output_attentions
@@ -231,7 +232,7 @@ class T5Attention(nn.Module):
             ret += (n < 0).to(torch.long) * num_buckets  # mtf.to_int32(mtf.less(n, 0)) * num_buckets
             n = torch.abs(n)
         else:
-            n = torch.max(n, 0)
+            n = torch.max(n, torch.zeros_like(n))
         # now n is in the range [0, inf)
 
         # half of the buckets are for exact increments in positions
@@ -242,7 +243,7 @@ class T5Attention(nn.Module):
         val_if_large = max_exact + (
             torch.log(n.float() / max_exact)
             / math.log(max_distance / max_exact) * (num_buckets - max_exact)).to(torch.long)
-        val_if_large = torch.min(val_if_large, num_buckets - 1)
+        val_if_large = torch.min(val_if_large, torch.full_like(val_if_large, num_buckets - 1))
 
         ret += torch.where(is_small, n, val_if_large)
         return ret
@@ -259,7 +260,7 @@ class T5Attention(nn.Module):
         values = values.permute([2, 0, 1]).unsqueeze(0) # shape (1, num_heads, qlen, klen)
         return values
 
-    def forward(self, input, mask, kv=None, position_bias=None, cache=None, head_mask=None):
+    def forward(self, input, mask=None, kv=None, position_bias=None, cache=None, head_mask=None):
         """
         Self-attention (if kv is None) or attention over source sentence (provided by kv).
         """
@@ -273,7 +274,6 @@ class T5Attention(nn.Module):
         # assert dim == self.dim, 'Dimensions do not match: %s input vs %s configured' % (dim, self.dim)
         n_heads = self.n_heads
         dim_per_head = self.dim // n_heads
-        mask_reshape = (bs, 1, qlen, klen) if mask.dim() == 3 else (bs, 1, 1, klen)
 
         def shape(x):
             """  projection """
@@ -311,8 +311,9 @@ class T5Attention(nn.Module):
             position_bias = self.compute_bias(qlen, klen)
         scores += position_bias
 
-        mask = (mask == 0).view(mask_reshape).expand_as(scores)               # (bs, n_heads, qlen, klen)
-        scores.masked_fill_(mask, -float('inf'))                              # (bs, n_heads, qlen, klen)
+        if mask is not None:
+            mask = (mask == 0).expand_as(scores)                              # (bs, n_heads, qlen, klen)
+            scores.masked_fill_(mask, -float('inf'))                          # (bs, n_heads, qlen, klen)
 
         weights = F.softmax(scores.float(), dim=-1).type_as(scores)           # (bs, n_heads, qlen, klen)
         weights = F.dropout(weights, p=self.dropout, training=self.training)  # (bs, n_heads, qlen, klen)
@@ -338,13 +339,13 @@ class T5LayerSelfAttention(nn.Module):
     def __init__(self, config, has_relative_attention_bias=False):
         super(T5LayerSelfAttention, self).__init__()
         self.SelfAttention = T5Attention(config, has_relative_attention_bias=has_relative_attention_bias)
-        self.layer_norm = nn.LayerNorm(config.layer_norm_epsilon)
-        self.dropout = nn.Dropout(config.dropout)
+        self.layer_norm = nn.LayerNorm(config.d_model, eps=config.layer_norm_epsilon)
+        self.dropout = nn.Dropout(config.dropout_rate)
 
     def forward(self, hidden_states, attention_mask=None, position_bias=None, head_mask=None):
         norm_x = self.layer_norm(hidden_states)
         attention_output = self.SelfAttention(norm_x,
-                                              attention_mask=attention_mask,
+                                              mask=attention_mask,
                                               position_bias=position_bias,
                                               head_mask=head_mask)
         y = attention_output[0]
@@ -357,14 +358,14 @@ class T5LayerCrossAttention(nn.Module):
     def __init__(self, config, has_relative_attention_bias=False):
         super(T5LayerCrossAttention, self).__init__()
         self.EncDecAttention = T5Attention(config, has_relative_attention_bias=has_relative_attention_bias)
-        self.layer_norm = nn.LayerNorm(config.layer_norm_epsilon)
-        self.dropout = nn.Dropout(config.dropout)
+        self.layer_norm = nn.LayerNorm(config.d_model, eps=config.layer_norm_epsilon)
+        self.dropout = nn.Dropout(config.dropout_rate)
 
     def forward(self, hidden_states, kv, attention_mask=None, position_bias=None, head_mask=None):
         norm_x = self.layer_norm(hidden_states)
         attention_output = self.EncDecAttention(norm_x,
+                                                mask=attention_mask,
                                                 kv=kv,
-                                                attention_mask=attention_mask,
                                                 position_bias=position_bias,
                                                 head_mask=head_mask)
         y = attention_output[0]
@@ -410,13 +411,41 @@ class T5Block(nn.Module):
         return outputs
 
 
-class T5Stack(nn.Module):
+class T5PreTrainedModel(PreTrainedModel):
+    """ An abstract class to handle weights initialization and
+        a simple interface for dowloading and loading pretrained models.
+    """
+    config_class = T5Config
+    pretrained_model_archive_map = T5_PRETRAINED_MODEL_ARCHIVE_MAP
+    load_tf_weights = load_tf_weights_in_t5
+    base_model_prefix = "transformer"
+
+    def _init_weights(self, module):
+        """ Initialize the weights """
+        if isinstance(module, (nn.Linear, nn.Embedding)):
+            # Slightly different from the TF version which uses truncated_normal for initialization
+            # cf https://github.com/pytorch/pytorch/pull/5617
+            module.weight.data.normal_(mean=0.0, std=self.config.initializer_range)
+        elif isinstance(module, nn.LayerNorm):
+            module.bias.data.zero_()
+            module.weight.data.fill_(1.0)
+        if isinstance(module, nn.Linear) and module.bias is not None:
+            module.bias.data.zero_()
+
+
+class T5Stack(T5PreTrainedModel):
     def __init__(self, config):
-        super(T5Stack, self).__init__()
+        super(T5Stack, self).__init__(config)
+        self.output_attentions = config.output_attentions
+        self.output_hidden_states = config.output_hidden_states
+        self.is_decoder = config.is_decoder
+
         self.blocks = nn.ModuleList([T5Block(config, has_relative_attention_bias=bool(i == 0))
                                      for i in range(config.num_layers)])
-        self.final_layer_norm = nn.LayerNorm(config.layer_norm_epsilon)
-        self.dropout = nn.Dropout(config.dropout)
+        self.final_layer_norm = nn.LayerNorm(config.d_model, eps=config.layer_norm_epsilon)
+        self.dropout = nn.Dropout(config.dropout_rate)
+
+        self.init_weights()
 
     def forward(self,
                 hidden_states,
@@ -426,10 +455,10 @@ class T5Stack(nn.Module):
                 head_mask=None):
 
         batch_size, seq_length = hidden_states.shape[0], hidden_states.shape[1]
-        encoder_seq_length = encoder_hidden_states.shape[1] if encoder_hidden_states is not None else 0
         if attention_mask is None:
             attention_mask = torch.ones(batch_size, seq_length).to(hidden_states.device)
-        if encoder_attention_mask is None:
+        if self.is_decoder and encoder_attention_mask is None:
+            encoder_seq_length = encoder_hidden_states.shape[1]
             encoder_attention_mask = torch.ones(batch_size, encoder_seq_length).to(hidden_states.device)
 
         # We can provide a self-attention mask of dimensions [batch_size, from_seq_length, to_seq_length]
@@ -444,6 +473,7 @@ class T5Stack(nn.Module):
             if self.config.is_decoder:
                 seq_ids = torch.arange(seq_length, device=hidden_states.device)
                 causal_mask = seq_ids[None, None, :].repeat(batch_size, seq_length, 1) <= seq_ids[None, :, None]
+                causal_mask = causal_mask.to(attention_mask)
                 extended_attention_mask = causal_mask[:, None, :, :] * attention_mask[:, None, None, :]
             else:
                 extended_attention_mask = attention_mask[:, None, None, :]
@@ -456,15 +486,18 @@ class T5Stack(nn.Module):
         extended_attention_mask = extended_attention_mask.to(dtype=next(self.parameters()).dtype)  # fp16 compatibility
         extended_attention_mask = (1.0 - extended_attention_mask) * -10000.0
 
-        # If a 2D ou 3D attention mask is provided for the cross-attention
-        # we need to make broadcastabe to [batch_size, num_heads, seq_length, seq_length]
-        if encoder_attention_mask.dim() == 3:
-            encoder_extended_attention_mask = encoder_attention_mask[:, None, :, :]
-        if encoder_attention_mask.dim() == 2:
-            encoder_extended_attention_mask = encoder_attention_mask[:, None, None, :]
+        if self.is_decoder:
+            # If a 2D ou 3D attention mask is provided for the cross-attention
+            # we need to make broadcastabe to [batch_size, num_heads, seq_length, seq_length]
+            if encoder_attention_mask.dim() == 3:
+                encoder_extended_attention_mask = encoder_attention_mask[:, None, :, :]
+            if encoder_attention_mask.dim() == 2:
+                encoder_extended_attention_mask = encoder_attention_mask[:, None, None, :]
 
-        encoder_extended_attention_mask = encoder_extended_attention_mask.to(dtype=next(self.parameters()).dtype)  # fp16 compatibility
-        encoder_extended_attention_mask = (1.0 - encoder_extended_attention_mask) * -10000.0
+            encoder_extended_attention_mask = encoder_extended_attention_mask.to(dtype=next(self.parameters()).dtype)  # fp16 compatibility
+            encoder_extended_attention_mask = (1.0 - encoder_extended_attention_mask) * -10000.0
+        else:
+            encoder_extended_attention_mask = None
 
         # Prepare head mask if needed
         # 1.0 in head_mask indicate we keep the head
@@ -474,18 +507,18 @@ class T5Stack(nn.Module):
         if head_mask is not None:
             if head_mask.dim() == 1:
                 head_mask = head_mask.unsqueeze(0).unsqueeze(0).unsqueeze(-1).unsqueeze(-1)
-                head_mask = head_mask.expand(self.config.num_hidden_layers, -1, -1, -1, -1)
+                head_mask = head_mask.expand(self.config.num_layers, -1, -1, -1, -1)
             elif head_mask.dim() == 2:
                 head_mask = head_mask.unsqueeze(1).unsqueeze(-1).unsqueeze(-1)  # We can specify head_mask for each layer
             head_mask = head_mask.to(dtype=next(self.parameters()).dtype) # switch to fload if need + fp16 compatibility
         else:
-            head_mask = [None] * self.config.num_hidden_layers
+            head_mask = [None] * self.config.num_layers
 
         all_hidden_states = ()
         all_attentions = ()
         position_bias = None
         encoder_decoder_position_bias = None
-        for i, layer_module in enumerate(self.layer):
+        for i, layer_module in enumerate(self.blocks):
             if self.output_hidden_states:
                 all_hidden_states = all_hidden_states + (hidden_states,)
 
@@ -498,8 +531,9 @@ class T5Stack(nn.Module):
                                          head_mask=head_mask[i])
             hidden_states = layer_outputs[0]
             if i == 0:
-                position_bias = layer_outputs[2] if len(layer_outputs) > 3 else None
-                encoder_decoder_position_bias = layer_outputs[4] if len(layer_outputs) > 5 else None
+                position_bias = layer_outputs[2 if self.output_attentions else 1]
+                if self.is_decoder:
+                    encoder_decoder_position_bias = layer_outputs[4 if self.output_attentions else 2]
 
             if self.output_attentions:
                 all_attentions = all_attentions + (layer_outputs[1],)
@@ -519,27 +553,6 @@ class T5Stack(nn.Module):
         return outputs  # last-layer hidden state, (all hidden states), (all attentions)
 
 
-class T5PreTrainedModel(PreTrainedEncoderDecoder):
-    """ An abstract class to handle weights initialization and
-        a simple interface for dowloading and loading pretrained models.
-    """
-    config_class = T5Config
-    pretrained_model_archive_map = T5_PRETRAINED_MODEL_ARCHIVE_MAP
-    load_tf_weights = load_tf_weights_in_t5
-
-    def _init_weights(self, module):
-        """ Initialize the weights """
-        if isinstance(module, (nn.Linear, nn.Embedding)):
-            # Slightly different from the TF version which uses truncated_normal for initialization
-            # cf https://github.com/pytorch/pytorch/pull/5617
-            module.weight.data.normal_(mean=0.0, std=self.config.initializer_range)
-        elif isinstance(module, nn.LayerNorm):
-            module.bias.data.zero_()
-            module.weight.data.fill_(1.0)
-        if isinstance(module, nn.Linear) and module.bias is not None:
-            module.bias.data.zero_()
-
-
 T5_START_DOCSTRING = r"""    The T5 model was proposed in
     `Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer`_
     by Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, Peter J. Liu.
@@ -620,7 +633,7 @@ class T5Model(T5PreTrainedModel):
     """
     def __init__(self, config):
         super(T5Model, self).__init__(config)
-        self.shared = nn.Embeddings(config.vocab_size, config.d_model)
+        self.shared = nn.Embedding(config.vocab_size, config.d_model)
 
         encoder_config = copy.deepcopy(config)
         self.encoder = T5Stack(encoder_config)
@@ -631,7 +644,6 @@ class T5Model(T5PreTrainedModel):
 
         self.init_weights()
 
-    @property
     def get_input_embeddings(self):
         return self.shared
 
@@ -646,17 +658,17 @@ class T5Model(T5PreTrainedModel):
         for layer, heads in heads_to_prune.items():
             self.encoder.layer[layer].attention.prune_heads(heads)
 
-    def forward(self, encoder_input_ids, decoder_input_ids, **kwargs):
+    def forward(self, **kwargs):
         # keyword arguments come in 3 flavors: encoder-specific (prefixed by
         # `encoder_`), decoder-specific (prefixed by `decoder_`) and those
         # that apply to the model as whole.
         # We let the specific kwargs override the common ones in case of conflict.
         kwargs_common = dict((k, v) for k, v in kwargs.items()
                              if not k.startswith("encoder_") and not k.startswith("decoder_"))
-        kwargs_decoder = kwargs_common.copy()
         kwargs_encoder = kwargs_common.copy()
-        kwargs_encoder.update(dict((k[len("encoder_") :], v) for k, v in kwargs.items() if k.startswith("encoder_")))
-        kwargs_decoder.update(dict((k[len("decoder_") :], v) for k, v in kwargs.items() if k.startswith("decoder_")))
+        kwargs_decoder = kwargs_common.copy()
+        kwargs_encoder.update(dict((k[len("encoder_"):], v) for k, v in kwargs.items() if k.startswith("encoder_")))
+        kwargs_decoder.update(dict((k[len("decoder_"):], v) for k, v in kwargs.items() if k.startswith("decoder_")))
 
         # Encode if needed (training, first prediction pass)
         encoder_hidden_states = kwargs_encoder.pop("hidden_states", None)
@@ -680,7 +692,7 @@ class T5Model(T5PreTrainedModel):
 
 @add_start_docstrings("""T5 Model with a `language modeling` head on top. """,
     T5_START_DOCSTRING, T5_INPUTS_DOCSTRING)
-class T5WithLMHead(T5PreTrainedModel):
+class T5WithLMHeadModel(T5PreTrainedModel):
     r"""
         **lm_labels**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
             Labels for computing the masked language modeling loss.
@@ -704,14 +716,14 @@ class T5WithLMHead(T5PreTrainedModel):
     Examples::
 
         tokenizer = T5Tokenizer.from_pretrained('t5-base-uncased')
-        model = T5ForMaskedLM.from_pretrained('t5-base-uncased')
+        model = T5WithLMHeadModel.from_pretrained('t5-base-uncased')
         input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids, lm_labels=input_ids)
         loss, prediction_scores = outputs[:2]
 
     """
     def __init__(self, config):
-        super(T5ForMaskedLM, self).__init__(config)
+        super(T5WithLMHeadModel, self).__init__(config)
 
         self.transformer = T5Model(config)
         self.lm_head = nn.Linear(config.d_model, config.vocab_size)
@@ -721,11 +733,12 @@ class T5WithLMHead(T5PreTrainedModel):
     def get_output_embeddings(self):
         return self.lm_head
 
-    def forward(self, encoder_input_ids, decoder_input_ids, **kwargs):
-        outputs = self.transformer(encoder_input_ids, decoder_input_ids, **kwargs)
+    def forward(self, **kwargs):
+        lm_labels = kwargs.pop('decoder_lm_labels', None)
+        outputs = self.transformer(**kwargs)
 
         sequence_output = outputs[0]
-        lm_logits = self.cls(sequence_output)
+        lm_logits = self.lm_head(sequence_output)
 
         outputs = (lm_logits,) + outputs[2:]  # Add hidden states and attention if they are here
         if lm_labels is not None:
diff --git a/transformers/tests/modeling_common_test.py b/transformers/tests/modeling_common_test.py
index ddc0f9f3de..42bf9ac3f5 100644
--- a/transformers/tests/modeling_common_test.py
+++ b/transformers/tests/modeling_common_test.py
@@ -73,6 +73,7 @@ class CommonTestCases:
         test_pruning = True
         test_resize_embeddings = True
         test_head_masking = True
+        is_encoder_decoder = False
 
         def test_save_load(self):
             config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
@@ -114,10 +115,9 @@ class CommonTestCases:
             for model_class in self.all_model_classes:
                 model = model_class(config)
                 model.eval()
-                first, second = model(inputs_dict["input_ids"])[0], model(inputs_dict["input_ids"])[0]
+                first, second = model(**inputs_dict)[0], model(**inputs_dict)[0]
                 self.assertEqual(first.ne(second).sum().item(), 0)
 
-
         def test_attention_outputs(self):
             config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
 
@@ -127,31 +127,42 @@ class CommonTestCases:
                 model = model_class(config)
                 model.eval()
                 outputs = model(**inputs_dict)
-                attentions = outputs[-1]
+                self_attentions = outputs[-1]
                 self.assertEqual(model.config.output_attentions, True)
                 self.assertEqual(model.config.output_hidden_states, False)
-                self.assertEqual(len(attentions), self.model_tester.num_hidden_layers)
+                self.assertEqual(len(self_attentions), self.model_tester.num_hidden_layers)
                 self.assertListEqual(
-                    list(attentions[0].shape[-3:]),
+                    list(self_attentions[0].shape[-3:]),
                     [self.model_tester.num_attention_heads,
                     self.model_tester.seq_length,
                     self.model_tester.key_len if hasattr(self.model_tester, 'key_len') else self.model_tester.seq_length])
                 out_len = len(outputs)
 
+                if self.is_encoder_decoder:
+                    cross_attentions = outputs[-2]
+                    self.assertEqual(model.config.output_attentions, True)
+                    self.assertEqual(model.config.output_hidden_states, False)
+                    self.assertEqual(len(cross_attentions), self.model_tester.num_hidden_layers)
+                    self.assertListEqual(
+                        list(cross_attentions[0].shape[-3:]),
+                        [self.model_tester.num_attention_heads,
+                        self.model_tester.seq_length,
+                        self.model_tester.key_len if hasattr(self.model_tester, 'key_len') else self.model_tester.seq_length])
+
                 # Check attention is always last and order is fine
                 config.output_attentions = True
                 config.output_hidden_states = True
                 model = model_class(config)
                 model.eval()
                 outputs = model(**inputs_dict)
-                self.assertEqual(out_len+1, len(outputs))
+                self.assertEqual(out_len + (2 if self.is_encoder_decoder else 1), len(outputs))
                 self.assertEqual(model.config.output_attentions, True)
                 self.assertEqual(model.config.output_hidden_states, True)
 
-                attentions = outputs[-1]
-                self.assertEqual(len(attentions), self.model_tester.num_hidden_layers)
+                self_attentions = outputs[-1]
+                self.assertEqual(len(self_attentions), self.model_tester.num_hidden_layers)
                 self.assertListEqual(
-                    list(attentions[0].shape[-3:]),
+                    list(self_attentions[0].shape[-3:]),
                     [self.model_tester.num_attention_heads,
                     self.model_tester.seq_length,
                     self.model_tester.key_len if hasattr(self.model_tester, 'key_len') else self.model_tester.seq_length])
@@ -214,7 +225,6 @@ class CommonTestCases:
 
                 self.assertTrue(models_equal)
 
-
         def test_headmasking(self):
             if not self.test_head_masking:
                 return
@@ -268,7 +278,6 @@ class CommonTestCases:
                 self.assertNotEqual(
                     attentions[-1][..., -1, :, :].flatten().sum().item(), 0.0)
 
-
         def test_head_pruning(self):
             if not self.test_pruning:
                 return
@@ -411,7 +420,6 @@ class CommonTestCases:
 
                 self.assertDictEqual(model.config.pruned_heads, {0: [0], 1: [1, 2], 2: [1, 2]})
 
-
         def test_hidden_states_output(self):
             config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
 
diff --git a/transformers/tests/modeling_t5_test.py b/transformers/tests/modeling_t5_test.py
new file mode 100644
index 0000000000..b8bb828ebd
--- /dev/null
+++ b/transformers/tests/modeling_t5_test.py
@@ -0,0 +1,176 @@
+# coding=utf-8
+# Copyright 2018 Google T5 Authors and HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from __future__ import absolute_import
+from __future__ import division
+from __future__ import print_function
+
+import unittest
+import shutil
+import pytest
+
+from transformers import is_torch_available
+
+from .modeling_common_test import (CommonTestCases, ids_tensor)
+from .configuration_common_test import ConfigTester
+
+if is_torch_available():
+    from transformers import (T5Config, T5Model, T5WithLMHeadModel)
+    from transformers.modeling_t5 import T5_PRETRAINED_MODEL_ARCHIVE_MAP
+else:
+    pytestmark = pytest.mark.skip("Require Torch")
+
+
+class T5ModelTest(CommonTestCases.CommonModelTester):
+
+    all_model_classes = (T5Model, T5WithLMHeadModel) if is_torch_available() else ()
+    test_pruning = False
+    test_torchscript = False
+    test_resize_embeddings = False
+    is_encoder_decoder = True
+
+    class T5ModelTester(object):
+
+        def __init__(self,
+                     parent,
+                     batch_size=13,
+                     seq_length=7,
+                     is_training=True,
+                     use_input_mask=True,
+                     use_labels=True,
+                     vocab_size=99,
+                     n_positions=14,
+                     hidden_size=32,
+                     num_hidden_layers=5,
+                     num_attention_heads=4,
+                     d_ff=37,
+                     relative_attention_num_buckets=8,
+                     dropout_rate=0.1,
+                     initializer_range=0.02,
+                     scope=None,
+                    ):
+            self.parent = parent
+            self.batch_size = batch_size
+            self.seq_length = seq_length
+            self.is_training = is_training
+            self.use_input_mask = use_input_mask
+            self.use_labels = use_labels
+            self.vocab_size = vocab_size
+            self.n_positions = n_positions
+            self.hidden_size = hidden_size
+            self.num_hidden_layers = num_hidden_layers
+            self.num_attention_heads = num_attention_heads
+            self.d_ff = d_ff
+            self.relative_attention_num_buckets = relative_attention_num_buckets
+            self.dropout_rate = dropout_rate
+            self.initializer_range = initializer_range
+            self.scope = scope
+
+        def prepare_config_and_inputs(self):
+            input_ids = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)
+
+            input_mask = None
+            if self.use_input_mask:
+                input_mask = ids_tensor([self.batch_size, self.seq_length], vocab_size=2)
+
+            token_labels = None
+            if self.use_labels:
+                token_labels = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)
+
+            config = T5Config(
+                vocab_size_or_config_json_file=self.vocab_size,
+                n_positions=self.n_positions,
+                d_model=self.hidden_size,
+                d_ff=self.d_ff,
+                num_layers=self.num_hidden_layers,
+                num_heads=self.num_attention_heads,
+                relative_attention_num_buckets=self.relative_attention_num_buckets,
+                dropout_rate=self.dropout_rate,
+                initializer_range=self.initializer_range)
+
+            return (config, input_ids, input_mask, token_labels)
+
+        def check_loss_output(self, result):
+            self.parent.assertListEqual(
+                list(result["loss"].size()),
+                [])
+
+        def create_and_check_t5_model(self, config, input_ids, input_mask, token_labels):
+            model = T5Model(config=config)
+            model.eval()
+            encoder_output, decoder_output = model(encoder_input_ids=input_ids,
+                                                   decoder_input_ids=input_ids,
+                                                   decoder_attention_mask=input_mask)
+            encoder_output, decoder_output = model(encoder_input_ids=input_ids,
+                                                   decoder_input_ids=input_ids)
+
+            result = {
+                "encoder_output": encoder_output,
+                "decoder_output": decoder_output,
+            }
+            self.parent.assertListEqual(
+                list(result["encoder_output"].size()),
+                [self.batch_size, self.seq_length, self.hidden_size])
+            self.parent.assertListEqual(
+                list(result["decoder_output"].size()),
+                [self.batch_size, self.seq_length, self.hidden_size])
+
+
+        def create_and_check_t5_with_lm_head(self, config, input_ids, input_mask, token_labels):
+            model = T5WithLMHeadModel(config=config)
+            model.eval()
+            loss, prediction_scores = model(encoder_input_ids=input_ids, decoder_input_ids=input_ids,
+                                            decoder_attention_mask=input_mask, decoder_lm_labels=token_labels)
+            result = {
+                "loss": loss,
+                "prediction_scores": prediction_scores,
+            }
+            self.parent.assertListEqual(
+                list(result["prediction_scores"].size()),
+                [self.batch_size, self.seq_length, self.vocab_size])
+            self.check_loss_output(result)
+
+        def prepare_config_and_inputs_for_common(self):
+            config_and_inputs = self.prepare_config_and_inputs()
+            (config, input_ids, input_mask, token_labels) = config_and_inputs
+            inputs_dict = {'encoder_input_ids': input_ids,
+                           'decoder_input_ids': input_ids,
+                           'decoder_attention_mask': input_mask}
+            return config, inputs_dict
+
+    def setUp(self):
+        self.model_tester = T5ModelTest.T5ModelTester(self)
+        self.config_tester = ConfigTester(self, config_class=T5Config, d_model=37)
+
+    def test_config(self):
+        self.config_tester.run_common_tests()
+
+    def test_t5_model(self):
+        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+        self.model_tester.create_and_check_t5_model(*config_and_inputs)
+
+    def test_with_lm_head(self):
+        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+        self.model_tester.create_and_check_t5_with_lm_head(*config_and_inputs)
+
+    @pytest.mark.slow
+    def test_model_from_pretrained(self):
+        cache_dir = "/tmp/transformers_test/"
+        for model_name in list(T5_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
+            model = T5Model.from_pretrained(model_name, cache_dir=cache_dir)
+            shutil.rmtree(cache_dir)
+            self.assertIsNotNone(model)
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/transformers/tests/modeling_tf_t5_test.py b/transformers/tests/modeling_tf_t5_test.py
new file mode 100644
index 0000000000..fac6763432
--- /dev/null
+++ b/transformers/tests/modeling_tf_t5_test.py
@@ -0,0 +1,190 @@
+# coding=utf-8
+# Copyright 2018 Google T5 Authors and HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from __future__ import absolute_import
+from __future__ import division
+from __future__ import print_function
+
+import unittest
+import shutil
+import pytest
+import sys
+
+from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
+from .configuration_common_test import ConfigTester
+
+from transformers import T5Config, is_tf_available
+
+if False:  # is_tf_available():
+    import tensorflow as tf
+    from transformers.modeling_tf_t5 import (TFT5Model, TFT5WithLMHeadModel,TF_T5_PRETRAINED_MODEL_ARCHIVE_MAP)
+else:
+    pytestmark = pytest.mark.skip("Require TensorFlow")
+
+
+class TFT5ModelTest(TFCommonTestCases.TFCommonModelTester):
+
+    all_model_classes = (TFT5Model, TFT5WithLMHeadModel) if False  else () # is_tf_available() else ()
+
+    class TFT5ModelTester(object):
+
+        def __init__(self,
+                     parent,
+                     batch_size=13,
+                     seq_length=7,
+                     is_training=True,
+                     use_input_mask=True,
+                     use_token_type_ids=True,
+                     use_labels=True,
+                     vocab_size=99,
+                     hidden_size=32,
+                     num_hidden_layers=5,
+                     num_attention_heads=4,
+                     intermediate_size=37,
+                     hidden_act="gelu",
+                     hidden_dropout_prob=0.1,
+                     attention_probs_dropout_prob=0.1,
+                     max_position_embeddings=512,
+                     type_vocab_size=16,
+                     type_sequence_label_size=2,
+                     initializer_range=0.02,
+                     num_labels=3,
+                     num_choices=4,
+                     scope=None,
+                    ):
+            self.parent = parent
+            self.batch_size = batch_size
+            self.seq_length = seq_length
+            self.is_training = is_training
+            self.use_input_mask = use_input_mask
+            self.use_token_type_ids = use_token_type_ids
+            self.use_labels = use_labels
+            self.vocab_size = vocab_size
+            self.hidden_size = hidden_size
+            self.num_hidden_layers = num_hidden_layers
+            self.num_attention_heads = num_attention_heads
+            self.intermediate_size = intermediate_size
+            self.hidden_act = hidden_act
+            self.hidden_dropout_prob = hidden_dropout_prob
+            self.attention_probs_dropout_prob = attention_probs_dropout_prob
+            self.max_position_embeddings = max_position_embeddings
+            self.type_vocab_size = type_vocab_size
+            self.type_sequence_label_size = type_sequence_label_size
+            self.initializer_range = initializer_range
+            self.num_labels = num_labels
+            self.num_choices = num_choices
+            self.scope = scope
+
+        def prepare_config_and_inputs(self):
+            input_ids = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)
+
+            input_mask = None
+            if self.use_input_mask:
+                input_mask = ids_tensor([self.batch_size, self.seq_length], vocab_size=2)
+
+            token_type_ids = None
+            if self.use_token_type_ids:
+                token_type_ids = ids_tensor([self.batch_size, self.seq_length], self.type_vocab_size)
+
+            sequence_labels = None
+            token_labels = None
+            choice_labels = None
+            if self.use_labels:
+                sequence_labels = ids_tensor([self.batch_size], self.type_sequence_label_size)
+                token_labels = ids_tensor([self.batch_size, self.seq_length], self.num_labels)
+                choice_labels = ids_tensor([self.batch_size], self.num_choices)
+
+            config = T5Config(
+                vocab_size_or_config_json_file=self.vocab_size,
+                hidden_size=self.hidden_size,
+                num_hidden_layers=self.num_hidden_layers,
+                num_attention_heads=self.num_attention_heads,
+                intermediate_size=self.intermediate_size,
+                hidden_act=self.hidden_act,
+                hidden_dropout_prob=self.hidden_dropout_prob,
+                attention_probs_dropout_prob=self.attention_probs_dropout_prob,
+                max_position_embeddings=self.max_position_embeddings,
+                type_vocab_size=self.type_vocab_size,
+                initializer_range=self.initializer_range)
+
+            return config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
+
+        def create_and_check_t5_model(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
+            model = TFT5Model(config=config)
+            inputs = {'input_ids': input_ids,
+                      'attention_mask': input_mask,
+                      'token_type_ids': token_type_ids}
+            sequence_output, pooled_output = model(inputs)
+
+            inputs = [input_ids, input_mask]
+            sequence_output, pooled_output = model(inputs)
+
+            sequence_output, pooled_output = model(input_ids)
+
+            result = {
+                "sequence_output": sequence_output.numpy(),
+                "pooled_output": pooled_output.numpy(),
+            }
+            self.parent.assertListEqual(
+                list(result["sequence_output"].shape),
+                [self.batch_size, self.seq_length, self.hidden_size])
+            self.parent.assertListEqual(list(result["pooled_output"].shape), [self.batch_size, self.hidden_size])
+
+
+        def create_and_check_t5_with_lm_head(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
+            model = TFT5WithLMHeadModel(config=config)
+            inputs = {'input_ids': input_ids,
+                      'attention_mask': input_mask,
+                      'token_type_ids': token_type_ids}
+            prediction_scores, = model(inputs)
+            result = {
+                "prediction_scores": prediction_scores.numpy(),
+            }
+            self.parent.assertListEqual(
+                list(result["prediction_scores"].shape),
+                [self.batch_size, self.seq_length, self.vocab_size])
+
+
+        def prepare_config_and_inputs_for_common(self):
+            config_and_inputs = self.prepare_config_and_inputs()
+            (config, input_ids, token_type_ids, input_mask,
+             sequence_labels, token_labels, choice_labels) = config_and_inputs
+            inputs_dict = {'input_ids': input_ids, 'token_type_ids': token_type_ids, 'attention_mask': input_mask}
+            return config, inputs_dict
+
+    def setUp(self):
+        self.model_tester = TFT5ModelTest.TFT5ModelTester(self)
+        self.config_tester = ConfigTester(self, config_class=T5Config, hidden_size=37)
+
+    def test_config(self):
+        self.config_tester.run_common_tests()
+
+    def test_t5_model(self):
+        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+        self.model_tester.create_and_check_t5_model(*config_and_inputs)
+
+    def test_with_lm_head(self):
+        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+        self.model_tester.create_and_check_t5_with_lm_head(*config_and_inputs)
+
+    @pytest.mark.slow
+    def test_model_from_pretrained(self):
+        cache_dir = "/tmp/transformers_test/"
+        for model_name in ['t5-base']:
+            model = TFT5Model.from_pretrained(model_name, cache_dir=cache_dir)
+            shutil.rmtree(cache_dir)
+            self.assertIsNotNone(model)
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/transformers/tests/tokenization_t5_test.py b/transformers/tests/tokenization_t5_test.py
new file mode 100644
index 0000000000..9362487d8d
--- /dev/null
+++ b/transformers/tests/tokenization_t5_test.py
@@ -0,0 +1,77 @@
+# coding=utf-8
+# Copyright 2018 Google T5 Authors and HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from __future__ import absolute_import, division, print_function, unicode_literals
+
+import os
+import unittest
+import pytest
+
+from transformers.tokenization_t5 import (T5Tokenizer, SPIECE_UNDERLINE)
+
+from .tokenization_tests_commons import CommonTestCases
+
+SAMPLE_VOCAB = os.path.join(os.path.dirname(os.path.abspath(__file__)),
+                    'fixtures/test_sentencepiece.model')
+
+class T5TokenizationTest(CommonTestCases.CommonTokenizerTester):
+
+    tokenizer_class = T5Tokenizer
+
+    def setUp(self):
+        super(T5TokenizationTest, self).setUp()
+
+        # We have a SentencePiece fixture for testing
+        tokenizer = T5Tokenizer(SAMPLE_VOCAB, keep_accents=True)
+        tokenizer.save_pretrained(self.tmpdirname)
+
+    def get_tokenizer(self, **kwargs):
+        return T5Tokenizer.from_pretrained(self.tmpdirname, **kwargs)
+
+    def get_input_output_texts(self):
+        input_text = u"This is a test"
+        output_text = u"This is a test"
+        return input_text, output_text
+
+    def test_full_tokenizer(self):
+        tokenizer = T5Tokenizer(SAMPLE_VOCAB, keep_accents=True)
+
+        tokens = tokenizer.tokenize(u'This is a test')
+        self.assertListEqual(tokens, [u'▁This', u'▁is', u'▁a', u'▁t', u'est'])
+
+        self.assertListEqual(
+            tokenizer.convert_tokens_to_ids(tokens), [285, 46, 10, 170, 382])
+
+        tokens = tokenizer.tokenize(u"I was born in 92000, and this is falsé.")
+        self.assertListEqual(tokens, [SPIECE_UNDERLINE + u'I', SPIECE_UNDERLINE + u'was', SPIECE_UNDERLINE + u'b',
+                                    u'or', u'n', SPIECE_UNDERLINE + u'in', SPIECE_UNDERLINE + u'',
+                                    u'9', u'2', u'0', u'0', u'0', u',', SPIECE_UNDERLINE + u'and', SPIECE_UNDERLINE + u'this',
+                                    SPIECE_UNDERLINE + u'is', SPIECE_UNDERLINE + u'f', u'al', u's', u'é', u'.'])
+        ids = tokenizer.convert_tokens_to_ids(tokens)
+        self.assertListEqual(
+            ids, [8, 21, 84, 55, 24, 19, 7, 0,
+                602, 347, 347, 347, 3, 12, 66,
+                46, 72, 80, 6, 0, 4])
+
+        back_tokens = tokenizer.convert_ids_to_tokens(ids)
+        self.assertListEqual(back_tokens, [SPIECE_UNDERLINE + u'I', SPIECE_UNDERLINE + u'was', SPIECE_UNDERLINE + u'b',
+                                        u'or', u'n', SPIECE_UNDERLINE + u'in',
+                                        SPIECE_UNDERLINE + u'', u'<unk>', u'2', u'0', u'0', u'0', u',',
+                                        SPIECE_UNDERLINE + u'and', SPIECE_UNDERLINE + u'this',
+                                        SPIECE_UNDERLINE + u'is', SPIECE_UNDERLINE + u'f', u'al', u's',
+                                        u'<unk>', u'.'])
+
+
+if __name__ == '__main__':
+    unittest.main()

From 070dcf1c020e28e96e8e4b5acfb29fb818b0b4dd Mon Sep 17 00:00:00 2001
From: Diganta Misra <mishradiganta91@gmail.com>
Date: Thu, 7 Nov 2019 03:45:43 +0530
Subject: [PATCH 017/505] Added Mish Activation Function

Mish is a new activation function proposed here - https://arxiv.org/abs/1908.08681
It has seen some recent success and has been adopted in SpaCy, Thic, TensorFlow Addons and FastAI-dev.
All benchmarks recorded till now (including against ReLU, Swish and GELU) is present in the repository - https://github.com/digantamisra98/Mish
Might be a good addition to experiment with especially in the Bert Model.
---
 transformers/modeling_bert.py | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/transformers/modeling_bert.py b/transformers/modeling_bert.py
index 7c2c6f4602..2baee71f82 100644
--- a/transformers/modeling_bert.py
+++ b/transformers/modeling_bert.py
@@ -138,7 +138,11 @@ def swish(x):
     return x * torch.sigmoid(x)
 
 
-ACT2FN = {"gelu": gelu, "relu": torch.nn.functional.relu, "swish": swish, "gelu_new": gelu_new}
+def mish(x):
+    return x * torch.tanh(nn.functional.softplus(x))
+
+
+ACT2FN = {"gelu": gelu, "relu": torch.nn.functional.relu, "swish": swish, "gelu_new": gelu_new, "mish": mish}
 
 
 BertLayerNorm = torch.nn.LayerNorm

From ba10065c4b44d733d135ad6dc1b8a77f88c6dbb9 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Thu, 7 Nov 2019 15:55:36 +0100
Subject: [PATCH 018/505] update model, conversion script, tests and template

---
 ...t_xxx_original_tf_checkpoint_to_pytorch.py |  10 +-
 transformers/__init__.py                      |   1 +
 transformers/configuration_t5.py              |  13 +-
 ...rt_t5_original_tf_checkpoint_to_pytorch.py |  12 +-
 transformers/modeling_t5.py                   | 129 ++++++++++++------
 transformers/tests/modeling_common_test.py    |  41 +++---
 transformers/tests/modeling_t5_test.py        |  12 +-
 transformers/tokenization_t5.py               |   1 +
 8 files changed, 135 insertions(+), 84 deletions(-)

diff --git a/templates/adding_a_new_model/convert_xxx_original_tf_checkpoint_to_pytorch.py b/templates/adding_a_new_model/convert_xxx_original_tf_checkpoint_to_pytorch.py
index d50d129cba..9d389deaad 100755
--- a/templates/adding_a_new_model/convert_xxx_original_tf_checkpoint_to_pytorch.py
+++ b/templates/adding_a_new_model/convert_xxx_original_tf_checkpoint_to_pytorch.py
@@ -26,9 +26,9 @@ from transformers import XxxConfig, XxxForPreTraining, load_tf_weights_in_xxx
 import logging
 logging.basicConfig(level=logging.INFO)
 
-def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, xxx_config_file, pytorch_dump_path):
+def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, config_file, pytorch_dump_path):
     # Initialise PyTorch model
-    config = XxxConfig.from_json_file(xxx_config_file)
+    config = XxxConfig.from_json_file(config_file)
     print("Building PyTorch model from configuration: {}".format(str(config)))
     model = XxxForPreTraining(config)
 
@@ -48,11 +48,11 @@ if __name__ == "__main__":
                         type = str,
                         required = True,
                         help = "Path to the TensorFlow checkpoint path.")
-    parser.add_argument("--xxx_config_file",
+    parser.add_argument("--config_file",
                         default = None,
                         type = str,
                         required = True,
-                        help = "The config json file corresponding to the pre-trained XXX model. \n"
+                        help = "The config json file corresponding to the pre-trained model. \n"
                             "This specifies the model architecture.")
     parser.add_argument("--pytorch_dump_path",
                         default = None,
@@ -61,5 +61,5 @@ if __name__ == "__main__":
                         help = "Path to the output PyTorch model.")
     args = parser.parse_args()
     convert_tf_checkpoint_to_pytorch(args.tf_checkpoint_path,
-                                     args.xxx_config_file,
+                                     args.config_file,
                                      args.pytorch_dump_path)
diff --git a/transformers/__init__.py b/transformers/__init__.py
index bf896276d6..601a068592 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -97,6 +97,7 @@ if is_torch_available():
                                 DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
     from .modeling_encoder_decoder import PreTrainedEncoderDecoder, Model2Model
     from .modeling_t5 import (T5PreTrainedModel, T5Model, T5WithLMHeadModel,
+                              load_tf_weights_in_t5,
                               T5_PRETRAINED_MODEL_ARCHIVE_MAP)
 
     # Optimization
diff --git a/transformers/configuration_t5.py b/transformers/configuration_t5.py
index 9db918e59f..96e67758ac 100644
--- a/transformers/configuration_t5.py
+++ b/transformers/configuration_t5.py
@@ -57,8 +57,7 @@ class T5Config(PretrainedConfig):
                 (e.g., 512 or 1024 or 2048).
             type_vocab_size: The vocabulary size of the `token_type_ids` passed into
                 `T5Model`.
-            initializer_range: The sttdev of the truncated_normal_initializer for
-                initializing all weight matrices.
+            initializer_factor: A factor for initializing all weight matrices (should be kept to 1.0, used for initialization testing).
             layer_norm_eps: The epsilon used by LayerNorm.
     """
     pretrained_config_archive_map = T5_PRETRAINED_CONFIG_ARCHIVE_MAP
@@ -67,25 +66,27 @@ class T5Config(PretrainedConfig):
                  vocab_size_or_config_json_file=32128,
                  n_positions=512,
                  d_model=512,
+                 d_kv=64,
                  d_ff=2048,
-                 num_layers=12,
-                 num_heads=12,
+                 num_layers=6,
+                 num_heads=8,
                  relative_attention_num_buckets=32,
                  dropout_rate=0.1,
                  layer_norm_epsilon=1e-6,
-                 initializer_range=0.02,
+                 initializer_factor=1.0,
                  **kwargs):
         super(T5Config, self).__init__(**kwargs)
         self.vocab_size = vocab_size_or_config_json_file if isinstance(vocab_size_or_config_json_file, int) else -1
         self.n_positions = n_positions
         self.d_model = d_model
+        self.d_kv = d_kv
         self.d_ff = d_ff
         self.num_layers = num_layers
         self.num_heads = num_heads
         self.relative_attention_num_buckets = relative_attention_num_buckets
         self.dropout_rate = dropout_rate
         self.layer_norm_epsilon = layer_norm_epsilon
-        self.initializer_range = initializer_range
+        self.initializer_factor = initializer_factor
 
         if isinstance(vocab_size_or_config_json_file, six.string_types):
             with open(vocab_size_or_config_json_file, "r", encoding="utf-8") as reader:
diff --git a/transformers/convert_t5_original_tf_checkpoint_to_pytorch.py b/transformers/convert_t5_original_tf_checkpoint_to_pytorch.py
index 608027ebac..2b74d2dd93 100755
--- a/transformers/convert_t5_original_tf_checkpoint_to_pytorch.py
+++ b/transformers/convert_t5_original_tf_checkpoint_to_pytorch.py
@@ -21,16 +21,16 @@ from __future__ import print_function
 import argparse
 import torch
 
-from transformers import T5Config, T5ForPreTraining, load_tf_weights_in_t5
+from transformers import T5Config, T5Model, load_tf_weights_in_t5
 
 import logging
 logging.basicConfig(level=logging.INFO)
 
-def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, t5_config_file, pytorch_dump_path):
+def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, config_file, pytorch_dump_path):
     # Initialise PyTorch model
-    config = T5Config.from_json_file(t5_config_file)
+    config = T5Config.from_json_file(config_file)
     print("Building PyTorch model from configuration: {}".format(str(config)))
-    model = T5ForPreTraining(config)
+    model = T5Model(config)
 
     # Load weights from tf checkpoint
     load_tf_weights_in_t5(model, config, tf_checkpoint_path)
@@ -48,7 +48,7 @@ if __name__ == "__main__":
                         type = str,
                         required = True,
                         help = "Path to the TensorFlow checkpoint path.")
-    parser.add_argument("--t5_config_file",
+    parser.add_argument("--config_file",
                         default = None,
                         type = str,
                         required = True,
@@ -61,5 +61,5 @@ if __name__ == "__main__":
                         help = "Path to the output PyTorch model.")
     args = parser.parse_args()
     convert_tf_checkpoint_to_pytorch(args.tf_checkpoint_path,
-                                     args.t5_config_file,
+                                     args.config_file,
                                      args.pytorch_dump_path)
diff --git a/transformers/modeling_t5.py b/transformers/modeling_t5.py
index ce443cf882..6ed241761a 100644
--- a/transformers/modeling_t5.py
+++ b/transformers/modeling_t5.py
@@ -65,34 +65,40 @@ def load_tf_weights_in_t5(model, config, tf_checkpoint_path):
     # Load weights from TF model
     init_vars = tf.train.list_variables(tf_path)
     names = []
-    arrays = []
+    tf_weights = {}
     for name, shape in init_vars:
         logger.info("Loading TF weight {} with shape {}".format(name, shape))
         array = tf.train.load_variable(tf_path, name)
         names.append(name)
-        arrays.append(array)
+        tf_weights[name] = array
 
-    for name, array in zip(names, arrays):
-        name = name.split('/')
+    for txt_name in names:
+        name = txt_name.split('/')
         # adam_v and adam_m are variables used in AdamWeightDecayOptimizer to calculated m and v
         # which are not required for using pretrained model
         if any(n in ["adam_v", "adam_m", "global_step"] for n in name):
             logger.info("Skipping {}".format("/".join(name)))
+            tf_weights.pop(txt_name, None)
+            continue
+        if '_slot_' in name[-1]:
+            logger.info("Skipping {}".format("/".join(name)))
+            tf_weights.pop(txt_name, None)
             continue
         pointer = model
+        array = tf_weights[txt_name]
         for m_name in name:
             if re.fullmatch(r'[A-Za-z]+_\d+', m_name):
                 l = re.split(r'_(\d+)', m_name)
             else:
                 l = [m_name]
-            if l[0] == 'kernel' or l[0] == 'gamma':
+            if l[0] in ['kernel', 'scale', 'embedding']:
                 pointer = getattr(pointer, 'weight')
-            elif l[0] == 'output_bias' or l[0] == 'beta':
-                pointer = getattr(pointer, 'bias')
-            elif l[0] == 'output_weights':
-                pointer = getattr(pointer, 'weight')
-            elif l[0] == 'squad':
-                pointer = getattr(pointer, 'classifier')
+            # elif l[0] == 'scale':
+            #     pointer = getattr(pointer, 'weight')
+            # elif l[0] == 'output_bias' or l[0] == 'beta':
+            #     pointer = getattr(pointer, 'bias')
+            # elif l[0] == 'squad':
+            #     pointer = getattr(pointer, 'classifier')
             else:
                 try:
                     pointer = getattr(pointer, l[0])
@@ -102,9 +108,10 @@ def load_tf_weights_in_t5(model, config, tf_checkpoint_path):
             if len(l) >= 2:
                 num = int(l[1])
                 pointer = pointer[num]
-        if m_name[-11:] == '_embeddings':
+        if l[0] not in ['kernel', 'scale', 'embedding']:
             pointer = getattr(pointer, 'weight')
-        elif m_name == 'kernel':
+        if l[0] != 'embedding':
+            logger.info("Transposing numpy weight of shape {} for {}".format(array.shape, name))
             array = np.transpose(array)
         try:
             assert pointer.shape == array.shape
@@ -112,7 +119,11 @@ def load_tf_weights_in_t5(model, config, tf_checkpoint_path):
             e.args += (pointer.shape, array.shape)
             raise
         logger.info("Initialize PyTorch weight {}".format(name))
-        pointer.data = torch.from_numpy(array)
+        pointer.data = torch.from_numpy(array.astype(np.float32))
+        tf_weights.pop(txt_name, None)
+
+    logger.info("Weights not copied to PyTorch model: {}".format(', '.join(tf_weights.keys())))
+    # logger.info("Weights not copied to PyTorch model: {}".format(', '.join(tf_weights.keys())))
     return model
 
 
@@ -163,10 +174,13 @@ class T5Attention(nn.Module):
         self.output_attentions = config.output_attentions
         self.relative_attention_num_buckets = config.relative_attention_num_buckets
         self.dim = config.d_model
+        self.d_kv = config.d_kv
         self.n_heads = config.num_heads
         self.dropout = config.dropout_rate
         assert self.dim % self.n_heads == 0
+        assert self.dim // self.n_heads == self.d_kv
 
+        # Mesh TensorFlow initialization to avoid scaling before softmax
         self.q = nn.Linear(self.dim, self.dim, bias=False)
         self.k = nn.Linear(self.dim, self.dim, bias=False)
         self.v = nn.Linear(self.dim, self.dim, bias=False)
@@ -312,8 +326,9 @@ class T5Attention(nn.Module):
         scores += position_bias
 
         if mask is not None:
-            mask = (mask == 0).expand_as(scores)                              # (bs, n_heads, qlen, klen)
-            scores.masked_fill_(mask, -float('inf'))                          # (bs, n_heads, qlen, klen)
+            scores += mask
+            # mask = (mask == 0).expand_as(scores)                              # (bs, n_heads, qlen, klen)
+            # scores.masked_fill_(mask, -float('inf'))                          # (bs, n_heads, qlen, klen)
 
         weights = F.softmax(scores.float(), dim=-1).type_as(scores)           # (bs, n_heads, qlen, klen)
         weights = F.dropout(weights, p=self.dropout, training=self.training)  # (bs, n_heads, qlen, klen)
@@ -378,34 +393,35 @@ class T5Block(nn.Module):
     def __init__(self, config, has_relative_attention_bias=False):
         super(T5Block, self).__init__()
         self.is_decoder = config.is_decoder
-        self.layer_000 = T5LayerSelfAttention(config, has_relative_attention_bias=has_relative_attention_bias)
+        self.layer = nn.ModuleList()
+        self.layer.append(T5LayerSelfAttention(config, has_relative_attention_bias=has_relative_attention_bias))
         if self.is_decoder:
-            self.layer_001 = T5LayerCrossAttention(config, has_relative_attention_bias=has_relative_attention_bias)
-            self.layer_002 = T5LayerFF(config)
+            self.layer.append(T5LayerCrossAttention(config, has_relative_attention_bias=has_relative_attention_bias))
+            self.layer.append(T5LayerFF(config))
         else:
-            self.layer_001 = T5LayerFF(config)
+            self.layer.append(T5LayerFF(config))
 
     def forward(self, hidden_states, attention_mask=None, position_bias=None,
                 encoder_hidden_states=None, encoder_attention_mask=None, encoder_decoder_position_bias=None,
                 head_mask=None):
-        self_attention_outputs = self.layer_000(hidden_states,
+        self_attention_outputs = self.layer[0](hidden_states,
                                                 attention_mask=attention_mask,
                                                 position_bias=position_bias,
                                                 head_mask=head_mask)
         hidden_states = self_attention_outputs[0]
         outputs = self_attention_outputs[1:]
 
-        if self.is_decoder:
-            cross_attention_outputs = self.layer_001(hidden_states,
-                                                     kv=encoder_hidden_states,
-                                                     attention_mask=encoder_attention_mask,
-                                                     position_bias=encoder_decoder_position_bias,
-                                                     head_mask=head_mask)
+        if not self.is_decoder:
+            hidden_states = self.layer[1](hidden_states)
+        else:
+            cross_attention_outputs = self.layer[1](hidden_states,
+                                                    kv=encoder_hidden_states,
+                                                    attention_mask=encoder_attention_mask,
+                                                    position_bias=encoder_decoder_position_bias,
+                                                    head_mask=head_mask)
             hidden_states = cross_attention_outputs[0]
             outputs = cross_attention_outputs[1:] + outputs
-            hidden_states = self.layer_002(hidden_states)
-        else:
-            hidden_states = self.layer_001(hidden_states)
+            hidden_states = self.layer[2](hidden_states)
 
         outputs = (hidden_states,) + outputs  # add attentions if we output them
         return outputs
@@ -422,15 +438,36 @@ class T5PreTrainedModel(PreTrainedModel):
 
     def _init_weights(self, module):
         """ Initialize the weights """
-        if isinstance(module, (nn.Linear, nn.Embedding)):
-            # Slightly different from the TF version which uses truncated_normal for initialization
-            # cf https://github.com/pytorch/pytorch/pull/5617
-            module.weight.data.normal_(mean=0.0, std=self.config.initializer_range)
-        elif isinstance(module, nn.LayerNorm):
-            module.bias.data.zero_()
-            module.weight.data.fill_(1.0)
-        if isinstance(module, nn.Linear) and module.bias is not None:
+        factor = self.config.initializer_factor  # Used for testing weights initialization
+        if isinstance(module, nn.LayerNorm):
             module.bias.data.zero_()
+            module.weight.data.fill_(factor*1.0)
+        elif isinstance(module, T5Model):
+            # Mesh TensorFlow embeddings initialization
+            # See https://github.com/tensorflow/mesh/blob/fa19d69eafc9a482aff0b59ddd96b025c0cb207d/mesh_tensorflow/layers.py#L1624
+            module.shared.weight.data.normal_(mean=0.0, std=factor*1.0)
+        elif isinstance(module, T5DenseReluDense):
+            # Mesh TensorFlow FF initialization
+            # See https://github.com/tensorflow/mesh/blob/master/mesh_tensorflow/transformer/transformer_layers.py#L56
+            # and https://github.com/tensorflow/mesh/blob/fa19d69eafc9a482aff0b59ddd96b025c0cb207d/mesh_tensorflow/layers.py#L89
+            module.wi.weight.data.normal_(mean=0.0, std=factor*((self.config.d_model) ** -0.5))
+            if hasattr(module.wi, 'bias') and module.wi.bias is not None:
+                module.wi.bias.data.zero_()
+            module.wo.weight.data.normal_(mean=0.0, std=factor*((self.config.d_ff) ** -0.5))
+            if hasattr(module.wo, 'bias') and module.wo.bias is not None:
+                module.wo.bias.data.zero_()
+        elif isinstance(module, T5Attention):
+            # Mesh TensorFlow attention initialization to avoid scaling before softmax
+            # See https://github.com/tensorflow/mesh/blob/fa19d69eafc9a482aff0b59ddd96b025c0cb207d/mesh_tensorflow/transformer/attention.py#L136
+            d_model = self.config.d_model
+            d_kv = self.config.d_kv
+            n_heads = self.config.num_heads
+            module.q.weight.data.normal_(mean=0.0, std=factor*((d_model * d_kv) ** -0.5))
+            module.k.weight.data.normal_(mean=0.0, std=factor*(d_model ** -0.5))
+            module.v.weight.data.normal_(mean=0.0, std=factor*(d_model ** -0.5))
+            module.o.weight.data.normal_(mean=0.0, std=factor*((n_heads * d_kv) ** -0.5))
+            if module.has_relative_attention_bias:
+                module.relative_attention_bias.weight.data.normal_(mean=0.0, std=factor*((d_model) ** -0.5))
 
 
 class T5Stack(T5PreTrainedModel):
@@ -440,8 +477,8 @@ class T5Stack(T5PreTrainedModel):
         self.output_hidden_states = config.output_hidden_states
         self.is_decoder = config.is_decoder
 
-        self.blocks = nn.ModuleList([T5Block(config, has_relative_attention_bias=bool(i == 0))
-                                     for i in range(config.num_layers)])
+        self.block = nn.ModuleList([T5Block(config, has_relative_attention_bias=bool(i == 0))
+                                    for i in range(config.num_layers)])
         self.final_layer_norm = nn.LayerNorm(config.d_model, eps=config.layer_norm_epsilon)
         self.dropout = nn.Dropout(config.dropout_rate)
 
@@ -518,7 +555,7 @@ class T5Stack(T5PreTrainedModel):
         all_attentions = ()
         position_bias = None
         encoder_decoder_position_bias = None
-        for i, layer_module in enumerate(self.blocks):
+        for i, layer_module in enumerate(self.block):
             if self.output_hidden_states:
                 all_hidden_states = all_hidden_states + (hidden_states,)
 
@@ -724,9 +761,10 @@ class T5WithLMHeadModel(T5PreTrainedModel):
     """
     def __init__(self, config):
         super(T5WithLMHeadModel, self).__init__(config)
+        self.model_dim = config.d_model
 
         self.transformer = T5Model(config)
-        self.lm_head = nn.Linear(config.d_model, config.vocab_size)
+        self.lm_head = nn.Linear(config.d_model, config.vocab_size, bias=False)
 
         self.init_weights()
 
@@ -738,15 +776,18 @@ class T5WithLMHeadModel(T5PreTrainedModel):
         outputs = self.transformer(**kwargs)
 
         sequence_output = outputs[0]
+        # Rescale output before projecting on vocab
+        # See https://github.com/tensorflow/mesh/blob/fa19d69eafc9a482aff0b59ddd96b025c0cb207d/mesh_tensorflow/transformer/transformer.py#L586
+        sequence_output = sequence_output * (self.model_dim ** -0.5)
         lm_logits = self.lm_head(sequence_output)
 
-        outputs = (lm_logits,) + outputs[2:]  # Add hidden states and attention if they are here
+        outputs = (lm_logits,) + outputs[1:]  # Add hidden states and attention if they are here
         if lm_labels is not None:
             shift_logits = lm_logits[..., :-1, :].contiguous()
             shift_labels = lm_labels[..., 1:].contiguous()
             loss_fct = CrossEntropyLoss(ignore_index=-1)
             loss = loss_fct(shift_logits.view(-1, shift_logits.size(-1)),
                             shift_labels.view(-1))
-            outputs = (loss,) + outputs
+            outputs = (loss,) + outputs  # TODO(thom): Add z_loss https://github.com/tensorflow/mesh/blob/fa19d69eafc9a482aff0b59ddd96b025c0cb207d/mesh_tensorflow/layers.py#L666
 
         return outputs  # (lm_loss), lm_logits, (hidden_states), (attentions)
diff --git a/transformers/tests/modeling_common_test.py b/transformers/tests/modeling_common_test.py
index 42bf9ac3f5..ee75da605c 100644
--- a/transformers/tests/modeling_common_test.py
+++ b/transformers/tests/modeling_common_test.py
@@ -59,7 +59,7 @@ else:
 def _config_zero_init(config):
     configs_no_init = copy.deepcopy(config)
     for key in configs_no_init.__dict__.keys():
-        if '_range' in key or '_std' in key:
+        if '_range' in key or '_std' in key or 'initializer_factor' in key:
             setattr(configs_no_init, key, 0.0)
     return configs_no_init
 
@@ -83,20 +83,24 @@ class CommonTestCases:
                 model.eval()
                 with torch.no_grad():
                     outputs = model(**inputs_dict)
+                out_2 = outputs[0].numpy()
+                out_2[np.isnan(out_2)] = 0
 
                 with TemporaryDirectory() as tmpdirname:
                     model.save_pretrained(tmpdirname)
                     model = model_class.from_pretrained(tmpdirname)
-                    with torch.no_grad():
-                        after_outputs = model(**inputs_dict)
 
-                    # Make sure we don't have nans
-                    out_1 = after_outputs[0].numpy()
-                    out_2 = outputs[0].numpy()
-                    out_1 = out_1[~np.isnan(out_1)]
-                    out_2 = out_2[~np.isnan(out_2)]
-                    max_diff = np.amax(np.abs(out_1 - out_2))
-                    self.assertLessEqual(max_diff, 1e-5)
+                with torch.no_grad():
+                    after_outputs = model(**inputs_dict)
+
+                # # Make sure we don't have nans
+                out_1 = after_outputs[0].numpy()
+                out_1[np.isnan(out_1)] = 0
+
+                out_1 = out_1 - out_2
+                amax = np.amax(out_1)
+                amin = np.amin(out_1)
+                self.assertLessEqual(max(amax, -amin), 1e-5)
 
         def test_initialization(self):
             config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
@@ -127,27 +131,28 @@ class CommonTestCases:
                 model = model_class(config)
                 model.eval()
                 outputs = model(**inputs_dict)
-                self_attentions = outputs[-1]
+                attentions = outputs[-1]
                 self.assertEqual(model.config.output_attentions, True)
                 self.assertEqual(model.config.output_hidden_states, False)
-                self.assertEqual(len(self_attentions), self.model_tester.num_hidden_layers)
+                self.assertEqual(len(attentions), self.model_tester.num_hidden_layers)
                 self.assertListEqual(
-                    list(self_attentions[0].shape[-3:]),
+                    list(attentions[0].shape[-3:]),
                     [self.model_tester.num_attention_heads,
                     self.model_tester.seq_length,
                     self.model_tester.key_len if hasattr(self.model_tester, 'key_len') else self.model_tester.seq_length])
                 out_len = len(outputs)
 
                 if self.is_encoder_decoder:
-                    cross_attentions = outputs[-2]
+                    self.assertEqual(out_len % 2, 0)
+                    decoder_attentions = outputs[(out_len // 2)-1]
                     self.assertEqual(model.config.output_attentions, True)
                     self.assertEqual(model.config.output_hidden_states, False)
-                    self.assertEqual(len(cross_attentions), self.model_tester.num_hidden_layers)
+                    self.assertEqual(len(decoder_attentions), self.model_tester.num_hidden_layers)
                     self.assertListEqual(
-                        list(cross_attentions[0].shape[-3:]),
+                        list(decoder_attentions[0].shape[-3:]),
                         [self.model_tester.num_attention_heads,
-                        self.model_tester.seq_length,
-                        self.model_tester.key_len if hasattr(self.model_tester, 'key_len') else self.model_tester.seq_length])
+                         self.model_tester.seq_length,
+                         self.model_tester.key_len if hasattr(self.model_tester, 'key_len') else self.model_tester.seq_length])
 
                 # Check attention is always last and order is fine
                 config.output_attentions = True
diff --git a/transformers/tests/modeling_t5_test.py b/transformers/tests/modeling_t5_test.py
index b8bb828ebd..2c67b83c25 100644
--- a/transformers/tests/modeling_t5_test.py
+++ b/transformers/tests/modeling_t5_test.py
@@ -57,7 +57,7 @@ class T5ModelTest(CommonTestCases.CommonModelTester):
                      d_ff=37,
                      relative_attention_num_buckets=8,
                      dropout_rate=0.1,
-                     initializer_range=0.02,
+                     initializer_factor=0.002,
                      scope=None,
                     ):
             self.parent = parent
@@ -74,7 +74,7 @@ class T5ModelTest(CommonTestCases.CommonModelTester):
             self.d_ff = d_ff
             self.relative_attention_num_buckets = relative_attention_num_buckets
             self.dropout_rate = dropout_rate
-            self.initializer_range = initializer_range
+            self.initializer_factor = initializer_factor
             self.scope = scope
 
         def prepare_config_and_inputs(self):
@@ -93,11 +93,12 @@ class T5ModelTest(CommonTestCases.CommonModelTester):
                 n_positions=self.n_positions,
                 d_model=self.hidden_size,
                 d_ff=self.d_ff,
+                d_kv=self.hidden_size // self.num_attention_heads,
                 num_layers=self.num_hidden_layers,
                 num_heads=self.num_attention_heads,
                 relative_attention_num_buckets=self.relative_attention_num_buckets,
                 dropout_rate=self.dropout_rate,
-                initializer_range=self.initializer_range)
+                initializer_factor=self.initializer_factor)
 
             return (config, input_ids, input_mask, token_labels)
 
@@ -130,8 +131,9 @@ class T5ModelTest(CommonTestCases.CommonModelTester):
         def create_and_check_t5_with_lm_head(self, config, input_ids, input_mask, token_labels):
             model = T5WithLMHeadModel(config=config)
             model.eval()
-            loss, prediction_scores = model(encoder_input_ids=input_ids, decoder_input_ids=input_ids,
-                                            decoder_attention_mask=input_mask, decoder_lm_labels=token_labels)
+            outputs = model(encoder_input_ids=input_ids, decoder_input_ids=input_ids,
+                            decoder_attention_mask=input_mask, decoder_lm_labels=token_labels)
+            loss, prediction_scores = outputs[0], outputs[1]
             result = {
                 "loss": loss,
                 "prediction_scores": prediction_scores,
diff --git a/transformers/tokenization_t5.py b/transformers/tokenization_t5.py
index cff6a41baf..ae898ba0d3 100644
--- a/transformers/tokenization_t5.py
+++ b/transformers/tokenization_t5.py
@@ -18,6 +18,7 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 
 import logging
 import os
+from shutil import copyfile
 
 from .tokenization_utils import PreTrainedTokenizer
 

From 8fda532c3cbab9e31fbbfa860f232b69e0f80633 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Thu, 7 Nov 2019 17:09:50 +0100
Subject: [PATCH 019/505] fix python 2 sentencepiece tokenization

---
 transformers/tests/tokenization_t5_test.py |  7 +++---
 transformers/tokenization_t5.py            | 26 ++++++++++++++++++----
 2 files changed, 26 insertions(+), 7 deletions(-)

diff --git a/transformers/tests/tokenization_t5_test.py b/transformers/tests/tokenization_t5_test.py
index 9362487d8d..aabb21e443 100644
--- a/transformers/tests/tokenization_t5_test.py
+++ b/transformers/tests/tokenization_t5_test.py
@@ -18,7 +18,8 @@ import os
 import unittest
 import pytest
 
-from transformers.tokenization_t5 import (T5Tokenizer, SPIECE_UNDERLINE)
+from transformers.tokenization_t5 import (T5Tokenizer)
+from transformers.tokenization_xlnet import SPIECE_UNDERLINE
 
 from .tokenization_tests_commons import CommonTestCases
 
@@ -33,7 +34,7 @@ class T5TokenizationTest(CommonTestCases.CommonTokenizerTester):
         super(T5TokenizationTest, self).setUp()
 
         # We have a SentencePiece fixture for testing
-        tokenizer = T5Tokenizer(SAMPLE_VOCAB, keep_accents=True)
+        tokenizer = T5Tokenizer(SAMPLE_VOCAB)
         tokenizer.save_pretrained(self.tmpdirname)
 
     def get_tokenizer(self, **kwargs):
@@ -45,7 +46,7 @@ class T5TokenizationTest(CommonTestCases.CommonTokenizerTester):
         return input_text, output_text
 
     def test_full_tokenizer(self):
-        tokenizer = T5Tokenizer(SAMPLE_VOCAB, keep_accents=True)
+        tokenizer = T5Tokenizer(SAMPLE_VOCAB)
 
         tokens = tokenizer.tokenize(u'This is a test')
         self.assertListEqual(tokens, [u'▁This', u'▁is', u'▁a', u'▁t', u'est'])
diff --git a/transformers/tokenization_t5.py b/transformers/tokenization_t5.py
index ae898ba0d3..93842d29f0 100644
--- a/transformers/tokenization_t5.py
+++ b/transformers/tokenization_t5.py
@@ -18,6 +18,7 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 
 import logging
 import os
+import six
 from shutil import copyfile
 
 from .tokenization_utils import PreTrainedTokenizer
@@ -96,18 +97,35 @@ class T5Tokenizer(PreTrainedTokenizer):
         self.sp_model = spm.SentencePieceProcessor()
         self.sp_model.Load(self.vocab_file)
 
-    def _tokenize(self, text):
+    def _tokenize(self, text, return_unicode=True, sample=False):
         """ Take as input a string and return a list of strings (tokens) for words/sub-words
         """
-        return self.sp_model.EncodeAsPieces(text)
+        if not sample:
+            pieces = self.sp_model.EncodeAsPieces(text)
+        else:
+            pieces = self.sp_model.SampleEncodeAsPieces(text, 64, 0.1)
+
+        # convert back to unicode for py2
+        if six.PY2 and return_unicode:
+            ret_pieces = []
+            for piece in pieces:
+                if isinstance(piece, str):
+                    piece = piece.decode('utf-8')
+                ret_pieces.append(piece)
+            pieces = ret_pieces
+
+        return pieces
 
     def _convert_token_to_id(self, token):
         """ Converts a token (str/unicode) in an id using the vocab. """
         return self.sp_model.piece_to_id(token)
 
-    def _convert_id_to_token(self, index):
+    def _convert_id_to_token(self, index, return_unicode=True):
         """Converts an index (integer) in a token (string/unicode) using the vocab."""
-        return self.sp_model.id_to_piece(index)
+        token = self.sp_model.IdToPiece(index)
+        if six.PY2 and return_unicode and isinstance(token, str):
+            token = token.decode('utf-8')
+        return token
 
     def convert_tokens_to_string(self, tokens):
         """ Converts a sequence of tokens (string) in a single string. """

From 28d0ba35d73d5b8b31fdadd72686a3ac078a6143 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Fri, 8 Nov 2019 11:22:19 +0100
Subject: [PATCH 020/505] only init encoder_attention_mask if stack is decoder

We currently initialize `encoder_attention_mask` when it is `None`,
whether the stack is that of an encoder or a decoder. Since this
may lead to bugs that are difficult to tracks down, I added a condition
that assesses whether the current stack is a decoder.
---
 transformers/modeling_bert.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/modeling_bert.py b/transformers/modeling_bert.py
index 7c2c6f4602..6bd5ab6a2e 100644
--- a/transformers/modeling_bert.py
+++ b/transformers/modeling_bert.py
@@ -656,7 +656,7 @@ class BertModel(BertPreTrainedModel):
 
         if attention_mask is None:
             attention_mask = torch.ones(input_shape, device=device)
-        if encoder_attention_mask is None:
+        if self.config.is_decoder and encoder_attention_mask is None:
             encoder_attention_mask = torch.ones(input_shape, device=device)
         if token_type_ids is None:
             token_type_ids = torch.zeros(input_shape, dtype=torch.long, device=device)

From cd286c2145221f3d1372aef103d0bc3ed03879da Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Fri, 8 Nov 2019 11:31:16 +0100
Subject: [PATCH 021/505] add condition around mask transformation

---
 transformers/modeling_bert.py | 20 ++++++++++++--------
 1 file changed, 12 insertions(+), 8 deletions(-)

diff --git a/transformers/modeling_bert.py b/transformers/modeling_bert.py
index 6bd5ab6a2e..893ec51015 100644
--- a/transformers/modeling_bert.py
+++ b/transformers/modeling_bert.py
@@ -656,8 +656,6 @@ class BertModel(BertPreTrainedModel):
 
         if attention_mask is None:
             attention_mask = torch.ones(input_shape, device=device)
-        if self.config.is_decoder and encoder_attention_mask is None:
-            encoder_attention_mask = torch.ones(input_shape, device=device)
         if token_type_ids is None:
             token_type_ids = torch.zeros(input_shape, dtype=torch.long, device=device)
 
@@ -688,13 +686,19 @@ class BertModel(BertPreTrainedModel):
 
         # If a 2D ou 3D attention mask is provided for the cross-attention
         # we need to make broadcastabe to [batch_size, num_heads, seq_length, seq_length]
-        if encoder_attention_mask.dim() == 3:
-            encoder_extended_attention_mask = encoder_attention_mask[:, None, :, :]
-        if encoder_attention_mask.dim() == 2:
-            encoder_extended_attention_mask = encoder_attention_mask[:, None, None, :]
+        if self.config.is_decoder:
+            if encoder_attention_mask is None:
+                encoder_attention_mask = torch.ones(input_shape, device=device)
 
-        encoder_extended_attention_mask = encoder_extended_attention_mask.to(dtype=next(self.parameters()).dtype)  # fp16 compatibility
-        encoder_extended_attention_mask = (1.0 - encoder_extended_attention_mask) * -10000.0
+            if encoder_attention_mask.dim() == 3:
+                encoder_extended_attention_mask = encoder_attention_mask[:, None, :, :]
+            if encoder_attention_mask.dim() == 2:
+                encoder_extended_attention_mask = encoder_attention_mask[:, None, None, :]
+
+            encoder_extended_attention_mask = encoder_extended_attention_mask.to(dtype=next(self.parameters()).dtype)  # fp16 compatibility
+            encoder_extended_attention_mask = (1.0 - encoder_extended_attention_mask) * -10000.0
+        else:
+            encoder_extended_attention_mask = None
 
         # Prepare head mask if needed
         # 1.0 in head_mask indicate we keep the head

From 727a79b305364522b6853679c5523efd9de7f772 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 8 Nov 2019 11:35:03 +0100
Subject: [PATCH 022/505] added TF2 model and tests - updated templates

---
 .../adding_a_new_model/modeling_tf_xxx.py     |   2 +
 templates/adding_a_new_model/modeling_xxx.py  |   2 +
 transformers/__init__.py                      |   3 +
 transformers/configuration_auto.py            |   6 +-
 transformers/configuration_t5.py              |   3 +-
 transformers/modeling_t5.py                   |  79 +-
 transformers/modeling_tf_pytorch_utils.py     |   4 +-
 transformers/modeling_tf_t5.py                | 783 +++++++++++-------
 transformers/modeling_utils.py                |   6 +-
 transformers/tests/modeling_tf_common_test.py |  23 +-
 transformers/tests/modeling_tf_t5_test.py     | 116 ++-
 11 files changed, 646 insertions(+), 381 deletions(-)

diff --git a/templates/adding_a_new_model/modeling_tf_xxx.py b/templates/adding_a_new_model/modeling_tf_xxx.py
index c661975768..b58817e453 100644
--- a/templates/adding_a_new_model/modeling_tf_xxx.py
+++ b/templates/adding_a_new_model/modeling_tf_xxx.py
@@ -26,6 +26,8 @@ import logging
 import math
 import os
 import sys
+import copy
+import itertools
 from io import open
 
 import numpy as np
diff --git a/templates/adding_a_new_model/modeling_xxx.py b/templates/adding_a_new_model/modeling_xxx.py
index ee705e753c..9c3505f0cf 100644
--- a/templates/adding_a_new_model/modeling_xxx.py
+++ b/templates/adding_a_new_model/modeling_xxx.py
@@ -25,6 +25,8 @@ import logging
 import math
 import os
 import sys
+import copy
+import itertools
 from io import open
 
 import torch
diff --git a/transformers/__init__.py b/transformers/__init__.py
index 601a068592..b882f4d968 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -158,6 +158,9 @@ if is_tf_available():
                                     TFCTRLLMHeadModel,
                                     TF_CTRL_PRETRAINED_MODEL_ARCHIVE_MAP)
 
+    from .modeling_tf_t5 import (TFT5PreTrainedModel, TFT5Model, TFT5WithLMHeadModel,
+                                 TF_T5_PRETRAINED_MODEL_ARCHIVE_MAP)
+
 # TF 2.0 <=> PyTorch conversion utilities
 from .modeling_tf_pytorch_utils import (convert_tf_weight_name_to_pt_weight_name,
                                         load_pytorch_checkpoint_in_tf2_model,
diff --git a/transformers/configuration_auto.py b/transformers/configuration_auto.py
index edd21a670c..3bee5b84a1 100644
--- a/transformers/configuration_auto.py
+++ b/transformers/configuration_auto.py
@@ -27,6 +27,7 @@ from .configuration_xlm import XLMConfig
 from .configuration_roberta import RobertaConfig
 from .configuration_distilbert import DistilBertConfig
 from .configuration_ctrl import CTRLConfig
+from .configuration_t5 import T5Config
 
 logger = logging.getLogger(__name__)
 
@@ -64,6 +65,7 @@ class AutoConfig(object):
 
         The configuration class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
+            - contains `t5`: T5Config (T5 model)
             - contains `distilbert`: DistilBertConfig (DistilBERT model)
             - contains `bert`: BertConfig (Bert model)
             - contains `openai-gpt`: OpenAIGPTConfig (OpenAI GPT model)
@@ -114,7 +116,9 @@ class AutoConfig(object):
             assert unused_kwargs == {'foo': False}
 
         """
-        if 'distilbert' in pretrained_model_name_or_path:
+        if 't5' in pretrained_model_name_or_path:
+            return T5Config.from_pretrained(pretrained_model_name_or_path, **kwargs)
+        elif 'distilbert' in pretrained_model_name_or_path:
             return DistilBertConfig.from_pretrained(pretrained_model_name_or_path, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return RobertaConfig.from_pretrained(pretrained_model_name_or_path, **kwargs)
diff --git a/transformers/configuration_t5.py b/transformers/configuration_t5.py
index 96e67758ac..83aab66fac 100644
--- a/transformers/configuration_t5.py
+++ b/transformers/configuration_t5.py
@@ -27,8 +27,7 @@ from .configuration_utils import PretrainedConfig
 logger = logging.getLogger(__name__)
 
 T5_PRETRAINED_CONFIG_ARCHIVE_MAP = {
-    't5-base-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-base-uncased-config.json",
-    't5-large-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-large-uncased-config.json",
+    't5-small': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-small-config.json",
 }
 
 
diff --git a/transformers/modeling_t5.py b/transformers/modeling_t5.py
index 6ed241761a..6be0ae6863 100644
--- a/transformers/modeling_t5.py
+++ b/transformers/modeling_t5.py
@@ -41,8 +41,7 @@ logger = logging.getLogger(__name__)
 # for the pretrained weights provided with the models
 ####################################################
 T5_PRETRAINED_MODEL_ARCHIVE_MAP = {
-    't5-base-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-base-uncased-pytorch_model.bin",
-    't5-large-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-large-uncased-pytorch_model.bin",
+    't5-small': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-small-pytorch_model.bin",
 }
 
 ####################################################
@@ -442,7 +441,7 @@ class T5PreTrainedModel(PreTrainedModel):
         if isinstance(module, nn.LayerNorm):
             module.bias.data.zero_()
             module.weight.data.fill_(factor*1.0)
-        elif isinstance(module, T5Model):
+        elif isinstance(module, (T5Model, T5WithLMHeadModel)):
             # Mesh TensorFlow embeddings initialization
             # See https://github.com/tensorflow/mesh/blob/fa19d69eafc9a482aff0b59ddd96b025c0cb207d/mesh_tensorflow/layers.py#L1624
             module.shared.weight.data.normal_(mean=0.0, std=factor*1.0)
@@ -502,11 +501,10 @@ class T5Stack(T5PreTrainedModel):
         # ourselves in which case we just need to make it broadcastable to all heads.
         if attention_mask.dim() == 3:
             extended_attention_mask = attention_mask[:, None, :, :]
-
+        elif attention_mask.dim() == 2:
         # Provided a padding mask of dimensions [batch_size, seq_length]
         # - if the model is a decoder, apply a causal mask in addition to the padding mask
         # - if the model is an encoder, make the mask broadcastable to [batch_size, num_heads, seq_length, seq_length]
-        if attention_mask.dim() == 2:
             if self.config.is_decoder:
                 seq_ids = torch.arange(seq_length, device=hidden_states.device)
                 causal_mask = seq_ids[None, None, :].repeat(batch_size, seq_length, 1) <= seq_ids[None, :, None]
@@ -593,7 +591,7 @@ class T5Stack(T5PreTrainedModel):
 T5_START_DOCSTRING = r"""    The T5 model was proposed in
     `Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer`_
     by Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, Peter J. Liu.
-    It's an encoder decoder pre-trained transformer.
+    It's an encoder decoder transformer pre-trained in a text-to-text denoising generative setting.
 
     This model is a PyTorch `torch.nn.Module`_ sub-class. Use it as a regular PyTorch Module and
     refer to the PyTorch documentation for all matter related to general usage and behavior.
@@ -634,16 +632,13 @@ T5_INPUTS_DOCSTRING = r"""
             Mask to avoid performing attention on padding token indices.
             Mask values selected in ``[0, 1]``:
             ``1`` for tokens that are NOT MASKED, ``0`` for MASKED tokens.
-        **position_ids**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
-            Indices of positions of each input sequence tokens in the position embeddings.
-            Selected in the range ``[0, config.max_position_embeddings - 1]``.
         **head_mask**: (`optional`) ``torch.FloatTensor`` of shape ``(num_heads,)`` or ``(num_layers, num_heads)``:
             Mask to nullify selected heads of the self-attention modules.
             Mask values selected in ``[0, 1]``:
             ``1`` indicates the head is **not masked**, ``0`` indicates the head is **masked**.
 """
 
-@add_start_docstrings("The bare single stack (encoder or decoder) of a T5 Model transformer outputting raw hidden-states"
+@add_start_docstrings("The bare T5 Model transformer outputting raw hidden-states"
                       "without any specific head on top.",
                       T5_START_DOCSTRING, T5_INPUTS_DOCSTRING)
 class T5Model(T5PreTrainedModel):
@@ -661,8 +656,8 @@ class T5Model(T5PreTrainedModel):
 
     Examples::
 
-        tokenizer = T5Tokenizer.from_pretrained('t5-base-uncased')
-        model = T5Model.from_pretrained('t5-base-uncased')
+        tokenizer = T5Tokenizer.from_pretrained('t5-small')
+        model = T5Model.from_pretrained('t5-small')
         input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
@@ -752,8 +747,8 @@ class T5WithLMHeadModel(T5PreTrainedModel):
 
     Examples::
 
-        tokenizer = T5Tokenizer.from_pretrained('t5-base-uncased')
-        model = T5WithLMHeadModel.from_pretrained('t5-base-uncased')
+        tokenizer = T5Tokenizer.from_pretrained('t5-small')
+        model = T5WithLMHeadModel.from_pretrained('t5-small')
         input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids, lm_labels=input_ids)
         loss, prediction_scores = outputs[:2]
@@ -763,31 +758,73 @@ class T5WithLMHeadModel(T5PreTrainedModel):
         super(T5WithLMHeadModel, self).__init__(config)
         self.model_dim = config.d_model
 
-        self.transformer = T5Model(config)
+        self.shared = nn.Embedding(config.vocab_size, config.d_model)
+
+        encoder_config = copy.deepcopy(config)
+        self.encoder = T5Stack(encoder_config)
+
+        decoder_config = copy.deepcopy(config)
+        decoder_config.is_decoder = True
+        self.decoder = T5Stack(decoder_config)
+
         self.lm_head = nn.Linear(config.d_model, config.vocab_size, bias=False)
 
         self.init_weights()
 
+    def get_input_embeddings(self):
+        return self.shared
+
+    def set_input_embeddings(self, new_embeddings):
+        self.shared = new_embeddings
+
     def get_output_embeddings(self):
         return self.lm_head
 
     def forward(self, **kwargs):
-        lm_labels = kwargs.pop('decoder_lm_labels', None)
-        outputs = self.transformer(**kwargs)
+        # keyword arguments come in 3 flavors: encoder-specific (prefixed by
+        # `encoder_`), decoder-specific (prefixed by `decoder_`) and those
+        # that apply to the model as whole.
+        # We let the specific kwargs override the common ones in case of conflict.
 
-        sequence_output = outputs[0]
+        lm_labels = kwargs.pop('decoder_lm_labels', None)
+
+        kwargs_common = dict((k, v) for k, v in kwargs.items()
+                             if not k.startswith("encoder_") and not k.startswith("decoder_"))
+        kwargs_encoder = kwargs_common.copy()
+        kwargs_decoder = kwargs_common.copy()
+        kwargs_encoder.update(dict((k[len("encoder_"):], v) for k, v in kwargs.items() if k.startswith("encoder_")))
+        kwargs_decoder.update(dict((k[len("decoder_"):], v) for k, v in kwargs.items() if k.startswith("decoder_")))
+
+        # Encode if needed (training, first prediction pass)
+        encoder_hidden_states = kwargs_encoder.pop("hidden_states", None)
+        if encoder_hidden_states is None:
+            encoder_inputs_ids = kwargs_encoder.pop("input_ids")
+            hidden_states = self.shared(encoder_inputs_ids)  # Convert inputs in embeddings
+            encoder_outputs = self.encoder(hidden_states, **kwargs_encoder)
+            encoder_hidden_states = encoder_outputs[0]
+        else:
+            encoder_outputs = ()
+
+        # Decode
+        decoder_inputs_ids = kwargs_decoder.pop("input_ids")
+        hidden_states = self.shared(decoder_inputs_ids)  # Convert inputs in embeddings
+        kwargs_decoder["encoder_hidden_states"] = encoder_hidden_states
+        kwargs_decoder["encoder_attention_mask"] = kwargs_encoder.get("attention_mask", None)
+        decoder_outputs = self.decoder(hidden_states, **kwargs_decoder)
+
+        sequence_output = decoder_outputs[0]
         # Rescale output before projecting on vocab
         # See https://github.com/tensorflow/mesh/blob/fa19d69eafc9a482aff0b59ddd96b025c0cb207d/mesh_tensorflow/transformer/transformer.py#L586
         sequence_output = sequence_output * (self.model_dim ** -0.5)
         lm_logits = self.lm_head(sequence_output)
 
-        outputs = (lm_logits,) + outputs[1:]  # Add hidden states and attention if they are here
+        decoder_outputs = (lm_logits,) + decoder_outputs[1:]  # Add hidden states and attention if they are here
         if lm_labels is not None:
             shift_logits = lm_logits[..., :-1, :].contiguous()
             shift_labels = lm_labels[..., 1:].contiguous()
             loss_fct = CrossEntropyLoss(ignore_index=-1)
             loss = loss_fct(shift_logits.view(-1, shift_logits.size(-1)),
                             shift_labels.view(-1))
-            outputs = (loss,) + outputs  # TODO(thom): Add z_loss https://github.com/tensorflow/mesh/blob/fa19d69eafc9a482aff0b59ddd96b025c0cb207d/mesh_tensorflow/layers.py#L666
+            decoder_outputs = (loss,) + decoder_outputs  # TODO(thom): Add z_loss https://github.com/tensorflow/mesh/blob/fa19d69eafc9a482aff0b59ddd96b025c0cb207d/mesh_tensorflow/layers.py#L666
 
-        return outputs  # (lm_loss), lm_logits, (hidden_states), (attentions)
+        return decoder_outputs + encoder_outputs
diff --git a/transformers/modeling_tf_pytorch_utils.py b/transformers/modeling_tf_pytorch_utils.py
index 88ce4d4610..6330c2748c 100644
--- a/transformers/modeling_tf_pytorch_utils.py
+++ b/transformers/modeling_tf_pytorch_utils.py
@@ -156,7 +156,7 @@ def load_pytorch_weights_in_tf2_model(tf_model, pt_state_dict, tf_inputs=None, a
             e.args += (symbolic_weight.shape, array.shape)
             raise e
 
-        logger.info("Initialize TF weight {}".format(symbolic_weight.name))
+        # logger.warning("Initialize TF weight {}".format(symbolic_weight.name))
 
         weight_value_tuples.append((symbolic_weight, array))
         all_pytorch_weights.discard(name)
@@ -269,7 +269,7 @@ def load_tf2_weights_in_pytorch_model(pt_model, tf_weights, allow_missing_keys=F
             e.args += (pt_weight.shape, array.shape)
             raise e
 
-        logger.info("Initialize PyTorch weight {}".format(pt_weight_name))
+        # logger.warning("Initialize PyTorch weight {}".format(pt_weight_name))
 
         new_pt_params_dict[pt_weight_name] = torch.from_numpy(array)
         loaded_pt_weights_data_ptr[pt_weight.data_ptr()] = torch.from_numpy(array)
diff --git a/transformers/modeling_tf_t5.py b/transformers/modeling_tf_t5.py
index deb453846c..c1de4745c2 100644
--- a/transformers/modeling_tf_t5.py
+++ b/transformers/modeling_tf_t5.py
@@ -22,24 +22,21 @@ import logging
 import math
 import os
 import sys
+import copy
+import itertools
 from io import open
 
 import numpy as np
 import tensorflow as tf
 
 from .configuration_t5 import T5Config
-from .modeling_tf_utils import TFPreTrainedModel, get_initializer
+from .modeling_tf_utils import TFPreTrainedModel, TFSharedEmbeddings, shape_list, get_initializer, DUMMY_INPUTS
 from .file_utils import add_start_docstrings
 
 logger = logging.getLogger(__name__)
 
-####################################################
-# This dict contrains shortcut names and associated url
-# for the pretrained weights provided with the models
-####################################################
 TF_T5_PRETRAINED_MODEL_ARCHIVE_MAP = {
-    't5-base-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-base-uncased-tf_model.h5",
-    't5-large-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-large-uncased-tf_model.h5",
+    't5-small': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-small-tf_model.h5",
 }
 
 ####################################################
@@ -48,33 +45,294 @@ TF_T5_PRETRAINED_MODEL_ARCHIVE_MAP = {
 # - TFPreTrainedModel for the models (it-self a sub-class of tf.keras.Model)
 ####################################################
 
-####################################################
-# Here is an example of typical layer in a TF 2.0 model of the library
-# The classes are usually identical to the PyTorch ones and prefixed with 'TF'.
-#
-# Note that class __init__ parameters includes **kwargs (send to 'super').
-# This let us have a control on class scope and variable names:
-# More precisely, we set the names of the class attributes (lower level layers) to
-# to the equivalent attributes names in the PyTorch model so we can have equivalent
-# class and scope structure between PyTorch and TF 2.0 models and easily load one in the other.
-#
-# See the conversion methods in modeling_tf_pytorch_utils.py for more details
-####################################################
-class TFT5Layer(tf.keras.layers.Layer):
+class TFT5DenseReluDense(tf.keras.layers.Layer):
     def __init__(self, config, **kwargs):
-        super(TFT5Layer, self).__init__(**kwargs)
-        self.attention = TFT5Attention(config, name='attention')
-        self.intermediate = TFT5Intermediate(config, name='intermediate')
-        self.transformer_output = TFT5Output(config, name='output')
+        super(TFT5DenseReluDense, self).__init__(**kwargs)
+        self.wi = tf.keras.layers.Dense(config.d_ff, use_bias=False, name='wi')
+        self.wo = tf.keras.layers.Dense(config.d_model, use_bias=False, name='wo')
+        self.dropout = tf.keras.layers.Dropout(config.dropout_rate)
+        self.act = tf.keras.activations.relu
 
-    def call(self, inputs, training=False):
-        hidden_states, attention_mask, head_mask = inputs
+    def call(self, hidden_states, training=False):
+        h = self.wi(hidden_states)
+        h = self.act(h)
+        h = self.dropout(h, training=training)
+        h = self.wo(h)
+        return h
 
-        attention_outputs = self.attention([hidden_states, attention_mask, head_mask], training=training)
-        attention_output = attention_outputs[0]
-        intermediate_output = self.intermediate(attention_output)
-        layer_output = self.transformer_output([intermediate_output, attention_output], training=training)
-        outputs = (layer_output,) + attention_outputs[1:]  # add attentions if we output them
+
+class TFT5LayerFF(tf.keras.layers.Layer):
+    def __init__(self, config, **kwargs):
+        super(TFT5LayerFF, self).__init__(**kwargs)
+        self.DenseReluDense = TFT5DenseReluDense(config, name='DenseReluDense')
+        self.layer_norm = tf.keras.layers.LayerNormalization(epsilon=config.layer_norm_epsilon,
+                                                             name='layer_norm')
+        self.dropout = tf.keras.layers.Dropout(config.dropout_rate)
+
+    def call(self, hidden_states, training=False):
+        norm_x = self.layer_norm(hidden_states)
+        y = self.DenseReluDense(norm_x, training=training)
+        layer_output = hidden_states + self.dropout(y, training=training)
+        return layer_output
+
+
+class TFT5Attention(tf.keras.layers.Layer):
+    NEW_ID = itertools.count()
+
+    def __init__(self, config, has_relative_attention_bias=False, **kwargs):
+        super(TFT5Attention, self).__init__(**kwargs)
+        self.layer_id = next(TFT5Attention.NEW_ID)
+        self.is_decoder = config.is_decoder
+        self.has_relative_attention_bias = has_relative_attention_bias
+
+        self.output_attentions = config.output_attentions
+        self.relative_attention_num_buckets = config.relative_attention_num_buckets
+        self.dim = config.d_model
+        self.d_kv = config.d_kv
+        self.n_heads = config.num_heads
+        assert self.dim % self.n_heads == 0
+        assert self.dim // self.n_heads == self.d_kv
+
+        # Mesh TensorFlow initialization to avoid scaling before softmax
+        self.q = tf.keras.layers.Dense(self.dim, use_bias=False, name='q')
+        self.k = tf.keras.layers.Dense(self.dim, use_bias=False, name='k')
+        self.v = tf.keras.layers.Dense(self.dim, use_bias=False, name='v')
+        self.o = tf.keras.layers.Dense(self.dim, use_bias=False, name='o')
+        self.dropout = tf.keras.layers.Dropout(config.dropout_rate)
+
+        if self.has_relative_attention_bias:
+            self.relative_attention_bias = tf.keras.layers.Embedding(self.relative_attention_num_buckets,
+                                                                     self.n_heads,
+                                                                     name='relative_attention_bias')
+        self.pruned_heads = set()
+
+    def prune_heads(self, heads):
+        raise NotImplementedError
+
+    @staticmethod
+    def _relative_position_bucket(relative_position,
+                                  bidirectional=True,
+                                  num_buckets=32,
+                                  max_distance=128):
+        """
+        Adapted from Mesh Tensorflow:
+        https://github.com/tensorflow/mesh/blob/0cb87fe07da627bf0b7e60475d59f95ed6b5be3d/mesh_tensorflow/transformer/transformer_layers.py#L593
+
+        Translate relative position to a bucket number for relative attention.
+        The relative position is defined as memory_position - query_position, i.e.
+        the distance in tokens from the attending position to the attended-to
+        position.  If bidirectional=False, then positive relative positions are
+        invalid.
+        We use smaller buckets for small absolute relative_position and larger buckets
+        for larger absolute relative_positions.  All relative positions >=max_distance
+        map to the same bucket.  All relative positions <=-max_distance map to the
+        same bucket.  This should allow for more graceful generalization to longer
+        sequences than the model has been trained on.
+        Args:
+            relative_position: an int32 Tensor
+            bidirectional: a boolean - whether the attention is bidirectional
+            num_buckets: an integer
+            max_distance: an integer
+        Returns:
+            a Tensor with the same shape as relative_position, containing int32
+            values in the range [0, num_buckets)
+        """
+        ret = 0
+        n = -relative_position
+        if bidirectional:
+            num_buckets //= 2
+            ret += tf.dtypes.cast(tf.math.less(n, 0), tf.int32) * num_buckets
+            n = tf.math.abs(n)
+        else:
+            n = tf.math.maximum(n, 0)
+        # now n is in the range [0, inf)
+        max_exact = num_buckets // 2
+        is_small = tf.math.less(n, max_exact)
+        val_if_large = max_exact + tf.dtypes.cast(
+            tf.math.log(tf.dtypes.cast(n, tf.float32) / max_exact)
+            / math.log(max_distance / max_exact) * (num_buckets - max_exact), tf.int32)
+        val_if_large = tf.math.minimum(val_if_large, num_buckets - 1)
+        ret += tf.where(is_small, n, val_if_large)
+        return ret
+
+    def compute_bias(self, qlen, klen):
+        """ Compute binned relative position bias """
+        context_position = tf.range(qlen)[:, None]
+        memory_position = tf.range(klen)[None, :]
+        relative_position = memory_position - context_position  # shape (qlen, klen)
+        rp_bucket = self._relative_position_bucket(relative_position,
+                                                   bidirectional=not self.is_decoder,
+                                                   num_buckets=self.relative_attention_num_buckets)
+        values = self.relative_attention_bias(rp_bucket)  # shape (qlen, klen, num_heads)
+        values = tf.expand_dims(tf.transpose(values, [2, 0, 1]), axis=0) # shape (1, num_heads, qlen, klen)
+        return values
+
+    def call(self, input, mask=None, kv=None, position_bias=None, cache=None, head_mask=None, training=False):
+        """
+        Self-attention (if kv is None) or attention over source sentence (provided by kv).
+        """
+        # Input is (bs, qlen, dim)
+        # Mask is (bs, klen) (non-causal) or (bs, klen, klen)
+        bs, qlen, dim = shape_list(input)
+        if kv is None:
+            klen = qlen if cache is None else cache['slen'] + qlen
+        else:
+            klen = shape_list(kv)[1]
+        # assert dim == self.dim, 'Dimensions do not match: %s input vs %s configured' % (dim, self.dim)
+        n_heads = self.n_heads
+        dim_per_head = self.dim // n_heads
+
+        def shape(x):
+            """  projection """
+            return tf.transpose(tf.reshape(x, (bs, -1, self.n_heads, dim_per_head)), perm=(0, 2, 1, 3))
+
+        def unshape(x):
+            """  compute context """
+            return tf.reshape(tf.transpose(x, perm=(0, 2, 1, 3)), (bs, -1, self.n_heads * dim_per_head))
+
+        q = shape(self.q(input))                                          # (bs, n_heads, qlen, dim_per_head)
+        if kv is None:
+            k = shape(self.k(input))                                      # (bs, n_heads, qlen, dim_per_head)
+            v = shape(self.v(input))                                      # (bs, n_heads, qlen, dim_per_head)
+        elif cache is None or self.layer_id not in cache:
+            k = v = kv
+            k = shape(self.k(k))                                          # (bs, n_heads, qlen, dim_per_head)
+            v = shape(self.v(v))                                          # (bs, n_heads, qlen, dim_per_head)
+
+        if cache is not None:
+            if self.layer_id in cache:
+                if kv is None:
+                    k_, v_ = cache[self.layer_id]
+                    k = tf.concat([k_, k], axis=2)                             # (bs, n_heads, klen, dim_per_head)
+                    v = tf.concat([v_, v], axis=2)                             # (bs, n_heads, klen, dim_per_head)
+                else:
+                    k, v = cache[self.layer_id]
+            cache[self.layer_id] = (k, v)
+
+        # q = q / math.sqrt(dim_per_head)                                     # No scaling in T5
+        scores = tf.matmul(q, k, transpose_b=True)                            # (bs, n_heads, qlen, klen)
+
+        if position_bias is None:
+            if not self.has_relative_attention_bias:
+                raise ValueError("No position_bias provided and no weights to compute position_bias")
+            position_bias = self.compute_bias(qlen, klen)
+        scores += position_bias
+
+        if mask is not None:
+            scores += mask
+            # mask = (mask == 0).expand_as(scores)                              # (bs, n_heads, qlen, klen)
+            # scores.masked_fill_(mask, -float('inf'))                          # (bs, n_heads, qlen, klen)
+
+        weights = tf.nn.softmax(scores, axis=-1)                              # (bs, n_heads, qlen, klen)
+        weights = self.dropout(weights, training=training)                    # (bs, n_heads, qlen, klen)
+
+        # Mask heads if we want to
+        if head_mask is not None:
+            weights = weights * head_mask
+
+        context = tf.matmul(weights, v)                                    # (bs, n_heads, qlen, dim_per_head)
+        context = unshape(context)                                            # (bs, qlen, dim)
+
+        context = self.o(context)
+
+        outputs = (context,)
+        if self.output_attentions:
+            outputs = outputs + (weights,)
+        if self.has_relative_attention_bias:
+            outputs = outputs + (position_bias,)
+        return outputs
+
+
+class TFT5LayerSelfAttention(tf.keras.layers.Layer):
+    def __init__(self, config, has_relative_attention_bias=False, **kwargs):
+        super(TFT5LayerSelfAttention, self).__init__(**kwargs)
+        self.SelfAttention = TFT5Attention(config,
+                                           has_relative_attention_bias=has_relative_attention_bias,
+                                           name='SelfAttention')
+        self.layer_norm = tf.keras.layers.LayerNormalization(epsilon=config.layer_norm_epsilon,
+                                                             name='layer_norm')
+        self.dropout = tf.keras.layers.Dropout(config.dropout_rate)
+
+    def call(self, hidden_states, attention_mask=None, position_bias=None,
+             head_mask=None, training=False):
+        norm_x = self.layer_norm(hidden_states)
+        attention_output = self.SelfAttention(norm_x,
+                                              mask=attention_mask,
+                                              position_bias=position_bias,
+                                              head_mask=head_mask,
+                                              training=training)
+        y = attention_output[0]
+        layer_output = hidden_states + self.dropout(y, training=training)
+        outputs = (layer_output,) + attention_output[1:]  # add attentions if we output them
+        return outputs
+
+
+class TFT5LayerCrossAttention(tf.keras.layers.Layer):
+    def __init__(self, config, has_relative_attention_bias=False, **kwargs):
+        super(TFT5LayerCrossAttention, self).__init__(**kwargs)
+        self.EncDecAttention = TFT5Attention(config,
+                                           has_relative_attention_bias=has_relative_attention_bias,
+                                           name='EncDecAttention')
+        self.layer_norm = tf.keras.layers.LayerNormalization(epsilon=config.layer_norm_epsilon,
+                                                             name='layer_norm')
+        self.dropout = tf.keras.layers.Dropout(config.dropout_rate)
+
+    def call(self, hidden_states, kv, attention_mask=None, position_bias=None,
+             head_mask=None, training=False):
+        norm_x = self.layer_norm(hidden_states)
+        attention_output = self.EncDecAttention(norm_x,
+                                                mask=attention_mask,
+                                                kv=kv,
+                                                position_bias=position_bias,
+                                                head_mask=head_mask,
+                                                training=training)
+        y = attention_output[0]
+        layer_output = hidden_states + self.dropout(y, training=training)
+        outputs = (layer_output,) + attention_output[1:]  # add attentions if we output them
+        return outputs
+
+
+class TFT5Block(tf.keras.layers.Layer):
+    def __init__(self, config, has_relative_attention_bias=False, **kwargs):
+        super(TFT5Block, self).__init__(**kwargs)
+        self.is_decoder = config.is_decoder
+        self.layer = []
+        self.layer.append(TFT5LayerSelfAttention(config,
+                                                 has_relative_attention_bias=has_relative_attention_bias,
+                                                 name='layer_._0'))
+        if self.is_decoder:
+            self.layer.append(TFT5LayerCrossAttention(config,
+                                                      has_relative_attention_bias=has_relative_attention_bias,
+                                                      name='layer_._1'))
+            self.layer.append(TFT5LayerFF(config, name='layer_._2'))
+        else:
+            self.layer.append(TFT5LayerFF(config, name='layer_._1'))
+
+    def call(self, hidden_states, attention_mask=None, position_bias=None,
+             encoder_hidden_states=None, encoder_attention_mask=None, encoder_decoder_position_bias=None,
+             head_mask=None, training=False):
+        self_attention_outputs = self.layer[0](hidden_states,
+                                                attention_mask=attention_mask,
+                                                position_bias=position_bias,
+                                                head_mask=head_mask,
+                                                training=training)
+        hidden_states = self_attention_outputs[0]
+        outputs = self_attention_outputs[1:]
+
+        if not self.is_decoder:
+            hidden_states = self.layer[1](hidden_states, training=training)
+        else:
+            cross_attention_outputs = self.layer[1](hidden_states,
+                                                    kv=encoder_hidden_states,
+                                                    attention_mask=encoder_attention_mask,
+                                                    position_bias=encoder_decoder_position_bias,
+                                                    head_mask=head_mask,
+                                                    training=training)
+            hidden_states = cross_attention_outputs[0]
+            outputs = cross_attention_outputs[1:] + outputs
+            hidden_states = self.layer[2](hidden_states, training=training)
+
+        outputs = (hidden_states,) + outputs  # add attentions if we output them
         return outputs
 
 
@@ -85,6 +343,19 @@ class TFT5Layer(tf.keras.layers.Layer):
 class TFT5MainLayer(tf.keras.layers.Layer):
     def __init__(self, config, **kwargs):
         super(TFT5MainLayer, self).__init__(**kwargs)
+        self.output_attentions = config.output_attentions
+        self.output_hidden_states = config.output_hidden_states
+        self.is_decoder = config.is_decoder
+        self.config = config
+        self.num_hidden_layers = config.num_layers
+
+        self.block = [TFT5Block(config,
+                                has_relative_attention_bias=bool(i == 0),
+                                name='block_._{}'.format(i))
+                        for i in range(config.num_layers)]
+        self.final_layer_norm = tf.keras.layers.LayerNormalization(epsilon=config.layer_norm_epsilon,
+                                                                   name='final_layer_norm')
+        self.dropout = tf.keras.layers.Dropout(config.dropout_rate)
 
     def _resize_token_embeddings(self, new_num_tokens):
         raise NotImplementedError  # Not implemented yet in the library fr TF 2.0 models
@@ -92,51 +363,56 @@ class TFT5MainLayer(tf.keras.layers.Layer):
     def _prune_heads(self, heads_to_prune):
         raise NotImplementedError  # Not implemented yet in the library fr TF 2.0 models
 
-    def call(self, inputs, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, training=False):
-        # We allow three types of multi-inputs:
-        # - traditional keyword arguments in the call method
-        # - all the arguments provided as a dict in the first positional argument of call
-        # - all the arguments provided as a list/tuple (ordered) in the first positional argument of call
-        # The last two options are useful to use the tf.keras fit() method.
-
-        if isinstance(inputs, (tuple, list)):
-            input_ids = inputs[0]
-            attention_mask = inputs[1] if len(inputs) > 1 else attention_mask
-            token_type_ids = inputs[2] if len(inputs) > 2 else token_type_ids
-            position_ids = inputs[3] if len(inputs) > 3 else position_ids
-            head_mask = inputs[4] if len(inputs) > 4 else head_mask
-            assert len(inputs) <= 5, "Too many inputs."
-        elif isinstance(inputs, dict):
-            input_ids = inputs.get('input_ids')
-            attention_mask = inputs.get('attention_mask', attention_mask)
-            token_type_ids = inputs.get('token_type_ids', token_type_ids)
-            position_ids = inputs.get('position_ids', position_ids)
-            head_mask = inputs.get('head_mask', head_mask)
-            assert len(inputs) <= 5, "Too many inputs."
-        else:
-            input_ids = inputs
+    def call(self, hidden_states, attention_mask=None, encoder_hidden_states=None,
+             encoder_attention_mask=None, head_mask=None, training=False):
 
+        batch_size, seq_length = shape_list(hidden_states)[:2]
         if attention_mask is None:
-            attention_mask = tf.fill(tf.shape(input_ids), 1)
-        if token_type_ids is None:
-            token_type_ids = tf.fill(tf.shape(input_ids), 0)
+            attention_mask = tf.fill((batch_size, seq_length), 1)
+        if self.is_decoder and encoder_attention_mask is None:
+            encoder_seq_length = encoder_hidden_states.shape[1]
+            encoder_attention_mask = tf.fill((batch_size, encoder_seq_length), 1)
 
-        # We create a 3D attention mask from a 2D tensor mask.
-        # Sizes are [batch_size, 1, 1, to_seq_length]
-        # So we can broadcast to [batch_size, num_heads, from_seq_length, to_seq_length]
-        # this attention mask is more simple than the triangular masking of causal attention
-        # used in OpenAI GPT, we just need to prepare the broadcast dimension here.
-        extended_attention_mask = attention_mask[:, tf.newaxis, tf.newaxis, :]
+        # We can provide a self-attention mask of dimensions [batch_size, from_seq_length, to_seq_length]
+        # ourselves in which case we just need to make it broadcastable to all heads.
+        attention_mask = tf.cast(attention_mask, dtype=tf.float32)
+        num_dims_attention_mask = len(shape_list(attention_mask))
+        if num_dims_attention_mask == 3:
+            extended_attention_mask = attention_mask[:, None, :, :]
+        elif num_dims_attention_mask == 2:
+        # Provided a padding mask of dimensions [batch_size, seq_length]
+        # - if the model is a decoder, apply a causal mask in addition to the padding mask
+        # - if the model is an encoder, make the mask broadcastable to [batch_size, num_heads, seq_length, seq_length]
+            if self.config.is_decoder:
+                seq_ids = tf.range(seq_length)
+                causal_mask = tf.less_equal(tf.tile(seq_ids[None, None, :], (batch_size, seq_length, 1)),
+                                            seq_ids[None, :, None])
+                causal_mask = tf.cast(causal_mask, dtype=tf.float32)
+                extended_attention_mask = causal_mask[:, None, :, :] * attention_mask[:, None, None, :]
+            else:
+                extended_attention_mask = attention_mask[:, None, None, :]
 
         # Since attention_mask is 1.0 for positions we want to attend and 0.0 for
         # masked positions, this operation will create a tensor which is 0.0 for
         # positions we want to attend and -10000.0 for masked positions.
         # Since we are adding it to the raw scores before the softmax, this is
         # effectively the same as removing these entirely.
-
-        extended_attention_mask = tf.cast(extended_attention_mask, tf.float32)
         extended_attention_mask = (1.0 - extended_attention_mask) * -10000.0
 
+        if self.is_decoder:
+            # If a 2D ou 3D attention mask is provided for the cross-attention
+            # we need to make broadcastabe to [batch_size, num_heads, seq_length, seq_length]
+            encoder_attention_mask = tf.cast(encoder_attention_mask, dtype=tf.float32)
+            num_dims_encoder_attention_mask = len(shape_list(encoder_attention_mask))
+            if num_dims_encoder_attention_mask == 3:
+                encoder_extended_attention_mask = encoder_attention_mask[:, None, :, :]
+            if num_dims_encoder_attention_mask == 2:
+                encoder_extended_attention_mask = encoder_attention_mask[:, None, None, :]
+
+            encoder_extended_attention_mask = (1.0 - encoder_extended_attention_mask) * -10000.0
+        else:
+            encoder_extended_attention_mask = None
+
         # Prepare head mask if needed
         # 1.0 in head_mask indicate we keep the head
         # attention_probs has shape bsz x n_heads x N x N
@@ -148,14 +424,44 @@ class TFT5MainLayer(tf.keras.layers.Layer):
             head_mask = [None] * self.num_hidden_layers
             # head_mask = tf.constant([0] * self.num_hidden_layers)
 
-        ##################################
-        # Replace this with your model code
-        embedding_output = self.embeddings(input_ids, position_ids=position_ids, token_type_ids=token_type_ids)
-        encoder_outputs = self.encoder([embedding_output, extended_attention_mask, head_mask], training=training)
-        sequence_output = encoder_outputs[0]
-        outputs = (sequence_output,) + encoder_outputs[1:]  # add hidden_states and attentions if they are here
+        all_hidden_states = ()
+        all_attentions = ()
+        position_bias = None
+        encoder_decoder_position_bias = None
+        for i, layer_module in enumerate(self.block):
+            if self.output_hidden_states:
+                all_hidden_states = all_hidden_states + (hidden_states,)
 
-        return outputs  # sequence_output, (hidden_states), (attentions)
+            layer_outputs = layer_module(hidden_states,
+                                         attention_mask=extended_attention_mask,
+                                         position_bias=position_bias,
+                                         encoder_hidden_states=encoder_hidden_states,
+                                         encoder_attention_mask=encoder_extended_attention_mask,
+                                         encoder_decoder_position_bias=encoder_decoder_position_bias,
+                                         head_mask=head_mask[i],
+                                         training=training)
+            hidden_states = layer_outputs[0]
+            if i == 0:
+                position_bias = layer_outputs[2 if self.output_attentions else 1]
+                if self.is_decoder:
+                    encoder_decoder_position_bias = layer_outputs[4 if self.output_attentions else 2]
+
+            if self.output_attentions:
+                all_attentions = all_attentions + (layer_outputs[1],)
+
+        hidden_states = self.final_layer_norm(hidden_states)
+        layer_output = self.dropout(hidden_states, training=training)
+
+        # Add last layer
+        if self.output_hidden_states:
+            all_hidden_states = all_hidden_states + (hidden_states,)
+
+        outputs = (hidden_states,)
+        if self.output_hidden_states:
+            outputs = outputs + (all_hidden_states,)
+        if self.output_attentions:
+            outputs = outputs + (all_attentions,)
+        return outputs  # last-layer hidden state, (all hidden states), (all attentions)
 
 
 ####################################################
@@ -173,18 +479,26 @@ class TFT5PreTrainedModel(TFPreTrainedModel):
     pretrained_model_archive_map = TF_T5_PRETRAINED_MODEL_ARCHIVE_MAP
     base_model_prefix = "transformer"
 
+    @property
+    def dummy_inputs(self):
+        input_ids = tf.constant([[7, 6, 0, 0, 1], [1, 2, 3, 0, 0], [0, 0, 0, 4, 5]])
+        input_mask = tf.constant([[1, 1, 0, 0, 1], [1, 1, 1, 0, 0], [1, 0, 0, 1, 1]])
+        dummy_inputs = {'decoder_input_ids': input_ids,
+                        'encoder_input_ids': input_ids,
+                        'decoder_attention_mask': input_mask}
+        return dummy_inputs
 
-T5_START_DOCSTRING = r"""    The XXX model was proposed in
-    `XXX: Pre-training of Deep Bidirectional Transformers for Language Understanding`_
-    by Jacob Devlin, Ming-Wei Chang, Kenton Lee and Kristina Toutanova. It's a bidirectional transformer
-    pre-trained using a combination of masked language modeling objective and next sentence prediction
-    on a large corpus comprising the Toronto Book Corpus and Wikipedia.
+
+T5_START_DOCSTRING = r"""    The T5 model was proposed in
+    `Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer`_
+    by Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, Peter J. Liu.
+    It's an encoder decoder transformer pre-trained in a text-to-text denoising generative setting.
 
     This model is a tf.keras.Model `tf.keras.Model`_ sub-class. Use it as a regular TF 2.0 Keras Model and
     refer to the TF 2.0 documentation for all matter related to general usage and behavior.
 
-    .. _`XXX: Pre-training of Deep Bidirectional Transformers for Language Understanding`:
-        https://arxiv.org/abs/1810.04805
+    .. _`Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer`:
+        https://arxiv.org/abs/1910.10683
 
     .. _`tf.keras.Model`:
         https://www.tensorflow.org/versions/r2.0/api_docs/python/tf/keras/Model
@@ -206,67 +520,50 @@ T5_START_DOCSTRING = r"""    The XXX model was proposed in
             `model({'input_ids': input_ids, 'token_type_ids': token_type_ids})`
 
     Parameters:
-        config (:class:`~transformers.XxxConfig`): Model configuration class with all the parameters of the model. 
+        config (:class:`~transformers.T5Config`): Model configuration class with all the parameters of the model. 
             Initializing with a config file does not load the weights associated with the model, only the configuration.
             Check out the :meth:`~transformers.PreTrainedModel.from_pretrained` method to load the model weights.
 """
 
-XXX_INPUTS_DOCSTRING = r"""
+T5_INPUTS_DOCSTRING = r"""
     Inputs:
         **input_ids**: ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, sequence_length)``:
             Indices of input sequence tokens in the vocabulary.
-            To match pre-training, XXX input sequence should be formatted with [CLS] and [SEP] tokens as follows:
+            To match pre-training, T5 input sequence should be formatted with [CLS] and [SEP] tokens as follows:
 
             (a) For sequence pairs:
 
                 ``tokens:         [CLS] is this jack ##son ##ville ? [SEP] no it is not . [SEP]``
-                
-                ``token_type_ids:   0   0  0    0    0     0       0   0   1  1  1  1   1   1``
 
             (b) For single sequences:
 
                 ``tokens:         [CLS] the dog is hairy . [SEP]``
-                
-                ``token_type_ids:   0   0   0   0  0     0   0``
 
-            Xxx is a model with absolute position embeddings so it's usually advised to pad the inputs on
-            the right rather than the left.
 
-            Indices can be obtained using :class:`transformers.XxxTokenizer`.
+            T5 is a model with relative position embeddings so you should be able to pad the inputs on
+            the right or the left.
+
+            Indices can be obtained using :class:`transformers.T5Tokenizer`.
             See :func:`transformers.PreTrainedTokenizer.encode` and
             :func:`transformers.PreTrainedTokenizer.convert_tokens_to_ids` for details.
         **attention_mask**: (`optional`) ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, sequence_length)``:
             Mask to avoid performing attention on padding token indices.
             Mask values selected in ``[0, 1]``:
             ``1`` for tokens that are NOT MASKED, ``0`` for MASKED tokens.
-        **token_type_ids**: (`optional`) ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, sequence_length)``:
-            Segment token indices to indicate first and second portions of the inputs.
-            Indices are selected in ``[0, 1]``: ``0`` corresponds to a `sentence A` token, ``1``
-            corresponds to a `sentence B` token
-            (see `XXX: Pre-training of Deep Bidirectional Transformers for Language Understanding`_ for more details).
-        **position_ids**: (`optional`) ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, sequence_length)``:
-            Indices of positions of each input sequence tokens in the position embeddings.
-            Selected in the range ``[0, config.max_position_embeddings - 1]``.
         **head_mask**: (`optional`) ``Numpy array`` or ``tf.Tensor`` of shape ``(num_heads,)`` or ``(num_layers, num_heads)``:
             Mask to nullify selected heads of the self-attention modules.
             Mask values selected in ``[0, 1]``:
             ``1`` indicates the head is **not masked**, ``0`` indicates the head is **masked**.
 """
 
-@add_start_docstrings("The bare Xxx Model transformer outputing raw hidden-states without any specific head on top.",
-                      XXX_START_DOCSTRING, XXX_INPUTS_DOCSTRING)
-class TFXxxModel(TFXxxPreTrainedModel):
+@add_start_docstrings("The bare T5 Model transformer outputting raw hidden-states"
+                      "without any specific head on top.",
+                      T5_START_DOCSTRING, T5_INPUTS_DOCSTRING)
+class TFT5Model(TFT5PreTrainedModel):
     r"""
     Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
         **last_hidden_state**: ``tf.Tensor`` of shape ``(batch_size, sequence_length, hidden_size)``
             Sequence of hidden-states at the output of the last layer of the model.
-        **pooler_output**: ``tf.Tensor`` of shape ``(batch_size, hidden_size)``
-            Last layer hidden-state of the first token of the sequence (classification token)
-            further processed by a Linear layer and a Tanh activation function. The Linear
-            layer weights are trained from the next sentence prediction (classification)
-            objective during Xxx pretraining. This output is usually *not* a good summary
-            of the semantic content of the input, you're often better with averaging or pooling
-            the sequence of hidden-states for the whole input sequence.
         **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
             list of ``tf.Tensor`` (one for the output of each layer + the output of the embeddings)
             of shape ``(batch_size, sequence_length, hidden_size)``:
@@ -278,27 +575,68 @@ class TFXxxModel(TFXxxPreTrainedModel):
     Examples::
 
         import tensorflow as tf
-        from transformers import XxxTokenizer, TFXxxModel
+        from transformers import T5Tokenizer, TFT5Model
 
-        tokenizer = XxxTokenizer.from_pretrained('xxx-base-uncased')
-        model = TFXxxModel.from_pretrained('xxx-base-uncased')
+        tokenizer = T5Tokenizer.from_pretrained('t5-small')
+        model = TFT5Model.from_pretrained('t5-small')
         input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
     """
     def __init__(self, config, *inputs, **kwargs):
-        super(TFXxxModel, self).__init__(config, *inputs, **kwargs)
-        self.transformer = TFXxxMainLayer(config, name='transformer')
+        super(TFT5Model, self).__init__(config, *inputs, **kwargs)
+        self.shared = TFSharedEmbeddings(config.vocab_size, config.d_model,
+                                         name='shared')
 
-    def call(self, inputs, **kwargs):
-        outputs = self.transformer(inputs, **kwargs)
-        return outputs
+        encoder_config = copy.deepcopy(config)
+        self.encoder = TFT5MainLayer(encoder_config, name='encoder')
+
+        decoder_config = copy.deepcopy(config)
+        decoder_config.is_decoder = True
+        self.decoder = TFT5MainLayer(decoder_config, name='decoder')
+
+    def call(self, decoder_input_ids, **kwargs):
+        # We allow two types of multi-inputs:
+        # - traditional keyword arguments in the call method
+        # - all the arguments provided as a dict in the first positional argument of call
+        # The last option is useful to use the tf.keras fit() method.
+
+        if isinstance(decoder_input_ids, dict):
+            kwargs.update(decoder_input_ids)
+        else:
+            kwargs['decoder_input_ids'] = decoder_input_ids
+
+        kwargs_common = dict((k, v) for k, v in kwargs.items()
+                             if not k.startswith("encoder_") and not k.startswith("decoder_"))
+        kwargs_encoder = kwargs_common.copy()
+        kwargs_decoder = kwargs_common.copy()
+        kwargs_encoder.update(dict((k[len("encoder_"):], v) for k, v in kwargs.items() if k.startswith("encoder_")))
+        kwargs_decoder.update(dict((k[len("decoder_"):], v) for k, v in kwargs.items() if k.startswith("decoder_")))
+
+        # Encode if needed (training, first prediction pass)
+        encoder_hidden_states = kwargs_encoder.pop("hidden_states", None)
+        if encoder_hidden_states is None:
+            encoder_inputs_ids = kwargs_encoder.pop("input_ids")
+            hidden_states = self.shared(encoder_inputs_ids)  # Convert inputs in embeddings
+            encoder_outputs = self.encoder(hidden_states, **kwargs_encoder)
+            encoder_hidden_states = encoder_outputs[0]
+        else:
+            encoder_outputs = ()
+
+        # Decode
+        decoder_inputs_ids = kwargs_decoder.pop("input_ids")
+        hidden_states = self.shared(decoder_inputs_ids)  # Convert inputs in embeddings
+        kwargs_decoder["encoder_hidden_states"] = encoder_hidden_states
+        kwargs_decoder["encoder_attention_mask"] = kwargs_encoder.get("attention_mask", None)
+        decoder_outputs = self.decoder(hidden_states, **kwargs_decoder)
+
+        return decoder_outputs + encoder_outputs
 
 
-@add_start_docstrings("""Xxx Model with a `language modeling` head on top. """,
-    XXX_START_DOCSTRING, XXX_INPUTS_DOCSTRING)
-class TFXxxForMaskedLM(TFXxxPreTrainedModel):
+@add_start_docstrings("""T5 Model with a `language modeling` head on top. """,
+    T5_START_DOCSTRING, T5_INPUTS_DOCSTRING)
+class TFT5WithLMHeadModel(TFT5PreTrainedModel):
     r"""
     Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
         **prediction_scores**: ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, sequence_length, config.vocab_size)``
@@ -314,183 +652,66 @@ class TFXxxForMaskedLM(TFXxxPreTrainedModel):
     Examples::
 
         import tensorflow as tf
-        from transformers import XxxTokenizer, TFXxxForMaskedLM
+        from transformers import T5Tokenizer, TFT5WithLMHeadModel
 
-        tokenizer = XxxTokenizer.from_pretrained('xxx-base-uncased')
-        model = TFXxxForMaskedLM.from_pretrained('xxx-base-uncased')
+        tokenizer = T5Tokenizer.from_pretrained('t5-small')
+        model = TFT5WithLMHeadModel.from_pretrained('t5-small')
         input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
         outputs = model(input_ids)
         prediction_scores = outputs[0]
 
     """
     def __init__(self, config, *inputs, **kwargs):
-        super(TFXxxForMaskedLM, self).__init__(config, *inputs, **kwargs)
+        super(TFT5WithLMHeadModel, self).__init__(config, *inputs, **kwargs)
+        self.model_dim = config.d_model
 
-        self.transformer = TFXxxMainLayer(config, name='transformer')
-        self.mlm = TFXxxMLMHead(config, self.transformer.embeddings, name='mlm')
+        self.shared = TFSharedEmbeddings(config.vocab_size, config.d_model,
+                                         name='shared')
 
-    def call(self, inputs, **kwargs):
-        outputs = self.transformer(inputs, **kwargs)
+        encoder_config = copy.deepcopy(config)
+        self.encoder = TFT5MainLayer(encoder_config, name='encoder')
 
-        sequence_output = outputs[0]
-        prediction_scores = self.mlm(sequence_output, training=kwargs.get('training', False))
+        decoder_config = copy.deepcopy(config)
+        decoder_config.is_decoder = True
+        self.decoder = TFT5MainLayer(decoder_config, name='decoder')
 
-        outputs = (prediction_scores,) + outputs[2:]  # Add hidden states and attention if they are here
+    def call(self, decoder_input_ids, **kwargs):
+        # We allow two types of multi-inputs:
+        # - traditional keyword arguments in the call method
+        # - all the arguments provided as a dict in the first positional argument of call
+        # The last option is useful to use the tf.keras fit() method.
 
-        return outputs  # prediction_scores, (hidden_states), (attentions)
+        if isinstance(decoder_input_ids, dict):
+            kwargs.update(decoder_input_ids)
+        else:
+            kwargs['decoder_input_ids'] = decoder_input_ids
 
+        kwargs_common = dict((k, v) for k, v in kwargs.items()
+                             if not k.startswith("encoder_") and not k.startswith("decoder_"))
+        kwargs_encoder = kwargs_common.copy()
+        kwargs_decoder = kwargs_common.copy()
+        kwargs_encoder.update(dict((k[len("encoder_"):], v) for k, v in kwargs.items() if k.startswith("encoder_")))
+        kwargs_decoder.update(dict((k[len("decoder_"):], v) for k, v in kwargs.items() if k.startswith("decoder_")))
 
-@add_start_docstrings("""Xxx Model transformer with a sequence classification/regression head on top (a linear layer on top of
-    the pooled output) e.g. for GLUE tasks. """,
-    XXX_START_DOCSTRING, XXX_INPUTS_DOCSTRING)
-class TFXxxForSequenceClassification(TFXxxPreTrainedModel):
-    r"""
-    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
-        **logits**: ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, config.num_labels)``
-            Classification (or regression if config.num_labels==1) scores (before SoftMax).
-        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
-            list of ``Numpy array`` or ``tf.Tensor`` (one for the output of each layer + the output of the embeddings)
-            of shape ``(batch_size, sequence_length, hidden_size)``:
-            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
-        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
-            list of ``Numpy array`` or ``tf.Tensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
-            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+        # Encode if needed (training, first prediction pass)
+        encoder_hidden_states = kwargs_encoder.pop("hidden_states", None)
+        if encoder_hidden_states is None:
+            encoder_inputs_ids = kwargs_encoder.pop("input_ids")
+            hidden_states = self.shared(encoder_inputs_ids)  # Convert inputs in embeddings
+            encoder_outputs = self.encoder(hidden_states, **kwargs_encoder)
+            encoder_hidden_states = encoder_outputs[0]
+        else:
+            encoder_outputs = ()
 
-    Examples::
+        # Decode
+        decoder_inputs_ids = kwargs_decoder.pop("input_ids")
+        hidden_states = self.shared(decoder_inputs_ids)  # Convert inputs in embeddings
+        kwargs_decoder["encoder_hidden_states"] = encoder_hidden_states
+        kwargs_decoder["encoder_attention_mask"] = kwargs_encoder.get("attention_mask", None)
+        decoder_outputs = self.decoder(hidden_states, **kwargs_decoder)
 
-        import tensorflow as tf
-        from transformers import XxxTokenizer, TFXxxForSequenceClassification
+        sequence_output = decoder_outputs[0] * (self.model_dim ** -0.5)
+        lm_logits = self.shared(sequence_output, mode="linear")
+        decoder_outputs = (lm_logits,) + decoder_outputs[1:]
 
-        tokenizer = XxxTokenizer.from_pretrained('xxx-base-uncased')
-        model = TFXxxForSequenceClassification.from_pretrained('xxx-base-uncased')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
-        outputs = model(input_ids)
-        logits = outputs[0]
-
-    """
-    def __init__(self, config, *inputs, **kwargs):
-        super(TFXxxForSequenceClassification, self).__init__(config, *inputs, **kwargs)
-        self.num_labels = config.num_labels
-
-        self.transformer = TFXxxMainLayer(config, name='transformer')
-        self.dropout = tf.keras.layers.Dropout(config.hidden_dropout_prob)
-        self.classifier = tf.keras.layers.Dense(config.num_labels,
-                                                kernel_initializer=get_initializer(config.initializer_range),
-                                                name='classifier')
-
-    def call(self, inputs, **kwargs):
-        outputs = self.transformer(inputs, **kwargs)
-
-        pooled_output = outputs[1]
-
-        pooled_output = self.dropout(pooled_output, training=kwargs.get('training', False))
-        logits = self.classifier(pooled_output)
-
-        outputs = (logits,) + outputs[2:]  # add hidden states and attention if they are here
-
-        return outputs  # logits, (hidden_states), (attentions)
-
-
-@add_start_docstrings("""Xxx Model with a token classification head on top (a linear layer on top of
-    the hidden-states output) e.g. for Named-Entity-Recognition (NER) tasks. """,
-    XXX_START_DOCSTRING, XXX_INPUTS_DOCSTRING)
-class TFXxxForTokenClassification(TFXxxPreTrainedModel):
-    r"""
-    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
-        **scores**: ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, sequence_length, config.num_labels)``
-            Classification scores (before SoftMax).
-        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
-            list of ``Numpy array`` or ``tf.Tensor`` (one for the output of each layer + the output of the embeddings)
-            of shape ``(batch_size, sequence_length, hidden_size)``:
-            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
-        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
-            list of ``Numpy array`` or ``tf.Tensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
-            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
-
-    Examples::
-
-        import tensorflow as tf
-        from transformers import XxxTokenizer, TFXxxForTokenClassification
-
-        tokenizer = XxxTokenizer.from_pretrained('xxx-base-uncased')
-        model = TFXxxForTokenClassification.from_pretrained('xxx-base-uncased')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
-        outputs = model(input_ids)
-        scores = outputs[0]
-
-    """
-    def __init__(self, config, *inputs, **kwargs):
-        super(TFXxxForTokenClassification, self).__init__(config, *inputs, **kwargs)
-        self.num_labels = config.num_labels
-
-        self.transformer = TFXxxMainLayer(config, name='transformer')
-        self.dropout = tf.keras.layers.Dropout(config.hidden_dropout_prob)
-        self.classifier = tf.keras.layers.Dense(config.num_labels,
-                                                kernel_initializer=get_initializer(config.initializer_range),
-                                                name='classifier')
-
-    def call(self, inputs, **kwargs):
-        outputs = self.transformer(inputs, **kwargs)
-
-        sequence_output = outputs[0]
-
-        sequence_output = self.dropout(sequence_output, training=kwargs.get('training', False))
-        logits = self.classifier(sequence_output)
-
-        outputs = (logits,) + outputs[2:]  # add hidden states and attention if they are here
-
-        return outputs  # scores, (hidden_states), (attentions)
-
-
-@add_start_docstrings("""Xxx Model with a span classification head on top for extractive question-answering tasks like SQuAD (a linear layers on top of
-    the hidden-states output to compute `span start logits` and `span end logits`). """,
-    XXX_START_DOCSTRING, XXX_INPUTS_DOCSTRING)
-class TFXxxForQuestionAnswering(TFXxxPreTrainedModel):
-    r"""
-    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
-        **start_scores**: ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, sequence_length,)``
-            Span-start scores (before SoftMax).
-        **end_scores**: ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, sequence_length,)``
-            Span-end scores (before SoftMax).
-        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
-            list of ``Numpy array`` or ``tf.Tensor`` (one for the output of each layer + the output of the embeddings)
-            of shape ``(batch_size, sequence_length, hidden_size)``:
-            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
-        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
-            list of ``Numpy array`` or ``tf.Tensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
-            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
-
-    Examples::
-
-        import tensorflow as tf
-        from transformers import XxxTokenizer, TFXxxForQuestionAnswering
-
-        tokenizer = XxxTokenizer.from_pretrained('xxx-base-uncased')
-        model = TFXxxForQuestionAnswering.from_pretrained('xxx-base-uncased')
-        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
-        outputs = model(input_ids)
-        start_scores, end_scores = outputs[:2]
-
-    """
-    def __init__(self, config, *inputs, **kwargs):
-        super(TFXxxForQuestionAnswering, self).__init__(config, *inputs, **kwargs)
-        self.num_labels = config.num_labels
-
-        self.transformer = TFXxxMainLayer(config, name='transformer')
-        self.qa_outputs = tf.keras.layers.Dense(config.num_labels,
-                                                kernel_initializer=get_initializer(config.initializer_range),
-                                                name='qa_outputs')
-
-    def call(self, inputs, **kwargs):
-        outputs = self.transformer(inputs, **kwargs)
-
-        sequence_output = outputs[0]
-
-        logits = self.qa_outputs(sequence_output)
-        start_logits, end_logits = tf.split(logits, 2, axis=-1)
-        start_logits = tf.squeeze(start_logits, axis=-1)
-        end_logits = tf.squeeze(end_logits, axis=-1)
-
-        outputs = (start_logits, end_logits,) + outputs[2:]
-
-        return outputs  # start_logits, end_logits, (hidden_states), (attentions)
+        return decoder_outputs + encoder_outputs
diff --git a/transformers/modeling_utils.py b/transformers/modeling_utils.py
index 063f52365d..5b1d3bb458 100644
--- a/transformers/modeling_utils.py
+++ b/transformers/modeling_utils.py
@@ -160,8 +160,7 @@ class PreTrainedModel(nn.Module):
         base_model.vocab_size = new_num_tokens
 
         # Tie weights again if needed
-        if hasattr(self, 'tie_weights'):
-            self.tie_weights()
+        self.tie_weights()
 
         return model_embeds
 
@@ -458,8 +457,7 @@ class PreTrainedModel(nn.Module):
                 raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
                                 model.__class__.__name__, "\n\t".join(error_msgs)))
 
-        if hasattr(model, 'tie_weights'):
-            model.tie_weights()  # make sure word embedding weights are still tied
+        model.tie_weights()  # make sure word embedding weights are still tied if needed
 
         # Set model in evaluation mode to desactivate DropOut modules by default
         model.eval()
diff --git a/transformers/tests/modeling_tf_common_test.py b/transformers/tests/modeling_tf_common_test.py
index f636c42889..6c3954a088 100644
--- a/transformers/tests/modeling_tf_common_test.py
+++ b/transformers/tests/modeling_tf_common_test.py
@@ -69,6 +69,7 @@ class TFCommonTestCases:
         test_torchscript = True
         test_pruning = True
         test_resize_embeddings = True
+        is_encoder_decoder = False
 
         def test_initialization(self):
             pass
@@ -156,7 +157,11 @@ class TFCommonTestCases:
         def test_compile_tf_model(self):
             config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
 
-            input_ids = tf.keras.Input(batch_shape=(2, 2000), name='input_ids', dtype='int32')
+            if self.is_encoder_decoder:
+                input_ids = {'decoder_input_ids': tf.keras.Input(batch_shape=(2, 2000), name='decoder_input_ids', dtype='int32'),
+                             'encoder_input_ids': tf.keras.Input(batch_shape=(2, 2000), name='encoder_input_ids', dtype='int32')}
+            else:
+                input_ids = tf.keras.Input(batch_shape=(2, 2000), name='input_ids', dtype='int32')
             optimizer = tf.keras.optimizers.Adam(learning_rate=3e-5, epsilon=1e-08, clipnorm=1.0)
             loss = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True)
             metric = tf.keras.metrics.SparseCategoricalAccuracy('accuracy')
@@ -189,7 +194,7 @@ class TFCommonTestCases:
                 outputs_dict = model(inputs_dict)
 
                 inputs_keywords = copy.deepcopy(inputs_dict)
-                input_ids = inputs_keywords.pop('input_ids')
+                input_ids = inputs_keywords.pop('input_ids', inputs_keywords.pop('decoder_input_ids'))
                 outputs_keywords = model(input_ids, **inputs_keywords)
 
                 output_dict = outputs_dict[0].numpy()
@@ -216,12 +221,24 @@ class TFCommonTestCases:
                     self.model_tester.key_len if hasattr(self.model_tester, 'key_len') else self.model_tester.seq_length])
                 out_len = len(outputs)
 
+                if self.is_encoder_decoder:
+                    self.assertEqual(out_len % 2, 0)
+                    decoder_attentions = outputs[(out_len // 2)-1]
+                    self.assertEqual(model.config.output_attentions, True)
+                    self.assertEqual(model.config.output_hidden_states, False)
+                    self.assertEqual(len(decoder_attentions), self.model_tester.num_hidden_layers)
+                    self.assertListEqual(
+                        list(decoder_attentions[0].shape[-3:]),
+                        [self.model_tester.num_attention_heads,
+                         self.model_tester.seq_length,
+                         self.model_tester.key_len if hasattr(self.model_tester, 'key_len') else self.model_tester.seq_length])
+
                 # Check attention is always last and order is fine
                 config.output_attentions = True
                 config.output_hidden_states = True
                 model = model_class(config)
                 outputs = model(inputs_dict)
-                self.assertEqual(out_len+1, len(outputs))
+                self.assertEqual(out_len + (2 if self.is_encoder_decoder else 1), len(outputs))
                 self.assertEqual(model.config.output_attentions, True)
                 self.assertEqual(model.config.output_hidden_states, True)
 
diff --git a/transformers/tests/modeling_tf_t5_test.py b/transformers/tests/modeling_tf_t5_test.py
index fac6763432..33f6f895f0 100644
--- a/transformers/tests/modeling_tf_t5_test.py
+++ b/transformers/tests/modeling_tf_t5_test.py
@@ -26,7 +26,7 @@ from .configuration_common_test import ConfigTester
 
 from transformers import T5Config, is_tf_available
 
-if False:  # is_tf_available():
+if is_tf_available():
     import tensorflow as tf
     from transformers.modeling_tf_t5 import (TFT5Model, TFT5WithLMHeadModel,TF_T5_PRETRAINED_MODEL_ARCHIVE_MAP)
 else:
@@ -35,7 +35,8 @@ else:
 
 class TFT5ModelTest(TFCommonTestCases.TFCommonModelTester):
 
-    all_model_classes = (TFT5Model, TFT5WithLMHeadModel) if False  else () # is_tf_available() else ()
+    is_encoder_decoder = True
+    all_model_classes = (TFT5Model, TFT5WithLMHeadModel) if is_tf_available() else ()
 
     class TFT5ModelTester(object):
 
@@ -45,22 +46,16 @@ class TFT5ModelTest(TFCommonTestCases.TFCommonModelTester):
                      seq_length=7,
                      is_training=True,
                      use_input_mask=True,
-                     use_token_type_ids=True,
                      use_labels=True,
                      vocab_size=99,
+                     n_positions=14,
                      hidden_size=32,
                      num_hidden_layers=5,
                      num_attention_heads=4,
-                     intermediate_size=37,
-                     hidden_act="gelu",
-                     hidden_dropout_prob=0.1,
-                     attention_probs_dropout_prob=0.1,
-                     max_position_embeddings=512,
-                     type_vocab_size=16,
-                     type_sequence_label_size=2,
-                     initializer_range=0.02,
-                     num_labels=3,
-                     num_choices=4,
+                     d_ff=37,
+                     relative_attention_num_buckets=8,
+                     dropout_rate=0.1,
+                     initializer_factor=0.002,
                      scope=None,
                     ):
             self.parent = parent
@@ -68,22 +63,16 @@ class TFT5ModelTest(TFCommonTestCases.TFCommonModelTester):
             self.seq_length = seq_length
             self.is_training = is_training
             self.use_input_mask = use_input_mask
-            self.use_token_type_ids = use_token_type_ids
             self.use_labels = use_labels
             self.vocab_size = vocab_size
+            self.n_positions = n_positions
             self.hidden_size = hidden_size
             self.num_hidden_layers = num_hidden_layers
             self.num_attention_heads = num_attention_heads
-            self.intermediate_size = intermediate_size
-            self.hidden_act = hidden_act
-            self.hidden_dropout_prob = hidden_dropout_prob
-            self.attention_probs_dropout_prob = attention_probs_dropout_prob
-            self.max_position_embeddings = max_position_embeddings
-            self.type_vocab_size = type_vocab_size
-            self.type_sequence_label_size = type_sequence_label_size
-            self.initializer_range = initializer_range
-            self.num_labels = num_labels
-            self.num_choices = num_choices
+            self.d_ff = d_ff
+            self.relative_attention_num_buckets = relative_attention_num_buckets
+            self.dropout_rate = dropout_rate
+            self.initializer_factor = initializer_factor
             self.scope = scope
 
         def prepare_config_and_inputs(self):
@@ -93,61 +82,53 @@ class TFT5ModelTest(TFCommonTestCases.TFCommonModelTester):
             if self.use_input_mask:
                 input_mask = ids_tensor([self.batch_size, self.seq_length], vocab_size=2)
 
-            token_type_ids = None
-            if self.use_token_type_ids:
-                token_type_ids = ids_tensor([self.batch_size, self.seq_length], self.type_vocab_size)
-
-            sequence_labels = None
             token_labels = None
-            choice_labels = None
             if self.use_labels:
-                sequence_labels = ids_tensor([self.batch_size], self.type_sequence_label_size)
-                token_labels = ids_tensor([self.batch_size, self.seq_length], self.num_labels)
-                choice_labels = ids_tensor([self.batch_size], self.num_choices)
+                token_labels = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)
 
             config = T5Config(
                 vocab_size_or_config_json_file=self.vocab_size,
-                hidden_size=self.hidden_size,
-                num_hidden_layers=self.num_hidden_layers,
-                num_attention_heads=self.num_attention_heads,
-                intermediate_size=self.intermediate_size,
-                hidden_act=self.hidden_act,
-                hidden_dropout_prob=self.hidden_dropout_prob,
-                attention_probs_dropout_prob=self.attention_probs_dropout_prob,
-                max_position_embeddings=self.max_position_embeddings,
-                type_vocab_size=self.type_vocab_size,
-                initializer_range=self.initializer_range)
+                n_positions=self.n_positions,
+                d_model=self.hidden_size,
+                d_ff=self.d_ff,
+                d_kv=self.hidden_size // self.num_attention_heads,
+                num_layers=self.num_hidden_layers,
+                num_heads=self.num_attention_heads,
+                relative_attention_num_buckets=self.relative_attention_num_buckets,
+                dropout_rate=self.dropout_rate,
+                initializer_factor=self.initializer_factor)
 
-            return config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
+            return (config, input_ids, input_mask, token_labels)
 
-        def create_and_check_t5_model(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
+        def create_and_check_t5_model(self, config, input_ids, input_mask, token_labels):
             model = TFT5Model(config=config)
-            inputs = {'input_ids': input_ids,
-                      'attention_mask': input_mask,
-                      'token_type_ids': token_type_ids}
-            sequence_output, pooled_output = model(inputs)
+            inputs = {'encoder_input_ids': input_ids,
+                      'decoder_input_ids': input_ids,
+                      'decoder_attention_mask': input_mask}
+            encoder_output, decoder_output = model(inputs)
 
-            inputs = [input_ids, input_mask]
-            sequence_output, pooled_output = model(inputs)
-
-            sequence_output, pooled_output = model(input_ids)
+            encoder_output, decoder_output = model(input_ids,
+                                                   decoder_attention_mask=input_mask,
+                                                   encoder_input_ids=input_ids)
 
             result = {
-                "sequence_output": sequence_output.numpy(),
-                "pooled_output": pooled_output.numpy(),
+                "encoder_output": encoder_output.numpy(),
+                "decoder_output": decoder_output.numpy(),
             }
             self.parent.assertListEqual(
-                list(result["sequence_output"].shape),
+                list(result["encoder_output"].shape),
+                [self.batch_size, self.seq_length, self.hidden_size])
+            self.parent.assertListEqual(
+                list(result["decoder_output"].shape),
                 [self.batch_size, self.seq_length, self.hidden_size])
-            self.parent.assertListEqual(list(result["pooled_output"].shape), [self.batch_size, self.hidden_size])
 
 
-        def create_and_check_t5_with_lm_head(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
+        def create_and_check_t5_with_lm_head(self, config, input_ids, input_mask, token_labels):
             model = TFT5WithLMHeadModel(config=config)
-            inputs = {'input_ids': input_ids,
-                      'attention_mask': input_mask,
-                      'token_type_ids': token_type_ids}
-            prediction_scores, = model(inputs)
+            inputs = {'encoder_input_ids': input_ids,
+                      'decoder_input_ids': input_ids,
+                      'decoder_attention_mask': input_mask}
+            prediction_scores, decoder_output = model(inputs)
             result = {
                 "prediction_scores": prediction_scores.numpy(),
             }
@@ -158,14 +139,15 @@ class TFT5ModelTest(TFCommonTestCases.TFCommonModelTester):
 
         def prepare_config_and_inputs_for_common(self):
             config_and_inputs = self.prepare_config_and_inputs()
-            (config, input_ids, token_type_ids, input_mask,
-             sequence_labels, token_labels, choice_labels) = config_and_inputs
-            inputs_dict = {'input_ids': input_ids, 'token_type_ids': token_type_ids, 'attention_mask': input_mask}
+            (config, input_ids, input_mask, token_labels) = config_and_inputs
+            inputs_dict = {'encoder_input_ids': input_ids,
+                           'decoder_input_ids': input_ids,
+                           'decoder_attention_mask': input_mask}
             return config, inputs_dict
 
     def setUp(self):
         self.model_tester = TFT5ModelTest.TFT5ModelTester(self)
-        self.config_tester = ConfigTester(self, config_class=T5Config, hidden_size=37)
+        self.config_tester = ConfigTester(self, config_class=T5Config, d_model=37)
 
     def test_config(self):
         self.config_tester.run_common_tests()
@@ -181,7 +163,7 @@ class TFT5ModelTest(TFCommonTestCases.TFCommonModelTester):
     @pytest.mark.slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
-        for model_name in ['t5-base']:
+        for model_name in ['t5-small']:
             model = TFT5Model.from_pretrained(model_name, cache_dir=cache_dir)
             shutil.rmtree(cache_dir)
             self.assertIsNotNone(model)

From 4321c541254bdabbda631520cff0a5a376ad9f48 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 8 Nov 2019 11:49:32 +0100
Subject: [PATCH 023/505] fix tests

---
 transformers/tests/modeling_tf_common_test.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/tests/modeling_tf_common_test.py b/transformers/tests/modeling_tf_common_test.py
index 6c3954a088..83a15c137a 100644
--- a/transformers/tests/modeling_tf_common_test.py
+++ b/transformers/tests/modeling_tf_common_test.py
@@ -194,7 +194,7 @@ class TFCommonTestCases:
                 outputs_dict = model(inputs_dict)
 
                 inputs_keywords = copy.deepcopy(inputs_dict)
-                input_ids = inputs_keywords.pop('input_ids', inputs_keywords.pop('decoder_input_ids'))
+                input_ids = inputs_keywords.pop('input_ids' if not self.is_encoder_decoder else 'decoder_input_ids', None)
                 outputs_keywords = model(input_ids, **inputs_keywords)
 
                 output_dict = outputs_dict[0].numpy()

From f03c0c1423d4635f3e71a6c24053f01f6f02063c Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 8 Nov 2019 11:49:46 +0100
Subject: [PATCH 024/505] adding models in readme and auto classes

---
 README.md                                     |  3 ++-
 docs/source/pretrained_models.rst             | 20 +++++++++++++++++++
 transformers/__main__.py                      | 18 +++++++++++++++++
 .../convert_pytorch_checkpoint_to_tf2.py      | 13 ++++++++----
 transformers/modeling_auto.py                 | 13 ++++++++++--
 transformers/modeling_tf_auto.py              | 13 ++++++++++--
 transformers/tokenization_auto.py             |  7 ++++++-
 7 files changed, 77 insertions(+), 10 deletions(-)

diff --git a/README.md b/README.md
index 40b08583b1..d6f6e426d8 100644
--- a/README.md
+++ b/README.md
@@ -122,7 +122,8 @@ At some point in the future, you'll be able to seamlessly move from pre-training
 7. **[RoBERTa](https://github.com/pytorch/fairseq/tree/master/examples/roberta)** (from Facebook), released together with the paper a [Robustly Optimized BERT Pretraining Approach](https://arxiv.org/abs/1907.11692) by Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, Veselin Stoyanov.
 8. **[DistilBERT](https://github.com/huggingface/transformers/tree/master/examples/distillation)** (from HuggingFace), released together with the paper [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://arxiv.org/abs/1910.01108) by Victor Sanh, Lysandre Debut and Thomas Wolf. The same method has been applied to compress GPT2 into [DistilGPT2](https://github.com/huggingface/transformers/tree/master/examples/distillation).
 9. **[CTRL](https://github.com/salesforce/ctrl/)** (from Salesforce) released with the paper [CTRL: A Conditional Transformer Language Model for Controllable Generation](https://arxiv.org/abs/1909.05858) by Nitish Shirish Keskar*, Bryan McCann*, Lav R. Varshney, Caiming Xiong and Richard Socher.
-10. Want to contribute a new model? We have added a **detailed guide and templates** to guide you in the process of adding a new model. You can find them in the [`templates`](./templates) folder of the repository. Be sure to check the [contributing guidelines](./CONTRIBUTING.md) and contact the maintainers or open an issue to collect feedbacks before starting your PR.
+10. **[T5](https://github.com/google-research/text-to-text-transfer-transformer)** (from Google AI) released with the paper [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://arxiv.org/abs/1910.10683) by Colin Raffel and Noam Shazeer and Adam Roberts and Katherine Lee and Sharan Narang and Michael Matena and Yanqi Zhou and Wei Li and Peter J. Liu.
+11. Want to contribute a new model? We have added a **detailed guide and templates** to guide you in the process of adding a new model. You can find them in the [`templates`](./templates) folder of the repository. Be sure to check the [contributing guidelines](./CONTRIBUTING.md) and contact the maintainers or open an issue to collect feedbacks before starting your PR.
 
 These implementations have been tested on several datasets (see the example scripts) and should match the performances of the original implementations (e.g. ~93 F1 on SQuAD for BERT Whole-Word-Masking, ~88 F1 on RocStories for OpenAI GPT, ~18.3 perplexity on WikiText 103 for Transformer-XL, ~0.916 Peason R coefficient on STS-B for XLNet). You can find more details on the performances in the Examples section of the [documentation](https://huggingface.co/transformers/examples.html).
 
diff --git a/docs/source/pretrained_models.rst b/docs/source/pretrained_models.rst
index 43c08228bd..c6240dc850 100644
--- a/docs/source/pretrained_models.rst
+++ b/docs/source/pretrained_models.rst
@@ -144,5 +144,25 @@ Here is the full list of the currently provided pretrained models together with
 | CTRL              | ``ctrl``                                                   | | 48-layer, 1280-hidden, 16-heads, 1.6B parameters                                                                                    |
 |                   |                                                            | | Salesforce's Large-sized CTRL English model                                                                                         |
 +-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+| T5                | ``t5-small``                                               | | 6-layer, 768-hidden, 12-heads, 66M parameters                                                                                       |
+|                   |                                                            | | The DistilBERT model distilled from the BERT model `bert-base-uncased` checkpoint                                                   |
+|                   |                                                            | (see `details <https://github.com/huggingface/transformers/tree/master/examples/distillation>`__)                                     |
+|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   | ``t5-base``                                                | | 6-layer, 768-hidden, 12-heads, 66M parameters                                                                                       |
+|                   |                                                            | | The DistilBERT model distilled from the BERT model `bert-base-uncased` checkpoint, with an additional linear layer.                 |
+|                   |                                                            | (see `details <https://github.com/huggingface/transformers/tree/master/examples/distillation>`__)                                     |
+|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   | ``t5-large``                                               | | 6-layer, 768-hidden, 12-heads, 82M parameters                                                                                       |
+|                   |                                                            | | The DistilGPT2 model distilled from the GPT2 model `gpt2` checkpoint.                                                               |
+|                   |                                                            | (see `details <https://github.com/huggingface/transformers/tree/master/examples/distillation>`__)                                     |
+|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   | ``t5-3b``                                                  | | 6-layer, 768-hidden, 12-heads, 82M parameters                                                                                       |
+|                   |                                                            | | The DistilRoBERTa model distilled from the RoBERTa model `roberta-base` checkpoint.                                                 |
+|                   |                                                            | (see `details <https://github.com/huggingface/transformers/tree/master/examples/distillation>`__)                                     |
+|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   | ``t5-11b``                                                 | | 6-layer, 768-hidden, 12-heads, 82M parameters                                                                                       |
+|                   |                                                            | | The DistilRoBERTa model distilled from the RoBERTa model `roberta-base` checkpoint.                                                 |
+|                   |                                                            | (see `details <https://github.com/huggingface/transformers/tree/master/examples/distillation>`__)                                     |
++-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 
 .. <https://huggingface.co/transformers/examples.html>`__
\ No newline at end of file
diff --git a/transformers/__main__.py b/transformers/__main__.py
index 31dbd24908..6136d768f6 100644
--- a/transformers/__main__.py
+++ b/transformers/__main__.py
@@ -6,6 +6,7 @@ def main():
         "This command line utility let you convert original (author released) model checkpoint to pytorch.\n"
         "It should be used as one of: \n"
         ">> transformers bert TF_CHECKPOINT TF_CONFIG PYTORCH_DUMP_OUTPUT, \n"
+        ">> transformers t5 TF_CHECKPOINT TF_CONFIG PYTORCH_DUMP_OUTPUT, \n"
         ">> transformers gpt OPENAI_GPT_CHECKPOINT_FOLDER_PATH PYTORCH_DUMP_OUTPUT [OPENAI_GPT_CONFIG], \n"
         ">> transformers transfo_xl TF_CHECKPOINT_OR_DATASET PYTORCH_DUMP_OUTPUT [TF_CONFIG] or \n"
         ">> transformers gpt2 TF_CHECKPOINT PYTORCH_DUMP_OUTPUT [GPT2_CONFIG] or \n"
@@ -21,6 +22,23 @@ def main():
                     "https://www.tensorflow.org/install/ for installation instructions.")
                 raise
 
+            if len(sys.argv) != 5:
+                # pylint: disable=line-too-long
+                print("Should be used as `transformers bert TF_CHECKPOINT TF_CONFIG PYTORCH_DUMP_OUTPUT`")
+            else:
+                PYTORCH_DUMP_OUTPUT = sys.argv.pop()
+                TF_CONFIG = sys.argv.pop()
+                TF_CHECKPOINT = sys.argv.pop()
+                convert_tf_checkpoint_to_pytorch(TF_CHECKPOINT, TF_CONFIG, PYTORCH_DUMP_OUTPUT)
+        elif sys.argv[1] == "t5":
+            try:
+                from .convert_t5_original_tf_checkpoint_to_pytorch import convert_tf_checkpoint_to_pytorch
+            except ImportError:
+                print("transformers can only be used from the commandline to convert TensorFlow models in PyTorch, "
+                    "In that case, it requires TensorFlow to be installed. Please see "
+                    "https://www.tensorflow.org/install/ for installation instructions.")
+                raise
+
             if len(sys.argv) != 5:
                 # pylint: disable=line-too-long
                 print("Should be used as `transformers bert TF_CHECKPOINT TF_CONFIG PYTORCH_DUMP_OUTPUT`")
diff --git a/transformers/convert_pytorch_checkpoint_to_tf2.py b/transformers/convert_pytorch_checkpoint_to_tf2.py
index e673b77dcc..19629172ff 100644
--- a/transformers/convert_pytorch_checkpoint_to_tf2.py
+++ b/transformers/convert_pytorch_checkpoint_to_tf2.py
@@ -33,7 +33,8 @@ from transformers import (load_pytorch_checkpoint_in_tf2_model,
                                   OpenAIGPTConfig, TFOpenAIGPTLMHeadModel, OPENAI_GPT_PRETRAINED_CONFIG_ARCHIVE_MAP,
                                   RobertaConfig, TFRobertaForMaskedLM, TFRobertaForSequenceClassification, ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP,
                                   DistilBertConfig, TFDistilBertForMaskedLM, TFDistilBertForQuestionAnswering, DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP,
-                                  CTRLConfig, TFCTRLLMHeadModel, CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP)
+                                  CTRLConfig, TFCTRLLMHeadModel, CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP,
+                                  T5Config, TFT5WithLMHeadModel, T5_PRETRAINED_CONFIG_ARCHIVE_MAP)
 
 if is_torch_available():
     import torch
@@ -46,7 +47,8 @@ if is_torch_available():
                                       OpenAIGPTLMHeadModel, OPENAI_GPT_PRETRAINED_MODEL_ARCHIVE_MAP,
                                       RobertaForMaskedLM, RobertaForSequenceClassification, ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP,
                                       DistilBertForMaskedLM, DistilBertForQuestionAnswering, DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP,
-                                      CTRLLMHeadModel, CTRL_PRETRAINED_MODEL_ARCHIVE_MAP)
+                                      CTRLLMHeadModel, CTRL_PRETRAINED_MODEL_ARCHIVE_MAP,
+                                      T5WithLMHeadModel, T5_PRETRAINED_MODEL_ARCHIVE_MAP)
 else:
     (BertForPreTraining, BertForQuestionAnswering, BertForSequenceClassification, BERT_PRETRAINED_MODEL_ARCHIVE_MAP,
     GPT2LMHeadModel, GPT2_PRETRAINED_MODEL_ARCHIVE_MAP,
@@ -56,7 +58,8 @@ else:
     OpenAIGPTLMHeadModel, OPENAI_GPT_PRETRAINED_MODEL_ARCHIVE_MAP,
     RobertaForMaskedLM, RobertaForSequenceClassification, ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP,
     DistilBertForMaskedLM, DistilBertForQuestionAnswering, DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP,
-    CTRLLMHeadModel, CTRL_PRETRAINED_MODEL_ARCHIVE_MAP) = (
+    CTRLLMHeadModel, CTRL_PRETRAINED_MODEL_ARCHIVE_MAP,
+    T5WithLMHeadModel, T5_PRETRAINED_MODEL_ARCHIVE_MAP) = (
         None, None, None, None,
         None, None,
         None, None,
@@ -65,6 +68,7 @@ else:
         None, None,
         None, None, None,
         None, None, None,
+        None, None,
         None, None)
 
 
@@ -85,7 +89,8 @@ MODEL_CLASSES = {
     'roberta-large-mnli': (RobertaConfig, TFRobertaForSequenceClassification, RobertaForSequenceClassification, ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP, ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP),
     'distilbert': (DistilBertConfig, TFDistilBertForMaskedLM, DistilBertForMaskedLM, DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP, DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP),
     'distilbert-base-uncased-distilled-squad': (DistilBertConfig, TFDistilBertForQuestionAnswering, DistilBertForQuestionAnswering, DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP, DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP),
-    'ctrl': (CTRLConfig, TFCTRLLMHeadModel, CTRLLMHeadModel, CTRL_PRETRAINED_MODEL_ARCHIVE_MAP, CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP)
+    'ctrl': (CTRLConfig, TFCTRLLMHeadModel, CTRLLMHeadModel, CTRL_PRETRAINED_MODEL_ARCHIVE_MAP, CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP),
+    't5': (T5Config, TFT5WithLMHeadModel, T5WithLMHeadModel, T5_PRETRAINED_MODEL_ARCHIVE_MAP, T5_PRETRAINED_CONFIG_ARCHIVE_MAP),
 }
 
 def convert_pt_checkpoint_to_tf(model_type, pytorch_checkpoint_path, config_file, tf_dump_path, compare_with_pt_model=False, use_cached_models=True):
diff --git a/transformers/modeling_auto.py b/transformers/modeling_auto.py
index d98110d4bd..a2129176d3 100644
--- a/transformers/modeling_auto.py
+++ b/transformers/modeling_auto.py
@@ -27,6 +27,7 @@ from .modeling_xlnet import XLNetModel, XLNetLMHeadModel, XLNetForSequenceClassi
 from .modeling_xlm import XLMModel, XLMWithLMHeadModel, XLMForSequenceClassification, XLMForQuestionAnswering
 from .modeling_roberta import RobertaModel, RobertaForMaskedLM, RobertaForSequenceClassification
 from .modeling_distilbert import DistilBertModel, DistilBertForQuestionAnswering, DistilBertForMaskedLM, DistilBertForSequenceClassification
+from .modeling_t5 import T5Model, T5WithLMHeadModel
 
 from .modeling_utils import PreTrainedModel, SequenceSummary
 
@@ -47,6 +48,7 @@ class AutoModel(object):
 
         The base model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
+            - contains `t5`: T5Model (T5 model)
             - contains `distilbert`: DistilBertModel (DistilBERT model)
             - contains `roberta`: RobertaModel (RoBERTa model)
             - contains `bert`: BertModel (Bert model)
@@ -70,6 +72,7 @@ class AutoModel(object):
 
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
+            - contains `t5`: T5Model (T5 model)
             - contains `distilbert`: DistilBertModel (DistilBERT model)
             - contains `roberta`: RobertaModel (RoBERTa model)
             - contains `bert`: BertModel (Bert model)
@@ -136,7 +139,9 @@ class AutoModel(object):
             model = AutoModel.from_pretrained('./tf_model/bert_tf_checkpoint.ckpt.index', from_tf=True, config=config)
 
         """
-        if 'distilbert' in pretrained_model_name_or_path:
+        if 't5' in pretrained_model_name_or_path:
+            return T5Model.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'distilbert' in pretrained_model_name_or_path:
             return DistilBertModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return RobertaModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
@@ -171,6 +176,7 @@ class AutoModelWithLMHead(object):
 
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
+            - contains `t5`: T5ModelWithLMHead (T5 model)
             - contains `distilbert`: DistilBertForMaskedLM (DistilBERT model)
             - contains `roberta`: RobertaForMaskedLM (RoBERTa model)
             - contains `bert`: BertForMaskedLM (Bert model)
@@ -197,6 +203,7 @@ class AutoModelWithLMHead(object):
 
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
+            - contains `t5`: T5ModelWithLMHead (T5 model)
             - contains `distilbert`: DistilBertForMaskedLM (DistilBERT model)
             - contains `roberta`: RobertaForMaskedLM (RoBERTa model)
             - contains `bert`: BertForMaskedLM (Bert model)
@@ -262,7 +269,9 @@ class AutoModelWithLMHead(object):
             model = AutoModelWithLMHead.from_pretrained('./tf_model/bert_tf_checkpoint.ckpt.index', from_tf=True, config=config)
 
         """
-        if 'distilbert' in pretrained_model_name_or_path:
+        if 't5' in pretrained_model_name_or_path:
+            return T5WithLMHeadModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'distilbert' in pretrained_model_name_or_path:
             return DistilBertForMaskedLM.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return RobertaForMaskedLM.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
diff --git a/transformers/modeling_tf_auto.py b/transformers/modeling_tf_auto.py
index df0ad6e401..b24623dcdc 100644
--- a/transformers/modeling_tf_auto.py
+++ b/transformers/modeling_tf_auto.py
@@ -27,6 +27,7 @@ from .modeling_tf_xlm import TFXLMModel, TFXLMWithLMHeadModel, TFXLMForSequenceC
 from .modeling_tf_roberta import TFRobertaModel, TFRobertaForMaskedLM, TFRobertaForSequenceClassification
 from .modeling_tf_distilbert import TFDistilBertModel, TFDistilBertForQuestionAnswering, TFDistilBertForMaskedLM, TFDistilBertForSequenceClassification
 from .modeling_tf_ctrl import TFCTRLModel, TFCTRLLMHeadModel
+from .modeling_tf_t5 import TFT5Model, TFT5WithLMHeadModel
 
 from .file_utils import add_start_docstrings
 
@@ -45,6 +46,7 @@ class TFAutoModel(object):
 
         The base model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
+            - contains `t5`: TFT5Model (T5 model)
             - contains `distilbert`: TFDistilBertModel (DistilBERT model)
             - contains `roberta`: TFRobertaModel (RoBERTa model)
             - contains `bert`: TFBertModel (Bert model)
@@ -68,6 +70,7 @@ class TFAutoModel(object):
 
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
+            - contains `t5`: TFT5Model (T5 model)
             - contains `distilbert`: TFDistilBertModel (DistilBERT model)
             - contains `roberta`: TFRobertaModel (RoBERTa model)
             - contains `bert`: TFTFBertModel (Bert model)
@@ -133,7 +136,9 @@ class TFAutoModel(object):
             model = TFAutoModel.from_pretrained('./pt_model/bert_pytorch_model.bin', from_pt=True, config=config)
 
         """
-        if 'distilbert' in pretrained_model_name_or_path:
+        if 't5' in pretrained_model_name_or_path:
+            return TFT5Model.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'distilbert' in pretrained_model_name_or_path:
             return TFDistilBertModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return TFRobertaModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
@@ -169,6 +174,7 @@ class TFAutoModelWithLMHead(object):
 
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
+            - contains `t5`: TFT5WithLMHeadModel (T5 model)
             - contains `distilbert`: TFDistilBertForMaskedLM (DistilBERT model)
             - contains `roberta`: TFRobertaForMaskedLM (RoBERTa model)
             - contains `bert`: TFBertForMaskedLM (Bert model)
@@ -195,6 +201,7 @@ class TFAutoModelWithLMHead(object):
 
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
+            - contains `t5`: TFT5WithLMHeadModel (T5 model)
             - contains `distilbert`: TFDistilBertForMaskedLM (DistilBERT model)
             - contains `roberta`: TFRobertaForMaskedLM (RoBERTa model)
             - contains `bert`: TFBertForMaskedLM (Bert model)
@@ -261,7 +268,9 @@ class TFAutoModelWithLMHead(object):
             model = TFAutoModelWithLMHead.from_pretrained('./pt_model/bert_pytorch_model.bin', from_pt=True, config=config)
 
         """
-        if 'distilbert' in pretrained_model_name_or_path:
+        if 't5' in pretrained_model_name_or_path:
+            return TFT5WithLMHeadModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'distilbert' in pretrained_model_name_or_path:
             return TFDistilBertForMaskedLM.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return TFRobertaForMaskedLM.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
diff --git a/transformers/tokenization_auto.py b/transformers/tokenization_auto.py
index ec056de17f..5be2562448 100644
--- a/transformers/tokenization_auto.py
+++ b/transformers/tokenization_auto.py
@@ -27,6 +27,7 @@ from .tokenization_xlnet import XLNetTokenizer
 from .tokenization_xlm import XLMTokenizer
 from .tokenization_roberta import RobertaTokenizer
 from .tokenization_distilbert import DistilBertTokenizer
+from .tokenization_t5 import T5Tokenizer
 
 logger = logging.getLogger(__name__)
 
@@ -41,6 +42,7 @@ class AutoTokenizer(object):
 
         The tokenizer class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
+            - contains `t5`: T5Tokenizer (T5 model)
             - contains `distilbert`: DistilBertTokenizer (DistilBert model)
             - contains `roberta`: RobertaTokenizer (RoBERTa model)
             - contains `bert`: BertTokenizer (Bert model)
@@ -64,6 +66,7 @@ class AutoTokenizer(object):
 
         The tokenizer class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
+            - contains `t5`: T5Tokenizer (T5 model)
             - contains `distilbert`: DistilBertTokenizer (DistilBert model)
             - contains `roberta`: RobertaTokenizer (XLM model)
             - contains `bert`: BertTokenizer (Bert model)
@@ -101,7 +104,9 @@ class AutoTokenizer(object):
             tokenizer = AutoTokenizer.from_pretrained('./test/bert_saved_model/')  # E.g. tokenizer was saved using `save_pretrained('./test/saved_model/')`
 
         """
-        if 'distilbert' in pretrained_model_name_or_path:
+        if 't5' in pretrained_model_name_or_path:
+            return T5Tokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
+        elif 'distilbert' in pretrained_model_name_or_path:
             return DistilBertTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return RobertaTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)

From 15e53c4e8712260b016225310c397e19a5f7b21c Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 8 Nov 2019 12:43:21 +0100
Subject: [PATCH 025/505] maybe fix tests

---
 transformers/tests/modeling_tf_common_test.py | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/transformers/tests/modeling_tf_common_test.py b/transformers/tests/modeling_tf_common_test.py
index 83a15c137a..20ccfd8ce0 100644
--- a/transformers/tests/modeling_tf_common_test.py
+++ b/transformers/tests/modeling_tf_common_test.py
@@ -131,7 +131,11 @@ class TFCommonTestCases:
                 with torch.no_grad():
                     pto = pt_model(**pt_inputs_dict)
                 tfo = tf_model(inputs_dict)
-                max_diff = np.amax(np.abs(tfo[0].numpy() - pto[0].numpy()))
+                tfo = tfo[0].numpy()
+                pto = pto[0].numpy()
+                tfo[np.isnan(tfo)] = 0
+                pto[np.isnan(pto)] = 0
+                max_diff = np.amax(np.abs(tfo - pto))
                 self.assertLessEqual(max_diff, 2e-2)
 
                 # Check we can load pt model in tf and vice-versa with checkpoint => model functions
@@ -151,7 +155,11 @@ class TFCommonTestCases:
                 with torch.no_grad():
                     pto = pt_model(**pt_inputs_dict)
                 tfo = tf_model(inputs_dict)
-                max_diff = np.amax(np.abs(tfo[0].numpy() - pto[0].numpy()))
+                tfo = tfo[0].numpy()
+                pto = pto[0].numpy()
+                tfo[np.isnan(tfo)] = 0
+                pto[np.isnan(pto)] = 0
+                max_diff = np.amax(np.abs(tfo - pto))
                 self.assertLessEqual(max_diff, 2e-2)
 
         def test_compile_tf_model(self):

From b4fcd59a5ae8d12102db106d3b03849ef86109bd Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 8 Nov 2019 14:38:53 +0100
Subject: [PATCH 026/505] add sentinels in tokenizer

---
 transformers/tokenization_t5.py | 31 ++++++++++++++++++++++++++-----
 1 file changed, 26 insertions(+), 5 deletions(-)

diff --git a/transformers/tokenization_t5.py b/transformers/tokenization_t5.py
index 93842d29f0..3847aeefbf 100644
--- a/transformers/tokenization_t5.py
+++ b/transformers/tokenization_t5.py
@@ -18,6 +18,7 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 
 import logging
 import os
+import re
 import six
 from shutil import copyfile
 
@@ -31,7 +32,7 @@ SPIECE_UNDERLINE = u'▁'
 # Mapping from the keyword arguments names of Tokenizer `__init__`
 # to file names for serializing Tokenizer instances
 ####################################################
-VOCAB_FILES_NAMES = {'vocab_file': 'vocab.txt'}
+VOCAB_FILES_NAMES = {'vocab_file': 'spiece.model'}
 
 ####################################################
 # Mapping from the keyword arguments names of Tokenizer `__init__`
@@ -56,15 +57,27 @@ class T5Tokenizer(PreTrainedTokenizer):
         SentencePiece based tokenizer. Peculiarities:
 
             - requires `SentencePiece <https://github.com/google/sentencepiece>`_
+            - `extra_ids` add a number of extra ids added to the end of the vocabulary for use as sentinels.
+                These tokens are accessible as `<extra_id_{%d}>` where `{%d}` is a number between 0 and extra_ids-1.
+                Extra tokens are indexed from the end of the vocabulary up to beginnning (<extra_id_0> is the last token in the vocabulary)
+                (like in T5 preprocessing
+                see: https://github.com/google-research/text-to-text-transfer-transformer/blob/9fd7b14a769417be33bc6c850f9598764913c833/t5/data/preprocessors.py#L2117)
     """
     vocab_files_names = VOCAB_FILES_NAMES
     pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
     max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
 
     def __init__(self, vocab_file, eos_token="</s>", unk_token="<unk>",
-                 pad_token="<pad>", **kwargs):
+                 pad_token="<pad>", extra_ids=100, additional_special_tokens=None, **kwargs):
+        # Add extra_ids to the special token list
+        if extra_ids > 0:
+            if additional_special_tokens is None:
+                additional_special_tokens = []
+            additional_special_tokens.extend([u"<extra_id_{}>".format(i) for i in range(extra_ids)])
+
         super(T5Tokenizer, self).__init__(eos_token=eos_token, unk_token=unk_token,
-                                          pad_token=pad_token, **kwargs)
+                                          pad_token=pad_token, additional_special_tokens=additional_special_tokens,
+                                          **kwargs)
 
         try:
             import sentencepiece as spm
@@ -74,13 +87,14 @@ class T5Tokenizer(PreTrainedTokenizer):
                            "pip install sentencepiece")
 
         self.vocab_file = vocab_file
+        self._extra_ids = extra_ids
 
         self.sp_model = spm.SentencePieceProcessor()
         self.sp_model.Load(vocab_file)
 
     @property
     def vocab_size(self):
-        return self.sp_model.get_piece_size()
+        return self.sp_model.get_piece_size() + self._extra_ids
 
     def __getstate__(self):
         state = self.__dict__.copy()
@@ -118,11 +132,18 @@ class T5Tokenizer(PreTrainedTokenizer):
 
     def _convert_token_to_id(self, token):
         """ Converts a token (str/unicode) in an id using the vocab. """
+        if token.startswith(u"<extra_id_"):
+            l = re.match(r'<extra_id_(\d+)>', token)
+            num = int(l[1])
+            return self.vocab_size - num - 1
         return self.sp_model.piece_to_id(token)
 
     def _convert_id_to_token(self, index, return_unicode=True):
         """Converts an index (integer) in a token (string/unicode) using the vocab."""
-        token = self.sp_model.IdToPiece(index)
+        if index < self.sp_model.get_piece_size():
+            token = self.sp_model.IdToPiece(index)
+        else:
+            token = u"<extra_id_{}>".format(self.vocab_size - 1 - index)
         if six.PY2 and return_unicode and isinstance(token, str):
             token = token.decode('utf-8')
         return token

From 268d4f2099f90bb62949988c3b78596242e1d753 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 8 Nov 2019 16:41:55 +0100
Subject: [PATCH 027/505] fix position biases + better tests

---
 transformers/modeling_t5.py            | 11 +++--
 transformers/tests/modeling_t5_test.py | 62 +++++++++++++++-----------
 2 files changed, 42 insertions(+), 31 deletions(-)

diff --git a/transformers/modeling_t5.py b/transformers/modeling_t5.py
index 6be0ae6863..2a74333d31 100644
--- a/transformers/modeling_t5.py
+++ b/transformers/modeling_t5.py
@@ -408,7 +408,7 @@ class T5Block(nn.Module):
                                                 position_bias=position_bias,
                                                 head_mask=head_mask)
         hidden_states = self_attention_outputs[0]
-        outputs = self_attention_outputs[1:]
+        outputs = self_attention_outputs[1:]  # Keep self-attention outputs and relative position weights
 
         if not self.is_decoder:
             hidden_states = self.layer[1](hidden_states)
@@ -419,11 +419,11 @@ class T5Block(nn.Module):
                                                     position_bias=encoder_decoder_position_bias,
                                                     head_mask=head_mask)
             hidden_states = cross_attention_outputs[0]
-            outputs = cross_attention_outputs[1:] + outputs
+            outputs = outputs + cross_attention_outputs[1:]  # Keep cross-attention outputs and relative position weights
             hidden_states = self.layer[2](hidden_states)
 
         outputs = (hidden_states,) + outputs  # add attentions if we output them
-        return outputs
+        return outputs  # hidden-states, (self-attention weights), (self-attention position bias), (cross-attention weights), (cross-attention position bias)
 
 
 class T5PreTrainedModel(PreTrainedModel):
@@ -564,14 +564,17 @@ class T5Stack(T5PreTrainedModel):
                                          encoder_attention_mask=encoder_extended_attention_mask,
                                          encoder_decoder_position_bias=encoder_decoder_position_bias,
                                          head_mask=head_mask[i])
+            # layer_outputs is a tuple with:
+            # hidden-states, (self-attention weights), (self-attention position bias), (cross-attention weights), (cross-attention position bias)
             hidden_states = layer_outputs[0]
             if i == 0:
+                # We share the position biases between the layers - the first layer store them
                 position_bias = layer_outputs[2 if self.output_attentions else 1]
                 if self.is_decoder:
                     encoder_decoder_position_bias = layer_outputs[4 if self.output_attentions else 2]
 
             if self.output_attentions:
-                all_attentions = all_attentions + (layer_outputs[1],)
+                all_attentions = all_attentions + (layer_outputs[1],)  # We keep only self-attention weights for now
 
         hidden_states = self.final_layer_norm(hidden_states)
         layer_output = self.dropout(hidden_states)
diff --git a/transformers/tests/modeling_t5_test.py b/transformers/tests/modeling_t5_test.py
index 2c67b83c25..091bd742b5 100644
--- a/transformers/tests/modeling_t5_test.py
+++ b/transformers/tests/modeling_t5_test.py
@@ -45,9 +45,10 @@ class T5ModelTest(CommonTestCases.CommonModelTester):
         def __init__(self,
                      parent,
                      batch_size=13,
-                     seq_length=7,
+                     encoder_seq_length=7,
+                     decoder_seq_length=9,
                      is_training=True,
-                     use_input_mask=True,
+                     use_attention_mask=True,
                      use_labels=True,
                      vocab_size=99,
                      n_positions=14,
@@ -62,9 +63,10 @@ class T5ModelTest(CommonTestCases.CommonModelTester):
                     ):
             self.parent = parent
             self.batch_size = batch_size
-            self.seq_length = seq_length
+            self.encoder_seq_length = encoder_seq_length
+            self.decoder_seq_length = decoder_seq_length
             self.is_training = is_training
-            self.use_input_mask = use_input_mask
+            self.use_attention_mask = use_attention_mask
             self.use_labels = use_labels
             self.vocab_size = vocab_size
             self.n_positions = n_positions
@@ -78,15 +80,18 @@ class T5ModelTest(CommonTestCases.CommonModelTester):
             self.scope = scope
 
         def prepare_config_and_inputs(self):
-            input_ids = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)
+            encoder_input_ids = ids_tensor([self.batch_size, self.encoder_seq_length], self.vocab_size)
+            decoder_input_ids = ids_tensor([self.batch_size, self.decoder_seq_length], self.vocab_size)
 
-            input_mask = None
-            if self.use_input_mask:
-                input_mask = ids_tensor([self.batch_size, self.seq_length], vocab_size=2)
+            encoder_attention_mask = None
+            decoder_attention_mask = None
+            if self.use_attention_mask:
+                encoder_attention_mask = ids_tensor([self.batch_size, self.encoder_seq_length], vocab_size=2)
+                decoder_attention_mask = ids_tensor([self.batch_size, self.decoder_seq_length], vocab_size=2)
 
-            token_labels = None
+            decoder_lm_labels = None
             if self.use_labels:
-                token_labels = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)
+                decoder_lm_labels = ids_tensor([self.batch_size, self.decoder_seq_length], self.vocab_size)
 
             config = T5Config(
                 vocab_size_or_config_json_file=self.vocab_size,
@@ -100,21 +105,22 @@ class T5ModelTest(CommonTestCases.CommonModelTester):
                 dropout_rate=self.dropout_rate,
                 initializer_factor=self.initializer_factor)
 
-            return (config, input_ids, input_mask, token_labels)
+            return (config, encoder_input_ids, decoder_input_ids, encoder_attention_mask, decoder_attention_mask, decoder_lm_labels)
 
         def check_loss_output(self, result):
             self.parent.assertListEqual(
                 list(result["loss"].size()),
                 [])
 
-        def create_and_check_t5_model(self, config, input_ids, input_mask, token_labels):
+        def create_and_check_t5_model(self, config, encoder_input_ids, decoder_input_ids, encoder_attention_mask, decoder_attention_mask, decoder_lm_labels):
             model = T5Model(config=config)
             model.eval()
-            encoder_output, decoder_output = model(encoder_input_ids=input_ids,
-                                                   decoder_input_ids=input_ids,
-                                                   decoder_attention_mask=input_mask)
-            encoder_output, decoder_output = model(encoder_input_ids=input_ids,
-                                                   decoder_input_ids=input_ids)
+            decoder_output, encoder_output = model(encoder_input_ids=encoder_input_ids,
+                                                   decoder_input_ids=decoder_input_ids,
+                                                   encoder_attention_mask=encoder_attention_mask,
+                                                   decoder_attention_mask=decoder_attention_mask)
+            decoder_output, encoder_output = model(encoder_input_ids=encoder_input_ids,
+                                                   decoder_input_ids=decoder_input_ids)
 
             result = {
                 "encoder_output": encoder_output,
@@ -122,17 +128,17 @@ class T5ModelTest(CommonTestCases.CommonModelTester):
             }
             self.parent.assertListEqual(
                 list(result["encoder_output"].size()),
-                [self.batch_size, self.seq_length, self.hidden_size])
+                [self.batch_size, self.encoder_seq_length, self.hidden_size])
             self.parent.assertListEqual(
                 list(result["decoder_output"].size()),
-                [self.batch_size, self.seq_length, self.hidden_size])
+                [self.batch_size, self.decoder_seq_length, self.hidden_size])
 
 
-        def create_and_check_t5_with_lm_head(self, config, input_ids, input_mask, token_labels):
+        def create_and_check_t5_with_lm_head(self, config, encoder_input_ids, decoder_input_ids, encoder_attention_mask, decoder_attention_mask, decoder_lm_labels):
             model = T5WithLMHeadModel(config=config)
             model.eval()
-            outputs = model(encoder_input_ids=input_ids, decoder_input_ids=input_ids,
-                            decoder_attention_mask=input_mask, decoder_lm_labels=token_labels)
+            outputs = model(encoder_input_ids=encoder_input_ids, decoder_input_ids=decoder_input_ids,
+                            decoder_attention_mask=decoder_attention_mask, decoder_lm_labels=decoder_lm_labels)
             loss, prediction_scores = outputs[0], outputs[1]
             result = {
                 "loss": loss,
@@ -140,15 +146,17 @@ class T5ModelTest(CommonTestCases.CommonModelTester):
             }
             self.parent.assertListEqual(
                 list(result["prediction_scores"].size()),
-                [self.batch_size, self.seq_length, self.vocab_size])
+                [self.batch_size, self.decoder_seq_length, self.vocab_size])
             self.check_loss_output(result)
 
         def prepare_config_and_inputs_for_common(self):
             config_and_inputs = self.prepare_config_and_inputs()
-            (config, input_ids, input_mask, token_labels) = config_and_inputs
-            inputs_dict = {'encoder_input_ids': input_ids,
-                           'decoder_input_ids': input_ids,
-                           'decoder_attention_mask': input_mask}
+            (config, encoder_input_ids, decoder_input_ids, encoder_attention_mask,
+             decoder_attention_mask, decoder_lm_labels) = config_and_inputs
+            inputs_dict = {'encoder_input_ids': encoder_input_ids,
+                           'decoder_input_ids': decoder_input_ids,
+                           'decoder_attention_mask': decoder_attention_mask,
+                           'encoder_attention_mask': encoder_attention_mask}
             return config, inputs_dict
 
     def setUp(self):

From ef99852961b9f3bb87a18a58093c9f513c86b683 Mon Sep 17 00:00:00 2001
From: eukaryote <eukaryote31@gmail.com>
Date: Sat, 9 Nov 2019 16:32:40 +0000
Subject: [PATCH 028/505] from_pretrained: convert DialoGPT format

DialoGPT checkpoints have "lm_head.decoder.weight" instead of "lm_head.weight".

(see: https://www.reddit.com/r/MachineLearning/comments/dt5woy/p_dialogpt_state_of_the_art_conversational_model/f6vmwuy?utm_source=share&utm_medium=web2x)
---
 transformers/modeling_utils.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/transformers/modeling_utils.py b/transformers/modeling_utils.py
index d51eefab58..61dd2546c6 100644
--- a/transformers/modeling_utils.py
+++ b/transformers/modeling_utils.py
@@ -417,6 +417,8 @@ class PreTrainedModel(nn.Module):
                     new_key = key.replace('gamma', 'weight')
                 if 'beta' in key:
                     new_key = key.replace('beta', 'bias')
+                if key == 'lm_head.decoder.weight':
+                    new_key = 'lm_head.weight'
                 if new_key:
                     old_keys.append(key)
                     new_keys.append(new_key)

From 90f6e73a35ee85e94b898a6867f19707b264d387 Mon Sep 17 00:00:00 2001
From: eukaryote <eukaryote31@gmail.com>
Date: Sat, 9 Nov 2019 16:46:19 +0000
Subject: [PATCH 029/505] Add DialoGPT support for Pytorch->TF

---
 transformers/modeling_tf_pytorch_utils.py | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/transformers/modeling_tf_pytorch_utils.py b/transformers/modeling_tf_pytorch_utils.py
index 88ce4d4610..aa74fcc10e 100644
--- a/transformers/modeling_tf_pytorch_utils.py
+++ b/transformers/modeling_tf_pytorch_utils.py
@@ -118,6 +118,9 @@ def load_pytorch_weights_in_tf2_model(tf_model, pt_state_dict, tf_inputs=None, a
             new_key = key.replace('gamma', 'weight')
         if 'beta' in key:
             new_key = key.replace('beta', 'bias')
+        # DialoGPT format
+        if key == 'lm_head.decoder.weight':
+            new_key = 'lm_head.weight'
         if new_key:
             old_keys.append(key)
             new_keys.append(new_key)

From 7246d3c2f93c4461f3ec8ada7a26a002d8f196ea Mon Sep 17 00:00:00 2001
From: Michael Watkins <me@michaelwatkins.eu>
Date: Wed, 6 Nov 2019 13:18:16 +0200
Subject: [PATCH 030/505] Consider do_lower_case in PreTrainedTokenizer

As pointed out in #1545, when using an uncased model, and adding
a new uncased token, the tokenizer does not correctly identify this
in the case that the input text contains the token in a cased format.

For instance, if we load bert-base-uncased into BertTokenizer, and
then use .add_tokens() to add "cool-token", we get the expected
result for .tokenize('this is a cool-token'). However, we get a
possibly unexpected result for .tokenize('this is a cOOl-Token'),
which in fact mirrors the result for the former from before the new
token was added.

This commit adds
- functionality to PreTrainedTokenizer to handle this
situation in case a tokenizer (currently Bert, DistilBert,
and XLNet) has the do_lower_case=True kwarg by:
    1) lowercasing tokens added with .add_tokens()
    2) lowercasing text at the beginning of .tokenize()
- new common test case for tokenizers

https://github.com/huggingface/transformers/issues/1545
---
 .../tests/tokenization_tests_commons.py       | 31 ++++++++++++++++++-
 transformers/tokenization_utils.py            |  5 +++
 2 files changed, 35 insertions(+), 1 deletion(-)

diff --git a/transformers/tests/tokenization_tests_commons.py b/transformers/tests/tokenization_tests_commons.py
index a921696b77..287e6fc7b3 100644
--- a/transformers/tests/tokenization_tests_commons.py
+++ b/transformers/tests/tokenization_tests_commons.py
@@ -110,6 +110,36 @@ class CommonTestCases:
 
             self.assertListEqual(subwords, subwords_loaded)
 
+        def test_added_tokens_do_lower_case(self):
+            tokenizer = self.get_tokenizer(do_lower_case=True)
+
+            text = "aaaaa bbbbbb low cccccccccdddddddd l"
+            text2 = "AAAAA BBBBBB low CCCCCCCCCDDDDDDDD l"
+
+            toks0 = tokenizer.tokenize(text)  # toks before adding new_toks
+
+            new_toks = ["aaaaa bbbbbb", "cccccccccdddddddd", 'AAAAA BBBBBB', 'CCCCCCCCCDDDDDDDD']
+            added = tokenizer.add_tokens(new_toks)
+            self.assertEqual(added, 2)
+
+            toks = tokenizer.tokenize(text)
+            toks2 = tokenizer.tokenize(text2)
+
+            self.assertEqual(len(toks), len(toks2))
+            self.assertNotEqual(len(toks), len(toks0))  # toks0 should be longer
+            self.assertListEqual(toks, toks2)
+
+            tokenizer = self.get_tokenizer(do_lower_case=False)
+
+            added = tokenizer.add_tokens(new_toks)
+            self.assertEqual(added, 4)
+
+            toks = tokenizer.tokenize(text)
+            toks2 = tokenizer.tokenize(text2)
+
+            self.assertEqual(len(toks), len(toks2))  # Length should still be the same
+            self.assertNotEqual(len(toks), len(toks0))
+            self.assertNotEqual(toks[0], toks2[0])  # But at least the first tokens should differ
 
         def test_add_tokens_tokenizer(self):
             tokenizer = self.get_tokenizer()
@@ -160,7 +190,6 @@ class CommonTestCases:
             self.assertEqual(tokens[0], tokenizer.eos_token_id)
             self.assertEqual(tokens[-2], tokenizer.pad_token_id)
 
-
         def test_required_methods_tokenizer(self):
             tokenizer = self.get_tokenizer()
             input_text, output_text = self.get_input_output_texts()
diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index cd14cc4582..fc31c10d25 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -512,6 +512,8 @@ class PreTrainedTokenizer(object):
         to_add_tokens = []
         for token in new_tokens:
             assert isinstance(token, str) or (six.PY2 and isinstance(token, unicode))
+            if self.init_kwargs.get('do_lower_case', False):
+                token = token.lower()
             if token != self.unk_token and \
                     self.convert_tokens_to_ids(token) == self.convert_tokens_to_ids(self.unk_token) and \
                     token not in to_add_tokens:
@@ -605,6 +607,9 @@ class PreTrainedTokenizer(object):
 
             Take care of added tokens.
         """
+        if self.init_kwargs.get('do_lower_case', False):
+            text = text.lower()
+
         def split_on_token(tok, text):
             result = []
             split_text = text.split(tok)

From 14b3aa3b3c300eb1fcc4f3a0c046c87bdabe0afd Mon Sep 17 00:00:00 2001
From: Louis MARTIN <louisrtm@gmail.com>
Date: Fri, 8 Nov 2019 16:47:21 -0800
Subject: [PATCH 031/505] Add tokenization_camembert.py

---
 transformers/tokenization_camembert.py | 120 +++++++++++++++++++++++++
 1 file changed, 120 insertions(+)
 create mode 100644 transformers/tokenization_camembert.py

diff --git a/transformers/tokenization_camembert.py b/transformers/tokenization_camembert.py
new file mode 100644
index 0000000000..9facf7d911
--- /dev/null
+++ b/transformers/tokenization_camembert.py
@@ -0,0 +1,120 @@
+# coding=utf-8
+# Copyright 2018 Google AI, Google Brain and Carnegie Mellon University Authors and the HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" Tokenization classes for Camembert model."""
+from __future__ import (absolute_import, division, print_function,
+                        unicode_literals)
+
+import sentencepiece as spm
+from transformers.tokenization_utils import PreTrainedTokenizer
+
+
+class CamembertTokenizer(PreTrainedTokenizer):
+    """
+        Adapted from RobertaTokenizer and XLNetTokenizer
+        SentencePiece based tokenizer. Peculiarities:
+
+            - requires `SentencePiece <https://github.com/google/sentencepiece>`_
+    """
+    vocab_files_names = {'vocab_file': None}
+
+    def __init__(self, vocab_file, bos_token="<s>", eos_token="</s>", sep_token="</s>",
+                 cls_token="<s>", unk_token="<unk>", pad_token='<pad>', mask_token='<mask>', **kwargs):
+        super(CamembertTokenizer, self).__init__(max_len=512, bos_token=bos_token, eos_token=eos_token, unk_token=unk_token,
+                                                 sep_token=sep_token, cls_token=cls_token, pad_token=pad_token,
+                                                 mask_token=mask_token, **kwargs)
+        self.max_len_single_sentence = self.max_len - 2  # take into account special tokens
+        self.max_len_sentences_pair = self.max_len - 4  # take into account special tokens
+        self.sp_model = spm.SentencePieceProcessor()
+        self.sp_model.Load(str(vocab_file))
+        # HACK: These tokens were added by fairseq but don't seem to be actually used when duplicated in the actual
+        # sentencepiece vocabulary (this is the case for <s> and </s>
+        self.fairseq_tokens_to_ids = {'<s>NOTUSED': 0, '<pad>': 1, '</s>NOTUSED': 2, '<unk>': 3}
+        self.fairseq_offset = len(self.fairseq_tokens_to_ids)
+        self.fairseq_tokens_to_ids['<mask>'] = len(self.sp_model) + len(self.fairseq_tokens_to_ids)
+        self.fairseq_ids_to_tokens = {v: k for k, v in self.fairseq_tokens_to_ids.items()}
+
+    def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
+        """
+        Build model inputs from a sequence or a pair of sequence for sequence classification tasks
+        by concatenating and adding special tokens.
+        A RoBERTa sequence has the following format:
+            single sequence: <s> X </s>
+            pair of sequences: <s> A </s></s> B </s>
+        """
+        if token_ids_1 is None:
+            return [self.cls_token_id] + token_ids_0 + [self.sep_token_id]
+        cls = [self.cls_token_id]
+        sep = [self.sep_token_id]
+        return cls + token_ids_0 + sep + sep + token_ids_1 + sep
+
+    def get_special_tokens_mask(self, token_ids_0, token_ids_1=None, already_has_special_tokens=False):
+        """
+        Retrieves sequence ids from a token list that has no special tokens added. This method is called when adding
+        special tokens using the tokenizer ``prepare_for_model`` or ``encode_plus`` methods.
+
+        Args:
+            token_ids_0: list of ids (must not contain special tokens)
+            token_ids_1: Optional list of ids (must not contain special tokens), necessary when fetching sequence ids
+                for sequence pairs
+            already_has_special_tokens: (default False) Set to True if the token list is already formated with
+                special tokens for the model
+
+        Returns:
+            A list of integers in the range [0, 1]: 0 for a special token, 1 for a sequence token.
+        """
+        if already_has_special_tokens:
+            if token_ids_1 is not None:
+                raise ValueError("You should not supply a second sequence if the provided sequence of "
+                                 "ids is already formated with special tokens for the model.")
+            return list(map(lambda x: 1 if x in [self.sep_token_id, self.cls_token_id] else 0, token_ids_0))
+
+        if token_ids_1 is None:
+            return [1] + ([0] * len(token_ids_0)) + [1]
+        return [1] + ([0] * len(token_ids_0)) + [1, 1] + ([0] * len(token_ids_1)) + [1]
+
+    def create_token_type_ids_from_sequences(self, token_ids_0, token_ids_1=None):
+        """
+        Creates a mask from the two sequences passed to be used in a sequence-pair classification task.
+        A RoBERTa sequence pair mask has the following format:
+        0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1
+        | first sequence    | second sequence
+
+        if token_ids_1 is None, only returns the first portion of the mask (0's).
+        """
+        sep = [self.sep_token_id]
+        cls = [self.cls_token_id]
+
+        if token_ids_1 is None:
+            return len(cls + token_ids_0 + sep) * [0]
+        return len(cls + token_ids_0 + sep + sep) * [0] + len(token_ids_1 + sep) * [1]
+
+    @property
+    def vocab_size(self):
+        return self.fairseq_offset + len(self.sp_model)
+
+    def _tokenize(self, text):
+        return self.sp_model.EncodeAsPieces(text)
+
+    def _convert_token_to_id(self, token):
+        """ Converts a token (str/unicode) in an id using the vocab. """
+        if token in self.fairseq_tokens_to_ids:
+            return self.fairseq_tokens_to_ids[token]
+        return self.fairseq_offset + self.sp_model.PieceToId(token)
+
+    def _convert_id_to_token(self, index):
+        """Converts an index (integer) in a token (string/unicode) using the vocab."""
+        if index in self.fairseq_ids_to_tokens:
+            return self.fairseq_ids_to_tokens[index]
+        return self.sp_model.IdToPiece(index - self.fairseq_offset)

From 6e72fd094c98901cc90d146d3fe3cd5a0e879911 Mon Sep 17 00:00:00 2001
From: Louis MARTIN <louisrtm@gmail.com>
Date: Fri, 8 Nov 2019 17:09:48 -0800
Subject: [PATCH 032/505] Add demo_camembert.py

---
 examples/demo_camembert.py | 59 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 59 insertions(+)
 create mode 100644 examples/demo_camembert.py

diff --git a/examples/demo_camembert.py b/examples/demo_camembert.py
new file mode 100644
index 0000000000..df28f4f267
--- /dev/null
+++ b/examples/demo_camembert.py
@@ -0,0 +1,59 @@
+from pathlib import Path
+import tarfile
+import urllib.request
+
+import torch
+
+from transformers.tokenization_camembert import CamembertTokenizer
+from transformers.modeling_roberta import RobertaForMaskedLM
+
+
+def fill_mask(masked_input, model, tokenizer, topk=5):
+    # Adapted from https://github.com/pytorch/fairseq/blob/master/fairseq/models/roberta/hub_interface.py
+    assert masked_input.count('<mask>') == 1
+    input_ids = torch.tensor(tokenizer.encode(masked_input, add_special_tokens=True)).unsqueeze(0)  # Batch size 1
+    logits = model(input_ids)[0]  # The last hidden-state is the first element of the output tuple
+    masked_index = (input_ids.squeeze() == tokenizer.mask_token_id).nonzero().item()
+    logits = logits[0, masked_index, :]
+    prob = logits.softmax(dim=0)
+    values, indices = prob.topk(k=topk, dim=0)
+    topk_predicted_token_bpe = ' '.join([tokenizer.convert_ids_to_tokens(indices[i].item())
+                                         for i in range(len(indices))])
+    masked_token = tokenizer.mask_token
+    topk_filled_outputs = []
+    for index, predicted_token_bpe in enumerate(topk_predicted_token_bpe.split(' ')):
+        predicted_token = predicted_token_bpe.replace('\u2581', ' ')
+        if " {0}".format(masked_token) in masked_input:
+            topk_filled_outputs.append((
+                masked_input.replace(
+                    ' {0}'.format(masked_token), predicted_token
+                ),
+                values[index].item(),
+                predicted_token,
+            ))
+        else:
+            topk_filled_outputs.append((
+                masked_input.replace(masked_token, predicted_token),
+                values[index].item(),
+                predicted_token,
+            ))
+    return topk_filled_outputs
+
+
+model_path = Path('camembert.v0.pytorch')
+if not model_path.exists():
+    compressed_path = model_path.with_suffix('.tar.gz')
+    url = 'http://dl.fbaipublicfiles.com/camembert/camembert.v0.pytorch.tar.gz'
+    print('Downloading model...')
+    urllib.request.urlretrieve(url, compressed_path)
+    print('Extracting model...')
+    with tarfile.open(compressed_path) as f:
+        f.extractall(model_path.parent)
+    assert model_path.exists()
+tokenizer_path = model_path / 'sentencepiece.bpe.model'
+tokenizer = CamembertTokenizer.from_pretrained(tokenizer_path)
+model = RobertaForMaskedLM.from_pretrained(model_path)
+model.eval()
+
+masked_input = "Le camembert est <mask> :)"
+print(fill_mask(masked_input, model, tokenizer, topk=3))

From e44b939e7198fac5ce4085a1bda8e17b3934e67f Mon Sep 17 00:00:00 2001
From: Louis MARTIN <louisrtm@gmail.com>
Date: Tue, 12 Nov 2019 17:11:00 -0800
Subject: [PATCH 033/505] Add configuration_camembert.py and
 modeling_camembert.py

---
 transformers/configuration_camembert.py |  33 +++
 transformers/modeling_camembert.py      | 257 ++++++++++++++++++++++++
 2 files changed, 290 insertions(+)
 create mode 100644 transformers/configuration_camembert.py
 create mode 100644 transformers/modeling_camembert.py

diff --git a/transformers/configuration_camembert.py b/transformers/configuration_camembert.py
new file mode 100644
index 0000000000..07ebd6e82e
--- /dev/null
+++ b/transformers/configuration_camembert.py
@@ -0,0 +1,33 @@
+# coding=utf-8
+# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.
+# Copyright (c) 2018, NVIDIA CORPORATION.  All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" CamemBERT configuration """
+
+from __future__ import (absolute_import, division, print_function,
+                        unicode_literals)
+
+import logging
+
+from .configuration_roberta import RobertaConfig
+
+logger = logging.getLogger(__name__)
+
+CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP = {
+    'camembert-base': "https://dl.fbaipublicfiles.com/camembert/camembert-base-v0-config.json",
+}
+
+
+class CamembertConfig(RobertaConfig):
+    pretrained_config_archive_map = CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
diff --git a/transformers/modeling_camembert.py b/transformers/modeling_camembert.py
new file mode 100644
index 0000000000..982c349531
--- /dev/null
+++ b/transformers/modeling_camembert.py
@@ -0,0 +1,257 @@
+# coding=utf-8
+# Copyright 2019 Inria, Facebook AI Research and the HuggingFace Inc. team.
+# Copyright (c) 2018, NVIDIA CORPORATION.  All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""PyTorch CamemBERT model. """
+
+from __future__ import (absolute_import, division, print_function,
+                        unicode_literals)
+
+import logging
+
+from .modeling_roberta import RobertaModel, RobertaForMaskedLM, RobertaForSequenceClassification, RobertaForMultipleChoice
+from .configuration_camembert import CamembertConfig
+from .file_utils import add_start_docstrings
+
+logger = logging.getLogger(__name__)
+
+CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
+    'camembert-base': "https://dl.fbaipublicfiles.com/camembert/camembert-base-v0-pytorch_model.bin",
+}
+
+
+CAMEMBERT_START_DOCSTRING = r"""    The CamemBERT model was proposed in
+    `CamemBERT: a Tasty French Language Model`_
+    by Louis Martin, Benjamin Muller, Pedro Javier Ortiz Suárez, Yoann Dupont, Laurent Romary, Éric Villemonte de la Clergerie, Djamé Seddah, and Benoît Sagot. It is based on Facebook's RoBERTa model released in 2019.
+    
+    It is a model trained on 138GB of French text.
+    
+    This implementation is the same RoBERTa.
+
+    This model is a PyTorch `torch.nn.Module`_ sub-class. Use it as a regular PyTorch Module and
+    refer to the PyTorch documentation for all matter related to general usage and behavior.
+
+    .. _`CamemBERT: a Tasty French Language Model`:
+        https://arxiv.org/abs/1911.03894
+
+    .. _`torch.nn.Module`:
+        https://pytorch.org/docs/stable/nn.html#module
+
+    Parameters:
+        config (:class:`~transformers.CamembertConfig`): Model configuration class with all the parameters of the 
+            model. Initializing with a config file does not load the weights associated with the model, only the configuration.
+            Check out the :meth:`~transformers.PreTrainedModel.from_pretrained` method to load the model weights.
+"""
+
+CAMEMBERT_INPUTS_DOCSTRING = r"""
+    Inputs:
+        **input_ids**: ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
+            Indices of input sequence tokens in the vocabulary.
+            To match pre-training, CamemBERT input sequence should be formatted with <s> and </s> tokens as follows:
+
+            (a) For sequence pairs:
+
+                ``tokens:         <s> Is this Jacksonville ? </s> </s> No it is not . </s>``
+
+            (b) For single sequences:
+
+                ``tokens:         <s> the dog is hairy . </s>``
+
+            Fully encoded sequences or sequence pairs can be obtained using the CamembertTokenizer.encode function with 
+            the ``add_special_tokens`` parameter set to ``True``.
+
+            CamemBERT is a model with absolute position embeddings so it's usually advised to pad the inputs on
+            the right rather than the left.
+
+            See :func:`transformers.PreTrainedTokenizer.encode` and
+            :func:`transformers.PreTrainedTokenizer.convert_tokens_to_ids` for details.
+        **attention_mask**: (`optional`) ``torch.FloatTensor`` of shape ``(batch_size, sequence_length)``:
+            Mask to avoid performing attention on padding token indices.
+            Mask values selected in ``[0, 1]``:
+            ``1`` for tokens that are NOT MASKED, ``0`` for MASKED tokens.
+        **token_type_ids**: (`optional` need to be trained) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
+            Optional segment token indices to indicate first and second portions of the inputs.
+            This embedding matrice is not trained (not pretrained during CamemBERT pretraining), you will have to train it
+            during finetuning.
+            Indices are selected in ``[0, 1]``: ``0`` corresponds to a `sentence A` token, ``1``
+            corresponds to a `sentence B` token
+            (see `BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding`_ for more details).
+        **position_ids**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
+            Indices of positions of each input sequence tokens in the position embeddings.
+            Selected in the range ``[0, config.max_position_embeddings - 1[``.
+        **head_mask**: (`optional`) ``torch.FloatTensor`` of shape ``(num_heads,)`` or ``(num_layers, num_heads)``:
+            Mask to nullify selected heads of the self-attention modules.
+            Mask values selected in ``[0, 1]``:
+            ``1`` indicates the head is **not masked**, ``0`` indicates the head is **masked**.
+"""
+
+@add_start_docstrings("The bare CamemBERT Model transformer outputting raw hidden-states without any specific head on top.",
+                      CAMEMBERT_START_DOCSTRING, CAMEMBERT_INPUTS_DOCSTRING)
+class CamembertModel(RobertaModel):
+    r"""
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **last_hidden_state**: ``torch.FloatTensor`` of shape ``(batch_size, sequence_length, hidden_size)``
+            Sequence of hidden-states at the output of the last layer of the model.
+        **pooler_output**: ``torch.FloatTensor`` of shape ``(batch_size, hidden_size)``
+            Last layer hidden-state of the first token of the sequence (classification token)
+            further processed by a Linear layer and a Tanh activation function. The Linear
+            layer weights are trained from the next sentence prediction (classification)
+            eo match pre-training, CamemBERT input sequence should be formatted with [CLS] and [SEP] tokens as follows:
+
+            (a) For sequence pairs:
+
+                ``tokens:         [CLS] is this jack ##son ##ville ? [SEP] [SEP] no it is not . [SEP]``
+
+                ``token_type_ids:   0   0  0    0    0     0       0   0   0     1  1  1  1   1   1``
+
+            (b) For single sequences:
+
+                ``tokens:         [CLS] the dog is hairy . [SEP]``
+
+                ``token_type_ids:   0   0   0   0  0     0   0``
+
+            objective during Bert pretraining. This output is usually *not* a good summary
+            of the semantic content of the input, you're often better with averaging or pooling
+            the sequence of hidden-states for the whole input sequence.
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``torch.FloatTensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        tokenizer = CamembertTokenizer.from_pretrained('camembert-base')
+        model = CamembertModel.from_pretrained('camembert-base')
+        input_ids = torch.tensor(tokenizer.encode("J'aime le camembert !")).unsqueeze(0)  # Batch size 1
+        outputs = model(input_ids)
+        last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
+
+    """
+    config_class = CamembertConfig
+    pretrained_model_archive_map = CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP
+    base_model_prefix = "camembert"
+
+
+@add_start_docstrings("""CamemBERT Model with a `language modeling` head on top. """,
+    CAMEMBERT_START_DOCSTRING, CAMEMBERT_INPUTS_DOCSTRING)
+class CamembertForMaskedLM(RobertaForMaskedLM):
+    r"""
+        **masked_lm_labels**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
+            Labels for computing the masked language modeling loss.
+            Indices should be in ``[-1, 0, ..., config.vocab_size]`` (see ``input_ids`` docstring)
+            Tokens with indices set to ``-1`` are ignored (masked), the loss is only computed for the tokens with labels
+            in ``[0, ..., config.vocab_size]``
+
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **loss**: (`optional`, returned when ``masked_lm_labels`` is provided) ``torch.FloatTensor`` of shape ``(1,)``:
+            Masked language modeling loss.
+        **prediction_scores**: ``torch.FloatTensor`` of shape ``(batch_size, sequence_length, config.vocab_size)``
+            Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``torch.FloatTensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        tokenizer = CamembertTokenizer.from_pretrained('camembert-base')
+        model = CamembertForMaskedLM.from_pretrained('camembert-base')
+        input_ids = torch.tensor(tokenizer.encode("J'aime le camembert !")).unsqueeze(0)  # Batch size 1
+        outputs = model(input_ids, masked_lm_labels=input_ids)
+        loss, prediction_scores = outputs[:2]
+
+    """
+    config_class = CamembertConfig
+    pretrained_model_archive_map = CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP
+    base_model_prefix = "camembert"
+
+
+@add_start_docstrings("""CamemBERT Model transformer with a sequence classification/regression head on top (a linear layer 
+    on top of the pooled output) e.g. for GLUE tasks. """,
+    CAMEMBERT_START_DOCSTRING, CAMEMBERT_INPUTS_DOCSTRING)
+class CamembertForSequenceClassification(RobertaForSequenceClassification):
+    r"""
+        **labels**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size,)``:
+            Labels for computing the sequence classification/regression loss.
+            Indices should be in ``[0, ..., config.num_labels]``.
+            If ``config.num_labels == 1`` a regression loss is computed (Mean-Square loss),
+            If ``config.num_labels > 1`` a classification loss is computed (Cross-Entropy).
+
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **loss**: (`optional`, returned when ``labels`` is provided) ``torch.FloatTensor`` of shape ``(1,)``:
+            Classification (or regression if config.num_labels==1) loss.
+        **logits**: ``torch.FloatTensor`` of shape ``(batch_size, config.num_labels)``
+            Classification (or regression if config.num_labels==1) scores (before SoftMax).
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``torch.FloatTensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        tokenizer = CamembertTokenizer.from_pretrained('camembert-base')
+        model = CamembertForSequenceClassification.from_pretrained('camembert-base')
+        input_ids = torch.tensor(tokenizer.encode("J'aime le camembert !")).unsqueeze(0)  # Batch size 1
+        labels = torch.tensor([1]).unsqueeze(0)  # Batch size 1
+        outputs = model(input_ids, labels=labels)
+        loss, logits = outputs[:2]
+
+    """
+    config_class = CamembertConfig
+    pretrained_model_archive_map = CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP
+    base_model_prefix = "camembert"
+
+
+@add_start_docstrings("""CamemBERT Model with a multiple choice classification head on top (a linear layer on top of
+    the pooled output and a softmax) e.g. for RocStories/SWAG tasks. """,
+    CAMEMBERT_START_DOCSTRING, CAMEMBERT_INPUTS_DOCSTRING)
+class CamembertForMultipleChoice(RobertaForMultipleChoice):
+    r"""
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **loss**: (`optional`, returned when ``labels`` is provided) ``torch.FloatTensor`` of shape ``(1,)``:
+            Classification loss.
+        **classification_scores**: ``torch.FloatTensor`` of shape ``(batch_size, num_choices)`` where `num_choices` is the size of the second dimension
+            of the input tensors. (see `input_ids` above).
+            Classification scores (before SoftMax).
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``torch.FloatTensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        tokenizer = CamembertTokenizer.from_pretrained('camembert-base')
+        model = CamembertForMultipleChoice.from_pretrained('camembert-base')
+        choices = ["J'aime le camembert !", "Je deteste le camembert !"]
+        input_ids = torch.tensor([tokenizer.encode(s, add_special_tokens=True) for s in choices]).unsqueeze(0)  # Batch size 1, 2 choices
+        labels = torch.tensor(1).unsqueeze(0)  # Batch size 1
+        outputs = model(input_ids, labels=labels)
+        loss, classification_scores = outputs[:2]
+
+    """
+    config_class = CamembertConfig
+    pretrained_model_archive_map = CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP
+    base_model_prefix = "camembert"

From fb6c70a91d3183742ce0a6d97add68103253ca3a Mon Sep 17 00:00:00 2001
From: Louis MARTIN <louisrtm@gmail.com>
Date: Tue, 12 Nov 2019 17:11:49 -0800
Subject: [PATCH 034/505] Update tokenization_camembert.py with urls

---
 transformers/tokenization_camembert.py | 19 +++++++++++++++++--
 1 file changed, 17 insertions(+), 2 deletions(-)

diff --git a/transformers/tokenization_camembert.py b/transformers/tokenization_camembert.py
index 9facf7d911..0a6e751351 100644
--- a/transformers/tokenization_camembert.py
+++ b/transformers/tokenization_camembert.py
@@ -11,7 +11,7 @@
 # distributed under the License is distributed on an "AS IS" BASIS,
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
-# limitations under the License.
+# limitations under the License
 """ Tokenization classes for Camembert model."""
 from __future__ import (absolute_import, division, print_function,
                         unicode_literals)
@@ -20,6 +20,19 @@ import sentencepiece as spm
 from transformers.tokenization_utils import PreTrainedTokenizer
 
 
+VOCAB_FILES_NAMES = {'vocab_file': 'sentencepiece.bpe.model'}
+
+PRETRAINED_VOCAB_FILES_MAP = {
+    'vocab_file':
+    {
+    'camembert-base': "https://dl.fbaipublicfiles.com/camembert/camembert-base-v0-sentencepiece.bpe.model",
+    }
+}
+
+PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
+    'camembert-base': None,
+}
+
 class CamembertTokenizer(PreTrainedTokenizer):
     """
         Adapted from RobertaTokenizer and XLNetTokenizer
@@ -27,7 +40,9 @@ class CamembertTokenizer(PreTrainedTokenizer):
 
             - requires `SentencePiece <https://github.com/google/sentencepiece>`_
     """
-    vocab_files_names = {'vocab_file': None}
+    vocab_files_names = VOCAB_FILES_NAMES
+    pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+    max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
 
     def __init__(self, vocab_file, bos_token="<s>", eos_token="</s>", sep_token="</s>",
                  cls_token="<s>", unk_token="<unk>", pad_token='<pad>', mask_token='<mask>', **kwargs):

From f12e4d8da783a535cd8978656f0a3ca108a2e2ed Mon Sep 17 00:00:00 2001
From: Louis MARTIN <louisrtm@gmail.com>
Date: Tue, 12 Nov 2019 17:14:50 -0800
Subject: [PATCH 035/505] Move demo_camembert.py to examples/contrib

---
 examples/{ => contrib}/demo_camembert.py | 0
 1 file changed, 0 insertions(+), 0 deletions(-)
 rename examples/{ => contrib}/demo_camembert.py (100%)

diff --git a/examples/demo_camembert.py b/examples/contrib/demo_camembert.py
similarity index 100%
rename from examples/demo_camembert.py
rename to examples/contrib/demo_camembert.py

From 3e20c2e871db82f81c3b2b814265a481be15273c Mon Sep 17 00:00:00 2001
From: Louis MARTIN <louisrtm@gmail.com>
Date: Tue, 12 Nov 2019 17:16:24 -0800
Subject: [PATCH 036/505] Update demo_camembert.py with new classes

---
 examples/contrib/demo_camembert.py | 17 +++--------------
 1 file changed, 3 insertions(+), 14 deletions(-)

diff --git a/examples/contrib/demo_camembert.py b/examples/contrib/demo_camembert.py
index df28f4f267..28144d5167 100644
--- a/examples/contrib/demo_camembert.py
+++ b/examples/contrib/demo_camembert.py
@@ -5,7 +5,7 @@ import urllib.request
 import torch
 
 from transformers.tokenization_camembert import CamembertTokenizer
-from transformers.modeling_roberta import RobertaForMaskedLM
+from transformers.modeling_camembert import CamembertForMaskedLM
 
 
 def fill_mask(masked_input, model, tokenizer, topk=5):
@@ -40,19 +40,8 @@ def fill_mask(masked_input, model, tokenizer, topk=5):
     return topk_filled_outputs
 
 
-model_path = Path('camembert.v0.pytorch')
-if not model_path.exists():
-    compressed_path = model_path.with_suffix('.tar.gz')
-    url = 'http://dl.fbaipublicfiles.com/camembert/camembert.v0.pytorch.tar.gz'
-    print('Downloading model...')
-    urllib.request.urlretrieve(url, compressed_path)
-    print('Extracting model...')
-    with tarfile.open(compressed_path) as f:
-        f.extractall(model_path.parent)
-    assert model_path.exists()
-tokenizer_path = model_path / 'sentencepiece.bpe.model'
-tokenizer = CamembertTokenizer.from_pretrained(tokenizer_path)
-model = RobertaForMaskedLM.from_pretrained(model_path)
+tokenizer = CamembertTokenizer.from_pretrained('camembert-base')
+model = CamembertForMaskedLM.from_pretrained('camembert-base')
 model.eval()
 
 masked_input = "Le camembert est <mask> :)"

From 694d4fcbb61f15b66781219954112791248d832e Mon Sep 17 00:00:00 2001
From: Louis MARTIN <louisrtm@gmail.com>
Date: Tue, 12 Nov 2019 17:28:37 -0800
Subject: [PATCH 037/505] Add CamemBERT classes to __init__.py

---
 transformers/__init__.py | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/transformers/__init__.py b/transformers/__init__.py
index dd9e18a050..cdf0669b39 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -42,6 +42,7 @@ from .tokenization_xlnet import XLNetTokenizer, SPIECE_UNDERLINE
 from .tokenization_xlm import XLMTokenizer
 from .tokenization_roberta import RobertaTokenizer
 from .tokenization_distilbert import DistilBertTokenizer
+from .tokenization_camembert import CamembertTokenizer
 
 # Configurations
 from .configuration_utils import PretrainedConfig
@@ -56,6 +57,7 @@ from .configuration_ctrl import CTRLConfig, CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_xlm import XLMConfig, XLM_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_roberta import RobertaConfig, ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_distilbert import DistilBertConfig, DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_camembert import CamembertConfig, CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
 
 # Modeling
 if is_torch_available():
@@ -96,6 +98,9 @@ if is_torch_available():
                                 DistilBertForSequenceClassification, DistilBertForQuestionAnswering,
                                 DistilBertForTokenClassification,
                                 DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
+    from .modeling_camembert import (CamembertForMaskedLM, CamembertModel,
+                                CamembertForSequenceClassification, CamembertForMultipleChoice,
+                                CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
     from .modeling_encoder_decoder import PreTrainedEncoderDecoder, Model2Model
 
     # Optimization

From 035fea53157b43d311b1e1164395398c4ec0dda5 Mon Sep 17 00:00:00 2001
From: Louis MARTIN <louisrtm@gmail.com>
Date: Tue, 12 Nov 2019 17:41:41 -0800
Subject: [PATCH 038/505] Add CamemBERT to auto files and docs

---
 docs/source/pretrained_models.rst  | 6 +++++-
 transformers/configuration_auto.py | 7 ++++++-
 transformers/tokenization_auto.py  | 9 +++++++--
 3 files changed, 18 insertions(+), 4 deletions(-)

diff --git a/docs/source/pretrained_models.rst b/docs/source/pretrained_models.rst
index edb47e7f1c..b0a578fd80 100644
--- a/docs/source/pretrained_models.rst
+++ b/docs/source/pretrained_models.rst
@@ -155,5 +155,9 @@ Here is the full list of the currently provided pretrained models together with
 | CTRL              | ``ctrl``                                                   | | 48-layer, 1280-hidden, 16-heads, 1.6B parameters                                                                                    |
 |                   |                                                            | | Salesforce's Large-sized CTRL English model                                                                                         |
 +-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+| CamemBERT         | ``camembert-base``                                         | | 12-layer, 768-hidden, 12-heads, 110M parameters                                                                                     |
+|                   |                                                            | | CamemBERT using the BERT-base architecture                                                                                          |
+|                   |                                                            | (see `details <https://github.com/pytorch/fairseq/tree/master/examples/camembert>`__)                                                 |
++-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 
-.. <https://huggingface.co/transformers/examples.html>`__
\ No newline at end of file
+.. <https://huggingface.co/transformers/examples.html>`__
diff --git a/transformers/configuration_auto.py b/transformers/configuration_auto.py
index edd21a670c..2906136139 100644
--- a/transformers/configuration_auto.py
+++ b/transformers/configuration_auto.py
@@ -27,6 +27,7 @@ from .configuration_xlm import XLMConfig
 from .configuration_roberta import RobertaConfig
 from .configuration_distilbert import DistilBertConfig
 from .configuration_ctrl import CTRLConfig
+from .configuration_camembert import CamembertConfig
 
 logger = logging.getLogger(__name__)
 
@@ -50,6 +51,7 @@ class AutoConfig(object):
             - contains `xlnet`: XLNetConfig (XLNet model)
             - contains `xlm`: XLMConfig (XLM model)
             - contains `roberta`: RobertaConfig (RoBERTa model)
+            - contains `camembert`: CamembertConfig (CamemBERT model)
             - contains `ctrl` : CTRLConfig (CTRL model)
         This class cannot be instantiated using `__init__()` (throw an error).
     """
@@ -72,6 +74,7 @@ class AutoConfig(object):
             - contains `xlnet`: XLNetConfig (XLNet model)
             - contains `xlm`: XLMConfig (XLM model)
             - contains `roberta`: RobertaConfig (RoBERTa model)
+            - contains `camembert`: CamembertConfig (CamemBERT model)
             - contains `ctrl` : CTRLConfig (CTRL model)
         Params:
             pretrained_model_name_or_path: either:
@@ -116,6 +119,8 @@ class AutoConfig(object):
         """
         if 'distilbert' in pretrained_model_name_or_path:
             return DistilBertConfig.from_pretrained(pretrained_model_name_or_path, **kwargs)
+        elif 'camembert' in pretrained_model_name_or_path:
+            return CamembertConfig.from_pretrained(pretrained_model_name_or_path, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return RobertaConfig.from_pretrained(pretrained_model_name_or_path, **kwargs)
         elif 'bert' in pretrained_model_name_or_path:
@@ -134,4 +139,4 @@ class AutoConfig(object):
             return CTRLConfig.from_pretrained(pretrained_model_name_or_path, **kwargs)
         raise ValueError("Unrecognized model identifier in {}. Should contains one of "
                          "'bert', 'openai-gpt', 'gpt2', 'transfo-xl', 'xlnet', "
-                         "'xlm', 'roberta', 'ctrl'".format(pretrained_model_name_or_path))
+                         "'xlm', 'roberta', 'camembert', 'ctrl'".format(pretrained_model_name_or_path))
diff --git a/transformers/tokenization_auto.py b/transformers/tokenization_auto.py
index ec056de17f..4510159905 100644
--- a/transformers/tokenization_auto.py
+++ b/transformers/tokenization_auto.py
@@ -27,6 +27,7 @@ from .tokenization_xlnet import XLNetTokenizer
 from .tokenization_xlm import XLMTokenizer
 from .tokenization_roberta import RobertaTokenizer
 from .tokenization_distilbert import DistilBertTokenizer
+from .tokenization_camembert import CamembertTokenizer
 
 logger = logging.getLogger(__name__)
 
@@ -41,6 +42,7 @@ class AutoTokenizer(object):
 
         The tokenizer class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
+            - contains `camembert`: CamembertTokenizer (CamemBERT model)
             - contains `distilbert`: DistilBertTokenizer (DistilBert model)
             - contains `roberta`: RobertaTokenizer (RoBERTa model)
             - contains `bert`: BertTokenizer (Bert model)
@@ -64,8 +66,9 @@ class AutoTokenizer(object):
 
         The tokenizer class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
+            - contains `camembert`: CamembertTokenizer (CamemBERT model)
             - contains `distilbert`: DistilBertTokenizer (DistilBert model)
-            - contains `roberta`: RobertaTokenizer (XLM model)
+            - contains `roberta`: RobertaTokenizer (RoBERTa model)
             - contains `bert`: BertTokenizer (Bert model)
             - contains `openai-gpt`: OpenAIGPTTokenizer (OpenAI GPT model)
             - contains `gpt2`: GPT2Tokenizer (OpenAI GPT-2 model)
@@ -103,6 +106,8 @@ class AutoTokenizer(object):
         """
         if 'distilbert' in pretrained_model_name_or_path:
             return DistilBertTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
+        elif 'camembert' in pretrained_model_name_or_path:
+            return CamembertTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return RobertaTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
         elif 'bert' in pretrained_model_name_or_path:
@@ -121,4 +126,4 @@ class AutoTokenizer(object):
             return CTRLTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
         raise ValueError("Unrecognized model identifier in {}. Should contains one of "
                          "'bert', 'openai-gpt', 'gpt2', 'transfo-xl', 'xlnet', "
-                         "'xlm', 'roberta', 'ctrl'".format(pretrained_model_name_or_path))
+                         "'xlm', 'roberta', 'camembert', 'ctrl'".format(pretrained_model_name_or_path))

From 26858f27cb352b6bb1cda2b090413d1d2206a9ee Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Fri, 15 Nov 2019 23:23:31 -0500
Subject: [PATCH 039/505] [camembert] Upload to s3 + rename script

---
 examples/contrib/{demo_camembert.py => run_camembert.py} | 0
 transformers/configuration_camembert.py                  | 2 +-
 transformers/modeling_camembert.py                       | 2 +-
 transformers/tokenization_camembert.py                   | 2 +-
 4 files changed, 3 insertions(+), 3 deletions(-)
 rename examples/contrib/{demo_camembert.py => run_camembert.py} (100%)

diff --git a/examples/contrib/demo_camembert.py b/examples/contrib/run_camembert.py
similarity index 100%
rename from examples/contrib/demo_camembert.py
rename to examples/contrib/run_camembert.py
diff --git a/transformers/configuration_camembert.py b/transformers/configuration_camembert.py
index 07ebd6e82e..3ff64454e5 100644
--- a/transformers/configuration_camembert.py
+++ b/transformers/configuration_camembert.py
@@ -25,7 +25,7 @@ from .configuration_roberta import RobertaConfig
 logger = logging.getLogger(__name__)
 
 CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP = {
-    'camembert-base': "https://dl.fbaipublicfiles.com/camembert/camembert-base-v0-config.json",
+    'camembert-base': "https://s3.amazonaws.com/models.huggingface.co/bert/camembert-base-config.json",
 }
 
 
diff --git a/transformers/modeling_camembert.py b/transformers/modeling_camembert.py
index 982c349531..a0f5933bc2 100644
--- a/transformers/modeling_camembert.py
+++ b/transformers/modeling_camembert.py
@@ -27,7 +27,7 @@ from .file_utils import add_start_docstrings
 logger = logging.getLogger(__name__)
 
 CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
-    'camembert-base': "https://dl.fbaipublicfiles.com/camembert/camembert-base-v0-pytorch_model.bin",
+    'camembert-base': "https://s3.amazonaws.com/models.huggingface.co/bert/camembert-base-pytorch_model.bin",
 }
 
 
diff --git a/transformers/tokenization_camembert.py b/transformers/tokenization_camembert.py
index 0a6e751351..de587ac863 100644
--- a/transformers/tokenization_camembert.py
+++ b/transformers/tokenization_camembert.py
@@ -25,7 +25,7 @@ VOCAB_FILES_NAMES = {'vocab_file': 'sentencepiece.bpe.model'}
 PRETRAINED_VOCAB_FILES_MAP = {
     'vocab_file':
     {
-    'camembert-base': "https://dl.fbaipublicfiles.com/camembert/camembert-base-v0-sentencepiece.bpe.model",
+    'camembert-base': "https://s3.amazonaws.com/models.huggingface.co/bert/camembert-base-sentencepiece.bpe.model",
     }
 }
 

From f9abf73e319d99ed74f46671e98b00b9328cd245 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Fri, 15 Nov 2019 23:40:20 -0500
Subject: [PATCH 040/505] [camembert] realign w/ recent changes

---
 transformers/modeling_camembert.py     | 10 +++++-----
 transformers/tokenization_camembert.py |  2 +-
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/transformers/modeling_camembert.py b/transformers/modeling_camembert.py
index a0f5933bc2..05538926e2 100644
--- a/transformers/modeling_camembert.py
+++ b/transformers/modeling_camembert.py
@@ -37,7 +37,7 @@ CAMEMBERT_START_DOCSTRING = r"""    The CamemBERT model was proposed in
     
     It is a model trained on 138GB of French text.
     
-    This implementation is the same RoBERTa.
+    This implementation is the same as RoBERTa.
 
     This model is a PyTorch `torch.nn.Module`_ sub-class. Use it as a regular PyTorch Module and
     refer to the PyTorch documentation for all matter related to general usage and behavior.
@@ -94,6 +94,10 @@ CAMEMBERT_INPUTS_DOCSTRING = r"""
             Mask to nullify selected heads of the self-attention modules.
             Mask values selected in ``[0, 1]``:
             ``1`` indicates the head is **not masked**, ``0`` indicates the head is **masked**.
+        **inputs_embeds**: (`optional`) ``torch.FloatTensor`` of shape ``(batch_size, sequence_length, embedding_dim)``:
+            Optionally, instead of passing ``input_ids`` you can choose to directly pass an embedded representation.
+            This is useful if you want more control over how to convert `input_ids` indices into associated vectors
+            than the model's internal embedding lookup matrix.
 """
 
 @add_start_docstrings("The bare CamemBERT Model transformer outputting raw hidden-states without any specific head on top.",
@@ -143,7 +147,6 @@ class CamembertModel(RobertaModel):
     """
     config_class = CamembertConfig
     pretrained_model_archive_map = CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP
-    base_model_prefix = "camembert"
 
 
 @add_start_docstrings("""CamemBERT Model with a `language modeling` head on top. """,
@@ -180,7 +183,6 @@ class CamembertForMaskedLM(RobertaForMaskedLM):
     """
     config_class = CamembertConfig
     pretrained_model_archive_map = CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP
-    base_model_prefix = "camembert"
 
 
 @add_start_docstrings("""CamemBERT Model transformer with a sequence classification/regression head on top (a linear layer 
@@ -219,7 +221,6 @@ class CamembertForSequenceClassification(RobertaForSequenceClassification):
     """
     config_class = CamembertConfig
     pretrained_model_archive_map = CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP
-    base_model_prefix = "camembert"
 
 
 @add_start_docstrings("""CamemBERT Model with a multiple choice classification head on top (a linear layer on top of
@@ -254,4 +255,3 @@ class CamembertForMultipleChoice(RobertaForMultipleChoice):
     """
     config_class = CamembertConfig
     pretrained_model_archive_map = CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP
-    base_model_prefix = "camembert"
diff --git a/transformers/tokenization_camembert.py b/transformers/tokenization_camembert.py
index de587ac863..ae1b322941 100644
--- a/transformers/tokenization_camembert.py
+++ b/transformers/tokenization_camembert.py
@@ -87,7 +87,7 @@ class CamembertTokenizer(PreTrainedTokenizer):
                 special tokens for the model
 
         Returns:
-            A list of integers in the range [0, 1]: 0 for a special token, 1 for a sequence token.
+            A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
         """
         if already_has_special_tokens:
             if token_ids_1 is not None:

From 0477b307c7501ea76e01b03cb387a2312db752b3 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Fri, 15 Nov 2019 23:54:11 -0500
Subject: [PATCH 041/505] [camembert] tokenizer: use additional_special_tokens

---
 transformers/tokenization_camembert.py | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/transformers/tokenization_camembert.py b/transformers/tokenization_camembert.py
index ae1b322941..41d3d74cff 100644
--- a/transformers/tokenization_camembert.py
+++ b/transformers/tokenization_camembert.py
@@ -45,10 +45,12 @@ class CamembertTokenizer(PreTrainedTokenizer):
     max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
 
     def __init__(self, vocab_file, bos_token="<s>", eos_token="</s>", sep_token="</s>",
-                 cls_token="<s>", unk_token="<unk>", pad_token='<pad>', mask_token='<mask>', **kwargs):
+                 cls_token="<s>", unk_token="<unk>", pad_token='<pad>', mask_token='<mask>',
+                 additional_special_tokens=['<s>NOTUSED', '<s>NOTUSED'], **kwargs):
         super(CamembertTokenizer, self).__init__(max_len=512, bos_token=bos_token, eos_token=eos_token, unk_token=unk_token,
                                                  sep_token=sep_token, cls_token=cls_token, pad_token=pad_token,
-                                                 mask_token=mask_token, **kwargs)
+                                                 mask_token=mask_token, additional_special_tokens=additional_special_tokens,
+                                                 **kwargs)
         self.max_len_single_sentence = self.max_len - 2  # take into account special tokens
         self.max_len_sentences_pair = self.max_len - 4  # take into account special tokens
         self.sp_model = spm.SentencePieceProcessor()

From d08a338c3bbe9964a4d44e5bdb15a45e985256c0 Mon Sep 17 00:00:00 2001
From: Yohei Tamura <tamuhey@gmail.com>
Date: Sat, 16 Nov 2019 18:47:37 +0900
Subject: [PATCH 042/505] modified:   transformers/modeling_utils.py

---
 transformers/modeling_utils.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/modeling_utils.py b/transformers/modeling_utils.py
index d51eefab58..e7fd593bce 100644
--- a/transformers/modeling_utils.py
+++ b/transformers/modeling_utils.py
@@ -728,7 +728,7 @@ class SequenceSummary(nn.Module):
     def __init__(self, config):
         super(SequenceSummary, self).__init__()
 
-        self.summary_type = config.summary_type if hasattr(config, 'summary_use_proj') else 'last'
+        self.summary_type = config.summary_type if hasattr(config, 'summary_type') else 'last'
         if self.summary_type == 'attn':
             # We should use a standard multi-head attention module with absolute positional embedding for that.
             # Cf. https://github.com/zihangdai/xlnet/blob/master/modeling.py#L253-L276

From d32ce2c8df7053c19061b709465cdcc765e45a15 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan.schweter@bsb-muenchen.de>
Date: Mon, 18 Nov 2019 14:14:19 +0100
Subject: [PATCH 043/505] camembert: add wrapper for
 CamembertForTokenClassification

---
 transformers/modeling_camembert.py | 38 +++++++++++++++++++++++++++++-
 1 file changed, 37 insertions(+), 1 deletion(-)

diff --git a/transformers/modeling_camembert.py b/transformers/modeling_camembert.py
index 05538926e2..f302346f2d 100644
--- a/transformers/modeling_camembert.py
+++ b/transformers/modeling_camembert.py
@@ -20,7 +20,7 @@ from __future__ import (absolute_import, division, print_function,
 
 import logging
 
-from .modeling_roberta import RobertaModel, RobertaForMaskedLM, RobertaForSequenceClassification, RobertaForMultipleChoice
+from .modeling_roberta import RobertaModel, RobertaForMaskedLM, RobertaForSequenceClassification, RobertaForMultipleChoice, RobertaForTokenClassification
 from .configuration_camembert import CamembertConfig
 from .file_utils import add_start_docstrings
 
@@ -255,3 +255,39 @@ class CamembertForMultipleChoice(RobertaForMultipleChoice):
     """
     config_class = CamembertConfig
     pretrained_model_archive_map = CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP
+
+
+@add_start_docstrings("""CamemBERT Model with a token classification head on top (a linear layer on top of
+    the hidden-states output) e.g. for Named-Entity-Recognition (NER) tasks. """,
+    CAMEMBERT_START_DOCSTRING, CAMEMBERT_INPUTS_DOCSTRING)
+class CamembertForTokenClassification(RobertaForTokenClassification):
+    r"""
+        **labels**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
+            Labels for computing the token classification loss.
+            Indices should be in ``[0, ..., config.num_labels - 1]``.
+
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **loss**: (`optional`, returned when ``labels`` is provided) ``torch.FloatTensor`` of shape ``(1,)``:
+            Classification loss.
+        **scores**: ``torch.FloatTensor`` of shape ``(batch_size, sequence_length, config.num_labels)``
+            Classification scores (before SoftMax).
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``torch.FloatTensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        tokenizer = CamembertTokenizer.from_pretrained('camembert-base')
+        model = CamembertForTokenClassification.from_pretrained('camembert-base')
+        input_ids = torch.tensor(tokenizer.encode("J'aime le camembert !", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
+        labels = torch.tensor([1] * input_ids.size(1)).unsqueeze(0)  # Batch size 1
+        outputs = model(input_ids, labels=labels)
+        loss, scores = outputs[:2]
+
+    """
+    config_class = CamembertConfig
+    pretrained_model_archive_map = CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP

From 33753d9139307d9635db0309b6ddb9c53192c60a Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan.schweter@bsb-muenchen.de>
Date: Mon, 18 Nov 2019 14:14:54 +0100
Subject: [PATCH 044/505] module: import CamembertForTokenClassification

---
 transformers/__init__.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/transformers/__init__.py b/transformers/__init__.py
index cdf0669b39..5c7b0a6197 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -100,6 +100,7 @@ if is_torch_available():
                                 DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
     from .modeling_camembert import (CamembertForMaskedLM, CamembertModel,
                                 CamembertForSequenceClassification, CamembertForMultipleChoice,
+                                CamembertForTokenClassification,
                                 CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
     from .modeling_encoder_decoder import PreTrainedEncoderDecoder, Model2Model
 

From 44455eb5b61b2fc4a0bdbea6775d328c45a597e5 Mon Sep 17 00:00:00 2001
From: Sebastian Stabinger <sebastian@stabinger.name>
Date: Mon, 18 Nov 2019 10:07:05 +0100
Subject: [PATCH 045/505] Adds CamemBERT to Model architectures list

---
 README.md | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/README.md b/README.md
index 40d6acb76e..d1b37dbd48 100644
--- a/README.md
+++ b/README.md
@@ -122,7 +122,8 @@ At some point in the future, you'll be able to seamlessly move from pre-training
 7. **[RoBERTa](https://github.com/pytorch/fairseq/tree/master/examples/roberta)** (from Facebook), released together with the paper a [Robustly Optimized BERT Pretraining Approach](https://arxiv.org/abs/1907.11692) by Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, Veselin Stoyanov.
 8. **[DistilBERT](https://github.com/huggingface/transformers/tree/master/examples/distillation)** (from HuggingFace), released together with the paper [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://arxiv.org/abs/1910.01108) by Victor Sanh, Lysandre Debut and Thomas Wolf. The same method has been applied to compress GPT2 into [DistilGPT2](https://github.com/huggingface/transformers/tree/master/examples/distillation).
 9. **[CTRL](https://github.com/salesforce/ctrl/)** (from Salesforce) released with the paper [CTRL: A Conditional Transformer Language Model for Controllable Generation](https://arxiv.org/abs/1909.05858) by Nitish Shirish Keskar*, Bryan McCann*, Lav R. Varshney, Caiming Xiong and Richard Socher.
-10. Want to contribute a new model? We have added a **detailed guide and templates** to guide you in the process of adding a new model. You can find them in the [`templates`](./templates) folder of the repository. Be sure to check the [contributing guidelines](./CONTRIBUTING.md) and contact the maintainers or open an issue to collect feedbacks before starting your PR.
+10. **[CamemBERT](https://camembert-model.fr)** (from Facebook, Inria, Sorbonne) French language model based on RoBERTa released together with the paper [CamemBERT: a Tasty French Language Model](https://arxiv.org/abs/1911.03894) by Martin, Muller, and Suarez et al. 
+11. Want to contribute a new model? We have added a **detailed guide and templates** to guide you in the process of adding a new model. You can find them in the [`templates`](./templates) folder of the repository. Be sure to check the [contributing guidelines](./CONTRIBUTING.md) and contact the maintainers or open an issue to collect feedbacks before starting your PR.
 
 These implementations have been tested on several datasets (see the example scripts) and should match the performances of the original implementations (e.g. ~93 F1 on SQuAD for BERT Whole-Word-Masking, ~88 F1 on RocStories for OpenAI GPT, ~18.3 perplexity on WikiText 103 for Transformer-XL, ~0.916 Peason R coefficient on STS-B for XLNet). You can find more details on the performances in the Examples section of the [documentation](https://huggingface.co/transformers/examples.html).
 

From 3916b334a86484af8442d1cfdb2f15695feae581 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Mon, 18 Nov 2019 09:29:11 -0500
Subject: [PATCH 046/505] [camembert] Acknowledge the full author list

---
 README.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/README.md b/README.md
index d1b37dbd48..a49e8086d1 100644
--- a/README.md
+++ b/README.md
@@ -122,7 +122,7 @@ At some point in the future, you'll be able to seamlessly move from pre-training
 7. **[RoBERTa](https://github.com/pytorch/fairseq/tree/master/examples/roberta)** (from Facebook), released together with the paper a [Robustly Optimized BERT Pretraining Approach](https://arxiv.org/abs/1907.11692) by Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, Veselin Stoyanov.
 8. **[DistilBERT](https://github.com/huggingface/transformers/tree/master/examples/distillation)** (from HuggingFace), released together with the paper [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://arxiv.org/abs/1910.01108) by Victor Sanh, Lysandre Debut and Thomas Wolf. The same method has been applied to compress GPT2 into [DistilGPT2](https://github.com/huggingface/transformers/tree/master/examples/distillation).
 9. **[CTRL](https://github.com/salesforce/ctrl/)** (from Salesforce) released with the paper [CTRL: A Conditional Transformer Language Model for Controllable Generation](https://arxiv.org/abs/1909.05858) by Nitish Shirish Keskar*, Bryan McCann*, Lav R. Varshney, Caiming Xiong and Richard Socher.
-10. **[CamemBERT](https://camembert-model.fr)** (from Facebook, Inria, Sorbonne) French language model based on RoBERTa released together with the paper [CamemBERT: a Tasty French Language Model](https://arxiv.org/abs/1911.03894) by Martin, Muller, and Suarez et al. 
+10. **[CamemBERT](https://camembert-model.fr)** (from Inria/Facebook/Sorbonne) released with the paper [CamemBERT: a Tasty French Language Model](https://arxiv.org/abs/1911.03894) by Louis Martin*, Benjamin Muller*, Pedro Javier Ortiz Suárez*, Yoann Dupont, Laurent Romary, Éric Villemonte de la Clergerie, Djamé Seddah and Benoît Sagot.
 11. Want to contribute a new model? We have added a **detailed guide and templates** to guide you in the process of adding a new model. You can find them in the [`templates`](./templates) folder of the repository. Be sure to check the [contributing guidelines](./CONTRIBUTING.md) and contact the maintainers or open an issue to collect feedbacks before starting your PR.
 
 These implementations have been tested on several datasets (see the example scripts) and should match the performances of the original implementations (e.g. ~93 F1 on SQuAD for BERT Whole-Word-Masking, ~88 F1 on RocStories for OpenAI GPT, ~18.3 perplexity on WikiText 103 for Transformer-XL, ~0.916 Peason R coefficient on STS-B for XLNet). You can find more details on the performances in the Examples section of the [documentation](https://huggingface.co/transformers/examples.html).

From 0b3d45eb64607158977f546d57f90eae268c7836 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan.schweter@bsb-muenchen.de>
Date: Mon, 18 Nov 2019 15:49:44 +0100
Subject: [PATCH 047/505] camembert: add implementation for save_vocabulary
 method

---
 transformers/tokenization_camembert.py | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/transformers/tokenization_camembert.py b/transformers/tokenization_camembert.py
index 41d3d74cff..bf2a6fe993 100644
--- a/transformers/tokenization_camembert.py
+++ b/transformers/tokenization_camembert.py
@@ -16,9 +16,14 @@
 from __future__ import (absolute_import, division, print_function,
                         unicode_literals)
 
+import logging
+import os
+from shutil import copyfile
+
 import sentencepiece as spm
 from transformers.tokenization_utils import PreTrainedTokenizer
 
+logger = logging.getLogger(__name__)
 
 VOCAB_FILES_NAMES = {'vocab_file': 'sentencepiece.bpe.model'}
 
@@ -55,6 +60,7 @@ class CamembertTokenizer(PreTrainedTokenizer):
         self.max_len_sentences_pair = self.max_len - 4  # take into account special tokens
         self.sp_model = spm.SentencePieceProcessor()
         self.sp_model.Load(str(vocab_file))
+        self.vocab_file = vocab_file
         # HACK: These tokens were added by fairseq but don't seem to be actually used when duplicated in the actual
         # sentencepiece vocabulary (this is the case for <s> and </s>
         self.fairseq_tokens_to_ids = {'<s>NOTUSED': 0, '<pad>': 1, '</s>NOTUSED': 2, '<unk>': 3}
@@ -135,3 +141,17 @@ class CamembertTokenizer(PreTrainedTokenizer):
         if index in self.fairseq_ids_to_tokens:
             return self.fairseq_ids_to_tokens[index]
         return self.sp_model.IdToPiece(index - self.fairseq_offset)
+
+    def save_vocabulary(self, save_directory):
+        """ Save the sentencepiece vocabulary (copy original file) and special tokens file
+            to a directory.
+        """
+        if not os.path.isdir(save_directory):
+            logger.error("Vocabulary path ({}) should be a directory".format(save_directory))
+            return
+        out_vocab_file = os.path.join(save_directory, VOCAB_FILES_NAMES['vocab_file'])
+
+        if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file):
+            copyfile(self.vocab_file, out_vocab_file)
+
+        return (out_vocab_file,)

From 56c84863a1a20dfb82b928c5c9f77c21d9def8c7 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan.schweter@bsb-muenchen.de>
Date: Mon, 18 Nov 2019 15:50:16 +0100
Subject: [PATCH 048/505] camembert: add support for CamemBERT in run_ner
 example

---
 examples/run_ner.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/examples/run_ner.py b/examples/run_ner.py
index 4359e587ae..127d63a6cd 100644
--- a/examples/run_ner.py
+++ b/examples/run_ner.py
@@ -37,6 +37,7 @@ from transformers import AdamW, get_linear_schedule_with_warmup
 from transformers import WEIGHTS_NAME, BertConfig, BertForTokenClassification, BertTokenizer
 from transformers import RobertaConfig, RobertaForTokenClassification, RobertaTokenizer
 from transformers import DistilBertConfig, DistilBertForTokenClassification, DistilBertTokenizer
+from transformers import CamembertConfig, CamembertForTokenClassification, CamembertTokenizer
 
 logger = logging.getLogger(__name__)
 
@@ -47,7 +48,8 @@ ALL_MODELS = sum(
 MODEL_CLASSES = {
     "bert": (BertConfig, BertForTokenClassification, BertTokenizer),
     "roberta": (RobertaConfig, RobertaForTokenClassification, RobertaTokenizer),
-    "distilbert": (DistilBertConfig, DistilBertForTokenClassification, DistilBertTokenizer)
+    "distilbert": (DistilBertConfig, DistilBertForTokenClassification, DistilBertTokenizer),
+    "camembert": (CamembertConfig, CamembertForTokenClassification, CamembertTokenizer),
 }
 
 

From f3386d938348628c91457fc7d8650c223317a053 Mon Sep 17 00:00:00 2001
From: Kazutoshi Shinoda <kazutoshi.shinoda0516@gmail.com>
Date: Sun, 17 Nov 2019 18:08:51 +0900
Subject: [PATCH 049/505] typo "deay" -> "decay"

---
 examples/run_squad.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/examples/run_squad.py b/examples/run_squad.py
index d7fdc32ae7..69088d73c3 100644
--- a/examples/run_squad.py
+++ b/examples/run_squad.py
@@ -402,7 +402,7 @@ def main():
     parser.add_argument('--gradient_accumulation_steps', type=int, default=1,
                         help="Number of updates steps to accumulate before performing a backward/update pass.")
     parser.add_argument("--weight_decay", default=0.0, type=float,
-                        help="Weight deay if we apply some.")
+                        help="Weight decay if we apply some.")
     parser.add_argument("--adam_epsilon", default=1e-8, type=float,
                         help="Epsilon for Adam optimizer.")
     parser.add_argument("--max_grad_norm", default=1.0, type=float,

From 4193aa9f813c7868117214c1261b01c79fb9420f Mon Sep 17 00:00:00 2001
From: alexzubiaga <alexzubiga>
Date: Tue, 19 Nov 2019 10:32:08 +0100
Subject: [PATCH 050/505] add TFXLNetForTokenClassification implementation and
 unit test

add XLNetForTokenClassification implementation and unit tests
---
 transformers/__init__.py                     |   8 +-
 transformers/modeling_tf_xlnet.py            |  53 ++++++++++
 transformers/modeling_xlnet.py               | 100 +++++++++++++++++++
 transformers/tests/modeling_tf_xlnet_test.py |  26 +++++
 transformers/tests/modeling_xlnet_test.py    |  56 +++++++++--
 5 files changed, 232 insertions(+), 11 deletions(-)

diff --git a/transformers/__init__.py b/transformers/__init__.py
index cdf0669b39..ecfb8aeff2 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -83,9 +83,10 @@ if is_torch_available():
                                 CTRLLMHeadModel,
                                 CTRL_PRETRAINED_MODEL_ARCHIVE_MAP)
     from .modeling_xlnet import (XLNetPreTrainedModel, XLNetModel, XLNetLMHeadModel,
-                                XLNetForSequenceClassification, XLNetForMultipleChoice,
-                                XLNetForQuestionAnsweringSimple, XLNetForQuestionAnswering,
-                                load_tf_weights_in_xlnet, XLNET_PRETRAINED_MODEL_ARCHIVE_MAP)
+                                XLNetForSequenceClassification, XLNetForTokenClassification,
+                                XLNetForMultipleChoice, XLNetForQuestionAnsweringSimple,
+                                XLNetForQuestionAnswering, load_tf_weights_in_xlnet,
+                                XLNET_PRETRAINED_MODEL_ARCHIVE_MAP)
     from .modeling_xlm import (XLMPreTrainedModel , XLMModel,
                             XLMWithLMHeadModel, XLMForSequenceClassification,
                             XLMForQuestionAnswering, XLMForQuestionAnsweringSimple,
@@ -136,6 +137,7 @@ if is_tf_available():
     from .modeling_tf_xlnet import (TFXLNetPreTrainedModel, TFXLNetMainLayer,
                                     TFXLNetModel, TFXLNetLMHeadModel,
                                     TFXLNetForSequenceClassification,
+                                    TFXLNetForTokenClassification,
                                     TFXLNetForQuestionAnsweringSimple,
                                     TF_XLNET_PRETRAINED_MODEL_ARCHIVE_MAP)
 
diff --git a/transformers/modeling_tf_xlnet.py b/transformers/modeling_tf_xlnet.py
index 4733ea8589..b5e824e924 100644
--- a/transformers/modeling_tf_xlnet.py
+++ b/transformers/modeling_tf_xlnet.py
@@ -939,6 +939,59 @@ class TFXLNetForSequenceClassification(TFXLNetPreTrainedModel):
         return outputs  # return logits, (mems), (hidden states), (attentions)
 
 
+@add_start_docstrings("""XLNet Model with a token classification head on top (a linear layer on top of
+    the hidden-states output) e.g. for Named-Entity-Recognition (NER) tasks. """,
+    XLNET_START_DOCSTRING, XLNET_INPUTS_DOCSTRING)
+class TFXLNetForTokenClassification(TFXLNetPreTrainedModel):
+    r"""
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **scores**: ``tf.Tensor`` of shape ``(batch_size, sequence_length, config.num_labels)``
+            Classification scores (before SoftMax).
+        **mems**: (`optional`, returned when ``config.mem_len > 0``)
+            list of ``tf.Tensor`` (one for each layer):
+            that contains pre-computed hidden-states (key and values in the attention blocks) as computed by the model
+            if config.mem_len > 0 else tuple of None. Can be used to speed up sequential decoding and attend to longer context.
+            See details in the docstring of the `mems` input above.
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``tf.Tensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``tf.Tensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        import tensorflow as tf
+        from transformers import XLNetTokenizer, TFXLNetForTokenClassification
+
+        tokenizer = XLNetTokenizer.from_pretrained('xlnet-large-cased')
+        model = TFXLNetForSequenceClassification.from_pretrained('xlnet-large-cased')
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        outputs = model(input_ids)
+        scores = outputs[0]
+
+    """
+    def __init__(self, config, *inputs, **kwargs):
+        super(TFXLNetForTokenClassification, self).__init__(config, *inputs, **kwargs)
+        self.num_labels = config.num_labels
+
+        self.transformer = TFXLNetMainLayer(config, name='transformer')
+        self.classifier = tf.keras.layers.Dense(config.num_labels,
+                                                kernel_initializer=get_initializer(config.initializer_range),
+                                                name='classifier')
+
+    def call(self, inputs, **kwargs):
+        transformer_outputs = self.transformer(inputs, **kwargs)
+        output = transformer_outputs[0]
+
+        logits = self.classifier(output)
+
+        outputs = (logits,) + transformer_outputs[1:]  # Keep mems, hidden states, attentions if there are in it
+
+        return outputs  # return logits, (mems), (hidden states), (attentions)
+
+
 # @add_start_docstrings("""XLNet Model with a span classification head on top for extractive question-answering tasks like SQuAD (a linear layers on top of
 #     the hidden-states output to compute `span start logits` and `span end logits`). """,
 #     XLNET_START_DOCSTRING, XLNET_INPUTS_DOCSTRING)
diff --git a/transformers/modeling_xlnet.py b/transformers/modeling_xlnet.py
index 658048a660..2f4f883905 100644
--- a/transformers/modeling_xlnet.py
+++ b/transformers/modeling_xlnet.py
@@ -1046,6 +1046,106 @@ class XLNetForSequenceClassification(XLNetPreTrainedModel):
 
         return outputs  # return (loss), logits, (mems), (hidden states), (attentions)
 
+@add_start_docstrings("""XLNet Model with a token classification head on top (a linear layer on top of
+                      the hidden-states output) e.g. for Named-Entity-Recognition (NER) tasks. """,
+                      XLNET_START_DOCSTRING,
+                      XLNET_INPUTS_DOCSTRING)
+class XLNetForTokenClassification(XLNetPreTrainedModel):
+    r"""
+    Inputs:
+        **input_ids**: ``torch.LongTensor`` of shape ``(batch_size, num_choices, sequence_length)``:
+            Indices of input sequence tokens in the vocabulary.
+            The second dimension of the input (`num_choices`) indicates the number of choices to scores.
+        **token_type_ids**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
+            Segment token indices to indicate first and second portions of the inputs.
+            Indices are selected in ``[0, 1]``: ``0`` corresponds to a `sentence A` token, ``1``
+        **attention_mask**: (`optional`) ``torch.FloatTensor`` of shape ``(batch_size, sequence_length)``:
+            Mask to avoid performing attention on padding token indices.
+            Mask values selected in ``[0, 1]``:
+            ``1`` for tokens that are NOT MASKED, ``0`` for MASKED tokens.
+        **head_mask**: (`optional`) ``torch.FloatTensor`` of shape ``(num_heads,)`` or ``(num_layers, num_heads)``:
+            Mask to nullify selected heads of the self-attention modules.
+            Mask values selected in ``[0, 1]``:
+            ``1`` indicates the head is **not masked**, ``0`` indicates the head is **masked**.
+        **inputs_embeds**: (`optional`) ``torch.FloatTensor`` of shape ``(batch_size, sequence_length, embedding_dim)``:
+            Optionally, instead of passing ``input_ids`` you can choose to directly pass an embedded representation.
+            This is useful if you want more control over how to convert `input_ids` indices into associated vectors
+            than the model's internal embedding lookup matrix.
+        **labels**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size,)``:
+            Labels for computing the multiple choice classification loss.
+            Indices should be in ``[0, ..., num_choices]`` where `num_choices` is the size of the second dimension
+            of the input tensors. (see `input_ids` above)
+
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **loss**: (`optional`, returned when ``labels`` is provided) ``torch.FloatTensor`` of shape ``(1,)``:
+            Classification loss.
+        **scores**: ``torch.FloatTensor`` of shape ``(batch_size, sequence_length, config.num_labels)``
+            Classification scores (before SoftMax).
+        **mems**: (`optional`, returned when ``config.mem_len > 0``)
+            list of ``torch.FloatTensor`` (one for each layer):
+            that contains pre-computed hidden-states (key and values in the attention blocks) as computed by the model
+            if config.mem_len > 0 else tuple of None. Can be used to speed up sequential decoding and attend to longer context.
+            See details in the docstring of the `mems` input above.
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``torch.FloatTensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        tokenizer = XLNetTokenizer.from_pretrained('xlnet-large-cased')
+        model = XLNetForSequenceClassification.from_pretrained('xlnet-large-cased')
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        labels = torch.tensor([1] * input_ids.size(1)).unsqueeze(0)  # Batch size 1
+        outputs = model(input_ids, labels=labels)
+        scores = outputs[0]
+
+    """
+    def __init__(self, config):
+        super(XLNetForTokenClassification, self).__init__(config)
+        self.num_labels = config.num_labels
+
+        self.transformer = XLNetModel(config)
+        self.classifier = nn.Linear(config.hidden_size, config.num_labels)
+
+        self.init_weights()
+
+    def forward(self, input_ids=None, attention_mask=None, mems=None, perm_mask=None, target_mapping=None,
+                token_type_ids=None, input_mask=None, head_mask=None, inputs_embeds=None, labels=None):
+
+        outputs = self.transformer(input_ids,
+                            attention_mask=attention_mask,
+                            mems=mems,
+                            perm_mask=perm_mask,
+                            target_mapping=target_mapping,
+                            token_type_ids=token_type_ids,
+                            input_mask=input_mask,
+                            head_mask=head_mask,
+                            inputs_embeds=inputs_embeds)
+
+        sequence_output = outputs[0]
+
+        logits = self.classifier(sequence_output)
+
+        outputs = (logits,) + outputs[1:]  # Keep mems, hidden states, attentions if there are in it
+        if labels is not None:
+            loss_fct = CrossEntropyLoss()
+            # Only keep active parts of the loss
+            if attention_mask is not None:
+                active_loss = attention_mask.view(-1) == 1
+                active_logits = logits.view(-1, self.num_labels)[active_loss]
+                active_labels = labels.view(-1)[active_loss]
+                loss = loss_fct(active_logits, active_labels)
+            else:
+                loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
+            outputs = (loss,) + outputs
+
+        return outputs  # return (loss), logits, (mems), (hidden states), (attentions)
+
+
 @add_start_docstrings("""XLNet Model with a multiple choice classification head on top (a linear layer on top of
     the pooled output and a softmax) e.g. for RACE/SWAG tasks. """,
     XLNET_START_DOCSTRING, XLNET_INPUTS_DOCSTRING)
diff --git a/transformers/tests/modeling_tf_xlnet_test.py b/transformers/tests/modeling_tf_xlnet_test.py
index 12a8fbe36f..a00a965570 100644
--- a/transformers/tests/modeling_tf_xlnet_test.py
+++ b/transformers/tests/modeling_tf_xlnet_test.py
@@ -30,6 +30,7 @@ if is_tf_available():
 
     from transformers.modeling_tf_xlnet import (TFXLNetModel, TFXLNetLMHeadModel,
                                                         TFXLNetForSequenceClassification,
+                                                        TFXLNetForTokenClassification,
                                                         TFXLNetForQuestionAnsweringSimple,
                                                         TF_XLNET_PRETRAINED_MODEL_ARCHIVE_MAP)
 else:
@@ -42,6 +43,7 @@ class TFXLNetModelTest(TFCommonTestCases.TFCommonModelTester):
 
     all_model_classes=(TFXLNetModel, TFXLNetLMHeadModel,
                        TFXLNetForSequenceClassification,
+                       TFXLNetForTokenClassification,
                        TFXLNetForQuestionAnsweringSimple) if is_tf_available() else ()
     test_pruning = False
 
@@ -258,6 +260,26 @@ class TFXLNetModelTest(TFCommonTestCases.TFCommonModelTester):
                 list(list(mem.shape) for mem in result["mems_1"]),
                 [[self.seq_length, self.batch_size, self.hidden_size]] * self.num_hidden_layers)
 
+        def create_and_check_xlnet_for_token_classification(self, config, input_ids_1, input_ids_2, input_ids_q, perm_mask, input_mask,
+                target_mapping, segment_ids, lm_labels, sequence_labels, is_impossible_labels):
+            config.num_labels = input_ids_1.shape[1]
+            model = TFXLNetForTokenClassification(config)
+            inputs = {'input_ids': input_ids_1,
+                      'attention_mask': input_mask,
+                      # 'token_type_ids': token_type_ids
+                      }
+            logits, mems_1 = model(inputs)
+            result = {
+                "mems_1": [mem.numpy() for mem in mems_1],
+                "logits": logits.numpy(),
+            }
+            self.parent.assertListEqual(
+                list(result["logits"].shape),
+                [self.batch_size, self.seq_length, config.num_labels])
+            self.parent.assertListEqual(
+                list(list(mem.shape) for mem in result["mems_1"]),
+                [[self.seq_length, self.batch_size, self.hidden_size]] * self.num_hidden_layers)
+
         def prepare_config_and_inputs_for_common(self):
             config_and_inputs = self.prepare_config_and_inputs()
             (config, input_ids_1, input_ids_2, input_ids_q, perm_mask, input_mask,
@@ -289,6 +311,10 @@ class TFXLNetModelTest(TFCommonTestCases.TFCommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_xlnet_sequence_classif(*config_and_inputs)
 
+    def test_xlnet_token_classification(self):
+        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+        self.model_tester.create_and_check_xlnet_for_token_classification(*config_and_inputs)
+
     def test_xlnet_qa(self):
         self.model_tester.set_seed()
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
diff --git a/transformers/tests/modeling_xlnet_test.py b/transformers/tests/modeling_xlnet_test.py
index d97ea6a425..8f35d34e14 100644
--- a/transformers/tests/modeling_xlnet_test.py
+++ b/transformers/tests/modeling_xlnet_test.py
@@ -28,7 +28,8 @@ from transformers import is_torch_available
 if is_torch_available():
     import torch
 
-    from transformers import (XLNetConfig, XLNetModel, XLNetLMHeadModel, XLNetForSequenceClassification, XLNetForQuestionAnswering)
+    from transformers import (XLNetConfig, XLNetModel, XLNetLMHeadModel, XLNetForSequenceClassification,
+                              XLNetForTokenClassification, XLNetForQuestionAnswering)
     from transformers.modeling_xlnet import XLNET_PRETRAINED_MODEL_ARCHIVE_MAP
 else:
     pytestmark = pytest.mark.skip("Require Torch")
@@ -38,7 +39,7 @@ from .configuration_common_test import ConfigTester
 
 class XLNetModelTest(CommonTestCases.CommonModelTester):
 
-    all_model_classes=(XLNetModel, XLNetLMHeadModel,
+    all_model_classes=(XLNetModel, XLNetLMHeadModel, XLNetForTokenClassification,
                     XLNetForSequenceClassification, XLNetForQuestionAnswering) if is_torch_available() else ()
     test_pruning = False
 
@@ -107,10 +108,12 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
             sequence_labels = None
             lm_labels = None
             is_impossible_labels = None
+            token_labels = None
             if self.use_labels:
                 lm_labels = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)
                 sequence_labels = ids_tensor([self.batch_size], self.type_sequence_label_size)
                 is_impossible_labels = ids_tensor([self.batch_size], 2).float()
+                token_labels = ids_tensor([self.batch_size, self.seq_length], self.type_vocab_size)
 
             config = XLNetConfig(
                 vocab_size_or_config_json_file=self.vocab_size,
@@ -129,14 +132,14 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
                 num_labels=self.type_sequence_label_size)
 
             return (config, input_ids_1, input_ids_2, input_ids_q, perm_mask, input_mask,
-                    target_mapping, segment_ids, lm_labels, sequence_labels, is_impossible_labels)
+                    target_mapping, segment_ids, lm_labels, sequence_labels, is_impossible_labels, token_labels)
 
         def set_seed(self):
             random.seed(self.seed)
             torch.manual_seed(self.seed)
 
         def create_and_check_xlnet_base_model(self, config, input_ids_1, input_ids_2, input_ids_q, perm_mask, input_mask,
-                target_mapping, segment_ids, lm_labels, sequence_labels, is_impossible_labels):
+                target_mapping, segment_ids, lm_labels, sequence_labels, is_impossible_labels, token_labels):
             model = XLNetModel(config)
             model.eval()
 
@@ -164,7 +167,7 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
                 [[self.seq_length, self.batch_size, self.hidden_size]] * self.num_hidden_layers)
 
         def create_and_check_xlnet_lm_head(self, config, input_ids_1, input_ids_2, input_ids_q, perm_mask, input_mask,
-                target_mapping, segment_ids, lm_labels, sequence_labels, is_impossible_labels):
+                target_mapping, segment_ids, lm_labels, sequence_labels, is_impossible_labels, token_labels):
             model = XLNetLMHeadModel(config)
             model.eval()
 
@@ -204,7 +207,7 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
                 [[self.mem_len, self.batch_size, self.hidden_size]] * self.num_hidden_layers)
 
         def create_and_check_xlnet_qa(self, config, input_ids_1, input_ids_2, input_ids_q, perm_mask, input_mask,
-                target_mapping, segment_ids, lm_labels, sequence_labels, is_impossible_labels):
+                target_mapping, segment_ids, lm_labels, sequence_labels, is_impossible_labels, token_labels):
             model = XLNetForQuestionAnswering(config)
             model.eval()
 
@@ -261,8 +264,40 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
                 list(list(mem.size()) for mem in result["mems"]),
                 [[self.seq_length, self.batch_size, self.hidden_size]] * self.num_hidden_layers)
 
+        def create_and_check_xlnet_token_classif(self, config, input_ids_1, input_ids_2, input_ids_q, perm_mask, input_mask,
+                target_mapping, segment_ids, lm_labels, sequence_labels, is_impossible_labels, token_labels):
+            model = XLNetForTokenClassification(config)
+            model.eval()
+
+            logits, mems_1 = model(input_ids_1)
+            loss, logits, mems_1 = model(input_ids_1, labels=token_labels)
+
+            result = {
+                "loss": loss,
+                "mems_1": mems_1,
+                "logits": logits,
+            }
+
+            self.parent.assertListEqual(
+                list(result["loss"].size()),
+                [])
+            self.parent.assertListEqual(
+                list(result["logits"].size()),
+                [self.batch_size, self.seq_length, self.type_sequence_label_size])
+            self.parent.assertListEqual(
+                list(list(mem.size()) for mem in result["mems_1"]),
+                [[self.seq_length, self.batch_size, self.hidden_size]] * self.num_hidden_layers)
+
+        def prepare_config_and_inputs_for_common(self):
+            config_and_inputs = self.prepare_config_and_inputs()
+            (config, input_ids_1, input_ids_2, input_ids_q, perm_mask, input_mask,
+                target_mapping, segment_ids, lm_labels,
+                sequence_labels, is_impossible_labels) = config_and_inputs
+            inputs_dict = {'input_ids': input_ids_1}
+            return config, inputs_dict
+
         def create_and_check_xlnet_sequence_classif(self, config, input_ids_1, input_ids_2, input_ids_q, perm_mask, input_mask,
-                target_mapping, segment_ids, lm_labels, sequence_labels, is_impossible_labels):
+                target_mapping, segment_ids, lm_labels, sequence_labels, is_impossible_labels, token_labels):
             model = XLNetForSequenceClassification(config)
             model.eval()
 
@@ -289,7 +324,7 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
             config_and_inputs = self.prepare_config_and_inputs()
             (config, input_ids_1, input_ids_2, input_ids_q, perm_mask, input_mask,
                 target_mapping, segment_ids, lm_labels,
-                sequence_labels, is_impossible_labels) = config_and_inputs
+                sequence_labels, is_impossible_labels, token_labels) = config_and_inputs
             inputs_dict = {'input_ids': input_ids_1}
             return config, inputs_dict
 
@@ -316,6 +351,11 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_xlnet_sequence_classif(*config_and_inputs)
 
+    def test_xlnet_token_classif(self):
+        self.model_tester.set_seed()
+        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+        self.model_tester.create_and_check_xlnet_token_classif(*config_and_inputs)
+
     def test_xlnet_qa(self):
         self.model_tester.set_seed()
         config_and_inputs = self.model_tester.prepare_config_and_inputs()

From 337802783f0ba87a58e2e9adddb47ac5cb00646d Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Tue, 19 Nov 2019 19:50:32 +0100
Subject: [PATCH 051/505] distilbert: add configuration for new German
 distilbert model

---
 transformers/configuration_distilbert.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/transformers/configuration_distilbert.py b/transformers/configuration_distilbert.py
index 2a8a149acf..57b9e57fe0 100644
--- a/transformers/configuration_distilbert.py
+++ b/transformers/configuration_distilbert.py
@@ -27,7 +27,8 @@ logger = logging.getLogger(__name__)
 
 DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP = {
     'distilbert-base-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-config.json",
-    'distilbert-base-uncased-distilled-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-distilled-squad-config.json"
+    'distilbert-base-uncased-distilled-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-distilled-squad-config.json",
+    'distilbert-base-german-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-german-cased-config.json"
 }
 
 

From 22333945fbd1f5eff8cfb664ac6f63363e6b72d4 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Tue, 19 Nov 2019 19:51:01 +0100
Subject: [PATCH 052/505] distilbert: add pytorch model for new German
 distilbert model

---
 transformers/modeling_distilbert.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/transformers/modeling_distilbert.py b/transformers/modeling_distilbert.py
index d30f493c69..cb2fa1915e 100644
--- a/transformers/modeling_distilbert.py
+++ b/transformers/modeling_distilbert.py
@@ -42,7 +42,8 @@ logger = logging.getLogger(__name__)
 
 DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
     'distilbert-base-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-pytorch_model.bin",
-    'distilbert-base-uncased-distilled-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-distilled-squad-pytorch_model.bin"
+    'distilbert-base-uncased-distilled-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-distilled-squad-pytorch_model.bin",
+    'distilbert-base-german-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-german-cased-pytorch_model.bin"
 }
 
 

From f21dfe36baaf316675a4d2f8c918e9e8afc11db2 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Tue, 19 Nov 2019 19:51:31 +0100
Subject: [PATCH 053/505] distilbert: add vocab for new German distilbert model

---
 transformers/tokenization_distilbert.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/transformers/tokenization_distilbert.py b/transformers/tokenization_distilbert.py
index dfa02926d8..61f967bf0b 100644
--- a/transformers/tokenization_distilbert.py
+++ b/transformers/tokenization_distilbert.py
@@ -33,6 +33,7 @@ PRETRAINED_VOCAB_FILES_MAP = {
     {
         'distilbert-base-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-uncased-vocab.txt",
         'distilbert-base-uncased-distilled-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-uncased-vocab.txt",
+        'distilbert-base-german-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-german-cased-vocab.txt"
     }
 }
 

From e631383d4fb55ed2dff53f2b80a29e51cd2f563d Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Tue, 19 Nov 2019 19:52:40 +0100
Subject: [PATCH 054/505] docs: add new German distilbert model to pretrained
 models

---
 docs/source/pretrained_models.rst | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/docs/source/pretrained_models.rst b/docs/source/pretrained_models.rst
index b0a578fd80..3b25a79802 100644
--- a/docs/source/pretrained_models.rst
+++ b/docs/source/pretrained_models.rst
@@ -151,6 +151,10 @@ Here is the full list of the currently provided pretrained models together with
 |                   | ``distilroberta-base``                                     | | 6-layer, 768-hidden, 12-heads, 82M parameters                                                                                       |
 |                   |                                                            | | The DistilRoBERTa model distilled from the RoBERTa model `roberta-base` checkpoint.                                                 |
 |                   |                                                            | (see `details <https://github.com/huggingface/transformers/tree/master/examples/distillation>`__)                                     |
+|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   | ``distilbert-base-german-cased``                           | | 6-layer, 768-hidden, 12-heads, 66M parameters                                                                                       |
+|                   |                                                            | | The German DistilBERT model distilled from the German DBMDZ BERT model `bert-base-german-dbmdz-cased` checkpoint.                   |
+|                   |                                                            | (see `details <https://github.com/huggingface/transformers/tree/master/examples/distillation>`__)                                     |
 +-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 | CTRL              | ``ctrl``                                                   | | 48-layer, 1280-hidden, 16-heads, 1.6B parameters                                                                                    |
 |                   |                                                            | | Salesforce's Large-sized CTRL English model                                                                                         |

From e7cf2ccd1567615513013d5fc9f9002733f70e13 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Tue, 19 Nov 2019 19:55:19 +0100
Subject: [PATCH 055/505] distillation: add German distilbert model

---
 examples/distillation/README.md | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/examples/distillation/README.md b/examples/distillation/README.md
index 8efd1ea6f4..c2765c28ff 100644
--- a/examples/distillation/README.md
+++ b/examples/distillation/README.md
@@ -2,6 +2,8 @@
 
 This folder contains the original code used to train Distil* as well as examples showcasing how to use DistilBERT, DistilRoBERTa and DistilGPT2.
 
+**November 19th, 2019 - Update** We release German **DistilBERT**: 98.8% of `bert-base-german-dbmdz-cased` on NER tasks.
+
 **October 23rd, 2019 - Update** We release **DistilRoBERTa**: 95% of `RoBERTa-base`'s performance on GLUE, twice as fast as RoBERTa while being 35% smaller.
 
 **October 3rd, 2019 - Update** We release our [NeurIPS workshop paper](https://arxiv.org/abs/1910.01108) explaining our approach on **DistilBERT**. It includes updated results and further experiments. We applied the same method to GPT2 and release the weights of **DistilGPT2**. DistilGPT2 is two times faster and 33% smaller than GPT2. **The paper superseeds our [previous blogpost](https://medium.com/huggingface/distilbert-8cf3380435b5) with a different distillation loss and better performances. Please use the paper as a reference when comparing/reporting results on DistilBERT.**
@@ -45,10 +47,11 @@ This part of the library has only be tested with Python3.6+. There are few speci
 
 ## How to use DistilBERT
 
-Transformers includes two pre-trained Distil* models, currently only provided for English (we are investigating the possibility to train and release a multilingual version of DistilBERT):
+Transformers includes five pre-trained Distil* models, currently only provided for English and German (we are investigating the possibility to train and release a multilingual version of DistilBERT):
 
 - `distilbert-base-uncased`: DistilBERT English language model pretrained on the same data used to pretrain Bert (concatenation of the Toronto Book Corpus and full English Wikipedia) using distillation with the supervision of the `bert-base-uncased` version of Bert. The model has 6 layers, 768 dimension and 12 heads, totalizing 66M parameters.
 - `distilbert-base-uncased-distilled-squad`: A finetuned version of `distilbert-base-uncased` finetuned using (a second step of) knwoledge distillation on SQuAD 1.0. This model reaches a F1 score of 86.9 on the dev set (for comparison, Bert `bert-base-uncased` version reaches a 88.5 F1 score).
+- `distilbert-base-german-cased`: DistilBERT German language model pretrained on 1/2 of the data used to pretrain Bert using distillation with the supervision of the `bert-base-german-dbmdz-cased` version of German DBMDZ Bert. For NER tasks the model reaches a F1 score of 83.49 on the CoNLL-2003 test set (for comparison, `bert-base-german-dbmdz-cased` reaches a 84.52 F1 score), and a F1 score of 85.23 on the GermEval 2014 test set (`bert-base-german-dbmdz-cased` reaches a 86.89 F1 score).
 - `distilgpt2`: DistilGPT2 English language model pretrained with the supervision of `gpt2` (the smallest version of GPT2) on [OpenWebTextCorpus](https://skylion007.github.io/OpenWebTextCorpus/), a reproduction of OpenAI's WebText dataset. The model has 6 layers, 768 dimension and 12 heads, totalizing 82M parameters (compared to 124M parameters for GPT2). On average, DistilGPT2 is two times faster than GPT2.
 - `distilroberta-base`: DistilRoBERTa English language model pretrained with the supervision of `roberta-base` solely on [OpenWebTextCorpus](https://skylion007.github.io/OpenWebTextCorpus/), a reproduction of OpenAI's WebText dataset (it is ~4 times less training data than the teacher RoBERTa). The model has 6 layers, 768 dimension and 12 heads, totalizing 82M parameters (compared to 125M parameters for RoBERTa-base). On average DistilRoBERTa is twice as fast as Roberta-base.
 - and more to come! 🤗🤗🤗

From 2e2c0375c3e725483d2ec297632dc52ed6a9f5fb Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Tue, 19 Nov 2019 20:41:18 +0100
Subject: [PATCH 056/505] distilbert: add German distilbert model to positional
 embedding sizes map

---
 transformers/tokenization_distilbert.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/transformers/tokenization_distilbert.py b/transformers/tokenization_distilbert.py
index 61f967bf0b..c82ac09727 100644
--- a/transformers/tokenization_distilbert.py
+++ b/transformers/tokenization_distilbert.py
@@ -40,6 +40,7 @@ PRETRAINED_VOCAB_FILES_MAP = {
 PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
     'distilbert-base-uncased': 512,
     'distilbert-base-uncased-distilled-squad': 512,
+    'distilbert-base-german-cased': 512,
 }
 
 

From da06afafc87b03f7588b6bd319e1b7592b091339 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Tue, 19 Nov 2019 21:57:00 +0100
Subject: [PATCH 057/505] tree-wide: add trailing comma in configuration maps

---
 transformers/configuration_distilbert.py | 2 +-
 transformers/modeling_distilbert.py      | 2 +-
 transformers/tokenization_distilbert.py  | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/transformers/configuration_distilbert.py b/transformers/configuration_distilbert.py
index 57b9e57fe0..1c35f5c0dd 100644
--- a/transformers/configuration_distilbert.py
+++ b/transformers/configuration_distilbert.py
@@ -28,7 +28,7 @@ logger = logging.getLogger(__name__)
 DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP = {
     'distilbert-base-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-config.json",
     'distilbert-base-uncased-distilled-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-distilled-squad-config.json",
-    'distilbert-base-german-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-german-cased-config.json"
+    'distilbert-base-german-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-german-cased-config.json",
 }
 
 
diff --git a/transformers/modeling_distilbert.py b/transformers/modeling_distilbert.py
index cb2fa1915e..4ba75248a1 100644
--- a/transformers/modeling_distilbert.py
+++ b/transformers/modeling_distilbert.py
@@ -43,7 +43,7 @@ logger = logging.getLogger(__name__)
 DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
     'distilbert-base-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-pytorch_model.bin",
     'distilbert-base-uncased-distilled-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-distilled-squad-pytorch_model.bin",
-    'distilbert-base-german-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-german-cased-pytorch_model.bin"
+    'distilbert-base-german-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-german-cased-pytorch_model.bin",
 }
 
 
diff --git a/transformers/tokenization_distilbert.py b/transformers/tokenization_distilbert.py
index c82ac09727..6574e0bc4d 100644
--- a/transformers/tokenization_distilbert.py
+++ b/transformers/tokenization_distilbert.py
@@ -33,7 +33,7 @@ PRETRAINED_VOCAB_FILES_MAP = {
     {
         'distilbert-base-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-uncased-vocab.txt",
         'distilbert-base-uncased-distilled-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-uncased-vocab.txt",
-        'distilbert-base-german-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-german-cased-vocab.txt"
+        'distilbert-base-german-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-german-cased-vocab.txt",
     }
 }
 

From 3de31f8d287da44a40566fb1d5c44107708b87ea Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Tue, 19 Nov 2019 18:14:14 -0500
Subject: [PATCH 058/505] mean does not exist in TF2

---
 transformers/modeling_tf_utils.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/modeling_tf_utils.py b/transformers/modeling_tf_utils.py
index e08605d154..8be7eaaf67 100644
--- a/transformers/modeling_tf_utils.py
+++ b/transformers/modeling_tf_utils.py
@@ -454,7 +454,7 @@ class TFSequenceSummary(tf.keras.layers.Layer):
         elif self.summary_type == 'first':
             output = hidden_states[:, 0]
         elif self.summary_type == 'mean':
-            output = tf.mean(hidden_states, axis=1)
+            output = tf.reduce_mean(hidden_states, axis=1)
         elif self.summary_type == 'cls_index':
             hidden_shape = shape_list(hidden_states)  # e.g. [batch, num choices, seq length, hidden dims]
             if cls_index is None:

From 454455c695ff38df1ed3670a43677fdd1abcedf3 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Wed, 20 Nov 2019 09:42:48 -0500
Subject: [PATCH 059/505] fix #1879

---
 examples/utils_squad.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/examples/utils_squad.py b/examples/utils_squad.py
index c04dacf6d3..4f1c581588 100644
--- a/examples/utils_squad.py
+++ b/examples/utils_squad.py
@@ -240,6 +240,7 @@ def convert_examples_to_features(examples, tokenizer, max_seq_length,
 
         # The -3 accounts for [CLS], [SEP] and [SEP]
         max_tokens_for_doc = max_seq_length - len(query_tokens) - 3
+        assert max_tokens_for_doc > 0
 
         # We can have documents that are longer than the maximum sequence length.
         # To deal with this we do a sliding window approach, where we take chunks

From e70cdf083ddb8bfe298d43e6d70d698a3a2f56d3 Mon Sep 17 00:00:00 2001
From: Jin Young Sohn <jysohn@google.com>
Date: Wed, 20 Nov 2019 22:33:26 +0000
Subject: [PATCH 060/505] Cleanup TPU bits from run_glue.py

TPU runner is currently implemented in:
https://github.com/pytorch-tpu/transformers/blob/tpu/examples/run_glue_tpu.py.

We plan to upstream this directly into `huggingface/transformers`
(either `master` or `tpu`) branch once it's been more thoroughly tested.
---
 examples/run_glue.py | 35 ++---------------------------------
 1 file changed, 2 insertions(+), 33 deletions(-)

diff --git a/examples/run_glue.py b/examples/run_glue.py
index 19316cb0ec..527e440075 100644
--- a/examples/run_glue.py
+++ b/examples/run_glue.py
@@ -158,7 +158,7 @@ def train(args, train_dataset, model, tokenizer):
                 loss.backward()
 
             tr_loss += loss.item()
-            if (step + 1) % args.gradient_accumulation_steps == 0 and not args.tpu:
+            if (step + 1) % args.gradient_accumulation_steps == 0:
                 if args.fp16:
                     torch.nn.utils.clip_grad_norm_(amp.master_params(optimizer), args.max_grad_norm)
                 else:
@@ -189,11 +189,6 @@ def train(args, train_dataset, model, tokenizer):
                     torch.save(args, os.path.join(output_dir, 'training_args.bin'))
                     logger.info("Saving model checkpoint to %s", output_dir)
 
-            if args.tpu:
-                args.xla_model.optimizer_step(optimizer, barrier=True)
-                model.zero_grad()
-                global_step += 1
-
             if args.max_steps > 0 and global_step > args.max_steps:
                 epoch_iterator.close()
                 break
@@ -397,15 +392,6 @@ def main():
     parser.add_argument('--seed', type=int, default=42,
                         help="random seed for initialization")
 
-    parser.add_argument('--tpu', action='store_true',
-                        help="Whether to run on the TPU defined in the environment variables")
-    parser.add_argument('--tpu_ip_address', type=str, default='',
-                        help="TPU IP address if none are set in the environment variables")
-    parser.add_argument('--tpu_name', type=str, default='',
-                        help="TPU name if none are set in the environment variables")
-    parser.add_argument('--xrt_tpu_config', type=str, default='',
-                        help="XRT TPU config if none are set in the environment variables")
-
     parser.add_argument('--fp16', action='store_true',
                         help="Whether to use 16-bit (mixed) precision (through NVIDIA apex) instead of 32-bit")
     parser.add_argument('--fp16_opt_level', type=str, default='O1',
@@ -439,23 +425,6 @@ def main():
         args.n_gpu = 1
     args.device = device
 
-    if args.tpu:
-        if args.tpu_ip_address:
-            os.environ["TPU_IP_ADDRESS"] = args.tpu_ip_address
-        if args.tpu_name:
-            os.environ["TPU_NAME"] = args.tpu_name
-        if args.xrt_tpu_config:
-            os.environ["XRT_TPU_CONFIG"] = args.xrt_tpu_config
-
-        assert "TPU_IP_ADDRESS" in os.environ
-        assert "TPU_NAME" in os.environ
-        assert "XRT_TPU_CONFIG" in os.environ
-
-        import torch_xla
-        import torch_xla.core.xla_model as xm
-        args.device = xm.xla_device()
-        args.xla_model = xm
-
     # Setup logging
     logging.basicConfig(format = '%(asctime)s - %(levelname)s - %(name)s -   %(message)s',
                         datefmt = '%m/%d/%Y %H:%M:%S',
@@ -509,7 +478,7 @@ def main():
 
 
     # Saving best-practices: if you use defaults names for the model, you can reload it using from_pretrained()
-    if args.do_train and (args.local_rank == -1 or torch.distributed.get_rank() == 0) and not args.tpu:
+    if args.do_train and (args.local_rank == -1 or torch.distributed.get_rank() == 0):
         # Create output directory if needed
         if not os.path.exists(args.output_dir) and args.local_rank in [-1, 0]:
             os.makedirs(args.output_dir)

From 2cf3447e0a3e3fd04b715f1e6f4ee43575e1e7c9 Mon Sep 17 00:00:00 2001
From: Juha Kiili <juha@juhakiili.com>
Date: Thu, 21 Nov 2019 12:35:25 +0200
Subject: [PATCH 061/505] Glue: log in Valohai-compatible JSON format too

---
 examples/run_glue.py | 15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)

diff --git a/examples/run_glue.py b/examples/run_glue.py
index 527e440075..ea5ac5bbb7 100644
--- a/examples/run_glue.py
+++ b/examples/run_glue.py
@@ -22,6 +22,7 @@ import glob
 import logging
 import os
 import random
+import json
 
 import numpy as np
 import torch
@@ -171,13 +172,21 @@ def train(args, train_dataset, model, tokenizer):
 
                 if args.local_rank in [-1, 0] and args.logging_steps > 0 and global_step % args.logging_steps == 0:
                     # Log metrics
+                    logs = {'step': global_step}
                     if args.local_rank == -1 and args.evaluate_during_training:  # Only evaluate when single GPU otherwise metrics may not average well
                         results = evaluate(args, model, tokenizer)
                         for key, value in results.items():
-                            tb_writer.add_scalar('eval_{}'.format(key), value, global_step)
-                    tb_writer.add_scalar('lr', scheduler.get_lr()[0], global_step)
-                    tb_writer.add_scalar('loss', (tr_loss - logging_loss)/args.logging_steps, global_step)
+                            eval_key = 'eval_{}'.format(key)
+                            tb_writer.add_scalar(eval_key, value, global_step)
+                            logs[eval_key] = str(value)
                     logging_loss = tr_loss
+                    loss_scalar = (tr_loss - logging_loss) / args.logging_steps
+                    learning_rate_scalar = scheduler.get_lr()[0]
+                    tb_writer.add_scalar('lr', learning_rate_scalar, global_step)
+                    tb_writer.add_scalar('loss', loss_scalar, global_step)
+                    logs['learning_rate'] = learning_rate_scalar
+                    logs['loss'] = loss_scalar
+                    print(json.dumps(logs))
 
                 if args.local_rank in [-1, 0] and args.save_steps > 0 and global_step % args.save_steps == 0:
                     # Save model checkpoint

From aac35514075290a46419b9bd969e6f94fef9d43b Mon Sep 17 00:00:00 2001
From: Aarni Koskela <akx@iki.fi>
Date: Thu, 21 Nov 2019 12:37:39 +0200
Subject: [PATCH 062/505] Add download_glue_data.py from kamalkraj/ALBERT-TF2.0

Original source: https://github.com/kamalkraj/ALBERT-TF2.0/blob/fa90194e5fe729dbb19f32ac29c8d6d6372c0f93/download_glue_data.py
Original license: https://github.com/kamalkraj/ALBERT-TF2.0/blob/fa90194e5fe729dbb19f32ac29c8d6d6372c0f93/LICENSE (Apache-2.0)
---
 utils/download_glue_data.py | 141 ++++++++++++++++++++++++++++++++++++
 1 file changed, 141 insertions(+)
 create mode 100644 utils/download_glue_data.py

diff --git a/utils/download_glue_data.py b/utils/download_glue_data.py
new file mode 100644
index 0000000000..86a4e8951f
--- /dev/null
+++ b/utils/download_glue_data.py
@@ -0,0 +1,141 @@
+''' Script for downloading all GLUE data.
+
+Note: for legal reasons, we are unable to host MRPC.
+You can either use the version hosted by the SentEval team, which is already tokenized, 
+or you can download the original data from (https://download.microsoft.com/download/D/4/6/D46FF87A-F6B9-4252-AA8B-3604ED519838/MSRParaphraseCorpus.msi) and extract the data from it manually.
+For Windows users, you can run the .msi file. For Mac and Linux users, consider an external library such as 'cabextract' (see below for an example).
+You should then rename and place specific files in a folder (see below for an example).
+
+mkdir MRPC
+cabextract MSRParaphraseCorpus.msi -d MRPC
+cat MRPC/_2DEC3DBE877E4DB192D17C0256E90F1D | tr -d $'\r' > MRPC/msr_paraphrase_train.txt
+cat MRPC/_D7B391F9EAFF4B1B8BCE8F21B20B1B61 | tr -d $'\r' > MRPC/msr_paraphrase_test.txt
+rm MRPC/_*
+rm MSRParaphraseCorpus.msi
+
+1/30/19: It looks like SentEval is no longer hosting their extracted and tokenized MRPC data, so you'll need to download the data from the original source for now.
+2/11/19: It looks like SentEval actually *is* hosting the extracted data. Hooray!
+'''
+
+import os
+import sys
+import shutil
+import argparse
+import tempfile
+import urllib.request
+import zipfile
+
+TASKS = ["CoLA", "SST", "MRPC", "QQP", "STS", "MNLI", "SNLI", "QNLI", "RTE", "WNLI", "diagnostic"]
+TASK2PATH = {"CoLA":'https://firebasestorage.googleapis.com/v0/b/mtl-sentence-representations.appspot.com/o/data%2FCoLA.zip?alt=media&token=46d5e637-3411-4188-bc44-5809b5bfb5f4',
+             "SST":'https://firebasestorage.googleapis.com/v0/b/mtl-sentence-representations.appspot.com/o/data%2FSST-2.zip?alt=media&token=aabc5f6b-e466-44a2-b9b4-cf6337f84ac8',
+             "MRPC":'https://firebasestorage.googleapis.com/v0/b/mtl-sentence-representations.appspot.com/o/data%2Fmrpc_dev_ids.tsv?alt=media&token=ec5c0836-31d5-48f4-b431-7480817f1adc',
+             "QQP":'https://firebasestorage.googleapis.com/v0/b/mtl-sentence-representations.appspot.com/o/data%2FQQP.zip?alt=media&token=700c6acf-160d-4d89-81d1-de4191d02cb5',
+             "STS":'https://firebasestorage.googleapis.com/v0/b/mtl-sentence-representations.appspot.com/o/data%2FSTS-B.zip?alt=media&token=bddb94a7-8706-4e0d-a694-1109e12273b5',
+             "MNLI":'https://firebasestorage.googleapis.com/v0/b/mtl-sentence-representations.appspot.com/o/data%2FMNLI.zip?alt=media&token=50329ea1-e339-40e2-809c-10c40afff3ce',
+             "SNLI":'https://firebasestorage.googleapis.com/v0/b/mtl-sentence-representations.appspot.com/o/data%2FSNLI.zip?alt=media&token=4afcfbb2-ff0c-4b2d-a09a-dbf07926f4df',
+             "QNLI": 'https://firebasestorage.googleapis.com/v0/b/mtl-sentence-representations.appspot.com/o/data%2FQNLIv2.zip?alt=media&token=6fdcf570-0fc5-4631-8456-9505272d1601',
+             "RTE":'https://firebasestorage.googleapis.com/v0/b/mtl-sentence-representations.appspot.com/o/data%2FRTE.zip?alt=media&token=5efa7e85-a0bb-4f19-8ea2-9e1840f077fb',
+             "WNLI":'https://firebasestorage.googleapis.com/v0/b/mtl-sentence-representations.appspot.com/o/data%2FWNLI.zip?alt=media&token=068ad0a0-ded7-4bd7-99a5-5e00222e0faf',
+             "diagnostic":'https://storage.googleapis.com/mtl-sentence-representations.appspot.com/tsvsWithoutLabels%2FAX.tsv?GoogleAccessId=firebase-adminsdk-0khhl@mtl-sentence-representations.iam.gserviceaccount.com&Expires=2498860800&Signature=DuQ2CSPt2Yfre0C%2BiISrVYrIFaZH1Lc7hBVZDD4ZyR7fZYOMNOUGpi8QxBmTNOrNPjR3z1cggo7WXFfrgECP6FBJSsURv8Ybrue8Ypt%2FTPxbuJ0Xc2FhDi%2BarnecCBFO77RSbfuz%2Bs95hRrYhTnByqu3U%2FYZPaj3tZt5QdfpH2IUROY8LiBXoXS46LE%2FgOQc%2FKN%2BA9SoscRDYsnxHfG0IjXGwHN%2Bf88q6hOmAxeNPx6moDulUF6XMUAaXCSFU%2BnRO2RDL9CapWxj%2BDl7syNyHhB7987hZ80B%2FwFkQ3MEs8auvt5XW1%2Bd4aCU7ytgM69r8JDCwibfhZxpaa4gd50QXQ%3D%3D'}
+
+MRPC_TRAIN = 'https://dl.fbaipublicfiles.com/senteval/senteval_data/msr_paraphrase_train.txt'
+MRPC_TEST = 'https://dl.fbaipublicfiles.com/senteval/senteval_data/msr_paraphrase_test.txt'
+
+def download_and_extract(task, data_dir):
+    print("Downloading and extracting %s..." % task)
+    data_file = "%s.zip" % task
+    urllib.request.urlretrieve(TASK2PATH[task], data_file)
+    with zipfile.ZipFile(data_file) as zip_ref:
+        zip_ref.extractall(data_dir)
+    os.remove(data_file)
+    print("\tCompleted!")
+
+def format_mrpc(data_dir, path_to_data):
+    print("Processing MRPC...")
+    mrpc_dir = os.path.join(data_dir, "MRPC")
+    if not os.path.isdir(mrpc_dir):
+        os.mkdir(mrpc_dir)
+    if path_to_data:
+        mrpc_train_file = os.path.join(path_to_data, "msr_paraphrase_train.txt")
+        mrpc_test_file = os.path.join(path_to_data, "msr_paraphrase_test.txt")
+    else:
+        print("Local MRPC data not specified, downloading data from %s" % MRPC_TRAIN)
+        mrpc_train_file = os.path.join(mrpc_dir, "msr_paraphrase_train.txt")
+        mrpc_test_file = os.path.join(mrpc_dir, "msr_paraphrase_test.txt")
+        urllib.request.urlretrieve(MRPC_TRAIN, mrpc_train_file)
+        urllib.request.urlretrieve(MRPC_TEST, mrpc_test_file)
+    assert os.path.isfile(mrpc_train_file), "Train data not found at %s" % mrpc_train_file
+    assert os.path.isfile(mrpc_test_file), "Test data not found at %s" % mrpc_test_file
+    urllib.request.urlretrieve(TASK2PATH["MRPC"], os.path.join(mrpc_dir, "dev_ids.tsv"))
+
+    dev_ids = []
+    with open(os.path.join(mrpc_dir, "dev_ids.tsv"), encoding="utf8") as ids_fh:
+        for row in ids_fh:
+            dev_ids.append(row.strip().split('\t'))
+
+    with open(mrpc_train_file, encoding="utf8") as data_fh, \
+         open(os.path.join(mrpc_dir, "train.tsv"), 'w', encoding="utf8") as train_fh, \
+         open(os.path.join(mrpc_dir, "dev.tsv"), 'w', encoding="utf8") as dev_fh:
+        header = data_fh.readline()
+        train_fh.write(header)
+        dev_fh.write(header)
+        for row in data_fh:
+            label, id1, id2, s1, s2 = row.strip().split('\t')
+            if [id1, id2] in dev_ids:
+                dev_fh.write("%s\t%s\t%s\t%s\t%s\n" % (label, id1, id2, s1, s2))
+            else:
+                train_fh.write("%s\t%s\t%s\t%s\t%s\n" % (label, id1, id2, s1, s2))
+
+    with open(mrpc_test_file, encoding="utf8") as data_fh, \
+            open(os.path.join(mrpc_dir, "test.tsv"), 'w', encoding="utf8") as test_fh:
+        header = data_fh.readline()
+        test_fh.write("index\t#1 ID\t#2 ID\t#1 String\t#2 String\n")
+        for idx, row in enumerate(data_fh):
+            label, id1, id2, s1, s2 = row.strip().split('\t')
+            test_fh.write("%d\t%s\t%s\t%s\t%s\n" % (idx, id1, id2, s1, s2))
+    print("\tCompleted!")
+
+def download_diagnostic(data_dir):
+    print("Downloading and extracting diagnostic...")
+    if not os.path.isdir(os.path.join(data_dir, "diagnostic")):
+        os.mkdir(os.path.join(data_dir, "diagnostic"))
+    data_file = os.path.join(data_dir, "diagnostic", "diagnostic.tsv")
+    urllib.request.urlretrieve(TASK2PATH["diagnostic"], data_file)
+    print("\tCompleted!")
+    return
+
+def get_tasks(task_names):
+    task_names = task_names.split(',')
+    if "all" in task_names:
+        tasks = TASKS
+    else:
+        tasks = []
+        for task_name in task_names:
+            assert task_name in TASKS, "Task %s not found!" % task_name
+            tasks.append(task_name)
+    return tasks
+
+def main(arguments):
+    parser = argparse.ArgumentParser()
+    parser.add_argument('--data_dir', help='directory to save data to', type=str, default='glue_data')
+    parser.add_argument('--tasks', help='tasks to download data for as a comma separated string',
+                        type=str, default='all')
+    parser.add_argument('--path_to_mrpc', help='path to directory containing extracted MRPC data, msr_paraphrase_train.txt and msr_paraphrase_text.txt',
+                        type=str, default='')
+    args = parser.parse_args(arguments)
+
+    if not os.path.isdir(args.data_dir):
+        os.mkdir(args.data_dir)
+    tasks = get_tasks(args.tasks)
+
+    for task in tasks:
+        if task == 'MRPC':
+            format_mrpc(args.data_dir, args.path_to_mrpc)
+        elif task == 'diagnostic':
+            download_diagnostic(args.data_dir)
+        else:
+            download_and_extract(task, args.data_dir)
+
+
+if __name__ == '__main__':
+    sys.exit(main(sys.argv[1:]))

From 05d4232f63f121baefec9a87704ea7a15933f6e9 Mon Sep 17 00:00:00 2001
From: Juha Kiili <juha@juhakiili.com>
Date: Thu, 21 Nov 2019 12:38:17 +0200
Subject: [PATCH 063/505] Add valohai.yaml

---
 valohai.yaml | 94 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 94 insertions(+)
 create mode 100644 valohai.yaml

diff --git a/valohai.yaml b/valohai.yaml
new file mode 100644
index 0000000000..2573551b4e
--- /dev/null
+++ b/valohai.yaml
@@ -0,0 +1,94 @@
+---
+
+- step:
+    name: Execute python examples/run_glue.py
+    image: pytorch/pytorch:nightly-devel-cuda10.0-cudnn7
+    command:
+      - python /valohai/repository/utils/download_glue_data.py --data_dir=/glue_data
+      - pip install -e .
+      - pip install -r examples/requirements.txt
+      - python examples/run_glue.py --do_train --data_dir=/glue_data/{parameter-value:task_name} {parameters}
+    parameters:
+      - name: model_type
+        pass-as: --model_type={v}
+        type: string
+        default: bert
+      - name: model_name_or_path
+        pass-as: --model_name_or_path={v}
+        type: string
+        default: bert-base-uncased
+      - name: task_name
+        pass-as: --task_name={v}
+        type: string
+        default: MRPC
+      - name: max_seq_length
+        pass-as: --max_seq_length={v}
+        description: The maximum total input sequence length after tokenization. Sequences longer than this will be truncated, sequences shorter will be padded.
+        type: integer
+        default: 128
+      - name: per_gpu_train_batch_size
+        pass-as: --per_gpu_train_batch_size={v}
+        description: Batch size per GPU/CPU for training.
+        type: integer
+        default: 8
+      - name: per_gpu_eval_batch_size
+        pass-as: --per_gpu_eval_batch_size={v}
+        description: Batch size per GPU/CPU for evaluation.
+        type: integer
+        default: 8
+      - name: gradient_accumulation_steps
+        pass-as: --gradient_accumulation_steps={v}
+        description: Number of updates steps to accumulate before performing a backward/update pass.
+        type: integer
+        default: 1
+      - name: learning_rate
+        pass-as: --learning_rate={v}
+        description: The initial learning rate for Adam.
+        type: float
+        default: 0.00005
+      - name: adam_epsilon
+        pass-as: --adam_epsilon={v}
+        description: Epsilon for Adam optimizer.
+        type: float
+        default: 0.00000001
+      - name: max_grad_norm
+        pass-as: --max_grad_norm={v}
+        description: Max gradient norm.
+        type: float
+        default: 1.0
+      - name: num_train_epochs
+        pass-as: --num_train_epochs={v}
+        description: Total number of training epochs to perform.
+        type: integer
+        default: 3
+      - name: max_steps
+        pass-as: --max_steps={v}
+        description: If > 0, set total number of training steps to perform. Override num_train_epochs.
+        type: integer
+        default: -1
+      - name: warmup_steps
+        pass-as: --warmup_steps={v}
+        description: Linear warmup over warmup_steps.
+        type: integer
+        default: -1
+      - name: logging_steps
+        pass-as: --logging_steps={v}
+        description: Log every X updates steps.
+        type: integer
+        default: 25
+      - name: save_steps
+        pass-as: --save_steps={v}
+        description: Save checkpoint every X updates steps.
+        type: integer
+        default: -1
+      - name: output_dir
+        pass-as: --output_dir={v}
+        type: string
+        default: /valohai/outputs
+      - name: evaluate_during_training
+        description: Run evaluation during training at each logging step.
+        type: flag
+        default: true
+      - name: do_lower_case
+        description: Set this flag if you are using an uncased model.
+        type: flag

From 6f70bb8c69cdbe1207e4909ccface32c8f51297b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Wed, 20 Nov 2019 18:01:03 +0100
Subject: [PATCH 064/505] add instructions to run the examples

---
 README.md | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/README.md b/README.md
index a49e8086d1..9cc783fd71 100644
--- a/README.md
+++ b/README.md
@@ -86,6 +86,18 @@ When TensorFlow 2.0 and/or PyTorch has been installed, you can install from sour
 pip install [--editable] .
 ```
 
+### Run the examples
+
+Examples are included in the repository but are not shipped with the library.
+Therefore, in order to run the examples you will first need to clone the
+repository and install the bleeding edge version of the library. To do so, create a new virtual environment and follow these steps:
+
+```bash
+git clone git@github.com:huggingface/transformers
+cd transformers
+pip install .
+```
+
 ### Tests
 
 A series of tests are included for the library and the example scripts. Library tests can be found in the [tests folder](https://github.com/huggingface/transformers/tree/master/transformers/tests) and examples tests in the [examples folder](https://github.com/huggingface/transformers/tree/master/examples).
@@ -253,6 +265,11 @@ print("sentence_2 is", "a paraphrase" if pred_2 else "not a paraphrase", "of sen
 
 ## Quick tour of the fine-tuning/usage scripts
 
+**Important**  
+Before running the fine-tuning scripts, please read the
+[instructions](#run-the-examples) on how to
+setup your environment to run the examples.
+
 The library comprises several example scripts with SOTA performances for NLU and NLG tasks:
 
 - `run_glue.py`: an example fine-tuning Bert, XLNet and XLM on nine different GLUE tasks (*sequence-level classification*)

From 26db31e0c09a8b5e1ca7a61c454b159eab9d86be Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Wed, 20 Nov 2019 18:13:38 +0100
Subject: [PATCH 065/505] update the documentation

---
 examples/README.md | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/examples/README.md b/examples/README.md
index abb4cb6e5a..622fa07f8f 100644
--- a/examples/README.md
+++ b/examples/README.md
@@ -3,6 +3,15 @@
 In this section a few examples are put together. All of these examples work for several models, making use of the very
 similar API between the different models.
 
+**Important**  
+To use the examples, execute the following steps in a new virtual environment:
+
+```bash
+git clone git@github.com:huggingface/transformers
+cd transformers
+pip install .
+```
+
 | Section                    | Description                                                                                                                                                |
 |----------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------|
 | [TensorFlow 2.0 models on GLUE](#TensorFlow-2.0-Bert-models-on-GLUE) | Examples running BERT TensorFlow 2.0 model on the GLUE tasks. 

From ea52f82455a7ca0f979768204dfeb38b5fff13ad Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Mon, 18 Nov 2019 14:42:59 -0500
Subject: [PATCH 066/505] Moved some SQuAD logic to /data

---
 transformers/__init__.py                 |   3 +-
 transformers/data/__init__.py            |   3 +-
 transformers/data/processors/__init__.py |   1 +
 transformers/data/processors/squad.py    | 318 +++++++++++++++++++++++
 4 files changed, 323 insertions(+), 2 deletions(-)
 create mode 100644 transformers/data/processors/squad.py

diff --git a/transformers/__init__.py b/transformers/__init__.py
index 5c7b0a6197..b859e18c53 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -25,7 +25,8 @@ from .file_utils import (TRANSFORMERS_CACHE, PYTORCH_TRANSFORMERS_CACHE, PYTORCH
 from .data import (is_sklearn_available,
                    InputExample, InputFeatures, DataProcessor,
                    glue_output_modes, glue_convert_examples_to_features,
-                   glue_processors, glue_tasks_num_labels)
+                   glue_processors, glue_tasks_num_labels,
+                   squad_convert_examples_to_features, SquadFeatures)
 
 if is_sklearn_available():
     from .data import glue_compute_metrics
diff --git a/transformers/data/__init__.py b/transformers/data/__init__.py
index e910d6da2e..827d96ed29 100644
--- a/transformers/data/__init__.py
+++ b/transformers/data/__init__.py
@@ -1,5 +1,6 @@
-from .processors import InputExample, InputFeatures, DataProcessor
+from .processors import InputExample, InputFeatures, DataProcessor, SquadFeatures
 from .processors import glue_output_modes, glue_processors, glue_tasks_num_labels, glue_convert_examples_to_features
+from .processors import squad_convert_examples_to_features
 
 from .metrics import is_sklearn_available
 if is_sklearn_available():
diff --git a/transformers/data/processors/__init__.py b/transformers/data/processors/__init__.py
index af38c54beb..4e322a2ca8 100644
--- a/transformers/data/processors/__init__.py
+++ b/transformers/data/processors/__init__.py
@@ -1,3 +1,4 @@
 from .utils import InputExample, InputFeatures, DataProcessor
 from .glue import glue_output_modes, glue_processors, glue_tasks_num_labels, glue_convert_examples_to_features
+from .squad import squad_convert_examples_to_features, SquadFeatures
 
diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
new file mode 100644
index 0000000000..c1a1034f17
--- /dev/null
+++ b/transformers/data/processors/squad.py
@@ -0,0 +1,318 @@
+from tqdm import tqdm
+import collections
+import logging
+import os
+
+from .utils import DataProcessor, InputExample, InputFeatures
+from ...file_utils import is_tf_available
+
+if is_tf_available():
+    import tensorflow as tf
+
+logger = logging.getLogger(__name__)
+
+def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
+                                       doc_stride, max_query_length, is_training,
+                                       cls_token_at_end=False,
+                                       cls_token='[CLS]', sep_token='[SEP]', pad_token=0,
+                                       sequence_a_segment_id=0, sequence_b_segment_id=1,
+                                       cls_token_segment_id=0, pad_token_segment_id=0,
+                                       mask_padding_with_zero=True,
+                                       sequence_a_is_doc=False):
+    """Loads a data file into a list of `InputBatch`s."""
+
+    # Defining helper methods
+    def _improve_answer_span(doc_tokens, input_start, input_end, tokenizer,
+                         orig_answer_text):
+        """Returns tokenized answer spans that better match the annotated answer."""
+        tok_answer_text = " ".join(tokenizer.tokenize(orig_answer_text))
+
+        for new_start in range(input_start, input_end + 1):
+            for new_end in range(input_end, new_start - 1, -1):
+                text_span = " ".join(doc_tokens[new_start:(new_end + 1)])
+                if text_span == tok_answer_text:
+                    return (new_start, new_end)
+
+        return (input_start, input_end)
+    def _check_is_max_context(doc_spans, cur_span_index, position):
+        """Check if this is the 'max context' doc span for the token."""
+        best_score = None
+        best_span_index = None
+        for (span_index, doc_span) in enumerate(doc_spans):
+            end = doc_span.start + doc_span.length - 1
+            if position < doc_span.start:
+                continue
+            if position > end:
+                continue
+            num_left_context = position - doc_span.start
+            num_right_context = end - position
+            score = min(num_left_context, num_right_context) + 0.01 * doc_span.length
+            if best_score is None or score > best_score:
+                best_score = score
+                best_span_index = span_index
+
+        return cur_span_index == best_span_index
+    
+    unique_id = 1000000000
+
+    features = []
+    for (example_index, example) in enumerate(tqdm(examples)):
+        query_tokens = tokenizer.tokenize(example.question_text)
+
+        if len(query_tokens) > max_query_length:
+            query_tokens = query_tokens[0:max_query_length]
+
+        tok_to_orig_index = []
+        orig_to_tok_index = []
+        all_doc_tokens = []
+        for (i, token) in enumerate(example.doc_tokens):
+            orig_to_tok_index.append(len(all_doc_tokens))
+            sub_tokens = tokenizer.tokenize(token)
+            for sub_token in sub_tokens:
+                tok_to_orig_index.append(i)
+                all_doc_tokens.append(sub_token)
+
+        tok_start_position = None
+        tok_end_position = None
+        if is_training and example.is_impossible:
+            tok_start_position = -1
+            tok_end_position = -1
+        if is_training and not example.is_impossible:
+            tok_start_position = orig_to_tok_index[example.start_position]
+            if example.end_position < len(example.doc_tokens) - 1:
+                tok_end_position = orig_to_tok_index[example.end_position + 1] - 1
+            else:
+                tok_end_position = len(all_doc_tokens) - 1
+            (tok_start_position, tok_end_position) = _improve_answer_span(
+                all_doc_tokens, tok_start_position, tok_end_position, tokenizer,
+                example.orig_answer_text)
+
+        # The -3 accounts for [CLS], [SEP] and [SEP]
+        max_tokens_for_doc = max_seq_length - len(query_tokens) - 3
+
+        # We can have documents that are longer than the maximum sequence length.
+        # To deal with this we do a sliding window approach, where we take chunks
+        # of the up to our max length with a stride of `doc_stride`.
+        _DocSpan = collections.namedtuple(  # pylint: disable=invalid-name
+            "DocSpan", ["start", "length"])
+        doc_spans = []
+        start_offset = 0
+        while start_offset < len(all_doc_tokens):
+            length = len(all_doc_tokens) - start_offset
+            if length > max_tokens_for_doc:
+                length = max_tokens_for_doc
+            doc_spans.append(_DocSpan(start=start_offset, length=length))
+            if start_offset + length == len(all_doc_tokens):
+                break
+            start_offset += min(length, doc_stride)
+
+        for (doc_span_index, doc_span) in enumerate(doc_spans):
+            tokens = []
+            token_to_orig_map = {}
+            token_is_max_context = {}
+            segment_ids = []
+
+            # p_mask: mask with 1 for token than cannot be in the answer (0 for token which can be in an answer)
+            # Original TF implem also keep the classification token (set to 0) (not sure why...)
+            p_mask = []
+
+            # CLS token at the beginning
+            if not cls_token_at_end:
+                tokens.append(cls_token)
+                segment_ids.append(cls_token_segment_id)
+                p_mask.append(0)
+                cls_index = 0
+
+            # XLNet: P SEP Q SEP CLS
+            # Others: CLS Q SEP P SEP
+            if not sequence_a_is_doc:
+                # Query
+                tokens += query_tokens
+                segment_ids += [sequence_a_segment_id] * len(query_tokens)
+                p_mask += [1] * len(query_tokens)
+
+                # SEP token
+                tokens.append(sep_token)
+                segment_ids.append(sequence_a_segment_id)
+                p_mask.append(1)
+
+            # Paragraph
+            for i in range(doc_span.length):
+                split_token_index = doc_span.start + i
+                token_to_orig_map[len(tokens)] = tok_to_orig_index[split_token_index]
+
+                is_max_context = _check_is_max_context(doc_spans, doc_span_index,
+                                                       split_token_index)
+                token_is_max_context[len(tokens)] = is_max_context
+                tokens.append(all_doc_tokens[split_token_index])
+                if not sequence_a_is_doc:
+                    segment_ids.append(sequence_b_segment_id)
+                else:
+                    segment_ids.append(sequence_a_segment_id)
+                p_mask.append(0)
+            paragraph_len = doc_span.length
+
+            if sequence_a_is_doc:
+                # SEP token
+                tokens.append(sep_token)
+                segment_ids.append(sequence_a_segment_id)
+                p_mask.append(1)
+
+                tokens += query_tokens
+                segment_ids += [sequence_b_segment_id] * len(query_tokens)
+                p_mask += [1] * len(query_tokens)
+
+            # SEP token
+            tokens.append(sep_token)
+            segment_ids.append(sequence_b_segment_id)
+            p_mask.append(1)
+
+            # CLS token at the end
+            if cls_token_at_end:
+                tokens.append(cls_token)
+                segment_ids.append(cls_token_segment_id)
+                p_mask.append(0)
+                cls_index = len(tokens) - 1  # Index of classification token
+
+            input_ids = tokenizer.convert_tokens_to_ids(tokens)
+
+            # The mask has 1 for real tokens and 0 for padding tokens. Only real
+            # tokens are attended to.
+            input_mask = [1 if mask_padding_with_zero else 0] * len(input_ids)
+
+            # Zero-pad up to the sequence length.
+            while len(input_ids) < max_seq_length:
+                input_ids.append(pad_token)
+                input_mask.append(0 if mask_padding_with_zero else 1)
+                segment_ids.append(pad_token_segment_id)
+                p_mask.append(1)
+
+            assert len(input_ids) == max_seq_length
+            assert len(input_mask) == max_seq_length
+            assert len(segment_ids) == max_seq_length
+
+            span_is_impossible = example.is_impossible
+            start_position = None
+            end_position = None
+            if is_training and not span_is_impossible:
+                # For training, if our document chunk does not contain an annotation
+                # we throw it out, since there is nothing to predict.
+                doc_start = doc_span.start
+                doc_end = doc_span.start + doc_span.length - 1
+                out_of_span = False
+                if not (tok_start_position >= doc_start and
+                        tok_end_position <= doc_end):
+                    out_of_span = True
+                if out_of_span:
+                    start_position = 0
+                    end_position = 0
+                    span_is_impossible = True
+                else:
+                    if sequence_a_is_doc:
+                        doc_offset = 0
+                    else:
+                        doc_offset = len(query_tokens) + 2
+                    start_position = tok_start_position - doc_start + doc_offset
+                    end_position = tok_end_position - doc_start + doc_offset
+
+            if is_training and span_is_impossible:
+                start_position = cls_index
+                end_position = cls_index
+
+            if example_index < 20:
+                logger.info("*** Example ***")
+                logger.info("unique_id: %s" % (unique_id))
+                logger.info("example_index: %s" % (example_index))
+                logger.info("doc_span_index: %s" % (doc_span_index))
+                logger.info("tokens: %s" % " ".join(tokens))
+                logger.info("token_to_orig_map: %s" % " ".join([
+                    "%d:%d" % (x, y) for (x, y) in token_to_orig_map.items()]))
+                logger.info("token_is_max_context: %s" % " ".join([
+                    "%d:%s" % (x, y) for (x, y) in token_is_max_context.items()
+                ]))
+                logger.info("input_ids: %s" % " ".join([str(x) for x in input_ids]))
+                logger.info(
+                    "input_mask: %s" % " ".join([str(x) for x in input_mask]))
+                logger.info(
+                    "segment_ids: %s" % " ".join([str(x) for x in segment_ids]))
+                if is_training and span_is_impossible:
+                    logger.info("impossible example")
+                if is_training and not span_is_impossible:
+                    answer_text = " ".join(tokens[start_position:(end_position + 1)])
+                    logger.info("start_position: %d" % (start_position))
+                    logger.info("end_position: %d" % (end_position))
+                    logger.info(
+                        "answer: %s" % (answer_text))
+
+            features.append(
+                SquadFeatures(
+                    unique_id=unique_id,
+                    example_index=example_index,
+                    doc_span_index=doc_span_index,
+                    tokens=tokens,
+                    token_to_orig_map=token_to_orig_map,
+                    token_is_max_context=token_is_max_context,
+                    input_ids=input_ids,
+                    input_mask=input_mask,
+                    segment_ids=segment_ids,
+                    cls_index=cls_index,
+                    p_mask=p_mask,
+                    paragraph_len=paragraph_len,
+                    start_position=start_position,
+                    end_position=end_position,
+                    is_impossible=span_is_impossible))
+            unique_id += 1
+
+    return features
+
+class SquadFeatures(object):
+    """A single set of features of data."""
+
+    def __init__(self,
+                 unique_id,
+                 example_index,
+                 doc_span_index,
+                 tokens,
+                 token_to_orig_map,
+                 token_is_max_context,
+                 input_ids,
+                 input_mask,
+                 segment_ids,
+                 cls_index,
+                 p_mask,
+                 paragraph_len,
+                 start_position=None,
+                 end_position=None,
+                 is_impossible=None):
+        self.unique_id = unique_id
+        self.example_index = example_index
+        self.doc_span_index = doc_span_index
+        self.tokens = tokens
+        self.token_to_orig_map = token_to_orig_map
+        self.token_is_max_context = token_is_max_context
+        self.input_ids = input_ids
+        self.input_mask = input_mask
+        self.segment_ids = segment_ids
+        self.cls_index = cls_index
+        self.p_mask = p_mask
+        self.paragraph_len = paragraph_len
+        self.start_position = start_position
+        self.end_position = end_position
+        self.is_impossible = is_impossible
+
+    def __eq__(self, other):
+        return self.cls_index == other.cls_index and \
+                self.doc_span_index == other.doc_span_index and \
+                self.end_position == other.end_position and \
+                self.example_index == other.example_index and \
+                self.input_ids == other.input_ids and \
+                self.input_mask == other.input_mask and \
+                self.is_impossible == other.is_impossible and \
+                self.p_mask == other.p_mask and \
+                self.paragraph_len == other.paragraph_len and \
+                self.segment_ids == other.segment_ids and \
+                self.start_position == other.start_position and \
+                self.token_is_max_context == other.token_is_max_context and \
+                self.token_to_orig_map == other.token_to_orig_map and \
+                self.tokens == other.tokens and \
+                self.unique_id == other.unique_id
\ No newline at end of file

From 72e506b22e90feab6c410136bacc27f3d65284b9 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Tue, 19 Nov 2019 09:49:55 -0500
Subject: [PATCH 067/505] wip

---
 examples/run_squad.py                    |  29 +++++-
 transformers/__init__.py                 |   3 +-
 transformers/data/__init__.py            |   2 +-
 transformers/data/processors/__init__.py |   2 +-
 transformers/data/processors/squad.py    | 122 +++++++++++++++++++++++
 transformers/tokenization_utils.py       |   4 +
 6 files changed, 157 insertions(+), 5 deletions(-)

diff --git a/examples/run_squad.py b/examples/run_squad.py
index 69088d73c3..d4219c3096 100644
--- a/examples/run_squad.py
+++ b/examples/run_squad.py
@@ -23,7 +23,6 @@ import os
 import random
 import glob
 import timeit
-
 import numpy as np
 import torch
 from torch.utils.data import (DataLoader, RandomSampler, SequentialSampler,
@@ -45,7 +44,7 @@ from transformers import (WEIGHTS_NAME, BertConfig,
                                   XLNetTokenizer,
                                   DistilBertConfig, DistilBertForQuestionAnswering, DistilBertTokenizer)
 
-from transformers import AdamW, get_linear_schedule_with_warmup
+from transformers import AdamW, get_linear_schedule_with_warmup, squad_convert_examples_to_features, read_squad_examples as sread_squad_examples
 
 from utils_squad import (read_squad_examples, convert_examples_to_features,
                          RawResult, write_predictions,
@@ -309,6 +308,8 @@ def load_and_cache_examples(args, tokenizer, evaluate=False, output_examples=Fal
         examples = read_squad_examples(input_file=input_file,
                                                 is_training=not evaluate,
                                                 version_2_with_negative=args.version_2_with_negative)
+
+        examples = examples[:10]
         features = convert_examples_to_features(examples=examples,
                                                 tokenizer=tokenizer,
                                                 max_seq_length=args.max_seq_length,
@@ -319,6 +320,30 @@ def load_and_cache_examples(args, tokenizer, evaluate=False, output_examples=Fal
                                                 pad_token_segment_id=3 if args.model_type in ['xlnet'] else 0,
                                                 cls_token_at_end=True if args.model_type in ['xlnet'] else False,
                                                 sequence_a_is_doc=True if args.model_type in ['xlnet'] else False)
+
+        exampless = sread_squad_examples(input_file=input_file,
+                                                is_training=not evaluate,
+                                                version_2_with_negative=args.version_2_with_negative)
+        exampless = exampless[:10]
+        features2 = squad_convert_examples_to_features(examples=exampless,
+                                                tokenizer=tokenizer,
+                                                max_seq_length=args.max_seq_length,
+                                                doc_stride=args.doc_stride,
+                                                max_query_length=args.max_query_length,
+                                                is_training=not evaluate,
+                                                cls_token_segment_id=2 if args.model_type in ['xlnet'] else 0,
+                                                pad_token_segment_id=3 if args.model_type in ['xlnet'] else 0,
+                                                cls_token_at_end=True if args.model_type in ['xlnet'] else False,
+                                                sequence_a_is_doc=True if args.model_type in ['xlnet'] else False)
+
+        print(features2)
+
+        for i in range(len(features)):
+            assert features[i] == features2[i]
+            print("Equal")
+
+        print("DONE")
+        
         if args.local_rank in [-1, 0]:
             logger.info("Saving features into cached file %s", cached_features_file)
             torch.save(features, cached_features_file)
diff --git a/transformers/__init__.py b/transformers/__init__.py
index b859e18c53..9a767913b3 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -26,7 +26,8 @@ from .data import (is_sklearn_available,
                    InputExample, InputFeatures, DataProcessor,
                    glue_output_modes, glue_convert_examples_to_features,
                    glue_processors, glue_tasks_num_labels,
-                   squad_convert_examples_to_features, SquadFeatures)
+                   squad_convert_examples_to_features, SquadFeatures, 
+                   SquadExample, read_squad_examples)
 
 if is_sklearn_available():
     from .data import glue_compute_metrics
diff --git a/transformers/data/__init__.py b/transformers/data/__init__.py
index 827d96ed29..50f2e768f4 100644
--- a/transformers/data/__init__.py
+++ b/transformers/data/__init__.py
@@ -1,6 +1,6 @@
 from .processors import InputExample, InputFeatures, DataProcessor, SquadFeatures
 from .processors import glue_output_modes, glue_processors, glue_tasks_num_labels, glue_convert_examples_to_features
-from .processors import squad_convert_examples_to_features
+from .processors import squad_convert_examples_to_features, SquadExample, read_squad_examples
 
 from .metrics import is_sklearn_available
 if is_sklearn_available():
diff --git a/transformers/data/processors/__init__.py b/transformers/data/processors/__init__.py
index 4e322a2ca8..924b4a1245 100644
--- a/transformers/data/processors/__init__.py
+++ b/transformers/data/processors/__init__.py
@@ -1,4 +1,4 @@
 from .utils import InputExample, InputFeatures, DataProcessor
 from .glue import glue_output_modes, glue_processors, glue_tasks_num_labels, glue_convert_examples_to_features
-from .squad import squad_convert_examples_to_features, SquadFeatures
+from .squad import squad_convert_examples_to_features, SquadFeatures, SquadExample, read_squad_examples
 
diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index c1a1034f17..1900e9f0ce 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -2,7 +2,9 @@ from tqdm import tqdm
 import collections
 import logging
 import os
+import json
 
+from ...tokenization_bert import BasicTokenizer, whitespace_tokenize
 from .utils import DataProcessor, InputExample, InputFeatures
 from ...file_utils import is_tf_available
 
@@ -11,6 +13,7 @@ if is_tf_available():
 
 logger = logging.getLogger(__name__)
 
+
 def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
                                        doc_stride, max_query_length, is_training,
                                        cls_token_at_end=False,
@@ -265,6 +268,125 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
 
     return features
 
+
+def read_squad_examples(input_file, is_training, version_2_with_negative):
+    """Read a SQuAD json file into a list of SquadExample."""
+    with open(input_file, "r", encoding='utf-8') as reader:
+        input_data = json.load(reader)["data"]
+
+    def is_whitespace(c):
+        if c == " " or c == "\t" or c == "\r" or c == "\n" or ord(c) == 0x202F:
+            return True
+        return False
+
+    examples = []
+    for entry in input_data:
+        for paragraph in entry["paragraphs"]:
+            paragraph_text = paragraph["context"]
+            doc_tokens = []
+            char_to_word_offset = []
+            prev_is_whitespace = True
+            for c in paragraph_text:
+                if is_whitespace(c):
+                    prev_is_whitespace = True
+                else:
+                    if prev_is_whitespace:
+                        doc_tokens.append(c)
+                    else:
+                        doc_tokens[-1] += c
+                    prev_is_whitespace = False
+                char_to_word_offset.append(len(doc_tokens) - 1)
+
+            for qa in paragraph["qas"]:
+                qas_id = qa["id"]
+                question_text = qa["question"]
+                start_position = None
+                end_position = None
+                orig_answer_text = None
+                is_impossible = False
+                if is_training:
+                    if version_2_with_negative:
+                        is_impossible = qa["is_impossible"]
+                    if (len(qa["answers"]) != 1) and (not is_impossible):
+                        raise ValueError(
+                            "For training, each question should have exactly 1 answer.")
+                    if not is_impossible:
+                        answer = qa["answers"][0]
+                        orig_answer_text = answer["text"]
+                        answer_offset = answer["answer_start"]
+                        answer_length = len(orig_answer_text)
+                        start_position = char_to_word_offset[answer_offset]
+                        end_position = char_to_word_offset[answer_offset + answer_length - 1]
+                        # Only add answers where the text can be exactly recovered from the
+                        # document. If this CAN'T happen it's likely due to weird Unicode
+                        # stuff so we will just skip the example.
+                        #
+                        # Note that this means for training mode, every example is NOT
+                        # guaranteed to be preserved.
+                        actual_text = " ".join(doc_tokens[start_position:(end_position + 1)])
+                        cleaned_answer_text = " ".join(
+                            whitespace_tokenize(orig_answer_text))
+                        if actual_text.find(cleaned_answer_text) == -1:
+                            logger.warning("Could not find answer: '%s' vs. '%s'",
+                                           actual_text, cleaned_answer_text)
+                            continue
+                    else:
+                        start_position = -1
+                        end_position = -1
+                        orig_answer_text = ""
+
+                example = SquadExample(
+                    qas_id=qas_id,
+                    question_text=question_text,
+                    doc_tokens=doc_tokens,
+                    orig_answer_text=orig_answer_text,
+                    start_position=start_position,
+                    end_position=end_position,
+                    is_impossible=is_impossible)
+                examples.append(example)
+    return examples
+
+
+class SquadExample(object):
+    """
+    A single training/test example for the Squad dataset.
+    For examples without an answer, the start and end position are -1.
+    """
+
+    def __init__(self,
+                 qas_id,
+                 question_text,
+                 doc_tokens,
+                 orig_answer_text=None,
+                 start_position=None,
+                 end_position=None,
+                 is_impossible=None):
+        self.qas_id = qas_id
+        self.question_text = question_text
+        self.doc_tokens = doc_tokens
+        self.orig_answer_text = orig_answer_text
+        self.start_position = start_position
+        self.end_position = end_position
+        self.is_impossible = is_impossible
+
+    def __str__(self):
+        return self.__repr__()
+
+    def __repr__(self):
+        s = ""
+        s += "qas_id: %s" % (self.qas_id)
+        s += ", question_text: %s" % (
+            self.question_text)
+        s += ", doc_tokens: [%s]" % (" ".join(self.doc_tokens))
+        if self.start_position:
+            s += ", start_position: %d" % (self.start_position)
+        if self.end_position:
+            s += ", end_position: %d" % (self.end_position)
+        if self.is_impossible:
+            s += ", is_impossible: %r" % (self.is_impossible)
+        return s
+
+
 class SquadFeatures(object):
     """A single set of features of data."""
 
diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index 4fa26a26f8..ba10e6b311 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -605,6 +605,10 @@ class PreTrainedTokenizer(object):
             vocabularies (BPE/SentencePieces/WordPieces).
 
             Take care of added tokens.
+
+            text: The sequence to be encoded.
+            return_tokens_mapped_to_origin: (optional) Set to True to return the index of each token in the initial whitespace tokenization. (default False).
+            **kwargs: passed to the child `self.tokenize()` method
         """
         def split_on_token(tok, text):
             result = []

From 9f374c8252330bffd669c43749b5e937ed31d90a Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Fri, 22 Nov 2019 16:27:15 -0500
Subject: [PATCH 068/505] `encode` and `encode_plus` handle attention masks and
 padding

---
 .../tests/tokenization_tests_commons.py       | 51 ++++++++++++
 transformers/tokenization_utils.py            | 77 ++++++++++++++++++-
 transformers/tokenization_xlnet.py            |  1 +
 3 files changed, 127 insertions(+), 2 deletions(-)

diff --git a/transformers/tests/tokenization_tests_commons.py b/transformers/tests/tokenization_tests_commons.py
index fdaf8cc137..d5b70d5266 100644
--- a/transformers/tests/tokenization_tests_commons.py
+++ b/transformers/tests/tokenization_tests_commons.py
@@ -335,3 +335,54 @@ class CommonTestCases:
             special_tokens_mask = tokenizer.get_special_tokens_mask(encoded_sequence_w_special, already_has_special_tokens=True)
             self.assertEqual(len(special_tokens_mask), len(encoded_sequence_w_special))
             self.assertEqual(special_tokens_mask_orig, special_tokens_mask)
+
+        def test_padding_to_max_length(self):
+            tokenizer = self.get_tokenizer()
+
+            sequence = "Sequence"
+            padding_size = 10
+            padding_idx = tokenizer.pad_token_id
+
+            # Check that it correctly pads when a maximum length is specified along with the padding flag set to True
+            encoded_sequence = tokenizer.encode(sequence)
+            sequence_length = len(encoded_sequence)
+            padded_sequence = tokenizer.encode(sequence, max_length=sequence_length + padding_size, pad_to_max_length=True)
+            padded_sequence_length = len(padded_sequence)
+            assert sequence_length + padding_size == padded_sequence_length
+            assert encoded_sequence + [padding_idx] * padding_size == padded_sequence
+
+            # Check that nothing is done when a maximum length is not specified
+            encoded_sequence = tokenizer.encode(sequence)
+            sequence_length = len(encoded_sequence)
+            padded_sequence = tokenizer.encode(sequence, pad_to_max_length=True)
+            padded_sequence_length = len(padded_sequence)
+            assert sequence_length == padded_sequence_length
+            assert encoded_sequence == padded_sequence
+
+        def test_encode_plus_with_padding(self):
+            tokenizer = self.get_tokenizer()
+
+            sequence = "Sequence"
+            padding_size = 10
+            padding_idx = tokenizer.pad_token_id
+            token_type_padding_idx = tokenizer.pad_token_type_id
+
+            encoded_sequence = tokenizer.encode_plus(sequence, return_special_tokens_mask=True)
+            input_ids = encoded_sequence['input_ids']
+            token_type_ids = encoded_sequence['token_type_ids']
+            attention_mask = encoded_sequence['attention_mask']
+            special_tokens_mask = encoded_sequence['special_tokens_mask']
+            sequence_length = len(input_ids)
+
+            padded_sequence = tokenizer.encode_plus(sequence, max_length=sequence_length + padding_size, pad_to_max_length=True, return_special_tokens_mask=True)
+            padded_input_ids = padded_sequence['input_ids']
+            padded_token_type_ids = padded_sequence['token_type_ids']
+            padded_attention_mask = padded_sequence['attention_mask']
+            padded_special_tokens_mask = padded_sequence['special_tokens_mask']
+            padded_sequence_length = len(padded_input_ids)
+
+            assert sequence_length + padding_size == padded_sequence_length
+            assert input_ids + [padding_idx] * padding_size == padded_input_ids
+            assert token_type_ids + [token_type_padding_idx] * padding_size == padded_token_type_ids
+            assert attention_mask + [0] * padding_size == padded_attention_mask 
+            assert special_tokens_mask + [1] * padding_size == padded_special_tokens_mask 
\ No newline at end of file
diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index ba10e6b311..3214699e12 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -190,6 +190,11 @@ class PreTrainedTokenizer(object):
         """ Id of the padding token in the vocabulary. Log an error if used while not having been set. """
         return self.convert_tokens_to_ids(self.pad_token)
 
+    @property
+    def pad_token_type_id(self):
+        """ Id of the padding token in the vocabulary. Log an error if used while not having been set. """
+        return self._pad_token_type_id
+
     @property
     def cls_token_id(self):
         """ Id of the classification token in the vocabulary. E.g. to extract a summary of an input sequence leveraging self-attention along the full depth of the model. Log an error if used while not having been set. """
@@ -213,6 +218,7 @@ class PreTrainedTokenizer(object):
         self._pad_token = None
         self._cls_token = None
         self._mask_token = None
+        self._pad_token_type_id = 0
         self._additional_special_tokens = []
 
         self.max_len = max_len if max_len is not None else int(1e12)
@@ -696,6 +702,7 @@ class PreTrainedTokenizer(object):
                max_length=None,
                stride=0,
                truncation_strategy='longest_first',
+               pad_to_max_length=False,
                return_tensors=None,
                **kwargs):
         """
@@ -722,6 +729,8 @@ class PreTrainedTokenizer(object):
                 - 'only_first': Only truncate the first sequence
                 - 'only_second': Only truncate the second sequence
                 - 'do_not_truncate': Does not truncate (raise an error if the input sequence is longer than max_length)
+            pad_to_max_length: if set to `True`, the returned sequences will be padded according to the model's 
+                padding index, up to their max length. If no max length is specified, no padding is done.
             return_tensors: (optional) can be set to 'tf' or 'pt' to return respectively TensorFlow tf.constant
                 or PyTorch torch.Tensor instead of a list of python integers.
             **kwargs: passed to the `self.tokenize()` method
@@ -732,6 +741,7 @@ class PreTrainedTokenizer(object):
                                           add_special_tokens=add_special_tokens,
                                           stride=stride,
                                           truncation_strategy=truncation_strategy,
+                                          pad_to_max_length=pad_to_max_length,
                                           return_tensors=return_tensors,
                                           **kwargs)
 
@@ -744,7 +754,12 @@ class PreTrainedTokenizer(object):
                     max_length=None,
                     stride=0,
                     truncation_strategy='longest_first',
+                    pad_to_max_length=False,
                     return_tensors=None,
+                    return_token_type_ids=True,
+                    return_attention_mask=True,
+                    return_overflowing_tokens=False,
+                    return_special_tokens_mask=False,
                     **kwargs):
         """
         Returns a dictionary containing the encoded sequence or sequence pair and additional informations:
@@ -769,9 +784,37 @@ class PreTrainedTokenizer(object):
                 - 'only_first': Only truncate the first sequence
                 - 'only_second': Only truncate the second sequence
                 - 'do_not_truncate': Does not truncate (raise an error if the input sequence is longer than max_length)
+            pad_to_max_length: if set to `True`, the returned sequences will be padded according to the model's 
+                padding index, up to their max length. If no max length is specified, no padding is done.
             return_tensors: (optional) can be set to 'tf' or 'pt' to return respectively TensorFlow tf.constant
                 or PyTorch torch.Tensor instead of a list of python integers.
+            return_token_type_ids: (optional) Set to False to avoid returning token_type_ids (default True).
+            return_attention_mask: (optional) Set to False to avoir returning attention mask (default True)
+            return_overflowing_tokens: (optional) Set to True to return overflowing token information (default False).
+            return_special_tokens_mask: (optional) Set to True to return special tokens mask information (default False).
             **kwargs: passed to the `self.tokenize()` method
+
+        Return:
+            A Dictionary of shape::
+
+                {
+                    input_ids: list[int],
+                    token_type_ids: list[int] if return_token_type_ids is True (default)
+                    attention_mask: list[int] if return_attention_mask is True (default)
+                    overflowing_tokens: list[int] if a ``max_length`` is specified and return_overflowing_tokens is True
+                    num_truncated_tokens: int if a ``max_length`` is specified and return_overflowing_tokens is True
+                    special_tokens_mask: list[int] if ``add_special_tokens`` if set to ``True`` and return_special_tokens_mask is True
+                }
+
+            With the fields:
+                ``input_ids``: list of token ids to be fed to a model
+                ``token_type_ids``: list of token type ids to be fed to a model
+                ``attention_mask``: list of indices specifying which tokens should be attended to by the model
+
+                ``overflowing_tokens``: list of overflowing tokens if a max length is specified.
+                ``num_truncated_tokens``: number of overflowing tokens a ``max_length`` is specified
+                ``special_tokens_mask``: if adding special tokens, this is a list of [0, 1], with 0 specifying special added
+                tokens and 1 specifying sequence tokens.
         """
 
         def get_input_ids(text):
@@ -790,13 +833,24 @@ class PreTrainedTokenizer(object):
         return self.prepare_for_model(first_ids,
                                       pair_ids=second_ids,
                                       max_length=max_length,
+                                      pad_to_max_length=pad_to_max_length,
                                       add_special_tokens=add_special_tokens,
                                       stride=stride,
                                       truncation_strategy=truncation_strategy,
-                                      return_tensors=return_tensors)
+                                      return_tensors=return_tensors,
+                                      return_attention_mask=return_attention_mask,
+                                      return_token_type_ids=return_token_type_ids,
+                                      return_overflowing_tokens=return_overflowing_tokens,
+                                      return_special_tokens_mask=return_special_tokens_mask)
 
     def prepare_for_model(self, ids, pair_ids=None, max_length=None, add_special_tokens=True, stride=0,
-                          truncation_strategy='longest_first', return_tensors=None):
+                          truncation_strategy='longest_first',
+                          pad_to_max_length=False,
+                          return_tensors=None,
+                          return_token_type_ids=True,
+                          return_attention_mask=True,
+                          return_overflowing_tokens=False,
+                          return_special_tokens_mask=False):
         """
         Prepares a sequence of input id, or a pair of sequences of inputs ids so that it can be used by the model.
         It adds special tokens, truncates
@@ -819,8 +873,14 @@ class PreTrainedTokenizer(object):
                 - 'only_first': Only truncate the first sequence
                 - 'only_second': Only truncate the second sequence
                 - 'do_not_truncate': Does not truncate (raise an error if the input sequence is longer than max_length)
+            pad_to_max_length: if set to `True`, the returned sequences will be padded according to the model's 
+                padding index, up to their max length. If no max length is specified, no padding is done.
             return_tensors: (optional) can be set to 'tf' or 'pt' to return respectively TensorFlow tf.constant
                 or PyTorch torch.Tensor instead of a list of python integers.
+            return_token_type_ids: (optional) Set to False to avoid returning token_type_ids (default True).
+            return_attention_mask: (optional) Set to False to avoir returning attention mask (default True)
+            return_overflowing_tokens: (optional) Set to True to return overflowing token information (default False).
+            return_special_tokens_mask: (optional) Set to True to return special tokens mask information (default False).
 
         Return:
             A Dictionary of shape::
@@ -883,6 +943,19 @@ class PreTrainedTokenizer(object):
                            "for this model ({} > {}). Running this sequence through the model will result in "
                            "indexing errors".format(len(ids), self.max_len))
                            
+        if pad_to_max_length and max_length and len(encoded_inputs["input_ids"]) < max_length:
+            difference = max_length - len(encoded_inputs["input_ids"])
+            if return_attention_mask:
+                encoded_inputs["attention_mask"] = [1] * len(encoded_inputs["input_ids"]) + [0] * difference
+            if return_token_type_ids:
+                encoded_inputs["token_type_ids"] += [self.pad_token_type_id] * difference
+            if return_special_tokens_mask:
+                encoded_inputs["special_tokens_mask"] += [1] * difference
+
+            encoded_inputs["input_ids"] += [self.pad_token_id] * difference
+        elif return_attention_mask:
+            encoded_inputs["attention_mask"] = [1] * len(encoded_inputs["input_ids"])
+
         return encoded_inputs
 
     def truncate_sequences(self, ids, pair_ids=None, num_tokens_to_remove=0, truncation_strategy='longest_first', stride=0):
diff --git a/transformers/tokenization_xlnet.py b/transformers/tokenization_xlnet.py
index a4f1a6e3ba..3ea71f4438 100644
--- a/transformers/tokenization_xlnet.py
+++ b/transformers/tokenization_xlnet.py
@@ -74,6 +74,7 @@ class XLNetTokenizer(PreTrainedTokenizer):
 
         self.max_len_single_sentence = self.max_len - 2  # take into account special tokens
         self.max_len_sentences_pair = self.max_len - 3  # take into account special tokens
+        self._pad_token_type_id = 3
 
         try:
             import sentencepiece as spm

From a7dafe2f41222469797f1a67232961d67bd2e519 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Thu, 21 Nov 2019 11:30:40 -0500
Subject: [PATCH 069/505] Padding strategy (left and right) rather than boolean
 flag

---
 .../tests/tokenization_tests_commons.py       | 43 +++++++++++---
 transformers/tokenization_utils.py            | 58 ++++++++++++++-----
 2 files changed, 77 insertions(+), 24 deletions(-)

diff --git a/transformers/tests/tokenization_tests_commons.py b/transformers/tests/tokenization_tests_commons.py
index d5b70d5266..40d68d0ab2 100644
--- a/transformers/tests/tokenization_tests_commons.py
+++ b/transformers/tests/tokenization_tests_commons.py
@@ -343,21 +343,33 @@ class CommonTestCases:
             padding_size = 10
             padding_idx = tokenizer.pad_token_id
 
-            # Check that it correctly pads when a maximum length is specified along with the padding flag set to True
+            # RIGHT PADDING - Check that it correctly pads when a maximum length is specified along with the padding flag set to True
             encoded_sequence = tokenizer.encode(sequence)
             sequence_length = len(encoded_sequence)
-            padded_sequence = tokenizer.encode(sequence, max_length=sequence_length + padding_size, pad_to_max_length=True)
+            padded_sequence = tokenizer.encode(sequence, max_length=sequence_length + padding_size, padding_strategy='right')
             padded_sequence_length = len(padded_sequence)
             assert sequence_length + padding_size == padded_sequence_length
             assert encoded_sequence + [padding_idx] * padding_size == padded_sequence
 
-            # Check that nothing is done when a maximum length is not specified
+            # LEFT PADDING - Check that it correctly pads when a maximum length is specified along with the padding flag set to True
             encoded_sequence = tokenizer.encode(sequence)
             sequence_length = len(encoded_sequence)
-            padded_sequence = tokenizer.encode(sequence, pad_to_max_length=True)
+            padded_sequence = tokenizer.encode(sequence, max_length=sequence_length + padding_size, padding_strategy='left')
             padded_sequence_length = len(padded_sequence)
-            assert sequence_length == padded_sequence_length
-            assert encoded_sequence == padded_sequence
+            assert sequence_length + padding_size == padded_sequence_length
+            assert [padding_idx] * padding_size + encoded_sequence == padded_sequence
+
+            # RIGHT & LEFT PADDING - Check that nothing is done when a maximum length is not specified
+            encoded_sequence = tokenizer.encode(sequence)
+            sequence_length = len(encoded_sequence)
+            padded_sequence_right = tokenizer.encode(sequence, padding_strategy='right')
+            padded_sequence_right_length = len(padded_sequence_right)
+            padded_sequence_left = tokenizer.encode(sequence, padding_strategy='left')
+            padded_sequence_left_length = len(padded_sequence_left)
+            assert sequence_length == padded_sequence_right_length
+            assert encoded_sequence == padded_sequence_right
+            assert sequence_length == padded_sequence_left_length
+            assert encoded_sequence == padded_sequence_left
 
         def test_encode_plus_with_padding(self):
             tokenizer = self.get_tokenizer()
@@ -374,7 +386,8 @@ class CommonTestCases:
             special_tokens_mask = encoded_sequence['special_tokens_mask']
             sequence_length = len(input_ids)
 
-            padded_sequence = tokenizer.encode_plus(sequence, max_length=sequence_length + padding_size, pad_to_max_length=True, return_special_tokens_mask=True)
+            # Test right padding
+            padded_sequence = tokenizer.encode_plus(sequence, max_length=sequence_length + padding_size, padding_strategy='right', return_special_tokens_mask=True)
             padded_input_ids = padded_sequence['input_ids']
             padded_token_type_ids = padded_sequence['token_type_ids']
             padded_attention_mask = padded_sequence['attention_mask']
@@ -385,4 +398,18 @@ class CommonTestCases:
             assert input_ids + [padding_idx] * padding_size == padded_input_ids
             assert token_type_ids + [token_type_padding_idx] * padding_size == padded_token_type_ids
             assert attention_mask + [0] * padding_size == padded_attention_mask 
-            assert special_tokens_mask + [1] * padding_size == padded_special_tokens_mask 
\ No newline at end of file
+            assert special_tokens_mask + [1] * padding_size == padded_special_tokens_mask 
+
+            # Test left padding
+            padded_sequence = tokenizer.encode_plus(sequence, max_length=sequence_length + padding_size, padding_strategy='left', return_special_tokens_mask=True)
+            padded_input_ids = padded_sequence['input_ids']
+            padded_token_type_ids = padded_sequence['token_type_ids']
+            padded_attention_mask = padded_sequence['attention_mask']
+            padded_special_tokens_mask = padded_sequence['special_tokens_mask']
+            padded_sequence_length = len(padded_input_ids)
+
+            assert sequence_length + padding_size == padded_sequence_length
+            assert [padding_idx] * padding_size + input_ids == padded_input_ids
+            assert [token_type_padding_idx] * padding_size + token_type_ids == padded_token_type_ids
+            assert [0] * padding_size + attention_mask == padded_attention_mask 
+            assert [1] * padding_size + special_tokens_mask == padded_special_tokens_mask 
\ No newline at end of file
diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index 3214699e12..dbbabd0e1a 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -702,7 +702,7 @@ class PreTrainedTokenizer(object):
                max_length=None,
                stride=0,
                truncation_strategy='longest_first',
-               pad_to_max_length=False,
+               padding_strategy=None,
                return_tensors=None,
                **kwargs):
         """
@@ -729,8 +729,12 @@ class PreTrainedTokenizer(object):
                 - 'only_first': Only truncate the first sequence
                 - 'only_second': Only truncate the second sequence
                 - 'do_not_truncate': Does not truncate (raise an error if the input sequence is longer than max_length)
-            pad_to_max_length: if set to `True`, the returned sequences will be padded according to the model's 
+            padding_strategy: if set to a strategy, the returned sequences will be padded according to the model's 
                 padding index, up to their max length. If no max length is specified, no padding is done.
+                The strategies are handled by the following strings:
+                - 'left': pads on the left of the sequences
+                - 'right': pads on the right of the sequences   
+                Defaults to None: no padding.
             return_tensors: (optional) can be set to 'tf' or 'pt' to return respectively TensorFlow tf.constant
                 or PyTorch torch.Tensor instead of a list of python integers.
             **kwargs: passed to the `self.tokenize()` method
@@ -741,7 +745,7 @@ class PreTrainedTokenizer(object):
                                           add_special_tokens=add_special_tokens,
                                           stride=stride,
                                           truncation_strategy=truncation_strategy,
-                                          pad_to_max_length=pad_to_max_length,
+                                          padding_strategy=padding_strategy,
                                           return_tensors=return_tensors,
                                           **kwargs)
 
@@ -754,7 +758,7 @@ class PreTrainedTokenizer(object):
                     max_length=None,
                     stride=0,
                     truncation_strategy='longest_first',
-                    pad_to_max_length=False,
+                    padding_strategy=None,
                     return_tensors=None,
                     return_token_type_ids=True,
                     return_attention_mask=True,
@@ -784,8 +788,12 @@ class PreTrainedTokenizer(object):
                 - 'only_first': Only truncate the first sequence
                 - 'only_second': Only truncate the second sequence
                 - 'do_not_truncate': Does not truncate (raise an error if the input sequence is longer than max_length)
-            pad_to_max_length: if set to `True`, the returned sequences will be padded according to the model's 
+            padding_strategy: if set to a strategy, the returned sequences will be padded according to the model's 
                 padding index, up to their max length. If no max length is specified, no padding is done.
+                The strategies are handled by the following strings:
+                - 'left': pads on the left of the sequences
+                - 'right': pads on the right of the sequences   
+                Defaults to None: no padding.
             return_tensors: (optional) can be set to 'tf' or 'pt' to return respectively TensorFlow tf.constant
                 or PyTorch torch.Tensor instead of a list of python integers.
             return_token_type_ids: (optional) Set to False to avoid returning token_type_ids (default True).
@@ -833,7 +841,7 @@ class PreTrainedTokenizer(object):
         return self.prepare_for_model(first_ids,
                                       pair_ids=second_ids,
                                       max_length=max_length,
-                                      pad_to_max_length=pad_to_max_length,
+                                      padding_strategy=padding_strategy,
                                       add_special_tokens=add_special_tokens,
                                       stride=stride,
                                       truncation_strategy=truncation_strategy,
@@ -845,7 +853,7 @@ class PreTrainedTokenizer(object):
 
     def prepare_for_model(self, ids, pair_ids=None, max_length=None, add_special_tokens=True, stride=0,
                           truncation_strategy='longest_first',
-                          pad_to_max_length=False,
+                          padding_strategy=None,
                           return_tensors=None,
                           return_token_type_ids=True,
                           return_attention_mask=True,
@@ -873,8 +881,12 @@ class PreTrainedTokenizer(object):
                 - 'only_first': Only truncate the first sequence
                 - 'only_second': Only truncate the second sequence
                 - 'do_not_truncate': Does not truncate (raise an error if the input sequence is longer than max_length)
-            pad_to_max_length: if set to `True`, the returned sequences will be padded according to the model's 
+            padding_strategy: if set to a strategy, the returned sequences will be padded according to the model's 
                 padding index, up to their max length. If no max length is specified, no padding is done.
+                The strategies are handled by the following strings:
+                - 'left': pads on the left of the sequences
+                - 'right': pads on the right of the sequences            
+                Defaults to None: no padding.
             return_tensors: (optional) can be set to 'tf' or 'pt' to return respectively TensorFlow tf.constant
                 or PyTorch torch.Tensor instead of a list of python integers.
             return_token_type_ids: (optional) Set to False to avoid returning token_type_ids (default True).
@@ -943,16 +955,30 @@ class PreTrainedTokenizer(object):
                            "for this model ({} > {}). Running this sequence through the model will result in "
                            "indexing errors".format(len(ids), self.max_len))
                            
-        if pad_to_max_length and max_length and len(encoded_inputs["input_ids"]) < max_length:
+        if padding_strategy is not None and max_length and len(encoded_inputs["input_ids"]) < max_length:
             difference = max_length - len(encoded_inputs["input_ids"])
-            if return_attention_mask:
-                encoded_inputs["attention_mask"] = [1] * len(encoded_inputs["input_ids"]) + [0] * difference
-            if return_token_type_ids:
-                encoded_inputs["token_type_ids"] += [self.pad_token_type_id] * difference
-            if return_special_tokens_mask:
-                encoded_inputs["special_tokens_mask"] += [1] * difference
 
-            encoded_inputs["input_ids"] += [self.pad_token_id] * difference
+            if padding_strategy == 'right':
+                if return_attention_mask:
+                    encoded_inputs["attention_mask"] = [1] * len(encoded_inputs["input_ids"]) + [0] * difference
+                if return_token_type_ids:
+                    encoded_inputs["token_type_ids"] = encoded_inputs["token_type_ids"] + [self.pad_token_type_id] * difference
+                if return_special_tokens_mask:
+                    encoded_inputs["special_tokens_mask"] = encoded_inputs["special_tokens_mask"] + [1] * difference
+                encoded_inputs["input_ids"] = encoded_inputs["input_ids"] + [self.pad_token_id] * difference
+
+            elif padding_strategy == 'left':
+                if return_attention_mask:
+                    encoded_inputs["attention_mask"] =  [0] * difference + [1] * len(encoded_inputs["input_ids"])
+                if return_token_type_ids:
+                    encoded_inputs["token_type_ids"] = [self.pad_token_type_id] * difference + encoded_inputs["token_type_ids"]
+                if return_special_tokens_mask:
+                    encoded_inputs["special_tokens_mask"] = [1] * difference + encoded_inputs["special_tokens_mask"]
+                encoded_inputs["input_ids"] = [self.pad_token_id] * difference + encoded_inputs["input_ids"]
+
+            else:
+                raise ValueError("Invalid padding strategy:" + str(padding_strategy))
+            
         elif return_attention_mask:
             encoded_inputs["attention_mask"] = [1] * len(encoded_inputs["input_ids"])
 

From a5a8a6175fb5cc1e993366add026ba06386bde10 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Thu, 21 Nov 2019 19:18:20 -0500
Subject: [PATCH 070/505] Works for BERT

---
 transformers/data/processors/squad.py | 507 ++++++++++++++++++++++----
 1 file changed, 432 insertions(+), 75 deletions(-)

diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index 1900e9f0ce..a0f2408a16 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -3,6 +3,7 @@ import collections
 import logging
 import os
 import json
+import numpy as np
 
 from ...tokenization_bert import BasicTokenizer, whitespace_tokenize
 from .utils import DataProcessor, InputExample, InputFeatures
@@ -13,10 +14,68 @@ if is_tf_available():
 
 logger = logging.getLogger(__name__)
 
+def _improve_answer_span(doc_tokens, input_start, input_end, tokenizer,
+                        orig_answer_text):
+    """Returns tokenized answer spans that better match the annotated answer."""
+    tok_answer_text = " ".join(tokenizer.tokenize(orig_answer_text))
+
+    for new_start in range(input_start, input_end + 1):
+        for new_end in range(input_end, new_start - 1, -1):
+            text_span = " ".join(doc_tokens[new_start:(new_end + 1)])
+            if text_span == tok_answer_text:
+                return (new_start, new_end)
+
+    return (input_start, input_end)
+
+def _check_is_max_context(doc_spans, cur_span_index, position):
+    """Check if this is the 'max context' doc span for the token."""
+    best_score = None
+    best_span_index = None
+    for (span_index, doc_span) in enumerate(doc_spans):
+        end = doc_span.start + doc_span.length - 1
+        if position < doc_span.start:
+            continue
+        if position > end:
+            continue
+        num_left_context = position - doc_span.start
+        num_right_context = end - position
+        score = min(num_left_context, num_right_context) + 0.01 * doc_span.length
+        if best_score is None or score > best_score:
+            best_score = score
+            best_span_index = span_index
+
+    return cur_span_index == best_span_index
+
+
+def _new_check_is_max_context(doc_spans, cur_span_index, position):
+    """Check if this is the 'max context' doc span for the token."""
+    # if len(doc_spans) == 1:
+        # return True
+    best_score = None
+    best_span_index = None
+    for (span_index, doc_span) in enumerate(doc_spans):
+        end = doc_span["start"] + doc_span["length"] - 1
+        if position < doc_span["start"]:
+            continue
+        if position > end:
+            continue
+        num_left_context = position - doc_span["start"]
+        num_right_context = end - position
+        score = min(num_left_context, num_right_context) + 0.01 * doc_span["length"]
+        if best_score is None or score > best_score:
+            best_score = score
+            best_span_index = span_index
+
+    return cur_span_index == best_span_index
+
+def _is_whitespace(c):
+    if c == " " or c == "\t" or c == "\r" or c == "\n" or ord(c) == 0x202F:
+        return True
+    return False
 
 def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
                                        doc_stride, max_query_length, is_training,
-                                       cls_token_at_end=False,
+                                       cls_token_at_end=True,
                                        cls_token='[CLS]', sep_token='[SEP]', pad_token=0,
                                        sequence_a_segment_id=0, sequence_b_segment_id=1,
                                        cls_token_segment_id=0, pad_token_segment_id=0,
@@ -24,57 +83,184 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
                                        sequence_a_is_doc=False):
     """Loads a data file into a list of `InputBatch`s."""
 
-    # Defining helper methods
-    def _improve_answer_span(doc_tokens, input_start, input_end, tokenizer,
-                         orig_answer_text):
-        """Returns tokenized answer spans that better match the annotated answer."""
-        tok_answer_text = " ".join(tokenizer.tokenize(orig_answer_text))
-
-        for new_start in range(input_start, input_end + 1):
-            for new_end in range(input_end, new_start - 1, -1):
-                text_span = " ".join(doc_tokens[new_start:(new_end + 1)])
-                if text_span == tok_answer_text:
-                    return (new_start, new_end)
-
-        return (input_start, input_end)
-    def _check_is_max_context(doc_spans, cur_span_index, position):
-        """Check if this is the 'max context' doc span for the token."""
-        best_score = None
-        best_span_index = None
-        for (span_index, doc_span) in enumerate(doc_spans):
-            end = doc_span.start + doc_span.length - 1
-            if position < doc_span.start:
-                continue
-            if position > end:
-                continue
-            num_left_context = position - doc_span.start
-            num_right_context = end - position
-            score = min(num_left_context, num_right_context) + 0.01 * doc_span.length
-            if best_score is None or score > best_score:
-                best_score = score
-                best_span_index = span_index
-
-        return cur_span_index == best_span_index
-    
+    # Defining helper methods    
     unique_id = 1000000000
 
     features = []
+    new_features = []
     for (example_index, example) in enumerate(tqdm(examples)):
-        query_tokens = tokenizer.tokenize(example.question_text)
 
-        if len(query_tokens) > max_query_length:
-            query_tokens = query_tokens[0:max_query_length]
+        doc_tokens = []
+        char_to_word_offset = []
+        prev_is_whitespace = True
+
+        # Split on whitespace so that different tokens may be attributed to their original position.
+        for c in example.context_text:
+            if _is_whitespace(c):
+                prev_is_whitespace = True
+            else:
+                if prev_is_whitespace:
+                    doc_tokens.append(c)
+                else:
+                    doc_tokens[-1] += c
+                prev_is_whitespace = False
+            char_to_word_offset.append(len(doc_tokens) - 1)
+
+        if is_training:
+            # Get start and end position
+            answer_length = len(example.answer_text)
+            start_position = char_to_word_offset[example.start_position]
+            end_position = char_to_word_offset[example.start_position + answer_length - 1]
+
+            # If the answer cannot be found in the text, then skip this example.
+            actual_text = " ".join(doc_tokens[start_position:(end_position + 1)])
+            cleaned_answer_text = " ".join(whitespace_tokenize(example.answer_text))
+            if actual_text.find(cleaned_answer_text) == -1:
+                logger.warning("Could not find answer: '%s' vs. '%s'", actual_text, cleaned_answer_text)
+                continue
 
         tok_to_orig_index = []
         orig_to_tok_index = []
         all_doc_tokens = []
-        for (i, token) in enumerate(example.doc_tokens):
+        for (i, token) in enumerate(doc_tokens):
             orig_to_tok_index.append(len(all_doc_tokens))
             sub_tokens = tokenizer.tokenize(token)
             for sub_token in sub_tokens:
                 tok_to_orig_index.append(i)
                 all_doc_tokens.append(sub_token)
 
+        spans = []
+        
+        truncated_query = tokenizer.encode(example.question_text, add_special_tokens=False, max_length=max_query_length)
+        sequence_added_tokens = tokenizer.max_len - tokenizer.max_len_single_sentence 
+        sequence_pair_added_tokens = tokenizer.max_len - tokenizer.max_len_sentences_pair 
+
+        encoded_dict = tokenizer.encode_plus(
+            truncated_query, 
+            all_doc_tokens, 
+            max_length=max_seq_length, 
+            padding_strategy='right',
+            stride=max_seq_length - doc_stride - len(truncated_query) - sequence_pair_added_tokens,
+            return_overflowing_tokens=True, 
+            truncation_strategy='only_second'
+        )
+
+        ids = encoded_dict['input_ids']
+        print("Ids computes; position of the first padding", ids.index(tokenizer.pad_token_id) if tokenizer.pad_token_id in ids else None)
+        non_padded_ids = ids[:ids.index(tokenizer.pad_token_id)] if tokenizer.pad_token_id in ids else ids
+        paragraph_len = min(len(all_doc_tokens) - len(spans) * doc_stride, max_seq_length - len(truncated_query) - sequence_pair_added_tokens)
+        tokens = tokenizer.convert_ids_to_tokens(non_padded_ids)
+
+        token_to_orig_map = {}
+        for i in range(paragraph_len):
+            token_to_orig_map[len(truncated_query) + sequence_added_tokens + i] = tok_to_orig_index[0 + i]
+
+        encoded_dict["paragraph_len"] = paragraph_len
+        encoded_dict["tokens"] = tokens
+        encoded_dict["token_to_orig_map"] = token_to_orig_map
+        encoded_dict["truncated_query_with_special_tokens_length"] = len(truncated_query) + sequence_added_tokens
+        encoded_dict["token_is_max_context"] = {}
+        encoded_dict["start"] = 0
+        encoded_dict["length"] = paragraph_len
+
+        spans.append(encoded_dict)
+        print("YESSIR", len(spans) * doc_stride < len(all_doc_tokens), "overflowing_tokens" in encoded_dict)
+        while len(spans) * doc_stride < len(all_doc_tokens) and "overflowing_tokens" in encoded_dict:
+            
+            overflowing_tokens = encoded_dict['overflowing_tokens']
+
+            print("OVERFLOW", len(overflowing_tokens))
+
+            encoded_dict = tokenizer.encode_plus(
+                truncated_query, 
+                overflowing_tokens, 
+                max_length=max_seq_length, 
+                return_overflowing_tokens=True, 
+                padding_strategy='right',
+                stride=max_seq_length - doc_stride - len(truncated_query) - sequence_pair_added_tokens,
+                truncation_strategy='only_second'
+            )
+
+            ids = encoded_dict['input_ids']
+            print("Ids computes; position of the first padding", ids.index(tokenizer.pad_token_id) if tokenizer.pad_token_id in ids else None)
+
+            # Length of the document without the query
+            paragraph_len = min(len(all_doc_tokens) - len(spans) * doc_stride, max_seq_length - len(truncated_query) - sequence_pair_added_tokens)
+
+            non_padded_ids = encoded_dict['input_ids'][:encoded_dict['input_ids'].index(tokenizer.pad_token_id)]
+            tokens = tokenizer.convert_ids_to_tokens(non_padded_ids)
+
+            token_to_orig_map = {}
+            for i in range(paragraph_len):
+                token_to_orig_map[len(truncated_query) + sequence_added_tokens + i] = tok_to_orig_index[len(spans) * doc_stride + i]
+
+            encoded_dict["paragraph_len"] = paragraph_len
+            encoded_dict["tokens"] = tokens
+            encoded_dict["token_to_orig_map"] = token_to_orig_map
+            encoded_dict["truncated_query_with_special_tokens_length"] = len(truncated_query) + sequence_added_tokens
+            encoded_dict["token_is_max_context"] = {}
+            encoded_dict["start"] = len(spans) * doc_stride
+            encoded_dict["length"] = paragraph_len
+
+            # split_token_index = doc_span.start + i
+            # token_to_orig_map[len(tokens)] = tok_to_orig_index[split_token_index]
+
+            # is_max_context = _check_is_max_context(doc_spans, doc_span_index,
+            #                                         split_token_index)
+            # token_is_max_context[len(tokens)] = is_max_context
+            # tokens.append(all_doc_tokens[split_token_index])
+
+            spans.append(encoded_dict)
+
+        for doc_span_index in range(len(spans)):
+            for j in range(spans[doc_span_index]["paragraph_len"]):
+                is_max_context = _new_check_is_max_context(spans, doc_span_index, doc_span_index * doc_stride + j)
+                index = spans[doc_span_index]["truncated_query_with_special_tokens_length"] + j
+                spans[doc_span_index]["token_is_max_context"][index] = is_max_context
+
+        print("new span", len(spans))
+        for span in spans:
+            # Identify the position of the CLS token
+            cls_index = span['input_ids'].index(tokenizer.cls_token_id)
+
+            # p_mask: mask with 1 for token than cannot be in the answer (0 for token which can be in an answer)
+            # Original TF implem also keep the classification token (set to 0) (not sure why...)
+            p_mask = np.array(span['token_type_ids'])
+
+            # Convert all SEP indices to '0' before inversion
+            p_mask[np.where(np.array(span["input_ids"]) == tokenizer.sep_token_id)[0]] = 0
+
+            # Limit positive values to one
+            p_mask = 1 - np.minimum(p_mask, 1)
+
+            # Set the CLS index to '0'
+            p_mask[cls_index] = 0
+
+            print("new features length", len(new_features))
+
+            new_features.append(NewSquadFeatures(
+                span['input_ids'],
+                span['attention_mask'],
+                span['token_type_ids'],
+                cls_index,
+                p_mask.tolist(),
+
+                example_index=example_index,
+                unique_id=unique_id,
+                paragraph_len=span['paragraph_len'],
+                token_is_max_context=span["token_is_max_context"],
+                tokens=span["tokens"],
+                token_to_orig_map=span["token_to_orig_map"]
+            ))
+
+            unique_id += 1
+
+        # tokenize ...
+        query_tokens = tokenizer.tokenize(example.question_text)
+
+        if len(query_tokens) > max_query_length:
+            query_tokens = query_tokens[0:max_query_length]
+
         tok_start_position = None
         tok_end_position = None
         if is_training and example.is_impossible:
@@ -82,7 +268,7 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
             tok_end_position = -1
         if is_training and not example.is_impossible:
             tok_start_position = orig_to_tok_index[example.start_position]
-            if example.end_position < len(example.doc_tokens) - 1:
+            if example.end_position < len(doc_tokens) - 1:
                 tok_end_position = orig_to_tok_index[example.end_position + 1] - 1
             else:
                 tok_end_position = len(all_doc_tokens) - 1
@@ -101,14 +287,19 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
         doc_spans = []
         start_offset = 0
         while start_offset < len(all_doc_tokens):
+            print("OLD DOC CREATION BEGIN", start_offset, len(all_doc_tokens))
             length = len(all_doc_tokens) - start_offset
             if length > max_tokens_for_doc:
                 length = max_tokens_for_doc
             doc_spans.append(_DocSpan(start=start_offset, length=length))
             if start_offset + length == len(all_doc_tokens):
+                print("Done with this doc span, breaking out.", start_offset, length)
                 break
+            print("CHOOSING OFFSET", length, doc_stride)
             start_offset += min(length, doc_stride)
+            print("OLD DOC CREATION END", start_offset)
 
+        print("old span", len(doc_spans))
         for (doc_span_index, doc_span) in enumerate(doc_spans):
             tokens = []
             token_to_orig_map = {}
@@ -183,18 +374,20 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
             # tokens are attended to.
             input_mask = [1 if mask_padding_with_zero else 0] * len(input_ids)
 
+
+            
             # Zero-pad up to the sequence length.
             while len(input_ids) < max_seq_length:
                 input_ids.append(pad_token)
                 input_mask.append(0 if mask_padding_with_zero else 1)
                 segment_ids.append(pad_token_segment_id)
                 p_mask.append(1)
-
+            print("[OLD] Ids computed; position of the first padding", input_ids.index(tokenizer.pad_token_id) if tokenizer.pad_token_id in input_ids else None)
             assert len(input_ids) == max_seq_length
             assert len(input_mask) == max_seq_length
             assert len(segment_ids) == max_seq_length
 
-            span_is_impossible = example.is_impossible
+            span_is_impossible = example.is_impossible if hasattr(example, "is_impossible") else False
             start_position = None
             end_position = None
             if is_training and not span_is_impossible:
@@ -222,31 +415,32 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
                 start_position = cls_index
                 end_position = cls_index
 
-            if example_index < 20:
-                logger.info("*** Example ***")
-                logger.info("unique_id: %s" % (unique_id))
-                logger.info("example_index: %s" % (example_index))
-                logger.info("doc_span_index: %s" % (doc_span_index))
-                logger.info("tokens: %s" % " ".join(tokens))
-                logger.info("token_to_orig_map: %s" % " ".join([
-                    "%d:%d" % (x, y) for (x, y) in token_to_orig_map.items()]))
-                logger.info("token_is_max_context: %s" % " ".join([
-                    "%d:%s" % (x, y) for (x, y) in token_is_max_context.items()
-                ]))
-                logger.info("input_ids: %s" % " ".join([str(x) for x in input_ids]))
-                logger.info(
-                    "input_mask: %s" % " ".join([str(x) for x in input_mask]))
-                logger.info(
-                    "segment_ids: %s" % " ".join([str(x) for x in segment_ids]))
-                if is_training and span_is_impossible:
-                    logger.info("impossible example")
-                if is_training and not span_is_impossible:
-                    answer_text = " ".join(tokens[start_position:(end_position + 1)])
-                    logger.info("start_position: %d" % (start_position))
-                    logger.info("end_position: %d" % (end_position))
-                    logger.info(
-                        "answer: %s" % (answer_text))
+            # if example_index < 20:
+            #     logger.info("*** Example ***")
+            #     logger.info("unique_id: %s" % (unique_id))
+            #     logger.info("example_index: %s" % (example_index))
+            #     logger.info("doc_span_index: %s" % (doc_span_index))
+            #     logger.info("tokens: %s" % str(tokens))
+            #     logger.info("token_to_orig_map: %s" % " ".join([
+            #         "%d:%d" % (x, y) for (x, y) in token_to_orig_map.items()]))
+            #     logger.info("token_is_max_context: %s" % " ".join([
+            #         "%d:%s" % (x, y) for (x, y) in token_is_max_context.items()
+            #     ]))
+            #     logger.info("input_ids: %s" % " ".join([str(x) for x in input_ids]))
+            #     logger.info(
+            #         "input_mask: %s" % " ".join([str(x) for x in input_mask]))
+            #     logger.info(
+            #         "segment_ids: %s" % " ".join([str(x) for x in segment_ids]))
+            #     if is_training and span_is_impossible:
+            #         logger.info("impossible example")
+            #     if is_training and not span_is_impossible:
+            #         answer_text = " ".join(tokens[start_position:(end_position + 1)])
+            #         logger.info("start_position: %d" % (start_position))
+            #         logger.info("end_position: %d" % (end_position))
+            #         logger.info(
+            #             "answer: %s" % (answer_text))
 
+            print("features length", len(features))
             features.append(
                 SquadFeatures(
                     unique_id=unique_id,
@@ -266,7 +460,48 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
                     is_impossible=span_is_impossible))
             unique_id += 1
 
-    return features
+        assert len(features) == len(new_features)
+
+    assert len(features) == len(new_features)
+    for i in range(len(features)):
+        print(i)
+        feature, new_feature = features[i], new_features[i]
+        
+        input_ids = feature.input_ids
+        input_mask = feature.input_mask
+        segment_ids = feature.segment_ids
+        cls_index = feature.cls_index
+        p_mask = feature.p_mask
+        example_index = feature.example_index
+        paragraph_len = feature.paragraph_len
+        token_is_max_context = feature.token_is_max_context
+        tokens = feature.tokens
+        token_to_orig_map = feature.token_to_orig_map
+              
+        new_input_ids = new_feature.input_ids
+        new_input_mask = new_feature.attention_mask
+        new_segment_ids = new_feature.token_type_ids
+        new_cls_index = new_feature.cls_index
+        new_p_mask = new_feature.p_mask
+        new_example_index = new_feature.example_index
+        new_paragraph_len = new_feature.paragraph_len
+        new_token_is_max_context = new_feature.token_is_max_context
+        new_tokens = new_feature.tokens
+        new_token_to_orig_map = new_feature.token_to_orig_map
+
+        assert input_ids == new_input_ids
+        assert input_mask == new_input_mask
+        assert segment_ids == new_segment_ids
+        assert cls_index == new_cls_index
+        assert p_mask == new_p_mask
+        assert example_index == new_example_index
+        assert paragraph_len == new_paragraph_len
+        assert token_is_max_context == new_token_is_max_context
+        assert tokens == new_tokens
+        assert token_to_orig_map == new_token_to_orig_map
+
+
+    return new_features
 
 
 def read_squad_examples(input_file, is_training, version_2_with_negative):
@@ -347,6 +582,124 @@ def read_squad_examples(input_file, is_training, version_2_with_negative):
     return examples
 
 
+class SquadV1Processor(DataProcessor):
+    """Processor for the SQuAD data set."""
+
+    def get_example_from_tensor_dict(self, tensor_dict):
+        """See base class."""
+        return NewSquadExample(
+            tensor_dict['id'].numpy(),
+            tensor_dict['question'].numpy().decode('utf-8'),
+            tensor_dict['context'].numpy().decode('utf-8'),
+            tensor_dict['answers']['text'].numpy().decode('utf-8'),
+            tensor_dict['answers']['answers_start'].numpy().decode('utf-8'),
+            tensor_dict['title'].numpy().decode('utf-8')
+        )
+
+    def get_train_examples(self, data_dir):
+        """See base class."""
+        with open(os.path.join(data_dir, "train-v1.1.json"), "r", encoding='utf-8') as reader:
+            input_data = json.load(reader)["data"]
+        return self._create_examples(input_data, "train")
+
+    def get_dev_examples(self, data_dir):
+        """See base class."""
+        with open(os.path.join(data_dir, "dev-v1.1.json"), "r", encoding='utf-8') as reader:
+            input_data = json.load(reader)["data"]
+        return self._create_examples(input_data, "dev")
+
+    def get_labels(self):
+        """See base class."""
+        return ["0", "1"]
+
+    def _create_examples(self, input_data, set_type):
+        """Creates examples for the training and dev sets."""
+        
+        is_training = set_type == "train"
+        examples = []
+        for entry in input_data:
+            title = entry['title']
+            for paragraph in entry["paragraphs"]:
+                context_text = paragraph["context"]
+                for qa in paragraph["qas"]:
+                    qas_id = qa["id"]
+                    question_text = qa["question"]
+                    start_position = None
+                    answer_text = None
+                    if is_training:
+                        if (len(qa["answers"]) != 1):
+                            raise ValueError(
+                                "For training, each question should have exactly 1 answer.")
+                        answer = qa["answers"][0]
+                        answer_text = answer['text']
+                        start_position = answer['answer_start']
+
+                    example = NewSquadExample(
+                        qas_id=qas_id,
+                        question_text=question_text,
+                        context_text=context_text,
+                        answer_text=answer_text,
+                        start_position=start_position,
+                        title=title
+                    )
+                    examples.append(example)
+        return examples
+        
+
+
+class NewSquadExample(object):
+    """
+    A single training/test example for the Squad dataset, as loaded from disk.
+    """
+
+    def __init__(self,
+                 qas_id,
+                 question_text,
+                 context_text,
+                 answer_text,
+                 start_position,
+                 title):
+        self.qas_id = qas_id
+        self.question_text = question_text
+        self.context_text = context_text
+        self.answer_text = answer_text
+        self.start_position = start_position
+        self.title = title
+
+
+class NewSquadFeatures(object):
+    """
+    Single squad example features to be fed to a model.
+    Those features are model-specific.
+    """
+
+    def __init__(self,
+                 input_ids,
+                 attention_mask,
+                 token_type_ids,
+                 cls_index,
+                 p_mask,
+                 
+                 example_index,
+                 unique_id,
+                 paragraph_len,
+                 token_is_max_context,
+                 tokens,
+                 token_to_orig_map
+        ):
+        self.input_ids = input_ids 
+        self.attention_mask = attention_mask
+        self.token_type_ids = token_type_ids
+        self.cls_index = cls_index
+        self.p_mask = p_mask
+
+        self.example_index = example_index
+        self.unique_id = unique_id
+        self.paragraph_len = paragraph_len
+        self.token_is_max_context = token_is_max_context
+        self.tokens = tokens
+        self.token_to_orig_map = token_to_orig_map
+
 class SquadExample(object):
     """
     A single training/test example for the Squad dataset.
@@ -423,18 +776,22 @@ class SquadFeatures(object):
         self.is_impossible = is_impossible
 
     def __eq__(self, other):
-        return self.cls_index == other.cls_index and \
-                self.doc_span_index == other.doc_span_index and \
-                self.end_position == other.end_position and \
-                self.example_index == other.example_index and \
+        print(self.example_index == other.example_index)
+        print(self.input_ids == other.input_ids)
+        print(self.input_mask == other.attention_mask)
+        print(self.p_mask == other.p_mask)
+        print(self.paragraph_len == other.paragraph_len)
+        print(self.segment_ids == other.token_type_ids)
+        print(self.token_is_max_context == other.token_is_max_context)
+        print(self.token_to_orig_map == other.token_to_orig_map)
+        print(self.tokens == other.tokens)
+
+        return self.example_index == other.example_index and \
                 self.input_ids == other.input_ids and \
-                self.input_mask == other.input_mask and \
-                self.is_impossible == other.is_impossible and \
+                self.input_mask == other.attention_mask and \
                 self.p_mask == other.p_mask and \
                 self.paragraph_len == other.paragraph_len and \
-                self.segment_ids == other.segment_ids and \
-                self.start_position == other.start_position and \
+                self.segment_ids == other.token_type_ids and \
                 self.token_is_max_context == other.token_is_max_context and \
                 self.token_to_orig_map == other.token_to_orig_map and \
-                self.tokens == other.tokens and \
-                self.unique_id == other.unique_id
\ No newline at end of file
+                self.tokens == other.tokens
\ No newline at end of file

From c3ba6452377f085d0f59e15b97ac247bca24367e Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Fri, 22 Nov 2019 14:36:49 -0500
Subject: [PATCH 071/505] Works for XLNet

---
 examples/run_squad.py                 | 38 ++++--------
 transformers/data/processors/squad.py | 84 +++++++++++++--------------
 2 files changed, 50 insertions(+), 72 deletions(-)

diff --git a/examples/run_squad.py b/examples/run_squad.py
index d4219c3096..634b566a46 100644
--- a/examples/run_squad.py
+++ b/examples/run_squad.py
@@ -16,6 +16,7 @@
 """ Finetuning the library models for question-answering on SQuAD (DistilBERT, Bert, XLM, XLNet)."""
 
 from __future__ import absolute_import, division, print_function
+from transformers.data.processors.squad import SquadV1Processor
 
 import argparse
 import logging
@@ -46,8 +47,7 @@ from transformers import (WEIGHTS_NAME, BertConfig,
 
 from transformers import AdamW, get_linear_schedule_with_warmup, squad_convert_examples_to_features, read_squad_examples as sread_squad_examples
 
-from utils_squad import (read_squad_examples, convert_examples_to_features,
-                         RawResult, write_predictions,
+from utils_squad import (RawResult, write_predictions,
                          RawResultExtended, write_predictions_extended)
 
 # The follwing import is the official SQuAD evaluation script (2.0).
@@ -289,7 +289,6 @@ def evaluate(args, model, tokenizer, prefix=""):
     results = evaluate_on_squad(evaluate_options)
     return results
 
-
 def load_and_cache_examples(args, tokenizer, evaluate=False, output_examples=False):
     if args.local_rank not in [-1, 0] and not evaluate:
         torch.distributed.barrier()  # Make sure only the first process in distributed training process the dataset, and the others will use the cache
@@ -308,9 +307,11 @@ def load_and_cache_examples(args, tokenizer, evaluate=False, output_examples=Fal
         examples = read_squad_examples(input_file=input_file,
                                                 is_training=not evaluate,
                                                 version_2_with_negative=args.version_2_with_negative)
-
-        examples = examples[:10]
-        features = convert_examples_to_features(examples=examples,
+        keep_n_examples = 1000
+        processor = SquadV1Processor()
+        values = processor.get_dev_examples("examples/squad")
+        examples = values[:keep_n_examples]
+        features = squad_convert_examples_to_features(examples=exampless,
                                                 tokenizer=tokenizer,
                                                 max_seq_length=args.max_seq_length,
                                                 doc_stride=args.doc_stride,
@@ -320,29 +321,10 @@ def load_and_cache_examples(args, tokenizer, evaluate=False, output_examples=Fal
                                                 pad_token_segment_id=3 if args.model_type in ['xlnet'] else 0,
                                                 cls_token_at_end=True if args.model_type in ['xlnet'] else False,
                                                 sequence_a_is_doc=True if args.model_type in ['xlnet'] else False)
-
-        exampless = sread_squad_examples(input_file=input_file,
-                                                is_training=not evaluate,
-                                                version_2_with_negative=args.version_2_with_negative)
-        exampless = exampless[:10]
-        features2 = squad_convert_examples_to_features(examples=exampless,
-                                                tokenizer=tokenizer,
-                                                max_seq_length=args.max_seq_length,
-                                                doc_stride=args.doc_stride,
-                                                max_query_length=args.max_query_length,
-                                                is_training=not evaluate,
-                                                cls_token_segment_id=2 if args.model_type in ['xlnet'] else 0,
-                                                pad_token_segment_id=3 if args.model_type in ['xlnet'] else 0,
-                                                cls_token_at_end=True if args.model_type in ['xlnet'] else False,
-                                                sequence_a_is_doc=True if args.model_type in ['xlnet'] else False)
-
-        print(features2)
-
-        for i in range(len(features)):
-            assert features[i] == features2[i]
-            print("Equal")
-
         print("DONE")
+
+        import sys
+        sys.exit()
         
         if args.local_rank in [-1, 0]:
             logger.info("Saving features into cached file %s", cached_features_file)
diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index a0f2408a16..fb3d2ae4d4 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -83,6 +83,9 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
                                        sequence_a_is_doc=False):
     """Loads a data file into a list of `InputBatch`s."""
 
+    cls_token = tokenizer.cls_token
+    sep_token = tokenizer.sep_token
+
     # Defining helper methods    
     unique_id = 1000000000
 
@@ -136,24 +139,24 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
         sequence_pair_added_tokens = tokenizer.max_len - tokenizer.max_len_sentences_pair 
 
         encoded_dict = tokenizer.encode_plus(
-            truncated_query, 
-            all_doc_tokens, 
+            truncated_query if not sequence_a_is_doc else all_doc_tokens, 
+            all_doc_tokens if not sequence_a_is_doc else truncated_query, 
             max_length=max_seq_length, 
             padding_strategy='right',
             stride=max_seq_length - doc_stride - len(truncated_query) - sequence_pair_added_tokens,
             return_overflowing_tokens=True, 
-            truncation_strategy='only_second'
+            truncation_strategy='only_second' if not sequence_a_is_doc else 'only_first'
         )
 
         ids = encoded_dict['input_ids']
-        print("Ids computes; position of the first padding", ids.index(tokenizer.pad_token_id) if tokenizer.pad_token_id in ids else None)
         non_padded_ids = ids[:ids.index(tokenizer.pad_token_id)] if tokenizer.pad_token_id in ids else ids
         paragraph_len = min(len(all_doc_tokens) - len(spans) * doc_stride, max_seq_length - len(truncated_query) - sequence_pair_added_tokens)
         tokens = tokenizer.convert_ids_to_tokens(non_padded_ids)
 
         token_to_orig_map = {}
         for i in range(paragraph_len):
-            token_to_orig_map[len(truncated_query) + sequence_added_tokens + i] = tok_to_orig_index[0 + i]
+            index = len(truncated_query) + sequence_added_tokens + i if not sequence_a_is_doc else i 
+            token_to_orig_map[index] = tok_to_orig_index[0 + i]
 
         encoded_dict["paragraph_len"] = paragraph_len
         encoded_dict["tokens"] = tokens
@@ -164,35 +167,40 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
         encoded_dict["length"] = paragraph_len
 
         spans.append(encoded_dict)
-        print("YESSIR", len(spans) * doc_stride < len(all_doc_tokens), "overflowing_tokens" in encoded_dict)
+        # print("YESSIR", len(spans) * doc_stride < len(all_doc_tokens), "overflowing_tokens" in encoded_dict)
+
         while len(spans) * doc_stride < len(all_doc_tokens) and "overflowing_tokens" in encoded_dict:
-            
-            overflowing_tokens = encoded_dict['overflowing_tokens']
-
-            print("OVERFLOW", len(overflowing_tokens))
-
+            overflowing_tokens = encoded_dict["overflowing_tokens"]
             encoded_dict = tokenizer.encode_plus(
-                truncated_query, 
-                overflowing_tokens, 
+                truncated_query if not sequence_a_is_doc else overflowing_tokens, 
+                overflowing_tokens if not sequence_a_is_doc else truncated_query, 
                 max_length=max_seq_length, 
                 return_overflowing_tokens=True, 
                 padding_strategy='right',
                 stride=max_seq_length - doc_stride - len(truncated_query) - sequence_pair_added_tokens,
-                truncation_strategy='only_second'
+                truncation_strategy='only_second' if not sequence_a_is_doc else 'only_first'
             )
-
             ids = encoded_dict['input_ids']
-            print("Ids computes; position of the first padding", ids.index(tokenizer.pad_token_id) if tokenizer.pad_token_id in ids else None)
+            # print("Ids computes; position of the first padding", ids.index(tokenizer.pad_token_id) if tokenizer.pad_token_id in ids else None)
+
+            # print(encoded_dict["input_ids"].index(tokenizer.pad_token_id) if tokenizer.pad_token_id in encoded_dict["input_ids"] else None)
+            # print(len(spans) * doc_stride, len(all_doc_tokens))
+            
 
             # Length of the document without the query
             paragraph_len = min(len(all_doc_tokens) - len(spans) * doc_stride, max_seq_length - len(truncated_query) - sequence_pair_added_tokens)
 
-            non_padded_ids = encoded_dict['input_ids'][:encoded_dict['input_ids'].index(tokenizer.pad_token_id)]
+            if tokenizer.pad_token_id in encoded_dict['input_ids']: 
+                non_padded_ids = encoded_dict['input_ids'][:encoded_dict['input_ids'].index(tokenizer.pad_token_id)]
+            else:
+                non_padded_ids = encoded_dict['input_ids']
+
             tokens = tokenizer.convert_ids_to_tokens(non_padded_ids)
 
             token_to_orig_map = {}
             for i in range(paragraph_len):
-                token_to_orig_map[len(truncated_query) + sequence_added_tokens + i] = tok_to_orig_index[len(spans) * doc_stride + i]
+                index = len(truncated_query) + sequence_added_tokens + i if not sequence_a_is_doc else i 
+                token_to_orig_map[index] = tok_to_orig_index[len(spans) * doc_stride + i]
 
             encoded_dict["paragraph_len"] = paragraph_len
             encoded_dict["tokens"] = tokens
@@ -202,23 +210,14 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
             encoded_dict["start"] = len(spans) * doc_stride
             encoded_dict["length"] = paragraph_len
 
-            # split_token_index = doc_span.start + i
-            # token_to_orig_map[len(tokens)] = tok_to_orig_index[split_token_index]
-
-            # is_max_context = _check_is_max_context(doc_spans, doc_span_index,
-            #                                         split_token_index)
-            # token_is_max_context[len(tokens)] = is_max_context
-            # tokens.append(all_doc_tokens[split_token_index])
-
             spans.append(encoded_dict)
 
         for doc_span_index in range(len(spans)):
             for j in range(spans[doc_span_index]["paragraph_len"]):
                 is_max_context = _new_check_is_max_context(spans, doc_span_index, doc_span_index * doc_stride + j)
-                index = spans[doc_span_index]["truncated_query_with_special_tokens_length"] + j
+                index = j if sequence_a_is_doc else spans[doc_span_index]["truncated_query_with_special_tokens_length"] + j
                 spans[doc_span_index]["token_is_max_context"][index] = is_max_context
 
-        print("new span", len(spans))
         for span in spans:
             # Identify the position of the CLS token
             cls_index = span['input_ids'].index(tokenizer.cls_token_id)
@@ -227,17 +226,17 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
             # Original TF implem also keep the classification token (set to 0) (not sure why...)
             p_mask = np.array(span['token_type_ids'])
 
-            # Convert all SEP indices to '0' before inversion
-            p_mask[np.where(np.array(span["input_ids"]) == tokenizer.sep_token_id)[0]] = 0
+            p_mask = np.minimum(p_mask, 1)
 
-            # Limit positive values to one
-            p_mask = 1 - np.minimum(p_mask, 1)
+            if not sequence_a_is_doc:
+                # Limit positive values to one
+                p_mask = 1 - p_mask
+
+            p_mask[np.where(np.array(span["input_ids"]) == tokenizer.sep_token_id)[0]] = 1
 
             # Set the CLS index to '0'
             p_mask[cls_index] = 0
 
-            print("new features length", len(new_features))
-
             new_features.append(NewSquadFeatures(
                 span['input_ids'],
                 span['attention_mask'],
@@ -287,19 +286,15 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
         doc_spans = []
         start_offset = 0
         while start_offset < len(all_doc_tokens):
-            print("OLD DOC CREATION BEGIN", start_offset, len(all_doc_tokens))
             length = len(all_doc_tokens) - start_offset
             if length > max_tokens_for_doc:
                 length = max_tokens_for_doc
+            # print("Start offset is", start_offset, len(all_doc_tokens), "length is", length)
             doc_spans.append(_DocSpan(start=start_offset, length=length))
             if start_offset + length == len(all_doc_tokens):
-                print("Done with this doc span, breaking out.", start_offset, length)
                 break
-            print("CHOOSING OFFSET", length, doc_stride)
             start_offset += min(length, doc_stride)
-            print("OLD DOC CREATION END", start_offset)
 
-        print("old span", len(doc_spans))
         for (doc_span_index, doc_span) in enumerate(doc_spans):
             tokens = []
             token_to_orig_map = {}
@@ -382,7 +377,7 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
                 input_mask.append(0 if mask_padding_with_zero else 1)
                 segment_ids.append(pad_token_segment_id)
                 p_mask.append(1)
-            print("[OLD] Ids computed; position of the first padding", input_ids.index(tokenizer.pad_token_id) if tokenizer.pad_token_id in input_ids else None)
+
             assert len(input_ids) == max_seq_length
             assert len(input_mask) == max_seq_length
             assert len(segment_ids) == max_seq_length
@@ -440,7 +435,6 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
             #         logger.info(
             #             "answer: %s" % (answer_text))
 
-            print("features length", len(features))
             features.append(
                 SquadFeatures(
                     unique_id=unique_id,
@@ -464,10 +458,9 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
 
     assert len(features) == len(new_features)
     for i in range(len(features)):
-        print(i)
         feature, new_feature = features[i], new_features[i]
         
-        input_ids = feature.input_ids
+        input_ids = [f if f not in [3,4,5] else 0 for f in feature.input_ids ]
         input_mask = feature.input_mask
         segment_ids = feature.segment_ids
         cls_index = feature.cls_index
@@ -478,7 +471,7 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
         tokens = feature.tokens
         token_to_orig_map = feature.token_to_orig_map
               
-        new_input_ids = new_feature.input_ids
+        new_input_ids = [f if f not in [3,4,5] else 0 for f in new_feature.input_ids]
         new_input_mask = new_feature.attention_mask
         new_segment_ids = new_feature.token_type_ids
         new_cls_index = new_feature.cls_index
@@ -497,6 +490,9 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
         assert example_index == new_example_index
         assert paragraph_len == new_paragraph_len
         assert token_is_max_context == new_token_is_max_context
+
+        tokens = [t if tokenizer.convert_tokens_to_ids(t) is not tokenizer.unk_token_id else tokenizer.unk_token for t in tokens]
+
         assert tokens == new_tokens
         assert token_to_orig_map == new_token_to_orig_map
 

From e0e55bc550a16289763b4f656790e30ed86e428f Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Fri, 22 Nov 2019 16:18:18 -0500
Subject: [PATCH 072/505] Manage training example & refactor the refactor

---
 transformers/data/processors/squad.py | 368 ++++----------------------
 1 file changed, 51 insertions(+), 317 deletions(-)

diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index fb3d2ae4d4..3d8f48c1bb 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -92,31 +92,14 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
     features = []
     new_features = []
     for (example_index, example) in enumerate(tqdm(examples)):
-
-        doc_tokens = []
-        char_to_word_offset = []
-        prev_is_whitespace = True
-
-        # Split on whitespace so that different tokens may be attributed to their original position.
-        for c in example.context_text:
-            if _is_whitespace(c):
-                prev_is_whitespace = True
-            else:
-                if prev_is_whitespace:
-                    doc_tokens.append(c)
-                else:
-                    doc_tokens[-1] += c
-                prev_is_whitespace = False
-            char_to_word_offset.append(len(doc_tokens) - 1)
-
         if is_training:
             # Get start and end position
             answer_length = len(example.answer_text)
-            start_position = char_to_word_offset[example.start_position]
-            end_position = char_to_word_offset[example.start_position + answer_length - 1]
+            start_position = example.start_position
+            end_position = example.end_position
 
             # If the answer cannot be found in the text, then skip this example.
-            actual_text = " ".join(doc_tokens[start_position:(end_position + 1)])
+            actual_text = " ".join(example.doc_tokens[start_position:(end_position + 1)])
             cleaned_answer_text = " ".join(whitespace_tokenize(example.answer_text))
             if actual_text.find(cleaned_answer_text) == -1:
                 logger.warning("Could not find answer: '%s' vs. '%s'", actual_text, cleaned_answer_text)
@@ -125,7 +108,7 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
         tok_to_orig_index = []
         orig_to_tok_index = []
         all_doc_tokens = []
-        for (i, token) in enumerate(doc_tokens):
+        for (i, token) in enumerate(example.doc_tokens):
             orig_to_tok_index.append(len(all_doc_tokens))
             sub_tokens = tokenizer.tokenize(token)
             for sub_token in sub_tokens:
@@ -138,56 +121,19 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
         sequence_added_tokens = tokenizer.max_len - tokenizer.max_len_single_sentence 
         sequence_pair_added_tokens = tokenizer.max_len - tokenizer.max_len_sentences_pair 
 
-        encoded_dict = tokenizer.encode_plus(
-            truncated_query if not sequence_a_is_doc else all_doc_tokens, 
-            all_doc_tokens if not sequence_a_is_doc else truncated_query, 
-            max_length=max_seq_length, 
-            padding_strategy='right',
-            stride=max_seq_length - doc_stride - len(truncated_query) - sequence_pair_added_tokens,
-            return_overflowing_tokens=True, 
-            truncation_strategy='only_second' if not sequence_a_is_doc else 'only_first'
-        )
-
-        ids = encoded_dict['input_ids']
-        non_padded_ids = ids[:ids.index(tokenizer.pad_token_id)] if tokenizer.pad_token_id in ids else ids
-        paragraph_len = min(len(all_doc_tokens) - len(spans) * doc_stride, max_seq_length - len(truncated_query) - sequence_pair_added_tokens)
-        tokens = tokenizer.convert_ids_to_tokens(non_padded_ids)
-
-        token_to_orig_map = {}
-        for i in range(paragraph_len):
-            index = len(truncated_query) + sequence_added_tokens + i if not sequence_a_is_doc else i 
-            token_to_orig_map[index] = tok_to_orig_index[0 + i]
-
-        encoded_dict["paragraph_len"] = paragraph_len
-        encoded_dict["tokens"] = tokens
-        encoded_dict["token_to_orig_map"] = token_to_orig_map
-        encoded_dict["truncated_query_with_special_tokens_length"] = len(truncated_query) + sequence_added_tokens
-        encoded_dict["token_is_max_context"] = {}
-        encoded_dict["start"] = 0
-        encoded_dict["length"] = paragraph_len
-
-        spans.append(encoded_dict)
-        # print("YESSIR", len(spans) * doc_stride < len(all_doc_tokens), "overflowing_tokens" in encoded_dict)
-
-        while len(spans) * doc_stride < len(all_doc_tokens) and "overflowing_tokens" in encoded_dict:
-            overflowing_tokens = encoded_dict["overflowing_tokens"]
+        span_doc_tokens = all_doc_tokens
+        while len(spans) * doc_stride < len(all_doc_tokens):
+            
             encoded_dict = tokenizer.encode_plus(
-                truncated_query if not sequence_a_is_doc else overflowing_tokens, 
-                overflowing_tokens if not sequence_a_is_doc else truncated_query, 
+                truncated_query if not sequence_a_is_doc else span_doc_tokens, 
+                span_doc_tokens if not sequence_a_is_doc else truncated_query, 
                 max_length=max_seq_length, 
                 return_overflowing_tokens=True, 
                 padding_strategy='right',
                 stride=max_seq_length - doc_stride - len(truncated_query) - sequence_pair_added_tokens,
                 truncation_strategy='only_second' if not sequence_a_is_doc else 'only_first'
             )
-            ids = encoded_dict['input_ids']
-            # print("Ids computes; position of the first padding", ids.index(tokenizer.pad_token_id) if tokenizer.pad_token_id in ids else None)
 
-            # print(encoded_dict["input_ids"].index(tokenizer.pad_token_id) if tokenizer.pad_token_id in encoded_dict["input_ids"] else None)
-            # print(len(spans) * doc_stride, len(all_doc_tokens))
-            
-
-            # Length of the document without the query
             paragraph_len = min(len(all_doc_tokens) - len(spans) * doc_stride, max_seq_length - len(truncated_query) - sequence_pair_added_tokens)
 
             if tokenizer.pad_token_id in encoded_dict['input_ids']: 
@@ -212,6 +158,10 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
 
             spans.append(encoded_dict)
 
+            if "overflowing_tokens" not in encoded_dict:
+                break
+            span_doc_tokens = encoded_dict["overflowing_tokens"]
+
         for doc_span_index in range(len(spans)):
             for j in range(spans[doc_span_index]["paragraph_len"]):
                 is_max_context = _new_check_is_max_context(spans, doc_span_index, doc_span_index * doc_stride + j)
@@ -254,249 +204,6 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
 
             unique_id += 1
 
-        # tokenize ...
-        query_tokens = tokenizer.tokenize(example.question_text)
-
-        if len(query_tokens) > max_query_length:
-            query_tokens = query_tokens[0:max_query_length]
-
-        tok_start_position = None
-        tok_end_position = None
-        if is_training and example.is_impossible:
-            tok_start_position = -1
-            tok_end_position = -1
-        if is_training and not example.is_impossible:
-            tok_start_position = orig_to_tok_index[example.start_position]
-            if example.end_position < len(doc_tokens) - 1:
-                tok_end_position = orig_to_tok_index[example.end_position + 1] - 1
-            else:
-                tok_end_position = len(all_doc_tokens) - 1
-            (tok_start_position, tok_end_position) = _improve_answer_span(
-                all_doc_tokens, tok_start_position, tok_end_position, tokenizer,
-                example.orig_answer_text)
-
-        # The -3 accounts for [CLS], [SEP] and [SEP]
-        max_tokens_for_doc = max_seq_length - len(query_tokens) - 3
-
-        # We can have documents that are longer than the maximum sequence length.
-        # To deal with this we do a sliding window approach, where we take chunks
-        # of the up to our max length with a stride of `doc_stride`.
-        _DocSpan = collections.namedtuple(  # pylint: disable=invalid-name
-            "DocSpan", ["start", "length"])
-        doc_spans = []
-        start_offset = 0
-        while start_offset < len(all_doc_tokens):
-            length = len(all_doc_tokens) - start_offset
-            if length > max_tokens_for_doc:
-                length = max_tokens_for_doc
-            # print("Start offset is", start_offset, len(all_doc_tokens), "length is", length)
-            doc_spans.append(_DocSpan(start=start_offset, length=length))
-            if start_offset + length == len(all_doc_tokens):
-                break
-            start_offset += min(length, doc_stride)
-
-        for (doc_span_index, doc_span) in enumerate(doc_spans):
-            tokens = []
-            token_to_orig_map = {}
-            token_is_max_context = {}
-            segment_ids = []
-
-            # p_mask: mask with 1 for token than cannot be in the answer (0 for token which can be in an answer)
-            # Original TF implem also keep the classification token (set to 0) (not sure why...)
-            p_mask = []
-
-            # CLS token at the beginning
-            if not cls_token_at_end:
-                tokens.append(cls_token)
-                segment_ids.append(cls_token_segment_id)
-                p_mask.append(0)
-                cls_index = 0
-
-            # XLNet: P SEP Q SEP CLS
-            # Others: CLS Q SEP P SEP
-            if not sequence_a_is_doc:
-                # Query
-                tokens += query_tokens
-                segment_ids += [sequence_a_segment_id] * len(query_tokens)
-                p_mask += [1] * len(query_tokens)
-
-                # SEP token
-                tokens.append(sep_token)
-                segment_ids.append(sequence_a_segment_id)
-                p_mask.append(1)
-
-            # Paragraph
-            for i in range(doc_span.length):
-                split_token_index = doc_span.start + i
-                token_to_orig_map[len(tokens)] = tok_to_orig_index[split_token_index]
-
-                is_max_context = _check_is_max_context(doc_spans, doc_span_index,
-                                                       split_token_index)
-                token_is_max_context[len(tokens)] = is_max_context
-                tokens.append(all_doc_tokens[split_token_index])
-                if not sequence_a_is_doc:
-                    segment_ids.append(sequence_b_segment_id)
-                else:
-                    segment_ids.append(sequence_a_segment_id)
-                p_mask.append(0)
-            paragraph_len = doc_span.length
-
-            if sequence_a_is_doc:
-                # SEP token
-                tokens.append(sep_token)
-                segment_ids.append(sequence_a_segment_id)
-                p_mask.append(1)
-
-                tokens += query_tokens
-                segment_ids += [sequence_b_segment_id] * len(query_tokens)
-                p_mask += [1] * len(query_tokens)
-
-            # SEP token
-            tokens.append(sep_token)
-            segment_ids.append(sequence_b_segment_id)
-            p_mask.append(1)
-
-            # CLS token at the end
-            if cls_token_at_end:
-                tokens.append(cls_token)
-                segment_ids.append(cls_token_segment_id)
-                p_mask.append(0)
-                cls_index = len(tokens) - 1  # Index of classification token
-
-            input_ids = tokenizer.convert_tokens_to_ids(tokens)
-
-            # The mask has 1 for real tokens and 0 for padding tokens. Only real
-            # tokens are attended to.
-            input_mask = [1 if mask_padding_with_zero else 0] * len(input_ids)
-
-
-            
-            # Zero-pad up to the sequence length.
-            while len(input_ids) < max_seq_length:
-                input_ids.append(pad_token)
-                input_mask.append(0 if mask_padding_with_zero else 1)
-                segment_ids.append(pad_token_segment_id)
-                p_mask.append(1)
-
-            assert len(input_ids) == max_seq_length
-            assert len(input_mask) == max_seq_length
-            assert len(segment_ids) == max_seq_length
-
-            span_is_impossible = example.is_impossible if hasattr(example, "is_impossible") else False
-            start_position = None
-            end_position = None
-            if is_training and not span_is_impossible:
-                # For training, if our document chunk does not contain an annotation
-                # we throw it out, since there is nothing to predict.
-                doc_start = doc_span.start
-                doc_end = doc_span.start + doc_span.length - 1
-                out_of_span = False
-                if not (tok_start_position >= doc_start and
-                        tok_end_position <= doc_end):
-                    out_of_span = True
-                if out_of_span:
-                    start_position = 0
-                    end_position = 0
-                    span_is_impossible = True
-                else:
-                    if sequence_a_is_doc:
-                        doc_offset = 0
-                    else:
-                        doc_offset = len(query_tokens) + 2
-                    start_position = tok_start_position - doc_start + doc_offset
-                    end_position = tok_end_position - doc_start + doc_offset
-
-            if is_training and span_is_impossible:
-                start_position = cls_index
-                end_position = cls_index
-
-            # if example_index < 20:
-            #     logger.info("*** Example ***")
-            #     logger.info("unique_id: %s" % (unique_id))
-            #     logger.info("example_index: %s" % (example_index))
-            #     logger.info("doc_span_index: %s" % (doc_span_index))
-            #     logger.info("tokens: %s" % str(tokens))
-            #     logger.info("token_to_orig_map: %s" % " ".join([
-            #         "%d:%d" % (x, y) for (x, y) in token_to_orig_map.items()]))
-            #     logger.info("token_is_max_context: %s" % " ".join([
-            #         "%d:%s" % (x, y) for (x, y) in token_is_max_context.items()
-            #     ]))
-            #     logger.info("input_ids: %s" % " ".join([str(x) for x in input_ids]))
-            #     logger.info(
-            #         "input_mask: %s" % " ".join([str(x) for x in input_mask]))
-            #     logger.info(
-            #         "segment_ids: %s" % " ".join([str(x) for x in segment_ids]))
-            #     if is_training and span_is_impossible:
-            #         logger.info("impossible example")
-            #     if is_training and not span_is_impossible:
-            #         answer_text = " ".join(tokens[start_position:(end_position + 1)])
-            #         logger.info("start_position: %d" % (start_position))
-            #         logger.info("end_position: %d" % (end_position))
-            #         logger.info(
-            #             "answer: %s" % (answer_text))
-
-            features.append(
-                SquadFeatures(
-                    unique_id=unique_id,
-                    example_index=example_index,
-                    doc_span_index=doc_span_index,
-                    tokens=tokens,
-                    token_to_orig_map=token_to_orig_map,
-                    token_is_max_context=token_is_max_context,
-                    input_ids=input_ids,
-                    input_mask=input_mask,
-                    segment_ids=segment_ids,
-                    cls_index=cls_index,
-                    p_mask=p_mask,
-                    paragraph_len=paragraph_len,
-                    start_position=start_position,
-                    end_position=end_position,
-                    is_impossible=span_is_impossible))
-            unique_id += 1
-
-        assert len(features) == len(new_features)
-
-    assert len(features) == len(new_features)
-    for i in range(len(features)):
-        feature, new_feature = features[i], new_features[i]
-        
-        input_ids = [f if f not in [3,4,5] else 0 for f in feature.input_ids ]
-        input_mask = feature.input_mask
-        segment_ids = feature.segment_ids
-        cls_index = feature.cls_index
-        p_mask = feature.p_mask
-        example_index = feature.example_index
-        paragraph_len = feature.paragraph_len
-        token_is_max_context = feature.token_is_max_context
-        tokens = feature.tokens
-        token_to_orig_map = feature.token_to_orig_map
-              
-        new_input_ids = [f if f not in [3,4,5] else 0 for f in new_feature.input_ids]
-        new_input_mask = new_feature.attention_mask
-        new_segment_ids = new_feature.token_type_ids
-        new_cls_index = new_feature.cls_index
-        new_p_mask = new_feature.p_mask
-        new_example_index = new_feature.example_index
-        new_paragraph_len = new_feature.paragraph_len
-        new_token_is_max_context = new_feature.token_is_max_context
-        new_tokens = new_feature.tokens
-        new_token_to_orig_map = new_feature.token_to_orig_map
-
-        assert input_ids == new_input_ids
-        assert input_mask == new_input_mask
-        assert segment_ids == new_segment_ids
-        assert cls_index == new_cls_index
-        assert p_mask == new_p_mask
-        assert example_index == new_example_index
-        assert paragraph_len == new_paragraph_len
-        assert token_is_max_context == new_token_is_max_context
-
-        tokens = [t if tokenizer.convert_tokens_to_ids(t) is not tokenizer.unk_token_id else tokenizer.unk_token for t in tokens]
-
-        assert tokens == new_tokens
-        assert token_to_orig_map == new_token_to_orig_map
-
-
     return new_features
 
 
@@ -592,35 +299,35 @@ class SquadV1Processor(DataProcessor):
             tensor_dict['title'].numpy().decode('utf-8')
         )
 
-    def get_train_examples(self, data_dir):
+    def get_train_examples(self, data_dir, only_first=None):
         """See base class."""
         with open(os.path.join(data_dir, "train-v1.1.json"), "r", encoding='utf-8') as reader:
             input_data = json.load(reader)["data"]
-        return self._create_examples(input_data, "train")
+        return self._create_examples(input_data, "train", only_first)
 
-    def get_dev_examples(self, data_dir):
+    def get_dev_examples(self, data_dir, only_first=None):
         """See base class."""
         with open(os.path.join(data_dir, "dev-v1.1.json"), "r", encoding='utf-8') as reader:
             input_data = json.load(reader)["data"]
-        return self._create_examples(input_data, "dev")
+        return self._create_examples(input_data, "dev", only_first)
 
     def get_labels(self):
         """See base class."""
         return ["0", "1"]
 
-    def _create_examples(self, input_data, set_type):
+    def _create_examples(self, input_data, set_type, only_first=None):
         """Creates examples for the training and dev sets."""
         
         is_training = set_type == "train"
         examples = []
-        for entry in input_data:
+        for entry in tqdm(input_data):
             title = entry['title']
             for paragraph in entry["paragraphs"]:
                 context_text = paragraph["context"]
                 for qa in paragraph["qas"]:
                     qas_id = qa["id"]
                     question_text = qa["question"]
-                    start_position = None
+                    start_position_character = None
                     answer_text = None
                     if is_training:
                         if (len(qa["answers"]) != 1):
@@ -628,17 +335,20 @@ class SquadV1Processor(DataProcessor):
                                 "For training, each question should have exactly 1 answer.")
                         answer = qa["answers"][0]
                         answer_text = answer['text']
-                        start_position = answer['answer_start']
+                        start_position_character = answer['answer_start']
 
                     example = NewSquadExample(
                         qas_id=qas_id,
                         question_text=question_text,
                         context_text=context_text,
                         answer_text=answer_text,
-                        start_position=start_position,
+                        start_position_character=start_position_character,
                         title=title
                     )
                     examples.append(example)
+
+                    if only_first is not None and len(examples) > only_first:
+                        return examples
         return examples
         
 
@@ -653,14 +363,38 @@ class NewSquadExample(object):
                  question_text,
                  context_text,
                  answer_text,
-                 start_position,
+                 start_position_character,
                  title):
         self.qas_id = qas_id
         self.question_text = question_text
         self.context_text = context_text
         self.answer_text = answer_text
-        self.start_position = start_position
         self.title = title
+        self.is_impossible = False
+
+        doc_tokens = []
+        char_to_word_offset = []
+        prev_is_whitespace = True
+
+        # Split on whitespace so that different tokens may be attributed to their original position.
+        for c in self.context_text:
+            if _is_whitespace(c):
+                prev_is_whitespace = True
+            else:
+                if prev_is_whitespace:
+                    doc_tokens.append(c)
+                else:
+                    doc_tokens[-1] += c
+                prev_is_whitespace = False
+            char_to_word_offset.append(len(doc_tokens) - 1)
+
+        self.doc_tokens = doc_tokens
+        self.char_to_word_offset = char_to_word_offset
+
+        # Start end end positions only has a value during evaluation.
+        if start_position_character is not None:
+            self.start_position = char_to_word_offset[start_position_character]
+            self.end_position = char_to_word_offset[start_position_character + len(answer_text) - 1]
 
 
 class NewSquadFeatures(object):

From 041a901f324eea7e7ee04b0f7a563c7ed5c8a03a Mon Sep 17 00:00:00 2001
From: Nikolay Korolev <korolevns98@gmail.com>
Date: Fri, 22 Nov 2019 22:56:43 +0300
Subject: [PATCH 073/505] Fix typo in documentation. toto -> to

---
 transformers/tokenization_gpt2.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/tokenization_gpt2.py b/transformers/tokenization_gpt2.py
index 4bec515903..a4798b0129 100644
--- a/transformers/tokenization_gpt2.py
+++ b/transformers/tokenization_gpt2.py
@@ -184,7 +184,7 @@ class GPT2Tokenizer(PreTrainedTokenizer):
         """ Tokenize a string.
             Args:
                 - add_prefix_space (boolean, default False):
-                    Begin the sentence with at least one space toto get invariance to word order in GPT-2 (and RoBERTa) tokenizers.
+                    Begin the sentence with at least one space to get invariance to word order in GPT-2 (and RoBERTa) tokenizers.
         """
         if add_prefix_space:
             text = ' ' + text

From 176cd1ce1b337134425b426207fbe155099c18b4 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Sat, 23 Nov 2019 11:18:54 -0500
Subject: [PATCH 074/505] [doc] homogenize instructions slightly

---
 README.md          | 5 ++---
 examples/README.md | 4 ++--
 2 files changed, 4 insertions(+), 5 deletions(-)

diff --git a/README.md b/README.md
index 9cc783fd71..52e74a6f80 100644
--- a/README.md
+++ b/README.md
@@ -89,13 +89,12 @@ pip install [--editable] .
 ### Run the examples
 
 Examples are included in the repository but are not shipped with the library.
-Therefore, in order to run the examples you will first need to clone the
-repository and install the bleeding edge version of the library. To do so, create a new virtual environment and follow these steps:
+Therefore, in order to run the latest versions of the examples you also need to install from source. To do so, create a new virtual environment and follow these steps:
 
 ```bash
 git clone git@github.com:huggingface/transformers
 cd transformers
-pip install .
+pip install [--editable] .
 ```
 
 ### Tests
diff --git a/examples/README.md b/examples/README.md
index 622fa07f8f..0c57990ea7 100644
--- a/examples/README.md
+++ b/examples/README.md
@@ -4,12 +4,12 @@ In this section a few examples are put together. All of these examples work for
 similar API between the different models.
 
 **Important**  
-To use the examples, execute the following steps in a new virtual environment:
+To run the latest versions of the examples, you have to install from source. Execute the following steps in a new virtual environment:
 
 ```bash
 git clone git@github.com:huggingface/transformers
 cd transformers
-pip install .
+pip install [--editable] .
 ```
 
 | Section                    | Description                                                                                                                                                |

From afaa33585109550f9ecaaee4e47f187aaaefedd0 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Sat, 23 Nov 2019 11:34:45 -0500
Subject: [PATCH 075/505] [doc] Fix assets urls

---
 docs/source/_static/js/custom.js | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/docs/source/_static/js/custom.js b/docs/source/_static/js/custom.js
index 2c7836fd20..ec804b3704 100644
--- a/docs/source/_static/js/custom.js
+++ b/docs/source/_static/js/custom.js
@@ -1,5 +1,5 @@
 function addIcon() {
-    const huggingFaceLogo = "https://huggingface.co/assets/transformers-docs/huggingface_logo.svg";
+    const huggingFaceLogo = "https://huggingface.co/landing/assets/transformers-docs/huggingface_logo.svg";
     const image = document.createElement("img");
     image.setAttribute("src", huggingFaceLogo);
 
@@ -24,10 +24,10 @@ function addCustomFooter() {
     social.classList.add("footer__Social");
 
     const imageDetails = [
-        { link: "https://huggingface.co", imageLink: "https://huggingface.co/assets/transformers-docs/website.svg" },
-        { link: "https://twitter.com/huggingface", imageLink: "https://huggingface.co/assets/transformers-docs/twitter.svg" },
-        { link: "https://github.com/huggingface", imageLink: "https://huggingface.co/assets/transformers-docs/github.svg" },
-        { link: "https://www.linkedin.com/company/huggingface/", imageLink: "https://huggingface.co/assets/transformers-docs/linkedin.svg" }
+        { link: "https://huggingface.co", imageLink: "https://huggingface.co/landing/assets/transformers-docs/website.svg" },
+        { link: "https://twitter.com/huggingface", imageLink: "https://huggingface.co/landing/assets/transformers-docs/twitter.svg" },
+        { link: "https://github.com/huggingface", imageLink: "https://huggingface.co/landing/assets/transformers-docs/github.svg" },
+        { link: "https://www.linkedin.com/company/huggingface/", imageLink: "https://huggingface.co/landing/assets/transformers-docs/linkedin.svg" }
     ];
 
     imageDetails.forEach(imageLinks => {

From 7485caefb09cb7f4c4b720b40ec69fed345a6b1c Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Mon, 25 Nov 2019 09:33:39 -0500
Subject: [PATCH 076/505] fix #1894

---
 examples/run_lm_finetuning.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/examples/run_lm_finetuning.py b/examples/run_lm_finetuning.py
index 52a1b75a65..aded521c1d 100644
--- a/examples/run_lm_finetuning.py
+++ b/examples/run_lm_finetuning.py
@@ -68,7 +68,7 @@ class TextDataset(Dataset):
         directory, filename = os.path.split(file_path)
         cached_features_file = os.path.join(directory, args.model_name_or_path + '_cached_lm_' + str(block_size) + '_' + filename)
 
-        if os.path.exists(cached_features_file):
+        if os.path.exists(cached_features_file) and not args.overwrite_cache:
             logger.info("Loading features from cached file %s", cached_features_file)
             with open(cached_features_file, 'rb') as handle:
                 self.examples = pickle.load(handle)

From 99f750d64e78d20fa5213ea11235b6a1b084481e Mon Sep 17 00:00:00 2001
From: Evpok Padding <evpok.padding@gmail.com>
Date: Thu, 21 Nov 2019 10:35:07 +0100
Subject: [PATCH 077/505] add Camembert models to modeling_auto

---
 transformers/modeling_auto.py | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/transformers/modeling_auto.py b/transformers/modeling_auto.py
index d98110d4bd..a5828148ad 100644
--- a/transformers/modeling_auto.py
+++ b/transformers/modeling_auto.py
@@ -27,6 +27,7 @@ from .modeling_xlnet import XLNetModel, XLNetLMHeadModel, XLNetForSequenceClassi
 from .modeling_xlm import XLMModel, XLMWithLMHeadModel, XLMForSequenceClassification, XLMForQuestionAnswering
 from .modeling_roberta import RobertaModel, RobertaForMaskedLM, RobertaForSequenceClassification
 from .modeling_distilbert import DistilBertModel, DistilBertForQuestionAnswering, DistilBertForMaskedLM, DistilBertForSequenceClassification
+from .modeling_camembert import CamembertModel, CamembertForMaskedLM, CamembertForSequenceClassification, CamembertForMultipleChoice
 
 from .modeling_utils import PreTrainedModel, SequenceSummary
 
@@ -48,6 +49,7 @@ class AutoModel(object):
         The base model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertModel (DistilBERT model)
+            - contains `camembert`: CamemBERTModel (CamemBERT model)
             - contains `roberta`: RobertaModel (RoBERTa model)
             - contains `bert`: BertModel (Bert model)
             - contains `openai-gpt`: OpenAIGPTModel (OpenAI GPT model)
@@ -71,6 +73,7 @@ class AutoModel(object):
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertModel (DistilBERT model)
+            - contains `camembert`: CamemBERTModel (CamemBERT model)
             - contains `roberta`: RobertaModel (RoBERTa model)
             - contains `bert`: BertModel (Bert model)
             - contains `openai-gpt`: OpenAIGPTModel (OpenAI GPT model)
@@ -138,6 +141,8 @@ class AutoModel(object):
         """
         if 'distilbert' in pretrained_model_name_or_path:
             return DistilBertModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'camembert' in pretrained_model_name_or_path:
+            return CamembertModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return RobertaModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'bert' in pretrained_model_name_or_path:
@@ -172,6 +177,7 @@ class AutoModelWithLMHead(object):
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertForMaskedLM (DistilBERT model)
+            - contains `camembert`: CamemBERTModel (CamemBERT model)
             - contains `roberta`: RobertaForMaskedLM (RoBERTa model)
             - contains `bert`: BertForMaskedLM (Bert model)
             - contains `openai-gpt`: OpenAIGPTLMHeadModel (OpenAI GPT model)
@@ -198,6 +204,7 @@ class AutoModelWithLMHead(object):
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertForMaskedLM (DistilBERT model)
+            - contains `camembert`: CamemBERTModel (CamemBERT model)
             - contains `roberta`: RobertaForMaskedLM (RoBERTa model)
             - contains `bert`: BertForMaskedLM (Bert model)
             - contains `openai-gpt`: OpenAIGPTLMHeadModel (OpenAI GPT model)
@@ -264,6 +271,8 @@ class AutoModelWithLMHead(object):
         """
         if 'distilbert' in pretrained_model_name_or_path:
             return DistilBertForMaskedLM.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        if 'camembert' in pretrained_model_name_or_path:
+            return CamembertForMaskedLM.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return RobertaForMaskedLM.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'bert' in pretrained_model_name_or_path:
@@ -298,6 +307,7 @@ class AutoModelForSequenceClassification(object):
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertForSequenceClassification (DistilBERT model)
+            - contains `camembert`: CamemBERTModel (CamemBERT model)
             - contains `roberta`: RobertaForSequenceClassification (RoBERTa model)
             - contains `bert`: BertForSequenceClassification (Bert model)
             - contains `xlnet`: XLNetForSequenceClassification (XLNet model)
@@ -320,6 +330,7 @@ class AutoModelForSequenceClassification(object):
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertForSequenceClassification (DistilBERT model)
+            - contains `camembert`: CamemBERTModel (CamemBERT model)
             - contains `roberta`: RobertaForSequenceClassification (RoBERTa model)
             - contains `bert`: BertForSequenceClassification (Bert model)
             - contains `xlnet`: XLNetForSequenceClassification (XLNet model)
@@ -383,6 +394,8 @@ class AutoModelForSequenceClassification(object):
         """
         if 'distilbert' in pretrained_model_name_or_path:
             return DistilBertForSequenceClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        if 'camembert' in pretrained_model_name_or_path:
+            return CamembertForSequenceClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return RobertaForSequenceClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'bert' in pretrained_model_name_or_path:

From c8eb8157b86291413fe8096217f4defc168aa73f Mon Sep 17 00:00:00 2001
From: Evpok Padding <evpok.padding@gmail.com>
Date: Thu, 21 Nov 2019 11:01:20 +0100
Subject: [PATCH 078/505] fix docstrings

---
 transformers/modeling_auto.py | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/transformers/modeling_auto.py b/transformers/modeling_auto.py
index a5828148ad..ce36f6dc4a 100644
--- a/transformers/modeling_auto.py
+++ b/transformers/modeling_auto.py
@@ -49,7 +49,7 @@ class AutoModel(object):
         The base model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertModel (DistilBERT model)
-            - contains `camembert`: CamemBERTModel (CamemBERT model)
+            - contains `camembert`: CamembertModel (CamemBERT model)
             - contains `roberta`: RobertaModel (RoBERTa model)
             - contains `bert`: BertModel (Bert model)
             - contains `openai-gpt`: OpenAIGPTModel (OpenAI GPT model)
@@ -73,7 +73,7 @@ class AutoModel(object):
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertModel (DistilBERT model)
-            - contains `camembert`: CamemBERTModel (CamemBERT model)
+            - contains `camembert`: CamembertModel (CamemBERT model)
             - contains `roberta`: RobertaModel (RoBERTa model)
             - contains `bert`: BertModel (Bert model)
             - contains `openai-gpt`: OpenAIGPTModel (OpenAI GPT model)
@@ -177,7 +177,7 @@ class AutoModelWithLMHead(object):
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertForMaskedLM (DistilBERT model)
-            - contains `camembert`: CamemBERTModel (CamemBERT model)
+            - contains `camembert`: CamembertForMaskedLM (CamemBERT model)
             - contains `roberta`: RobertaForMaskedLM (RoBERTa model)
             - contains `bert`: BertForMaskedLM (Bert model)
             - contains `openai-gpt`: OpenAIGPTLMHeadModel (OpenAI GPT model)
@@ -204,7 +204,7 @@ class AutoModelWithLMHead(object):
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertForMaskedLM (DistilBERT model)
-            - contains `camembert`: CamemBERTModel (CamemBERT model)
+            - contains `camembert`: CamembertForMaskedLM (CamemBERT model)
             - contains `roberta`: RobertaForMaskedLM (RoBERTa model)
             - contains `bert`: BertForMaskedLM (Bert model)
             - contains `openai-gpt`: OpenAIGPTLMHeadModel (OpenAI GPT model)
@@ -307,7 +307,7 @@ class AutoModelForSequenceClassification(object):
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertForSequenceClassification (DistilBERT model)
-            - contains `camembert`: CamemBERTModel (CamemBERT model)
+            - contains `camembert`: CamembertForSequenceClassification (CamemBERT model)
             - contains `roberta`: RobertaForSequenceClassification (RoBERTa model)
             - contains `bert`: BertForSequenceClassification (Bert model)
             - contains `xlnet`: XLNetForSequenceClassification (XLNet model)
@@ -330,7 +330,7 @@ class AutoModelForSequenceClassification(object):
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertForSequenceClassification (DistilBERT model)
-            - contains `camembert`: CamemBERTModel (CamemBERT model)
+            - contains `camembert`: CamembertForSequenceClassification (CamemBERT model)
             - contains `roberta`: RobertaForSequenceClassification (RoBERTa model)
             - contains `bert`: BertForSequenceClassification (Bert model)
             - contains `xlnet`: XLNetForSequenceClassification (XLNet model)

From fa963ecc59a1dea59c2d0e952b2c4483e1828176 Mon Sep 17 00:00:00 2001
From: Evpok Padding <evpok.padding@gmail.com>
Date: Thu, 21 Nov 2019 11:03:12 +0100
Subject: [PATCH 079/505] =?UTF-8?q?if=E2=86=92elif?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 transformers/modeling_auto.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/transformers/modeling_auto.py b/transformers/modeling_auto.py
index ce36f6dc4a..5866420001 100644
--- a/transformers/modeling_auto.py
+++ b/transformers/modeling_auto.py
@@ -271,7 +271,7 @@ class AutoModelWithLMHead(object):
         """
         if 'distilbert' in pretrained_model_name_or_path:
             return DistilBertForMaskedLM.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
-        if 'camembert' in pretrained_model_name_or_path:
+        elif 'camembert' in pretrained_model_name_or_path:
             return CamembertForMaskedLM.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return RobertaForMaskedLM.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
@@ -394,7 +394,7 @@ class AutoModelForSequenceClassification(object):
         """
         if 'distilbert' in pretrained_model_name_or_path:
             return DistilBertForSequenceClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
-        if 'camembert' in pretrained_model_name_or_path:
+        elif 'camembert' in pretrained_model_name_or_path:
             return CamembertForSequenceClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return RobertaForSequenceClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)

From 07bf43074f39379d8c7f6afcca105e69685f7531 Mon Sep 17 00:00:00 2001
From: Bilal Khan <bk@tinymanager.com>
Date: Thu, 21 Nov 2019 20:52:06 -0600
Subject: [PATCH 080/505] Fix GPT2 docstring

---
 transformers/tokenization_gpt2.py | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/transformers/tokenization_gpt2.py b/transformers/tokenization_gpt2.py
index a4798b0129..5fda709448 100644
--- a/transformers/tokenization_gpt2.py
+++ b/transformers/tokenization_gpt2.py
@@ -107,10 +107,10 @@ class GPT2Tokenizer(PreTrainedTokenizer):
     """
     GPT-2 BPE tokenizer. Peculiarities:
         - Byte-level Byte-Pair-Encoding
-        - Requires a space to start the input string => the encoding methods should be called with the
+        - Requires a space to start the input string => the encoding and tokenize methods should be called with the
           ``add_prefix_space`` flag set to ``True``.
-          Otherwise, this tokenizer ``encode`` and ``decode`` method will not conserve
-          the absence of a space at the beginning of a string: `tokenizer.decode(tokenizer.encode("Hello")) = " Hello"`
+          Otherwise, this tokenizer's ``encode``, ``decode``, and ``tokenize`` methods will not conserve
+          the spaces at the beginning of a string: `tokenizer.decode(tokenizer.encode(" Hello")) = "Hello"`
     """
     vocab_files_names = VOCAB_FILES_NAMES
     pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP

From aa92a184d2b92faadec975139ad55e2ae749362c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=C4=B0brahim=20Ethem=20Demirci?= <dmrc_wr@hotmail.com>
Date: Sat, 16 Nov 2019 16:49:37 +0300
Subject: [PATCH 081/505] resize model when special tokenizer present

---
 examples/run_lm_finetuning.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/examples/run_lm_finetuning.py b/examples/run_lm_finetuning.py
index aded521c1d..c33aa94a32 100644
--- a/examples/run_lm_finetuning.py
+++ b/examples/run_lm_finetuning.py
@@ -215,6 +215,7 @@ def train(args, train_dataset, model, tokenizer):
 
     global_step = 0
     tr_loss, logging_loss = 0.0, 0.0
+    model.resize_token_embeddings(len(tokenizer))
     model.zero_grad()
     train_iterator = trange(int(args.num_train_epochs), desc="Epoch", disable=args.local_rank not in [-1, 0])
     set_seed(args)  # Added here for reproducibility (even between python 2 and 3)

From 5d3b8daad2cc6287d30f03f8a96d0a1f7bc8d0dc Mon Sep 17 00:00:00 2001
From: manansanghi <52307004+manansanghi@users.noreply.github.com>
Date: Fri, 22 Nov 2019 10:31:54 -0800
Subject: [PATCH 082/505] Minor bug fixes on run_ner.py

---
 examples/run_ner.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/examples/run_ner.py b/examples/run_ner.py
index 127d63a6cd..1ab1236d94 100644
--- a/examples/run_ner.py
+++ b/examples/run_ner.py
@@ -127,7 +127,7 @@ def train(args, train_dataset, model, tokenizer, labels, pad_token_label_id):
                       "attention_mask": batch[1],
                       "labels": batch[3]}
             if args.model_type != "distilbert":
-                inputs["token_type_ids"]: batch[2] if args.model_type in ["bert", "xlnet"] else None  # XLM and RoBERTa don"t use segment_ids
+                inputs["token_type_ids"] = batch[2] if args.model_type in ["bert", "xlnet"] else None  # XLM and RoBERTa don"t use segment_ids
 
             outputs = model(**inputs)
             loss = outputs[0]  # model outputs are always tuple in pytorch-transformers (see doc)
@@ -217,7 +217,7 @@ def evaluate(args, model, tokenizer, labels, pad_token_label_id, mode, prefix=""
                       "attention_mask": batch[1],
                       "labels": batch[3]}
             if args.model_type != "distilbert":
-                inputs["token_type_ids"]: batch[2] if args.model_type in ["bert", "xlnet"] else None  # XLM and RoBERTa don"t use segment_ids
+                inputs["token_type_ids"] = batch[2] if args.model_type in ["bert", "xlnet"] else None  # XLM and RoBERTa don"t use segment_ids
             outputs = model(**inputs)
             tmp_eval_loss, logits = outputs[:2]
 

From 0669c1fcd15051ec6fe2d950079886faccf2fb33 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Mon, 25 Nov 2019 19:22:21 -0500
Subject: [PATCH 083/505] SQuAD v2 BERT + XLNet

---
 transformers/__init__.py                 |   2 +-
 transformers/data/__init__.py            |   2 +-
 transformers/data/processors/__init__.py |   2 +-
 transformers/data/processors/squad.py    | 180 +++++++++++------------
 4 files changed, 92 insertions(+), 94 deletions(-)

diff --git a/transformers/__init__.py b/transformers/__init__.py
index 9a767913b3..f3f81f1dbe 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -27,7 +27,7 @@ from .data import (is_sklearn_available,
                    glue_output_modes, glue_convert_examples_to_features,
                    glue_processors, glue_tasks_num_labels,
                    squad_convert_examples_to_features, SquadFeatures, 
-                   SquadExample, read_squad_examples)
+                   SquadExample)
 
 if is_sklearn_available():
     from .data import glue_compute_metrics
diff --git a/transformers/data/__init__.py b/transformers/data/__init__.py
index 50f2e768f4..b351bf625e 100644
--- a/transformers/data/__init__.py
+++ b/transformers/data/__init__.py
@@ -1,6 +1,6 @@
 from .processors import InputExample, InputFeatures, DataProcessor, SquadFeatures
 from .processors import glue_output_modes, glue_processors, glue_tasks_num_labels, glue_convert_examples_to_features
-from .processors import squad_convert_examples_to_features, SquadExample, read_squad_examples
+from .processors import squad_convert_examples_to_features, SquadExample
 
 from .metrics import is_sklearn_available
 if is_sklearn_available():
diff --git a/transformers/data/processors/__init__.py b/transformers/data/processors/__init__.py
index 924b4a1245..1e52776629 100644
--- a/transformers/data/processors/__init__.py
+++ b/transformers/data/processors/__init__.py
@@ -1,4 +1,4 @@
 from .utils import InputExample, InputFeatures, DataProcessor
 from .glue import glue_output_modes, glue_processors, glue_tasks_num_labels, glue_convert_examples_to_features
-from .squad import squad_convert_examples_to_features, SquadFeatures, SquadExample, read_squad_examples
+from .squad import squad_convert_examples_to_features, SquadFeatures, SquadExample
 
diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index 3d8f48c1bb..39ee00ae56 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -46,7 +46,6 @@ def _check_is_max_context(doc_spans, cur_span_index, position):
 
     return cur_span_index == best_span_index
 
-
 def _new_check_is_max_context(doc_spans, cur_span_index, position):
     """Check if this is the 'max context' doc span for the token."""
     # if len(doc_spans) == 1:
@@ -92,7 +91,7 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
     features = []
     new_features = []
     for (example_index, example) in enumerate(tqdm(examples)):
-        if is_training:
+        if is_training and not example.is_impossible:
             # Get start and end position
             answer_length = len(example.answer_text)
             start_position = example.start_position
@@ -105,6 +104,7 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
                 logger.warning("Could not find answer: '%s' vs. '%s'", actual_text, cleaned_answer_text)
                 continue
 
+
         tok_to_orig_index = []
         orig_to_tok_index = []
         all_doc_tokens = []
@@ -115,6 +115,18 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
                 tok_to_orig_index.append(i)
                 all_doc_tokens.append(sub_token)
 
+
+        if is_training and not example.is_impossible:
+            tok_start_position = orig_to_tok_index[example.start_position]
+            if example.end_position < len(example.doc_tokens) - 1:
+                tok_end_position = orig_to_tok_index[example.end_position + 1] - 1
+            else:
+                tok_end_position = len(all_doc_tokens) - 1
+
+            (tok_start_position, tok_end_position) = _improve_answer_span(
+                all_doc_tokens, tok_start_position, tok_end_position, tokenizer, example.answer_text
+            )
+
         spans = []
         
         truncated_query = tokenizer.encode(example.question_text, add_special_tokens=False, max_length=max_query_length)
@@ -187,6 +199,34 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
             # Set the CLS index to '0'
             p_mask[cls_index] = 0
 
+
+            span_is_impossible = example.is_impossible
+            start_position = 0
+            end_position = 0
+            if is_training and not span_is_impossible:
+                # For training, if our document chunk does not contain an annotation
+                # we throw it out, since there is nothing to predict.
+                doc_start = span["start"]
+                doc_end = span["start"] + span["length"] - 1
+                out_of_span = False
+
+                if not (tok_start_position >= doc_start and tok_end_position <= doc_end):
+                    out_of_span = True
+
+                if out_of_span:
+                    start_position = cls_index
+                    end_position = cls_index
+                    span_is_impossible = True
+                else:
+                    if sequence_a_is_doc:
+                        doc_offset = 0
+                    else:
+                        doc_offset = len(truncated_query) + sequence_added_tokens
+                        
+                    start_position = tok_start_position - doc_start + doc_offset
+                    end_position = tok_end_position - doc_start + doc_offset
+
+
             new_features.append(NewSquadFeatures(
                 span['input_ids'],
                 span['attention_mask'],
@@ -199,7 +239,10 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
                 paragraph_len=span['paragraph_len'],
                 token_is_max_context=span["token_is_max_context"],
                 tokens=span["tokens"],
-                token_to_orig_map=span["token_to_orig_map"]
+                token_to_orig_map=span["token_to_orig_map"],
+                
+                start_position=start_position,
+                end_position=end_position
             ))
 
             unique_id += 1
@@ -207,86 +250,10 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
     return new_features
 
 
-def read_squad_examples(input_file, is_training, version_2_with_negative):
-    """Read a SQuAD json file into a list of SquadExample."""
-    with open(input_file, "r", encoding='utf-8') as reader:
-        input_data = json.load(reader)["data"]
-
-    def is_whitespace(c):
-        if c == " " or c == "\t" or c == "\r" or c == "\n" or ord(c) == 0x202F:
-            return True
-        return False
-
-    examples = []
-    for entry in input_data:
-        for paragraph in entry["paragraphs"]:
-            paragraph_text = paragraph["context"]
-            doc_tokens = []
-            char_to_word_offset = []
-            prev_is_whitespace = True
-            for c in paragraph_text:
-                if is_whitespace(c):
-                    prev_is_whitespace = True
-                else:
-                    if prev_is_whitespace:
-                        doc_tokens.append(c)
-                    else:
-                        doc_tokens[-1] += c
-                    prev_is_whitespace = False
-                char_to_word_offset.append(len(doc_tokens) - 1)
-
-            for qa in paragraph["qas"]:
-                qas_id = qa["id"]
-                question_text = qa["question"]
-                start_position = None
-                end_position = None
-                orig_answer_text = None
-                is_impossible = False
-                if is_training:
-                    if version_2_with_negative:
-                        is_impossible = qa["is_impossible"]
-                    if (len(qa["answers"]) != 1) and (not is_impossible):
-                        raise ValueError(
-                            "For training, each question should have exactly 1 answer.")
-                    if not is_impossible:
-                        answer = qa["answers"][0]
-                        orig_answer_text = answer["text"]
-                        answer_offset = answer["answer_start"]
-                        answer_length = len(orig_answer_text)
-                        start_position = char_to_word_offset[answer_offset]
-                        end_position = char_to_word_offset[answer_offset + answer_length - 1]
-                        # Only add answers where the text can be exactly recovered from the
-                        # document. If this CAN'T happen it's likely due to weird Unicode
-                        # stuff so we will just skip the example.
-                        #
-                        # Note that this means for training mode, every example is NOT
-                        # guaranteed to be preserved.
-                        actual_text = " ".join(doc_tokens[start_position:(end_position + 1)])
-                        cleaned_answer_text = " ".join(
-                            whitespace_tokenize(orig_answer_text))
-                        if actual_text.find(cleaned_answer_text) == -1:
-                            logger.warning("Could not find answer: '%s' vs. '%s'",
-                                           actual_text, cleaned_answer_text)
-                            continue
-                    else:
-                        start_position = -1
-                        end_position = -1
-                        orig_answer_text = ""
-
-                example = SquadExample(
-                    qas_id=qas_id,
-                    question_text=question_text,
-                    doc_tokens=doc_tokens,
-                    orig_answer_text=orig_answer_text,
-                    start_position=start_position,
-                    end_position=end_position,
-                    is_impossible=is_impossible)
-                examples.append(example)
-    return examples
-
-
-class SquadV1Processor(DataProcessor):
+class SquadProcessor(DataProcessor):
     """Processor for the SQuAD data set."""
+    train_file = None
+    dev_file = None
 
     def get_example_from_tensor_dict(self, tensor_dict):
         """See base class."""
@@ -301,13 +268,19 @@ class SquadV1Processor(DataProcessor):
 
     def get_train_examples(self, data_dir, only_first=None):
         """See base class."""
-        with open(os.path.join(data_dir, "train-v1.1.json"), "r", encoding='utf-8') as reader:
+        if self.train_file is None:
+            raise ValueError("SquadProcessor should be instantiated via SquadV1Processor or SquadV2Processor")
+
+        with open(os.path.join(data_dir, self.train_file), "r", encoding='utf-8') as reader:
             input_data = json.load(reader)["data"]
         return self._create_examples(input_data, "train", only_first)
 
     def get_dev_examples(self, data_dir, only_first=None):
         """See base class."""
-        with open(os.path.join(data_dir, "dev-v1.1.json"), "r", encoding='utf-8') as reader:
+        if self.dev_file is None:
+            raise ValueError("SquadProcessor should be instantiated via SquadV1Processor or SquadV2Processor")
+        
+        with open(os.path.join(data_dir, self.dev_file), "r", encoding='utf-8') as reader:
             input_data = json.load(reader)["data"]
         return self._create_examples(input_data, "dev", only_first)
 
@@ -329,7 +302,13 @@ class SquadV1Processor(DataProcessor):
                     question_text = qa["question"]
                     start_position_character = None
                     answer_text = None
-                    if is_training:
+                    
+                    if "is_impossible" in qa:
+                        is_impossible = qa["is_impossible"]
+                    else:
+                        is_impossible = False
+
+                    if not is_impossible and is_training:
                         if (len(qa["answers"]) != 1):
                             raise ValueError(
                                 "For training, each question should have exactly 1 answer.")
@@ -343,15 +322,25 @@ class SquadV1Processor(DataProcessor):
                         context_text=context_text,
                         answer_text=answer_text,
                         start_position_character=start_position_character,
-                        title=title
+                        title=title,
+                        is_impossible=is_impossible
                     )
+
                     examples.append(example)
 
                     if only_first is not None and len(examples) > only_first:
                         return examples
         return examples
-        
 
+class SquadV1Processor(SquadProcessor):
+    train_file = "train-v1.1.json"
+    dev_file = "dev-v1.1.json"
+
+
+class SquadV2Processor(SquadProcessor):
+    train_file = "train-v2.0.json"
+    dev_file = "dev-v2.0.json"
+    
 
 class NewSquadExample(object):
     """
@@ -364,13 +353,16 @@ class NewSquadExample(object):
                  context_text,
                  answer_text,
                  start_position_character,
-                 title):
+                 title,
+                 is_impossible=False):
         self.qas_id = qas_id
         self.question_text = question_text
         self.context_text = context_text
         self.answer_text = answer_text
         self.title = title
-        self.is_impossible = False
+        self.is_impossible = is_impossible 
+
+        self.start_position, self.end_position = 0, 0
 
         doc_tokens = []
         char_to_word_offset = []
@@ -392,7 +384,7 @@ class NewSquadExample(object):
         self.char_to_word_offset = char_to_word_offset
 
         # Start end end positions only has a value during evaluation.
-        if start_position_character is not None:
+        if start_position_character is not None and not is_impossible:
             self.start_position = char_to_word_offset[start_position_character]
             self.end_position = char_to_word_offset[start_position_character + len(answer_text) - 1]
 
@@ -415,7 +407,10 @@ class NewSquadFeatures(object):
                  paragraph_len,
                  token_is_max_context,
                  tokens,
-                 token_to_orig_map
+                 token_to_orig_map,
+
+                 start_position,
+                 end_position
         ):
         self.input_ids = input_ids 
         self.attention_mask = attention_mask
@@ -430,6 +425,9 @@ class NewSquadFeatures(object):
         self.tokens = tokens
         self.token_to_orig_map = token_to_orig_map
 
+        self.start_position = start_position
+        self.end_position = end_position
+
 class SquadExample(object):
     """
     A single training/test example for the Squad dataset.

From 8e5d84fcc1a645d3c13b8a2f64fa995637440dad Mon Sep 17 00:00:00 2001
From: v_sboliu <v_sboliu@tencent.com>
Date: Tue, 26 Nov 2019 12:04:31 +0800
Subject: [PATCH 084/505] Fixed typo

---
 transformers/modeling_bert.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/modeling_bert.py b/transformers/modeling_bert.py
index 7c2c6f4602..81d92d8f1b 100644
--- a/transformers/modeling_bert.py
+++ b/transformers/modeling_bert.py
@@ -278,7 +278,7 @@ class BertAttention(nn.Module):
         if len(heads) == 0:
             return
         mask = torch.ones(self.self.num_attention_heads, self.self.attention_head_size)
-        heads = set(heads) - self.pruned_heads  # Convert to set and emove already pruned heads
+        heads = set(heads) - self.pruned_heads  # Convert to set and remove already pruned heads
         for head in heads:
             # Compute how many pruned heads are before the head and move the index accordingly
             head = head - sum(1 if h < head else 0 for h in self.pruned_heads)

From c0c2088333e2e8ce2b24d0c7f4bf071dcccbd7ea Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Tue, 29 Oct 2019 19:57:38 +0000
Subject: [PATCH 085/505] ALBERT model

---
 transformers/configuration_albert.py          |  72 ++++
 ...lbert_original_tf_checkpoint_to_pytorch.py |  44 +++
 transformers/modeling_albert.py               | 331 ++++++++++++++++++
 3 files changed, 447 insertions(+)
 create mode 100644 transformers/configuration_albert.py
 create mode 100644 transformers/convert_albert_original_tf_checkpoint_to_pytorch.py
 create mode 100644 transformers/modeling_albert.py

diff --git a/transformers/configuration_albert.py b/transformers/configuration_albert.py
new file mode 100644
index 0000000000..c86c9565cb
--- /dev/null
+++ b/transformers/configuration_albert.py
@@ -0,0 +1,72 @@
+from .configuration_utils import PretrainedConfig
+
+class AlbertConfig(PretrainedConfig):
+    """Configuration for `AlbertModel`.
+
+    The default settings match the configuration of model `albert_xxlarge`.
+    """
+
+    def __init__(self,
+                 vocab_size_or_config_json_file,
+                 embedding_size=128,
+                 hidden_size=4096,
+                 num_hidden_layers=12,
+                 num_hidden_groups=1,
+                 num_attention_heads=64,
+                 intermediate_size=16384,
+                 inner_group_num=1,
+                 down_scale_factor=1,
+                 hidden_act="gelu",
+                 hidden_dropout_prob=0,
+                 attention_probs_dropout_prob=0,
+                 max_position_embeddings=512,
+                 type_vocab_size=2,
+                 initializer_range=0.02,
+                 layer_norm_eps=1e-12, **kwargs):
+        """Constructs AlbertConfig.
+
+        Args:
+            vocab_size: Vocabulary size of `inputs_ids` in `AlbertModel`.
+            embedding_size: size of voc embeddings.
+            hidden_size: Size of the encoder layers and the pooler layer.
+            num_hidden_layers: Number of hidden layers in the Transformer encoder.
+            num_hidden_groups: Number of group for the hidden layers, parameters in
+                the same group are shared.
+            num_attention_heads: Number of attention heads for each attention layer in
+                the Transformer encoder.
+            intermediate_size: The size of the "intermediate" (i.e., feed-forward)
+                layer in the Transformer encoder.
+            inner_group_num: int, number of inner repetition of attention and ffn.
+            down_scale_factor: float, the scale to apply
+            hidden_act: The non-linear activation function (function or string) in the
+                encoder and pooler.
+            hidden_dropout_prob: The dropout probability for all fully connected
+                layers in the embeddings, encoder, and pooler.
+            attention_probs_dropout_prob: The dropout ratio for the attention
+                probabilities.
+            max_position_embeddings: The maximum sequence length that this model might
+                ever be used with. Typically set this to something large just in case
+                (e.g., 512 or 1024 or 2048).
+            type_vocab_size: The vocabulary size of the `token_type_ids` passed into
+                `AlbertModel`.
+            initializer_range: The stdev of the truncated_normal_initializer for
+                initializing all weight matrices.
+        """
+        super(AlbertConfig, self).__init__(**kwargs)
+
+        self.vocab_size = vocab_size_or_config_json_file
+        self.embedding_size = embedding_size
+        self.hidden_size = hidden_size
+        self.num_hidden_layers = num_hidden_layers
+        self.num_hidden_groups = num_hidden_groups
+        self.num_attention_heads = num_attention_heads
+        self.inner_group_num = inner_group_num
+        self.down_scale_factor = down_scale_factor
+        self.hidden_act = hidden_act
+        self.intermediate_size = intermediate_size
+        self.hidden_dropout_prob = hidden_dropout_prob
+        self.attention_probs_dropout_prob = attention_probs_dropout_prob
+        self.max_position_embeddings = max_position_embeddings
+        self.type_vocab_size = type_vocab_size
+        self.initializer_range = initializer_range
+        self.layer_norm_eps = layer_norm_eps
\ No newline at end of file
diff --git a/transformers/convert_albert_original_tf_checkpoint_to_pytorch.py b/transformers/convert_albert_original_tf_checkpoint_to_pytorch.py
new file mode 100644
index 0000000000..04877d41b9
--- /dev/null
+++ b/transformers/convert_albert_original_tf_checkpoint_to_pytorch.py
@@ -0,0 +1,44 @@
+
+
+
+from transformers import AlbertConfig, BertForPreTraining, load_tf_weights_in_bert
+
+
+
+def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, bert_config_file, pytorch_dump_path):
+    # Initialise PyTorch model
+    config = BertConfig.from_json_file(bert_config_file)
+    print("Building PyTorch model from configuration: {}".format(str(config)))
+    model = BertForPreTraining(config)
+
+    # Load weights from tf checkpoint
+    load_tf_weights_in_bert(model, config, tf_checkpoint_path)
+
+    # Save pytorch-model
+    print("Save PyTorch model to {}".format(pytorch_dump_path))
+    torch.save(model.state_dict(), pytorch_dump_path)
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    ## Required parameters
+    parser.add_argument("--tf_checkpoint_path",
+                        default = None,
+                        type = str,
+                        required = True,
+                        help = "Path to the TensorFlow checkpoint path.")
+    parser.add_argument("--albert_config_file",
+                        default = None,
+                        type = str,
+                        required = True,
+                        help = "The config json file corresponding to the pre-trained BERT model. \n"
+                            "This specifies the model architecture.")
+    parser.add_argument("--pytorch_dump_path",
+                        default = None,
+                        type = str,
+                        required = True,
+                        help = "Path to the output PyTorch model.")
+    args = parser.parse_args()
+    convert_tf_checkpoint_to_pytorch(args.tf_checkpoint_path,
+                                     args.bert_config_file,
+                                     args.pytorch_dump_path)
diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
new file mode 100644
index 0000000000..b006cbe8fd
--- /dev/null
+++ b/transformers/modeling_albert.py
@@ -0,0 +1,331 @@
+
+import os
+import math
+import logging
+import torch
+import torch.nn as nn
+from transformers.configuration_albert import AlbertConfig
+logger = logging.getLogger(__name__)
+
+def load_tf_weights_in_albert(model, config, tf_checkpoint_path):
+    """ Load tf checkpoints in a pytorch model."""
+    try:
+        import re
+        import numpy as np
+        import tensorflow as tf
+    except ImportError:
+        logger.error("Loading a TensorFlow model in PyTorch, requires TensorFlow to be installed. Please see "
+            "https://www.tensorflow.org/install/ for installation instructions.")
+        raise
+    tf_path = os.path.abspath(tf_checkpoint_path)
+    logger.info("Converting TensorFlow checkpoint from {}".format(tf_path))
+    # Load weights from TF model
+    init_vars = tf.train.list_variables(tf_path)
+    names = []
+    arrays = []
+    for name, shape in init_vars:
+        logger.info("Loading TF weight {} with shape {}".format(name, shape))
+        array = tf.train.load_variable(tf_path, name)
+        names.append(name)
+        arrays.append(array)
+
+    print(model)
+    
+    for name, array in zip(names, arrays):
+        og = name
+        name = name.replace("transformer/group_0/inner_group_0", "transformer")
+        name = name.replace("LayerNorm", "layer_norm")
+        name = name.replace("ffn_1", "ffn")
+        name = name.replace("ffn/intermediate/output", "ffn_output")
+        name = name.replace("attention_1", "attention")   
+        name = name.replace("cls/predictions/transform", "predictions")
+        name = name.replace("transformer/layer_norm_1", "transformer/attention/output/LayerNorm")        
+        name = name.split('/')
+
+        print(name)
+        pointer = model
+        for m_name in name:
+            if re.fullmatch(r'[A-Za-z]+_\d+', m_name):
+                l = re.split(r'_(\d+)', m_name)
+            else:
+                l = [m_name]
+
+            if l[0] == 'kernel' or l[0] == 'gamma':
+                pointer = getattr(pointer, 'weight')
+            elif l[0] == 'output_bias' or l[0] == 'beta':
+                pointer = getattr(pointer, 'bias')
+            elif l[0] == 'output_weights':
+                pointer = getattr(pointer, 'weight')
+            elif l[0] == 'squad':
+                pointer = getattr(pointer, 'classifier')
+            else:
+                try:
+                    pointer = getattr(pointer, l[0])
+                except AttributeError:
+                    logger.info("Skipping {}".format("/".join(name)))
+                    continue
+            if len(l) >= 2:
+                num = int(l[1])
+                pointer = pointer[num]
+
+        if m_name[-11:] == '_embeddings':
+            pointer = getattr(pointer, 'weight')
+        elif m_name == 'kernel':
+            array = np.transpose(array)
+            print("transposed")
+        try:
+            assert pointer.shape == array.shape
+        except AssertionError as e:
+            e.args += (pointer.shape, array.shape)
+            raise
+        print("Initialize PyTorch weight {} from {}".format(name, og))
+        pointer.data = torch.from_numpy(array)
+
+    return model
+
+
+class AlbertEmbeddings(nn.Module):
+    """
+    Construct the embeddings from word, position and token_type embeddings.
+    """
+    def __init__(self, config):
+        super(AlbertEmbeddings, self).__init__()
+
+        self.word_embeddings = nn.Embedding(config.vocab_size, config.embedding_size, padding_idx=0)
+        self.position_embeddings = nn.Embedding(config.max_position_embeddings, config.embedding_size)
+        self.token_type_embeddings = nn.Embedding(config.type_vocab_size, config.embedding_size)
+        self.layer_norm = torch.nn.LayerNorm(config.embedding_size, eps=config.layer_norm_eps)
+        self.dropout = nn.Dropout(config.hidden_dropout_prob)
+
+    def forward(self, input_ids, token_type_ids=None, position_ids=None):
+        seq_length = input_ids.size(1)
+        if position_ids is None:
+            position_ids = torch.arange(seq_length, dtype=torch.long, device=input_ids.device)
+            position_ids = position_ids.unsqueeze(0).expand_as(input_ids)
+        if token_type_ids is None:
+            token_type_ids = torch.zeros_like(input_ids)
+
+        word_embeddings = self.word_embeddings(input_ids)
+        position_embeddings = self.position_embeddings(position_ids)
+        token_type_embeddings = self.token_type_embeddings(token_type_ids)
+
+        embeddings = word_embeddings + position_embeddings + token_type_embeddings
+        embeddings = self.layer_norm(embeddings)
+        embeddings = self.dropout(embeddings)
+        return embeddings
+
+
+    def get_word_embeddings_table(self):
+        return self.word_embeddings
+
+
+class AlbertModel(nn.Module):
+    def __init__(self, config):
+        super(AlbertModel, self).__init__()
+
+        self.config = config
+        self.embeddings = AlbertEmbeddings(config)
+        self.encoder = AlbertEncoder(config)
+        self.pooler = nn.Linear(config.hidden_size, config.hidden_size)
+        self.pooler_activation = nn.Tanh()
+
+    def forward(self, input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None):
+        if attention_mask is None:
+            attention_mask = torch.ones_like(input_ids)
+        if token_type_ids is None:
+            token_type_ids = torch.zeros_like(input_ids)
+
+        extended_attention_mask = attention_mask.unsqueeze(1).unsqueeze(2)
+        extended_attention_mask = extended_attention_mask.to(dtype=next(self.parameters()).dtype) # fp16 compatibility
+        extended_attention_mask = (1.0 - extended_attention_mask) * -10000.0
+        if head_mask is not None:
+            if head_mask.dim() == 1:
+                head_mask = head_mask.unsqueeze(0).unsqueeze(0).unsqueeze(-1).unsqueeze(-1)
+                head_mask = head_mask.expand(self.config.num_hidden_layers, -1, -1, -1, -1)
+            elif head_mask.dim() == 2:
+                head_mask = head_mask.unsqueeze(1).unsqueeze(-1).unsqueeze(-1)  # We can specify head_mask for each layer
+            head_mask = head_mask.to(dtype=next(self.parameters()).dtype) # switch to fload if need + fp16 compatibility
+        else:
+            head_mask = [None] * self.config.num_hidden_layers
+
+        embedding_output = self.embeddings(input_ids, position_ids=position_ids, token_type_ids=token_type_ids)
+        encoder_outputs = self.encoder(embedding_output,
+                                       extended_attention_mask,
+                                       head_mask=head_mask)
+        sequence_output = encoder_outputs[0]
+        print(sequence_output.shape, sequence_output[:, 0].shape, self.pooler(sequence_output[:, 0]).shape)
+        pooled_output = self.pooler_activation(self.pooler(sequence_output[:, 0]))
+
+        outputs = (sequence_output, pooled_output,) + encoder_outputs[1:]  # add hidden_states and attentions if they are here
+        return outputs
+
+
+class AlbertForMaskedLM(nn.Module):
+    def __init__(self, config):
+        super(AlbertForMaskedLM, self).__init__()
+
+        self.config = config
+        self.bert = AlbertModel(config)
+        self.layer_norm = nn.LayerNorm(config.embedding_size)
+        self.bias = nn.Parameter(torch.zeros(config.vocab_size))
+        self.dense = nn.Linear(config.hidden_size, config.embedding_size)
+        self.word_embeddings = nn.Linear(config.embedding_size, config.vocab_size)
+
+    def tie_weights(self):
+        """ Make sure we are sharing the input and output embeddings.
+            Export to TorchScript can't handle parameter sharing so we are cloning them instead.
+        """
+        self._tie_or_clone_weights(self.classifier.word_embeddings,
+                                   self.transformer.embeddings.word_embeddings)
+
+    def forward(self, input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None):
+        hidden_states = self.bert(input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None)[0]
+        hidden_states = self.dense(hidden_states)
+        hidden_states = gelu_new(hidden_states)
+        hidden_states = self.layer_norm(hidden_states)
+
+        logits = self.word_embeddings(hidden_states)
+
+        return logits
+
+
+class AlbertAttention(nn.Module):
+    def __init__(self, config):
+        super(AlbertAttention, self).__init__()
+
+        if config.hidden_size % config.num_attention_heads != 0:
+            raise ValueError(
+                "The hidden size (%d) is not a multiple of the number of attention "
+                "heads (%d)" % (config.hidden_size, config.num_attention_heads))
+        self.output_attentions = config.output_attentions
+
+        self.num_attention_heads = config.num_attention_heads
+        self.attention_head_size = int(config.hidden_size / config.num_attention_heads)
+        self.all_head_size = self.num_attention_heads * self.attention_head_size
+
+        self.query = nn.Linear(config.hidden_size, self.all_head_size)
+        self.key = nn.Linear(config.hidden_size, self.all_head_size)
+        self.value = nn.Linear(config.hidden_size, self.all_head_size)
+
+        self.dropout = nn.Dropout(config.attention_probs_dropout_prob)
+        self.dense = nn.Linear(config.hidden_size, config.hidden_size)
+        self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+
+    def transpose_for_scores(self, x):
+        new_x_shape = x.size()[:-1] + (self.num_attention_heads, self.attention_head_size)
+        x = x.view(*new_x_shape)
+        return x.permute(0, 2, 1, 3)
+
+    def forward(self, input_ids, attention_mask=None, head_mask=None):
+        mixed_query_layer = self.query(input_ids)
+        mixed_key_layer = self.key(input_ids)
+        mixed_value_layer = self.value(input_ids)
+
+        query_layer = self.transpose_for_scores(mixed_query_layer)
+        key_layer = self.transpose_for_scores(mixed_key_layer)
+        value_layer = self.transpose_for_scores(mixed_value_layer)
+
+        # Take the dot product between "query" and "key" to get the raw attention scores.
+        attention_scores = torch.matmul(query_layer, key_layer.transpose(-1, -2))
+        attention_scores = attention_scores / math.sqrt(self.attention_head_size)
+        if attention_mask is not None:
+            # Apply the attention mask is (precomputed for all layers in BertModel forward() function)
+            attention_scores = attention_scores + attention_mask
+
+        # Normalize the attention scores to probabilities.
+        attention_probs = nn.Softmax(dim=-1)(attention_scores)
+
+        # This is actually dropping out entire tokens to attend to, which might
+        # seem a bit unusual, but is taken from the original Transformer paper.
+        attention_probs = self.dropout(attention_probs)
+
+        # Mask heads if we want to
+        if head_mask is not None:
+            attention_probs = attention_probs * head_mask
+
+        context_layer = torch.matmul(attention_probs, value_layer)
+
+        context_layer = context_layer.permute(0, 2, 1, 3).contiguous()
+        new_context_layer_shape = context_layer.size()[:-2] + (self.all_head_size,)
+        reshaped_context_layer = context_layer.view(*new_context_layer_shape)
+        w = self.dense.weight.T.view(16, 64, 1024)
+        b = self.dense.bias
+
+        projected_context_layer = torch.einsum("bfnd,ndh->bfh", context_layer, w) + b
+        projected_context_layer = self.dropout(projected_context_layer)
+        layernormed_context_layer = self.LayerNorm(input_ids + projected_context_layer)
+        return layernormed_context_layer, projected_context_layer, reshaped_context_layer, context_layer, attention_scores, attention_probs, attention_mask
+
+
+class AlbertTransformer(nn.Module):
+    def __init__(self, config):
+        super(AlbertTransformer, self).__init__()
+        
+        self.config =config
+        self.layer_norm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+        self.attention = AlbertAttention(config)
+        self.ffn = nn.Linear(config.hidden_size, config.intermediate_size) 
+        self.ffn_output = nn.Linear(config.intermediate_size, config.hidden_size)
+
+    def forward(self, hidden_states, attention_mask=None, head_mask=None):
+        for i in range(self.config.num_hidden_layers):
+            attention_output = self.attention(hidden_states, attention_mask)[0]
+            ffn_output = self.ffn(attention_output)
+            ffn_output = gelu_new(ffn_output)
+            ffn_output = self.ffn_output(ffn_output)
+            hidden_states = self.layer_norm(ffn_output + attention_output)
+
+        return hidden_states
+
+
+def gelu_new(x):
+    """ Implementation of the gelu activation function currently in Google Bert repo (identical to OpenAI GPT).
+        Also see https://arxiv.org/abs/1606.08415
+    """
+    return 0.5 * x * (1 + torch.tanh(math.sqrt(2 / math.pi) * (x + 0.044715 * torch.pow(x, 3))))
+
+
+class AlbertEncoder(nn.Module):
+    def __init__(self, config):
+        super(AlbertEncoder, self).__init__()
+
+        self.output_attentions = config.output_attentions
+        self.output_hidden_states = config.output_hidden_states
+        self.embedding_hidden_mapping_in = nn.Linear(config.embedding_size, config.hidden_size)
+        self.transformer = AlbertTransformer(config)
+
+    def forward(self, hidden_states, attention_mask=None, head_mask=None):
+        hidden_states = self.embedding_hidden_mapping_in(hidden_states)
+        hidden_states = self.transformer(hidden_states, attention_mask, head_mask)
+
+        outputs = (hidden_states,)
+        if self.output_hidden_states:
+            outputs = outputs + (all_hidden_states,)
+        if self.output_attentions:
+            outputs = outputs + (all_attentions,)
+        return outputs  # last-layer hidden state, (all hidden states), (all attentions)
+
+# config = AlbertConfig.from_json_file("config.json")
+# # model = AlbertForMaskedLM(config)
+# model = AlbertModel(config)
+
+# model = load_tf_weights_in_albert(model, config, "albert/albert")
+
+# print(model)
+
+# input_ids = torch.tensor([[31, 51, 99], [15, 5, 0]])
+# input_mask = torch.tensor([[1, 1, 1], [1, 1, 0]])
+# segment_ids = torch.tensor([[0, 0, 1], [0, 0, 0]])
+
+# # sequence_output, pooled_outputs = model()
+
+# logits = model(input_ids, attention_mask=input_mask, token_type_ids=segment_ids)[1]
+
+
+# embeddings_output = 
+# print("pooled output", logits)
+# # print("Pooled output", pooled_outputs)
+
+config = AlbertConfig.from_json_file("/home/hf/google-research/albert/config.json")
+model = AlbertModel(config)
+model = load_tf_weights_in_albert(model, config, "/home/hf/transformers/albert/albert")
\ No newline at end of file

From 91ccbae788de11f0f5ffb862421e345b27a20a76 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Tue, 29 Oct 2019 21:21:57 +0000
Subject: [PATCH 086/505] Accepts multiple sizes

---
 transformers/modeling_albert.py | 143 ++++++++++++++------------------
 1 file changed, 60 insertions(+), 83 deletions(-)

diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index b006cbe8fd..f3cebdc3d9 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -5,6 +5,7 @@ import logging
 import torch
 import torch.nn as nn
 from transformers.configuration_albert import AlbertConfig
+from transformers.modeling_bert import BertEmbeddings, BertModel, BertSelfAttention, prune_linear_layer, gelu_new
 logger = logging.getLogger(__name__)
 
 def load_tf_weights_in_albert(model, config, tf_checkpoint_path):
@@ -32,14 +33,14 @@ def load_tf_weights_in_albert(model, config, tf_checkpoint_path):
     print(model)
     
     for name, array in zip(names, arrays):
+        print(name)
         og = name
         name = name.replace("transformer/group_0/inner_group_0", "transformer")
-        name = name.replace("LayerNorm", "layer_norm")
         name = name.replace("ffn_1", "ffn")
         name = name.replace("ffn/intermediate/output", "ffn_output")
         name = name.replace("attention_1", "attention")   
         name = name.replace("cls/predictions/transform", "predictions")
-        name = name.replace("transformer/layer_norm_1", "transformer/attention/output/LayerNorm")        
+        name = name.replace("transformer/LayerNorm_1", "transformer/attention/LayerNorm")        
         name = name.split('/')
 
         print(name)
@@ -84,44 +85,22 @@ def load_tf_weights_in_albert(model, config, tf_checkpoint_path):
     return model
 
 
-class AlbertEmbeddings(nn.Module):
+class AlbertEmbeddings(BertEmbeddings):
     """
     Construct the embeddings from word, position and token_type embeddings.
     """
     def __init__(self, config):
-        super(AlbertEmbeddings, self).__init__()
+        super(AlbertEmbeddings, self).__init__(config)
 
         self.word_embeddings = nn.Embedding(config.vocab_size, config.embedding_size, padding_idx=0)
         self.position_embeddings = nn.Embedding(config.max_position_embeddings, config.embedding_size)
         self.token_type_embeddings = nn.Embedding(config.type_vocab_size, config.embedding_size)
-        self.layer_norm = torch.nn.LayerNorm(config.embedding_size, eps=config.layer_norm_eps)
-        self.dropout = nn.Dropout(config.hidden_dropout_prob)
-
-    def forward(self, input_ids, token_type_ids=None, position_ids=None):
-        seq_length = input_ids.size(1)
-        if position_ids is None:
-            position_ids = torch.arange(seq_length, dtype=torch.long, device=input_ids.device)
-            position_ids = position_ids.unsqueeze(0).expand_as(input_ids)
-        if token_type_ids is None:
-            token_type_ids = torch.zeros_like(input_ids)
-
-        word_embeddings = self.word_embeddings(input_ids)
-        position_embeddings = self.position_embeddings(position_ids)
-        token_type_embeddings = self.token_type_embeddings(token_type_ids)
-
-        embeddings = word_embeddings + position_embeddings + token_type_embeddings
-        embeddings = self.layer_norm(embeddings)
-        embeddings = self.dropout(embeddings)
-        return embeddings
+        self.LayerNorm = torch.nn.LayerNorm(config.embedding_size, eps=config.layer_norm_eps)
 
 
-    def get_word_embeddings_table(self):
-        return self.word_embeddings
-
-
-class AlbertModel(nn.Module):
+class AlbertModel(BertModel):
     def __init__(self, config):
-        super(AlbertModel, self).__init__()
+        super(AlbertModel, self).__init__(config)
 
         self.config = config
         self.embeddings = AlbertEmbeddings(config)
@@ -129,6 +108,7 @@ class AlbertModel(nn.Module):
         self.pooler = nn.Linear(config.hidden_size, config.hidden_size)
         self.pooler_activation = nn.Tanh()
 
+
     def forward(self, input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None):
         if attention_mask is None:
             attention_mask = torch.ones_like(input_ids)
@@ -166,7 +146,7 @@ class AlbertForMaskedLM(nn.Module):
 
         self.config = config
         self.bert = AlbertModel(config)
-        self.layer_norm = nn.LayerNorm(config.embedding_size)
+        self.LayerNorm = nn.LayerNorm(config.embedding_size)
         self.bias = nn.Parameter(torch.zeros(config.vocab_size))
         self.dense = nn.Linear(config.hidden_size, config.embedding_size)
         self.word_embeddings = nn.Linear(config.embedding_size, config.vocab_size)
@@ -182,39 +162,47 @@ class AlbertForMaskedLM(nn.Module):
         hidden_states = self.bert(input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None)[0]
         hidden_states = self.dense(hidden_states)
         hidden_states = gelu_new(hidden_states)
-        hidden_states = self.layer_norm(hidden_states)
+        hidden_states = self.LayerNorm(hidden_states)
 
         logits = self.word_embeddings(hidden_states)
 
         return logits
 
 
-class AlbertAttention(nn.Module):
+class AlbertAttention(BertSelfAttention):
     def __init__(self, config):
-        super(AlbertAttention, self).__init__()
-
-        if config.hidden_size % config.num_attention_heads != 0:
-            raise ValueError(
-                "The hidden size (%d) is not a multiple of the number of attention "
-                "heads (%d)" % (config.hidden_size, config.num_attention_heads))
-        self.output_attentions = config.output_attentions
+        super(AlbertAttention, self).__init__(config)
 
         self.num_attention_heads = config.num_attention_heads
-        self.attention_head_size = int(config.hidden_size / config.num_attention_heads)
-        self.all_head_size = self.num_attention_heads * self.attention_head_size
-
-        self.query = nn.Linear(config.hidden_size, self.all_head_size)
-        self.key = nn.Linear(config.hidden_size, self.all_head_size)
-        self.value = nn.Linear(config.hidden_size, self.all_head_size)
-
+        self.hidden_size = config.hidden_size 
+        self.attention_head_size = config.hidden_size // config.num_attention_heads
         self.dropout = nn.Dropout(config.attention_probs_dropout_prob)
         self.dense = nn.Linear(config.hidden_size, config.hidden_size)
         self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+        self.pruned_heads = set()
 
-    def transpose_for_scores(self, x):
-        new_x_shape = x.size()[:-1] + (self.num_attention_heads, self.attention_head_size)
-        x = x.view(*new_x_shape)
-        return x.permute(0, 2, 1, 3)
+    def prune_heads(self, heads):
+        if len(heads) == 0:
+            return
+        mask = torch.ones(self.num_attention_heads, self.attention_head_size)
+        heads = set(heads) - self.pruned_heads  # Convert to set and emove already pruned heads
+        for head in heads:
+            # Compute how many pruned heads are before the head and move the index accordingly
+            head = head - sum(1 if h < head else 0 for h in self.pruned_heads)
+            mask[head] = 0
+        mask = mask.view(-1).contiguous().eq(1)
+        index = torch.arange(len(mask))[mask].long()
+
+        # Prune linear layers
+        self.query = prune_linear_layer(self.query, index)
+        self.key = prune_linear_layer(self.key, index)
+        self.value = prune_linear_layer(self.value, index)
+        self.output.dense = prune_linear_layer(self.output.dense, index, dim=1)
+
+        # Update hyper params and store pruned heads
+        self.num_attention_heads = self.num_attention_heads - len(heads)
+        self.all_head_size = self.attention_head_size * self.num_attention_heads
+        self.pruned_heads = self.pruned_heads.union(heads)
 
     def forward(self, input_ids, attention_mask=None, head_mask=None):
         mixed_query_layer = self.query(input_ids)
@@ -248,7 +236,8 @@ class AlbertAttention(nn.Module):
         context_layer = context_layer.permute(0, 2, 1, 3).contiguous()
         new_context_layer_shape = context_layer.size()[:-2] + (self.all_head_size,)
         reshaped_context_layer = context_layer.view(*new_context_layer_shape)
-        w = self.dense.weight.T.view(16, 64, 1024)
+        print(self.dense.weight.T.shape)
+        w = self.dense.weight.T.view(self.num_attention_heads, self.attention_head_size, self.hidden_size)
         b = self.dense.bias
 
         projected_context_layer = torch.einsum("bfnd,ndh->bfh", context_layer, w) + b
@@ -262,7 +251,7 @@ class AlbertTransformer(nn.Module):
         super(AlbertTransformer, self).__init__()
         
         self.config =config
-        self.layer_norm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+        self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
         self.attention = AlbertAttention(config)
         self.ffn = nn.Linear(config.hidden_size, config.intermediate_size) 
         self.ffn_output = nn.Linear(config.intermediate_size, config.hidden_size)
@@ -273,18 +262,11 @@ class AlbertTransformer(nn.Module):
             ffn_output = self.ffn(attention_output)
             ffn_output = gelu_new(ffn_output)
             ffn_output = self.ffn_output(ffn_output)
-            hidden_states = self.layer_norm(ffn_output + attention_output)
+            hidden_states = self.LayerNorm(ffn_output + attention_output)
 
         return hidden_states
 
 
-def gelu_new(x):
-    """ Implementation of the gelu activation function currently in Google Bert repo (identical to OpenAI GPT).
-        Also see https://arxiv.org/abs/1606.08415
-    """
-    return 0.5 * x * (1 + torch.tanh(math.sqrt(2 / math.pi) * (x + 0.044715 * torch.pow(x, 3))))
-
-
 class AlbertEncoder(nn.Module):
     def __init__(self, config):
         super(AlbertEncoder, self).__init__()
@@ -305,27 +287,22 @@ class AlbertEncoder(nn.Module):
             outputs = outputs + (all_attentions,)
         return outputs  # last-layer hidden state, (all hidden states), (all attentions)
 
-# config = AlbertConfig.from_json_file("config.json")
-# # model = AlbertForMaskedLM(config)
-# model = AlbertModel(config)
-
-# model = load_tf_weights_in_albert(model, config, "albert/albert")
-
-# print(model)
-
-# input_ids = torch.tensor([[31, 51, 99], [15, 5, 0]])
-# input_mask = torch.tensor([[1, 1, 1], [1, 1, 0]])
-# segment_ids = torch.tensor([[0, 0, 1], [0, 0, 0]])
-
-# # sequence_output, pooled_outputs = model()
-
-# logits = model(input_ids, attention_mask=input_mask, token_type_ids=segment_ids)[1]
-
-
-# embeddings_output = 
-# print("pooled output", logits)
-# # print("Pooled output", pooled_outputs)
-
-config = AlbertConfig.from_json_file("/home/hf/google-research/albert/config.json")
+model_size = "base"
+config = AlbertConfig.from_json_file("/home/hf/google-research/albert/config_{}.json".format(model_size))
 model = AlbertModel(config)
-model = load_tf_weights_in_albert(model, config, "/home/hf/transformers/albert/albert")
\ No newline at end of file
+model = load_tf_weights_in_albert(model, config, "/home/hf/transformers/albert-{}/albert-{}".format(model_size, model_size))
+model.eval()
+print(sum(p.numel() for p in model.parameters() if p.requires_grad))
+
+
+input_ids = [[31, 51, 99, 88, 54, 34, 23, 23, 12], [15, 5, 0, 88, 54, 34, 23, 23, 12]]
+input_mask = [[1, 1, 1, 1, 1, 1, 1, 1, 0], [1, 1, 1, 1, 1, 1, 0, 0, 0]]
+segment_ids = [[0, 0, 1, 0, 0, 1, 0, 0, 1], [0, 0, 0, 0, 0, 0, 0, 0, 0]]
+
+pt_input_ids = torch.tensor(input_ids)
+pt_input_mask = torch.tensor(input_mask)
+pt_segment_ids = torch.tensor(segment_ids)
+
+pt_dict = {"input_ids": pt_input_ids, "attention_mask": pt_input_mask, "token_type_ids": pt_segment_ids}
+pt_output = model(**pt_dict)
+print(pt_output)
\ No newline at end of file

From 139affaa8de89c6ef16a4712d24a22a8fea91eda Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Tue, 29 Oct 2019 22:45:24 +0000
Subject: [PATCH 087/505] Albert layer/layer groups

---
 transformers/modeling_albert.py | 80 ++++++++++++++++++++-------------
 1 file changed, 50 insertions(+), 30 deletions(-)

diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index f3cebdc3d9..440ccf2bce 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -30,17 +30,19 @@ def load_tf_weights_in_albert(model, config, tf_checkpoint_path):
         names.append(name)
         arrays.append(array)
 
-    print(model)
+    for name, array in zip(names, arrays):
+        print(name)
     
     for name, array in zip(names, arrays):
         print(name)
         og = name
-        name = name.replace("transformer/group_0/inner_group_0", "transformer")
         name = name.replace("ffn_1", "ffn")
         name = name.replace("ffn/intermediate/output", "ffn_output")
         name = name.replace("attention_1", "attention")   
         name = name.replace("cls/predictions/transform", "predictions")
-        name = name.replace("transformer/LayerNorm_1", "transformer/attention/LayerNorm")        
+        name = name.replace("LayerNorm_1", "attention/LayerNorm")    
+        name = name.replace("inner_group_", "albert_layers/") 
+        name = name.replace("group_", "albert_layer_groups/")    
         name = name.split('/')
 
         print(name)
@@ -104,7 +106,7 @@ class AlbertModel(BertModel):
 
         self.config = config
         self.embeddings = AlbertEmbeddings(config)
-        self.encoder = AlbertEncoder(config)
+        self.encoder = AlbertTransformer(config)
         self.pooler = nn.Linear(config.hidden_size, config.hidden_size)
         self.pooler_activation = nn.Tanh()
 
@@ -133,6 +135,7 @@ class AlbertModel(BertModel):
                                        extended_attention_mask,
                                        head_mask=head_mask)
         sequence_output = encoder_outputs[0]
+    
         print(sequence_output.shape, sequence_output[:, 0].shape, self.pooler(sequence_output[:, 0]).shape)
         pooled_output = self.pooler_activation(self.pooler(sequence_output[:, 0]))
 
@@ -246,18 +249,18 @@ class AlbertAttention(BertSelfAttention):
         return layernormed_context_layer, projected_context_layer, reshaped_context_layer, context_layer, attention_scores, attention_probs, attention_mask
 
 
-class AlbertTransformer(nn.Module):
+class AlbertLayer(nn.Module):
     def __init__(self, config):
-        super(AlbertTransformer, self).__init__()
+        super(AlbertLayer, self).__init__()
         
-        self.config =config
+        self.config = config
         self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
         self.attention = AlbertAttention(config)
         self.ffn = nn.Linear(config.hidden_size, config.intermediate_size) 
         self.ffn_output = nn.Linear(config.intermediate_size, config.hidden_size)
 
     def forward(self, hidden_states, attention_mask=None, head_mask=None):
-        for i in range(self.config.num_hidden_layers):
+        for _ in range(self.config.inner_group_num):
             attention_output = self.attention(hidden_states, attention_mask)[0]
             ffn_output = self.ffn(attention_output)
             ffn_output = gelu_new(ffn_output)
@@ -267,42 +270,59 @@ class AlbertTransformer(nn.Module):
         return hidden_states
 
 
-class AlbertEncoder(nn.Module):
+class AlbertLayerGroup(nn.Module):
     def __init__(self, config):
-        super(AlbertEncoder, self).__init__()
+        super(AlbertLayerGroup, self).__init__()
+        
+        self.albert_layers = nn.ModuleList([AlbertLayer(config) for _ in range(config.inner_group_num)])
 
+    def forward(self, hidden_states, attention_mask=None, head_mask=None):
+        for albert_layer in self.albert_layers:
+            hidden_states = albert_layer(hidden_states, attention_mask, head_mask)
+
+        return hidden_states
+
+
+class AlbertTransformer(nn.Module):
+    def __init__(self, config):
+        super(AlbertTransformer, self).__init__()
+        
+        self.config = config
         self.output_attentions = config.output_attentions
         self.output_hidden_states = config.output_hidden_states
         self.embedding_hidden_mapping_in = nn.Linear(config.embedding_size, config.hidden_size)
-        self.transformer = AlbertTransformer(config)
+        self.albert_layer_groups = nn.ModuleList([AlbertLayerGroup(config) for _ in range(config.num_hidden_groups)])
 
     def forward(self, hidden_states, attention_mask=None, head_mask=None):
         hidden_states = self.embedding_hidden_mapping_in(hidden_states)
-        hidden_states = self.transformer(hidden_states, attention_mask, head_mask)
 
-        outputs = (hidden_states,)
-        if self.output_hidden_states:
-            outputs = outputs + (all_hidden_states,)
-        if self.output_attentions:
-            outputs = outputs + (all_attentions,)
-        return outputs  # last-layer hidden state, (all hidden states), (all attentions)
+        for layer_idx in range(self.config.num_hidden_layers):
+            group_idx = int(layer_idx / self.config.num_hidden_layers * self.config.num_hidden_groups)
+            hidden_states = self.albert_layer_groups[group_idx](hidden_states, attention_mask, head_mask)  
+        
+        return (hidden_states,)
 
-model_size = "base"
-config = AlbertConfig.from_json_file("/home/hf/google-research/albert/config_{}.json".format(model_size))
+
+model_size = 'base'
+hidden_groups = 1
+inner_groups = 1
+config = AlbertConfig.from_json_file("/home/hf/google-research/albert/config_{}-{}-hg-{}-ig.json".format(model_size, hidden_groups, inner_groups))
 model = AlbertModel(config)
-model = load_tf_weights_in_albert(model, config, "/home/hf/transformers/albert-{}/albert-{}".format(model_size, model_size))
+
+print(model)
+model = load_tf_weights_in_albert(model, config, "/home/hf/transformers/albert-{}-{}-hg-{}-ig/albert-{}-{}-hg-{}-ig".format(model_size, hidden_groups, inner_groups, model_size, hidden_groups, inner_groups))
 model.eval()
 print(sum(p.numel() for p in model.parameters() if p.requires_grad))
 
 
-input_ids = [[31, 51, 99, 88, 54, 34, 23, 23, 12], [15, 5, 0, 88, 54, 34, 23, 23, 12]]
-input_mask = [[1, 1, 1, 1, 1, 1, 1, 1, 0], [1, 1, 1, 1, 1, 1, 0, 0, 0]]
-segment_ids = [[0, 0, 1, 0, 0, 1, 0, 0, 1], [0, 0, 0, 0, 0, 0, 0, 0, 0]]
+# input_ids = [[31, 51, 99, 88, 54, 34, 23, 23, 12], [15, 5, 0, 88, 54, 34, 23, 23, 12]]
+# input_mask = [[1, 1, 1, 1, 1, 1, 1, 1, 0], [1, 1, 1, 1, 1, 1, 0, 0, 0]]
+# segment_ids = [[0, 0, 1, 0, 0, 1, 0, 0, 1], [0, 0, 0, 0, 0, 0, 0, 0, 0]]
 
-pt_input_ids = torch.tensor(input_ids)
-pt_input_mask = torch.tensor(input_mask)
-pt_segment_ids = torch.tensor(segment_ids)
+# pt_input_ids = torch.tensor(input_ids)
+# pt_input_mask = torch.tensor(input_mask)
+# pt_segment_ids = torch.tensor(segment_ids)
 
-pt_dict = {"input_ids": pt_input_ids, "attention_mask": pt_input_mask, "token_type_ids": pt_segment_ids}
-pt_output = model(**pt_dict)
-print(pt_output)
\ No newline at end of file
+# pt_dict = {"input_ids": pt_input_ids, "attention_mask": pt_input_mask, "token_type_ids": pt_segment_ids}
+# pt_output = model(**pt_dict)
+# print(pt_output)
\ No newline at end of file

From 12290c0d5ce8475e884190b6ba480a8b3e671b3e Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Tue, 29 Oct 2019 23:19:02 +0000
Subject: [PATCH 088/505] Handles multi layer and multi groups

---
 transformers/modeling_albert.py | 30 ++++++++++++++----------------
 1 file changed, 14 insertions(+), 16 deletions(-)

diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index 440ccf2bce..90e9b162e6 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -136,7 +136,6 @@ class AlbertModel(BertModel):
                                        head_mask=head_mask)
         sequence_output = encoder_outputs[0]
     
-        print(sequence_output.shape, sequence_output[:, 0].shape, self.pooler(sequence_output[:, 0]).shape)
         pooled_output = self.pooler_activation(self.pooler(sequence_output[:, 0]))
 
         outputs = (sequence_output, pooled_output,) + encoder_outputs[1:]  # add hidden_states and attentions if they are here
@@ -260,12 +259,11 @@ class AlbertLayer(nn.Module):
         self.ffn_output = nn.Linear(config.intermediate_size, config.hidden_size)
 
     def forward(self, hidden_states, attention_mask=None, head_mask=None):
-        for _ in range(self.config.inner_group_num):
-            attention_output = self.attention(hidden_states, attention_mask)[0]
-            ffn_output = self.ffn(attention_output)
-            ffn_output = gelu_new(ffn_output)
-            ffn_output = self.ffn_output(ffn_output)
-            hidden_states = self.LayerNorm(ffn_output + attention_output)
+        attention_output = self.attention(hidden_states, attention_mask)[0]
+        ffn_output = self.ffn(attention_output)
+        ffn_output = gelu_new(ffn_output)
+        ffn_output = self.ffn_output(ffn_output)
+        hidden_states = self.LayerNorm(ffn_output + attention_output)
 
         return hidden_states
 
@@ -303,16 +301,16 @@ class AlbertTransformer(nn.Module):
         return (hidden_states,)
 
 
-model_size = 'base'
-hidden_groups = 1
-inner_groups = 1
-config = AlbertConfig.from_json_file("/home/hf/google-research/albert/config_{}-{}-hg-{}-ig.json".format(model_size, hidden_groups, inner_groups))
-model = AlbertModel(config)
+# model_size = 'base'
+# hidden_groups = 1
+# inner_groups = 2
+# config = AlbertConfig.from_json_file("/home/hf/google-research/albert/config_{}-{}-hg-{}-ig.json".format(model_size, hidden_groups, inner_groups))
+# model = AlbertModel(config)
 
-print(model)
-model = load_tf_weights_in_albert(model, config, "/home/hf/transformers/albert-{}-{}-hg-{}-ig/albert-{}-{}-hg-{}-ig".format(model_size, hidden_groups, inner_groups, model_size, hidden_groups, inner_groups))
-model.eval()
-print(sum(p.numel() for p in model.parameters() if p.requires_grad))
+# # print(model)
+# model = load_tf_weights_in_albert(model, config, "/home/hf/transformers/albert-{}-{}-hg-{}-ig/albert-{}-{}-hg-{}-ig".format(model_size, hidden_groups, inner_groups, model_size, hidden_groups, inner_groups))
+# # model.eval()
+# # print(sum(p.numel() for p in model.parameters() if p.requires_grad))
 
 
 # input_ids = [[31, 51, 99, 88, 54, 34, 23, 23, 12], [15, 5, 0, 88, 54, 34, 23, 23, 12]]

From 1b92564330aa2f40b065a0b9a2a94a28a595bbc6 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Tue, 29 Oct 2019 23:23:18 +0000
Subject: [PATCH 089/505] Reorganize and cleanup

---
 transformers/modeling_albert.py | 287 +++++++++++++++-----------------
 1 file changed, 132 insertions(+), 155 deletions(-)

diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index 90e9b162e6..c6662cb6d3 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -100,6 +100,138 @@ class AlbertEmbeddings(BertEmbeddings):
         self.LayerNorm = torch.nn.LayerNorm(config.embedding_size, eps=config.layer_norm_eps)
 
 
+class AlbertAttention(BertSelfAttention):
+    def __init__(self, config):
+        super(AlbertAttention, self).__init__(config)
+
+        self.num_attention_heads = config.num_attention_heads
+        self.hidden_size = config.hidden_size 
+        self.attention_head_size = config.hidden_size // config.num_attention_heads
+        self.dropout = nn.Dropout(config.attention_probs_dropout_prob)
+        self.dense = nn.Linear(config.hidden_size, config.hidden_size)
+        self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+        self.pruned_heads = set()
+
+    def prune_heads(self, heads):
+        if len(heads) == 0:
+            return
+        mask = torch.ones(self.num_attention_heads, self.attention_head_size)
+        heads = set(heads) - self.pruned_heads  # Convert to set and emove already pruned heads
+        for head in heads:
+            # Compute how many pruned heads are before the head and move the index accordingly
+            head = head - sum(1 if h < head else 0 for h in self.pruned_heads)
+            mask[head] = 0
+        mask = mask.view(-1).contiguous().eq(1)
+        index = torch.arange(len(mask))[mask].long()
+
+        # Prune linear layers
+        self.query = prune_linear_layer(self.query, index)
+        self.key = prune_linear_layer(self.key, index)
+        self.value = prune_linear_layer(self.value, index)
+        self.output.dense = prune_linear_layer(self.output.dense, index, dim=1)
+
+        # Update hyper params and store pruned heads
+        self.num_attention_heads = self.num_attention_heads - len(heads)
+        self.all_head_size = self.attention_head_size * self.num_attention_heads
+        self.pruned_heads = self.pruned_heads.union(heads)
+
+    def forward(self, input_ids, attention_mask=None, head_mask=None):
+        mixed_query_layer = self.query(input_ids)
+        mixed_key_layer = self.key(input_ids)
+        mixed_value_layer = self.value(input_ids)
+
+        query_layer = self.transpose_for_scores(mixed_query_layer)
+        key_layer = self.transpose_for_scores(mixed_key_layer)
+        value_layer = self.transpose_for_scores(mixed_value_layer)
+
+        # Take the dot product between "query" and "key" to get the raw attention scores.
+        attention_scores = torch.matmul(query_layer, key_layer.transpose(-1, -2))
+        attention_scores = attention_scores / math.sqrt(self.attention_head_size)
+        if attention_mask is not None:
+            # Apply the attention mask is (precomputed for all layers in BertModel forward() function)
+            attention_scores = attention_scores + attention_mask
+
+        # Normalize the attention scores to probabilities.
+        attention_probs = nn.Softmax(dim=-1)(attention_scores)
+
+        # This is actually dropping out entire tokens to attend to, which might
+        # seem a bit unusual, but is taken from the original Transformer paper.
+        attention_probs = self.dropout(attention_probs)
+
+        # Mask heads if we want to
+        if head_mask is not None:
+            attention_probs = attention_probs * head_mask
+
+        context_layer = torch.matmul(attention_probs, value_layer)
+
+        context_layer = context_layer.permute(0, 2, 1, 3).contiguous()
+        new_context_layer_shape = context_layer.size()[:-2] + (self.all_head_size,)
+        reshaped_context_layer = context_layer.view(*new_context_layer_shape)
+        
+
+        # Should find a better way to do this
+        w = self.dense.weight.T.view(self.num_attention_heads, self.attention_head_size, self.hidden_size)
+        b = self.dense.bias
+
+        projected_context_layer = torch.einsum("bfnd,ndh->bfh", context_layer, w) + b
+        projected_context_layer = self.dropout(projected_context_layer)
+        layernormed_context_layer = self.LayerNorm(input_ids + projected_context_layer)
+        return layernormed_context_layer, projected_context_layer, reshaped_context_layer, context_layer, attention_scores, attention_probs, attention_mask
+
+
+class AlbertLayer(nn.Module):
+    def __init__(self, config):
+        super(AlbertLayer, self).__init__()
+        
+        self.config = config
+        self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+        self.attention = AlbertAttention(config)
+        self.ffn = nn.Linear(config.hidden_size, config.intermediate_size) 
+        self.ffn_output = nn.Linear(config.intermediate_size, config.hidden_size)
+
+    def forward(self, hidden_states, attention_mask=None, head_mask=None):
+        attention_output = self.attention(hidden_states, attention_mask)[0]
+        ffn_output = self.ffn(attention_output)
+        ffn_output = gelu_new(ffn_output)
+        ffn_output = self.ffn_output(ffn_output)
+        hidden_states = self.LayerNorm(ffn_output + attention_output)
+
+        return hidden_states
+
+
+class AlbertLayerGroup(nn.Module):
+    def __init__(self, config):
+        super(AlbertLayerGroup, self).__init__()
+        
+        self.albert_layers = nn.ModuleList([AlbertLayer(config) for _ in range(config.inner_group_num)])
+
+    def forward(self, hidden_states, attention_mask=None, head_mask=None):
+        for albert_layer in self.albert_layers:
+            hidden_states = albert_layer(hidden_states, attention_mask, head_mask)
+
+        return hidden_states
+
+
+class AlbertTransformer(nn.Module):
+    def __init__(self, config):
+        super(AlbertTransformer, self).__init__()
+        
+        self.config = config
+        self.output_attentions = config.output_attentions
+        self.output_hidden_states = config.output_hidden_states
+        self.embedding_hidden_mapping_in = nn.Linear(config.embedding_size, config.hidden_size)
+        self.albert_layer_groups = nn.ModuleList([AlbertLayerGroup(config) for _ in range(config.num_hidden_groups)])
+
+    def forward(self, hidden_states, attention_mask=None, head_mask=None):
+        hidden_states = self.embedding_hidden_mapping_in(hidden_states)
+
+        for layer_idx in range(self.config.num_hidden_layers):
+            group_idx = int(layer_idx / self.config.num_hidden_layers * self.config.num_hidden_groups)
+            hidden_states = self.albert_layer_groups[group_idx](hidden_states, attention_mask, head_mask)  
+        
+        return (hidden_states,)
+
+
 class AlbertModel(BertModel):
     def __init__(self, config):
         super(AlbertModel, self).__init__(config)
@@ -169,158 +301,3 @@ class AlbertForMaskedLM(nn.Module):
         logits = self.word_embeddings(hidden_states)
 
         return logits
-
-
-class AlbertAttention(BertSelfAttention):
-    def __init__(self, config):
-        super(AlbertAttention, self).__init__(config)
-
-        self.num_attention_heads = config.num_attention_heads
-        self.hidden_size = config.hidden_size 
-        self.attention_head_size = config.hidden_size // config.num_attention_heads
-        self.dropout = nn.Dropout(config.attention_probs_dropout_prob)
-        self.dense = nn.Linear(config.hidden_size, config.hidden_size)
-        self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
-        self.pruned_heads = set()
-
-    def prune_heads(self, heads):
-        if len(heads) == 0:
-            return
-        mask = torch.ones(self.num_attention_heads, self.attention_head_size)
-        heads = set(heads) - self.pruned_heads  # Convert to set and emove already pruned heads
-        for head in heads:
-            # Compute how many pruned heads are before the head and move the index accordingly
-            head = head - sum(1 if h < head else 0 for h in self.pruned_heads)
-            mask[head] = 0
-        mask = mask.view(-1).contiguous().eq(1)
-        index = torch.arange(len(mask))[mask].long()
-
-        # Prune linear layers
-        self.query = prune_linear_layer(self.query, index)
-        self.key = prune_linear_layer(self.key, index)
-        self.value = prune_linear_layer(self.value, index)
-        self.output.dense = prune_linear_layer(self.output.dense, index, dim=1)
-
-        # Update hyper params and store pruned heads
-        self.num_attention_heads = self.num_attention_heads - len(heads)
-        self.all_head_size = self.attention_head_size * self.num_attention_heads
-        self.pruned_heads = self.pruned_heads.union(heads)
-
-    def forward(self, input_ids, attention_mask=None, head_mask=None):
-        mixed_query_layer = self.query(input_ids)
-        mixed_key_layer = self.key(input_ids)
-        mixed_value_layer = self.value(input_ids)
-
-        query_layer = self.transpose_for_scores(mixed_query_layer)
-        key_layer = self.transpose_for_scores(mixed_key_layer)
-        value_layer = self.transpose_for_scores(mixed_value_layer)
-
-        # Take the dot product between "query" and "key" to get the raw attention scores.
-        attention_scores = torch.matmul(query_layer, key_layer.transpose(-1, -2))
-        attention_scores = attention_scores / math.sqrt(self.attention_head_size)
-        if attention_mask is not None:
-            # Apply the attention mask is (precomputed for all layers in BertModel forward() function)
-            attention_scores = attention_scores + attention_mask
-
-        # Normalize the attention scores to probabilities.
-        attention_probs = nn.Softmax(dim=-1)(attention_scores)
-
-        # This is actually dropping out entire tokens to attend to, which might
-        # seem a bit unusual, but is taken from the original Transformer paper.
-        attention_probs = self.dropout(attention_probs)
-
-        # Mask heads if we want to
-        if head_mask is not None:
-            attention_probs = attention_probs * head_mask
-
-        context_layer = torch.matmul(attention_probs, value_layer)
-
-        context_layer = context_layer.permute(0, 2, 1, 3).contiguous()
-        new_context_layer_shape = context_layer.size()[:-2] + (self.all_head_size,)
-        reshaped_context_layer = context_layer.view(*new_context_layer_shape)
-        print(self.dense.weight.T.shape)
-        w = self.dense.weight.T.view(self.num_attention_heads, self.attention_head_size, self.hidden_size)
-        b = self.dense.bias
-
-        projected_context_layer = torch.einsum("bfnd,ndh->bfh", context_layer, w) + b
-        projected_context_layer = self.dropout(projected_context_layer)
-        layernormed_context_layer = self.LayerNorm(input_ids + projected_context_layer)
-        return layernormed_context_layer, projected_context_layer, reshaped_context_layer, context_layer, attention_scores, attention_probs, attention_mask
-
-
-class AlbertLayer(nn.Module):
-    def __init__(self, config):
-        super(AlbertLayer, self).__init__()
-        
-        self.config = config
-        self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
-        self.attention = AlbertAttention(config)
-        self.ffn = nn.Linear(config.hidden_size, config.intermediate_size) 
-        self.ffn_output = nn.Linear(config.intermediate_size, config.hidden_size)
-
-    def forward(self, hidden_states, attention_mask=None, head_mask=None):
-        attention_output = self.attention(hidden_states, attention_mask)[0]
-        ffn_output = self.ffn(attention_output)
-        ffn_output = gelu_new(ffn_output)
-        ffn_output = self.ffn_output(ffn_output)
-        hidden_states = self.LayerNorm(ffn_output + attention_output)
-
-        return hidden_states
-
-
-class AlbertLayerGroup(nn.Module):
-    def __init__(self, config):
-        super(AlbertLayerGroup, self).__init__()
-        
-        self.albert_layers = nn.ModuleList([AlbertLayer(config) for _ in range(config.inner_group_num)])
-
-    def forward(self, hidden_states, attention_mask=None, head_mask=None):
-        for albert_layer in self.albert_layers:
-            hidden_states = albert_layer(hidden_states, attention_mask, head_mask)
-
-        return hidden_states
-
-
-class AlbertTransformer(nn.Module):
-    def __init__(self, config):
-        super(AlbertTransformer, self).__init__()
-        
-        self.config = config
-        self.output_attentions = config.output_attentions
-        self.output_hidden_states = config.output_hidden_states
-        self.embedding_hidden_mapping_in = nn.Linear(config.embedding_size, config.hidden_size)
-        self.albert_layer_groups = nn.ModuleList([AlbertLayerGroup(config) for _ in range(config.num_hidden_groups)])
-
-    def forward(self, hidden_states, attention_mask=None, head_mask=None):
-        hidden_states = self.embedding_hidden_mapping_in(hidden_states)
-
-        for layer_idx in range(self.config.num_hidden_layers):
-            group_idx = int(layer_idx / self.config.num_hidden_layers * self.config.num_hidden_groups)
-            hidden_states = self.albert_layer_groups[group_idx](hidden_states, attention_mask, head_mask)  
-        
-        return (hidden_states,)
-
-
-# model_size = 'base'
-# hidden_groups = 1
-# inner_groups = 2
-# config = AlbertConfig.from_json_file("/home/hf/google-research/albert/config_{}-{}-hg-{}-ig.json".format(model_size, hidden_groups, inner_groups))
-# model = AlbertModel(config)
-
-# # print(model)
-# model = load_tf_weights_in_albert(model, config, "/home/hf/transformers/albert-{}-{}-hg-{}-ig/albert-{}-{}-hg-{}-ig".format(model_size, hidden_groups, inner_groups, model_size, hidden_groups, inner_groups))
-# # model.eval()
-# # print(sum(p.numel() for p in model.parameters() if p.requires_grad))
-
-
-# input_ids = [[31, 51, 99, 88, 54, 34, 23, 23, 12], [15, 5, 0, 88, 54, 34, 23, 23, 12]]
-# input_mask = [[1, 1, 1, 1, 1, 1, 1, 1, 0], [1, 1, 1, 1, 1, 1, 0, 0, 0]]
-# segment_ids = [[0, 0, 1, 0, 0, 1, 0, 0, 1], [0, 0, 0, 0, 0, 0, 0, 0, 0]]
-
-# pt_input_ids = torch.tensor(input_ids)
-# pt_input_mask = torch.tensor(input_mask)
-# pt_segment_ids = torch.tensor(segment_ids)
-
-# pt_dict = {"input_ids": pt_input_ids, "attention_mask": pt_input_mask, "token_type_ids": pt_segment_ids}
-# pt_output = model(**pt_dict)
-# print(pt_output)
\ No newline at end of file

From 67b422662c7002319e54679a73d654ba313702ef Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Tue, 29 Oct 2019 23:33:53 +0000
Subject: [PATCH 090/505] Documentation + improved AlbertForMaskedLM

---
 transformers/modeling_albert.py | 85 +++++++++++++++++++++++++++++++--
 1 file changed, 80 insertions(+), 5 deletions(-)

diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index c6662cb6d3..9b3c51fe25 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -4,8 +4,11 @@ import math
 import logging
 import torch
 import torch.nn as nn
+from torch.nn import CrossEntropyLoss
 from transformers.configuration_albert import AlbertConfig
 from transformers.modeling_bert import BertEmbeddings, BertModel, BertSelfAttention, prune_linear_layer, gelu_new
+from .file_utils import add_start_docstrings
+
 logger = logging.getLogger(__name__)
 
 def load_tf_weights_in_albert(model, config, tf_checkpoint_path):
@@ -232,6 +235,70 @@ class AlbertTransformer(nn.Module):
         return (hidden_states,)
 
 
+ALBERT_START_DOCSTRING = r"""    The ALBERT model was proposed in
+    `ALBERT: A Lite BERT for Self-supervised Learning of Language Representations`_
+    by Zhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut. It presents
+    two parameter-reduction techniques to lower memory consumption and increase the trainig speed of BERT.
+
+    This model is a PyTorch `torch.nn.Module`_ sub-class. Use it as a regular PyTorch Module and
+    refer to the PyTorch documentation for all matter related to general usage and behavior.
+
+    .. _`ALBERT: A Lite BERT for Self-supervised Learning of Language Representations`:
+        https://arxiv.org/abs/1909.11942
+
+    .. _`torch.nn.Module`:
+        https://pytorch.org/docs/stable/nn.html#module
+
+    Parameters:
+        config (:class:`~transformers.AlbertConfig`): Model configuration class with all the parameters of the model. 
+            Initializing with a config file does not load the weights associated with the model, only the configuration.
+            Check out the :meth:`~transformers.PreTrainedModel.from_pretrained` method to load the model weights.
+"""
+
+ALBERT_INPUTS_DOCSTRING = r"""
+    Inputs:
+        **input_ids**: ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
+            Indices of input sequence tokens in the vocabulary.
+            To match pre-training, BERT input sequence should be formatted with [CLS] and [SEP] tokens as follows:
+
+            (a) For sequence pairs:
+
+                ``tokens:         [CLS] is this jack ##son ##ville ? [SEP] no it is not . [SEP]``
+                
+                ``token_type_ids:   0   0  0    0    0     0       0   0   1  1  1  1   1   1``
+
+            (b) For single sequences:
+
+                ``tokens:         [CLS] the dog is hairy . [SEP]``
+                
+                ``token_type_ids:   0   0   0   0  0     0   0``
+
+            Albert is a model with absolute position embeddings so it's usually advised to pad the inputs on
+            the right rather than the left.
+
+            Indices can be obtained using :class:`transformers.AlbertTokenizer`.
+            See :func:`transformers.PreTrainedTokenizer.encode` and
+            :func:`transformers.PreTrainedTokenizer.convert_tokens_to_ids` for details.
+        **attention_mask**: (`optional`) ``torch.FloatTensor`` of shape ``(batch_size, sequence_length)``:
+            Mask to avoid performing attention on padding token indices.
+            Mask values selected in ``[0, 1]``:
+            ``1`` for tokens that are NOT MASKED, ``0`` for MASKED tokens.
+        **token_type_ids**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
+            Segment token indices to indicate first and second portions of the inputs.
+            Indices are selected in ``[0, 1]``: ``0`` corresponds to a `sentence A` token, ``1``
+            corresponds to a `sentence B` token
+            (see `BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding`_ for more details).
+        **position_ids**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
+            Indices of positions of each input sequence tokens in the position embeddings.
+            Selected in the range ``[0, config.max_position_embeddings - 1]``.
+        **head_mask**: (`optional`) ``torch.FloatTensor`` of shape ``(num_heads,)`` or ``(num_layers, num_heads)``:
+            Mask to nullify selected heads of the self-attention modules.
+            Mask values selected in ``[0, 1]``:
+            ``1`` indicates the head is **not masked**, ``0`` indicates the head is **masked**.
+"""
+
+@add_start_docstrings("The bare ALBERT Model transformer outputting raw hidden-states without any specific head on top.",
+                      BERT_START_DOCSTRING, BERT_INPUTS_DOCSTRING)
 class AlbertModel(BertModel):
     def __init__(self, config):
         super(AlbertModel, self).__init__(config)
@@ -274,6 +341,7 @@ class AlbertModel(BertModel):
         return outputs
 
 
+@add_start_docstrings("Bert Model with a `language modeling` head on top.", ALBERT_START_DOCSTRING, ALBERT_INPUTS_DOCSTRING)
 class AlbertForMaskedLM(nn.Module):
     def __init__(self, config):
         super(AlbertForMaskedLM, self).__init__()
@@ -292,12 +360,19 @@ class AlbertForMaskedLM(nn.Module):
         self._tie_or_clone_weights(self.classifier.word_embeddings,
                                    self.transformer.embeddings.word_embeddings)
 
-    def forward(self, input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None):
-        hidden_states = self.bert(input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None)[0]
-        hidden_states = self.dense(hidden_states)
+    def forward(self, input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None,
+                masked_lm_labels=None):
+        outputs = self.bert(input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None)
+        sequence_outputs = outputs[0]
+        hidden_states = self.dense(sequence_outputs)
         hidden_states = gelu_new(hidden_states)
         hidden_states = self.LayerNorm(hidden_states)
+        prediction_scores = self.word_embeddings(hidden_states)
 
-        logits = self.word_embeddings(hidden_states)
+        outputs = (prediction_scores,) + outputs[2:]  # Add hidden states and attention if they are here
+        if masked_lm_labels is not None:
+            loss_fct = CrossEntropyLoss(ignore_index=-1)
+            masked_lm_loss = loss_fct(prediction_scores.view(-1, self.config.vocab_size), masked_lm_labels.view(-1))
+            outputs = (masked_lm_loss,) + outputs
 
-        return logits
+        return outputs

From fedac786d401a9da31847b45df842bbee185afe6 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Wed, 30 Oct 2019 14:39:16 +0000
Subject: [PATCH 091/505] Tokenization + small fixes

---
 transformers/modeling_albert.py     |   2 +-
 transformers/tokenization_albert.py | 210 ++++++++++++++++++++++++++++
 2 files changed, 211 insertions(+), 1 deletion(-)
 create mode 100644 transformers/tokenization_albert.py

diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index 9b3c51fe25..a1c0d5f610 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -298,7 +298,7 @@ ALBERT_INPUTS_DOCSTRING = r"""
 """
 
 @add_start_docstrings("The bare ALBERT Model transformer outputting raw hidden-states without any specific head on top.",
-                      BERT_START_DOCSTRING, BERT_INPUTS_DOCSTRING)
+                      ALBERT_START_DOCSTRING, ALBERT_INPUTS_DOCSTRING)
 class AlbertModel(BertModel):
     def __init__(self, config):
         super(AlbertModel, self).__init__(config)
diff --git a/transformers/tokenization_albert.py b/transformers/tokenization_albert.py
new file mode 100644
index 0000000000..f2e37222f6
--- /dev/null
+++ b/transformers/tokenization_albert.py
@@ -0,0 +1,210 @@
+
+from .tokenization_utils import PreTrainedTokenizer
+import logging
+import unicodedata
+import six
+import os
+from shutil import copyfile
+
+logger = logging.getLogger(__name__)
+
+SPIECE_UNDERLINE = u'▁'
+
+class AlbertTokenizer(PreTrainedTokenizer):
+    """
+        SentencePiece based tokenizer. Peculiarities:
+
+            - requires `SentencePiece <https://github.com/google/sentencepiece>`_
+    """
+    # vocab_files_names = VOCAB_FILES_NAMES
+    # pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+    # max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
+
+    def __init__(self, vocab_file,
+                 do_lower_case=False, remove_space=True, keep_accents=False,
+                 bos_token="[CLS]", eos_token="[SEP]", unk_token="<unk>", sep_token="[SEP]",
+                 pad_token="<pad>", cls_token="[CLS]", mask_token="[MASK]>", **kwargs):
+        super(AlbertTokenizer, self).__init__(bos_token=bos_token, eos_token=eos_token,
+                                             unk_token=unk_token, sep_token=sep_token,
+                                             pad_token=pad_token, cls_token=cls_token,
+                                             mask_token=mask_token, **kwargs)
+
+        self.max_len_single_sentence = self.max_len - 2  # take into account special tokens
+        self.max_len_sentences_pair = self.max_len - 3  # take into account special tokens
+
+        try:
+            import sentencepiece as spm
+        except ImportError:
+            logger.warning("You need to install SentencePiece to use AlbertTokenizer: https://github.com/google/sentencepiece"
+                           "pip install sentencepiece")
+
+        self.do_lower_case = do_lower_case
+        self.remove_space = remove_space
+        self.keep_accents = keep_accents
+        self.vocab_file = vocab_file
+
+        self.sp_model = spm.SentencePieceProcessor()
+        self.sp_model.Load(vocab_file)
+
+    @property
+    def vocab_size(self):
+        return len(self.sp_model)
+
+    def __getstate__(self):
+        state = self.__dict__.copy()
+        state["sp_model"] = None
+        return state
+
+    def __setstate__(self, d):
+        self.__dict__ = d
+        try:
+            import sentencepiece as spm
+        except ImportError:
+            logger.warning("You need to install SentencePiece to use AlbertTokenizer: https://github.com/google/sentencepiece"
+                           "pip install sentencepiece")
+        self.sp_model = spm.SentencePieceProcessor()
+        self.sp_model.Load(self.vocab_file)
+
+    def preprocess_text(self, inputs):
+        if self.remove_space:
+            outputs = ' '.join(inputs.strip().split())
+        else:
+            outputs = inputs
+        outputs = outputs.replace("``", '"').replace("''", '"')
+
+        if six.PY2 and isinstance(outputs, str):
+            outputs = outputs.decode('utf-8')
+
+        if not self.keep_accents:
+            outputs = unicodedata.normalize('NFKD', outputs)
+            outputs = ''.join([c for c in outputs if not unicodedata.combining(c)])
+        if self.do_lower_case:
+            outputs = outputs.lower()
+
+        return outputs
+
+    def _tokenize(self, text, return_unicode=True, sample=False):
+        """ Tokenize a string.
+            return_unicode is used only for py2
+        """
+        text = self.preprocess_text(text)
+        # note(zhiliny): in some systems, sentencepiece only accepts str for py2
+        if six.PY2 and isinstance(text, unicode):
+            text = text.encode('utf-8')
+
+        if not sample:
+            pieces = self.sp_model.EncodeAsPieces(text)
+        else:
+            pieces = self.sp_model.SampleEncodeAsPieces(text, 64, 0.1)
+        new_pieces = []
+        for piece in pieces:
+            if len(piece) > 1 and piece[-1] == ',' and piece[-2].isdigit():
+                cur_pieces = self.sp_model.EncodeAsPieces(
+                    piece[:-1].replace(SPIECE_UNDERLINE, ''))
+                if piece[0] != SPIECE_UNDERLINE and cur_pieces[0][0] == SPIECE_UNDERLINE:
+                    if len(cur_pieces[0]) == 1:
+                        cur_pieces = cur_pieces[1:]
+                    else:
+                        cur_pieces[0] = cur_pieces[0][1:]
+                cur_pieces.append(piece[-1])
+                new_pieces.extend(cur_pieces)
+            else:
+                new_pieces.append(piece)
+
+        # note(zhiliny): convert back to unicode for py2
+        if six.PY2 and return_unicode:
+            ret_pieces = []
+            for piece in new_pieces:
+                if isinstance(piece, str):
+                    piece = piece.decode('utf-8')
+                ret_pieces.append(piece)
+            new_pieces = ret_pieces
+
+        return new_pieces
+
+    def _convert_token_to_id(self, token):
+        """ Converts a token (str/unicode) in an id using the vocab. """
+        return self.sp_model.PieceToId(token)
+
+    def _convert_id_to_token(self, index, return_unicode=True):
+        """Converts an index (integer) in a token (string/unicode) using the vocab."""
+        token = self.sp_model.IdToPiece(index)
+        if six.PY2 and return_unicode and isinstance(token, str):
+            token = token.decode('utf-8')
+        return token
+
+    def convert_tokens_to_string(self, tokens):
+        """Converts a sequence of tokens (strings for sub-words) in a single string."""
+        out_string = ''.join(tokens).replace(SPIECE_UNDERLINE, ' ').strip()
+        return out_string
+
+    def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
+        """
+        Build model inputs from a sequence or a pair of sequence for sequence classification tasks
+        by concatenating and adding special tokens.
+        A RoBERTa sequence has the following format:
+            single sequence: <s> X </s>
+            pair of sequences: <s> A </s></s> B </s>
+        """
+        sep = [self.sep_token_id]
+        cls = [self.cls_token_id]
+        if token_ids_1 is None:
+            return token_ids_0 + sep + cls
+        return token_ids_0 + sep + token_ids_1 + sep + cls
+
+    def get_special_tokens_mask(self, token_ids_0, token_ids_1=None, already_has_special_tokens=False):
+        """
+        Retrieves sequence ids from a token list that has no special tokens added. This method is called when adding
+        special tokens using the tokenizer ``prepare_for_model`` or ``encode_plus`` methods.
+
+        Args:
+            token_ids_0: list of ids (must not contain special tokens)
+            token_ids_1: Optional list of ids (must not contain special tokens), necessary when fetching sequence ids
+                for sequence pairs
+            already_has_special_tokens: (default False) Set to True if the token list is already formated with
+                special tokens for the model
+
+        Returns:
+            A list of integers in the range [0, 1]: 0 for a special token, 1 for a sequence token.
+        """
+
+        if already_has_special_tokens:
+            if token_ids_1 is not None:
+                raise ValueError("You should not supply a second sequence if the provided sequence of "
+                                 "ids is already formated with special tokens for the model.")
+            return list(map(lambda x: 1 if x in [self.sep_token_id, self.cls_token_id] else 0, token_ids_0))
+
+        if token_ids_1 is not None:
+            return ([0] * len(token_ids_0)) + [1] + ([0] * len(token_ids_1)) + [1, 1]
+        return ([0] * len(token_ids_0)) + [1, 1]
+
+    def create_token_type_ids_from_sequences(self, token_ids_0, token_ids_1=None):
+        """
+        Creates a mask from the two sequences passed to be used in a sequence-pair classification task.
+        A BERT sequence pair mask has the following format:
+        0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 2
+        | first sequence    | second sequence     | CLS segment ID
+        
+        if token_ids_1 is None, only returns the first portion of the mask (0's).
+        """
+        sep = [self.sep_token_id]
+        cls = [self.cls_token_id]
+        cls_segment_id = [2]
+
+        if token_ids_1 is None:
+            return len(token_ids_0 + sep + cls) * [0]
+        return len(token_ids_0 + sep) * [0] + len(token_ids_1 + sep) * [1] + cls_segment_id
+
+    def save_vocabulary(self, save_directory):
+        """ Save the sentencepiece vocabulary (copy original file) and special tokens file
+            to a directory.
+        """
+        if not os.path.isdir(save_directory):
+            logger.error("Vocabulary path ({}) should be a directory".format(save_directory))
+            return
+        out_vocab_file = os.path.join(save_directory, VOCAB_FILES_NAMES['vocab_file'])
+
+        if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file):
+            copyfile(self.vocab_file, out_vocab_file)
+
+        return (out_vocab_file,)

From e3ea5d1d8db36091ab840328cf6691c4f46f2e89 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Wed, 30 Oct 2019 15:03:30 +0000
Subject: [PATCH 092/505] Docstrings

---
 transformers/modeling_albert.py | 40 +++++++++++++++++++++++++++++++++
 1 file changed, 40 insertions(+)

diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index a1c0d5f610..ad8b979cef 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -300,6 +300,25 @@ ALBERT_INPUTS_DOCSTRING = r"""
 @add_start_docstrings("The bare ALBERT Model transformer outputting raw hidden-states without any specific head on top.",
                       ALBERT_START_DOCSTRING, ALBERT_INPUTS_DOCSTRING)
 class AlbertModel(BertModel):
+    r"""
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **last_hidden_state**: ``torch.FloatTensor`` of shape ``(batch_size, sequence_length, hidden_size)``
+            Sequence of hidden-states at the output of the last layer of the model.
+        **pooler_output**: ``torch.FloatTensor`` of shape ``(batch_size, hidden_size)``
+            Last layer hidden-state of the first token of the sequence (classification token)
+            further processed by a Linear layer and a Tanh activation function. The Linear
+            layer weights are trained from the next sentence prediction (classification)
+            objective during Bert pretraining. This output is usually *not* a good summary
+            of the semantic content of the input, you're often better with averaging or pooling
+            the sequence of hidden-states for the whole input sequence.
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``torch.FloatTensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+    """
     def __init__(self, config):
         super(AlbertModel, self).__init__(config)
 
@@ -343,6 +362,27 @@ class AlbertModel(BertModel):
 
 @add_start_docstrings("Bert Model with a `language modeling` head on top.", ALBERT_START_DOCSTRING, ALBERT_INPUTS_DOCSTRING)
 class AlbertForMaskedLM(nn.Module):
+    r"""
+        **masked_lm_labels**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
+            Labels for computing the masked language modeling loss.
+            Indices should be in ``[-1, 0, ..., config.vocab_size]`` (see ``input_ids`` docstring)
+            Tokens with indices set to ``-1`` are ignored (masked), the loss is only computed for the tokens with labels
+            in ``[0, ..., config.vocab_size]``
+
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **loss**: (`optional`, returned when ``masked_lm_labels`` is provided) ``torch.FloatTensor`` of shape ``(1,)``:
+            Masked language modeling loss.
+        **prediction_scores**: ``torch.FloatTensor`` of shape ``(batch_size, sequence_length, config.vocab_size)``
+            Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``torch.FloatTensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+    """
+    
     def __init__(self, config):
         super(AlbertForMaskedLM, self).__init__()
 

From ee20201d339b62948acd05783efc48c4e2b466bb Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Wed, 30 Oct 2019 16:19:49 +0000
Subject: [PATCH 093/505] Tokenization tests + fixes + init

---
 transformers/__init__.py                      |   5 ++
 transformers/tests/fixtures/30k-clean.model   | Bin 0 -> 760289 bytes
 .../tests/tokenization_albert_test.py         |  78 ++++++++++++++++++
 transformers/tokenization_albert.py           |  30 +++----
 transformers/tokenization_xlnet.py            |   8 +-
 5 files changed, 102 insertions(+), 19 deletions(-)
 create mode 100644 transformers/tests/fixtures/30k-clean.model
 create mode 100644 transformers/tests/tokenization_albert_test.py

diff --git a/transformers/__init__.py b/transformers/__init__.py
index 5c7b0a6197..152d520e7b 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -42,6 +42,7 @@ from .tokenization_xlnet import XLNetTokenizer, SPIECE_UNDERLINE
 from .tokenization_xlm import XLMTokenizer
 from .tokenization_roberta import RobertaTokenizer
 from .tokenization_distilbert import DistilBertTokenizer
+from .tokenization_albert import AlbertTokenizer
 from .tokenization_camembert import CamembertTokenizer
 
 # Configurations
@@ -57,6 +58,8 @@ from .configuration_ctrl import CTRLConfig, CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_xlm import XLMConfig, XLM_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_roberta import RobertaConfig, ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_distilbert import DistilBertConfig, DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_albert import AlbertConfig, ALBERT
+from .configuration_albert import AlbertConfig
 from .configuration_camembert import CamembertConfig, CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
 
 # Modeling
@@ -104,6 +107,8 @@ if is_torch_available():
                                 CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
     from .modeling_encoder_decoder import PreTrainedEncoderDecoder, Model2Model
 
+    from .modeling_albert import (AlbertModel, AlbertForMaskedLM)
+
     # Optimization
     from .optimization import (AdamW, get_constant_schedule, get_constant_schedule_with_warmup, get_cosine_schedule_with_warmup,
                                get_cosine_with_hard_restarts_schedule_with_warmup, get_linear_schedule_with_warmup)
diff --git a/transformers/tests/fixtures/30k-clean.model b/transformers/tests/fixtures/30k-clean.model
new file mode 100644
index 0000000000000000000000000000000000000000..c91b8acfa56ccfc80e1cdd854ddcaf9b6c44ab2a
GIT binary patch
literal 760289
zcmZUcd7RhN_s5k&L$Y)#Q6X7UnrY9TB`SONgwHlJpPA3r=l%K2G+9#FLzbju2}KB5
z$`(rYuO&r<5?PW$2?_Z<&pq$Y9goNF^VdDE&t1+v_uTE=bML!s=d$L9=OXbVWgC9&
zwk=b-n5hlvrb-V#_M~AW{`3FvVaK25=~iXi4L^C{u%R9(D{FF~MQ&C`_@MuU_x(?}
z`+vfRl)^ieZTs~Vx0R?J%bLFWK#9=5#?3BKWl)pHO8a&@EhuYQwjJR@bwc9fj$2T+
zLs^q<vY^{Z3(9sXYtA3pN?G<+u@o(miA&Sj4K7+B=k**@ik8esuMe&$QH89uTsD3|
z*^Xt+$RZ+{<q+h2fE=B@y_8pdB#^t-O)jN@ft#mJFHy-rc074TDQaalFY{uMAx|0k
z{Ir=Rifm$6LG4_&J%!aL3)QmMb+bxzp*m8K9acaaGf8=IzlTdnlLcw$^GK<bY6`u1
zGNhFfZII1tA1_hl^5yPxeAGl<K0D-@QX!T3Y+A0I4oxKsa`{QmmLdjJYwin>mS-d*
zyS00z^k;P>E-fc`vLRcP^H)HUNnC#ZY(XhiG$B_vUFh&b<-!54mWr)U<P$PF0d0*K
zEf!?_*RPfSE@*|e%zoV^OiF&|H%bvBq??j@vy^|jL3Y36ozkC~Y(aA0dor6z$eR7$
zEyblJzPiLy>4bbd2dY)s4l4WVWUN|(=K-uSsv~s?xpJ@f90w6;dKn@62YqXhE3UJD
zj*@F85VA^CC)3hx;rlL0LcZK)iT{I!KBO<C6;_$d2L<VRv;RMmO3Bta{(nkH=H-Qd
z0j$VKD#*&!<CiX=Uu~PrRLO*I2)8I}nuyC`g^x<bWD(-AcR*T;6?5{|AO6q!>a1Ko
z<YUL&VKVg*LY8?nk;up&?Ut3ei&Pd+k+-t|7NNeH2nT=R>12jZ^8Ut8O2uXp^&)>m
z()ek)qwS}qoTvw_6`ctXcGdc<bX&H(L?#P?G+753wn4U-ip%vYKPwfSt%H##zWKaF
z7m_sEz%|fTL4=Y%`is)PK_nTM!4Iw|WfzI6JeIuVfvoZ)->m_#^qH)5&aNz_PiJ9v
zY5A8W?qX>CuahCER)h5VnUGh2jJ$lws!}%f33=!iDB6P#{P?zCl_EwXv0B-2zNhQ8
zLq8Y!+C`<X;r|jeR+5!6SXP%3rjoS@d0{v-d8QiV!`BJIMn;zQ|E3fvo8g~RAN4e~
zOUu_Uert-6m)q77wA|xanZL((?hi)0Ej0wKDnS;BuHEnZlFVYR369DFP+V4ynonq#
zvKI7^crmY<*Ll&JQe^akN|d+$p_E%q5w&jJ2ilTE6O4WdxuMP@L99BPP08S|eVTM6
zo{+DOUh4|%A!AM?R1z}{&^86VYgri#6sq&tVpX;5GY2TFS3<@qW%|>Qh~xv=F2AnS
zg6RmN8wW|RK{~R>7lbYIc(y(x`~3}F(!D&ow_W@7r3mRPbxEBJZKX#Og?jBM*@Zqz
zeUffF<5OtU?o75`X4U>w%CeY=qi$R8^#Bz}FdWYH0E12{A$?Z>7+3WPa%{cR&rTRU
zrB#}c{S!+T8l?Sazm(>J3fcS<gfX9rXpikJ8(g}))FlX66GV_16W_?YK`f_SBwvsY
zyKgKdjwZBO4;285P^<Mgw<EMwD^9mMvA6#_C^Ar9Gts9H3Y0DjZuJq00i)9Mw?P{V
zxkNTcHG6ynVzP`>MUolBTiN_q$8k)~JeH6ZpN?c2<n>zU-O6^P*)x$sGMkB{WZ0!(
zmSvELWvSJOk9;a7puBY3ZPS9X)@3b8Rm>zAJOY(kurH|P9c2iRp;ex)PB4viUJh-7
zL`ecvYPa7?2~?$I+X2uz*alKHjG&379_8w_5Ykk!Dl3<4_j@U$Vy2Kx$xm$nEOu3b
zsr(Q~x>bcd+>4O0l&C|g9vSXas`N85c)CxSt*n$??)N0(k^e3B$?LQETG@Mvrz?}G
zv@C1&M@dh5%58^3SXqTcBrSWyp{=?BOr~U?w|#_E7FD?POXyM^^YUen&8`D!!8>0_
zxYV~}5yp-)MuUX8B{Q*vG#%FnF#cA{QCImiG)e*8sF(<1QY!}X#-C8eUv-l1z2a|Z
zE4~^<d|{hEOKKBM#3Cq{+*bx<F;rK++Y=J8?jzS%5;A6-ipk0pfT<D8=94<4AG`w0
z>QEU;CmBPJ_{t}%&T0Vzn{ROqj?#@9LeR3z!(Bn{y9gj`x&}0&_b7|YYGKF?1dXqN
zj{f7jknmM2_Z{$8smLG^39=bE;s}tiQL@=u**qB9ipga2NNv&OK0+#4nUIN#8ne(2
z7s$PDfY`#Ymz=%HN6%=pyx8jRQvQL`nSP#5MCiX4PJ*`1h>+32ylkEdWZcJ5?XCAh
z8b`TQgogd&X#jcwA{_mi1yP<&iwRmq(mF8A*y8DGom<ZD_)jUDDu$w*EZzrT_p+Vn
zqv%Vj7}wF~-cx+6N;h7Odl0~~Wx*GbMejmWIHK$SnFZ-KyRIOJgrm}VJ)w|!;<ReM
zwUnIv@-k>APcxcjWNRBwSL?pu;QpRYB`OPY%aPFGSjC!BmkrI&12VOVv0OuqcTa<u
zgsQSyi%YXUmxzuI=_Sy{R6R`f``G6Ypo=u&rz?C8>0*$KNv|J#4(SL>fa6>ITPieD
zOh;j3kbszA-0a)B%G0R`L+Y*LppElbHl5S{G2}`Q1=`YR^@y8%s+w$dMjrkE8bS7w
zCzca3o)ZXW<vK{@&TgaKPXt31bqcek$2JQ?Ad-p}(=xah08_1Q2F4ys*!YYl;Op5_
zJYCS5CC`M`2F=Uz8iJ+~mDyA(Td%V0lJ@bdRDoONEj~-O=~oei*=pHxlaEBb8sy_?
zp03Jf<MQXT7BeM_7ZS38Af5~gj06Rl`X!jPC>q{Ci|+U%h*e$(;e(a5I@Cup1zETS
z+9VK7RM8KHZQG=jYd(>pXOl;#GB7H}5_umX2%}YUY6lDTk$hJ|R%j|&C@?RM831C1
z7P48{coZb*85z$yfuQy9-qQU<h*HC|RQdB<PX|esP(N2fhx$nuV&1#H!6%Q&)XO1^
z_d<1+F|P9!9$=1R6kM)=@u6ejXLmqad6nAzhdt!uvpLJu%9Qy4VNDW|s$xPqyai(X
z?l1ej>(ke(0(9QuX*Y2kxqZ`8aS=8<2k#AOQ+Jgv`!neT67PF3sBs;nb(2&z$OGv}
zp7G|MD1fn9&^BvU1#LVO<-WlG$w(8GifPc+IS>MC$%2e|3<#aoQK0f&!n9XW=By$V
z){WtrW<B%=5GyB@WLx>bPta}4c5%ZPt1rJwxvc57rHC~}BGxzebiG#Ti5;MghpI%z
z^~Mj&e3U?^ym#9{TZae|xkAEq{KxkJgt^|bv3uj+46)i8Rnnq2pv|eQxX(Eh%H$St
zbNU`f16boCSQ#=1(rVM73y6_}p~*Zar=3p7<d>G;;}Dijy<1`iIRMKhU_tX)BU#BF
za7mtjt!mj=A*;hx?}CSPoybB?4jK<+W!DtZyPa<LQDSueL{&oWngV3SWKs3~ra@Y3
zM2c0q$2js{kK~Hc6eDMqB4H8+{J{@F8s|}FuUZ-KB!KZ&6Uh`Kj1~)D@tIUc+!+7(
zQjerf>L;uNF=nHxBK<e`7_sW2D&Ur~X0E?xB69Ff1g#BXS$U#0RM-Z&JgdRy_XV)L
zGKo<Ps6C*qT0uiVGfVq&5UWHLmW+kHA+0U?$f3tVSmqVwGGv&iQW3fHY#%Wy$D9vg
zHBDt>G#dh}0<x0Wkg*Vwd-EX1OFoMMXjPr3S)NqM{L4L^qC3iV<2}u^!rt^MXscXa
z7ur&NlaIhSSdfV`e1wc!H?4aTAY9Rtqtofd^T0w0pu`M@Prc%k={^NJ%WD8(wfo8q
z|AR6CMbMKTUwAUcIz90%q>UlDY_7<zNcC;MPR&bF%*h`6LKxpVhc(D$`vVwT(PA>?
zRwMf#0A^)Ulh?XInW!qMs*JDjDdLHkRpgxBK1D?zIqe9aBASgi$dsd?tx$G+Y?jJT
z{7(c{e<xh%>6ESpU%Cj|<YMELq$@q5Rzf!M$K^o6OlgsfR0aUn?kq3rl8o7p)_F)%
zsX1g_1CZ6N$a;P}l+`VpBahCv_<t+<N%K3QtcX-3kG!u_z{Fq3=H&0Y{QvO;JxG4L
zA0T9*K=Yw`ZD;y8RV?HRGGmShR6SEtcov}4El6?n2R?!>W#cSP075OH4h#sNd<w#F
z)mOHDZK2Wf^v{IoEYWN}BiA(Fu9Rn1Hzj|y@N_=QmT#{%&`h#4@9P@0N~Y1vxO}C6
z@#FH!90W~a<ua=)A*)_ApDZMUY8lzbr_SbeyZ_|?fKXq1O6(XY)2<kM{dgW3cNpZ8
zsKyoK`BMO_z<8GRP}w;?LON0dP!j<#u?H+GyTl-2Xp|H*WPik&e`hOc<3*($RtaIu
z*0Iae?$WW&|KC@7T?=81)~031Lr|7e9vK||EF|(#Q}zOa#$Bx%`bNCwX?kd_T>dV!
zO;)*lLYE>;a+BZpk<x7SrREc8%Ra@tan`qxCYH)XH19^^O>4lYRi!+zk&v}WB+$8c
z`)#%_DL|o^&wz~B76kqnF}jx#wAeXzGI_aO0V_)_9WB~<GRMEq?Fnhhq-#3zdh{R=
ztDr6e<;t$m##|&7cE5hzJ(j1>GYrI~rXQG9Jf5tq#1uxwA2<TgxS^Nk<8t1S0HN~I
zD@ix(L=Y=4qt=DmlcBAlupbABRHY0%*XL51RMW>21uRWv0+m^t@M&TcTVx<Rqz1^?
zE@ZJb=$@rX#^+O))d68u&I8!};kUNl15vkp?Qj`@Nl*8wl}ehH^emUvS3-sDmdzHT
zx+1H+&F7%z*+8D00uU;Uu1vnW51O{Z{#G-Kpa~<IU=WMT5=AU>Jc)2_eHc0{K2k3)
zJ>tn6gV#lKAx%^^KL^t76(BY5!R-9S7RK%)&3+^t%DsTutsuYr>=SdiP{<;)Q~w3B
z(yJ4xoa`VwloF+F*Y{&PkWd%vBedtw9RNa2hQSNUeF_c?w7L%|Vx6#`yj0<n=hz|^
z<nLnuti(v3vx_z-dRmQm@~P7Fg8uS;ln`RWxIE`tf>sx8Z5p=r9uRAYc(N`T=sfr2
z3@~Gr4H%r<F%!Vba{IMO^P!FPO7!lcS0F9dNGzWX0{Qbz5DM=nZI%(Tyma>!$nIYP
zguPrBT$~MI$N2%!n!~p830bxt$ZFqT+HQa_1`4{4t5w=^O}RBeQ32}6>B#C9ElU6C
zjy#a(S^|UvX@42o8p?{oM4c(ncmvx2S?(!anr*wMr|C3tIiowYm6qj1BO}M01Yi|r
zO&mvL+0#KxA~8%8(;7+L-+z8K|5}+rE)mlWd;5q_P+#33HF=+7CEcIZ;FvleA<z8!
zZiCMv!TLsLjG9Y;OhPsC@TEQj%zWBIue{R#T_wY>f(Uy<a+H)`4QVyTTu_j+ulID7
zo>ncG2yG%`%sBLB|8GIf$`2`R@<4=)jMb9>tW6`a7}J-k?`|`EtVmi{g1bKIv*}sR
z%JUwjQIdK|KW3p3Qu!<)Q^zb;GB;#p7J*rDY-%WO?%M#yePt1T4td{4sA53P$Xg#a
zMqrgH)ytqQ*HFDsA;rQGfLTrTzV$agS`;HnCV?h@{e#a6TUH?JmA0m55t3rpandG_
zBr@(;VdZZis*HJ=+oG%_$co-_{_apFC)U1cHDWf%+_r$B_UWqd^Sz)O6*R_rf7*UP
zR;aTt<fYC2AjocjeBa$CuZ}R{Kh?v>)rA<^cVRh5m?52Btv1K>29mO!*>lF}(Wz|C
zZIb^!1~}9bj1pHI2Wf(@&#KY(`V#=CSf0b|GYA^j9ARhVw_(sGV&-mUBUE^=v%wfq
zb@;D6*T+xkdB|fIKwI8v*MrAj1VG+wLPkd{m`F*7%LrP{vDxZ=q3UuFQ_XZzy4(mA
zszQ?OYX3==v=EWWVUQE6in~FrUKx0aJ?68h=q2+PSdw%^9$rk)a>MMJk!wHkG#XnW
z-z<j?6`xup<eM)6j6JMus`X7*d4!Ebf)V!H)gV+1%gW2&6AVY93O4>eYE>ey@=g3=
zQIip=+vxv{q%Z+8Vl>;a#2iBeHsa2nuGabFg#)0Cmt^3^u{KBiCkFlGos*%%VW^nP
zX;)f)IuOiZURZgyPajFSGpno41F`C|*<(oQ838Cw7A#+w`ec`ck7Nf#6Ker%SQ;Qd
z6(B;17HOZcS9ltU_LSyRp^a&0eF&sR5$hv~c#&PF_ROcJ0a~7Qa>i_ku;)hdm_OsP
z^Bf@KR6A~!9Q2Hj!}(4j&q?9JML@<%&+_vA((4UqE1Ie>XvuAj2!~2)3Tw8MI9B1H
zP}Bu!AWJ{>={U*LIdseSAfcR+38tqzegr_O%miIF_}F^D0<XJn1St&y9PVU_fMb3I
zq=Jzs3(*z-LYTTnlT}jRY^PH8g{0c~_Swz@96Qm9Ew={<dqIL;bMB5lg4uL#+Svou
zMb)&GyLq6-P0O2F1C;u2xHi7{Krk<{0WzpNbT})d<>+4iUmbO2b#G`Y&D!I#WBgx;
zy07#--u{Y5WcpzLS6b)RcZNY*p2?)lI@2OzhUqfgN2K*QaUB62j!kODzYwxbSt~}c
zSal?wLra3{Bvz*=xV1$gTVSYrQ~^sD!BqP|0@CV<McbsaxYDP|N8-t>bgBig?SDL>
z)&n;>KBcJXou1bwIRQIxG>{iOd*bV$ZCs56K{l2&^^M*DOzXN+`#Y5|4sGq&=SH7N
zjFq$O{UEfl$e~h_?QerDn(Y(ivkg?M=Ntel37aTtV+*OmGr-1UB*g@U&7kQ^9>PxH
z7I<S7Dm5%a;4S}&63Hhyxso>*5e1grKyu()Mj-5ZS*L{c8@547jN{ub1qsI$UEaXn
zn&m(!1FV0til8s7x19FXf9Udlvhq8hI+xE@<q_+RYdoZ9CfcC~{q9p$tCjbvZCaL^
z6)V4-*9_7M$!9Ur&1j_KS(aPdHRpe(q7)7{C+|SG)bXp_@Uk<AwKof#M7Rw8a2Fph
z#nKoies2w8!%$Xg+IV8P^xf0vKqVNr+_1I26KJS0bbig^k<LKY`fBDc$n5R_#(7TZ
zFMInmsz3==2!qe?P(9WHU4b?k=2O)s5;-|90d1V6WM1I^)B`42emQigaqdj4!&M%@
zltOk#UJqc5>grM#!6PRCnHJ`<oPMjEE}Q71sU;p;wcK<Ekg*?RiBfkbv~@S-bB>ah
zaV0B7-AbN59mvG*c1D5By5C2N=54n$<N+Y`2fv)t9=Bj^L`KXZ7<O)sell{yGXPe4
zTulfwp0&71wmL6aFrSs@U-8LkT*k+17W!oMdL%mj18C!tJ`>X&$q^p{8JAUwteUJ|
zUebuzOt;fA5Mzg>9F?#5#7A+RKPAgP2eF*diyEd(Nw4;iYS_t9){q}SLQy;A0c-gq
znAHqDiOW)@5eUnwzc%;`botGQQMM7prj%Tk5rfm|lw|(^v=JhiQDd!*MSZsdhayw2
zwx@U5xkS^>1!>+I+BlA>nGIf6w*@lgs+D3#2rDa@)-{BlXT8-KFdS8LIp!o%{nZ7`
z3euIbr1yt5X~o&##-(ddAEB5-dNU985u#aL(cEwtfaMz6Ap>cC9GJCBT8)(j=|0$J
zgPJB}-btQjiP2ZSy5PTbANlJ-pF=d_&dr{Vf*5ByM+dBFe3S&%PL%e>N}owpGM|!E
z$|s3pXpKv|kv>UKOsiqM%Q%l{qp*CMbt8z0R*kbd9L&4fL+Zz(Ej{3NAnQR@WISv#
zBqJAw23hkEVUv9%!(Ib}WL(xh0caA5V>PSF$h0RxO!Cnb*1SnCL8CkL>f>Mbe_=I1
zoQt5XFpf*5&)bkDaXUaI|GPf`+o7yUoou(5kd@@TyD~CQ5u3exl}qk(DASX2S--+Z
zuFOZ&bh_^<5HDsn=94x6s7`04&F=(_4<gHhfBL8~4x<BEwOy-{=&(Jzwb@T)K$PaT
zY1g%aAS=wm)qDD=&g8`w?Uv5=FDJ{Jx)8LXo`s=iq`La+L7>K3+PRuM*TVxrP7f_V
z>jx0_M?H0ZxIZ*fsgda6giI=&8>-&+KguVj)6rwkRRC6<4^W>IAi_<+0C{96lu0~M
z%(G*o_V1qzRvI+eOrhjG&H%DPf{|D`6Efgz5L*B-vQ)BHWdx#|T{YY%R+l{O;*Ujv
zLcue>;=@tn0oGvbHd?2AB36Uy-69`kfWm6o)@D@RN3sPop3C(>#=Ubk(HWukN-(Qm
zJTE&>fHIXuJ7f7|6mgGH&=jRwj=vScB+GgzC%aFAHdTnSgYI|-WXMOpLHo@6_ke`G
z0OJMwS!(ddOu(>R0T}Nxpz9w6wVLZnIWE6E?sH?=m|<8S{FINvYKx^ppQi!na11W8
z;Ta1jgR~UpHYR2B&VY5y^B~4~E>cWM@0Xxa!fYTbUh#ji9WKby1<+RGynL>NwQs--
zuYF^;_koO2hQ>rzcKe_)1J%Eul~x5$OS8rPuPQZ$y|conWvXB$-fbm-Nw5KP<%*wt
z{CZsuRR7}t<_t-?Y=X4*>e)vIZ-Fv#r<3tG3-8E(kSyILBe|`OaX8?oS3S~Xmy#X?
z+J7|GVcU6(HI*%hy6ymG5_3D9^ILeDMJ(>^EupC;UgxWK@qg)Qirlfgr*&g3|0->j
z<g`H20%@8U=V-TD_hsYSf|*e9;3bcX_X9A&JLe!&@!|>}r;t@|r71@MgdJTCKrN1f
zHi6_5x-pVDM}t^VY+)FeDe9ELU@(q>;@Oi4hP{_l0M*!I&Id9^P*;|x_f=YS`q}XG
zfBw_G>@^vHus>+mAD@Lb&2|TkwjLgr2enqiRs6{!lnH}ld>LH_Y2~R{@jk5wF!eh^
zKE4#f__uycp>K|}2zVq^$%yNG@=7MEyqr7{fG(ky&LwvfHujt}mfd6jB^kJWyw4|u
zRd^pg3&8lQC@+`Uk3pJfSm-&giR|M(iMkZg<(>ebEOq10&mnAOv5)K{-JgZF?3le|
z%e(*in^VT}4=t%S$aS9(wo0;`tt#rNRri(lKl^wYuHjJ6F%VP-a@$uPaTfMKZu%O;
zWSq{+h;N~2V*IP8tnpFv>YPxJSJ#4A%@U*7j?!=X`~>8gFP9HCc$%r7iG8C_l8+?S
z#v^}h0y3szYG2}f@-Lr=gVDHbY_V%82IESR1Dua{1+kJt&mnoLjgQ2sQyQE2L+yQ}
zG`ABXtbuRe6Uf+O(G!=)IzUrB<g|ZBA3I1gDAF+x+XpD*ERu<^k^fu~OOIDifxZQ8
z0!PQ`7bw!*-N7`^tc>jKGho-K7QvVLd8C%@G?hL6NS_5Ww{2)2RLlx?4hM00=s1wD
zf4Q^yyc{qH$n-t#{O7hC0$|J))N614P-s(}fF;7{Qz4DlNIY(Po3hiu&^w(l=MN)n
z`MJYZIrBmQ<06wt#NjoW1qnyZ3i&`O`=5D+vp)GW0HMb@KkU<s5XMeicWc*O4oz8V
z{@vwD!enHt^XrwelIb8!)#qO80o~;%WcwQcLV;lWLXn!@>=8ZGVK6*z3W$k_`5>>l
z|M^rPnu1$2>FGXx-n>;_QN+ZP!dqnRgFbnzn0Kd<`#%a)8c}q;dxZj41ilMQS#i1P
z>Hk1MPS@fao&};*_9&k`Pta!VFjE$0kG%k>f<PZGdXca(p=*S%=R;Z>vA8KD>sZ@d
z{|1<G!Tifs{`>bmz(7Yezxf0p)apb6*Rs!{{ou*@rZd_1Tn*}#D#oNgYMn>0GOO<1
zwB92)fT((Q{RzbKroIt5=nv?S;Vg^HKmFfnUBylL3%a!OaW;1HSg4@YfDsTcv*cC)
z^rWB{I?mf+w~|oZlu3e(3L(KvQV!aQpmB|LTKyS@w*=9|dYfghwh&g;@O}ddB(ErH
z1;!IPrnYYfV6|XO#T&3awAF%jq}<uTlTpT-BRfJ;JOjk&a-R_2fk0|2{u8r7mh^=T
z4RS233;ECn5)1%02J}>)1HeN8Omh%4Iv&V($N4z)w?w7<d@?||#H(X)beg~5R3D38
zOZ8tl4Z!+fMK?L_3@EDzD;H^YCS)irHbq2#?_7_#Rc}TXi~unmqwY~kX^@E*fl*^d
zjnN5$CehMilx`*z4V5s`K!0jp17H<0J3ba0g~)_+*>-6kUHzeO^H#v3GorZ-!$ld3
zjzN5R-bc5i)CF}Oz&1vAKj~77T$`1+1}!TF^}McLs}w>d2nW*h3ZIrG6Z1gxu>fHw
zBmv_#Q|0vWfF{OxQm(!l%ETC8XPa;hq$)sChFt3tM>sv;5E`MLub`<D*4A42=sIZ2
zj7g^YdMINM$F({=mwtE>kO>tRd|SO|<OfBKG2I6z3p)16-GJ6S`Gk6ZsLJ0D7OGQ1
zxqoN|fEC!Yx6FRhC&8K#Bz44mMX|6(oTP44GL!-9c(tb`<bY=Z!$wW$j;PNI0LFPD
zmke0=%AT(PSyk~Et(7i~G^cXhw0PD3ALzRLf!BQ&oZz|%@vb+3kS1q(Ti^Dv@#jg&
z=Ecy)fgUw*Mdu?Qp;AZO7rucuj*5YeW81FtY4k!yAloZYl6RJYEs*2ad#KXQ7lHiy
zE0A>!ypQ4xn?0KDUa}Q2J3C{M?n?e`1|0IFn<JUNJ%HuU8sB-COx@8#Tzhl1JgyTE
zqdFt0e6u%U6I#W90p$|!;^{<+Y2n`kpp6%|WDMli3J@zaij^dgiM@RiJSS`A&;dRP
z`lBby%?<@A%@q-rrF4Yr6tjY{jThvuqo5IeG$Pp(2wDjVclLO|5TBlL08=k@FAN1M
zX-!m~xd76#W7zB^Kg4_<c#`2Ja!@sZRfJhMz=ZyM7Q}=#vY3q3GIneYe5zPhZ%@2a
z^Z=ckmZ+`s0M{%L`7WaYtZsN)GpjNWZny-@n4t@#<&`neR#Jw`W$K6f*;pX!7KJ)V
zkApJis)~A|AO~OLlSC7#B(i5$4rCo8R08_j>>B`0Rv3urZ<kC3u(CPkEy#s;LK}b1
z@j$KrS4{J9blWZE_d{FpSlo02H1`3Iq#K+bojU`Bfr~SSQL^7d5HwbGL~>6M49n94
zo>_A|jT2cQdp-wkT;ocgw0`#SB0%Gs2?dARKVJ8#@F6J3ON%|N-bym<Lr)hPa$F2r
z4jrncGt0>A&p}MaYPZ)NZvG1(xUAv0R)hAhnrn3dv3fO-3C8SlNHh73k7ft?fy`S2
zWc6dG##QAF(AIFY{-J+ESw*-y%W2Br3RpopOsFf#7h8eCf;a<bOvfhBtaS-XX6mvj
zbr-u~J21-&Lkr8?E;|5NUS<r5%LT1GWX7yOF5bmMID>IsStW<>3S=ef-rSjCr7f8C
zmpTT++mu8$47mI6LD0%Y)vz+IZx3L?tt@J9mGe7*7$=b!*EpFs&*=mdj_W;T!@f`^
zrCJ=>W#E3$VO@%7tDJw3k3hGH%Yq)zp~{)<;kF8&q$a|37Wu0efN73e%FO4ibzeZn
zF8t{wuHmpln2C+OaG;>gUpPMK*_1krZ5jYtl5GPG_w}IwR(@b6f`<<SK_Ym*d@#`e
zi5jR8cbj8;`WR<TDS700fKbHTKpU+$1zHaVGa=MyCu%tanjCqZV3Y>!uPSZYMMFJZ
zF`$3BbUoG6%w2e_D{btlS3i{<Fbssu8X{6QoS-$O_LX0h<R7e_UtB=Yioq68kV8g5
zhuRpLcqS=iQR-PH%P$u}TRX;!a(n_xh0srWRr-kW2>u$2YN5lt+}++l#-@Nws(Se#
zft|z0iA8KI99i$7N_J-I-Lhm9P$+vnp$O#Au^<$ii^%a;5wtwh%<3umaH7wEJPYb%
z`sY+2<GENX*Uy5o>?&D?(6~=L0%8Kfh%DnChcrE5%46@ZJ@9}zJ{49s=X^RB#3Yl_
z9e4e6&{kn{+2X3aLe}b)46byIxcqOvPZrgM!oUTd4w9oeym|}TDx}T;)w1oo9;kHF
z+i4#HSV7MFn%QvJCqQ;Mq%5QS?tpLPx4>b=*eiUeq}7~_rTW+0yB>t@p)We@vXQXW
zr&8^TZI!lSumDDMBieP7kHRrgQqKL|)2U=Gkad4U(=P0TcWox6ex+M>b=XG1&T0;7
zWfZEJQ?K3uTJ@*DyuBkqD~`R8JMucNrH{=dT_Jz&0&V<9^(?DecG)9L-&a<*_mS~g
zWQTl!r%BvXCLHK#Tsl+IsT|tc#*H6(zxRRuU{-*;)5)2}VIVd<;Ag>>)Kz2Z;h@GV
z#&zee^Z1cIVj&T$&SX>Bss_1ppwE!gU}pgu>Jd%?l8iv#pXM{uGo(1HtPjrsGTuT5
zw?M|94Q6`5T7xzpIvfDmvAFqFgES*6J!yLZ;ZQp;AM5x&QlYSqu$5&FxDbF2QR&=U
zURBB(g%h^^GCb;|<K9TC-Nr^Y%#zzoMeUPw6*A4@8mW9=?XzGLn2-b0(D23v<W~(E
zKRS1-m_N+{S@rS>?a_3L9Re^Dbiz4Rx}ANGI#6THttu|6hqeZC(`8oPyu|0l_Q%%I
zZN~UmdXb8DxJHpsx0SzkW1&ljsrZCochQPII1U)Os9$_Sve!eobuZ67OduR;MkJ{X
zdG}2osnhEU_%~1T2u3>H%#5B4!Xn5i&8t%&Ol0b%diZomYm{8I4u)qaU_4d~=vOYs
z-3N^fv(@s}41!j=ZdEV|-#6R-tcpm(;{<JV>)A`X&4IFPF`;ln{O%_K!m6v@w|^Si
z^3ofybfQi3L40%A1a|rYFk?2#p~9#|kh%?3;eGWEgtaKnG{`NGpB1u9dAdVA)sydo
zgo9{xo+(oe%Kbh7w32Z8h~;@uMD|+jlVXQZbv^wf5ah094gJKY;BsItqUWJaz67%x
zVU|cp#VVfzRwZ=_8Tvg4q8lBNBi0ZMyK0dgBwVyu2V`7u?VHsd*a7Q(9=hP8?G|kS
zfeQ_vw#lc!=uzi9Y6kxXW*Ji`Yw5V``@7GCYrc7oM5q7Z5snU3CI0vmBy4tkb66lM
zVBF{y+#N4%_qUJ4+a)OG@+NIc))<@@BAMi9voBt-EqH0FQtxM7K(19VQ~~zG)l$Aa
zfVECMp&lZ#Z<$A!V{|`t@{S&H*RccHucgn1mw=GbTMAe<L6RkCUe0L+V!XSJmRfPP
z><(sfQSTOQ(2rVsNRI?_#JaVehn%Q@Htq#v&8h}usow|Msxh*tUM>Ch^$}3*26<WO
zum#u)>lOYZ_xDi9kklXGv(X#gTtHL6%E9c1yKy&YtCl`9K($`!4x&v@Ax|Gd*b33T
zn(E3!y@0I4p}%GQpe(z9@mn4{9NMy@uTp<$IS|C8RE?#@Iq~*V%u;c;F)rT^f<|Mb
zS&_jO<i6O_(|t+?O1sZE?<}9PDhvBZo(FAWVu#7-dbR@Ab?dNvyfPe;oEbEpxPYLI
zv8;34#S|uwmld@d#yI=X1=BGXd5AR#+sV8{Vjg0JfjY!xr%E6O1=q$0)IeB;SV{7r
zkkZCQO*SemM?;p1rjObh{*6mPjrSPcE-zEZ0FY(9T1Z|U3t^1uV6Ox8wc~)OOe!hQ
zY0%PU*tDeN;OhWP@OpiQ?fH#2f*21RzsMaoLt1|7O0AkX<rbet4Qed@{+teCjf)5I
zQ};tzAuLLe_H8o(Pz`OfrVskWY@0N@`bR*lJ8ExdiA0V)p8~Yz=XDa^%3*c1_8BlM
zrBAv1I2X$JtYU*zGY^tVVud*CIf4|2OQg)3Z(+t@DJ&oy3XaO><<^A&RyP&_m_agf
z%SS+D6GUYBGJ>TsCaz|>23h<GSlA8Jkay~5p4O38K3@S%(Yh3@{)(^_N5v^_+*$xD
zF3lAht@}yqfx>FyM1=oVT%P&~%*s&bc=_>XpAVKob`xL!;Q`#V7=|AC3m_b0^fimP
zJfn!^!P8=l0)In?TH!9o2GX`k+Y-b<6T6>4_HPPQ5-C1{+d)zp9V4F6u>DobT#?%e
z((=cmg$GAVXv;s9jOKNBRoBWxmH6XoZ3naivf8@0LUi1HdQUKXNJdKiUW6zvg{Tr8
zd}3C-?ym6t9ev_FHV)_K@lY2qV?!k>OZWH5Ftxf(Ui=^*g^4gDYm~NDQ75KAUhd`r
z)|Lzf($8nYDwPZ4Ol1Qa<7hKUkHZ1N<wIY2O({56kEik@EKFW9e;{FNmwLTvnURHx
zm;|sk<r{Pf^K)Y?{07v;{Nv+&EW5qW3iZho!HjPlF=^s2hXPQjKGJs9sf4Ydh#S;p
z#4sOSmudLYodpnT02_b3J2Oe4uoC^H<JmqBy?u`@?EZ5+f}60idck=hHXt#<mfZBp
zE(A5!*z(jhNVkgttd^y{pgdL$7M5R(Ijd~z8X(HIt8#5d`Uos@oMZl)1|T{kXD}^B
z6Qs7v*J+m#w(|6XKac)h4q)Zsxs-9E@z-N4EiNhhjP-vr1>)<2DyFXlGtJ@fr6Aub
zZS0|4MLq7l{Td*OR8Qb#w-Po1Gg7Fr;-}j{;4Ph%4!2uGJs&&g4#JkNt3qUoPl!gU
zP4(C5AXXA%o%6hU?s1^7A^ORKbD*?b)}g(gB5VTZp(vFamwn{qxu8~GCV5tg44BIm
zvw|6v;}Uz;a?u09zn=HGIF}Dw20izJ&xMJtpa&sx>MLN>ihCZj->_h$T5eiG&;+hy
zHg!<I#KALO`Z~tMr69(VT?fQ_eey?O#ynF8$E@&)O?sI}bxuq=pDn3e4r)apC-nUv
z1wxg^nFi#7FF>q@j5+<~t5wj*NnLDr{D!azHN|PY4uPE;AvNq}nCxCyODtm&ewetJ
zo7O1iTDc(2))Nky;4q^g{eSk^vb%P2?)nQ565{^msT#CqOUGo}4N%4&9_$4ft~9f<
z+mfBR5yFN~%$4%mCLgzAK;Lqi`v<fY9CvPls}-=?+h$LKtsqtyjt2+->nU7<m$Yc-
zioy~Sm$5rRcvV&1u4w5ITo%<W;|T$w(t3NiS4YAoIyD+<=UChc$k-|1$q~q@`!wc2
z-N@nmeL)Zdhl~XchEj9Z+gV*az#DHIUOb_IDMD!{z(V8bgFubZfU764zOn~M*gmR`
z>MwdhPcRCK6y^6`1g#+YJs11;ILzmuu^Afr9SdS|=CZTIj`t~W=;k2k_Y(n(lakW|
zIxyuFP%BqoLgVZ^CMORAGkG&7T&q-xGgW3(ey5AUoDc(<9U*MxCK?hlvI^3kGhywT
z&yHd)!l}MNwy*Z_QtT-1P}11Z(^$OT*o(`O6sWZboBvn@&ynU?AlrJ=QB4(z`EU+6
zlt>Pr3|Xc?m{gZIKu?bHar6<SKyGhnjDz!xnx3~G17v()wbEPDbwqq_JSY{^#~{XB
zN!YZT1#?<nyvnC=5A(T!Ri<Cp7)uwUEhl)IF56dLyaC$8lw+H4(5;XZpew2WY1meQ
zrI#ZDx%oE0(BgnUF`G^&#XggOjhSj1L)zaC9cq$W%E)X*RBLMG-n)EcXSe)aX_J|{
zD6ma@z^CTorXClrSHz@f_N%-cGt)=n)iizTruU;hAAN!;kVTIHP-b03wwX)ND#Pj(
zmpNS!ZSyQ3wbl0kFMYw{Vh-H$B0=M)-W}xa_cDO#T0X&{nKt67idl_Wf+ytlSD{Ui
z+DXb6HU6y=@PV)SEZmY>7XTMGV)PzXBP>?pZxPRwJTXcZyaQ#N<RW^WkounobR`<d
zg^PXs%82w>0%f&~yE7M^hw7IC8u#iG<F48qy9~^@Pw8!(CM%$wVFYt}b0w@)#%WI5
z|5U(o;N}}+61{QlmtfYy+|5u4to;ha@=WOI?X6!!hiw+lDKD)CqCvWLmAAels9Pl6
z*?;&wMA$mcMMJwlehr}IhpP6IxtpNFF-v!IW{UkmQ7eTTH*Podn?jV5%5vL)Af-6h
z>?gNEm~4BO%MVTWC}qzCT5mL|DtgP@ZTYJtF|~NpSnoIU7{dowX=shFntP1irkms{
zY25<M^nlf_ENcmAmB5silbO4DI+NhAWTVnB$Qp*b3A;m}k6PXvZ3u@l#}tm^0lt3i
zz|5Hir#hRx_3&|BCrhVO%R~DRHhI8PJc%3ZMf=*{szASMSZg^z?%dDEWMSh*rUhN?
zU#_tJ)!l+@EspO&(DXrXEoJ2RasX?4?$~LqA5+BIMwbwYyt{6-M^8|zgSv;hs|9!T
z1~Y-M$_(Vo!=Wv6HHP!n<B<ReK|K%e9Z1*;au(Z!{G*6vi2;kvG<|J{L0~qvqp$wb
zT_A%_@W}&ATa5M#6*1}1inw8<x%%fJJGzWZlT(0<Cwe<WVY@~^mvRj<Qvp-h#3**d
zZmOGkh6nxXN_w8<^WxNsCC|FEJ;0k&9JX&bAHXW9k2KKsk>Ma=&(UY|m{;a1WV*}g
z^FJ5(2x{w5QJoqGGQHygIJr?t6Ej{OfefpJwnEgLg(h5^Y($vq7y>t>d=@&x?UwTq
zIIz(kd0!sHGS$N)c{_kMIpH)0<3ASwObygRge#A>&!bm>ny|tb!Dxx{u|6##MGUEN
z0AZ!w5d}AV8X=zIbvIs~zLHqzFN59h8Z=dK??CBT*Wp@^={ZGOU8jyw%&MMZsA9Nz
zRS~O)9%snW*ZCN{Ql-nWt&Io^Rb2wCn?M|^tFv#)0XG6zXVfe6v1AUj8jFywH-TGi
z$s3!b&S(QC0$PcTWj%3S_PKGyOZIo(^4|!3WWnvw#<aVp!&9u2J)*W-My!YL@M)4k
zG$Z%k>0`L_+d!VV7bF}Eih1|U-SzhYSv~YA)j;l4z{IFaqP%)aG`-(PitF=Na_>w4
zV<(=)fR~V?AM{9|PB^>-qlh(rViaSIT_N~*mXGHiH;v0?Mf^C<lG5$%UVa$V_`>^#
zA!Weh046dXA}dJE6Q0&Biu8UG+UjOo+EFq=A(Vq&UH1$@V_5eZXyKo8L2Tus*O8LE
zMxiZp+_S)8zoB>TKOfTAVAsPrX4jWJz~yxM&6ouMx<sNwv|2<sR4qMDkIScugk414
zsSkYJ)0Mgh>G%e8s7676)&HZn09<xR<{iS;$$Jivjqm!TF}(uU<O83SN91UwvzG!`
z%i)M-4|To$5tvmc$}Vi5WspduHYw{rB^b&pyq|vgav)<r&|PFfuKwKOGwF9<LC_EB
z?)aF_)aOf3;~z5z6+ZbZ0IM*@Nygl-eFSE5*naE3fZBUTuOWt&nof*La?NV;I-du<
zqqqF=laIkRM13CK`vt^0l{Plk5|($JHUoxj#g&FY?r#J*0W9w(?Y9uaB;?*{P`$cq
zoAyp5JREv*Q-YQ&j@=yX=))QJDrk++t3u9i24zx<a}P>K=U27|GX85YROhPY{uUr6
z1AU4|UMTa4>{ud@UONNnyxU8TXho0~*Ow4S?dD^$UUN5GH@60|TJg*cqgAw>2h>s?
zNL>d2s}+x?Ch*7(<X6RvDLMxx2>GQWNU85f-5t-{_xCxe#VU{`3Rs`8MI5~r5nX>U
zu*pU5s%6|{e)b_gKX<-B(>|_H*pg@{!rQe6h%v9%>uRN-fN`OZr|Z&qK)Hw1w@lqb
z->mQ%v53>zX{92Tksj6sQta!aV5!t9PdEm|8VU!B>yGvRGlZ*Cz&|H?B$IWXxfh=X
z!eWcXNFc4wfS?)Fd^29dRvL<f*KV(~0E}(zD{9^Ab~X@Z(fwtY5rj%w<($;w^7=&{
zE3HA~$rzaR#dIVli6o>+D^SZ*m26VPie<NqL8W&MfECO8cJ2wMrbQrY3%eGh33sda
z@kVp@qxR2NE(bCl#ACQYre6VVrA09<a)a&gn|z|si$liV41_@3%Ix%8{r~DStL03%
z4ajP$ugbb{bkHO)6Fa`XE0wZ3X7LX?^A1Skf}UI<Q>Q|QYMG(ua%A@ah-IlZE(Qd)
zwTf9L0lS_~Gkl^13sa0>O=g0ap!D>L>rhVulxmA*zvolXVSAyjddBk6LLl0g4ZFl&
zwXoV)hb#)iC{^b-eKNP8rkM|X3n*N|gl{E)PEpjjR_9>nzcT9`FsqwgNtQYP^RZG9
zuEFY!?Pc$S8E5P|-PZBK4}jX0?TSxfByM-L%oHaJ^q+77=qA&b`s{)t3ap<)nDB+i
z^f_4fxWsFUSt`!mZdeJ$NK=nf(JF$zHF`-zX%nOFR|4txl?S+s#13ST0+v;(C<lBE
zWi?EUim8L!DT<gJ>XLc3eVoj_@C_J(*6q%<-xH?rnxq`LmY@~Jtq8`kL5+a3F3FGU
ziDB&FXb{K^KSNvL>9pMSr%z9BL33Je0kFXe3mgZ`YOIaR9)ANHllX+FFX&$X_>`)0
zfsEVgfv8(+22J-Yl^o47alE9oRf~MN^{wZR&B2U+#)O{Tq{&VI)RV=jEZK#y70M&H
zY8}0ESCCRa<@GOjxc<p*9;9G)2|C!+@9xv0N-?Z)t6GDYj<_uWwOT4b2oi-I+|GhE
zS*h$qkp7D1bd{}p8=lL`r~48#c336r%WwJ49!bQVH~;W1AmM0)xhTCqqzRv+bl$c-
z(EqKkzPIiMS(+MEz13G_se&dFJ$|p19~uE<)T^8PS?-e|dOcpHJs$1_7<MT&3Nq7D
z!asdLjh|?Clr-%NX)@q&-8lek(+|k?U>;xU@AF{Wi0g=Zo9~PPptc6r_fTvpNiR9(
z2=G$dl@H(+MkC<79AvQqei))Q=-;zPf|(R~giu{V4>=0NG?F`!6$AbMTtLfF&R55R
z&`R8OEI*N8IEK}7GCJ7Db<S;hxqb+U@x%nj!n@PSmKbx{h*SK39HXO6T~4+Cf}+IE
zA!rR1Q^%JV&Vx2-mhAeh1uwq<6wPq9iasMCta9PY6&$K5X4$A`xU9Pfx}<aF->@YC
zguK-=`L{|7s}JVFBw?!>&(6wGHISB*yVqY(+Hz6@ryjIUX@u;GL?E|hiDlKdj~q13
z`D8WDvo0ljDP#?xwj7F{SOlQ~V3{ttR;f^BSO9cufHX!JKY3gC68|?gd)e+%NXsp%
z8_{aHqY;Vg1I~FFIfgi;j$g$n!^CmI6=23GgENtS8S68uw0HgPALpY4xYhGA>6HM+
zSryj#TdsjLk?8KD3Uwbe0SMu$W5(qgE~#fl?`|^uMgXfxPH(m*Zt^sJuU_`M724Qz
zFQ)1~v*m4IRz0(5O2zFUmaRHyG{`gsLJ1@|F66b^$sVcX0uz1u`Kch5wV4c2>ocZ#
z48yazZeOUFv6!vJ-mq>aG^)>hy<(OHE0gljV+4(bRJO<}>vIK&hPVG+kNf|ax)8(B
zPXbt$Y-rf6D(wbXzQXg2|1YiI3i@)c{ikEoKN_}l6$9`)_-P)rRUtI$1ai%@VAcZe
zu>+FLeGbTY(-DSE?Q1?C7DRlA=?!QTmwulJLG1NrV-l9@897CPQbVvX)U&*Ki+v<q
zL{YPqp8?pY6cnRDf)l}vE{>mC<rA}Qz&L`=KL54H;Ju1xh5{L~+GDzrcAoRQd<$mG
zV^?7n9bfAKj^gMXr~cvr#^OHR<g*O`3<=x>u981~gRqe-nv@g&fU;_^`!-j|BmV@W
zrP%ZSqCpd$8mrSx{FiU>5Mw220=Z5hYX<eJm2<aR1}t^AZL(Ks_F#(3Fr%`8Keaia
ziPv87keDJ?DSTV|ca!Qe4+NaX<mAGZ0Mvm3;<Blg&%jhwGP{F><1-)DQp!3ZR~zIb
zC9M{9xTxuqMVGhsIdIUX`;6Nau}W~A2`1b6L|iiG8d5~gYUiUQIS!EX_VhGP>;vR4
zrNgFSDb8MRF955kdhoCq|Fkm*vaL(XYy0`sjB!<RQ5Wd2Db(1m9%?rq0A@MnottWT
zcb^ESdtTN##MAW=98w1M`cDGgP}TMUpg^8KS=x`FaZ_*iiau&Y*iz__w@ML{g4>uJ
z)*m|D`<E`{aVO+3Q2M93haGjePr*RKMF*vg<s|Na)zbe+4`2si`FG_p0LDi|R}XQi
zKFK32=L)n>w=+G0-l#X!&<jAU5i!0ntnp>Ti-D}snW(mpL0Q3-*%-6JU2%ZY=+M8v
zJXiy5b>Rj_7H>_~335U$sEJ)~#o-!;J8=puRC{N=WEfE>EGsE9^H5d=*3ShBTcwC)
zV<um!bxQzd9AWMx(l28GEMI#9F_5w=fDom!GG;vCP;JrjXRd;@LfmcpyqtZtN7y)O
zPdHZ*V+)nq_8KTF5VI_;|LOGrCIa_ufc%ZnwwO~Rs(rL<k%Gn*kJ&Kw#bxiCd`3)L
zblE!_fjC1`L{7ii=N!|^hbt#RQxE+d&`-Df4Bfj(DEIIwK051PK07lF+BjuiV*z>W
zJpfksnruQQC~0lQ$ft`kHR^t<D233uWYl!R#x>6>xf7Rd?*|GOfr(5eXh?C>hkeNj
zGr;Ksd=_x&Lj+AAyphK%tB*okv3liZ;cQ52MZ4#tt@5O$=JAm?G-zBivi9vJmpu(&
zOz457cH_Cv`ZzWC<plEAD*#r6KDvF(8<1f)XE&tJE6NjZ`!ss5fG<V;55St9=|OwZ
zo$rEJNxb)?HcOeI5Y@^T<@5ImTD9DB_|Gl2zu55}u+)OQFxh!IL1UkNSAl{0GXUcy
zSKVOe3&TGLv!Yq@a4ahWzXma-3G7);G^EKlfL3C@h!wb1NS7RW&sfX<G?iLbTdwmN
zH15{+`^9Hat5)2Pl{PWyMt$B!|924K3AFE(Hn!B=gQd~d%^uM`EIbY0;t?+BsD<h2
ztss`Edf?99rbDT|`1`oMj6B^8%!I;6^vL($<^XWca!qb*;s0d)tj{Vn0!&o>y2*oO
z#4tHjau6rG?FelmRo^MMIoWP!u+olD?+L4(T;9q@4BT3hb-~`df`ubgk|r`Qfm0Q=
zEZuz>O?`7)FqqUAh)!!y7(L)L>h8S=8aGw?tYA$CXsaA9`Bd(dPCf$eBYYHQA0Hu?
zWD1aOoqdFW$E0P9(%Oz}CEIl&94Z1gpv_fjxq_h%X+PEDp}GUWLX}Tc$+-tYTHE26
zmSyB#)QIrq0o@u->rNcXVW5<gdJrz{@7*Tc-IJcu6WA)J57SAex6hWN6Uvs!eL%wI
z=C#ga`$1am`5pltr@I{vV7#jF*u7vf??{g^9wvBBA}+@b^a=IFOV;dz0IUZ3q!Bz$
zKNZ9VUfj$kp95vRvl^cZ&12O0K93-qM|wvO_i=RS)>iB=!pCuwhRmq+KuuAP&bwE6
zK%Zm_WO+3J<sy7CXj0?<4x9_+-6;TLBu;JAe9<olWYTPOWWnMyvN5SX*Dv?v0cZhi
z1Ts&<Ca5$^yriClbq~NO*-ac0D?)_@VdKz~t34Zl!k(mwWXmN!4z5ZuUHLwHxsMYt
zu_R=(0w`}RDa*$YG$GrWq3l1_XTU;1&!psJg{%<giy}j=0x(vIl5&1lL$3y7vB(lC
zBgJbV;M*Q*d-^(`o>v=KSB$vc1C_~2mM6Iz0Zh$U)%BHICqf%r9Aj|dLLxVVSds2#
zj}FE^C>D+btli|Q$<QVrr{HMBB~yGxI;qmS3ls@8TA$?DeVUJ<-jk%MQp7Ymir4v%
z_cW4`Dp`6TB=vN6b3RtW_~ZykO+Po@4`Q;?(};ktmRNi>7|Q6G5Z0S{f01|i^lWU|
zgMh|1t9?!s@;vGX<mJ&%-k9an=@%T?I#QFP9|oiu>D$X5A!vN?DNi-<l+OVP+l&L+
zT50n<fMv<iY*NxnTLdn=Fd_b=fa)Ss_N@8-&m6ZEvU1}~9>CJ5?eQ{zwFg@#bp{CJ
z^jCZ&PL5z}_yPdyusD$4w+PBts<$+G4ceGyp{^&`i{J7H<GQNe+_yofW1lW^@O%DG
zOp@wmKj{OHIFk@FmLeuf*7qE%2XfS6AS;#c5~StqMp~aFlHf!Br`jxerDCa1&Y6K;
zG?RvpfMB3!g-p?4sDiw=$B7i5oB9~c>cD}nZt3?~210&%-y!`eVJlnD>*TgZI$mUE
zOE34SosA`s!e=1XDlxtBB<(&2uo~4Saadp&obm;b@quC3SvcgU6+Rt`Nf*0$C4eb$
zJX<ACukvY}ky8h<_TTt4am*17(p>@5PGpaxf-3Z3-+E9#M50!{f*kk*pfxNEUGpQ9
z?sfG7@^9;Wd~B}lj@4W;`6ocD7QGigllOk{0KR}IbKi{smLqr7vPvO!-d(<0gP|h3
z1{?G{K&Z%i_Hsuf&2dm56`T1#><juLW<mD;)2FD^b=_%yL6dt8E%Fb+k{YW~oBlXM
zF<TLEhe&UM>pVATE2w3wPwBAisA<-*)HLc{q^FagHTMvvH`WroIkN+hrNb`9T}e5j
z%tzv|-FatE+8L-MA6@ed6c4cKWzDlo8vv6Jmfi_%{olnr*8$|*y#RdQ>LcwtK~qI8
z>`vL2pj9V)2v+YO%~Z6cM7RvAOWeZF9-~j|gT;JT{17l>zlt*oT6;}z5YrNUY{6XP
zcI^Xb<ukyiWaqv<E8FFwUI+H`SS9B(&VBf<L%~eFxiRMU<ToD<WcjoHi^QGxaBu`5
zD&}k%OOJvuX*u61=|0f@b>G;2TnX*5y`{r3J|Zq&7<XnXV8z)d^%y!83e_oxva~xE
z+5}yL-B>Bjte5P0oKNel-AvE-o#@lDL#vl|LwpqXushq@4-{fJzzij;hY~hf>LDtt
z^ixjuNUrF<da_Ot>$iCO$?(%5O&Dx0N0~`~je^E9M*|JA<C#7eUFjm~A;Uqep6<(r
zj74K2K2c6@JzrMo6V)W@wOf?c_!!&)E69#Yn~LD_ir*m{{+}~o#vYp_`7`U2)ad}%
ztq5&3Y%nAKF?BwZNG$Z;d`B^BKYe^MkRvV!2y3a<31<n8j`4Z9eLzMwU*Yo#5AGVI
z!`Q}HFwKy>KoOcOTPMekCuoAgf{nlC)jl%bdTIe2sffj3w5^q?*LoUrV1q=jgND`c
z1yn8v-T*-Eyj*zKZ3IhWN0v|e(1lh`viO`4>FHn~LvIJLme2=*be3D8kd>L^2^i@-
z)zj`lo%g0en-Hm{K6<p~UJ&cT9NU_^b0F_b_wnf1Iuv$!5X3kNUk?amn}@(kHN-yp
z%Pb#<ywpXC_fj4Pvj(BoSQeg}4PecjMv0G|<MY5?la<?_^nY{YAWzNxZ*tW7fO-D^
zbYzsQSJGsl--Feu`;2FS=oUE+Lw0|TFnt@Djh5vvLKrLRo*c+#1;S3lTaa|4aq~e!
z<;G<~y%Fjb0$C+dRj!@93LTDl`ns(RHlHo>vDh9I<c7CA%>q0vFZ>VM%2pe7T7q{0
z!j@pw!*byJMx-jm>tHEq_8xJl38SN&rQt9GcS`-J0^ul&yCc><Hcan>nQSql$o@+`
zS>R;wuaA5d>SM|`vOWf|s&O4ECHF4(5nvMa*`k0ksdw&>#_`{Sn1Go6+&gJwe+CL0
z6cg5a8z60v>)S^j`4x&P=z^xxCc@SgxuiCl<vL@<GzDoPwjp$hKOw9<3{Sd{sNDh*
zvZ!y(_V^1rRKrvQ{va`2ApiE5`k1)+X2MpVj#YmznXGik72m^a+No5Id%`x53$_I@
z@nQ(T6!f<O#;ki!TKDfF+xv9cjJmH#popa_)T@c&9|cT6`oNUd;qaC~RMA;r58v7U
z&wAV$X(#UP^Wc~pXRQ}o0~oVhB)y}pPod74nqy@<AeAazewRH6+kQ|xo%6B=9<Q(;
zL2Pf_s=deD=?{03Wkw^WpNr6|a&l!y;)T6HzktN3|3oJbajM{|^}yahFri;d?YS@E
z5)*o(%!~;Ig)O7d4~r9R4{<}rxdct@3}mft?{G@nu09e%a-Jgj?mm(lkXZOt9O4ny
zQPtdmI7JcD#5(kCWH}@Z>f6c1{)Da9vXW!q!~3ZqLk9RP_|~&NAG7RGkLUo3Aln=c
zVj74cpk9_A0Uh=L_hoC&1&#zVC5RW*^Y<DBEDlc`U`A(tI0^`6)a&?`V+f-VOk)xG
z>sScOKdoOh*yT9rQXg>NHQM`l5X()M@~Qw=C}h=9=OMfv^*&PHL7>J5#wr^3uAx39
z>r7`B_~9fVlaU&b#-0jkYGRH_?7u!%%s7Z*R%GNDa+=S_o}>vR(+Dxps(#6JrxVMH
z4|2H7%F_y%9;4NXGkmr+`j{K`u`@x87hP!Xr=&@OKHF1fo$YBxUL3p5g*IMwhILom
zdY%VnW$I@%xSu&3fHL`BDGf1ygiqpb2dO{uDHnq=&DIp<1`S$G)x5%@Xigl&^5Z=_
znChSKIfP$f3FL<=Fk=)e4QC|PK91U<+<QD-lOA*1clEygtOm?7w>8?BEP$1dLzK=N
zyX8Qvt<gVsW9Tl0%nYuMRqW&GUhJO!R2+>`<LO&@!eP#lDq6lp0AMWY;!dtp+J-YO
zRq9n{Cf7F=wBptCk8b{HV>F$uu?eD9BS#Spf#s5N%OwQ0OEG(VH`ZW|9fNTMt$6m&
zjDl^(1DJlYSJn&NTK{_$waVe>LqEhIa;1kNQT_T1pU0a3WL%k#ft;y`=@HLG?R_)U
z|M#+DayV80DJcjKf2d3N;kSU9j@97JvFELjmbtCc<#|P@9p`)B-bT=>5pf5TH%<bu
z46&&&Vj%Vblfg`q^eMfmz{I<13Sijb`68XJzqcr4Y2D{N)kEm_d%&y!lSrp~0m8a+
z3_#D`(un9qJCI@b5ht8BS>@WN)9zNVRG@Am1L^vJkB7ws%dCDF{)idAfNExRI^7eB
zP+wk}+V3HPw5C37dqx=6dnMm%xYQB*4UoCBpiTZ5a5gJtMPsUPO}6}D59!xj^wRF#
zvwa@)5gzb;%;$k$#NChkf1TQYtu&l)rYu*@A#C}xRn!HRjDN~QJUof|;^y)UP^j_A
zQ54v0o(J^9&vCg@0n3L^eshyDE|)I=3OlZT$#36Rp-r^NFk;_bIz&NBr~QpqkQWuQ
zw!}prvm|*neI3k%!m{<oH=&FjeJnSSnQ!?ddN||O(6it6k)la`+2+jm8Ud=#4MGKs
zL2e%4g75*fF{mp%hCE8%cQK$TjdKa+W27H~c#-KHCV6owP{>eHXTT*2lp2-e<0Z>{
zF77KefgJV;h}BYW=W!iBE<3LBse)>D{O2fb9ow0e^d2iuPbz90@zDh?v`O%d&x^SM
zyCL24JH@Or?1r^Ej0y2?0j+=Pxk2rENaHfl+k0y_`o9Y-_s`!1X%%&!3MIvy-#x^b
zrdA|5`wyV7?z*CpwwnP;b?;vxPj2-QI5lA;Jmy~y(8XB!RyN(cR3kMW&{p%Bfs|Gd
z*s$DlX>$1Xp!BAyqzv1^M<_jpf;O*i0cvW(Yd>%71ZAA@Z3=uJo3#Wm&e))(B&W1R
z&?!8S$9M4&Fj1xChFyJxN<1m}R7Yz7IH)elLJgMGv`X(Su#HyCnkHRzM>C&yY((^)
zXGS{i3u02``T>tZ@F0~!mW;y!eX??4XAoQBHC_S2?6E&^NwocX$}**`J$VOGIv)UO
zS+jw58<#l;`V@S3hmQqxg|?!iygq``eR;4?L=V(QUnJQbDAdh3$G|$oOgaRNK|?=i
z+PVkfP`BK36l!#wSpjIgsaKAibd;xg!K%O9FwoQ5X=T=N(8dXFfZFk6$Ag5k2ZAa;
z(I?=9BqddYp;^f3*Qw;_Ax7xwr7Tv$+EEX;3i4+ofK3wM-l4>Rd$!kd@yQnE7(xC$
zjj$;U#=2k6fHL{jC!%yDuHX&>GDUQs^T7EOoj>wS(6ArTeuK|}wDR=@d0Yq12QXPM
z(&gp0;hxqjbaM9yXcHh_XIKJsiLw4-KodyN5Z7rUAMw$c>G8<sa+G|oU>J>sH9W3r
zM7TVnv)I5Waf~7K8d;_lutaJ*#7H!)5(J5GnM>wsIBbd_uOARtq=;3Jxl^u7LZbFP
zdrGmEplK$3ITM7Ze$&#xCdfijeo>0}xRnJNf|R4JFfvQn<ZCYJSd4N&rfTkdTpeU5
zkF?aaMcJo7u%z%@Hyu+CZJe?bVpAypG+5+FR4%^6|4A>>!Mj$Guwf$l$b7R)0Zj0$
zSutVjrmpQ6K;wbol$8PNqW2ZE+mLW>FCjT=R@W=|!|I^#bgB&`KMsg0>TYg=hK*qa
zmy_48gbwS+kuc^74g;?OGZ7{-IDls}F6SMu0k*ou@e;w7euW~X+V|@%S6%C4(A~MT
zB5fytSbf|k(=F47-vDT}#xokz%M%J%Yeu6nxo9G!@x+|3cUt%W>CIqPl)g<S1C+Lm
zx!}cHS*mFBZv`|CD^u=uw$774tSGL7FqB`ZfaQ(cP!MLv@9qGzKIE)fc0*`~yL{3r
zZrTL$paLdY%;{BXzFMsidc@<RtM4IbW%H#WeF1En`#`*O`pS_?TYCM5sy?0g*8M<M
zaSRP|;DeAf6z?d@w1<7_>LRk<;SnEId+e~sAT2Yu4?uedKH;IjJr9M8fI?PV_e`Z4
zDP}+EF=r^${`lcj9xL4tl6fzHg(HhwF*zk@_adkXlr=2F>iGEp#t55)sD3@cK}e;d
zi`?-F|5`KYj*1bT(ms0?&;-M}rd9^N1x>YiDo*~@u!*slaTcX(-}8u$m9YjXTMR<}
z(V|RSLeTOry~u)x=_eixidDGf>e^+Ig2uRes#)Io!UHTL*-=hk31BsGPePFBnJ<Bi
z8zic;$c10|RQlmjj@rKluz>{^{WDiX!5U6a-)JyYJ*MVrdHNdwYYd&d^~gc?{tnFe
zN7dO#=`b`-L5rxXemv=Uk$4T5aZ<2%5S~!Pl5(b!#E7p9H-a44*(GUtPeJ1(U#!AW
zal#KiGkbzcdj8^(m>zZTG2=#rw`=s_g;O>Vhh|D*`MYc*VQa#AwZzr@3T-Uo+Kwdz
z%R#5#eH=b~oRD=&n|u+aI|18qGZ0y(IMaAe!<Hp}B$cu_{7=6DKj3elG(0epO$u4p
zQTzG7N}7Ba4%9UB>Q)e%f*UTA|0QUm;yONkS*Nq5+w4=?Nmt{t!S02pJyC<tnt+?M
z;`t(r_m#~7d_V3d9k=%p^f-=B^(as(n^VS={LsQ9_K2QbQwCz}R&o#Gf|iiRQR(F*
z+&No$P+uIbayF<(b^|Px-k0}IcZW79>cfNDpRQ^RWGWcrMveFgI@hXcdyzun<d=}`
z+Cy26*=i0xTkYj(os8vHrLE)ZL90Hn$Yx}^qE@b1C%8$`0f^QMxCzmz6Je79ivV?9
z&hG8AVYXvA?`|<2v5$|ZQb%QPQ^e%WYMYVooX!9yDjvR*{C<!n0rc)8pDJylL&aI3
z;&?CZy8zPIeY?n^2l@DXDkv?@yF$Yy3jx_j!$fPyO8LPS?Acf5XwW#&3+{PYp+Kn3
zJVb`q6L+S1fEgQfRNe~f1??)Il*0xP4jY#z@lHGf(nMu0&|G3T5{R{-n{Rp_?f+*b
zTF??(47C59j8aOJYTU1ne6O_S=T^WtY9D+okO{%vq;xmM4?GUg+JHkM^{I?00+Y4O
z7*jQDwNzj426=RlkM6F2=Vkl}An1u6Z}EDSQkFm`Ku#=&0F*j8OJWsEL@_OaN}M~C
zFn1>pmJ?4RWCGDu?CMh>ja$rvw3@D>7M})awZb~XxmR4yIvvPLR5yk|ZdSnJu+4Bb
zqSqSZa3zDc+Bcj<9OEh*b6@v~%UkDwS^1PEr=1Jw1=>r_9S&{fnA7?b3K)Z>H(D6u
zF7P1F@-x8dba%oCKpo)gWa5PcE!UdJ=+Sa_1R7B=$H`I+TZVcdYzFX7Q4hLTmo-b7
z83VLNi`qDQQ5A?)+I{_%F4?vg$hsug`_(e04hR#|pcKd<hxEx1HYqTx7qn10HtRFO
z`;CLixXe_@O5-8nB3khBND$*Re9nRqWu=0~0?!KSZ6Ensp^^-FglmJ+#u8Q##(bx6
zN9RE;f4!84*Wwe3ATS<PD+>sE{fRQ!VT37ySxa%>U%tB(k}7ctRrb1!u$4Gk_3@_5
zp^a1hb}GguG<?JuK;yKYSzL$1v&RBiG5R!v`~1#a#jM56Aop@(B=xOxwzbT7BH6kZ
zPe(WGX?|ZQ{NL+mY;$z3eer6aCx`O5Q;xe1fU2^=c<OqB#ui?&Y4y?hdjgQvRQH5m
zP4ubx)=qs~rcMShwiz$&R3$ML%tVmW1@}ayO$53t!*N3SxXb6nB9bd1n;QXy#Oyrq
zZl5)8194N0R4wiWqqBCEN)4H8T)|SOfR$oj9aJf-x)0D=*7-sNvd>Hqn^hQ=k5kH6
z#m}6NIn06%B@lkD=YdCn!fuE8G$Ma0V0k(R)j;-s%;!?8A6JPhZ87++rq*_?A|^)N
z&y1M^88$Z0%Hii6$keBOCRn5zWYqJ}#tKifxGD3C1wIO!{|Z^5G)<R{$qfrFIGRhg
zult0wqWh@e-|qs2I-F-kt2%h<2OiT8+_5#t$Oj*SG2nBOnUO9ZL0Bz#3RrLc4gDCz
zSW-Vb74Y$&`&_UFab<u`*Y*oA6EW?&N+~Pbxp?a~xqOAsi2W(5mXMvk1fmbsF%HEv
zY#H?GDf5)FHe!V5^KYx5O_tcR<*L<?#uWF!ot@{P?>uBzX?W%Jdm!zioT#6>hHyA|
z*h|l}P|F_xt+se->wBD$wIE0rHR-XQpcSc;^AaVkNPL!9It6mh1|aKhdK;aOw%XA0
zqT-?Cxpc~rD-W4%1hb;_dk-LwD`FYy;!LH#wh@ZNv(Aut$L~H}6}=y(x7iG0qQO5S
zA=m%y^Wl;=4=u;#?WX&dqOdTmlD&6;wyGp~E@?uU|C_f32gu|dJ&k)d-y>8y6iYVE
z2*yVQd<^FoGqELf*uKT^t)?$pfms7`OZ2N<p~4=-!V-(&Za#^#g{va<R1DQ~55G@V
zg4V)<Blqv_Q|pk!%gL<)OjzL(O**v!qN&OU$k7@!k+_!yy0-PP(Q1akyj;-^h}=2Q
zk!$uKOtI?NF2}d`ae1JG$kj4pPaxx6zj6~u%e??7Dw31M9V|FHDP<ik$fLe{??=$s
z)5%xsc6b*cQ#g$1?#r5i^xq%Q7>(kyDm6+ov<;Up4j=@7+zgez<%B7&R~NaeH^Hz!
z(loewOzI0_MHMsJBQx@?Le{)RoJx1<4{7Y{2PasY915W9vsF2aU}?a#w`6gQJ=*^7
z*-y4R-luKAYlQofg8<-*!wYG3g8x$&{5+E}7@(xoDK+5q9s*+AagyU6=DB$&kS-{R
zAb%)ftnwilNuT7S>-tic3MZTlWO+uj?!!GZ6fvQpuKi@wX+Du22kN4_^%+2>dtB0+
ztCTUpOskIK^4Bn+P^I0o`O@S}AIS`dIHsKqWP-yfo3XFcIRG}z>g{oT;Z9k6`+Q(y
zC`tL!YlKh7YL;{hF9a|tGx@QR)9K*mivW!)_5ah^uG7Up#+5te4x~-gXT{bKQ_g`3
zSSoH`QI|lbDHNts=XGwuRDmF3+=WKUze>>B+>9Pl?PJ>7P99JsT%K{gQ168xo#r*5
zR-yVV9~rnv>9F%UkE3q20IHFF<f#-v6LmvT#%1geY$JPQEr@M#pOF^q*+s@_(D--P
z;ox?19>{b6BcTi`Kw5Qp^Flvs*tQPD##MDBz<6vY%Ci*@E6zrbz4RJI!k)@}iJwos
z2iPg_-pk-o05(!#8@Jb_ij7$K(Yiq1YapJ}^Q?5ajIfo1Qu5f7RF1WO)wSza4V$pg
zc)TdaLt6<YGp<ZiEab}_9yHtpU|FI?>iO6((I<2JBDJL5dovhf9?(S&yoI1wAr^i5
zgwFK}hQqFX5j&6<{)6Gr;HHGfZnJ#qveJGMLF1OCT1u8R(zQwYcgHFGPrY>}Nom-)
z#(2x=#hR%A)>0{2>a{!l{|pB-`SN=j0ll&(zbRm??cO(n=N9(@8PB1wZ6GtIgN1Yb
zp)zm=lqzgmsx)X~!d9wJ8vdh5C<oVZ^*g=s2LX+*TBgnGm9*@5lj7-FP@(eDJ#y0P
zVF0SAE)Q>N*s}BIfYR|%pT06Cb&o?C4{CM9qI>lm5UWDCr)iK!6bt35{z&CdL0hRg
zJfsWqj{;Vbe;ZLwe+JOx6IyYk$6O$706txM_B_JoBH&*XVd2I|F-!kJM*X_T%IAC@
zjBffii0u8m{hKXH@<snIdt-Ikm&kmeP#N7>+~!6gQkP`~{PIi0uqyNEJeG8_RgqB1
zb<@^!k&oeQl7XE28i-Y;G>^<w$mA5!Z?37udf*#AT3rIWHZH;^ya~i;Rf%;%&QrpQ
zu4kV>iy!<Bh!xFh6C=?7d<-3x>5Wa_1+j8i*y^Q(Dk=LA%qoe~4MwxROMEhXWiwn-
z9=g;=;ge)oFHiXBKT%vaJ^B+6Yc#BPT;b5HM}O+0MN(0<vn^4?<^{fD%s1|=k=lI*
zY@9Pu^_HiVHZ^4*CQqz@G>N$<I9LPjyb8#;=3JNqjRzD!8*n3+LEjR#?C=2Q%K}Q9
zn)7wWx7R>fZthAEUG4ZEfDmj$f)}}eQ-YlID|1(`C5-m)#h$-3XvuH^;b5*odj1Gx
z<z&<!>O`eOZBXkc4?wT;iQK-6L43AC)&^u^4mV$~_ZbDqMb7*gI_&0s<b;hrg|h}`
z+5gll=BkU0PQ#%>VZ~R;@4E@e#KSoTGsEt`djR_rBUH!D0LFVP%eig+U(kp^jgnIF
zkB^;GOO9+-pmY@FkBiv7^lRGL@u8=!FK<h@Bq8pup~OoRGCmklP_Cnz`8Zf@(&|b$
ztvL`f?#<Y}9YL!%2X8lQ?-RS{#bllWRsgmm+$G}Dt_6@)L!WSy2})ZOwuS00D*eiU
z$iHWQ8NVYzV+Wsl7OY<>VA0*{?XYw1?m*!XNe{nKNo!M1m-KDwmaRd;&Z=+AsJuTC
zFk^!^%~+r8;p4D~;$8+zt;HR|{*S7&4zRPh+BlFxaqU493M5zvkd238En3_up547S
zyPH+^?k3sbS{#B?tdy4E?!lb^!GpWIgy8c1o_XFom+!AT=ehHaoH=7>&YW?=<(|fP
zMkK}J>mODk=wh)YW89p)GJt!NJZ9^J6+2sjg%mf&cY~m4FVq2}<Al{ATn+Y>SpD?G
z>1zNXAQhze>mh`*GypueHUN2Z(6YojwNK^78-|S%l>2g_X6!!{(CeR)-o@29Mkx*w
znoZT%74Z7Gw#%mrcY#=DVV+Woesw@*WHOIJ={^iZ4Kp^j9G?H=N|>xLCkvtcj3#l`
z2*TzZXX5e6s7#J$#64&Xr1Ro*ii1+Nr(=O!1C*M{K{yJ-^#IKU22;2&VSNBM8h6^e
zW7!RJgt82sV#Z%HkZzB}Mw<c@O~|;^E7DbxEh6>q+8iXcO5M=-UPK`}Jk!o(bmJD)
zzlV;Av$rw9+H<X#CBkK41RNEeJN%v=8UGPoNG|eJj?Z??Kr1d#>SL!}0LWU-`Qpyq
zAlzETI$EVW{~w?(P<on>#oM0#2>)MAvG+a@&cjwdJpx>)R^6ulUUVV(>6uknz0Q6a
z<Dx(3GGONmGn-IO#cYwZ^O#j*#5jVM;Yv%aQlBXduoo;rIeP|i#`q5eSh%n$EZ#(8
zP<f~f?gpQz^DjpKR|UCdJf|Tqx-;4`(#XqR%$}dOgSa+Ly|@ZHd4FgmjD=rp*+ba;
zDo<{1X<)kozfSE1rQj?|4xXGp=DtxWz7(yGQE*(}mp|t4QiekM_<Nz%E+E}~CGMC4
z<QgKW)MWGlIS#uYY|lF5&>GaB(VmpnJeXK1FBJNR<RrNEM6IqApg&p8?R;2Ht}tPb
z-46#cV=z;Zfnf8IIR%}L_r<uQoPs<pwJ7Ws#zm&=xVTt&bS()T6Y=c#KSAJ2&y+av
zr2IizjLU)A3}G}fo_=YXGfswbrMVYD0X{kfpePcWTl(3|sUYsI=?Zt6IvzL;(EP)r
zLPaC!U(U!-2~}Mi8?_?@xl!fSGj0^^+}d;#DK<J2fWlyT#9fYiMfvX%PZ3t)(6az6
zs+<_cN#{U%o|7XxmPn}7dl-@T{1<<?&@38SSjzlxE{MyADRCMNH#<K^(xoTrdh!K1
z66;c?9BhdNIY%8M4!RJM_LFJc?h3=#7FmTF?xRtb{x@M4y27|ripvEMuRJB*QqX;b
zU9VgWvFN4m4E!Isr-xDT&@>1azL#4JSQ%|{DTvv|JtCa4iFReIm}GZ$IS>VFr3Bk4
zOlryf+zMAfxB}eFjo~6)y2dt~B*iXQIx_1Ho_-VIe~~t(*Y>LbBpgY(bYVd;2DbOc
zMOX7*AsOY#8iB7DVtLduFFoY<lQ36pWGr%xlOq9EyltY21n_hLA6^jRiZ-+$0Lx#S
zY37R7cwe-?Xzt~@c_l8p4y2gqxTA^)r`7oNpcYXE)bZR9r*(S-niCoa*0=%EbwzPX
zQo9Op;+>PxBb4I(T9ip?hcx&`qPTpFT9xP)?S_y;YYods+zjOUj*Zq^A)F;U{@CX>
zNJ#=OIp0n&PsPl7JaMnl<I(BhYDWf$&+l+pZlBM%hp<z^kS4x=5Ym$W`v?7a+V&y-
zAbWJhJ#mwWqOtLD)DZ_gnt@8oWL8w4J`UhIvj@<*=i)OnlAibE1$H5ld0g@B^AzEt
zn`7Ln3Lhefzgfw~#m^E}muQJIUm%#5B@Cc=r;klu0(5pL0H<ItdHBm2>0rr;rQ4FP
zfw-ovY_<RPI;0uZ)!f3P3Ak!+M!C7g+c}zS1=o2uN7Km0P%O)oa(pD*1ulAAG=Bn=
zg`;@PW|F{SWYTJ5?oV?hv>ipH`2<ouwhFBZnzgLyh7O6_1X%VNi7U9!=!>O3%ZVgK
zL$cS|QlN7h-;3eg)n7rogsr&V5=#ev4J0<U#w!XswqAybZD;4$S{CV<y6@%y6--4<
z1~T?Lf6q(LYyL>k#gcn~DeQ&fMq%zV$vz$fMY_j<^9-uy#QlYGtk@6Q<<eoa%JrQf
zn!3VEJ8RBQ*n30GJ>5Oh8urCIf}I~Dh<q#Ka(*E&Gra2QoN?!OAfMV&@sb+I@`RCF
ztJc*P;d8G-%Du7t;!tKhM>{MxM=t?Dp`^%}e<{MQSur81PsP%pRT;2F{BeSoTX0DR
zqW`afz>k&ws<vz$&e8t63_!}DB{o>rVa`FKyFcMVWa`^HV&MUq#^9nHo3E5<p5?>$
z)XLCqC!QwLiQEB$atzeWt+Csx&@Lzn7;f9+!9x(4bA3E!Dr`QcC$M<mx)fKf259bZ
zSA53mnHt|4Zx4ZTS88vW9Cxe@iCB#o9nY*w&}Fw}!LUrT;p~Y$hiAHO?1&L@&<JQV
zMYny?A&&%bJJYWyDn(guRE{(uJ%)I#07v3@SRTc3#wW}<V|t2BgJ@UE2}#hd9#i{}
zuAFQ%E*uL+MST`9T~s00jPSgS(i3m32WIxk@sUo~&-+V;Sal~oX6O1qu1Yh8QXTPD
z4UnVHN*u5OG2Bp`3-ii-{98zoTRLN;jCX99V<p!?%$`E3k~5TR6en#20K;3`W6g~T
zYFyT!bo5_yL@w1L2t)n`P&5<@$elNZbau&#R|@jSgt<`oywTkJy%2K}e|hCtZ8K<B
zU(QVR%-6WhYaiJM-KkrERAtC&R30b)JD>}W(Hg3>wFS656;?3YiY%lBEmZ8d6*NLA
zKmFa?5-y~=4Fy3@EVCVmQ{(D5vhw3D0PdB#E{Gw7+W6_+ax`6vY>oT&04SutMeknV
z&|z;NCr(nj)^X&1AVsCM+L|renN?5{lwzrIIYNauM50S{p~EViQfQ@fNIh`Tq8!n}
zPt2(sGS*d(mn@k)eAft8tsd$FE>n~<($X=Zm+uBQWrVpxj<s(RLTlrOq$7T9hH$-f
zUxjPSu~#dQ`OBrwShy{d<I($V-wx?6%?N(Z{!p$ly5gQ#pc~pHVg<=t;W1u_V_>$2
z_1mNzL*A2E2wXTBq$<C>3a4fK@qJ)s3{F`w4OZE(J~;r`rR4&5OFQ?CZa*+X8t65o
zZgmh46-+lxPd^yKZIDg|;&K6Q@bL^%#TIM$A)xdIG$8vb=u8TUh_?>|a<y<A!VWh|
zNBj;+$6w=*1VQekh<;V!LNr(!A{sj$1yaZl)|^r^eo=!sALUtYJe~a0X;fg`oeKL4
z-3pLY?spuJTSjC58plJzT3rNIQ@5W0ZANmXl~YJIms_0#=H_a}_1-B^PBXdJWym|Y
z1{L1*;)1EfGRsOhp5;crfI_a3^ms;eZGFhyQ0rcgJrx`lV3-%{oR$;8yAvOMd1g+e
zM~}WWoek~MN`ahOZnbm33NFb(6?OjfLLjq}b*OsTpo>9Vr3v-%;eVluE+m^T#t6y4
zwwL73xcICQX_09l9$a`}f-RK>yR`&5L)PNyiOC%WQ3)9nj=YR8lEmGTrr1D)xzFW=
zI7ei`S@aK=LzzV=U53SDqDh6NdAy@AvW4HtH?J@(H<6>SBJ7&!`VK|B;c5`G#6GU*
z9fPj}%LxyUmv4l2DTXnMuXhuq<hCx3Qqbw?%!HZY*PB6HyN)tXYLUf_w*a|qbx6)j
z*tY>tP;8Rok=q@{y=crgov_(OFJL=^kj3Hxo%u)%&^M})4ZVHw>K**+w7LF??s?ri
zYXC;>nC0IE@VmxjL31yNi|9A~@SlAjP%$3ZGaGh0_h+nL_u=Fs^F?9iOu9C<>jOC{
zRzR?zFMfPDLmj<oNg7jU<gZu=wa54;bCgEjq+}}@z0Uwyq~Ic4BSTablb+3AV*Y}e
z>?~-Pv$H|&a*)mUgt<&i&3O3~jR0vG9<47zm_Nx#B(8c1K)h#`eOzI)2K`uxCEA%U
zgE)PLZRw^;G3He;w@UTyK_!k7=ybbyPO3Ae)c|%SH1cAv5knF<GWL6epi9^#k4N9X
z4NV!ek~s5S!fp(l3BbhWctubl2I(%Ij=I))AJ7@<;h&y3Ux16Mkq7hdO2h{l(~+{C
z0*-ZrIW_sw!mgA$?fem7Q6e25?EEpbQ)f<LGWethpjhEp<&jT*_XXu%>{DoWMa-0r
z`V5N7Vq*7F5p3#>{l5Tp*4$n1X-?O!&JbA58vB^Vzk+tlOVyl)5(;pPU`K0Ci<z&X
z%_k3BmDolQnfA5CTi@0GO{&dGEdG6l*cGJ3U%MbrLo&nC@ddot;HQjY!OBi2`G|h%
z7f?&lh}dTCRZ~tpta;u%1dFLmZxA~6{QWY@m?kqcv3|@SLCqs{Td@F?xseX87&hBt
zSAnjUBwnNMzy*QGIh{n0UL=21ueY9A#g$^GMKdTZgUqYD;{}112YL!_y@`0u;$Rl=
zPQN_!wh)gd+Hia+GS!l2ZMRtp!Yr4eX51%Q<J_3o@lW|L9XrOzrJ*e^xJy89u!Ddi
zcYHV|?b#l}oL7Y#U=7RUD8*IzN;EH<(_yj3>c4dWfcYUC9`WOy6@iM?qMod0KLYxd
zP!|^8Vx957m2xZ}J10HbS}TKA%c*0G?FE=4N$=hfE3E=TEt|VzO9hJ-q6u}mu@d_W
z^cQV?EHrqRdmxC*nBGldx>;;cj?~qOlQ6tptO`;cNtg@Osa9Pr$6|igOK-cbnGsH#
zRj<p1IPv5lfifSyHjpbZfvpwl?sWjnSecD~D6;5LZT!7w9W=v>_A+M=Crs@yEqq)-
ze<cSs--)E@TT1oi_<96{OIw^)SK@D@!KzK+H*lU2gh1Q*{+eT;oS^hH$BE?6u5^F$
zt=gaUjA?6a$e(V+!ilXaySGp?TqicxJFLEOPPT!i@irpOe~uix*u=ad#NCP&*$taO
zxtN$-rZwAlf}E@`3Cfb7{}%bH#@<+F%S`b;%PN0|@)*Q{RnkqVc+<B6HskU0#>w=>
z+X0x<jk*duN3_|dE}xc4>ue8Jt!dTid^{%9U$TE57SHck`%)_-WjAgoFxQkty3R!w
z-WkL-9Xm9x+9gvRz448xqQ}uEQK;kbdbf;e{hF9@r!X_aZwP2)I$-x4k=iD|5wY%G
zU@juAbep1Ev@4R%2HdHC6pC=s4`Z&q37bFid4^5BfU3yQPswTieH|Zxk8j2imdMG`
z)XEJIh1AMmX#>$nDiwWGLE^Cl?28K;b36uqv@GMH%@fWPuATrzad=xXerhFbdBmqj
z{H?wCtCM$S6gF!ydXuTze>!Ujml7C2pA_N(;S~lSbzPaBz)~#k*&n(P5{*N}s(054
za4|eF^LJJ{@#P%n_JF!9vg=%GBBUFOONBbrY>c-jffZvDBW80PI0b-eCzap32jqX#
z(mBQ+1mJpet8NcbRne45@lS{NV@_>2$2=SYUbak$-xMsm2R=<DXG4z#f@hp-$6hB8
zb~+5eu=}MG0W2=Ey6ff;rW}`?0_eQ)!ZLg+6sh-?V)SVQt#)o<+M`}g-V5~U4>MDT
zYH{D0&@P6Z|5W0Lvw<jv3@TSXm$3Px8zPvRm*N6p{Ez3KxaK^<E+!UOT-|%^B7i~*
zY!!lZLX~(&xJx>EXl#8kl$$*{TW^mmrh&j?t@mQTOY`5!Ni3S4%RpRgy6N!P=yGVv
zAm`-mSLE0g-Rr^*`l^h``#)uVNJ!BE&_JmJta>$&<%9dR)m`<hYe7j^c1lC9BTSj_
z?l>&Qh$tARQFRm1u7cFv+z0;q^*IL5PqxLF8=+~A5o2Prn+TdsJfX^cjvWNJJUr@+
zlvQH=n}N(~E~&5=r_ui<j1p9k&6p#?qC1IhH2M}u7ogP+0XDl8#6^?`s@V88XlIhF
zRoGOYC#Yzz^k(>|+X1LAYN$A5dXA6KB)%Pe2auCwDC~>ZL_10L>1iCB_f8;}o1Ry0
zl0(wR?*=1PDIoW_m$3P)(UX%&HOOc8ffiGN&bQMsfBpS{MJsV4sy@*t$dwk4r#=Ab
z@?cff7Slx+OyHP<grBP+dIb~YD<Kr711){|LBh@fQ(f+sV&D5vPNKcD0sEr(MTjHm
zB_0)H;KM+!q|{<;gZkoUVa{8FI!T=~qxPYmt7ez@^W*tL-SU^0n^@`zFj^H^t&ej=
zIF(7fKh7Yr&yyL^&I6=Hh-)nul=1mfIR;xv&OD;)X%Onjb3D=U4B;%;jhH>7wU2lX
z)E$s9beCCBh2ar}Pph|d=yB<b{830MyZ%x%z69W!H<Z!C>LmJLVN@Ws(7~_PzCdQs
zyS)bBv}r_aS5VKTn9ID*AFc*^fx37|wDa<n^S1@Kv9Z^R_1}Obn<l->zM%-0qu4I-
zES@mO*JG9|z6I&@xjfh!i@uXTLCvfkL@C~`VT}K+JVd}dQ}5<@seyDn(k;v~P#kc?
zY3~EUV^*JW=Ldw%&91H(_8}BK9k192e@w7i#}T7q=%>&wBDV~p>odn3IwI~;u&696
z_tq~UNnLhgzkW$LCxS`KSJ1AzY=IcPD$(@~nA$)m825Y&K>@h&xc+wp-TGSY*Q38c
z<jO*wB3Eph;!Dwd%^^JF#P<*uL1uQIVT*A;{4P3$yX{A4a|0_Sbawb;{V7M8#9Q*Q
z)-TX*^};m0n1^@JAI^FLtG*aG*J{<CkDVq{nGQe)33LWnCd$?-)|m&$-|-MYDGsaA
zJZdyH{@9QI3xVK~Kq`XW`MhARNn<CIRVC&TQluhItvp};3Q_h|>dpU{vFaVmxJ4Ko
zWxaL4`~-`#ac^J4!BjyWy%;gtYy|PZLV(5Cs>?fU2LYV7yo$!*HJTcXijPFQsn8H5
zeG9`%+ftx3O`4>7m(EF`^+1ch3^Y>0dp&XUviUz=;+Yg@F9&Ub;#eWR5a}}3%hO9T
zmb<Ed{)PnsmtY1!n_<}NsA|^<AqONQwqB92#A;~#V-*L>jdA%vg60P<3%GFJ9^VOa
z))TmoK5KPIbzEh!{~8cQSLux>)`D`Yp%h1XPnOFLb#xR%!();N@^4^NTda;S`A_MK
zUq%otn9G55U&KiN#|7${3YyRGbxbTi2HG{J1E5;bp!=;rm%6(w1q_?qO@Lf^U6w+K
zcit4ls~4W&Nq)4`^Flugrw%PsRNmp6<?rz3DHHtNw#Z1j*U%PQZV6Iwzj|dL{w2^E
zvMz68%AI#BkV3ZP!~m&TW@{jr2RDYrrXrjSc0{NX<Vp2?f&NG~SUNmecN?HW%sCp1
zZA2GCD;2}#`E~&2doo8>J&q9OR+9sjV?=r#Ep?Pm1EiY2QLuAkHcA8eTeT16UiI3Y
zbK3a3@*MZ?-N1?hsW&h2kDQ8xl~FA26Xd$VMph_tn7{cRfMyXVY7AiU=3XE!R!c`S
z&h3~^m}@WD8`w3(j-DFsun&NHQ}J9L+nLyMUr<XQuN)rI2$dyebQ~$VDz|;HcL}oE
zXgzdn9X1Uu10saHxH(lp|DPtcbRlWfc)+3{9Zh^Ny$OIUeGV|H847tuz~W1Vi-yyg
zD!qNr7NFGHb;IK}5o9%VR4g)qpi`~1%iDOj09R!As5r6>$`!$vcTda|&6lHwMngA2
zXM)XMa<~153Q#dr>Jdnt^{v+fmPV4cn5wWdYwD%_V%v!TZUv4fBr9z~d=udR4evCB
z{xMS~X~szd<dTyBi}Gp3vXy8vnsF$8>w~1Yjg2wo0D>;A#vkQ%o*<`)S1}odqbB&_
zkQ~p(Y~}dhp+K4Q>=ds!0-)G@P!vz*2<W@QU4I0-Ejjdi{zx!au|ZFY^uz~8fw;{0
zGjEQqkAZfX*)qqiBK?8iu8oIk2;+xNMoY&N$0hfbDpluwfrUuRbs%>BC*+iy(99n%
z5^mrjaKOm~i)NRf!I)>N|EqMxISM+R_7*H~@I7@Z2yMvw8*#P5F0I~YZaB?<p+$=E
z3OmsbjNEy${j3adG{6|W_`fnxY1jC1%enr%2W`Uk7Z7#}*|}>aMqUV1@Iv<#Dsib0
z>M(pPXHq$SBi&VpFWX)QWY)HIvOQh*@=Rj~Scw+V&LSPm(wzrf0pvn!UKkR$2yh1&
zv0jY1GJnC|8G9%!%&r1*ChQUG;tkOhi3QrLR}*x*B!zcca}2u%&{Dw@wcZpid2LQZ
z%bfT^v<DTI@3Kr#i;lVulunzT-<f`WP6nSJ&6rPi$JaOHWE!|c5+iSfHowv{!Z^pe
z352GTuKCeh^1oc3pU6SvMz>}}*FIp!jzY{1=?d_<eVY?t6p10X6L$5uOpi`e{%=Rz
z;Rx7a#o#-01Rd@9eC8^Fg}k#8iXnGrnrU=cJRsVIwj0Cv{2mZDK?To|_d!twUS*G8
ziZC~kW90|(cPKVw9=gs$AnuoFto&?&w%}b4gS!M=#bVL@2(*WK-n-OOaOtkmZjS<!
zkw&IR6|N?QTEo_Oyk$58+>OaCBQz!~h*o_J%*nHpVSDyCw7UqNS)%I+NOO&)D%X_d
ztKma|<b`Eg%u(2BChep43J=Z9AJI{C`-g?yU!DX+9_ppQI`Jt8vkgVDtSujT8pQc(
z4cGb%q?^)n-sR8c|7FQ=+H;VkhAef*db1#^jli`Q4Dp2c^9s^a=IqV!=_Ylt`U?=|
zj~>P7Ojq1U^zXu}Ba>c<kQ^WT8lV}3lBpcGi*|*RmtP6q^TOywTpudEPSDLiu|5tL
z<yx_8i0|Kobg`0`k|@0eAn{<O^`64ch^{$02EPqW6?LUIhKq0&*$~KG(pm3-I4#<o
z`*q(7sLpBff7lrpewgF5qHgJkOFqiU;QYH4ahf57638BS-j4}Wjj?0nAO)RYdnc;t
zNpaSv8NoVE_BhLZmXS2m^G1*m^M=b9cz|T!KKpYpGh36Q>@3qc&|AXESgpOw7lg^U
zp&@Qouo&g^4l3Mzt@xKq!tW{Q7LvNhR^A)U&R{yf!p~AEzMTVT_UKfsK0f#!fc!dX
zmZN{jU(lv)C0*{1w*{L2*k_mHkH12@B+2}aMECg(sAx`Dvf$jSS4)%a?xItmImd2A
zkM=$+#N%Tzut=`@%>(M>QFoPEQ1oshsOW6Ixy<71+kSxNX*vqjw@>{6%tb|&-xdqa
zpJ`SWt#Qu+nbvDQvDbpo7RuI1vCKkHMJqLPpI9jlUKkAS=vAwxMF`Ux!|USo#R+<_
zDO`W29oKDw)pqoU#}p<u`}o6`G|c_#_@~0vx}e=m?;_n6fX><X2I1)uOJ@v)Ai8@v
z*0v0o`7Cd;mn;Wa6~=D4zj#fETNmR)^nTwg4`3;iCAz#+ast;EtE>QSPU##qrirGs
zTywc<06}L`nple2BB_&>F`KMJ*vYfxW`)szWdIjEc_`JIf1y=!q#iV~tucEbK<<Dz
zW$24}2W3cy4C3@Eg4`CvhsU*|ii#x9!V9emP;~VAcA3=n7ebNIjI^$v|HSo$w$Dqg
z3F7+U3>l;U8jWFHQ=BQ<mF(p3e$%yc^hU|};i4^?xME?8khXyr2z71K%`_GJi*>+U
zIj&@DzKtc<1!90o)mnEbgfl`}8>iPnx;?NYX=>(TJ_Rnt^}?MreH)usr1#3<`BRZ!
zi8Kd%(=HYq1p>=3eWyd5Evn#9588H|{EP)r)3J(LY(2sf-qG<d1<hLq3OKpdUjST2
zER5n3k^a94tIVFbcYOdelgWK#jMxy`O~qONl^a2sSH%qw@n!2x@;4N|J#N_yx~L%g
zX9kw;cwh@K%EVsvtSt$;2*a`PSaa+AwJdOAPtmR$!>2rG-zJ2*wP1WSbekNTIj}WG
zYzOVc<nLq94vwmK&c0IE#XwEXii~cw#;#!1#@2%=$rAN8q3X+MOTJLp3`lE|`*w#m
zhcLln4Lo2E0CgYiLiQq9-E!tdMl7}uuz7})+YR=GvecyKCt=1-f=Gh#^QQd>W`2)|
zo^d%glU%yawOT!p8In#P+Tx5x5Z6LtD1|X?Ij9Wk`l1`-7BdUx+iMu(uBwX@%keo2
zW?($1Vu&orSy+7#Khz?@hGZ2utBFW(Ni)gm%??kYHe*@|n?pQ%#85lA&3|(K9$T~%
zcKvkx5ly0t)|KHc?>csZ<WW*>!iae3{`oUp-F3wM-Ov&N)-hizT+ABse#=^F@g5*a
z*YFsrVCs>KS))W0xp4rDfb1-!D*qkPERuBF*%>|9>n~wSN7PM(D9Wv;(0&kI3|S*$
z<RmDk!Wm4gcA&pZuazDx!e374C8d^wpo<~Vo|GdF1}bC<gVIW@b_hVxV=(++z{loa
zm?L30!qb&E9|}Mv`^piA5p-#wrigM(d7f&G3k77M8XgzbBFNaf600BXq=v2<2OZ^L
zeM?-Zplek(Y)HIX{4-)~)E&b=1<P4;uw0M733477JvFimhIS#6C0!g;i%>U~?cScp
z<p{dMhbL<RP88YH^cYPi0;wY+0&gl@@U2q8oRt&BlQM)26@xvta3=$qbDbR4vn#^U
z(9|4lk}hkq=No$}kW0(6F+6j38gwBTJax?Ob-OcjLh^i;R$VKf^*gK??@XMJTqVr)
z#L@uqAM!5%%P+bxEYh_uT=ZN(7e3jXF`;iQh~^tRI(AafpD^2CkS>?2f4KnA+-G+e
zyNE2hB<og&zDfTE@qCgWQ`7MKt#FS9I`P22h^#cPya*H#VR46fPrv^_IY*3zInCfL
zp}63GfUXMr9FG70ehGl{w1<ZHrw|&1Z9v?uu%nO1{Q~aRYSGc6G~yfNu*---9Whni
z`3iz=7}gI~*o+tE6u8SxOHH~Gz-+<@iggdxfY;<WoVvibu4@5kI^XSNZWLW@fKF}#
zcH9U+8TE$hFADolG|o)WOWh1WiH47f#}%{;U=i04OW#s_hh@XY3j05H0bB*V6}mcb
zq~15~HW0>@^tk=~x97k8ic9P<JtLTG_aPUT3MrOjy%;D|D)#0(_zv*uc!%mR`PdmM
z(B&q3j`AkmnbTv`)`K?3-33BTxTv?=-2|O9V?WPx>ir*h*L83Hh#S?ifmr)Ke>7@r
zyr!UeElUBN(4TxiklU!Etb>0T*70DDhJ|R-?A}p>CUwgfeJS>Qh*+)!izX6Nzq?qV
zM<rgIKHy;}%X1^z&hsA0B#)`Zuo;kMaC1vZhnELEmXYLS67L{0K?-rfabhW+6hKQi
zu=HHyNy2V1H1HiU`Z;LV1s4Duv@@5_o&{D6(~K2|zX<8+TAPsM_KZ}I6I`uJQU=m}
zo_z@rZey0*_%cCq<tl1SdDUT_AC0#a_IJ|u;L(I)RbB^lz1$gjt}`Q?YwJG6FGBM4
z&7hH-Gj07As7oiy!4_ShV!3gr;GzyarTF4)NXo^oa*_84QZ6l;o_e3Kle8U2TEzY)
z)I3QQ)(KkA2VfQhZq25lV%?8{NVB;u?)@zP%T%5WSzZz3M3VnA4W}Q~FuX{#w6WEW
zO+P1|*-<D@>AVLBbZ#mE3Z*!!7SFFwX$-sRcVMAo<AIeg`Pfa2Hh^bQzRrmyH@56;
zW(soQbTuPA%==9(8V?1gW#7rOiRM<(2@V-vH3tOd$sF_G?+D|}Y=h|cJ)vS=?Tqt9
z6-?D5Kwd09CA455NBq5LwFdnJ1Ovy6j;5dUH`sqOi?E#Oj@yO$8yZ!bx7h3#psFz9
z%vr8j;9~z@fhkwdB>XNYY>wj1oK>{8Vdwn@=t5OJXi|mpTx(QQvy}-=2m9yD4NPh>
z03I|CVG2pttXHrY?Jy;X2SgW5H<=A>GKX074}j*ypt|^EUMNyhb>}GTewlh|{3w92
zk;Nb0YUYD*V%bwg93#*TkRBQ3IADRC2<rvpVyT4yEa-K3VjQ;!q+3?XPKKsZd@alw
zH*lM{63Z->fpSwHPJ0g$KuIvg84(LEk^l8o*yVo*aCC*wU;)k(U0`e=5*D>gWFJq^
zwcuh2hRwP~5laEO785WyVsW?EpFmt9ox=@@dzJ<;*GQ#dStz)stM22L%TalEjG^LY
z0ZvWt%poSHEe~Qb(IXa;)A?7(kz|O%HRnf#c<@gSdwqOfiG^1LcTsc_!`=EC(12C}
zH@*VQQl1~>fCCBX`7;>hW!cT*=KGbPT!3U7&3N(5%0LApbPS(tz|USK$Ezf#oy~E}
zKp=R>q0{z*9OUTwc?B(L6Lk9jrfAohi+H{8v~4vISj56LUQ^h4p?;r)Y8Mu7ygHyu
zjeSU3Va0F4%w-NdHe3UWMrT@UjRQou3OHirc5x-{734y20@55Wi>9c(_0cgT$CkqZ
zO@KEFA~ktvjyWQ#a!a@1^7K0Xc+7~Hrl9}CnhDeHctS`apggnM5nHbd;Ks`8F*WPW
z!d*&@)av=42(m0PAH>+9kWO5@V7z}oh!bbvMEa)<18}`Ktm=uIMVnJxz?B6EJG)vO
z{0z0m=n=&6@S)>vjZQSelB04;<Xwt$MY|!`V9Jdh;Qs`|wy|Skm(ltE>aD3tGzoNx
z)5#l70mt}@Qg=*JP;!>$y1`?iimG($Tzaek^CS5RrZbe+19I)SmDLod`~@08;a!Z_
zW&QjgT8g@O<!{hVPY*gziY+z)C`QTjLL=HLmXbo9WCQ*|V}VVfNm7T?3vEu=CE#ct
zp~hX;7C`2wG~`K9^PVv0)SHy{OKq7WF>Pz{#w;pK>9VVQSYaQCrn_`$9!U*e+M0ig
z0gqcrZ6e_iO~PpSswfw=t&?7`#kSC9ZiPog=*{m5p*eUIWSQ;qpR5PvYH@iXPP?0T
z@7JkCz@J2gwa^Yb<WHEV<R{?y9W%niXl1y(-aj&ewJ<x2xI~DV)X<WieTcz(0ZA%G
z#>EP{jJ#s0&hqg-Ilh~<6i1B%GE>tog8f;X*8m2;aJ?3vDr^R$vSD)irV+pfLu@pq
zsqN(o0rI#=i9hJ5x~>zyl%0gM89O&QjGb;gFoCeMOP6}p`JNSK-s<LUJkkp7T4Bb=
z`hY`=HYcHj{ihUmXJC`6H4NN;vjfm0P(ydTFRDmE*P+<I9o3n?;*}Yc6dKEp7g$sd
zwMsco>;fo6EID;T62B2<HX?~SalT=H5NB4J5Vwdbir<2c=cD3Z_6hLv%I^F%N)2vj
z#by-{=b^*y<g9xKVYC_>_cIiPjp+{WxwW{B_>WAo^r<k*QWsk7fxVEfYm2VS(h`RY
zDx?js?Jw72=rJ96yLexaD=zEBO8hh#fF{?Ykr($7uC@^drc!2oF$K&_SF5x1p{7F)
z26WaqR!>U%$7@jGK|VSUA(r{ao>5(8*~3B1G2X|n#2KQg{g|<F(Gdiz{aH)fj|F&L
zmaG!l8S@*b9tBK!d$<Vn#?gd}zS+y9cHeRN6Knukrs<kwDFz=8T=Z|wrT;0i=)pL7
zVX<<a5Q>2%)Zh~cI?1}RC`B7i%-;^<lzxMg^0y^=CgquVG6=0!H<pvC9Z!KUU+{F)
zgQx%KKNX0yM-7kh3YxuaoNP+<htt5EAEzVbI7Bqj%boFxf-bgP$<BK^q#L3zvS6+n
zcm|-k){SbL6IuZ|mUQT_y|)$%wKW{M_Do_G-Gm*ejEUAf3&>nanvhb=do~C~!F2CF
z1s&hUB%JFCq8-|q5Pv%-e~ax4*6wk~xgd0u^uS)+d>)i*g8nGBKOYj7u{InT%Uu9L
zI(h~9fqxS=Gf;c9#?=>Rx}nU-_~U<}U0dI3962pV!0e?GGcPx-rL%0vUbzCAoYGu(
z+*J^+gzs?iR^c^3mNCZOG*iop`}J#qs}h3IeI-u34#YXr1q9p(K+5!o=D6x+2-i<b
zxy^3P<S^vz3XyJhoj7s=-4gfT4(6K5Y$;Zp4qX%^+p(2n2f>~_^$2}o2XM(98Rc=#
zHg3*I%HOyX)YU8Xb+E(ii_PxIkS_Ua$ZZ$osxdXlCT0D5b0j^1qDaf$mm}dVLweo$
z9|W;jRi6;RqV%EsC2Okm5Xwc5<}aI>s7Y(=3=r3JLNAW4_Ymy@a0;zm-vvU<y$Oi)
zJWoKn9*8~K#pc*akh$N=)}}8y1R$E+|G8vlj;>4B+_{WL1UbtF>_6j+C-dhx_hztP
z`l<X)Qgbufd?v_Q=vaP8v_75Vu=3JF^zpA}GL&urQ1+dk1#-JdS0J+%DOHXTTpj6k
zcu&;n!Vq-c2Z|4hFt)&P;q%~Th}X&StPnF~7(1Egen;XSZmwDU>yqkOrAq82z}1z5
zusB_`Mo1R&OTR$a0+bXFcJ<Kj#he7I2?nV@)&OoXk;Ysv5yOqZ{TwU;cmzg}hJz9D
z>&pbyt(YmJ{Z-?`N5nY_g15;KYIZ>|V3gycH~7a@(jb25n~<)<#7>>FUGp{oOkma2
z#`!aHv(7spWY^UkYrdCb;{<}D9Pt4_ZJy%ZRcGw@p;N%`4DX{Ug_ni64wONU6USGA
zNOV$5jQlvK;oGt_(e|H!xmXq4FZRTMPXWx@<P!BQ(R|NKn(u!`u;^~gG};y4@Hr5S
zmy`8JiZF9%oc0AcY~eJaCwBM}!kmz&_G`a}be75dBOaO!Ko(rBoBtbvMPJp9s}ySr
zsaB~&Z(_utIY2HV&I)mV^<Ab@GaU6jv?YMEQu|LT$K^kPQdHgS-uq|5P86RhEd9FU
z+h2iPswR1Gi}`+Ye3suabgnh4HQ+VVSblCuXTT+1L`6e$Y&#FAo2WRVg}uVOdP1>x
zX}_Gz1Vr!(k;OWL8`muRsqZR(05k`1*e2oLPe|2jofCS72<vMu)E_P~Zq6}T`=kcY
zmaInSnU@%@C$n#xT&!$8e}>ZbiuL~Eg5dw?VR8F{`Cl2R=yJuNMS)x(ahD=C2=NHO
zu3QIuNqh$`R{OR%KD=0v8I=|^tVKD59=#MO@}M^;7#SX2I)CTa>DVDG=NR&+gdNRC
z%R{=sxwDGp%L@7LIxJ=z20&7NX2a1d=6{B9zjJSq{4;WRoT8waB|poZ@y1F3Rlm`^
z6U1S0gCC`SZDsy(cAV=r$2+3^DYux=Vb!eyP)u*d92Hv&BP(1ZEjNf@A+j8D>SST?
zYWa&wSr@;NhP8wtkaY;;)HU+|v|L}3r{xBNx!cH;{xFd)Rg?V9KPlSn!T>Kr?PJ!;
zF&gDf;kO~sMU`>2fyKZ&If5?c$B=cQ3xnC@09IxjX@B^!{&C`Z7M^d<ugl+Zh*gP;
zMLTi&Pg7@XKRkcJ5>pSgOdSDYK4{+40CmEs3?ZPr`%@oZjs~(cY0N5|7R2piYss-Q
zVvX^$5IBt4#wC9t=$^yYP#0wIg168H`3ss;KJQi*;%FF@^G1>;(u)MTRGO}({%GG2
z$gJh1z1G<3Z<%h8)yf|>`Q0b7IWNchg8WH3I{H?$877CNH2I;M<&WACbeXKYEr=H9
zOf@#yf-toiJ1Tk=EV7q}`01jZ7GsM(T<-5V29NiZ;<Bxv&9k(zmk}cMICNWJOUy9*
zcipueq^nPh=@DB#U3w=#cS$Lo)O(S|4TbtkUJi~CJ432)!(-sC1YH=`%yOu|*lzxF
z=*U=6!D>gqGKlk-xL=q*U~<<kdf@IL&JqU^e$2BH$L#53#te%k_9j@!63peiX&-><
zxP`tQ7HqmNkSoKrQf79}J>mv|{*c#DQ8w&74!}AuyD}`C5)ZG>Uo+}4i%1y93#2(<
zakMB*jWIjfrGcO;Ak_;Oa_5vlJc>h2Ekn6*tS`C7(h-Bl1GzmalUjPPOh`TW5Wy}p
zx_IAUi&4#hE^|ZT=l2r;%*6Ecl8W|^Rxm^Y+vN1<`sJei5t7Ub=>2v8H`wr@L*j%E
zNSC>Z{)+7H-UZ^k8Mc#rRN`%c)Dm}YPxcUWigLn+{qH0IxWRqp*tiI5RC!!sSDuHj
zm!AUVd|8Lb8wWya`J-X+ClQ4-a3vAL5Il$+3??2k#a^qh|EK-1RHh#-b|{#c$ucrM
z0rl4-fLs?wN}fcBTaR=iGVnO{pM(n;M^(S%asG=fMr^FG|KemQP88{+Thg<_xFbFu
z$d%?kC)%<}Cjz)gJST2l<LkoQm~8)b=7?P53+!ZIr>^_xH;HsoGTbV~4+5xZBQK|a
zb_!uv0dZqn9+#d9BF=Zlk*7JL?n4bei?F*sh6<x%&^h_T37yRym}0zeu0QPSh>r6J
zx<VK`GbWber1ODX3N+BHTJ9F$QgSL7uUwG-pU~3Ib9!Sg25^R~%{pU~|3XtPyk73C
zpm|~APE1w~6jqEDT~x)k37h^eC&C4o$uZ9*(5|kuO?2RSF3mA`&<@R;Xvct8db56;
z%RyXn`6gkIy`lyrwZ|e?0$764CM9vI;3pOkF7tY9%(;fJ`Jt6QDuVdwS|C@yj*Y|4
z*Fid4c?6+bbO|XK#(Im3{{k%Dth@W)nE$rl=9O*%FjG*4>wLa1?iE%H2e>A=6^at^
zJVNYyTmHPUk1H(kvH;gXTQ&M2jkd;gK$i;3k`~khlZCic+%<$5OWy@Rze2?_Caw|T
zHlxvb-lIm>jjiMVyZK*zj1}Jt_Yij3FgsZHUMOd;9>+d|?5`A9496NqlU_5v5o*q%
z7l?K4&*`w^eDnb*mmnPk&VCfSnl47g>0;x9GXOpI7DoV9xNr0rIK^wh(dRn~BW}D%
zwA$kYUB%HcQ-l*_R_bKvU~KqBpxam0cxYbZgqc8v)Nq>xm#k0b4~tCMT750hvBuZS
zvDK8P0L+!1mN@tsC|9)${r}&eg`|A-lyAdXgq<Y=cG^=-6Xe9ue=#5qd=a3Kc6?{F
zmEsB^j*@1b821u@WP&-W^m6{wTl;twfWqLI4m<KEUV}1g$K$p-o_ie{4zZyc_y$2|
zB;6i6pS^{+cybu0Q=<NFWoW#{Pel6uxAS*xJex!JTlO6gc&&?naqD~eZyrn3G&)NN
z#g#2m=Ldvco>n!&MA2qlS4;fxAyjp-#uk~uj(tlU{1LF5ucAxLNz8v1XgSltLoa(o
zHnJ;zlB2SDYwMFrHU@kK=z3Y_S&4yy%u`-@X8n7U02tGS4a%1UDHDz<d7a}PQAIuo
zPx5F~iBVsJx*m+IW8!7e)RFhf;#Y+W#`ZMp=K1QgbL=L#7`uN1?QYM4OKK#pV$R7}
zeTn6Uv~4Zlf?1S0F}uC7MviCn{6eHl<`+Sk&nj{Gcc3(e91F)%-$S_5XicNbBFQ3a
zm~d)~vzEDjAn2N)5o?Q!M4Q>H`+GUOWK6of7OAa_Sls-RzaKd|W+-S*GJ?}u7ySYN
zE3~ql@oP>DSLQ4?)tnpu21p4hU`zaIp21aSReF=P%nTte0P9L!-lijN-4D<llM!RO
zBsX+EFz4LJmYP+I0CKLN`FwqW+Mj9L{@sGL6!5~D=C|t>u3@sFROudA9LzlD+#r4y
z>Bbx{BUxNtE&*a`((813C}LMZmST?H`!5Ama5Oz3v<x)GlELqq%Mx~(yVJq@Zvvbe
zyX<tgYOCcPA5C+dy#iqxidQBd=x?|lJB8I&B<xaANG<7B{c}b*2+<+ZTq}XNA{y3*
zh$JP>6t7UwUvvFNU*EJch}o+>X}Zz8$|_(qlQvgzm<ZQN1{*POQ2qc-C9DHmPnfHs
z=N^^r3#$Q{F$3%3<kg{yjv+mEN6f4tysfBB#`{8?QIkA<y}l+i%)=XWj2@i-Yh(o+
z>x(us^#Wlf-dzj8RUXE&^{{mzsq&QGXdOz>S*ImeTih(fA2TvC_&;0+;K7H3F3i`F
z2V80`KLXfk_MnJ7Z)E<gjr)#$aiIVgn5**`-H(DMQ#=u$s32u;>8p<?MiX{2dQ1A;
z^&rhF2DGFjUi&XV@H-vG)`=*@qRb*4UhW~tyd1V(oF&TX>3ZQUB3(Zjb)X=8s}_Np
z#AWLffnv2b$Cev8$a~<oY)r5yMzTMsM8n^JoG@FNI9g;;knTQh^P*u>AgaaH_ZYt=
zVM*1P7`hcf_gi+P9g{e`EXOo~#6^!BqqcUqwR2(&-JY;>sbeoWYX?Z`j<4wEn7AWE
zktGTa^|?Vi0Xc1%<TA`~sIW&3#LHA$DP2LhL7)dk&1!`=^f^M!JFX5avNM#W9sQi0
z+26Jc7zLFAu6I|$l!naf;|>w#tkmt)xpg-Xmwg!ep>y_tbZYpu(iL@f(83(08CA(a
zdqKKTEMKcxKD&30BoD7E?~^0Rr59aFy8Jo&f|?l>+5RC2kL?F!iA#NRk#W$p0n&MT
zeU6P13d_uLd??7}$BQVh97cB|h?&8DVZ{<u(TT86YmQw76auZMMEZ}<5!ijAuXtDh
z?9sakF}oR}5Ip8<I>3f4`Fpiq99N@J`Zve!6KenCXFG<sLc0Jxovl2se!l=0fVp*a
zyxa~=)39udVHLv8E_p((#1%D!y{m*}sUG4KENk|m;Kmd4XL>FUCQcVpL}F8cmL=|-
z1Oy+)4vU#ZuoLfnlR-T!uq{G`#05{<<%lW36p}4m8n<2-T};LZK}DB}OEKes{IL$G
z<@-iQz;_=4T=X%HNVhl)k_yU0-hm3cT#RXQa}@&)2XdWZs2-K+65>cr_1GXYQ5+)3
zOk?KfG_e*19RvHDN}O;cQPfw5#ViGtBlpr?RM_%c-9*IC!dyT!!m;AfkWL(@BHX(=
zHh)qzA&bev9H*p*f_@e4GO~0|OT*?9fZWp3Voi>Jo|pliu)|N<nF5>$uEludR<zeI
z+&fBd=cGN-%_oC9aRxWIG%1Fk0^}l(K;_w6BkM*EjblXm1Fj3FZuE;Fcz`K(Y%<m1
z3L^H=sW~CuCW>87gDi#~Ou;d&m6GaoKqtqVFl_>sJR^UG4O2WX5@wg1V}-K_S~Tq0
zGo5H%cs3}#h&|ZQ_(4QbA5M34zvUM}6i)`%)6Q|2#b?~Cu<O~unfAXffOKbWs*nEv
z&VS<0Hukv)(#dk!3Jp<TGzp?yXjlHJpqm39VKmvJ{{<+jTDaNfPWdH(l%pQA`u(Qm
zFF5yOq}=SX8o-USQml1(2D)3AS7T=ZwH2tHLo?4Geg(Lzfhj|@UkOPyIArCcKU@Xn
za<_2H5}Tc95>!Y=J;#ezi*~)amplFH{BNa`MeNttK*QV~S>`Qs9Yof+aN5k9&=hsZ
z^}y~{Qm28OTtnz*=+xptA<l*C^zVp*cX*|I;Rb>(A-kMp5%h+jBE2$oD#fceg1F4s
zOZCO-H$xYzEG=tg&QveVanSql%86(vEc<)C2Xx3SK(qkHHnI7wge^-r43X@WI82@@
z)CucIOouUXzA(PRTs%fhCv0Z7qlP&74oIwm2gO=<7Lm{rd{Tt#@t3geUEr=u(o1y2
zmqJ`F3=h#<_QgN$0kXJMy)af{dJXJsz`qcNIxNNo`r~rPgyekUg?oW47hN@%g_T(0
zK5*yEk#IfL|KULpR}NM)fWo!49|BY5m|S0fgfMNS-35LWW|)$us4qSeMT=pQa{prl
zU0T_S@Z=tyX_LnRi`-ZTkn(&_<Y?&%nbe&zQlN`qLoGJT?+bIoaTw4OSI>l|7%0lv
zv!C%4R8{Qt-Z<tLZA^Z%g*zpEl~$0?KLh4AMNDMftn-LNpUp90H?s+Q(SJM-<~&(q
z^~A7Q&=$|;-dOkrD7P~AN8_XyA<cA@-f21h@Jl&H7uQMGeHqdNz<3U5n&ZkE(3P(0
zPIx7MfKS@E?NvzFIeJu#el367%!q|jT>!b4JMdlH6n}mL(iP>_h@U!|_Ga<<@EG+D
z!KyeRO^o%O@BR+bQl%2lz6a#`vPbWWYd?f`eJlIN>!Oknw8y<45q8~WoVwn}ko5nw
zk@@0N2p5GN1!i<~?7O~l1hhdNJX{0e!qD8ESRutfW&@Ek`e7Xp#ZjUiWfJBklAVjb
z0kS}8{J;|d*GOXypjnFN4~DYqzXfpS4LtJLtV{Dx2y<dwapCH9Pb~REhLReGPIHSO
zSE&IdOTQl>&5H)<c6Sg>TQK!~t01*aPg)%F69hc6$J!r6S-^Q0j*Aa5>t~Rn(a}Zp
z#oINYaf;q(xae166kNxeD|Vb~txQYF*(%z_$GkLc&wB+q8$Dh?q|byn8!pyR{%>nQ
zQpLwoa}$FKjAdgu4?!~tgBAKFEzr{sOhY)&y~nYC$Pw_-i**3P`1pKaE@rtMo6cFH
zb5+x|(Ld&|FgW3YD5{uVVTFn*JzmTa3^VNE^8;J5+8Ux~0VpS(cI3~CM$+VKV$6d1
zKO8gQi(WMUv8s$`79w0spd3Fie4|jfcriei3p-_L>sMSnCxHf`Csq|rRapv*jL{-I
z8c6NTE7KCTmL))|^35Z)OF}v$od7EuPpt_w7pt#i#P&;NOnwOZl-9<70&{kw;x`d)
z%r3?NvFNh^a?@dCNzWX#G=OV>F^djAeil+lFl$oX>YryBAexhT@{ncozg^m@A!NHP
z59G4R%Pl+oO8l|Ezvyj<%@uSGr7{`_Znq5paUqi9fOuYjv#ctBeq1p_n1iO<TPOV)
z$jz%m$?o)4NF~k}OqDT^8x{Mm3{iC4PAqPdiPOMU01N4n1B~u?X&^|o{~^#dk{K-i
zGKhcOOzDg?txaQdp)O>pUj89?vQm%}<{FReFyC7>N0NgVuF1!Q)pBad$*lJj9YW0m
zty;=)VGU?+;lsE_fb%OnxW+fD13ABRD9G978X#0&zIkH#H6dIpEVtqzk<J2lZ>)sp
z84Tct(&<52mW;dn4rA?=-tX8~7>VjpRU8fERAjX|goj=MT!{3(PhWg2#QZ{G)GO^b
zBk!F<L7iK2DX!;5`qzQEdHe>1_Ab8&^n40~(vA1D5g0lQ+zc3x24CX+>u?}fiLEw0
zbb}E9nzcs9Ulnw841TeEV#PWNsNk1u<9Q5Qh*>1pD7%k_ba~RzfegmuLV+$ghSsc9
z&_3-n1`L)oqkoGbf5{Q5PTynE^}#%>Fl^$+dV@@J(lRzat<e(kXx@<jDF{0}Ifnar
zV@MZ-Z3sQJrP3Zd{uQvOt(Ft9rvS#IP8z#UVb?^;#6SEE(pA;^6dB)Kh)Yy2d+`CA
zLc3ZTspyD239@<qibX$u>tdP^b5Xl5$@GA~1CgtiVC!v7xL}yChazbE3Up4G0gu7z
zKc}IV*Hk%EkfU%6#G8xTLDO!${MfmDj*UiGSEl0Goq;@J<6d~#U7%bwR8$&Ne-`4p
zbBf3lPLM7XD|_s5wBUbEpug&F?(1o8k5hLCsS1c*59^6D1W`5ZPd`)GAC$Ce<z<<@
zK`3NXQ!KYHLFd>?uUUFO$ZGHHWkSDJfIpUUjHHho2U65w1osE>2N+Yqfa~h>2kmTR
zdb?sogFi^-PVt2(GQ{3bwioj?Lb{6SkyiE#yjn0G&=u0zpE}6ff{@cTY3q+`f~fX)
zZi;CQd~pkyi#5JI)}N45V9AP3sxLMdWScRj)n1vjVuMT4v-Jnu7&{`yw&u_Lzz+7$
zZ5b-sAm$h3>^VJ<_eTuU+riAoMs~PUJ0S}Nj1GxuO-&F#2__boo22xaxPSf#v5j{{
zR>h46Bo^w1c4IVRkJ%j8R{$t!OGDK45X`b86SOHp9Gi1jGW%{KfJd+7UQGToD)H!~
zjFveFW2lti{bzDc4u|Tx|JSb%#FZohs~0%6j<8(7(Q)IHOk>~C84rs#6WGtifP*rr
zXH#z%=^|@<MGtqt!64>RVLF`d$(}FRO*yfGi3C)^?OyIyC%dOZ!HNNdf-!O(4xj;q
z)0syUb}2b=D#t-b_&+h}vLgvQG3}K(nCF`IQDEj1^AV1hlQR1lp)T!shB8*boBb2S
zrKLq>c9~WLj|nDg42EOqF%HW``|}Dr+2lHe8JCgn!DB&PINb@2=R{j7(W$fV6$i9!
z*z-94a37Ed0kMCP<AJIS?3L&0`Az^ar!cC1SyYjBC$3I<PlVQVC)>@-MVMt814{9u
z0QWrE*!Y;M-$^+M)R-e<Z_#9#ycea}?_D8OS8CcnpW-lv7cos?7Y-{D?NDY5aSimk
zt7<THDi9^&IHnvMoC@LMCWm6ziFRcslw+1Cw-o9+&52a^^!!ELSl&sWBAWhyXZdLg
zI<v-3#GkRO=?oy(SPoFjvBa4GE+XAc<=IP!v*bc^x2R&+spmm3oSgNX1LQK+jX~!%
z-?`9Kpj)=hJ6#B2aqq4lAHDy9bn;!=f0bgT|7FCJUmPf;x}snorLnCN8(s=p<TAEy
zL>w%-5O0*h<yhhJ90AWcn6!v?1hz-)Tt2-5AT>2DTe%9N;6*(h9BF!85Ur+}@WHDc
zX1I=(t|9E4@pxE{o34j;9d!3fIzb_P$xbm^Zy@Z1>roFhQId}YxtVY;g-)8x7QG3~
ziE{O-GaeP~&$#d@Rc3rF$Q3}#%38a8O9n8A#ZmP<w*inZ79w%;?Ky!$KNC++&+*%m
z+o_RvI=)mNzbfq1S?W=u5qAT)T9v-`2HY&4CB$vVnSS)$2k9d6s$nwGTjl=zja<Mm
z?c5}!I(STAq@anPcmT}3h>LLaTrITrcrbs;Xd*MyJs!%Ravp(s2*NjICYXy*-`yB<
zKM7eBK`xeQsj?t=AZIJ{JeB`uNh}W!u}KZW{S%XKC5{(Vj8S?=i>;5W^<JsP#GP7`
z>}wu;nwV~+j>h`>nEVX1D?$gt!cI%w)t&`(jd3CG<?mY0gPFNWFBYf1m?LTP($qwK
zUwswK3}>wG<@MHU-^{TJ+s{f=-T`x-a$RicIpy8_>BN?9Jydwodmt_<+uqK2?0smu
zeP3@heL&EK;aCaQpC!N<Gt29Q<Z>bAYj1}yiGEy*#N7+|t@-}<2sR(hQTd3^G*qSS
zum~DLJHR(SA?!kzuyWJIm`beqIbcn7R=5=Gj_tnihn(od%&!Pj4c4gIzQxO;J^0`N
zezUI~vyGAQN`+ksooj0{sUhR_mU?&05mFsViknVYTYZzioY1R>U8Apt=rjVk8$E|O
zW+&$6yf!M@a>(v4juz=QuU=+}Pldq@8GyyeZ}a!;0EWh<qFq`h7R|C-*J8BFc=!U*
zt|>Ph@Ba?U`8A>T{-s8iSgo$`y;Ijnq4MwZ1GJ0N+1gf*-i}wYe*)8(#2Vu7KNGgZ
zCr^%i)rD2gNxuRkBV&fgPYSxen1+mx<$r^A1M*aY%r!7hXcg#eD!u7#FK$rJJ*0|h
zZQ81(^C`kCP3(&&%CGl`d2%$rltJs9_XjZNq0J_)ieUT-^X7<ncR`e4)uzIV#?VV(
zhs+P{W-YEYNG1G(P%6v-8$T)R5;!}qLo5IUNAZ6oJz$qe$G`wkxBp^~g@I_mJ~sHV
z?4l5Ej#6W+xdfDJ&O(PR7OVazgb^K!s_yu;79qX)r>flXCnC7b8TRy$@=gKbc3{xy
zD6#aCj%u>tqPJjk;cF2`3UsAeWen#@XlSxymB1{$3<PyW4HD}s?Cy;ZMdppSmh+!F
zwi&!UVOKL<uA*<w7UW{FyklaCCHiLsY1UHrCLzwe4Mj@aFWTbRfFL|q`@2auV`mM>
z2~1)&Nnc<1&mb<8o?C2)vH(ilK>M9m`;#*XP6Bpcg+I++Y-Oy$Uv^*(!|?+{a$9UT
z2+XzdZ9Cp26I93~!$P`9caSjGkdtHP?AU%)5SZ85DlP455QsfX;=0xI*K!`gP=!W9
zn2TOwq{eV!4FHQ9CS6QRRIwE22z4~tl9^`FH33{WJQ*cnK2VTHydrrY0X7fLQ5jv*
zJ&8kp2lbTMhTt8(2I=HSnI5nfA6A88*qB&)2z0Kg4t<Xhg4}u~z1Z4NE+o4v8g`-p
zi*<Kz?66M$oXJ}|#M6YR6S1P$d0oP8^a)s^+$hp%@gN*sXk=XmIIyB!2M+^q=}>c2
z;)xN^5?P&3o;ec2MV0Xel_`$`anmIok^0O`Va^060HoP78o*pl>Sj`!EvTrp9{Q=o
z(PKbdEAFd}8e5|}<DT_$Y!sTh9(?xtAgUuOq$wNZe_4%k(<3esf^?v}-{`Ldi-khE
z4rZhJ<2C~>a_fx)wt^}OSZ23Oz0VR<y?IqQZA#mq&-F(!s<H@T(xR)ZyLARJ=PC}?
z9u(+0%Agdh&Ly@1F>^5&X^m|~Q_fL!(O&xt!Qt7TZTYL{PWDn)i50iYpO^G(+QXs?
zMr+LyF9>jJ59d<nJ2etN%h9$y|GGR)J{fpp2N3daZ;T%lgrQm_Y`!C5v%@dA*2io?
zh2U`U8b#nP04|>v(8<lfAYsnFkt46~MViMv>RH+q%83?kBI9F0MZd2OnLq9hX7=Jy
zs<UmPBx<F9WKfT*VkpVBr$wNP#M_CCKs^GS9Bb{eG%E4yo?t~)<eiqiWW!!SZUlLH
zi*su91bq3$kD{H7yi+%-OKr9{SWz=fF6M}I5*llxd7n%(&X0(5MOW*}F)*vM8}|jO
ziVfaZ)1-KZKxdV1>ByLJx&1Px7Y}tJ7Pks>dX)~D$d%%)`W&sqsE~}*{?q_wj!a5s
z^y>+5IU6Wwtk4KeXJ)+tyF~a;870OYC1^<i*A0#;6D}l+9bhG{9}i##RPgPvU~~Sz
zzM%or$)yE2F*q<RR+^A$Y5(Jc*8Dx?kIzMtm$X4Ew-I(;y<PGnyLLN(yR1xlT8bM^
zD|G-<_+*9NEW+$eCWA?B)6|*4?42%d-yhJ)YlTvYCA)L1Y?&MX5Lj@S2c%`ecDW$R
zsAYboC;!cCm!87CTM*JMTO!;Q^g<Qo!Jz<}6*&B*Kqrc4m*jWgr%6DCAhE)4o(xGI
z=-lJ2DTJvXet)}T!2=-7gX;0UI>P&ciyp&`^E6XVKM2f8qHc@-9-5=baYDBq5PkZH
zjNvz@y-Qk)`;W|@F(}7%M?ty*+5>uy`$V97FrGALi^}7fj+G8P8pN%x^Z1x9n)rV2
z?PpQuymYlZFnTP2`H>!YMvON<F2~_MDU56r;5h9vC?HOkAm_ql-X6;z56!66)g4<a
z2>;}R_s0_mn-91}il0w}EHa~PdNiy5NkC2rKiT`ofK#AdK$aq!IR%*ONR{3rO?tWk
zr{<)1O&WK+eX-hU8N%v3Z8yfA4&=5$XOPaeVq0P6YN=kn<26}y3v@2kX9Q%Lccx&M
zMpr+EMBf?t<Mb@L)-18^nSibq25aenXFWmw5E~3m7Dk<&0VzAM$oN+V@aV)gYpej5
zl~szafi%WL=YqL;v2fQrAdd)vyFDB#ey4D?gY({C^Lfy&M5zU}S~4Q2#F>JNggg4|
zB>t=OoeUe-So1={mMD&$j}S#m)=!C36r}Z#a$RkBOQh>0ab$V$pA0l4oxuzNZf^Ny
z-0XjlE`3)or<$Fy@+BZHF^ju~I9)VZV*o~%xKfleKtYa_EjA56-A|k14@6XJJ8VSE
zb(R0bysRhU8VJ}TN7koa>py6*5pjhGFETNx$MhRxKi&hq&e2(K#j-az+{CBfDQwx`
zecH4`Yl;PL1f&$n0DF!ISCLEqSR+BYimdZGI8<&FVqw6rg*`4WpWg~Z;X29fV}+fA
z9VuxD`O|d3B2m_w!$dk!y&ggjyzUOC(2YxyrxmW&D0%3UwxoP#M%%GTz~)j;4Vr~I
zB~~z+7C#W8valE1^De>`3@-nnc!%Ye+zsaPRd{vMy7!arseuSQ<{U`<arXkcwGc)&
z)0^I(W1$zEXpPPN0?7|uV7#Sp(J7icTe@Syg8*<=uQA2$4>=}Rp=0xh{YR(#D{Uo$
zB$z+qkVgo+c&v6i5MO4=+XND!2g|MAM+v*eXjf!-+J6R!bM<llxdL2mc4^I>rOwIm
zjxhHDF6H-s9Et>yQ5<~`u8JLZY)_a8?c{jep5^*00-T&*RwU0~g%zc+l4kBFYe*^i
z@Lf%a6Jqbjz1#-{(DN9|o>$O9)Z5CrE5k$or-3XZn5D#Yk<Qh(d1&qZo&|HRIJuD_
zLfvy9E_@5`s>G<5po<pJJ3Dc`01FeUA#~}<N9KaBfI4w{0hd%re7G=wYIi;R3!w4n
zX2av`+81emq1mkcH6pkY(qKr?KNnPG=CEP0_UoBuUDp;TiiQPT9p2(Cf-VE9hGa$>
zr@Rg3GH`#8Lx5>DK<nMEm|X*4D%$IT@9+r?i;)o8=~qO#NInLu#C7ih!B<Y{wT9dN
zgB%0Hie&eA%|}4yQCetNu-_Hxs&Hyvj?YB9ecJK;u;9lgG2YN?KFyy%!mMtomA(^!
zc=2v{Z1fpnkJ-t;PT77~?-xvbdMzeVUqHCj#c@RJ{3TG8k-SVX>ucyj;BZjM#J=?G
z{1GdM<n`d+-vC+2Iki}64ph;gTs1=1QUmmSW+fi`HYY0mI#b1Z--EbO;5S-$y64}1
z$g%W@4qA<2KY~;XH*9olCfeo4{HZ}|pyhrBqM^}X9IPPR&<rr<7sA!vo9@lrDa35m
z0R7CbkS;W5>&!uQzX7;$>}ocx4z8bT?dn%0#^k@w4VeoxGQJk=Z14)MYk$Yj146p#
z5KiOibODrVB4f%u3cCzaeScH@HHqPLpML!1>^av;n=T4<$h@Ggu)NAi@-LkasLJhj
zF6qAd#|%``-L;_$0Ju<e@G`67^@Pw~m}wvduPy}TM7gNJQ6zf#g)_uzi-k;I{9RDB
zszXP|WYNWp#tRrMx)8u676Ei&xRb*Ibx#5Q3fB#-aqObdMLVNZPb<at#lc)YI+F|?
zrU{{Z9L_TzOcw>mv}oI83Bt}57c*1X-;?V^f!5k^Pan;TOvMnOQ)-b}f=(>UI4}k+
zMP%oOp`f&t#|Uv5F`{UWpKCM*5~Jej;D5IgiO0mHA?ZQM_~x@^@_$^%i^Z4CNhSAz
zYl=2Ma0J6?-Jg~N`2Wj63B|t4=XjGiXQT`@07v#*51)GlVwiC}^o0L)nxsKk(S13j
zQC(&Uf>$_ikEQ$n{%?HD^u_*yTvbkInq#RIp$o-_UT<ticfr2261V)BKV0Vq^iDsl
z3~9F8bV{C}mRTLp+3Qk)gz^3wq{o2;?KHTS5|%G_i!?`intSP`U#yk?uWx9K`PPAS
zJsDgYnFn!AD6A@a!|LL>S{$@yv?~=~f8Cs}?u9hQdjg6vi6?W0Lb;;yPMA)9liR)l
zb^PI)G4QYkY2w?>1QspM;M63tdJe*oGWI&jqV$Mi;7CI1Wd{z2aEDFqO*&A^R^p|R
zpe}l;6V>D?qo5g{rc8;AMiW#gWIctU{d!OqOIE;oNNsZ==2Szn${F<+0Q0+^M`rax
z*m@i0IFn!mM%SC<WVj)u%ZDwSg17{z1v+A~Xs5!?C7rr_wgphZbH-8bsMN>yTY{M{
z%+rX3EEMkC3J~eidDV{F5Uw`L(6MoZXiK)vf;t&>wGDf0TYsYS!v5P6c8!v!wb)L8
zD~)|UMzc{C;#4Pe(hm+6U2qj`RmHZn7wrh@yozJPG&n!B6R?|Y0&BfpcZQ^SSV!Ps
zeOCxqn{)SM`cStUkgJ$RXHM$cV!_=3T}Cb)Gq{WrP>fETnz5z3bFcgzL$mf=@N43}
zfV3v2o4TpBACya?rQYaqkWN&$jJaYgz{1Ij<StRgpeVBxX@>3><|;DJcc9E_k;92M
z>w#Sw6iw;%mB#!rDtBg1+WbEyFr=&(6Voz5n9!uxo7NZ);d0bt;9ic;1<;^c()DX1
zY!0!rOpcqI1vxia7r~y#gz$kB0-r1Fwqk6@b9FPcE5*g8w?q|uqf?RlYAqnnU?StR
zjxF{R#3y)K#HeykEkX~MWaNV3z)m8#jief3%44*=QlOKjDQT1VM^{cdZ9|!Pnrjed
z;=DY-gBklfH@r=8h2}F+FbUVHvlSE@&@5cvO}Hx8!$!rcqMex@SYm$Iu>#`ek`WY}
zueNxv2h5zMmoYZ?_GW-r&oEwmUVv-Qs|oU58%s?Ds?PD8O|$-E70@m;w|p9HG>#6K
z1Y9lG&=Ju(8QLPoIz3(%2`5Jmk>9g7MY+%np)#tutS=*oY&{nvj-3MH0?DO_M)WrY
zxfX1**kBGh0Dx4JsZ|^+%70_dPoj4n2;efYAsrvP9Rv*@wdv@phH=Djioz}qD_?E-
zcR3uyveMNU4~z2ucqQuQ$Y=K>K%5m*k~VQi0;q|o>zPO8KXF{cGVH$s%nLeqd@Iu9
zwSA>(bfA&mKH6zu`4s2<lW-xt($1f9Y^LExM_eJ=ZPV4+rOWpk&HjEIs54^s6#o{9
z2$fr6_2cvRbz{a1jpJ%G4S+}W6Zl`O;h18;iG-`~*V-3=<eC!heG>mzw3x$WAcS%D
z$zVmlgcCaQSm=}t>C$;Sda`2$nPGa<xDu;Q1)yy0-LaX1t}8A~OR@i{&_yk1i(_hJ
z6GP0sr}3{!$}P5!6t*ndvdmkQ4NeDkuf&0GQh@So#hHL^L{@Q9a`rnbBf45mKOJ8~
zxH=(S^I<|5pBm&i{B02~nAA`|{tMFCun5J{Osbm~&jYOX18m6aE`W9s({(6Yu*G$P
z&5u&Ki(b3OMF3_aNhKxTb{7M=#CY=RVz{|U2z941;)VYbF1RLBmiqrer|E>|9W)K1
zNS;LlSCKCTa9WMJy3!GUy&S|6v_C^<%lKsS-|q@g%O=I>;3lzm#=XL6uKJ0w$(00M
zQXFe>K3|R#1UWlqGp=bLcU1;3TxZlc{2G8_4yTKy6NjPK=7d-Za0_sJv|X3K<6MFo
z?|D4{wPJb9yJR=yM06d7ajQGpZv>-Y*q<J&pi9wS)>FA>-IU{3nq?#dBfb>oHmv9H
z;GJ9ZH+WDXj}@i^m@Qg@Hn2*;-SJSNW*^-}@xHtxBT0!@iH+|DDdvK9nZ!#)+9J>&
zcAz_CZhi0`5Vsp!|8jgHIyXGax1QMM0U#>PC4|xO!h;YVitxxV?ja~=B0m|(J7e!z
z4?Dgb7{wPN%xU{crqahflE3024r>{Xq~cD2mdwJJEkk^pVdsG5yXJM~((@iCZXt>-
zl@6Bb{}F;Ha_utaDC{iLjeDIjE&T+TtEHWliusZts?}VE@w?1~a69Ngea9yuNe*ui
zar;va>jYtqrwO}FwW#Fkser23!{8auL6fLf$8pm05SCuNa<^2lV@osg9<y>P6Uv(Y
zT3-M_YB&`?Q9%;W4l(+_m?KIL-5URU$^UZYk2kYLST-72lCi^6*Lqf{Ymc2<S9<Vq
z+ABF7ZTQ;bwHhF2(`d?GB}Ty?9i_AQe*86{g2~v(5b4l2K+GgAlf{Z}L*}t%WUM9H
zAIP|uDf=KHPF2T_I_S9fU7%u=DskZ|J!drcdw?25C$_{I3OiMphRN**&`y=ZQM?z&
z;~!=O4_NXiHRYp>;9x_$^m#u4aW?EPSdPhT{-{p@%^G=TL-5Mc?=vuG$1qfh3q%(m
zqDSD0AoaNSbHIYrzQHhs<aYlOl%`@GggbQ+o~009Otq_T9K=`p#}&X%rh^AFulpK=
z>S_hO+BbwPQO&9Sn&$wxklH7vb=_&g_@I%4?1jE1>;k6MTo<O2v6VnG8nw<8iT5so
z_(r2rlfqRgr9KOO5A8e<VqLFmi|d3r4%U=i@!XI89#M-2ej@DT;U>lEiTi%d5X-vc
zvAp58oC5wx>m%k~r<x+By2IlQ(Jr|!L#UXC&jV<gLh%;!^n>*9(%cohigG#Sb1rSa
z_7!G^VHC|?f2sf%yo0L8qoSP+yNME$2?O^5^8&gB*@d*Q8Ial5O7rEkc!iJtHBNvl
z%A<7TSBhH%!N&&F({C#5^2nzGIwh|8*3!U96K*E{{6``o7WkOrU1kwZkP(<IBx17w
zkmaG-UTiK}Fh{EA#XN0j=2{4d5uvdyzF#>1n~Yp2(=Lku<yjo1tbNctA=oufZsq0p
zC*D{T(AgE^^NhtnTnqYSa_$ftFP^c9m?I@iq4SmiGp8Fm8ydI^RRfp}nLOqQFzfNM
zRUZp41?|Sf-$I-!(v`2EWYs!r(4T-@MD&VqNbl^OSOeL;P1J4&Tc99jG0}{SH<yNj
zsicE*x@93;FpYV5h*$x@b)49lW{As#I0`qC7$P?84?r_AFvVU9yIft`B{Se3DX5T5
zF7h$jl22(sj+gFXG9n+fQijqFi(TRFD+4(z-tb|;`jY^ce9-V%XCM?AYpK2Ms)Sum
z)~b@OSA|e7{w>8bt3ed(RImPI^-MEc*T=<cK&Pa-Vu3XYJJ<9EShNc8$LQheV(qn|
zt;5mE0o9-`yD;KhsXsg`rStRRUSoVJJPA5()z=~HhGa^Quh)e%cYF>a?mSip=*nPb
zp}Vu&49j1YdOPC;kySa<=?HVydBT(lSDSi=6L#G(tQO(oGdB0cog@4w3%I)YRD|=#
z8C64Py2WzHNI-LdshKvsv<A>o*i;=qiWshi&fH!UNq%~}GnO6=;ryh<ixWk=uq}E+
zL$)Q?3bUMGG}2aL3;x$JPJ=PC61%Pk;XwtDc|8>$gRQVZPN}_<17OXZ(>DYwItJ74
z2O^z0LjsO5MLTor^WoI;8|6=!!(*aI7^kJuYYMwuXgSiFt7l^%bI9JJHEo?DkP6q0
z<ynoPf33y9!I1BIJSeabNX9Pe)cm)MXlF!0CT^M$3WsVdF5V2poJapfl(#kqpm3a2
z#%Bt<F6jX<iOQB+<RoNA&vZF(OAs?eFLqua(&b2dP)uL80x2pqxsg4Idg*Thi+LDZ
z4L$z}`tH^_D%?!7M4v%tpr;IX--b9Ya!02w?KKH-C&gIe)gANS@^a2`)J_0qv0ki}
z{<w8lAeXU=<rnQRbvF<<V}q_RzPv}K>4L-JW6@>>Zg^Qupe$|P6U;+*^-ha)38xBn
z;u?_{w{O@7#3iN=;I(syeF0n~+Y?t}h9G~14wMcayN?5Lmc{tm!*#s;At!k0cx{#-
z8ccV1XEx-&txLryq6Fl0Tcr-_ju}FVs_MY~fHJg8$G!y#KTLp2SK<g!c6ze}ITh|+
zvyh7A#)FWp-h&$5>@d5%*j-_#F6U%%zi2leKI7wYk**X6?a4qaeiY{Xv_I7;`~O;i
zT&86B9=o*WIB2oyIi+~5EywAQ`;%yIuK^rX@)o!N3m%=E4QCH+|Aasn5Sjzr&P=oQ
z!{NVZmynzBI=sLmoG_;Fa#MVzpyjrL(Hwd?$=Hrvpma~^oML|wPN8~HIM&`D%(BAR
z$->ju948Bgi%qD#uj?jUZK>pjet`-IsiIo#i)kXv5(W|WNjvxC_z011R{UWih%+ws
z&!yN^Sk(!9ak0R<j72Ba62!qlH+KmEC=AbW#PGiSr@RBRc@w~&l7=Ow5QZJ>EcQIG
zfVm0&aS?7}X*>ELaEd-&-b0r?7{WDUbtw_sQHU$Ui9$n5oO%d=woQ%koq`qzPJlRZ
zXp73BIr^kBZxhDA!!m#cGesIv1K5J{$zewlqiAMMyW$nmPPLhHdMs6A<zs=|l=A3S
zoYc<}Y9?xfl{PhN9tY;)>S|6~^gABF%~?)fm6<eInvNFiOc^zX$JQr67dhBeMuXh7
zLfs`W5s6h!ge(NEo5w%&8r@i7&QP;TIkpnuZ=^MN{p3tCfJONfNDDeTAJm?*w2G;K
z)f%_8DcRcvxqLlx6*BEqXo(-Xu~Dbz{}6`W*ju!Vg>>1_>sDcINL>o(L^E;*2<_I^
z6(=a@{L)DiJA%G5fr|XFAB#gp`+Mv&G%v4qHV7r)?XNgiVPuB$>{2Xo4g`aGSDLs6
zpKA&W1jdvh=b1vC-N>UKqWDbf$JkwjbHtYb3XB?!WnxEscs~ETR=irt@eWh=xC@;?
zNqeV$|Auf8aUG;PBIAV=ZNba;-Eq-HISM18+Wo5kfWU?Qd*dzz-4wnAtDVu#7Xw;c
z(2>hOUpltgTQDV0hbw1^aLF;w!foz<p`9()g<G0<C@DuM;VdqexdhrG(ZEfcA4R%m
zW$p(g^V#{Qfw?j);?>7@ycEQZlx{oF7q1cIj94Zh2~#cua7nlkG#p3PITibZih3cm
z8q;9>bOmAO+tkZyPK$(bSAtpYi!(HFVb!ZZivn|wR=OIp=tKEouS!h32H3^X-QCvM
z{8|8Wlup{_SCdCw59pf7>n=m^0ylwZfa1>ZAcb8^?Qj`IVhuq>7ip0Zzcd892y{)+
zEa8&(W@x7>FZY}ipCqKZ!e+im{ea=+JmE!aN$E1HMpNp}SmYM|FItcFI*yJ6R6BiF
zj1%cK3x|C@qMTZz_O9pN3T^(?T!_Suw}HA=Ec(#4^~EDK$kugu8NZ!a$YSsK_^*QM
zQ0n`0Pls?GEWD|^X!ylqBmQ&;VdvM-k`A=xVdcQPK%E~al-xlm#lP;(P<>ltr#3>%
z+>;@O%e0u@{$3!91dc&DRY2W6Lzr2Mk5V0qJ|@HkYSxQR+uoNGLg$p0Kr;nZqhOYH
z^8GoH&IXy+#jipvL`WK4o)s5L;h_%!R~g9Bw)JUmKMd+}X%9oM>WSqZ0fR>rydzG2
z6vC}m>Xf_J*2e%~66PZD$Hxg5?BIHa)FO`y^2Q9)=fbI`7MOp0f<H)=(=INCh#(P-
zmrp5-7_h3q;3N9age*GkWciv#QU(ZfrDX;ZKh$U(-nGW|Pv-yY#*7}ywNhx8OCy#>
zv^|~zvalon#d`@WKb_+x4GCEddJc$UYGD}fh^Tfvo#lQffO1QV+Wox491X|66}Bjg
z*H^y)DJheU$nh^h6mrkB(HTp=0)U`%X`Zu)heQ<vgI;9xj(z-V8SUT&!Qu_$`QGuD
zw*Ov3lxvBh2&$`cd>{x(?`n?~-z8{1N=JzO1XF=9m5Vi6{8A818H3`8_xz9CXvNy^
zLzvwxCEDY8(JFA4c(CpVQ1AdQ#s&&k$FCOdrsfY3rLNfbLui+<9WS*q9UA!&kh#Hl
z!Vq}c#{gouyqqro352^u@=VQBP(qx9oV>{dWV#?HLGhRu=K2(%7&Z(420Ea-{xe{6
zKz6Bkr>ewqpMz0&tm)$a$U5sN%Zuaf<1PtF;ch_|SQmo(;;spBXXcLVD063acXqMG
z3GM-c1OfziC%8lK0Kq-M{fCG5^YqiVFXz1H+&{XW+uxRIudb@TKw+mdDXGv-`m_L;
zFEaUk^mzf8L+SkMd;tJ|Y4U@<B<K#MZqAN`wdA?43MhY{l27C8<7+@y0k7uLbBgx_
zxzT&Dg351kFY|w(8VPof`fes^AVzU){|*0o-bIeYl*;v~_D&xPcUo;7{arS3jEUcY
zx(#J+9`An-P06v9{zgG~Ak)}Qe<;zgz-W*AMCV?^N}o<1tNaM&Mq&L%03Fx<1mYTT
zACpw6M*a+x8>P3x;`5i8*Gjoag8ibxPBqyqtKDax)&G+~G|f)fg++s6AkGm@3A<Uu
zx1ED<-Y6ham(sdE-*`^otgR$&aNi@N<1YW<f}s3ELC#Q2^9Okx3Wls}E%8DFmE5se
zM(nlNY3`D$_Kes7$z%y8-=seln`a^IuQ<@?`ca1PvR`YOhrc`qCQ+9IL^^+@G!*EJ
zc>&BUni0FIq4=*LwEzl5Uo2Suu2tL8Cg{Ci0Wl<SGl3%k5yTzfVt3Jn2<JY>ZHwP{
zJ^we3jnI;rM}*rd`FFvNPe|rH7j^Ki7FYkaU?_)Sb&TcDhCCdheDtvMpu#S)UMB8A
z)O)HRug+6D;_FPwB{J6$i<Ia(QA#6r^x_3-<;q*~nx&QmGGoy9g$v6Ha4tz#g@)gz
z5eJoYZnh2*paItsGcQHZm1jcZojY}dJq5bH7>MbvaMaQuuJ1rQuJd(G*T0NY>6sRb
zE=$m*K(b6nxI=(bkP;0%+<rL_s_Lhsz8B^GXXi_F2U_*$6@c>`De3086``FjcBZUj
zcM5O>)>>ZXUJ05;)te--#>x;>Yp^5MQ7~64SrqGzcg|J(|N6g+eHF~3k1K!pcH^oA
zVNR{!`DesxAm$`%N(-;O$K=(4a`u|jcN+#x^J*`(<QjzYl9S|ONl1LonxL*|nqSh0
z8j8NPfSvy&33+;A$Kg&&&k_DjVV9o5OJv1}m}_MDBt79e5G|t%!jn6iLI#E8l>>{D
z$H$a($eecoH0ahJ3uvAqfs~5kH~<&XHiAAhd?46`C|izLbUdK@f$k=7{D*PX?+b>T
zY|b2XxrKz88J;?m_Di2oXWiA+JJb?S3xI#gh-MwaW$w{|iOrM>ywtk<;XyKbTr9mF
zluN)>j&Evi7UpyjrBI~@)(3FYaWCJcQvP*=5{In{!+zrp0bD<fciHIbx^GaR8+fqF
z?Zn?3G(sCJ1~w|GaC5_^<f8^q>(uUY^Nop7&sxg1Xu@n#AUR^w2zybGE6aH)9UV|l
zVy--Svl6$pqmAP=f+9h#9a24BYPxD7fLVawD=TBJy%iXo#4$k}yKVWv)6A<4pwiWc
zqvxjWiQ(?28J=sc({=!<_tP{FBvq-icPdG-J0srB0@UManDurh?5d(^nw+d1FUTFR
zzcMKn-UZS;o+3A7$*yqft_953^Njo%yOp$1SCHU5P7>s$xkt(qI0i_sI3BVmcpker
ztALy5?p2T)H&M(sd;Fmwm70v;W0!qEm>iJf|E!=B(1H)w$2Wo0YoCq;bi7*3+YD^Z
z`;@j<3xI2kh)bRU;@fJ8LpNhD_V?BjhgmzRq+BP+873DkcnEC+p)`G6aY+Y3v#f@h
z?>{F&!r(z%6TjO@IQQ#{94x%m1>l5`Q0tMI#kw86iHoMa6|UDJUAZPD+I4z?-KQt%
z)ssp4`4jAhVv~NtPEeMYjNGq<<b^$*meb;SNUfy7S-Ce(5Y4=s90ZOYfN(=46Hq-S
zqrN;tI91U(^*4irT@{Qu={&azFe`C;g@Cpeznu)^eyt&q7MWv(xqt|$<WK8SA^uQ1
z0k!|PQ-EAU-QC9AQ%e-i3YgZ{;{2f!g~hccHkbx&u5+Ik--@gszhx$+n=ewjYC5=!
zog5Ct`}+e>CuB$Q#DRpJ?KIv%i~A3PcKI<|=SdH?)g3bm#{EgPTlWZu915n*x%$|>
zhe0?8B;fR)TMh^Cm_d-Dm(*zZS&snBJqQ^zTZf$*2<w%u(q0qdl+X)r;R)EI0P5vQ
zCm{SI9Rub*!mSA`_8c36ojy9Q9o5!o6XIEymp@Ku{ErM+j=TPF9RI@zsW`oJ0^z(=
zq|>FQ?G;Z1q>EPv;|c|dJDDYPhm#1Ct}G)4;tNqO4nA>oqDs^7;->)T-oo8VCC(J!
z=ITV0`BCG4WU#pT9ypc%U4C{?tf+CI5VB{|-u^TP(Sy8R!MsT8@P^hR`=K3A2c<!}
zrB*rVOb80g3uE!3!nv>MzLE{sD}u~awBV)3(GriJ+rZfFBpdg=&jZUP)}RyVlw^mF
z4M-dzh+La>tXll9B?)%I$rD1XazVkG5hW!{jOGi$Tv05mbdKy5lGiar)Lh##B38Hv
zF!!onIn?~*;_{vL;56C-mz3}7w+Fm;Au#tLDpHMMe+PgYY|dkeOUs{Zi`jk6a~VK=
z>gtUrE{8O0@@qlNPgnZ;X~_EbRoLSt$wN1~s>IjSnN;x46XrNg+VFiP+66);OS>L=
zb%}s$wehjxwIxEeqFejv0$hrur+`{u+#}4Lc?w?5L}~Jpl;eTxAY6wY#`NqrK)Rif
z=ChaWkG&d@b%$`*aAWzX9hbhli6p^x=?OkB!Y#x&K^rKRy1As8u1D!H#|m;kq<h<B
zxE5F50_vh?kp<)Ij)F1swrdiM!-Sbh)izEZmy326u!kUZf%gPD3+uF`t^H#E@Tck~
z^2N8=ozUj*04gs1@zz}cE<UzSN@$L|L0l@kcV%!l38Py`@%Cedodo-n(ZhHY9>B%p
zo)QV*{f?}wqZbs;Qw5jDh)kGe`yK$Si?{5bK>I~@B@gn43nSZZGG6r|r&Z%PH0hC&
z7BV@#g~<4vCeWq9?GUoF%A)`@4*EmYI9h}gOlF0(Sl}@bYN9K<`Tt4S1;=Y|d$fpl
zY#LIg-~SVm2OEBGn6Vj2Yd={)O=1|oV2HZvQ-I7roa-hjNEV&!yf%5-VWb=Jy236h
zXE`Krb3X&%{P7Vs5Wf-a{1HOp4KX^N1)(xYCOIG?&rY09Bz$dqu8~5uX&{al?T$}h
ze@avy8a?cSbwf7S3nh^(&xz$<1ah{#xWY|Q{7DEkpjXDV|0Z1DWzc>y!=Vi~eYwDu
zX6^;h0#$o`AlPM;;fDItkFNl^+Q}z$eEKSYnJWG<%~GyYUIQe{W@IZ1zE0TXmaZP6
z9lV6U>Hl@MoawESK-$VDHRkQ!2CPeV@{U#u?qA*k_qb?6jLB9-7s1a6HS2qLgera%
zZ5~K3iizj^cR}jhw!@=^|5HAK1-zXgx?U04gzD&g#RnxN7|cCmto|WLorC508#fE~
zOoZ)G+#}NY;i89>7!Ds7h&v>$NAa~Fmy^e5#h61sDM&I6?~1XXf|%jDrD%`41mwz9
zv9fE8H9iB$vmo8<IFa>b8GlTyD$2j>i}GCy7MyHvw)?Um$$xq+?)X2a#E8^$ajSm|
z=~gJRAIXmn5KhO`O)(lI-<4#fCr-Uq_yNT2fn%l4m@b;`(8Mj|n+m%=D7BN=w*u;X
z&mLCe%^v~X^~a2k??pLEkxaz}GYzk2!jZo}?iXz?=yvJVS<BzD#)x%igU)<RTm9<<
zn2+o^WD9n^5Jvsvq4-2W@<(1m`DdNOf3)KranACUtee;q9XuBZqq;*1hKGxAqNttp
zB8Zr6ZV-(=j`RZx&o$4?E5z{%c~D3{CT%8Hng`6alr)Dai#FOqpjnB*C_7HkuB8mZ
z=A0ig_cN4nDk!`CW`UASday*g?XkQ-=ZC2QiuG)pKmA3-bkQ*c4>n(rh<SiXk&aCj
zS5Hp}b<wo(<oQ2lhXhiTn*82Z;|;|zgK<p)*M|S90CWmw=nWPE5L461X4c=7zegmi
zoD~H)6SSmp5z&P|+f{z+Z;&#M9TjH<6uj2(iM<v!(JKAlSi}UXPvhgAMIp=;_Ge7!
zY-kq)(vFfB<Q2@lNfOPZFS_~S1@vK8h9p}r0hn89Dmza3ss8g)U^#6rptyO9s|2}C
zT2L*2LZtgZP1hYQ@$k~+8+oSCK;L;8ASaEL|6t<C`pbd2)8nJ#t>vNUIC#H~F)I*u
zlG6WY+1gf!xy@k3@I3)JSK_eTovJ7lD@$S?Sh0M;{(%PjKLM@@remo|H&_Ws^ETb%
zxRo82q0?f&Bka;L=G(B9ABgo<0n0s+=u<?xqqg<Kg=JUu=hez!OkACC-Y{wZEX_c6
z7D980SwWe{pA+IVx-m(M%Z5SI+(~GCjR@DcP4@#aum*tVx%zW7@`QMfU{^~na#);W
zp*4Xt1`#eVrf@EH?_m61qzi#QgeLgwh4_1V1`^kn*n2pTv+nJ0Vq&ZcFhAI6Vl6}A
zZ}<h4yp%Dw{bxi;YXC93{NOJ>5=doXE?iw<mjzFMu}S_50ec)iioeQapkqAc_~)3C
zoP0BAei$_#D08LS!$4?S7l8C7?|D$+`mn`hAYJ=CEzo60F4E4E08?VbdL^Q*B-KU#
zwLV~;{fl2&7_{gHB{H1>4+iSu8wzx)h_dN_lr1&{%gtG-b;eo^x~YRL;QFE|Dnk4C
zPGQ%hvr4gI$&CSMdzlNxG9vu17IZXc1@NhqKt5MEH@G!)V*gG2aq843i7=OYFrgmW
z6w;+aV$KU$qDiU^A<V#L4)^wS#G{)#%=?<L<`#sV1zjwqIykXFoZpi!%>#m5N>=#x
zj<{n>Cou+_!$)(tf(CofR^T2FYS?}jHD;Dmg}Sv+%*x6KF9~#-%p36KsjUHA6lUjI
z8=7@%Zd2mac)kf0Z~?AjtJLeh7VZ34Ls+wuSZKTL%4ZyOcocp9?E&0|d51C-O9-PG
zH0{Kw9U#ofDfIZIcZ7s<a%#4t!Y&q9e3*|+js5fAIz>&`iN9S9uI)KWye+^*?CjuW
zj>UI|&Uy2K8yeZwxKtQ<GpCP>@w-5Hpkn#Q@({ZUA$u8awJGe+y(eQ~K1`Ui=+b(1
zex}ivMe@}i|Js%R%@u4=d21o{tlM@kDIu}Mn+_?Rv<Fz8B=F6PyTRtYKs;5qcEog1
z=F5nY!{aWInHOjxNl3BAAAl&UJo-=kBjKDHgNO9j*c-&9m+#+n3B0DToCAt!jTqgs
z>DF;(zkSPBEpj6y4U?OMk&kZKr&kEO7z}E<H*=#PvxM{agqYL>ohJzVy4Xk6kU*D4
z%9lx|^N=vtkX?S<+zhGe;9KfFg<YK*yJlv+t6M<K*yQon=CSt)b>2wPJEOZ5nhK`t
zwm40c8WVS{<J;=tu6RIUr=?91R#MXZzO)^bo<T3{iaRrr7L^@4_{SYhx9dIe2LUb~
zf*Z}E4+|kHGyvl#g<Ys7UG`o!$^T*08h`2}ock;il+O6Bs{kCz>BOgW19%0Z(`V1y
zwHVg}T<?r|A9t^?oRL%!hx(w+&>B~bbL<D{hOqZT33-+knB%Y;nbaU#FkN4<zx<2U
zd{lo)VrL*ym1Lp6vPReyVXx5Jic+Y7a)A0qfSELD5Q!5ncMcYW9iMItx1L;(3i|)_
zrZ<FWZje6MN>d<Q18G3XBRG;0fs{gOGxH9W2w9Pm2q!kcRvXJPuZ)>a%-o}RpP3pw
zD#TscxBNV!Ajr(AFn4{`h{5(luYbhQ{v{dtum(A=fwXc%339U#Hw&WCj(A11YcV8^
zq~rmn{~$+VT&BI-G|_I~v{9Fbqi+u`ALb1`6?t)nKV<XD`M)jhJOrACNT+uRA09mv
z0RAxD#a@Rwj0Pem+z!i%^u!|w*E=v-)uQF7g5bwm#teNz%#2oM-uI7&tPh%=W}a)+
z{5SDWpk&4%h>MQNVPue}9_uhRY;nf%gk3c@OPpKb(Y_}XEInQz)ujn1`WsEtLpdx_
zZ{;L#7ZwX<oqzr;Bo7A3U^$M*E5g`}cQr@t<nn<hu^4_T2)zmizRhvF2&ad<1K(5e
zNd6l|jLvu?6J*^JGo4Ocw;df!>d+MaEP;6tqu1UU$DIM7fzinYp~7Zc7kj}o&T1r}
zN8pnc^gQPQx`*&s8Gg-0J3)zghT{42O9b>Fc>MdX(Ah9h*I8O?EvG+M|2U~MWM;jf
zB*h_(^Wm!k+|99jO80GBURWUM)aU`j{sx5V7nh&?@mK?D!X>9PL-+d|AA0c|8@G#a
zMX`m8O)i3T|6oGU&5<_r6NQ@B+*!-%;p;-+m5zXSTte7oqx&XV&>DXSa?_$D)glwi
z_`5KD!eDB@jIeWU#+mH=mzNZ}xVRJV#t3qjlGk#m8+%1T@=KKKTnR!&(v{|&BJxC_
z4K;&m!K;AWAUKtx=KTW9X5DDTtD;GOL(M?!dUc7d_dK{9xwZf`**?Uy>kEL;8V#~f
z1US7B@*XkEjnJeg%kp?O6C);|WVhQ*{9}foe9QF$J$SpD!Q3lREk@Z!f}ewKbrS3|
z@jiYV1fzj%#l+hQ!accl-s&HOT`wLI$6vtd0xW3dcKsKR4ENsIh{<~tJi>EV0XTM`
z68Y5K0Iq^wl!~84J8e`CX!{@U18_q$GeQ|Suig)0PIF9Pzk#Isfda8BU<Ihf#Sa2y
z-i#kRJgyh*gr($Ji>C!pBg}MS_D9N}$eKqDi_rqS#~Cv={{ASG%h0OFqMPGWA-My=
zwSPSZNm^358+hDdU31KMf^cr|-WEJoiH4!A{jube1oH@IZ0`FMq#H;#i%IM<+tXlP
z#94Z=g!EutNw8~ze+#C$zGndPps5YX`$R7<oxcApXq^GPium($&}JKxXOd;!_5zSI
z;F*Cm)^ynY?%%+!d^5gYUxISwYXfLV55&eVgSY^dirxx6>Xj0OeUNUuZWrRd#pac3
zOS#CO`BgyIv3pQ<I2@5)eyx1U<&K=atok}g&b=MGZBy(lxjlSHl*`gS*rNT`q3?jW
zmVA!~op(!&rVjM?`i;0KO}#@|<L*L5qro(jU-TYf_ufH?uhF;W`i(X4=Kt^q!%3Ez
zalwB}!bn6VC0PA^5VB6U7gX<E(Jmam1W>Puk3Il#8*>bc2R|$kxCPYi;d3Fmn~jS3
zK7uMOHZDF9?SDIaV)>67U-zf`|BZyWMY$f7I#2xblLATiL}m%Id<s-j9UI4qcI)Zz
z!5y{$mz!xys+9A820}M&YLDRxI*&eLOK)<w&w<>GsQ*lsjciO7mg}CJ8Aa!pAaIx~
z>{#imk_d|jZqPUa3GxSUTdULk4Mayl_7}5#O&n)b??f^F|G;v4%K-7QZ=fma;I#Nk
zK_|&Rk2h=L-fv3+*=iOal;455&m%9wKa9uP8-gh=Zc*YJh0P(}eBnGd%l80ug!B~I
zDnFEenJY#Pi>C!RUldPdg!qgQw|ghsnA(q!?t#AB)t={f|7(D_{qU0BDKO6hI84AF
z9&^*rV3Z80d#s+rhzH^kh0TfP3hP^SrV;gGU?ixc-Aprs@Fm(TvDPevU50d8=}{Su
zo3(s`rk9NNIMUW7D=gqrTI3dkq41+{lH%?wVh+M^BN--NHYbGBWkmORTe{ZVfM#GT
zuUg<?Y90WyNDS{m(_>x`+6`^IW#%if<)E_`R|qis<WFt%{N-cHB8`q)g*ZYJ=Zy^)
zgsgLddjCT-1>`i}854dD;r3=U3?v6~EJTM1cE4sqr>0%8-$MS3<)I}W6yYi`p&`WY
z{|!LdxoW>C4)z788;m)%n(j_!Sp>|LWm(k|q00(!d6|FWbCE7D$NfsQFA5EFb<Eph
zu@bd~u6W#%kj_<}bUIKYrf`=Ftq<WItP{3g2EbXeIo2CXX9{s11B1BedbL4!Nw;FV
z<@nzP_k};n?k~*WaBa~O%PwD1#k*XMZRyKGoH2%E)PJ8909-Si0>&XCnX!^JcRVSI
zHm>x>+A9*wLpk}SWZpSdn6pdW_0^T2-BMD*Y~l7n7X*v10_-f9DPjeYWTCfm+TsOK
znyF~MRSCjHseVsbov@RZB3n!q?W8NMZOk^bIPihPfa|S?`vIjmX$>$s52BqhF@8;m
z%xJD#cpzAS3nN+4Eo(V4<?W94;Sg>Fv{khqUUvkD)7QaLSCG*wjAB-XDsl8k!tR7g
zb(g;T{%HR#KX@M~?7t;Z)ta-@SRhK5cDaj<Gl9Y)E=fgrv7j&CG9Jnd=+aH=?U`0O
zGA~?s0{`a@qcbx)lS2Icw83~?RK5GFhi$bsfGaMqx<~vT(nU)rSc-O<AjiNQAifrv
zTUVwNJ?lc3o{VSKbgTHLP&W+j5+?WRPKV?-To2gAL*z`EW3}}`Ts*E(ZxQAHwc(`i
zJ|YChunzK!(FPEiaWs>hI$hs@Y8`mor<2UIVTqL`8!?|C&t_Kg&>im58v)YCnvo@M
zLXbKlu5P1Fi)`wD@W-+5W(3{Tj6i-AW1h_mrt_B`B8=Y(Gncpni|ra@H<|&p#urGF
zq&mmLLR;`f?#rmb#tx$4M6&UZJw=%lJWedD;Z`9Ocd$9O*^(fA4L|pBzrt=&Eo0nE
zZB+p7JvfbQC%{RyV{+Be5o>M@l6%o0-Z8EaU2j9)an!rtJdV6g`L4CX8&teACd6#i
zJFz@{M7c(6TR_||^<orAiem(7>LZ=?6gF!*xyiUoq>Gb0X<*p5T_Xa0{27tCbTW@P
zX?tkbP*PK#Hy1#k(4I8D7vT;xV(s{B2PpEbRj0<(9SOS+rg^J>Fg+b{yI^-BN%onG
zV!oY#NDzgYj<{lHh_WEin+J8xy4fzEE(DK&aFx&^K;LL%^(PUzW47xg+Y=Y>T9V2B
zVYR~DA(TeMa``0%{f)LYQmwvu51@Siq1|;)KE`jkXGu#emMj^@UoTX3MUmr<y$HK_
z9X)dOcDDeh-kUrd;2u(t8Sh(^*!T}1<}?-{ta_rIN?TVitIM$hoXTKdOX{`{3(5nc
zqs4Dcd@PJwP3eu#6@)EV(c+owj}Y~qhhHV+`TK&nO)#S7=_=I`Z{QVR7er^l_?PI+
zPdnR;rJG7xOdL|qSyhNjJAjINEw*besUWW&h{dZV0y{dM<k?IB<s9gblNHR|$G}Y;
zd6rfn_wN2G^6MnRzek`uFS=84aT}y-&m&vhD$>#~3Uib;Tn^vc4oO|K=cc5wVh1$M
zq!P_g*q`eeV>CRuMx6xaY;`4&o|cZ;I|1DZaZezn2klBO>;`p>xh_HY_3Iu0($>2Z
zH}6N-JVeqyDqa%pBy<4Fw!L2ptq;bYII|xzk7##7N_2xjrOG|je-w6!Ie&8iMVB;c
zV9p7JitZQ^%@Aa{VWm4xlp7t54$X9r32`ReW#NNLYxI8x0m+2ZTwFMrunUC|Q9Lfv
zZIib5oI`e>0^}?w_44)jsnF)c<PP<qy@mitZh$?~1q#y#^{hx-EW&Bao7xv5bA{`E
z65>D80V%Q=oz{n#u|G&=9;<tPOJlwRKuI{=y=^YS3D;<fW|1x-7qkPo*eo<FKqXE+
z5IRpQ3?v>&(R{G;3{Z-#%dO`QA?!M}54FIDqYedd4n4@bH$MW>O_4-b+SSZ`B$!#n
zE|NDdhGL{33duIBH|`hV#$bynHya-aAxGUZuX<GZTH{|=RND&5jO9X{5;g#NG;WC_
z1(5zU^zP%RMhe_r@v1RuosAEW%Cl3S{b)XL8FU_^3}*<b&jM1^?@#{uJ`?Pze8rKm
z`=6jP!?MF&DtziOpejCRpNYp3cEUV!B<JpT{{k5pk}=uX<D9146|~2WqRc`bV6|4r
zX2*l&@yB&Jwx3k;$`b(HnzdexC|?!L%+WU#KQ#VgIP^x}iTq`bu^?$nbrOJ^t(}Ll
z<~teE>04r=_b?6;==2#3IE@!gwfb?OvD+zx^W?__DM8Ftrvl|s=rtDiMW=%~T~>9j
zza{+I@(e&1hF-{G`stqwz}}l_;*&E0oO<F3XNJF>4dm20`17CyP4xG_lxQ@MWT&y^
zIVCB?lr-!W4FLWl9r{^79=JTIz>AHs*ttNosa|Qv1N3<%2{~`5#zyA@xRLt?)dABz
z@kF5-zzA)kMFgzD`DI+Guv;9Ty?XHV%LbCxqFNmFSK^oj^fi2nT`M5>KPi2%m|p;5
z)=8|0(1xBeLtwo(CEwj22q`^~Q)W7vtZ@-2g=4r?;tCOY)N@%FKV1xMR&tNY7-v8p
za!J7ue`pZa;#FZ(3Q<7(?(YsynKC6VQaHDLQg0uKn*_N&lG*pg)0YBJfIi6rJ{D2$
zUGl1|33u9M<rBHhtHq*M0JxawTB6->Wr@HE5T)F?uPOj1SCqFW3&_(Z2B5274M{V~
z$f6d*Mdi^UHPeZrT|_B1@}{W(hRa}kOum+&*~)#F>={110fh2v`HRJGgmC#~@W#ae
z<)13h>EpY1LYyd?FPJ4;V)U&L7VB|A!}d#_E3{91Pk%TmoB~Lwigf>dfi9Z-UJS(w
zw*zDjw^!57dhUM|NOMtzL0%P9F9F`ld+sdBu))Q7*f#>qpC$~E7r(3gkM<NjeG36j
zV2ZTG4ilXx1w<PcH{v(9OJU;A_c;C(aey=W@O#VmBgT&%9nGTYFufe<mcEa0z5B9b
zkP&Je@xU)W;D~$dgV3I1WwVVmFDcl5F4)y)m68vHSo$F#lGY>Idnjyn^`uv|M?DPS
zax-1VJ0fX`bj>mOQ4>iwYM#d+NCnO6cw1qQg7lUvHHy9d3FeHk*h+lg?uimjo)s#*
zKNs^p3Ff+S0>pbwI=c&WUDzL`8mnzrdm5N5i4`NBC0tkE?U8;A8zvUje+YMa`L!f8
zqFxi|KH0`;htDg|f#e=7d*P~Rmlk_r7VDunPY`{Y(Q>1Lo)~Z>B=62T+&&}JtZGA?
z_Q}5>orxUnauf8!^Cd!0dsiIxLWv+fe<D07#0%h<anboAl-q)<LAElly;MG6ik1xg
z(pNy>Ko4ik1zsauW~)~1WB6t*p=J1qQ?I=a;>OA9)7U}_bM^bAs5kiy=v@7dt~C3<
zCa9h~QgFHxO>p&Fp!5eFY2((n%U6;!spXb`r$E>Xv`fdJ_FjP)wk(MQ@xXt9oOZge
z;E73+`&R=c-az|)gF1H|^geOnQ4J-_miR`Lo3|#_4ETtriVq5iZwndr13oR#jG#d!
zbiMjRkUT_Xoxk))(B}Twk+JZ{P-c^qr4W(#$GXBaQYhUf3cIzr_>H|9(X&-7&o+KS
z6n9(M(^grI#HK%d>QqK?vsY0V{z<_mZWQ6lqyuZ(zkVXnP2uawq3HO`AEAWY9p`-x
zk^3?iWE_Lu5K?bo{9DNB#KvEM)mu*wMx}N3v#&v&c2l(j!}+#Y`~QGkts3&Z6Gb}p
z=6<yFtMQW%a}6(tZPE2jNrQQqPRYxQf*hrjO;|kqT}gzux3TAozkXj3zdaU*{s7`a
zqy5pFtS^@P5zJ-LOSAWj%$*c##@OX2XtRqu<;l9Q;1NNAu9Ehe@;>+L{{gw~#Ysdh
z&KBy{p4=M`it=ZzQn#YRtT)rh6ahz8Fz_}}&J5SP-PJU6-Yw82Vx1Zv^UMsL`<SM<
zf!Iul)8)#6YqqX<qycs2Ez5_3D2o~eYoS@7oK^B)nwG@31=95@+^udrD`8Iz2)MC(
zS#LG~=i6r!(rtt&I!}S!qOiLmx6RlXsTZPTF}tIz&Ex7PD{SsFS2JXgc#NJC%!y8B
zGU2tlE+MWA8+!&sEj||H577%r_HEHNH<-)MXqJ)-QmH@AQ&LkeVG}m%d?02XQnZ0M
zcK%;{hK6T4DNJ4fFqd7DQpwxc^}H?E-!kHR;`s%kT?|BiEO*rJSHA)iQ)P}2{}h$`
zHDmC^UqgyNEgkW+!u|^#tN2Q!xr;|x4(;151mOBGjAbvta(j3KY(<Ng8H+{!4uSRl
zAg7|Dom6_2JC6MgfV(2Q5osk$v--;4HZp))Jn701xn7`|E!`4LIUfq5&^?s$j|&ra
zp}W!4`rD$AW&kR6$^JRUEavDqr-~+p>m?gICgxcD7c`G2o+{ek%TokO0zI+z68^rw
zqciqY*fl|B*3})={HH7v&uje26vv}d151|JP044#*#h#w7^Hs47=$qKX{C;*MdaMI
zuF!xn(^BOV`6_@-vkK8%DaFu*mxf?9u1$*#mm%n~S+j3P0q{xAcvf4Mu#-kHLDOto
z+#oD>P_oxL*%DtY2k53r(lNQRMQdq!P_ubjZ*RQ00whJ?5@B@Qyds1vB~@$E`1MK!
zsZPb3uNwQWToA^>V1H9wCWOMVafk(1A?(!Azm45jgLDzl3!(IMcwSQ%>THmAHZh@{
zw>pTYt`$eero+nr+{eqXdteO^mvU6>y%vNSVBcYsbB^I)FmIqaey?C&qd39K!@DF+
zD~|xK&q3;_;@fTlofJC<U4?TmFU+MwomDcfJw}#T=pt1x#~FyeQDC_<qiD=-C_WVA
zjN!LLxhdn(qf2_p{0UFgLd+!8kK-(n?#Yabp4feCp?iAS1|2jW+F9|SMY`m?Z9+*w
zHzM<|4XI(o#IdEqK1?vRWS279SZuF9$R;T)(jC7L?cyT$s8!`nnH|BWzjs>NL-wyj
z*g4}LD{<_Ib%ES9r6=7VUkNa~SO;a;{=Jag_jxW7nfg%tdcBfTR_%*ng8U&ev_Z1F
ztbvRep3GVP+xQF>jHI^um-YF~U4;%d--b}soR*6vHzHhSgb}go#?USYz8Z120TVtF
z<{BjfQ$G8q5sgJxMwHl`Zc4OL%f5lA2p}KyBI6o`UGP>ZH({c%Ifxrn{sLnQNM~Yg
zM~<t4=)fo?wZ)SnoI0Ws<SUzQ2|%6LxKqY2wkn^a9)vexjK#JF@;IvCtthF=V=XI`
zPLJx#(c6@7l1~#F@^eA1ezIX|X2Kb@Etpw^r}k>Ji}s*E`oO!Y(#%+5d%)77x{}&Y
zh%3}7PhyOm(*-$o+;sRM;t4wxOj-?T$Me#TU@)$aeM0L_B^hd=j)6A~AXiM=rj8`Q
zM3KEjWLv8%jc95>wgMT53A;FH1pD#G?u2uj$Pq_NydgwA1K*Sv?LpXmhMSyo_AGJr
z^a(ZIV=oZ1MLWKj_YWloJ8MnOCkP=0bLc@)dHq!Bm@4tozBgd*Gkr46*hGM1C0SlN
zKC}yTQShervM3iDy>HDvZwql(LKYX}_kr}phil{tMCLiua}w46wF%6*_Uf_0cu@fL
z=I$`&Y%YJIl*OK9r#1jmPtzD4mf9hmJ&e$3ZfzjaG+;iNxsy1~G}-OzrKih<xj*n4
z81GuQ#qYa;%tCCj@C(=r?Q+3UTBuh*?!q)Qw*1=C+$5CJqSO(uDr|nS)=1%&<Jd}l
zB^?@zSkoH-y2N_vQ-D8{ZjrZwOY8?|HliREANE7mr-m`Z<1YiyZUy|Qekv-j7+EeW
zZKH(-f%7o#>Po`MX9PKBt!}kW$DU;hFtw2T%bgT<&KPCH;Ub*~r#)(bG<2$Cck*RV
z4oe#OiNY>qa-=zIsL(76oiS0gE6OY)%>Wp<%QQe|hquSraC(UYOU>n&<sd*h5ihtf
zl^k3?7zrmP%z&im+OO`WuruIFH@ShsS)D*<fQC%0dPt7X;O#t|unUZ;?UZ=^2xxZ|
zg#Gp=IO<3+OEJ<3Fv~IGW#R4)9AS7kQi^EthEQkL(bN+M9}Ved8apC3I0i~wW`}!`
z!tS{}_`}2eYL(+aoOyagP4`&$3FCX(s%0Ra5LNF4oQJPH-v9RS3{`)ud=eCs>C{SG
zsh}e=zxTyjCquhbylEG^i*(m!*EC^RoGrlddCZ~`N1OufHs{68<j!-xQ%e+fX9Ka|
z>Cn_g$A&M@AncM;5F2Ow?$2Q4p*t1!b1gALfb)>aZ=508A5GS#KpV^tgt?rE^y0QN
z3)zBD_RnV(5|f;GS)@7GH`v!fISx7-#GRWPIxAh1^J0G~5t~r3W^6AbBoAflDM=S|
zTq9l|r9u3#ju7{!Npjw-CmUn?Mzki}MRB5hUI@HsqFM`|N7&s3F$xx)2+%dX+#}po
zYr;_Ti}NAPVeb%m$lwB}B2~?Q|BZ0HLGZpj*G15-E0c)$zjFhTJ`2TvN=Qz!4|_Rk
ze&8iQE<c)+a9QA|0$q}>j%l&a-yyXIGqMzPzd)$06Se@lUmC43&!td|!nBKd_i_m8
z+{Jd`_$vszv<OB=#R;NaTDFl|{C>U~Bo~B5TJ9Av&A0~8J)n;%IPK1UYG9}{c#nJ5
zwZwD9*y2gt9iIux{HaW;#1hv*Q?#Tcajpnwk62UpV-GZt-l;=Ovhn2gB?0F4DZJ|C
z1UP47$#nyS`7nta9Gu?X2w;|R<BEe|NY@5A(|`;)a7rN1MPd`&AD4*E-LoC53RZ%f
zON=I@+_6f7=8((zOSJ2WUkVhHEAgJ7H2!7pIPa}c_5R)6fhJVipMQB9u=$974&D{G
z_L=z(e~46#+TJS4WkU&IAYKriy9yJ8oJcJ44<Og6JS!Fb0v%5}XzD$G5=4ons2SsB
zQS<@~I^yWN2zyYaiBjuJH7;nNl(%KjMPBr7;(C-wdxH1<yy{&q+!^XQ7HwR%yaz~q
zFiC*k5v6X&jmaPGcen?U+x!m@rUPh@#&aTE-EJh<apZ&0x%n6<&x>?HrJpJ1d<@WS
zA1bH>o@l7a#3iBTt>u;sy_p{-CS_pP7_%zu0;7+bT)SNUNP$ozPsjFa1myztvJk!~
z+I@rD4Q;^RX~aO6r7!k*lo+{9*{>$)$$P@cmo|;XA9EP-YAp6RVMZ|C+BQ+pJ(u+m
z8Ouuoa#wBV(b{H75xy7b;!t@W^m(Fug~W)7cl475;PEOhcoumIKv`5`DFtoWQtLp@
z!2zJk>5IMjnfm7!E2XZ2)CDdQPPt_)cjME9T^=c#v?Q75<YxfQM5J%B0*^}t@eOv4
z@rlCp0P4+Da`3ZIE<A6#)zH0%c5B*~ZgRnJ&vT#@xTiUuRL~zZ58@x_JJBSKOf&xX
z0$~?NH}&4qefA<C`BhL78i@D*4du?$$*^AQWq-Z)T5H9`R|x0EM>R(q%Zac0Z#~}9
ztT1I5>WEGSbIfY1B=vEOAh)pYRM~*n;;j62b8{u`QrLZDNJiT&vEXYUZrr|#BpC|}
z@O)5}Hhr_*KCUG=b4cFec6%L~`e1-IZfv|M%0<DKKgWKw>)!y%%VAU6I7ub{@;5;#
zJ_?2}E9i{)fLAp<ihp_wm;#w(WG(nMfOFPkDKXzW(6G7%_i>Bm@Bk}$@4LoXlVaod
z2)mwm<<?p82qAeQK`zd;Arb8I{{>})Q;QYilJ}i}v>_f;*lc0;rbi@z7DFznbJSuV
z@ITdOIi^VKeB>{u(Lg;4`%7si^v4MT+#L}9U>d3!Z_Q6iECeSQvz;%%v9g*ov2Onq
z(ACiF$24)!XCPD_Ie5INFua`HAIE%7(Cwc5wCP3azYEJPHh{m24@DEbw-Rf8Ns!7Q
zw$bH$RitC$?qDDe5KSh?dE$J9-LlPklZ=EP`U=QH5;-XYh|6fnlAiw>9A@G)q849>
zAPq^vJO7Wc>n3ll=&#ZfR{RFgd@^v#w*bzmg)@#`RQdEfAd1}2Zu9#_be1FDnEwI5
zLkJb&_~Va|o<VSA%#_5@nz2W5k81oQ32~)bY*V-ViEqr__93*$IX4J#8Jc>#n>2+V
z^D~gkDP6IlIBKR*_1aK<78wM^vC}LClpcA~#6x}$2t#BY9Dkh+!kM9YsR!Lg&0Y|>
zt|m^C5I*RvAkJTZP6#;I#GY}-xd^-W@(dIL0j9{I_@B^R+T^lmD1J4!Q^06!LQEA=
zMi>i{+v(4O%{lCk;vtc)Alh@ohsB)p0Aw>m$>I^)W&OmR^-pG4rCYnyDGg*my#uA$
zd5Pmf`8?JfRYBCGjec>j!cLq;p{J&Oj*Io^d^r|psMF>ziRdC&>&x|mNMwM?d!q%)
z-*R}YX6hDX4)x)Kg!d<OvcGRZ(E8>^mq(t=rv3`p9i9akGcVDkjc4~bL*d*Fr2#;d
zHu*JBy{iqyWg^WmM2YB9eUSg}#*ZTd{`H0U--D6sZgrzdRDV-2nUD6iw$@@$m~-OR
zfRTk8p8%o*Op;OfX(Dn-dUY95i+?Q)<OaZzO%?0pgBS4^O<nY;?G`P6*Q#_e-4}qk
zrX<i4BNl_E?5(^jluR)u38*($2d>y^vHRjcuC8=>UT=`nM*UKx2Qv&sqAv0IsY?QC
zpmxTi3OYfZAvBEpmnxqigU1)J=FF!B=5?kn*I;xlyL5?*3_K6v*96i&Si;90%Q)P`
z9I(`~g!8mF*wP}4m-t>-9%q^e+GDTfK-^wz-PyVgTONSYws*&<6-sQo&W;C#Ftei}
zh9cRx6`?Z!q=kH{Xs2Zd3m0a96GYpf6m>2W;EpKGy#D4PIb}b1Rj0~AGpQr0t3lDs
z9F#}I#UgTN<eEJ05bZ3IAT<_Uy%9kwB&&$d-H|zKLy;6X4W~Frl&jCNnVnhuxF!h9
zlU4KWYY}!{t-54?NVFRZ(<heU_IPG^iNamyxY%h#iNgHHe9v>Pg3NXr7~v>${JSFy
z2FWH2;lXFZ+*R=Aq74fDCq@;pw`o#VFK83!Qcvz35*z!3cz-cw?bvHHR2JW}yY8=Y
z_mu=;3ydirqKP2Q<Bf#4O4Syw7LbL_G8QOz0HjZxa|O5pD5-L%$KmEUVP4gkkk~*o
zE2&~@j4L1aRi&-jA;2H&R+f3DE&7GI{Je^W!eKnv_zW#^Y(>5h;8x@5lvrhaAyL4I
zHAFg54%HnD93Gik8%zzyGCWG*)MGG(t%-1b*y6D(cufGz!i0yJa<1P)xskccpB8hA
zb~9uhV-g=Oki>f_>z)cT9$K4Y<8=tSBAj7*SY-aRZuy$=KF~YBTU~Lsz#L8PDQfYz
z^*|_~?rmf5^~*PvZtajz0FT=?03=6=Sq|8cu#1Sqn?5Sq>9O`}=H6jrklaLh`xqw(
zbJH?YGhVClfuQ_hrBjvpy73{Ll$JTqqisekvgUp}j!@W%=rJ(0-YxlW*5z1cbN)7O
z`v<jsJ5YezjX5^n66xVBl@{#l)Tw9R0@TF->t$qh2_Z`)6>+V?&eCEPq)HQkoN6~m
ztfxf!d!!Bc(2FIv1aUc#cq5I!u>sVk$c|*jRwYJ1Lh_;br~&BFx@w%V4KbXrd>v_6
zUAb+^h?B64^=8`+go5H!A$k?gOy?>#ZG0~i=H6pNGjVLb9RTZtrDqTY7D@Zj`xIX9
znS%%kPZZ!`jNE>a9ieiqIYX>2vc5mTgJTvwJS*5aqSnEtvll(QcLh?+L7MpBod`P@
zH2M*TpC`cGslUpBuy1E*=ZcQsN4r4f{@;W9n$#uM+ZE7nOeDcV{kXvr^!wfT%axOQ
zA$kbb*Vsh3o1VG0Zsh%i&^#Sbm=5IpvD)rnE@0wvtg}Y}xEyAUeRfX(H@L4@<HZJ&
zE*@Lrn!SkQTJhQsqx=uhuwXR$M3etm5=5lMN#+3obt7nQEVEn$Uu^n=fhk9&(YH5*
zS%G0AyAvIyPZj89kjfoW%Opv=U$85IPkSEReM^9A$^*PT$u8~CeZcB-4I-}vn;JCC
zi>E|8!=ZFmlB@1l8<9Au48*d{B@(tUHE##6Z2@Fbn!=l)Rf4%&qI{C3>oZ%y{=XjB
z67LCjpZ9f6OZ-QWdmWbzv0NKuZX8~X8Hfi3z*nhJ|4(6Om2Rih*-vjT7;inJ!p=5!
z?@3@zb27)plSI11F}Hsx3Z^7Yd1i;zJE7g!cobBUj|(BbJdw1=kD_RzF5N1w*5$uY
z5MeIAK$Q8*YCIwayP-?hXK(e0fYP}K*?_VhT-?)$g%FfG(Ov+UIXE@u=_6R5z}O<R
zB<J~~g=%==8}II#!xcoz3l0!=ejE*%E;juIz&Q@TuY)B$_750NHmQP#O$JTt71Du)
zr$FQ>w^!;CNCyPfi(MJyMFWP-rUK1;`Cvf=SB))%nRi%~qp}_gO$DK-x<G$kVSmx>
ztJRj6dm5N?YfpC`wK!god$=Ck?&dMw)bXbS*IAJq^y&a}$Nr#hT3JiqaDbCTp$*j!
zj;}4T?14be6&Y(64t8sCi7<a97l-$Wbg`5FWOTO<Djy*Ml0{l)JS)rv(Df`%0@-=Q
zJ3=*LSqg9z58-s>MW_7`NT-Y5rxcKnXdr%>G?qTJeA0%|&0Y->F--hXBwbIB^9;n2
zhe4TH+%Bftts<!2G!!9$ZygTkd^z(>X0`dtk>yJqLMJ7fa|D^qy*xhNkBc+O1x6hO
z>^{U9n<9@n8lZIQv2mVg*AYi4(SKY?ramicV}GPzVrdmZ{CtW?r`j^a?mJqK2XLwb
zGTPqt1Zet+9=2#xxIV1ekVx<Ecwu5hx=LB*L@1cfs}Q`^Ai_y?Dz7(Bf_Cj$k$7(+
z&N&&x39=jY<^Q{<l&=sx^vBw#mN+ykr)B#6X2K{G8<RL(VYj{xxl&_h^;zXKP=;y?
zY8<CKvIfiug<S*`7v*=fSBM!P-8)uq0d7m2oM4O89p^W`N{_>!9@~h-5`hbk{-~V6
zcP@ve8EG)i|L2lC9%tCme9{2;66gB=IFlGI5>icx)t?dK5^!4~%bR##kQ0|qpIk-K
zHnW{oKIPmYor{%)z!*fsamLw%J)4dhA3yvB%2UbsSp8f_)6QkY`Tk={6;-r1{tDp+
zk)N?70zTmaFw((1tWu53MY(W#i$=%R^DYE(J4>Pxm;MbpGr7_%KPW3+4B};G5Q~{+
z)L?4z-}=KP;O??aCUM+lkZudzIo9Hy%K@AjO9n2%I-~CjAQ~1&k*6t`MmYU;^Q$1t
zq2%*YYHCc7SA)V2wncG<!kMiW9>h7<0GUPmRpJ^^E;lbZv1f@ph4_;+5LpxsxDLog
z?qm`E$Muk|J2#G<@t$ZiP|~TYF4KM#=4wgy#VeLK061B_d#jm#H9<~?DVR6mMLQjy
z^W=EKmESTqmUOrhXE%3<0H;$)YUh>sNKh^h+l{#W=8}%zl%+hszXi-w6K4^5H>LnX
z0-YAyvh?og2@MFH!em;o{jJ0@i+CPnpuHA92!X?TIO2rc9ByOG?0S1ikBinGnPhk0
z0hH&OBz(M30N--Li1q$Kn8r-~WxhMh$JA7M>d`ES|I>2Tl?mi?W8zQ|q*5E|j%O8i
z`E>n954hlNkle+Z@p5wyr0cF_h%0cI{j|XPxfoqqS!*XR2z%XIV#`yoyk=D60D-yn
zvSaGS?kn+<+N3siaoq!eZVH^oHb?Ix&~!E(ZMx%3Q63VAm6IK7()4*)xEsE?H+hn5
z;|8i=^3s7KIohoa49y&>!Lj7vM~Ua!qb@hZqH&@SYVVnmhcHV7rtemU&5#fm5=P_J
zR<zql*Gu@gZvcpW7?X27P7Dt;rtl74q?H_MiGMnsR^(EW?7uHM4`ogY{qgT70L(SC
z7v<DBW`7chCS^ja#8FR||2YmOmqkAaa?_#N5zjsY>DuW9c=jznH;_s)bBR5kC623u
zm*|1GQFNZq^rVljAfFXRw$jmyUq1&y9?9U22}h(`CB1%3gQwf}4~4tiFgmac(WTFb
ze*xx$7t>X;8R8f<r2*4<nPT<yHu_M{6Ps?z<2~`z3xr+FBzZXKMQB$|LKC){PYI!r
z+8M=W|AxrCV<*vzFs4nAzretKG6Jgpmq1(^_Nqyqu5H7af*sM<xpAG4JVm7klaPqc
z^9mS!({GNA6y>C7TedD9CQk@<d$i!R;k4HwsSry~9Q6iamyleN)!Y4ooK4bn)g-ge
zn_#(jv0bf3>stV31WwPZagu1uwUIg^<?Kh(AYJwB@izZB^Rb-V)_Dg~ji^T`re#8c
zq_`^+@-z|WzDu-Rdu}n9XV&-+h>O6)!Y*ar{{pxOoDD|DrJ}tij2#zOzYpaO(Su==
zt|8$n`}GgNseC&}p*s|Ip&MGUdT24``jCHI=<eR8Xcj4sv7+CqusN2-FH_ZH4Tz`S
zlE0ewKPHw-iw6)6IUfm766`@&{gkj%Z^rk=CL&!|7_QatZ6R*Q>QqG4`Q~D=&wyQH
zwy%A0+vm`vi3<~63w5}Yu5*#XE~4J3lQ3<gFM&J`IB<7xlgJ?J6PS6#b!mIuDDMQ}
zo+&slXnQAn#uL5*B{v)b#F}3dc1chWZHhfay9ykFNGqlZaW9f%d+w=ugyjDUrY#$G
z6xFyvSiPZ>GUP2n%$-iQ{8PVy)a)w*z<5Ixg_tJgtZxasnzD(K75vQK0l8ds!uHs^
zL8GM0n{%SghPo+UOI#w546zND)$FkEAzgCC!B07-mmk2~oXo!%r(P$3a>$<PA%$H9
zEE4dx6i@#MlDiLEulaw5%)CH10-fV(EH8`%^a|Hhg`K)?IeCdfK<U~e<89G-l;UMZ
zmQJ(HG&;{S5RZy*wPcE7cLWE_0_b9?NvOppvx1PP7MR@?&Sm1LpIW$In7?3`uXk{6
z6T)bn)XNJE_lxiV8#5smoeio!ZAp&NA*0-gx%>rdWlyXiLg((uv4Mj14#;etHcvax
z4d{ANK2EnS(Jm<WtO~EhjGhNNZ=e~!NjG8uGaFEIjNK8}B1Z_wLkdaWHzHjV>_ntb
zH}AaV3r;-pMkxShXg}UGA7K}c1vm5WEP*bGTu#OfqMbzt4_SO5vc7Lkol}~ro6Qdj
z!{FaQ^oVc@lco6bq-Yq(Eg96C3qZLJ2+io6b1&$>xg;4oHhvIA`FO%19{3euXRYhd
z%0iHNdf?JAy%xmL>3hNCoxG;Z{Tm2#K(a7e|2rZ4t*va}maJYh`bKBmEV{mv(xo<o
zJ{gwwEX*Hy#Q6L|Bb_YJoku32=?>sjVXiE~R$iW71lol}OD^qJbuB+#u-R7k_yOCl
z6lz|uCt}xl@1g)=m9{0Z)nZWa0z0?3RAJ2rGK7l-mw?I*%}c=axt1kCT;OhWIHx5k
z?&$)}K-q>axfEnQJA5+4ga#l*%JvwsborngDadxqK{^|zb9L=GmM=+g*knV}DZqW%
z>V&m;Nsw8>5Ua#GE0k}Tuu$>8Q2_PQ#%u2t37au3JP=%0yI%>E+E%!lY>Pug5x=9e
zH=0)_>^9&qK)0m_yd%tIU>~8^ed7J!0nzGo`QG^Tst`I{_r)g)lCT%)%U6SPNjc9C
z#oVhyyYXzxZxQfjg1unKssovjFO(M=2Chshh^G%&(*!~ZHcTSSxOS{=P|00%EfC7k
z+!q@uNOkZm6;l*;(I?9wh%N59f^rwu?h4aLA?m`s`W!Dc{+5*@e?K$4q`{M9yu5=3
z;Rql%{3NVsZWQT;AE->FJ3cAEc{FL-?2p%kxE<I%!S=L^`FK=8le}DB!`RG?0i`^+
zVeF4{$3jrC-l14}9KqZ+Xu>5i&<tU&V0CIcVnSZ#7Nl;?O0~jx!kJO5It+o6gi!Qq
z4>EJ!uA5LI;LwwuU2H6*-n|$llSJ2hAc`zh=2$_lFHbbHh_APHiPMKkp{%+7Eyy`D
z6y#*)-E}~m@Q~E@w_3M+ge-(DKBiUc15tC<@k;b&0-;`4TrR>{avw)APuc*$43+0D
zcDS<VJ-dNoNX(UCV$v-_J$SKOP<xQq6Ah446#j~2YW1x^XQOAr@iVy*fNPHqU7x0c
z0|mifR7K-fg<U6|iQ_iW{+>ryFaY^7M`qI;XKc)$W{G_Y#pOcWqPkYY;qXA5x=D$J
zG=P;mUK8Tp*3~f>%WMh>lPcY)$Q>lYX(op(+{<oO5Ke}Y$xYn6eBa7_(K{k@F)$Wp
zSM^yVLMOL%FKs~tvo^cCl<VNPCW4Z!&W^GDmJskjXU0)mmGn}Eypy#xh?yn@AZ>Mi
zx2-=wAuXQRzWj|JYEF&a@tYlhG$GLghhjBRF5zUhXS8;^klYF7R-5w75ahL`*b^mL
z-!VJ#hx$LP>)45~tAY@N=haw2u`>`#41^f5&d!8$ch`BgDjH_zjs1qZK<0%?Ze7#G
z<evpn+P6-h!AAQbQO={WB*i~>bvneUp(XmfNVvytuo|=O=I^EThD1|T?gS}Uy8CUy
z+>bgrE%wA7y8}2ARKv1^*7)xpB^vK9ae&%)PXKd{;jL5kg+l71j>&%D574fV9z)in
zRkb)>pxN8XETHS7=LEqcY4^|aN5b{?;1YqyqlCE3cv+Y8q-tEhHyATjWomq+AhnfT
zGS=A#!a2%$RWqg``0EqsVzF`Ri{nJ|MU4iES4GrY+fDZ1zQ2eMKQ3x2G>@2}6(QQ6
zAg1n+&fwC`KrS`Xb-HbQA%qgLaXYGopc$+~T6c`E0=P!H%SdcFP#ArILdnI!tgR+#
z(jIiY|0<Fq!;ngB)n<ZYKW<Y*ctLA!mmgK;7WuNdydB&nLS(BG;a(je<_#h+EcUe7
zU1AcT%O>Fhq7DJB9P<}iAiR~^31t2u3T=tgx}e=ZlKvbs2`l$Z-6bVt4@h383&_#f
z>0mNQkBg&)=JFyYNe)%jNC}5u7~Y<vD9(?^2AK^+=OXDrZw^uG^#Cy{{OI<7MEQem
zyyV1=z0kSty6eC+Rfwy}qtOU4UfZt#5_DH%f&TK%fR>e7EGNW+-X>1)K27oaOy&<a
zACi3bt~7UtusTl=%3FvxqY)EzZjG8+oGHwOMf%hk7fyku^K<NpHx>5mt9P>A%YStY
zOpA{db`7|ZMic0(2A~T~-aMOH62{PdC^|)(3kWAVu;f^4T1g(?gfUU1r<QRe<6Tjd
zL(UMUOegHjtN?bz{s3+Qy)2~rjQa(-JF($Hu%=c#<p4mEWOb*AXNscNCg1swi!g`k
zLLw&HI}QYO!i@g3?|(^<6K>U^jH3nD?FW^w(*pr+i8T%`U*k?8T^StRfLPfj!j0Kx
z5Gzk0>FowO#5Ka4FBh)3?&d^x2oPl-pG<V`5#{1aqQ#pfX(~As)QoA805%S90Eh^t
zp}lfgiGgz1_+hco;Q($1sc@hd`h^fU-OoO2yCVp@Y*_uFh)UUB5au4D8=iRLNa)<R
zB$8%9hyS}ARlbxHrQ}cLXJMp_v?x|SnsDw*bgfpAE=H{~S#`Neh_h+oC4kuLPyPVa
zFa3SZvE_S2D6z(IkgmB}x)zr;06k*dj2#*=T$OIRh@Q8L1&#-+k4$FwJw&^0(THGT
z;)E8@3d}`A5Y1dN;{=dA;yG8UI2W7<<kXo%r59F><4*;1>KrNu;u6u0!)di6J)F1v
z8DQ?@_!C+)Q#cuC-FZ=zGuDL(3;kC@a>i0<r{}G57LYT+jJZGFKfBO~EGx10U!dva
zsG0Cyg9xX{ZbWQ4?i>)GLpYKaCHWQ41<o0<h-jyKg&-Pl8WPpX=QYxyaXDQq_E!)W
zqgBr|JSo}<U|%9RX}r^j!ug#0)W7)?3&v5{6io)Wn~Pg6DxbF|J&9UucrlR6p1phR
zFNCT|$so45r2O5dqu7+#U5JZN)eK3)-SzhZA*4(GwH^`VWYL}ICeneImPjnpGUbmu
z1yMAVg<_k_2)h-f2#h9hEt)R}qa;%0kJGP&aL<;)*Y#IH<_79Owe4Qf&afL%CKEu<
z)dfM*T;j={t^si#m~1B#xVS@@zr$m-<R!;m3qr08y4H9n6I}d^irKH@ACCq``M6>6
z==A_z*6Ooc64pL1+&t>(^&46j-T>s`jAzy4?29tKBQ(z}pzU0_0f;4dpkMhW!me=(
zmtFgc%r#{5K-0f=Gl;X1zLN&b#9RD*4~7O`DeMo>*3+6g&#nGf2HhJg?5ta*@Ef1y
zKe?ZarSEV=rid7m!;*scD$IIaZHi|Vtk3LZ=ec>we}KA3T<2i4Rf|Uixfty{JwiJ?
zB?J!Z^bw2S3E_t1e1dt5XgJW1e%-8hJ0jd487qq551GpyuCUWgx(Vot2+1>5I{D6g
z4?v#EeF3c7i5Ws2t+z?7z3jaJ&K|WBURe<BnFCv$i$%E!`q0Uk<NgvE=L*%>MRevN
z3U33^)rf#>e^eYTn$&dA=g=-nBa7yXH6MU5r_hG$kap6wf}9+ySaWae_F##FH4OWU
zTKrd#%ZtawuJ~LuC2YqikX2-rhar^(va?uA1i2$rV5a^|6#Q%Fy{|<cA)L7-U%GL&
z08i2Rluu$$JnHYqrnT`cQ64=wudq7a!H<ENKO=^Zh<}Q7vtY*|Kb7l04&sbzC?<8p
zJ_4LktunQjSJD{U6@lgj%AIM=JxUN|mm(prP>FKhtjp5vlgGM(`DJ#ng`NQOyxXRE
zsK?vvT?AA0DLkxp*^?y-Q}cix71sgfdBNo80&VgXRL*Y*pXt4bWVR4SjJHmo(T0<2
zQ7}4b<Inaqg!@`A_F7p4dDt_ct`)7*CD*rqdJf3t<930oHJ5Lte*sfI^qb<9=LzS<
zvKLiW)-*hQ2s9g67*GSa_5}d4XJm|v_eJ<VNb~P6F5`M%<R8XsUu!J-GC_A7-YM#i
z;iA3KP4}ZRy-64I=jsoaQ8O`mobopw&*kqqLZlajM!!Ndvxxg7J>kfB5SGi%OVpF;
z)jUNdFgGr0p7H~`-m4`Gj-DuzU`)lC?-If8K3F)(QRN*%{4Hw@3nmW{yaq%wRJe2M
zif2V-#w3UHqhAMbk8Ed14Sxd?&dSH*8*e&{KZ0m_i?FleajY6f7&i+cW(BLH+Z4{j
zzP%;(cpK8yvZr-)tKt}exn~ct?TAYSI7#e6cqS+Q`!0yPg|)QOUUl?)B_3?!%n(!m
z1CmFUZtk@m{F^W*kEa6iyH9{wftd+A7dGR9V1=1_?EB>lBvmZ0*EWC=6Go327Poys
z4CkZgG*fA|`w+}^Q;%d}Vig(j5n$p)U(EM0VRwflK_q2Oq)P~P`M5tu)xE0$;NXF!
z>=^-RgdoECUPNQ1)U{ViX6a8Hi7A_VHW6wp*7t`Mc3Baja^U9w?}eG4NLn~1B(K}c
zed@2I^02kSZUYQ$qyw_aX9a;Pl7{Rd#Er|86eo!^Us<l&WCOd*=RmGoI!(ls1|Y^`
z<uAa42?+<PqhCN%axUIt%9n&)h8kychQ-lB92<!!o)blz=Q28`b+t@>RU)+y4Di6o
z|3QlxtOJ{VOV~wjVq<W-$U5)4*eH$ruKX`QG;51=3FLAXA>sOs2yJMsy!|~9T<4Y{
zHa&3e9YIbLaW{;6_ZM+`2f0W^AO8p9(4s?3z_GIkrj`M86BYE#$P=P1qMSI~ONtjC
z<qw&yMo0C>^1tk-q|kMfAZO0TYV5FhtN}FZd7F6RCt|ot90F?i=UC={AY_V)Gs=fY
z{tN}TdSTCJ3cKsHN^ey?deTf|>V?Aoirq?A>^n0MwL#3yoAV-^ev9TJe0R<Qq7+%Q
z{xVznnga|g)<9e*s6LY;LzP(zyWEEb=NWLYi!HRYfmfa#(CyEYOC52gXp&}njbG1E
zK9ITu$6-3)U2_7Oh3>S7Oy??Tj2J&|eDux@P3h$}28+Rxhz}cL>TLpDB^Ftkp<O!<
zNSTkMQ=(aSf%%*gjWZ%X6j6p)^18ra|9bv{BA;$c&sFuu@C87ffplOQ7LN#_pGq?r
z^1KBhU5#3mgCR9*{uNMOg{o7B;z-f3LK~FR8vn5~9E=|tbayY^`jmx;>-_NXh^|H)
z{u>ZYF=&Ry_}@CIZmuNPQP^F8nL-B}loO(0I6`E%Oeeo%;etuVpJwCni-9>Cyv|7U
z^{fy#subPnFkJn8u{fantQ2CZF=B}ltJ1<F+R3bP%OwHL0_LYAnCued^n1|nfW2P|
zaZ+p(=zF^^1&|v`W}|35Fo@S(x<uo)gZ*T;05Pw9FglkZoL6lft`J-4Xtw#X<(p~}
zlOHL-Q;9BreiBK0u$?1gMD5E#D@oSK+VbVw0UT@%#MuH|ndXju?Bi<jvmi5?RX}P7
z_&;8ue8-lR^YQ)y@>Zk%P@c*cAL<Wso5VKZofQewmTAWPT!eYV=2(e-zOtjDMi6uV
zj<DHNlW348fs50Hk^uK*)%dH3-1V41RMeMN17gBph@r~3?XZ%@Bz7qr`t=t3&1)2t
z>t(%2qLd#MYMyCneqSUr-?a8vdrg9xg!D3MY$3}1M203_;cgOGZ(Mms+Gi~Q%E|V0
zD4IpUzZUHT^S<#=VQxxk>LIdar-RqI5x_3R*wMq{VUZ*wCAnxH4dG1Wjd3VW72x_)
zcLpsYctI|m7IO7to|X|<7_FIb^D!W9(R6CT^fKNSrnb@b(XL}5T#hMKj-?{q7(9Qf
z4TLhAF4U<o*8E7`DdWqB(s{Xk0;IF%epS2Nm)8c#ljjf**d<Z-Z`K9OlczQTGp!E|
z%h^0LuWusCg=xn@!*r2OL2J2Fcw_@WmjxYTMiV`+x*?ePgKbxGKDGHq1>%fB;<eaD
zkn`sD4i9G=LzA1v7bE_RO-#PUiGXQ-rA;B-2vZPApuDs>fJ@U;oub!9xs=?!0VZ)(
z>S6Z^G!q#zgR;lK!sWh3R8Cx20Z6F*+bv3_BXA8nUZg96NFb?4&oi+k#@2~z|1{Qi
z*%H+3#{Hp8<aXN%#C+kni1#tN`utm$ScnsOD<_r_lKL5o?yB2BFklg}#8C>nx>i-W
zQ-HaZrp=|cbp%~JvfXTnkpdinNx9v$Q$QC1hCZkWZcVh&V!7=g%dLUm(9!^$a|dur
zA;H|^3OfzNz9?qRyAuGTa|(9G3X)t_Lz^g)&&DChctK%*hR+`cOzaFmH8H^(iUUNL
zsWKCgeb2(X0GWNYYBIFCcvq0jzU-<kz7R&|q(`hUGFI3P%JpT<9uphwUcTU|E_=|u
zSC|Welo+|pzIy;TBm7RV6^lM0&M2!G)Z%bq^-)*j?3T`7j|!#2bY7Ved@9n_(5X;U
z_)2>MQOy>#+hZ>g{$FoRNGtZ12E;j=oA>_Majz1q3CW$N(k4N!*c4e!Rz<towa%E{
zpvI1mpBkS_cCJDDKS6G0y#l+$AN;kdfXbRE(&n+~v|P^hN9fFWPNJv`-XX;NkQdD~
z^-kEk#Nx23cL;if(ApTm#T@$(&YdJ(1br@m0<$fmAKt%j`8XN))MCvFNM?B#Q+YR%
zp&c8Td<Syy$81fQn<2kcq56H$1ey;!SUqR=1!(9sgPY&UCV>RzYXLGVxOI)0tB|fN
zErv~XtSrPummx-594sJjJpH5#*Mle$3wA;@T>LE3EmZ4bfw-g%+6nXEdna7nVp56G
zJjnKFl}>0E4*T_VCv%!0SUA9*b;qvqH|ew2t;K5H{+neoR_Y<_>?d)bcwH}~`9i6E
zdC`KmLr`~<Y1|2laz%W$@7GIe@q8l^ujj7SPbAl)dx)mGT(lcT$3kpuhvF(>^yzjS
z;lyAK!WCuul0VWB1LYI6G6vHVeh)V=&Go6Hy)4X}*7kB*JTVAB^;@t+dRgH-K{Hg^
za6iGMG1KINqO-&)E@`~nB-BNa8Y)ZXzl6BdoDdP*)6YAmly9><TuaI^LUTKDalnW@
zzk#qwcEf{fh2&w4!T~3&=$H-!3z8<#eEXOG`6Ag{FC)kWpcC;hHyvqzVSGaYt8tVF
z=S#<9Sj3zMfWQv+SCzz{<qs-pXj0^}T0AGr%|4!4b)gxM<`R!#vZqgb(7S~?b7T(v
zNlrcf5HQyvJyyes>ODa&0vGPmseJq}5Vwf-E7cfwc=^gAzgk=^D6g^dU7B*7^@#Fe
z4dE7UrUg(}7T1orzws}pP2?hbA6cTK<jU^gKmqFbNSF>+*u}!9pCn-)3UQVU1^YsY
zw~z8CgS=Kb&(Va<73oK=AkuvpMQo0aqA3iT#F$ky;&xKdtNz4)t}>S09PQ#5fZQPl
zd$3uIsX`nDZ;_Pt5&`O`%%!7_Bb-Mu3W3ttI7bi*VlUpe8voLS@D$rGLAh}lhG{oa
zJs!*zZ0hJ`Z=jb=zc`_!+*O&9jKsD+5r~Y3I7m!Vn6b@t%XDynD3`js17Ag5@v#th
z4V(a|18P;Bd=e<EkcXdmT7=owiNVVKCqueqY-w9Nc%>s=IVH!JF4wlFL8L6Dc(A)D
z3e(3VFzIx{RJpS=9#$|jR96MD;GY5VK}nC{u*uXJeWYOLhSUPJJxaHWFf&uW|LD$f
z;h6<W4}&w1r=3;4<N~gf&CXEVEsO#T^~c8wQpFJy#!iS|pIzem2+M~1Sz+dk*g&Uf
zW@No2)SZUC!-&}VFTbQG#JQryInFJy(>V}uu2pth?XkPa^6l6-Rdnt{Qmh|{M}@eB
zD=o=)G}S%f+!9S^273A;=YizTAxX40U#ALl;`l&geiH4ZywTI#a+N?grDUtOh;$>g
zrfZ6!^8w7+3fJE{Wc^8y^U@SmlY^h9gt_3{)kx*$_zOx(xJ&08ccuWEy@HMO;Dv-;
z6f_-pHc%(r&jjX%<<2JQzwG<Bk_w}YzV<f(t`v(#3l@bX0*7g8{hk09I9;5wNt}2w
z5Zy0ngT(ft9D$Qx^6GY$AlF*k@b;MH?*+hv6D#PX2EZ*Lmtn67AWeED%FbV029?_l
z^Ld8)7MFu~XHgtVr9;sb;Pi%GmW=o@6MBs^MqkN4?(RDEF_;s#Cko~Rq#j+-A;NsC
z^&)IY{<m(u8qnR+jyuy^2s>T_YHqSlSQxtTwFOi1RT)}dC(K_<cL$DWExSdiE5tKQ
z>L>%(fzW)-JeB&I!g(Ozzq<$BYsT`_>p{)MYF85%&{y6FU{0grEW?wS_oiP!c-??g
zw+OP#fr8@Io24Pu{C79=hr1?Q@Yqfy9B0RlzQgyTTtF^0q=$UTEsl;tH!`kR=2l1s
zb_=JO+X$K!x@%{X)Dy1@%RP{JMKTpM!EOh17Mz!l6QvHzboRc&v;wN7)wt^ph+KVN
zkZV{>`Uj{>gyO5D497GOE!|}BKm%#j+oS#Qr4Ta>FQ6z#O}MipK6w!L&%9K?S@h3B
z>GpC{B^OwCiKaX##&pC?cbU+t7f;^};mW2-N-FUm3!^5?3-Q0k|4HB`vzZO=;s0E8
z?va?(_7vh_#__!~*1WgGK&mQ^ye=fiKqoHVZO~I#hvL7DFPITfRErJn;|o`{saLAe
z+X?U(N&A%kd^O!GoXR108Hkfa<O#?|-tmec7njqzR;L~BF9=sz@|(8!10YVDmwR-&
zw%BaCaMEUyA0Kasa6&TTg|A0H2$H8m)O1f4nFkR^<0Q!Vr!ap+H8eC{6k=A?bTC}<
zA!zq4>kn#P+C`v0lP4CUwFx0F+Ecfr2a2R*O-k}I5xEtc5v(y-;|^i&o!u?*u_#X$
zRBrMR5*}UHJ@+uM%ff1bbL~e8J!1Um;c<XyHv(1Wbn(ff0C~P-J(E0o=Es0s)VAJU
zuJ>YYA?iS!uJ%yaIWQ%+#Yv(mH+E+o@qXh^oPc)5evkW438Uj25u}2eRdoH6aGo3J
zP})_TCn)!ier_(}`X~G`vm>u}h~Q&A#<|XugmbeX-b|~<8^UrwVFw^N@()h|F$MJW
z#V$`1bd@A3huv#D1LA7nJgG0X67Ad(?Xnzo3UF7zm^BU)nG1}ESos_~S&*9`yMRv;
zwN0J{rlj(_w7tUqN}l%BwD125$Qi*hGWu8m3~bjzdinDZd2X4;D<qYUNpNE37aWD=
zVWJV?y68y6{o&6-+%xtYjDuf<B7Y7vG4H?q;gE7ZR|J1(lX%ff4iE4c%dcK0?EY#e
z28z6fK$l#eEVMs;R1n`|YZs3xoY{r-Y&BjN;5N~%14*El{c1^rhlbPj&f|h8Y7I@6
zo>=m=@=?0WQpVGT(Q4gY@$ef2b5~>k9>={|=t+`hB-)(7n^-?j4=nIjiGl7>E%p*^
z&h$>>5j7|{&1%>mtB7#>HgRt3ZIB#lV#9a&*VREiCfzi}mcraNMIeX(RVaz_AUIx>
z8lPeGpKg_6{`ZJLg+{Cut0)KuQ9Ot-3O5!}tUzTxaYh5}=;1zD7eBRF>3!n5qVg5V
z`t`C9x0$rAtu6fE2PIlEKJ1SjJ_Iobaq>Mr+C;nAIPP<M`<{T(%pBdi;;fH<`OVna
z`V(k(b4sHHG#O27{wXl^PM%&)7vXfZPe*bjz$Mn(9n@j;XCTxad#N}|Ve=lV%D7u(
zy+^0F&>#66#ChtvM^#p;{a*mPRgx<<3C36a63kii6cBx24FQhCF54ccpA_aCxbc@i
zjB#IqP<*vyd@I6fp!mRCwEWiqv{`RQY_Fg>#I_lm9?psX2jr$>%c99fvV%>8=7MlP
z)fqp213-=2InYh}mN3~>d$6o}Uqoq@aq+QedTFvGjOpJ&<*iAT;|-#-ftdMwKsc5r
z?OH4?Qpss9-Bd)q4K0g1L=ah{uMubd;P5!!U0C=>!e%^n?Xi+bXT@-=PUHMniG2mS
z9-O+l!02y$#O5sa{K@I{p|~{n{|Gy4IWgB;H@_3)&w2fWb?eTV#?~{z9e=tb8;Db8
z2FycS4x4A21)2hDqp*X*b@q%K86VAB=t&iv+x;looujXp{dw9hqVYPLli>Nict>Hk
zcY4YSpNg}WL<U$;Vud-N%?TtVv5iRQ$?l~Zi_BRPkV^sPi^b+DF}PapjA3(^1X|g~
z?j+JVaw^h0lGh2*0MV)Sj(N)0EayDqHZL?yp4@H6?V^ZE%t|~X!YOe>rsa3t`9R#G
z*|S;2eivctGpHW5#Ggf}Kh&BBW7GK|)c6!@dxbNfyD*NMz5uj&*P|T+N<tb~lNJPa
z4hZKZl<tVde+8C#IGE&B>k1)ZTw28wzb5RKN3)ta0h6tT{5J|6v6aGRFKP>MtjIiw
zXm06RO#Us9v#btbxhD^3%Lf?M7&&~z_*i>k2y&Bx++P%?wz|J<iG3FFKa<&+j$V|o
zOR1HeDlfQLL9i;OEn)*9W&#_Lp?FDjp3yKJ*W{@k{5p#R*T+#eny3S@fgq<YZLyx-
z=n~@Y&0Q;-=bIWpe{XtBVu?l!R!%IWmUJRDoT1r+?X;ypm|hU=pRb@Z<v$d&@jI|I
z5WGWX)r?qlevvRG`GfmZ1Wf5;M*ZWmgk3mv#<fCUu^b2$<+?h_?|&3s&Q{z$a85&X
zkC(CK`NO%k<+=58L7oew<i#l#x#@B%0J^rx*;>3IAm>T>7-~N@q9i+pxMd}xP{eMm
zf1Xs>`5+q7y+ORRa!Hb<f~R#3_+0@eAp&fPiv+kNh(&oJOy(04R{<ndrtf%HVMl1k
zmP@x0czj&d5#+(OZ8gH4pkaF5@8JVNb2q?tCLV3XlBKX-$RPLaS1*anQ(g^u7Q6Rh
zfO&i%5Wwi6Ck_+ly6Ngk!{~`MfheRljJvK?KEY29De?bv!u(0vL9lA9GaSeVm9%Et
z@TVmz!W~tI2wl~#GXluvXHjg48KUVJsCW*<xgy+Bu$ftO*hl~|tfMK`Q#em{Z2qR_
zzp&bmc}DS9W|c%28VT5)3iSxl3Wud{8X-3bCe1+{b>7^FqMOj{qB6#eCW_l8?H1C`
zaZ`aVIdafS>?GQ~i1Td|8>cA^q>n)Z%P$n7DNU!8RmVV>|5|6%En<hUCDE=SM)|#>
zohZ@`?CD~aaUiZX!p_)zJR~V%pfC`Nt_@MIA$kYMA~c6B^m|~Jl&67OqFJ<)Kn@u5
zuLJ4Yab#u9)Hf%r`-^YbJe($)rqiJ(Yv4R5$jNAGeLcrOmVupQ654+v$Tg(&T%U1O
z)V)3!>}Ee5`zcJ}(fw?Xjt%^WB0ZGDQgfcZ5qRF1)UD0X;ucCWx^&%jW5RjXoytb8
zgA0V+Hz^ou0ad2hTI?&V-YwY{vSm&?(b{I<OvEsQ)UMeA%GK%`Y>wq7LOOdALLn-3
znhk_HOM9nBLuQ&V_wJ-#%W3U8L8MI6F_XQQDLAy^nAmK~l5h{os@Wg@eL~&kxyoW*
ze^h`AEn%s}IyBJCTZ6l;>!zO?jaLgMNAf^V>swK7TA7`*lrXuiyDea@Aj{A{L~6th
z;GtvI?I7G$q!1QMigp5scG&4i4*sz~X8Pu7vGn!?-9OQ^kh7m%b^xJano$edM`5Ri
z-D5wFU7^i2bf5K5L`&R~<Eslips?$NdNL!BzKE=IsU3mM6WrvmJs!VPNtY{~YD^Vv
z&f^!EDZDH0YCxRm(~IJ3?o2EVbUM|2yOb1~CEgg^725TXZ+JS({X$%GX>idqnAn!v
z9WYN|*b?Z>eR2arIXd;Ty9BwsdM0QnuGymiST*y}K6?Vd^T}$mw?#N7+)JU&E1Abl
zdjaMSW)Vhz-0%l5=ibVSpl`SQV}ZDHPZxk!3v%vBX(JZc8z9diavX(@)IK1&9fo@F
zPffRP5|&$@3)py1bgn9w!4g~jdtV?Jh|`kTx<Z&5AafcU^EW}{nSlF;{Y2L5uSNYz
z0qSBnV2)ue5Tw$<)p%c(aGt28ZP3HEYbZvxg1W#}+$OQUJ|rY_gISO}lFq2KIUOCE
zn-q2d5J|*!?Ik*Pi+0I<a|f6cWwX}G?DL=y*O*I=(Xn@DNklHoHtjBn<Qw8z9NhzE
zrpQ1?7h+R-%U9h4ljCHOx$c8<Sh7W50i^cKBh&(1XLd$dXQiHXV*}0hx!S7ywnp4i
z#Bs5aXhvRJSF|YTjC7+_i<bq!<Q8cwUOfPj`xKI(3fXX&7zA``O)?|vpInmQ`EDW{
zC4eMU&^V{@FGr@4!{TlMFqu_FoqG5bM;<YOGuT)W&W{^@2|~E%`BtFw>!>k$MoxwH
z+=MW78&UqGySyN^`4C|4yliuM20(zPgmT?U557aNKTc1_(@9$5t;WZlI5kHlZW^(u
zPn#~c7oHB`(%@FU8e5BYR!kzy-_sir<f<TMh<1<Zt5pYE*jMadkZKDrhOna%;tr8i
zb(oAM336Y>rHS7Ind^Xp>AaoxO{0Z5`F=fnjJe8zAn>P4hnVOR<?^9G&yjWIgGwUU
zrH}|^Tt^t6NS&6sNEAb>1rg?zGYIEpvr^;wCf%zH9^x3fFF!(I7l3(P?vZ9X6vQk-
zMWi#95ba^bJyY!8Amvl$<wk6praN92?eq}yNLg>)!$2sHydD3faOM>97MTmJdw78m
zJ*Hu>$`L@hSJ!F`uj2*yGaewW=uzC$j|9qnjgH3U4in*+qsxb_Rcxu_@IL|MW<Y0x
z*_-?6V@hIqKN4FAax|$GATY<8>sT<e0Wm7e%W=@|gV{VO?I-6w9@yOJ;~widkzziZ
zhQm)FoZD6wP8gI6am82@$YlEy9i0_5jy{=iy+*j-N5a7M%qjnmtg{Z2w7A|j?gTwR
z&_#mn0z2!A1$Wor^i222j_&l#&d%T#G`K@>celj}k^mtDcMHC_`|p10seYU5`~K;x
z&eQdlobprWoC2h25%ts2z80mS*~XfA$f*!6G#8ldNpbl5(~2~j*<F?qqPE8FX8`6-
zAobW1x8c<oIJ5X%*Clf49fQKm=K*=e;57W<Sp_5qCfY)eJR8WIU_SwpPG7N!U{|-N
zf-_CDrG>zZQhRGu&mrvEA@%aRz_IgrfVEM=HcYRCd{@Uxjk2`ju)6X2#G~HnntY}U
zAk0J#xY)SdRtH$3SrpzBK)yV$7^_`K*zMDfLh+>{&5ObQmTt_D<VSGi#h@-ad1{TH
z`4SMfzGlT}5^c_~4)9)ZXWV}&P;F?AXKH+|4#?B+K=fQe4CmcgYK)bxgrpX%mgD0N
z5zc_^cqx7no%?*c5v^PWV20rY7PD7fvmGyxB$AwRjjJKdJ^6`4yx1F82&*xtv2idC
zy%xX~rkUyRxP8E#*>#|~b8A}MLbR*bjCPvF($Rv<6Zs!yuAkw05a);G4&HrOn>G|i
zf%t^C9Yi=44xQ|bs&VWMKsh6hm$AW(^)%RUBtLeS3zJB4DDF})Hw1SGIM#SXh$|$&
zvFKmaQ=sp4B~67zZz7U2mU>KudQZ3+h=S20%5ks=Hw4~t;+#6!+}IXR-dcRE-mK<-
zQIM-DjWAAS<!HGbj9#T}!8&&k*0AhXKkpLd<U8pYCy8{+u^Ud0T=ClCoq#T4iA!Zo
zM8^p7XXtZA<t|9ql(&Z2;GQMGpCFfmp*Y1C<Z|Kt*01?JI}OakQJ0OZgc63tLU#i@
zb)F%JS?+;!q0sDQE39qn3Ig3aY<_qNUbNY-sh`m}C?vNzQv9?Nf98HL7*S=u+x-E;
zE;P##Rx)=9fbZz<@#0P<P-O3l#UJ!%-K-i1D-8ceC9}@a4;6`&8p~YWe_RJJ!bf-T
zn1_jx`vqH7TwovYNb${N25498^(X-K#qGA*WRgf1j7zP_Z81fF*(7iCdS^zB@VQ{~
z1#vVi)ta>2V}RzoG#Iq4*x{c*7C8M*D8P`M?_d1kqDhWL)Fp+$0_1G%@mCRMU>CB$
zQWfQ_3x(NOQmG>=(ivTXx{LFtvapP`pCFoB2X}tem{O<HRr%?nYrLo7F-HZUVFp=;
zSNu0&H^012F7;HA7S&|mEZU`JTF^%L`g#nOg1%VwKTeF@OY}cYnC2RvnrpphpnMpm
zbF)ljDPu)A*_RN_zpJp>N^7K}H*cLh>tr~VV$!&%C^MNeVA`E+T8A3ZB2a%?<2hnc
zkL=~@X_5XY&w0`CJW%f0df|sV5GFW*<{d97x5tH|&8s1afn(_x04S`47O%cYnCi)C
zb4#r85>)0~Z+~ngvev%I!0M>qAr*Y3Rbtmyia1GSj+^q=KwK!yDG^6WOKzmVTw^3P
zIt;)0I#6wG&R%IMvD+J<Do}Dqy!@L`6je$T|IOhtH(c|*1)e7<i6pdoPOW2HgXx%w
zDVH$5*IU_}zU`E00+`$ME|g0tw`hY&jCq7WSD3p;iS)<52V#y?abxs_NVk7dUan~%
zEdD;Y>w*DPk`hqE9-;oYi|deAKY+}gS|+Q@e+W&(V7_k&^0oq8CO_t<v;Lyf9gAXf
z;$_XhT7Z+mlDIJ*5KSGq7K<+x_E635dZ)pKR+9dxO#!kv)^4*J<391Xs47OQ!Y(bF
zq8R(B|HA29Ed3c_8ig`6<J;wP|I^4vD=O@E?O{KzTdf4SzwrnS^-KT-lp6Q0`CCcw
zZ&BD0v3}$6>wkO!;QFJP-Z}*>tV+E5B_LA@Qx>;GnP`;V`ix)k59#uPQ4IT<a9-;%
zmrQy%_=6Y9@>=eQAAT2sr9zVrN1hG8Ap&*QA(=VoFCxv_CY}I^PSHgtX=-QZ+8L7t
z=J~@OYLEQ3h=yQR215+Pvjv(nT3Xrj;<XmYZ0zidCq?CSl3iAOB*YAXm2A9!ssmF>
z^0nXiony$JJ=uXDEdU-2G0}bceet=}hIugH2LKlV4??Jx^~S|N0=Wni$3^@M<tE1z
zKh_fI>dSv&>?XR_Kodrfh*kkkP?9(tKN+HV(zU_7zZU6DX={t|zZKuhzf(sXBfz<j
zn;0jH%A*ahG`#0I&kW;h8Ft{DawrxM;AZ7W+L1M6FB0m}4{O;`bH4c7j9@SVFMjc{
z!u~4h$kDXS58uxM>e}$~DR(gU&I;hcqRo6KQ*rXK7q`yFA5I?y1g1>dZpAskTy?fY
z$TR!m9YM9-2ao!t%_APu=K^-M&@996ZX7f>h;v3{2Gh2m2Y`ObD+DJg2>*~wc0}{M
z5N;3Y%AEU$;yc;yj)*Dq0n~EjJ|gc#mz&>daJcD>;UbDS7_o4VLKz+xoVz1?T0M)h
z;sOOkLO>m{x1ikN)4NPET-|v=Kv*kT`Ed&u-{^fK*#_Pq$c;K-Z2UQ()F0TwM*kuZ
zx#9y-$eC#|04kUjo5qQ(F-CH%;{>>rHN(OM7Y8ys<>Z4JUa$m6(foQh6yuI1!OS`7
zGj%TwNhx@*u_Iy`h}`ff50vBHVE|M}FS7h?+2T)5hopW@9pG(Ww&9OwK&(u{%NLg8
zlRQh{buE4tZH`NHrS}sxdgfgoSS^e4d@Liv-{9V@678bR=+uRofU0p$9pvsRO9Y+}
z=+aBWt~Z)jr~@UeO5;iaPK)l^$YWoePNuF5Xig%#lz~xvCMeI5%<vo^wC4XR)SPRC
zbL*}0|3%`E^t4bHUwfulZdGX2jj1$N6ydDcicxhYUO{kJC-?nw7?E3SJsjMbX$$=q
zkxmx(Ref=WXs6UDiyn$MazqgYWtWL@y6FGs4ns2XRaOH}CBkZM;ngABX9i@qhZ)Km
zAZ{azIlSpyU$9w+WCIHp2_9w|2}t3!haIV~RuHZ<8wlp9b3$At%3s0}dH7-dxLTmg
z&qgxc6W=Dt88>5rRogKBAQ(nuTf-$sLApz8x}gK1|311%jAC&uh0V{H0!l{Z6!wj=
z1(d9pf$kd*ln3O5Sbahfp;0$z6mt_n=3&yKta)%hwH}qXR+G)gOM;vhu1+qW2<7%d
z@o)lfDnhH7*(`p!mcw$Gd+OSRb9LxtY3n%CI$&;-^zKbHwh=;;v0iq^KI=l5bv*ru
z$!crd^5^1<X1%F^7sU07NRzm}RR5SJh$7&qAeLOe_&_T)&3@d5AntFxG!jQ`1WArb
zMANtlge!x$oKh^ZDKsfz?H)@jtllK6>}^GO1WL@<71IRNX81N_81b?Yr_~|b`|0%<
z$Q{%H58R9xq^{%Os+$+FTL*i(qf4|iXvP<EN1QAGuCpQ`m3&=PUXJ8MO|QYdy`^LL
zntYb6pqxFQw8u`O$&5R-cwlS7R0Y%9O1vY&5t$mx@v-RIfb*lP|J=sk;#nhxZ%f#%
zUmb{hMUg5`ZNxVU=ho!jVyF^>+krSK9g3tXc$XmOhCVlL+qN%gG_uOEv1p1WClj&d
z4iK&-FZyup0n;|w3CwBswQ8S_ZQ#y8xiOe(c+p7!r9wS1?pD~R%Osvimex9E;-8g&
z$O^3k%n35V7QY%>?#e&zA}wrDJ`!15K4rp#y%-HT?k~VDiM8c;-$sxNjBW+e=$7s{
zZ8tEph3i7OTo}D4i027}!kTm8-g^Qa4by%(<r=*g2({2#sfQ_?TN6)gG27nt*ph`-
zv91JRR>Bj~=zR#f`06>|iuAW!`?SaG`}%Jg1~C_JC6a7-+ajha>{$RqIgXEM*UULe
zc;+xlepBOh0q_o4TsihCLF8J?VGRz$<9T6p6LgljC7YoUD)%XN2pXi93nJS_Ch7+j
zcJda1oYMqA@@dOiEjy;J_OH)jJZr}374W>q($YNTB^UCsOEa*mE6?3=Xq|4DFnVPC
zvxWb2<4U+9($NtGw8pBf#owrZ#Lgn!ESl5lTwe*H;L^-Eu#K>nPhB|6tb$ekb)jxw
zydmg*nGW`mK-Y0fTh|~@kxwdO;RQA`x}_b^*~v*V%VVdIJll?r;T=UPB+N!;O93u^
zr@W{AAlfC$RzzC)e-R2l<#8e=cS2Bs4z#L{Qkc{mwG71R`5(3`aZHy#;n3S17j`=u
zhdrK=Jfp{dA-NqBb5DkF)8c+xD{Jg3$Q<tBO|Noq5knK74zLFbq6pnwAbqE>86#!=
z5wUr-h~LESPEQrSEy%S*65kW+_d~nKktu^225%(H1?j*JNqy`9K~x(_I`eV;e@<SQ
zvUG9T$Nw%UmyR*}K#>BIyv&SZ_NhS5NZ!ln&E;6Pj>#sfQKyt;2c1*$lssHSZhs4E
zDBfoR-Li-xbh9|e5Qqzmv4_T6TqTGC_v`9wt^JB*=z2O!kj#q$-RVgqc_q;7@w#A_
zv#W%txB@n`KbX@;-5b}xeQ}qd+?mG2e?*a6i*C*4IRL^ep41Jyb`)Kkq_mfjc}k~1
zw-w7TbKuDWiv6P>G3QcMsOw6Lck;UW%m)^TgtAoRZXv~&-Yiw4=imZGeG<tm8c3aS
z>LCT14846XcB#<X&@6G#i7$kxn<qEyBIslIagJ9GCG5%!RLZE4X}$bbU{OkmJ|-Rp
zVjj?tOoyULqG7_ws2)z3v{6c7w{!%AkBKz}6c#v+R0|x*A8v>K9u0lYO2dS?m&rft
zHX_|Dh~CPw&1yTUfFv81l<IRq<}}@f)oRJ30i1GsWzf$AE^rJOWs=A@&JyA3RJw-n
z!W$<Z2jW_{qD_kPM-IiG9bZIi9K`=WJ}6HBp_YuycuHX>qJ1gI+d`ZO8_r&Grn{eY
zViAp3`y^~V^<)sUj>Gq0+*YTX&~A?(L}zBBXHN%M_ml$R$pu~dd@96gB!5ujPla}0
zFFeA~O1}warQ<1sM@}Q?{#@;f6;6jTXAwiN+wP2ZK~A#7d(g4-83oOx&>S0{4NY1d
zQ=(l#GZ@t~pVwwQ7tE#BbJ{4$93{k!*@+yh<?lrV9{tpuv%&cwUP4(<bdYQ7>ZiI_
z2sa;luvm~eC+o)r#phUn$WC>!3qdH*6x4eabam1@aKwK~kOvLc0L+YO=KE174ag;W
zEPoMUmw3QF16IEngieA}`BH2ng5t9TtFXt5bQZj7gep^4oGl2JA|~gS=TcF%4M96f
z_MGlz>07x3JWm8{!rG%<Kpr}hcpp)ZfW(`f`(*-Lf2L=;%pmj0@|Oado9rZ_TcqP~
ztw=P9rydpvN1B@BQ3YMx{IW^>a#;b%2cNtOPP-h)Wn-U*)<#?b0DlHqy)RT4ZqUH3
zG44tTv$Kk8X$mE#v<h`ckku$6UsN)$@@K42V`I|Q5IF;;&QhE!z+ceocsrOq<UPXN
z5Xe?_d35|W1yYrN!me2By5c*jhoA;A7`q6gu!zZHw(E;8y#I|kg_r}ZMQJ0oVLcM6
z%RE~d``kbzm#-0F2XCO?2tb=6^kFmbt|)Sl&xygC2)hAVIIzXdknnoy)Hqf_mtRif
zu$*bT6~uX?=!#A>Hcqz#IVWUutZ5B*0OSRxA&$PY_(Eq>=}5mO$Y1Dcsv6%-1IS%~
z#v64vB!!#Y-yH)A*X9&<a%<dE&}E5JyYDUNlAdTeMYM~IUli`8qV0YVH!!jTM1EDV
z{p|-p$%D(ZIO;*d6qs!eDem|WD3=cXo`w-|g#ahX@lnQ}^E?dV@~*n-s<Em_&r!XV
z)c=ys#&E&r6~`Qmue#!hM}U+o2b+%+cDdMxq}TP}=FE?Ry8ELa(S{c!z^MXV0^XFC
z!tZf)q_Mk-5<rw5E0SZv(Tf$@6TB&u4B3yzTK^)fX{9OFR?ro|4Q*T}(&;l$d9f<~
zD}*6}T&ypeABTWH&1?{xVn0!Qi$~(5yZ9ueYuR10%Eje^%t0znub(c!=~uaCWz@~^
z?;?th^@H)mQ$;fPh+-{kiBAQQ3>zPGM@VM;(*RVammBJP{#zt~FAW-gx&W6$#%EeT
ze-q>(>4)z~V~b~sXj*wCs60^+d?`1^rS*T=4fF3{yi$i)KGDv_)=-eELqj4RjwhZ4
zagB9hRF2i2E3%OimEHh2UyxhYUT{^%yX)~7-Moj+%UXi+ge0k%Z)D<mP#T2UsT6+}
z;Sn~}9XDr+=i=y_FNtzC^n+&BD_U@c7Xhgv3Y7nQiJ*&G8qmgLua^OG<6GV793hT^
zh?^UkM+LY@eaX#Cnsm;371Zq29+;OK`r=w){<NwoFXA--bwF&ymwuhFi_J0aus5K*
ze_~v6L(NaROn>i9{%{SrB1{u<C8i5>W@Z#jUi&RD&kcHOD4r4Rn$Z5z=w9aS;;UAs
zaXMo)7Jdhe65$su{-toP0*BP3Z8PqFV88R*Tla*!+PnNi8?;Ft{MYxK49dT;)BA*-
zG4{T)V3_>_5Z9p(V~Q4OdB=hR-9?$Clj8ePf+&pk5hp8b-Z6>O^;Z25fc)hBIL3*{
zm8D0;ZlZJ1*&*W%xi_Z%E{?S3M}0&b4|8tbV;_<5pQa+gdqt5tzMVUxcRGa2>nTi~
z^OjElDO?F@_spOA+ky1{<SL?aQBi#Fj>83b3X~JV+Cw$VeOAQdVhR!0pa3$ZA*$F7
zLbwV#SRrfR+0@U$Tx8AzG5m`nO0|+6qOQhYg;A__YWxp{UGAa_MZ=fHN0h{i_nyLB
z;>P|m+t>XCxchNvFQbCN%T1CPKcYWiG`6_OEanN(6ghd~8unF@3^vbfCeIMSznI1L
z#@ixX8XkaQQkUiM{NMPKNhMs_>lws3z6EeMun8_cTm_jE1Kr)_q`0=tcVJYr6+Njg
zh2aV;Qo~%G|HAAvu2I;Z_V?uf&-)&%HVWmK`U?Rbz_?q{iAwq~OZ-rerhBz032^!K
zQY8xw^VWHFu+-6l30YrjXKejb@o77E<dqotGk}Kxf=|42MWYbP&-*sfqOhw>*G@aE
zNxu{jbBgYn;y6L>;@ly#EQxm2usGq6-Q6Bjeg$*YIxv{yrLNxqoN*Zq9uByx2w}40
z9qub<m{99YI=^+#2+2Q9opH5-PCL1t6<ar&35>SH2P1DUieRvCfJ0ughbVH<L}zoW
z_i|(%v}4AZ`P(Ju+s=4s7N>+;t4e$z!X?x~kH4Bpvw>(}v5>u`Fl8!DjvZ$&BI;#w
zR0Hmu1B7;G>m{dI&x_2%ZcyTZHwB2tl_|0GoP;So+tX3eBf<=)Jjv7MG(ov>xkQ%#
zyh@bk0we>BHKf|uEYcZBznm9x1h`@_aAaIukHFO<UYg#Rn+WEkv_u&~IFy(N%$c(R
zgMHOFOb`Wa!&dUpKM;07S{Pw`f>Pprb+AVYj!31;@qxhHA^L}gqG!J1%Vbl|hdk^j
z(Cn2_D(n7qAuf|vXZAsOuAU!^=1iv7an1roj0&=&)cL;^<}t#A+h-3;vG9VRp4ez<
zj(3RnNN(<hfXyg+F7mD?1h}Z!_IF43!qBw_K#s(zYE%?a8Yb}OX}f6ifd_25@SKBS
z_CA5MB1&>yF?kUPR}Iy3y#}<$Vg(`z5|y3nk05zyS5V;gzHgLJmrovqc@(1#B)^Zb
z#p1*uqei{lu&W3<AxdF4E<w-*N1Vy|0;9yGfLyn3-3A{gI@gV-nVCG}--1YuZ2)t@
z(nWkKIzCPpox3e3mY#T2fNRdJtbFJ&@6Ni6lR$T+E!GrK8=ie^yBMAuE(>OEu$)Sx
zZ6hIGEZBC(XChrr1{&`6en*S9yD!K86jJ^UW6>2LoC6XxWPPl_e-h?$X-|m|gKV1x
z(slb$()&SSXV}RzJ!`KBnG&Qh4~xjfYimRkB^Fx=#N}y`oObq=ix@q0XU5w!A*79O
zzgT=#!t_w~3W#BM5ara^W)H&Et%f^pBNwEvE9`$chUqeXjS(Q`F{g8sJPr`xy7UdO
zvBH5+Ydj&)ou#E4Z^c#&TxGQ)s?H5|>wc=hq6Z<8U3PT<*QST1EdC<e46Hpd8utlv
zrf8jEStW-pv#tTCvCj=ftHO@JWjGrPiR_-q5hS6#Xe1F}X)<rQPK2}MEjAvcYk)TU
zHREj}(tl%8jAK~Q?w3`);mnqE@limoFw35%0xfB02rcS{XY&gMkSmKge$PcX7rC0-
zZ!{$BB1Nm#7%>*gwdNKHRTq|uU&n#D98L7hm~(tVOQI78i1r+!o2N9rF=ZSl*riCK
z=IEUO;PP|k;eCg`=o94Zbl7H3-4%}r6F-~UV!eq)8d!iciiQMG6WOA~q_rX3O_+$=
z=w!zVF+VsR=#eXptOwQs&5YBVd8G4}pggT0CPCX*d&U{oEl`9U>=Dvk$7(`dzkJt`
z{1_c9n4(B0H;(=@gp0z~uQbT%ObFG$%X4(CN7&WuY~qZrz57uc_;WpX5jTl2i~CzB
z-X<FsA6TJ{o5r;_0;{tb;~zQ^TAYHNb|2ygZtbeE=f+^}hPFG=RqUGrUBt59CGCxE
zHUTjc?5#FA?^<S4P}(iI&_Yi&Mhl=OXfeln3g<36fV|?f{GT3q8@5^TpY|>>T(nCj
z7nn(Syqhp**;4TuL=H!74qEFO7$l>j_N1Wvao42o<n-lvVXi?{&x7E|PzXiq#;g1E
z`oByuT9t3uqKGdeV^*A>gg7H~Mz~9FjXAdj@^Gl49W<B7Jc1F2>iyJeY$ecRq^~z7
zi*m=tXoLH-7q$X$xsV~SPl(oS`~lKY9)3Jnlyl`mBiZ>~ylwFbZ;|NUVUg`YoU!hE
zq@7ufLv{dk;W(VfIU=1i{hzTi-;Mz066X)JU`S}n&R~{j>K;N~b<!5$Y~jv;t5-cD
z%NFEvfi7Aya^=i%wIDZ9x=rSs0y=#c&|2@|LEmcpao6J0?42ui7sNW-Q;yaDQpDHn
z%!SycLey4tqY1kac3)PfmAQd-R(A+?b$Fw&AD2O0aoFx)3^N=nl;Z*sE(MI0r0;bh
z?vOO_U~IKVL2K6${}$~e>2N3=ljIA+T<LZOOe-UF+dY9aLo#^dT@lU&m)NqL{91^=
z;3@zKY&m`ul>1GkDJJa&soJ&o$4v@nURBc5sLSpRLh|V4^~ISYh>F{nxK&}N(u?P@
z&vHCAxuavTefZB+L*b}BR^Asn_mJe_S99BQjeu@#y^Yt=He?l_9h$)1Rm!{!Fc^mj
za34m>b(tvV%MLc46X~RpMq!HDj~3!_6~N4FwBo4qLd_tw@8~t+0tH>Tc3o_o*aq!h
zlT09x(y|7eE;M&oxW+zJ0E}So(G$ZuAk5WCT<q%TFi}vRU%1VQ!$iCJO0;K>POvj|
zISs`7JaZz#g_Q3A3VBd>f#esrKNsT4q8Ak>_CUH&_^#xORVM>D)20&g+HFPCKs+24
zKPc=nCV^#hHOBT9(b0+IrcAVx$0R526$z6@tr5>F=)(2cx?+44#EfpP^rGse&Cr^C
zfO!-Tq6)#h9_vm4%Ok`-hha}gR0YF7ROJU^g#i;BpE_fd2p1H6y>zpCsUT<1>o}7*
zs7@2&>^b<c)zZK`X=(v-&p{7TzxqL-lS7+6nZx{J(8;AkqK;Ux&X5uI4m=Ac2VjQ@
zaRzAQvF<n1V3+L&Xoj;}N1=9_09RNVUdhYkPjyVY2)2nGvB>^KJdXZ7+#Z!<m@qe1
znQMsnUUY8F4jrq172xJz^BAuk@cT4SR%HU%<G=#LLe_3$juDpIN=i3)Z4d&llSKM{
z5pEAGkhI39_Yr117?@FIALBQRa_!k4%5^EV-{%lOmq)J!T_@6v<qU+)VmaOsL`E1#
zHN~$YTp!$3!sPo81;{g48<&Q-D2urc1M_H0#uj#I(S10mt00%+$e{;>)P@h!cq6&p
zEG%E_QVj-bc2)mX{c~ghEv?n49#JHyS96%V&JmJ(Sz5QsG53)`6qqYN&X4S(MrA-d
z%5*pLg8+ZT_~I&{BUU)7NI9v`U>j9O%Cd_^A@S(qv+5ugpu8dU%rQWDmXW(3y*Q8Z
z(y{f3Y1!aK-Q$7GMfRGgcl`YX0Jk4I!<rL_c;!T3H;ztCQ%{D>=_i%eRH2EdfYLtN
zPsXhx%u+_M#&35la%zDgsjW!b7fT3p36PMo#H=iUSp^NW*z7dIq==xo6l<MPeBC62
z3|+xoE6h#F*{a0$=0+iAVsZvcq(^@jDaoy4hO>y|@*_rOht$OwiIs$EK4V9HjKWTa
zCygjqN8BpNy>_6j53z{MX}&xg(8cGHgK@OzIRGAabh);e@!TRIrjkCr7B)f<`7lw(
z2KB%DFg}$s!a0BEZ`X-yTU|bUc0LH)&szO6UH}PK*n5{^B@rZsGECflp~LI};=zk*
zk+`Zp=wgUm%GM5ilU*!;$rq{qB?^*SW2v<(=D!5O?c6YKWE>*WeB$Aj^aL3l?oOdD
zpezZHNImyEq=|QPO7WTy^AGd-F5NJEAt<k`oiX*&A`yf?jeK&!WgyO*^*{b6GEZRA
z7XPkJBZ5H`eR+{UnQ4d_WCbDAB-u5M6H%LEl1FrU!UTa%lNqiruC7O-cJxtA>2qH}
zBsi*fJ@-`DWy0({-p_w^AiuitO8z1@HhH~q)zuL0?Z_OYhhB}{t|`9dMTa<Cq@yI2
z4P~|OwP1N1;jxsFIs|{qLd{=U%7Oe>h`%lOXQQ$et^;!g>3Lcrk<pwi(4Vq}mWP0W
z>p`3a#(?Q5P&OQcLMeQ@ocSyhT<*5Tfj1N>Nb!Y@^3aWrt4qANScDmZ=i#_tWG+}+
zRqiROvHVS7W-@Mjx?+rI$H6LDGTw6Z3v=Olp^lNPW92|SA{Uh~k@5ASF!v*g`6PsW
zQ;_SZ#e-UpycvWR>+PNzyD043Boj$3dW^8lP<bXS$0KzF!&Hvv@vD&B6qrOL^*uC%
zZvl06`>N<++**)wInXWAJ-Rd$AJ?gg<739#{Ga69P4T!Wmq1rT49SMuL0nEYd=ehw
zTJMfJhH`0|RJX5Vy!(`<w5SN9zGzOwe0LV#@Dw4Neb((&?*hwn5XXl2>$D;c$4lIp
z)Ajx>%q8M(Q7_+}csB@rf(KdKVvBnr$U?pvZ@kZcFtD5AIT4;p%H7f-d|rT?lEVmd
z9P*xcU!YU3GLh4etK458jW`xUPIIylH*Gfyv$kz-34)<=4c8SHJOJga*yxXEKlC7g
zv$C$AmeFqn=EkMN(;cpPsEAhLWQPESa^E1({HABggIG(<`42Ev6943@C`{I@NnFm1
z7L_N=ejTq8P+J$dB+(#b96k4N5f7mNqP5QjxWjOL9V<Oj{LhA!J@Y04T-wI|bXtE`
zNM32CbaVVJ$6Aj9xi7+rQdyTJytgUTwdYjJ#f;{>V;=*|=Nh(tC|<bRd??(ha<;9M
zxKUp7pFplc5B43LHj}96=YQ2>@=TND|KlGAav@vV8Z|UO5~3c*;cK2J2(xyyHpS)&
zx}!-D6ubP6mOA!{qWMgUo1f<}j*#N8e}kL7+-phEK!EFJ=IT^`+EYd95`@LCqUoTD
z5$pU1A~(CN$2yE;M#>Z#=YkWSCX)ISPdjqs@LwqB%DWcG3I6m9fH^`hVMpR#xVm5`
z)ZfDfo6)+HAom_tye2fzdET^+NhOjFg&Trri%&HcWdgVCb44cXsPyCreWw!V33f-(
za~cOfUyu!GCZG2LBt0Oz*5PohdG0x(xfB@N<65vBUkS@I6td{`UW9}#^)5On#I<35
z#=xvU?tcl0W|3#}bzUV5Q`y|MMxO{bgRDT?xVXd+PN3N~AcgnX`85Ezs%x8a{8N<o
z$=qZ1mYCxyEh2}RU*`|E5Mz4^+s31Xs24HcF7XCocb#;9v&owU%@AyiLq+>*T2Df$
zrQZT^MX(i3r;VKiDM8*2?o&7~J5sE9Tr{O*d)?9(U)Mj@%CFJ?ogf!n%DkD?u=KFE
z`NI`1m6@7i@Errq*r}(>;I9R^a9T9kobLM`h?|ButT8m#Jnw_Kz`W?51SBH_d1NDI
zLGGQVy%|0Lr495FZ0u7IV<*IwAM%eYg^QUM)6SSzYju16m}0;J?9XD%M}%GFacjor
zqFfDBN$H6@3Gl~KuAyna7vd5kC+a{l5NCc2<Q%J7ZO;{5`zUEftT7!V&qGwS6Jc68
zHWf$}l5N{wBAhdVzJ7EB1vnoMw9a^1kkiHbRFa|(a}v@qi}gP#l4wCu9RDN&%wQBa
zcnX#;=Ki!u7O&;yMz+1j3d;BSwFi<DkEiJmcw=>?**}AF9ax?QO8wk{#Lu4>U=wGN
zM(%?U$9)0le9`-q#Z5UL7UnUBBj)(#OGuhvhzsy7zak7D^h#5_B7#5I631p=6LyZp
zdxEjQK=UQ39<WZ$^o<i}!Yf;h6p?#SHbOo|ka??HO)QD)ajJMDQDe;fEpg0SgjBjR
zURa0|!n`1L{}w^67;ntd`b_-2!YHuzI<tM}FijjC3VWht4#S4a*Ne9cc3I@Pqa06v
z4?ykp!eWg10m6Br<wAEyTJs~2^VG8p+V@4jFf&v8u$6y;blFja;+?yAUx?Ge{TmPB
z(%XIpawpYO!m-dVMH&cqV-b<624iCX`d_%?iN{4|*7WyEAhge~j*K>U{8wRDh#dzu
z=Wo{mB*d)W9|~}3yLc1qfZrf<9rK$ehYE5-ryA<kEnb;nVq%PL;bX=bAt)Z+lVT=?
zYuy4JVQoL+8G*2|QTC~?{Vs;xbI`(D&g2++Rxesan5)U~L`og62+K@ka3S1|muL1j
zC?R7IJqv_eH0|2+Zu4`&<jm`m$fjpCfeSNyt7Rhj_4?;*{Nrg%(nU?+tR5TA4r<<_
zv#A04{2U<8#m~BWb8^Vsz`4K-?hb9^66TNly4f-I#o_Z1%dD=l3%y7r+YWp+JgA^6
zXLX-y>@qJ91()OcIOq=$POFDUhe`b;A^xIa<jCZ#;~YV*1G@{}z~xz!-Z*yw&^+Pd
zW;tFEO<UoUKlWd!NJO`@QV+RX5ZR`C<F<uO$S^p1MYs%DrgtKYx<QC*-7jYxT7zz>
zW2hA(vH}|u#7o_y9LFe3`ZU4BI6*{hF<?)6o#@=gDAY=|{2@VFnv<s4W{VUbq!;m?
z5lw1{yzm|%!aa}ul#~wHRn4*}AbF#y5TzVOks%gatVoHw{IRi%XlI>0ImbbN1oGlD
zaa63ZI25hb(-MseI&KxSLEVyEBFL<$aHdICAg?X~M&Y24YAk7@UwwJRQV>?flOBC<
zGHg$Wqa&Agw2JQiZxLaR%hN90m?nhURB^yQ+c3fu&jus&Ws0i+=J*Xn<tYd&rQypu
zs<uElfe@7&QmTQyv563p9pv(PFo)^IF=aX9tz7ucx<c_4rwb{5{`F5Fa0DNZ_bKT7
zkR-)Nb+W146Pv8azuwYQT~r+KHKD<Dpl~y|y*GXnWfqouOX*fPURbF}S4a8#RxYAw
z=f}A3j008ygUc<rW{z(}<jF@WWh|7ObXEoQxWM><C(0A2uNm&|pqa);KzN$uHiqr-
zfB+|?Wy+6H9U@SD3ER1&>R-^<Eax`~fCbrE?$)cBl!P2k-d!g#Ws5~u=imAvCks$H
zoeOt4-~p-5u|@%4p}^ueLV$-1{dP>u-vDjSmubC<$echrLo7O~0MNYRfOlmb;AWkf
z<WT|CO7@(eD(uvCUmhbyL%XX?>gM6+QZ$SOq0L&c@`bgcoO=hgd~SRZUFPiU#}=Ic
zLgGAn7<(w}#3lQWPejvlTv@cnyb~cvms0@kahWJr1I+`3K2HlE9jV!GvL<0?T4gJ?
zxkxuJ+H|_v&}n2ys2h+T*{UAJHUC<GX1n}nH4Y-y(zU=M!X3HFAFTDvT8jy!L`e9t
zJ=|E7Il+yDRLM(1C<Z5rD(BYkf0qUi3X%fgfom6EX<H4qZx-SrWZ(5%-ILoi;d%a<
z%9emSud+_@F<0npY2qj$;${o)|E^28*3vw@7$XHZg><6}_iq%G3tj4xSH?Gl<bI8Y
zyyuQP{tW1mhhURY1OM(9M(vq*=`^kDL6Na8brxN}NTRcX1_gtx8XE|75}oN5it@MC
zF^RrXHRs=;h}SG8F*nA)g?UY7deBym$1Pf;e?#D0L2jowv<Pq>*bqw(0DqhtfoTDd
zx0u&6Aq||^Y-9d$Z<BMKm~j(mD#U=CI64|c(Rk7cqAQNs6w<uo<fMac>>!AgS*XUv
zc_N%Orcj<h{<ImGnQJ*!?6f(Evuvxf`P^SLNuUqZ9<yv&{E5tgKSis1p+Sz!S`-W}
zaTU7OR)k$~e1bDJ{<JlKbHcrCJ1OraB$o%Zx3tPm6-I9C^gCjn?IFxlc{_m9Gy(6$
z2zK+gr)sL7ZM6fSi%q<~7!plc5VOR`3VW<gVLG~UM@Z+3dUYEMe0SU-%*Dm!1?$Vz
zI{~=3?B;Zf{;?2J#5Y#FzjG0NkST@dmIP$p%Pf-kkL?2FL85sCT}$aGR%6axi*M62
zCcJqJK*ALk)uz7`f1*Ol?(Iqet{2^>J02AcgE}g_<TuN15XJnE?mqBMEs*47$?=H7
zuBsk2O09Iw?qDt#Cap-e@X;jYq<e((9ljCb8-;W25jOMha(jT}o~V=cL!#Xin|k8e
zOmR44LS1D~{-ccA7Q`hYNS5UxF5AmtG(qF8TA0)FX9}|jms?}}-UQtcoV(GDJZT>Q
z&y2|Q`=o5a(dWc{L0v%P5K?5jRtT9U^T#_xIF;%EdUP?aRHQ;Lk`27(vA2X$wpO%)
zk7_JFkwVpA{8xZ`4W@i?RuiNvJYiJ4A<CTR>OVb;gJZ5TAVo-5(6@`o{av@`7=H+H
zAL*05FGJvoN)d_GL}&lonn9dvJKAb@iF8He|Bw3@0Zs%RWM=V7R9lKfI5mqGrwM_d
z%&@=I|3ti~;WMGNND>a?2XuTJ09@dzDMqvtrk2U&V_YT5jgG{#D?S%Z%t?eF>>%v=
zC8L6Bd@979P2SP8fsBPa!Calr?$+2zq!a8xe<^P7g3e5x7|V7SksDcv$B4|GP%npe
z#55r=m}}okY}5l$>nY>MkBD<81DG{>bz2M4AA9}5U_VRuY9jo9b+8kQe(njYKvYQ2
z8g9}@*x4cZm{O`lmyle2mA0ebu}9%M`dWqEEjfE*E}4AE%r)@4G_WDTkVA;m;NGSy
zJ`qh7QuFUMwTPbf#??4d2n><tRd0MRisCVS#2JHxoe-AYeKF^d|HCRYHr<b~3&YBv
zZcdT*Z(B!M4{$z~u8Q^)>U3}iGcgw1AKH_w-E~ybM7)Mz8ZO-ea<6{?Bqiym(>{8T
z!)+|Zk13p&Vh)><o#EE!5J0-$_%&iR1#=sgcok|e=06mKgq!;l$2Jowez1te=OR2c
zb>r7IKSPNA>|x**AIXdWBe;C&f3g0^O^Rz=Juv+bLGBjJ47jJ7=Lmq@ExA<UTDcsz
z{uPW|lW(~Bk1isl=i}u^d?R7*w^%mBcl9WogL;w1bHRU15v9!QRD&X2Qw*6A;>Q<4
zY8`YO>iCNv3y}8+6(%7`oZ-lJg54tWNg96_O&^d#P;WFH2L<Ob&y1D+R>YK7-`HKW
z^OW^79sVF8xhHUi$MhiQ({~6ZjRrdIJN0j*(@sTx6yib1v^UT3P%cd?9*1M@6QJ|3
zz$k(84hbO!k837o&UO;C8$nYPPcFs6g7UZ?6YZkR%qn+*w~BO%+&W0?apB1zxy?FC
zXp4R%z)`q!)Z04?ol+n;@zi$bOd-ye!ILg+_B##8wNF=aez0cI(~F4gmy-jG(wRW+
z6Ygj*dVzDmT=BH|zFM^7AQzx}vMV_LTre2fgsoon?}VK{0=}erceo&r21z{nc;%Vh
z-FblU!s;bWqH4L|TlLlRq3MYn1$j>R0*K6O)ZQr6dl!P3Gsp~C@;AN+z`3G+lk6Km
z5M=h!tKyc6i%(E$;k32KCH`MZnW!y1Dv}&g1Zs~@MC68Qi0?&FIF#LzmD4Yxof%Gw
zD^2X{+Aalhfp9~`6gqgBzhO&Skt2y)M0>0jX9GMulw+35`6GADnq-4DQ>e4(sdOh>
zn6v5-9&k7~m1D6hidfPKZD*mpMi7%s6Wv?E+`llCdQ_x~NN+^0G3iQxJV`Y(rjkC`
z@mB#lWBE|h;JHnZTb$|FZ&l27HJH<GW_DRuq-)1)E{~M?-hXrbk-x>3P8#m<t}thT
z@qW5#nCluKmrW{3z45Jp+{>72*)pMWe_erbj;wIt?Tt@`(et{v44C73!tN&Y30($V
zUq`s>kudxoA;hO&#m5S}&vgtk9UOE6G(DhU;^>KSjtKv$7rW!!{AUYq`@}OM+-4YJ
zVKSPOeV-Sqw2`d;^+v+3BeM-VaRl~)oMH=4cf|TP<>=f}E`GDaQ_(E>Rbg7NZF0<e
zYmP!|H7RV4aBkyGB3h;_%&j2TH;0RK4)|Q@ikC#Yn1vL-8s7-bU6oCHC4Ld$vclK&
zT;kHV6%h7xsTZ6h$ayu48XGT*bSkCRP891NydA)q(FNqlQm+_H7n*oFt|3O=0YS|M
z`s0B+2{OQ>KhFCsqMR5f0{YVjcNIXXT;ft|zi9w2qhA53#^r)spjL!7&xmyEC$CoA
z7uCN(KaH-7Pv^UfNV$M3-vdGc8+9YFwFr-htSHYn+X*z|5vok#l8pnTFwYO;Cq~1)
zP+q`_8*u7=w_q1gHkI#-%mw78AHB*|?*nmmop`0Ou%P??0+j7~J07F@qDNq!0sG4X
z`26V=;>7Xp6MYXsnm=sQ(drqDeg6UEa%V3j)i_m{8C@Znu2|qv0IJ`^SwB7#k*klG
z!8@Ms1Zr+aM$z;bVHcx{tp}a8SBMjwf}{*l{XYRnkcpD##6&njHd);FYvrE#Ux4l?
zL_%8thaKG9Jzk*9vNb_oLY+W`Ud2da^L+L1qm~eur_p!%6C~g%D7gT`zl+cE<DTRX
zvj(S3xZ^8Ff@nU!c(9`)m{Iu#otgFDV8kFlioB*M4=k>9dg8pNpy3Hd4e_?ZR9LP-
zV$uIVm>ozb*)}u?aMxhRfjJX(V>W3M>;#eYHN<lNg*H#Tg-b`zA)zEaAYRV@EJPjw
zN#6*s?t*GPDk&vx_*{X|2YL}jTqndFl)oEX+)$snUjTIWdSE_w743w$<`IYX5<(4I
z5mgT;OjnanNoJK7@?Ug-nBm174cqv+ULx#7o8_W&p*r1K!pqHNqH}4r#lvs?%LUSm
zYZ=ZqQ-zQ%+W4{5D@A(p2ou|g=07fA<2Z%g`y08~+WplcD*dSxd)8@`uu5@*XtP(3
zqDj2WeE;}spl)})za-<*cutsFQ-)X{i*T3X(j^~eq}njk>-=G+a$(HbUT@&uEztEx
z52+Gsy#Y;mkvYaT3K#7?A=*Tf4TdIdal)GrZdu9BxiN@O1esSd-9-qu>02OL6(+?c
z3cB(wsDD2o($z?xBh!vAg^@SLUa{uegq=5c<fYhMwBsj*0kPsrVU&jf*%vQ};A4!D
zqVFBT<W))vfAK^n+0``gSodopU2-~CvMF8ne?Vpp*K+Ol{=Tn3C9iACTk0{WY%e=V
z2F{-E6T^i>)QyF)Xs0}YL`_TmP9Ffd-0au7@M^-Fioz%=kHW`e3j32%8T%G`@ogUh
ziOo7CZ~PI2ha#H+q#ayH6K7Mw9<tgapbPvlfNQ{J72OEwYu_Od_E#pwlG6$1a!6TJ
zTblbn0dsp*^~MLjOoi0Oot9Ot*DrqxSX(2Lk9(nC2z9|xH13u^wb?%_(v$N(mdLq<
z)EFX#-oe;WP@Zghc^Nl8{~RE<wC3}*zbYc=N*hH+0YsouX-!}K>mmYfV@8gModm!z
zG+eu4wr?PErKBV>HTs08H`oK?mfu2>T=JYiMPB*NG`%Cu>DBLx1eht+wH_5hd(k!G
zONGs$o)T{`ql?!Ttv>*&0LU`p2NA^9ZheWL9Ok4N!+s&`dLpClkB>#WMR<J>+b9;M
zJ$@}P?EksX=R9@uZw1CptS)BIftq2>TCYGMP{+k|{yuglV6(fUdk`tXV7w@Zq9HUM
z74L}1MW8dqtTRJ<Okx)m^N6J6@?ddV4mTp?oM#qrmsA`{)6jClYCXkv-Mz729W3=v
zMI5VL(dn}ilM-Og!Rr$u+?RAG&efEjn)_6!GpKaY2ejv1dp0l%GYNNBCn)SbgLNOL
z)*l7<15{(Ud#4?*nH`MENNI<?*Zgxpx`6Uxr(x3@FAH=+%$4${JtqK#>qABpCy7v&
zwDV&*+{m0h^IYK0jfuD|8A{wMEVH^akc{E~HaC!SZ0_b}3>C;~#5{mx!2Sl2sR%M?
zrA~J#tZijS{HS2A9#_4+an!s3wGqdT6agm{{F_i`shhY)=K8k;`Ah8#tMOGm4(|ac
zo0`4nBTi-nvQn&N*zoKp&|5(~eZ4q8l=}rL{^+7C0PQh^`HuXZZn>c2r}-yZMAgd2
z2B{S-rmh&fFlc5;@@Y|x$$}hF+Z$e<QAMs0s(!}Wd9%Xm9onG2Aj0{wB4Vr_*r(Uh
z<k~GsIIb1sG?O+rt2Pm5T%?G|SkkbJT?9EbsU5Oi8^0)s>xsHUdps=KtZAbXja)OK
zY$uQs^|YdBz2ssLW-zBdT-Yu1M*zw$#p#a}CMGQ6YJ71Br%fNld7S7x522e!^!6n{
zJeAkZ8sf$=g4J)6wrD&rk_>PY5tEiI(qO{oqu6;VM`p6Z15}hmI#ZsL>W)(~ZF5%S
zb4%BwqdvX6NJr-_=!x;pGSIo5F(0FhXAdh7<R*C1iYJA*RdI{m$vbqh-m*Xxu?tIz
zsS4MUXln28iED*;v|zJsP3Vo519QeKS3C?Q+KF;G$DW+$$?6b-JE^P1_kzf$4~xSe
z6?UrdK@W+dZ?FQGr>};QV`F!bZYw=#jdiR5mn}Izz<yi^Ervh3_BiWL5Ef8!is2w2
zJ65(pE3F9btfbb<e}l%H)Kt`INPHE=auv|aMHqumR6(RZpbN@xMYwRSs8jW>1nI)z
zY6(^0j|4bH1`ls5#DOapU$kQgahFI^OSjhdW`b$FA^s=AS+f&M65ekFxhHdN(}L)M
zB6h4&B#4!MT(T-8V;(aW#`wI$p-7&gG8m7GaOoKEoB$3R0Z^-5^7$IC2yz@0w@}Uh
zMu7P=AR&{E2sf_==32L)3i0mh#W&=`8RU>Pif`<i3%1=f63lr`LbR<l?<+x0TKXAD
z{qg(;FgKSxMk4#Q#PxpR#T>)bpBi!0C{U`0sGuu06+uT*3653RU6I3LbNpR&t~z%5
z9KZ#@rYbq`yvk@OmjsUux``f)_3Bs~Z%C)-(YwYF&kc~I|LLv6{|O{tj3i>%Si;Qr
zJyYV3;|NkVx?3q$6XD{shfEGq*f{Mo-e30Ne)r&sgv}pqFtqVJUJz`ylr7E^=|XZ@
zuGNr*XH77t%Qb@38?<HKSg6~l2d$4=)`G0HyPk@FaBTq3Q8l4GPxHj$>l83&3f`0R
z(vo2F64}mRtgdJ-2NrsLX}5N}Fr7)UgYSx8*M;&I%(iS~{tWF=P?bVlH4a`61fF#w
zE;wd=!Y*w)JH;h8@L!n9Mu);A+m7e|rZ`fRS;TRk^|Bf-YzX8AVK%8&c@(rWHroi$
z1*F|rCt%58!d#?*q=&vSw0Xk86A}KY0&;{(V`Ds0ryJpR?6V2~yAG1aFt!?nxH#QN
z4^^!pK~$@qdzK3oCI|eYQB>BwOX?|b<0G$TJ8Vi6=gCofT>K{5t*I?3+S{6yF5Jus
zrPV+xKgVni;Ebync*!gLc3S|sbZiK;!0jT0?#F#&W1LX`yNcgLW{qh=Tz{?oNCUP6
zfYp*}SE3@yCBt*^U^Hz7?O};UrF%vjE6FKagS+jzFn&*@ONP%B)?@j_FJfRk8Q*OK
zS<4qWK&-JXfIs2Q=Sna8z83|-#WwCT7Tk_-W;->Z<Ygf)8rPo~f{S(nZC&yfiD!m-
z6oeX`vBUOGkgN0fU<bl3zyJ-?9?R}nq=(rg$0^Y;j!AfI9KBN!oAWG<zU|HcZX&%7
z#tu!Z`y0CyX&^@Ki-ujH%@)KuahXVXOAAhHh&x45JtPrv!e9JP6UKx*yqqcGaj$-d
zD3`P|of1dw2H;fN8aty|r2j|ENs~{i|F3Rxl}Iy6kAuah_4r&&5dXD3h>-hHXLow%
zdCHzZxq8TGI&hcH(dY`H9%Z`6;l&g-UI}#es1lZX;#I|>UacH5cHfJzhai$2yLEY1
zn2X=bf$gk)AoC<79fYjj@`_*=3h`o_)`^ApEg&9jqny<^vs6H(uBj*)pdZu(l#@sF
zuxp)c$FJN0BApwD;w;0%+gTadJmPH79ZS?{M&zitU9?M+jQAMW4-3lEUu$-X{#gag
zY-1BRl>sI9N5h(n#I>u933Ymsq%a4F&V#CUGa=Xgw+PSEC3i~nx+PmcT(?PB{TwPX
zXD8iCJF{544U~?{Do}}&L^us;7SY-ZP6BaN*jqKX^>)Tz1UXrjlNfA=bj^CCRIZ+V
zmoOb*s_}t>Zg7mVIo$jvz}<Sn=-8&CPK}SUC>IgQ9tGgdcHWcg1a+y(-FjRgrU;=3
zW35Zv*HuKoFii?!@79q4d`)mWcXT&#$X51%?eU^0Cz*_UDzRn{2z+SK2ECLClo%^<
z<7ECZ3pz_&ozQFM?FGuMiCg@%2YF4HlSI>qYs#2b1<6(K=8`s+?gMbYVTmcl`l20y
z^@f@b>j!ZC@!KTDam^l$Lg8j_Yb-W}Afsg>Uf>SMVLcza^+1u*APR~MB>^s>+L`se
z8jDRWB4N#f9@hi`E;3s%$y3f0;u4~5S&2(TdknKvXu*ygr#E*9q!2R6y<cJHQ<C|d
zT7Q;7Ff9rw(ebdJC@0z6KdCLIi{@*bk;hU)gk4{Hmpr@5sZNtnVsOZaL4{oadBL!=
z>N5nA!RVw(e!VEyLJmi1>)UbylTc&7N8vo-aH~>^X9c)9)fMy%$m07KDKriCAd8Nr
zgt%I=yJqZ{V?|*qIH|6wFddL(Pq()>))SE5MXV+^Jb(!9<<-I1S(IymV>P5pI!jCw
z=#r&3k>i{L0mN5E$yWyvb}4%+T!k^~uY7O;=^|6jvyC7+tt6E(`Va_bxx%>E`%oxn
zsmu4aK3ToqBhXp0igMfjpa2r!NUaC9O|XWIiDgC93Q%p5cKZr~^4F|arMOXm^YDH}
ztv|!zfNt-`4y*x|5bfsa>rQv0Hwy6tQblF!rTPbDM1wKuh$4X&y<K#MXxPHYj44MJ
z|Dj3P6Bmfi?UWpqGdMesDlmD~Pgmy$3Ukq0n0>JhKe~v7U9XNxl{iya9u7#3D9TL7
z6rW+_jtu1>0p>Acx<!wLa{V<qa&jc);R2o2K&vDs8w#Mh%xLn=(3DA*>mjZ)4nK~6
z&1b!>t*hee{su&a@hKfE98Xx&k{%Y=Rs^}TePg4)y(lvpAwpZ+EZV$~+&8VvKMHdR
z#Rp_VlyHd?09`_lpw$F8mR?4;&GaOJ<_7038M{6sBxiuVCQ_M7ymum)=B6nz_elg@
z9tIwF3a1naAXLV{A!ZRm<d&w`AqUxKopmaxbJq*%gQ%pT33M8uIwr0S<6sf)FsSCn
zc_LkY6x(prvhwKwFidK-t19e5N`s(Du6TA3Mq*rbmEv#_E+*4icYGn*^-Q~1PMAWR
zekXVL)i~lz0Mc(5KOx4SMc4)7VQCRAIyR8pg>eFlway0gq^1P(4rQ`X*P)3c-+3b4
z?8t5P2tllM4v-7hg4I$CiuPp<mQ=0W4`h4APxJ>xXlsH|o##Th2#EAhdY7NK8wI)s
zllpPu8KcfC0G+i<(I>$9w<3CzK>jvC=A10uIaT2<`+P8J&gLzKUr?lB8O#&`PJ`!W
zl84rb7Xs13ZT&GtK`{+`+HVv#kGXx9j^q{>fp}+??s!;TwB6WSu#=RLjs(E>3CiDM
z%RqC?b1?`h5AibD#tNIQrKyO8L^`rMJ0o`YOF&#Ji7)7fWn5wSqcPq-{2~&5w&;ps
zhf5*cdS%u~)<yaEoar)9vTR|OI;gNY&ZsX}()HqH0&9J89DQ=?<^G}xvDc8o{sN&<
zQb|P@R-nI7#+>%AC}JTH<&dD=`PTv+i^D8e>{nk2;POpDlA3$hf>#&ONQ=bA3os*a
zd#1DNF+!*U4uRrXh0W%Yo|a~HS?C)7tsT#Z9G2I(uN8IyP>^HVneo~JkY^iiRs}dQ
zW-zWQcs@dq1{~J&agm7JuVvEuK>icmbmk8cZXT{dyO5%^MD@A?(`g9{mvX#R$8=?m
zeiwSE!pv6gn52!($BSGKSTqD)B_|5VbxpTuy|F<(4rV6Yc0Ey#!<g#R!z#<)P<+Nj
zhuJC)P6U~!vgBrX>~$lEMxw3=UJ>D(QCCN0X78H-TyY++ug15c-B&T2jL|nkdVEV}
zS$}N%ulmFBu;-3P>m(O(@!Bo?o4X&<JRO&L=t`h_Ma_VOX87Z+pt<eX!{9AcK<?ah
zohA-|?9iHo`a}W5YX{le{Sm^c9YV=coNyb2D<jL?I7PGvFlkhJWl~2S&lJoOWz~=K
zMX4W1%YUWYA@bPtOgTzO9v6B!yiYQh=kN44Rh+TJ@Vkl_>K&|1Fm6|Yt_7#Mw5rlW
z%0kVx9ttw+G)UJSAL4A;Ce{HgZ&;tU6TpDrX{*>@VTz5zUj)h*iE`bvuFFA@?!&Lo
zA7CaBH!JMQurKC7AD;>#LuL#gf?DndaP1I`(QDN`@3{xirNq5ujJ($$u^DJYP=Tmh
zfZ04K&ChM`D<X83(3*=r0nT_*e@7cD!2Uwq!ISDByomdYR60w8rFdR6?ZA6>JWBEa
zM6DArym+99PH>=5^Q(&PS$jMvfNyxM2D{x%Xwi@FGl5+-E%adV9dbT)z7+v{$IT17
z8xbxod1GW9FV+!eXnB_(dWbl*Ln{THrm#za4>s;^|C9gZedL(%52xHL=Mfu;@WjEc
zlB-Xcd#W(kOjk;HyU{-TJfZI1+?poSOlH!KcqAu>ZDwyQ_b8EQ7w$T>*MCT4(RNcY
zd1Pbxvp{Ob{&;lk@EC+k3fJl0o&RzQH09Wle~Q+}i~mv7B8PqfHMY}h?JTtI2~aZ#
zch_=TaEuV~U2F0qBIuX4vHCFo#ja;$9QP!Dx!JR7Ykd0@kUNiDG{nfKp>vz!&a4tY
z3dj*;;6Cerq0L0ipXs%(#f0Ua&88D(EF;8UA>J4r>poM&V1pxPQ5!r9ViQq&1!ZFL
znRBZCsSPDsL<-$>p>v)C_AqGDeYZ%uKxwEgR(#%JULlV03cHTIcw18&Z}9>U>}f_@
zIPQKC!e!+l6e-@_C&Xn%H>w;@h&Brlj<Vl=Q-FH|H&#i^oBIEc!krcOxH6WB1z#%C
zLW6!>oG;o@Bn#4I^QD5A!&#4HTiezs;kRTK7rx9Nt{@VNcv__U6I=Hr*}_2R6~H`&
z`nl*Nhu>MS<7%(K5Mi=<L!j$5Nxf~!SE1dT@b-(xz9jP+5bD};QO#OA{xtx&g+~x(
z$Tv(NXKsxy5oVp{d>ls3_GW>g0g7Q(>?*{`^>k}XzncKMBYnK3Cw>-1;+<`b+y&o)
zaDkZZcw&g5SB{m0(y(~ijopiIQ+%T^pYqgCU(EClgn6aMO8y|Sus6;IxT%dc(?9O6
zI2hF1oILa?%=KWeEd!rpg}9lLyBAvHEI}^!#4)k>yQcJnKzl4FiY@EdHDWIX{bf~4
zINfc3LAfz{xwfKTRO8s3KqCUvmKggUQF32q@7*RccO;Zju*He}1(}bX7{vF*bpmRm
z0=GfTI_kV@y<fybcMMs@h5}q+8r_BzogaX~(rRfS9{RBOABi$b(0KWq^CK`=i{eOI
z%rBzWWvfXEX0#v|S8qveDAHx)01L17sDD!GU~@>74*A$grg`pt5oUM?txPBTvW^U~
z?z0t|Z+ekvD>h)LX$tU1lv7GcPu8PIJ$XcYD}Zt%)$NVhK82u?4ZLeyQP@3efM;>>
zW~j|y^JkzmX|k-2k3>0rDI~H1?)@Bu0izoqNl(reP4YOk?v5`-xY9bxlwzqb9h;6l
zrLC<kb`hN?#<abmryVOSw_=GUA#SMy$rEKfA;6ubOJ^5`5l51*ilio?;f(|EO6>A=
z5mCo-UR>iSlfYakH01D<uUIfaPU-ekI2WjusZWd4*54Eu`e9n@s<H96U`_;`CRY0k
z1?16SU+=N|4?v{LWnFYD>~!UYhif_PHGc&2T!SdRFU}Y3DHBJW3=9@DG&p0zPvEdr
zlX`qD!Wk!P>X`Xw|I36}iN!>?6ll9}i0g}K`G0opJh&jjS@NO_F7oT~H5X~1KbPYt
zt*W;~WOiUIolbRE{aWO~H&mn=(*#k|QLLBO{YE&m7gd*z7(T;VwSm9`ezD|?kp6!X
zCjVUN2=M-#Lmsm``<2v{{x~y#xU==(_Mc^-NeG;vCy$H|M0m29FfOLd0_Bo2@=INM
zE#{P2!7_u{_eRrf&`waoC^oA%2%+GO+)J%KM-dyjel_+I?OulaJ$k3wZ1y<;&EP&x
z0NN(cDJZu{xx_Wah60>zwUtfX-lCn5{MMor#B6roTtz}HSm@|*al<@CJTBLhO&01B
z0zHlBG3r(3g(Ph@1j#~Y9no-vD)+_SBB)I(b+0JwQlVN-FF)W904}+l%<E0%;|1l`
zmFB%v#qJTNPRrW0@qEQ6Jg1q8J#u~^*R5Nww2lx>{%H)Xx_}8DZ|sZXL@?iXw8V1?
z=7UdmI+z6Ik1Yu9#F(3V<JN_sT||0T60ChH$dgj-<#_t&a#4KPSi+0eXQEXKhE^Q4
z2!w0Sc9KqhhX7THCE`B{o1JJr#wLqGQpm23*ljU_q>(!GaUyb0L1&7!Xv05(cqvJC
z8F}Bew_vA77l~s<I&a<K@KX2U#s9r%|NOg7;#MqXU4nmIVWdZ#ys(NE<j%s?L|%&)
zToTa5!8e(1h?fvTorc(8x92eaqT;Egz}=*^18OY4bn$g#cUnr%T?WMYqVvyOS&Bym
zxh=>7qn}~Wu2?(W=~t1pxv8(Ms-vBz`^}aGryP3r`0N~>gx%}O%YnPyu%U%DUHI3J
zS(gWO<q$1!pQDj5n?M%^!AqG(IipLEYthjyp)03?=Y+X{Qm=`3R)jXIIcs1RGxJJC
zIv6M<rR>$}5C@mkp)MAb>%~~TQY0OqwUq};2-ao?HtB;gaa9m!Iawt=R<yJ1X&V|U
z#oYp&m?r8N4(aM8byCf!v0fc)Bh?meC3@q0f&P$tkhoD~PB=XfxYP&$^CXQt*0Mtc
zIS#JZ5U5TQU?wReM0Vvkay2lj(>oM@S1^k`(6dMz=d|5^UVk`ke48e*+ZL+>Q5@Ng
zaFRPDlTrs^jQ=&0+KqA|yhx-+g3i)3S=uy?TLaj|kqQk`r@=^=v*K!6rySlyAsWN1
z)CK>35#c%|-&xT)vH+&E$?xw&0+{3x5z_155JfrlEENxmFVFz#PR5xkIen!|E-I8R
z*~xf2Phn?}+^@x)qYFS9fsL`KfLfa&9ZJ2rQ6Pmx9lte}9s}WuHOc{^&b0dpbIx3?
zu(L^Gsu&CE+;nst5sl-VPGgDAKTQPX;_1CuWP<<D<~ptx;m$xaMgK%d_qrkF-?%2U
zONf+2s}N_VBL%upXfop>k!|gJYk`pg9@(S3Heu4N4kF6wj82hGQx2r#cG0j^PH8zk
ztOKQ~Py2;+MYuGbO*l}^|52fd@sSA60u%+)&o&u-m1FjGom`U1#BHKTA!|hcvK}PG
z(lh2G)+g)^k+qDYOOTnSMFce!0p=~m)7u?V?0Z9~D=bwi+PfU*ZUE*H%i^VN+0Gk+
zP?K^u#C{^g4^FsmC=3IV0xZki!W%(56ZYQidP{&C(T#+<EO^#rExED|kOOTsY~y;2
zN)u{(69srQNcx2HSUu>1qY+?pk!#Jkq5c_%A*|0=*o6PZaxGAsig4>p()HnGqWy8h
z__6VZNSBJ^if#)wb$mULqYiY7Xex@k&scaf!g;A6G78E=bp*u*y@C7W=EU*NsdFHn
z5=BvvoVCUuwt&bpG&UA`kQl3{Egd7hCHt)i7mnBaVwtTVonLw{wkzHf;+8^!)YmE}
zqHAwc#F923b=X~qWAOqHcTr3<cM5YwrO(+FGi?Vz(R2?R{UTg6`Wf#o>MnJ^Iw&6n
z*uc>6*9wFmm~6!o+ZSJP_aP5)RA>`{W_Y?f+^bGEw{ds=>F++0Nlu#JPT8UO2r<R@
zI8(IyGE)K);=!0<C!mxu2R?-<dpdDmDZ&{JN=AFX=vuGA<qgAE!{d3uxz})JY}~o{
ziX&w;mfNNHir(0T%i**E-AAws!otVXMZIyXAa`%d#tVpPnEoo%ZN(Hlh5heYyB447
zaaWnXto9cmmtm;2KPn>KP8xQDa^3!xF!~Rd<7_PF+6{_Iv-gcP6sFQry^1|W<bI3H
zpgaB{z)9<(YscLo&6H|)b15Dd?c!yBLwD>^#6TvCd{s0FU|m4F#iM&dk~77Knf4-F
zYYqG?bE~}X?;-4k;QQFUiB%imCDdkS+Q&&$@Mp1%!g*LAoED|3$V&WD5yc#gheh3;
zj@j2SkeODaEW#~XdnZTxqiKR&M|nMrjT#GD?2Z|l3L429-U38ZC9IHRw8E+_M>%$6
z>xeXabbRE#>l-1iGx{ssZsJ6t4CMSeWze>}Xws7k*vN{*dZpxOg|$1!jbt-Hr;=QW
zlgc%Ms5`rX!5Gm3K{2!?r^|G=nq)VCJS<KX>92eGSsj@}&Jg6zhr$Ktb5x#qtVgIz
zCfQ`n-d4~pQj%G05;S=-9k;~iBFt|zyV~WX2KTD%fVuIw3F?k_1ehZgbPL&VJlFw3
zS+k+u%Oa^g-lAi4m%|-&obwbW8g&>bO^t6v)>=2&dN70z>n=dHMfo+>-vs7y#Ziu4
zti^ba9#E>LLzo(UFVT4(OKR>13h*G}bZg7?V}-f5F}XEPb`s($%^k0Zb_r#Sc~}+F
zCCuLH=yEIf6`&6D^a-6}1-c*BPt%ncBRI1PRXVsiK?wO{XosEf38LUW2hUQRmjA-t
z1FP(>qF^E}`D2BC!Y&rBp5nwQkWLk+x{{U5Hvr=7>9*SVqgUDzy9qV}x^TyqmdJ#e
zm#myUSQzT9?}U<@yo7Y?D+<LHrW_kig)mcbNs7{7y8!b<yIbT>#9C~y$VjJzR+(lF
z2{WhrTjM=Z?yy$$WfB-FK0=JeO&{yicEU2plY2sK+m5YAWHe|iHWXI~BXuufZ`VIl
zrMfZv=Bd8@aug1^mDp{6C>H>shHi!9F+r*{{wUv2*!4umA9q5Z34rHq&FRGdQ~h&o
zuq5SQ;lLurq!vY;=ROF?4A)jEZJ}-vmivHj_)3Qqv2Z@&>$FpZxp?VTm^;Yx1?B$C
z>c!FH&?1hkv2^lyUXU}SfBBy1iNnAc7d7LIiw-a1Fr<8dd{r3i<APY@UX+W-5}@tq
zDMu9F(4Fv3hpPF=Bf;Er(mKd!hYNcMrMHOhaju9w!r4<^BN7I43l}#ZMc5V8*-D4q
zZw0xrxzNIXL49_cqXAtb+<|k@rm?3V1LkBKMvjTU9b0_Iez>WUIB~5&YAx$a%?4AC
zgLZyQ2D;*m2Lw53nE|IsW7zTjj2QqQtvIGS0b1h|e_MDbIuVM>C4=SzM5Nj@#_<Zf
zM4cEk-!9UbHL68vnF%KW)%F^!Ii#*)&3`hmS&jA<m%=ehNNsY%d4RpWY+pyK{X7(i
z9Eh(QPAR_53iKxlan8vP4l<JkPX(%RYgAN3QU^V?7l&m6jqLU~^fdl4KeUO2`@EPa
z%pW&$O*;SSkWQW_b9p1R65VG2!69~iRCDn&p<GQQ3+T72R#*lIhF4`4?4K0&64cXy
z7x^vEf_8;5;bi}t+Uq2tER9o8nNu*&ZycMLUhg}*{z+SBvR{Q8i=6{XCb%DrNeZje
zOF@LK|MjAs2FsW_*5c<Dhz?Bjhw(zFiA<`nz7^%(s8be~Asp`3|9kOKH~V}>#-2i8
zr>4{Rwf;9=hNNQp&-3^@4@jqi%b4?vL>js0V~?@F5TEF1N6tKIT~^4OFDSm2@lVof
zSouOQPXmlr%Cw3A^RrtL(|tu#4Z85u{&-VV9;My=O|41ICqBOjlu{$hrUfsC$bGPv
z`JgE-76K<nA>h4AVJFXqdu$wfiT|5K`5sr;|DisQlEI|~&3wnZK%(<FsB|=oUoQ$G
zX)ZR<0dlwjCmbUcc1IhJx^J^c^R1CNO0K8kUSTe04~DFa$CfKWoIdLR%&%M>3UZOq
zI%Ye0RT0Gw8msYVLC#7dtZPJ?p~;vkZm)lm9vR^N=FO{%Pf!kxm9K%U^$m9H3~`h}
z1X45lW1M|$5ua0Sd@Pc`P|l9w*AaFtTHF&$Lim>{CazJ~#b@R!#p|MLV^UKwi{hKY
zXgnU<;&nn1t_()r@wrGhx8@zh_3G>MUk~ak%J(|AF9OK28)wBgDC`1p<D`ex-w|TA
zvWhTU;)+~Qp1%Ahb-EEg@&@4Cy=7a}hO2{vZ!9oBoWiQG<V|4Yhd<?5sTSszv`1kV
zjK_M=9lkLC$+cPhPhskb&|_R|bTfp@BwuM79jn~}<hsfyBh^=3w->4bsk2GTtq|nf
z$c@v{3TrA)7DIRCf6A2NGlk8=Dgv!}Zi6&Kq=8ex!bXT&r3>4j_><_oQc0JI9?E5u
zK=LZFC;lXd)k1A|fV<=zQ1SGS@wXt#IG`ttuFQlU3yOzDIH#U|WU&2rI_98s=I#^W
zl*(-_X-oIEAeV$in7yX%ln<K*XvV;B1~iH=yz3;COv~EYj=#qQ(%5obA|ek61g`%P
z=`ygS_r!#Ip>vOAR8#&_h4=$0!Mrchtteq-SG;mRfSQ|Z=X;?3Cx$+KafkpHnGJJ(
zdH0M50cjkZYjHq)QIvaI(wEGp=3fbBO<18JW_yUB%h;jkAy@oILF1I4dtK2kVZ+FA
z@v=ylkV%L}d8G~@GlsYC2%yNEC1Rb237gM4C`f~I@ki<jBnPtJ34ohu)%L~gk3yJ3
z7*VJfP7{(VSwSvLtZIBMkUV<OBYf*I!s?A|nmS|Ge?sNyuF`=d9r@Fb!d%5Zy#_GL
z<Ng{A)>uVh|A*FLZ;Z?TVaFG9Kf!;L4AYQOw1`kkl%~YwTA1}Po&?WLhEp8|o`5{Q
z&=k?7^S=bSRwb#5Rb$3~J2j4uv4g^7gZKPu?41d2(7IztgufeQozq+ENYd`F#_d9=
z{v<qAw8S%_T%9hN3@`i?v<uUM@JtN7Ly$Q{;pFZPpRmGQ3G7msW}X*N>zCDN`41$O
zZ@^0JJca#zm6vijsQ1PF!kjHznQAmV4V}~D@r~BlUw|XvWN>6WE7~<fMk)=!UxbjO
zZUQOMjL#Gik~+a)OsFH+rAt?DPa*W;@=!eWyFdFI^%~Ni&+=!US(pR891jR|w!9gt
z!~MlV@=U;@uGYF+P%i6)hEegdXjfQvxTynf_gwKA`%-Ba#NVC=a-F-lokcse4q!Kf
zR^rw#6dyIUqU^%{;4ML!57=-sBcA;t2&`uB7LUK|aKpH<@ruGOs4QBT@Z!~1fLui!
zsWBLJYc=DmfG!4ZSdp@=E`Sb#??@y+&7#!VO1Qq9UXRNJL+@YYwc_LEq(3xFfO)~*
zjTe>bH1mB|+#?#^&<U_ugrPLu<k!Kd0ng&a3ksWe=vOpR#;(}s4KRw(t7&<WHz8bH
z+#l1ZZwYXYat0rNe#^0YIM>JeBAgsnfgBc5MVKtiEJR`|LEW|WSbF1v6H#w$`8Kig
zsFX8l65Hq<r`#>MYFR|?@c1e@_<u#B{XNMO+)F~7wH}sCV$glx1@rR2P~t>Q2kMQV
zg;GCVj-!fWf-A0;I7Ebo5gvkH&0%&5amWYY&Rk-_n?9_QSX4hDvNo8|D2@X@0&u&l
z)%6;9Uz}YBd%AnFg!C4H{<JsQs=Sbanm%K^E5LkVuG0Dz(*;rfPIZy3KZfvz8qp9J
zY}$?@IXP9hxt=|>UM+-%OAh1MF<h7FBt~aHIchrpQ~46xxcwC-i-w7#M#j#cm_YY}
zJ=CX8rFIm#^)r9fQ*MqYzaZ?|=y_D$LH!cIj1aRq43f{00$sBHp5)5$8X;ycvdnlv
zBm;uAfibq(SN>D$6HRolNOwPO!dPR!hIXF#;pbBH6#<UGKB+h6`39P5$pP?M3VXHX
zI+$Y{yKU0kNw7<#M{g-(laSml+{;v>@4Euv7{gAY|NA1T+Lnw{Ph26`y-<(XrYrV0
zh2@zp+k=*3rXK*^uE}<Q{yyiAMLZf&*NY6+($9b{n*?<X&qsvhJrDbFR9<8){O#{i
z4sji6ao*z>;!=4GuHrm}&E;MW#QXgUnR*eoIQNV2e>}z%kL3T-6!@eF^8x`89=$&n
zK>J9V*&5&1|HdOOGK)QCSi4p}-sphIoZawbBQ)1T3rabj6+*h6cbIDzpAn$e)w`SY
zHui>s{FN<kW2br~?r@S`(|0ow$vp2VC9P-7LuPj3l#xyKgc0#g9pL_r_2;!&h~X+X
zAqM+IWa{x_Cd7)f7Aa!lWJ%vrvpIerE}-Z$TZuGR(y1t}&Hs}lo70$Qc1MTo#^2A5
zb&|KL<5H2D^V5TQtIq+I=Zgw@w;a(qy9vylGPymD2ZT^WJ?0&6ig2aN6%<C8*<#$B
z1txFtsq;+`=42$rpyWRY@c=<Ls2bgKL3>azZKmV6MBAmgf!!LVuEFG${Z2uS$?V>*
zH)DE8;6H-RNohenEz-5g4rCcn!{z~WF<7S<*>Q&uf2Ha1ZIS*82l33{S`?b+1*H2*
zjjmf^7rWHPi2PUnuQH`OrYr2m=t|Z>@BRTG7v3+u#N_#$7IOQzMq$S<S@<Qu+)OvN
z;?uJ8gTZdhk-B1|OmGc`8qI>m=X9>|BVr{1{v0;~99=2<F~VHG9#lFz7J@Vb(&H{@
zn+kCRo_LK}7lx#+Wmq^$VW-E-f}Qx(sl@ui%pgf=7(+;y9v0}i0Lx9DZcnd93=b#O
z&c*x-In0$p?7s-O2X1vr93jdjVrwBEn@cSUQtMN=M-UgLELOzov!Lf{A&vz%G#GID
z^hdB<jvg&R9K|ZJ{NkWliKm~*f$o)D7aS?vP030-GQJa?OQJ1GH6|@lAbn^GGZP$C
zN4SdVg~8<|iQ_h|4&ampXw#*D@<7r=UeOyR=L>XslbBB4sApQbz}UptrgBwbE^RuB
zSzug^&6WXoXBlFPh4Q`tCs--7C6o;9WnnH11`#afE%Aq81=P-+lAx6YIVre7ukR7y
znwL5-FR!5B1VT6H=6q99*#9Cg)Q<9$<v`3iSjj9PnjLm~N8GqPLDwChaojsA!<kk9
z^oXU?NGm(-j!*s*u*=cM9YHCk3!ty#TBQ`*t!P35l=!=dTn4_yrsnpQK%6F`Z!2A}
zop?sDBeq~`#VKr&l|jrWIa*GV*Pn!$Pde(VjI*p#M3a&+`aL{$EUdPx!I4vUmoC1l
z@qlo+(~h3OiwaX_Bo|%rlL!x{QX`K#tqN&AX+nqz!=YVaY5sM`T>^3&$`(x$-D$$i
zK{iq)G=%6<!$uS-p~Kl8Yl&8_>RMAXp<Qh3x*Gpb1ihO;YLi7$RAe}DjKZ!p;_&46
z?7keEZ4DzlJ{Rc%%A;pj3|}39vD&EDmSbB{P5@yKVwkCMtsoD%)<K?zu<2Do9Wi<0
zq9A`*gNSYz`oI*#9v2ItXb9%xq>)a!%6&WuOZ}oe$`DHzhJWeEaQbK<vx^H01~K=v
ze;!jrlfzbgBN%cth9lY!Y=%=b7={k*xL9Uvkq)y25^#po1;X4`&EoKp<DlL48p8v0
z!z9MuZ#=l`!HGe4cTWmQJ%DSj`6fW*-lkoglnl2Omgiv99WJa#;W$?5#jiokIguz@
ziRI9X2cV$fOe6m+v}W<~BwUCbEYh5nkV8v!`&vM50!BhBRw8Qn8-yxFv|f2Ma&0I#
zEZT7FiAM==I_aHlg#6pAQ=}tho$PM$JfZ#)ePCn-cdZMMYbf6Z8a#*mxk!zzAcqqy
zNd=L=E?RLxxgM0ci*E>|#-de2%Kw|fPLLyO74O(xvBdgd=CY2l2w_KW00Ng$6s*Kp
z5&oohF&rNXBSX&KB>GjRH~|p9x*`ANMGhwZvrfwLa@~#iH}e#oL6oB62tlrOQ-347
z!g~d{0^GWG#Zem<G?$&FxVTPtmCF5bTb*WkVhg!Rkq0AMH~VpgAPUS43zu<kh@yUI
z)pf;Vn?k@I$@srj*rltsv&*J!+cpC-cW^MTTflLfgA|Q|uWDxLi-oxks4Osvek8y%
z3;m1<LR*T<w<y4x{Tiw+0-4ZT+oNeqg6iFxhQAQuY;*&f5?F03FqcoxEZ7so4uW77
zr`hC+Xizk1pgAx#CT|VlVxcTdQyn3|jngo4bo{6OKeG}F9MAn8U`>q0x8Vcl*euts
z2xFGp);X{=@=US_=P<zCWHoxX1Mpau<sKI5@}SZZmkOs`?UK;GBSO;={J(#D!ftI`
zj&Um~+9^^bW}Sm}05BssVzMZ(^qem&GeXX|DAqkY77(|0>AvJqL9{y48hfGlMUiPk
z!?;*&C&F$Xj@a4YcsGI0nO!;~=dU{hxa~F0IP9G+#NXLaLsffGAlZybPE5+XKv7Yx
zyzw^?PLll_%iK!>Jb_4?jG0|$$ftG%H4oB5q@UL5PM%YUd;Y@zUVwEMidd1ucyFA!
z8*uKf60^`<J9Y;mcP{99<BUBZoI9duwz1RdKoiE;INaZp7_JZ6;^=W|eC@VZJz1u`
zR!(>cay&gb!$iYGmp~Yb=r5-08<&tV$99kgpg2DWcE#jVCWh?|4Xes&r@4tJm!D%q
z61MIj$c?S_KV8I37e+peff&0FVHX5b5hiV>*eDBhJ0WP%-f^XUL6lFLGM8#BJ}T)o
z3D(Khct)Uel8c07Q2SdwB2M_Ehs$iZY!eallENwSRgo?<ZgaF{h;Pe4<c|n68<xya
z0inn(_|P7%uz5h0`eTG>Cy-tJ#(m92lxB&?e-iCl*50?lS8fZSyAI<jY39*_r?-N-
z?owo>`HpS_$xR{0JuNKpwAMp{|BtD&4wI|M+Bhyjg5&_f7S}*B@gTd{q6;kU(mmZH
znVIg%bWbJ|ba8iCZ1Lb5U~!kl-CY(4`yq?VV$1h?>%D!O=jlJ{yj{2I)G0r8>J-^B
zZ<X*~qVedO$v{Yn;RS{8EK%kYFR1vV=z`9e(6#t9bZ*uSNc)mE;`p-&=iOImE1+Pl
z3CUVqGCX}!fQ5%CL>$(We<3q`+UONREDb#lphk0-ASc2wi6-Hx=2sj{tKI%0q%k;R
zP^5$A9V>ZCw45|E6$CN1wlkS4NR5nc85K7+WAX~G)&=pTAPX(|WI;UY+OG(92{p(d
zvkj+!m^q!@jfme1u-G{A$1PhQG$q!ej)adE2@93B_(DN{%E}l`_plm(%g#;G<eA_*
zL6%pM{nQ$DlO62jk<4x)omZdCc&CVVicI|sBp(UT;G;{!=T9|3-{++i5ti8WPOKM-
zI3T!fAjb+!-l<KjIt|P`p`OMjo@LQs4oxm=`&WzG1<`Wk=!sV{MLF%Z?m~MqG@a)F
zFgRfmI4-7%FyAF<<z|R>)8gh3>sVF6-<w}di2rMT!Cl|(_)fI5s1>*&SHTo?(F4I;
zIT$7BY=5*6GR1NVlLrx5%hC)?<vrhoCYRe-?F`0M2NBC6WX;J4J?dZp8cG*Rn@)u(
zW^3Od3zZv0Ilo@+fg_n$gk-hF<>&WAI>(+)COWnuTO9&qzI9`S3pe#x@lY^w;4!#3
zM`35rd_c?oM1UJ2^Ka^_YKZIp1vqPE7=g!gheKx#6O(Ir={N$!nalANmelc?AdeX7
zz#K&uqvhAlh)l2;a`1?#Ag9I|sHX9WM}m-_{84FRe`ZEAmG)|Vv;9&0-(XX>0KHrL
zS#Tx+ZBWnyEpRlDWmIBCzEG1a_P54gj^STw48>9{4i}LL1Jm%qSngN=w-X$Gvj&&h
z$wmVCgeU&GqD_Q3FQa*+mJSf&M0j-#1|G}ev-v2*xa0WO^-)A^p5F6*UYMC?_O0mo
ztrrBTAtA>&`gjPJfHi--C6fQ7_K7u5fN0!-L2<~mXA8}f^@lTJW5ihwLOoTOGs1d+
zfvgl43Ua-+mAU0U;iR0VQ&eJ8(X^EohOG-uCTwXit}_A11bOdM04)G+;Z->wxKxO9
zK|78^61D(>T!Ib^6!$+ZkC7fyq8!%=%6h02UN4AF`A&>mPbZvBJPqfcgK_H_p!5w|
z#3~cw>x{A;?TcsfU&bk`Uz>w1dKO@#3s-pJg*j(sVdfs4ic~GfG-Gkzj~#%{Ergl5
zsq*gobCc#4V>uQ(hyTsW%41^BbD>~eJ}+k|?1F3e{)|YoKz*pCmtu$Wz}#MHNpo=7
z`8iNePnq8*fC{fliT<JpOPAxyw0r)n`GJ}-@6umD9Oue*IvyA4LUQ6OYoDitxIhD)
zJahDdXbPZ*k=tYF-|~OuN+mWHZLZOr>p2|W5xo%160Wi5(OtB1%q@_T_YGF!uL@_Y
zfXt!QbOzazE&^8L#l|8o7GXB%Tqu-8=PL}3ebbV1tBXOi4u#*xdh8`YontCJ<(7<~
zU>y95$onf3rdsF+0$u#<O6fJ1fKXi9>l}M&{t?3?D|e}VP7!F<l2i9zL{qe0q}P5K
zVT+eLH1U;4bB%@}`9Ma9`&<s1ag9EOH?9RZMJWa2^cm1*iw2KJuDC;p<;)^UrxCUI
zpo!rb3&#`8G8<k&Ja}nkIJ;J1H;10WwDk0K`AQ(?tVLvZyez<R5HsrC{&?dmAQwSq
z=5t*GiA0l1bC?K{A4JsaDeO<#P05o~oF*vaQWnB`zv=pG^JtW<Dn}ka2_i{W)%7^^
zIta5gm==Rq3UDIa1JE__zWCtpVAKlN*w{|}Ta?AanoGv<UkOnaaCY+iOv2{BgUq(O
z9+J$Y8;^<zf5|O66nru74InN>!}Hn3Hv*BE4o)st*ojpr;8!C3Jv%c^Mw|QtB<ldv
zlG3FXJKhZDrskEFX32d8`ctVgIN<-A5H}89ou<E1)wP)K7EsFr(?jmT>><F712g5`
zxLq`|pmOM*!)}E#|M<n031u812>#iP#k&d<nUS#47pva}MQLhWP^3?tF47EO4BmL<
z;cKC0fDsIvp8f6sFauq30oZ#dbXLoLdXJ$|m`jKAQk`m4?gGg=EbDU_e$XZFX`<3K
zQoEAPeRt<6@-m>!`WC+@PqDkePNq8^65@Q>+F<;J_=n#MM(tv*TZs!qQ1E(ZSKN|?
z%iu@fmnVgfElpip-49}^X^7H1w7no#TGFSbBl}$hl43h1f`>G}85q>#bnUU^19>vN
z%=og%*+39qpl*w6vvAA!QSs4(dGfTJPQ9Ws7<2s-(5(vtli0aQ)_5H84w2?49RV<c
z)MM9&0JHYWU@LRZB_0OKS{9ojDt3$zN75x#74Ks~S#{*P_Za>Ns2Qjg+G2f?{w}FU
z7{R&(c@e>5M%7|x<7J1p^#?5r&sfaSOFasebqv|ta;Kw-*oHR#DFo59jGhorDcopN
z%;_JAcHN?h)m<Aj;E#ck1FWEXxhtb+$vDk_Q&fXfRC_G@ei7s(c@3{SmV6wVra?RF
zk9|Zqerkve)$>0IWR|;eJ;^Au=~H<mR-d@gC`D0FLu}GXYb_a9G*L$5$#gVrS-%%*
z26QGC>plbRjwJnPIZhYg3T>^*TdJm@Gn=twM~NCn0Ak|yP%Qi`VKwjM_*_I*2Dt`;
zkt3f2au%{uCeP;q%)?~aknAYhg+zmb4ggar5Kq3m`JipNbO~{y;H*MWw8qsh<{wI#
zL04PsB@pL@RU1I3fGq6}J)|AqWl^MTN(!gvU(Qoaqt)Om`6tW`eZ1~_gb)k32mh10
z(uw1M8-=<#Os;nLI{PZGf?0a>1lgXT0C`KGdB8|VBg}}`@&ri^Gt9IB-$-a0C)j*m
zD#9fhK+#o;F9oP&P8c6Ay^(*9DTUO0QlsDYX8wr`AG)IaRt})DKzls&Z2&ble67U4
zGJ?sg94EZPKZq9XRJFgoBmO4Z1!;Iys>Fi=U62yKMyROe{srQpyR{fKQT|Ae7Q4mW
zAkZJtTBys|?Y$h4_iZ>|?tKs!c3i5Hl?7PhT%>Pn!%UU>KVPT>gndNJ_5p;4g~`mQ
zONrz^=I_qfLIkqn95|j(IBStSda9>c-W6m%d4&f3rL43LnFZ*^Rnon`k3~EEcC??f
ze+bza*lDBE{4XcWMU!V?M!rXdIAt9jqLko_=Q)8+xx2kskLR18u^^Iv@?rl+44Btx
zY0MO1Vb~k*>rH?=bo8i0KgwgsYN0QF_c4H(L(##)lVm0eYp_BAQ`>%8pbJ%KYmF;E
zh0I1Mu9L{pY|4M~q*xr<Z2CfBE+@xJ=n7)<&p;XytYprhb`(fsV5Fi2?beK<uE182
zu70*aXGk*XnaX<vnWa8fcH=*XbWzedGbSr>fiTs(Ovv~C0wNpjP$D2<ouM2nlqjf6
zCdR!Y%r6#H(sz&e62zI*v83o038!j~c<T5Hs<9U@^kezMBC`|+3eFm??y2DiM~F*}
zW)rs;iuR;Ha>hapjrZ3kIM6P{>EA#&UB3b9yNI_4b_r3>X^A%&L%s#GT)54gPQEu0
z)My5M+=W^5I{=qKF3C0$Y4(c!B@DOYZ6R=?J5#ab_xa!LdKnWl<^O`19qB)joXQy!
z1)H68+fLQAi!hXKBU@ti9|+QwSd_%t3Y$yLI8ztcMUamxSntS-9N#nMCEVQU)RI%B
ziefq+7PY(qLbYu5V~(+A?oaELpPN{EY^s3ZNB@+^ODj;uytr7{Z)QCjFEJ<moX70I
zkxKHxwXHC-?VNEtB`ApsgUvlMg4OrvxKD(oJef1&ng4@y-<>SY@`JxZyA(D20+%(K
zuQJ<)jr2SE{i5zVf-LJcEfteaqZZo<&gv9{gL3TN1mtRkqg(-5Z{W`KVzWb15%sn>
zV2=D-+%DoKY?c6O2G`y(-<<iM=>h3jZ7zUp%#jbvPSKVY8ewLz>8j^Lgt|0Lyj;Db
znanj07@}#FEo)tbs}c9p%zSa05OUSye2dOU*n(wlr6-nRmmxsTCp{>K*&40ky9N9a
z57x&b3lb)My`&yHECgXTIgYExaZQ?yMIkN{O>X#?i*FS+FBLfeWzoMdh#Q}5H99N(
zT>2)vr-fV0NCR84r56EkIi(?BTxOvo4AB-R;|Pf$EHNvV<8l%1<m_0vl%Pvx4+zX!
zK7iD%8!QF{YX!XbZ>6wVW8o}Qy$&JHk=2*nDE7sHi-S4kbZ;}+mfbFpytxaG&Ezl$
zGtL=Htv?PHKqj0p#J)k;nb1{`c2tG%FUrzbU<tw$2ls`SDd>pBDl5dk*k?%)#LT>f
zD-wpOjAosx?Ni(@+WF%8j(RwKDFC=Cm2pddfe5$BG3*l7S~@3rxHqPYbW>;Ksvz!i
ziAnyHCY*WnkVQ)@zf2w*qdn&FLb|GQreI3K8EO2eutmpY=0ovsmIZ4FMt3geTMob#
zqN{dl%ohk^Q0JVCwlU}Oc`C_#mo;(h^1J+_qy<6`e3T$3j7=kEE_Vxn6P&xn?^Yme
zLCWbp*^X}_JdCZA@q>go0WAS|ZDmD(Y?7#OH8j<7MW9pCZ3(951Dl9uDA|&&L>zKv
z2NQo)*fQV%3(Fds)W^c86I`2fr)3ogvMu&wGNICMw<<uEEgM!E3}BDItjVxSq<@|%
z#O(U+#ODIshtdWv^YI(~tApG3B*tT`1n_N!YKbxHF#Q9cofylvt?`?$rN)-KQ+Qo?
zO>nOZd$0@4DB0d(mbVuF(HME9Aa$f6YeTy#@qn6kDTfF$myC5B5<Miq;{?yc^hC=D
zXn5eNdJHI>^-UQXQ%xTVa*NgbMGLM2*%*|HsX0(l9YW3HKx-_zE)>#5IX@WxYY?&?
z9y*eLoIaPE%l$0<CkV1!HSn-pzkHMvL)TMjjVUcqSrg_^4$VL*mKdF5a^#WLz1InY
z1?}W2@m@0qZni2s(a{|nk0FL7!8pRQbU=W!W*Ac?AK#3^#+niA9RY3%x_zcY|5_Br
z0$K*?LNh4Z*qz6L!a%J=^NzbRO54|BJTjjDGPx$T;9yLh0EAqV*QvocP$ad&@;p`B
zQWK%oci2#TuCPUruI*~P9yQ7FNxx;{m<iD>+C0(9wH(nFb-VRIUF1@A5VlH9KzI5F
zqqKgWPCA<#x&gHF!uZR}r+b7^pS6Be`iE}_;f%7;SPk+2gl3%*LnC_CT)zjQe2njs
z?4=^rBUx^Z{zIMw*9p6OF(#x$PYZRs>|peZg*Jj#g>oi`UR{*`O;3ogE7}#)%1S}z
zPTg^rK(|#HSwFEcBz$wZ607n}^8b1nh(BwR18mjT*_3}RC|36@e)biRHM?#`xFS1}
zH8ull%#qTMY9~b#zfU;%V3<*gwf~rZ$XLWgOncp4Sc6&Cb&Rp)=oiQ=!)EVX1<iNM
zgi*2l=8&ES(1US6&eor7))(B!9=E_clo7szaI=cim-Q$)?ko)9(S?{m&dI2B1`{(y
znHzSz=*p!px?Z4LC=*AT1YQ>8I^|lVG&JjP0b)Mob-P`pzmaoG@iJ48zhPGG;!tzl
zEkVq+cGtOwlySn%HAnnxOR<AkY%4GXiHLaNM1*S`YjXTOHfbz0iZMkrEvj#@KW=OO
zM{k%IkBN45Y6De|$F~MB7qV7Dr0O5F_@8h`Z0#OoQe9>n0Id>>vA=?3z{SV7R$(|}
zV_J`4e}Zr!bqfId>I;OxE5}~R2Kw)!E#d;E1@6*|ZVQ-oM%gaMu>xE|u6(z|8%-K#
zS9n_4j{oNyGW)90su0<;tZs=5L}WdXv9D%d97}Ew>JnzpI{Fy&7~!s@LU(c!5<C1k
zkIMOVXLsXu#YK0>qq18~6Vz})W|0MFA+{BbOv?4?v5UfP4MpaWy+x+VLyUbQT)U`2
z=$p(W(}g)Lc0Bm8Y{t>ZpDy>ly<;AShunxGnnc-K)?&S#AS^&R$LAjW&H%0m%vb6i
zHGE2em?z!Q{ACwNvr)ysIaT?pyXH}lT%0b_T(N4Y#3iEL`82yVZW=AP8?bAXduNRJ
z@`<#UKr@6H61(f3-Sb%3Ot*3GoIbJV9)OgArw96Ch6uBqUUjd<U-tCha+eXui}2rQ
zU(n>n<w9H{ZnMPIB3=3D1W_p_^Ze0!{pPcx=BOz`Qbkqcc7@G5due&mqAzc=cMj?v
ztSqi-@u@(|3*|GD9SX}e_5m|Xa__^Z-9nlN%}nf>E1<{pN0w2tEY`^Xu|RWdhmc4Q
zW8M9JgQk~!bcy+4p$uCr;+UHj-ye!H$obCs3L``g>bv455te&V=KgU@7L#Lvcu`^V
zi_QlnORQ1=$;K(vIa-?TE64>+&ZNX>MVLRVaNx!SGPDR{))_UV!jXaH?5#N}4|v>^
z(p0;kP^YLWNjK*g6Xv$XO0zGfigpg%lGCB^uR<sm3f8`Op#(vB`#B7FLt!V!SsPO*
z^Uu-c{Bx>Nl{lmg#EpZeMB`5)NsWpt#T6nfBo=;LLb^wQYe)KO?W1ZjrXA2MA!cpH
zdWAS=94C}@<2Gu-;0z^PjOifkJW)r)S0Wu<rZdUN?YmAes*x)fth+Xu43+gTlqGSX
zXfx*3_!$DsgO(iduyq%RHk2IdD(F9ZG<gih>q1~gN51QJ=l{~QyJ*5k3CZ~Bs178*
zd1niQ<sJ+TIY4<|q=!moR`j9rYEEMF_JCVJTE8<bk=rnV&YxGKwRsp?0r3bSIpRf=
zOH=h2QU!J^lQ}mVm{?3uCdx9ZR~h*%F3fGp4-)J2DOPC4%#P-2(cQ$X0w(XHdik0f
zuH!Ky?h)Y}S$eedepkelJV#~$S|`O?QkY9n>*wNlB?g3$ikfts()=5XDP6#OyNMLb
zsb%iimnWgu?qTF$Az5kB3&?)tM?qPy>_?*&zY55z0;71YuHn+44o2m1p*VW`2{)=1
zwHl+A5H}TRlj9Z9u939OPlx!YP6ecKup7EbLC3FYn!axU+AI{9n0Zcp8i3iPbtbLL
zN`m~2mZq_aXmiPBA3O3GEzIfQ0)^q{X8~GIF;LXv`RPz*oHut^aK+jOfMly29Ltm|
z12i5Q@@~Q%lLHWS^1lipCEb~;#Q7Q3TN}g#W$6R?&#Xwz+sjB9txb$~w3(6+(<uk#
z@#JqfO>3_Tb52vaR>p>8g+oBx(pVBU4EtAZqP2e9KiM#14WSlFI^TOvwA;9BgS*Q%
z+<d?JG3F(7owWWv`%of#MCj?59+Un8Nfp=WJ$EQfX&I)`oJ~6nDw9TX#T*w4X*6MO
zsAC+u_Tf3irRlU1-A$0Iqh6iHDgNLQ02UOtcq-Vfw8uI}f{|^ND|zD-c5$>DitV%i
zdN9?yQeku3GHOD6EYfNBRp~?VZ5Fr6lfR1`#h=ax#c%<)0z9aDG#L4`Od1vYE9?UG
zXl}&6gAlqF%Jb1N=~xJ}k@g_FH))os<KiySE<G+gIMw8dy5oRcX-SQXcP={Y^2dXk
z5BeX1XS&)zm=mv8@ozFHI^zX}G`@>BJ|U06lvIeTL^~C>6w}dM7Gkp#!CZVDbzp{l
zQV!J0I4_7TPX@5O(7{Cgl>C48ZN(}cw*bsB8^UB5!eB7(X`to?oeV7{mJ#B@7SLA+
z8X?G;<8GGBs{)W7k1G}83K3?PzRPuN)>6Bj4u<f%c~I#hg&Xr3YL)aV-zx&mt&Ze7
z&TxE;45IZ+!WLLj_B5x6R@>C16AzpP<rc3;*E?ZQhn_Rf1~!ZGTwjWF&IKR?PDyoK
zG()7B<oJQ^K(Y2d&r#T$j~O>AZWE~n!)3$w6eiD>bTl{Wd?+`Tu~E1H!tBvFCPkNM
z=ggZ-uu>7=@)qbRyxWIS$^F?^y_^_Gr_uShJQdVW%ypt0t$kt)zYx;w4$vdhC5nTE
zxdI!i1D1WK359F<L%&2}vw~9}=DM#h0m!6`A;V81DU}8|uIyY2<$@%o%Fd#*F_=MG
zHt{zIqTuw3fw<$cJicCdi8U{W&Kk1Mc&7+){!G;@^ZzTrENiA_64z#~Gy~Kb`4S@6
z{RKI#0zQWr6>q%)L|N<B!YmP4laQs_LRaQAJF?L+^eSjo9ghlE73M=O;+5hs5td_m
zE&)wIJR^+6vd5s}Ezz#8HqEQ)V0z7~0kdsdvN{*@ERt>(&ZO4X8<Vaf=(H)V9?RgG
z+O=So0QoU9J|w_p#k4eSI#;+3$Ym{YWrShl9wDx#B6_4ZM7p#X4R^;ve}`^}m4l3!
zV`dJtOlo1tc+B<r7nm1x#WpuUlOTpk*pbG5BAp<Go*o^dX$S1h@GyJ7D6_~36}}eW
zK#*CKuA2VyUm-{szlw48jf9;XvtwucSG2n%yJ#LkqEAw)H~#@_`S1b}nl5I^Z-r$E
zWBgAiT>GXxG2JAK6GT&exGv@$mzyDAlpb2?h)+fN<Nj_fVq=|KK%7N#ua|Z`!pt3d
zlyWS5D>Mn}3QVji!gXCO>hM=1^q&Q06=^f*O8h9yRiXvX9JfI_YerKU{H!YkfwVJv
z?V>PQpsMSSyYGN-7Rg(DlEs{N0uCRt7zJSf)KO%9K(_f^{FSv3B%nj~L+%DLL-cgc
z7wPB2?g1kYEoEbC5!qNSlO@DWZ@oZhCJo-CVJFSIlJ|nS9JOjO{wNZGB3*X>cZqVT
zSe45e&)Y)GJkF>v{hp4}ZH4>tXnKT?ZP_1$xJ9yx;@VRJoXvDh9Jk2+m#qU<MjYD+
za3Qh9Pm8k!@6Xd<-!pDRtRui_bm8OqpCa?REpuz96eAu0bKRnG8;p$}&g0lPjA3na
zVOdrkofx<c#!iB$tKPxB*jZuZ$izP;4ib@7s}|6eI7~>!1Ut1w9))y4ndYhX76BGe
zQiibM>x&ZvnyoJGCd6%zL7N3#Z`JEgTRaZrbak+Oph#z&lzr?NguoZqEC%C85zdxF
z2>Q}(PXJJY>C>xm?o)&<m_lhXR$d6^8DUNy+t>I+WYz}Rlxjp{3;DFuz*kzVuW(lL
z+}S`Udzg@{Q>%*qDLU)ajm;Msy`LpgL5uRmGkGF<yqkmdMV<xmsM*m^OUBAau9o6B
z;aPDRNEt^Ldk(}57uvd`F4DO(hL4ZGiuMOsP)mT<2tfjq+2&sVn}5qP^NNF?=U+=)
zr!m_8KP1R9<Vd`0FdlgUz;a9016VZ06)%E0AKcLMlwSNS#Gh3<ry=^eU&;aWowx;9
zMgTR1c~NJqEFvohUnerF`DvUNvkqNFF<`uek<LhkT}=Fi@)U*Wtey+~jMUQv@O^R}
zY_0X}+4r0r#z)Qg?O03}V*i(k?|}eym)4t{O2>6Vv%170Gwu?dbtUfqvi>N=EMd-?
ztLEKpvB4_<W}V#=`_yurEXb*2v&D3lrh;W(1*Jf&vtqKsSq%A_WtW1;9un$<H10}y
z_kzQHI_$Ih$C+SCqH8>{(rZ8&vy2z9iD)y+W~7_N6SjBFIL!CBxPARMaeCx)4oj6L
zh7&$CmlW~=VcBpth|-D0*kH{02B7O)cis4OT_GwF7A3nYY`L(9<1Xo6o6#|$tuY>c
zCd8>?=fRAs9$0)c2l3F^C6`6DnAHTi2#p3q+Q^>;G6o}1wmIdup)3(;vos$)BZvmZ
znNn|D_YQ>l!Q_s<bY~N&_salqlXvqNr~xR>eE$Nl1a+}e#uzb7SeA{nJrrZ*-$WzZ
zbUk#uFz1F28;gr4-Uo2)bZG*e^+BGFN+g~AkXb;MXQe!#`Ri&S83XJHlCEjgzroy6
z{QO7SS^OXr$>WkIX8*rD9VrVLB)%7tRUcEe9E0Hd;-fqo_9QZ{pDBbj*<f$Ht)SVJ
z;agnsF*NBHS--|7ge`(}PJ-xG6=X)4yXnuAaCKqs(M7KAY}$;X6H0Ws<=9ggHJI#N
z;$TrOKQ>P|l#Ra$$+*z%jPIK<n9nI&Ec_`kG=^|?8^?-(Np8x;g8w1xA}6I)T8thg
zP^Dlldq-iuq<Y52cb`ExqlzvptnfKBywlYCFhc*6NGH}lNH?kJ{PQ4zE;IW;dsi6#
z1(>V7*exTN^95K|_&~?g8c(o%+5C=;yvB&wQJBBu?RYxN2LfQAm&YThtItF_nIgv$
zy2motS3qW&+}YTQraWy7E<CCSfnfC-T`g`F<*%_kNVkzM{Tj$4OJ7;GQYfIa)u#SY
z&hq!)8rOdVY0lW!qCR@*TY#*V3Yc>8bnbT`<ct=UUM0d(VW6IxW{^*X`4g=T7}!4k
zF9;PuqmCngAnbOc#n<dVLAre>?~p##Sw^UfAR`D36k7>$PJS*GEPe)aPPzflwU{-9
zz#My59#9nFdPFs!Y@KWIy}(S6D0pJTFVIdm9df8G?<cI$oyScW5p(|!z!~A}TqlrA
z3UM+WI#^ptG#Mk<j#x#6)9fp!ORl`wD~viqRl!o{Xi=G*c|DmX_$z?rkSzgp%yHUm
zzi%XzZVgD6dRY_0QwRDv#>N|jkvQw*fns#b4&}sghs)aTXaN?dtk2a1PZwks<v}mm
z9c(^Fj+F-G-x?drLxj2HdM~_pFcz3IM_AO#Qs;MbfmqI(D=^Jq{@z%i7eZ{4nHbsA
zbMY|2PLIv9{KCGci10os=?uOQmC3crvFOor=V{}Ro7dX(PUZgd06LN6>$@KR5<;uO
zZkR1XtTnI6nk+}k(Js;_I_MMO&h`U1&D9^7V7TX}P2~ae0l6?bO4UNV9Ipw4M?6<h
zhBrlJO-SC77oR^*g+(ilGOF>?5FoemUMk@E1t6U$?MD+nTDv!!AR~&5fN59%U67b=
z2U1^a`%Hs;ej!j=7j7Z?V%~*Klyru(kqEk$Tu^VWu=(O>tvwDC?PX==51pZeRyF5P
zaOXjD$VRi+Ola03nBUVDZ#!XbNt|0ykcR|VU|8ic)DFf!7s<a%j?tLx<4IvIuY7VO
z?R|`06c8bDRg^x`FUlzuD$~*of2APjCR^8Xd?|p!O-H#pYB9nt9F8t{0z|Y6p+1Ly
z3$2)UTO3sFz&eAML^=~q8GYzH-xbs-ybN;u)Wk4QcpFef#0R3x2-on~I1USdEOR^<
z#6LvC*qBkP#<EKg&Ke^xhv61rQz4FxF<4t1xa4nQqzB1=Y$6;$NiDm^QpBN#rI?7<
zMZgW+8UI8-h_W!{vnRch8cPn(!LH5{R*up;t=j~1%Ziw1X=1uwdGE(-kClYEb7Fko
z!D@h26)s~0n{l=iqoXR?MI&SR6PkY+kZknkKw?pm78LGqxo4jMXNFA)n`P#RiHhXR
zwDj?N&$zS|Oo`L=s*OdsqNdVCHS=sIh~)_uSMx6Gs9Znmi!DVsDx2?KsZDoS4#;)N
zJirPz`3l-qs5(vZKDvDVl@zu*O!_}TmTRtDi2wY~AC>g*#D^<DAP*@+)eV<l5!&+5
zXhuufvx$`O$<5&Kj1Y1~8l7?RN)YCiqYXL&574d*MAN~HC=OHD_1x7Ok2fjw{U3`$
z(mtw-<5q#RU@J0x(<PTXR|T{5SwS)~?68{C;}DGUohgc(T5z{NY<0q}H{5T;?jrpq
zt8PZUmNfxfM~tZ1o8z1iujS9Uf2+Z150P2z)_B$y3+_0%iS=l7#)Wcxxi<0Kj<{XI
z`;#M}k%?Tc$Cz~>D6~uu(6Na!CsSBp_r<mA`rB;8ifczgyO#L657PwppPvXMjq21I
zJJi)iL3wy*cI7#1t&uevjME<$zn1vg&HuT@U+Sc(C$14i?u>z4dx5!cGk~6lfd=-#
zM~)_jn-&*0(hVB8?i!n;^gBlGv|UeVR@;27dA>&xDLLK)@vf8+ESL-NxQMJ7XXikt
zjstR0yQQ74OL70OK*y6dQ1(u9j|Xxy>_c_VN@uDNjYsI<zf#yA;JFN$4aT1)_-{%S
zb%ibPUe=?1@udJa4VrLlI4S=h178@}Qh+P7O_h1MXy=|;_RkdJW`V9p6Bg^4^}sAQ
z&0}aaigeVW>w_W=Gz;a}UIdkBAEx_k0A;qj{SsxoA&64->#2ba2|HI7db*l(rx5<-
zo@mVTdxwkFt{ASc`IT4nnEemXWMfaF!$p}__>!}LQXC}ABH>Om%c;|v0G7LSk8_#;
z2RZEe|1N;Rsn{0xD`?L2`XCa0O$ZXT^WS(wWFyO}H$NW<Gn1?*c}IRDXqO(d@U-)L
zR1oFUrP)y%6LubYS&5P5z$PM_5mao;Y(g9jJpF3N4MO*t=G~Nk%mB{0we}y37GX}Y
zv%=9B3v40CN#j~yoGD7OFLuT)3Y$5u<uM0gBlE``VuDZF*Ovsj_10KDwx`p^?kxaG
zl>USp&ih1p8^T#NXXveXS4~6N0$cJA*$uFaSyy3K2O7??BjWD@oPQk?z&2J7$881V
z%-AQlCl7`^g$Rh;nNnqKbJNz4ma&>x+%Gz7rab&1dzaYvPk9_>ol3kdI_n*Hc<76f
z+X6Tv1_>$6Nckv%Zroi37AGA{DV2CcC}qX-04aaJ9h7t9qzbQ%G^1JDgJnYR!5<EW
zr9zxMOI1X#^;rTPtpoKci=EYWAR2Xx(gcl@2xO%<*N5%sF#DUBq_98Z%t#HpCd8?;
z4@?_ACMbcHCi^+|0w}cv!MR*il;gvlpvVl1>5f=#7YH(A=!=6D_Lr#3SiD#8(gkEm
zREz26s7@7M+7;ANBnP&np9vt95)XlWt1z|1^$M7Z8M{G}n+EN;NrcCu7M518iFApu
z6V-lm@!dgOr&^H4A4RLt>%8@<Js>Rd9&|o5{ZbshCzyFg#gZ1ar#2xr{Atm@>RwKU
zy*4j3itr~j%m<}KJwN{+mG41&=h+n4Or?X0tA)V_T4}c9B3%4Zzn%e`vJZgE!7URy
zP@KIlNVX%vZKSQ?wZ!<R{$OC~=#7UJ^jDa%Vucc)3UL9Yc9%z)cz=JetblR=QtOIY
zO$c3TGSU4>5EURJ=^cuM%^nxx7>JK=1)%y-825I>C!(Bfnr`~yCm~r!Dsx+TlM*zA
zMXkmuM;XGYO~t81IbLl8aHC^o!R}WY#jV=`&1Jn4V`I9@?t+{TuYq@>fZM1e|1$Hk
zO8eMSpfhIzF2^OJGpVrfMYZ!>CkX5$E#+pD^S?Q!la6~AA(W2YJr~n{5@klv&Gt&|
zHlhp2ZJ5;~W+_Q|^!sj5%R1>(rTksK2Mj6JQ9!PwuoJ>LS?_cPt{VjT6BY!O&JykK
zCSkCTrDamXty_V13uo6Idx>;u2Ff)Zx_l$V;$zLgnuRr0%vW_P45KJ=$HZEqNd+yA
z-uT;IBob(qS@;Pe%$SV)xp)>U_X0Uh**&t{pfGy~OntH*{Zj~g`IkJaCo^WWpMAiY
ztmQ$Z7B32+c}^S|v)2idX&V=ZVsR0!CC)~BV>Qtvt_#DPD-5&kR5A;^<3%=_eljmp
z4`MlcP4*2=&0}akk({%xJ29rLD1>&Vr8@fx?xzT_wB-hC5Y2IXCoEgWatnoSqrc*?
zsd-d(Pnbnf>njg{Y0SgdWLK;;4GJFRcAMMz>xp!uuBAoLT>@M_7It{|6pfIR1||)x
zF*pchmTA4xLVPa7b<HVBtUn#n3~O7ThWd+yx$UJ(PIOj=n;V59Jxm_rx#m~6rc4*4
zVwD5<AJOAx3Gdk=h+O3Guc~lX4cb6oCK?Xth<fy)`#?yy1KKQF2+j0o3pC$4YykN}
zh+9yRZO&9$*pdeUsXhmIhV<Y(m8750qCZ*%x^VprXvy2abHZSgGsBh`aR`K2>(*Nw
z`yL97{4qLY`+LY=pj>fGvy|Jz>P3Rdwu{r5M-?`o((uFQ+d{GyfE)a@kXZR}e^F#r
zI8I?o-=jrOIVOp80*pV>r2IjM<$;fd_DVY8*|!<3q8oa(xL63~z|R<0Z!$uTsN;kq
z_{Y_k^b6d-C)qiUbo>%V5ivxB%9-31%n;><g}?t%BjRxZnOHC*h;Nz^TC3Q;$4N&y
zu~t3U^MMGLSpAgqIv$KV8Y~-#xEHEPq)iZFYZ=`VdmTgAIigg7yIK>FG8(&F0hWSR
zWK7Cb_-e-jIwx7u$bnZ}Doi6OtA=`9B?_q^%dYt3IEbv{VJY5N9^7(#o{C&Jv5l_9
zQ^L$IeOV)ayxWA@s!|hnMBfR-q6kv-|D>?LV9c(tYKgf{1ajBHGCh_&2@=j2AKQDP
zcuM|f1&z;Yr$Sl+?7X<sxVL~tdbA8`^;<y{rcxLz#1f}LGz21p16g#hE07rS1<Q+x
zCOW52AB=Vp=4F7%{=JM$7F4{rDw2X=Y#;NS;c$uchAQk#aHqyy8@03*1iFc0%*);`
z))Zo1_3W390<LX-iBgq4juiL52s9I%hw<PdB)Q@PFs7YF7|#1y9K>fLoF2!0X-how
zY@kNZWXU(=uh6dgw2ssP#RI}fuY&`pw-k1KPTdPJa1ONV-kw^0dHo^5PR6h5(bQ(0
z3&!*@eq;<kkDwW`vaS}J339^IItyHH5bfT@+@W(bsWm2C0IGIJe~(rXPKEV6akv@#
zHz3QfRsJ9<@qmydjB@<Y{I?4?@EB8$yf9BfuQh2J870VdQz#|h2BbURMS$jpd(=9x
zmHuJ_p>RW^=hoWA`FB#RY82z(O`zo<g))nhS{x*d+R~vGV);a*3og^6SoRX=#wgLk
z`NLe7IzH{YhKGsWM7q>GM1*UAN*peTBxIPYAz+4RdIU}iIBVSKGXKl>3`#w5v}nty
z*w?Q!!|R2($Vn$mb1cQ&mp4D_XC1M@3~2M!Tj=fQE~5Z5qP3Qm%44qpa&bw*4rWdk
z1~WYP3TvBP3FWMKSwIKDj3!lqS!=;MGF^x{S{U_KZ7;-e3cGzz;t8!MMbeP=SwH@*
zptanrG-(9ws+dyjcon!))N7ULF`So#x!gJ#pd0=w#I>4fLN%w(b2TVA7yG;79fcbM
zsC-%%t^sguVR6U9cGp5@<xzF&UTo|l&?6GQ7q1iLE@Ee_TG<W8>em6gP?O<!FOlY0
zXW_lNCU}CdEO$Hx55_qHGLdUVOY^z18E0~JkR|v7O~7mV-oNva%gyE(Rc#z2q*0Za
zh{nePTnd&2G9~B4cBPqs<_K5ua{1O5lbe_ng=u;JtT2nDi(56Z&h^k#Pjzx!ub`7i
zix=&>UlQi%13KuPcmuSVL6eV2H_HI8nrWW53ZO!2oCLh9uts8)ea9P{N$EPzfuf06
z=<1Jk|KW)A)<Wzf0$HQT)#~v+kuEN$rhRqk2jg{tmMvo9P6}43HvyToQnG6q5Rf%1
z?A3YQw-omXa}#8xC2xnGGv2xx98U4()gJTRnt#}Mx+O}&+yDl!Ke}I}Gm}=l7GK?#
z1096{jtSdu&jD=CxvaO(9ROyh#%RI;Pc_<vA%i|%Wj*Omhr1a2TkdjrDmJg%-%XgD
z`wGc3{k|fdGmZ>crJpOnEe0F6KC*teiO35VO>u#Hh~vC@-jbmWQJgQ#BGe@gSsq+0
zNL7Lg;d+IUh;-O-s|e?zQ$xne?}a!;rq8m>=Hl>s!7QVeQIq0!ky(M+Z>BeXUdf{1
za!%Fp^L<2d^Fo<ck7Ms|eu8xaH|5%6s|Ubf1hqpUCTD~e#BsL>i-N;>RGA+=n17xW
zCfbD_^G`6!B-6|H6m~J>SDEp9#fL!L-FY-DoviB=X=TCY3-90P@E(Sym$XieBNepR
z(w2f5{WL-5V=~(P?zrL+0JSFhn7&1XrGjoLUK8o0b*(SPJqq2Z8r>#DEEfyo6PBGg
zw2Ci8@=4O4@%MI*`ETY33^w9Gk#IzV#;z?LUA!a6iL$ZeD3RfQmN54t<_#7ta*T2I
z<A5GiHJRWaUn<90LXjC~V)2>6l!Z6YCdIfXAj~}Hah3R^=uDv8jb+I*LkO%F%W4+!
zsz{4Q*Vs`3JPBa&vFBxN&vhL^X3Q6nG&wEzR1UJOWc*N5*hC<F@rY#96;6Dy%Rc>S
zh^$gr7HF-!@H0SJ&br+diwQ8VZJp^vc4;Bb*t3zuwAizGG_6x`ka~y^3zIRrj^n6u
zto|GroQ%V)epiKE!KjR4^z)D~CNtsa6G6k7G%^->fuQq8?<nQlT0)4ad%3a9i}~Md
z`(*_Ej1ZSV_r+?l<janaU!a~?TZCq6HmKe4v?vRJolP=3TkMtmbBu$K^LqlahR8}h
zUWLrgoc+v;wuiMC@)|f?F^BO^sR&m%3$C7+lKsg=%|e{i{1cBOg;@V}A~))mLa>E6
zPms%y+GgA@z*3OEJx*x<EyQe<*#nf`$iMJCrT++UEm^%I1!DJ`fS$CcbjBQS<=-&T
zQPfcaoD0e%t|rIpLR=86CCc%hXeThWKfY?lXOp1?)#`8azjNi0=ssI{ZYR*arB>kF
zNCK-yLY<iw0<6Avc!zjdi{XA5+aw%i3$t)IsZM5av%i}|GCG!R_m+Y@9;t(oOHx|)
z3(ne3xk9^)6NS(KsqNV6UwJx=qR9;XPC;g$eqFQ~_)YIM<0X|#JrYD^>!M)Xq5ny<
zi-B?YEW(XOpBD7(@ue_xfuCrp$@l*-N0PM-?wp1AyV@W}_Bhvw;r|9B-Ih@;algWu
z7?=Y3STU^m5fD6-3LV`sUPRU>z0qeb`jbF6kuHo(?-1!m)yZ`RY#ju+a(TlG8<S@I
zmX>j``p5a-I#G@@MK^emm&$saBgjSL5*OoO9Qp}JCIGpG)O7XTr+F;KeqFa);XgT2
zkRy>&Ocdf8WF0o0COwtOE{+i_*|H_RUt!uF9yidqKau^z;Yj@W8UMJTcxH{RFCbme
z$!I*<qV-DvM3t^sek{TqR@3b)WO?dW`4>7OsKz$}$QPTjnCEN43`RI)U0FeY+gHxc
zj879rRGw0d3%@075tYhq1yrK(h#)h=<5A2Ko!<ets@i2cn=Do?6zC7ptR_Y89nE-}
zqHM;o&VPyLD&zi3A$Aq*3P|Vg+?Wy4n3_;jvC3ulako%QsJpv0Hva)KYi{c38k^Vs
z5sY%Q;l$}&g`G{ys8OrN4Wj)yr80>`Ciljz*;jg9<wJ#CvU0j1rYi}<egbsKuw9Z7
zcKlJ0Q<G6gUl!vxTbs)TS!ldD3v-V)Q=z$V&dpT!bDmo(hE_OfWvXlWB?nQX+MqPX
z2s8`Dbb*$8w*PBpz_l>0NHqZ-2BHz&Ex?jxch4{05MrTp;UX-q{S_L4aep{IRoGdy
zO2GpwyUq58)aba`h4ruq;<h8bE`=?8+^9)VI1>5|hSxnb=e2mhiIE2DWfDDWcD~I9
zD7q}xX$a>ofzCWFr0Ck?<^VFUxFjsbKBAo`w$$jK7z@9eGskKL&IwRz#X@s|S@3cn
zI3iAJ0(#D?9<$F)47aajSXD`WCgz?ePm2Oc`8yc%G_l$sI&W!lwiagoQE=9H%}fnI
zpflw{K2yUJ0y1f`CDb|GO7rF!aJqtxJl{Pl%u{j0o*5ajRuoQm@9RX@^^J&364DGT
zJs)&q>B<~4-OBCD1!r?q?z}HX&JXIATE{YUdXrS=dsn0<!;vFn*bpc)-QGW$|2cIK
zWT`Vzu}p^PX~Jj;c;t(#6^6HL2^w7sK*RWCY3yc+u&5f3!E!r(K|p81u8v3W`r>Ip
z&V-YESu0~sFH8w(U-m*Xf{uUE9r$4j5h1HZSWSvkJO&CCZ#1e?6n0Jn<ziB#&$e(5
zVI9O<seN&PAWA?Am^>8XsG-oV65Nu=jgeh0Ex#zZt9URS_#7#~$@g>drJvo_<$|)w
zq@a_Xcu|POpH^bcZ!#mp*myBwz%g&g#DNOCrdfIP#PgyVrN)koRfiFDZl$aq{v1H&
zNam6Qv8Mpbo4ab9t?G1qsU-m2T2du-GJtT-u#;fdJkLaYF3PE60Gw1nt1p>D=`I0D
z{UFH2Lm!nKsw@TK9FtzX4LD8}m<^7cj@;KIyRrTFT%@^X>&754*Khzhu?xzfqRg)C
zh3OBdVRjMh+Lnzg7T7w+oO@|dvyH7lk8HqK66EBuOc*^PUJ_t#`g(h!V;M*vZOQD>
zHLP1`8!gKpxtx6HMPt83u(QyhAZPLiF9+l-P>f-WeqIyMMc!ImC%|c8szt}aoS!-R
zy5)i0&N+KxvqSn5eg|guYW-8=6OpurI{W|gRv_FMX|#yOTUp#9&~=~--K4tliXi3?
zB~f}(@-ab<Bqasew5|j~Hna<M;mXR8<|Jv1#K}X#{1Nkk)VRM0$;6AZH@%&{z$!p=
zOO~<KxM@`r$+jdBZo8Nzm!iIUo*cV*ZTEf_;^b;oE(wvsscYnrRyJZ|yETDa(^Kii
z>7*BJtWZl-?jIVfo|V=DrdVD@^Xz4w9I6-0?FQpcK?szir+8IiTAJ0FZ><ew88Eo(
zR_|cEFU(ad|7<801n@Z?PUDz$2wOlriY`m(ONEf2UcSKQSd{BaFNY2C=u9Pk6)1tJ
zXRNa>gp)@(OWlnUKmwyjjAOgs-u#>Sn<?W$0pvcJv0$Z<ge_m}9Z=bb=D+bQZoW`B
zbMH4DW6^4@;p}3aQT#*pHJ#_}CBmYWzr%J9QF{wDQ#we+a#n!Li2*TpW1^)c|FE}%
za>;TJTN9xOzZEajtcXt&NjjLbvqK^~ysR}i=AvXh{vevstec086?C4oPTt%(dJKT`
z?5&oy(K}-tNY+|764Em3YC(?FS{*1>2cu&=2rTHmKL)5*MOsGM{KPxIrTcp04-@#`
z0$^>W`|IkKTMLDm<Pb;mR#kwr?qTVn4)dTOXDw9#Bm2S=L69u>jqvO#!sTJq88aeo
z7XT;hmg<@(?-T8u(TK_XRkn7!Oajfu2{*x1^aMM99i{21xQ`XnCBxNyTZ~-~+Wbt{
zHV<1nA&rkJIDak09qR)j9n?j-F#C>3H+YUxX<zI?{@27<u@#x`Dlz{C#G{_3m1DAk
zE>|B{DEi_o0p^3*yF2FF5ZbvW2O?D5u7aFvOUvkZK_u-|u2<&%Jz-?l-^W9hS7ihm
zkM?*{gn7dNh&>k;6$|_UtkEXY^u6syAkK|xDk(dj-57}EwdL7<6T;?}^=tpMI7Wb3
z>8wwU+eD`RuhsYPO`+Ufc|RG|iB_kd3v~HfCbENEVl!y-kZc^dBPqn3`k6jzcX?qj
z(ThK&y28$LvaZ866TnAC6;0-IS!B!{V)j208G-2BEY=Z`4M@o)ELAvsb4S3&u@WOi
zWZj&N38%W8BUQC$eL?@YEOrPim-gRo2?$pc7$JU8I4dkqarnC6@T~yJor9#lcvOUA
za7387!(L@;KueURGFBHC2%vPyGPo2oMY>t%Mkz>k8&J}fVN(=DIEgy9q8Ky<xDshm
zGNO$a1zF1N@^{TEs(;E68+gSA+k#}R7qhD5Xm;-Hz%1(urw)sWq`>_w{=VFvu!d$F
z0pJkg&roobOpa%YP?O<G`{xSh^*uJ$-vQdW=$+AIvVEU0OC~LH*m{&=+>U@QmQ>*7
zxK)5tf|o-4D4JY2IgQzN$|K{;hV$DH06%Ec%CW2nSe>5EAZ`@Z;0U9PU3Z4Ii1AQ3
zol`elt;gO%oxNINEd~TwgdK(HIw5^ch~>=5CpJdt!UU0Ro0KzOW`tf^izj#CpR83U
zLpF+d!>(X(($6MGR}XF!Z8`LCbKBPxZrTmdjBpE&J<}Sy=PAi&G{a+GyeG_MO7FmZ
zA=<2YAm`9~4<Pf;Af&GNg%FsXE-yth_k?Jy&X}zG(i3a3>R!NJtg|4L-h+99fhOT*
zmCFoBOe1yzz4YVUx8_R@nz+*+W`|ael(}`Jge55*Z~Q32|1(J9O%cniy>kc!Xljgm
z39=-xJK@$%rvMj;gBK}RF5f3dv|^#|pV>D@Si)n19bX8cHE_B`y==H2R5n8-r;&08
z*caCcrg6258y`<746BS`^;mR&2s6P%gz?Ht0{FL3#0=nd5w2y<$Jqujg8$gWbVCKr
zK`AyX5YL<sFauG89j#5o4!P<=)Qiv^oS22X%N_BR!WLGK>^o_jCzXJ*X_v>wSw84w
z^+ur-lzkFw=%+>bGxSc$ir^bT=C+Od!#Zz`p=B`l_;yV99u;Xu(T(zQK|CYGAC)nh
z<}kPv)7x?kT{ePcusytqaXr&D_!~@aJ#c;vmx>3MT%u)?jSeWJ$>Oy=R>}z74v$G9
zoG&g-b>>)$orGl#Ku!;7HbN{3T3X{$d5=(+9rv9K34JlG`Kf)m$EiY8<>SZ43(bE^
zg|t$fuhaj^?)+AToe8WoTx)(L)ROD&FUDq*A>Bi0iv8{F>E?w%_fs7272~jGBw6D7
z+2y&rh~!dXZ-{T0*iJ~23N>_n^KUtYX9jpih;!x+SZvx2>73I&qGU%fTMuBP)o_f%
z{5+>1OEh_Bq$$M~mHb<u*-1-tfIzqTR@R<{I9GrZ!&sU_;!<2B2o^cani&1n=KmeC
z6XD@lK`uF4zwWq5v<H^osN?Y>BHT=|n{Jci$@r-GEi3cRc(yl>!@{DH28YF`fSD!C
z>2>u#)^0*w+(hc(aQW>fgqd@Cu<1QPmUdS^woCW`?#q)$jVRmmj#yWi+1HD4(IeU-
z(Vnmpr#2(-3|mLMCfdoS1p}^4Yk9J0(mIoi<5lZmw7GT^s{8j7b{4W7jG3aH0zE|o
z_M<{98<~Kycj%5~raFZRdaQ8@TL#@#c31SSKM86~ZgRm;m+@D-P$$H3xt>yJkM8Ej
z3{%W1T%Ho-hLnyllgZm619^)2u3xT>TjNln&RS=;s_`cUnSZnZa(p&y8VEvT$smu4
z+>aHIb(u2i6_Ng;(1t^WO57yG$rieGoKTHrrUS8fF80Lf3O3plo@9RzjWmmc<rsMY
zVW*|7Hdjqw7LrwA71zcO9+cB;%j)r_XnoDm@x+4(J6}2R(?du<38H~(Q5bt4;<zp2
zCdAM~3A-tj(Bra=6_8Z}L!X>k-6V*DvXj$^)x9Fk3YUVr<5AIMjdmw~RoHcteA9L^
z8$I)vJU;VdUwk9l6~j!;PDAU!B@P3$Fb8;TceqHi)rL7QehLLxBIt_MIEEb#l4XS_
zHSGc8Az^MXs&g8uH1yjZ0cwVOLE})-4BBXyi~Vu$kx**I9p%{eD8g{T#d%G1HIa>F
zg>2DL_W_?SxKT*Xs&wKtQy8gKnAvYu*t{|V$O>o0qd`1gpgr$HJz}Noc;U($tHVnb
z_E)Hqr4nB67!V6euF^EsPLF#9@<}V^bt@iAIIBGN+~qh*fJ=&9#X!t)Tu#$y*_Dbm
z=W;KWv}B7M4`#`?F#2sNG82}jv5T62<G?6x7D@IzAQGP|Y-!^Tx>e#~U;kqh?XJ?n
zG>S7do<LlhRr0`wk0w&S{6rwyWIDsXN`$i=Bin}6PJ$+%sceJRQ#fNurrz6T|8)&k
zV`qge0v$y6#^WagSZK`oG9jn=pL~i_YZ*Twu2tA_W|P~_%f$<wmVY9fGHR6OHS`QX
zdTLKwtgK+8W>S6bEySG!PkmT4l;Rsf7A?AZyoI4f{XrNFO`FF;Y<wn^i-CH-7Jn7(
zMuAmIy7HJ7(&Nqs&$@3%wa$w^@s1#uvc^o%6%+mn;M{b}Fx{L!Q&=V@b*Vd>G+qx|
z<6+Urn~PnQ`13gs$hRBq!A%M~b-b{#|F}nhQ{oo!*tq)KobE#>ADf?-f1rgAt^a%g
zl4jxG7sD>dW0$387%AG#M6(elgxIgej?Gvs$)`tAbXLT)wjJ^}0E>Z|O-koQ1d$D%
zrwcJ%gg@Zufj!*`0$fcT70Q9)MHhm&96fA<V*Eupjge&`wh&Ev7)+I-da()Ae^aAR
z1U1JptlAdWigH${YSG}Xc?p2U&&$eQDQ_)|PulS`-yXf9vO?=2OPv((+^|q5gx4a}
zbOV<HWKD}ZW^84?7Lp}hW}&F5Y`x3CNSd=fhQgghIR=99z2>s8tSb)SZX;R5#i>G_
zozBF{@#o6{D6mw+@nS}>=$R0Y&fp(2La*0q?+GEqC6CUCD<GV{^m$x2SCkE|bQJ7v
zD)Foc50-i$4W(yt`0|Qi4=<=4<A|#uv&LQUO72)ej?shL{PRVc6ZXG)so{nuLUYrp
z#lNp64y||Ev}y5^!e)&JQaOZ6mJzF6lcS8=oSh~?t|8Q-;~=iuCwE!@7Dig~?6~2z
zgd6?Z8l)EmS%%!~=Pu)$0*Eggl~LCbb{A&Zp*66_^?u>ziH#KULIc_t3H8u~?{^xx
zTt=O&s3g}ia&*i&ldvVh^W!l>q?%^|73l9pxGkgFESWZ|J-S3so9j>%B{i`n4#>j1
zoE@hs?0m7&<4&vGf?VDNH4C?PuqTOsHbIOmQGX(%cv+w;g!Pv+t?vslFD&*@a9y9%
zs0;AYCmP0kJ7{KSi^!Ue_Cj_TyUh)t@QPhbT&=J(KzYJr@EC}KxSp_@EOe@URd38=
zp<FK_x<`blyjY<9S7C<h_TE_IAHRv1bRCBaadO(Yq!r_bH~E8p3}+QoufUgE%y%<{
z>)YLJ4FTkWPs7+<VPayHUW+YnF+sDK6jMaFCDeGTg~`<|t}dJ^P3q)wTqu&F_M*qa
zLKl(WBqST7wDdmfR%o&-RQls4g<Z$2bkoJvGj4MNvLmlE-__z0Az7=+#XIlqfKHXI
zCKvNWTg2GOvNwrKnlXB{O}bvR+n%13d+iQL_hxp9v{ptUzS{LpU`q#oNlY^L34q-q
z?e)yN2xonJptDW8_zwiZhp)MzJiZ&6T;wj3v+9u|NiNwq%Pqi>0@QMJ)8QHsq}_ok
z|FC-;#%v~@SD4|@)86QNq0AEc2?pma1i0*6Oe)2mqO<m!o_dIOA#M+Y{p}owFyh~z
zB|~XhB(r5-@O~m$Aj~E)Tx3@AvQ~L339!DXVGWEmH?vN!{(x<oK^(3y<zbgsh~Gb8
zf&q7Y+$+MuLPeX*1RfW}hb=g`I{lx7ojWEzahXWR_S4$cSnMGnlBUO@782n`kUUw-
zUOq-X4CsREIFaS~W<p#D?#Z#wU*M7aJ9*mSG=cT<Sb@%aI{klpkxoX}D_Ah>C&b+F
z2vmnAr4zDm810yTqVOm&+<ejoxd-hNRZ)8kSk;RON$m7E1l2gTs$-?EMLCzA!L%zY
zJ_+J3MR%$DonLlq(}bJ7b{4^}i!?LI1v_f9{hk6MWp0zuL{AgtlsR+PK>0Tz76WRq
zmJxB4fNUv-PNc*;w8>=X6XC97&Lr_}t}81lNcw|<P?h82E5%~$=`F^*PZNYUoux~S
zB+_GFelA&y`-QrJwzId4x3Z7u)7+ztRi5E*7pRr>Gs><w<9Q$#f$gWPu2z1*X_R}J
zN+|NqqG1SS4QhB1E}w2I)2&|0zn8BW>}{8P83ab<X@xqzP^5c=IshwYJQ}=`r&3GK
z2zL`e9^8(v$5|pMbjOsqKtY$1IZ!u_B;?<{3TmFq({S|P8offYMvRtEZm$j%M3O8r
zFt%&{&vcZo!rs(`Sk`pm{wD4aL@$|4r@Slsyn^%oc<(j-pw{&ecPTb~9m+BoDAP#e
zF#+x(9q57WmT~7d^H{LJD)1oz=8<FS<W`~QEg&+K?&tx9v%2B!b8Qlbyba`XqAZfR
z{3=4S*}tAPnNs&1CDi$|s!YCQ=r#h~mT6vbu}CM62?dRpqwRNc2upOP?m`?asL^~|
z@KyMj0N0+>PRUmM?00i4xlh!I{r$oup~>2@JeyHm>fp#=(!cnRn$V^>Hp>Xj*0GNW
z4NK!E#<>c+u;`!IW5;<y%uI6iA5-1~Ff#+#Kctn(I`4y#cmX?@I7I}_--_H)%<}=X
z3oXA;TJUZm$cbVJfa3Zz0a>4C==e^gdk}rT8W+xjwrpiQZ~viB3v?nZD^NtdCcs}|
zE{c<o524K|=eQ^p@ceY%zrn<*jDxTLKM3c_W7RSAqx^erRN#R_w7;j_Yq{7Ly9moy
zJJuB8t4(OeAGNA|H>rONh+MnyQ}n_ogw5KZE=#>H+A=Bk^>a_A6zhEo<S`0gLCFb8
zYM$E&ciPyvq-$-p_)#D=ZI^AW|ADe3SfetH|4V?g<It@yj`=K)z$p_Z+>3t>;LLb`
z7QaYov095ggr|yRMYi7;Ce%=)PXyz^imS)>3R-OaJUqO^mylW6Scu1eL}y~tVL2!0
z@yJ(T&X4=AsJQXdjF{pR;pRNOWEWoxAZ=OTBa~gfhHSJ)&QJzpKOz2tYp)pYPZf}L
zS<Z?o>}C!S<lNG1*?0(hW-~SN*Lvq!A(X1E7|VW3&}`c`1=(LO3~`{B=f&*rAk2##
zJTj5K{JsC?W>7i45uss(>R#gq!kGv%E5Z0unEBL;JzRo_%8x)!k-<otm465!5gqcs
ztZ+V{q-RI5;QPrDT3PoWp)j)Qst(BA`KmwruXgN&_EXqZ#NCcKO{DYYmR))>^`KvX
z%p&`9ZDFz15=N_Hg@j;V{2x@t9nyl|p}+p-8+0#Patv=`-iR?ZtR>7}4kV}8Q5NEE
zBMq#Y8}V*wew*3fVHG>uMyW<B#rRA?*x^NF=GsMOhmvS0>S+p7FHB=N5T66WxwW$*
zpr)Ri(;qV7a+78BTu^Q?U3l%S$IAjpY0Q{Waq8R-kDfRx9#EM4x;ThK8!yT-<LRWt
z@9=qm+|p5bcg3GY`$WFnsvR?$57lgUOq!Q}$a6q@0Nu%(Z9V|<Y#BQtj?cn)QHmJ~
zoB64oU7D(Y65@QBrYR8XvH6`E<-x-M%fStY0A#CoT^Z-foa8x6u%)bF5}S8jsTsBa
zsM|uX9_1G4LZDb`kB>w<Kc?T*Htt>!$V{<!ln<EKgm|IE_##(x*?{rvLi|CwxYrq<
zC`@OX%mLQI1gRiYkPLPrh|^0G`$%Eu)}~Wf6z_`;1u{dWev(;RwChX0FsjiafW+{K
zP>D-LI05vLdZ0F9Q6TCqU0;tqM45T~P+(9RZQ1`QJ}{J?CCc4ust$+KzULQ#u-M%>
zE#_T}aF&B?q+n?cK`y?|)O1jaE}d2FCi=s@h{>{tJ$7HLzBriqXQf4_K1Bfh<LSLG
zt`_0in99Qc_F<6OK#X%2sgaUSd+h+HOl_S!VIJ#<%G@$>L`}0NQc@;P)juvRx_8d`
z;|w9rpl2%jQjz8cGn46s_)37q&EQyxt(I_n_6r#K#dacLmX+`5xK>0qj=@;cC*Pk%
zpdE4|w`87}wij{9QqXRgGUP3kZECXga3@OX3bBm{H3t+jlNGkCnPJllE6XgMLmZ@Y
zldcx03o^?%$-st2bXI7zRk-P2V;K;avWI<HYiujPoVV$An<k;Ng;BJ$*xh(p2&akq
zh`~lQ0;i94O^v@S2bGOisEk>*=*_7UggO^_J1WG70?Y_?ud|nXmj@vay&oL2MELJ9
zzS6hU3LxZymWGjQTTw1OI*U?`wz8KXv!%zdV&CiowIr^cteAa{u<Eh%N>Jp*kdDXI
z2GL0+-X;PO_x0n>aFr~~8Ps1C&Q>ohiBN=VP5fV>E|#1Y4aQ=t`V&^MF-&1Igok=`
zs;fh1HC*qJmP3Zqv4>!%f!#e9;@=ejb9(tH=34_I8-h?0?j#cSFn5cS6}Ajy^Po2Y
z?h)j=lQN4@sTK#U36@QUwn1bdzf+(y#qlIcvj+wEOZH>kj3%X6X)Q1q!7j2AYp)G~
z{LvyZWmiS{d(Nu4jF$l7a+9Dlb{fIIS>^KnCrKUFM7VFH&a?hH#K{E8we6aAW4Nyn
zn(>96XnGtcz}$Cnb(3CziO|980-9Oro6y;o<3wQ|4${8bN<@5>30GBfG8>1EgrE?3
zx5pdQnW9`F$tEZ5hU)Qw;EV@OyK1pOOAavVp_Usa!1-4=c9BcASh^Xhos(gFsS9xG
z4E$1^FcX&JW5I3}Xw|CGIT|_}Guff&p287Lh$g~%sS>{*ljq2(C6Zx>xM>q(b&FtF
zIe#vU_Q4gZD2>g(VdLYAA1??)0PI9*`LBplW5>MwpmBsPXiTrwX;8s^F3_UT;}|1i
z)$#fFjL3(HLbiHGCqA73;R>aX>foV+MdwdKoossUCp}y;|3pB`k9k%XsY}r>%xwl&
zgL*{jcp;WlT1VH{gU&`gI-gF-&tD%XYebx1(krwM_)4fFavqYdlHa~T9+7PmR|H!%
z>pw5lovQ=a6R{zrGi;fN_TpF36pz8Q9-I8Z1X|K+>?Oh<s_p$nWTTMDQyd-U1%U{+
zj#_BkM*diO@;F9>OJC#8fTpAP!^U8Zf>fqr>MH)u6H0BR=c(dtQCY8IcEx_=<LrN?
zIviolxe5Q93u!7Cni1=4n*mY=?C)zahX}K+7h$&%=?~cA;k`Dgxc&%e;gJgW3^0Ed
z<czDmtX<2o+2$ZFLMJ-HWa)ISFxD_#%+v}t1j(B!b8ML>Fj*VjzIaJUUS51z)pZA9
z_N_pjHxnkx8~pMJa!zfG-1G-b-h^eXGTDER*cyOBq8q`hw+R2+E^CRnTmTipmEc%@
z8^SIM3yw0r6OI+)=#+#r)GGy8DmX~#^mWotnz5wj>WdZr<U~i09~B2H>_i#aG~E3n
zq)|L<xg?-kJi0BY%OZmmtCF?@OC3_F?vzpVX0BNFiGuMm+uoBDwn$rf;vsGnO;t)`
z&xdb_bcU!<@GGCL1oUhVnza}P78>Q%LfisydmCRhN$w-ZkH3je3nJZx|D4C4j0x&4
zBH^fo_gIcp7@-eHfw0XE{)4<@ABD*S|Bh<JSBWGZ%c7Pj?g){!QR&0d60%*O<pT5F
zIv<OKnRV3AH5{FFG-F{0E{72H7!XF<+;`}UV?~I$@nhp!g<Y%N>}wc}cXCY4AhE>G
zj@g9>E>_r*qrFLqsom`}Ld{ZvD-5_ktjDv$+&T+5w3F%nds#wN>`*S)g}==)X2BBn
zn@xn#N_GqH3xUPaJQe<_!cGn|)8xYCv|WKH4y#MvRue%HTJdY0I@s9V02&iYW-39a
zo+X%%(j~0%yF<8sJE)2+MY`2sO+v>MO>1MRR_l%_dqBC6J*_B!b&+GnZ!qLWI~<G~
zggH@Nx!re9NILq+v2mw@E_2boc~0F6Bx|Rb494Z6T@fsk*$~VSz}NNOxJf}cm%=fA
z65*ukWlqP~Gw%)J89kZEr-LGfqveEWF)?VSvCO{@i0crYzZ^fwGv1+1v=fgG+&@*#
zPT7~Z4Jqo;t+dyR;{+oa<`A0Fb^Ae4PuTm$oeEO|)+jtE#zgmc_Gg)u$omRgg#2r#
zbD!)FMzLt2<@iNJ76HWv+BFo!1u#;_yH&i=O4zL++0s*?XO@7n0f?6$WUPaAa5)Dv
zJsu}hrwVl9*~8(rI9r&@mmj<`A}!Je+z?tisjmoeL?(Mqom!KBwHw>>G#R0g8C`hw
z4lq}0eUSNXwCKjf!Ojbj)#6HlPLJ&nHeE6+8;oZJyMUb)J<m~(gF1l_5G%p1I9^0!
z8`rp3MpHdvGPoJm0~f3Zt`p*v(xZHoYThm&b@_2)V=0AUQZ8!5^mx%unCUdzAtK(r
zggY4?NJ$F$q9B)trzqIV>H7l&TDaPvGKS-|OPHBbhh-UiNH+)^^m6a)EQS3AHY*<A
zH|qh*m}Usp8Sp7hh%1FjH#$?0zf5m}DF2z29K-oKFFVl+_O61tB6w}97Uzm~!`IP-
zjMCuga-r^X;u3q>UH~(Z+<S3r?<7GkNvoeZ`?iTtpLz%6)+xksjp9v==@QK%HM&I~
zaCY>gm7*55X-B<HEl(3MwqqZ+Tph&xDNPx7k?!w>x@NH9!-q<h3#AOOI}1maU39Es
z^Ghy=)0NT?e7sOvPEq}Eh6uBVt_LRC1|_#2vQ%2eOo-k4iQ@dTyUEg_Upy6*zK(}3
z`uYX~P?kLxeW_=5r_+QnNDcJHlhX)hz0n?V850&B%t01O7(sU8Y9N8;hsSE!S=Kt}
zSg~j`F4pYK17cxSfv!ldM$}{T>Chfxr*cTxFVdOf4!j>d(jc$&8~})T#^XWduUQxy
z8w?=xw5AoS)9_~r?b_Cbw+ipWopCUbYa3M#1{N{u5GOX7{<ww0ZZG9-)MT-zfK0O8
z*_$GC$=weHmfX7WZyeW&%6Me%Ks8C#&T4`TDQpZ?BikKDOc=szg7jZHQ{O|7%g@+c
zj<ZBNF^<3WayeV78w9#$e6t2e1CIzp2u0RFFDmSU;#~ue3RvSD4&;LBnK4q{O$bc}
zwRK0lCBoTG?#Faw;t_dt?o;UU#~Me0P!Vn1#bo<(zes119xI@|Y=2A>;(-yli>wOD
zW-593WHp;s^%n_u^Wzw#v6g&AsCks_6-jP>Y#uN7Qdx=VLMdt$we4{VH|lgC=^kzr
zL~RtSRgU)mPgGVLa>|sh`L`bjNZHYcaF^!zJO)0@87Oq{xQ9U3G{;3+Hl|I$;wJ!`
zMV%tCh%O4riil0AtgB?}6h{h&$?j@7UdqBP<0eM&MB^<O$<0vM%yY(?M%+6DIRl=h
zmH+2;Ps$_7u1bw*f*{uxqYSRpno+p2K;Pw|GeJ&g^r)7xTs$iX$%e%KVlu*`ztx1g
ziH}VlqMm7f$nI6f5Ti~eidq($pUtvxE2SE83b?r_^`iwnP_#<}qs)CgD}O3bLtKo*
zJ+ZiutOr%3=V6qjMIY6Sh<dn63p`l}xwed(6xW?bnCeP;fh$jkaJJ}CyJN-~(C!S_
z{Hb5@=K7g=96Y1+#8lC0>-5kVa+V1un5nUl2y@u!>G4b<WXxG({QEcm>I|+F3!m+e
z(Nq><7ZDa-y-MD3xd8LV&Yat9e}#7Hyb{Be{&-S|Kk3!^UmST30Ia2H^eGYk7h_vY
zdp`Uvpym2cO`y<SrA8JwH_xKR=)gB4&I55haF~k1v^z!$awpI`TDTfw^?IK`Ge<X&
z!{6fhAlYCmgM5+ZrdH7Njwf6IU{Q0xhxIV+{~}?`2)sFanSxGmGPl+-8Td{}R-WY3
zWyIg|C|DtA<^P-z#La8M@uI?6XHIkWVC;V(kU3AzcG2c_#koz8t*M+{GFIFs(7Ds%
zN4JcLmW%S_^n#?i#f*!AkT%2O^mtAL#g$VfRtKMmbm>aG^(I5`IQ|kaS18J>a`avb
z?M!&&QBC|zA^xh)D=TSS|F(&-LPApPp0~X$kB1vPY353ChA__?%$=9cfO1|M>|&!U
zppgtNOJl0S@I6_c=57<=h*R)@63>Y?3+drzc!)2Wm{eBje81~T;<<WRvc$zAoy_P_
zydiRnXiJ~Q$P%{}k2k;Qz_)VuRm5=Vdw3#=5=?0#?U-+Y94&-06|mN*#NEyR)BPus
zY$CcRf&tfDSM!CdSeG}M2iXm6E6`=qlCeV%=MK9D%nWmwj%H4@hKLDcW0h+Odl)Wb
z?vzZ=v_%_ooxkB~cDgpxBHAqyPe<^(hXD93Nr@7}{to3O99j`=xYOpS7PHUf5Au>`
zq(foLQT_#%ydKhQ75f={7P<l2<w3p3@TRGLGl3cJxHIEfX=w9-c_fOilLc6|$!UEp
z?h#^M<QGO0rws&ezL7sH8@VCD9YZaa_y-vL>B;Kp3j2R{2!3+s5n&$v+nCMpBCS4i
z+)bda@_LmwT1n__K``2Z_CoSm`DSPj*gn=b4=uL<yO8KuIepc_bF@HbjI9_quK9k!
zTftmVUWEWys0p-kL|cuO1-RsCfioD}2x+v$LN~Tuah)JK083uZn?z)1XVq3N>1nU3
z<~H;;{xDa%bEBz+qaJ~XrdIBXi?VRb*zxhY!sZMe?)VWg<aPkFUqVGw(j|?AkuQ@v
z=J<CI!JM+`l|#Rh5Lc`2(nhyvi&>`>8Cxsf3Cb8btr{H)x@b&vJbgMCbwTjQyIFCw
z!Y%+Fe)S+4ONoW<0?e8QY@nwZEyUT%&r_Tt+MH`Yl1^Ov?gld}B!%L=E(5;@g}Tu4
zir*ctH39U$w24#hAx73F<lP9az7XW(xq~IC@)*p$c|x>TW~Td_KygsU4IA7?3^R!|
zr$#|E<-}cW9H1~oVr=T?z9%E=Uxb*k9`t&Zc&iDdWm(LBzrSxqf77n8)24xU$6)p!
zZ!6>Zlm9dVB~|`?4?tOX104OZbRGI2h)csoYdv0sTLNL0JEKjusT%la{wc<#y4rfK
z5X*+kmUwdGakPhl%(oOx7=#OOQe35COQEYFQw72k25eAwipr9~)rlOfz9lHDQ9bp!
z&?C?uW3We#KZ+z5ZUbt|-yQ7&oJ3L90Hk_c6Vg$LTJMX3TzKh!sJ0=Gf=~%9V<yJs
zdAKEBRM^GnW!p={$MOuAtZ4o?YHdP2yb3Feh93n{E_s#2cf;dQ4avfSPTp1*M!~rE
z%yrLtY%LlFdYFOEQaE)kroaD)u-x>dAPtYh-{DW>Ag4Xp+pC3En_!WpqV(paI7J}&
zB<ma^t@I?c3q<e62>&nvmIdpBF)d@`Xd#lt_(^f)Q-m!~Ic`zET<>WhcOdCW<kFOu
za<x!b48CsT@6AX!OD=b(C3?*B43S)$9Xuu*3yN+Crotg9+uzc2U=%)WVJ{SsP0cLo
zo)qc$wemDymS6XIFc+v))>d%~0gNg*zu)5pg3gc0zYq_KCT-nbjU`@$ppMXOaszky
zm!K$gE0g0j3R@)II3i*vyT{9UTuzF+MVYIXG2>$CS0P<PR!(R|wTwWg(d*sG+^>On
zbWT#s=g^Jx2g&JT4ztCRB4LtdQjfV`hj58;bxREPc00TQ<~pfyNxs_MWyrbFzSJ15
zRU{X^9dm=YuK7_i##eWHlPDBf9v#+ti?F3AGmLutLx5AnR78)lyqF<vtO)EuZ@IUL
z<AQ6uY8#ks1)5`;rQG1s*x$)huN0=EFqr*a0A%lGv8^b}vlm{r`WGYuV6{??+eEmQ
zRvsJAi6SN|de*J+g-9e}r?yEa)c+n3g<?e*$0+P{b)-L2q^E41cj?NaPX8Vi>{8$w
z3B|9B7Zd1Guq@TBlfn2QOS%j5(i7izHht`dE>+lZd$<IhW|`ZBS$qt3@;f@)2OzEx
zE)x&N??h8d?HA%s5sofhK1F*<h$~rBSB>fB89}NqEN2#-Mc8GL^+%~wcihXdjbK}^
z$UIT%YN|y`fHU=vnbd9Z0X^5g>W5%vsiH?g{w`Xzq8wwce?vJ7&U&&rn`PYpIZ{sn
z)7RKk6s4G|?)ATqAS_9CiDa#gu+GPhAqSSxCW0K;6q3V%qFhF9Wy%9=9F;|2Y^69i
zi89kD_<acfnlSYinL}Xs@d+eCsn<GV?N15A44V5w93jGm>#CsTy01ysbTQ_$EHdi2
zO3eNrB3u4kt&;qQ2;p-Mn0O7bjz`fv2J=eyAkRI24r1<;Apmv!wjlG%<!^4%vM4*}
z3otjrF6=^{Xp${0EwS#G{A-cPS0(x{bglaeq#Uw9j)z6K1GEpsTg~`V8?k+3(tG|Z
zqPX+cI&>G|?5_>LJg-vF)#v-B86$0mR}tXKC6{#mhbA+IeG8oRdK6mh{?=|H=&q)-
zZah>-<5PJEp@G5l_kvwcy<fh_caY{uPb=2qH~}sxUdm{$(r?@^&;n!mO?zQk_q;Gl
zpSID9d=HT|5yYvE^s^w(KN?5LUu*15egHLBl)3|-BdOUZ*lfsmRy8(NJeQE;CG~<W
zgwTT6GsF%Gr@WcO?-JnwyKLPgDgrcK@h|=1qUyTwN<Tto;^$OsFg6z8JlLD)jg)Ck
zghwglcW3-h#G!qmhe^jMJ)$ihJ)hClDXZxR1v=%f3WhQ9h=7KW>#@kskQN2w4g-Nq
zB4ZVymYE)!uD0_ks33A{c;ebav^m7$LHdT-e*w|pLbm%UY*{iTQ7`ffTZ<$9=P0cC
z;$nr($aKiBMP?#E6~|q?Ujdu}WzkNrI@`wXs4PWqSC}@f+m$SqL=l%qePk<4GhA&B
zBgqOA4Gg!Xguq@I<A|;~2~$b<|LlzOMVOJ%qgIW3GqPMkyMVifx$=b4La#3_65{w=
zhmWP_hBV{4#9faS<^izqu;ypN<qn=8k0h9RjE&Po^G%78rki9#=FJo7OGD=~^8sYy
z<%xNE$Fo9QQ*!mp_$rxvFVvNe)d|`NfpgBEN6j8w5!yQh)O_{I7@I1qECA$OP|mWf
zj4PT5M-y7tKDr=rT)uARCQLdO0zmYfcyhyhswk33FZj}Sy5mG4mQS6<-svKdJCB#d
z2MU`7?3lUS+Y|FI?C*6xtORQ@RERUAcJVxTrhta@T1LfN%}<JbESKvs|IqvswoC|X
zh5(NW^kuG>(n!7+Mx|>BA7d7QFeig@je>mk5M*J{2l132$2E~6vlIPtRuf^VJFaDX
zY_e#cljcjk;1PQZBPTB9wc}P4!d;<_<&4G@<u?0bz*#A|FkOi`1bD1S$E^(wN$j9M
zbTTv$mBk5~YxEE4pzS<C7S3d)d!J~hOZ8#jHs>&aOcXjF)r`7q6Ju=gQtWeK*@s=S
z#G0{0{$Vo9(VgWhmdqpJH-MGP`vRQ3?p(xwL{r;ER02IqIb6oo+A_ll(`bt6J=MoV
zHCnCpC}_Z!26W*DGB1J$3L>>uEW=+_*ww?FBsY2=3CTFd)}RmzE|W(o6tVCr#y3p_
z?*pov1(zj`vu5CA$)l!G5@;D=!;HBV3#w_q!Pqy}tIR-03o{StV7@2*C!|p$ZB!&?
zdCN5;$x3Q4O3UZT_vnEhb*oE-`9s~k!Ov36_B$Z6rVDN~TUI9tbRl64PbYILR(34H
z-7g+^@z!^2^J_bTQfv4@AT5!VQz4dK0m2;<v&L9KWL6m7X=mO%NQje`K^~9y`F`-x
zD}radY8}Ta>o`G{7y}%N5sGx8Ftdu~4Gyh_tOU@QXE}3|2Z&M>ggP1v2adRxSsB25
zQR?_vq&tl6kRvZ{W3|TGs{mU%_I;xD;xz*Osa)ROCz4hsm!5_AUX(NH9@M&E?NxJt
zXD+DE0|mGgG;y}44EQIk2IeF&dy0ESrcncJ&>vTaaDnuY9|t0u_-<MQ)O~<Phl7zd
zp|g4{bnE^Ik=7KL@7=PyLF$Wl>JM|-g9jIu$7|)uaUfWYLq$6|*~u1o1gaEQ3UrlN
zAxJ`kEH_Od`0Yu{g6{;Y32^2WE3FM-R?(=GV=d86oBbEI)f6DE6zGQ0Q$<Hok2{1&
z2r?i2sF?<nSxbD^j9%#HXbo?WBOILtY$1mJKUL=)U|CUgaX^tENGvjfq$LR(*(HNW
z&Pb9VyqWiA=1q7r%$wQSSwP7dBqw3X8Id3Y5+zHPAUPvJB#QzPzTfH7@813X>pHh@
zhw4;aU0p3g%SF;ruKrUfM)bIeQ=+D@YmNewC3^lf3IHW1{Qj+g%t<6w*_~mmB9OwQ
zSBNiP6T)ScdngiFer%D*ROE1MBuaoyA1B{`5<mj-?i9C*a2H_GM{UY-@{%xfk}Y$+
zUHWV?{zfo0NxJknavYR%CR^FIo+E^GQF!w_j0jgwTCZjdjxPZ2523GlNPs(<-uQe=
zBu!&qccU@;1k)^x9BOb&NVHqLt!9<`S%RDgzS-%|^G*aXza*#WehbQdfv#bbdz<MF
zV5czm1FmlJIId`lj&`v#PM%bzqCL}y{I$+Kxd3s;I^2hKFHa7YAi@GaYm>Xj6k@ud
zD6%+5;qtg37oiL0q?Ka(wZZb_z&1>0y9f_&uT#WCoUcvSBsTAQV6G-dLyVB^1+XGk
zJL4P$Dfo0e6JT6hL_}LpPmc8goNovB@#L;F_7g@A?8O)C7v*;}=BT{ruh(u+BqHZ5
zqp{Y8MI!3MD6qIzE{tqDSwG@!5qbQ#ca6jcqG?#2)U!ibdn3r)6ZBYb+$6vaUqkcq
z7m?z9g}HF{jUilOZN6bYmsN#10a<DCz@32H^`z^@K2->n8e!jbz|RPq|Gbi!+;80^
z$R(^`l@e{66bW?J<xeiI65>2i>P^Bct^hi_tyBIk%0<?}{(?V;bb73nSh$I%3LM4O
z;!Y9HpGV@8l=_Y!|6XCgblIjQiI<>RHuK-?!}{WgOt6AXh~qNRkg4tuHYb{E&Fw37
zjNODdCk*=JFFWoLL`itrtR9b+e=xgsp!Aq^3qI!#p<S#}*<i~e739q%!fq_cS+RNH
zd@{O(z+hZr#5W4N_RN^Qar;)#9(9bAsa}q|w*jOe=_q=!Z6TbkjtrByxvDTC;Geo0
zn~QKwhA~+i7#NL4`Hv<lHqO;JH~+)sN?uqI;X+ob-Ljhgs}RP;aCf|}AkBxHO~kep
zw}VQv?$}r(hq(Y43vXW}F#^M2MnYT=z0@dsv%M6F>W&}lYcZ#aCb3SG>09kU*j1Mw
zU_Ugttw6UnkG#c6BHgt(tYG&w>y7}F4L@u&>vU0$J%k5=w?(=D^6Z9^KmMKLr)%2r
z!<~rWYV#tRr2qAWn7drYW!NxX{i?)dGLcRwe=RK64Zr*nS&k^7<XH~r0zponKBEhn
z4-0Vx7}A_L&-V)e*IFmeEExzF(*?Q|NanUpI!Tz*;)08$i$!)WVzCtY^mq+nj)m$-
zCp4D{p>OC$L%hBVL>~SqY?|?o0LMp!l?m<iT^+xzpUL)a5tcNxW|m^yQoH9^y6SL&
z2&W3a*xKA8fK+>sIdnXRd@Y2A?m+1@=9h#u<#xxm3i{6xR<_YtY7fVgCxQ<YrZxTW
zLCmse5tpq8Yh(gkcFdvb91Nee7g4AyyP0~tBqBEvE=agR9SiIYWM&tBi(?gm{3M6E
z(f%t4I;hO#HC7K4O}XnV2|w6}a9(9_)lf$oqm9=P>QbO4lu5veLR_gTLy0%gN`R?|
zd;KTuOANO!1H*Q?6DnZjgp?XvDol}(j(I4dt%%%-*QA>K;XT<7=87O`N}@n4Odu)f
zRoJH$!6RXSdHkxXNxiz-&O^8&UFIPy2&l!oH4y&oK~cS62Vs}Lljh^Rv;;;*CG{LC
zpb*A&+9|R0JfUW|&Ud6&x?PZq!G(xet`m}6q$a#2hsRACAJ645uR+ISUBsd8>AAXf
z>c#&Z%$6I7G~eXeKDnA)wA=B~#tlYAL~e7;vUIflmLP9Lc<Q?cM%i)Q{XO8GdD-{(
z#B-uuYw0rk<1GPBt5U@|8);4J1<KnA>Fo1sT}KOcQfZK%P}1lcxvo~yXshre(N|=}
zgu=4DivTw#`i0!Ft`_J<8j-GXD8@HHnC$Qy#ay~$i;`pBI7?KkYje_{UQ^j%05JDN
zBr0w~3UHbT$vStxT8L}O`$@^ib%MyI*`uA$@`FS$&(n?{<CVBUnA64B4ZYVp0%%wi
zCb3MDaI15W2QM7)%cDXyG<zAEuZVzQNH9o|jktHnbaFI7PkKOf?ly=7?~8QZ*hk6^
z7r|}fZ~?L)>$!6@%exA6{gNp>ie4ePQzG>DV3_oRAU9WMKN84XQ=!vnqnrzhaOt_~
zEn^F9>bDb0f%HTY{onwRR2Q@LcqsoZb$L`q91lSo<AkV*bWX$Uw&FC=xiBoIT*8R{
zX+X{o?GjQp9!R;MDAW-V8XCHad}IlA@Y=jom1hca(;%^X<9v=Va{#T9Jn?@Z#Dz%L
zuD%jY^Ge#P#g9j!a<^A~<L3feeGxNw%vna`Sb<K1ecIabmy+()W_hvce0Q##ztX4U
zOd+lTWz(T!%)UQRt0!^9QfGcmn6q!P0M%k*8Hs1GhA<n!=1Q0&Nu7&?DZu@MJNvQU
z0g$P05&19?>Wc7xj{^z2SZpSB%VxR|7fbI|vQ3L81nF-C@OVxH6=nOJRBK!R8roU+
zRpjNjR{+VQe2bTVL%5g@+Up3&^2YL`gMiJ!+BD8BQbOYe=AtA|>#>s%7e(p<?O4tc
zl=U%eTGL^MZnZs-f2N-^?MP1fcmBC$YCzwg_2A-X9mF!vdrbU5xN}CO+DFXugt$^N
z02+ux4*|%Nhu(YO?QoQ}0yC?0Q3Y}QAwhH<x_6^H&N&ncR`4xuSJ;_wib*-+wG!d<
zNe}33bXXAwHNU3ob|J7{K0jjh!y#NMc#lAMngFNCg(Yc6c0U4ycCTvNjIu;z<{+*)
zYiy?GJF-Xw=U!~I+e(0wW+cGjql#3y!BS(Nbhr>o!R*6Ps|fdWG?$f_d^B{<Lpo(j
z{cAz)$4FwDA2>ojrijK8K$~~Q*@E0jIZ24$iFEZCs)#Rl3vgqj_G&4}n(<bd6f;mt
zJ&q+Rg<^(aXTr8fwA)KJh_yNFjE#>gGGKN9cuW9)aWkh&+P-;@ho-$a&52HhU7AJ(
zJIa_*0=$38y#KlYGXhmKF6K}0-{U4th+}_C7>=?3Lq{*d>_|o~dXn$H5|c$hdb#Em
zVQx+)6$vaICxYZ2*3Y^?nYTR&$hp!-Fkli*PRM3l{}<u-a^TR#DZpMQ19|$?J1&_2
zOA<LiXr35lBq|m3N5Y&v4)W!S_J2Z1n_j^)q^Ce=zcUiEoLc;ywS!#16Xw60SSZlH
z?-k<0==v$guyN#RK%Niupf)=zD$diLrx*XCgp}(0Y$18lN4<zIFF7hbU#L3_CLyu%
z8IWcL>Wg}8D%y>%Z3rEMm1uK;nW=O|D$*^)X|wm!p|1`kPY~+T;z<E{qAMN{<Srm1
zxUN|8OaS!{SiOzH=Aa}JWQS^8B8(r%y>CEleq8_z)TNA=^(=p7ZlFIhe19l_<m51A
zi5#9VacqoLn7>%T`lBrq$i)-m&mv}e=!CV9b6H(DS@~?fxHvjn;_@eV%7iI*UAXE0
z9fa%0n?R`Eur(57*63ibk9#CS@(^TfH{x6YIa$`mq<LQ8959k)Dj$rx2#-<BSWxSX
zw%2jJANL4%TOdHE8&}OZ^<2Q5A<GmjzD7u~xMKmDM)O@lT?{;%uq%600Gz6#BVKAo
zk%&}t)MHm6?p3%WQdQ)RFq(p$iiTJ-D(5+s8ZY1sD9l8i)SPjVD0eI#35m-@npYg&
zr8k=9Iv>ad)iFP^qP8|G33Y`U^&Z{3xx9?0XWFC^!j}E;3y9~Q%GpL=JR=$wqN4AN
zw?$BtG-t=67eY~0j)LR23g@26)ZE^c6sBhhbz*%ya2cQHKR6wavo7LWOYxmwb7v)J
z#(kIY!OURVU;=ue1mq{P60ZnwQ=_?Mmrc=rd?^^^<kD^<ju24{FHZ8cCarf_0ix7r
zP|FUaUm!JNCGO~mH!g=VH^n-g$bTe=O7fIU%yossm7e}sA&2`oraxF=5@D@m+j)m5
zCz1@*=z71r637K+1C4w@&U*=?z-W>O;)SaqT8tW;7VnGZBU-(B%zib5J53U*(=NOj
z4Z)cUFoC|*jK_tUcQ{dH;C?57gwp+k*{^}fC6nt|*!EjN=9HXQppvGqoGg$!_YU{O
zSqkS|;BDGeob!hQ!+npnWp@k91E$iHHN_Iwf;j0mU1v{609#!L=mIwNh*6v*pw(Mp
zU^)nm^Mz7^5$@W|bv<FHTwz&YwOv{WV*t}7**C8&vNcs>Phn{(#tC-H>E@8lHr5qt
zPB$fYY$@9HvqVOM`{y`mI-vP(6-}Y#mW?H-J(jxxz(pEvbVweaD8$K5??PSGCOVCj
zVa{4RG9_7SB#sv4g6W11x)M}_vDF{_2XR)I5<7|VWTm#%Rjy=iv5Ro$tgDEWWnUq*
zFj|ggoFpQ5G&JgIih8%i(l$|+X>B(W&&6fJVSW|u|L}QT(?za0yTmZtlWx8mj|<Dq
zhtpqtua%MHQ9bz*@B9;yav9Sjc5%{8#cz^=dgF5ev~Lxoyr*v_oVy?sxw>UD9=iq5
z1(cCz{O`}uq?w)!Nu6-ZTaCakuH<P%og@TKB&To--)4dnk^Yz<g6#Tfsm*h^!GOL)
zVGjs4W~{BvXusW22RW$zv%+pDjv+KQSl#2VcN8d1qq6`Vc|0UES6Wt#jrjGQAT6HC
zcM2-8`-GW)7zNR_{w9Dj(>dGYA0nK_xJgN@S^KU6Nf&}f<3J%)WjgcDL*>`$SWRGM
z)c8vgzcwx1AAd@S>!1@G#0p)8*zIn|E3A*=PeRm<WcoIG4@7H3;X$}I$o_=`4Z&V}
zrnBN6c`qbI88>n5*y29It}HuG?Va#=Cd}1I%gbnt3Q=u&uQv`7Vdm2%(|U1e2{A>>
zr(zr}h&*+2$z*VbNT;dc%3k_wA<nrC>-nU*-{pQVcPYd<-p^ql8`lb?xXdx}^9KmG
z+Ci@u|3d)%4aI2EcXvGq?Nl{E(=(cJt}yq>e$Kh#bJ3YgS%)Y|YKOl9J9EYi(}gq~
zlO6(eh69yRR-^p|kQnyrX(H}?7}^zU<JBU%M!yiZT;c4N<3gb>Sn|!ON&HJ;PP{Fd
z;;i-v0DQ&?c{6r=)L)BBP%jAJD;80l0zOv!%eIVl`BMRI<SGi@p2s0`ereTY!+M!8
zm(_>-F~<`iBw4E<uxum3#bKmUlS>3R4UW?W(0g+9S4Ki!<c;S&|5hZ5Xr%+A^PVga
zslfQ@$)`XvOR$|y`n6iTBGk$2vPwG0d_$PqN-u}Subzf>qexf4W{gSZc!7?D5MGb}
ziFU!e`qeAec&3QJk#RLX`8zbcv3*wj<XK4bffmvU@diS2OC-hK#OI0>(&JvQi*}!3
zoU~j4#211+7UYz#*@6GA=Yi<1IK_(JDC}fWDlzmf7vL(h*6QXyPMBT*b7|nRw2Gey
z$*I;j2SdF3Mi4dNCarqTUN4sEFf$}qEf)&&Up^<S#_NKd^Z<8Q=;Z$q;u7hRlF|6+
zr2;?{!2s;ZmjPTv?ah-Dr^YK_Zok^pN;{8^YrqT&b%x!;_&KM?GzB@s<|r4zD{-n2
zXNV4_8>60A3xMesnR#gePDT$i_h1RMvLLUb%qN;6wcHNHAIpDPELb=9e2s{ici0Qi
z)MpD}4>?ke84Bic-tw~w@0NTW*d@Xah-b?iB|xKLyLjUp#PCpB0sFLl-h`yPaV;a6
zsNNvj=~8F7&k+H8@H++PR^x85n7hqe1td=zvSd5Bgz$-+Yy%$=<m|?cn;cJw)R>fs
z%bOxxSe^Xw%Jx41Tw}xw-D2HZNY-I!Gtjz-H)AFoqp(80TKMfEg(@d*v|5i43ee$q
z1+N#GmmX=@>sZCtV)z}<+)PQkG!)YWx!|~ai-X>UbXjpuiuURS0WhwWM9+KBx$GR9
z(tZi?FX)>X@AoT8P9xn8-|wGJ1973fAN!8mM7tQsZ|U%U&i4z5i-;&Su_+PcX#=*S
zW0xJw5B>$1I}7g>!pymZz-qJ)6>P;JTpPVoGC`y>vPJ`U(H|5EjH0$gi}#`svr12Q
z;F6ESn%L*VBBi!I<VKO6uCPGDdq+v920Xuv&eOzku}GJlR~*@CYLEUvj)Q+HrquZH
zN5pYeuueg|jj=-93(#8HL|ub)>c`+zv72pZocBrb8;7;}yvnBqQe}ofAlg!hOUnit
zYt7G~nGZN=K|{emboDcY5l2ckjmw)vn^y=NsFk=KDaf^uDVDr3RAYgE10r3^pKgEL
zC`t=EI;7VWZjFHnvHs_fP7q-}sn)83=ozF&e?rgyuQE=ulc$CME68&fx;#G_{;U6h
zr96=`;to;dInv)B%YH%FnP4EAyeo|eGGDn5uJLw&kesB%a2-8-F3b~vN`sp;eTnaX
zS->cZ(6(A{GP~fM4R3VFJ9w`k_wuO~u9!q~rZ{Lt*c``kf+*=!E&;CozanJ`bZnI5
z%sX}y?55&ELtFewbgM#k3v-7c7r&2_y>yW9xG-0ejk-K1%>6ZpxsL&2>?hL2V<QI>
zP7#oMusqpaDB3A#g=fZI;2RK4S`ECMz9|BhOPm}u<x(MrBd6kM^V{No%+_FP)P*=p
z6fCG#ddnz!au%+iC&X3Jo$I9Ey}OLnWgC)Lzatg}K|$6N?}&hh6PU%nQrI0*v#$1L
zBXP$pKWk~b(p!A^Xw6;x-};hg5N2Zb?o?&jSpi8L^U<#O+iWIKWAI#;2y>{}$J@h8
zeh=Cc15c_^n6zwPFVw7%)v{ib{iC2(k3+@Rj7NpIo)UTJ?W3{o_X`XyvP>eH@pEBj
z8CD_e>9!L<={n?N;x!R@$(GeBM|_ld^6bSw*r1TeHYL#H5D7;jKOlz7!wY^gLYpqc
zDd`MWyC&kDQpS^l3t{ITAuhKKR%x?M<^W3!!|-Y=5l*N%+}@57K%&e+Lj4;Zp*LBG
z^v~(vaCSH;hD6YtIDBA1d`(p54jm(j1<T9@W=_avurGEO;Ks}H3S~KPZa_0%6N>Ww
zL=df(t~U|oh#waJjnvcY<4lpa2+XN-zksI}<^iBrVT&I#6sB;Tgd@1Um#NWy_A~Ke
zrdYZ<+oC=%F<}LDnj9yIaChN=Z7|LjZLTxlNpt$S5SKPx8=@e6KLW~fhYUfO3N-m1
zBA6o7>(YuIQAX0{O|qfp`SS!j1CD968eX4&BPHgpRp;Y7rAz0ATZmvfpe>|y_<JTv
zuaA3!Kb@biW-i*dA!KrPb6W~@g@&4(@H{U%w`!%gUB^DSiY|U^<27ztlci(@f#z<9
z7MHlB1RAJ28*#k=ibkgzis=hNm@Ok&gY~E&XPumW{%E0s*5i_~oM;yX$AI)y8h*Zo
z!DvGH1=&Dh(wAr*y(03k>C+OB?6i&&>MHB)dYm{>u4D6msjVMvB_dm#OM~c?2(2G7
z&0sty+J)BJzcN2~cM%{Ot*tKaT8l3RMRCSWSUaXEOmTSLwGpRgLWWrJx`@mwuA4Pt
zfyDt_EG))&8cltAIf3d9I;4!JM7i?$$qkm~CBV!u96<EM|3tfZjCVQ8iqR#3%m$jE
zs@)kg!OKGZ<EXf}{Zi1*f3VhaT6&d0M`G{L6|XD}?X{)Ku6EuZ7ysxQB{M0V{m>E|
zZeT#i!qFLr33Rd8<>3@%D2@~4v^vn5JS@_k8l!TIaAVzNiht-W+}az6i9acz?%_5j
zJ<hiU<sOWKAH5A%HE8bgnj^K{La}Iw&N^Gn%QC^+uv|JWlSFx+^bXrEM`V|b=Skxf
zk^GB@UyoBom?NBr_QYb#L#wqXj*E2_Zn4l-NX@9N0O&?TeO+tF@#^>$!AOyH4>h0&
zr^q>fe_SY9DXu;--pJn&3b`QG8|$uA#KnozKx`zMxcE_uLlriQ=`A?Ks>YuLX+mS$
zJ3Vd}We&3}>$vzOA$efyQ5R`Ez7(cCS7#hNh9Ifq?JaIqIQQ4&&-l?2;3^BJ59hB;
z4CjxHU7E}95#*lN$!08H=$N<)DCMwYY4wgRg_s50+QR-#G#O808h%P)8UxL^USIi7
z85P5cwPNd4{h4{W%6qWqi*zxO3z*BU6hP8!&!nA?r$obg&gm=ho(OY<jknBszqcBQ
zo3PQ}NXNrNae_ccL$=iW)gKCRU8kZ<)d}r4WhCC!(H?Tb>O^wwX}{Knvr{Pb@r9(=
z^rw(&?THiOTZNs^0OrmZVl2HTNN!W+DyD&$B#0lhKJ1~e`G!`W2V+W_<J{5sp!~gU
z7*kg2zx-H!X%&h?nW5y+eVS0Xf{A7`CX9n{_vf5DHW!&YNhLpFzn?I(nfuR-qFKiS
z<QCNCWC_u^yI|DS9~VpjAP4q8&A3fOS%_X{V3w^#CxYh6)<(HGGZ>2tq5~(XJjRQn
z>PXp9SGYB-5M)%hsrh@e(jK2F?8LO$(LVC@N&a02iI3%QKNl*0zE+VG>%Kjn?I6s*
z+fp3wT7n8aM%o0WPX37S++}Loo$4$#f#wGn^T<z;)T3Dk)}2h){O!iE>xLqo6}O^T
zW4PNUNb`eM+V!VE<Sxy-d%dxP0Cy7(E_KRtoDd5WoFWdlo|L#ouzAGc3(5qf{rAhC
zc@aQt`Bn%#<}MHV)gP{1{L>sxZ(nc_D$FHhMP#1ml{rC_kZVx&ctC_J)S%JEtm|(G
zM-yTj5iTr_b2aIlC&cxUW4dNsFMt9do}goTB~uLO$?HVVdd2S&C|Q5s6GCj}6Evqc
ztPe%j$X!~*hBkmSBjh}YOEXEi^)JD#iJ2pr^u${NU8|}Tjj&?IhQw2?^uFOoB3gZ%
zqQ`GH0&rD&?FVY{jX`KttOH|nh0O%~Z^xM;9l1w-4x4eA5a)*^Q_%)?#?OH0CCDMk
z+HI~){1aY7x{w%X*{g))aU^TPct?Q75teOO^A}&a7%(3F?WP5QqaKbPZxY~QV7;~$
z((3m&1EI{kg&8j@>>{<}*afv;Y`l2^F^jipX4_m49H&K^IENEuULs&%Xru|evIU?A
zH2Pp{G-UA^Ckllnv<<Jp#+f21`t&-zV6iPB)Qrgy&HP(IdAMNdT8*7VyV&yF#<X{t
z5HmvW?+wL5Tl<$JVX#fuOLS&$@|aSM4nZ!J%q!4>tE}^G1L{$Mr$43+=GlXVxl#-h
zO#(zrVE(-y8I4m%Mqfgm51QHGUe3I3(ciFiYDaJUHHYz|7ISV3o+km>qbw%cb>PJx
zj4W0Z;G$3t?ID{nW;-w!h4m`GWxdY!1*naHl_(ntG?UP#skdw`B&RprY){*-gn39<
zW6t3%8o)k9ZB;`O>1f;|NIip<@p*+^3LGYN4Plqn7EA8{==z{<9*8@3gl=_C`VSWS
z0JJduxf-AT0>U}y=9UhLSKk@P<&yzg^00ZJFh|2+k#W$7tIKGdU2uB}+4sg0V~gT*
z*uMxPyEJc{w~GmOB3<!NCODj}#G$(uxn}c>X53g}sCO*(eJPCOWX>E%>;_@x@}_4E
zZ#K6IapAR0;zVgEKHMG5g@+^9%vWN$J%G$P4j0(dG5BMQKywbMWlHQSnvp?MsHfc`
z(mib|D^k{4d@R(7!znsQXZ&JMAUCG;9+)=tYJYj4U<Q*+Zg8t1Dt9{`Y>B%{8Uyco
z+%4KwLve^xop^PxB3a6u-qp~C<wtu1yT&Lg@WI5Z0xJn~_3WcZtKD`pfywSpyP~>4
z7o*b4^Q#Azky!BCtiJkh8HvYur^L^HRb+`_mmaa;&1_*gOH3WoRz6-4NMo_cjEh6|
zfhZbFew=<+BFuZ(?Zylt<|ZwbHrwY3$_eXnDYnu)i&XyEUKwUFjXQ+6b@Z&KlsL=-
z2kr|>0T?aSxLQOmKxIU}YwpRP@nXQ`-U^?+Ht03_!$tBdp2-JeY#W3JkuFSNaxK7-
z^=j?!%kKym+JFi$Uokwa#Gk9s&YLsezRI*%yjG?_cFUGzFSe;Gh%_T`B+Pq|JW3^q
z0wI&KsB}P>D_FM4DHQVE>Yae@c-gyOa`9XhY}POVO+|Q-ZQP?m^IQ#EaVOCQK>xH7
zcl|n)n{KV>7U6^ul68JZ&%-U2V0S$(QS&@99*U0X26JXK0(qsi{dl1+Zl#?IjcJZQ
zXC_b*=YXz}{6U}@J8s;>n7;>-jZS-KbSP++Ab6*d{hA;OiB!Z>8odx^Ii5G$Vjj`V
zl)YGJj8)hsRJx^GXAyP5w&>%76UAQvd&gSrEy(F`dt)><Ye1V(n6<#ULrS1`RI|w}
z{lp+AI1pPZ?DF?ueY)`gq}idp1bo&Earb@&3P0103bZ~~4}zNI<Kt!#7Gx0%`AxcQ
ze!Wk)6Ku!k4+y)&_)w_JJfQ7qdlQ;c$aaaY^QK6D?&5ehwjP3pt5bX9cM6gzyV^=z
zDZ=pvd3Gr77p=vHz2zecn^V3#$dh>`gh+{f?<kiRh}Ehu#U8?tq!xz{gSw+27sz>Z
zuQAddW}k4azB^wgm<jQyXtTUCZ74qxkZaxFdPX6Jr-D*;nasyCBM{Dmr?zZ<^_4L4
z(DUO%(+GP6a%faR)JX^(egx%iBhJ!y664l)UwkbB22C4{7$um?BEM9z>GT4SwsiH_
zLx4NC=3ka+=C&?@B%T}<$A~CruG@-CASVfNm6HbWEYU7L4;auDk?CgNzrY5t2IxWz
zGe$A-%ikZXC_)}q)wC!{ge^hLa4=()B(drN#N;>K2aIb)P$hc?VCX-8paF!(-hMq(
z^p*gR7sghjT5Wl<UE<e#a4%rW(G8uL_cuVUNqTu^%s~a+&Go!ENVE&dMH2di%JYCg
zGX$x#XLtZF<RDI&y2*D?^c)PNCXv4XRuTSQt@ObjHhYKoyPO-vGnq&y7cYr``*u<G
znJD)j7HABhlE2@>4h40&>?upy&Z7jnwK3&_H8TW=-%{1zQT`$u0jKkxK=MLDVYXQ5
zFen#O>SAdGw=N+@JDxc3!6is7;7jH_doIA!2F{1$5|L&P?;%Ucu;LK~f)ZC7>1~9#
z{KJjmyyu_oNI*9Rd2u@~(~LM>d5dL_D!!v-!@;R&=SvgEdLpT9j}#TdBHSfW3~~HV
z$9S~F`q=5{0kC*pn0bay1^vu=M>`d^0E6*CCi?W=)WOG;QE7_)_~EeyAm?QD*jRvT
z&Y3W)ee5NKBI24z4Kn$-A_7u|c3am7$@%xzdFtya0l5vZudP>cDRcr*9$tuF>>b+U
zY+;UuKa}LZV};)WnWr|zkji7sNku%Vt8@~&lc3CIESKUN(eRBIni_G)$;HQJjXMy>
ziDnjJXBRgr><kc1*$$(UjdxA~bm7QAhc1=)w=i;GQA}^^%zr90IgAdI!*4{m#(I04
z-MN6=eHcrdpH_UY)~8P&jV%P^s$yTREzD7ZNRX2~uFHy`4|B4}4EHBd6u6sd`f-Kx
z(yw(to77%#I&f=3Zg7V#*=Z!y(?)5tqg5+?Fuz!zdgF8bE`If}vi#-@h}K6=)rQZ6
zHapPm^$qBXTgO>o6bV0y@r1&zOdU_3axpyL*+A|!Jrxb`I7f&Jh{*+|Npi?TWxTG+
zuuhg|JBN5VGx}Rn7Of%71(G3P)Xs&5J4}x~ZE=z)%8p~d-grucGnR^x&Ns&l5I4n?
zu`9*`B3)Se28yRj?0udSkn_AaS_ILVJtS9NBvSRn$H4US9oE&?+Z3h<L;ZR*{9pMC
z2E4J&1$=Qu<?zPxerzX{1#F}v`W007o3vv5Sz(fqs8x+I7ecx4Tuon_Bil2Ds5H3z
zr(0}#5u_Q>gk3RJG*w4QRf$hTm~+WvGYe<zaIyb~Q+=33Un$a+M-#%b&zps13{+R~
zY8eA%Xxe{%R6@4D$Gh7lMON~F!(kdtekoX?>GM`amnOEn4BR7(gP)-x6eO5f2&CRM
z`UVb8E{AflFwsiZ!W)-RbAT~GNsqsxh=rj>x_q&~l|Zl(NiNn^7+xUlS7M+1p;76M
zHicb){P-`j-RZ4Axi$9MRYY(P=|%GY{?$&Dw=Cj9h25Chr>sWL3qms+c}QN+9=`{2
z%M7s^#*w0(X7XTh4WznDx^4eIQEq2BG{FMwTOm%7xe!O~opH$@3P|?nXnf*BK~%R3
z_2KWYB}{cWU8}~wM0oz;Y>hSkqU#*92QBJL3KO#jktIGC;jFPgl6lACHx&OOKH=JW
zj}l;+(WPDi>SY`U#3drkO)38^KOgW%K(i@%-scV_x{OPNng_aQ@rFqEPpM;Y`_qU;
zZv^vfmdq${h=5sgOcw0=G#eN+#q$DkvHMwMSU~;+;&Nakz`Hedy2yaQ%tmsM4e5v=
z|Jgmv1u5Q0EfG`_Tmg((ZX!-=Sh419B-&{Yqf4sGA5|P8FgH3)j3@RuQxFx@9L&hS
zxlBgJtnr>`_{Iv=6OCIST%|Vp40rYZ3}ANf7=Z20-Vl}xiyv&Jqg8GNF@usRlsrIF
zhaSQ7gkfGFJWOGie%$1>C&clh&31(QxF|<OHi*9{?DX4*na1!Na2tqg#oRts9dCmY
zqR(TDjdwvoWQTcp%z3-Rofwa-tFYU>WpG7**+;0^hc#CXeJJ(6SzyunZESp-SgDs-
zawb7mcM4O#Mlg)KGr_&{Mtmf~9PDDz?TiU`7Qocx53eGi;KbPY-Awcp-V^>xv^mqm
zQ`Fns1({2aB3o7+NSp@?WvH^wq1ZS@l#31P2h-z7SLPT@s+?t>em60kDKqX+JTDqv
zwK0`^t*|*EqXw^ybKC=Hmb1L~HQ1z27Q{a&DRD1OnEcaLF|I5lx7@6}DhLj7vOYcj
zEy69zS)?Y@!F&BbE}rO<cZy=&>A|Y*K7U|7WrqA%6j>r>#s3t}e48?E?U;N&v^lMH
zXq~u8fEhQ6{lhIHDFJGl{&+})tAon1jzi}7$pb(x6h>JxvY9M|KM>k`V)sn&z;Gq@
z%S4hscYUx(tF;L~N01u{A9|dvqz(HeLS5qN>Rp$WKUUg!kpn(FTK>oxW)kcflk@(S
zAE^mu)UwQ)Br>-JV_Mz+4}#p{$+@n)UGDOb|Kb9qgn}1Ex_-!vb@WZMJq$wqa6B{^
z^NYx1iux*n(*&8LddQ$R?i0{rk!09d<&h!^X5}19tRcWrm|!F#jtFTLExWt8SD2dy
z54`b3`4esd^ziOMj}k!*joM|i$BHCz;HyKhy@j}7a^OE2_Y0u=VW`GXzVGqk2P_ce
zwtB-S3WPyfr6(OwA}BnyC%#088Gt!40%0}2|F;6d2Mvd9JW?;nC7;M^Ty>FdXl@>G
z_>o=;I9aG0y3WJCaYO#8y^9?XZ^%AbBr9Ex^y_B|a<c7+X4i{!vc>MZ8ZQggfI|`S
zfx^xL`!DoMPeFUoO7h3V3fXV2rvYhTj+h4HYY}9TCT;FWJyRsWG1)*2izaS*f9CE?
z@ZNPLz7gRD<k--z68`Xa$L2~ySKOHidN4O^MC7?gr`XaD&-N^s3(LcMs3e{*;>d!N
zhh*_fAdDuYagan_5k+Ybn7DxZLJ?h;cUd{BvEYkf?gHGS(EKfc|CwV+?Fdo)ujiLy
z_Lm^cO<fSl{-!S#3O9LjGHz4Y++zbvmtv-mO<o3cP4bx%+el%S7xPITI)8fbF~Xey
z7Rs<+TfPlm1*8zkMR!evQ?uzg8^Du<nJ<jUM(p$&G);uIWB_+Uzse+2_5hb?dPTa*
zO0gC@z7C)oa6k7Lg+1=@OsFa0d?8+h2lYG#)0}))T_>1A!N{(d?F|TbyFs>#bBpv4
z;O$ft57k&kP@Xkpi$2YWw`aJzmBuwYyIxaKT|UkQY4$S&kQD+<HO9YLq>V{RMcS*4
z1iAXGxYJ_G{}%<5TkISKnXl}?v07ng5)TNZ-aXvxKKm_)Yu)`=;1hQ78~6;VA^&_L
zhzg>E6;Hk_V=z828mY#2C6v|yCgJ8koMt!IbIw)Rm10&vgncc4tYXyut-@3gP2o`N
z_;wMqo1=Q3BYp=!Jz7t##08>Ull1sf>LK?Dbf@T{$URzh{w0h=la2Y3??SjrQX9(y
z^ooKq*SKvQx0O-)aTcoaw&;6AfkC4P(wpb7u3a3Wuv25)CAn#Ze-@CR?Q2G#APRsF
zww}0NgxkL=uNqh{zYpU4xr^5s3;e5~QM;}cJC!s!)Z$ptPC*xNXq2ag<gr`9?mvF;
z0e}Z6`&ZulOw8#O>e3=2)X?lipCB_4vp*C$@`KVZupC+Nuo+`N1al#~hOw2qMRe}Q
zS`)ajAjE&A?Tord!kj!7OWDwM*N=*%*j%>9<DzqyWaMS1<sSp)c4A2ANNo>6G(Urp
zI$t45Oyzhv#(YvF(d-&#ldL`M8UkHJOuCR#VnZPwopNY<xky@dIz8{73Y&$Uy)xuH
z{8IpOOILB?B~h(uN*A{C)t^N^19vfT`;}~I7cC)VG1&&hUV_MXihVs?Au{L10~ABC
z>*oNu6?9WJ-6i;!z}#9quc?!rk4vbEn+BaJuKgcknd?;A>XiG1x%DL|F;Zn<`hZZU
zhg8+4_e$FS3j_m(IQTkJVRK70fqyrD0ip$9P@2wLMfrD`yQy{W5|r~ss>M97H|`ad
zdmg39PGTP|f5ufOqQGE0DaiRHHD5DU`4WT{WxDT(HNJ)*^Yq4493{$`bL^7bU%gR6
zotzMHz8BvK%EJJY12%%+0=R%=iOQX$kc2rY-Fed;`iB+GkZppSvFt3HB+hj8^+mVB
znm_8uxnGKKv9(8~TOK;AqvAC?{-AJ9iA7I)+m}j6!mOmNR|Gi?1W+Nb2_Xx;l@wEE
zGr<W`CDsw)Dll0{2pNt2gwbr6CB=xs@QE?XL|gwJl=J6_=%!pX?f-os_brrJoYJc0
z?iA>F%$*do8QnhsGV?g4#17zg0qPt&Why^%c#oRrp8}i!dtN4f9`qGNNwC3ZAv|#o
zC~73dX`Ca%D<=v`8DUV`X51y%b(4dk)U1={1oIF@eHD9(bg6XnvOkV50Ty*G5nWaS
zbV@qb*tz({++||Y^DMUsX)WBzwlqzLPYO0?5Gpb8)+F;*85aw(>{P88U(8KhDmT<W
zjW(VChfq{bTZ~vjg!982i)8H~A)YHS$-qbWJkYsSlBH-fmKQ|5WSd--Sa6~M7p7|z
zkqsq9dt9AkA+HSayf3bh%aj<TXm0xAJ3*91?<vKP=FNn5LmP;2!MFyc4d$gn%uRN5
zQ`U*8KPmu*0%u!i2xu`6?@NvNQiuzWi&JU&;)wZ5h=(DyCe1ZJ5S?Y*<VmrD!fZB2
znfMfhZM|4#^hS>;&kae7p-b-=5tpbe&t>z$jQ+Gxcqp$vGL!gJv{){?^05m*xZ}u8
zI8z~ljW9EvyZ${fU9@Y$DkJgXN+D*k&nhKS?z>=-l+J-U^&O2qVJ;^<fcpi^FZdZO
zV`2~M-m=3&#ox5OZW{4!uQ2M2Kk?>p3@;4jOsh5BG}Z81F96{dr>L>eB95*$=4C)p
zd>QT;jwKX!F|;k#JuYf-qEOn9`y9P2LdgD$0=R*6(;J7ev6P_P@^S}Mjg5pbtfuy3
zqE(HD%YXEsS)CKIMHefQ!E&vuie<)<g3MB6r-3*|bgTP}8@Eo(wL}3lWo8pw3h*C@
zCTx*KyY^TD$%qp1;xmCRFw0X1Ue)5uG9i?@$bNG!StO+Pjip69YbwM$lQRT(o@nw|
z#ZyZ`(ml{-^~Op|L%>8<(x@n`58Swl8$_5dwO($i^A^=IAWlZ!`1TcPmb1lY22ZD=
zb)l}6jwj-W%R;MHqnxQkMU)d|<1~o3j8lZTg81L6sm16236RUr{Kw{aaRIqsF!5=r
ztVO#(mu7%F7{`cIHzChs6?QK2DMNoeS%`DNmI>zxqcMK@0zv@dN?uo-D#-cNxN1cI
zW?nf%APJ9PSF!F2gkg{#o{MutxTffdQGDl*D7`S%Tak}uj?5#Ffiy6?S1M3MC6?Kw
zl253MD_?gM_k6`8ALO&4SZ@r32e5A5q2lDFY+)`beMc;zqaG!Yzp#UVA0k|1^pV)7
z;*dv>3(Pwf?J<63X!BXJWi7Uv3E*k7uILuvlA&H0;av1yA+8rznQXw)yRYx80!q$!
z>SkVDwfKomi*C}-vl@tN!2FF41Z-Jh{8DGD6k|oWuvLUSsV)yF|3RGN?Qo83t`J6(
zv)F61e^VJ{c!09V*Ts`%N-SP_uOprkL{9jR=FyAQA!yACHxAcRnEGQk&cw+)y7W(-
zzTNuA#v+L;Rm%<vGf($+$1fE$@6oqp5qz&4cfyph@kfPQJetZGX;SCRxhAN&-Ce`e
zRs2|hE3NH>dj8MKI24FGoOcOulO;pW#l}KAZxk1DQbK3AOJIw8z6iGPI3O2cxRDRe
zt1&DzHz&86I+}5F8P5(T($IQWXg1u@fpzOinR&(+|5xit7rulL=QB|6?Ueg47RD8Y
z!g0NU6|0GG6_85WV!iy4Ii@Q%$pl*Sdh9WQn63wlgUs@xSCD%SJN;U6eXznrFf$UZ
z-{jR}iU8Va7<Ey%!p;pN1XOvV=`zU4GB`L^Bqf94wYVgI%jb=M66w6ORit6DvK3~2
zqq*xJNoT7Gbi@WHUYuq&9nl!Kmf3`n{zLlN1(h0mFn%-%z!j7=pXSts1+_+4MK9lc
zx)z9PgN}6X$up&Zm0-*e;1t@pb}Bu`m4cjeO-@uKm>oBz_-~-X#y!pz;K>r-5b?A~
z`)kwd8gdkq*63!utuHPQa{`wLW7D-koMp13VWt}r<kT3ham+eJ4BP}I<?U^PT&NDF
zLtbF$jVbGbIq4DHbWqHrg*fTyx*x}ieUYHnk7<Y2jLXX(F~FFHZOLEOBbGa;O%UX$
ziRNmoUnGdv`E(Tfq#!4VJ=hSY6z%b;Fb4CQ<Ku-52)e&$u8QFep<T5e4udZc>Gs#s
z?@{_sfvyOhQWE~yjX+!xiqWnu;O>Il_?&F9&@HtwK$41i2{?8ZMMq4#fqg~zcl1oM
zSdQa!Y$Tm(Uu?7q5pu6)wu-Gvx>GaR<)U3o-7WVH?@57V!Et9j-V@OpG!w?I9^VR}
zo3*1<m~&IY6phSygJL03&J6n)WcHHg1PoVXqRr7H1xnt#a<d|Z_K|o<l$%F)1M>?q
zcor2-iP(=LD~oW6WI4M|%)2=NCF(=Xx1z$i8G46f?@ZdGVeBW8bhtT%x@ijtPeIx(
zaTY8=@4FJ^&RPRU8k%2YOXB91Lxjh4z9Ga#^#y=Bul^x--7K8K;1NGw%3+39oUv7r
z+tfNYO*P!u;R$uRRjf!^-ZtGD#I@rkJ=U|X3`jvF8;UqofHj#&2+ZZ&I!S}~4Sg_w
zIhc|E%&wSqn<4||B`ui?3ULOS%&PHh83l&p#3()%qM3v<=yBT?3CQ&uBW*~C6X0e7
zv(pCxB&m(k(c_lwpwvOJ+>J*>@QL*t-mJcT@h2}9Qk97UTw)EPB(_(C@jp9jJWuQZ
zVg4kSHqBU3h`T(yHJpD|VlP3AN*vU_R(`ccyb}BG$XDmVhQxWCEUdNA;#n5{As8;Q
z(@aTo(TP4%7v#DQrAw)&{{n#0>FPrKSA;wKAX9M6w{sDj69g;(Y62*}9fe3hn<fO-
zVuu{3E9{c7G1mEhGtMtD?gOLo!?XJ(!ki6zo4(j;mx3NQW$aq9e@XMb8mEc&80bS3
zm?6@2VA&hDdi+m-S%|VD-7dauS0IX_8;I1COG)i<z;0j^b=>OX<2;3(1~MW$gLegZ
z<ahMMQoB1Ujv`{R!s3t|ZJZ*)qa<ze^vtHl)11Eqx3YkFr?3saSg>n~1lx$WMZ5VC
zj?v4;J3{h=$yP^doo)62a@qR&(V9h<0P|I5h+0}=KVhEeFr~m0L1y633Z$s`OsL0m
zBB+sv@vVEB;#>(m%RGCT>c<`(mANRVPo=Eg>JaAilLz1AmF#ALG=wheFiu_&N$oH(
z!J|SgcGw%*wZrw8u0tLpq*Y2?>yvbTia<))Sz+=!Qv{9Dfo}c<g<WwpPPjaHSAe;#
znE`KOo*2FrZnpLgb6|4Lub`<aGY8}O?0um~f10eNyW1d~rkpr5W1e;ZcczJBSBzCf
zn*TZ??m^$NgCL5;72f!Lj-4G8-zd_Y(ykvRKaR$#U><w)`ev7{F%A^!bh+}$G|$7W
zHK#ks#_V#1%@}sGNWr@3G)G7A53Wj+@8*RBxu)Ej;e1gvbz}EXi5*0^Zal^aGpCiY
z*&tQoGSMzX9~XqUc@Q7w1ZWJrA$9WI)oDZ7(j(2^rX_|{CFc}_aj>vl8LB7=?%gsP
zk3VSB#9?3;(OfAWOw)~az!e0#22-Wkyg{_nqR>qg!0|#ID0K)#s`a`d%uduj$pZta
z>>)IFEoz3Q=4v5Mx7mxe>pnfuPM3!<(5Z9bUXTYR=|~aEf;MW{X<@^x5qDT`krors
zxOhdh8H>YIEcO=b1E7O(*MUiVb5Uf?twTJ6U`Z^xm0%jedo@7lLZkSqJzZgc<v9Yy
zzgUQiiaRiF0w5um5XakiwW-CMg7T=x%}dh2F53^r%Fx{&YbfZ?=plLVk3HS}!eAuB
zAAb=8CYWV-uR(-s#0CHdgFEN1@(CPs?#EXbkpZM4o-M@O?c`=coG;o%<}MFM_oMMg
zLB&|8p(k&|O~NQZa%%JAK?s;8KQ{OP7UdG4ePgt5&~#kpn?_tE!s#P|r}u526y$-5
z<P;kZK~kcg{ASl?nMPC_(0=x4(PW)0<c}5Mva#WyLKNV4GyOnsi1S1{%kBz~tH7lj
zN~n+BK%MpLIYI7@40gTPr|EwCVNlA}jw38qmS2jbrra`R6}?1M>lYO2XyEu|!Kt8>
z4|PX->>$GIZL6TPJ6tpkkezkI$=igubkcRmGUkq)Lf24VZ~S?L@AR-XvcFwnHy`tY
z2Hp3jfw%~8lR=y=CT<~gra2v5Aoj2VT@-xU48^^ob4xH}G(7Gb1@buJNDQ9};&GLG
z5CuHY9Hq&b=P1qog7d7DOv*1V0qn->J@Nhhi!AhZrLLN6ASgE&*K&|X*ntT1>@S^0
zqm#v4J?X6ScmC}|qaF1Fi{B;o!{VVhO_(c^-aBN1y+DxKfL&>PBqCS5A2rxlWg6M_
zc{Aq!HF0QImIAgaD~ocEtW+w|SCRw$wK!EIiPMvKn(8-2;v8SIj)*29x)7u5KcZYT
zY;!c^;s*x-xeXcVHB@Bt3ZaE%WVorqc~Tsu3vy!-M*ODy2U%067&C;JjnWWcZ^A2(
z!pICS*=ZZek536=8X3E0{OMqV<kE-j|E`A+rsXv04&lx$wm8%PV~i=DF}t?_>d02V
zFBU!wB6laQitBi<J$4tETc0}$y-^k5qE~u(j|eyI>;W$jYUXo2b3*KRI5hl8?p|Z|
zBcaG<I_E&Kh=@W6=%^z@rrp6<;m)avYrAG_eiVSascu<OoMiPf=4fCuTTejhWxSOI
zWp?6HE^iZmT1KRkqnN@HL=Xj`a`osDVdmOU6#?f2VJ-wq1G}4t1-N1o=i(F5^ch^m
zNM(KRF^=D%J76-tSo&ChXDyM*z=a~=gLb<kai1usuG55`_`C%4E=My~JB}E6iQ`Cn
z2*nU~M=?&YOVQxqG-{%ALdYVunDcmm)LGdgE-S*Zr^<2ZEhSCkOpbqucHiXgtIo#8
zpWxrwzE$FlO!RTdGZy(RU#NiWYH`pns@2S6C&WjhU5*MpsxPLVSOAD8<bA6EGa1)>
zRjnQI{gc3)f3pi2g@e`-VE}PTA1j^ggrw<A-D87O0Ngv-_jj^BF$te_Dj=o9GdH5n
z?V{-WdN=-Qh22ZqDtLxOQu#tqZpvZqjxTW<w0jNG6ZUa6-CNp8D2Yr!H@d_bgyAap
zIMA?(aH-|MiMm}X#0gjxo>bg#l<_#39OhK-O(C9V>TH(|I}=i4gM%78RfuvTQ@Op8
zukx&PR`JUKGRJ-*&F2o5YutvOozq|j?v3q4QRXfVFRoOWsL8CFTT(IRcK~jIbl|Ej
z^^?NP9j@RrJv8HMLDW^xuf#Fu6#oo2BuVnmKTFIz2(UNLEz-eQpH{&E_>2N#-cCn7
z&k7=G)~$N1ejbFYg4&89w1EH$&&zBvsxUE;#p`j22(zsR;r8nM4aZlNxUT%hL!j}3
zNUI#Mbi}FEfb^NH5ViC9kSi~RuMAsm7iNy@QLrJd?ot02h35Vx6P={WIP*e47nTDj
zwnR4waIB<TNXohAOHeBfvn08EdJ!?zcUa_Nu8SdPa;_>c3(Y4gPqe;GNBmO-x_2@t
z+TxlrQuZH*^zyPmc-5%k5aklWZbS7xrl0sgkkdfR+Y<wqLc8|JH66*q?@D3*ja30B
zmi=+<Owrh1BSLe`BaE6iN3jBsR_h%BPJ5D`Dp~q6M_^{-6d6Y&8whX&l;wSKp=cN|
zDuL+KD<C|0xa6;$&&5LcELWH@$CVJ~h`ih4^or$qPl5iY8xwjaj=hEC9)UN7`L2S5
z8%Qg$^wosDsvt4QoYK?ADZ-ry%KiaN+s+W;Hsy{3wzVG!pd?UZW9#4ht9nMd%yEAP
zsHZGUa20b6G2DIpFlADmpDmO`Fr~$Zk_hvi^$kTTGyA!J0CN@D`DuA2;`govbg4Pn
zXOYtui&bQG;jJ=`pSXJ5SOTaDS%#kx;AX^xI8BDH2{LyYcx-)Lq?ZzQ-<=~mIcCcI
zvP`MT?zR%6*A*!-vU*2(;Nk*7Y8yF2n0h^gE2Gm}Cg0IGRG4`n^$3a}nF>w10o1%^
zLe<u+TS)GJNU|8p;N-T9MQWNyj}Vf(t&SX?5l!}d>yOX>XkxH}x8&z;gm4vk%qE_^
z36f!@>m#*T>So83uhUpdgxRfUTWw@8B+zW_t8}oZ8WCVFGC6UBn8zl|A7w*}luGA6
zQJ`yv;)Ko>X9{u6IO-hiXX=kjgt>IcJ@I8p%HH?JTlh+05#(rw*>8n%QJOU@NMlX`
z{uw`gafwLhs3D0A@PrW8h<yvD4BBcxH4}(TZ5P@v3N$~N4J+}g=vH2sPQ>d1T=?Wx
z9+5S^yA6!|HG?6oo^l7I^XKSR9p^zI?oWDd2T?uU+L>}EFxBGPHsZp$qWl~3fSmR$
zcNd83HNbt#jYZP7XwN#MQ-sUWkJqIEos+cv#SyqC-V^(aaGiBN;>P%mz}y(gT~wSY
z#Oa_=^Va%BVQyOa|CTO)vAcm<0%bcF4y@x|MX=e0*VJ0<B${cZ7oU0?-b2`|<H;}<
zlzjwH{vHW~XNqu1I+#hR(d8wALU3qwFs{3oI4%b}aqJ_wp(QBKlzJh4*87WOFn`4(
zV|-7D>pddjotp%_$}G@j;h3I+z9azFOEHEW-UCoA(NUIF_QSR1gTOg|PE56kRRuYJ
zgsn<EA=<g6XC*nq8jTMH!Ue3`qWV|DZbK=}FDgl<x42HEf5(<p#yiXfHwrXsFxwSt
zo)$uB<*o=9X%7`YRd8!XuVgyhRbZZ4dCP~6sAhXzD7kaCChu3DW|9ecTD=~NJj_?u
z4?%`|QX}zWL9QQ%t-W!cXm=020h-$29$_wYL!McZjJeCB1&Cb5Q<KtWHv~EZ7Fh~}
zxFN_bfe{;X=pv65aWKJ<qqyi6lm}fqPa#CFfZR#DQ0GgBvCiXQFnPk{v2nk`t~|0E
zu0`4SHsf=lEy<=h(1e}aKfwoAfMbtA-+4=**ApzkbT`?n;4X^nyk+h@DD$9y{te6p
z#BiJ~>&;IB<etKmr0vVLPZj^kS}$GfU-)S-+K~+plkPX7s6n&V5vM;xICm2ynA=3s
z40y{-Hjpd+9XdCI{DY|*>@F-f1Cm&}GZaGtT@ciM+EH^C0*GPPkRk2Jvk)#h$2+LT
z9~O|ShZz_%Y&GV3&VO(`R*79in70kRZ`Gz}lj1O;E^pUxJ#H51Zi972CGN^8px>><
z^CB{H*!k%sY~|;HoGaVEv=@)9g=u<_3FjV};07f2qas{rylrD@wZ#hsz_G`K)niuy
zt+Q)&0pu7C;mB#;7x_X#En~-Ykz~SwStBOD1mRNf6c^1+-*~pfSk!z%__;6|svVP}
z#a|}u!AbYk0=JY9O6S*jndgoa;GTrmXh`N5UAo+K-YbsC=#wA2W;`gsX>r;)84pB@
zz6z3upQg9<MZ1KYI(c^9S{PNJaqF>{h+G_uTPe<puYtH_MsNi;UL;+POc8v4Ez0RJ
z$7*ij_MkA*<LS@-crr(4nSvSHzCKg*PTh`f#3Lovmb@}a=6Zv8E=840v8vu$2&K^C
z&I0_4GJ@<NP%-T$q%~w&M74cu#%Ds^Re1IdXBn0F(VIouEFv;l+D(Wn;p17h03QpZ
znti-KG#GQe1!Xo)#Y#U`7VQqim_SV0S%C9lN@G-4<H!F1a;lV>m#NqlZ!0WM6tcZZ
z&)aS%P+h2w>C?_4oHA+_8vi5#&YD$kDCU3LvC)U3hZEu0xVNL&Zwa8-EE^R2r|&@J
zfijGuljPPEln0hxwBAItIn8$|&@L-u;1F&g{vetplV|gIx%`U54K}~;@|6;#ZK#xO
z8wkifjN3Hyp;~M&%xxp3gYKP@(XK*Wl>Q1AZ;uk~(&`4ArkWE3xqb+0h}zc+aAzDq
zt2X^TNCpypJ(m7w@k#e6Hx&t2`*5QZ11&<2k?t=dH>Qq2cYYt*nY5+Jw<>^Cr*n(+
zG=*Iq^=|Bo^@Iahs}TM1evYoE8K(S;PfnYgBIDvD(G~=xrDAKC)|$rzo9mc*YMS^)
zh<ip7ztV-)1wH^IqcnJGF-c@=<{V-sOKbT_LS1~^CURu^iU22-oYJPN8asbj#F9-E
zcEti*F_t%s&}qBph2{Q&(_#rauL^QqD|FBI%NR0+lZ&zWJ|afyzKAm?iJ<<PGPOZ_
zL9|np3aVKdiLoC8xfVkxQM3E^XN9UI$4-hJJ|XN1V04Hl7(Hsbhrm|9LS-x`Ex!{+
z*4Vxb$BktKt*b1V96o(Y1oNheeGsp}d<Nk1aD#(0OU|GET|kI;^(t>3>{UV-tg*Wp
zia!gYVtPeGO1{5|hA$k(YNQQ+4#^<lK$vXL6-8y*I6Le84`G*Ej&d|D#k&9X&&;6%
z7|cH>(w(Lqoqc>Qn#7pw(Wi8NQG7+<>yL9plemm4&s7)(sBgs|GokCHanP51am^8&
zSS%L!3cy{i4o{gwR~16G7~-m2v88A=0Rm9B!fpbDc={g?yT@Sufns+twBu0GxfkR5
zfIY%vLMV^o$EN=)K652X?+INh#D&r&$!1LX8UQYGrl3LaOOgKG#l;G4ObT#cVQT82
zlrlWKzC_z<Jmk#wmYMC3LS35n!BH-W$CBTGP@0B3f9)coRaj}7;%Fi63pIpsy+QoP
zG7>_k+IjTbBGLf+0oh^QDkuy0Y&NB_Q>o_p4mkDT@$oZ-$vGwQ7ZI)%YZYq2t!DXo
zD`O7Jum<PdC_$Ms71@`ZB!Cpgt&M)?LJ{smXkX~tC|RzWwfIB68Zl51K(CSXiDGRw
zDEFx<3Xb@?q}#P$S?_y%&+O14Irer!%xlE7YE6DgslQNX&8jBdZN^&TQ!l19(4-!u
zh%_atIJTqli|<2H{`4;B9wM9w-q`Kk^Cz=|xyDlVtRCY9I1<aERseL-KX59tcZ=`l
zfN-`bIwX<N>h(DRT@Sqi19G$w*MpkN9hK^_#av}PO2~w=<J=(5Pwx!E^ky6<3?8B;
zLJIkZD5u?98PQy~<PSmE0U%y@TVR>K8SBhbMCIwlHZ}{}2yycLI6h5^sojLZ6b^G3
zO(NX=u#!nH9-l49jfO=p@5qX#OHIVyck6i_9%RwIT4BfIqG?}zEZURw`0-<7;*TKB
zTExKd>%?>c{*JGWSYSR#*Ne6j&#s#vgp80<;sJ%7I49rOU@TD3J-TXqh3GsrNb7)$
zkB+!_LBQO2T-@a*ZfCqIEO#-kG$CBBwGfC4Hh|fFC4MWwQP@xMT2L&%a1o`epJd{9
z0&-nzG|^k4%{FFud0`KnUbmnQgl<h7*9vlZWbihm0dkYDTx}gPF^LOtRd5N5!bX$O
zR{~u~bURF?S1bZRt5q6IYI7_K0jEY8F$XB@O#%)GxCKAhp8O&;;|P7>7an!Md8P;_
zi2o;Yct?PnV+6Ci_eE;$!1x`D#WHr=C<l)*VKF|tkenR$!8W?{Jp!F(ds5%@EdfA|
z>3Y*;BH$VFG%vs5_)D~hggk0%K+wIf6KZ~7$fw8Ye!istlW3vEe>c&bWw!Oi2@1Lt
zeZ$u0V%Q}RW*}F`7Ye&wFv`&hX?$D$!}#szIs>cD8%q(-m2York3E-$bVKR%KAkG|
zl$ajXO-sYEKPFyU6M5())?NlmO@gIRJR>5{<M}y=ZI%V1YRyp=`E+uAix8^P*%9;n
z#6L@a8V`zaan>HYVl<Y6G+*}{W?77X3Lqz>fNCtaJcK#M>?j#_h7gLyWgA{ee?%0K
z+34|D?FtZ1xef334~V2qP+TW}`dh6C?VNPOmdms=0qMTO*#caz^tv?6+jk|W$>fOQ
zAod+oq`}QXn)E>d{-qn2%qZUDD?whZco|k+{dDIf=3Ke_m4Vk2V+7Dgk|6JMK25Je
zsSzH#D)EO*pp55;UxYWMx*pFgQs-huXZ%OFi-1;?WxpCHtl~tb>x^yoRUtf{u?LzG
z3yUTPSWu0<M3}+J`RY&%2{D`4nd1fN?*g1Y9hvJEeX-zbK<?gh$RV9GyjonS3&?Vj
zZmlI>L{o&j$D!mLOwK~rD^tiGxHa21XNrjuLdvvwNSMcg%!`;mnd|3Y9Z>T$@0V|`
za2}b%gB+cY`6)nV2(NyK`R599BKRCaW3P?VVru~AiH>8w7$@2ZVL*lze*(-LOi^T)
zGf7ceZOinNPwx+yDJ8ajaf)b{Af3lhf;-E^dCL^-(lavwh!|MEuUQm_wXQ#&7oCfP
zLZuO#k2Szhw_W_);{fRJ01ckAK0%}_DI;d;O$R++V9}iT1YK=>0q{gH)7?V?;AM5X
z?rIPw#X2tvvvYq)Bwd9Qm3llT!j+OGryNJGGZBcv$bBKY?kl3qE}l@(sp;~QK-}_N
z>+MYjG{*>YG8h6eF(QizazoTHs)!dvyUfhc)O>--0Ir$rVYD$Fjm=9CJ8;><(Lh@Y
zbdJa->=j5pVQw@OFiB~yW1IxLxU%-)=QH&)wU>bohoZ|&fu#0&|2Q@g(X#9A8^|*<
zwVtLAu1_-4U__oS2#)p-48($KJCPxEpG8Gz;i;>it|ThgiWZe)WVR6kogCdXox`3c
zC>PX{)O%$lb{0+9Vl2N7kzDakc8ST<U{7J@qt5to0V5!{5Wc)<{KM8QqR2z!P>fx#
zh+=c5SVxGbRUT@IIoF4DskKw2>#Csz%9Ib!Itsc1Sh}$XL>nx~yeqBoC3E-cLzW!t
zJn*jjQtdG!6!z)<-?$A3=PKacy&3xopiy+CE|%O70)DbTVsCc4D0O7~`+cpjYwBJW
zV>SYDxoz-*$QcI+BsEURx?;k{#Xq=d#vUQ&`I$eXPHM)gnZSIpKh_iB^as)|CC)4n
zG|JiwZ?*|>a_`rqqtow`g}D@{lr^8C`9D=4)$l_&XOvNpx9d#|l1JlQfo35Vs<LHm
zHj{*Vtv*m`G@bF-&x^!q8XbzWVBWTA0j8JM(zWyV1oGD?*Dzx<!telXN+li^;WF!O
zcLpt<B{p}2_EENYXN%xZtU6<@EeO*y#@naE2_mTmyA$S)8KT^BLs)GL#bW~KBOTLX
zgRKa9z;nfNFn(XsT{V`*Cq+{vW}%)~e(Oxg_UB{~=DVE7v8(v6kjzmZo51?vZ2;W~
zk=IA$cJl~96ugPe-lqyX6<c`llmZAjqO$0TW+p0C)a$VwU)=qBxU0OFNEbb+kp@JQ
zY^B;2=a=7jIuVx!`P*Q<JKj~84A=+P<0}z)NKNaHEw=Z!9xd0~iEz=`C30rg8<Teg
zauwU9FH3{__7cP>Gzr|zxKALtq?-n9JDK3pHTHcXTxIrX-AKU;|H7Y{PzU0Wogthn
z$4+=V;yS({FX!@%u8l|Ptqbl7?&*|XhCJw(a1-o8((!W7PGFvnTe|o$`oA^kIkA-J
zT8(Xmny>io8z0Y$cK6{f6Y8@=b_1Yy`R2@}qTnDJG08x4?hfhV*Ssx=g#_hn{F0d>
zj`?K~F<JhwJX|9Pp5Yy(7W3=@;bOo9?mUVnS9zVph*OlAh_SUioc(N1kX+2P)#wvI
zYM1~F#Umoj6Wx6Gqih{}6^YQm8lpo&TvV8x-fPvWSjOw&_C9wm|0cwlaTvlT@VOE|
zzXX#O-kTV?cj9U$4j1h|5S+M#aLum@pff%C_^JSR#mdk?d{KVqScjoG%RYS1&B}Qb
zGa(b!0VT*872=;>3p+<3BcYeWu=^Eudmyyc1`ydE7X&BM$!5G)Mv&+$Rlscf62Z)s
zlUIiJCPJJFW~l5uxqwwEAY5)}$%-yP{tqcY4~YJ!jMRa{P}$fFwGqiX$K>*aU0U8b
zp06*?guSC)!i=i~(S`_kwfHI%bvZt&wEK%r>*5j-&X`7%W&N_%A}!|oq^JC04G6Au
z;eT}ng>zQ*X=&%Ro*)g_E*7oNGQonv%Gb#k_4lN@-$)ekD(uf<nh1};^px~8OpfTC
zBVEAmMxCR54JL|udo!*R>LG}AmS0%7RTzcppoib7Fe$T3>WS;>CV1`>SK!?cZu-JR
zFrE<T<gAfleya48m@bBCpmhpkwx8OF(|ZYeBB2?vHLUtx_DCN%Id|dqB$jPJBt9V)
zDC|5~HLHlJJ#lS`)ubuch0m9Txq!U7i!Ud&*DC@^wTXNFn6n=u4_FydVXG^Ie8!{p
zzFc9CmcDl8$oECNpq)~lH?+c1zxf9ok3737^snVayB#^BNek*~CB%Y;DNQ5R&wnAD
zptRy{sAzJY&Y9l93cCeS2D3@Dus(J_V7DNrFQDn{qCV&!$BkV(rYlVCc=rIgK}4Q>
zboV31HUVIIFIy=ET}CY9vHe4NFDNrZ%WN|?8Y&R2JxoQj4HpPIIIO^8DIu*)@V68j
z37{T09*%7lrjz39knP(6nZhW5hB!@Bu}+{<aI0@Q71*WV+*t1g?>7SCX5hirdORxH
z&ClvUGtdMZO#>q%wA3gdwiDHw2=PT42MX{n8j8u&=F2697>el$)5kue#4}gg`bXR4
z5bKE&q6a1a(N79;fpif#)|?JajioKxS>fD7)6x>6-hPfySBlF*O{A(N_Xlx_<QITQ
z&1e^vi;WfBP#ji9^0iZnIOhSykFc@IoptP?gt<U%Se3^6qG{*V#>Mb~1U-NEq2D`R
zWbVS+;4>rAwVKiVHMp}z`OA6x#R5oT7(X+=A?PY)`wdlTrGr3Qqd%QjurvF23E~&7
zHmDP(IO$+wnyKtNlC$2^OK7SdXkeO75KKg`6Cdld>%%PA5nqXRB?iztFLFrnSw10R
zo09IUjP}Q|qCJ_XEr{MJu{F~>`r>LL_cf-)!G{(p<4ig(7n!>n7fa)t!~8iJh~mQM
za7cPtC%aqrf1)Vyxbc%>_Z%KV1pMt0;EciaD&|cha%YvR&KjEeCyoR&Lu4SMo$_f%
z0eLXs2UfR{Il<sfLBWnWfL(?7zUt9Mf~>5}@BIDKV~f8%F`>U=!oNeX1_j=d`eWDQ
zpz=zMkja~AaJ?#!vT~+`1Iy!!7z0ub)?`HYjX+q^%ZV`8=wqQ1pmSH|wmnMm#f6ZK
z%d)d5SC|<UGbL8nx-eHm-ow&}5Y*~aKC*zN*9vV7emYMjrvA1-7=H4|cA^jxvxA~^
z{<O@AhL}p(Kr%|Jv9X{$Bw$9&aT2s!9%r>|6hna7!K#<vKABX;!X%h+Q5BFI0G4W~
zjFfRkNVzL+7T~N<piPR$%764|k?aRIJeeqFip*q_Il{TZTJvA73~f30DZnnM3?}-c
zUqGt}_M!e-j>AoO@e?<FrxM3}L$3oIzZc@ah6gmep-(ss$YURCTAusNR>0$g(|#yh
zFf9J*>83h+M+ds%GEq5m?zpPuJ{H7al;ty582>BMS#u*)``XXW0C8p0^Ca?PHttL?
z`k7w3k4a~l&@oqBE+Y4*;pxLY^|<|PkYZ-hlb0m=```H=REg8$I}v$4sCC9>=a}q5
zp~4z9Akr!Fz^F9#8=qUmAH=oJwj#}9e3fw!vA+OHM{gdC7e%-^(evOaIDbr{_>wb<
zU)$=;zp;q`7dA<*$z1+;fiA&xj6vD$E_+__AC^KmUAkL<89M}{<t6W*=g*W>qr>(7
zRG^tDU2&hRVHdan%++u3gvVr&&JF2FGj6vKCr(RqIfXY-F9f2IQN=MJe|Zs<IlzI`
zgmvO@0nR2repQPvguxXK>>6?S#l;`Ije|m|a!CR3P8jw+*9kD+no<>fDVpCg=&r_Y
zmzuyFczWz3!UdL(HN=*yg?Kg?ZgBR2H7Oc_wJs~-NvSe08b1@_ipdYGZoV}xFCyt&
z71N(Hg-}%=GF>CeRZaaf{wu&$)iGmjly&R)E5MvC(+8J+c%|XWBATBhqZVrmQ`grf
zyB3>?^az$IG1Evo8QEK~6XOCw>R>ktvU(<YQ$eojP0yFyqYq}q2yc5aymq+?gjHn_
ze;DlwXO_w!;B?V0`hd*Y`Wac1uP#!+2v{5QS$|)|!c&p9+RZpzn3u!>@`k&NUR>o2
z`AcgWN-km8V{+p~KeXhS#AzL=A9qcG%GzW|0x<4`gt`)3F4YixMu-clr$r<z$J2i(
z{*_H@t*K=$uDQ1Ox2-ZwmK$FQaf9i)O*PiK4!}dK^~B_8Y%P%9#iLwYOA?`#c<jXJ
zzrOgL(@|`74=I6OCgagKS%8bnT@qZ-#fL&Xpt^9ZIq?Q3*N&a$#yQ->O}YmZHc#}v
zGuu*m_j*>Sr&VNiHdVacAzmfi#YPFz$t4X2UIHoh2<noU^N;-Lj=~)+j5=zuh9JIk
zDp`qM5p?A4njRdvMU;!p%OJV{xzLRubdO|RPK>QYn{zm*Vp*6bfO>V+I8hjhJ48AK
zPM&nt@?#-nhn@k81QC+J)YJduFvdV}>P>`Q%;AA5t6($E66Bn!H65-$CcqqQvJakc
zvq?M&sp$_E>3)nY?ofQ5KeJv><K*B!B0V~inVfh0wA7Nf@WBPfW`@2NhYN8^xRt8J
zynlwK0%-EPVjmGX8=h~EF}Fgy1$2O#j)P-kfz%Qmd+evMf9S37*w7oIT~nqPnh2|b
z+h&5KZmq?vw*$FzvuvZarCoOw=1zbGNm9`yuOu_!ob*USV)w@=?@)YqrQ(>V8B>K)
z8>xcgz)WC{+Z$u=;7jhaIF{j##nwWohjg56cltm4DOO@Ygp(#q<mWbJc<^10LI*|A
z6XAN|e77GPZ$qvW2P2r7vZ*>s(ab;29@_k*<^q2KbV;mqiKT=%wa$JXT;nm*?FE?~
z3^_Xq)R?%k{8`56bd8x11S9Su#*2#MTqV8tVkyAZ=x#uFI2IRxsRA5}8MTQ*p3}+`
z1iB=g(dZgT%(w^0P25#s?Yu!WESBTxSmfR!5lr)SD}X=h0$p};W=7Al_^@y%E=Qu8
zX>S(fUf;o`=*P<#5-G*@uS<xQN5O!-=zaX;s>oRe{%Hj`VOiHn7ryQNK#Y8OOT?p(
zs5}hdA{vGj9sqG0YtCWH)l9UZP`4pYx<`@|h2u+%O@Oh>g=x@8nYG%3#ZUNO$F>`z
z<Zb^72GdAe;#Yhs+9hn0LvTzj=6ndqC6M!M_%A!UgnBW-hEcLeU0yO1OoqH%KNS0w
zf3no(_h@-3g7n;;O>|#eb7Mds^)Mit&^0H-I|{m0^l2E^LFq8)bK$vTaAB_&-wJ3g
z%IuDF>&^BEFh#(8Ywh@n2-?j`l-O9L(~v-o;hg}+7T#y#r;ieL!FYSB$IW_t3DRWf
zFJ;<1^qBwZ=-{H1!f+5TD|{P`D<6j@d%AO5ypsvN9~d8rFmD;FY?y{(g(rYq9(KbR
z$nGe>%pPtajc^)8@2d%Qx6o}_M6jbu1o=So<r;q@j=MjLa3j_c?SI(%>rK$}gtXE|
z+uVsY80xtvLEV*5;|-*9(!+%LU-~f@ui{%Fq|Vb7>9FJBr-~RDk}93mo(6I4da#{j
zG{po#t{tyrA@84A#=%1@?ue@~_8H>9t`VKP?Ipr#;2o7V#!>g*fyfvSa}=)@&xt0!
ztXg8mvk)%3TZ}<{)^lLF-t1$@;sYW4(xE}ncpk!?M*3(pT`Y6VCxUaoXNgpW?h#~m
zHSJ|Z=;!7C$uTFRLDSvWg3bLdxgK2lMQDaKW<dP3o+y_|mqanW*iK06Ae1R2`v${3
ze*O_(M_5i?ktq&f;pfAdVliD|?U?iuNl>kB*)?4#A`dDwyG-EBfGfWY=o;uiF>eA_
z5z2p5vL4%sV0`2YG7sc$JYL(3C0^m1d6LfBffjoe$SsPYFVes+0!WN=6@0CU$h^fc
zvomQ;aa<(Wb!D1qpeNvvT}<2kH9oYM#(~!-qL~DGkd0q2J`HMAT`1ZaORL(9Zv>Dr
z%8N>jdBd?W%H~3Rd-RDmzbkltWh%N=h*`nls7=Q^I~tGO-vo7mrNWj)itI1;7wq3h
zv_Rs;@huRhiSd)xihXlfK6LvOcCt8p)HJLsMYsI})P=<yiPh>w0p_G+a<+xB;M+iM
zFq~IQ*+?z#5J;tDnG(;4aOX}-oI27f?-UqPxgW5u3d@70GJrQJ&Nu|+nGEqpXG_(%
zRG_)dOw}9nz6<Tjb2E^wBU=R|u{8y|@(dVd+9AY0cFN=L?C(M6ev!{MHWcI<PvuCF
z$3+CVdFpYI2p6Zf(!rS23EHOuT|KGxX~uz=`=4O2xeX`Ai!1D((qkIvu#kJ-FF?t8
z+-F20n=VjXqquRj2$uwfTMxdv1h_t^zHq$}#|oi7*{nG}$TY(W9Zn@W{>67nfuq5A
zOktM-p?#|6QJgY-0B8mz6&}nTBgl&uq6}WYn384P`I2z)fk$Oywhtl5m>!Ii%b^$%
z?cSrq5ScRb%M~AiI!g%&DA$=!migGfQpPr7#do49BUkEp$bqnBlg|E1TrZMO<Hk;i
zPZf6M*(Ye+{OnT@3dE6FTbwV#pKDdDK-dj@2I9=ybO{oV=8ZT>pri48qi&U4C4?*y
z`4A)GYLV_CNblTC`8PC0=;ZKYhQh6HteqQt4sGUf^wu2Klr6}aCWDOhM&)?|Nuy_U
zIG+6vVK-`5a#{140BXUpRd0;@0)n}g$()g16ID#`!#3KiMqM!3a(Ab~$&@srpMD9H
zy9!-{%a~sQkfm(*UQ*Z}HHY`aTLR1kM!syCzAO=0e=$;qzb20B)`LnEcQUcvH^pxq
zTz{ndT~H!y<F)v_^(}GCBAkujI{7<j=RL@kr`|YLKx+n-!M>(Xsed+{Wz*Im(mPX3
zf4p8fYk?tR>dx9|d{$yfQKnYiYc}Gwx~#6p3?e5mrQaspxx*Ez@TdUS9QA-?^i{tH
zl6p0by^e_7JCsZ{ZW7}9BtwSVMZ0}{Rv=#8C(zmSrHg<Mmw#b#HW)99cDh(xN?2Rr
z`yie&Y+TM<{k$-WLL1d%!`UG`pi13(Ec^pdb1&U#VKyC3KHN6a2R9c6Tby+AQk)=X
z%7i41`&B|*cy9GJMw3Uf*UM;asQtJb{85RZtkPQqOU%KKDw#YJ?kmE*63a~Sx{fp%
zcxG)S-mQ0`l9lB40?ipIA9~_A0gN`J0PP3go#{KLEVcNs1n|Vh(@tLta4oQu)i&bd
zxj@W1MZur}i~6~XSS)?oh%P-(fnaH&RytJ(Ljnig@ppx(5K~tz{w>06!8eAcZ@|^&
z1#~Vn2NNLJX2M(+c^5^9I8jJz)#YkjazDn*^dn%mhK{?G`>lc;Go2=6J<PcIfSnEJ
zZs;l4j~pnhHO+9b=VH-jq6`js_oDoXCX~na&iF)_v*d&oLxX1gS5WQ%{i8BNK5c#w
zrkA8gi~B^m-fSd?^gIm%SRh=g&~2AnfUwJ~4KY(_5-?X6tVv3Tsd0=bSjlu(i3>6z
zmtk?E2&d1UUXO{8{%i{Y=8if%eY)i4$$~uT^QvL&DAI#j#bTq((K>}+eqnG*hDBX$
zps-nj6uQ<rQ7r*pJg)V}s{+i#25#bFjew?@$?f2!i$D}ym#HDoZY>H%DL66~KX(#I
zZc}g?bcMn$GZ(ker&Qu*K^~rM+&=5_^8JcX*BFIU9s|E$43LD=MO=>K7KfxD5_)0-
z5vmg#sIL5h<(6j`ML2z&Z_}E|+<MC;oD@svVE^=ZUL*|xiQaY7C80=|DX0;9ir`Zx
zUbZXIE{bv?vr0qn#d)8l9EDf)Sd|;1%u!Y%y7t=wDB3VO#BUWg12|nnG{sN;kAciz
z)PlXShG=r{>B61BX8EHmVRfl$yE5{CRF_BP$XHo%^*$3MT#2_uz%1knZM!yE2GR`0
zCkYq5Wm&YJK;rnGZ%h;El3-*tF}}BK@edbsb^d^H_x%Fh;tX$mq9C&*%pJ|{H;|%a
zu>IFE0gMu}#o>zrsUJ26z439G8b^1`4s@6w{e%d4w9$l|T^=jMEamAb_O&wsJW_#?
zzzO9)(y$|q{8@;&&N98pa)e!ZHX2e#(@pmjsD?%K$FHR*btZOd@v4YC1)vQ>EQ+~T
z0D?CI?DpYZEFju!z`2JGhnlgVFc%!(IdTN_i~JY;s-Y_{b<ysO$RK)@9Bo^iCX~E6
zb8C<LGJ*T(TKr^1zPLqt#IEf{(_lP`B&RYPuH+w+{rAQqT<CfOy*Qeq%?0JQV$n|;
z&5AG=5=Jo!;}Id|j+B|&*7e3_V*tsClC$fMHjyyPcJ6zx3<Ym1-M9==*j4hA6Uka@
z^Hm(9%Hyt0g`Fb@Y@N6nT60wp7ge{FxX2Sn2qMXeYmJRrRwoQUIE5aK6-1Z?OzA`M
zH_=4YW%j3XL>^Baj(0`4lOW_{M)0|SJl6Vgsm%yk|EEP_^zbQhzUVwEh7onrS>uB_
z5;7r9yN2Q^(YdK5!E5RKOqkPVI>r!(ZSWdJH15CZl1nX?6Gma>>||77YQ?L}U2%^H
z_{si(J?{TRd3j*%veJkh>vw`_16|8y9$9xyNax8;f>j7XN05uAck3CVow37MFxM$5
z7<st1o!24&E$;k31GkuM93Lp*0LN>~C`^9Z_A+|c7wP<v`I5$GTS2Y-d>Ru6lo%6r
zatjm>3M17nrpzZ5cDs-&z>M(#?pk=sjekuj(&R!eeUy&9<3uobnJVi(0_Cnka$#^z
zE!%(x%SaV8Jsh8`Gl@vBxM!F@W{7Yan8nd>alQ~r)yAQ5buGf~Ryd3qh`(l<n~l1n
z7|)1q$s;Xy6V2FcGO&xmY=t>eBhD)!4z$?|55;{Yq{n4-fIW6fkso)H24Z*7W<ndY
z>R#pN3hIHlu%xBLj1|`*x+~KLL$GL=D8$W-(PeKxhncl_`v0gp>o{Ah>i<iLfPe~1
zIwhs<+@z$tJCyLuoHH}$&Y3gAoQXT5q{JmJA}L5phY0vpM7p~}Qc_Z+L5$z~vp;*D
z=l=dVuV;PE&b52(wbm9$4V$df^Kam9napjiZwF3x1nkyYvAaO0IgIB6_RSY>2;%1D
z0S6R@Se`x<Mv`rlro_}A@lCHy2I_H>sNAPI*p%Xf(juHCH!70h+?%9Kv6L`R6FvvC
ztL1Sap=M`0o=~(yUa<_A7g-5cHxx=K24$|RZ<i7c?9Q}i{BC3a-!^ShY_<vCX6;a`
zBc_XV9_qt24)7NV${QojZ91MN+0%iW7C~WWKQ|X~k{~k)ZEH;jzZBwXvyH=Rb06xD
z+E$&RKj?k3Ly9ZPr0^0{l`75cWm5DA`9fP{Gh$?ZBk5-gU5&$q(J{4m(uwTJOfwPd
z5-ir+od0QqLHXQ|^+mxKT-e9q`X+zeg!ja6^B*#3ygWydBhG4EFDlQIoLO5jx<vu7
zv@!ReDPVkuO~dLf33}Y6f?e8d$2S^J3-d6-YK~3mme6@ZO$GvL(h~(zNL<~<!}>Nq
zu^5&++$ZwC1GTaE?pFLu70~dl#9<;F6g79rSiTnGYizE~v@5iT5{s{``6riy<v*?$
zO(nDIEEGt#F#sDY4efo-*|q@K#lYyl3UCi?p=iPEuS=C>wgaU^TpUb}M?{!0SqhU(
z>?Usyoa-ccnXC>#%;Vt-VzlVr0l?*B`!y8*E9tgr)27CWJI<EGfS#c}M+oiQU+1;J
zn6(oW#b7c(TWn{D%w%~@Oxx^t1k&qSdj_c9zeU1x-p`2(cj4QmX-BbhG!EUhNPz7j
zKGvk9bhkj)Ul&T~s%amP(A?-!2-9NnphBj?5N+4pinK76fN|=?oh4Y|$|#8@;vs=V
zV`-cbEA0W{QK_eSrP;NKAZI00AJk-9vCy6chC%N@QZV|CFgPQ9+j^`oid1=Vp%EvF
zKv>*A5|j7h6Mk^%q*bx8HzW<f!3!aJBU+-JKVvandmU0j+-ftTTk$VJ<}^YVnflU}
zbL|5t)^fU7U*8UZ!xx<0Nn*R7z&zt{uc+(?1r`0XFGez*bn|9lM6)mdyNT%7-FO22
zF#y%#YNr{8iZEx`_2)ylFjW;-Yx;EJNl3i;PYRkh%Ia~6X!p+n`OD<cB?J~q8d{4@
z_JeYHy>Z9Mv>?x+ofyl`66xYm9c)=HIRGGc3%#<SI(}GUNyb#eE5h6d7U5tZ^CjND
z<`d?``fFW!URc2R781?0R@y?|A=IVjE`-@J-YyXm>!MuSwiijsrmz)1=_ml6d!Vrf
z1*o$i-s_1CszsVo;$b5<BE)q^NxUjK<{^S$y(HxEXeKayuf&rgTx(of=!OOlEW*ru
z)c+>OM<v~_iFm_W5l0SM(lR|$m^r{*Kh8jo6+j}rysRQ6m{ZCC{aAVR#`ywVOhi&i
z4)T^D^QWqJKdP}x7YGICbWn@UL^uZB7B}Vy_XN54Qa546sm5(N4i5?q#A6~{7u587
zV~%d<+`GA&qvNsueIhWYg7JFl-W%47fV>5X-i`Xhs6YqwTi`@qaXOd_jh76mozEx{
zj-hJ9-<Alz{#BPddy3@QAE&Dk-268HJsH<H?yzl3ukfxc*!9*TtA%;8AO`Hjwkffd
zzFh$6iqgWfdW&>$FsweBO8=x_w-KiaQh8PY+-1h>h<8PJe01X;eV!&HsSaU^!!a3N
z^I(!Yk@lr65iS$X-MG7IVTa%tIvG-HdgAhZAl{V7*;`Mm3-Lc}JXttBn2<Au<8)DG
zs&xAj_x~%*&Bm6cJ8tWTb{&x@s6Ra^gvp|}J9avVPuB}~^7vXiK!8(ZQIiXQ{Co_6
zInVa~I8YS)!-HWA>f3oSMGnRTq8*#@U^lvt<nZi!;-BUJNZ2J}=jpl@aa=w<dqMdQ
z8+5926r|(T#Y52WdOD^n@$4`Jg~7Ul?e>?VJXz?RqA{zB?h!}88nPY}MVQ?Txg^#7
zg&=3yZJX4sXE+9L=nlrtBHWj(?;%y6-Z&zM;fU84JB$*BCfDPX6nTG<W(4nzG~%zK
z%~!U^^q2Pq<c=tb1KBJ(2IR6QSH9gzf%*fXuy<<Pg!o+FW+zWhdsn*1&%km{LoF;G
ze-7=ABJZ8aP)Qx-OyP-7D00NN4~BA$*-pqFrdNp5>91pNaGhwF%+_Yz7&rvN8T&mO
z*2r1H9D#wt!x*B+cfM%h;MzkAVk;&dS_GV~dj46otIqWSVz(JF|6xGmb36lz$Q1If
z5|uYY>CQ1H&3iausfaz3o+OH5R%9k{nF!a4DNSx;7CfRr@aCgC(nW;0*l3FI#JXtD
zV)PiEKb5F-zh3~lJTixt0YAIFgqjPSTxd0xgL4(q4a-(vn*sq&zg=#Z);zMHIX%|m
zfRY}qU=z|I+T7qBs-)$4rXbQxud1l+&J&<2C+AkX99_gf)|qVTjun)f6P<5Xw&R4j
z&!QE=t^wcYbh{e`yL7DX)wo|YDa)xvBaS`>$`$l;^RVKUnP9FU4%C^X7d#dKj`KEO
zGZqu!-bUfk4#`@d>j^hknHN}|b;|wLaeywRcA5;6-AkldOFOSMj(6$|l1dydBJ-xZ
z#uH5gvG556l-x6s?3f@5G*U-jWcw2#@}fiwStI8Ffz%rJfLcW2IMFT$hr2P-IJu0L
z9>;`Ncw!Z3cF1)J@Bg6AT1Lf}m6rXdgpfEdTS`wLKFKsQGHX#^TzV4!Q@ms$|NU7I
zu2HA-QJC-E6J!R%KBkgYPA&j$Vps&vD*?8Un9={>mxOR7Q7DwMDV?MxkkYaP1LKE@
zZ>v+l$QwVhOgWk0xKxR6{)&HcLFn`HSh0d2`hui67$}wjIO*d`uNij;A~n{y(fCkA
zG1KCEkX|<$^PCFml5uIp<0V?46X+VEx{a)AS%q@ux)@dIH!q<US54C5+f@)#DpRn`
z3+p1CiWFjL_c97Mj6H+MY2Fe<38fbp3!V<)zRo^sFur{Tv~%s^MOn4*R)X@l;Vv6-
z=e9!J<}5u!X=ijmiM3-vT<v3e{MVU;Bd6q7=HRo6WYl{YByW}ojjZ#_E59a;i!Wos
z8L|G^1;D)_3)A@m%+X@HYsK;B6v6P&0w3-d;9y+$FbgKbizkJ$hEK;*^f!FE-duhR
z$2y{2Zys4<Z#NnXp9|z>LY#%|j^x4L6=-Hi^N!u?Iyu<?XVAE(Kx)qUsuqulQ0K34
zzF`Jl?L6o_aWaOu<IzEB9idK{6+yRI?PWN`eq0PCTgg|1y5u-0uE(tN3tI1E$Bm-R
zLoVbnJh)kav%}tA9yrzCx-KZt_Bt-#DmtykpM*LxZnbdUzxIUyE)_TYir6iL9><`P
zF!=$IMRuG`w!Wx{!K;@N^xrLn)w!*0TAcn{$7Vl(!bMyl(mhFcYs{8|al0@#1)Yt4
z-xm*-5W&#atj04Xq}#kY&m_%#aS@Sjsi)E7dxBg#nwhzLdjaO2yl=!$N*ar_O57#d
z1yVPr(FbGn5-=JQA3^A8i!e7k8NB%#a_P%J%_eq$J=E-HLa21DITlao8<t6Xlj->*
zkuFpR>;FKka(NM3qHQ|EjzU}vPBT)c<6tJt(NGdbuD{k5MIc?sq-Vo>g^?8#t31#T
zWs-9ocZG3g8Ka$v8JScwZYpC8VnV&#l|=%)Vi#|gBo5N!yI1iq?30SN^k)tf?Q-&@
zWmh`HwuE`yNP3Q14O@UAfv}@KHWugV8)j){@2InZiyMWwRTv~#4VFnTne*a?>fd>F
z8Ay}GyP`?59Z#9P*YNE;S(K@oCaxz1(r(FN>(e5<*5j{)jSt5i*&Ya%44@frmuckt
z0~+Ezfw`@c2fc~E10YwqCjVI9WH5&NsMz;f2zLRQO|b6&Nr>aJIif)SEdjI_=n3)r
zA_NN#hCR;+$lZ_myQfAUp5r>O+~YWKBCZlQ))MNF@K|<NyeBHm!aeLAbP+IK&7Jj+
zyTw3fN9--qWo36H9gstWX!n#99j+HeC3!+6=DVJ6vo`Ix)3S1%KqsIL95csXg{Y59
zH>NvQz5x=}pt4=zkp&YxdfXce{{e!u>%801t8Wi==`3lDTk!uJmQ7w?EGfbq9AWlt
z#Vi4iAg>5=d+@X%vq#cywnW;~_Wcpm6&~ZR;2M#Rmv409=05>t26B+^ljZf|Hx{v`
zb0fa1$kc&w&~|$M4`pp#9A1#gYE4{;4@Ks|IU;8^^ZmtfF($-8vj}J3ne7*F8#B&U
zT8dq}b_Uy*;SgDJMjDNu3i8n4y1yB>mO=0kgTCcxeE(*Gz)^OG@w~p>Fgj$>$M(7f
zgi5e`9gHP!g>a?wyXCcbP?)!;uvku&?UVEw;ZBPC5)Mgm&21oLglI1gxSelTiHo$l
zoE8oXa!b@uK*vl~TelVN0L^1tE9e`dJrEI4w&Ujs8AdA>xD(hVYoRrOF3o5xD=fEm
zUt?I^V?#lyt~_e0Z}S{E67C&ZWq2;jyL8s=zsup<QOmng-%d`C4k@|U1i8rUpkxjw
zbhW#Sgb|<6*sXhtm~y$qvsrPhAP);p;EX=%5$6ho*Ho+$i`@&skirRetrFLYa^@^y
zvKx=b1-Z)fTs^$fipTGB3QU`|SoeMiw*Ygog!20eftlDXr?bI)4?xq+<$-{u^IDPS
zh*aF`F~@^N3jL_;GBD;7LJ_f~uE!dgV1~q>d_ges_7g$E5<@c2pCFQi$Mq>=;zI^l
zXNGzAW+2)H&@f|ci~p=|YR=2+#Jy9L`xhO!&Mw%8^0tS;DT$0G;twLmk79(9Iv73R
zA;GQ?vMC-enf8byG3dHt2NC4SYl1Q7qkNm0NrbL;TuK-@bVx+Ig9tZ3y~$pOg}qCV
zQ^zktuZ#=;+#LpL%xXF}R<Sk!b=eB*DpEi7F<`e2u0T*>h^vKAo*o2$jmQ0sSrirm
zd{Yh?IL4e$K)9eB>$$NbsS#nGtvdRL;&hRD5bK?V(YRfRY9d9o4@JP<Zaj33#gwO@
zTxKq1+DE-l-At%+!Jfz`yhDZMg_%9CyG#N-Jb4SG8K)|c*`&g0*YHpoPn&7_I!t*|
zpcV|dT8?d=hN1$a>6jF4&p<l&4$iyDQG81n5z<T7@#|-yoGsEP?AArQ%BT@)qWkl6
zAYvK=HC__oQnAvp*=fa-fA#-8^r!gl^AKhNoesfB>OD&fb*!d#lGLRsh*C%djNnof
z4OYWY`xbpu0_{qqse{x*i1{Ut{4H&=;%6@ak|yVMJW8Z&t`<aHXkE-zoAI3&p<Nhh
zm^Zs{6IzW~LfzTeevtfaCD4iFl{b1`A_S$G*fupn-?_qrH722fcwJELI9RBs6HWbP
zFqnkbM=nf7n5VqHhy%p@=O{&p|A=s*(h_a^p!NR->W<e$b7rpBAak!_{Y`J~gfO=Z
zXXXYkYVRz>HNgulOU&zK5RTq06l3Rmogi$h5PhDaPiN7KiPwg2K)R4^)7D)t-V^O|
zNneNiNa=TfA=CxovX5g8d!vN8AXxM<GGq5Q36&AqhVl5F5So`wG@F{^M45^0&{AJH
zPY|WiRXe@-?7u@gceaDsYef9?ZBV#MS09X%Gl9ZaBhD8=6LoYq;x>J|jM&8cnYX3i
z0aNnYwf9DkNQy(HXpf(ZGT+c)W`+3bUH^@@IrPovd(Uyv|B}A&X(Dqk8fs5oeJJq#
zLUW(Azh*kz>jHB<8N&>y_*9Tt123e0-+Uh=v$Q)ef6ohZMCJp8H;q{R10XjSN3@tG
z(q+bxJ6rQz1UO;buILTRgN4A8ejW?BvJ8$tF1p6i|KLwg{tPS}^-?~WLp&+mwUb1V
z`BNk3ZK0lN*`sU3rs;9ghecr4Jz4l^P5P5iO3Jvb)7bBb9Pe1`O^L<+X*%Z_TZpEZ
zC<sSQ->%hkyn?WG5uo`M&!6$_k07$VZrqkhrMuu`aOZ&UhgLi#nn<{+AB;IZfiNqP
zveP^9VkF2MC*9VdtVGu-F(&Htj$m9T%;|CqGZv4OVWe5d=BgEc6GmP&Y1S<MDFjs+
z!`x&wee(w{U<PB72se(++vAwo5V(Zoto<{>IJdqA%l^EdL%T=u5D1zI+CFb3&_&M|
zPrb2|Fei?(M>pos#|d##9Ai<z7i}Ks<#`#_F7yQu#p{*Q{)S&dnC&touf^d4GSlh<
z9g-@aE6BN^_(My_!e4>V&fLlA)?itYq{wcq8LNnJH)4cwwLYl5<fzc`L4qF$Ivqnj
z$3u*Q3t<uN<*2j9*CJgGTxCgI6bt+dXuP(_qN;JM5Hdyt+l-4uIDy{ENIWExfzI)a
zy~uN-oWGQSd`B_%Z%4=VO+9`oB6oj^E&oN#danvK*Cm^*3^Rac!?3k)aXZr$Zz-I5
zzOOpp%PDgUF)dd85C56RdVLF5P9@L}(^+Z-xYgRGqG5S=8Gwr!dc<b}ax>~0jg9SB
zf^x~^`9x>2Z~Yfcb!OtI#6F@NDc!xnu<0elom8^sJwi|(5;S<aDQo{9*!V;z{i1Zj
z-02%zji=W!im;XGayLPq9XK}77dq2J1N#W3U09Dh;zJR+`t7_f6MN19;C5t-Ec4`p
zgt&DO(y~+6yQ1B5&K8XAVl`eALbANIRFAFaf?&ky#(bu}T?b9vTx$=;&B9y<X|i(2
z)9R8?=gqK{l=^Ok^VFgjdD8@^L(;-?gF7S^obn>`%XvU@E3G?WV!Tz-9Buky?r-wH
zYE37I14Wqm-5lt(eL75#ONmL>K-w)lB+NCJx_m9E-mE!q5m!2j+W2fBh?bS*OINgu
za;$W@m^9y7WvoFxS<du$jv(jQ;sPOg*1bU(rPNF4vHZ6n@?h0aW<SPCu=sqSWX4J2
z2UFryQFQHT_IUb;NVtG`Q5|`Ad?=dq<Qy{E=J%h(oPjoH>?GPuq}xicc8m}giMLU7
zsW=*E3v&~c?~o&nPPlo9(k!l`DzW$i1f?^zP`gT(TWbm+dwJZ7r$sm+u8B}OWtXRo
z={nyAr>``)yS-lDt}>UV=&&W-gk=_Th*55X*V1?98u$Kbid#n**)X7|#JM6|F5Ism
z5~{{;O9<^_EHy@B!orR+UA9`AWr9LgVn-3qK*F8$l;rLu#@<Q`Q7f9lC^i=oalF1g
z(vndJCuz32(eWK%PkOv@7B7gT<<L`S<o#0=@!PlxUFEwDNPptL*jc1|5Zh)J4=Gw6
zEO2~q)a6y^Fd?p~j4i~=vjuq+432UlK1Z}mpnJlHMVh@MNld6$-(!{UmFY+T$(*-p
z3F&b*iJpHY$b8Yl!R<26=7mv*iAbtfTh!lGwr0=k+p`LO4A}xNA4Oyg_fc+ToNO)3
zg+b4e3mQ#PaX=YTk3@|ktByIpUxc*LRC?K`PpGTktl=A#hCEM@t1la=*lMvNhzzsE
zp??X=&5J}kt==7r7tv&w7YB<r|C2EvXA<tNmjI-$1K86qxg_5%x)l&oGCvaP7SdwN
zhWj`nu%B64Vv%b^I$adY2++c4_=%+e>AY<8upwO<!U;>&K#Yb{Pc7phG#NI)%R)F8
z&XU|M$6-Ru?deF5yW?a5ZuPEef4nY|9LRv(!E#0PiBporz<NT=kxF|Od~Yj^%C3()
z`!0PGy{Adbo|lPXy}^CO^87<SltK&slocFFcMCh{TQ!&WhEbC)A}2#8_Gk4!iQ?B{
zAeLGY!Y#?jO0NyyT0+UgFrl6jlxxM-`2~?qlK0`5!Jqm8fa}4N<;X;sq$aOYr0jFi
zSWFXS7ITTsnv$&iI)pkh23A^*n$fvx5t4?;9w6dTp>D)9`!LyIA|Q~mwQHM(fz4`=
z?uSfujksa;A`VJmy07G9aHl{AYV^_T_WL1p?wL5GN*Y%e3UinqdSTom+MKeX-YBmj
z0BH=aHzDR+lW+3`6=P<gSpsrFc%LH96;0qC*61(w?MaKPPR?!?Gp(=|u>04Drr&C;
zR3ehyqA#_(5L!=*Ok62~X-%*Ca2qAc>>XEFrPDrBCV^`QrkHp^h&4M9lH_}Sb@pY~
z;g5W#;R$BWoh+8Try|%j!H!~Td@34_BYV(cf6WPyZk1lt<T&1w>)8VHfYeGb5Z`Ze
zEWDK}ujwLTM?YiwMSW8zt;I1k5yHvo1tFdvgbzmxBm~mR{^*?q;bP)4gZ9L3niSWX
z3``54xMqJ_hYKK!3QNn0`p$(!h=V>la*pQ(k}9XP{@8sAglmblPAw)*g?67{t1L5R
zEJ20253q$#8<lqixiCqCCADNLzCX>Wv6)Kz*j%)`hXlG<>j-c>n5c6YZBdKk1&`-|
zmIbH!Qv|_oPW?Q`9={byp*ghn#C0N^EJs|r#}@+3ch*{|lwZ28!=ozP8^h~CkQV!p
zUc}6gi_D?%-A(>q3o)P2cIZm#MzO{EfG(8gJ$bh{xrDS6<59afTae3whj9G$Us;Ah
zM9GCL4i7gVOdjj_qexmRs|#}>m}NQH$Av=N6k|1p(nT9Wn^!HiR%|>pDa6Zy$9W|c
z-fFxl$T2y#Ap-ajw0apfw8!b2!NLZpif-aYP%f<`P&mmIkQbIN2AhxhhY2-5kcCTU
z3F*k;0y95Y{k287NDvwE{8HSYZ&#8XvUPwy7v@T$;we+Z?`;g?rl2FmWRWfp;`Q!U
z{3`#S6$9TY|Ho$+<7wqhiWGH$LFIoeL}eI5<yhajdYoslnG@hb@c??0120-K!pyB|
zA99@Qa{_G>rpA;_`A>r!EkQ&U{}AaW(A3L?NitOaw_tNZTX&88xi<qso>m=>#r5e*
zceCEpmn%Y?V^gy94$<D2vgo!Khm1S70MF@ixYY1IV#@-;<yKPOxkivHi9ZWA9q*L@
zGEdg$56eiB2dn4KyA_cbKZqw#As69B;0da@UZe}0jP3^FZ6VGAM|NcKNf`v|L^Z@&
zTNA`(mFl3}J<qib5G^sC<#_)5JsE|+cl4bb8TUf?n=An~H*|x;wk<-8&`~>i<N$!;
z*7b7wIU-1RjGe|(JMc}>d+i_nha$~ccK19=A$okm>tLy6LmsUX)LX>%X?KJ`H|@9t
zTpOq(1-n`;wkBy7E+I4<SY3H4Sx{~Q`SXb{N<a_%;8I~nLU>$BaSV~TS~H=JS7knI
z#ZC(5mZ&lt-zk#8svf};raKpD3`-vql>5eyg}Sv+nv%%s^D<mjGCBKB8rh`?r{$kM
zdV>(>E|E)Lyexq1`qLw<Uz8D;HDx`u%C7wHO0Zm`2bE@<NkTJ=y3;W~w%Z-Z%#<=@
zI-MaCn|BXj7ZrmYJ^!*=iC`43LLDo_Ek+g7ftzitKsReAcPMeb=yATuo5w+W0?;KU
zwM~rI^-V2UO^~sly*Cs+r_#>#*@pW-xLh!gyF=?8Y%bWzvxfG?_4`7*Mu@l(LyyKj
zKQ_ep(M+J@^PnKj4eYgPox??XYC+~MW!k}5?k8XrMQ8VBtRc$P@2m~U>S!||Ztfm>
z>pyaEtxLl(aX<bX?=n*ON&Acg_Xnk{hzxWLjN>v*94_3r4icSJ3iKvcX)pAK{xN5{
zC6dxNz2{Sb&IC!7#0Od}7e4^h{6(#fRBtH(ULZl)?Rf#z42b}n@>fMssO<bU_Bs%n
z=D-0~RP~)g>W1{P{1117alO8Y#nOSMPIufd+Wg_|5H9oLzMq27ct}+HV!jH58LVfM
zIF<5RoG>?NjRk{dNg|3P1e5<LCb4Jhn>6seRgbmWA<TmjCI}RW*y9WfbixSC<qM0W
ztuS+-i-oE;{#M3BF3y`tQV02;KzAJOMkd6f9nk7*&FOj!!2}zkj<{Du?y;Ci;C{Rc
zK-1SXj0WOABJ&cEgwgrVb)Fiyd4dE8-`TOI5a)?+6$I3x^MHYAmE_`d-PxkC$D{l^
zmxzvZRQrJ05|ZlGsiHM*xDu?yqMfGd)O8@%6GdrJw3A$Tcai28?pCC5-W_`i8y{QT
ztTh_RedS?-T>*JlV(EEU2>Ixw!l9~6%8q+kk=N=XgvXx@o@kKRPLOL2FcQy-hSPFT
zA8(7u4M-PGiUMnNgOLlvhADqglta+ln*DY5o`(o?$uJohCf_S^9C=tBjQd5o$yntX
zE6ddZ=v9*^PKa6h_UVCQ=R<-9)cyLyu^41Xt{)cQk|WF+igTw!Q!plL)|X2}QIDi_
zyG)P&s-@RrH)#|RxopJKh-f#o#;NB2u2`%W&=te;Es_$dv8*u4$pfabeEzO7Ra~U+
zJOQvvL5YRH4+!)WRE4jbH6dNvbT#^{XfvsXsT*t=>H~3(x*%c~R*fTsxxwH!qM5On
zIa@GvC9t*SSyW-U04N35lcmauv&H2)LyDYN3eqq_$tE5YVTN)w&rsQ+9{>gp;%K-d
z{ws>GoG0jm0}#$Yx4kuOUt+lypobPWQMy+~4HP-%3BsLl4-QWH;z9u=)gc}5?+iki
zFE#X?*AdBo%@($VTZnK>8ip?YW&RiW-^93LsQ9<5-i`~F7%l)_{ABVxT!4$oNkXD_
zczA|DTCEq~95<H%SQA-abg<oZ!~yEloAFb9Qyi)H#+oxA*lUh8Vxm65nGBn;t_T-X
zLI9kg3UFgK`sm}8I3tHoo&xSE|Hf0X{2Se$|4!2UAC4Bu+4kxe0`kH?G$#GuSau93
zk3^XlXv$wjSS|(^@H1i?0jepcy@%*qjme$oK%72XP>$7=xKV&PrCY}a?iTR>{4+r3
zt0!Eg<GQ~fT8nu{)|$N{$zhm@tfg-%ZCRBz{Qnl>KHjbla>>sNx($agvD(4VPQR<(
zm0o^YM^LT<y{A8p5-?t{W~WWUE-OQEX2*ne!XZUUsIbrk)!0msCl1D{w&#7mRIKX{
zr^sEo<l%&FI~*|YxRS?>bk#CRXwJOSnYMzspAwj70*=zDMVlqo*DgsjNsi7ISv#b3
z0r#vx`VEhd48;2)T-gpx3%FuF0>BJGWlb6qF;qfaH_1s-943fv*xep~(kIQqyoCfi
z_WebXU}Fe#v$*%j0_aSV%b}wH+%x5Ejn!z2qn!l%njS{Pp&}_D(r<DSLGzBHop?q5
z*6scL$AG(G>7o6xjA)PHe&)7)MAFx=wqnsZK$O=j{GVWB%1VMn`Kg%>G*Fq&y784L
zbDFy=32@=yqsIc89aJ#QG4Ba7^AR8F{VwdYj|0;f#ffAbC&F<swB^L#9ZwuzK-~D_
zHJVr0OGrxXYOdD=xiQ#CFiTB60l=)P@&Yi=V%|C%0(<Du`<z%L&FYubPuojK4`WHj
z{+J-Mhg-3DStR9^^-aui5(F$>9}h;;^li2xXRX9$qA5~Uw<;2uzm@}}OLOQAW)Z;L
zNUDY^(u=~1K9mI7yPph1hUp#AqeReq)|nE={*q7USDC@Ib(3hyiBTwWhIm#a(Rc&6
zJAQBqgt^f|RgKKQ62jkHeX;rc6@<Bm={?Q6@u>jg_FSCci!Ecf{%}INw4v>PCPY({
z){8h>RAz$S;9K%EXs0Yc9M#xHK<>hRG-tmuNPm3>=j_V`P&C>R#dQ&8NH?b|9!QES
zP6u*>tTbst+$z#pB5_<V{w|s<=@_&h)7+N=NRMGW6i1v<ghu?x6^Ll3&&^tzH@<gf
z5d>F~&DdKs@l#^cMR=-dO5Vs-D?RcAp*fX-O1gi%QIIRz+rg7*jhOo^5DF$uGwr}*
z*$gmv(N*Rk`@R5IWu-Qz*z11{?H)3I42ypYgWr0~u@)Pg1LfhV-Hvu-QmZ&fFhy=L
z@5Pu1`b<xI%>5fab3efDg9&6AA#Q0NSU?oaY>UjM8fWPbk7?B3s~W%fW}Pj*90~Q#
zjv#>@n!#OG83y61&Pn$OaVOwWiOK87O9CifUD`qmo?HCOxtQ(ATIT^UsD^l&c4K`L
zwKd9aTN)e}3m_`5KGfo&{I3|uW5Sh~c0T{+*&WGUy6)IX7!1>uRB!xQl*`hdOk)2b
zz${@gm*vZb7Zi!0tiuKp+3BvrC=+s)Xy{urcvp;-LC4?PjvItILCyhHo}5|vLJ&G^
z)~nlJqysQ9%ILl+#Jon($+=il!C;QX4U{fN@5yAUYb7@REm5cyjm38Q7*Vb_x)D7C
zcqGFlrO1X?L?l&zt_;PMxaMiR&s<C>^P>YDk!I{FfSUGU0NkT*Co6Fb{Heyl!rVtY
z^iJ*ZqFtLo6o`=vTv#G{Lz8{SMM9hjwWHVVc?kd<Vf1KU@~p_*C_@c|x<_3KK<yaA
zbchu%gK{wtYcOm%YR@V$HgPN-ObizZ^A?tE3?|-LX!U=>JxF@FSH|2(53R*Bf<2FP
zc67utR}{J6K}9NH)mTj!m6c7D2E`_27`<Q3^DG0gn=p@kZN7AN8x-XBv;#0W7sm>O
zv7OkwUZ8I?je#X)+*ivea60?y-R4Sy(b+Lk>55ZCI8k&{Sx``bjlT)>bVTOK^$9^g
z7Mi;tk9m;%r$Wraq$La!V)LuYP+0NOEZYciJFvr>vVI&=0tjVsB>o!#Zv93DUnJkV
zx=1%&lc=B}%(=^vIbwpvgt+K>no0|QH7+iLNduJzxmw5)^M8*nu4ZPqs;!a~X|8Md
z%lYDvk_R9}!%S&<{ZZd;#}N$2ZWoz*G<tcXvBvL;|GUs&u8DMElrN5_yLr4^V2-2#
z7sp%+0P7h}ykISY+{TDL^Y{LPL(J6pKm=34N)zME>x|<Apf5JP0m8}Q0iOk8YXRne
zi{ovh3jePb<~p%rbi^B?$z>w`l)C@Gw>uU7DYzx=ik}I}Y^CopQGHn=yhx&K7WpG#
zoUi0Oow02R@H_&&wJN}=P))^oW_Bb=#@{R>(GlU`5`Q8R!x7)5Ou-XGIg=jtkGwX2
zBLD>_Y`m`TeB#IN7)oXNUTBw_90)ywI9}gQf&C(){ssQxKPOF_636S?RnK>6aoEj3
zZZavn;{K4m*MkDhQaKQzzl_EU!W<WtC6Up`CDbL)J#g~mTZ-f)sKogIY`9w>b;Lb7
ziLZ1U6wQwqgq@rS^Q_g)y+u!KDa3h`0)Dv=Je(=aLlMsbXmPQ7<9V^$b3%Ijc?}V6
zacmm7;SlYF8x?uoh+ZMCfK=l2#=@*J63?nN;w#Z`TOJw)?wBosWMMZ`2w5<NaAv_B
zg=i`QfBG;yanwd1JwxBFLNWl?uKsReZWlPkR&3T?0HmgSG;!ix(XOMMM_RTW^WP1s
zu_BfAqeM6jyx*dLexiW!Q4L>MNj#`q@!`FWC?)4O>plplt&8DWAJ1Ra;x3_P2seNz
z2h`$KK_2g2DERa};6$o2P+#;xh`a=JBdtIli~%T1uyAvF1RuHDeU}Jsxn$5WQwY^a
zeR}SPAaYOY)$<&<Iv3<+1g{BPyP*Tb_Hd$Lc!ZU-j!dzcXxe9Ja42@r_xQxeDy%(P
z94?TaHXZeMG`B@5*W}CYVH4b!aJ~BoL>?tIt%GkrS^#YGn4~@x;6kEDkW{%gc&rFh
zYYwB*vttQpN<|?|02!iLI}ocr4&ef3<s6vxJArNq7HEvINVyiBPZlZnFs<_*iU2Cg
zu02lBxB1&g*XF+bS|RT1xFD;=Bch#C(#O!_7_rAwfUts%AbYb`rjkFI_)wHPsMghF
zm$}!|U~YO0t7GnGAe}20D6ER{T_GgM64EmmGs^!p9inJ=l@OPLEdq}ms6pm@7SK60
zyulcV1%+ivvo(C|&|*NiI{}+z`9w+b+sjK>;y!kkNgHDG=Lqcrwve}SCAjHd3j{wR
z^6FX<;%08qe_HXm0JERX5#sE0;(p|LP@0WBDyG&iI1XB8%+U1aI|a}dO*C%WUgSGx
z%38y^{-}^VtJF9iGB4fy5|9~DVaJ6;pUGmwmjTJ6V~G31_eA88R~gky9<jtL1tih6
z_K$}OGVeIo)G_Pnh*`p10J$;cEIjfmh^wR<3;6?NwtlHZ^PWeokx=K`iw&lvL2oLc
z6Vzju%##VpHNuUd;(b=8*55q}|C2`3(ytMf4uMb>LEikYLydPz=F-Wc&HG^_E{%9V
zfcS`PcAhu*&J18K$2KE26Xf{qNIs%f2B9f9z2bONkQs|ziDpU4w%VHoNKHBX>C*HN
zfp8nGy{<UF41v>goE~UtzE+@_%hnEQB{ST;!rTGrP<>p@rJi<QPOJxI(3tubVO;V~
zxLxH+e=9*Q`Sd<quWnriL4*lQTTv0@!L<^y`_s!1+JNXF_ZLBK2WC4KIi8LHaS74w
z>e9p=GlfwvG(!jD`7!`km#CX^H}f}PPDV0jM*bH<V0)#z5%a(8=qz>vv9E|6RP*aV
z+>rm~u5~nC6hV=b%dUZVIseyzvT!W*4*!)hjHJ?>_L9r$k35eLVf4yeD$S+U1iLRF
zszV^XPMHAL&fbhBt%V(C3pvK2Cw3R$QtDkEZfu_~gUGicj)Th}vJ_*TcvA>vmoFAB
z2Y>voNi{Ni;Zc!htegQS@$Cn~$|+PjCcFWaLG5Jf(zMmc+W0*}x{8JW7ZM#2?0zAR
z>4BTm1-aJKlg8iK0-OoH8JJttfBzwnCS@|h&yol=79Mood7p2F7uSrfp;+SsC@02^
z6RGbgz+I}v3o|E(bd#_ap}vDWJ3U6V`#<=DBq!4EUeUJ;KoKQoV+Q!ahk%|yq_b%w
z={QO#(Z&#p-=c3yqP3zDYy1<60<+f`jBP%G$fH)y5?Z3o9dS*k05#FDV7DU9x}>dk
z;m07Z4-O1if&VE$<($wK`+veWS@g=Y&_^QjNachp4T_lO(;{|vWrSO%?+I`qTo%gZ
z{b+<RXWFaxjvMhKA#No0+o)mI;wL2}g;ee9?iJ*sv2o;h+4&iO>d(8u{qbi}PPW#Z
z-p8|*p9wJ&kZSbB(Vs)RFkZ1p<5Xd0eOiNYg<p-!zbKMo!clM7{!0+&-PtFp-=F{&
z8Ev0*yLO8pIK`bIrY(GHLXei#un3n7U0)?PD#RrlmM!cnqH`}o0irL4{{`UWk(4L@
zTNes)NhN~j)hRmW&R;trdw>{>-9$THZ?+6u;NL(_8}S`Y+?fG&8(d5c#IXV#360h?
zdw(LxMdrpwr}Yj03qtMsaC5zzzVnuo<0RSy+G#N2g`2@-gn*e=TljxqZ~~3u?pR#}
zgTvm>V!m%|J)VnwuEe53Xwr5bpy-LIqMROEiE6S#*jN}z;Lby3zf<J+&Y@EU(ZsgY
z97V)rU!4{tVdevSdoI2w2%u#0Q5R>6@UZR2CFy4+S>ch!pUzpt<#ar~FCG&BH!36b
zxNa`KT?AD*Sw-`>o=`Uz{Usk8ZV+g;PRD^RJD*$2U|LQQp3yz_nWsoG?E~#kswUK(
zM&735LeVaA*AVY(<iF8ENrIc(=jCtL7l8|{^VYYDRFho~4>2)8R_6nS<6IJ>&2o&W
zJQLN?^VFOf@0K7MTpVR|n0Qa1JB?nRV9ORC3(B2FN}Pl7A0aMsXCJbgwdRL*nbZaP
z;`aim15<;#=6(x8=1mhTQmd7Anx6=zh+VuMfyc~+OyPfo>2@Je&NDf-uEsBf<UCR6
zkCndzO*1oyy5nFGE_FW-Eb0`TActo=$FoZcLXpr?=C&>VCDI(xyYo_-8jSx5bUBl{
zJX0!zaq4$L-Cen4#?1jW`ca9o%ppMLOuDTw7eeldC73%`kgJgdgne;I8L=jhW*UQ2
zzDFpk(vB`*thz{%o1_D2(K<#L0m!EjSBWsQc^7~a{wMigy$T*%Eb5?`TJcPSh+HDP
zCkXmLP#(+pE=*(Z+X3ckrzJ^hUai<ms5#V$y9bUfY_<mkI=XCCP>kw`gN3D*pfF#T
z0i-9AE*sbSJ^|bmFa{T1q8VlVxG&m&F}}z9n?_1kd@jftbAZ(}3HZas0iCfn?A*3C
z<JuC#Nw#0&Sbd2i9H#@C1zo&$VQz)=NEZ)B2ubsi&b^B)2}K14a0C&1i*SLk=1B!Q
zNRY(joiRh7DtcomF4MQG)h?Z>&qb3c{;jwj{mxPjsbzYqzCEOSF~HbQWUdP)xUhl!
z_|jl*ReAZW#jgcWDDJ@VRk92O+4i!-++W|W4fk2n2s%=T3&LG3t_ejutt4KJ`IjxC
zbg~m-d@U%%wd9P;%fF+sPzm8^tW&$>50)d83!2QE)exTvb7ovpVa~Wbbl!qU-8pN|
zKPfnOlyOz{l~!<6dQS2qb(LtZH>@I9BxFOUKMMEeicM&m`+ahJSbw<ca!1Cj@r)3f
zrgl)=zapP`TqJ*8bN`?S&Caxk&e;^=LRI87RQ0(-n4__VBZ<04K<)r&rzp%5WjvMw
z`B7}fGs0*e+Ik>9$N}t)mG)%XY+s20dCs*$KQqjY0zK3?bn_fG;wUU^RxW~~@Sw9i
z%>T(MfG#ZOD0J7@B5${9foZ?Tq+N^ch0!-ql#G4ycN=!5UHYbg-F0rCju&Chb;zuf
z>n<Vcty*y7Tv0hQZB%tX@;hO!fEGaBf?<-IxSFG(UB}ZcgQCnUFLS)2C5UVhsrAO3
zt3$YjIo%`JkQq4)-AXu}i1`D{+FHCWnmR~&_UaG$c2SU><4lQVe2z81oDV|hVI?t5
zkZZ(|Uk$fW83^H8QfWO=5b0u?7C&2)Zxswjd~ux!$3{XWfmPpHAnpfrea75XC7|R6
zW5czJ5QsnsF-w3MM5FN<sCJN73QR+_8rSLDt<4z|v13xGdZUbn2Pm9XOQUF!bqGr7
z22odvEk!tC`4f>g&}eKY&;t+VquMT7?T*lD`UJjdOk@lkHQyHHs_0sv5(~E#QRN#u
zHW8f*KFp0yGWcJ7BB1%sdCYcTs|X`q&hUuX4wwW*j`9cjq`uuEoB(iM){1up5jeGT
z%$y9B$5jKp>6m{?0dTlx4NZVck?xk5Yq{c8s9dxetVT(N<z9iVIvaT;aH3sQmJFLL
zEHV|$1;VEmu7w)$eL+r(k*!NKhSi1w9XEO3Q(xM7T`<qjGwQ=JoJpAjG7c79kAIyz
zd*W&=CwhDg%c~DFG*Wk=G$8uzm3U5sm2%T#f|BW5>NeN<{Nt=TM*CwCk?x00`WRb=
z4TP9)t#rEHN&sBfJ0jPWfA+97tHdV)NRAV)T%>$a{*Oiq-D>U)iuCpL71P2oLd+B?
z!_&8@(pLg=6|p;r$2Wx5Sm{F0x95)_T;E|1Ns^6oFdr65YcoO9BKK|N2<+YKan{BV
zt^&&jE&Uq-uBFBrJ#fM%AWoA*UU$8{F7wqjHU)Lru;N7Gq`hkbDO);vF!A;&(0Bvl
z*GF5B6NJGA)*sm>+_4#?^JALqWpSIdIS3WvspeXYh;aD^&<N;@8wEIDEyi?Cc=AAw
zqW%?+iJ%dXoyF2y@a>+K-cezGT33)0pnvuxd%JkG42M1)nqMtL<}(X_dwi}SF6eN5
zKw`T&wgh28>l}$m`gGvB97FwBv{zcLY`bwJo8*E^Yz0i+QK-{Ne{YerOJBz*^ZL76
zLpg)w1gkwh6_UrL#CNgVHUO|jn$+<-5oRtEHQkmJ-xoG6Ku))rj3h=)YzywRaSx)o
zZ*3vs5|hxM^zG8?;k~$DG!+{`_OP9UO>A2?D%<ltUYsfpMei2iOBVSsWNyqir|4Dg
zz+dKaI-E+h7&{BhCGJ6wL&t$#ggI~YUV3AXGE7o%lP*oCFsjY%=cKryjDRK(t|)fo
zPnU{4AXDGF0^DC2HA#02-`O7uh1c@v)ED)gonmLbvl4$0;Rtj=dZ}Ch#=N_Ly08P|
zHUozVBM~|t#kw?8exYZ9U5iw@{dAwUWj`)a`4-1YMHd3^3Z>G_G-*tC?gs5Tp!E)?
z-Vxw_OOYB{Ll)c}hz9FpJG+Fw^SIJtC7I4`!ra6|%umQEAKn85wy`flHEk~lijGYY
z>F>5Tl$(d+t&I6$(McuPo}}A4QC=$0Mbp!c`0NxwLP(3*I=?5%9X&1WwPtdvvFtv;
zxuu8NyI^R4{6!eW;Q<aN=6xYNx6YVRk8?$mD8~>Ed@qV}?c4jXj%L)pCdl2SZ5m2S
zpZplwwPjDhRwX9?1jJRU%1TeXjwu3N4ix*i#X!Y%YKh6`zmDoLO_+N!CIuXPG@8^^
zV3q!I&Rq8O#Lq>$a9Qmz4lh$o8dGBCRl>|SW>Kc*dG;&f;ieXCXY?ZlIcbJRZ>)Df
z5d?8LC&*@eR6_C>!84w*!GVMtZ+w-L32^35!N?7Pp#%-BI8gw-Wqq6;oThJ=7)Bvd
zZpFER%s%WEbO-#X5E=xX77fcqDvsaJ14sYY&UfyeFq_B91voZ#EnTQ+eImrnr+FKr
zBl5Msc?Td3&Z!%-gsKU~a9dm<!YK~ox&x1OH2@cp8!~1LM5`r)BsyL3Y}-jFdI>L5
zA!<KLl)D~Nk5tyvrr{33F5IYWXt{_I;=&<t;lXI#?mSZllV%R-t<goW%-575d|FqS
z%f(%1+EUEXT|nGnaB$_J3qj732}q`sYYA}$4;sSWJT?&kFFGZJ?-1e6KipT3Q_8@s
zG_=L_8DNuPx$9d3T(orR=O7SY3v;bTI5F8?c%M3;OTy!rlC4Y^;uQfCS6WM^+^A&y
za5}h4IGr+$#*G4;2f|dm>uRqd%+;^Wu#WV^9w2X*(j{SO!?eEsp^Ea96F(P0@(6~R
z=xN~v8UQXXUfXqx&9bzVKo>%H4b9k70Nob1tZ|9H%_X#1V~*Y;EM6iwN^dS87Z``8
zN%U}~Fc+B3Ij5y+yeEiGgI7kz(;Cep06S-1+G#{pka^ZUq<v<L3ZWG8lXF%d-+A~m
z6UD=#DOHwAKQ59=^S~=kANwJk(qPgwj*Eo2)-5hY<$SFgYY*f&^=dPwh%gIK0pV%v
z0|lh1umcsL{_<~@LzbmO1mtF~H;~D~$~aG8?jPuiqz-;r8B<C)_)jkb<yo~E3$=(y
z@#zzTvBY4JN_rArEws@Pn2U!%4L{m70eNS`o<#GH&R0ExUG*Wf*l5bbgt#dX*kWae
z_cuZ0hV?rh$1;Hh7klgBGCHagED&prfSB)Wt&#PJb_R%Cw6v};1B8~49uDM?NE&T`
zd&bN3?Z|9*`r_FfwANA!#HyqGO{LPLAz{OQ0`itYPNd~+1TEDuP-n@}4&hd8C4@Yt
z@G9ir%74*NmlWX#KjUBLN=K^2UZQD98UGB#x<7{^Qx-!XtTz*o8d@sv4N*i!sc?uV
zhvJJ&qf;<a>FSEFM7xQT=1)2c9pV(^A+-@p917*0k+$;53EV0I-3n4<m$~GiATes7
zJ08`yTT>Ray2TM|zZUGwRQSdadyeZ51EaX?5V>@laRijp)j~VKzP<7bAXi_uHF2|O
z+6S{Vshd72(rI=en&vdqz}5iFMMi+v>eeYPZaoT+jPSXS6lU(Dq1*wHx3Hwo5a3qf
zs4p|M<1&Iq4Qsp@bFmOMupJ}uh(76aY>682L?*gq`?Ah4{G$p<;VaG%<)SbSS$gFR
z_;G<$K^=uH*sCH*i^X~%Ce4H}2RO>7z3e7};6R=~<5bbEa<Z>e(dIoCj9lRfGyR&P
z%rm5F^;r5iXqO`CtFVvZx?7-G(4QRa{X&3Q!1$9=-b^8+iq;VdQzwXWzSIq=4EgpR
z59WEB{h<5?Sxt#Iu}<J0Dp^Cjh<fcH(ur#ag2`7^h-=EJCAl6uNRW$!+6iOjp8|5@
zvmK2kPlP5>x^Hh+Z)_%VyhGuZMq8Bs2y;eBt-Kmnom52X>BEVrNC#k3pnmy$iO8~r
zjoaHoT#&AGmoV?F0vO|-w<{JF;H;%{lH9aU6*j(gDAgGb6Fh!-&tX;aOpJg=<H0gK
zB|*H-4Gib%1e-C;ZxX}A6J>7loQ@9^%qWG?ZK!rnEPOHqqfbg|zty*k#;(&(fL|kw
zzzv;#pAg~NHL#iE^^;!$xPRD)j$LYOEYQ3jM1yO4k!}jy$PdI%%Q)0X8w`BX+)@T&
zbsmZjOPa~772iIEM9lkPHjRv(g?<H+d8Vr+?PRFjse)Yxz3Y>18BP;szKkTL;?YP@
zZUuxe2m<=zbOm$mS~|$bon<_^Jd_{LmjqE6JtL@Y{F76m&45;q-agh@_E>?=PnUH%
z4Xkn+5Cej<5u1;LPKR>t)V3}&>8PFoW;U@as>GmZCxR{+YZy0@We|LIU>YUbOc{_4
zM>ReZLTgS$E%!TT@~xRhR=Asr;Ezi0P#l;4!0Qx>a3Uxe1*QAmB+9!paajvPx;b3+
ztQ;!IJes0hRKy6a_`PVCgL8s>q95>U5Opz>9C_!%1eXU~5{htGyvvFi)pJ0c8riXQ
zi*_^gB)>ol{|3N?;!uwl7e=e@W<t1=(x#SCzqb(cjME?DqtO@><ZhzfCN@ZE`gu~Y
z+YL3DbXC@hp>qM<Ec9k3cR2b>VGbzw1oFF%FDZe}XUYUSvwKVsot{99ym-2d#!ABS
zEm7I00^#*^-s5e?M&}iY$VZEgP4CSHu_=}tuC>o6l)0mq2V$ydm$B8?>2wb-F$R<L
znOpI!FqfOx3~KlqL0PeKL6I2!v=WD22<>_#nF7c5s|9)M!7_w(ae7Df<}zUSP&?CO
zHEtC|6SuT*u62<U<~);*dpnDE)1(`Lfp{T@K=Cw9>tnwq1iWCjLv(SND9u91$hN(>
zh&$9oB^4!kA!IdzpQ%{wlHzZ<5Mb^&K?r~IPBFI;B0Ohy<95AOo57x1@`yDqoh=@R
zr)FFw#Pig&DJ#dbB6I5_AwwSw%l6B_%%Hv&Cw5l9Z(af9?kP8It=LF_;}MBTmlFS2
znCs8(kSl9m=P07FJ)^JQSORoEqzreLfS$eNp$q}k3)xOzOuiDr?ZWvOiEV0^p9!UR
z=+QM2iMK?%yy&x3YrV~w=PDrAvW|hrS|Z_eXJtnGsQjy!xowfF`HTE`47w82MYx*m
zs+rVI5<tWWoR_cHcix{SMbEU$S>+ng@u8LuipY*o;u1_5*+?b{>x#dFHg86BH0X^@
zg_whA$28KrGP_>uSS;8{KDe7`hv5#8AqJNZ5~fL&tKNY)T$Hm<p4_ue;4HzptV0bh
z3etr3!{38a2)hnfVqHQ!(0FmRH{KTQ0mN0Zyn1u}{CNo?y=MX7%ocN9M@Wjnia>Su
z5j8&W7@!ojJ7$z1?%vqeXzlo`K=|FoO=~Q9J%qDQ7Z4YSrr9JmXG3*^NVq=<$7xsJ
zz_*)*-4>GiJ^lcYd84aA_6c#1Fc(8RH*1mX`^O??+Vw|8fQrs(6|Hqq6sTAIbEQ8)
zC`a1+R(+Gf2+x1;UN!n~U2(>ZV9o%q2E2ZyLr((TwzU};AGP9Q1#_z;Q30D-iP^6f
zOda&dq~0@)KMQdExf^ba1^*09_HAvGVnpAraA)5ru}>0E^uKfw!Fuq`n?TKm5gdZE
zqvcvan6pnCSSei7*n0_firnwA$-k)t@Z-s-{_bB0k*mk#EaQe|oFx!;;kghFy(h{=
z<`6O#ABlG5F;|hY%bYiZ<ZO~Gcsl_(RrEOV$jqQ^33Ry;veApU{TvbIHE((!YF*@)
z1eq5JC(m@FffVmP(ar%^XO#N`0koms!HKE2I5dNZgh!*dq%r2|i@%8W+|Xa?ij{7K
zbOjLcs*CO@L|s%p;sX&blXPLZ%dXb-blk$Xfx8TvR}pXtFw@a;N<&VGZFoDdC#Ur2
zIZ<~L<Q(gAbA(;Z9Ux`~vMaW+tYy8zI6kegLY$({%uy~>@bxBuCaDd}+rovS##@bN
zL(*9PKY`9fvUCJqua~J{_Q42xuLLS1m<h&VcM``{&<!Q`R!Nxms$g>KW=a^1>+gbc
zh2+Ujin;%o4Pvh?q1AWpCX_oX3bp8Lq?DEr>UwgA#fcjR&k&Z=t9K<ur%MGf3?{Zs
zj^pnsqVS@Hv|PL5@ArXe$V!g4>HQEeCmHH#I4=2s0oq^8#V_cC4xaq*zbHcYMW~QG
z#HUNcbW=^2hqnkb+q%%!;R645A#@i%uNMnF3`yQ7TJ*$$B0Ly4PV+jsXvgM8SbNil
zg}CY1Kp>Ml`w;*oN0T50S^QB*YA>T5F5@)oJt)Y{so_kApocH?n8Q&iuK9G*jay1c
znX4bYFNnhNKvgA{db~&nBNy(;by_-ApiA5?8R;@lKvQDw<1w<PXU7>razD;?)Dort
zR<KK@`+wcNRAP}Q3y{?7)6o%T5q-)@QdAC9BQW7DA&$%1F<eV7V^@3{5Js@u;vz1d
z6Y0j#i48k}(fGzQ1%`xrkOKF_k<Ws;0y?h6=c4JHve)j46`q4~XOf^*!p1EH<+0V*
zWUs}_bc!%dD4cH8p{@{}TPbZZUid431{i0ofmq;q|BX7m-9P<Z0DSK0<H@H<FBFkw
zDuEY$0WL(Pfkr9@1xvpO<XNNx*<cHso*}$B#y-Mb8_ZM&<1a5k)AJiBSZwk#-!RA`
zf|x1NL786~@n6yIYgG&w(26<t6%gmE>ps1*vd-U}2Ii_f)ECuPix6G7R@KeQs30ek
zwt9VWq7d?B&mh&>dqh%KmJDtpMVL{T7Yy<G+-m^N6HmR^iZtU3K}1m9WAWD^s8AP9
zEtY?SZ)c@TXlAZmg=l7BJ{pWGL^*MG?kGsC{U(4X5e^_ckqqNnai~xygS0^siC+s*
zj&-=d=UXP&zE$FA5oTI#jNN7Z@6awTGaMQ~gYk1g&O}#yjXK6!I07zF2^;cCpHOGr
z)Js5>SmJGBrxkl*to07x{uh&Q9ho*0;u^`23jY;l5M-dbetz}cB75{glHx8Kh!fuf
z^@zbIpA?B0-4_XTYSSw{^!>{O<l4z5aIX(cw{U8w$>m%BfX)-#KxHJ&ohu2`{MQu?
zeZn@aw6Xe!MOZzrBiqQR3o~mNqZs+8Jy1e3|NGiWG@kuu5qE$|lXmKecZB7R(TC>=
zdeWqifHbX7m=s&<o0iwHpb<4ulq5Zf)X2dmO`H--eq5wNrBLk~i0cJ;C=Cu#AO-{0
z3s^9J0`9_U+|#_UErx}XXBFkEL-d^oA?Jjoc&E|7<)`43g&SHbeUB(tPp5h&<nGU$
zM%$W`;@<pS$Nqhn&%qfetgi_2qMk`lkMWpD&!c)#4S7Ub0$2M2*logdq|z<K)?Ao*
z)j*`7*98;g4j|`QIQ@pFt1FyAoK)h=UiQly33MlDn}!kX=$Fv&GI=n3MFiDH2_-(!
zw{w=eF1(nb=JAySck6cO`hP*V#PAONEZtG@^3B)4W=KaZjqrtpkj0ShLKYL@LW$!v
zJ}g~AAgo~xklIF7q$`N9Bbjl1^WQ*}j9dLUM&HhXi#u&q?<^5^`|Mxh6Cvcu;(+ei
ze;}MEr>#CDMI2LJ6=+5b>tWla{|n7<LH@|vNY6P+fMX%n?&wK2c&7+-ECzmmJe^}v
za!frignzA!rQ@Acm+>Gf(4CEii5la6qdVRgY6c<(PXe2?g)EaCO#U5~{U71n68NXV
z#~ypHrZD)?i}3I!ebXV+i+S<5DD$$H-nh><wi%xVbgjoxa}EHPrGp((Mb9bHr{)B8
zz4}Mt-zlP9c@D|i%EGco1-c+`NlP=%d&-FH2)g5`GU8B^RUWx(HGVV~5#6AbAx08c
zD|JB>Q%65e&ew>f4jqj?mWj_qQM+y$Z_&B=9xv~}AiEHz_4t)gmlvsWrkw*0RMTv_
z`8-9C_CDJ4e$h;2xcmAr|0S7%<TOA1CV%DHaw*8N#OimQKqm>KbPA8wyg(EhVWiwN
zo-C3O<1Yy^LOv0cCyi{sz$UX5Z+{EieRfc5D;D_kIWj#5Z4fEB#d3nEb_b0*S>NMb
zCOzn>&FU#RM77=%r_Nu5z{eRDKy)_Dw1kc~oSaN(d|fWkRY6Ho@07-qg4~wUJx(S@
ze-)OQDns6n%e0ch0h2eT0}B*^xR&Q-Ytf9fE(yOs7U3?9jsW6T3H+-O-v;(lgcO3a
z#L5DQgqO3vm?DB|53}M9>KhKDealTi{H~1HsC03~_HYU4J-^Xdb-^NO^cv9h*j|7+
zU2CS3%!5K)Ei6EKVzGswU3E@2eQ3fp<L4#D0{~{nvBI2E1G5QiBQF%<vDL{@@mC8M
z39*xHjOw6wt1xF)uVL)}sAyN6D$$3Q{!S5Td}kcX2qPz%Kj`r`_9#NgNs9&A3Xh3|
zUv2C1u+y_5;8!R5{Y!n1Z#TDxq?)@!pzn>+4Pmm*)8SzDf3Tb9g~(=o%nMD5<8tCV
zk->hn5^+Tsl=x98b4q6)GQx+pK=)D^Q_zt$nOrV3*STN2@IMNm#FM5@iTm|!?n@CR
zIlNk65is}9Dw{0Lf<F{QjkWWQturA(Pb|DB|F~HYHqyUh2O%UsapL57N8iNbKDZw9
zMG;i}jzN8jkX$q#3v9*d0?aGKLI~>gd_Zb~?Y_@HxlnXRo@<H61m(F1NgY<$V)suM
z1I=yFKuwf+pC_0Ex|DjVoRje7i-Wl0yba^1*9fDuD3Mp<pCVi~rg>WH8%qMX1nLKv
z{=|WToGhn0iW;+oxKH*VCp=FiDY9Bs<Bc+OCs%&R+&?UlK{UuH{k%&NA?J!3<UnjH
zz*%%QkR)N?k&)bCXyN08xGvV30K2D5sD~|KBYwVgk#c%TF=h%d`;wS58C1r~%M_??
z5A57HhJ`u@)bm*!$zkbbi;%5)1B-Rt`5QI+y^L!7?5m)zRT4r*mjD(RPH-J#@j|BB
znbt>RvE}%W{xcaTVJqv~5%u~z4-p7(p_4ma9+#teC+8?*ECVO#+YFOLiOzez5Upl?
zQjL2~6itGtxi7xGJOp{QwM~ij^-a^|7oUz1$pY0g5=XDVrxK9-DCS-f%0=K6pE@hA
zCe?X`x_6-Jj72?<*9&q<+HsM^ytastaYI@^ZDJL<<@E={Tl&*5J!WK@<%o_o5Lbz&
zoDKX~t@8uEDUFUG)i_2JrRG{oB8xRwf^<VAok8`fgN1qSXfjOV3DIzX`DS{2V`T_)
zm@O?zTw?-U>MBk+SYtlRLDH-$@!tlk6hU+mNV9w_#Hr|^6;!cTErM{bC85%aC6eTt
zyjmc{jMRYgRaY#!S`m|zCGX8epAeV7%2VTUqG*q?%1~!qCX(vtKoLu<ZUQ@>dTcMk
zeUWENeP<TCmLPheOv4&eCkS+<Y0rK-&D%oUo=6>7xiN|OVHt=w0I2qFgt*!$oZ}(w
z&jMUj?%x}6muT7wEm<TN`>$b&Rvqb$iYVv6SyDPj@vSw1oE{eqtl$Zt-I*<7G5yK+
zM9vA@>Dx84GQ9Qz=LmEq<ob>BJt@S!K@LH1HpCmYYk`^TI+r78P#;OCb7qswvb^@%
zMO1E%(duHlT1S}o6U@ytncPIT;%9<gk)(qH_bwHb`Dwp;yR1_L(%}Xl@$pMRRA>sX
z;9RC}GmZ(GM{e_f99m+R2}NSaF?86O&<5f<sAV|IONZ-8q2BGud3ZbagXxy%l!@RP
zA~1=A^>rd$RTRrPz}{a1dOf8Q+fF8g>zTT$T;EXMMWz7L(R7m@2Z(T`dAg=0-)2_`
za-$64+DtCR@VZ=v!%AAFCWc&qu2CNSifa`n=U1&s(LG(t-%>_pV=UF8xV40ETY)FD
zTHG!OE_S1}_2g8(%|#rxjl|zYtIM>t#a`3+CRgj%atyjrfF?p%8c&Gu%!C>ZyAcfo
zDYHymmp@=$XN^S!HQ!9qg<CvveY8mWHycu}B#sf4vu9R}cT2h=50MM5$N#Rl#`|Dw
zDj@e*+L0i43USMFsi#YGxq*L3s1s)8oDm1EUnC^;ou23xKwUcfMjG+T2F1Tz=F?X5
zZdd?pi`ZsN_)!7KzOp9{7LZph$*Fb5j^hQobSyKPT}~9@M#m47R0rd-G7xgHE?Bw9
zMnzib4kl^xek70rCa2kRZfv4kqQEOfI8D^SYb^5dl%PCBSy$PWaZ4Q^2&FidWo!Ge
z#3mpvH5)4Wu9Pt^-_*fqp=37By={pgSkWr+rI0+@U`Wil8T9ygO1fy86V})qRO6FZ
zEjHMqNEM4uJ!`S65Sm^~IIa#vIYpM0YIKXvb2RhdRU%zdy<LZ>hH5?`aJ==C0RMn3
zji3RIuH7)ls(vAksJS|MZjOTlI`O{hSez!3ZpCG~*W!!I|1s{2w?%puOc#qXJX&Nc
zV=_WdTF-Fq1Yxuh?i|MA+O3O7ohW}M=LmNR%T2{bT6>Fk1-ah59nup=Yy+(U!GQ_U
z*tSsSYu!G{wHr#Pli^kbL%Di%DWGQnRN-b|A(S)#`UIOp>;mdc0J9-Hq@`=f%M^xI
zLdh_uZpSw@Kmt#*i{Ot*D6lWReUT8R4d}$hib9;DE-7k*a^#Qa55ca_NO!#}-WE-Y
zy}WxJ?}{+rI~WZw?co17q159e5v~-*;Oty^A8AJ*^BvtQ+6K|s(!!jr{J?XxNtZ^e
z3C`K-renJ@YF6Q9-z#jK`*b2k=s_ja&RkyMdhu34l(Ef9E{p60=@v-4c;(iLlZ56L
z=pL$Ssk})L73I1WYw%qlJW(YFU2%K~Fex&FEWT@*uBB~i*}aA49?i8P$o@h!15cb7
z6@9xe^ZW^BIXW%P6qp!HPyM?Hb4>2L25QKD-pc`)=x~C#o0H>J@wivtc{b(xq%Xd?
zI{>-Ml>x^#`o`u$ay<ua3%$f1K=6m@U4ox(k-3KrC1(U%?+M}-l5z+>EK*WC3-;p5
z8GT4r&B^fh*TUT=nhcxUM7jp8p?2O<V$HZ;nCsn%qz$JU^uG`G0%V}j?YN*?e;=rE
z#vwDWM!OLH@4`(>9m^LeHzP7iz3w4_{3k-4Fl!SYt*h{B<9!QoB)Kop)+m84R;2^c
zhzw#jRY2E>odLEXdUAwImB)qCA`(u<%OW(oDkE6X_<It%%is7h-{$N99<Mn-#m9oE
zLptP-NPd05PmC}XBeJT33<_~Mh8ZP@x^eouADG**#q&eF94BC0i4t`~)PsEnUUz@~
z$Q?+|pOR<Y^#q!A_}*?uF0!8xH_1wzYZ@Y5RIUc_dWb_GL3vbEB`H8}3KM9dW{3_)
zGJZ>JUgQ99crXPoKUICZ0%`xJUVoY}58g37CUU@m&^dMKp&-#=%$+F|wql5a*y^Vc
z6cL9iTF5#@@*heGmH3~CoJBWc5eXQh;s8ioCd8H^oc=J+!sWSgJE7(Sesi>6+ewHS
z+mFHxdfPdbVRpxHLMFQOM%9%f-09QKw$6*bbg&l%kF$<Rl1Xiw_9A3DHRDJ`YvuNW
zsR8bvC&Z6MxPS=ddYL;Fvs0*xgOHSs!kGeGV)j<Dq&>Hzh=_J0rbtQju~rS#3_(H6
zkGHQa%!#qY!NSW+KodE776RN^=t9hmYWNL00iC4|$GDuntgA>!jn3P@bpZDQ%r3Dz
z>*_`!408hE)DXEf{=!8#9ds0VsVgoJ;!c3vfVD!jv-jiM;?jxJ0U6ZngS&g9s|U*5
zr7zJ9Tk(H_97_*eVCO>Bo)Jj@>qdrjWP@+_ZHxfo6p?v3;h~Zad#H@t1-k)cQ!fKQ
zT64Z$(5%4C@r7G+23E4YSVjLhyo@kAbd^<&RfW2e-F>)gsKr`B%=JNT6)RB_;3}l!
zHnFyE0=Y=(BA0{c7@6-Z*jY%f<(KDU524Nil}m{ycI*T3;Kx**=BD%vGW3c2`2*gi
z{d=q-%6#ZW>wSHZPQB7E{VUEDwOHvOP&Z|J-8=eput>Ob?H{VOGtDK0#+6&T=vOqg
zG`EIvnh1A3mL%I<ydZ46E{H&IwnF1BI#A@6PEAr=()4hWaFRnJiWQ;=bB9$|FI2Hk
z@etl%k)lN)DdrD`KwVA>DPNzHgt&h&5KsW1+r%LPT^wvoSas7mcbO4TdQIDU6XPs>
z=a~RGLoJ>Y;NFUNgTa_<1~lx<ujI}x(&5qmWIwWs0P3LUlj3?2q+I6>zIXLatLSE_
z7E?x{Twv}+FpQ4Q65;VTIU{*qh#4x8p<MLEx?^DOMv|6D5o-rQx$Zc)=P-Aw5X#IY
zQPL&c>F3b@e@D{|`+I`jfl&ELy7pc@RyvqJ@`R9Ncxs#k=0t|t-SZiT>y%+rzf2R@
zT8EH;i^Rg9mHqc6!ebv?{>39g%sF&=xM7_CPylt?v^dXo7!-WQag^@d#zex_Vd)9a
zeYgphuz^@XM4lUl&{$#4Sw)bW8Q};6BR%GoP_wgD(ZjT~?&=EZ!NFx>Y*VHsg<GAa
zzWIwH#7f*$A0(1;RP<ip=^|VOW^YcPAC|z7wlA^Kk&dm~QpgKLnz8lt^q5A&>PG>(
z0zE^B4XN!91(|<ccFBCAknz62iCgRB`NCXH3Gp%Rl3RvXgsOyy*WyP<L%4)=7f!rG
z0!W!fBUvI3#<dw?>mf4-M#=w#xIiOZGYp_@vC%O=x!|n!(!%6b?Gj`Uko8xd6G)ez
zIAOiGVkX}%Ae!wOajNfH$AY>e$$1jqUXm*91%e%zX)kRx2z<9tmj%THge%q9=ePpG
zI0!c(197Gx7mw{7My?2sVx!{$H4<^{8Rwi(MC#{~Smn4v5dZUX4lc$;Wd0Iep08q;
z6TvikTZ~S9n~{2cq+-AY!+;NYHm<l7W_*M&pfm@Pb({q383cFPjo4T;$@X;P8%y7A
zX!@I~a-R@-WWJJlPNd6)RL&Mx>&*gF3pX()*-eD2&McA+s5qMu=wy-ENGaogLKs|j
zCi~r!p<Ej&QH;jP0!Xqx%HHDM@?REYX7<+vxM_yQ+G{*gClU1qzXWyB@CSs^Fg(3e
zm`j5sjP+kj^}hw?($K%S^?mh}BAB*yEUj!O{vi-C`p^(YCVZM@xZSA6-6hTYa+6}!
z(~9^E@Imj!Hx}sf^w%3S3aUk02y@9>%xx`;Fb@>!wqsLBZ>YqDf?xo9mS(&zf&ya%
z*TJ<OLt@v{L1-1+gz7A`>=}?W;dIV1H|m?1$(r@%{6`zteNW~;ctl||hR!6EYt$nN
z@hs7)J*1U$iYOOd_u5I4wdh#|mZYze$Zjocys>%uJ+0B(lmU?rbD*bf_7-NA>xD^>
zkrI*Te{5(D72-~SN*7zCEq@JQwlH_Gp+ZnB$i+tMWLoTSc0nUl7>dV4=k7_tQxE<~
znA1hOokvTLJO{w(O1}jgj+peDA|3R}hT>PEVIo>{eQ~LX+>i(}al*wLJ?DaXY_Vra
z{slF*7CsLcHcE%J65EM#X)!EqkCy0^Dw>Bs7vW+t72@jm4FS%E(VkvbiEo_`M#FL*
zjOFy51)VtX#q>3;j;**&f4Eceq$`*AM8^%u1qI5A!w^aP#o5qtYc~2%LfwwcP8vJ#
z<R?nN<>7i;K4Ax9e_>`FCmyD)O59pPND*c86AuV-YwLJ`lg1q{1aW222t^8URSC!l
zxE0%9L<oCfkS(V~nXPWiJZ9=Iv$2;~#^ZR=PTqe0;`}m1tHC|*N*5ENU=>ce>RAn{
zQ-640nRa@~PF%;h5@$n?u8>wM%pVi(tXj=cNn55Xq!F|!q?B`5nVFvAODssDp(Dzm
zQU$>l`A8vjk6zUEk)Mk4V&BfdeO9Ea$-2~xqb`A_w^U{z5<5qP|HK3s|AeC5ICyo(
zl{Q0a>ZM?$hmNrPf^BwL5rmzf?r70c6y}B=-@<VhEYRJuKkg9WLLJ1^GwpInH=Lis
zI$1!jKfMA^Fi!|^_e;Ffai|tg3nfv`z3awXWu#6`aIM(t3jWTOl@{B9CCx=}R}6@D
zzO25?`Hu)7JzA$0Yg`H8La46Tg={B;j@gfO4nZ`->?FuZRPifUiRG^<Vqr+dPBhLE
zLVIu{#^uQ!qTIRr<$RLe{as}gy*SGVKj3Ntnem9SaL~z4@lauT8sr56bY-wANT5?g
z9yuV%P=Z`+wsZ1Pwc0f<K++Y7DWVwd>ra^y_vzbdvIUc)Ws37=4xTPrV%qO=<TTay
zh|1NaI-}U=IJGWB-V6VpZx;qpqsDUVDTo$e6c5XU{4UWh99n*C592u@PEjjbB|e%h
z71V?z|Cs+ehoS#rFb3CG7T{#0IL(MTPl)qIKQiqUo7V%ma_LQDS<J=g4S+O;F944h
zl`~KFu%8KV!ss+`ERK#p6bRCkZZ;nm332D;CD_>N{~(<WegrZ0*Op)}fw_Pfny{Ii
z>yIGg1CM)+B}6+!Z`#qtR3Yx-GzRy093iT4zff0<A(=D_;zl5PDJPw3?D%I0r`tY+
z6NbG-J3O1QWKK6G$hE|bh1=(cOMtBoqL%q?a*8}9hR@tgXx9=aig0o~FVDScS6nT~
z$+6O*RP{Rnxd?1PV!ppXyZ1`;m%Jw4a&v(VR%Y~9Wb|`uiLnHZa<MrO-?)Wv)J@8_
zvAzf=%nJaO*f7&9Dx3y-V@Px{OgIxWOpX#p_nwLra;Cn;DD|1IZ-sENsRzsWg0}&<
z1!RPZ_;_I<gx4KIye}d*7*1(<<per+Cn-%+_0gCh&?T@hS*oy&Ftd!yvpT*h;%Pzj
zy6%deu-fCUB2v;BkQbHbgt;)Xfna2Pni0;xOg7T|_)JJ{+J+u<p?BO3G``f}xt!aQ
zBpP@_I5mOnj=k=IaLLg#CV<A&X7>WpizV7+pza_NwlzB&@hg3+4LUmTGAF`pL+PE#
z`egy;OHZY{8SC8#?UJL+E1~|@C8FIhOaFZL6UHTIH8{4VsdL>D#NI4jdgCQQpc%?5
zOA<^~<Gd2%n8Q8;Cf`{`ZVYt}jxe|J601Nnh_hKwjV;t%4-}>2l6V+%g@uJU5dNmx
zC1PGdkopWa?OW;F1&}M>Budyb|KHX&IXcSdgXlABKvc^pvVLO$Z^ei(vrUhiq6JXK
z8$iKY@t!FmSc36%E2chJ6i#OZ^@BAY0;2CA{i5DmipqmejEtKFxH_`$*E3`<J`CiE
zL`t&=aJBd?_y{PKoH$|K*i7HFFpEng4#<Cu(NI-==i2KzcNOziVQz3*sW%>Y6xum*
zE07!zJp33CWvo^7w)EqoNS!Ob-nis(hn5o$`lsAmZuErz$Buq5Iz%}2Ug?z`E;^4&
z#1$~$;}StxS`TwB`Xphf%A^UC;s^S6Wi^T=&!o$=3C-;1a6B3l1vpvsoV0g1NQkS!
z18;2ndG}5b1xKZ_8ZU{UP7D}sSl$sumOQu`-+G#FGfDnLxK+Wwj4-0I6Im}-dIrK}
z;>v>enV7a)v4Kzz0?t<qYh93TA~2T%S8`Yw#118--LqEk(*)&a;c|wvBif8*a+1}i
zyWM!DjEjJtcLwmj@hox8;7S8$bn|Cg?xtF3D|LxBtI$@0Rowz|^`!JJE2pD{QT@gc
z@4My%key(KG!*U672)B3Z9oFcZeioyRNiN95<n?<5i)hEFK3H{g0V!+vB&d-@$MX{
zM+c3qN3@eg<^KAM=;QulW{Ec2I#Cpt?eiIeoF&XaAlCmv5e5|{35w!!K~!UUUoF1S
zxB0?GtQJSU2%YD5Hg1T)m~tNyNIp8i>opm+bBn$NL`%axo*NM1Zk4w4)wo^=S&Y<h
zkF0MesfQyfSltao^JPGHj$U@%NiPy_#6q1A;f6T!99bf~N7l#S{91^QgPtyYIlRy-
zz^*uZF2ZdkAor=JHZj%M@^1x{e3k)?zFI&$7BQl|#&LpJ6Lhd(v-Yq^mx(bT8}bi?
zjQ6GXdbGV(q|w{a%UfT~m?;SM$OJGh5@80kqkf$hvUpLTCN$lH$H!$5J?oj|J+boZ
z1fd`v&vBAS(!|YFEvCK!F+Sk45FMF^P&0u|8E2P0N`OAh+Yti-$QF?$YC*S(a@}ZY
z?YW;VBeeI8QnJ722si;69*srcB!WkUrZW5yF?io9GQ}32+ljpexNPasPZk$a{;kkF
zpb!vsOGbIAFsj6g+81|;aD(AJk0Wm-?#chlaX_5(cmAj7{J|mR_o9fZ=Mv*X5su1T
zR2o_z334|X_tvKy<;L4ZNUr~qf#ys>Fd<#ZBB&gU_k}pF=5{2;sn0L_4zO#0)&ok9
z2u_w4=B<auoBZdcUCPD6T_V3ug^)`SjgvjfRAbV+01T2bESGBf&f`!vubraZXE-sX
ziE=b<FG1!+cA}Dl+);usHLc}caZd@RSvI{|I~%A|KN3}TKrg*VMs%`t4KO3tejn21
zX8fVCH5fAmjW-+uQmqjSe*i{Xq?^CENK|gQ9(Am0EdF5u)tY0Ya)vlnkUJauR$iEo
zO@+XKiIdu5roP>E^@=18um$94!|HCd%yR!MLSd)PDu25GXRUV~^px9u9|4(_q%<0v
zd<^aKr_+qef4wj_8yW~K5reP^X8uRGlbcR+vLeP@p8&ZWoV<AQwSutXn*!;5>FKtM
zMK}XIPx~7255nA8@mlb<C?~6DXtgz(`%@ruL)kNIY9+$4Tw93vj4<Yj6yLDG`Md!7
znFSir5I{DlRAON?-xpBMhDQSA{wr1#<gzhU^OQLohn0jOZEf$0RrTo{<iL9%CT9Tm
z1(LLJ^btV+=#maEdw7v#SWU0_z9q`6Zbw5VKKRmsxipq3$XZ`P!X+%ho3VokVzQ($
zJMSyXTt%opiUOMi83J8cHo}Z^qWx8v`9y2V>F);rDi9s4Xz`7|2JwWW1H9i#HEM3K
z^~2Dq5?d>vGe~<>b%hziDA#~(aYQ*g^-;RqLqc3q66gIV$@{<k?;;qQC&?~Z$aoXi
zP!1l9-37s&$=R3Ap=A(h-^lXrAwlCa4d%K?O_ls3f?+zFiy;h)-V^C!r@f{`PX7?5
zhTv|e8EgNi2qAVRosSv9+&De#zv8t2LT64U%a6E5i2251(Q4r!gfn{Wf1tS+agfzm
z>?$Goeq-v1GX>=al-Do{){1`#%-ty&WPM!%%vk8sFEhus<I(V$=I1`*YJsj_)?365
zLfCjIYMhEk<10aMh2z@*Q|7PdERy1ZTh>vvW@Row^|FRkfQLjmVQww8FpS2L!kjQR
zf@xyBRgl>|9d>U!Hzc)S1WCD~Bc{w#BqP7AtQ({8nlSpd{G7%j--Ix0laIA#EGvX^
zNUDv)f_b6L7f$S4c=XH$aKOPJL4euG!f4CYxJ;nir-uGq7t&_B)t`jA&!DL}*iC;|
z&EFTyWZE@^EcIJ_k2ke+5M$2yLGpx`rZGebnhTB)PBjou)Z-`-Br&ONYTT&r+<W9%
zb>9V`%{H7+qxgb_lOS`qmnAug!k-uBXoKVs?caum52z1vxp}@QmkY74&cqj$2p4)<
zIsQ*bZg{q1I^RE-6OmxOKk5q-z|7<xlTFSo0&*)OeofNET^9m#IjSf+#2zJ0QM%$W
z(Pkp6AS*0$4b$KJ3j@1=1GVWKhoUAVR|tVgoFLkn>G3z@M$NceSRn(KvJxwp+T_zR
zF#3*2#pr(tbEE276eos7$UMuiu-S>hJl`Q*vw{?p1Zk()Au^(K{Y}t#GvK~NP4kv8
z*e*FiA20hr<?hE-hTiWffD!aJxFQme>&;an%ifp}7gHu%c{BXZBA_m2Z3cn?R0wOa
z)S`gqWjE{8_@OSrE|*@O(7G5aD5&$n_@arUlg)%UA7p*<dan_;cNq@#51eVmbRn(-
z&XTF()F=QwmC}r>1<;k7*z3k~BAgK`5pE^pw(o;@1Qgm^5;i<9Ja?x)J;?U50P&8E
zc{3he42o2-CByqhCUoZ=>n&a+NU5_gP*$TYgqug|3cBOWpqWAc98FJrq=4hyrkBS)
zlL{Z3`}BtcBVwx$j?ps{Hr{RIKPGjsKPoRq>?msX-AnN8D&rzh`pZ2+T(u4w3VBQk
z^dq|OiGK;u4h%8)BL9!8^MI19sQy2Kh>}&VNS3gYac6dC6G#>j5CK7gfb>lF<mn!!
zduFzm97GmD2_i{EB`8T`0Ragr0)nE5NLZ4JWCT$V1%coH^Q~{qdwc$K&Ybh^x9iob
zx^*kvy7$(aMQTPlUf*AUi=k;%dhvKym|p1Yi1)3<GhG8=O$wqlMZ-A78R1$bQ7{f|
zc0FdU4Uwg(d(g(z&Dc|@KcH)6#?S!A4Z>Ut7E|gP>#pOB5UfhvB!VuX1uT6`wla_8
z-&xxX$MYimx`KEjC$9)G6Ie2`c3W{>0Qcx#`RA>_p0i}K<#0RQZ?U}~=fV-C%pMy;
z+*Y;udB2qw`YU{*_`&lMq3zRB@;my$hGYmf(*R7xXM+L!GX0Xv*v5RW9=})uA<8<3
zEbFXaq+$D{x4ybTf#?f#rf&#=!+jNEl$<NV98TiR5djO&2Zb`TO(wU^`0$2)L&xrM
zs0j0c7(=zFigs(rWsNw|t+;a|u&GhMr#67zHbE{u@j*DMiMy4EcAv2YQ3K;`QCX@v
z(Z&*ug|t|B^|T^?20d9+qVN@lI5qIscv{b7ESp0@4Wq?g`jbXznPl+AIf)RLd4ez$
zF?|zgmruHKc~k#aQ08u9fv$5{cGCip?GIz{G$C-VQsXN}ipT>)>DAJXzY4eDaBSW&
zFIL>lZ)q-$%|&?4z-?`@r^q~tq?zP=0mUrc$1y_MFJ3Cxg9Gt(xZz1#<9h_U78(O|
z{IuET1=Kx|qSC|}g4{j~A)a7rTNKH(?qkayhYNE3vFFtEIVl7t&zdtmp4PLACC6nn
z94WfndRqeL>4$!f%88EhErG5R{fvxW72tAOD4Vh4RzUt5PV)GB$Dj~+OV|&N3qA#b
zXeh`aR@b6Pj2Re_-_<$q6EZdWGfy-2(}@>OcRJcV-E@x;w-6VbQ^*EJ<FzUy)Sex{
zE;hCl1e%Z+^cMuW*l-E<?k&J&WJ9ZM)15-H7_p&?=!B-sDBLid65*EZZY+p1W|VIe
z{fAlL<`N;GgwBVa6vF#h4KlgT^!prCO>*uqD$-w17e3(+B_#6|lU5eRT>!2feHTxG
zo^gr)(SqIJcH?DKeNdR+W%!XD_v`}Dg1Q;iIRLcwjE))c<GDPWKU$P{?{ZBEqQk*h
zQQNEpX5ms$lx_~fZ&gB<9L^4LWOmv-5c7=@g?+7PniriQd2GHl)YKGJ1nb&UD7<Ou
z9;QP@kO8BfJeK0*Of&G}$WV<3a^eMw+q2O&ev<+cFjRyKgZ-VtSXG5k(IK?2ujm=J
zOgSUOW}o)E%td3dK!oMCO>A;j6!Em6sj?<LtG54}ZtH9$c&ri$PniLq;4_cU6Od)e
znvR`?<eca?^m2_t2bQSf7;}fIVh&B@?8qN&6Glpl1+W#L7>T#_2rpX<Q8;;S2W3uk
zVnf#{X;vDQI72k4wZ9mNvqZV|nlh}D!Qm6aTxo*ya4=MjZ|?v!C1;p@&}f=qe<j#&
zbMweRO=Cvv2$lx~BGcMPH-WAyLm($S-B@5bDGiKLhZ6?9S14APrIXnQn|9%rTRhF_
zCH(@-NG5W1)9>sA;@o9a#vvZpH^+fHgQ}wA(nfqol-W{6M;I50maxy55!-%-XLZ66
zmWw-OLZ<NXxn20h73>`+ke1B4b`$0bP7)N5!3a~Z1ZGYsmRpsr+QCX_K1v6o^?vKK
zKzVA%fJUPzoRERa75Zh0ZM~c;X)*q+Kv$KJTndVO$gV}6yozljL3#z5LBnlLjjxw!
zdeu4DX0G%((wMpI2a^9za<BQQaCju$Cx)$}Twn^TJHlf#L9h+kz?hmyC8GT?jS9^;
zNPu~ty#IVWF)q|2;qdr^I6|bkL#Dd(9^3lez;b`k?u>h{gqQ^>9$r#8{Z}Au!yblP
zf_I0QDj>Q|IC{GvejjdQjq|98+@o-{%~RcbLUX=~?ZaWfa2&M<AT_UZ{J~WgqMR|)
zTCzCZ<?}!m`6)wttwp{s*ePW?K?}svdjh#8y$r1g<Z(jGrNLxgkpL&b#;}j=>s3mU
zB|-71v`u0l;R`?2FIf@>nzA4xXiWmm&_OJ3{-ZS1nk%;!#AaWBaFJx>tLWATmXHR>
z7PtF6EXcg6Yo}C8s-9;`G}&4;Sz+U(u~(6!I-+EwV{d<vX_bwR2<L?znD&7hd48}D
zs4IznMvk=0d=Z3B#(8bb&@&7tzAeqRhbWgB{gG@r;}k);nYcyPE8ffZ1N54Z*9LTN
zq8G{QeyPZSB_BcW8*y+6^>8>R)1aCl3NQyB(_uYZ2wS?sL3+b&_6PH*Fg8wrna_x(
z3aAIQ&SWavSBOi+(gs#4M9F~yb1#^zAp!}YatzO{`7xezfg*kr!%<vQ=7I)F@&uQ!
z6G$UgWa99z10dYMeFQ~1_rT%{lSDzQ#_9(FP?J7Rkh}EE*BXTQ9Eh)pGTW=V$f9Y)
zHC2!eLI94$@Rv#DChOLPx^>)8N8)UuZU&uAv)Yb#1Ub9z@e~nc+k-(|TREcFVow3;
zlSDF!?;HX_!8%zrT%qSIy0+b>4ek->*O}j#Umh30>*H83#Zret@Hqre#=lIIANdLh
zoWLoXn#Zt6Dx2&UqbbV0vVz$nhVDsgagksOF-*_CQqL9{Mb4CE8cv3Exu59=*R0mV
zj*&Bf65;EL+tR3z#x+30l851+2{e~DQqgT0@vb0BWz(as9w<JKgHUXz=iD1PnAJ47
zNRWA-D%n%hB5iSrqG04`lVPB`B$_mU>Im)li_*G9(MHltwDKlZFtq}9i?Nvq^M%-~
zEmq>mL}-p+^Io^GV4NnvJV5i-kKy)M4MhB9E*Xzqd!bx^9kjt*hUzE#09_^6P1u7<
ze5@ZReH`gtON2SD^R|XhpP4?l)(?EXp3GpGJ3lKx!qPb_mLA|)QaqmlB=!>Jo`_a}
z7-1#Nbrg+wR5Tw<3fzwmI-7A8T6^f(Rb<yfAAUsu`Eo%B+x``Ypq#HFtH&k7(5|3X
z%${er9|3d)Ie<tWcc%z4qb6l$+8>LB_(Lpxne(@*15lZnctPH)XE%`6^}X__>1-4)
z;9k9;65|4#NEZdC4Zx+sTs}sMM(jQcZC?6dO?q^#FwfQxep%5D@zZN9iKUB<6_Yll
zbHS<L1eRFm=}lP2?R%r~Rwi`EP3(O*zfgmY&S`O)p3MWrTxQu2kCc$2SIM(w=P^<_
zcXVf5UxEJ>h}pub2fyvrn$C3Q?CJ3fJ)7IfoiW|(a($}+HQBOkMHlmoKnjVid?i*L
zcNU$~XUEBUcK(=X67y^238%zVWOnSFiH6*G-x6V7c8{@W-g*++UzeYRbgQ+vT_ANP
zs5O39Je#MZY_Z0et7DBeSZ;U)Ng0bHgitgLAX{<o0tiN@0lXm5`0srLq(9lhjw;R-
z?Y9T&Qdh-IM}i=Z=;m3x-Y<%;S0}KL(X-#3)avC8(Jl>4;RYUBeeI*bTm{rJS~vk7
zEYJz1+?R_-9Dg(zg+yUVG&>P|rE~hsxaU}&J+Tb+FNpg^k`AZ*N_^`$2<J{nR2F!;
zXzn6`yrB8E(i2A<4@q9A+2Vy0c&2tb`AlAn@xc>~Xu?Nr#PVemA+A+(ymugW679lr
zFe=yR{&?oAU>*>M=!9}oV;7iEOCWkq%)y)Syb_w7gsH6|T`yz=PUx0kM%R~@NK<Qz
zX-l2NTe;ndzkpd23aL*BbnaX@u02S6Qjm**nv(9u89*~eh5DP7?lBo0J}m^EVaU{o
zBfbXV`r$*`ihqlCzBsZn-H*oSP6mQyZN|pK^~}2iRo?xF2&WY%=EJV{Ed;W7<1Ii4
z@^uC9LS4Jw7e$zP3<G?hSb|%g0+y#h_EG7?fJ%1>E-FJyrcvW!p{|U=xbwj+zFuUh
zbFXyG#;7nCf<sWuv2GXON^;)?SFoK5O`$LW<R&Z;W-o5%9IT4xM+)Q(R!9TV#@;O2
zV%N>3X`7q|=|XGRO~>X;z<Yk9fQeLta!Q9mJuxnrQgbZ`3ogRT#)J^pj;$vv|5Upp
zg_=Da|50V?cT9<~qGeS_L!T?mC5Fj5GiC~ZP#`5tOU<76$TuO)1ICV8BSHYJIG3w1
z-q15?367^tz)Gh>`n!~w;hvFkJ7H#-_WW^>XkH)X{PDYb&itk1bpuXJ6nm~RR|4@@
zR#wNJ36UiVy)!d;+$P8wFu!tadzS#WtM*E$bKVr<EU`i!jgGURT}JIbQ+VFCFlQsz
zl)jkzEdajTF$WD_OGK{0INrz4igr1$B~$p5IQMKIDm5_FA6MvEu2+Z0a^L1L^R9+*
za!fA)b_VF%z9zusm2)!Xy;umZk4?s!=kPdHVTEAAtw@m5>pB8Hy>FQ)txOml-W24|
zVsc2k{O8=_t$JmY(K1#!uRw+ds(`V!5KEHI%%x*y5c!8tMroGp8_tS<igLy%sNmBk
z=L1ZYkX0oM_6r5Uiq07`WAh7mHm`MopzdLfeS}%&2I_<9IJg%3m*6NJoYNrI5eEpQ
z5jcCP#FZjk@j=DZ#E-4`nT`DOB3uv#VS@U#;?4ZK_LXtacZ$~;7$p0*3n6#b-ovrf
zg%ECHBI2}SL(#72z!>4q<6;5uf#HgU{trdDUK7wMHv4nJ$PbG=;*ftA!j;67Q@ms!
zc$z?e1RX*vz9pKENN3$*0}+=AppJ53z?gwLE`Je-^%JsWW^S?G=sUulzkH(^H^~2H
zfi6|@V6QQdyeTa8Hg4i8f5*~^P{_1z<9qyGIOi9BS>*PFelbT`U*PW7AMXfq1*9H`
zp6^3jWVDLTCz>1l0F*HSI|`OHv8hN(Q0LO~pXk|*m_m6Iy(1?^zAC-lvBJe9FyFbW
zx`h#_0DpqRUMV0nsQyE!Ys#(>v15BQbV-pnH+)HNGQUK+8<UdF?+T#;>{v0NAug$C
z+KhWlxY7R7A_E3Zo%Nh^8Hkw2)co@w@@$SLqm^_3^Cy9>4@yIYjaltSATBTCxfVJ)
zq@63&=~$L<R%ygFNCDHdaoI<yU>r-bG4?h`iF5-n#AptPg(bw~fs0@@PR}pFz!7vF
zZ;3=si8V=;j=3C)`f+FVQ2amyKcT5$)3+#73OroF$BExuY&H@-98E|#re6W-Iyc+g
z;G2}hy9o4*r7bkU9h3HNM6gSX!*8-r`ja3UGhHC6$f0jv36Pn?+2pt^H>)x4suGnE
z9ZGn<eUVVJY?Sya%$)ZL@fT3wq8Xu!3zD+8JMOuf$6QuaAzW}vd;eRY#-dp>WBniV
z3@`feritT3@B^V1W9Mslwp6z&oS%PP^b{{@n6jrSb@5L?oj<!s8bmUeKyf>{jeOu*
zlDJLjkPPNjW1g_0f9b*^88XnlW7|di;tC=uZ0Y0cLbC8D>*!Yn<i&y1VChaVT-C2D
z(s8bbYJ9N(iSano&r;VzAVmm7y}ud9iKM1I3)*q=4LrN1gE~-t|4*U$U^hXD*l*Ng
zT!0f0+Ls`AqFre@0g#48NcBd*Ec7+botjY>0vptc2V;>aHOfRhmVY2t3L!3+zRAfj
zl;`xdA<As#j7Mydie{&q3$(__Ec+DN@t6{=7O^?vQK4B93E)OwON{+b2~X*iq@F(Y
zXQX!PFkaC6`Pvb;f=!7Ou@Pvns|2}NoFhwTF4;TqHqbOs(3?*b;UrvKMEqM^EPCU?
z5~ZE6aD`nXM$Ig~ie$&fZ!f+ng|BR6HvBn|8(x8*YH>vgunFMc@Zn#OVv6V34X~U1
zB?u*wr$l^01T5ct^Eh9RW(p&Kj8&4NWV1Vf-Lkk(pd1rThUpYLsa<aV6^IMLH8W@)
zK76N>(5j5&;?1}a=gkZ|b%Ao6T*5WNMq`JHiy4;9cahq{qopl+b#@!O3Uoz?8=*1y
zuVpg!BdkCbwe;A#i)1>vpfs-uLB>YWV7B6_UuQ~Y|8ch{XDU}dI_6>bfVdM7Xe=5c
z=>)yC@i<e@BvyD$92r=l{!<97WaB(D=HClpHn7O%K%`v)>^>PE?k#~{9LAdQKnduU
z%7IwvH{{`(SE|W21NnPYpqba?ZiDOqZP;aRjrV95yI2Wx+1VjHBa(^_v&-XJby1dB
zsVil~_t)P7xx}2Rcg~Bg?*nj_3W5kbtG@$r>(>W{8d3Xw@e125<a@3A0W8w#%u^#M
z-Y!C646<B_qeQs>R9Q%~MB~Z<VeqwYU_7?@1J5qTL_?#}NdojX{t2=A15Qbd_UP6#
zHELt9&=)WM(G)jDVCyjNK`65hZI0~rNb;~ivM$_ko)rL#(lQ2%dXXB<`^V!|J?D8y
z7cwx+?f)nLj3CAFdZtkQZJgGBJw<5PU+p1&F^`gB{~MwyC3ZR}C7%+-C$u@}dDy=U
z<7Cl_V?~)S%mT<i0{Q$SVCDsiR#p#(Jz69}KBaG|#(*$?5F-lAVnzF<NmO8X4wMOK
zfZlk!q}vTz@wmms3s`D!uh@$M%sjdYiXC-@j>ig6%WA1IIJXpN5n)Es@c6OEi_EY>
zV+p*S0ChOdG}-ExKhr8GOy3d!BXvp<)1H9vhewFejk^#2o;wNjH`t14i+z9)vr3oy
zrg`;HVdT?5M<3%N_-2ZO7FUTfLkI}Z&)1iLJa`7;s3-k0W=I3^?o%c(53R&ie>TB=
zJ@{Y#rN~HDm9<zPfbTH04GqQxqEIcSgC2$|HljYb`M!Q|d8Gtlb56i5VeTpEb`|Lx
z+eV!dD~&3QZp9ClNXedKlAR{aaf{GA<qRv=YWzh|ZZ?i3kRq<w-SV%1=H4*v!<(Nj
z5lvtSWlIR%V1|P(wna{D7IG{Y6|VWTlQ}SWtoS!bm$j+zpXy;B6-FyiCDyPatZJRm
zHJwa>K8s!IS$=TYsbN!hVZC1vMR2gMm?hFxnKE2#$83RSnI~jURhx05P>+Y=xy+iR
z&b?SLCD9RZEEb_Y$Y!}6n>^>A*!B*`4kBC@dK3oQm~1~^K#Uz|J@BLvL<YkRon|&f
zS)gPiioKTr54986Je(mix1y}e<X!h$VRjzlsFkfozJ8i%`>+@J$^7PUlT&bR_0V<d
z33tb?4B|l+FOiNaF~7`;9bSfDAi)A4zM<zVUkXE}?flAr2QvrZNxi8xa{N&s<s|Tr
z(tP}t;yu}<Y2_4`33J_gi745QI|NX5RCR3K{w~VwV%AM(TpxKAh>~=2AHiyR_S-sy
zBLJ*`+-C9)M=Q=CakkJxc#|>UuLQXc3QCj~ahLrA%yp2JnUB677f5-u!WoLqUxR{2
za>8h!$J$Y}1W>E<`!m7zu`gZ`K`{qdk8b);=YcM)zZR3C_+=c^l~eW1`|PEOhB-yZ
zxqGpQTo~euIJ(fZe*wCB^ex>?t7ET(nhWWOFYZz@a_yC=$yTpJNR}u+7YM%LpD^Sd
zh@XhC95wrxO8#w9UhxcsCyS(<9n+`9srh*(mt|h8XZJJq6KS%uO~%P@@`H=4DaJaz
zb4ySb@mjAF152PA1dX2dU2&@*H%R|DdPt1`bN>TIJr>Bq<5CfRN0#@Bd%yaBf&9Tf
zbh{icGDtot5N=qbtZ_<!<$6dmwb5J93{ff8Wa_*_Ld?L?0Rof7B_)8j2(5Xe0O!Hl
zv_IY!?JxIW9Ii9-Y8>%4APmDPHy+b-F-%}{LxEWkzarR`=+Exb-~V3$u}otet;Hfi
z=0zQA#-yKmU6{YvBs}GCio<@yJAfW-F-hkHC+;bc)Ca5a{&$PaxcQK^HABuG!eFet
zI^u2-2ne@i5QTf4_n=%;R)vk&MYPL;Cz8y_`(kflye@aRXo_&ba21lWgW+)B|0_VO
zWXHL_V+u4+bT&M7S4T=~?&I!+E!YCRJT-@;W0p@XwOitBjV?1&&wLzf?!MSfg!#!e
z%k)VW4(|i9crqR0qQG$<x_2U{>IbAf`Q%pPdXav06rHSA)FVp+`NOJ5Up?AFTsB$z
z4#%;}0KlGpRL)wFEEMfm<R8yy{jv}<EExt!P(Sp3zm@hR3`pCEHqXZEqYj@C+Y5AW
z92p;sXGCT$pr%J~8_O05hrLy{-uIM<oVFPY#y&urJkO7_j_gT$%Y{O7DSeXs{ttq<
zEqZA-CIFTX+>k2RujBMhb$YZV+Fhu*(ATIiKV2>WNw0A=+R`85*>%DHxz6pBtX-EW
zL3V-zb%rV3n&h_>PNGB4R2IJ*-FNr6NE%f({_&yZAu<OwnW%65d<C%FMP<toca#7D
zi8$zMujnj>B(?jl1VQ#hKGR~iM`Y&ZIEzkhx)qW~6O{DYAg#1=0VzD3I^s-0QzfbB
z0@uD66h_@UXLNF>Q1QEZ4~ynz{6m<(HOO?$u5bBOKys&77+)4eLnR~=d*yqgAip<>
za*MY7#Ht`}3_PA?L8D8G4iE~flB(q;5ze|=(P3F7KK)@Jw+FGy#&BKZrH+pPI&*GS
zlk(%}jIfqt0Mo*L*^iROJ-^EJAhC;R=iQxxQQ|frNFARLW3l3D5Ue8@i?dgNJn`Gn
zB|IgfEiV;e4kE)uoJX9l5Jq9xjHg*_I9?axPvLr&G*CN)6VYW@3e-87{;m*<th{AL
zl1iD&BLw>ka!n>26p_|G2A0Pp?D8ld!_Ojt<^k5-TA1&)I*30@s0N}ZVn{5A!H{TC
z^xLW^+Ij{~4h!^5wRD$Nd{u<Su_v9w@A`3o%yq4*(pKppff91vrr3&#NPiABG`gi$
zv<0C7sZYdGYZN)4n~_!sp_nd^kD_tGW=n+MN0BHC*Z8j>iVK&%d`R5378J$h_$S`d
zvlDA%q=jSMwf*yKtfxMsXaCH?y{!%WnS#g*F-CE*NR)*SUTA8Nv)gZkdW;&PsoyRW
z6&sLdEL?{~Fh^0`v0&s%`E>!z97dG3LJ;@GP6Ay5H1JjJuy-z@bo&Km9u{Oq_qRvT
zKaLBajxEmnzoBP0y}a1O^f*UYmSU!uxV229n7e5(`L!^(I;e6yE`n-vm;n<y*MoAY
zI3w$Aad>)zAXSH$V98T@#wP&$367jb;uO&oX>Ql_m|Wk9X%J4M`rC1tfGl<CF0Yuj
z0f;%v3_`4mv<6uu)F0M-uhX3y1X<t<-(>pf7dHg<=d}!qwrID>1e()jH}cOMus8XS
z4^4xnvfREm9G@3qcJghkYxdf>0EhuLHf$g_iz23BYx`BDF`E$s6lfIUVPO;kH-%(6
z{9LA~H%onV=AvC~He4L~#@9C~5ZesLAB0#ew2)Oy2=0X2)ESczJ#r+r*bExx&c^)Y
zGCk+Em1g$Al4iTth-Eh~GL#p9baE)H1-kS}@YLq}2+KpsD4|DnuEJpTZNXkTB)4Pi
z%<PnxesCVd2-IceL;zG`*v_CcjCKEhTR^(PZ6a&2$o-xWzeL|j=h?C5mSF17avxZI
zD+se%aX=LB>F!VY_YvK#^pFVuPLHWJQJPIm2XgySY!+{)37{0+ED(^@_jly?B^XzU
z@cWE`Xg~U6#ZDkNm|PQ4imxr2*AvY~MWkDV{*0PXA!fG~C;?KtS46l{Bg9QlUK)=J
zbD`j)w6rq{08vKDrq2YR5j#6)#9Tc)HAb5(V)_NZb+lRBuDC?t>0|VRI|{aRj7mwT
zeY{YLgJA<#M2n#O_~N(P!|{nOlh&!n{vt^_z>X$v&%Ys9lasx%-YoxyMIVD*Cb(L;
z5({VZi`j-%M@1g_7YH(6WLbh56juB}AeHK1d%Wcw=hi!p2i6`UC=_2B!z1X+A{k&b
zY1ZOQQI-?ghY*-$afz^kWXy~=N`$#j*46!Uo&PAe2gSuA%)y}ww=l-z0w_rzD!av7
zJ0&}vO1ys?2)7oSFE%oq<&S*2z%UY8fW1PWF^VWXN<HQ)g=?!&3hfbTOinIA)CFkI
zw4^>&pjpK-dl=)4*l=4QH`*k|DCts#X3P-ko{Na|#KEG?1&r(ErEs#4+{M%(b!2nW
zc3{+tePeuA&+f7e%2QGjF9`PMuxTg!e+nb?q3uDP{h*xx#$rVw)JYa7c-1hhG1hD@
z(8br@R4cwMN=Va}ri6XV%toyF+F}Q1Mgy}FA(fUFj|+4KSPLpT&Tu>{Y-*?{siQG1
zfi^M9oNUqt?RZsbt(KY6j-(i{tL_Nyy5M52DJ6vXOT>(7V#mgPX=N%|g3ztDERj(h
zEp)msgzg}FnD}z}Gb=)rsqvVQ+?eC_L9}P_lbs8Qos@1cjx#<3WZ}T53%f`SD=!JO
z#4G6IKQrkSUo5{1zq;`;scOZlqWvLa{iPVyhm{b4gBW0|@v@*?OSXEc8yxvruq-xo
z9a4Qqi1XH=TihVpVv@X3*?Y)F`Q%-J5pXL6za`2A;EDvTtYh})3W(hx=K$3>QxI9_
zgYNi?Xbn95oW6d3Hwdj76cZ+fMA0LekJ%v8@Z<Ss<d~mN6XmK96^M?Ag5bx(GPem#
zsM~ZF39_`{k~h?xL?G+&SecLkLn_61T9DZw+fG#IEAI~CR>5*q63NWBfxygXMsjKu
zNACfYnTk1la!`0!n3FLDBMAh95=8AfwwT4?#vY%C@^~|vnEq`6&IP9z^e9ZJj|p>`
zaxlKT_XHu|nnpD;Mz_5E7l2$$bW_;B4HGM(BG9=dpCy<W*9mi(SaYQbc89%y7!3M}
zxwVU)^U4UL6ATX1S>Qs!E-0>n9N#=$0_dL6%r4xU6#kaXbyF0qZTBfK=1{B**bDAh
zVj3V>HgeC_7fI(1nwBdJHOvk*<E}Z4Gs3yu%S!3eYjLrF$p6dj3zQRaM>BKICxuvY
zIlF4c#-b^_Vuhx_lKW(UQHGAkwPjZeKmeNyxCQgcOki+{&7_E&7yH7Py&rUL3}zID
zx~sqB6xv&9P>cPY7qZfh7e!d`P_|>+gi#3V%mYAOi*B7Jv5#H?>T6W725K8}lQO!Y
z5R&KRe&wr}(`dMjg9J?tg*3(xlR0;+2z9}G=t;3iv>O&r6=r>TMYZEm!PLB0^X?i4
z7VqGBQ;jc|bhFLxBciiBX&)_x*TrQTt`+W$H$@{%v$>?^Ej^oARk@i|W9NfFvcSXn
zYFsFQ?+?`Qg4z7bCb$`>5|4;5ldyAZ)Zykcg2>PIBRclo>R=G($2eAP#Et^UPe+My
zYbFRyQ;&N^I6wAxbZ+EMkU1>Z=85DHoj|hemGo-~^LY8<B>Noc(ViCM&teKinw<_Q
zGG|Z78nqc`3!)@g9oOSY5e(x}HBjsa4uyn~nzquhIX(D~!WbFYG}0y3{|c0wkCUxR
zoLth}crqNf9menEj<&fH8&x38=MnC0=I7oLA$~Fj{&Bew7a1uf$Njtgg=$+G>z;Tb
z)4jSbf~H!ohu<l3t&a5+$`O%S@VUr<Iv*&6BA|(3!zjW+sHj3XBWv|@eht(DGXhuY
zx-I%qVP<-o%hSH`F@bKJJ~pFj+UErMoJEWL<X(k8H`}Cmr%WwvownG0y`+YFQdHOD
zN22^)W&l>W417Np=1+hm&F@`;sDYMT@v;ari0l8@^K8&(fb$(<sgc968DQ9uKV>^!
z5MZ8g{GjlKYxftKaA67h$n!FQT_`F;W|+qWctUS796nc)M0Shg2l$l|v8o{GiikY+
zDE8DR2BBR827dGo94rhu3CC#3hJG#4E)j#UqB348gdEd>*!?1$14>O)ADFic7jH12
zXhHXY5PEQ95^diI&t?j-j&bY%Rvo~zGeUKxD7Pg|l+FFF7nnr>%lLd+@vKl+rmw=Q
zA8bHpc}z-@I|VpFOa2$@jzYUC1MIdDxXpz4YseVhWO$jyF#_QyU0R0YYaI?r;XCHc
zj%j+%eMn>Z^f72yL65D(=S5gVdWjd5E<!wB80^tqrW}>%P;&Vukc-K!yZ}ECU>3^J
zU7G4yEg)Ao*~$^Gk^aBMIH=3dNsiWCcTRx#liDvQ1HPU~Fb~T~HAS0gGx(Zde?`k`
z-C(oId>~4%>o8(SL}nIch&S5MQ&}?ws7tK6z%MZCU^*+prD0{vX=t3Z1cEw-Uh_*K
zuB%qh)%d3Xc&mdNl=v$g;S8~3s>H@3TtX^6u22Def{+c{>t8L(MH-<e=xTy!AL$e<
zziH3YMEQGFcI7y<9aR9hkI`9{I~ssGbG+AxX(HVGY@u;T(h^}WfzE@&5DIam0GA81
z1C*AnxI>WJhZiXdT0mly90N+;qq4``PlVs<VJRr5r5_4%N99u0_(z#Sn+1*VJ0D95
zWDPxUT&ia`2$yhA#GRte?|k9kGRJ|q6x!Ob<kw2_{X*UTEQMGM*ls?-uoB5caJN*V
zDQQ3|X_{8~9_&viS=!ev5=^DgQ_hVoj)$O3oT%U$+m4!OYNS;q&HZ(eZlDxdN1E!p
zgt<5@x6_fL42ka+Oty^$3kKphCqTF~oZO_X>2FU2a_##xTfHqhw+j1`TCDk1XE0go
zAB(j!!F4zTv9X9!=qBR=0l8SLmKgFIadCcw)lDPjokRkE9~DFDMNL62Hsk#`X8vil
z@h8D~UCD4cM(Y!^vlUMWFFFG*I`NVaH%_h2=1NB=)mZstP*)2p)K+XF8a7~e#e^t=
zCQA3%$BczVQjGl;#4km=kb}DJoM`3ioC4&Ejn^l!ed7!33Uloljal_<cq)Jkf@_>k
zIkarrL?|tX8<})jr;Bv?b(W79ZTAfjL~&l%-1xqp-A`m9&Ver?bfrLlJh@eCZ2Pe=
zvxiVc=moUfCi_=~Q!_<%YR7WlgmO(R$<?R}GGB0i=#8(6c6qomOHOL%2_XwK2gLk-
zQWU!Q{)yP_^djjn`UB?OF9~sGg9>msF4`=qayo%EUHJ;VnUneWqY&Bwzt%>~KEnj6
zqgE^yk-Hs(7A4vI%mR^e6k)zl2)ydUu;6w*XSO6iTx9irVX8h$)`9rNSx!AZI1t-^
zi)V9|4WmpT7-shtD521Z`ArcnJUU;1a|O6S9Li`|{+^J`PgW7c!-^|%7X7W(Xf)2|
zcd@R&%|h-PQO;0XNba8bHh@d4$%X(EBXPnxU@jL14j6jL7~(#G{!X)j2_`j5kPDNJ
zu+fJpk&$kzbM879#2>`2fq7@8^8hT7avjo*rqp6(p)Lll{6o4EIF2t<qLE{roV2<3
z33Z0rDGtO3&j+A_SckC8Yl%WO#^%S}dW4hgYH$?4zy!_8Sp6Lcf32s%QcR=zp~9x7
z10shF%K$2YnYl#7W?>tHO6PGLv++1Wq>DpPWCDA9U4U6YbLgD;Jt3CJ9*zg&;0vMs
z^%TlY{pp*+s01R5ZtF%-&VyC1Opg-cJPwxve3kDOc{mh5>fgD-{1&!uM3Iljg}887
zqvE*pv;aQCp*d~;;zdxdvKI?dH@zaz^&3RpQu!XFn~M{a#&|C}7lI!Ponjz?v+?FJ
z=9C1ZN_)vEG>M<y`{4I^k0v3Y$Uv+w!k_16X3ohph}<L4m7u9)M844v{0_Ue{y12K
zf7W$H>|ig+KQl$mi7#Eu&lEo0KZ+(<WHA<0;;c)c>8c7!{cSzF3RR3?lG6PNVHA%A
z$ozO+1jR$Ei%C&ljP7wMAm7hUHyGc1O^9n@^Ac^VPcA{cY*`G+=Rbk40%r*{)vqWC
zWz$_Jl7Q7OgQl$<)0l&qE&&t<t1iy3MNlz<(#T6|)gSsd2J4A9RD=fWnVqpf&!m$I
znW6Xo{L>@@<3oBjGkS63i_XiTE!zwwI(LZk1o256J;r0P2(y|~c9~Z$cLj)Bm{m_b
zHSt)iBa~_({)~oiiL%IH`$s(wyAr@hHOxpX9_6Z}P3z-V@xxRDZ~?f!u+G(BW+rn@
za-!*r?jHkMBDt+To)pRVbR}O;yele8A_}_{5qphmz&w2yM%b*-HV|wE%-1G*P_#r-
zo(FO2PoP|IR#M37de;`YFwr}{Cl&mXV1E!5wPujvXbUqdaS}{}*X4r5m1IzNn<#&p
zz2iW9aS^mR$su}+tDLmj`uufeHsdYYuNhw!l)t8L$>He`VP+C5KPjs@cbG5GIZ9L+
zQzcZF3gx@(P~{&Nn_ds?N>V2V&yxkXV4R{3#ZPX4c1?&+pUmtx{wa{<0F9*vqvwVA
zCAOfw(RX9<5(WSm*fH`5$_+_$B+5uvd0wC^BWn>hox|~ViB(#xSu~}sbW@QZQzl^+
z$D+H0I6qDXiMI&yCs0Y~E?8P(zncN6p!9voCvMlD0Z`U_Q%5WiojS{mSos#7T{8t3
zVHbIz5LcfL!CkA(7%UTE!buGAIN??jxf*l>8je&6b74>|$U6@3<T4>^pK98&EEMK;
z>Q+pgCRQ)~af47wsUYQ=#r`N7={dYoV-dH`ZBQ<!P6U{&J}Ja4Fv8#!dy96FWU`*N
zk%tJ&O(}~P#NlUxsJyO`xFbK$#m3;k+rcds6;=WD_`=VfP^zPKe#x`OY@JhWEy6`X
zAu&X++NMNU5b2EJz!E`G#<_VsF2vj+CW<uUdhrfG>WKPd!`X4^ub`N>yC>rAJ9%_v
z*ukV-SFCpzn7>PwsA1zQAzp1tn3`?e=;hRThj4!jl__Jl93MOH2BhL$Gp5HvJ?Fm1
z`hC8wAHOYBf|Iu?ajgJG9o>@-NB)?9r?iTGQj1-G&F`=!UBY?FJrLA|@fbGWE^2DX
z(>}q$xVS3%UVfk!D4%F1xvMM?l1AJfoa97k%otNI+vqo@SmjWy{Y04oX!6;A#o<C+
z4;+hh4WaaE#|h1?j#s^m<);4@D6h*|wv8#UpLFgk-Useh=Mqk&a1{Z57q+3GX~ZW4
zS<tW<!yHPqxjU&Y#=t_f!ZIV`1j6ohe^;dPrSdwRzDyvW;~a;3s~PUL`aJ-R(%CsD
zeyC@-T^UXn2tFawpT$nQ5zpkL^JdPBcl4awkBOA%>~y$zPpHezvY3NMVyoX@KvFnJ
zG}jPBzMKq=ARTeEXmgbs6ILK@6heaxXdU(+5tMM8a4j4CfoFd}_7C!~J4+CS#wkhL
zz<-If+%`FhMaM^Uya&L{V@1EJ#Q}dTfM$w%u-t<Nyc9;ox4HnylDbLE%s(S{oG4=#
zyuSdK1Ib`z-i$s$ZWN+F<?eq=&PKCaCEhCE;@ECdyU~sQ#9J<|%qn#wOzb6$+N188
z9!HAsmk`)d9Y(D2Pyu1**4Avab_ualmPJ)m1-XUjz>4Ic9(0FLiMni2H+&c(i!*NN
zi2HUz+~nFa$5%_5ecMpnk!hCD>1b`QM@a8UzCyI!?#ci+><q-wf}PC*7A<k7XujGx
zkMpsO9)-vw3+ARAxNw9Zj5eOd&`{O0YuwFEiP~^ID#!)J@dAmDC(BGIH|~OrWv>hK
z5~bO$#zz+yFG+pQOI<>;^l<jACfQ{Ps5T}$v8a_0(XI&a(2Qpu<3$V51hYyjKJ~cs
zPKOqgnk-g%qDa!xN-n-Az&yYxVL^<FcC~dc9i!xMyjWtGxG<hHh>{`9?Z$#6g}>#v
zYr`i&;UGJ~YFr~CXNBG;=>j%<3QYZY=B(IF&t^oM7>Lh_r1Z%mOEb`+e|FkF?42Ic
zvs)0pGVppNzWlGid~b4`rN#bFLzr(w5kebZiIyN20j7;4gVpPVxs`B#W=9ln{LQI2
z{-{@bbqx$>%g=yOILxod>57+$G;5RUHLexloVmJ?cuS}Qo)+e@T)nMQ+-q^uv;2@{
zB}Im+#@_^)@2HQvqwBdMKXl0~6A<&+!pvhrx1jlxy`E8L#{4^hzIaI}?URwkt@W;J
z*-ZdMq<v6`dp-w&h?+Xu!vYyVQtXJ>`}raX0nL*mG2vZa0CY*a<+{Dpi_kRTAS2*?
zdd_*$n{hLx*l*?py<Eg}<V(fNS^>v3qWwa0o=q`*|6GC`nJY3D!kj?2cydTk<=KiY
z0drwwnk{9(kl;MftCcnJH{!3#*o{QM(jlUNRO4x3)Hq!y6)%YLe5X?$2k9o&w3mxF
zbt0qYjO&G21}bBNxa@u4?**V(Db4tjfGiZqTUwRB?-elEHNrKDbM=h44iTm_wiB_$
zqC`$Y68q2O7dmE4i`Vq*YV{IIHah^X_$nx}i3Uc-IV=1F(hWq+z@}-xl;X9p1V)#3
zic(o1SoF&>Vcpk2C|M6m$o2K?c9~!e!S0%oYCnONG?<@mYW&qdf#4p(fmh{fuS2;9
zV-p&~B3=Ckx2{waV&>n%oUu+L`xX$_CFZ<QfI6{jrIkSf-QYcKjTfwe8GPm{t;;=D
z(^cBB=-(hNmM+EBisyDg6ohbHn)kl?CL|nToM5URi<3mVoCraz_a8_qK1w9fww~Rn
zia6egy9KxsY!kXUENI3j{|iQ~Q7tj?%@#!~;+ixN+le4+o$JvP_sA(HvD4&UO;OH^
zGaWovtDN5ob2TTWPU3)koE6WyZxu<2f>0mQVr*Yw<{hGscb<F!Zh5-^nJ~en%YCZ^
z*@&S4lc~$2|0AWVmPWCd^$q~lkVjxG?h@sKa3rb7TIrPkjCX-uW<owz2C>7Bgk>h-
zWZsU;mB=E-N&HyLhh*=TuM%j=s)`T1M@kE4$Ly{-aiM6nl&mer;$#0;{ESrM&MMli
zjbk0V<2XYIjWtFL{NL)C3Mjf?CH^HUiyNz|WEXn$QoB#R&ncrWxL;;z5Em8JumRE)
z3bDOVvo;$UGq&tdCQa*3d#fKMG<O^d&8T<4G9VPGvukD?s%MJClA5EguZTi4*ow5{
z=KRw{n!-EY&rkkhiYc4+U4IqG+X^5*8LKS|#lB{uH#XBFpU}3qqGuPNSsBr$_edeR
zjruvhlFihwh4JY=9bm8c0SI@IrVN1j{p*5M=;_nq4|>jhiXAunNQk+FROuv0M^pb&
zI*W)6kkr`-gIVUT^9tE=B*-O7k^VK1t*{(#=Tg$klY8l>h50)u*a?x>ka5R;Wl99c
z=PJX(Tm_;jas<+j6+Q^$V)yBW&9PYJLm=?4M)cFMctjLctatasoy+qKJ2-_SB$^1n
zNDZ`8|DO=VGF6LF=6x%eX6f9aTU9?QI`drCg_YPqh--slmJ1ljWlLf17ucgA8NC8%
zWX&bW05dGLo4F#OI6t#1_SbXnlN_|SqF;bNG>MLwTGUHK*FK@xIZ}w3#`Z;C#A5F)
zWxDEEavZx$h*_3YdVG8Nm5NtcEF|}T2IWr+r9^`$@bA^LON1H&MevIm;L0CGJ{pem
zS0;tKFyn12{wmr<PWKhn<7FWpU;A-RTz3`7Tn-jfQsC__$lP<tP)hM-VgB;r%xv}e
ztpHlHhr_wy_*40cwiM~~=`X@u6oT$@;mE4co<n8hL{v?$3l9)(!6mFEvm!7oOR)PO
zt3k?zJ|+%Q+T2&Pu4KpYWkK*qE=dD=_B%-C@Obh)JzSuRPW%w&+~dnk7)(=ii60B{
zYLN33WI+Zvn$fP;4;nJqMb_e<qRdSxB@pvHJ`9p2U-#2dh@Dp}645s$`BfpvH<HLQ
zSA^Mu=Vdj1A=<^`YBq+V6#~HQXG-*EwkR<MdRFJmqqBwOrnN?WpO1mK`h023ftn5w
z==>SCC{3kICEq+;3}Ecn1m+%-%r=q7{|HluWieQdFRY%aNyW$=_&8MlxGWZX;?#`b
zC?U-T=l1^-;ym$#vyavEHNdF7mJ(>PW4kp?)0Y((EIuPzQ;zN#<u-GXa1LGhj5sX+
zMp`U_;#QGnJ(@#>q>V=fQCGrUAucrE;93TFrLLohXbM1D<`HD<COz{l?N4ajjn;;A
zelnUt^AdZP5J5*!?exd7>ypZqf_Gk8y)JBu#mG}kTMxhuK`B@ZsJ)I62s=|a*|<<t
z9w)r>Ro8n&sLR06P8ef?T9%Mz?>1vh?<bs7w}zb9e0?Z41)4kzLt;*eNTg`I=Y{xd
zcuQ1?{?pGoregz8iz0@gt=L+$<W%Q^2a3p&O^?^g<u`(07%F_3HBH+H5?<=k<va9D
z<(SY|%lA)%GDA3>W^y`706%y4>)3z(#!%*=RC#HOe1tHwjMLs6dhj0tU2HW8TEBj7
zo!Vp*P^RAbJ@L98U3Z0h;Y+J*3gWz_Fs0uT#Y7kdB|KH!t7p|w*JCcVS@AlGV(w0h
z4+_aeQ(QoeM5}KB=Gw^CxD^KpV6<RVk~`BFoy|MutMo(Wnb!L(QI~*pWjMNr_?=Qw
zoX)vju~^Uk9%i)dbZq_P61iX^Cr!`_TawIUB!T(ZtDxjS%ea|fHz2#YMzlmbw-gIZ
zd$XSibD0RhO<eC<?6Oq>p><AHE{6&-Ba>9MW4%v-z$liB&A3v8Sun&EyUlo4fD6>D
z%x`E-Dfu1jD6+vSzmFN}NFljBu**(ckzWXNZd%^9<Hk;aJa7?Zh?b_+R-RF$nq)%3
z^`-<kR-$1}F99xsVRv<=fGK_u0QSa`ZdGvhd%b)?2MLWheP;0ub|-|e__YA%p!4c@
zwM-yC^r46@5}2Lx4AJ`J96=Ip0?+kF|17A?NiRVk5aL&ee9XBKo}RM{NKt5MgjjhF
zka<9LRX>($>j-plx*KScHxTX25m@?mbO|vpbD+n00PZscN!Au!$FDC6_2kQe(Inrb
zD<O+3Z(ZO-eq+VGz?1S;!rg3|!qA{j#@ubdax0Cp9ZDnYega)NyuH$Ncd?+%fkwJ^
z;jwKCgh)d(r+39mLb5=iqayajc103w&^0ApRU)lgKh{!lwGcWEe*5u`o*9d<b)W~v
zyCQS+|0UHq&Dd&telb4?(~mn0jf9J=P?xc;qj{|Jgdp!&LAByY5qzm*&a7Ch=iK7l
zwICOwJ9jAFX0Xya;I<t>XsAK<{c=C(*vT)USz@n)!U_GNAjE`|UTj^S6GgGKT!Q7@
z_KT?R3~ILba@|#Gx=~^Nz7(YLa6VFy#iDPdyFY$d(j#*J82=2vo4Z_Z&xG`Q0d9W9
z$|`hnAKRtKTcLK;sV@2~P!=&Ph-FS7B!6vucr?DaD>S*Yvi64JasjjjbGokdS(JZg
z(xor61^S%;4R|x=#AAB45OH)zcohxMZwmA`y{CxnK35`Wf@tBd3!!}mSk}ciMDRro
zJgfnKB+@M6OfoGWci*jmIEqVAguW%nT;ZGo9pJSBs2GjJwu>&fdyxR6bLNK=1-P&Z
z`RX&hlLXR^#8c+D=(0@N8K!Fwexs&}y0nv?EqLj3O$tkdC!x9it$wt+lAeDE)Wph)
zmX*SS&qJCWj0{o_yeq^N<c^(Q8Pu=1r{96gy=<^ML^=f>GaWY05aa?-!70CkMwt*l
zA0p6UJ|T!w&tuQO^cQ$GlL+v|DsDo6CQR*j_S?%TJ7;#pH}y;@YbdB%I-8DH^RM$d
zJ7a^r`PGsvZ!re`c5E%o1=1<CkF(=5Liqs2Zp2STP!z1q<Dz|vEYOMI0!IPv6y|C&
zh^2k}V}d*tNrf~ObLdsUW@4+KF|QTNeGw#&5u>bO&@iC8)r#U%`h^w6=wKYEN0+-P
z|1&ku>B4e%(ySKemjJmij4r(|Df}_+UsAY)6@<8;SVysMuf*DdGHbanp%N>9sd$Uc
z8A_)JA^tj{*-;oF4c!M6uPIJ_J022X5yI0<Ycam}jzAdKK`6TY4&)h*a?*!2^Fa`1
zkH(tuY-hbdFzI{HXFsZE7YwsV&LS@OatUCy-;M_a_>;+0Pa!4aHG!@o7JzJhs<H3E
zKvN%+2`vVs4CTXu`B;O~$&2)CR`qkVgN|<5Lkfhxi-IJ!7UBYN8!r5Pp-e(Ql+n%W
zCBjH4O~<Z>7B8_iV-?yfAdg!dpT*p-K*JVfuM&rdaN)b*phmv)1v!21KreSb9|nz7
zd}>atp+|nIFt)c*7M>xR3Uy7N7VB1cc21n+5fZ2#8<Y@}8@7*xof70~6A@f0>bHdC
z@j)9i4(sScBi*131)S$-uX>h9aSqqwOZ5y_FlojKg&J>O1)x;2zfthRtwd)5t8g_}
z>@6S*p^k=0Gb)G-$E)Kpp5JE(K}Ft(--xD+jN`=np?Ac6LM)Fc_PKSCebVVQFn=XQ
zVvKVHI1v}B%b2<uuL#RuCibxIt9(rmHREeD=go*O^g=R>YAza%GeqSUZwz<0<E#>Z
zg*dO=(pTij29(|5So}f|4a)AYBl`P`-?gpP^7w}XK;lnx3|E4I7$1a&<J!2<z-(%C
zta7VR%Pe6c#t9xe#=qgI6$Q2v?8mUi)-ab|aNP*V$PS=l$Sk0@*9QoC8n-E5_ZeAC
zY0~YBI|Rc1S)E<+ks+Qf5}K6J>}@L~50D%b_HdmDr-#=FrJ5CXPjOfI5;HbihWObq
zKf_n-Br#@+4~{_cg;7SFb@e<Y5OU7kxCDl@NsoaNpr>Iywr-tQa^>V(N|`qj<Wk`g
znLJKGH!TxZYdR<Aav(vjkaS<P&_^2u(yahmMEnrq{Izsb{zHP??>HeCiZex1Ar2L!
zJ^7<ZSAtQLN-T3YfIriXLKSn~RfL$;oC`9|YVV&we`%u*2fFJo)3GPlSs?j;Lm;1H
z$j~9{b7PRXfLe#s&sJ*|5I)vPeK$wYl+33K_V6ccfn`4Fl2Na)4;7k4Ls#Oc#=jKi
z8guMD%Jm!t&9VMKG)23H!@A<K6;EXZjZ)7L9>2BtKOyv=o_@`X_e?+`BaOb{@pwrD
zA~3IOcD#Q+&n{GjC}y(#Vif3TgSt?%+0$lmNsQJ*7fZSsN@r$Bomk_$-AlYGrM>3Q
z9}45^1g>qwZ6c@~F#_o`{qaKnk)22_R$5SG$W|y>C2uGQ{&Q}pA?G-euDjwnQTGdl
zP@Rrh)8ZpXI59S>qpSlzn@Pr$CNt8FBGpsmM1PM6uX$DJu_=3n2ZdXnF#lv>zVVR&
zW(I;{NjXlK6UlHB^H~AvR(O|QbQI6Ih{Nn@N8$rVgHQ*RW4tHU5a}F;6o5!VG$M=)
zy0{o1jwzD&va!8DWX_MVmrdlALb8Bhy&a!9)-R8Yj6{Ea?&PAqwx0dn9#%}}iF7|;
zZB*-(V`BVOU~X%=DQ5+EY#tBhb|ug(8#kOC1X)n%&#X(Cx!X&ktOJN6pIpCET3(lL
zEmNlsC7vvEB!(KXuv+n~AXkg)M<n-emPj9}W&8o(72?7nFUqVhwm$(3PIY$9j=S{C
z9Nf8iJgUcBz=39;E&w3$vL}N2BRYhFo9BEL#JuaTU{m`e(f$Npfvo6w>uF(pLR-<1
zlZy01DC?pwnvT^<>-*Eb28BrV%d+%B5oViiv#d1w<2pgP0@NEne?n!Q3})d;YM3};
zAplvj8bZR)6vfBqP4A3r^*kkK6Sx?^B!JRmZJD-4?~3O8tZTV@=@bZ87K6+N5lTKO
z#FfS9U?9ds=Y}QbhU$96l1bRSE-I7MdRQ-`az6D`-l7updo;C3Aid%=Va}O-;ZR&6
z+9jn!V13ec8i0kZk~BNgGn^yPC6ql7?EJY9RY<!m?5!^QMv;$BG4aLtCIIZ<M1VU#
zcy=3WP$kTuR9MHCsBd4zE~pug3(f7SEwvV!bIt&x`q=<O(|eyFm#EIN0c0R2qO-EV
z#(1k~2rlDpGY}2^?u-%AzNBh8xy(g@k5TL`EOQy@ab%I7m5I=>N_l&?5b8FED^EUp
zW|1v13nZ)6=o02JibY9_?g`WAO>=}JFl?MqS|5-pT{YZ~Uy6TI7C0T81x3kO8w^BE
zM8OhWSHVRN!pu_kJQMiENGLWv+pl4rU5{-<SZFjNDz3;Lf+#=@cgb4(Sd?Xfz{r(Y
z^V<cDg9sMbqNl#ggreY;JC+y&JvBq}6Xs)<6qW?*DBGVc@k2N9`58WP`0+0)7pf<3
z>G+Zems1hwG(H_tLbB}A+_v61q_Uu*LSs+4r2v05T^mxwAHQ5b&^kjD_v!pRZ^rD{
z_*`&HHpbFpcv@m1RToI&4o)IY$j=oHFIPAp+;vyb>|~DpNnw<l+az%^7NHk$wYyu-
zu(e|gCYtz-s4VT+q@bJCx@G+fKrOIr99Zbb6++04(8Z1T&V>*z9J(IuUr}uSS)kuh
zKoVy2-+vdxT&N7OQjFOb6-g>1%rZl<%=bXdPOd^7(Am}Ug7^{^29>y01O=bqnD7}r
zJ3AbhFymY2`%Z~{_DFnE1ivD~tQ+Wu8w;SG+<?bnQI{weSyy~&U(2mn0?jvW@Wii#
zrknc%usnLuePG6^LR`icwbhV!pCHOOoxgZQ1ZC`;IU`oRm}g5V%Y%XVh-kl?t`wo`
zpDsxKO3^Rli<dyT2;Ewp$}jY2VJ-sO2m)Wk`9j=E+CY!hWb++=7V6ALF-xWVhFncK
zGbUsGCBcgP)j5Hz#9#C>5-E)-Dht|PI}yJ1+>&2sDrv{_LR>~p0>&g_G4oO|ReDZW
z?4{?-AhtD{0Q&^NAnXD#U>8A+*<8{tr{td|weP-Igxi!}jHI4*836hA(8XWYbN&(Q
z!@gMhhXCZNa6a*45w2>hHouixm77igT@)5>$v<M19|4)0ItZeMtCom%?Px)MB*b6h
zB$4=`e-OZz6j>qua5+R?#SNn)8OYXzt6fp#tjn|+T(-Uv#LeI6Zq(xUqO&lOf<Cdu
zRUp)_WA2=INYDPN?low{Qv#fvFj%9s((`2!g-mM1@>l!)p4w1U^z2=uPTgw650;j)
zCD_?=2+4`a*95@-WYUYv3lwFfBd0RoDmv$aLVwDF`whWf)@?*cyVb6Noa#S>9ZXA@
zwM(!vIiz6+>HV!h^MHY&5)YR&&hoW*RkSN9haGnFYhDXt=}s4dvO=M;HxTNkWb7ay
zd@BwSB%xz~JVMW|B_^0?$gW=GALn&-#va!d?-D5^d0)>LWFh8MmzV`uaW53+Jerjj
z3xg*FxNKYtz<3l-l?Y3Pev|+&3ZW1ar2p9Uex8SU>ad>SazCn|R=n>9C<F){Mk}U?
zFfUuRzUUBX4zWJM%ujTZiP`apo?QZlu_1QDyZscz(#MclZ(=!y*Ge<iyAeE(9GtP>
z&5Bu`Fv@_<2SeE|qFg;@NX%P*D8QL_<0e3o<!=IUQBY~IH5To8ieZ8Fnv(T~exN=$
z*Awlxd<(%Fh?_;b0d@M)tfiIq7j6b+-Q06{9HK{mm@WAjLi(|vIX~%T<C#n(pS(9k
z<k5zxKS^cryIa6q6VxhPO|s0b04_Jf##rnn+6k)TSQAIBOwchK1D*LLtssMOr)cU!
zT%Jn&T7-pL$My&}<MjTw0b2fuN<;U;(NmDC(M%znVzf+z$$dJQxl54CuN7EQ*L>sl
z;zjgx@;cu3=ODEC42*n+^z6@%>Lw-@fR%VppueR%P8m4=@(U0O<WRv$Z}f~Ha_O2o
zJ!ag&vm1ff&}s$gfOim@^G>HLvIpB!AeDzRb7Grc74Hos^#ke^<{+Ws&oEQ+oV!dk
zbZ@~qnZP^|<KnydB}<edj||5)LJ&cWu(({ZD$1YNAz89Xjm>`z=#ruVuhte|soFH^
zCPap+1M#Agxd9k&P|Xj;o%a-w<VXwi2L(~kNv7tQc`pR~7{{%Bzx=E|Th}v%B6ti+
z*%L(Z3YyY^_+BRP@$8BNe#0*mmj#5*#Q2?o=njD{80smsrkDLTzt1u@{w~7Jt=&aW
z1L0f!K4-(~UJ>O8uTM791-d#qmX{Us5)iR@Qi!_!mCAxWq~qjGeg~Z;pH=yAY$1TE
z=(wA^Vnmt$=#J!h^Qqs1P&wjn4aIRHNQ~JXs<{7&GT*s0kLF9trsn;iQxd{nCpqSw
z@CUz`B8$a0MZs2r&BuB97nVy%#<QYayuOOOn*W^v0?}Xzi}rY>2S`DAv@VF1Mfh8s
zoAmVeph=3=1m=#$P_0>VEkWuX>{+(dbMES>DB}Rp60mWm;y6PTLo0SEI@G>Sr1_zX
zKj}3$3vmf`uGbeg{1HHTcg&7E^qj?4tCL3jpAaXn;`>pJr5^;~18fr&c5yF}REKVj
z8d?NJlhRW|#_T^qPf2XL)G2ll#Me6J&W)2E^7DA#KrGaAt~5bTn2BTchk=TN3wX!+
zjqSxJ1oOoq8I0~E!UY~=eq!TveTkr>CQxJCBg9!VVPFM?(b!ZTidMDPBhG`BGCF&)
z<SRlvKlPwn8ho_)2ofzz(}hCJ6`duph%1wL(1^Q)xS6<uhC@ctE(KQzGk;g({>4SU
z!~xOt^OBG}-Jq^a8?LkgdHJ#8Z6Y2KY!l0HLGEZ6pQKAb<2_-PXqFPJ3Yo)Narxur
z%j^>*c=rf$1*ESfe#;X8F0M)!A=<^on}QLzC)O84)6K(dchx70pEa~<Rkgk#n3n8L
zdScn9AoCaGQ4k*%;HKg%bd12H>j?2T382)9Ipr&Cmk{QI%2zP<oM3e672*y<Bo7^o
z!L=&|@?|aWk-{tg4C$hy48z-ni?vIrLNBQ!xQhk3IBFlpj&=S5V&*kjXw<rOD_^WH
z)VcN5SSBGJn+eLVa%M=ewiQANe5w&UXOi)50!`tYB9T}E^VH+mzd~55H6*Y~PV>ug
zg58?zI^};ubWdTfESoIN3@me&dm7MuWh=_5+dcy5|MX^t|8qsTxLB4a!<K6WIhz!;
zI#~t1B9Jl-aGVie`x}Hmj-5*tm0UB<5SAMU6?vQ^+T78KD~0f>#=+0zZ0ZbLH4)Ax
z#Vq7TtGHhnLNk*Y`KIT1cF_qF2QQu$;1{$_C15?CBhQ1m?=_n!f<^irosNveR|UX)
zMq&<(Qoto4lr#k*V1%3bBD5EXBtj;1yvnlrm%=k2h#JNag&5pjf?Aqbx%E&9X6dG2
zYre$m=B$i?BpW?~U?Tb`E<qOIlIldJ9ml*}yny>LqZcuH1kqLvHg^+o(%<vzlG?LH
znBVDk;(z8JaRuc{pI3^k=+{hD1F?}H_#j6kJjf82E+NigQ1i&X<x7m^1XO9qRj(HL
zaFZADerj=(APaMS9Hr1e+$+Qt=^kst(3f)(R={~+|M5Ra;)Y`c#YhSEUMtM6)lxvS
zFTMug!K_MPY7EF|3u)5o`oVAO%#VK7iZg}zZIt^gQ64H2r5iM~czZ<{!qko9$s2lx
z_uW#$uJo_sZ8RC2%gz$u#?p3PQAyT#-7m1PnHfil;CJk`bU&-g{}VwL4;)r;0ns8X
zfi6}n-7Ro)nF?J5mOrY{vTyLJ`!vTZMWB^y1XD!D47@EwAlGaLWFflkn~+qcLM-G<
z{=;)Ftn8|hhQ0p_WErOXSnF#U%$KeZp4)~6UNvsdKr&5`_1J>~_~-~h*5X|e&UuvC
z0HtwU_7;%)u{>_lgjkKW-ky4kqCWLM2o3a{W;~;3S1qlgQ#9dr?D`IH9s-HuM_+zY
zh(E>JBP~T=5>yU#s2b8;4ePuMO!l2~I$}c4a0Ocs%^^pMr0@i|8;`Gvph0x)IyY8G
z15V_4f&Pq6Z`DRWEg?n`LWyB?{vN65XJ|7>aG5B-gVKpbA?F~1%mtZDF+XYEU-$n&
z6YtpiZITJ(Wp?bD2|4M-kO)^EHJPT<mJq(jW{h3%flKW%^?Os{*~a22A#OIrjV|@7
z0Qc2uWuhH_dmpsR(}yV?K}>3lbr&xU>alQwBfU38QoNd09Lq1`9N2K`>cH4Xw8a<y
z7}f&~&cH4fNV#P0h8u?{e?*<4Av60X*RT{Q(jdQ2O*3l?<xS*{see0B<{;Bh>Igds
zDtE@ZKGxp*|10y;*%Yf<v9^0zl2i6k9Ejos5w2I2JGGAdfXQyJ*jHT+GV_An1FJr1
zaV{=VZUSjF8?ppi5=ELBVAA$-g%6TL?#djcv6!R;8wup23i>XaP?0o=tO3)0WI%vh
zbu<O*jfFy}bydTwx+O-U9|Cf_VgJNcIHLU$olEMTf0ijm6{~OC<w-%sxyNfDUKRmY
zd6z57-nRl29bo}3eS>=D7i)@Vtq5fXwU{Wb7n$1@%^GtjEp&@O-X3nuk6)At?2)6%
z=75z*;6{`6e+3Jh5kVB6UY`X1f=sheA$-z!Tq`<H$P71Q+8jSrX2TgG*OfNojq)}8
z#X0a>*+m&bQ+S4+b7vdivm}315G?4I9puuhK$v6LUSm8Es|%s(Ow`EnQKHOYxs%2{
zq9yclO?abxnV3s*E&TMV{0yfQkQjM*Qe@6uQyxodA+$4^TTMRISoOmI@U4SsV4<E}
zU)l=$0J6A5n6t1NDGvJxh}qATD}DV6kQyfmq+E#Nc>G9&->FrJPg{-Gg_sS!x?qE$
z>jNJJ%FIQ}oVvxJFlx)UQ}B$tL{lAPQei}wS<OGwJ{Z^#eONRf9jFq2w^syw)0L{#
zm@mqMt@PY!$%8mvmUF;7k=cGMej}RN&%hq=RXvm25T5MuY(SXrIBZQ6ba(bVpZ76-
z&x1#@2TyU9ek9b9UejE{*vjfK9k{%vpHO!#n8Z(oM@ouMF$&MyMVVP_r{wAWgb<fq
ztNJvOFbS^ralgiSJzBzzMfo+3F|`ZZOo&HSOzz;pNo5lGqO+zyU5J}*f(|U(on|~I
z5GLb^ubFGPHOh?HBD0hm5~7aD6#9^!DLdAuwBc)_U|CBW$F0_c@b@d-+(j8D2ymkl
z9ym4dX@cD797W4DkC<^~vPQMd?1Qp)Es~i7s5x<mr~T-A33i<sUKF8g=e0@b3Uy=2
z(O|~ui5CS@-gy}F?z0Zheg*Bd#>ieFE++DgzUHbj2^$dhVsWn!H!l8OQ?m1jQ1z6V
z(_-Isd8Xo=&T<K>2p2*dB3+h99Ig_q)dw5xG!kDS5bkO9%Wb=&%nYI_upnic{mOb^
zE-yZlGFZG^h`5i7`0w;=zA$C6#bJ)oSpEs%%z5lDS$qocY}7X{2Ntas<a~FXesIkc
zI9O*6eBwr-)CQ-wM%*WY+Vm{YbzvB1>&$<>^#L;*@tfuovxMZv&6DJ31vztkF$p@z
z`lcC|2vz17#Kj&PI-hiYnzr)a6+-no2_d%cMm!^lDYkhtVj86R!%-W<FVeYFi7`Z>
z+okxo2y@}+``9<e8lTLWcFv8p@-r6tagLsyG6e~&#@QQ#@JZQGAz^om%=Kh_#+`%r
z3&{hp;?*9u33MLGP)X>J?o2_ns=`BI;UbFq>X!6)Mg+AR8($D>Z|Y>Sh8>HuMfpp;
zI%TTGa+`rzj1<epBEN+|aZl$^KNn$!q__q+>%?8cTppAjq~dU4^8y;HF_^JXUao|6
zjf<j0e?|~OBlDkrxrB`f@b@JfTC@`NQJ{;NV$Y1m<y(Nb5cmnmTVW{P6!!leLP#)|
z-4dMY&zn9YcGt7t!$ggcZTx<)Fu2cvJ~rAH!y^5$@j8|&80zN?Iy$DsZF<fU$6Q%w
z0zYA^;{A;}I^$-Me6C~mw3zlOo;8(7kw`_lM7!&-h=U29j2{r}PbwlQ61G@~I};cF
z(uS{;Npy~e6t6U$B)KFE{ut`T#|6<&Xp08oI1$d2QBDS2Hwtk@N7#?uE7C%x_#d%l
z2eiM57YWwPLnY7{mfWAz={LF+29;j*J<-mvhLO@TGaw~`$tQ6yQDjHJbT2|TnCYL{
zcq77_it^X`YV0IpOhA?^-883o|0kAMvY1tKJt@pQEG`73?#~LI+MUVEH+$%u*j0Ro
z^&OhG^=1LM+T0>D5c`QXpBQ!(VrG#La~4*&<VCR3Y#=zxedPG%i6Dno+cC(G6%8A*
z8{Hp8!Uk>~z^rfEIersSuEp^p{B2fN7{`eAw{e<c;*DE{@N&l%b2zsDl_*y#EqK}K
zrS;NQbHUw9QWUim8T1Z;PK;NtcF@w!w&L%?T^)A9Rq4k5DahZ|6>bRe(epq&SSQCz
zA0H$9|I-iDyf#7k=WY#=hkg##5W}<_-DwFJOO=Yk@Y7O<ZbNFCm_0-VgAe@I1mw<*
za*@eji*tc+Pj_^)J=*xwkbI9#Pc^<Lg1$6+=Ja??&wiQxqGmEW)HT}{Am-`)!x$Zm
z#5)4j04#f|@uBUYELW_zTC(7XhOo?cPT~-s#oHGM3Mx!$V}%_+C`bycw3`UeDw^=n
z7N~*;3FSj0HKa8@v?CN18Neu%ee;VVS!*3maQ2;ebkXH_ulVE5ctKz;SGoX`4z=>m
zMZ!9JX2t00Gw6gI_ee>HON0`-0^~NpbxPX{aVxVmS6r3&T$x*=JNd~|(d|A%I;w?X
zV1ImU7YHgqOe_v84-}PqhR!X1BHABj-BiVkZzMMVESNu|I3+UUyHrrFoh_#TpDWWb
ziYMpHopvQ1jft&JoRy!)7&dOwbMd~8C@JPGIROFc;)9<ffn}2(nj)*%W_W%5U@p|q
zld%}zNQjF(ULUSZruc8ds4*Hqt|lxKu=Gneuf-uHHjK_c87?mr=HW@&WHgw&0Z@7r
zYjL`sv-s#tuO7b?Vz#7<dFePy?Os5bgmSD?i;t90pKPG$iB|}6fvc!=wbG6ogt?bB
z+gR+OUX;Ppi-M^K!O7;tn<88UmduD`JiEu#&p6}F`aA?IqnF6apKP&KlL_+q`eAB$
zgyZy3S`yCM6O`IExtlEMgXUy_IZ<IMhhrZBE~n%>S&&^L4C#?Fx5n)_MGABlj+An$
zFjBxGs^7C0lpC6!iM5PgJ5eBf?cw>X{5%hf))(|_j<N=oH*b7tZ=l>J>EQTk0dS;S
zTH^)#K&UimH7jwkD3>?w+u&eaB5bPe1MEi9dj9iY1a+zEnB?kku^9<8d+|JDe;+3)
z8J#03WLfgOC_1+|Mk;Kt;uS$IUz@5m<6i<ad~vuxdtaWNBZ_z(t?VkqAEue}5zXNO
zd1WRie~;=p=QqyvIkEhHMIJN(x}x~F5I6ui5c`YZU9FXe^z7ovJ)ZmG;_HH(U0OxO
z`2yVEIpl0adw*yb5yLO7*gktekxSn=0bST<)C8HaG-3~D#O>%UlWDc01<arz3dW7t
z12HVZ8FM0AXJ~n$d?Q`DOzUoXpcCQFi9~A=yPpvEgCgPuTzFJ44PD_jZm!}y2+HLd
zZqreuvV2OIzsEfyS}_0e%LRhLh$h;_2ZOlswQ-i%a}I%?T0byvB%>0Fb&PNe4Z-fE
zltkK%<}NBj1{s3WMN?vPy}~Jey*(1g>V2RjH$?4VS9X%1sd1LfE*I^F(PT6hD;)~J
zOhT(Fdgl%z{X54;aNs`zU@j*km3ZG*AY8z7Cz7Rl4WVWMdM%vMV_1lD>FFP9H{)3W
zE-XtGro5qO99E=Z35Zen>fJ>)EHpHR9VLXf`Z(Ib+N&QaIbV>Ofei{Fnxz__G@LsY
zyEPPij8eBNt=)w<*MOhWXeAfx-{p)Mau~{4LSjuR$W<c-oOF)bqWq^!xWLxZu~QET
zVF<x12I4Le{;n1Y1iFET4+u0rCJ3BN*ZEBq1SWGG4-P!>$r_|9ff_5FGfI?p5lkvd
z)yCUtlU_)=*o=;jSgdE44iAkcM|<k?F9`M5@wsi*Qe4TU`--G2FbVy^Zf78F6q*Ns
zBL4kN0^KU8i^ZPUs~-q~$6Eun!~vq*j`&|uriB7rH4Y^d8s%p~oCl4M(DlUAC8T!H
zamPmoNR@j)9ewwUqMb2|eZ=rz0;uc|RxAJ3Gu=$GuCpLD#43YeluJS|7Tbs-b%S6n
zZ|Iq<QIMspT6P?UHW%xJB-vl&lt6RybzCmMoERRT&wfW|2oDQ%Evm3b=RvOs;)}8(
zX~k>hTZG=!JB&qd33RK#JpAudNaYPjKoJmK3KAnCTyZw0^W)U~Be4-G@lN^UAhT9{
zpiW|QcY)pMP7>k~2ktt+-7UhXd<U^yzt`ZI%HmNjCE!m)QVpziwCB?>agU(f6^BU_
zdmQf9J7;pNp=Xzl4Mro*6K%oN@_@NDO<QjYHg~v3hvS82tT_hcqK*$U6!ypNCBh{-
zS}=b}h&f&O2*j-d%`+TZlMy=1KDG&(TZz*R97S6I{v`1olLFxlLH;B<d+FPDp8%ou
z+L9v`<073kqNA>;nS9Mj&@5_ooekCmxRNXjWPP8mP(4Yovrld>G_MdOpf;>k;>v9(
z*CquZ;~1nKn+c??IV9?gt3<f5h_Bz>(t_~J1z=OG(_N9(xIhruVi-xGM~;A^HD__@
zT3yfnL|Ppsd(`;?ox^1IRr{%+sohtvwg>it+~!Dd7of?iC=GQ9%LQ=2+SpymT+{AG
zbpSy=aY<xSg*>f9dWjwevg%R11+%dV*R8T^XObY((kt&M(s?#|II|$4i4a(W0WD6*
zBHRe<(6FH%6F{}+A;iy=?`m@3*L{xWSFxx)iiJ~Ml=B|q5_p#0>m37<8P!i`O*^Cq
z1ZI(8XQ(pocq|ZK>73OOV|sQjIt#!U?>HeIDaLynags=j54Y|zpyD)vyNFP_$kg%H
z@5@vuyi)(5mvtZKR7B7omnB=j=v<Z_BA{Y__Z>l+H|Un}jtGBx0lUc6j)!!8#t3{^
zPkXTqPbff!Km}Fc1Y}&Gdn}!&5+{i^W3}HLj$aALLRcfxgVw$KpIBf%31V*jt}s|U
zf-z++t`=qHNR_|>?bY%fR<Au=%`+SieU()1IXNCmQU?4>FkGJCB&z$QA|r)g;GC*Y
zh}lX2DKbx2ubf<>U@S$_bequ3Hr?rhxmSYVRPQ+EGh=Z|nSW&*w+8}q3gPQ)%Hljd
z7k!=h0*k%|KrKgE<27Qbg~iW|kld^|ARtNMtawJxv=5en6S3SW5NxQ&4v&xN(M{D|
z8I0>hsu>laa{aGExNazCZ4FS3YlY@+h*v&#c}-((w{E?;NNI9O(ZUVI%K}_=jGzgP
zQ;AmuxlwR4N&%El1+kp)J*G!|c;edvoeZ6XG)OC)R^){OFS)mVM3BGF_N*NXMNgHH
zs5%-cuMtMo*!0bf+ePH5pfQTM&sf^5-I4Q2vD}zV9uQ5j^Q9)T6Wi?@KzSHuQDLQf
zY<)T?Eus6K$Kn=I{`?@3tdNNwoW}}`5t##JmIX^-+7e8tm<e-(j9^{$f?k8H<6*$i
z@wP~FquECBJ3rnNLbW(ztHc-1fS|=^%;<_4XZjgsaO|vS7f1&%$s_Y@Va{sE%b1r-
zM8{&-0xWwLY20i$JE40(!!F46uWGyOyG36X>IUzTGB-NUE|L*xB83NmF>}8Ss`7Qr
zjCp#7W%kK^;2bD_37eq0?gMyRki~(WM?X@77tHPFI)~1gbEEe>o?R<VSGrm@`h>xb
z;Q@kMoGKy@A8ZQ;Vz~<dGW*yHQ=;{R<gu2(|7jH+T>`VTqR1q^M(ixeS#btUxPb98
zD;6U<<ys`fLWs>>C2kU}?l-+7Zqc)g%A%fCGCI7Umzd5z*tS>WDPdO7<WV{+Upv23
z{E&>5H8!Kz<DOLgK*1D{F}3T5E(FMJmY2ON2y)>Sy}lO5Es=))BDp(D5Y55JfqHVM
z??O@b>C>mh{ELbV6+BDV*#Ak8g@9cY8<@0^s(ugDm87u-I4EsL!pvVS+X!jE_I^+a
z-3&6_V#XLQ5nk<qo5z%ybyH;SSs9KeODx^I$$?iK`hD`sQi5owXiM=Gf$)Is8Cws+
z;M8KiAnHf(Ti#kE${Zng)<2y;D6&bu+bK#;mq4c)(V2FfCIGg~B$&c`de%^heRp(T
z3gN0t$B~Zlb`&<%5OS;I^j#2zpw{$}t3_otpu<Hcc$W|lOT;Q-lOmh4;drCWsgHx&
zZf2tGFC!&i9~ti-j?aovwzOR%z9q`pG9St??sr0R`=dQ4{7;#L&>-4ub^VYeW+O3y
zwDBZ-&agnL+u79-l^^l!+V~8n8H-AYF^+fey%l6W5NgR$PS(2|jM6!{)cT^FkL>1T
z>~p3dOA{05D5Jin<Cla|!v)gL{9S}fGOIJ*%FhEm<FUaN;O=u+q_?6bI=7fYY{z8+
zoHv@H<XB0sU-3#%vk0jq)@-aPBzI69Zm3JFEzFtsv=N%EL_703H)*#GP^wwnaWG3i
zGEk!|Wf2g2iu89YRhDoxFW+qo^;br8<+GHLe-q}f^jEmcZLO<{H#uaWAU!R*`N2ZX
zHtk2WO}tK!n~cFjOAp%dJ%KQr<psOJk*lFBPnol6GjPA)T#;l*uvma|Z&oI8Hy(~<
ze_W(#s;u!h0Zv22vUCVH{h9)j)uwjG*O!pCbdtV@1-a7Wto>NB)#Dvuxd8^oW$*Fs
z5-+mbmzJ^fC(d5wA^eZ({A)qZOOX)M=DF)yFwdC7-Mw+CNVBCE<$%WTdVGHoVD4B9
zFkEq&6zRVd3?opAvht>hm$?qW`D64zgiprggk_oP!Ad$-zrOf_cYbQxrv<vf2kI%n
zfexTMZvgdPhV`C2Y8IpxNO<lA*m=v2eU_3sgJcxON}BfEr9_)3!mtUJ+v@H@U8E)s
znxpYG0q}r@#RB5i{pF{SW-q!SnSp%pMi31V>GGXvH$i1-VmM^$9S8gj$TJU9NxI{n
zDTZD!s2|Kaf@Ibe_n(NhhFvqXFuKgqC$K4bZXvXq=-tL*NLg4_rDo*T7>1W5{3q%c
zIyr|qJ@K#zN~OCfP)KTCcq0Ekh~GhU-OBIg5Th2ubiZ_)dk7^9BIk9*10u|rzOe?{
zsy%K4po?~Obwz)Et`B0Rt7j)fA3V?(j|wnLuyJK^UyZNa?sw<S=!_rfIsZ&qb;xtc
z&kKmidCAK5CP9|H!^agnTQcyNKpH~^@fZZrnb-UUh(F%ukfs%r0?ahsz>_-QPlaj7
zn=v=u*0Xwb|2U@4^<SD|u}M6*S48nHhNMy_rBA;94ls9#v34zP5a}k=N@L3H=ZRkd
z=P`=@tJ|dM!GF0E)HS7|oK5z`>UR|hX(;v@d}lZ|7ix(?4Udfy>)f;?IYmF@PJ{tY
zU9EJEFy~5^EkTO~<t7+Yu*|;;pe0kxie>IDGQq1@<^lb&g)qON>)7VU^j`xwS4MaA
z#G}zE$dyVLn_$kyvhjC9VIc9&8JxGc2TDTGF)b$bY*w`<xx6j^)L%I~el(Sg;r#(U
zyRzBMATP}BzLy_d*>2g2%Zq4(-xQbzAVtuLxx(^jjlu^F550S#K#GO0DF(5B73JDU
zv5t7h5x)hRYB?+q6pj(^*RusXGaXM8E|ae=66lg(hlO&TLFL2u6$xc~Nej`VrwR0T
zIJQqmPKW;v$ebk!PM&jrU!=iaG<jOY{XjI&q?CKBi*N~;r4;Fg66{xkC>M0!6`GQR
z1iDygwdf~wiNl1^0P{L$#^?V~<dlxLQw-ukfu0oQdl~15o)VvSvcNm9OtL_>B|Q(2
zBscuH4{7;&pU^zJqI8U(il(9j?_pAb8Qp&bP*G;ii47j~kFsBZgSdEQ0IdNUV)%KX
z0HvNWE1uUg{aOQPno!~+e*%F+(h$bxBKWdm0qO+kP|;IuHi8hTAukYU=HSJ^^)k%7
z*9mi5$lIzNca(|Pi`3Hr=&cVqcj@@$Lvhx_&=jnSS=3W{Mx;?XXbSo8BajFg>w4zd
zK2cNSJbF|GO-W<MlZ*MmEKWu)EX<aGMz}?YD9!PdQduB%nG6D%UjFkEFE=;<Vej~!
zpkjFFMfRB1b#OZL7(Zl2sNbXRug4n#)j@TqRIKwj6pUoruf%;KvMexs#j;O8Q`H$n
zvpHPP=3kB`TZ^ZJIsK%3=JtFN+Wh3iYNQ<p381F9nxO+`nyJS^L3~h`lEx__JbUJ7
z=<f>2-4KHi3{cXd?5L-l5sg5%#J&&zy+0RFWrV)Z;J&^f*O&9xiCWrjeZEX5+0ojj
zuY~d)0`dQ-Iu9sIi{kA|kf0#3Kr+HENRr)|*+dW}=LIAR0{70`$uoDDJ7Mnv5=1}|
zR0Jdk0VN2EC@MKi6p##mat;zCDe?O~{q(%=e&@`Y^VV}aR;TLf>gt;X<r%siGk0H;
zcvz?_t+UQFYri54{xP08p*r(fDA%C7k}Qe;B_y|7vr$+QE%6+%GmzIQ8UHd{?kN-w
zw@sZM%RJ9F+*Ie$csXCR6VgT$o7bdBd{3~MEY~z$vFr-~a1`BfYiufl9_bg`Vk?oZ
zfSf_4nX2<eFmp-jO<P?i&;fa{QxED5M8e!pn>`&&KHY{K7pVZrKnj;GC0Mi>gT_5|
zP_UC6(GwHWARZ#j?2@RCp~SI5oC(u#y2Jg9Ae3gk{jt(Zd{P}t_VG=T?r>FICYvI<
zC=SOa+Na+u(43YeND-wedP1mk=4HhO8|EiVge^33MNEE~FfK7Z`+O_x6k#qOHpR*A
z_W4)mNoq_7jO)Ax;x=Ip$2NI00cMGvdNb4aH{vqEUIMWni<Mr7q~*9Q-xE8Ea9vQi
z*BQW!$g@h28>JW*N^W>gpo@uk#CDzO?NwnG64_JgsEM5&%y>_@t6i-$F=&nT-T+Y#
zK~2Lx^21CvYU#$yM{gG4@h8gb$??HkAkIzB!_r3JBViMK$IV@mw7>Q?Ahp2;vl6pJ
zkV`M3A?wwvBArO3Q!i(L?fi~|Fs3JAE4BH%1=&}x#uXx6a84!kaMpb96=~#qV7SE;
z=7MNn)UNZ3_;wjkM_s&SH4=LXb2D~ul&PbqPC?EAPtP5kN{+q{;%3x#xg*y90Kmn@
z=et|~G-344Nv$#ZVG$m8nRK$-1u&f3M=P=6|M<?sx6+T<DQ(B7{<&bL4@^`rrn^lP
z5jL9E8iODEm%jAq_zy+pu8xo#zY*=^>T{4k<(&`p(S?FtLOdzZ$tcAhpMbfpxUfuy
z*#m;8Ytp!&C?APmWkfToF0Au%O&QJ2jP|Jv)IQ3sis4x5Qxc;1EXC{-wh=`ub6pBQ
z88I%>oiI<`h#&|I{}Qfs1XloG{fuwdgVm@<8es`ETM4G~ATw8Ddl9gBtSe5^r<(*P
z1IQ;Y3drLT_a*6$L>UH64ztW+pBG^mG2J~F1?>0*kQ<vvsI})<>PtJNnxuEEV;xcC
z!%sY{*)bf`1yG>mKxUQ*b6NG~Y7>_Ygt_ZsAm8CzzUR#c=xi`6;Cwe;6hhsZa$>#t
zi(e7VwPWuV0y|mZ+4$cg3U;cd97{eXlBm6v@%TjF6bE(<#-a;Am<M@f<r<g6PQj<l
zkiO|Ub<SU~_Kd4DE&oKCKVB2<+OQ(xsDp#G^%n$lZIbWqv;rm2`E>K1iE2GhxMXvg
zO!hZWWQWtsNV>|ijUcK*_o{L=@oJG2Uzg0{ZV_e@uOzY0iHFMs29l8zFN~DXTzZjI
z28(^U$R>qCO58_~M@eIF6zfLOPP&(BAy|$Hfy0$Ka#w!cLQooSQ)WiX!v1dMd&Nch
zcJ<jeBO4-O#Nh&6E!45NQyY#W1WgP#$upP~993c%0(RrEhi8O^x!a;qMKt_-nHG``
z>oDF?1UXNeUut^z@>j|{DTb~pE-%P+?9&>&nP~Nyw3X`@<r4YMHH~?LFz22w!NfxX
z+#0j60@?bjka=~|WWylUVdKMs#T-0<FR&QjI=JeMujvz>vaPAN(}VEuC1gU^;>~uF
zP1j!>Tw?}v)ObP!#YVDBx9d0inv<bNXpo_D*jxyulCMAn-h(Dc^se-YheWz!h#qL!
zm|zm-f~zb^xk1NxLP5RaV;q5C!Je!ZFA8;EXds=!@wJu!$^BGTgYbKOK@=H_4t!IK
zaF&R__6B>TFt^eW=a<qhcEmjb;Z*C?nX%TAd{YZO4aq9CyGZlCgNdRUw+P5x6uB}!
z6753h8ioq7$WlNgH#o{ecKUXy*~IB6WpHT_G9Y)Us8?79k`iOv#A{+AGLs}XN}YSE
zFqasU2W$nD#ahb(x>)RZF$ix6a7>*wpgn5FnI+a3tX0xs-oimRvO~I@8Y_w*6;6l|
zkQslQ3vg+xwn^HmL|CP&tSc7@QGcxu<Kk`2<)GYzqk}^na!Wth5i<lkA#^gj0C}Jg
zqS0UG48@oza}o10+`*RNIXmW(PAs}SAz)$~zD;h^w>!iDf?VC2;;(-l)ZJFo8QsM0
z{|;RN+^p!5Z%56MKNRLx;>0Umc{@Rni=fFqz7XvqbE>JMjr&(DA~m}4G={<CTY_8?
z9-o5ENg22BN(IWHjI2`BMn?;E8;#(LwId!8;6h;9n4D)^w=$43Wf_wg^}Z15J;oC^
zU90d-V-2u_h+YxS8aZ8eqGTm{tWZw{oM^<SB3&t-HgCq~qH{Z7EKNt*dsUFk9UiAn
z4igV7F-+j`9L?Z*q{MU`D%l7<D$I>7RVS59PVOEPJTcqKghE2Xy8`pz#G+XYS$H)d
zsy2#vT+z3ggZpflT@&DP$$<rSo?NpRM&)Eeg>A{ZB6An&j>hUxnVZ^J(tWQNWENp<
zoVKXQ(*F?Z;xX&Uv<NmoCvalUY{%24v^gE|reK$X=qTk_B=%erj8@R;YBWSR*P)(%
zxkHSr1)0Z<n)+6>PAXFA!1^u;igyX4Wpqf3)nYs%npFF=i{4@~MCKq@x2#~<ULfq+
zh(n|W)^cdh;M?LinW$>{cw=pc8sf~m_D#NBdQ7glE(j0TUB@x!%5Pbl2r~%NNi<Dq
zGC4u0qse%(Uq0j>66SiM1!tNZi5E+V=P26cNa`IyRH6&bbbOqNezK0G*5#KxIJxtf
zp0QFNSXy}Q7O<Ry*Vs%D^&Dsp$GQ5ZTi}@8=i`&s1IVMYCoOB2tY09y7r~1CmXL{o
zHpor;#Wnz-qsVVDS6D^4dhMe_7VJ8|1xC+;3vz;Wx=5FbS&kdywRo+B+OcSqp!236
z3fjWW|8u7B?SaMwJRIkVb_$%!bKKC3+XPJv*lym9@uE5v6qaEMHXKzEq{yY^xI^Er
z1<xeV?>-aYP9aYfx;avd)mt1LrTxsfS%eFLVc48lrLCaZwbf%Y(QX8thz<8tb>pQg
zb`)+66NeIG>id@L(+?gM9G_uy<)g<0yC&9q4R&BaolsYKkS5n*V1jZ#)`dG##gJXO
zIUKWNl0@4}XbAT%tXT0wL2e}}mn5FX(Zb9UroiseI7fifXRDC)Z~s>&R_T|e7%vo;
z5FQhxVCQ)#qPinhYxp=Bj-980&=T!3-Tz)Dc!|3wP7*=0w9T3v@8=(|XKTh1)A_*+
zAF7P?$9bZiDEkuWtge^=;>x1pk@7A+669J?FZMiF%q+qnr=ig9h))EO<1mIP^;yNw
zjoL^u-n~SSnL|S+!>hQw42CfeFQX5~A4&*6Hq0`$xJ8gT#w5zDI>LtR<ubVjJK3ay
zUUMS?dLe1=?2mmnhE!iiU;F#~jXoy6(06Vo_DN~3Sbh^QDww7=&UfQ%0eMEmmVx@M
zx+w^Z(@VqZfp?0|A=ojH@b2FRA;XQPOpnX;Z6<OZ7_)1(6oQ<u9zNkfs2NLbRwPpA
znM$0wd|in9Z-v_lJWYdR#P(QKuqxBm7HfWoZ{`LDHJo@v6u->E@7UA&h69W}ro~wK
zyO3_S-sFO4Z2>NojN!TwYNnKs+>vk%Ikpf)O|e43QP)6xCYn?`Fb}5Ut85OL`KS3=
zv;PLdC{j({aN+<_L>TVo@OU8ZD<fbWgLa5)f3bxl;OP`MlRItc2uM~<Ok{Tx=^SvU
z#0fC|ZpVdr0JAXYVg)19Xf6vL*^{H)6w(;y2^QuKQEXQnZ-Qyhc|r-XJm92&6;%+d
zoQooOsjc`n?>Jk+zIiVJ6Qd6$4zK+R(bB<s7&Q^@rB;_}+}u&dLD#PB_2WWvztH0y
zaBIfaK%U}oA;P3FE>{dA(G>2HDsQ&&3(>g%NCE6mZYV?aV(b_*w<$uj%a<BsK#0?3
zMn@`@-x9jh<HB89w!4Ep>`a+oww)(BTh)4OxLpxUOdn2;YMw4ZIrGARj}6{S0w>0c
zjsoKiAzH!o_*?vMp5Hl6p=hgaU&N@eahI~DMA#&1|F`H4gmELFaO0NSe+8)PC)2}I
zcPxHRi$EN_6M(tfjgD11wFPztqLJo~;-X`=2zL!ij#h<(#cgMUxnrrr=Tp?}gyxi(
z--hV`^FX+o8`PA$r$Xh9id(m2nR}=(ml<^?$7LUs0OJyE;EKBt!n|TaQ?GbJh?_&=
zsM@J7o)+lotbtr{@UA8+df}-%ev?U@=ILV9BH!cJ%qh-^=EhS3TmfGM8IE_#FgT$@
z!dYi`!oVB0RrNSSgzJJ%PCl}maiuUP!chaW!<=~`{4OTL#J?+)=DlgtW0yVnb~*Go
zPd#=OKpt}X9mk7MNG|iQv!_GmTSDxPGr%N(9*$F_Hv|y6bxK<-x0gfnBXbm^<1Yf-
zY;ut?XzzcY2qib9+QC?AZ-|Md4gu#|qWQ7fL<%`rgbPZ4<ml~?GIU>^X8Um&fMdv>
zYP?ti)P?d5?Ng+W>#|xrD%w*wFQMY^R$g^0vD){6-Q38B@&vPi5DFr#Lmad(gxlU8
zj5WuO37lvQT{>lUI!PEJPdm0XvKYi{H3>a&vVLGtp!3AM`K<k*+-5j#O{Yu0738`i
zZE`35F9Mu=9|jPZ6yZ5Sn2U)}Dy8SfKPV8r49dp&rTsx@krv)=`_=(`QwgkK`(g_b
z6afP``sSD@lET>&Qnu=If)E!<2c~t>N{}0{+QU^Rc3sR07b~Dq9pzAbDYUN73DWvJ
zi6Xh`jWR(FRM^-K#|;M(m0C)w!Y&ch{EC6}$g&bYBm>0}%$ij~h;!Ox$4#Q#Us$I`
z;ttW~NPj&&+xV^!HC#LIpDxx8k&CLsFkG*a=}tmXQ}vHWL!X+gFwAMj?Hy3&IcMEG
z_R7xZO=0DMEb=B!AH=UO^n>|@g$*YGxF@ZZ5F<b@giIC048ihHW60ks(mCi97c<Fb
zY}5%xiP38J#~ED?4;x3&;I7&Y>BT`0ad7Nr3xkV<6NepTJ#G+@ntVp|_44iFvO!L!
zZEUQ_<skjwJ}hM={f;#zf#wfhBCrUcFVyA@+Qp6>?H!tuam6Q-HcU=W&^Ni)^uqlY
zBAkX!=hQ{WZppqPJ6mc=6j)6l@#e}G(~cq>4=;K+_DUjKzfky98N`gB8C&%`1`g#1
zVh<6H!M3u!4?7^X<9$M1F-ft=TkMVR6-c|1D3)sH4??+@P(C><drF9hTc6(5j6LfH
z)KfKd<9Y#37CDUBPP4*G0&@xQO_7cX2IFO+?g>aj$;ZPxf^1TS$?Lqg|CS!&7xw_p
zpKCgliB*NstNVCWwWob<98iYg*%$5<)3?qLHZk^je-+PGCDM`H(Z%8i5r#JELr3#Z
zeS2x*$TYScb{6z2rirn*py8iq`FW#%iS)Q?uT@9m!Vzd%iu*mh!Y~Svi(K#P!``|c
z`wBBB@G8K0N4ku$pHOoGC#GD!<hd0=E_a9Yqu6$pfx1VrIc6+fEXdj8ssbl~qTM`6
z`4uzA09*t*nGadt7v{N~(-|z{xlhGhzQlMD)CX7iG~Y;=(`(4RT3jc<r9uJDs~PdS
z5Ngb<Qi-kR6ltKVpneF79l~I4E2e6<>D%csYS?VO*#vNJX$;^)FKIj;oC}x-uf#G&
zd%_<T?#jpul(rf1xG+~nUL8?&BvYlA1v?{N-C)$bDZm9sEZ6D8Mh6!``gOZH&Jp0`
zEvC|YUOc44*j3P%J7Tj#!Kg9&Vy<L}AS0Y`THCn94*}ei*fkE|g*Wy*42VHBWBqBk
z>^M-A10=&Bq^61>HzO~tV4rh1v<o{_>E$|d>?*{Obj?-T;P1~9hy$#lM*KjCE8Wh)
z_)wfQPY|7>4aV9>7AZ?Z$u{jDLUNO!w{OOa0?Y=aA|@u9eW4!}fzs{hSXhADZElaQ
zEy~zs$r42J&QF5wE|BIQW&N(}8_sbW-i%F;f`BWXcz@(Brzq!IS956R^@ksWxza3^
z_^ja7aY4>EnQJo*^cl<?mB&5`{CpWTP1!6+8y-zm_Y_t%oxdl~23rfJOw!y(90Nfq
zxXq=PIXOD&kAsD|aaim)S3DM)3bo8=iEr!Mzj3fviJE8^kJVj@8_ImKKv#|PD&0}T
zc(M%0&D6Y;d$|N<kC^0!Sn(&$LIM;?Mu%w1J0x?plSH`jSl_w^W5J&S<e@XF*CJI3
z`s`}LiN#idGY%0hr52F!o(=??t&b~`s*YB1&$^ue7rR;StLsth*hv`4pw6#Emk1h=
zsj<B`-u)RAY-vNH-RgM0-G7rg^Ke`(gs|0g+`iTc4%>pCinaCauo&Iw!q=b>GSL<z
zru`hk^QCrUBeM4Di|vKxNt$Dvc9Eoxkv*coxuTpR)^JMkQXv|osGrjD*gXPVHx@2z
zmtfHoIg#W!JJ$S#qfj|~eX@@k65xU$9`L4C8W^_=H7l_##=U^{FA1bS>M{olj#lu+
zeo43qX$Ug`Sv+weq%T&`4^)<;BP@YM<at0G-le3GJNjcU(QaMNjhS$A>+W)55f5j1
z_(c%y>frzvHzycBjR_+o)Qt6bSA_eG^r<5;`&a(8Pp+g+I0+&*Uw68~cC`>1cO#5;
zTTgaSTtsk5Q3NH{{H&`i*NM&v)AvW>DFH4@r=FkI^Dq0J0_OA)5e5+;J4%G1S3~qT
zu|$xtBvAfJh?5>1mHzZP0dxY5C=_#_{2G!@pqDXtH~ds6vx+@Xy2z9C)B6ZFi-*w5
zMP0O0ttX4UHBJX`fi>YLkF0l;SQ3>m_K(%iAe<*1lqgMk*ScGfy9n3Ty5;_W9(+%U
zA|WNgAF(i@6rO8?oY}^M3S^dyCHEHbj1XAfisSaz^zEsOb7(4KS#iNLfnD|yZW(vP
zx&mA#?5&a4hU0EQj#|}y15Tf1$?~aS*Aiz2Y|2p&FLM@{S%Z^~{#Z`5Q%jwd+eYUK
za*jOUG1QDp1;F%lbvpiB{#+#+eBR}sr>@_|%D2YZ&HxDqok*>qkx8-D=b?K2PU`8t
zsm>lYI0wL`M4QQILY&yP#4v~BXo!Vkm-C8n(lFD@`eKg~OK*)xVREuCxICc@X^;1X
zXjI5^$;rQg@L+B<byq+qG?$(4I7|dw{lV5N{&WFI=97FRJyg;SrZ--y;&+;sB^c2v
zR=v<925poqR{w389*&w=tIiVQT2`?H#5%n%9u;Or^AsrC+Q$XBY(By0p}n6LXh)v&
zI_uVi^Kbp+4mFC|A!g`{{0_v;$?{8AXvA`Y@<IRuIx%po#eM=EbAVZYyf1DM;=<^i
z8t$221ONx{<ez528%5^;+ONf5F9v|E?A`|BKbM%mMx+vFTxx=&vgvWB2oD(9`f&`a
zW2*yy59(6M+<}|UETRd_V~a_hg;jRO33Z*g{!icx<=Ape#0$*09#WDMgBn3t`on}#
z#wO>Pahw8}ljuuTDb_m#emdc)`pNugj$+s_7?%s7fwB74!SS1!#M!O3P#a%Xghv=<
zd5P_X<Zdj79eQr?@4^_~Q)f<&`}9pMSp7&KNhKFt4&><zbCvQ4=tkjWO>dkt&>we-
zbXV-J^mWHeqMZ{Hg<UBu`G+EmPIF>s(V1r)I;jS$UQq;T;0`1<674E;uU1!j4;SJV
zuBIo~&lBLBdMhwq4z=GXqv7FHcPUo;Q<>0cKbwu%RtOVDFYVeM$BJ@Ubo+!Yb1ZeG
z18~%Xfb)VVm!xJ7iL?F;0^?dR;`^??`4uL!Xh_bvPYC?!!B;<v-p~JHQX&eo?0#24
zPSk>JbFBPV0CyJNc-Nb9jQHyXI!`<&v)POrg^;S=d1BZ+C6XeH^LXdi|Hij#;9Hpp
zwt}2!-(Y=?wRgJ+bfO$1ps$X-g_wimn9la8?;KG^<zg~#(qje`XOvM<E9fG|>VGeC
zK%B`3ZL3^efXU?_Z)Nw#K?2=DkXY(*a~Vjc@641;K%bW&f!e!M51f4sF*P&ln(*Ty
zU_cG)0|xOX*Fu_)tVEct?k9j&<@yTK!|S5V0D5$PGfCtNT?c4JbvL^(5voOk9FE<f
zs=+`{pxMGy*vk;gnBxv$`zY(8h0EmFMWsiZ7Aql}Ko7^#f;?;JfR%R-N3qK273#r-
zS}Gl&r#C517M|N^w4Vte-Vx%8HY?qAZjN6MK+{i~IxWuEw}W?}W}#i469OAJ<X{r)
zy`hMVD8QKi;6?yf4KY^hR3o;ysR*QvK`s7V0_lcITq}Tvs5Ch{e@cYgv#RaqiZ>T@
zXT7T{ekj^i(<GlBFw&vOuZ5Ecdj!<sB3y9#GkdN0gAlg_V@a$+lk=KDbDc4d$tf0q
zw*XQ7RvyG&^VZ_$sne_txls_g;EZ>W3s)vM3!f5u-Ui_;kzb@7IYS6ZwajRZSM}{?
zN~#AY9k{u`?Vx5Cw{UfxcO*6}G4@U98VR?VF!Q7jBegicq{q4T9nXq(BQhs3>c8?2
z04g?I(?f9Sux&d*ba0&$v##`uEZB^uP_&EUn_^U8-8;a{GYZW?g_hm*1UfJ5beO<S
z6X1%YBG4B0dLd-R)|CU9=(!UTp5cRQG(Nf;!ujM~a7TQ6515n00Vt<W-?`T@+VPnj
z-xc9snFv_E>jLuNV^Z#taZX$>(A`2-`r!7snsDcbt3!^qn8j~USYs7LXWe&-n}p@A
zQ0cC4S{hFaq8iNfc!ZCZe?r1B<TUPabB|JVE(UfA$mqPcd8$C-^s$d-wYp6tU7!uO
z4!7&uU4Y%W+FCQyl>3TQSU1y(JQM>0T^OxjTEPBVLJirkv-JE|5M!}|PN@>h+z;gj
z#rZu?zLhjym^$O&2l(A$GFAaey3kUv$AiEW809Dm`xq7JoOu~ej)CG!4;N5&IvN`Y
za+7mK3TY#@7eX5&^h}Et9)WOC<mx517d??^9X~$JIKB+RQ*3kMZP9KLG<K*(Q55g{
zD40vkkky%cnzsM_SOKFXlz5!_cY@8{F)sGA_^PSrMQdD><Q$cOo6wva3h$~KVzI}I
zLZWw&RyV#`Lg`US23C(C=b}qA$oNu=pDdILV<4TDn_oNutx06cj5t}}nlb0FH;UO$
zLYZ^4lI&6wV&>2RlE&{e1v3An6NzU^dTJ}XjJyA@NJQHtmX8CU2AP;%sJf16@KGbw
z4AjXHp8Gjf9W?AR$c}*N@<IhPKRD!+5qv#f7M6##EYSwz&}RyScj;vBahebpv|mRQ
z*NUbWjB?%**JXn8RpV_DZ~@_KcqrmIDEKp@bxL&R?~U*aQh6TS)fnedAM79`;v6f~
zB^hXP7mXD}h<T>XUcK5AEiVFD{lH;8x2q;DD(|5m9G1LMZ*t%v$kj#H!tkOQzIYjI
zVhTmU8k1fDU|=A&%kjdHNdJzylK01t%)hg!OMtpqBvsm|b!yx*@r^696JG_V;+zO@
z=Kp{w=PZeiX^KmNuYtMJ_?kcykS3CZy6u=PhOnGWy*$C@1-a^;^!<u1&a%*T#942c
zLiNXS@6)1EKb#)Rz3K0b=#y9I8=h!b=}?<4{1-tk8GYYo9#0A*X`Dv0Df~bbjosZu
z@Mdp2`z>fV;(O(E!rk6BfG8@v#DO?20|<G%K0%bz1-LCFzi7Gpq=fW<wx*<b#~G0h
zT6;W7il!yxtCn*7LZlO6ykXFT3*>pAZr+V!;NK<4jM70Z??<ihZh^?YmJYp&kSr7E
z-K}=D`=11o8@V6EBJlSz7DvKb0C)Wl0buGh41*umx4G7Ze@ESokJa7-V-Td%o(n|~
zc-G7*(f@uC825{|_gwUY0+UXlot;^%SYixs8%wM&%$)>-GM)KKet1QwE7?IzkjWo{
zxSMjEfr9)A0TX;cpWhMZe^dmK;=Mm^7GO5WG9LelB((W|i%_zsL}w|0gfJ3<QPR_n
z2yx{RGIb8$jQan;+&D-`+MLFs9|OS;y<Hc3i-5a{SCu$SB*pF2qpVtNE)n3epxNX?
zs&ujmbVf{1QZaSJoeF4%)VLnNA&R<6LOp7v6;96fsLX>3Oc8py85fUcqN$_C7eu5{
zUY8~3H$QPeO!J#@oCv!AIDLZyK2fe2GrtbJW5rJ$nXWQ8jF0@!9Jy^H=IEBsA>0tW
z_*|*R;Q~mF&WX~n@kQ}7Dh6hT(*@8`W?<WSjlSJxeHAV`4#tf_Ci*;X9H>)*%oL_D
zlmJLZyL@Tq38v`knyd^MPYHE$bW)y<kBb#9BP<|HiL>V`5DZeaXjc{lah9CN=uM_r
z|I0=4<LEJT_C@k93(ay+H61rfHWBFQSk~M4E*MOzT>sz}qO(zQj@~E2J2&*c=n!iT
z=|XZWz&6u}dCG}Dw&cq#L@u;za`io1L|#17qowIS%PvBRGr;}W<OOvf0WJX3Upkjo
z$2%e?)mvc?a-0aC(ZVsNFYZlPA$h-kaVA`!)jRWp^!l|H236Z~U%WeZ5#?2}r-I~l
zqi9#eTkv=zhv(8BOUKwE{7xUlE<nyQ7F^Vc(W};+8M8$>GOHJs>amv)l209+j&Pu8
zbA`2h90yDS+#m>b%;pk#o)%~Zb0U%s)ITpHO4r^W8+^qnr-_##|DI@fF(xPc%#zzT
zz6$2@v`m>I<DLVAxVm_=?T#ZwJNv?mQ~Xw-wNF)?63nZ8x<NnXl5wkmxjB|x49FQv
zL6AJfO%~>A_Vys4$NmCfIvW!lz8#Z)mN9gX6b*kDplPQ*)EU2Dyhsqw)0N~(@)2RK
ziChZM<^LzdB~Ozv3vqu;`x>CD$ILSpeWG2CdKWwE__+Ww!k}{??q34JtH<EbaO|`s
zBza)ugvo^(d*`J<$OG3~WXi5t{xfE05XQ6T<^HSA<rC-~4T=uNa<(i?xI1uKM0JeD
z;5<}mm}jU%?RWL}!Zf?FZZT3%UmDU?;fCsPTqv5VbjcD>-fmyY0Ojap?rz2x8DNU_
zHl%BrB8f(24(f8z&Xt~qIuS|wxMjgSM)2S@;dbUxp^`-qkWLSO_cDpTr*Fz=_x$m>
zNVv&HVrrbSoWsq;3F&A0&fTg$$Z9i8ffMNLb$FW8&tH@Qb;2o8xE6;k?_}9n^u#$L
z+=Lz2Yc}H&0l5}vY0|dm*sp_8-WIgD(^lZyT@<r)-3yQNN~nebq}I0X6^m5pP~LyU
zN={`ARm=MN_83SyfV4~PV*7%h%a!@bm4k;vNn2*n9TdkhYy{(v3ML-no%C<|cA3*v
zWpO@XE+=O!>2kvZ3g%wKG_0QDt1Zt6*XT(Pg?%B)5!;*Ta3Z!^1<1K`q{*f57+VcQ
z9GTV<_vzbhj3al{0QesmiNV#&IFjGF1pEyUcUJiaOUel8y4L`9elpHd*?R@KK9rKX
zX<&1NxeT~tl*Z{KA@0KRm=q6|389EfuM<8Y$VJOeo|M$vWkC7~Be_2oT9be-8&}*L
zynmT2z5glLWkcHNV*ZSG1i8b|49q=klK@-{{7Uu3Hzz~8Y0G0vOYG~c#ZRt-OdW9r
zdA5+;*z%c{jHs?!8_-$spd~xSe+bCy6kTL^NW#=`Ec?v@W)q)!o`li_=R%{Wkz~EV
zI)uy_qg7#HT0@8f4IwXxExQYHF4zcT_~k<Hwk~iQh1~zGi=eB>a&a(D73FrSRAoiL
zH1R)SPIv&X71EG?XFVWKgyWda{841C9t#Q95bFbwDB3EHtq$4%D&;qQM%*pJ4Jcm#
zIP{YglTdT5zn6P{GIKQO!mtU%1dg3r3g{|fMS>Gx+Lkijr&JAVy6Z>u5CfL}7O=Ub
zXCW~JJwk{pLyeP}{VxT%^i>{t<eg3d)J|74IS*T7!{YZEm*|G$$SDBsA1o1Fac@aa
z)j8qfQ~BMwqQTGs{?$U9D`EvQSw}qC0+hzvtY~lL+qLHv0v3|t+T8e239@8lt#$%k
zR=v5{AAeIQ=f&wmGVf^vaS$E=gg0AE1AzSl=#OIu5e}d3;Sa}N8KKW;;Gk@}PKZ;$
z*pjl|Grfq!nubwa>^u|1B}UMb(+n1=&1V66HsYCfP1m|>IS}K*-CKq@#W+T!<8peO
zMX#R<q>x-frvM^cNZd1W?K?inKO^(jW95yCpCzd>aZD3Je@jObahE7fq_d{QEBba_
zBp|XS6+mHZRvDj(ba#}q(!8{uzA>oVOYU7z-Ng4dDH5t;bIjAr0^B>%R3V2kDqr6e
zjGU*nOpU9*UHr~PQCo(6Tw-%;s3xS%O;~PxX|*U=Ef(C2$Od~^#DLR{6hQ1(yT)Q=
z{q8PU>tLw-Jk#tlHR;5kMZ4NqKyZ2Q9RVI7n02Myum;F7-vM^@Sr`)gK&J?G>P>vF
z##GU+s*b05W)QXCcfnve4ry`l_-9dhSaRsto_B03Zvjj(^tg`H=Fu&{;h10OZ`R|k
zz9n$xCG$(V_nENV25d3th_OQ%jEeT4cIc143(M@Ip?nYK8iDRWJPX6R6nB<UW!sf}
zdb}mfIq;GvVrwPdD<LT=b$qqjRz(I$hszZ|y1GCYm~bQAl1bZz!L_~~Mk}4NL$uof
z|7L6<&liyMV-nCry3N)F#8^RT(h*fbE;~;NsO;lHoKO{aSC~zeL40*2mfwaT44*z6
z55*HA9G*#!D!eDa8RM%UN%{wGTZ9=vBi<ipW`OyfO%j=`x}B56DG_fUig2Om>K#4o
zYlOIupdB2ImqoiL(XnKxq$%O`GSploub8qup(gBz3O83|Fjqgg8Iz3GBOiJR^n_Bp
z$E@`*;m(VF8pjgp+TT1$pbDXL$z)<45CfmVF6>;o;$US)*)^@|%x>8oip=SeH4bZ#
z0e>&hvsQbpJ#NV)+az5mUVcY@)%4Ac+1dJbXV7C~>ijVx%<Mq}ITDwNb`$X2Bj*FW
zBD)h1ZKlK1I9`Nv8{`E0mm<w?)X3VyUn#_y4DhVqfV5e+?p$C^UA5MT{`zb%3O0S#
zl-MJGw^s+_41JTEZji_0BFt^<<vUO(GRimEr2tu3uuw(!X>elRpi!&3%-l5W7C^FK
zFs9R^O_hSnD1#aqRmGMi#x@DtCbo0i33F^+h~+^=nZoWa)GT5Hg9;6$bP2KACGGxU
zf~W^h&MNWK@^|GXne@*kB>Qd5y5j~xx%2iUhxS9e7Fpq4P?kn<s-QgVuuY*n_m&89
zJiF?6MTj*Ata>`(!EJNe6}`P1zqq+4$w+p$??Icv!|7pqvEn*`6rB^rsj=Jc5X_%G
zhPhP~b)UF<&jp|P_H+Op&T()nDr>;gO|%Bb#!{H3A120~-U~?k#jhy1M~+^`!nq5>
z5f#{)!d*)?)~e-tLh|H-G6(M~0^AH}aXFzJj&=7c!k|@QB7R1IGu6#VoT+O+`GdVd
zQ~hSlh;|XV2V>Mq%RIUd2sQ5In1X)#eFsJw9*fB$T!=CDU_CKK0EIw-<TuQ4`zc5x
zPFIun<vUOMGLpf5mq(jRFuOjbQzy_}61#NNs5JGR3Yh1|^f3B=1-Q!N-970&#EbR=
zat-sLi7sN@DwG=HRG<<)KPZBrLrd?LP|LG~I<0p4wc7pv_IE6F#@s*<;mi^1Sk+HH
zpa2-^7>8aiAa_^P{!GO)D<G}}+f(g@JK{uP&POMLjQzN-gc@}_<akAp)55bM=QFWd
zwTQ=g7Mg?M*jSJQVU?>v^lc#%?SVokZY*QTyP&KfN7@PFLh&?-A0EE529~=$<3iS}
z-1X@!l82T2#dShR9<8tk_4^q>HKEC|9{Y3^iC5{l9Ef7rCCo#B=W<vq>v(R_a*%#-
zuGsppnU*7*c|h*VF^F)IX3)cQ)`s}DnWH2p_Rq-%$97q#2{%VNx$28R_LdNOUq4Q2
zKhd|j!>*Md&+9INGZ(T0iuHSpVC%-zh;}i~>~#VyjE#%*?F7<A^LRi2{feu9xJvDV
za3MQz!WNr|Hlt;|i4F<h+`^m!z7biD59%*~;YyXx@M{5{xI6ntnPnS#k~Y>I0Cv9~
zuE`5al7td!4sy*Jt(@kH^%OJ@oe9q{ZNl8^^fV|=eq!S?MMkSU!)eQ~qfiFBERE4U
z)}b_ZdAaWzeN)%|9yIhf4?#F%NrSfddQhO(!b!L~{LVp;ZVV1Obq=vzi44i-Mn`;K
z$V6w9-j@Y@Oqh8jbFO+EDj;Xp#9kR&sDZdlpoX+=R9#y}>10pSANLINtJ{L}jz)YS
z+8suwIJoz=RoBPDoiv*0I(I`TR07Q`HqNY9{jqF=aIht<R<W-r*fNg$k^S>`FXO#e
z-)^K%Zu_y~YJDCSYCh<$Jwvb|#JRF?Nbo1u1UgqKbtSB9J>q<t^{zNa-{uMyLL4e?
zISK#+5I^g2kO)}cX@5oykxnZ;heY!8j{%t@dQu_|5uJw$clEgbx$!s%2_gZG#xWxN
zvy{DfXTLy*3$3lXp0DnRwdMfkfhbLc?@-oaec^6b+y<lDp{w5|%!R@wG<J5i_?Mu}
z8VqmgSPu%Bn7<n%QpYls-Vo}%^+F)W9?TSbHjDH??230T5LI9>kH+gFa(-zmhT`jB
zAeu2XYrG<g&^$uNGbo2ZI17fCY(KG*T2~+q-O9U<d+R$VSCMn*j_4FLA^dg5wucsE
zwWl}s6KUSya+zVLA<HP(@Q2`eUPv!cv(%hZVltJYKqGNpiE)5a<AK~}TqMkC>XfjH
ze`zYcShx#k0}&JzWl%YsFOJ?nDwE}a4;dAacBR9J?#_j;bu>eh1UNlT!scRV9yLKZ
zYi^`6DN!Rt%1edAqcM5g)Hj`sRW-eQz1`tOGOXb!%sQf1ko)0)c4Uhl0qyyYIaq>F
zQVlO9m|!^SOTJ+*$h402D;R*kA=*8MiIx0f*B^nn4qYfd2jfWrn!ILBjjtWWw_6vM
z@rsh>)R7XjK0=HBKLzGw@YKLfydML&z_Lq-ABc9N_oD7OMx;Az=?)++Jlc_1XV~yI
z;tA0%uHRdXE01xAv=3*7={OeJCG<{R8tB&qI+~_g8fL?vfVg-p1NcQ=^ruCT>Tr6E
zWV(>tQIj$fx1%M*_~FiDJ$4r4a-$o<V+3+>d+Z_9$#X$|bSS=eTmdvk8>5|3F9Eh1
z>@?>886jYEFUP4n<?j}}KacC%W$%?AM9dV0F#4F)S*8{}-V}Wud$rv~nafkB<0y1I
z(^c#|wWytNf>T51$`OMWfXjtwY__$=)<5Ul%wy{0=u(F|YyF}CWy8<OH|ZqMgHL*+
z<RN-}1<Y9T4Fcyk5SE!(BRhDrwIElB%{+$*abk%up{e(rS|S{V>Y43Be@QG?OpZt5
zIMH;U!~~2aL^20W)A95zC;B(!BkCE?iF6s#9L!p>%Sk{k0}o-T{ZA3%K)M>CLM(W)
zlR+x$8|{moM7mA!`b&?VEr23*k2Z7<D-IN`7R1oBu5U6(R|F9hM&d&u9`|fG(*`~1
z#^*l;e4?8rzp0U+Jj$52u@ixlODh}|*6VR}f;B`^VfNd7(JF$Hpj?-D8rz9>^^j<h
zePWGYgE&c=UGHO(<hBB7Fdx{+5MnrfCCtS^v5@$1TM6-KsJyDg5vLN$t-%}Z_4rUU
zrKw6>KjkzC4+*&x=8mpppqGVe8sVL*SMzu4)LHSqzRgz7a<Q}e=@|g-M#!!75ays~
zh2?EH2R-sqE|Fj!n3m;{dBz@DJzu6UQtun>;xci(Aj}M~$zzS4vFlkyR%tEP!?uf@
z4d(I;SL)oOxk><x$56V;2)q9rNa|6ow#TXG^6g@wUFl$qrArbwo(F99H^(%qMndwy
zLb)X6P$O<pFtb|Lf12O!5JX<}Dz1Zu<HIrp4;1N6bbOpcw4-HXQ?Tl9h+t;vkV=Ls
zv6V144yS`1afE2+lP<^6el6z%xrDuS`Fe>}F91>Zk{V$+E)i)ipwDAne?kB)%=^Ce
zcuj<(V+$<x%m){OIM4nj<=W)8(2{>8c*L2aNL&*g)mrIykmd%K!Q;BNf<3N4*97+*
zv?TG)7v_==(G-$aZW82zjbR?$A3I)D#KRSQ9~bC4Vq6%x%XS2R8W%&kv<-}~(2CuW
z5jhG-?~cW@LR=&{SYXe?1hDWWfSCzc)nm>mz~$+a&l<YRUV_SXR~OPHbc_lnPg#+$
z!qX||zZ3)pavF#JRD}0(NI>$QozJE&$)Pa2Z^WU$cN7Gxfw)72IXrbb2Uj0vn#M^-
zj4NHn?=CFXaSV=cWPtR$>m68_%oIQ?FrSTfcSgTR4^xcD*+iwq{8*u`B1}NQuf$IT
zW!|tPSb6+AfmEkUs(^b$n8S=i`f@x{CPPOa6F1)%<Ob&XUoDe;vB>3M&JC_5v-=eU
z(RMmB-~l92PI;)&WKoLEg)si`9};`&+jUNFipJ3bsA89Pk~n-837eA>dHlCy`r}JM
zm*@xQtMe$PnfS9H_)BrqqbXmAcCFc5utaF9-S~q8vG>-q74fiW^AI&>dax16bl*P~
zQMF6bwVlDJ3xr{8t8g_T!Uad2LGQRyfD;<SflEHtOQ?(4Ta$*qEAGF7faW5aXkL>+
zfBU2`wS{aozx5{wSb$nctM|V$iJ7C`1A0j0#6-${1szc_`q#M<*p<}*3}zZA^y>mS
zZ&|M;PSvLi!i=uHVf?F%!fT_L@~d>umEl-+SlowWn?DoP1(woK&ydCUgt_{703zOx
z1W;%xk9i8=((-pk6Ye?h65#yxz_=E#@B9Ubeqy`CQBkf}x@^GSR#Ngqf}K%KRvxkL
zRgNW3r(>~+sN8S8*6k&PjJi0mzzGCK7jcRpctgcid_Ny@xj-k3I5#{L2_T|8w##ap
z%<oYcO4uchcl_Y5Q04=Vv3E0Uc@Ok&MJfa69+_bN?kISnp;LTMRE{ENk~#q1yTlM9
za8JORJrBnH8lK`{LQa@DhHG8vKccCqE;B`9WlMm=wB#!wS6%G_z&+;HEv|v0S#$%4
zro2m}ryyP@>xmn#g-&w-dzSsKgK%O=CrmFtSrEU=Z+YA-0&Zac#vPtX*F(81gNS^U
zI9PytAlnxiT`-GoaYF%u$m9*SCeTeifbfI)9H&<(+RLz<G4sqHD^F}C(1qfKT%Ocy
zL{*TPN;x(5yW){D*dPN+Zv?D;Bf%(cC!2wh__ipgQ^8zCuSXpq2;OLu$zZujq|46+
zzN208M7%DLxvI6LHCDWdZ)$9_ChBps0CA5R#M2@&_jFT&-<P==hzt-{Wbx1^Qu5RE
z>D>46cayTuxrJ}nM{k&7@!Szt3Uj~K)Q&{B(5)cmw?wjG9!~;rTlS2$<K#lL12iO`
z#4!R~BFWfd)6;?|!WgzG>)&36M+e>?ZwjEeOkF$-9sd<+e#n`Qr<qy*C?cg<E)Eyq
zf@(EIbGXeN1%h%(;^e+UTvEi(k=XLiB1AtgNy5CBO9ZpMmMN|AW{IHDZg2s&>n>uM
zP3b-{lXa|qH<%|~{EXvBf%OG*4WYC=!UJ;8H|{BNlM6gL`jP(x$$deGi44kf1(|u=
z45P75xevf)?qlN>qxVBQcaA7g1c5F6K#>kRoOEAi)e>SFW^DJz3_(u1(qBcfd5!>g
ze;J@n+ywYn8Loen#|h)%66nB0w<n$zke5;<Rran2{R@OrV!lrc_r+tP>5)CWJr*D2
z9|w8Lf%z=ndeA>s^|0gt4?(y-a_5I*-exTFFqrv*8?E#}_3?s8kduH~eEAUw4Tjdq
zF`{o*B5gi+s7rJD5|0*1pk82!?TB3k<&H~EmFO1WtmXJG*?b%-%vIweh+aKCqC_g(
z(?V|lX_*kly-aQw2yhS8B?PqTbn&Gg19T;N>$nJ`|E>FYkzs>}JF%qdh`WVpFk;77
zdxCHJOGlmSCh>(R*NiDr+OKpr^pz(ap>@ivSXtjhK-bRxdp}VV<4m(gHTp}4$wgf(
zZp`spFd^Fc|D4t|p8Gsi-=sCH7kA>Hq8NjnY?_$r-WTZ_R{H14evFhp73cyMiWZLB
z66|7dnKL~LnjpGIIwpoUpO?w$_<>~(@C$*Chm8l@;^A2EsWMx=Z(Eg#?!p2cuDdqG
zj6E1jln~N>8UQy6%JTu$W@vpR#LbP1B*b|}*CS7Zd313@4u=%a6lv3I<ok;IIYmP3
zIOLXX*=Gwx2JS4cbA-SxJ(f<fXFUg*`-9XtX(qi_pu3;^ShUAqpLa5JTNVbLp1&`|
zvj|fw&T!IM@CQP@-RB7xoT}<t1FFzZg<JBM4vwvCcB7hpa58L|b!5rykLY@V;AWc~
z-^o1b5J3#8D%PK|$cqqWMYlXNt|XfB7KaJrF9DdNX+$;SJR!vI(`&%jzg+yxp|b{D
z;}sC6IIf2^c6=4u6_GnIu2fr9`i^k3OS+`NntI&30$p64iE5AjnIPB8Z($_`*($Gr
zx>=cf2+54|7h&c#BNki2C0{R6V=Q6Y$<0nd6uvT-edKi_==|8h^3(U<@K5M0`r^kT
za(BzF2A6yjh>CGqr`r>;n`pA*Pze&o{xAn%f7D&+#`ak>{R6s-_sjkx(iLH)8gAkO
zoI?NmEl{^7F8b%uyrv&qo8J1|T4%hakREvGX?PqD*D?&Cw*{LM1C?Ga1nu!3VJ`gu
z*QiDt?eX6-m@Y%5r$iQiyNHTWla`Q53<=94i1AJdf0jda;Iv&AnWd^aTbMH#uZ$!m
z{BO&M@`lYB!j10~Nug@c;{2EpxZkCF1pj&01oJ>WrvC?m39%h_pkw+b6dv(wvBG;f
ziZ&ygh%gHqxIkn6(c#BuLd~f@67Gwo-v^+j^u`_&VEkCL18`)+)rflqI2}3ZNirKg
z!i2g<GT^weBliAVplucqu<;x_R*_d`WbJqKi+O;7IE=5y(jOG*cB8XqmQxa|2+dOv
zUaK^mqg5F79pFL6ZS?ICVk_kIvJwvr2zLfuJylfccq0*%=NR+?$uc*t6zJ;VQ~_CK
z)eiw&y+T8QvPx*?f}X&My#>%ewR7fh{y33tdAa?7-9Im5aEl0`>|OzsdivBE@qxZ`
zTQmo`jGmT^$^Qd&CJk9s$8sM7xDw<eXNA!yAzqngUTem`1(`)UK24S?@0P)&NtQgn
z#3uxE5!=ySW7jcN2qkL64gbUCueyau&zS$yB6Hb|s!kn(oJ10lFhneoA><mgXs4El
zwwF4O{;Ws{Rc+FjY%7RfI#<SC*NdPSt-R*)SpIIAIWx}w0-T<RnQIz1pNh^EU^PG;
zG5?olPcV#!=Q+q+NdW1hqpZX?L}b43C|p`E-WBF~fa7trZ5p_>NN}D9+A$zQ31H03
zA7K)}Ii#Vyd{4Z+N5~IzF{tifrN&IW*nH%|+{om0P<u=gMOT3xy0m`L{E+U3?6)yQ
zl<;u5Kxd}Smed<P3jk3UEOw?xbwLPsjVjmK@G~U<=1iYr_W@gf*)+RSw#1mPWtt*z
zsy-6GS;(pJTFPj=D#G(7hx%OYK~tI9<s<!&hX|t|_Y-u41iG{8)vBZ~>o@Aw=&v{B
zt$WdhiEJjK$K()<xNX8*uVjh@v~w9u7jTlY?@VE^NS8Yaa=l1;#4NlYthXrNc~H<|
zSUd}Wxpl@!B_91s@iRO7dL@qkDuAni+E&_s=JYET19V|I!<5O@4MNoEa8CT9zFk4t
zji8~|#f~p64(jT(+ht@U<`d?0s`V~T1LJ@)RMi{WMqFKnlGwr)_IV-XHgn44*yd||
zXCC8HUN%NFwm>J`RmY;9TSh|6IcBe9UN_$oK&eMen-vR-nBX?cIFH!J8NzZkaehz6
zHwvMfXjvT{?vF`J79p63wN1WFi1SIdeH7--rGQ*Qj_I|*C)~nI1Clux9qTbgggcOo
zBeR7EySlF|!w()I+FLLkF}`><U9dADJc3L!RuklsHAZo0Fj=(QS*9GEZ2o_7bd4y+
z1xesWZHK??Wua6ec7}1fh>0Aqx@D@I6hJV&WqNCzrEgb>;n1!NEGVc1I!m4(<K2fv
zmUE1`JjigAzH`6RgDdg-c>r#|b;tDO9l96Grt|gf*5LRcX{utquY);HY=*EHRzokp
zLJ^R8Mg|;IM3~Et)f<N^_#71E+y^;&sK#1lJQ_r^7EeNz7-Jp*ge7BlVeSG5kgC+|
z6$=D6TAa+sJwj4Hn$Z$Zif}3ISy?GBuu>6cqzBDyEv5>|>=@xjD^LCaxX@g}PAklN
zE6)S9r+w-7R{^4xxn)l3BXN{yxj~*2E3C?=Q>007>bsTzidB<4zaNO89{HXIv+P={
zfjGlntkBfL>j-k$*`FX-=n!phf#!hr7VI_l6=E3yeGF0zBAn#efik)zEEY*Pf1yBU
zgY6c_aD#EPAUaYT4>&CS4ZfWvvviXCy5h7F#Hg3O0;jTZuRyBKz9jhpS!fMts)xOH
zPmGFiA=$u-<-c9CNVvU!v});OgGqoAvXgPk2cq1`&~)PCh*f>&WU$N?-Gk*cKS7x*
zT=-$*z;H{T<6_du#*p1X!aP_xzd{T{3X$2Zl5|kGFSn4DI7XPui<4D3?7N}Nn0K+@
zYe(EB$Qfhrh(1j9>Rua=YIh9t>fKMjX@boS8cGpr;t8Xh_>B;%+m@WYZoUqbtJaz3
zQH;+85nyP9<A}IYl=ldzhGV;RAzf|^1h7-4iaV?a<|5&njMLW|0#gD#-*fhpo{V5*
z?x>*7Zgfz7Df(k4L9TVz=v+2nv8xd0f%3~Y%Xbq-_vpq0^7-ZOHt?qVT_VgKofU&E
z^W`Pbm!1N6rUbai#kTf;0<wI_^Qvs@#mb%6cSfzPZE>u=C#Fvt{a(>7Jq@O1B^nz5
zQTpWPT)L__>RUx1Zr-IW)OR-of}^NoF}2)5l#9ws&an6%0p<YK?z)tH`xFo}eF#sd
z4~ir{a~Z2)+tea>x>(qZijd4O42aSV$Eq+lij+U{K*6a@8IXH7%qOfS2Mcqh5CQ3(
z!*Qq}7YNtOmEL;XB?L~$91=M-UJy-(V@1RZXbXfZgn28&dtU)rgp=VweK4K8{YkKM
zl9w2*4quc}c>}YPIlh%JwDYu<$?>7S;eyp7aY`E`d2+jhCm5zdxM{f4H5hM-mT*&J
z*%^%arkj*$G)jiuMY=-F15#)nk`XDZQCEWeFh^<Ep$-DGNEep-3P~?Ndqx3aZmoB<
zW3C``87u#E(8i;#0-YOrMkemqU}h1M!lJC<_O2k>ROiP$sUwPNw@~d_v-qYeC|dQ{
z!hc1YnTS<dR2w=2KYSzLJm_lU>@@Y{>cty7Dw4>gscrF+NQdz@i>Wl8*aVQQX3m%r
zb2jChmaFvhetDBKuahz`h&^ah53TWShagM$<u(B@9-(DO17L<|ODWQs*`craxvzdO
z19cHi6U9Y2ZiRy<`P%#XcL-ueNkoI|+X<m?GjO~+tZ#DCqQTKf{8_Ynxr}O);rczo
z;1zmI-4%Lab4XX1RfTuKHJvQIh0}p4yd^ReCyI6;L=N$v=J>5Zmo}|PID@*P3`dW^
zxK_1FpqIH(uDQ0eO-*KK3HM}TCkkm^zOqb}8$xMH|GOY}T&^z+#?_)7v0C$i9OL2|
zq2^2#UsP=7uPc+{=vUn!t{0S>m#rq(R*+3@6bMJCZIyQhUKh!rfr%7;%OceVF4gJ3
zD{T#8p6baj&cqsVvoPn4bc3-#th5b?%Z&255}S(7A#j^TdABP=3{<$zx4&qo+eHXm
zvG>PW!f0Due#QRV@@+O^_9;E#;DF@OA>p}4jOYSgTqVSL(+Yhtc{^y=ZwSv*dy8~N
z43{RY&O?bM=JOM<Z*5OF^GZ4_2{FyMOrX=AI#riCc7SF$$fsC3;5<`+Gn-o*iwi~O
zdSeht-&%4<5bD&5-Sc<#4XZi6<qa?q4vNeqIqE?n?qTD#k+j(!-3f@~*|ADrx8cr3
zh(0z@b>7G95#$o$5Qg``${=!YkGL0$&L)U+;($p~!(&2nzraIB4e_@l-V^A4*4JcY
z%thZYXw-#5Uh9@9yTC3*dOXg6BSh{3ftbgfHQ99RRAEeC+VjkfS424(?i|R7JMP{U
z$cq9u6&SuB3xHLlb4TOK-S~!8oSPx@e-8pKaD*f0mD6`GB6i}UzZTmFFb}vYfF(gm
z<CLZzf8B%MU0rGaZ~@uV6)Wrs=<33OcJBLVty@#5%h^?DM}s`Iwjh_YnjBy4BEYqk
zwuNzAi~Y-ZC|}c^j}rt@Dm<vAp?<4qs#KMd`kK8Ua_i&FpJ6^}?*d^0mr%K}5F+3z
zp*5Q2&z;=jU=)5R#Km9|^EI8eeZX>eq@~71<{qW1&dhCVejmuZU=Jt<e&qC@eE}z?
zn8BJ}KxgQ0wqFs{{*0vToh{J0*5+V}ui^9uVa^5{E{x;4<8Okfp9hK}-josg1IltH
z?082Ah1C;mNLwF>q{H+icXtc?0BVBCXm)jga<eeEGp;*E;+`BvvOPRTgTwQt!pvB%
zW=XNN?fxL;W;k6qMT~n*FxhcC0o#RlM42sCU~0~K_5eVafroPYb@d{4JP<53%Jf-r
zcm={;Lk3592@~KL+OnsoYVQ~3{1M*RnJ|qa=yq0t&GSl^lwE1tbdXS&gH;BVC9-lH
zA~4T|a!G@YEvGe&_*EGg=N#>QGDD5um0-HXh%E@tvIV+a4SWfua_!s>mPd^=qsw#@
zG*{TC#5FbO%o4doOeQDGcLFj-Td4gC`ZlX|`I|PrUWoaDI}@g~*UKQ1IvHRe<ODkD
zBs+Bx9A5HhAr3r6kaBy+P&}t^iiXaS!G->+8`{O^dd~=!7aj3SVJ;r8*3<oUfR#Y^
z#9j>j^b*B9;V=%+!2y%+JmFwmyUvDxrS#0?cKoiPDWFbB3}*^=EqT&`*U#hH5|Phd
zmfi(=in1`5vsc&CQA-OnGrAD3`(t$hW;oZkDzU$43d`xd_Czw<Jhu$eg{^R0C)$}~
ziKZps9U*BJLH)R9pHq<F!g?TS>i|LSZP*%d(4*Uq9dWQ=Cxn|Bgu%E#h<Q3;ZWaG1
z(;KTdi8ve!^%q%apQR1Zl7bkwST`xhnn)+a8=W4hmkFbkh-^J^y9n1ChxI-2SQ#O?
z`IU!--3JI`W?`y;ynBNHQb3HsRqh~!r?sKlK+F_LWk`Lzri+=ei4eEDuBPjC)J+AE
zz=m442Sm7WB`cyu*EsBq8-%;;vNb>}LC(U=!D`Qh^Segen3G_#kv>go4rZK2+$P*v
zle6xX&kI#+&2+!;PKC;YjcXP0Kp6}3u3oYA89{EMs;};?Q3o+ohI-gY$I?Ru!0`=N
zZ+c2#pw`pd6Hf|YRzrD-BEIz?hsR}-U%TB>fa@!x<TRwW73O^9(}u3G?=XnDWzz)2
zNtN*P2Dpb@zb<v9qpP2d08%q9-X|9jXNz{A+F&oeNLt*B1Ulbd-Qhu%A8!kD?W$~_
zI877af@rrwko87E%v)>*F&ASeE6A-gjy!_cpY%3I31;?~f%n7n^zGg<f_tu;L~31;
zg=s854wZW==VmnJK&&pz88fv?#F!$)8RH!;-3}fXL=N?i(O7&A-)>Fr)^frpnyQf6
zV28%UHUgXr`%R`ShTt!9AO<V9Fge@3SODdengoXpw~BNrWG;m9>#IV%L@<^cI(8<V
zuQfrPd0!p>xyy^DV6AP_<Jq}<lP^bi$#Jq4nr|EoM&X8<x_yn{kbpc`@*Zt1g>j2>
zid|k=+*r8lfe6YRMfC3qa|0tkup`k0>zxI<@Y*JDsq7XZ6caZfsq?IO2sD*Jmo;4H
z_BwUJPP9!Zp*R9UQC>6cbR>Q<+is%}8PRdva6BNy9fu+@d2D$ofJ>S5hw7`p5av>H
z+KGHCaK#@2lAEmKw65$b+9{!gM~pj1fHOn%lMwcBjzX*9o_{Fb6m8~l<AY0O&3Ic7
zGmVrt@ot%dj@b~SHad)-X`hyvt+D>$e7mIBt&XSZbR%JIWvqF0qg6YPJ&!0*WCpy8
zBpaB7k`<>>EitYj6YW@)<~fclV=+&$dBujKBhD*Py;{qOrA}kkI<km9h`Ccg+rI^V
z1VoR-HWNpzv9w5Mpjne;Hdf0asx>sM9nmV%T()_FCgY6-rb6=Y$G8afPTu)FP2XmB
z6;Cw7ah3qH8xdIQl52%HOU#<mxOr2M^FXB!PbG4GT!zybI`3!4qDK*qY;lZ$np%WY
zw)>Viszj>EbUBv$F=5;oGCE2-=5>X6aB2G#w~40s?d|RHiN2kt{GZbDUp?BPCrs#f
z6Ak~c@{oq;L6NR6j?i(|`I-P%1)DKW#l)`Tj{$W1%VSt&DA`0L)QsSIPI@2s1O+ri
zYwIPixK)sgL<dJV7q19$7Ob54-sn=t0>T>JMCpiSMK}ZwSn9)7jf}koPGpDLfl<{Z
zC|4J6Y^W2CEMsEc#;GX^l(@Bw$&nI^Vtd?^W9GX!G}l5uaRG3k*dL3FATMNIYLS#k
z9X|!ioWhc&Cx%Oag8~c!-YNlj$)@fT0Td525^ZCY$#Eb|^l0P8aTO}ct3V$+^Sgcq
z2@Cq<r2K<Sbc`bR{_J>uG5?ttSXH&I$3KLcArdGwjP4TTr6)bh3Ey-n{oXQuYCko7
ze=K|gfn6gcaDdGPOvsnaW0?1*jP@5!50kgOrGCkGDkSd(|5Zfh36te;+$6vS&09CQ
zT)0)J1{mf8|EF*FUWCx0IOxP837jTVtA_-T0YkYfe)%g1rzgiKGI^l`&3{sX%D7N^
zc{XPY2zA|c?IuY-B=Kdz=2Anw4XAG(DDO1>Dk%3jnvPYHx9T1c=;~GaRJM0d2B2*C
zsThuDe+?lyrVAsZSpHNfmz2|c^vj8J-9iZ?FPl8mBnm#X%xaC*PUD-T&=_&WO9UN-
zTMaCLa%Z-ikcm~T!PB6O!tb75VA{az%xPXs{#Vl_p)D0I*SH7kb0ok7xq8e+TxFzy
zv27U+AvgKi_(_=s$|G^B62B2fsi#kw8t?1d9HGacd06m_A}`i8ytfU;qJkWz%CR*^
zy(NH=7i|2hfV>#u7A{$kE-B0<)UkwR!Cwe;XPGNIy8o1Phlc5@XYxBmlgIbJ=^Iuf
z9*xY6+eA4g4?E(}LNtZafM#{UDj`R0;nnlc^zHKG(?Q(up5^3NRM(sp7mD&kqsKmJ
z{YB3PF_YPyr)upgh~Yd?o1=U0Tb=`HJ`5lg@IF8ZaLI$=v&^}UKpxfZq1Z_z5p?e$
z+36iAfIh;-@Ui&Hc@Pu<-v#Irq{f{g#2KJu;<%{)0stzVPUn~nR=&_tNS~!p^TQ@W
zGMCwT`Q5>51(Hple2rcw!u=R033Q&c>AUT>psp?Vu<89c!xLntq2wReEE$i?gW;D)
z$IsFH9pPvro?3}Z^zFpCF)d|}d`Vp=)QRib2*PN*SO(N)DW6GfaS;LC3XFO#)x`IO
z(C7>XP8n9YxcHsLQrESAF2osD)8-TlGESz1sz=Y59JgHJ6r?92Ur{a!jhKYOl`eJQ
zHau}G{d>OQRI+>AUj*D9$K*+~%Hg8T5abSSpm1`z@?{0aMBHDO1$MVEnl0^r5O?EB
z0Tic=UVMYTT>uu)E<KO3>E%GV)=1&Vt?feGxtqL685fJ5n73(XPB4-f{uklyAaiT@
zs*Gm@z%HrdhvP?ofFj?%N^d;;N4|3d>w-CQ?k}$ZQf1K+#iJt3NCY+v%#wcQdsh~*
zYWPypgfI`pp#s~ViXm4(7Ngv8+e7hO0q&>Ijl)D!F8UDWSS$S*igBbH*JvHr6zN)U
zI#6ev2*^E;>7!G6woYM`Uzc`ST(!hE1UcJ243wBaA1;w@`6}mn$X^KKg76F{9^yos
z4RXVyW?knhAh^`pIx|kzw}aE?2V%p&7Qtl)ErIGMg5VlkNY2MZc(~CdO=;+wG5R+^
zuTz{F6o*tV2sVdNL!lH;=8wnyy+GLs(j$?wF%A~$B=C!X?=w57`<Gy6hq@ReJBb_{
zUJd9fA-Sjbhg$`iGh;pJ+Tw24fSAV(48w+_Ccr`H-)uAD7$MX>+snr}qG_dMB+!g=
zu7#W^emW`F?&YTIK(XRDs573@CvBV*vhji_Qe!?&^}OkNXx9j<4qg>NRYv?bZYW@w
zlDrqGO%j|ZY;GwcUt!#+pcbYE2YTX1H$u8bgEedn;xYlQQKQ0q*%M3M1ma<%>7uyZ
z`6uCSWqD(Y({6!Q517>!m+CvWC=vli$|wwPxfRe!OIJ-V_~vZ|(!kd=UEz-<!j4!^
zVLm9tNwJ;cx-mBs;&XxI)x!0QHE-uT_j81y3e!hyEX*gBG8@Pa|5?wztsk5kSFv@7
z&!+zXfo^*a`ZX!UlK&_m-1ea(>5HZy@|?*70mqa-v&LXd-yb)Y5Er(RhJLv_iQ&d(
zG14xWp%UK`N;=c0w!{f{@ogSXosR3|3q?~Be1m9Xt46#pL#$46rl*x);kyY%AL&RF
z><*$_dYnC^1BG1$IawTq*5e<d$#Mq%iascV%L9qdi#{tMbU|3ku$8%o7-lLtavlut
z5@z0U^h9@;ka9v9P%}5v1qRu|Tp|g0Qr+KOLYxMtik)(Akp*WzXc3r<t}HQb9y0Zi
z+Bx?Vj>_1~rahh$?P82_JJ2fduRj22R^!%J!jfd&4TVzgPF$7Tpl`-lZ&y67Pxy!O
zhEvCx|AO+EQUf7_rtA363wAcRiICv{-QvX(%y%YW!Ak;N-HysAra*)7h9GzK_PXAS
ziVuY_?sOPSs~++oq+1BvTFz8T8XJe<`07La?%dfb=YzOu0?8=PEfP*I6XbMIk<W=m
zABJXL?XO|LbAkx!saKKu<8o0jqh)4WocoBsr!jseJ?8H=-o$Rz_rwroS>b`FjEJX5
zW3)$)5KkF;LAMe|32+%OdB@mBwEK?rPv~$z73TI4lat?vM$CE~)U{-^%aH)nL;@Y1
z`KH3-MstnjJh@EL9UCd81JO4JlF|Pr!eH?@)^C_yb&7PeG!V};yIdp4r9coH!XFds
zQLOv~s4I;@U!C>s8X=CFRzT#lTgpI<p{APfb|GZPkddAC{7)8fWPhJd(*(hP<=zu(
z{U4Nz%s6F0vB>>Yn1lyb>}Y=q!lg@RfHqk>Tc}!8-i)smp<cw|0~bCONt4dRIm%+s
z@C}Q2y@)YQJ53Rg+Iv=Pqi-k8(@G=FM!IEsqR_lP$`v3U9nn2Sf?ZI$ZkqAqBn5R#
z(U*9|Lp$qvK)n!yWiDnVXDXz5hs2LZ%0?{xED)vU(PZ>1(JInuF&D^Oo83>Fz#OeU
zh!1gGj|+0657p46#Q6f;P-xy|y>!HLAaJ;kn-J}BwW!1@jvc=9JcKKUf{t@0?uO9V
zHNj3)cJQ?r6_C3La;I8$(HEUgJ9h=+ED`SAIF#;=`Cfun6R?rvL4fT=QYI$&POh3E
zKD{j@vrqH2=CF5iBqs5}&h}XK<syg<@sf-+Nti1K&t&1Vvk>!%cQ@0ujz;WL2J1q{
z&sun(5Necn^TTn!05>+Gl%5ArrT!;$q8_}?_s%QOE+#dx5LbT{j4spKIwOvr_-?_t
zNZ*djZn_>nd<`15S7@ewt~wRJ4x!$Rk3?q{qk`w|&FcVe3i+)=sglkc?iQT;fu1)?
z=5Y6wpk7fONjs5s-Y8NWulHjsaj5`Wy1R$dw?Acq1I2pWl8IR}Sf-yBk#nHjx*E#6
z;co(FMx)Eai&;V(46ltkzoQ@TtYGFMSM`{PlJ;cp5@njkpK=m@zxHp!<{1#H$k<S{
z<I1Bd2Mt&y#B`y~7_AK^M)8^uXN*>safKQ&THh*yrPAO7MTnaUQMijUP2IEtqY1dd
zijB>3Z=2!-lyic~qMWSW8NtahJC4l+I(HfEvL<iv4#-41aXcG$3dr%aaBE1%PyPd#
zcTMOzSdC=u+8%!>ab{7}w!F(BlgEV9G`v1ki+dEvgG=^U5?=l(NYm1+miW9(1C!-J
zJWDVwQ;^1cML~F02`A+gC7J?VTy7L{v+523j6|LM(Crp^ACipGsJF#oBFr)bDH+sr
zY*S)Q86q(_V(tfo%Wc9A4)#AQgtla5(^*?A@FBFTz#5o@!_@@2BvPzl*D8Rlxw}>8
zY7>MTpZ1fTwvrhU>WpwQq>1H6LNqI~-Jo<wi*$*RV3-s+vN-3zV5G}sN8VNv;Q*`*
z+*D!)UiyDvZZVw_;Gv%r$z=q(0H`LCf)80Jp)NqX4EAxOq;Reoo9N_Cbw5FriFs-g
z4<QbUbeusheKA|UDa1W}yfVhA&&Nd?4ZT!CMJj?Eq<0i|7te`yA}H0gRsF^%1p<>L
zuWum49A%@z-hKa10bDWmXz7w8FI|0B#6uR5PT(yeo>_FDfIHPR9-ChPn^m0uyI4mG
z%bgq0ijSWN%&o#v3Kd-HOS>eN>rKV@g9!TS4E!%&sqYDHRk~$u_=O<H=I)A=5G&3H
z;(GYuvK<BFnW55+Hz(1#m07aVTWf>!7Z}dFJM|PwoGr|`Ve5c08~slSC69QBqjw8(
z0fwYQ?1&W?DC5z8ISG$TONiruWVE~Yf`powpJbAbexM^pz6_c>7}AMOZsQVROfwv^
zvjzI6D92+C(HXmpl%Fr-F|5*Uig-m>9yPKxhg+|f!FZZ?6p;=02MZO!*y6HG55)!x
z1Gz~jZhlDCXcMd@hr=BnXB1_&Vn5UwM`apq076|Y9v5xKj5L$qui-^N=-Jr6;0P$@
zUldZa6fVb1xLWBr0DiM8x+4eS7@-pHh%j%kDw3$m(ehWo+@ri%(K!EsK+0JkZ*c4M
zZc*+J9I-IxwtN+U9#uz9p0XI<&X$F#o?p+{NvLzuEtmA>bHcox)n!ns(D1{^M<}r8
zyw+HLaR@U7!?M0OOtdp+a3C(;EWk;RaB!PEHNGIs4TZDo^<^&fm9GI6dNO9$WFk?E
z>B1eI>k_&P!d7&NBCLFtaueMbfnOJ%c{A*1@P^~kB>|mRlKYqhZxS>yK493BqKT*d
zGTRL=x}Ri_S!tA852NyFAJdj1B%Ew(X^C_74bz!TP;sojG}J^}4KmqbI&iJ9iMGN{
zBOOg7(7mLARt~HGgt&xs*$(;Dct|mGPaf@0CwluYQ)EC#&|!La^a@iE>O3nF?}_p}
z+FPr}0?Ycxsp%fZCZe2e^4G^`7vkX6@y@t7Ae|-4YTDbpxm*!v>eSRq?+M9Wiz%`%
zzO}p);o2>Vk~mVNYs{+=G6$hgufIY8rkhboy4p~nM;v`iI!iioBfc$oqL`@jar7mG
zI&y@9%yC8jUG6NkIqrx*2{Ge)B*JSv+$YS5vwYT(Pk0ieBc`qhoM#fYr_8*vZQ4et
zD=yc3GRiqwkkgbblYBFtEsS2TTP$&783!xvbbR=%Aa^RRl2_ySm7vWkR4Dk`zd?X`
z!;X*VSTxnmS{abqG5X==?V{*jv$$RSy1sKiv0u?0Rsl(~2@~U394#{UIGyKe8*rj9
z_0OrTvG%Hbdql}VHC?usNOOzuG<>Jcj6GI^aH%W0oOg_9v#3WZCONAdiF1WIE5s#)
zD$07Eu-t3uf6}jBA;>u<pYao(RqiUIv%-0-J}7LWrzDw{ruaahbH`vuzCSeOUG)u6
z%7~*o3_Vc{uK^9uP+}@mQB<c-7fIh*69T3oYv@sCl3ZkxBXghv7e=C6G)$Bx8g1=y
zlXC=VU}V)1u@;ESFjDD7xYWdK)EtlwKP~v1E2OK?&Qd1B3w+<q6P<vtMe2o%g!&GB
zqlx5|6i;b0zWCb3FWxbtytet)0s6(svs$L*Y$X2pO+Xr$s}Vf4#J5|3;o88pZk#7b
z?a#wCm#@b+HAU^NQMkbRkgfqTl?__f6b2`m<HmbqwkT)FdWQy5(%=aKNwE)G|MkDc
zH>^gBQHvcngrHS)BB-Y}ju!3y%B@<QPsK&^AgIikuf9CRambHN$3Pq*I%`akjj{IA
z4gk4v(N=zNVHqk2-b9`UgL5d{8R*|tI2T!mXuve_wgM>`TH|CrwvT9cOWgWvlD7lC
zM}+4doR69pZz~|_0o8OH333_Pav;;QKS?dOm41L5J$)=8JBrB5UUQhUy4?i026!&<
zv~-g&&$yFVK>kt2VTMPi^?(2m#;G$<%)Tkw1!V7*o{jlXkhv^pBt4z+<7q`89Cjr;
zrPl;G9r%triJ8zY97hry($AU&KtV8Blt{U=$lP=o$S`Vl5#m9h-KM%L;?OZdJ+1ce
z#2y2=gk4qkB67Xg;h=69+i7qG%Aiwafp0X>KjFyiSf%RHaRC;$7{4qt$7z-h8+3!|
zDxqeE-0SetP~1}nOnMgVCs>1?5bE0C%4m$8-E(DBt{`xPHXLsYa{2VGFcT`KE&Fd&
z#Dvx{?!iJ_Vx&yj!pbCW;f+CEV)=!t#xw!BY&yOp%nO1v5_!mJ!A<#giL`ZNR3#+M
z!H7Ov>t}ZhMpm36lHYiyyC*jOHiT1_t2O*p#iE-5F=%<WMng!H%fbaBdjIn!fbE<H
z-FFD#+;wb#UFQlyT={`^wCY@!?~l_2x)t$3&0dh|@AO?Tcci`w_XPJ5t<`F%PItIe
zl*?V^rp%)vbB{;*O3J=zn}eBu+Q!atKbj%bWk&0P((Ne$<fHAR*mKPm(BvaE)!cYk
zl)Ig_W!Z;dJS#XCj?H%Lyd^ZL%F2t=*^$_BD*$J%Ys(B;O&JxT<|r$$uARlN1bKpB
zlZKKrixxlK8r-vXw##Pdr!N1resNuC1>actTo^6h+A=*(*@kb|5GO-udw#Gj0FBy>
z@;WXM;gWMnzB^tP?G|ETk<NJ0?LeG9lV69PM~k__TyRzpMg>pR2yzSb)y8AP?ID$k
zY}C)p1Uot2O%_28T0fiU!JF~l9YDllbj|hHbVn$cSK&M2-vV3!j(WT5@<_YnPGD{a
ztSq$L$&b8Y_a<gw<`1cPQdp8tPrSxbis>NQ_%WQ>E+x>lMyA4jijh1&u(i!@371u5
z*ISoGxtJHL3(BQq8BfKSD~ueG8<^6L6lKm#LYUfqXGl0B|4H%X*+pzA%F-5ghOpcO
zddaIjb}U25hy=@r!Prfh*@TB(c4!xu(RgNFM^jG-AtyvYY+^)E3ZzwCFSu~mB9nd$
zAQ5NR-OT}55V^t{+lX}f<H_aeuLQVQGA$;JKb8n`r%XMc7GnNr8q*oT;P-&cNW6n1
zfe^VFrwa8-#>7z9`GxKl<{z9wzqY?*b48hgkAaiK^MFv*r<q>hSz~uddVEXU<mlG7
z(_r~yq&4E{{Bu`xG-mJNn0n`#W1$N~x`5IiCVf#Y9@rDq<AR&0^>|J+yzFBG7RT)c
z0dF`GVl*&GEWEb?cvt0djL|q<fTOcy)#E|YYV-k`qIDk=EEG7r5#j#B?OkbE+v9X$
zt^fibEr!EBLAjKywR(z?e!J%PK|Nbkd$0#eibC3<A>3`HZr0w22ZT5q46vE_8HrB|
zqnmak()7f;WgLdR*5dws`JF<e#2SpH_JeSZSjfg=RJ6+|lNE{PhY50*(VIm>G2i}<
z!#I|p5~ql!xN;rK{Ku-zn#E|oOSmH<YXjUTfReGBMqReg0Z=Y`vQ|)D7YcK6w0VkE
z4}{i?H*;!iuWwHdoZPYlf&~b2R|tkF>FO>6b0n_J2+h{dgB;BMe=8AQk{iG|;=@AB
zcof;(Zc1AE11nC8n6#N@sf&cTkTUb(CRj5b5~fZ$b#m-b<=X{jHLpZXwCAC8NQhih
z<Hi$U?zy#Uk6seEA;;tW&LqBnD7t9K_I}v~Ufls^{&nHS0cW)(fMSKixm{|6Aa`AK
z8^#Q=FVmD?qC`9`8iuo<;!Ni~QF$7bR#8Uh@u9$pX{<Gl?u2v~!@n?>BVzZi0?`d!
zim{InmsLK<2Bf*_6$VdqJn?gVn<vb~wRkxHHiU`tWBN8L=3;PG??SK9T?9`r-Axfq
z|Ku4qq+Su`0S!!tJvE1xh6^Ud<7I4|Bn(Dh4-s5}bWdkE*6amwaoPCTHSly{Fs8j5
zyV7{D48i1tVwFm+(MJeZqOXpv!GLH=HC2a_U+Z^ttcek-w-@D(iR@r2PBwlI4*<K1
zv2&2;9SH{s&07U#g@&&(=Fl!TcHQlmNu`i(F4V9ZSqz$|l&CAi<XXdyt|QhP1anl>
z3+kbpl*m96b!kV`gg74Bdo;>~Y6x>%=uIjOou8Eu`)i=V*t||CR|9n*ClblrtuB=I
zo{q1}mV=74k;>R+P^Q_!VAqID5nd88(Z^=Smxdj;tz~k2Pv886@=^W)cWXd;0g*tJ
zJUJ(W%|GdfyuQ<MSyt!^eIo^ku#Tot!a@Sw8u$g4sFoI%1iP}M_-&VqiUj2b(tOWu
zj&-SDiJF_Nz~a-O5T}TT0rE+g4dXb$^bq99<W2IQWkgf~Rm!*FSdmmOS13vR1R=yp
zuDj@EH;ATMvdgZs-$~oP=LNc`LwZ`Cc3FH*krItUuZ<m=AoSq=<ivPaQLajDj=XuM
zIv*g^39+?H$1&}K%ttOlk8rV?&hc4Ji_y(fxEstRh<ko*E|<l6Mbp%6GbhJY`gV#G
ziz73xWE>1cOJi_95MLLOJ4^Bgjp#ERs|t1J!?~HfUw6eUVNMA>$(W$M1)0x?ygCX>
zBIw70-5>~xdj4jgLqN2SuoJ|klPIU2?B=ACxk^}0ol{TkvR@M9nFNuXzV?m)v!Y+G
zhhxCkh}xk=C{{6+mzM=lBW54@)6UZQ+J}KUeNa~D!Pr3%Y|9=l;*X*!Bq!mRUL1co
zlyhg(&!Rbe1b|D;ks|3bW1KC_sY)LsC%)&F!4M;CclmmWr5C_t>N@#IqI!tQc}TlR
zr!J#f?v&!2r^M1j(_;BrKPqz6qa<<$IsGUgH$P1}5~qtc`(dQCWvd=tL`wHUNoPQi
zXEk>GGC#31gx7@AcV@Lti9L@g5|WQGTIpgT?kx=e9*p)ifHyuC)a@dtVYn~ilGP>x
z9aMa&4Rfr_wf-_FuEXgL;%oxVMXp}bZfrgj&ONK}+loMzV2709g{J*S6w-X`<J9K1
zpFq;N2Xz6r?KlW?f=vfzb}{K^Am&x47SnFgYV+|57T+I<BCQ$Il4kqT<00Jz*ecSs
z#!di$IV~6>|3u#tlCu;N)``^OH2vVBPnh1S;hqv)Fo|~|q)T{4V6Kn$qSWVw65>83
z4!GLmRY5MtU=J_e#>zhj$Spi7*L`tziJ(~0t;P?8IBA^?D7U?T0p!Zd%~t$IbZ#OH
zmGUi$M}<zb2AUzHvCjp$bLt{byDT4ve+lOL;22YiIuxaM3iO&gE0#PF!o`+*o1VBv
zfQ!xD2OYrPBt$dS^l9;tzTJ@hjnUZsWJni|rGg_-Of`n%aG}n%E4>Ya&(}Dj42o#T
z8k96%33Z`)PDOn+hjiU^RZ7>oPF7?O1SuPF@@<Q#-wMwIl4&;Wq3#w&5oN&9-HbOy
zx;}_L*fSY%`q(hjc!RO>DMdQ+KisHt%KV`)bA~tR8A0p*+6f~D_V$cMkI39_`zz!0
zio=Asx3O{K{36|W;fHf_Oaw}8>ibTeCuKGRaioBWGRkHmZW2P5ml_S3<uoW)GbxXB
z8gSkjU}k-z#?;Fp6&-l(GmDhQm@$a^O(Cu=8=howxo#PV;m*5BF|!P$rxkP)Yo9U@
zSA2)jJUk@C48iW2j<ow(Mbap!(%Sf(Acq;iZDTXW&MtrvJYujFKL@}gPbwkiP%cDa
z2qxGOIY4A@d%ggA-3&ybzvS;$e2)J$e@~mz7T-9JFy@=CQqy;i5~8WXFU|f2%5~?J
zmn5A%A;_#m4}vc}+T%%K?yp*pcpn3Oz#9U!@W>%H<9o65%Rs0Y^^Vm@Y$}lKMsT87
ziLFF2usABMM^l9J)_rw+-Nef!gn$pDd*Yr890lv_@%TuDhkjRmY$Vpc5SpHx+>eYz
zRitany#rQ_`2BA|^3vYKr!G5exeR?raOR~vOh@PM09+Q`qUdKmdiElq|HsvNfLT@)
zZy!N23IYWZCF~-ZjkCK%m7E1!B;%c#J2Q9E9p+BhJ0uAr8AL=71VIszAW8;NlpvWu
zIjZCwM9D$E-`j7`IpfE(56|>_yZdxkS9e!eS5@aZs^DIjp7irvPyx!uL#DR`-4<v@
z8y*054B77sb(OqSs@tYGQc#}Dib*Jm;q0<T*&ynC-Wz-VmW*y+HxgPi1_Zb=Y!&f6
zXpfN+N^wtE8P~tGC{`f`HvA_5f?#o_U6%2<Ux=2fsWakj{cURY$Rg&6%ZfDd3M#y>
zL)uH(Bs70>m2@}R36p}H_TT<HXTk|v@2kJ2q`OGmEMJJNcLkUS0m&*k*J~!OFW5^K
zf{b)C$ie8USN;RI`^q#+0atet<QiDGWq^{RxBl|VA}1@Rd_m(6f@rhjWW_5=`Wq?e
z0}A0(a+#~Z%_*bUbwJ_Y2{Fl&5p=wMH2{>FI%jfBy@tQtq&|$J(BQ>WB}B`)%^Qqm
zuO*iWG{7AILX8}6HWBJIPeHHHwzMk1BaeORhs>K*uG!e!B<0*H*bRXlco+WJ>j0QS
zy)~JD{Z}auT8-Da;&Y`<a7mL%n0)?vF!#>a46)!1AZBJi8A3Z^x<I!#5hq2&hz_BA
z4)#XHzUz&o_*}PQuF$%pm4+(QXUvLk{E@#++`7C1(wPUA*!d=Ke+;`k77JPHohopg
zF*q7(#W+ov$vTRuj+o+#G8>YXZf7cSwJ>)F4nc;AMg9b0B5)=fiS9p>HqD6ehP~6J
z0$ejvK2hWFNV&Nv!!-$SyDIHg!c%f(<@4tWq<j6X*s}eFzhNcWBGK5@|4Ndzi^r0B
z<5VRwOLG^2;M7{9&L8DGsC?(e-;@xkFc5#{7XEg7bmh*CODu9LklT!|5_{zk;0Ez}
ztnM*Upl>74ih5}_v0LT5n8eKqzp#;Nb;33@EzH}N6{XOpY2RrR=oYYBCF8Rn38F{!
z5v*G-xs8O0tT+WU@V`P_0{vhzoA~DK1tb-#7KuIo2IN+AY{kk$QdQhlRK{dc@SX1q
zp)!O+-atn8#MASFAON=YLkXDgZh*Xw_G8n;W6f+>3H1k10_lk*mYIc=723`qCihbM
z+wH_NW<&v`Ly*hp@>J@<YOEp<#_p=?hHcBT6bj39G~8cMUKEf{c+&WKA$*48aC}l0
zz=w{0@?!BlMFH8qa}M2Ch}+^<0NUb2L2e6yv$Bo+g97s;hx>T?L4a2YPoqw$I9ZpT
zzLy`wb~SWj8{JnFYbWe+h4*Fv^7pV8Rmax5A1F_L&WFg*o0kZY^>s)1st|uZy;c^x
zJx~<V=a~tI34vNEetul4lxHuV5csvB^t(xzhf7hpvV4_LcXt#ID1~RpSzVkbu)ETu
zs((^(H(fCRkg4L!4+6Pt2cfTc-<ExDX@Sl+s>=p;>oo*<_Gx$r70d@IWsjUQpWd|U
z93lwnReHD`IjyWoiV_U><3d4RNc1{Z+^BS3ll2Y*kELjHcto&i$^{_<7OOr~Agp|H
z8oyEqpPD)w$Bz$`Fo7LD_fr94*eTOvgNKVBvFIe+(_8_xLff>Izr#6=94)yc9#)#M
z;C_J(<!z<-ke*_TQy(GWmJMSggxGtL5P!OXx|^FE0eM82v<maqi2pna=)q$qk3UYg
z?PH)Uc~a5R$?KJLRoJ6&M7djlzovbVL+{kbL1?;+n6bBx-Ib<s91=#OO9_9uI!}+E
zYl%HXpm{+row6Pb8uLz~_kjCRVW*%IDX+(Vs8u2EYR;OWIxesXd54uX@+Ki~BhkNO
zg*jUv5%c9~aMA>bC0lQ-og&EJ?W^kpV_ree7M{kEM|K}B%PF>l`mo58MeVFC2=;gm
z2{I>fpHYvSl+GN$YR-c@0-$Z8R3+AViX>yIpl<c}sZypCK~&|=!`A-3Kvy}y%Zahc
z)1*x~CMlyxtxpQ`h$v1iTH!7J4no@rh{-zn4-)3*jtZh_RwL~rI8z59LS5=FAvr?m
z`$K}LiKsP$af1>xzhgL!%(nj|>9+9n6!I9e{5#Kp(WMT=Wg^BX<!_*6OjG{aoPjMG
zGpoCm5<wWA4xXS=(gdTa(BusP>GSZ_e$N)Q$=Mc#SRC@4%l8xA88yJMO8R?D+^XUx
zJ|xHl?86^2J}a{*PKnGB8_$zPn>0r5ztrEfp9#Ww>j|Y?xV6&7dkad_TqfmkZ(j_&
zNRkDcm5(=mYv~A-7VH!B1GV(<?)X^zRVhlbLMmwJy-J$l*gS0*qyxzpUjp}Z96*mv
z({zGeP^h_rHH}V5pZFDlu0ks{&nDa1<Wk|qWW_K4vkb`fHF>)(`YJgfC}usiSS%;n
zhY>|zR3fi$(%ZMigRg<mZXUYo=l=RhB_Rfa4tHPe@pq*&zozG3#6n+vqXZ?XaJn9h
zE`eqg9wy<knbP+NGgo3rhT%X>j|dtbUoMC`mV6FVUN;Ta!+taFQ|ZjSM5xE2GoBaZ
zu^DCiZ+VksUZ!w0<%@?1VVI}QnH5LtZwQr6vi+>?7YlF`p}jT}%vjzkF`|j48;Q4t
zK?HVo*<8djZ#xY~^lNoYX|3_-)mAEh>*Wo@Sm9qq-nQfgue6(pdxBn#IZX)7@@*#O
zF4qb$l@t|&DbLbTji3Gp*i_<O$X*7QshU9NO-z9u5sYcrUm4{Nibe1ol^&O|Fgx~f
zXr*^SoeyU2f?&G}F_CrP(J}X~5@I&<AYM;ADadqh^bE#_O2S|q?<+WZkT5?Y8Zo>-
z65uK^V@b-b8w8CrI8xe>;@8ah0MHW;qgS5p`LIZ1jM*9QEA5IqP>$52>mz{NRur+i
z+uPw|AdAiIIyNzNSfs^!`|6j>J#Ac#LHVj6w?YpB4yA)g0?ibdQede}*$!nimC<|l
za@vyPNYn(6XC+jOczHya>8#k6ImZ6a%AB-(BrQ_Qeo}lL^)yGoc35(nK+~Y7A6fi^
zPm4@CswJDuIN-m4?uy<WV%dl*KL>&~tYFf0RN{RBv<;VX#$O)(ow4eCyNy@QPQw+W
z(u^V}C7sIF`U1(^bT(po2tlGns5?0#e~*5i*_ePe(A6LkS7R&Xg9oN>jOzW$vQUF-
zGLE>talf$KYK+>n(Ofw{(D;01Lv6(4LbNQ;m=UKg;2PO_2jem&JXTto{5nn77XeKz
z4EKiPG^L$s{57p=d}%>IddcG%l5fY9<O_W~)qR=%raXMER}<q3r9I9H??s@yB)L4A
zvq>z&v?Fd@h-`GVK||tRCEN$OnIW1zoe?SVIN$Tg&EpG`$6ePcQe*L%AQL;C!EGhB
z_adO~5CQjl;s~Wpd|7a$_g~%@<_}^C-p4$MMM1nXh~cVz3OYqYD#F>3lBI2&t&|(V
zomMM*R-7xyGm4!Lf56MiJiY8w;)=<Om3dfmi9Ia9ZA9nZ8Lueqo}f~Z7EE?U?+Vqz
zJ9}FEU~&F-FS5r?d)z6^t+DYwR_jXuxivh>z+wI)0b&N+?brU&_}?w@sQxwu&^K~~
z!=(bm0|y(;lU<5THJ%Y{e(0-77?g2Q0{vxP5Fz+_hMbQ%&%UM;c*^JIw074qY=1+w
zxga;=aAjL8yJP`L6)HQojZ287H^n5|OOR_K{s=|}R|<fH{Z;NQDD#}s?$D4PR<Fc<
zOF0MQDa}<3DD4Gz(zIDIV`-A6IqJ@X<o?9sF;l3&gC))&(XIrzUSwl!4FAYkWJH_R
zgck(3Q;n*(<NcN`sv$}QmwQJGa5ar#c^Ag@LQF8-P3r}LzPLULkS&%IjWm%7bW7Qw
zy6KC?U_#ckZ6VF$l&$i)DMS>gq3nN9Q2uhYA?fmQA$e4i26xVKMHWQ+G2&Qv31NC-
zDN76*CE%esX!CYlp1=J87SeRc>@6V_naRWHbwQMvHFH)R{#E`q=W3Udt}B*Wp{NCe
zP^=HQ&{`3QiH4Lyn3mgCBIW5geuKu*hWB4r1~-#-5N=^y9hhK$gUiR{G!oaBSz+4-
z@zNA^BpwuOVlu#4c^a__s3+toHd*mzna0|uCl*|l-{GD5yrj#Stz`Zl?c>0Q+Dw?~
zthadBj2Hb{Q7O8&j@V%}(xx3M_M|qc3UX`YFjQ+K``EKq2Nl7mOpYbi;BQyU(Lm9?
zHWcFFgIKk`m|Ftv_^<TGQvw*h+SphvHeQp2`HZuPylfbNy#!`9=b`j!>@UDQN)It>
zeH|87tceXYZ4!kEha=p?f*psWSynbwna9F)mjHLPufqI=3pbur)ZfP$L=P0kkA%5v
zIDklFdSQuRyeG5moz@}`gs<@A6^G76)+U)*7rzg#<OR?c^ln&-4#bpooMxXHtl(_5
zjneLYKiADL8{6o<0^L5YmU?Jm^a!H=Wju?aIJPV+KM{gx(zoxfTU3bE5Z)x3p$Rng
zSg+cW(a=81Xp9Grt)4%k#5BEi4vlk4Offm7%a6x|xtjw-I>k<LJpiv=1ASbAD`^o;
zM_w*BI|;5RdF%xJlE+53fiEa+8snX)2{31U5O<zKRDGb)jAb?e^5@Xx5Cm&|0Zc{&
zEo_FbRtk1vAMTH9%Dmj=!401ZplRKZjsY9qkfcAzTkkR}dS?oVwxH>A;ww{K6_3$3
zVg)5kA-4TilpNLAti)Kn4jPulT*5qtsBkgzlWH@8UI|e3Gjo#jOth;!ADh`SIePWC
zX~DGWiFx_AX+#AY)8B+(8R?EMP2*1!ggcXDy0eKOccO`BDVl*D%S`OhDF{zpkekmM
zZ^Oo4mDn(W@Cn0&`;suI+u)KS`JpX6y(lO{pmx~}P7>xmaxE>N6=;2gK#zK7|8RV)
zq+7*eh&C@K%>Z%J(`!R8z?w6G+$t<2y$$au%-Q5Tr-Q(yf*4b7t?RLD3kg?@YLp1G
zv6c{X1d5NgS`3tc0;SjEQUR`$MKm1`mYL-`<v$;tO3?P!nl3S7o|1Vi2q_UyDxEnT
zJtPq#X9Ktmqikg`>LNnCHa42Pw$4ogi^e3Od0JHvp7eI(WMSq7oOswLY^0cASBcub
z4b$Q2%IH>1n}Q3*+_Eywy`*UA5N4te^6FD_4oMG815J{n>ps7+^Kh-hop-#bq&R%W
zoM>!P<d>bO!%O^FAayFZ1wA`mN&1i%t+caUAdET&2rGElru^-O^X~f8DbwOFo4Ewb
z8ah8VJ}A8E=3ul;hKdU7kmS`}1o9=l#fqN!5G6B@vi)%{GB1|e0xXlIQbi_?WrcX5
z*HJ^%lLkFjFLNeE;Al({B5_!Uy4XdjalRTBF`}!4bkQh5<;w9gdL23*P)dtQ7gs^m
zZ;){FvLg*I_X*1zG{-Ww=n9<P{>>t1Pfammk=33M=8iCDw6^W@Es)Grs6m_2+|pU(
zc}<TfwBc4oA!K?vgd8OV${&O;(k1%a6%dh%;OH+1$ZW33%Y0sKYao~?nSj9B-%{EQ
zCQ2Cb!;>fLe}$TFFjSXf0zxOyEk-)4)348!%@n8(;IE<>V9Rd<CQ)b@+qqb5+ad=>
zrfeik@<&RH$%kHv-TE0}uDf57>R|j=h#ST#kI{(cP%OROgnXRX2x=+BePl}-QJBun
zw+Et+=$+H?ZM7YWEIet;$#EdA6+|nMqS&C`SBe4+If+A*b#^4}DXG5Wj*;s5n_zQV
zS93t#94roS+S@yUn_7%x@-)ZTPoUc+m1Co|8T0Q91oyPpurdfG@?iB#M>$IX%s<3s
z2ZMjHlBO+(p}wSBSZx<D*9m{72)rlm3Y6EGVJtq8QW1zw6-xKGcjkEbu2K*evdUl;
zJ{tZu2rX&V{lzM~lkk@a9Phn8VQ!uLj~g*VnY2nwpT<ix{GET)<zp<lM^S^8&o+f|
zstTh!x{t>IWJpO5aJOB=Ul3xdaj3=|SbC)dx-C4C%6W~eu7uIp9wuZg@EuB-1O$;y
zmJRC)$`vK=C#`BT1?FotRD0}7c?x|)Ke+KcFv}UZUe+k<EBvQp&fMZ7=s`Fn_Q#j^
zEQ%@4AyQKbaL>rG_#Xn?GSog)L|pp4fEWgLI~I@al=AxBr_dDFDNVn4_k%~U_bEzu
zVfNFD8uuwd9$MVJ)7}oVCDar`x=L@G#Y<&jW*H`K?XkqZ6gE+~3dRUlCIm|gbd8)e
zw3=(~tygA+{S`|X5*#4V^aH^q1ib<(o8CA?KZtek9EqEhfGcM3SnbRD+YM8Q%rsp$
z*$>PgMARFwH*lOJH1m&wy2U91%p6>kqMwh`%Pa^2OwpSr0&Li^!2U%gh;~F0Us!;9
z%#o>Ip8D1JxXh&|zU%R;1Bzl8MR9byQ-FsJd9g=*Klwl)dNpm%?6^gL`)3(^5*Rsl
zr~oml@E==xMJcZp)p=<lPvXQQg57h}5qc;<jZUaLIx4g7Q6r{WQtnAQd{kwP5n4;n
ztr6;0nA<U0^*QeEf;1&(PKza5`P&t9!_ImWQ-rt@?27yNT2E|R<^q$lVh16vRzbzY
z4qbxip5oxP#Suz*te6S}5q_)$`iEI>4Y;Q^`AGg(OSVzYC`%P{h?54pk5KAvpp>k}
zmP)Zi;Cv!&>b^?454>K3qBzbJ;?}X&C%N^CGE-YU88f_4B6t+HGo-!k6f5XQBon@T
zoG@3;ny<a|mMRE6W1B|MWPiI;fEvfM9nILPqbz{X!r&bxB(X(Hyxv*-DwBT(?ZU1C
z;TlXfk53A5GjQ{-4axrVRbeJ4l4yM_X4U}QMX3RG<oI(55w0x7x&MnGw;r(zK8;U>
zFb(SjFpl}UNzn2hR_dWRA*a~Puoyl`DQ{Y+iF7lQB)hn;7o63|R+DY9VBKl7&gmWM
zgOv7k^1CRa*?9uZ>{2x9PE^mIJtWwRMcOUwM8K6*`eA%&=F?~vniD|@b#_CXkh8ss
z2#`WYKB9U}LB51`r&5o{g~2o_7D`t<sif&w>0~4z&9VbPZY1twX#*#cN4zfBG=%w?
zkFnT6ARdf%9w@5E?gHFg!bi)G60OY&gGF^*(<^2{l2vaL?CLnY4H3+3@u8wL(GWQ4
zN!WR81kmln(g~|do|+Tn_9Y_*E-5tco)GMhGvQ^?16REw&=f!bM5v?uj-z01HnRvS
zCjiErLv*F%^N*uQ0!hw1Q-B+Q_8!(D|2!brWe9E2Q2Z;Wc>b43@atw#d4o&ufp|)Q
zRv9kDmpYigVY;caXHJh!CHP^+?3UQz5dKc{T6@`UN_p<{7$WM_hGdSWPzcsqZ~Do`
zSnf~|%|K2XO`a@{djv4HDRwokM@pI^a9i7Os~Vrc6&71shl%8oRR+>cKhX+Zdp6=7
zmG%Jfl4~8U>&HS&if&G`G6Pxaup$#{66->IO^CY1vpKW$w+mnmf(uQ1eBp2)nwTED
zug7*uI){S65hgs^N=Q!qiX64__sQj*e~@K1&R5!{q=b~k#Os0}En#^&<F+G6xZ{=n
zd0A$AU0^=va2|xHNP5t9?vdc`XJ^{e<6<GEKLVg07}eu~FCSf?JR>ZLM0-s_^Az(t
zT|6wIoW{6}Ma+fi-Y!dXSu#`|K_02cia#j6oz{cFm@dR!?HNWXHoMG$p`Xkl4iGZl
zIR+lVjL-sQwxK4w_fX6@hHMa;ah^1FPV_37g(r@9ai7xewPKVbwD0yq5Kjwi*I5Vl
zEdd1~YmL7Ma2bRFj6ULT$AZ8y#1X~u>`A5kdxN(~v|I}j^W%y1mE-u^R1r-vW<xt7
z%zZ&cJ2;fi7MmPjWMm1@jKhXNm`5C`BWX|fJE6!X)1u@laiK8Rr?AxYjD_J{fnG1L
zI7lXkIBFYqe|wvX3%1LNMTL4DhXAbYafUF+#=F+(qTzO>>Cns>lVkIf_}dl8nYt3Y
z3-H%rW~q}9hRzj8i7rfVetv3Ef+>tZ@reM@hZiP(^dtUu7l~}jnnt9(5|Wp`RFzMi
zMlSOv3I;CI($>1bk3qeGh>UV}Kwv=H{9^e7tFgjZWka1M^aEoB+jVo&WP4s(h+9uU
zLez>n9xp2}ub*6HB$gi###F<2gjdH-C*f9iqKuR~`uaa9Al7=BSjSXBZnc!D_@8U^
z_ZKSe!91fYPCSE@Kh<6*c8%g*taK)rIf{E@&GQq_axM<g938QygL9Yv85k{>y`o|m
za)-z{T43I4rFmgIbr?#p+pAMWFGge&a$;1WCw`Mo=BK$<CQw#7U=jFpf#WP*!+j~P
z7vip^tx_s>ym-zAb!W1xZzGmD2h0SvJ+t&9lLcn_!Nc-sTkz*V;;|_+V~+mzz{*oI
zS)8TT{Y*db0W>EK`j(6Wo)P9L&wI`BoRV%C=d!fx!Vx3qfkJd5Y{5<+|H9SKNzA$k
z`~f4Ze?HhaiQ1BH{)U3w^rYlZcXv|-X5PTU3<uA_*sLszMk5J!TMEi!C@)JcOG+fI
zBhC1Gi4Y}CK`EaV0+Y2O`o`wJB0-xa#PZCoQdGz}Ejgz&<18Vx*cKd#x6Ts;EjXts
z8tV!dItTqkR7uxVp9mVCzKB+_*+l?uD3V4iE8rGFGTS2Yz|GS4Z}S^a6Qc)ha05kQ
z2EDt$29is3U5Z(~GMh-;e{tGJ+`C&C13Y8KOkTX;@7xNWn?Oq>KtIl&8CU(5zrAQ9
z6Zga;N;`9d5M$N&)g=IQT(2&1Ak}vEn@bJRQzi(j#!&2W8IZfdvc?H59YkLc>dLW+
z!vUd8^O{;SCSA_&X-6eg!s5Rxs%S;arN=_y@;0HFEtqueUAzNnaD3vSo?^WwcQlnZ
zbIWk2ue#1JYwbgc$*Lgv>@Ot>Q*{#FMWLQ%T|s$sWzu@_`P&4!N)+ZO#}Gf}{Jy}X
zb7J@Cj5~#8vT3D`rv-Rc@fZU~9yCTtKlp!t;0IHRdp(++w0YfEf|^o1^V=B>rKwM@
zAqw#w^IrwPm`!h)8VBod+R4HymuG$dRzaSZqiv%+VnGtZ67OR$u2I5$K>dcT4U)!@
z*MPYdXjHJqO-uO%w1MFs+$HHVq2@cjXmbe>@if(8X71+hie>I%A^wt1aBwL>4TPC8
zU7Q9R@nTs_rvdKJJ{DrSH;CV>{;Yc~kcVB5^YW4};0%EfiBq`D1Y*~mh7aYC7pExg
z&kt5x5iuFSk?YF3YxB~PWryp5Amyy7ljCRl+eMmOw{v^Z6MNnOmbq4Uv&Sn<{+ZLK
z#~u3H7h~gcTMke8Z~a0)u>)%&-1fzpHv*VE?KqMmFP-qm0)fA|GKxiRDv+dz=IT(0
zS(_DtNZ@c22hIxxn?%G^L9}TYGKu6Ctana1;+Dme*Nsj>C>s;(w%{EsH^RRuqdSS1
z#mVSTq_rks)U%WRc5bc*xFCqn^RLs1-M8AGi>6@Vr}&96D9BB*x3$1W1)3$%jmEBj
zB`F4;Ju9xz-+6Q^1Z~5bqY|$R%(G7c#?k*Db&IP&@EXFx{um{-nIjpt#idGR#^MgG
zJ)X+|$_&B|!XNwI>Kw>|ob}-Ps{~~Jl*e!lGTtpLFTgU4^fgC9ol~K<5zGj1s+^hN
z7_Z<(j6gO4Oi_JU{n`z>G@7|ul6_uKX?IUCQrK2v?rlYd?F49Y?CP@xn)s4OVBtRr
zaYqq&*=ZKNy#RE@I~1!4$aQPha-Vp~-+*!lxU=Wh5kF>Ot{RQG^vY-4St6W<dEo30
zA?80sCWL6I`rG5f5|-w!ohWb!L9yOl<kq0Jw8TOB+k}*qBHbRs$GdkIC_JCENT}Hp
z>IM_%Cq>p>=^iq=amY-1)ZjQF&V*hw&F1q2xmVKpC#T87?gfJ@su<4?#uE3DqD8Yf
z4pj6vgp=nUHu3~xz8|0%YdfCuSf)@aX!OZR?X(9-xgFf6j#27M4;7_Qoe#5#u!zQL
zLfsUcWO^|2s!YVNzsFXBvsB9GUT~&B@7^A(t0=28A-aZR-LiHz&LN3Ch)~}aYF^VD
z8O&n%e=mWqO0gjrR70L*{X9WE<6&~R@i=MBW8mfqaWj!D`t`s|9A4(a{#YmW<4Qzk
z^pv}yMB0b-Xj?oWMARfSSc>}jv;dl;!NP1$`nfG00fPEnweHwae`kK+D~S3ELf$LX
z{DrZToucUPpkVVj7f=lThL0Bc6tYU2uLRe71m}FH7JA~=#{e>u@K~i5o;gndnF#6H
zpE&G7Oihe>hKTE+*&xu{8i5rCiF)CX9p8Qu+!TXvhy||U5C)4(Tc1b9V=txrfkt{y
zGJYb&t-#$iMND4mDWGCHU}u`F-PRZCmf&>C{-_53P@p?Cb;`zk;35H;&AF>dBFo(q
zFdm&vw_PheU3_EQgMgvitVGj;m%c9Q6v~3HfX?EupXW67C(EvnmG(EWsZNqf%=o*@
z$dz7)#d%6=su7B`8HYYg$~{CZL|#a)c*hI%0JAm`IY0&vjW|iLX;zn|4N)wFn01Gk
zPk&X?l##`xEfgQld0I{8{huR`$;{PL3YK+>AZp<;Bi8ts`#i}!G@~`e;6J)V*dz$w
z5w{3&EwZQ968oB<OmL<Zf!KQD*DpAme2!E1dt;Xuf%2v~!Y2BJ(r)wkD}n9tfk3T^
zoZk<4iN9$bk;b$*4Jw(-4|mq$7Ny-1Ew$1nKJ+q>c>z;m#E4^F0Wg)(HXx9p;;+Q@
zLR}Ikm1IqvAQPT#xx!mWxDH~*r<}QKIvg~((YRTdv!dRWPAP??dQ`B#jrBP<`LC8=
zX=O{6vb6DP@nO~_f>T%G2PMS4ZDOv2Un@Q;3lt2k+T&zlta@lBr^V??m~u=35t!BE
zJb_Sd#`LLix&DS!S<Mi4D4mBM{y`-jy{`kAGWo@LMGATU4bUvD>dntSRn6YN)|;+i
zl$|zn*YSc(niT$4j+HkG)9AGjljyDDclcyTYfvm%Vttt3q}!%--gedL<VSZ=$s8-l
z{X+z^3N8LE&}Ra9WJzy_+W?|W`XTdC*Kh~#&ao;aFb^?{J$K3RfhsWnv{#4c#g_jf
z?E%B#j8N};3dl>JO<n&jg#L{3%Jb3h6xoT`Bx?i#u5W6~>^X6Z(*C5*!Mf?V_TNPr
zJ^_7?{||tt?Ese}FDWT{O`8)7zsuj5y=8r|xY9JwPdL)g+3x|E;M_C9jC8ZD#BfsQ
zdLO}Uz4u*Ef<0DSgE2*b3DgPIkv%v%KQ7Q*%sSWB(+U?c$u{}`I4`^$L#2@7u#!-B
ztphtgtqch=rwsEx9DG8wQTS>CZWS^T4?=xxqKe%J6fU~j=Pdj~fvz{5X7Das>BF)C
zOgFt8jtIW0P!A1Z>SQCig%H>0N0M~&a=k#)kkj^9eDo1%$X#b5Fk;JpLedn}Z7mk{
z1fLKFnOSZ*!f=lLPylsENYGvRQlF95yqYpQuF~Ibude(=ngjn^K&^Ex5pB^W$TTFx
zA^w|5EnF9Bg6km2y1~q9#vx_-UInpaP>&VlUMNak{6uMYeyD;yM=gF|A{AWD(7><v
zc~PcCn_x}pOzv?>G&<(neY|2BAVsgzo>J-EpdmeKiIOfJ(hn}pj>pMcC)tiK0J`%-
z^>)^f!8l%!_;=RqSZRL$J%=F3=jd-!Kv%;Ox9HcT1&Rt}GKRDkI|=f#))IT?1P7LS
z?D0i@F}V<;`G+=@;TjNZ+Vo4MeX7#q$7^IznapT4{KJCa?g5%enM`pay0QdeE0hW-
zJ`B2BY*bBM8egq^uGGP&kryPjeq|wW^A8gZ#}KqK1H#zVd*tRHH!Msl@1S-kd{qcN
z93uimEU_pF=rwcJ<hW}w{&r8;Z#rYY#Yt098^U@eDoc=ZMYe>AJA~vO4Py_d3)lFt
z1QA)=txow!AS}&cn!^k_;AOvLKuc_j%(9UH*NzH`H^pKzA>$*43}y-9G@-e}(g=vV
zFBC>QrJZeyyObKA5h&H5$aY@_qLZ_DGVU$?O;Z{eyC-v=U6urJFXU&RUeSnq1(J*Q
zA$Mcfbzhka?`dKHXr?|Z(7ZA=-YbjfJxweWm-4Tbc5D`QR3i5emqYH}*ZK;`IP<2E
z4$&ou#Q_rOWtnG`q+cALSg225nw04QKgnrS-1iHiG!(1fT6(yxdl`UoQtL@bEehXS
zwkXYMPLiBl94-{f7qoTN;(4VqtMEJ{*ZQ9eF>Q3aX+e9T0Q3;x7Y`S*Y<{=I(iNeU
z6e}%9zB~|3MOEvK)xHX(*@<UBGygqj)|7Z&fBVbYFImigw?a`ky`{wv+$u<oB{~K%
z`pXQcRFbLw*H?4~6rU)n9#$%Ue%SikWmf_+3DX%8eqnB%Qle5TKnn?O33b<*>ph(0
zpke~ubzN_zHRNs;7Zc1v)pCy#nfhFAvVoys<5^n4?jO<06(?ZY${?O=c!i>HRT|37
zKqK>W{q2@Z>!mPb7p&s!=`jqpfIlkj3cIV=wPRDYDu~<5C39Bf+#t}|rp`zK%1-|p
z2rM?nd(OB@wyZ|d9pW}^v|ZsN#)Q$776jg(=f7vSOo;{705`YKio=xf=aShl7M7iH
zra(6wT?c2;dORY?AK>a^4AoF0o+-;}AxB?;jPRB~H<cqD&ss#RS$qO5RHEfjOc&@+
za9N(thX<DV*v7allG*4*0_i7vCL3P-X;M);*SOMy-z>xx<)=8~JwcxIupn*PXfl91
zi8m6)5IPgj5Xi_i>X`Sk&-Jb4G)qOJg4*g7rQHz;&&XVl3Gt_pU%4i$#4cY4GAGJ+
zko`fks3|n_9P?A!ZP;F*<_wWE@6g|TTOx?m1<$QbnqErq<hJ$Xbx8U{a+6V9%Qx07
zq2v#OyX1O6d0|b)XX)&@-1?wtYEF%N^tT6%_^e13Y?YgD0A}JKJYj%+#)f4k1yYi$
z$+V5iOoTJVac2sE*&K0fIMz_wwcts?rn=@-0C%|!!<rQ7C{7UQS;Rq}hlC&ylt{1}
zK8&~mDDf_#u7TYTYg4VvC#nX2r<#<&KNY~@(Vv1oEI+OMD$Jm*yfc0y&;y5?3b)Pi
zUm>oA<0)(Kn$rQGcU3<2n9$B3>E2;K&E2%pn#4RLefUiN_E=!$n_T~17vv8}WQ2y>
zwSZ)zR@zzBP8UFDQW(mbeC?4Tx1I&aS33!-%_91gk|qPjV7+?KlGeXC8`NWlb&GxS
z6K1J_>m_2>5!~$m%iG6&H`o8zq^rC&;hdmp9ihhwBEh0Jp=LhUgU@=-?KZH(YnfBM
zFR6|^rv?UaHmbxcf;>7}rX0F<fjNLC1djwDjmB~tgV05;kYzHx?j|Hn7xqflf~f*b
zX<S}<VuvyV4;&yZoRR@Fde*PJxhW;wLX?--pTj}z5t|hivpn&<QAGe1qxf#aWF`8x
zC^B%aB&_6>0$BBUM!OR8e}e=AJcszeU)SI6EDyjkhnulu3CXHg9v26fP<lJB(HBPx
z@<{0#0Y9N?{8AWSsaEjY_`MQ&fahtWO1C9{6zWNV1{Z7Ae)O5mSmc|)rX{f?<Ug{M
z5W0j!tKfODk<xA`tFmN==n>>mVZtzVIcUT$1k$-#EtBIO{mpb~)Q1Vo`ERA%;^eWf
zUGu-+Dypauw1UNvL-{#dx{B$urpJq0@wb^znqnSm@eSTTw=OEn7sJiCa9c3<SQB0=
z1GDXNp>CyK*TB!DrxCvsXfj}z$5vta=8uBSApL{5MBG}IZ4-+nWye1!fW`j3-gcDC
z6IEWFtot_#a&x4|(E<Uw?e_kJ4brM{wbEu%;&2nLcby#yfFmZsxPMRr%=Xc0th^&R
z80KzGjT?6&;SZBNUB`bY$W=0liO6^3&IN#>I23+IfDcGHqy=1{ZE4Y6`N70W=ktcH
zjVFMLQGi)F^LcUQ^he;WdM(cSHUPBacARamFYZ@bTS3e8XxWXwX)*7S^u>@8<~0S8
zVk<va2=i8QavRrLN_rGJ>&Uo!?@roX#LzN{O9y|)S!C(Q${uehZCbPHAQ~mL%rbL9
z%{&Cp*8PE3uE9OQ-Ai=M$&#f#5@`0stE;znfWX?r%9yLwV<~7YIoh5jG`EL4Jr?o!
ziy%{@6P58$+$F%Dmi|KXCB-rP>UWE>G82+Z?qorx#|RI?ur6LH#7%*1;3f9@7lpY(
z0z{|XCRvevYcFtrnZ1x(l$wA{Ue-#gyM1pEMxuqi{}cVqNbsi5%(#Ca63|_j^$h7k
z`xYhJDoyUJVgn&$fTQ)i)x!Ic%uJ3EsUkR?DacKQyl7fjI<6Du$(5op)wEu)#_V5|
zM#4&x{Feloz6vy)W#PY1K(S`fRS>l%)E{rf=u^RfmN<~&bY^G_mF8MX_;=o!kkx5C
zCdk#xZZ16&)D??-ugHjhR?<v-Q;?`Kn-B;ev=+Y;tb?U=AYN(%%FJG8d}C@mfLXL>
zm{<bqRf{z5APvRFN}B{a6{UkZjJ}g#4`Q+<kcon}yo8Uhq__sK5b~&%GQ0IME<$+K
z`xpt$+%jB6*<vJfOJ33m_fJ`4Sp0b}`%Z~Le~gJ~(Eh^cIF7+x@wgJM8beNnzns6L
zXo4a`=2dceq0Wa0IS<o|tIB+>l~!D6lZ<;oC;3b@Hb#s&t`T6W(1PTiGt54+Y*&$w
z$VAMh6$L;h2X;?;RMOPEESw?MBETt|kHq8}5N%0sy9{vw8T$z_A2iU-RtO6Ba5s>p
z5-Fassq58U0@PfjhhI!5X=K^piE`Hq=2prYn6oK&Aq~SGqps>$t2Ck5F_d|ksD-eT
z;Jm=ps+}^MZI4@&cYLbw*fS=YF{cmAq-@C2u0OsZ#GJ*ZJ{(&s%?ROmjtgIzozmQY
z6Rp3fl(Eo0h6?<kqEfbM1|wb*k}D->a;!P%O1o<qW^bc}2_}ONNyFO;^2o7(v{vJ?
z20-qR0%O3vg!~Z5F2ah-0>O|lr<1HVbBG_zxj67pBini<66!iwAvsUUWg~%}5i$}$
z<TSHRz>|`JzO^Zae8N3Ac>dzU-<s4f3HBFt^k;Cjp!KL7EMHpAImL|RWS3sO8;rA+
zbbAOQliw-7wrqNPRo0NmZv*kFP=A$`mvN?%=q45%2KMOTW1+2K_YsiXa|Qmcv4UVk
zD%9lcAm9S;!H)uHP|>Bu;5-thKd*JM#~@iXfjnN_mEr!lPHC;+2vx0P{(JhyQ)A78
z`P(d&EYed%0Aa39FPPLhQCu!)oE^~^%9qw4!Wqs~+-qH-EN(bUXBtR4c(qXXTldz3
ztZm}tR|Jo<2Mvyu4gqjGTJ;#7EN6!f1@lL7@6+l*$^OHOk_4qf&vvH(m!zqFl6uM!
zU{Gg(_hDk%ktE!mR%WRDcy<)zw!(KDpE-_HW4)t5Ez_j9n~rzO@-j<5xFN%MGI_;K
zpufvq7t@?zFIy_3yVX@`w)e-j0-OnBc<f79s_zj-<9OX;My!0a6V)-~+qac)@jhPC
zs>bjS3V{8JEB?0xxG#FlLu=oP#{e<X(%VXvyNA;Lq>T!h;};8aF-4V=Y<Zoaael|d
zy_UAfO@9dL7I19hLhvF1<|FpkdR(ux1ae+uL5@6@l)ExDwm!~TS!a^5;Yr7nc0;h+
zYoL}THnXtt$;#>^_g~qPCfpS6Ym+Oc9-m5JZVXSfb;MOF?QRVBSG8goazWj*=s_~1
z)$+^CDvr1f4uEA@>*;Xt7sTY&nM`8-2_#(=3slEYLrUybP6RZWnDPS(0}N+Q6Y7RE
zFvypM`Rjt*TjJ&tSde9O{*%EVRx5HzC6-Z&&oPpSIWg-LlKxz-1MYrXU}9V}hsT^s
z0@|lL0>1h?rKtdWZmFBU_#=|0WkrU=$l*yOnl9WeX;siy@!XWt3QT8eS=ktK*U~vV
zy@FxX&BA$=lZ~YPRNP<^<c4A9iFKxkZn9^duJ7dCh#i|#=1xNB1bX@OFwnkAn{<r-
zkgk$X5;RWZwBgDM`lYg}fy#iMtoX{0U6syw>eQJ^i~D9xpB1kw0dZ{V5+5k(&-7HA
zc&An43#S8#brtO{HdV@u-BTN_#c!3SGQGnRS1XZQ)P>=N-E{sb)SYW*4x!Hb<xfCZ
zRc9d=&pxC06-Qw07F+z(zs?|<%q{vm4~oKo>8P~TnPC13cH{)$O$I>Mo&{=JEA*|;
z8xz4MHs~x4jYy!^E2vBS{{ejKB+{fG#<w_aeqMAF>^5*Qi46)2_g8%_w~K`h_o&g>
z@MmD2cQ~kEbUIf+9#y?9(;gQIfm*$?AG+*p=VyQQW%d&SEG?z4>*V5nz5N`1VU!3-
zpvPdkb8;X(02Dv@IluZlQ|HWSA#UH4b6qv(I+zywvqJ<p%b4PX55{dm7*=S?RooL3
z3b4I5;wb@q5qSgSmiSmn)0|z3<9Jfjt#Cf5sn9`?t{&|^-x6jbqD0ZtAc^S`?BUU&
zNB4qN=XLoZMGw&Q+@`EEHB{zqy9sk)Y(_Zh$b@Bgfsh*=$6)%zKQln5wa?aKkqevy
zU{p5Hv5WwJk88Lj2FD~}=J-CudPxnNln`$cVqDiBjk2U-)O9Jo&z-_(fEFf%;=VZl
zmjDo1ax*7&8oukVK*Z==&)t;&p3WNtw_OMh@zD+;B|LKxDU*yPu8?IP_zkdoP1qka
zD%xxl=q}V*qei3=n2-^@aSF88hzC_Vj}6|Haw&^vg?aK0405aWFQwggZdH?RakGno
z^5((SWh1r`0LykO*x63MCE-eO0+m_IVC*8yl_=b(gpErD<#j-zfs=C$rt6>(V<xvp
z8(vyeq#2dW$Tt(_Por4Fo@ejN0H6}8RZa&7C`IFS0u^nKReGEfh%`F<w_O3|4<jz7
z>n)bygqptSzfgP(b`V7@I+Qv0o<osBc^)mu)1=)mIi4WIU*Yh>$pdD1<o5+eW18uN
z^olUrI&FPybmQnhkjgY6e213!hXlEIGB3chS%AOQR_Ds9J?6U-L<{q5+yoX@B2Spw
zAfla4@@HNRm|KGvZ8Ece;TkY1mJ4~(Q@<?0MBp}sIoYaL#jm**G>Z`Uyf6tdcjKcU
z33rvu3C>(>=^vL^AJ*;=95=q}it;Q8Y@ChiJV_FZ3U{NY#STii(VDzmo^jN>M4&s$
zGafqS%y&b9bdwKF$(Mw<Ab}<cdm$60<peU~v!_f+mtSYyNLm{?XIS)Hhu&1wf*Z3A
z7Pko@Ll5EA7^nUIMAD-me|H?T;;Vlypsps1L<Q#(K^_=Ja%?DmF2L-<ddrT<u8&Cb
zT3J*tO^QJ$-Aqo;Cme7GV|%4x0vRe+;^*bBwOUt4Eb|wBH66JBs1RM?W+DDk9}VVH
zX_`LxS75q<si~AdA1mqctnlg_It>X3&09fDF^<&)(8abd?hxotOr4%+_J)wmF+6HR
z$W2ThZYzprQwHYwl)n`qx1|`9X|+k9OS68UhT|K1DkH5UB(Su+9c4N7I@$Th(E?3*
zu_QAf+U@`{t$J$BdR(it$<AFEjP$ku6GNBPI_PRlAHTE6hiHvNy7pZFX1H1Ls1h!u
z(4(w~iK!308`vL1-!L5W-Amf-=|Rkv6f{YYNri!<>^621!ur(R)f)r)+Z}I*-DJcP
zpOzTPM{H#1{iF9470cR36@Pv|5cFvbX=RAn|L0`Vq4&qE2b^S2T65}@*g{Dyac%4=
z@ja#TvH^pk3yMnwnM^FCcoe=Mz+IDTLp<~#X+EcE!_<0LN#`Fz^EwojhX7nBl0cfu
zH9>hO>fJnr`*2Z)V0_4D-xL6eM=Dsg9ifB?g&nGVJS&fYjIXkK9x9p5b_;brofJ5=
ztp6y8Kb^EbtTs~xnQK{Jb#upIcc=uBsBz;YNLvgG^hfm64}nGHotRL%mz=zGg*sAx
zBk9}t#t#M2dPI!jI7x}j!6{fc8Gch@dizjDIuo%5f#y-sRpZGrA1YRzRIBltFf$y6
zC2*)V4j$8f3^;Rhy%(>;0ReeILy_v>tuH-MGzO;$DMqVt%#&b@A2h=%KK}9)Nf@F{
zy@`jEG6j(fG$CTn--|j>{zy~O9xn^?*X6I?6K@Id*KP17#$5RyfF7VeY>GK|uT>(f
z3({C`BE+mw)wSZ>GRLTcM;s&s>LM=;#}AayT7XWPHRSWMbdStpSNJDCd(VM*rJ8Cx
z;cm$Q+9=k_n@fb-CaI8?e}+6BI3f#}#ddtIz}SpAUXZUMEFb5336RZO1?f1~Ej(?=
z)7g-lO66I=h;?#T`x8NKgyN-eEMN0^5U(sPv8xg;p;s7nFnLyx4-US|9T(3((+_-0
z_jUC+<^@s^g(==cSo7syB$=0Q?hl}gM5H@}<`F_<Bai@sN&?MX2y(p6!GW3s+=RT+
z(%H`<fu>KR9~XG>LBiZ-#WiAak301pvj%Ex){Dbm0&ur6mtu{HH-+RZ+`ywx5#XwI
z;2~lnj(;x~nWTDd#%U!$m<9xlpOyfj3At`s;B|^*nqwfP2Y!w#p%hDoihnG~ALH<X
ztskRxpD@kf`Y5h@N|+_5w#=9o3%*G@({q4L-8KiKLNhtKc!o@d8;1xpjk&DCZc@VX
zp#t51ZY9}L;;=HC%}wMOaiqXJ5u_MAUTL#gqlxpfRE&y7f3k3YHy!cOiN$5Y+;N_d
zNrs~L3UZ_IU25n3X%_QX`z=t~lEQ){w%A#Sn~ad7+gjG4PJwO<BKJtF`8H|Tje{L4
z;O<Jf2t$s~oVxx8fo>M3HJFB`_Wi5CYy`~-;6h>JlK`2pHSQ5Y<qrLP*ndcwH<AHD
z9R4nV=|BU~ZRv?;$hz)(;O?Lnb|Phpp%d!1jnrfcs&y!Vu2Z&PSTU)r`-jH5D^5__
z^pg>SV!&S`2#(;X!{K;b2~V)jN_V`Xq(8z=ftj1c)wSOT%>2-k{eN49xn3l0xy8nA
z9{~B&Qbot5N>g<Yw(2MnZ&VUubfV<J2=GJysE)%Tw<rniSc2#ANjAn!KQ_Y5m#Vs}
z9@~Fngd*12MMH>tuYi~6S3WHYwQ(!P>qgHAF)a~2h;NZt<5{6@y)>fC?|4p-JK37#
z3r7Cc2^gwPV$XNQJHoWE%_0cmil32i1K77&wdV>j{UpK3;I$Eb0_iz-uS8rt_`gM#
z6eV0%UvZs4Xl1p0{8>rY+mNqE+&uptsR}}IA=~^zDcXR}6w~(wNSN*LzLN#X)k65*
z2tkyparzfY<=LhKBqylEWoHY|T!vdx-LHwBBh-{pR8sQCxq?iY6v0P(^!dWPx~7X<
z$STggu!K9S*zhuPx&-{4;5@BhXAJ&f-QNrJ_s6Z{5duG75EQ|22KK4l3-LEZfzvhT
z9#GP?B%g+gR>9td$MesKm-M%LDyI)rY`(63UAQU4vN(&lT5l`@Ld~d_WMTQXlI|h5
zT*yWoQTi7JgT`3<6Zc@Tq6Xd@hDl>LAuhrLOcr`?7sdo&l~8Q7=v<t%N3lD-%ed1L
z0PaB>>IS!N7lGtbm=q3qyf-eu&BbS+);}-y72=QK(agmk4LC_yW?_ZqLH2vIEXbjQ
z*Z<-(A>L(KQ5AkEdpB<R<>DJtThig?Ga)byFE>+guO&(Oi(K)^O1>E@d<Dz{hDwd6
zSJDCNK%t)Xh_nPSUYc~C0enzb1dUi}SwMHGt%*$9Ya~~O%5bDAckur~abB0NJ)DwC
zyGjX|?AcIhWmV}4be`w#Kv^dVHB}JSWbG0=mfvpW1r2RmLfko&ivxs!YQ%HG{0){-
zo#?6UvH*172z}BMYc59`=A1rrPF$_Op;r$_-_Ce)c~S+7&7LzSR$sw+XH0L2E%dh;
zlPA}Zf0--N6b=ga@YuL!+)?k@mH5G}fj0Ht5w#|P?i*$;=?RJFl+O^W=l~BnM1tHo
zj*6JMaGU(1vKbQ_NcxSpgt^6)R*be9vGB@8Wwaw%vAre81xL8rNhDrr6)+P8m3TdN
zDAPnjXpM8r^t${Pzfjt<Aenu|Vyn6W4_j~CqcoF|^;ycDca((4Evx`vT+Lb81p4aT
z^|)GTH=3AAy3N^tbr6%bw2pv1+J)yaCLW58(y{9rMc#DL2p7ZI?QzMP;HDxnBGD83
z4Y@ehIvol}{6829hGcSfnTlpM#(r-I*%~(9VG^i^4@D!#JWZcn1v)EnXi>B?hSk_p
zs6WKUi0Lb;kmJhY@-T!`I8>b?(2YdR!zsEqmYNI%rDhX(YgPU2pRsvPO1-uc5_fU0
z9zPNUsavZkUE1RbCH?Vk-i1g3c{g7RjCD_OCZk7*Ohw&vvZ@m9yTDLlXfh7QaqEJ)
zZ&)36N@)vSS6L6#qXh-(DBG(-a&LK*oaL?=^MtwO$;b#%Cqd4J8XF}I<_`%o8)0w6
zN&CDFC}*Hw#u~BDh5#_8A{I&>IYVh?C{j1u;6wRWy)mZ7J*u>e*Rg3Ij18v%xE&aq
zCR(gJ6-ax13(u#$tAq=*bxG-X*0cg3<P-iSPYG}h4W7@CRJ!$aAaeq4VQ_+`-u41b
z7jEe)=woXof~io2xZ(*hMcIaBv8fTCkZ+JphDKbZl5SN85w9At=?nuHViNrc(c&Zl
znKQ6fl*RpPIS(?84Y^mGS=55ZH<S8_Sp^{Pka~P7!0n;II=^ebziv*Mt=U@ZjmaAq
z*$_W*T*Lmh8P5sL2Ys!skXI7zs+$y<Io5ONO^1qvIy0vO-{SMB1iKq76yoq$Q{_!u
z8kE^{vOwnp#7dwM8wxTzVNZ#w2^nc4VO}@TlgL@Sk*ostD2ul_NNIdWkV#zaRbb;M
z1(+5|f@4o%-CAW+(7ZOW7Q<o}2!SC9N|9F1-zu#wwkGfReK#ZJYI#^0`;1`$Oitce
z!6h|z-Q1;F`$iSQWsh%=X1344{r>>{P1YI?e`qxiRnnj4@=Zp<D}1X!`mo;MJ;M)#
zxY69Y$i0XAYojJMi#MM^4*<ndW?IezUcBzw#KM)rw4BHiyg<G9mZVHic4HPtMCE0K
zne1I$siKOAwFOZ~Z}RoW#!68or>+6`Cl1JIo$Q;nk@z^LlcyBKTX-w-XR5Q@VU;3)
z>eBf<zNZw`QG99(fn)AgIuH8*x*fKL<F^KaNGWOw&kAotG82hikhxP4LIzgHDls?T
z+?J%fh0>Y3G1mO;wgaOTEJ4lmV#oFZOvV%s#Pi@Wp{^StNGg|SOJqo52q8PRCy&U%
zF`MV|b|5vr^|K%mCeISgL;9guF)%FExh8>bTD4!NAqez{GP)(5b+oz63cm7aSzMlZ
z$rU%A5NP6X{!Vf0*WIxw&3q;tOHHvHTZCqIV9?kgB_74$g6SRsa<Yl-O+p|Y<nOKy
z$9_AJ^k5OthQ;*RodMinbh)}_s>bWW+zLVyVzDzCp9wPE(uFe5<L*)v?8h%>RL>B#
z3NuMK@vwI2g83|gE{eKMK4SmM2-K0wD=c%tw~NY{<Y~8$FYE@!nvR*ya6G64m9YF|
ztN2BC2Y~`?ya+scgDAavySiZ5ZG=q{y?^YebUA!-gGD96+J2RO$RmX`l#V|WFkCf=
z7syJUFgG6wvO!y}Q^~xVv690jY}j{}AnPN<*Vc89KruGk2{Zful+2W<bTW{!@g5*~
zIB>ja692l$(!&`TK3VfSAdo>KSsbl|8;!1cfb%@JJ9EK2T0E1^IzALv3i5|wBQCCD
z>YgC}5c>u)>U04v#e&{cl%!?%D)Mxun8m*qKs9pcMHS3J>G_PHGU8H8b_p-<?L2fE
z&9pX)-S+{ZjW#Iqv^z(bzlI`>n+~1?5v0noZ$Wh8RCGly2&)q67oyVpS8}d_df^SR
zud62)*Kr&i5s!8d=0RqiXCg?hk)y#s^@FD#`gI)@H03wiuRys~PbSq2)(?bwq3Wtv
z=0)>B(rz=7rHqZ^3*RdsnfzB#M14(=nL$yh(8Kb9H7bCnHg0WPVg8?h%rZkwoc8qw
z-S=AoJ%=F*CaV_P6YLF-!+yHh?vV4>5z6jC9n<@tZE=(kbHltEk`^LfoG#39Pt$=U
zxx>JsaQtR%{Nz><%Zk<F3++Yus$=dA#(*%Wpb!s=opZU;Y-Y1&OpdGdw<&@qzajB^
zn<|j&#QYCnPDcTB4stygTXh1s*<3wx!4r4ZKp+NTWfgvf8H5RQZ#UpLwH5q>S8fGm
zqB&q8;5>ouGwLNmLvlazVTo}D9!&?LxjjW;MqU?}nq2~EOA0QMPU0W+f-plxX_~fp
zKS_E!gLC7>`rBXPgvu<UclQW$cd>oNS>c+40Ngvgtq8gm`wxP+)huP)&&gaMfq75Y
z&Wt*nm9e6%?sJ7_m-QT9yP)yzDIzOfKDo@+&(o8N&G!dEc}uD4eN>sj#p^=N!Ca(Z
zan~s7M<JUWa#j=M4zwYNjl~QB^tc6K<&XN?osq!F&fGEt;u?FawbuAJr?LG<4@<0_
zQGSPf9At3Aj05M9Was2?#d`8ula%>elA8jmXfxk!4A{Nqq=FUi|3F>!_D&uBQ!e*V
z+rco(zyC!!sBjER8|?0?(Rt+%iV?6*;d3W^XAr+phWE%4V*avU_r^8{7p-B9VQcP>
zy9JpHY)^^-_J>12@~B{y!kX3=R}1qvBb#&3kgzcIP(WHTaq2bW)WblbqGBN+i^Tmo
z%@Wq5P!mfX&hI9aw5-x2%j&XU@Hj8jdy-4wCBi&@C^{H{+rAIL_COSN?pco@;mv-m
z*%`+u>17ccH0?q8G_d`V{NN8VC{=CV5IGX;k?CSFS16JZ<#ZclO2FMue;g^y*_c*I
zU_POQxWh?yNT-w#(}xADH!c)p!Zx^9gSGD!V*cu?jNxH?R8ciW`lSolLJ)nbV13O2
zY&RuM_Q84+L@V1q0>^a=@LHR8d8$7{ILtkJ>g0%{`P(0_H+YJUaTemfb)=&bB~JSx
z7+>Rw7+w0Fr=-cPAkFQu!Lb048>RB>Sm-zsCWO{QXc0|8>G$a~;t?fWdQ3Ek_XRK@
zl5x;yNsWTg!kpOZ_@btPD%*a2D7F@4rq)_JIHWx-wh`(o`*?L38#e)TB6UDg5F6=N
zPA~?4rHg7x#<9Z2d!l>llLeR_xQ|yD*B_q<;_f8N5k7Z%33XsCq)W^51kuB`v1U)~
zdQwqqD|4<M#|m(*lV&D$<>Dt7wIaTA`oo%4kTWH-(Q3Re1U^7mN!@(<)FKC4F|kr(
z&S@YfM~#g<yH;&2)J?|Ev{kERf?PRER44mvLOjlWgnmt)=i8`YUYn}2W3!yHZCM{H
z3m2qav7I0@Tqo|qcCJVu3tjRwj)TimN$rU*&@qDCigsMJ36Ord5Dx%Xmzb43RF>j7
zXKrx2;v+$DUoS@|y0Yu(MJ0r>RaoGko>3s^R+t~L*iS(c!%dkHt0-Y2sriH4>7mcM
z<;()L`yI>kKwNiLfif;oJ?Z+l{~4%ztJ@pvxf5t&Rp%krjVJ{CP$ZZ^L-b?GLhw^r
zXTyC0`QJ1>K(I;CO2kQAKx#s~F3J9zt!p%e>5pOk;Ne6lMF45Um>@TZ_gs1NS)0RE
zxk?!WvaWHd|GE(Od<-kp{I=7gXBUOJi0GHWVtWh;G&`_urSL~_Uy12rCq-s><6QE&
z-AH<z<>H?6K(d&L;Stk;HBvF_H~a-ZFm9tLfNArQ=a(h91z@sWA%ww49bDll>uM}_
z0f4)uU}&b)h63F(-7%(TD|Qp+6>@a6R*ONU%|<mEg{k5%iwqd1_r%*uyAoVj*d&pU
z7r7A3bk!vR_x^F-MMXXMaMPk4eglx_4G*56Tew_^n4Cb-)p+V+Qd&z91h@Mwf4lY4
zim`NalH9)p<<G@TV5BC1Se2P8;sE9Jr;^ab0lzGp+;=!sp@S5dTa6hu`pc*ba_97B
z9fp_n7!>AFfwH{mim=N$1Y1e{;Lr5o2+d@<U_y=B50QcI5kx1`dt32FS%im0F}H|?
zFX4B0wF94FIV?5fj{+GK6qd;l=7>v4yQM?<+VvDc{y2Oh)6Fx=ObWD(A~LQL1mEEK
zuH(hy6Uy-J3YwX?%pXz=J-JD@Tu$2k?_@GmbOPI7Snej3qLyCm_aNpG4##MuzVQcu
zye^<%=3O%hQ+-0m3CH&lrnMdQ`iD$-!ig)EPs-~L<GYS@LKuXE<!mxSLk-<fmClAj
zPxcvuahR~2O(Db(&3+_^e&b()xnEpR=0F9;Qq+i-1i4cj5Qcll;+88xJQ0Y6QIA!w
zDyo&+88-1TCdl2C5}jDfa&h~sP^dwmBt39BN}S@NrmjkNcXV7$+KooY<x&yaTp^6F
zwWf#Sn(?^OEVCV0t;E1JB(h2pjg%~1A$&7Fxfa|##VtQw9L>27h@Q3%LPk!@N@iMM
z{3sb^hwFj7YVwe`4LcZvc{c!?@36%qmaI1RBLq4xW~&Y_g82MDu+&hn20L|vT?L1R
z^vq<=>lUC3mFWkug{jqK>%d4zG9uqPscZlWndBAFh_lPAjUF|sO;(wYmRXsTxW9=l
zSGlq105KdI1WbY#P7ugg@cJR4N#TSOZvyZ*>q-sEUr-_)+#q6Ov*4eKJlqP;oYNA=
z3o(Ne?F$C~_MZXVRZRJ0BgOi)r$Bd=NI_hu6Ndg~Ack>(S7T%7FU9Y~H-V}9<Bx(^
z4bpJMihm`=;2}rxVq|nG>H2!4$Rb4c&je+D$80Gn6wfZRDLfc8$FEC>!IZ~STr9{f
zW<|~qo2+pQp!<a*BlZK|6kzgn^F7#QD<R(RXT}jqxJ9x?k^p*j3Gp6WD)RMPi%QTW
z>!x*XiQux&Y8fvG@t38QtH;8(6@>;W^F*6Nh45t}XxHP%xBF*=1P%a_OwScU4bqFi
zKWsXGbCz}<i)AwdSyo8?2%!uHV$K}^<8zu?SGtR|iIU=PvSIK@T?tBV<}k?gxJIC7
z0!MRnMO@ji0wmbu!Is1%#8EC`dA4Hihl0$z?X{}H#N%=FRD1#?4^26GU8h_g0rVMq
z>EeDNUN)18t0Gfzod_4Xn;-n~N`s499%+chgn4dN+Pf8EQl{X4x(8HyDTnTd^*1a>
zxcRm*OmIp0bFGL*u<Xb8fusSSF+JA4p9J03^Tk?U@?1SzFoXr*B`^fZwf+x4e2lua
zKmMtd`FPT_8PWa#N!2(TSEz0!7y<@Tt4xe3O;tLGVt+Cmr#}eb9<)|jTO_Y>DJRrK
z<rvd%6$CW+AG|Tt&SPll1~lQO4Lc3aUF3aCb!6^M>b~~)hahL`;o>%(E#E3Jte2P<
zgn;^3n8{RSVS(?z{1Av)hpmVx`ML;h73eQe8^q$SQIJbwDvXvRt`cId?BrgvD^`CP
z0Q#U)>5VOwaEnG2f<Qa@zl6Ek0Y%zn9Xslg0vdPS{IMXHl#2-VMP2b|nGMB4TBXl<
zv?$5B6;G1Es67s3l5@xx5oCgXuu#TS&I8R@?#ZHd^p3s5k`6u*mZwkhJ=B}yp9-Wm
zwPv*|=6{NWTdvncwN9@r$j#K<JIAB;*hUx?DVjLKeQH7+@iZWuf+9G%6h5Q0DT&%d
zm!5O}?i{R_qHWx-v~zG}E?bx-{{fPxEt{Ve_ejHn-3IYpI!zq*ESPv&%Vj$TM78Gt
zOckuexiL-J=1I>NHKy5^mzo5R6JK7q^lGNEnmkfhNieU(mI7TX_dP?|;(BYL+O|;|
z!vs&iKvI(gp^z)UxcEiVCL=LWhmcTW@=HZM<A)zqI2#Ey1-Nt|;8Yy(GDz9gWGDHb
zSHMgat*5?Xs>bA3i>w|6`mp9}6Dq{oq!N1w$sItp(-VSd4|aRKK%tWySL8`E0cDx4
zp@Wc!`#&f_Oj#b`=!A~Xok1)cdH6?_x;L`ELLIV~5cfv*PTZGNV_#t=Gg49AdW!=y
zD02>!U>>_G%KB1x4<6}$gR)+!M(f?NypsMP-c%fv;+sO;D&2re`q-+32Dp`wLBYyz
z7P$!ANYmS5?-C;FE30Q`{Prz!u?mmyrH}M?<`<l?IehlT8E=E-g+Ue{@jC&|#Q6vo
z$5`@aVQ!+I&xy7E1u`xjj&k-#NuWi4ws2RcaB73G`M&{}&~>8PHRIb#d0Rp2i&`6T
zDBX@8`JbXb83Ckw5H&yHCIyeoHWd2s7-cmhbr9DwF#`hB34-&wfKDU1Z^y|3={;79
zaj*U!=ONDR*weC^{-02Hmo<&+UaYU)73Pm}^5M>*KjwQENZUyZ5Bb!UfE+0Gb^pg$
z94f>maBxd+4lMLu39*ea<Kyi2flM3(I1cpplz^@gi5>D0IcV<;Uch)tf4f7Bx44ex
zF8VQ`+A)h9mngxE)l-LEu<qnfNYfWQqHyHW_*dgvVIC-*<5K84or@A~DsugdhQR`H
zf=!O}NS1t|?o<`-9KtHWNeAPfCB&pjkEABZ&Bf9)-Sk%D4dqK@n>`(K*rJ|#AK~5+
zJeC&3@1wQup4eUqx}%k~+Fgl#l!h1s0|RlP{>~1xHX9wpQC_zMi+si}=1p!A_1f;o
zLQG~%XYdpm_`-L_JCKYukR4tsA(3eimxlWmAeZ?=@)uG5bi4VEP<Ly9PU*g`5&srw
ziZsv`i}61!^T|KX7U>@eG&QAyN!s9rzF1(oU0~DejIRsxr}2lxP9(+6-%;o|MbT8t
zqb2@Yf^uHq6vPqxiiKQ0DZuHEWxu|KL20Em(<zSOupn0{X;OiD8gYt1Q&5kMbH>|t
zQI~v<m!uk9NujjlVnFVS-t$1<X~qYOJDW6ADOO}NHd>;{R&8jXM+8b(o*d~-;<1>b
zY`GmgNN7*w1Ues1Z;8;GDxX<mjzSghoYP#w;UJ9_Ls<_tICWns94Sz8A?9;kaaJyZ
zl&zy&Tv--LHbF_)dHhQhVM)ecjWLN|f0?A0sk*GiErT*^ZeNlgU@bOak-0ZET?)V*
zfpnZHC4N1&G+-v^Pz5u**kBo#gP=XQl~&?F0q!fOHUzR-oG&P?l&p;LVwoX@=frgC
zpk>M6AwU_R9T1yMVXkKg!4(dj2ua>WIE_Y`%f4$}@hd@QIb19oNRzsU|4gX6!!lg2
z4v#Ry-UvVs0zqp>V%6nHGih79Q9P`*d{GY~&A5&8ZbE%Q%aIM+3I!at*8BXg@{?yL
zfm<NqMk@fAS_3=?z;w7q$hfG0bzwgi8=R3gToIUdjB*nwgQg!S&7k$@TB0{DQ_}UI
z=;m@RZV*C4XKqXgsy$aGVd^1hq!+g0T47!sSxb6Q%5dCZ{1ZN&7l~MIGJjTHQ=y|a
z(i$%*od+3jZHO1EfQ<7mG8UY0g&=3bu^x}&Sbo)_nt9j>#x6>`P`W>4w9XRbLd-0!
z4;Kru#8Abo?VPWXgj0H|IDo`^N|~S|sHYln+iC!=vadoYDQJAf>OhQZI`C-a)E01q
zKo>*@j*G<20&<shq_4&@Yk;`ZOz@%jjnWk1CRvg_4(<YS5t%7)yF3ZREX4f7=3X8|
zzY~~86tyS%uZo-j=AR5~`Xe*tFqpI~;m$sS3GDcSmS7JH+7FJo5NiQdVN4`P-0d`G
z0`LsX5Y37&Dx2Az@a3|3l8SRdp*f@6IM8A&RMwXsfoAT<BEq~JaAj7j+e?|XWZ5!v
zAyTed)+y1Gg%WGn;^PbeVj;V2Z=6<UB*MII65^b-$OttdJ~tI?UeUTXh&Iq!Jk@ZX
z5@v(`PL7sJy6dvQO%XJfS|{fh%g<KAgTo~WPpZX@LOh<mb*!TgS+C5IF4y7?A>)KX
z`s7>$p)Ot@)C1M2Ye#Hw(*4RM!p%bUF^N-m2ypT8_#V#-bwhf(nkXmY+Z%wm?Rr~#
zNO81x6Xvpg$cBi4a>Lp~Xy#@-b7iHxpCEr#@fx90nTK-)lOL0g3@PX+i?Z0_PC;&k
zR8cq_KUkJ)t#M4oV&|PQ6UPQ!bj;q6Vs5IQM&%q%&~ag|3Q11k+8-*B<SKyA|3>6-
z!;|I>&1kP6CN76C6hSAJ8758hJ>s!bif=)Cy=>GLhff7#%n1C=!qtekm4+q8^eWc6
z)BGbF3wq{xF>N~OJo`|>rKiSY#TkI^0hu5O+Od^{nTD9<u;x8oW+F_tlo0*#tC?i-
zh+rp)G7y#HfKhi0hp?nmx=<Mnv2C#8G~>^L{80%i>4o)&g_&%+fnxD`RtU`)K;e;=
z-c?%wEa>STPsehd)mRiZnZ++|40rEXfTb$87YcRQt`LzcaqMi;d1Md?Fc8^i<Dy(r
zVKJ{5!MU4)y2U6i24wpE>1IHtosN^0WWSU^FGWmI4n*v`@M?nn861{7I(TAn^P<9Z
zc?j!nD99Be5pr3m;rm7jrVEnfO?O*aShj8QvHXW1(+?R@RsyKWgwa{7t@Ye<?7Kzr
zO&khk2pvZYB8%=Y*>K~RO2aJ};$kcD4H9`wuu0`po$<?Wx&RzRq?dVd`L{?znB+7B
z-I-qh-V!KJ%p?IL`{2{CRZ*E@I&xpiD_PqVIXPA6DKKtFjG7L^#MtAog7JOInhDE!
znY-0w(zLH?80S@3LR-rBgt<Hp5_$Kk3G@Im-#I0>M_rHyrIVeLqsvJ`ykd4X2jhZm
ziws07l9m4a+Z70g!GtD=C4^|jojrX{RF&{7K@i6D_-O&mU{pp37;h+L!eB*?C1qmN
zIXi&5W!><M^iQV?a?9A!^_k!ASY+a~!IC-@+wKhHZgPTC>_2$ux?M^Tzc{uyzMH_z
z8<nv!32zfK_6_&2dUSK2IiWQ1wb^w#QKlr&?He7G16kk13h`taOPkU4`m(cO^WVlj
zQ+y)GW2VbG*dUo2?!0S}lc~>!U5RfC8ed*ooBc6Y0E9zL#bsEJQmzO$M3il*aWN{`
z{o-m|uJ!deQ<$5<70aj=jc<G#2x>A7l48Fa9YUOGpaL^+COSQ5g0GR_Ww!jL5HrIh
z9?{usch{mfeRYf<ik*beXk<sRX<O_fV7$caSZ5EH&|<5%C+o5GcR-kJAbPqhPM+%;
zNLP8iZZKBZv&b-1!>ao|rSp;?Ef|`&W~}mEK(kSLTuW~WVErtVx@YAWDVr#r34xg`
zn*Z2G5be`ripUvFC0#jtR}GhW7P@Q8VjSdJWtwtpE~XbW$?>d`rh2+?lN#}ay}%eQ
z>56OdoKo&4?uCSv*t;l!;h)yTfw)ANJ3-`XRNrYXY_Jcg`GD{o%;lDS0T?@U#VL+H
zF92jjRF#bx4^);2Rdf7_rT23wsWfdv*pd-S$-#(q{s<wa5BD&LUOh3gKadHM^yL(7
z#8(dhglO=o?gI8x(#3qR7>tDv1VY{`B^X^6vi|O$f8e8%q@^|(!a#{o5BkCUED~~t
zNY{%AHf0sGQbEPzL}j%orI$!$yMW5H9_I>oUUdDa(+ER*k+96xiao)Qzga@OY^BX2
zJ}FBof=ue<*S<$duVS;~7fN^#*!ytjS*i_y(M9xvxEr)PFTr@G#NJ9|F4G+bLjSoP
z&Q0Kb8gOe)wc}dWjD0%!%@pP`Nq6qts{Tiy>%j6)i(<Sd$i1TyNC)LV5@w=dz*|H_
zTeu6@t?k4imevZ$RLHlvs|!lq<GyH8`8&&xq|<1O|K2ilX6)Wwl!v>RalAGx$jwG>
zKwG0r*+&GL%iAk>&_9;bNY;r?|54fvL$!c&>BT(-(9IUzjQRTjXlZGn5pN5E-?Ivx
zCE=ca5ZcbAN;6(o!rV{@CtA&`1K@5TD@0E>ZRjy*LbTx}sSUTI%%_tg7O@P<1P~d3
zg<BCMI+fF{#}KY<Ai7GRr-sv6^b7DN)n8{<sm7>~yfCq_OU9xDhYNMvuz_Knf$8G+
zWmfJSdfQvm<C2>k<c}u%cf4GpOMv-6T9rD2R=g+7-(~#J3&#0_#TRtz!W?0lI(rn*
zjgcgl)LaC+od+sTXqnX#9ZHy*=?;`>l%khBE!>3DYiN3XG(l!QmQI!#IYMOEIf+21
z=)3WE6?RTN)|ag4VDeYXIy-CjI$F9(PM;GitsIfmn`FAVx_%g62T)#Ug!dL^K^I%U
zsysflxpUy?)GzR%m{pNmRv05~dN+m%yvOa;VEj#>hm%EQgeMLTE&yDEdOPB~0#Ki>
zzWVC%>qGe4tj+Wrju(_R$ysydoO0Kp&OtB@Ss_K+VWhPIu;kX__3x9)WI;D6tI=;B
z0Yn2DO#%x@vG!d7Ccegp(e<$_;r@2B4|=x#Z@x5+`r={2O(a=1;HWbkX9#mU8%=J4
zaC6KEM`lh{x~e|qNb;CGJnz&?XiQpnrodboO-T~c@kbR{a*)AxFjhGlj9ze4M#Nf{
z=Pfhv96r7nu`@#jjFW0a4Sh|BJA}RWKy3YkB2BEYTHL2Jq~jV5L2u5nq|CO+K5h6(
zN|N72s2k1o+YlBi$m)+CS7b$X0Q05DcN0)fDcUhOM>wx5NG3qZnEDA-;9l!yg4l9x
z@nng$VZWV(>}P~|I63K|;cUj+C4>=8IxjDCe346r7jo6SQji9nz?KN2T`_h7Kpr5h
zxrirurVzGQH02#gzSR?50(uT$^E(uW3vdZ^lX9=9o>XM%#@e|n?mW53(owBSyT08i
zMHWo2Ik_KmY60+?8kgmd{-`L^nv7&)$<siXjmerbNm>$AIL_>&AKZIbMHUcIFCmVc
zm>l+C@9JEb^XU+(yZo6y2J)P19qvI{bgcjng6{kXC(m|uhd>X4?B6ha>4`VXj5;fk
z5k5NhbTUE-g%Os`d}rJt#P#VeL)JrteHEBb#JG$kxr+4=ChGhNzgRpS8*Yu`lytuk
ziFkyGhsp$*`BGSKA5a!O3)G*Xb-E*lD$oC{C=K(q*RUy0kXJsr0XG!T2|a9rU7UD4
zRH-A$3o7njbN`|1?5Z(qfr&SZlTeo=qhv~uvNK0G3W_}Cx2s%T$%twKG+0AF!S&X^
zl`GFgtio}1S@LWU&&1BkXiwa)v|HAP*;hI>$MXU`Y8Y~HW53}!0H%{{=yl7pry#c;
zyL825ti+n<7L`#IGeR*D*ZW10vB5ftYN-+RUltf9Rhnc3;uB`VQy4*0R)<FY8TMpH
zhDtfdsi_l{;)pcMDm6XZdG<O^5$5jm2}P|s{8u3CeIp9y`@9nV8Ev2*{G0FBAaoqX
z9vAw^{a^k~kp)*kZRIar4C0Tonkv)?hnO$_*5#P8OdB@nqb>pP%8s%_LAE8^H{uJI
z7C5gu6E;3wBUX)@Z_qxAm`NI*U)c?Fe3pZBT5%KXHo&zg&!qiWP<c(i6eZnv=^-?a
z9Gjkkge_mTL1Hr}nuZJ&<Ez3=>V8@7%JO{mGA|8CXUAmOx~Fh<%GS}5*h_!=6X}`8
zj!JZu5KaX6iZ<d<K_<5LN5tY2g)kP}3=PEJ%M!RA>v$P2l#uj{dcbI(%P7M1MQM*s
zU0kB1sjPr9LtQAiv?yFF+$~~^kocQ%T?ujL!@UB29+wxnn7*wlxY6%`+(?YC(PqWl
zLg?sB;@n?!Me*xE+S238zXvhR$t>NA2)VmZ6LQkbd=&k;FfTll2q^N7l9}~%bRCMg
zvZ#uKJPRYfDD_xdsHbap9pSg*D$?#*b&RVY)Tn0)avOL$Q2I_CC=%*6pn#W&Q^S}G
zb8$j^foA-ytgi!KEX}xc1-WTFFGq~_IA6#(yFtC=HF#lJl7n@MAN6ZN=2qVqh`XC{
zS(%ejVspUuuB?&A0eAzILw0;x=0ow0y)=e@*A#8(uGHj{-Xp|I6LyRYY*T<c#T6e>
z2>aq5L0Q_UR9a*0Ye}<m_SX=zW6$eIxqqxZgvyN_uLp7eaA3n8SFEtr4S@a-3kR#k
zivp<CE~hL6O7r1l3((P2na6Jg@=#!eghndO*6BBax_WHK5uLSnB+#uG7-qw+7|A2X
z!4PFu(-_|%AQ4B2q#l@2)~4H2?HhgZO@Xehf;kb)y_*n|B{_f+uV8OMu%NEI(S#mT
zmTA-w6zg%15VrtriET5V5=NJ3QDQBdFVFoGsJm31m+X=li7QG_dlXtKwu$Ejx@D{%
zuzIY0GYE9Vl~WsbJoQ)7Gz$g40ueWEC21yNM@>;$B;h}J8*t`9OfdaK*FysH^2mCH
z(*{;)yss<RLx*!5$I`cN2XG?@a*IwOZuuLC$-`8@myd^01mzLIX&ytI&+Y{A;e<!h
zncMu24V~u_?&1eC*mT{oucWlco%xUb8eW|Jan3!UCMRkozjWWY7YtTxS)YX=UQ;T!
zi{%~3e82mOnpnA#CKT~uM5xy}TAOduVabHM|C(4BV4^{5H|)+a5V*+J4Y@62T~)gT
zr_#>GdRrkJZp_n%Sd#1f7w0QM>-0%U|ELx22hq@RL}Y9*J9Q4WJ&bA*i{pV3L3WXY
z)_Wz=ULzn!HP(25Qh7c!kj@dE+hVf7+?Xz|ve-jsmynE75C>wl2gxN~LEx;8#yd*7
z{~TM9@wr)nTwi+(czhaiOl5`Pm@*NSF3^)Rp&kPK-*MY_K9j1O=m~h7G$&$xZRz5%
ze<H?t5PcM;ChO?SuC28yBZ#<INFIyfb|mp;yjW%$sU_Q>4|Aq=<}Q{sI=Wx~cv+5!
z%L?LfmmqhGhvJ8-DQ;rScmgy}T12i^?Gi1*GGFl^EBu}ia}dH6<Cj;HYn9W@L#fO2
zAbqj)Q$S|!CK~*>Lur2?U$PST<!Lb37@Z+gc5S6xf)i|OT<{OlZYat!&g3$Lj(h(J
z>QD9PiWv>}eS%yyF7h-9=DlB-TZ*ckb11l+q8};qiWwyd#q+{o2hRAi1l{Nvl5WRn
zKaXW>qcp8V6s|Ype5L#`PFcJJK&2}_3uZ!bH&2YWc=S0Cw-Unu4jqJ?ejbRg<RERE
zyhzgQn+$?d+_TpO=52{GT`~9NA{WPQ+$XdX9{viTKR1Td&h^SsLh`bPYkgnZ2)BN<
zsEk**t7CDY05^*BQHNfQ*yJ@J_qEZ>RQRFNuBfNhj($N$`tQF1NXti)VOkvf7D+df
z#eywo`+pUcXupj^mCn3G_>?3_w7gSbc$-R0hQEaQV|p9Kp|EF_pxg<#d0{jYm<i04
zCS^BOwv2F$8mq_V<rlh|0~~u;4t6dvs*-x85xWXA6^4*>w9@TUR>pM&*Xi+~5LU2*
z^rqB(|0V$gP%YG$?>$mb!>W=v@O_eQ2<9-%^H}Z!5NL||H!7ZHtfMqJ<Oj)p4voJ^
zNTwiJdScE;F2V}f!$Yglp|mN}K}<h%QL=p#=r2_&#Q(=!Qb?W)Qa&U9y(h@DZ{k^}
zSBT=HGMip!X@xK=e_YgsMp(~^ts}&rR3?XN*+5`kE4Y@$$U}?Kj%8s?I;7J(y+oM*
zy5&Brtd0OHqCE5DOP^3(X23d&B-GqlkQs1<C^d0dnFTSJQ#ZcFpBAM!DWy0AhkOPE
z0dcrwny?tIG2h%IU(Z0kKJ^RyZR*x|q`ogM6To01Hg?8SN|-<hf_VP3ru<?40&5bk
zveHkD33JzVj?A<6$}(p^4?3k&`2-A!f{~)lSC#papodIv%)RH7!4f?bw<w!;9!}ai
zm?itF`4<58NVhc+;5*}TAzsedGtnFBxgt3C0pT7sPG4E(d{>}rO^drD@W$d_ENkU)
zA_nO%LLf}Lj6lD;kP}l|=1h$%l`sv{t#N!l5l|$K&P9qG$RylKKPN!TWFOHW;&Y{#
zC8%;m;e8h+nWs9k1ZQxnn74Qd@_4haG!y8?x8om>Oi0G_df2K1+~*H)nKGgLP^H>4
ziXl<mplK&(YQ(k+^qFxYCks&`7<R=fO8~oz!@bh0P7>fxD8!Sj&*JxA0&{IS9_{H%
z7R7q%?YO$0E5OB&+A7Ffl{jvxqL^YnFfYfJaWMjUYK^SL^2-8vMANAh1HvSje@pl{
zrD{YB=3gYxT`2UEXkjMg<(N%xxeGU#8C<px!nVwnvfWDtU@v9#w-Pv>F+0+1TVe$L
z??=ev3{bX+aK3c0iB!6mz)WUj79O)Zutapt)Y})Amz5z=V57*AwQ3aQcpkbPd1PlH
z9*5+i62BIZR|x`zNZ+#Paz!bQh&sTXAjGBcxo5tfEx;r~1}2cGBq@A!mIr2iX{~hi
zckyo5NlLrZysRoimPWkx)uJee8VpRCnh<Sl1z>lPEkOsf+gB_KbFSxV3MwwRa#@($
zYq{r04fT~(fZd5I_msM{N{~Oz<s9cC6p0D*D#OtT4oH$`Ld^r14H3spd)b5<eOa7K
zJ0Dsh46e~5#nSq+8B5xX#*l7%SflqSTLe#{_rxAyRkC{VounsLPW&2a6P4pA?0>s}
zOf6y^Lh1fkV0AENwj9ejGak7{QAdT4zVXPKq<z{+M$wK60`Xp)%nxqYaK9Rcg^M)I
z3D0cUOYA5S`eo&HLprO(Z;rPGxTSr1-E-r$NYjYc>R63-o}i?M3Svn+b*2#4&|MuQ
z_Q$(93-*07&tXmaFxN4_g*7vD+Se(O*^ZlncC<E|uII0?Ew^?L#6C)*#F<wcj!EnD
zw@Ht_K->m%|4X1hN=Q_6`&VyJ<jSsr7z=zL(7F1P%imEO7P%_z2&Fh6H{&LOdDTi*
z5=ku-e=1?b;dV|S=}znaQFSKpSru3RpC=&+D4WZ;vPdhfw4$P-Pyt23iWMssRII4k
zAR!xM0YXcC0tqAvDgr7ht%$4=7Fh&j5i70Q#A+*V!OHtW6%{L0th8cF`~Uvt+&nk`
zd_I%=+~1w`%$b=pXJ*a>*fP)Z4NF?Z{uTMOWVr7gIpkFVM(84JCWn)5zmB5|wWT#7
zNb}nmMmQ};Z=ppE|GPL!PAlf*_|q><MJdCJVJYNNVWb=EFsd^539!7OUuGgzWRT?z
z(STMoxKGLt$~Wc<h5lJV_+bT8M`1X<7XcEInBr&qd#}I;On@{A*Z9ZvYCOW0qcUH8
zCGU)GMKY1_s9sE$&)}0Q9+8rbCipLXv6>9j51aH(?F&fMs`-*}{$afmRWO11x2t&3
zK*v77e4lkSZ^qVPZHpaN;gH)eg&RvUkZ1{Ktq`L>t1oO?Vxs=n7?Zx1{<+*=+b<Q8
zZGX(W{Av9`jDjq*F(Kwqv>>BkcIi}3AhrELDin*%>bGBv5QJ7l8gcw%*Wibo7ir?+
zm+H;vRjM<&zlpwTAR6n72BZ?;v=S}qBLUC=Yx$*q*g#&abV%PiI<-)Um5vFQuA0g7
zzqk&}Xv4c*SeVUFQ6t=V8`%m`-G;vDf-MUAZXzL*Glf~vWg>?Onx7YY<0>)M>X4}a
zB+U9zZ2{1Q4vdfgjRSKbEqah_|AWD)G-$VIyiszFxE{<nKvw3okt5Ilc?cl1%ALw^
z=`)A&X65CO_?7+p`tC7OhCMl$`j;^Rv1NGb{2y8zmtV{d4?pK7-o((D#==@vu5UjK
zAWByTSlH;D@ACy35zup$1lw%2!YuU)t|B0JOK(mg<8^=ubG)~J7{^YkD)J?_@=E1G
z9X9ip`VK!%1wd88sN(Mtf)BQO<mFW7`wMU5-6$RG^T3@QZwInvB(yhX996qm;fCAw
z$p{{lT4qZupcV(4R)Z`Xpq6o&X<no;Xyvn7Hwz+Cw2G3J8O}@5TaHvhR(90)<JoK&
z5F;D=k1E(24Dm<Fn~@E1ihTtTDMwgDU(6wD&(1hsBGB>^3;?hdOWUVk6bv<(FqE>8
z;a`ozvYAquMuH#t6FjwaTT>Ac+(-Z`uS`p5FU$FYZ2Suoz@QCR#rD<*%MJQl1P}F%
zBL$Kyxj-(=^?OFyyEG9h!Y}UN-AXUp+|=_yAr=jTD93&k&3^={wpk~XTKLL4Q^Dw!
zHEZ2G8U%KxR=7gp{JVK0-(g1(96}?w@~1#np_&_Tr<iuR+#u9iFsn8}FbdTlaLYvz
zri-QFDKyUp!jY9zrAjo=NW@YB`7LS<aY#6vTOFcwtomphYCc>AF3h51ajHBgtpD5^
zgPJkrvIDrn=L%Hur&gC@ip}r8Clyk$A<{88|4SeZB&>av`HV5VTE6H7#!9<YCCI8*
zJDLmqj2K`9#s1Wt_oV_&CPx|*9~WeS=vJ{uy*CHIiXt;1Sn}cW+*~lIU&h*BssB=M
zR4RR6C1&ub=6?vWFyXSVNo<<@K^~YDU)xdgwcgYkL~+0tFZCnAXVg8IEIwRe4WaRC
zwsWfxV|&;kW!vLtaU#sM=n<tP-zm^YL#3B%^V>&)td!O4kelSc5MU`VaG`4OY55>g
z8R&i4Mp5Yp2#cyvT6kuPSf-9W!okRl=$)P$i^r%t(P>FL$vL1Pl?bB&joJ?rViD>d
zpQ5mMpH~RT$iz-0w%qB(VnP{%C={kEe-cXmTP%zEM~VO}|FZLmz}2$^#7bR+ail8N
zar+9gB5^(s(S@oCnB^d}o=MrXx2s~0O0={L_U#{IjK)EG?5`1KnM2f<crNQV%Y_<o
zF$sn}$E`6yLqa)oMJYaD@UUfvPVwgxQt^~?gbx9T9@D>E4r*n}W$O(wpHY!Q=)g6p
z=_16^LKs7SSzHNV<tZ*J!iWvMf$uvB(6~k0D>%3w*z_vF*0*s^P}Lco5C|X4Kefai
zA~G3S?T5?1(NCR_Rz3WP7(?>}XS2y^q%g}kLKW&-KS~IVmCj1)syp>+xn?sz#C;|P
z*w&1`;Wq-T_AvZesq2QGnhX>%pH8mol%n4vkcbiNDzm5hkyBEBka5Z?tLd8kHv;h^
zgNiWr&DFd{td4lhF5DhJ1EE|g0|xNLML!3#EEa3Ug`Hjia3zzEQlB}^UO2>E=s(ko
zaRJL-OwmIhe$4}*My!%tl2Nk@Sk!8n3S(t6j0yoJE$z$-TI~S}Yo#Y3J02_iWI<MX
z?tVv)rg5S!_BV+dm4)VS1X;dOs-P{qG6`T^J`?S@jz<fp4@1GhMXm4X&2q~@1Dyb2
zyIc(zxrpwvm6)(Bf*?1&wd^?f(RKJDL20FQ(8v$02e3kN3}4k7HbLDe*r-%kO6P}(
z0XV-E3Ok``3KQmlMyDRl=32@~+s%r`GOCrsK!)Klh9pdF!Exh+MI=;L%l491kYrE1
z`9I573su9GAcoFFLkf(n!yzl2@nEZTHP=N1_@OXM!%l-%`oj+y6y3DQHV6N)pr|$E
z=+K4#n-H2mhbLKR#-jRc-l+;M?&Z$oCcSC&({>m7$a!<@EBh5ERQNX^=FMV56yWfH
z|42y0sT^v|_R}8$fK$gpec$O(ULxky7E=USx{W*Lf?6HfO0BwbcJByosUvM+fo_p!
zBRMw#a*Z==7JM34;fD?22s4`f->DdkBy8;smVh3LA+|7-GTiuQ1X-4`!3KB#N`RHB
z3QJtj`=!_m8%iP44gpccGlvW##hw`B=<S4HH0zSbVt?A2qLtaZgjuc_W;jDz?VlH9
zDROt7cHk|TXP#OXHlfEkkEc8_on{*5N5lx-g4T2IcoILBFD-!5vySuMUkFAGu(hh#
zcYTT%<1xCz=+c)Wff4<XA+#+nXitiTr|8EuT`YaaUbPOW{U@OgYlU0FT1=z+$4qvN
zY42*d*=m_EE9q2Jl@#l5LM#!sRu-yr{Z9<BnXrH{&p-HdDqLwHoAk2%dLhOh=(coj
zTKg(8o&mK{1PMG`5@PeePWoVd^Ca$ey;ZMpIWq~Q{=4;Nx#u86kg9#<FTpIQVZSj}
zp`HbSkU6p!|J35tXOwF)r#<pHkZ4H2OammKbahJrqfQnXw<?;A&J=8gMy3qrK>}oX
zK(IgyO?Orwho%9BlL?x7J}B6-R#1jwpu*1<V#F|me4#6!Bh+#fZOx*em>UPKDB$E{
zk$+5xHGsm2EZ+HF3NU`4g`skgDId80L>zP~({?hL@jO1Q9PD5XL(Tny;$~Z38uVdn
zgj$xdLZc%iRZCMyF2~GJDs5g4V)<a#85Dha1%M<IEw4e$6)Vj{*!UsV?z$=!2Rj3$
z2VE|NYS#3CioM|lUL&FwmuM>-jcs)b%i%DhL>I%XnpB%Ku8`~C)yl#iXL_t4_yjcK
zSTqjY>)bj{_LQ*20p(@zX4J<_%Otk(uu$77tQNd<amg~}$BJDwH3!xTfW<$ys;H{0
zDBG`y^DNaUTLbd_YGF`^>5;Da@~^~xI9LFQQGLBF%%XyVXlk&z*a&1bfyHU81o$E$
zWB|Q^nXS5bO)8Xn*iw$1<q0A&w8P`IGv_ZGdAGvR-kB)LZo?a!0F7Ky+%SolO3+p)
z6<J*@g<*t7JjT`T>Kf@zO3k-r54#*w8B*R<td<lt7;Qg)OQ02uV?kw_zPu;M>QIdZ
z0XoQ)AI30iKFI&3pgkg(4nW(COZ?|?bX;~4ltZgG<AD+*U9iDyoPSgAmN8~V(n<~7
z0%CbEc_Z7WZA}GYj$P#s=-s@~({RR^rL1j0R^1b_CsDrHFQ))ZD?_jjJ1i0vK`Uu@
z=pEJY<*Yx>h`wQ&4mBgIG&}ibg_eS`6uCvp;=c>FZV>j)fjy>LHyA#ce|8G#csY8;
zO7kCe-#TL@vt<8N9G1~auAM6KX}bW7BK@?bsGZ)8B0228VP`^5Ay%o>3L{V+Q+HVP
zhp|u0eQ68xhzLnpk??DU7^wgQsV2MvWVvEK9*(H{3A@3_RblB^NB}**`)Vqj7QwYz
z*h^5vhs;&z(sks#f9#QE!yv1@Cx{-sq6mfjy?ZQXDtuhN8qX^NEcv1`bfcR6f9v&B
z)KxiH1VV0v+yOP8RoMDQmyy>RW8sy~1MEVDCO!5SbA&)I#@AT1Vg^X<`N47gNVIff
zp~o2#dqpG9T_@UWbCW;`YQ;qi?0fZQ<de-ys@7lo29VVT3ZUY!Ta#5Be}iz#Z5bjM
z173;0BlfE!kH{(7d{XKszL|=a74*<4)(EoT+4OWY-D<y37zN{mYO$~V9WN1In01`l
zLdfdn7heV6%x_A4$DlPk(fkKOEK0fv?W`^HXEcLZ?m3`Ci4gfE=N9}PJgVH5X3l9j
zp~kFZt7J`J_S-3M$WTEoK|*)D18NBs$Qz7im_-Ov$rirYD)_X9!p)0}b!tS}21jJ3
z*OMalPQ_{Axe6I2WUwW0SBwUOH1wnFFs>46!F7*?w99od#PSSR9hdpX1Q}<j7d1n@
zUkQ}NqDwS=k3Xa$LabE7_l1It?BfcVj-r?SW6D!`l}tTy*N`B_N0i<9l=7eN+N;cN
zF!9RU%d1h5BS~zVDi&Zl(IxW?AbC~((f0tY5EGc@u{F^kt1QMpY>Y_B244iPMjIx-
zy0$^9A_28p;6}l0?o=sk?NdwhVP6wQN|jVg8`0{OM*XJ+T0t@YPfzHV2(bvmL`6Es
zWx^~=Y%Pk-JiZ{@axc52itsNnREeQX9Z~wPAdPownGmAQpHrc-brO{E=71o)Ol7@D
z*V3H-KJS*-aHlUlMv<R?0MK&AqyRM)M4$J+VALdLB4z01|9GQDP!Y&h&$_?xY82DX
z_i&RVhC4q1wf@}fjkEPOOk~3ciysXPEtzWHE6;|>s$7nWc*=)hMl)SulRw2jEX3$g
z!mYk+=k`klS=x;AEz&^IVNfd<(a|%AYa0YcJivI#X&is~M?l6rC@WF=-w*?nP^NRc
z*2nm;OvnT_i%*MUgiAZHZQ$P*LM&k?za;4Ie3D9u#;FY>>`?;>l@kPfKlq5flMzN6
zmRF*8Sdng?J*c1ZodI+L7fa00@0N_CDyJhI|5qR*E(^Hoe<Wnk$j&%w@v@P#nypga
z>)pZ@U>Zwm=>S=gB=w;TJy)KLT=^Ow!c3~WK&uiuS9T6jC1YE~hOvK{4c`(&IuTuo
zK6}RJmIsL;RV+3C_Y2-F!C(oi+TU?3g%n8Ssqps;u>wtE=P)%Ejs9i&0C8#kTJORT
zUjz`l?tQBAo4+(48eGHfj;sC-KtHI7sF}yonynx64`3sOMn;M6WrCuD!-l44sTOD~
zrF*o)dPs#|5qm{JN@u~8(Z3?l8Z5JCyK7Iq#U9}%c3ds7Bm1AcThEM+lW{@Tx$^u2
zUjrLqFmk~WR(~=^&~9nke*6S}j74<}u*^4plM19`o%H5!35gmMd&`g>O8o=>0;7qd
z3Xs*+ReH6uvt5XNB@_HDCxI+a#bs4_j5k5~y86GsMv{tr?%B{53?D!X&Y4Rj5sn|o
zpHYeaD_F9oL~Dea$D;D;3Cc-;tN^eKD>gf0KUy?b>-bF!F<!HIx5EEPP*gP>gpgvi
z((jM`p?0g(D)O)09WC@>4H=E*zajwJF&L01Pgf-H0Vz1oRa(Z5I=xzc(Egy3cqk2k
zzO8TAJpN!hZ$?apDD=NHqSG_LQk=sMOsq)?iiUa}58?PlST;(uOL;|oSSZUL0kV>?
zek~yminaoo{jx=jZE^;}5A11T8=;szP0>aDqWVf1O-PTEkcdW{{+!rbxPeSPLBJ?B
z?WB~h?lyU88SXg!iOH_8Q{uP6jR(ak6;u5iQT!D_SfKhV{a!&<*^KUGL05(O&{IH-
zEF5g0Bg^$8W5}9gzW1l%i+rGp!SbNLO|KeBw4{d(Jo~7V+i>#84!l}AG+t!JaOK*=
zok5Y*`droTO8>=aDK9x?*!4$WPIdhBbkK-frXYecIQ)Mgxkg`%5hC*$b!)-o0ZP;m
z+6q!Y|EkX7{y4(GH)z5!|02+ubv9?de3{-YE_SIR#v{S57G`nj=#Tb3{W%IIAy=&J
z&ah~h@MB1%(zn-plzdPdU_2-C!qIQKg+jkc?=Y>M31q-L=S%=AuvAHC=Hc|gX8}TL
zNWx^&->+ArCL)0}LK2VLpKVd8cV-Qk!%yd=yfNM*2?o{Rw(~(-Mi`_Amgd6!8~$Z|
zh-#e@Pt70ix5tnUf@m2qKoM!!po!!bVU}llUq(~9690<Ann#XvNI8o6|2RH2Cvh)&
zzW=u%BmLy;f?~g-OPmYIpp<^CAd6lKRpyl(^FJie!eJ|tn@jX=tzSB949j%|F#cyg
zsP&EeVniHEE1V?+2FynN>?d5n3k=NN1km{DZ}SGprGcZLZGR!J)~T|aLLy~2v(mjM
zuoaSxLMSoWW#0-5=bY8f0v7cu6hR8LAToVB4xhtTzHsL9dohNw4K3BcpslcKO~VlX
z=L)kD(<L%IpZXmDV~dJv47AcE!IXW3Qmq;^3)#JZ4D!r(Qyvw2FeJ#e_d+R4e(5Ca
zMVhKC?IH`Tj!OpP7|HGz1YLtMw=fG@Bg8Vot^zu;$9koJ9IQBV$;9^oAZA%{wK|i*
z7xQXV!@L&~2;BSNC14TXb7Z<0?R%bq=CNX`HmD%#29H)sjZZci^K%qDLJP||X&<H`
zYlTw3SIbu2f=lBRn1gEJq}tEC4A8iQi}%FfAN5JaG!{GJ3`g3(^xex-{@53DMgJ>(
zz7R63HXuD)Sf4p55Mr}8CrlgYuKQdGMA_NC!8HOtTd$Vjg?+-!PW63Lh)lmB1YUd<
z2%N}O=A3&jxtcdP9b1-kfO&qo-i?=}2G1^uZvFX2xHW9uw?>X6F>qgI_+Y$?U|-2)
zHri)zun|}5dvwdituc6vbEp*Hq~ayV=+>A3>Ks#gKai+j<E|WWS)m|{3}!U@6A#1~
zVKJ=4_LaBdpv+xlBh&Z)fyGb1#daMI*!p{gSp4)Q;T+?St^tB1h!q?Qerf=3a2yHg
zCel2gIVcqee5tDfe=-=vGE@TFaEU;HFB29KLP{bux!U^?;Fj(S9c);<Zw)eDqQ8ep
z185nWoSj!)kv-W@h|`fi56xzUUnk7GX=zo{ncqd;&}NiYFiY0H<j;j!7gju;)AjFO
z&%3e41Wn!x{g*<boE4OzGu0}3pCJ~EPOXrYb>H?z-eEAN0J_2Vf*<l~iRYlK(D3os
zn^RD}Ru~xu17zfZr^e-@eKpL;Bkrl-(6Lsof>+B#WQLb0KU0(zp0R}+e>p}KuxMD)
zfvdoQKJg=bT5y)}B5PXb2((OcuUC<_wswml-WA;6D##+F_d;Sr^PlIR78*5KwxhDF
zKIE2EY|IM7WgX86vOcMZ**{wA+W;VT{uEh0%lDPHn+J|cgniKdR{~m!Sz0zF7~v}#
z2}oL6=`~v!#X|6=6@Bg-vtA^CMh(sRb?2uK7#us1X-}p9nSxoPWhaZyrdY@)q69Q(
zmxN8SPYW^9a62Rlj9h=pC?JcT^AMq7`X0i}+t}>tpys(ukP)U7Ll_J@bA%WHQC72y
zma!f!<(;5Ln{gQVsrLN@SU^?*SuFA!g;+q{mxw%}u5o7+Q2V`@DwX?x+=Vye7|I<q
zEfT8>?gl2mR2maZc9>9W+sA-G{|eeO1~L4Xdbgkq>fr>DnofWt#$Y*B>q!B!ib0Dn
z4Jw0dE09i6*2ziZE5&Jnv0lXO&>W`})It*k1uPuKNZEvJS#3M>Ui=t&nO8D%A*!Fo
zSUA`qecN02rQ*T_3@y~3dOrwFPithXFQ;XvJTREBBJ5wwPa$-HFtzVE7R2%xtS+Nn
z{KYsRD`2TCL4{TqK+Cvn<Dn3nE^mfZ6c%ZvpiYQoT=pxpE!RI3M}~wIHMb{YkF2xh
zvpFlzKQGL<07W1DNC2$PG4-gE4J<$SBl2dca&lTn>ADtxP-s{-r(}x%wqEg|!yQ4K
z^p!#^=HlvH77>;RFy_yf9V5oL3f=d%Eea~;w|fct72?yVquolfO?15=OK$39tU!(D
z)riHttGWaZgKM290$SG5aj*j>n_Y^AnpfBm8B~+D;5DLa^*CnLt-WDJw4AkJ+DNMy
zbw&mOUtMo2lyL}yK8TeS{xqd-nWozySNZ+|K~^(nMu@@Vq1aoFELtIvJ|QToQY6wU
zmK9@9=ubGR$T{&iE(!1&Vz@6i6+cW>BBLJul5#L3g4ReY%*J1?P;#QpqG2uz4PK2f
z=&r-LC{2oe_X_+`X;zK!$07k%Dwvu1=_LPH6%bWXiGgug<9m~MrH3oWVqcEmHkmhM
zt-?IYzrLE6mhz)9m(;mpY6{CM(^m1a9K$Tn?9yS`JnXD*<ujX`4w(g2_*DvR*?}*Z
zzL)y7g4Crz5_LNpV=rtc57W@M;xswI$vlaCd?3uyRCfX-d%NF6k&zgeu;u$HKf{+X
zaXw9^)E^XJb(LGpYD>N!^K%f&pNIAht^c&tcW!r;hUe1<KrGA1)H<i*cPCR2f*UIV
z`MyVu#lj?s#or6{W`vU*9xO}@ihM9o(eoAQjqIZoqobhX3kxAH%`j9|is?1FM2T)g
zYg4gJt}363GGbh7ERPNl)?xWZ{;Z``RH6Mt6ZHRq*_<#d;%?IlKTJMbxT0K!!VI!3
z+x|azeMFpael?PW&g<MB2Svt$rAmE`AZk#vdUo>p&-4yOIiZW-d!R0r7CA@OE%Jv1
zSz3shl$^;$hnXpFGFiub!k5efB3_LwGzGt^-rm*Gun*T?(ZGAu_vzLIzj_XU5dewO
zZf5YCVh}+SaYn-85rLKi77%m;OQm5Zc{AiiH#AFC6Ff&8FoC@z^x3VTaj5*YBP&2r
z;<6H2sC_4`(y5|4wJX6e<6K?ZEJ=WN|2O5xQr93Hl$Aj*ddkCmh=y1en6zn0Gp+B&
zu(T)AT59Lh6@o3VVK_$4e_D{0wW>6_dhdcqVs8l08e<O&GHPm*W~HC;r~$eS96JQL
z{&)mHe=T0+`Jd0TSTuuUcZIAFY!wu>ud4E~?3I1&3G+lKO+B=uc%C4m0Ed)yIg<=n
z1l005sgP~9SY`50$fsen^1?hBFj^?g#vv9SGy#{9dB&&MFEb&@I(h!|CsT=6aK=f5
z9?}YuNlm1fje;x{1SeSl9p`t&-k8M(+NLc`MTYrAscZbRF~m8PFtOV$$oPP+pSCLA
z`bwaYB`n->H}X>;MuRYgvE!uYJ`GCW%&}ya9{jn_@M=uK08!4>9@_CqHUCK;@E2CA
zu*v-^UabHKC~!GLLH)B}#tGTl*^O@X#Rw|NMu?G#2v&y|<IA$aCVE*%*5YeGtql2E
z=~NGEpp`)afef%HuqDCbAo`ptgQA>p%N3>)O8xcnW*EbOZI8KMm_=4D@vYcTj3c8{
z;p(DN|MM8)6bPqeCi`iEj7!mrRB;Kg06Lo@cx$YGMsE?F^D*y|>(|C{$O|eYze$jl
zPU62QScX3b2t{=>488t(y;{*Y(X7k4{7ymf#H!`U<}*uxEk~rsHsZIQ2Ve+g<+{?}
zxs(^n5QjvN?fiBj=D!FDr^uhS41f|$Wppm^mn=_x9*@>F-*1ZmA~&8ZUzv&$73E}1
zPt*vp7C{lQI6zMWyZGa)z^xEeE*dZFtHux}@wL=okg<T4={e;p{%_@xExIhy75c9g
zHkv0?vG+*}GF0!s1Y7koYqAMZ?-#(L>Xq&&-{%Oiyi`mUi{|+%K~@h`LUe}zF@Z*&
z3UrX;{fN~77QMC~ha_)b17^`{3nX-Yd}GQN`UrXj`l5Bh@I`~+bcjZ(9YQQ}b{*rz
zZvwEgvuy#HvzW=bzaU0Y9Ld_57Fn}5rvh`Rj4}T)0hSAH4MuI?-;0rOS|{88XFGl@
z_hX?et*gS9?@VD@n$UH1zB0z>w^6-v7G;L8mYzhKFfBsY3A98RIP<WRF~3z%7Vbm;
z@J1X|Tk>pCy6-N6O46zd8(RI_dN*q1R7tJ#Cjl0>cCymuuY3i>vagG~;m|xkURcBi
zk@7`a;j?}XY)lw7;nQJ3_Huvq>)=)jtc@~GNq-+eqkGuqY&)p>$tyg9SY6GGK*L;r
zL1Z8--ZH|7H3MeZ%;9ht-C;mlqSKLvz$F6}*2*`Y@>TmmQDh87>?UKc8mGQ*jIsP&
zgfVG0$HxJ=@P<}XEx=M_8Cw@!d=-0{fEbMdu{SL5wj(S^bDiHbh~#whRwd`T8=F%Q
zJ3I0?kKjkWm5Mv~l0702BYJiY#^t6m(6spjn3a%42CZ?ix@f3X1$tek->jQ6+62#3
z^mwcjYiQ3<;Hd5Cgf>!dfUHDZT`RRfJ9)CWf>Jm-D{*NI(>rjs5-x@b$^{#3%bCE@
ziwLj?nPkYO##MhzA?jn<bNJA^AmmG;fb`+F>}}<vT-jPC#G*y0oxoLP`^<|DoRw8@
z8q9b7b1G0UyjSes6=EEgQ=K=iOiE_|KQS0Bn~>`O8fEFeR33&>c_c~QA1*zg-zOmQ
zq~i{f<A%KtW)!50r1vWK9||&h;X_8gLZ#ny0N9#Am>sq(l&=zQv0~LsJAaD(E&m5b
z&)+XBZGNRUO9SDReJO>0?x7SUO*EEg{m8>Wk(JNXjY%QpwfbUYz>Ec3z0_6$XcB}h
zaUg0$y>|Fu3<YNfwT4DZ#bKxnDs#ClZON}|6*D?L=GRQE0%#RpSsm=NwniWlT~p;p
zL}z~30502{e-*+o%k*3uCBD#WL>1KZY@R~C`#2Sd#+C<Hye=f--*8=Bwm<So$`iGy
zsovVZT2$;iN)1iT!lFWIJDf(mN`K*JsjzITtHMg`{G%y<>^kLyAon?=cl$i$kL`RD
z#^?Kog&2uV=0z~_j|jBfu?sMqfL++yA1cqVNb#aT;|v0`O-Y>%J^EY1jRwqG!ge8_
z{Y5Gvbcc~ig%<=yO-%D*RMbL@2N7JeTkiK8^Chqq8<nTdk87O%;om_ay?VcjDgM`A
z@rsyIg~rME`3EoLje)b04a~aDBkXYe!#@qT78@k>fR~>Dw}jDap`fC(W)usskz$;>
z1v<)#*e4sIf|6F-xB}e5e()IjTWThtMqjF4Yf5M#suY8Hovih~JSO|&`~yNGuVK?M
zarixPVoZK9*Fo=%klX(o{2JldeatxOe=Ed7kJA)95a@RTE%#bxW3Q<n^RE;Xjszew
z+#5r>Xir-hHvb!6)@!l73OdxC1c21s`J+3AD*nr>rC(A-=TYgGxI0^>%(=3d@`2v*
z!J%bFP2H$9Az=uVFJVg@3dZ+^kiBrqB2<$9M39k2Q%R1gXiLYKbWmd`Z4wA}{#s!<
zvWvwHnLa!Amm{+k>eusy84VE2*-cjN%LEy%H141?WwJI=U<*l^AF$^?jt%v*36YzG
z8JnPz)wSs;<V*dZ;uv$VC@(!nzW+p^rHu9_?AGnm2B?LPC(&lY*55%wEkDdUILf8n
zroRv<*@oVLd6=)r;+?J)!G_h8WqOO^*5#$-;q<mZQQWF*W}9q6pDHveR%Nwjcx~H(
zSZc_ExtQ%?mE-U!fTYIh*Kx8c(x*eplUCCUkn>sh5NPe488mW5Fr=7oxK)o7n)LL!
z{>O0yvIQ-b0A~*c8uiA7ZCIuYygl|?iG>pxzRB~W1lrJ=Q#^^yrh2zTSQbYngr)Lg
zOqOxz^Nan@VvNazW?iNJzfoNHVKXDU?m7~e5oJ7^|LAL95)yG3b;qHneE*%(EeD}B
z(U#pAywkswbIFYFsTYev^Ivoa?+7uvhPjD0ru%Eo0<=n@Wy#{U^8Kbz#xVBB!5I=R
zHR~tyH=dn}LzA0Q&R$fs28olaX=UIl__XvnlYspQf2+b;5$LkjBDi!wnAKRc8e<e+
z7N=wOJGB&YLZCJJsjS!`s+<D=lZ7jxt;@eokR`${WDc358+=8WWd-Y9B;Xs)i+z=`
z=TBw+pc|OQPIHArmUjn$oc**2zE3X}FZ%;DJ1g}!^hjY$uCx|F%?XQIGvWz17x*DD
z#ALjnk`>DUMODE`X!eCsyjCD=ke45{jsY|RA!Q>$)0#D-X%(rAjLA6bP^Dv$(zwRS
zPG!EoQJ|HbD|E{cEpCakqn$LYvbKh_4F#*G{#N<2bW_Xp>V<9>ZdBp|W)?4h6#*6b
z=oF-nTJ<e_7zL#NmbJPW-v)}9R|jo#xvFfXu&5&qm&S(~U+#se=!;o}VQ<$(DQ~c=
ztPW@>`r-Fe5i;CrKFY3WVHP1wOTQ8JN)HoiMA77yJ#tjs%>pR|M?eshA>EH+Upjbb
zgVQYntsoPMtF@ZpM+!1(qF2;Z@a{OIv^mwpm+S9~L(*4p?@6AYBFK6R6t)xm8+vEF
z8qb(Vjy~3F)HibFlBQD>GxK6l%K>w{a9-s)LCU%jgPUG20f-ngzqC9k*o!W+aL_?E
z!F6EqhRdUHMf4mcC^@nPP!w#jF!fbxIah#?B~4b@FH(D@c^C&rU=9i`^N7A6mPw8z
zqwZi`1eLW=%Rp`w8%UVa8`dHNOY{mt4rB<PtvZmW&^c;)+X__K5lk2@3KXp|Ak)V3
zO6FQ*k_{8Wt&2wIQqDe;a7eSqRlr87(!%0!Y0T4ttg^HXnu*7kF`}KaOqGkSP6Z>I
z7|i2rU4I}boX4)2u#B)H0;Hb~=My_$YY`)3ilGWhe72CN>9GpH5*J&aas(P-5N=p-
zM++V>%cj!Pq%A)VtCM21aKZ-A{7tTAc*CGif&5WDoTik@|H(LZgej>iSp*XnHGwMG
zb`0zO>*543a+6QbMJsJ{9bSz%=q{N(`rbn9h%__WaB^K2`D)g5l|EUc#~3yO;)cPg
z(6EoTE0<mm!sxGEv*mu@4ZM+ksi<`lmAoAu0@Okclo(W=pE3;0$Tpsyg2^T0YAa9!
z7sm}~paHb(V&`5b?y40wrlfWb%n)Q;hxI=h{0+iWwSPz-EdLc{)m2md!*SqnPLnJ@
zEy(i1AT|LmCXuDFzwvCSL6PELh#{6|u-BRAUyUJVk!<_S_U{RzvS1M%2zc;jUZc#?
zBVyjr7v7SBur9&=Q%=wbvO;p(h9<!?f1E<JwnAF>j{q?;Y7whKzV-{Fcv45O29xJ6
zzui3Lmj-D(^sM`YMS>yz7}mqqgatoI!8CLUvkW`cz7<?VSBiw8Jvs&(gRwJKt51?c
zf@jMU<6t(+Gf@N)D<UpYv1M4R4L%KJR++=9xTKv{Xt1EF^AfrKaplY^1ObojQ=0ib
zDbSKaaLNlO#5V}D+AS{|&m1iP#s_4Pz6Vj0txeo4IWpz50`sT^ewhGzWgVSS^e^b$
z(qtP1m-%SO+bqzc!&IXd&!XAOxKVtFn34!`3;jiRfLJ|OV5p;#!}qK1OhqYu0z(x-
z@81Q&9_0nqbiO|OE(^vvv?u}p!A}jMHaN<Hc+dL9J*k*5bpuYx{MIq%Pt!$=zG7p@
zAgk%17Svf!m?HQzII5+fw+ygViLGgc<2eB}LR4b>hT92gA^|kw%R(NTYPrZcPM`d+
zKOSq#F%&j@BE2@=o605K1INWm{I>g2r6F8&<K=z#r~I+oo33zz&y69}3_&K^o{cY~
zJ4V(?!}rTeg`>;TsW;yyABd{eLMEme{IAAZAoZSXhe0>_!8i~r4%f*saKjJ*vvSKq
z3Y(-TPymfY>`!EOSD3rD@@l5}MUIaYr!j!0NitR5r2r_B;Ht{iKxwl~(|<hPGDOWG
zB(SB<FA`#jGhoQ-6cTbx2_PLS(_Ff^M<?(GueaP;=F`dnT6&h^Y7BJvGh(Pn`$cHC
zXT=a_8>xc4){w-sN(P-C@?`|k$;=>CT^eH?;-_ze+IPmWA@ef7FYzx4vNp)lEHwM3
zRXE0f-N)xouSn%JoLQt_FB23sIKpX5QicDyK<ZbP!ZB`4vAb1*Sk9P3*}2c|!YoN9
z{Y6-N6+ojJE9FMH-!nP&9aSi6an(}+qO^2Ni~28r^UndTcBHG*iY2W(!Y##;T)SA=
z{2Jd?R<X^t%n+k~F>P1J*nXr4qkvqq%C4m1)|fVv&`0L^TjejRY*}oR`8hvYpv8x>
zn#}`wzCsZ7-{-PES6%8W^=4VX?2@)+O^Xq9|H{+@LM#VNFwv4ij1|HxHH<=OT!aJv
z8H3sjZ*2XqI5gWeY(wgIrV-jm%^Dn1?-v51`p|&`?eY9dSVRjg-y*~PLl8Y@tDVu(
z9aQ?tWXdlpTjZ)-Kc^-I$%HIxXTJAzAnUQ%GJqDuOrFh}32Zb#fnAJ(Iv`fMl3W&i
zuu)WLpyg3zWAV7mKdhjJq1{DukNow7Fv~aNFelh@{X#)-5_!{k;f;E=_OGKp43Wqs
z{-s&wNfQtT<~8-a(%+bstZF~|LEfYK#fK>b*9f%qkqn>&iR}_bVkl*>tL^{x5U*A`
z`gjc0`5(?n1=2>)Ad~*?Q83Fi3|n4Cs|=9kQunAASB33x0ksq{^~{N`oP6^c1*XJ0
z=-j0E%$o`aAvog_W8v-|HJA6IbeSU2&xU^R{n%RtAZsNYs}yF1=fD~Ls4P<aJqk;Q
zr}O!x*w4ClZd{BqtTVtGcf?Q*r<Jkr?XzPD$w}Kx`~*RkU#+~d7iF@aBg_bZ%BZry
zl;*z|8qpJ7CX(|lkAcvlCNfTx`0;u(V(5f0SGD_h9#2KYWCZDnlIHuLKADOL*?`Nb
z{9lDoa|ljVtm=NWkXOr7Zkg(dj_a(a!CF`tIj<y$4+dLSOM*S%P*=foO9ZUFXk+==
ziohrt-MT|<1u&{WS?r)t=gXfJb8J;55|r;Eq@`e!LIPt1S+dj#3w#iy(x1Nw6nf|~
z$0`1`=Td%jqD-AFhf>BZO-05jDP87dkmVN5gf6(jAd<XUZ>nWhY1)GD-zs=G#Kcjd
zR#0W(c&w~PiL^8s4%#x<RQO6oX(0RcwPl*Q>7+P2NE{5GEEEW`>@Wa~(*_E^EXKmY
zJ-R~wLTe1cNZrD3g;_1-6tXg-4hh=-S#ZQUFcke>+Okw;U=}!Amy>@k(7b8a4gzh-
zau6c}Cknz;N(UtUrd8ndqrqT<W?`2$8W5()%JEOu09Xy7xi2h5=?oX`TAPYbt8#dR
z;W3jM@Tfjn@`&~|1=R9WP&Gw6Azx8=qoLx&R2_r}pmpkHMJ!kwVBt`)7DJca3fEG_
z(k0VAUl&F{*C)DgswvJ}5Z=|PYX1Je!YwxLqt?O1PlXsmPAWrMM&<Bj9GF5e8;)k!
z-(H7T6S+A+&9!e*?kV-n`VetG(~5B%Kl)gZC8Pt%=&n9opYp^J`p`+8y#dGyD~qnh
z{{77W>aNSgXOsM(t-Px(vyYixc$Z$S23dRL3N@8Vg2*;-qc{tD=#v;qOAR&O+D3%h
zom44aE8p53hhKI00X3?YvvsIKH>tG3(Ans2@-oAWD>=xovzwt2xAH1sk{w!D5>U2m
ziPC1iG^b*$lveml^u-cGI$`rHJ=tY(A;!w=x+Z=V{(7OtBsrRLXzQ3C8iOiqb-6_M
z`vn>?8G1@8eA^uWRs~YMF?Q4mF)wHYiYxt|T>w_2x#64<@?}sL&UlRvmTUoM45U*o
z_5rk9m*g@N5c{<9NN<B+#ni4fSfOQxpauy(tvYmvpfvkrip%npl*Ni<O?Su6q}mu`
zbAm34uQtqzlT#Hej@8A!;BOMg<YK@mVk9(T5g{%KQCEa9$UiC2;+i0hSEc`z5MyD?
z@rSlDc2CNa*o4)=RYHs^NLQNa!F(?Xv~q=|F*HpG?#!QmoAOMr8?CFH-2@1Ov)wZs
z?D(LWcNjdIH73fP`xdVeBctHQBAOWXbD@?3@u0kkRG|TnC<lLX1gO;f8hc_xli3a+
zRoBWZ>La+J6yxlQ)`D{fpU%?#e@V&Odr;YGni~8<uw2Y++4;&Z6uPAyXua0vvjJ}D
z)WXRSQa=`_#GFG`EUN#1j9x}P@xSFSss#FbZ6`Sy2Q9<Q3Nvx`8~q+{R(rAwfH1a5
zh%xq*Nri=lewP3%n~Xf8#;Ww63A3!J_h5e;S<BfEXcP}8Sy%z13(pg3!I3se1QG<y
zf^%vEYH3SyD-aDB!=My?V-?zpfxcf`_5Ju5(&h^3SIY!hDJt0dgc@>!kSK$+IBtic
zMyACfC$qf(MXqlaWTYoW&Xvd(&;)-%uw|OHYi-Lc_E*0L2ES=Kn?Jt9H|X86#;S+R
zR{69)0~zg@?qKxtQUUl8Ltrsao#WGjEU#3bt`ivee;`C0beD|%#nHR$aO>d4&Vzin
z%uFhaw$PM+02Y;xqg?FcgADHrjJhBly5TreB*DhMbvLPwCl#4*%RhIFaPBX=bp$NI
ztFtMu3M)4uW~<ss3>DH4F}(_?WejUsg>1yi_Xm`eVeH!HU~ztq|GzkSZ5)tJY@9zB
zd(6=#&Qj}rAk1hpwz`ZNoBuFII5Jz#g!;M<Q-1O|mx(l+EyR-40=qO_{!w8@AJp8Y
zHuTTO9!oK@6_$@y3A2)+htgg;)&I)FDU`~mw3^lVw?6_Udo)pwx@+cg|HmLmmRDS?
z%Oh6l%_xf95Dwi{8ARS%?s(Qdg>Qe7@}{HC49~vT5sO(f2Zm>A=ZBwySib01!@Z{v
zXw2W@pwNdKtPHY|5r5j|O)Fy@#rlOj!;RQT1%UPLjYFd?WH<0+gCdHT73m~PD^DCT
z=6YTZH!7TYD#_O^lYVv_Qw7s))#uU};@F5Lz5WA1Rxw<*P=IO+6X&0U!R!b!sN72Y
zYQ0<BZ0W?RiT@z-0WIj-r}*o?Nco@-)lrzvUmC*o-;u=e#NVS>i^pM8INJ~S%HD}%
zDnhlt@;L9-%ta#d%g|iyKl~clh|HV<z6>oipw^6Zxf3^Xpd2oQpy7itl{RUx9fck=
zfY#<GRI}B_0OJBiKjxHETc^Vqg1J?#QuxD4-8^QqL@Jf=$AWBlVu4+P9VsIP1UzDK
zCb=Y>?f)*n(eT9edM%gFrk?<|Oww^hjV(i<6*TZiesvC0stbi$tf<2zIdMQjm}QC~
zC!B5iP>@v->K17p{v*W7JdU;90^j$WR4ArOTE+F(3bK6Y_}OH}pc0>#epgFAQ1^vP
zD=1xsP$M|H2BvK-yJy0Os1axz6FWW0Vk;CaA{z}kGRw|iMPh}4O;|8uNcn{@%Y)|p
zToa65>H&d9x3DFLnUpLw-`57%3XbJl858j11X=Pj(H->-g+i^4V%;@$m&gD&GD(Tc
z$)tV}0+UiGb^cFe;>R*1?PnN-{E&8lE&V?O&52Warz90{1yhnRk7(P`FbzHk4xE$2
z(r1iSB8Y^A*?>j0Pk$zLW3_p-qDd#Hg@Sf**a<QwrMX(btN<LYlA^6s><dGcrC1Q*
z(qmypKBf)Q`=BQ&_B{k!9fnKKWgbn3;QUYY!E)4cn{uVk7ic-6vykBJ9~NZ94q_EO
zF6(f|Vvn$AIGuiaCp<>kREy2_pPmk4BQtGsod4!*-lG~SM`+fTcCOj<OV0t1Sfqej
zFJ+n(u&8H%N7&;KxM7*RMkGUPDh68_hsH*Sj{Tzc7iM{p8I0;OzeR`@S8UH!F3@aS
z?2WVfY8|h}k&R_Ggy6v`(%6@-v7!UL|J+n$VNMYyH$|;LdRenV(Dpoh8ubthIg&&l
zm+waljaZWX<#gpw2(fYngK-R%!_EgX3UH`H)6gI+-Sll>D+rCgblgOB@&Rn&Bo3Ow
z<gFlPPiRxBTSE+z7^<rF!{XR<ygr}4WwIYB(7KTdjKdW9#|2mgA+Va!jLR-ed6L<j
z3jcWw6rqu>@H2bjBPu1LC?evX?-)W~!a%G(<Mi)=SjG{GxQVaa&+iQsl^b&Dx+e{G
ze-|8eMvTpHv1Twt5k44+)$Np3gvn|^EtY5zFHFY+Z0XSp6L)p%@KvQW9XiUESgpvC
zg2G+t$CU-cCg)eP%1x2}Da=x5?2+AgGUd;@1k{p2j~h<!o)<%i!fXc0^+RI_nJ_4w
zY6Q__;AE~xz!0I{t+wf@F%ly)1pg9fWs@q6g^;i*xMv^WC@bW`Ox^q4SAa78;!sTy
zliBuH^J;C7NpCpt69}`kAGjqXWyUzB;e~#+K&!#=oYaFXgZrltb~ZCj^FxHRjLJx+
zqC%KTKd28z1LW|~Av_a@Mz<n4a_$dO@iE_~Z}Llp7?If&CMzbne(XT7Xe=*drPY7$
zT7ak#&`B|`WNvZq4S-gD%~6n@bt_3g%`4(H%SPi!wv|`Nhq5<xv;nmoGpR(JX7o{<
z5gV}ZSuBm`SYZ|;d;jvKqs{Y$F{mTyI@H1O34s<N16Qy>G&zot#ZBX(HL*7gc<bPZ
ze=UYs_mGV;f9;S|WWf@&YD19K53)2<13zSF$`ht?QP}1B^cz!<)_N4xUj;!=4n}Aa
zmFrL30$|0Y4Po?+#r1BtrZBASGk2zY3|Lek&VMsO^gZRxsKHDvJ8!IX>E97(8Ot8a
zDqNZG6UPR3Ndr>qhYPdV^4O%Jj%>6b8!{%9jl&8<988<pIg5w6&^XuxjxfR3ew`po
zuM*R2RMLtcry?4cOPe{McVi*WmeCph=(dyxnK|^udbi9VCP+u)ci*0ZxP&g4etAm}
z`IEGY)Z=IVgm)5Vwi`BHV*5x4KA0Bgv4h4xc^B_3tZExR%lzROWK>2Rl;W}#sO>%Y
z#ZpE9t!B7w8a_qM7_g|JqXyF+UrO8xtzydp#|71owF=MBRYm<5`;U~mQH%5R?6UG7
z$NAAHFfPn*|0>Xu)>#7VUu65wVhjmfc3UuKGCCC<s|>|l@zg<x(K46>kOo?iMJ6LX
zm;^5OZSPJ6ga+tb{2(C~P?KC$!eK$sGZH@hE%(`b;aKzz%sXe@&pY&`Mwlhoj&}cX
z1R|E9V$#Msvali7d;&7-Vf^Q`jo&EHXesHCH6*cNKrOfQ7@T`&&~Jq?^ONp2!Ynkh
zEd#lBk(p1ca-<4o8Vq%QlM=F`m6fv%ni_jaP{g%WS~xSrJh6QM2`=C7jy*AB)&=r!
z3bC0bi6TY86k#gt&&lJ9WgDy06IjXTq8nkx_~X^u{V)NRDyRCTrN3K9lq&OM?P|{z
zWCeq;lDgGRY!qtole=KW$oI+zW2!x&YHC5YpQbnSp{=CM<x722f%)KI1$I4eD&mdt
zh<=w%(!NBmRBu^%4y!xGzHhO8MmS-u!OzjFWmenyrxg0jN<kQJ)Wss5g@u$bJ$x|w
zK=q0jg7KGNWgvX8Jmrn&{&1Eyd@Ij1Ko)M9xrIj9$}=)6{UzKGFe3*0%7Q4y>?oi%
z3N!qo-azwEfVC@S!(eWSo`O_^#D`%l!NwC~HDa*crr6IB7}XbgkDv(+n0d7I8_Bqj
z%418Nm2)`AuMlF@%0^1;YyUcq>%z<XT5r{{3{O_E(u1k%|NewjxbYYp%J#hlP#=(O
zTmcsg>D7YB=rFbb{9A&oL^LM5qLR#?KM|OUpUgg&JpV$uy<L7W>ofj)lkM$_i?87A
zylUQ%IgpK5YvxQ~r2ZNwoZ_d<8#46|m=zy$iWLm7uxyJ*@)!($s8Rw30kW$Q=dq0I
z3!>Q6yo-yp2PWSa#uz5jDyukF=BEj>6p;2<*;n8Dpg=2jh0K~`W#pyUpY+`lCn|mB
z)Re!n91><^$r1z|=&#7?<$8<SG=~a^`W7LU9;21C2Xq|eu{X&O9KZaHkd_w5Dh%^M
zh}=I3%2sggZSzB?^J20a^YLKb2rg$`xfa~m2!%5J4V|5^)1OE?<>pJPD#|ZKL7ZKD
zDfj8IN9IzDf8<{Z!_z#GQc5o!g0fTRJu5Dq%;gy059Wh<V~!=e^wRMEJ_e3UDyLMQ
z-tLOaitNKVi3`?!nz+pUz>V7a?+32EE7|y$HWOZ{9XQRo>#j{EI@c%C9!(s);K@Yd
z*+l0RiNv}@VoM_NO2U1a$oT0ExBO&`o<A9S8_&$24j=rJAC37Yk+=xg3->KtZ`>u{
zB+~zqocB#4?bRjvIem3~TH?_(_exsg^|b%C{3YIQ^?P5M`%7BJ1%IiTb-|};83UNR
z4Ei!HeL%&aZ_?TfxU^!>^z@7Y$P9zd`{%BXr?f3Csl2arY*Fs%AMEPrHl{gO>-3jQ
z;F|uO=C))wH|tyc{*d1To%@-7e`feY`aQt8=Izcs-OeR{iCc`D(UrjGI#+`m*wrQb
z;)dgf;*x#9anIwgf#<-x(%pcY(%p5Zxa6RMbeE_j3yFkFpW|FwgL4@RoNF`JxvWLb
zWiE8C-Ad=$E_UwJM(0jh>RgAd&b8m+Tqj&dexJ6-xz4+tJEPgT)8BB;?{)6Xcbq%>
zfOBW<ckbN7&Yg3}x$}=Ycis`_x*m6~%Q5G=pLDL<3For=wR8F8GxxfVF6WE3F8`uV
zE-%^9joa7Bjs2pNE1b~Y6<pTd6@Ag(jo;VamE6?96?f_2$~vF!N{c(V8!5+x_Gh??
z6=%5e&(3gF7j<@(w{&)s-#*PvI<vE@e)}vp<*Uwa3~5aL>MU2qvjq3Rw1k`4-X#Yo
z)7<}_>XJ8{<J|SHI5+er=Z4(x+z%hkbT`%}+^~x>-A%;__oIGU?&dGD-0%rm?v^20
z?#ExWakn<Lam%6WZ5L;_pFDetyS;OU8|6-MBX2v!-T8J~cSq+_-RPUzy1TZtb+xBC
zclUs{E=l=o7CE<%G<%be8HH(X`p`62(>INZOLMbI(%sBa>8_?f^^%|F&L)o!Cfc}$
zz3Hz0Xu5l(opUP)+Y5X)zc1z4&?nu^<#*FQ=SC6dG~AB!T=EuZlguZs4#YXoxfxy4
zT~p^Y*G##l_riZ=x?56{>Gt3^xt=`Wn)zMBx24qSUVb-ylJ1)CPjj~u*9z+GE}r+|
z@^A;_w~%<=Z{wz)mT-r89}j+%=Sf_~`4pGu258juY#Y~%zoxbcx0&~)<Z&ip4!Q5U
zgFpYS`=I`&w7<0Pm$nFe*4%{K#Iq5X+@F9pr14ss`*cWN+5)f+#QS+cURup*S#J8V
zg!|&{hte8GLD$0x=lM44P=ZS4y#r;gc5V`Rs^OVjkZ?`?60Z5ogqyuP;eHOjoxCjZ
z3HR&yXS-LQIn(W_J=?u@+u827pPlV~lRVwcA-#J1{SH2Q^Wa&8J<Iil=D**5fqUzj
zbKN^l7rM9i^>lyy<U+T8M8aLex8(=2+z!J1&-M%5{@1_j_PzTp_vf#_>v}71N69ns
zHGSR2b>+P${${<A>DG~txrE!wvnTij{Pg8n$M@N)>l2yoRG!JbnXYDcW}ruNn+!L7
z1ZAwsf>_{NGu_O-S#DOZELS_&xrd2kIqhyB@zfv5at+N{Za&x}yR+PE(t3*DpTbMa
z$B;fe2@5Be;g$@}bVCU<3O9{3hw~hRYn+(r@_Ftek4<007Z<0ySNXk!Hok}F?BUM6
zNciRO$rhgL2WGli#5;2VJlmB#H8{5ueELf2X0LONi7fXf?{neJ=?fF?b>e7lp8=0#
zxEioobqQC$BjKt@a}RV_LtbaY8-vNy4{^<4$>Z<`{-)O?+>AvD(nz@1h;!!3gqt<S
zxqA4Zfqaz^hj@QB*z^w0y^WtohB>$Hh;!5U_62b-=>^~PB)(AzR|mGdFZ}X6{6e_@
z!T%gE@#6dZUehk&KH=H)3EwYDa|`&kgmTwrIJb!3f8kpr&lQ_!OA1H2n|L<;8~+SS
zJNUgvcwM?Xjk44nPIrGJ%&a5n?suZ)Qs~R~Z@^aZzE^%1(v}aTyV*PVb|HDY0H^kR
z8Eu!obQZjj90hGhJNFggW>CJ_@WEli9>BdtJT=qsGm-Xm0dcIO{ZC7G?|?5kE!`c3
zo;AG_Zv9E;meZ#+4o-K6!Isb`o#c5Oca(UR^PYqbHC>%s&+q!)&OJ#PmM;J!?j-5Y
z;JM^D@8qem#<}B!Ti-6-H9&{jfuv76tfjs*W+mIHuXFAsewyCSa!vHJ8C~Fu2K?QU
z>6+WX?-O`d^1Lp~J>9#LYg&=znpbDJnbhU1QJvuTPOcqwI6cwc&EP#r-PO={JWswJ
zp^j%Cb?z~KAD|2mk+1raPOgElod~m0;d(kZT=f9Ge!|mtqo2X`#C65>#%17=y(kmz
z--PKv9&4)^^LW+_hj$hy+&SP&PLjWV)C=QmSAByX%|B0rmpZr$`97NXSG>{5&Fb&m
zpP+ri;!Z9deC}z^{gQ9*fi2|yK7J>+cA{N(a(l`5j5(d$Kfrq`UHGgw&u+Z;<@r6H
zHH(O|2lUt^`jGx7q`9#@{cd}=teHC9LmBb=NKGd<yP=a)yS<EX1HlIHT>l1PiFXw5
zLvgdHr!hPyl8>wSz9fUZoS;rtq`Q3Zo4{rdqn?k^zIdOEzw5!o2h(`3!Zodi&t7-#
z0A-q0=-fe`H9Kg7(Cq_ZTNCb9!pvQ2{IrNLBYCeUA4~cD^jyYG-ao>x+WsH-{TtjX
zxF+0;p~OWTHMr!`gd0d5&7A1m{d}t>Zx4Cr9;6%%(7GPKkC6YdeA|YbO&ZJTvzHA^
zxP0=yjyx?H&DgLKp2XG2QI;gGabm(1;paKUI|n+F&QRPE=skw#C|o|SrUd`E^%?LU
zY4!!Hqa4MATS0mY_&o>L*bX|bBj2>;ByL7C?G1brZU?SuJN*RXbOZfCQy1|v^1&YJ
zly=iJ09U~8A!%;LaQa^S&)QGj^(Jk~SOXtTC#`bQ+{?FS+yTW2Pu4?^`a_mS$rw$w
z#B*E1Eg72bYWp*W<L{{azE8c~l;%pv{}E_Eg}CbYy$TwrjK}fw3$P{cP}ks9xTg9x
zuBM8(aW(q3iTGU`*E}-Koy7kvuv*&59G)5Qa|c}0+Z|og#gxZoxaNB@+-IcqIj-r+
zj*N4JZ-SQx;0L*(`5B(?c5=;A3~TO(pTpEUZU$~T?p?yx@{|lPnD`r(GKTDjes5&C
z{=ARC4O6;TCfsrSCJ&I;BhKBR{_-SwZ|41kXqe>+@t2Kjpo~oyC0sSXC*lr~e}(@u
z&*T{V=EFA$@{rAMcrQ83xjy(i4_du|AIT;c^ZRn#XmI+?<Zk4mACu<vu98t0zt7F0
zFQh+%pJw%<Kb)K8z9fGS7Q##WGhIE;hT~ao7-1e^?4G@!@tpX+#!oNe`xno-i-~72
z@!-^s-yn_Y#Pbf%{kSCykPqszoMeKgN2&LN8E*X%@(!Ix5${iN*|>4I{}A8O{495f
za7|m%-HK6J?g+n2_;!qEg1)*P?u34WEoo%lz;kxbOqZqSaOYN%ujwVw1UX=O0{H|w
zEe4y#bE}@GAs;j{-tfEWHu8gvS3|ue-+-1p)&D0?Qg?e3ZXteVGhQ_Gb*_T=1CK%b
znZ2PGSnV|CBIIX!9bx&M)Yw9QaV}Ws{|`A=O}Gber{L;4Ku_v<IXpj?xE})R2R)bc
zOt?Wj*Nx^oxZ2Tfo{c`?9s%z|d^5r7@ZW>qy>QE^n`8&lg9m2yfu6&tvvX~1?$7s*
zgqw^1#u3o{UHJIxgj+G#xxsvEY0ouB)7=yJS%jNS-j2dIOZZ*0hxSifU&-(3{2oS_
z+BebzkIW>k(Y%YsNzxd>?^)T*sdyIRJo#2%-Mrf51C7%ih^I5|4BTw!wu$G<xb=G}
zC(mlay+OG3;_=H8ZV2(-jC)6MwMn?&;%6d$mM?YgkNjRThccZ>-?Wu_#x1X7zKH+n
z`Ml##Y5ZMreb&V_|C{k2Y)zwc`w2Uj-%YPDuD^}^xdVDVn&tivyoUO3I;X2^z97wM
z-t<?n<}Hl<t25nV!W@>LUeK7|D{)JX;{Q7QBMWWieK)SA9sJKT;BTk774!jTk}sv9
z{<P_S(#^|s$B3&7?j8Kj?UisR6?b!Yx1aZa^7{zSLyG4${FB!jXrA29{FQIt!fEW8
zF%+KVsWE8gn1nk9t!jB5;n~cbvgr%@+(Z}GP@QlI`pHK|F~(#k+zGzd9O&k1e0R5y
zG-}3lce87{yIF&~Q-12LfU#kCcUQBtyX#3>Ggk5)zq8-z<}TuW)~4=^O?+R7gHAJc
z@LtF7hVJed@k!2lnld-={RHnCQ!}XxvXV@8pngs=zvuUo{oUP?>TYfcb=RKnjeWbj
z<=Nfba?)E&_*(of=eY#@`=r}Q{IwIodUSW|8F1H=<^sMSqCBS){w3A-M(TTA7k3-@
zddhPL&u75TrTnMidf|HDl3OW1Zh8;sh|E?CUHX8}Y@jW{6ZN#=h9e2LpST}73Ed9U
z&y$~a_<7_2VW9b3`j8s@Cgm3M8+U@V{)-b0PNR$s3li?>8oo6X5GP$r^B1&>!{{jZ
zUUQUoyHx!m?eiYy;a$;L5Pro8=X&t{yZpX~GUVWT@h<tR56>mEn;v}YuQ)~!{wu=2
zop8OuBtMM+zk=U`d0s1waHIJ>3^(f-^89|B<oda^Td+#vpM8Qhk9<FCD>%RV6Rs%_
zJpt{0Hhv2g_8oKqbNPNC)6G4Of7iw}e@4HMPkEr#j|ejne^ZHXsKTEEE&EgFVAZ&$
zdx)#2bC*z`8h<78FX^A|`tW`wZZUqAk0xGp1ogC~B_q<^O0Y4+HKU}1TfWG-WsTrG
z*LT1l<KtewE#Cx=dqZi`f7I;Fa=+GZzW<)*V91}#v$4UsLY~QH{7{cgLo~j3bG65r
z8_EA&yx&57(za_(QqRO$All%!lxI`hZf^E5`UK>mX5^s}#M?j{sL9B5!}tyRCP(86
zaWklgWDmYk{-kKY@BZKoq67J=VSFtk&6<<M2i+Hf&&93BC42B4zFb1RZRLG8?hRZM
z!ejGwY3`*i>8^^fvzS}&2Y(0G75X>Vi<g*p9LNam>MYUuV46F`_j+hjGn~BOMiBN0
z*quD*_Rn%x<F9^CmRrjEgUBq$`1UmUpLICPO(f4XjNeJz5AZkr1iA$D0g{1}=n%vU
zGY@6DpAoheUTnbsgN*qPEku5(=6zA7YlL=>j7EMa$#N%%yD3ROb77iWhrBb8IBN*g
zgZx+XtR<f09Oy+IH^E!YpR{oazC8%OWLTDalxI8Mm+Z)Lr|~?WY<x`iPUZRh(k%Da
zHu%kOq(D2$hCUsTk@@{P-P-xc7mbB@?hSnh(x%5y$H?Mp|62&R8aIpI^|bAVO$qnl
z4#wYJ-fbtouB!hjwBLQ`3E&6GTpyB`=_d()u5;h!`z-k80{Od=c5zu7_c7mpRp{M!
zz-BYPG^2lMqW!NQmT*1DhV&8A13kOYy9(5FH-gn3r`#72{^B<7*I;YuzgJA7f1(_#
z3KQ;j%CHf6VN);oqg}$K5%;H=F8Mdyix~;GVPL{-CypH_(R=K7t{-url6G5h+i(L&
z;|AQ!_Reisl<r<^NOv3XqxpbHQ;a*EG(IH0k8yqZ)=)=V`!LJxTAc1ONbgT74@7PL
zusyun-Zdas)Q@7`15Z4NtJ{QJkDIjs9bgUmUh>`smxX)sV)VAu$1>(;&9mYC6`8KC
z5Zx_U9~?iEO<$zDES`OF?~vB|1<v&(>;|4M;`;KtAMQHbHMmP~Rrp;-etYtKlKgk=
zfzD`Dy1RjI=knX*vM+!Z#m>#o@8BMDY3>>7dC}5@d#Ztc8u#Qv+9A)T3mA9Xkhkl2
z-oq23HKU7d+Mea=lu&{>X$SPg#J7zwbs5f0<o9UY&N_Ib7rKC+om@We(|dGsb-kS1
zNIG@&6(wLBm<PXD*U7DaBjMJeJ9vq?@n(2u6K*4}CZ9eCHzOOq)V!LBaxHw2JVd$&
zI=L$17=vqi9X^3^>*y2fJYxc7?2l{u<XqRh@m%-(n1s8L@;^5`jrnpX_w1;ITTJ)|
zsMAttbPet(ZLXN-ah{WT9^<*5@AEn^|K#@xT<zZWuI@-jH=oLxb)ci$1h$ZMgdIHh
z;MU^?lI~WXzp6~Q-8|=>L&9^IE7f##GwV9K1+(FahK}xuA&e=|VdWM_<<icN(r@E`
z7~T)_tQ+0Q&1gVozp{g?8QPH#9o-S|rB_0~Y-sVtnXW4$$JW=-+0m$*nKOp@)NXhL
zeAlBHfzIncgOQIDKloy3wFzG9&G(n6(<Ei6p*&3&(Vi(svYGc29Ya0Mp#G9o)WLe5
z$9V2dxRr!Q3`we<*Yi6`oono$KFzs#e4l@ayg{#pd_z1*O1C5(Q!jqcgU9CgMz6p(
z`RNnI)5v>&{f0ND<F~G_BR#?$LO$l(BHUnTsx-5C&LB@=T<QgXenDTdF5zy0W@Gp^
z8n;Bhi7y|Qr0=dFtrC7$>7Ddzx{?k&lFfTFV{akPiMVQ9Ju?1M;t2Dz?a05(3w}&|
zGm!%p@jXc!t~m*v6)*X$8Ae`d^Ud!=1D;FqvlxfymTYGHoln0qhxnSwXH$D<wUw}>
zF@yeN-co)y631TRSj2NBV+4M0Cq30w@(tW3@ET|WKBKS=dN}HNBY9Aq1JIq|cV40m
z{Rg_Huc$xDs(j67Onrwu?8PBoGG0TcE~F2?Nr$$Q@S=4R`A+k+<^p(VK5=wzhd%Wz
z*K}7$=K1i&nH}ioPk}$sV=*ymI(RmY_kO3k=C9fY{u)Idk|P*XafdD-Uys5+)YTDw
zH|-;y&ydN5J8;r$W?iWnokH_S;x8cWV&qG{XVcc^FQjjI<r-ed*M7=PUYqmqHwAw`
z2Wy0vadr9B;Ua#I0K?5cM8AZrufAd49_I0cKTMqUJ&YfZD*bb)gMGxe(ZToR7nxxp
zw5e+c#`8FSPVzj!GXvKF=W*?D30xyIJB{bLxTY<%7i13&sM5KJ|GR=Kt~&T2f!~#)
z3uAu`1DwjOd2u`Ny3y!b@p}RXQ9?QAsJ&A+wMWoV@q6ZB=BN5jU5bAf>3dh_hJh{B
zZ|FNh-$(Ji#L}NPlzPXn+E#zQ&Bb46f5f{m3O^KlaTI>I!XJTt$KYM!TPa-m$Bp9K
zKz%RaJ2Vxq%vhRm9h85{tMpIfcLuHp{+eEg9`C2Q)sz|WJ{jb^{;Xl~n<-ZE9o&B5
zltJY=#P6Ppe-H0_q5mt4^~ga<e%FB25q8EK^kc-oR&*4uype8iu#<$ZQFt)%bO!Y{
zZz1L1${bsDnvdUsv?u(`8-tEm82*|kz~_}2Zhm5bo6EOqT=PZn2+uh<)zbo=({O!B
zr<a`Yo|dPkuPNtiv_ab5i{z_@aXmSQ_t&5i?YV9Q{)S0zfnP7i^@Hc|n=BmUYH-7d
zZz!(0-vHM%a)7IOXMmfxXpn2V3BRHtefw~}&)+)8-KBCbCLC@$u71t{H;Qi><3c~y
zgS7Tu>lRS9<WcBw41LT})=+t`T}fFlBu?@%ocH<@gIwLQYuzY*55v{%xfZ#M@11d9
zX&j)Q;f3byjPvOFn+|HcC+`J3S&P+Ny?F{aI{#*L|4pCpd>#3B__c1vN$QQY)0!h`
zZUk|U#w9Zby1Hr13&?M8=%+QmUJPM1jOj_n2+?nPGf!wVU1JP<mK=>^tPk~193lVE
zZ#w0YPB47$#qWvGm$t0_K;h@$SADv8tAXEhaT7^5tkHZpkaE!9d_tJ^#LpNd{l;{7
zWyaxaT|4}qhC3J6hcMqJk5xPe>YaQubxvLkmZY3DgRgbdcrWC84bOa@$)4c7u66U~
zLWgP2eFwkuY5z_8s7v%dCHR>)ioA(W$=@dGbQ+laG!jSX_ZM5<#mBYO-zIR?;|`wn
zlzsX;gx!yyW*mH20}a$(_wao^ZYyp#Zr*zMpE%y&cW7rm^4}Z(1JS=Jzv2_()A%=|
zhTm@wbbT+N{i5s97*>Z|KNo+OQjUc@8_<IsP<+IpaV3<agf%UtNw~SB`6#?L@4z+g
zu|ylUfPBwiaE&|4_s1!h`jUB+^$q+U2UGifhu`y$GoRxb^i~JR&tdumz8{gFBhJzx
zJ^9~*zhlCoUo*c?MDgt9_xjx#ZXM(44t~GLbJP9|x3QVAn!KwI*~(bGnBNO<FGKrg
z+}V^ZK^z%8|Hyk&eVRL!-^~XH!Rzom?Pdb_!??lV({M@3+59fwKOE#{@ZPkY-&+RJ
zu`w1SBTOfqnT5^`;#(bKzW5vUIPxcS9dNCyr+gZJ5c`t#(Bm4OlX2qT1w7TiJ<W6G
zA=)tI|DnE-kHva+W*$bJ)%Sw0=zoR_<EeZu=XYIi<|NOOpJMW}m;6xv=?P>EWR|8p
z^3s&%hD7l-wI@GqX{)CZ*E4)yN4SmjKd<q8g=f;r)rfzsO*B18KXDuQ1>j1fz4{H@
z6kHq1HQG6L(vTnM_FBT-%<nz8-zof=gl|h5eVzQP-ogGP{37&(Jf#DkNx70;nKw35
zK6s*O05Vr!;zBl_iG0Ualw>`9!c3mahh(@Mo(~<xFHYk~eNW=ZC)@<eRY`i6@nl|l
z5B_H3G!EUzvk6`MR6Xer)fdeqoaSk5!DiN8=3b+&ljsFzp3HJj${%A<J#FTf{4T=j
z_w&MUB7WM)ZTwz^d*YQ0_c&w9f^(P{Q~&esVopq&_2lW%uLro9(C3#cGF<IY^d}1c
zZkl^)BYwxQX1hAWE&r^It5;p@BVNYOCEMG$b@F$AntPRJDeiS#W7{_FEuJrY)rR#N
z@|NM;yS%^jNxIwIrH$Kk5jgX>4Ll#JCjDuwtyi{jGvT$J&okes0_*DBdxY6`3-!)i
zVavbakrl)R9k+vPZEBWez<y~ipZBIVY3?EN^a$T4^8U!V<dc4)lHYTfyVn!vldNMe
zyck^=7_vvQ>6SF;LHrT@9#cF~emCg1aN^s{?^(y$m+&rnEyh!(A<6xe`43DN_u~GH
zo8K3D_CZ%&kmlHql-y6;^C|cL@|*VT=F=W->x|CsEavfNyW}5uDvqo9y=h;%+sNEm
zaqPb<&D{;Au=__62kH0W{a__}jn^r=hc-M9Ceb^B*E~yIf0pK+e;vK`J$%RiLiw*v
zlm5dk;`j4kr@4b)p|@#DcMINT4H9g_!F0EbFw4oqWBZ)T1KYXVxgwrV@GR$9jZ5Ox
zewwyl>vkfrKg9d?Dec`3esBA#y?c3Nd$)z>W8j-{FKub>w(;D=MwAzyMMv>$d$*YP
z_1C4lHSqmf+W3|X)-P{jod|p_Zp9Gj3Ju=l_i~=kL&K%@@Jiow_tL<0_Za!y#B<>o
z=K8~#^V7zjDntjs^BLT)>Jsjk%<X19lje3dI`{Ze=bqR>{a=aRZT8i!X-hxGw|?$d
z3)oKtUoFD#Gsn@ntWUT#Q_!=K?oQmof%M-zpN4myYDai}{}FfarZmU?Ik)<rG}j0n
zzXAIP?mxIsaI@}Dcc1fIOMUI6eLMl()6OB>>NJ<hbLQ9RR6a|02REj<j=ZlWzBP|h
zCj;<LeVqdK1Zf`JM?X@V?j9$OU9P=bRN~w-+3c0$`+E3k!)@#x!T(u`lm74^blCvk
zYkd6{Sj+eC@cTSmH=NR(Ih*upLrQA_?f!Av-1qpl@e}AuoXYRz^80R@+xQiEQeK`-
zbC*QeLc+hqH}rGvDzG2m9)u5Hx{~$=JqPiAJ#GWvZ;YNl<@pRWTZ9|V?~ypIq37@{
z!BygZhWphp##5dRxUslLaZlj(<9?6(6n7B!OWX`(p_&s3x0>G@aNBUZaj)Z+;*uMU
zZw|GEzU1o|7<;zd9P0WwzYjTh9Uk~#erKoo#}f_65NFblOrei;ZQPANVEqbrEADpO
z9k@l%>hswJ?wQ32cQ^0%;+|VbIe3o6jmMSZmN(F5cuvBt9F}lXd9GgR+%I@e$IZlT
z=t&#q`7rJ=+)K>&Hupz=)F<H{=lv<%))UA9r=dSO7oO(*IovYbDqM0MV^2M0<M#~4
zsoIhBmG`u9FY;c?dU)~x<7sc`Nk6m`-r5Gf6E};tH1nA@ZUbZQtGxe4IOW~hCBtn-
zCfKw;-MtC+GGp+zW9jaxyZ)G{J&4{A8D%Hmb`bA&+)j6nTT0tq)*;=!h5y<v=eR%c
z{2$yB!mVZ;Z^W&^&)Rc&fByWm#`n)lTSdAr^d$T(=enIYo#)=e|4uZ(8xNkBw&~rj
zX`3(Wmi9Sy_P^kVo<-l;@4U1R=(Dyt<jTA*X*<8_lD5lrPMh22kBLV={72&8kc4}z
z?Vz;3;Qu4sr?|i2cCPN?zU282+&8%Y;L@0$XW~x5)vPOli0J3K-0r6Do9yZ?c*@m2
zy3@^S`+M5RN3Nmx>O}p)ixUr>^NYlT`_4$rIk+@2`-{61kNo?uiHGlcB(ZMq*=~KJ
zz`Z!92pw#Z+c>PqZCX&~IuiH62?=*P&$Dpn;kx0Hv?J-0Li^<Xd$`3*)7*3D!<P)s
za?i7OZYi6!ma*Y#Ig9)&5?$O%1j|*!y1N$=o!#o=XSl}3Gu)cQnQrZgv)zh5nQkR+
z)$mOB!q7~&nr9<(_BC@d-P$_z`AOROIqB{a(zzUW748SPLAdL2ThQZgt;%xSre(R8
zd2Z)@2YQ8_tPSkKty`JN`dp@au{z6bz->gaya~5?OqP3T6i?{6VK;kSiR(MK&0RCy
z%{*_#Z35p`nCV_7U$^sq2X5y|;>B&p?Z7P|?z_S7#pU3};+Bnu2H@lQ{X&AW_0Dvq
z{4U3>J(}g#;2LqOaVtn;W$%PrMLT+7ApHQ(M&8%(TuZy2M3||#U*M+WX6m_vI>04y
zTNa^DZ-l>=Cfv*Pt=p^MuW7Uup1XKoM?bOtBy;86j19Prn`oOmXrnw|BHV+-_b~1;
z+y?rC$9Znzxf%Bqzh}Z*vydydk++xoqR*!d@4)SxmT<eO(EA^yP2=x5+%nuMJ;9dZ
z*6{m<YRWc|^!dFRx3)hxeM93Y%8J_twiEX%?l-tMafenX+*>?9xIfeVf#?6=-ot%9
zGSmGp&%fYyZamL@#Pd_!-*7Vvy0}^N>0k2uAGmLD|G~BBLBy|V186h2j=0ltXW>?K
zWt<+I?pEPmz^%qL4x_zKq`l*6FX-yd!(TVtw{YLVeGhjD?sD9gBlK;@o!fTOxtEVY
z|GCf~x0Ake7jE4l^t8AaaT{<Oahr}Jf9+?i!F|wPIy>BD^x<u@+*QQ+1Kc3o^|%{x
zH{))_-H!YG$t-sV&%5=z9-eMUbNBK)2e&hMo*T<^GuS5Fcz(Z}4~=o9{4U4sLdMw1
z`wr@EJMN`{nQk*~6K-RVOt+ygZ3VX;zFpTY)9r$ncaF$(J9uuVZN6MWo2broTPJ3^
zEj+{gxsm>0sdH0_{};IFxS6;Iacl9nhBnklxYf89YG@A&;rY3lZUxUw{7k|<j9ap&
zt9y*+Ql86jkMnz`qx~aat?Jdyy?|TYzng2st>O3Dp55FAem{l3U*ev_EyHa_IkgG5
zir+8q@8-6l{942B7je4^ySts)-Q5nJ+jn<!D~@$@D{-rEFW^@9CH{f<=ef3bH@BHE
z+i*K^ui}1#dlUB-?hm*vOvJY`+1|$UWhTbknLzKDOB%RcbGo~Aqu3uXrki__=Z2Hr
z-A4Rv!vAKTFEO$If27@gaFy4c;Q51)uwgAZFop@VkptCGfemb60uvZZj@iHlwqXMs
z*uVxhW@9#{0u|VXYFHzKT8Kag5=fu}ZJ2<fJJOZz)$6@_z4rzxP%#y#h6>b}3Y3Kc
z6*C3}s-XfEsrU1U;C65K)cmo3eCzi+=XvgV&Uv2m_B(I4YTGN}s&KC125usjc_mci
zHofksx>Hz{-_xf5g*rjHlM}=-dKbGG(F@fDC|@am9W4%b{3f+myq69M_vjDs2%XcE
z-$}l?a-1Qv$c;9iw^;hcEiR6AWIcD{xH76A(XdN<tSmn9`vK&YXm>CIMHq#4d8uQ=
z(>&ItPUz;&a7TG8Cx16qtA7&8KiQH||IkyyPal@|R;xcO>>MMUaVW+_#K!CM`2LgW
zb#LW`_#1g)Dt$T{R_ZJ8OQ3#JUYN;mHs+!f^RNJmumqnh|I4^nVGZ`^yH|cAKUDlh
zept)B4i)(Yq4zUG!g_j7jWR>_{dip1Om4+?l%X7z*oT&rL&E{`5RTv&(mV7&aDv{o
zQvYM=&~S<#Lry)BMKmYct*)tAYHgD=#ib`oYkf-EGSWykB-r9L&xN?(CDDv%?Rc8p
zNk;P(?aFfpI*0qd__%(i=g@e?9PjnK&`h?x<^8LVQ%KX>-mw3m9lw4h^nMX9{F!=Q
zUdI`n$3<MmRa`^%)#2d=*;nyrA)@M}?LXX=zsnDM7-zTXu^ql2?$GO^v`OPV`U5od
z%UejG{&jKEPM+ic2v0D8^j`JR{$U}{a7veYqLX_VeFSn3)koYJ^+-2a#BUVFU>u4u
z5tGqZF+NNsd-MyZlf8e;H`wUCjepvInaQu_zPbqULUmJtItq#5>MZqPgL-hbd*))W
z{C#R@D5cNC0xUxNNOj>@aa4D9q5GkJfia9Mnfp-xW3jq)xjJ>Vx|M9@PR&r~q77Bn
zGAt3!GOWNV#P(YMu>ZNRhF)j=Vw`&|eH|Kvo8(UH(tjY=^K0qRzaTdweQQ|QN^VE$
zzIbKC%eyS&F2`WIt!#E5_72tmHRGj!($hXl9a%5k3F%0RZ-cm2`prHZz#$w#d51oE
zlW}*l{jU7CTmIWG|B>Ctm47nJog*8C*HkF|-fPP(`OkMxk!iGDQU0q7!ZBf<z$wHL
zJ6RB_Qw1SOuY0c`#2*)gN{j0DysS;0plr-hMrJ50mkL6&ds^=2g*4fTw0ZU{c?PK-
z{cUY{*S4YI9QS!-XABLw5`BMVGJ~2)%HJ$`v_$zM>!%6(p!OdPxac>RaTV84um667
zyouXrKRQ(ZOBt{BoDIU=B3xxZOXj#6?+*=4kF9MuZQKB@P3ne>aex&2vvz2>BaC}^
zfJb<O0pz_V|JIBPS&M#pemo=$<L-S~ds`^~Zjyg7f?w6cp`nNzg$nlcI5MWZRo@yK
zis^OAVSEdF`Ip|`=L$oEcbAmT#Bymx&sT?qiSFxt$^Mj|zYr$Vr(!y0Vm9WY6!WkE
zi_rI1e-^TTcrh%YFT)D7_~xs~)KTA;z9#Zlp4XC{@<P{Q`D3~Kv0DBha|?Zc1H5Bg
zUx)SBh#0@><wapLy>3?#+o33IrEf>W?xK)9TI3rQg?e(QK5jHG6|EyKbG`4=!$Ub)
ziGApOwJ;nYd%kWwC-OH|w5Qv61pOFlgi|Y=_<Kd6&cJECfw07B&p^Y&qEPWW<A?4$
zg&5*Uq8VwlpMEiPpc7r_Mh02r9=;eFtvhI%Ulf`R8n=+GtF2j(zBJlUv()ofdcOPW
zj(fiJCFn`AAyX7OJtK=VIEVANh|9Q&cJc2J|IS1aJH`9;ZZl*SQUAR0tmj|we6mH_
zTFF$(^M&0;wiFbFYr?sKo4Ad1p?h$L-Zfm<LyE#Z`UB)P6@@H!hJW|SqEPPrN8fO8
z9`sFX-+Zel?i)ZgJw9K$7F&1mm1s;Ngd}&vEa_Y>o&0N7D}QguqxSmM$*1)j<jp2|
z6b+~_KAZO?{SJ&k5k_GQ#-V+a@`u54PIg~W{;n&3WbT6UcS-w)X0)Jnsq%+3+E6Tv
zmJjp7L^8WG-}qc!m`tCF!TMulUg*Mf?wQCQ6fU}lSaX8e{N|#H4O~iAYqRH(3lQ6u
z7Z#CAP$!*nBo<2}8pJV)nk&+O-L+}#ze(D}5^W=y<Zd9B`Rxj<!Wyi_dThjIwC`5`
z?p6ODSH2GW{!P9=vfJ1nXp|qCZs~vZ=zl#{|9+_cy{>#-QU7MN{l8bXgwy*{QP@sC
zvn$K!<@)u}7}B1P?6c#p#6BEA-#3fGA@T^0VfQz7gcIZ`^nS~_2(qV@?ML?gRDLF7
z>a~hLsI%#5bfQI@vyRN-49+3VFQx50PwzUZ4b#3}q+dpk&5%uKyU?Aou3rA@v%c+F
z{>y%2Xt?UW%0I|&-!cA5zk!>$jXUW5a(=i+_I#iHUOg;4pg+PB3?Rm@TG)B7`)=C8
z$X*#nAAtsONlw#7qFx+}__Y{k7)6djdf{_n99fK%^%$M#!bI-L$j;I)m}C54zJ5Xp
z`(GHf+Q>Lrr$1AFT-f_Pqe=UVseUsZGcg;3^?#H44s*HNy`K*6Vz8YT=WhDP^{e`}
z@sjoz%~!O)uC=-+)uI1UtN$ULRheO-R5<fcQDL0)htZlA=^t$zgkK#QN9FV0=iBn+
z4sqKgj@!g_$gr@)JuM0Olw5&ytu}yMgB1VH4f3UWXf5|TWaY=)Ecx_J`E<Gdhq6Cd
z{>eJjBcTpR9#=*Vvtj*aBQ|3zwxbN?s6@MaI^5fNQaMHUCgmSl<j{Cm`G2VVqvfpf
zudJm~`U_3gu411s4&V@~ZkanFtI1f8c_R7=oI-q(_?3v?H1QiRext>2hx?*yNTT9P
zZ(FdUJ^EAsvYB7@qqjpRS^1S6#^l6{eg@}o9yQ0MU%cX{g^zlA;)3w+3m+HVa~W50
z4etg05Ar5%qy4n<k4|(c*WDMq*N4h~zxRCH_kUgaM>AT`iWJglL(dn7g*(FOwa(xk
z`2dgb1QkEt5wi7fhXH!tJasU#?eByUWZ$QTgd%bj#$X(Z(euM0VItZ4W#jQ=G<F-!
zUH9#jXXjXN&2J_iX&+{jJ&o^(i~E25c9_dO8KtOfm*0M^E~PKPA}ql&tU%vq-VUqC
zmc2W|8gebxVLj6Ocd#FJgpKsB`DLLKo9SDTqi2y>UZyWnX0OVQu$_My%2A2f;vJzH
z`{;GM%R(Fn=!eiiPa<)+EYu^ae_CdgeO&*f@4pi&%u5_`Z)$>PUw9{UU3%wfJ^Jft
zT<#t9@;mwo@8~DI6KZa~6KWs66XN~vggUbR?mHnd%QH&KLIZhBI45ulF~pHXGty{(
zzbth0cn-SIeM{Otl*XfFA&17rW%gy4g=Vy%^<h~^JuVAr_q3r(d^&}b#TlGKEL9e&
z-+U*Wr`H{p{=H@4BK<NN#49Pi3GbjDSNTQhzee6b`s|Q!le~@8CHe9_`4V@y?;$IH
z=Cn`U{4%JyzC-$VNdFGwFFQiLvXvMj|Ldnlebonk^9WBcfWdLR|4}_SQaMBWNacU5
z@;^cONB0%&@^$qXnNyA$7e;HGmH(B>KUx<n|1*?-v>|JJscO@X!8wSi{g3U=3nPSI
zgzAIJ<zD3ymF9K#yr9kA|Fm4z^Gjfq`zn5E-O;ZK*f&GN8159AzB@FGqZeZ$I@tnU
zZ|XPfP(P7bL~DX4yEYZMkRPhlG1KYQWNezT4f<@%MLZ!M)#`_g=aNZ8^<$}P^RNJm
zumruo8XA_7(KzY~vhU@gVHLRsYq1XNu@ODreJ3o4%3r=R!o5B6`|*%aM)rPvh)FB&
z8Re6d;or*d{43G-L0Q;GwhS2(4n%&xv*+$}`QZ>ft$t4x7)z!{W1^kLJde>&AgA5R
zUNZh6%x?a}_-(~0#Bg#*e)#D~?_z8Sr_Sbw8tqlBHY)z1_OV}^b64AQMH_To`>6eF
z)Q&bSf6n-tvCs|1_qQ0|A8lNJyzzaqeVX}!S=!+l+TlX&?pW<Evct8*i1zx&JvWJF
zq!HVq{hO~%rPnpFzfV3Fe(>IgkmYW;&;CBh{yxh7=70G5mhj`#e;NMv!gs@wY5#XP
zxcb}Shs(bme)RrdhC_?r3de8#Rd{#&{}cYBxSi<uEBpA~2|ro>-SBrO-wwwfza7fo
z_)hq1i_Xso`y47qZV7u(VN!GV{%_m6`<LNw&TbCpUAu?_*UQ2WX6y*xFWD6KAKw%X
zjolG`l-V9GyVm!^;o*pP^S4W_uUow>93H+c9NSb7{w`AxuDX5=H_&4}#Jk3TdY#L0
za<}rzugYS{Q`Uc14|(3cUHe;Vhlf~Zc&M!&9cs+g)FIwKIwU$qhWe(Fp~1XP^5myN
zlZ~y7x5kH-566b)A)gMZA%7lPcZ>}cFZs?G+&^(x{!vff@tn%9+N1rg5bn_*p!16F
za@}`2?K>emBR}NO?LBAEobX*!zBAcW?YkcLT_5_+^weSB=B?+9o9RcP7D+Uq9?|}a
zN1pKn1IYWbHVh+Bgi)v&SrBT+8)KML5b97rtso@4zvRM#(C~Ue7~`ICD8@ug##Btl
zOti07|GcICVcU11o2{S0pq-G?{}^tMDVlrKKWN>e{@J1aA=@_aYswE5K|N$s!ffHq
zMfy^HNLA;DQhL{gd~Ha6m`7iLT!*lmJY$w;VG+NohxuU%xeT$x`C%2g2Gu?J>aTqD
zm%6K_FjW5Pg^+w`T%+GShrUq)k^SHEh37)=m&`?c{W<HYl=tht*A?@@!dQp(*oe*8
zitXroSAFoy=Rz6194)idkz^`mPKi8#XpSlS5xbk-bx>PiTp+Vl-M?2GKxS8}|L+!t
z+Wx{2e^?mmHn0n~unTvv7sv+g6@GgJ$8Z9t5JMbEv_EA3_OnsRuGiV$$Sh=kZ(@I=
zaWwmPJo}eyxxoIt!nVE6{w3SEt9BHIX5pmKi7aBmtHv36-TQ3VOU40e^1?aphOy$t
z?oE*Or`f%Wy|3lo7h0CS7|y#V<+*8Gq*s2RACu2MU2S~8xeeX3#BYxH(R21?)GTNJ
zul9_??Bs+nQ^L-82HC)USr}Jw4L5KTw{Zve@bUPs^4x{)x4dIy^#yZR^#86%f3@@r
ztL3Eh(^K^Haq0K|tHu|F2f}%T+fi8|V<pm$0eamV;wHU$^OYSmyd`e@5?jP|TTvLn
zuLz?s2IEkSiI|K*du6ouqfb1Qdpc&K;;+O}{m}c;i(xKz&j*E}A{dizlviIe&dP5d
z7GM#UU>W*;{$f}`uEH9u#X77<i#B#68I8TCw+s!N>07ZK(HMLey5;|j{Ga_$``a@#
zl(|-pO6<b{9KsPCL(L}r^KDT-!2BTUk197v(i_mItPhUwEjGS4-}v6^#upZPzOvPZ
z_6f!((1|W|BZDk*7~H2l()&R)|Ip{1obcRJ$bQOv#4m#}6z}KnjR}y=NTU;3#Ko-+
zgX8btRBxTuFL+NsK_4h_#TqvERLz%n$@7S*uP&08QN7%`F61@bz)jr79o$1-g|VLc
z{Lu6F@(6eDOGCmV@(Bizx4=7>&xVmBP=rxP_iO(#jvm=0ov8RAKNNH4PWtW{c@*8K
zDbT);e`;gaO?zrnCrY&Mv$XG1hJ~2tl&kM|-BLGUj4&pm@@3_!(fggm{ynR$zF?f2
z?A9;HkXh~=*|?JZi)LZ9pmh%WcM|&-ZJt|od{~$)ys4OunTTby|Cmj$b3Kl^^inkR
zh+Drt2<kD9UyJxHAQvHh*LXg;3@P#L+^>DpUs%Dt3fV3C9Xs?xc4^<p8u6$Vuef;C
zp`M;Va+~z)hph3NwOEJs*oe*8itT9E-gjJ=e(~xOpYB`IuaA@^qp_04Lg^nV{b=Fe
z`cV4amnPe|tD20Vu+MfWqtP1M+1_`oR{BqxGZ22A@Z;~fho1d5`(GTBGsH29>)Byp
zAODv7)*5~L>G(i;wf-Q#Lr5+6?H8(-aD@99vL))J3BLUtd0>{c1N=^*LLHgi`ErQS
zdtUZTvNuS>^V36;-i)ZdNt2z(;tbB=JgT&*7s=}PtzjguB6cY+TqAFw?yNG5M2E7G
zQbzQrqcO(X4F#cQrMAD&b=M->KC(d?3S8%x{r329)9-KN4hGkCN8`;$&98EI^vHu#
z*dAo(hsrh?^#P)~C8}eZ7PCRn!ZvB1rtP1k4=_jDKT;onUZcF$if0^kw>-OF97Zea
z<Gt%bhxd8L6AU2le^Liy1d1>U?fTgr!@XzsclllS4DS_#_Wwimzc%4x``@=|ov&;z
zX8)6IzD4Cv$6McF4pvxw|JfOJWHBa2?q3WG72kO|^!$S|!rj|wJc^9$gXxj`%ktD%
zb_#tq=3@6ppVPK_hW<b)ch^~Q>mL&4(H9{1p17SBN3z>}RW`OilD0*zEx|IxMi+-_
zhb^t3*A<wXHXpx=z6K4$&3_LuM@`ndFIw}v3z1#8*7bE*k3FB#mb|S!pl`-jY)2W&
zQHg!1seUEYCd{#ub>`gb$pm+jY#9AYIN+W`ID%t1fm4Vfj`l+H|6|SnPcZ+FZgx#%
z+hh&Y<W3uZ-Q<h_G^1s{`Ty6=|1UQGUt<1$7C&nYlEP_58l92kmqM1_m9l;TXXxjU
zGq<1JT^uq8i$iy{a|5Ore}Bt+(7r}xI7UD3zKf`)KU>#unY-cx`O2Dyh6~<Zk9T;*
z`Ul^j_LgzlyT+f%dfzHRCb=8P>`&zp-{-2|Uqj`ml=HtF5^m6MqJ2j+{=~*Z7rK#=
z_UwJ>N8@q%Z@%#-d9H<QT`g~rY3?@iwlMDC9v<Kk`o54C`aUWSPv||rA0K)@$P2xn
zFAhBwL&5;Ryssz+7=adL=*Pyyis+*-22r_+^4~amm-5vqj~3G>BFCM5-xz`N)=f_4
zHx<)S^`X2>RzGHc(`VD?q7?Bf%BT9Ee)Q8iVsQMAd*>V8JEHtI&wYzf5%TqI^vUU!
zKi8l6Dtny10;{kFYq1VB3FSYd{HLPuz5mhP|9J18jLLsB|FPbE8?hN%u^nY7N4tAF
zrYU2Sl>ZA)%T(sN@1}g^l(WVi-v4s%f3^2dwl4PmOT7PC-oM}Ve7ZQAo7vg>#p1A!
zJb*(uf@3&=Q^<ZZKg7sgTeK7C`{0#9n=Z~TiNXHI+w2^AGzQp7_Wo2pAkW|&`dULc
zPhP}jTt%wD+B-5GLBB@cK#O_dn`CCJHci`=dqbN>-sX1)UBk6`6SQCb1>qj|13bbL
z#8zv+FhH+csx8@OEPtUqgTeM|q4w)_?H5_&dhJU02)_>XNFa#@jPRQxjKUZ^9e=lf
zsY(BayL~acb-8wawRWED=FVv2v)cCDCU)ytcIyRpEZK6M-Abmo(`4H=ejh#;iiI-~
z6+idg{<Ck>;~xH#`3<IXi1GLDDR;(ma?+jU&XCc2nd+XZ#B*UfITNvJYskpCsOBG^
zFI|g`@pI2Zk~^B$uU{^mWR0}fzA611+(-U@(;uz>S>ShzumsDn0;{kF?FZ#QZDFT=
zT^G7HsoRl74vqKaKQup-|K!Wov+`e)G@=bHx9lkq&N{5eMx=WRq`x3+rgtgFo!Cm>
zj@%aIcZl*kT=|`l7s~imSuapdRw5?f>?04LI>D~rq+H<;clNXT?en$mGnC^umE*<A
z?-k`YvUhuoQMo5t|8s(j*8iL$qq-<Y4z54=w*J4mCeEEiGg|iR>yw?R*!e=plIaW^
zV3&2t^e%QmC(hB&BS+67gKpHkuKbHfT)d*SGW83^YlpsvbT)KE{g1+M(eEzfDk{G7
zLbyiWz)iF(&mGEjC!)1}-O77rwD&*K`$yw@#y`=lu4_RnJ%zOU+O{bFf5rCy2|KLa
zIKVfQFJa!nJv_i8Ji!3+7KNQ3Nq^rr@H=ILJ8FBQI<RWDHTz`sUi$&~k3uZPF8Pvm
z2fuhB)Jb2I#suoofaqJ)>}UTT{FD7#%cf+XCfKJ*vVrYAOgLjO4#k*=$(V}iXy0P~
zeTVsXvdeh_-Dla?7ueTiuAluc%qF&UGg@}Df7!h$c6u6ZXkq`(6wYkSMJdwL+=F@a
zu1&@kHWY;g^hL-W7WPW>-%HJZvxArLtKwcpu0V`^yoy|dYI=MMyMC6leYw{m>3I#8
zUJUh<q;s_S?~L?cav$oZsQ>4v|B<9O;FI?AdcWU@&B*?xGC_X4{$`i&kM7;R|9;<J
z*^0*58&@iSXucbjHRZ2g`6HjLzu7K~XwEKL^IJx5S!hn5to)+-<~PEaAuRF8Ob|BN
zCG2Re&k^~d(tTB9i^4wg01n{@Vvnu=y>I<6y-t~r;{^Q_8s3n$ZPK_&TFDr{zTdnU
z;$#xdNTU;3oWVJq$3<L5@9&DjRkA1AV8(8a<^cxn=P&8+^S^<c$VPp@sQvx2bvE4h
z@BokS1Ow>%^SqGvKWiH>0!0{wF&Kxa{VgUZVlsL^vKEo-`Jz5B*|L0Sm`TpYT$Ca`
zUirs7de?B}f6mabfW8PHum5|juQXEm-=+L3Gx6iff0MFtMH#uSY>*8X)K3@e4;e0f
z#>1ER{W7dV_PYgP4cYG7cA)bu-v-^@X$Dz(PJPg*Z_%`0`QNSlZ}9zH`*{7Idn!NU
z{XDM@7S=kf$3|>MRL-OH_t?z0mAlvb-cFXG9F?eAZVr_^fI~Qf!S$Qt&7acihD+NV
zdyVL)@bUV;$G-na<^L`1527}@4)yc|lE<}Ehlhm~QFw?WiDsnHi7eXLu^o@~1@^Ol
zcWV><E<<LKOR<0NvVYyv+|O1aTe(x0*uQ8)%YC-3e)F!BdJJcx@IQ1<k1&31%w~qL
zr&y!N|2%S|`EzF`coun)UzO)n8mHNVSi+ihepgXl{ik+yi+&CF4J0>6*HY;u6K`sN
zXG#AY_wAB?)FXi;8iae(Z*Job?%@F*;Ry!NF1>^E|DN?PzQ6DL@%(?3Rt(PnXQW@)
zZQ|N8T;J#~<bRAn5z<4XpZ_R&*J$a-82UKme533p=`WCe6!WWEYz-NCTX~AjXDiaD
zqMCnPTBAPLbnclL9M>t4&ex?AyV%Il9-oTu*)z*#u55fY?D?K{?`uQDT=$n^9u{B`
zmS7oHU=@0QZ;XWO`8-=Syc*Wh*I_+2Vl#Gsp(Jc2x1+Dom^E3BO6<b{976UNuZAOJ
z`Eg|q(b!x~q4N2rJiJ5s+@gH0R6db32H1ec$i_!Ao4*CE^c2#ojsK||+BX>gGZxv2
zF5`pU{G+k^!F7mJ>`U^z6F7x9dcJAj-9HQsy<heW?q>A98u|VE(DO(2Hg_koID^V>
zy%zR-+g|@KOtAi2KZD=1^@Zo@7jYS#`Wap5Mh02raMiVIxPhCvjXSu9zCW-(zWr+G
zdDR*j?r8ntBk~Cbkhj>kA3Zb-BS)YJqY#_p43`<^Kk0SqusE{cDh<WlQJt1Vw8o(x
zHRDS|?WEEWpH>>`$oi7fkRX%X4P?x7${*U}@2r5G7$b~{7+jb1uJ`M{jv1w)6J6-O
zRLZt14O!$al!nIDrJ-p;X=t8PYOg?PXq{3TQb><24Q-?ORhNdz!kdcen29vMRH`)0
zrgz;E_R-QXmtKn8U11}GZp`CX)l(W4kc$wbFCmwq`o45smo8^Vtl(aSWUX|e9tr%Q
zU{Yw2&NZ&BMS4nUSVyi$6wl5b(lx&{Y~<dIta#-zo-JJ&@;~nQVmR^UoN)a97sI<t
zzYzXz<*aaQ>8x<{{^!F_3TK4BeRD=Qa{BY(@XBf7$BU<hAN71L9D09BIJk03_~HG}
zhSut_A$9%JVXLq!o@cKg937g5j2@iJYsrj#+W$x>o7F?ZcK1~2<DD8`5>DD!`qNFX
zhM%3aR`>P%5HoN1pL&Le*}m7whx*PR8r!*UKFyx5#MzM{`Mx>BIiCtOZ$2Mt%`?Uw
zhE_McFzlN#J^ande;3MxU5*1C|1SJs<m~YM@iW8z#WT%2PY!=w``^ny>WFXVhsu-x
zL)bHVdZ-vXHS9h(H5_}RH2j_KQ29&k*l&i0eZuJZMrn9=uQC)$!=blkhaYYG_u&Bl
zzE78iBa8n-_}h-x%-Q~jaCr6HP<y;2)a)$@bsZ%ko+$B6N<#g~5@YBk<~K`1_ILT_
zakXnWh7&l27~<$ySrXcpmY91f37v12n0qPFFDwZ;?<h+)?J5b4!fxT;ythQ#TM}9~
zm4r6#G+8y_)sPfUGt%fpY}%`#8d-YX>#u4XUkzvI=g`pM{-#%rW4@{`dNrKqcM&Z+
zhK9>zwD0jMc@5D%$Q$HMbe&ZPA=9A_N*SN@O{00l+pgWgJ@kD;T^zNA#*(7_mVMtZ
zecJy2u{1nzz4ztP@Pv%U{{|xWXY~6Y4h?xr!p=S3Q{{K^!!UYmrSiXYXc$4SW8=h8
zL?4BQNo+9nLqh#fU!uMj&LfO*Z5)a*5tA_$HQJims2$m_9oenj(B37;XZr&i`?c?{
zv#-&DR-_iPui4&hXrI77N9QE=`5gB76!!UO_W5}BdB5LG(l&X{Ow7hyJX?QQO7CJ%
zN9zyg(H9`crq1%qq=bb<{Hort4vt)g7<~n~3f1BsUoGA}*79?&MUvjYwy(c$ei}7R
z(wA}H1@~Q-J~EnHNIum5v*UK^C+*S)EI+GH_bv9+5A`#GJu1T5i1thRfA8!6_2~bh
z`=I`xb}M^W|8KAMU!SjOi~j!({eQBRJEd=)MjKkR*Xx9{8C$U(=^fgC{$=#8CSlil
zKD`pr+Py4y=A>tl`}kE285RzZhY-`3IYJ&oHUGhJ#fSR;+@~-&uDD6x@9xua#-7iz
z0r>ZRRXY(~M>EpsL>6ao4(AcAeUJ8__PxvA+!n$`ewR`4OZ{8<`x-q(rt!4@&;I<f
z{rk1r-)eb3qYRvM4Y|Ye;z;k=_l?#vCZhOO`wskS=FbVWo92WY!n=vvxPyCmfJb<O
z>|Ob2vAl3ep138i7~|<WE#J-3FS{ZidLM0t`d#u_irzd!pQ}W_bg6uc8hzth#CPdm
zpdQh>qL0@<f2S}EM9=wVVaWSm^wTf`MHq#T?f=!j|8n1Cv2Wu$5AO3I8&B&0C-ncT
zy`K*De=Ym}IQ##wIS95u&)3-fzcvQ&#bKdW-)M|5$DtS#QSn=Q0?4VDj+vN^zVG-3
zzqNj5VP2TaU5a^Fgc!eSETPv;k+w<l?N@w<-@M>mOWQPQM7?pC!S(NK{WzPyj;v=7
zC&(ms1G)2-a|Z9e7|O{7!dr$FScNrczry}!_ji(AAF}@+v;UE$=g@ea{clY`^L)1c
z>K8*RnOe;LC)<SEGKp<2oOM`_jY#uLvDr7%yLJnkjk}e;9l5>2-sKr&_rVuK8NVvy
z0Oe#QV)T9F0aP11h_9Be(=UcY+((c+`y#uJ{V%==vWDGX`@Z|;7=K6o6y*X*WjNaZ
zd(3Z6;1pumS@_geDbE;(;%;|OhkO5N{LT2gGLl6OjmlZm!#|C`sY6<?EB~o~8h=X)
zqZut7^32cW4P`pb-HB)qaOKXY`+vK{t#g9BHC7%IzuY$WkJQGiL$dC@DvzEa&ms09
z`(Isjo?fS{M`wjzq+dqE3~5^_jZ3BVjUnMGziYUGn~28UZj-%i`8#Cpv!la3vd?(O
z1M(4`U;uevRUXj$m;V2P{m}CT>k+t%FbZQZ_`bAh(fGhP?qc-)?8PvVoQj&Uf7<^$
z-~7J5dA+tOp)E~r(Z4_L{G*?Z4%xT0Q_t&Pylfx3-^|2p%tiad=!_%tA7s~lHvKN|
zA6ewkc+39ZyY~N*Eo}Z)dg_urd{^xMz2KY)Va&q<w4AnIfE*m3yRZDvmth6|+5S!6
zAv?)8@GWAs_DFoi`6}*-)+??dtB&S{wd6V^cgRPtd)LMS624h{i+sj!y=xn>8C$U(
zW$62@cSTlWA9_Dk6b_I*U$@?$jMnlFj<<NH*+y#~g?Ehqo|o*=(*B>IM`=rW-YI(5
zpWaQ39!Kt^-^sg~>)sV=v^BLU^=?MJOV%gU!`kPh_PK#P;x|b&BaKdEaR%qmu1)Vi
z=X>g1bic12M)rbw`K<AGeSoHI`UeZO^=Nf(YLa?cpP&s@>gn^sxrocSir8WM@6^}V
z=ye%m=&9$z4f;(qXup&C4T<C0aNOp12lwy*k6?dj(CrI5@q~K-c}vx$Kh^I1jJ-r3
z+@J7)^q-afE7Fg;ThiYz{ZSb~R8~f~w+N#!2IEkSiI|Lb<+(!{@4WSNY_9w8C))r0
z>VM@d`d_AIw4n8&^53tFlWl(2vf3PiaHeA>W+NJ#Pi?X0fgbHC=|m}g9&&T!y&0ad
zTRnj4Y36VEFG5uAmypX)RiGb4Cd98EgX4bcfmQq_dmnWd<VCFEUW<xfXph4`gmv^!
zzQ6U{8?hN#w()kd4CSbd_`tq@>xTMXcrNVYK7d0wf@3&=Q-~pso}cSmk-eXjC&)B9
zk;NID!+Bi9Wz-nQuSFbns7C@xG@wzv(WK63CR@}Wt@LN>kJ{8N?QHyxS^C~3(O8oH
z#}NGwbw#uXIHxXX(LP@FoNKs&n@CSk5A(11>iE#5d`Ej5?(n~doU)zemqB-vbHo<Q
zXXtyEjkel;o!@&Oe>Gm;XMA|z{zrI%0pu-{esgof$X@Sq1X+Yp7=v*r#zag;pMHMD
zN9O;(E01vZe$|*KITN!n7tx$)DLD@dun5uKiE92!=+E|VFQaE`UJn1?{o6G&>_0~w
zb*Nu$UV6Sc@x|t;$=l*xe!-dk(cyLd&e0dnDh#gw`Gfa+wm5W{&+a6<uA37lGu&A+
zx6=Om68rB9?7uIx|DJ3eZ*JZCfi%6XpI=9DSR=f(ScmmU`)$hH_(pn{`SQ-k#bGmj
zD{`)7-}j75o<(lwS0$V>vK+Bp=2OXis6HrN8R^<x91d_FLh`tDxu>2?95(+xMfwq6
z?!FEDxA5N~{bYmr`y+mH3@303F~pHXGul6t{)gh&Fa2cqU2*f=EOKZZFa5WqpMMM4
z>b?}2=58ZfuG59ni7d|G+4%_-UstD05Vmw?KWpyRwH$YLj%Up9Eb>|Yt2%5gg8WyV
zDh}t|cM&oAR^RV3y-r%<;*;1WjhmjPElCcJ;~1ZB*`*w~FGXG>Zy-(op82Mm^e%a-
z^R9e<OTK?>?CO$oK)>zU9aIP-`+a=?ZQ(uc-e3*EA72R%=sm9%hrZufL-5U4!Xxe{
z7(m|t>Rh7&eNg47j*KHws9#Dp%+gOq#diwAF!zl>5k_GQ#-SJ;AIgJ0%JySroXl`{
zlR0%kmTa10{m&$Q?-G6R3CgwpZ>zp<+xu*kk@g)5V=|_qYPbFcSxv?;lRg)vhzqyw
zminh({Uh8Y8m_R{7iw#kYqL?MPRRZw-*}9E0p|H#<p%{}4+i%as=Mn)>IW2PUkcU9
z<O26B!V>iTCWK|=3apCU-}8S2>dBsulo9UUmy5$%GP3{IMefh5zoc#VH*bga+#9hO
zY4v&vTj^b?r+w4y^fE*?X?B<PgN*iy&*oQ-O6<b{9KsPCL(LHDx)I-Eoj2RNUR#{d
zCMRq4574NsZ(@TrKi0lK)V?3pz9YR?`@V}EMs}ccgZ|l@`W!p-J;*F~j*RA3b{@6n
z(zBv9s;9^p;z**V(E9Kl-rr(t#7=6TtF_PiD1-A8+UmyL?6`yMx>|OfK1{3YDc93v
z+csyUEMy0wbEWZlbkj3<c0N(m{x*9~<!7CJ_c`q{J$+ml=%jb;C54-%pFyslj=}Q}
z_M4aBS2b22hP;T_`{p0WtEi^OAM00K(3j!9fn*1J7xnK+$65BLK4dN8`$MQZ?mRd$
zk+DxV6+**xWzTPJ;|}iO0UqHA2GDN)w*#GQ@h)`JGss?72ISAiTK!)%3$sO-t=@GC
z>6QAw$bQcJ+t>8tF#<&xg>fjxV1IL#{9WQaIPS_HvLk<I-41;!qBXbE$(fjqxfncu
zgMTT#YnpP2dGrOyjaM!at-<ZqHrHLywqucNOAxJ%Pa<(zS$ogkplWM(_^-e!tU>I!
z@h$#q>Fcl_8?hN%@yY(ncJ8vs4;9+n%D)%?&l{)YK7g!o{zK#u97B)#?gZIeGc24U
zD?akyb!6Y)izk^xGt%fpG^dv(&)^)+Bdv@_{qKwPt`g;Mmj8{TUq!Cp{fO2GbmJPo
zo<BNQiR^v#xp0%bjXSu92grWsh46$NK;ClS3M0_>lKGp}$`7h;86#4rqH~{$TpNYh
zQMO~v3!%~&LY;abo>BK4R|lbCi#iB3Z0lOY+1$}QdA)h}#8UP@8c^w(k1$4f<4}x=
zn2f1tzia*f!#}P6-)~O-u(|o;)-97c?#3DH|2gLWhnW9I>tnV((#8eau9$z{=J|;4
ze^EQlE<esLXWu934IM9r>7FqYvoRN?n1=;egmz;99Y>83oHYJ`ZfgoM#u>5;jX$7~
zjo-x9Z#G8JLbh_Jnv6f74blGlCBj*T6<CF|c&0`dg*Ei9k<vEY7zMrJr)<70(#D-}
zPd8bvtPSqJSC*px--*q4UAw)`{ng%Ie6jbs+<P=P8Lb~mayNJ<^<=`m>)p2zo3Rz!
zQHC+zc{y3*8MTOy6%Oj@i8p*7MB@`5k1xnmEf3ijzJJPheYU>jUmc&iuiPVtO3&Db
z12}{ui01c?k&!(go#_{?4>-Yn3NgfyL^E3M7lt(1i7d__y{j;!dJ4lide>6@hJ|_I
zJpCdD`wiUD99TD5<<Cqm^N+?4u8}tod%ZAJPbdsG>2(M73-;<~(C?r@o1NrN=$k})
z0H2-1Q@+@qCf9quH7fM}g}%n~|6Pdo-#y^h_o}|gr^kdx^vbXL&$q7(4NvF;7@YqX
z);}Bn<4*3@2N~}_RaPi(7=a>;!WfK0MSDSL_pUp3DC1<;aAjIuK6pMGJvYSnQ4ch!
zBbvz;_qD#|`>ghTmb3pB4++J>nTW}lis_h%zAvyTeqI!!x&NNOc+vTi)(E{@6lS|t
z9<`f0lzYtOE=Bei+TZ#1>(WzXda<z#dgYg-_lNoVZsMno=<XLkGTS45SDbTt`1x?6
z;Q4U;dSQ6C<c08eGhPVC>;*b{_fz2~C7%j^du3!e(lk08E_f;Y_}1v~qtw`NXvU|*
z!8xA}KP-GX{NRn3!&UFGh5fuFdbYN40l5rm?eYq86}k><N0C`>i~u=gvBtHv_!sjF
zyPnR!SJe2wOxM?=YEnViNLG`vY1XIFw_-cWP>xFM!vP$^5gfw_oI+J?K{$0OKb(9&
zKm7Df>((ZW4}U$&d1tQ0QF-{&VbArK%x{eiyRoluboiU5mqOCDW*jh=_5=Tc^8E{+
z^54uO!=X0{!;iKUI)_pl?D~<j!|g@P4?o`jg88lC_ORRkhst06W7y-K=)YKZZgU3f
zg#5sU4kyq07JgSX<d5Oh`|SDA{o!Y+{_vk1+8G=22W9^cq2};o{o{wmXdi||(|u#*
zzn3?EZ;bYSXzIBe8i#zOZv9<o?*B;M`p_7|hvAGc&*3~S;xew{8gAewZlh*@eyBa1
zAL6L1&JXpO{E+Z&l8^I4!~J~y?>~k+?z@Kvc!VbyK;HkR{@qq!|3`uS9|fU{>^A3^
zA+y}MyUr#*njf0bywF$`TCMd@Ax&>X)%g67{na1CFky{A5n?OzL-nluFp6HcTHG*(
zJ`N4X?Xf`u^(f}oVmy5!ITdO3&2(}mQXNn0p04H5_;-KU^YcH3ESn<77U)(+GGxtl
z>0hXAH(yq_L7&68W8$zrh(1O`hyMM3^FPW<x$-hwm~&Bzd02pU_jb6qlk7_9)1>rm
zGWs}V&T|{}VVbTOpFoT2t^A*@4`}0FB#b4fx}={&RzDmXR*<U@8$UFxA=jeLSVp`+
z+%}2haB(A(+t^#5eJ-qXz4y1yyCi!)>ny6sZLi2?ax1o@@9WOzA<I!2xqoTz2wAhZ
z-}p*@h+patb*R7JZ(giFB+-D$(Qih|w`1koOaB-Sa34a&UrHm|w+yKN2GoB8zW;#l
zKj8Zhu>S`_?%_aaJlL;)-XEILg4X2&^5}s2Z$SMwU{82|I3m2tdhH_{@)$k6sXwH)
z^@kJmuI2rq6Q}4g<c<papl8hQ_YK9HU)6Z=C7YvrX7z_O*@@~B*GEbhvfO8oyfP3P
z#Ib&ebfV_MKZaWAilc4<e<YAZ1J3!)d0fP0T*Woqz)iI8mwxf-JS_bQ=}(DUhd9<s
zKN@=ml+gj>&(ivM!1@&F7l*Vsw7IWk<v)ho!bvMncgTB@e$qXD`+(jxN!WNqe}ddd
zVNdf6bk7(VT$ffg{vZ7r!_#jR^PB<q<$YcH?+%3c8v~&Z72i_Nm@iJgC0)y<bBlD+
zN4T%|3j?8u?0LKY>HfdxhlWwyV=xZIsQ7d9|9_qzCer)9G7u({QxTorGo8$SSvmev
zL6}KTD~~DTWV7jA+UU-VIVyT7ax3K<Wi3N?YujtbD}$ri|C651Q$=fg>Xn-Y?N?G+
z&VI|>uiq`eA}ql&tiURCJeCLVD|?HTK{E5YaD^S6J(8t24Oa%0^A_c>d5CzuuMA$Y
zHi$c#pO41l)(EHf_x7)l>+nhYVLkUoY{qt!p&Wgm(|01PK4eRf&-VWvpby$7N1ZWF
zkM=Ca)dxrD$1phlCS8N+Jm`!<ey0#a97!}IjXwRzil642BmBqE$=zGw8<J;m4(HLL
z&A3SJ{?*uUnY@Z<ZaMXsZ9(s<);Dp_4f;*wjtc`B{@n@ve{-(2h3wRkPv>3hXGG)Q
z?CUAo-$~B0^PBSb?EgFcoIjy3-p+jogX4dH^zQGnuODjLwfkLUH+N>~)3!aDmu+0l
zKA*tGp2I#DZtGaKHPYAF=U40>IO^OJ;XS}3JVC6<{(%nr2k6n5e02U#-b(#2H27U~
z_CSJvbQZw~el6lvM2<q*n8O%y9HMcE!To=4TI0_>5rg{zxVv|lr`Tftf132q@{AJc
z7r()E`SU#k(f+^5elr!*F%z>f7p0ho_7DG+^dB@vaM=3C{eRm3C$7!n+CsMWOFx<B
zZX=_$iwlIa2urXGv9r=Ijl-h!dw0?PzZLvfp@BO&=jk`9C;!>`qx0ou_YCgup|7P^
zeo4LhEq#+I`hSyr^Ag{Ff;@oe{LyvpS&zZ}f8mADtNrNvz`1mv(vI-&`=NPh{pZc}
zo?jbNBDbRq<*39yRGrQX2gqvj5P1YK`Z4kZ>P{-(3HiQS`PMEsDErYF8vo7lKlbaX
zC@jR0L^IOp#Q)#%zrp=KSHuk;@Bh(vADsW0<~@1F8Jx#OT*g&g!ws~1j~)0I{cHX0
z+%5fo{rslY`u}8$@LFdnn{)L4>21EpPTyl!wegjN{=RhpH-&i{chIM8&wkGNgZBU3
z<L>>T|Kuj4@xMor`vrMQ{rQBx=gZC)_zP<l=;`-7<2u`AmGZCbcV05*Kp%lzMw*)B
zfev}$^pJ4Q{YCsoVGPEh7!xrWHN#)j2em$YEc+bw^aPT%>~l1}&%VCPzJAP>e#pK)
z%)Vy3r_q*rF|@PmJKm80(1mVfkVOtvs|){W|Lb^r)IBq{*Zy~72-E3xN7=py?Gd2Q
zMngZ__dXl=F<W?7QJBjwdT!<KoKxJbP8XL_e)EvJWn5sJae>9wE^seGc9wC0NyZ5l
z8Yh_lBKwp5&wm+KU=?ChJbwcFpI#>}@eKQ)z77pNe|o-W*Bjqh&u=3(V=K0!4CScA
zJ{-Uy9Kqmx_!Zy#y8SNPHLn~0TWV|ub#EG**<p<4pmrJ!IN_dC7;L|_;g#B}Jr&j-
z@JpiIJss#o7rL9gUt|%T_0o9X_}^pWf2WQAp_N|>X|y5QlhQ1l?02=7+VC_zJyIX_
zp>~_zRia&;<r@7Aa$D%gjMmQL9KWjL?0fPeV#}?ABd?;mNx9mjU%icO&wT^Q8TziM
zU#c&QnmOk8@LwJO^qbqbgL`;@M|gq({8z_6<@+4|TjQU3|GRbtBT$6CFZ*u{vhpX!
z?SE?=!QPzc?1`Q~*l))F<MWKCg;2~L?WvqdPR3MBM>O^l)xk6AUCMd1c3?JrE^_0Q
zKRnydQOd99tM<B(y|2or<RUD=GOWNV^ct63L$1X-RQ!Xp3Vtgu)*8!0G^Uq0EFT?}
zkFef78xd0%Y$mrN+7nXVZ_b_EN$z^6t>nKQWhh4__C>N?dHSL6Q)v8utnvR1#{adO
z_1et@nM4Eoy>XiH|Fg#b^$-5V{yNWScdbMJq4Ry~1MuwJ{_JsM0Ehj?^#h)F2uE-X
z(cZk&73cNPyCytcTX2dVLvD&Xkza=Fo+V!Vs<in@vN^hko+dj{ebDv&Pvf2CK7+w{
zOH2JG=^$%HD}TGZH`JjXiM{IoChryJ{N_9^;xew{8gAew+V#6TjJtM{UB+X&$qaXv
z%qioIC)NK6^?$YczeD}+Tm94d+acT*M!9+~TJyORcewB20itoliu%0ph`#4}W&ium
z0;k8KyiewRL;hQ+ZYM{e2&0hZm%<o&*A)40lDtbVMzm%nD?ers?N!hI=b>RD|Gq!y
z>yT41J#y;<%p{|>W;WSt{m<OUtz48wZu9-~$OTx0sQs@n{};9YOSqRIYX4V|tMFfI
z|7+9*wd$DoEcL`f^}<T^0+RFw{rAEDM;xxcWPh#^&RVQTbhc#W%P;GHdjI#m|A*fH
ze((RV_rJ^gM=s<2507jBeSi10^!(}ke=>dR>G}T~g|!)5(W1WEPNuFKKhqy5qnD!+
zozfERW$xB~XXbcs-fIrA{r2CuegM_G^?k@_PW}ja49QLM5$ZR{S8qh+Q$GVITswsr
zdMfns$lgEKw~O4~d2{4uV`P6fKBVcL=re{~@keV5e*bdFM((dX7tWC9a2^-&EFWE_
zM|+|xzhphaFJ2DUxNqPlZsQK_p~f6>ZGrvw_R-go^+W8xA8!Bs9Q*Ha)jbdJ2v0D8
zyw&PVj6nN)_TOKz|NfHw_s-t$e(e1HyUyP~>im5)PO|@gn*H~}YO&wGm7emfv}d(h
zPtdaal~5#{Q5b`9NbhwIis@09ojtFFiS)_Hy(esb8GhZKGnHS}=~u#ZawcLIUJ0|w
zxu~YctEDUPN+{)?ha|mWzjW40r)xFl?`si%Z2lkh^aPS6zMF6T$@RY#-^>dOgs}*F
zjIUI_>-@V*`hV{$V?FxU_!sMc7c2kEeb3d(KRLMn`@a4k2G{?ZgI^-NWmth#NWW!`
zz#Ha#>0Jkfy;onJz7B)yf4MWeJd0c(`THhS)&gv#S5L84g4~LjwF2A8GSrD<{Iopy
zp1epl>`=#%<*rp?9}eIUj^G$h;1ptrBZ+2oA{(L4(DPe!)aLwqfA>m=_T5Kgo4XLr
zi=1)Kv;BYP=;v_}m(lkbd5gS;8@L(y3FkI>2lwy*(fa>K<P!`aZ;kR1t)U@%%|o`R
zCq~eVFbZRk<`<m-G>$$v|2WAx!}N*B>9b_<Z2qxk78_)jehU7z^}jp&|7rdARKKq@
zmcA#yAWWxMd^bO|FZBJ<xzhJXH$8*wHuVDr*Z<b)V`aoOW$gjGCxx_Y(fZ$+!kLY^
zD8)Q1z#{bheevM_hn{c9Biy~#o-ZR;U=`LNTK~J2T!;18h-d46H`D8;Nt>~it@Q0^
z7|rfNVgfsCim?~AZ!Oz5zMTF0Ci`~>+mamIC-;Un!Eee?j!F#nU-cWK^KbWY58A(M
z;LgW?vj5q=nd6_Ze|y;fXhusv`<Km|V*5tt<hQ-g@4B_?!a0N^IELt2>fx|(g5EVr
z*r?PG+hdF{XIvsX!86EiaX;mrXbd4n#t}Oy?qoBnjUzleS0K&ZiR7E?`VG>#Njk|v
z`@h6}d!--s!W#VlBYFd}esc!ra2^+N8CP))?fcpP(j3igbXAK}M%*^}{utbU@4Yl#
zkp3vGXnjxGkak~Nir)}(zrwkR+qi@DSoe%%^U(*-9}vF>^hd}QiW4&E#uI*3^Mx0c
z4|;U=Lf-!&ZyN8dUM^kRq-(L?sIP|cPcD=$L}xD~u+u#IuJ^26yKb)ig0nZ>6P@7`
zqZiRDeegX$);^z&>|^b-vXdCEeI^^Yd*1!Gq4(3q2wtLp!Pvp4oyqBUV=xZAfA8!M
zvd8{{iR5JTea?NqofM|hqqDdZOP`MQHM}W54pAP5%d;yzZ@E6jM}=X!`(|P`vVSNH
zbIDT7!{Gl2=r3g7Fy@E$yT;?$70=Eai~b8Wb%k9heQjhTdwFm?p1XxRI^S!7aH9X=
zSVUHQ&6)b-GOWPK5A(uLM;oWH-r$t<_8CK6#eWUfVjb3F_ZKFI=q!Sb^wd$$qi?2H
z{@i(sKlZ<dyQ48m{Re5vtTwJm=D4@JrwrxMz1W8X_`&u65~A~evLCz{dcHp_9OBoz
z)A!QP*`@DNjwAezZg?g9<m@Zq@80{faBSPSaD2<S@b1|^3;*%4{<C~^@U1c7hfV(_
z{Ak0!4TmoL+i>{OzX?BnYm_r8M};F7{!RGn<-ZN(um3h|71lBA**`v19{hCJZGNxf
z@YwJ-BVYC(eJ_U-uARaUc8?ARj+*<jm+bq8$A=#+eJLE8|5Av#7RTSd;S9_z&xgbQ
zr0>U%N1OXHpO`2NN!OZ@Mklg(_rCve+)x<K(5r^zhf}-d*`xX4X9x4ce;WJ65HpWd
zJN%2GX37_pP5ZEh*mpJL^PztHtD#}a=grZ4UfuEq^Iu<3jy`Mt>kGzr?Ww!;*>Fyn
z=W!92aTV9FGwN$c{mJ)@iE`JBG`?Y6BW}$=-E!j_#tIUfjBji)zOmuOaML|^@bv#j
zoH<2*fJb<OcJufhORW(=7rNiHHeiE!b22yH`iJqxN9;dnMhjZc8Xvh}e1vS<W=+JE
z7sG&X^1i80g%L>KcaQ5u^sWM97{*ve(Z?WX>?J$I*u`jL7#PQ|>fno^n4E}M>cudb
zoQmpQ(sf+AUbl{ddnS_erAr#?mr5sUb~*nManzw62_&2RSH($ZV*1TolwuwhpuEF7
z)VJP=_F2x~EHPG~zUVqEfA5#SQ}Va=E=M+QP~PP8=K0z`?NF<GQ{%ONqpiU!G)FDG
zC1@E^7?zPKaur#j&Pm&!nEhOSSi{|=eU8@QW*&Qn{`25Eg8SM&*VbV@s^%Aljbt?$
zTU;16(<?vFUjHmV#BZrj&T8|nt8c_@@c(_pw?>?5XQ^+*>!1FA#C2_SG)B2q7~4^X
za{P)NUP<o50kn%tM?&1HrC%SvdyDp8-zrPyxEsZ>Nt~NcOFvqhq~G(?-c4IAdqEhz
z6-D6)+4B$9U63bm3NciC#@aVBiDsnH_mTWYX3_G{8H42D{KNP2*#-F_ngdNq?|J@R
zJH%~^v)SmEk(=Y&NN;BO)4Hf?RzX<7{~WI38e#|SDV<ahZqVy8p6|b4Zqjd~;j#JG
zhvr`on}5Yl^~^4mv$I?FJImZX(b~D_KZNelAK(!>7qkE8v)Na(-Ct+JBZnuh4d8ZE
z2HsExFakvwg~9%bHZuN}b`15}v%&sJ)Q%fNXuS2O^@n8iKLD*C8aqIGrS*r#C)(Nh
z9q2?CqO}JZafsF)MDr6>r;UMl&NvifB4WlLsxLYJhh8__vps(*eL5Pnxk)5YuZ{lX
z`DZg-&;EqXj@k6NsI(`3&o8Xsr_aLzEW#2jLyh*jc7ta~&%ZYQu);m7um)?f4(qWI
zo6){p|8uqeC)xEL`}ckJFS7L9Y4&du`=f*Xv6uaU)=kC+klv#Ixq+WK`mMs*jxv-Z
zt^G}*lKw3J?4ut*RHn0t+OTe9*+kKribJj)K}=acMxH=5Jw8LZlIKowSA3w~pgct9
z>er7|4w3zydfl~W|K}O)NloalC#UIOkb~`AzxrBQs`OQoQMhCb&GgD|dcR*&SI|3=
zMX&L?Gi1-tjNg!b-%u9Fi@1!dxP}|JiQey<*CKmf%@23Tdw76HsQ8TWb#efCYvoV$
zedoC_f-FMIG<}pWmxNLDaVSQbUkVfHUGFJ>%Kc>eROGm`+!?a_l5xf8{GXp&+hhM<
zbpFrB>zOV&tKa=IF&o*Ry&C3{rRaOvT!DF+dGx{b=a+;9^p;KfFl1`1?>pRH5c)E#
zK<5zo1Kr3Vi)ejP&F<1rn<`btOVy91>c`TMAd{y{LqmUQSmn2CuommE9viV4ThYGs
zHSOPP+P~Mdf3K<kUQ_?Qrv7_P{r6gE+-v_onva$`AHn{AdJ5@nuZ1@E#Jtz4Ii+E{
zu*M+vdTA&l%Tc{lycUWVD!KO|NpC<s5*VESS}K2_<=|`KfO}FMuZ8sd*TNxs<(JqE
z`TB$0U8}`!vG|pU-xTqi@LH%@Ed9$p2X$`>cZYEIN`Hsv;D|7e;RH@0hB%UFM*ADe
zA3D*6Ze)-}4viP2|BCb@I_seIr1aNHzjx7wD(^cjoK9qM2C?C#&R;AI=je4K#f|$s
z{URDRh?{sOHi_$y(r}sIRb0aj+{A6%!99FDUR@Bv1MaMTZUvsu2k_74zm&Q9S<0U>
zm?Rsx^ZrkH9wShMQ5b`9C`S8-`rGK-tAC4becueS^c)(O8-H7^tdcDsntOk&KaMoL
zt%u)g<wQ7@pVdzP%6a_s!SlN^L&H>hm#{hyT4PL~iRiz>vdEx&mvjHaOQCA4wFa)u
zMNHVG<UCZ<<5!F;UU&Wv_aa1VW*Sg`RyxVvua$<L-@N9G-BR<M&W><TRKAvxE3gVt
z`CLOr^Phd+EDdYv>(Te+*Pfm~xIzBgVQel~=k<H(c-Q);Pg$$xp3T^b?I=SzDzOg-
z(7v}MbfEL7GYk$oyP(4v2FPA=h5;H2<ex(Mhin-y|Bxx}G}-39n&a|sLLRA>e>3s~
zncz;64Lh74P@g||hVUWJI)Y<3fm4W~Mtfe1_!R9AqB-U0Y}vv2f9`eei6e<-q|u2i
z&S22~J;?sW$M&yw{#pC}@%g`Z*}DDgUov%F|KQ4>>|gf!IpLhgMO;R@!#z#b%+tH{
zB|34H{z!h!jb#&w*P#6`o>954x~1NHU7r*&&$!{<o2Y)MU437=Q27=0z<126X=@u+
zOXm*hq}K>z@chBIKGDC}CjHzExGkJJxQ7RLgeMq4-nXRxP33=s@=tc{Q2xmbca|KS
ze;UjFNAoV}M=MfDqYcr22M!a?2ozxy(vRJ9-<d!3uIti|ar9#3*!o#ymb3pckze#5
zN|VW{h|W%)PR>O1A4<{Mi8{>Yo{QuV=|a?ZNnr5*Z!T%4Q8(>hmj6<}nTP1?gGJ;L
zEW-+{!apnjyOqh~%H}R*6uH!YyZrCmVEmz4`AMk1m)k!p%v6*50c+_M<{{RRox<%p
zEd2e#pCNqY5S9J)uJ8WtTVW%)8L@uz5ZFqudtV)J$y`CS|K0p>!wz*DqCLj-r|oBS
zPZ`QliM~c-6Th&ghkgLPZyVcSi}bYHhvPi9L;Q~57*606VyLi>Ap7;gFgQLoOL|Jo
z(eh8C8H4}-_YLC!^eoQc9M0n+F5@b$;RbG^kG)?}Uugf2e8}Cae{hGqhX;6sXwCN%
zasYX6s;d#5EgYROH-bJm|2V;(0QxB89_kw+Gf7^Zrv0C0efcbN+$En_|G(S3A{siZ
z(--a-zZ-{Q>>R6{A(~s5$lbou{5}Tf*|)Jt-(s60JJviq8mq1UueJVPSS?M~|8HX3
zB08VH4b|q>v(|o36;_PBJei!1s)RXzGWsuq=syV>5^QoT5TDun>-fbZ`?<sU_1uH!
zU%#O4r$=WCrdGZf=Fz*hh#MBr7a_;~%_8HT?l;-Lo1`Cc)S>>MwFO9`0ZaU58CGBw
z)?h8xVLjTtr;h#XF|zA8`=6eP(v)KVZ;R|d=|?kK&PxAjdkc_88>$``g^j}5jIG#?
z*huRS#ukM#dbB4aj&gb>8VbY>3HL;27!KyQ(Z0uc<FT$Ez#&waL&~D^Oa2$;x7Hf)
zJAqS(A&w+!_OgEuD*uO-{|^1~EBfVU*}rIb$ZmB{8lA{SyzN{~@*K{iT{-Gdjyf^;
z{~KiHJ>~Ph@`=X%?BDs?|HayRvURogpG<SNk<t2<i^92#tGI^reEnzrf*bU%UBX`I
zKW^!_k>k!D^^Aj0`x<xnRjst=kbHpH5c8$v6I81k;)T+6c4!#j&RZwV*VO^s^<?78
zQ2(!EeE+U_5ap$>-T4LX>0z%`{KTBbubpemo&6|3j3WD<4`IyUb>l8%@3-vh(B|*?
zwspwdEg5|oGL>N8SL^fBr(!xfU)MgMTiMJk_HD@AW8?p>&Bk1mVjdP?5&Aw5kMHD%
zo?pv{+`TV5=aO83Rak?n8OA`!ia)mwm|Tz86#IWC<b{p&XZyT2)3;-A-_LvIfliwT
zB5OX>{v-O2*E-bG6NuKBMQ8O6^PBPuPuB`WYX-J*m!b00PuB;uUo!svzBZ~y`;YE}
z+NR^$sKeI3?X~`Gq4p2|?EiP~YyZ%O#vb-NnuoLB(K?p>jx@augXaqrvfuf2p_^X@
z&(8NP7ndWxZ6&!62XH9z!!gv{w|;>=A7|&+T{rLl-Y51ip7hKq<{8{~3NgfyUakF|
zpbtRrnlEi=j{K3!NE<SU#`-(?Rc+u$o<VG^{tS5@)o(r>JE$AtoI>u)NIv#nQI7;_
z{I<5q+}uLz59EpZ)xHgxlt&uKtA6vz^#RwoZ{Q}fUmFtIcbWg*Z~l9?`R}9b|Gnz=
zj5<EW{?{jMEMfomSpS0-v?BGP`ET}R8@dXVulL#d_5tqsoHAReeDQm>z9;(6oI9R%
z4;8LIAk(*$Pj$m1`rx^luiH;UA3&1cuu%D0s(c~u|MGs6g<<3f#7>%5AV(ouA5>fG
z42F~TK8kN#yc5^$fkeZ7eYJk+_}t53jC;qS7!xrWQ!yPKJB&dhI(xel(b?PG$kC(!
zI%_)Y9e5`#i@gJSYPENOHg4yAg_**bjkzerJS@N>EWs!3r)Au$um*en;GMlaUcVqe
zWPkax|L++V*79G6_1K8b*oy5aLpdt3579Y@2gpM>f@6`Nxrh_wDa0bTFymyjKQKu~
z=l?ZFZuNRPa(`=B*g5h~?eP?8?Dr1%KU-gtrJq5?cgD*f>Yt<D&0g<lsds}Ms%8|2
z74A8Q^SFrEO6UJCEe@CI(SOp%7uqLoO?b5CqTzUPi1yGYP(QyoTy;;&@K?e$@&?lM
zo8)bz{$IlGKgP=IJoNqAbZ`R4;6MWn)QkfP+?X3^-~>+KgmH{&RHJ|j3MfVe1yoSX
z4cx#D+`u(sm=5Eb8WM;hffyPX&+g2g*?WHOJ-_z^6won>(E$}yP(c}{puh=CK>;05
znB33W80w_w-v05`^So=n@7nKr-*^3Zp7o2~G1$0(VK0Q6j&CFHxJP)`D1G?C8?G52
zaLf1rve|fm7BYLo_<%9S2cYdm;|q)%=$IoOvTKCv#u)=3jvjH{73Y0i_w79-2QcVc
z+CdnKYU2`WP#6c${RjIC9DX5GE+`Jeoi`GrF&0T-RVb#{N_*<%;xK_e3H8D=j?=>H
zJfn}-zmxqw71L4P;u)joSBWs&afPuOZ<*6o-eoOt$8(Wyk^graf3#ozM=Y!B)JJJ#
z==CSf4|m>tEW~0g#WJkGDiqeAKQ8~H6J7Yd_5Z!^Ml`+1*N2ul+J+N-KT{r$KkZz{
zKZP~oT8pRa^IztlLYz0hUie0AM!9b>w?+Qnpzq3YN?jAzm@1{0p|Ef0PVI<;+7YM_
zR_Xj*<Q^o&w~su4su|W0Tlc?(zu7%GR8*@+eP0y^r-rva{IgI#Vrtk)-!*<@_`4TJ
zhC|Ljf<1djhwqOY9o`u+GVC5PI_%ppHvI6~*l^5m{qJrL2ksPyA6Zx7-B-tk{bP#5
zA@}wZWmwAZ88o5yvm?WMchv8n9UcCD{0rfT@gs*1{%Lq`%AeX7PCr`rpM-<6{v`ak
z@lV2!-u~lopyrRme!RPBYWU%-KQsUB&%)luKMOyY@xO=frzeHJ@4XPt=@;nzD{J!0
zV>$n^{f>48^401t$9bGUeE<Dlcp;pkpGNQd#i389Lf;$qSLia9U)VWZK>2%YU^4%0
zHZ6G-*U|qe{=cH)aFc!;cX1yNF@Qm@`~EQ$+vg}BhG-vtV2r%u$_M%}wabo9`=jvF
z59PmCH-yH-8=+~)8=-mpw$PIJUdWF6UYIGKBu2U>|9S14XW7s1*k94{SQKLdCgG^^
z)ZEZexh;hE;~sXS%$dITZ1{)7Cqwmk>rI~gBp>>BL+z07@>zX1q{nRvnJwSphxtyZ
zzNY<gNBaY{Q?x&3Xn)Mn{vhiePZs}FOvg;jMhWI(9@-vhf9zHd@6`TCtCQQ+&4_yx
z_f%<LEV2LFa{D5XPsiV%^$nx2{_vKfFkd_iu^3B{>n_qqQxulbI}aA|Z4`wS^i}8?
zrR_0G8)TZc$d010Mp)&XqOg`+kK}`*u#w!1DthXCk#<K>*y^|xnQg{RF47J;X`bCC
z;}2hdyl-A=i}uMy?SgyS1=3VlfAE^}{R8h0(cUU^eZ_b5Co4yG(f6QjoAeKs{$bKj
zc8!yMGL8}GAsY^Be{E84IIo#(S)^^VT>6(t{{mrmimcHoKjILMAh*pNJshKVHk#Y_
zn)#OW6nYNwZS6P5@2I(cHQtl3%4|_+B6CR2az2?y6+Jc1d%ZU}oN#;!nX}&OW$*cx
z_Z;7SyFNJWfZ7k_Q8KMesKb->KTf;m49?*KF5wEU;yMcB|JnG__gGl}<M3l2M8QrF
z*0@9;B%*Jj3DNi1;`?vC!3H6l=UAIuJhyQd4-tJ2*>*MqJ^CEt+`fvx<v;l8uu$kf
zo5p{?)Bj-5+E~6>Q;Hml<UMO`k|Pm)5YhKgi_wn9qOh;PHfcPreT41WZ9BBz%E<oz
z&exz_(D$37(EG`_?nhDR*Duij*&^#!JsXN$KLNevi7<)m`)B=9<W!XZY*=XifIUIa
zsz2f!fSL5!C_#t1q%*DlZ&&{#Pw%N>pR)fOUtyoV%sxd6J-dK?x`BO)wjJzSbf6Pm
z=tdqrxbGRw75_ZUM}^~s<YFvEwRL7|kXpySUd?7@&!@=@>X)#8cd~7#cm|_FVf^bd
z*Cd5i?Kl3OUfaw5H3okbeGTewxK>#Dp1819So2Qf|G%Ytdfq#wuNSrv*}ZK4-NyeL
z>%ZCYR^+GgOW?`)|7v!0Vf}+w-QO$XTrTdH{RiqdvB6zah6?P$9_+&b975Ys&j%gD
zJm0~d?>Nsl;rXI4|9hC{`+?_s(epj;`8ImK$klkhuB$vgJRA|vF(i?SQdpI9GxS>5
zr{eYW9O6C2y&2O9_v*I}VR@XuDHPfxGkj0V{nL)m;2g?*GkH{eQ9I;+hHy#P6;vPB
z20ttBp;o)Tu>aKzJ~H{Dp1kUu>$r*AxQqLEhyk?uw%g^unD;u-wfC`YUf6$=-oVap
zT*CfeF0YY=^||gm&U>w7XQOLt<fH#XI?CA}qu3vg;~Jq|WV5!zGkzP2;TVbBe)&^g
z9!*c(lE3a5>qjp}#&P{3_6M21W^A*tN^Oow<YXkz>PsM}qiTeH1F~%VN8g`su)jzf
z?OkJ?g_of3pX@6{#_|92$%R;qrC5fFZ)&$#zkdZit_8A+%x;taHkn&RUyJqV(AMrm
z*8|_jS?>Zp*yy)P`EWD270GV<K#*mq%4!!L)GkDY<6X#%(Jn;YH0?xkkFYpyU?17{
z>ETvz(++v#*>E5p|K-qdh&+P4GNk{D!@@Co66IfVTxWdBEcF8l{^=U^gSzwS{`5z!
zpL0X~f{qW=FDKP6m(?%F)i3CoW8Xb-ecV1xInJO77knSp>WdnrUgd{I-D~PMJQ;^p
zomT&4+2`HrKeEno`m*u-+AH<P`2ob!feJptxTjd2egda(8g1+J?+@0-O7IV;o4eG@
zv2M=4sDB>~JNOsSq@HbF&A*`j&8`$L;`o5(8EkX$oWliNLc#u@sLe%xV*g*IUq@_H
z<UOPA#>e*mO<|QAjCCOIBHI7=$%m*~`EU0Bfa5{m_G})p|9d~O|L<x4Pt^XO;XP}Y
z*OGNpyw|<|Hvi`t*9^sQjKpY^b@Q2HJKEU%?P=}*D((Mu+W(uh|Fy&NWRK(j<@}#+
zwr`dVTw_g6@fBkN%D3}jKW~2xdd2^ILw}sU?K=6z@pR0@Y?NRw<{^$hnNKdnKHuJA
zGAIAV@h3~^o!jK^P4>U0uRzZ%?{J3vy+r;-<qZBS;cKuK>yg~ce?QTjLwc<=rm}qb
z^sT6WplrG)Z||4Kk^h-JM1_~30v|8`b~)aIeet*dCOyIYa{3V*LlP-u(1aX1kjDwc
zeE`aT{YHrP_9@4GU;LwRnmmJZxPa!@&9f!1;3}>ow~~z|{3g9H|6;sxWAwXtI{(Le
zkN)@j!X9D(u^bq*&UZs5)5?gizY!|7yKnc9y5;-0=KJXNecWIlDm&^i)cL*agz};{
zL*M7LAO7)~Fg%8Ru_%lr``<M-$F~soNQ+}tMhhE@;`p0>g|^+Lp?!a;Hcn~itSaTx
zDpfv}DxXS~Po<$@b!li^R~nkOl!oRFrJ?1uQoi-lkaJ#Zwlq{PDGfC%OO4?u4YjC4
z8X44Mg8w-nZIdGOXD}I4F&)+7tU(I3sN4Id@BdBy_c!_9-wZRIGaDtCi+Pxjg;<QX
zoo{OYzp4HICja}J{O@n_f4#~7^=9Z<{-*E$P5$>c`QP8<e}9wz{Z0P&H~HV+<bN0T
z;LWg9Jj<{GtB@NZjblo~8hYmp&jxGh>(MjVvq86Wx+dC>eH6RL|HVC+;xkTq9vhvP
zH%7f``M;%Wv#_nmyzIH4@SL1mO|MzxJbLXm;huNe^Ugf*e(xyv<*B#6q)u>c87i;~
zd$13`xBd(};opt_+0Xtz%>E~v7qI`A_%@fTpUAlWO!MV!;edD!;Ruc)cWYb7B1!Lj
zjelT~_5nSE!upqcwSN!t54@~REv)k3wvZz`kgVAj^5h9r(Nm+eE5_OX*70d%&THFW
z*3KBLtxuj2b`BSC2~Wpo&I;j*<Ew~k5Zxqi<1X&wAqEiV91U8}AICpc@#~`S488wv
zjbkQ<Bl^$#zGO@?z4z<pM3G}rj0vdb`>5fsPgS|kZudo|9ml!Jv7U|l?@n^wWK6|$
z%*1Szplz7`-x2(7csjrOyz4Hj-^m8$KqH#Ac-9*{|JOW!<YsyPGlWgCW{-I0VLle(
z$@rng^w?(az%u#@{NeolkMDmS*92eXoc>?G8Tz|?zx3WuDHr|`|5VDJcr&bxVP7i^
z>&d>)m4@q{;Zk}nTQP;gxHht0y%_h3sZeIV^?Bv=PV4PBuZrAEZbgzF_wOmCm!Sf?
z@MQm<J@n3QZ3pb5A3*FA$nQ{pl3n}Ne=qWXBQ;C=08jVdUG5&(r-xi~1jmp>3K=vZ
zhqfE)e{?)h|D%iEjr?-;zxw@Y`zFEG9jARGZCSQ)-0QIQqVa!?%Eyn{A06UM))@bH
zR2=l$N7^PhK`(#bH}QeC$#HEIvhKWde&HQT$0@&`My^JCf;@+~4q;(li1NR%{@1g<
z3woS47U$k|<7xU|lYX{&oi;-pmypTI|26u5^Z{KE*CkxRRb0nS+{S-6{^+9b@2Kw&
z&1gYkjq$y{KUA{E?~3O>9%2AVwt1C4ph5qq^skmT<lkrLLs2ikWQ3(R$txHxY$Qfw
zEQ&D!lQ0=i{ZIOgr#haF*~ouIUoBZ-&4IYz_*{A%k3Wyh&S4+Sur>~TA$r~!Du1yR
z(52n-w0}dpXR+V=zMBY3$>={@My|jrtif8W$3|?%R>ZZ;OUW`+FOdIZ`G(pWY2uHF
zdlYBr^(eRge}(gQVGs7<01hGl-C?0^o%}yo9#sxJwf`LF9rut87xjOl$$8DIJ^yW<
zKbb@866^mPi`Dy=Z2p$$1N_9}@qa_uZ_2+T{^J-@$e;<g)dxAUpX~o@`SY8epY{DJ
zmB}5#tF~yvktdMcpxh--BhIBrA^kub;*M|YhBgG5^xI6F!8u&OC0xN(R4-v4zs3eV
z%0BL9bEesz``Mrm*vI$S4}00SXhJid+P97DTeJylKg>>bO|-+iUeouEC+qVy4;db=
z`=6V*jl0OrU?UIaTcCGtaxEUx2hg*G?*ZMbACI*k^d0rj0{aw_Ly_DtJPapCqUvS$
zJ<WZi{C#ajWJb7e)Qy)8R98u#G^NPecHx(WAD2GFwFb7|dhEw7^Q=aTvoQa@#e7<N
zG1@-x{t?$$={)<9edihH4|@Omz5kt_k9Yrc{WtGD>$vd#o!2~Nc$grbNtleO$h|E6
z(m0*osSN0FJd-{fJ&yD8a5vf2Xk4+d%Gb>wC+8tKczBpkE<}}drX1Hw!(zuvk-6sG
zJFdH@9ZCM%{CjpZe^%XLZEG@v`mA=lYgS+t)?h8xV<R@BO`GTcZ+(Y`583}!?0@!r
zGg|1`wEV4&--<$eT_09lV`{7T`hO#Dl4Z#A=k$GA9#o(AZnxKouszs^xBi>@-gw>4
z1pDH3>+1+Rgq|VZv#_WAm#eL@Bm5YWNFgUIyGa`9P4O5VsC--AA*&9W|3p7QFV`=6
zicIZPKHO41JcxNn`9QC~W?Zq~&LIEigG2e(28VO>3%G<UxQgqziQD-2_?f$odp~1-
zk2U`Kj6I0u*h66h7_>oK8HM$k|9Ws3N*|7q7>)inhKK%Y-_IuBqT^yrz$E0R`Nl>i
z!en~qINuL_Dt$V7dfA4Ll%eRteg8F6_-vG*!tq>k9_FKZxB6NAmO89{x~6`5q<%W9
zenLGO8u<^}`43)YOQYpA_Vp|5YsCF^pYH!NOZ|+({y#6OH^-=-5!e4;=zkVtDV8C*
z!TL8#>?=pF)eeYr6IapmKlM$jmowdu`|#_??TfU#U-SL=4x82YJ2^nI+x-6~;~`vA
z5uR5*dInSc|3v>kNBe-xBl_~I_evj9s70N$q>)jN*JG_X)?*_!<Jfq0=8Q!6Dc|<d
z5k=wY{7-G_r~AJs2cGzM8mc@W?SLk-us$)pFh9aGjq^X_+QHvHyd`WEcPYvc_wcEj
zrfjFjIS+9T<1YFh)W7E5gvC7?;~d5#Z+|lU{qe!!y$^<j!#H$wSolfzv-+;A;n+Sb
z95}im{OFYp;oVm@h5d~i!oH)Mtl_vR{NUP_u=m3)Vb}7F;qPAAs2|Pq-Tr*oqpbb@
z>fzy?_KjgTy=>Oj@HaT%_d}@IusOVSaC0ahvNh~nw>cc*fBA`e<%r*oA&Fk|RNs3|
z|Jo<VhJ8bx3qM>vG^B+0e|faIVb6pg;oWUR!~PlnDb(y99jbSZ4z-C-DXT|^^oEh4
zE;};Rw@(e3#23QRZStCHD~}e3V>S9^8^?zC@4XQI;o6^uB)$9>Z1k@^ADWz#LkIFW
zfm3KSFRbC>)X=PtvFXO-ki9o4v?M+k%D*!%oc4QRelHs?`mkd9&j>q*x7zeweq~&^
zM8AToxQ?5+jk~yyhZw-1jq)q{wwsSk_WsnEJF?%{!I9)>j72dfU=k){D&qcr)5$*m
z|C#Zak9KxEuG9ac9T0uv@&5Xs*LN&@F7j<J*aKMpzrp|FIL`5}`1SK)AwB2WXSHV*
z)8m}O4lJcFLr=SJ^WtM4WY<04-;3kI1@}MCZ!540$%ErU)!uPo4ZYTUz!cWf*Q0)m
z{%`%N=@G@DZicz$!kX8;5H^!rkz4&jC?(5~72ctLwG$PNcOgGfzr8;Fu4(%5QGHbZ
zJW{Ad9n#349(!D~4+n4vM{o>Dq|lZ&J|OFV^zC<&U5>k#7$306|18kIDUL=op&2bz
z#uwBWU!d>5b-(`q7p<GGZ5qck`dC6H#y#G5CcKGWySg}}P?-PsTXp9;>0c@Ro1`B(
z=QJNI4jp73Ipx3!@)WX-!e3P`;I!j2$j{L(P=<8P@(z?E)vpwXnmEP+@xSV}iTj}Y
zc-w#N(EhIr;hgI(;1aIjDz4)u+S1w|+7un@{0~{!f11oY?jajK(EdkLjc=t&`?Fi!
za8&yPt^181di{C*7te=|<IjgqbfFu0^q}Ew&ks%GpAXIEfwqi$UitdG^7Z-9y6*W<
zz0&(b3bm-a==~w{f%kX(g>c(F-o<@9!~h2UU-AluVmL-3+W3`=#)Z-Jsx9L}a{r((
zmR^hrcsf7-j{5hJGU1;3*S9~}Z@s@!zma`!Bno?R?ARo~Y%{h;SYiEv_lv?z`fQY7
zF6N>C9et0lYX2j(T>Y*btJ@-eWV*$l)%N)5;4t603$YmauA;D%T!s~Bo8tL<j~(7?
zCmDTu-RC|3%bx#-p8tp1|L3*;FYALOTO7x_IXBMpCwt5F|9|_1&}ZyG?>DUPFU~bs
zi}l!u&De@k+=%V|-`G##Yrb2@&HI&8<Sy*NKIGc{N27UI^e5{t7xri5%jntb-lX@*
z`pc7?Q@LnxI6@vkGWPpMiK<Qf_Ph1p>hDcC&O{v7tB>o*H2Z)1+r|c>?4bHz{n_N4
zC;bOGdI$1RE>Qn!6a3-)zv`FOzY8AM@&9i9fz{816Rtgl(>Q~3xPVKzg0|P#|LAyK
z{OH=Een)=w<Mu_vN&f#2`Txo0i`pk2XrCbG+*VXxw&t36uHz<dBYBPef6rQL^e6NG
z?$aOQ-_8HyW8V?&HRZ>E-v@nHoAJk<aa^N<Uhy;jJME95^x+tZ(HM(j6z2c!=Knv;
zzq?<%T=|?vMxL!dF8}}3IBVxk!emUvbj-wTv~BVIq2ojT|C8Eum$mK4JmOgI24zSi
znqHOv@znoBZ;fH2>>neZxtNFf$j$QoOtX#;z4I0I2Nu(pqUW$SGP==)oo|@Kq^?=!
zw-rck<IgA8plXwRm5{Hn*715|F3MM^LmKge-hnv2VWZ#k`~h3Zx2m=8yPgfD^fFXn
z7xrKu4&V@u;24rfq3@fX9U1H2CNla0b7TkdIDu0*jWeizUH(UEi~B?!J&h;%KVSE3
zIOn_zxP+(v$r}Di$8C!||JOYKmGVEj#>oG8YXASQ^8W_;AFb&7dF-##*Z)1f`!4%U
zi?jSQd=72w`me<K&(DRM!fxX(;@F@2<in_c*BY7RpiRnC#PP>N$tUBFhto%5G~zq#
zT<)Eb-P!|rGLApq?E0#0=1XrH7OvCBI=2`TP`S>0@crh4A2uJHtbft?eBqOka$LLH
zn1gH9>2o|CGcg-EVcC1bLkYd}DEs{)`;<NpJyq;?c6N6o`<<*F#r~eiPGtwzl67;~
z-(<#dJz201KWhKXcl|;vM!CM2rDU6XYDdQ`{snZ=yODpSpZ}iu2g}*t;%O3J+-tB!
z{MqsP-|4Mm^uJd<7nX^q;`7$k`<nGL=}BQ#+2_J4dTp;c^aq~{Yv^lHf6Y9E2j<S-
zF^8U9FRb~7y$Z?A$nCX9J6VeCHTNZ*ozhw6xB^e+1$G^DpQv^pHSVpj|H3Tkckk&X
z((k_G_=sJu*@JyJfI~QfV@RUyp6^fo?m#EH5chV8YdrSg>HZ5tq#rH4>VM?Ws(z|W
z8=E1X44RNb^1Su;*%2M|+DCrB=R2gIK>b8{1L@1Y4QyX7|0)B^R<pkrvAy48TN@9P
z8>6qx`Q_&CoFNPQZ}I7M3GYV!qP~7q@3j8cetmqX<r}Ee-xtT_XU?+qP_Tc$<lLYB
z=WtFu7jOwza20JEj6XmJI?;7r`KJublRe7BhAjVsIGc9xKcHoi`$O!bY(>9v=DK)p
z;x@|lf8HhU<00~27#s%3L7U|<^nF$Pp6va+xdCLQJT;OWjj<?3Qkfddz6ta=_c_ji
zo<yIF`tzP2($1+vpYN_W@Bw_1{g3UBKB`Z&`=>g8I%cB;bJ1^ahI!=tsGENf#~duA
z7v}#zJ2)(+=Y&Nc(^7ipEq%jSMqhy*&n2(C?dAvQA`A5o`?9d_YPWiwom$vmhpZp-
zi7?YOtFQ)Zu^t<-8C%iD&Td!Vb_`Sh4p#q;Q~#2A^vvLYSI0J@iA~<@+!pp?*7&y^
zT2VPg|ABbQP=Q@Y3XA*N?V-m#hf>%_KY;rE?Ef0}KR-Yn`SJ4Mkl&Bs7|Ory8~ZE$
zar6`_ex)A%n(vC<gd94MM_*A<I6?Nd3=XHr(>Q~3xPVKzf~&ZWe*H{0$=hhYWo;hv
zJ|1EKao>Y%uXT9-Z_j_XJ^(yJABvur<xTZP9E06OR!`Lak=7LXwEtrf8~;_$U)yNA
zZ}>_3Vz}!^qWlZuFa~fez3q|wFV3g)OX!95r_?Ql^@oP=KcGooZ$7F1*DlSHagV50
zvf`&-331)^VsTX((>H;vN*nKIjNT;rWaK|@jy@klE#fzu9_Je->xZdZUp79#Inxo_
zoHNPhqsGsWCCI-b?=FxpH_7*82j5%gI{tpY&BJ^w#3^}fF}W1WumY=4t)8k`^0D^0
zx+^VDXBw3c{D=+8ipI<GH`#nk{=O%FKa#)6R>y71n8NrZ=XNSz3i}U}dBpvP<Nn!e
z{Lfmf$3_(L?=@o%=uh(RR(dJw$4L7K_dwQ-*M523`$1uRrr&pA5B4F>BQF0o-{rfB
zaKLe&@%@L$BdGX>=l+*-!ZCUhDU_Gn^PFr#4*mKD^56bS=vB{mIPUvv$K(l|!f7<0
zHD;JRhYPrb++Ock-F}7M+3nlFRr+=GY!U~a>>pX(Jtx$dPn;T465?F*x@jdLy{IH)
zP`|*ud)It?e9ukCw{aKu(PmzI`{k0*L3Y0Sl5hGY{(bZ7={+yMr2ozQWi*Mq87+6r
zUq;US_SRe4j(hECE}jbI>03WD=8~Q~TVnn1k}&9dvHo#yc!oX{^)uWX;u-~Y7%r^&
z@JnGNIU2dDm%>=G7}@koq2u+JLZ@{GCODpiynE}h7C_fJ_qqC|Fj?4COvg<0ebYJ%
zWN(-gO341-&IxnLd6<udSd67uhDy)vssCVudE&y7<K6#@=84m5r8Om;<)8E1QGeY1
zdtT{Y_kW`#R7-0OQmDNqOnCe)Bi?!x`v1#ASSzme*oe*8ic+*~^ZXN@KRVHcZtDf)
zr<EAzSfc#({5N_2&TA%HUi184@%+8pR<e4h{Iy^HLT#0AS{pKb-Zy?${zjSqsK74t
zeC~Q;54jHqa0qQn<geG{>vi(iYWZuMd`{*a_mB-)_OCdbcCdfZvO)et4y}mu|Bi^~
z7?Ozl0OgjjKV}%;=UeF1Chyqb{`{6foKqb4)_AghO_Q)V|1U>&ASrL=$rGrG*NLZg
z4j;bb)5zRWZlKP2X;jxpKT_?^LmfSh%q;H;+m(|~%HK1tKZm@1e=d<%&~`)liw<qB
z&YjBi!}8V+?;1T>?XOqaUrX2`%ay<6AFlt{|7m-?iKj6C;hzVG>-6$(s9(tbpO=K&
zWG{J_?E9iM3X~!DW0-csLo&B*P#7QwZSlQt(oah^uVx33!%^@<taM+O-Pd{ddC`3;
z_ec6|G%AhnzTo<?^kPgv(zxy_b-^Th+>1Rm%s6BERMcnHP3n`lS9={fUD!;_MhQON
zzcJTw{@eO=$wL3P{+#*rg;<QGsBSm@4=L@3T5X5AA?kMRhRihM|6cq=DAzt)=DZbH
zg*8}<_1K8E5A;8w<0AX_tTxR9Z5!kdYX6~Ox%S6+?GH51(f$x;7P*A>2by=<<3K!H
zQHnC;q%qFXub_8cQ}!O@f2Z$3&n@E@(0#-Bg?q*|2s?m7ID%tHB8C1R6om}g`_JlM
zvhUaYB`>NkqCKO(E6-mQ_v!8MTcP}8f8^=O8QMRd?+JSJiKO%qoT8sbT)Qx?VNme@
zZ!*?EShMo|9C-n8UCB%26=dm!b%A_)R~=tR-m~dJVIC@3Esxa1e1clkWqtqhYhit?
zH2>GHpA9!%cN=$c9}h8rL0dh4eQ)jP(C^lXr|o-s?2BtSFaKYb|H<ap)z4c#vJdF3
zD~+2M$50GM<vnd_vTB<CH;kr_MKR(SgxXQ=Pn#p%_)+~&J{fman2+wfihpG{e^VLp
zsb|Av$7_7sQ_1O=iP<Q@T+G9NRQ}C(&O$84QY^y?til?!ZIb`DdCp{~vZ6~Ki*pe3
zAISg5<^Sv-lz&Uq&ECh8@mF!5g>q{X#I^j_im&{~&xZBnMr_7bl%fn3h`!D3=>M!a
z%j7O$d$14rpFJCPw%9N1Q}#R4_U#L!LoYwnA;0x~!~ACZ3m%~#Lq+&xc#Dsr`S7qX
z$#D`XWRPS3WwXOV6TR~RJMfM@-sm0Z8N;^LuiWjNF64z(UbKcEc?!wB!@_Cu460n0
z5>IWnJ>DE&K!#0Q&$g{|UK-W(nw9Kd)S?bg_qU^8a?KT7#dX}oZQR9uv@Njyz#`-P
z(TT1N#up%u9yA>PgZ1|uXOTlInr}Q89*QTwJrM@T1IYC{NBE%sNBRe|;W3o{40`B!
zbn{zup|Jn%%ihPg>|>xWvf^Kjum4AVKh7PA(J23pK3cLEy}#7oPxk$V=T26>{#=+$
zPQ`T0L~<qngYen(+MS*aO6YS@PmgOe#4#9km?x}xnm&JWA#%RC#pF_CwGHE(*tiGG
zGRG^Br}w0Nr;WZ-eE;$!>%k8TtNgYGasK>Tay{bs_`>{o^~pxZaSVAJ`xD0i)hR2|
z`1_;Jh9g^shQqH94ey=(WcW$r%y4Ma%y97Rm%@*Gr-vV{n;s6Vp02O&rSR^`lJLU~
zC1Kx;{~Gpgm>qtwWp>!UUwL>~d3aR+zyIj}bM58tJ{L;qeb3wX&a*6|_x{rN`&Y_K
z`RSPFef0A1@Y8m_oEkm?znA~4DCCaP|J8W>e<@O~Tf;+m_XcH>vZjZflBe$xmQ!{`
zUqOemsO+-&hl&4cZUme4XUhB$FNL?pyk!32e+@gA&knn^oBnR%e-8UxcL00ZXN2#M
z`Y+*~8UH!#-XZ+xe+fUlI3(nMJ|-NB=QIwpw&ru;-7Wtq>|gnx!lA*>g`a4TAMx8U
zB*8rq-qTJ>(R=@bKZ%TU`<uudI*`W+G`{>N`mp{aG}ru(L{s8diEQH6iIx%nBhmlP
z@NmlSH7E6<C7$PF)qlD-_PzXZs5|?|e5`*GGK2p(9KB<Vg7Ydr_(V8%oL?gGnehJV
zPlSIss{i!CpBjH=eExF&{Uz+Xb?mz>{QAhy>sK1<=h`zkhYKj!+l}5AE;(*{dvIt+
z$6@=w=_lxFH#ZD<Hf#?Xt_=>2x6HpIn;+T#jh<y&=FZ#yt=s-@^ef`IigI%duan6K
z+8=iYhnw_T_m>)PU8T7GFJE~5822VDjk;-r!)@m@uVjOh_mLZKO<ZySS#@X!-$Cc@
zPlQ2l$iK+bdp7V>@FR57^X|0=Pwu;e-ibD}<H<gCE!*7pCh6EB9cUOOUBWWUJ*W8G
zUU5(KRMpT>OICLe4K?YZVW|Hcj*%FRu_(p_OhRFQ1?`^H6z!i^wSN|9|9C%{M)p4@
zJ7+4UV<u*!1amPDZA;A0Fh-(-?A)MDqV3RqNBx5yG^E-8yR|>GbD9qu|47d|FQ@I$
zsvS{TRrGt~ALolRd95f^oh%9q>9wy))9d_7^kt~eO0%#u>hv2`eAD=fPui!>Z_PXT
z{>iw9c-%XD4Sg-v<NvFDP`5-mvRr#(rE&$;>h~I9h4FZcv_I;sqwD(3*owIReJNRn
z3UtZGF^|W*9ryhDLw{h)UhNG$8E@1;Z$uNCu}d6#unz}t2yxxOBV?aCA^)xC!!ddi
z{r{vbMrP2294g;dhD>?fri*<89m0~k&0j#CUOQM?aDsjs_4Ev$`2Q+C>-qhBTsY&m
zbGU#@xPpH3ey@_f)*HM|_I>vg;U;+-cX1!h%9e-Z00wRIjU(5njnS^pi5|zK#6IZZ
z^pSYF{?9Aw08}orU%YEa`)w?Wk=&sFTR;B<dhPz=kb1i~OrlRl{S1A(Nbf8Tb=bb{
zqxP({_y2ZGnChJAn2Fh__?dj`9hT7NVjg;}=`f${`-T4FwEpi8{l`W9-)F^nr<iZO
zIArD&hx!@Tmj87K3!S?dOR)?qunKF?c2@uYW&QuR^#5Pe|KF?sA9?hk!F>J3Mf&p5
zjFvn4|B*v0s`cmB?DqfCQXAJaAk&UB?fySo9M<}e_1K8b*osn=p#p8LX;16tuPWBR
z>6vQVb(4klU9<@s&ia2ep&2dW%yt_$K(@NBd04SJ#y5z4IDnk6EDq5-M~NFp=*Q6G
zz2r9(^RX6(E|huxm9KgaOS}g@_#{35p8iDVR?$<7qziAop>6Ut^NMCjmuFiS*Dye}
zyjP<>Pp#Iz-lkrEJ+`lXqo_wYUuQ;KO~|1Gd7Qv0{9%6UIxj!+$K}bM54Hc#YX71M
z&G<wAd-E;zlX%V`zr84&BXiT(|Gk641$yTw_CGGsuORyD^5~X_x^PukC42okc@xP^
z)&e5$qN?iO%8~nyAL2>BLfy;KiRz2eKi=~~E$Wa)2K5+l&7e~G3bFk&lpKzc7>l;u
z+W+!k2Vy&`tK0k6F87Q=57}^y{eO%7|A75}hy8z1-b7CRY|Zlhf5ti!;_3VR3t<Ae
zqbj)Z5GK*f$U^^zy?!P;o{H(1iT-aEhuLHa=3?i+{w6VxoR7Z0)+a#5a%?f#&u6fd
zT!y9cdiih7SER2(T>oY?xdv;o9<dJ13g1ZYtWmcdWh2wKqDLK)M|YLFM%~ie@`+F?
zyzhJJ5puh9?_k@OjdJgf3;F*W%31mz?85={KU)+Ik(J^<LLNgBsVMzlmGBHbu1T13
zKTY%;>c<I3dLsYB6!#`9jyuegC*nEVJ`qllr;*j)(82zX<NnS#K8Lt>P>;H{Fb<H9
zp?VJg`wsu1eyrUu&USI*$@squuDOIOxQgp2tbeWjv%THgGoC?vxBJMdfB#qc|FZhm
zvubi)vvjo}&I`!#DYW{(%3;=Y6X$K*MO=64$@ssRk84*xw*M<WsT}1mh-(tYJq*(7
z@3<E7L)SF>jt0m<rXfYU@)>d{;y9m<4Zh1Q%74crk)NhrFvoYg$aji=*Z-|7Nbi;Z
zQIFBC8H-{}z$8q@R7}V3uYawry-D8GwvBs)_3#1yN9$kD6vu3opz@KqFl5yY&tROr
zR_ODw5UE+pC*{JEb%Qfo+@HL@*l$a*4CTfS6#P#^e1Eg}Vuh{3TC7L@m(PZc<YrV@
ztM9F6^*<h$|36etoD{#ZpzZ_dcOQlEMV`}E=a!-@UV~lOgMG+<@!8OJQvPRSJstmj
z$8$wt{P#iSzi*;(f&5Q4s~cL#tmC*wU+W|J-~9dq;_6e^A0m(77?Mb#yjtHo*@PVO
z4QyR9whi*+2_%(Kr^wUj`^yl{ka7IOIkKPc_!9ZnXXE~W&xP%C<j?V*&qU)Fgg1XU
zEUY0f;3}>o_lW&}d|0?i@6>19kr*Cs)9<3kao%w^*~JfVUsxZX#Y3|9x7M;J2W?kg
zVkm}VBt~N_iqZegXTt<?5-JxA3zNyIn2wo9Zqxt#vV9)th5h$-uz%@uQU5mk*KwMR
zYxTwYuX%=k(~s)EoId~Zf97*^Zq^upj<fs%D3qiB9Lr^Xfz|v3Z2#^p*8kI=Uvo<z
zzrOrh)Ul=0$e<qc#j_BLu@uX&0;{kFZS2tYUe5@f=(^_F-0+O_?f0<H8x|RVfF{Io
z0xgf&tj@{ZGX8+9<fB+Cp7q#>%}6$y|IfDFO0N}13Z?Wi)Q@*>v)tnx>E{Qi5Ej>c
z*hTI^?twKb$pgq9_g&V=|H_9$j*p<Q?!$ihKkYlMlK-z+pMj4jwo&6ey@G#jj_-el
z{I#5|?V2P~$l&At-6qFvQ>=gQ+zxrO6J6(xFS*PIPxi=H4KJ$y{oW+bX0pZk*-`9+
zan`>lJH%1`p7Y7d9oorc)!XuaBcDCJ;<H8kDQx9l_xH&CU6Vg=xj!<_BaHogdHt6=
z-+7dX{eGui(@dTr&mpfYD)c|<i-_e?oQoXiBwz5`C0xN(lowgYe4=&)y^2h_&ztny
zxQn=V>C^s^D*uoA{nkM9+d~Xs&<@`{`t)NBC41|XyJWxmZzMSy1^fA${C7+LZ9M0j
zd>mgj4(QALcHbKwiiJ%;{#|uG`DFb6RQhzxME`G%{Wt!9HoaLHUP9hi{?8@nAva7J
zGsJoY^v+&3$r$Yc`eO8)l?QI{kt6!bbMAYR@TFLW6<CEeSc_`)RgL;8rB1A6kJk07
zdv2+NuCY%~vhUS>jj!;pvs;_hwJl_<cVj)=I?niib#J@+wu9`Pqu%9v?;fST9-$sa
z<p}%C`k#&1jIBs6Q2)PbTmZfHDBJio>-^FyP`~%1{{I@bvVMb)_y6zm`yT8=+&?0Z
z1N1{E^gp~~P8Izal1QQY@JH#Zl0Ipvlh(rghV#<*p>#T@2|08ij}thB(`Zwsx5s>j
zPQ*P`yXD8H_IG?fyFaqOS1bRwc{gMnqtQw>PvOHC&pBMcCFB-qBTTgZ0KL<5=uppJ
zrC&$SQE|8XkH%PsYdZ+5)F!x1-bHeh^#{m@D6Bs)N4ka?C+2uinLMn{slTS3>%8<m
z_kG@bx$He1_nwfSRup>Bb;o-`Q`&pY8plC4R(Y?7z1Mr*Gd;WCyLp}80#DX4%Fydk
zhcvdgTdROAT*e<4`*JG&^0$fTFUWsM9&ta_tJS}o)W6&0FS2f({GH|B@77O84tLH-
zjK)|LV*(~&GTIh>)c>e2p%Yyr<nwXjA0^)p(f(kAe0;y3!hS#-{;mIUsyL=2mKih2
z?5#fvv&j<V=4g+Q^Uyg)S(Q+>4N=CCJ>AMIa=zadqWtHN68WF~HnEt#6wA=B9$G=h
z{{K~EpY<!&kTLJ?z;?v?zn5RWPzJoK55Vv1u@PHQvEB3f)uTiyy(}JM7y7jg`}HeT
z&fwE@yblL(2+4c=d=t%kqSv1FY_(~Q#c;$nQ^s-n_~W+DW??C0(1iZ?A0={R2lD9q
z-lN0`viGaMO`IZ6;|$K>0-CQ4BtGu{xFjqmEPHPtafRNw@^^`jivx+P^zwi4PG0|A
zA}_3a)9(^pTYi_Qel(D%nK+P0O&Lhkl67+i5@|ByxSq^^fltu2h4J@%0XONlaToW|
zw&Qn+c61#4U7{0RZ;LzYKN^3R=t=)B(Xf0V(Rl55iKbh>OEmZXF402I(sLJomuN+0
z)$bAy#WR3Gf1~_HQdpHZhSF=tyEpL;r;kMai|$QWdWL(QHINuBtog=3Vk}vV-2Q>Y
z1acCx!aEuV5}oY>iOG(qqOdQFGN5bkK%yJf<D_3&Q_@;1?R8{&k^ewFrn_b)W}^gi
zF%R>x5N#KgKj^rn{OJ`hx{*f@8m39V`)o=KB%0A8&u5hZIkNRe9RK$yu~<C$ujywc
zS0H)zQKCxtDthgsM~T$EM~OA`wW!w)&#33q=N~2Nu>FDjjj{zH?ELO$!+PhG+f%gs
zAB)09dc{}O-(NSDhQ1Z0D2w69^VL_2s|In5T%mox-}t}7@;_Ox+}P#3J=li>C^uI2
z5P1a0(8mAYjyMOf6J31z-N@5>&>&woDg&BM^7p0nKeqD+B1dnnvX6*3QplhQIdq`^
zSK=Uh^`D&}`@XGhL2lP>-k}dSj)^(rxcPzoj>(E|>G%07ed+Yva%~)3qIbTk++eF*
zp<hLhG9{00&#`Nf_J*+DulUZb_1X8Y)(mlc8+UOZ4>5p2<-Sed=riO{3`hUJ=*K5V
zqf(hRmaIB&zcq3KlJrUBWYo&<sXN|VqkMl++a>EgqW7QWYw`E}Wl`w=+amj|7|$Q=
zCw9VA*Tr!w)5)2bjS}>J&AI<#&C~A|sq^haBCIfv`<L3?^f<RZ%RZe??^M5c3^B)#
zz8F2~)jYZr{0G9H&M(}ro>B+bsV~zf)w`&_XWx6*EW-+{!Wyi_dTd17CieF>|3P+s
zsC{u>y`?_OlRb_b)V+=3ZaU2VCJXD|sW)RC+ltC@pU|K4>%?YpZAB@PuNeP7!<+_s
z?ILyDtDguJ^j#>7Kiscwds!VP{KzX05`SOyAo1SbM~TA=9wZJed6f9c@<)ju55AK)
zs4O^8bvN;&m3I>FuDh4mKjU6vU+;~?4|m>A{9wti6ML86NHjK%3k_A{Li5{WLetCU
z)+9a`TJDVrd;Hfv96(u(Jf?3ejx{@`-=T8wpb+<@JKAjy;z@0L=Tx%ej;YtHthxC9
z<ze9;&JGDl{ZciFPn+ZX$xwUp(;@ZYC~cb&p-x{>ed76$*)+m9g#N_eyx5;OB93FI
z=)ISCYsCFT`K+6Xox5))c5S(z_`BXaiMak;%I|v~JW70Dnext@yNTT^?<V%WI*|C`
z;Q@8}v!UYK)}3m6nE26y!QtK49wzqV(8_1RPmX(^u5CgN9e6Lck;v+FTmw)a!-x8x
zF6&bu(~tBq+|#cz&boolIfc_WgLAlmOSpozP5NoJ>3<rc-v(W-=}zc>LeCuiG#m8O
zpb5=rS)?BaIkYYi_DWH>DxT}OiQC9+D+*cRcj=wkBEFrXaG(AVJ^PFJAB)&d+DI4>
z7Wb|jv{T+hauy#UIUH5OQ+J9&Ek-(ydl_cV7KM7$^%jLRdd>MOe^&qh=Y6yM7h|2*
z&z>oa|Np|UQ0%x*`(Xmv`&(mr<<&{?od2e8h@6V)h<gvtB(vmfvIM!TahBvfbn5@?
zcvaa*cD<%-B=e3dU$IuFG>mr6d@MwAn>9RzFQ(VN?Y-}KHY}wtL%rjS<1|^fm%nAy
zb79B0=R(<ded%n|70!7wzJ3*b4c4OLIQtk~=tdqrs7~wuXA7s;!nJJUIx@{pjy7&R
z*1L8iHe)NE&cEokCXVB_5B2|_XXm4nE#JlF@7}Nfe=j>94X?7l(exVo8!hxKa?WeL
z_gtu4=ibE=*T&mL?m<%DdzJ8g^xEBg11PjVzOC&{&mfID9B@vv`#(e;K^*^oj7%ci
z%P)|X?(W!!uZ}@p-5=-WJstl);-mQw<E3AE>!wKm66s$q{d@-H)z+<cT@D@bdYr&1
zoJN~^znwkbfx`NZS@r)Wc`N=8*)T);`3jnbNIzQC{c&%I99qw-|7(VYGvYai3%G>B
z`2VV5;R?Mr`UZ>-xJti{`tj-wb=TAJ|HIUWe7&(gyy^GbxQn=Ve8n5`%CC#UL&pOc
zR3VRGD28Js3iC4DV__Z2iQbF!w;r}nBAIpE(e1r@&TUoRGd(}WdmZIHd&k`qp9`bK
zF&4#`fJvB)shEybmFMamRG;*mFMF<sJy*}SPIw$cQRr_RrEWy+0(IkS(uNGZ{uTWn
z)+L@Pp4litd0SDKOU}c5v_0bgc-iwuC%W+a<FENY*bGe{@_(G<|G2DeK;|5`lD+@P
z{{NJ*|JLy9ebzev;#`cSScVl?h5YA=!dg-jIQ0KHej#r-E?dq2r9LnCzt*wm4;!l?
zd^589wLRYEmqA6fcGxeykL7F^>56-|#W9qX%FFxS%^K%#MJbZ6>3>%~meFgKkttNr
zccFfVviYDgdatsY+#@XZA?za$AV)t$9zpCw=uoG}b$O3DP9pEP$8lFy8-T37#{Ry=
zzd+VLQvZ`r{R_*5yC#Ds<j{dUPT&-tuK)9h9eYpxe@p$Z{?0E@|C0?9)&En}|77zV
z^*@<)oFfbC|E&A{^?y!_GbyYJXXv#z)c?Jo2<PZ|KGk~18OLd|j(vPVc%1iGm|tLA
ze$M;9B<u>Zd@~&jd>0dqzjJ&Yd3w(j-^n!J2^!M;!zcNNkMjkfg`WLTJa^<zv?b(k
zbf6P|xPRMCao)yV+((6X`H&pIptsbo3p`)`la#dA&hUKY|MWHeTd0ru|NP+ajPr(K
zI7T8lnEfa}jHcJFlsA08W9h}He_h@X_GJB^3Bo2}GNxiWdaYM6lPrvP)E89f4<h@2
z^Guj6yaaQRH;1eI=WJ+tb1(g8+CS|1g^m{^r=D6$F2j@YI^9`*EA9FoGVXD(!f(CN
zz9#$Z^H=d5{psq@HICO}JvL%9wxSees6hYs^#92NHS$7Anqpe(&Pp3Hh~uAkId2d4
z;Q$Wd2#z6%w%5HMbPVx+@O1tky|DgI_M`ECWLzt-WsCQV9KCfV|Nqa~{_ku1xA6nK
z@yug?){MvP@09p6XhQy%*6@!K{onK~e`S0qJ@$Q`B2VKC&LNJ0j(eY6pg)<vQ1KU@
zi~ja{??3v0n0KLWy!7wX|Gif~F>2HLg|qsF&+Gp_tDpF!erj=C!WCS_b=<^l+(p|;
z{qN{NC%Vv$JbKX3t^eD9H0{>cO|~4?w@v08x020cio<>JJj4J7{jK-D#5r?{!!z{G
zm-T;dDGo#F!_nh7|C;{qMf$(Vk-{pM7l+a0SR_a2PbDWHj$ufNr*^P$1n+C_@IPh7
zNtfd~GCfNFw=~yCa|*Q|2)`&C8G8Nk;@?~UXOcK8tZnhucde~LpNju*{hwF<fBJtO
z$n)-Rx_D+{HcF6_uB`OUrFXvV+uC6t0Q!9NIL<qcbNssY^6Ls)jLJ>+aUiS46o+Nx
z3dCnI&of;`uRX4h`vdQtjC%~kxy5mg@fyFa#d>VSW|V)%+WurI%20t_*n@pIgd@m*
z=dn#vnE$fHT$fN}{GEE{yVjWZd;ha9ge2Kp=lPO-|LVDtO~|1Gd7Qv0oJRlG%@-uk
z;R5>JXGfE9{hur3Rb0nS+(z6}=q`C54>5q8u<R1+0>16}jep$k&Hq%p+;Pu~z5`+1
zGqi(e>HnK0|0A_q{@*G8t3T6Z2GK??tp9Y+7(@(reT6aIZxsy=Bk7~j=KZ!S<2%UC
z<=)AwAJzZzcn=Em|EtvhY5D)C`k!8ypX+~Gw;2~Ij$%x}BuvIsOh^CUn*Xi+)cf_~
zFw=40*X<=imS8UCq4E{wC%F)du@p&RRm!zx^jdj5^{Rc<=&MlwqO=K1&yZH>F6^(X
zZb-3(YuBkC_?ip*>(c92o44kgHCT)F_;~-%M#pW}*xxtQ^AFVX=%UB5@o|hm4|}^|
zoch-}O}Es)WXmJ<uReqvTG4#bx&NJiNL*V{ik$P}{&!{c&hg@QPGS7t_muk+oj=Qe
zO!Hr4g>x$X*Di7o;@E+G<N*}+zw=uhLx0He5oESW*GlOmV;_3;E%OhwQ&QrqMcrQE
z$e<p_T$4m9UW+E=(1ASKc1r(l_ejRQ3cKiWZ=k&PP2v6Dk$yCx`K0t8mp0_kisp9Z
zxOh(CG|nJ5!Z~A}3FqjY>g*0&pkG2<(?5@H=X6cfPx7*U0^wJYeD#?^TYS5`SlR3O
zjk2GPu$b1k=V1orUosa>9*OTdh5Rq9fA8A3j>K&;d(r$?@;-9vzl!a~(|?(7?WDFO
zo~*0YGfF<fL+1^k{~O}|syw$#8y`b493wFrV^NIg@18*Rea?S>hYji5Oe3QXuJ_GW
zkCPt<`({xqti1Z!Fv<CoF%|i)VPQHs6SL8^Kwg&D^N90}+HUX%qGP#x7S{5CvH-b@
z{C#K`qdrCxy%{Cq==&FIe3Nngz<Ff!eeFQ(E1&OpAr@m9R-oTATScxx<>fyLYsrdV
z#d!zzAEze|8vnP~xJ7y`dpw2B^sR_}sxhr`%wru&g_WTKyRZlQZ~%q<zpR1K`;PC|
zaoqps2zd-iq|mGlmm!;wLr0YAh3rA~1pUeUzZ3M+c)EV*VPzGn2eXUVg{g7uq*3f5
zHfI_cHgY|dx~A+yHa=T<JDd5G<1?r*xACop_`LOhqvH`f_a6J6E!@3?owkUbwt)T3
zrftMi`*RGt4!L3Myo>t3*<k0ycLA4h1<AAezx{TV9(@8ST&Lee{bl!tG~!x)w}myk
zue;=Z<aQX(PYxh^kY9l9*tyzR{=f5G;%Wc?HumEt{r~LDni>9Mn)IV?iMU_!ABb!1
z4|UCOjKpY+MKLB|5(??R_Xp|U%fEphG)QOTHR<<U;#vv+uK&ON_xC@TEY9Rf>6f;t
z^jdy_xc>iidY*mpwEsWe>nwf1!sD2qtommrJ<c&JAuGP;{g?yNGs|}{O`SB+`$i`{
zu1R`bo}KHwd6<v#BEF!vt@%NJvi{v-de7%)g{5R_r~G)!H~YZ1izn;f^?k!!lb^9&
z)m6)!w*srM25Ye%8?hN%(f|24=iPV)dhefm_E~xUb#p`#^865SlW~ks{WSkUkLv=J
zIky74um}5a0Ef_i%{Pcnbf6nu=s_Nh>beHyOY=l|9@!Vw!NTI$gB%Luf0T2nd(t;7
z_KRvG4VI@6=l>n?AIFeF22IGJ19=qI|C!^N-tug(dB%@CYclV+hisUl{WpVs=bUD;
z#qr12|Jg1)j`@rGpZ42pprXP2g@5Ar6n8)0Kw<wszJXJY<9vwIWVCnA#AE(~xIfc5
z`UPCV735C(kIQUMdS|11tMR_+H&Iys$8q;j_ev&(-A3va^`H8oj!Z95SCIAUjk|uk
zkB6u{Yb^vau1%LzXAXKtp2ElH2M={T93v6?-$#>UQSlen5ms&%(<fjOCZoDW{fm^o
z{9430{b^(p`jPqIr#fdkW@0u<Fc<SMA8k`&zqK(A#xr!HF#pf^iadG_8-K}v-?&^q
zw)2`N>bIVw-<r&!b(An^Sty>xSc+wcK8(2k-wJx?LG|x`>krb`phvx$f1v)oqy8n=
z3ai{?3<J3lNp<RGaw`h^|DE+*FB{9?xD1(Y&$Y^PCezvI|FHi5J#BN}O)XjHyGi>_
zGQNv?a=S4aJ3jP%oYcPlrtj)|{FLH(tIoHgZ>r6Bj`oYvZ=6RQ>(RB>*beE=@0b3!
zrGJ6%Z;9`3x$lpBJpYuR=#KUO(Y$F;*dxAuIDkXQW#yCA=H=2m_dfO^9iu1FGef>o
zPIT|!Ln2ec;vSS4vfQ|r<Q@5po<o&<mg48CJueQ&d1M^d>#M3G(<lA5+rEFOL0pTp
zzS|gOq<7i}F#g8hRf`j@J%!UagLAlm&TZ0;C;I^QT=t9*=iqgrxt;&CF@(mWo<&-k
zmrFaco=*!diQ@{c;yQNBQ11`p%cPgl^L2B=ZTem8{MMZCkQ_k8m%kF;`k8V?dw<aG
zu>HTg$7jf)$O+40IKA^l`ER0rI{IkzjB|hJ9`FAA*7v2E{C;fz2Ib;c?DOoLVobm!
zlz+io`k$JoPM?bYH~veQPR@+_XV?Q|2`au-B7eOU=F;b3KKkD^cAkvo#bPp!4O~i=
z-FV5GOD~1(cU}sW=8Atm@wsqcnwI%(1q%Dm3tvUAHGe#{x+JWjuSNax6783gkT#dA
z&TmInemVU8rZ0#0j(;f}&dv&ls%C_rY%=$2%NN7J4b#Gb@l(Q&c1{cLuKPmRzxoU2
z%YHumuxd*9!J^NFz4XRmpAHR!KOLIQOKMv3X}*!qgqG}Q!g~L;5u33Ur6|L(vtKbz
zD-nLWX_o%4uY{u?$Opn>AIvd%zG_^N^&yMGKRozkNa_o#Njw{>*XjTIP(KFJXFv70
z|LgpSkh%CN>o!abe=}ub*d>lVs5t(G@YaYap?utnVdtn9!>$*ngujdT?Y9H*oN-?a
z-<Q^RE`BlW9ycxQTR9{A@WUD5kl&8Pb5_m@KRWp3@b1wshyB|;kCk)6PX=4(!EZ^V
zkimP(swOgr4&-qH72d-s@-)t%dT$B;jCWuSf!bcr9O)b0!Pyf2Df9amnct5lG^1rp
ziDy_6a_h|R_uksZTOR-&v#bw*t{1HjFv|J>BdiZFtR!sLR^5TJ!PXh{e|dfNy$y51
zC3@dity}PAeb@A>=>N@2;W~K}6+dPNo6p@WZRMXV3b%#b#eL+4%rX8<UZHp1v$s74
z=!5q7{x93x9^KdMb&qF+4MpXL!hc>As*cYI!yS)Al0KRoi`qGJ_;Tii^prWF4)vp~
zm4V#r`n&x;0a^MaviVi(0+Le^_Y&$s*KutFbUe^jz;wUOMD-Nw4;gPzSbr$3T*?}M
z(9KUo)?eh`P!>Gt-<a+E63oS0){3ckuPC%B``XblgP&cQ_Iu+WKhXAYPmO4LSy_RW
zS;|)AMreazzPS3nR1y}Fy>%sFG1>nHzbCm2D-ib|Uq$x5As>?czbXl9$@SQX&8TGS
zY$dBU*$;&*Ly}%W?n3P~Wiis1wXg7`|G%1@S);CtW9)0^sFx<HmzJylm#C+-1LnDQ
z5BA{z4&exnA&G+hze62~g8lopx)J%K>|Zp9vvC;vcQ9M`7W?;}y0!Ns`~R}A8g}ZG
z$Ip2CIOV-KbHd{o`3$}D197Woo9N}n7ZmJ&b$0hG^{=`$?j2q!-V5=&b58Q0^T|A_
zZb(<RbnWIR_zUgfZ>w9>)dl;1zjSU>|GMuQzo%SR82?X?^>ShUU7Bq!o>MrDGdPC}
zxP&WcQ<t}|RK}wdUFgP>b-o%dN<W&UGp4o0{bW6t99j<x8)0mpc&_6nZX-9tIiA;D
zdgnOl!F~Ee^sJWVEz-YD`Y|A^a*4Ls_qD|k*G?Wv4oCD2#IcjL80mO4GIOMVy!Sjy
zI#I1|uJJvjUbOxe>Sp==k)hXPtZRxf0h2Km(~)n}-%PgMVsG!1uXe~+h<oC7vv>22
z-W3{N(*{SA^P08CTOP5;$sAhG%U|l{9f<RP^L!1n#Tn;7l#p{#7+3Ne<4WlBu@H-~
z6y@yC*l)0mz5=VT2Du0R;|{x=-pM}iz<T;d^sIDmTil~INE|D)S=d&TqH>D$PRJ^4
z>ZJT!LEnWvNXe_U_`~%P`hISW@=sa6&ssfkp5FEi(t}u@?sHwm&-ewtJ}exfA3?u+
z?hoeT(vwIbgC^wAfjmy2|NXfB<1lNVTmQjv->0l2OP;|wTtM?Z{Y~T*T*Y<dwE1HD
z<0idxk#CcJn|>Eh?ZKCQtHPh!zjHpae|NBdce8(YvVXI~!vWXa$3qO@<Nm)vKk)t!
zv;WiV-zxS$d%uhA-@Tmu|0?_cW%mDv?BDb3U-oQa|2uJJyV?I_tLrMsn1`Pc*H8>c
zav1w}$nY?dUOR&QORtFYUyQ-3-{RgjxW@$hSNLe>j72e;v(JSIWY#_8+`}aIJIV26
zbQtU3i7tKnPw&q?R=>^$MM|8th;s_k$e<omT{|5!F&ibAi+PxjHqWVjxAc>po<$eE
zTl(_S_1OO>{dc4v%_rIa$E6=Rw4(BXK40-H#!@Upa)k7cF_wZ}JHx$U6@3lr2fH_<
zol`e)`0v$EYyG|+8&R&!mq*1{jZ3p8`c`43C_@EyVGpWz%m4f3f7GhK>XZkuj)-GK
z>mS*x*Et7p2uIKpJ{^vcNu<!WP5wVFkD(KVdiDc(O#bUZ!%p?zEAsyWd6jHgE^m@K
z$E{>s|0g4!Cgji&<#Kf-^7L4ybf`B^&`+VTujfJecd!2<PYa9d|C}MuA<1`efxLv+
z2UhSSp8TV5#qm{Slm+#U>&Upy;7wtDzhGMx>5J$8+x`K2{Ly27fBsYEm-{{c`J!-_
zypM;7Yj6&bgZ6rsztj%0XTdY{q3AOQW;ogV%jd#Kvj5Y@D3D`Oj0u>8$(V}in2Fga
z!CcJ4d_@1>&Z3dV==nw*$29^9>&K`kWBIaJSo1V~ezN=*=Jb)PkbBYdV|%Tk$2Cdf
z8W3yg>rq($OB%b6%l~Bc4dbtndSLt&>gZ`?)D88xE}ctVrw=H^{nIwmH)AVG(WVY=
zzo?GtRsW-Fue#~5I!gT-_xfpgRsDyi<?1^1RLcT&+a2{ET2Z;dI0NxiU>Eiv`MU83
zD+h;t^jLSNaDaXYg?)*{o8F-gBaa9>hGu>8Nis{O$P9AVJ`py%|0a58w>bz$jSVn%
zuZw@7hs>ioq5hxfKgO#kQHS&t^}n#V&T-DQANOx`IL_lllwV|jdq(Xj?2q_?^m`WF
z$kTh!Ag;z~(mzA`he-bz=|_&<ia7tV+&a`1=6}BR=gN2SCex4SKc1o2-g9p^^aao_
zp#HLZLmF`|#wB6Rqx1ogSCO0Josl<@^`9LVwM%f@@m)L}f3(dvu*v&B>HUA;{a^I{
z`FHAk18HP@L$R;uzH1(00E700?e9c8mmG@WXuGBTE01?5M>=<C|I)kZ`EKJ(lno8?
zW+R%M*R1{7^2oPI=FqCFiEB@f6whdkMKO|Fyno-t1bXdZc^H%Elks%?k?%hCUzLq8
zPF48!)#}|L*8HaDe{2jKS%UJ<83RYY^(}4Tcb*UP=<~4<i?I~T5c}o|;}6d2D|5UG
zIrq1QT#L@*(s)nWM#wi8rH#ywNrd%&+lbBBic*xJ0{!~*cagp9ggs=Rb^P{`2XF{S
zQ1I{Y@x=Um%yIr3&xi7FSlgfPw%?fGUOt}Q-x?qA73=x9KINPYDt@d#|2NNv%GW(>
z{j5#Ga_B&uI~mKqJiT^^=Ql_BNk4^p-$L9wFzzuQ*FdY@J1*25w8sC@aiJD<XUF-r
z#)ZuJaiRW$abaeR52tYk=Wqd+a0P8!%zsD6hvvVZ)F*#g-#nQ|&%NT%&^|6S?jC0h
zn=!5n^#3o>x4m5dKiT?facKL{_<{4r4_r2Wfb6<s{J<mQ2W}WYfQBmT&o@dRns-~j
zf4}+r_srjSPAgfhzq<yhDf;jArPoc=e^1Yl_2Rth9<SpjZsRUa`IhgK57B(>g)l%4
z`l0qSh9Y;%e_%MhbB45yH_m`Q8a*4NO}~1#|LpP~Wv}aNJNiP{{+jP^k@@#EFNCqq
zD@GMPwc35TrZA7|*Pgv}$8`<rR=Q8mp?aG8N9twgp)fDsbIG`-9uvef36n7u(=ijX
zQG&J?^*^IymV7?Lw>d?=KC3N_IM?gx`v2Z{Gg_4SS>(nTe}ZQ3c&>QnVLlcjCoGG_
z^v+?vdn~0dLk~TVZggRVuuAWD6}bjU`dV^5s_3!rxfUB8Z$^e*k0<?p1^ar0?;l(J
z-v0}IVq_W0f50d91^@SXK8b(g_Y_v~DfR6?n)5>6hvt>$c#(&21jmrmu8(UBCh48p
z8?i4iMbDt8*Rzq|y6-&p6<5mlP4OCZAdh6b{L?7U(4Uk)r|72<*93@b1E#0RKjax<
z=g|MxFN6!^C0xN(T*uS&jp9>u-22Pt!)>zfZ(j&^$-<ti@9_Em1%JX9io!!-1L%L>
z8vnm34!r??X@7t(*e9T&I1GB%cZ#7Hj*%FR^75h(_ZJ*X&+;$kMkPWqeF7%o>G-jG
z(*EfGrR?rwtgi0G->(4%g~L$c2o>s}7%}3naD)_Z@j57`7%^gs5hJD;F=E7@c*PV`
zjDrkk7=$BKs6!npR2XN@JncP?d(Y$ETTC%xiYca;Vv6ZF#fTB7aEn*G!nv=t5w5m9
zx4-+xJD<;5`}<w{`?S_)eb%FAK<18Xb7gDlGs=||`?sG>VY?DY(wp}hW9;1Nn28&3
zBj(^%+=dJL?-m(<r_H_(8Dx<|%SQGOZ7a03FVp^Bs{NfzBksRj;a;m}n=2r#&*`;G
zw8zu$M%)9aVZQ6ay~6h+?)}qTqo06Gr1)Da&XxOd4bBJr{t)6C)Q^ylq2d<xmIp(4
zg8meq#xq!hb-1wqPq+U;Q@#5~(lcmAg=02h3$|f9c3>CwpvyDwMt+_>*?0I{-72eo
zmE)UT*dJ*l`;U%=@*mRZMC^a4l}-baXv6vbhcvynlnq}6dKS4#b_W^HsSj_6i~SF8
zl5ZpSKfFV}i+uls^y9k1<xS2re)0Uiu}u@)C+d!<{~u=~*yF#OUt2OXyf2*(@FB7f
zj|v}=pWri`M3=gLH+l}M{~uH*Kdo+#Y?b<dCHt?A->R<P7VG-*VF#HKj$?E>9apwU
z8^C<OEn~~+vESiy>3)ecJ@5bj`CQvV&qitU2V%WH?*|A+pZ%<3hg8ZJ7>PJ{w+5r>
zjjpYMUow_H9?il@;RM;_dI$c%#y0WaUtwc0(J|%L2AD)%g==sfreh`sFW1jX-WYZE
zIgYhBzn)`Ue_)RI0ev&Kl5zcwJK}LqYNLKjJHTs2;V$93aWC#ikGfYcx@rnTx492l
zb*~)y=oxjhHu3EV^BBg+M@Z8<cbdnLGLL~C_lAmVMAYq+k1Kt*^nCl)O79mB_>YJ1
z2p+=|cnaB;QQ>K_ZZiKD4QMPm=Tj!8@iF<9ajyMsj(G-aunra2ge};HF8%!7eC-}S
zckc@R^(y|gzV$4b<7>C@Z(Dcsf6<N(r0V&<Tlv3er*D_ed*0y=au?DG$6yb=cNyQ>
zb=A@vkXymG7MEGY*GA0u)yMfovHnM|J?7dS*M=H;gZ6^hhn^PhL6RQ#F*rZZziz4f
zmrg^e<H#oICr(M@jN`}(<IXmCk6Guwfj99s-od;0*7}d^YV32$>~lQw`G2$fuXO)A
z-K+ckyZv|l$9vLwA0OaDq*uBBqsBbYdl$O@`J=-p^v{s1lJ-{rG0**rKPj%d)jAC1
zm#CdFI-DhkY<K^jbwiQ+8fSk*;gLul(y!<mo5aU9lsb98ewsWoMf;!p-b5yZlVr1e
zHrhF3F&+~!30EQeQ*G>I*Jk;jFW!R->uaxKUrN~*vgMfkk2ZO|{e=8~SRO~ZTmDbU
z`w8zY#r|jXMeeqixHMB_T+44JJ?<ZH1KBfK`shPus`6!wG62=b&HaCUSXkfje}o$y
zHwU%iYH%yPafN$B1^b$BKT-xIkyz$l#oy+bfnYxUeUDqq<~!jI;k$4*?#1r<q2bjt
z=85bzrm;vpL>u%V-+n-ubidypz(Xh>HY{ZE2tDrk^%(gCp2C3o%F|^3J?c2*8mvPF
zzS+O9Nq7sk;pbbw7Y1K3|Km@}rr(=iB5vR>qr)z84=OIxuJhusP<8aX;br{ln6(CH
z|4Z0;#yX5A{w4fce%vKrZk>8Xc=5pZ!Y`&=5w@voS6UnJrK$fKemUX0_G0_j@IwD3
zVe^zr!j@^5grA=_)=Jvjv1#_D;Xf)b4dur!3+oFn3l&E%3mZ;c8fyL4fORu25C1;y
z^6*Oiu#gmPLmEBEV#mQ@;aBESy<~ll%A*&D?f7Nt;_!yw-o&=Wqrxw+dus@93s)CB
z8}?M`zw3TByteh(@Sjyb3$^{Fp}uT&sN4Os(75rZp`qZXzS$>3)8e0o=Jij8<jyBU
zYyab+rQmV>hmVG~;>Y<0kLqK1G`u6tckv$H#|QWjAK?>xhLfl(G5$b1Km!`J3p5>3
zuh?h&0h)`oyF2Ea{{NT4XEEfL?o)q!7aQ7*o~hbDv>EhG=Kmv$99opYt!P6#^6ek9
zl;b7*e{BQp#TSK<(#o&@Yp%~|dfYoPrEWBq-kXv(#?vPvcS73g5&89(YKDhN;;I*4
z6s{t#L9II1b>wu^jB#I%Z^TUD8<0G}K1eUWjzd4YQP2NpOByo#e|}=q6zzWtwaK9w
zH#%nyZpCdVzsVZT<XyNMT{Zmw1RsIy&F~4xjBu9B`~Nfe|7hcnwxfd&eSUwM&Yk=U
zY21$o&_25`JVd6{$<yj3kI)~(6X-cDef06eGl+9&bIbS@eyc5G@6Q;sOs`(XhTivV
zsQAOPVZ$rWYQu6bOWl+Iier1@8Xj@4?=_BHhkv&IZ`pcnf~bhcALoORTd)n=@vZV-
zJKH~9|5u&woAtjP(%6MPs73$v>LU-E6L78i$Zyo!t}G1=;s^Cz4PM5+)7y|n589V_
z7Pmhevh+9bCgQlJxEA=^^j`NL*JgQ#{x0I4<XL1K(}(JvrQvDu@8NxXfZ9W)p(a%t
zKBPCE_WvhJ!$<T_&^&HUNFw2wCahhc{9dH|Mtkd;@R?&$%bwH!{9HImueimypJD97
zvFAcB`c6Nme)ya|;O9c_z;oeq$9#$5IEx`K(=igGQMY1Es9&-sG%Q*Z8rMG;n#!IF
z2_(^smbuS`R<xlV9jl&GhdlTHon+UN=R)_i=R(gE|6k(&>6v2xKhFQ3Qg=A>To~(r
z#$zHTp?2~b>-Vn-SJ4}5{5P(lUx(&0_rKNsr`G5TS`(&=n~586Bj(^%+=e@lUvEjf
zRII1mC44u^^%Z4X%;lulC5->^J{oXg{MiKc^bzbd^6du0jIDF*19%9J;4wUbE<RlM
zdShTV8~;Q0?d0E)#uJ1b*|OUBAGA5H{irc7e7_W#M&}`MzMrS0^E95p8l<ayKk~{t
zdd$;3sGx5`j{VKfW`k$2#bor&t0x#k<hO08Ju)P0CwHLcu)KC!p2IHTJxG?SS47)O
zCRREBtzn^7d;_w-(e6mLA&nk<^Z8|k-$1$X$ZwPHAkN`=m#ny5yTCox^`XCy5AY#A
z!YB9)Coy1t!RKWEE!NE>&&If`p9@2(#3?U^lOr)2V^RKiVHi(N#3WpW!ONZt*O1qt
z{lt(kot%jqF$d{mo?pMUL+HIL`G(8YA?+hr@!MhN#$zs)|C9r<uekq~=fZ7{8~EWG
z?T^F4y2J9hJptC%$XE0`;%~3}cF6wU8^=T5i^1|W;eN6~nbYXod4T>9^6R_oRKHXI
zjO)8RBCdLoHJ!;PP`kkUCZ9&lJadQK{|)qKum<Z;flb(gx+;DJ8jiCgXV{IS{0e%0
z{huT38ynP`VBg3%mc4@>ZC8GOp9R*HS!H})skV2rcfR%kwkm@xJ<iW-pQlaK|7^z&
z>_U2hG_Z%BU;kz<KRovTX?)JL#&g*}zsL1&)-92T3;1F4JTH3f2-oVoBx<5w=(>ws
zUraZ~)0<DR^W)ry-}Ce17C8=$8{J>4_>B9jcN}V+(}OJDz?*m*@8Df@&FB9kj;Za%
zh4~-q*s-i_@r*7!pUs|+XPH8p-nrg0^ZR?!c^@C(L&Um4YPNQKdhc@fe}Z*>=$|3C
zg#AZmvHM?XjDWam>3vRqiQ3i11du~^xc^f3<=Dn)=5Yv*L{eJKXqx3di{1aBulxUf
zuZ?J$C0~}PBa_Y3)Xxv_(Z4RgMmv8j;{2fOPmK+rUl?z{QT=?qGGXUAA3a;9ey)Ds
zvPk`ZiTXd;&JXA~t^SWRy|Z7v{7QcDCB}tH>ndD>>o6UI=JU-Y`{jol$N_u#Wxp^+
zfIbJeB97(h;a6l;q;C_>`L8T8=#zf?2xDx---WwzFVaU`1Ma8SdB^q0uW!}uxzO|N
zKjy_YkJG2-n1}EP9>WuO3Qyx1bWPy5PqY5OlyiQ3-{Et9d-ec7ejmSmMPX=N#g9k(
zJbwEUetU`a2WIi(rxu2Ob0YiiVec<x@B7T(@T}L?hp<N4aSg+`X23f7hUcxXV4Z+X
z^eq_t?L}emH`dRZIy7t(-i{sEh1xUc>O*_zaZSU9gT^z^8_-;;K7<6C_G<s<Ti5eZ
z8`kp^`KWPCxqSbBH~%-q|8-6hZAhaBS-gQa(UsBvZG3$X-#?DC@8kPt$gFUVY&pdL
zJ;MJb^ZtJ&-`4-6$xi><&JTTCI`80JyoWe9Acgnoz52m>#*7Xh(8nXUnE%fw&inuT
z*AK;2^MgMkKS3@1Gx8*A=nae9*AjlzYeU25;*#UsSF!u#KgV_c>RR1D8V)#apLo}r
zAd`sxe}4Yg3H`zN()n@B?S>BPZPJJQ%Ka;6^Xos7z4CmY@;GybeXo-LE9L*q?td=(
zi*|G%g*Z>R6YZ{JxO7HhG{zz=F6Fw%)8n2p(f^-FpM;!n78z;x;VN;}p2Icdb*QCJ
zCugE&%-8<^UTs#wHzG-obN`wSvJc3Qe=Ff5%kPcya8s51k0iZW9-iZzTX7riz+Jc-
z_u_tZCHVi#)IV0Je~^8p>K|lQI7hbZm;cdLFE7@}{~392_c{N*vr_qzQEq=-zC0kk
zhmidP|CxLO<xT7rx#44F#i(K7Y5Fr*gLR1Oc2$s@um#%?=MBes#M|k;3zQGoLEnYk
z6y?M0uj?=FdylwUG$4t=`@R#}$TWH|@S^XH?6>ab8|0gK8}Hy<#JxDm+;iNEqx`Z$
zdu(Vk6`#G-y7uJ2?}mmC$$st7ACVQmQ%||i+V=F%a1x*6OPs}!SJZzo5`#^~|9?8n
zejcO4XyG{jZ!9?;6EO*K&D5*NYj7Q=qjs_Uzfjppk82t<;0F4QXr3bf&zArB1WhIK
zf7RFaJ;6?7lz;4af}SK}`}cdUDfZ38wzai;FA8&{aVz57&)di@cDx(;b#~ePJ~ltI
zl8t8XbK3k{l&7s|Q?|CVtrzUyYVH4sebjeI=Ptx?Zxz?G^FQ&elnf7di_4F5Tg8{2
z$L`?~*K@D<UU9Kqyn5O2aKG>acnFUmwt?5+F?zIp4S0h76rxZ6&Gm!p7P<e$j&Xc!
zM{Qc}{*m-tn(?%Ap1~TdLj^Wr3$~%_H2)uQkH6maZ1PUFnam=WQa?K8{?X=`c688F
zN7(=U>_6h1$nDbEfnC^xbUcP{UQ6$F?LBCqCy|>VZQ%_1ux?!Je^mcj>K!4D9cXi0
z+<&;H#(B%#7s|C^_QW*kaW3Hb`^@{ln|&jd?0>g!LVX}V{)cQnpx(b$Kl8)#Oic5^
zq2UekO}veF(6vhbFO|p0UirN5u)KvVJ-3(r&+z}H-?pCrFCTU+k~f#g|753d9RKsK
zbjmMQcPHP+`SCv=(9f@*^&$Nu<f^2-)qjw2{j8SY{(bMC`0X>q@joZY&ygSh;~Q@*
z8WO$~K8xfb<=PSD+_-q3<}2=$-!Kw`O~b=zaxBJUBC;P}9ww1j;W|u5|JsYfOmg5s
z&t07}KKFHKFICQpPpweCFEciuUU6ly{BG|4Y4(8)$@dQ(R))8-4;XwkggK786}RCI
z+=aUl_d6+n{PNKMTm4_casJ=H?e_Ad_g}BhPgYCwA#%NOS&xv9q4t<}-*0RJz0ox_
ztSSgk(Vs?h3EvWlMQi}-j_d!|U*0fV|1+AV>X$}xr~c=y*8O+RGgyOlsEG2kd)9v%
z_bTnG*Z+^6`TFnCw^;xEJpK0z^xu;$C(QppZT|mWbN$iLt^XhCg#P~;;{(R%-$xtT
zwNZ8;rOh(EMF0N6D?-;w;{#S3A23(?=&NL3cCs&Ij*q*^^`yPmE#x+A#|~6{q21wk
z`Zu+C)*DyQaM1qr8T->G>|G$uB%0Cgo_9HZ4{FhXwEIrYi_gIKmvV1ti}A>9*8g9o
z|KBlvuD?fIb*VV=4b&d8p91+dY6?78_uq(jgx^JSpXa*YbME(?9b4xe)R!6ej7Bse
zvC%l?R(qY`J?FfS5AY#A!YB9)C(*^mcC)oTWSk4rM`nbxWX|{9vfJF3gt-PazRzy{
zzq)10^V1)ap9}N3G}aY+pNQkHzZ5=;A-jD0m)b{;EU(j-MvldJOvEJoT;6M+KQvrL
z&+k9WR$oI;i%a1;dhb+a1DiXYJ`=eK^6xbH7k%0gZV*?!*?1jt4&vS!x01IZ?xE43
z4&QiOo1^euNFMt-|DIA$p=Zm7hP%bzi~I2a9>ODd4Bu=&ctZFoJcIHd7$;A*PgA!e
zHyC$NaVH<0o)(wFCVFr5?N%3rE%a^3W#nCCy5;4C{P&+-5w?pT_}vv@2e}LV#$E1-
zagQ2<KsF$W!ROV3$aK`3eA8qWZ{SV5jd$=a-oxOJCWZIO{y$C%ACLo=PYNHBpWri`
zMA>rh&AYBX>OM-fpOc@9FTZqHsHMlTi(d*imMI66H;I+<eU&=kD&+v`CbNI+NW&C%
zqJ&)_^W)#xh&cZLY3F`~3gz>L7wpeTAM$HuCPt!*t;)~;-~2cBPdK-p{gdXm#{aW{
zDWs3He@Fd(!2G;J=I0&M7I5770KR$ZjP`?L>Idl39^XAh{bHK-12VsEKbaNIk)vJ1
z*mymzYdkp-lW-NTLESj_$7VK2yHUDLWI~!r|JmH_KMRd<aNKmv#0|*%-{Nni$31m=
z*vvWfTajDL{x1C5|6XQ(gt+Rt>M-P8i2nE8WH}!pKmLB2=Q_(d`VsFJpL9L34w4^d
zNM2Zf!7*{1e$y(?zs9>sct_;gWAY}}osplf)*g8MckFLvtb?>3Lf1a||A72Y_8yV{
z=^4l7$9L>h|30GqZwmjz_uM|2|A7?J=tO>8oBH4*(s~TJeSB<W*#9`rt$k-<ctZSB
zcp7p1U%F8H){(;SjBtM4UbcM=eI1fz>J7-Rqp?+<5?9S9-b8Le?ObDu$?d4Aieo<?
z3;&KC!n;t>Jw9wW_FemLjSuUq#)nP2$A<sF9>3LMi*onpizkH_s{S=>9y2LyTRAcO
z;^@TC;J3k#uMXQ!3=hBb?p_jC>AmlmIwbtc{UrU~hBW%G{a$$Gh_OXiekc5TvNasl
zYj@5PH|=}jSN6r)ar%l-HT#P2a_bf0mjz?ZVf)umsn6`Ch2z67^qXy)@!hbs^}FH4
zssA(leBu8L1I7=O4>L~xa^q2oysuf_SBZDxzp{8E$}8=QN4|r>JB@p}PrJv1_Qex^
z5AWjxl>g*A;X`szd)Hw3*W>?M*!@2l|M##qA-{cs&u|i-<4a@%U*Q&gzw{xm$`2Tc
z(HM*In22iipFP;E?0r=py4Ue<Vv^sk!Zpwa7|ON=bHmtA{de^{tpQhQj`kG(aS8vJ
zUztQRT1xr1eD$_MzWg-(+q3zu$gkt(_zUX;>@_|fecgQboqVcF<Ky|p?IqU#_n*@-
z6E`5;Dh<EiNbfzy|F6;x@jSo4cc0^*XOTH9trPYIa7^`T`w*6^E14%z%V)n;+-;~i
z$p3Y0W5PQA!gnDlt!6Z(+=uIEE1acY>Ib1^#r5uER=C?S_oAcVTJ1Db!~OIJ@DLus
zV|W5j;c4ud7uV3TPXvAU0`sZ0`}}9Y55sHbBi5AaS1Y(FG|nptb@OM3`lGXzxidm@
z>5R~{#9X$5>(#%sKe*;N$Dt8T6I>UPMXqz2dGgX-hYD=M7Hq?I>_Ats_7~5&$FaR+
zUx{ZU&&NKV+yU(mi@cvD?vu>FpIM%*<IcaI_Oha|OFDbdfF#mY_J6Q9WA>h5p_kr6
zuXvu{+NVz6FW!29nbY=vbnd$OY|p+T>&9v)q}N(|v@NEAnnS)DG>YpH&LWvAQmz$+
zCg&tF_J36Wub;>NSfZZ3NV&0Ey&U=WhmHIV=e&ux@ebnNHSdz|;eB*DrhAz>#tQxi
z`FHyd^FdnChIVuyg)}<xfiynENB9JvA&%QQNe)=oAp7XB@HzcU#IZkT$sxPtFN{Qc
zr7;rZSWLtu#IX&jR`nu!uY2o3e*L}Iv<cF)IKO{!+z+z<`_})w*cyQP`Sar;4k|xz
zjq|U=bj-vJxDj(O_`ER--#33`ocfaR`rrJIa2t6CYG)h&Ab;OQZ`|kE;%@rAXr?ET
zP`*cBudaxFWRvpk#EaM;_A7y8H#>##&bc2C;2}JM$M6K6Le~QJc@g`(kbPdxKJ)1_
z%h=}?>@(T2pZ!PM5x(vr{_g?ye--;)%D$5^?rG^fgEd%(^l8T^$1CW)uCE9A_W$q8
z|FQjl1p8aS|0Uy|o7G$SVRMyPsLdF2?Yu3>um3ZjUBx!x?MN<gU#HXq$izJLk5<=*
zwgawjuj@l<zxs*)kMk3{3bpSQsRNR|#p--yMmS66)B$%$a~JlY77a*7-FMPP)=l$#
zXL-Je{qRi({V$SeMr{92JFW*=yn*x#*D%|dK6>vs_m8*f?;y8AT|&Jhv&#LiHZMS2
zZ2y0cd>^&qKOjFuzWslp|HVhbpCCEe|4wxsC9cE&{9XJ1LiK7S(TvZWa}wE4t@T5m
z#gIL|e~iRGZU3Jk|DkP`{HMPDx9$IH<=Zm8NnHOTjsYAk&D0_JlkYN?9>@QTC(qB@
z?c)o?zQ-(b*x<S<9=<8Gsh3Q2+$0QKJS<#A_J2A&Ttf~9bv$x9X5t3ih&i|wx8V-l
zg@IT36J&p#{#kPHFXl!U`?e7Mqx;1@fQRrHvUiq*C&=pEuIZMWLfkX;DdDFP+ak&z
zH!gu5$5zB~_=!{UojOS~8TS~faL$H~lCZ`x>rjDB*n+_y`EJSn-)kcx2S!=@oZN*y
zs70Lr*FYxGhIC|}I^I#=EIq&8%T(Vh{S91Le_(=dc}i^mHNSk;9M?8SJ#|iK5>Bk2
z6Ow4|Hs9DeZ{lsdgKy6Ndsn#2G2Q5ye>4C8X8!-pA%pCKn?r8y&7mb>etOFM^j7oJ
zH=CbcWqvx+8_iEgwcq0Wm-nRgK0ZKg;T&yObHa!8#+9yZ`J8k4pFeeB{ei`<b=jQo
zkz?Ao-W)z5KSSDopCms=s{3Z`Qm(7?=J2KPS!7*PysrFpE^>X9u5YL7t9O0=vtzgG
zLmHjvI_mz8yFaq`jQR8Q%#k_ltNHVu;gD*10V6RQWAV8>GM=1>x+R|P<ZnFRah|Vq
zlBYahX;n*SlH;zzHMkD7#qQrR)9H<~ojYSrm`T3@&4tcIVvPHDTwGf-);VwV+qwzb
zlE_<;pZ|A>ai?+qUoZ|;++DaE_u_s$fWMpnr_P^1ew_|>;vvU8g2(U#V*URq@@YJS
zu449=9sO_T|83O%hqkHmKRVc^6w>rg?E<m=Z;f=;p#qzbUd+GHwzq}eyMX<~Hu`qt
z&d6uTpbtC5?ZO_^qJNa<MGn;69Fk<2didHI>e4^HIiHW)#AUx27AmyWZ@AN%qoeo%
zzg1U{=bHbQCEviCcpLAa{QK%0<a>A@asJ;2<cEmse|<!Lg3oXg=hy%GoPK`&uP^Cm
z(X0$e#&Q6${e8%5z8{RlXpF^pOvK=|z74Yfra9p%a^QyrVeM39(KzL$_kNwYIR7t>
zp@?G#rVG!+`T2i0(9h5RyOBNz|9|KIrCswi&db-ojpezOejDz<UHFId|F*JG$l-rw
z{@>lwxfl220n{!(H~;S;`uX{PkI)~(-_HMgLR`DHgQv)+kv?%zc!peq6u+*=Z*hN{
zb;1?M^7(V<+ozsFb}iR_fu4=pFVM%9W)R2N<>xPsQ9meFKVXB~9oNC;rjVw0E@eOG
zv+s+(QGbf{3N|>olAXnE)xPa-*pE%FVGFj!Yk89IP{~K7=g0ruZoMgbZA$;edgCAH
zjjpA^F}3stG@p?+5@;%;i;MIB+Q@XI#J!MNd~^Qa8^UkK_)_(jWy*dsKmRZD|2_Zj
zZRfm$ckv!#{(qnR03YJQ{69Ybx90z~8lQ*%mHB@kN#_%MhLfmWeQy5W=k)XQZ@#3T
z#ox~VD_bFdV@QqXT&UgSC!@kh`e+Qswe{5x`Y$)fP<Y_0i^6zvA}apC|F*W?KO6ry
z$+1`A8eE6zn28&3Bfd5MZ@=ey*mFgep4-d+N-6&`%0JJi9e+FjV~#Xp+rh2mZOC3e
zD%?R1T3>JQ)vx{k9~mF;q&9$;&5iNf-MAO`BaYvCfP4s#;4z#Z|Mvv_{P@48=+EHq
z$Nv?D;o_gh8mvQp{day^1^xW`@A>vW{Xfm)_@+qk>zc6E8vA9|*;}{U`u{bf!zSmX
zcJnXIwckSDhVAH?XMBM1$bDqSIssWSXIyn1zqFoz+OU-WTgv}kChgRBLNfE6&`j=d
z?k?;>EgFzS8`9`9&bfP)@dw5o^pbtX9c0ExA33xf_>T5PeY!_nAK4+Ca$I_)_5V?Q
z%(>FZ;tjlsT0U~kaciy7&#yo54*gv;Ynw>&)6cIz@SeDD=I{4~Kfs5G^Y7!ln~&%l
z{-o`Wo%w|R8BSum@40S;``@SzS7jX6R&BGsnS^gAd0P7)%3lrPbH{#(vlvp#USK3f
z<Da(w&DH*g|91Srm?2@TbjD*MCLuka{l!)E-u3J+uAyIt96gInrS?IrziDEK<IAS|
zZ6<2<wcSA8h#GpsUVc612;Yh%y&31%*^YIV+r*!*SKdLt3wPsw#P$CkARodbcnnYA
zDLjqAkL{C4=KCLhtWBT3F2=7_w<i0&gH7bXecJG&G}eDxjDNH!Y$xkhYJWfj8tYwG
zH`|OPy&2_Ag<*%|c3}@{(SRh{(6!<l{^11qe}w#BBLB~l{}0Rm>Qb#}b6mS}_rm^f
z^v*^44<0m^!1@evZhx$Qq@@|h{gkP*#J-Cj;n;?sCEtj;y~1MsWAJyz%?Rh$A7KCA
zp|>k*-X&9|#!s(Or>4J;574tv`HVjPLVkP-nLEzM@!OkdXf^&D`EhZ@><XLGTx4w;
zzkP&HP`$`{2xN^m;o8c=aFX7B%-A#XOPs}!|MdOzvytM5(|cRlquth|p^rv>|KP1`
zYz<pm&wl!^`pIk@F3i8D=lxW2taHX=A|~M~T!ZT{9bHGYKWNkMA$zs&_mLUlESb|b
zkzfCR7W>F9#<_MK?khk4-o14e#QFE?2I^B~DRnE%lxFR2>;H?tf!?S-5!e4Of7Muc
z;pW(PpiYsHev|%*IpX5JLizRoFZ2FV_CpeP8&VnfK3}`RBK2?KyO5pjxub8M_dwRU
zu6lKb{5m}g+4PlcJCfb<KVtcDxAX4B{dfQm;SoHBu6o}e;uzCj^sSaIvh*DC^M4P?
z|7frG{Yf+K4WFiW?pFUesC{2LPvL1igW419@9Dy@h91|vX^@ZC(JSz+@nZ$n#}~H+
z+prz^@0LHZgC4)*UF05Quhy4GCeen9KNwe_{+*`xAd5Kve&9;`ZPEMRDhO|qgV!6!
zPQHV8@gCmC2lx=N{p}<26MTk~IN$#EIlXrw-(<eF3;J2aG2e0Do6KDP16eoiziEGD
zvzz6o?4!kDNS!hlBN5vG^5d22&D9X@(g)bR*z;fN`J-=)=Z`Gn8UgwBzYod(N7#R|
z<AAbtrRR@M^xyaO_`eoqfboB0r8yB5mw7(-6o*Olt8fjjL-vcy!*p`+FT=y&pM1ZS
z#`y@}fEzIf=jSiqN^g|s;~ebU=y#y`6x$*k_dbtvu<KTsf4*u`XwZk<sBgQeQoppm
z|0KOxKlga&+=aVwFYd<!cnFW6Yoq>u^i-MuuWzPL-+l(!QvLs{^#AMcZatv?oNPa$
z|DQ|=r^!yowV#?49+S=!cnVJ=UF;Y<L+>3SZT;wL=<ARZ&I)J9KK(y&|KIA?Nn!0G
z?`4_xzr~Y6x%D4%ztjIlZ<y!47JNPb{$=+r+$`L*$bFKVq`3v<S6i=%+>RZn^{rHp
zbt%thy>vE9hivj}5}r%)jC5Rki}=<`_lNcx_m^?)-*~_I?_;6w192^;zb${;r?{Vs
z^+EZsJ@J2$o-B=7{*&Ij()};@fAluQF@+cQKUnHHimN7j$Si7?O$u+2Z=%Mz`E?Pb
z^|tUk$nSq}P(8!{#{Cca@0l3h6+iHkk>Ne^eSClq@ew{j)*d*Yktgx<8P|m`$-&?J
z-{CCzi>1@Uw&H&aTMu6sUYvhj_~l{!Wvj-9$|C(`d#8tA73)u%J3ds+A0J*W`fm8O
zYuYvDdts-3yH{4LJM6YM<xc%kl~;sUH=5)AyfrZ|c3qF!8}lo3;Qwfh%(KRXY&;t}
z3cd>M1z&pBUxwC#FN_`eJT&)z9+Cxr4T&iyL(}n-p|RjEp+TLl{*bYqM@H+5vA@;+
zk?M0J^~qcus{8p>C+wYe)YzYT-&Dc>2>*`Z^pU8T@o!<n#(xXt8~;aGKW}>2<X--x
z@L#n*#rcEwyBYJ{@N>M-s{e7x_^_=+|Kr^6h4QxwLwUF;Y+tC)5HBtKUZ|WsG3+Rv
zWFDM$O=*usnX+Q-qJr?sV(p5)hwNK?67dr;30L77T!+EmPYTn?{;TZ~K@L1>d=Ysg
z=HOQJE>}NWEPvDQK(1Jsh0FxDex)+!xHX-{-;H~5KhmxzCH?_=b&>slyvK*=kKqy2
zvhOwGpP)Bx)Q8n=Ttw{ue^&kfj5!*JYs55dEeva0?H9<8UZ?Fnj_-TQIjP0k_h(pp
zhyD!Kpob0ZMISQABFC22v6uDT?B6Q(e?6PJlWk?kliKK;CrH=171)F=*oN)cfnE64
z{DMO17fauNXOTmTK8aSet=89o4*ECeCsg~dJ<_Q~1CmiM<$sS;XP`GGTpQB#9>hHX
zlEQI}WD~OD-oTr98yD;Y+xL#}yLcb@{fG4{eL(*ZAK??!Z8rX=O8x(^`hU!qC)EED
z*O-XwPsjTIXO26G&+#SBVn_p<gsubn{|~7rqW7r&|1<jkE7`c6`bx-_`Re~@Tg?8W
zgC5t0OFORfl=BLQh29avLYMc^9p4K*hdyM`w%K!2*J(+4j^0mPQ>jz^s^f&`hz516
z#zo%aeDf%gJmS6Wi~TF!;~{e!UE^qs#du7_BwU4S(0<T7Vsbj-{&zFU^a<bRX?_I#
z{Q4s|(&ylB*B{A`e`EvKE>-@cHf7$8<L*FB=A7+|>yPB;-#6L2j{eQ{NB-gd!>#K7
zXhvLr?=ETGjeBuF9>7C*1YI-OfAk=>wg27xzk}-kXhXc$jxnA;(&$9LzWv9fGZ5P<
zZuw4lf*$)HpNes>>l-7V!5R$e2l!O`%etsPYzzUp30trY?cL^Hk*Rw1e|dZdeHZp1
zmJz+^bN?CS$ER1R|E)4_!tb?cKr-sBMWJSyeq4HE<+;B9>}%?3!g24kBogcS|M=GY
zzh!)B$E3(KKRip%&;Pqr9gIKS%ir%)HfQ+ZS^jvAtXs_g)ppQ;Ml>OTB%1MtG~UG9
zcn9y|J-m+(&{e?yM-O_7rC%idS<)|w>HGd&ODliBjj!Lnfd9XU|G&ib(L3kz|EHQm
zAf1o!2|h#ZWd6VSlk`UZeuMUm&*@(xjva_=118A0hum3lLmK5pjKtv8MPW2K7UMC%
zPn$^gUuJK2a_u<ZDW5mCRb3@qcToO6EdNiH{}I=eNgzpYMpl`Bo#QIhhc`4?>xw=T
zH{eG6!})(@@;`EDSuFpf?YQz;`S5r1|JdI-(zzA4;SQvIN2ykJgWkK7eZk%Id-1pP
z|JHwT%Gfh&(E05F#JMLAk&mF}l>B#?Uya9vpTOVF|LZ!Y{6f!i<rn%^E5Fp^vg&m?
zvZdAd#ctyl-Ag+<P8+|7wEO9FKXv;2>sRqDN{tgF<DCA)LE|UUjHmqP(|88&)yqfZ
zI#gg2)}FE6-w5wxLI_)gw;}r@d57GEIR9V|x#0_Ch_--QdIOSZL%VxSlRe1d4aBuw
z;u>df(tEYD_N=n58~q*RV%q>R>HvM{fATxwUGW25=BT`G{=i$td<wsh4`TdHzE84W
zefA@={HGU%Psl;v#b@M6)T}nY=YVm4^tF?HUs$Icj%$N8?f<$xiLcj;FCB9h4aa<M
z{q};P4{37kw^+lFthk<?$4L5U)Xi3|qoGJS&$c!dD%a^rvY9<?8Nn`~Z2~(nhFx%M
z3Tgk}xnKMGX?6uY2iX6;U$?u9&$q+3I;X}wcmFdU6EO+tgfws!z4y3taSi=C#5zM3
znWNGYKV4ikyET)%0ky|23OAB-P(yF1KR2)cR^i)_ta9DR?=!IbqHu?}yKpz|#r=2y
z58)A<?_Yb29>0+%$fuC~b3u5TtoW00N&Nn2=xeYp#$gi%TZV-#WbD7)Mh^UBSlCYP
zz%J}TUDen2zsUP2l>dvpFEYRX#4Pszd1Hzk*MKD2kVX%(=-Mm)pa;F^Lk3yokYE3Q
ziu^zM>+yf|IJWh?{dq$gZ=!uOKbws6e%~SAMOxf@<ooD7rW`0^JLK;^<oekVaUb|C
z?lWDpPgotXnx0?J*|Umk<1{NvqAg9-_!iMfk7dG#&ie#|Kd|o*x#2SX_}~93oTPt_
zFL4$_63SML#Aplzb708+KNW=WWKYVtpNzSP^ht<o`{vMB!^bDD@qM>D<|@BkgY;J8
zlF8{vIi`*+tVhFQ*M|JMWUE~BO7?$+dwTRt$iDVpVW#76z>S!LTX7q@PFw!~`F5W9
z^7ImUd#*f=Tq*y59Q$9){*&$d+5f}JR;1~j7`T%EuMQCRHygNQ__xLn-XZ<Fa5wJ9
z1Bh!0Jw!ed^*enZFPsUF(W|S5geS<Fjm964Pop-Y%zFJR<L187w(K4qlUU>)9TWFj
zN}{gxY^X0g>$kI^5l#H81d`ony}PsFUg<o8HCTrVY{C|7Ls#l-=+2zg{_$VhKmIH9
zA%iS(XxV=@w4&|ES@j=2oHSENZ#^42(LUyE*e;zN*o8evk8n)ES#w!@cSX`h13ihH
z>&+rlcvc-%y5g!ANssJ7?E=4(Z=hzL^X9uRlwUhEyeTd@+kK(Qaf!KS&2Lryv%%}o
z?z-M~Ov-hp51kF~&?_EfchMt%_R5!i@@Iz33dj8&-*wD;cpo3&L-fCXCVWH=yzo`{
zg!~LAG1&1{_?-L_XE7ux|4;S&(NN+#XSpu-l348dyS`@EKin}RF&bkr9uqMMSK(Xp
zkJ+6I?SB(o-#FjjKK9=|wvp{e)c?u+{-R{(>Vj~MG_J#RRCk{VGs&8RXF~0vGok!<
zL&IF*tT8yT4Y0A$x8qw*kV)ZYvTVM5K35ye2>$F6<Ig7y4fC8cAGKrT>DP@L{4?8J
z#6}}gz(yDIBgv$5TKbI-LL1^-z7BqQ${3&g`rsw@_9-;Jz<E7nuk-rIjBu9BWkOis
ze->g97ULsj#u9QVmZ9#1`oU@cUGEyu)a@FOq&L_2&$&awa>uQ}O02^9@dvBv=g04r
z(#w#WuS{@GtV8$7KkLP9#Aa;8;0r@TC0T`?7<gRU2HF2(&xWi=_Gg8ml`Maf-=Yqa
zAK!giy+&LGo3Y`Z!q810yv27;_W$H;*h|LwfBVS&IDms_XFm>+hj9e`Nc%?P8h}UX
zy(PX~9HSpcuAgo2U+339%#Z(H$j3$F0Y2>^zU>h{F4?@F?wk`ih0{2Lg66Qc5F_xd
z_5aif^8R-L8;$(>e|-B^{CDgBRUaN6ilj3J<1hhn4Zs?dzv7$Yi#H&)9~b+rnaxgi
ztAnUJG?8n4fAJe!w^;p<A3E7FsZ#Zi<rjr1^om>9XX`@r_^o%U>zl3afGj;XSskfr
zc$n(AX_$dosMR-7qb*}Lz4563cT5R=E}D;fHsTVeJS+77urSOMKXCQ1FrQq2g;<2e
zSc0WkhUHj+l^Fc%u&|1(JFWd=g#Q@h`A?JPY-vyNAM|E&wPQ+AhV?k#PZs@$jlx|=
z`JczVW3u;*{7+_tvt(|u{O>uo>PKlqJ30{eWsZ9ccB1Ye`^O%|yxAyE#k`pylW3kL
zpFKPzZ1q2tsKQR{Mh)uGHAViM=J}Gn@@pTNU!O~!%b{g~{5@Iz7T@mo4l?EVG}-CD
z+SeO{^OiD2T5%3wE14FTssug1uh&N9$i@2WUy<f9*QBh-kbUy@`cFrO8+?=12V95a
zYmWGh+>2Uq`^f!hOeh2Py0_iu%7A7vmH`L+b`Xbf7=v#O4M)g+9K}G#&~Pk<)xVFE
zajaF@D&Jda2q%P3;WT33UUnGUP8PJtABb}RD{d(Y8>}bTd&)cVelj)QQMY#^-)89}
z99M)fDA)cmj+}sE4F0wtOeXukKRiq!W8X;h|EJQYVFqTQT3voNS%NtCZ!S3xHS~s+
z>XoYt!hGQc$orM-LX)~>VygZpW5Mf<9d982ZvDT?(cz4HTj;z+Sd1kw4=|KrExPpM
zck`w5{rijg-}Cw3$nUeq=WaP+{P}6)&-u#j-R1|>pBvks-f8}SDgR#^K>ZQ^{|Np+
znkMl73;F;2_-1mc|67LT$Qqluf?SCWKl4rK2kjc?|BL<qWd1+;v}s+K|H4OX;p4Zu
zzBWF8JKw*9o|?t~pTYmfDrpT`3vo4BiZZOnMr_7bRH6zyu^R(*+DFL#7m7mrUeAx7
zKr2#6?{kj_jmxL^?l(4HI78oyoc4sQxQz4q$T)X@pZNVafP<(l)c!DLbT~wha{?NK
z57VPB+Pw1H;}6y@^e&Wr>y|42_9@$aGbu8CK>N9K^5g$5_3iEFAE58B=YuRg=eYBB
zxqgJYaJTe}<O?LwggEE-sC16uI8NXcP9uB0H7};h`xE3nbQQ>-@?SRQLB#oean5gC
z`#-J$8q0#t+48>l)LdnO@EPgUZZ%%LYG^2E^(~Wy<OmcYu0fbo79^AjO=zUYy?QqL
zF30$N944R`lQ9KTF%2^?3$syzxwz0iD=*Fy&gpxZPcFouKA=Tp`Ab8>VsZ&KsQXk5
zv!+<MEG!jXhW-}|!g6w8Z3rvKl~{#X2CODy8IWF45K8G~SdV=9d|KX@Co{;e51A<l
z8~wH!Tao?EgiuLVVJ8N)`Rpd+`u{cL!1ca0GJ#f<Uq3AGW97#KYFCPL08`@Q9Kc#O
zpqpNC%W&%i@`GJdEC-sF%lAmm(;iSF{~s~_i%s~4{qH-6hXvBeU@ywASN|aQ;{Xn#
zYqRllmHN3I+e_y6zfG~%nXl)+vzM(W*dMebKmOnEX}@=pbsN84e{U;49r<;AwFxBI
z`PkQf$bTHh5%l9Ij^Q{?Aiw?q+kRpFftBn#`=4e1b7+xPtG0wTb%l0yh7SLcLVC6K
z37nF~X`DepTYTn)q5t&}p^!fC`h*be`UrZz_SwNdUYqZ$s#bo!I-xM^-fT^m#p*A9
zAA@n2fb=ppcGd7uOz&lDdpzIC^eM=V^K6h|+vEKGSpWEW)mLGv-)3MI^7RkTW;VU?
zn0M1}d_mkF;1&M#p>uV!#8Gtx&vw^}5#iVF_mv`RKOP?uc9@&<tJy`?g#1gWQig8t
zKM{U;>_m8J-zj}^|EIpX|113B#HsM&nbTqG@zY^l@qg+6`ZARGueqog_j%Yb`}0sf
z;>)o9#9zavy?+V+p>J)T-{xaW|ETcuaihZvvrdN11=3554!<b7I4tnnLTo=aEc|kn
z`FE>FhRPEo!;XW)!>^Pzi~P12WyM2f)}i{pv=f|?p9?OLpT~##&ErEInk%R4&-=bU
zps}H;YHX;knH2V5x3-8Q%8LJ-FwxkPiJ``Raj)&XA}o>SQY^!AtiVdF!fLc1y)Lw&
zHSuqu<>>TKDy|Hvf~&Q~Ul;0>U-h$;DJ8CjY?`L5NhxzO>ObUq$8N-CY{iB0qdSC3
z;jRVz|0VqY724NF@ChcI>wC-6a|Omf9#HljQU)GT7Lpypac#cXKHo`JNu&I&;;@^H
z>-g7@^%%U}_jaGPKjZiS<NgMWV;Ts)gG-DHc(yoX|EmAiaji%%ljbSV;~q=hWCnYY
zZ>L$TJYJ|gCVQ2;J>)*W?Z-6i!~q<{AsogL^rJ3eeSnm8+|f8i|96T1E!H2c@3?uD
z{_i6F-_!Je`>kDHct@qa>z(?g$xh*}l<@%>{nKQx_`db}?iX1f!2JFk*<M^6j{2Wt
zIF1uYJ3i(7Q}o{b&OJ6MoTi^aZnge$eaIQI@1Xwwsmi^@#l9(aP#IWzdQvE8SN}qd
z-y0J8?On$R;kbWTa+d2ZaUar8kahZ;>*)=yt+7?Sa}tN$-vP(%n-q$iHwO9f&yN>`
zar6l&#$<FYQO@pVU(t&`WcDlHPpIRe#l5sjJ3l@^8GK><7yHmD+%U~^omCvBNNXyR
z!p)VJ`ws2HFr`@gjy@WDG)(i`4AhP(4ztMFs4*#}ebJ;)LZ6Fyn2+=X&kqaey=2eq
zNns&<5pw;mjUCFkw!VT%VX?R+Sc+=<KrAC`4i)qNi^FpI3amuKDc32k>457h;s4K4
zSD40sU3Tud$1#xQfluSu)!6W)`q@{*LK%HMHexfjq7qfui8ywBH(7&vB+!Z!2LEX7
z2QrS~%a8+?7lgg!KJ3Q<^se$fFIVS~Uk(cAlr33g&?jHSJa<U^VH`m}(xvjsK7PlO
zh3dcjANLaH<}{K`?kTPnnyhqxDP@#z>Zo(p9zNGk@Ei5-W5OqJ3SIT`KQ6TYEq1MJ
zYZm$TzXRXs|8rc2I#vp4dgpF?4qa=k!w-ym`l<YVneSYhXHd~~t{tHKhZlu{4&O6I
zAZxC15jpr<V<X-&e&W=SFh+PBCZHI#OSH!x9TFzf<DAMk{$~n(D!w)UV4n8BrTkBv
zUq`%&p1}Ea#E-j{Y0jI0S(uFy%*8x(vFY9DK`-*_-?XxE$f1RuYGtR|$o4brKbaCv
zA7=mAvN(QlzH}B~Ar|5M_`${W^Y(KIeJOJ6X<S=3L-x)1di}qIcGG#j!zJ>6&G4|y
zam!I7zF`@^dWCY6Z?)kT`CqtMI6u#xT;aIEXA8qhaursi6lGYCjo6H>s6-V89xV(z
z$$srP?FWq|q1T`u38dX;3a#|sde3IJFufZ&_C6~vgFervuKVlp4-;G`^6T=Ic>c3o
zr+dse=bPh;_X_XBejG%Xx@0%<>&wX37wVVFrQ9Or&lJ}&&GRSYno1q?RI%qj&hz(8
zSFH71{yZccl2)~GABV{r<3MY1gx-&%XgF?s=uu-n;@A%|iRMGru~UaX=J)d7$W!x%
zhU4_umvDkSg*etOhd9<QgP2!)aoTTZP>}MyZdCs-Q}3r&`&LJgMHqwT&E6RjXhPnP
zPYeyEzRPibpMYYl?<fhA$tjqMX_$dHhBn{7r+k_voNXBzW|JF!sC@e4s4$m45A(4A
z3o-bU;b9TE7}<}_&3$QjSVHeN7HKKD49l?s?F$RSN^%ueqZDy$Uuw1Tf_}dJZ9RP>
za%Yqcr<4(e-W4{B+loq*-!dvxkvp*)gYtR}*?+Zfgd8xJtKvoP>#I?rRXByUe7`vE
zZ5{uw`iO5qT#fW={ZEFz7yA&`{%AzgM)iLrU0<{7Jm4La_<cVPAc2E8gu|%YY5sp~
zADFFuV1_<#erbOGzp?tw#;&(a;NPRoZ|&$f$oD;L{Qm*|J-@f>3_n=7r{4T@^kw+I
zHGE+8#vEB!#{WkH8qtKrX8!;BbL0PyxQ2cl#W5Vm37o=dbnWE-C;0z0{C~dvxAq^@
z4$*Ru|Bp7bqa(%t&+z}rPUp4rug^%QAgzqW2;|ov6kkN|RoCso82UKmR=f6<?txF<
zhy3`H1)~*C-UH(LgA*K=y~OxK@eK<-H<VwkEk}HEmgiRDInMPQQ8$}Uhla)Hj%$*3
z0?Do34=UK34do$Baqd)1!wk&AY<zqE$=~?*d$oV~7FvA^ZDjib^^Zlq%_aO(vQxOn
zGwanZ(nWU9`?}r6v+1K}R(Wm(o)KDRc#cTZJBvKWF`gqDs$92Y>yEk)dUCG!hNgwy
zBT8K3T+G9KEWko6!eX>9EDB4=rC5gLNUya1!0Mu~g5Jwk_UJQQNneGW`gL~9xjJ?q
z`?*?N^|+!?N|vE^R#8|_ZbVIqu`BFj;~C?4J{}sjii>?-&Fo_n`<Ni>R<QpGHnPTX
z8S!L7I7v2_8mq^TE<15<{l}~|d@H3<g$-;#g?U9?`{g<0_gOnAPb-5n$d9k7RQ}52
ztwo+c*<S4VlPTde+3C3chrbj0Uoi*Z(m4PBJ7K4^`)?W@c9R2t)=twjEY#5JF*wTF
zP-JV=A2k2J!#kjNqx?t1Li<wBg`OhQ*h}As{pi`Oe!AZE?R0%)RyeLDS)C{hoBe(O
z2XP3s;%ac19_P+C%pD$%(EHIWoD@!wO$&#IqvDR?I8NY#e>6LUQ^Kcl2Kn)CS8C(w
zl+Q2%MW|zg>r>h%GGC7`7@=){2LB(;(~P^X8ydzqZX70{7?UvtQ}Iv7ziAt2ox=Z@
zR(lEmpG=M8{}=QBwG~d2M!#{_GsuCz7KT~mY?NRw=3zb-U?CP^F_vKPW6zwgyKb7i
zubvq1eX01kc0g*cdI-Ij?e9@XSWeIKAL5+I3-+JwuTBjQEB#h+t$coGL0CnvouU7s
zK-)jPajEA&W_T#2m!Wx~=fB+ZM_j96y}0&_b|i8$(#rL%WF?{v>XFZTmlT95;hl(m
zKe?mIiFv+*Gse%X^!+10|AqY9{cX!Xy(#Q=UJdGzKr2${M%NPlrh4{&yZ*)@_1`1<
zzsdF${NL5;364+GJIP|-Mn*b&5$m%1$eJ0(vtU2{01l#|SpFR6`ikVwlsZ8x|6DyF
z?|+P7|0aZR$Z?0!kL)Ekg`;G}>o<iBzPn@e<2Zp+IE^#-XY>CGx_qw~fg+5-I7~n>
zzFq#xUxz){eexHgt^d3Ee}7Z{_P?fWNIC;Q_Fb&KDU@G+6M1u(B5o?CVFqSlHcBuT
zaqRy*vaHm1w=#tJ^w=k~kX#ghzcYlzWNfQkLdN+|OUY$ej*9zkR*$$TwC^klM?8lG
z;#OcK(ueHdo+=5e=)I@y-+rPbtfrSDH(`$QcTUJSrVnM}st=ch_2fp>ZY>F$$*rhy
zUW0TRGbN!?xC+TdbCkbxLX+bXs1vR?m%m}@oY06SB#=ZiV*cOhyxlQ9)FXjbq|i0*
zX6-NL`J2n%yTZJF^ZhgC`)A3VaLeACLo3?Qjt-=dMkm&GE8h{HUAOf5*Ukxh$$@K2
z!alM>8}Ei;#-LwZ67~xpz(E|sVYJu#Kk^9raTMv@{zLpRdhfhBp$Et5Cy=9Ok#XOB
zs9t_^I3@lx&Y+;%GhBLes1aXCZ(Q%$U<ADg&Gfi7L!#2N!WePw-pM#}0@9P`gko|s
zQq~RVVY_-y7lbLoQ;}bHQhCwGR%K9kuq4!@!MkicR-%6Iy&NqG$>SxVc?|#lwDma-
z=-c1NU)Bd3=i|+g&W0xS=e}W~i#_h1#zs%k{&h_I*J(BgacocyEy|SE)$;#b`5ztf
zb*hAYn#KQ}%1>QU5@tznHfl>tLJ7IT_gHPsq`CCOW_cdX{cQ4ay86&O@%eo`68z^c
z^lb?b{FRM=eOOpPUx-Cmj3ro#Wmt|CScy2VXcbwfOsRJd4X*#f{wLG?mv9^-v)VDG
zD8qVe#AcMY3=3Pyu7mFXu={s?z3AKO{&%|nO839f{Y$IWy|$sf(ETgNQYYL$I??~L
zo5R2l;+%&$_7At7gR(5H|6M8VxQ192xf?aeUO6n(lY@Vl69#|l8=$wM-E+>*e{}!(
zy$8f~BPG7aGwMybCgHuvy0+X(*I6UZHBJ+^6Z^0q2XGLFa2WaZ^VqNk<ktsIusvmL
z5dSvzg|}4lFL&}U$#%bWuxD}JTbu{iDcmK#yNV5*um2t2TK`|TN&iEFZ<+TeNANlQ
zXFnSFmW>Ar!%=#Tc?PvOMn8@dIE89)?dl<?>1R;TqpU`n%}mW39!Ai6XZvsVH2ZJ}
zMZ)>@bJ^3(1pfDA_IIN>2FmO*tkrfm#xeQ%f9o#_<LHg2)lE+DyDl~k=vj60!|Emn
z)ltZ%{)@r{$F!f))=y4GdcD3{aw<}uOMYG86$N3M@C;;^u?xa|tK>hjj$d9s&GU7g
zjjl7+sS<O~)v=ma4G*)NGaDtCi+Pxj1z3nK_M}^yJ&0pX`^cDXw%GrSV}FWb`-}SL
zQT0#usDn%ir^!yoRrB)~NoO&#x69wwrK{bm{&%=AED_fzk2dTt3`^<D&@7ys%m$N9
zGmM=Pw*o7%3TxTn^W&RW3zsV=%E<iqzTb`t>**V@8Cy}O@2|c}`}=;ry)q`hFZW*d
z9nIRbD;-mXo!E^U)FXjboZEl=>+z4q`}V2dWRR^=&q2#(?eFu>^+Eo_{m19ZkJ9Nz
zZWy13OvkuI>d*AJra{kq^%DC2c#LqiO5WKj50MAN9mF9V#^7tl50L#hih)biJ!1Gi
z?W0!~`7k^m#5SE1F`nJ;f7Sf{SGDo~_@c1>rBUIO_`%!17f#3DuO1T4kONOn3<bTu
zZ;ZfT9h>yAcCA;f2RW+9cRVB%iLa5r$B^StD{cZ=j5r6P!L=tAdv6PsZNhPnr@CeQ
zwpIL3{``gg?^66YetKO0|G592?EER1ifQ;EZo)0-(y!jVg8#jYAIUH58|N7o^CQWe
z{`?m00Ii3#4;;}xKz0bHR`Mg!>Adz+qr;D+b35+DkCC3`82;ov^xiT2Z@&0_^q(Nd
zH_ZxX^qcpQdH=ih8~!)n`9a6!{qNoGix1woXmof~T)FWRN#W)b{8KV9Z*=(eyo<xG
zq7h*yUa9(*@T*m`!j3VsLRHoOX1>~t@XIMP!gf@am4ugyO2RKT-W;}#`(fBxGCRD8
zpD&vd2Aiyv5Qc>fuNH*I{m+y5DF#0-2tOnHjd6XJjOF~=D2-()kMS25h26Ug!t?ZM
zW#1lk;+g})!fRu!Ay_;i)S`ZOu{MGUZ1d&DuxWeWIwmyfGixrm)cYF~T2GJD#yU<r
z>T<sNqakH4w2sxw!wb@U5ij9oyn<H|$LZ~v;D65)g;%j|$~EEN3vUds`K=BW_3F#B
z)Vqpr4C_n&_poW>9Q}VYLzCZHu%-0};pa1M2rrb(44bFT4BMvM5Pos+;!yr0We5AS
zecZo<UwYmzEz$qCNdKSj>sK4ScjtDY53l2uvJid~<JhL|j66pE!~U~XMWMCST$K67
zTA+i?ier@1WannzZ$|r+HiU*n))`!GPKsj_=B^}HSZ8odQFz#YtW)0UD#`n2Zwdbv
zrM~N|_W$;s-o*=r;dkQxfIlK#GfMq;RQMCU*PeJiC+xj(t@(dfTEB0C`y3O;H;f8>
z__JfG5045Tlb@ni|L9-HzoN!@4bo}Mj0#@}e}&{C<r<%<scKY6AoPW`!*DT1;SyYm
z%P`ni5H2VCpB){pBnN(65UwV##rJVN+C7IKkUzvtxCLp?GxjmuPS3CZu~hv)Kg$Mf
z3EyfznQZ^BMb=<FWDegxW9e|Gb81VK|EmkbkLiu4&Hvk781A9phh}ymd63N}oB9jG
zPsFt^z9>9MK8)yBKT1B1=x6tgVE6H)@K2FtPjl!)2EX@AekA^9cou6>j_2_L>h#^`
z+XqtG2adD3&PgCiZ$7Gjzn*`Ow)y;fbS&oIBTer_*J|?*=DCi!u49S+rDqlxA3)}2
z8~;~4JiO?CUc$?G1+|O$-|X<K^u`5jGhU<Dp;_G~i3FOkc7^g>oh|yGO@41d`OEAt
zzpH~@5%kyo#y)}cKD>_K;4S<XbxYX)O8tOntkFKJZk{-#oKiP$#_t^Sd;9@^#GmkI
ze2h=gwV3@!kMTae`+QDE*grChoc4{Dl=dfSww1B}=vaNOZ92W2{YPAX>o3yzE3(%5
z|AI^}^8L&j627AMu9i1f4w0zu6gl5k7U$QmyjWaZe}5Et32Mg-373+Wp~iU)(uwuz
z%Z0B*@-*9nCdVakt+?+aJ4~N;m=vz3{{TP4O<2p8m$|of?x&m|U(w=uzBDx4;<q2+
zcHD^{;~oq=J|x^n_A85jLJnR!Bs@qyj7RY}p2Sb_GxYz3{rv&k%U>S2Y<PH9+**|5
z`KYUdpMSqE2)`JA>njd}FI$5{-&|aOOSL0x_(O51uvY)eem}qd)+_Yh1D+pVrN4$8
zJ&O$b(B}QOFH?_>^|VFm&86ziJM9xh#xW#4n~kwj56<royxusZeaay9u|D<VjQBcf
zHlYO_=t3V}$8V6|*E^%1sX#s=lj`oxn|-@gzTrys-<8GTEyw*9zr*iQySzBmEG-Uy
zpf_&T_m4l)|Ac0G5(zZn&*It-O$r~ApCYZ_{4eBRk<x$Ovs=GEz7YNj*_ufqSFi70
zUwZ~ui+c%P{@U_k+H>OjMqJ-)n7F#h`seX?`x6xCpGON?(S~+(6zcmPVf=x<{mxm&
z9~2pXFwS)pyAHBXICIqagX6{@kkwnS2p9XGifh&HZ!ZX==(QV-!zjBVTtaVjOam^Z
zUxsGqCHw6maLQf+xLjQOv`OJg@@m9!57(04M;r&y;~sl)z3>l^U8P=Ls-C{vb)fE`
z`#bFZ`i(z8(?a79kVG?n=$xBy3x0&#aVLI^d(d@6{%7Cv<I6Tnw@UpV=i7H%W_Uhm
zLpwT7$^U26|H)40ReRU>N#`ec5D%mFp!a{cI6O*kJmvl4ar%?U@2BaTPaO6Bkzap)
zvGPX!XPy3!pE>4PtVMa`)1l#ca>MiT?x(}T3-lN95?)5#diDRxb7T4%YuLq%zP>~1
z{|DJgyyBQw@mf3<O=v*}x*{vJKcWY{=sTpSQU+$poN}WjrT#C?HvNn3=vb)z5ovUy
z{~rDCO;?;7U!yO6;6Z6fxBuc0UMB}%92$N@zJ=f7clbR9@4O;J;`&>E5dI_ngpYB4
z{jE>w=hxpVzgph?6Z?Be{>S<KH@mhL^N;Nhvb$J5Db_Ai?E003hl=I%;?Od}`$O9V
z?++dH{P&$H4xK4=!+PHj8oGTyXrd>OT<!b8+8X&OAwQ9SaXtC{hrSpV{!0G>KgFN%
z6+%uKkBc!1m!NJX`-BEGqDdJQ*UinZqf6%3_dm-2Kdu}jJI*M#$n)#_cOBvXuVjB$
z{bT>X{qXQ`ssFhQm*YyLoge$pucr4FC<7gHE&cn*31=tp-^cLZ$?L^ctBd@A{2^jH
z!A;~Xs5znTlG2WlDGWanz8%Sh`mGVyE=VBmof~a;et)C>f^er}YM1GEExjoGnBG{%
z|94DW|0mnTCJV<ogo&-{BIG@eX|Lypear@YU%q0C?-TbEq;|V@`91pO4+=kwIQAxo
zyq`|i9rS$NUw)i|^1f+>=ev{ruk>H?bNQzQ;Zf&3jwkU`{0z@xExOq7ZuE@sKZVjQ
zmj4Unf8=KI|K;-w{eO=8mi=EUeQ7+87f`J}_aa$yOnVF4{1W|Tyn=6yk7?yQBF>M0
z)o-sMtL!c}#-fhigcfulKVDwnNc4BRg!}M1Ht-i>|9e~$_$}ez;&=Ex(k1dJ{y^_l
z-|iVRH2jhNC*->2?N)i5>?;@={wywz5BiuK&_44i`4{{ZU*Ic**L~Nx7^83rF2&%T
z{D(!_|K#BYb>v3;ZU0BsI$@VN_HtZ_tFcagig{`+x;(>f<e$HH(C4{dc>eSjdHw(G
z|M2{M4=LY6zW(LB{zuh6{=)x{>;L@KcP{<!<9hrMH{lli2)84v@BdD6@NxBzYuWGm
zA>qfu6^|MhU>(7G=(VNVRdF9Z&S`8|I3)ap{vhI-a>*0wm}Jvj_rIS1znPy*=J)@j
zC;0x!6hE2I{sZScj7RY}p2Sb_Gdzo~g#LZU_Bf}P?9*qQDdnS+IsSc1J^x>Po8Q~X
z4#&r}{qy7h$+-XDTIrPIdAxvhiDL>!hZpI+{P~_0{7L%D$Su_WzMTKR)PIpN@5cTA
z^8G;c+SQ}OD~@{=HFMq9RQEMY|GTy8UlW&{=Dx=1e<u@D^uO<P|22+FIF4*`oeAv*
z$s>+CVBZ041{)rhPo3L@7IdHseNnFV{?T(l{@dsMAMyUleE*NU*CM^v&F+7_d!6t7
z|39SNdvKQ7o!I%KTN^x5H;lmskLZR8OkhT2U;?wEBX!disG_&%TBtx3WMLO-p#r<G
z0~x4=9q1NqakGUR++c$ZHdsj3<<$+{UP-TBK>{0?h#5#A3khsc1X-9s1PLUNf%bmR
zi(z`~BvY0B<6FPqInVQ+=Q+=NuD^3z>i<Uvy%X);<Fs(jqGQ&Z;XK)o*3vh_B{GiH
zro=Cf)y8xGuJAi6P51Z?-M&HP>NmqRem8Inw^7Z%YV(`nE<Lk;K)Cnm57FfP8qt6_
zm*X+Nrx?2QR(MW^Gx7?CAs-`<UXv5XkcAjnq%TSiu6i>}B+Hh{3;1-c_IZUo{ZK#C
zeh-sfE5bC)Kru=%8*?xZ^RWPnumsDn0;{kF>ri8EP%Y}%2<wrcH=t4Z5Z5O)tHWE6
zIbU?U`hB#rW1RY(>@2V*!F{jS4?a+4=o4k^gZH!#u)#AnVGCMI*=fl3sq72v|7G;F
zasEAIcNhBvQVr_vo$7)Y?C4M2^P%`J7Jrln>-T+w^cde@llcM9bHiXsPN?wvxc+ZH
zc@T$j6vuHAHAtWtZAc=8p)1n5JU*~V8sU!X|IU)<(T_{0e3Tuokk@bnw@`gg{vXH=
zx9Rnp<$v6z-^c6r8~a~Ec~>(<{e!yFU)Z17)8p8F<5v9yl>1GzvAwbnJmfBy9v+iV
zk-VUON35fJkfQe@t<LO2^KktdoEocrfHrzN;vWB<Xl;KZJQt2N^C8a93;#eqxZs{M
zZ-iWW&qZNl7(E|-^fY>r!U%qqkF+7kF{s}6Mkpl5BkrMFw^zI{k$W;4=}kzW0U7)I
z%3svK+xf5Nzh3;wrekk}BEOl287M{xW@8TKA^Al7wN<*wU)eu+$A4x1qk6y7`_C7~
ztNDTp=y8tZB610qp>LM>P8VM?j#s=I&lu&Kt#G}v!Tf`*FVkRIOoPkJ{o)trKCW>+
zp8Xf+cJ7+@N!Wec`T_1Zw=*+;VB;HMoqL9Eu+Na2um#&thH~tQ^0+ii_D$Rx_LGBa
zz7h_S1DpJRm$SoRdZzv_+a8Y6JM{147}s%n9DC|kN1deCpif?XZ9g{#RhuUb%2#o0
zKaTOoxy5l`>?U<q<8t|I&GwM+yJoZ@i4;!bEV>@a<M*`v$#_OUW}Z8D-$$O~S#j*G
zd5W>Ote0bO?U$7iW7!VLj_bxXr>H*;#We}Op>$r|t34vzepIz9cOEErgnfzo8X9jZ
zcf4z2;AKB$*NW|7_m=I(HnpX~Cn28gdc%FU5X<=6<X!ae@BT>Lz%S0d=$oQEpQXLx
z+R%+R!b9>gp5i&eKjc46{ww-83>fRr<&Jx!4U2M92>IlI_GaaUtT2K;3OhG!4P(ga
zGv=Q0FQnI#b?36ec=|-dJwO^;l>hEYsIzKrZucL29O|+@4)sW&0gY(F74I_H?}{)D
zGw_$z|4sckbPaqQy3vCadXYvSnm_nBw0!h&IJNZS(7O2J&^Ghq&_3y7<?F|xbM(id
zrr{Ig&!2=k)Iaz{pZb%~F#40wIPw$r+fPEV=apbK=3v)D`;#O?m`9(Fq<gx?eiFJT
zd=h%_E9?I@eiB-63axWL32h(vk3RY&bd-J)I%o3pUJHb?2urXG9f!Q*_K(8~dQZE!
zVHJH1`u2+3PH{XSu2mn0b^I#FeH=EBn^1lE<FJL?hN@Y<>m%P4W!&XxJmb4w^qqa1
zhO54hHe?Ii)+uxKTG2*tzi9p5Bm3JRsZZU7Zv4vme$wh5VOC&2R{5_El812=$5ErL
zQHwftU;Q06$`|I4k&S4&8p28Ue0hFx4R->~NH1bHE>oA?&kjlM6i(wT&LeFN=HRA}
zLqB~WxjkGWuizSP;1+J<F7D$YhE9JR9+OY;9C7_`IIGQpVaUg;^}i$NJ@Q$%`aV5T
z8PENC{qGa~bY$}HnVi4Kzw|NgD@2t%UUyOcWmg!_z4I2^H9gK7O0>wo7<j6!;@aS{
zcR~?4^f)I>BX`+ncenMHgWEn1Gq?xJvO+Ogg4r0l{k1TMY~7d@=8^NU0E-a!R%u_U
zA4u<!$K&2$%jhf6H&(uPy_ZZ)$O^0Yt-(5Mz$R?LHVmEp#K-pR=O2f1a_}K{fpTiI
z_Qv)Q_VBB~epEgaZWtEgn%~3RM^Sx5nN*V(j??4Z-?}dO{(yX6t8Ym*y8gQUSHILF
zF-6_5P#v*Y{r8c2({JOr*-7#sDhHfxcqBKR7@Hf8KVp~kn=ZsUy62oa>ZZDiOr!6S
z`fsNCzf}ECwic-WN2}Y%ssG8&;nu#LSQlFLr8D+7`wMoQu$P?YB+%Z%Zg7X)fUUoo
zyA5f2+yf}}gdLy$lV@APkMC@;pXAnXbi%gqqet7q4_9tCez!3kzP=&+U|>V|{?<+5
z(E3f`AU<zc>&&<{;lS24VgI=QJ$yQ0efZX$KL|--r?BgzHDULm&7u6r+OQMfuG$>F
z)3rIAcI_;_yKzJKY{#}xF>7PkhwpjT=e66zdDr^!{k2=e4}71416#tO?OW}AnP>f*
z@g>);;2Lh=Xl)3$$Vz8u{nY<)0=Kz;)-pU)+o!V1xoann7}Gj1THkDRs5RbKvvGK6
z8aE;|-W#cp_HrDs)jGISj|xM}8MY$!qBeQzu5j81#<0bV4fp8}@fc6>dVc1Moba5x
za&%q@=fr!D{-d_)FnaYQ<3BU=LO#8IEgQ<pyfA`33QbGpA$|==Y|IN|_!Xi&H$RLg
zCt_foek?h7|BpfuISn%~RP#nCCQC3Ib5P5^U9*FYronTL2=|0=yVz)|Jg3&4n)=`K
z+&3Q!un0@A4C%-Dq3c6w&^PY!UfpCbcZ%%uZfWCKEy8FPW~(qyk?o$*#;<d$wF~S&
zHCN54xtXhc%nkLElnctExYi&uzfk$G!gE$(4c6hy^BXpBC#M^~M7MTEJWDa9Z_wMW
z-O{UnKsK+{|Hte3PqVa*&@oaQ1iOy=?(+-67GZ6}(8)K9!}*?3?`Ica+Zf<CP;8En
zbqRa;S71L5;xLY)b*lVG#x)ox$r^Of6J#@b?s!M^-t>-$@0LccYn#x9BvLqyvpA2M
z>H5d|k9Ar4$9O$IUH|g+{QpV%xB9WIWZMLNcXXf=$#d)j`q6O>e2+eLitOc1+bf_C
zt+j84e$TmtE4YS^%fh(+X1GC*X9jk^cr)Chr;GI2{5~DSa8JrTvENzw;LVWfhc489
zJM?C_?cTenI_AB*y!TH1c>Aiw{qq`kcyA<JYjE#<_q85)Gdv_8qoeB0@RWRx_6E-$
z@ZJdjmpTgRd-~#6J^P7wpjTFXH4NjQj}aJ!>WlUVz!-Y{dSk#Sq>o2asXZvsusEMB
zmA!1Z?~A%|-gC6~n&7?Ah$c+*o5?7`G|WITN-!HqZMd$h>KOg`9x}z<yHH(#J~RuX
z1*azYfBcVaQ+!*`>F~Tx_K7*dn1}hORNgEgtEvk0A)NDf!<;(y`z8G9?kS6qFlV5_
z8Gwywx@kVkT=mgM>LWIQO6hG`{QU>cy~PT8X8)g{9-gWmo+Le%%3ow7nnv4mV5vU9
zPS<z1j$h7ytAw!z>#zZvum#&thU9Af`}NugWY2bO0<xDoP4?Yn|F6;qXkZ^8TeX+k
z=<UbUGl$p*mh&4fj>6f43hYNmfjo+X^vwRc`pJjsN73hf)ADmK;{LkF`JF@!5{ToD
z&15|Lw~frq3)ioRc7fQBPV!6PG~)b&xG&_uLG%BMVg7&Id*Hll{kViHxP}|Jh1<A`
zA$H-Rr*DN_E40sYm2ZdC_5I4X&H5CsKST%pG5Hj6@9FNP(#8`ukPoE|GM*8e>HjX)
zM?~D~HqoyB?^XYkP26Yw_Bp~olKwFa`51vwNUk-$j#uOSsj=#1qzly3#vGfks{geK
zPLZwdYujpk9&vx3&XvaBT^l2uLX5{mRF~?%Vlq9hiHq~!i|EtP)XV;k1|+iHFu_y$
zTQB{s)_z{-x!3%^8-<S<?%#DwzGsih><=(YyJ3p=GcO@M&ih>#|3mB_$2^1l&&9u3
z7$umEIhcp}Sb#-HZj}GfjUK$-|MjYGEH7r_-y;5K-KYKIIqeJC|Impg!l)F-W#kG}
zYfCNiy;sqz=yjvL*G7GL?sdqt%MNJ6#XkJ<H^K&fo3I7jP=<2sK?U~XAP(awzFdDD
z=Poakp3C&9^4abZ&YEYBuHip;*!Lxy(T2DdElH+u8fS4H{kViHXqD%#k+0_W-=N>Z
zZFI*z`W^Z6h5Y$I{?s<_!(G?z;~^g7DV`&om)=mLOsK`{@`2ufMl_-M1Nr+S`FpOs
zyjY&w>i<W_Sos^tDar?Q&+>ev=$ZYAT#IvgGUGlYm0c5F?ps(z4)ff6R8PqcBgiqR
zKfyl1_SaCW@888fafp2a<;D=w^TvhUtBn8Ow~m($tI++i92idy9nA?7$;l|fG|WIT
zN-!I9Fc0&w0E@5$Lv6ABJ}wMwlRmiP89*z@Rak>{Xw@FyKyJboY(t#e+5RNUA0*$O
zkT$QY4{jHRJ=}c_^eW{*i?pggP&3;2;aKB`Q;Z)Xv%Y_l@x!Uc56Mw}^Xap9jLE+v
zQtyOv?h5R3KGAM#DU)Z6ucI41NTC;L^r3mV@k`;H5?<WvzHNf>OV5dG)H_jm;_a|s
zcn5J9M^Rn%wsnthhvW1(CSG^s?QoJ_gQf@G?T#_dN8a@r__ZE!&H&klj^1}dl1!n!
z-5Kky_24x3S){ib_eTo7sPRs<3q518_^%b-R$-%&-h}gh(~nEIf@`>eTeywnb@BI3
z-QJ}Kse9s&^fB>2B>rQ?AE)@WqRl<+=%9C^^5)Au3|_4tx+~1Mhe4J2-KW>PUWb$N
z#6#}5cR?fK-Uac$ihE~8+fzC}D?D|*gZ`Wh7sOxr5^YXBldL10=l?Q)EKTp5>c5{Z
z4ItNj!w~0Qr1kal=|eT<K9Hl(H+i}~{ySj|eW3X5P)Lr)L`+5zrlInpd1z#nwq-F{
zf@*%V$vKE;>%_Bw8nlZOXp*1e89TMwpf%<o)MvRq+;uWB)^%;_Mtb?WyfDx2=VJjD
z;nQR0^Pvn~%9JEB`-G9bNTCmDv`o+j|Ijy@qz#VtDdynw>m02OjvDV+d)c#5U!Yta
zt&g0iZ=9u%j3u783@fk(@$A5L<OXa)(*0fIeCH{?Gk*Dehs^%#i_O1UZmt#C=KpA)
zsr)Zh{`>yBtSQ`mJ}+z$)=(;hZDbkBqr1ZYTVxNAhi`{H+=GSAJtqhB3HC>~F~C;e
z>LC3vj-s3mpyQlp+{_BcxqB{qw~PAk^cwWFNCOxA_h-B-S$RD#tm2<QGulv{r~Vyh
z?moR<-CL*Mo}!;dJkzdGzr8`-n;_5f`?CG$Ja<1X;R-VQf1gz^&~M-tZsRWQBQw8G
zKfez3NN77W<f-GeADYOA?s<%-c#iPL>IDo#K9W<lKQ6M5ons%<-b>M6@Bej7`|m^Z
z0~ebgxX@fJb^q)Ai6?1)Xa}^aV@C*Q6vm(saSng`fj7c<dJh@r@K2;qM&Apz4fG=A
zz9N2=+s#8DXQ29wJ}y~;s;j=;1@$*Mn>()mdv&gI!hH>0>e_qkbEtb}{vQ(b2E5w;
zzf2o##|!P7Ga<|oM%?pk_a-(c`eXV0zqtPKfW8v`s{Q|mj5*I2)&eX-To<#1T!s}G
zy!Dl^ij4NJHPLN6YF%_ovm3}b|7{bw1>ZLhZX5Y({#zOS)%>?|`X2o9{I?2z`*9GJ
zFSN7Cst4Bp>8l>4AIC}5DGOp5kw7dXV%gAyFV8QiabE(>NH59{Npja|{p*R!1o~;5
z#d-AO65`lHt$bdGdL-oi2IW9z|K(Y+{w)Ys+;<H(a0|C_7x$5zD*wxGug8DotIYl{
z{>wPeBeVbVLiu0bZ=E9l3$uNa{EyDj`u64=J`~QNxrC3&r-*AXo|EAp%g^iH3b|x@
z#am$*nU4_|(iRv+mW`G^#!4Skjs5iNL#)uA<{#T1h4DUP1>?!g`3L3>Or{rM8fKtU
z-?PoLCen*hg6aa}$CyozebRU~{T%u{G_BM(-L8*{MAn;OKEK$9UqCKG2Ym^-4DIx8
zYw&xpf_oLx^gg`WC$MI@{_lFvSgrpX!z3HH8_A}wx565~S%(eSge};HGL$1}jd_<f
z2;GZ4XRUC}J?OO-A#LtK-%9fjE?BpYQ`gxV&_-`Zhx_9Be|v-x*Rgk?B8ICinE9T@
zg<3}oZ=CRilSZH4?03&Wbi39w-1z|wZ-v9$nfX8Go$Erc=2wN|^!ks)4JYX}Xrjk8
z2n~p95NbAxKk87A#4+h_r}T%Wc4hT_HhaHoMjMhy;WW-h_e$vxug{;N_acoxG%NdC
zs-(XY(jVIB?G4i30qGB|&kDkM;q>DYuAn1NJ*jTIM(<h4ZlqqmLBEB*wc5M<dN=F;
z;x@lZbMWty_fh?U`NZU7R4r7mek9+jcb{@UM`m5!RQXuB(SRKff<;~NJUeSJ$0<Pi
zmF?J7tY5nzC*;#dU=+q6?)zOx4q7)qo}7ruC_?K2wtKR@OZ$I&PMAS2MhW7cS3P*O
z4@p{?8vDRCbLIcV-V^n!z1Mc{)uMbsoZl43|7ZLC9L&RfEWjcxL1zD_^}hdV^?!jn
zb*%4?G`%m+_rJ{khu7zSZ1(-P`~GBHe>Jd0o}FRsf07-*JpX0FT!B?sgLT+|E!c)K
z4BfW&-`s$5dgU?w7Gpy@*ummnkmX0s-P&mUzdbwbaXqvDkGSoc=X>_&hS&3d#5;jZ
z|F=p#rLSL4#yRZ`C)DxurkihsO~R<aejLPM9K~^*L}vfr&HDe_+3CsDdj0Ej`qvlr
zugPZNv<Ul@x-Om%)TTa-b!*3R^{;xlb&@(zI0-bP4ILBQGgJG6-m_NyyHZ_CKaIY{
z>R<G(=C@w^gI}fV=gEFl^S?x1LDfd_5`NrQ_Zs&NG>#N6WX3r>r>0i_A9dcZzJ>oq
z@t-dLp$TOdv^DW5Gx;sQPjAsC#clds{FU=>7s`J)g}7JDFVDY?XA<5QPG<f8jjZsH
z-q9kB%zYy~ruQ6@w(ykx9P#YH^mJ)#r?iG#ewCZw2;ri35vrdUBPR0^_YkPNtX{(i
z?onuz?&7`{nR7<ZWQ8&O3NaoNF&Raeh8ZYE31(vs=3zb-Aicr<c%dW=oHb58$5>g#
zS6=Uf64T@m8`#b{u8+{BaR1=7-wC@Gl!o2L*p_h*owRNw%Gudr6}blMumPK}1?43r
zVH;T%b?m{Q@yH4?j(@i<%nAGH2XPoj(Xln!fO5iddhGLePiGUS*Wi`itC#ID7446u
z^8XI`|A_p5K%USZ$ejNv&o-gF##(YD{I(fwNFs&PQD3P1MK_|2Ew#~qgEac^E9?LH
zwd2+Ke`W3RZm;^AJTJUm&DQ1LDhd7cp+#(TMb-r`lBT%__k?hTyoMXNh02A}BH4HC
zx58cWKB~9M{~MLV^!i!iHa$B$rawjI{6A@=VYIkn|I~@$K;Go=y@JW%^Xrqt!JSjX
zp_xVD`{#<n589`P!&%e9k<rt_4-d@@KbkTl99=&n96K^I{J69@{A6ZvSR~AjtM7&O
zbMJ+-^w!Joh391W6X_9s*WU|i*Lz)0p@-gm@4Yb0wa=DM47GQLvq20C^?B^%1?=M1
zwk4d|*7SjKuHF&h#DKXD?ybzq4L_w<9m@?T59NlRbq&jmlO2CDE1bwH4L`kC5-OjR
zh7s-^g)t~Z%b8K3+1OcToNW83(0<jp7{<Fc5uZ+&6uvcOQkYCH!me{u!tNbKp?vI=
zuoK_Tni{_IY-*V1S~0$xH9dT`cSfi<FfHsGH$8lB?#%G{*qLG1UE{@Uvfp1iEBs((
zaX9#7MmR+O;fa#)qq(J_#BXO~$9Cm3j($**nKM>%@BL8w;(cwb_k9EDY~1_$KhCZ%
zcwhhL{V>OG=3zb-U=fyJ8CD?azOK>sKS0kU`yZ6rkKl!G{^Y&T_u#$Iyxsl>AK3o@
zt!P90NA^E(U+2vCL*<T=uu3><unrqgy|W}#9V!W%==Jovu9C2Yz70)#OSDZ&jGdH(
z1j^5|yV)zC%(ZfK*h63sS%G$XcSC9Du?NI{?t@4lD-C@Uv>mESL+>H`KUh;)JI?+C
z6YM`wYX5;r_8*vP{{gaTsc;MA4{hAoHaP6}M{yiAF|4G%zx>*zJ<>ha_#0B%D!oX1
zMtn}Quv%u8hEr(GD-CU))qX<XA94S^O6wgH!pf}gQ?4}As~f~0ZS?v_;xBC_=_xd=
zW_!FY{+Io3pd_5;*E%IToF&gA&VA}9FCm^`)_v7`-84s!`x??)l?zCn^Pc1lezy?M
zP`XW4%`{h!ypL-7L-H}|N6N?I*U+x)7QfgwY{cM0<wN5AFfjW)a~0kXPu=?*;ZNm1
zWWJv^ZCp<?jC-i&y^v3ii2C^VLd{hB|E@Roy87i9Y{Go`2J__`$tGj6CxsRFPD<a(
z38VaG*9f-a!{%7f3z1Ymcf~qdT@&Z(r}EUz1uy5U^{LyMuNs5BY3!A3bzj?7bvruf
zoh#M9v(&%l)Yl#P#s0rt`u}LWuK$1aD`C86OvDWDF_|pFwCF}LlG+em+7jKy6MB#`
z=Fp4R=MOaK|FbKcLhDZb|6|4<v{O2?S2{QH(-tTZ&TP!VJal-ccJDr)-lJ~q_O1))
zixBrTOfMD3h2pBMv4mfx_Q*1F1>)I(tH?E|^4mIXkGQUX9rp$_j??~lqMb28J7cu?
zcj^D*_4!|=%H~P_+o|ela+BYD*}k%cdmGA--e6tf4)cTdv48Aj|2V?_v6uZreUet6
z^eJbWmur70H&3AzZD_w|e$i#~i;%G&R%M6Get45#=8HYTu0T8kES~+<Gr|AAPCXl+
zN$+Drk9&LfqH=)!iT`07#c{;Gdp!H=B)wjKiRb;)&=Y7{FaPjs*les1wcferfVoK*
zL#R7pZqiNt9W=4$H1bOq6@+HLYeNz#oW@z4N8EcgxmDcBZtb*;UF*6$c1|6P7T=~B
ztsnRXABsEKwp8574(@(oT*4Jx!wn2shkuK_jkGbgL1%8=r4MY&4Vm$i;qsGb-skra
zt!uT<$nvwk-#GSHdbs5KjrLB%wF|K8u5$p8zWMUqGV{c*YQMQ}81gX!Lsvo=MUIL3
zeEFM=eW1<wKl{s|b)|*;$73QUqX^Sbd55jyv~lMnSz!iuF{%$_g%WZ$>UT<GXjrQJ
zSSW3MD2<Viz3Uv;=V3k;U=fyJ8EV$rpHDxz4)yx9iFW(C^}eNiu+Lkq{<S`J%M<<U
z7jZ4W{`CX>>qGj}$MmoF+E-2=zH6uc`M%gcmp*;BUZm-L*l|pKh!vg{+t{nfHCTrY
z*cHd7v==wgw_qE}P#%fj{)j$2y#o7j5FN|C0}j)B7JByfx582Sar8~sm;X?o-X3+)
z22r_G-<*F95@<&CG4>C%(d!TBoA;W>Ku@9Rk^cK${r4UE>f~vDXK^0=`11JLCGK)|
zm@DM&yKyhS+;EM41GjJ+HPT3}G*d^`Px9ZC{^k6C*FE>~5RdT`&tcJNSuWUULl?TQ
z%LDTBuk3#z%ob(IDY8}AZS;6HOo!)nK2s+MBOfEsdcfLU^40#dW9Wq#k8U=(IEN(l
zLcK+%5!bR+KgtdhU7w6N=e&rlypSEHku%WPtFGFoty!h+qSv)+XOqRQmEiSyPyhXF
z?zHwq`IwwAk3Js@un0@A3@fk-yW$*sV{2>Z>#zZvum#&N)XX+WMq7C~8P^5xAuF&S
z2XPp2-NRAxI8LGl0}1~F8SU$3)1}eUm+g<C$NK5(?0FgQDUy!j{sFV}5%SITzw3Qg
z<b@kd7_ZJRO}ekgH;Qe56#X>%HY=CVdyXH@#;}BYo*Z0YE)01IS8xqCa0|C_7x(cH
zk1=%ntx&z1ZH4|ERqL}u_$R)9gLysVFf<Ox>xg@PG@$MS-yiv|jX=#2^)Kpb)l*1x
zy|ha=(wi{KJzuVW#&8#6Jkp!AzZ%s4?dnK&gB~))-OJXVMjx6_=>OvsTG58MF0P|W
z{qMPRymzz##QkI2=pChO?ZTdnXak6S{~qib#_p~Elcx6}wb64B`~F4lorW1GM)g+X
z7p|4i>xCU{0JG_H5YN<UM1!#7Ub~s~_v_uWTKxC%^G*#r-9v98=lRWiEWjcx!7{AC
zDkR<0<=$?;dwu>vgEm5y_uQ`k_d@*7iT_3MC);j{KiR?ENwz*PwlAD@*nmyw$aBxg
zH^LTrX8q$V>(c3E_~rGF<@_rBPkYD;RDWPU1F~GYuUaf#+r?|H_fZZV=HIwnytor&
zgX=X%^ncGt7pO-94f1cJeBFc{#_G#()Nen1W{$vq>+k*V$GJ};dCvIyMSX0t=caZ6
zz1J9j`iXYJv0seO8>2skR<t3W{n3F=w0@{<Ae;o6(S{Cw?MTvlb_yFQ`f2ph)95|K
z4`=yR7ML?b_M@79iM)a;dOY)|9@n^UAhWL!cjB@(2x6T&aPs9o`-5?>ec$Pp`vxzC
zaGSh~`xq*=em2U->P+%!)SsGvqAq*YKPeC6!gK!NpUN8}#)V<zuG79(VqD0lkH9F5
zK_SMY=bU+q<|THSlbAFgvG1aJi|D=Za*coMIB5qhXhx@dJG9T+(T2DuR~_Qo{u(rG
zmv+&B1SWdMWE5c<W}p})n2nkT_V+h7UpL+OH!}O;A2Y^m?7C^++acFIb1)C{u>gy(
z1j~@DHU5q6<;K|;8du+FTpj7n#@W{zmw#dW`=ashbH=|LjDL3-|3(MB(>Q-CeT8sV
zVGR!YejVf7gLT|J!-YNT?XZC!_wVo9ZfqTy{r^WAJJ*J&eDU_nd1wpG|6A(1dSR3M
ztLSy^t^erlm*d~c^G0DcAz{2fv;Y5c<Nxd3x7vN~O^^-TjbxK>w+Lq&%Hs2}2Nl?l
zgLu9EcB^(CQrd#O$h7mW%KtmXf1mi1t;fWlZ0GJEJGI4HAG{q73+E_~<0LxxwWEgK
z^P%{E;JxV0=;Kav_mZiF!sAz2Yo7x$g=+d~@+_+8b$Q}dU>*Z^KN`84xD%trldPGb
z4URh0BY_4qq6wG$<_fMMy}~#Ec?)-OAIZJ)Kf2}ho|q@Q<V~dIi#{~pF$S;vI7Md8
z_oKHfCpxam|GmnVTKPv`<<nkm^s1cjP?+&t&W;9a%;@E})Nf?>RN+q&eyQ-uH2OyC
zf7~;V+4Ws()W6z_;filCU>_y^x%B$A;<nPfW_muFmWms{1|&A-gc1BkVGIf}9s?!s
zgo)(fY<oG7MVN*-9zTODMhRwP4nCbE4kNV{##xubJ$O4O%qL?zVgWg%4PU-Wn@Rg1
zv;KZgURc6^8CGBw)?ghrU=y}rs9D?Pq&b2&-VWQi2S>0+ljYch3PhXoe)1p=<0vxg
z57a%!>2XY=8z<>C=zF0KMI0lKXT@jw-|Clm#$EjZWk$QQ<FYd3g0kd{xoCcqKr`Bq
zu_v_nzmVikuGIfVH+uA`Q>*pAjRB;|zH#dRy~-|}(l`H=`G4-IRL`Y^a~g3T+F7!C
z;mi4d=jpHZ-^tYf+oZ|fFU|j>_q*p3;@&@3$hh~<HSz{Jv_Ec<x22Ds54@v(fA3uH
zh(5$UfNHiXJMn6rV5a`}Enck?yzTdQaUT!y7*FvW;i~xWQ~w`Q|D&f$TzX@><V*J7
zXX3A}KXqMQk2c@69k0$ms?4&sOE|-jj}fRIEB?OAD0=;H?^bF|mtKgb0`G<fBru-e
zL`+5zzTE$q#$7HA#(fKR-wk2c{I^0ezY@&G9Ml|B|7(ZVUHnD;Pj8@S_UVz==DBA+
z7GM$5CHY|qxeP0iydeLh`?~z8K21H7KaoBnf1-Krm*(d!SO0Uje<XjCo$hO$%l0pv
zHCTrY==ezek4^NRy~5s+6SmN|p^rPw-AkqpC_ngBW@&?ydr&>x*Z{d7RrI=Z%C(C*
z;UM>6G;%j_CzL-8WM+TwXkV0n%Y!)w?in~K?~sGWUTeq%nlUtu?SM=oh11w|%lv}J
z+VS-B=*J}t^}i9WkOR%qBN@;3y+Pi>ZQR9uJj7!>#dCyzreB9)i2KLw9AkfQVdZlV
z7HV%yQPzy}zWPfe_+|PZ{q~<*V;?}{`vaH$NlwQjeY|JJc3khdF8^Pa25y^!&K=vN
zY4jqVRa5iG{%NQ~JrdLPZPAD(ob*nk{BArZVls*_4Kt9uXMR7r^@V$OnisF%+^cV#
zMqj(W(qi+^mz#f1wz{wFnfd4Fpmz?KKfkpg6bq*WvoQzNE&AVGZ-#mF`Xl<^2h1~|
zFF@QozYz^cc*Y`rt!xfU$YtobXRZOc3hnf6Hj5s%iW%auhF_X(qL0lW#lQE2ed{L}
ze@7jD<@`hMyv}bnU=y|=mhn4=zpSseaVKYqKe{i7KjNB%Uc@v0`p`U8{AY^4_syJ-
zFk1Y_ivMu&M`i7s4iJ`4%hV5SJUiJ4svE=~d-&Hs@^1I6UzlusVGO(2P4T}h{&&Qm
ze}#KmAE*P!gNSoK50gg`_xa3>ORtxnxlbZZ?_2J_k9|;jjXC_a=JnT2kpGcD1L7W^
zO{npk1e(!?pN?cR$8qm{Lccb-m;D_XJK-UHT7CQ8&GP7W`JZgQWBmJp@o%#Ag}Hb1
z_M7Zve%tBZxPQfG=e`+|!cO5ds!y;V@;^(@>_6%~&eQwR#0K6t#XBO-A^7nmdpP<Q
zNAG<#92;%F!MR^I-tyIOc<I-}k%eCm-{1J_;Ridu77i}{^>FCHuZQnle%m;gIj8O4
z3<tCU%WK~WJ14ypc42qR*TN;wxPniY{zmxL!rurLbALVTtNKRxE<Wq}Mkt&0=ULZW
z|Msk}h3_;7*PfAwq@(YzeJlLn=3C*0YeV+)|K5dP3!iWQwQy89H@M}^aEt%IVfzJ=
zas2p)%B~+N4+nD9$9wX^;Eg}ax$D|}Jj9T)XQ;xQfDL%g?x5_6;p~{8oiWR`wg)+J
z?PFZO@YMa8_D@k>cus${ej)td^8Yz$0J-#Gc%A>XP2*hka`!(*zH1{e3S&@+@u<<q
zsJ&x+U_f6333_I~-pj@Z(5(O6f>UTkn>wx?aSeaxnK*|~{fO?Z+6Tz&zc)$UJ5n87
zp#J|bFHH2D$tc1!RDWds8D`MycdGwTOfNx`y08&(@3I7D^P7Wtn2!Zmge6#p_^oyz
zmX|BI<Csx7*3j3X=A8Qfs%Idc^_zGh>~U<PquEUgtj~1M2JEu-dUqjvJbeqcp$xCj
z-`T7EkJJ(Q3Tb*Dn&q<=W$h`l^&b11aygdG9rR9inbzfOAHvy#3Y4c@JLaCv`tjU7
zUH-q8tZ<Ni7=5L}MlVt)%s1i}*AE;gPony|JWVE0HC?=(uq(ds9^7rnocpqsJxDpx
zK-N4{|0A<LM}41|q+Xn=jZDU~H<Nyo!fBkvm)n2mxs!*~|HstPC)EE)ZPd0z8hvQK
zr~b#OC+d22Z`)1vf4lm>Mg5O-e@^HZ&Lv#K4cx+Q4Ba*sMn)UneR8nK{0Q<fGW!V^
zKF{g-$g}a3-*fbh6y{d#gN@$hgO|@Nzx$j4K)s4#=uq#sBcFaU>v>Mqq@U+hU;lZ|
z&$6E8jNo@-v^`SC>06X4V@CeRoQAB2IgL~QeNI!>zt5=|`R6&cNB%sgF6(hl{kZ>_
z)0TBNr+xLm$w{Amo>PwItou1Fi-&SfeKeHQy7AxUj1t}$6k<FkVls*_4VnFaXXLUg
zvhUx0Bh27ej1tU7*-YuTRQe_7anHvB#J<7MY4$JW>LTu4;<$U;i<~9&YGrJkL%EDz
zzutd~74%ie%&T6i{9Gt*E0yp2zR0OP^hHkHu`hD!$wbQ+ISsvE<TTcPk<)}ZezOMa
zumPK}1=~=Dq;J<X_2)U=WKZeOb5dk4cbe?G`171*VYJ}XjxTar(T4WDU*vSG{drF3
zxS!{g3u9pX^PD~8;Fjk(736*##9<u8ahyaAGWD-?^(lt(U*xnd_Rjp9k)Ex6Ms{rX
zjICefB<Vd<zsTuEW`15mnk@JtC!P02PVeY1a#E-~<elAj8fS4H)xAH@sp|T9PCvc=
z!OwH*aEX2eP2$ve?&moTSAU+9z%_oYkG{ycLEb`#?|GZNi*|bVT;;+?(l7TzWX^@-
zPA!&)$r|rgJJa_s_5D8-mk)e@vT>vDkH>!V6weW^OT!q3e2hTyL-S9B)xF$v$doX9
zH@|G-MO$;TvcKh`@`r4_sr(_^x#Jx1&h6%Zx4+04C7dycHrPV4`ofExDvYPst6$>W
z+sX8_@$n|^IQO=JjC**NeJFh{_J1I*|0{A|6<LjG^vwFdQTpIhw8JMUUz9Pg?F;hR
z4EGeH1hX**^DrL^P_te6q+F>(JrZcBRX!b1KB0Ms@@1d$Wv9HZylUGkZ}%!+lrf##
zm0#YYOZm`E_AFN>kiFb#vhNvlfik1DKpyg}C5UzBGI9mt8o+LOq(}Z)#k~e`?O)$=
z?VQcpIrMe>HeeIBU>jopp^S`cfy&7}sKAhVCo^C77wz}`{4(u6bE$WaVH+sQwPszK
z$MJZ7jea{hu<47O8glTV^g%YG4N0VM8fVdZAS;|FU#&0er(ePqbo;M+q=(dc>Eng+
z`H8%6Pyc3$`fss%0vY@FA@$!ebu1b82|ehy*Kh;3a2t1V9}kh4|Cgt4tuF5&Q~Lb9
zWM=%mRR8~p_JKO}6xq7k`1^L_?`#Dfi1YsjN92Xa!Wnp&8=jJ{+t=#s=iK3+`$oT*
ze~?Qbvc_>3xtl$9m-cX_cg^P>fl(NPYISNA3hDJ*#ciX0GJPWAp23aJ%rSUmt%3GR
z%>nU89Wv|xy2QUq{2Rodoa{G6n1&fBMhRwP4w9?oKi@Rd&i0Kn`}%Qb_W4&AH{TS0
z;hYj)E14PpsMS6?@^btm?uj%{IP<Xpi_kGi8)~|_1@t)f(e0jP^cCn^&z8UX<=97R
ztGNdJD#djTxenFt-$2Gah%@sKT#I8MX=!o`|HcX8<v&l9iswvmKc^08(e@|n5c~cK
zvgwI@j%|KZ7Q@3HRA4_2;xJ<0KY2#px-O5AvG1Sa7yJI{NAjk87Ht4=@BO$&FRo9I
zHh@k<8$j8I*5Kh&c{ADq(l_kAO?Kd@@H6K&-WK*=&p=AJy)kThv;|ZuhmO1NBx;aA
zv;oAve>1)QoU#LL^dy?-(FPFr7>zc76u;9ri;2?Pkny|o^nP5zAiK#Ga-bqBTqAGb
z7VhFcGW$PoR{m{gKbUQv?L+o}Ed}ABYmf01&k_EGvKzyYj}aJ!F(|}%OvGdqVH$?|
zwa@SC)7uAV2KV3?dytSNn2kATy<;s1Szi8Dm`^UitMz|N=yC7K%=*9M`C%D%-vnuM
zmNH<vv`W@I(Ernis?$%cClj%6O=kAVBX4{6PmSyExbv2SzTOIpgs}p<4jYfNW;C<^
z-(3CdkMy(26ng1tvTv*T0n7D&pXvYV&$d?S|M%+ul5w4V=b<;lDq#(Y;~H`uHeeIB
zU>nlDe;HYhJ*dE-c@+D}0eenV-my0z{V<N=II8(oJz=|`$2A3Y`r|e91mc?d##t|I
z90~IE`Fjh+AFuZ}*vS5YX1{4e5-EJyzIHnLU)KLeH+nobHK6~GH2TmyQv7iWt!R5B
z{^&p_TD{wO;bi9jCe25o$7|6>a*5t^RoJ*fzlOe<!ba~o&%zCUmB;ii$=j%Ix9=Z$
zA633}w4KB`3=g><qjBTQctzVuoWt<C|F5lEzfayjq6|ReUS$B{`EyVG?m6QA+XlGH
zaxo0~c-{Z^o1O#e#V++CVjH{<&Cm4zaZ0(;s$6NissB%QoYDVxUu(NHw!#^OF(^dG
zSodgKjHmZ_uWn4FPewdbJMDU}>+!7ZB7SiXylG@H;@P()<ZM(eR;FzgFYKD9jun@<
zr&yD+B|$d0{#Tv<HAgt}FdqvL&jKtvq92&a7jf>9@*l6)|J;<fw5?w6|0sSfz2d(^
zT=Dw+MeVk@|KlQIF2OPkU6J3_hb!o-qFetzeV_gRzVj=%2MgJw$k-p+5Z(4*jQbC7
zqHn=Al;PF>!{zi>`w#D-SD^2@cf0HzNBe$=XTHwhzaIy27}e(LRhj>Hl%CoD|508z
zPCtnzby1u{-{77E$`)$>Emi)lWQ)*dsBzD$_4f&SX8qqH<@W>a19fi-ao?WwW&Qt)
z+6V4wLlP;R##s!_&JE|ufoW{~<lw>FaEZKvYq)`1h<)GN<XzmyLsY-8e=i=>>y10q
zX=^;CKSz^sfX3zOe|`zqYP>_Ocd0{MC-9ff-#unt{Y~i|!;p_J=l>Dh$-V3w$c+DM
zL!`RY%}6iQ|3~v3_5TC){|oiMHc-Ysa!&n!M*Z);8vjMD|D<lXemW9k^~=#nZ$kMB
z|JOFQhZ^l2b<h~kEW~(BjL&yZ7rOnXXXMK=y?2)1Xxqm2ZL{*f<pb^ixzX03{ZF=k
zr2S8Jx^J>DiZBiF+};`FtMfIA=_Qzr%=sM3gVZs3Z>K!y-aay}@tx!PJXFco^U2Dq
z_M{~jp^@Hn^JSZ&fvkHVKVD|9e{5e7*K4L5e?%SXkvL!uAR5tR-rh3ztiUR)!8&Zf
zCTu~n*Zil|=0B}B|A|bghkMC1cOTiTt()0j?V|KqtAE&K4q%lzIw#BlL|L!(9mA#H
zvDPjMuMFuW=0lFMp8u%x*Vl~;XO(?>_*cYhslu?IJcz?+eWsm49>+=4AnxrD_jpa;
zFedKXcW?E-@^40_{w<aNkaEw!25n;ggTvU$$f165AW!3L^taygJUMVFEA*3@`4^81
zLp=ZQ5_hGt?FxAfH*gEp(p1%T`$^I3&q!O(?2k;pkIepm;@p6Qd&>EJI$9pSV|_m!
zazDmXJV(v&ckEAN5BzcOg!<9%ghYY8?~U&_q6y6--!Z@Jop6f%wDrO};_^;tKl6_C
z{SNz|wfN_p;m*JNs&7$ien6M`0Z7yPPWX*A`r-eP$1n`}=(sG5Dr?Z`J@L0s?7vSR
zgFeqm&k{zdu-sS3Z#*VqGKw&;;+-&!9BeBLGst3;U?_|Wv&lJ_hxw?S`F2=9F2WKl
zL$z`Bs!8T0(Cdw}*KKAeuF(hNZmKdjpv%~KuQ7D@m3{OV{omFL*2uc2-FNMvucPle
zuAP)C4X6`(8vN(&{(Ij(O~$<#YL3YNs4I~F^ZX~{{5ND~{pn2Qyf8Ll3$~#Q<=BG?
zBqz!LQ-wcE_()9=KGHX(|EtpfNAkaWPPwO*Y%7&E$&Lx~zh~_i#z7oLJVWXzS#?Cd
zrXQ!DL=EcX^?G?X@kG7(;$`2UDY{*2MjK-LH%X>&8fTHQ7fN5Bj+2hL`*8`o+v5J;
zxf%Q4HSQa@g`vCV!IO7!AA`@$fhS}7e@s5bbA*2>&7gIqv3)WhBQOfD*8hy5$3A~I
z3hCp~H=t}#CiEhXe>ZojpK)rq?}N6nzE7U-gHGQnNoLMhTkrd>_I+3SzIe61a_6$I
zgo(nPj3SJZXQq)eP>h=C>=(0?dl!5^B(D2@XrwnG?%6X?_I4<7?_kPV1ol{%9d+p@
zJt8;EqtC}szH~}1!V)Ztew*Gl|IPZpSO<TgEK_E#;Qwm<&nkNTa?eKQ{GSpw`vGa|
zg*1i)YIf=a?$iGz>yN1a)MasB+{Rx0#1`}Ogs}$eumPK}1=~=Dq&l*zLH$={{Cl|i
zeXRN&nf-tA)bE$oe>jC!{POvLaZkW<;p{;L_M>CC{%;oB1HEUouyL4v6n$IOtsB*`
zdFomm=QnUXH=HB~muZg__!cvzw^FtZehD<Ae1$&aI(g!uF}D5c0e(Y=?f<{;E1`Ut
zx;5yxRhZMjzk09s!wzww*B{b0z-jtfG_f5tvK=(+)PBHuey#W32>s+GbkMJm*U(OX
zW&gXueG9*0|I;6@MP{91sqa6@_n+$ff9U&vpl#zfcX1yN@fc6>9O0Jl|Dp7WxVKYg
z{n=LO6KUx<bN=7C|BC(3eYwIIhI~|>u+JY^MOI@J{j&*wmNSN|E0BKkq~CGUFWJc5
zME<~-+V@u*TbnyF96a{*@cG8Chwm*OZGXjole2&6pXK~y<s0G0GxNf+x!(*&*M2kn
zXz^HMbZ>_vOU-+lYj2wsAruOKaQ=H?JXzV29e#S%eu(W~3CH(JJ6Xk{dH8$UmhXmB
z``$Hn@LQoRYofJ<?}qfW{7`%0_e0H@-w*ZHsnreqckCB`QyKMpq3PLsp)so@oEXb?
z8N*%wrt{TezwoQ!r2Q3tw)59Q^^vdo-_6UP;~cc%|0d^KD4+gkIXg#>3cF@~BkaET
zjZl8PAQbuiw<nDV-zoT7s5thuu&<yne0S_S;WKsN6=l&-$O+$DKPG(s;ctY4+kYb*
z%5x6d(yxRceE5|x&2MMmhjZTuKbmUY*Fxig;#|ycKzg)DGwjHU>r4G-(qIYy*_eYN
zV@~tPxc+ZG8TbEQKrX@(EJHl|a|O8yYp@O-$Mm0$WQPs(p1t}{z4{3BEy(O&won^x
zskYn!?f)qSVH^K4lw%L7ms>}$v>;T_>pwI$yj_2rei%(h?EQAY9&lu0VZmQ+|4%iS
z+CA-L#|7)(>6!NbFl8`z&pGyiGwcIMqYo3^a}Xy{g9Mt<h9qj{8lT@`{ej=tBY}oV
z)*pCA6WP4d`h)e>AFwsFqHV3Y|L7QJ{lRGK58T%^lYIg`d7jU|_mQ#{eW+ZiAMH8i
zXS2d-@+_)%N+<l!)9d%@Lv#1jFQI84`vDq|z!iSia09n+8+UOZ5AhhW?u&CaqHXgj
z_j7~+WkH+rw?X|cz1QI{)&B+B|I)-Ra{_j=`wVk`K1N^^#-I?%onKo2b4;Da-Mdfx
z$iCiRwExw4r|xL~zEJKx(Eh!t{i|)z>3Zxxju*~EOhyqp)b;I$^r`7_&RyL5eg?f5
zne*?au)!g&0ZRAhhDv#?#I@O|KBGTC&O_B@b<YdshB|sa_X0GoXSY`$$F)BV^cvyR
zqHdP<KN9o?G#=CbKNP|uzgdE1Sb<gev{s(+-O7+WBK{59|5e)m>&1Vw_-_|~vX8r2
zd|RG~KU&X;zqq%HcSpPU>kkZU(FWJAj_vHh68&u9uS0rNZrDJUC$hs9a`zVY`TV@F
zjb4T!d-V*>)Bk>?4(Hy33hYPb4Bi3jqUdo?xbC&)WYEix=Z3yn^7&F_v_3*?OB{{<
z6}jPfl%Y7BBx{gBGwP(jm<DUQ<bxJz?*;qKfV9Ux)bQjNcD2>)Ol&+YRqRUaU7g%*
zWcy~eqvg>qRS=S7*WzfGDq#07(Do?^y-x~4JpXl*=d>Y-6spZdj&;CkdR)6!H^n-8
z`gz1N0UF1uqeiN$)LZ@hF5wEU;mh{>8{D^W8=3t#*$eN|@8cmJqo(Q?=lk{Qe_v3i
zURSrGk=}GhJ3>3?sr#NI{3~rf3`0Ifz*_9kb=Cim9%($b>3R?O|Bw0q$!2wV%WCWY
z^!Ho0>ub^5*BgHzJD*4sGtILT&KMM8JUW(Ye=Ib2n%;9l9sZGV0D2MnYPCQ3^^&P}
z`~L8ofuV<ap_nYeY|Oztl$FZ+v41%(g!$Y9=dC3q2kpnOh^!RG5;9}|n{A(Hdi8bV
z7nkkzMz8mc>n^hM)7K!{?xPLAfsD3_b^Ka;^%KZV=uoF`A-AEO-o046V0l(3<1WW5
zyFlE7zL$NVW~25g>QIkgIsYb66889A1@_}04&x|}Be~Z28oGzG&m%QX*^sAPC}961
z|I+z4?t8sIAH7q#a8ejGNT5>RqM59kuK$5HdJ-wbvjO560SWDnxNrBX{dtGRX#bA(
zjpg-di$Cq&^Vs#YM7e2h0sRuL;2H+keI?{dO9N}Hz2*L^?0>i1a~pSY9}n>uPw^bd
z&DvkveP^=gqCBk}@4YFHU(o(K!~WC$U$FoAUphZh_E_5&2EEr~^FYb+F{NP`nU4_|
zg)u0^cud4(3|XsCMDExw-Wx-hMvrq`ipi3AUA>WBWbcLjSz$K!pgjcVkn=DfyRP{@
zkFA+MRvONV#|(Z8un5&xOWFTQ!xDP^Gi$RSm4;>X6=)i(T{ZH((BPf~R`F{+@m^R%
zu0uz|dtn2)3GFS``hRF$I<|0cLz>=)6nas!_`OiO`aN?L-V60)V!N={dPbM^?<n(|
za_m6`_TwN9<0z7gt$$x`{X5ySUbvfuyWO4*WFL3)1?%5&>bmvsXrs5I!#$lF-_N9-
z$^p-5QKz8#!TaI3aMSvmRgdDjeD8(w+s0z~H`1GsKm%HDmV_GqnfV3AYZCMh>*Cv=
zl!Rt_kG1pNE!L9LljvLO-4=RB_oPs>Q2ddJ|4QMa0k6;hIxbE5-D#Y~dGzBFt{|x{
z>_V(#dXSRmGwpwRADUl$sr|3MZ0pj$MF+i8+q3mfX}Bhw8@Pqr=oqj@-rD`U^mrz3
z_pI!2pZ*Yi+u8omyH)#fq<tm$Rh}pfPs!(~u2SILR!*a;LE3R|{YUoH<jzN<u;Ljs
ziN*49X|^#Xc^9#KOrOpPan8>O_w8mg-c^#TUS{*;9(Zc)D>)t$F&Raeh8ZYE-2bP9
ze6|12Z2BC`!+-1kKXL!T`F_6ui?9UC5ZC^#Apd{w|1<mjFsLmtILtc!b?=2$!d!!O
z*nmygf^8^6TKN_C1sOW}Uf9E(+5cy)Z@5x9KvwX3wg1n4`s@9F&gX{R_jC1&^#32}
z|IhUwpr%UyQ(czX|E5cwLT^B$I;aWd!Z?V-IEv#qi5eu3+-UwkGUxl<)Mq}Y|IenL
zCi`CK{|l={9d`<UvHz|5ccJ<hasQiU;k3oDkV3~4^{+baG`(k|`ggtd2mL(y^u^Qa
z=H8|1-<8@Q{NnyMm&hxqcJDRv2C7!SY`fH}CvS1zM&oGlx~!csNj%4yPhH^KsXt=+
zyX*RWJVf~o_9k*y8T;LM>$~aC5$;Ia7>0a|z~H9bFp7+O;*B8-F&+~!8AX_e87Rh3
zTW%;JXQM`+|MmQfTIGkl*?>lR6XN{+IqsW>`B;EOSb}9pu9yGy6T0;wdyt~{7AX6t
z%l~M;F8||{`&w7a|LUrCvICv!pdGzAY~aS;jpalec-)(Qg>cjIUxzw)4SgMY)V<w^
zV|%F<w*Ok?0J*`n>b1t(*`dnyUn}=%Px9OKlszS`*LcKMa8F*>9!`+4EQou8ZE?>w
zl%X7ZP=P_;dOsQ0;T<I7-av=Rqd1O}s6hhFXnmwjL5}j@B*_#y?wMyWkP}YRd&~ps
z###D#^c~VRL2s3|3Tmck|DaC1<6b!ldYoI_xKuf%j4He68x5B}e4Bp1zl8KQHiI^E
z?hEZP=o@t@$1?l>-}7yr`9?_7`_Sxq3r->KmDo1&<=Ez{{r|5C?*?w+Htym+9%AT<
zJ)p?g|9?sjp0qbY{^;<W9{#m-h&c8+jLgRfjKZt)zsAt(+r<rq^znGLZ*fDFxZ?Hv
z_Yd_mmg-lm)vrL~5&aBY@vMn{HyK5kh8ZYE36lCCT`%-A$R7Qa%=~}-<uo$$|L^IS
z;FNn>AL*A|)i1fJpF(zC<afp%{;~bmuKdDmVaBxq?fN)reX^dZ!bbV3oG{0=zOnpA
z8vmZ;SyQYL;8*EcyY|SZ_Na(+=jZcVfT}8e_qF1MMchl!xLCaOg%az<Qy;%|uRYiJ
zufQs-LB|fy=f95LbI9}ey%{#pH=*yD=Zi<LXQr@)-!_z?9E14<VGlVlhW&&bO5}(A
z<Ut(9Q5?rf)I_rM>+t$~1$qM-5!X5xk9wW{67FwCd5QdOU&kapInMX@eYz+4{`chn
zXR%%HI=v6gANu|aeSfr~ZKk*hw}b5TyUM-#g2LHlY&@R-ahhK3Op+@8XX*8a*njrf
zBaq&Yrf2Lw)`>LiWdFe>eyzvk{dw}$?A*|?I44};cMX~SLpJIcm<Mo!`xesE^$U<%
zC=a6ffH^`(%ndrB3_{ywc^4gsXLKc>=wCdOhc61}oZsE_J2K7PN8T3ZUEIe*?A$gs
zJSLxFm9$$kLH?HS;vQJ9?U#@AAJBy7?hF5~yo6yGT428^as)<Uusnn@WTyQ=zTR(B
zTYEfrrFWl9X6D~p-&aJhUZ^~5H~*1dul-ZEnf-)bjHa2&7R0#+u?<whZ#L#&9_C{K
z7GX%6C%rB=43ybZlY4NK{vo*ntFQ*^umPJ8=XY))+qL=P`Oe$uWhh6SL*BDf`Zy+i
z%oJ{^^g-6<8H08<O1*lqPJP%AZP18k|HQT86a6;Ujj!xid$=pG9|zI(fqmhRvHg?X
z?46nQb!?xn*4M?e|Kb_J@r>M4S?YCdjJ6hI-wnpV$;t!T7!}4RCd<Fte}{!x?Vc*v
zj?(Kdc&``i{PdH^*dI1}&#m6`vUlg#dPIFrHY2tx+Q=j_=LhNQXZ9uJK8?(|gmbl1
z`1i5_oaJ{O{kViHxP}`Ts>utt$blJo;WjzwOvk(AeLTctRE~50-qe?Eh>U%YU-eb*
zug&qCUjNAZ-_yst>wlQ6PCw(j-t_(h-k<+}cKw0>qEWa_WcohapWkNIKaQ6+=_4=-
zW02f#{KbFRUEn_;Q@(NUaKD>n{{b{>54CLe?cLW(wk`D^%=PX4znx^w0r}&IG<rh*
zK;p8rc~cs_`f_}#&@;wkBKnG~hb4<J4Kt9G-@BH}pZfDXtL0C9`d*~bhs^leKK0*D
z^&fj*+h+NDyZlXdau*Au1g-MvY_eV6Dc_^4r+3I7ZSrS&a-sC4j@_(oZSf3sb03-R
zGXHOFc9`eh`B;EOs2(T(jkV_>Ju`o~)SO=W3N$H8V%~2c<NW0sbzt0syslUL$wY(r
z%NLFEM^nrvMTKFN->tzqY`_-mUSobh;kb~Lr@HoOe~~@%R*L?5e0MAR(?abpyx#xu
zk@god=da%}hyR+kg?evCmHH3cgk6Sm44t*F13ADpP(cnBTO(k8!2akbEgd8eBQxL6
zw<(`psEua~m)~)mMBi$0K<`!WhF9x>YFta8(i!5L$Y!)5iE3lwRaMR+r`L}*{*BZ0
zvuJ96J2avp&-gda^XtbYT)~&e->z}rz%8^ZFE@VEd$;+<*R+ASH*?n*|F0eWrTGQZ
z%`Y&v-)K%j(}Z`}NB-jet-tj9?cDZQ30>FO|Jb;DkUGHrhcq(#Ukjr}n5WQ+w&CU$
z;MMthned%+Gqh-2@F~KM_>7ciUh<Apj1Nq5-r0eiaANehaJ;~HRoUBNr+Wrd^t00R
zRcrON4`ZD6{^GYC_6>-8{3mXi6X#oXS8<E)KG*y9S-+27ekoLrHs<^4c_TykZ0)2_
zUF!SKdOI9&fBk0Pf3<t!cVMe-`pEZ31HZUe!H;f^3O~FvIvmOR1{?E;@RN=2gdcAm
z8;(628;)lEM)>^B)bPELzZ(v$pAz=t2hT=^?>`zD4(%+mc5SM;0Bqm~+0Y*SR`}Mw
z-wx%6emm?OC<?pA{f)4D>)!|o&pqTlzP<Q&!*{0qx^vI0ha3B~@ZCqF!e`^Yo{85H
z*T1*;8{zW<W5Pjy@}Xzn4BvlN7=Ezb{tUNr!sowc{A;my^*<ezPQR5ezazQ#t09q<
zt1Y3t+o3OKpU{4Nh7Q+`3G*M2=WrfD7(1$dFFQrgm7ach$avb~zm}akpuB5||F`l^
zosv|hX4YeKpAbfy^0xh=@(!KK-OPS0ca(?6*nY{Ki~dngmHT!-*PhDyYG|O}^Nbd<
z;=mt<eJ9Pkd8j>asOt>&j&k4FuZHjBhwx3!v%iBb_xFK!*@4qp;UCkJ>%SWQguH@(
zj+?jz%MQyX3+G?cyXXTa-wwaSJ-Eoe3={q^^xK1N$Td@9%l;jnKv$})zsz`o^#gxS
z|K2F~tgwbLWQQ-fzv}nNM_&yE<WGF35#-(}zZSkm)_O*2n!5h1azI`3b$<QUaU{=u
zHFVF>*BtJ>`Tg-5q3mDc-=UEEKauZ{U;Robn~E{`5BOcw@_(0XBL91`T{-vn$j*(w
zo&ER87Pbgmrj@l8Ov&Cy_TCHOIQgH@fUdE>ll>3Lo~5I*-x(Il{;qc@A%EAm{2$3#
ziozY_M)Dt#zeA39Pbs;K{M+RJM9y{Xf5nH~jy)**JLFdWOUd8#ZvQjcQ}CPFm+1YY
zen0yi?!ML1ihS0~R(!`hl&#gj+id)GV^;Xq%?a7%Yrhh9?r<I<cJteFA}j1|$O_*!
z9{nAWsbGKDSCtk12Vs3i?jOhs2OfMSe9u7>pR;!ybkCuMZ^&a=;RpKNhc9P^BkJ8B
zs(*i^{5`t%4g05#%Kou?euCqhM`fSrdL#VwLu+w%PR%|!f!%Q`J0$;~wX2^`jLfcG
zI4V1yPt-C&JW?~WFP6;A_WEUYy6y=y#V)*SXYnuauW-Tjw~#!cZQ#G`gsP+LPh8V9
zExVcLKjqh*r;NHPJ>0Vg%?o?aTu`^(`9^lCKwX=q-X5^W6`J+yTg(;hu>P~ty=_nI
zLG!@6&{A#3GtSW?YiHUw@}fB=t|iEZb9v#X{44YF!g0?yG0VJyae3iq+$YJZ(e}IX
z4nOky_sGTg@y)#Olj*s}$+N?;t9jw?^V^G)-g(+@#;M@yHu`7&yMK#+hyDKZZ2$io
z_`dXikh@7<`~%m%gCF3hsEfGv=hE4)za@R6Klg8DUu>S1-9MhZMUKE4ESZ*lu_Efz
zvoDUCp51>X%HPlK-!MM=;-$Zy-9L>!k$-+k_Qf^tXZMeJKl|d0C`+?1p8mb;{y8P|
zS=s%?zn6WnhI<eB^lxVOZ}Xhl<YaOfd459n#fiU_-Jihh;_Uvzld><Ke>c0oaANkw
z5fiif%m0?&|1H<~r}!t_^VIWi6=z@ERGj_S#o<5rJ}ag1J-9gjxBd2avg`O2`Ck1e
zz3*}N7ynlFALnuxzbpL+Z^W$Zi!0vC?)Nd%rN#mu6lMQ2*L#$yxxP_X>2GHLsW1%1
zl~wUK4WjH6`3KUC=7{s2L)m6}Q?K*vj{J@6hN}GV54h{k{bqKK>t|j6Yvd)8S*&ad
zeK{(##$+Go_f6rB!%vpyhaWG@H$Nml{B`ahEuE0vM7E#}N&Ghc1efsn)Zfbfr{p^S
z^&jA~y!`Opjrn2U>in>k`#W>TXRq-7duRS`_WwfuJ#6E@f&4=-b(IC*WrydMtwfOj
ztt0w0<c3`S<s;<(QT7|VJuK`lv1eRSUf6pwFMK;}3Bz2kn5Yd-j=~5el_dwx{!#ee
zy#F(NK7VT{<kxQw+xH(14@m{t{?o(5M1J4n?tiKbZ{yGAI62-|DBo!p%!vAJy7osg
zy##+7e-||s!$R%tKME`9^;e?3(|vybzj61gxBq}_Df^?4)CNhiqx3)5Zt?Da7_V!S
z{2yd|M)y%?WG=I(RM-}JAAb_kPq&A@=YKzx4SzR#N5Q+<pJu(A{jI!rv&+Z6o4s?~
zyV<)&zMH*!^t;)6ro5ZIcj~*@-=6Sp_ID<|n_V&M-Rym(?`D5@`n%bm&6FP1jR)?%
z6}}ap*YDlNhVV($Z<rf&$NY<#-pl)b7X7?ie3#FoKF#$gFJ*@xkw1z09(#tGuTizw
zzIMys3Nd~)weM!vHoTi{&y7&&o*3TG;=K#>-`hRE$h?8*HrCf_en*@69e>K*G2HwQ
z#Ph?G3d`hN-=<jKVai+KpSqqlccg!by(RCM+i_R;z2*{Jh;_a4^`v}R_M6!kx5;w{
z<@4Mr*%$B2GppQp$9-qi30LjUfFHEj8&dq5gxQKt{C|ACdstl6*|7hx025+}!2}f*
zn~-2c#RL@_tTfWMu|~xjYgDZGHnwStZ}DwfNyUa_7#LuH;d~lkU<PIe1{lt#If1dJ
zH8$9wL4z$;L{!w6#EKOw>F=J{_PT!G^<Cc|_w}sPUVHDg_BuanZGRegkZ(Y9+PBo7
zk>h@M8nEN^8U5eE@38;D?=~&+N914q=1JB8{PG;(j^Co*_nWE6kAHBB`p9o44IKM;
z7rz^M>z}h9Ne*Q!7b-){=lJJ#M)!z>Z{nMXKsP3aIk-j2H+nk06y|y2SF;uzDsFJW
zC}@1EB%8i~Hatjjj;*5q6)d_L^rEI#q5p*aA-;e57KNO*T@<=fio;flf%CdchPJ5E
z`BhSuwvxW(O8%+4QYzD_uh_p(9l26!bX<%1IAdpH;!3Hfuh+o!ja+Zy`r~)ww-z1K
zeXGPmU#SH-8oWZReL=i0>%Yj+Q1tV&vGd(@7&#+6mHr~n;dI~^jebD<O7sg?O5Zuc
zJWpTTuu=wNu2VxBSE1j!N=9IG<Hh2NUMZu@Pk87Pc|#+WZ-(z8?qV4S4euI3dNo@3
ze>HUjQ^y)zM2O@WL--GHh!o6*h(0=6DeA1N;3i0dhu~*W=3FJ^yModG3YN;f!BPd)
zm}~Y0OKn)N)W!SXKUnIK4WaC*2uBx=wyz1<974P32ocLN?c^fuWB~0VZKI8Fc4QmZ
zwIdzaJCL0bw5$3M-rs*dGyei!4w2q~5b4X}eB}(<X|5kk3qg01cAD@bTtA9*%|<IX
zZXWPXMk`-p2>WwF*vA?oIopE8@+|*R?mHyU@jh0}w!7)$VE!8v9So8g+T3*kv@drD
z$<@dWuo)PFYHkF&mzsWIE$1u#xrmuzm}dB0)_2aXkoPcu1n##<|3>Cr3Yd3k=f6(4
z4IDWnt;d685%ZUjBSG>{q>l9z+Xm`m;wrc=NJ5Zn;Bx4R<=I|DM`R;&J~1nVX+RBI
zM`;Hb!|q!l?;s;-XZ{NwgWo_2yhgZ>A(*h&09_1?mv__7+@NyiZ%~?L%E8hN>O#Jg
z_qnsH7zgp)lio#80D36Vg~)fgws@9uU<i>4*A1#NX^Sc|-k{2lUMw}7`%#;*MO7UQ
zmFmgOsvdj8^!2I^Y0O+9k8tf%(71fDWFgDob1?I6R%nO0pD@-#_5vzm=o(QD!e}=z
zd*OL_5nh7hZ~{)kDfkP#4e!DSK$l;07CwV7;45I*pwTuSk|3mc@gZfIen_<}JfvC|
zGsJp2gR^fBh`suNwD}I8AACR@F<GpSWHEo5C0&tOjQz9F|I9)^EsOmnS<<)b5c3;{
zm|s4`{LUffcMi#L+94T%(Ue2t0(Z<I%HJXJZagGD7>hV0<4{3co<O_&1~QpCj%5|<
zGW4^kD_h97F6e>4O*B>Di+)zTLyxN8$59q<XDinAxWO@)hFLfbi?9qKH_-2Z4A4O}
z*kA~zK$Ce`va=717II97CD(9RbbW^<5Ava4_OR%o5Q>%$OY!_+F`PauCCJj5!%}8D
zEalF_Qh}`W9hNF&HTIhHLsA=iNb155iIMlLUpOQU(71GnvH2ls!rVN6NX(Ik#1ef-
zTA(%YkXVs6uCpWCQV&V{zC+?5t`1};>Dq;KV(&)w;NKg4i2WC&KXQQU2a!Xh^Ki;<
zWMnEsMwc_iwU8n1r3~>*XNniI5BIUj3>lx!pdcKO?8F12-E}~6b{>%2qywT$J|KBZ
z&q#jO0meVfYhW)-IlzC=pOa$38BEVfNgwC-L_Q~F<qxa!m;+LgeLyNF4zOSRfK)FW
zpd1{K+Q}4EXE`9oeUGX7a|fg$`7zZvc|c5ipOdEShgI{`0WmZFuoxawE%A@2){p~Y
z-FSfa=ljz?rrL~;sdm?6%8_wEI#Pcl)8y9-gx$zD1G`{1?18<I1=(N(6WCw~cE^fl
zAK1VJN8l(-fiYv1e1Y726ZIH80FT3u;3uH{_&Le>$#3N#W)0*$^IOUH{#Lp$kHYit
z5}bnf;S5~0m9YhEfr?)qlpBz@K>2+K<y**m;b-ty_#1Q{KPc_*9u&u44$9xL|7M$L
zeha10^87)mL^ePx9KUmynvh-U_*eHZ?zmTV;x+=OVHv_G7bQ>+9njx(PzLTi$XYJ*
zDeyA90)xLeC__O9DG)0p+q6Qo$Q%c{@GknlkVnTq9|{&$h#m@;82>^sW&@P4hF%I~
zbkxhCg7Ir5RHX#)-wXbOfw>myG8z9uJ>~{z#Lomxn47`uLf0EwjxqiP>-5)qm+Y9^
zpnZuo3h2Px3I07xPUyzm1HCThSD}9w>)SAx!u%=>WA^W_{LlVO%pUOK=YuiK<M1kJ
z@jAQ-6*nW{J@^nlfluK}I1iWJLY)J*!2x&<w2%iSPyxf>hS%XucnjWv&*92jX?q|F
z?t(lhgbr}S3-AtH6wka0JP1F9U%_+088DhWD1;WU!wc{-yb7<wJMbZV0-wT{a2}d&
z<6D3x-D-Ic`5}A?8s^ZmL3=Ea{-6K63jWyxd6@H|U@1`aP>8t*it#f*2@P*4>u+TY
zsLJ_YO9ktUl~9$khWuYc{?j*xy394?{~Gdt4f)SGUHF-x33D@;U7RBZEyq}k1nV@q
z*I>uo2JK6%k3t9LPUynV3Eh}`px4FvCiL%OeG>*#Sl@(U%p)+G$@(a`F?+y^pAW_`
zkNfSJLE?ikeUOZs=vQV0q0bk@dPb0VU~qqs48bsrKri$`KMX(@VVuy7xd+;lg2VwG
z&<R!({kgqCVuv;`W4Gw&?-}UtAsbWa5ALHsm`;BXS%=-Ylm22d{Y7L|B>hR;Yw)i{
zmSHasroS0Ve-l|8M|p%2D1`#hLm?DFF6bZ+@<9XHpoJW`_M3dua2xyz4!|MMLLT(P
zFpR;g@H%`7U&48a+Af;!z=QBGJOPDJ0!{D&yaI2*=WrfW0^c~Sg3I7ah=Om!ci=8~
z0-l7Y;1Fmb4@#f{YM>sPpc{S<FT?BbCcFjjz=!Y&d<y5`BKiz#;8M5}G%m*fpgqR?
zALLH6KM3+L=R?5~V?8LuTm;4V8K4AnDU`XG|APwV0V<&?PNW*HA?_%+9&Uz2`Y!uu
zi_#!^Gw<^wB}RWo8U3Pc&Qj36dy(W61WE46t3+3IrQ}_~dIi@Mym*c1WvvvpbKYE6
zm=w!(Vi>$ZN=C1jQg00ZhhT2{8P36D+&9O(u=W@1tLHi+{iw0;;!a;m`*ZHaxZ8vK
z8OG>-ooa0^I%JF?=jb5L(K#H)Z;pK$zw=AR$$#)m*5Ur>zlJ&dKlvs0Gk36N^<tpB
z=C_|@FAJUOYxoY{@Q25Ifp-Jt&wl$T;U9<l0_Cs%@ONYWj4+tr_uFs9%zVFpZ}&%j
zJNymi27~19$bY~;;mj|BB#<;)0cRc!k~0qk$;Fs2gDb&DGR}PwBrhQ2uwMf#b!g@q
zgJfm2X#NMf!GGPG$e?R6-w`A+NKMGSD(~vwTu3V|={WX)l5gW~45Qr&C6BvEQ##6q
zpVMX?QBrge$=YC33~k6t<}jGQFs9PxA)EHl2Kw#HCzm6~?pD$~s^l)>GY#UOOdHty
zb$BE8v0U67%=H#4d6F<z4Sv|4!ffGME%)wu7fD#{<M=W6;J>H#AP9AhmFt&q`z3Qe
zR|iRx4l`@k{lq&=|7@&_@n1}cTsWVRHC}BoW55rBM9W;zueitPS*~F{*qF{9t-Z)h
z){TkVZ}#7B>`v~9e-jgT7rs~9zx;jJuf~5p{;aEv-ACL|KSNk(K6;VNt;WrZ{}TQ`
zVZGur{LkUf`b;x4O(UlW7ett7=EpY@=d+j-zjo7pm+-jz-|b8z&lF7f@QcL1KeRrB
zXT`4NTEG7sG~w19O1uGIKjYl3#D(2|X0^XeoH>g<o@>hZ2KnxrxL+=5&~S7G>55L_
zLEM<fZ^rKbhD{H!jyg#_>A&_Q*G^z2jQ@VdnH9Vz*9iCiIoCOf!^L%&{bBs??$3Lp
zgKHk;`WFdfB)mT^z6Wg;b_05xm^~t-J35rCCpGOyZvwmaA?DtLD8IxvKp8XkG57yV
z+GyT)WIN$3tZChfKhM(19G2gH=5hc1&>2`uIG)|Pp6hfxR~l)+Gx+n=|Bn8;z;5Ol
z%LtF(9N)A54E#9;oO28$=NL3zs80->Nx6;ZJ^uAIM$URX!*ic`i1PnS*bLVZmN=Vy
zgyo$t<R9zj7p}K)k6Upg{f*3r`San-SA?Zr^1r`7P0L7s|NdTv+ZX)nus22e)#1^b
zi@yFs+WG4-|F`4+hPv@T^E`17pY{sk-p)Np10(frJ@uT?UnZYHSDA3l(WDdc8W(UI
zWDblvtbT%LKE?HvZSAQ*SteXIdE@`b&NsIBpM`KdPcClS>(EJAtK@pz3+~{Wjl%aO
z@*MUv*U|UkK4)$utqZt+2KNsasi0o0K0&(2VxJT#A?(;o=r8`9vp%WkwD>nA5l;+y
z0r+Wizs`rTmE0HmMc47%yj#<0(%=N~?*BSoe>wC2m)XcOoFOmQ6RwAER2zhD;eA(2
zuUySJsO)jV+>}cEd%52Gb@(N|r~TB0{=CaT`tG<|v?!{KANhu4`s>KS$glH@J@#{F
zBQLZQS-4-wSJLl7{_);=zneb~`5ydvwBWB#|HAY0JuO{9Ja<IO#w!_vU=E0&jl&!R
zu@DFGkN{__N)r8YbEG68`;LkP6Mg{MKf~Dn7<;sius4f)3?uz#JB)&7f;nvP5ylJS
zgdZbp2-k)}GVFyg2!}K4pA9@CcL&a%N)DXe-mc~z&kUR!d?3(?UmpyC3oiWMlXW@m
z4DEk9d5_fXC+`E$hoi4m5Y7B&JoEpF!BRvYteAN=1O2cP%%yx!Wkb~UKH7g|WisbX
z@8KIjAEpLbi@h#`?*V<6dSnCkM%+y)d<VM%>Hh}O|3$X|T8^O0GQ)lt`qVc1)^=pu
zGWx0XxgF3k$vii7#i6?b-OvNQyV0M6e(?Vbb8tWUH82b#FuEK49B_jN{{7#sf9v7-
zNP{DA6pq0JOu`gQ!wk&A*$qnOkqdAd&cPxyF5V`l<=bQlb9tGPWuyak!Y;ogkKJjf
zWEV0SM*D6P7wpEo2T~vvb`y3lav$u$oC2w^57HnV_Cp3_LKb9$4)kDva;OG>oohs<
z5!YGF=|~fH3)mn4doYwg<NvqTw^Rt`xl>Up6d4BL5CM^3V-9H#^8*g#5pY5cW4SXg
z?NCFQXRH3EX6s&6b5(y=b8COEW~+wPY?(>T<+iA~mH8@y`$fW;(E)X4dyz76&D+Fx
zW<#wy^Y+_H`>v8S>?@gzb03|SHAsrP=1L&RHP^z8a5HR&Z^L)t9(Vv|HP5QE^d0Bc
zKc%#ql`?C-M9r>T$Nkr+xht+w4-@9rK<;<zx73+!-%z(>zSD1CeW|(&8O^iB09`cA
zgAj{34ro?1Prw1tLIIRO71RUOndT=DPuK)V$NV_=O2nK5RF4{}1`XAa<~gW?C!qqK
z2C4zgNzkwco}Eg&yodfT?NM$fIvJ3MIUfpAE)u<k@r;e}jBkY$BMr2-CCE>?{$qF>
zDniyvCHuCk&awUuHBbw6leB?QpRisUkd3VCo1iI+wmxz_bv1Jo(2&R+1(+~5!baY~
z-=8a|KT|e&kyJ)g#-SR&DyWTQjsx`Pc*cc`qzGBKbdebP(AC0T>bOWsT)Z3Zn$x@+
zQj5QCl6S;EADK7pKRdyHh8ktlM;QelW-p9W7mPt4=3W@UuOEgm55kCvG7kQK2e$Vy
zH<G~|3DOzP90~NyP)C4&ZpBQWvQ^I<3;zE9!&%FjTfy%CKio~^+Y6-MEASq?2|>8^
zBENtr%&VXYvj&ENCQWl5@_?pL(}AShp(#dQhP)a=_@+N%F6?FIGjdnZwgpqJU^CnT
z-+_DK?2Y_G7MTh^g$kIv`{}?2<m`P<2U?N*b0@G3*#S<t>vrZo?qjUChjHuAxo$sY
z@;Szr#f(9ld9HSn+|9jSWSwS$`@hb7*I)UrJ?O1;^DH*eEKv^i%q3co^N<ury#V*~
zyi<(PE~1PVGOl?NdnEP>WD{<eZl`YIIvs9b68_o*{INTs3A>kfK89ZkY{z^fl$$S-
zPq;S1ABH$H@S7lgdU==GlRE-uAH6bg?k`&d`Ttko?D|gwXYYJ1aPH0z;dg-pgc*i2
z&wdnm=J8JhU6|+n_V2zGc&08oaCWdLaQ2Scz`27hfpc}3w<XE!16_f0S09$Sg1h7c
z!hYlrzx}sz=Itc;1T)8ZF%B#YJhLG`P;=x8$)34Fw5*NgK<>gC(Lo;ThxzBOp#8f<
z^rx?s!uhMDXk(-lCq#+?b4l7-DV@4f%5+yrc{1mn@4Q4RS+lJ|R%5S0)@Jb!KBN(Q
zJ+dL~D*k(WiI|`Xn!$YJ60txFw1U-jvDm=wxLDdIIG=x!{|_#+{~y_jy=$8N|ImGm
z{r}MGV*fw%r?LMZ2KTc6ABGcH|Ao;o_Wy&Me&GL2KmIe);_vV!EQ3&7H^6KV-#hl#
zU=8+L;a>PT9EKX8cxgU|O9*=@-+0nW>Sx%2Ibk#NDbNZpf$Lh<Y~Uq04kuvib<~AW
z1NATte}Fkyge6!8_rhxuPXB8g+zW}scMm)QKY;^~1J%$5BXAsEhj-x<_y??xrhLNx
z!U5O}nh56qKpVsSALJ%5{|9-P^PvDgJrt%f{|CjH%>O|N=29rbuN*4qFIC3U|B0vn
zLtS1&onDKqOJ}T>@h|-+WFvM{JmWORdCf@k1bTW?=s!`<wj!-d^nbv<K>r8Yr>Wzp
zH~o9iJE1E9-D$Y+ACE-p#+}rW$zPu>I*>};i2XnR7s;Y--9`TgF6{ryLO*&R`hq-z
z{|r?P@9zlb8w=pu;v4M%C*Nro^jO&EfTEiv<6<%Iy;xdLUo0(rleYPb#R_eFTlSQd
z;v(EAgA@<aO?V&D>k5@P;vU1^hrM??l)Vk1(vN)zIf(xVav1w|!rcu&gnsZsIauHX
zWS<VDG7FWQWzHx89p+rfC(gW$S5US>r4AZc<F98u-;^1Oj%BDcr-d?}43%;{XLn?W
zN~JNBH5%3=u~#8$vqGf?`8MCh2JT~k<M0-&<lBse@4#8?U&3n45fBIW!=rEj?t!1e
z)1ZN3sDf5_3mTyx+Q0`V;T;IrLSGB6g6~5Gyze*jeW$`+*av3`dlT;8hPz=C`~ZFo
zhv1h`124cau)}LG1A}k{<!2RqggF}dDe^101@oP74?F}<z^~vi<UteIpdbDSQ}8|n
z@JwqU0ye>|@GbZO{EoWdy&&3#VAg|R-9^*`@Br}1X>y?f<_O17Qxi_u>)=~(KkS8P
zARp>Mh~pBt24dk3`0+~0EgT^pmfJOvA^!Seo!o-F9`{&?^M}FjFSI=`Z=)YiTjN3Q
zq<z{#xHBKGlZ%jl;nVsPVRQV~5dJ>+ig;IHuY&6d69R7&?qm2fZm3~s!U>m$+=Kil
z;qA!s&Eyqf-obqfG{U10&h?++K96}lG{XmQIUIu)_!zE&<6wus!!n>MpeaPAAOiul
z1x*s{g4^H<_=31UBTOvj)$lI%SFx8sD%Who?GpF^_bJ@U;YYYTkv~Sx-F+#2XZT{B
z%#Fgwn9cADc=l2`^L!|C2Jj5#2cRAP0;Ktw4eNxZFb%)w(X1gpR9iHsFr%iW*@$@s
zem_I5Mjk-kiu@*YU>*ikSr~V$<$nL-nK0%|Ft3Dn;j^`J#tU!5ao7%P;g|p7KEmYk
z%pbru%=pdy2=kp_g+ZS2CGg8SI0%o!eXt%rSWDe<hh(!?UAugT<REjg>%tQyFEUZ`
zkp=ueT_2yw_%~6Ckj2;yxR>BxiY!Y{l=6&3smMx{N@NxGYTRq^uSM1o-WZ=q-#Sqm
zkd4?)xHsY7j5ME46w6|wv@9n|E7FSHhPxgAHe@^D9r5fV=lV`$7j`G^-T3z)d(-*9
zdj|h^&*J~?$U*EwxDSUV$_R|cC5j8&-~n$K|DA_17zYjichBbCwY+;y=nlyZ-yyol
z9g>I4$6gS>L-dI|q!3wzy%=``{w2s#>}44{q@1$^Dv*`ft8lN5*da9`JERus5_X6Y
z>M=J!BYq}m!rTnzr5$2{7R;?+#m@$I%x%z~utOZsfw>d9@N+^p<{s$H<a`6@$2<Uo
z_zl4@<`Ec8;CuvdWA=a-KOc-?9tX`5?+@BtiIM}kDT$(kJk0q}kV*bSA?6||#?JsH
zm`kB7f&7OH?SC?7w(Yry`48HEWG(i(4BCI%UH_bI1NKJTP53t<o3WdRX#ah*|H#$}
z+W#r`PR(2-{ykJ}*xMJ`AH7WbkL<+Wg}W2~Ze$Pk-XZ2jx&Ffb@?`cm?O|>-6@7o?
z2=>to##mYCKqEcay}0}EA485~?;`J<Hqz5cdLrAgJNA*@>7+N(N?zM`k{`+B2hxn)
z5=?%P*R4peiS)FQo?JhU9L4U^k=_Q<8#y>be&9Zg{|K@dd*4y=Yl8ej7Vo=V4C%K^
z39>Znb}86-JNg#4OChoddoJ!e{PU3c*fqho3;LR(Mdn~{Ox-S~dA_I9d{4+a?8cqj
zSwGw^4ah3jcGeHKvwpap^~3E_#&zXfUxBO)NRa2LgHFL)a25_wKSfbrG9|5Hn5=mS
zvkCb}@WG4JZGWL{`YmnID-eQxE8GP?hUcIh-k^Ou3n7G!f_S(aeh5!PHdMivfEt|U
z3fKa-!A`gz9)TZ08dO6Qr1Fdpuc1wV+ewbDRkSe|a~<YyLnHI2%=HNCg3=6gtLew!
z{v6}3bF_gi$j4aAKgyWT#W<{fTfBT6OFIf5z#HgaH0Q?4tBlu7-;I$c;|aq$`RlNX
zYp-GacYt<O5l1)NMYt#Ny9@VUF&11PUIXzK66RI>Yw&vk_bS@r=dkBtzmj+^=2^yw
zdr)(w4CO}1aKU>rQvB|Px!k@g{ue4&$Ux&&=<9wgKI<9Qbv}`C>uDL4x5dR_B5n@G
z@o-p^^?4DStQPxAte3M1;vM2&!8-(Ve<zbPU925C-(4r&A6zOuKMIrfv#Z2$b%=C)
zahY^}wO+2|nXcg-uIKq<7@r=2Sf2maJfD_0UcvuU{7Z<}4lncUpAfc%cN&A67#BYc
zUh?Hy-sdjDJcs}NgtL&In|PMD@Mj%Tqlo7=%=P%aM;H_Szx1aI_qmDtJPDT(?%QDZ
zKM&8*$ozE&_6eTr7}Rt9_1K@_KHY?QinuP}Ssz1AlJ?&vzTa}so4Hp%BooKCxh8>Y
z{~Px{!hFL0in!)d;_)kdW`j0itK=**{+Va|bDHrFGC%7kDM-3W^c$5FX0rb&ntuzj
zm(YML!L2lsHH>}eKQnJ!fvn6zzZ+SNy#`r}KkEF<G5OcB&~xV6Mx+UQ6S5h9^E~=b
zr#bTw*}9DWGt&R>9y_uv8U5!B_9tec|H>RrC;Xr1L;UA_2gbJ-{x6joCI8<4J8~NR
zmnrrq&agib`JeM=nG+f#u5qxCJ_hpPJZbe)()8P;qnq@3557SfU4z>WWH<i7q+vE3
zg@^G!Pr80i8s5V-kC7Mg#I*pIlg@2ibBO%>4q<BWdya4A6?hL8@UwIMCtNp2evK3U
z&+tFEKg~66c#8X#aP7^+UjuIu_GzBwBH}s(3p_(O_gaPy!heomAaVQ}zDYd&#Q!GO
zUrRjUgiYZc+qu_a{J(?S6Y#V@|97+gAHZ5uD)aw)SpQFE{y&5D|18%3kp<ZGxEGpk
zkfIdU|97$e4<)2wDU=;!ju0xQng53>%+*k{#2g{iVKzcNehtuw*#u24^q;`Ii#(`i
zZf+-YbUcF<Y2%)DWE;=Wj&xw}$YhL~ME-*lx}gVpSvTl|ei(p37=mFK0sp@_T;TsV
zhX=gigE1IC#@-3i;7QW+9rzS(B7b5@FD+?zKlYcQmNYv@x!H~T-K62qp_VlMJ?Z`m
z>G&LJe<k6jace<_k|)ume;iy&n0w%U!gav)#Jh%j-p+kKBHn(&ClPNkao$Y0cgW}O
z<K7GjT<^yJ9RAb9`7B{x;2z<)KM9}1J;XT%j}hmW{_lh9U+0=s!rxDPGsu^bPZOpP
z|1tbJxc966_vHG%giX>LM%smwhA_nZ#UPBpFu1}=L-2qb{Qoxeg8$!!V_;!iZ7$~=
zJx3_}L^(IX6^agfsI+0XBOUm+<9^{k#7Ji*b@G1JZqk^aMV3b}J`ZDj9?SR~S)IuE
z9BQ%GfH9KsJv5{<zK09r`)qWl(!L(wXA{3)F*b+$X^CIQ{d4=6AHbf6)Z<@(dlBIa
zk%mmprPv?J+)XI^W<uq9(&qQ1-S5el58+DEsRg&6l8&LI`%R?TB61%7M*6QK{XZw%
zt(Y$&+&uOV3EM%sUP7MjC0smZO%GG}O=6EC&iCLRuKOuG1&82EuK5I6!95-(j#b!i
zgZs&w2jO9G!%evT9^!~=52Qgf`5FiNA)0$7!7k82IXuJj74bX+;DPcl*2-(h!#q<D
z%)n#hzb1zA584FEKjdQ8L0$^w9}1?g6MZ`6KPFm=_EP@A0N3#Dw?kP3`WncJaOza#
zTTmTO`A?+$BkMvb|M=BMQvQ*R*iDO+|C#ILLDIw=93VH71}y;r@-Xs2uo0&{BtRa+
z`~)0;4&v)1zOLA>=lHvcuZQ@0dA7a`^dGaR|B-{(hj1Upe*`&7cvn32Ki7MZ-szjf
z$NP<Ok8yBkULzhDo!}fB<XH7J=)ti!ISCyEWIua&2X<bAo;Uw7XD{+_D*r9VK8PG*
zPw&6~FW!!yBlH@1j%RIxfB#>+h4a%}!3-9#L)#MV9MT5$?1gUNo^?p0?iy**Un9-X
zh~0#}BAETdi|nHcp{y7Zq^9p0#+8gSSR=0vxLOK*%u6OQPpM;`(!{(Z=F(lvQyxnY
z1M8jIaP}M1&T4{{WFzw?XxIB_=dtUKQtp?~mtjJ@57|4*o(UAnhr_4`BB%$jd*Uhg
z;EJXm0RKP7$H4y&axb(pY1{(-Jsf6eGqHCAY~a`Z>tw>dqmMlt(Obm-FU=k!dpF91
z#fj{)pu-n{E+6+U&q$E6MRfeI*F>Q67l!U1vVnaX^$EdZj6*jN*&K<kAo%x?G(wJ^
zbDOe*B^&9V*w1J3DGzjXm~*qQl_JcApdVr_8%i;kfC2wvc!j*Wo$~)3co+`BE1)Qo
zPr;>>%j@9@coOW;0gCeYCR{`LybW?FTV)?o&bih_Tj8UOX8Fr0?1OUHNLfvUFv{mf
zNCYFi3uj?9W%g>gn7G!#f5DHT5;~y)df*IvkMjEi_#xy#2xU1Eo}m0a40&)F@Ai&g
zQqG=)L(l@hhnL|)&=@HH)s%k|<sX@gT{lGe_fh_l1=#gRBBjuX{vWbfNBPIC1X;RF
z`;RP-h?I(Q)*mrf5l^-LW~oVzklMtnq>eQ=Bh<%UEe*uin0U3A9F%|JYX<XtB>SIe
z|M$`UXVLy6ZP@MSBBhObrrpB)H~jzpe@8ssPTCL~^WS?Tq>nlB{%OiT{no)nzFqpU
z!^jcrqem(KlYGN_BiZN6H%mBQ;^i`Sly7-5K)xgm){suC{NDi#!+UTI`FKCv%s22D
zyaJbz?@{m_xC<VHr{O)gjBnu{cnXGL4d2A|PzYQ2E~4Nen2jN?_&!46b?kv3(O-wl
z7^7Va-(wu{AnbvA;C^_7G3dE9<R!chp98<d(e(42_g+F?LN;uL$KYr1zhHtFU=)1N
z0ex_uYXgxhL6b!O$5Brtlm9!(|5WlHnTI`pBlSw^H$+dqTv-3uPyL_4`o|vfKZWsn
z3i(f6S`HNn^#7m=b2ZeE&KLH7FOdJhGJ_@xu7{i94oHH#-~reJsqhs13Vs6zf!}&*
z^iT<PU<Ml;g>S-@@FbK#D9l_&e!v2pfjRgX-iCFTlg_Y+`BPvBsL`O8mJR>FoWr_$
z?hxmr5+<(?ePJjlN1qr9p$Ll4vHlJvPzq(`=>J!v|IZ{rCGk}0Sbt~YpayC&*ST1K
zXX2oqcpLgyf9D=1^y2*baLrs}-n&99`&Mu^JO2VsWB*?U`_J~T5IeMCZr{cEz{#A!
zzmxNUli2f>%HFRO&IiW52lw7M&i9SyeBVU&pCxeiPc-NIZsdGH7{NRmfo@$SXA_09
z|16CCXQAvr3qcp}96Cg_E1D4UKa}zxN&ZKW|I;Em^o#PQIFAS&t^&g830Jr*?0?{V
z#RbmspzSG{XHV-a?UIT8zeaSwoap~L(ElBxUFs9sDUn*tb$WEr4Cry`L>jU=|AO!)
z!Z+dGjJr8Q#Im2gPFd`4%@nbwJ)msq4=DRy_PL_>)Skkg&s4_1dqg^SbIwRI|Hs+I
z|8Wv2{|S`;m<Lo}>;sD9KQt@gJoy-kTm=y@{rH3G8^~ygg6rT0xD{@KRtxvG@eJiW
zOEvf8ne9Bg&BwF2xF^rp!LvGe_IBQ(2f8skaqGfwU>UuwCC=DBhfdofdRs>HwoK@8
z!60Uj6TK}5dR*LN6fT@cL%rtz_qaEWGd+2RN8nDNTGHGN_rvWledBI*FY+PC<h^G3
zzT?(Xw?Gs`LS>T37UYev4r)_Kd(x^JS(Qqf?<eik$b)_41vEp`Uh*LB2dXihXJR7C
z1#a+w7j|>MF+-5N$NhR|=o=&lN&imt{ZrWwfE>a;jQj}yb_;F$=_KiJ($?E((|xq<
z$ZqUCNV9{!L3t8uhj)tq{|Z+4_x}~zF#nS{uHd;Jg;Zc!U1Qo4EQgWJDYVtdN@()S
z;D7B0K-PrPS3r*9{zo_i8^fqW;U-9g`#`gwz5!?<2Xc4QH-Nm<VA+j-3LJ!1=z%}M
zNtlLpml7Y`4EMl`aOFDI&R{#-3oIvSo`zpTJ+#0GyajWRxSsX`z6U>m|AwEyFW_0I
zfLdq<D>$GV24EEa2q)lUxG0>pO1Kfe3%lT9_zj?csTqXl;V4YNC$N(5=5n|d5+Dig
z0)7v#c>**}%0KO?)<^kA=3>`PQUA|S|04@d)BZ0~|1Z=2Ba5*ca4*5X6j?_2@*(Pf
zuCGK^VXxkU{$DEkf5^IY^#5ox>a)=QLpEYJ;ogLQGt%s23^|1UpO3yI(u&<S#rk&y
z^Le!E?dj-0Y-In-Ec$=Y6~q3QdB%?T_dsto<9`$Fe*pVm%4z>2+5ZyE{udYRKa3`@
z|Ha_%!?OP+jQuXzlsy>Br2h}Oq(K=pLMOZg6YvT7|4%rGnc}V~0;&g1Gf<p01At0~
z=1U0T+slCGAQ$H0L%5vlZ-MRb9oS8Ngc9GSa6MSS4r6eZctW^F4_6cR4ZgJ<gx`wa
zkKuRlJ=}kSdo%Py7~cdcT$&g#@U2zAB)ksGpm0wk+|w`uUU-2x{s1-50M<m(dnf6g
zjJ^TVjNKAVe#DX=$gV8(EphM0zX#cl-La4USUUNE93=eEQTkgGq&Kn`dtWs9!Sw@3
z?=oWr-2Z)E;V5?3Jo#~&IR#{H@M_V8u9iGxe&lLiWVK{FR*N=bwdCw(tj-w~1}K42
zD2Q1t`ta3KNZ2A|RpM%?CjJ`kUyCdYSpD^WEMz4#az7L9P53t>>#!T=`5w7{15%R~
zB-x<dA4L5SB)O&_(dmLDkNPhkS%4zC9&;h)BK(Sx1_!!rOF>c!W%?j=41(zY21zAU
z^##%X1u_4{`j0b6>Z*gp$YOPU8tXsMn92GNG@<|2%pPcSDC<9w=vG9q{<9O^ieT1%
zlF|Q(XZ>f0Jq}^4|3C+H#<SllHb|TSL7Yt(Bt4O=|4gv{lg|222It!m*C1oVA>4<N
zBiKiY$Cb$X57L9(i@Oj1G2}RQ%_9AuW%TrsIkaE7r)mEg|K&kG6vQ#EghD8SVjcaT
zaK@RDj5CpC>Hcvk{U6*bP5<iuV6VYm>!bgJG-9tuHsIfgG+}Q-HtT3}vuJy>`QD%v
ztY8B>v_U)l7Ke`S&A|7D>@x8^TKFEJ2YNH;|3E+er2!rL|0jZFXp#N@%Zz1@QT~_s
z-oQ;;<ADqFe@6r5tuX3m$Zh<GTuWKG0rtXA;Q-}C3)N5$JAox{%_I2TOZjXCAG`=<
z8T5a6hYH%+N~j9>iu&(7_1}5wzw^|8=f#+IUg|T?GnPFsjhIc)lyY90!E8G(7Uy~T
zAm?cx&(l7h7aL|fZf(eR<9X)L&P&I_dFecTUb+^~i}T!h&hR_W`JLyb*Kl6?s?SS5
zVFqAOe_r^%JoVps8S$N$QQdj!zw?~WcV0ZE^WsJNvd_yHjDvlee*p#4?>w>+-AKmr
zoF~x=X2#Q&p_S4dKtB_@=w~{iBawb+H2qSfBc6U{7X3`Fzwlpce?A=J9EN@LOUaKu
zxbPoruJ>}S58U)iJuu4kF6?>qS@R(`mVJ6iZ7AO%@oK{OE{V5<er_ogC-dD+F_u4q
z9y%0Y*6-&#L)P*RRms6pO<$#gw!LzieS%BODUjby`&UYH7Uyp4<=hRZ$KH_1T*A?n
zv`hSZn>_pnD87Yk@W3G4Kz?5V-++hV`;ZF9;2V4g`N-db6*?giw+G-^I1G8v2446b
zprWAp9L~Uf)XQ0r3l~$zN5G}Pk#d@7WE>oX9H@jJLK-xK9iD`HX%8NRzhV9orr~Ya
zgxl573s3W2eb51K!5gracME}1cmg!Eh1sBm9LP1%{_AM}^|b#`K%1_oZ7)O?8EF5j
zY5xgRf-J>eKpfxX{VSGf{}*ZhPt*S6S3{UuWSxWYAJn^;|AR)%Ci)0XHpc&iG4Esk
zKmGqQ{>Qy-hCcT^eQz7(-%0sLcKIm($p4K0XJ~KcX>X{{2atorKeWexzU05g|5KEI
z!nlzh;_y!VYyO{j#)kss1kYXq-z8uEN&0Um4Xa5zJ&b@0dlBvdm|rH$i<n=<%?V#%
z_Q5{VpdEgR+X>S6QQV$Hew%pT$88<g?Z@pqxFtX#ZnOA(&Nbcm{SR?}i2WS=0=FQf
zou(;t6U6{fpVj<<{>Bip6MGN1He!c4ipp<8_~sx9_cn+Jru8+Euo0pm24W!&{>(Lh
zg-PPnB$NNU$$!YfoSV%0-$v^Ho#cO3fV@GNmO$PQX1L}x>{H<5Su&9;crU6yO+J*v
zYU~ZjE08wiW@Ha?8`6c0NB)Vv&K}aC4jSMF-uYoL!*TckK7zkNC~2`4eho)pn?Jpv
z7Mh`#@2HROsUPY856mEP2>UQ{B$fH!ea!!)GtZ0k{|nrU{4>|x&U2)JCfFs}v*@Nn
z4&*{R?=}GW3+T2_qkq5DCxwfB=(zXM|LqgQqQCt|$9);yc>K$u0{=$htI}7f>T~FS
z5T+LD@H0;JOZ`m0G|+EobkY7p6Xxb@^gk?&|Dk1LzqHz@hiMaSj4SO)12RQ??X;DS
zYU*7R^)B*$zMYrhHq!0ed^f$z)U!J3UBdkw`=F6}6FE%S5!^<HsAqlDyGRdqFYZ3#
z8bgj_*JPlhUu{-eqgmxZuFkA<KC{Z}Gpl^e1unDFmz!0g$*hX-FUIY{{2z5i=}z81
zf%m6>Q4z-;0jP>a|7D8z=e=uZcz@E<2=%Aw|B#-Iq=|`iYa)%BF`Gl^|3C}<p4I^L
zUn1%MV7BAdM*pOpI33WzJ^cSM=!!xA1-hfre<5t|Uiv@KpHBZLjs8yt{huuQKl|zb
zU>@B?{|DSWt0#&6Pb&Q%%wxFu^<Ol7j1NJ}xHZSc_|V4qkg;nXZAU)l0we7V6jsyj
zlrv{+VD6aqr-Xh+DQ#WZ5bX@tRuD%e<|@X!)lkE=wZu_pV(vQTU+dqjnKve~{=LBZ
zH|yKYv#ftJCvS;o{X2p6Z~Ux;v4K67^>40gXa3$1!TL9{GmLX_LfPAx$zH$=_Qr;v
zmEN?o(uaFL@eB}dka>OT1sRTIE+U%oeFSq4vwZ)j`Tpnm{+9ye6yMbhj4$S?HrB!3
z!Tdh_iSjuO1C;eyK8-Bulz&7~CQs_r@wavApO~{_D8rj+TfhKb(#8j4yz{uxi!N}m
zxWKKazXgM&!O&En4A1n*2w{5r`lXMw>px0Afplvoju+u~*HI_IIViYRG&=5+1~K@>
z!t_tfivJL3rXS!NMRN3zrV#SsW9q>(Jlki;x_zW~SigLQd3u{!RV9?G>SXd`C;5>|
zex;FLP?17@fo9n&*;9>5yU!~*N4=6;?iHQMD|t4r<Qu$F;H*^oc`xf<WvXb%E5%`D
z%24f<lGt7;J=&<s7Q9k^<Rz&{DpQr`OsZ<iE7gRrne|F-a+#{zSEh_vWvbrgrTnjA
z{EPm-Z<YLw=WS-f)9gh5KbrHuI0LMeGr+8&tLXosTOE9oHB@x7jf{VxBZl>F=tA$>
z3Eh}`pf?5mf77J&S5MOaeO(6oCS?eQr%%cVj4qty{F{^FUOFir@M8ACnCqmBgJ##1
zWP>(kN^&3<vo7m|<Yk<Y{PYu2u;+y6Q%^`?#*`E}r=)mjN({a!DM6MdoshC4CwTu8
zQUR4v1=UakwNMAfi4#(P^n^6boRG%ZDZYQsznEqX7tB+CX8*&qv`kM+E7HmyFx$km
z*pY1!)6%|oS{%@EY?|{grrG~6&HjgJ&e@!n9>VvAOiN$*G~fTU3^?Cp{6Ee6zsdUl
zUu0xvMn>mn#I-mh?&TTU{~7V}^FN<$27R>|Im<VqJ@=;Mz~9M_cR!%t7Iunp&neo0
zH$?}9F{h*mdeAAgUH7JxU@nDXFhIrLQ&I^FT>A;s>^mj3>8GSB;}m)@r=&jh6#u<E
zC4a;3EE~l7w*6Jg&r%O?ZT|_zJqfIx?`~4&Xmsr{+xM`3&89Icc3Ud{9$?M8Gk|g6
zUe?lg)9;}XZDY*Qh242n$rr?%e>_mmKE%0t54ffLRFE`$_Gc--0==7C|173wf@Ca`
zdA}dg7iRx^F&oAVT|u(Ub-u+y)xL3D9I@ll0iB8C)UV^>+%qoSyT_Sd9v5@OO0h5*
z+CsS2xB;;e#uhy;_Rs-oLyj&Ch->1HoEI`69#@EX>1+EqRCg?CHS4IH&z3Mw`{YAE
zbDZ^=Kcb&G&iu@{44)og+%V4g(#QDHCxs{^7e)J|7=>$tkMs4S$D~xhTGRq{PH2c|
zZT|TH&LzuUDLTkAaz3SVwbbqPi4p44Ls(xnsYVtCOrc)RDqw%j{tzh(ACvO1W>pd2
ztSUpBRaMBCR71^vpVWeOG)QV`cU>=%hL57lbRW7*Z1Bn@jd~L~>yrDg0IDhpL!QOo
z9mbe%GwJv)<NhvmonUm7ekS=@!=^FL1(wVFVV@;F^1O!4WIbGe6>b|K25yEOp!<S<
zLn8Av^tql_l7jiC@N1|6PXcG6Bbz}F<K(43KYie(p09pBQmS73hSa{qnbYXZm2Y59
z{v>_zQ>+~eYi(~|DvjvCHPfkTLhs#pHbUw?yGk0qxLoRGJ>w(#;h^F9x`=;b6@7};
zj15AWYl5fX0H8LZIS7YA3*}G?^<aVxl%GE!Ux%;YpRk5?CZ^mpOv!1k!2f0>({P$M
z=%1`*j`>aGsN5*7;;rH?xJlf!J9&+l%kgKahwEZR_xR;9g4@ebAm}07xL!^Np({hR
z{a3h>=KD@~1fGE+*v_lm0e8Z8;cmDOtn~dJK(=JC&PX57O#j6K-u-Uzl@H4pa@;m7
zqvuA&mEjh*@fGnP2lu*VDBUf?nQr<MZt0D1)84xIp4~DK?B*Y^ZgK7(ro6bNhj8ta
zqpVlBS+5ypy}~Wl{WZ$Aw?^5KZ9X^qeHxVIoLgFsxux}}oB5!6Wr}rkF2F0yht#M#
z;x!J{sQLuAH0XypPsJ@?Y~))=b2Fb*qsn-W@_0A>EjRO9ZuGHU5ySkjltk34Qj1#(
zK%X=$g~+0eVaW{`WjyDWytG%C&vMh>aZ5JOp>?<=$A&$d{%@#9wBaL?6W}4dM|2xU
zBoCRN=#hdDkLWQMMtGQS^GNZ2j~KE%Qj+G8(tRE&v$&)@+9MUm{v?&is*N7%{}HJ{
z)-H}n-NK*5xO0T@UxR8Oo<^4k{SOcQKaVsY@rXIhL;tTzwe0dptD#m|xu0#QR@oCg
z?9m)y{ii`Wc)pI<KQaERRb8R=${FWD|Jy^E@<{Jwt?D~ctNM`xkscY`?UJDxj|?9j
zk&#Ic`RQT(cSPJs&xD8dUyt~3A3Nq@4?P<qjaJEyE>K#BmG!T7`61t}&S#Z86YXDk
zfhsWRl-^=x{o5)<)mACqSfC7X1*!yDdfH0+XP5G1t5oc1m&zooR6+ILcBw(uW?7|f
zmsO0I>r<@Ku*WKmaaJ)UTBRw*D$UVWG4HjDW!@?+At$62X`Qu-EudZO$hP=)X^%O<
z{8u~rZ@H@Th*i2Wt^A+CD&54>6JTZkyH)yj6{-FZs|-vOs6p;Gv{;~q2{&@Ao&I00
za`Aj_+H%i)f$}EnlyAS4{;yTWcUmc+dX*iZS6XDwjE(+Jo9H5K!Z8<`{4g8!vrY72
z4(9)C@+fYPg8}!F5Sx?++n8^)p-*j-iU^xjhTE9`Y?B`mre=B){U4ju5!QIlA@#_H
zqYi0Ynv|d5-{jD#X57rXZ0M8Qq=k4|<7{F@+9C>-J;jDDiC(oQ*m!@v>fnBzE{AmO
z(kUm;*PS>iJwtlco0Y5jl5EnSX+!_VhW?*k4NdCRFmi<F8r|C_uG2R8IrsF;+QbXK
zBMuou{sOm=bh`{s<*A{V4jI(FDg&+##u9evGjvEVa(rW<8k@J%XV5E-{m^&}c6kQr
za@l2c(Js$owrAMUS+L7N%o=D(wo{I%v!Ex$F5UC5N-^dxgOxs<T{@!ejA`to9d&4e
zU25X(QXOiitmmsrW1*@DdsWK){`>M&v(t{QWC!z#tWm`B4e`CyI}4T3SE%Zc`B`?!
z<9@nb+%woNIX=5+gY#7OQU`<VSEY2ZLrQkq#c-@p74w`$=L%IJ_tR(SRYAI)#;8rQ
zH@1m3p^bUBHp$)HCUx8^Z&w@tPoVFAw_Z&Y=vATPH7VNJCdEl@Vz6|gXVfO8)t%^n
zz9u_2G5=%a*@>sh*v9y;O=>3Fq;{6L&b6`rU!dw8ZPJilq#766SpO+fP19}Y@O6qg
zGhbO^+t^R=nzZVQlr=U_*}~ey&U3Xzx3RaoNI5J;ssq`X)y6!eO`N_q>9%!BPhT75
zzm2umPU%MuoNklBr8XHl*CxZ<XJo!jMlIyOvrXKtHt`I#vHykquWn=g=QSBeYRb*%
z4wWkHF*EyLidC-OEV}q&mFF@`en=PNe~r=`%=AxA(myFy#fEHU=rc<R;YxL8DcfDF
z%9Dy!1+vm)ma54XsZKRB-sqCr-DauVZx-Y3F6JH0(vWPH#uT%daBtdWmgb#iF&{CD
zWx`DV*DS45X4;Duu`QZe|DykoY@amqt#nBTvNO1g{{KmFR(DBvnnv~D)*E7$KH}+5
zGRwezvkW4KQcBfugqi+tu^P=Zi|c5ya&te=QkQr`G|I>Gjb+jQUo2J{pGIY8TSV)!
zu>NP^yEQ0Xm_g+s^Cv8{C#|B7wMb#Sll{*Y>OYGZ0-REUERD1<p0Y^!UW-&@TBI_=
zB2{4)sm8q~#KQWlMe4F!#fYq5n34u$V}M0WvrcJ3HcvSj|4-4sbV^G~wrZWVh&9C`
zws4EsiKlI&h4H^d95E%TgZp$k4XP`~!uuOkcZ5ZH98UU|*{UzyBK^rzGH}$O2Gcca
zXt#y)S1mHK&qDogP_9{naw9#77V++H72mQ&#^#+ee$K+2L5uV=Zt6q!o+iy)-7=is
zB14958AN)LTg1JnRj!b3*6&|u&ZI@gd@bU0wTO4IMOyK1N$D0#QVVUnS=z`yJ8ri0
zZm}XeQ(DknXb}hI_MI(^*;<&pXkjkBh5t;oNM&`mRA4W6l&G@s7O6FssG6|XIYX@l
zU3#r*h-i^|+>Nm<)TKG9d2fp}IZKquSE3q`IceR@U9?DcYB!zK7RiUaV6D<EbxZD^
z*QF@6MGDhfM9=#baG%n1C8{L8MGP4RRgAkPtW~n(dPIxNiEWkKgjUhTwMt%0tK>(w
zO2J<8KfM+G|5o}Ftx`Nm{_kl;|N9LoMV1k^e6olBNUK!N_i+B{8&aLoBQ^RQRf}7l
zgZ!UwmHOFMX_#zf{MRa`rDE0OYGq!&RGA}M$^TN-!u?u}J?MYrC|hDH``_M>w(3&V
z9;sE1jjhr_xX$oabOuY6v#(TjBYSwR-Xsh239ZsU)r0<bEA@Y?49)g1{%e(y<yILz
z){6djtGK6I(f?`{Zy)*ZYh}KomHu9<{EYT~l71t@EX_;IdHfoA2~vlwLKY$Gk)y~g
z<jcsTNJT&P`^fdk2ay|)W%S8siR+Th=xRlyYjP9w#5b~L5X&6=9&|}!L~i~jbIIre
z{(c*4xHn_Q{|8&?hTztW+iB#L@ytmR&dGK2`2Un}f%t8OHQSlvNB)CreswGD&BkuY
zwsnbi*D3mQU6O0*5*>9w9x^|}ECp$&M1Pd}YZ?8WxZg8>iT(}~za>oKmL3!8e~V-%
zuNUn`=6`2LRc=Cu=pZl5r}B5NXMUgg|EW<`I6?W3^{C>#>%}l_P$lX7|7PEMDa&O2
z-!-Z#cKKB0PLHYzT`$#4^43H!|L+)8b*UX<4B#I<#NQCSp8l>+nRKJ7Dbu5x^`pw1
z?NgS_QPsjBSgqCTDgWJK)2|o1k@<fn*V~!TclbQ2!-f7Y?p?<9=#H+JZp=M!VgCnZ
zvwwm4-zkq8obafj)9Yop+N(xp*Hizm7Z<pXSj1yn|F!;K#HbpZ>_GnycA^UizWB?<
zn>4OGF_*Jfa8$VhM%5_uaU-XlGK?IGyqvj(%Vhxip$~eQqw7f-Ro%(V$)#P6ZW42J
z$(J)%>l8=Q<<g$KO1jCTN%GK^b-7rf)zZyek5*Z-w91TZPUx1V_-@9m-SiDhRfAEh
z>cid?V_dh??JH5WLoHGh+>K79R#g$MvQMijGP)T9WUI2+ZuZP|OG$D!dosGEm~chD
zZYc~dQTipy`<^!?e;4(Eshc^S7Rg2C7^yEZ-jr;lCbCP8x96&f`|{NB4KAch9lx_g
z-NpBIyiUh|GIBZ3K1WSFkfV;9k;TYu8vfUz;oly}T#c%~g`p$w7st1%ryr0EkHG&x
zC1ijJ_QGK}ewO>>a-RoA)kK#^X(l-Tk#A8uBa(y6#jaaK|BUZ7A6bC?5x(^!++zxz
z$qASN7h~0pl$El5jK$MgCn~sJ;;|>fPS^#>&?4J7WBnE}f51N5kM9us+1sS;v+Kkv
zw~Fm63K~j4+4N7esbP|{Crol<f<(9AQ+a$N`Qc-#AS+DtJHya_43eU0pDGUR5d+th
z>}UO-@MXJ3Rr!#Q?|)2Hx_qi?_o%9N_*BiV)l$oKb&gIkW`?o;&%8`>7~g-8n9|2o
z6ZU45kIarRu|SI}Oj^OJ4-#8Sr`TOV(xw|z?Y=SPFoj9Syi0YKhe_ADFxG#=q`RE{
z|6-W*E(A$mj8FAb=MBK%^r#v#hSC2Cl95=S8Vzno-(pO;iQ5xE|KAqI{7e}8UxH*j
zz^60;F8+TpuCy%3<fN~q{};@>msjOQcT0ZeTIvtxzmqwCbE8WY&3ILD+_*9v@v4#p
zmnub;p`%fb;#Ng=hg9xfD^*-yjWSFPvUcaVs@u<eY|>h(hX!Z_6Ey8*?zgW)%*+p4
zOyjDhdR(;*trhFcsIocMihX)5`af%>T~Gi2^jhi2M!zh{tGf7ZozUIqQa$I^O7A)J
z)8f6VKi0wcXIu?tua%($`u~QtGNM~6qtnbsM|hQc!KFO&<I3w?OaE^z^PkLDhj^7M
z*2x|Lr+9X*6L+v%`DO{j+QK;3jM0bgi(JPXyGISAuVY-WPKGkqQ4iQ<WNKXbbu1hi
z>)4~VPC9iC+IiL{cJYj?JM^GT(9QE&lGd@t$=b$#uWCWs_N)^tvJGhmV+LuY^Qs19
zy}_+a`gQ0?FgL1O#~xval<yf=W%I19Osqr4#3|L&>!fO#wUu4ts%FTo^u~4QsB$)H
z{J1Jy<ou;+@@UGfN?rV;EPI_~J6Y^H?N&Jv<navWFP-zM+^`PDwuLIs@0Zvm*?TV)
zZ4&G5`!1E-2#?b7z2rqu|1a?UZ}cer^q49{38IJva6=4baDnz|qJ#FMNR^q2RJpB4
zRgjmJP&Mt~868ry<d9mZtET=hFH-eGMXJF-|6kuBCTL<z*bL^44zWN>R)@4^(*I8`
zQnrjD^glbK4ccQI^#4fP5C{9e8UO9<5NA@6>fTqRdQyv2FKtyH^e2)3Fu30#LogiM
zAtNzG)Zs<S6<nm;yt4<qnGW&6Sc*f&K})5TPCc<73^2yK<r_Dl)6AR~_Ls01xi*Uf
zvwmr_lvHnGUUHN4`NQisqxZH+!Z5p_e0Gzxe#DsW=q9N;w@GGkGZIfj`6gNNn`bu3
zGO{@&Rzj%HOol1O{hP!@m?Z3$#Z9sk*%q;h`QOcwj=AIXX30dhM{W{F=4R;*-y}U;
z(-pE=oY0@RiFq;R?BaMPD%s&Zn`C6?CemRO`@lDeJ7p8^&HE6>g?&sPE7h32Y<m0)
z@1}7@OLkV2XtSee8=@rF87;c$XvuR$Nq%{h6jVoX##gk|&?Xf1MN4sZv>2eo6eXox
zQ%1P*g(#^QijvC7D5+YCV*D2+H8W9CI}s&y*p1lh@oO+fiGzFh!7TUoA*sIEI}*kC
zKT297qU0!Uwv=eGCqzkGSd_GfbItx}nZUg>E?T+*qS1efmhQ+X=0Bq7|3}exkCOiQ
zC>hutC4-sl|J)NL!>Lg+k{l(Y*j?D&_<16tWYK?bI1T4OW7tanX{%_R?0@xbW&Lj}
z>wjBW|Jf?}Gh3x#eyiwDZ{_@_t+GV;0NS}=2!(JcE9c*WPyv-t)wflup=M|+_1`U=
zO>>JF_uRty54T7|#4XZjyoK-o7RvTEY2LMs@!vMF9NWhDk9LX;W7dgnV&fot`}{WM
zf3`{cP@Fhianf-*PCEPQrE9WAoPD=T_t5PU!To!as_6gL(EqKc|64=<w}$?2J^kMr
z`oA^we{1Of*3kd0q5oS$|F?$zZw>w58v4IAa@3#xSsSD!dxOLfpVhfVVvz~ZR=z<}
zk&bg)`2WHd?!QI4vbTs6x=kCThiiIW8>DYxgY*w=kiCQ*T-v~T-v-wEHps}t2HA<5
z3wtusW8A=617!^|AQLnRTO@mVgJ=UbNKV8CNykl>vPJR|Hn0w}K?=e*py#uN^xs1N
zbqn(-TNr<CLH}%nlyXfO;mXrCNJabx+Mf+lmAOHx_iRA_Y6I(U8>9}q5qmv;4G|lt
zTW=Cm@az14;U;MYjdO=&5A9$KwS)05=U-sgxn4s5o%1gyIR9di^DmY;|HAkZb0?fv
z5b!PT|1Bvoyd<RwLA3ut{Qrh^fE4uq_5`7a6oekq9a2L(Uz-smb+{Wh2GRee{f|E>
zjY!k<d+dLv{f`V1^FokV7K5aPd$l61;k5rtC&j+;CC0xmG5$YEngvM*_v%c3;cNZB
zjF<TSgQSOh^v2Ns$DyB{7$gIP86@n`GWu^0^xxn=>#>=v$Fc_M4G0z=jKMhge{Pr^
zkKR?{*LqhM^spT0Ww9obzi*23Ptbp0{k;&1SdTAWIwpop^dC~tf1tf9OFKqil=c5a
z+9>3O_5Y)+|EK;x^nW>bq$ve`OfWmpH-#1#`hQ@>Yy<l=`hU>AfW9eoETR9ijQ$Tg
zK2CIfx{*EUzmr~c0Q=xS|IfxeH1VDcqw6zbL;vsCd#p#J|C5IP4|p@t|A8^g<De1x
zpL}y!Pz~=bP(LAuzz37?8GMBMSr9&%AXo`s;l37WiJ{JdD==?>MZ(>Iyc6z&FR?#_
zd=$P(U)Dj`Utng*Lz4?ZA5xDXLy(QgIxs^o41$yJFCve_2{;LTxUC`l5c2QX{{jDm
zDcYYaaDOwN@gnzf;r}h{p}6nB{yp&F_Cw_VK*z?<`0p?0e%LtoW8!`3-gy?iO3qo?
zb(T4b&!lDFJJL$oFuOSS1MJnD17hM_5W?G#?vOj=e~D)l>4M*2&WEGi;|TQ6aLz|C
z=hP&hkv_`QaMoGoSl*=%#W^);cSt19R!aC*XoC^>Bm55j0NSB{$e)m!Wy&dKaSrqQ
z@K0F5IJn-&xiZTWQinA19!)8n0|Jd;f{JX;rAb?0ocgYm_i-)__S)>vSf_Yjs*OA3
za^koFZsa+(Ag_X3;9Ky0h{OF6<UM>R_agrbj`3dqjr<w>FZ>uLaLY&DM)(2bB22<G
z?o%)cCfa@rEa5&2-Q-c?_2f+)V;P8m?{obOVdF8!z=M#2+Y`t{_$fREPs6X^5Ili@
z9&(=RDv&?MUW5Gq*t!!SO_RIM?|+D}!w><4vAqr)-WY+s7+4`RJ%=>1fz^p5G+F`D
z2(Y#GAPF(j%;*AfgyrbE@B6;*>Z7{4yY9PcnW|EE&q9rc4y2hLBhO6NnThORW7!Dp
z25Wzww_2<>?5Ovb`5gIV<~N_`ec$KF{N-C`*2l%yX`6n}_W%1#Ux)ZR*1sZN{UQD+
zTKBxZL7SGhX!{BNc)I!L`|hv*jK7q>o)+u)>uK4(6}o(8eEtr9{;U2!#g+ZIw$Gpc
z=l(y1_{R5)?;G$N`$hlXLOe0IACrIcqcKI(bcuSO@%xC>|HpnGkp?Xf(ePin2bNFk
z_foWd+3zcU-R~>@f!|l;pXjo`oA05AY8_KY^*{7u(Ljwq@ng|M%|G&I(efid=1b<s
zqKwKv<bU*0<!}91R8h6%8mj*_ziCgk-}{DfGrw!`Wxs3jUHv4Mo9O>Df0pm2=Ih3X
z`E07h{l?%2{@m|ev>E5Bx);ONN30M2C7+MD{YU&R2JcR*xXm%$;-0_tI~n3G>sR$H
z%)Fn^M||ZmzMfC%Q?PzbJo^d0pig}xCVqfV$i6YhOk2LR|4K~q{7ljM6MjRTR%wm4
zUv=K7G|y2yFW&{_+!W8rWBDiaY0%<7)TeRwLL}G0zoa+m-_eh{&-(}TQ}omHhpv$y
z7XK2xhra8Y`gQUDYWw5juTYugYWkYxdMc;Ck*--hO#L)XyRN-uaf7rkh`Yq!68~%J
zxBSEU5`L6MEKkzUS$-G&t-O9o{5|ng;@_s<rLWN6*;X&UBK@@Zy7)h%7c4J`SG{jn
z#arZOE-wB@bj$J!^k;PO2fh<!f8+a)DyZ_ma9*`dwYWz6*8ZdK@V|T~>a}eHHS%ON
zQL{E|`TC2|`s8=?e>>0A@!da-&aXSKpLL#p@}Eb~Gp>u@;=_3Oh3Nl;>*VKM7tX<u
zb20p|>x4#Mc3x?mCTQ|+oL8ErOLX~V*9Bdr8M^kc>w>P+4Z8Vj&MVE+0xf>od8K7q
zq19h=UC{dX{`c5Whs`I=YgdPDb=WE5?|Rnt+_)*;@|?MThfj`H_y0w#iI;6(v42s#
zWdD2<pPluaFaH%^2Or+wc-EzB_RZ@2Ed5TE=hMsko{h!R(q0lz+CL>8w>}{r{T|=n
z&|k%{c!UONi27-Oda3x|38i@y;?6JsRdk8l9or$VR&krWn#C<u?n}glFSA}98h+rf
zqE?-X-zKj9Q@=G#m6ofhyzQ2LPMx%8e2@G8y!8G5vg_an_0#>w^yBoO{+xSzdd>Eq
zw*C{AU$=ZjiMnn3hqjH2r|JK)?O)Nh<)5_ON7lu^u>39hN7nzH^?zykXD$Dr&-{O5
z`G>{-iSOi}692sIZxcUGAE1Y=|G0hsB|T>OzZU;9+kaC0dGXiir!D_uYPbA$aliPV
zB=_yo&s$!mcUxW;|L5)t9+T!>^nPhROy9BnU+ZJO_~-n;f64#*SNea&mDa2NmjC<j
z_<zN<*6VDqx4+?Ue<&L7@PAXY-yUf3f9hM`|B&(ThoW8FVXknexXXICxaaE+Men<B
z>Hm31|Ib7Ee;(5R^N{`@|D&Bop8h{#bnBrQ+keRY??e0$59$ATD5k{I)-Q=K|J_4z
zMSRuzjQE=Uv*PR4Z-{R?e(wGLzx|UB`JD#;-~R9J#L{m)<o_Q%<o_Q%<oWNRSfh2z
z8~glUpLi&?KJ`#+i+3FJ!;d@f^rQ3>p6`BK{152AqMxE))@DB~ei!``{T#iAo}$O;
z!}L-5b$Xw)AEgf!()M4m?Q8VE(eKz+F8&SillJ`@{ifwl(I+gA+jo_&(>g8F3$#t&
zpcm;c=(nW3C;kq7%JTmy{+vAio%jcx&!=tsQE`{g#b<5*GJV#zXXstFf6aQkxSOui
z#U<x|%lQ{qIM<cm{Z~=-vhz<hRQos1Kh;wMHNNcpQ!}+t>(_UEukA+rr@j{*^68x7
z|DWam7x#Du=>4quR?nGlB_8;?`G4Xe>%-!aCwBG!x&NnenxILVqG`HBm)~&zPk%rE
z&+;r?ryKPDd;dRXJ{qmi>eGB#v`!neNn5l%wHrHQyRlxhfj0SzGwp2o1v;Q_TfZl+
zvag<g+PKRFdKa}=|0VJ7(pTtf^gNYQHFetGE`Cm%_lx_if7$Y1Sf2D=OyXhNF4G^_
z{>Sv=Pq@#w{&maq;tSSS#2*%y+4kGw-=*j2YgA?1rv1~F=c&$m6D?c*Q`_4te?$B)
z=$q7K{a@SnZ|Dd8&cF}TJ=<Hox7)n8+r=H<b^iX=`TNs<8>Ro>{=I)2y{?Hq`}(CR
z{r~nm|2B%>Z+gG{T70)&bdOO+<!7!Rs{EeM5306YL$&tRQT-47X*9Sl8kbx@u9IeQ
zi}g`?jL|qv&?HUKG+m<0WWHrwr5U<Lvvi$q&`p}7c`_$67HNr=X@!dKf0oHur{aIj
zY|<8OQ}MrM;w|x(KP#gOs-!BarW&d(^k)}KVUecx&&!4N)Ic4DN_#s>IdqivcG}um
z+S*y#+F9Cq$=1uX5N|2g|6D1r%H!3>La(+7n+lbdnk_Y#mRc;el$Kg8wU(BOv~926
z6t>4(?w9u!L4W*Y{G`_}$4|w76aQ`eqxk9g?-=HO=HZ{Ew--Ta=|ZveND(|*1dkQL
z<3;d}B6w#JJRvBOT_}z!k`^h8ghjd{S&^zp^lo|&y_ft>;Lp6DK0v=hAEXb_hv`@8
zBlIMFl>RAwjDC%No&0d{&wQLdK~K>q={M=O=(p)<`W^Z-eTF_uze}H^-=k;f^YjIJ
zmY$>M>5KFw`Z9fmzDmDOe?VWOKcqjRKc+vSH!3fd!tzpBQ3|U{VRb32DTQ^Vu%Q$-
zmcr&z*is5xOJQ3nY!@D`7I_}77P%d+7RMg0>FoWP!}a3W!}ZQ0+?XiB&9YKh9D7tX
zQwnE`uzs-!n~J?hO-1^n)}|tC+bqI?fg&9A4|RU#Xm(n7+%r{#gSDlwDDimkauJSR
zEy9~cZYO0${wH;7ML1BDelodTgi}S`PNs@-Piu<wr!@;jI6qc|3q=c@)f6>MHw&&@
z-dltXlSMdG)F)pm%FQ>6dj56!wIb}QEW+N&zaD+C@ZEvZ!f?@6e?3tg_18Cx8vgZK
zQMd0_6ghvlv1p&~cFF%|<LBZZ^N9Yuj+eI;ueZf7#M|TTZ;iLVE#CfSS>+?~_LK6p
zcsL$@Ydrk6c=+vJMWiK)aIjWbN+3nC|KZY(hf6yiF73Dw4_~n3LTSf^(vAzI9goDr
zkJ#}@X~!d_O^?RIkJ|KTY15;nO^?OHkJ<ECY13n+O^?UJkK6QkY18ASP49?@-(k}`
zN}JwM+VswN_?<Srv$W}*rA<%7!%x`sL}}9#rA-jyVL<^VQrQx85vq((`9ehsl_Hd}
zP+p~MN<c+c7vjQO1L=jgdlgZ!`$DO4E<79;N~tfDn&g6{rRKO$YKRMu#D!8LTqrfd
zg-88jSLxVN16+74E|lOeJRTQH$UF`gN}#?lFO=|nn_VbDT~IN(2_03+qo`h~lt<!`
zw}$rC`|?PsYL7e|kCZC)$c1>s-bdq+$KsL4<B@m7Bkznyo`^?eV`C|^BDAr{@{uCT
zQcWL?N8cLG*Q0NbM<0$yFT|sd#G^Jo7LPt2kG>-weP=xSL_Au`phWa&acGI^v3TsQ
z;k@dCf6V^JOW`}C1o~K!s&u9wkH_B{PIU?U@nUo7Y`-Jk@z!v<-%+YmsVUx3>@1!4
zcg8#48cuwvVcuD6E}i=);)%D0lV574CyLFb^Z%}R*Smf({`>eJ;uphDK>s5BQv7oK
z@=wMu|8o43cz3+}6Y=ghYrCF{_r!a?9Pf?yzAxVU#ka+KZTrIK;(hVH56Am{JKpyv
z@xJ{n+K%_f2jT<o{>}K6_?7p>uYB-N;#WTQbo|O6#|PtsABzt@6(9Uu*!ZFNQ1PM<
z$A>-<A9^Z2^qcYF_|^E;_r$LXpNx;hM?M@M`S|<dBcG3tys#2a#*^=fC*K=SzAv7X
z!IN@*@~L?8lkw!imH23U^uzJd&&Nl<82>ar79aayeC%US(o^xVUyqM{{QV*Iuf?y&
zuYWauBYxw3@f%OZZ&>(Rd^|q>{*T1RW%%(Y;}h{zJoO>3k_092PsS(T8=w5`_~h@z
zZ^mza$m>J#Tk+fRKgRzQpNfAL|8x8=@pL@>w&&g!PyfQF>8tVdi`(%#@jFj{B|aUW
z{)I2nAH=7hel|XB^)vD5=e#b^Bk`H|%zNT9PseATjn6#ib%AXCllY95{A_&osrc+y
zJ{P|mzgxVXp=YCrgwMt2;&Zb6z4$$BzxQ-J6VDVc`g}a|d_42Dc;=7d^YMlF!n@-O
z_`V?77oLhQd@{aJRK>~rg77)73-m~Q;rHVUTJqU=_Ve-VbMfr+@$45r7tj7tJgf1a
z3#lY}?%na+Q}Ns<<GJ67=ROzDJ@afl=Vf=Xzew;!t6z=hwA_R6LOdVOzdN4)Ks^6c
zJpajf{`2wtbMgFF<BRddr{YWTrJ~`!^zQi52jWZCKN(;8bbRTn@#XkReC6rS6e{KQ
zAifY^jjw(szWV#|`|$_ywfNdIUL`#fe;9uhfAoykGx5jqC*hw;$JgV9@pxe(UYLp(
zrsIWc@xsmb(gy<5?#0+Gi{0|rt%%*q*sY4)>e#J`-P+i#i{1LzZHV2**lmj4=GbkC
z-L}|mkKK;g?Tp>7*zJzpzS!-L-GSI0jNPHw9gW?w*d34EiP)Wt-Kp5U6uXyW_e$(u
zjoq2py%xK(v3ossZ^Z7+*qw{r`Pf~E-No2lirwYdU5nlI*xiWT&Dh<F-JRIG7<*;0
zR~~y6u~!*;Rk2qcdo{6F8+&!JS08%~vDX-TO|jP;do8io9(x_J*BN_VvDY1YJ+apt
zdwsFjAA19_HyC?Eu{RuhBe6Fcdt<RT9(xn9HyL|Vu{Rximtya7>|KeytFbo|d)H!b
zHukQ^-i_G18GCcFw-|d%v9}z1E3vm0d+V{c5qq1lw-tNav0oAUm9bwH`_-{u6Z^HX
z-w^wavELN?&9UDS`>nCx7W*Bs-x>Q|v2TE|C-!?|zc2RtV}Bs_2V;LI_J?DCB=$#R
ze=PRLV}By{Cu4sq_Akf&mDs-;`!lhBE%s+)|9b4-i2a+fKNtJ+vA+=ei?P2H`^&+h
zzrPy$Yq7r`ocQ~j;S<W+>Ko1Rjn??aXnbQVzA+QuxDnsD8Mm6^R&(5HiCe94t1WJ|
z$E}XI)fu<C;#POu>WN#uajP$G^~bG&xHTBJhU3;q+!~ErV{vOdZcW6k>9}<%Ze5OB
zSK`*yxHS{EuEnj{xOF{l-H2O?Sd3dsacebht;MbNxU~_tHsjV-+}e&?J8}DB+%AjT
z<#D?rZdb<bs<>Spw`<~dZQQPl+x2m~A#OLu?WVZh9JgEIc5B>ji`(sSyCZIQ#_g`S
z-5s}k;&yM`?u*;~aeE+c56A71xIG@Xr{ngexP3WpUy0jS<MvG4z81G<<M#EqeIst)
zjN5Z@dp>S2#O;;1(;9c$;!b<q>5MyFai=ft48)zmxHA%W#^TOI+?k3y)4}|I=StkU
z8h2*m&TQPd5qECJow>L(A9ohw&SKnIiaV=uXDv*Hc(E*AERPo};>Egnu|8gGh!-2<
z#nyPSEnaMo7rWxcp?Gm5UR;Y8H=hrWH=b=g8VNe$pgIm}<Df1M>f@j>4w~bjB@SBS
zpe+vC<3NJWIOvLl?l|a)gWfpki-Z0+7>I+xI2ek9;W!wHgV8t`i-U<cn2dv|IGB!u
zOL1^HeBvEkjf0suxE2Sqad15jZp6XOIGBrr`8ZgJgT**lii71iSc!wxI9Q8=^*Gpw
zgH5k@(+B7&`Xqgho~7sLt8uUu2itM56E9tim&)R$@_4BtUaE|js^X>Uc&R2{s*RWG
z;-&g{sUco!jF+0?rRI2PFkTvpZ(fXVmc=(K;+r+`&Bpj<Q+%^IzS$SwT#IjR#mn{a
za$~&Q8ZWoS%N_A@PrN)BFAv4b!}0QRn2vI{Ebf-a-PX9<7I)j@Zb#hhj=Mc^w?FO<
z#NENTI}~?^<L*e@9gn*cad$HAPQ~5nxO*w?UX8mmarau>osGNK<L-^Pdo%9N#ohV1
zyAXF5<L*-2U5&eI@t0HIDv!Th{L85#m??tUB3O>U+!lPh)$3a;MX*`~n}Tl-#<$zN
zij}r+?Zmg+y^56%uOjaBD&nqhtrdZ#?jq<dg1&Dp7J;SyA{g*0#YH}YUd75#e0#Wv
zM~h&r2*%^v6Gc3E9!NV~EL|#s%U(r%<w5Z6veNRm%SEN-V(F?^v40-RD`WA>c)T(Z
zuS~`(Q}N2Q*DIHb;93#P7Qyu*m@9&L!K+n8P+bHyMNnG=bwyAwc&*s+TCwA`V#jN(
zMbH&RmX}_wc;#}D&!tx@{Yi7lBfVfTUb*a5lz6!)QBdS{xhPn0zU;WmMfsPD@-G+V
zUoNt^Toiq|NOC1!x$0HqA}CT7k-bG^|J5RE!Fht&^8`w`s)SPF>*t9}<y|f6d9^6-
zYH8=IRi(p3MJA%+z*nnEM~I5-iz;0$s&w^yt8Fu0#c?x5ZZk!0GevGQMbeof=}eJy
zrZ`-xYen9IVp|avTVJhwOA(ZsO!^X*pp^2wcx&e=i=&DqnUoj>rPZQDv_-B(RHQgx
zwuIt(QS|lVs8TbO8llt#H{z9>UV2W7+RweR@M`U=l@>}HUTZ9>6a~_PSCNOH$VF6S
zzi_@>s>wo8j^KQIDf5LQ&xN9Dg3{6FP}a^jf?X)3FDmQ;Su7F>N~=ZQiv`VjRHP`1
zSS&JKENwkMLzd6i<hEGkwpbj#<W=M#co3Dg7C9^xIV@@Kwc<Dn#X+U$Eu~Z7s(p)n
zm9N%Yc(qm)D<33$t-VNCI;d3U()p2!TlOj{zg!e1D0UQ4v8yOkP#m{h6un$pDy1m)
zdwZ@FN39f_1rL@Vtd&v}X;+G+Rj*>lnpaVrptO2kS?lL(rOH_@jw>QZuAMJiTK6iB
zT`$rKz8^_jtQ9GWwCk2$z39CwXf6e>wiQuJDL7wlEtZQ8I8iBRk5|jQWG|R4g6pL~
z8LyT-2#S2mlpwHF?p35Lf3;HUilh{rZz}3mtd$ZLhm;pf6<)7gDgwn+6e%l;Vk%s!
z0!IkG9~H@q11pLH?XUDXD3~aMsUnb1Ws#5IK~x-St#m+<d1dJuEJZRZ){3l(5(TBi
z4}wzS^U4+<e5LEEI5AP|uPRC^-ULNZq%0zNh|YJH3M_F8&eIjwz67FIZk#U^d7U3s
zyrYZd^CZRF?Hn5U`Leo}ma5~`>LRV6C{|P)QbbY|kt3_m_ms-Be7>i|SzVM?{mNVs
z%nM4#%D6c8LH<7HEaR;{-zq;VHAR^<MRLJ|<?}U%))a--_>A$r^7n5rpFE3EWKmb_
zkf6ReOmM!e?)B$wUtiRzzNk}uQK$MMTfz6E;>e;--n~U2!}?MVMeYvNl8#)9SJ%U&
z*H^dW)t$I^G455wy~?;(9rtSDUTxfKjC)OSuQ~3u#l7CRHx~CM<KA@KyAt<i<KB(9
zHy8I7;@*1P+lYJH@!G|Btt?(EkJqZgx1;aI*J>2#>+!X^c&#B`YYg9#uJG47;`Pa{
z*C#hc-RDuyc{F$)jh#p1=g}3(X3iIHoJVuEEXM28W3NxkV|x5Nx>1TQHN@+e8sqiL
z-LGG^^K#F5G<Y7(oJaHL(P}BWHum~8DX)!}qS@MbeO4Z`^0?j^uivOA>o=^=*|%_!
zYN%MYZ^6DL@lp{>v(it4G(^Mk`dZWLtMl>t`poO=edp0wyuQ)&`nrvqj@i2S`u38j
z{Ppb>QR8{kb{=(~M+4{4-0NEo+7f{mZL7+5Yw7Uq_EM7V!SiVBJeod_X3nGezdvUF
z{LqEJ->^_@c;n)9yiqY2`n#!&%Bg}XsfwzphH9yf>ZySmsfn7Yg<7eN+NqPesGo|w
zhiI5aX^h5cf-cb&nxShnOV{ZJ&C_zcQKep0%CAyhRS)%2AIYmqURCm{l2_FvO%a|d
z`Bq&fc~@N}`B%xmO8!;yuabY&O`0R+R4vdVEyWwvxT|qj<E_S1ji(w<wfa{puUdZ9
z@~W0swLGfjQ7w;Zc~r}zS{^m(TjTf|b*|AqH65hxHR@fXEow%{aW#&+rhTtTdriAu
zlm43Y*OYl}gSO(0S^Trk%dB&<=-e!Jle`wS>!Q3CF)vBGtj?>_u1dQq-KzAfSXPVe
z%2-vWRpqQYhpWn3bq-gRx9S|ODs$C2Ty+jtF|R6rRh?IHuR4#b%3M|Es<vOnu<E?7
z4v;#ms>7<bS;e&KoUCf2Rp(^YIazg1R@HCSIazg1RxgutvZ@VNos(7VxT+mjos(7P
zWL0~vY8UUFH&(Uhs`gyfo~y2d)fHN$HCiXktDB^aR<~&<-dMxGhJOwJ8vZr>Yxvjj
zui;<AzlMJe{~G=^{A>8v@UP)t!@q`q4gVVcHT-M%*YL05U&FtKe+~Z{{x$q-_}B2S
z;a|hQhJOwJ8vZr>YxqqLe`5{*8vZr>Yxvjjui;<AzlMJe{~G=^{A>8v@UP)t!@q`q
z4gVVcHT-M%*YNuhp*Pm>ui;<Azm9($|2qD4{OkDF@vq}w$G?t$9sfH1b^Po2*YU68
zU&p_We;xli{&oE8_}B5T<6p<Wj^78`8|ygNajxTB$GMJk9p^gEb)4%s*Kw}nT*tYN
za~<b8&UKvYIM;FdIDBIr=Q_@Hoa;E(ajxTB$GMJk9p^gEb)4%s*Kuy(+`zeka|7oF
z&JCO!I5%)^;M~BufpY`r2F?wf8#p&`Zs6R&xq)*7=LXIVoEtbdaBkq-z`5Z(Z#d5z
z&hv)zyx}}=IL{mSH}G%Z-@w0te*^yp{tf&a_&4xx;NQT%f!{BGzp;UT1OEp84g4GU
zH}G%Z-@w0te*^yp{tf(_I5%-_;@rf!iE|U@CeBTqn>aUdZsPQ{^2R34O`MxJH*s#_
z+{C$wa}(z#&P|+~_%`uv;@iZxiEk6%Caz6fo47V{ZQ|O*wTWvJ*Cwt_T${Kyac$z-
z#I=cQ6W1oLO<bF}HgRp@+QhYqYZKQduFZnWdEIngH=Wl_=XKM0-E>~J@NVJV!n=ic
z3-1=*Eu338w{ULZ+`_qqa|_=VzAb!P__pwE;WMG(jV<SK3*#2PEo@uP+ZMhpd|UXo
z@NMDS!ska4-q^yog>MVr7QQWfTllu{ZQ<L(w}sEQ@f*I1-`K*mg~yMlys?F&_%7X6
z_ic6GR`+dn-&Xf+=W5%z+QzVrVH?9XhHVVn7`C0OZRctm%Qlv6EZbPNv20`6#<GoN
z+d13Dv+bO1W7>AkwsCE{CbzL|W822Ijcps-HnweS+t{|TZDZTUwvBBY+cvgsY}?qj
zv2A19cJ8+EZ98|{7`HLHL3v{v=eF~<jddIA4#pjfI~aHH?cm$Nw}Wj5*AA8)EIU|s
z)PG0)chrAJ{dd%VM_qT+bw^!y)N@BYchqx7J$KY|M?H7cb4ML_)MrP1cGPD_eRkAm
zM}2nGWoI+qyx2+|)JgMR52F8I>3n7B&5KL+7Ex)b{7Sr8?Qmc7Z~B^lvsyOQ{WL(=
z<IS2TYNnQWv$mWnsfOyPksMd+xY|jYrWvmX(SNXXzH<G|TA9`sQE92JEZ(e>Wt}YR
zWLY;#<1|ZB*4?0+G)If1@^!1U7H`&HB*)h~zP^fTNgnmr$no`#uXlWdJR8PHz6}#}
zmDI68x(4YwVRcK>(-3b?%V)Ye-kcfwGo>tz#r=zAR8F;2NA=V|jnqWV)IzP)M(xyD
z=r`{Bd2}(G!}pig8y_qkH4J@!>H7Eg9@lj|*mm62^Zlj4YY&zTSATzL;KAM^_ipNm
z`(<iVrY2=-QKklEYEWihSugd|APv(fjnfoeqAPTh=4gQyX_;1Njn--7ei`7hP1=h4
z<+b<AwL!TOSZxZY&HZvYmustXIhT*$uV}kpv3b8jgB2??_ba*{?7d&HaKED8*7H5m
z%vhSB$wK!l8j2L<>Qk&ptL@9xw0yeI{qid|T2k-w%Z2W1lkyv-!kiiFiY-^3n!2v<
zmtQSZ$`VSkq(lu+UX(ghDx*l@02x=DSJ=wz{fZ)~J=Y4AHWbBLQl$cQw$S~GslQKX
z<-xXD)i32&lv8ZJUoloHxAq+Ne3LpqDDS#t*UwA3e!t)^%C)2(7tdF2*dV?7g3~NT
zLDpr2rpZdtT>rkz2TN$q8_S6;pDWU`JKQgyr-DT*pGT#$1WS4KrR_zoLzWh4sZdd9
zsRE^y(pf1LT>!dL%3UetE2WlraOzfVD9Uo_ly4L&wOes2&zEW~6_mxbv`ssKwb1u#
zS)BHYa;l(8s-kMDDO4)4sP+TG3a7rJuGG2(LPh=kie^y*HIf!8tvBC)(Bu^@wzQO*
zvdFu&PzlKSow4FQe0~A{`|3Wppo`?*TNNFJ&J%Z9?7Z(XcO@3}DKfBLoUijGtz6-K
zT6zn7|2iuw>@6`-DzyOVD(xwfb{D#@rkb?;(s?dDws{*@^cMM+@05-za__bKynVb4
zEBZ>USG1)!Wbr1`yrSZgD(Y;>>2X?$B^QH}R54Jfbd?m5^Iqh7zEsp@uu!RtqK-p_
zO0DK<DQzg7)`G=bqGGt%RdBmJ&sR#h4c{+Xy=ZYuF5w&JORD0M!B><sQt15H2OC^X
zrESIg@ci5qHFLpL_>l2_D&Cl-=<lzA|Mz>W!pBR+1WgtyffcnpUn(;4iBjPcrNSqQ
z&!?gam*_HGp`vKzs`i8PX{Fd+9I3rKWq7`%!JMjMb5WgZ=M@wc?c<$xo^#P01+<na
za<t^_TH#Gu;jLIIdr|B?R8gFf;{15qRBTXDinm9*)@`@F-L?)*)#+*)@ToCi|FAd8
zsNBa8k6E9nA>@;TB%jODT~WCiE#dk4{<XfiKdUXSJDE3KQghxYi};sacAih}d*-<B
zUiiLyibLObhZlQkEe^}7<FLFt4z=o`eu+aj0*AGi;;?=+4jap8F%Fyh;;^NQ#H|f+
z*ftY~ozyi%V`RCfhve706o>uWaX3&DhtrO^G#!UmuG4ZHUY(D_YYTCBqd5*23eD1L
z94;x}v;CoGd+tH1qk2-7=l4U;<cFTI4?SNWdY(S?{Cwz{`Ox$7q37j8&&Y?Kj}JW?
zA9{Y}I*dclh=-mH4?P?5A5te_@+^1gIqlH1*P-XEL(fi!o^Lo3sVxrOLm#@=J#=4t
z=sxz){o~>G2JOU=dxfKmGjZgb^T_w(k&FMxTk^;?c2s$lq|u*qR8>Vy)Jc;xMbhZA
zInrmt@kq0@5=Yh2SIf6rdN*lD)m_w0J*13k$5%_QPv%J9%TcvsYU-$w25Fc^$-Y{7
z)mBgqIlgv?lvk_1wHt9%S53;Pn<x3WqdKY|pb1)_#W-qkY=i9$>e3+Z2IV)%yJ3yw
z(}=B6K8@<xsEkJZjjJU8M*K}>R7uh`$)~A}v_q40P3qgEK20}Cc}?<d+9Jm_YlG(H
zIBHS<mRhPOj4dOiel6OtMLV?IpgD4Ut9rCb+lr~xxoE}PsxGbS+9tm?``Wa*_t{bV
zI4P?`8J(?h)Fpja2jS?Ff44e!WAE1H-SX*HR<|;`rSH+MJ@V_3N3SybF!jl|PkRk$
zg8^+gpv*z%ZP5ClIuBwTZiu6iW|D4H9Y@uBRNIfqZ%liPsn@uC#+}=7Wli*wV<(h7
zah>EbA^n7NHL3hbWlw6CNp+rdyw8H8Np+fZZB1S}o`o@~d8eREL7!61DNR2m$&~Cb
z$^Md}F01Zkhg`A!%5)rEbsBtx99>hDYuj-&t9fQ&%-VNdRj)gh*R9`hYHrBmhGTE2
z-VJ%)(A+nizMHCcQ<L75=S|yg+BP>z*5_P8b57r!W9Br$yr$6OeB>X79WBXoNqI|<
zm+W8KjH6}9%j<EpqMj>bB##xxuc*(8<5!%<74`ISaOACjq|50@m(tPN!f~yv^eP?c
zK|0cVbfnklNUzP29+@LOF-Lk}j`W-y={Y&lJ94CV<Va7*ksgdAy%a|qvegT5q_^Qn
zkHV3jf+J7iN1ncqJXs%kvOe+zedOu+XiI)RA&xxh9(jU2^5lBt3G~QQ=aHw!qaB!@
z43Bm+@6K8rdwMv&I2gxeOL1IYMxAl&w&&Olz_BmCV?EHvRh3l4jnqt|beY!UxZ1kM
zxnrLQ$JIAzK90TnkM*)1*DS}ecmHv%W3|C??Eqb;MY6B1fjVd?j_bQ<E{?s+j~iOZ
z_6Bun*re?^Zd9kntvGIyPm{Dw%5U0;<K_xdpXMuZ+@hW>RirMhm|Dl<xD8*MeQhfw
zpZ0!IUWc@u>f5O<-gU=a-E@ss<JddyxVwk?NM7AHNn7>EyGL2xamPIqG!@6a&Bybq
z(W|<>cJ<C4wQHq56yAr&{mA><NNxJ94^)%f28v}R4Qd@O)8j$uy+@A+wa}neA4E5#
zWrt)tq-H}}bVxD7YA`I_$O38oQOlzvbhXgYX3ODbKO=b@kHH>yI>#L_E+cQ~<H^oA
zo+^*yX;r`EfXkLIE8wbqGvXPAU(+nta9mS~S?On?dIulRI-RpCq#3R&@Ve7@T_vuo
z#B~)jZgYI2i5zo7KHhc5H!a^>jALK=$8(qBcpm8dP8=_2pap3cG~a@<b*>*TI%W~%
zqV$XEzohO<PTP`tE-80OW9e)^UQv#5pW_wVSJXwP__0pxW1YFjI%1F4HMB3$V;z#m
zIw6mBI3DXvJk~LHtRwE&Tjy9O+p*52W1T3+J2SK$Cl?23nAYN?te$4*Caus`ocLlr
z@f3FAhU%okvH_J7U#KTlb#y&Wst4mluf<7CYn-^$Ph9FJwU_BC$*XpWmT7}_;zaku
zNu6|c@~V?o7lg?GaiWXi#9QG+m%@pzgp&sA-U26$^6|xV(zG5Y&6VW%W@R@w&=k$b
zNsIdX@;Pa>-KBrhCQX}qx%^Ms=HjGXecSD8?~RiVj2-rMOvg#5V>)f?9HB{4?@sKU
z^6Xq7ZP2OSo!Y5OeY=#`Wq-G_yW8W$+u@`aOYacLqpzHl*@w4J+x2Uoe#`^P^rd9F
z16?B9y%A0Z%ShdQL7fcY9g@zM)XA_kBhrt^-vxg%rhUh><#;2t#mR(p6P72`ZSpE<
z8zY^?^?%}<=41-%lzRDv71Jt68B^*$r7qLToNgj{PPdS{Os|q-FUi-p(#a)dUDifd
z)ai<CSERe5zJ9g&<f`MZYSS6@oN?@obLV|?a$TAm(%;agH<W)<e!g)|=G1>qJ?FL8
zylo5O1@&6c77IAN4^G^4o-8_cQJohtFFOBA+Hy%7EJ?R4pJm6cIIk<}VZiCcP3eh2
zniDsfCk9<kbo!p?*gerfd!p0yWJ|qtmY(PoJ<$nzqI2`aSHj7*^P)5FRA<xa#VTr~
zc3Oy2H{YkaZBEN>&`sK=oj5J8qXwF#`8YLTd0J^-O)pK*Wx7iC*T|=4Gfr#euLti`
z&)sRQblw-IwIgw=tL;=*+o_wrQ{8K)uKm+`dDpkm5Y3T1>bK&wVTz{Xw9)oP<uoqG
zX_LA(b&&OD<uyy!qJD1BPTioLwhYoGQidC|(^mW2>~HId(+*{K*w!Kc4t4D8j?=Em
zIPGa7ZQ^F?v{(MU^6kalXM0~O$+J(MuI<ylnK<>~e%fz&U_4F-n+gAr?L%0G<uNR;
zk+C=()doJKPu*ahj;Y(2wjERNiE?Tob($C^X(zPBgySZaH|4k~^_tSgx(H9Fwa>J=
zUQ)M9j=wCwtMa<4&9ADbYw>i(z8Ps|)O|)?GmgC`pKIzgtNq<MonF^QH&^0xPM*fu
zPv^Qw+Ij6euMYE==5Z`IZb2Q5!Jis~KV8t?i(;SEr%UR!<l6Are7dY1m$jqM=F=7J
zy<)oos8hEyrv`aW4P=}eyf`%wacThK)a}Zt&h%5A=%+fsPYueS>hwOlI6<p%=JWSV
zH{n^?JZ;8Vc?-2sXPoI-IrI2`=7ILCVk^!JexFqi#F@|GGlR}&9!|~-2%ptdQzJRH
zMjE#qXSMS7Ywl-t@@(j#-Z*QljI$=?wXDTiYge3gt;AXPMQWf~vfevL*W;|uHrK*g
z-xA5IPd@$1=~wT5`SwfOZ+pMI1}bQr>>IdD_7B)UXnhdZP%WvKFZVNF?q@^N4q+Zq
zhR?jSkx^2<&%Lveoj7w1osCwJ{iEtVD(&b-oVi&)^WA<n;TSjRXA|<7xIyYRX}de`
zGuO=7w04=kMAzc%QaNeAOWSdFMVtBlKbw)~jN=W0pIw_Gd0bOZc8#;yezHENUFLe?
z%=h-$g0^3h-?DaHS&uV=(Psut&)nLc8MHiGmzTlGvvqYeSa@b2@XSEqneXE>gKTFT
zj(1yoW?=2iz}cDa=QFpqXPfG3K<vzQa%M2<%;3|Rfu=KqHfIKH&J4hu8F)D}@R9~E
zMH;#h>Do%9+4e9Yj>@Q<DyWjGsG4f1mg=aU8mN()sF_-*mD;GttAjeJi@K?YdZ~~4
zX@CZ4h=yr|)am9Jjnf28(iBaTy577@SLiCu&^4N+>xARxO`4;5TA)Q*qGiJ92`1fK
zqjlP#P1>Su+KJ@+q&fU^_~-D?;h)3reV*p<d#|TC{B!u{@Xz6&!|(nq&EcQJKZkz~
z{~Z1~{5~6!Un@(_dz!;Phu<%tr8)d__~-D?;h)3rdoIo4_jHx!@O#2ap0Ls!e)oTA
z4*wkfIs9|@=kU+r_XL*a@Xz7*B$oVlqU8HB&EcQJKZkz~zt5;Nhu?osNWM$c9R4}{
zbNJ`*&*PuRKaYPN|2+PA{PXzd@w@*_^Z4iS&*PuR@2M}%<DbVrkAEJ&@8~p-e;)rl
z{(1cK_~-G@<DbVrkAEJ&&&)KBe;)rlexILd9{)W4dHnPE=kd?upU3aBHO=Gq`I_eO
z&*PuRKaYPN|2+PA{PXzd@z3L*$3KsM9{)W4dHkL^(meio{O%```|9NBHTld=KC{yT
z{ssIC_!sc|3{O78(*k~8s>Dl`+=nOk;Yp`gTEM@6e*wRHljO5LE#P0k?>;3h;9tPM
zfPVr10{#X33-}lCFW_Imzkq)Mzx$fBfZu0DTEOofC%MN-3-}lCFW_Imud^*J;9tPM
zfPVr10{#X33-}lCFW_Im@BSz);9tPMfPVqM&!Mz{e*wRHr?iNlVJ9u(U&OzNe-Zy8
ze&1bb5&t6oMf{8S7x6FRU&QaeD=p$*#J`At5&t57&nJobCoSUFJCGLf`wmRZK#3VB
zd2UII_!sdn;&<<s7V$6QXB0|{_!sdn;$OtSh<_3PBK}4Ei}>B^rA7RU_!sf(Bu$=$
z(jxvv{EPS(@h{?E#J`At5&t57&r4|${}O)BP02lD(u0wf@Gs$C!oP%n3I7s)9=gOs
zmw4#X68<IpOZfG8q$T`I_?Pf6;a|e9BQ!1HU&6nHe+j>P(zJx%zgSO8_?Pf6;WvPp
zmhih@P3{rW68<IpzAMrae$RBtGhNcNl9uo<;n%~GmhiiuNK5#a@Gs$C!oP%n3BUV{
z<QXt6;a|efe3kUdq$T`I_?Pf6<6p+VjDH#bGJemDX&L`A{$>2j_?Pi7<6p+VjDH#b
zGX7=!%lMb^FXLavzl>jJby~*1jDH#bGX7=!%lQ49`Lv9G8Nd7Uw2Xfl|1$n%{LA>4
z@h{_F#=nez8UHf=W&F$dm+>#-U&g<TUr$qVPnCSPre*xg_?Pi7<6p+F7b-2|ch8lU
z@p~pn%lKFDui#(7zk+`S{|f#U{GJ)o3jP)REBM{3r4{@u_*d|+;9tSNf`0}73jP)R
zEBIINui#(7zk=WMchU=%crO$0W#YX|EBIIN>lsUW#?lJ@75sY0k{+_Of`0}73VzS-
zX$AiZ{uTTy_}w=q_l-$UT3W%sg5NVvTEV}9e+B;ve$V{L{YF~Bzk*+{Thi;6R`Bbr
zPde+9=c1&;J~84YM!dv`ml*L9BVO{nlyu@Jo%o3nFX_lnjChF=FEQdJM!dv`ml*NV
zDt<=1#E6#|@e(6m(&?Y{IV48Bw2Ge*FS++gtN0o55+hz>#7m5Li4iX`;w6Iwi4ia9
z8BC0Li4iX`;w47BWZ)n%;w60;i4iX`;w1wKi4iX`;w47BWKbb_#!Chkl7WT9h?f}g
z5+h#H)0i0Xl7WWAh?hJECPuu(h?f}gl4ruih?f}g5+hz>#7m5Li4iX`;-xkGjChF=
zFEQdJM!dv`ml*L9BVJ;}ON@Al5ic>~B?B3W5ic>~B}TktP$MzoB}Tl&h?f}g5+hz>
z#7m5Li4iX`;w47B#E6&H@H65iM!dv`ml*MqzMI5|ml*L9BVN*Lni%ntF4N?mGBM&M
zM!dv`ml*MqZq&qxmvp5jM!dv`m)v(I{XdBjFBxP>jChF=FBxn}23ry%URuY`h?f}g
z5+hzR0FxN;5+hz>#7m5Li4iX`;w47BWN;=i;w47B<eoM$;w47Bw2prrzrLolj(;7$
z``u(9Co$+H2EDY7-=I!%51e%MCI-F4pqCi*(mH<6N67$KV$e$leiDOT(jA-_^b&(!
z(k+}A^b&(!(mk9Q^pe4{#Gse-ZzcEDi9s(h=p_cd#Gsc993=+5#Gselb0-G9#GscL
z^pgIt#GsdSGbhhwN#9s9$dufdC->!vK`$}rCHLn^7j)7EojkiG2ED|fml*UCgI>}t
zof!0zKDET4ml*UCgI;3LOALC6K`(9KXV6Oquo8n_V$e$rdT9ecgI;3LOALBR_jY2?
zOALC6K`$}rB?i63pqCi*lCJN>pqCi*5`$i1&`X{j(gyww{2TZ^M<mY?X#+pwUNT^o
zHt{p=C0*vp^F?CZON@JoaW8G+XWUD=(i7ueV%$rfKa%H<#JHCXwk5{B<XI%?FHDSk
z$umjX#LvK&82AzcUovQ!82AzcU)sdKiQjWeV&qHu7?VE6#K@Nz`I6_~#K@O6@iX!z
z&%%k3FER2ZM!v+zml*jHBVS_VON@MpkuNdwB}TraPbV?*CC|@^kuNdwCC^8RkuNdw
zCH*~#kuNdwB}Tr)$d^1jB}Tr)$d|V8Z{gp<&(N1VQzeGJ#L$-*`qCEuE&PTQl76G)
zc`F%ONQ`}nu`g}mXY5N`_?i3C7Jdf5w1uC^FKyvx^h*XQli`Tu`93lFB}Tu*=$9D%
z5~E*Y^h<^*5~E)-RFN3{5~E+z7n`>5Gy0`1{Q9kuVT{D^m$vXT{iQAZjDJagR??r9
zw(v6mrY-zi__y#IAWh7Gi5W04114s`q~9wUG)>Hai5W04119}o$q-30M3OvLCeM|L
z889&eCT76I44Ag@GXp05Wr-Ou8A?gafQcC}=|4;QFOp}`#0;49V<gX{sdy%3229L=
zi5W04114s`#0;320TVM|G8B`T0TVM|Vg^ixWs<?%#0;49g(POc#0;320TVM|(l?Uy
zjU<M^#1NPm0uw`EGWeUA0+W8cw2j|DaN5DYgP&0_8R|)lf@uf8e(uC9n3x6A4t|Eg
zw1b~%Fc}U?JNS3-GY}>Q!o)zBcJS}u-@(sJm<$;uX2Qfwn3xIE4*nhdJNS3-@8D-J
zObmvJ!7%OMXEIDohKb2AF&QQ%!^C8mm<*GC_rz$JcJS}u-@(6we+NJFVcNmZe3%SV
zB?iRAfS4E%lOe05-!kpsXBtR5`1RkXot?-RFH#wmQw3F06;)FW)lwbRQv)?p6E#x{
zwNe|kQwMcY7j;t)^->@8(*O<95Dn7^jnWv6(*#Y@6iw45x=dH-D$UR}nx*S>gKp9s
z&C>!c(h@Dx3a!!_t<wf=(iUwK{xbY!_{;E@;rF98xeUMl++2pg41XDZ{cD-OF0=6E
zGW=|OxePxmUoOMX&X>#Zv-IUM{A_)>3_ojMF2m2>m&@?8_~kPEY<{^6zdqw!hM(Op
zm*FqN&-Rzg@U#BqGW`1Wav6T(E4d7R8U8Z-`kJ%8=FAqD_4{Skz|0z$4NGSBz-$~Q
zm*Hm-%#04X48Oj?Y-}d$pU(QHb2)xPn7JH(IsS6|<@n3-m*X$TuMayj6K4I`xg38v
z{&M{K7;`y({fxOBe>r~rjkz2@BT6pEUyk26P-Z~P<@n3-m*Zzf%;osY@t5N-$6t=0
z$1pP!=5qYy_{;Hg31%+A%n_K&@t5N-$6t=0uP&G4*SDUT<uXTH=7`HoZ<*;WvruJ*
zs?6t>`P?#3TV}e-JZ+h$Emz>@WXl!!EAVr&WlpwSf!~mBHg=VH*D}*u=2puU_$%=1
zQ_U6lIny#TT4vwMEN8g_KPy@05z7_$4GCvsUzxcivvOsAt<0sB^~YwOs?0By4H;*4
zsLZ&N*`YG)Q)Ye2%qW@RDKj=@hNsNMl)0EP_g1dJ&$5&mm9l=_%&3$Zm2w6C3j7uL
z8J04`Qf64n3`?0|DKji(eZZM%DOcb(nJSx1l})C~#wN1KRJjtr$yC{7s%$7bn@p7}
z@taJQO{U5wQ)QE>vdL80WU8z`Ih#zCEAg99l})J1mG~?1SK_b4Z-Q0U=bTNl$|hN5
zldN(jeiN;-alu@P-$biiiN6wmC4Q5xawYys{7gC7<g0A*RW|u5n|zg-oN^`pO8h2a
z<x2dO_$%?7kd-U(SK_b4Ux~jGe<l7({KjB1lTT*y$yNBP@K@nCY@dx$<|_Oqb7e#L
zSwDHM!e52oaDHY8%7*lF75*yxRrss$SK((A%2oKQ@S6mdO#;g%f#oXvRrss$SK+V1
zUxlA7Ae$VPO%BT@hh-+C%w&{J4$F*2nPng|8fE?Tnb|0t9F`f5vhmYwa#%JwESnsb
zO%BT@hh>w)vdLk&3O`d)uEJl1pY0%5;jhA9g}(}a6@I3rT#cV`DOck+sVtjRmaFks
z<2Si1SL3h7UyZ*SKZ8?dVaV0^*%&fcdS-aaT<O^aw9NRF8J{xaQ?ABud?9nIXKwXu
z%r{r#H%Tp5<FCeVvRbajUyYw7B3I+D#$S!U8h<r@li4!!RA!#aCbZ>h{Hzn%#I|f=
zTV|oiCbnhc7@4^$SL0`}%GLO*@mJ%o#$S!U8h<tZYWy|$Yw*|Lufbn~zXrdtkZf{X
zHaRXcY2_OHHTaE>WN!Ox+&UX4$;Pg;v65Va-{iSmgTDrU4So~pvT?p_GF`60UxS~~
zE1OuC**&t!b(!HS*Wj<gZ#*#9;IF}7gTDs9@xfe!-xy)8!C!;lSYfWgUxU8}zX^Jo
zbtD@<%r*EK$TAyAuEAe}zXpE|{u=zo7jq5%8vHf*SxvI>#>_038P2kC$6SlQ7C-Y@
zW<$wl5#(C@wfK!m=34x<_-pYSx6fu5WU~u$E&f{kwfJlCGp=Rhm$?>yEq>#exfVY&
zTdu`#Of%Qwuf<=BzZSn43b_`)SqhnXG4lgt#<$E9ka+@fE&f{kEHSwje=Yu6{I&RN
z@f)+sd;^(pAlKqIewAzSGtOl`f?SKg7Jn^%o`PJ5-}HuDhrbTLX%4v#KZ9N7H^_DP
zO?k+5_*rjq9e&2UT!+67KR-gQ!(WG=EhpFEuft!5zYc#Letv~qho5I5*WqW~$#wYa
z@Ymt5!_Ui*>+sj%uft!5zYc#L{yO}1`0Mc3;WzCf8w<|Hf^!{y(=M_x;arE`^owlz
zMK)%a83VI%;%uBa*Ws_jZyYbz;jhDAho5&Mo1T&D@z>)wO(UD8k?Zl-<2PL+GZ*H1
z{KlGdJ^p(9_4w=Y*W)+-oa^z|<FCihi;?T`*W<6pUyr{YKZ9bf$8W4U*W<6pUyr{Y
ze?5NY#axfS9)CUldi?eH>+#p)ugA~hkxd)Prj2A?k6e%6xOc9{Z|plWKIVG-_4w=Y
z*W<6pUyr{Ye?5NV<GCJxJ^lv#4fq@IH{fr;-+;dXKLcfMz|UTl8}K*aZ@}MxzX5*(
z{s#OF_#5yy;Ag+e4fq@I8!ycb_#5yy;BUa+fWHAhLuPKkZ`w;X?IoM`l1+Q@qm+c-
z*nMum-+;dXzv(d9beL>9Og0@RH{fr;-+;dXe*^vo{0;b-JaYs72K){989p<^XKujH
z7n2+Cvxnsd{EhgTgL5PPM*PNhGh=9G49)Cgxe<RO{zm+b_#5#z;%~&?h`$kkBmPGG
zjrbe!GnD2={7j{}5q~3o=F;4VzY#xALT<$0h`$kkBmPGGd<wY{e<S`z{Ehe<@i*e<
z<;jis8}T>dZ^Yk-zY%{U{zm+b_?et@BYu8|+=!o{H8<jK#Lw888}YNm<wpEWuDJ<6
zqib%$--MsxH8Z^CCj3qKoA5KgX5-zt34as*Cj3qKoA9&NWyaXdTakGyGJ|YpwaZQT
zoA5W`XS&Wzv$+XB<7{ri--N#je-r*D{7v}zGIA6CCj3qKoA9&jWk%cFgue-Y6aFUr
zP57JeGvMYX{7v{7adQ)Xo~PV|zX^X6{wDl9P?_~FH{oa0%}w~3b#pWRX8cUMnJqB$
zN#$nz&G`AHax;FOsoad8Zz?zAZ^qw@zZris{$~8m_?z)J<8Q{_jK3LwGyZ1$&G?(~
zH{)-{-;BQ*KO16h#@~#;8Gkc=-mKh=zZrisem2G2jK3LwGk)Hc+>E~&e>47O{JdPb
z8GkeWX8a7uxfy>m{$~8m_?z)J<L7(H&G?(~x8QHV--4fEIkQ7%cF5d<zXd-7bLNrB
zJTkcjKc7r)!QX<v1%C_v7W^&vTkyBwXRyyL_*?L|;AghaE%;mT^VsAT{CsDbAv(9<
zZ^7S!pE)|SVP*#D+=8DMC%52l!QX<PX*##yZ^7S!pHC;Z;OEuJE%+I!a|`|!{4Mxf
z@VDUS)5tCOTkyBwZ^7S!zXg9Q{#N|0_*?O};%~*z6PH`@Gi7JqxZH~0gn`_OpGiBn
z;%~*@ioX?qEB;pet@wFDax4B;{H^#~@weh{#m{1znY(i<{#N|0_*?Pw-sM*Ot@vB<
zx8iTb--^E#e=Gi0{H^#~@weh{#ovm*6@M%KR{X8_Tk*HzZ^h3<o|(vVEB;pet@!zx
zax4B;{B8K#@VDV_!{3I#4SyScUclUjp9d<p;cvs=hQAGe8~!%@ZTQ>pv&`l;{B8K#
z@VDV_!{3I#4SyT{HvGJWxeb3C{x<w=_}lQe;cvs=hQAGe8~!%@Oz)W&F}LA2StGaM
zZ^Pe)zYTvIem=$AhQAGe8~!%@ZTQ)Da~u9P{B8K#@VDV_!_U^6+wiyJZ^z$`za4)&
zexAqNj=vp$JAVGh+>XB;KcjAL$KQ^>9e+FicKq%5+wr&KZ^z$`za2l{WNydLLY&+2
zx8rZe&&Z$K@wek|$IsNC+wr&KZ^z$`za4)&{&xKB_}lTf<8Q~`j=vp$JN|b3?fBdA
zx8rZe-;Tc>e>;8?b8<U=lXG%Aev=1sJN|b39r!!&ci`{9&*Gdr@OR+vz~6zt13&L)
z?!e!HzXN{<{to;d_&e}-;P1fSfxiQP2mTKH9r!!&ci`{9&p(<w@OR+vz~6zt1Ahnp
z4*VVXJMi<E<_`QF_&e}-;P1fSfxiPkKVI&@-+`YmFL&VQ&C4D5JMfzvkvs5r;P1fS
zfxiPk?_Tb}-+{jae<%J<{HE9DPW+wtJMr_k=1%;b_&f1;;_t-YiN6znC;m?So%lQP
zcjE8F--+Mkr_3LiJMr_#=1%;b_&f1;;_t-YiJx~icjE8F--*8yKQC?W#NUa(6F*;V
z?!@1T-$bR{iN6znC;m?SJh!<MKi_TUyUncWxf6dUejeQ1iN6znCw_k1+=;&vKVNR{
z#NUO#3x5~>F8p2iyYP46@50}OzYBjC{x1Ao_`C3T;qSuVg})1b7yd5%UHDBz$*l3Y
z3x5}W{@>h%zYBjCev_?o7yd5%UHDDB%3b)o@OR<o70zAwyYP46@50}OzY9OReeS~F
zg})1b7yd5%UHH53cj52C--VwqFL&YZ!rz6z3x5~>F8p2iyYTbv<!=1l_`C6&Jd?Ze
zcjIU2&)xXD@pt3r@5|lzyYYA9@5bMazZ*YKbMD69jlUazH-6IrayNcc0&+M0Zv5T&
zP0GpL_`C6U<L}1bjlUbesROwie>eVa{N4Dw@pt3z#@~&f$1r!}@5XPMLGH%ijo&1n
z+>O5*zsWwC4>5P+Hw__o<L}1bjlUazH~wz?J@|X@n-r9L@b}<1Wg++A@4?@LzXyL0
z{vP~2_<Qj8;P1iTgTDuV5B?tfJ@|X@_u%ir--Evge-Hj1{5|-4@b}>F!QX?w2R}b}
zHa#Nu;P1iTgTDuV5B?tfJ@|X@_u%ir--Evge-Hj1{5|-4@b}>F!QX?w2Y(O#9{fG{
zd+_(*@4?@LzXyLW{$BjO_<Qm9;_t=J&zpPk_u}ux-;2K&e=q)C{Jr>l@$<0fUi`ej
z*?f-Ni@z6tFaBQqz4&|a_u}ux&lj3`-g7U0zV~eYNAAVni@z6tFaBQqz4&|a_u}ux
z-;2K&e=q)C{Jf>P7k@AQUi`iId-3<;H*qod;_t=Zi@z6tFaBQqz4&|a_u}VM&VBg%
z@blm2KKy<7`|$VS@5A4R-^7&Mho6@<_u=Pf&3*X$@bfrllgM%({yzMD`1|np;qSxW
zhrbVhAO1f4{Lr}%e;@un{C)WQ@b}^G!{3L$4}TwizUgdoTIQY3efasOb07Xb{C)WQ
z@b}^G!{3L$55EaCxevd21i24C&u#9*--o{sKmTnuzaaPF@5A4ZzaM`;{(k)Z`1|qm
z=;nU>{Mxx6e?R_y{Qda*@%Q8J$KQ|NWS!iPzaM`;{(k)Z`1|qq<L}4ckG~(kNj<qA
ze?R_y{Qda*@%Q8J$KQ{?AAdjoe*FFT`NeZT{(k)Z`1|pjACdd<_v7!!-;ci^zX?OR
zA3tw#<}J>=#hJG__v7!!-;ci^zlp8c{E9q)e*pgg{sH_0_y_Qtbd(4158xlbKY)J#
z{{a30`~&z0@DJc0z(0V00RI5~0sI5_2k;NzAHY9=e*pgg{sH_0_y_P0;2*%x_ni5j
z^8o$<`~&#S_sD$Ec>w<a{sH_0_y_Qt2a*Tyn;4Y`@DJc0z;A9y9>C8#od@unFqH@J
z58xlbKY)J#{{a30{3cUnb4l_b{z3eM_y_S1;vd9sqE$B0Di7ix#6O6C5I^sJ9>hP0
ze-Qs5{z3eM_y_TuK$%U>%7gd^@ekrRQ7aGPAH;9&N;Y>T58^j@D-Yu5^Uj0#2k{T$
z=T*+;2joHggZKyW58@xhKZt)2zsZ?-5I=u-9>hP0e-Qs5{z3eM_y_S1;vd96h<_0O
zApSx8L->dA58)reKZJh>{}BEm{5<D*2>%fNA^b!5hwz(dmWS{U;pa)uL-<WR%R~5w
z@bjqWA^b!5hwu;KAHqL`e+WPSdLF_*gntM>FMA%sKZJh>{}BEm{6qM8-188A^Ca>R
z{vrHB`1!c=5dI<jL->dA58)reKZJh>{}6t2EAkM2li~ck6T&})e+d6DeiPL5F#ciu
z!}v{@%jRO_Vf@4Rhw%^NH@_$k;~&O9jDHyaF#ciu!}y2s591%kKaAg`yF84482>Q-
zVf@4Rhw%^NAI3k7e;EHT{$c#X_=oWi<2RQn591%kKa77E|1kbx{KNQ%@eku4#y^aI
z82>Q-Vf@4Rhw%^NAI5K9Ngl>OjDHyaF#ciuBlt(~kKiA{KZ1V*{|NpO{3G~B@Q>g(
z&mfQBAHi=ja2~-wf`0`62>ucLBlt(~kKiA{KZ1V*{|J6_6!HlE5&R?gNAR1okj;0=
z=4<5<{3G~B@Q>gh!9Rk31pf&B5&R?gNAQo}AHhF@e+2&se)Ar(3Cei{{|NpO{3G~B
z@Q>gh!9Rk31pf&B5&Wb0&6CKZ_($=N;x}g^kK!N2KZ<`8|0sU*DDo(N6QJ`b{!#p+
z_($=N;vdC7ihmUUD1H+yvw0VJ6#ppxQT(I$NAZv1H@_#3;vdC7ihmUUDE?9WqxeVh
zkK#9PBah-A#XpLF6#poGKdO*N@tcgDO~%fn_($=N;vdC7ihmUUDE?9W=6>W+{G<3s
z@sHv+zbKF4AHzR}e+>T^{xSSx_{Z>%;UB|4hTmMIJci%Ax;%z|4F4E@b4T+S{xSSx
z_{Z>%;UB|4hTkOdJcfS^{}_IAQ1Tdl^P2J){xSUKH{~(>WBAALkKrG~KZbt{{}}!;
z{A2jX@Q>ji!#{?94F4GZG5llr$MBEgAHzR}e+>T^{xSSx_{Z>%;UC9u&TAgWKaPJK
zzxl6u9RE1}as1=>$MKKjH;*un<2OGxkK-T5KaPJK|2Y0}{Nwn?@sHyl$3KpL9RE1}
zas1=>$MKKjAICqAe;off{&D={_{Z^&<2O$ykK-T5KaPJK|2Y0}{Nwn?@sHyl$8X|&
z9>+h9e;off{&D={_{Z^&;~&RAj(;5gIQ|Lz6Zp+5%M<t~@K4~Mz(0Y10{;Yl^Ct5I
z{t5gO_$TmB;Ge)hfqw%31pW#96Zj|ao1>N|@K4~Mz;C`$p1?nWe*(X`M|lFjIc(V+
zq&$It0{;a53H%fIC-6_;pTIwXe**sm{t5i%EM-3|kSFl_ae+L6e**sm{t5hkY#^H(
zmnZO>AD7Mf%;q@d3H%fIC-6_=pTzG+3GyWVN&J)e{XjvU#6O9D68|LrN&J)eC-G0>
zpTuv@U7o~0iGLEm`O?{s8stg*llUj`PvZ9@2YC{|A3DgB_$TpC;-ADniQkVO<VpOK
z_$TpC;`d_+*?g>QK2|nYHBaK7#6O9D68|LrN&J)eC-G0>pTs|je-i&B{z?3k_|0?8
zllaYb&6D^i@lWEP!f(!CHpe?p;h(}kg?|eF6#gmvQ~0OwPvM`!KZSn^{}g_6WwT#r
z$W!>I@cWg9JcZx<**t}R3jY-TDg0CTr|_Fso2T$k;h(}kh2Jkb<SG2-E@ty>^A!Fm
z{8RX+@K52N!as$73jY-TDg0CTr|?hV_sbA@3jY-TDg0CTr|?hVpTa+de+vH;e)A)<
zIlXxrzj?iR8viu@Y5aa(Et@-;r}0nYpT<9p-(1Q(jei=yIl$Q*;5?0g8viu@Y5ddp
zr}3LVm#6Vh|36jTOKxS2x@Kjv7Zhm&P0SQo@TC9<K>@@$=ifVgw;?Hk5DnrRujD51
z?1Y4*;2B;aYel@F2zz<XIi9dzM|Xeh|6~6j`~TSg$NoR||FQp%{eSG&*xw)f|JeV>
ze(l};vHy?#f9(HbzZUQQ*sndlKlcBz|BwBD?Eho`AN&8<|HuA6_W!Z}kNtn_|6@Oq
z`Tp3iS-wB^zu5m`zrOEY?0>QU#r_xjU+jOe|HXcN^S#*rV*iW%FZRFK|6>1({rb{-
zvH!*X7yDoAf3g3?{uldS?0>QU#r_xjU+jOe|Hb|n`(NzWr{0VGFZRFK|6>1({V(>v
z*#Bbxi~TS5zu5m`|BL;)#Cx&-#r_xjU+jOe|Hb|n`(NyTvH!*X7yDoA*D>CU{d(AY
zwV!KnulB#%|7!oM{jc`F+W%_*tNpL`zuNz5|EvA4_P^TyYX7VKT!eeI|JD9i`(N#U
zwg1)rSNmV>*M{G#{jc`F+W%_*tNpL`zuNz5|EvA4_LEKS)&5ueU+sUj|JD9i`=Lqq
zYX7VKulB#%|7!oM{jc`F+W%_*tNpL`zuNz5KXm(E?SHlZ)qb7&z1ja}|C{}9_P^Qx
zX8)W0Z}z|0|7QQ2{crZa+5cw$oBeP0zuEt0|C{}9_P^QxX8)W0Z}z|0|7QQ2{crZa
z+5cw$oBeP0zuEt0|C{}9_P^OrbGbMB-|T<0UvqkI_P^QxX8)W0Z}z|0|7QQ2{crYb
z?eES0H~Zi0f3yG1e*OKu+5cw$oBeP0zuEt8|GWL~_P^WzZvVUe@Akjj|8D=g{qOd_
z+y8F=yZ!I>zuW(A|GWL~_P^WzZvVUe@Akjj|8D=g{qOd_+y8F=yZ!I>zuW(A|GWL~
z_P^WzZvVUe@Akjj|8D=g{qOd_+y8F=yZ!I>zuW(A|GWL~_P^WzZvVUe@Akjj|8D=g
z{qOd_+y8F=yZ!I>zuW&||A+k__J7#_VgHByANGIP|6%`!{U7#!*#BYwhy5S+f7t(F
z|A+k__J7#_VgHByANGIP|6%`!{U7#!*#BYwhy5S+f7t(F|A+k__J7#_VgHByANGIP
z|6%`!{U7#!*#BYwhy5S+f7t(F|A+k__J7#_VgHByANGIPuV+8b!hP8PVgHByANFh9
z@5BC2`?c=(Y5%AFER*}R|I_|Y`)NY=X+QM)KJ5p8->3cX@B6g>)BaEUHPH8I|EK+*
z_J7*{Y5%AFK(qU_|I_|Y`#<gfwExq7{=;!o?$iEH`#<g1Qs1ZjpZ1fC?$iEH`#<gf
zwExroPy6-O_i6v9{krP|lHI5MpZ0&+uW`Ol`*r;HX+Nm^wAlwMKRVKV+W%$$m;GP%
zf7$<K|Cjw=_J7&`W&fA`U-onNkGp?g_J7&`W&fA`U-oOM@5_Gv%YE7ZW&fA`{DAwi
zUk`p?_J7&`W&fA`U-p05|7AZj=DzIzvj59|V$*%u|7HJ|{a^Nf+5ct#m;GP%>&x%U
z{xAE#?EkX=%l<F>zwBqv+?W0QnftQ;%l<F>zwH0A|I2>1!F}8RZU49Z-}Y0Y?%V!v
z`@ik~wx54+-}Zmo|84)b{oIcGw*TAyZ~MRP_kR-HxBcJtlcw(5{%`xg?f<s_+kUX!
zecR8(xo`Ww?f<s_+x~C+zwQ6F|J(j=`@ik~w*TAyZ~MRPXX)Ix{onR~+y8ApTj##*
z|F-|z{%`xg?f<r)xpT~&`?mkv{%`xg?dOi%kNrRPQ?c&H{vZ2UB==+gkNrRP|Jcv#
zxgYy~?EkU<$NnGtf9(IU|Hpoo&;8i{WB-r+l&$-*|HppD&;8i{WB-r+KlcCF|6~7;
z{Xh2q*#BeykNrRP|JeUy|BwAY_W#)bV?VX)e(e9T|Hu9x`zc=cWB-r+KlcCFPxZPV
z`+w~JvH!>ZANzmo|FK{HfBOH^|DXQ<^#7;-KmGsd|4;va`v24apZ@>!|EK>y{r~Cz
zPyc`V|I`1U{@;J0>&<@s|LOlv|9|@b)Bm6T|MdT-|3Cfz>HkmvfBOH^|DXQ<^#7;-
zKmGsd|4;va`v24apZ@>!|EK>y{r~CzPyc`V|I`1U{{Qs<r~g0w|LOlv|9|@b)Bm6T
z|MdT-|3Cfz>HkmvfBOH^|DXQ<^#7;-KmGsd|4;va`v24apZ@>!|EK>y{r~CzPyc`V
z|I`1U{{Qs<r~g0w|LOlv|9|@b)Bm6T|MdT-|3Cfz>HkmvfBOH^|DXQ<^#7;-KmGsd
z|4;va`v24apZ@>!|EK>y{r~CzPyc`V|I`1U{{Qs<r~g0w|LOlv|9|@b)Bm6T|MdT-
z|3Cfz>HkmvfBOH^|DXQ<^#7;-KmGsd|4;va`v24apZ@>!|EK>y{r~CzPyc`V|I`1U
z{{Qs<r~g0w|LOlv|9|@b)Bm6T|MdT-|3Cfz>HkmvfBOH^|DXQ<^#7;-KmGsd|4;va
z`v24apZ@>!|EK>y{r~CzPyc`V|I`1U{{Qs<r~g0w|LOlv|9|@b)Bm6T|MdT-|3Cfz
z>HkmvfBOH^|DXQ<^#7;-KmGsd|4;va`v24apZ@>!|EK>y{r~CzPyc`V|I`1U{{Qs<
zr~g0w|LOlv|9|@b)Bm6T|MdT-|3Cfz>HkmvfBOH^|DXQ<^#7;-KmGsd|4;va`v24a
zpZ@>!|EK>y{r~CzPyc`V|I`1U{{Qs<r~g0w|LOlv|9|@b)Bm6T|MdT-|3Cfz>Hkmv
zfBOH^|DXQ<^#7;-KmGsd|4;va`v24apZ@>!|EK>y{r~CzPyc`V|I`1U{{Qs<r~g0w
z|LOlv|9|@b)Bm6T|MdT-|3Cfz>HkmvfBOH^|DXQ<^#7;-KmGsd|4;va`v24apZ@>!
z|EK>y{r~CzPyc`V|I`1U{{Qs<r~g0w|LOlv|9|@b)Bm6T|MdT-|3Cfz>HkmvfBOH^
z|DXQ<^#7;-KmGsd|4;va`v24apZ@>!|EK>y{r~CzPyc`V|I`1U{{Qs<r~g0w|LOlv
z|9|@b)Bm6T|MdT-|3Cfz>HkmvfBOH^|DXQ<^#7;-KmGsd|4;va`v24apZ@>!|EK>y
z{r~CzPyc`V|I`1U{{Qs<r~g0w|LOlv|9|@b)Bm6T|MdT-|3Cfz>HkmvfBOH^|DXQ<
z^#7;-KmGsd|4;va`v24apZ@>!|EK>y{r~CzPyc`V|I`1U{{Qs<r~g0w|LOlv|9|@b
z)Bm6T|MdT-|3Cfz>HkmvfBOH^|DXQ<^#7;-KmGsd|4;va`v24apZ@>!|EK>y{r~Cz
zPyc`V|I`1U{{Qs<r~g0w|LOlv|9|@b)Bm6T|MdT-|3Cfz>HkmvfBOH^|DXQ<^#7;-
zKmGsd|4;va`v24apZ@>!|EK>y{r~CzPyc`V|I`1U{{Qs<r~g0w|LOlv|9|@b)Bm6T
z|MdT-|3Cfz>HkmvfBOH^|DXQ<^#7;-KmGsd|4;va`v24apZ@>!|EK>y{r~CzPyc`V
z|I`1U{{Qs<r~g0w|LOlv|9|@b)Bm6T|MdT-|3Cfz>HkmvfBOH^|DXQ<^#7;-KmGsd
z|4;va`v24apZ@>!|EK>y{r~CzPyc`V|I`1U{{Qs<r~g0w|LOlv|9|@b)Bm6T|MdT-
z|3Cfz>HkmvfBOH^|DXQ<^#7;-KmGsd|4;va`v24apZ@>!|EK>y{r~CzPyc`V|I`1U
z{{Qs<r~g0w|LOlv|9|@b)Bm6T|MdT-|3Cfz>HkmvfBOH^|DXQ<^#7;-KmGsd|4;va
z`v24apZ@>!|EK>y{r~CzPyc`V|I`1U{{Qs<r~g0w|LOlv|9|@b)Bm6T|MdT-|3Cfz
z>HkmvfBOH^|DXQ<^#7;-KmGsd|4;va`v24apZ@>!|EK>y{r~CzPyc`V|I`1U{{Qs<
zr~g0w|LOlv|9|@b)Bm6T|MdT-|3Cfz>HkmvfBOH^|DXQ<^#7;-KmGsd|4;va`v24a
zpZ@>!|EK>y{r~CzPyc`V|I`1U{{Qs<r~g0w|LOlv|9|@b)Bm6T|MdT-|3Cfz>Hkmv
zfBOH^|DXQ<^#7;-KmGsd|4;va`v24apZ@>!|EK>y{r~CzPyc`V|I`1U{{Qs<r~g0w
z|LOlv|9|@b)Bm6T|MdT-|3Cfz>HkmvfBOH^|DXQ<^#7;-KmGsd|4;va`v24apZ@>!
z|EK>y{r~CzPyc`V|I`1U{{Qs<r~g0w|LOlv|9|@b)Bm6T|MdT-|3Cfz>HkmvfBOH^
z|DXQ<^#7;-KmGsd|4;va`v24apZ@>!|EK>y{r~CzPyc`V|I`1U{{Qs<r~g0w|LOlv
z|9|@b)Bm6T|MdT-|3Cfz>HkmvfBOH^|DXQ<^#7;-KmGsd|4;va`v24apZ@>!|EK>y
z{r~CzPyc`V|I`1U{{Qs<r~g0w|LOlv|9|@b)Bm6T|MdT-|3Cfz>HkmvfBOH^|DXQ<
z^#7;-KmGsd|4;va`v24apZ@>!|EK>y{r~CzPyc`V|I`1U{{Qs<r~g0w|LOlv|9|@b
z)Bm6T|MdT-|3Cfz>HkmvfBOH^|DXQ<^#7;-KmGsd|4;va`v24apZ@>!|EK>y{r~Cz
zPyc`V|I`1U{{Qs<r~g0w|LOlv|9|@b)Bm6T|MdT-|3Cfz>HkmvfBOH^|DXQ<^#7;-
zKmGsd|4;va`v24apZ@>!|EK>y{r~CzPyc`V|I`1U{{Qs<r~g0w|LOlv|9|@b)Bm6T
z|MdT-|3Cfz>HkmvfBOH^|DXQ<^#7;-KmGsd|4;va`v24apZ@>!|EK>y{r~CzPyc`V
z|I`1U{{Qs<r~g0w|LOlv|9|@b)Bm6T|MdT-|3Cfz>HkmvfBOH^|DXQ<^#7;-KmGsd
z|4;va`v24apZ@>!|EK>y{r~CzPyc`V|I`1U{{Qs<r~g0w|LOlv|9|@b)Bm6T|MdT-
z|3Cfz>HkmvfBOH^|DXQ<^#7;-KmGsd|4;va`v24apZ@>!|EK>y{r~CzPyc`V|I`1U
z{{Qs<r~g0w|LOlv|9|@b)Bm6T|MdT-|3Cfz>HkmvfBOH^|DXQ<^#7;-KmGsd|4;va
z`v24apZ@>!|EK>y{r~CzPyc`V|I`1U{{Qs<r~g0w|LOlv|9|@b)Bm6T|MdT-|3Cfz
z>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq
z|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq
z)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ
z|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJ
zr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c
z|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUc
zPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnx*Y@lG
z)BmUcPye6(zkhAN{y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ
z^#AGq)BmUcPye6(zkg%D{y+VH`v3I*>HqsT_Ur%C|EK>?|DXOp{eSxZ^#AGq)BmUc
zPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?
z|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?
zpZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v)
z{y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(
zKmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp
z{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7n
zfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH
z`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D
z|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ
z^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ
z|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*
z>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq
z|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq
z)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ
z|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJ
zr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c
z|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUc
zPye6(KmC9D|MdUq|9i0i!TtyPAMAgy|H1wT`ycFou>ZmS2m2rFf3W|-{s;RX?0>NT
z!TtyPAMAgy|H1wT`ycFou>ZmS2m2rFf3W|-{s;RX?AQON|4;v){y+VH`v3I*>HmAM
z|H1wT`ycGr|EK@&!TtyPAMAgy|H1wT`ycFou>ZmS2m2rFf3W|-{s;RX?0>NT!TtyP
z_5bPr)BmUcPye6(KmC9D{~qmswExlmNBi~v>HmAQU;p2u{g3uP+W%<(qy3NeKidCj
z|D*kn_CMPHX#b=AkM=*>|7icC{g3uP+W%<(qy3NeKidCj|D*kn_CMPHX#b=AkM=*>
z|7icC{g3uP+W%<(qy3NeKidCj|D*kn_CMPHX#b=AkM=*>|7ib{{ZIBk+5cq!ll@Qj
zKiU6e|C9at|MdUq|9i6k$^Iw%pX}HFr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ
z^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ
z|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*
z>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGqd$#}C{%8B2?SHoa+5Tty
zpY4CP|Ji>1fBOHP?SHoa+5Tty_5bPr)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH
z`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D
z|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ
z^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ
z|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*
z>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq
z|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq
z)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ
z|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJ
zr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c
z|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUc
zPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?
z|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?
zpZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v)
z{y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(
zKmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp
z{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7n
zfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH
z`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D
z|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ
z^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ
z|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*
z>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq
z|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq
z)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ
z|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJ
zr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c
z|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUc
zPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?
z|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?
zpZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v)
z{y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(
zKmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp
z{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7n
zfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH
z`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D
z|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ
z^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ
z|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*
z>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq
z|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq
z)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ
z|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJ
zr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c
z|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUc
zPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?
z|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?
zpZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v)
z{y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(
zKmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp
z{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7n
zfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH
z`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D
z|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|DXOp{eSxZ
z^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIa
z#r}V>U;m%}KmC9D|MdUq|I`1c|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ|LOnJ
z|EK>?|DXOp{eSxZ^#AGq`^)~n?AQON|4;v){y+VH`v3I*>HpLJr~gm?pZ-7nfBOIQ
z|LOnJ|EK>?|DXOp{eSxZ^#AGq)BmUcPye6(KmC9D|MdUq|I`1c|4;v){y+VH`v3I*
z>HpLJr~gm?pZ-7nfBOIQ|LOnJ|EK>?|L^VJe_6r+gaHTx5C$L&Kp2290AT>a0E7Vu
z0}uuv3_uuwFaTiy!T^K;2m=rXAPhhlfG_}I0Kx!-0SE&S1|SST7=SPUVF1DagaHTx
z5C$L&Kp2290AT>a0E7Vu0}uuv3_uuwFaTiy!T^K;2m=rXAPhhlfG_}I0Kx!-0SE&S
z1|SST7=SPUVF1DagaHTx5C$L&Kp2290AT>a0E7Vu0}uuv3_uuwFaTiy!T^K;2m=rX
zAPhhlfG_}I0Kx!-0SE&S1|SST7=SPUVF1DagaHTx5C$L&Kp2290AT>a0E7Vu0}uuv
z3_uuwFaTiy!T^K;2m=rXAPhhlfG_}I0Kx!-0SE&S1|SST7=SPUVF1DagaHTx5C$L&
zKp2290AT>a0E7Vu0}uuv3_uuwFaTiy!T^K;2m=rXAPhhlfG_}I0Kx!-0SE&S1|SST
z7=SPUVF1DagaHTx5C$L&Kp2290AT>a0E7Vu0}uuv3_uuwFaTiy!T^K;2m=rXAPhhl
zfG_}I0Kx!-0SE&S1|SST7=SPUVF1DagaHTx5C$L&Kp2290AT>a0E7Vu0}uuv3_uuw
zFaTiy!T^K;2m=rXAPhhlfG_}I0Kx!-0SE&S1|SST7=SPUVF1DagaHTx5C$L&Kp229
z0AT>a0E7Vu0}uuv3_uuwFaTiy!T^K;2m=rXAPhhlfG_}I0Kx!-0SE&S1|SST7=SPU
zVF1DagaHTx5C$L&Kp2290AT>a0E7Vu0}uuv3_uuwFaTiy!T^K;2m=rXAPhhlfG_}I
z0Kx!-0SE&S1|SST7=SPUVF1DagaHTx5C$L&Kp2290AT>a0E7Vu0}uuv3_uuwFaTiy
z!T^K;2m=rXAPhhlfG_}I0Kx!-0SE&S1|SST7=SPUVF1DagaHTx5C$L&Kp2290AT>a
z0E7Vu0}uuv3_uuwFaTiy!T^K;2m=rXAPhhlfG_}I0Kx!-0SE&S1|SST7=SPUVF1Da
zgaHTx5C$L&Kp2290AT>a0E7Vu0}uuv3_uuwFaTiy!T^K;2m=rXAPhhlfG_}I0Kx!-
z0SE&S1|SST7=SPUVF1DagaHTx5C$L&Kp2290AT>a0E7Vu1JJwu@Akjj|8749APhk7
z_P^WzZvVUe@Akjj|8D=g{qOd_+y8F=yZ!I>zuW(A|GWL~_P^VY0SE&S1|SST7=SPU
zVF1DagaHTx5C$L&Kp2290AT>a0E7Vu0}uuv3_uuwFaTiy!T^K;2m=rXAPhhlfG_}I
z0Kx!-0SE&S1|SST7=SPUVF1DagaHTx5C$L&Kp2290AT>a0E7Vu0}uuv3_uuwFaTiy
z!T^K;2m=rXAPhhlfG_}I0Kx!-0SE&S1|SST7=SPUVF1Da^kM&p{U7#!*pC4S0}uuv
z3_u_Df7t(F|A+k__J7#_VgHByANGIP|6%`!{U7#!*#BYwhy9=Sf7<_P|EK+*_J7*{
zY5%AFpZ0&+|7riH{h#)K+W%?)r~RMyf7<_P|EK+*_J7*{Y5%AFpZ0&+|7riH{h#)K
z+W%?)r~RMyf7<_P|EK+*_J7*{Y5%AFpZ0&+j{yh+(5L;M_J7*{X+H)a3_uuwFaTiy
z!T^K;2m=rXAPhhlfG_}I0Kx!-0SE)om;GP%f7$<K|Cjw=_J7&`W&fA`U-n}F!T|JT
z|Cjw=_J7%r0SE&S1|SST7=SPUVF1DagaHTx5C$L&Kp2290AT>a0Q6=5m;GP%f7$<K
z|Cjw=_J7&`W&fA`U-p05|7HJ|{a^Nf+5ct#m;GP%f7$<K|Cjw=_J7&`W&fA`U-p05
z|7HJ|{a^Nf*^dDT0}uuv3_uuwFaTiy!T^K;2m=rXAPhhlfG_}I0Kx!-0SE&S1|SST
z7=SPUVF1DagaHTx5C$L&Kp2290AT>a0E7Vu0}uuv3_uuwFaTiy!T^K;2m=rXAPhhl
zfG_}I0Kx!-0SE&S1|SST7=SPUVF1DagaHTx5C$L&Kp2290AT>a0E7Vu0}uuv3_uuw
zFaTiy!T^K;2m=rXAPhhlfG_}I0Kx!-0SE&S1|SST7=SPUVF1DagaHTx5C$L&Kp229
z0AT>a0E7Vu0}uuv3_uuwFaTiy!T^K;2m=rXAPhhlfG_}I0Kx!-0SE&S1|SST7=SPU
zVF1DagaHTx5C$L&Kp2290AT>a0E7Vu0}uuv3_uuwFaTiy!T^K;2m=rXAPhhlfG_}I
z0Kx!-0SE&S1|SST7=SPUVF1DagaHTx5C$L&Kp2290AT>a0E7Vu0}uuv3_uuwFaTiy
z!T^K;2m=rXAPhhlfG_}I0Kx!-0SE&S1|SST7=SPUVF1DagaPpOf2ysc0Yn3c1`rJ(
z8bCCFXaLawq5(t$hz1Z1AR0h4fM@{G0HOgz1BeC?4ImmoG=OLT(Ey?WL<5Kh5Dg$2
zKs1180MP)V0Yn3c1`rJ(8bCCFXaLawq5(t$hz1Z1AR0h4fM@{G0HOgz1BeC?4Immo
zG=OLT(Ey?WL<5Kh5Dg$2Ks1180MP)V0Yn3c1`rJ(8bCCFXaM~?`)L5t0HOgz1BeC?
z4ImmoG=OLT(Ey?WL<5Kh5Dg$2Ks1180MP)V0Yn3c1`rJ(8bCCFXaLawq5(t$hz1Z1
zAR0h4fM@{G0HOgz1BeC?4ImmoG=OLT(Ey?WL<5Kh5Dg$2Ks1180MP)V0Yn3c1`rJ(
z8bCCFXaLawq5(t$hz1Z1AR0h4fM@{G0HOgz1BeC?4ImmoG=OLT(Ey?WL<5Kh5Dg$2
zKs1180MP)V0Yn3c1`rJ(8bCCFXaLawq5(t$hz1Z1AR0jb`1?r%hz1Z1AR0h4fM@{G
z0HOgz1BeC?4ImmoG=OLT(Ey?WL<5Kh5Dg$2Ks1180MP)V0Yn3c1`rJ(8bCCFXaLaw
zq5(t$hz1Z1AR0h4fM@{G0HOgz1BeC?4ImmoG=OLT(Ey?WL<5Kh5Dg$2Ks1180MP)V
z0Yn3c1`rJ(8bCCFXaLawq5(t$hz1Z1AR0h4fM@{G0HOgz1BeC?4ImmoG=OLT(Ey?W
zL<5Kh5Dg$2Ks1180MP)V0Yn3c1`rJ(8bCCFXaLawq5(t$hz1Z1AR0h4fM@{G0HOgz
z1BeC?4ImmoG=OLT(Ey?WL<5Kh5Dg$2Ks1180MP)V0Yn3c1`rJ(8bCCFXaLawq5(t$
zhz1Z1AR0h4fM@{G0HOgz1BeC?4ImmoG=OLT(Ey?WL<5Kh5Dg$2Ks1180MP)V0Yn3c
z1`rJ(8bCCFXaLawq5(t$hz1Z1AR0h4fM@{G0HOgz1BeC?4ImmoG=OLT(Ey?WL<5Kh
z5Dg$2Ks1180MP)V0Yn3c1`rJ(8bCCFXaLawq5(t$hz1Z1AR0h4fM@{G0HOgz1BeC?
z4ImmoG=OLT(Ey?WL<5Kh5Dg$2Ks1180MP)V0Yn3c1`rJ(8bCCFXaLawq5(t$hz1Z1
zAR0h4fM@{G0HOgz1BeC?4ImmoG=OLT(Ey?WL<5Kh5Dg$2Ks1180MP)V0Yn3c1`rJ(
z8bCCFXaLawq5(t$hz1Z1AR0h4fM@{G0HOgz1BeC?4ImmoG=OLT(Ey?WL<5Kh5Dg$2
zKs1180MP)V0Yn3c1`rJ(8bCCFXaLawq5(t$hz1Z1AR0h4fM@{G0HOgz1BeC?4Immo
zG=OLT(Ey?WL<5Kh5Dg$2Ks1180MP)V0Yn3c1`rJ(8bCCFXaLawq5(t$hz1Z1AR0h4
zfM@{G0HOgz1BeC?4ImmoG=OLT(Ey?WL<5Kh5Dg$2Ks1180MP)V0Yn3c1`rJ(8bCCF
zXaLawq5(t$hz1Z1AR0h4fM@{G0HOgz1BeC?4ImmoG=OLT(Ey?WL<5Kh5Dg$2Ks118
z0MP)V0Yn3c1`rJ(8bCCFXaLawq5(t$hz1Z1AR0h4fM@{G0HOgz1BeC?4ImmoG=OLT
z(Ey?WL<5Kh5Dg$2Ks1180MP)V0Yn3c1`rJ(8bCCFXaLawq5(t$hz1Z1AR0h4fM@{G
z0HOgz1BeC?4ImmoG=OLT(Ey?WL<5Kh5Dg$2Ks1180MP)V0Yn3c1`rJ(8bCCFXaLaw
zq5(t$hz1Z1AR0h4fM@{G0HOgz1BeC?4ImmoG=OLT(Ey?WL<5Kh5Dg$2Ks1180MP)V
z0Yn3c1`rJ(8bCCFXaLawq5(t$hz1Z1AR0h4fM@{G0HOgz1BeC?4ImmoG=OLT(Ey?W
zL<5Kh5Dg$2Ks1180MP)V0Yn3c1`rJ(8bCCFXaLawq5(t$hz1Z1AR0h4fM@{G0HOgz
z1BeC?4ImmoG=OLT(Ey?WL<5Kh5Dg$2Ks1180MP)V0Yn3c1`rJ(8bCCFXaLawq5(t$
zhz1Z1AR0h4fM@{G0HOgz1BeC?4ImmoG=OLT(Ey?WL<5Kh5Dg$2Ks1180MP)V0Yn3c
z1`rJ(8bCCFXaLawq5(t$hz1Z1AR0h4fM@{G0HOgz1BeC?4ImmoG=OLT(Ey?WL<5Kh
z5Dg$2Ks1180MP)V0Yn3c1`rJ(8bCCFXaLawq5(t$hz1Z1AR0h4fM@{G0HOgz1BeC?
z4ImmoG=OLT(Ey?WL<5Kh5Dg$2Ks1180MP)V0Yn3c1`rJ(8bCCFXaLawq5(t$hz1Z1
zAR0h4fM@{G0HOgz1BeC?4ImmoG=OLT(Ey?WL<5Kh5Dg$2Ks1180MP)V0Yn3c1`rJ(
z8bCCFXaLawq5(t$hz1Z1AR0h4fM@{G0HOgz1BeC?4ImmoG=OLT(Ey?WL<5Kh5Dg$2
zKs1180MP)V0Yn3c1`rJ(8bCCFXaLawq5(t$hz1Z1AR0h4fM@{G0HOgz1BeC?4Immo
zG=OLT(Ey?WL<5Kh5Dg$2Ks1180MP)V0Yn3c1`rJ(8bCCFXaLawq5(t$hz1Z1AR0h4
zfM@{G0HOgz1BeC?4ImmoG=OLT(Ey?WL<5Kh5Dg$2Ks1180MP)V0Yn3c1`rJ(8bCCF
zXaLawq5(t$hz1Z1AR0h4fM@{G0HOgz1BeC?4ImmoG=OLT(Ey?WL<5Kh5Dg$2Ks118
z0MP)V0Yn3c1`rJ(8bCCFXaLawMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifp
zG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C
z4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(
z0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy
z07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=F
zfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfP
zU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR
z7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR*uS=)1~3}HXaJ)Dj0P|o
zz-R!Y0gMJP8o+1(qXCQtFdD#U0HXnn1~3}HXaJ)Dj0P|oz-R!Y0gMJP8o+1(qXCQt
zFdD#U0Q)!g(*Q;T7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=F
zfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfP
zU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR
z7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|n
zMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y
z(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifp
zG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C
z4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(
z0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy
z07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=F
zfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfP
zU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR
z7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|n
zMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y
z(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifp
zG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C
z4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(
z0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy
z07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=F
zfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfP
zU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR
z7!6=FfYAU(0~ifpG=R|nMgtfPU^IX|*iQo(4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|n
zMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IX|+D`)*4PZ2Y
z(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifp
zG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e7Y
zll@QjKiU6eKMi0sfYAU(1K5-OPxe3A|78D@{ZIBk+5cq!ll@QjKiU6e|C9Ys_CMMG
zWdD=>Pxe3A|78D@{ZIBk+5cq!ll@QjKiU6e|C9Ys_CMMGWdD=>Pxe3A|78D@{ZIBk
z+5cq!ll@QjKiU6e|C9Ys_CMMGWdF1M&-Opt|7`!W{m=G4+y89;v;EKZKimIo|Fiwi
z_CMSIZ2z<U&-Opt|7`!W{m=G4+y89;v;EKZKimIo|Fiwi_CMSIY(EWPG=R|nMgtfP
zU^IZy07e5C4PZ2YJ=_0mKMi0sfYAU(1K6|u&-Opt|7`!W{m=G4+y89;v;EKZKimIo
z|Fiwi_CMSIY(EWPG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|n
zMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y
z(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifp
zG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C
z4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(
z0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy
z07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=F
zfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfP
zU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR
z7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|n
zMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y
z(EvsR7!6=FfYAU(1K59MKMi0sfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR
z7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|n
zMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y
z(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifp
zG=R|nMgtfPU^IZy07e5C4PZ2Y(E#>e+fM@+4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|n
zMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y
z(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifp
zG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C
z4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(
z0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy
z07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=F
zfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfP
zU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR
z7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|n
zMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y
z(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifp
zG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C
z4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(
z0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy
z07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=F
zfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfP
zU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR
z7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|n
zMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y
z(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifp
zG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C
z4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(
z0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy
z07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=F
zfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfP
zU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR
z7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|n
zMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y
z(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifp
zG=R|nMgtfPU^IZy07e5C4PZ2Y(EvsR7!6=FfYAU(0~ifpG=R|nMgtfPU^IZy07e5C
z4PZ2Y-~ZBuLWlqUABPVI*LfU%9Grb|cyjPM-r;}$$Kn5Q@TJ|s&;1Yo#eeJV@Bet`
zzx)5a<AJC7e}2}Fp6>s9@IU>>5B}%>_`$#P;NSkQKJ*_?`G3FT!9Skq|9$X}XZpAQ
z+kg0wfB(*Z_>lkbz=!;kt6RVS<fX5J=lUlvB!B<OSES#665s#+lQQ`CpY(0N|0D$Y
z{b%0-{Qk4gd%yqeAN}t?3lV?+Sxxi%&+fnd{bx4_I{f)he*eYg8i)VU!3nV6fAN{;
z_g}>94i64`o!@^E2>kx#(*5sW&V&B`<%HAkU(V7v_`Uq)e22q-<KR2W-@knN{ryXU
z{ri{B>-R6=(C=UW&Y$vM?$!PM%S}$df4OVr_b<18{Qk|k>fgUP3HkdsXJvl>=4Ha~
z-~7A#{TmM8_it{(|NWa=!hiqf-VO)9_}^Vw|NXnmln(!kgNuR=|9H>o_20icC-D1s
zU-W<f?kmOb-+c=H{kvrK_wQP;-@kwRHU92a_20j{-}?9OZch3A<BH1fAD0Jy|2R+o
z`^QPv-#^ak{r>T-*6$zROZ@)v;_>$nQ^LV-?a!y@`}2?A+8=jl|Ne11?e8D=6aD^i
z1Ih1;E4aTePV@Y}_`3W1A~*ki5##>8xCQ<9#l2m>FK#&cef3Ju!5`SG&q2Sh`ikFI
zciR5Gxor6R=9<p$n~NR4Z_b$fzWFxz_swh3-?vYH&)!`%{(X0O_4nQNliznI&VS#X
VIs1M0GVDM7SO3|+`p^H9{|m$tW0n8_

literal 0
HcmV?d00001

diff --git a/transformers/tests/tokenization_albert_test.py b/transformers/tests/tokenization_albert_test.py
new file mode 100644
index 0000000000..dd63f6756b
--- /dev/null
+++ b/transformers/tests/tokenization_albert_test.py
@@ -0,0 +1,78 @@
+# coding=utf-8
+# Copyright 2019 Hugging Face inc.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from __future__ import absolute_import, division, print_function, unicode_literals
+
+import os
+import unittest
+
+from transformers.tokenization_albert import (AlbertTokenizer, SPIECE_UNDERLINE)
+
+from .tokenization_tests_commons import CommonTestCases
+
+SAMPLE_VOCAB = os.path.join(os.path.dirname(os.path.abspath(__file__)),
+                    'fixtures/30k-clean.model')
+
+class AlbertTokenizationTest(CommonTestCases.CommonTokenizerTester):
+
+    tokenizer_class = AlbertTokenizer
+
+    def setUp(self):
+        super(AlbertTokenizationTest, self).setUp()
+
+        # We have a SentencePiece fixture for testing
+        tokenizer = AlbertTokenizer(SAMPLE_VOCAB)
+        tokenizer.save_pretrained(self.tmpdirname)
+
+    def get_tokenizer(self, **kwargs):
+        return AlbertTokenizer.from_pretrained(self.tmpdirname, **kwargs)
+
+    def get_input_output_texts(self):
+        input_text = u"this is a test"
+        output_text = u"this is a test"
+        return input_text, output_text
+
+
+    def test_full_tokenizer(self):
+        tokenizer = AlbertTokenizer(SAMPLE_VOCAB, keep_accents=True)
+
+        tokens = tokenizer.tokenize(u'This is a test')
+        self.assertListEqual(tokens, [u'▁this', u'▁is', u'▁a', u'▁test'])
+
+        self.assertListEqual(
+            tokenizer.convert_tokens_to_ids(tokens), [48, 25, 21, 1289])
+
+        tokens = tokenizer.tokenize(u"I was born in 92000, and this is falsé.")
+        self.assertListEqual(tokens, [u'▁i', u'▁was', u'▁born', u'▁in', u'▁9', u'2000', u',', u'▁and', u'▁this', u'▁is', u'▁fal', u's', u'é', u'.'])
+        ids = tokenizer.convert_tokens_to_ids(tokens)
+        self.assertListEqual(ids, [31, 23, 386, 19, 561, 3050, 15, 17, 48, 25, 8256, 18, 1, 9])
+
+        back_tokens = tokenizer.convert_ids_to_tokens(ids)
+        self.assertListEqual(back_tokens, ['▁i', '▁was', '▁born', '▁in', '▁9', '2000', ',', '▁and', '▁this', '▁is', '▁fal', 's', '<unk>', '.'])
+
+    def test_sequence_builders(self):
+        tokenizer = AlbertTokenizer(SAMPLE_VOCAB)
+
+        text = tokenizer.encode("sequence builders")
+        text_2 = tokenizer.encode("multi-sequence build")
+
+        encoded_sentence = tokenizer.build_inputs_with_special_tokens(text)
+        encoded_pair = tokenizer.build_inputs_with_special_tokens(text, text_2)
+
+        assert encoded_sentence == [tokenizer.cls_token_id] + text + [tokenizer.sep_token_id]
+        assert encoded_pair == [tokenizer.cls_token_id] + text + [tokenizer.sep_token_id] + text_2 + [tokenizer.sep_token_id]
+
+
+if __name__ == '__main__':
+    unittest.main()
diff --git a/transformers/tokenization_albert.py b/transformers/tokenization_albert.py
index f2e37222f6..0785e55ad2 100644
--- a/transformers/tokenization_albert.py
+++ b/transformers/tokenization_albert.py
@@ -8,6 +8,7 @@ from shutil import copyfile
 
 logger = logging.getLogger(__name__)
 
+VOCAB_FILES_NAMES = {'vocab_file': '30k-clean.model'}
 SPIECE_UNDERLINE = u'▁'
 
 class AlbertTokenizer(PreTrainedTokenizer):
@@ -16,12 +17,12 @@ class AlbertTokenizer(PreTrainedTokenizer):
 
             - requires `SentencePiece <https://github.com/google/sentencepiece>`_
     """
-    # vocab_files_names = VOCAB_FILES_NAMES
+    vocab_files_names = VOCAB_FILES_NAMES
     # pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
     # max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
 
     def __init__(self, vocab_file,
-                 do_lower_case=False, remove_space=True, keep_accents=False,
+                 do_lower_case=True, remove_space=True, keep_accents=False,
                  bos_token="[CLS]", eos_token="[SEP]", unk_token="<unk>", sep_token="[SEP]",
                  pad_token="<pad>", cls_token="[CLS]", mask_token="[MASK]>", **kwargs):
         super(AlbertTokenizer, self).__init__(bos_token=bos_token, eos_token=eos_token,
@@ -142,15 +143,15 @@ class AlbertTokenizer(PreTrainedTokenizer):
         """
         Build model inputs from a sequence or a pair of sequence for sequence classification tasks
         by concatenating and adding special tokens.
-        A RoBERTa sequence has the following format:
-            single sequence: <s> X </s>
-            pair of sequences: <s> A </s></s> B </s>
+        An ALBERT sequence has the following format:
+            single sequence: [CLS] X [SEP]
+            pair of sequences: [CLS] A [SEP] B [SEP]
         """
         sep = [self.sep_token_id]
         cls = [self.cls_token_id]
         if token_ids_1 is None:
-            return token_ids_0 + sep + cls
-        return token_ids_0 + sep + token_ids_1 + sep + cls
+            return cls + token_ids_0 + sep
+        return cls + token_ids_0 + sep + token_ids_1 + sep
 
     def get_special_tokens_mask(self, token_ids_0, token_ids_1=None, already_has_special_tokens=False):
         """
@@ -175,25 +176,24 @@ class AlbertTokenizer(PreTrainedTokenizer):
             return list(map(lambda x: 1 if x in [self.sep_token_id, self.cls_token_id] else 0, token_ids_0))
 
         if token_ids_1 is not None:
-            return ([0] * len(token_ids_0)) + [1] + ([0] * len(token_ids_1)) + [1, 1]
-        return ([0] * len(token_ids_0)) + [1, 1]
+            return [1] + ([0] * len(token_ids_0)) + [1] + ([0] * len(token_ids_1)) + [1]
+        return [1] + ([0] * len(token_ids_0)) + [1]
 
     def create_token_type_ids_from_sequences(self, token_ids_0, token_ids_1=None):
         """
         Creates a mask from the two sequences passed to be used in a sequence-pair classification task.
-        A BERT sequence pair mask has the following format:
-        0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 2
-        | first sequence    | second sequence     | CLS segment ID
+        An ALBERT sequence pair mask has the following format:
+        0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 
+        | first sequence    | second sequence     
         
         if token_ids_1 is None, only returns the first portion of the mask (0's).
         """
         sep = [self.sep_token_id]
         cls = [self.cls_token_id]
-        cls_segment_id = [2]
 
         if token_ids_1 is None:
-            return len(token_ids_0 + sep + cls) * [0]
-        return len(token_ids_0 + sep) * [0] + len(token_ids_1 + sep) * [1] + cls_segment_id
+            return len(cls + token_ids_0 + sep) * [0]
+        return len(cls + token_ids_0 + sep) * [0] + len(token_ids_1 + sep) * [1]
 
     def save_vocabulary(self, save_directory):
         """ Save the sentencepiece vocabulary (copy original file) and special tokens file
diff --git a/transformers/tokenization_xlnet.py b/transformers/tokenization_xlnet.py
index a4f1a6e3ba..c01fbbbeeb 100644
--- a/transformers/tokenization_xlnet.py
+++ b/transformers/tokenization_xlnet.py
@@ -185,9 +185,9 @@ class XLNetTokenizer(PreTrainedTokenizer):
         """
         Build model inputs from a sequence or a pair of sequence for sequence classification tasks
         by concatenating and adding special tokens.
-        A RoBERTa sequence has the following format:
-            single sequence: <s> X </s>
-            pair of sequences: <s> A </s></s> B </s>
+        An XLNet sequence has the following format:
+            single sequence: X <sep> <cls>
+            pair of sequences: A <sep> B <sep> <cls>
         """
         sep = [self.sep_token_id]
         cls = [self.cls_token_id]
@@ -224,7 +224,7 @@ class XLNetTokenizer(PreTrainedTokenizer):
     def create_token_type_ids_from_sequences(self, token_ids_0, token_ids_1=None):
         """
         Creates a mask from the two sequences passed to be used in a sequence-pair classification task.
-        A BERT sequence pair mask has the following format:
+        An XLNet sequence pair mask has the following format:
         0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 2
         | first sequence    | second sequence     | CLS segment ID
         

From 1e5b31c3881cb1313216ce0f3cffae89b0845d4f Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Wed, 30 Oct 2019 20:25:32 +0000
Subject: [PATCH 094/505] Several fixes and improvements

---
 transformers/modeling_albert.py               |  36 +++++++++---------
 .../{30k-clean.model => spiece.model}         | Bin
 transformers/tokenization_albert.py           |   2 +-
 3 files changed, 19 insertions(+), 19 deletions(-)
 rename transformers/tests/fixtures/{30k-clean.model => spiece.model} (100%)

diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index ad8b979cef..371a2e535c 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -7,6 +7,7 @@ import torch.nn as nn
 from torch.nn import CrossEntropyLoss
 from transformers.configuration_albert import AlbertConfig
 from transformers.modeling_bert import BertEmbeddings, BertModel, BertSelfAttention, prune_linear_layer, gelu_new
+from transformers.modeling_utils import PreTrainedModel
 from .file_utils import add_start_docstrings
 
 logger = logging.getLogger(__name__)
@@ -37,18 +38,17 @@ def load_tf_weights_in_albert(model, config, tf_checkpoint_path):
         print(name)
     
     for name, array in zip(names, arrays):
-        print(name)
-        og = name
+        original_name = name
         name = name.replace("ffn_1", "ffn")
         name = name.replace("ffn/intermediate/output", "ffn_output")
         name = name.replace("attention_1", "attention")   
-        name = name.replace("cls/predictions/transform", "predictions")
-        name = name.replace("LayerNorm_1", "attention/LayerNorm")    
+        name = name.replace("cls/predictions", "predictions")
+        name = name.replace("transform/", "")
+        name = name.replace("LayerNorm_1", "full_layer_layer_norm")    
+        name = name.replace("LayerNorm", "attention/LayerNorm")    
         name = name.replace("inner_group_", "albert_layers/") 
         name = name.replace("group_", "albert_layer_groups/")    
         name = name.split('/')
-
-        print(name)
         pointer = model
         for m_name in name:
             if re.fullmatch(r'[A-Za-z]+_\d+', m_name):
@@ -78,13 +78,12 @@ def load_tf_weights_in_albert(model, config, tf_checkpoint_path):
             pointer = getattr(pointer, 'weight')
         elif m_name == 'kernel':
             array = np.transpose(array)
-            print("transposed")
         try:
             assert pointer.shape == array.shape
         except AssertionError as e:
             e.args += (pointer.shape, array.shape)
             raise
-        print("Initialize PyTorch weight {} from {}".format(name, og))
+        print("Initialize PyTorch weight {} from {}".format(name, original_name))
         pointer.data = torch.from_numpy(array)
 
     return model
@@ -177,9 +176,9 @@ class AlbertAttention(BertSelfAttention):
         b = self.dense.bias
 
         projected_context_layer = torch.einsum("bfnd,ndh->bfh", context_layer, w) + b
-        projected_context_layer = self.dropout(projected_context_layer)
-        layernormed_context_layer = self.LayerNorm(input_ids + projected_context_layer)
-        return layernormed_context_layer, projected_context_layer, reshaped_context_layer, context_layer, attention_scores, attention_probs, attention_mask
+        projected_context_layer_dropout = self.dropout(projected_context_layer)
+        layernormed_context_layer = self.LayerNorm(input_ids + projected_context_layer_dropout)
+        return layernormed_context_layer
 
 
 class AlbertLayer(nn.Module):
@@ -187,17 +186,17 @@ class AlbertLayer(nn.Module):
         super(AlbertLayer, self).__init__()
         
         self.config = config
-        self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+        self.full_layer_layer_norm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
         self.attention = AlbertAttention(config)
         self.ffn = nn.Linear(config.hidden_size, config.intermediate_size) 
         self.ffn_output = nn.Linear(config.intermediate_size, config.hidden_size)
 
     def forward(self, hidden_states, attention_mask=None, head_mask=None):
-        attention_output = self.attention(hidden_states, attention_mask)[0]
+        attention_output = self.attention(hidden_states, attention_mask)
         ffn_output = self.ffn(attention_output)
         ffn_output = gelu_new(ffn_output)
         ffn_output = self.ffn_output(ffn_output)
-        hidden_states = self.LayerNorm(ffn_output + attention_output)
+        hidden_states = self.full_layer_layer_norm(ffn_output + attention_output)
 
         return hidden_states
 
@@ -352,16 +351,17 @@ class AlbertModel(BertModel):
         encoder_outputs = self.encoder(embedding_output,
                                        extended_attention_mask,
                                        head_mask=head_mask)
+
         sequence_output = encoder_outputs[0]
-    
+
         pooled_output = self.pooler_activation(self.pooler(sequence_output[:, 0]))
 
-        outputs = (sequence_output, pooled_output,) + encoder_outputs[1:]  # add hidden_states and attentions if they are here
+        outputs = (sequence_output, pooled_output) + encoder_outputs[1:]  # add hidden_states and attentions if they are here
         return outputs
 
 
 @add_start_docstrings("Bert Model with a `language modeling` head on top.", ALBERT_START_DOCSTRING, ALBERT_INPUTS_DOCSTRING)
-class AlbertForMaskedLM(nn.Module):
+class AlbertForMaskedLM(PreTrainedModel):
     r"""
         **masked_lm_labels**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
             Labels for computing the masked language modeling loss.
@@ -384,7 +384,7 @@ class AlbertForMaskedLM(nn.Module):
     """
     
     def __init__(self, config):
-        super(AlbertForMaskedLM, self).__init__()
+        super(AlbertForMaskedLM, self).__init__(config)
 
         self.config = config
         self.bert = AlbertModel(config)
diff --git a/transformers/tests/fixtures/30k-clean.model b/transformers/tests/fixtures/spiece.model
similarity index 100%
rename from transformers/tests/fixtures/30k-clean.model
rename to transformers/tests/fixtures/spiece.model
diff --git a/transformers/tokenization_albert.py b/transformers/tokenization_albert.py
index 0785e55ad2..7b16bb573f 100644
--- a/transformers/tokenization_albert.py
+++ b/transformers/tokenization_albert.py
@@ -8,7 +8,7 @@ from shutil import copyfile
 
 logger = logging.getLogger(__name__)
 
-VOCAB_FILES_NAMES = {'vocab_file': '30k-clean.model'}
+VOCAB_FILES_NAMES = {'vocab_file': 'spiece.model'}
 SPIECE_UNDERLINE = u'▁'
 
 class AlbertTokenizer(PreTrainedTokenizer):

From 5680a1106302b1ebeb960de0700d6379c0aeef5c Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Wed, 30 Oct 2019 20:42:49 +0000
Subject: [PATCH 095/505] Activation function managed from the config file

---
 transformers/configuration_albert.py | 2 +-
 transformers/modeling_albert.py      | 8 +++++---
 2 files changed, 6 insertions(+), 4 deletions(-)

diff --git a/transformers/configuration_albert.py b/transformers/configuration_albert.py
index c86c9565cb..15437dbbea 100644
--- a/transformers/configuration_albert.py
+++ b/transformers/configuration_albert.py
@@ -16,7 +16,7 @@ class AlbertConfig(PretrainedConfig):
                  intermediate_size=16384,
                  inner_group_num=1,
                  down_scale_factor=1,
-                 hidden_act="gelu",
+                 hidden_act="gelu_new",
                  hidden_dropout_prob=0,
                  attention_probs_dropout_prob=0,
                  max_position_embeddings=512,
diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index 371a2e535c..7e9f7f1c46 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -6,7 +6,7 @@ import torch
 import torch.nn as nn
 from torch.nn import CrossEntropyLoss
 from transformers.configuration_albert import AlbertConfig
-from transformers.modeling_bert import BertEmbeddings, BertModel, BertSelfAttention, prune_linear_layer, gelu_new
+from transformers.modeling_bert import BertEmbeddings, BertModel, BertSelfAttention, prune_linear_layer, ACT2FN
 from transformers.modeling_utils import PreTrainedModel
 from .file_utils import add_start_docstrings
 
@@ -190,11 +190,12 @@ class AlbertLayer(nn.Module):
         self.attention = AlbertAttention(config)
         self.ffn = nn.Linear(config.hidden_size, config.intermediate_size) 
         self.ffn_output = nn.Linear(config.intermediate_size, config.hidden_size)
+        self.activation = ACT2FN[config.hidden_act]
 
     def forward(self, hidden_states, attention_mask=None, head_mask=None):
         attention_output = self.attention(hidden_states, attention_mask)
         ffn_output = self.ffn(attention_output)
-        ffn_output = gelu_new(ffn_output)
+        ffn_output = self.activation(ffn_output)
         ffn_output = self.ffn_output(ffn_output)
         hidden_states = self.full_layer_layer_norm(ffn_output + attention_output)
 
@@ -392,6 +393,7 @@ class AlbertForMaskedLM(PreTrainedModel):
         self.bias = nn.Parameter(torch.zeros(config.vocab_size))
         self.dense = nn.Linear(config.hidden_size, config.embedding_size)
         self.word_embeddings = nn.Linear(config.embedding_size, config.vocab_size)
+        self.activation = ACT2FN[config.hidden_act]
 
     def tie_weights(self):
         """ Make sure we are sharing the input and output embeddings.
@@ -405,7 +407,7 @@ class AlbertForMaskedLM(PreTrainedModel):
         outputs = self.bert(input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None)
         sequence_outputs = outputs[0]
         hidden_states = self.dense(sequence_outputs)
-        hidden_states = gelu_new(hidden_states)
+        hidden_states = self.activation(hidden_states)
         hidden_states = self.LayerNorm(hidden_states)
         prediction_scores = self.word_embeddings(hidden_states)
 

From ce9eade29c75fa676ac528d1fe21d9f4ac3c5622 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Wed, 30 Oct 2019 20:50:44 +0000
Subject: [PATCH 096/505] Initializer range using BertPreTrainedModel

---
 transformers/modeling_albert.py | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index 7e9f7f1c46..b45208b696 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -6,8 +6,7 @@ import torch
 import torch.nn as nn
 from torch.nn import CrossEntropyLoss
 from transformers.configuration_albert import AlbertConfig
-from transformers.modeling_bert import BertEmbeddings, BertModel, BertSelfAttention, prune_linear_layer, ACT2FN
-from transformers.modeling_utils import PreTrainedModel
+from transformers.modeling_bert import BertEmbeddings, BertPreTrainedModel, BertModel, BertSelfAttention, prune_linear_layer, ACT2FN
 from .file_utils import add_start_docstrings
 
 logger = logging.getLogger(__name__)
@@ -362,7 +361,7 @@ class AlbertModel(BertModel):
 
 
 @add_start_docstrings("Bert Model with a `language modeling` head on top.", ALBERT_START_DOCSTRING, ALBERT_INPUTS_DOCSTRING)
-class AlbertForMaskedLM(PreTrainedModel):
+class AlbertForMaskedLM(BertPreTrainedModel):
     r"""
         **masked_lm_labels**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
             Labels for computing the masked language modeling loss.

From 25a31953e8820eb5c88d8ad35ee547efccfe577c Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Wed, 30 Oct 2019 21:18:06 +0000
Subject: [PATCH 097/505] Output Attentions + output hidden states

---
 transformers/modeling_albert.py | 58 ++++++++++++++++++++++++++++-----
 1 file changed, 49 insertions(+), 9 deletions(-)

diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index b45208b696..52cab2ea69 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -105,6 +105,7 @@ class AlbertAttention(BertSelfAttention):
     def __init__(self, config):
         super(AlbertAttention, self).__init__(config)
 
+        self.output_attentions = config.output_attentions
         self.num_attention_heads = config.num_attention_heads
         self.hidden_size = config.hidden_size 
         self.attention_head_size = config.hidden_size // config.num_attention_heads
@@ -177,7 +178,7 @@ class AlbertAttention(BertSelfAttention):
         projected_context_layer = torch.einsum("bfnd,ndh->bfh", context_layer, w) + b
         projected_context_layer_dropout = self.dropout(projected_context_layer)
         layernormed_context_layer = self.LayerNorm(input_ids + projected_context_layer_dropout)
-        return layernormed_context_layer
+        return (layernormed_context_layer, attention_probs) if self.output_attentions else (layernormed_context_layer,)
 
 
 class AlbertLayer(nn.Module):
@@ -193,25 +194,45 @@ class AlbertLayer(nn.Module):
 
     def forward(self, hidden_states, attention_mask=None, head_mask=None):
         attention_output = self.attention(hidden_states, attention_mask)
-        ffn_output = self.ffn(attention_output)
+        ffn_output = self.ffn(attention_output[0])
         ffn_output = self.activation(ffn_output)
         ffn_output = self.ffn_output(ffn_output)
-        hidden_states = self.full_layer_layer_norm(ffn_output + attention_output)
+        hidden_states = self.full_layer_layer_norm(ffn_output + attention_output[0])
 
-        return hidden_states
+        return (hidden_states,) + attention_output[1:]  # add attentions if we output them
 
 
 class AlbertLayerGroup(nn.Module):
     def __init__(self, config):
         super(AlbertLayerGroup, self).__init__()
         
+        self.output_attentions = config.output_attentions
+        self.output_hidden_states = config.output_hidden_states
         self.albert_layers = nn.ModuleList([AlbertLayer(config) for _ in range(config.inner_group_num)])
 
     def forward(self, hidden_states, attention_mask=None, head_mask=None):
-        for albert_layer in self.albert_layers:
-            hidden_states = albert_layer(hidden_states, attention_mask, head_mask)
+        layer_hidden_states = ()
+        layer_attentions = ()
 
-        return hidden_states
+        for albert_layer in self.albert_layers:
+            if self.output_hidden_states:
+                layer_hidden_states = layer_hidden_states + (hidden_states,)
+
+            layer_output = albert_layer(hidden_states, attention_mask, head_mask)
+            hidden_states = layer_output[0]
+
+            if self.output_attentions:
+                layer_attentions = layer_attentions + (layer_output[1],)
+
+        if self.output_hidden_states:
+                layer_hidden_states = layer_hidden_states + (hidden_states,)
+
+        outputs = (hidden_states,)
+        if self.output_hidden_states:
+            outputs = outputs + (layer_hidden_states,)
+        if self.output_attentions:
+            outputs = outputs + (layer_attentions,)
+        return outputs  # last-layer hidden state, (layer hidden states), (layer attentions)
 
 
 class AlbertTransformer(nn.Module):
@@ -227,11 +248,30 @@ class AlbertTransformer(nn.Module):
     def forward(self, hidden_states, attention_mask=None, head_mask=None):
         hidden_states = self.embedding_hidden_mapping_in(hidden_states)
 
+        all_attentions = ()
+
+        if self.output_hidden_states:
+            all_hidden_states = (hidden_states,)
+
         for layer_idx in range(self.config.num_hidden_layers):
             group_idx = int(layer_idx / self.config.num_hidden_layers * self.config.num_hidden_groups)
-            hidden_states = self.albert_layer_groups[group_idx](hidden_states, attention_mask, head_mask)  
+            layer_group_output = self.albert_layer_groups[group_idx](hidden_states, attention_mask, head_mask)  
         
-        return (hidden_states,)
+            hidden_states = layer_group_output[0]
+
+            if self.output_attentions:
+                all_attentions = all_attentions + layer_group_output[1]
+
+            if self.output_hidden_states:
+                all_hidden_states = all_hidden_states + (hidden_states,)
+
+        
+        outputs = (hidden_states,)
+        if self.output_hidden_states:
+            outputs = outputs + (all_hidden_states,)
+        if self.output_attentions:
+            outputs = outputs + (all_attentions,)
+        return outputs  # last-layer hidden state, (all hidden states), (all attentions)
 
 
 ALBERT_START_DOCSTRING = r"""    The ALBERT model was proposed in

From 870320a24e28f187d0dfd10b82c4d60d5269374d Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Wed, 30 Oct 2019 22:30:21 +0000
Subject: [PATCH 098/505] Early tests

---
 transformers/modeling_albert.py            |  59 +++----
 transformers/tests/modeling_albert_test.py | 191 +++++++++++++++++++++
 2 files changed, 219 insertions(+), 31 deletions(-)
 create mode 100644 transformers/tests/modeling_albert_test.py

diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index 52cab2ea69..f906352311 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -11,6 +11,15 @@ from .file_utils import add_start_docstrings
 
 logger = logging.getLogger(__name__)
 
+
+ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
+    'albert-base': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-base-pytorch_model.bin",
+    'albert-large': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-large-pytorch_model.bin",
+    'albert-xlarge': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xlarge-pytorch_model.bin",
+    'albert-xxlarge': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xxlarge-pytorch_model.bin",
+}
+
+
 def load_tf_weights_in_albert(model, config, tf_checkpoint_path):
     """ Load tf checkpoints in a pytorch model."""
     try:
@@ -39,6 +48,7 @@ def load_tf_weights_in_albert(model, config, tf_checkpoint_path):
     for name, array in zip(names, arrays):
         original_name = name
         name = name.replace("ffn_1", "ffn")
+        name = name.replace("/bert/", "/albert/")
         name = name.replace("ffn/intermediate/output", "ffn_output")
         name = name.replace("attention_1", "attention")   
         name = name.replace("cls/predictions", "predictions")
@@ -114,29 +124,6 @@ class AlbertAttention(BertSelfAttention):
         self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
         self.pruned_heads = set()
 
-    def prune_heads(self, heads):
-        if len(heads) == 0:
-            return
-        mask = torch.ones(self.num_attention_heads, self.attention_head_size)
-        heads = set(heads) - self.pruned_heads  # Convert to set and emove already pruned heads
-        for head in heads:
-            # Compute how many pruned heads are before the head and move the index accordingly
-            head = head - sum(1 if h < head else 0 for h in self.pruned_heads)
-            mask[head] = 0
-        mask = mask.view(-1).contiguous().eq(1)
-        index = torch.arange(len(mask))[mask].long()
-
-        # Prune linear layers
-        self.query = prune_linear_layer(self.query, index)
-        self.key = prune_linear_layer(self.key, index)
-        self.value = prune_linear_layer(self.value, index)
-        self.output.dense = prune_linear_layer(self.output.dense, index, dim=1)
-
-        # Update hyper params and store pruned heads
-        self.num_attention_heads = self.num_attention_heads - len(heads)
-        self.all_head_size = self.attention_head_size * self.num_attention_heads
-        self.pruned_heads = self.pruned_heads.union(heads)
-
     def forward(self, input_ids, attention_mask=None, head_mask=None):
         mixed_query_layer = self.query(input_ids)
         mixed_key_layer = self.key(input_ids)
@@ -225,7 +212,7 @@ class AlbertLayerGroup(nn.Module):
                 layer_attentions = layer_attentions + (layer_output[1],)
 
         if self.output_hidden_states:
-                layer_hidden_states = layer_hidden_states + (hidden_states,)
+            layer_hidden_states = layer_hidden_states + (hidden_states,)
 
         outputs = (hidden_states,)
         if self.output_hidden_states:
@@ -367,6 +354,8 @@ class AlbertModel(BertModel):
         self.pooler = nn.Linear(config.hidden_size, config.hidden_size)
         self.pooler_activation = nn.Tanh()
 
+        self.init_weights()
+
 
     def forward(self, input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None):
         if attention_mask is None:
@@ -422,33 +411,41 @@ class AlbertForMaskedLM(BertPreTrainedModel):
             list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
             Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
     """
-    
+
+    config_class = AlbertConfig
+    pretrained_model_archive_map = ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP
+    load_tf_weights = load_tf_weights_in_albert
+    base_model_prefix = "albert"
+
     def __init__(self, config):
         super(AlbertForMaskedLM, self).__init__(config)
 
         self.config = config
-        self.bert = AlbertModel(config)
+        self.albert = AlbertModel(config)
         self.LayerNorm = nn.LayerNorm(config.embedding_size)
         self.bias = nn.Parameter(torch.zeros(config.vocab_size))
         self.dense = nn.Linear(config.hidden_size, config.embedding_size)
-        self.word_embeddings = nn.Linear(config.embedding_size, config.vocab_size)
+        self.decoder = nn.Linear(config.embedding_size, config.vocab_size)
         self.activation = ACT2FN[config.hidden_act]
 
+        self.init_weights()
+        self.tie_weights()
+
     def tie_weights(self):
         """ Make sure we are sharing the input and output embeddings.
             Export to TorchScript can't handle parameter sharing so we are cloning them instead.
         """
-        self._tie_or_clone_weights(self.classifier.word_embeddings,
-                                   self.transformer.embeddings.word_embeddings)
+        self._tie_or_clone_weights(self.decoder,
+                                   self.albert.embeddings.word_embeddings)
 
     def forward(self, input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None,
                 masked_lm_labels=None):
-        outputs = self.bert(input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None)
+        outputs = self.albert(input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None)
         sequence_outputs = outputs[0]
         hidden_states = self.dense(sequence_outputs)
         hidden_states = self.activation(hidden_states)
         hidden_states = self.LayerNorm(hidden_states)
-        prediction_scores = self.word_embeddings(hidden_states)
+        prediction_scores = self.decoder(hidden_states)
 
         outputs = (prediction_scores,) + outputs[2:]  # Add hidden states and attention if they are here
         if masked_lm_labels is not None:
diff --git a/transformers/tests/modeling_albert_test.py b/transformers/tests/modeling_albert_test.py
new file mode 100644
index 0000000000..46a2eeb729
--- /dev/null
+++ b/transformers/tests/modeling_albert_test.py
@@ -0,0 +1,191 @@
+# coding=utf-8
+# Copyright 2018 The Google AI Language Team Authors.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from __future__ import absolute_import
+from __future__ import division
+from __future__ import print_function
+
+import unittest
+import shutil
+import pytest
+
+from transformers import is_torch_available
+
+from .modeling_common_test import (CommonTestCases, ids_tensor)
+from .configuration_common_test import ConfigTester
+
+if is_torch_available():
+    from transformers import (AlbertConfig, AlbertModel, AlbertForMaskedLM)
+    from transformers.modeling_albert import ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP
+else:
+    pytestmark = pytest.mark.skip("Require Torch")
+
+
+class AlbertModelTest(CommonTestCases.CommonModelTester):
+
+    all_model_classes = (AlbertModel, AlbertForMaskedLM) if is_torch_available() else ()
+    test_pruning = False
+    test_head_masking = False
+
+    class AlbertModelTester(object):
+
+        def __init__(self,
+                     parent,
+                     batch_size=13,
+                     seq_length=7,
+                     is_training=True,
+                     use_input_mask=True,
+                     use_token_type_ids=True,
+                     use_labels=True,
+                     vocab_size=99,
+                     hidden_size=32,
+                     num_hidden_layers=5,
+                     num_attention_heads=4,
+                     intermediate_size=37,
+                     hidden_act="gelu",
+                     hidden_dropout_prob=0.1,
+                     attention_probs_dropout_prob=0.1,
+                     max_position_embeddings=512,
+                     type_vocab_size=16,
+                     type_sequence_label_size=2,
+                     initializer_range=0.02,
+                     num_labels=3,
+                     num_choices=4,
+                     scope=None,
+                    ):
+            self.parent = parent
+            self.batch_size = batch_size
+            self.seq_length = seq_length
+            self.is_training = is_training
+            self.use_input_mask = use_input_mask
+            self.use_token_type_ids = use_token_type_ids
+            self.use_labels = use_labels
+            self.vocab_size = vocab_size
+            self.hidden_size = hidden_size
+            self.num_hidden_layers = num_hidden_layers
+            self.num_attention_heads = num_attention_heads
+            self.intermediate_size = intermediate_size
+            self.hidden_act = hidden_act
+            self.hidden_dropout_prob = hidden_dropout_prob
+            self.attention_probs_dropout_prob = attention_probs_dropout_prob
+            self.max_position_embeddings = max_position_embeddings
+            self.type_vocab_size = type_vocab_size
+            self.type_sequence_label_size = type_sequence_label_size
+            self.initializer_range = initializer_range
+            self.num_labels = num_labels
+            self.num_choices = num_choices
+            self.scope = scope
+
+        def prepare_config_and_inputs(self):
+            input_ids = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)
+
+            input_mask = None
+            if self.use_input_mask:
+                input_mask = ids_tensor([self.batch_size, self.seq_length], vocab_size=2)
+
+            token_type_ids = None
+            if self.use_token_type_ids:
+                token_type_ids = ids_tensor([self.batch_size, self.seq_length], self.type_vocab_size)
+
+            sequence_labels = None
+            token_labels = None
+            choice_labels = None
+            if self.use_labels:
+                sequence_labels = ids_tensor([self.batch_size], self.type_sequence_label_size)
+                token_labels = ids_tensor([self.batch_size, self.seq_length], self.num_labels)
+                choice_labels = ids_tensor([self.batch_size], self.num_choices)
+
+            config = AlbertConfig(
+                vocab_size_or_config_json_file=self.vocab_size,
+                hidden_size=self.hidden_size,
+                num_hidden_layers=self.num_hidden_layers,
+                num_attention_heads=self.num_attention_heads,
+                intermediate_size=self.intermediate_size,
+                hidden_act=self.hidden_act,
+                hidden_dropout_prob=self.hidden_dropout_prob,
+                attention_probs_dropout_prob=self.attention_probs_dropout_prob,
+                max_position_embeddings=self.max_position_embeddings,
+                type_vocab_size=self.type_vocab_size,
+                initializer_range=self.initializer_range)
+
+            return config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
+
+        def check_loss_output(self, result):
+            self.parent.assertListEqual(
+                list(result["loss"].size()),
+                [])
+
+        def create_and_check_albert_model(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
+            model = AlbertModel(config=config)
+            model.eval()
+            sequence_output, pooled_output = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids)
+            sequence_output, pooled_output = model(input_ids, token_type_ids=token_type_ids)
+            sequence_output, pooled_output = model(input_ids)
+
+            result = {
+                "sequence_output": sequence_output,
+                "pooled_output": pooled_output,
+            }
+            self.parent.assertListEqual(
+                list(result["sequence_output"].size()),
+                [self.batch_size, self.seq_length, self.hidden_size])
+            self.parent.assertListEqual(list(result["pooled_output"].size()), [self.batch_size, self.hidden_size])
+
+
+        def create_and_check_albert_for_masked_lm(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
+            model = AlbertForMaskedLM(config=config)
+            model.eval()
+            loss, prediction_scores = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, masked_lm_labels=token_labels)
+            result = {
+                "loss": loss,
+                "prediction_scores": prediction_scores,
+            }
+            self.parent.assertListEqual(
+                list(result["prediction_scores"].size()),
+                [self.batch_size, self.seq_length, self.vocab_size])
+            self.check_loss_output(result)
+
+
+        def prepare_config_and_inputs_for_common(self):
+            config_and_inputs = self.prepare_config_and_inputs()
+            (config, input_ids, token_type_ids, input_mask,
+             sequence_labels, token_labels, choice_labels) = config_and_inputs
+            inputs_dict = {'input_ids': input_ids, 'token_type_ids': token_type_ids, 'attention_mask': input_mask}
+            return config, inputs_dict
+
+    def setUp(self):
+        self.model_tester = AlbertModelTest.AlbertModelTester(self)
+        self.config_tester = ConfigTester(self, config_class=AlbertConfig, hidden_size=37)
+
+    def test_config(self):
+        self.config_tester.run_common_tests()
+
+    def test_albert_model(self):
+        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+        self.model_tester.create_and_check_albert_model(*config_and_inputs)
+
+    def test_for_masked_lm(self):
+        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+        self.model_tester.create_and_check_albert_for_masked_lm(*config_and_inputs)
+
+    @pytest.mark.slow
+    def test_model_from_pretrained(self):
+        cache_dir = "/tmp/transformers_test/"
+        for model_name in list(ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
+            model = AlbertModel.from_pretrained(model_name, cache_dir=cache_dir)
+            shutil.rmtree(cache_dir)
+            self.assertIsNotNone(model)
+
+if __name__ == "__main__":
+    unittest.main()

From c14a22272f3fc17bb2eaeca62986c31a7d26bc85 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Thu, 31 Oct 2019 14:04:10 +0000
Subject: [PATCH 099/505] ALBERT passes all tests

---
 transformers/configuration_albert.py           | 4 +---
 transformers/modeling_albert.py                | 9 +++------
 transformers/tests/tokenization_albert_test.py | 2 +-
 3 files changed, 5 insertions(+), 10 deletions(-)

diff --git a/transformers/configuration_albert.py b/transformers/configuration_albert.py
index 15437dbbea..04f9fa8d60 100644
--- a/transformers/configuration_albert.py
+++ b/transformers/configuration_albert.py
@@ -7,7 +7,7 @@ class AlbertConfig(PretrainedConfig):
     """
 
     def __init__(self,
-                 vocab_size_or_config_json_file,
+                 vocab_size_or_config_json_file=30000,
                  embedding_size=128,
                  hidden_size=4096,
                  num_hidden_layers=12,
@@ -15,7 +15,6 @@ class AlbertConfig(PretrainedConfig):
                  num_attention_heads=64,
                  intermediate_size=16384,
                  inner_group_num=1,
-                 down_scale_factor=1,
                  hidden_act="gelu_new",
                  hidden_dropout_prob=0,
                  attention_probs_dropout_prob=0,
@@ -61,7 +60,6 @@ class AlbertConfig(PretrainedConfig):
         self.num_hidden_groups = num_hidden_groups
         self.num_attention_heads = num_attention_heads
         self.inner_group_num = inner_group_num
-        self.down_scale_factor = down_scale_factor
         self.hidden_act = hidden_act
         self.intermediate_size = intermediate_size
         self.hidden_dropout_prob = hidden_dropout_prob
diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index f906352311..9bb38dead9 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -202,17 +202,14 @@ class AlbertLayerGroup(nn.Module):
         layer_attentions = ()
 
         for albert_layer in self.albert_layers:
-            if self.output_hidden_states:
-                layer_hidden_states = layer_hidden_states + (hidden_states,)
-
             layer_output = albert_layer(hidden_states, attention_mask, head_mask)
             hidden_states = layer_output[0]
 
             if self.output_attentions:
                 layer_attentions = layer_attentions + (layer_output[1],)
 
-        if self.output_hidden_states:
-            layer_hidden_states = layer_hidden_states + (hidden_states,)
+            if self.output_hidden_states:
+                layer_hidden_states = layer_hidden_states + (hidden_states,)
 
         outputs = (hidden_states,)
         if self.output_hidden_states:
@@ -247,7 +244,7 @@ class AlbertTransformer(nn.Module):
             hidden_states = layer_group_output[0]
 
             if self.output_attentions:
-                all_attentions = all_attentions + layer_group_output[1]
+                all_attentions = all_attentions + layer_group_output[-1]
 
             if self.output_hidden_states:
                 all_hidden_states = all_hidden_states + (hidden_states,)
diff --git a/transformers/tests/tokenization_albert_test.py b/transformers/tests/tokenization_albert_test.py
index dd63f6756b..59eb3bceb0 100644
--- a/transformers/tests/tokenization_albert_test.py
+++ b/transformers/tests/tokenization_albert_test.py
@@ -22,7 +22,7 @@ from transformers.tokenization_albert import (AlbertTokenizer, SPIECE_UNDERLINE)
 from .tokenization_tests_commons import CommonTestCases
 
 SAMPLE_VOCAB = os.path.join(os.path.dirname(os.path.abspath(__file__)),
-                    'fixtures/30k-clean.model')
+                    'fixtures/spiece.model')
 
 class AlbertTokenizationTest(CommonTestCases.CommonTokenizerTester):
 

From b21402fc86257feca05ab050d061e15441a49929 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Thu, 31 Oct 2019 14:19:31 +0000
Subject: [PATCH 100/505] Python 2 tests + licence

---
 transformers/modeling_albert.py     | 16 ++++++++++++++++
 transformers/tokenization_albert.py | 19 ++++++++++++++++++-
 2 files changed, 34 insertions(+), 1 deletion(-)

diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index 9bb38dead9..b6d1291725 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -1,4 +1,20 @@
 
+# coding=utf-8
+# Copyright 2018 Google AI, Google Brain and the HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""PyTorch ALBERT model. """
+
 import os
 import math
 import logging
diff --git a/transformers/tokenization_albert.py b/transformers/tokenization_albert.py
index 7b16bb573f..7cba99b9e4 100644
--- a/transformers/tokenization_albert.py
+++ b/transformers/tokenization_albert.py
@@ -1,4 +1,21 @@
-
+# coding=utf-8
+# Copyright 2018 Google AI, Google Brain and the HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" Tokenization classes for ALBERT model."""
+from __future__ import (absolute_import, division, print_function,
+                        unicode_literals)
+                        
 from .tokenization_utils import PreTrainedTokenizer
 import logging
 import unicodedata

From c4403006b8301a24bfeec99c39ce8d5d47df570f Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Thu, 31 Oct 2019 15:30:11 +0000
Subject: [PATCH 101/505] External MLM head

---
 transformers/configuration_albert.py | 17 ++++++++++++++
 transformers/modeling_albert.py      | 35 +++++++++++++++++++---------
 2 files changed, 41 insertions(+), 11 deletions(-)

diff --git a/transformers/configuration_albert.py b/transformers/configuration_albert.py
index 04f9fa8d60..b72bbb971e 100644
--- a/transformers/configuration_albert.py
+++ b/transformers/configuration_albert.py
@@ -1,3 +1,20 @@
+# coding=utf-8
+# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.
+# Copyright (c) 2018, NVIDIA CORPORATION.  All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" ALBERT model configuration """
+
 from .configuration_utils import PretrainedConfig
 
 class AlbertConfig(PretrainedConfig):
diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index b6d1291725..487455e561 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -401,6 +401,26 @@ class AlbertModel(BertModel):
         outputs = (sequence_output, pooled_output) + encoder_outputs[1:]  # add hidden_states and attentions if they are here
         return outputs
 
+class AlbertMLMHead(nn.Module):
+    def __init__(self, config):
+        super(AlbertMLMHead, self).__init__()
+
+        self.LayerNorm = nn.LayerNorm(config.embedding_size)
+        self.bias = nn.Parameter(torch.zeros(config.vocab_size))
+        self.dense = nn.Linear(config.hidden_size, config.embedding_size)
+        self.decoder = nn.Linear(config.embedding_size, config.vocab_size)
+        self.activation = ACT2FN[config.hidden_act]
+
+    def forward(self, hidden_states):
+        hidden_states = self.dense(hidden_states)
+        hidden_states = self.activation(hidden_states)
+        hidden_states = self.LayerNorm(hidden_states)
+        hidden_states = self.decoder(hidden_states)
+
+        prediction_scores = hidden_states + self.bias
+
+        return prediction_scores
+
 
 @add_start_docstrings("Bert Model with a `language modeling` head on top.", ALBERT_START_DOCSTRING, ALBERT_INPUTS_DOCSTRING)
 class AlbertForMaskedLM(BertPreTrainedModel):
@@ -433,13 +453,8 @@ class AlbertForMaskedLM(BertPreTrainedModel):
     def __init__(self, config):
         super(AlbertForMaskedLM, self).__init__(config)
 
-        self.config = config
         self.albert = AlbertModel(config)
-        self.LayerNorm = nn.LayerNorm(config.embedding_size)
-        self.bias = nn.Parameter(torch.zeros(config.vocab_size))
-        self.dense = nn.Linear(config.hidden_size, config.embedding_size)
-        self.decoder = nn.Linear(config.embedding_size, config.vocab_size)
-        self.activation = ACT2FN[config.hidden_act]
+        self.predictions = AlbertMLMHead(config)
 
         self.init_weights()
         self.tie_weights()
@@ -448,17 +463,15 @@ class AlbertForMaskedLM(BertPreTrainedModel):
         """ Make sure we are sharing the input and output embeddings.
             Export to TorchScript can't handle parameter sharing so we are cloning them instead.
         """
-        self._tie_or_clone_weights(self.decoder,
+        self._tie_or_clone_weights(self.predictions.decoder,
                                    self.albert.embeddings.word_embeddings)
 
     def forward(self, input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None,
                 masked_lm_labels=None):
         outputs = self.albert(input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None)
         sequence_outputs = outputs[0]
-        hidden_states = self.dense(sequence_outputs)
-        hidden_states = self.activation(hidden_states)
-        hidden_states = self.LayerNorm(hidden_states)
-        prediction_scores = self.decoder(hidden_states)
+
+        prediction_scores = self.predictions(sequence_outputs)
 
         outputs = (prediction_scores,) + outputs[2:]  # Add hidden states and attention if they are here
         if masked_lm_labels is not None:

From 4f3a54bfc8fa8749f6d5b29f110148738a646fcd Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Thu, 31 Oct 2019 16:37:34 +0000
Subject: [PATCH 102/505] ALBERT can load pre-trained models. Doesn't inherit
 from BERT anymore.

---
 transformers/__init__.py             |  2 +-
 transformers/configuration_albert.py |  9 ++++++
 transformers/modeling_albert.py      | 44 +++++++++++++++++++++++-----
 transformers/tokenization_albert.py  | 25 +++++++++++++---
 4 files changed, 68 insertions(+), 12 deletions(-)

diff --git a/transformers/__init__.py b/transformers/__init__.py
index 152d520e7b..bdfb1a0922 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -107,7 +107,7 @@ if is_torch_available():
                                 CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
     from .modeling_encoder_decoder import PreTrainedEncoderDecoder, Model2Model
 
-    from .modeling_albert import (AlbertModel, AlbertForMaskedLM)
+    from .modeling_albert import (AlbertModel, AlbertForMaskedLM, ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
 
     # Optimization
     from .optimization import (AdamW, get_constant_schedule, get_constant_schedule_with_warmup, get_cosine_schedule_with_warmup,
diff --git a/transformers/configuration_albert.py b/transformers/configuration_albert.py
index b72bbb971e..c35426768f 100644
--- a/transformers/configuration_albert.py
+++ b/transformers/configuration_albert.py
@@ -17,12 +17,21 @@
 
 from .configuration_utils import PretrainedConfig
 
+ALBERT_PRETRAINED_CONFIG_ARCHIVE_MAP = {
+    'albert-base': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-base-config.json",
+    'albert-large': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-large-config.json",
+    'albert-xlarge': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xlarge-config.json",
+    'albert-xxlarge': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xxlarge-config.json",
+}
+
 class AlbertConfig(PretrainedConfig):
     """Configuration for `AlbertModel`.
 
     The default settings match the configuration of model `albert_xxlarge`.
     """
 
+    pretrained_config_archive_map = ALBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
+
     def __init__(self,
                  vocab_size_or_config_json_file=30000,
                  embedding_size=128,
diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index 487455e561..4da10ed1cb 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -21,6 +21,7 @@ import logging
 import torch
 import torch.nn as nn
 from torch.nn import CrossEntropyLoss
+from transformers.modeling_utils import PreTrainedModel
 from transformers.configuration_albert import AlbertConfig
 from transformers.modeling_bert import BertEmbeddings, BertPreTrainedModel, BertModel, BertSelfAttention, prune_linear_layer, ACT2FN
 from .file_utils import add_start_docstrings
@@ -274,6 +275,29 @@ class AlbertTransformer(nn.Module):
         return outputs  # last-layer hidden state, (all hidden states), (all attentions)
 
 
+
+class AlbertPreTrainedModel(PreTrainedModel):
+    """ An abstract class to handle weights initialization and
+        a simple interface for dowloading and loading pretrained models.
+    """
+    config_class = AlbertConfig
+    pretrained_model_archive_map = ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP
+    base_model_prefix = "albert"
+
+    def _init_weights(self, module):
+        """ Initialize the weights.
+        """
+        if isinstance(module, (nn.Linear, nn.Embedding)):
+            # Slightly different from the TF version which uses truncated_normal for initialization
+            # cf https://github.com/pytorch/pytorch/pull/5617
+            module.weight.data.normal_(mean=0.0, std=self.config.initializer_range)
+            if isinstance(module, (nn.Linear)) and module.bias is not None:
+                module.bias.data.zero_()
+        elif isinstance(module, nn.LayerNorm):
+            module.bias.data.zero_()
+            module.weight.data.fill_(1.0)
+
+
 ALBERT_START_DOCSTRING = r"""    The ALBERT model was proposed in
     `ALBERT: A Lite BERT for Self-supervised Learning of Language Representations`_
     by Zhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut. It presents
@@ -338,7 +362,7 @@ ALBERT_INPUTS_DOCSTRING = r"""
 
 @add_start_docstrings("The bare ALBERT Model transformer outputting raw hidden-states without any specific head on top.",
                       ALBERT_START_DOCSTRING, ALBERT_INPUTS_DOCSTRING)
-class AlbertModel(BertModel):
+class AlbertModel(AlbertPreTrainedModel):
     r"""
     Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
         **last_hidden_state**: ``torch.FloatTensor`` of shape ``(batch_size, sequence_length, hidden_size)``
@@ -358,6 +382,12 @@ class AlbertModel(BertModel):
             list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
             Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
     """
+
+    config_class = AlbertConfig
+    pretrained_model_archive_map = ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP
+    load_tf_weights = load_tf_weights_in_albert
+    base_model_prefix = "albert"
+
     def __init__(self, config):
         super(AlbertModel, self).__init__(config)
 
@@ -369,6 +399,11 @@ class AlbertModel(BertModel):
 
         self.init_weights()
 
+    def _resize_token_embeddings(self, new_num_tokens):
+        old_embeddings = self.embeddings.word_embeddings
+        new_embeddings = self._get_resized_embeddings(old_embeddings, new_num_tokens)
+        self.embeddings.word_embeddings = new_embeddings
+        return self.embeddings.word_embeddings
 
     def forward(self, input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None):
         if attention_mask is None:
@@ -423,7 +458,7 @@ class AlbertMLMHead(nn.Module):
 
 
 @add_start_docstrings("Bert Model with a `language modeling` head on top.", ALBERT_START_DOCSTRING, ALBERT_INPUTS_DOCSTRING)
-class AlbertForMaskedLM(BertPreTrainedModel):
+class AlbertForMaskedLM(AlbertPreTrainedModel):
     r"""
         **masked_lm_labels**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
             Labels for computing the masked language modeling loss.
@@ -445,11 +480,6 @@ class AlbertForMaskedLM(BertPreTrainedModel):
             Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
     """
 
-    config_class = AlbertConfig
-    pretrained_model_archive_map = ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP
-    load_tf_weights = load_tf_weights_in_albert
-    base_model_prefix = "albert"
-
     def __init__(self, config):
         super(AlbertForMaskedLM, self).__init__(config)
 
diff --git a/transformers/tokenization_albert.py b/transformers/tokenization_albert.py
index 7cba99b9e4..acf67c1154 100644
--- a/transformers/tokenization_albert.py
+++ b/transformers/tokenization_albert.py
@@ -15,7 +15,7 @@
 """ Tokenization classes for ALBERT model."""
 from __future__ import (absolute_import, division, print_function,
                         unicode_literals)
-                        
+
 from .tokenization_utils import PreTrainedTokenizer
 import logging
 import unicodedata
@@ -24,8 +24,25 @@ import os
 from shutil import copyfile
 
 logger = logging.getLogger(__name__)
-
 VOCAB_FILES_NAMES = {'vocab_file': 'spiece.model'}
+
+PRETRAINED_VOCAB_FILES_MAP = {
+    'vocab_file':
+    {
+    'albert-base': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-base-spiece.model",
+    'albert-large': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-large-spiece.model",
+    'albert-xlarge': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xlarge-spiece.model",
+    'albert-xxlarge': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xxlarge-spiece.model",
+    }
+}
+
+PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
+    'albert-base': 512,
+    'albert-large': 512,
+    'albert-xlarge': 512,
+    'albert-xxlarge': 512,
+}
+
 SPIECE_UNDERLINE = u'▁'
 
 class AlbertTokenizer(PreTrainedTokenizer):
@@ -35,8 +52,8 @@ class AlbertTokenizer(PreTrainedTokenizer):
             - requires `SentencePiece <https://github.com/google/sentencepiece>`_
     """
     vocab_files_names = VOCAB_FILES_NAMES
-    # pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
-    # max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
+    pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+    max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
 
     def __init__(self, vocab_file,
                  do_lower_case=True, remove_space=True, keep_accents=False,

From c9875455929a63f12b81e2dcc8fe30f72137c06b Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Thu, 31 Oct 2019 18:48:02 +0000
Subject: [PATCH 103/505] Converting script

---
 transformers/__init__.py                      |  2 +-
 ...lbert_original_tf_checkpoint_to_pytorch.py | 36 +++++++++++++++----
 2 files changed, 30 insertions(+), 8 deletions(-)

diff --git a/transformers/__init__.py b/transformers/__init__.py
index bdfb1a0922..db98d5fd44 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -107,7 +107,7 @@ if is_torch_available():
                                 CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
     from .modeling_encoder_decoder import PreTrainedEncoderDecoder, Model2Model
 
-    from .modeling_albert import (AlbertModel, AlbertForMaskedLM, ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
+    from .modeling_albert import (AlbertModel, AlbertForMaskedLM, load_tf_weights_in_albert, ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
 
     # Optimization
     from .optimization import (AdamW, get_constant_schedule, get_constant_schedule_with_warmup, get_cosine_schedule_with_warmup,
diff --git a/transformers/convert_albert_original_tf_checkpoint_to_pytorch.py b/transformers/convert_albert_original_tf_checkpoint_to_pytorch.py
index 04877d41b9..5bbaab8c21 100644
--- a/transformers/convert_albert_original_tf_checkpoint_to_pytorch.py
+++ b/transformers/convert_albert_original_tf_checkpoint_to_pytorch.py
@@ -1,18 +1,39 @@
+# coding=utf-8
+# Copyright 2018 The HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Convert ALBERT checkpoint."""
 
+from __future__ import absolute_import
+from __future__ import division
+from __future__ import print_function
 
+import argparse
+import torch
 
-from transformers import AlbertConfig, BertForPreTraining, load_tf_weights_in_bert
-
+from transformers import AlbertConfig, AlbertForMaskedLM, load_tf_weights_in_albert
 
+import logging
+logging.basicConfig(level=logging.INFO)
 
 def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, bert_config_file, pytorch_dump_path):
     # Initialise PyTorch model
-    config = BertConfig.from_json_file(bert_config_file)
+    config = AlbertConfig.from_json_file(bert_config_file)
     print("Building PyTorch model from configuration: {}".format(str(config)))
-    model = BertForPreTraining(config)
+    model = AlbertForMaskedLM(config)
 
     # Load weights from tf checkpoint
-    load_tf_weights_in_bert(model, config, tf_checkpoint_path)
+    load_tf_weights_in_albert(model, config, tf_checkpoint_path)
 
     # Save pytorch-model
     print("Save PyTorch model to {}".format(pytorch_dump_path))
@@ -31,7 +52,7 @@ if __name__ == "__main__":
                         default = None,
                         type = str,
                         required = True,
-                        help = "The config json file corresponding to the pre-trained BERT model. \n"
+                        help = "The config json file corresponding to the pre-trained ALBERT model. \n"
                             "This specifies the model architecture.")
     parser.add_argument("--pytorch_dump_path",
                         default = None,
@@ -40,5 +61,6 @@ if __name__ == "__main__":
                         help = "Path to the output PyTorch model.")
     args = parser.parse_args()
     convert_tf_checkpoint_to_pytorch(args.tf_checkpoint_path,
-                                     args.bert_config_file,
+                                     args.albert_config_file,
                                      args.pytorch_dump_path)
+ 
\ No newline at end of file

From 0d07a23c04c9837234cda402b32246d7581e5bc4 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Fri, 1 Nov 2019 15:07:01 +0000
Subject: [PATCH 104/505] LAMB implementation

---
 transformers/optimization.py | 93 ++++++++++++++++++++++++++++++++++++
 1 file changed, 93 insertions(+)

diff --git a/transformers/optimization.py b/transformers/optimization.py
index 99e6cc75e4..90f3dbca9b 100644
--- a/transformers/optimization.py
+++ b/transformers/optimization.py
@@ -167,3 +167,96 @@ class AdamW(Optimizer):
                     p.data.add_(-group['lr'] * group['weight_decay'], p.data)
 
         return loss
+
+
+
+class Lamb(Optimizer):
+    """ Implements the LAMB algorithm (Layer-wise Adaptive Moments optimizer for Batch training).
+
+    Adapted from the huggingface/transformers ADAM optimizer
+    Inspired from the Google Research implementation available in ALBERT: https://github.com/google-research/google-research/blob/master/albert/lamb_optimizer.py
+    Inspired from cybertronai's PyTorch LAMB implementation: https://github.com/cybertronai/pytorch-lamb/blob/master/pytorch_lamb/lamb.py
+
+
+    Parameters:
+        lr (float): learning rate. Default 1e-3.
+        betas (tuple of 2 floats): Adams beta parameters (b1, b2). Default: (0.9, 0.999)
+        eps (float): Adams epsilon. Default: 1e-6
+        weight_decay (float): Weight decay. Default: 0.0
+    """
+    
+    def __init__(self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-6, weight_decay=0.0, correct_bias=True):
+        if lr < 0.0:
+            raise ValueError("Invalid learning rate: {} - should be >= 0.0".format(lr))
+        if not 0.0 <= betas[0] < 1.0:
+            raise ValueError("Invalid beta parameter: {} - should be in [0.0, 1.0[".format(betas[0]))
+        if not 0.0 <= betas[1]  < 1.0:
+            raise ValueError("Invalid beta parameter: {} - should be in [0.0, 1.0[".format(betas[1]))
+        if not 0.0 <= eps:
+            raise ValueError("Invalid epsilon value: {} - should be >= 0.0".format(eps))
+        defaults = dict(lr=lr, betas=betas, eps=eps, weight_decay=weight_decay,
+                        correct_bias=correct_bias)
+        super(Lamb, self).__init__(params, defaults)
+
+    def step(self, closure=None):
+        """Performs a single optimization step.
+
+        Arguments:
+            closure (callable, optional): A closure that reevaluates the model
+                and returns the loss.
+        """
+        loss = None
+        if closure is not None:
+            loss = closure()
+
+        for group in self.param_groups:
+            for p in group['params']:
+                if p.grad is None:
+                    continue
+                grad = p.grad.data
+                if grad.is_sparse:
+                    raise RuntimeError('LAMB does not support sparse gradients.')
+
+                state = self.state[p]
+
+                # State initialization
+                if len(state) == 0:
+                    state['step'] = 0
+                    # Exponential moving average of gradient values
+                    state['exp_avg'] = torch.zeros_like(p.data)
+                    # Exponential moving average of squared gradient values
+                    state['exp_avg_sq'] = torch.zeros_like(p.data)
+
+                exp_avg, exp_avg_sq = state['exp_avg'], state['exp_avg_sq']
+                beta1, beta2 = group['betas']
+
+                state['step'] += 1
+
+                # Decay the first and second moment running average coefficient
+                # In-place operations to update the averages at the same time
+                exp_avg.mul_(beta1).add_(1.0 - beta1, grad)
+                exp_avg_sq.mul_(beta2).addcmul_(1.0 - beta2, grad, grad)
+                denom = exp_avg_sq.sqrt().add_(group['eps'])
+
+
+                # Inspired from cybertronai's PyTorch LAMB implementation: https://github.com/cybertronai/pytorch-lamb/blob/master/pytorch_lamb/lamb.py
+                step_size = group['lr']
+                weight_norm = p.data.pow(2).sum().sqrt().clamp(0, 10)
+
+                adam_step = exp_avg / exp_avg_sq.sqrt().add(group['eps'])
+                if group['weight_decay'] != 0:
+                    adam_step.add_(group['weight_decay'], p.data)
+
+                adam_norm = adam_step.pow(2).sum().sqrt()
+                if weight_norm == 0 or adam_norm == 0:
+                    trust_ratio = 1
+                else:
+                    trust_ratio = weight_norm / adam_norm
+
+                
+                state['weight_norm'] = weight_norm
+                state['adam_norm'] = adam_norm
+                state['trust_ratio'] = trust_ratio
+
+                p.data.add_(-step_size * trust_ratio, adam_step)
+        return loss

From 6637a77f807615ef2427c0390a015f4eb4814fb4 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Fri, 1 Nov 2019 15:17:31 +0000
Subject: [PATCH 105/505] AlbertForSequenceClassification

---
 transformers/__init__.py        |  3 +-
 transformers/modeling_albert.py | 77 ++++++++++++++++++++++++++++++++-
 2 files changed, 77 insertions(+), 3 deletions(-)

diff --git a/transformers/__init__.py b/transformers/__init__.py
index db98d5fd44..51995942ce 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -107,7 +107,8 @@ if is_torch_available():
                                 CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
     from .modeling_encoder_decoder import PreTrainedEncoderDecoder, Model2Model
 
-    from .modeling_albert import (AlbertModel, AlbertForMaskedLM, load_tf_weights_in_albert, ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
+    from .modeling_albert import (AlbertModel, AlbertForMaskedLM, AlbertForSequenceClassification,
+                                load_tf_weights_in_albert, ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
 
     # Optimization
     from .optimization import (AdamW, get_constant_schedule, get_constant_schedule_with_warmup, get_cosine_schedule_with_warmup,
diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index 4da10ed1cb..bba6767079 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -20,10 +20,10 @@ import math
 import logging
 import torch
 import torch.nn as nn
-from torch.nn import CrossEntropyLoss
+from torch.nn import CrossEntropyLoss, MSELoss
 from transformers.modeling_utils import PreTrainedModel
 from transformers.configuration_albert import AlbertConfig
-from transformers.modeling_bert import BertEmbeddings, BertPreTrainedModel, BertModel, BertSelfAttention, prune_linear_layer, ACT2FN
+from transformers.modeling_bert import BertEmbeddings, BertSelfAttention, prune_linear_layer, ACT2FN
 from .file_utils import add_start_docstrings
 
 logger = logging.getLogger(__name__)
@@ -510,3 +510,76 @@ class AlbertForMaskedLM(AlbertPreTrainedModel):
             outputs = (masked_lm_loss,) + outputs
 
         return outputs
+
+
+@add_start_docstrings("""Albert Model transformer with a sequence classification/regression head on top (a linear layer on top of
+    the pooled output) e.g. for GLUE tasks. """,
+    ALBERT_START_DOCSTRING, ALBERT_INPUTS_DOCSTRING)
+class AlbertForSequenceClassification(AlbertPreTrainedModel):
+    r"""
+        **labels**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size,)``:
+            Labels for computing the sequence classification/regression loss.
+            Indices should be in ``[0, ..., config.num_labels - 1]``.
+            If ``config.num_labels == 1`` a regression loss is computed (Mean-Square loss),
+            If ``config.num_labels > 1`` a classification loss is computed (Cross-Entropy).
+
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **loss**: (`optional`, returned when ``labels`` is provided) ``torch.FloatTensor`` of shape ``(1,)``:
+            Classification (or regression if config.num_labels==1) loss.
+        **logits**: ``torch.FloatTensor`` of shape ``(batch_size, config.num_labels)``
+            Classification (or regression if config.num_labels==1) scores (before SoftMax).
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``torch.FloatTensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        tokenizer = AlbertTokenizer.from_pretrained('albert-base')
+        model = AlbertForSequenceClassification.from_pretrained('albert-base')
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        labels = torch.tensor([1]).unsqueeze(0)  # Batch size 1
+        outputs = model(input_ids, labels=labels)
+        loss, logits = outputs[:2]
+
+    """
+    def __init__(self, config):
+        super(AlbertForSequenceClassification, self).__init__(config)
+        self.num_labels = config.num_labels
+
+        self.albert = AlbertModel(config)
+        self.dropout = nn.Dropout(config.hidden_dropout_prob)
+        self.classifier = nn.Linear(config.hidden_size, self.config.num_labels)
+
+        self.init_weights()
+
+    def forward(self, input_ids, attention_mask=None, token_type_ids=None,
+                position_ids=None, head_mask=None, labels=None):
+
+        outputs = self.albert(input_ids,
+                            attention_mask=attention_mask,
+                            token_type_ids=token_type_ids,
+                            position_ids=position_ids, 
+                            head_mask=head_mask)
+
+        pooled_output = outputs[1]
+
+        pooled_output = self.dropout(pooled_output)
+        logits = self.classifier(pooled_output)
+
+        outputs = (logits,) + outputs[2:]  # add hidden states and attention if they are here
+
+        if labels is not None:
+            if self.num_labels == 1:
+                #  We are doing regression
+                loss_fct = MSELoss()
+                loss = loss_fct(logits.view(-1), labels.view(-1))
+            else:
+                loss_fct = CrossEntropyLoss()
+                loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
+            outputs = (loss,) + outputs
+
+        return outputs  # (loss), logits, (hidden_states), (attentions)
\ No newline at end of file

From c110c41fdb3b363528336ceda3b9fb46f026ad9c Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Fri, 1 Nov 2019 21:59:40 +0000
Subject: [PATCH 106/505] Run GLUE and remove LAMB

---
 examples/run_glue.py         | 14 ++++--
 transformers/optimization.py | 93 ------------------------------------
 2 files changed, 10 insertions(+), 97 deletions(-)

diff --git a/examples/run_glue.py b/examples/run_glue.py
index 527e440075..550a0b8175 100644
--- a/examples/run_glue.py
+++ b/examples/run_glue.py
@@ -47,7 +47,11 @@ from transformers import (WEIGHTS_NAME, BertConfig,
                                   XLNetTokenizer,
                                   DistilBertConfig,
                                   DistilBertForSequenceClassification,
-                                  DistilBertTokenizer)
+                                  DistilBertTokenizer,
+                                  AlbertConfig,
+                                  AlbertForSequenceClassification, 
+                                  AlbertTokenizer,
+                                )
 
 from transformers import AdamW, get_linear_schedule_with_warmup
 
@@ -66,7 +70,8 @@ MODEL_CLASSES = {
     'xlnet': (XLNetConfig, XLNetForSequenceClassification, XLNetTokenizer),
     'xlm': (XLMConfig, XLMForSequenceClassification, XLMTokenizer),
     'roberta': (RobertaConfig, RobertaForSequenceClassification, RobertaTokenizer),
-    'distilbert': (DistilBertConfig, DistilBertForSequenceClassification, DistilBertTokenizer)
+    'distilbert': (DistilBertConfig, DistilBertForSequenceClassification, DistilBertTokenizer),
+    'albert': (AlbertConfig, AlbertForSequenceClassification, AlbertTokenizer)
 }
 
 
@@ -99,6 +104,7 @@ def train(args, train_dataset, model, tokenizer):
         {'params': [p for n, p in model.named_parameters() if not any(nd in n for nd in no_decay)], 'weight_decay': args.weight_decay},
         {'params': [p for n, p in model.named_parameters() if any(nd in n for nd in no_decay)], 'weight_decay': 0.0}
         ]
+
     optimizer = AdamW(optimizer_grouped_parameters, lr=args.learning_rate, eps=args.adam_epsilon)
     scheduler = get_linear_schedule_with_warmup(optimizer, num_warmup_steps=args.warmup_steps, num_training_steps=t_total)
     if args.fp16:
@@ -317,7 +323,7 @@ def load_and_cache_examples(args, task, tokenizer, evaluate=False):
         all_labels = torch.tensor([f.label for f in features], dtype=torch.long)
     elif output_mode == "regression":
         all_labels = torch.tensor([f.label for f in features], dtype=torch.float)
-
+ 
     dataset = TensorDataset(all_input_ids, all_attention_mask, all_token_type_ids, all_labels)
     return dataset
 
@@ -361,7 +367,7 @@ def main():
     parser.add_argument("--per_gpu_eval_batch_size", default=8, type=int,
                         help="Batch size per GPU/CPU for evaluation.")
     parser.add_argument('--gradient_accumulation_steps', type=int, default=1,
-                        help="Number of updates steps to accumulate before performing a backward/update pass.")
+                        help="Number of updates steps to accumulate before performing a backward/update pass.")     
     parser.add_argument("--learning_rate", default=5e-5, type=float,
                         help="The initial learning rate for Adam.")
     parser.add_argument("--weight_decay", default=0.0, type=float,
diff --git a/transformers/optimization.py b/transformers/optimization.py
index 90f3dbca9b..99e6cc75e4 100644
--- a/transformers/optimization.py
+++ b/transformers/optimization.py
@@ -167,96 +167,3 @@ class AdamW(Optimizer):
                     p.data.add_(-group['lr'] * group['weight_decay'], p.data)
 
         return loss
-
-
-
-class Lamb(Optimizer):
-    """ Implements the LAMB algorithm (Layer-wise Adaptive Moments optimizer for Batch training).
-
-    Adapted from the huggingface/transformers ADAM optimizer
-    Inspired from the Google Research implementation available in ALBERT: https://github.com/google-research/google-research/blob/master/albert/lamb_optimizer.py
-    Inspired from cybertronai's PyTorch LAMB implementation: https://github.com/cybertronai/pytorch-lamb/blob/master/pytorch_lamb/lamb.py
-
-
-    Parameters:
-        lr (float): learning rate. Default 1e-3.
-        betas (tuple of 2 floats): Adams beta parameters (b1, b2). Default: (0.9, 0.999)
-        eps (float): Adams epsilon. Default: 1e-6
-        weight_decay (float): Weight decay. Default: 0.0
-    """
-    
-    def __init__(self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-6, weight_decay=0.0, correct_bias=True):
-        if lr < 0.0:
-            raise ValueError("Invalid learning rate: {} - should be >= 0.0".format(lr))
-        if not 0.0 <= betas[0] < 1.0:
-            raise ValueError("Invalid beta parameter: {} - should be in [0.0, 1.0[".format(betas[0]))
-        if not 0.0 <= betas[1]  < 1.0:
-            raise ValueError("Invalid beta parameter: {} - should be in [0.0, 1.0[".format(betas[1]))
-        if not 0.0 <= eps:
-            raise ValueError("Invalid epsilon value: {} - should be >= 0.0".format(eps))
-        defaults = dict(lr=lr, betas=betas, eps=eps, weight_decay=weight_decay,
-                        correct_bias=correct_bias)
-        super(Lamb, self).__init__(params, defaults)
-
-    def step(self, closure=None):
-        """Performs a single optimization step.
-
-        Arguments:
-            closure (callable, optional): A closure that reevaluates the model
-                and returns the loss.
-        """
-        loss = None
-        if closure is not None:
-            loss = closure()
-
-        for group in self.param_groups:
-            for p in group['params']:
-                if p.grad is None:
-                    continue
-                grad = p.grad.data
-                if grad.is_sparse:
-                    raise RuntimeError('LAMB does not support sparse gradients.')
-
-                state = self.state[p]
-
-                # State initialization
-                if len(state) == 0:
-                    state['step'] = 0
-                    # Exponential moving average of gradient values
-                    state['exp_avg'] = torch.zeros_like(p.data)
-                    # Exponential moving average of squared gradient values
-                    state['exp_avg_sq'] = torch.zeros_like(p.data)
-
-                exp_avg, exp_avg_sq = state['exp_avg'], state['exp_avg_sq']
-                beta1, beta2 = group['betas']
-
-                state['step'] += 1
-
-                # Decay the first and second moment running average coefficient
-                # In-place operations to update the averages at the same time
-                exp_avg.mul_(beta1).add_(1.0 - beta1, grad)
-                exp_avg_sq.mul_(beta2).addcmul_(1.0 - beta2, grad, grad)
-                denom = exp_avg_sq.sqrt().add_(group['eps'])
-
-
-                # Inspired from cybertronai's PyTorch LAMB implementation: https://github.com/cybertronai/pytorch-lamb/blob/master/pytorch_lamb/lamb.py
-                step_size = group['lr']
-                weight_norm = p.data.pow(2).sum().sqrt().clamp(0, 10)
-
-                adam_step = exp_avg / exp_avg_sq.sqrt().add(group['eps'])
-                if group['weight_decay'] != 0:
-                    adam_step.add_(group['weight_decay'], p.data)
-
-                adam_norm = adam_step.pow(2).sum().sqrt()
-                if weight_norm == 0 or adam_norm == 0:
-                    trust_ratio = 1
-                else:
-                    trust_ratio = weight_norm / adam_norm
-
-                
-                state['weight_norm'] = weight_norm
-                state['adam_norm'] = adam_norm
-                state['trust_ratio'] = trust_ratio
-
-                p.data.add_(-step_size * trust_ratio, adam_step)
-        return loss

From 70d99980ded1565c9e8efa2cd04c21572b664f2f Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Mon, 4 Nov 2019 11:34:30 -0500
Subject: [PATCH 107/505] ALBERT-V2

---
 transformers/configuration_albert.py          | 12 ++++++----
 ...lbert_original_tf_checkpoint_to_pytorch.py |  5 ++--
 transformers/modeling_albert.py               | 16 ++++++++-----
 transformers/tokenization_albert.py           | 24 ++++++++++++-------
 4 files changed, 37 insertions(+), 20 deletions(-)

diff --git a/transformers/configuration_albert.py b/transformers/configuration_albert.py
index c35426768f..de665c9b1c 100644
--- a/transformers/configuration_albert.py
+++ b/transformers/configuration_albert.py
@@ -18,10 +18,14 @@
 from .configuration_utils import PretrainedConfig
 
 ALBERT_PRETRAINED_CONFIG_ARCHIVE_MAP = {
-    'albert-base': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-base-config.json",
-    'albert-large': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-large-config.json",
-    'albert-xlarge': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xlarge-config.json",
-    'albert-xxlarge': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xxlarge-config.json",
+    'albert-base-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-base-config.json",
+    'albert-large-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-large-config.json",
+    'albert-xlarge-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xlarge-config.json",
+    'albert-xxlarge-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xxlarge-config.json",
+    'albert-base-v2': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-base-v2-config.json",
+    'albert-large-v2': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-large-v2-config.json",
+    'albert-xlarge-v2': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xlarge-v2-config.json",
+    'albert-xxlarge-v2': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xxlarge-v2-config.json",
 }
 
 class AlbertConfig(PretrainedConfig):
diff --git a/transformers/convert_albert_original_tf_checkpoint_to_pytorch.py b/transformers/convert_albert_original_tf_checkpoint_to_pytorch.py
index 5bbaab8c21..b6476b4fb6 100644
--- a/transformers/convert_albert_original_tf_checkpoint_to_pytorch.py
+++ b/transformers/convert_albert_original_tf_checkpoint_to_pytorch.py
@@ -26,9 +26,10 @@ from transformers import AlbertConfig, AlbertForMaskedLM, load_tf_weights_in_alb
 import logging
 logging.basicConfig(level=logging.INFO)
 
-def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, bert_config_file, pytorch_dump_path):
+
+def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, albert_config_file, pytorch_dump_path):
     # Initialise PyTorch model
-    config = AlbertConfig.from_json_file(bert_config_file)
+    config = AlbertConfig.from_json_file(albert_config_file)
     print("Building PyTorch model from configuration: {}".format(str(config)))
     model = AlbertForMaskedLM(config)
 
diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index bba6767079..51cb0a6d23 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -30,10 +30,14 @@ logger = logging.getLogger(__name__)
 
 
 ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
-    'albert-base': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-base-pytorch_model.bin",
-    'albert-large': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-large-pytorch_model.bin",
-    'albert-xlarge': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xlarge-pytorch_model.bin",
-    'albert-xxlarge': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xxlarge-pytorch_model.bin",
+    'albert-base-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-base-pytorch_model.bin",
+    'albert-large-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-large-pytorch_model.bin",
+    'albert-xlarge-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xlarge-pytorch_model.bin",
+    'albert-xxlarge-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xxlarge-pytorch_model.bin",
+    'albert-base-v2': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-base-v2-pytorch_model.bin",
+    'albert-large-v2': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-large-v2-pytorch_model.bin",
+    'albert-xlarge-v2': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xlarge-v2-pytorch_model.bin",
+    'albert-xxlarge-v2': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xxlarge-v2-pytorch_model.bin",
 }
 
 
@@ -538,8 +542,8 @@ class AlbertForSequenceClassification(AlbertPreTrainedModel):
 
     Examples::
 
-        tokenizer = AlbertTokenizer.from_pretrained('albert-base')
-        model = AlbertForSequenceClassification.from_pretrained('albert-base')
+        tokenizer = AlbertTokenizer.from_pretrained('albert-base-v2')
+        model = AlbertForSequenceClassification.from_pretrained('albert-base-v2')
         input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
         labels = torch.tensor([1]).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids, labels=labels)
diff --git a/transformers/tokenization_albert.py b/transformers/tokenization_albert.py
index acf67c1154..2f9af0b0bc 100644
--- a/transformers/tokenization_albert.py
+++ b/transformers/tokenization_albert.py
@@ -29,18 +29,26 @@ VOCAB_FILES_NAMES = {'vocab_file': 'spiece.model'}
 PRETRAINED_VOCAB_FILES_MAP = {
     'vocab_file':
     {
-    'albert-base': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-base-spiece.model",
-    'albert-large': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-large-spiece.model",
-    'albert-xlarge': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xlarge-spiece.model",
-    'albert-xxlarge': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xxlarge-spiece.model",
+        'albert-base-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-base-spiece.model",
+        'albert-large-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-large-spiece.model",
+        'albert-xlarge-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xlarge-spiece.model",
+        'albert-xxlarge-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xxlarge-spiece.model",
+        'albert-base-v2': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-base-v2-spiece.model",
+        'albert-large-v2': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-large-v2-spiece.model",
+        'albert-xlarge-v2': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xlarge-v2-spiece.model",
+        'albert-xxlarge-v2': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xxlarge-v2-spiece.model",
     }
 }
 
 PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
-    'albert-base': 512,
-    'albert-large': 512,
-    'albert-xlarge': 512,
-    'albert-xxlarge': 512,
+    'albert-base-v1': 512,
+    'albert-large-v1': 512,
+    'albert-xlarge-v1': 512,
+    'albert-xxlarge-v1': 512,
+    'albert-base-v2': 512,
+    'albert-large-v2': 512,
+    'albert-xlarge-v2': 512,
+    'albert-xxlarge-v2': 512,
 }
 
 SPIECE_UNDERLINE = u'▁'

From 4374eaea786399a948b4b34d39fc614ded5e1de6 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Wed, 6 Nov 2019 20:47:42 +0000
Subject: [PATCH 108/505] ALBERT for SQuAD

---
 examples/run_squad.py           | 12 +++--
 transformers/__init__.py        |  1 +
 transformers/modeling_albert.py | 93 ++++++++++++++++++++++++++++++++-
 3 files changed, 100 insertions(+), 6 deletions(-)

diff --git a/examples/run_squad.py b/examples/run_squad.py
index 69088d73c3..59683c0668 100644
--- a/examples/run_squad.py
+++ b/examples/run_squad.py
@@ -43,7 +43,8 @@ from transformers import (WEIGHTS_NAME, BertConfig,
                                   XLMTokenizer, XLNetConfig,
                                   XLNetForQuestionAnswering,
                                   XLNetTokenizer,
-                                  DistilBertConfig, DistilBertForQuestionAnswering, DistilBertTokenizer)
+                                  DistilBertConfig, DistilBertForQuestionAnswering, DistilBertTokenizer,
+                                  AlbertConfig, AlbertForQuestionAnswering, AlbertTokenizer)
 
 from transformers import AdamW, get_linear_schedule_with_warmup
 
@@ -65,7 +66,8 @@ MODEL_CLASSES = {
     'bert': (BertConfig, BertForQuestionAnswering, BertTokenizer),
     'xlnet': (XLNetConfig, XLNetForQuestionAnswering, XLNetTokenizer),
     'xlm': (XLMConfig, XLMForQuestionAnswering, XLMTokenizer),
-    'distilbert': (DistilBertConfig, DistilBertForQuestionAnswering, DistilBertTokenizer)
+    'distilbert': (DistilBertConfig, DistilBertForQuestionAnswering, DistilBertTokenizer),
+    'albert': (AlbertConfig, AlbertForQuestionAnswering, AlbertTokenizer)
 }
 
 def set_seed(args):
@@ -128,7 +130,7 @@ def train(args, train_dataset, model, tokenizer):
     logger.info("  Gradient Accumulation steps = %d", args.gradient_accumulation_steps)
     logger.info("  Total optimization steps = %d", t_total)
 
-    global_step = 0
+    global_step = 1
     tr_loss, logging_loss = 0.0, 0.0
     model.zero_grad()
     train_iterator = trange(int(args.num_train_epochs), desc="Epoch", disable=args.local_rank not in [-1, 0])
@@ -537,7 +539,7 @@ def main():
         torch.save(args, os.path.join(args.output_dir, 'training_args.bin'))
 
         # Load a trained model and vocabulary that you have fine-tuned
-        model = model_class.from_pretrained(args.output_dir)
+        model = model_class.from_pretrained(args.output_dir, force_download=True)
         tokenizer = tokenizer_class.from_pretrained(args.output_dir, do_lower_case=args.do_lower_case)
         model.to(args.device)
 
@@ -555,7 +557,7 @@ def main():
         for checkpoint in checkpoints:
             # Reload the model
             global_step = checkpoint.split('-')[-1] if len(checkpoints) > 1 else ""
-            model = model_class.from_pretrained(checkpoint)
+            model = model_class.from_pretrained(checkpoint, force_download=True)
             model.to(args.device)
 
             # Evaluate
diff --git a/transformers/__init__.py b/transformers/__init__.py
index 51995942ce..81e659329d 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -108,6 +108,7 @@ if is_torch_available():
     from .modeling_encoder_decoder import PreTrainedEncoderDecoder, Model2Model
 
     from .modeling_albert import (AlbertModel, AlbertForMaskedLM, AlbertForSequenceClassification,
+                                AlbertForQuestionAnswering,
                                 load_tf_weights_in_albert, ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
 
     # Optimization
diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index 51cb0a6d23..2540218e69 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -586,4 +586,95 @@ class AlbertForSequenceClassification(AlbertPreTrainedModel):
                 loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
             outputs = (loss,) + outputs
 
-        return outputs  # (loss), logits, (hidden_states), (attentions)
\ No newline at end of file
+        return outputs  # (loss), logits, (hidden_states), (attentions)
+
+
+
+@add_start_docstrings("""Albert Model with a span classification head on top for extractive question-answering tasks like SQuAD (a linear layers on top of
+    the hidden-states output to compute `span start logits` and `span end logits`). """,
+    ALBERT_START_DOCSTRING, ALBERT_INPUTS_DOCSTRING)
+class AlbertForQuestionAnswering(AlbertPreTrainedModel):
+    r"""
+        **start_positions**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size,)``:
+            Labels for position (index) of the start of the labelled span for computing the token classification loss.
+            Positions are clamped to the length of the sequence (`sequence_length`).
+            Position outside of the sequence are not taken into account for computing the loss.
+        **end_positions**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size,)``:
+            Labels for position (index) of the end of the labelled span for computing the token classification loss.
+            Positions are clamped to the length of the sequence (`sequence_length`).
+            Position outside of the sequence are not taken into account for computing the loss.
+
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **loss**: (`optional`, returned when ``labels`` is provided) ``torch.FloatTensor`` of shape ``(1,)``:
+            Total span extraction loss is the sum of a Cross-Entropy for the start and end positions.
+        **start_scores**: ``torch.FloatTensor`` of shape ``(batch_size, sequence_length,)``
+            Span-start scores (before SoftMax).
+        **end_scores**: ``torch.FloatTensor`` of shape ``(batch_size, sequence_length,)``
+            Span-end scores (before SoftMax).
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``torch.FloatTensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        tokenizer = AlbertTokenizer.from_pretrained('albert-base-v2')
+        model = AlbertForQuestionAnswering.from_pretrained('albert-base-v2')
+        question, text = "Who was Jim Henson?", "Jim Henson was a nice puppet"
+        input_text = "[CLS] " + question + " [SEP] " + text + " [SEP]"
+        input_ids = tokenizer.encode(input_text)
+        token_type_ids = [0 if i <= input_ids.index(102) else 1 for i in range(len(input_ids))] 
+        start_scores, end_scores = model(torch.tensor([input_ids]), token_type_ids=torch.tensor([token_type_ids]))
+        all_tokens = tokenizer.convert_ids_to_tokens(input_ids)  
+        print(' '.join(all_tokens[torch.argmax(start_scores) : torch.argmax(end_scores)+1]))
+        # a nice puppet
+
+
+    """
+    def __init__(self, config):
+        super(AlbertForQuestionAnswering, self).__init__(config)
+        self.num_labels = config.num_labels
+
+        self.albert = AlbertModel(config)
+        self.qa_outputs = nn.Linear(config.hidden_size, config.num_labels)
+
+        self.init_weights()
+
+    def forward(self, input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None,
+                start_positions=None, end_positions=None):
+
+        outputs = self.albert(input_ids,
+                            attention_mask=attention_mask,
+                            token_type_ids=token_type_ids,
+                            position_ids=position_ids, 
+                            head_mask=head_mask)
+
+        sequence_output = outputs[0]
+
+        logits = self.qa_outputs(sequence_output)
+        start_logits, end_logits = logits.split(1, dim=-1)
+        start_logits = start_logits.squeeze(-1)
+        end_logits = end_logits.squeeze(-1)
+
+        outputs = (start_logits, end_logits,) + outputs[2:]
+        if start_positions is not None and end_positions is not None:
+            # If we are on multi-GPU, split add a dimension
+            if len(start_positions.size()) > 1:
+                start_positions = start_positions.squeeze(-1)
+            if len(end_positions.size()) > 1:
+                end_positions = end_positions.squeeze(-1)
+            # sometimes the start/end positions are outside our model inputs, we ignore these terms
+            ignored_index = start_logits.size(1)
+            start_positions.clamp_(0, ignored_index)
+            end_positions.clamp_(0, ignored_index)
+
+            loss_fct = CrossEntropyLoss(ignore_index=ignored_index)
+            start_loss = loss_fct(start_logits, start_positions)
+            end_loss = loss_fct(end_logits, end_positions)
+            total_loss = (start_loss + end_loss) / 2
+            outputs = (total_loss,) + outputs
+
+        return outputs  # (loss), start_logits, end_logits, (hidden_states), (attentions)

From abb23a78bab17ec09dde4635c32f4aa21c15fa83 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Thu, 7 Nov 2019 17:09:16 +0000
Subject: [PATCH 109/505] Head pruning for ALBERT

---
 transformers/modeling_albert.py            | 42 ++++++++++++++++++++++
 transformers/tests/modeling_albert_test.py | 12 ++++---
 2 files changed, 49 insertions(+), 5 deletions(-)

diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index 2540218e69..89ece4b61e 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -145,6 +145,29 @@ class AlbertAttention(BertSelfAttention):
         self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
         self.pruned_heads = set()
 
+    def prune_heads(self, heads):
+        if len(heads) == 0:
+            return
+        mask = torch.ones(self.num_attention_heads, self.attention_head_size)
+        heads = set(heads) - self.pruned_heads  # Convert to set and emove already pruned heads
+        for head in heads:
+            # Compute how many pruned heads are before the head and move the index accordingly
+            head = head - sum(1 if h < head else 0 for h in self.pruned_heads)
+            mask[head] = 0
+        mask = mask.view(-1).contiguous().eq(1)
+        index = torch.arange(len(mask))[mask].long()
+
+        # Prune linear layers
+        self.query = prune_linear_layer(self.query, index)
+        self.key = prune_linear_layer(self.key, index)
+        self.value = prune_linear_layer(self.value, index)
+        self.dense = prune_linear_layer(self.dense, index, dim=1)
+
+        # Update hyper params and store pruned heads
+        self.num_attention_heads = self.num_attention_heads - len(heads)
+        self.all_head_size = self.attention_head_size * self.num_attention_heads
+        self.pruned_heads = self.pruned_heads.union(heads)
+
     def forward(self, input_ids, attention_mask=None, head_mask=None):
         mixed_query_layer = self.query(input_ids)
         mixed_key_layer = self.key(input_ids)
@@ -409,6 +432,25 @@ class AlbertModel(AlbertPreTrainedModel):
         self.embeddings.word_embeddings = new_embeddings
         return self.embeddings.word_embeddings
 
+    def _prune_heads(self, heads_to_prune):
+        """ Prunes heads of the model.
+            heads_to_prune: dict of {layer_num: list of heads to prune in this layer}
+            ALBERT has a different architecture in that its layers are shared across groups, which then has inner groups.
+            If an ALBERT model has 12 hidden layers and 2 hidden groups, with two inner groups, there
+            is a total of 4 different layers.
+
+            These layers are flattened: the indices [0,1] correspond to the two inner groups of the first hidden layer,
+            while [2,3] correspond to the two inner groups of the second hidden layer.
+
+            Any layer with in index other than [0,1,2,3] will result in an error.
+            See base class PreTrainedModel for more information about head pruning
+        """
+        for layer, heads in heads_to_prune.items():
+            group_idx = int(layer / self.config.inner_group_num)
+            inner_group_idx = int(layer - group_idx * self.config.inner_group_num)
+            self.encoder.albert_layer_groups[group_idx].albert_layers[inner_group_idx].attention.prune_heads(heads)
+
+
     def forward(self, input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None):
         if attention_mask is None:
             attention_mask = torch.ones_like(input_ids)
diff --git a/transformers/tests/modeling_albert_test.py b/transformers/tests/modeling_albert_test.py
index 46a2eeb729..979e0488eb 100644
--- a/transformers/tests/modeling_albert_test.py
+++ b/transformers/tests/modeling_albert_test.py
@@ -35,7 +35,6 @@ else:
 class AlbertModelTest(CommonTestCases.CommonModelTester):
 
     all_model_classes = (AlbertModel, AlbertForMaskedLM) if is_torch_available() else ()
-    test_pruning = False
     test_head_masking = False
 
     class AlbertModelTester(object):
@@ -49,9 +48,10 @@ class AlbertModelTest(CommonTestCases.CommonModelTester):
                      use_token_type_ids=True,
                      use_labels=True,
                      vocab_size=99,
-                     hidden_size=32,
-                     num_hidden_layers=5,
-                     num_attention_heads=4,
+                     hidden_size=36,
+                     num_hidden_layers=6,
+                     num_hidden_groups=6,
+                     num_attention_heads=6,
                      intermediate_size=37,
                      hidden_act="gelu",
                      hidden_dropout_prob=0.1,
@@ -86,6 +86,7 @@ class AlbertModelTest(CommonTestCases.CommonModelTester):
             self.num_labels = num_labels
             self.num_choices = num_choices
             self.scope = scope
+            self.num_hidden_groups = num_hidden_groups
 
         def prepare_config_and_inputs(self):
             input_ids = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)
@@ -117,7 +118,8 @@ class AlbertModelTest(CommonTestCases.CommonModelTester):
                 attention_probs_dropout_prob=self.attention_probs_dropout_prob,
                 max_position_embeddings=self.max_position_embeddings,
                 type_vocab_size=self.type_vocab_size,
-                initializer_range=self.initializer_range)
+                initializer_range=self.initializer_range,
+                num_hidden_groups=self.num_hidden_groups)
 
             return config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
 

From 16263f9685eaf459408f33c9790a967012b93fa5 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Thu, 7 Nov 2019 17:29:29 +0000
Subject: [PATCH 110/505] Headmasking

---
 transformers/modeling_albert.py            | 11 ++++++-----
 transformers/tests/modeling_albert_test.py |  1 -
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index 89ece4b61e..6682930d89 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -224,7 +224,7 @@ class AlbertLayer(nn.Module):
         self.activation = ACT2FN[config.hidden_act]
 
     def forward(self, hidden_states, attention_mask=None, head_mask=None):
-        attention_output = self.attention(hidden_states, attention_mask)
+        attention_output = self.attention(hidden_states, attention_mask, head_mask)
         ffn_output = self.ffn(attention_output[0])
         ffn_output = self.activation(ffn_output)
         ffn_output = self.ffn_output(ffn_output)
@@ -245,8 +245,8 @@ class AlbertLayerGroup(nn.Module):
         layer_hidden_states = ()
         layer_attentions = ()
 
-        for albert_layer in self.albert_layers:
-            layer_output = albert_layer(hidden_states, attention_mask, head_mask)
+        for layer_index, albert_layer in enumerate(self.albert_layers):
+            layer_output = albert_layer(hidden_states, attention_mask, head_mask[layer_index])
             hidden_states = layer_output[0]
 
             if self.output_attentions:
@@ -283,7 +283,8 @@ class AlbertTransformer(nn.Module):
 
         for layer_idx in range(self.config.num_hidden_layers):
             group_idx = int(layer_idx / self.config.num_hidden_layers * self.config.num_hidden_groups)
-            layer_group_output = self.albert_layer_groups[group_idx](hidden_states, attention_mask, head_mask)  
+            layers_per_group = int(self.config.num_hidden_layers / self.config.num_hidden_groups)
+            layer_group_output = self.albert_layer_groups[group_idx](hidden_states, attention_mask, head_mask[group_idx*layers_per_group:(group_idx+1)*layers_per_group])  
         
             hidden_states = layer_group_output[0]
 
@@ -544,7 +545,7 @@ class AlbertForMaskedLM(AlbertPreTrainedModel):
 
     def forward(self, input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None,
                 masked_lm_labels=None):
-        outputs = self.albert(input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None)
+        outputs = self.albert(input_ids, attention_mask, token_type_ids, position_ids, head_mask)
         sequence_outputs = outputs[0]
 
         prediction_scores = self.predictions(sequence_outputs)
diff --git a/transformers/tests/modeling_albert_test.py b/transformers/tests/modeling_albert_test.py
index 979e0488eb..466f473332 100644
--- a/transformers/tests/modeling_albert_test.py
+++ b/transformers/tests/modeling_albert_test.py
@@ -35,7 +35,6 @@ else:
 class AlbertModelTest(CommonTestCases.CommonModelTester):
 
     all_model_classes = (AlbertModel, AlbertForMaskedLM) if is_torch_available() else ()
-    test_head_masking = False
 
     class AlbertModelTester(object):
 

From 9d5c49546fedb3d52f76f97f4043a07e08ded918 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Thu, 7 Nov 2019 17:32:52 +0000
Subject: [PATCH 111/505] Tests for AlbertForQuestionAnswering
 AlbertForSequenceClassification

---
 transformers/tests/modeling_albert_test.py | 45 +++++++++++++++++++++-
 1 file changed, 44 insertions(+), 1 deletion(-)

diff --git a/transformers/tests/modeling_albert_test.py b/transformers/tests/modeling_albert_test.py
index 466f473332..da87709df1 100644
--- a/transformers/tests/modeling_albert_test.py
+++ b/transformers/tests/modeling_albert_test.py
@@ -26,7 +26,9 @@ from .modeling_common_test import (CommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
 
 if is_torch_available():
-    from transformers import (AlbertConfig, AlbertModel, AlbertForMaskedLM)
+    from transformers import (AlbertConfig, AlbertModel, AlbertForMaskedLM,
+                              AlbertForSequenceClassification, AlbertForQuestionAnswering,
+                              )
     from transformers.modeling_albert import ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP
 else:
     pytestmark = pytest.mark.skip("Require Torch")
@@ -157,6 +159,39 @@ class AlbertModelTest(CommonTestCases.CommonModelTester):
                 [self.batch_size, self.seq_length, self.vocab_size])
             self.check_loss_output(result)
 
+        def create_and_check_albert_for_question_answering(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
+            model = AlbertForQuestionAnswering(config=config)
+            model.eval()
+            loss, start_logits, end_logits = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids,
+                                                   start_positions=sequence_labels, end_positions=sequence_labels)
+            result = {
+                "loss": loss,
+                "start_logits": start_logits,
+                "end_logits": end_logits,
+            }
+            self.parent.assertListEqual(
+                list(result["start_logits"].size()),
+                [self.batch_size, self.seq_length])
+            self.parent.assertListEqual(
+                list(result["end_logits"].size()),
+                [self.batch_size, self.seq_length])
+            self.check_loss_output(result)
+
+
+        def create_and_check_albert_for_sequence_classification(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
+            config.num_labels = self.num_labels
+            model = AlbertForSequenceClassification(config)
+            model.eval()
+            loss, logits = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, labels=sequence_labels)
+            result = {
+                "loss": loss,
+                "logits": logits,
+            }
+            self.parent.assertListEqual(
+                list(result["logits"].size()),
+                [self.batch_size, self.num_labels])
+            self.check_loss_output(result)
+
 
         def prepare_config_and_inputs_for_common(self):
             config_and_inputs = self.prepare_config_and_inputs()
@@ -180,6 +215,14 @@ class AlbertModelTest(CommonTestCases.CommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_albert_for_masked_lm(*config_and_inputs)
 
+    def test_for_question_answering(self):
+        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+        self.model_tester.create_and_check_albert_for_question_answering(*config_and_inputs)
+
+    def test_for_sequence_classification(self):
+        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+        self.model_tester.create_and_check_albert_for_sequence_classification(*config_and_inputs)
+
     @pytest.mark.slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"

From d9daad98c744e115bfb425f316a5e7d4f405a9a5 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Thu, 7 Nov 2019 19:55:43 +0000
Subject: [PATCH 112/505] Re-ordering of group_idx/layer_idx + Python 2 tests

---
 transformers/modeling_albert.py | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index 6682930d89..640af537d0 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -281,11 +281,17 @@ class AlbertTransformer(nn.Module):
         if self.output_hidden_states:
             all_hidden_states = (hidden_states,)
 
-        for layer_idx in range(self.config.num_hidden_layers):
-            group_idx = int(layer_idx / self.config.num_hidden_layers * self.config.num_hidden_groups)
+        for i in range(self.config.num_hidden_layers):
+            # Number of layers in a hidden group
             layers_per_group = int(self.config.num_hidden_layers / self.config.num_hidden_groups)
+
+            # Index of the hidden group
+            group_idx = int(i / (self.config.num_hidden_layers / self.config.num_hidden_groups))
+
+            # Index of the layer inside the group
+            layer_idx = int(i - group_idx * layers_per_group)
+            
             layer_group_output = self.albert_layer_groups[group_idx](hidden_states, attention_mask, head_mask[group_idx*layers_per_group:(group_idx+1)*layers_per_group])  
-        
             hidden_states = layer_group_output[0]
 
             if self.output_attentions:

From f6f382532bf40a5869dc14e3eefa451646c19ded Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Thu, 7 Nov 2019 23:40:45 +0000
Subject: [PATCH 113/505] ALBERT in TF2

---
 transformers/__init__.py                      |   6 +-
 .../convert_pytorch_checkpoint_to_tf2.py      |  13 +-
 transformers/modeling_tf_albert.py            | 723 ++++++++++++++++++
 3 files changed, 736 insertions(+), 6 deletions(-)
 create mode 100644 transformers/modeling_tf_albert.py

diff --git a/transformers/__init__.py b/transformers/__init__.py
index 81e659329d..a409ef772e 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -58,8 +58,7 @@ from .configuration_ctrl import CTRLConfig, CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_xlm import XLMConfig, XLM_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_roberta import RobertaConfig, ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_distilbert import DistilBertConfig, DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
-from .configuration_albert import AlbertConfig, ALBERT
-from .configuration_albert import AlbertConfig
+from .configuration_albert import AlbertConfig, ALBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_camembert import CamembertConfig, CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
 
 # Modeling
@@ -169,6 +168,9 @@ if is_tf_available():
                                     TFCTRLLMHeadModel,
                                     TF_CTRL_PRETRAINED_MODEL_ARCHIVE_MAP)
 
+    from .modeling_tf_albert import (TFAlbertPreTrainedModel, TFAlbertModel, TFAlbertForMaskedLM,
+                                    TF_ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
+
 # TF 2.0 <=> PyTorch conversion utilities
 from .modeling_tf_pytorch_utils import (convert_tf_weight_name_to_pt_weight_name,
                                         load_pytorch_checkpoint_in_tf2_model,
diff --git a/transformers/convert_pytorch_checkpoint_to_tf2.py b/transformers/convert_pytorch_checkpoint_to_tf2.py
index e673b77dcc..d1776e9c14 100644
--- a/transformers/convert_pytorch_checkpoint_to_tf2.py
+++ b/transformers/convert_pytorch_checkpoint_to_tf2.py
@@ -33,7 +33,8 @@ from transformers import (load_pytorch_checkpoint_in_tf2_model,
                                   OpenAIGPTConfig, TFOpenAIGPTLMHeadModel, OPENAI_GPT_PRETRAINED_CONFIG_ARCHIVE_MAP,
                                   RobertaConfig, TFRobertaForMaskedLM, TFRobertaForSequenceClassification, ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP,
                                   DistilBertConfig, TFDistilBertForMaskedLM, TFDistilBertForQuestionAnswering, DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP,
-                                  CTRLConfig, TFCTRLLMHeadModel, CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP)
+                                  CTRLConfig, TFCTRLLMHeadModel, CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP,
+                                  AlbertConfig, TFAlbertForMaskedLM, ALBERT_PRETRAINED_CONFIG_ARCHIVE_MAP)
 
 if is_torch_available():
     import torch
@@ -46,7 +47,8 @@ if is_torch_available():
                                       OpenAIGPTLMHeadModel, OPENAI_GPT_PRETRAINED_MODEL_ARCHIVE_MAP,
                                       RobertaForMaskedLM, RobertaForSequenceClassification, ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP,
                                       DistilBertForMaskedLM, DistilBertForQuestionAnswering, DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP,
-                                      CTRLLMHeadModel, CTRL_PRETRAINED_MODEL_ARCHIVE_MAP)
+                                      CTRLLMHeadModel, CTRL_PRETRAINED_MODEL_ARCHIVE_MAP,
+                                      AlbertForMaskedLM, ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
 else:
     (BertForPreTraining, BertForQuestionAnswering, BertForSequenceClassification, BERT_PRETRAINED_MODEL_ARCHIVE_MAP,
     GPT2LMHeadModel, GPT2_PRETRAINED_MODEL_ARCHIVE_MAP,
@@ -56,7 +58,8 @@ else:
     OpenAIGPTLMHeadModel, OPENAI_GPT_PRETRAINED_MODEL_ARCHIVE_MAP,
     RobertaForMaskedLM, RobertaForSequenceClassification, ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP,
     DistilBertForMaskedLM, DistilBertForQuestionAnswering, DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP,
-    CTRLLMHeadModel, CTRL_PRETRAINED_MODEL_ARCHIVE_MAP) = (
+    CTRLLMHeadModel, CTRL_PRETRAINED_MODEL_ARCHIVE_MAP,
+    AlbertForMaskedLM, ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP) = (
         None, None, None, None,
         None, None,
         None, None,
@@ -65,6 +68,7 @@ else:
         None, None,
         None, None, None,
         None, None, None,
+        None, None,
         None, None)
 
 
@@ -85,7 +89,8 @@ MODEL_CLASSES = {
     'roberta-large-mnli': (RobertaConfig, TFRobertaForSequenceClassification, RobertaForSequenceClassification, ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP, ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP),
     'distilbert': (DistilBertConfig, TFDistilBertForMaskedLM, DistilBertForMaskedLM, DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP, DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP),
     'distilbert-base-uncased-distilled-squad': (DistilBertConfig, TFDistilBertForQuestionAnswering, DistilBertForQuestionAnswering, DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP, DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP),
-    'ctrl': (CTRLConfig, TFCTRLLMHeadModel, CTRLLMHeadModel, CTRL_PRETRAINED_MODEL_ARCHIVE_MAP, CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP)
+    'ctrl': (CTRLConfig, TFCTRLLMHeadModel, CTRLLMHeadModel, CTRL_PRETRAINED_MODEL_ARCHIVE_MAP, CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP),
+    'albert': (AlbertConfig, TFAlbertForMaskedLM, AlbertForMaskedLM, ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP, ALBERT_PRETRAINED_CONFIG_ARCHIVE_MAP)
 }
 
 def convert_pt_checkpoint_to_tf(model_type, pytorch_checkpoint_path, config_file, tf_dump_path, compare_with_pt_model=False, use_cached_models=True):
diff --git a/transformers/modeling_tf_albert.py b/transformers/modeling_tf_albert.py
new file mode 100644
index 0000000000..8861b7add8
--- /dev/null
+++ b/transformers/modeling_tf_albert.py
@@ -0,0 +1,723 @@
+# coding=utf-8
+# Copyright 2018 The OpenAI Team Authors and HuggingFace Inc. team.
+# Copyright (c) 2018, NVIDIA CORPORATION.  All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" TF 2.0 ALBERT model. """
+from __future__ import absolute_import, division, print_function, unicode_literals
+
+import json
+import logging
+import math
+import os
+import sys
+from io import open
+
+import numpy as np
+import tensorflow as tf
+
+from .configuration_albert import AlbertConfig
+from .modeling_tf_utils import TFPreTrainedModel, get_initializer
+from .modeling_tf_bert import ACT2FN, TFBertSelfAttention
+from .file_utils import add_start_docstrings
+
+import logging
+
+logger = logging.getLogger(__name__)
+
+TF_ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
+    # TODO FILL THAT UP
+}
+
+
+class TFAlbertEmbeddings(tf.keras.layers.Layer):
+    """Construct the embeddings from word, position and token_type embeddings.
+    """
+
+    def __init__(self, config, **kwargs):
+        super(TFAlbertEmbeddings, self).__init__(**kwargs)
+
+        self.config = config
+        self.position_embeddings = tf.keras.layers.Embedding(config.max_position_embeddings,
+                                                             config.embedding_size,
+                                                             embeddings_initializer=get_initializer(
+                                                                 self.config.initializer_range),
+                                                             name='position_embeddings')
+        self.token_type_embeddings = tf.keras.layers.Embedding(config.type_vocab_size,
+                                                               config.embedding_size,
+                                                               embeddings_initializer=get_initializer(
+                                                                   self.config.initializer_range),
+                                                               name='token_type_embeddings')
+
+        # self.LayerNorm is not snake-cased to stick with TensorFlow model variable name and be able to load
+        # any TensorFlow checkpoint file
+        self.LayerNorm = tf.keras.layers.LayerNormalization(
+            epsilon=config.layer_norm_eps, name='LayerNorm')
+        self.dropout = tf.keras.layers.Dropout(config.hidden_dropout_prob)
+
+    def build(self, input_shape):
+        """Build shared word embedding layer """
+        with tf.name_scope("word_embeddings"):
+            # Create and initialize weights. The random normal initializer was chosen
+            # arbitrarily, and works well.
+            self.word_embeddings = self.add_weight(
+                "weight",
+                shape=[self.config.vocab_size, self.config.embedding_size],
+                initializer=get_initializer(self.config.initializer_range))
+        super(TFAlbertEmbeddings, self).build(input_shape)
+
+    def call(self, inputs, mode="embedding", training=False):
+        """Get token embeddings of inputs.
+        Args:
+            inputs: list of three int64 tensors with shape [batch_size, length]: (input_ids, position_ids, token_type_ids)
+            mode: string, a valid value is one of "embedding" and "linear".
+        Returns:
+            outputs: (1) If mode == "embedding", output embedding tensor, float32 with
+                shape [batch_size, length, embedding_size]; (2) mode == "linear", output
+                linear tensor, float32 with shape [batch_size, length, vocab_size].
+        Raises:
+            ValueError: if mode is not valid.
+
+        Shared weights logic adapted from
+            https://github.com/tensorflow/models/blob/a009f4fb9d2fc4949e32192a944688925ef78659/official/transformer/v2/embedding_layer.py#L24
+        """
+        if mode == "embedding":
+            return self._embedding(inputs, training=training)
+        elif mode == "linear":
+            return self._linear(inputs)
+        else:
+            raise ValueError("mode {} is not valid.".format(mode))
+
+    def _embedding(self, inputs, training=False):
+        """Applies embedding based on inputs tensor."""
+        input_ids, position_ids, token_type_ids = inputs
+
+        seq_length = tf.shape(input_ids)[1]
+        if position_ids is None:
+            position_ids = tf.range(seq_length, dtype=tf.int32)[tf.newaxis, :]
+        if token_type_ids is None:
+            token_type_ids = tf.fill(tf.shape(input_ids), 0)
+
+        words_embeddings = tf.gather(self.word_embeddings, input_ids)
+        position_embeddings = self.position_embeddings(position_ids)
+        token_type_embeddings = self.token_type_embeddings(token_type_ids)
+
+        embeddings = words_embeddings + position_embeddings + token_type_embeddings
+        embeddings = self.LayerNorm(embeddings)
+        embeddings = self.dropout(embeddings, training=training)
+        return embeddings
+
+    def _linear(self, inputs):
+        """Computes logits by running inputs through a linear layer.
+            Args:
+                inputs: A float32 tensor with shape [batch_size, length, embedding_size]
+            Returns:
+                float32 tensor with shape [batch_size, length, vocab_size].
+        """
+        batch_size = tf.shape(inputs)[0]
+        length = tf.shape(inputs)[1]
+
+        print(inputs.shape)
+
+        x = tf.reshape(inputs, [-1, self.config.embedding_size])
+
+        print(x.shape, self.word_embeddings)
+
+        logits = tf.matmul(x, self.word_embeddings, transpose_b=True)
+
+        print([batch_size, length, self.config.vocab_size])
+        return tf.reshape(logits, [batch_size, length, self.config.vocab_size])
+
+
+class TFAlbertSelfAttention(tf.keras.layers.Layer):
+    def __init__(self, config, **kwargs):
+        super(TFAlbertSelfAttention, self).__init__(**kwargs)
+        if config.hidden_size % config.num_attention_heads != 0:
+            raise ValueError(
+                "The hidden size (%d) is not a multiple of the number of attention "
+                "heads (%d)" % (config.hidden_size, config.num_attention_heads))
+        self.output_attentions = config.output_attentions
+
+        self.num_attention_heads = config.num_attention_heads
+        assert config.hidden_size % config.num_attention_heads == 0
+        self.attention_head_size = int(
+            config.hidden_size / config.num_attention_heads)
+        self.all_head_size = self.num_attention_heads * self.attention_head_size
+
+        self.query = tf.keras.layers.Dense(self.all_head_size,
+                                           kernel_initializer=get_initializer(
+                                               config.initializer_range),
+                                           name='query')
+        self.key = tf.keras.layers.Dense(self.all_head_size,
+                                         kernel_initializer=get_initializer(
+                                             config.initializer_range),
+                                         name='key')
+        self.value = tf.keras.layers.Dense(self.all_head_size,
+                                           kernel_initializer=get_initializer(
+                                               config.initializer_range),
+                                           name='value')
+
+        self.dropout = tf.keras.layers.Dropout(
+            config.attention_probs_dropout_prob)
+
+    def transpose_for_scores(self, x, batch_size):
+        x = tf.reshape(
+            x, (batch_size, -1, self.num_attention_heads, self.attention_head_size))
+        return tf.transpose(x, perm=[0, 2, 1, 3])
+
+    def call(self, inputs, training=False):
+        hidden_states, attention_mask, head_mask = inputs
+
+        batch_size = tf.shape(hidden_states)[0]
+        mixed_query_layer = self.query(hidden_states)
+        mixed_key_layer = self.key(hidden_states)
+        mixed_value_layer = self.value(hidden_states)
+
+        query_layer = self.transpose_for_scores(mixed_query_layer, batch_size)
+        key_layer = self.transpose_for_scores(mixed_key_layer, batch_size)
+        value_layer = self.transpose_for_scores(mixed_value_layer, batch_size)
+
+        # Take the dot product between "query" and "key" to get the raw attention scores.
+        # (batch size, num_heads, seq_len_q, seq_len_k)
+        attention_scores = tf.matmul(query_layer, key_layer, transpose_b=True)
+        # scale attention_scores
+        dk = tf.cast(tf.shape(key_layer)[-1], tf.float32)
+        attention_scores = attention_scores / tf.math.sqrt(dk)
+
+        if attention_mask is not None:
+            # Apply the attention mask is (precomputed for all layers in TFAlbertModel call() function)
+            attention_scores = attention_scores + attention_mask
+
+        # Normalize the attention scores to probabilities.
+        attention_probs = tf.nn.softmax(attention_scores, axis=-1)
+
+        # This is actually dropping out entire tokens to attend to, which might
+        # seem a bit unusual, but is taken from the original Transformer paper.
+        attention_probs = self.dropout(attention_probs, training=training)
+
+        # Mask heads if we want to
+        if head_mask is not None:
+            attention_probs = attention_probs * head_mask
+
+        context_layer = tf.matmul(attention_probs, value_layer)
+
+        context_layer = tf.transpose(context_layer, perm=[0, 2, 1, 3])
+        context_layer = tf.reshape(context_layer,
+                                   (batch_size, -1, self.all_head_size))  # (batch_size, seq_len_q, all_head_size)
+
+        outputs = (context_layer, attention_probs) if self.output_attentions else (
+            context_layer,)
+        return outputs
+
+
+class TFAlbertSelfOutput(tf.keras.layers.Layer):
+    def __init__(self, config, **kwargs):
+        super(TFAlbertSelfOutput, self).__init__(**kwargs)
+        self.dense = tf.keras.layers.Dense(config.hidden_size,
+                                           kernel_initializer=get_initializer(
+                                               config.initializer_range),
+                                           name='dense')
+        self.LayerNorm = tf.keras.layers.LayerNormalization(
+            epsilon=config.layer_norm_eps, name='LayerNorm')
+        self.dropout = tf.keras.layers.Dropout(config.hidden_dropout_prob)
+
+    def call(self, inputs, training=False):
+        hidden_states, input_tensor = inputs
+
+        hidden_states = self.dense(hidden_states)
+        hidden_states = self.dropout(hidden_states, training=training)
+        hidden_states = self.LayerNorm(hidden_states + input_tensor)
+        return hidden_states
+
+
+class TFAlbertAttention(TFBertSelfAttention):
+    def __init__(self, config, **kwargs):
+        super(TFAlbertAttention, self).__init__(config, **kwargs)
+
+        self.hidden_size = config.hidden_size
+        self.dense = tf.keras.layers.Dense(config.hidden_size,
+                                           kernel_initializer=get_initializer(
+                                               config.initializer_range),
+                                           name='dense')
+        self.LayerNorm = tf.keras.layers.LayerNormalization(
+            epsilon=config.layer_norm_eps, name='LayerNorm')
+        self.pruned_heads = set()
+
+    def prune_heads(self, heads):
+        raise NotImplementedError
+
+    def call(self, inputs, training=False):
+        input_tensor, attention_mask, head_mask = inputs
+
+        batch_size = tf.shape(input_tensor)[0]
+        mixed_query_layer = self.query(input_tensor)
+        mixed_key_layer = self.key(input_tensor)
+        mixed_value_layer = self.value(input_tensor)
+
+        query_layer = self.transpose_for_scores(mixed_query_layer, batch_size)
+        key_layer = self.transpose_for_scores(mixed_key_layer, batch_size)
+        value_layer = self.transpose_for_scores(mixed_value_layer, batch_size)
+
+        # Take the dot product between "query" and "key" to get the raw attention scores.
+        # (batch size, num_heads, seq_len_q, seq_len_k)
+        attention_scores = tf.matmul(query_layer, key_layer, transpose_b=True)
+        # scale attention_scores
+        dk = tf.cast(tf.shape(key_layer)[-1], tf.float32)
+        attention_scores = attention_scores / tf.math.sqrt(dk)
+
+        if attention_mask is not None:
+            # Apply the attention mask is (precomputed for all layers in TFBertModel call() function)
+            attention_scores = attention_scores + attention_mask
+
+        # Normalize the attention scores to probabilities.
+        attention_probs = tf.nn.softmax(attention_scores, axis=-1)
+
+        # This is actually dropping out entire tokens to attend to, which might
+        # seem a bit unusual, but is taken from the original Transformer paper.
+        attention_probs = self.dropout(attention_probs, training=training)
+
+        # Mask heads if we want to
+        if head_mask is not None:
+            attention_probs = attention_probs * head_mask
+
+        context_layer = tf.matmul(attention_probs, value_layer)
+
+        context_layer = tf.transpose(context_layer, perm=[0, 2, 1, 3])
+        context_layer = tf.reshape(context_layer,
+                                   (batch_size, -1, self.all_head_size))  # (batch_size, seq_len_q, all_head_size)
+
+        self_outputs = (context_layer, attention_probs) if self.output_attentions else (
+            context_layer,)
+
+        hidden_states = self_outputs[0]
+
+        hidden_states = self.dense(hidden_states)
+        hidden_states = self.dropout(hidden_states, training=training)
+        attention_output = self.LayerNorm(hidden_states + input_tensor)
+
+        # add attentions if we output them
+        outputs = (attention_output,) + self_outputs[1:]
+        return outputs
+
+
+class TFAlbertLayer(tf.keras.layers.Layer):
+    def __init__(self, config, **kwargs):
+        super(TFAlbertLayer, self).__init__(**kwargs)
+        self.attention = TFAlbertAttention(config, name='attention')
+
+        self.ffn = tf.keras.layers.Dense(config.intermediate_size, kernel_initializer=get_initializer(
+            config.initializer_range), name='ffn')
+
+        if isinstance(config.hidden_act, str) or (sys.version_info[0] == 2 and isinstance(config.hidden_act, unicode)):
+            self.activation = ACT2FN[config.hidden_act]
+        else:
+            self.activation = config.hidden_act
+
+        self.ffn_output = tf.keras.layers.Dense(config.hidden_size, kernel_initializer=get_initializer(
+            config.initializer_range), name='ffn_output')
+        self.full_layer_layer_norm = tf.keras.layers.LayerNormalization(
+            epsilon=config.layer_norm_eps, name='full_layer_layer_norm')
+        self.dropout = tf.keras.layers.Dropout(config.hidden_dropout_prob)
+
+    def call(self, inputs, training=False):
+        hidden_states, attention_mask, head_mask = inputs
+
+        attention_outputs = self.attention(
+            [hidden_states, attention_mask, head_mask], training=training)
+        ffn_output = self.ffn(attention_outputs[0])
+        ffn_output = self.activation(ffn_output)
+        ffn_output = self.ffn_output(ffn_output)
+
+        hidden_states = self.dropout(hidden_states, training=training)
+        hidden_states = self.full_layer_layer_norm(
+            ffn_output + attention_outputs[0])
+
+        # add attentions if we output them
+        outputs = (hidden_states,) + attention_outputs[1:]
+        return outputs
+
+
+class TFAlbertLayerGroup(tf.keras.layers.Layer):
+    def __init__(self, config, **kwargs):
+        super(TFAlbertLayerGroup, self).__init__(**kwargs)
+
+        self.output_attentions = config.output_attentions
+        self.output_hidden_states = config.output_hidden_states
+        self.albert_layers = [TFAlbertLayer(config, name="albert_layers_._{}".format(
+            i)) for i in range(config.inner_group_num)]
+
+    def call(self, inputs, training=False):
+        hidden_states, attention_mask, head_mask = inputs
+
+        layer_hidden_states = ()
+        layer_attentions = ()
+
+        for layer_index, albert_layer in enumerate(self.albert_layers):
+            layer_output = albert_layer(
+                [hidden_states, attention_mask, head_mask[layer_index]], training=training)
+            hidden_states = layer_output[0]
+
+            if self.output_attentions:
+                layer_attentions = layer_attentions + (layer_output[1],)
+
+            if self.output_hidden_states:
+                layer_hidden_states = layer_hidden_states + (hidden_states,)
+
+        outputs = (hidden_states,)
+        if self.output_hidden_states:
+            outputs = outputs + (layer_hidden_states,)
+        if self.output_attentions:
+            outputs = outputs + (layer_attentions,)
+        # last-layer hidden state, (layer hidden states), (layer attentions)
+        return outputs
+
+
+class TFAlbertTransformer(tf.keras.layers.Layer):
+    def __init__(self, config, **kwargs):
+        super(TFAlbertTransformer, self).__init__(**kwargs)
+
+        self.config = config
+        self.output_attentions = config.output_attentions
+        self.output_hidden_states = config.output_hidden_states
+        self.embedding_hidden_mapping_in = tf.keras.layers.Dense(config.hidden_size, kernel_initializer=get_initializer(
+            config.initializer_range), name='embedding_hidden_mapping_in')
+        self.albert_layer_groups = [TFAlbertLayerGroup(
+            config, name="albert_layer_groups_._{}".format(i)) for i in range(config.num_hidden_groups)]
+
+    def call(self, inputs, training=False):
+        hidden_states, attention_mask, head_mask = inputs
+
+        hidden_states = self.embedding_hidden_mapping_in(hidden_states)
+        all_attentions = ()
+
+        if self.output_hidden_states:
+            all_hidden_states = (hidden_states,)
+
+        for i in range(self.config.num_hidden_layers):
+            # Number of layers in a hidden group
+            layers_per_group = int(
+                self.config.num_hidden_layers / self.config.num_hidden_groups)
+
+            # Index of the hidden group
+            group_idx = int(
+                i / (self.config.num_hidden_layers / self.config.num_hidden_groups))
+
+            layer_group_output = self.albert_layer_groups[group_idx](
+                [hidden_states, attention_mask, head_mask[group_idx*layers_per_group:(group_idx+1)*layers_per_group]], training=training)
+            hidden_states = layer_group_output[0]
+
+            if self.output_attentions:
+                all_attentions = all_attentions + layer_group_output[-1]
+
+            if self.output_hidden_states:
+                all_hidden_states = all_hidden_states + (hidden_states,)
+
+        outputs = (hidden_states,)
+        if self.output_hidden_states:
+            outputs = outputs + (all_hidden_states,)
+        if self.output_attentions:
+            outputs = outputs + (all_attentions,)
+
+        # last-layer hidden state, (all hidden states), (all attentions)
+        return outputs
+
+
+class TFAlbertPreTrainedModel(TFPreTrainedModel):
+    """ An abstract class to handle weights initialization and
+        a simple interface for dowloading and loading pretrained models.
+    """
+    config_class = AlbertConfig
+    pretrained_model_archive_map = TF_ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP
+    base_model_prefix = "albert"
+
+
+class TFAlbertMLMHead(tf.keras.layers.Layer):
+    def __init__(self, config, input_embeddings, **kwargs):
+        super(TFAlbertMLMHead, self).__init__(**kwargs)
+        self.vocab_size = config.vocab_size
+
+        self.dense = tf.keras.layers.Dense(config.embedding_size,
+                                           kernel_initializer=get_initializer(
+                                               config.initializer_range),
+                                           name='dense')
+        if isinstance(config.hidden_act, str) or (sys.version_info[0] == 2 and isinstance(config.hidden_act, unicode)):
+            self.activation = ACT2FN[config.hidden_act]
+        else:
+            self.activation = config.hidden_act
+
+        self.LayerNorm = tf.keras.layers.LayerNormalization(
+            epsilon=config.layer_norm_eps, name='LayerNorm')
+
+        # The output weights are the same as the input embeddings, but there is
+        # an output-only bias for each token.
+        self.input_embeddings = input_embeddings
+
+    def build(self, input_shape):
+        self.bias = self.add_weight(shape=(self.vocab_size,),
+                                    initializer='zeros',
+                                    trainable=True,
+                                    name='bias')
+        super(TFAlbertMLMHead, self).build(input_shape)
+
+    def call(self, hidden_states):
+        hidden_states = self.dense(hidden_states)
+        hidden_states = self.activation(hidden_states)
+        hidden_states = self.LayerNorm(hidden_states)
+        hidden_states = self.input_embeddings(hidden_states, mode="linear")
+        hidden_states = hidden_states + self.bias
+        return hidden_states
+
+
+ALBERT_START_DOCSTRING = r"""    The ALBERT model was proposed in
+    `ALBERT: Pre-training of Deep Bidirectional Transformers for Language Understanding`_
+    by Jacob Devlin, Ming-Wei Chang, Kenton Lee and Kristina Toutanova. It's a bidirectional transformer
+    pre-trained using a combination of masked language modeling objective and next sentence prediction
+    on a large corpus comprising the Toronto Book Corpus and Wikipedia.
+
+    This model is a tf.keras.Model `tf.keras.Model`_ sub-class. Use it as a regular TF 2.0 Keras Model and
+    refer to the TF 2.0 documentation for all matter related to general usage and behavior.
+
+    .. _`ALBERT: Pre-training of Deep Bidirectional Transformers for Language Understanding`:
+        https://arxiv.org/abs/1810.04805
+
+    .. _`tf.keras.Model`:
+        https://www.tensorflow.org/versions/r2.0/api_docs/python/tf/keras/Model
+
+    Note on the model inputs:
+        TF 2.0 models accepts two formats as inputs:
+
+            - having all inputs as keyword arguments (like PyTorch models), or
+            - having all inputs as a list, tuple or dict in the first positional arguments.
+
+        This second option is usefull when using `tf.keras.Model.fit()` method which currently requires having all the tensors in the first argument of the model call function: `model(inputs)`.
+
+        If you choose this second option, there are three possibilities you can use to gather all the input Tensors in the first positional argument :
+
+        - a single Tensor with input_ids only and nothing else: `model(inputs_ids)
+        - a list of varying length with one or several input Tensors IN THE ORDER given in the docstring:
+            `model([input_ids, attention_mask])` or `model([input_ids, attention_mask, token_type_ids])`
+        - a dictionary with one or several input Tensors associaed to the input names given in the docstring:
+            `model({'input_ids': input_ids, 'token_type_ids': token_type_ids})`
+
+    Parameters:
+        config (:class:`~transformers.AlbertConfig`): Model configuration class with all the parameters of the model. 
+            Initializing with a config file does not load the weights associated with the model, only the configuration.
+            Check out the :meth:`~transformers.PreTrainedModel.from_pretrained` method to load the model weights.
+"""
+
+ALBERT_INPUTS_DOCSTRING = r"""
+    Inputs:
+        **input_ids**: ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, sequence_length)``:
+            Indices of input sequence tokens in the vocabulary.
+            To match pre-training, ALBERT input sequence should be formatted with [CLS] and [SEP] tokens as follows:
+
+            (a) For sequence pairs:
+
+                ``tokens:         [CLS] is this jack ##son ##ville ? [SEP] no it is not . [SEP]``
+                
+                ``token_type_ids:   0   0  0    0    0     0       0   0   1  1  1  1   1   1``
+
+            (b) For single sequences:
+
+                ``tokens:         [CLS] the dog is hairy . [SEP]``
+                
+                ``token_type_ids:   0   0   0   0  0     0   0``
+
+            Albert is a model with absolute position embeddings so it's usually advised to pad the inputs on
+            the right rather than the left.
+
+            Indices can be obtained using :class:`transformers.AlbertTokenizer`.
+            See :func:`transformers.PreTrainedTokenizer.encode` and
+            :func:`transformers.PreTrainedTokenizer.convert_tokens_to_ids` for details.
+        **attention_mask**: (`optional`) ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, sequence_length)``:
+            Mask to avoid performing attention on padding token indices.
+            Mask values selected in ``[0, 1]``:
+            ``1`` for tokens that are NOT MASKED, ``0`` for MASKED tokens.
+        **token_type_ids**: (`optional`) ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, sequence_length)``:
+            Segment token indices to indicate first and second portions of the inputs.
+            Indices are selected in ``[0, 1]``: ``0`` corresponds to a `sentence A` token, ``1``
+            corresponds to a `sentence B` token
+            (see `ALBERT: Pre-training of Deep Bidirectional Transformers for Language Understanding`_ for more details).
+        **position_ids**: (`optional`) ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, sequence_length)``:
+            Indices of positions of each input sequence tokens in the position embeddings.
+            Selected in the range ``[0, config.max_position_embeddings - 1]``.
+        **head_mask**: (`optional`) ``Numpy array`` or ``tf.Tensor`` of shape ``(num_heads,)`` or ``(num_layers, num_heads)``:
+            Mask to nullify selected heads of the self-attention modules.
+            Mask values selected in ``[0, 1]``:
+            ``1`` indicates the head is **not masked**, ``0`` indicates the head is **masked**.
+"""
+
+@add_start_docstrings("The bare Albert Model transformer outputing raw hidden-states without any specific head on top.",
+                      ALBERT_START_DOCSTRING, ALBERT_INPUTS_DOCSTRING)
+class TFAlbertModel(TFAlbertPreTrainedModel):
+    r"""
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **last_hidden_state**: ``tf.Tensor`` of shape ``(batch_size, sequence_length, hidden_size)``
+            Sequence of hidden-states at the output of the last layer of the model.
+        **pooler_output**: ``tf.Tensor`` of shape ``(batch_size, hidden_size)``
+            Last layer hidden-state of the first token of the sequence (classification token)
+            further processed by a Linear layer and a Tanh activation function. The Linear
+            layer weights are trained from the next sentence prediction (classification)
+            objective during Albert pretraining. This output is usually *not* a good summary
+            of the semantic content of the input, you're often better with averaging or pooling
+            the sequence of hidden-states for the whole input sequence.
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``tf.Tensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``tf.Tensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        import tensorflow as tf
+        from transformers import AlbertTokenizer, TFAlbertModel
+
+        tokenizer = AlbertTokenizer.from_pretrained('bert-base-uncased')
+        model = TFAlbertModel.from_pretrained('bert-base-uncased')
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        outputs = model(input_ids)
+        last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
+
+    """
+
+    def __init__(self, config, **kwargs):
+        super(TFAlbertModel, self).__init__(config, **kwargs)
+        self.num_hidden_layers = config.num_hidden_layers
+
+        self.embeddings = TFAlbertEmbeddings(config, name="embeddings")
+        self.encoder = TFAlbertTransformer(config, name="encoder")
+        self.pooler = tf.keras.layers.Dense(config.hidden_size, kernel_initializer=get_initializer(
+            config.initializer_range), activation='tanh', name='pooler')
+
+    def _resize_token_embeddings(self, new_num_tokens):
+        raise NotImplementedError
+
+    def _prune_heads(self, heads_to_prune):
+        """ Prunes heads of the model.
+            heads_to_prune: dict of {layer_num: list of heads to prune in this layer}
+            See base class PreTrainedModel
+        """
+        raise NotImplementedError
+
+    def call(self, inputs, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, training=False):
+        if isinstance(inputs, (tuple, list)):
+            input_ids = inputs[0]
+            attention_mask = inputs[1] if len(inputs) > 1 else attention_mask
+            token_type_ids = inputs[2] if len(inputs) > 2 else token_type_ids
+            position_ids = inputs[3] if len(inputs) > 3 else position_ids
+            head_mask = inputs[4] if len(inputs) > 4 else head_mask
+            assert len(inputs) <= 5, "Too many inputs."
+        elif isinstance(inputs, dict):
+            input_ids = inputs.get('input_ids')
+            attention_mask = inputs.get('attention_mask', attention_mask)
+            token_type_ids = inputs.get('token_type_ids', token_type_ids)
+            position_ids = inputs.get('position_ids', position_ids)
+            head_mask = inputs.get('head_mask', head_mask)
+            assert len(inputs) <= 5, "Too many inputs."
+        else:
+            input_ids = inputs
+
+        if attention_mask is None:
+            attention_mask = tf.fill(tf.shape(input_ids), 1)
+        if token_type_ids is None:
+            token_type_ids = tf.fill(tf.shape(input_ids), 0)
+
+        # We create a 3D attention mask from a 2D tensor mask.
+        # Sizes are [batch_size, 1, 1, to_seq_length]
+        # So we can broadcast to [batch_size, num_heads, from_seq_length, to_seq_length]
+        # this attention mask is more simple than the triangular masking of causal attention
+        # used in OpenAI GPT, we just need to prepare the broadcast dimension here.
+        extended_attention_mask = attention_mask[:, tf.newaxis, tf.newaxis, :]
+
+        # Since attention_mask is 1.0 for positions we want to attend and 0.0 for
+        # masked positions, this operation will create a tensor which is 0.0 for
+        # positions we want to attend and -10000.0 for masked positions.
+        # Since we are adding it to the raw scores before the softmax, this is
+        # effectively the same as removing these entirely.
+
+        extended_attention_mask = tf.cast(extended_attention_mask, tf.float32)
+        extended_attention_mask = (1.0 - extended_attention_mask) * -10000.0
+
+        # Prepare head mask if needed
+        # 1.0 in head_mask indicate we keep the head
+        # attention_probs has shape bsz x n_heads x N x N
+        # input head_mask has shape [num_heads] or [num_hidden_layers x num_heads]
+        # and head_mask is converted to shape [num_hidden_layers x batch x num_heads x seq_length x seq_length]
+        if not head_mask is None:
+            raise NotImplementedError
+        else:
+            head_mask = [None] * self.num_hidden_layers
+            # head_mask = tf.constant([0] * self.num_hidden_layers)
+
+        embedding_output = self.embeddings(
+            [input_ids, position_ids, token_type_ids], training=training)
+        encoder_outputs = self.encoder(
+            [embedding_output, extended_attention_mask, head_mask], training=training)
+
+        sequence_output = encoder_outputs[0]
+        pooled_output = self.pooler(sequence_output)
+
+        # add hidden_states and attentions if they are here
+        outputs = (sequence_output, pooled_output,) + encoder_outputs[1:]
+        # sequence_output, pooled_output, (hidden_states), (attentions)
+        return outputs
+
+
+@add_start_docstrings("""Albert Model with a `language modeling` head on top. """,
+                      ALBERT_START_DOCSTRING, ALBERT_INPUTS_DOCSTRING)
+class TFAlbertForMaskedLM(TFAlbertPreTrainedModel):
+    r"""
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **prediction_scores**: ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, sequence_length, config.vocab_size)``
+            Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``Numpy array`` or ``tf.Tensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``Numpy array`` or ``tf.Tensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        import tensorflow as tf
+        from transformers import AlbertTokenizer, TFAlbertForMaskedLM
+
+        tokenizer = AlbertTokenizer.from_pretrained('bert-base-uncased')
+        model = TFAlbertForMaskedLM.from_pretrained('bert-base-uncased')
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        outputs = model(input_ids)
+        prediction_scores = outputs[0]
+
+    """
+
+    def __init__(self, config, *inputs, **kwargs):
+        super(TFAlbertForMaskedLM, self).__init__(config, *inputs, **kwargs)
+
+        self.albert = TFAlbertModel(config, name='albert')
+        self.predictions = TFAlbertMLMHead(
+            config, self.albert.embeddings, name='predictions')
+
+    def call(self, inputs, **kwargs):
+        outputs = self.albert(inputs, **kwargs)
+
+        sequence_output = outputs[0]
+        prediction_scores = self.predictions(
+            sequence_output, training=kwargs.get('training', False))
+
+        # Add hidden states and attention if they are here
+        outputs = (prediction_scores,) + outputs[2:]
+
+        return outputs  # prediction_scores, (hidden_states), (attentions)

From 7bddbf5961f35d03db5151a4747085f7a57d7ff7 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Thu, 7 Nov 2019 23:50:05 +0000
Subject: [PATCH 114/505] TFAlbertForSequenceClassification

---
 transformers/modeling_tf_albert.py | 66 ++++++++++++++++++++++++++----
 1 file changed, 58 insertions(+), 8 deletions(-)

diff --git a/transformers/modeling_tf_albert.py b/transformers/modeling_tf_albert.py
index 8861b7add8..410d83ff78 100644
--- a/transformers/modeling_tf_albert.py
+++ b/transformers/modeling_tf_albert.py
@@ -479,16 +479,15 @@ class TFAlbertMLMHead(tf.keras.layers.Layer):
 
 
 ALBERT_START_DOCSTRING = r"""    The ALBERT model was proposed in
-    `ALBERT: Pre-training of Deep Bidirectional Transformers for Language Understanding`_
-    by Jacob Devlin, Ming-Wei Chang, Kenton Lee and Kristina Toutanova. It's a bidirectional transformer
-    pre-trained using a combination of masked language modeling objective and next sentence prediction
-    on a large corpus comprising the Toronto Book Corpus and Wikipedia.
+    `ALBERT: A Lite BERT for Self-supervised Learning of Language Representations`_
+    by Zhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut. It presents
+    two parameter-reduction techniques to lower memory consumption and increase the trainig speed of BERT.
 
     This model is a tf.keras.Model `tf.keras.Model`_ sub-class. Use it as a regular TF 2.0 Keras Model and
     refer to the TF 2.0 documentation for all matter related to general usage and behavior.
 
-    .. _`ALBERT: Pre-training of Deep Bidirectional Transformers for Language Understanding`:
-        https://arxiv.org/abs/1810.04805
+    .. _`ALBERT: A Lite BERT for Self-supervised Learning of Language Representations`:
+        https://arxiv.org/abs/1909.11942
 
     .. _`tf.keras.Model`:
         https://www.tensorflow.org/versions/r2.0/api_docs/python/tf/keras/Model
@@ -695,8 +694,8 @@ class TFAlbertForMaskedLM(TFAlbertPreTrainedModel):
         import tensorflow as tf
         from transformers import AlbertTokenizer, TFAlbertForMaskedLM
 
-        tokenizer = AlbertTokenizer.from_pretrained('bert-base-uncased')
-        model = TFAlbertForMaskedLM.from_pretrained('bert-base-uncased')
+        tokenizer = AlbertTokenizer.from_pretrained('albert-base-v2')
+        model = TFAlbertForMaskedLM.from_pretrained('albert-base-v2')
         input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
         outputs = model(input_ids)
         prediction_scores = outputs[0]
@@ -721,3 +720,54 @@ class TFAlbertForMaskedLM(TFAlbertPreTrainedModel):
         outputs = (prediction_scores,) + outputs[2:]
 
         return outputs  # prediction_scores, (hidden_states), (attentions)
+
+
+@add_start_docstrings("""Albert Model transformer with a sequence classification/regression head on top (a linear layer on top of
+    the pooled output) e.g. for GLUE tasks. """,
+    ALBERT_START_DOCSTRING, ALBERT_INPUTS_DOCSTRING)
+class TFAlbertForSequenceClassification(TFAlbertPreTrainedModel):
+    r"""
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **logits**: ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, config.num_labels)``
+            Classification (or regression if config.num_labels==1) scores (before SoftMax).
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``Numpy array`` or ``tf.Tensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``Numpy array`` or ``tf.Tensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        import tensorflow as tf
+        from transformers import AlbertTokenizer, TFAlbertForSequenceClassification
+
+        tokenizer = AlbertTokenizer.from_pretrained('albert-base-v2')
+        model = TFAlbertForSequenceClassification.from_pretrained('albert-base-v2')
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        outputs = model(input_ids)
+        logits = outputs[0]
+
+    """
+    def __init__(self, config, *inputs, **kwargs):
+        super(TFAlbertForSequenceClassification, self).__init__(config, *inputs, **kwargs)
+        self.num_labels = config.num_labels
+
+        self.albert = TFAlbertModel(config, name='albert')
+        self.dropout = tf.keras.layers.Dropout(config.hidden_dropout_prob)
+        self.classifier = tf.keras.layers.Dense(config.num_labels,
+                                                kernel_initializer=get_initializer(config.initializer_range),
+                                                name='classifier')
+
+    def call(self, inputs, **kwargs):
+        outputs = self.albert(inputs, **kwargs)
+
+        pooled_output = outputs[1]
+
+        pooled_output = self.dropout(pooled_output, training=kwargs.get('training', False))
+        logits = self.classifier(pooled_output)
+
+        outputs = (logits,) + outputs[2:]  # add hidden states and attention if they are here
+
+        return outputs  # logits, (hidden_states), (attentions)
\ No newline at end of file

From b18509c2085dce16e655dd86b7ea0129335da4e1 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Fri, 8 Nov 2019 00:12:21 +0000
Subject: [PATCH 115/505] Tests for ALBERT in TF2 + fixes

---
 transformers/__init__.py                      |   1 +
 transformers/modeling_tf_albert.py            |  18 +-
 transformers/tests/modeling_tf_albert_test.py | 229 ++++++++++++++++++
 3 files changed, 237 insertions(+), 11 deletions(-)
 create mode 100644 transformers/tests/modeling_tf_albert_test.py

diff --git a/transformers/__init__.py b/transformers/__init__.py
index a409ef772e..baf430c17b 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -169,6 +169,7 @@ if is_tf_available():
                                     TF_CTRL_PRETRAINED_MODEL_ARCHIVE_MAP)
 
     from .modeling_tf_albert import (TFAlbertPreTrainedModel, TFAlbertModel, TFAlbertForMaskedLM,
+                                     TFAlbertForSequenceClassification,
                                     TF_ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
 
 # TF 2.0 <=> PyTorch conversion utilities
diff --git a/transformers/modeling_tf_albert.py b/transformers/modeling_tf_albert.py
index 410d83ff78..a3f183b192 100644
--- a/transformers/modeling_tf_albert.py
+++ b/transformers/modeling_tf_albert.py
@@ -126,16 +126,8 @@ class TFAlbertEmbeddings(tf.keras.layers.Layer):
         """
         batch_size = tf.shape(inputs)[0]
         length = tf.shape(inputs)[1]
-
-        print(inputs.shape)
-
         x = tf.reshape(inputs, [-1, self.config.embedding_size])
-
-        print(x.shape, self.word_embeddings)
-
         logits = tf.matmul(x, self.word_embeddings, transpose_b=True)
-
-        print([batch_size, length, self.config.vocab_size])
         return tf.reshape(logits, [batch_size, length, self.config.vocab_size])
 
 
@@ -460,20 +452,24 @@ class TFAlbertMLMHead(tf.keras.layers.Layer):
 
         # The output weights are the same as the input embeddings, but there is
         # an output-only bias for each token.
-        self.input_embeddings = input_embeddings
+        self.decoder = input_embeddings
 
     def build(self, input_shape):
         self.bias = self.add_weight(shape=(self.vocab_size,),
                                     initializer='zeros',
                                     trainable=True,
                                     name='bias')
+        self.decoder_bias = self.add_weight(shape=(self.vocab_size,),
+                                    initializer='zeros',
+                                    trainable=True,
+                                    name='decoder/bias')
         super(TFAlbertMLMHead, self).build(input_shape)
 
     def call(self, hidden_states):
         hidden_states = self.dense(hidden_states)
         hidden_states = self.activation(hidden_states)
         hidden_states = self.LayerNorm(hidden_states)
-        hidden_states = self.input_embeddings(hidden_states, mode="linear")
+        hidden_states = self.decoder(hidden_states, mode="linear") + self.decoder_bias
         hidden_states = hidden_states + self.bias
         return hidden_states
 
@@ -666,7 +662,7 @@ class TFAlbertModel(TFAlbertPreTrainedModel):
             [embedding_output, extended_attention_mask, head_mask], training=training)
 
         sequence_output = encoder_outputs[0]
-        pooled_output = self.pooler(sequence_output)
+        pooled_output = self.pooler(sequence_output[:, 0])
 
         # add hidden_states and attentions if they are here
         outputs = (sequence_output, pooled_output,) + encoder_outputs[1:]
diff --git a/transformers/tests/modeling_tf_albert_test.py b/transformers/tests/modeling_tf_albert_test.py
new file mode 100644
index 0000000000..85fc62f34f
--- /dev/null
+++ b/transformers/tests/modeling_tf_albert_test.py
@@ -0,0 +1,229 @@
+# coding=utf-8
+# Copyright 2018 The Google AI Language Team Authors.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from __future__ import absolute_import
+from __future__ import division
+from __future__ import print_function
+
+import unittest
+import shutil
+import pytest
+import sys
+
+from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
+from .configuration_common_test import ConfigTester
+
+from transformers import AlbertConfig, is_tf_available
+
+if is_tf_available():
+    import tensorflow as tf
+    from transformers.modeling_tf_albert import (TFAlbertModel, TFAlbertForMaskedLM,
+                                                 TFAlbertForSequenceClassification,
+                                                 TF_ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
+else:
+    pytestmark = pytest.mark.skip("Require TensorFlow")
+
+
+class TFAlbertModelTest(TFCommonTestCases.TFCommonModelTester):
+
+    all_model_classes = (
+        TFAlbertModel,
+        TFAlbertForMaskedLM,
+        TFAlbertForSequenceClassification
+    ) if is_tf_available() else ()
+
+    class TFAlbertModelTester(object):
+
+        def __init__(self,
+                     parent,
+                     batch_size=13,
+                     seq_length=7,
+                     is_training=True,
+                     use_input_mask=True,
+                     use_token_type_ids=True,
+                     use_labels=True,
+                     vocab_size=99,
+                     hidden_size=32,
+                     num_hidden_layers=5,
+                     num_attention_heads=4,
+                     intermediate_size=37,
+                     hidden_act="gelu",
+                     hidden_dropout_prob=0.1,
+                     attention_probs_dropout_prob=0.1,
+                     max_position_embeddings=512,
+                     type_vocab_size=16,
+                     type_sequence_label_size=2,
+                     initializer_range=0.02,
+                     num_labels=3,
+                     num_choices=4,
+                     scope=None,
+                     ):
+            self.parent = parent
+            self.batch_size = batch_size
+            self.seq_length = seq_length
+            self.is_training = is_training
+            self.use_input_mask = use_input_mask
+            self.use_token_type_ids = use_token_type_ids
+            self.use_labels = use_labels
+            self.vocab_size = vocab_size
+            self.hidden_size = hidden_size
+            self.num_hidden_layers = num_hidden_layers
+            self.num_attention_heads = num_attention_heads
+            self.intermediate_size = intermediate_size
+            self.hidden_act = hidden_act
+            self.hidden_dropout_prob = hidden_dropout_prob
+            self.attention_probs_dropout_prob = attention_probs_dropout_prob
+            self.max_position_embeddings = max_position_embeddings
+            self.type_vocab_size = type_vocab_size
+            self.type_sequence_label_size = type_sequence_label_size
+            self.initializer_range = initializer_range
+            self.num_labels = num_labels
+            self.num_choices = num_choices
+            self.scope = scope
+
+        def prepare_config_and_inputs(self):
+            input_ids = ids_tensor(
+                [self.batch_size, self.seq_length], self.vocab_size)
+
+            input_mask = None
+            if self.use_input_mask:
+                input_mask = ids_tensor(
+                    [self.batch_size, self.seq_length], vocab_size=2)
+
+            token_type_ids = None
+            if self.use_token_type_ids:
+                token_type_ids = ids_tensor(
+                    [self.batch_size, self.seq_length], self.type_vocab_size)
+
+            sequence_labels = None
+            token_labels = None
+            choice_labels = None
+            if self.use_labels:
+                sequence_labels = ids_tensor(
+                    [self.batch_size], self.type_sequence_label_size)
+                token_labels = ids_tensor(
+                    [self.batch_size, self.seq_length], self.num_labels)
+                choice_labels = ids_tensor([self.batch_size], self.num_choices)
+
+            config = AlbertConfig(
+                vocab_size_or_config_json_file=self.vocab_size,
+                hidden_size=self.hidden_size,
+                num_hidden_layers=self.num_hidden_layers,
+                num_attention_heads=self.num_attention_heads,
+                intermediate_size=self.intermediate_size,
+                hidden_act=self.hidden_act,
+                hidden_dropout_prob=self.hidden_dropout_prob,
+                attention_probs_dropout_prob=self.attention_probs_dropout_prob,
+                max_position_embeddings=self.max_position_embeddings,
+                type_vocab_size=self.type_vocab_size,
+                initializer_range=self.initializer_range)
+
+            return config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
+
+        def create_and_check_albert_model(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
+            model = TFAlbertModel(config=config)
+            # inputs = {'input_ids': input_ids,
+            #           'attention_mask': input_mask,
+            #           'token_type_ids': token_type_ids}
+            # sequence_output, pooled_output = model(**inputs)
+            inputs = {'input_ids': input_ids,
+                      'attention_mask': input_mask,
+                      'token_type_ids': token_type_ids}
+            sequence_output, pooled_output = model(inputs)
+
+            inputs = [input_ids, input_mask]
+            sequence_output, pooled_output = model(inputs)
+
+            sequence_output, pooled_output = model(input_ids)
+
+            result = {
+                "sequence_output": sequence_output.numpy(),
+                "pooled_output": pooled_output.numpy(),
+            }
+            self.parent.assertListEqual(
+                list(result["sequence_output"].shape),
+                [self.batch_size, self.seq_length, self.hidden_size])
+            self.parent.assertListEqual(list(result["pooled_output"].shape), [
+                                        self.batch_size, self.hidden_size])
+
+        def create_and_check_albert_for_masked_lm(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
+            model = TFAlbertForMaskedLM(config=config)
+            inputs = {'input_ids': input_ids,
+                      'attention_mask': input_mask,
+                      'token_type_ids': token_type_ids}
+            prediction_scores, = model(inputs)
+            result = {
+                "prediction_scores": prediction_scores.numpy(),
+            }
+            self.parent.assertListEqual(
+                list(result["prediction_scores"].shape),
+                [self.batch_size, self.seq_length, self.vocab_size])
+
+        def create_and_check_albert_for_sequence_classification(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
+            config.num_labels = self.num_labels
+            model = TFAlbertForSequenceClassification(config=config)
+            inputs = {'input_ids': input_ids,
+                      'attention_mask': input_mask,
+                      'token_type_ids': token_type_ids}
+            logits, = model(inputs)
+            result = {
+                "logits": logits.numpy(),
+            }
+            self.parent.assertListEqual(
+                list(result["logits"].shape),
+                [self.batch_size, self.num_labels])
+
+        def prepare_config_and_inputs_for_common(self):
+            config_and_inputs = self.prepare_config_and_inputs()
+            (config, input_ids, token_type_ids, input_mask,
+             sequence_labels, token_labels, choice_labels) = config_and_inputs
+            inputs_dict = {'input_ids': input_ids,
+                           'token_type_ids': token_type_ids, 'attention_mask': input_mask}
+            return config, inputs_dict
+
+    def setUp(self):
+        self.model_tester = TFAlbertModelTest.TFAlbertModelTester(self)
+        self.config_tester = ConfigTester(
+            self, config_class=AlbertConfig, hidden_size=37)
+
+    def test_config(self):
+        self.config_tester.run_common_tests()
+
+    def test_albert_model(self):
+        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+        self.model_tester.create_and_check_albert_model(*config_and_inputs)
+
+    def test_for_masked_lm(self):
+        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+        self.model_tester.create_and_check_albert_for_masked_lm(
+            *config_and_inputs)
+
+    def test_for_sequence_classification(self):
+        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+        self.model_tester.create_and_check_albert_for_sequence_classification(
+            *config_and_inputs)
+
+    @pytest.mark.slow
+    def test_model_from_pretrained(self):
+        cache_dir = "/tmp/transformers_test/"
+        # for model_name in list(TF_ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
+        for model_name in ['albert-base-uncased']:
+            model = TFAlbertModel.from_pretrained(
+                model_name, cache_dir=cache_dir)
+            shutil.rmtree(cache_dir)
+            self.assertIsNotNone(model)
+
+
+if __name__ == "__main__":
+    unittest.main()

From c9cb7f8a0fbe784665b00bfdca6bfc54ad10d5f5 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Mon, 11 Nov 2019 15:12:54 -0500
Subject: [PATCH 116/505] Torch 1.1.0 compatibility + FP16 O1 + TF checkpoints

Co-authored-by: wassname
---
 transformers/modeling_albert.py    | 4 ++--
 transformers/modeling_tf_albert.py | 9 ++++++++-
 2 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index 640af537d0..ff20ca78dc 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -203,8 +203,8 @@ class AlbertAttention(BertSelfAttention):
         
 
         # Should find a better way to do this
-        w = self.dense.weight.T.view(self.num_attention_heads, self.attention_head_size, self.hidden_size)
-        b = self.dense.bias
+        w = self.dense.weight.t().view(self.num_attention_heads, self.attention_head_size, self.hidden_size).to(context_layer.dtype)
+        b = self.dense.bias.to(context_layer.dtype)
 
         projected_context_layer = torch.einsum("bfnd,ndh->bfh", context_layer, w) + b
         projected_context_layer_dropout = self.dropout(projected_context_layer)
diff --git a/transformers/modeling_tf_albert.py b/transformers/modeling_tf_albert.py
index a3f183b192..ee8712eb28 100644
--- a/transformers/modeling_tf_albert.py
+++ b/transformers/modeling_tf_albert.py
@@ -36,7 +36,14 @@ import logging
 logger = logging.getLogger(__name__)
 
 TF_ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
-    # TODO FILL THAT UP
+    'albert-base-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-base-tf_model.h5",
+    'albert-large-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-large-tf_model.h5",
+    'albert-xlarge-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xlarge-tf_model.h5",
+    'albert-xxlarge-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xxlarge-tf_model.h5",
+    'albert-base-v2': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-base-v2-tf_model.h5",
+    'albert-large-v2': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-large-v2-tf_model.h5",
+    'albert-xlarge-v2': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xlarge-v2-tf_model.h5",
+    'albert-xxlarge-v2': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xxlarge-v2-tf_model.h5",
 }
 
 

From f873b55e433e1c9ef364e597ad1d4662db4bc038 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Tue, 26 Nov 2019 10:28:41 -0500
Subject: [PATCH 117/505] Warning for ALBERT-v2 models

---
 transformers/modeling_utils.py | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/transformers/modeling_utils.py b/transformers/modeling_utils.py
index d51eefab58..5d51caa951 100644
--- a/transformers/modeling_utils.py
+++ b/transformers/modeling_utils.py
@@ -315,6 +315,10 @@ class PreTrainedModel(nn.Module):
             model = BertModel.from_pretrained('./tf_model/my_tf_checkpoint.ckpt.index', from_tf=True, config=config)
 
         """
+        if "albert" in pretrained_model_name_or_path and "v2" in pretrained_model_name_or_path:
+            logger.warning("There is currently an upstream reproducibility issue with ALBERT v2 models. Please see " +
+                           "https://github.com/google-research/google-research/issues/119 for more information.")
+
         config = kwargs.pop('config', None)
         state_dict = kwargs.pop('state_dict', None)
         cache_dir = kwargs.pop('cache_dir', None)

From c536c2a4809a51356fd50ff62a80955fea79f790 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Tue, 26 Nov 2019 11:22:52 -0500
Subject: [PATCH 118/505] ALBERT Input Embeds

---
 transformers/modeling_albert.py    | 77 ++++++++++++++++++++++--------
 transformers/modeling_tf_albert.py | 45 ++++++++++++-----
 2 files changed, 90 insertions(+), 32 deletions(-)

diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index ff20ca78dc..7882356d24 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -433,6 +433,12 @@ class AlbertModel(AlbertPreTrainedModel):
 
         self.init_weights()
 
+    def get_input_embeddings(self):
+        return self.embeddings.word_embeddings
+
+    def set_input_embeddings(self, value):
+        self.embeddings.word_embeddings = value
+
     def _resize_token_embeddings(self, new_num_tokens):
         old_embeddings = self.embeddings.word_embeddings
         new_embeddings = self._get_resized_embeddings(old_embeddings, new_num_tokens)
@@ -457,12 +463,24 @@ class AlbertModel(AlbertPreTrainedModel):
             inner_group_idx = int(layer - group_idx * self.config.inner_group_num)
             self.encoder.albert_layer_groups[group_idx].albert_layers[inner_group_idx].attention.prune_heads(heads)
 
+    def forward(self, input_ids=None, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None,
+                inputs_embeds=None):
+
+        if input_ids is not None and inputs_embeds is not None:
+            raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
+        elif input_ids is not None:
+            input_shape = input_ids.size()
+        elif inputs_embeds is not None:
+            input_shape = inputs_embeds.size()[:-1]
+        else:
+            raise ValueError("You have to specify either input_ids or inputs_embeds")
+
+        device = input_ids.device if input_ids is not None else inputs_embeds.device
 
-    def forward(self, input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None):
         if attention_mask is None:
-            attention_mask = torch.ones_like(input_ids)
+            attention_mask = torch.ones(input_shape, device=device)
         if token_type_ids is None:
-            token_type_ids = torch.zeros_like(input_ids)
+            token_type_ids = torch.zeros(input_shape, dtype=torch.long, device=device)
 
         extended_attention_mask = attention_mask.unsqueeze(1).unsqueeze(2)
         extended_attention_mask = extended_attention_mask.to(dtype=next(self.parameters()).dtype) # fp16 compatibility
@@ -477,7 +495,8 @@ class AlbertModel(AlbertPreTrainedModel):
         else:
             head_mask = [None] * self.config.num_hidden_layers
 
-        embedding_output = self.embeddings(input_ids, position_ids=position_ids, token_type_ids=token_type_ids)
+        embedding_output = self.embeddings(input_ids, position_ids=position_ids, token_type_ids=token_type_ids,
+                                           inputs_embeds=inputs_embeds)
         encoder_outputs = self.encoder(embedding_output,
                                        extended_attention_mask,
                                        head_mask=head_mask)
@@ -549,9 +568,19 @@ class AlbertForMaskedLM(AlbertPreTrainedModel):
         self._tie_or_clone_weights(self.predictions.decoder,
                                    self.albert.embeddings.word_embeddings)
 
-    def forward(self, input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None,
-                masked_lm_labels=None):
-        outputs = self.albert(input_ids, attention_mask, token_type_ids, position_ids, head_mask)
+    def get_output_embeddings(self):
+        return self.predictions.decoder
+
+    def forward(self, input_ids=None, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None,
+                masked_lm_labels=None, inputs_embeds=None):
+        outputs = self.albert(
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            head_mask=head_mask,
+            inputs_embeds=inputs_embeds
+        )
         sequence_outputs = outputs[0]
 
         prediction_scores = self.predictions(sequence_outputs)
@@ -609,14 +638,17 @@ class AlbertForSequenceClassification(AlbertPreTrainedModel):
 
         self.init_weights()
 
-    def forward(self, input_ids, attention_mask=None, token_type_ids=None,
-                position_ids=None, head_mask=None, labels=None):
+    def forward(self, input_ids=None, attention_mask=None, token_type_ids=None,
+                position_ids=None, head_mask=None, inputs_embeds=None, labels=None):
 
-        outputs = self.albert(input_ids,
-                            attention_mask=attention_mask,
-                            token_type_ids=token_type_ids,
-                            position_ids=position_ids, 
-                            head_mask=head_mask)
+        outputs = self.albert(
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            head_mask=head_mask,
+            inputs_embeds=inputs_embeds
+        )
 
         pooled_output = outputs[1]
 
@@ -692,14 +724,17 @@ class AlbertForQuestionAnswering(AlbertPreTrainedModel):
 
         self.init_weights()
 
-    def forward(self, input_ids, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None,
-                start_positions=None, end_positions=None):
+    def forward(self, input_ids=None, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None,
+                inputs_embeds=None, start_positions=None, end_positions=None):
 
-        outputs = self.albert(input_ids,
-                            attention_mask=attention_mask,
-                            token_type_ids=token_type_ids,
-                            position_ids=position_ids, 
-                            head_mask=head_mask)
+        outputs = self.albert(
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            head_mask=head_mask,
+            inputs_embeds=inputs_embeds
+        )
 
         sequence_output = outputs[0]
 
diff --git a/transformers/modeling_tf_albert.py b/transformers/modeling_tf_albert.py
index ee8712eb28..ee122205b9 100644
--- a/transformers/modeling_tf_albert.py
+++ b/transformers/modeling_tf_albert.py
@@ -107,19 +107,25 @@ class TFAlbertEmbeddings(tf.keras.layers.Layer):
 
     def _embedding(self, inputs, training=False):
         """Applies embedding based on inputs tensor."""
-        input_ids, position_ids, token_type_ids = inputs
+        input_ids, position_ids, token_type_ids, inputs_embeds = inputs
 
-        seq_length = tf.shape(input_ids)[1]
+        if input_ids is not None:
+            input_shape = tf.shape(input_ids)
+        else:
+            input_shape = tf.shape(inputs_embeds)[:-1]
+
+        seq_length = input_shape[1]
         if position_ids is None:
             position_ids = tf.range(seq_length, dtype=tf.int32)[tf.newaxis, :]
         if token_type_ids is None:
-            token_type_ids = tf.fill(tf.shape(input_ids), 0)
+            token_type_ids = tf.fill(input_shape, 0)
 
-        words_embeddings = tf.gather(self.word_embeddings, input_ids)
+        if inputs_embeds is None:
+            inputs_embeds = tf.gather(self.word_embeddings, input_ids)
         position_embeddings = self.position_embeddings(position_ids)
         token_type_embeddings = self.token_type_embeddings(token_type_ids)
 
-        embeddings = words_embeddings + position_embeddings + token_type_embeddings
+        embeddings = inputs_embeds + position_embeddings + token_type_embeddings
         embeddings = self.LayerNorm(embeddings)
         embeddings = self.dropout(embeddings, training=training)
         return embeddings
@@ -603,6 +609,9 @@ class TFAlbertModel(TFAlbertPreTrainedModel):
         self.pooler = tf.keras.layers.Dense(config.hidden_size, kernel_initializer=get_initializer(
             config.initializer_range), activation='tanh', name='pooler')
 
+    def get_input_embeddings(self):
+        return self.embeddings
+
     def _resize_token_embeddings(self, new_num_tokens):
         raise NotImplementedError
 
@@ -613,28 +622,39 @@ class TFAlbertModel(TFAlbertPreTrainedModel):
         """
         raise NotImplementedError
 
-    def call(self, inputs, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, training=False):
+    def call(self, inputs, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, inputs_embeds=None, training=False):
         if isinstance(inputs, (tuple, list)):
             input_ids = inputs[0]
             attention_mask = inputs[1] if len(inputs) > 1 else attention_mask
             token_type_ids = inputs[2] if len(inputs) > 2 else token_type_ids
             position_ids = inputs[3] if len(inputs) > 3 else position_ids
             head_mask = inputs[4] if len(inputs) > 4 else head_mask
-            assert len(inputs) <= 5, "Too many inputs."
+            inputs_embeds = inputs[5] if len(inputs) > 5 else inputs_embeds
+            assert len(inputs) <= 6, "Too many inputs."
         elif isinstance(inputs, dict):
             input_ids = inputs.get('input_ids')
             attention_mask = inputs.get('attention_mask', attention_mask)
             token_type_ids = inputs.get('token_type_ids', token_type_ids)
             position_ids = inputs.get('position_ids', position_ids)
             head_mask = inputs.get('head_mask', head_mask)
-            assert len(inputs) <= 5, "Too many inputs."
+            inputs_embeds = inputs.get('inputs_embeds', inputs_embeds)
+            assert len(inputs) <= 6, "Too many inputs."
         else:
             input_ids = inputs
 
+        if input_ids is not None and inputs_embeds is not None:
+            raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
+        elif input_ids is not None:
+            input_shape = input_ids.shape
+        elif inputs_embeds is not None:
+            input_shape = inputs_embeds.shape[:-1]
+        else:
+            raise ValueError("You have to specify either input_ids or inputs_embeds")
+
         if attention_mask is None:
-            attention_mask = tf.fill(tf.shape(input_ids), 1)
+            attention_mask = tf.fill(input_shape, 1)
         if token_type_ids is None:
-            token_type_ids = tf.fill(tf.shape(input_ids), 0)
+            token_type_ids = tf.fill(input_shape, 0)
 
         # We create a 3D attention mask from a 2D tensor mask.
         # Sizes are [batch_size, 1, 1, to_seq_length]
@@ -664,7 +684,7 @@ class TFAlbertModel(TFAlbertPreTrainedModel):
             # head_mask = tf.constant([0] * self.num_hidden_layers)
 
         embedding_output = self.embeddings(
-            [input_ids, position_ids, token_type_ids], training=training)
+            [input_ids, position_ids, token_type_ids, inputs_embeds], training=training)
         encoder_outputs = self.encoder(
             [embedding_output, extended_attention_mask, head_mask], training=training)
 
@@ -712,6 +732,9 @@ class TFAlbertForMaskedLM(TFAlbertPreTrainedModel):
         self.predictions = TFAlbertMLMHead(
             config, self.albert.embeddings, name='predictions')
 
+    def get_output_embeddings(self):
+        return self.albert.embeddings
+
     def call(self, inputs, **kwargs):
         outputs = self.albert(inputs, **kwargs)
 

From bdfe21ab242c2175972f53ca6b4af29ada02a2bb Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Tue, 26 Nov 2019 12:54:36 -0500
Subject: [PATCH 119/505] Change param order for consistency

---
 transformers/modeling_albert.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index 7882356d24..5b7b2d3900 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -571,8 +571,8 @@ class AlbertForMaskedLM(AlbertPreTrainedModel):
     def get_output_embeddings(self):
         return self.predictions.decoder
 
-    def forward(self, input_ids=None, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None,
-                masked_lm_labels=None, inputs_embeds=None):
+    def forward(self, input_ids=None, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, inputs_embeds=None,
+                masked_lm_labels=None):
         outputs = self.albert(
             input_ids=input_ids,
             attention_mask=attention_mask,

From f2f329408db66285fd59e6628ca394381bb7f94e Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Tue, 26 Nov 2019 12:59:28 -0500
Subject: [PATCH 120/505] Fix input embeddings

---
 transformers/tests/modeling_albert_test.py    | 2 ++
 transformers/tests/modeling_tf_albert_test.py | 2 ++
 transformers/tests/modeling_tf_common_test.py | 7 ++++---
 3 files changed, 8 insertions(+), 3 deletions(-)

diff --git a/transformers/tests/modeling_albert_test.py b/transformers/tests/modeling_albert_test.py
index da87709df1..976feff9db 100644
--- a/transformers/tests/modeling_albert_test.py
+++ b/transformers/tests/modeling_albert_test.py
@@ -49,6 +49,7 @@ class AlbertModelTest(CommonTestCases.CommonModelTester):
                      use_token_type_ids=True,
                      use_labels=True,
                      vocab_size=99,
+                     embedding_size=16,
                      hidden_size=36,
                      num_hidden_layers=6,
                      num_hidden_groups=6,
@@ -73,6 +74,7 @@ class AlbertModelTest(CommonTestCases.CommonModelTester):
             self.use_token_type_ids = use_token_type_ids
             self.use_labels = use_labels
             self.vocab_size = vocab_size
+            self.embedding_size = embedding_size
             self.hidden_size = hidden_size
             self.num_hidden_layers = num_hidden_layers
             self.num_attention_heads = num_attention_heads
diff --git a/transformers/tests/modeling_tf_albert_test.py b/transformers/tests/modeling_tf_albert_test.py
index 85fc62f34f..fbd519b8f6 100644
--- a/transformers/tests/modeling_tf_albert_test.py
+++ b/transformers/tests/modeling_tf_albert_test.py
@@ -54,6 +54,7 @@ class TFAlbertModelTest(TFCommonTestCases.TFCommonModelTester):
                      use_token_type_ids=True,
                      use_labels=True,
                      vocab_size=99,
+                     embedding_size=16,
                      hidden_size=32,
                      num_hidden_layers=5,
                      num_attention_heads=4,
@@ -77,6 +78,7 @@ class TFAlbertModelTest(TFCommonTestCases.TFCommonModelTester):
             self.use_token_type_ids = use_token_type_ids
             self.use_labels = use_labels
             self.vocab_size = vocab_size
+            self.embedding_size = embedding_size
             self.hidden_size = hidden_size
             self.num_hidden_layers = num_hidden_layers
             self.num_attention_heads = num_attention_heads
diff --git a/transformers/tests/modeling_tf_common_test.py b/transformers/tests/modeling_tf_common_test.py
index 2bb7cc9c5f..31a30766cf 100644
--- a/transformers/tests/modeling_tf_common_test.py
+++ b/transformers/tests/modeling_tf_common_test.py
@@ -426,9 +426,10 @@ class TFCommonTestCases:
                     try:
                         x = wte([input_ids], mode="embedding")
                     except:
-                        x = tf.ones(input_ids.shape + [self.model_tester.hidden_size], dtype=tf.dtypes.float32)
-                # ^^ In our TF models, the input_embeddings can take slightly different forms,
-                # so we try two of them and fall back to just synthetically creating a dummy tensor of ones.
+                        if hasattr(self.model_tester, "embedding_size"):
+                            x = tf.ones(input_ids.shape + [model.config.embedding_size], dtype=tf.dtypes.float32)
+                        else:
+                            x = tf.ones(input_ids.shape + [self.model_tester.hidden_size], dtype=tf.dtypes.float32)
                 inputs_dict["inputs_embeds"] = x
                 outputs = model(inputs_dict)
 

From ae98d4599179b299563b679dd33f8a86da12980d Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Tue, 26 Nov 2019 14:12:44 -0500
Subject: [PATCH 121/505] Release: v2.2.0

---
 .circleci/deploy.sh         |  1 +
 README.md                   |  2 +-
 deploy_multi_version_doc.sh | 22 ++++++++++++++++++++++
 docs/source/conf.py         |  2 +-
 setup.py                    |  2 +-
 transformers/__init__.py    |  2 +-
 6 files changed, 27 insertions(+), 4 deletions(-)
 create mode 100644 deploy_multi_version_doc.sh

diff --git a/.circleci/deploy.sh b/.circleci/deploy.sh
index 98151963d8..c4b802c9e9 100755
--- a/.circleci/deploy.sh
+++ b/.circleci/deploy.sh
@@ -23,3 +23,4 @@ deploy_doc "fe02e45" v1.1.0
 deploy_doc "89fd345" v1.2.0
 deploy_doc "fc9faa8" v2.0.0
 deploy_doc "3ddce1d" v2.1.1
+deploy_doc "f2f3294" v2.2.0
diff --git a/README.md b/README.md
index 52e74a6f80..bb6a8eaa1a 100644
--- a/README.md
+++ b/README.md
@@ -58,7 +58,7 @@ Choose the right framework for every part of a model's lifetime
 | [Quick tour: Fine-tuning/usage scripts](#quick-tour-of-the-fine-tuningusage-scripts) | Using provided scripts: GLUE, SQuAD and Text generation |
 | [Migrating from pytorch-transformers to transformers](#Migrating-from-pytorch-transformers-to-transformers) | Migrating your code from pytorch-transformers to transformers |
 | [Migrating from pytorch-pretrained-bert to pytorch-transformers](#Migrating-from-pytorch-pretrained-bert-to-transformers) | Migrating your code from pytorch-pretrained-bert to transformers |
-| [Documentation](https://huggingface.co/transformers/) [(v2.1.1)](https://huggingface.co/transformers/v2.1.1) [(v2.0.0)](https://huggingface.co/transformers/v2.0.0) [(v1.2.0)](https://huggingface.co/transformers/v1.2.0) [(v1.1.0)](https://huggingface.co/transformers/v1.1.0) [(v1.0.0)](https://huggingface.co/transformers/v1.0.0) | Full API documentation and more |
+| [Documentation][(v2.2.0)](https://huggingface.co/transformers/v2.2.0) [(v2.1.1)](https://huggingface.co/transformers/v2.1.1) [(v2.0.0)](https://huggingface.co/transformers/v2.0.0) [(v1.2.0)](https://huggingface.co/transformers/v1.2.0) [(v1.1.0)](https://huggingface.co/transformers/v1.1.0) [(v1.0.0)](https://huggingface.co/transformers/v1.0.0) [(master](https://huggingface.co/transformers) | Full API documentation and more |
 
 ## Installation
 
diff --git a/deploy_multi_version_doc.sh b/deploy_multi_version_doc.sh
new file mode 100644
index 0000000000..bd567213eb
--- /dev/null
+++ b/deploy_multi_version_doc.sh
@@ -0,0 +1,22 @@
+cd docs
+
+function deploy_doc(){
+	echo "Creating doc at commit $1 and pushing to folder $2"
+	git checkout $1
+	if [ ! -z "$2" ] 
+	then
+		echo "Pushing version" $2
+		make clean && make html && scp -r -oStrictHostKeyChecking=no _build/html $doc:$dir/$2
+	else
+		echo "Pushing master"
+		make clean && make html && scp -r -oStrictHostKeyChecking=no _build/html/* $doc:$dir
+	fi
+}
+
+deploy_doc "master" 
+deploy_doc "b33a385" v1.0.0
+deploy_doc "fe02e45" v1.1.0
+deploy_doc "89fd345" v1.2.0
+deploy_doc "fc9faa8" v2.0.0
+deploy_doc "3ddce1d" v2.1.1
+deploy_doc "f2f3294" v2.2.0
\ No newline at end of file
diff --git a/docs/source/conf.py b/docs/source/conf.py
index 00c020ab39..f762a89cd2 100644
--- a/docs/source/conf.py
+++ b/docs/source/conf.py
@@ -26,7 +26,7 @@ author = u'huggingface'
 # The short X.Y version
 version = u''
 # The full version, including alpha/beta/rc tags
-release = u'2.1.1'
+release = u'2.2.0'
 
 
 # -- General configuration ---------------------------------------------------
diff --git a/setup.py b/setup.py
index f49aee68d4..d8dcf7b898 100644
--- a/setup.py
+++ b/setup.py
@@ -38,7 +38,7 @@ from setuptools import find_packages, setup
 
 setup(
     name="transformers",
-    version="2.1.1",
+    version="2.2.0",
     author="Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond, Google AI Language Team Authors, Open AI team Authors, Facebook AI Authors, Carnegie Mellon University Authors",
     author_email="thomas@huggingface.co",
     description="State-of-the-art Natural Language Processing for TensorFlow 2.0 and PyTorch",
diff --git a/transformers/__init__.py b/transformers/__init__.py
index baf430c17b..dff2479d4b 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -1,4 +1,4 @@
-__version__ = "2.1.1"
+__version__ = "2.2.0"
 
 # Work around to update TensorFlow's absl.logging threshold which alters the
 # default Python logging output behavior when present.

From b6321452738925983ba1306fb5396982d64f408a Mon Sep 17 00:00:00 2001
From: Lysandre Debut <lysandre@huggingface.co>
Date: Tue, 26 Nov 2019 14:27:15 -0500
Subject: [PATCH 122/505] Update master documentation link in README

---
 README.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/README.md b/README.md
index bb6a8eaa1a..67fb80aecf 100644
--- a/README.md
+++ b/README.md
@@ -58,7 +58,7 @@ Choose the right framework for every part of a model's lifetime
 | [Quick tour: Fine-tuning/usage scripts](#quick-tour-of-the-fine-tuningusage-scripts) | Using provided scripts: GLUE, SQuAD and Text generation |
 | [Migrating from pytorch-transformers to transformers](#Migrating-from-pytorch-transformers-to-transformers) | Migrating your code from pytorch-transformers to transformers |
 | [Migrating from pytorch-pretrained-bert to pytorch-transformers](#Migrating-from-pytorch-pretrained-bert-to-transformers) | Migrating your code from pytorch-pretrained-bert to transformers |
-| [Documentation][(v2.2.0)](https://huggingface.co/transformers/v2.2.0) [(v2.1.1)](https://huggingface.co/transformers/v2.1.1) [(v2.0.0)](https://huggingface.co/transformers/v2.0.0) [(v1.2.0)](https://huggingface.co/transformers/v1.2.0) [(v1.1.0)](https://huggingface.co/transformers/v1.1.0) [(v1.0.0)](https://huggingface.co/transformers/v1.0.0) [(master](https://huggingface.co/transformers) | Full API documentation and more |
+| [Documentation][(v2.2.0)](https://huggingface.co/transformers/v2.2.0) [(v2.1.1)](https://huggingface.co/transformers/v2.1.1) [(v2.0.0)](https://huggingface.co/transformers/v2.0.0) [(v1.2.0)](https://huggingface.co/transformers/v1.2.0) [(v1.1.0)](https://huggingface.co/transformers/v1.1.0) [(v1.0.0)](https://huggingface.co/transformers/v1.0.0) [(master)](https://huggingface.co/transformers) | Full API documentation and more |
 
 ## Installation
 

From cf62bdc962c53d9fb7a5820217f1bf844bb6da3b Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Tue, 26 Nov 2019 14:37:32 -0500
Subject: [PATCH 123/505] Improve test protocol for inputs_embeds in TF

cc @lysandrejik
---
 transformers/tests/modeling_tf_common_test.py | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/transformers/tests/modeling_tf_common_test.py b/transformers/tests/modeling_tf_common_test.py
index 31a30766cf..232991915c 100644
--- a/transformers/tests/modeling_tf_common_test.py
+++ b/transformers/tests/modeling_tf_common_test.py
@@ -426,10 +426,15 @@ class TFCommonTestCases:
                     try:
                         x = wte([input_ids], mode="embedding")
                     except:
-                        if hasattr(self.model_tester, "embedding_size"):
-                            x = tf.ones(input_ids.shape + [model.config.embedding_size], dtype=tf.dtypes.float32)
-                        else:
-                            x = tf.ones(input_ids.shape + [self.model_tester.hidden_size], dtype=tf.dtypes.float32)
+                        x = wte([input_ids, None, None, None], mode="embedding")
+                # ^^ In our TF models, the input_embeddings can take slightly different forms,
+                # so we try a few of them.
+                # We used to fall back to just synthetically creating a dummy tensor of ones:
+                #
+                # if hasattr(self.model_tester, "embedding_size"):
+                #     x = tf.ones(input_ids.shape + [self.model_tester.embedding_size], dtype=tf.dtypes.float32)
+                # else:
+                #     x = tf.ones(input_ids.shape + [self.model_tester.hidden_size], dtype=tf.dtypes.float32)
                 inputs_dict["inputs_embeds"] = x
                 outputs = model(inputs_dict)
 

From 8742baa53136b906e392fdae57f1c191d1b2370e Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Tue, 26 Nov 2019 14:39:47 -0500
Subject: [PATCH 124/505] Improve test protocol for inputs_embeds in TF

---
 transformers/tests/modeling_tf_common_test.py | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/transformers/tests/modeling_tf_common_test.py b/transformers/tests/modeling_tf_common_test.py
index 232991915c..ea8cd1aecd 100644
--- a/transformers/tests/modeling_tf_common_test.py
+++ b/transformers/tests/modeling_tf_common_test.py
@@ -426,15 +426,17 @@ class TFCommonTestCases:
                     try:
                         x = wte([input_ids], mode="embedding")
                     except:
-                        x = wte([input_ids, None, None, None], mode="embedding")
+                        try:
+                            x = wte([input_ids, None, None, None], mode="embedding")
+                        except:
+                            if hasattr(self.model_tester, "embedding_size"):
+                                x = tf.ones(input_ids.shape + [self.model_tester.embedding_size], dtype=tf.dtypes.float32)
+                            else:
+                                x = tf.ones(input_ids.shape + [self.model_tester.hidden_size], dtype=tf.dtypes.float32)
                 # ^^ In our TF models, the input_embeddings can take slightly different forms,
                 # so we try a few of them.
                 # We used to fall back to just synthetically creating a dummy tensor of ones:
                 #
-                # if hasattr(self.model_tester, "embedding_size"):
-                #     x = tf.ones(input_ids.shape + [self.model_tester.embedding_size], dtype=tf.dtypes.float32)
-                # else:
-                #     x = tf.ones(input_ids.shape + [self.model_tester.hidden_size], dtype=tf.dtypes.float32)
                 inputs_dict["inputs_embeds"] = x
                 outputs = model(inputs_dict)
 

From 668aac45d206bc0c8ca95e7a8b565d1ff701da7f Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Tue, 26 Nov 2019 14:52:42 -0500
Subject: [PATCH 125/505] Pretrained models

---
 docs/source/pretrained_models.rst | 33 +++++++++++++++++++++++++++++++
 1 file changed, 33 insertions(+)

diff --git a/docs/source/pretrained_models.rst b/docs/source/pretrained_models.rst
index b0a578fd80..d017a3be6c 100644
--- a/docs/source/pretrained_models.rst
+++ b/docs/source/pretrained_models.rst
@@ -159,5 +159,38 @@ Here is the full list of the currently provided pretrained models together with
 |                   |                                                            | | CamemBERT using the BERT-base architecture                                                                                          |
 |                   |                                                            | (see `details <https://github.com/pytorch/fairseq/tree/master/examples/camembert>`__)                                                 |
 +-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+| ALBERT            | ``albert-base-v1``                                         | | 12 repeating layers, 128 embedding, 768-hidden, 12-heads, 11M parameters                                                            |
+|                   |                                                            | | ALBERT base model                                                                                                                   |
+|                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
++--------------------------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   | ``albert-large-v1``                                        | | 24 repeating layers, 128 embedding, 1024-hidden, 16-heads, 17M parameters                                                           |
+|                   |                                                            | | ALBERT large model                                                                                                                  |
+|                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
++--------------------------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   | ``albert-xlarge-v1``                                       | | 24 repeating layers, 128 embedding, 2048-hidden, 16-heads, 58M parameters                                                           |
+|                   |                                                            | | ALBERT xlarge model                                                                                                                 |
+|                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
++--------------------------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   | ``albert-xxlarge-v1``                                      | | 12 repeating layer, 128 embedding, 4096-hidden, 64-heads, 223M parameters                                                           |
+|                   |                                                            | | ALBERT xxlarge model                                                                                                                |
+|                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
++--------------------------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   | ``albert-base-v2``                                         | | 12 repeating layers, 128 embedding, 768-hidden, 12-heads, 11M parameters                                                            |
+|                   |                                                            | | ALBERT base model with no dropout, additional training data and longer training                                                     |
+|                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
++--------------------------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   | ``albert-large-v2``                                        | | 24 repeating layers, 128 embedding, 1024-hidden, 16-heads, 17M parameters                                                           |
+|                   |                                                            | | ALBERT large model with no dropout, additional training data and longer training                                                    |
+|                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
++--------------------------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   | ``albert-xlarge-v2``                                       | | 24 repeating layers, 128 embedding, 2048-hidden, 16-heads, 58M parameters                                                           |
+|                   |                                                            | | ALBERT xlarge model with no dropout, additional training data and longer training                                                   |
+|                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
++--------------------------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   | ``albert-xxlarge-v2``                                      | | 12 repeating layer, 128 embedding, 4096-hidden, 64-heads, 223M parameters                                                           |
+|                   |                                                            | | ALBERT xxlarge model with no dropout, additional training data and longer training                                                  |
+|                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
++--------------------------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+
 
 .. <https://huggingface.co/transformers/examples.html>`__

From 7c6000e412288ddce7d0dbc4f4d9a12f49d28690 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Tue, 26 Nov 2019 14:55:29 -0500
Subject: [PATCH 126/505] Updated v2.2.0 doc

---
 .circleci/deploy.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.circleci/deploy.sh b/.circleci/deploy.sh
index c4b802c9e9..73960cbe99 100755
--- a/.circleci/deploy.sh
+++ b/.circleci/deploy.sh
@@ -23,4 +23,4 @@ deploy_doc "fe02e45" v1.1.0
 deploy_doc "89fd345" v1.2.0
 deploy_doc "fc9faa8" v2.0.0
 deploy_doc "3ddce1d" v2.1.1
-deploy_doc "f2f3294" v2.2.0
+deploy_doc "668aac4" v2.2.0

From ee4647bd5c1343304f7fdf20f06db1fc2524f892 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Tue, 26 Nov 2019 15:10:51 -0500
Subject: [PATCH 127/505] CamemBERT & ALBERT doc

---
 docs/source/index.rst               |  5 ++
 docs/source/model_doc/albert.rst    | 71 +++++++++++++++++++++++++++++
 docs/source/model_doc/camembert.rst | 50 ++++++++++++++++++++
 3 files changed, 126 insertions(+)
 create mode 100644 docs/source/model_doc/albert.rst
 create mode 100644 docs/source/model_doc/camembert.rst

diff --git a/docs/source/index.rst b/docs/source/index.rst
index 4cd1f48ba8..55ead33b4d 100644
--- a/docs/source/index.rst
+++ b/docs/source/index.rst
@@ -47,6 +47,9 @@ The library currently contains PyTorch and Tensorflow implementations, pre-train
 6. `XLM <https://github.com/facebookresearch/XLM>`_ (from Facebook) released together with the paper `Cross-lingual Language Model Pretraining <https://arxiv.org/abs/1901.07291>`_ by Guillaume Lample and Alexis Conneau.
 7. `RoBERTa <https://github.com/pytorch/fairseq/tree/master/examples/roberta>`_ (from Facebook), released together with the paper a `Robustly Optimized BERT Pretraining Approach <https://arxiv.org/abs/1907.11692>`_ by Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, Veselin Stoyanov.
 8. `DistilBERT <https://huggingface.co/transformers/model_doc/distilbert.html>`_ (from HuggingFace) released together with the paper `DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter <https://arxiv.org/abs/1910.01108>`_ by Victor Sanh, Lysandre Debut and Thomas Wolf. The same method has been applied to compress GPT2 into `DistilGPT2 <https://github.com/huggingface/transformers/tree/master/examples/distillation>`_.
+9. `CTRL <https://github.com/pytorch/fairseq/tree/master/examples/ctrl>`_ (from Salesforce), released together with the paper `CTRL: A Conditional Transformer Language Model for Controllable Generation <https://www.github.com/salesforce/ctrl>`_ by Nitish Shirish Keskar*, Bryan McCann*, Lav R. Varshney, Caiming Xiong and Richard Socher.
+10. `CamemBERT <https://huggingface.co/transformers/model_doc/camembert.html>`_ (from FAIR, Inria, Sorbonne Université) released together with the paper `CamemBERT: a Tasty French Language Model <https://arxiv.org/abs/1911.03894>`_ by Louis Martin, Benjamin Muller, Pedro Javier Ortiz Suarez, Yoann Dupont, Laurent Romary, Eric Villemonte de la Clergerie, Djame Seddah, and Benoît Sagot.
+11. `ALBERT <https://github.com/pytorch/fairseq/tree/master/examples/albert>`_ (from Google Research), released together with the paper a `ALBERT: A Lite BERT for Self-supervised Learning of Language Representations <https://arxiv.org/abs/1909.11942>`_ by Zhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut.
 
 .. toctree::
     :maxdepth: 2
@@ -89,3 +92,5 @@ The library currently contains PyTorch and Tensorflow implementations, pre-train
     model_doc/roberta
     model_doc/distilbert
     model_doc/ctrl
+    model_doc/camembert
+    model_doc/albert
diff --git a/docs/source/model_doc/albert.rst b/docs/source/model_doc/albert.rst
new file mode 100644
index 0000000000..cf52f35eb6
--- /dev/null
+++ b/docs/source/model_doc/albert.rst
@@ -0,0 +1,71 @@
+ALBERT
+----------------------------------------------------
+
+``AlbrtConfig``
+~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.AlbertConfig
+    :members:
+
+
+``AlbertTokenizer``
+~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.AlbertTokenizer
+    :members:
+
+
+``AlbertModel``
+~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.AlbertModel
+    :members:
+
+
+``AlbertForMaskedLM``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.AlbertForMaskedLM
+    :members:
+
+
+``AlbertForSequenceClassification``
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.AlbertForSequenceClassification
+    :members:
+
+
+``AlbertForQuestionAnswering``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.AlbertForQuestionAnswering
+    :members:
+
+
+``TFAlbertModel``
+~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.TFAlbertModel
+    :members:
+
+
+``TFBertForPreTraining``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.TFBertForPreTraining
+    :members:
+
+
+``TFAlbertForMaskedLM``
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.TFAlbertForMaskedLM
+    :members:
+
+
+``TFAlbertForSequenceClassification``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.TFAlbertForSequenceClassification
+    :members:
diff --git a/docs/source/model_doc/camembert.rst b/docs/source/model_doc/camembert.rst
new file mode 100644
index 0000000000..82ca9de945
--- /dev/null
+++ b/docs/source/model_doc/camembert.rst
@@ -0,0 +1,50 @@
+CamemBERT
+----------------------------------------------------
+
+``CamembertConfig``
+~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.CamembertConfig
+    :members:
+
+
+``CamembertTokenizer``
+~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.CamembertTokenizer
+    :members:
+
+
+``CamembertModel``
+~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.CamembertModel
+    :members:
+
+
+``CamembertForMaskedLM``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.CamembertForMaskedLM
+    :members:
+
+
+``CamembertForSequenceClassification``
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.CamembertForSequenceClassification
+    :members:
+
+
+``CamembertForMultipleChoice``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.CamembertForMultipleChoice
+    :members:
+
+
+``CamembertForTokenClassification``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.CamembertForTokenClassification
+    :members:

From 44b82c777f0a4e173fa9c68de630a63fbac1b946 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Tue, 26 Nov 2019 15:15:11 -0500
Subject: [PATCH 128/505] Updated v2.2.0 doc

---
 .circleci/deploy.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.circleci/deploy.sh b/.circleci/deploy.sh
index 73960cbe99..8081153b9f 100755
--- a/.circleci/deploy.sh
+++ b/.circleci/deploy.sh
@@ -23,4 +23,4 @@ deploy_doc "fe02e45" v1.1.0
 deploy_doc "89fd345" v1.2.0
 deploy_doc "fc9faa8" v2.0.0
 deploy_doc "3ddce1d" v2.1.1
-deploy_doc "668aac4" v2.2.0
+deploy_doc "ee4647b" v2.2.0

From cf26a0c85e0520d3e3770c9ad8dce94318ea3440 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Tue, 26 Nov 2019 15:40:03 -0500
Subject: [PATCH 129/505] Fix pretrained models table

---
 README.md                         |  1 +
 docs/source/pretrained_models.rst | 14 +++++++-------
 2 files changed, 8 insertions(+), 7 deletions(-)

diff --git a/README.md b/README.md
index 67fb80aecf..dd2c865036 100644
--- a/README.md
+++ b/README.md
@@ -134,6 +134,7 @@ At some point in the future, you'll be able to seamlessly move from pre-training
 8. **[DistilBERT](https://github.com/huggingface/transformers/tree/master/examples/distillation)** (from HuggingFace), released together with the paper [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://arxiv.org/abs/1910.01108) by Victor Sanh, Lysandre Debut and Thomas Wolf. The same method has been applied to compress GPT2 into [DistilGPT2](https://github.com/huggingface/transformers/tree/master/examples/distillation).
 9. **[CTRL](https://github.com/salesforce/ctrl/)** (from Salesforce) released with the paper [CTRL: A Conditional Transformer Language Model for Controllable Generation](https://arxiv.org/abs/1909.05858) by Nitish Shirish Keskar*, Bryan McCann*, Lav R. Varshney, Caiming Xiong and Richard Socher.
 10. **[CamemBERT](https://camembert-model.fr)** (from Inria/Facebook/Sorbonne) released with the paper [CamemBERT: a Tasty French Language Model](https://arxiv.org/abs/1911.03894) by Louis Martin*, Benjamin Muller*, Pedro Javier Ortiz Suárez*, Yoann Dupont, Laurent Romary, Éric Villemonte de la Clergerie, Djamé Seddah and Benoît Sagot.
+11. **[ALBERT](https://github.com/google-research/google-research/tree/master/albert)** (from Google Research and the Toyota Technological Institute at Chicago) released with the paper [ALBERT: A Lite BERT for Self-supervised Learning of Language Representations](https://arxiv.org/abs/1909.11942), by Zhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut.
 11. Want to contribute a new model? We have added a **detailed guide and templates** to guide you in the process of adding a new model. You can find them in the [`templates`](./templates) folder of the repository. Be sure to check the [contributing guidelines](./CONTRIBUTING.md) and contact the maintainers or open an issue to collect feedbacks before starting your PR.
 
 These implementations have been tested on several datasets (see the example scripts) and should match the performances of the original implementations (e.g. ~93 F1 on SQuAD for BERT Whole-Word-Masking, ~88 F1 on RocStories for OpenAI GPT, ~18.3 perplexity on WikiText 103 for Transformer-XL, ~0.916 Peason R coefficient on STS-B for XLNet). You can find more details on the performances in the Examples section of the [documentation](https://huggingface.co/transformers/examples.html).
diff --git a/docs/source/pretrained_models.rst b/docs/source/pretrained_models.rst
index d017a3be6c..a6bf508d18 100644
--- a/docs/source/pretrained_models.rst
+++ b/docs/source/pretrained_models.rst
@@ -162,31 +162,31 @@ Here is the full list of the currently provided pretrained models together with
 | ALBERT            | ``albert-base-v1``                                         | | 12 repeating layers, 128 embedding, 768-hidden, 12-heads, 11M parameters                                                            |
 |                   |                                                            | | ALBERT base model                                                                                                                   |
 |                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
-+--------------------------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``albert-large-v1``                                        | | 24 repeating layers, 128 embedding, 1024-hidden, 16-heads, 17M parameters                                                           |
 |                   |                                                            | | ALBERT large model                                                                                                                  |
 |                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
-+--------------------------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``albert-xlarge-v1``                                       | | 24 repeating layers, 128 embedding, 2048-hidden, 16-heads, 58M parameters                                                           |
 |                   |                                                            | | ALBERT xlarge model                                                                                                                 |
 |                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
-+--------------------------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``albert-xxlarge-v1``                                      | | 12 repeating layer, 128 embedding, 4096-hidden, 64-heads, 223M parameters                                                           |
 |                   |                                                            | | ALBERT xxlarge model                                                                                                                |
 |                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
-+--------------------------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``albert-base-v2``                                         | | 12 repeating layers, 128 embedding, 768-hidden, 12-heads, 11M parameters                                                            |
 |                   |                                                            | | ALBERT base model with no dropout, additional training data and longer training                                                     |
 |                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
-+--------------------------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``albert-large-v2``                                        | | 24 repeating layers, 128 embedding, 1024-hidden, 16-heads, 17M parameters                                                           |
 |                   |                                                            | | ALBERT large model with no dropout, additional training data and longer training                                                    |
 |                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
-+--------------------------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``albert-xlarge-v2``                                       | | 24 repeating layers, 128 embedding, 2048-hidden, 16-heads, 58M parameters                                                           |
 |                   |                                                            | | ALBERT xlarge model with no dropout, additional training data and longer training                                                   |
 |                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
-+--------------------------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``albert-xxlarge-v2``                                      | | 12 repeating layer, 128 embedding, 4096-hidden, 64-heads, 223M parameters                                                           |
 |                   |                                                            | | ALBERT xxlarge model with no dropout, additional training data and longer training                                                  |
 |                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |

From ce02550d50f2074d54d58b37cbc9845d9a159818 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Tue, 26 Nov 2019 15:47:02 -0500
Subject: [PATCH 130/505] Fix pretrained models table

---
 docs/source/pretrained_models.rst | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/source/pretrained_models.rst b/docs/source/pretrained_models.rst
index a6bf508d18..a312346df7 100644
--- a/docs/source/pretrained_models.rst
+++ b/docs/source/pretrained_models.rst
@@ -190,7 +190,7 @@ Here is the full list of the currently provided pretrained models together with
 |                   | ``albert-xxlarge-v2``                                      | | 12 repeating layer, 128 embedding, 4096-hidden, 64-heads, 223M parameters                                                           |
 |                   |                                                            | | ALBERT xxlarge model with no dropout, additional training data and longer training                                                  |
 |                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
-+--------------------------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
++-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 
 
 .. <https://huggingface.co/transformers/examples.html>`__

From cc7968227e08858df4a5c618c739e1a3ca050196 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Tue, 26 Nov 2019 15:52:25 -0500
Subject: [PATCH 131/505] Updated v2.2.0 doc

---
 .circleci/deploy.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.circleci/deploy.sh b/.circleci/deploy.sh
index 8081153b9f..61b9f90909 100755
--- a/.circleci/deploy.sh
+++ b/.circleci/deploy.sh
@@ -23,4 +23,4 @@ deploy_doc "fe02e45" v1.1.0
 deploy_doc "89fd345" v1.2.0
 deploy_doc "fc9faa8" v2.0.0
 deploy_doc "3ddce1d" v2.1.1
-deploy_doc "ee4647b" v2.2.0
+deploy_doc "ce02550" v2.2.0

From 361620954acf16b27727d763a591257b03f90b5d Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Wed, 27 Nov 2019 10:11:37 -0500
Subject: [PATCH 132/505] Remove TFBertForPreTraining from ALBERT doc

---
 docs/source/model_doc/albert.rst | 7 -------
 1 file changed, 7 deletions(-)

diff --git a/docs/source/model_doc/albert.rst b/docs/source/model_doc/albert.rst
index cf52f35eb6..92970c9328 100644
--- a/docs/source/model_doc/albert.rst
+++ b/docs/source/model_doc/albert.rst
@@ -50,13 +50,6 @@ ALBERT
     :members:
 
 
-``TFBertForPreTraining``
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.TFBertForPreTraining
-    :members:
-
-
 ``TFAlbertForMaskedLM``
 ~~~~~~~~~~~~~~~~~~~~~~~~~~
 

From 45d767297a78697e4ff75ec3d83bdf35e43e4fff Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Wed, 27 Nov 2019 10:12:20 -0500
Subject: [PATCH 133/505] Updated v2.2.0 doc

---
 .circleci/deploy.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.circleci/deploy.sh b/.circleci/deploy.sh
index 61b9f90909..a32581baef 100755
--- a/.circleci/deploy.sh
+++ b/.circleci/deploy.sh
@@ -23,4 +23,4 @@ deploy_doc "fe02e45" v1.1.0
 deploy_doc "89fd345" v1.2.0
 deploy_doc "fc9faa8" v2.0.0
 deploy_doc "3ddce1d" v2.1.1
-deploy_doc "ce02550" v2.2.0
+deploy_doc "3616209" v2.2.0

From 88b317739fe56888528c857fc8e90967148a0051 Mon Sep 17 00:00:00 2001
From: root <huangyingbiao@pku.edu.cn>
Date: Wed, 27 Nov 2019 09:58:38 +0000
Subject: [PATCH 134/505] Fix issue: #1962, input's shape seem to cause error
 in 2.2.0 version tf_albert_model

---
 transformers/modeling_tf_albert.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/modeling_tf_albert.py b/transformers/modeling_tf_albert.py
index ee122205b9..b2bf66f750 100644
--- a/transformers/modeling_tf_albert.py
+++ b/transformers/modeling_tf_albert.py
@@ -645,7 +645,7 @@ class TFAlbertModel(TFAlbertPreTrainedModel):
         if input_ids is not None and inputs_embeds is not None:
             raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
         elif input_ids is not None:
-            input_shape = input_ids.shape
+            input_shape = tf.shape(input_ids)
         elif inputs_embeds is not None:
             input_shape = inputs_embeds.shape[:-1]
         else:

From de2696f68e20019fef3a5e1b54de10351abb4145 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Tue, 26 Nov 2019 16:30:27 +0100
Subject: [PATCH 135/505] suggest to track repo w/ https rather than ssh

---
 CONTRIBUTING.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
index 136ef8df81..8228dd59d8 100644
--- a/CONTRIBUTING.md
+++ b/CONTRIBUTING.md
@@ -106,7 +106,7 @@ Follow these steps to start contributing:
    ```bash
    $ git clone git@github.com:<your Github handle>/transformers.git
    $ cd transformers
-   $ git remote add upstream git@github.com:huggingface/transformers.git
+   $ git remote add upstream https://github.com/huggingface/transformers.git
    ```
 
 3. Create a new branch to hold your development changes:

From b5d884d25c96069962030b9e4c9dfb4b1f2f6fa0 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Wed, 27 Nov 2019 11:05:18 -0500
Subject: [PATCH 136/505] Uniformize #1952

---
 README.md          | 2 +-
 examples/README.md | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/README.md b/README.md
index dd2c865036..dd20b80590 100644
--- a/README.md
+++ b/README.md
@@ -92,7 +92,7 @@ Examples are included in the repository but are not shipped with the library.
 Therefore, in order to run the latest versions of the examples you also need to install from source. To do so, create a new virtual environment and follow these steps:
 
 ```bash
-git clone git@github.com:huggingface/transformers
+git clone https://github.com/huggingface/transformers
 cd transformers
 pip install [--editable] .
 ```
diff --git a/examples/README.md b/examples/README.md
index 0c57990ea7..e109a12171 100644
--- a/examples/README.md
+++ b/examples/README.md
@@ -7,7 +7,7 @@ similar API between the different models.
 To run the latest versions of the examples, you have to install from source. Execute the following steps in a new virtual environment:
 
 ```bash
-git clone git@github.com:huggingface/transformers
+git clone https://github.com/huggingface/transformers
 cd transformers
 pip install [--editable] .
 ```

From 71f71ddb3e04aa18aaa1e3633a2881493e67b438 Mon Sep 17 00:00:00 2001
From: VictorSanh <victorsanh@gmail.com>
Date: Tue, 29 Oct 2019 11:50:42 -0400
Subject: [PATCH 137/505] run_xnli + utils_xnli

---
 examples/run_xnli.py   | 534 +++++++++++++++++++++++++++++++++++++++++
 examples/utils_xnli.py |  93 +++++++
 2 files changed, 627 insertions(+)
 create mode 100644 examples/run_xnli.py
 create mode 100644 examples/utils_xnli.py

diff --git a/examples/run_xnli.py b/examples/run_xnli.py
new file mode 100644
index 0000000000..ee37296832
--- /dev/null
+++ b/examples/run_xnli.py
@@ -0,0 +1,534 @@
+# coding=utf-8
+# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.
+# Copyright (c) 2018, NVIDIA CORPORATION.  All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" Finetuning multi-lingual models on XNLI (Bert, XLM).
+    Adapted from `examples/run_glue.py`"""
+
+from __future__ import absolute_import, division, print_function
+
+import argparse
+import glob
+import logging
+import os
+import random
+
+import numpy as np
+import torch
+from torch.utils.data import (DataLoader, RandomSampler, SequentialSampler,
+                              TensorDataset)
+from torch.utils.data.distributed import DistributedSampler
+
+try:
+    from torch.utils.tensorboard import SummaryWriter
+except:
+    from tensorboardX import SummaryWriter
+
+from tqdm import tqdm, trange
+
+from transformers import (WEIGHTS_NAME, 
+                          BertConfig, BertForSequenceClassification, BertTokenizer,
+                          XLMConfig, XLMForSequenceClassification, XLMTokenizer,
+                          DistilBertConfig, DistilBertForSequenceClassification, DistilBertTokenizer)
+
+from transformers import AdamW, WarmupLinearSchedule
+
+from utils_xnli import xnli_compute_metrics as compute_metrics
+from utils_xnli import xnli_output_modes as output_modes
+from utils_xnli import xnli_processors as processors
+
+from transformers import glue_convert_examples_to_features as convert_examples_to_features
+
+logger = logging.getLogger(__name__)
+
+ALL_MODELS = sum((tuple(conf.pretrained_config_archive_map.keys()) for conf in (BertConfig, XLMConfig)), ())
+
+MODEL_CLASSES = {
+    'bert': (BertConfig, BertForSequenceClassification, BertTokenizer),
+    'xlm': (XLMConfig, XLMForSequenceClassification, XLMTokenizer),
+    # 'distilbert': (DistilBertConfig, DistilBertForSequenceClassification, DistilBertTokenizer)
+}
+
+
+def set_seed(args):
+    random.seed(args.seed)
+    np.random.seed(args.seed)
+    torch.manual_seed(args.seed)
+    if args.n_gpu > 0:
+        torch.cuda.manual_seed_all(args.seed)
+
+
+def train(args, train_dataset, model, tokenizer):
+    """ Train the model """
+    if args.local_rank in [-1, 0]:
+        tb_writer = SummaryWriter()
+
+    args.train_batch_size = args.per_gpu_train_batch_size * max(1, args.n_gpu)
+    train_sampler = RandomSampler(train_dataset) if args.local_rank == -1 else DistributedSampler(train_dataset)
+    train_dataloader = DataLoader(train_dataset, sampler=train_sampler, batch_size=args.train_batch_size)
+
+    if args.max_steps > 0:
+        t_total = args.max_steps
+        args.num_train_epochs = args.max_steps // (len(train_dataloader) // args.gradient_accumulation_steps) + 1
+    else:
+        t_total = len(train_dataloader) // args.gradient_accumulation_steps * args.num_train_epochs
+
+    # Prepare optimizer and schedule (linear warmup and decay)
+    no_decay = ['bias', 'LayerNorm.weight']
+    optimizer_grouped_parameters = [
+        {'params': [p for n, p in model.named_parameters() if not any(nd in n for nd in no_decay)], 'weight_decay': args.weight_decay},
+        {'params': [p for n, p in model.named_parameters() if any(nd in n for nd in no_decay)], 'weight_decay': 0.0}
+        ]
+    optimizer = AdamW(optimizer_grouped_parameters, lr=args.learning_rate, eps=args.adam_epsilon)
+    scheduler = WarmupLinearSchedule(optimizer, warmup_steps=args.warmup_steps, t_total=t_total)
+    if args.fp16:
+        try:
+            from apex import amp
+        except ImportError:
+            raise ImportError("Please install apex from https://www.github.com/nvidia/apex to use fp16 training.")
+        model, optimizer = amp.initialize(model, optimizer, opt_level=args.fp16_opt_level)
+
+    # multi-gpu training (should be after apex fp16 initialization)
+    if args.n_gpu > 1:
+        model = torch.nn.DataParallel(model)
+
+    # Distributed training (should be after apex fp16 initialization)
+    if args.local_rank != -1:
+        model = torch.nn.parallel.DistributedDataParallel(model, device_ids=[args.local_rank],
+                                                          output_device=args.local_rank,
+                                                          find_unused_parameters=True)
+
+    # Train!
+    logger.info("***** Running training *****")
+    logger.info("  Num examples = %d", len(train_dataset))
+    logger.info("  Num Epochs = %d", args.num_train_epochs)
+    logger.info("  Instantaneous batch size per GPU = %d", args.per_gpu_train_batch_size)
+    logger.info("  Total train batch size (w. parallel, distributed & accumulation) = %d",
+                   args.train_batch_size * args.gradient_accumulation_steps * (torch.distributed.get_world_size() if args.local_rank != -1 else 1))
+    logger.info("  Gradient Accumulation steps = %d", args.gradient_accumulation_steps)
+    logger.info("  Total optimization steps = %d", t_total)
+
+    global_step = 0
+    tr_loss, logging_loss = 0.0, 0.0
+    model.zero_grad()
+    train_iterator = trange(int(args.num_train_epochs), desc="Epoch", disable=args.local_rank not in [-1, 0])
+    set_seed(args)  # Added here for reproductibility (even between python 2 and 3)
+    for _ in train_iterator:
+        epoch_iterator = tqdm(train_dataloader, desc="Iteration", disable=args.local_rank not in [-1, 0])
+        for step, batch in enumerate(epoch_iterator):
+            model.train()
+            batch = tuple(t.to(args.device) for t in batch)
+            inputs = {'input_ids':      batch[0],
+                      'attention_mask': batch[1],
+                      'labels':         batch[3]}
+            if args.model_type != 'distilbert':
+                inputs['token_type_ids'] = batch[2] if args.model_type in ['bert'] else None  # XLM and DistilBERT don't use segment_ids
+            outputs = model(**inputs)
+            loss = outputs[0]  # model outputs are always tuple in transformers (see doc)
+
+            if args.n_gpu > 1:
+                loss = loss.mean() # mean() to average on multi-gpu parallel training
+            if args.gradient_accumulation_steps > 1:
+                loss = loss / args.gradient_accumulation_steps
+
+            if args.fp16:
+                with amp.scale_loss(loss, optimizer) as scaled_loss:
+                    scaled_loss.backward()
+            else:
+                loss.backward()
+
+            tr_loss += loss.item()
+            if (step + 1) % args.gradient_accumulation_steps == 0 and not args.tpu:
+                if args.fp16:
+                    torch.nn.utils.clip_grad_norm_(amp.master_params(optimizer), args.max_grad_norm)
+                else:
+                    torch.nn.utils.clip_grad_norm_(model.parameters(), args.max_grad_norm)
+
+                optimizer.step()
+                scheduler.step()  # Update learning rate schedule
+                model.zero_grad()
+                global_step += 1
+
+                if args.local_rank in [-1, 0] and args.logging_steps > 0 and global_step % args.logging_steps == 0:
+                    # Log metrics
+                    if args.local_rank == -1 and args.evaluate_during_training:  # Only evaluate when single GPU otherwise metrics may not average well
+                        results = evaluate(args, model, tokenizer)
+                        for key, value in results.items():
+                            tb_writer.add_scalar('eval_{}'.format(key), value, global_step)
+                    tb_writer.add_scalar('lr', scheduler.get_lr()[0], global_step)
+                    tb_writer.add_scalar('loss', (tr_loss - logging_loss)/args.logging_steps, global_step)
+                    logging_loss = tr_loss
+
+                if args.local_rank in [-1, 0] and args.save_steps > 0 and global_step % args.save_steps == 0:
+                    # Save model checkpoint
+                    output_dir = os.path.join(args.output_dir, 'checkpoint-{}'.format(global_step))
+                    if not os.path.exists(output_dir):
+                        os.makedirs(output_dir)
+                    model_to_save = model.module if hasattr(model, 'module') else model  # Take care of distributed/parallel training
+                    model_to_save.save_pretrained(output_dir)
+                    torch.save(args, os.path.join(output_dir, 'training_args.bin'))
+                    logger.info("Saving model checkpoint to %s", output_dir)
+
+            if args.tpu:
+                args.xla_model.optimizer_step(optimizer, barrier=True)
+                model.zero_grad()
+                global_step += 1
+
+            if args.max_steps > 0 and global_step > args.max_steps:
+                epoch_iterator.close()
+                break
+        if args.max_steps > 0 and global_step > args.max_steps:
+            train_iterator.close()
+            break
+
+    if args.local_rank in [-1, 0]:
+        tb_writer.close()
+
+    return global_step, tr_loss / global_step
+
+
+def evaluate(args, model, tokenizer, prefix=""):
+    eval_task_names = (args.task_name,)
+    eval_outputs_dirs = (args.output_dir,)
+
+    results = {}
+    for eval_task, eval_output_dir in zip(eval_task_names, eval_outputs_dirs):
+        eval_dataset = load_and_cache_examples(args, eval_task, tokenizer, evaluate=True)
+
+        if not os.path.exists(eval_output_dir) and args.local_rank in [-1, 0]:
+            os.makedirs(eval_output_dir)
+
+        args.eval_batch_size = args.per_gpu_eval_batch_size * max(1, args.n_gpu)
+        # Note that DistributedSampler samples randomly
+        eval_sampler = SequentialSampler(eval_dataset) if args.local_rank == -1 else DistributedSampler(eval_dataset)
+        eval_dataloader = DataLoader(eval_dataset, sampler=eval_sampler, batch_size=args.eval_batch_size)
+
+        # Eval!
+        logger.info("***** Running evaluation {} *****".format(prefix))
+        logger.info("  Num examples = %d", len(eval_dataset))
+        logger.info("  Batch size = %d", args.eval_batch_size)
+        eval_loss = 0.0
+        nb_eval_steps = 0
+        preds = None
+        out_label_ids = None
+        for batch in tqdm(eval_dataloader, desc="Evaluating"):
+            model.eval()
+            batch = tuple(t.to(args.device) for t in batch)
+
+            with torch.no_grad():
+                inputs = {'input_ids':      batch[0],
+                          'attention_mask': batch[1],
+                          'labels':         batch[3]}
+                if args.model_type != 'distilbert':
+                    inputs['token_type_ids'] = batch[2] if args.model_type in ['bert'] else None  # XLM and DistilBERT don't use segment_ids
+                outputs = model(**inputs)
+                tmp_eval_loss, logits = outputs[:2]
+
+                eval_loss += tmp_eval_loss.mean().item()
+            nb_eval_steps += 1
+            if preds is None:
+                preds = logits.detach().cpu().numpy()
+                out_label_ids = inputs['labels'].detach().cpu().numpy()
+            else:
+                preds = np.append(preds, logits.detach().cpu().numpy(), axis=0)
+                out_label_ids = np.append(out_label_ids, inputs['labels'].detach().cpu().numpy(), axis=0)
+
+        eval_loss = eval_loss / nb_eval_steps
+        if args.output_mode == "classification":
+            preds = np.argmax(preds, axis=1)
+        elif args.output_mode == "regression":
+            preds = np.squeeze(preds)
+        result = compute_metrics(eval_task, preds, out_label_ids)
+        results.update(result)
+
+        output_eval_file = os.path.join(eval_output_dir, prefix, "eval_results.txt")
+        with open(output_eval_file, "w") as writer:
+            logger.info("***** Eval results {} *****".format(prefix))
+            for key in sorted(result.keys()):
+                logger.info("  %s = %s", key, str(result[key]))
+                writer.write("%s = %s\n" % (key, str(result[key])))
+
+    return results
+
+
+def load_and_cache_examples(args, task, tokenizer, evaluate=False):
+    if args.local_rank not in [-1, 0] and not evaluate:
+        torch.distributed.barrier()  # Make sure only the first process in distributed training process the dataset, and the others will use the cache
+
+    processor = processors[task](language=args.language, train_language=args.train_language)
+    output_mode = output_modes[task]
+    # Load data features from cache or dataset file
+    cached_features_file = os.path.join(args.data_dir, 'cached_{}_{}_{}_{}_{}'.format(
+        'dev' if evaluate else 'train',
+        list(filter(None, args.model_name_or_path.split('/'))).pop(),
+        str(args.max_seq_length),
+        str(task),
+        str(args.train_language if (not evaluate and args.train_language is not None) else args.language)))
+    if os.path.exists(cached_features_file) and not args.overwrite_cache:
+        logger.info("Loading features from cached file %s", cached_features_file)
+        features = torch.load(cached_features_file)
+    else:
+        logger.info("Creating features from dataset file at %s", args.data_dir)
+        label_list = processor.get_labels()
+        examples = processor.get_dev_examples(args.data_dir) if evaluate else processor.get_train_examples(args.data_dir)
+        features = convert_examples_to_features(examples,
+                                                tokenizer,
+                                                label_list=label_list,
+                                                max_length=args.max_seq_length,
+                                                output_mode=output_mode,
+                                                pad_on_left=False,
+                                                pad_token=tokenizer.convert_tokens_to_ids([tokenizer.pad_token])[0],
+                                                pad_token_segment_id=0,
+        )
+        if args.local_rank in [-1, 0]:
+            logger.info("Saving features into cached file %s", cached_features_file)
+            torch.save(features, cached_features_file)
+
+    if args.local_rank == 0 and not evaluate:
+        torch.distributed.barrier()  # Make sure only the first process in distributed training process the dataset, and the others will use the cache
+
+    # Convert to Tensors and build dataset
+    all_input_ids = torch.tensor([f.input_ids for f in features], dtype=torch.long)
+    all_attention_mask = torch.tensor([f.attention_mask for f in features], dtype=torch.long)
+    all_token_type_ids = torch.tensor([f.token_type_ids for f in features], dtype=torch.long)
+    if output_mode == "classification":
+        all_labels = torch.tensor([f.label for f in features], dtype=torch.long)
+    else:
+        raise ValueError(f'No other `output_mode` for XNLI.')
+
+    dataset = TensorDataset(all_input_ids, all_attention_mask, all_token_type_ids, all_labels)
+    return dataset
+
+
+def main():
+    parser = argparse.ArgumentParser()
+
+    ## Required parameters
+    parser.add_argument("--data_dir", default=None, type=str, required=True,
+                        help="The input data dir. Should contain the .tsv files (or other data files) for the task.")
+    parser.add_argument("--model_type", default=None, type=str, required=True,
+                        help="Model type selected in the list: " + ", ".join(MODEL_CLASSES.keys()))
+    parser.add_argument("--model_name_or_path", default=None, type=str, required=True,
+                        help="Path to pre-trained model or shortcut name selected in the list: " + ", ".join(ALL_MODELS))
+    parser.add_argument("--language", default=None, type=str, required=True,
+                        help="Evaluation language. Also train language if `train_language` is set to None.")
+    parser.add_argument("--train_language", default=None, type=str,
+                        help="Train language if is different of the evaluation language.")
+    parser.add_argument("--output_dir", default=None, type=str, required=True,
+                        help="The output directory where the model predictions and checkpoints will be written.")
+
+    ## Other parameters
+    parser.add_argument("--config_name", default="", type=str,
+                        help="Pretrained config name or path if not the same as model_name")
+    parser.add_argument("--tokenizer_name", default="", type=str,
+                        help="Pretrained tokenizer name or path if not the same as model_name")
+    parser.add_argument("--cache_dir", default="", type=str,
+                        help="Where do you want to store the pre-trained models downloaded from s3")
+    parser.add_argument("--max_seq_length", default=128, type=int,
+                        help="The maximum total input sequence length after tokenization. Sequences longer "
+                             "than this will be truncated, sequences shorter will be padded.")
+    parser.add_argument("--do_train", action='store_true',
+                        help="Whether to run training.")
+    parser.add_argument("--do_eval", action='store_true',
+                        help="Whether to run eval on the dev set.")
+    parser.add_argument("--evaluate_during_training", action='store_true',
+                        help="Rul evaluation during training at each logging step.")
+    parser.add_argument("--do_lower_case", action='store_true',
+                        help="Set this flag if you are using an uncased model.")
+
+    parser.add_argument("--per_gpu_train_batch_size", default=8, type=int,
+                        help="Batch size per GPU/CPU for training.")
+    parser.add_argument("--per_gpu_eval_batch_size", default=8, type=int,
+                        help="Batch size per GPU/CPU for evaluation.")
+    parser.add_argument('--gradient_accumulation_steps', type=int, default=1,
+                        help="Number of updates steps to accumulate before performing a backward/update pass.")
+    parser.add_argument("--learning_rate", default=5e-5, type=float,
+                        help="The initial learning rate for Adam.")
+    parser.add_argument("--weight_decay", default=0.0, type=float,
+                        help="Weight deay if we apply some.")
+    parser.add_argument("--adam_epsilon", default=1e-8, type=float,
+                        help="Epsilon for Adam optimizer.")
+    parser.add_argument("--max_grad_norm", default=1.0, type=float,
+                        help="Max gradient norm.")
+    parser.add_argument("--num_train_epochs", default=3.0, type=float,
+                        help="Total number of training epochs to perform.")
+    parser.add_argument("--max_steps", default=-1, type=int,
+                        help="If > 0: set total number of training steps to perform. Override num_train_epochs.")
+    parser.add_argument("--warmup_steps", default=0, type=int,
+                        help="Linear warmup over warmup_steps.")
+
+    parser.add_argument('--logging_steps', type=int, default=50,
+                        help="Log every X updates steps.")
+    parser.add_argument('--save_steps', type=int, default=50,
+                        help="Save checkpoint every X updates steps.")
+    parser.add_argument("--eval_all_checkpoints", action='store_true',
+                        help="Evaluate all checkpoints starting with the same prefix as model_name ending and ending with step number")
+    parser.add_argument("--no_cuda", action='store_true',
+                        help="Avoid using CUDA when available")
+    parser.add_argument('--overwrite_output_dir', action='store_true',
+                        help="Overwrite the content of the output directory")
+    parser.add_argument('--overwrite_cache', action='store_true',
+                        help="Overwrite the cached training and evaluation sets")
+    parser.add_argument('--seed', type=int, default=42,
+                        help="random seed for initialization")
+
+    parser.add_argument('--tpu', action='store_true',
+                        help="Whether to run on the TPU defined in the environment variables")
+    parser.add_argument('--tpu_ip_address', type=str, default='',
+                        help="TPU IP address if none are set in the environment variables")
+    parser.add_argument('--tpu_name', type=str, default='',
+                        help="TPU name if none are set in the environment variables")
+    parser.add_argument('--xrt_tpu_config', type=str, default='',
+                        help="XRT TPU config if none are set in the environment variables")
+
+    parser.add_argument('--fp16', action='store_true',
+                        help="Whether to use 16-bit (mixed) precision (through NVIDIA apex) instead of 32-bit")
+    parser.add_argument('--fp16_opt_level', type=str, default='O1',
+                        help="For fp16: Apex AMP optimization level selected in ['O0', 'O1', 'O2', and 'O3']."
+                             "See details at https://nvidia.github.io/apex/amp.html")
+    parser.add_argument("--local_rank", type=int, default=-1,
+                        help="For distributed training: local_rank")
+    parser.add_argument('--server_ip', type=str, default='', help="For distant debugging.")
+    parser.add_argument('--server_port', type=str, default='', help="For distant debugging.")
+    args = parser.parse_args()
+
+    if os.path.exists(args.output_dir) and os.listdir(args.output_dir) and args.do_train and not args.overwrite_output_dir:
+        raise ValueError("Output directory ({}) already exists and is not empty. Use --overwrite_output_dir to overcome.".format(args.output_dir))
+
+    # Setup distant debugging if needed
+    if args.server_ip and args.server_port:
+        # Distant debugging - see https://code.visualstudio.com/docs/python/debugging#_attach-to-a-local-script
+        import ptvsd
+        print("Waiting for debugger attach")
+        ptvsd.enable_attach(address=(args.server_ip, args.server_port), redirect_output=True)
+        ptvsd.wait_for_attach()
+
+    # Setup CUDA, GPU & distributed training
+    if args.local_rank == -1 or args.no_cuda:
+        device = torch.device("cuda" if torch.cuda.is_available() and not args.no_cuda else "cpu")
+        args.n_gpu = torch.cuda.device_count()
+    else:  # Initializes the distributed backend which will take care of sychronizing nodes/GPUs
+        torch.cuda.set_device(args.local_rank)
+        device = torch.device("cuda", args.local_rank)
+        torch.distributed.init_process_group(backend='nccl')
+        args.n_gpu = 1
+    args.device = device
+
+    if args.tpu:
+        if args.tpu_ip_address:
+            os.environ["TPU_IP_ADDRESS"] = args.tpu_ip_address
+        if args.tpu_name:
+            os.environ["TPU_NAME"] = args.tpu_name
+        if args.xrt_tpu_config:
+            os.environ["XRT_TPU_CONFIG"] = args.xrt_tpu_config
+
+        assert "TPU_IP_ADDRESS" in os.environ
+        assert "TPU_NAME" in os.environ
+        assert "XRT_TPU_CONFIG" in os.environ
+
+        import torch_xla
+        import torch_xla.core.xla_model as xm
+        args.device = xm.xla_device()
+        args.xla_model = xm
+
+    # Setup logging
+    logging.basicConfig(format = '%(asctime)s - %(levelname)s - %(name)s -   %(message)s',
+                        datefmt = '%m/%d/%Y %H:%M:%S',
+                        level = logging.INFO if args.local_rank in [-1, 0] else logging.WARN)
+    logger.warning("Process rank: %s, device: %s, n_gpu: %s, distributed training: %s, 16-bits training: %s",
+                    args.local_rank, device, args.n_gpu, bool(args.local_rank != -1), args.fp16)
+
+    # Set seed
+    set_seed(args)
+
+    # Prepare XNLI task
+    args.task_name = 'xnli'
+    if args.task_name not in processors:
+        raise ValueError("Task not found: %s" % (args.task_name))
+    processor = processors[args.task_name](language=args.language, train_language=args.train_language)
+    args.output_mode = output_modes[args.task_name]
+    label_list = processor.get_labels()
+    num_labels = len(label_list)
+
+    # Load pretrained model and tokenizer
+    if args.local_rank not in [-1, 0]:
+        torch.distributed.barrier()  # Make sure only the first process in distributed training will download model & vocab
+
+    args.model_type = args.model_type.lower()
+    config_class, model_class, tokenizer_class = MODEL_CLASSES[args.model_type]
+    config = config_class.from_pretrained(args.config_name if args.config_name else args.model_name_or_path, num_labels=num_labels, finetuning_task=args.task_name)
+    tokenizer = tokenizer_class.from_pretrained(args.tokenizer_name if args.tokenizer_name else args.model_name_or_path, do_lower_case=args.do_lower_case)
+    model = model_class.from_pretrained(args.model_name_or_path, from_tf=bool('.ckpt' in args.model_name_or_path), config=config)
+
+    if args.local_rank == 0:
+        torch.distributed.barrier()  # Make sure only the first process in distributed training will download model & vocab
+
+    model.to(args.device)
+
+    logger.info("Training/evaluation parameters %s", args)
+
+
+    # Training
+    if args.do_train:
+        train_dataset = load_and_cache_examples(args, args.task_name, tokenizer, evaluate=False)
+        global_step, tr_loss = train(args, train_dataset, model, tokenizer)
+        logger.info(" global_step = %s, average loss = %s", global_step, tr_loss)
+
+
+    # Saving best-practices: if you use defaults names for the model, you can reload it using from_pretrained()
+    if args.do_train and (args.local_rank == -1 or torch.distributed.get_rank() == 0) and not args.tpu:
+        # Create output directory if needed
+        if not os.path.exists(args.output_dir) and args.local_rank in [-1, 0]:
+            os.makedirs(args.output_dir)
+
+        logger.info("Saving model checkpoint to %s", args.output_dir)
+        # Save a trained model, configuration and tokenizer using `save_pretrained()`.
+        # They can then be reloaded using `from_pretrained()`
+        model_to_save = model.module if hasattr(model, 'module') else model  # Take care of distributed/parallel training
+        model_to_save.save_pretrained(args.output_dir)
+        tokenizer.save_pretrained(args.output_dir)
+
+        # Good practice: save your training arguments together with the trained model
+        torch.save(args, os.path.join(args.output_dir, 'training_args.bin'))
+
+        # Load a trained model and vocabulary that you have fine-tuned
+        model = model_class.from_pretrained(args.output_dir)
+        tokenizer = tokenizer_class.from_pretrained(args.output_dir, do_lower_case=args.do_lower_case)
+        model.to(args.device)
+
+
+    # Evaluation
+    results = {}
+    if args.do_eval and args.local_rank in [-1, 0]:
+        tokenizer = tokenizer_class.from_pretrained(args.output_dir, do_lower_case=args.do_lower_case)
+        checkpoints = [args.output_dir]
+        if args.eval_all_checkpoints:
+            checkpoints = list(os.path.dirname(c) for c in sorted(glob.glob(args.output_dir + '/**/' + WEIGHTS_NAME, recursive=True)))
+            logging.getLogger("transformers.modeling_utils").setLevel(logging.WARN)  # Reduce logging
+        logger.info("Evaluate the following checkpoints: %s", checkpoints)
+        for checkpoint in checkpoints:
+            global_step = checkpoint.split('-')[-1] if len(checkpoints) > 1 else ""
+            prefix = checkpoint.split('/')[-1] if checkpoint.find('checkpoint') != -1 else ""
+            
+            model = model_class.from_pretrained(checkpoint)
+            model.to(args.device)
+            result = evaluate(args, model, tokenizer, prefix=prefix)
+            result = dict((k + '_{}'.format(global_step), v) for k, v in result.items())
+            results.update(result)
+
+    return results
+
+
+if __name__ == "__main__":
+    main()
diff --git a/examples/utils_xnli.py b/examples/utils_xnli.py
new file mode 100644
index 0000000000..f0238f4664
--- /dev/null
+++ b/examples/utils_xnli.py
@@ -0,0 +1,93 @@
+# coding=utf-8
+# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.
+# Copyright (c) 2018, NVIDIA CORPORATION.  All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" XNLI utils (dataset loading and evaluation) """
+
+from __future__ import absolute_import, division, print_function
+
+import logging
+import os
+
+from transformers.data.processors import DataProcessor, InputExample
+from transformers.data.metrics import simple_accuracy
+
+logger = logging.getLogger(__name__)
+
+class XnliProcessor(DataProcessor):
+    """Processor for the XNLI dataset.
+    Adapted from https://github.com/google-research/bert/blob/f39e881b169b9d53bea03d2d341b31707a6c052b/run_classifier.py#L207"""
+
+    def __init__(self, language, train_language = None):
+        self.language = language
+        self.train_language = train_language
+
+    def get_train_examples(self, data_dir):
+        """See base class."""
+        lg = self.language if self.train_language is None else self.train_language
+        lines = self._read_tsv(os.path.join(data_dir, f"XNLI-MT-1.0/multinli/multinli.train.{lg}.tsv"))
+        examples = []
+        for (i, line) in enumerate(lines):
+            if i == 0:
+                continue
+            guid = "%s-%s" % ('train', i)
+            text_a = line[0]
+            text_b = line[1]
+            label = "contradiction" if line[2] == "contradictory" else line[2]
+            assert isinstance(text_a, str) and isinstance(text_b, str) and isinstance(label, str)
+            examples.append(
+                InputExample(guid=guid, text_a=text_a, text_b=text_b, label=label))
+        return examples
+
+    def get_dev_examples(self, data_dir):
+        """See base class."""
+        lines = self._read_tsv(os.path.join(data_dir, "XNLI-1.0/xnli.dev.tsv"))
+        examples = []
+        for (i, line) in enumerate(lines):
+            if i == 0:
+                continue
+            language = line[0]
+            if language != self.language:
+                continue
+            guid = "%s-%s" % ('dev', i)
+            text_a = line[6]
+            text_b = line[7]
+            label = line[1]
+            assert isinstance(text_a, str) and isinstance(text_b, str) and isinstance(label, str)
+            examples.append(
+                InputExample(guid=guid, text_a=text_a, text_b=text_b, label=label))
+        return examples
+
+    def get_labels(self):
+        """See base class."""
+        return ["contradiction", "entailment", "neutral"]
+
+def xnli_compute_metrics(task_name, preds, labels):
+    assert len(preds) == len(labels)
+    if task_name == "xnli":
+        return {"acc": simple_accuracy(preds, labels)}
+    else:
+        raise ValueError(f'{task_name} is not a supported task.')
+
+xnli_processors = {
+    "xnli": XnliProcessor,
+}
+
+xnli_output_modes = {
+    "xnli": "classification",
+}
+
+xnli_tasks_num_labels = {
+    "xnli": 3,
+}

From d52e98ff9af4509bb803641f0c0d81d67ce73cc3 Mon Sep 17 00:00:00 2001
From: VictorSanh <victorsanh@gmail.com>
Date: Tue, 29 Oct 2019 11:51:15 -0400
Subject: [PATCH 138/505] add xnli examples/README.md

---
 examples/README.md | 40 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 40 insertions(+)

diff --git a/examples/README.md b/examples/README.md
index e109a12171..6f8d6bd26e 100644
--- a/examples/README.md
+++ b/examples/README.md
@@ -21,6 +21,7 @@ pip install [--editable] .
 | [SQuAD](#squad) | Using BERT/RoBERTa/XLNet/XLM for question answering, examples with distributed training.                                                                                  |
 | [Multiple Choice](#multiple-choice) | Examples running BERT/XLNet/RoBERTa on the SWAG/RACE/ARC tasks. 
 | [Named Entity Recognition](#named-entity-recognition) | Using BERT for Named Entity Recognition (NER) on the CoNLL 2003 dataset, examples with distributed training.                                                                                  |
+| [XNLI](#xnli) | Examples running BERT/XLM on the XNLI benchmark. |
 | [Abstractive summarization](#abstractive-summarization) | Fine-tuning the library models for abstractive summarization tasks on the CNN/Daily Mail dataset. |
 
 ## TensorFlow 2.0 Bert models on GLUE
@@ -600,3 +601,42 @@ python run_summarization_finetuning.py \
     --do_train \
     --data_path=$DATA_PATH \
 ```
+
+## XNLI
+
+Based on the script [`run_xnli.py`](TODO).
+
+[XNLI](https://www.nyu.edu/projects/bowman/xnli/) is crowd-sourced dataset based on [MultiNLI](http://www.nyu.edu/projects/bowman/multinli/). It is an evaluation benchmark for cross-lingual text representations. Pairs of text are labeled with textual entailment annotations for 15 different languages (including both high-ressource language such as English and low-ressource languages such as Swahili).
+
+#### Fine-tuning on XNLI
+
+This example code fine-tunes mBERT (multi-lingual BERT) on the XNLI dataset. It runs in TODO min
+on a single tesla V100 16GB. The data for XNLI can be downloaded with the following links and should be both saved (and un-zipped) in a 
+`$XNLI_DIR` directory.
+
+* [XNLI 1.0](https://www.nyu.edu/projects/bowman/xnli/XNLI-1.0.zip)
+* [XNLI-MT 1.0](https://www.nyu.edu/projects/bowman/xnli/XNLI-MT-1.0.zip)
+
+```bash
+export XNLI_DIR=/path/to/XNLI
+
+python run_xnli.py \
+  --model_type bert \
+  --model_name_or_path bert-base-multilingual-cased \
+  --language en \
+  --train_language en \
+  --do_train \
+  --do_eval \
+  --data_dir $SQUAD_DIR \
+  --per_gpu_train_batch_size 32 \
+  --learning_rate 5e-5 \
+  --num_train_epochs 2.0 \
+  --max_seq_length 128 \
+  --output_dir /tmp/debug_xnli/
+```
+
+Training with the previously defined hyper-parameters yields the following results:
+
+```bash
+TODO
+```

From c4336ecbbdbb8bddcbfb31e0fcc4d382b430a9a5 Mon Sep 17 00:00:00 2001
From: VictorSanh <victorsanh@gmail.com>
Date: Tue, 29 Oct 2019 12:04:20 -0400
Subject: [PATCH 139/505] xnli - output_mode consistency

---
 examples/run_xnli.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/examples/run_xnli.py b/examples/run_xnli.py
index ee37296832..a9b2e46c13 100644
--- a/examples/run_xnli.py
+++ b/examples/run_xnli.py
@@ -247,8 +247,8 @@ def evaluate(args, model, tokenizer, prefix=""):
         eval_loss = eval_loss / nb_eval_steps
         if args.output_mode == "classification":
             preds = np.argmax(preds, axis=1)
-        elif args.output_mode == "regression":
-            preds = np.squeeze(preds)
+        else:
+            raise ValueError(f'No other `output_mode` for XNLI.')
         result = compute_metrics(eval_task, preds, out_label_ids)
         results.update(result)
 

From 84a0b522cf1b3f69bfbc92eb2309ba5ff652e521 Mon Sep 17 00:00:00 2001
From: VictorSanh <victorsanh@gmail.com>
Date: Tue, 29 Oct 2019 18:53:45 -0400
Subject: [PATCH 140/505] mbert reproducibility results

---
 examples/README.md | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/examples/README.md b/examples/README.md
index 6f8d6bd26e..3e5fd03c45 100644
--- a/examples/README.md
+++ b/examples/README.md
@@ -604,13 +604,13 @@ python run_summarization_finetuning.py \
 
 ## XNLI
 
-Based on the script [`run_xnli.py`](TODO).
+Based on the script [`run_xnli.py`](https://github.com/huggingface/transformers/blob/master/examples/run_xnli.py).
 
 [XNLI](https://www.nyu.edu/projects/bowman/xnli/) is crowd-sourced dataset based on [MultiNLI](http://www.nyu.edu/projects/bowman/multinli/). It is an evaluation benchmark for cross-lingual text representations. Pairs of text are labeled with textual entailment annotations for 15 different languages (including both high-ressource language such as English and low-ressource languages such as Swahili).
 
 #### Fine-tuning on XNLI
 
-This example code fine-tunes mBERT (multi-lingual BERT) on the XNLI dataset. It runs in TODO min
+This example code fine-tunes mBERT (multi-lingual BERT) on the XNLI dataset. It runs in 106 mins
 on a single tesla V100 16GB. The data for XNLI can be downloaded with the following links and should be both saved (and un-zipped) in a 
 `$XNLI_DIR` directory.
 
@@ -623,20 +623,21 @@ export XNLI_DIR=/path/to/XNLI
 python run_xnli.py \
   --model_type bert \
   --model_name_or_path bert-base-multilingual-cased \
-  --language en \
+  --language es \
   --train_language en \
   --do_train \
   --do_eval \
-  --data_dir $SQUAD_DIR \
+  --data_dir $XNLI_DIR \
   --per_gpu_train_batch_size 32 \
   --learning_rate 5e-5 \
   --num_train_epochs 2.0 \
   --max_seq_length 128 \
-  --output_dir /tmp/debug_xnli/
+  --output_dir /tmp/debug_xnli/ \
+  --save_steps -1
 ```
 
-Training with the previously defined hyper-parameters yields the following results:
+Training with the previously defined hyper-parameters yields the following results on the dev set:
 
 ```bash
-TODO
+acc = 0.738152610441767
 ```

From cb7b77a8a2ee52812c4358817a6a586f19687cda Mon Sep 17 00:00:00 2001
From: VictorSanh <victorsanh@gmail.com>
Date: Wed, 30 Oct 2019 18:13:52 -0400
Subject: [PATCH 141/505] fix some typos

---
 transformers/tokenization_xlm.py | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/transformers/tokenization_xlm.py b/transformers/tokenization_xlm.py
index 01f8721d98..ba994dc356 100644
--- a/transformers/tokenization_xlm.py
+++ b/transformers/tokenization_xlm.py
@@ -12,7 +12,7 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
-"""Tokenization classes for OpenAI GPT."""
+"""Tokenization classes for XLM."""
 from __future__ import (absolute_import, division, print_function,
                         unicode_literals)
 
@@ -758,9 +758,9 @@ class XLMTokenizer(PreTrainedTokenizer):
         """
         Build model inputs from a sequence or a pair of sequence for sequence classification tasks
         by concatenating and adding special tokens.
-        A RoBERTa sequence has the following format:
+        A XLM sequence has the following format:
             single sequence: <s> X </s>
-            pair of sequences: <s> A </s></s> B </s>
+            pair of sequences: <s> A </s> B </s>
         """
         if token_ids_1 is None:
             return [self.cls_token_id] + token_ids_0 + [self.sep_token_id]

From 289cf4d2b7cf090942e2e7dc4e6134a81095adf6 Mon Sep 17 00:00:00 2001
From: VictorSanh <victorsanh@gmail.com>
Date: Tue, 5 Nov 2019 10:10:02 -0500
Subject: [PATCH 142/505] change default for XNLI: dev --> test

---
 examples/run_xnli.py   | 6 +++---
 examples/utils_xnli.py | 6 +++---
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/examples/run_xnli.py b/examples/run_xnli.py
index a9b2e46c13..7fbbf7d152 100644
--- a/examples/run_xnli.py
+++ b/examples/run_xnli.py
@@ -270,7 +270,7 @@ def load_and_cache_examples(args, task, tokenizer, evaluate=False):
     output_mode = output_modes[task]
     # Load data features from cache or dataset file
     cached_features_file = os.path.join(args.data_dir, 'cached_{}_{}_{}_{}_{}'.format(
-        'dev' if evaluate else 'train',
+        'test' if evaluate else 'train',
         list(filter(None, args.model_name_or_path.split('/'))).pop(),
         str(args.max_seq_length),
         str(task),
@@ -281,7 +281,7 @@ def load_and_cache_examples(args, task, tokenizer, evaluate=False):
     else:
         logger.info("Creating features from dataset file at %s", args.data_dir)
         label_list = processor.get_labels()
-        examples = processor.get_dev_examples(args.data_dir) if evaluate else processor.get_train_examples(args.data_dir)
+        examples = processor.get_test_examples(args.data_dir) if evaluate else processor.get_train_examples(args.data_dir)
         features = convert_examples_to_features(examples,
                                                 tokenizer,
                                                 label_list=label_list,
@@ -341,7 +341,7 @@ def main():
     parser.add_argument("--do_train", action='store_true',
                         help="Whether to run training.")
     parser.add_argument("--do_eval", action='store_true',
-                        help="Whether to run eval on the dev set.")
+                        help="Whether to run eval on the test set.")
     parser.add_argument("--evaluate_during_training", action='store_true',
                         help="Rul evaluation during training at each logging step.")
     parser.add_argument("--do_lower_case", action='store_true',
diff --git a/examples/utils_xnli.py b/examples/utils_xnli.py
index f0238f4664..482e79a81f 100644
--- a/examples/utils_xnli.py
+++ b/examples/utils_xnli.py
@@ -50,9 +50,9 @@ class XnliProcessor(DataProcessor):
                 InputExample(guid=guid, text_a=text_a, text_b=text_b, label=label))
         return examples
 
-    def get_dev_examples(self, data_dir):
+    def get_test_examples(self, data_dir):
         """See base class."""
-        lines = self._read_tsv(os.path.join(data_dir, "XNLI-1.0/xnli.dev.tsv"))
+        lines = self._read_tsv(os.path.join(data_dir, "XNLI-1.0/xnli.test.tsv"))
         examples = []
         for (i, line) in enumerate(lines):
             if i == 0:
@@ -60,7 +60,7 @@ class XnliProcessor(DataProcessor):
             language = line[0]
             if language != self.language:
                 continue
-            guid = "%s-%s" % ('dev', i)
+            guid = "%s-%s" % ('test', i)
             text_a = line[6]
             text_b = line[7]
             label = line[1]

From d5910b312fbe02a4b08097646e46399dac179084 Mon Sep 17 00:00:00 2001
From: VictorSanh <victorsanh@gmail.com>
Date: Tue, 5 Nov 2019 10:21:25 -0500
Subject: [PATCH 143/505] move xnli processor (and utils) to
 transformers/data/processors

---
 examples/run_xnli.py                                        | 6 +++---
 transformers/__init__.py                                    | 3 ++-
 transformers/data/__init__.py                               | 1 +
 transformers/data/processors/__init__.py                    | 2 +-
 .../utils_xnli.py => transformers/data/processors/xnli.py   | 2 +-
 5 files changed, 8 insertions(+), 6 deletions(-)
 rename examples/utils_xnli.py => transformers/data/processors/xnli.py (97%)

diff --git a/examples/run_xnli.py b/examples/run_xnli.py
index 7fbbf7d152..0c1743e6c4 100644
--- a/examples/run_xnli.py
+++ b/examples/run_xnli.py
@@ -44,9 +44,9 @@ from transformers import (WEIGHTS_NAME,
 
 from transformers import AdamW, WarmupLinearSchedule
 
-from utils_xnli import xnli_compute_metrics as compute_metrics
-from utils_xnli import xnli_output_modes as output_modes
-from utils_xnli import xnli_processors as processors
+from transformers import xnli_compute_metrics as compute_metrics
+from transformers import xnli_output_modes as output_modes
+from transformers import xnli_processors as processors
 
 from transformers import glue_convert_examples_to_features as convert_examples_to_features
 
diff --git a/transformers/__init__.py b/transformers/__init__.py
index dff2479d4b..a133425a9c 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -25,7 +25,8 @@ from .file_utils import (TRANSFORMERS_CACHE, PYTORCH_TRANSFORMERS_CACHE, PYTORCH
 from .data import (is_sklearn_available,
                    InputExample, InputFeatures, DataProcessor,
                    glue_output_modes, glue_convert_examples_to_features,
-                   glue_processors, glue_tasks_num_labels)
+                   glue_processors, glue_tasks_num_labels,
+                   xnli_output_modes, xnli_processors, xnli_tasks_num_labels)
 
 if is_sklearn_available():
     from .data import glue_compute_metrics
diff --git a/transformers/data/__init__.py b/transformers/data/__init__.py
index e910d6da2e..46615608a4 100644
--- a/transformers/data/__init__.py
+++ b/transformers/data/__init__.py
@@ -1,5 +1,6 @@
 from .processors import InputExample, InputFeatures, DataProcessor
 from .processors import glue_output_modes, glue_processors, glue_tasks_num_labels, glue_convert_examples_to_features
+from .processors import xnli_output_modes, xnli_processors, xnli_tasks_num_labels
 
 from .metrics import is_sklearn_available
 if is_sklearn_available():
diff --git a/transformers/data/processors/__init__.py b/transformers/data/processors/__init__.py
index af38c54beb..1c41553ba4 100644
--- a/transformers/data/processors/__init__.py
+++ b/transformers/data/processors/__init__.py
@@ -1,3 +1,3 @@
 from .utils import InputExample, InputFeatures, DataProcessor
 from .glue import glue_output_modes, glue_processors, glue_tasks_num_labels, glue_convert_examples_to_features
-
+from .xnli import xnli_output_modes, xnli_processors, xnli_tasks_num_labels
diff --git a/examples/utils_xnli.py b/transformers/data/processors/xnli.py
similarity index 97%
rename from examples/utils_xnli.py
rename to transformers/data/processors/xnli.py
index 482e79a81f..96fac761a9 100644
--- a/examples/utils_xnli.py
+++ b/transformers/data/processors/xnli.py
@@ -20,7 +20,7 @@ from __future__ import absolute_import, division, print_function
 import logging
 import os
 
-from transformers.data.processors import DataProcessor, InputExample
+from .utils import DataProcessor, InputExample
 from transformers.data.metrics import simple_accuracy
 
 logger = logging.getLogger(__name__)

From d75d49a51de28df48b39280b2f06df78d3fa04bf Mon Sep 17 00:00:00 2001
From: VictorSanh <victorsanh@gmail.com>
Date: Tue, 5 Nov 2019 10:32:20 -0500
Subject: [PATCH 144/505] add XnliProcessor to doc

---
 docs/source/main_classes/processors.rst | 25 ++++++++++++++++++++++++-
 1 file changed, 24 insertions(+), 1 deletion(-)

diff --git a/docs/source/main_classes/processors.rst b/docs/source/main_classes/processors.rst
index a85c126956..a093e621ad 100644
--- a/docs/source/main_classes/processors.rst
+++ b/docs/source/main_classes/processors.rst
@@ -55,4 +55,27 @@ Example usage
 ^^^^^^^^^^^^^^^^^^^^^^^^^
 
 An example using these processors is given in the
-`run_glue.py <https://github.com/huggingface/pytorch-transformers/blob/master/examples/run_glue.py>`__ script.
\ No newline at end of file
+`run_glue.py <https://github.com/huggingface/pytorch-transformers/blob/master/examples/run_glue.py>`__ script.
+
+
+XNLI
+~~~~~~~~~~~~~~~~~~~~~
+
+`The Cross-Lingual NLI Corpus (XNLI) <https://www.nyu.edu/projects/bowman/xnli/>`__ is a benchmark that evaluates
+the quality of cross-lingual text representations. 
+XNLI is crowd-sourced dataset based on `MultiNLI <http://www.nyu.edu/projects/bowman/multinli/>`: pairs of text are labeled with textual entailment 
+annotations for 15 different languages (including both high-ressource language such as English and low-ressource languages such as Swahili).
+
+It was released together with the paper
+`XNLI: Evaluating Cross-lingual Sentence Representations <https://arxiv.org/abs/1809.05053>`__
+
+This library hosts the processor to load the XNLI data:
+    - :class:`~transformers.data.processors.utils.XnliProcessor`
+
+Please note that since the gold labels are available on the test set, evaluation is performed on the test set.
+
+Example usage
+^^^^^^^^^^^^^^^^^^^^^^^^^
+
+An example using these processors is given in the
+`run_xnli.py <https://github.com/huggingface/pytorch-transformers/blob/master/examples/run_xnli.py>`__ script.
\ No newline at end of file

From abd397e95483cc7486bed47fd271ab417e1a288e Mon Sep 17 00:00:00 2001
From: VictorSanh <victorsanh@gmail.com>
Date: Tue, 5 Nov 2019 10:57:24 -0500
Subject: [PATCH 145/505] uniformize w/ the cache_dir update

---
 examples/run_xnli.py | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

diff --git a/examples/run_xnli.py b/examples/run_xnli.py
index 0c1743e6c4..e8457e542b 100644
--- a/examples/run_xnli.py
+++ b/examples/run_xnli.py
@@ -467,9 +467,17 @@ def main():
 
     args.model_type = args.model_type.lower()
     config_class, model_class, tokenizer_class = MODEL_CLASSES[args.model_type]
-    config = config_class.from_pretrained(args.config_name if args.config_name else args.model_name_or_path, num_labels=num_labels, finetuning_task=args.task_name)
-    tokenizer = tokenizer_class.from_pretrained(args.tokenizer_name if args.tokenizer_name else args.model_name_or_path, do_lower_case=args.do_lower_case)
-    model = model_class.from_pretrained(args.model_name_or_path, from_tf=bool('.ckpt' in args.model_name_or_path), config=config)
+    config = config_class.from_pretrained(args.config_name if args.config_name else args.model_name_or_path,
+                                          num_labels=num_labels,
+                                          finetuning_task=args.task_name,
+                                          cache_dir=args.cache_dir if args.cache_dir else None)
+    tokenizer = tokenizer_class.from_pretrained(args.tokenizer_name if args.tokenizer_name else args.model_name_or_path,
+                                                do_lower_case=args.do_lower_case,
+                                                cache_dir=args.cache_dir if args.cache_dir else None)
+    model = model_class.from_pretrained(args.model_name_or_path,
+                                        from_tf=bool('.ckpt' in args.model_name_or_path),
+                                        config=config,
+                                        cache_dir=args.cache_dir if args.cache_dir else None)
 
     if args.local_rank == 0:
         torch.distributed.barrier()  # Make sure only the first process in distributed training will download model & vocab

From 3e7656f7ac369de08a2ebac3cd1d6870e60605bc Mon Sep 17 00:00:00 2001
From: VictorSanh <victorsanh@gmail.com>
Date: Tue, 5 Nov 2019 11:58:53 -0500
Subject: [PATCH 146/505] update readme

---
 examples/README.md | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/examples/README.md b/examples/README.md
index 3e5fd03c45..960b218f11 100644
--- a/examples/README.md
+++ b/examples/README.md
@@ -623,7 +623,7 @@ export XNLI_DIR=/path/to/XNLI
 python run_xnli.py \
   --model_type bert \
   --model_name_or_path bert-base-multilingual-cased \
-  --language es \
+  --language de \
   --train_language en \
   --do_train \
   --do_eval \
@@ -636,8 +636,8 @@ python run_xnli.py \
   --save_steps -1
 ```
 
-Training with the previously defined hyper-parameters yields the following results on the dev set:
+Training with the previously defined hyper-parameters yields the following results on the **test** set:
 
 ```bash
-acc = 0.738152610441767
+acc = 0.7093812375249501
 ```

From 73fe2e7385af4c6062f366825af570d44cd22fd8 Mon Sep 17 00:00:00 2001
From: VictorSanh <victorsanh@gmail.com>
Date: Tue, 5 Nov 2019 12:51:43 -0500
Subject: [PATCH 147/505] remove fstrings

---
 examples/run_xnli.py                 | 4 ++--
 transformers/data/processors/xnli.py | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/examples/run_xnli.py b/examples/run_xnli.py
index e8457e542b..952128f4ad 100644
--- a/examples/run_xnli.py
+++ b/examples/run_xnli.py
@@ -248,7 +248,7 @@ def evaluate(args, model, tokenizer, prefix=""):
         if args.output_mode == "classification":
             preds = np.argmax(preds, axis=1)
         else:
-            raise ValueError(f'No other `output_mode` for XNLI.')
+            raise ValueError('No other `output_mode` for XNLI.')
         result = compute_metrics(eval_task, preds, out_label_ids)
         results.update(result)
 
@@ -305,7 +305,7 @@ def load_and_cache_examples(args, task, tokenizer, evaluate=False):
     if output_mode == "classification":
         all_labels = torch.tensor([f.label for f in features], dtype=torch.long)
     else:
-        raise ValueError(f'No other `output_mode` for XNLI.')
+        raise ValueError('No other `output_mode` for XNLI.')
 
     dataset = TensorDataset(all_input_ids, all_attention_mask, all_token_type_ids, all_labels)
     return dataset
diff --git a/transformers/data/processors/xnli.py b/transformers/data/processors/xnli.py
index 96fac761a9..a4807dd901 100644
--- a/transformers/data/processors/xnli.py
+++ b/transformers/data/processors/xnli.py
@@ -36,7 +36,7 @@ class XnliProcessor(DataProcessor):
     def get_train_examples(self, data_dir):
         """See base class."""
         lg = self.language if self.train_language is None else self.train_language
-        lines = self._read_tsv(os.path.join(data_dir, f"XNLI-MT-1.0/multinli/multinli.train.{lg}.tsv"))
+        lines = self._read_tsv(os.path.join(data_dir, "XNLI-MT-1.0/multinli/multinli.train.{lg}.tsv".format(lg)))
         examples = []
         for (i, line) in enumerate(lines):
             if i == 0:
@@ -78,7 +78,7 @@ def xnli_compute_metrics(task_name, preds, labels):
     if task_name == "xnli":
         return {"acc": simple_accuracy(preds, labels)}
     else:
-        raise ValueError(f'{task_name} is not a supported task.')
+        raise ValueError('{} is not a supported task.'.format(task_name))
 
 xnli_processors = {
     "xnli": XnliProcessor,

From bcd8dc6b48a335b899f00b59f016f740c5230d41 Mon Sep 17 00:00:00 2001
From: VictorSanh <victorsanh@gmail.com>
Date: Tue, 5 Nov 2019 12:53:08 -0500
Subject: [PATCH 148/505] move xnli_compute_metrics to data/metrics

---
 transformers/__init__.py              | 2 +-
 transformers/data/__init__.py         | 2 +-
 transformers/data/metrics/__init__.py | 8 ++++++++
 transformers/data/processors/xnli.py  | 7 -------
 4 files changed, 10 insertions(+), 9 deletions(-)

diff --git a/transformers/__init__.py b/transformers/__init__.py
index a133425a9c..b29ad38e73 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -29,7 +29,7 @@ from .data import (is_sklearn_available,
                    xnli_output_modes, xnli_processors, xnli_tasks_num_labels)
 
 if is_sklearn_available():
-    from .data import glue_compute_metrics
+    from .data import glue_compute_metrics, xnli_compute_metrics
 
 # Tokenizers
 from .tokenization_utils import (PreTrainedTokenizer)
diff --git a/transformers/data/__init__.py b/transformers/data/__init__.py
index 46615608a4..b811a35807 100644
--- a/transformers/data/__init__.py
+++ b/transformers/data/__init__.py
@@ -4,4 +4,4 @@ from .processors import xnli_output_modes, xnli_processors, xnli_tasks_num_label
 
 from .metrics import is_sklearn_available
 if is_sklearn_available():
-    from .metrics import glue_compute_metrics
+    from .metrics import glue_compute_metrics, xnli_compute_metrics
diff --git a/transformers/data/metrics/__init__.py b/transformers/data/metrics/__init__.py
index c9ebaac38d..5a46eb05d3 100644
--- a/transformers/data/metrics/__init__.py
+++ b/transformers/data/metrics/__init__.py
@@ -81,3 +81,11 @@ if _has_sklearn:
             return {"acc": simple_accuracy(preds, labels)}
         else:
             raise KeyError(task_name)
+
+
+    def xnli_compute_metrics(task_name, preds, labels):
+        assert len(preds) == len(labels)
+        if task_name == "xnli":
+            return {"acc": simple_accuracy(preds, labels)}
+        else:
+            raise KeyError(task_name)
diff --git a/transformers/data/processors/xnli.py b/transformers/data/processors/xnli.py
index a4807dd901..ce582f31a6 100644
--- a/transformers/data/processors/xnli.py
+++ b/transformers/data/processors/xnli.py
@@ -73,13 +73,6 @@ class XnliProcessor(DataProcessor):
         """See base class."""
         return ["contradiction", "entailment", "neutral"]
 
-def xnli_compute_metrics(task_name, preds, labels):
-    assert len(preds) == len(labels)
-    if task_name == "xnli":
-        return {"acc": simple_accuracy(preds, labels)}
-    else:
-        raise ValueError('{} is not a supported task.'.format(task_name))
-
 xnli_processors = {
     "xnli": XnliProcessor,
 }

From d47402263964e30ee17cbc06811622bf2df50d6d Mon Sep 17 00:00:00 2001
From: VictorSanh <victorsanh@gmail.com>
Date: Tue, 5 Nov 2019 12:56:03 -0500
Subject: [PATCH 149/505] cleaning simple_accuracy since not used anymore

---
 transformers/data/processors/xnli.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/transformers/data/processors/xnli.py b/transformers/data/processors/xnli.py
index ce582f31a6..efbe3762f6 100644
--- a/transformers/data/processors/xnli.py
+++ b/transformers/data/processors/xnli.py
@@ -21,7 +21,6 @@ import logging
 import os
 
 from .utils import DataProcessor, InputExample
-from transformers.data.metrics import simple_accuracy
 
 logger = logging.getLogger(__name__)
 

From 07ab8d7af6041f0d3562badcb3d6f173062329a1 Mon Sep 17 00:00:00 2001
From: VictorSanh <victorsanh@gmail.com>
Date: Tue, 5 Nov 2019 17:33:14 -0500
Subject: [PATCH 150/505] fix bug

---
 transformers/data/processors/xnli.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/data/processors/xnli.py b/transformers/data/processors/xnli.py
index efbe3762f6..958bdf62f9 100644
--- a/transformers/data/processors/xnli.py
+++ b/transformers/data/processors/xnli.py
@@ -35,7 +35,7 @@ class XnliProcessor(DataProcessor):
     def get_train_examples(self, data_dir):
         """See base class."""
         lg = self.language if self.train_language is None else self.train_language
-        lines = self._read_tsv(os.path.join(data_dir, "XNLI-MT-1.0/multinli/multinli.train.{lg}.tsv".format(lg)))
+        lines = self._read_tsv(os.path.join(data_dir, "XNLI-MT-1.0/multinli/multinli.train.{}.tsv".format(lg)))
         examples = []
         for (i, line) in enumerate(lines):
             if i == 0:

From d5478b939d64db58972e46b7218c765c918b76ac Mon Sep 17 00:00:00 2001
From: VictorSanh <victorsanh@gmail.com>
Date: Mon, 25 Nov 2019 19:40:48 +0000
Subject: [PATCH 151/505] add distilbert + update run_xnli wrt run_glue

---
 examples/run_xnli.py | 51 +++++++++++---------------------------------
 1 file changed, 12 insertions(+), 39 deletions(-)

diff --git a/examples/run_xnli.py b/examples/run_xnli.py
index 952128f4ad..a3bc0d4604 100644
--- a/examples/run_xnli.py
+++ b/examples/run_xnli.py
@@ -13,7 +13,7 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
-""" Finetuning multi-lingual models on XNLI (Bert, XLM).
+""" Finetuning multi-lingual models on XNLI (Bert, DistilBERT, XLM).
     Adapted from `examples/run_glue.py`"""
 
 from __future__ import absolute_import, division, print_function
@@ -42,7 +42,7 @@ from transformers import (WEIGHTS_NAME,
                           XLMConfig, XLMForSequenceClassification, XLMTokenizer,
                           DistilBertConfig, DistilBertForSequenceClassification, DistilBertTokenizer)
 
-from transformers import AdamW, WarmupLinearSchedule
+from transformers import AdamW, get_linear_schedule_with_warmup
 
 from transformers import xnli_compute_metrics as compute_metrics
 from transformers import xnli_output_modes as output_modes
@@ -52,12 +52,12 @@ from transformers import glue_convert_examples_to_features as convert_examples_t
 
 logger = logging.getLogger(__name__)
 
-ALL_MODELS = sum((tuple(conf.pretrained_config_archive_map.keys()) for conf in (BertConfig, XLMConfig)), ())
+ALL_MODELS = sum((tuple(conf.pretrained_config_archive_map.keys()) for conf in (BertConfig, DistilBertConfig, XLMConfig)), ())
 
 MODEL_CLASSES = {
     'bert': (BertConfig, BertForSequenceClassification, BertTokenizer),
     'xlm': (XLMConfig, XLMForSequenceClassification, XLMTokenizer),
-    # 'distilbert': (DistilBertConfig, DistilBertForSequenceClassification, DistilBertTokenizer)
+    'distilbert': (DistilBertConfig, DistilBertForSequenceClassification, DistilBertTokenizer)
 }
 
 
@@ -91,7 +91,7 @@ def train(args, train_dataset, model, tokenizer):
         {'params': [p for n, p in model.named_parameters() if any(nd in n for nd in no_decay)], 'weight_decay': 0.0}
         ]
     optimizer = AdamW(optimizer_grouped_parameters, lr=args.learning_rate, eps=args.adam_epsilon)
-    scheduler = WarmupLinearSchedule(optimizer, warmup_steps=args.warmup_steps, t_total=t_total)
+    scheduler = get_linear_schedule_with_warmup(optimizer, num_warmup_steps=args.warmup_steps, num_training_steps=t_total)
     if args.fp16:
         try:
             from apex import amp
@@ -149,7 +149,7 @@ def train(args, train_dataset, model, tokenizer):
                 loss.backward()
 
             tr_loss += loss.item()
-            if (step + 1) % args.gradient_accumulation_steps == 0 and not args.tpu:
+            if (step + 1) % args.gradient_accumulation_steps == 0:
                 if args.fp16:
                     torch.nn.utils.clip_grad_norm_(amp.master_params(optimizer), args.max_grad_norm)
                 else:
@@ -180,11 +180,6 @@ def train(args, train_dataset, model, tokenizer):
                     torch.save(args, os.path.join(output_dir, 'training_args.bin'))
                     logger.info("Saving model checkpoint to %s", output_dir)
 
-            if args.tpu:
-                args.xla_model.optimizer_step(optimizer, barrier=True)
-                model.zero_grad()
-                global_step += 1
-
             if args.max_steps > 0 and global_step > args.max_steps:
                 epoch_iterator.close()
                 break
@@ -214,6 +209,10 @@ def evaluate(args, model, tokenizer, prefix=""):
         eval_sampler = SequentialSampler(eval_dataset) if args.local_rank == -1 else DistributedSampler(eval_dataset)
         eval_dataloader = DataLoader(eval_dataset, sampler=eval_sampler, batch_size=args.eval_batch_size)
 
+        # multi-gpu eval
+        if args.n_gpu > 1:
+            model = torch.nn.DataParallel(model)
+
         # Eval!
         logger.info("***** Running evaluation {} *****".format(prefix))
         logger.info("  Num examples = %d", len(eval_dataset))
@@ -383,15 +382,6 @@ def main():
     parser.add_argument('--seed', type=int, default=42,
                         help="random seed for initialization")
 
-    parser.add_argument('--tpu', action='store_true',
-                        help="Whether to run on the TPU defined in the environment variables")
-    parser.add_argument('--tpu_ip_address', type=str, default='',
-                        help="TPU IP address if none are set in the environment variables")
-    parser.add_argument('--tpu_name', type=str, default='',
-                        help="TPU name if none are set in the environment variables")
-    parser.add_argument('--xrt_tpu_config', type=str, default='',
-                        help="XRT TPU config if none are set in the environment variables")
-
     parser.add_argument('--fp16', action='store_true',
                         help="Whether to use 16-bit (mixed) precision (through NVIDIA apex) instead of 32-bit")
     parser.add_argument('--fp16_opt_level', type=str, default='O1',
@@ -425,23 +415,6 @@ def main():
         args.n_gpu = 1
     args.device = device
 
-    if args.tpu:
-        if args.tpu_ip_address:
-            os.environ["TPU_IP_ADDRESS"] = args.tpu_ip_address
-        if args.tpu_name:
-            os.environ["TPU_NAME"] = args.tpu_name
-        if args.xrt_tpu_config:
-            os.environ["XRT_TPU_CONFIG"] = args.xrt_tpu_config
-
-        assert "TPU_IP_ADDRESS" in os.environ
-        assert "TPU_NAME" in os.environ
-        assert "XRT_TPU_CONFIG" in os.environ
-
-        import torch_xla
-        import torch_xla.core.xla_model as xm
-        args.device = xm.xla_device()
-        args.xla_model = xm
-
     # Setup logging
     logging.basicConfig(format = '%(asctime)s - %(levelname)s - %(name)s -   %(message)s',
                         datefmt = '%m/%d/%Y %H:%M:%S',
@@ -495,7 +468,7 @@ def main():
 
 
     # Saving best-practices: if you use defaults names for the model, you can reload it using from_pretrained()
-    if args.do_train and (args.local_rank == -1 or torch.distributed.get_rank() == 0) and not args.tpu:
+    if args.do_train and (args.local_rank == -1 or torch.distributed.get_rank() == 0):
         # Create output directory if needed
         if not os.path.exists(args.output_dir) and args.local_rank in [-1, 0]:
             os.makedirs(args.output_dir)
@@ -512,7 +485,7 @@ def main():
 
         # Load a trained model and vocabulary that you have fine-tuned
         model = model_class.from_pretrained(args.output_dir)
-        tokenizer = tokenizer_class.from_pretrained(args.output_dir, do_lower_case=args.do_lower_case)
+        tokenizer = tokenizer_class.from_pretrained(args.output_dir)
         model.to(args.device)
 
 

From 10bd1ddb39235b2f58594e48867595e7d38cd619 Mon Sep 17 00:00:00 2001
From: VictorSanh <victorsanh@gmail.com>
Date: Mon, 25 Nov 2019 19:41:00 +0000
Subject: [PATCH 152/505] soft launch distilbert multilingual

---
 transformers/configuration_distilbert.py | 3 ++-
 transformers/modeling_distilbert.py      | 3 ++-
 transformers/tokenization_distilbert.py  | 2 ++
 3 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/transformers/configuration_distilbert.py b/transformers/configuration_distilbert.py
index 2a8a149acf..f929a9bc39 100644
--- a/transformers/configuration_distilbert.py
+++ b/transformers/configuration_distilbert.py
@@ -27,7 +27,8 @@ logger = logging.getLogger(__name__)
 
 DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP = {
     'distilbert-base-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-config.json",
-    'distilbert-base-uncased-distilled-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-distilled-squad-config.json"
+    'distilbert-base-uncased-distilled-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-distilled-squad-config.json",
+    'distilbert-base-multilingual-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-multilingual-cased-config.json",
 }
 
 
diff --git a/transformers/modeling_distilbert.py b/transformers/modeling_distilbert.py
index d30f493c69..62c623ff6c 100644
--- a/transformers/modeling_distilbert.py
+++ b/transformers/modeling_distilbert.py
@@ -42,7 +42,8 @@ logger = logging.getLogger(__name__)
 
 DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
     'distilbert-base-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-pytorch_model.bin",
-    'distilbert-base-uncased-distilled-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-distilled-squad-pytorch_model.bin"
+    'distilbert-base-uncased-distilled-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-distilled-squad-pytorch_model.bin",
+    'distilbert-base-multilingual-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-multilingual-cased-pytorch_model.bin",
 }
 
 
diff --git a/transformers/tokenization_distilbert.py b/transformers/tokenization_distilbert.py
index dfa02926d8..832f0c3d0b 100644
--- a/transformers/tokenization_distilbert.py
+++ b/transformers/tokenization_distilbert.py
@@ -33,12 +33,14 @@ PRETRAINED_VOCAB_FILES_MAP = {
     {
         'distilbert-base-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-uncased-vocab.txt",
         'distilbert-base-uncased-distilled-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-uncased-vocab.txt",
+        'distilbert-base-multilingual-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-multilingual-cased-vocab.txt",
     }
 }
 
 PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
     'distilbert-base-uncased': 512,
     'distilbert-base-uncased-distilled-squad': 512,
+    'distilbert-base-multilingual-cased': 512,
 }
 
 

From 3c28a2daac43386dbc63b0bd014a966f22888850 Mon Sep 17 00:00:00 2001
From: Yao Lu <95luyao@gmail.com>
Date: Wed, 27 Nov 2019 11:45:22 -0500
Subject: [PATCH 153/505] add add_special_tokens=True for input examples

---
 transformers/modeling_bert.py | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/transformers/modeling_bert.py b/transformers/modeling_bert.py
index 81d92d8f1b..34bb8f89ba 100644
--- a/transformers/modeling_bert.py
+++ b/transformers/modeling_bert.py
@@ -597,7 +597,7 @@ class BertModel(BertPreTrainedModel):
 
         tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
         model = BertModel.from_pretrained('bert-base-uncased')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
@@ -760,7 +760,7 @@ class BertForPreTraining(BertPreTrainedModel):
 
         tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
         model = BertForPreTraining.from_pretrained('bert-base-uncased')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids)
         prediction_scores, seq_relationship_scores = outputs[:2]
 
@@ -836,7 +836,7 @@ class BertForMaskedLM(BertPreTrainedModel):
 
         tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
         model = BertForMaskedLM.from_pretrained('bert-base-uncased')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids, masked_lm_labels=input_ids)
         loss, prediction_scores = outputs[:2]
 
@@ -919,7 +919,7 @@ class BertForNextSentencePrediction(BertPreTrainedModel):
 
         tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
         model = BertForNextSentencePrediction.from_pretrained('bert-base-uncased')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids)
         seq_relationship_scores = outputs[0]
 
@@ -984,7 +984,7 @@ class BertForSequenceClassification(BertPreTrainedModel):
 
         tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
         model = BertForSequenceClassification.from_pretrained('bert-base-uncased')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         labels = torch.tensor([1]).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids, labels=labels)
         loss, logits = outputs[:2]
@@ -1060,7 +1060,7 @@ class BertForMultipleChoice(BertPreTrainedModel):
         tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
         model = BertForMultipleChoice.from_pretrained('bert-base-uncased')
         choices = ["Hello, my dog is cute", "Hello, my cat is amazing"]
-        input_ids = torch.tensor([tokenizer.encode(s) for s in choices]).unsqueeze(0)  # Batch size 1, 2 choices
+        input_ids = torch.tensor([tokenizer.encode(s, add_special_tokens=True) for s in choices]).unsqueeze(0)  # Batch size 1, 2 choices
         labels = torch.tensor(1).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids, labels=labels)
         loss, classification_scores = outputs[:2]
@@ -1134,7 +1134,7 @@ class BertForTokenClassification(BertPreTrainedModel):
 
         tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
         model = BertForTokenClassification.from_pretrained('bert-base-uncased')
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
         labels = torch.tensor([1] * input_ids.size(1)).unsqueeze(0)  # Batch size 1
         outputs = model(input_ids, labels=labels)
         loss, scores = outputs[:2]

From 8da47b078d92bee2de3e5fb50a37483d8cb02f13 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Wed, 27 Nov 2019 23:11:37 +0100
Subject: [PATCH 154/505] fix merge tests

---
 transformers/modeling_ctrl.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/modeling_ctrl.py b/transformers/modeling_ctrl.py
index 849487655d..3a252941ac 100644
--- a/transformers/modeling_ctrl.py
+++ b/transformers/modeling_ctrl.py
@@ -373,7 +373,7 @@ class CTRLModel(CTRLPreTrainedModel):
         if inputs_embeds is None:
             inputs_embeds = self.w(input_ids)
         # inputs_embeds = embedded.unsqueeze(0) if len(input_ids.shape)<2 else embedded
-        seq_len = input_shape.shape[-1]
+        seq_len = input_shape[-1]
         mask = torch.triu(torch.ones(seq_len + past_length, seq_len + past_length), 1).to(inputs_embeds.device)
 
         inputs_embeds *= np.sqrt(self.d_model_size)

From bd41e8292a4bd7db10eb036112019d93c50adcf5 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Thu, 28 Nov 2019 16:03:56 -0500
Subject: [PATCH 155/505] Cleanup & Evaluation now works

---
 examples/run_squad.py                 | 44 +++++++++++----------------
 transformers/data/processors/squad.py | 14 ++-------
 2 files changed, 20 insertions(+), 38 deletions(-)

diff --git a/examples/run_squad.py b/examples/run_squad.py
index 634b566a46..545c3ad55a 100644
--- a/examples/run_squad.py
+++ b/examples/run_squad.py
@@ -16,7 +16,7 @@
 """ Finetuning the library models for question-answering on SQuAD (DistilBERT, Bert, XLM, XLNet)."""
 
 from __future__ import absolute_import, division, print_function
-from transformers.data.processors.squad import SquadV1Processor
+from transformers.data.processors.squad import SquadV1Processor, SquadV2Processor
 
 import argparse
 import logging
@@ -45,9 +45,9 @@ from transformers import (WEIGHTS_NAME, BertConfig,
                                   XLNetTokenizer,
                                   DistilBertConfig, DistilBertForQuestionAnswering, DistilBertTokenizer)
 
-from transformers import AdamW, get_linear_schedule_with_warmup, squad_convert_examples_to_features, read_squad_examples as sread_squad_examples
+from transformers import AdamW, get_linear_schedule_with_warmup, squad_convert_examples_to_features
 
-from utils_squad import (RawResult, write_predictions,
+from utils_squad import (convert_examples_to_features as old_convert, read_squad_examples as old_read, RawResult, write_predictions,
                          RawResultExtended, write_predictions_extended)
 
 # The follwing import is the official SQuAD evaluation script (2.0).
@@ -304,28 +304,20 @@ def load_and_cache_examples(args, tokenizer, evaluate=False, output_examples=Fal
         features = torch.load(cached_features_file)
     else:
         logger.info("Creating features from dataset file at %s", input_file)
-        examples = read_squad_examples(input_file=input_file,
-                                                is_training=not evaluate,
-                                                version_2_with_negative=args.version_2_with_negative)
-        keep_n_examples = 1000
-        processor = SquadV1Processor()
-        values = processor.get_dev_examples("examples/squad")
-        examples = values[:keep_n_examples]
-        features = squad_convert_examples_to_features(examples=exampless,
-                                                tokenizer=tokenizer,
-                                                max_seq_length=args.max_seq_length,
-                                                doc_stride=args.doc_stride,
-                                                max_query_length=args.max_query_length,
-                                                is_training=not evaluate,
-                                                cls_token_segment_id=2 if args.model_type in ['xlnet'] else 0,
-                                                pad_token_segment_id=3 if args.model_type in ['xlnet'] else 0,
-                                                cls_token_at_end=True if args.model_type in ['xlnet'] else False,
-                                                sequence_a_is_doc=True if args.model_type in ['xlnet'] else False)
-        print("DONE")
 
-        import sys
-        sys.exit()
-        
+        processor = SquadV2Processor()
+        examples = processor.get_dev_examples("examples/squad") if evaluate else processor.get_train_examples("examples/squad")
+        features = squad_convert_examples_to_features(
+            examples=examples,
+            tokenizer=tokenizer,
+            max_seq_length=args.max_seq_length,
+            doc_stride=args.doc_stride,
+            max_query_length=args.max_query_length,
+            is_training=not evaluate,
+            sequence_a_is_doc=True if args.model_type in ['xlnet'] else False
+        )
+
+
         if args.local_rank in [-1, 0]:
             logger.info("Saving features into cached file %s", cached_features_file)
             torch.save(features, cached_features_file)
@@ -335,8 +327,8 @@ def load_and_cache_examples(args, tokenizer, evaluate=False, output_examples=Fal
 
     # Convert to Tensors and build dataset
     all_input_ids = torch.tensor([f.input_ids for f in features], dtype=torch.long)
-    all_input_mask = torch.tensor([f.input_mask for f in features], dtype=torch.long)
-    all_segment_ids = torch.tensor([f.segment_ids for f in features], dtype=torch.long)
+    all_input_mask = torch.tensor([f.attention_mask for f in features], dtype=torch.long)
+    all_segment_ids = torch.tensor([f.token_type_ids for f in features], dtype=torch.long)
     all_cls_index = torch.tensor([f.cls_index for f in features], dtype=torch.long)
     all_p_mask = torch.tensor([f.p_mask for f in features], dtype=torch.float)
     if evaluate:
diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index 39ee00ae56..3d5a3eca80 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -74,26 +74,16 @@ def _is_whitespace(c):
 
 def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
                                        doc_stride, max_query_length, is_training,
-                                       cls_token_at_end=True,
-                                       cls_token='[CLS]', sep_token='[SEP]', pad_token=0,
-                                       sequence_a_segment_id=0, sequence_b_segment_id=1,
-                                       cls_token_segment_id=0, pad_token_segment_id=0,
-                                       mask_padding_with_zero=True,
                                        sequence_a_is_doc=False):
     """Loads a data file into a list of `InputBatch`s."""
 
-    cls_token = tokenizer.cls_token
-    sep_token = tokenizer.sep_token
-
     # Defining helper methods    
     unique_id = 1000000000
 
     features = []
-    new_features = []
     for (example_index, example) in enumerate(tqdm(examples)):
         if is_training and not example.is_impossible:
             # Get start and end position
-            answer_length = len(example.answer_text)
             start_position = example.start_position
             end_position = example.end_position
 
@@ -227,7 +217,7 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
                     end_position = tok_end_position - doc_start + doc_offset
 
 
-            new_features.append(NewSquadFeatures(
+            features.append(NewSquadFeatures(
                 span['input_ids'],
                 span['attention_mask'],
                 span['token_type_ids'],
@@ -247,7 +237,7 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
 
             unique_id += 1
 
-    return new_features
+    return features
 
 
 class SquadProcessor(DataProcessor):

From f671997ef74199823db83ed7b43340764888e129 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Thu, 28 Nov 2019 17:17:20 -0500
Subject: [PATCH 156/505] Interface with TFDS

---
 transformers/data/processors/squad.py | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index 3d5a3eca80..52c2c28add 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -246,16 +246,24 @@ class SquadProcessor(DataProcessor):
     dev_file = None
 
     def get_example_from_tensor_dict(self, tensor_dict):
-        """See base class."""
         return NewSquadExample(
-            tensor_dict['id'].numpy(),
+            tensor_dict['id'].numpy().decode("utf-8"),
             tensor_dict['question'].numpy().decode('utf-8'),
             tensor_dict['context'].numpy().decode('utf-8'),
-            tensor_dict['answers']['text'].numpy().decode('utf-8'),
-            tensor_dict['answers']['answers_start'].numpy().decode('utf-8'),
+            tensor_dict['answers']['text'][0].numpy().decode('utf-8'),
+            tensor_dict['answers']['answer_start'][0].numpy(),
             tensor_dict['title'].numpy().decode('utf-8')
         )
 
+    def get_examples_from_dataset(self, dataset):
+        """See base class."""
+
+        examples = []
+        for tensor_dict in tqdm(dataset):
+            examples.append(self.get_example_from_tensor_dict(tensor_dict)) 
+
+        return examples
+
     def get_train_examples(self, data_dir, only_first=None):
         """See base class."""
         if self.train_file is None:

From 0b84b9fd8a728ca46e4109aa38a11b25f87a09bf Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Thu, 28 Nov 2019 17:38:52 -0500
Subject: [PATCH 157/505] Add processors to __init__

---
 transformers/__init__.py                 | 2 +-
 transformers/data/__init__.py            | 2 +-
 transformers/data/processors/__init__.py | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/transformers/__init__.py b/transformers/__init__.py
index f3f81f1dbe..aefa3f1921 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -27,7 +27,7 @@ from .data import (is_sklearn_available,
                    glue_output_modes, glue_convert_examples_to_features,
                    glue_processors, glue_tasks_num_labels,
                    squad_convert_examples_to_features, SquadFeatures, 
-                   SquadExample)
+                   SquadExample, SquadV1Processor, SquadV2Processor)
 
 if is_sklearn_available():
     from .data import glue_compute_metrics
diff --git a/transformers/data/__init__.py b/transformers/data/__init__.py
index b351bf625e..ea3a4e9fbb 100644
--- a/transformers/data/__init__.py
+++ b/transformers/data/__init__.py
@@ -1,6 +1,6 @@
 from .processors import InputExample, InputFeatures, DataProcessor, SquadFeatures
 from .processors import glue_output_modes, glue_processors, glue_tasks_num_labels, glue_convert_examples_to_features
-from .processors import squad_convert_examples_to_features, SquadExample
+from .processors import squad_convert_examples_to_features, SquadExample, SquadV1Processor, SquadV2Processor
 
 from .metrics import is_sklearn_available
 if is_sklearn_available():
diff --git a/transformers/data/processors/__init__.py b/transformers/data/processors/__init__.py
index 1e52776629..2470e7a06d 100644
--- a/transformers/data/processors/__init__.py
+++ b/transformers/data/processors/__init__.py
@@ -1,4 +1,4 @@
 from .utils import InputExample, InputFeatures, DataProcessor
 from .glue import glue_output_modes, glue_processors, glue_tasks_num_labels, glue_convert_examples_to_features
-from .squad import squad_convert_examples_to_features, SquadFeatures, SquadExample
+from .squad import squad_convert_examples_to_features, SquadFeatures, SquadExample, SquadV1Processor, SquadV2Processor
 

From 1e9ac5a7cfeb48ff6a1cf20e07941fc8c59b391d Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Thu, 28 Nov 2019 17:43:47 -0500
Subject: [PATCH 158/505] New -> normal

---
 transformers/data/processors/squad.py | 106 ++------------------------
 1 file changed, 5 insertions(+), 101 deletions(-)

diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index 52c2c28add..f414d41925 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -217,7 +217,7 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
                     end_position = tok_end_position - doc_start + doc_offset
 
 
-            features.append(NewSquadFeatures(
+            features.append(SquadFeatures(
                 span['input_ids'],
                 span['attention_mask'],
                 span['token_type_ids'],
@@ -246,7 +246,7 @@ class SquadProcessor(DataProcessor):
     dev_file = None
 
     def get_example_from_tensor_dict(self, tensor_dict):
-        return NewSquadExample(
+        return SquadExample(
             tensor_dict['id'].numpy().decode("utf-8"),
             tensor_dict['question'].numpy().decode('utf-8'),
             tensor_dict['context'].numpy().decode('utf-8'),
@@ -314,7 +314,7 @@ class SquadProcessor(DataProcessor):
                         answer_text = answer['text']
                         start_position_character = answer['answer_start']
 
-                    example = NewSquadExample(
+                    example = SquadExample(
                         qas_id=qas_id,
                         question_text=question_text,
                         context_text=context_text,
@@ -340,7 +340,7 @@ class SquadV2Processor(SquadProcessor):
     dev_file = "dev-v2.0.json"
     
 
-class NewSquadExample(object):
+class SquadExample(object):
     """
     A single training/test example for the Squad dataset, as loaded from disk.
     """
@@ -387,7 +387,7 @@ class NewSquadExample(object):
             self.end_position = char_to_word_offset[start_position_character + len(answer_text) - 1]
 
 
-class NewSquadFeatures(object):
+class SquadFeatures(object):
     """
     Single squad example features to be fed to a model.
     Those features are model-specific.
@@ -425,99 +425,3 @@ class NewSquadFeatures(object):
 
         self.start_position = start_position
         self.end_position = end_position
-
-class SquadExample(object):
-    """
-    A single training/test example for the Squad dataset.
-    For examples without an answer, the start and end position are -1.
-    """
-
-    def __init__(self,
-                 qas_id,
-                 question_text,
-                 doc_tokens,
-                 orig_answer_text=None,
-                 start_position=None,
-                 end_position=None,
-                 is_impossible=None):
-        self.qas_id = qas_id
-        self.question_text = question_text
-        self.doc_tokens = doc_tokens
-        self.orig_answer_text = orig_answer_text
-        self.start_position = start_position
-        self.end_position = end_position
-        self.is_impossible = is_impossible
-
-    def __str__(self):
-        return self.__repr__()
-
-    def __repr__(self):
-        s = ""
-        s += "qas_id: %s" % (self.qas_id)
-        s += ", question_text: %s" % (
-            self.question_text)
-        s += ", doc_tokens: [%s]" % (" ".join(self.doc_tokens))
-        if self.start_position:
-            s += ", start_position: %d" % (self.start_position)
-        if self.end_position:
-            s += ", end_position: %d" % (self.end_position)
-        if self.is_impossible:
-            s += ", is_impossible: %r" % (self.is_impossible)
-        return s
-
-
-class SquadFeatures(object):
-    """A single set of features of data."""
-
-    def __init__(self,
-                 unique_id,
-                 example_index,
-                 doc_span_index,
-                 tokens,
-                 token_to_orig_map,
-                 token_is_max_context,
-                 input_ids,
-                 input_mask,
-                 segment_ids,
-                 cls_index,
-                 p_mask,
-                 paragraph_len,
-                 start_position=None,
-                 end_position=None,
-                 is_impossible=None):
-        self.unique_id = unique_id
-        self.example_index = example_index
-        self.doc_span_index = doc_span_index
-        self.tokens = tokens
-        self.token_to_orig_map = token_to_orig_map
-        self.token_is_max_context = token_is_max_context
-        self.input_ids = input_ids
-        self.input_mask = input_mask
-        self.segment_ids = segment_ids
-        self.cls_index = cls_index
-        self.p_mask = p_mask
-        self.paragraph_len = paragraph_len
-        self.start_position = start_position
-        self.end_position = end_position
-        self.is_impossible = is_impossible
-
-    def __eq__(self, other):
-        print(self.example_index == other.example_index)
-        print(self.input_ids == other.input_ids)
-        print(self.input_mask == other.attention_mask)
-        print(self.p_mask == other.p_mask)
-        print(self.paragraph_len == other.paragraph_len)
-        print(self.segment_ids == other.token_type_ids)
-        print(self.token_is_max_context == other.token_is_max_context)
-        print(self.token_to_orig_map == other.token_to_orig_map)
-        print(self.tokens == other.tokens)
-
-        return self.example_index == other.example_index and \
-                self.input_ids == other.input_ids and \
-                self.input_mask == other.attention_mask and \
-                self.p_mask == other.p_mask and \
-                self.paragraph_len == other.paragraph_len and \
-                self.segment_ids == other.token_type_ids and \
-                self.token_is_max_context == other.token_is_max_context and \
-                self.token_to_orig_map == other.token_to_orig_map and \
-                self.tokens == other.tokens
\ No newline at end of file

From 41aa0e80039d3148c55f4fe967247d4f7bbbfec5 Mon Sep 17 00:00:00 2001
From: Juha Kiili <juha@juhakiili.com>
Date: Fri, 29 Nov 2019 15:33:25 +0200
Subject: [PATCH 159/505] Refactor logs and fix loss bug

---
 examples/run_glue.py | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/examples/run_glue.py b/examples/run_glue.py
index ea5ac5bbb7..8749593a1f 100644
--- a/examples/run_glue.py
+++ b/examples/run_glue.py
@@ -171,22 +171,22 @@ def train(args, train_dataset, model, tokenizer):
                 global_step += 1
 
                 if args.local_rank in [-1, 0] and args.logging_steps > 0 and global_step % args.logging_steps == 0:
-                    # Log metrics
-                    logs = {'step': global_step}
+                    logs = {}
                     if args.local_rank == -1 and args.evaluate_during_training:  # Only evaluate when single GPU otherwise metrics may not average well
                         results = evaluate(args, model, tokenizer)
                         for key, value in results.items():
                             eval_key = 'eval_{}'.format(key)
-                            tb_writer.add_scalar(eval_key, value, global_step)
-                            logs[eval_key] = str(value)
-                    logging_loss = tr_loss
+                            logs[eval_key] = value
+
                     loss_scalar = (tr_loss - logging_loss) / args.logging_steps
                     learning_rate_scalar = scheduler.get_lr()[0]
-                    tb_writer.add_scalar('lr', learning_rate_scalar, global_step)
-                    tb_writer.add_scalar('loss', loss_scalar, global_step)
                     logs['learning_rate'] = learning_rate_scalar
                     logs['loss'] = loss_scalar
-                    print(json.dumps(logs))
+                    logging_loss = tr_loss
+
+                    for key, value in logs.items():
+                        tb_writer.add_scalar(key, value, global_step)
+                    print(json.dumps({**logs, **{'step': global_step}}))
 
                 if args.local_rank in [-1, 0] and args.save_steps > 0 and global_step % args.save_steps == 0:
                     # Save model checkpoint

From 2421e54f8c354fc110a7f8819a9161163813f7ad Mon Sep 17 00:00:00 2001
From: Juha Kiili <juha@juhakiili.com>
Date: Fri, 29 Nov 2019 15:39:28 +0200
Subject: [PATCH 160/505] Add link to original source and license to
 download_glue.data.py

---
 utils/download_glue_data.py | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/utils/download_glue_data.py b/utils/download_glue_data.py
index 86a4e8951f..f676a71c76 100644
--- a/utils/download_glue_data.py
+++ b/utils/download_glue_data.py
@@ -1,5 +1,8 @@
 ''' Script for downloading all GLUE data.
 
+Original source: https://github.com/kamalkraj/ALBERT-TF2.0/blob/fa90194e5fe729dbb19f32ac29c8d6d6372c0f93/download_glue_data.py
+Original license: https://github.com/kamalkraj/ALBERT-TF2.0/blob/fa90194e5fe729dbb19f32ac29c8d6d6372c0f93/LICENSE (Apache-2.0)
+
 Note: for legal reasons, we are unable to host MRPC.
 You can either use the version hosted by the SentEval team, which is already tokenized, 
 or you can download the original data from (https://download.microsoft.com/download/D/4/6/D46FF87A-F6B9-4252-AA8B-3604ED519838/MSRParaphraseCorpus.msi) and extract the data from it manually.

From adb5c79ff2ffcd2e4a43a12f082cca55f7630a96 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Thu, 28 Nov 2019 15:51:43 +0100
Subject: [PATCH 161/505] update all tf.shape and tensor.shape to shape_list

---
 .../adding_a_new_model/modeling_tf_xxx.py     |  6 ++---
 transformers/__init__.py                      |  2 +-
 transformers/modeling_tf_albert.py            | 27 ++++++++-----------
 transformers/modeling_tf_bert.py              | 26 +++++++++---------
 transformers/modeling_tf_ctrl.py              |  2 +-
 transformers/modeling_tf_distilbert.py        |  8 +++---
 transformers/modeling_tf_gpt2.py              |  2 +-
 transformers/modeling_tf_openai.py            |  2 +-
 transformers/modeling_tf_roberta.py           |  6 ++---
 transformers/modeling_tf_transfo_xl.py        |  2 +-
 .../modeling_tf_transfo_xl_utilities.py       |  4 +--
 transformers/modeling_tf_utils.py             |  2 +-
 transformers/modeling_tf_xlnet.py             | 13 +++++----
 13 files changed, 48 insertions(+), 54 deletions(-)

diff --git a/templates/adding_a_new_model/modeling_tf_xxx.py b/templates/adding_a_new_model/modeling_tf_xxx.py
index f1d898b47a..59f798bdbf 100644
--- a/templates/adding_a_new_model/modeling_tf_xxx.py
+++ b/templates/adding_a_new_model/modeling_tf_xxx.py
@@ -32,7 +32,7 @@ import numpy as np
 import tensorflow as tf
 
 from .configuration_xxx import XxxConfig
-from .modeling_tf_utils import TFPreTrainedModel, get_initializer
+from .modeling_tf_utils import TFPreTrainedModel, get_initializer, shape_list
 from .file_utils import add_start_docstrings
 
 logger = logging.getLogger(__name__)
@@ -121,9 +121,9 @@ class TFXxxMainLayer(tf.keras.layers.Layer):
             input_ids = inputs
 
         if attention_mask is None:
-            attention_mask = tf.fill(tf.shape(input_ids), 1)
+            attention_mask = tf.fill(shape_list(input_ids), 1)
         if token_type_ids is None:
-            token_type_ids = tf.fill(tf.shape(input_ids), 0)
+            token_type_ids = tf.fill(shape_list(input_ids), 0)
 
         # We create a 3D attention mask from a 2D tensor mask.
         # Sizes are [batch_size, 1, 1, to_seq_length]
diff --git a/transformers/__init__.py b/transformers/__init__.py
index b29ad38e73..de25c24b9e 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -118,7 +118,7 @@ if is_torch_available():
 
 # TensorFlow
 if is_tf_available():
-    from .modeling_tf_utils import TFPreTrainedModel, TFSharedEmbeddings, TFSequenceSummary
+    from .modeling_tf_utils import TFPreTrainedModel, TFSharedEmbeddings, TFSequenceSummary, shape_list
     from .modeling_tf_auto import (TFAutoModel, TFAutoModelForSequenceClassification, TFAutoModelForQuestionAnswering,
                                    TFAutoModelWithLMHead)
 
diff --git a/transformers/modeling_tf_albert.py b/transformers/modeling_tf_albert.py
index b2bf66f750..164dc74320 100644
--- a/transformers/modeling_tf_albert.py
+++ b/transformers/modeling_tf_albert.py
@@ -16,18 +16,13 @@
 """ TF 2.0 ALBERT model. """
 from __future__ import absolute_import, division, print_function, unicode_literals
 
-import json
 import logging
-import math
-import os
 import sys
-from io import open
 
-import numpy as np
 import tensorflow as tf
 
 from .configuration_albert import AlbertConfig
-from .modeling_tf_utils import TFPreTrainedModel, get_initializer
+from .modeling_tf_utils import TFPreTrainedModel, get_initializer, shape_list
 from .modeling_tf_bert import ACT2FN, TFBertSelfAttention
 from .file_utils import add_start_docstrings
 
@@ -110,9 +105,9 @@ class TFAlbertEmbeddings(tf.keras.layers.Layer):
         input_ids, position_ids, token_type_ids, inputs_embeds = inputs
 
         if input_ids is not None:
-            input_shape = tf.shape(input_ids)
+            input_shape = shape_list(input_ids)
         else:
-            input_shape = tf.shape(inputs_embeds)[:-1]
+            input_shape = shape_list(inputs_embeds)[:-1]
 
         seq_length = input_shape[1]
         if position_ids is None:
@@ -137,8 +132,8 @@ class TFAlbertEmbeddings(tf.keras.layers.Layer):
             Returns:
                 float32 tensor with shape [batch_size, length, vocab_size].
         """
-        batch_size = tf.shape(inputs)[0]
-        length = tf.shape(inputs)[1]
+        batch_size = shape_list(inputs)[0]
+        length = shape_list(inputs)[1]
         x = tf.reshape(inputs, [-1, self.config.embedding_size])
         logits = tf.matmul(x, self.word_embeddings, transpose_b=True)
         return tf.reshape(logits, [batch_size, length, self.config.vocab_size])
@@ -183,7 +178,7 @@ class TFAlbertSelfAttention(tf.keras.layers.Layer):
     def call(self, inputs, training=False):
         hidden_states, attention_mask, head_mask = inputs
 
-        batch_size = tf.shape(hidden_states)[0]
+        batch_size = shape_list(hidden_states)[0]
         mixed_query_layer = self.query(hidden_states)
         mixed_key_layer = self.key(hidden_states)
         mixed_value_layer = self.value(hidden_states)
@@ -196,7 +191,7 @@ class TFAlbertSelfAttention(tf.keras.layers.Layer):
         # (batch size, num_heads, seq_len_q, seq_len_k)
         attention_scores = tf.matmul(query_layer, key_layer, transpose_b=True)
         # scale attention_scores
-        dk = tf.cast(tf.shape(key_layer)[-1], tf.float32)
+        dk = tf.cast(shape_list(key_layer)[-1], tf.float32)
         attention_scores = attention_scores / tf.math.sqrt(dk)
 
         if attention_mask is not None:
@@ -264,7 +259,7 @@ class TFAlbertAttention(TFBertSelfAttention):
     def call(self, inputs, training=False):
         input_tensor, attention_mask, head_mask = inputs
 
-        batch_size = tf.shape(input_tensor)[0]
+        batch_size = shape_list(input_tensor)[0]
         mixed_query_layer = self.query(input_tensor)
         mixed_key_layer = self.key(input_tensor)
         mixed_value_layer = self.value(input_tensor)
@@ -277,7 +272,7 @@ class TFAlbertAttention(TFBertSelfAttention):
         # (batch size, num_heads, seq_len_q, seq_len_k)
         attention_scores = tf.matmul(query_layer, key_layer, transpose_b=True)
         # scale attention_scores
-        dk = tf.cast(tf.shape(key_layer)[-1], tf.float32)
+        dk = tf.cast(shape_list(key_layer)[-1], tf.float32)
         attention_scores = attention_scores / tf.math.sqrt(dk)
 
         if attention_mask is not None:
@@ -645,9 +640,9 @@ class TFAlbertModel(TFAlbertPreTrainedModel):
         if input_ids is not None and inputs_embeds is not None:
             raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
         elif input_ids is not None:
-            input_shape = tf.shape(input_ids)
+            input_shape = shape_list(input_ids)
         elif inputs_embeds is not None:
-            input_shape = inputs_embeds.shape[:-1]
+            input_shape = shape_list(inputs_embeds)[:-1]
         else:
             raise ValueError("You have to specify either input_ids or inputs_embeds")
 
diff --git a/transformers/modeling_tf_bert.py b/transformers/modeling_tf_bert.py
index ad0815e2ca..5aa7bb3da2 100644
--- a/transformers/modeling_tf_bert.py
+++ b/transformers/modeling_tf_bert.py
@@ -28,7 +28,7 @@ import numpy as np
 import tensorflow as tf
 
 from .configuration_bert import BertConfig
-from .modeling_tf_utils import TFPreTrainedModel, get_initializer
+from .modeling_tf_utils import TFPreTrainedModel, get_initializer, shape_list
 from .file_utils import add_start_docstrings
 
 logger = logging.getLogger(__name__)
@@ -145,9 +145,9 @@ class TFBertEmbeddings(tf.keras.layers.Layer):
         input_ids, position_ids, token_type_ids, inputs_embeds = inputs
 
         if input_ids is not None:
-            input_shape = tf.shape(input_ids)
+            input_shape = shape_list(input_ids)
         else:
-            input_shape = tf.shape(inputs_embeds)[:-1]
+            input_shape = shape_list(inputs_embeds)[:-1]
         
         seq_length = input_shape[1]
         if position_ids is None:
@@ -172,8 +172,8 @@ class TFBertEmbeddings(tf.keras.layers.Layer):
             Returns:
                 float32 tensor with shape [batch_size, length, vocab_size].
         """
-        batch_size = tf.shape(inputs)[0]
-        length = tf.shape(inputs)[1]
+        batch_size = shape_list(inputs)[0]
+        length = shape_list(inputs)[1]
 
         x = tf.reshape(inputs, [-1, self.hidden_size])
         logits = tf.matmul(x, self.word_embeddings, transpose_b=True)
@@ -214,7 +214,7 @@ class TFBertSelfAttention(tf.keras.layers.Layer):
     def call(self, inputs, training=False):
         hidden_states, attention_mask, head_mask = inputs
 
-        batch_size = tf.shape(hidden_states)[0]
+        batch_size = shape_list(hidden_states)[0]
         mixed_query_layer = self.query(hidden_states)
         mixed_key_layer = self.key(hidden_states)
         mixed_value_layer = self.value(hidden_states)
@@ -225,7 +225,7 @@ class TFBertSelfAttention(tf.keras.layers.Layer):
 
         # Take the dot product between "query" and "key" to get the raw attention scores.
         attention_scores = tf.matmul(query_layer, key_layer, transpose_b=True)  # (batch size, num_heads, seq_len_q, seq_len_k)
-        dk = tf.cast(tf.shape(key_layer)[-1], tf.float32) # scale attention_scores
+        dk = tf.cast(shape_list(key_layer)[-1], tf.float32) # scale attention_scores
         attention_scores = attention_scores / tf.math.sqrt(dk)
 
         if attention_mask is not None:
@@ -502,9 +502,9 @@ class TFBertMainLayer(tf.keras.layers.Layer):
         if input_ids is not None and inputs_embeds is not None:
             raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
         elif input_ids is not None:
-            input_shape = input_ids.shape
+            input_shape = shape_list(input_ids)
         elif inputs_embeds is not None:
-            input_shape = inputs_embeds.shape[:-1]
+            input_shape = shape_list(inputs_embeds)[:-1]
         else:
             raise ValueError("You have to specify either input_ids or inputs_embeds")
 
@@ -939,11 +939,11 @@ class TFBertForMultipleChoice(TFBertPreTrainedModel):
             input_ids = inputs
 
         if input_ids is not None:
-            num_choices = tf.shape(input_ids)[1]
-            seq_length = tf.shape(input_ids)[2]
+            num_choices = shape_list(input_ids)[1]
+            seq_length = shape_list(input_ids)[2]
         else:
-            num_choices = tf.shape(inputs_embeds)[1]
-            seq_length = tf.shape(inputs_embeds)[2]
+            num_choices = shape_list(inputs_embeds)[1]
+            seq_length = shape_list(inputs_embeds)[2]
 
         flat_input_ids = tf.reshape(input_ids, (-1, seq_length)) if input_ids is not None else None
         flat_attention_mask = tf.reshape(attention_mask, (-1, seq_length)) if attention_mask is not None else None
diff --git a/transformers/modeling_tf_ctrl.py b/transformers/modeling_tf_ctrl.py
index ae66dbc82c..6d0d6a57ad 100644
--- a/transformers/modeling_tf_ctrl.py
+++ b/transformers/modeling_tf_ctrl.py
@@ -95,7 +95,7 @@ class TFMultiHeadAttention(tf.keras.layers.Layer):
 
     def call(self, inputs, training=False):
         v, k, q, mask, layer_past, attention_mask, head_mask = inputs
-        batch_size = q.shape[0]
+        batch_size = shape_list(q)[0]
 
         q = self.Wq(q)
         k = self.Wk(k)
diff --git a/transformers/modeling_tf_distilbert.py b/transformers/modeling_tf_distilbert.py
index 6d393bb95d..b3d4889475 100644
--- a/transformers/modeling_tf_distilbert.py
+++ b/transformers/modeling_tf_distilbert.py
@@ -137,9 +137,9 @@ class TFEmbeddings(tf.keras.layers.Layer):
             input_ids, position_ids = inputs
 
         if input_ids is not None:
-            seq_length = tf.shape(input_ids)[1]
+            seq_length = shape_list(input_ids)[1]
         else:
-            seq_length = tf.shape(inputs_embeds)[1]
+            seq_length = shape_list(inputs_embeds)[1]
 
         if position_ids is None:
             position_ids = tf.range(seq_length, dtype=tf.int32)[tf.newaxis, :]
@@ -160,8 +160,8 @@ class TFEmbeddings(tf.keras.layers.Layer):
             Returns:
                 float32 tensor with shape [batch_size, length, vocab_size].
         """
-        batch_size = tf.shape(inputs)[0]
-        length = tf.shape(inputs)[1]
+        batch_size = shape_list(inputs)[0]
+        length = shape_list(inputs)[1]
 
         x = tf.reshape(inputs, [-1, self.dim])
         logits = tf.matmul(x, self.word_embeddings, transpose_b=True)
diff --git a/transformers/modeling_tf_gpt2.py b/transformers/modeling_tf_gpt2.py
index 5e416a5e3a..aebe790114 100644
--- a/transformers/modeling_tf_gpt2.py
+++ b/transformers/modeling_tf_gpt2.py
@@ -92,7 +92,7 @@ class TFAttention(tf.keras.layers.Layer):
         # q, k, v have shape [batch, heads, sequence, features]
         w = tf.matmul(q, k, transpose_b=True)
         if self.scale:
-            dk = tf.cast(tf.shape(k)[-1], tf.float32) # scale attention_scores
+            dk = tf.cast(shape_list(k)[-1], tf.float32) # scale attention_scores
             w = w / tf.math.sqrt(dk)
 
         # w has shape [batch, heads, dst_sequence, src_sequence], where information flows from src to dst.
diff --git a/transformers/modeling_tf_openai.py b/transformers/modeling_tf_openai.py
index c553d92317..dac3b17590 100644
--- a/transformers/modeling_tf_openai.py
+++ b/transformers/modeling_tf_openai.py
@@ -98,7 +98,7 @@ class TFAttention(tf.keras.layers.Layer):
         # q, k, v have shape [batch, heads, sequence, features]
         w = tf.matmul(q, k, transpose_b=True)
         if self.scale:
-            dk = tf.cast(tf.shape(k)[-1], tf.float32) # scale attention_scores
+            dk = tf.cast(shape_list(k)[-1], tf.float32) # scale attention_scores
             w = w / tf.math.sqrt(dk)
 
         # w has shape [batch, heads, dst_sequence, src_sequence], where information flows from src to dst.
diff --git a/transformers/modeling_tf_roberta.py b/transformers/modeling_tf_roberta.py
index 450c0c72f2..954279f873 100644
--- a/transformers/modeling_tf_roberta.py
+++ b/transformers/modeling_tf_roberta.py
@@ -24,7 +24,7 @@ import numpy as np
 import tensorflow as tf
 
 from .configuration_roberta import RobertaConfig
-from .modeling_tf_utils import TFPreTrainedModel, get_initializer
+from .modeling_tf_utils import TFPreTrainedModel, get_initializer, shape_list
 from .file_utils import add_start_docstrings
 
 from .modeling_tf_bert import TFBertEmbeddings, TFBertMainLayer, gelu, gelu_new
@@ -51,9 +51,9 @@ class TFRobertaEmbeddings(TFBertEmbeddings):
         input_ids, position_ids, token_type_ids, inputs_embeds = inputs
 
         if input_ids is not None:
-            seq_length = tf.shape(input_ids)[1]
+            seq_length = shape_list(input_ids)[1]
         else:
-            seq_length = tf.shape(inputs_embeds)[1]
+            seq_length = shape_list(inputs_embeds)[1]
 
         if position_ids is None:
             position_ids = tf.range(self.padding_idx+1, seq_length+self.padding_idx+1, dtype=tf.int32)[tf.newaxis, :]
diff --git a/transformers/modeling_tf_transfo_xl.py b/transformers/modeling_tf_transfo_xl.py
index 8a8d11cfbc..fd325e218e 100644
--- a/transformers/modeling_tf_transfo_xl.py
+++ b/transformers/modeling_tf_transfo_xl.py
@@ -337,7 +337,7 @@ class TFAdaptiveEmbedding(tf.keras.layers.Layer):
                 emb_i = tf.einsum('id,de->ie', emb_i, self.emb_projs[i])
 
                 mask_idx = tf.cast(tf.where(mask_i), dtype=tf.int64)
-                emb_flat += tf.scatter_nd(mask_idx, emb_i, tf.cast(tf.shape(emb_flat), dtype=tf.int64))
+                emb_flat += tf.scatter_nd(mask_idx, emb_i, tf.cast(shape_list(emb_flat), dtype=tf.int64))
 
             embed_shape = shape_list(inp) + [self.d_proj]
             embed = tf.reshape(emb_flat, embed_shape)
diff --git a/transformers/modeling_tf_transfo_xl_utilities.py b/transformers/modeling_tf_transfo_xl_utilities.py
index d7666a650e..e6a6dfe686 100644
--- a/transformers/modeling_tf_transfo_xl_utilities.py
+++ b/transformers/modeling_tf_transfo_xl_utilities.py
@@ -105,7 +105,7 @@ class TFAdaptiveSoftmaxMask(tf.keras.layers.Layer):
 
     @staticmethod
     def _gather_logprob(logprob, target):
-        lp_size = tf.shape(logprob)
+        lp_size = shape_list(logprob)
         r = tf.range(lp_size[0])
         idx = tf.stack([r, target], 1)
         return tf.gather_nd(logprob, idx)
@@ -159,7 +159,7 @@ class TFAdaptiveSoftmaxMask(tf.keras.layers.Layer):
                         cur_logprob = self._gather_logprob(cur_tail_logprob, cur_target)
                         cur_logprob += cur_head_logprob[:, self.cutoff_ends[1] + i - 1]
                 if target is not None:
-                    loss += tf.scatter_nd(mask_idx, -cur_logprob, tf.cast(tf.shape(loss), dtype=tf.int64))
+                    loss += tf.scatter_nd(mask_idx, -cur_logprob, tf.cast(shape_list(loss), dtype=tf.int64))
             out = tf.concat(out, axis=-1)
 
         if target is not None:
diff --git a/transformers/modeling_tf_utils.py b/transformers/modeling_tf_utils.py
index 569b2faa4b..e4ba55e25e 100644
--- a/transformers/modeling_tf_utils.py
+++ b/transformers/modeling_tf_utils.py
@@ -494,7 +494,7 @@ class TFSequenceSummary(tf.keras.layers.Layer):
 def shape_list(x):
     """Deal with dynamic shape in tensorflow cleanly."""
     static = x.shape.as_list()
-    dynamic = tf.shape(x)
+    dynamic = shape_list(x)
     return [dynamic[i] if s is None else s for i, s in enumerate(static)]
 
 def get_initializer(initializer_range=0.02):
diff --git a/transformers/modeling_tf_xlnet.py b/transformers/modeling_tf_xlnet.py
index 4733ea8589..215d906f57 100644
--- a/transformers/modeling_tf_xlnet.py
+++ b/transformers/modeling_tf_xlnet.py
@@ -112,8 +112,7 @@ class TFXLNetRelativeAttention(tf.keras.layers.Layer):
     def prune_heads(self, heads):
         raise NotImplementedError
 
-    @staticmethod
-    def rel_shift(x, klen=-1):
+    def rel_shift(self, x, klen=-1):
         """perform relative shift to form the relative attention score."""
         x_size = shape_list(x)
 
@@ -135,7 +134,7 @@ class TFXLNetRelativeAttention(tf.keras.layers.Layer):
 
         # position based attention score
         bd = tf.einsum('ibnd,jbnd->ijbn', q_head + self.r_r_bias, k_head_r)
-        bd = self.rel_shift(bd, klen=ac.shape[1])
+        bd = self.rel_shift(bd, klen=shape_list(ac)[1])
 
         # segment based attention score
         if seg_mat is None:
@@ -192,7 +191,7 @@ class TFXLNetRelativeAttention(tf.keras.layers.Layer):
         if g is not None:
             ###### Two-stream attention with relative positional encoding.
             # content based attention score
-            if mems is not None and mems.shape.ndims > 1:
+            if mems is not None and len(shape_list(mems)) > 1:
                 cat = tf.concat([mems, h], axis=0)
             else:
                 cat = h
@@ -252,7 +251,7 @@ class TFXLNetRelativeAttention(tf.keras.layers.Layer):
 
         else:
             ###### Multi-head attention with relative positional encoding
-            if mems is not None and mems.shape.ndims > 1:
+            if mems is not None and len(shape_list(mems)) > 1:
                 cat = tf.concat([mems, h], axis=0)
             else:
                 cat = h
@@ -565,7 +564,7 @@ class TFXLNetMainLayer(tf.keras.layers.Layer):
 
         if data_mask is not None:
             # all mems can be attended to
-            mems_mask = tf.zeros([tf.shape(data_mask)[0], mlen, bsz],
+            mems_mask = tf.zeros([shape_list(data_mask)[0], mlen, bsz],
                                 dtype=dtype_float)
             data_mask = tf.concat([mems_mask, data_mask], axis=1)
             if attn_mask is None:
@@ -590,7 +589,7 @@ class TFXLNetMainLayer(tf.keras.layers.Layer):
             word_emb_k = self.word_embedding(input_ids)
         output_h = self.dropout(word_emb_k, training=training)
         if target_mapping is not None:
-            word_emb_q = tf.tile(self.mask_emb, [tf.shape(target_mapping)[0], bsz, 1])
+            word_emb_q = tf.tile(self.mask_emb, [shape_list(target_mapping)[0], bsz, 1])
         # else:  # We removed the inp_q input which was same as target mapping
         #     inp_q_ext = inp_q[:, :, None]
         #     word_emb_q = inp_q_ext * self.mask_emb + (1 - inp_q_ext) * word_emb_k

From 4a666885b501ed6bfd344ec2c4c16d80da8aab79 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Thu, 28 Nov 2019 15:56:53 +0100
Subject: [PATCH 162/505] reducing my level of enthousiasm

---
 transformers/modeling_tf_utils.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/modeling_tf_utils.py b/transformers/modeling_tf_utils.py
index e4ba55e25e..569b2faa4b 100644
--- a/transformers/modeling_tf_utils.py
+++ b/transformers/modeling_tf_utils.py
@@ -494,7 +494,7 @@ class TFSequenceSummary(tf.keras.layers.Layer):
 def shape_list(x):
     """Deal with dynamic shape in tensorflow cleanly."""
     static = x.shape.as_list()
-    dynamic = shape_list(x)
+    dynamic = tf.shape(x)
     return [dynamic[i] if s is None else s for i, s in enumerate(static)]
 
 def get_initializer(initializer_range=0.02):

From ecf15ebf3b7f5d2b0144f1a428e2d9f39494c8ba Mon Sep 17 00:00:00 2001
From: Elad Segal <eladsegal@users.noreply.github.com>
Date: Fri, 29 Nov 2019 12:48:41 +0200
Subject: [PATCH 163/505] Add ALBERT to AutoClasses

---
 transformers/configuration_auto.py | 15 +++++++++-----
 transformers/modeling_auto.py      | 33 +++++++++++++++++++++++-------
 transformers/tokenization_auto.py  | 15 +++++++++-----
 3 files changed, 46 insertions(+), 17 deletions(-)

diff --git a/transformers/configuration_auto.py b/transformers/configuration_auto.py
index 37b8c0e3df..43f251bd0c 100644
--- a/transformers/configuration_auto.py
+++ b/transformers/configuration_auto.py
@@ -28,6 +28,7 @@ from .configuration_roberta import RobertaConfig
 from .configuration_distilbert import DistilBertConfig
 from .configuration_ctrl import CTRLConfig
 from .configuration_camembert import CamembertConfig
+from .configuration_albert import AlbertConfig
 
 logger = logging.getLogger(__name__)
 
@@ -44,14 +45,15 @@ class AutoConfig(object):
         The base model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertConfig (DistilBERT model)
+            - contains `albert`: AlbertConfig (ALBERT model)
+            - contains `camembert`: CamembertConfig (CamemBERT model)
+            - contains `roberta`: RobertaConfig (RoBERTa model)
             - contains `bert`: BertConfig (Bert model)
             - contains `openai-gpt`: OpenAIGPTConfig (OpenAI GPT model)
             - contains `gpt2`: GPT2Config (OpenAI GPT-2 model)
             - contains `transfo-xl`: TransfoXLConfig (Transformer-XL model)
             - contains `xlnet`: XLNetConfig (XLNet model)
             - contains `xlm`: XLMConfig (XLM model)
-            - contains `roberta`: RobertaConfig (RoBERTa model)
-            - contains `camembert`: CamembertConfig (CamemBERT model)
             - contains `ctrl` : CTRLConfig (CTRL model)
         This class cannot be instantiated using `__init__()` (throw an error).
     """
@@ -67,14 +69,15 @@ class AutoConfig(object):
         The configuration class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertConfig (DistilBERT model)
+            - contains `albert`: AlbertConfig (ALBERT model)
+            - contains `camembert`: CamembertConfig (CamemBERT model)
+            - contains `roberta`: RobertaConfig (RoBERTa model)
             - contains `bert`: BertConfig (Bert model)
             - contains `openai-gpt`: OpenAIGPTConfig (OpenAI GPT model)
             - contains `gpt2`: GPT2Config (OpenAI GPT-2 model)
             - contains `transfo-xl`: TransfoXLConfig (Transformer-XL model)
             - contains `xlnet`: XLNetConfig (XLNet model)
             - contains `xlm`: XLMConfig (XLM model)
-            - contains `roberta`: RobertaConfig (RoBERTa model)
-            - contains `camembert`: CamembertConfig (CamemBERT model)
             - contains `ctrl` : CTRLConfig (CTRL model)
         Params:
             pretrained_model_name_or_path: either:
@@ -122,6 +125,8 @@ class AutoConfig(object):
         """
         if 'distilbert' in pretrained_model_name_or_path:
             return DistilBertConfig.from_pretrained(pretrained_model_name_or_path, **kwargs)
+        elif 'albert' in pretrained_model_name_or_path:
+            return AlbertConfig.from_pretrained(pretrained_model_name_or_path, **kwargs)
         elif 'camembert' in pretrained_model_name_or_path:
             return CamembertConfig.from_pretrained(pretrained_model_name_or_path, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
@@ -142,4 +147,4 @@ class AutoConfig(object):
             return CTRLConfig.from_pretrained(pretrained_model_name_or_path, **kwargs)
         raise ValueError("Unrecognized model identifier in {}. Should contains one of "
                          "'bert', 'openai-gpt', 'gpt2', 'transfo-xl', 'xlnet', "
-                         "'xlm', 'roberta', 'camembert', 'ctrl'".format(pretrained_model_name_or_path))
+                         "'xlm', 'roberta', 'distilbert', 'camembert', 'ctrl', 'albert'".format(pretrained_model_name_or_path))
diff --git a/transformers/modeling_auto.py b/transformers/modeling_auto.py
index fa33dbc0c8..b63e43d73b 100644
--- a/transformers/modeling_auto.py
+++ b/transformers/modeling_auto.py
@@ -28,6 +28,8 @@ from .modeling_xlm import XLMModel, XLMWithLMHeadModel, XLMForSequenceClassifica
 from .modeling_roberta import RobertaModel, RobertaForMaskedLM, RobertaForSequenceClassification
 from .modeling_distilbert import DistilBertModel, DistilBertForQuestionAnswering, DistilBertForMaskedLM, DistilBertForSequenceClassification
 from .modeling_camembert import CamembertModel, CamembertForMaskedLM, CamembertForSequenceClassification, CamembertForMultipleChoice
+from .modeling_camembert import CamembertModel, CamembertForMaskedLM, CamembertForSequenceClassification, CamembertForMultipleChoice
+from .modeling_albert import AlbertModel, AlbertForMaskedLM, AlbertForSequenceClassification, AlbertForQuestionAnswering
 
 from .modeling_utils import PreTrainedModel, SequenceSummary
 
@@ -49,15 +51,16 @@ class AutoModel(object):
         The base model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertModel (DistilBERT model)
+            - contains `albert`: AlbertModel (ALBERT model)
             - contains `camembert`: CamembertModel (CamemBERT model)
             - contains `roberta`: RobertaModel (RoBERTa model)
             - contains `bert`: BertModel (Bert model)
             - contains `openai-gpt`: OpenAIGPTModel (OpenAI GPT model)
             - contains `gpt2`: GPT2Model (OpenAI GPT-2 model)
-            - contains `ctrl`: CTRLModel (Salesforce CTRL  model)
             - contains `transfo-xl`: TransfoXLModel (Transformer-XL model)
             - contains `xlnet`: XLNetModel (XLNet model)
             - contains `xlm`: XLMModel (XLM model)
+            - contains `ctrl`: CTRLModel (Salesforce CTRL  model)
 
         This class cannot be instantiated using `__init__()` (throws an error).
     """
@@ -73,15 +76,16 @@ class AutoModel(object):
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertModel (DistilBERT model)
+            - contains `albert`: AlbertModel (ALBERT model)
             - contains `camembert`: CamembertModel (CamemBERT model)
             - contains `roberta`: RobertaModel (RoBERTa model)
             - contains `bert`: BertModel (Bert model)
             - contains `openai-gpt`: OpenAIGPTModel (OpenAI GPT model)
             - contains `gpt2`: GPT2Model (OpenAI GPT-2 model)
-            - contains `ctrl`: CTRLModel (Salesforce CTRL  model)
             - contains `transfo-xl`: TransfoXLModel (Transformer-XL model)
             - contains `xlnet`: XLNetModel (XLNet model)
             - contains `xlm`: XLMModel (XLM model)
+            - contains `ctrl`: CTRLModel (Salesforce CTRL model)
 
             The model is set in evaluation mode by default using `model.eval()` (Dropout modules are deactivated)
             To train the model, you should first set it back in training mode with `model.train()`
@@ -144,6 +148,8 @@ class AutoModel(object):
         """
         if 'distilbert' in pretrained_model_name_or_path:
             return DistilBertModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'albert' in pretrained_model_name_or_path:
+            return AlbertModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'camembert' in pretrained_model_name_or_path:
             return CamembertModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
@@ -164,7 +170,7 @@ class AutoModel(object):
             return CTRLModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         raise ValueError("Unrecognized model identifier in {}. Should contains one of "
                          "'bert', 'openai-gpt', 'gpt2', 'transfo-xl', 'xlnet', "
-                         "'xlm', 'roberta, 'ctrl'".format(pretrained_model_name_or_path))
+                         "'xlm', 'roberta, 'ctrl', 'distilbert', 'camembert', 'albert'".format(pretrained_model_name_or_path))
 
 
 class AutoModelWithLMHead(object):
@@ -180,15 +186,16 @@ class AutoModelWithLMHead(object):
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertForMaskedLM (DistilBERT model)
+            - contains `albert`: AlbertForMaskedLM (ALBERT model)
             - contains `camembert`: CamembertForMaskedLM (CamemBERT model)
             - contains `roberta`: RobertaForMaskedLM (RoBERTa model)
             - contains `bert`: BertForMaskedLM (Bert model)
             - contains `openai-gpt`: OpenAIGPTLMHeadModel (OpenAI GPT model)
             - contains `gpt2`: GPT2LMHeadModel (OpenAI GPT-2 model)
-            - contains `ctrl`: CTRLLMModel (Salesforce CTRL model)
             - contains `transfo-xl`: TransfoXLLMHeadModel (Transformer-XL model)
             - contains `xlnet`: XLNetLMHeadModel (XLNet model)
             - contains `xlm`: XLMWithLMHeadModel (XLM model)
+            - contains `ctrl`: CTRLLMHeadModel (Salesforce CTRL model)
 
         This class cannot be instantiated using `__init__()` (throws an error).
     """
@@ -207,6 +214,7 @@ class AutoModelWithLMHead(object):
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertForMaskedLM (DistilBERT model)
+            - contains `albert`: AlbertForMaskedLM (ALBERT model)
             - contains `camembert`: CamembertForMaskedLM (CamemBERT model)
             - contains `roberta`: RobertaForMaskedLM (RoBERTa model)
             - contains `bert`: BertForMaskedLM (Bert model)
@@ -215,6 +223,7 @@ class AutoModelWithLMHead(object):
             - contains `transfo-xl`: TransfoXLLMHeadModel (Transformer-XL model)
             - contains `xlnet`: XLNetLMHeadModel (XLNet model)
             - contains `xlm`: XLMWithLMHeadModel (XLM model)
+            - contains `ctrl`: CTRLLMHeadModel (Salesforce CTRL model)
 
         The model is set in evaluation mode by default using `model.eval()` (Dropout modules are deactivated)
         To train the model, you should first set it back in training mode with `model.train()`
@@ -276,6 +285,8 @@ class AutoModelWithLMHead(object):
         """
         if 'distilbert' in pretrained_model_name_or_path:
             return DistilBertForMaskedLM.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'albert' in pretrained_model_name_or_path:
+            return AlbertForMaskedLM.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'camembert' in pretrained_model_name_or_path:
             return CamembertForMaskedLM.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
@@ -296,7 +307,7 @@ class AutoModelWithLMHead(object):
             return CTRLLMHeadModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         raise ValueError("Unrecognized model identifier in {}. Should contains one of "
                          "'bert', 'openai-gpt', 'gpt2', 'transfo-xl', 'xlnet', "
-                         "'xlm', 'roberta','ctrl'".format(pretrained_model_name_or_path))
+                         "'xlm', 'roberta','ctrl', 'distilbert', 'camembert', 'albert'".format(pretrained_model_name_or_path))
 
 
 class AutoModelForSequenceClassification(object):
@@ -312,6 +323,7 @@ class AutoModelForSequenceClassification(object):
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertForSequenceClassification (DistilBERT model)
+            - contains `albert`: AlbertForSequenceClassification (ALBERT model)
             - contains `camembert`: CamembertForSequenceClassification (CamemBERT model)
             - contains `roberta`: RobertaForSequenceClassification (RoBERTa model)
             - contains `bert`: BertForSequenceClassification (Bert model)
@@ -335,6 +347,7 @@ class AutoModelForSequenceClassification(object):
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertForSequenceClassification (DistilBERT model)
+            - contains `albert`: AlbertForSequenceClassification (ALBERT model)
             - contains `camembert`: CamembertForSequenceClassification (CamemBERT model)
             - contains `roberta`: RobertaForSequenceClassification (RoBERTa model)
             - contains `bert`: BertForSequenceClassification (Bert model)
@@ -402,6 +415,8 @@ class AutoModelForSequenceClassification(object):
         """
         if 'distilbert' in pretrained_model_name_or_path:
             return DistilBertForSequenceClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'albert' in pretrained_model_name_or_path:
+            return AlbertForSequenceClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'camembert' in pretrained_model_name_or_path:
             return CamembertForSequenceClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
@@ -414,7 +429,7 @@ class AutoModelForSequenceClassification(object):
             return XLMForSequenceClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
 
         raise ValueError("Unrecognized model identifier in {}. Should contains one of "
-                         "'bert', 'xlnet', 'xlm', 'roberta'".format(pretrained_model_name_or_path))
+                         "'bert', 'xlnet', 'xlm', 'roberta', 'distilbert', 'camembert', 'albert'".format(pretrained_model_name_or_path))
 
 
 class AutoModelForQuestionAnswering(object):
@@ -430,6 +445,7 @@ class AutoModelForQuestionAnswering(object):
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertForQuestionAnswering (DistilBERT model)
+            - contains `albert`: AlbertForQuestionAnswering (ALBERT model)
             - contains `bert`: BertForQuestionAnswering (Bert model)
             - contains `xlnet`: XLNetForQuestionAnswering (XLNet model)
             - contains `xlm`: XLMForQuestionAnswering (XLM model)
@@ -451,6 +467,7 @@ class AutoModelForQuestionAnswering(object):
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertForQuestionAnswering (DistilBERT model)
+            - contains `albert`: AlbertForQuestionAnswering (ALBERT model)
             - contains `bert`: BertForQuestionAnswering (Bert model)
             - contains `xlnet`: XLNetForQuestionAnswering (XLNet model)
             - contains `xlm`: XLMForQuestionAnswering (XLM model)
@@ -513,6 +530,8 @@ class AutoModelForQuestionAnswering(object):
         """
         if 'distilbert' in pretrained_model_name_or_path:
             return DistilBertForQuestionAnswering.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'albert' in pretrained_model_name_or_path:
+            return AlbertForQuestionAnswering.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'bert' in pretrained_model_name_or_path:
             return BertForQuestionAnswering.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'xlnet' in pretrained_model_name_or_path:
@@ -521,4 +540,4 @@ class AutoModelForQuestionAnswering(object):
             return XLMForQuestionAnswering.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
 
         raise ValueError("Unrecognized model identifier in {}. Should contains one of "
-                         "'bert', 'xlnet', 'xlm'".format(pretrained_model_name_or_path))
+                         "'bert', 'xlnet', 'xlm', 'distilbert', 'albert'".format(pretrained_model_name_or_path))
diff --git a/transformers/tokenization_auto.py b/transformers/tokenization_auto.py
index 2e15e38073..b7c5046961 100644
--- a/transformers/tokenization_auto.py
+++ b/transformers/tokenization_auto.py
@@ -28,6 +28,7 @@ from .tokenization_xlm import XLMTokenizer
 from .tokenization_roberta import RobertaTokenizer
 from .tokenization_distilbert import DistilBertTokenizer
 from .tokenization_camembert import CamembertTokenizer
+from .tokenization_albert import AlbertTokenizer
 
 logger = logging.getLogger(__name__)
 
@@ -42,16 +43,17 @@ class AutoTokenizer(object):
 
         The tokenizer class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
-            - contains `camembert`: CamembertTokenizer (CamemBERT model)
             - contains `distilbert`: DistilBertTokenizer (DistilBert model)
+            - contains `albert`: AlbertTokenizer (ALBERT model)
+            - contains `camembert`: CamembertTokenizer (CamemBERT model)
             - contains `roberta`: RobertaTokenizer (RoBERTa model)
             - contains `bert`: BertTokenizer (Bert model)
             - contains `openai-gpt`: OpenAIGPTTokenizer (OpenAI GPT model)
             - contains `gpt2`: GPT2Tokenizer (OpenAI GPT-2 model)
-            - contains `ctrl`: CTRLTokenizer (Salesforce CTRL model)
             - contains `transfo-xl`: TransfoXLTokenizer (Transformer-XL model)
             - contains `xlnet`: XLNetTokenizer (XLNet model)
             - contains `xlm`: XLMTokenizer (XLM model)
+            - contains `ctrl`: CTRLTokenizer (Salesforce CTRL model)
 
         This class cannot be instantiated using `__init__()` (throw an error).
     """
@@ -66,16 +68,17 @@ class AutoTokenizer(object):
 
         The tokenizer class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
-            - contains `camembert`: CamembertTokenizer (CamemBERT model)
             - contains `distilbert`: DistilBertTokenizer (DistilBert model)
+            - contains `albert`: AlbertTokenizer (ALBERT model)
+            - contains `camembert`: CamembertTokenizer (CamemBERT model)
             - contains `roberta`: RobertaTokenizer (RoBERTa model)
             - contains `bert`: BertTokenizer (Bert model)
             - contains `openai-gpt`: OpenAIGPTTokenizer (OpenAI GPT model)
             - contains `gpt2`: GPT2Tokenizer (OpenAI GPT-2 model)
-            - contains `ctrl`: CTRLTokenizer (Salesforce CTRL model)
             - contains `transfo-xl`: TransfoXLTokenizer (Transformer-XL model)
             - contains `xlnet`: XLNetTokenizer (XLNet model)
             - contains `xlm`: XLMTokenizer (XLM model)
+            - contains `ctrl`: CTRLTokenizer (Salesforce CTRL model)
 
         Params:
             pretrained_model_name_or_path: either:
@@ -109,6 +112,8 @@ class AutoTokenizer(object):
         """
         if 'distilbert' in pretrained_model_name_or_path:
             return DistilBertTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
+        elif 'albert' in pretrained_model_name_or_path:
+            return AlbertTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
         elif 'camembert' in pretrained_model_name_or_path:
             return CamembertTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
@@ -129,4 +134,4 @@ class AutoTokenizer(object):
             return CTRLTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
         raise ValueError("Unrecognized model identifier in {}. Should contains one of "
                          "'bert', 'openai-gpt', 'gpt2', 'transfo-xl', 'xlnet', "
-                         "'xlm', 'roberta', 'camembert', 'ctrl'".format(pretrained_model_name_or_path))
+                         "'xlm', 'roberta', 'distilbert,' 'camembert', 'ctrl', 'albert'".format(pretrained_model_name_or_path))

From b0ee7c7df3d49a819c4d6cef977214bd91f5c075 Mon Sep 17 00:00:00 2001
From: maxvidal <44881831+maxvidal@users.noreply.github.com>
Date: Fri, 29 Nov 2019 12:32:37 +0100
Subject: [PATCH 164/505] Added Camembert to available models

---
 examples/run_lm_finetuning.py | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/examples/run_lm_finetuning.py b/examples/run_lm_finetuning.py
index c33aa94a32..4acea00c55 100644
--- a/examples/run_lm_finetuning.py
+++ b/examples/run_lm_finetuning.py
@@ -47,7 +47,8 @@ from transformers import (WEIGHTS_NAME, AdamW, get_linear_schedule_with_warmup,
                                   GPT2Config, GPT2LMHeadModel, GPT2Tokenizer,
                                   OpenAIGPTConfig, OpenAIGPTLMHeadModel, OpenAIGPTTokenizer,
                                   RobertaConfig, RobertaForMaskedLM, RobertaTokenizer,
-                                  DistilBertConfig, DistilBertForMaskedLM, DistilBertTokenizer)
+                                  DistilBertConfig, DistilBertForMaskedLM, DistilBertTokenizer,
+                                  CamembertConfig, CamembertForMaskedLM, CamembertTokenizer)
 
 
 logger = logging.getLogger(__name__)
@@ -58,7 +59,8 @@ MODEL_CLASSES = {
     'openai-gpt': (OpenAIGPTConfig, OpenAIGPTLMHeadModel, OpenAIGPTTokenizer),
     'bert': (BertConfig, BertForMaskedLM, BertTokenizer),
     'roberta': (RobertaConfig, RobertaForMaskedLM, RobertaTokenizer),
-    'distilbert': (DistilBertConfig, DistilBertForMaskedLM, DistilBertTokenizer)
+    'distilbert': (DistilBertConfig, DistilBertForMaskedLM, DistilBertTokenizer),
+    'camembert': (CamembertConfig, CamembertForMaskedLM, CamembertTokenizer)
 }
 
 
@@ -432,7 +434,7 @@ def main():
     parser.add_argument('--server_port', type=str, default='', help="For distant debugging.")
     args = parser.parse_args()
 
-    if args.model_type in ["bert", "roberta", "distilbert"] and not args.mlm:
+    if args.model_type in ["bert", "roberta", "distilbert", "camembert"] and not args.mlm:
         raise ValueError("BERT and RoBERTa do not have LM heads but masked LM heads. They must be run using the --mlm "
                          "flag (masked language modeling).")
     if args.eval_data_file is None and args.do_eval:

From b90791e95026dfa95a8cea72605257e4b9355956 Mon Sep 17 00:00:00 2001
From: Rostislav Nedelchev <rostislav.nedelchev@gmail.com>
Date: Sat, 30 Nov 2019 15:57:51 +0100
Subject: [PATCH 165/505] fixed XLNet attenttion output for both attention
 streams

---
 transformers/modeling_xlnet.py | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/transformers/modeling_xlnet.py b/transformers/modeling_xlnet.py
index 658048a660..476d9ab13d 100644
--- a/transformers/modeling_xlnet.py
+++ b/transformers/modeling_xlnet.py
@@ -581,7 +581,7 @@ class XLNetModel(XLNetPreTrainedModel):
             of shape ``(batch_size, sequence_length, hidden_size)``:
             Hidden-states of the model at the output of each layer plus the initial embedding outputs.
         **attentions**: (`optional`, returned when ``config.output_attentions=True``)
-            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            list of 2-tuple of ``torch.FloatTensor`` (one for each layer, one for each attention stream) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
             Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
 
     Examples::
@@ -878,7 +878,7 @@ class XLNetModel(XLNetPreTrainedModel):
                 hidden_states = tuple(hs.permute(1, 0, 2).contiguous() for hs in hidden_states)
             outputs = outputs + (hidden_states,)
         if self.output_attentions:
-            attentions = tuple(t.permute(2, 3, 0, 1).contiguous() for t in attentions)
+            attentions = tuple(tuple(att_stream.permute(2, 3, 0, 1).contiguous() for att_stream in t) for t in attentions)
             outputs = outputs + (attentions,)
 
         return outputs  # outputs, (new_mems), (hidden_states), (attentions)
@@ -911,7 +911,7 @@ class XLNetLMHeadModel(XLNetPreTrainedModel):
             of shape ``(batch_size, sequence_length, hidden_size)``:
             Hidden-states of the model at the output of each layer plus the initial embedding outputs.
         **attentions**: (`optional`, returned when ``config.output_attentions=True``)
-            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            list of 2-tuple of ``torch.FloatTensor`` (one for each layer, one for each attention stream) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
             Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
 
     Examples::
@@ -993,7 +993,7 @@ class XLNetForSequenceClassification(XLNetPreTrainedModel):
             of shape ``(batch_size, sequence_length, hidden_size)``:
             Hidden-states of the model at the output of each layer plus the initial embedding outputs.
         **attentions**: (`optional`, returned when ``config.output_attentions=True``)
-            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            list of 2-tuple of ``torch.FloatTensor`` (one for each layer, one for each attention stream) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
             Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
 
     Examples::
@@ -1093,7 +1093,7 @@ class XLNetForMultipleChoice(XLNetPreTrainedModel):
             of shape ``(batch_size, sequence_length, hidden_size)``:
             Hidden-states of the model at the output of each layer plus the initial embedding outputs.
         **attentions**: (`optional`, returned when ``config.output_attentions=True``)
-            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            list of 2-tuple of ``torch.FloatTensor`` (one for each layer, one for each attention stream) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
             Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
 
     Examples::
@@ -1178,7 +1178,7 @@ class XLNetForQuestionAnsweringSimple(XLNetPreTrainedModel):
             of shape ``(batch_size, sequence_length, hidden_size)``:
             Hidden-states of the model at the output of each layer plus the initial embedding outputs.
         **attentions**: (`optional`, returned when ``config.output_attentions=True``)
-            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            list of 2-tuple of ``torch.FloatTensor`` (one for each layer, one for each attention stream) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
             Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
 
     Examples::
@@ -1292,7 +1292,7 @@ class XLNetForQuestionAnswering(XLNetPreTrainedModel):
             of shape ``(batch_size, sequence_length, hidden_size)``:
             Hidden-states of the model at the output of each layer plus the initial embedding outputs.
         **attentions**: (`optional`, returned when ``config.output_attentions=True``)
-            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            list of 2-tuple of ``torch.FloatTensor`` (one for each layer, one for each attention stream) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
             Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
 
     Examples::

From 76c0bc06d549e0ecf746fdd3d72eb235a9c4aec2 Mon Sep 17 00:00:00 2001
From: Rostislav Nedelchev <rostislav.nedelchev@gmail.com>
Date: Sat, 30 Nov 2019 21:01:04 +0100
Subject: [PATCH 166/505] [XLNet] Changed post-processing of attention w.r.t to
 target_mapping

Whenever target_mapping is provided to the input, XLNet outputs two different attention streams.
Based on that the attention output would be on of the two:
- a list of tensors (usual case for most transformers)
- a list of 2-tuples of tensors, one tesor for each of attention streams
Docs and unit-tests have been updated
---
 transformers/modeling_xlnet.py            | 24 ++++++++++++++++-------
 transformers/tests/modeling_xlnet_test.py | 18 +++++++++++++++++
 2 files changed, 35 insertions(+), 7 deletions(-)

diff --git a/transformers/modeling_xlnet.py b/transformers/modeling_xlnet.py
index 476d9ab13d..56d755c11b 100644
--- a/transformers/modeling_xlnet.py
+++ b/transformers/modeling_xlnet.py
@@ -581,8 +581,9 @@ class XLNetModel(XLNetPreTrainedModel):
             of shape ``(batch_size, sequence_length, hidden_size)``:
             Hidden-states of the model at the output of each layer plus the initial embedding outputs.
         **attentions**: (`optional`, returned when ``config.output_attentions=True``)
-            list of 2-tuple of ``torch.FloatTensor`` (one for each layer, one for each attention stream) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
             Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+            When ``target_mapping is not None``, the attentions outputs are a list of 2-tuple of ``torch.FloatTensor``.
 
     Examples::
 
@@ -878,7 +879,11 @@ class XLNetModel(XLNetPreTrainedModel):
                 hidden_states = tuple(hs.permute(1, 0, 2).contiguous() for hs in hidden_states)
             outputs = outputs + (hidden_states,)
         if self.output_attentions:
-            attentions = tuple(tuple(att_stream.permute(2, 3, 0, 1).contiguous() for att_stream in t) for t in attentions)
+            if target_mapping is not None:
+                # when target_mapping is provided, there are 2-tuple of attentions
+                attentions = tuple(tuple(att_stream.permute(2, 3, 0, 1).contiguous() for att_stream in t) for t in attentions)
+            else:
+                attentions = tuple(t.permute(2, 3, 0, 1).contiguous() for t in attentions)
             outputs = outputs + (attentions,)
 
         return outputs  # outputs, (new_mems), (hidden_states), (attentions)
@@ -911,8 +916,9 @@ class XLNetLMHeadModel(XLNetPreTrainedModel):
             of shape ``(batch_size, sequence_length, hidden_size)``:
             Hidden-states of the model at the output of each layer plus the initial embedding outputs.
         **attentions**: (`optional`, returned when ``config.output_attentions=True``)
-            list of 2-tuple of ``torch.FloatTensor`` (one for each layer, one for each attention stream) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
             Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+            When ``target_mapping is not None``, the attentions outputs are a list of 2-tuple of ``torch.FloatTensor``.
 
     Examples::
 
@@ -993,8 +999,9 @@ class XLNetForSequenceClassification(XLNetPreTrainedModel):
             of shape ``(batch_size, sequence_length, hidden_size)``:
             Hidden-states of the model at the output of each layer plus the initial embedding outputs.
         **attentions**: (`optional`, returned when ``config.output_attentions=True``)
-            list of 2-tuple of ``torch.FloatTensor`` (one for each layer, one for each attention stream) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
             Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+            When ``target_mapping is not None``, the attentions outputs are a list of 2-tuple of ``torch.FloatTensor``.
 
     Examples::
 
@@ -1093,8 +1100,9 @@ class XLNetForMultipleChoice(XLNetPreTrainedModel):
             of shape ``(batch_size, sequence_length, hidden_size)``:
             Hidden-states of the model at the output of each layer plus the initial embedding outputs.
         **attentions**: (`optional`, returned when ``config.output_attentions=True``)
-            list of 2-tuple of ``torch.FloatTensor`` (one for each layer, one for each attention stream) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
             Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+            When ``target_mapping is not None``, the attentions outputs are a list of 2-tuple of ``torch.FloatTensor``.
 
     Examples::
 
@@ -1178,8 +1186,9 @@ class XLNetForQuestionAnsweringSimple(XLNetPreTrainedModel):
             of shape ``(batch_size, sequence_length, hidden_size)``:
             Hidden-states of the model at the output of each layer plus the initial embedding outputs.
         **attentions**: (`optional`, returned when ``config.output_attentions=True``)
-            list of 2-tuple of ``torch.FloatTensor`` (one for each layer, one for each attention stream) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
             Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+            When ``target_mapping is not None``, the attentions outputs are a list of 2-tuple of ``torch.FloatTensor``.
 
     Examples::
 
@@ -1292,8 +1301,9 @@ class XLNetForQuestionAnswering(XLNetPreTrainedModel):
             of shape ``(batch_size, sequence_length, hidden_size)``:
             Hidden-states of the model at the output of each layer plus the initial embedding outputs.
         **attentions**: (`optional`, returned when ``config.output_attentions=True``)
-            list of 2-tuple of ``torch.FloatTensor`` (one for each layer, one for each attention stream) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
             Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+            When ``target_mapping is not None``, the attentions outputs are a list of 2-tuple of ``torch.FloatTensor``.
 
     Examples::
 
diff --git a/transformers/tests/modeling_xlnet_test.py b/transformers/tests/modeling_xlnet_test.py
index d97ea6a425..a5ee9b1e0e 100644
--- a/transformers/tests/modeling_xlnet_test.py
+++ b/transformers/tests/modeling_xlnet_test.py
@@ -163,6 +163,18 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
                 list(list(mem.size()) for mem in result["mems_1"]),
                 [[self.seq_length, self.batch_size, self.hidden_size]] * self.num_hidden_layers)
 
+        def create_and_check_xlnet_base_model_with_att_output(self, config, input_ids_1, input_ids_2, input_ids_q, perm_mask, input_mask,
+                target_mapping, segment_ids, lm_labels, sequence_labels, is_impossible_labels):
+            model = XLNetModel(config)
+            model.eval()
+
+            _, _, attentions = model(input_ids_1, target_mapping=target_mapping)
+
+            self.parent.assertEqual(len(attentions), config.n_layer)
+            self.parent.assertIsInstance(attentions[0], tuple)
+            self.parent.assertEqual(len(attentions[0]), 2)
+            self.parent.assertTrue(attentions[0][0].shape, attentions[0][0].shape)
+
         def create_and_check_xlnet_lm_head(self, config, input_ids_1, input_ids_2, input_ids_q, perm_mask, input_mask,
                 target_mapping, segment_ids, lm_labels, sequence_labels, is_impossible_labels):
             model = XLNetLMHeadModel(config)
@@ -306,6 +318,12 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_xlnet_base_model(*config_and_inputs)
 
+    def test_xlnet_base_model_with_att_output(self):
+        self.model_tester.set_seed()
+        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+        config_and_inputs[0].output_attentions = True
+        self.model_tester.create_and_check_xlnet_base_model_with_att_output(*config_and_inputs)
+
     def test_xlnet_lm_head(self):
         self.model_tester.set_seed()
         config_and_inputs = self.model_tester.prepare_config_and_inputs()

From c356290c8ddd9037bd08c854b118673983ef6def Mon Sep 17 00:00:00 2001
From: Aditya Soni <soniaditya922@gmail.com>
Date: Sun, 1 Dec 2019 14:08:14 +0530
Subject: [PATCH 167/505] typo fix as per Pytorch v1.1+

---
 docs/source/migration.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/source/migration.md b/docs/source/migration.md
index d04b66d5e4..f50d1dff0a 100644
--- a/docs/source/migration.md
+++ b/docs/source/migration.md
@@ -104,6 +104,6 @@ for batch in train_data:
     loss = model(batch)
     loss.backward()
     torch.nn.utils.clip_grad_norm_(model.parameters(), max_grad_norm)  # Gradient clipping is not in AdamW anymore (so you can use amp without issue)
-    scheduler.step()
     optimizer.step()
+    scheduler.step()
 ```

From 5ab93083e4d9b610df1dd082f17a455a0b7193ac Mon Sep 17 00:00:00 2001
From: Aymeric Augustin <aymeric.augustin@m4x.org>
Date: Sun, 1 Dec 2019 18:25:15 +0100
Subject: [PATCH 168/505] Mark tests in TFAutoModelTest as slow.

Each test forces downloading the same 536MB file, which is slow
even with a decent internet connection.
---
 transformers/tests/modeling_tf_auto_test.py | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/transformers/tests/modeling_tf_auto_test.py b/transformers/tests/modeling_tf_auto_test.py
index 2cda3abc1c..fa90906e86 100644
--- a/transformers/tests/modeling_tf_auto_test.py
+++ b/transformers/tests/modeling_tf_auto_test.py
@@ -38,6 +38,7 @@ else:
 
 
 class TFAutoModelTest(unittest.TestCase):
+    @pytest.mark.slow
     def test_model_from_pretrained(self):
         import h5py
         self.assertTrue(h5py.version.hdf5_version.startswith("1.10"))
@@ -53,6 +54,7 @@ class TFAutoModelTest(unittest.TestCase):
             self.assertIsNotNone(model)
             self.assertIsInstance(model, TFBertModel)
 
+    @pytest.mark.slow
     def test_lmhead_model_from_pretrained(self):
         logging.basicConfig(level=logging.INFO)
         # for model_name in list(TF_BERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
@@ -65,6 +67,7 @@ class TFAutoModelTest(unittest.TestCase):
             self.assertIsNotNone(model)
             self.assertIsInstance(model, TFBertForMaskedLM)
 
+    @pytest.mark.slow
     def test_sequence_classification_model_from_pretrained(self):
         logging.basicConfig(level=logging.INFO)
         # for model_name in list(TF_BERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
@@ -77,6 +80,7 @@ class TFAutoModelTest(unittest.TestCase):
             self.assertIsNotNone(model)
             self.assertIsInstance(model, TFBertForSequenceClassification)
 
+    @pytest.mark.slow
     def test_question_answering_model_from_pretrained(self):
         logging.basicConfig(level=logging.INFO)
         # for model_name in list(TF_BERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:

From f3776df0f3daca86634862fe3ba7da6ae2b9a663 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Mon, 2 Dec 2019 15:47:00 +0100
Subject: [PATCH 169/505] WIP debugging

---
 transformers/modeling_t5.py | 61 +++++++++++++++++++++++++++----------
 1 file changed, 45 insertions(+), 16 deletions(-)

diff --git a/transformers/modeling_t5.py b/transformers/modeling_t5.py
index 2a74333d31..1bf55611a2 100644
--- a/transformers/modeling_t5.py
+++ b/transformers/modeling_t5.py
@@ -132,6 +132,21 @@ def load_tf_weights_in_t5(model, config, tf_checkpoint_path):
 # - PreTrainedModel for the models (it-self a sub-class of torch.nn.Module)
 ####################################################
 
+class T5LayerNorm(nn.Module):
+    def __init__(self, hidden_size, eps=1e-6):
+        """ Construct a layernorm module in the T5 style
+            No bias and no substraction of mean.
+        """
+        super(T5LayerNorm, self).__init__()
+        self.weight = nn.Parameter(torch.ones(hidden_size))
+        self.variance_epsilon = eps
+
+    def forward(self, x):
+        variance = x.pow(2).mean(-1, keepdim=True)
+        x = x / torch.sqrt(variance + self.variance_epsilon)
+        return self.weight * x
+
+
 class T5DenseReluDense(nn.Module):
     def __init__(self, config):
         super(T5DenseReluDense, self).__init__()
@@ -151,7 +166,7 @@ class T5LayerFF(nn.Module):
     def __init__(self, config):
         super(T5LayerFF, self).__init__()
         self.DenseReluDense = T5DenseReluDense(config)
-        self.layer_norm = nn.LayerNorm(config.d_model, eps=config.layer_norm_epsilon)
+        self.layer_norm = T5LayerNorm(config.d_model, eps=config.layer_norm_epsilon)
         self.dropout = nn.Dropout(config.dropout_rate)
 
     def forward(self, hidden_states):
@@ -316,13 +331,14 @@ class T5Attention(nn.Module):
             cache[self.layer_id] = (k, v)
 
         # q = q / math.sqrt(dim_per_head)                                     # No scaling in T5
-        scores = torch.matmul(q, k.transpose(2, 3))                           # (bs, n_heads, qlen, klen)
+        scores = torch.einsum('bnqd,bnkd->bnqk', q, k)                        # (bs, n_heads, qlen, klen)
 
         if position_bias is None:
             if not self.has_relative_attention_bias:
                 raise ValueError("No position_bias provided and no weights to compute position_bias")
             position_bias = self.compute_bias(qlen, klen)
         scores += position_bias
+        special_out = position_bias
 
         if mask is not None:
             scores += mask
@@ -346,14 +362,14 @@ class T5Attention(nn.Module):
             outputs = outputs + (weights,)
         if self.has_relative_attention_bias:
             outputs = outputs + (position_bias,)
-        return outputs
+        return outputs + (special_out,)
 
 
 class T5LayerSelfAttention(nn.Module):
     def __init__(self, config, has_relative_attention_bias=False):
         super(T5LayerSelfAttention, self).__init__()
         self.SelfAttention = T5Attention(config, has_relative_attention_bias=has_relative_attention_bias)
-        self.layer_norm = nn.LayerNorm(config.d_model, eps=config.layer_norm_epsilon)
+        self.layer_norm = T5LayerNorm(config.d_model, eps=config.layer_norm_epsilon)
         self.dropout = nn.Dropout(config.dropout_rate)
 
     def forward(self, hidden_states, attention_mask=None, position_bias=None, head_mask=None):
@@ -363,16 +379,18 @@ class T5LayerSelfAttention(nn.Module):
                                               position_bias=position_bias,
                                               head_mask=head_mask)
         y = attention_output[0]
+        special_out = attention_output[-1]
+        attention_output = attention_output[:-1]
         layer_output = hidden_states + self.dropout(y)
         outputs = (layer_output,) + attention_output[1:]  # add attentions if we output them
-        return outputs
+        return outputs + (special_out,)
 
 
 class T5LayerCrossAttention(nn.Module):
     def __init__(self, config, has_relative_attention_bias=False):
         super(T5LayerCrossAttention, self).__init__()
         self.EncDecAttention = T5Attention(config, has_relative_attention_bias=has_relative_attention_bias)
-        self.layer_norm = nn.LayerNorm(config.d_model, eps=config.layer_norm_epsilon)
+        self.layer_norm = T5LayerNorm(config.d_model, eps=config.layer_norm_epsilon)
         self.dropout = nn.Dropout(config.dropout_rate)
 
     def forward(self, hidden_states, kv, attention_mask=None, position_bias=None, head_mask=None):
@@ -408,7 +426,8 @@ class T5Block(nn.Module):
                                                 position_bias=position_bias,
                                                 head_mask=head_mask)
         hidden_states = self_attention_outputs[0]
-        outputs = self_attention_outputs[1:]  # Keep self-attention outputs and relative position weights
+        special_out = self_attention_outputs[-1]
+        outputs = self_attention_outputs[1:-1]  # Keep self-attention outputs and relative position weights
 
         if not self.is_decoder:
             hidden_states = self.layer[1](hidden_states)
@@ -423,7 +442,7 @@ class T5Block(nn.Module):
             hidden_states = self.layer[2](hidden_states)
 
         outputs = (hidden_states,) + outputs  # add attentions if we output them
-        return outputs  # hidden-states, (self-attention weights), (self-attention position bias), (cross-attention weights), (cross-attention position bias)
+        return outputs + (special_out,)  # hidden-states, (self-attention weights), (self-attention position bias), (cross-attention weights), (cross-attention position bias)
 
 
 class T5PreTrainedModel(PreTrainedModel):
@@ -438,8 +457,7 @@ class T5PreTrainedModel(PreTrainedModel):
     def _init_weights(self, module):
         """ Initialize the weights """
         factor = self.config.initializer_factor  # Used for testing weights initialization
-        if isinstance(module, nn.LayerNorm):
-            module.bias.data.zero_()
+        if isinstance(module, T5LayerNorm):
             module.weight.data.fill_(factor*1.0)
         elif isinstance(module, (T5Model, T5WithLMHeadModel)):
             # Mesh TensorFlow embeddings initialization
@@ -478,7 +496,7 @@ class T5Stack(T5PreTrainedModel):
 
         self.block = nn.ModuleList([T5Block(config, has_relative_attention_bias=bool(i == 0))
                                     for i in range(config.num_layers)])
-        self.final_layer_norm = nn.LayerNorm(config.d_model, eps=config.layer_norm_epsilon)
+        self.final_layer_norm = T5LayerNorm(config.d_model, eps=config.layer_norm_epsilon)
         self.dropout = nn.Dropout(config.dropout_rate)
 
         self.init_weights()
@@ -515,11 +533,11 @@ class T5Stack(T5PreTrainedModel):
 
         # Since attention_mask is 1.0 for positions we want to attend and 0.0 for
         # masked positions, this operation will create a tensor which is 0.0 for
-        # positions we want to attend and -10000.0 for masked positions.
+        # positions we want to attend and -1e9 for masked positions.
         # Since we are adding it to the raw scores before the softmax, this is
         # effectively the same as removing these entirely.
         extended_attention_mask = extended_attention_mask.to(dtype=next(self.parameters()).dtype)  # fp16 compatibility
-        extended_attention_mask = (1.0 - extended_attention_mask) * -10000.0
+        extended_attention_mask = (1.0 - extended_attention_mask) * -1e9
 
         if self.is_decoder:
             # If a 2D ou 3D attention mask is provided for the cross-attention
@@ -530,7 +548,7 @@ class T5Stack(T5PreTrainedModel):
                 encoder_extended_attention_mask = encoder_attention_mask[:, None, None, :]
 
             encoder_extended_attention_mask = encoder_extended_attention_mask.to(dtype=next(self.parameters()).dtype)  # fp16 compatibility
-            encoder_extended_attention_mask = (1.0 - encoder_extended_attention_mask) * -10000.0
+            encoder_extended_attention_mask = (1.0 - encoder_extended_attention_mask) * -1e9
         else:
             encoder_extended_attention_mask = None
 
@@ -553,6 +571,8 @@ class T5Stack(T5PreTrainedModel):
         all_attentions = ()
         position_bias = None
         encoder_decoder_position_bias = None
+
+        hidden_states = self.dropout(hidden_states)
         for i, layer_module in enumerate(self.block):
             if self.output_hidden_states:
                 all_hidden_states = all_hidden_states + (hidden_states,)
@@ -564,6 +584,8 @@ class T5Stack(T5PreTrainedModel):
                                          encoder_attention_mask=encoder_extended_attention_mask,
                                          encoder_decoder_position_bias=encoder_decoder_position_bias,
                                          head_mask=head_mask[i])
+            if i == 0:
+                special_out = layer_outputs[-1]
             # layer_outputs is a tuple with:
             # hidden-states, (self-attention weights), (self-attention position bias), (cross-attention weights), (cross-attention position bias)
             hidden_states = layer_outputs[0]
@@ -588,7 +610,7 @@ class T5Stack(T5PreTrainedModel):
             outputs = outputs + (all_hidden_states,)
         if self.output_attentions:
             outputs = outputs + (all_attentions,)
-        return outputs  # last-layer hidden state, (all hidden states), (all attentions)
+        return outputs + (special_out,)  # last-layer hidden state, (all hidden states), (all attentions)
 
 
 T5_START_DOCSTRING = r"""    The T5 model was proposed in
@@ -707,9 +729,16 @@ class T5Model(T5PreTrainedModel):
 
         # Encode if needed (training, first prediction pass)
         encoder_hidden_states = kwargs_encoder.pop("hidden_states", None)
+        encoder_attention_mask = kwargs_encoder.get("attention_mask", None)
         if encoder_hidden_states is None:
             encoder_inputs_ids = kwargs_encoder.pop("input_ids")
             hidden_states = self.shared(encoder_inputs_ids)  # Convert inputs in embeddings
+
+            if encoder_attention_mask is not None:
+                # Apply masking
+                encoder_attention_mask = (encoder_attention_mask != 0).to(hidden_states)
+                hidden_states = hidden_states * encoder_attention_mask.unsqueeze(-1)
+
             encoder_outputs = self.encoder(hidden_states, **kwargs_encoder)
             encoder_hidden_states = encoder_outputs[0]
         else:
@@ -719,7 +748,7 @@ class T5Model(T5PreTrainedModel):
         decoder_inputs_ids = kwargs_decoder.pop("input_ids")
         hidden_states = self.shared(decoder_inputs_ids)  # Convert inputs in embeddings
         kwargs_decoder["encoder_hidden_states"] = encoder_hidden_states
-        kwargs_decoder["encoder_attention_mask"] = kwargs_encoder.get("attention_mask", None)
+        kwargs_decoder["encoder_attention_mask"] = encoder_attention_mask
         decoder_outputs = self.decoder(hidden_states, **kwargs_decoder)
 
         return decoder_outputs + encoder_outputs

From b3d834ae11381ca493da97f717d77d185ca7d780 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Mon, 2 Dec 2019 15:01:52 -0500
Subject: [PATCH 170/505] Reorganize ALBERT conversion script

---
 transformers/modeling_albert.py | 30 ++++++++++++++++++++++++++----
 1 file changed, 26 insertions(+), 4 deletions(-)

diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index 5b7b2d3900..49d120ffae 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -68,14 +68,36 @@ def load_tf_weights_in_albert(model, config, tf_checkpoint_path):
     
     for name, array in zip(names, arrays):
         original_name = name
+
+        # If saved from the TF HUB module
+        name = name.replace("module/", "")
+
+        # Renaming and simplifying
         name = name.replace("ffn_1", "ffn")
-        name = name.replace("/bert/", "/albert/")
-        name = name.replace("ffn/intermediate/output", "ffn_output")
+        name = name.replace("bert/", "albert/")
         name = name.replace("attention_1", "attention")   
-        name = name.replace("cls/predictions", "predictions")
         name = name.replace("transform/", "")
         name = name.replace("LayerNorm_1", "full_layer_layer_norm")    
-        name = name.replace("LayerNorm", "attention/LayerNorm")    
+        name = name.replace("LayerNorm", "attention/LayerNorm")   
+        name = name.replace("transformer/", "")
+
+        # The feed forward layer had an 'intermediate' step which has been abstracted away
+        name = name.replace("intermediate/dense/", "")
+        name = name.replace("ffn/intermediate/output/dense/", "ffn_output/")
+
+        # ALBERT attention was split between self and output which have been abstracted away
+        name = name.replace("/output/", "/")
+        name = name.replace("/self/", "/")
+
+        # The pooler is a linear layer
+        name = name.replace("pooler/dense", "pooler")
+
+        # The classifier was simplified to predictions from cls/predictions
+        name = name.replace("cls/predictions", "predictions")
+        name = name.replace("predictions/attention", "predictions")
+
+        # Naming was changed to be more explicit
+        name = name.replace("embeddings/attention", "embeddings")    
         name = name.replace("inner_group_", "albert_layers/") 
         name = name.replace("group_", "albert_layer_groups/")    
         name = name.split('/')

From e85855f2c408f65a4aaf5d15baab6ca90fd26050 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Mon, 2 Dec 2019 18:00:19 -0500
Subject: [PATCH 171/505] Fix ALBERT exports with pretraining + sp classifier;
 Fix naming for ALBERT TF models

---
 transformers/modeling_albert.py    | 17 ++++++++++++++++-
 transformers/modeling_tf_albert.py |  8 ++++----
 2 files changed, 20 insertions(+), 5 deletions(-)

diff --git a/transformers/modeling_albert.py b/transformers/modeling_albert.py
index 49d120ffae..0f67bf8f36 100644
--- a/transformers/modeling_albert.py
+++ b/transformers/modeling_albert.py
@@ -99,8 +99,23 @@ def load_tf_weights_in_albert(model, config, tf_checkpoint_path):
         # Naming was changed to be more explicit
         name = name.replace("embeddings/attention", "embeddings")    
         name = name.replace("inner_group_", "albert_layers/") 
-        name = name.replace("group_", "albert_layer_groups/")    
+        name = name.replace("group_", "albert_layer_groups/")   
+
+        # Classifier
+        if len(name.split("/")) == 1 and ("output_bias" in name or "output_weights" in name):
+            name = "classifier/" + name
+
+        # No ALBERT model currently handles the next sentence prediction task 
+        if "seq_relationship" in name:
+            continue
+
         name = name.split('/')
+
+        # Ignore the gradients applied by the LAMB/ADAM optimizers.
+        if "adam_m" in name or "adam_v" in name or "global_step" in name:
+            logger.info("Skipping {}".format("/".join(name)))
+            continue
+
         pointer = model
         for m_name in name:
             if re.fullmatch(r'[A-Za-z]+_\d+', m_name):
diff --git a/transformers/modeling_tf_albert.py b/transformers/modeling_tf_albert.py
index 164dc74320..d1650d41a8 100644
--- a/transformers/modeling_tf_albert.py
+++ b/transformers/modeling_tf_albert.py
@@ -31,10 +31,10 @@ import logging
 logger = logging.getLogger(__name__)
 
 TF_ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
-    'albert-base-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-base-tf_model.h5",
-    'albert-large-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-large-tf_model.h5",
-    'albert-xlarge-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xlarge-tf_model.h5",
-    'albert-xxlarge-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xxlarge-tf_model.h5",
+    'albert-base-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-base-v1-tf_model.h5",
+    'albert-large-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-large-v1-tf_model.h5",
+    'albert-xlarge-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xlarge-v1-tf_model.h5",
+    'albert-xxlarge-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xxlarge-v1-tf_model.h5",
     'albert-base-v2': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-base-v2-tf_model.h5",
     'albert-large-v2': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-large-v2-tf_model.h5",
     'albert-xlarge-v2': "https://s3.amazonaws.com/models.huggingface.co/bert/albert-xlarge-v2-tf_model.h5",

From fbaf05bd92249b6dd961f5f8d60eb0892c541ac8 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Mon, 2 Dec 2019 18:23:00 -0500
Subject: [PATCH 172/505] Remove annoying tokenization message

---
 transformers/tokenization_utils.py | 2 --
 1 file changed, 2 deletions(-)

diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index 60df822677..5d683629f0 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -981,7 +981,6 @@ class PreTrainedTokenizer(object):
         return (ids, pair_ids, overflowing_tokens)
 
     def create_token_type_ids_from_sequences(self, token_ids_0, token_ids_1=None):
-        logger.warning("This tokenizer does not make use of special tokens.")
         if token_ids_1 is None:
             return len(token_ids_0) * [0]
         return [0] * len(token_ids_0) + [1] * len(token_ids_1)
@@ -994,7 +993,6 @@ class PreTrainedTokenizer(object):
             single sequence: <s> X </s>
             pair of sequences: <s> A </s></s> B </s>
         """
-        logger.warning("This tokenizer does not make use of special tokens. Input is returned with no modification.")
         if token_ids_1 is None:
             return token_ids_0
         return token_ids_0 + token_ids_1

From 66fc8d25a5de43c08baeea8b22b9bcf57346c325 Mon Sep 17 00:00:00 2001
From: Juha Kiili <juha@juhakiili.com>
Date: Tue, 3 Dec 2019 10:49:50 +0200
Subject: [PATCH 173/505] Change ref to original GLUE downloader script

---
 utils/download_glue_data.py | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/utils/download_glue_data.py b/utils/download_glue_data.py
index f676a71c76..de8cfa9e73 100644
--- a/utils/download_glue_data.py
+++ b/utils/download_glue_data.py
@@ -1,7 +1,5 @@
 ''' Script for downloading all GLUE data.
-
-Original source: https://github.com/kamalkraj/ALBERT-TF2.0/blob/fa90194e5fe729dbb19f32ac29c8d6d6372c0f93/download_glue_data.py
-Original license: https://github.com/kamalkraj/ALBERT-TF2.0/blob/fa90194e5fe729dbb19f32ac29c8d6d6372c0f93/LICENSE (Apache-2.0)
+Original source: https://gist.github.com/W4ngatang/60c2bdb54d156a41194446737ce03e2e
 
 Note: for legal reasons, we are unable to host MRPC.
 You can either use the version hosted by the SentEval team, which is already tokenized, 

From 572c24cfa25c167e48dd08a7d2429afeb69acbe0 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Tue, 5 Nov 2019 15:34:37 +0000
Subject: [PATCH 174/505] PPLM (squashed)

Co-authored-by: piero <piero@uber.com>
Co-authored-by: Rosanne Liu <mimosavvy@gmail.com>
---
 examples/run_pplm.py | 782 +++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 782 insertions(+)
 create mode 100644 examples/run_pplm.py

diff --git a/examples/run_pplm.py b/examples/run_pplm.py
new file mode 100644
index 0000000000..30853c68c3
--- /dev/null
+++ b/examples/run_pplm.py
@@ -0,0 +1,782 @@
+# coding=utf-8
+# Copyright 2018 The Uber AI Team Authors.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+# TODO: add code for training a custom discriminator
+
+"""
+Example command with bag of words:
+python examples/run_pplm.py -B space --cond_text "The president" --length 100 --gamma 1.5 --num_iterations 3 --num_samples 10 --stepsize 0.01 --window_length 5 --kl_scale 0.01 --gm_scale 0.95
+
+Example command with discriminator:
+python examples/run_pplm.py -D sentiment --label_class 3 --cond_text "The lake" --length 10 --gamma 1.0 --num_iterations 30 --num_samples 10 --stepsize 0.01 --kl_scale 0.01 --gm_scale 0.95
+"""
+
+import argparse
+from operator import add
+from typing import List, Optional, Tuple, Union
+
+import numpy as np
+import torch
+import torch.nn.functional as F
+from torch.autograd import Variable
+from tqdm import trange
+
+from transformers import GPT2Tokenizer
+from transformers.file_utils import cached_path
+from transformers.modeling_gpt2 import GPT2LMHeadModel
+
+PPLM_BOW = 1
+PPLM_DISCRIM = 2
+PPLM_BOW_DISCRIM = 3
+SMALL_CONST = 1e-15
+TOKENIZER = GPT2Tokenizer.from_pretrained("gpt2-medium")
+
+BAG_OF_WORDS_ARCHIVE_MAP = {
+    'kitchen': "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/bow/kitchen.txt",
+    'legal': "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/bow/legal.txt",
+    'military': "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/bow/military.txt",
+    'monsters': "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/bow/monsters.txt",
+    'politics': "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/bow/politics.txt",
+    'positive_words': "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/bow/positive_words.txt",
+    'religion': "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/bow/religion.txt",
+    'science': "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/bow/science.txt",
+    'space': "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/bow/space.txt",
+    'technology': "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/bow/technology.txt",
+}
+
+DISCRIMINATOR_MODELS_PARAMS = {
+    "clickbait": {
+        "url": "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/discriminators/clickbait_classifierhead.pt",
+        "class_size": 2,
+        "embed_size": 1024,
+        "class_vocab": {"non_clickbait": 0, "clickbait": 1},
+        "default_class": 1,
+    },
+    "sentiment": {
+        "url": "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/discriminators/sentiment_classifierhead.pt",
+        "class_size": 5,
+        "embed_size": 1024,
+        "class_vocab": {"very_positive": 2, "very_negative": 3},
+        "default_class": 3,
+    },
+    "toxicity": {
+        "url": "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/discriminators/toxicity_classifierhead.pt",
+        "class_size": 2,
+        "embed_size": 1024,
+        "class_vocab": {"non_toxic": 0, "toxic": 1},
+        "default_class": 0,
+    },
+}
+
+
+class ClassificationHead(torch.nn.Module):
+    """ Classification Head for the transformer """
+
+    def __init__(self, class_size=5, embed_size=2048):
+        super(ClassificationHead, self).__init__()
+        self.class_size = class_size
+        self.embed_size = embed_size
+        # self.mlp1 = torch.nn.Linear(embed_size, embed_size)
+        # self.mlp2 = (torch.nn.Linear(embed_size, class_size))
+        self.mlp = torch.nn.Linear(embed_size, class_size)
+
+    def forward(self, hidden_state):
+        # hidden_state = F.relu(self.mlp1(hidden_state))
+        # hidden_state = self.mlp2(hidden_state)
+        logits = self.mlp(hidden_state)
+        return logits
+
+
+def to_var(x, requires_grad=False, volatile=False):
+    if torch.cuda.is_available():
+        x = x.cuda()
+    return Variable(x, requires_grad=requires_grad, volatile=volatile)
+
+
+def top_k_filter(logits, k, probs=False):
+    """
+    Masks everything but the k top entries as -infinity (1e10).
+    Used to mask logits such that e^-infinity -> 0 won't contribute to the
+    sum of the denominator.
+    """
+    if k <= 0:
+        return logits
+
+    else:
+        values = torch.topk(logits, k)[0]
+        batch_mins = values[:, -1].view(-1, 1).expand_as(logits)
+
+        if probs:
+            return torch.where(
+                logits < batch_mins,
+                torch.ones_like(logits) * 0.0,
+                logits
+            )
+
+        return torch.where(
+            logits < batch_mins,
+            torch.ones_like(logits) * -1e10,
+            logits
+        )
+
+
+def perturb_past(
+        past,
+        model,
+        last,
+        unpert_past=None,
+        unpert_logits=None,
+        accumulated_hidden=None,
+        grad_norms=None,
+        stepsize=0.01,
+        classifier=None,
+        label_class=None,
+        one_hot_bows_vectors=None,
+        loss_type=0,
+        num_iterations=3,
+        kl_scale=0.01,
+        window_length=0,
+        horizon_length=1,
+        decay=False,
+        gamma=1.5,
+):
+    # initializie perturbation accumulator
+    grad_accumulator = [
+        (np.zeros(p.shape).astype("float32"))
+        for p in past
+    ]
+
+    if accumulated_hidden is None:
+        accumulated_hidden = 0
+
+    if decay:
+        decay_mask = torch.arange(
+            0.0,
+            1.0 + SMALL_CONST,
+            1.0 / (window_length)
+        )[1:]
+    else:
+        decay_mask = 1.0
+
+    # TODO fix this comment (SUMANTH)
+    # generate a mask if perturbated gradient is based on a past window
+    _, _, _, curr_length, _ = past[0].shape
+    if curr_length > window_length and window_length > 0:
+        ones_key_val_shape = (
+                tuple(past[0].shape[:-2])
+                + tuple([window_length])
+                + tuple(past[0].shape[-1:])
+        )
+
+        zeros_key_val_shape = (
+                tuple(past[0].shape[:-2])
+                + tuple([curr_length - window_length])
+                + tuple(past[0].shape[-1:])
+        )
+
+        ones_mask = torch.ones(ones_key_val_shape)
+        ones_mask = decay_mask * ones_mask.permute(0, 1, 2, 4, 3)
+        ones_mask = ones_mask.permute(0, 1, 2, 4, 3)
+
+        window_mask = torch.cat(
+            (ones_mask, torch.zeros(zeros_key_val_shape)),
+            dim=-2
+        ).cuda()
+
+    else:
+        window_mask = torch.ones_like(past[0]).cuda()
+
+    # accumulate perturbations for num_iterations
+    loss_per_iter = []
+    for i in range(num_iterations):
+        print("Iteration ", i + 1)
+
+        curr_perturbation = [
+            to_var(torch.from_numpy(p_), requires_grad=True)
+            for p_ in grad_accumulator
+        ]
+
+        # Compute hidden using perturbed past
+        curr_pert_past = list(map(add, past, curr_perturbation))
+        all_logits, _, all_hidden = model(last, past=curr_pert_past)
+        hidden = all_hidden[-1]
+        accumulated_hidden += torch.sum(hidden, dim=1).detach()
+        logits = all_logits[:, -1, :]
+        probs = F.softmax(logits, dim=-1)
+
+        # compute loss
+        bow_loss = 0.0
+        discrim_loss = 0.0
+        kl_loss = 0.0
+
+        if loss_type == PPLM_BOW or loss_type == PPLM_BOW_DISCRIM:
+            for one_hot_bow in one_hot_bows_vectors:
+                bow_logits = torch.mm(probs, torch.t(one_hot_bow))
+                bow_loss += -torch.log(torch.sum(bow_logits))
+            print(" pplm_bow_loss:", bow_loss.data.cpu().numpy())
+
+        if loss_type == PPLM_DISCRIM or loss_type == PPLM_BOW_DISCRIM:
+            ce_loss = torch.nn.CrossEntropyLoss()
+            # TODO all there are for (SUMANTH)
+            # TODO why we need to do this assignment and not just using unpert_past?
+            curr_unpert_past = unpert_past
+            # Get the model's token embeddings in order to compute our own embeds from curr_probs:
+            wte = model.resize_token_embeddings()
+            # TODO i is never used, why do we need to do this i times instead multiplying
+            #   torch.sum(unpert_hidden, dim=1) * horizon_length?
+            for i in range(horizon_length):
+                # TODO the next two lines can be done only one time, and why not using probs instead as they do not change at each iteration?
+                curr_probs = F.softmax(logits, dim=-1)  # get softmax
+                curr_probs = torch.unsqueeze(curr_probs, dim=1)
+                inputs_embeds = torch.matmul(curr_probs, wte.weight.data)
+                _, curr_unpert_past, curr_all_hidden = model(
+                    past=curr_unpert_past,
+                    inputs_embeds=inputs_embeds
+                )
+                # get expected hidden states
+                unpert_hidden = curr_all_hidden[1]
+                accumulated_hidden += torch.sum(unpert_hidden, dim=1)
+
+            prediction = classifier(
+                accumulated_hidden / (curr_length + 1 + horizon_length)
+            )
+
+            label = torch.tensor([label_class], device="cuda", dtype=torch.long)
+            discrim_loss += ce_loss(prediction, label)
+            print(" pplm_discrim_loss:", discrim_loss.data.cpu().numpy())
+
+        if kl_scale > 0.0:
+            unpert_probs = F.softmax(unpert_logits[:, -1, :], dim=-1)
+            unpert_probs = (
+                    unpert_probs + SMALL_CONST *
+                    (unpert_probs <= SMALL_CONST).type(
+                        torch.FloatTensor
+                    ).cuda().detach()
+            )
+
+            correction = SMALL_CONST * (probs <= SMALL_CONST).type(
+                torch.FloatTensor
+            ).cuda().detach()
+            corrected_probs = probs + correction.detach()
+            kl_loss += kl_scale * (
+                (corrected_probs * (corrected_probs / unpert_probs).log()).sum()
+            )
+            print(' kl_loss', (kl_loss).data.cpu().numpy())
+
+        loss = bow_loss + discrim_loss + kl_loss
+        loss_per_iter.append(loss.data.cpu().numpy())
+        print(' pplm_loss', (loss - kl_loss).data.cpu().numpy())
+
+        # compute gradients
+        loss.backward(retain_graph=True)
+
+        # calculate gradient norms
+        if grad_norms is not None and loss_type == PPLM_BOW:
+            grad_norms = [
+                torch.max(grad_norms[index], torch.norm(p_.grad * window_mask))
+                for index, p_ in enumerate(curr_perturbation)
+            ]
+        else:
+            grad_norms = [
+                (torch.norm(p_.grad * window_mask) + SMALL_CONST)
+                for index, p_ in enumerate(curr_perturbation)
+            ]
+
+        # normalize gradients
+        grad = [
+            -stepsize
+            * (p_.grad * window_mask / grad_norms[
+                index] ** gamma).data.cpu().numpy()
+            for index, p_ in enumerate(curr_perturbation)
+        ]
+
+        # accumulate gradients
+        grad_accumulator = list(map(add, grad, grad_accumulator))
+
+        # reset gradients, just to make sure
+        for p_ in curr_perturbation:
+            p_.grad.data.zero_()
+
+        # removing past from the graph
+        new_past = []
+        for p_ in past:
+            new_past.append(p_.detach())
+        past = new_past
+
+    # apply the accumulated perturbations to the past
+    grad_accumulator = [
+        to_var(torch.from_numpy(p_), requires_grad=True)
+        for p_ in grad_accumulator
+    ]
+    pert_past = list(map(add, past, grad_accumulator))
+
+    return pert_past, accumulated_hidden, grad_norms, loss_per_iter
+
+
+def get_classifier(
+        name: Optional[str], label_class: Union[str, int], device: Union[str, torch.device]
+) -> Tuple[Optional[ClassificationHead], Optional[int]]:
+    if name is None:
+        return None, None
+
+    params = DISCRIMINATOR_MODELS_PARAMS[name]
+    classifier = ClassificationHead(
+        class_size=params['class_size'],
+        embed_size=params['embed_size']
+    ).to(device)
+    resolved_archive_file = cached_path(params["url"])
+    classifier.load_state_dict(torch.load(resolved_archive_file, map_location=device))
+    classifier.eval()
+
+    if isinstance(label_class, str):
+        if label_class in params["class_vocab"]:
+            label_id = params["class_vocab"][label_class]
+        else:
+            label_id = params["default_class"]
+            print("label_class {} not in class_vocab".format(label_class))
+            print("available values are: {}".format(params["class_vocab"]))
+            print("using default class {}".format(label_id))
+
+    elif isinstance(label_class, int):
+        if label_class in set(params["class_vocab"].values()):
+            label_id = label_class
+        else:
+            label_id = params["default_class"]
+            print("label_class {} not in class_vocab".format(label_class))
+            print("available values are: {}".format(params["class_vocab"]))
+            print("using default class {}".format(label_id))
+
+    else:
+        label_id = params["default_class"]
+
+    return classifier, label_id
+
+
+def get_bag_of_words_indices(bag_of_words_ids_or_paths: List[str]) -> List[List[List[int]]]:
+    bow_indices = []
+    for id_or_path in bag_of_words_ids_or_paths:
+        if id_or_path in BAG_OF_WORDS_ARCHIVE_MAP:
+            filepath = cached_path(BAG_OF_WORDS_ARCHIVE_MAP[id_or_path])
+        else:
+            filepath = id_or_path
+        with open(filepath, "r") as f:
+            words = f.read().split("\n")
+        bow_indices.append([TOKENIZER.encode(word) for word in words])
+    return bow_indices
+
+
+def build_bows_one_hot_vectors(bow_indices):
+    if bow_indices is None:
+        return None
+
+    one_hot_bows_vectors = []
+    for single_bow in bow_indices:
+        single_bow = list(filter(lambda x: len(x) <= 1, single_bow))
+        single_bow = torch.tensor(single_bow).cuda()
+        num_words = single_bow.shape[0]
+        one_hot_bow = torch.zeros(num_words, TOKENIZER.vocab_size).cuda()
+        one_hot_bow.scatter_(1, single_bow, 1)
+        one_hot_bows_vectors.append(one_hot_bow)
+    return one_hot_bows_vectors
+
+
+def full_text_generation(
+        model,
+        context=None,
+        num_samples=1,
+        device="cuda",
+        sample=True,
+        discrim=None,
+        label_class=None,
+        bag_of_words=None,
+        length=100,
+        grad_length=10000,
+        stepsize=0.02,
+        num_iterations=3,
+        temperature=1.0,
+        gm_scale=0.9,
+        kl_scale=0.01,
+        top_k=10,
+        window_length=0,
+        horizon_length=1,
+        decay=False,
+        gamma=1.5,
+        **kwargs
+):
+    classifier, class_id = get_classifier(
+        discrim,
+        label_class,
+        device
+    )
+
+    bow_indices = []
+    if bag_of_words:
+        bow_indices = get_bag_of_words_indices(bag_of_words.split(";"))
+
+    if bag_of_words and classifier:
+        print("Both PPLM-BoW and PPLM-Discrim are on. This is not optimized.")
+        loss_type = PPLM_BOW_DISCRIM
+
+    elif bag_of_words:
+        loss_type = PPLM_BOW
+        print("Using PPLM-BoW")
+
+    elif classifier is not None:
+        loss_type = PPLM_DISCRIM
+        print("Using PPLM-Discrim")
+
+    else:
+        raise Exception("Specify either --bag_of_words (-B) or --discrim (-D)")
+
+    unpert_gen_tok_text, _, _ = generate_text_pplm(
+        model=model,
+        context=context,
+        device=device,
+        length=length,
+        perturb=False
+    )
+    torch.cuda.empty_cache()
+
+    pert_gen_tok_texts = []
+    discrim_losses = []
+    losses_in_time = []
+
+    for i in range(num_samples):
+        pert_gen_tok_text, discrim_loss, loss_in_time = generate_text_pplm(
+            model=model,
+            context=context,
+            device=device,
+            sample=sample,
+            perturb=True,
+            bow_indices=bow_indices,
+            classifier=classifier,
+            label_class=class_id,
+            loss_type=loss_type,
+            length=length,
+            grad_length=grad_length,
+            stepsize=stepsize,
+            num_iterations=num_iterations,
+            temperature=temperature,
+            gm_scale=gm_scale,
+            kl_scale=kl_scale,
+            top_k=top_k,
+            window_length=window_length,
+            horizon_length=horizon_length,
+            decay=decay,
+            gamma=gamma,
+        )
+        pert_gen_tok_texts.append(pert_gen_tok_text)
+        if classifier is not None:
+            discrim_losses.append(discrim_loss.data.cpu().numpy())
+        losses_in_time.append(loss_in_time)
+
+    torch.cuda.empty_cache()
+
+    return unpert_gen_tok_text, pert_gen_tok_texts, discrim_losses, losses_in_time
+
+
+def generate_text_pplm(
+        model,
+        context=None,
+        past=None,
+        device="cuda",
+        sample=True,
+        perturb=True,
+        classifier=None,
+        label_class=None,
+        bow_indices=None,
+        loss_type=0,
+        length=100,
+        grad_length=10000,
+        stepsize=0.02,
+        num_iterations=3,
+        temperature=1.0,
+        gm_scale=0.9,
+        kl_scale=0.01,
+        top_k=10,
+        window_length=0,
+        horizon_length=1,
+        decay=False,
+        gamma=1.5,
+):
+    output_so_far = (
+        torch.tensor(context, device=device, dtype=torch.long).unsqueeze(0)
+        if context
+        else None
+    )
+
+    # collect one hot vectors for bags of words
+    one_hot_bows_vectors = build_bows_one_hot_vectors(bow_indices)
+
+    grad_norms = None
+    last = None
+    unpert_discrim_loss = 0
+    loss_in_time = []
+    for i in trange(length, ascii=True):
+
+        # Get past/probs for current output, except for last word
+        # Note that GPT takes 2 inputs: past + current_token
+
+        # run model forward to obtain unperturbed
+        if past is None and output_so_far is not None:
+            last = output_so_far[:, -1:]
+            if output_so_far.shape[1] > 1:
+                _, past, _ = model(output_so_far[:, :-1])
+
+            unpert_logits, unpert_past, unpert_all_hidden = model(output_so_far)
+            unpert_last_hidden = unpert_all_hidden[-1]
+
+        else:
+            unpert_logits, unpert_past, unpert_all_hidden = model(output_so_far)
+            unpert_last_hidden = unpert_all_hidden[-1]
+
+        # check if we are abowe grad max length
+        if i >= grad_length:
+            current_stepsize = stepsize * 0
+        else:
+            current_stepsize = stepsize
+
+        # modify the past if necessary
+        if not perturb or num_iterations == 0:
+            pert_past = past
+
+        else:
+            accumulated_hidden = unpert_last_hidden[:, :-1, :]
+            accumulated_hidden = torch.sum(accumulated_hidden, dim=1)
+
+            if past is not None:
+                pert_past, _, grad_norms, loss_this_iter = perturb_past(
+                    past,
+                    model,
+                    last,
+                    unpert_past=unpert_past,
+                    unpert_logits=unpert_logits,
+                    accumulated_hidden=accumulated_hidden,
+                    grad_norms=grad_norms,
+                    stepsize=current_stepsize,
+                    classifier=classifier,
+                    label_class=label_class,
+                    one_hot_bows_vectors=one_hot_bows_vectors,
+                    loss_type=loss_type,
+                    num_iterations=num_iterations,
+                    kl_scale=kl_scale,
+                    window_length=window_length,
+                    horizon_length=horizon_length,
+                    decay=decay,
+                    gamma=gamma,
+                )
+                loss_in_time.append(loss_this_iter)
+            else:
+                pert_past = past
+
+        pert_logits, past, pert_all_hidden = model(last, past=pert_past)
+        pert_logits = pert_logits[:, -1, :] / temperature
+        pert_probs = F.softmax(pert_logits, dim=-1)
+
+        # compute the discriminator loss using unperturbed hidden
+        if classifier is not None:
+            prediction = classifier(torch.mean(unpert_last_hidden, dim=1))
+            label = torch.tensor([label_class], device="cuda", dtype=torch.long)
+            unpert_discrim_loss = torch.nn.CrossEntropyLoss()(prediction, label)
+            print(
+                "unperturbed discrim loss",
+                unpert_discrim_loss.data.cpu().numpy()
+            )
+        else:
+            unpert_discrim_loss = 0
+
+        # Fuse the modified model and original model probabilities
+        if perturb:
+            unpert_probs = F.softmax(unpert_logits[:, -1, :], dim=-1)
+
+            pert_probs = (pert_probs ** gm_scale) * (
+                    unpert_probs ** (1 - gm_scale)
+            )
+
+            pert_probs = top_k_filter(pert_probs, k=top_k, probs=True)
+
+            # rescale
+            if torch.sum(pert_probs) <= 1:
+                pert_probs = pert_probs / torch.sum(pert_probs)
+
+        else:
+            pert_logits = top_k_filter(pert_logits, k=top_k)
+            pert_probs = F.softmax(pert_logits, dim=-1)
+
+        # sample or greedy
+        if sample:
+            last = torch.multinomial(pert_probs, num_samples=1)
+
+        else:
+            _, last = torch.topk(pert_probs, k=1, dim=-1)
+
+        # update context/output_so_far appending the new token
+        output_so_far = (
+            last if output_so_far is None
+            else torch.cat((output_so_far, last), dim=1)
+        )
+        print(TOKENIZER.decode(output_so_far.tolist()[0]))
+
+    return output_so_far, unpert_discrim_loss, loss_in_time
+
+
+def run_model():
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        "--model_path",
+        "-M",
+        type=str,
+        default="gpt2-medium",
+        help="pretrained model name or path to local checkpoint",
+    )
+    parser.add_argument(
+        "--bag_of_words",
+        "-B",
+        type=str,
+        default=None,
+        help="Bags of words used for PPLM-BoW. Either a BOW id (see list in code) or a filepath. Multiple BoWs separated by ;",
+    )
+    parser.add_argument(
+        "--discrim",
+        "-D",
+        type=str,
+        default=None,
+        choices=("clickbait", "sentiment", "toxicity"),
+        help="Discriminator to use for loss-type 2",
+    )
+    parser.add_argument(
+        "--label_class",
+        type=int,
+        default=-1,
+        help="Class label used for the discriminator",
+    )
+    parser.add_argument("--stepsize", type=float, default=0.02)
+    parser.add_argument("--length", type=int, default=100)
+    parser.add_argument("--seed", type=int, default=0)
+    parser.add_argument("--temperature", type=float, default=1.0)
+    parser.add_argument("--top_k", type=int, default=10)
+    parser.add_argument("--gm_scale", type=float, default=0.9)
+    parser.add_argument("--kl_scale", type=float, default=0.01)
+    parser.add_argument("--no_cuda", action="store_true", help="no cuda")
+    parser.add_argument(
+        "--uncond", action="store_true",
+        help="Generate from end-of-text as prefix"
+    )
+    parser.add_argument(
+        "--cond_text", type=str, default="The lake",
+        help="Prefix texts to condition on"
+    )
+    parser.add_argument("--num_iterations", type=int, default=3)
+    parser.add_argument("--grad_length", type=int, default=10000)
+    parser.add_argument(
+        "--num_samples",
+        type=int,
+        default=1,
+        help="Number of samples to generate from the modified latents",
+    )
+    parser.add_argument(
+        "--horizon_length",
+        type=int,
+        default=1,
+        help="Length of future to optimize over",
+    )
+    parser.add_argument(
+        "--window_length",
+        type=int,
+        default=0,
+        help="Length of past which is being optimized; "
+             "0 corresponds to infinite window length",
+    )
+    parser.add_argument("--decay", action="store_true",
+                        help="whether to decay or not")
+    parser.add_argument("--gamma", type=float, default=1.5)
+
+    args = parser.parse_args()
+
+    # set Random seed
+    torch.manual_seed(args.seed)
+    np.random.seed(args.seed)
+
+    # set the device
+    device = torch.device("cuda" if torch.cuda.is_available() and not args.no_cuda else "cpu")
+
+    # load pretrained model
+    model = GPT2LMHeadModel.from_pretrained(
+        args.model_path,
+        output_hidden_states=True
+    )
+    model.to(device)
+    model.eval()
+
+    # freeze GPT-2 weights
+    for param in model.parameters():
+        param.requires_grad = False
+
+    # figure out conditioning text
+    if args.uncond:
+        tokenized_cond_text = TOKENIZER.encode(
+            [TOKENIZER.bos_token]
+        )
+    else:
+        raw_text = args.cond_text
+        while not raw_text:
+            print("Did you forget to add `--cond_text`? ")
+            raw_text = input("Model prompt >>> ")
+        tokenized_cond_text = TOKENIZER.encode(TOKENIZER.bos_token + raw_text)
+
+    print("= Prefix of sentence =")
+    print(TOKENIZER.decode(tokenized_cond_text))
+    print()
+
+    # generate unperturbed and perturbed texts
+
+    # full_text_generation returns:
+    # unpert_gen_tok_text, pert_gen_tok_texts, discrim_losses, losses_in_time
+    unpert_gen_tok_text, pert_gen_tok_texts, _, _ = full_text_generation(
+        model=model, context=tokenized_cond_text, device=device, **vars(args)
+    )
+
+    # untokenize unperturbed text
+    unpert_gen_text = TOKENIZER.decode(unpert_gen_tok_text.tolist()[0])
+
+    print("=" * 80)
+    print("= Unperturbed generated text =")
+    print(unpert_gen_text)
+    print()
+
+    generated_texts = []
+
+    # iterate through the perturbed texts
+    for i, pert_gen_tok_text in enumerate(pert_gen_tok_texts):
+        try:
+            # untokenize unperturbed text
+            unpert_gen_text = TOKENIZER.decode(pert_gen_tok_text.tolist()[0])
+
+            print("= Perturbed generated text {} =".format(i + 1))
+            print(unpert_gen_text)
+            print()
+        except:
+            pass
+
+        # keep the prefix, perturbed seq, original seq for each index
+        generated_texts.append(
+            (tokenized_cond_text, pert_gen_tok_text, unpert_gen_tok_text)
+        )
+
+    return generated_texts
+
+
+if __name__ == "__main__":
+    run_model()

From 83b1e6ac9e81cbb053ee272a4a4fcb0b6fac06ab Mon Sep 17 00:00:00 2001
From: Rosanne Liu <mimosavvy@gmail.com>
Date: Sun, 3 Nov 2019 04:51:57 +0000
Subject: [PATCH 175/505] fix the loss backward issue

(cherry picked from commit 566468cc984c6ec7e10dfc62b5b4191781a99cd2)
---
 examples/run_pplm.py | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/examples/run_pplm.py b/examples/run_pplm.py
index 30853c68c3..59ae8a9299 100644
--- a/examples/run_pplm.py
+++ b/examples/run_pplm.py
@@ -36,6 +36,7 @@ from tqdm import trange
 from transformers import GPT2Tokenizer
 from transformers.file_utils import cached_path
 from transformers.modeling_gpt2 import GPT2LMHeadModel
+from IPython import embed
 
 PPLM_BOW = 1
 PPLM_DISCRIM = 2
@@ -246,8 +247,8 @@ def perturb_past(
                     inputs_embeds=inputs_embeds
                 )
                 # get expected hidden states
-                unpert_hidden = curr_all_hidden[1]
-                accumulated_hidden += torch.sum(unpert_hidden, dim=1)
+                unpert_hidden = curr_all_hidden[-1]
+                accumulated_hidden += torch.sum(unpert_hidden, dim=1).detach()
 
             prediction = classifier(
                 accumulated_hidden / (curr_length + 1 + horizon_length)
@@ -257,7 +258,7 @@ def perturb_past(
             discrim_loss += ce_loss(prediction, label)
             print(" pplm_discrim_loss:", discrim_loss.data.cpu().numpy())
 
-        if kl_scale > 0.0:
+        if kl_scale >= 0.0:
             unpert_probs = F.softmax(unpert_logits[:, -1, :], dim=-1)
             unpert_probs = (
                     unpert_probs + SMALL_CONST *
@@ -270,7 +271,7 @@ def perturb_past(
                 torch.FloatTensor
             ).cuda().detach()
             corrected_probs = probs + correction.detach()
-            kl_loss += kl_scale * (
+            kl_loss = kl_scale * (
                 (corrected_probs * (corrected_probs / unpert_probs).log()).sum()
             )
             print(' kl_loss', (kl_loss).data.cpu().numpy())
@@ -280,7 +281,7 @@ def perturb_past(
         print(' pplm_loss', (loss - kl_loss).data.cpu().numpy())
 
         # compute gradients
-        loss.backward(retain_graph=True)
+        loss.backward()
 
         # calculate gradient norms
         if grad_norms is not None and loss_type == PPLM_BOW:

From 0b77d66a6dac359c3473d6d4b7e799af5196ae4d Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Tue, 5 Nov 2019 15:35:51 +0000
Subject: [PATCH 176/505] rm extraneous import

---
 examples/run_pplm.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/examples/run_pplm.py b/examples/run_pplm.py
index 59ae8a9299..2d4dee72a3 100644
--- a/examples/run_pplm.py
+++ b/examples/run_pplm.py
@@ -36,7 +36,6 @@ from tqdm import trange
 from transformers import GPT2Tokenizer
 from transformers.file_utils import cached_path
 from transformers.modeling_gpt2 import GPT2LMHeadModel
-from IPython import embed
 
 PPLM_BOW = 1
 PPLM_DISCRIM = 2

From d5faa74cd6d7de66a058a9b3368e5cbc6dcaf4d6 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Tue, 5 Nov 2019 15:48:00 +0000
Subject: [PATCH 177/505] tokenizer white space: revert to previous behavior

---
 examples/run_pplm.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/examples/run_pplm.py b/examples/run_pplm.py
index 2d4dee72a3..4b1a6a2b6f 100644
--- a/examples/run_pplm.py
+++ b/examples/run_pplm.py
@@ -373,7 +373,7 @@ def get_bag_of_words_indices(bag_of_words_ids_or_paths: List[str]) -> List[List[
             filepath = id_or_path
         with open(filepath, "r") as f:
             words = f.read().split("\n")
-        bow_indices.append([TOKENIZER.encode(word) for word in words])
+        bow_indices.append([TOKENIZER.encode(word, add_prefix_space=True) for word in words])
     return bow_indices
 
 

From 34a83faabeb8f1f0e487a70f31a4e6b1cc0185e1 Mon Sep 17 00:00:00 2001
From: Piero Molino <piero@uber.com>
Date: Mon, 25 Nov 2019 19:15:25 -0800
Subject: [PATCH 178/505] Let's make PPLM great again

---
 examples/run_pplm.py | 908 +++++++++++++++++++++----------------------
 1 file changed, 440 insertions(+), 468 deletions(-)

diff --git a/examples/run_pplm.py b/examples/run_pplm.py
index 4b1a6a2b6f..2f853d15c1 100644
--- a/examples/run_pplm.py
+++ b/examples/run_pplm.py
@@ -1,3 +1,4 @@
+#! /usr/bin/env python3
 # coding=utf-8
 # Copyright 2018 The Uber AI Team Authors.
 #
@@ -37,10 +38,12 @@ from transformers import GPT2Tokenizer
 from transformers.file_utils import cached_path
 from transformers.modeling_gpt2 import GPT2LMHeadModel
 
+
 PPLM_BOW = 1
 PPLM_DISCRIM = 2
 PPLM_BOW_DISCRIM = 3
 SMALL_CONST = 1e-15
+SmallConst = 1e-15
 TOKENIZER = GPT2Tokenizer.from_pretrained("gpt2-medium")
 
 BAG_OF_WORDS_ARCHIVE_MAP = {
@@ -65,7 +68,7 @@ DISCRIMINATOR_MODELS_PARAMS = {
         "default_class": 1,
     },
     "sentiment": {
-        "url": "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/discriminators/sentiment_classifierhead.pt",
+        "url": "http://s.yosinski.com/SST_classifier_head.pt",
         "class_size": 5,
         "embed_size": 1024,
         "class_vocab": {"very_positive": 2, "very_negative": 3},
@@ -81,6 +84,30 @@ DISCRIMINATOR_MODELS_PARAMS = {
 }
 
 
+def to_var(x, requires_grad=False, volatile=False):
+    if torch.cuda.is_available():
+        x = x.cuda()
+    return Variable(x, requires_grad=requires_grad, volatile=volatile)
+
+
+def top_k_filter(logits, k, probs=False):
+    """
+    Masks everything but the k top entries as -infinity (1e10).
+    Used to mask logits such that e^-infinity -> 0 won't contribute to the
+    sum of the denominator.
+    """
+    if k == 0:
+        return logits
+    else:
+        values = torch.topk(logits, k)[0]
+        batch_mins = values[:, -1].view(-1, 1).expand_as(logits)
+        if probs:
+            return torch.where(logits < batch_mins,
+                               torch.ones_like(logits) * 0.0, logits)
+        return torch.where(logits < batch_mins, torch.ones_like(logits) * -1e10,
+                           logits)
+
+
 class ClassificationHead(torch.nn.Module):
     """ Classification Head for the transformer """
 
@@ -99,234 +126,175 @@ class ClassificationHead(torch.nn.Module):
         return logits
 
 
-def to_var(x, requires_grad=False, volatile=False):
-    if torch.cuda.is_available():
-        x = x.cuda()
-    return Variable(x, requires_grad=requires_grad, volatile=volatile)
+def perturb_past(past, model, prev, args, classifier, good_index=None,
+                 stepsize=0.01, vocab_size=50257,
+                 original_probs=None, accumulated_hidden=None, true_past=None,
+                 grad_norms=None):
+    window_length = args.window_length
+    gm_scale, kl_scale = args.gm_scale, args.kl_scale
+    one_hot_vectors = []
+    for good_list in good_index:
+        good_list = list(filter(lambda x: len(x) <= 1, good_list))
+        good_list = torch.tensor(good_list).cuda()
+        num_good = good_list.shape[0]
+        one_hot_good = torch.zeros(num_good, vocab_size).cuda()
+        one_hot_good.scatter_(1, good_list, 1)
+        one_hot_vectors.append(one_hot_good)
 
-
-def top_k_filter(logits, k, probs=False):
-    """
-    Masks everything but the k top entries as -infinity (1e10).
-    Used to mask logits such that e^-infinity -> 0 won't contribute to the
-    sum of the denominator.
-    """
-    if k <= 0:
-        return logits
-
-    else:
-        values = torch.topk(logits, k)[0]
-        batch_mins = values[:, -1].view(-1, 1).expand_as(logits)
-
-        if probs:
-            return torch.where(
-                logits < batch_mins,
-                torch.ones_like(logits) * 0.0,
-                logits
-            )
-
-        return torch.where(
-            logits < batch_mins,
-            torch.ones_like(logits) * -1e10,
-            logits
-        )
-
-
-def perturb_past(
-        past,
-        model,
-        last,
-        unpert_past=None,
-        unpert_logits=None,
-        accumulated_hidden=None,
-        grad_norms=None,
-        stepsize=0.01,
-        classifier=None,
-        label_class=None,
-        one_hot_bows_vectors=None,
-        loss_type=0,
-        num_iterations=3,
-        kl_scale=0.01,
-        window_length=0,
-        horizon_length=1,
-        decay=False,
-        gamma=1.5,
-):
-    # initializie perturbation accumulator
-    grad_accumulator = [
-        (np.zeros(p.shape).astype("float32"))
-        for p in past
-    ]
+    # Generate inital perturbed past
+    past_perturb_orig = [
+        (np.random.uniform(0.0, 0.0, p.shape).astype('float32'))
+        for p in past]
 
     if accumulated_hidden is None:
         accumulated_hidden = 0
 
-    if decay:
-        decay_mask = torch.arange(
-            0.0,
-            1.0 + SMALL_CONST,
-            1.0 / (window_length)
-        )[1:]
+    if args.decay:
+        decay_mask = torch.arange(0., 1.0 + SmallConst, 1.0 / (window_length))[
+                     1:]
     else:
         decay_mask = 1.0
 
-    # TODO fix this comment (SUMANTH)
-    # generate a mask if perturbated gradient is based on a past window
-    _, _, _, curr_length, _ = past[0].shape
-    if curr_length > window_length and window_length > 0:
-        ones_key_val_shape = (
-                tuple(past[0].shape[:-2])
-                + tuple([window_length])
-                + tuple(past[0].shape[-1:])
-        )
+    # Generate a mask is gradient perturbated is based on a past window
+    _, _, _, current_length, _ = past[0].shape
 
-        zeros_key_val_shape = (
-                tuple(past[0].shape[:-2])
-                + tuple([curr_length - window_length])
-                + tuple(past[0].shape[-1:])
-        )
+    if current_length > window_length and window_length > 0:
+        ones_key_val_shape = tuple(past[0].shape[:-2]) + tuple(
+            [window_length]) + tuple(
+            past[0].shape[-1:])
+
+        zeros_key_val_shape = tuple(past[0].shape[:-2]) + tuple(
+            [current_length - window_length]) + tuple(
+            past[0].shape[-1:])
 
         ones_mask = torch.ones(ones_key_val_shape)
         ones_mask = decay_mask * ones_mask.permute(0, 1, 2, 4, 3)
         ones_mask = ones_mask.permute(0, 1, 2, 4, 3)
 
-        window_mask = torch.cat(
-            (ones_mask, torch.zeros(zeros_key_val_shape)),
-            dim=-2
-        ).cuda()
-
+        window_mask = torch.cat((ones_mask, torch.zeros(zeros_key_val_shape)),
+                                dim=-2).cuda()
     else:
         window_mask = torch.ones_like(past[0]).cuda()
 
-    # accumulate perturbations for num_iterations
     loss_per_iter = []
-    for i in range(num_iterations):
+    for i in range(args.num_iterations):
         print("Iteration ", i + 1)
+        past_perturb = [torch.from_numpy(p_) for p_ in past_perturb_orig]
+        past_perturb = [to_var(p_, requires_grad=True) for p_ in past_perturb]
 
-        curr_perturbation = [
-            to_var(torch.from_numpy(p_), requires_grad=True)
-            for p_ in grad_accumulator
-        ]
+        perturbed_past = list(map(add, past, past_perturb))
 
-        # Compute hidden using perturbed past
-        curr_pert_past = list(map(add, past, curr_perturbation))
-        all_logits, _, all_hidden = model(last, past=curr_pert_past)
+        _, _, _, current_length, _ = past_perturb[0].shape
+
+        # _, future_past = model(prev, past=perturbed_past)
+        # hidden = model.hidden_states
+
+        # Piero modified model call
+        logits, _, all_hidden = model(prev, past=perturbed_past)
         hidden = all_hidden[-1]
-        accumulated_hidden += torch.sum(hidden, dim=1).detach()
-        logits = all_logits[:, -1, :]
-        probs = F.softmax(logits, dim=-1)
+        new_accumulated_hidden = accumulated_hidden + torch.sum(hidden,
+                                                                dim=1).detach()
 
-        # compute loss
-        bow_loss = 0.0
-        discrim_loss = 0.0
-        kl_loss = 0.0
+        # TODO: Check the layer-norm consistency of this with trained discriminator
+        logits = logits[:, -1, :]
+        probabs = F.softmax(logits, dim=-1)
+        loss = 0.0
+        loss_list = []
+        if args.loss_type == 1 or args.loss_type == 3:
+            for one_hot_good in one_hot_vectors:
+                good_logits = torch.mm(probabs, torch.t(one_hot_good))
+                loss_word = good_logits
+                loss_word = torch.sum(loss_word)
+                loss_word = -torch.log(loss_word)
+                # loss_word = torch.sum(loss_word) /torch.sum(one_hot_good)
+                loss += loss_word
+                loss_list.append(loss_word)
+            print(" pplm_bow_loss:", loss.data.cpu().numpy())
 
-        if loss_type == PPLM_BOW or loss_type == PPLM_BOW_DISCRIM:
-            for one_hot_bow in one_hot_bows_vectors:
-                bow_logits = torch.mm(probs, torch.t(one_hot_bow))
-                bow_loss += -torch.log(torch.sum(bow_logits))
-            print(" pplm_bow_loss:", bow_loss.data.cpu().numpy())
-
-        if loss_type == PPLM_DISCRIM or loss_type == PPLM_BOW_DISCRIM:
+        if args.loss_type == 2 or args.loss_type == 3:
             ce_loss = torch.nn.CrossEntropyLoss()
-            # TODO all there are for (SUMANTH)
-            # TODO why we need to do this assignment and not just using unpert_past?
-            curr_unpert_past = unpert_past
-            # Get the model's token embeddings in order to compute our own embeds from curr_probs:
-            wte = model.resize_token_embeddings()
-            # TODO i is never used, why do we need to do this i times instead multiplying
-            #   torch.sum(unpert_hidden, dim=1) * horizon_length?
-            for i in range(horizon_length):
-                # TODO the next two lines can be done only one time, and why not using probs instead as they do not change at each iteration?
-                curr_probs = F.softmax(logits, dim=-1)  # get softmax
-                curr_probs = torch.unsqueeze(curr_probs, dim=1)
-                inputs_embeds = torch.matmul(curr_probs, wte.weight.data)
-                _, curr_unpert_past, curr_all_hidden = model(
-                    past=curr_unpert_past,
+            new_true_past = true_past
+            for i in range(args.horizon_length):
+                future_probabs = F.softmax(logits, dim=-1)  # Get softmax
+                future_probabs = torch.unsqueeze(future_probabs, dim=1)
+
+                # _, new_true_past = model(future_probabs, past=new_true_past)
+                # future_hidden = model.hidden_states  # Get expected hidden states
+
+                # Piero modified model call
+                wte = model.resize_token_embeddings()
+                inputs_embeds = torch.matmul(future_probabs, wte.weight.data)
+                _, new_true_past, future_hidden = model(
+                    past=new_true_past,
                     inputs_embeds=inputs_embeds
                 )
-                # get expected hidden states
-                unpert_hidden = curr_all_hidden[-1]
-                accumulated_hidden += torch.sum(unpert_hidden, dim=1).detach()
+                future_hidden = future_hidden[-1]
 
-            prediction = classifier(
-                accumulated_hidden / (curr_length + 1 + horizon_length)
-            )
+                new_accumulated_hidden = new_accumulated_hidden + torch.sum(
+                    future_hidden, dim=1)
 
-            label = torch.tensor([label_class], device="cuda", dtype=torch.long)
-            discrim_loss += ce_loss(prediction, label)
+            predicted_sentiment = classifier(new_accumulated_hidden / (
+                        current_length + 1 + args.horizon_length))
+
+            label = torch.tensor([args.label_class], device='cuda',
+                                 dtype=torch.long)
+            discrim_loss = ce_loss(predicted_sentiment, label)
             print(" pplm_discrim_loss:", discrim_loss.data.cpu().numpy())
+            loss += discrim_loss
+            loss_list.append(discrim_loss)
 
-        if kl_scale >= 0.0:
-            unpert_probs = F.softmax(unpert_logits[:, -1, :], dim=-1)
-            unpert_probs = (
-                    unpert_probs + SMALL_CONST *
-                    (unpert_probs <= SMALL_CONST).type(
-                        torch.FloatTensor
-                    ).cuda().detach()
-            )
-
-            correction = SMALL_CONST * (probs <= SMALL_CONST).type(
-                torch.FloatTensor
-            ).cuda().detach()
-            corrected_probs = probs + correction.detach()
+        kl_loss = 0.0
+        if kl_scale > 0.0:
+            p = (F.softmax(original_probs[:, -1, :], dim=-1))
+            p = p + SmallConst * (p <= SmallConst).type(
+                torch.FloatTensor).cuda().detach()
+            correction = SmallConst * (probabs <= SmallConst).type(
+                torch.FloatTensor).cuda().detach()
+            corrected_probabs = probabs + correction.detach()
             kl_loss = kl_scale * (
-                (corrected_probs * (corrected_probs / unpert_probs).log()).sum()
-            )
+                (corrected_probabs * (corrected_probabs / p).log()).sum())
             print(' kl_loss', (kl_loss).data.cpu().numpy())
+            loss += kl_loss  # + discrim_loss
 
-        loss = bow_loss + discrim_loss + kl_loss
         loss_per_iter.append(loss.data.cpu().numpy())
+
         print(' pplm_loss', (loss - kl_loss).data.cpu().numpy())
 
-        # compute gradients
         loss.backward()
-
-        # calculate gradient norms
-        if grad_norms is not None and loss_type == PPLM_BOW:
+        if grad_norms is not None and args.loss_type == 1:
             grad_norms = [
                 torch.max(grad_norms[index], torch.norm(p_.grad * window_mask))
-                for index, p_ in enumerate(curr_perturbation)
-            ]
+                for index, p_ in
+                enumerate(past_perturb)]
         else:
-            grad_norms = [
-                (torch.norm(p_.grad * window_mask) + SMALL_CONST)
-                for index, p_ in enumerate(curr_perturbation)
-            ]
+            grad_norms = [(torch.norm(p_.grad * window_mask) + SmallConst) for
+                          index, p_ in enumerate(past_perturb)]
 
-        # normalize gradients
         grad = [
-            -stepsize
-            * (p_.grad * window_mask / grad_norms[
-                index] ** gamma).data.cpu().numpy()
-            for index, p_ in enumerate(curr_perturbation)
-        ]
+            -stepsize * (p_.grad * window_mask / grad_norms[
+                index] ** args.gamma).data.cpu().numpy()
+            for index, p_ in enumerate(past_perturb)]
+        past_perturb_orig = list(map(add, grad, past_perturb_orig))
 
-        # accumulate gradients
-        grad_accumulator = list(map(add, grad, grad_accumulator))
-
-        # reset gradients, just to make sure
-        for p_ in curr_perturbation:
+        for p_ in past_perturb:
             p_.grad.data.zero_()
 
-        # removing past from the graph
         new_past = []
-        for p_ in past:
-            new_past.append(p_.detach())
+        for p in past:
+            new_past.append(p.detach())
+
         past = new_past
 
-    # apply the accumulated perturbations to the past
-    grad_accumulator = [
-        to_var(torch.from_numpy(p_), requires_grad=True)
-        for p_ in grad_accumulator
-    ]
-    pert_past = list(map(add, past, grad_accumulator))
+    past_perturb = [torch.from_numpy(p_) for p_ in past_perturb_orig]
+    past_perturb = [to_var(p_, requires_grad=True) for p_ in past_perturb]
+    perturbed_past = list(map(add, past, past_perturb))
 
-    return pert_past, accumulated_hidden, grad_norms, loss_per_iter
+    return perturbed_past, new_accumulated_hidden, grad_norms, loss_per_iter
 
 
 def get_classifier(
-        name: Optional[str], label_class: Union[str, int], device: Union[str, torch.device]
+        name: Optional[str], label_class: Union[str, int],
+        device: Union[str, torch.device]
 ) -> Tuple[Optional[ClassificationHead], Optional[int]]:
     if name is None:
         return None, None
@@ -337,7 +305,8 @@ def get_classifier(
         embed_size=params['embed_size']
     ).to(device)
     resolved_archive_file = cached_path(params["url"])
-    classifier.load_state_dict(torch.load(resolved_archive_file, map_location=device))
+    classifier.load_state_dict(
+        torch.load(resolved_archive_file, map_location=device))
     classifier.eval()
 
     if isinstance(label_class, str):
@@ -364,7 +333,8 @@ def get_classifier(
     return classifier, label_id
 
 
-def get_bag_of_words_indices(bag_of_words_ids_or_paths: List[str]) -> List[List[List[int]]]:
+def get_bag_of_words_indices(bag_of_words_ids_or_paths: List[str]) -> List[
+    List[List[int]]]:
     bow_indices = []
     for id_or_path in bag_of_words_ids_or_paths:
         if id_or_path in BAG_OF_WORDS_ARCHIVE_MAP:
@@ -372,8 +342,10 @@ def get_bag_of_words_indices(bag_of_words_ids_or_paths: List[str]) -> List[List[
         else:
             filepath = id_or_path
         with open(filepath, "r") as f:
-            words = f.read().split("\n")
-        bow_indices.append([TOKENIZER.encode(word, add_prefix_space=True) for word in words])
+            words = f.read().strip().split("\n")
+        bow_indices.append(
+            [TOKENIZER.encode(word.strip(), add_prefix_space=True) for word in
+             words])
     return bow_indices
 
 
@@ -392,327 +364,308 @@ def build_bows_one_hot_vectors(bow_indices):
     return one_hot_bows_vectors
 
 
-def full_text_generation(
-        model,
-        context=None,
-        num_samples=1,
-        device="cuda",
-        sample=True,
-        discrim=None,
-        label_class=None,
-        bag_of_words=None,
-        length=100,
-        grad_length=10000,
-        stepsize=0.02,
-        num_iterations=3,
-        temperature=1.0,
-        gm_scale=0.9,
-        kl_scale=0.01,
-        top_k=10,
-        window_length=0,
-        horizon_length=1,
-        decay=False,
-        gamma=1.5,
-        **kwargs
-):
+def latent_perturb(model, args, context=None, sample=True, device='cuda'):
     classifier, class_id = get_classifier(
-        discrim,
-        label_class,
+        args.discrim,
+        args.label_class,
         device
     )
 
-    bow_indices = []
-    if bag_of_words:
-        bow_indices = get_bag_of_words_indices(bag_of_words.split(";"))
+    # if args.discrim == 'clickbait':
+    #     classifier = ClassificationHead(class_size=2, embed_size=1024).to(device)
+    #     classifier.load_state_dict(torch.load("discrim_models/clickbait_classifierhead.pt"))
+    #     classifier.eval()
+    #     args.label_class = 1 # clickbaity
+    #
+    # elif args.discrim == 'sentiment':
+    #     classifier = ClassificationHead(class_size=5, embed_size=1024).to(device)
+    #     #classifier.load_state_dict(torch.load("discrim_models/sentiment_classifierhead.pt"))
+    #     classifier.load_state_dict(torch.load("discrim_models/SST_classifier_head_epoch_16.pt"))
+    #     classifier.eval()
+    #     if args.label_class < 0:
+    #         raise Exception('Wrong class for sentiment, use --label-class 2 for *very positive*, 3 for *very negative*')
+    #     #args.label_class = 2 # very pos
+    #     #args.label_class = 3 # very neg
+    #
+    # elif args.discrim == 'toxicity':
+    #     classifier = ClassificationHead(class_size=2, embed_size=1024).to(device)
+    #     classifier.load_state_dict(torch.load("discrim_models/toxicity_classifierhead.pt"))
+    #     classifier.eval()
+    #     args.label_class = 0 # not toxic
+    #
+    # elif args.discrim == 'generic':
+    #     if args.discrim_weights is None:
+    #         raise ValueError('When using a generic discriminator, '
+    #                          'discrim_weights need to be specified')
+    #     if args.discrim_meta is None:
+    #         raise ValueError('When using a generic discriminator, '
+    #                          'discrim_meta need to be specified')
+    #
+    #     with open(args.discrim_meta, 'r') as discrim_meta_file:
+    #         meta = json.load(discrim_meta_file)
+    #
+    #     classifier = ClassificationHead(
+    #         class_size=meta['class_size'],
+    #         embed_size=meta['embed_size'],
+    #         # todo add tokenizer from meta
+    #     ).to(device)
+    #     classifier.load_state_dict(torch.load(args.discrim_weights))
+    #     classifier.eval()
+    #     if args.label_class == -1:
+    #         args.label_class = meta['default_class']
+    #
+    # else:
+    #     classifier = None
 
-    if bag_of_words and classifier:
+    # Get tokens for the list of positive words
+    def list_tokens(word_list):
+        token_list = [TOKENIZER.encode(word, add_prefix_space=True) for word in
+                      word_list]
+        # token_list = []
+        # for word in word_list:
+        #    token_list.append(TOKENIZER.encode(" " + word))
+        return token_list
+
+    # good_index = []
+    # if args.bag_of_words:
+    #     bags_of_words = args.bag_of_words.split(";")
+    #     for wordlist in bags_of_words:
+    #         with open(wordlist, "r") as f:
+    #             words = f.read().strip()
+    #             words = words.split('\n')
+    #         good_index.append(list_tokens(words))
+    #
+    #     for good_list in good_index:
+    #         good_list = list(filter(lambda x: len(x) <= 1, good_list))
+    #         actual_words = [(TOKENIZER.decode(ww).strip(),ww) for ww in good_list]
+
+    good_index = []
+    actual_words = None
+    if args.bag_of_words:
+        good_index = get_bag_of_words_indices(args.bag_of_words.split(";"))
+
+        for good_list in good_index:
+            good_list = list(filter(lambda x: len(x) <= 1, good_list))
+            actual_words = [(TOKENIZER.decode(ww).strip(), ww) for ww in
+                            good_list]
+
+    if args.bag_of_words and classifier:
         print("Both PPLM-BoW and PPLM-Discrim are on. This is not optimized.")
-        loss_type = PPLM_BOW_DISCRIM
+        args.loss_type = PPLM_BOW_DISCRIM
 
-    elif bag_of_words:
-        loss_type = PPLM_BOW
+    elif args.bag_of_words:
+        args.loss_type = PPLM_BOW
         print("Using PPLM-BoW")
 
     elif classifier is not None:
-        loss_type = PPLM_DISCRIM
+        args.loss_type = PPLM_DISCRIM
         print("Using PPLM-Discrim")
 
     else:
         raise Exception("Specify either --bag_of_words (-B) or --discrim (-D)")
 
-    unpert_gen_tok_text, _, _ = generate_text_pplm(
-        model=model,
-        context=context,
-        device=device,
-        length=length,
-        perturb=False
-    )
+    original, _, _ = sample_from_hidden(model=model, args=args, context=context,
+                                        device=device,
+                                        perturb=False, good_index=good_index,
+                                        classifier=classifier)
     torch.cuda.empty_cache()
 
-    pert_gen_tok_texts = []
-    discrim_losses = []
-    losses_in_time = []
+    perturbed_list = []
+    discrim_loss_list = []
+    loss_in_time_list = []
 
-    for i in range(num_samples):
-        pert_gen_tok_text, discrim_loss, loss_in_time = generate_text_pplm(
-            model=model,
-            context=context,
-            device=device,
-            sample=sample,
-            perturb=True,
-            bow_indices=bow_indices,
-            classifier=classifier,
-            label_class=class_id,
-            loss_type=loss_type,
-            length=length,
-            grad_length=grad_length,
-            stepsize=stepsize,
-            num_iterations=num_iterations,
-            temperature=temperature,
-            gm_scale=gm_scale,
-            kl_scale=kl_scale,
-            top_k=top_k,
-            window_length=window_length,
-            horizon_length=horizon_length,
-            decay=decay,
-            gamma=gamma,
-        )
-        pert_gen_tok_texts.append(pert_gen_tok_text)
+    for i in range(args.num_samples):
+        perturbed, discrim_loss, loss_in_time = sample_from_hidden(model=model,
+                                                                   args=args,
+                                                                   context=context,
+                                                                   device=device,
+                                                                   perturb=True,
+                                                                   good_index=good_index,
+                                                                   classifier=classifier)
+        perturbed_list.append(perturbed)
         if classifier is not None:
-            discrim_losses.append(discrim_loss.data.cpu().numpy())
-        losses_in_time.append(loss_in_time)
+            discrim_loss_list.append(discrim_loss.data.cpu().numpy())
+        loss_in_time_list.append(loss_in_time)
 
     torch.cuda.empty_cache()
 
-    return unpert_gen_tok_text, pert_gen_tok_texts, discrim_losses, losses_in_time
+    return original, perturbed_list, discrim_loss_list, loss_in_time_list, actual_words
 
 
-def generate_text_pplm(
-        model,
-        context=None,
-        past=None,
-        device="cuda",
-        sample=True,
-        perturb=True,
-        classifier=None,
-        label_class=None,
-        bow_indices=None,
-        loss_type=0,
-        length=100,
-        grad_length=10000,
-        stepsize=0.02,
-        num_iterations=3,
-        temperature=1.0,
-        gm_scale=0.9,
-        kl_scale=0.01,
-        top_k=10,
-        window_length=0,
-        horizon_length=1,
-        decay=False,
-        gamma=1.5,
-):
-    output_so_far = (
-        torch.tensor(context, device=device, dtype=torch.long).unsqueeze(0)
-        if context
-        else None
-    )
-
-    # collect one hot vectors for bags of words
-    one_hot_bows_vectors = build_bows_one_hot_vectors(bow_indices)
+def sample_from_hidden(model, args, classifier, context=None, past=None,
+                       device='cuda',
+                       sample=True, perturb=True, good_index=None):
+    output = torch.tensor(context, device=device, dtype=torch.long).unsqueeze(
+        0) if context else None
 
     grad_norms = None
-    last = None
-    unpert_discrim_loss = 0
     loss_in_time = []
-    for i in trange(length, ascii=True):
+    for i in trange(args.length, ascii=True):
 
         # Get past/probs for current output, except for last word
-        # Note that GPT takes 2 inputs: past + current_token
+        # Note that GPT takes 2 inputs: past + current-token
+        # Therefore, use everything from before current i/p token to generate relevant past
 
-        # run model forward to obtain unperturbed
-        if past is None and output_so_far is not None:
-            last = output_so_far[:, -1:]
-            if output_so_far.shape[1] > 1:
-                _, past, _ = model(output_so_far[:, :-1])
+        if past is None and output is not None:
+            prev = output[:, -1:]
+            # _, past = model(output[:, :-1])
+            # original_probs, true_past = model(output)
+            # true_hidden = model.hidden_states
 
-            unpert_logits, unpert_past, unpert_all_hidden = model(output_so_far)
-            unpert_last_hidden = unpert_all_hidden[-1]
+            # Piero modified model call
+            _, past, _ = model(output[:, :-1])
+            original_probs, true_past, unpert_all_hidden = model(output)
+            true_hidden = unpert_all_hidden[-1]
 
         else:
-            unpert_logits, unpert_past, unpert_all_hidden = model(output_so_far)
-            unpert_last_hidden = unpert_all_hidden[-1]
+            # original_probs, true_past = model(output)
+            # true_hidden = model.hidden_states
 
-        # check if we are abowe grad max length
-        if i >= grad_length:
-            current_stepsize = stepsize * 0
+            # Piero modified model call
+            original_probs, true_past, unpert_all_hidden = model(output)
+            true_hidden = unpert_all_hidden[-1]
+
+        # Modify the past if necessary
+
+        if i >= args.grad_length:
+            current_stepsize = args.stepsize * 0
         else:
-            current_stepsize = stepsize
+            current_stepsize = args.stepsize
 
-        # modify the past if necessary
-        if not perturb or num_iterations == 0:
-            pert_past = past
+        if not perturb or args.num_iterations == 0:
+            perturbed_past = past
 
         else:
-            accumulated_hidden = unpert_last_hidden[:, :-1, :]
+            # Piero modified model call
+            # accumulated_hidden = model.hidden_states[:, :-1, :]
+            accumulated_hidden = true_hidden[:, :-1, :]
             accumulated_hidden = torch.sum(accumulated_hidden, dim=1)
 
-            if past is not None:
-                pert_past, _, grad_norms, loss_this_iter = perturb_past(
-                    past,
-                    model,
-                    last,
-                    unpert_past=unpert_past,
-                    unpert_logits=unpert_logits,
-                    accumulated_hidden=accumulated_hidden,
-                    grad_norms=grad_norms,
-                    stepsize=current_stepsize,
-                    classifier=classifier,
-                    label_class=label_class,
-                    one_hot_bows_vectors=one_hot_bows_vectors,
-                    loss_type=loss_type,
-                    num_iterations=num_iterations,
-                    kl_scale=kl_scale,
-                    window_length=window_length,
-                    horizon_length=horizon_length,
-                    decay=decay,
-                    gamma=gamma,
-                )
-                loss_in_time.append(loss_this_iter)
-            else:
-                pert_past = past
+            perturbed_past, _, grad_norms, loss_per_iter = perturb_past(past,
+                                                                        model,
+                                                                        prev,
+                                                                        args,
+                                                                        good_index=good_index,
+                                                                        stepsize=current_stepsize,
+                                                                        original_probs=original_probs,
+                                                                        true_past=true_past,
+                                                                        accumulated_hidden=accumulated_hidden,
+                                                                        classifier=classifier,
+                                                                        grad_norms=grad_norms)
+            loss_in_time.append(loss_per_iter)
 
-        pert_logits, past, pert_all_hidden = model(last, past=pert_past)
-        pert_logits = pert_logits[:, -1, :] / temperature
-        pert_probs = F.softmax(pert_logits, dim=-1)
+        # Piero modified model call
+        logits, past, pert_all_hidden = model(prev, past=perturbed_past)
+        # test_logits = F.softmax(test_logits[:, -1, :], dim=-1)
+        # likelywords = torch.topk(test_logits, k=10, dim=-1)
+        # print(TOKENIZER.decode(likelywords[1].tolist()[0]))
 
-        # compute the discriminator loss using unperturbed hidden
         if classifier is not None:
-            prediction = classifier(torch.mean(unpert_last_hidden, dim=1))
-            label = torch.tensor([label_class], device="cuda", dtype=torch.long)
-            unpert_discrim_loss = torch.nn.CrossEntropyLoss()(prediction, label)
-            print(
-                "unperturbed discrim loss",
-                unpert_discrim_loss.data.cpu().numpy()
-            )
+            ce_loss = torch.nn.CrossEntropyLoss()
+            predicted_sentiment = classifier(torch.mean(true_hidden, dim=1))
+            label = torch.tensor([args.label_class], device='cuda',
+                                 dtype=torch.long)
+            true_discrim_loss = ce_loss(predicted_sentiment, label)
+            print("true discrim loss", true_discrim_loss.data.cpu().numpy())
         else:
-            unpert_discrim_loss = 0
+            true_discrim_loss = 0
 
-        # Fuse the modified model and original model probabilities
+        # Piero modified model call
+        # hidden = model.hidden_states  # update hidden
+        # logits = model.forward_hidden(hidden)
+        logits = logits[:, -1, :] / args.temperature  # + SmallConst
+
+        # logits = top_k_filter(logits, k=args.top_k)  # + SmallConst
+
+        log_probs = F.softmax(logits, dim=-1)
+
+        # Fuse the modified model and original model
         if perturb:
-            unpert_probs = F.softmax(unpert_logits[:, -1, :], dim=-1)
 
-            pert_probs = (pert_probs ** gm_scale) * (
-                    unpert_probs ** (1 - gm_scale)
-            )
+            # original_probs = top_k_filter(original_probs[:, -1, :]) #+ SmallConst
+            original_probs = F.softmax(original_probs[:, -1, :], dim=-1)
+            # likelywords = torch.topk(original_probs, k=10, dim=-1)
+            # print(TOKENIZER.decode(likelywords[1].tolist()[0]))
 
-            pert_probs = top_k_filter(pert_probs, k=top_k, probs=True)
+            gm_scale = args.gm_scale
+            log_probs = ((log_probs ** gm_scale) * (
+                        original_probs ** (1 - gm_scale)))  # + SmallConst
 
-            # rescale
-            if torch.sum(pert_probs) <= 1:
-                pert_probs = pert_probs / torch.sum(pert_probs)
+            log_probs = top_k_filter(log_probs, k=args.top_k,
+                                     probs=True)  # + SmallConst
+
+            if torch.sum(log_probs) <= 1:
+                log_probs = log_probs / torch.sum(log_probs)
 
         else:
-            pert_logits = top_k_filter(pert_logits, k=top_k)
-            pert_probs = F.softmax(pert_logits, dim=-1)
+            logits = top_k_filter(logits, k=args.top_k)  # + SmallConst
+            log_probs = F.softmax(logits, dim=-1)
 
-        # sample or greedy
         if sample:
-            last = torch.multinomial(pert_probs, num_samples=1)
-
+            # likelywords = torch.topk(log_probs, k=args.top_k, dim=-1)
+            # print(TOKENIZER.decode(likelywords[1].tolist()[0]))
+            # print(likelywords[0].tolist())
+            prev = torch.multinomial(log_probs, num_samples=1)
         else:
-            _, last = torch.topk(pert_probs, k=1, dim=-1)
+            _, prev = torch.topk(log_probs, k=1, dim=-1)
+        # if perturb:
+        #     prev = future
+        output = prev if output is None else torch.cat((output, prev),
+                                                       dim=1)  # update output
+        print(TOKENIZER.decode(output.tolist()[0]))
 
-        # update context/output_so_far appending the new token
-        output_so_far = (
-            last if output_so_far is None
-            else torch.cat((output_so_far, last), dim=1)
-        )
-        print(TOKENIZER.decode(output_so_far.tolist()[0]))
-
-    return output_so_far, unpert_discrim_loss, loss_in_time
+    return output, true_discrim_loss, loss_in_time
 
 
 def run_model():
     parser = argparse.ArgumentParser()
-    parser.add_argument(
-        "--model_path",
-        "-M",
-        type=str,
-        default="gpt2-medium",
-        help="pretrained model name or path to local checkpoint",
-    )
-    parser.add_argument(
-        "--bag_of_words",
-        "-B",
-        type=str,
-        default=None,
-        help="Bags of words used for PPLM-BoW. Either a BOW id (see list in code) or a filepath. Multiple BoWs separated by ;",
-    )
-    parser.add_argument(
-        "--discrim",
-        "-D",
-        type=str,
-        default=None,
-        choices=("clickbait", "sentiment", "toxicity"),
-        help="Discriminator to use for loss-type 2",
-    )
-    parser.add_argument(
-        "--label_class",
-        type=int,
-        default=-1,
-        help="Class label used for the discriminator",
-    )
-    parser.add_argument("--stepsize", type=float, default=0.02)
+    parser.add_argument('--model_path', '-M', type=str, default='gpt2-medium',
+                        help='pretrained model name or path to local checkpoint')
+    parser.add_argument('--bag-of-words', '-B', type=str, default=None,
+                        help='Bags of words used for PPLM-BoW. Multiple BoWs separated by ;')
+    parser.add_argument('--discrim', '-D', type=str, default=None,
+                        choices=(
+                        'clickbait', 'sentiment', 'toxicity', 'generic'),
+                        help='Discriminator to use for loss-type 2')
+    parser.add_argument('--discrim_weights', type=str, default=None,
+                        help='Weights for the generic discriminator')
+    parser.add_argument('--discrim_meta', type=str, default=None,
+                        help='Meta information for the generic discriminator')
+    parser.add_argument('--label_class', type=int, default=-1,
+                        help='Class label used for the discriminator')
+    parser.add_argument('--stepsize', type=float, default=0.02)
     parser.add_argument("--length", type=int, default=100)
     parser.add_argument("--seed", type=int, default=0)
     parser.add_argument("--temperature", type=float, default=1.0)
     parser.add_argument("--top_k", type=int, default=10)
     parser.add_argument("--gm_scale", type=float, default=0.9)
     parser.add_argument("--kl_scale", type=float, default=0.01)
-    parser.add_argument("--no_cuda", action="store_true", help="no cuda")
-    parser.add_argument(
-        "--uncond", action="store_true",
-        help="Generate from end-of-text as prefix"
-    )
-    parser.add_argument(
-        "--cond_text", type=str, default="The lake",
-        help="Prefix texts to condition on"
-    )
-    parser.add_argument("--num_iterations", type=int, default=3)
-    parser.add_argument("--grad_length", type=int, default=10000)
-    parser.add_argument(
-        "--num_samples",
-        type=int,
-        default=1,
-        help="Number of samples to generate from the modified latents",
-    )
-    parser.add_argument(
-        "--horizon_length",
-        type=int,
-        default=1,
-        help="Length of future to optimize over",
-    )
-    parser.add_argument(
-        "--window_length",
-        type=int,
-        default=0,
-        help="Length of past which is being optimized; "
-             "0 corresponds to infinite window length",
-    )
-    parser.add_argument("--decay", action="store_true",
-                        help="whether to decay or not")
-    parser.add_argument("--gamma", type=float, default=1.5)
+    parser.add_argument('--nocuda', action='store_true', help='no cuda')
+    parser.add_argument('--uncond', action='store_true',
+                        help='Generate from end-of-text as prefix')
+    parser.add_argument("--cond_text", type=str, default='The lake',
+                        help='Prefix texts to condition on')
+    parser.add_argument('--num_iterations', type=int, default=3)
+    parser.add_argument('--grad_length', type=int, default=10000)
+    parser.add_argument('--num_samples', type=int, default=1,
+                        help='Number of samples to generate from the modified latents')
+    parser.add_argument('--horizon_length', type=int, default=1,
+                        help='Length of future to optimize over')
+    # parser.add_argument('--force-token', action='store_true', help='no cuda')
+    parser.add_argument('--window_length', type=int, default=0,
+                        help='Length of past which is being optimizer; 0 corresponds to infinite window length')
+    parser.add_argument('--decay', action='store_true',
+                        help='whether to decay or not')
+    parser.add_argument('--gamma', type=float, default=1.5)
+    parser.add_argument('--colorama', action='store_true', help='no cuda')
 
     args = parser.parse_args()
 
-    # set Random seed
     torch.manual_seed(args.seed)
     np.random.seed(args.seed)
 
-    # set the device
-    device = torch.device("cuda" if torch.cuda.is_available() and not args.no_cuda else "cpu")
+    device = 'cpu' if args.nocuda else 'cuda'
 
-    # load pretrained model
     model = GPT2LMHeadModel.from_pretrained(
         args.model_path,
         output_hidden_states=True
@@ -720,63 +673,82 @@ def run_model():
     model.to(device)
     model.eval()
 
-    # freeze GPT-2 weights
+    # Freeze GPT-2 weights
     for param in model.parameters():
         param.requires_grad = False
+    pass
 
-    # figure out conditioning text
     if args.uncond:
-        tokenized_cond_text = TOKENIZER.encode(
-            [TOKENIZER.bos_token]
-        )
+        seq = [[50256, 50256]]
+
     else:
         raw_text = args.cond_text
         while not raw_text:
-            print("Did you forget to add `--cond_text`? ")
+            print('Did you forget to add `--cond-text`? ')
             raw_text = input("Model prompt >>> ")
-        tokenized_cond_text = TOKENIZER.encode(TOKENIZER.bos_token + raw_text)
+        seq = [[50256] + TOKENIZER.encode(raw_text)]
 
-    print("= Prefix of sentence =")
-    print(TOKENIZER.decode(tokenized_cond_text))
-    print()
+    collect_gen = dict()
+    current_index = 0
+    for out in seq:
 
-    # generate unperturbed and perturbed texts
+        text = TOKENIZER.decode(out)
+        print("=" * 40 + " Prefix of sentence " + "=" * 40)
+        print(text)
+        print("=" * 80)
 
-    # full_text_generation returns:
-    # unpert_gen_tok_text, pert_gen_tok_texts, discrim_losses, losses_in_time
-    unpert_gen_tok_text, pert_gen_tok_texts, _, _ = full_text_generation(
-        model=model, context=tokenized_cond_text, device=device, **vars(args)
-    )
+        out1, out_perturb, discrim_loss_list, loss_in_time_list, actual_words = latent_perturb(
+            model=model, args=args, context=out,
+            device=device)
 
-    # untokenize unperturbed text
-    unpert_gen_text = TOKENIZER.decode(unpert_gen_tok_text.tolist()[0])
+        text_whole = TOKENIZER.decode(out1.tolist()[0])
 
-    print("=" * 80)
-    print("= Unperturbed generated text =")
-    print(unpert_gen_text)
-    print()
+        print("=" * 80)
+        print("=" * 40 + " Whole sentence (Original)" + "=" * 40)
+        print(text_whole)
+        print("=" * 80)
 
-    generated_texts = []
+        out_perturb_copy = out_perturb
 
-    # iterate through the perturbed texts
-    for i, pert_gen_tok_text in enumerate(pert_gen_tok_texts):
-        try:
-            # untokenize unperturbed text
-            unpert_gen_text = TOKENIZER.decode(pert_gen_tok_text.tolist()[0])
+        for out_perturb in out_perturb_copy:
+            # try:
+            #    print("=" * 40 + " Whole sentence (Perturbed)" + "=" * 40)
+            #    text_whole = TOKENIZER.decode(out_perturb.tolist()[0])
+            #    print(text_whole)
+            #    print("=" * 80)
+            # except:
+            #    pass
+            # collect_gen[current_index] = [out, out_perturb, out1]
+            ## Save the prefix, perturbed seq, original seq for each index
+            print("=" * 40 + " Whole sentence (Perturbed)" + "=" * 40)
+            keyword_tokens = [aa[-1][0] for aa in
+                              actual_words] if actual_words else []
+            output_tokens = out_perturb.tolist()[0]
 
-            print("= Perturbed generated text {} =".format(i + 1))
-            print(unpert_gen_text)
-            print()
-        except:
-            pass
+            if args.colorama:
+                import colorama
 
-        # keep the prefix, perturbed seq, original seq for each index
-        generated_texts.append(
-            (tokenized_cond_text, pert_gen_tok_text, unpert_gen_tok_text)
-        )
+                text_whole = ''
+                for out in output_tokens:
+                    if out in keyword_tokens:
+                        text_whole += '%s%s%s' % (
+                        colorama.Fore.GREEN, TOKENIZER.decode([out]),
+                        colorama.Style.RESET_ALL)
+                    else:
+                        text_whole += TOKENIZER.decode([out])
+            else:
+                text_whole = TOKENIZER.decode(out_perturb.tolist()[0])
 
-    return generated_texts
+            print(text_whole)
+            print("=" * 80)
+
+            collect_gen[current_index] = [out, out_perturb, out1]
+
+            current_index = current_index + 1
 
 
-if __name__ == "__main__":
+    return
+
+
+if __name__ == '__main__':
     run_model()

From 0b51fba20bd88c4cc4acbb3e9dce82980719895c Mon Sep 17 00:00:00 2001
From: piero <piero@uber.com>
Date: Tue, 26 Nov 2019 13:15:56 -0800
Subject: [PATCH 179/505] Added script for training a discriminator for pplm to
 use

---
 examples/run_pplm.py               |  19 +-
 examples/run_pplm_discrim_train.py | 582 +++++++++++++++++++++++++++++
 2 files changed, 583 insertions(+), 18 deletions(-)
 create mode 100644 examples/run_pplm_discrim_train.py

diff --git a/examples/run_pplm.py b/examples/run_pplm.py
index 2f853d15c1..217c131b8f 100644
--- a/examples/run_pplm.py
+++ b/examples/run_pplm.py
@@ -34,6 +34,7 @@ import torch.nn.functional as F
 from torch.autograd import Variable
 from tqdm import trange
 
+from examples.run_pplm_discrim_train import ClassificationHead
 from transformers import GPT2Tokenizer
 from transformers.file_utils import cached_path
 from transformers.modeling_gpt2 import GPT2LMHeadModel
@@ -108,24 +109,6 @@ def top_k_filter(logits, k, probs=False):
                            logits)
 
 
-class ClassificationHead(torch.nn.Module):
-    """ Classification Head for the transformer """
-
-    def __init__(self, class_size=5, embed_size=2048):
-        super(ClassificationHead, self).__init__()
-        self.class_size = class_size
-        self.embed_size = embed_size
-        # self.mlp1 = torch.nn.Linear(embed_size, embed_size)
-        # self.mlp2 = (torch.nn.Linear(embed_size, class_size))
-        self.mlp = torch.nn.Linear(embed_size, class_size)
-
-    def forward(self, hidden_state):
-        # hidden_state = F.relu(self.mlp1(hidden_state))
-        # hidden_state = self.mlp2(hidden_state)
-        logits = self.mlp(hidden_state)
-        return logits
-
-
 def perturb_past(past, model, prev, args, classifier, good_index=None,
                  stepsize=0.01, vocab_size=50257,
                  original_probs=None, accumulated_hidden=None, true_past=None,
diff --git a/examples/run_pplm_discrim_train.py b/examples/run_pplm_discrim_train.py
new file mode 100644
index 0000000000..cc52234281
--- /dev/null
+++ b/examples/run_pplm_discrim_train.py
@@ -0,0 +1,582 @@
+#! /usr/bin/env python3
+# coding=utf-8
+
+# This code is licensed under a non-commercial license.
+
+import argparse
+import csv
+import json
+import math
+import time
+
+import numpy as np
+import torch
+import torch.nn.functional as F
+import torch.optim
+import torch.optim as optim
+import torch.utils.data as data
+from nltk.tokenize.treebank import TreebankWordDetokenizer
+from torchtext import data as torchtext_data
+from torchtext import datasets
+from transformers import GPT2Tokenizer, GPT2LMHeadModel
+
+torch.manual_seed(0)
+np.random.seed(0)
+EPSILON = 1e-10
+device = 'cpu'
+example_sentence = "This is incredible! I love it, this is the best chicken I have ever had."
+max_length_seq = 100
+
+
+class ClassificationHead(torch.nn.Module):
+    """Classification Head for  transformer encoders"""
+
+    def __init__(self, class_size, embed_size):
+        super(ClassificationHead, self).__init__()
+        self.class_size = class_size
+        self.embed_size = embed_size
+        # self.mlp1 = torch.nn.Linear(embed_size, embed_size)
+        # self.mlp2 = (torch.nn.Linear(embed_size, class_size))
+        self.mlp = torch.nn.Linear(embed_size, class_size)
+
+    def forward(self, hidden_state):
+        # hidden_state = F.relu(self.mlp1(hidden_state))
+        # hidden_state = self.mlp2(hidden_state)
+        logits = self.mlp(hidden_state)
+        return logits
+
+
+class Discriminator(torch.nn.Module):
+    """Transformer encoder followed by a Classification Head"""
+
+    def __init__(
+            self,
+            class_size,
+            pretrained_model="gpt2-medium",
+            cached_mode=False
+    ):
+        super(Discriminator, self).__init__()
+        self.tokenizer = GPT2Tokenizer.from_pretrained(pretrained_model)
+        self.encoder = GPT2LMHeadModel.from_pretrained(pretrained_model)
+        self.embed_size = self.encoder.transformer.config.hidden_size
+        self.classifier_head = ClassificationHead(
+            class_size=class_size,
+            embed_size=self.embed_size
+        )
+        self.cached_mode = cached_mode
+
+    def get_classifier(self):
+        return self.classifier_head
+
+    def train_custom(self):
+        for param in self.encoder.parameters():
+            param.requires_grad = False
+        pass
+        self.classifier_head.train()
+
+    def avg_representation(self, x):
+        mask = x.ne(0).unsqueeze(2).repeat(
+            1, 1, self.embed_size
+        ).float().to(device).detach()
+        hidden, _ = self.encoder.transformer(x)
+        masked_hidden = hidden * mask
+        avg_hidden = torch.sum(masked_hidden, dim=1) / (
+                torch.sum(mask, dim=1).detach() + EPSILON
+        )
+        return avg_hidden
+
+    def forward(self, x):
+        if self.cached_mode:
+            avg_hidden = x.to(device)
+        else:
+            avg_hidden = self.avg_representation(x)
+
+        logits = self.classifier_head(avg_hidden)
+        probs = F.log_softmax(logits, dim=-1)
+
+        return probs
+
+
+class Dataset(data.Dataset):
+    def __init__(self, X, y):
+        """Reads source and target sequences from txt files."""
+        self.X = X
+        self.y = y
+
+    def __len__(self):
+        return len(self.X)
+
+    def __getitem__(self, index):
+        """Returns one data pair (source and target)."""
+        data = {}
+        data['X'] = self.X[index]
+        data['y'] = self.y[index]
+        return data
+
+
+def collate_fn(data):
+    def pad_sequences(sequences):
+        lengths = [len(seq) for seq in sequences]
+
+        padded_sequences = torch.zeros(
+            len(sequences),
+            max(lengths)
+        ).long()  # padding index 0
+
+        for i, seq in enumerate(sequences):
+            end = lengths[i]
+            padded_sequences[i, :end] = seq[:end]
+
+        return padded_sequences, lengths
+
+    item_info = {}
+    for key in data[0].keys():
+        item_info[key] = [d[key] for d in data]
+
+    x_batch, _ = pad_sequences(item_info['X'])
+    y_batch = torch.tensor(item_info['y'], dtype=torch.long)
+
+    return x_batch, y_batch
+
+
+def cached_collate_fn(data):
+    item_info = {}
+    for key in data[0].keys():
+        item_info[key] = [d[key] for d in data]
+
+    x_batch = torch.cat(item_info['X'], 0)
+    y_batch = torch.tensor(item_info['y'], dtype=torch.long)
+
+    return x_batch, y_batch
+
+
+def train_epoch(data_loader, discriminator, optimizer,
+                epoch=0, log_interval=10):
+    samples_so_far = 0
+    discriminator.train_custom()
+    for batch_idx, (input_t, target_t) in enumerate(data_loader):
+        input_t, target_t = input_t.to(device), target_t.to(device)
+
+        optimizer.zero_grad()
+
+        output_t = discriminator(input_t)
+        loss = F.nll_loss(output_t, target_t)
+        loss.backward(retain_graph=True)
+        optimizer.step()
+
+        samples_so_far += len(input_t)
+
+        if batch_idx % log_interval == 0:
+            print(
+                'Train Epoch: {} [{}/{} ({:.0f}%)]\tLoss: {:.6f}'.format(
+                    epoch + 1,
+                    samples_so_far, len(data_loader.dataset),
+                    100 * samples_so_far / len(data_loader.dataset), loss.item()
+                )
+            )
+
+
+def evaluate_performance(data_loader, discriminator):
+    discriminator.eval()
+    test_loss = 0
+    correct = 0
+    with torch.no_grad():
+        for input_t, target_t in data_loader:
+            input_t, target_t = input_t.to(device), target_t.to(device)
+            output_t = discriminator(input_t)
+            # sum up batch loss
+            test_loss += F.nll_loss(output_t, target_t, reduction='sum').item()
+            # get the index of the max log-probability
+            pred_t = output_t.argmax(dim=1, keepdim=True)
+            correct += pred_t.eq(target_t.view_as(pred_t)).sum().item()
+
+    test_loss /= len(data_loader.dataset)
+
+    print(
+        'Performance on test set: '
+        'Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)'.format(
+            test_loss, correct, len(data_loader.dataset),
+            100. * correct / len(data_loader.dataset)
+        )
+    )
+
+
+def predict(input_sentence, model, classes, cached=False):
+    input_t = model.tokenizer.encode(input_sentence)
+    input_t = torch.tensor([input_t], dtype=torch.long)
+    if cached:
+        input_t = model.avg_representation(input_t)
+
+    log_probs = model(input_t).data.cpu().numpy().flatten().tolist()
+    print('Input sentence:', input_sentence)
+    print('Predictions:', ", ".join(
+        "{}: {:.4f}".format(c, math.exp(log_prob)) for c, log_prob in
+        zip(classes, log_probs)
+    ))
+
+
+def get_cached_data_loader(dataset, batch_size, discriminator, shuffle=False):
+    data_loader = torch.utils.data.DataLoader(dataset=dataset,
+                                              batch_size=batch_size,
+                                              collate_fn=collate_fn)
+
+    xs = []
+    ys = []
+    for batch_idx, (x, y) in enumerate(data_loader):
+        with torch.no_grad():
+            x = x.to(device)
+            avg_rep = discriminator.avg_representation(x).cpu().detach()
+            avg_rep_list = torch.unbind(avg_rep.unsqueeze(1))
+            xs += avg_rep_list
+            ys += y.cpu().numpy().tolist()
+
+    data_loader = torch.utils.data.DataLoader(
+        dataset=Dataset(xs, ys),
+        batch_size=batch_size,
+        shuffle=shuffle,
+        collate_fn=cached_collate_fn)
+
+    return data_loader
+
+
+def train_discriminator(
+        dataset, dataset_fp=None, pretrained_model='gpt2-medium',
+        epochs=10, batch_size=64, log_interval=10,
+        save_model=False, cached=False, use_cuda=False):
+    if use_cuda:
+        global device
+        device = 'cuda'
+
+    print('Preprocessing {} dataset...'.format(dataset))
+    start = time.time()
+
+    if dataset == 'SST':
+        idx2class = ["positive", "negative", "very positive", "very negative",
+                     "neutral"]
+        class2idx = {c: i for i, c in enumerate(idx2class)}
+
+        discriminator = Discriminator(
+            class_size=len(idx2class),
+            pretrained_model=pretrained_model,
+            cached_mode=cached
+        ).to(device)
+
+        text = torchtext_data.Field()
+        label = torchtext_data.Field(sequential=False)
+        train_data, val_data, test_data = datasets.SST.splits(
+            text,
+            label,
+            fine_grained=True,
+            train_subtrees=True,
+        )
+
+        x = []
+        y = []
+        for i in range(len(train_data)):
+            seq = TreebankWordDetokenizer().detokenize(
+                vars(train_data[i])["text"]
+            )
+            seq = discriminator.tokenizer.encode(seq)
+            seq = torch.tensor([50256] + seq, device=device, dtype=torch.long)
+            x.append(seq)
+            y.append(class2idx[vars(train_data[i])["label"]])
+        train_dataset = Dataset(x, y)
+
+        test_x = []
+        test_y = []
+        for i in range(len(test_data)):
+            seq = TreebankWordDetokenizer().detokenize(
+                vars(test_data[i])["text"]
+            )
+            seq = discriminator.tokenizer.encode(seq)
+            seq = torch.tensor([50256] + seq, device=device, dtype=torch.long)
+            test_x.append(seq)
+            test_y.append(class2idx[vars(test_data[i])["label"]])
+        test_dataset = Dataset(test_x, test_y)
+
+        discriminator_meta = {
+            "class_size": len(idx2class),
+            "embed_size": discriminator.embed_size,
+            "pretrained_model": pretrained_model,
+            "class_vocab": class2idx,
+            "default_class": 2,
+        }
+
+    elif dataset == 'clickbait':
+        idx2class = ["non_clickbait", "clickbait"]
+        class2idx = {c: i for i, c in enumerate(idx2class)}
+
+        discriminator = Discriminator(
+            class_size=len(idx2class),
+            pretrained_model=pretrained_model,
+            cached_mode=cached
+        ).to(device)
+
+        with open("datasets/clickbait/clickbait_train_prefix.txt") as f:
+            data = []
+            for i, line in enumerate(f):
+                try:
+                    data.append(eval(line))
+                except:
+                    print('Error evaluating line {}: {}'.format(
+                        i, line
+                    ))
+                    continue
+        x = []
+        y = []
+        y = []
+        for i, d in enumerate(data):
+            try:
+                seq = discriminator.tokenizer.encode(d["text"])
+
+                if len(seq) < max_length_seq:
+                    seq = torch.tensor(
+                        [50256] + seq, device=device, dtype=torch.long
+                    )
+                else:
+                    print("Line {} is longer than maximum length {}".format(
+                        i, max_length_seq
+                    ))
+                    continue
+                x.append(seq)
+                y.append(d['label'])
+            except:
+                print("Error tokenizing line {}, skipping it".format(i))
+                pass
+
+        full_dataset = Dataset(x, y)
+        train_size = int(0.9 * len(full_dataset))
+        test_size = len(full_dataset) - train_size
+        train_dataset, test_dataset = torch.utils.data.random_split(
+            full_dataset, [train_size, test_size]
+        )
+
+        discriminator_meta = {
+            "class_size": len(idx2class),
+            "embed_size": discriminator.embed_size,
+            "pretrained_model": pretrained_model,
+            "class_vocab": class2idx,
+            "default_class": 1,
+        }
+
+    elif dataset == 'toxic':
+        idx2class = ["non_toxic", "toxic"]
+        class2idx = {c: i for i, c in enumerate(idx2class)}
+
+        discriminator = Discriminator(
+            class_size=len(idx2class),
+            pretrained_model=pretrained_model,
+            cached_mode=cached
+        ).to(device)
+
+        with open("datasets/toxic/toxic_train.txt") as f:
+            data = []
+            for i, line in enumerate(f):
+                try:
+                    data.append(eval(line))
+                except:
+                    print('Error evaluating line {}: {}'.format(
+                        i, line
+                    ))
+                    continue
+
+        x = []
+        y = []
+        for i, d in enumerate(data):
+            try:
+                seq = discriminator.tokenizer.encode(d["text"])
+
+                if len(seq) < max_length_seq:
+                    seq = torch.tensor(
+                        [50256] + seq, device=device, dtype=torch.long
+                    )
+                else:
+                    print("Line {} is longer than maximum length {}".format(
+                        i, max_length_seq
+                    ))
+                    continue
+                x.append(seq)
+                y.append(int(np.sum(d['label']) > 0))
+            except:
+                print("Error tokenizing line {}, skipping it".format(i))
+                pass
+
+        full_dataset = Dataset(x, y)
+        train_size = int(0.9 * len(full_dataset))
+        test_size = len(full_dataset) - train_size
+        train_dataset, test_dataset = torch.utils.data.random_split(
+            full_dataset, [train_size, test_size]
+        )
+
+        discriminator_meta = {
+            "class_size": len(idx2class),
+            "embed_size": discriminator.embed_size,
+            "pretrained_model": pretrained_model,
+            "class_vocab": class2idx,
+            "default_class": 0,
+        }
+
+    else:  # if dataset == 'generic':
+        # This assumes the input dataset is a TSV with the following structure:
+        # class \t text
+
+        if dataset_fp is None:
+            raise ValueError('When generic dataset is selected, '
+                             'dataset_fp needs to be specified aswell.')
+
+        classes = set()
+        with open(dataset_fp) as f:
+            csv_reader = csv.reader(f, delimiter='\t')
+            for row in csv_reader:
+                classes.add(row[0])
+
+        idx2class = sorted(classes)
+        class2idx = {c: i for i, c in enumerate(idx2class)}
+
+        discriminator = Discriminator(
+            class_size=len(idx2class),
+            pretrained_model=pretrained_model,
+            cached_mode=cached
+        ).to(device)
+
+        x = []
+        y = []
+        with open(dataset_fp) as f:
+            csv_reader = csv.reader(f, delimiter='\t')
+            for i, row in enumerate(csv_reader):
+                label = row[0]
+                text = row[1]
+
+                try:
+                    seq = discriminator.tokenizer.encode(text)
+                    if (len(seq) < max_length_seq):
+                        seq = torch.tensor(
+                            [50256] + seq,
+                            device=device,
+                            dtype=torch.long
+                        )
+
+                    else:
+                        print("Line {} is longer than maximum length {}".format(
+                            i, max_length_seq
+                        ))
+                        continue
+
+                    x.append(seq)
+                    y.append(class2idx[label])
+
+                except:
+                    print("Error tokenizing line {}, skipping it".format(i))
+                    pass
+
+        full_dataset = Dataset(x, y)
+        train_size = int(0.9 * len(full_dataset))
+        test_size = len(full_dataset) - train_size
+        train_dataset, test_dataset = torch.utils.data.random_split(
+            full_dataset,
+            [train_size, test_size]
+        )
+
+        discriminator_meta = {
+            "class_size": len(idx2class),
+            "embed_size": discriminator.embed_size,
+            "pretrained_model": pretrained_model,
+            "class_vocab": class2idx,
+            "default_class": 0,
+        }
+
+    end = time.time()
+    print('Preprocessed {} data points'.format(
+        len(train_dataset) + len(test_dataset))
+    )
+    print("Data preprocessing took: {:.3f}s".format(end - start))
+
+    if cached:
+        start = time.time()
+
+        train_loader = get_cached_data_loader(
+            train_dataset, batch_size, discriminator, shuffle=True
+        )
+
+        test_loader = get_cached_data_loader(
+            test_dataset, batch_size, discriminator
+        )
+
+        end = time.time()
+        print("Building representation cache took: {:.3f}s".format(end - start))
+
+    else:
+        train_loader = torch.utils.data.DataLoader(dataset=train_dataset,
+                                                   batch_size=batch_size,
+                                                   shuffle=True,
+                                                   collate_fn=collate_fn)
+        test_loader = torch.utils.data.DataLoader(dataset=test_dataset,
+                                                  batch_size=batch_size,
+                                                  collate_fn=collate_fn)
+
+    if save_model:
+        with open("{}_classifier_head_meta.json".format(dataset),
+                  "w") as meta_file:
+            json.dump(discriminator_meta, meta_file)
+
+    optimizer = optim.Adam(discriminator.parameters(), lr=0.0001)
+
+    for epoch in range(epochs):
+        start = time.time()
+        print('\nEpoch', epoch + 1)
+
+        train_epoch(
+            discriminator=discriminator,
+            data_loader=train_loader,
+            optimizer=optimizer,
+            epoch=epoch,
+            log_interval=log_interval
+        )
+        evaluate_performance(
+            data_loader=test_loader,
+            discriminator=discriminator
+        )
+
+        end = time.time()
+        print("Epoch took: {:.3f}s".format(end - start))
+
+        print("\nExample prediction")
+        predict(example_sentence, discriminator, idx2class, cached)
+
+        if save_model:
+            # torch.save(discriminator.state_dict(),
+            #           "{}_discriminator_{}.pt".format(
+            #               args.dataset, epoch
+            #               ))
+            torch.save(discriminator.get_classifier().state_dict(),
+                       "{}_classifier_head_epoch_{}.pt".format(dataset, epoch))
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser(
+        description='Train a discriminator on top of GPT-2 representations')
+    parser.add_argument('--dataset', type=str, default='SST',
+                        choices=('SST', 'clickbait', 'toxic', 'generic'),
+                        help='dataset to train the discriminator on.'
+                             'In case of generic, the dataset is expected'
+                             'to be a TSBV file with structure: class \\t text')
+    parser.add_argument('--dataset_fp', type=str, default='',
+                        help='File path of the dataset to use. '
+                             'Needed only in case of generic datadset')
+    parser.add_argument('--pretrained_model', type=str, default='gpt2-medium',
+                        help='Pretrained model to use as encoder')
+    parser.add_argument('--epochs', type=int, default=10, metavar='N',
+                        help='Number of training epochs')
+    parser.add_argument('--batch_size', type=int, default=64, metavar='N',
+                        help='input batch size for training (default: 64)')
+    parser.add_argument('--log_interval', type=int, default=10, metavar='N',
+                        help='how many batches to wait before logging training status')
+    parser.add_argument('--save_model', action='store_true',
+                        help='whether to save the model')
+    parser.add_argument('--cached', action='store_true',
+                        help='whether to cache the input representations')
+    parser.add_argument('--use_cuda', action='store_true',
+                        help='use to turn on cuda')
+    args = parser.parse_args()
+
+    train_discriminator(**(vars(args)))

From 7469d03b1ca3c1c920e4cadc8a007609d17aff50 Mon Sep 17 00:00:00 2001
From: w4nderlust <w4nderlust@gmail.com>
Date: Tue, 26 Nov 2019 21:30:57 -0800
Subject: [PATCH 180/505] Fixed minor bug when running training on cuda

---
 examples/run_pplm_discrim_train.py | 49 ++++++++++++++++--------------
 1 file changed, 26 insertions(+), 23 deletions(-)

diff --git a/examples/run_pplm_discrim_train.py b/examples/run_pplm_discrim_train.py
index cc52234281..7f10e861a8 100644
--- a/examples/run_pplm_discrim_train.py
+++ b/examples/run_pplm_discrim_train.py
@@ -18,6 +18,7 @@ import torch.utils.data as data
 from nltk.tokenize.treebank import TreebankWordDetokenizer
 from torchtext import data as torchtext_data
 from torchtext import datasets
+
 from transformers import GPT2Tokenizer, GPT2LMHeadModel
 
 torch.manual_seed(0)
@@ -89,7 +90,7 @@ class Discriminator(torch.nn.Module):
         if self.cached_mode:
             avg_hidden = x.to(device)
         else:
-            avg_hidden = self.avg_representation(x)
+            avg_hidden = self.avg_representation(x.to(device))
 
         logits = self.classifier_head(avg_hidden)
         probs = F.log_softmax(logits, dim=-1)
@@ -203,7 +204,7 @@ def evaluate_performance(data_loader, discriminator):
 
 def predict(input_sentence, model, classes, cached=False):
     input_t = model.tokenizer.encode(input_sentence)
-    input_t = torch.tensor([input_t], dtype=torch.long)
+    input_t = torch.tensor([input_t], dtype=torch.long, device=device)
     if cached:
         input_t = model.avg_representation(input_t)
 
@@ -428,7 +429,8 @@ def train_discriminator(
         with open(dataset_fp) as f:
             csv_reader = csv.reader(f, delimiter='\t')
             for row in csv_reader:
-                classes.add(row[0])
+                if row:
+                    classes.add(row[0])
 
         idx2class = sorted(classes)
         class2idx = {c: i for i, c in enumerate(idx2class)}
@@ -444,30 +446,31 @@ def train_discriminator(
         with open(dataset_fp) as f:
             csv_reader = csv.reader(f, delimiter='\t')
             for i, row in enumerate(csv_reader):
-                label = row[0]
-                text = row[1]
+                if row:
+                    label = row[0]
+                    text = row[1]
 
-                try:
-                    seq = discriminator.tokenizer.encode(text)
-                    if (len(seq) < max_length_seq):
-                        seq = torch.tensor(
-                            [50256] + seq,
-                            device=device,
-                            dtype=torch.long
-                        )
+                    try:
+                        seq = discriminator.tokenizer.encode(text)
+                        if (len(seq) < max_length_seq):
+                            seq = torch.tensor(
+                                [50256] + seq,
+                                device=device,
+                                dtype=torch.long
+                            )
 
-                    else:
-                        print("Line {} is longer than maximum length {}".format(
-                            i, max_length_seq
-                        ))
-                        continue
+                        else:
+                            print("Line {} is longer than maximum length {}".format(
+                                i, max_length_seq
+                            ))
+                            continue
 
-                    x.append(seq)
-                    y.append(class2idx[label])
+                        x.append(seq)
+                        y.append(class2idx[label])
 
-                except:
-                    print("Error tokenizing line {}, skipping it".format(i))
-                    pass
+                    except:
+                        print("Error tokenizing line {}, skipping it".format(i))
+                        pass
 
         full_dataset = Dataset(x, y)
         train_size = int(0.9 * len(full_dataset))

From 821de121e86574504ec648f76ccb924e38125b52 Mon Sep 17 00:00:00 2001
From: piero <piero@uber.com>
Date: Wed, 27 Nov 2019 15:27:49 -0800
Subject: [PATCH 181/505] Minor changes

---
 examples/run_pplm_discrim_train.py | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/examples/run_pplm_discrim_train.py b/examples/run_pplm_discrim_train.py
index 7f10e861a8..9438cbbac1 100644
--- a/examples/run_pplm_discrim_train.py
+++ b/examples/run_pplm_discrim_train.py
@@ -72,7 +72,6 @@ class Discriminator(torch.nn.Module):
     def train_custom(self):
         for param in self.encoder.parameters():
             param.requires_grad = False
-        pass
         self.classifier_head.train()
 
     def avg_representation(self, x):
@@ -122,7 +121,7 @@ def collate_fn(data):
         padded_sequences = torch.zeros(
             len(sequences),
             max(lengths)
-        ).long()  # padding index 0
+        ).long()  # padding value = 0
 
         for i, seq in enumerate(sequences):
             end = lengths[i]

From 4f2164e40e803677869865b140b2b3f5a96d4bcd Mon Sep 17 00:00:00 2001
From: piero <piero@uber.com>
Date: Wed, 27 Nov 2019 16:32:45 -0800
Subject: [PATCH 182/505] First cleanup step, changing function names and
 passing parameters all the way through without using args. Identical output
 as before.

---
 examples/run_pplm.py | 270 +++++++++++++++++++++++++++++--------------
 1 file changed, 182 insertions(+), 88 deletions(-)

diff --git a/examples/run_pplm.py b/examples/run_pplm.py
index 217c131b8f..0d9ed86f45 100644
--- a/examples/run_pplm.py
+++ b/examples/run_pplm.py
@@ -109,20 +109,40 @@ def top_k_filter(logits, k, probs=False):
                            logits)
 
 
-def perturb_past(past, model, prev, args, classifier, good_index=None,
-                 stepsize=0.01, vocab_size=50257,
-                 original_probs=None, accumulated_hidden=None, true_past=None,
-                 grad_norms=None):
-    window_length = args.window_length
-    gm_scale, kl_scale = args.gm_scale, args.kl_scale
-    one_hot_vectors = []
-    for good_list in good_index:
-        good_list = list(filter(lambda x: len(x) <= 1, good_list))
-        good_list = torch.tensor(good_list).cuda()
-        num_good = good_list.shape[0]
-        one_hot_good = torch.zeros(num_good, vocab_size).cuda()
-        one_hot_good.scatter_(1, good_list, 1)
-        one_hot_vectors.append(one_hot_good)
+def perturb_past(
+        past,
+        model,
+        prev,
+        unpert_past=None,
+        unpert_logits=None,
+        accumulated_hidden=None,
+        grad_norms=None,
+        stepsize=0.01,
+        classifier=None,
+        label_class=None,
+        one_hot_bows_vectors=None,
+        loss_type=0,
+        num_iterations=3,
+        kl_scale=0.01,
+        window_length=0,
+        horizon_length=1,
+        decay=False,
+        gamma=1.5,
+):
+
+    #def perturb_past(past, model, prev, classifier, good_index=None,
+    #             stepsize=0.01, vocab_size=50257,
+    #             original_probs=None, accumulated_hidden=None, true_past=None,
+    #             grad_norms=None):
+
+    # one_hot_bows_vectors = []
+    # for good_list in good_index:
+    #     good_list = list(filter(lambda x: len(x) <= 1, good_list))
+    #     good_list = torch.tensor(good_list).cuda()
+    #     num_good = good_list.shape[0]
+    #     one_hot_good = torch.zeros(num_good, vocab_size).cuda()
+    #     one_hot_good.scatter_(1, good_list, 1)
+    #     one_hot_bows_vectors.append(one_hot_good)
 
     # Generate inital perturbed past
     past_perturb_orig = [
@@ -132,7 +152,7 @@ def perturb_past(past, model, prev, args, classifier, good_index=None,
     if accumulated_hidden is None:
         accumulated_hidden = 0
 
-    if args.decay:
+    if decay:
         decay_mask = torch.arange(0., 1.0 + SmallConst, 1.0 / (window_length))[
                      1:]
     else:
@@ -160,7 +180,7 @@ def perturb_past(past, model, prev, args, classifier, good_index=None,
         window_mask = torch.ones_like(past[0]).cuda()
 
     loss_per_iter = []
-    for i in range(args.num_iterations):
+    for i in range(num_iterations):
         print("Iteration ", i + 1)
         past_perturb = [torch.from_numpy(p_) for p_ in past_perturb_orig]
         past_perturb = [to_var(p_, requires_grad=True) for p_ in past_perturb]
@@ -183,8 +203,8 @@ def perturb_past(past, model, prev, args, classifier, good_index=None,
         probabs = F.softmax(logits, dim=-1)
         loss = 0.0
         loss_list = []
-        if args.loss_type == 1 or args.loss_type == 3:
-            for one_hot_good in one_hot_vectors:
+        if loss_type == 1 or loss_type == 3:
+            for one_hot_good in one_hot_bows_vectors:
                 good_logits = torch.mm(probabs, torch.t(one_hot_good))
                 loss_word = good_logits
                 loss_word = torch.sum(loss_word)
@@ -194,10 +214,10 @@ def perturb_past(past, model, prev, args, classifier, good_index=None,
                 loss_list.append(loss_word)
             print(" pplm_bow_loss:", loss.data.cpu().numpy())
 
-        if args.loss_type == 2 or args.loss_type == 3:
+        if loss_type == 2 or loss_type == 3:
             ce_loss = torch.nn.CrossEntropyLoss()
-            new_true_past = true_past
-            for i in range(args.horizon_length):
+            new_true_past = unpert_past
+            for i in range(horizon_length):
                 future_probabs = F.softmax(logits, dim=-1)  # Get softmax
                 future_probabs = torch.unsqueeze(future_probabs, dim=1)
 
@@ -217,9 +237,9 @@ def perturb_past(past, model, prev, args, classifier, good_index=None,
                     future_hidden, dim=1)
 
             predicted_sentiment = classifier(new_accumulated_hidden / (
-                        current_length + 1 + args.horizon_length))
+                        current_length + 1 + horizon_length))
 
-            label = torch.tensor([args.label_class], device='cuda',
+            label = torch.tensor([label_class], device='cuda',
                                  dtype=torch.long)
             discrim_loss = ce_loss(predicted_sentiment, label)
             print(" pplm_discrim_loss:", discrim_loss.data.cpu().numpy())
@@ -228,7 +248,7 @@ def perturb_past(past, model, prev, args, classifier, good_index=None,
 
         kl_loss = 0.0
         if kl_scale > 0.0:
-            p = (F.softmax(original_probs[:, -1, :], dim=-1))
+            p = (F.softmax(unpert_logits[:, -1, :], dim=-1))
             p = p + SmallConst * (p <= SmallConst).type(
                 torch.FloatTensor).cuda().detach()
             correction = SmallConst * (probabs <= SmallConst).type(
@@ -244,7 +264,7 @@ def perturb_past(past, model, prev, args, classifier, good_index=None,
         print(' pplm_loss', (loss - kl_loss).data.cpu().numpy())
 
         loss.backward()
-        if grad_norms is not None and args.loss_type == 1:
+        if grad_norms is not None and loss_type == 1:
             grad_norms = [
                 torch.max(grad_norms[index], torch.norm(p_.grad * window_mask))
                 for index, p_ in
@@ -255,7 +275,7 @@ def perturb_past(past, model, prev, args, classifier, good_index=None,
 
         grad = [
             -stepsize * (p_.grad * window_mask / grad_norms[
-                index] ** args.gamma).data.cpu().numpy()
+                index] ** gamma).data.cpu().numpy()
             for index, p_ in enumerate(past_perturb)]
         past_perturb_orig = list(map(add, grad, past_perturb_orig))
 
@@ -347,10 +367,32 @@ def build_bows_one_hot_vectors(bow_indices):
     return one_hot_bows_vectors
 
 
-def latent_perturb(model, args, context=None, sample=True, device='cuda'):
+def full_text_generation(
+            model,
+            context=None,
+            num_samples=1,
+            device="cuda",
+            sample=True,
+            discrim=None,
+            label_class=None,
+            bag_of_words=None,
+            length=100,
+            grad_length=10000,
+            stepsize=0.02,
+            num_iterations=3,
+            temperature=1.0,
+            gm_scale=0.9,
+            kl_scale=0.01,
+            top_k=10,
+            window_length=0,
+            horizon_length=1,
+            decay=False,
+            gamma=1.5,
+            **kwargs
+    ):
     classifier, class_id = get_classifier(
-        args.discrim,
-        args.label_class,
+        discrim,
+        label_class,
         device
     )
 
@@ -422,49 +464,68 @@ def latent_perturb(model, args, context=None, sample=True, device='cuda'):
     #         good_list = list(filter(lambda x: len(x) <= 1, good_list))
     #         actual_words = [(TOKENIZER.decode(ww).strip(),ww) for ww in good_list]
 
-    good_index = []
+    bow_indices = []
     actual_words = None
-    if args.bag_of_words:
-        good_index = get_bag_of_words_indices(args.bag_of_words.split(";"))
+    if bag_of_words:
+        bow_indices = get_bag_of_words_indices(bag_of_words.split(";"))
 
-        for good_list in good_index:
+        for good_list in bow_indices:
             good_list = list(filter(lambda x: len(x) <= 1, good_list))
             actual_words = [(TOKENIZER.decode(ww).strip(), ww) for ww in
                             good_list]
 
-    if args.bag_of_words and classifier:
+    if bag_of_words and classifier:
         print("Both PPLM-BoW and PPLM-Discrim are on. This is not optimized.")
-        args.loss_type = PPLM_BOW_DISCRIM
+        loss_type = PPLM_BOW_DISCRIM
 
-    elif args.bag_of_words:
-        args.loss_type = PPLM_BOW
+    elif bag_of_words:
+        loss_type = PPLM_BOW
         print("Using PPLM-BoW")
 
     elif classifier is not None:
-        args.loss_type = PPLM_DISCRIM
+        loss_type = PPLM_DISCRIM
         print("Using PPLM-Discrim")
 
     else:
         raise Exception("Specify either --bag_of_words (-B) or --discrim (-D)")
 
-    original, _, _ = sample_from_hidden(model=model, args=args, context=context,
-                                        device=device,
-                                        perturb=False, good_index=good_index,
-                                        classifier=classifier)
+    original, _, _ = generate_text_pplm(
+        model=model,
+        context=context,
+        device=device,
+        length=length,
+        perturb=False
+    )
     torch.cuda.empty_cache()
 
     perturbed_list = []
     discrim_loss_list = []
     loss_in_time_list = []
 
-    for i in range(args.num_samples):
-        perturbed, discrim_loss, loss_in_time = sample_from_hidden(model=model,
-                                                                   args=args,
-                                                                   context=context,
-                                                                   device=device,
-                                                                   perturb=True,
-                                                                   good_index=good_index,
-                                                                   classifier=classifier)
+    for i in range(num_samples):
+        perturbed, discrim_loss, loss_in_time = generate_text_pplm(
+            model=model,
+            context=context,
+            device=device,
+            sample=sample,
+            perturb=True,
+            bow_indices=bow_indices,
+            classifier=classifier,
+            label_class=class_id,
+            loss_type=loss_type,
+            length=length,
+            grad_length=grad_length,
+            stepsize=stepsize,
+            num_iterations=num_iterations,
+            temperature=temperature,
+            gm_scale=gm_scale,
+            kl_scale=kl_scale,
+            top_k=top_k,
+            window_length=window_length,
+            horizon_length=horizon_length,
+            decay=decay,
+            gamma=gamma,
+        )
         perturbed_list.append(perturbed)
         if classifier is not None:
             discrim_loss_list.append(discrim_loss.data.cpu().numpy())
@@ -475,15 +536,40 @@ def latent_perturb(model, args, context=None, sample=True, device='cuda'):
     return original, perturbed_list, discrim_loss_list, loss_in_time_list, actual_words
 
 
-def sample_from_hidden(model, args, classifier, context=None, past=None,
-                       device='cuda',
-                       sample=True, perturb=True, good_index=None):
+
+def generate_text_pplm(
+        model,
+        context=None,
+        past=None,
+        device="cuda",
+        sample=True,
+        perturb=True,
+        classifier=None,
+        label_class=None,
+        bow_indices=None,
+        loss_type=0,
+        length=100,
+        grad_length=10000,
+        stepsize=0.02,
+        num_iterations=3,
+        temperature=1.0,
+        gm_scale=0.9,
+        kl_scale=0.01,
+        top_k=10,
+        window_length=0,
+        horizon_length=1,
+        decay=False,
+        gamma=1.5,
+):
     output = torch.tensor(context, device=device, dtype=torch.long).unsqueeze(
         0) if context else None
 
+    # collect one hot vectors for bags of words
+    one_hot_bows_vectors = build_bows_one_hot_vectors(bow_indices)
+
     grad_norms = None
     loss_in_time = []
-    for i in trange(args.length, ascii=True):
+    for i in trange(length, ascii=True):
 
         # Get past/probs for current output, except for last word
         # Note that GPT takes 2 inputs: past + current-token
@@ -497,7 +583,7 @@ def sample_from_hidden(model, args, classifier, context=None, past=None,
 
             # Piero modified model call
             _, past, _ = model(output[:, :-1])
-            original_probs, true_past, unpert_all_hidden = model(output)
+            unpert_logits, unpert_past, unpert_all_hidden = model(output)
             true_hidden = unpert_all_hidden[-1]
 
         else:
@@ -505,17 +591,17 @@ def sample_from_hidden(model, args, classifier, context=None, past=None,
             # true_hidden = model.hidden_states
 
             # Piero modified model call
-            original_probs, true_past, unpert_all_hidden = model(output)
+            unpert_logits, unpert_past, unpert_all_hidden = model(output)
             true_hidden = unpert_all_hidden[-1]
 
         # Modify the past if necessary
 
-        if i >= args.grad_length:
-            current_stepsize = args.stepsize * 0
+        if i >= grad_length:
+            current_stepsize = stepsize * 0
         else:
-            current_stepsize = args.stepsize
+            current_stepsize = stepsize
 
-        if not perturb or args.num_iterations == 0:
+        if not perturb or num_iterations == 0:
             perturbed_past = past
 
         else:
@@ -524,17 +610,26 @@ def sample_from_hidden(model, args, classifier, context=None, past=None,
             accumulated_hidden = true_hidden[:, :-1, :]
             accumulated_hidden = torch.sum(accumulated_hidden, dim=1)
 
-            perturbed_past, _, grad_norms, loss_per_iter = perturb_past(past,
-                                                                        model,
-                                                                        prev,
-                                                                        args,
-                                                                        good_index=good_index,
-                                                                        stepsize=current_stepsize,
-                                                                        original_probs=original_probs,
-                                                                        true_past=true_past,
-                                                                        accumulated_hidden=accumulated_hidden,
-                                                                        classifier=classifier,
-                                                                        grad_norms=grad_norms)
+            perturbed_past, _, grad_norms, loss_per_iter = perturb_past(
+                    past,
+                    model,
+                    prev,
+                    unpert_past=unpert_past,
+                    unpert_logits=unpert_logits,
+                    accumulated_hidden=accumulated_hidden,
+                    grad_norms=grad_norms,
+                    stepsize=current_stepsize,
+                    classifier=classifier,
+                    label_class=label_class,
+                    one_hot_bows_vectors=one_hot_bows_vectors,
+                    loss_type=loss_type,
+                    num_iterations=num_iterations,
+                    kl_scale=kl_scale,
+                    window_length=window_length,
+                    horizon_length=horizon_length,
+                    decay=decay,
+                    gamma=gamma,
+                )
             loss_in_time.append(loss_per_iter)
 
         # Piero modified model call
@@ -546,7 +641,7 @@ def sample_from_hidden(model, args, classifier, context=None, past=None,
         if classifier is not None:
             ce_loss = torch.nn.CrossEntropyLoss()
             predicted_sentiment = classifier(torch.mean(true_hidden, dim=1))
-            label = torch.tensor([args.label_class], device='cuda',
+            label = torch.tensor([label_class], device='cuda',
                                  dtype=torch.long)
             true_discrim_loss = ce_loss(predicted_sentiment, label)
             print("true discrim loss", true_discrim_loss.data.cpu().numpy())
@@ -556,7 +651,7 @@ def sample_from_hidden(model, args, classifier, context=None, past=None,
         # Piero modified model call
         # hidden = model.hidden_states  # update hidden
         # logits = model.forward_hidden(hidden)
-        logits = logits[:, -1, :] / args.temperature  # + SmallConst
+        logits = logits[:, -1, :] / temperature  # + SmallConst
 
         # logits = top_k_filter(logits, k=args.top_k)  # + SmallConst
 
@@ -566,22 +661,21 @@ def sample_from_hidden(model, args, classifier, context=None, past=None,
         if perturb:
 
             # original_probs = top_k_filter(original_probs[:, -1, :]) #+ SmallConst
-            original_probs = F.softmax(original_probs[:, -1, :], dim=-1)
+            unpert_logits = F.softmax(unpert_logits[:, -1, :], dim=-1)
             # likelywords = torch.topk(original_probs, k=10, dim=-1)
             # print(TOKENIZER.decode(likelywords[1].tolist()[0]))
 
-            gm_scale = args.gm_scale
             log_probs = ((log_probs ** gm_scale) * (
-                        original_probs ** (1 - gm_scale)))  # + SmallConst
+                        unpert_logits ** (1 - gm_scale)))  # + SmallConst
 
-            log_probs = top_k_filter(log_probs, k=args.top_k,
+            log_probs = top_k_filter(log_probs, k=top_k,
                                      probs=True)  # + SmallConst
 
             if torch.sum(log_probs) <= 1:
                 log_probs = log_probs / torch.sum(log_probs)
 
         else:
-            logits = top_k_filter(logits, k=args.top_k)  # + SmallConst
+            logits = top_k_filter(logits, k=top_k)  # + SmallConst
             log_probs = F.softmax(logits, dim=-1)
 
         if sample:
@@ -673,16 +767,16 @@ def run_model():
 
     collect_gen = dict()
     current_index = 0
-    for out in seq:
+    for tokenized_cond_text in seq:
 
-        text = TOKENIZER.decode(out)
+        text = TOKENIZER.decode(tokenized_cond_text)
         print("=" * 40 + " Prefix of sentence " + "=" * 40)
         print(text)
         print("=" * 80)
 
-        out1, out_perturb, discrim_loss_list, loss_in_time_list, actual_words = latent_perturb(
-            model=model, args=args, context=out,
-            device=device)
+        out1, out_perturb, discrim_loss_list, loss_in_time_list, actual_words = full_text_generation(
+            model=model, context=tokenized_cond_text, device=device, **vars(args)
+        )
 
         text_whole = TOKENIZER.decode(out1.tolist()[0])
 
@@ -712,20 +806,20 @@ def run_model():
                 import colorama
 
                 text_whole = ''
-                for out in output_tokens:
-                    if out in keyword_tokens:
+                for tokenized_cond_text in output_tokens:
+                    if tokenized_cond_text in keyword_tokens:
                         text_whole += '%s%s%s' % (
-                        colorama.Fore.GREEN, TOKENIZER.decode([out]),
+                        colorama.Fore.GREEN, TOKENIZER.decode([tokenized_cond_text]),
                         colorama.Style.RESET_ALL)
                     else:
-                        text_whole += TOKENIZER.decode([out])
+                        text_whole += TOKENIZER.decode([tokenized_cond_text])
             else:
                 text_whole = TOKENIZER.decode(out_perturb.tolist()[0])
 
             print(text_whole)
             print("=" * 80)
 
-            collect_gen[current_index] = [out, out_perturb, out1]
+            collect_gen[current_index] = [tokenized_cond_text, out_perturb, out1]
 
             current_index = current_index + 1
 

From 7ffe47c88861ccea7e4b28538ea3dcb504c497a2 Mon Sep 17 00:00:00 2001
From: piero <piero@uber.com>
Date: Wed, 27 Nov 2019 16:39:49 -0800
Subject: [PATCH 183/505] Improved device specification

---
 examples/run_pplm_discrim_train.py | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)

diff --git a/examples/run_pplm_discrim_train.py b/examples/run_pplm_discrim_train.py
index 9438cbbac1..519e2de29a 100644
--- a/examples/run_pplm_discrim_train.py
+++ b/examples/run_pplm_discrim_train.py
@@ -242,10 +242,9 @@ def get_cached_data_loader(dataset, batch_size, discriminator, shuffle=False):
 def train_discriminator(
         dataset, dataset_fp=None, pretrained_model='gpt2-medium',
         epochs=10, batch_size=64, log_interval=10,
-        save_model=False, cached=False, use_cuda=False):
-    if use_cuda:
-        global device
-        device = 'cuda'
+        save_model=False, cached=False, no_cuda=False):
+    global device
+    device = "cuda" if torch.cuda.is_available() and not no_cuda else "cpu"
 
     print('Preprocessing {} dataset...'.format(dataset))
     start = time.time()
@@ -577,8 +576,8 @@ if __name__ == '__main__':
                         help='whether to save the model')
     parser.add_argument('--cached', action='store_true',
                         help='whether to cache the input representations')
-    parser.add_argument('--use_cuda', action='store_true',
-                        help='use to turn on cuda')
+    parser.add_argument('--no_cuda', action='store_true',
+                        help='use to turn off cuda')
     args = parser.parse_args()
 
     train_discriminator(**(vars(args)))

From 6c9c1317800499bd5aaf3f1a7492a72961bbfa91 Mon Sep 17 00:00:00 2001
From: piero <piero@uber.com>
Date: Wed, 27 Nov 2019 17:27:39 -0800
Subject: [PATCH 184/505] More cleanup for run_model. Identical output as
 before.

---
 examples/run_pplm.py | 308 ++++++++++++++++++++++++-------------------
 1 file changed, 171 insertions(+), 137 deletions(-)

diff --git a/examples/run_pplm.py b/examples/run_pplm.py
index 0d9ed86f45..27ead3c3c5 100644
--- a/examples/run_pplm.py
+++ b/examples/run_pplm.py
@@ -39,7 +39,6 @@ from transformers import GPT2Tokenizer
 from transformers.file_utils import cached_path
 from transformers.modeling_gpt2 import GPT2LMHeadModel
 
-
 PPLM_BOW = 1
 PPLM_DISCRIM = 2
 PPLM_BOW_DISCRIM = 3
@@ -129,8 +128,7 @@ def perturb_past(
         decay=False,
         gamma=1.5,
 ):
-
-    #def perturb_past(past, model, prev, classifier, good_index=None,
+    # def perturb_past(past, model, prev, classifier, good_index=None,
     #             stepsize=0.01, vocab_size=50257,
     #             original_probs=None, accumulated_hidden=None, true_past=None,
     #             grad_norms=None):
@@ -237,7 +235,7 @@ def perturb_past(
                     future_hidden, dim=1)
 
             predicted_sentiment = classifier(new_accumulated_hidden / (
-                        current_length + 1 + horizon_length))
+                    current_length + 1 + horizon_length))
 
             label = torch.tensor([label_class], device='cuda',
                                  dtype=torch.long)
@@ -349,6 +347,13 @@ def get_bag_of_words_indices(bag_of_words_ids_or_paths: List[str]) -> List[
         bow_indices.append(
             [TOKENIZER.encode(word.strip(), add_prefix_space=True) for word in
              words])
+
+    #bow_words = set()
+    #for bow_list in bow_indices:
+    #    bow_list = list(filter(lambda x: len(x) <= 1, bow_list))
+    #    bow_words.update(
+    #        (TOKENIZER.decode(word).strip(), word) for word in bow_list)
+
     return bow_indices
 
 
@@ -368,28 +373,28 @@ def build_bows_one_hot_vectors(bow_indices):
 
 
 def full_text_generation(
-            model,
-            context=None,
-            num_samples=1,
-            device="cuda",
-            sample=True,
-            discrim=None,
-            label_class=None,
-            bag_of_words=None,
-            length=100,
-            grad_length=10000,
-            stepsize=0.02,
-            num_iterations=3,
-            temperature=1.0,
-            gm_scale=0.9,
-            kl_scale=0.01,
-            top_k=10,
-            window_length=0,
-            horizon_length=1,
-            decay=False,
-            gamma=1.5,
-            **kwargs
-    ):
+        model,
+        context=None,
+        num_samples=1,
+        device="cuda",
+        sample=True,
+        discrim=None,
+        label_class=None,
+        bag_of_words=None,
+        length=100,
+        grad_length=10000,
+        stepsize=0.02,
+        num_iterations=3,
+        temperature=1.0,
+        gm_scale=0.9,
+        kl_scale=0.01,
+        top_k=10,
+        window_length=0,
+        horizon_length=1,
+        decay=False,
+        gamma=1.5,
+        **kwargs
+):
     classifier, class_id = get_classifier(
         discrim,
         label_class,
@@ -465,15 +470,9 @@ def full_text_generation(
     #         actual_words = [(TOKENIZER.decode(ww).strip(),ww) for ww in good_list]
 
     bow_indices = []
-    actual_words = None
     if bag_of_words:
         bow_indices = get_bag_of_words_indices(bag_of_words.split(";"))
 
-        for good_list in bow_indices:
-            good_list = list(filter(lambda x: len(x) <= 1, good_list))
-            actual_words = [(TOKENIZER.decode(ww).strip(), ww) for ww in
-                            good_list]
-
     if bag_of_words and classifier:
         print("Both PPLM-BoW and PPLM-Discrim are on. This is not optimized.")
         loss_type = PPLM_BOW_DISCRIM
@@ -533,8 +532,7 @@ def full_text_generation(
 
     torch.cuda.empty_cache()
 
-    return original, perturbed_list, discrim_loss_list, loss_in_time_list, actual_words
-
+    return original, perturbed_list, discrim_loss_list, loss_in_time_list
 
 
 def generate_text_pplm(
@@ -611,25 +609,25 @@ def generate_text_pplm(
             accumulated_hidden = torch.sum(accumulated_hidden, dim=1)
 
             perturbed_past, _, grad_norms, loss_per_iter = perturb_past(
-                    past,
-                    model,
-                    prev,
-                    unpert_past=unpert_past,
-                    unpert_logits=unpert_logits,
-                    accumulated_hidden=accumulated_hidden,
-                    grad_norms=grad_norms,
-                    stepsize=current_stepsize,
-                    classifier=classifier,
-                    label_class=label_class,
-                    one_hot_bows_vectors=one_hot_bows_vectors,
-                    loss_type=loss_type,
-                    num_iterations=num_iterations,
-                    kl_scale=kl_scale,
-                    window_length=window_length,
-                    horizon_length=horizon_length,
-                    decay=decay,
-                    gamma=gamma,
-                )
+                past,
+                model,
+                prev,
+                unpert_past=unpert_past,
+                unpert_logits=unpert_logits,
+                accumulated_hidden=accumulated_hidden,
+                grad_norms=grad_norms,
+                stepsize=current_stepsize,
+                classifier=classifier,
+                label_class=label_class,
+                one_hot_bows_vectors=one_hot_bows_vectors,
+                loss_type=loss_type,
+                num_iterations=num_iterations,
+                kl_scale=kl_scale,
+                window_length=window_length,
+                horizon_length=horizon_length,
+                decay=decay,
+                gamma=gamma,
+            )
             loss_in_time.append(loss_per_iter)
 
         # Piero modified model call
@@ -666,7 +664,7 @@ def generate_text_pplm(
             # print(TOKENIZER.decode(likelywords[1].tolist()[0]))
 
             log_probs = ((log_probs ** gm_scale) * (
-                        unpert_logits ** (1 - gm_scale)))  # + SmallConst
+                    unpert_logits ** (1 - gm_scale)))  # + SmallConst
 
             log_probs = top_k_filter(log_probs, k=top_k,
                                      probs=True)  # + SmallConst
@@ -696,53 +694,88 @@ def generate_text_pplm(
 
 def run_model():
     parser = argparse.ArgumentParser()
-    parser.add_argument('--model_path', '-M', type=str, default='gpt2-medium',
-                        help='pretrained model name or path to local checkpoint')
-    parser.add_argument('--bag-of-words', '-B', type=str, default=None,
-                        help='Bags of words used for PPLM-BoW. Multiple BoWs separated by ;')
-    parser.add_argument('--discrim', '-D', type=str, default=None,
-                        choices=(
-                        'clickbait', 'sentiment', 'toxicity', 'generic'),
-                        help='Discriminator to use for loss-type 2')
-    parser.add_argument('--discrim_weights', type=str, default=None,
-                        help='Weights for the generic discriminator')
-    parser.add_argument('--discrim_meta', type=str, default=None,
-                        help='Meta information for the generic discriminator')
-    parser.add_argument('--label_class', type=int, default=-1,
-                        help='Class label used for the discriminator')
-    parser.add_argument('--stepsize', type=float, default=0.02)
+    parser.add_argument(
+        "--model_path",
+        "-M",
+        type=str,
+        default="gpt2-medium",
+        help="pretrained model name or path to local checkpoint",
+    )
+    parser.add_argument(
+        "--bag_of_words",
+        "-B",
+        type=str,
+        default=None,
+        help="Bags of words used for PPLM-BoW. "
+             "Either a BOW id (see list in code) or a filepath. "
+             "Multiple BoWs separated by ;",
+    )
+    parser.add_argument(
+        "--discrim",
+        "-D",
+        type=str,
+        default=None,
+        choices=("clickbait", "sentiment", "toxicity"),
+        help="Discriminator to use for loss-type 2",
+    )
+    parser.add_argument(
+        "--label_class",
+        type=int,
+        default=-1,
+        help="Class label used for the discriminator",
+    )
+    parser.add_argument("--stepsize", type=float, default=0.02)
     parser.add_argument("--length", type=int, default=100)
     parser.add_argument("--seed", type=int, default=0)
     parser.add_argument("--temperature", type=float, default=1.0)
     parser.add_argument("--top_k", type=int, default=10)
     parser.add_argument("--gm_scale", type=float, default=0.9)
     parser.add_argument("--kl_scale", type=float, default=0.01)
-    parser.add_argument('--nocuda', action='store_true', help='no cuda')
-    parser.add_argument('--uncond', action='store_true',
-                        help='Generate from end-of-text as prefix')
-    parser.add_argument("--cond_text", type=str, default='The lake',
-                        help='Prefix texts to condition on')
-    parser.add_argument('--num_iterations', type=int, default=3)
-    parser.add_argument('--grad_length', type=int, default=10000)
-    parser.add_argument('--num_samples', type=int, default=1,
-                        help='Number of samples to generate from the modified latents')
-    parser.add_argument('--horizon_length', type=int, default=1,
-                        help='Length of future to optimize over')
-    # parser.add_argument('--force-token', action='store_true', help='no cuda')
-    parser.add_argument('--window_length', type=int, default=0,
-                        help='Length of past which is being optimizer; 0 corresponds to infinite window length')
-    parser.add_argument('--decay', action='store_true',
-                        help='whether to decay or not')
-    parser.add_argument('--gamma', type=float, default=1.5)
-    parser.add_argument('--colorama', action='store_true', help='no cuda')
+    parser.add_argument("--no_cuda", action="store_true", help="no cuda")
+    parser.add_argument(
+        "--uncond", action="store_true",
+        help="Generate from end-of-text as prefix"
+    )
+    parser.add_argument(
+        "--cond_text", type=str, default="The lake",
+        help="Prefix texts to condition on"
+    )
+    parser.add_argument("--num_iterations", type=int, default=3)
+    parser.add_argument("--grad_length", type=int, default=10000)
+    parser.add_argument(
+        "--num_samples",
+        type=int,
+        default=1,
+        help="Number of samples to generate from the modified latents",
+    )
+    parser.add_argument(
+        "--horizon_length",
+        type=int,
+        default=1,
+        help="Length of future to optimize over",
+    )
+    parser.add_argument(
+        "--window_length",
+        type=int,
+        default=0,
+        help="Length of past which is being optimized; "
+             "0 corresponds to infinite window length",
+    )
+    parser.add_argument("--decay", action="store_true",
+                        help="whether to decay or not")
+    parser.add_argument("--gamma", type=float, default=1.5)
+    parser.add_argument("--colorama", action="store_true", help="colors keywords")
 
     args = parser.parse_args()
 
+    # set Random seed
     torch.manual_seed(args.seed)
     np.random.seed(args.seed)
 
-    device = 'cpu' if args.nocuda else 'cuda'
+    # set the device
+    device = "cuda" if torch.cuda.is_available() and not args.no_cuda else "cpu"
 
+    # load pretrained model
     model = GPT2LMHeadModel.from_pretrained(
         args.model_path,
         output_hidden_states=True
@@ -753,76 +786,77 @@ def run_model():
     # Freeze GPT-2 weights
     for param in model.parameters():
         param.requires_grad = False
-    pass
 
+    # figure out conditioning text
     if args.uncond:
-        seq = [[50256, 50256]]
-
+        tokenized_cond_text = TOKENIZER.encode(
+            [TOKENIZER.bos_token]
+        )
     else:
         raw_text = args.cond_text
         while not raw_text:
-            print('Did you forget to add `--cond-text`? ')
+            print("Did you forget to add `--cond_text`? ")
             raw_text = input("Model prompt >>> ")
-        seq = [[50256] + TOKENIZER.encode(raw_text)]
+        tokenized_cond_text = TOKENIZER.encode(TOKENIZER.bos_token + raw_text)
 
-    collect_gen = dict()
-    current_index = 0
-    for tokenized_cond_text in seq:
+    print("= Prefix of sentence =")
+    print(TOKENIZER.decode(tokenized_cond_text))
+    print()
 
-        text = TOKENIZER.decode(tokenized_cond_text)
-        print("=" * 40 + " Prefix of sentence " + "=" * 40)
-        print(text)
-        print("=" * 80)
+    # generate unperturbed and perturbed texts
 
-        out1, out_perturb, discrim_loss_list, loss_in_time_list, actual_words = full_text_generation(
-            model=model, context=tokenized_cond_text, device=device, **vars(args)
-        )
+    # full_text_generation returns:
+    # unpert_gen_tok_text, pert_gen_tok_texts, discrim_losses, losses_in_time
+    unpert_gen_tok_text, pert_gen_tok_texts, _, _ = full_text_generation(
+        model=model, context=tokenized_cond_text, device=device, **vars(args)
+    )
 
-        text_whole = TOKENIZER.decode(out1.tolist()[0])
+    # untokenize unperturbed text
+    unpert_gen_text = TOKENIZER.decode(unpert_gen_tok_text.tolist()[0])
 
-        print("=" * 80)
-        print("=" * 40 + " Whole sentence (Original)" + "=" * 40)
-        print(text_whole)
-        print("=" * 80)
+    print("=" * 80)
+    print("= Unperturbed generated text =")
+    print(unpert_gen_text)
+    print()
 
-        out_perturb_copy = out_perturb
+    generated_texts = []
 
-        for out_perturb in out_perturb_copy:
-            # try:
-            #    print("=" * 40 + " Whole sentence (Perturbed)" + "=" * 40)
-            #    text_whole = TOKENIZER.decode(out_perturb.tolist()[0])
-            #    print(text_whole)
-            #    print("=" * 80)
-            # except:
-            #    pass
-            # collect_gen[current_index] = [out, out_perturb, out1]
-            ## Save the prefix, perturbed seq, original seq for each index
-            print("=" * 40 + " Whole sentence (Perturbed)" + "=" * 40)
-            keyword_tokens = [aa[-1][0] for aa in
-                              actual_words] if actual_words else []
-            output_tokens = out_perturb.tolist()[0]
+    bow_words = set()
+    bow_indices = get_bag_of_words_indices(args.bag_of_words.split(";"))
+    for bow_list in bow_indices:
+        filtered = list(filter(lambda x: len(x) <= 1, bow_list))
+        bow_words.update(w[0] for w in filtered)
 
+    # iterate through the perturbed texts
+    for i, pert_gen_tok_text in enumerate(pert_gen_tok_texts):
+        try:
+            # untokenize unperturbed text
             if args.colorama:
                 import colorama
 
-                text_whole = ''
-                for tokenized_cond_text in output_tokens:
-                    if tokenized_cond_text in keyword_tokens:
-                        text_whole += '%s%s%s' % (
-                        colorama.Fore.GREEN, TOKENIZER.decode([tokenized_cond_text]),
-                        colorama.Style.RESET_ALL)
+                pert_gen_text = ''
+                for word_id in pert_gen_tok_text.tolist()[0]:
+                    if word_id in bow_words:
+                        pert_gen_text += '{}{}{}'.format(
+                            colorama.Fore.RED,
+                            TOKENIZER.decode([word_id]),
+                            colorama.Style.RESET_ALL
+                        )
                     else:
-                        text_whole += TOKENIZER.decode([tokenized_cond_text])
+                        pert_gen_text += TOKENIZER.decode([word_id])
             else:
-                text_whole = TOKENIZER.decode(out_perturb.tolist()[0])
+                pert_gen_text = TOKENIZER.decode(pert_gen_tok_text.tolist()[0])
 
-            print(text_whole)
-            print("=" * 80)
-
-            collect_gen[current_index] = [tokenized_cond_text, out_perturb, out1]
-
-            current_index = current_index + 1
+            print("= Perturbed generated text {} =".format(i + 1))
+            print(pert_gen_text)
+            print()
+        except:
+            pass
 
+        # keep the prefix, perturbed seq, original seq for each index
+        generated_texts.append(
+            (tokenized_cond_text, pert_gen_tok_text, unpert_gen_tok_text)
+        )
 
     return
 

From 08c6e456a391cd463b0a52d125166aa766a2c7e4 Mon Sep 17 00:00:00 2001
From: piero <piero@uber.com>
Date: Wed, 27 Nov 2019 17:48:46 -0800
Subject: [PATCH 185/505] Cleaned full_text_generation. Identical output as
 before.

---
 examples/run_pplm.py | 103 ++++++++-----------------------------------
 1 file changed, 19 insertions(+), 84 deletions(-)

diff --git a/examples/run_pplm.py b/examples/run_pplm.py
index 27ead3c3c5..b85998d706 100644
--- a/examples/run_pplm.py
+++ b/examples/run_pplm.py
@@ -401,74 +401,6 @@ def full_text_generation(
         device
     )
 
-    # if args.discrim == 'clickbait':
-    #     classifier = ClassificationHead(class_size=2, embed_size=1024).to(device)
-    #     classifier.load_state_dict(torch.load("discrim_models/clickbait_classifierhead.pt"))
-    #     classifier.eval()
-    #     args.label_class = 1 # clickbaity
-    #
-    # elif args.discrim == 'sentiment':
-    #     classifier = ClassificationHead(class_size=5, embed_size=1024).to(device)
-    #     #classifier.load_state_dict(torch.load("discrim_models/sentiment_classifierhead.pt"))
-    #     classifier.load_state_dict(torch.load("discrim_models/SST_classifier_head_epoch_16.pt"))
-    #     classifier.eval()
-    #     if args.label_class < 0:
-    #         raise Exception('Wrong class for sentiment, use --label-class 2 for *very positive*, 3 for *very negative*')
-    #     #args.label_class = 2 # very pos
-    #     #args.label_class = 3 # very neg
-    #
-    # elif args.discrim == 'toxicity':
-    #     classifier = ClassificationHead(class_size=2, embed_size=1024).to(device)
-    #     classifier.load_state_dict(torch.load("discrim_models/toxicity_classifierhead.pt"))
-    #     classifier.eval()
-    #     args.label_class = 0 # not toxic
-    #
-    # elif args.discrim == 'generic':
-    #     if args.discrim_weights is None:
-    #         raise ValueError('When using a generic discriminator, '
-    #                          'discrim_weights need to be specified')
-    #     if args.discrim_meta is None:
-    #         raise ValueError('When using a generic discriminator, '
-    #                          'discrim_meta need to be specified')
-    #
-    #     with open(args.discrim_meta, 'r') as discrim_meta_file:
-    #         meta = json.load(discrim_meta_file)
-    #
-    #     classifier = ClassificationHead(
-    #         class_size=meta['class_size'],
-    #         embed_size=meta['embed_size'],
-    #         # todo add tokenizer from meta
-    #     ).to(device)
-    #     classifier.load_state_dict(torch.load(args.discrim_weights))
-    #     classifier.eval()
-    #     if args.label_class == -1:
-    #         args.label_class = meta['default_class']
-    #
-    # else:
-    #     classifier = None
-
-    # Get tokens for the list of positive words
-    def list_tokens(word_list):
-        token_list = [TOKENIZER.encode(word, add_prefix_space=True) for word in
-                      word_list]
-        # token_list = []
-        # for word in word_list:
-        #    token_list.append(TOKENIZER.encode(" " + word))
-        return token_list
-
-    # good_index = []
-    # if args.bag_of_words:
-    #     bags_of_words = args.bag_of_words.split(";")
-    #     for wordlist in bags_of_words:
-    #         with open(wordlist, "r") as f:
-    #             words = f.read().strip()
-    #             words = words.split('\n')
-    #         good_index.append(list_tokens(words))
-    #
-    #     for good_list in good_index:
-    #         good_list = list(filter(lambda x: len(x) <= 1, good_list))
-    #         actual_words = [(TOKENIZER.decode(ww).strip(),ww) for ww in good_list]
-
     bow_indices = []
     if bag_of_words:
         bow_indices = get_bag_of_words_indices(bag_of_words.split(";"))
@@ -486,9 +418,9 @@ def full_text_generation(
         print("Using PPLM-Discrim")
 
     else:
-        raise Exception("Specify either --bag_of_words (-B) or --discrim (-D)")
+        raise Exception("Specify either a bag of words or a discriminator")
 
-    original, _, _ = generate_text_pplm(
+    unpert_gen_tok_text, _, _ = generate_text_pplm(
         model=model,
         context=context,
         device=device,
@@ -497,12 +429,12 @@ def full_text_generation(
     )
     torch.cuda.empty_cache()
 
-    perturbed_list = []
-    discrim_loss_list = []
-    loss_in_time_list = []
+    pert_gen_tok_texts = []
+    discrim_losses = []
+    losses_in_time = []
 
     for i in range(num_samples):
-        perturbed, discrim_loss, loss_in_time = generate_text_pplm(
+        pert_gen_tok_text, discrim_loss, loss_in_time = generate_text_pplm(
             model=model,
             context=context,
             device=device,
@@ -525,14 +457,14 @@ def full_text_generation(
             decay=decay,
             gamma=gamma,
         )
-        perturbed_list.append(perturbed)
+        pert_gen_tok_texts.append(pert_gen_tok_text)
         if classifier is not None:
-            discrim_loss_list.append(discrim_loss.data.cpu().numpy())
-        loss_in_time_list.append(loss_in_time)
+            discrim_losses.append(discrim_loss.data.cpu().numpy())
+        losses_in_time.append(loss_in_time)
 
     torch.cuda.empty_cache()
 
-    return original, perturbed_list, discrim_loss_list, loss_in_time_list
+    return unpert_gen_tok_text, pert_gen_tok_texts, discrim_losses, losses_in_time
 
 
 def generate_text_pplm(
@@ -821,11 +753,14 @@ def run_model():
 
     generated_texts = []
 
-    bow_words = set()
-    bow_indices = get_bag_of_words_indices(args.bag_of_words.split(";"))
-    for bow_list in bow_indices:
-        filtered = list(filter(lambda x: len(x) <= 1, bow_list))
-        bow_words.update(w[0] for w in filtered)
+    bow_word_ids = set()
+    if args.bag_of_words and args.colorama:
+        bow_indices = get_bag_of_words_indices(args.bag_of_words.split(";"))
+        for single_bow_list in bow_indices:
+            # filtering all words in the list composed of more than 1 token
+            filtered = list(filter(lambda x: len(x) <= 1, single_bow_list))
+            # w[0] because we are sure w has only 1 item because previous fitler
+            bow_word_ids.update(w[0] for w in filtered)
 
     # iterate through the perturbed texts
     for i, pert_gen_tok_text in enumerate(pert_gen_tok_texts):
@@ -836,7 +771,7 @@ def run_model():
 
                 pert_gen_text = ''
                 for word_id in pert_gen_tok_text.tolist()[0]:
-                    if word_id in bow_words:
+                    if word_id in bow_word_ids:
                         pert_gen_text += '{}{}{}'.format(
                             colorama.Fore.RED,
                             TOKENIZER.decode([word_id]),

From 7ea12db3f539af414579c0e27b87e399645c58ed Mon Sep 17 00:00:00 2001
From: piero <piero@uber.com>
Date: Wed, 27 Nov 2019 17:49:39 -0800
Subject: [PATCH 186/505] Removed commented code. Identical output as before.

---
 examples/run_pplm.py | 7 -------
 1 file changed, 7 deletions(-)

diff --git a/examples/run_pplm.py b/examples/run_pplm.py
index b85998d706..f4d697de3b 100644
--- a/examples/run_pplm.py
+++ b/examples/run_pplm.py
@@ -347,13 +347,6 @@ def get_bag_of_words_indices(bag_of_words_ids_or_paths: List[str]) -> List[
         bow_indices.append(
             [TOKENIZER.encode(word.strip(), add_prefix_space=True) for word in
              words])
-
-    #bow_words = set()
-    #for bow_list in bow_indices:
-    #    bow_list = list(filter(lambda x: len(x) <= 1, bow_list))
-    #    bow_words.update(
-    #        (TOKENIZER.decode(word).strip(), word) for word in bow_list)
-
     return bow_indices
 
 

From ef47b2c03ad3d901f4b7454b004eb44cb01cc3e3 Mon Sep 17 00:00:00 2001
From: piero <piero@uber.com>
Date: Wed, 27 Nov 2019 17:50:21 -0800
Subject: [PATCH 187/505] Removed commented code. Identical output as before.

---
 examples/run_pplm.py | 14 --------------
 1 file changed, 14 deletions(-)

diff --git a/examples/run_pplm.py b/examples/run_pplm.py
index f4d697de3b..b18e97b5a8 100644
--- a/examples/run_pplm.py
+++ b/examples/run_pplm.py
@@ -128,20 +128,6 @@ def perturb_past(
         decay=False,
         gamma=1.5,
 ):
-    # def perturb_past(past, model, prev, classifier, good_index=None,
-    #             stepsize=0.01, vocab_size=50257,
-    #             original_probs=None, accumulated_hidden=None, true_past=None,
-    #             grad_norms=None):
-
-    # one_hot_bows_vectors = []
-    # for good_list in good_index:
-    #     good_list = list(filter(lambda x: len(x) <= 1, good_list))
-    #     good_list = torch.tensor(good_list).cuda()
-    #     num_good = good_list.shape[0]
-    #     one_hot_good = torch.zeros(num_good, vocab_size).cuda()
-    #     one_hot_good.scatter_(1, good_list, 1)
-    #     one_hot_bows_vectors.append(one_hot_good)
-
     # Generate inital perturbed past
     past_perturb_orig = [
         (np.random.uniform(0.0, 0.0, p.shape).astype('float32'))

From 61a12f790de9795af1c3dff5fad7e2c4f6808d05 Mon Sep 17 00:00:00 2001
From: piero <piero@uber.com>
Date: Wed, 27 Nov 2019 17:54:49 -0800
Subject: [PATCH 188/505] Renamed SmallConst to SMALL_CONST and introduced
 BIG_CONST. Identical output as before.

---
 examples/run_pplm.py | 25 +++++++++++++------------
 1 file changed, 13 insertions(+), 12 deletions(-)

diff --git a/examples/run_pplm.py b/examples/run_pplm.py
index b18e97b5a8..4d335a9241 100644
--- a/examples/run_pplm.py
+++ b/examples/run_pplm.py
@@ -43,7 +43,7 @@ PPLM_BOW = 1
 PPLM_DISCRIM = 2
 PPLM_BOW_DISCRIM = 3
 SMALL_CONST = 1e-15
-SmallConst = 1e-15
+BIG_CONST = 1e10
 TOKENIZER = GPT2Tokenizer.from_pretrained("gpt2-medium")
 
 BAG_OF_WORDS_ARCHIVE_MAP = {
@@ -104,7 +104,8 @@ def top_k_filter(logits, k, probs=False):
         if probs:
             return torch.where(logits < batch_mins,
                                torch.ones_like(logits) * 0.0, logits)
-        return torch.where(logits < batch_mins, torch.ones_like(logits) * -1e10,
+        return torch.where(logits < batch_mins,
+                           torch.ones_like(logits) * -BIG_CONST,
                            logits)
 
 
@@ -137,7 +138,7 @@ def perturb_past(
         accumulated_hidden = 0
 
     if decay:
-        decay_mask = torch.arange(0., 1.0 + SmallConst, 1.0 / (window_length))[
+        decay_mask = torch.arange(0., 1.0 + SMALL_CONST, 1.0 / (window_length))[
                      1:]
     else:
         decay_mask = 1.0
@@ -233,9 +234,9 @@ def perturb_past(
         kl_loss = 0.0
         if kl_scale > 0.0:
             p = (F.softmax(unpert_logits[:, -1, :], dim=-1))
-            p = p + SmallConst * (p <= SmallConst).type(
+            p = p + SMALL_CONST * (p <= SMALL_CONST).type(
                 torch.FloatTensor).cuda().detach()
-            correction = SmallConst * (probabs <= SmallConst).type(
+            correction = SMALL_CONST * (probabs <= SMALL_CONST).type(
                 torch.FloatTensor).cuda().detach()
             corrected_probabs = probabs + correction.detach()
             kl_loss = kl_scale * (
@@ -254,7 +255,7 @@ def perturb_past(
                 for index, p_ in
                 enumerate(past_perturb)]
         else:
-            grad_norms = [(torch.norm(p_.grad * window_mask) + SmallConst) for
+            grad_norms = [(torch.norm(p_.grad * window_mask) + SMALL_CONST) for
                           index, p_ in enumerate(past_perturb)]
 
         grad = [
@@ -560,31 +561,31 @@ def generate_text_pplm(
         # Piero modified model call
         # hidden = model.hidden_states  # update hidden
         # logits = model.forward_hidden(hidden)
-        logits = logits[:, -1, :] / temperature  # + SmallConst
+        logits = logits[:, -1, :] / temperature  # + SMALL_CONST
 
-        # logits = top_k_filter(logits, k=args.top_k)  # + SmallConst
+        # logits = top_k_filter(logits, k=args.top_k)  # + SMALL_CONST
 
         log_probs = F.softmax(logits, dim=-1)
 
         # Fuse the modified model and original model
         if perturb:
 
-            # original_probs = top_k_filter(original_probs[:, -1, :]) #+ SmallConst
+            # original_probs = top_k_filter(original_probs[:, -1, :]) #+ SMALL_CONST
             unpert_logits = F.softmax(unpert_logits[:, -1, :], dim=-1)
             # likelywords = torch.topk(original_probs, k=10, dim=-1)
             # print(TOKENIZER.decode(likelywords[1].tolist()[0]))
 
             log_probs = ((log_probs ** gm_scale) * (
-                    unpert_logits ** (1 - gm_scale)))  # + SmallConst
+                    unpert_logits ** (1 - gm_scale)))  # + SMALL_CONST
 
             log_probs = top_k_filter(log_probs, k=top_k,
-                                     probs=True)  # + SmallConst
+                                     probs=True)  # + SMALL_CONST
 
             if torch.sum(log_probs) <= 1:
                 log_probs = log_probs / torch.sum(log_probs)
 
         else:
-            logits = top_k_filter(logits, k=top_k)  # + SmallConst
+            logits = top_k_filter(logits, k=top_k)  # + SMALL_CONST
             log_probs = F.softmax(logits, dim=-1)
 
         if sample:

From 9f693a0c4831d09ba2c177128452984091fda619 Mon Sep 17 00:00:00 2001
From: piero <piero@uber.com>
Date: Wed, 27 Nov 2019 18:16:30 -0800
Subject: [PATCH 189/505] Cleaned generate_text_pplm. Identical output as
 before.

---
 examples/run_pplm.py | 125 ++++++++++++++++++-------------------------
 1 file changed, 53 insertions(+), 72 deletions(-)

diff --git a/examples/run_pplm.py b/examples/run_pplm.py
index 4d335a9241..bd03bbe5e0 100644
--- a/examples/run_pplm.py
+++ b/examples/run_pplm.py
@@ -471,59 +471,49 @@ def generate_text_pplm(
         decay=False,
         gamma=1.5,
 ):
-    output = torch.tensor(context, device=device, dtype=torch.long).unsqueeze(
-        0) if context else None
+    output_so_far = (
+        torch.tensor(context, device=device, dtype=torch.long).unsqueeze(0)
+        if context
+        else None
+    )
 
     # collect one hot vectors for bags of words
     one_hot_bows_vectors = build_bows_one_hot_vectors(bow_indices)
 
     grad_norms = None
+    unpert_discrim_loss = 0
     loss_in_time = []
     for i in trange(length, ascii=True):
 
         # Get past/probs for current output, except for last word
-        # Note that GPT takes 2 inputs: past + current-token
-        # Therefore, use everything from before current i/p token to generate relevant past
+        # Note that GPT takes 2 inputs: past + current_token
 
-        if past is None and output is not None:
-            prev = output[:, -1:]
-            # _, past = model(output[:, :-1])
-            # original_probs, true_past = model(output)
-            # true_hidden = model.hidden_states
+        # run model forward to obtain unperturbed
+        if past is None and output_so_far is not None:
+            last = output_so_far[:, -1:]
+            _, past, _ = model(output_so_far[:, :-1])
 
-            # Piero modified model call
-            _, past, _ = model(output[:, :-1])
-            unpert_logits, unpert_past, unpert_all_hidden = model(output)
-            true_hidden = unpert_all_hidden[-1]
-
-        else:
-            # original_probs, true_past = model(output)
-            # true_hidden = model.hidden_states
-
-            # Piero modified model call
-            unpert_logits, unpert_past, unpert_all_hidden = model(output)
-            true_hidden = unpert_all_hidden[-1]
-
-        # Modify the past if necessary
+        unpert_logits, unpert_past, unpert_all_hidden = model(output_so_far)
+        unpert_last_hidden = unpert_all_hidden[-1]
 
+        # check if we are abowe grad max length
         if i >= grad_length:
             current_stepsize = stepsize * 0
         else:
             current_stepsize = stepsize
 
+        # modify the past if necessary
         if not perturb or num_iterations == 0:
-            perturbed_past = past
+            pert_past = past
 
         else:
-            # Piero modified model call
-            # accumulated_hidden = model.hidden_states[:, :-1, :]
-            accumulated_hidden = true_hidden[:, :-1, :]
+            accumulated_hidden = unpert_last_hidden[:, :-1, :]
             accumulated_hidden = torch.sum(accumulated_hidden, dim=1)
 
-            perturbed_past, _, grad_norms, loss_per_iter = perturb_past(
+            pert_past, _, grad_norms, loss_this_iter = perturb_past(
                 past,
                 model,
-                prev,
+                last,
                 unpert_past=unpert_past,
                 unpert_logits=unpert_logits,
                 accumulated_hidden=accumulated_hidden,
@@ -540,68 +530,59 @@ def generate_text_pplm(
                 decay=decay,
                 gamma=gamma,
             )
-            loss_in_time.append(loss_per_iter)
+            loss_in_time.append(loss_this_iter)
 
-        # Piero modified model call
-        logits, past, pert_all_hidden = model(prev, past=perturbed_past)
-        # test_logits = F.softmax(test_logits[:, -1, :], dim=-1)
-        # likelywords = torch.topk(test_logits, k=10, dim=-1)
-        # print(TOKENIZER.decode(likelywords[1].tolist()[0]))
+        pert_logits, past, pert_all_hidden = model(last, past=pert_past)
+        pert_logits = pert_logits[:, -1, :] / temperature  # + SMALL_CONST
+        pert_probs = F.softmax(pert_logits, dim=-1)
 
         if classifier is not None:
             ce_loss = torch.nn.CrossEntropyLoss()
-            predicted_sentiment = classifier(torch.mean(true_hidden, dim=1))
+            prediction = classifier(torch.mean(unpert_last_hidden, dim=1))
             label = torch.tensor([label_class], device='cuda',
                                  dtype=torch.long)
-            true_discrim_loss = ce_loss(predicted_sentiment, label)
-            print("true discrim loss", true_discrim_loss.data.cpu().numpy())
+            unpert_discrim_loss = ce_loss(prediction, label)
+            print(
+                "unperturbed discrim loss",
+                unpert_discrim_loss.data.cpu().numpy()
+            )
         else:
-            true_discrim_loss = 0
-
-        # Piero modified model call
-        # hidden = model.hidden_states  # update hidden
-        # logits = model.forward_hidden(hidden)
-        logits = logits[:, -1, :] / temperature  # + SMALL_CONST
-
-        # logits = top_k_filter(logits, k=args.top_k)  # + SMALL_CONST
-
-        log_probs = F.softmax(logits, dim=-1)
+            unpert_discrim_loss = 0
 
         # Fuse the modified model and original model
         if perturb:
 
-            # original_probs = top_k_filter(original_probs[:, -1, :]) #+ SMALL_CONST
-            unpert_logits = F.softmax(unpert_logits[:, -1, :], dim=-1)
-            # likelywords = torch.topk(original_probs, k=10, dim=-1)
-            # print(TOKENIZER.decode(likelywords[1].tolist()[0]))
+            unpert_probs = F.softmax(unpert_logits[:, -1, :], dim=-1)
 
-            log_probs = ((log_probs ** gm_scale) * (
-                    unpert_logits ** (1 - gm_scale)))  # + SMALL_CONST
-
-            log_probs = top_k_filter(log_probs, k=top_k,
+            pert_probs = ((pert_probs ** gm_scale) * (
+                    unpert_probs ** (1 - gm_scale)))  # + SMALL_CONST
+            pert_probs = top_k_filter(pert_probs, k=top_k,
                                      probs=True)  # + SMALL_CONST
 
-            if torch.sum(log_probs) <= 1:
-                log_probs = log_probs / torch.sum(log_probs)
+            # rescale
+            if torch.sum(pert_probs) <= 1:
+                pert_probs = pert_probs / torch.sum(pert_probs)
 
         else:
-            logits = top_k_filter(logits, k=top_k)  # + SMALL_CONST
-            log_probs = F.softmax(logits, dim=-1)
+            pert_logits = top_k_filter(pert_logits, k=top_k)  # + SMALL_CONST
+            pert_probs = F.softmax(pert_logits, dim=-1)
 
+        # sample or greedy
         if sample:
-            # likelywords = torch.topk(log_probs, k=args.top_k, dim=-1)
-            # print(TOKENIZER.decode(likelywords[1].tolist()[0]))
-            # print(likelywords[0].tolist())
-            prev = torch.multinomial(log_probs, num_samples=1)
-        else:
-            _, prev = torch.topk(log_probs, k=1, dim=-1)
-        # if perturb:
-        #     prev = future
-        output = prev if output is None else torch.cat((output, prev),
-                                                       dim=1)  # update output
-        print(TOKENIZER.decode(output.tolist()[0]))
+            last = torch.multinomial(pert_probs, num_samples=1)
 
-    return output, true_discrim_loss, loss_in_time
+        else:
+            _, last = torch.topk(pert_probs, k=1, dim=-1)
+
+        # update context/output_so_far appending the new token
+        output_so_far = (
+            last if output_so_far is None
+            else torch.cat((output_so_far, last), dim=1)
+        )
+
+        print(TOKENIZER.decode(output_so_far.tolist()[0]))
+
+    return output_so_far, unpert_discrim_loss, loss_in_time
 
 
 def run_model():

From ffc29354051ccd4fa3fd12010abacb0ff2e0733e Mon Sep 17 00:00:00 2001
From: piero <piero@uber.com>
Date: Wed, 27 Nov 2019 18:30:42 -0800
Subject: [PATCH 190/505] Fix for making unditioned generation work. Identical
 output as before.

---
 examples/run_pplm.py | 49 ++++++++++++++++++++++++--------------------
 1 file changed, 27 insertions(+), 22 deletions(-)

diff --git a/examples/run_pplm.py b/examples/run_pplm.py
index bd03bbe5e0..e337add46d 100644
--- a/examples/run_pplm.py
+++ b/examples/run_pplm.py
@@ -481,6 +481,7 @@ def generate_text_pplm(
     one_hot_bows_vectors = build_bows_one_hot_vectors(bow_indices)
 
     grad_norms = None
+    last = None
     unpert_discrim_loss = 0
     loss_in_time = []
     for i in trange(length, ascii=True):
@@ -491,7 +492,8 @@ def generate_text_pplm(
         # run model forward to obtain unperturbed
         if past is None and output_so_far is not None:
             last = output_so_far[:, -1:]
-            _, past, _ = model(output_so_far[:, :-1])
+            if output_so_far.shape[1] > 1:
+                _, past, _ = model(output_so_far[:, :-1])
 
         unpert_logits, unpert_past, unpert_all_hidden = model(output_so_far)
         unpert_last_hidden = unpert_all_hidden[-1]
@@ -510,27 +512,30 @@ def generate_text_pplm(
             accumulated_hidden = unpert_last_hidden[:, :-1, :]
             accumulated_hidden = torch.sum(accumulated_hidden, dim=1)
 
-            pert_past, _, grad_norms, loss_this_iter = perturb_past(
-                past,
-                model,
-                last,
-                unpert_past=unpert_past,
-                unpert_logits=unpert_logits,
-                accumulated_hidden=accumulated_hidden,
-                grad_norms=grad_norms,
-                stepsize=current_stepsize,
-                classifier=classifier,
-                label_class=label_class,
-                one_hot_bows_vectors=one_hot_bows_vectors,
-                loss_type=loss_type,
-                num_iterations=num_iterations,
-                kl_scale=kl_scale,
-                window_length=window_length,
-                horizon_length=horizon_length,
-                decay=decay,
-                gamma=gamma,
-            )
-            loss_in_time.append(loss_this_iter)
+            if past is not None:
+                pert_past, _, grad_norms, loss_this_iter = perturb_past(
+                    past,
+                    model,
+                    last,
+                    unpert_past=unpert_past,
+                    unpert_logits=unpert_logits,
+                    accumulated_hidden=accumulated_hidden,
+                    grad_norms=grad_norms,
+                    stepsize=current_stepsize,
+                    classifier=classifier,
+                    label_class=label_class,
+                    one_hot_bows_vectors=one_hot_bows_vectors,
+                    loss_type=loss_type,
+                    num_iterations=num_iterations,
+                    kl_scale=kl_scale,
+                    window_length=window_length,
+                    horizon_length=horizon_length,
+                    decay=decay,
+                    gamma=gamma,
+                )
+                loss_in_time.append(loss_this_iter)
+            else:
+                pert_past = past
 
         pert_logits, past, pert_all_hidden = model(last, past=pert_past)
         pert_logits = pert_logits[:, -1, :] / temperature  # + SMALL_CONST

From 61399e5afe15915f5e074fd25faf85a5346e093a Mon Sep 17 00:00:00 2001
From: piero <piero@uber.com>
Date: Wed, 27 Nov 2019 19:51:42 -0800
Subject: [PATCH 191/505] Cleaned perturb_past. Identical output as before.

---
 examples/run_pplm.py | 202 ++++++++++++++++++++++++-------------------
 1 file changed, 111 insertions(+), 91 deletions(-)

diff --git a/examples/run_pplm.py b/examples/run_pplm.py
index e337add46d..77758759d9 100644
--- a/examples/run_pplm.py
+++ b/examples/run_pplm.py
@@ -112,7 +112,7 @@ def top_k_filter(logits, k, probs=False):
 def perturb_past(
         past,
         model,
-        prev,
+        last,
         unpert_past=None,
         unpert_logits=None,
         accumulated_hidden=None,
@@ -128,156 +128,174 @@ def perturb_past(
         horizon_length=1,
         decay=False,
         gamma=1.5,
+        device='cuda'
 ):
     # Generate inital perturbed past
-    past_perturb_orig = [
-        (np.random.uniform(0.0, 0.0, p.shape).astype('float32'))
-        for p in past]
+    grad_accumulator = [
+        (np.zeros(p.shape).astype("float32"))
+        for p in past
+    ]
 
     if accumulated_hidden is None:
         accumulated_hidden = 0
 
     if decay:
-        decay_mask = torch.arange(0., 1.0 + SMALL_CONST, 1.0 / (window_length))[
-                     1:]
+        decay_mask = torch.arange(
+            0.,
+            1.0 + SMALL_CONST,
+            1.0 / (window_length)
+        )[1:]
     else:
         decay_mask = 1.0
 
+    # TODO fix this comment (SUMANTH)
     # Generate a mask is gradient perturbated is based on a past window
-    _, _, _, current_length, _ = past[0].shape
+    _, _, _, curr_length, _ = past[0].shape
 
-    if current_length > window_length and window_length > 0:
-        ones_key_val_shape = tuple(past[0].shape[:-2]) + tuple(
-            [window_length]) + tuple(
-            past[0].shape[-1:])
+    if curr_length > window_length and window_length > 0:
+        ones_key_val_shape = (
+                tuple(past[0].shape[:-2])
+                + tuple([window_length])
+                + tuple(past[0].shape[-1:])
+        )
 
-        zeros_key_val_shape = tuple(past[0].shape[:-2]) + tuple(
-            [current_length - window_length]) + tuple(
-            past[0].shape[-1:])
+        zeros_key_val_shape = (
+                tuple(past[0].shape[:-2])
+                + tuple([curr_length - window_length])
+                + tuple(past[0].shape[-1:])
+        )
 
         ones_mask = torch.ones(ones_key_val_shape)
         ones_mask = decay_mask * ones_mask.permute(0, 1, 2, 4, 3)
         ones_mask = ones_mask.permute(0, 1, 2, 4, 3)
 
-        window_mask = torch.cat((ones_mask, torch.zeros(zeros_key_val_shape)),
-                                dim=-2).cuda()
+        window_mask = torch.cat(
+            (ones_mask, torch.zeros(zeros_key_val_shape)),
+            dim=-2
+        ).to(device)
     else:
-        window_mask = torch.ones_like(past[0]).cuda()
+        window_mask = torch.ones_like(past[0]).to(device)
 
+    # accumulate perturbations for num_iterations
     loss_per_iter = []
+    new_accumulated_hidden = None
     for i in range(num_iterations):
         print("Iteration ", i + 1)
-        past_perturb = [torch.from_numpy(p_) for p_ in past_perturb_orig]
-        past_perturb = [to_var(p_, requires_grad=True) for p_ in past_perturb]
+        curr_perturbation = [
+            to_var(torch.from_numpy(p_), requires_grad=True)
+            for p_ in grad_accumulator
+        ]
 
-        perturbed_past = list(map(add, past, past_perturb))
-
-        _, _, _, current_length, _ = past_perturb[0].shape
-
-        # _, future_past = model(prev, past=perturbed_past)
-        # hidden = model.hidden_states
-
-        # Piero modified model call
-        logits, _, all_hidden = model(prev, past=perturbed_past)
+        # Compute hidden using perturbed past
+        perturbed_past = list(map(add, past, curr_perturbation))
+        _, _, _, curr_length, _ = curr_perturbation[0].shape
+        all_logits, _, all_hidden = model(last, past=perturbed_past)
         hidden = all_hidden[-1]
-        new_accumulated_hidden = accumulated_hidden + torch.sum(hidden,
-                                                                dim=1).detach()
+        new_accumulated_hidden = accumulated_hidden + torch.sum(
+            hidden,
+            dim=1
+        ).detach()
+        # TODO: Check the layer-norm consistency of this with trained discriminator (Sumanth)
+        logits = all_logits[:, -1, :]
+        probs = F.softmax(logits, dim=-1)
 
-        # TODO: Check the layer-norm consistency of this with trained discriminator
-        logits = logits[:, -1, :]
-        probabs = F.softmax(logits, dim=-1)
         loss = 0.0
         loss_list = []
-        if loss_type == 1 or loss_type == 3:
-            for one_hot_good in one_hot_bows_vectors:
-                good_logits = torch.mm(probabs, torch.t(one_hot_good))
-                loss_word = good_logits
-                loss_word = torch.sum(loss_word)
-                loss_word = -torch.log(loss_word)
-                # loss_word = torch.sum(loss_word) /torch.sum(one_hot_good)
-                loss += loss_word
-                loss_list.append(loss_word)
+        if loss_type == PPLM_BOW or loss_type == PPLM_BOW_DISCRIM:
+            for one_hot_bow in one_hot_bows_vectors:
+                bow_logits = torch.mm(probs, torch.t(one_hot_bow))
+                bow_loss = -torch.log(torch.sum(bow_logits))
+                loss += bow_loss
+                loss_list.append(bow_loss)
             print(" pplm_bow_loss:", loss.data.cpu().numpy())
 
         if loss_type == 2 or loss_type == 3:
             ce_loss = torch.nn.CrossEntropyLoss()
-            new_true_past = unpert_past
-            for i in range(horizon_length):
-                future_probabs = F.softmax(logits, dim=-1)  # Get softmax
-                future_probabs = torch.unsqueeze(future_probabs, dim=1)
-
-                # _, new_true_past = model(future_probabs, past=new_true_past)
-                # future_hidden = model.hidden_states  # Get expected hidden states
-
-                # Piero modified model call
-                wte = model.resize_token_embeddings()
-                inputs_embeds = torch.matmul(future_probabs, wte.weight.data)
-                _, new_true_past, future_hidden = model(
-                    past=new_true_past,
+            # TODO why we need to do this assignment and not just using unpert_past? (Sumanth)
+            curr_unpert_past = unpert_past
+            curr_probs = torch.unsqueeze(probs, dim=1)
+            wte = model.resize_token_embeddings()
+            for _ in range(horizon_length):
+                inputs_embeds = torch.matmul(curr_probs, wte.weight.data)
+                _, curr_unpert_past, curr_all_hidden = model(
+                    past=curr_unpert_past,
                     inputs_embeds=inputs_embeds
                 )
-                future_hidden = future_hidden[-1]
-
+                curr_hidden = curr_all_hidden[-1]
                 new_accumulated_hidden = new_accumulated_hidden + torch.sum(
-                    future_hidden, dim=1)
+                    curr_hidden, dim=1)
 
-            predicted_sentiment = classifier(new_accumulated_hidden / (
-                    current_length + 1 + horizon_length))
+            prediction = classifier(new_accumulated_hidden /
+                                    (curr_length + 1 + horizon_length))
 
-            label = torch.tensor([label_class], device='cuda',
+            label = torch.tensor([label_class], device=device,
                                  dtype=torch.long)
-            discrim_loss = ce_loss(predicted_sentiment, label)
+            discrim_loss = ce_loss(prediction, label)
             print(" pplm_discrim_loss:", discrim_loss.data.cpu().numpy())
             loss += discrim_loss
             loss_list.append(discrim_loss)
 
         kl_loss = 0.0
         if kl_scale > 0.0:
-            p = (F.softmax(unpert_logits[:, -1, :], dim=-1))
-            p = p + SMALL_CONST * (p <= SMALL_CONST).type(
-                torch.FloatTensor).cuda().detach()
-            correction = SMALL_CONST * (probabs <= SMALL_CONST).type(
-                torch.FloatTensor).cuda().detach()
-            corrected_probabs = probabs + correction.detach()
+            unpert_probs = F.softmax(unpert_logits[:, -1, :], dim=-1)
+            unpert_probs = (
+                    unpert_probs + SMALL_CONST *
+                    (unpert_probs <= SMALL_CONST).float().to(device).detach()
+            )
+            correction = SMALL_CONST * (probs <= SMALL_CONST).float().to(device).detach()
+            corrected_probs = probs + correction.detach()
             kl_loss = kl_scale * (
-                (corrected_probabs * (corrected_probabs / p).log()).sum())
-            print(' kl_loss', (kl_loss).data.cpu().numpy())
-            loss += kl_loss  # + discrim_loss
+                (corrected_probs * (corrected_probs / unpert_probs).log()).sum()
+            )
+            print(' kl_loss', kl_loss.data.cpu().numpy())
+            loss += kl_loss
 
         loss_per_iter.append(loss.data.cpu().numpy())
-
         print(' pplm_loss', (loss - kl_loss).data.cpu().numpy())
 
+        # compute gradients
         loss.backward()
-        if grad_norms is not None and loss_type == 1:
+
+        # calculate gradient norms
+        if grad_norms is not None and loss_type == PPLM_BOW:
             grad_norms = [
                 torch.max(grad_norms[index], torch.norm(p_.grad * window_mask))
-                for index, p_ in
-                enumerate(past_perturb)]
+                for index, p_ in enumerate(curr_perturbation)
+            ]
         else:
-            grad_norms = [(torch.norm(p_.grad * window_mask) + SMALL_CONST) for
-                          index, p_ in enumerate(past_perturb)]
+            grad_norms = [
+                (torch.norm(p_.grad * window_mask) + SMALL_CONST)
+                for index, p_ in enumerate(curr_perturbation)
+            ]
 
+        # normalize gradients
         grad = [
-            -stepsize * (p_.grad * window_mask / grad_norms[
-                index] ** gamma).data.cpu().numpy()
-            for index, p_ in enumerate(past_perturb)]
-        past_perturb_orig = list(map(add, grad, past_perturb_orig))
+            -stepsize *
+            (p_.grad * window_mask / grad_norms[index] ** gamma).data.cpu().numpy()
+            for index, p_ in enumerate(curr_perturbation)
+        ]
 
-        for p_ in past_perturb:
+        # accumulate gradient
+        grad_accumulator = list(map(add, grad, grad_accumulator))
+
+        # reset gradients, just to make sure
+        for p_ in curr_perturbation:
             p_.grad.data.zero_()
 
+        # removing past from the graph
         new_past = []
-        for p in past:
-            new_past.append(p.detach())
-
+        for p_ in past:
+            new_past.append(p_.detach())
         past = new_past
 
-    past_perturb = [torch.from_numpy(p_) for p_ in past_perturb_orig]
-    past_perturb = [to_var(p_, requires_grad=True) for p_ in past_perturb]
-    perturbed_past = list(map(add, past, past_perturb))
+    # apply the accumulated perturbations to the past
+    grad_accumulator = [
+        to_var(torch.from_numpy(p_), requires_grad=True)
+        for p_ in grad_accumulator
+    ]
+    pert_past = list(map(add, past, grad_accumulator))
 
-    return perturbed_past, new_accumulated_hidden, grad_norms, loss_per_iter
+    return pert_past, new_accumulated_hidden, grad_norms, loss_per_iter
 
 
 def get_classifier(
@@ -532,6 +550,7 @@ def generate_text_pplm(
                     horizon_length=horizon_length,
                     decay=decay,
                     gamma=gamma,
+                    device=device
                 )
                 loss_in_time.append(loss_this_iter)
             else:
@@ -562,7 +581,7 @@ def generate_text_pplm(
             pert_probs = ((pert_probs ** gm_scale) * (
                     unpert_probs ** (1 - gm_scale)))  # + SMALL_CONST
             pert_probs = top_k_filter(pert_probs, k=top_k,
-                                     probs=True)  # + SMALL_CONST
+                                      probs=True)  # + SMALL_CONST
 
             # rescale
             if torch.sum(pert_probs) <= 1:
@@ -662,7 +681,8 @@ def run_model():
     parser.add_argument("--decay", action="store_true",
                         help="whether to decay or not")
     parser.add_argument("--gamma", type=float, default=1.5)
-    parser.add_argument("--colorama", action="store_true", help="colors keywords")
+    parser.add_argument("--colorama", action="store_true",
+                        help="colors keywords")
 
     args = parser.parse_args()
 

From afc7dcd94d2480e1fa6ef91ec8e9029142566612 Mon Sep 17 00:00:00 2001
From: piero <piero@uber.com>
Date: Wed, 27 Nov 2019 20:08:53 -0800
Subject: [PATCH 192/505] Now run_pplm works on cpu. Identical output as before
 (when using gpu).

---
 examples/run_pplm.py | 28 ++++++++++++++++------------
 1 file changed, 16 insertions(+), 12 deletions(-)

diff --git a/examples/run_pplm.py b/examples/run_pplm.py
index 77758759d9..0d6b0d635d 100644
--- a/examples/run_pplm.py
+++ b/examples/run_pplm.py
@@ -84,9 +84,11 @@ DISCRIMINATOR_MODELS_PARAMS = {
 }
 
 
-def to_var(x, requires_grad=False, volatile=False):
-    if torch.cuda.is_available():
+def to_var(x, requires_grad=False, volatile=False, device='cuda'):
+    if torch.cuda.is_available() and device == 'cuda':
         x = x.cuda()
+    elif device != 'cuda':
+        x = x.to(device)
     return Variable(x, requires_grad=requires_grad, volatile=volatile)
 
 
@@ -182,7 +184,7 @@ def perturb_past(
     for i in range(num_iterations):
         print("Iteration ", i + 1)
         curr_perturbation = [
-            to_var(torch.from_numpy(p_), requires_grad=True)
+            to_var(torch.from_numpy(p_), requires_grad=True, device=device)
             for p_ in grad_accumulator
         ]
 
@@ -290,7 +292,7 @@ def perturb_past(
 
     # apply the accumulated perturbations to the past
     grad_accumulator = [
-        to_var(torch.from_numpy(p_), requires_grad=True)
+        to_var(torch.from_numpy(p_), requires_grad=True, device=device)
         for p_ in grad_accumulator
     ]
     pert_past = list(map(add, past, grad_accumulator))
@@ -300,7 +302,7 @@ def perturb_past(
 
 def get_classifier(
         name: Optional[str], label_class: Union[str, int],
-        device: Union[str, torch.device]
+        device: str
 ) -> Tuple[Optional[ClassificationHead], Optional[int]]:
     if name is None:
         return None, None
@@ -355,16 +357,16 @@ def get_bag_of_words_indices(bag_of_words_ids_or_paths: List[str]) -> List[
     return bow_indices
 
 
-def build_bows_one_hot_vectors(bow_indices):
+def build_bows_one_hot_vectors(bow_indices, device='cuda'):
     if bow_indices is None:
         return None
 
     one_hot_bows_vectors = []
     for single_bow in bow_indices:
         single_bow = list(filter(lambda x: len(x) <= 1, single_bow))
-        single_bow = torch.tensor(single_bow).cuda()
+        single_bow = torch.tensor(single_bow).to(device)
         num_words = single_bow.shape[0]
-        one_hot_bow = torch.zeros(num_words, TOKENIZER.vocab_size).cuda()
+        one_hot_bow = torch.zeros(num_words, TOKENIZER.vocab_size).to(device)
         one_hot_bow.scatter_(1, single_bow, 1)
         one_hot_bows_vectors.append(one_hot_bow)
     return one_hot_bows_vectors
@@ -425,7 +427,8 @@ def full_text_generation(
         length=length,
         perturb=False
     )
-    torch.cuda.empty_cache()
+    if device == 'cuda':
+        torch.cuda.empty_cache()
 
     pert_gen_tok_texts = []
     discrim_losses = []
@@ -460,7 +463,8 @@ def full_text_generation(
             discrim_losses.append(discrim_loss.data.cpu().numpy())
         losses_in_time.append(loss_in_time)
 
-    torch.cuda.empty_cache()
+    if device == 'cuda':
+        torch.cuda.empty_cache()
 
     return unpert_gen_tok_text, pert_gen_tok_texts, discrim_losses, losses_in_time
 
@@ -496,7 +500,7 @@ def generate_text_pplm(
     )
 
     # collect one hot vectors for bags of words
-    one_hot_bows_vectors = build_bows_one_hot_vectors(bow_indices)
+    one_hot_bows_vectors = build_bows_one_hot_vectors(bow_indices, device)
 
     grad_norms = None
     last = None
@@ -563,7 +567,7 @@ def generate_text_pplm(
         if classifier is not None:
             ce_loss = torch.nn.CrossEntropyLoss()
             prediction = classifier(torch.mean(unpert_last_hidden, dim=1))
-            label = torch.tensor([label_class], device='cuda',
+            label = torch.tensor([label_class], device=device,
                                  dtype=torch.long)
             unpert_discrim_loss = ce_loss(prediction, label)
             print(

From 611961ade71042ff759712e6f680544ec5ff68b9 Mon Sep 17 00:00:00 2001
From: piero <piero@uber.com>
Date: Wed, 27 Nov 2019 21:34:49 -0800
Subject: [PATCH 193/505] Added tqdm to preprocessing

---
 examples/run_pplm_discrim_train.py | 206 ++++++++++++++---------------
 1 file changed, 102 insertions(+), 104 deletions(-)

diff --git a/examples/run_pplm_discrim_train.py b/examples/run_pplm_discrim_train.py
index 519e2de29a..5291ad4b51 100644
--- a/examples/run_pplm_discrim_train.py
+++ b/examples/run_pplm_discrim_train.py
@@ -18,13 +18,14 @@ import torch.utils.data as data
 from nltk.tokenize.treebank import TreebankWordDetokenizer
 from torchtext import data as torchtext_data
 from torchtext import datasets
+from tqdm import tqdm, trange
 
 from transformers import GPT2Tokenizer, GPT2LMHeadModel
 
 torch.manual_seed(0)
 np.random.seed(0)
 EPSILON = 1e-10
-device = 'cpu'
+device = "cpu"
 example_sentence = "This is incredible! I love it, this is the best chicken I have ever had."
 max_length_seq = 100
 
@@ -109,8 +110,8 @@ class Dataset(data.Dataset):
     def __getitem__(self, index):
         """Returns one data pair (source and target)."""
         data = {}
-        data['X'] = self.X[index]
-        data['y'] = self.y[index]
+        data["X"] = self.X[index]
+        data["y"] = self.y[index]
         return data
 
 
@@ -133,8 +134,8 @@ def collate_fn(data):
     for key in data[0].keys():
         item_info[key] = [d[key] for d in data]
 
-    x_batch, _ = pad_sequences(item_info['X'])
-    y_batch = torch.tensor(item_info['y'], dtype=torch.long)
+    x_batch, _ = pad_sequences(item_info["X"])
+    y_batch = torch.tensor(item_info["y"], dtype=torch.long)
 
     return x_batch, y_batch
 
@@ -144,8 +145,8 @@ def cached_collate_fn(data):
     for key in data[0].keys():
         item_info[key] = [d[key] for d in data]
 
-    x_batch = torch.cat(item_info['X'], 0)
-    y_batch = torch.tensor(item_info['y'], dtype=torch.long)
+    x_batch = torch.cat(item_info["X"], 0)
+    y_batch = torch.tensor(item_info["y"], dtype=torch.long)
 
     return x_batch, y_batch
 
@@ -168,7 +169,7 @@ def train_epoch(data_loader, discriminator, optimizer,
 
         if batch_idx % log_interval == 0:
             print(
-                'Train Epoch: {} [{}/{} ({:.0f}%)]\tLoss: {:.6f}'.format(
+                "Train Epoch: {} [{}/{} ({:.0f}%)]\tLoss: {:.6f}".format(
                     epoch + 1,
                     samples_so_far, len(data_loader.dataset),
                     100 * samples_so_far / len(data_loader.dataset), loss.item()
@@ -185,7 +186,7 @@ def evaluate_performance(data_loader, discriminator):
             input_t, target_t = input_t.to(device), target_t.to(device)
             output_t = discriminator(input_t)
             # sum up batch loss
-            test_loss += F.nll_loss(output_t, target_t, reduction='sum').item()
+            test_loss += F.nll_loss(output_t, target_t, reduction="sum").item()
             # get the index of the max log-probability
             pred_t = output_t.argmax(dim=1, keepdim=True)
             correct += pred_t.eq(target_t.view_as(pred_t)).sum().item()
@@ -193,8 +194,8 @@ def evaluate_performance(data_loader, discriminator):
     test_loss /= len(data_loader.dataset)
 
     print(
-        'Performance on test set: '
-        'Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)'.format(
+        "Performance on test set: "
+        "Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)".format(
             test_loss, correct, len(data_loader.dataset),
             100. * correct / len(data_loader.dataset)
         )
@@ -208,8 +209,8 @@ def predict(input_sentence, model, classes, cached=False):
         input_t = model.avg_representation(input_t)
 
     log_probs = model(input_t).data.cpu().numpy().flatten().tolist()
-    print('Input sentence:', input_sentence)
-    print('Predictions:', ", ".join(
+    print("Input sentence:", input_sentence)
+    print("Predictions:", ", ".join(
         "{}: {:.4f}".format(c, math.exp(log_prob)) for c, log_prob in
         zip(classes, log_probs)
     ))
@@ -222,7 +223,7 @@ def get_cached_data_loader(dataset, batch_size, discriminator, shuffle=False):
 
     xs = []
     ys = []
-    for batch_idx, (x, y) in enumerate(data_loader):
+    for batch_idx, (x, y) in enumerate(tqdm(data_loader, ascii=True)):
         with torch.no_grad():
             x = x.to(device)
             avg_rep = discriminator.avg_representation(x).cpu().detach()
@@ -240,16 +241,16 @@ def get_cached_data_loader(dataset, batch_size, discriminator, shuffle=False):
 
 
 def train_discriminator(
-        dataset, dataset_fp=None, pretrained_model='gpt2-medium',
+        dataset, dataset_fp=None, pretrained_model="gpt2-medium",
         epochs=10, batch_size=64, log_interval=10,
         save_model=False, cached=False, no_cuda=False):
     global device
     device = "cuda" if torch.cuda.is_available() and not no_cuda else "cpu"
 
-    print('Preprocessing {} dataset...'.format(dataset))
+    print("Preprocessing {} dataset...".format(dataset))
     start = time.time()
 
-    if dataset == 'SST':
+    if dataset == "SST":
         idx2class = ["positive", "negative", "very positive", "very negative",
                      "neutral"]
         class2idx = {c: i for i, c in enumerate(idx2class)}
@@ -271,7 +272,7 @@ def train_discriminator(
 
         x = []
         y = []
-        for i in range(len(train_data)):
+        for i in trange(len(train_data), ascii=True):
             seq = TreebankWordDetokenizer().detokenize(
                 vars(train_data[i])["text"]
             )
@@ -283,7 +284,7 @@ def train_discriminator(
 
         test_x = []
         test_y = []
-        for i in range(len(test_data)):
+        for i in trange(len(test_data), ascii=True):
             seq = TreebankWordDetokenizer().detokenize(
                 vars(test_data[i])["text"]
             )
@@ -301,7 +302,7 @@ def train_discriminator(
             "default_class": 2,
         }
 
-    elif dataset == 'clickbait':
+    elif dataset == "clickbait":
         idx2class = ["non_clickbait", "clickbait"]
         class2idx = {c: i for i, c in enumerate(idx2class)}
 
@@ -317,31 +318,33 @@ def train_discriminator(
                 try:
                     data.append(eval(line))
                 except:
-                    print('Error evaluating line {}: {}'.format(
+                    print("Error evaluating line {}: {}".format(
                         i, line
                     ))
                     continue
         x = []
         y = []
-        y = []
-        for i, d in enumerate(data):
-            try:
-                seq = discriminator.tokenizer.encode(d["text"])
+        with open("datasets/clickbait/clickbait_train_prefix.txt") as f:
+            for i, line in enumerate(tqdm(f, ascii=True)):
+                try:
+                    d = eval(line)
+                    seq = discriminator.tokenizer.encode(d["text"])
 
-                if len(seq) < max_length_seq:
-                    seq = torch.tensor(
-                        [50256] + seq, device=device, dtype=torch.long
-                    )
-                else:
-                    print("Line {} is longer than maximum length {}".format(
-                        i, max_length_seq
-                    ))
-                    continue
-                x.append(seq)
-                y.append(d['label'])
-            except:
-                print("Error tokenizing line {}, skipping it".format(i))
-                pass
+                    if len(seq) < max_length_seq:
+                        seq = torch.tensor(
+                            [50256] + seq, device=device, dtype=torch.long
+                        )
+                    else:
+                        print("Line {} is longer than maximum length {}".format(
+                            i, max_length_seq
+                        ))
+                        continue
+                    x.append(seq)
+                    y.append(d["label"])
+                except:
+                    print("Error evaluating / tokenizing"
+                          " line {}, skipping it".format(i))
+                    pass
 
         full_dataset = Dataset(x, y)
         train_size = int(0.9 * len(full_dataset))
@@ -358,7 +361,7 @@ def train_discriminator(
             "default_class": 1,
         }
 
-    elif dataset == 'toxic':
+    elif dataset == "toxic":
         idx2class = ["non_toxic", "toxic"]
         class2idx = {c: i for i, c in enumerate(idx2class)}
 
@@ -368,37 +371,29 @@ def train_discriminator(
             cached_mode=cached
         ).to(device)
 
-        with open("datasets/toxic/toxic_train.txt") as f:
-            data = []
-            for i, line in enumerate(f):
-                try:
-                    data.append(eval(line))
-                except:
-                    print('Error evaluating line {}: {}'.format(
-                        i, line
-                    ))
-                    continue
-
         x = []
         y = []
-        for i, d in enumerate(data):
-            try:
-                seq = discriminator.tokenizer.encode(d["text"])
+        with open("datasets/toxic/toxic_train.txt") as f:
+            for i, line in enumerate(tqdm(f, ascii=True)):
+                try:
+                    d = eval(line)
+                    seq = discriminator.tokenizer.encode(d["text"])
 
-                if len(seq) < max_length_seq:
-                    seq = torch.tensor(
-                        [50256] + seq, device=device, dtype=torch.long
-                    )
-                else:
-                    print("Line {} is longer than maximum length {}".format(
-                        i, max_length_seq
-                    ))
-                    continue
-                x.append(seq)
-                y.append(int(np.sum(d['label']) > 0))
-            except:
-                print("Error tokenizing line {}, skipping it".format(i))
-                pass
+                    if len(seq) < max_length_seq:
+                        seq = torch.tensor(
+                            [50256] + seq, device=device, dtype=torch.long
+                        )
+                    else:
+                        print("Line {} is longer than maximum length {}".format(
+                            i, max_length_seq
+                        ))
+                        continue
+                    x.append(seq)
+                    y.append(int(np.sum(d["label"]) > 0))
+                except:
+                    print("Error evaluating / tokenizing"
+                          " line {}, skipping it".format(i))
+                    pass
 
         full_dataset = Dataset(x, y)
         train_size = int(0.9 * len(full_dataset))
@@ -415,18 +410,18 @@ def train_discriminator(
             "default_class": 0,
         }
 
-    else:  # if dataset == 'generic':
+    else:  # if dataset == "generic":
         # This assumes the input dataset is a TSV with the following structure:
         # class \t text
 
         if dataset_fp is None:
-            raise ValueError('When generic dataset is selected, '
-                             'dataset_fp needs to be specified aswell.')
+            raise ValueError("When generic dataset is selected, "
+                             "dataset_fp needs to be specified aswell.")
 
         classes = set()
         with open(dataset_fp) as f:
-            csv_reader = csv.reader(f, delimiter='\t')
-            for row in csv_reader:
+            csv_reader = csv.reader(f, delimiter="\t")
+            for row in tqdm(csv_reader, ascii=True):
                 if row:
                     classes.add(row[0])
 
@@ -442,8 +437,8 @@ def train_discriminator(
         x = []
         y = []
         with open(dataset_fp) as f:
-            csv_reader = csv.reader(f, delimiter='\t')
-            for i, row in enumerate(csv_reader):
+            csv_reader = csv.reader(f, delimiter="\t")
+            for i, row in enumerate(tqdm(csv_reader, ascii=True)):
                 if row:
                     label = row[0]
                     text = row[1]
@@ -458,9 +453,10 @@ def train_discriminator(
                             )
 
                         else:
-                            print("Line {} is longer than maximum length {}".format(
-                                i, max_length_seq
-                            ))
+                            print(
+                                "Line {} is longer than maximum length {}".format(
+                                    i, max_length_seq
+                                ))
                             continue
 
                         x.append(seq)
@@ -487,12 +483,14 @@ def train_discriminator(
         }
 
     end = time.time()
-    print('Preprocessed {} data points'.format(
+    print("Preprocessed {} data points".format(
         len(train_dataset) + len(test_dataset))
     )
     print("Data preprocessing took: {:.3f}s".format(end - start))
 
     if cached:
+        print("Building representation cache...")
+
         start = time.time()
 
         train_loader = get_cached_data_loader(
@@ -524,7 +522,7 @@ def train_discriminator(
 
     for epoch in range(epochs):
         start = time.time()
-        print('\nEpoch', epoch + 1)
+        print("\nEpoch", epoch + 1)
 
         train_epoch(
             discriminator=discriminator,
@@ -553,31 +551,31 @@ def train_discriminator(
                        "{}_classifier_head_epoch_{}.pt".format(dataset, epoch))
 
 
-if __name__ == '__main__':
+if __name__ == "__main__":
     parser = argparse.ArgumentParser(
-        description='Train a discriminator on top of GPT-2 representations')
-    parser.add_argument('--dataset', type=str, default='SST',
-                        choices=('SST', 'clickbait', 'toxic', 'generic'),
-                        help='dataset to train the discriminator on.'
-                             'In case of generic, the dataset is expected'
-                             'to be a TSBV file with structure: class \\t text')
-    parser.add_argument('--dataset_fp', type=str, default='',
-                        help='File path of the dataset to use. '
-                             'Needed only in case of generic datadset')
-    parser.add_argument('--pretrained_model', type=str, default='gpt2-medium',
-                        help='Pretrained model to use as encoder')
-    parser.add_argument('--epochs', type=int, default=10, metavar='N',
-                        help='Number of training epochs')
-    parser.add_argument('--batch_size', type=int, default=64, metavar='N',
-                        help='input batch size for training (default: 64)')
-    parser.add_argument('--log_interval', type=int, default=10, metavar='N',
-                        help='how many batches to wait before logging training status')
-    parser.add_argument('--save_model', action='store_true',
-                        help='whether to save the model')
-    parser.add_argument('--cached', action='store_true',
-                        help='whether to cache the input representations')
-    parser.add_argument('--no_cuda', action='store_true',
-                        help='use to turn off cuda')
+        description="Train a discriminator on top of GPT-2 representations")
+    parser.add_argument("--dataset", type=str, default="SST",
+                        choices=("SST", "clickbait", "toxic", "generic"),
+                        help="dataset to train the discriminator on."
+                             "In case of generic, the dataset is expected"
+                             "to be a TSBV file with structure: class \\t text")
+    parser.add_argument("--dataset_fp", type=str, default="",
+                        help="File path of the dataset to use. "
+                             "Needed only in case of generic datadset")
+    parser.add_argument("--pretrained_model", type=str, default="gpt2-medium",
+                        help="Pretrained model to use as encoder")
+    parser.add_argument("--epochs", type=int, default=10, metavar="N",
+                        help="Number of training epochs")
+    parser.add_argument("--batch_size", type=int, default=64, metavar="N",
+                        help="input batch size for training (default: 64)")
+    parser.add_argument("--log_interval", type=int, default=10, metavar="N",
+                        help="how many batches to wait before logging training status")
+    parser.add_argument("--save_model", action="store_true",
+                        help="whether to save the model")
+    parser.add_argument("--cached", action="store_true",
+                        help="whether to cache the input representations")
+    parser.add_argument("--no_cuda", action="store_true",
+                        help="use to turn off cuda")
     args = parser.parse_args()
 
     train_discriminator(**(vars(args)))

From b0eaff36e6aefd45c0fe89bb2ab86a495e4e735f Mon Sep 17 00:00:00 2001
From: piero <piero@uber.com>
Date: Wed, 27 Nov 2019 21:43:43 -0800
Subject: [PATCH 194/505] Added a +1 to epoch when saving weights

---
 examples/run_pplm_discrim_train.py | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/examples/run_pplm_discrim_train.py b/examples/run_pplm_discrim_train.py
index 5291ad4b51..fccfb14426 100644
--- a/examples/run_pplm_discrim_train.py
+++ b/examples/run_pplm_discrim_train.py
@@ -545,10 +545,11 @@ def train_discriminator(
         if save_model:
             # torch.save(discriminator.state_dict(),
             #           "{}_discriminator_{}.pt".format(
-            #               args.dataset, epoch
+            #               args.dataset, epoch + 1
             #               ))
             torch.save(discriminator.get_classifier().state_dict(),
-                       "{}_classifier_head_epoch_{}.pt".format(dataset, epoch))
+                       "{}_classifier_head_epoch_{}.pt".format(dataset,
+                                                               epoch + 1))
 
 
 if __name__ == "__main__":

From 7fd54b55a3f7c3134f8cc5a62f4cc447a5cd34de Mon Sep 17 00:00:00 2001
From: piero <piero@uber.com>
Date: Wed, 27 Nov 2019 21:45:19 -0800
Subject: [PATCH 195/505] Added support for generic discriminators

---
 examples/run_pplm.py | 77 +++++++++++++++++++++++++++++---------------
 1 file changed, 51 insertions(+), 26 deletions(-)

diff --git a/examples/run_pplm.py b/examples/run_pplm.py
index 0d6b0d635d..28aa66cc7d 100644
--- a/examples/run_pplm.py
+++ b/examples/run_pplm.py
@@ -14,17 +14,16 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-# TODO: add code for training a custom discriminator
-
 """
 Example command with bag of words:
 python examples/run_pplm.py -B space --cond_text "The president" --length 100 --gamma 1.5 --num_iterations 3 --num_samples 10 --stepsize 0.01 --window_length 5 --kl_scale 0.01 --gm_scale 0.95
 
 Example command with discriminator:
-python examples/run_pplm.py -D sentiment --label_class 3 --cond_text "The lake" --length 10 --gamma 1.0 --num_iterations 30 --num_samples 10 --stepsize 0.01 --kl_scale 0.01 --gm_scale 0.95
+python examples/run_pplm.py -D sentiment --class_label 3 --cond_text "The lake" --length 10 --gamma 1.0 --num_iterations 30 --num_samples 10 --stepsize 0.01 --kl_scale 0.01 --gm_scale 0.95
 """
 
 import argparse
+import json
 from operator import add
 from typing import List, Optional, Tuple, Union
 
@@ -121,7 +120,7 @@ def perturb_past(
         grad_norms=None,
         stepsize=0.01,
         classifier=None,
-        label_class=None,
+        class_label=None,
         one_hot_bows_vectors=None,
         loss_type=0,
         num_iterations=3,
@@ -230,7 +229,7 @@ def perturb_past(
             prediction = classifier(new_accumulated_hidden /
                                     (curr_length + 1 + horizon_length))
 
-            label = torch.tensor([label_class], device=device,
+            label = torch.tensor([class_label], device=device,
                                  dtype=torch.long)
             discrim_loss = ce_loss(prediction, label)
             print(" pplm_discrim_loss:", discrim_loss.data.cpu().numpy())
@@ -244,7 +243,8 @@ def perturb_past(
                     unpert_probs + SMALL_CONST *
                     (unpert_probs <= SMALL_CONST).float().to(device).detach()
             )
-            correction = SMALL_CONST * (probs <= SMALL_CONST).float().to(device).detach()
+            correction = SMALL_CONST * (probs <= SMALL_CONST).float().to(
+                device).detach()
             corrected_probs = probs + correction.detach()
             kl_loss = kl_scale * (
                 (corrected_probs * (corrected_probs / unpert_probs).log()).sum()
@@ -273,7 +273,8 @@ def perturb_past(
         # normalize gradients
         grad = [
             -stepsize *
-            (p_.grad * window_mask / grad_norms[index] ** gamma).data.cpu().numpy()
+            (p_.grad * window_mask / grad_norms[
+                index] ** gamma).data.cpu().numpy()
             for index, p_ in enumerate(curr_perturbation)
         ]
 
@@ -301,7 +302,7 @@ def perturb_past(
 
 
 def get_classifier(
-        name: Optional[str], label_class: Union[str, int],
+        name: Optional[str], class_label: Union[str, int],
         device: str
 ) -> Tuple[Optional[ClassificationHead], Optional[int]]:
     if name is None:
@@ -312,26 +313,29 @@ def get_classifier(
         class_size=params['class_size'],
         embed_size=params['embed_size']
     ).to(device)
-    resolved_archive_file = cached_path(params["url"])
+    if "url" in params:
+        resolved_archive_file = cached_path(params["url"])
+    else:
+        resolved_archive_file = params["path"]
     classifier.load_state_dict(
         torch.load(resolved_archive_file, map_location=device))
     classifier.eval()
 
-    if isinstance(label_class, str):
-        if label_class in params["class_vocab"]:
-            label_id = params["class_vocab"][label_class]
+    if isinstance(class_label, str):
+        if class_label in params["class_vocab"]:
+            label_id = params["class_vocab"][class_label]
         else:
             label_id = params["default_class"]
-            print("label_class {} not in class_vocab".format(label_class))
+            print("class_label {} not in class_vocab".format(class_label))
             print("available values are: {}".format(params["class_vocab"]))
             print("using default class {}".format(label_id))
 
-    elif isinstance(label_class, int):
-        if label_class in set(params["class_vocab"].values()):
-            label_id = label_class
+    elif isinstance(class_label, int):
+        if class_label in set(params["class_vocab"].values()):
+            label_id = class_label
         else:
             label_id = params["default_class"]
-            print("label_class {} not in class_vocab".format(label_class))
+            print("class_label {} not in class_vocab".format(class_label))
             print("available values are: {}".format(params["class_vocab"]))
             print("using default class {}".format(label_id))
 
@@ -379,7 +383,7 @@ def full_text_generation(
         device="cuda",
         sample=True,
         discrim=None,
-        label_class=None,
+        class_label=None,
         bag_of_words=None,
         length=100,
         grad_length=10000,
@@ -397,7 +401,7 @@ def full_text_generation(
 ):
     classifier, class_id = get_classifier(
         discrim,
-        label_class,
+        class_label,
         device
     )
 
@@ -443,7 +447,7 @@ def full_text_generation(
             perturb=True,
             bow_indices=bow_indices,
             classifier=classifier,
-            label_class=class_id,
+            class_label=class_id,
             loss_type=loss_type,
             length=length,
             grad_length=grad_length,
@@ -477,7 +481,7 @@ def generate_text_pplm(
         sample=True,
         perturb=True,
         classifier=None,
-        label_class=None,
+        class_label=None,
         bow_indices=None,
         loss_type=0,
         length=100,
@@ -545,7 +549,7 @@ def generate_text_pplm(
                     grad_norms=grad_norms,
                     stepsize=current_stepsize,
                     classifier=classifier,
-                    label_class=label_class,
+                    class_label=class_label,
                     one_hot_bows_vectors=one_hot_bows_vectors,
                     loss_type=loss_type,
                     num_iterations=num_iterations,
@@ -567,7 +571,7 @@ def generate_text_pplm(
         if classifier is not None:
             ce_loss = torch.nn.CrossEntropyLoss()
             prediction = classifier(torch.mean(unpert_last_hidden, dim=1))
-            label = torch.tensor([label_class], device=device,
+            label = torch.tensor([class_label], device=device,
                                  dtype=torch.long)
             unpert_discrim_loss = ce_loss(prediction, label)
             print(
@@ -613,6 +617,20 @@ def generate_text_pplm(
     return output_so_far, unpert_discrim_loss, loss_in_time
 
 
+def set_generic_model_params(discrim_weights, discrim_meta):
+    if discrim_weights is None:
+        raise ValueError('When using a generic discriminator, '
+                         'discrim_weights need to be specified')
+    if discrim_meta is None:
+        raise ValueError('When using a generic discriminator, '
+                         'discrim_meta need to be specified')
+
+    with open(discrim_meta, 'r') as discrim_meta_file:
+        meta = json.load(discrim_meta_file)
+    meta['path'] = discrim_weights
+    DISCRIMINATOR_MODELS_PARAMS['generic'] = meta
+
+
 def run_model():
     parser = argparse.ArgumentParser()
     parser.add_argument(
@@ -636,11 +654,15 @@ def run_model():
         "-D",
         type=str,
         default=None,
-        choices=("clickbait", "sentiment", "toxicity"),
-        help="Discriminator to use for loss-type 2",
+        choices=("clickbait", "sentiment", "toxicity", "generic"),
+        help="Discriminator to use",
     )
+    parser.add_argument('--discrim_weights', type=str, default=None,
+                        help='Weights for the generic discriminator')
+    parser.add_argument('--discrim_meta', type=str, default=None,
+                        help='Meta information for the generic discriminator')
     parser.add_argument(
-        "--label_class",
+        "--class_label",
         type=int,
         default=-1,
         help="Class label used for the discriminator",
@@ -697,6 +719,9 @@ def run_model():
     # set the device
     device = "cuda" if torch.cuda.is_available() and not args.no_cuda else "cpu"
 
+    if args.discrim == 'generic':
+        set_generic_model_params(args.discrim_weights, args.discrim_meta)
+
     # load pretrained model
     model = GPT2LMHeadModel.from_pretrained(
         args.model_path,

From 75904dae669249c9f5d4d4d57890fb6c537d1639 Mon Sep 17 00:00:00 2001
From: w4nderlust <w4nderlust@gmail.com>
Date: Fri, 29 Nov 2019 18:51:27 -0800
Subject: [PATCH 196/505] Removed global variable device

---
 examples/run_pplm_discrim_train.py | 47 ++++++++++++++++++------------
 1 file changed, 28 insertions(+), 19 deletions(-)

diff --git a/examples/run_pplm_discrim_train.py b/examples/run_pplm_discrim_train.py
index fccfb14426..db081e1a17 100644
--- a/examples/run_pplm_discrim_train.py
+++ b/examples/run_pplm_discrim_train.py
@@ -25,7 +25,6 @@ from transformers import GPT2Tokenizer, GPT2LMHeadModel
 torch.manual_seed(0)
 np.random.seed(0)
 EPSILON = 1e-10
-device = "cpu"
 example_sentence = "This is incredible! I love it, this is the best chicken I have ever had."
 max_length_seq = 100
 
@@ -55,7 +54,8 @@ class Discriminator(torch.nn.Module):
             self,
             class_size,
             pretrained_model="gpt2-medium",
-            cached_mode=False
+            cached_mode=False,
+            device='cpu'
     ):
         super(Discriminator, self).__init__()
         self.tokenizer = GPT2Tokenizer.from_pretrained(pretrained_model)
@@ -66,6 +66,7 @@ class Discriminator(torch.nn.Module):
             embed_size=self.embed_size
         )
         self.cached_mode = cached_mode
+        self.device = device
 
     def get_classifier(self):
         return self.classifier_head
@@ -78,7 +79,7 @@ class Discriminator(torch.nn.Module):
     def avg_representation(self, x):
         mask = x.ne(0).unsqueeze(2).repeat(
             1, 1, self.embed_size
-        ).float().to(device).detach()
+        ).float().to(self.device).detach()
         hidden, _ = self.encoder.transformer(x)
         masked_hidden = hidden * mask
         avg_hidden = torch.sum(masked_hidden, dim=1) / (
@@ -88,9 +89,9 @@ class Discriminator(torch.nn.Module):
 
     def forward(self, x):
         if self.cached_mode:
-            avg_hidden = x.to(device)
+            avg_hidden = x.to(self.device)
         else:
-            avg_hidden = self.avg_representation(x.to(device))
+            avg_hidden = self.avg_representation(x.to(self.device))
 
         logits = self.classifier_head(avg_hidden)
         probs = F.log_softmax(logits, dim=-1)
@@ -152,7 +153,7 @@ def cached_collate_fn(data):
 
 
 def train_epoch(data_loader, discriminator, optimizer,
-                epoch=0, log_interval=10):
+                epoch=0, log_interval=10, device='cpu'):
     samples_so_far = 0
     discriminator.train_custom()
     for batch_idx, (input_t, target_t) in enumerate(data_loader):
@@ -177,7 +178,7 @@ def train_epoch(data_loader, discriminator, optimizer,
             )
 
 
-def evaluate_performance(data_loader, discriminator):
+def evaluate_performance(data_loader, discriminator, device='cpu'):
     discriminator.eval()
     test_loss = 0
     correct = 0
@@ -202,7 +203,7 @@ def evaluate_performance(data_loader, discriminator):
     )
 
 
-def predict(input_sentence, model, classes, cached=False):
+def predict(input_sentence, model, classes, cached=False, device='cpu'):
     input_t = model.tokenizer.encode(input_sentence)
     input_t = torch.tensor([input_t], dtype=torch.long, device=device)
     if cached:
@@ -216,7 +217,8 @@ def predict(input_sentence, model, classes, cached=False):
     ))
 
 
-def get_cached_data_loader(dataset, batch_size, discriminator, shuffle=False):
+def get_cached_data_loader(dataset, batch_size, discriminator,
+                           shuffle=False, device='cpu'):
     data_loader = torch.utils.data.DataLoader(dataset=dataset,
                                               batch_size=batch_size,
                                               collate_fn=collate_fn)
@@ -244,7 +246,6 @@ def train_discriminator(
         dataset, dataset_fp=None, pretrained_model="gpt2-medium",
         epochs=10, batch_size=64, log_interval=10,
         save_model=False, cached=False, no_cuda=False):
-    global device
     device = "cuda" if torch.cuda.is_available() and not no_cuda else "cpu"
 
     print("Preprocessing {} dataset...".format(dataset))
@@ -258,7 +259,8 @@ def train_discriminator(
         discriminator = Discriminator(
             class_size=len(idx2class),
             pretrained_model=pretrained_model,
-            cached_mode=cached
+            cached_mode=cached,
+            device=device
         ).to(device)
 
         text = torchtext_data.Field()
@@ -309,7 +311,8 @@ def train_discriminator(
         discriminator = Discriminator(
             class_size=len(idx2class),
             pretrained_model=pretrained_model,
-            cached_mode=cached
+            cached_mode=cached,
+            device=device
         ).to(device)
 
         with open("datasets/clickbait/clickbait_train_prefix.txt") as f:
@@ -368,7 +371,8 @@ def train_discriminator(
         discriminator = Discriminator(
             class_size=len(idx2class),
             pretrained_model=pretrained_model,
-            cached_mode=cached
+            cached_mode=cached,
+            device=device
         ).to(device)
 
         x = []
@@ -431,7 +435,8 @@ def train_discriminator(
         discriminator = Discriminator(
             class_size=len(idx2class),
             pretrained_model=pretrained_model,
-            cached_mode=cached
+            cached_mode=cached,
+            device=device
         ).to(device)
 
         x = []
@@ -494,11 +499,12 @@ def train_discriminator(
         start = time.time()
 
         train_loader = get_cached_data_loader(
-            train_dataset, batch_size, discriminator, shuffle=True
+            train_dataset, batch_size, discriminator,
+            shuffle=True, device=device
         )
 
         test_loader = get_cached_data_loader(
-            test_dataset, batch_size, discriminator
+            test_dataset, batch_size, discriminator, device=device
         )
 
         end = time.time()
@@ -529,18 +535,21 @@ def train_discriminator(
             data_loader=train_loader,
             optimizer=optimizer,
             epoch=epoch,
-            log_interval=log_interval
+            log_interval=log_interval,
+            device=device
         )
         evaluate_performance(
             data_loader=test_loader,
-            discriminator=discriminator
+            discriminator=discriminator,
+            device=device
         )
 
         end = time.time()
         print("Epoch took: {:.3f}s".format(end - start))
 
         print("\nExample prediction")
-        predict(example_sentence, discriminator, idx2class, cached)
+        predict(example_sentence, discriminator, idx2class,
+                cached=cached, device=device)
 
         if save_model:
             # torch.save(discriminator.state_dict(),

From f10b925015b03612877fc2213e118d9507dd3ff2 Mon Sep 17 00:00:00 2001
From: w4nderlust <w4nderlust@gmail.com>
Date: Fri, 29 Nov 2019 19:59:02 -0800
Subject: [PATCH 197/505] Imrpovements: model_path renamed pretrained_model,
 tokenizer loaded from pretrained_model, pretrained_model set to
 discriminator's when discrim is specified, sample = False by default but cli
 parameter introduced. To obtain identical samples call the cli with --sample

---
 examples/run_pplm.py | 300 ++++++++++++++++++++++++++-----------------
 1 file changed, 185 insertions(+), 115 deletions(-)

diff --git a/examples/run_pplm.py b/examples/run_pplm.py
index 28aa66cc7d..8516454f86 100644
--- a/examples/run_pplm.py
+++ b/examples/run_pplm.py
@@ -43,7 +43,6 @@ PPLM_DISCRIM = 2
 PPLM_BOW_DISCRIM = 3
 SMALL_CONST = 1e-15
 BIG_CONST = 1e10
-TOKENIZER = GPT2Tokenizer.from_pretrained("gpt2-medium")
 
 BAG_OF_WORDS_ARCHIVE_MAP = {
     'kitchen': "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/bow/kitchen.txt",
@@ -65,6 +64,7 @@ DISCRIMINATOR_MODELS_PARAMS = {
         "embed_size": 1024,
         "class_vocab": {"non_clickbait": 0, "clickbait": 1},
         "default_class": 1,
+        "pretrained_model": "gpt2-medium",
     },
     "sentiment": {
         "url": "http://s.yosinski.com/SST_classifier_head.pt",
@@ -72,6 +72,7 @@ DISCRIMINATOR_MODELS_PARAMS = {
         "embed_size": 1024,
         "class_vocab": {"very_positive": 2, "very_negative": 3},
         "default_class": 3,
+        "pretrained_model": "gpt2-medium",
     },
     "toxicity": {
         "url": "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/discriminators/toxicity_classifierhead.pt",
@@ -79,6 +80,7 @@ DISCRIMINATOR_MODELS_PARAMS = {
         "embed_size": 1024,
         "class_vocab": {"non_toxic": 0, "toxic": 1},
         "default_class": 0,
+        "pretrained_model": "gpt2-medium",
     },
 }
 
@@ -345,8 +347,9 @@ def get_classifier(
     return classifier, label_id
 
 
-def get_bag_of_words_indices(bag_of_words_ids_or_paths: List[str]) -> List[
-    List[List[int]]]:
+def get_bag_of_words_indices(bag_of_words_ids_or_paths: List[str], tokenizer) -> \
+        List[
+            List[List[int]]]:
     bow_indices = []
     for id_or_path in bag_of_words_ids_or_paths:
         if id_or_path in BAG_OF_WORDS_ARCHIVE_MAP:
@@ -356,12 +359,12 @@ def get_bag_of_words_indices(bag_of_words_ids_or_paths: List[str]) -> List[
         with open(filepath, "r") as f:
             words = f.read().strip().split("\n")
         bow_indices.append(
-            [TOKENIZER.encode(word.strip(), add_prefix_space=True) for word in
+            [tokenizer.encode(word.strip(), add_prefix_space=True) for word in
              words])
     return bow_indices
 
 
-def build_bows_one_hot_vectors(bow_indices, device='cuda'):
+def build_bows_one_hot_vectors(bow_indices, tokenizer, device='cuda'):
     if bow_indices is None:
         return None
 
@@ -370,7 +373,7 @@ def build_bows_one_hot_vectors(bow_indices, device='cuda'):
         single_bow = list(filter(lambda x: len(x) <= 1, single_bow))
         single_bow = torch.tensor(single_bow).to(device)
         num_words = single_bow.shape[0]
-        one_hot_bow = torch.zeros(num_words, TOKENIZER.vocab_size).to(device)
+        one_hot_bow = torch.zeros(num_words, tokenizer.vocab_size).to(device)
         one_hot_bow.scatter_(1, single_bow, 1)
         one_hot_bows_vectors.append(one_hot_bow)
     return one_hot_bows_vectors
@@ -378,10 +381,11 @@ def build_bows_one_hot_vectors(bow_indices, device='cuda'):
 
 def full_text_generation(
         model,
+        tokenizer,
         context=None,
         num_samples=1,
         device="cuda",
-        sample=True,
+        sample=False,
         discrim=None,
         class_label=None,
         bag_of_words=None,
@@ -407,7 +411,8 @@ def full_text_generation(
 
     bow_indices = []
     if bag_of_words:
-        bow_indices = get_bag_of_words_indices(bag_of_words.split(";"))
+        bow_indices = get_bag_of_words_indices(bag_of_words.split(";"),
+                                               tokenizer)
 
     if bag_of_words and classifier:
         print("Both PPLM-BoW and PPLM-Discrim are on. This is not optimized.")
@@ -426,9 +431,11 @@ def full_text_generation(
 
     unpert_gen_tok_text, _, _ = generate_text_pplm(
         model=model,
+        tokenizer=tokenizer,
         context=context,
         device=device,
         length=length,
+        sample=sample,
         perturb=False
     )
     if device == 'cuda':
@@ -441,6 +448,7 @@ def full_text_generation(
     for i in range(num_samples):
         pert_gen_tok_text, discrim_loss, loss_in_time = generate_text_pplm(
             model=model,
+            tokenizer=tokenizer,
             context=context,
             device=device,
             sample=sample,
@@ -475,10 +483,11 @@ def full_text_generation(
 
 def generate_text_pplm(
         model,
+        tokenizer,
         context=None,
         past=None,
         device="cuda",
-        sample=True,
+        sample=False,
         perturb=True,
         classifier=None,
         class_label=None,
@@ -504,7 +513,8 @@ def generate_text_pplm(
     )
 
     # collect one hot vectors for bags of words
-    one_hot_bows_vectors = build_bows_one_hot_vectors(bow_indices, device)
+    one_hot_bows_vectors = build_bows_one_hot_vectors(bow_indices, tokenizer,
+                                                      device)
 
     grad_norms = None
     last = None
@@ -612,7 +622,7 @@ def generate_text_pplm(
             else torch.cat((output_so_far, last), dim=1)
         )
 
-        print(TOKENIZER.decode(output_so_far.tolist()[0]))
+        print(tokenizer.decode(output_so_far.tolist()[0]))
 
     return output_so_far, unpert_discrim_loss, loss_in_time
 
@@ -631,10 +641,167 @@ def set_generic_model_params(discrim_weights, discrim_meta):
     DISCRIMINATOR_MODELS_PARAMS['generic'] = meta
 
 
-def run_model():
+def run_pplm_example(
+        pretrained_model="gpt2-medium",
+        cond_text="",
+        uncond=False,
+        num_samples=1,
+        bag_of_words=None,
+        discrim=None,
+        discrim_weights=None,
+        discrim_meta=None,
+        class_label=-1,
+        length=100,
+        stepsize=0.02,
+        temperature=1.0,
+        top_k=10,
+        sample=False,
+        num_iterations=3,
+        grad_length=10000,
+        horizon_length=1,
+        window_length=0,
+        decay=False,
+        gamma=1.5,
+        gm_scale=0.9,
+        kl_scale=0.01,
+        seed=0,
+        no_cuda=False,
+        colorama=False
+):
+    # set Random seed
+    torch.manual_seed(seed)
+    np.random.seed(seed)
+
+    # set the device
+    device = "cuda" if torch.cuda.is_available() and not no_cuda else "cpu"
+
+    if discrim == 'generic':
+        set_generic_model_params(discrim_weights, discrim_meta)
+
+    if discrim is not None:
+        pretrained_model = DISCRIMINATOR_MODELS_PARAMS[discrim][
+            "pretrained_model"
+        ]
+        print("discrim = {}, setting pretrained_model "
+              "to discriminator's = {}".format(discrim, pretrained_model))
+
+    # load pretrained model
+    model = GPT2LMHeadModel.from_pretrained(
+        pretrained_model,
+        output_hidden_states=True
+    )
+    model.to(device)
+    model.eval()
+
+    # load tokenizer
+    tokenizer = GPT2Tokenizer.from_pretrained(pretrained_model)
+
+    # Freeze GPT-2 weights
+    for param in model.parameters():
+        param.requires_grad = False
+
+    # figure out conditioning text
+    if uncond:
+        tokenized_cond_text = tokenizer.encode(
+            [tokenizer.bos_token]
+        )
+    else:
+        raw_text = cond_text
+        while not raw_text:
+            print("Did you forget to add `--cond_text`? ")
+            raw_text = input("Model prompt >>> ")
+        tokenized_cond_text = tokenizer.encode(tokenizer.bos_token + raw_text)
+
+    print("= Prefix of sentence =")
+    print(tokenizer.decode(tokenized_cond_text))
+    print()
+
+    # generate unperturbed and perturbed texts
+
+    # full_text_generation returns:
+    # unpert_gen_tok_text, pert_gen_tok_texts, discrim_losses, losses_in_time
+    unpert_gen_tok_text, pert_gen_tok_texts, _, _ = full_text_generation(
+        model=model,
+        tokenizer=tokenizer,
+        context=tokenized_cond_text,
+        device=device,
+        num_samples=num_samples,
+        bag_of_words=bag_of_words,
+        discrim=discrim,
+        class_label=class_label,
+        length=length,
+        stepsize=stepsize,
+        temperature=temperature,
+        top_k=top_k,
+        sample=sample,
+        num_iterations=num_iterations,
+        grad_length=grad_length,
+        horizon_length=horizon_length,
+        window_length=window_length,
+        decay=decay,
+        gamma=gamma,
+        gm_scale=gm_scale,
+        kl_scale=kl_scale,
+    )
+
+    # untokenize unperturbed text
+    unpert_gen_text = tokenizer.decode(unpert_gen_tok_text.tolist()[0])
+
+    print("=" * 80)
+    print("= Unperturbed generated text =")
+    print(unpert_gen_text)
+    print()
+
+    generated_texts = []
+
+    bow_word_ids = set()
+    if bag_of_words and colorama:
+        bow_indices = get_bag_of_words_indices(bag_of_words.split(";"),
+                                               tokenizer)
+        for single_bow_list in bow_indices:
+            # filtering all words in the list composed of more than 1 token
+            filtered = list(filter(lambda x: len(x) <= 1, single_bow_list))
+            # w[0] because we are sure w has only 1 item because previous fitler
+            bow_word_ids.update(w[0] for w in filtered)
+
+    # iterate through the perturbed texts
+    for i, pert_gen_tok_text in enumerate(pert_gen_tok_texts):
+        try:
+            # untokenize unperturbed text
+            if colorama:
+                import colorama
+
+                pert_gen_text = ''
+                for word_id in pert_gen_tok_text.tolist()[0]:
+                    if word_id in bow_word_ids:
+                        pert_gen_text += '{}{}{}'.format(
+                            colorama.Fore.RED,
+                            tokenizer.decode([word_id]),
+                            colorama.Style.RESET_ALL
+                        )
+                    else:
+                        pert_gen_text += tokenizer.decode([word_id])
+            else:
+                pert_gen_text = tokenizer.decode(pert_gen_tok_text.tolist()[0])
+
+            print("= Perturbed generated text {} =".format(i + 1))
+            print(pert_gen_text)
+            print()
+        except:
+            pass
+
+        # keep the prefix, perturbed seq, original seq for each index
+        generated_texts.append(
+            (tokenized_cond_text, pert_gen_tok_text, unpert_gen_tok_text)
+        )
+
+    return
+
+
+if __name__ == '__main__':
     parser = argparse.ArgumentParser()
     parser.add_argument(
-        "--model_path",
+        "--pretrained_model",
         "-M",
         type=str,
         default="gpt2-medium",
@@ -675,6 +842,10 @@ def run_model():
     parser.add_argument("--gm_scale", type=float, default=0.9)
     parser.add_argument("--kl_scale", type=float, default=0.01)
     parser.add_argument("--no_cuda", action="store_true", help="no cuda")
+    parser.add_argument(
+        "--sample", action="store_true",
+        help="Generate from end-of-text as prefix"
+    )
     parser.add_argument(
         "--uncond", action="store_true",
         help="Generate from end-of-text as prefix"
@@ -711,105 +882,4 @@ def run_model():
                         help="colors keywords")
 
     args = parser.parse_args()
-
-    # set Random seed
-    torch.manual_seed(args.seed)
-    np.random.seed(args.seed)
-
-    # set the device
-    device = "cuda" if torch.cuda.is_available() and not args.no_cuda else "cpu"
-
-    if args.discrim == 'generic':
-        set_generic_model_params(args.discrim_weights, args.discrim_meta)
-
-    # load pretrained model
-    model = GPT2LMHeadModel.from_pretrained(
-        args.model_path,
-        output_hidden_states=True
-    )
-    model.to(device)
-    model.eval()
-
-    # Freeze GPT-2 weights
-    for param in model.parameters():
-        param.requires_grad = False
-
-    # figure out conditioning text
-    if args.uncond:
-        tokenized_cond_text = TOKENIZER.encode(
-            [TOKENIZER.bos_token]
-        )
-    else:
-        raw_text = args.cond_text
-        while not raw_text:
-            print("Did you forget to add `--cond_text`? ")
-            raw_text = input("Model prompt >>> ")
-        tokenized_cond_text = TOKENIZER.encode(TOKENIZER.bos_token + raw_text)
-
-    print("= Prefix of sentence =")
-    print(TOKENIZER.decode(tokenized_cond_text))
-    print()
-
-    # generate unperturbed and perturbed texts
-
-    # full_text_generation returns:
-    # unpert_gen_tok_text, pert_gen_tok_texts, discrim_losses, losses_in_time
-    unpert_gen_tok_text, pert_gen_tok_texts, _, _ = full_text_generation(
-        model=model, context=tokenized_cond_text, device=device, **vars(args)
-    )
-
-    # untokenize unperturbed text
-    unpert_gen_text = TOKENIZER.decode(unpert_gen_tok_text.tolist()[0])
-
-    print("=" * 80)
-    print("= Unperturbed generated text =")
-    print(unpert_gen_text)
-    print()
-
-    generated_texts = []
-
-    bow_word_ids = set()
-    if args.bag_of_words and args.colorama:
-        bow_indices = get_bag_of_words_indices(args.bag_of_words.split(";"))
-        for single_bow_list in bow_indices:
-            # filtering all words in the list composed of more than 1 token
-            filtered = list(filter(lambda x: len(x) <= 1, single_bow_list))
-            # w[0] because we are sure w has only 1 item because previous fitler
-            bow_word_ids.update(w[0] for w in filtered)
-
-    # iterate through the perturbed texts
-    for i, pert_gen_tok_text in enumerate(pert_gen_tok_texts):
-        try:
-            # untokenize unperturbed text
-            if args.colorama:
-                import colorama
-
-                pert_gen_text = ''
-                for word_id in pert_gen_tok_text.tolist()[0]:
-                    if word_id in bow_word_ids:
-                        pert_gen_text += '{}{}{}'.format(
-                            colorama.Fore.RED,
-                            TOKENIZER.decode([word_id]),
-                            colorama.Style.RESET_ALL
-                        )
-                    else:
-                        pert_gen_text += TOKENIZER.decode([word_id])
-            else:
-                pert_gen_text = TOKENIZER.decode(pert_gen_tok_text.tolist()[0])
-
-            print("= Perturbed generated text {} =".format(i + 1))
-            print(pert_gen_text)
-            print()
-        except:
-            pass
-
-        # keep the prefix, perturbed seq, original seq for each index
-        generated_texts.append(
-            (tokenized_cond_text, pert_gen_tok_text, unpert_gen_tok_text)
-        )
-
-    return
-
-
-if __name__ == '__main__':
-    run_model()
+    run_pplm_example(**vars(args))

From f42816e7fca8280927790f74c6e280c37d49b280 Mon Sep 17 00:00:00 2001
From: w4nderlust <w4nderlust@gmail.com>
Date: Fri, 29 Nov 2019 20:00:43 -0800
Subject: [PATCH 198/505] Added additional check for url and path in
 discriminator model params

---
 examples/run_pplm.py | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/examples/run_pplm.py b/examples/run_pplm.py
index 8516454f86..9ddd42681e 100644
--- a/examples/run_pplm.py
+++ b/examples/run_pplm.py
@@ -317,8 +317,11 @@ def get_classifier(
     ).to(device)
     if "url" in params:
         resolved_archive_file = cached_path(params["url"])
-    else:
+    elif "path" in params:
         resolved_archive_file = params["path"]
+    else:
+        raise ValueError("Either url or path have to be specified "
+                         "in the discriminator model parameters")
     classifier.load_state_dict(
         torch.load(resolved_archive_file, map_location=device))
     classifier.eval()

From 893d0d64fe008a63eca89b8750502fa5cb684439 Mon Sep 17 00:00:00 2001
From: w4nderlust <w4nderlust@gmail.com>
Date: Fri, 29 Nov 2019 20:19:33 -0800
Subject: [PATCH 199/505] Changed order of some parameters to be more
 consistent. Identical results.

---
 examples/run_pplm.py | 111 +++++++++++++++++++++----------------------
 1 file changed, 55 insertions(+), 56 deletions(-)

diff --git a/examples/run_pplm.py b/examples/run_pplm.py
index 9ddd42681e..57bed3890f 100644
--- a/examples/run_pplm.py
+++ b/examples/run_pplm.py
@@ -121,17 +121,17 @@ def perturb_past(
         accumulated_hidden=None,
         grad_norms=None,
         stepsize=0.01,
+        one_hot_bows_vectors=None,
         classifier=None,
         class_label=None,
-        one_hot_bows_vectors=None,
         loss_type=0,
         num_iterations=3,
-        kl_scale=0.01,
-        window_length=0,
         horizon_length=1,
+        window_length=0,
         decay=False,
         gamma=1.5,
-        device='cuda'
+        kl_scale=0.01,
+        device='cuda',
 ):
     # Generate inital perturbed past
     grad_accumulator = [
@@ -351,8 +351,7 @@ def get_classifier(
 
 
 def get_bag_of_words_indices(bag_of_words_ids_or_paths: List[str], tokenizer) -> \
-        List[
-            List[List[int]]]:
+        List[List[List[int]]]:
     bow_indices = []
     for id_or_path in bag_of_words_ids_or_paths:
         if id_or_path in BAG_OF_WORDS_ARCHIVE_MAP:
@@ -388,22 +387,22 @@ def full_text_generation(
         context=None,
         num_samples=1,
         device="cuda",
-        sample=False,
+        bag_of_words=None,
         discrim=None,
         class_label=None,
-        bag_of_words=None,
         length=100,
-        grad_length=10000,
         stepsize=0.02,
-        num_iterations=3,
         temperature=1.0,
-        gm_scale=0.9,
-        kl_scale=0.01,
         top_k=10,
-        window_length=0,
+        sample=False,
+        num_iterations=3,
+        grad_length=10000,
         horizon_length=1,
+        window_length=0,
         decay=False,
         gamma=1.5,
+        gm_scale=0.9,
+        kl_scale=0.01,
         **kwargs
 ):
     classifier, class_id = get_classifier(
@@ -454,24 +453,24 @@ def full_text_generation(
             tokenizer=tokenizer,
             context=context,
             device=device,
-            sample=sample,
             perturb=True,
             bow_indices=bow_indices,
             classifier=classifier,
             class_label=class_id,
             loss_type=loss_type,
             length=length,
-            grad_length=grad_length,
             stepsize=stepsize,
-            num_iterations=num_iterations,
             temperature=temperature,
-            gm_scale=gm_scale,
-            kl_scale=kl_scale,
             top_k=top_k,
-            window_length=window_length,
+            sample=sample,
+            num_iterations=num_iterations,
+            grad_length=grad_length,
             horizon_length=horizon_length,
+            window_length=window_length,
             decay=decay,
             gamma=gamma,
+            gm_scale=gm_scale,
+            kl_scale=kl_scale,
         )
         pert_gen_tok_texts.append(pert_gen_tok_text)
         if classifier is not None:
@@ -490,24 +489,24 @@ def generate_text_pplm(
         context=None,
         past=None,
         device="cuda",
-        sample=False,
         perturb=True,
+        bow_indices=None,
         classifier=None,
         class_label=None,
-        bow_indices=None,
         loss_type=0,
         length=100,
-        grad_length=10000,
         stepsize=0.02,
-        num_iterations=3,
         temperature=1.0,
-        gm_scale=0.9,
-        kl_scale=0.01,
         top_k=10,
-        window_length=0,
+        sample=False,
+        num_iterations=3,
+        grad_length=10000,
         horizon_length=1,
+        window_length=0,
         decay=False,
         gamma=1.5,
+        gm_scale=0.9,
+        kl_scale=0.01,
 ):
     output_so_far = (
         torch.tensor(context, device=device, dtype=torch.long).unsqueeze(0)
@@ -561,17 +560,17 @@ def generate_text_pplm(
                     accumulated_hidden=accumulated_hidden,
                     grad_norms=grad_norms,
                     stepsize=current_stepsize,
+                    one_hot_bows_vectors=one_hot_bows_vectors,
                     classifier=classifier,
                     class_label=class_label,
-                    one_hot_bows_vectors=one_hot_bows_vectors,
                     loss_type=loss_type,
                     num_iterations=num_iterations,
-                    kl_scale=kl_scale,
-                    window_length=window_length,
                     horizon_length=horizon_length,
+                    window_length=window_length,
                     decay=decay,
                     gamma=gamma,
-                    device=device
+                    kl_scale=kl_scale,
+                    device=device,
                 )
                 loss_in_time.append(loss_this_iter)
             else:
@@ -685,7 +684,7 @@ def run_pplm_example(
         pretrained_model = DISCRIMINATOR_MODELS_PARAMS[discrim][
             "pretrained_model"
         ]
-        print("discrim = {}, setting pretrained_model "
+        print("discrim = {}, pretrained_model set "
               "to discriminator's = {}".format(discrim, pretrained_model))
 
     # load pretrained model
@@ -810,6 +809,20 @@ if __name__ == '__main__':
         default="gpt2-medium",
         help="pretrained model name or path to local checkpoint",
     )
+    parser.add_argument(
+        "--cond_text", type=str, default="The lake",
+        help="Prefix texts to condition on"
+    )
+    parser.add_argument(
+        "--uncond", action="store_true",
+        help="Generate from end-of-text as prefix"
+    )
+    parser.add_argument(
+        "--num_samples",
+        type=int,
+        default=1,
+        help="Number of samples to generate from the modified latents",
+    )
     parser.add_argument(
         "--bag_of_words",
         "-B",
@@ -837,40 +850,16 @@ if __name__ == '__main__':
         default=-1,
         help="Class label used for the discriminator",
     )
-    parser.add_argument("--stepsize", type=float, default=0.02)
     parser.add_argument("--length", type=int, default=100)
-    parser.add_argument("--seed", type=int, default=0)
+    parser.add_argument("--stepsize", type=float, default=0.02)
     parser.add_argument("--temperature", type=float, default=1.0)
     parser.add_argument("--top_k", type=int, default=10)
-    parser.add_argument("--gm_scale", type=float, default=0.9)
-    parser.add_argument("--kl_scale", type=float, default=0.01)
-    parser.add_argument("--no_cuda", action="store_true", help="no cuda")
     parser.add_argument(
         "--sample", action="store_true",
         help="Generate from end-of-text as prefix"
     )
-    parser.add_argument(
-        "--uncond", action="store_true",
-        help="Generate from end-of-text as prefix"
-    )
-    parser.add_argument(
-        "--cond_text", type=str, default="The lake",
-        help="Prefix texts to condition on"
-    )
     parser.add_argument("--num_iterations", type=int, default=3)
     parser.add_argument("--grad_length", type=int, default=10000)
-    parser.add_argument(
-        "--num_samples",
-        type=int,
-        default=1,
-        help="Number of samples to generate from the modified latents",
-    )
-    parser.add_argument(
-        "--horizon_length",
-        type=int,
-        default=1,
-        help="Length of future to optimize over",
-    )
     parser.add_argument(
         "--window_length",
         type=int,
@@ -878,9 +867,19 @@ if __name__ == '__main__':
         help="Length of past which is being optimized; "
              "0 corresponds to infinite window length",
     )
+    parser.add_argument(
+        "--horizon_length",
+        type=int,
+        default=1,
+        help="Length of future to optimize over",
+    )
     parser.add_argument("--decay", action="store_true",
                         help="whether to decay or not")
     parser.add_argument("--gamma", type=float, default=1.5)
+    parser.add_argument("--gm_scale", type=float, default=0.9)
+    parser.add_argument("--kl_scale", type=float, default=0.01)
+    parser.add_argument("--seed", type=int, default=0)
+    parser.add_argument("--no_cuda", action="store_true", help="no cuda")
     parser.add_argument("--colorama", action="store_true",
                         help="colors keywords")
 

From a59fdd162703009e5774683ec65887a2ce419c1f Mon Sep 17 00:00:00 2001
From: Piero Molino <piero@uber.com>
Date: Sun, 1 Dec 2019 15:48:33 -0800
Subject: [PATCH 200/505] generate_text_pplm now works with batch_size > 1

---
 examples/run_pplm.py | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)

diff --git a/examples/run_pplm.py b/examples/run_pplm.py
index 57bed3890f..5e09427879 100644
--- a/examples/run_pplm.py
+++ b/examples/run_pplm.py
@@ -231,7 +231,8 @@ def perturb_past(
             prediction = classifier(new_accumulated_hidden /
                                     (curr_length + 1 + horizon_length))
 
-            label = torch.tensor([class_label], device=device,
+            label = torch.tensor(prediction.shape[0] * [class_label],
+                                 device=device,
                                  dtype=torch.long)
             discrim_loss = ce_loss(prediction, label)
             print(" pplm_discrim_loss:", discrim_loss.data.cpu().numpy())
@@ -508,11 +509,12 @@ def generate_text_pplm(
         gm_scale=0.9,
         kl_scale=0.01,
 ):
-    output_so_far = (
-        torch.tensor(context, device=device, dtype=torch.long).unsqueeze(0)
-        if context
-        else None
-    )
+    output_so_far = None
+    if context:
+        context_t = torch.tensor(context, device=device, dtype=torch.long)
+        while len(context_t.shape) < 2:
+            context_t = context_t.unsqueeze(0)
+        output_so_far = context_t
 
     # collect one hot vectors for bags of words
     one_hot_bows_vectors = build_bows_one_hot_vectors(bow_indices, tokenizer,

From 1efb2ae7fc4b1350b2bd46bd9898b9e5a31f8381 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Mon, 2 Dec 2019 16:01:57 -0500
Subject: [PATCH 201/505] [pplm] move scripts under examples/pplm/

---
 examples/{ => pplm}/run_pplm.py               | 0
 examples/{ => pplm}/run_pplm_discrim_train.py | 0
 2 files changed, 0 insertions(+), 0 deletions(-)
 rename examples/{ => pplm}/run_pplm.py (100%)
 rename examples/{ => pplm}/run_pplm_discrim_train.py (100%)

diff --git a/examples/run_pplm.py b/examples/pplm/run_pplm.py
similarity index 100%
rename from examples/run_pplm.py
rename to examples/pplm/run_pplm.py
diff --git a/examples/run_pplm_discrim_train.py b/examples/pplm/run_pplm_discrim_train.py
similarity index 100%
rename from examples/run_pplm_discrim_train.py
rename to examples/pplm/run_pplm_discrim_train.py

From 0cb2c9089074cb1bf158da701c9e9c027ed08258 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Mon, 2 Dec 2019 18:17:28 -0500
Subject: [PATCH 202/505] readme

Co-Authored-By: Rosanne Liu <mimosavvy@gmail.com>
---
 examples/pplm/README.md           |  48 ++++++++++++++++++++++++++++++
 examples/pplm/imgs/headfigure.png | Bin 0 -> 668261 bytes
 examples/pplm/imgs/wooly.png      | Bin 0 -> 679776 bytes
 3 files changed, 48 insertions(+)
 create mode 100644 examples/pplm/README.md
 create mode 100644 examples/pplm/imgs/headfigure.png
 create mode 100644 examples/pplm/imgs/wooly.png

diff --git a/examples/pplm/README.md b/examples/pplm/README.md
new file mode 100644
index 0000000000..6eb040a442
--- /dev/null
+++ b/examples/pplm/README.md
@@ -0,0 +1,48 @@
+# PPLM
+
+This folder contains the original code used to run the Plug and Play Language Model (PPLM).
+![header image](./imgs/headfigure.png)
+
+## Plug and Play Language Models: a Simple Approach to Steerable Text Generation
+Authors: [Sumanth Dathathri](https://dathath.github.io/), Andrea Madotto, Janice Lan, Jane Hung, Eric Frank, [Piero Molino](), [Jason Yosinski](http://yosinski.com/), and [Rosanne Liu](http://www.rosanneliu.com/)
+
+PPLM allows a user to flexibly plug in one or more tiny attribute models representing the desired steering objective into a large, unconditional LM. The method has the key property that it uses the LM _as is_---no training or fine-tuning is required---which enables researchers to leverage best-in-class LMs even if they do not have the extensive hardware required to train them.
+
+Paper link: 
+
+Blog link: https://eng.uber.com/pplm
+
+
+## Setup
+TODO
+
+## PPLM-BoW 
+
+### Example command for bag-of-words control
+```
+python run_pplm.py -B space --cond_text "The president" --length 100 --gamma 1.5 --num_iterations 3 --num_samples 1 --stepsize 0.01 --window_length 5 --kl_scale 0.01 --gm_scale 0.95
+```
+
+### Tuning hyperparameters for bag-of-words control
+1. Increase `--stepsize` to intensify topic control, and decrease its value to soften the control. `--stepsize 0` recovers the original uncontrolled GPT-2 model. 
+
+2. If the language being generated is repetitive (For e.g. "science science experiment experiment"), there are several options to consider: </br>
+	a) Reduce the `--stepsize` </br>
+	b) Increase `--kl_scale` (the KL-loss coefficient) or decrease `--gm_scale` (the gm-scaling term) </br>
+	c) Add `--grad-length xx` where xx is an (integer <= length, e.g. `--grad-length 30`).</br>
+
+
+## PPLM-Discrim
+### Example command for discriminator based sentiment control
+```
+python run_pplm.py -D sentiment --class_label 3 --cond_text "The lake" --length 10 --gamma 1.0 --num_iterations 10 --num_samples 1 --stepsize 0.03 --kl_scale 0.01 --gm_scale 0.95
+```
+
+### Tuning hyperparameters for discriminator control
+1. Increase `--stepsize` to intensify topic control, and decrease its value to soften the control. `--stepsize 0` recovers the original uncontrolled GPT-2 model. 
+
+2. Use `--class_label 3` for negative, and `--class_label 2` for positive
+
+### Example command for detoxificiation:
+python run_pplm.py -D toxicity --length 100 --num_iterations 10 --cond-text 'TH PEOPLEMan goddreams Blacks' --gamma 1.0 --num_samples 10 --stepsize 0.02
+
diff --git a/examples/pplm/imgs/headfigure.png b/examples/pplm/imgs/headfigure.png
new file mode 100644
index 0000000000000000000000000000000000000000..f4c11ad54d10b300e2051ef6ba2d209447bc92e4
GIT binary patch
literal 668261
zcmeFZXH=70*DeeQiiiRV(xeC*q$3?^fsKNKfRuoAq>41@Jrvy{T{gX=5Rfjt7dO2t
z2%!f-h!7wkkRT!CtnBxB&v?%`$8S8(-*0Sxp@U^|ueIi!*SxN4-qDYAH80b!(U6gm
zU4C%?t^pYt%^n#URRHyQ@Gl`ZsVT|G$e%c?t3Ps4*Hrg#_3$!${>0AVuCw=Z*C&CG
zq{zqwlI(133?7IG3_f{cV>37=EJ)+!Z}9qcoPkZ)VE2`m?X<SEG|@wSR0-tdDyn?6
zx7nCN+WNbx--)9mM-}*F8Xh!Qbt7zx_oid3?`GegE+nf;`Lb%yO}OydF!@r977GiD
zVYuu1k3T8s+s+MM7o<?28O%FpTcSS8uDbmjHJ<vh&iyx1Z$i%v8s8tkPnq@c_}t*k
zgWs>}Y1*pqP?5W*kFG}M^lJS<W=B!QcP^#KI3@fk>NIlr-RNQ5RVqyz0k}}W1|ND;
zfQ6MMOX_y`UpL;Jcyr5(u=b5}_hw<FZqM!#!5{aR83@?<SMjR37K#kEY}aXNBNu6D
zZ~0yS?K)rN9~luDf}-&eU#GmzpI^Rz{`}WVs>+}HoclFw0yQ{qoj*SlOeBUGgN@1`
z?cVeMi%9HZCJ>4K5|0iz&c8YrNXB&2VOM70GoFl0mF&S?HKPFX?WK#0zSE1h&zQza
z?>nbVa_g`KF!Iq{Of5CIFQ~p>z{!*<`O@kyyBNu$M@5fff*wWlCvF55-1nBAkS7up
zF~PgUmI+g|1)dhsQibm9LLan15O>hq9oTuKavuH^>oGR<Udw=*o?A7XjGTh<-}`|j
zwmZ4MmhsZf|6~!jYJnhkI3nRcTVoSPK@l+U+D-r8{A&Mt4$8gx3;*7;{P%;$>8p}q
zSZYLL{{3wLOT)>m@c+>QvIq$p>J<*&#(NC^{mK8<-qUxxaQ#OMxNmEalegRTHVXW^
zr}|(2Jw?FIx&Lgh`vthe{o`h@-}?`45Vxu-S#{`tv>?2jg^F@7fnnj@e|E6cD-Cr2
z))xQo?PVzpCtH@VHD~;{1LE(c<n7r1YypKGBYn4`hHEU(e-slSh*bWg1*-o9(SOD0
zKSA_Q5dAA#{}V+2RHA=n>whW{`9GEDzaY;)aE^lVA2|14Xz?F7NBvJv^iNOpuO986
zp6I_?u78Mye~5*DHQ)ab3;z%c|H_X45DWhh3;%y57P{N1{|DikAw{*@yq4gz<colt
zdvwSwd5wr}S`$x9%wRv)r)Y#LKAo~xGh00Wc*lFal8Kxm^rF(~Ue9J(dDXK#;nf_^
z31wW7Lz=1MObgFAyI$y)A^zD^FC}%Halz-OMZ(qXwShVPzRFbxvErS_t_ZA?(i-dL
z1HB*ZXHVA~G=9Ms^K05@Q0}Rmb)n#z2}CPv3O(3H-K61im$mi5q2hj*dzXyz@lW-W
zTv1W%Ts1s!BeMkaQ$wc5(tjy+`5+G$SJsl$=so)+3Z80W)lxndid&K+Z0dz=l|c?a
zsZTvCubhEaM|S&Mea&TB@*{&Qs&DZ_LT*CSlAn1H62FWv%{8|0>^tleht-r@QAG@0
zJs#9lnWX)+?b_v|vy(H9zhk&^IHji-#RUscIQn9kt=85t<CPzZeUMMsb_&*6m7?Hf
zlE*E2PoW^hpH;GQ3$AsCXSlAea;v^?G<V;G<%lP!oP77ebl|WSvs>lWc~wBa$(uTU
z?nR}mxKn>P<^ij7xd~S31mPEu%TT1%escclZ-kW#kf+tpI+veBaZF+tJdn_8+Y{=$
zssw#aj?~Vxqv{fN_vW`w0~^^Moj!d`CPngAOurHfY=SP0z9w4^TRLevJkBIerGIOm
zuxzW|ypY+vzw#m!=kjJfE-3GkSTlDYo&Qwjs3N9)6OSaqOq$MqPJ|U#hwiutO8(xE
zJ}7Xogn>1647*TFX>muIt!d6sznl%mbd*4m8q=8<I0LvjLiRp7x9v1)vwO7fzt75L
z*im|HdCm(p*?`D9MNh<$O%Byu%mJq;ShbV8-hGa4yW$7zPs%e~7*VdWc38+cnlCM*
zi0_Ms>~6LnG8E^?ZMD?$4nq^6Mzsn<A-c|dgZhN6^xLAl2s<8gF5myUByQ1m`C5X;
z%B)?YPTO{^cF=0{t!FaF+fI<S86W;ErHFtJGX8X}lUl+$&7WMVTKE+)t>`MnukHL9
z3K4^Ggff#3p{0<6u?EGY;M=X(QU2e4#n1Qi+dS;3(bvnTy7y+tM}*<9@zd{@e0@b>
z6<f5+@zFDn?c(ddzl-(LA>B9n=5K#$PsUoGuGTNOw>i{l=|!nrS%_LoP!v-}-I;h`
zefA?gnTrrLEe#>3kkVh_A|7jc8@-5^ofla0M}``9(h$KdnnGLQT00~C3#T}hGJfdD
z*(pB%e2-a%YbE-#WJ>0(=e-P?MNOSYl?^Za?JnnvJ1+aptYwPQBZcMW?32N|HAOS~
zuh9RI0{*v7Li!<MQyi#2Cmw}KxIMbC<DS8V8~0ui_qm8kZpB}a%<U^wbDPks8Cnt4
zD=}~L^NFXZl7n%@>bca(PUTd>4aYeX>rmf$uy<47Ca1WAW3QhvbAjKxZfc*wKc)>n
zfR&G)Rg-j7a$L5r$(pHia<8ms*d;tD$qRahw2VGz{=I-Jg2i&B0~6{IgWtC}zXzL_
z-1$Pi2r|<c1J7#6j2=VxLSmZpMcP~R11Qem?WElnR9+{WR}Qw%!S)vy{x+QAKzDOe
zG(?NC*ya4P!5h2ME{cE>dtA*YP6`n8xJecKb}}gyni`5WEqXcsTu;wGg7J3+e197?
zc}rzKnv@Sb8y~tQ^kud~k8;T(U{qqA$E)?Iyx`)42}l>x^p@|*0B5iW$MbIThBS!@
zrGp~lPs<>MzR5fJQZkb@ahKAz>(0sd!YLSjCw;g5K$lzU@k+sqE5iw1lPy|1sMzC&
zHZ~=1aFqOdD=)9<zW43=Dxcf#R^affXPTpsQV91y)3Hjs9(1r!r*WXPE;x6hZ>$p7
zvOcwHO>nOn!mBg~YY|VK_76&rxg|wK8)t3D!d{3FlQcOWG$CQZelrf~ma?MFE0J>(
zTSonRQ!PB<awYHFTQBDLOg?Z%<@)b#&EzmHw&Iq9OL!BNYF#+AKjJK}m4VCBeS>>1
z!M~li@A=GB<>`K6*u$pfAeTn-OXBU%=fQ=YYFu#7Ha2sHZ-lOZ+&tZ05`4NpP<QbD
zw>Q_CH;c0;tJ-$6@b{N!ugU-XmK>`LTVqFNsO&}16+w+n`9$kRB{$FT7>_*h)!-Ec
zbcJEg#D%{r3&N9aV08d;wqL*7iY{g{v8XwzE(se~&|T~J{TC&LdXzSTm7^P91W$i(
z;88!6@?fHFkEbJ@SaeIr6sdJ7;bB48*+ITGm)>I{DW4q|67HaKl8vnb*Do%7cZ#HD
z={)^0(&m^McJ^}$l$iFaASTLUtpe$gL$EmtausVvJCHQe^?pL7h5G&!S5Y^vl}nc*
z(s4K`cqT|mJsL`}AG){Oh8eF4u>Ya*?Txx4UZmD%pi2hih{hlfYWfU|h{*=!PP@F|
z-ImQEJ=l+mE2DDAkTGH+>k)Bi)GVC&S2AvQ5%uVIvgHR1pP%YQ7|1U2rCWG?``Gq5
zY4AC7roDo;eWs~R1eHR0$F18A6d<O#YDacyGlq4g&k1qCj=%b>r<(X3`Mmw(dX`I+
zLBgmpvsJ*8nWAz7DnO|7o32!c5nTK(0&w7Kvghvidn$Ms84CEpy|^zw-qM9Wt@6w$
z6>cS97fRygmYd7*oWtLFGWVI_c1J5O=<-AZ{QIllGyNYq=6|TF-Zem)Z3$STtKC|=
zVjx0AJ0Yl+^uqPr!9z%dMq!fe_aEbuAdp5S)C`%eLp)_OV1y%t9>_TUR$8k!BQML&
z7I@bM>w$e2WC&P^xKKN?<TIsHcO-O?J|h^ML~H}Xsx8yme>XLg)ur=`fj0cOWOR-~
zPig;d(00v0t?=f2o`e2_U)&nUqV8lz`)c1HEy~q_!V$4Hk+$t2n)e+3-Ov5jjjm>>
zu7I+kja%}!Mi`Y>HqPI;MyaoG-R{`<4-$ARwklwJ15(L|PAKXSd_!lxl@N(a*{E^)
zL7J3=se8<p(tVJ>fAjy9j5DMtDfh-%7d%B=7(TWoLe2vGOp}?qh0l4UnXgM<6QF9+
zo$-`lDu<BnUb7iL_&^Q5N3ELXj+qV`4b(gI+_$5s<*i3=vQIu6{Y(L0^Vx)_=M^Fr
zhJRsHh-lFPj)}#_zNp7Y12SvrJFKSi7?4<fz>fPL9SiCOYMFVD$+T@j=`(~PCTk>E
z2mbKQ5h5Sq1DRlyG2lmXzLJz;6#|5)W%p0p?Qhrb7kH2lI3w*He|7p(X5bU+mzAc<
zz2UCS!d1<-LH$yB5<kYR0>1uH`uV9(L)*vm>_xYfOP8z6hs~y@zyD@@inb<df%EJA
zP3*~uGyxMv3^uDo+a)Tez0*s!ES3!Szkxpn<udo|6b12@5Uc6E>O2iK*(u|4xaZn6
zo8TQBc=%Zp^)ynGGw`>*7B5{fPvz|7DykY4I>_l>{gV9?<r0kH6jfL)4lPey-@Q;W
za*}wWa0AgnN)S^HN5yUs+kBP=!fj#pfruSr>)VY>KJ&00<-{HNE%bnW{;dT(<|u&D
zmgqa0I57V!D^7Qg*lkQsXXK2B4+WgNk5pI-&F`CQ!^6K${#ezlH`G&>A$-?_Xdx`T
zbxy!DXYs^hrt;4Y>rb}I<%oUlm=KiW)ak(lva#rMNBhx?`P9W*IiE&vg!EjND}n&H
z3lgmRv-57sN2Pi?+M*3cK>!-3Tm2k1DF*dT^4{Kok1)d<-{nA*uP?e~BGlc?y=C38
zybIHb&sqk#HPNsF1AcMx^rjSrV*XKu6C7M{p^p7{kB8{$t~1wz_!AvUVy<UD>n1?C
zfvu}C!jl~Si=R(%0KpXvK4K(9a~}Zc^4^*FBfYELRPH`R3{;szkQBzVWGL>7wS4>2
z6tiG`3XV{(>*tuqeEB3#tjOf{fNQGEa_>G4pv}$}2w`13%=s)n6iuoGyCxGbY@H##
zi@D?bJ?D6tnfIT6Y$xCS;3`tmVG*+5x8vc|0o#zupKE~aKiNi*Lb76WtF!sXuyFRm
zUQ-vrEyhl;(T<W%lm4Fla>ClJ`A~oyKP$mzOr3IDb@YaPYLK}g{aU(dT3a9g#q2kY
zC~XFXZ%RDu=ZMcv141^jR6eS=t8KScc#M>r34TL-=?{e~KKvfEscoGFife5B*H`Q_
zu=j6ux%geO_F}w8z<)h`%Dy0%kaWz9Ujn%-`MRag=fnCE<;tJp<Ehzhjiv|I+Y3`~
zduy0ct=Ow{Dec=1;w}K^C!jz1cj1?Ouyamw6Y-?>#M?T-t!Y@OT$)%3N-e26yI1*}
zoSQ1*IpJ6kIkN;kTV126g*ES|=ip#<JVRr{Mnvi!#(3+wHZgY{_c`zPF>?iouC}w`
zHG-qD2s4*TY}p1niZe)n@a-1l0qNdMfiym=6xJ(oZGk@_9iW58I|kjPN-m2GIjCQW
zyyVxQU-lY(A4SmnvaXdw=S8RZa_IL<qAA%9E3Hx!N_A-R)W;}`765x=K0d9<omc)#
z$zLaUNZV>s=|aPN;=$3|D9$Vxt_=wXeafA`5s8bgYrdu59G_e55P#(Z7h?Pl%Yb%;
zc|ALh(ev_aZ%vDlH8y%;-a!tk$fRQnIu<q}-W9Tg)xOq$&dzh!O&mOE@WA+0L@_BU
z#$f$Fxo4D5nDM3-_*Eo`os;^WD~d%_*B=*nW4G}n?73-<JTdmpHFxXpxR-NlRe_WE
zmoI*&DZd_%a2+_acG#Zsv}n-Ex#q=}5IY{6mmxNe?n$^$&*^!iU)gV)w}!*Li-6rw
zcA6c4(GyqCf(AJJb5br`T0q>v5|79C!DqI%&@Bd{a>p<?By>VBufOiBwKCy8RtJu`
z6raMnd_5LBub5c+Lh%cx)pq&gZUY(I3DNShTWQT27D1cvF|?*@)3N(Z>u%F79Mt2l
z8%MlVg#Aa4b6i&0t|A81>f-wS|B{dLNo%%_ae+ET2jk_d?ArPzSnvUh$-TI|$4Gh9
zHaN90v^lCBjzi^m>nd!$FU&$}+jiK6#p8QsBsjv>lvvS<_oELq7KQ!N)G!|v7o5Yw
z023gCJHWOEa3C6*JLAiGJsEMVxqMY$4ey-jhR{VH=vid1S_$lqtj#kbKiWc~L*+W1
zUWJAX#I3<Vi0oR#UqlYT5P=qFYh1*^ab_e!d%!v8#p<#C|EtXX-E*-1_K3i(urtyF
zQS9_Yy(;%1v=OH+=%PE3E&JYWCpXaggp50t_*}FF)BCyi6i;|K+!0;GgRnC__q4a9
zb_n4sDx5~_;W~TX*jOGO(GRfv)O(oHh%RytwKag!VB(fDXYc7OjkIKg(i<1jd?=re
z`feD|xi0GAY!OB{>Jft#9S~QUy9TS&nTl?^+=0rr_eViaziKcYCg(OzonH0`P%oq*
z%akF$w=#nKkS!zos&TXyRtc*=ff800kE@9%)fz|Yp4OJ7ux>Y_2!T-1@`lkFq)FGw
zbVE8TDkfP^rKJX>WNgRiY?t4>b4e$zS}33q<Ipfud4wysCOq5mt$>BQ-|!l|8Bd}u
zFzdz<PVrj<hw~$BM=znO>WvfAiP%u_G8#5#BXaA8#L_h1(@Kgj>=!tE9u?B4qJvx3
zGTzbzJ$0=clZ_Y35JgXSiop?3RiCvKLtR!)|6s8-sr;W05ei0hFGpqdn3@+566RFb
zWUk1sgm+G8?p(T@pJU0?Jf~v1gMb!y_k(n~qroJd;U?EqG2YtDx!Z}~jTia^$9gT?
zs-QS(OS7cfXwaK{Pw|uu6EL92HvNLZa>@@|a@0v9?H00SnE%X3ExoP*8g%VdW#9#Z
zIaO$S7FRPYW34n)x$J_PBU=52(@uVqRg<;&n*967wA!hL;=}{*rQk{-mbw}w3@=j_
zj>&UtoOdUhQ+0#Cl+De&Maq3fU}e{iK*QEKm(>DlA-kuw8C8Y9&cYp!_UgzMw*2GX
zTWnCj{<(SQuxqFokSAO9sRU=9)djsN(^s~cTJC*p#RZ^Z!%4dL>sDWcrQE{ARz#a?
zn@1?x+$p=0X57Q*<A*tC?~ypS!+At!qdrx__{{hRk5$eOK>~&PRK|#1?eZ|_8@oap
z8;-_H=O(yt%$1$A2e>XGzRTt4ElsD#DJtyDX$Rxjq5`FBlkWVHTKz@Eq{+X6=H+OW
z2dagq)6kgil0(VZQ7Ic$F#|F=D0Q|P-{RK^s-U|Oo%GQ&NxbQz3+jYadz?x<s^eU2
z+?NwWx9xV4$AUsZ59L&nh+(U#)P&mR4;`Tgc@u9;^PD^@33^||F#aM$Guqpjse9@5
zfyUC+Db!3CvUQ6*y){;v<9`6}|D%_<twWh-MBpx<&v?XL+jQ?b+(OWycT8tN&oY;`
z{Z_7zffVlqN(I+qnpw?nreV(zGb#@y+q#>qdmL!*7W+^!X?+89@S83HhkK_V>0obo
z>UK!X*Tr<1s*7H`NtWLMWdy6cv>$kM&QI#Viq^%l<gR1yVA=_UcmVO`&wrjZ_~{Td
zRVAA+z|>%s<AUu+E*%UeTxZQjrw#f*hua@F#kXdf+J(i$1w&WtWMgOVj>HnsfZuW3
zwNCO?kSv=PY}v7P>cl#4RoMAncHJc=O;^xd$r5d|4mR(w()6nv3}G|9$m_lKP!vk$
zEhXi@9qXG|iuH}(wtyBq_71C}xI7QZnnWb+i%ZgJZ1f4ycVs&Ff4XwduU&|{JTq0i
zHctm(KxHj0#e<S0rk>o$gp}K4InMmp{6HX)d5&p)gR(4h)E7i0ECTGA;_Z~M6Z+{>
zdXSDl%jk+|!VO}QL^Vx@yQA9nXocdI%6Rk+l+F#F81ju*aQthBR!)BTO0!k6G)B1G
zX2pm~6aG}L_5^3nV;5r2n}}~6-<~BjLbFs7(9IW4<|bgq!Oe%wdZnBd_ZQrnEV1pJ
z2-H(4jVrIW6MVIp?&g$T2$}@Mh{VEuGT)8d=&8HsH@4oE7%5Yxjk#Kym9R{9(%kiM
z{0p-BnpZ3N2+GBKuD8R(K0RSOn~*j*&XH-%tLSa7n`@|T?^qp)%*OFT_uRmJzLWU8
zj6xyXG*C0ND#4|n)V`Ok%sIYVS^VMcyC2$2)2atmbw&($60y7fU2U16ZTT)vIk(o@
zY%GFHT%s{N5?K$NoI?@cE4aReE=t|-U-Y!B(?~-$PjLB1dwW}C+lIMMs9H5~CABDF
zYmU#@Gh2U2V1}!T13z?!Pl&8`!C{YcYM_&`4XPqqbB_mBpBR?Q_v&R_!3c{x!ObJs
zysY-^^p%I(uIRdaf~z;n==e!1hW8CGEKdBgwP{l>we<0)D3z45hkrDjQ{QS-9`tE(
z<51t|fApQ<HG5Bo-xZ9h)17_0K+2VS>^&*eG%;MQyaI*ppR}C=B|<z-1Gv;=uTk;d
zV_&0jC-HPs3&=Ap<41*_wAn76*J7i#ENjujAJ8o7H70YTL1O%`LgQ`1g+`XU2R!f;
zu+ufJ5|;BSuLSa(*IlgfW+wpjEBt8(p$O!ibs!RO`ChMtq`bVdjWV}1=F?dsgA3zj
zpIz}(o_wYG>2Yf2ZRng0o`&De+EwW>{Q?G{Yxio-1+@GH<IJIK@@Fq7%TpBCPz8@m
zKCnWgyFUo^Sa@d~s4c`jnY0G2K*G0nm8V>e^CA26x)3@@pKpcHDyPT6n4Cvm^=h8`
ztQRO$GUsIN0uJ=ksF&F`n$G}@J|FVtE#YQ!Gpby}hDCr{E@8?eWQ~37g0)6V-c)5J
zH7piCCKP+nZ5bOr959PM$93G@?baLTNykKoo^FZCdaAa+bd%GI?VkBHQJ2O10SONB
z4Nq&S7`xVK;!`=!88%1jrwvSDB(77AOQft;u*n1g<3K9&Jd&HKgYrBNT#J_*KaXuJ
z-aL}qMe@kE=SPJFHOOsEl*M66*`|3HT$SP!Vy-U8;lhnW5Q&(AH6hRCS?>wuY^33;
zTHF)u5T^uuDhbl}$U4A&!L6y<T)7U++!s6+^{dPT--}D4MO(kWE3`1Gt0Xl?bF#b(
z9;bV3kfJam9yy_VEp__MMk!lb`*C0hy&8QI>#az6yePx_qkg$g&%Z{iNe$S%ZLDb4
zToy@N$crs~C9KELd}o@ksgn1FqkTav?R`>@Z^E2z>X_*$+e7_*7s}ISKzT^}XNSmq
zyhz^khb=+acd^uGvsg##)q%-e|K%1|0R}aA$M7x4&ml%rvRguTvZFVgiAs{JQ4ZBV
zKq9_<Ph(SI-B~Z#%*=1j!LVLYKbmaIt!YUXU_49ZTO58I2<tEc<3fRv?Lyx^``ms|
z>f>ta;rQ?f>QgJw4hHEg)4qzfX@4U18_r_zH6)I974+NqTdoakRvj(5YvivK_^|7P
zj(4DZ2~&fwSTl9eDT7<GpC#s(b#oxgM^*j}COSXS)RVC*2#e@7>By|rClR2jto=!^
z<=$#+X<r^+@KJ$|seXeM>mec>8P&P>hxnIgWo++M`i|V*dBcM3JTLzQr5<4=^9e2Q
z-y7(~$S2vU;Wt4oX&iK^_XRTpt!a}I+H~TXWB-^A7vgPt=OJ9MHcQvYmdo60h|eXL
zclG2dB4W@65_1(7w4)aEqA{?a)k#`vb>K9~A(*jH>~2?8%SM)3n{*{A);lCe%A<9&
zIO$M|0(JiKSsVj}E#nIX)azpvMqJOeTxO4sqnVA9%XqSt`_DDL+^4h{P&Qv0t<wt}
zHawvjjyuo`3H$C#w0457Dvo9^9~MV&-raVENG1)mh+yLf>q9}e=%YJ{!#6iJT)^X&
zx_=v|M`n8S&9v?MO^hMSgmatLmhCP0wn1VXI6SHj>zLtLBeGpLcdfZt5?_D`ctc{v
zF^@4j{M#89e+j3Zd2eItH=`3Cx<nW}vZnl^t05>V4iku_)g&FaYixXX<Dd<@T6TFw
ztrkh<%Z3zaY80_JD#Gl?s3`7Q(z>wDg>I!BKM{vn#BYbeYq9e#CDhdMB1!w|u;A#k
zoS?1JF|K_t$nRnMvDOd8gFeS*LByIS<f;%!Nw?G81YkMw;hWg-TEI$(?{wgn$5u*!
z08lj7M}4(p@2^YQQ?P?<<fxw_MKSNeGum4gxAy9%@6(pEnWk%-`kMg4*ioZ`1*=Rk
zGOvBw%n1=2fIXsFeJ%1{BFC4Kn4(4bXMa$WR)x0M8`rOYFpoQ$#XPa$%@8;FOBOb}
zb5!VQ-x$As55g;7#-zkWeo^AfbIOjpj;n)Si=CBbBe6T|5vZg-X%pK#7PHRKWxyqh
z-+LbD6iw^J@MY4}*xapQyv!8Cd#n=arIq;y@19B|w{5^4HRYxVi7!_;J6`*nzvuY#
zbl#=f?R-E^z-n~pj(P3~>i}L}m*+k=^Lqxo-Zq*+gaPjxTwuTLUBkiqdjCVul@Ukb
z*t0MaGIosLR=Um~ykyI+*`+;)<N0*6*0dki{1=U|^toi-I&-ilbS>uw-ZKZHg?Fw{
zx2K3#{W|zfX_{|bJV~d{+r0mr7(*O;#6eBM4&SS<391AZ1w|7cdGS~7VpLS94c1~Y
z$?vI=PVbgv5EgZr9P$Zx>SB+bWY68`QAARWT-7xOvKtFEMz*1=ESOt1Z5@^J{@)U^
zsn5qg86%MbjEI#%HZH`%Y(CE;09D{*?ehNp(lvS;T1p{Pfmcpv2UOC=IAK5<@v{tj
zH%3x*TZ*H~Vnr%)@fM5O-y#PAXOX}0LbQcyCiFIUtpF}L^UnB_7MBpp)xL(qSM2VC
zz9pv2h`9i=TaC^+=Xh<Dv|OjCPvR@kkdx1ewYYQp1&<al^zL6nWUFcCgn1#>oKy}n
z$M9n;*OhnWc)hXf>DJ3WiG<&3<u<ey?Ru46ck^y)c)t$r2brzu&SFpwWKD@>z`<Su
zvpD`4wmsoVC%YwrYUvzM_6lg!q7>bbl!(%~!JZ5EK~CMVM29Z=f79A#X6S9{XLFI}
zdinZ5FXtLh&pP1L-c&WO#w@yywH_zXPt9CxbdkA=GxQV{1*tM%*7n6^*{AMpw_`$8
zb`1;lJjJi#@8;MkXe+rgMih}4Yi};hX~-80ubnH3utQ3o8$E%!BQa4Gt4i6zo|gwx
z1K#uR>Ui;S{vH!5P(|<@?UgC=B?bX$Saw3#vnm0fRhX!v5Nx5@R)g$2lX=KL)f`59
zFg_VH$iMr<4w8U2yMbT`u~z7}W#<H~<CtBI6pP|6Ztau_x3e!kqgWN6Zs})w&~8A}
znHc<HkXC6p?kE?n@Etz`<3xWlT~)Nry>L?z<<gPp9lZOovso{W;Az39q4rv24t^Ey
zPRZB>bY+iam-=0Y)z<3^VhKNIVKvsM;jd9#d8>Im*dB{$j7lz!{j>3|)$Oyyw?42K
z;)VZ@PM`k~)!0M)kG+WmkHwa+LYBl`;$g&8E5_gbbg@chAjk;PtA~d9`&FHMtkK_@
znG1vR*NhKo{^%U8&9by|f|PM#*^ocS{n?;o_a@ejAt$nBT&=Zk#^lDzC^zDV;jA<_
z-V-Plm2?@|q6U-Fq>47ySLTLmVSNI$K(FxR=50l!y@|1!MBkCioC%~?u53H+n%_F9
zkuXw&_F>D&+Al6V)p<nqyL4GWzpeT7qmp=`r^PaX`yMxfYc7|oes(pVrmJh1cNQ(6
z`E9t-@G^ces|bq};=Z|H++5wdRTd9A2S0P@2d_e!^Foa#RNV;o@{^=H3*#k#T$^{R
z=dXxlt##{DFedkp7uu_kah~XcKDt*(^F_Fk3BN2q4DN6GjyjX7{MXfTE@It5zh8Nz
z&8-{=|NF>5cIc;vjk&igBio)eNDW7q(dPD@8O^<7hZ-->$Ch*YJvAbC1JAHhGVd1u
z>@Oiz_-y{n9m%s->bVC2sVz<V%4A7yQdNSWx@cBbnW#bMWXr<;yojofe8UFqM{l2c
zE)zhfrt*Zn@}g~i1w~dCqQ@^0i-10V=<*{kOdBzXZCpLNqPYF(J;k@6Mqhvv<;tSM
zYSAoiMsqg7<H(;+nC@q|l@I?ucW3i4(&Xdn94r^;`VDw}Dt(8A?>^PFtFleLNAyO8
zG#V~eg{vp?3UUKKB}~fV2Kun&WC#tWcI0~sZtuErV-u>Rqs8JHJ*3})TiuvpkRg>0
z6aI@CMrh@OK2-~@eW09k^r?|rPG4i$XhmJ<^;gsCf^c^kba_-R8?^h>^`PN|F)R#_
zj7K?Pk}7e8_71vAgr~)!VHx+ft70YAEUbL=EMBOf)oBWP+oO<1(!+?CjB%lYlM+3r
zfsabJ3UV%_83$)FV3zuzDKu>xE#I{Z#H4tk4{{5}Ic<d)L;bRf?{g~9%-``_x=c#B
zZnqk(7wJqswlpln;t4{tO|w=vq@a5zFL(NCkn>?dgr(;_m)WP)3zZjQF|BaS<=g=u
z5#cJC>2nT!vTJSk7uD0T(Z`);C?d@}>lWW~Z;S~Q%n?;dQ4~u^pj!=gch7`AdRD2D
z&Q{HNf2h%1S>REY!?z=7|B)+~ZK$213Ef}Ps{wiMjp+W8ExT|z#;p(c9ac324{Zs$
z@A|IlIOku6rvJ;zkfBDAbmTtIOx|wT79B+84a`1Nycmnqr=|`KLT7u^Ctq(z2Cr(?
zw!()Wrn}X-(?3qVT*-2m4f1aKGNbbm0Oi9)SvynqR2ofbmM3jMKybuQNmke1mnE55
zCbun*7y=AVA`~L;pi`(LW5WlHx#o!)1sbUzYjvgSbgo52?CjKcRn_v2$+}g`LC=r5
zD*D2dJUp6)EfyaGF0{@Knz`|eD!RUjxte#cdYC88t!_AF`cB|JVZ;hs>&nhhKdmQ7
z_|Ao8pq|pRaT1Zk#tI2PZS-JJHNJKHHvbG&%pn~vH*Vg-T_{xUwMG(QMav#DN4QtR
zj>_w<?1%TrYg4{LMVi{OdjMMZoLjQc!Q=g}vD6`z*P0<(FBHhz37kZRLH!`rx5Iy#
zl;e9S_xjaEc2M0j=-D*r#=O}GCg=U|@9Az*^9kK)3J&kcmW8}s3x|Uqmy~Wu*Hjy^
z3d9o1YW-*2UEJwbDkZJuz<20;XOgNWgP#NX0Q4}ipn-aLr^G1p89Q|2n{76<7-=@9
z!`RLKu^T!wZsxWWuofRr{plHpCcKM5+6u=h%jquIz^dl7CbmYq5(?nd;_Kj|R=(Yf
z+8e$T%KO|_i^1#jrL}5uWut?>ZBj0+B<(%`vN-YLWgXj%kFokxd&YAPb;kR9rd>Gc
z%IU46Zj;=V0fSF=c-bZ3ka|XCuaO_L)YI7X6A1h5Gv}y3J5r7`PAr@Sy^AuiJu9VN
z5nbxiK(3vtJT2A8A2RFhxw$3)J-^bw8Y`Z+qrO;;yLO7)@=)5I@6=ha9yqGn4Xmse
zzqI1^01ssH|0ff*K^t)k3gH)^45B&xN*6W=<Lp@)_)t(NCgJ*=7Cu*Q6j5AC^V`@J
zsAX0)P1+cYv8QN8ki_^KGD*+MIMcp0cMy0uo#<(qe^mDRiC@<tT-@PSDNJB|M$gZ|
z*}N&uoQk`M)P5ZK{nZ7%US{N=IjYO7K50D1R+rykVw^b=)59}xyAY0DSoj`dNu?wk
zwYOW5h?}OX+P6{PEv#x4axpQnGy+<k_~1;ske&RB&`KkEO{hQ$3VB^36!0Eh7cFEv
zh@GTwN9Bi|-K~>@R7ctvrs{uinV<F79(ZeG+xg$`0x(4MLhse<vx&*Rx{~<zlX;II
zYGY!$Z?BgmJb7`!h0b~U2`q2-8a`s}ksG*;$#o!EsC=X9JjTqaOp94s<;MpA%dt}x
z`Ao6=)=ntXb|%M`@~eb|CpC6mn@vXKOr<oBM#hb)AVn;G!3wD?{O!ZP&zq!|sdlb*
z-cXkq_%NDmS~iNF+khOtrIU$@9*4=RIixx?1SG_-dCwY9eTIL>0VkPNTj>^gijl)l
z?aG_gicS@^KmCmZ^BHCV&1{e=BMW7HCzXGOEE7r&z}6+g4h!c_gu<1fYJtwVzB5BG
ztFpKQD+@@ONdDXd+xu9dV;Q{Z(y+WoZpe?JQkwZHnGr4F$`o5725M^^N2naGSa7K|
z&F|~L0&YnG9g2??%4j6)+&EI$oi6KcVZ>`LUB$hG6RO#U3XG*{!b~Dw%sJBm0=pjA
zjT|6oC@b=2*~<9#Z~K-7Y!80J&-qlxp$$Vk&?26TMsqXx&JyPaL#mSEt(&<s3v*`)
zt851GCvS8WHj9kn870jZ;BsJ-%_qFAMopBd-&|4nUciWR@YPymPVZFIMXV`|n*x@5
zZaOyjv)b@wxy3hS7gX9D<JX#L40g)$<6?lyBm$*_+h-?u`yO$tVe70?_rsX>#ZAgS
z#~Xx1{nB(rP9Yi*)OEX|80%lfC_p7>te#y${X2_qrod}yiXxuD+FiP$%CpqsMAd=l
zVUoAY2lTK$Ix{ppW^N7Btt#b)`VKdS&r|d!er@6EMc3c?E)BOCGY)3B-v-*7rsbJ4
z^UG6Nm&Gb1ql%*&k4Vmz1o_q1+qFFOt0aC5Tc?n^Z|9Z@Ep?gfI$am<3;#MEtxX^%
z5|0zdAH-tTF02V9dteWMDJ8jdcr=?(q3a+c{oJqi*(}O6)X71H@_Cu$nxAnAb;#@P
z;Z{Bhld_xd+V~T>Hh8=3J{{brOg3Tt(mEd>L%o<)#fPo2V3mG!lamLirJ}7_ajJdP
z*Evn1-7<wiQVf=ODk&H0WJxKDqDZl8ZN+;zWhd{SmFk5)E0YYM`*PR@XoY$pr1*`o
zqt=ILf4t^?3&g;7mln^C5KJ&~0gBY?GhhCZw*9Se){$iXE319sw19!O01>#PIchLf
zsL-#<%rZ)q@(Z2U_}O$dES021`qpE^kp0S*sr%CSxx1-Q$N3HAC(f<vbjtOI`jscM
ziTFSpKDH&mA_X^EI&A9Jh+_@4==9?8RyFx5bW(1V9#?`?;t7H#JWs2J9W(Jy8(SxP
z)PPbBiy>Q9(^g`v0sbNXF~DgNJAS^=MqH38c;~Z}7$scxnf|=h3zCV6B?A)UGUx=s
z&u#mM8$_+$tg}Mi;jw8s{GeYt)~p$Ofyc{1feQhTx~MDv^`()+y@^WQiB`Y|1kX*Z
zDL@jg4p?tS%r_H5(B?>m=W<wN7}1tKIE0P>xcEEI7aV_SV%!h`Ose+`HYcS)kZKQ`
z?NINI(i(~V$y;YC^UY<c*fF~*W$bCodRn$y>$sf%lCR0UQ+}vIc{YX@bi4BDxy_GE
zhqn#@?b&#H;G2<VV-3mnxe!2k<L>QE?sY&{`<1;Cofucp=btrL(V?75k^JevoKhNR
zGqlr$R5^GseBQ-hmj&OalewLrq$V@d;vT&{#&C?tq;egTK|F+$HMhE@7srNAD8PM0
zu-@C6{cQ&H&h`yu3+V=7ybGpP-zavyr*!UFZeGC8GM^XaI$gPt9dYjo8!BP2lby9k
z8-H9dA~?p3N_gz(l(s}&#s1q~SLhvvhis@biy^8ER581zrM(&ZZ{~i7qKhg&k`nFP
z?qq)a4t&pQvMi7)7Y0?ChtZ=WEs`;A*nHy8{FY*u>3eCt7SRNgP0>A)zrl}xempHf
z89O^(wt=T!<dA)=#QAET1YUPEADL3RUoku1f_RPgFYQ&gWB4-dAxOoLzUzi}3v1Qk
zqC4vqpnQ=sIG7WxDm<U{^kpKo<V3s@{SJrss6?M1`VcVt$E21!S6Xs>y!R?y__qpe
zUe&|8gk;~#(GQWpi&cGgq)^4%+~gC7eO4zu;iFQu_OWHL=MNwtc_8T~`d;5^AzP+)
z=d^#BYBd9&;OPl+znjq<#^HB2Ct$Y5r!$cuoaa*gFWqzW)Zm6-U*u89EeF`O;&<EX
zrVf)Gq!(1*)-p=7Bx^o5qg1?Npru}=o1!cSdmP8br#SMbw%8q?PY>!_w#v%=WS2s$
zhj+7b11M8F79j_j6O2){m37}I6*eer%y%q3`<OPr80xuo<Q~aUEOmku<WEI2o>h>@
zoODmYgy#Tr7yfNp@41TzucYF=;!5i95I{W`-0)_Yy#d)utJTo^(lmntRx54vx@#Ev
z{8$?HBP@YnK<`I+OY_t82e|aNAlji|V3)YEa9T8KJ=`7uh;62*oxJzF48Mg9-i(rt
zG7Td+6h?m_SRX3tv-gB>&TaCr+>Crh>ggCi4kpF#9^?~4vcBJpRu~bb`teOt5#j9d
zW$JFhfS9^tY^+<xC9*3Bi?l`kik<|;-P!ieK3|o{K;A(~&Kdc=?sHM@u-ZmbdouCd
zHndL=U=0R!(=M$H>B=*kXpmtF;D?XyQ9espwKxI<yF&0-oGZ}@6j0Trgf$6l6HJ>{
zD*w{6IKS0a#LsTsT&#BCy<X*>oX1*puxV37vwsjUy=3Mdk_^L*ozToSOxy&|CoiDJ
zyX(mAAp;8d()Cq33u&)CRe2uuDrX<GqRv6zk~Fh$RiySlyU;NNxB_)O0(#hAx!!>f
zwZ3Wx*oou4N6Yq5<*V~c)=sOQ9EB4pwG1S-+au}A3I1<)4#r3K2Kmej7!u>^@(u@g
znF0dFylek<^gQ}hiSv=5GW}6D^0e673v=-oZ)}NZF#;Aai`K8`DnG&d5*rw~M>XS;
zyi)7H9hZ_TUAu8v(%!GU23YSZj=UPXBz|+R{&R$uaS8*h4d*WNF11t)&SYKs5=UFT
zCIU*{puEiJ-rJfh@VNR8$+!3X)&@tvnDX#W^|N~{nks+78`yZ}a>q>8d<J*Psf6UK
za4i1J3|QyB?%nagpbjFiEMyP@{$Mg5obfll(w4wLpK&QCX4DLRvq}+v`!gPpq|Apb
z2h(#2*BT(K`4C*o#F#Am$M*~&!GN-Lz!A0|PE|-Yx>Z<Hs)VeZ<9d3=evdi*v2C=g
z)D&O190j{J{z>Cn<}fwIauB*wD!-_+Zr*`?Q=c`nXKj&*yn~^g9sl(M{IH<=nx2w`
z(G)QBe|>b2ej;Oug}*g+|I>-{aGss6^e%UhQ*`%|0rz86{k4w7;8eH9Iex{t0$#S1
zS{WziNn=(tH0b$TAIKNkV64L;*SB3JuF%8Jhk$@mRGfHN8MlIZGNC5SRNIJa#n4t$
z3yh|!;Rxf6JONFIEBd)n0}2b-S1@byOv_0v1WQbiNlet)s+G;Gq$z@+mWZ|@Z-2pw
zfZLn}6nH44PoC2T)*^BKQLm*=tE6cK6JTX7=4h;9>{i_V&QJ*x{m@Mzs4+B{s#dMj
z()i^U_Cw%!HwRZ?bQzYsj0gV1)0M|Gzq1z`CPfHI-dP|1pe}-X)f>NF0env_eKKgC
zOWVNaI(ZaGsSkXdoR&u~jC$#UTf&FChy=_FX}LWmxLXHERCKu<)hYLc)92)kxobb4
z?L|RGt47+8gk4txcPgB0Spf?VM-5TCa@jBOplxWWa)se5Nmysz4ACZxrj1UuU2#Xe
z2cWC}ww8^#SV{d^VO)7WYA&Htw*RNxE2>1*bc^V2LayI@Y37k9m`>2yUpxCip4hTJ
zTGYy-)heCiqjV_2F)UE~U?3=kQ{WDMyRM|?+03qbPMB7N;J%GUh{{tzhfkFY%K#0f
z<r<NBq&XNn*jtKR#;Lfx`zsbWs-*A_7TNDM&g-X6(&ONvp6}MLw03@Wr_Ie1hSP4h
zI4h!8BT`%5-o5?(1DXvnXR*oDjR%YteDZ}}A<as>O6vF%DVcCNl#;8ZYsuuf+#W03
z1?8$>hnF0O&0<u0SI}~E2h|NTW^)$;*A#+)y#Qerq7<?RVxie(KH&H(Q+2>S6N0^w
z1t|TSGv*#+vdY7Dojz_c48r|{Zg)Qi=+$32JPm=%F$Jf3p}0X6gtz3%x@5%VHc9w;
z-`;k0@DlpG`x!`<_eq%>aVpX4ZqF%_YJgoD6f>o{BG52ht};_yw+JNFTxc>@CGxr$
za8p-dc^`YvZnu4YYM(HYYB1AGtu#hZ<(jL0H8wo6oH$9hd+l<~4Yw4Eyr`k>h2^H!
z>K;Lgw|2zNpBveRrz|(r=zt}s+Q3<Pix$GgTJBBmCgDMuix7lY7&8<o3}0Q4V6Lz*
zBkGF<e`*A}NA(w%?~5P`Z-(xA=ewN69`t~oD6bK?dKr79(<xRj?2PziiBl+$M5GOs
z5a7C#sk_;a-?Vx)MzdhnWk7}jdfQBvz1bP*HckQBI&YN?cJ>ohLM@+HvF0%wK4H-P
zWsy?=@;RYlRC0u|di~m%IZZd0^qZ>s_DLD#@sbxQBt!R`xvYb`*OviwHgu|IQv2eq
zF6c0{>$Iee6YSvRe2vd*<{D28kQes1&dT<C*NRfc1kOFs;{C(^R^AJtwKmlp-qp7?
z&9EF`Zya-ZQ0*o&x<%)0fl6aXdaj<z-_INxI3(%$1hGS4$b?J>&i?q8tf$+8Y2ELT
zs#QoVOsUdgxU01No@R$2ZkcVHHTVd!9jb$`ZRh{L2&I2r2FK#=#eMDP4Q`zBhXr2*
zP6ZYwip}?As<jUwVQVnQ%$liwx)S)dW^$@SJtHG4dEF(^`cK_y@a`2WHIw}pKR{S#
zTW%a7tn;mo%1j&}>_tC-rFAQ-(xkYgDmtn{fXby493|&aVhjnEG)7GX^h?zq!CgZY
zw?FjUKR2K~mv^f+aVK>ogHde>?mJxdG;tWrF-hH%ZFH&bXk^V{SNoonByy&Fyz*j9
zqz{J-3p?;lbZk|L;VME?il-!GwvrD6nb_LEy-gE4GFz=-dVVRPa9h30?cViOTsum6
zlO1{7|FTX)V=Gx#q4cLXe*!>R8}SFLBw{Lw7OMQ@mWI3<<*R{C1~p%XW_ohQeR1qj
z0K`;Hk=Npns$w4YB>^ol_TYN<0iscQz$5NP(^D6>x>+kQMRW8RZ!%n9j&QPwQ1iu0
z85XnizpqW1h5_3v%jQ;kWD$?N;&@?86g<e+W;Rq5;in{De0ZJ3(Hgzvh0$B<EH~m2
z-m_013i|$}7H?TX7;HZSmM+mgTnQEpYfp^w(ZYH9&hI*9QCDDaO1Jg#a5VPr!Pwyh
zNqnp)je7*_m38#`ryu7@f0h79bfA7h7JVevwk?T1vhQL+GzfNb^>;E4s5uE9z4cc3
zb&bDcZc}w}CG_x1KKyX7N8MxW=S*2G{}>;6J44E~h{zVnU9#mT=gzbuK6cEk4GUMh
zp3UZ%6F)?UC7Y&M1#J$w!WtX3RSMbus!#d)o^jiZW@QQFN=G#AgNN2fwxrD_6T!fv
z`TU>;#ED2G8wt^rkc>B(W0E*^50`m0B1u-RkU@VyKc0Uj4%?5_<b1C1-LHa5BKlhB
zy0Eo#o}KQU^_0%Es~^QlGkpiv_OJ!OflUJVp56!wb%c~Xb~-}kR>I25!NKdSzIFjW
zsJAns)ml4@gN?|Y;Nc$wY`8gvGUG;Z@S~Uu0i^!S@m0rQ9E+)6iMk-GPE_aEP$)nF
zeCMq?48kV?MW*sRyM<Moyj{ZW@GY{tLmwx8!obTe&_Si)tFqIPva7d7URo+**m0^~
z6-+rcaKyVywFZ-n^ie#gtl84u>UQruR^M6^HDjm1=@<9e69(|xct01?61Mk(tLWSn
zd7H))tehv36X^@YsNaqkQ7TP!L6ggg=y^ugkG$m`?$QMYT>NfunTNk1<ES5MN3`&V
z`<rd)9q6V~9KT!et$U8GivceA=OnXtt7CMWOp=>dhQToZyzq*o<ZE3AK`@k{k*}qa
zwB*eQXY9_tWRM^dNK(Q4Y6>~sgFkgo{D~+SxJM}scp@&hGisHxn^|??(Di<H#0a;_
zwnAPgc5Y5AE{MJVcMxcER}yL@U+E^?e93zW14gD0qx5r_)wT9O;1QE0-m|fzbq@KY
zb^-n*a_3}6-T6d*mjpxpwg3yB+$l-ZcRY+;d1sN<NF9(sHt%V{{P%1nu)2$pT<cR_
z$z$Jc4wUdc9n;lG$?P<df|J4ZFAS2@zOlCmZ#wF`epAtszx!3GcaeAH9t1-&x{Wy2
z40Fl-b*We4lWry`{TKu&Ikfbc{+ms3-_s4Rhl%|)(^`w(e&_f4ovu;s^;65WU=-Iz
zJJ|_0K(wSW(PexCGiN95*Fp_{Dc+Yd2TXqPSEDN~R}~ERFNLRVDN?^1SI~IXwU`(P
zW)&S{9IhC8p|7eLbWj8kPY_*K>IUDUI5Z24%aiZq-=@Rsrx=iVxO3ROUK*61Gqs)F
zTs+L3IyyK5W09k}D&w^6r201rB-q)}M_CKRI*w7J{<9nE^(24i7_j+^r20|>@V-{4
zh5M#ai&bQ^mNSw;w#b$Tufl+B#MiGO)7g`s@>#Fn0vNi;TE=6#!py+TN$U*U{T}7`
z-A|*Wfz||IHg2tp4`~I1j-qaWR3C$d9F<#h2nG0)9P^mKiD3Tqw@Lby0rqN~P>rrx
zE-kmq-mv#|RSeB-;|geLDFr(Ac05S2;-O%$q@wQe)F^$#B+Rq~5C44Qy4lmW3U!D)
zehSE%LaJ~Qc@0t$8c{^RP8}99Gy?%bW9n0QWlBOac{V%R)iX3EuH3V_b3z4A%hb+3
z@XX$C$+;v~qw`9_=I1|2&8CO@!ML$qO4g)*MFy7SWX$2~^9v)9>QG1~m}GMQf+&6*
zW#G84R;82d{Lbx{o}p)y_A9w;IM$jk2It4E)T(<b9*ez}o?oy})_LVDOA>$t)0Z5<
zIEqe(<Zvq3<`2F%A|ikfWB3u;(tm^cWQ5A))x>)(aPsg}<1Mqhg6xxQE)}0;c0MqH
zNdQc(&LIGFE31o%lMIWD5zT~@wM0lKYr&h5>$7vZlzjL6K4QZ@Pmo6X`j8|hNs!`V
zA<5t@*y!RE{RD=l{M3K49{Jvmuyx)u-bIl#YbJ<7xUl=RK8(Ui)bEz49F;nLTJs3Q
zFF4X;1xs@N)Iz<r3cZ5&Z(rjW3k4m`v1;`L&9gbEiRCmc8AlQFbLOnOXZ+mL!uUzx
zQ%RDYwRy3j0%jE!;FB`W--JIqy(qkK{S;XKhj%9!W)-(Cu*{nr?Df#;k;L%25%+vC
zoC3wW)hCJaP3_<K#GT!nn2aCa1VoD6*yw6Y;FmE=3i!je{SSiz+O$hn&%9?ZeAJlJ
zbJNj_m{B%z!dwSxRgHZEVB3ExGAr{g4@lsO<vIij6U5UARYLrM`l5Bez;%fqnjx52
zH&4=R?BJKOht?6tx*q^FDu=OF3}!VA@J7bk*5@wX7beYIOAI#D?dW%Kfsj4P##>!K
zs?vE|@QePH9K6kT=bUK-b$iQp?MQ9Djxm{spgj8TQBf{mVeN>I`|6(61~jU84Dk%}
zV2EbATs-;b9I3m#JQLUbNk4_Upc8zNwyc<Kjh2;d7Jd~Io@*!pZOW5Vz!gr&vi9Vp
zY1yX^JGjP-))OfGG(R1Jp;}|wj99KzQlVzVhuM$q+$zB5c#jYHg#mw(wr{Q0<)G1e
zaUY9?8pfAPF6@ms@79h4o7dDf(P5xLM;K5l+<i0(#`@5t(UKDCy$W*@(sgvKmuvre
z+1H8nSu1J6#SK8_O#sg`KI9JGZ3Ls1vQnKIdrmhX`%R2RS(gOzpWn30;-(n1YZIv*
z<WU!|VGLnGbqx*Q2ms#;j_bsAT*E3yzBVc2Y<=q1H0!u8&RI2>Yf^k5oBiBD=e9wF
z!Nx~@C#Cqc+fB&6KoaqBav$-)&O5LBs<udyW$B;*2`4*2vKNHctQ0sulwNCRkND(L
z*<gIo$`xjgAT(F?9>MFrC9ID-x90~%kOoNDyV-Hq^YM$QNiZ<~ZB8L`q?7&HT}7ah
zwaWGdfr!(_b*wdhQ2KQQQ*C6CE#Jf8A3rnd3BF>!VP4%uo82SkQMS=_IH^4a+VUjJ
zQ+uktw8-<k*Vb;s8CjG*2Q7ku=ea)Y@O*aelPr_85RIc3t6;)5lj${C-?2njU5aE=
zf<ibBOank$!Mv~nGx7#Z@x)k&uR0oAGLBt7nQ3x77s31b2{3$mTv4+ark$QMMgh9Q
zi;@B%SonBg(Jzi`UmKE!*fc>&Gss^xjq`RuS2m49Cz@M<Dy`Vp1#DJrfw~FY@2EFO
zGgFyo#{-b!+XAV_+h88+A?dK|K1#rUp05Nw>QeA|OuuIXnAwz*I#dDnxh)q75>x1R
zVujAnee$OeFTE!tKxdS>!ldz`s-W+!0goKOs~$ueuSK+YR>h14kj9p*=>$_6O+{Q=
zU7J>8@HRyh@b1CI@6<ga%|55SKNMdKzA{Y+L_Lg{Bs@_1#6tIMHmbD=ZvyP_J|~+c
z{$YE>XT|{L9yT_0h<0EbfR}0)Yd4E#Gq5P2>ldTvZFB`zywZb+)&-9)Wk80Rj#ZrQ
z<i>O_k6)#W2kykQ-09>qm<349Ns~tp&M!Yej$Z#|FY>b6;#zmxipxN9#B&YTBeq--
zs=dAn!g-;Y#}+Z%34o~YbeW`1i|0BYB?t$?%Q*}<(sWB#!Mdk@tE`VqX>!D<p5yAY
zl7@c1FH9I7ubnnFcR+iqHAp<Eml?h~r{m2>0r8&u&3r1Yoqs;x>|g*n_$C<G(;7`=
ztTp#`!D|YR4oRkHbRA4cA?z+t@#Mn&XMID!FrwipT7;n<cf6KZLYh%JD2!Ov<VFJ{
z2tvk#@^w2Hx!W$w-(>RCjp2?y#sN82X$g==o0^aleFg9`4nzCa@zg;U0^ldxS;I5&
zAx6Fo(y$*xx16$Lx6{FF*HbA0Wkt34VS@ss&qYe`stK1VFiUS@rxT>-1;*T+Ze>)E
zEraNrDrNkw0mu|3pqUX=r1Wb{=17IRT~F3Y-<kRB;Bo6g0b>CpU2J@<1+Z4mZD@^5
z=vH`DF$pLOV8IANk^F#WTEIGynL7Oioysw>`g;Hq?ujPgox16oQmSrB9{8#1y=VfN
z#l?io!v{?&Pp;(c`%5TLZOly)kw4O!9zF;U*_@}x*?Q&!nr~_PCmYlLBRlsFyBPzn
zXVNZLe;Kw{t_Dors24MKcu4^#>m1CI|6h!K1yq%7(={L+k`mG-4FVz{DG1U?Nq2Xb
zbVzrHgoJ>kba#VvcSuQh!+)LUdB3;gUyJ2JL^$V)nb~{x>>H>;_7{J`v3G9`gSpMX
zo8!F<rKt_zLQ?A?_j8&R=NhcwG%(h{Y?G|B1A^ks@5UCIrs_Q(COQ%z)LT4EI;hTb
zp+X=daUD*^Q+E&`I0iBz8>B5s>@@TB-g0{P?UVL-!W$s+9er?`+67SGX?taHTbntb
z16Q8L`*MiqP-=Btb*;8Em4kzgVXHh^bfk+mG6OKv2A|TxP8IX|6k8&_N5P&jh(sMg
zB%^xHLtuJo%{cpJ1u~;nU4Zx|qIC%r#x%FCi~aICj}Az`=-{4CX|_x`<T~l)AIl{I
z%)+(7w%TQ<*L&4_)VLbp7y0JHoYClm!>rXPp5X}`po!xT?&0@=PXu699j$6f43sqt
zwPAKk8mvMPWbAS)tv9}<ssCOhX*q-Hg}4(yWa-9-aJx5P<4Q;!;d(ehMz>5{oI1w%
z;6}%{T?H7~O4--d)47kvwT}<C73jF=b3eweDM`2(4WUUKR5vOA<5m17_@LaMT*v&0
zPsm3;N{<dve?}zHSM0XWEM>O?G{T}OgcXg0$#>igRl|2AXK{eXgn;vF&XHTtUM(Up
z!1w8d=r$Bk1-u~DKj6pgGcVwiA(^Td-dt22bAiKx4(_`s*$#bSQH-VMA7pMMn>Xi8
zJPMSG#gDGBLck~Non*6>L-l;<Pork*7hQS3DF3FJAE*wdEwi{Z4fKD;_pYeYK-GYD
zliQx`g%f1zSPiv(39OAh?05+?$;%*}T|^VE$GcO618v?Y8$d#S15|~;d^fG%w+fVe
zdddN{l0DFt-r*>9l|IOxbBwOr1Ju!wjDRLvofeST%s!l}UxS|6YaU71%gFjK0F%)9
zD%w$($aWarGr>{zoO~N_8Yph4A4O^07Gdds{rr*)G7NlC@s=J;48}ujL6nOlS}kB~
z(ecq)en}kYwk|k0{?^NgwD;g;e0B-VUKLkzYg_6H`WvX{P)13A?B%$1Bv31-$4~^-
zRCG3}-}*g=bcV7tL0)}hOgwOln69ZG&QU%93>PZO0jru&B7OWwlp6>S4}$rBL@mh-
zz%^vkszH3$751$2!;d;P3qi}0pZN@MuPz8*&CGfhEWX*5LtY^kdAR^?NdiC6(C0yl
zj4q5nI?m&eX|prndcYPR5Vo&TBi9&-8k);yk$D0N&3IxNjs!fP{@jy6V}ABqxjV$W
zPHb5Nf!(|l;DRH&e7DJ2QSmNH19m+@0v7w%O&I_XIlZscYTg~08V56i{^8(|spKK$
z$T1{jUaSYR$<!Fp6SLnNtub2m$ieABG`SKGETy)Xe$MHux_lJCIrvFnQOn;Nm4Ubo
zIXd@>d)QSb<o+O=^nUw;X88SKRjG$%b@*6W(`+z8ei*7d#00=;Xu9dZarVZOURL1#
z%%at!l2y|;!xM~+=YfW3X2#_2l-DH<b?YoJQd%c5i4M{N7G2Vui*c3|Ft%F2`^6aY
zj=-3%i0#Qv`C<t1S(_M~UeFQO^o9hTo3quNmLn#PEATooqUnQ6Fr-oBo<@PBNL9bu
znomb<2o)LNK>d+S0W6hfD6W)0)os5bJcQ_A&gUoMWVGu8I-A56LeUf`YFpx}CeHQo
zXZuo}r!}2_zFtB;i#X)3%iJsIZ=V0X_afC|>=XC=vK*lLy!e1;iLrUZd3RL-^(xE1
zmF|Fp0de3Sp~gtIlb8(;K+E4(n|Un6SDG3gx(GC_iTlCiP66edgw@fv1p^IXbt-?p
zo<|*b&_3Pt?NUS2@Hn5G-RmkIrnzYV;l!B6brnT>hJC;i)^pQWmr|vIFI1uW7>PbR
z9L$)rie#D_|KY^(j=)B;4VKWN?5(kV*Cpom)pmBIlLv5R3C-K?cuDQsZg2TMHMhtv
zZGf=rjc95v>Oj*)-B5Gf5~<hS@eW8@gPwPwPG#W(L4MftH8K4<{v$YqF%+%$$33JO
zlexX!%KZUv%(IZRiU2P#OBe7nC&jX|_#{Ey%G8bXb^!3Xo);m|OZT9$z&JE%rSioD
zmGoWriTsiKVpk#w;|xeMflmYa&%ahKN}!kinxmZ*IZ`U!a~wBph`$k#^We3KW~f8p
zwHyB#?wBjJfH@P_K>B++KhDDwxN^wc^LvjSmTy(>Z2h_y5bf-Z=h7wtZc25LH{U}M
zF6KznLsy!6I|(nLauG6a)LQXrm^Fn(Z5JbVO?Mi$rCL10Y_Y=~<uLEk8!jIqlRJCw
znQyJ!4}%X46;vgrzg-FVk3CYJjV~kW4BfZ1f!XwN_sr6aYN`LE&_W!n1Au31!II}F
zB_3u^FQ}&v(Kl!>3gY{s>_r)5GDGSkxhqfe#2PjQ1Xbf;!D3nWHSDy>s^ttIlixuQ
zxO(e!(+ij(t)DWyzWrJPQle$%rLA>Zp^sq9BN3v+*`vaS6J%^WYVVTD!iJkM!khU>
zknAP-))0ZnMiAmaG5_trQ}ww}UB+M?W<DVS2alwIRzzmaTgf}Yhxk(WTrRgos3o}~
zeFr@-(LH>z0X%)ojnalZRY1g#gRmU4bd;f*sPnvOlBi6?TM0v0K3>sXj!AY9v|-F*
z*gPo1D`*HyEJ7w!^6ulHt_1AynO(As53j9Ns70q=9Pp{ga=DZFD7qux>THng1wuc(
zYPjlYC+iL5M15Mm)_Y@FGt@lks$DO0skZfIf2eP~5wxXb>sp6_#D&LWTbT5@+2p!b
zDNgGxK&z(t)jt=QUeX*smbz~+q!OGoOuxy6%qQ)X4=@jSC^nRw-IE`nQ4l^#(dU}?
zmI34gPYgy_MM9E2?`cbBP#?(-p4bDeQ{zd$1H}G81In5=GwxRAl}yNX2e4{O(fq{&
zMANe*)dMmU*=z?$0#|M5MtMPxp|r%pvjmF&v3+A-3G7+8Kb=0Uw#&^N0#-GYkSQoX
zjye633n|6@ShUprO91U9j<a=*gP}(8o10}^0d4XnjWJ-d@!<x{+ns;E166OT6L4eO
zqx#Vdkyl+*^nGWv+?_gELpZb9%ew=TV;Ri+SOze8p%mhU9S@-zX2mIlRT*LxoP0^*
zS-;#u#Q%FZfV98xpLN#nUKlBrTekK#pi8R8@g^A9R-#bJGMGxSKA%*!Zen~07yPav
zTJ;%ME&p)PhQ)1mpd9+>59che`)RLdw<EiLh!LSWMG%n+&n)J9qG(`F6}@LXUQqtZ
z!0N7-R2UUqo<eK|B=3PCwYlYTZ1fLxK&xa}Q6i#{OTNF$xCs)ujN9Ii_rv1rJ|R2u
zYj5Mm>PnSK*NF&)Ww(4m_@E4+%m|b3Cn9P4mAh+~=GkfM(odIn_Qe65jC5iqcl=At
zx)?WnI?lxf6+GyWT%EZY<F(aqZzVKpoIno=2c6h4!R2U=$=X%Yxav|EIqu>Sf`Ym@
zJYOUt3NgaPy{}&2yIf#LOEKjU{YJaE)wK;>2X(n)p;{mC#d*oKYp)KwSazlvSRJq|
zZFSwT-`}uCw@E$l4t3cs?KAce*-KqO_iMCTZ#Bl=^WQW~Yq=TX*#n1|ZxFLQhyu!1
zIs|HE;qrSDXA#bSD!aV}B{OgF5~gS#hN@t|0nvS8CYc#q)!0jwg0dY%`epY)jKqhu
zZZPJU((`WH1?VplJ{A@}plDwSvG=k~=jHM3o6@nxZhD^WnVGzM7QIRjqx~r!Q8dJR
z9pHw_X@d&>P{x>4i*oVTD24_EA%l&M60p~x15rLJehCIlaNbhf#{ryA^E0fhZq=oO
z7p7bve>zx#nT&U{ph2K&js;G!yZ)T;CGlI?=uyY4!3@Y8p80WC%a40ILmxQYUJ}VV
zQ*{t{HIA<#U0#O?x%X8siqAPCs+1eoo)t{=T#@s#ZT<9@2ejKdYkBb$-;&zy^Q}0S
z88SR7jO0PdxAw)??~iNZKHq>+&M8DE-E#oEu*HQ^h;>eSMS^dg+t90E(r#{Mm>+gq
zDGPd}Ab#TrS8u@>f#q1j@d=N0)KmeD)zGaHl|awS4#b69AoZC-wHYWjcUyFrx7zn8
zJbhC#TZi>lLPC{{Z85Q=%JwNkxXcGL)&sfMfUv?zp0(egB)t}+*R-}DFD#_NK)3vL
z;6_n)0x=Pr?tPVRmY3!vaXg||0!SimQ!qd(0s2dh2kUx^z)A3<d(hsa{-M;0VbKlO
zLq>!4<=ixdWHfcW#4IwoNbbA1v2@)1T#*=w+?;R>daAL2L|SV#Cb<}~cjV;;4F}D`
zZk|0nHFfO6?5BgDj1TVb1v2WKH+g}f9ryAU@79+sSI;f4m&ddV*jG~QzdPV7JV~*M
zv|U!=Naf$)KGRPs(QN1kE#58LWs^ezUR>sW&NZ!4^%^UyTznp<qST`%tHpXdfL!Xj
zm_nfz#tDn-{TdqqhP4AFmpaL`E~!+nP=?VINme@W4A>8npoGRE`yU_DWC28Ct;wwq
z_qWn~g6-Hkw5XD#gsPc6?2^-X<ZQg>DXxnC#^OOfO|D6I!09p9a=v<QH$OTbg#wWg
z+iOgmhCmllH7VKV28foJsPTm#SQ?LnB1r<t$^31K?YD-?oe#BYc!kSO+Ym^_ex`6N
zRujN^_W|^2QYj7w$k5-CgeT}ZE<&%hfFLA@g8$S1M<oAEJ}z?o!S2oN$XDTSS}jJy
z>HE(Y*R2t)!cnsjT-JC$o#l^i%ImI#;q_w}f6L2DoHIJ>Kz4KC<zNf)TWy<B7Jhin
zX=PDM$P157z}6X~<J#s6ck|rWrn~oO>>hYZu+HH`>tb*vF~v7I?W-|y38(5zXw~8(
zw#y<43OwG~PfOo$HWCYa6PvLYjvsvs6o;ZseAdAcO5NSyd}uY3Q2WjH>PUAwlkM>y
z2x?4Z*Lj@wj*{?#;)dVsGPZe9HUJJOS332zXsRz3{P3KKWH{<lH@V62i%h4z2Hg&L
zrcT%21$m6FD`B0do3x?vj~LyAv!H*#;)j4fm6Zncikj26)RGA*W`Im)u-47DP8u{Y
zNWzD6I%*F7kxwH$e+BCO-D<n>Z_)vBJPzBLa2o-H1o#-xQX>V*rRloQxx4_$U*rx*
zjZX_ftZXm#-^J<<-hf#p<T)|Q+nu}&hj~8FJD1Y?S08f))kBcfj9z~_Ght1|(W#cD
z@=ra#?Irwp)av0AZ;RTEgnufCvA3U&Z2I2q(T6S0xWgl?kU$!-ok~?oF$C66*>)i>
zMfzp(OI_deOtWgNP<sL*eIE8>tMKO>22(|<xd&Bgms`&N>oEF+h?B7`r6@{WV5*w2
z%h9myh8|6ttsd!3_nwkn1a5O+uwx(H(4&&D8-+1y)=!MgCv>P!<R)r;7s+%<)+|=z
z@X3TFWi++^$REeV&#y)H;BP!?lS)8}$gn#3bSZ`?ATgRry?R23*G1B7uChR}SWPL^
zR8GdFY_ZxhmGV=;;}HqB=@^wHh3XbDNU&=ApUwN@=!V*EczEN5ppd>ICM5PqJ_1fe
z&f1TL|K<DZBi{)r?*R9=RH#1L&rT~1*cuHy!6?w&6Gnd~a8PiAbZKP1<cpA$?4aYn
z6trePVS^t<9P`C^s#QcLBf{5K<$DUl(u%>{R#;}9PDU?GM9J<(w@)<90VcTaZ_cq1
zv)(xV&PmR4$GAudHv7Q>OGk@-;loL<ochM`JM%6`P<g}TyuNGHZ#oIM0<Rb~YDXOU
zOmhNxe`W}19|`fMSagRGyg3xrx+A;u+bRKktISAs=F^q&%(qW`qV9ls)mR4jV=lYw
z`P#6J>s5RC+R=B5U%Ag6;qDQ5(U)y|h(DDJfmc%HLe9b>tIdE>+Hw1WSAr0UZzX!(
z#12a3UDn$V;jgi`(6!=O-#+np(iqqg`pQ4O_TguzG`Q{KmV)^M@X*hB+X!3(0yB{}
zT2DrkI(^=3V<dmogmVic7<nkwYMOglIO?>7ilU2QvYCc0H+s0TZh?>DO)hP)J@Qqp
z^68Jb!qrdSc>0JLFQ>N0^CFv_RthkezK^zi^?Jtxt(%jq>+X9&CEuxQbRD*7qwz#k
zYkAaDxN*ng=p{;N0Y0fnXqDBX+Dn2b`g)WDeSM?YcPnYw_k=g<VIs0ztFW_m_IV`O
z*_!Xv(QWIK(Mb6<Uc!fn`Z=3Rb+SFr%a(2HM_hR#JA-k^X*MAyOWXeXY-`28&jlZy
zAozkoxul53zgDJIm<Q@xV>4yN)Y<;L@-pS8)G$O{H!6a39vxq#k7O^|f(BT3kQsjJ
zXav3qXMOsSRL~=!jMO7`_P&Zzx=6~0^<_A$nD_tq_=5DIbyv!7Sz7x|o1Z>y0(rnT
zHie+FpWp;Jj@_csc}+6@tTMTsynM_B1;ow_No%Iz`MXFAg|+oia!o`PvLkww(@q2y
z)>Yd{)v!Rk0}R63HA`GW{~&mlH}|ZKSMT|!i`9*5d9-fUpLE0wVLYm+s^RHj#XVUg
zD}=YU$G{{AMvdy;{F!>*6HVezPm|$ZzHVIbh=DGf!LO;UfI(e12cl>8=Lp0=PQ6C5
z0Fzm<ne7=oEfsQiU%_^zJ#Q|Ar$iX+rO^?ucwKM(RbYIDjYnB|ncdhYs9kzKn+)`l
zef)fd9Z)`1qNyHA6uwp;pb6lx+F;lxTuXQ#g3i9A4?3ID2LNhfy5F%P&aGZ!`)=Fd
z^F9;G_rI9TPhXpR=eC8h|Mo+KIj*k(J7yu}G6wWp6J+QD1_s1l+n>VQ3JLIX2y=9Z
z+(htCG0N80L*K?dEgd(>6SBm}60`_JdLkO|Kmfz;u9IK0jS@D;v~{nbpm0tFceXC-
zLG=5@3!Qg`=X<jXp_omRn(wb)j(&#GLO>h6C;g4tDI=PYmbO1#i){KeO_0a$v2p}c
z(4+ddDxN+T;sYyx@C!d0xgs0yc%m=!PFrj`N{H?wkJn~^C{4ZixVP&WjYGQ56NVE0
zf|5Mrs5vs8LI?j=j{{K(>X^z*RHMkoSDZ?Px}z$swOQN!_%rS`B{Newq7HA6xJlB2
zi4j+e5d&1xK6Zr7Lg7jZ!1O(6J>N_aUKUJRY{1I8b|zV!{OW^n+-@P=mQ;c;|NY}x
zoS+mO6KANUP^;7g(s0s8UY8Uq^XC?ux!uYvbtMSiGo!scgjz%$*jc6Uwo~7g3h(Bo
zw^fSOs>VoA!u_R1kNt?@SHnhQE{=@g(Dg7ZgrgQ}ZS`Bw(IWGYH1<ke@DKuCVrz*a
zpgsS^LF%$<Cenz)@@*iI<x;$Hmy1hFOuY)6GB=!OLOQ0^akZzcHYF>g(_%QmE*MSi
z3cFb%p_MOTmX^a%(}x-fw7zB7_0uh7#aq<;rY3~Z3{oe1I5nVU;EhxJ0gkAAXj!&g
z<1Q`f90j7niY|8&{9$qJ#b5##`sR-><GwxZw(H&82+5v-?~N+Q(_y19c(~r(VBB78
zk9IKeNqs+ZJn>hjc#BvIsFW&NOqg6jRIQL7XT1J<tK~H5E<$c`6?VKX6Q?hhx(rBN
zhDmr`${duIJi@!59@U%VHdV*ul?VBWiyFmZg`GTsX4%(<!Ekvy^K|%2+bhKE6}VBe
zWt>5M<;u#HT;R^26gp+TvgCF+sio(~C;ltNyks<%sOThg!X=5y|2QrF2YzQ{lGLVj
zqZ?>JZM!i~uv7%2ClQH-YQSK$xRr~YHt&8>K9k(e@a2Yndil|%+jr8?<iBq<<X?5(
zk-f|fmJ2}*HG?sN(b8_3Qu~bYeLXpsFev{?7FCcI3y?Y$cG-s_=rVRfVQ{|1(fkN6
zt17CcM$YKeSmuAJ5|pq?-KXqlgVN(Yy(Wm(&exdl1*+1j9vQE3^O>@*xOnY_Wi+xC
z>0@5NcrVxE_98ti&cbhdJ{8iyQN#Cbyw=C??ht0sdd-vY&1$whS63c`!<8LT{D(sw
z=htFNVg^HpPGT;b%p&cUrScCWKQd0eexb9ptnN5W<jW1EW?k^WMwpb&Rhi@25ZLU4
z#z93n*w{><`sPr;Fz;~VoWt@$+agFu8VzW@E$LV}pXj9!>Gg9$PdB`;OJ`k$7IJqX
zs-*B|TU)Wle*rCcbPvj16yB{H^CTi4>Md4Q1Z;dAu~Pk>2rF?R7e4DH-kKv5WPDJI
ze0A2%W=nF-dljNC*FIwWbtwDbdMpwDxIiH<qE%nuw_=sqWLVZq#$H&m_HJ05LM_Ma
z>*uYo;h<BmN>-2N5CebXE+v)9Vg9kz6z)3iCXML8xRqu^JQxgJ_!Mc>$!iZp1d|%w
zKiplP!ll=nMvFt~t5w`$eo-&V#O#|Y)fV6*po$XD2Pa%SER>a*Z5^|ZD2elbmRm=d
znckf4MU)R>M=GHkgW-%HB5OrWDF&rX2=jGj1}m&HcZ6nw61$I9LN;2C^@c*)Qr9}r
z1sFlC8wF6)<ca6Uu?w0A0@CSIRP=O}7iZU78HvEcduaXZN>h#V;gVX?OZ!bJpUf9Y
z@T)Hp=pcr*9$2@6m3+oSL`^ISZv)6?v^yk?8J~q2>Ow4EBKuUHg$f@*4Lg9YS*G__
zPT@4a#5U{ZYtc|I&CfALLLVq22Sj_!MzcO!%vVbvv^;w52hY+*P!5_A5FeADy=^9+
z@x|yS$91;gyhVSDTavLqUt>;Q(*grr=Lg;p#XccZ^TGcPk`=$(ug}J%s~v&6pZp)6
zu1c!L7b=%(sj2XT@-Kt8PY4RR^cw(PlLM#?pJHEQpS0@lFEzOhe6P)o2;~(LydM!p
z7YD&k!jk=FG)EkJ@tdxzt1)@A+`@-|ubMtIT4gOjW6OjxKxQP|@<z8EcF{^Oe@*Ri
zaEK7rV$|XYqe-U?hDoiPW3R6@-be~6i!Tgr^|%ZI_0D@-SVm(%eyAxUZJg*Zr(tS@
zC0pp6f!S*czs|aU-ufNMq~fJ1GPY!6{Pelh7a;cdbay(btSz3%tf|5I{@z+@&fECh
za3JCOmH1QrlV*?G-?#K(XaFL72!ZPkPx`QJb_5%F%c}MZ>*Z-X%G<{HV>#7CRLAG*
zb_?wKj<AKXgwfVTCTsv3%vKmjl2EENEz|9?onHOVC}Q}&^;<$|b>iY`F)_DsTVTv}
z*ezRw_<c!2e<YtiQ#4zUV`lb=V15FUbCsHa1=DfDPJMVwb|_Q=_9M7zaBEv>JkGym
zt*lBsS&>GWwH_P~mo(kx9bLr2{jrunDXK=cm4b(;5Qs{!_PoeTzt63)ccDzz54U42
zm#QBzni$sqNFHYhuly(n#r7Cr=}H%b$JM%W)e56es>XND?wpJtynT+ykXDi3bzQA+
zcDevRw-~`~pTm!^bwjf=EnY3Sk8ejphe774N<vyQbtfD8tV5^4;`?~LMY^wqwjoR$
z*o$sxANRxa{$^<Q^4d*QG~i+?<cz$vox77`vetGz#y-$~dyiml%kSOlQ&UsZxl=q|
zv}_V0Zo7A5<KtD+7D376MF7&5G+&f8R<aBv1rj2Zptxu`XwYB;W;1`~^M^vVL_Ar}
zejn-Gco9&jaa51|wYV^xh+94#Sk1_{!NI|mn~YL04T;`A0K5p7?nG>feG>?V@swAf
zImb{fM-!OTD`}@~4?e&ns@N<Y`8Yd$sY&557gf6$pb}_#d}t`y(g`%jpQdYMUH7}-
z!5X$t_CB%ufOtw~cL#nXoCgMflE%lTPmcTh&wTmaU#=p0z%j8Q;^5-`X2pD=u<~5+
zb}JuDLspma5?E!dI^trBy@XzG^?7m=o0yolOWiFIf!-~(^r9`E=5<$Bm+u9}eN*;h
ztFy-Z_UWjX&K;t4r^;me^bb2%J*D&kHp70LrW9Xu0@h==*vYJ)Bi&!MZ`52zy7^w-
zywJEzs{FN`I6||UK9ezNo)wO8D)~PO7b*-57@Q;cMB{=9+7+hHegYIhPzi+(0Pil?
z`IoT|adj|#f&TL-Q#2#G^o9Pd<ff<a0rDeRw~tQDJTEeI*)T-rhZL{U{`L+3sXl+F
zhv)+<mZ^v79jN0n>d4_z+7*-W>feOlqj)G@W)|&c(D!TeLr`ek75)5|*YaO47c|Q*
z6p|tr%<gpQv?O5zZX+NT+zWmqe{K=~7sSAbnZ)l=uiKbvRp0mWi_d@kxqmF>hf$=F
zC)OgR(2QQ6Da(w0KJQ@50MuB(Dk{~#p^ju>k&%%N0I0N*d*gh-2+*_xiZK*Y)c^eO
z79%2$V}T6KWizT(=DR-&$wi{~2@6B*I0Kzz%VaQ1y80SEMz9jg1t?-RKnFW$YL9*S
zA6`GkGgupzI=4%!nnp<X*-97nf3aJTJ>U}chFP-vhz;)F#!8#tMv)?;3>Tt-kr6Wg
zD=b=BSUWqreg#_<+<(}wATAh?5E8%*TV~Qe-<?hb{8+<h_J3LKe{&>O@O|)o|MLse
zM#OpctZ%I=H1~ZZiQTVW%o<JLaHAC*8R`80x00=_Fmn4oH8%6Oed+9rM;hG|`PAZg
zT6(oiDn4DN@2Q;oU`~Hc3@|?b_q+0r!H3W7qPHL~ot{apG77{x$#{CDUwIjxV;Mz7
zj2(KDh03y!Co*(c{@WAq56kq&@{-lVHST)F2%WQ**CrGTv#jvB2$pw;zkb8p@o$y~
zEFsx%B*C!M=xA~Po+n5d;0zxs$p6m^Amx>5OioUYN^n2dKkWU#1uP$Qk=|8#*R-c|
zpu<6`%?&U^dx7DXO3&s4?$<#3XOu;Z!7tnjH)lKRxI<7_K~6Az0RC&iZf5+$t?mbC
zM=I=#O*_&FFe^~&@5y}Y_v>VUr3NQ4`tRSu0us*t65VGWXR+gSO?mgTd@;{-F{z@N
z2?<0M54P&#V65r<;F%bPy2C+?e_kiZiSWs%5QMf?Ok+MwE>;Xiw`f%OUduWK5Vy(|
zNLLb*-<8m6Sfl>or9pmurXIElD50sLt@C8k4Awe>pAq6p%Kj}7dScM>JAANP${?@H
z7)o68x^R^xXO_Vl+Mx|@x)EE0YwE(qJM<v!{<nt`oOGNc(FHo)05P&xuoru?lz_N$
zk|Y}V+MbH_?>LDtM8VeGpJ4t-KAFe4#O4{Qbx!Ty1<40$otk-S<0<u+^UYm~I_^iS
zql@CQ4=2i=g#xPBS}_uw3opc#HGtQ(inW^^&tDE{oFvSAjWvMv*CL>vWx4WPX@%x?
zIm#q@;~-}7I<*xuW!)Y?T1h~6Zsf$uj6J4uS~G+7Vo22JSO0o(xU6lLUSlzvHT+{;
z0}lsBOhQ7U&$UFUFoC~?Yx?Sy`IHohD77a6{u|_fL^WCj*esz|!s*6b#rkWR44a;b
zk}U;yFN&Ut{2lw|JMO2K<UTqWy#oWWn3STm)RUmbt9c3kXPB$S_I}Tu?Nlk5$b1k@
zAoRC=*a&I`<4F|r)t2KdVr=Z}k`e)UI^aMxqY}QBVg}v97#q(&fBm03aU|2Kr9!q7
zxkP!b?X}8R9*Ey{dnvw&A}c=Bh@NwTJuL1%Fh!zgkNuZLdM5zQRjk(RUI)PhWiU+s
z4LdtKo_MIkoMdbJBVYX-_GaWBttY?7<38M^(K@3hwMJ9EKjYT=q4&8+v*BZC1^EhE
z=kMv#Qm{(A89-zc6BUIa7PF8Dju;G<_k&qS09w7Jkk=XLul0>)zt63$Rl&l>J~!|M
zsA^aINzyCd>O9Q`Cp!?KC8eaK;HbJO+$ePvD%|#$C~W_4EB`qc2zI^jelG?}+jvmk
zN%+B|<p5ZAR;&J_5cY3}7z4`Zj3!WgDev@&SewrD6|A_2O)b@BOScZC-(PMZ*&C$2
z#Rg?u7A?ks1Saet?#G+m(kMLG_CnB1Pytt*62!F$y`m`6N?%ekN=o4bX3g&cKtY<n
zA$0@{2}9#^U-~r3q`i4|$5;lbW{qGU9&W97mRnlxz)y;akI(I@H?sZNZa!5+2mHw6
zdBJ0=<Z~sSRhXcbdc)GL6noR8iM;qXN%5brIxQiL>#_&k&gC+0xgQOBvxxkm2bIAN
zyY64U5s?Ghh5ZirlNyX>iOhBA=70W@-X9wmN3~e*$bjjpC(q8r1$(hSPXpQ~lUvtd
zH{$Q^?pCPw-qRn2AfYE?oUV=6b0~8hJ)v|a_puY#IsZM$X0sx=KSO~J_fCM#@wZ}$
z86IlV1{E5z%RpbS0qfrfvH$crMMi5h;f>^K_xU%1Dy0JYV`Jy2+85GY32Lf)*RJx7
z){!_?HzLTK)@+NzuEHmogw*Fgt3z+^Wd3piFr~?Oc+^9k@}yJ0@$HM68O>LkW;dN)
z94x92ac>a?hlYku6p0!%hlGUusu<?S{$h++iGqSc$(A9V@`{doShL;%(`7d|8k9PW
zFiE^@PJ1c@ta|=9I5@CXN3T<*fl!9p6+G9(H-#BHR0a~Xwp{f~{<Q-7*Y}qdCOCS~
z^6O}05zU@hp(Z_NNk9XuUW@ECHtUB`fm<r)QQkgRuYehJQZ}g`%%q!bTJZy~@O7HU
z*gwk}!FO0(4%<;-qyk!APf7WSqFrANOaM^~3vfYfKqX*%_j{_?95;M5L{LjBeK3ht
zSj+ub<Qj}4WQB-t^v7?2C!@rE6%LdD$2Ob(aF7ux%G(RJ+IpE8QVjuZEhmVXwk(J)
z<i9>=CVsR9B(FMnn4IMQE$n~GgBYSP+!>85$-{iBo}HcV3XYHEgNvOTZ(sXixH%;l
zNV{s<ExEma=NiEFiW;Wm;sOC{`L|-n&;QHpKIK$Um|KOV&Gd(jrx6B*!1U%L48N0*
zkS~PAMsaO$+{Lx|v_S`rSV<$B&WQa&Dhe<+A|9hQp4N36xDX?e9S`o`!!J+-VGdwd
zp1F_Zh_kt0+4Kz#Iz(dlM!ofe-)YA1Kq}V5%Ji<$OLtC4`IqeWuQe6fAS_#)3#W3h
zizd@oN#}w$R^{ep)63U|&9P_i4>8>mjt4eoM5S|X+#WI2<zuER@j?0h$B$`$tJ4+o
zjLm9+OSMeL2L}%?a6L+wPxSI&aRbPc5`Z%fQ&CTNrJv<uy%Kny(X)L=P97efv2r{c
z&y`?2Eq1rd17aIBDM*dNhG;QYDSmUlO9N8PQ@4PT1mM#{6O;eKX^fu4Cn-ePAR+1h
z=e>*}&4h-;*{fn7mPe<%#@;=<uls2uxj>gk7~qB?wQZBGn08k(Bd>s#jk~|VJX`UN
z!0MD{Bt3(sYv+CLRWG>@-pYxS=bdi5$DePp5nIrU)$751XXZyXWP+8=da2QlZV4_J
zK>j`g<1geufK<4*+_KE`#ST{=L<Tje=`0bH{jPG}5NF$dP&aed0VXrocsL!xkJ&ll
zsuE%4yTF1r9c9`ZM|$jIW)l`I;4+?eeSQ55onMmitaU)*-<Dth^hJzl*pp@z3ftm&
zF^i^k3O281qu(PUX*pLb0#*%5uBKU3PED;$3ub~3m&<*lZdRDN0d~tGuFJ*TkCMR!
zO7kA6;z3Pig2<c?6mYwqQSD5rIX1Vi>6TiQ_eutOySoY;cE$FP!l_DJC4;TUKd#MJ
zCninwzo29Pb00D(FllLNqmtQ;`)7XWgVkVqR<K=P#9=luLNNvX1U~MzoJF<#9l$P>
zpgV3^$5Lnxy}rJ_(MOXJ2P77XdiwgBoTr>fQG)*1t({MTj>pHx2YB-!nP4!TPu=3R
zdh*t10z^`1WS5Sfl=l?>ABo_|$GfTQ1Dhy6Rq!HnTqjAB`waU@Ru&@yPbJny^%Rr+
z<Q8LGtr62gy9ZKl%2l~R3tyj2;;-F6vEPdws5RU$tH)ek*V8dI4$;57F5@rCc4s!n
zy{I+BO{T@MSPmG17g=!BH5z3b*xCLl5d_b$X%#(4{2Rb?qQdrOeo)?^;OHQgG~bv(
zjG}Wso&SdZhU416qFJvf8bL&d(W>NNzTDhYatS(hh1;|$o>)Q8VL$_?0zxarBqZRn
zx{`npd>}yiLU>TD7X&6kyH4K!h4tcyeX{%z;%E;U==f7!*fPmPzoRb7juw1{Njo$c
zMd^1VCVx1PGiG@S|0}#VJ(c10Mszh}k;$?ax4XD;hw${>$@|~}`AM79NkO-_%|y-R
zi8zN31D1H|X5T07EOP1H%IB-silqAPZ4!PBrcrkx_%=*9OQn5^;k_~vZ$5dB?7G<d
zCsOOD@!_`LG<myKZ8_g>41Le21X`I8kT9i1axX=Z<)jXZsSB<Ky{?d;eq~kTZy?>z
zt)Brgit3c=>M_VIjTl$p+jBr7wmVvR%fi~x!2RNdDCkZxSk6@<9s0OdvQc92NT?<g
zry>0J(%>jq@GF+lM)92}a*b6&yQO;0_MxU;VH^c3T>*dYy7jMi%a(yQ29~&mN9L)X
z8fq#Di+k)l>7l&)@^7*Um4OFQcasb`CWj=!d?n4DiY03b&z~`|`MU-pT*NF4<4HWE
zJd0^?X`Y!FyFkiImmB?%lz`X1>Gy;==8LO+uFg1obURG2<=88-==`0w+BOy#6XP{m
zZcVmyPpK*ob*dlc=s&LOcx?~7_+ygPsye2O-r&`44_n4GtO##q>MzG<a-B!qE_QgJ
zLdVXiIZ_>lQXjZ&Zb@3KV@tYTR~6Q$v%A&jpc8)QJDjViShz%a*;z7|>8rF%SK6HM
zdzW70Uc`HL^We{1=7})_cso?k64&y)Hsp4{DkG(fJJe)jV}qy#?6yb9cb1wOI=i|=
z41F#Smsv3KmVhj#+<dwupHY~WUGu{SvQRvh1f{}{razp%r@RvZLYE+37_Z0s8i)Nk
zB~Uuy*(O&MK2tZF$p3m95;Q=>j)Mhr)=G#~Bo+7Hg0`<Dne}8%4EMDX-Q77}hd=fk
zr-X_50|knjanu?FhNIq*-74=X?xqydN9DEW9JWu9Z!WXEF8bOA8iHvtQ#w|pt2c|_
zA~No*1<7QC)5a&b?W?cYhZ3GEY;hqdZ66Y_2r$f6EYo#m{+<braqt|L*S(XwfGz*l
z$ULjijWX9j|C?wQ3)^Ez5yVLGi^^bm$9^|Gh4k=tQ`do(gtJ>KM;?Q)l&}x>tJ7Ai
z53n<#p&Mc?>GGB4dGiOsi5lceqD`3{D4wb`1hRY-B0Mz>T_}cvCoAeyhTDGqgw_rr
z2||4VrjWV@FfFw7^jM6S_hXVQ;<~!JmAI5e*8$31b@p5RASMz`t;2sRBVae`1>;l<
zfU7`zRe=1%3zVo;CJo@Pa>`4Mjl4`F0Y|(vH1CgiK1$<CDy~}x#YFNPa523MtM_H?
zpvpwd$<6Hp&;Fqy<g^M?Apu4{33p()2QyW`h_M#7{nmh@*45bTzg0LN4v<<;MK{&0
z#pd`pQ$w7g$-Ope?5=JYsO;wm$@S%H*U>qTOZb!e_4Cn|<kGqtZd@etMNg$NugY%a
zqW~eod#LDBJ$HIabsKxRC^=ZaF+>F;M89a0+L=BgR^$9W9s_r4SUHq$$AIqY`h8#W
zc|vIeJkfG`tYuPwfRI6lzarVYKJi5KmkSmO>7L|AKjmZJj#-i5I~_B##h)I$YO7-y
z8hklO`Q*=1G%FN}cmLw|q$Ef*!p9}(c?SRi;s#Z04hBbvJ7O3u1FF^zC{sT)?sm8a
z)U1IgM1Ft-<1S23ujtYj{X&_1pp-Wq$@HP3p^>aG`uS^+eK-!l7V#RZ#T<Y+$C^Fd
zOc(2`CW=%QzAKk915@MVO#X2(GBw0so0}3D{2ml)jVKA=vcwbfIK4ypxXCUE^JVEd
znc%^Htth`zMLkS(#59<sZwMPmdy4Gsn95)K^e`RjbLbryKeC2BS17L6b$R(DJc2l4
z9liq9FgoGhBxPRr=R1EZOa?*0B~I<2Uz@e@agGCM;_qN*u;O}3-;4H*Hdw{^l71BL
z?$wpQzMk$hs!$!$w%D)FS_l-Mnm~$~DU94+PPlx^n8n94zy+;TqVpWKk(+cldX29^
zlG<ESA|cz1O|0#5K=QNO4d!^}SE?_QGHYt-6hFdKB`g*{=}F2-zGVowXw;wsEJ%pc
zBlTzX*E{y|wWfYSiQ>du52N!4l9S@>)Zgr$$p&t{{sV;hAig*PgKlQW5cxZ1TV2>f
zk)e@MKZN%IoZs|PD5wl5*(w5`632iiH_JDJr|Sd_YUdauw=Wu}yNZ4+)nckJ9-eBS
z78_tMSq0Aq7Ob|uv^0cwVZ8@KR+KGwC!rQgjjC>2;g#jA0#qwfP}sjl4?F=!HPnWA
zdk?y3ak?D=A{GxxKUOd^5tded7)boLTY#Ymqssp6ifYM`h}Z3cryu(ewr4o=epvIO
z&P@#o3Crw<ep@wDz(8BigdC9lyoc!`*s+=?ug@l?)KSTsd@)trTTG6$KOFCJ?z7?X
z<6XHH;kvz%WqcvMa@lOm{RLIVC7KCBtYMdBJUv`o(e4ndQla>y^0y$`wvvhskE)3n
zm+6m90aV%^eu}9a+Ga!AFXA&?WMXc$DUmGaP0@Q<In0nFFxB(YJ8LYM#PQA&CK^a$
zezWr@JP&P-{*`*@z+bd_+5WSof3(3Gx8cMa60!jc62uRKvG2G&@|Jc9?QWq7=%nW)
zccUqzYb4+k5(PjHmkxaW)xcP;<OUFIjtWhaQQ=KcrF%WNW^^?S5qKw#0HNJ5chIsp
z83htAcmOq<b^1BdiJ>N<ma?G-e7o}<@|KTtyFXHZUntmE?IFI)nI{=Xt6P8L^-fMs
z&ZIb%u?h+bJ{90j+dP0g4w4B1I!PRu*#F6--$Uxv2(aJMpN)gB<YhLiLdk$+S6@@P
zuxN!^r*>GR_zN=(Moxb*9L6MLci*Jt6>)0EzC&D8t2yRD+ae~1WtwaL$t~s;3GkV^
zO4$=+PgUb21B<2`K(*f`dM^sLGQhz-isr8@BfrL<VA}Y}QGH<h?2D8Eo`7L5U(AIU
zOF4JN(#IuT+4vk3mLKUmk34%n?}y*V*v<2qVF{t_-oj(Be^tPfJMdlPop_P%<_T5s
z{P4N6Ji9pYg}(tmdn1+5@p6MiGogo%O-&5e@{7v`HH%{;QMtx|)j9E|bx%5#22o_M
zhf)2z_HY5oZuUs=cw_Bj{7Aw-BO66%YZ9xz!PU`f|7ty7IU8<QosYBnw$f1xh)$St
zSiI)`c@i$j-{1mUL1&u|1fMwYq;sJmFiWvQ_1s^PA2igx6+mCq`WG;-LFOWUfx!=R
zdwH1H@^B+5!>6<;NlpFvu+`9kYT^{o(e+^w2g|zCfDKH2Z5i{iSnWCy>N4axj2zR7
zw5;s(vnqf@6MP;)HUYRSqqgf15IO4jPyP!3-D$jYJ=Ll$=s=lJRcS9L7p=Rj0W^Y=
zx9nrAFm?UvND0e3V<Jt6S<!+?NCw8ngjzkI4o0z;apA(spMeE6qmc_!I(R_YGeXoq
znJPJnRshU)hQn>JU9T01&V~5yucPJlH+g>)<PJUIP{1r!)8d=>Sms&0O0L&oW6|;C
zy`0YRs_;Rb(&gH#=qwMn9G6EeP8`<$!h9z3QE;H5g~=W&*{L4BM97yJ<qj!sR!~wk
zPZrGa6vb=I+1o{kC}0oR3#ik^m((xyTS51<`3GJ(rAfH8g$GA>hO(e@S!Tn=ur$b@
zwA^W&D7))dUN66-rk82<49cNBE@9&4TJm0(6&Nq{#v|yxb&vj4$t@XE+IX>icSQIn
zV(7&3rVON0h)Z-<8$lAxf!iPW!fl@eNWQnS%ZaeI5r?(wPwAMMFAO=ms|2Ca6B?ZM
z>0@bRUpv?skOyaq`gq$fH+w+m)!A+Quyt)kV)co_Q~`YB3U9G$g{YU!fi9AnQf|f2
z8%bG;TCnxGuBV&i1H!7*sMg0oXm6<~!<)=uo&cfR3_nNopkezYy(FgD4txtIlI^1^
z@Yg1qPdN<~`WIl16`Ic7KU=%XM+zwb0d+o5X*%BSrbF^iOf{JSY`Kw4Y?daE^mwHC
zl+brxj*s*LY!YiJDJ>dw3+3W={9g=HB3KAdri)&Nk<4dCLo(axBg>e*G&@m`6Q5D#
zJ@ZY^y<ySf5Z*CY<mNg3S=O0J;rr;M<zLW`$o%`Wsxc2UO?owCGAu)b7=8CG-pbDy
z4=Y5pLXDM$mfHntBpxCQsb!J$>^FbLTWnl&-Y%NP@Y#Pi7*3*?Jsgf&5ILnf+-0G9
z%;S)+G_ILJL0KG=al%9q9wAh3C{b_BgzI!kZ^g`x&~zMYi1{=v{1*F=;?Pz)MU7c#
z)0{<Y)asOf*iI-+><{`4LmozDKWS|-iXs6>9wb5DGB{YM`^py;I6-~{alWNnI&Y;N
zktQG@ume7#y=A%`Fw*oZ_g-Kw$a`E6#)jGUY<n~_r|X@sBvg8a+xhQYxO(7kQwABz
zo;xjBrv+?4x!EK=0T7~JoV{__9?8e4f78Ey)AC_47NXGuaB;Jqj^0z4w670G&h|S@
zzrlp-%3`f;cIFR=R{04%m-Al{65>;6RD96P_?{^F2GB(|Scdcoh<gil3cw7UKUu)5
zIU+Lf&0jbFPo(X$O#{_tIK_i4je_N-$6oL-r4d{Zm$go%xu6`e;8~1gHK$W6yE$<j
zvv5iOgBpHQ4#)QuWvKZ?CoPwW?q`V!C3l?`-A}gR!fqBFexj<Gd^GcXJsJlp>~w{1
z_C1C0*U)|ZPwSC7^CB+3+MJfgup8Ga)?XM!_5sSKc;cd6ds@==Mh-wzrIvi=xjN0N
zyq4|PRiF>Ymz?F&(ty#dNMR%pcWLk95TUehpb@G-)of@Utn{d#n$9RP?>X&8{kiv+
z8EZJ5U-hKrc8t*(EAK?X@lG8}p*X;l$cwXcf@kQa#LTc@&~_N)-lfWBbp4n=z_x#U
z>tV|y38tR#v`|k^&j5&bMQL9qzAH%{&2Q@~uWP%c`bVK<fQXCu&OF+rcQ=Y$_>Z#D
zrqnKQl7yVG>vwXr!RsnD!PpnlXRcPDxLoD#c8yfhF9n==+!y07yG2DsV)POPjbVhG
zGLX(L#Pz@6$mU_D-qC#IY2amo1{$*o*Atoi%uGSx6$$4G>@wh{tDkkaJqBab^n;^9
zhR$~u1u$J}zZE9&7U(aIZ3J&ied=8Rt*u|6)=B`w3$8(3(xWnDX0VM(KE9#9fn5W3
z{<`U33m-$-6_9KvDr{JT@;*mGk<bH^JIR{>2A=s95sRzxc1l1F|9yMG-y4eeO~}cG
zi9z*5zr5u_4%RXAnqaTG;~9fpzL%_=759Qu1*vrG7wXg8v5hB-)v9PCOEfiBufdJK
zqUt!(fy1@9nte&9K50gdnMOcT6!v7bui_|jIaei3lF^#ftKSq${m@0SuAHzbKigZm
zf?>u)qee(Gk14Iz4j*aBHWZy0LtWzgjA8>ExhR>D0!(=4QlfaoiRW%vQD3RJUZfUD
zX{0aOF41V@z4Em1RHv<c(Z@vkt7Eqx!2GKW)o|&y<^jjOMT5xG%kpVkky^>?hoXzN
zH);hA$es^!<x4Jgib<!DSFbH|`wJDQM9&$%zC_GIE4F(U9p+FYjq$vr!&YeC$!Z?|
zPcH7m1#ISI89fw=k`(}4jCtvwzCY+F3AvY<u_t(4PLj#bH@P_nkJy7okMG6?(E8$G
zeu_IgbEpF@RtDE68}e1(#WQ}Je`V_Q0<?5h?X1v5lt_VQL-iNWo1L4xHFwZA0#eJE
z7&{A?!cv5#7B8<p0Af^K@U<$8q~9UBkxw-PcNB+okau^?M}4tK+b?to#<ILQRbD-j
z-2y6%1khKyjmKEJM<?R?P!l{X@Ytwr`mSBZWV$2Xt5TshVs({k<~2}_A-njYakP(x
zK8I$Zr!>kK`D<RM2y6w@<&dHv<O`}&)E9R!t9Syc0OhE$@}$2WCCDcqejwuH5hv`q
z9unfI1uuS&@KjxsG1;r;E78>#mlt}?xNK#>eJX$`)~bFqFDhqU9v+iW+fDO40D)md
zs+I8fIy}SU!?!1+4hJ!CF??PnI93|kZ<3Sn#Sg}T4_=eH-tMOhJ5zyAIcLpRhR*2U
zO~|_>1XU{6Wjpi|FY66ewqL>u*%U6+i5~<WiXUu+I;4WYr4VesBmwEo+7`<^OZ`6I
ztfDtfk_+1ylS-*t+C-+5?)CFk^4+uLTFZfZlL&Pc+9!M8i>JT6FeN>(Z>DnEVGi!c
zwiXH;(hEhon<VzX-N`1Ao320ZBFTFJt(bOHJrffntSq$wH{2^v`kXvjmfxaMynH$N
z9ds!<0X!s}2+m^7G`dzaJ~sF9Nd#5ElN&!i+~;byc&gz)E9&UVXMGP&$9W#WkJNqb
zrpO=t(%Fq822(j@S4>KZPzx0<L1zG{E7@D#^*3)4!dSZOGNN?UwQ_gYb(_Wd$eh_R
z3j>k>j&w_jarG63ObGwg$`eSZ_+oCiQI@MEFE>{lNd65#^VV*`HvMp4D4NBE@o7Sv
z0xeJ(L1ca15yXTxMi8D>#m9M!iX(cV$F0g6okR!5qy<#vM7WV_5K>lf8c}woHht6z
zVVwWrO5fvYOsufWBkmw7J5vVI)df}O)~~uJ{}3+zK6F4n_f~?Lt;h?tvoyr#JHAVg
zaXGjSn8!%68(5@x(awS>82WUNb8&h;z&CssHGAg!^btK7$HHi!Ag6Wm<1(IPOQQ0?
zwZ&JW#>$GENRk8QC6TZ3JuidG(J%IylgF2if}pkB7Ey)zk*$oDoPDU9MXrn04L#=c
zBMVvLt!kye4?jfaESyK-94^z$VrwT<65b4aJuo&OHGPmZXEbInaU=w+c!T0y<0Er|
z(38m3B(iAkoo93^73GR|Y>#qLEUwy&(Py{TvBG$cF(BL^pF9sTl=V^waeg#Rf^`9m
zyPio`16R3P`Si~<Y^s83!<dnMp&DsKT9qat<VVycwxIw7U9vxx$a@%e$z>gl<IX{Q
zr(?^X@nEFdQr*YIVt1<e<aAhIAPAGvmwxFyE?gp>z7JSuP_Y{gMgfxeJxU@NGqR9v
z@OPL8l2B@h`=nUps@NR_;WoKlzVQsWdlaWhqk_~D7Xc0STQaUdgPsdiv~QmsraYsN
zS>&v-Ue+ST&Kd`ly*OYm+U+c|dB5I)|C}&5D3$$_91UfYJI|XlC$B%R6O0m)RF%aZ
z_sj58=TOKdvx!)NvH!A*C(WCqv=5rji_}0yz&VFAQKoL)qy*;1in|)buA<*hyz3_;
zoC^SE1h0Av*y_X}It;2bw9U`|nGMWr$12;}62EwVn4HiuMmEf1wf_87+%JZb*NXyP
zlg{7f8!c(C9Q}0T{KtpcP|-hTY9h@8YL``td7|evuL(X(B>!nrP9%6JH9F~u=c|n-
zcJmVk`N~V{j@qMkq{zGyBS|R-wC^@8KYu$w+8W<Pe<Jr#F2DsSbGW1x^WO^>-0d+)
z_^ZF)>u0cA%98!H9cBK@Mteh5Ie23KGd#tIyg`bztLptUp$L!9h*P?7Ek&_stCw`P
zdv!6VQ}5o}yn6^VUb3L>7jO{A>vK7#qnq1m2>J$CXHbTQ*j~rEwDiw#QdK)?uJ6A^
zWg2q0RKcIm|8Y_N<<sR$Ib?#4><!O1%YW678470#O80tm&e_a>>cXHWg4j2Ap4=PE
zQJHc*HUAWrlS9?ye+Ptym93|~D9kUz3H#A70{8%g|H|txL%2lQp$m~h6sgz9tdQf7
zcn!#tg`cz+ngCYI01W5Bl=i52;J_@~5O)4P!ZqVmeM_!*&F#?J?vdB!N6PCmq9-Ma
zLJd8!0fypW&oPoFyv-@U#B#&XK{c8OW=s2R9reZ1Bw{@}14#FKq5xa$0`2l1NO;gX
z2vqnHK}VH;h*I+H-?SS4kSIW|z}SGbYtf*;ILi*m4N@R^i#=D#v9L*$Wb>`gIsfk5
zJT3d?tm&vGJm7eJ-v48(dhNb5CN{LTdr((L0Nhhsogt7S09C#m5feqm%3)4$pg%ms
zhFU`nesk??j!dF$Q+eTrJ4)Q;%$LuklHKO|RQGk5c4s^9sm(8VVjc%%_0VJRs5?UB
zM16pmi(*pXK?93rUpdAPKp1NKvvrAHHy;;lhwh9|qJ5qbEZ577v3ux7ebK215_8f|
zsH_W4J{pWAa9#X3S^k*&z4NsNqitSww7-ld)mM8?@$ETb59X+=i!BMzc-+~qj#Fgl
zOYV_dyVZz)?6_x+Dk@lK%5X+ZXBVZ5l$BVpHt6%S#-!8|wS9ud@L-&wj%E8(edi_O
z#?Qv1O(v8By9%phoj*%4Z&_G~BBKrGDw&vQ>HV_%skG@teza;F8(Lti{mxjfTJL<l
z<CU^G@I0WYY<;DO;dHpfw2BePVKMW;E!qMIdh$cCKGXv-b<ECq-ar5vVI|F`H`b~C
z*7fPuKyNhV{ml5Irmw39DWBOLZ^z?;t*jTKRN9-#jK#(ScA(YBKk5v<*aHaTQY~GO
zQirYG?XW422AV7Hv-meq-8wu)?!Sy@ptPV)&WsBQ=0f!MnhRR*z56@;)IXmENa-L%
zX8f=ab6BEHk5=Et(R<`<u^x-sfvHmVKr?Qc{<@XjW^}n*@`pO+GdGF(`RexldDQGq
zB4us8-3j&u+!DbCHPM;$_yPgeNwfTo?N}Ir%HZuGh4x_LH=)U7(|}yAHUDs8mSlGZ
zn=)DaVcL{k=rhkl+@9(<5WR#trm=<6NR*{5Y)OdMt-gkI^i3;G<}tEiOO?`zDC>-?
zV#8P^DxnTX`$OWLcfx>jrUdJivZs5c;MI;9#`ts9gcF_J(%J7yi9nG@t%YQ_p7J^o
ze`8g{&G_PNlDhjV>5z-PgsimrTmW?Bujb*;y<)dccT;q!XNEja9hDwrserg6Cf~o^
z=ws1bc33pJcw0RwYIY_nUCn}FEfbE9$ak5Bl+(`89R7RiWx*o###@%3Ocz)1B@K`k
zDx}OHz__|Q9qzE3&}#bhd4tP`1Xf7PMdEXh2qF_Y5K)-KBwPe4PXXU0nQE)W=UBrL
zV5&((h1a~D-O$diJeMueFb9`O%>@0MN;%N!98NEE#AkrwQ8JQ*54NaXx2>oH=O|!3
z%#|OW_g9I^k9PYUsOolXYj^iXsbsd?xanD_7}Vb-v;uSel<y;8IW(s`Q-EeH!~G<f
zDy7H@7McZixUCW>X~n;$9I4}ywEL+({bnfT4Wp&cj7t4qn!7$_n6wL3!pNR?OGhpk
zG1QtP^ERpGI$yAk=HilY2Fe=LMk54rR9vsZ<w&TCpRa9|%CUG!B^*l;zQ!?+Ad!?Q
z4zLPUiu^K(wO=hkrzcrllxPZ}D3G4sBE2j5_)|-fsw)s(;=ybkPqqsIz1Jsn_?DA9
z9wGHjNoQ9m6SYYEkCaXlZSA#TuJ<d8*d!ldSc4c&n;%O1hwZ9z?n&rmcqlHy&b(q3
zd){-e51TL`S@Pt|x&_N^Tu!bHzv)?L&PQ=_+{02Zl&QOKj>#y|-ZSgdr}V><qkQ7*
z!q0SL-hfkhvdY`7SW13eJRij5ZD_E{jZ3k9g`0;@Je5~=2gv*ctojgo_5Pp2)Nhga
zJy?T?LMP5Y=KMdd&N3*ju*=qsySp`R!QI_La7b`>f`{Pl4#C|6fk1Hg#x;1*;O_2n
zPiDTETQ${xq3AC9oU`A()_Rt7kZkJ3fI?!KaPDWRhDfy=B3D61ojJhkSY`m|WqAzp
z9c)R>IK;s6rhBdM#hZ33WsmkOy8n;58r4~L{AyD61Nr;n*I*jg{a{mtN^|=)hKQ7}
z@<x&NE7EbRE((nC+t%b57{@GS@iB?pqm|5`u(RXE(OgA3?S<s6-7EW=j=oPg=i3Z?
z{6%-)$3$7J|FujReO`MbERXi>>5K-@(sdVXcli51AZEW#=3x6P%gJlv2u=><EB63T
zY4sbcNONF-ifLaJoUVG_FI`vUcB0we=}OqUh21G$FhM#IjLJVHR^(<-1#dB~yMEjB
z|F^_u(1Ql%BoqbsLiumF<pW_-;pi!VUw9Pg*P@EXXF*0O0*+WjJfD*lv-vsp2e&+V
zWxaq9lmcK&u>iKImjEU9=dx*=b00;x==G#jt>t)Tz?oTx1&V3+vE7z@-s&ySk}^~k
zR?2yp{Ytai^z^i(*QCu%i9BH3H3wpKlp=9rk?~|gnt)8f<nPwdt^(LTEH-*V)q!>A
zjPFvz)rml;jifRO@+{XkvDbSk3UEn^81$jdyZ^H_`#)Bkd}gF==KG=R^WhFw96g_)
z1GhaiOdUGcF4Zrhxm%?AmmT7sfr=lY*{SQccK)_$04j5s8I>|EVV#u~kqiqGlW29<
zcH0yI3%Nn9c(jpi`%9o2E}3R&U-F_$36n~SKUk;{dpHWaHA7dT6PIbDhz|xKB7@1m
zwWF!jVL3W^Z{j%K>ZqN>Gm*I=`0`S{S;kXJEie1d>K&zXHcfU}UvJuN()TO2dgjb%
z+ev`=ASPXFr8qS__|C+1vt{q7elsO)D*ceB*y&DrXV|3(_hgOUIX3fsJ2rY&hR9hu
zfyVp^M~VW92jz#~Ubcmoua9L5TwMA^(Mpk+Ola-O>JF_lR+Nsn2xlJ#-b4xpcx`@8
z6EP3_RteHIy9p(F2Hdhf_Am?^2zSw_2X3m`zNYEY=wR<(!b#2x(@M_wY<Ib*V28F!
z&xxJLp+`5(UwXc?yj{sqf9g6c77W?))BXM9D!FNqzJE-kcF-IJh34snD6#38sqAr@
zCQ{4LbTnFm-OzrCRK9IXu!dal-mxaxL*Oc7L;e1UEm80|GSRKZy6Q7qvWF2ajuC3m
zF|fUoFU+J!nB36UQ2EpO{eXq_P{Ajw?iyj!TSP#7&Sv&#(|2{bK8S}2j$ARjn$fxA
z3xvjP=U_gI_Gct2(n$`Th{rNz_let_>|T|D#=De-5@4p<S(+7y7JE@-y!Z1g^SD>L
zAJqI`FKkF2NSNpqn6pd-&VfgF&zEdt@ONO*quU1*9y{8w{URbd-P$(>W7sI=2!0|j
z2eLR4O2mety(Uv0xk<AhnAp0WYp)Mx<bZMycp!&uHKg|PZ=(-z3Pu8;U?jaNjOTTb
zJ*2!^nP~y9#?!*mbeyPc56=`3CT#&)1bkPxzD~^pb_^Ba3+a9IeDwkIBmf`rzP$mm
zfS$lbX9zpfXtyW{<;Wf=tE$U^m7G>Ogp7MaSSlN}^V8yc56>ulwuNj9)%Z91G4Uwr
zQ7sJCU;~VAmNeSAx>U@Gl`X2A5<0=YV3=XHBOi+E=d~uPInOu7mNO!o7g=Rx<xrcp
zm5DG%%7p2){3@hfX0tU0k9aiYofff}dRxMZDez2>6zEP{Cbv&uV_3GVC|_B2x*0Ww
ziczfC814b#aq@i`M#3LnI2@is4VpV|2s6_RRiHIxz!9fbr2aVb!-(xTFHzWF-j`Mr
zerS}Y=g+Mt`(}6VcS*Kr&dQ6`o9&~uE-Gy#;Ylcv8?z#gD3V3KjEWDd1r7PoZogBf
zQi$I)2)-79!?Qadw@nw}ljEfCq5|r+XTdBTQSPzmmATZrObG^$I2a>=(Mt-C3M#e}
z9kOZ$z3<jus%O(1#6SDDXnyvs|M+h0E}h2hw6eb~0wuw58je6=q=8SN5S!eKv+a`C
z)|~f-3dwR}B55p=Z4@v=DEw}}L+fu7%*95u1U!h-ryV-SqA@8O5P_{5f}5K2AJ$L2
zEZXA_PMY%Rb8I?iimbG&{eI^`**x|^HrN?SMk&`*y|j)8KXk9{v>uRBDOnJIzi>qC
z%nXXM9H^jQsk9$`RrSCY{c<Hv>+$@Tk^8Z0AZJ1GXD6?ccc2@K(kcC!Ek=h7!dn6=
z?V;$G8(Psw_bp{1sdr)mvY-7r)qlo-=_3tV!)4l!bkU!~?$(oq$iESzbS92WbW+vO
zH*rVpf=f>IC><K9R%1m9626N*CVLZ8qB3gV6-9hj$YGIJl67w2PrAg<@)&aaoA;j}
zj-<jmpnM@MU%@c*fXmJ{JJkMFZNP!;^cp_zuzc=bhU^2YTU36s`G5A$Q)1l2HO!LN
z$MOdeoA>C%-=zU@<tN<9`3MO*iNF{0p(-T6YUWJW>m37dyjkrKNO|qFwP8;*kwdKh
z0{FvN-O|A(`5et@w>X%MFHH`)dA$^(nt}glh1u?CBX@wT#>8;Io-jm>Sp)X)zuyQ~
z|M0bRDAo=*yVcq)ayBWcKIVO$mUr0bgk87z3!Kha0WrJ=$k!VFiva){f;vu?*>_-k
zQ}6Qg6}afb`L`wd0+5kDn3@){4L{)$#G+b$@B!T5wO@8VECTttSHMCl4@_XQKzyIQ
z_XtS!r75tsNPFELGo~<U-5wBHU_AgMeXYlZ>BGz2ngw8$U;4|Z?*|<4RMX0V=bn1G
zCd)tXgR%H*ksOA=Rt(p*0HTZ^WgJ8u7AS0QsL-yF%)9Ch$2hu*BN%3P3T67*EaM2c
zA-)3VK7VLujW{zx7KOligEW-D{C_s%BtRcl0S3%>*7?p?$=|+F&}TaZz2<{g03&3o
zmEk&^z~P_c#X8r_be^zQ^jjbjGnoe@=nlO?Im3I-3z4}@zhN}j58i}*F-gU*6xfDy
zCC%nwIQ`q!G+U-YkE?Sbq=Xu#A8MVhGgECM)x@pV2sqC6U0Q7CSswtw`9>YSIalpB
z-~~3E_Z-_k0(ds>kDd?b$`uZufk5d^i(C$>MULDHXXbF^5$Wh}ev-rIL|TQfIdVdQ
zk4S#$ClQTMH=V$g+WH%qT!(+RW%|kG41VE_6O2DX^pw;)EWNt-2hh3zQ46-SzDP}0
zb<a5|N-&;xu;Di~Tj|yi!l5_Ktmhb})Z4M$es~P3FMg$-mJlGf!JM+GAx!cC$vl40
z4Sw+D?Ci(Cd`14ES7TiNWmwiw5t-vbk$8$VD^wYH0KUK}m1(7N&3+Wmz1^f-N0U}S
z>RRf~$a-D~Pzwlu+?%Fv0HKY*L<4qKj#@yPC5nq?cR>1Nz&jGIL&)z7Xte*bHHWrD
zd)U6BgeEJ0>TNV$$nC8F`LOu&+%&jgZ2E?lbYAE;ktnd*o!&#)LmM5=?~nkBllJ~u
ziX?LSJ&yp->-SO%j~AZ#8kT8^CyQB!g0tS^aX1wGeLt`-EPsKobX$NArPx08qyyE&
zl=;3<b^X{%i_1;0U@{vQm&TRM8v?7ff6X)Dz)}HkeV4V$<a~Y}grB5+N^6~1wkyUB
zZq(c)J64oFaje1DX&ddKPkYIz_hk@pdLK0koq9SM+?yFkIfDl}2(Pwab_r279(x<Y
zJ7Ij+fhNhK!Pn;3$<FYRZ}jy_gar9#l0s2}0fLdv&0C@<X2vlRY_0oLITNZo2nU;@
zxMsL9Q>59GZj@rpvh0n=bLDBMU}5d~an)Vgk2XSg1pou6!}4?~e6thfPJbU5eAEce
zFB&i!)Ztg+{r>IHPURushl-D7iIr*1x$LY`dI3+<va!TE@~vZA`J(?8_^oTb_fXjv
zkPiMsu}6Hlu5eNKJMj9yW}eXY$}|B|%K%ua7;(tbnNr6v-B<OAaAMwD#zANLEN}DB
zaBb?|+EY)Xw|Q#~r=DbwChI#^7EKo=cK;U@b)WM!Z}9v{;4AC1OV$p}7d#k5=nhZ(
z#_aqkYb`SrkDc*{dtdpMS55hr9oC)xKUe&so&AiRq&?h~83;S+$s>txGsG+oz*!^w
z;I)$EQmg7)(TA_P_G3qtS*ZTKPru3|FtAQOk8PDND&}_WF+XdopLsU-k(9f%wv39p
zyb@iXg)YBgA}aZEAm-12JM{huYH5Fk2SQbaP9V$018rBdC6^q|je}%`=G2wJ@7Q9r
zO#fzmDy~TzafM)%Z=cf|)FUjZlZ^S5UVPn@N&7%c>k}U>?%01mmi4`TdlRN=M2Ag`
zzEeNTvOlg&L*-|jKhyC|&o41Gd&B8|-@mm)<(F3G{Ua`d|2;zIA0w?VS_4tSVWD2t
zUBt8e^9p|5+?M9rZXtBxJz{R_Z+?IFTSd~lF5(`5zKO}R)8G_`e`1`=l5z1KBUyV^
zWQe<F?xl4j+8HW9NqrkKix;5->=S3wGIjTXAyU*gKxxGga32A`=LFxOH3sl4tjda5
z=8~WRbux0zsAs*v{hIiacjANRzgRrS44&@d+K24&%GMG3_}<CJjT~)uw6iGdoumLI
zq=*k<Hxb{yNJhQ_<uDI9S^w<M|2#EKFC5=8ILyNI_zk}q0-CNMX)LN>PmS5&AwiQv
z0vhzWDOPX|(09pOYKnu%|8GN&L$AzCYt1_(Sw?{<pqdMu&uy`!%VM$4VHBu^0l$ZC
z`$oAtS35u8A@tq>c#|lZ?}_T#yOAY{!mX#Y6ag4pW(*;RBuXnK|4l!|P``qQR&e?d
zrD0I%20-abm^bn_i1yUqYJClJLX~uwwLtWWMYS;v8v-H=KOG}tHu=LPxX*zk&7qEO
zvkpp_#JM04;LlNsIIkccL2$;X#HhQ}QDV}*px}p$I$22xT>`=V6j+l<P#FK7hlxW!
zg}Tf#bOQVc8R{~iR99X#LlD3JHbV*mz?MB$fgB;~vUNBdrS6KNAac2!@an-VfkW6s
z;YvAbsT^`XhwOjwwbUcw-{hA3@O8h7l;lJyv1>nDYmvIHR`_VnT=IcKqyatt&fUF(
z96kkePk^M4L$>FAgggbS^8!$?U9$Z4_B#KFVu9!f_PgIm6kScwq<C}Q3KoTVICsY<
z?{jSSn^e~0h|t-acyo_bWqRTtgb>2vh>i3(p#$`KgN_QEC2)CI4c_m+00Xh8Uqkbb
zD=NGOJ~gGM0iz~gER1&@n8W8hgVJZTY?m8<<(6)?lRJE-SG=Fsz<rds#=6=apMe(L
zYqngYw39NB)U{yT?r*TaHpVlxL)<d94(5oH$ohO!FbR<Sem!|H)#;2AW`s&r0G?f^
z3s+vwzTk7G0D0)nxQDSz?DzLT$f9zvy3!`Uiig;!b%5(BmLqmb8Wr*V1;t;2-iLwS
zG2lk3r=opfVKjRLaI+Qvh1&DIs=Hy60l`^*US6@i`ZFPh+40qi%LJUT$?Q$KH5&)(
zfGhM=F<<aGw%tHHc~xjfM10K*42CdJI9;k%B)OT<dm?TKeU9Bb72c}sb5(0QybCyz
z4%{%d3dPYzCpY{MA1>!84rQOb0gx=beaxvXcx=;*G}el=bnBJXC9)Mg9?nA|pfOt(
z*aLe(1OA?Gq&sTUL+_ZXB2ylX%<6}^QK2FJzcp}(=LpB<OtZLCbjgXMiQf*K8y99?
zJ1|`S^5w66@X>ji*#X4-Ry|GtD{a4&Gv5-5z(gV8S@!F71h;A!@3~P89iwx_wq2Z>
zCi}0%&koK{cAvj|DP`HK3r!T=^lR!BpVfMfzuCjCMEaP=sfYbrc#vn|<PO5*MHq!y
z9^<Ez$ac7ezosGn+Dg7X^lLD8qivuZI4sH^dIj26lXfb3uIscdu?TBJ@s#*8U;K*A
z>mVkBk$k9cbDUkxh;2*@VXQHed{R!F?MUk@vu(6~Gkh5~Xa^z}S!;SfRy@MLqx)9V
zWL#d-y7VX{a=j?MMHN-nZ37D#&&gO*I{X{9jmc9BwxP7_#hH}w13u|5wJ(C&&pLbW
zQ}zs#(lZ;Lm)6#DqEhmoozAZM7GXR#X?=@F7~B&SP776N%x0+_!lP{e%KlV&oBZDt
z{suh=DltJLS9KfqckX&VZ6?|Tgp#J0t5cPj6r&s9-#qd7$9XT?Hw`B<uAXH)%P+f<
z3&VtR)mk?J`(l)Kwyo^&k#kf+^a9BpZP?#d>x$IcVRttkyhO}OQq`BA)TSj2Lbk-p
z1(S`PGlV^B^U)ZYdWF8WD_|5uHYl?t9yguf?LSwyh$BQ;z13gJb@Qc|JTP_XMXtx9
zv2C8KgALOvhz%P%F9t}K+Q6(!#)PnB>1A+%g;iXiC>SyYTuN>a!w4K+kz<0?(C4`+
zmb%5sb9a33&Zt--kiNlN26jA@<YAn=<}rGJNZE}8ZYS@7ZRQh;x%3wgj;MNy$r>u&
zXMcR=mtZS?53italZP?L)>j)6#JS*DOV{pTMnkXOUmc&Y;8V8)_A$XMvD5V*34grb
ze?K)P5pxplK_H*+g9+KBI|Vp|%(Ry-g3{8y{o{z2DDmZ;sG0wk?Z8d6quzG@_hkzw
z6TfhzggFR`L74oJX;tR2^y)pz=1=2!&sq6d&smjn8=*tk_E5-ZdbtE5ObrceVCSr7
zJ!C!fl~OgB0V8+WY3mgDi-Y|EB&(EDeyd++l$P|Y^B*EpU#4!2>6rQ+J0slHM1?;F
zdElLdWHcF<ldf}MyCUYQP{yEqcMNsHv_FR<(8t6(N&6&Qc|k!OW=sNXY_ZO$$YOpu
zt#aV^qz|RJ%?OYa=0T#5{<^Q7G<gsl6m;~#@NFZQwiuJNe<<?rAM-RywF^JTv_~GO
zN@zdEZD*1C8s4spcNW9}1{a7C0*s&uGq^ez*onW*33C1$<sJk!rvCMe`!hABFymZM
z_3oH&67`WBzpBC-@$b-@McFfk(;m&-m=Pm4E=USBjr!W&WB5)Zko)BK_mxfr#uR5?
zh2LW5x;md(%CtxOPO<is3oKBmB-ri8Pn|*9m<D=YqukYG_1(`yP}!RInVvajsq0yq
z{jcQOqs<D9q?SW!i0A}lqBc75x0Dn)&$Zg4Yx673F1b$U*%v7#_x$9L-FAavgCs$@
zs^iQB<BOJAioF|U+>H`ay(cnlZV?6sd$OsMjlpV8sy*rs&3{$fqx-h9ny~ds+r!@B
zmo=^JHK4lfQ1ZWI47LxT$cFgr2wd;uV1yL=y5dg??1^AruuczqDMTW7Xm6Z$q4u{e
zuylr=mHYHJ*)Vaq#&ZdOUL{aK?WIgbPocft-%0Wyco_28q6L-*fVt(P-#${=%i5g!
z%bZQ3kByjb%=c3Q`6%rCKj{VSXlvztadb(~bz&Y`QFXCe3dnfjPT#W%D>wL!I@Aa0
z-%oc#)LmZF61~>U)M!`XG_pqgi-CMoY`myOZg9%N-|cLYi>6Oil#Cd$-+tm)g~=qO
zM*R;9z@R<)d7I_=d$NZX%M(S)M{;9x%a;TtC5EE^_5n%5L-AuGrolae>GdD3zmoF%
z25@2V(<2p|%*hHPAg4mq#G5OGv#qLIB?K^h0LHB3;XcenKmiA03f$2{BelOr7~-H(
zeJv%3C8g_317?^bJE|K?ONVe(=r7O;LMg&6-R2R01bpL(db~@GNxmA^U&;r#cPsDV
zv56IEVr*MbSqebP!P7n1ZLfrJ7|!^HoA9481G<H9rTeW_*QdLf(ZB_{p1T?O=1@5p
zmX;Ftk!fe0#Fpm}ri`n^?amStTykhOL>5X;l!0lCb4}DL7wwTn^5A=tosfmGjSlo2
zdPD7}cbTrF<W#E6%V_al+5^kv!3dU!J1{&E#2jsCY~yMJ10zg^G{luQsOvBs5$R6o
zl(}JdLPltHS6)Dx@mdhW@!)(UN^lJb%Ny10;FC+j%ua|O3d2%iwgEHWgP?E362;~D
zC+?#mQeij2L!S<>fzeih&CC}$A-J=^K;lE;!^EQBB%bZ4*Mtf6Vvo`hX3!U+(U<u-
z;%yiS=&3!obL=Mh{%@~P(E$eWPyGZ&))MEn7NgItl?F4>$!nM|;sF>KXMx0(dJ&F%
zj;lam|5wvj=in$3-s~r1Vmk5-2npS6_ox>;Godflqq!V8R!k#5;m()ufsurO5p}Ld
zOD?_YjZ(SK1K-z}!dGOv8E}Qs67n@`M}fyE;X~6tWC}sW2u%4uGSz-n#(G(+(5bt=
z$Fb*Z|N8XVnqV>)b}Nwnk-pw}Kfb0gFrmq`{2>@YTXFVEivkIYGu0t#7NH2pRm>Si
z2frADed%E#_h->oU$52DLvDeAMpr}zyj8%`&F=zi)LVaD5eoE?)KH;no+6fPv_cML
zJ^ln{yj<sEDYhy`!@kn0emy5=T)Nonf>K^v=i#{hZmCT9=ihc}Ok22Ysi#n+jTW#U
zHTC_B@z9L3w0sW)eaMzegquniqa76Hh1^1#nJZ0j68dPSOqp(Km)@J>19nik0yD+A
z&+7ISx`B#F*3qx4uFReuKR!+H#xj>^lKs;NYl=1TBT{JTZWo;Gd7Jjn<D#~_xjb5!
zpZ{K=(3A2#TtZ=22~Q0CDJzF_>yJtkp*G-Rd3Gtsb$Bwk^nX>`5gFW~yz10_cTeNm
zHfh!_&X<YgAyRA$<kJ8-`W;8Z(Y`Rg!{1|k5&s)x#)Vq>=&HZGuVVQiQdH&=S*1w(
z?}xGRVSWr3&nPr|FBuNr_;#;zET~ZD@1q~lCmIT+m!Q=s@M%B#W+%|zbn54<)up&^
zZ=1O;l^k*(HbnQAwi2v0b|{qY<$12LCLL#JC{aT+zT)jT3;mJ)9z$mT0WpK~UNf@H
zP-9w7<Tw@ow>#vqA#tIgZn@DKeez}@9B%{L+mM;j9V2OB>rl?6#&uPV{NI)@RO&Q!
z1yGe<qpFInpU6+ae?I+<6D!Z5`N>OLrbP~M)yKEgI&r(LX!<}VSf=71Y3pNJ#y%Lo
z2{TY0A7ZB-zSaF7sn|FJM0dOSphZzmuP35S5vNfBrq;ip*z9*<s2Cx7*L$Vz5r7u9
zLB=@^0PUt<JlLn*l!}CNtP~3euehwCBrTjq;H-GGIhYpk?V#8$5#%Bvm#AdNm#KHx
zK^Lq8<gy-of$q2-Mv1s#DPg3Zecsgi!IN>J)72wPqz3n&anBz=RG1E8_@R&&3r`&Y
ziz)_&IxP0}?sa&uXZ6=WHn;12H3&0rc`g_bqsrk1SBXs#LnxTT<TU-^(mIoS8CZp_
zOmwUr!vf`ioNxV|028=|BBETk;pgrCE{px(EpQ&uM5*}Yb>72y>LKrVQ_LSybDkcB
zjyy_yZ;P2Ge!X7xZOU0cB9=lkfSr?tMJesC(M2V-5^B3s?`}B*9*j1#goCRXGa7oB
z@w)shC=k(bU8<hwCoK)B#odjVVWA#NRZ~6XB(#_Z>s9}H0)7&5g%^G}1N_av=bXW@
z5};aGBAOaTN0}777xZb5vN|{)`UJ607kIdmWltG)Gv!wCRW=KRM__`-D$?f!yKH9Y
z4E88ZgTmJ@t&PJU13H-q!;rAfyHOo0I|)<g!}upgVBp#n`vm$LEaygrL&^YQ4$s4y
z6(fuTsXdF>v&W6a2v}C0UQO|N;<#;W^a0CRJDWd(hv=!z;prn$l;ng8>+lmAMqAnV
zskGr9I7#Nv9KyyZ10nd^P+RaGN!N4B&J+)Kjcfyr>-AEezj!$Hx#BK3={|XG#&frR
zJt&lEKmU1LzCp=FccgQ$T-}E`P-Gm%k)L5g(%#ckC$vPc)CD_vw)uGOmKP+#e%if_
zd>mX;NFR(yvZ|%^OLr&93MSiK5fLJihbPt1*Guk*;r9*0!!rDR0}$QTOk|A=etZRr
z7PK=_1Kg1BXofH@cRs&IF_RwLrky{rxFNX<I1h0w#*hI6K8dm?nDX2q0FpulK=FSt
zYv;oX;M)&<qf8yZAO1iAZNi)GZOb=#wnd9ThnzCkjGCnPt5P-C+d<#m@@5#3R=5?2
zR`TpC_z{sso6x9YA#Jn0E!8hzvhR%LFrJUtXxJy=<84|X!q#CD9=;tuz4@0nU@P?9
z*3ffvDmCp_orFOfMqrn#c*q^^CtY35axlrSW~L;%?YW-XFshjylHD;Sm;vAdF-h!X
z!rDJURehClOdpJpOdKBfHxi3rZ7cEDcA3k*&fQc({N7j>8f9u%Gt67Xg#Alagg;`Z
zbWa2E*r_D3Hm{PzF0U5zj=M(O3ZA=J^bN<-@mzb~6oGBpmN%Y%PWqr$kV1<?1ZihY
zS9+w!hREXBrP4_5WQ~iw2+#67vv8%2ORv&W-B*cD{7*u10qYHiLmcWBeJ&#`rPv|m
zM<w^?lmZ?FHTR9vK(Mk?2W5{Y=(6K|ByPOuO@xhZN-{;qr9rP^Q=7y4&TZu3?B^)l
zv*n{!d1T&a{Yu#ObJbjqO^KRK9>GcrTlr7mm@Sp~)8eMT%0H32zemq{&-(s&T$qCm
z1l$?@m6|9)cjxJ!=V&$Ufk-g%3Gtum<AbD69nbDJcv&NQBdS+-k~02wJYz;<&YeQl
z?1~ZpL4HBubh|Jr9)^Z%3P&RqdwTN{j+lj0*;gEfyDvbr7ymH$|AQ}`75UsC+a%-n
ztXG4QnuB>R{vPx1oCursi?%oQdl#t^uGWdW2Yo^hwlBau*Y(9GJ+5UJVVuG9DUj*;
z847XAH@(r9WP(Q^FW}w>h}O4*>mIO6L`Cj`uiwLcgEsPq*34YGCb1*5q0WG{zd{Fd
zzf5>wLAim?MZa$T2Jz{U0#^}%U-LJL#~x$cQCc&3jrNZN6evB29?L=NeM3%DD?s&|
zSEN_biKR(8*}8mt^cM-5@Hg!bD;oz$Z2UFCEbU3-*#a}F+^`$ihPq|}gb}5)%xAeh
z9QgswkoU=fi+q@Eo55+T3`tU;2!aAkTX83-iI0|SlN&AQ`V}6VJrsR2mu;9B#zMp<
z*)E8=7nk^9S5eQ;0Lk(a$Rd1>ud>_;`ym&>3*$T=cNMxNK=PYRdeuWB#tX@L7>J+1
z?GfA%8tTTP1`hy&{6TR`g1=R;IMhLba)ASIGHBnC_{Q#eX|AzB4I7D>(_p{+M?fHc
z&Dq`?yRHzYSHWW&E)O1!c>kJxGT7ZL0z{F|jqSem*mn~}>-B3EE}jT&S~eS}Xr}=}
zXlPoqEg;O(8~yGQ<80u(^W;DloywxG_o2xYlQdsBE7G0J6<kQi?0TlV=Ufzqv4l^$
zJf?aF=<I2MW9%X4sG}7^B!|L%0hmw%hr%Ia*5Ml|c)6jmsLN_v6me3oIcW1}hwViy
zG8t;X@1tzM{0>bh@F~60=KcFq1kirK{cWXNK+R@2elE3Gr7L(-?V3;46)fp9>#0wI
zc@wfkk?w-W-aC^u6#`oMPHYg4*`&)et3!Rk`kS0?B>zMrcijMt5j!a+#T_NuJH|fK
z0*FbeaLT19{HcrMbR!WbrA5|Vs%aF@QvSe>zIOvcWY4(tkxR;8z?{V6oQrc-C`O(%
z%+2BX{vr;rb8MiIK(lIH0e6$ku(>tN32*GP!rY>q0Dh7eQP}KtxcEi}0^yq?3?5Eq
zNRS(7OH2hmp^^TAwbCSJ=A8V5V*osaFrZ-_JZYJ44;g`yozLA`Xtt3ekMVussL$HZ
zGC;dQwvc`c@Pp>!=172hkQ76NDqI`~3&#~<MB-wR4=7jiC9T7qrFoQw`q0tQeTK#C
zFdK_^h90poFQs1~pi(v(s7s#=6dh31j<0tD>~*W2UeAF$dl$yUjR3<=WI&SKcy=%U
zRWbxA(33wciE9&w(WE(5HQeCotK93l4f*`~Kb*(8-B~FN=30?h7$1hmGq){@)9c<?
z##Yjv*>yv1U)y*V-~5)_-~tGT)})jR)1pvJmW(~qF2D_x1V8y(Xk?S2{Ks4KtBB|2
zK+GwSXH>f5AS~Kgf})NIW{1M7>&$sFw(E^TOfXGP-zmfBUw~VN{I0<0PbctY7_hbs
z^CO2o%oXv@lwm}=<E}y9a-LOC^pf(AizGvr5(;ZK&W@-N)NUg_<s$b!=E$R8YAM;8
zaI64r16S%Tk}rPW?*1KnYL5ut0H3wiRM)!S7r?IKB^jgx2K!#Cu?$t}Kcwf6GJ|`8
zgrN`px4V}wGdb%Sd|Sm<icKUAs#{Emj-P@pnu!8w4IPwB#jhFO!&q+0^U8XLqNaM>
zEd<8BxZ>wae77{OK_e!Hdm>A{x**kvuW<g6O*gleu100u@C7p>9!JQ1J}+A)@BKje
zM6}4s`vU_!vxw9V9f%_doTN5l_=YRq<~$+oapPw<S4aDPyH7tqq$ErzsLxOP{!~I5
z&~Rush(qt1V95wF^813~f<E+i;k}nfyM`m=#`ti0l-4Xum-LXlPkJL*S#xp4nRWQo
zK4-H{+xUvqX*q1`v%*E{3*7L@dHmA82^&|dY#Izgnc2>B;hrze#0wTm-}pO?=JHC^
z{!n(h<2_R6RQwX~p1bUgVrp;6!rSvH5{c9q+;?U1X|<=z?fq>`SuwiNL(Pwd?&go{
zZ~hB&+tOSG{wtA6v2G@W|8d@7b=Pr4)cuR;bUaqg4bOL461@0(atgZ-e76=x2<7^2
zOhT$CuD5wMGFkZ$h)FoHf8l@^x2T_Y(l&5+=k6&qu^zCu=;44S_oaZnoeeECU_NUL
z+JD6K_H`t$zB%3zcI=v>vv||rPI&ZesiFK&@k3rzgNU<I3|M4I*IdkOqeCb+<H>d(
zX1$&!y}ZZU%kjifw8ba{Q#HrzV>-5}asBSC8d%(SqzgN;wb&%N5_@`SPJubzrd%S`
z0O^O~j0QoGfq^i1#%Z<Gk~^cM7#8+ctw?rpj$+J{p!e($uGS!uQgKGzl&SYclBm>h
zon%)ok_IqJrxi;vou#?aUb8DA?5TDr)+fhdf6FW>kdAYs#coVSr(y&YN}i>X1J~1)
zMRnZ|xHg$#E-xPoG25`$GD{*KodFhOW=XN(bgR7vl}XznyW-48K4Uk1jNcDVua(lJ
zB~Psms%(EwK6n^srg{ZBY886S|6TE;?&8<__Y>&K7A8Nhc*c{c{hL$gYA@YptyD^+
z4cus{*p*Zw7lcm?Ka*5?4qh0%ZLK^`E0<-7djpz3@Nr>k)dzW>3WO1}N;b0PX!p-`
z_fiNU+$=?cRecX`v>{xhnYueH^L{4{^O+@OA$|Y#K31JY{&>=!H<jxo0bP18ntVgx
zD_CU<3&2auX<iKLm|FL}Pc_Vje;&?9OOm(^o9oDBo6(szyNSs;$|iFptMbQR1^cVn
z*9~1wKFQ)u@5ch^RyW!K&T)b`i@9X&VRo|FzqQ=1b>``8*<w5<gOv^|$<$R;@j88O
zw4dwBX*E%2)G@vtH`9$NA`QF_uBLSB@Vs2CWi{yN6-TJwO3S~)E0)Qktu-$<R~W&M
zqM-~)X4V-|A5J}%d|w<Y{a%LQHTh45WoCw@=b3*epJS`bnt;2=-MBZ`-rYlc27ov}
zI?ML{-LmI1ardJE&+G{z0y7Xs*!&5XyNApSdpGl#0)vk&B8e<N3UScI8gK!y$2l2t
z$Z|T>**swK?Rp)5i)QZ}*~cMN)kx2CdN?OHBN6JD&`^Wp8zUN9HG}u6SYJDaH)DmV
zLb@6Zn5Jy3w!{y;jbRZC%>`OUF@?ChV^L0`18-Tj5>6zpMgF;xygneC6?dNY%$E`{
znSkURwBa97YJ!zCA)|S9y^)k-IXyN<Z*{>)l<<YjceZYjfTk6GO!Cgf$vm#}VjbKi
zu6ma#ceG8@b(K;69`IyWwba)YACuQ0M=^LXy7%6z9O*vSwHKBFb*J?xKqCn81e_q>
zi#Tx2+N^&EaxN3SO%fP0ZyrSG=v4pBU4Ud`3h6^)5uBIF2+ZdY^vp4MY^Fo9$l)0<
zYGANX39#x@NH;V2olPSG%UdoAe2$G9el;i47ef_8SrIRI$v#b{{uwWPTo&1pSk4&p
z-+^;jb<;O{TOM&Ug^w44&mC0qo>&op&pUMt?)R2@%U#XP`h5%KD`nC4t84j}I@{<t
z(vGN)iuLHQ+-IAe6;HELk8uo5NWG|!URNdFW#d-V6-)1d$aYNIFG&vvJzIaph#JhJ
z;7`r0%Sq%*)4LH79*wt$s*t=_%QK8}s^{<-)@2L-aeZ?7N`==YnwX~h!LyjvmOB}y
z&dq@f$9Z1;H$$@MPW^5rjK>8YuaJ|PrS$j{g4^AJtfy}6qWR{wL7^nBY+v#w(Ur?I
zm(fYpgGT?Y%9u|qy^x-u$=}fvKUIx9lTP_q=!Ck}AzhuF3Vu4a#F`=>iB<pWRVs|e
zZfi~{>u1txG^FQWvo(PJL(352qqun5SxQDm^uj*s)Vb{gI5#O}cZ+|8o1)F7SEhNp
z4XvgLYq@>b7UNx3xh&*r^FCL04G_U#d)N$Z=+O}vFS>Lzf{9sq@8x@WxG;w1ZM`sf
ztwD&T7?HsBpsU#z|Fu18(J&|jJe~`sSMdGNCV)&xt8gBqC?7P1K2~Ulqxi-EFt3G;
z$IJq<BCMQ1H(dtsH0&}_W2Zqp7<NYNRiu{CTXgq%)Y^dl0GrB@tC-F%A@tJgy&F!V
zoCN7>Q|r!0a8qxqNZ`Ndr^B2DVt{5JmqZrAf5D=5+DRoIp(z?0r+)209|pB9mGQC@
zU%*xj)hcVS8?%n*2&JJ59kUy%XOb)-DCz}_r_(>qDD$5{<!sE7rz6Nf<8!oQ7^iJV
zMi7nfr$M2HFpn8e^dS*de>=#ijByS*OE9X21T7}hd$1+!Y2#Eg1VLfiPmUxbrOP;f
zQ0c2hag!9H-<*7v-W5D&Ef^0zn5yjlDxmxY40P15`~&^qb6U8-r1SUmK@$@GS`r?c
z2rvQO0&n~)*+pOt27Qin5HCiUJ^S)ypF`H*0euXH-619x+LFMd4f|ffiyo48Z2~<!
z>xLKH5pk}T1r!P0DiXV!ZVgG$BHj&+X?XCd=Z)zs`G94D;nc*Y0ep13fkal%u%6lz
zOt7(V7Mf%p?A2dk2{D*36L$HaiwVVZ(wG>+E6<>rBI8_Kqy;<^<sk+_YL2*C8T2jc
z4UyGPec?5qXaHpuN94;k`mH>>VxiH1kRx6Ckx5)&Ql2p68JW4HhK_LNo1@TiH(^ot
zHfqN??3F3|))H0>y)pC7a4g?yX-!9?%}m%`a}ddb3dAY4aV)E=R>~$<d!LC0e~-^N
zZVzcU*BI2mWUL48BB)}l7Bm=PcRh-a!gK3WBy<Xqj=DpC5ddysSLzV7KLXvSv7%A*
z#&5Du)zs`KeNW2bLcQ69_M|bVmO-M(zG2qNcKCnJtg2rrGvm!x;df2w%@_vUVhRe+
z@Mlrt@wAMT>&3vqoV&LCW$|a!LP4FV2a%7Th1QDQrfZH6&Jcx!yKmy$u~HExD~}PG
zC)mY><{#Dg+bf*eNb?@6xbwNY|6nw|C_qm1iYk$XO7?p*kZXl6X=*6_D+jKCg_T_!
z#YymXvB4POH<__g9lo`Vh!f3^zhSGF<46^5x^$&(#PsBO&b(t1<LxOg4NuX-mN^Hk
z%hD3(#+0mb4c}ti$wnr&1hkw+NqG|!GcV@(HaI9nf)nF}OC$+b^2s8#6Yp5p^_-zy
zTqIeZY}E#&aWEFF7$N(WJK?@*OIRjZsSB0Y4+~c#D38mRM<HG}jNaEBjf01m0mp^c
zcSdDjVb(%;He;<A4c@p;TCYjp8MY496G9D0=kUCm{CpVFNG9B~;AOGS79JyU7b-aI
zNnE+486ou3yH9=Ua{3*u?_R$log^Ab@?PK_8pD6Rq?KT_HF{g$-7cF)&%Fj7Ul-z$
zR|NHNmXil{aEZZs+iCL!f8wI|9(KD_|8zujG#Jy|9fwT%1kr8wTU&O94Fa{_s0V5-
z|MuISaoe0*DI$B0SbJx>?vGYZR~vD_<ExzlgO8r7@PfOZ(36_f@g1k}cApmAOxw6B
z`J)J0IyJ*jLr)e3*Z}x3OF=A<Q1w(l4%jJW?+mDFqe<syN9M&cDFl`1=O6}rjv)#c
zbUP-*LWL<o>?7XL34&2vzXZw0Jj`8~W9$SCt3UH0^IphUdiDC%OpS0I1jZVJo{mA^
zd!0kyVBW*O^T-vS2mZ}lz!d?1&b(CecbGZgohCFUo;a2p-0XRKtPyHV%=ic=@jjyE
zS5Ii19XG{Q;@&s_xyj{f_jOQ2v3UV%|Dxw{BaAof7C~dk4X(ughhj$04f?llIpZ#B
zSqST(l6=oIY(9w`Y7i?B^sE1E@kYDqN@Ncf50+1&b6#F~oxL?$w7CM|QyI2VeZFER
zFdKY^U1}eyM_sj=GJ&mohSKjzzil>TkLy#<+OwQ4rUq4kP9o41&I?23-KHRfpps)y
z4F~J_*AYTc=(@V0;kCm)iuCR@$Kq{oNTR5(Sk)9fR5BOx06NG@##p21?OX(8$lu<B
zGY01o^eEYJ0ei(G)U9Ty9l$G--^*kB=RAo2ORq3?xJvduB7%f5YYqaxlu-=Es5bUk
znp5@o@E~izNK8h6kXSEONJh)O)1nD+ZcSB~Bh#~-cel0~Tw(=g#9z=c$~SHm!ZxRm
z7$X;#m?#eIGG?Sh2SBuS>zJd_%Y=V_`2%}yG(IJigYd(4(yzkOb5dVo^mXH@9Qu8f
zU$6mR0(?D#dIk9_Hl@dQ`;<2m4z#$#Oj%I_YDpkI;$PX(5cmk6qHXHi-6BDSz4Vp|
zz>pzrEjX&&#vQ$Uz%_Q@8RQxZKzP&dhNfe6vM2E+Ji!(*-t8VKS<~XKf5e9;#56x#
z1PdKIeuZ^EUEk%n$OpvIjJTe9k`aYAX#{?*-Hxv?ZxrcMd4HckR!TF>cEc*=E%7<^
zS?iTKEBrli!}JYB*U`WB$RoqKkr2wxyRNI0@J@UAio0HjJ1fq`4kC))tq64>RXaw9
zJfhu0qV6vLqE$i+rfg?uyDEzkQ%2DQ7TmLVyxN%pJiZ0iUV*&>Na#tE(8VP0ng5XE
z`aV*g5PzYBd*cwR&R8iCne%&*`Z@KKt<9$1Fl6+)jUbP)q+&OnTubg3H1;S=D$8J4
z`>i~0HrJQY*cp^fqWb@@=}#~qn_>J_ucl+}cTA!c8jotCAR$j})Qp{EgL%Jw2}{q8
zK39OcVcUmgxVawMzsmg{$2{8lWJN&`x%)`Qd$s2gF4Hx$9I*VxIRK2+9}v3-1gL&M
zP_uh*JvsnbveXp`qnO5i=F!mSBY&*z`hI}gTfIQmra)JkAc1!phzf921sxHQC8Zb_
zsr5|<EB^xQn3`%%d1jl`&`3};Jur(-)R+4V&~sqd<hFiY5C#PLGj4zp2%i@xIg1{I
zbdf&m;pgc93J6HZpf`kQgYbml^ttMmW(%Yc4Bfc-vJ-n{yKtcPC1G=c53;6M)Wb-K
z^*%6z5+52{iTdS?5)><ZTiIYcKP^Xoks*5A$Fz)b9UJTS1NtT6n$;9%_6OzXD}D}}
z@DpwFv{&MoxxyJq!)4)sk04{A1OW7tahqcN2Yx082da&@+e{b|2*m{nL{-ZY1H7q#
zIatCHqLc>}lAfCj-asI?H?FOW3X(y?8qLx11gc^zO@tb_M59;9Nduga?`QUKjEPqR
z5$`V^U-xGsZyWaOMRqkE0_!{+JuA|IzA5_E1;8k|=Hwc|4}$*6R9Xd)LY;y~OY4Ac
z7@YL^9AtEl0<qo59F8D@-)v2Q?QAJo%^;a|m;0V@4=G*l7a<>H`u7#gEhvCb=8TF>
zq=Te_J)!~!gsYze7arpg-X?uTB-FY|CfEz%Wul<Gq=I%TQfjs65O0G#XS+9_0v6>W
za=&{>y*6mU%+C&Y-x>f2%UbE(ff}Pa<?LiA7L%+|LsvUu3ihKtz2t6xJ(?}96&wHA
zOR)e341GF};dg%urF5x`YiELo2c8YS4AaWepR9hESP>slvk|xH{s>cyx}rsdJyzGX
z2wut=9WPu%Z7(RKGL1oat|uYwb9LXchPT9riun6|i!P?rm*2!5eK(V<&OEmH@1J)5
z8YushnITM4R;xw)*NseRr1c|E*2EP}A3E1k0_Nanlpb1M{sm<xIX~pR-y4rAHuI66
zv;X|+;UfKjwtGwy;CILRl8C(Kn{;U<EYn6#q|)NP_My#zP*RAH>*RKotjuGv5ZPqZ
zapBJKgMuvz_0Fr=e&PA}VlrPQ*>-;$8+PMfg7XW@$E#t2;h0z*T>X~OInZiL`$bWi
z)1T|h2%IeLqD+-8MWlwK)HWhENmRSMJK!YX^57fm!phL4N$>&Yzh5=wP`R|iRV}Vk
zjXON!)}a74SKtQV#(RD7gA(c`f)gN)k6?XX8Grk!Khe!{52vpWV2nsyv%)T5x{t!d
z$GDRY3+&N`{Iu-p$1d)-lT<##s$|Pm=7^J0QFXhJl~GbS*->TZK+1pVvGH@OLDyn@
zikMbAotg`PK0mxh3&^qNQr(O7W`r&=o>}++y4<v156T&09GXtY7ataOf>Jh#B;<~D
z_UPCPq{F07m-ve;uCuV>2Y9==HDv?xIa=0Jc(UO~+WlU<9LS;xS$a)Qo)H$}=?#S@
zqFGztBT{V_6S%=heSdFksb$_{fD3<0<&=Pzjs<cBQ{%%kv8Q`t1-zdEdr(2EUOqn_
zz~?iA>KtTX2O2ra;@mMntcG3qqxRr#)Tt-)5~(^~F#x{}9Wh_lY<~qbVG={Bn&1ZP
z<@w&~6x9`ZvyVm*hMN_S3%&amY86llboyCuM86S=|KqHFA=izaFeX6IsKQ!7$d5yu
z>`{u?LZq7x#@Ka4_9ElrAC~Dy3a8_>Tfj`f7oEO%ZNa<F1HHe1O@)sMO4>Cp;X&R-
zFjNwbe59>%4TI6*A3o5OHiLCs0sa;9PW<LPKuhWTR17gsl(!SqOU0rOHcDEdC%kCq
zU($*x0#g9>rau77uI#v2D|6PMYY@Hr3X}@aw)h%NVstaF0j7N(P0Q*R0(|_>sUOCY
zqdC%=K#M?_h|>F~j?bf&wi7|Dislwada)<88+;6w1MEw&-g#Bl^;pk8JW(Cc86d?5
zR3Fskr%te=N2j;{T1qRm&%B|wBpUhsn~s!D+9;p}Xj&L!V-G+7qWYUzX^V?_>1w)X
zVg0t-jvuXSpu_3;uxmPCy4w#e0+jjWd+&nx1+<QKx(g$YLmOo6S&UnprTT+4F!WPu
z2=Xmr5i$~}uLqP!9ud$A>RQCBYgHCJh5aDf82YOe`|!`!BZ|XQ+L2)+?H2=BFb)Q8
z_1+h=4I%i%#0QLY9&RD3&w7z)J5as74-0MgSLYU?Zv`R`$=(wu$u6S+6z%+>{?}Pq
zeZxyl{jcTWdKYn{5h9Kdw|H_H@7zr3(5g;F8S=5B&8p7t)2?C(I$26WM?N1)Keb8f
zUN~vA`ylOznfu*~1~%mAPD~VBOU%D>`$JedYy6vaX?)2M3nnw1wq-~DqwQmP>iVJ2
z0EI!Li7&r)=o@2Bn+7%v6qDM2PTdi{XxhK}n;D_bs_U~nOnFd;gr70R78yQ?2vu<$
z_g*0mWe<Mj2qC{2(-W&o|9RE+6Lr$9GIl>K0uO%GsZ64>9*{%NfBks-7kNPg3V$_7
zH3Ms^L~Q>spXPr5q&?+wOg)4z^rz6`N7?3;5sdsRJr+4*K*lePAIg<nOwb_XtLYEY
zB+bVAE^UG3Dma#Td*fL@+@~uEn6)uCcEe%P^*k>D*B*}GB6XaT!!!z=t&F9urws%R
zL<EX)I)n{|O@BJ?{eIQF6^8>?cM^y88^9Du-`{zS6a(mBEaxR>%)nmEMgW`YRv2+k
zUFhCq>Vn_kT%9{?j4Vac1(>6$FH!Olc@LSMk_q;2Iuxh><=N0G9GLUxx?{e;tOxN%
z2ugY*ED%?CSQ=}Y2^M<SKNKsZ$qH-Q=W;u#g_x$pD2BrX9DSPHh4k4bhoTS+NN4fw
zj*fOLzrM>ZS-xliVmwO_7*OB&YRDjJ%%}spc*qsbYy4d}LeJgfYsgmo?Hj<@P0ONr
z6vNWv;{y$1fm&pu@omJv7sZRVKKn=GCuEI*dN!L}fx3D@sh7a<mGX!HMZ=>CAe{os
zUZ4jdgy4W--c1;BuK-EFU7L0UJy!7rfRY(EbPC}7-U)YydJ;K0V{K7auzBFs%lU}q
zpUt7{f}MIPD-Icq6%oFz@~c2;^a4P){l;YnmkRvrws?`-^ev}C0fux|4`cryNXEhg
zfs#e(zE*)9*Fd_44}NN4U_$7u9nqa0-P#2tc!eG=7&%yOM4H({NC1}V-D1FZ;OO+i
zyB-9bPFB*~nlCl`ZMX<}|2nE!x4fmP;1RH-nju+^O?%lc=o{g#;ph6~Z!!q!+2-E?
zJ#R^ZjveX{j_%xvzwup`;B1<f0TY_E6pSTkb+%H3v1~CRroS2;ZhD$gm0Ag=@-uT$
zF)gm?75bkGTjNd&7#L{G2qSj-@F1Z^o32{_S2|O_SZv(tL*kbW(dOu`&J5KvP_Da;
zL(kWv5hcEQEz~fylyjnhV!UapLQW`t-fyPl-K1k{EQ+)z$5){fF&NN1RKK7g81cp0
zW*)5XxMd6e(t~Fsj!uUXbj;SH3_w+-N(v5ov`79W>s>w+F2nI_Sbm(e-wcvC;SE{1
zMoRlaA7o#p;a7Ragsvm$n`~4Tall*_u5(X!Ut;e>`)*NV@^1>`*M4-6cPbr7ITmgg
zWeF)$h4_)YaKUuKUwyeFBeq)mk(@3axQ#2BN6fZ<)PX{vx*EcPyVO#ok71LBS`xF-
zqz1C!Ob8NSQ=cM9Si953#OF0z7^f?kqyD#c1YUvK(;&x_IosEtg6A0Fm3obK!v^Hn
zg>C)O1USq_50pQ1b8PWwEGb|vKOCAX65TNCJ%i3Qd3mK8<2D&PM^;JL58s9RzJcz-
zd=1%7puN!+<}ZrGruQBI!4^M}`0z!bWf#kdA8F$1XwZ99L9Rtc3uf17dTLyuxa3cb
znVc4h+lC^H2UHm4)-HZGjwML)tG3w7wFZAC@*6;pG1^!?Q38&tEfOY6cn>fPJt|O1
z>?+lf7?UuHB)!a)2?db^vSB33rx4*^1YDi%J`oEdzjzZpoe%y>k9m2CwcB!ms;sXv
z3F23kV82!qQ+RY+_A!;ml7AVZgsnzg;Ji^+RTT%!`(#k0pE9Vt5KeaG`-p9t?19x+
z;3yMH8b{Z~^zKyC#|GYb4g}*8GND_x#3H|Q)=I;U%=ShBGXt$%Ux1n-Ov?s+uFG*A
zh6CrqppAc##OT){#(A|*Qwb=I;YT@Ex4B1CeOMI6KXzN4*I~tUwoq&Lv~oOsjXL76
z=zgZ%z7K0*@JYV{wHx`~q_F1|J`Dcrv^cqDKEyCKD55Z$!|O&d0Ii_%suu+)5gmUA
z#9xF56W;fE;9Yru^AWpR_N=S929zezTT{pe5Zwd?YZyoDl6`#IFN(i59IXP*;9psj
zKDQu{5bXbO5=Td%HIp#4N5%6^3+e|H&{No5^I02V<U_Dn#+pc}80pu6%~G{<zl|2`
ze_xX`Vvtk)ah1|@1YxiBxHs4&|1a_L9<PDBfmnJ}nw-OB=;c+ttD?}6Gbrl<Hs4fC
zCY`mek?PY3A*L=(#YDLx($3{#JWSc`QYieTrEN7;86=(c<b!y=J2;?85u_`l8_Ai@
zgLcK%W`Xz=t+~9^wjJ%1`1)p*znh6u!XCC2_WFZXAar`Jy!3b?wnux0kU?OpHH)zu
zQSceRK6WKRiU={~2BOlXZ!%Hnagaug^5tS$CRy}e_e0mM?6P7v99P~C5e;zdZ&9Jr
z1dmBb^4H^w<mp@74+jgO*tOeSygF7PXGPadMm7_5@fsIaSEC{{;OTVE!4(z8Uk{qq
zBO<lYB3!v8&;##S66>Sk)u&Cuqrpt>|8o@)&IFJv#&Zshn<^?{Ee1AltU|pZ#-o!~
z%bA7AJ7azd(fTxj^{!?oz$+DPJ#^<UknK$HfaPoC?7!GXps4)`b0Gk1b?CPYWth!T
z$EK>Mkw()U;I@^zn;sY!_%2PAJ|^>tf!bxE$<&rcL^`=C_<w?pLcb;GRdtiv*+nkM
zLrvo$2{KXd6Zn~2z5)X=KAg*ToX<sf_I`Vf&OuXviB_M(Gr%qUY$vy{Js2jcY-s$2
z$Pqe>kv-?KPr(yZ>4wsZ8%Wh1f>?LyAPu|D6Ae%TU#Xrws8EChn`KZ#EW5#mxzA2G
zJL3+zRJy)xAj7r;{)=$lE<cgAv5Yj{c%&E-O^0oR*<FZ7wCGZ7v(44N=pgX0KKPWz
zJHgbZ$5*vg7qK|Y8Zd9)Io=cK3qA><H(Vt*^qF$v%2M$-c`Hot7qVIJd$-FVe4%S)
z=!zR{7GkKfpJl-wAAKDX>{JLxH$^Ib)Bi|(xu<JBfVd{2{as=W6@MFmmnxZ|-%t|(
zT@oxC-4rV34Nzlc7i7ISYEN5Gomk*85>9FOtDiAG_jrTA^H4v&z_<g`9vbTWpAq+P
zrG*P<y^+7S-_vR_WihNi?wJm=4jwpKAh6x}^q@vus)&~C@T@e%pW(BDlnZHN&=ZGk
zw%X!c@H&b3R*(v{;*(+u#Z`IyRY_Y+s!;e|_xvRx9G1D!oS@rQeenMgb(T?4hU?d#
zVF+mu=@g`-yF(;Ix;q3xr4i{GN<d0VdO#2X>6RWE3F%Hjx?B3a`JZ*pD<AxV3m0pi
z=f3WJ?Y)2D7_`e_1KOw*=!N+jy|-(nXy_w+ECri{^83;I&*_H*jI?SQ+Po6yYF<wM
zYiP~qvYjZlWUV$)XwsT9PP(oClBw6#mFl|qq?gLCkpG~BywGj4Yw_J=)~)vCO|5pt
zdrwp)yI-G5&sO74$<c7aQr*{8zVT!qyO<4+?;$X}*ZceyrM(yZBaZec6vey{GM?x2
zidIIo+L2tn?u<no>kE?UZYEO=RE?DRT1mVP%|SLThp~(vq@kEcv`veT8q+-f#U`{)
zMJ4`6C}qLzh(VX#MmK%A<R&$!lsb$T5%jkL?VG)LVGQqx#_5h~1W1xmTtHukbiPZW
zpD{a(`$)Tb7MHp8NRR~DICRGlBQ66zTlIxMK!Y)Q6X2hew9Vk6D);oK@cEXDGS75Z
zrwJjX*aGED$yQ0ODAaKp@5W=pn)(%qKMBGHM1O&QoSKFTQ8KTJt>*cN`v-6|#ppTG
zH4l0n$Ou`x|A0bcdcK$b=_FnoUqh>CMaw|x<-`8(LC<<QT$q?CHPa!>P$y^5c~|9g
z@xE=X(JW8mb@DqQ5{mEY^PeaUy)`JiF{S|Gl({<U9^FkScU<T||B%-ih_a-F0WWf_
zP@vMU8gu>#;=^I8=NFH4>IOm$HeR1?b5lDe=)XPg68Ee-7nN)vPiy)+Nk&rS`Y##%
zGma&Cmkh+CFb=LlBu$%>l}QSnjr6nJrNM|C8vK(*glF`1;aa~#x^Bfcvh0T#A$E<9
z@XGOXe2Ii1TC&#jx85Gz9$9jPS5NVN{o0CXUF#!i=~6T`EZ=Yg@zYW8o1{-(ZujiG
z0(dp{-J~A<DkKF3;wp#p&zg>`i@Lr85~^Te4_Pz-=ELwF&Mr%C1xmRN7ET+bt^?ik
ziskZ`%oKxABwh@VHh3^P)5nUl7NE<WfwXE6?mJK^QB9lN4(0DYXPIGYR+9_u^l<J>
zoBf?3R_%<xkGsYuTorg|5)N)s@H+~Y8_OW>*o63^6r7NNQ~YEdirC;IHzF&qvVP*C
z9vQBe3iS;NXFALmm^y+mu24+s3$-!>M{`3G#Sy~mpcnLa**{StdB#{;2V%E}4&v4C
zq`U04e3fK+ly_WdSnE`hu8%4;1Ugh7z1t~qgPz0^ZWzaDv8;Y`OPJV~`Qdy?UXS7I
zMvmjmmp-&a_i}eQ^s2v@LEi})IoNo7`)Q}1O>bz=sFb1D?3(N7x|><Emo<Z>T*LC|
zjVm=;tz+Tmodq?iCvbfzq9(4RD^x@Cc%|S2M`c%!!E?foZ<KfC99vZ7ju&%_G|ttd
zn%A+KQY9d8lGU1dcW!gP&3Xat#=lB4R~J6G+Emd}oB5UnU3H}IJTc48>=<+X9bTLN
zHR{0`qA{mjo5q(^bz81{sV1TK->Vk_{Tf0Le)b_-@9CWuR;hpLs~2;B>4k(QRk-`O
ze;_z?I!-^w#(p)pS~mExf?l@C(tx;`y3&N0rX_9$2~FR8>Squ9hbAn{{MW>sjaJYu
zx!=xdB~{#$7s%U@SN(+yFDR|B2W=2L{MtUllkoHLijDQ@q(@CC!E@F&V9c7!nmUQ;
zLVk>@fBT}v@8$vZ5vT*<h;M)LYN63{1|QMd6&^Jz>3gXT_{PCSieCuD4&b;qQItXt
z$q@H2qsyD~#cEo?$iw*RKaPovQmU4aDgqV4YG=D1%jK<F*}|x_#CBVWUPyWJD*m8{
zJ&irB^z-qx;wrtvy8Ho4`qeKt^%hub?&U*k0|4Fo&MIn9^?mqgH+NF>IitAg(oJha
z(4DwbVf;VKc`js$ICP1Z;)ayVUd#btu!p)lBTp~Au7F)IB<s1@q39_QzGd(z@8(S(
z$^pG9)fO*`$DinWek>AVIqV1+2HunF>qn7=$)jGyW>Uu+Z|2DM?>h<y7K=$D52)bJ
zu0g(0&3pwzzv7;r?NN*KJNuCA%j2iVVoGCgyhL8-l`1hum|Z$0x<&<AYy?Funh0}m
zc9UI#iOU9qTkgw~6EBF7WdGcs&_kT1J8<RG0P%W(8q=mlkDv0H!F5<CpJ-2_?;whf
z)0%aW4u@G%i3+*U7Jl_E<=BWP7r_(R3cu&38)ct-4V1`MsF{0?hE{X^>+I3fNQ-*L
z&WcJ$F-geY3y3aWnMgb}W-9jq6SZ%8221Y^s>_A$8VVJ+vpOmjNkX1DAth;Rx(T6l
zuTz<{y!F44asL#1(H_WAD@Pi`QS67QKM4|-rccw)pYEE}BnRHfUltIA1$-ddn#jH1
zyb-p@?--@s#NkP$#>Gf};c-<!$!80W38I7>c|N>e%+wb4M$a{KW85#dtc}am)%s)0
zC$M%5lz*0$EkZ-8&PCtnb(8pQDDOIj$(IE75yRTY)6?3GK8l@h?&gWh|7p&+WsPYy
zyw41!)*?qzNWtOf{T?>74W*b@?!v|w{xx3^LVIpldvnh=lgxDX?9Ne;o@HpHd|B_E
z`lKjwxaLsG#z<e#m1%}fHRzn}lRUNDImo3<Uac;Zi%)p;(!rOPF|)td(+gjQvW6jy
z=yKt>e}%n7#+l$t%uXmG>l@RpU2+7Gu37^6f4&-RvXG(s0b58Ub{ji^caLezVGt{+
zz3h19mC$Nd5Z-%QeF9B4i9YlL^r^?oTL(1QKk`O0iq9;c@&$oj?hku)>oy&Vu3KZ}
z{aE^)<s@7cRd2}yRPaGmyXK=+<>q|3kkrBBiYsV$h$hr$7$}mrc@w=T4-TatG>=u$
zjlBJS58$@_PU*p35V-U1_tfi-{cvW)5F~S|AK;*ltAA5Oe&I#6W<;W(3!0<{w#i#p
zN!MFnV8EZ%$im9923`1PlO^X5T&?nAd^+eoFyfsdw%Td<wU0dS*ct|O+%Av!e}!d{
zU45inb4>AE0tw?;!GTD0q_z1(XF$vkyph$-$2&|Yzd8*ULvQ!MPGO?!0&r+u0<!+_
z${qU`Z7RA=TcLQ!0U_NuEToyhlVH`L24Tq~Nrm*@D#&{aRv*!qFSs8xg4UrK{0Lu&
zip8FPD;nwN_@K9m^XRgh5tqsad1Y+d4HFKOMLRr?*z0M14xr#S7B1*m1L@nq3JFmw
zgRBSv)e|s-Vr!<t+@`^{BV%x?4i%ffKQ17ON*OJ?hEz;MR|HIF2yo&Y<G*T+Po;V6
zv-bB{em~^JD2D3GiA9lSJYp{xL`?Od1!5Y|O}JSh68Zun9U-UK(_53sL1@c~!c4o3
zgJiRaRLwWy;^jig_?T?PSE_m0K=uzYHbzsXo`8{Jk1ef)c*0X=!)s~LiBAIQy}gYG
zDgqIFGeEY+q#X}v4ty+MtvRsD%upIiwosf{$n>AvJ)dOpM))Me8C7Nf{jz{303pb9
zzaztKjNr%mE|pt;g~Ffea~+;AQjSf`bv>3Wp2+RAzYw$VHO}#ioH(}FUnzc7OBwkl
z14Wr87|V7|18eT{H!~8SpkdTBFQ(e{1LC745V`B=xzzSgsqpN%dS*;FX?K<`*8EvI
zCO#edF^&CD3Evgf*dq63Jw$#?nEJ_jMZFd#LRrswX--2dekBacH==q3^GM>^M?o!?
z1>1@HFS=X)EO^G=HqJM+3TUG+%YJCSms_b$?cYb%ti2^~rXi*V|ET$>^WRwDnj>iY
zP>$G^nn*x=;SWUCvs04r`mm@!qV1<MM$ZBsW^UTg_E*;)%-84wp~oe+3VJXs<W*)f
z8#Y}Bj7qi=GfeQyO%Duyzj1ji0d}Dux~bg>RP}|4xeJ=^1HSlarFzEBX!F2zUNVr<
z7QCCls<n~w2Jv`$4L$u!)OyQPxa1El%JuQbb59HX|Eyxfr#q!3NjzT9T^~~+@#uV4
z&wEoS$fcH6ezKRPEv$^P)(mp(zs^Vfj;c2NmFWe4>3d{Ci_qVm$12baOY7a*wtHfg
zPyEhELW-m6yI6(4HSO4T_jIT@dL=~DUkN(A?IyhwPux$(^|S!cu6ndj>ZU`vQ}hwL
zU@>UIxte#IwZAXQslU6u+Uy`wtBNeBn1$?%VbiJ8Sn^{r6b>d3$K(c5dm+j0ixKQ`
zitUb7xkIV}=DU9Ni_u01si~8GobQ73<sG>DPx_USxoj8L+U$xye%z|#!wY+LNVI|&
z4XWfl&RW(46GK@k^l@1@^xs1~;UPa+Yssj{IMEh2mBe;h<x<KJj2gbn!w;Epy|^^I
zuM$_$%ii-tc|(0>P-}bR&eE%4r6?r&u}}C2vP@bVi@=rHcwaG8_O~Z2kVgaE`~9Y4
zi_<Y-zDejqo5@v4K2h}Sp&D6;`%Oa>%vLcf`Szf&>!CS&yTGNHObC}GQ32XS=<0$G
z?zop9o~Ou^y0EQi;!VAS|FxY1UTqN%5t)-{Yz1wya-ERLb4IG~^AcLLUuR1Fr5UAs
z)F6+NZYtdT#dolLJppn}vn%mTrQy4qaFmTohN{RU{=(n~qsO|31Be15*45lP3w%LD
z)-Gd(TWBUb>+H1<@ql;Krv+cFo;Z`o#Z@108?jAT{wHOt;fAqi)p4k}TI?0|v9tG(
zw1io|p5*J@#bC&_^X(1eqS-PKl>c5oRvdZuM}evwZ%Hntd>uAnWZtbg?G>-#^meJJ
z@;AMH@wryn7b#GLezYPnzHS8dIGs_Sw5G_vo@cC_(+PT0fgzjs)CHF9n)-1;O0*TM
zD{ZYs6{0BQFlaXX_~-$XYZg`VsE46CS|>AlT%Zk)Mm!n7wPiqK#|i2Jn$jL{7jA&J
ze(lSNLtyfkaH!=TFOHfWSE{;!$_Bb&(T#%&=c1HTw=|3@&qgUgXLY6w;WpNRR1^qu
z6)pVqg}7{UZ#n}!ge`_-W4cYDS&Jkt;2guMI*Vm+$kzkLmU1gp{x*UFtPf)v3c~;b
zwocwS`(1^y9^fc&8TD%2_~}MdA^Xp1JeQ)V6X!`8%!Uzf7zXqoac`6uR=*VM^chiP
zuKP_-lC?gVspdfo_E3fuhQcK~g2+PuzXnyGISU6nzA73x*OjoFNJC79@hx#O1nnt2
zq5B`Zhrk|rQ`5;E^ZPPjm<<7d&}Mc7=Q8J`nJDt5*YI)BJXOduKZDCWy9jEDREJc6
zrB5Qr9k4!C0i=oc)r#`}T`gKDSZ2?-gaW`+k|O6K(pSTs!ube2wUR+GRA9x6&2nw`
z=8;Cdxgah{kL8_8PtPNgAPi#?uP@iq+o5Pn%XCJwRSb`zaG~<yuN0%-GlZA&JT5Gv
z7zNv;EC-bx#+lkgK6lnWWrF#b9P#-R;YZfE$&J-~t9eK0VLH-aUt4sz%;h^I&=!RK
zb-pga`-T`5o=?eowa7GcJU14W<t=p=EVCm9bY|xP1w{4}yiv!`r{f`&)&%hH)n;Dt
znuFq8^d)W6d_Al|F7Nx|S*{SF49gYoSxk;$iS`%Wy++!Iq}#e!ENe-%hL6TC#0Ixy
z(D)EFXbURB{|J6a+~}y}Y5dB6`QKMHk`%Yc+PX7U*VP~bDVk(QSS;qHqTnXMAms;{
zY5X!8b5L$NSoVS=?Vu2RzG-*m<kJr!OL8LRU*-$-xB@#Ro6PZPmKMNTVKc+G4xKfW
zmW<2N`o@E~(VRH)LhHOd2il1=`sMmqGyuiTV3;q(;LLKwoY|jS<Pr(Pf14JFP0INM
zZfD;6h>kTcQ1!mwc~bfz_pmMFz3zh^RBg+KNT|gk^}`$DL1Ij0?E-JQ^#m!B5K_TC
z#8QwC)e!sL(?ge%h4@sk*8<!1bxi^n2^)LAc~R=VU|4R}0F0f;;rOR~TKCl%2r>(Y
z6?l2**f-mZ^C_#IZ;pl{r}pH{_o(-kfonS-Agw8!NU_z!h@AD%p85YKVsoPAiewjp
zM<a*!<sp8y!`UIaj`s`(!HFDN)F6>$sYKyXIB6MzCl{dea^l_a=@1&SzB~&pKp>)M
zx_Zym8^jfUK~T_-X4a@zmL5Qhj0VDk=zV$42{0)<fEH)dKQ=2v90DZtFxY=^p_+a=
z!Dm`mwnj`k{-M^Eond8V?cj4kN0t`FFX($jp&2xJRSPcE3AlvtJX$mXRWy^A7G)wU
z()Wj&BG8~S66on2uFKSKev9o7HF_RzfI=WwG^APG``ntP(EZt!m|A?gI1K`LY-!yZ
z#e1!DVScZEp6<8xdt(Lhf@UJ1uC&@N3RI8s?Q$^0JjJ`0_g~;b_(|rP1&*LDiupz{
z+k3s`RkCF4$Sy{)xDWp{w2y9T2~CxRiejANH$UsQK+!5H!ZBpfux(#r?}v(hUJQnH
zq^&=8gXF2N>`?!=3xJm~pqLP&R&Q=d#xGEee%|8(XK$v}g$Y!B|MItaIO*N1xEqE6
zSGD`_U{Z>*i2uM<M)XcBjy6Br_SaTV%>!Ci4u*gZ;^%uUM2TkBX{y4X@8${TdNKC)
z2Te{JU-tB87kSfuo%@<Zlm?e&dHmBAdljb{UcoG-^h-e~#=~Llvw9v)uA6Aa8Rx(J
zKp{v_pWv<T?;XngSa|{34EY9hgu=6SC>mVbHr-c}5o~2Dx+Jq!o%gRC6T3tGerz<=
zvh23?_0I1Ye{<Vae3yhbJO$tVhkwO-<qKmJ4BM+0!V6BYzs`8+cOCsw*H?vcvG`jS
zwz{j$`9gx{x^J+kUN$gM-hUA+QhcsE!&V8KI9@hVx}IA>UIAp&KYWw=OSHLolR*|!
z*Me=A-y#Z?C`>NJCe{)`C#p5tN%e`vwOeGgPE5r{wco)j&(tV>0pP|@w<lx1aGC8j
zh?cqggYd&80`K7g3~6M7*C3l4Xu(2=wp9GLH(x6cPY6S3{sYmg%Az0Gg4oc&PW&Tq
zKnyWw`p5uVK#JRswqM~TK)^+m{kmVGe4T+Xp}_Bn@-k9?Fs95g;23G8-89}sH3i{O
zDui^}YaMLqp+64H#=5{Npg2J2hYY!89S$T2JcEBN7#js+YPe8D3j_xz^YaIC9KFB<
zbP<qXt84?wGHH~ecUiwwhJ6S*L|Kvd=ttgpo&KZT*V(-J6TtkHl5k41R7k1>;Dv0k
z8?gz<`tx%ZqQ&-p9y!w#HHm+p{Gt8EZ^PtKUVo>=u%eSG<)hQifB82c>Hh*%oHUVv
zX1<B@?;+XiuNb9=_ZDP|90C2}a=M_pyFfJTstc(p!(FU4K>ZE|b%mX{^fs^ueD5C8
z!rP)=R9FJK{l=ga!!(xM?(DlGkM%jYaK?AbRexaBnGr(F^NjtQC*I|Gbi8!meRKsi
zZvsV#xIGd3n%pmDsbR>M{$1p6-iChgpmShrs`JEK;+d(Ye%Ho6w9zfGLB7+|7~<Nv
zZMx_#=BeqIfc~QrJy5Kc2y@CADj3g{Yt@#I!dlSw<`RRx%rblsV)_^QF3IOSN>Xgb
zb+dGv5yhjEr0q3#m;CF6fafy1^*)eNc+)hQs}m*d6Po^kuj+@U6{F~9jn`C(B{#pw
zs@*c(to=*G$o%9`q~EuF-GelGx0t@P0zwP(uVU3@5tsR3`_yVS0~98EH?N3cQc}IJ
z_xo{N5Bh&YBA$YGEu?KUbU_vk1u2Xt@z+=cUa{|;peg2FN;dVE6HxOY^CXpLu%Qyu
z-rLJcDkS!KaB6k_+P8NU`aj#Q2Di%1yR;GntOL7nl3mxEP^}j==l^_c6S5XVKcF9o
z+*Z*1g<qI|f^ii(cKsRypJ$?Q*(xIc!;Rr?mLiLkn&maeO=9fxo%sm%hPYrD-nahQ
zE=`*xnv%9c>Hism@}McUEi4(dba_*wGtE@csqoXj3^DhYee`F*elUF+DXyDT-(^VU
z#7&3unUQzf!1@Y1skDB~{%7M>ix4h+iEw*!N~o14oX976HsRR;fT5QXguTgr?AJ&%
z-0RJMk~1<@3PoF*FO7Ivw+zMmOlS}(lc>ppA8x@%*b>X2VAHP+hWb&#6RFD$2K$vU
zI~ZZw4+u&!)WfmyP+seRlk6b|4w*D031w{M@87_U;5Y+>ME+BUz#oF9bvgnT9S`om
zr`ueR%>MSE9p62{V0`UjOVb`vn&jakw<z&;h)(Ok0raKkgLaOw$~S7eknnh3-Q079
zu{AlX582Fle)ntf`rVq@;MO(>N3z-G&{r5~*%K^d@>RlBM#NmL;~V57MS2vJw)Y_#
zvbztD<*3No|LKi~bRTVf>(3V4f9+?NNK+JG?6sLYgYO4meWe38?OCRUovG86Z$`mI
zzJwaV{hiSUmUxrF2>gnM?1^Zof?K1%B)W!LgD}#fzs@(f=K;cq6jQf)*^_Pjlxzni
zI%Mc4bs-N}41Z<S`aTqlNtB*HkZAVJ71VcOwJ%HeGA0?z0F*3=>*h!ZH{J|!72|Mi
zQO~@?TyN&Rvqlm(x!j2<1x}l>#r3+Ye-9pzl*Fm;86?uc@>Tbn94f+J$blZRzaWRr
zax)h(CreI<9+^syN&Gj0My^xhtE!2EXe;&E+BE%lQ+T$!sUN4+;xRWPtt-0_uJ9Tk
za+wC2%&T6A2U|XTmJp6hsX@n)Arpc|=p6J+Xmf~(a8DQS89IHxzcP1oMNPVo9d;p0
zZ)OCf@oRd9cK?)|RhgLOLb{K(OH^@{_$xOk_v_m#)&$oTmZr-3TQwND9?rOQj$G+6
z{uUujYM;iFdR(|r%4&{3^&y#3Ej2WZ6sJBUQc;iV>zv7@(XJ@i>Lj?{J(=+x6(F91
zM=bcKjg(TThDVx+d`_2{K9>SVQ@Q~j7J8cFuPvPfpSEy9=Ts5Y<lM?sqRek5I~k~>
zWz@rvWjArr6RpPgXuHiT=vcddE8WE+RgY<UdcIXDp=({P>ai}>Q=<Lg{?Eh1Qv{u6
zYV^d2wEX!GQGp)ESM==HqmIqco{~3z%SFel?G6zAUSF!2iXGj@1H|Rb7W5mw)cF$)
zRJ_U%Gdu$KRN<~pk6%Vn6atN?>mPL>rX_~U3&WX~!p@~%Hi)G=`t_zM`-Vs1!i>s>
zYGS+mQT0@7FPtxzLH_Fdca=0O81yrb|E<rS3j{^Fh4Z1^tv@&(P>~GqNL37>Vbr{O
zM~6={!ep;3q2_cp_&Tk=07!HK`V${S@B;UkVa01j&ool6N8;+>Q6bYv3;pFpnD$tT
zY1J@=0{KpVI9%U{3e=>poPay=sE(e=DjCR~>ww4R^$|5rydV2cay7O-7eQ`#xo_tu
z;$VC}NOw9XLHJfbCtHp|bKz0teaHICL#JM|*U$cxwc|y+$N^_XLWkfpPtEbQ>1>fg
zGl(fd)v7K4>J32HsPY={mV0mp?$SFFsb+i+dWsmZU-Cwr=hs1zees!o%at>S*Qs^C
zyw>|i%Dm7g3?pO3TggO0usq6B#I#Gj>?^Y9518DF%b8bZA1y`-7jQlN7uD71xZDvw
z)6mw3K?aA4BP4xA|15nXI^bL;it_f*oo{je3S)c;<K-uq;}FzI*+7|N>DMw~y(3Ix
z3tuDbg8WjTRp_w9=HU36eX&o`q|~G)QjROsyqp`6AIVeKOTfqdE}!lp{3(M<gLBwa
zauC(fW3JyzoIE*ttd{Bvo}1HER{;(A{m2cY`c3>9-^C1>1J04&*UAO-=!+1_<9MSZ
za_4iuQ%0`B?+#WtW63V<1S4e9B9fzC=cACsbOmE(xFm3B$6Xa|%-0#a3q>oUs5`f^
z@#Vjsrx4Q6jo?nH<^3{2<Ez|fc(RvPwSA4yKKV2I=wVq3CBq8|`%Gf#Cb21N&$yX#
z>|8UK=bZ{ggQGAw^zQ(bL~~$fw);Z9ljAwl*9B<8K(ha@t~Cev!1MTMtDB0~tXJ-Q
zyUnYH@0R9)sGN+apq_C~qS-ah@W?~&y>G$Be^gtGXRwMC11lJ{<uiqBar1dOL30(R
zGObKUaCH1E)G{eGU~pgHszk_rW4*j|_uBn{JM!+jx1|n!r{p-fThlc&P~93d=4x<c
z?X*+$G@=~HS7YUTNd|wbj9&*ER@-#?mVt_sQCvQUtfzY(2dx;k{f!Vt3z+5;KowY)
zZ0pK>Vf{N@5nu8johOe7#HB#IU=K+HR*Byotzax3r#gvqvT#@BIix;@F8BD_>f}$G
z<0Dr?a-L5Mw7|ptEEb3c-o3gy6ny4hX|+?DU4lceIh5fN1d(<B%7D?lp{gAd+_;L&
zUs{1h&B<Pa&jF;(@8)zeHN~lxSTAv_zZh>Wm?*#@qpRBM)W-Ui=*Tr6vO%2hlx!OS
z;C`uCgN3K@xrgf`-+XVhgHI&;*=TdmTTq6cusb(iAN93~M7-VOqKIi_zXq2XJww-a
zI|{Hz@6(6mr$v3uAEb2PhCKGb;&CqAx%~-jXB6$6(vzVyy+ZiWZXero{je6C8!5#_
z){-qZ(j1!)I|MmL_(lr%7JrIounFZt^(Y>MC0P;Subp!_j9~8l;p;)?-7VLpAAnrp
zp(IB!qm?%ITN#XBaah`wGj97;oaEsXHBXGJLT8e#LIRUAalMIdxo&nzpIQ|qKVCC?
zf{BY~(LjrG#zvxwdO+04JStJ{{P&eWhuPWS;#r(xnfV`!ZI=+2^`SFU>bDrM<2?#z
zv|X~HB)4>6BVY7WNk%7=AH6hT@rlM@wg0Efb>?@-lgNEgMhy*eY^AJx+l?k6V6Ejq
zrt40whqofh>!Q3nk#_ZSW92;A@2=aI#I3_y=V;UhYwc6OzXPKjr5LBrpY9AAdZ`7{
zha?kaZ6J#-1}n7R$M&v8u|gm468JAn8NFNgl*s=Kee=B2yFKRTU3@C7TEc7Yc$+ru
zq<g0dZ`mfvZ||tUhO^4i33E^I0}I8Qvel%%skb7IyqQVC-6t<q>kR><-a5|h&>V}b
z?-dDhtx_UJ?b~Pb{xWhoG)G8TUe)#s%xM9(pM{kF^``vD325>j=r+w`T;1gpj$$yq
z_P@wIer~U8<?_<dg4pa|-HsPJWE@}GNVui^<j()@DzhNhSjaQcU=S#orD^QRAAj}s
z^7@w0bS0seVTAfF{h(t;c+g8+6`hL4?NnS2FNU9?S*@FNlA7R#>3zyci?K;F1QUO-
zRZ`tW{ryz^6U9l%N|BXs8iT}`+vXMDaB&u&7x?VfeSc4w7v{PJW|K9h=|a8I_mxe?
z_<~gBo>-Oh;^8Dc0ZDG`KVKc|cfgq{qZl~3S?<7w(dPL@%O}$63((m;3WR~30?#E&
zdmt~&3IFg;lRBHS#8kEy*RV(AB~H1(SmaPqLlh70d6&KXmj(8-HO?zx-RBtWdG<3^
zf4+40dC*1^MAm~#K@RY6+JxfYQ%~ywg0aeLDD(P$lq&!<Evllny8^8|&E1)W{>x8z
z86B1xI_lAbLL29=%hW%S@Xyw+2cf954~rXFo8f$BOP1&9CN7t{DzbQ~qO!^qo0Qb;
zTD$!d*bN#?LvE4uP;~YF`KBJLBole_sWnWJXNMWD9#!EIQU>%3mVq|-z!G&`v72XL
zUl?UDzW0jw*~gv@_olxvQG%$i!qa%@AGj(!kF6hau+I)ljVx5Y!H053!R$?&V}M9q
z#Df{3k1M;FYua0iA0fzqGPB~e^*EF&`+V=8A^+ZE`i6S9^n9!=e)R!<f)|gt`k2Ow
znN=uTcdx6F9YK^LFJW{7h=dzQqFsas{(kJ5ZsV5ijL3m;*NGRN{qEc+H{6!Chg1bI
z?4Q4=7Ep_BXY`h0n|cPEajrW0gueXrZm?+mJV4_r^VzJbql%{%xo{Mp<AWzp)R!Me
z^{oh)24l?!97Z1gtWy&WY|+rIf0e<?B9g8vQ*8K0sqZc~l8~goi~h^36Ka0^jyDa{
z#(=V>sz~BpGIjr~?<9xz<z!dbI+-OTn*G+Ad^Nfus^;{qf+v1dj=CgmSoDLDD>LPA
z$Sbh~3KR$zGO5mep~If-MfOPyZk6qn^o+=c_{_xVlNsk}5$%eP{!Ywwu0|+RDa=-P
zSMi<0nP>f;v4IUPo~B@jvIt?47h60&a!C-ncs3*P3THLCF6FG>r{rbKFG#+%O0x2%
zAaeTI_r7cI{VeLDrk&<1-I<2hk4Zgf&CMMdj^9fkNQ7WP<3F7JQ(*awC~xu8&v?`<
z>3t{B(A#+9fHijM$niPgH+0#RjkS12eK*BXYs8t^Dd)xJ2hm*o`j1!r5thHs+1~AU
z?|X;S)7u@ksj<E}Kud7h&DCyPE(}}mefGbdjQ}0x6f03O+4jt7y=gC-nfynG-)+ZB
zBPwnZddyj8l#Y#mG`E+{!L6z4_fysR%W&XT-0o4M&91hoeHS>j1)eEV3);CZ%3UWQ
zK`@)Yd75+ADp|kdr2$!ZTbs4<?M-tpNrt%gwzjzpF;j*oy#?p;g@8-@cR{2)c%a6~
zdbuyq@*g5~Al^5R;P!y)Zj6$JySe<K3ly<GU+IcEPGDDMg#`ByB}}sDUx%<dBIu+{
z(fZA%9;{*Ft0Rx~66Zmnv4>*;1lM}3MpwK9ol)VJ_tN{f(R>XC`=w6r5Vt#UrArPj
zs*iz4pHvROUg*$qGZ@UziZhvK3*Lc&^2NL3fBNW<u*o&h$)e>->6Y*ko#4ycXEI&M
z(6eW;9ottfpqbw(;shXm-p^+Uqhx)<*!tLh-+HVKESv`%fag<wZRih$Cx(S<?GJ8I
zKwcp_Y%>mfmqRH}j)ASfJDTE#Sxwd+-%k!=3LnueG5!_tyL=Y3A4>zc+1T38<ZTOw
z#S#;-`mG;@RNbqB5E6BdBT)Hzm~P(?usTzawcqg!V<U?M7Qt^f^;s2oGIy6gKY06A
zD9Z2V(zz)xg*;0&=QZ!d)8PiLZ82=Ct>K)|uJ@lRvT><g$230N3les_UE+V5K}F(R
zX9);X$cg16u`MTY>D~`RIUAfkPDAC)Of#cA*j#XY0SltK9>tQAB(^mhTlc)qW!>v3
z?-u1^z4#yBq@(LPS1@|&H*SyS#mv6#YY*>|@KCojt4kdV8G#GoCUfU7(M%zY=I%l!
zkU#7Of|(UV4<{Ks{r;d6*k>`(;uxMR7)RArgE%g0a%IHJySW!YlK2te>Tcr8gI0T8
z$z>?$K+0t#doid?$k^_xw%wkIfpq!Z^EKh_<Vh4}b!0%IlF+qG$bPfp?BK5I1<p>7
zs;8~<$D`x?rQA0mbQMpaqZ}xqsN1xL9g=s7BoL2Qal`%(c8tH6YpRcycSIn-CjC+|
zssq066#OfcRu1G6u@`m!;eS5BgA@*K74h7mpf(h{E6mK?lFdvsaqLofkrOxFePOh9
z8R6;HSSvGn+F^grRG_FeLS`TW1!A8nYy$;O(?6?{i$iC4-ToB)X>(#Z^gLU$tC{l4
zNVU}P9>yLP_z^Jz**iQ$^af^k$7mn_Tr2G7*h|cgMn_x@;i0=NI%_!2__>jydpIf@
ziSg&{Dl3REd{n<(!PkJoa!H#{pArxFCmrR9%Z5Z|{CtXilH|M_mCy-UiTplLkq`0O
zzge=M@m<lWqQ*eyluA14iuoQ$nm4?FM*GfgF8rhWQ+0*`{)LVp{IC79PidIgE!Z0`
z+aVxwr8$xByn6VkT(T!(N%~na=|a|yLY{`rekE&xmF>&TxeVU@RO#s$ak+p5ZLVR3
zV6^9+x9#WhKl|RZ7}N#655>(k3HqJ4oY-e2SGfGY6&e)RqZ(q(E@K5LjPwlak|?6N
zSHMT&QKSd(f!WWrkbc`>&+<u}!Jr?+qq1_>=;|;VgZN7+lxrf%lgEPkC!h?Stfe_>
zHU`SaI>I-b9-1DU+<#zAQy!MOKGwp};VJnEJ@_;gFRN^OsBJ(`_f(qeUc_9_76*d%
zw@m$4&t1Wl5fx$z&S;5lM-nI#f%yPTE3df(@|OT=I<Pz{MU=vZ=G4Xlv`*5%ay-d@
z<~i=&=hhI$rafu}V>y(t8DR)N2yq0jYod(hEYgMk2B>wR-5U|Gl9TABjUj~3Yu)|c
z?8};IG^M;Xj+hZAy92KU*1_=K$vi;-X`FfFaRQ>tac^JhP-V7$W{cDJ^tVvihZG*7
zVi?Op7!dx5Ss=c)xiB##t7$v(ZrR;yt;jrz^}2Db*5Pe^d~`kTISMq4`pffu#}Aw>
z`@nO0&d-5YheIto#%C$DmIT;~Y-;!zU1dwv7M)lQ^{X_DW>ii{=p5Ow!#vgkvk-f=
zIShcEqrL~t&CMPCg;ofNHLC9o|CQlC?+#3OJFLl1Gngs0$mc|4kRk5*9_`~h$XjSf
zXKnO`7&f+mut!(?8-(Y_b-P*IMUyE=zB1uyXHR_O?bNP2DRMQ_?~)ZdRq`e#vD|>(
zCarN_GsTr*<TL<-hUIl;P1-GR`U&`kUl*3EeIs8j;a^gR(L=bz(qY`mIGAP5%(?CF
z=JH1_5q++tBIx?fBN#XjB_@)MpmuW;#kjk|wUn#I?se!Bt7s3_)(+A`9L&>|?YvO$
zM2uiFd^fC7n6BP~TKoqqdxtqJKU3~ZNaqIB#0nJo{I2^2dZfQ3GKOOaWCx30pQ$&_
zr6xa)3b+fdchS#B{gN~@*eoMJ_Iaq}HBO`9PT~MzP^1Y-e+#4)z1E?Xm9X5uJs!2I
z-e|jgy$1{a?t$(fLCc|P;=NO$9#Y80Y<>C*vEL#YZYIu981W#CxM?BH&3@uL^vhuL
z)ie2Zim(TId&n8*&5x(ry~#dM8Lq3e0Rv_I(DpxA3qFiK2Dx{W8!o>kb!T$^VdVV`
zd1GI2vQCR>J|m$f{<1+KD(ikU|7J~m6({|@I((6$s72Xf_NzEKcN!^|fxcf%#F#@<
z6>W)<UTOA-ivQ9`HKZ}!%TO;0`tuBcAXU)kRK%`)qKKzlzXihO_8Xt&Zu>|3^fuQ)
zJya!fxN^<BVO9&Cv{q`V^3`T{>J22&69$Hvb<NAs^91i*s{}qfg2b|OrT<%<0wqrR
z+E)7P5`x#Qhtc}XrBoO*2T?}zm()pfyi#NKY!9TXq^(Ixb~+z!Dlwh4Mz~h=r0=?D
z;55<?$ab4l`R@pxa>v5L6cIj0inYVqHo`kA@JBy*<=mQBAdXfO-~Hcp<HGgEAaXmB
zw*|_6{X%n2K~6#TKvE$_)~ae@P$q`G!utMJ&Uzu^JHrt^i8jqKv0&{0nXl~O=DeJ>
zuji5g_epp^CZvBy4pM=htoo#axjOrU#Wc0O3MyxS$ZHzg)cGUGH>@);1Uzy7TfFMB
z#E&Bf0M{c53~j9OgwrAs><`D+!Z};6-^)-Ryav(T)D-&#J~4+Ou73Js)W}9=M32xw
z&oNhf@zdx*dOEz5)zh2S09a+ah}&YO{-_u{4iveDz14&$Mp23Iw#}Eokcy*uYU^G?
zyds~r25d8+JD8RGUdSZL=T7+k$gTPDQ$JkuIpkmz<U{BMw;Cr1G7Z$D?vu~iOqLXv
zw;mya0w&q?kA@X!S2?RJa$Eslu<^C8Gpvta?H<HEY5p6GX2Sx=3&7}Ivg+nD*X2J~
zxPVfv{Q{mBW}lC|YqJo!gXIt-RaHEIY)iZ%BYDxU>JkB8=X>+}36+l%Ls=wTL>`Fk
zT=LD0XWbwS8TZc2Lhz%^x`$8{N?A4O5MLJ@Y%OQVirogUI9I(sY_g09;CZgDuGU4m
znsHOTpK9{1<;LSP!G;xJ-0O*i6c5IhV$6V&D);Yn83MOl8*q=K`S!(aPcDJpTwXwl
z?It@QVQA>;*}re!e^2=jhrXHTpJBJH$qY{pB}xm&hX*y)@-i0RF%te_9xGIPw~ia;
zBqR!%GrZbNNtnsO&vn}^Fq!mre{8iNoQgu8BELU(C0#s|z;nX%<Giao;k}G)32m&U
z&6QSxyxQ#_Mh!BtJ=R@O%?#4TZyjFO?RIX<@JwxVc~!~7*9{9Hw?#7_3`;A|c@D8G
zozA5Ni3%HxWtktrYpYFDmB@>@9(;TA^W~#I56O=!Oi`u<PHWq?`(y1&2FUW;HBW}c
z{`4K)rQ~#JWVuoA()9m5q<c2eLi{^bRk6D^o8rxBUL>VM-cC2-tLQO_dK06RW`@YV
z#@S7bMpxpbmSGDv>*-iYLI<AqQvF0#iuU?E7fXhO(%YdSH)kbXA32dEL{2#ahPz!Y
zwW{ZTOO3)LaKpla=hmu3xerhn4gp`UK_`~m_<ZjSe;Sj4P+a|#T|F}7xFej1!$;U3
z=b+@R%8;j>EFEhshadEp)q3TRu5}EUqrP65xF@g&eRNz|z5{d?#VO_=cHn7EG=Uw)
z`prQ+<d{oNaDorN9`zUt+Rr4E{puqsi&9DE4gkFnxM@G9@;hWdGc(9XE{Q0)ive3<
zoioZ5odxbP`jAK<;fszQcMv7o@Y!qVknVKp9=(q@{;S=F`yiFe9*hNfsN;{Dn9p2X
zZkMkAfKs+Y%F{MKQ+^hq?*};Qx8Nu<<_fE=$5K7~H@x=h>E$Q367FKI@ij%5S_I_H
z+hYiBo&E999l*2wvY$yDUMpPw7a8A+LH&V1uiLZ>hdan<Ja+z^ndCIuZGb45*&iqc
zo-xYv2hlhb{9)V^6lkG*kxRma9@Kw4cZ$1&<<5nlSm^A#ZL1{6yE?V}jZ-onxN>N@
zaT|z(sVTUvRR}Hm@P(7G@BLiC)AN)ug+%-QjaBfNrY~!zah8?SWiC-rjLtr!**yYa
z71wku<;Q>4b!C^v03Ia`qVCE@zow&c5sP=}LiOVhW|7cOv__Q12U!&eEB(xmI%bPt
zmt!+8`tbT8V%DO0Roh8_`FzpeokF?ZpK8W-EDT?~a!YA|_qdzZ|MR$*;{+*#8d&|u
zya1yuCkQW3!d<CUm&QrS+5jqXBkbh0C>XM+#0jAbyWe8Y$86>VN2@dV6KL<3`Mwk{
z?2s63->W3#etbf)Q(<3OZbcokKAXqaC}uXxE*d_sf5{NQqcf4(L#D^B{5wtl_mNhP
zk)sw@V<;rh+IV&Mvo^h@jEd+ri|I)MyHn`;Qb~{FLx(4lBGx#~Uq&!Yo+}Q-<aNkL
z2{8Sw$I(vu@Pf#Qm1&|_p_)ngt0BI`GgkW<B7I*zPHGv~b-$P7dd3Qh-R0HV(IVYX
zy4+GF&Yp=b1u-Uwye30K4gFrk3q3Q)Q_jd1CZeR37cda9kvgP@?N5+;R_@|Cc?@92
zvwrpaCMPr-%N+xf<kcs5XJb28OoWhO&ETfc=@pu|hdPb2D7uFyN+sr5!-K3*r?wpv
z=@}0oyNc)#$3PXL7RX(miIain`_2d{X1LjD9M}X|Mb#^~%v{|XJ+fdQE$4&?Or~66
zjw235t)hJ4BU}w}nyp@bFD%>sy<#-<Z=3SmHjjM8-U==c)mWOhR69Ll;4*?%Y26W0
z(#u!E`CnIXuK=Xbh*z#^|1n=wZ!|W>*sD&Jnh$$(25t$fhXr1odLg^m`n4`fHWpvP
zcD&YhTT*fz*9}1V-;RftpU(*VtyI7LdwCQFeSgwjXq`%XsS{|^k0PHir>P4|dLI$i
z+}KgjrrV5O)PiJ_iZR4PW_8Ac;4oIJxA<N8*2e)$yQ9Ou2=BdN>1<W~4^9uUvdo8>
z-hpdtu(gS&=fN_Q#h<pnk3f|Bj0Wc+1nroqR@nLV>%Z9R8QrLnZ@2}p2|jAIb`*km
za4NZOWtr;M06W6NADug=r|}RRLeGTBAJBymL4{>dQCpds9+2%niSjsOvpI&0*6~<2
z*^ECe9xxh@w>iEt#FgHrU)uzct0=FNjXz|<AA|5gu6RvQ>BRr$@^nS=cvOR5(vpxK
zA$M|%6o0UQmWQGx0>LwAqSYO2-~4pa4Af~e<)h$Hfh&CY+N)6Nu2U2yx5!#mROQhj
zN;QZ2&F>H1Xk@gm2uvD3S0zp1PLdIIgZQ~^bI^~D_q+mLu@4glZ(`%y-E$OTzJr~l
zQ+4n~F7=liu-18p)7$t0nS}+J1Kja$Z}tyHjUMHKqyn7A!7%**q4wL?H&Vc98nS{r
zvkz2mM={niEVB>8ay2){;)0;8?=v$f#_gb*DS&er@R#TZh(b|F4VI;^WQ^WdsAD=k
zxfB~lVV_|vl1|3NGlUus@TFC~YDX6ZY%R?FNE1L6e$;wmS`?NqB~R0Gb7s^%jq2ii
z-ziD{im*V(jQx@`#3~u2<3)axN(152hewDCzT(HFOY73lgHzdtQJh8l-sTe_<b7n5
zaFz|DZhZ|_DS{2uQ|*jOqv)%YHigGUORB<NLfJ9-b`2)4+O6Y|`b?w0#bY#6$UV3y
zF@AMZbtB`s)KOp|q9o|eeHM1eU)?oMRF9y%rF8+Dk<?Uw%YGQiB|Div$IZut(5XRm
zM=Nc&ZW8Qojr=K3x24~In&8}OJ|P+RzT)exx58l@E7=yIH7^}#+_V01wQT&O<>Za%
zQarKt`Z#|6g14FGjg^$v{Peyih2`H5*mJmV2rf!*rYaNsHWQ7roTf52{OS0#ED1^k
ztVG9skBNS?CI?$#!I9OODRSYFf5xNWk41`Onog7XyX1}ccL!5AKXRKqX;6$tF_@;%
zdb-t(<FWVjNEj0}%MQ#T>htHM8lsk=s^(cA#2!9nwPs)LFD;>cFji#GoWI6=lS+sK
zZ;E-9WN)K>SkX!Ns`W$0Off0Q^jHp(%8ADaV2II~h<VD3n^%<x*prM~W8i3dBNTV|
zi>-!+5I!?}7tmgE56;9QBg;)Q&;L{qs#An<;F@PBe-E1wdHr1@#rc&4`F3dq8oY_`
zxA1efs{cLvv5Y#igb&2ew#JnXIx;HbU7Z_P|Jg(593+1fO0_^6_0>`ZY)LZQX5R4>
zTkL2AXv+t;Mr`)a_N1d6%&2Qpd6nM=M#sv8Cm(R<o69_0z`1X|-UM=3*1bJi{=EYi
zB{)DDH-mB4wpux>U;ld8_n^FW8elIk!3a_3$2XnLtST?71W6}QVM5}JRahxSt@hzj
zu6`nBdl0F^?b=JEUK1f&SYv;Y*AmMqsNZsZq-2^v=)heq23D8$mJ4`&iL^wSValBW
z&%t$j6U-EZ6-!^|mn8_w3pXRe&aZYV2T4*nx;p-rchgwfmS3xo`x@(9%oGj4l;2Fs
z@(FG*tE7Ib_Jj}@$yQ(ZQvA>>MWELZt0X{`?S`|_p#8Rt5G#~_=m#unZT1KZ)a8q$
zO(u#=XbFb*mXTwV9QY4Kpie=iyCUk|r(D#0QaZVX7vP7!P-k{SE%Jd3y(+Q5h_pJt
zA>5?S(kk;uNdPQi-Y8F!{raB9+6>;QQy9^i%Ub`R=yu^ESQC`M);f;z9kdbMH-I{k
zncKAuru>YVenKJg@WGqf8(+{fG}F!8<t=7WyRg%tlXwHU#&gPfOGC;j#?F7TJF5qa
z3z<K;b2f*<RU3F+x3nJ<t$~X1>f2MG9@d~X1lSW}Y#&4ulaRcR?|z0wl@RgW9EX5<
z=>e@zdmspCDP&|Bn>&vO&Cf@(2|KkBZS0O|spU58)hqRw7UGBc5kZVKe1yq{bCpPa
zKZn`+L@1{Ta<ob8z=K|s#mG%awA~)ufr6tEWHtRq3p?H^lVA)F;|}BKI!=U9K&f*+
zRw!xCg?-*+=*El<&JC!;kbB1V(-MbT%&o|619SgJqW)<oB+t^XD;~TF^|rBhiWrut
zTj?3$6XItkjY4m0{ji%|W!W2TCrieYLv~YJgZ%pioSkGAF~WMyb8R(*!7U%y+X@2u
z*+Mnu4IH(aJTu>!|4Ln9dPqiZX`*R}kae|gQDDtWw-GD6pU>y@7C<W)`V3zCaq=%^
z(2Uy)>2MpU<~v}SvfXax$&8C3sLX+VOa1V{#Wd~_Z<*bDiOxC(q&ueTmJNKU<0A%<
zG`Kf_oq~Z-^vH~KRP#+$RsS*u9p5g!8{Jw#(k2{j@hwhU0k!Hes-%z-zFimOMP7zW
zfIZyfp{*Ktp{PvX@sy-?minVJZ5L50R0>`YS{X~oSjv%<8^VgxfY}j*Eh;CXZKdvh
ze2+<uBo_%UC)9Hv#*@LRi{5HH7P&XC=bCt-F$%t8(pH-8T4Y78*+5Qg+ZL`+zgsp0
z)<5@u_+?q{TH;f`#tMsi10HeH%0h*raFoY&j62M3)OEoE?%k-RysoQRzxSrM?%KFF
z^O76-_Wy?~B>Cy+g`Lu_eqdE$ZNo_B@7K2n#Z`B7I1hKDdR~Bt5g!N2vANZJuR-LM
zPx}njnab-=w6;E2NVdw4>D`QNjKVSR%pUBBQKDV7f=;p$`TAErB-rP`%Eu0khZq|d
zg;BGo$2aPp9*%@_{)k)r`K*>2pbROV6FTx?TE{|rhp>}P%E>9!P-^5M%CeMgkl%lv
z!1{{=#~(<+%|=H?_{*>%T;_3TY{@T%QkK63yv<R2^t(`ujG;WXFD=2Mk|2vFAx|w+
zlERyo?D}44Yx~Hc+<A98X+|++d3$})Rs<n<fA0*+gzdn}Fa#H(0)Hv4|IQFtLYev$
zEq?T3fL}<^3@VE98TI=hv<ikyz<|G<Dtm0d&^S$2Jv8PU+-<W2N%gX81FNiPqZu$H
z4FmM}2B1}r&b<AZ5mZNN{;rA^G7PY()V!2`!BI)N<yS)gM5K+1durZ8c@9FECDkex
zDu}PO$gM!@a103!o)w>IvN_&+iq;RhE<lSRf)Y9-j(h@2OZ}+rX)YC7qMf1f=9Pjm
z&)E|@h3&0rDD1-JyJX=VMMzy^E4gVx>Cq%JKZ8P;<;X_Nkkl(<=te0hBtbo3k^BVy
zRPkWD5W&5XqJ5UB=c3l|aeUfsaY3(0IeB$5Rf=VT$i={Nn%xu(<2EWhxc&xlD7Vv3
z3~A*uOv~s>kp0+&kaQV9K*doRf;^w(MrwHt^YVFr07SJdIi^m2AXPx1bc2XzQQH3k
zU>Gr>Gq@D|SsX?+va&H{5XjT;F$jt6Nrd#<%w&9XPSLP{>*3HW>ZP&=GtdR{vG@mc
z*hq>64(9$Z=4PGO6#D@^J7QbZn{8?yw@^1FIQlSYj_<NG^`(7mmLn`|9y=$4#evNo
zvT*3|6`vPtZfLJuI?|aIuFJlWeJAbzv5seLMgARFlXR(qiXD7CkJouV$w80@ci;@t
zdk{g{&pM>G<ck*1u|~I{aHwj*L$*)kC!8G4Bilj$$&rp~s)C^hziSjYUJ={W{X@b}
z$V$8;JAeMRUi-j9V~K|aFAXE6inYKCxy6Q9HK0fflVv#%b+RN!WB_iLVOlW4DVfL;
z7ydV2s6TYrI?A46=gti;l58~IGlu8EZmb?><GRNk&w4l6r?a<I&j~rjKOS|#YC(Cb
z3Kcs_4?;*VhZB;D&O-4;cZlimxqr|lQfRX)in97VU9YAL4;tKkDaVg$X%UJ?9aj7K
z(^&v6>@NnyxADy%v=7{$bHo9q;~U@{y0pZ``q$P+%TMc2+-oL@rO>k42|pk7#Oo6Y
zVz5K}r01{H%5gQ>$<O^ZLN~!j#uj&jEuNJ>F(9DT!tex|CggX5mw0>45Q24<z{VJx
zQm^JP9H*%&o}x*9KxmiY_6^-qVZcl)_K}RGSYrP_8)hz@aW&0RhWs^Q^dZY&tlaHM
z_05lFEFEiwUz_A2b$;S8QhSlX%mo=Cj5!>T!8b7}(jYCVf>xQQb=bSAUGwhlnHx8h
zSx71%E=DRAL@r<3U&*Su_`^T)k;6$JS}gW`Z9}nOeph9dU!*+>g%7?g5`a4<yml*t
zNLegxG$}P_%?p`Q6P6|LG@`rP<;l81=>DCXB{(4i2>ptvMlSKQ0+SG5he|TE8!a#=
z#V*s?(RUVn*Hs?%PrqxxUzR87jt-Ba(3(iTV)dmH;})~$-MyJ=6UDJ~X7|}N0R63D
z75lyk0h2M2D<g$P_cJdGt~Nd{*<^0xm-?r$xU#OuYH6ts9NNbW5;}ut)%Pt`Uyq-o
z<596BHsW}@=?05A^b(r-e<0EgIV^#Kpc)~Z)|+{XD}dO6fduKzI@b-sj-Bd*)|s=1
zzld=-W}&bCK$(d}%0OWrVdcz#V@GIVQ+aBc@8wW8h3vu|Onz2(n<31502OjTKPjWe
zO@o90TfVr1xLY6RQMX{?bI>JKok2QHGg)jzYLS-NoDgMy9ZH65<Lv|ahl&ggbOOwB
z+yeq5NL4?dB^wscf!c!$?yXGRxW_djHmSAXw$DOOwfwBnAp|;A=gbR%&$2v^)QF^H
zzFI1yX6CCTv_Nt?h|AB{)dBz6mHoP0je;QI(w7JoKMBT82=kl=x$qdTDSF%k9yo<i
zqehGBj5tg%n2e6CLz<M(IcI=xsNvnh)T^K{ryhK;WX5^h9~EDY+YrUuA+0ngOs<r+
zD?E*%9#$UYvcfYFkamSZ<AHQDy*}uXwvX;T1HM4n*Fgc5Kb_6{51!(LwMV5v5ga+z
zOO*s^5SMn!?N8L7F~ct*Ig5a5{Imnk_Za<YWLTQ-LGTntv(>^<KzX-9u+Vqn=E2_H
z3gSE*wT<<gx5qdhR@1d~cRQBDHKt@9!U+h3!)7Lja4;Fc(|b#x(+#}}+bfK+pRdSg
zb9B^4*OO<1w+-%SUE<z;Vxh(1r?6P!b)Y%DsCF5cQhB~39kgx1Vf>RjxV|d>i<Z<5
zok7pG$9}!}9olP*Z{HGeT|&WHBV6p<(p<^NzzX}}<F4-<9BP-Mt?p5yHAr~Y<KfyY
zNe7|Hf2QUTM<XvDFAM1{f|;1K4^giUqRWDK=Y9@JX0w7$5&JOvdZ%NHexd7W>gUrR
zg9SfY81r?(gnZPDFkwaV!2(^YXEevEkwL?cx{!7a;>m9gKD=S)g7t(>wI&h3ATBG`
zfQm(;k(+xzQz=lefU5Sy-MRnA*;|HHxwc)SFqy~{q`Q%p2I)q+I|Kxz8<Yk~2~kQ)
zTDp-A=|(|PN;;H|NjH2KYpv&d_pyJx&)WOo$Na}J=XGEA8RH!17!*vi>7)~eD8D>B
zHcuV0s$K|5c0Ajz6p#OEBLB*In6z5*E9N3@e=4tl7W{NOt4i3(PB6PB-sg>VCW{!m
zR^8iLo5{M-q4@5w2O}sB4ld@_cH3Pveiy?c*!?Y6I)lDbF&_-^8Sj#x{5eRxK1k(=
znp3BF=3}W$C|d4_`F`paFKEc?1ZR12Z(6ORgk6^pzv%F9Ipps22NmVkhMMr#`{r!|
z3-LWH0`e2udczS|jNKT;SH)49Dgx5B3&h`p%P-1)e7oru=Xi!#ByC2ErLoO3M^B6$
zH2$XeTbd2R6@0ld=77=%L#GpFnVY}KyErYIc$1-*Fu&)rWAl}jkev*UsZgBp07uWF
zFEW4!^@of*ltsQueaC(5WD7+}!vy1jVxkfvgz|@JLeI#%s{1*3%2x!Ag70`7^~-q3
z)A^;)M@}Y7)il;(CgrB(Eh9oIkf0OTBRzg)XxyUAOVS1s#kkRy#)Y@oh?Z$qKax6l
zs`sSX>*w85M}9Yc@>~^QQ6c`t^VG3m@NEyNy|;-R)GpulXtE56VV+(kt+4X;yC^!3
z?Kop+ITx}eV|%8RZhIjSJH+zrAuj!NKN}Im%d+Mq@D&O6P6Sa{`*pt=b+3&s2f_??
z(v>9@nUE2WyC%o3wB*Qe&Y;-IxMtnDgU$uSfHA4CW6YW^iG9YR3purT0=O6ROW*GR
z_3of)Y6%Al<a(xg9QISsvfAHtFq6_PBu9(g>Y35HrOI*)zYQlzBrr#vAr*?Sxblrp
zt`q&zwBMQ<ib?s!<F#9)08f%oD&Kl)OdaWy*4g$Ir`~Z#KDHLI2aQwHF&!$~QXEcJ
zR0gIf#=@L=#6P73I-oz=J9=g7_BtAe+y9U-e=S3RsBZWiw4e}d>S0yvJezWd+BfKc
zG%Do2p~IYG{RlaZuNRg1D_}`&BpNtY9<H4Wlrk-mqPSrzzYMd4=}_*#)`&;ql1&^@
z9yqu!y#wS9InVV7$N;~No8_aNy#%mpUmW%Tr^=3tV4^WfX>4FbDZtl*VX~}GkIfEO
zPitO*>PhlFndcWxpT-&wpW4i?bzM~4(%I`0=R=;Dn2+)%aWb<qLYg>~By!*S^MCX1
zeobag6Or6zhI1?jTpEh$MA@ukP?_E!=LT0sf1o`&Z7TBo{%myBc0J>bt4s08HCcZO
zM=}nEy@}CntZBc$r|>R?%Tu#WHtw^MjQ@y&YrTJ9f-ZNO(y${^qoDqXLA{i)Lokmr
zh)2Hd#IiPQrN+>2eagnq`~pPi{qT=j-`{_ECOkbGKqr)(U7yDIbV1KdG~s!j=R3vI
znVNLJ4D20jBn-PmTV_Ss3hqchol_s`9O3CZgw<N5qT-uECGs9K!&kLi-AM!72zUVq
zIzjR@O4!{wnVQHZ_sd`tN*>-%kc<rNn>)JKmpb0INYT-;08yPZ1jhc7b~3Jj%!sd1
zDI77`F9ZOhZ!?-LopBft>7BG<@x$D;3!N{XffaK4BSWNVF1m5L0I^2a{?Xa)>{PXa
z37aoqhjvdF2={5@(=@<4(RivyHzem#XN1l0a)4M>C4^tD*6WIdoXh;(<l4*Dxeq20
z0X9|$sJjre=3FZV7KFGK?~|(41iSosopyyoX!TEoDlf$AmkX~!k7H-i2S!)r`<NJc
z2w0D(;>e#jxtEbUja`H3cYGqt^X%l)QyiPS+Z!7&&a0EK+2JlsWE0v{a$k$cZtf@*
zGMk%Ndo;oWMOayl5@Mr<VO%e9e8g#eur*O4YwUl+!^w>M29w|m(R$^%8&Dw7Kd<q7
zM8&yAY1SK&rDO@2`as2o@Qf2hg4j&L`I}{b{F9l-q=41c7W~x>^d|GrXxQ6Zn~TFg
z!MEa}oAhKdQpTZP4F(+@gm2JXx5mFa;ORHO=U^zB$BRgg)IA>P0R{ZsoB1zztw+R>
z8hl1Sq`ol-`3iPxYhN!@6&Of}VLUY~JVQhxQd<!_5DYclgZwmd#|RAliFYGw3<tb<
z+b~`WN*(;XB9Fc~tT2DLLqyjzN90{tm?p8$YXLCH^f2-dI3XYt*7?t#*+1K6B_xHT
zS7eTp&%Bz7%qRF}s=+$V@5{>9N`)Q>d1>B_Yh}lT2l|T7$^_$kF_7c0uXD8HC$_vI
zDSQiuNRNtoLq1^(-B{|Mwj3?yym))5x%rt?`KBoB??nOyBp?Z6ug0Jd;P$SVetIBI
zqLFG0?Udq%;FsGA{B2^C`O3SOC_5YZtx4Kkd>u+%0ZTG<CJs10t_n=5NL;5ckH6N>
z=2wPdR?)CuRb9HK8@`P(-5kpczN=*X`t|Erbsj?Vm^KM6b}xPaAqgqBHI^^f6qrG<
zJ~>3HAsHSkGvGttL0H1#W$YNT(}i~iVPh+Gn_SwhAS~JPq1WHG_d>DQd9M)`TpPbX
zOne%^2JjeJg!P8aw2GIQRNsMm)WZVK9n_o`3>>_iaR>T(s}C_Mo*W(`En&B}DlpkD
z1>>rW@TPKRLNc{WbjsdSDD53n$u!g4JM<isQ6L^e+s6jj+muYIYM8B$&d-tfkzaZ_
zWSn8FzU9tH(%IeWDiN#kvK-5NLspH}EY)az1n};k0j4J6j=!q%rgr4eFm;+SMbE{}
ztm;ffdm8YZAaJTn=YW;=I!9fEOK2e)%KLnkbdnAVW(Z~A_n&*;r|uc7{RzyQV0X42
z>Z={PH(P605NI>1`V?2aj&KPfqN^dm5A6t<!CUsm<RgN3KA`8}5@GM6jv=qIsdSrV
z1bQ*d9a9(A1Cq^VOOf5%2kkSrfNW!H7M^~K<}I6w$cV8kyJkqRi}Yv`Xjz5vesV<}
z>C=NPQO>}KFg#Z?v(D99$T>}ePtg(J=76-q`m^^u2=_XP@U1p3m1r_x(?rZI_t+Mg
zSe^GZorC$<=-gMACyxrxVr47dL%Kg4u9%=B)VL!%;QF#5<Ry^`IEZDtjzl@X18YQg
zxd*F#vyc$8y*GnAQ`VVwO^-^mk8fV1Xma(c1q`HDuxDnBCUS<Z9Nz+y1H;E3vRkJg
zK4sa$UWF3H?;a~&K3Yn}T->&vW&Aw<R_T)8+{nxUy4X8Hh;Y<fsTfbN;Q!{cBnqgG
zn|Pf#*{VLcmi3B!)JbT%m!bOSPn^Y?>LfxmexzcBj4Q2t><*#Uk=PSkHFCkqZWKrq
zh?23M`P)uq;4!f*iIxmS1s*@@Z**gs@LqOPupj%Og>OOXmz6efHM+nq_VlEH;y;N&
zN*BrJbmwCRl~eQ!Z))`jIGAN-RSyXcjG<=Evb?;R8|d(A+U)<bU92-)G4c@3HaU5X
zg@x5XC(AhyO_}*j=nC7?OJlg%4q>~cs)y|z;VvFagL`RFe?QHss|5WY=p8k)L(_r%
zud1Fey0f>%`eL?COWJd0RNb^j1Meh|NjpHsVRx8s)JAGB&0)V@P4<qMs-j+$mX@x$
zCtRP0aJ}6{7qg$_0W9ZEOcrsYX^-q8hO|Az0nDU{3hXGKq|<J5EE7xw7$U8BOUA2S
zVtoONHOx3B1de+lr`<4RRl~QvWb_UPcAW?#0EVFp*uN$^Hcw>8SPdFNEDa^Tmt8Zk
zTLhe|0uv6oFmiW2DN+wqgmA0*Z$Fw^2g7EC$_J6+oq>QF5aH0g2yG~~O+_Jk;DBo8
zqM{P@fC?>o@Z2-b_i{K>$JpuB7?Bx%KcSyLTpc6xQ7_sXIixuhDs658mE#cU1&$2U
zC=eA*Ubd`?1xzNe0-2o}lMcu(73K_zAAfR0GvaiJBoDuHz*}CfbxtBp=L$t*C!iiM
zi4m2#z%do@%ZtTR0C7^cPIOhkm&_4j_8TlNZ2+P>WtKyxe|wlOV3zUZzasvT<n9*y
zCP13_M%Obc(h2=H%8@+>SJ?V&5@D~RSg2}?#XzD>qR8##Hv6+C*{_*)ET9rc(95<T
z#o_P>`a|gAUxRe4n{>4rwx0r@qr4Mx-N+2w+6}^<Pei1TaL#<Rs6V3KfSyT-KIL;R
z)>@rLCg_sNaJ=~q6T>*z-x(<e-wkB|bsB*qh;=@nFw*&wlY5RbtWT|Ze<PaN?veBb
zpW|F1d3+}c62k4WdpIvC1vC7G_r2;vz`Py>`$%ZlV#D$U>Lo%%86%n>KSKD0ANSRa
zb<NX+(Es~J+`k#(dPvgy!#T~b>O-7pZVP}Vn#F>%PjC}?n3_DNgIEhXUB!j=JSIWv
zgNF%;8UelrjLUXZ-KQ;w1uD7OQ<jG!8-_p2i4=e;t;Y6*V0xkAU#kXz{bLA|q!GDd
zMn{K0fpQFRSk`$`l3>wDi&PlxFMZpf!6?%|;RfOr)Ny6_xkMHAy7<s5gNhVXge+r-
zU1<dqgF$;@aXkqWji27(<;n+kbE&))=OV-yp`c^Qd^`dI!I`l3g?%tNEB#pY?ME0V
z+m1ua>5N^}JC4X#QmYh`MiMwrG)*q504JF%N9$Ejz=IW%c2cBE&FBA_=KR~wcbw?J
z2PjRU^?C0wKGKAx*fBIOV>01xu9L^(w;GetvA~2uqYR=etE<KOtAoRynyW7#@n&j&
zj{Dd~i^jVROQF*0iag!UN7lu>uXW=em?|OhC+&}Cf7He<Mhzg;A$UoQ71niw6Jvrv
zONNfP34vjm%WEUTSKz`3!3fn?bW~0Vi0^R+EZvU|0q$o*(6W13zv|Q;fOrgO(9;LS
z*XOx`_c+>@#x16z%Jn~gp6dtl(8Gd6^+<wwlne*(Vy@@??`~WH7~6~dYlyXe?pYkY
za%>SKas(~$BwmXZomQ)CrEu&&qK^OeQ_BP0JXJV?=s#Kj^EXt;wv#I*ft;&*ttWy^
zZ%!<3UC^fD-O0muWi%Q2`Z^1|?PZa`h(LW8kY^6V8B)C$ufC%I0Syz>we{pn&3{fP
z{=xMG0t@T+M1EefvLXmoeU@aH?T;qEAdioLMAl_q@|l#9CA=THPyNxe+$sQ@5|~y(
zJcm@_Y_7;u4U}MRi;h-Yg*>?$ul41@bp^dqe`{(QYN=in0Fp&-&&_NI&^!JzXtv&;
zpU^cXCtP~5fv!Voq4@fB)efWf8_rCy9vm5{p!Wayvcwbwc0WM5=LKQU&i|3^M<BY0
zo>TO~jL3`WiroB7P*%e)iR^R!idnahM6Oi#v#sb2R~~xv47dIFdSlI@1VZ&+ldSo-
z<C6bu>>%E8S6D?~AFyWxrXQ^j7Z9-Oj{qU=Hz<YoGwKJJ6g$R$woG+SN;Mxe8CF}O
zyKRgF%0!XLhlGUSQ!2D7%FAZ~8E(zjrB^uvgM;(_G@HesK8Q8d7;TscphQ@Rjm4?|
zl?66pMi2iC6zP^5eOdX|-%rgU<bU{`p$2#wKvEq5?);)h`*|QBzfx|>PZiVg=Q4@V
z;lJeOiNPD}*3$n^$@uR-M@JBnUO(M$jy~LPkxViiofFWy+-|(9_8>RWz%w`-TKiT7
zmCk^m!l;=bYr&|d-l&Wn0j3b7b2O$)EhWq!*qqGS?+}lRh9eM>U=JeRJy-v3<Fb%`
zyIoU0m7GzW93}K}|E6_*^NaKLWYON%))&qC!of%n&C9ykCqF}Ea?duTY3t%Aw`?Wf
zUQE+L0*_spl3bry9B)bDPZ_IdG56JPu2U8rM|{-ccHy;f-QTRz8XAIeKq3%mOb8L_
z-ETvt!TY8HCIJF1K-tXwv1=NILFM}I-~2;P3^ese#@1y(6Z#8K=I{nN@1aq+!Lxj&
z<Yyw@$M7;nF=6<nO}UeJZqiD`qx~lVweJrav^Ts@wxofO4#~ZC%6;0t1YvUn^!Zd+
z27LC@;_wY4UEKnot097Y-KL!?w0n7oD6v~l63~Qs{^_sN*Ls$J)P>{(;A)(zFZS_5
z(B?Lu`CnaMi4kQ2uuY&u|K>v#>zjxl?L$7B#`+ftoxSC4nX#Ht1-w6<3QB!3HR$Xx
zP)50Zuh^tkKK1xOUq|tWjy1XF9DeB4bUu=?LF_=^E;=q9)#A3}wK&~nQ0@u@$JGu_
z?Uo9mGhr%LiNbR^-0kvs^B7EzFZW*Y;aPcBlZn3Zw2mq7D{D!YXl?=mM4N!vf^GNy
z_lNh_)&YQOe6{6E$a}RM_~~J@pCS6G>!rjJ^4CeUz%a|}bid0@qe4XzawDIm!QtTo
zz=h5&5+k)GN9<{pn8bHh`1ikDhDsg#8@US!z9A0{)3OdTYgU!KmHI@^*>mdEfOgVv
zKNQB97+PK=w|L~$xe%ybM%nJW=vyEsyl{TN^X9~BZ0Zy9{V$CW=X$&pW;IDtIZgi&
zdG;G1Vc*+GO97vsJixf_9rNrc5zPA_EeUU(jU6MIB|dKj1bS|OC~eKuGS7Q&m$51=
zfYd(cjD(Mew=}|cFoVSYE=Opm;;qwZm8l)f+|p8C(V9kEF?Px!DFIJD@P&m%UO`>e
z*$dTWB}NOSCFR036SIHyu>U%oW*I7vX~yNU6VB{TUfYQjnTfMS`9GgBR3#)ZBI#{<
zcAt#Y@$2J6xrB>*7YvP`j#@ERFN){Umlek<Fl0!a7{4i2>cal#IWJHzp;T69Q(wAF
z&64O%)d6x(3Xpxe7Xy%qrcefo)2$~zDDue!Uef|)14jzmh_YgyS{E-b0T5w6xX-)k
z3IdkDO-B$ibj)25t$8{FP$XIjw`)-rPIqTNL_YDfLR}<flR*sM3FHgid_ug-qfSBP
zwlmF1SGl5q%^CQq9P+Opd0@OOm0kHw;rYS(la^T(^=<vTg;_1{UPfE}dYNQ4hgp?)
z9ks%Ub(j;DoSmkW^x05WX;)}5{-oCSLfp34m0o)2f6iH0vK`a3V9E$}UaPLLF~!LL
z#*IYu4L?|njm%g5tg<~-?vXR5y#N*)Gr(H#rb1ZUkD`U4$jHbVz*<P*cnJZXx5@O%
zDN<mj<O)e(@T#h=wgDomcrd8&D53I0*nknL5Aq3)@_OSn6}$??Xnl7X{PIM2%UZJt
zj);{^HtDxh&iml>*B$)tA4XI$qj{wGVh5DP4Y$TxXQ$f=3Ce9ZN_49#xpilvD#Rl0
zQF5Ou&7$9NV8BrV1E^|hzS_`hFNmhweokj*&#epXXA?|)zvmfKb-G}j5i0plkP(&u
z2eE#yJOg|jASXbVX(^<JuEpz%L46c}cE+yHK)r_bEj^s7r>93SnV-Ul@Wy(5(f=;}
ze#4?A9SF`S3kV3XK6olu0=AkaX1xwpdV=kSq6Gzu)b#aJR2Jp|gCX6#KE2P*^q&3Y
zFxyVtEKBwD=}(tDr6o6vqNy`tmhisnb^%D*PYj6CQtexU0*ika{J_nTvxP#uE(9Of
ztR*uP9)2N}uW-#h8LhL=AA4A!HhTpPuh;!t!M!=}6PkV$-Ufq0EA!xEB~N1F6nUb>
zXxU5a68q_+3$l}=H6{amf7|fK{~dVSc_47L?b-l60iX=>f>roq=g}_Ey^))>uSN%E
zTv6iYAbL&R9UdOSY^tiNV8J}sL2$I;JgR3ax<9$ogVCtjZf$LCU90GYZ%hr2fHKyW
zk=rMBV-?`4QZhAhAVs-#1qwkmup%llYNZ=ax!df%t1tK`Uei@`L<b1Ojm~;C?bhXh
zmuP#}P)%)QZiwof#KFF%dbIaEgvcSnI0jv!zlZXh%e`9HEFki><>NE!sBsXDAJLA9
z8I}VMt5`bl6JWJWoa%ly_nGBDwjC3*+0D)Epj<CjKfo$-s*}RAMTYy;F=rs(tC=9y
zm;d_Atsg^Q_oc}YHy!644$!OLXObfQ=T8z;1moUYi79T*dN`=WwMy2nnTc8pKvC&m
zC?HCV6=(RiHLFQ*F@6>fe81-(#N{#n;(xB8;p3(HDGjV%z;-~n8BWNOPa~$ai19dE
z6!SCsbK__gW20N9pLa+1O7knP@7j-KZog~KDeoP|06tUp5D~a{c)rVvoxmvpcfa4q
zDU6t;zmUo)Srdo^@APA1iRvow|NV;=A^zcU%vV`|ri?Qji&N4<7R}klkbbrlas9zX
z3KS&b9U9q(NyFF|>#AAJ2XXOv))SrE8B{O$Zm)B=&}ILsH`-zRvh7Z$1ug>`G**s|
zfF?Yq$Zb86*`V`o_Lu+pKHxC}AK~mf<dFCPcL!Tr)Q7M9^|Z8p8$Gmhjzqxqd?C3}
z1QCNZgCF#hUg7^oiqOXitT`nw+-MbucbitTs8lZT%X&{*uG8xxH9#FBcKfVf$f>GI
zt*kjhft^1+<@WkTawyF|87S>>d$)>J!}ISJo(tr{o)rndTl1IQPw|Z!ulAa34$V*W
zddtTY6%|Ev2cbs8kKJ-P7#RB0Tdk9qu+w{pA%1gq;A=lOpp|5!hto?kS4RH(qi=nT
z1!e)fe1o}ZFLUB8##>HtCj5^BxiV<<2?DRHhYEz3BX#L!5(l-s7i`9B{L+~ohCTf|
zhVqMN=I8fHunpm$1qKB{yn(m(#=TldLXGsw*Yuk}0W8=ESbS3?Gy)cM$lW7bVY+Jc
z@Lq(Lkn9^O_~hhdUCc?5ZUiON=(Ii6dn#O=h8|8L2WS7c;W<LQM4BolQ3hF8rC`S_
z^a*ES^NIJE@PfPEb0XPpDXP04M~ohGP>aT;%UUhoS7gWo&1bv?Eb)Jzy4)*h(#RiQ
zBtk;MpZ8q4FtXS0&icu1f4jTG4CC!LH#fO`PVEK3P#`DwknjD&=Gzrg)m;F}W)Kcj
zc&1D0YiO(pS5q_iENLUeK{`N55h(MU>7Tp!FFqP%5sXGr3+Ba%ea~W;>@JTyihXX{
z#`rOSO%5_Hdsf=V0!B0^{S)Sd7p@pKI}%Fa*ni)26BJNc&qy%4&w%P6YQbk;WCj=?
znK3wa;LQy!1flO2-ZwyiGV4_`ozE-Q90Y85d-L(r<}jht;0-MloZ9Sb=><QbUN(E(
z9O9H+k2Tx(chUYY^5Tdo49dk-G<YoZuBtlELT5OY6@9(^{IMFxNlIZx9BJD{Y=8N=
z*l7k?8~sWwNO)8sSSF?g9+AHaR4MfvU7zd%Uedyo6FJ=aYFTkq9-u*x14!Fr@33i_
zHMi8_D04frv#G--oq5@@O(5l;tQ)8PoU}Z^T!2pw#iqbG0{M7m6YP5a_4i%E3OwFz
zQd^jI$^Dw@cbP%rf2&T?7kD5cewOy;L^+<v`rTN-CFVjNu#;Esb%FkU`2u*LNz*py
z1$sp)q8H}(bkQQf`m2oXrOTN8MLNCw+zX68Tt_5OymChJ5<zqLiG;ek8f8IFX$3nr
zHFevo5(NE>ZF^SKme#2XEGpJWn~Wmw8|Q(1+kw|D|Gw@dqlxk9Czdj8GD{w3H=;#b
ztKg@`!ZIqFne!yHz|eA%FuLlj);bya!LR>a`mfhA<{HLP`SS(oUd>OH<6csydr`!~
z%*^(5E=m^{p(dny0Q9(mq3LKWo#MR$<X*g2(yX1vV<j_4c^LZdXWDh9XIO+x;a7<3
z8(zDC^=5rhLy^mA52Hbh#BB|q7XKHOV;lAecO=tZl0=^Ahm)-f7x|K+(<ZseVerr%
zMv3dcB<&a7;3!H!U-bLy?oeX^^+MkIdD%@7A7=C-zYU6F-(6`7;r|47DcJxKE}1@A
z{M6)R`FnX|ARQbOia}nikjSw-H#C&G0yf*S@BKyr+oU=8DCw))w$3`f=5KRIIV6by
zqc;lA%gZYxu^(!*?(D6cxIdEh*2Rq9s*|Q2x)AI{%^UOI8^3lZXz^#yq~g!s5nBYU
z3i$PgvIf2B;#4P9_ub643{~}=kp@B#>7;y@0LLvX7avb!^oJ~CnWj}9C;rnf*`fPw
z+W&LAND%2=iz?N7$lky^j)8FBSOoFYZ_y5W+J71!uDjPo8v!xbO(2+=3&c0y3%YHY
z8%v=5Hi0;y1(5YoH`8l=N=WFb<o*7lX6lT@XGwW)`*Grl(o$pyR^LCN`G1Q)xx$dz
z{bmZCzWha9Y~g9zIPTlZ_i#d%b@@7TTfRD;N^H4VOQE7beTPWyG$dT4z~iV{jnUFy
zc-ES<e1%`4`q^6og7WANJu&{N7}cmj_M+p)#=dXgK(G&7@W1oj0$YGmZ>Uk6g$$g?
zW8C(?u2blLeN9IW1PLdf_CC&=%6+DlRZ+o#!Lv+l0rON|x&ev`7Cp{-kD$dRB}J{R
zqB2ZAqadwO7=2GndD#4Y($xN6WQKqO7*w%NZMEO2)AfPyf;X4aS4;*AUE^m4GLI(j
zc*%JK>LSQxly@GFqvSrfu#*%oDT>uiCU+fpC7(GG;W%Hk5rq<7Rh6rQJXMx;$jlHo
zIZhk7{?DZkl!I1CN~h~VJ8;MYGemsG?gd8y<2d_eSE#gVT4p792J;+XQ>g&R`6T;Y
z{2{zAIf!0=>m#e=sWp5LD2Sf=dSEfp!qU>aI4(gzw%v~F<nm-Yo%H8>mrIFwE6nC1
zUkQs`_mc+q|C+*UjG#dY_*yr=X~xf2&&F%qYgJ5xbURCLFtg^CbDqTXic&bcabH((
ztx(I;L!2r~<82q;2$W((E|z6JIQZ0BHs@olXDWJMg3>`EnOkFf^V~KVK``%rMpeCt
zVC!<@pyhWFqR~YnYV=I|Y(WK?P$i+)h3fAJDb5d3T0rmXdNT>=*!S%#D7T}yxMEYI
zx}Zib+qb?L>;-WQs`tw#F;f((cWgbMY0ASxzU%n=t0|L}8P)dz5+QEEYkBi<(2Nwf
z4)V^@VnghUUqEAhx=rfq(lY+rOBtvWBz{BqS>qSaCk?Fbc`Tb;r^3CP`l7uqkD^`W
zr=Z;+#hDevB=x^|lD%RR^4?*JMx$3O^LDjI3I{V*xx}k!;JhBgU_=83EnzCMnJCXq
z`1DanQ#n3e(k8@#5db*xlV4hBo(AFn(Yli17g$h>AGIS~jq%4N3*70DRbL23E4u7R
z)=;#|@Js@gb$S)1m<Gx4f6yF%n)ckGwn~_YAiN#PW-M2p+qC;#F2f$l$x0n>%=uM4
z01;h*#Qz4=p<E>W_Qm@)StDGu7W#-wgRAkWJvSdM9UYuYh^M|~Wt6zQeE3N>fu{TJ
z%!H|eN@B!)9wau1bg=wy{R0&$Bq4f1QTuF>m=0O}W<^@-^{9Z>o3k9eZ$nWjyc0C3
zw}$xYJ>sdqAH#WXv2DbdawiT?d$G|65TAsIcbzS2?D~Jsxr`RgT-p+gck8-I`+AiT
zFW@ZJKXZoAzcH$qJo#>fgtTyFrYJX|pTW5LRA5A|zfC?seBl>vkvz*}q0g0DQ7!&m
zYY-PSoUcXG%0?K4>+kyUIs_q3@U9i-wTsIpK=Ja!aTzp#gUW4`+(#P6?KToF<sE$&
zFsa#w?mm_bRina^+afqkX-_ScTFZ6pX6*G*UXoUCuoU!I<9v-bO_L3l8*s4Kcr!s(
zbJTXeg~|jxY<@V$Z%^`H3rctF!jvm#rNb{-)yvfuS|{IFm>QN$X$rM(dF9{v!{?G&
z#>@|^?5ERp;)Q9=yDV=lpruqJHEXKH&3^=ojSpKBemCkaT=wjThoJ4NCw#-UUJR@{
zo%zglMphWd`?S%GaN*nCAyeK%OwE&Qr&2$-qwTe0h?f5hZ#tvP{!QL1vZN$ffW4;Y
zN#ywXQFN!~5j(bwEqJ+KP4s9Tj|ii}vN6#9sZ7WM>A!!$Jd7_Q6?9$_7u?K^yH`Ok
zINeS6x=?-Ks_$d44^lrj&<PX->n9KUim34M@yAC-mTa_>+P9yieM!J4By5^|se{19
zgzGt2C?$6Gwtc_>i~Ptu+5+NKRZlg0pQ!)W@Y8;WRp`5pUfg_TY2Z2De8Gk5IaL<k
z&Z4G9gq>K3r~q*?*k{<s!AGhqIiXT(JC<Z>v&XTS$XpS`br`TL_E~Ma{G27xw*HjM
zwTd12txUx+-##eTV=~wxSKGTmQU6KR!7*K$&P*0-Vp57|2L8uNOJrT5pKy1e^Y8CQ
zB}+xLO8jIv;W`UmAq0TI^pRvH^C7)5fEG3aOk{BPQMP+=_|$?&U^FF#u_pRiVrew!
zG{nB7CU(<OSlpB6f0|ZTRcUy4yelxFE8+s3lTBBxWi=t2p&Vc&m_VUjVVDJ)sv<np
z!rTZY;k^;WxForhU}|(f&Zp@V_!}-RfEjxINjL_{OV+#%NX}V<sqQ=2I>O&&Qed}u
zW@V-N9gMl!^7{OM+jT?jIbImRpaRgK92YN7Hv-ExffvNmGiIw3oWTBmh53?*=M17S
zF`%-sW+G|@UT;zux5I)|ty6QZ57Li~U#W*G!Xc1fzj(QW862=Os0Y5(?aT8qHrG%3
z-w0UjE{GYvWEMno{Dr4JTZ2ahfAIP>%~lU009<KC;1lawynhU{L%oMf5HU-2Xw9N;
z<DXK$&gf7uQo;3ya=GtFgy<To;_DhS@#J4-`7mBBbv?=H8yfS(FuxBpnYA<~x1n*x
zp1;@Ee}a*>Y#4<%W4h-`c+JsJ+5XoNk{Q9LPvd}Ghbju?8;(fe;(q$eiZW>GRx4p_
zOk=nNG#w&g;|ZN0jD78AANT<^B4!%xYCdfTsLO1=>mnZc;3W0{bYXII=!y0KAdyDi
z9(}%(8ZQFAv)C`!5K3|4{`}FFu*^N5Lx%uUuE&r3JOCQXXUp#tO9Ka;5Ks;9#Z$kx
z;S;5fK*FI61Qw3*jUM|de$B_Ue!IV($z-}c&Lm-ngaGO1IUXVawTkWn8esfxP}1rQ
zOs5@2O9b(heX0rzXdj5>hmB87DK6&K&pI}BbH)O&CKf9&l(H=+_JH9dz1(4_mY7J`
z%))DSi_>3&u&V2l2`$ZE_;<2ex&=StYWD-R+UzU2TTg30>9YpvmXx8}p&(_NAGVhy
zBobMM-MxQkLm)$?jWr^5rvpSJTVp)ROygT6b(tD>MN#3N6HMk0liAicUTWQqhP|Xn
zN=0|D&wRhlUQid!YQjD+m<dPxD<JsOEQRS#0x8;`_rfDa7TwjY&1nh>3a|w<K8G18
zQeOK^kePAF*gc;(!)rm7#gbD$@8QGvfpoI-OY?4fzKPEQHhf+7yxDOZxKxdN`$%)L
zA7E|uqkwcbg4d|cuf-4DGhCoqC)+o&PR6I}Ems4bwcT{ET~;m3`Ak2_Zj~o{?-=;u
zOr}HR9gn}WoMOQN7=Ou;I1UawnqQ;I;p=@ZGtvoL>cL*nE^pcsYy`D>Qu}P@xy!4F
z07{sN^fJD#p#-MbHOI~M(Xi6e36lBbcUTP$6n=0JtH_q+b+oognU&l=t3Z-b^8G)-
zNlGH~aEO#zE~n$fuIP7jrPiB$vi_Y}c7?rfh3{tzG=F|>la;w<r@gy=qUF4HQS)c~
z<@ou{`**Hx9~SXnPDuyXhax1Oy%-lN0=7G^SKO)UPn2ZpL+xs7rz%hXtTQ&O4R3J3
zOdV!;JB8F8MQz+_@#bnqpLRNVe{$4v9ky3scBGtt5T%R?MP+MCBndwA^338>1eIr@
z<$`^`&@Pz^-siu{@sTZT4))GdNDP=Ls+SwJ#21UxA`_$14i^_Qb*f05mlYK~?zTzE
zzgy-P5lI)H>85P$Sm9$}2wS4#KE}+@eV$FT<YoYbtNw7i?@Skl1Tn(K+@+;M7J9AM
zx3+B4k4a}H;jwl#)d$RWR<{_G81;WaDx^`bnuo!zjWA!}!{*b+Cn@c-z&!R!#yKkY
z8OV(C6NdII;0n<)LG~Xyl@J;_BbFI{<x*ERZvwA+`vb`N^iu$UboN2ZQKor+EaifS
z7z2SV(J+XL1H;t`!B>39#ogNfkC7QB7EDK1{bv2M9FL85f)>}oBU=XHyO*g>%o<JB
zk0+Y>jgz^-6eYd+1N8aLSuBp{+wCd!itMy8)LF)pTwAH1iOOTe$lY~czf+|UxuAYB
z*U<cliki1f_6<f$1A*gxi7l0oWb_y>ivQ(=07Umw<7|Sy{j9mfw8ipud<z#hy4%zk
z>$i6#XOZg==#tgE%~cHFFS4|byG3w#?m4Ke#k;I!^P-rdF{~<@KXFXx;0vxs0RMgS
zr_TzZQulsBHP?x2%iW=(97;H0y~#gh0Bv@(^W=%512BBB7=b?)0_Tj~OGx?DyT+e=
zP$Q3XuLM#w>}EWAfQ4F`Zka-yOzY4L?y(8^AcW@47jF)Fj=>5eU3iZ@Jviy^XG{0X
zsetA6Ig!<XfceSa-5M5tATZnBxcLr5&&YYmW&f#;{^~0wGY<;(<r}axbPP*MGplSq
zsWl0hUhPF}WMqMwJM(Z;m(k7Qcuz6p(~DWM&!bjnYi36had7+ONgrJo=l;(qe9W-s
z&bL+??>ev8>{aR6JLLm;NQwEN;#yAEdj>WaURG&}7lei^k!Y|^)lhia2Qd~)gsl35
zu!M2=FD~2jEtebq2FFNlcHuos@;Cf{GMsVp_D8f+LXqM~14|LF)P}+tG<RKM_pfSs
zw7mKrF*RD~yw;5^yU$!Q?Pk#C)Bc%OUmu9XC%OYHcw?Ct+WC1sr_WNPKk2zcAZ75@
zm_{lZ)y)@koQ<$PPj=*C1rRsD?JSyaH539l6iyrn>`yqnrmP~Ki_naW!P3B1(|yZa
z=UQn<0S+~n3UW_b1!h|XCW8$M|5LUs5d@lipG;yRX)~5cIGdLMRzdknXwM9O)qF@;
zphfb!&~jaUf%ol)?c~RWBA27e3o1R5SvHppEq+WUVR-(ZKXlS3HI0IK=}Jr5&$_VX
zLj&l)#gD78*9uzwelq6#oC?;G+;lu~x8X0uZ`aA}{WCu@H^Kg~f?|QVBU@g0YZZ?-
z?WdQCTQA|IhHakM9k?yew-$ZlwnEm|z?7I<_CVmT_+3VTSl&`afQMJ3&w|^sGB+8%
z@a%Ng472%^Q?ZWhLUe~#OG6{po}+u-sQh6wK+8V(q)#LGV=v_?-L4GG{MWHH479jq
z=t8>bT4Dzr6Rdu!_{!Rtb9p`=wdfeHr%vK4CF?6Tjva(T(!(;K21UB~_<?*eEKs8a
zlm+ns@fFLLPz}UD@kG6Mned^4!yk6F3k|AVjeBy*HxD${_IxyCh8L6;YEg|)a}@{J
zGlM>qYfR=I?xng75B5XJI__h&IJ0ECU3n=OSd~5M_+7+S?{-*|psfg|1=_i>d@sJ}
zMQqe^fIdIFVc3?{@2t8w`C+rpM}-AfFunAejbil(5Klb+R{!SW@Ld`}WwBq9{qHd4
zc~HQ6=O<4}cV-leZhn@ldG|SVnM&-GdGFj-RE)%>naQf{srQ9F*z61J<i&m|6QHAQ
zLsnP3xU{A`cKP6S+}rEoCtnz1aFBw!LA99+fxu}q791FCFHhFrZE1<P@>!5O!;rZ=
z`<WqR7!|?dr&+RUYknx|U1OLuV|+KqA`<{Z|3fAuOdW;v7t&`P3G>`d;CcGAnx``f
z&uy(v<-KK<MJ^?U=WLC-zgdL^8N(?QQS5TOs)|e1-i}pa^2>{DFKxhtTl%V^lcWcO
zPepXwV%GIqLE#;Z{&%;cpE|ex_Xp1d(XWsvw|PIRpjddk;!-p(sj9i;zfhDi`6a)q
zgq=6<`ilF7XRN);uUdbUU^3*_N6&^jF>3|wG6K6?=<c4!#kOykt*2yC$?=t*zb9$C
zOC&*l%%b;P1>RfgbGtJe&2_aWK`eAyfoc6Bom(|CQ;lROFhabFI>rsYtVIKQ$s0Ji
z_Pjp>zEb13#stZop0u)<pKN5?olHkrs{Cx@pYp{mTbwUC;n3$q(DNF1a>kaQ?(qvP
z3>L==30#eC^|_yM=@=l$pNy30a-!&M?cWuOPuzH@onp&sQZWnt-=lQ^(9za7cUi4O
z68FiUI+c|CDtl{-yt5DPXlNHLKVKu&h(F`!Io4M&*mP7@d3F5uuOwKo=G;mvai0Xo
z)cy+1|69M43xi1b#EnNXw`En?XclgSHpzil92>qW$E@oq{(UGvJJssr`$E!o6|JO^
zqVi}2&62Wkm>oJ|2P+e1vu#y1xgiPkWxe-am)6w~X%NvQ(Z|~RF7xR6r?+Q5)#2Lb
zr{Ps0eCj0-oC>8ePipN4Xkm1mkgp(v4dP<rA|rpFqy=+RDH~~rC;K%nN}T@blm5-Y
zw#NQorG+(XX)SV$yAtvfn#|`jg4Hi@fG4pw|3v3?lqwgLtI;)Q?qh-4;7=E77EDJy
z)mnD8%sKS}neJDAPru911M}&(KVTb>@)qeBg8)+$s>(H4pmVUdk<SJu#cwmtysCTn
z9+!uJ5(4W5kkV^;M4<6Ba0((PpAihsZy~^5^shMfKY{oE{dUq4Lok|^CRp|53}h&z
zQ{az?49gdy$_Fi`)Gecp)DaQC>c(Mtey6Y0(}p!$^Y~ry@O-&lWmgRis!0L$aQ{vl
z-^;L4sklt|YnC|iFG85jz};T0$Kfu&@oX`%=u;?9>3OWe-9AhIT#Hz3vin=Q=;n^t
zGL093*`_x90BrdFUVPQ5ZQ}zj_wt}Vz4lY5to&NLLX-Zs6p=jMaKFMgGd<(ygP*Ec
z>N<l!MKbtQb;JL87#Umt*VwVU%D1fKY5bMg--NCp#l9C_zLgQb#<uz>GSgntq?WU~
z2>(;#>0`lD#=i$IbvM6mcC;FgofW)w*hLBCV;@cpDL;^9;9)VrY(8-616|Fa3k~SR
z1#%x2!~lR|iq;<Mznow$H^k;OjvZB70HpJ+x$~~A%}J`x9UD3y#O2lt@`u7B`&IEZ
z`n%m_3(){1;vq3YzOswB;?Mr?NrLN_dX4>Pm7g&wZ;vOxjwYAzG8@k4R-a15EBg|a
zGFg^SFVU`OXP^Y;$5}N&ZFQHyx@JYNv^0f5J9do=bX{zLW9M=B$$HJ7->bmG;-}k>
z>|a@jRF9l;St7)lsIc(PtpAivY0j6h_m-;cx*QcQSLelx9cWZqo3Jy9oM}~VynONF
zjXUmm+Z{&2o5l##@}`g~Op#7Uca_hXunnWMKIP2f=k?NhL~}KPEpVC4PQ?3WfdsqV
zRD?Y*E2~4%t#p6}3iuqa4SdG=*~USpK=zc6oxp@Rxw-Ky-wyuxp@f6CVwXP!HDU<{
z(7xmw=AOK(sDLB(s@BxC)&TEi>CWh(?_VrDVVzMmHQY^LPAVd(Kb;u0d$htB*@=s?
zBd3)%_w20%n@)n)g&LBjb-kw#o5Z{``NeY7BJ18a%&P!x>SGk6@jiBPYe|!Y!rgbR
z++YA(zGX0Jz(9YEgHJP%P!WSQg>;$th%m06;C*E}^oB8G|0`Py)i1UN<fr4~<Fu|f
zIDd?etK^+qd@kPb&Vquk<d+fX@PELFGO$-}S`nH?WKESm!RnG;b^b=qP%q+FkxN)-
zzg(5_JVS4B$GQ1RX<L%IDn%-`<k!<6ioL)-x76<yM~0X{sLx5P@n&7J=A=};^Xl_l
zS8nkdpr+7!F4oc#zuynQSU{|uAt`P+dQj~3O4N`^nw28sZ3#uPG!Hv#W@piqT9j&T
zy~*<@6aEKJcu=`2RTytCP*5Egh+Z04wjl<=g=4iAgpr6(849=qCGy&HE|dFc`m4!D
z)KvdWsZ3M{o|nchO%9F~a-X&{_k<wDp#@$JC#r`uVD7{?r|R4&{P~n7YwE1BlC^IT
zQoF@;j|JhYw7mtyxZe8LdauBAT*#Y6TTCFeW=w8>Z9-L4L4YAmDv*D;pq7Q&@Ne{P
z8RjX%9=UJB5*`*nnTGDTeO~{xb<~R$3<&I`<~JoNBL!67pq**-_xC4B*<z14wcSjb
zD%gFWsdGqmPD8g#bYT*L<vt2LyNBz6t!&~wIu9%g@b7g%YF@3%^aElXg^0Hrn7pL=
z13sbWjAg3B&XV~PRtonIgzWji^KBHM8scCh1O<Rc$rU7sxIgaw$Ob}0B*QTBa-sy^
zH@ffI%6rsE6$<q~#2<G8`;utrW6>`MK-Dh?Ry%|PPuhTZ!Ay-!oXS|xz^+QV;3xns
zl!1pQ89IgVbp3ahM?)E2YsB|s()&)>y9tr~5md-yPwAf77By;9kS+sgD&d7;KM`VZ
zM@20WC3a&ZF*s3V)4*cEl^`}Zx9O)B?xPME=v@b3CM+8*d<bJTZZ&Kp&?zUN5bZ{J
zV-j(>dk+t~@8}5vjENQ3S+`X0uh!ca&{i*bS^*;_y}beu3^n%`T29&mkx<5fc8wix
zU-c1i(v36n+2y%cq2C0}N*=IF+?b-QU}a*;cBr3EtXuT+xw#A^ui75@!r7psu2X1|
zeG-6#0Rx8+cyIYuW^8PYWVi~5UQ@GWU_V0e7Kpo)_(cL*@@O!J&>=XC&0Xu?h!k8M
zk{9V73*qm;6bTdjYj^BrR{VhQt$P=HHB7_d4d#0LU(W!}oA>^GTk~K5(Q`vXB_bG*
zif##P6FHmv$@%;N)HdEKQRH-5QgB2vbvTtzwas`m^t=QQ_mj29oMc09=p*818^c*$
z>>a*BRzqZ$;2xx2F&)<fM)0L?#O)iPg~|Qq+<Ea*v2BZ>_zX;RvUH7owc9M!t>&ia
z=;*u`gRKbn><rNwtfB<gh{*3ouKmdFZHO=3-rM2#^}4<Pgk;!9<+Lo=L-{4_EvBg6
z-7L6x_O-}gu2e-YXiHIhTd)WiRe*wB&bwBa&FP}QzihC9;ZLWQlf<YSz?m(IHsEmQ
z{6^bOK6M4qwt}<m!W@IxqSDqvBxhzu#`c#<OJ9%fw+5;#29B?=&ZytX5E*yW0h03O
zFz4I{Ag}#sfV3Rfkex^~7<DUDFcdYz0^ntYh&~Y^9JlRHTR<u43PeFh0m+IeMR<+#
zi{kN3gXQRA7Uk0~vxkSTMZ_*Sa{NR5j?1idg`3J^|GaWCB{LzwdVnbm7k8=1bD21Y
zC2LQG;`adYVdCWsbo@8`{D7Ls8twb+i2jl@BhUEFfpk)GarDau8FMsjXJx$>f=<*!
zIP?r@&!^eUBrQNpn3SqvU2?I+`DP{KK!0R7bGYXw_zYS2QvmzkljDyKQE(RC3?!l)
zqO-VpAHMNZz!n?@>tB*@kE<}iTT#K0wv^G=v<Pt8yy&tKF;n}fXj3BGoak<%o6mRG
z1-D^G=2*us+R6Qi?uc&mY;0_fjtI@9OzxunuO;_A0z@*jtLy}g+<Ju&7p|ODS4aa7
zhQcd_E``StnI{i0BK-D%X2(?d<6qSyC4vZw(@^_YzNGu=o_xIx59*dg{D9ZyR2x1S
zm}&|R86_;ndOLqMwGa33i+K9V-{etH^o8dFiP;6Ns<3*S>UABu-s%nh@a_F_D6vm-
zP9}Rt5}Nz+MX<*-BGQB0-!A=5Y0ukyp%BELXa27oEL|6qXBKX$5>zNusluLrJpLl2
zG3!H-6EF}`?k)o7eW==z7eqUCzUra#!(Ii6J=6jaQqc>d3p(Zfv9(fZUXYEq-tkVr
z*xOGA2+X8Cl$6np*%+>$NFR^7OW_93RxvYZ&Nn&Dd?$G7a1l`ggYN!V@XfP%YtY$-
zP8z;6m>hyFrvr5s)1G%Boe(}rszD8$6CT@wB0ipJ*`x8Q|HY{8*W>m{U1r!Q9*`gb
zUM*(?b$0lZ?#`<{*i4s=1CKFovwm!WDa!PJ>vdKzxvF=oshQdZhB@t|cXqNoR?EEr
zd~cK({@e@}uSVN;eTvc_HcyUp&q3FwTJV05SJXhD-Q8wm@2Og?gQ$ugb{5g6^dEkv
zO1B1NfDn2ZR>t@XrPp>-hAxrxaw>@q96R{I=*mj8U#lczim3ZVmV<FSUEe=++D&sA
z8to1?m(uSdIAR|I{L_;Zs+-Ww*Dln8Zo6OwaI*V%5}gn<OBXk8A%zNqoA3&_cFG-r
zI9)3cxh4KcF4wCENhyJjg={K{n8EOE)*&#yumu{NYEptOqN|2j88!E^LR!+@l}Q^D
zP>CN^DiT~H(7-3pqEithWG1?Q*IAH8Ts{oSp?`;-lEmm?cnubQ2-Uw(NrY|Unm@ep
z#{d29T}>$Knz2Ob)R5pdi*gK4r>nFOHj>)F<=~%X!x)L2qBB+ux8{xP58M)Z?)GAP
z1`_Tr;|#FQpqe>g)%>J%NHLLR6v}Ckq7%x{34J{4CPrcz%#SUFq|8D!0)0Lf%NhjD
zzPVbwn{*~8$!@gj2^f^anKI_|9H#)=A<8k8EEAk~{>QJZnaibP$Z7%lVjVR8a+bbJ
zR3c^Uese|{tBf_7;ayDB09QX=E|sDMaX3^!K?u+&xg{nmwRtnN3rU?{GnuM~*C<Jh
zqiKfwR2#SXjn&As47N3_A&>GYsS2wAnm^7<55UAsRmRSx_O*XmCw5FW92Tb{D32@E
zy7s0RDFTVTX@^U>Sy(!rwD{>vC6)c=DvgAr2L4YnHn|0oURg`u6|<C3eg^7yb=a4W
zZUEF@gPV|Yj#cnbLo3+m_*XbvgG|I_sy{G=G|IaGK0sP)Pq)?6(PDEo+CF0w6@()r
z&lIZX*q<qb9nPPhQ;nt>zBQ&$4gacpjHh}mt+H4SH+(b^Vm4=x1t|p%^%MrfsvGTB
z$yBkS#R^HQ0KC0tXIE=8JXP^jZ3J{fo=WE-^_G$S6u0d1YG%*HW?-ntQ)$BX?DO-s
z3$oSLd&fLe&d_<&oP0tl>s`R^pX`2yX26SxJ7*lu<*pUE9$5RhL2R$>&JW$p>}hMU
zO9WrqLbc4-;>CL*gRUmdJqhiLD(494gLw(8H`>JlY1nXR;MS+lQ<In*6c--ti)K@o
z`#RR7VSA*>z41;zA4=BrDwEp(sy%q;o9kUd@T0o4V7cD720aWje(U4ppn3hr{GDE#
z!~rcjd&4%JS;2!xQ*}hV%N|`FK85e&LUSltlDQt|qmLMP|1mGK9DO$GNLqWj-!@^p
z;G@WVb%w8g<S61)T4M-Jyxx9hv(W!_d66{%l^>|#azII@u8SN*+>+{EiDPblrmD45
zrZcE%RQ+U2dbX&YB9kx$TjVZm)8SgiD|>U9p#R6$!+K&^IBViv>(|Q=tJ*DORa@<7
zk<5fe4&_<z%kY_AU-$6my_1z)wN%lRI6Dfj9aSnC6=q#G5Nw*9L`2@)4thOFi@RqH
z1pby1=a;TL$zb(4!$8MH>xsjCr(@z~E#&oThsQq)pF40}d(N)BMYzR3{Y}?NvqmWg
z1aaAX_@{gCO|*bz;xr7#@HWE?)A51g7EPw}36*0T^b_BLr{viWCi+fl;q@M_7fNbl
z=zMdTK000HsoL~FlxnPH12rfs2mcXV>49o9Fw#6s3vpUPV9$gHp$?b9Pp!0LPJs^I
zq{vFUWQ;-Jl(A<9J}u$APy)X1N*7-Xu#;*e%m#iF`rmQtx%K29zte79{{?0uBUR*Z
zJWpU_5kr+A`&;4$Ib8n?W>J6+4cBuzN_A|MKa0oDnwCM>c{`=-nTLa$$@}^?0dL{U
zJ#lTm`l#TupRG_^!XtP#wA{GCXxU+^ICBuHE_p`skiDdcQSaf2|J|9tw?B$%9NcJK
zYe`!rCctgj2Ohef*M5b+N9lQp+pT6$ynKLj!zeuWYjf2+q}^@-xbGG%irw1-{3)Yb
zN^k@<Z6kw@VRSg#2F=n}J*;WB@)P%6X-X8#X?RNLc4t}U4e~49vmHh`zQGKN^o<@K
zG2ctq7lgt`46*HKx4;iE+ea}h%%+Jsa=8cJ?d`hN@e4m#uiR>1^i+#M*V8ic4i{8k
z&R$FFpHg@xY;vxb)6^-xo}#c7CTSl>MCM9Qxwgsf725d^PTc9ED@$I)t2CY%JvYS4
z75zB;snX4L8x-k-%Q{2I2U^KWYsyB+{IrlJ^>5A3(YW;Gm_(JJ(4`lkVm_M?nFY%z
zi|&iT^CEC10U$m?E}|KL*nDHEi)uJq-hBApf%MrcDdXT*h`nDNdquJYBh7@5D&qQc
z<>Txp@Xr_)LMm{(SnH*{ow<(dkJ%S^?nRYb_B)Pj5~jcR7G<h7gGDH}7eM~iy|`Cf
zG(vUhgvbolQ?rhl2!&0rLYxvM1FY+eiT;RE@r(_X_3-rQ8)AVg*vZ0KI&YApn}IWo
z%W+GGikIWhmD_d?^l@S{vyz~L2Yb(jDSuELmu+zryzN^Y&5_;qCCw;lK-q;|_P7j-
z=(Z7<ihSDL_)`-h4uu4YOT|!7?fFY``Rps~uC>Y$H|;q>rLU>KoCkWvFLYcwn(6zJ
zBP;%1L@D(K)*8*>3iRoZ>j>GyE519554xzTPney}4o;@EF3*<R3~b~S`fZ{NzBD3(
zqRvihXx^N^)nX7vD4|tTlMB&RDu{CORg(U>pP42hDb?hb`@_hWaXqGEe65&R>-t+Y
zo;C9#Yw>Mdbm6kk1#bkv%xrpIzhEK-(M=8t-6CMmzTd9u-KTCL0HcS5_Y^4%dwKqv
zDGj<+5B3y$j*O`1jCJS$)6Q1t4B7T>s9^jwkf8|PzZvhpVBSwIlK47`T;gk&jQU=9
zf!adDL#^}j?IO=1$j4J$rsh2zh5auJzQ6^9Ppf)KzQoeG5ekU8o-V4w&uuMM-<KCk
z)bRaHy(q&+l$buM**XvHj7@?$8!%WlHB|Pa^C{>!Qt$TDSC`j@zO3ggM|Od;RARR~
zP6LbRN+Eg>L6&wmijm=+-Dvc`D?;jLA3j%N3S<oaZrOw@IQ@*=fZ6nrAegK^$iX%8
zp<R9Bvm>-}&P#J)`ieAE`f`q}z0eAVW_S#IL%Nq>%-x}*OWd=Ic0*pO!<?m#zo#}l
z+?nC^Y9!y-@9%M`)us^QsU-N`jvT`S<GahLwhiC_=`?9`hVYxz{>g`3nR(Q2Thm5o
zU>4$B!o(u>En}LO_%a<;?lYCcrSv-&+fTD^sGcGnB|o)l)GsutLkfo3)7>oI-A<AH
z*&quRT)jn9?<fj@BUK5&XjPyW;t&Fg%5?LoqktV09IiTB6)g!d)aI7yMnrSknAW+C
z&ZozKIy?1n&hhhhR?=!K0fp8k;AlQVkDBF#rA#LhOFe0VQdEzQAjj^2@3gUTVKQlz
zW@Z>HN{Qx|hT5ySp%@aj987z#lKj54^SueFTa2$okRASS@tBFUQvt3PB4g7Z$cN6{
z0zjucchsqM(>~Q`*-N_^Gf*X5ktr>?y(CzW4I5Md8LisQ@Ra4^Hxd4o91^9Fx|Ry0
zsYNg{ms2S>PFQYmKky6jAZC~HRG$csAKm+;_b2lX<+-uvZgj%<nP)^BV1MMWvvj!h
zFn0u_5x{C)$wT^FpDi{3DdQ>aj+2NGL&?xbmC{n^tVFrTpqR&6=-c(PgH60On||vd
z81a8-I_r2k|No5-#~e&gO&!hA#+aIJrn_UhX6#_<V7j}zr<-HCnVN2<neN$sx6kAI
z`=1B5bKkeu`+8m1^J;m@Sh#)?&p`Q%pjv;NECrwZyB9uejOwJ^B4w3z4sSnA%89=g
z>jd^KU&J87hLw#q)T~17Ta7<|ZxolVcaJauAuJ||7~$8R*F}J&tG4uzmE-Q3H7A#2
znrfE8<M8saaPxr>tWX15T1+<lNQpZ4_TQ>Cd;a;PSE)_@mDGat-+s*wm7#x`QfXMl
zsM4!C9e|*0PHz)c_3sY>KBZ9l_!!y({M*vs7CvUDUJ?uo%k=#8@w$7g9Cvtz?Ay0v
zrnzWWR2=BS4kB9kFhYy<oro{M_81r043n>{^~N3AVJ&1sxV@2to~QD=)JpXXSiNZd
zYwy_v)coSj;@;20qu;8X_d{jXVJmD2EeggM^AM`SZDN%lTs<J&+l~am9iCEKkk&hA
zVpL(B`5ab1TD_P4%c|yybDGB&lN~GdCAkJ9|0$(x5q|Ug{nPkd#mCnatLgzdy04Lq
zni_}^X&<(3JMoYmcQ!3i!wknPy2y#Fp0q=OZO=Z^Fd`1hwc>UnU|Sn$ZFZmlMdn~Y
zk%=J5jsId6OZS?aTh@MC)~szmWiY@Gl$Wn5HoyZrHF+xTHviC-Qv`@S3CY9{TYQ7k
z5Ux-*UGsy6xM%U}1dPg;7M++lww<e$O``K2KnP+Azz|c#TLcCRB8H{(hl1|^?7sU~
zqxMLo5I`IT5U}_engd^9<Q3M?pk#$3AV0e6stBqIP7ToGjOFC)-$Ts=b?}!W)V?9T
z`fUn33f#5j_ywpfGs3(cT=O!5-G_8S*v0TafL_rWtWk~vsyXHgt#D-1rr?-9C1-1l
zYRZsFJOz$AxqxI<n&%MIbMZ>(Zjmr%N3^nDoDyHZTc!^K^n6)xBGnps3isZeg~JNw
z<9_gy*)}`k{t@`y1<o)qRgW;T*dPN402Po_yAy&w3WTEc<-;YLtNPqmFM)WP&W|k_
zahT=v{eBEF?5(f9f3~PV=hm!UFeH#V<M@X<N7h$j|6WAT8i8eB1KG_g;xeIW|Gz5s
zcuu-h)lzHLM66NI7ADHHJtL6%h)h244!hQ)X2~=+X~56&6f<x!#YCfa?#RdWgw&71
zB09q}NdRo2wKie#4LVZ9zyz8;wNs<roU(6Im?g$wt<DSc#Q7)}VWMxk6y@F6LYUt$
z?u{c#S1_{nbwpb&M~IrrF8|E)b`QBu$9D(eM{HH^p|5_Zkk!&k`m1b5UlIEplYxlC
z$Ufu#KD^KJbltY^kfoAZ$SS|x^;LgfcCb+!fh;Tu8)#Ax{To4l>sSN$XnMay!zqvK
z;b)RURTR43r%}ODY?C2Pu<<qbB?so&0vyDxnId*Kf}qP47Zd#(t4V(w%c|)aJDF0o
zS+#%h|FZzfpRH;;uK_M{BibCgsL&gdvHE;eaQ0nsrbOS(c;Pi%Xx0dKe$l|4&GMw(
zchpPa!{=(rM@7u~NMZL1&rMTXowGv42FfJiTc%-wJx*ksnc9$D^=<DaedVY?rNcS3
zA=E8a52UYrD=#)CN#1N@Q%UHeVbctUh)h~1c9OAj<k#8|v&c=h<iF2vY+7+zQ25xd
zER3q_<;zTjwKmKMhwE>zL8e&ii|ZkP3rvp$siHM(xUbs}*nxyFy{)*oeoDKRtmlm|
z=fJ3BiEYj~Qt)H>cp&-Xj;&dMLm=?^lE7D>IxV~^PjTZ;n?XlWML3!G-(a;|e>Kkr
zy|C9Fm)f-5+Ls`}%nYPo{xh0EX7H-d4g1)i-+u3mgMncyv&%ouoXp$S1k80aD)lnR
zS!f3y!k+DHA&3?#(~Xj2G45Y{Pfm6Nycx;X43(7;y}WEw+_+*jK?U;Fg*-x2{y+UH
zUPmIKGWa{kn|PVSQsdA!W6+HF^U-AqA+^%9eWu}x_ATgwfLCl)h!l;Z79uNjv+UX$
z3?<!o-fLmr=B4;SkT3>!<7`%O3ON)Me*k#hXZ`_U^uqU_%-S?y!1|5UEdd?O5)v#U
zQI7oHV9xM!Cb!3Umhe>tv%gI;yW!=V9uO%5BUkR92??Aak1=GC$wm(?qF?5{-Zv|Z
z^uea(>IP`-UN;P%N+2?w3iLvxPv|?faFus4Lt?mmUIBgUXavWP4>xn|zG57M4JjmF
zWgvvHV1mV`+MvITY#R8@Ku&h$b%QPDCq;L-K|LS|AL=t~33UH@{rt=P^~)P(M9T1Q
zf}eOsv@RX~zxFgZI|qc#800p;2j2j-xcB>W{Ebj)VH0A8HlPhq0($ee8ta&J%$#vk
z;)KOui~A(-ROtDaQ5wwG#b=x37KAaC7RV9nxi22+@5d(kuse5v@(<cebckjih2|WR
zkcx8p8>Jgo29^PNlEF`ql{C~jGz9Dz=;iekH0!1!IBQ-tRp$|B*erfetiL7~@_FZD
zgewd)=*A%NSrrgp+W!SIHw6#MHD_l7{YmZbar)$f+U0!}0t1dn9hMw%&1+<lDJv#@
zbJviEkO6!u`{lpA<Wogn%u2G$t$E~yADBp4IP(Vv^rs&p>`2F1zr6(QKS!PtsZ=#6
zhoI#;LTgEm23$3_@vz4!U46sPzD5XnOe^v2V=9ea!(sXOh|!!Xcm-4}8?~yS(;E<>
z?HEqrA%Rwuh?=tgci?ORH|Po=-$1XDx;K!F+&S_%Q1PBS_&wlBYFrkRXm2bKI835|
z5ql@*v2LuSd)#=HC^1grC(}C6vlq!jQpAP@be6f55U&l=P@flb4yvlLmCxspD^dQm
zKVC20lgPXwk~E7U6>HXxBtBgZ@`)jkKpp>5k5x%u4)>C<49-Ovf6Ka94TZRvvRV-(
z?wYu^S_luM^ekr^d>o>m>QjM?4>u;o1>O0ht;#w3mWj-6kcQQao^JG9oxK2O;Idlu
z){)FH^h?%y9RQZ3xpwD(3_C>h(1BkK*J``Q92F4}=^H<bU@3aYV%G;m^12ax;Ni*F
z2}P%{IDa|~Q&9~cRpUP=w-|TF#S)X}yO)g!7K07wyqXi|`N&g=bd-RL(*;`xgnU&Z
zthhZ?VuPIMi%4Ll#TU_W_NR~PAjB%#c1Cy+eG4N3>(9%8hzP5AMN!ti%D1pA(mC9z
zjm5q?pC`mzri$~*FzD*bPO24a2Z}l3Rk&gV7yD{yEcTZ3GGiy^941kGuF#CQas-yM
zaQAPD&&#SGK86Q(F_3z^GACQ^fi>}l+%&J9DStf^IAA^uIAwj^pD4X%UZ@7UenvKv
zg3G_PAR-%6elu^5)BTxD;$43eN5`Mq-}^cZ7q|%dt%Y!k3>^}raa>?_StmrbzH$S_
zr5NLHEN~a(=md@zDUp>N)I3Oh45T7bY>7L6|JUnC2UzuMxF<qp(5u>O3a+pIsde?1
zCJAAnDKuf6|K5MOT&-OC2C&v92;Y(n6ImF-*xz)d|3sWqK|9>bxP5b&-lUSIj#wF#
z*3<50GpcORPbXNG&}bY#i_6y)y`2RP{|wODd{#EVOME7tWAi8gO&7{a^9eKPjrZ>5
z(s1tUV6MpFEb?f{S1wGmqAbptAcnq$OGY<GRIWy3Pb9xI@_L;ZA@a4pY<ME?ax2Yh
zS8cV_N=>ikN@@>Jo*!ZgATBoO%DlFNWW8+dN*p)J6Ig`G>;CgeOf2GtT#t=26Tz~Z
zHgU&FeSYk|K&o2OyISc_b;}q*N3K0_QNAScNa?hL#Oiba%?KCEkQoENT~zuOaYp4>
zGtIZEJIH1WLuc%GxD=n%LLhcFX5tzIGm*Ch7UD`VwmdcTHlwZJ^KT4bQ9OYhV$a^d
zPZ|{PQ--ndKW2IQ1o0_pkWT3te_65SiC&95wd9oQS>Q_)g=HZKG3KklO?sgr8#A-l
zWy)L|CFprkYS|Df=`pHT%!7<+r;c<p^dcW^;3Z|D=yMaScmtc%Iq2s}8Zy=^&a7ZX
zyt^?Zrtu7%U3iR@GE`WGY~(!nYiMvTgU~<O3o?Houn%Vd3(Fgtagq};h6tpk$nw-G
z_@aiY5D>(e-o0q+4J>64DwMxSML}iOIb3Wg*TA_%|Cr6dusOCs^8-~^o`5pw>Y=-u
z_EH;nm51vsS?qfvh^&vim$=*ShB>3OQ)g!NN%F6U+>723?qaXXX+qG|mU<i8v@~H^
z$}wk!Z;)8T1}8vBeIHZJXbcF9-X`8pXorU0rrZY(2oy$5ym`6mz$<nTxe$%P9rd+V
z;^Hf1_;Hx|q6Ho9(iQXytQN!&GAgaM&5@-F2j~1GuuX^bK(B5IsbwjL{~|SR+(8Ca
zxWZ+!S!FqgJu#{Z%vP(piYVziFNnrhRO}<n#mZ(tvHfwBbns)ibpbh}Xcfd$ca~9B
zM6K;;Sy0*(pc>R_5ZiadEPgMyNKch?m?6b6$kNIiHb+%}D^XE)tvmMeQCp;sj7T-g
zq?V5s>UhOU)DK=3ytVY{paEpzyn4p^AMluOp+VyJ7lb$TS!3OH?F<Oe!MK?2$|QEK
zfOzyITF!Vqv^fieE?bH-?kIMzSq6wyHmg94S2I<Hh#EYkCY<XQkm(M3gjxG+@z354
zDwFyY^$7-<*4=62PCF2?g%qn9?!DilFUf|GDIAM?qq-;)WPLJEV?;#Qnx%UdW&!I;
z;WBKoRn46~Z+k1DuQfQoZCt%bclW|8LLe(pH<H#<7NV&>@)H!ROXl1=@#r&0J5(zc
z_a!-65lS046MKUKYf9MN4?Mjj9ogdC3XKy;NtKQUZkm(_tH`DFeBkFi9r&&_Q)NF_
ze}Y-8%D1DjGq_?IxlD*uOHw}{2|+mOOEoVO;?)|?uMIS2xxoxXZBVyRsmwHphQ8{y
z?yK@G53&1M5YpQZE%3o|bF=!?J8hUtRB4awGuAxQXs!gEOG~=w`v>3v&eA~|N`Y@L
ze3(zv8yn##ngb}g4v&7nr#gz_D!!X(T=b7jeKdW)^>uH}cDTs6bX4E-aa*nJP{gpz
zf#2RQtIFaWR&hQglPCbCY*r}iA=mLd)+mgMlsU@R?BF-X40BSiu^0(dPw*W6U_i>!
zrqyvyX8E^*II9$u?K^EcCbZ?&pwwvIE~%&@z0l_B+JNFwzgm|VaL%wriazog+ong&
zYcKKV$lR_Wwb=l!glY%ntJm)X5Ai&kB19G&DOpG-#wF(;J3KX(-Qo!eGZ^G>=ec)u
zZ1kOk?LNi*^c(wcYV6bpof=O3p**=>hH`W-rByfTfOR&c=WV<;su{Fm$xmJ*uV>EN
z)alp+O@p%8K3pl|;zrR3IanE>x$jbC&SzqH2mqeDx7W)27^QxecI7BSS#$`$7j!~o
zavUEOMZVu3ls6vu{g8hn7NU$+pq2=B%6k=3TpOHlDr;p|OJJE})SaHHK{Z@0pA;#z
zjyj|SJ5|?Yp&^eS>z*#3VlOW!W2r)@mG-dTiNU1Q4!9R-@xTa3lOBB`2F^RK!@ySb
zDybw!E3r-9Q_g?zPZm;7s>4qtd^jQ8)B0_9e*V``<vL-0;;dWocxbda9*nDn1Ce3_
z!O~B(4#W`rH7-eais7WDh7e<4t2PbL4C$WoUuU+H2+VySNw3B5r=|oc9Q+LfoC8@K
zIl?zmWGfZ(WkldAxcPlC365YyOdNo0QiXF(*_75JwPbJ|v%If$6?B3ax*v-?Db2(p
z0RfElCGi$XAX&KumFwGX+IkJfPnhG2JHLiqtH6lKdy5V4aeNzrpR%mIDtLWj&-;-h
zbj*S}SrEv2Q<e$zLe0<t&F+Wl*l&G46}YgVWXJ8-D0GmRbIV%bh^SD-74vA$78kcK
zp;KRzGao6MsMK>Y)9fVf8&oV|NvSKzGnJo9j9GgWa_UH8O4dK@K7i{!e96sruay6L
z$GFXW{^|NTwubM~qqkK<A3kJ=t@qw`M-cqrqEtoC+9=x%ISs2CCgLSG$HG0eY&1cb
zSa!t{BA)<gGY7e3({HC5tPSYyMs72j(zQengn}5d$6a0x^OHTX6t~WdJrt%S1NwLi
z4=seqRi|)Pwx!|QpZ3Y8JFGK7p1pc>V8hA9SLW*FC!t>rn>(sSRf8%LZ;U09i959V
zFU%O9k6^}{5`ee6z5ME<NQyDu|4UMgZX~hn4n$9zZ|{nuq{ZvcEBgyFM<SB$WC7VG
zvuVY%o5xtMEUGk^WYr#-7*?ox_hgsXUwK_6?5qgwkg(~L|5yUd*~;#`2~KZ!=}NPd
z2B<#Jss4QffXk<%BL4|V+AUi=F3Iq)4qL5$tM3Y7YPSumrzFGPFXnZOXjhq?Ah<6l
z_;|C=Hly!3R+uYywe?lALw>rFYF+|!cEBBmmm7mZ;!h-=<2dxN*pG5AjZ4cH?vt~@
zc!+5sF@~k3ZIk}|DtbcqU*c8JX6}6+h55$GtP7)b0;<|cQNV+ZQsRMB%jD%pgH39c
z$0Gw9!Tbs`k6g)3UK0IYz#9CW*6+q`<_4p9W8JixjuQE{@}yqL@{SS^%X-T#Sn8P4
zrl0Ai_hHlt`u-eVkJ5MR`_YrN+SZKy&F3&$zmOouWK^Pb=eya8LK$o{o^IriEi#Xo
zT6zovJ3Ns13{pHU@}OwT*vLR=p>C4_Jq@9a${}-X7JC(WY$6pdztw9|!Go<|7=|w1
zt(ET%`u)qa1F|k(XR3b&C?|4T-=62a5c`n_L=Sbw@|<Xy!4bm_(7>kll-6}31K3Jj
z5kc>(Cp<)<OIattz%^Lg(I(6WZ@`(9y6ZB&^Oh5OP5hpRzfJ>lC<Pw7MY74{7fNM7
zg!Yzhho?d{No&`uIqe0*4y738k$r#%J@9ZP3LKrm<NNoE4bcW$WG__7CL>c+taqyE
zju3}qWC7$66nMj9<H<7D%wvuW!uVM7(eCqxVO*fLl8n|WaEPP}Nt{9$)l)ZC_h~K7
z7cG?KJWfP$1B_)S*%ykIs-~tT$@b}=FEi(*`DTS*l_|yqT(Y&gtTDu`w=k5`+P};i
zv;JA4+BnwNxM&vpl_wiA90%!;+lmmikEInNBP%xTn8z@TG3f68mIXEwK^Jf2X(65~
z)u^GtHAlB^U?mU(Pc}soWno)R4s+I+y<a;gzZ<IsqWJi$7GqtF<2Y}B;d(z<3r&9u
zJ6*22>8ic7wyJR!B_eF%G+=D@(Hwsv<XSV+s>|7k2V!@v_KC78hOo(PnW18AlI1#B
z;vcuJ0IsfryTD*E?{u$J5m%B%PrhVoBi6AY@II$>HH{e;O4su`6NKaE7pAjayT9>p
zC0)0@aI?_N_+mpU3iA`6k_g?m;xX@N<Iv;S(Ye1!@+0?FGELn;YwnGG-^G0h!($@i
zW?Fk!-4K83P9`C7WR2(viPoSK8XQpm)A7=b4^Mmz8lObz8Wt^d3BWdo6{&?fP`OlJ
z5woboFv_uV#xlIV;hK_UN49)CR~9^$gzYx506Q%09WTHrl3zw3*uHlnUn>a{3F%8n
zS|h-sXzqAw<-x~9ipZ%p79#<sOaduRAEK-q^jm$xnnQS&%=LE`5%)M99z69KATf=;
z#$iE&2s{+9@J?WD$NO4Sa!SKW!Lptf?j}s)d7F$E5+o{m{)gw>`f75mG~$!8Z^Q6M
zn^{pK5g;&A=mP!*SXlE+R9+eLU%kSxD8DHBe4q%#7#x12^X**jY)u3d&2*?wFYb*W
zmjzqN9fRu~WcuoMR6NlR*F<V@G~d0R!uYi&_QEUujZ?v5lM%0Sj}IoJvDiMvar=ds
z7?3d^wla|VIgeX3&Ofw=cgLn$&h$F3#|OnBoH-_ykf@(ALNpJe@^G)9@x~Iqr(@fr
zIJ|rUf%<H|qidV(!G|g%w|FFY|8&8tY4~S|3#56Uos2)3z&Gik;)z<Cat$0F#vdfb
z$R}vP7F5(br!OyPfN110%^PRx{~XA4_{i?xBXt|}kNPcnnTzejo+3PsyF~UvQmRn)
z1R9S~lZ6bmQN@hT&!AI>1v_Rowkti$;r*VB=D8b^TVbJ&w2#SQ_2RP&j#Ec$1WX3|
z3Yf3n*pr}hxJ)(c_(NicL+l||th%*yG=kt>Gp~aXXAn5rmriI)7PwP4{oO~{nq(Qj
z#%RN`A`5A3g}|B03>;iTKt_^L-tDRKXX<pK6UeiLIyr4nt>NOBEdf1HRB1oYxn$>3
z))h4HDC-A*1mHSEqd!{1?IeIxAlo?LX)T;IJH)erSwk3pTjIRLrCG-9i4r(qq7Nmv
zY{kQU7X(z4Lr=S@5xepix-I>Mqy!KZso`f`%F?oRwzWFKGTr%kJa>x?c7+@6$WJH+
zWTe?jKiT+-6iEG4l#V1d!G?cK?`&^vNr@?dLk4k6(S#mQqR8#mj<CDhLy5Eo!6hUq
zo20l()q7Ic5q?ir2I-Mkyx5P0r$lT@TYYX&&2QkBZ759W(N7SpA=zdOP17#ek97}P
zf@PWRwhq=|HIgg^z>gThW|o#qvIR{AJMqXPx3IRanRctQn~Eh7s79iQ=}Ax>YD^TB
zS$(cQB$8xapy6EBH`D*9uoJ?zNIPY^%yg(-5tr!fJ5nhTv81~h-3cfc?;(AusNQje
zc^Pl1esFzPX^$_>LMuDN04LNQzSpW`RXm~-Yv!}SKlX_$<O}w<uF}h~R<;w#7V@Zl
zEBh{3N$2O38vMK1bu89a^0czxSPkhdtUwR<7txKiG<tNk+8<HSXK1YpnCMF{4UIVx
zK9OJw>bMqe!<opoWVN4ig9#P^rOr_*6Wp!@X!&X!>w>?Xsk;LfZ_zDGsOC-!&G=j0
zZ(Tv5?<X01xjz9k+IwVO-B!Wh%@RbTrV)M)AW?4hP%XW@uyZ<cI@QzyLYb2)Tcdwf
ztYntQuFouwZBCY~uf>AnV5bUKGrqY-F-y${+GXAnjxrnn45dj8Tjku!kf)bE=hVBO
zhgMiSfV>;35Rj0@*b@^{!4vVf1CNh}glmiNLjBT>;*-+f`8TD~G51ia_T*uw?@98-
zMxyk)k%erG|Ki_Jv`{$+e1Xig-y|&?oEZ#>?45=*o3y|?(=!QTlYmh`QZ(m&|E&hs
zaa_m)BfwUIwm#s=+o}H=9tC1GlE12ej%phWvoDfnWJsyr?)gXNzfRhC$A!eq!U@P0
z`EBxOVi<@dEslMwe<E5xuf}}+A5Ee(!f&oX*`%-cM=!$czX$fo-mC4&Tzd!MxT<5f
z0`osoQdHDSeq_KK`vL0@HM|w#pcJA=n)Mg$CV+7<Gt&Od^T@y#izSYafHVzDzM1?+
zgBJise(&fAiZ>|EX1@_YimucW&Kk0RYHG0TO6tbuUw>vJz_MB{Y7jRB7!bq(LUwIQ
zis>L)w00;{%4HUhO`OAYo{}@*$N4)PdTNb|Mb5ACiedzXgR=R_`M`7XgwLOAC2QF)
zN-v;Xaqbuj0cnkVgtz7DyG8b7CDEyqT*J;-8od(ZPJReoD9#<UeLi8IL8@|hl(b({
z4yK}fi)1MU9G%PTM6`@i)!M%K_KlPfs5P5%AQZ!nGa{#(|90DvRif8D<HM!Nz<-s4
z<6RqOj~A|p938>^-WuZ5Y$&!f=Vsf_IolG(kM%J%%z|8@7|$0lE5*$_F7iA4?c=SY
z>EmS#`{>t41jXrBt=(7E$;tQh4{_t@pE|Y_$|6-B@;Afeuvvv&R}D}Q?`wi#Kt0d|
zzJ2Ydy&j9lGEL+Y;LL)r`bJhmhSI}Ak{_*J2zmZdl%bhVBP+*pf6TP(GbuvCe~=0a
zcwzw%?tO?0byf#J4Z>2geIOCgSgl@((iUZJY^(lJ;%7S=dYg1a3o!ZEy@`A3<gN|g
z95tPjMekM}Y;hu1v@|qyxCCj&fM@6ZDN}?L*C|bOjhT{C<hk&WiXR1z1+sSl8cvdw
zE;bpTW5&l_T-*~&OinhI8z$=C!ZXbi@_0I6x{JTje~~HMG~uY|Wqq#mg;EV$Ve;8;
z$=jF*`>~qen6(?Hy8+odhy)}JZPjE*OP6T8I{~zcO_6B%)3+r9FzYnmZ5{ji23pOJ
zuC~dK*y&vq^N{;%!f&A$$QhIQ#M|Vr4N^Ju0>e7V*l;!vM83$Rfo_wO$pIh`cD~X8
zr6SqQp+x%g6D9rk0!i{5LgB;z9y!Wd|0$R0p3=><KQPSrwkEvt{SyUG=aai4^<N11
zsv<o%_*qOzsJvkFtoK{}3t$UkPor|R=dI!P5OHc?R#|kqm|kkKN&m9Crbn+iu47!Z
zOlgVzOPo4^<jvRj$&Z<XJ+$pOFWopX??>-(h*JaZgyirBMGj!4+e&%=9Q7W9d7@DM
zi;T%?f=($q?c`=XbgNusdAyC9c$-}R=<~CTEE(!qU`uQoD>Ocfnnp0#1+)#|Qy?*%
zN)H5^=hMcVf{YX+czAiGej=}`WH{D1Frlb|SBtO#EAW4y34Idtkx!{I;)FIx1z1fM
zwDTxtW%g!Z7l_H*Siq8u2Ur53C-F5}(*EFk-~*O<iJ1Y=PATvl8J`stBZL})@daBE
zikP((@m*12N1&w)n6k7;s+CRPzw0HZTUCro7e(cCR9Kq>-(K$(T^r>Fn_S~e;{x_f
z3mMHa;KQs$_?jE;LcKOAM1T=`1qCQ)DZl)Vk`=$fthu8!T3G=#`?}=clKw`LBI8;U
z_I(hP#kmrbk~1AiESRdK+@lfBe7*A7crNhy{3pO|C3}(!XuxrKSes&BI~4?kX1q_k
z(0F8EG68%KPdnTSTUU{e%IO_sR`Q&GDjuS!9~@D~$q03ocT{I=J)q+cS{H)t#1#iO
zG?`@hghKAx+S=2A?^i6Gh;6E!FyP1+C|zlNGq&$mex8KM(Od%x<XE_ZqY{$LrJx2j
z>(FuG^44rO>*mw}>@M=rUABm4A<ng2#E9^+`<LYmK-mn@uTC<K!5qN#ise567H_(1
zWoBoel`Dr~ca}5D_UT+ci%wE*j}Knlzx2VFp%!$XmN7HPr6Fqil|pdR!h}N=>&#|y
zM!7gV2Iog%>2T(|9P!y`F}`*T`TPpj>qgW3+(p=|7@O7J|MR@gi3cP&fSiqhqyLSL
zbI^*lMJ*7SxCc%a7OHHo+<|UsyzlS$3BO>fw!50C*2#<9{UTVbms55}Ia{bO@t*=3
z?rDDJ7^{2JhRKcsFp8I;DU3xXb4`%G<@0nrNbL|^g*l^cHD%E`M*4Se?cv3qg_{Au
z?kqQAw+Ok>$pgR&$88_q2Ll8mH|0O-u(BDtRX3~vx9C=5U(<}QkI#c10c%2T|B050
z^P}E|Wj!_P>JT~ti8$k(R2pwgJPTIwy?_UQFgLP`c+SCwSBYNCoCG~&l`#B+h&)j9
zoV$NSX+?#<&zb|kgO1B;5=7cBt`lC{C-AU^+_RSYi(dN&$RfyPS=5^b7sYLDKKyS@
ziH-+^zR<xb^QkGs@HvOEWAcCQS}t)JdRp28Cz;AM7q6)X+mtC-%8t6p2^2nomigH=
zY*qm;zosgAmKj)*9aFg)qpXqT8XoZ#{evQ`4?iYm-A#$XD7Kz67c;tZ-e{T?f@GQw
zYono6Se*19a!c){IrQhDp*0L(I(Ap1)DNNbI;sK1nbFAcROw9~MM$r&*f8GFN%MGz
zsN-3fVTD)*V($eIi*U#uOa$4T*Nuzxn5J}&yjh&7T>q|$msGKbY>HYQZSngr@bG9|
z4zOSy1Py)oL*YJenw6TnXxqi6b#_p$0b0eQm2U2TAs{;M93H7Jp>cBsT*a<dLl)KQ
zz59Y?GI$d#iv{F3@E!qThelEWaWKH7KJ<n7c+mS87P}wmH`=FOz76qm1(E_a`SPh+
zZfhm(Y${NUP?h}`a`;yh`_0s4YB;nTZwbub@xkBp)}i?V9?FmTP}ZvHL!$97G3H~B
z3xM(84Q^4NED+%^wq*iaLPJ*vE*Rla_j}SSTKF1VCwvK-6PZf2oaelA&)fO*<u?bi
z&QIzKomN0!QU{i>282-}-hE+&ji1&@;VVpe$Fimh8~^YH+q3f$gQC-brEX*r-3Rjf
z!FT^6E*VIoYs8C{Gr_*fYo->7S16x#-|le2be^I)V_iHz!Exq%=*q`x{jhBKjoZQl
z2d8^0)3p5wuwrxLJ_9Vf)=^W$Sf-|CLfgPu(cMr;_MCgut-dDG_8aD=+r#y-;Svay
zglY&e(!<UQvN8oNLvWdxG(O;HC9rHELcorRUP@g=8_?2+J#QP0>-GGC4J~98=(r))
z$8Nh787gp7Q*3;IO=e^M(IZZcT7gg`$;hB&d3hr0p{bj~JGvS*Inon1$@D|V!Uzg4
z*_r3PZBdGF{Wt2>tzlE=8OL84riXw5R2@`C*_>WRl}eZ-4B(tKKHcOeOq%ajJ98*=
z-w3AIW_i#kY6{=uX43(To|5ar-H=xYJ@VE%jVVvDUSR$V3c8Rxr}hhasrx`7d37Qq
zuv}`42s!UR2R;S8M7AeJRWZ0%@90R<>6?c=qrlr}um4VERyPitS^;yQRoIs`_i!=0
zq3E}g926nVE8Eq)>%Zw&pBqi`TS~wBtAqub!f$^Cc{U`uy{(iq>p;TrYcwlAhk!sl
zh2&gbBYy5wc~5fL3}lwCXeJ!Ek*mbuDEIoSa@FAbuOb{imikKilVlFux{boEgqcue
zr*DX|P}QmsA0}frrLy5D%Q|`EZ(nlHl&daOC@MBc%7Zkr`YbFUf+@`oGK>x(B8J=4
zGvVJc-uTbIdhcI6!SVf8+*X`sZB8wK?RocBmhJNTwLpV4JWFw=cq1)j_GF1!!fmZH
zy)b^p8)k~;G9Ag^LcEe-#mA8~V^l@dWEa3H+~^UECYGwx;utOXn_pK;ubH#;hB#G5
z!hABgEIMS{K}Fgbd0*Eu|EDwdB^HwAzmkLZUtagh^Z|w6m=*YKxP!gJk8~r!oH3tQ
zuBk#-Wy%g5>B)_VG$iHun)R8<vWzX^HLE~>MIMJWnV>tXFhGL(l^pyGW|@pWc(#Yq
zjz4pQJd-k8{?UBDOEPM@uQ<0=(|Zz#SIGn6EF7i9FW+9dmS&lckvfo<9H}KbNboXR
zf|&e{XX*7TQC=km8&sKKIz}3NLR|s=p40tTKQ9m8Pi?vy;KG;<O(1xQ6mC?KaYZr0
zyVbPxdiEeWci(AOmXO{MJA;LfloOZ?N`!9YW#4dREusE=CIAL!6|5o5{~C@+X9rg}
zP-UKAdJtcYExJNt=Z7E~@bZ5-MOS~93HFyh;TyqZz7oxSF%bMW1fY&(9YM7?i8p;%
z#@@<ZWy8NJ$@Tw(SNUOq)>(V(Oaa_@h-W^IG6~|PqZP;Gr=k%%WTya_Q<6(*hVDQM
zLh{)Uymo-!(Jvj3j;P_<d}#~%Wh^(dug~+GqTN<dk^`)z#S(x{o7#M(VKrMyV{cZA
z!PjbNR;2ILRCa!f8^r)%K@+9Dwh0GWgw<#%o}1|#Fd{oyLR|#V4|?_47hmeHj(`Zp
zjg%1Oi@yObQ4%4=>(OK0DC7*IU9L!lP0&2@BVe1D7LMjHY)v@50VX3acZ!r;m!&A>
zMO=_fjPMxzK}ITyvpCW%>d<hzU0BPAV-r|Qt+YY7%QlaaD$e`&FTXdc{{@QO+@f=u
zMO-rBLUeZ8f&%Tb<uvAINMYK|JrDeUM=3S9+~LHEu=Xa}(-)y^tt%n%UXw~5?iT{y
zm5XeQdY<(-gF$;_&MR=0P&>Zad1#t4tX1R$aX}*zV<HpydR$07Ra5#jh|d_@SO{bT
z1^r);LLsNhSwe4ln2vO$*HJ@<Bb)wIBj5HCA_Lf=m>s8~8)zvKHj15{-97+WnkFXc
zdbkLt@{#2}DZriywsaNL@~@RVq3wZ^Z-Gv{*1lXsu14qLpj@Bb-vS=!;%T73qd?F$
z=ar?U<t31FJMe}5*jRaP<N}m8lu@;LU+2C6ZsX!`p<H!_vJ+Gwo><4~5gybP&_pb0
z33K_n3yU}zg`>)!TH6!Nq?FUIn)=@FGm5|aHBeC4lJzBP5g^&t$=caRloagbqZviW
zi47*o>A48%a`JvTOXU%YF%~6lzn9(m273PT1cDUAxPHGqQea!Rf)7)oe4UP|isvN-
z2Y66+%0!;e)Xd^f8~3%4gB$|2Hv*V<OH2!#mMEo6AsiU~O|>-g%28&ZKQcM@Lkk*@
zr^2SFg+RB4!wG;IJkIo1$5AihM+c8j@2`9=Vwi4UJe~O{fAZ-YO>)~h!iAvZI_M6*
zK_%j=_OQ0gbb<(HQo2lK9f}l9+Dk5^$=b%sW-mMvlPH)3KsismjWTM#lV=blGu0+-
z{3<5}qlN4;&lG^L%y!7@PYmu-jU4uuI-pJw;yq9640k$aoy0rc(goOD`Q};jHRwv~
zE47y4348FN_Z$qW6>sQ{V$d58FV!>%J~E-aDPi!|eADbLE08&=5sQML`O$oZThcwE
z)U5w)%rZbS#NnI1Jdi+yw$G{jMQ6-dZzI*I2yn1EHgYvucWhEz>$1*8<}_^rVzAx4
zufci{%5|JwNKgGsbbQLT<?l@VHWHfeW?EdW=B`PUNKU>FD<1J2bvr4%xSXhhQCqX%
zsyk=CnuGdiR~nSYe?&0#*ik@_tj4pHZZGsROx9lAtH&hhi%)dGmp>RZ?diLWZIVBh
z&v7{IHHqRs0NKo*vK!zuK|o|ND66@*Wx2|Yh0O61Xa|>V64YRJ4|;}f*r@Z=CNhEU
zoxATMWIxa8ikm%vjx<v@Ue47@HPiirsySHvgfZhE5DUD&3bjMf7w}%jeju-jneANp
z^XF<LzGC4y%PycL@Hnjs30%*cU)_XSdz^kp>gM`e39zqhmm1k}*93vk=YEQL)Fd&7
z<VaOm_{|Nn<{i|l9s6XqBHao)Ni_sU)Cz2Rw0YDm2Ru1vM1c$B-05~r3Bq#@dM2bj
zqajlQzL#x%QK>IMmsP{9iVyR>H%iLjcHdW+Bq~;T9RzyAbUA!vy;z$H#~)9%p%75p
zp?s96UwrqI4)aT{)bvmD$o_2K5GE$bcikqhOWiV$E2%Q$#*~A1gn}c?vYn83Xkm+L
zo^9BH0tIf%050iYz*N&onHODOm8Lxk>bc?1O97?>B2ytWPGk{?%EjiFvev_X&M{K&
z2i5=3w@Ruy%~Xt}!d_CiFtQFjW!A6v66;E9?t&|rPRFkAgEVEzD`_%&y{4M0okiqC
zKvRnibJ^Yioiv2=yU%_wz)}?rdIG3G67dEJ!Uqv_r?plCy$HRDA4V8Wb;;HO)7!kv
zWLoQ9HN<NZnKL1-rMT529P0aA-C{pKgho9BAVX0&1JW{(Tyz8!SyAEh-`XLX&V!jZ
z^U?%;G^bB2O<CLhra~9)2(^ZEZ5AQDFhBF31H)BwbI9OOt#?!UH5^)Lu=m~O=Y*81
zKxZ4($j|CqnBkVGG~Ez7So<l_-Q`6iYXphyGS@)R8#(NKeurwXwW8hk@ZDWc1OfjJ
zxM%O(Fiw1iZn@X%e;chtD-Q#CmOahPiamd;cke)f)j(t~R5V5EAI3)yH#a);p6FVv
zOdt<#0P-g;$a~sf-Ei6zy!P2(oBs1+Z=sH{D?E=H&_Id+(A`gdAh7S^02rh|PE#LZ
zJ=UaNF`TBwhOM<6EF{a`yc>i8b=x!ONzJqCHLZoRt&@5Tv7BKaiqo5c^%mZ5^Goy7
zyllnteL$3zP2;soQl77OZC`&x*)spLv4ZKn6v~(-n=c)i2b=}ExHHcF8s$LOKB5{h
zT*h^lx6WCF$2&(Z+BhZ8Bzzc12}eh$S@TakN8YDs%<YdG?2Ar|UUD7@DgHWS1OLJh
zzn-L2YchPaqW#K|zS`zd&-PaFw<=$S!>@VaX46a!haE?+*1uni<Ya$Nb^j#}InmM6
zwULYiPJpU<E?&d$u2>&&RucmX=?gTI2ns`;yl+>N(`Asv3|@rvJs8QQoz9jgwBGtY
zs4n*7F-M9XcWak^?AJ7P8=d!lZ!37n1^>L96cdrvrb5w9qNR~PMC+tefX3)yl~o+?
zHik)WyCW>!Li--?#-|9ausRGz^Na(bhI}%huxUe-Joeu^HJtVteolM~I5BsKpSYk*
z0UG3`bp6Xc>fg-9z#raw<Tit%TiV!rBA8O=z!#>BmtCdL!^x3=K7l`vx>0V>ubwxB
z6KJ0_jPejPa6MjKUg{lQ8v;td)8J6ax<Q6-qc)?v+$YvMJU$uagGfdVWQ`@Gqobc1
zR_f=jw3K)}4Wo?jGYpK4mf-%GhR9BcN|&m`-^C&burLtf{{(nV@es5OX}#rGkH3rk
z>a@MHrr(6$csuO&C`m}52tQs<e_C>Ak`smr^g)d7;?TWN@?k6C(syiU;ao|{$Z@~T
zqlKkEFS`VPeE7sM5?*xpl-g*D;VVX>rd$e1->zs{<wn~#Vv<1FNW>3(6GI@5iYaCR
zS_J~}ygaEsFW&zGDw_c6rQa<ihUrABcClxg<#*v**PZ%#_GjFJ8m6U-GMfQQa8%68
zT@LW{5bJK17yEI`Zq_G2I+z6JAG<c_-XQ=*EBdFDDUb-Tz2#qj7@ctISDRH^!_>g^
z+0dj2=RVCM$5CC_?j@S>YOuGb$8=RIN*8kPZ*v^pK$J#`F1}g9mqSc+>*pI!0Wf4W
zw|n=0G5w`NKR+BDefeCQrfnRP6x4Z1L6b?s<FE)(Izh!#(R{~6l9VLZz&qt!w`&sf
zX*%e$VRv>kKJF56{QSnCB8g;dTJX}V2OU(cOf&FL%#S{OGs)eNZmwgW#@lZGyz{a`
zwo<?QcXnd<!K!~$r!F3hj;@*%*FL5!E?+J`MH7078HCH0@TT@syC6@ETEV79GAzNK
zd5P2fy5;6i_IT>=(M+#w*E!c+_KU~wMxS>wb~7$Ac0IBmGsZ7GR#|z^*tK4u84WPS
zWVll7o>?txv_Si=NEH`G)|g`>HF@*~?<@%WR)lgS1)>9~{7V|O3D`XQO4n!2Wd=X~
zVM;CYS|yoaCQvp(R`f#;#6xNi$~+o<l}47`<ZA6m>?=<Qt|Rl2*%>E7U#|7UgsG}!
zl<Rhuw~rxHug@&W>9$=8(a(b-y4d~7uk14S6drmKX2ptburPJ38(?Qm;U3x<R~gUo
zh(sX6;^W6e_NE!;kLxL9KkzWH_4NjH`CYL-EKm=^fLTuSrn<&_;GTXpl8DNG{WlD_
z9Z~D#TPyVsrTnsXnc0xwm4q&_y4BHk@1yroF^O;ksQ~=yjRL}s=T#}7Ci8q4B_Y<3
zw^D1a=Ad|$TSs(6>j{!Hnl4gOU$F)7pvTEDVyrmN%e`ETAAgvaERzkTxzT!oSF}MI
zTVQ(>Qdgz7Mwvz^@SgjSU4ZQiggTUMa9>NzACeq>Ka{0|Z8dW+McQRI@e29;;lDaw
zN(PpGvQ^*j%XVA?8Sl)p+W}-{iKW87)`@Ei!&uI7V4f|<I6}jr{3c-cpvMS)1X^fi
zaNC~?PBfbgL{ZpoHE31n-5HQK1DDto6*B<zV3fCFG!H8tX33n~pz57qR0CdBbgjLG
zYTd*3U`0u)qX)Lr*~ROwMIyxNDV>R80v~tX2k^k`v!LzzWhLVB*IKekRSgj1G)Gw`
zMy1cmCBL?b1T^G7@MyJ%<gvRGt>`v@5V%$Mq-ofQO>!1YDFYS@Q(X48E{uHVtHhwV
zY(|A09tK=aWK=X9mHTF|(4!w@I}Bq?Hh<GCF3&PH>`ay14h7lDLN3aA9}()US?Fg?
zYqW$=tM)r+S86&f0$tK@Hbb_#bc80_5N4^vABoQC52nv);*7}Dm2j2WI3`b9>dRKM
zX*a)Qe9snu;(Pwf@{uv;<}4iltYz`aeX=cB3vq+<nJvxEOlT%cR95iRmifkX#4s)Q
zN(_q2$XJ1McnhJx_9p<thD3_??_A3Lj6dzS3R)W?T)0Zqq?)@D#P^spSc5jdUP7I}
z0qdBUCsLaFTWX>9r&@KO+ou3foz#Kj0r&x&!7y3jMkfh~J%uwN1s4gkE?8W05Xv9q
z{;hUc&$$*PaD*kFT)ta*l#yE*lL!x2w|7C##1BxtNXNDBhHe?xO!3dgK-ab2=yT^&
zxbu6*JY*0>(^NKc?8y|<?cqd`M5+PQ-k|GUZ60~?n`1!OJ!gKuByMeeHl@I^AU*TX
zIli~O&^rL``}b}@jupq{brki$-c%vx@oIaU<awjR5|ef_Y$5y`<N%dyzyzs^Hg-L7
zD+_y9X~7+OY{fzr1Qg_^KK0!VMA@`GTii)}tkSJ3f>zrDBc!QuA1~4aH+d=b=C2q6
zQQAPDHh3CXhH_+gDM6#WLmpfxd#-L4?W*XPwI6||qRVbTWY!fpjh{Sy!H&44;CB{v
zBQXMs5Fdd-Ryk_(w>%2<2@r{ve&XAB1D%H5^D6vg^2p;IwZyKT9UK#$+I|n&Cf|Ff
zZ97@Sw%^Ewryi|p9vTs|FjqMC)vxNdR4|Qqf{wr>JgpQO>=yH5g(o>5RyU2ha4V?4
zHn>jKb3`8evCUrJI1NfRQZdA^tQ#;|Qc6<PqtTbJ&-mo?o`4XGZ>4*C8`&rX7i*gU
zU0n2m3HNt+7e+``1-FYkB_=GOL1}o?6IfK{Uf`^Xg~9l2FkO546E4q}ELgQylQeAq
zMV4-+V+7t}?{f0T+J&xOF=cQy(DX&W#Om4^XN}F&{KmH%VEK}d^P?iZyhe3I1j(rC
zOK&N%;icNH(teLM2R!W2fQ+fP{^*=Y47j17%Qs7LS1RP8%iR^qh|C^`@D>gy2suom
zhS=nWj>Z?e9g9I!ar|>vY%}WRY)#EqDE<xpu|QD0pZ0a^bIr;7l6*s@e?``b5)5ir
zPD(*ZTeU(^?MhY!?4f$wnKC(4gdP=ndj&{DU?fm=b*JZom|&t+HDKZh5D@8Hg`h@h
z!QzpffOal~f3gW*Yj!b*yVAD4?Ep{%GcL`!-9sZ>5VMekP7iq^&&=YOrYw)4;RAN1
z!{)E)%Y18Jwj9s=-apk-3IpH%#}(LdmYC%NCB9`3%-1tO+n(&WQf_rNy!}W3wW;U#
zN*h`RTNT6^1OZ=2A~0i#uhm1A?RT{!=BkX12}uN!RH7e=@}y*tI4ECkMA&f@JQXlz
zO4{R!6crT-5GfE!{Yc|2mlUA<(eZe#fAF(ZT~#2^A<C3MTcGFyb_!lk-RTV_P)20}
z?(CP(#6LERWQ%w&mI_6o^A~UGbU$MJ*0;Z5(ZeFY;?E)<!Ij%mFIK*iMPQ0CRW__U
zw*7^_s~oEBXUZL<u*~E%h2IN$oYQqBl4D^zQwBcN71sb2oUH)02z|~)*P25A>5gq}
zdU4%`ENr}JJEMB7EDhYNOW?@&^{<YG8mWu9{7vvVt@&@6Q~Ngg+DD4%19_8@1)Iwb
z=bZ1~gzWM<(cUaMF^%Nf-Q@{M`naHcC5l|cA>rxuMGgK}pFMVHDm8^WmJ}++gFS_z
zex;{zy0nTP&*uAA`R?IHH3643$(;mRAgeuujZkCFAbEx?uf>q*YyDC;C}fVPD6N6n
zCd+wshTq5G0`;InY<XH-@sP=}iSnKN&l0&Dt@0f*J`0d#l~49J?C!s#K^&@zO{<Gh
zyGU$=R+YrSfBEA=<kde^-|3s{+gg40G@t(n{_l>pN4_q?^GuK`xB4J_z9aMnc=@{j
zAi#PtiMy)VW51>_qTngsIU~sXJ}G%jT?-Y#SuWstSyC1WKSRB?<GCH+co~@N&;R4{
z3o+|ba<?$nLtqvj7_TvFdkb+%zy7GN<vn-8&9@-)iZ$mZm@_#Dm!N^&3^;weHI-;j
zt5*H_vdg{?F`6>ln8A>@*#Y)%Q-EP>r~XYlOzb6L7*$9nlw2BJ?}6B0alI%A=Z1Vk
z>>0c16Ab5rDIltv%?O`=C395i<6LcavuW@NDiK*m-euO^bITH6)k6j?c%sA>xfw`I
z_&G2T*r4T+*9rmM&a$tql0H$_WaEe@04*GN<Wc|?%CLKX+DBx;ndzw5gu_c`rNxyI
z6J|-qrZ|<vZYY0iBL!eoXC8@m=YW=2S=1f9k%O8wBiAur(2eB)0;1#k@>C!RMO<FD
z)1Si30-;fd69PX*MP&Qjmp)p!kqmIiSmSlJ$T6(DEnCUw{{+U@Hia{Ga<_+@ld7!O
zb~9z$nz8i<1W`9EMw-h3slXdmCJiu~>}0@xHyHYE*dL9J<f`mOd13OGYz22-Kzk8y
zRI5hC<|uyxojgy8ZEZ5k`C{|t`c>i#0s513BZbl5IzL1sy=Z!GKCS4j91nRCPqfbU
zwK|Wx0Vpo7R=uofbWoxnwbpOS@(`Rfe+;S1A_!){UZf1m3)PRr=kNeDrq&rvz3{h%
z`r}!-mw?b-xP-syL?O5Hs=*<j{x6w2G>N1s^4H@ng>KL~X=U^3*SKVbx7ihwMx#M~
zd}S``j7S{Z!!T2B`i`2spV=*uLqD+2RVcNQEfsf7kUeqzOt>CPYb@^CZ`1v1)ClE0
zw<)-b25*#vXAk8l`@<45K?ou+429g8hlB50PM*r^eJC&U?zn1^Gd0J*%X<8MA$%iH
z-co(BqFn68s8O{OjudaZ0e512$}<8J<X9=uax`;eyV5Jip}9Y>x8C5r_Dof%{h=rr
z^QzS4*REQ-r`rk(^8-WKgFHNix?T<#%rv~3mg}r=j#t`}f$VB=(a@zK6O@O<x!Q5Q
zI(PAc%ky17PAl0Zoz5$~kS)d85p$cL^aitC*ockVLh-r`YPR*nxBf`9${5b5K`MDX
z;)ek#=E>cElPu>d4X0Q;sZ+#RY;q0gY|*!zsShXfgTv{C(?9J^P>+B3)gr`@<Mxdj
z<;x!nq^bd?00CN%7LXJE$%?W}N-ZKv{QG-gQZp|(Dwk%NMl6lrvkjoyk-h?`|K!0B
z@UEQtND)v?VkjmtO7ZQ<H^xI|w!X2QA1y*-Oa+lJeK%tkeZedP!|&{Q+nO<l5fZ5N
zO=rV3fJOLPpWkd##BQP1l0FRGCCyv(%OMG_!Z0$<jkcj1A;4BD?X?J5eTrBVnRX<-
z)m;q#s=zu*D@L2+*3ks8z|<TJ{Jc+r6=D3xP6aleQ<B#oB0jY;hI*2}&-%tlzU%+c
zK8s1$d{n7h_jQ6GfCJBQnT*#~^7g4S+qTIVP{Ip=oU$R|<?uoD!wLWnngBu+$2<!&
z5Aa-xZT2|jR9H;T)pt?5M@oy*5mwKYi|)*49N#UempCRTNU35iun@YPEct^rx$>gZ
z{@Um!IQ&{IWFcHNa0UjpT$j~KDrTPv=dtZ0#+T#*vb&Vu)4y5$!v0g#O03?}MpbKE
z(CZw5t3bL*0A|mo7YE0Uzq`ZLg!zcrBu@MMh12?~{>{ovXmg|8c36>QUy0()2#z(@
zh#z5o=-Iy3E$_K&1MsqFBTV1daEopH6*)DlsK-`6cDv4b*gVe|1PMW6?uNq?Zj{dI
zim16l<BWX56=yDq(O@bW_;n;KT2QViJg;Pv;!pvG-cQX@X04Yw%N%owgS=K_@?rZv
zC{j(1+BBIm-NAmk26YjM<?=U$h+}=-PBM}7IE7(QiXRo|(*1)an$f#A($9!zyae-0
z=4XiwKdC-b7}IT|!39Sq8l9jSCbrcVi7`GiBnWJ6r0Q>X%j#=W4w-3kTC4nPza6ZK
zH0pY4=mMC`UJkC#tb@LN9`xLu_M+Ix$X>-p<gWiB|FJ~7JZvN|<wI54IV3&evJVtA
zUvtgHg>P(eE(ZZc*^0M0q0#QR^z*);Rsdk6z<#+#0s<x#(L&5-x#b%3;7?JaS@nf1
zm#=F8FxC`#{Q+a<f)ovAjzQ;{3;uG$DUv>srvsnvYDN)<KY-=7Yh8$>5KID&L)&#Q
zV)_|R7`BXzJ-@uIF!@2lINF233^00+^;3DyOePDH;Jk(D#G*>>r1Lpl>mtI1fcrh@
zp^OziWT-`{e38*fZ2X1VO;Z~CNtgmqaCfY$Qk0&@P5fWUQ8<(*xJk}{%PCvAn;#Ew
z^sfAVZ~IVy-^1;40+<rtlai7ieBqw^+)T45c$qX9BLkNfJzOrBb0q}DSYM*vSPGes
zWeRE&1`2#fcmYI>f)~{s-D3+>XYAawhp;#z?g)Z^_lF$!z}!>F+d0w=6sbM^_@a$b
zig9sZbF3M(jk{;doAn6TA`1QhFYiBq;5>Ebt@N`s8hrhLyu{l-=edw1G0ss-yisEO
zcz@+kmv~#{W;AW4X=X}kicBs346b1l^1M~v-<88~%A;NBW#qJlDQnenr(LTod-uqL
zO&gPyKWiJlO-p)5Q624l?DqTy{=O7#_Xgj7%K~UXzQT{6Z>RP5)DZ4w*3nej58<<C
zhA<Bz%L-U>ZD8}8iwbRK^J@)(cXz7tWWsP|qq(pjm;2)T@sx$j93pOPtAut_%OYla
ziZYX6WgdnYss5u~Ca+`Lg#y3A)UvsR9S1zPB{l=DEb?13<8dD^u`}qXY6AZSA@`h}
zN(NCcbwF?C64X<iwSwuU*Mhr1KrqBg=z)%T!QWM%qvAes`%2OVTz%28rk7OXypsBi
zfJp*$N#vgQneMYNT_QDtT3xff={2-yW<+C9L!cWI8HJuFV(#^+(fju+30an|iNvcm
zRmL4zMar4-P0JrKXG&B9OteBy%$~X<+5xBgCD7|3yQbSZJU7D=;%QoN1_0bj)_P1m
z37P8ZMaesO?S|Kco6B5v)BcaKw~ovD+uDRBB&1uAM!HKnB&9*5O9Z3@Zo0c$TBN&E
zy1ToP2I-dW+5Db4GoNQ>-uIk8Zz;dA_g?E-SLD`u%_iw$4q|S9C^7$l<OJ2M!{<zI
zm`rmK6i2UCD3ZB3nD9J++5w>bM}<DH#W^ZnUX#LCBp9r2V8SB1_zz*AIE#KJlpk+&
zv@*4>wOlG3?3OIEU}iMLmlvB|s2o%F3&NGDLj*BBFE-8W=4t3*@|-1JK<jF(N)T>y
zF|zu6!`Ju;)A1GQUtR$5`)cWybVf#F>~j0sFQE!ry_0@@MuHzh$OUrfiNYOc?Trj%
zA6&c0^?td68IO?t1E4!NP;p$U9Psb(D6|*!cfv&Nb9vciV<Ky~vh$&=i6<=@w(3%g
z&=~j)sWx`d*84dcfaS8(=t!-xua<;#9DwpB`BlK3tlH8}A8eBoh5=UPbN46jd<oJ)
zuJDBA>DOS(5MgiYUv0nTnDEMfv#x_+O-9#i8}Hs`q=O5b)3N}$tDnj_`4WIF=pK~_
zvv)WeI&M;(GVk$W>}LJ<nH$djQlIli-e?AZGZ1~m8rV%eZJXon*#LlN_F#N{XFO}d
zH3gi}GPA+kGeffQ2_JTokFdsZl4b<V%2)92!NjJ)&B@x%wN}YmMX1D;N#NMtI_>2w
zWGY7J?eCgbPqQ&z?*~!DgT667C6C#vbUT~cP3Nmu`PXg9{27gx{;e^fLETzo;E%gj
zg9LS0U*Xt@JxC>gO(L#iV``1h8L$q&oXaZycH6LsDWTBBu?KtL$L*(#A6Ly>wA(Q%
zu~$pDBC{9Eeylv=whjjVhHHZSB7!<iXVRW7geiEf_sxwdTS{UYND)LCO-n*bm((O5
zdy-A<<m(Xf!s}y3mO1@KggE1iw<nV0ZR0+dk{uU}*nAS@IHY4w?SbWg`@Pi)&4`~b
zqib<9%_SmJtA;5GcL+mNws>+c4*6kunA_;+`{&~CBEd1KjvsKC+B%5tI_?wYcazZt
zLCG|b2J_eP_-*nw&wrIl@-LX8UEq=UE~xfAJ(^M;j9;7ath$#klgsW1o*Xeu!K^Ur
z0#nZEMGS%I^*4H;m@x$kB)?;?%4I4T4Qe-YSKtF{`lm1ICmjA`@yZIAn%Hmu`O_E8
zb0`FzYC_rM-HqC`Jof6zxl^@m|C($rLC*UK--vn*yh#%|3An8sJ@37lnvF6xNhy<N
zRR~}^AyzTsPmd2q{bWgS*22|h)5!yYXsHmcOKJ#)@6JzAPiRdNq35P?xKy^8u^Sj^
zO(l^M8kjIb!+)|H4#$-se8lql^^<`#JDOAE`*2a5!`2YK-0T$On^K@lhp8Mcu;-=n
z)bT%!B4OO|A=2!<0Dh=;7^K1oApe(iZ?1wOV-QE1dA!?tdn84L^?UK7<{d4;pzyaN
zD-#_V8CxO$qykwMG~H(ar#58oFS^|2-gM|{RTv)4wf;5O%&G1qJ~R}Ev6GcN;FiWc
zI=qKr9RpH~oY>|itMNGQ2+7`ZL?@&D`lE(@LStsu2m`DeUO0l(_cS*fkwU!kAFo97
z#N-ipp%|f<8@wm>YL=5kItw&$P3kQ}K|-FHt|#R2FBjw?F8EVaS#57js}W-LX7z}H
z+)s`SIFx%wNnbB5T<`S=JpboT<M$lStT4Ztw(DW_-k=&I*Jq6zL(|K<;s?7aaVm7H
zrH1NrkO}IKRQ09%)g6NyaR7_%B?i*fHQSYan%bXEAMUNO50fiS`z5v68YU@?EWc)&
znFzvu6a4(ZenF0bV*d=Gwor6%wo1p*O#x-QNEVD|8h#TE+$cdiY}jVX@vm>w7pmXP
zv__UZ856X+yCr|@ygTemUmHteJ<}MNzZQ?eDkB$M+<a*Aok&ZQ&9mAO)({b<XBzw2
z9YW$l-F<PC#k1rg@+m79o2w19!yaFW0*0&i8isNf1<<16&lqw2*mvhK+;<HF2X{Z!
z*4S0r%G}mkQ1Cim7};%YnbRIdc*hFfunuTeez1ofS*VQq>*%HcdKEIr1(j^|-Us8_
zd-vGbd;VPxc|+(&^zu|bncc`G1;bj;?#QNuFRpkiPj@k`mV+UVDiQj_+{Zt#$4+-Z
z0D)!KlKg(3M#Yc6jyyuje8aEx5kc3kAr6B!Sl}sc;QveCH6p~45$_}#GB$cACA5)-
zuyC(UMAZgM9wgiyGJ<;csH$*z)Vad%?EHjq;&2(Mm(Nr(y$~m~InCyDLaiVW?@#;<
zD81}%HWP{j5Cp;Krf|0p^uHOBmkocsb=OWxwT=<|yb@a9H!e%Nc%tb;m6<O`z2p4z
zvdQF>H1s(Do|NI7n*XNn+aL)!zQ(HW;_hUgQB0GxRvjc`YM7dnQw2xFIZordae#@`
ze%7|nONS-X8%24$A|yOD)Ca^ieTA*x@U?j$ReJOohnUYH>_BK>U6eVw5`^;dwGy@7
zGbNMY|B$8Homf@<O=kYxi*1K2#4jmv27Zr!yxN!>PAqq4Yl*eIY+ZXpAzdb15h1A7
z$ol$3-{n%{SP*XDuXj?K`|N%l$PPpxe~;i|gwgkaErL=44{vp;>=|iZHRo3@ifhuB
zt|BpiuW1^wfOZ6dZ1@i{i^&q+Vj}$U2vu#?EInU7sI4JbaoPy4Y0rFu<aF61_U0P&
z5jq@hx1Wv`C{y3PwSOI6NjhhXEwiqVDc?ru4|^|=w$qPMvUm7~_V@M!Fd3VRbP))%
z6-pf?%u>pd@JnN?ilCI8-SFsEp-|9^pw_CVSZl>~6{y?TJa~M#%lQG_O|(rAGzuzd
z-Hu>l#>Hxavkyo#QpGAo@o(ck)n0)0z#}KBK}sCo4H)t3cB!$$+e|1oeb6XUhT#l`
zXK-v5kM}noMrgI`04eCFMN1ZiM<v{ewUqI?t5AJ7MczMV$9IwU%BpiuCxdqRBBJIC
zVOaYN{fG(KJSjBX-&j9Xm+%!E3UJm&3_#hU2|1iuV5PU$?r##_?p>0kUf|4)(J96L
zGW|97oy`lQrxi?cu4AOk1;=a~G(oCiHTsJCIm8_Dbxu`~k0=vzY+$&_+SQ}vKabQ;
zO88F4U#>W}!6Rh*s{;A<pnbNXu`^o{Z)AZqCtFuH-zXZ_Hiaiop6x5F6JIa#qxy;7
zNV1|7WhdDLVr<XjPLDqXKg6w*c>^*Hjz2^=n}<^`vIh(er^tO_%MGLEa0{BE;dc9A
z(HHgD<!e`?=g8|1huq9Nb_0J*I`z<T-oO476U%yNI^br3YQ*TOW$!S0=fJ-C)R0iT
zn?ql7`IYar^2dHVYtdcun-#0LTw8L`^vjY)BfyHx6<j>hMc4Uj={WHPF5sROH?hD@
z@ZC>a?rIP>P5LsugnQ>SdpBng8F!spvfXe#fFyv|5i2|jmJy!7UmFvwu?26u+#o`u
zK;b@53oqq7RkgxM!Q1>%l1|7(D!Xxnoy=fY9<cg}tkSZo%e|qu6ua6emq>V-oPVNE
zvzYnpTua_5peDo3sME%Tu^ta__Jy32^i<f@pZvfc%rt&y1iLi*5%$u09^UO><&jwQ
zF?Mjp$SBkJPhyJB?UJEsFv|U5CG^q!vu1>fZr))!na7Xn^MWZ)e8z&6UL$`%kR9Nv
zlPM{42-g3JmdQ3s<TS@!yWLRW4*6i=oUwqxcnV3sRMw7^3@0$6_nGtrOGzMNAB@v@
z<Dd{hS95;S`RG38lSJo7QR%dk5op+-yhZITd{~PPRsWl$Y2M00B^J?X4f+PK<S_|o
z<|My-A#c~8)#(#!K|fyL8midz#v=y<jbGvMH%~6f3g>>;tG;N`zkIKcfb?v_H1(}~
zBAYmG(uQr^OogBSL$h9cbK$Usrs`|3jx+?<e&%*B>m9A`4G@ujUxY_~O}a!#;&$TA
zu{0^h!4fN;!fT%`-}pmqzX45Y4<zOq&O0XuxN8+rzbV<ajAWlRm1s)8x_HDxX*+2V
z$F0ltZdC2+?VVh5TwzSP2L-N4%UKi?sZ>|Z<erM?z8hVCK+F#>8EEt;lYko2k}dL7
zSr*k2rj(M^G9m2-QYz#|q!Rq>!NOn>n!;omiZ#V_fykxUAE>iR6IBU)t2wWiNtY=o
z1>Xfne{B)z2{oJB+*$ZMWyDAWu!WF)NBZ|(uWfZ4Lz)YOC<TPlNVOgCwp>tXZl)dk
zzvmz39x$QJ+bPEtmO9Jevdtz4JsBfWe@3NBr}iGYVW)ixv<j`{;A&1D5Mmeo{YGhA
z_B{d1hY>Al`Q6r9l~En#ZG?7D8$?X>IpdKeC340^$<^DGv;M$Xhkd!&nOb@yW)ENF
zb(aK|TMTKP2Tqm66D-QsD-4w%S5%bpp23)DDN>ug3S>da8}O0obHrRq=PI*`Q#o?e
zXX#ww_|i!_c3SmFR?7~!C-Ijxs~4<LREoE->T?>mL%tOf%OmaS#MkyUCgUPkadj?S
zWp@J{Y<}~dz+XT2GM&@r&j%5OXiYdH0u5Ns`MCGN%r85+gquaJqAHd*ksLf+j;$i7
zq}xkL%7n^cMReOj_e4KXhL&j6^44ecZN=n-6AS!Y72bIbA&o}FU-jer(uL4FKW)ve
z+;ZCUX2SFclk88sIIND>0Q6)B>*^F77ompm=grX2BR@o}d)%b{flkwISWs4}@vpP^
znYK9S_!x){au!;xw0z16cp}jz4AC;nx;9}z5+HnMbkiUWaIPFXHZg>jWrkH>-Vt}m
z)zm8l>Q2rvbXXlUJ~(aJ>QaG-v)_?f)_uSbXw7y5^RDUqbpZYA2IwgNVV#B9QbLIQ
zcb*)59pGl~l}nL0<MArWzvw|$75<K6YI0f>2t5jjd+ouPB44-J<x-r1z~kxRN@aT@
z6P6i{w7gR^kKu%Tb4JT5PHvq;2+;(rw~u!&$ZJhuuH->Z)g4I>Gx6|?W#c&CfkW&N
zdxX?fhA4a_K}bL7vrHBN$4)ub#H~k3RyguZ%G&Y~Eo0Cz@vf-^!;kqNmLv_1ZB|tT
zv@XB0DWVieZQrtex%MK=JhJFy)5#`IuTE}~H_|&r=&A&CRd3anvetOG6qyr2_$Kob
z@*MSw>#@KthXQ7`ZtTEFNzMJL-$Jnl`+`aElL^9eraPu-YDC+-x(rPSUb$xYIRCKT
z=ncekT}ITaKZe@;?|5OOd;PeRl!H*DGvx3rM^+t|`MR2*$o^n13VQ96&xK8_0CW%b
zLt~1)Bx{2>{SCwV&B|U!+{D|iO2p31OlRkS-vy@>M;OqHYW=U%(gk9YXe&}0K6b0<
zDNWeg`i&L&F1pW<p^4W2#a_BJ3u$yFjq^M=ExaCwlXl*J%D=A*j=k<;|D=D*Ur6fL
zD20`$biO$vpNKk-6E4ANB1Vgpp!A9T#x*e$Uw{bxm=>%I@qnGz3JHqa>|AE#DMI@x
z{;}0&^>zf*Y*1E+e=Hw0aSVUQvx8ZAw63;r!0Y2nwS~Xlk7Yr|R(U7o8QzmN9#lW*
z)N;CL*;8!#bsteTE$4S)i6ao}wYZdid!@#*-dK~vD6JRI(riu+f70b>3fi?3XkS`P
zVxfvX+qplIOLZA}v$SPq>JaY>-}lo&f;Sus02~0k77?D2iE6bsFqZU0+-v1uwlZ(t
zxU{!uN^z&mTd~nR4tdb%d{zbl%=(-fDJ@=3MkWg|w3PN|N(unRRAtzYDR&sBKu^={
ze<Ju1oNO><ty`a6W|*<hCidbcXDe_24v);*pW3uje@6DeSWI|<+Ek+1ugBE6`*p3u
z|A@UJ2}Y;OtUg$z+=wNJ3n|)bOlDB&0N@%E*fi3WA7e(6d9r@jm?Nei0jKnizzhg7
zOP52G`-e_BeApIh`Q(#2GsuOH)?3NXh}uBwk$pi%2u<OZtqrn4)%?9S{(?h-^JVth
z5iTuO`wMhG81$WluO`*;zAPJ_Pj_jv;_a-jsKp(*LUVK!%IMd>c!Nbgxe*zPHxMn{
ziL4+7QzL%r^&(1(g;@TZ(f5Zyb)<;8%DFmgifZ%Ok(?8innf^7<Q(mHdJWm^%BxPx
zltE;K=hVReW(tuGFu4GxNxvwio;tm!l%b-)kRM)I0fuWQhDvTCnTWjCbTSvwmTq*h
z;QKh>2aBR3m?D-yn0+jpc+RQQt5;IN3^GTwURA1`P`hxpSR;!w^ykt3jmlOB5|@4N
zcSJzCX?FrxBSQG%;!6UQzL`Tao+iciLGoU`{5t6dx<=~Bsu5}@Vnm5mqoR0a8bZM}
zlf7^QI)bLvMT$`K(t3$o<UD)+HU3A?3Jz4-B-Q}KSThh6xFA8=Lae}~Ml69<U6=K0
zE+AEx!a^m01#hq!=u>On?t1<SHHVO90&r!*9eG~_22ri!OUd6RU_tZq_|@VPo=6ZK
z()o}zup0$PUf))2PR6~7Nyg@j&`PrugKy5UTb{94voAD6O-pQ^oecfb7{+2?m*3Qd
zhB7sLI3vcq3U*5HgUea&0?=7{NY*&gci=CLuc}Hdk3<ZGupFjRB!%=DL6OUOT&;?%
zZiP3X0+fm&+dOJfkSkGaTMI6=Uv<e}EvI6m^|WL+B_q#%JFCUBCUY7zUgR3Ct*e`M
zqDpI!6B2DVy$|j{th&cirm_Bu^yS`OTBGSFo%@Y|%5q&y^meOUu!4Z(DNJBgfs33*
zjbKh9bBo`{Ww*3wzfLOH@D{%cP5Wv0%Jl(en&0rO5WAvxKZ}O3GI>#{R+*s&Td809
zCq5Y&qOT8R*IRv;;j;Er+dKY?303y-`>|E2AT!BQDNJ{G-Gb>&m@zT+{;sn8?ppMV
z*`=`E7@CI+zJ8ZW<hV4B6uKZ=E7{)t4_iKjOcyq4BQzW#;}G-FmEDK4lE#ZuJJ93(
z5T?NZMdcMhlF@SLY&0Gz(ywZ^4hipdkeX{%^A+D;`CKos|IB;olQ9GEhp8(^l_J#~
zzT>8t6G7%h7}SpCmOMw5ZD9l_vZ<{xs`Oefynt`dWvM#&7r1Ujb*j{EL}VY5y5>4t
zc>h(S)i&?Kr0j24on+F31?XoRFd6acuQ%&}r=!YgmMrO7onYN<TZZJm2uPO4lCzFA
z(9PgIl71ans!2BG)i<spT<mADB@Mf2=A61E0Cli2H*4f=fZ*fe*3=VdqGo-0py6ZA
zCT6B7f{x%?22vf==KbzD+*smYUY>#&r02<voa^O*k|&6r%^5{ithDFOeczOc14$E6
z+fH3P&!8LyVk|_FNUN3`3HrfIyKnmt)Pp=H#i*AJ!(hBxv~UPFj<vW@J+fQbBtwtl
zBOioY<(ERkUc)IydIXI&4GDl}FZm(p&Rg$7+E&dhzae@af<uoW@1gkxvDaS2D-1Ud
z#CY2$|KW&m*OPg%!ToE-I^3LVFLDFMvem@vm*o;~hi@hWY3hv4L@&x!jDldnyD95!
zWcVoNE)Q-!GlAFdv!A&wgCXsvyy$JGDAt7xV{Im&MI#Qbs9+R&yf(NX?u$9{a}#KF
z+3O3EAV>|M{-Sq~1<>D14jFqvgeixk#ohDJSk)%Y?HCZpqjAXg_&kM6)*{mej7kLo
z!6)aJ^ZHw$BQU>=tuULWtJa5K%+Yx{qR+jOb&5wo$twZSE(FVs%77@%EBZP;F;TVy
zv{cv#G=7S7eP4QNb1hx43Dj(MKC~7kRL)(U5FQ*oF892!>^FIF4oEu-(b|t_Z(SL;
zjiz3b{{>F|zf0=g`1dyYfs_&1U#GY(3D&c9xW`zat{0?5I%o5;Wm<G1qqE1ULlfEH
zW10C~YaPjI*oUq*4{P*$x!HBv^~)kj{j2jn-v9yyfs0tXcTG;7XTw}MlZL_Q(dNNZ
z6;Bz~&_C+;523!;a+G&Ci_17gZ0lRSDe4;-GL>yh^{Tgoxf{>a7p?Eg)Nh6#2u~5t
zC8!Rm<b36*6Qr=j(<0&*_m<FXP~e(7SoKbPu<;Gcbn}mBF|J+OP&FmHqP8Y@k*jm1
z4)1qcd-B+``?w*;Tz=~=R~Q>9)0T7LI3E{rloNI<bX=9Y(PoN9+N>WeDG^*zHXm)j
z!?L%dXG*dJTe^lrV2Fh`d-&fM9G7KDlUQZ`jpP@H@Lf`VPjiAqF`3=m{oqK!O7zNu
z#XjXvV%28=gJvAYkB^CXecE~+sPsQ+=U`N&I!&1gl;Jb45w8Sgl{!!s1!Jpz#XkOd
z{jtk+9)d*47do=Smu#cTL>a@qP04)s<BQb}y*n{-Z9_eM8QHstAvva)k@!}Md-1mu
zA=dx&tE)oL==Kug;=2Pd0-QE}#84%Ui!Jfz*QXnnE92KcxtO={q1=x57VCc*v%9k5
z*-^DzzB;hvy#DHUY|SDq9GUIJ^(uytktn3`Uc{QWE3MLst|fkY`pQM`4Eax^hu$Ez
z^x!VzsuBFM;_n7;=1`uKqA_C>^ju->Y)U_G78K|lU~c5?a&=Vl3f+1=V1eH9^28TG
zebBi85F-k?dzKKbPgdIt)8z(t(K3k<*h#N_oHF10!udriTzes$Zp@TusjQ(jbWFVp
zDkjC8z~nY%#0%L)cYcrHh2|UqWer$y?kX5ae#zIX0dIuAQ3NGI+Pf<wNJdVCSLKTj
z>zIhIl3!?Qy@z;5PF21k4q7EO>*7cM{JF6`5%0(i^3DH_2DS)2$u#)4A^|tuS{=%j
zaC0?uFQ3OH_yJ9wir=H#gC{u&y3$#ak*fZN5rZ3F@ll9$U3Mu|I1n(5KpROTQTOnf
zpoVC%67uM?^SU?BARXi7NL0kO5&{w_hp_J2YZ@+C8uYw=P1bgM0+7_y&Qk;&3CxPb
zSZ;x@_4HYF@tYe(UbVH?DWLPG!;P_^u%kz9?*}G?DhX^0);d`dD7P})+#7X|U`ZiB
zYpF$4KvE|$OXp@n%?XLitG0YUzht`MR$UzmhI@&Kh*)?{ct=3c--bhg+Fz;KP{%4S
zB^9Mh4N+JPR!%9}AkZmWH6{Uy*Tmm-?SjMa!d&qX&Z@eCxFyraQT_T>BS;!{dRSi`
z<x@03B}`A(plb6~qyaZx3Km$inc{+w(p5o5B!s^5l8TitZ`SWa+=;f;lIhEj-w(mE
zPE;6-0J{@rj|g&G!X$J_z``0(ElLcK_y`&54rM86djO@6Ts0M6&IaM`2q$j11Gh9L
z#9mON{4>;=F$uRGh)&vG;#^?f-(?wzaboJIhdVlr@a{q6e(Jcv@)28P7ycf)-+EZT
z2ewu+4%O(Fx{Dk2$ag;&#?X=Pq7R%AW(lV}$I{=MQOIfAT1RvuiF3EC6L`U4b2Dd>
zr5Cru4}{zi^tM^6$HO3x&%GGcscaK?QF`X&(Fkgo>HV_9mn5mpZD=TskcXvOt2AZY
zI!Qk-k`CG4wn|X=Impf!5oXGANQf*E!i+}M9S3|SwtR5PQ!LXJsJ3)E^|x6Id?2`c
zJ5pcSW_tqy_Gps+#0K1~#**$E!H=qfSyYqD3L+E@!5zR>WSTuT)Vm(T_jyvijR-C1
zMFkY|6cHC4=UeN#AE6VR5$;d#y#t!_sjt_0NKo(~_hyKR6=mzrT}g0cbph))*xP6&
z#)^x2r8<o@T>^Lm0s9O`zaV7+7af#(^ka%I<;h)IgQhh}@)C%k#~I7Ao2-ZkMnfKc
ze-agPRG0kum26o~rzPH|<uNR5a{!5KciOVMv4$TqqVW2iw0+C4+ZoQlQj>Qvb;=6a
zYs?s$27A5`oQP&t9hbfJI&*r_{HMTQY{}lI^uuj6VI=%MmCLr4uo-y^5Yw69SsY&2
zO57H)o5m0#@r{5Lan_Vlv=nHwI5Dc3YFM?mp*`rqlUoz@YwrCvf66kWw~wz=+4Nkz
z2g2PeX&9&AW1F;jg)WKhS6H|#<fYDV5xK6~IE4|#iIo^jvwfEMl0&|FHgU`Sf&Hfh
zf<<q$g<4*4p2Ui5G}~w<S!P|;fsx!nZ>tp%fYF3$67o$Apm2g6^CCJXyFfml3%s%#
zGrxJm!U6JOu-b<Zg%o75A9Re%6K3!DfKOrct%bOPQuT)8&}Vj}>Wd`t4I2T`T%B=t
z&k_1AgdikC^=@FXf@3Y3j}X@I;722eK|&_5MJm`to9PN!nz7Quv(73CLW=4&Y7Y}Z
zpk*8c?0_{D!3g$HwDk$M?m+Ec4V{Qkv@anLwCF7(Jx7HAe^Ox#JQGvh4%U7n9u_6H
z>jb8xA3t^+G=@mU3@0`#6s`#X5B_~d**xv~dnrwcQ>f`8et%LqC_IgudpWGk7vgSh
z7j^s{0;-y*WanqxXeIdmPU1gnuf2+@KLp!ZcV?~DUBpTO%s`r<>?+8FA^o$K>qp{b
z8UMad7+jc^+wjz6Q%YhKaJx<GqD_N^tNf(&G3Zxy+9NT+B5w4<T6LMOl}_XQA}krJ
z_f+(GwWcC0MF{%-9L(eC873(?0xSkS$l^id*_(fBA|lR^pt97aF+_wS2g$lZUyX!D
z?AsCIF=JqpF@5w?m*g#jd^*by6X{x;ZVo`-uNzY1|3o)OO*mN8Tjvy&<Mc}b@qJVk
zp-Ay|C2FcHb|!FvnLCzd?};FyXACrPTUTT45{-tIPyNP~tk>;Z<8z1Wc!uPiSLu3X
zDSl2vmZF9b;YN1_;ua}}M^%Z(dFPAV#6>w0Zpicq`IZA_Dj3C7mcQmK+-qL7<x`9t
zF8i6>uf``?B&|LH_WZuwrN?k+<(gm{U~VFZ_J~*xRvYx9Qu6Zt&ayWU1eBv#xuBOg
zYvr`ff!*wB=*SHW3z_1f+57FMf2>Rdeyu%bafosR#my-7;LW}!3$x>EMc1V3P|;50
zb1b#fUAsM~2@yfxtzrPdIEZ9DDqTp%d7DP|f8U{8^^24rNwJ1(VW4nfkB_mEZepN#
zV3XM~YpR)|nk<bf0I(AyoZqg1Il^)L$NaY0T9ayyG>*=|dR<wN1aruB(2niA(|fAb
z5aGx6hXEnADdtm|w4350aS=<+b|Fvt8_x%gd%oL*FV*TIrO8Ga!ZY{$lopH4l$UwR
zTn-;!MG(;<O>gtU1qmNrEyTu@5YSMU)u}(^EV&nb!MXaphQ2!*!Aw_Ma-YXrjU1f0
zO>!&xUgO(l<-@3MBk$XRZ7DLguL>%Io(+zY-8Pq1)S+UF{5)Fkcr8TjwQ4N$)w{1~
zQ2$x3Z?Vw7DmDoDZX}Ta$|8ER&|XUl0FiMt;~0E%+OG(Euk>8PHnd7iR*!L%cHa1{
zMUl%=Ve21dkfI3;U|KpOOv0DdIkWxBTTyM&_rE$Nw1IZ<%tIHTA5iNmz>pGN&Wq>f
z73CbOrcPE_Ezgup2SP>LwbC8aHkDVXW{}bM`)tk7BVTvea4prMus`2#Cg>HS>ORgO
z$%{ty^y*Q0YZB}D0w0OmLeaR?ePOOY5Q|XQKvPINPn)tN(-5;QgJ^@Uw{2D)>hgMF
zZbNLtujN2xx<$3MPZ19yxR4H{t2Q2|v#sI6!E_;ej7+F$4dl{_3|tVAq)aq=&Xi;_
zB_dL_p<?g<A%x$e8p)U=aWM1Y)%Z_aF!eTboPRvZgVbO_L+%cAP#gJ5l@#hY_Lr8}
z2q(v*aZp<by{WxaV?Yl=IbPR&H~{hR_f2bzkD`zWQzZJi2H89I*jJ}Hhh3u^i)-Y9
zQd3sDBgkEwTceSB(Tvk3H!}%w6g=f1QP>ilhq&W0zu-??I^?gTb^+uH#E7XQX!&WU
z_`h#gHTMQEu$<dyaj(ehYF1y1b7@9Eaj9~9#$U9Z#pd<W5h|>@9ne9TFv8J4^&*#?
zy&g&7W`?6%4G3|Bty7z3hqS^<s-6?B%dP67m!iYKqn>|dtvwS2A)I>;V(E)vOSr-Y
za7`rbd^l>e2;t%~Z+*F3U`_+yjF>|p-gjlG-tICS-!Jp2;RR*~o#Tk0w{8;@*R!{2
zOP>^_>mGFJgw{MiS)OO<zij)ILy%}b%lJ5*uaisYC~9he52!!q_iob8Rw$J6iBeLN
zdYQ4^MZrW)qNeozY8u7Hql&l62I97z9d|qiIdrky_Dt>K?+8WeK5*kH8>_wD#&L@O
zQmPyo{WVlqBQr~e;XCW5)WE`xXmFJxVgHjewNdL;yk=F`Q}c%&!($h;SKYifA9M2u
ze-7ed9gV(hwN^W^;Z6q3IpJ({VP=R6ybJ#y1V-ql6j5{69TkEm3e+xFiF1+Q<BW)A
z?5stIPq$r@v1laZS?Bhfond)5%h7}kSIr)^GL_+?(2-`FB23JVnb-w+hHuVaEs4gZ
zYKHwsbt;FbP+2+F3AAae$HSB*+xTX-kauMG&ZG9eXjWsB|3=a8a2h76M_Aa>QGqK$
z#98vPhTwDKl$>UUuAkTWT&5FmBmI~oHaq=aW-5#m3|eDG!he6!fGvJiHaEMmvGJgZ
zCHD8)2`vNR;Ff~+B+7ZcE0`GZ7LAyHM|EtI`OP;j1ls$<0m5}W1WmH@a?YQ^js%r$
zy$&TuW=3a$W&{mbfsm!yQh_$vx>touuz_YoWP842I6iEiZ!%VSL)>A*HJgT!W(AY=
zPFU6zxVq2<GwBTud^-x%x+9KaqFN0Q@-#0Rj(;${0f!dv5~LHQQcB)*$7WtfIM%1s
zq-|gY{tJR5HDG``h@xb?lHA3}*QzaTKy#+(i;Y9QiZEF`2M~CaoMPlz;DNfZ;~x0d
zSfdTL5+FYPv-f^n3yN7aAOKZ!?U+-Ne&xdl&&@cse8BA|R_Ts83zo!p=2mWYx!Aco
zZl&fpDc9K-(5~wGW|Hsajx)-U+Vg|eV9K@@l5uH=k$2;#MV*m>QHQ9uo4KpGR3<Ls
zo0Tbj%l)E)9X0~pSHx-8{Bz^lGBLB3c;qXycj!Dbg(`!zGIlvMuKN2NStN*c@78%>
zu5gx@WBOEqoLA4jwE<+Y8O1E+t~qJee8WdnvuJ%IHUsny7G9hn61cS{aY5vc6(;o<
zXmun&{?7$%0?<hmUFB)^mU1Po3nE>eWlF-ert$4h6^Md>$!rk0d%e^*n#6n3WE3O3
zh~iFrdfNZ#=uwj95%;th7fbowdq{~r*=H64+CqJUkbP(NZiii|o~}L@v^=D@BV82G
zCVp{`=iDDhl9`hA5cIJ8X%>HA3W~^4@+}mU!o{>K{p2la3-!z2DnVFRv&PbD`s6{J
zBV@yjLn!zu8P`WN@yxie<LHNFGwt5rcrx{l30lmN3ZOU0p&xuQs(o&Y{|l0CTd^Wy
zL(NPO*I>(g86`|w^J}L*T8vrweKAfM0c=&v-JJ~l@s?QR&|c#gjw-z7*yx>c^(w8<
zTp*w;yk!_q%Do-NUI|%BT)Wmvg{Tv+kho8m)EAN<Iq8g#dlIy??%*@5PRD%R*~W>7
z*8PmLgjZN<tLOPpy%)4mkVGxGwYD(^2t%LN{8||UAZy-5#9cYu87^mQjbJwMjCk{U
z+}G`?QZ^u<JP7neVrF`v6B6;*_=$0C!5Wi0dSel4E(?e4l$=F1HaRyK!$!x%s03C4
zP{$WS;g8n04BM4`?S!(nxZ3VZ`|-+l+`iaFz9YP}h*wi=@D2dMlW+V4hZoY{!>BMI
zT$RN$3V3?Is=$LyO?qdM2~>a5&Gj|RFDy%~AMKIHtNrO1AhKw^*8D*t$RTjV=oW4E
zlRCoX(D0f;32ih}FeP->+9G{O8U8zhokirLU4iB#ut9aGw*;Ak7SHif)3*Gj!>Mq@
zC!`Kyo#;~;z$KCrx0^MelB3HWP}lM?<-^g-MduFb#JJc?sqo}P>sM9XR@<O{3V8~?
zccaJphRe9h^1^xoSa4*uObPqG4zrCFgi5A>+^@duQc=^{w1yf(hr41!Z(O^!#vT*7
ze~V?f(Ew#^Z$%*yBv?g!>29G>Vy~U?4Lp)%k-CJA8ymnv`6FMvw|gVzYO$=)-V#{3
zwg`KgB^S0UC13aWelY+}I^wR1dN8XU+0PoBuxERkehF@n+in-@R0p@*Nld8B_UxUO
z!I(_(bAFRQ(5yQ^O1C~;fz;0IBI=D!FgS2@SUu+olEUHnOGE3*gv;K|2{kphw%<b}
zdimfM4L%E<QJ);GfdJJk5NT6%X9}hrI)c=5Wk^@NFmT9<qeR@yi#CIdmocI?ujTQd
z6WDR+Rr5#ik6s1eksm0StX2U8Fft`$X)|=UF47gasZI6BUmwdw%BZ(QpdW=W@lBtR
z6JXp?I7oUB0`V(QbhW{D#0g(EmMK3vYK4U8Sq@#7W0}#-CyH+}6R-rvyF!Mlm-3S#
zdbMbj5Bw>up4=MpgtZE&iEBQskLmYBD@B{$zHS<5wpcffojx$FarPYAp5T{~C%<SB
z-y?EzOo)ogaI8sABY}mkNKs0jxYj+AtRCzFVY^poCIVzr0TQs9Y0n(U7Dwwo#6^%D
zL)^+2zfG!u;^)OBGcy?Mc45S%Vg<$QlK;p2%lv#eawd+w{)(i&uCd)AVg>;$20(wy
z_C5uas*=TfQIH7kbsQpRY9xv3krpCjyz#SwycXes!~~NFhKZi&-?bC5D01<6=c|$a
z?RqCl&+GQRtJm?qw;z7t6_f6cCf*lxh3<K88j5XNYOgnD2;a+xzunbYEVTCzKL)xl
z$EB;Z1f0k}06z6pk`gzS4_X&2grXBo<b~UJ=HWI|;!Hw9WArZ@f%@EA>$xc8(u9uO
zLLy;3B>6IyMEu%qDd=~69;<<8m_PJgMV#ocmg0set9r{Lg%6<HUlC}0Ztdp8=JH2p
znIoA*KXs)z`kT9DVqtFZZ9z{7SG%?%EpsjKQ1hsazK(0Nqlr79{LtD$(wvK>oM;v_
z^k>pR=B-CnRU9e|k+``vXS*P<aEj^YVZFiYgJx42!&YKqR6?kQ7Ja_jb)icPg!b|k
zktxq%&e{AsCb4Ql*%uv>$;S}E_6MvHvYIo*5ytiws}4@6aEdXSm$R^oqpN{-$&uE_
zUdH%!{QEcjz6pYmiEdWICm1wBnd|p%5j`J|{KlFRoF!je@z&TT9^QsNV2S8`vpPrs
z1G^mg4k^uJJjLa>smA)XMK3swxwO;h*Y+I`%pTk%8!gcGon<<W4tt|X0i4Tmxe&`m
zy|p=!#BaU0rT8?VhdNM?S>B&WbxUtQe$`8u2euO>y3IdV+rr$DxB1F@>@j>a`S%T)
zsyb!_NJiLeT$##ydFD=2ks|QCm+_>FmYay5YyanMz6F!!`hiQjQI6vWQ=i(epxxi0
zn0X_ooQ>&>3GdFS=&mjFf-l~fPi0O`>&1OQ@{c-*liBS!m!b^$Qmh+8*VGa#8z*zz
zQXmG!ApzLqd}ld$ed-0Ta!ToOoy~`cPy01OSJH>-NBO2F7qjsxThF|ed8z+dXXO4b
zs_>m8^BvRMTZd#n9A7{@QZf2~p~3t@nsGh%Xtl{WIRTGN_8F`^Ehn_ocBEBgAgaH^
zC|VBO+Yj_U?F2R6wE@_v*mwf3JI+N$+Zr9tzYng-eg;g{XWO-HM2~D1E1^~Y`AiIo
zRTTCkZS8+u^S@U5KR4k2{NRTTMJc3!bF7Jaq&8?&sFM(#P)sGN?e-xK#KI-lSFme5
z395ckRrZyLtvuMkG|ga(et5)eIO$-g3^6EF2?UytB-!c0gb!$2A8xM$X&fIkLo8RK
zqPJQwhBE!S8f<GVGtUE$?3hBiztVa93LWEvhyRDkgo$|naJ5o6AHHla%AB}UA4*||
z(B_8HEWMBaJiR}M3}R?Q@9^C9Q>P(Yk_}w{?u{fVYuuR(z!qrOb5Cwh?$9}|uDM94
ze%fKRIo}y?P`&d-6NK!AtIy&<zLJTyy=J3a^ZDo2|35GMzkLNZTzJdewx4R5(iS_;
z)6zjb#1TqSO6a(2oYZCJ)pXDsme>J>#P^2p@NmX${xI`8-o2o^R#4R8APa(tSAHrq
zN0U@?BrQ+eH@71LWA~XS^_O3old1o?&^g$#LHRFVGLO22nf(bf#7WTwGEW^)!!6uE
zmlvuvy6!qUMx!!M`%>GnH~bA=`78L$1g@fKg7Uusu01>L8ppzsHiSmLFOZxDdmOej
z*`7D;e|JT~F~X0fX*73Bv!S`v{45cf1e~-4Xn9cLZtAK`hTJ$jQ{s6Q0#><Q8f6&W
zJ-Bg8F1K$}8g0=*MqWJ)V?NBy&E8&I@lZN0#V1wE;*a4pR1%tm`@yX_yFm2={&)~J
z^l%c_Y%?7mllGW^!pN8qG^1wK2{TOpL{kc-qR;nF)n+OUzysK){Hrb{!q5*^-;W<Z
zLc>v@L+gqwL8~r9M(e%~|7=jx2a=D5?SA|KDk$8OfGvl$kAR@_@2@ZY69P>KBZvE6
zgN~Y-6W?!kji}gMe=a|2IFhQ8+pcuGh|GJ>_K@YeOG@ij`Gie}5G7-iedT>>BZ3Vy
z42to^59oSpfAO+|=!fWmG8{}IomyOG`Ms;i&ynEg1>*gIY^>5qW~nGTu^&auE^yB{
z0MN!^U$2J>0oKYv#~VKj4&Z}C!1_f@cy<VS*=yZ?ph+r*3L6J!GxkWK<5|Z}Pd}K(
z;55%d)&{-IGOwsD`yX7E7b2`rn?GJ&^u-(or@5#!U0m9|51lM>SA#T#!v2lz#Pn;-
z0jWt*#_`2+R!U}YD=mKG;mHLR{aXH^uzmbSK7K<obW+B-PrI?SLE?aU=GWpd&*yxm
zrs%`Z4}nBZl<M5U&khAZIi+uG34EY2Lc$7993sLF^Hq?7P>(^%dP#rsk&jouA$gEt
zkxG#S5G&<rR+}k-z^Vc;HrQk4<ctgJ`J$0J|1B5jrN?4vm3M}@R_vmF3C0wDD#%oj
zaHubbDI-D-Y*!x-l?HG-(Z`23!+6bU2TP5!t+GzL<)oicLqFP({Zn=SpS#h2Ta^EN
zu%rIoXpgm7tf>@hwa(7oM<dp>MmSK<VS0+PIUqAB$~vx2mKd$8DK>xHBA<w~jj_31
z&^`2Y%E!xBPEEOZk-iuG4PC@*0(r6<B~&$`aSbVCpRdlOsTlP&{a`nSrpVvQf#CgN
z&@unP*a35>qc|V*2#P8M@O%|<sELI<v}=7%9cEB+nDE>E9d{>k>_950&fm~%*)Lzd
zd>Sv2i#5N~0|0!Xa$ba1uS(I501OJ&i4YT0sb4zRShCq{X~`Yk*d1)(J8y9R#(@_@
zwegAm04NPb08i{vUrP6FG6J^DLubL1w(5WMim*VnBB)P}PWl`7;%#DZ*}WA9o{i&9
zb*So`uuVNI)~nUlTnK7q*9*rPe<+Lk7q8j0*p%c^jats3b5NRtN!yr)gDkb0JpZ^g
z6mqB$RqnHO4W^{v9Ki`1*5c!ZfvV6Q0c2kIGPT~@xU=0cCkP{q*fD;_9|q_}YQqd+
zhh)SU+4tC3F)&bY+rg#@qLh-7%6Oi$<tn7f15?wCv+dEzQ~_6$i@j+M^BIPQXP*!t
zK`zYaqN5wIkXma)TC)@p5h)09R4TXG98dy0%cc17B>Yvj!3?=oAZ?-MX57bju@V&+
z>~Y?e^m^HJTk$9P-|cz-)U)XCV40e@<BIc;Qfh|636wh?BK_p%b#SI!j*9JUxOq-~
ztF>vjLa8_bLe(PH==b&!qIsK>7db+9)G3YI?*>+Cdp)AOP?8E&K}M5R`zRl+=W(f5
z3B5MkJB0TE46#E<m4@;L&(1Mo*idySdpPI<V4f{qcEC+Sgt$|mqE!6O;`fd^?}#)S
zG&=Ef0*Q+uj;eF|cw4!Cr%AidOW3^0$vlNg&@(=(n)86f2aC|DlV=sX*w~K`O+cgX
zuTP^Buo5(iag+AlVAaEO7r}Ets_^1Ec!!Vo<G)Jwe{&}uG5VgV#2$b3^{K|kKRwQC
z7zTx0GEbswX!3Z<YeqwGPBPgb8{zw0ejn95#ZMh0s6^c=OgNr;k9QW!1?@bBXa=4(
zjZK#>WW$;>6qFn^teQS@Xsf~_cc{qWN?S%SBu@VGU%(K5(ZR(nyNB`Yc|=J~y&r|M
zox$$Sq|<;-m~eqm*X*z}_CB(h8!ZK~>HhCvLkOBZQ>x{>j&(N<htUA`?`Bsnpw-qp
z2h8*_9mgf{@Hbpk2viibOhEXqQ`CvtbY#8kdfa?u{1*}dhO$a=bSnJGNybRV(6@_H
z{_Idpe9s5$|LkM_zi-L_I#dw^ZrI6EL@6BW$o<EV{l!C_R;)8R9MB(So1+**m6eE~
zA5kryq35{WqI8EIe*GX{A(z?2_-1$Uq{Z&YJ1X|7vsQf91+D9{A@M^{3x>1SJH_S5
zg$rGA#lM3CA9r??=nq${aC9iOc6&(}t1OcITb`EJ8~9(ooY;j+X&j9!_4_ZF;XRdN
zkt*ZfLTwpS&TCGySf(2Y)h(9Shr5hG;t9$P;3%y4=~FC|vo+N%CQC4!!C(S&-g7|K
zN|<l*n@g^svF*_`1<43Pj4LOh31A*kpz-VLdnS!lwUvA&3LUlRA3yTVJ)(YS05o?O
zBzd1Snt>KE8;m_~QqYXiiD8jLf=+1Y5~Ba37fH5F4;p_mH8&Y8zF_`GQg!+(D+X#`
z`6K7Yc)7?;_S~?%WKQ~&N@*8U4Ds4fhF&g8>Oh;B(%ge9yqL{ZZ%q4lTxPP?qQ~h!
z_8az(N|RaL**1&a6m&>tl=V;d${`GcL!V6U9Ow_1w65<q@%&LqRCk8({b8l^^7U4P
zNjzRtz!LvX{eZVMoPOYl%ggv?@(x!z`QAslfq{%d6ay9Viip=x<>~3J3|JDcN$x$_
zZetIOP@U4>M^<)_%I|y&@fiEm+l}89Nh8>DA{1a@TC?`}AU9PIqyF9Z`3|IGd4on&
z8K|7Pc%+l_-NN2jHAM;D?of~<Y;{ZfzNoPRoVl**z3-z%MS86*doDk}D=x-(jV7M@
zKv6($&rP3|REG<-%%ArW?0RTvY3;!K4DwQ!!~p8`mRl&H`x?+~?9C0Kyd6C}Jk%G@
z;9a%wg5W&R!>xW-_2*@tA1x)}GN=m!O>>mO;F-wU^JceJJ6zZG?_X>1!k+}cSo}D;
z!d?ZLpHr9Cy=6@_pFX86Q`FhHr?^sF&Miw0QPUrP71Gd@qwJFX_Rm9Aip}G(z<~JD
z*;z=TQT}(43a+cXIzit)yC}gWJ*^0moL}yx)Rou`lkuKk;R;5Y#LZE{p{`%MU!}we
z-VKT9d0MD+H;Oo}DxDvgW?wCbX1v1DcFb|p-+H2Pz2TF0?pbF0KA>=|7kqwhCytQt
zs=L$mlBRI-xU5yv->a_c8x(&1`A3-HL@YU-#!4qHQZF85D#aeeCeX!mh;SFV(65IF
z2WJ3U&re|D&C2%6*17LkywTzCWB2!G)@4I6g=9oFc;|!n&0dxzY`&CEYXcG>Il!*M
zLP<eQcpPw7rx+mt9VgBek4^zNbfyYb=zVPMG|)K8#|HhmK>0#?v~?2JMmvsKCYSV1
zm&?V=$q}^AZgS_ywp~Nr(91#dvDz5_(MfEArD#*Rdyu|tDBvKLLaV>1j4fnRcL}Y}
zO9=#7pUPbWm3L#*iYaX$;N*9s;y(*`M%c`jWF$ulrTaWMjF@&*;-Gd_JOn7(-N}12
zGUW+fAZ&WnNOZ$TK?Vj!`_##J-a?Lk8{y~(;{~OO+s|F(Dne@9Z%x_deqwJ%b*kS^
zZqPWE{4#&L7#Zo{8rJeKL+62e@l@7IXISl`Oz^ZVhnoazU;pIs3mv)+m*TGV-3YMp
z@vfA8VVNQtfUHEMx@4WN@r&yHm^1KKL=WKU2z&wSHfL!*gS`w>N5`YBVAXt~O2BD`
z&1to~I9<%o2#ZL_4e|O78t}@@;RgHt!Y?`q4rR@kn$ohe$nQhi88Hcdz2`_vO+9ja
z|6D)#1f_M_LIe0aThRS>46tbgG<Wuj%5tCmqR-RG8pc;?lwsA|v~u5Zsm1>XdlF62
zKgWAzLQ9US&|b()?D=oZNZ63I9;A`-c%at7;WIk?%BE7f7fala_%aYp>F8F0_9|4N
z|L=BtW+6e5YS!GF+h9GCG#zK8$6MbCxJx>lweVeQEJ{JXUnWF30^eh@@_m*P^qA5~
zJnc$AQm~#`q4(ppD6P`awG(D(<G9_~@D}E7sl><<SMuN1mD~{<HX;9aIeL^H7Q1iE
z@ou-1`&d4Zx$xvR9<fdS8kH!pNBPL=R1KDMEGu+<ef@bCmcrO#y)9w~t5zG)FF<OB
zwZ<f5J2h~WeUox>{{-x*#^)*wQnVA+E`txjHyT?DG$*cGTW=z0LGw8tWrKqdqpPVs
z!p{%ZzSZY*t?Sj1-Dx)-eB%fGibrttj;ZPKT*<FntXVakN`S3wTf%&KOL=B%bZ{`M
zm&7w!*Lf4Cqnsf#Gt*?QT%W`B@*ACAOVc-8`A`Tn!)o{m%Fy&b_09iXivRZ$Rv0)(
z*jfG&#xx_ytaFmqctlJrl+4{&+TAaGD_%FoE75J`5Je`~&RRbdhIfDJn(+fb1C<E+
zWsI4c)j4xXw@R|xmWDy$)s(L1LI$^n0~$BF<4wV~E9pNji9&m13|6OT{kMkt^Ijso
z0;j?aLDD-;#oh?2HFD!mo;dUb`7#OG(>Sv=uwu<=?zj{U4LAd8%<GLKwG+MYOj*_2
zr8znpa?R>&kY5f*->}!~O&Si^j(B+?u{_3Lacg5-1N31R=&aY9+$1>2Kd3Zf1YSo_
zR3bd15_Tu6@9p_Sv)!jv)Rcf*8bwjG_8XOK0Z_C&1JF7^S@<a!bO_&{oydUAV>&7k
zucDouT|uc3rs_LT@)Ce7mKBvmZfkOVq1BV{Ho%8OT=)W#&kMg_Wri|k-d^l07pa!)
zfZ%U~wNfi}im$Y|ClW)se?tD<ZGQ%_WZM$GK7Lm)6}4or`OJ9L#g08|6?7Jc@jX+$
z#I8XuOC+Px-91dA@chB!E|yIInbm4}tBl5RQteZ@W9`78Vku#_-Jx6s*JW0IZh~u9
zugeibO7jix$44w%VA_%SCx;}4m2x_3^C58;6gI=j8p36*x}<)Tl(uZIgML@DhcKDT
zbv%rUUF_i?X_Tn?TQ;#K2tHKJ<>39S41Z(L1#dUn)#&+Xn$sQ1SoKAN_@P0sm_aPl
zQpfi*Cgt8?y)PsPJ0-OJf1wzD+}>a^hLN1GJ)9(oZUs-iI)p-Y4XoNlxR$1?PfPyg
z1z?BOc(C`v83?CgjlX#sfabC===X}7lQVdr$%y4%vx$g#|L*>I!9We3X2m}2(vfF`
zNWZq`!q1$$>4Azm%DQoN#OMlAN_c0q+%U<gh5xlhFtQT2DO;6yZ=i@~H_ovugOE2<
zRG=0PHpjT3%6eczQ80?m^17ck1uN)zvU{{uMthu8Mq_oSnwd<<Hr=d_;C%P-TVxVK
zrnXwt#nFE3%><J8FR^M%NvX3Ha$oe%u;h<7k{JccNJNfAt{Ek|OTwLjNhW_j{DM?P
zmw8TP^RE%BDt_X?ww*3CA=|TObdgg~+j)8%6*+EF7S^g!S7G1Jl)Zmlw?hsyU1~vk
zU#41e;t4jaWsk;mKhg3cmA0^>iryg(eG%;L9*M1r!@@^?U{Ux#5f&N&Xy<jf1Hj+S
zq?Cyt1KT-`kkHC@oM<@`-UP6L%9n2hWsD)7ahj<aGeQrr?901Ew#<F@j06tnyI+}7
zhk!*(8Q8z~>3MV`{@3sQf7axRa0tT439QDvo5D>+G-lxjR(lFS!V>5G8@?<=V{j!S
zEOE5(B}-Sga@9_>VCm@R5R49Pz-UMD_7R1rwKxeE>+nzLYKg5}S7y`s)S1N4VYUor
z5ZP58OUVP@id9oS3N7!%jv!JZ($v>Km+P^$M!t}pho8lH3QBg35rsdvZhwv#vPi~h
z3x^e;j@Dj~4Ia1R$c2(zyr5is_&i=m3}b5gN-iyxxxtDly|_4<r^d=EyTV|&qhMH!
zDyGKE{=o&w{y0@8(?U#XD3wUupoe>Q@Lgw%yQ+OkC>~yrWtK(3@>7|{q%flvfxmF~
zhXh@-bQvbrkG^o3O0pvTe~J0N=8%d0H97m*DAJ&=vppgKCSfLKbBavr4iz*CU;cM1
z<D~}+1Sq2PYNgqrlz9dXKqg;!ruZQFogO%7b8>Sr#nNl^K<C-@FqTuu{_WehVm+f(
z|0S%~T$Y5xKRE{<!U$|-(*#q9J@2i<3zDB7RO|DDVt%ii`v3o+&@*6*O^tUB`k)MX
zT;>V_`pXz|cU0AQa%_E`U{@PlZ+jvPAGZCJzXp#ErPt}!a_87WV|~X^^xJt{O!3YC
ze1&RnQ{PLir2UjpsX;--=sBxkmO?~tXf_t~G2Z>)$Iyl)0yg2bO5K^rP@^{kb%fHl
zV?k`kL1DJ(e3dW#_bVP5ww-sF8Vvrs9cw(+J@nX<1v0@mZutHEN}r$@wH^JbvT0V6
zK?fB~>^>!{rfW(kJxjoCEt0V{{L9}}{glubF41JZ^2fU0Qvl?kSi>MAL6<JQto2g%
zgLzu;Sb<{FJBfxZf_vGi!Z;-aHR)IOhugI%o_~Y=CBphBGyJXJ!~y=<7qQ8)jM5a}
zrLSG=|5`Bryx3wK-k{=FUZ}BfFWl(H(~|{~0E}k-^0D94UQOs&Ctd)k74Yi%_1}Kz
z|NbyQ0C4xniKFIl(tsH8PiOQro8dT>QqL+?A0gDSL~ntddI0t*4Y-m-)k|3H+5zGI
zC2Zx2o(P_<lZ0GILnbF>bB!Oeh&q0MRO6|PG>K!GGESDJwiJ%89w>kEVG)Gc*^w@a
z=jX-HGq(Pq^Pp~T(S2%Z@-~}gx)+xWJi7FLV|CPXu{Zje4N+~`1AD0=6tc;P{rzjp
zUvi`78#C6Ar%|mPK36n_Qoq(Ym;;Jc#qEZ_W#pv}<Yh@B5v^8mt8zJ&2dBERN!Qgl
zAMq{}siBtWXfgDZs0<1VJrT-CNp%k~;`+xk*N#=0-kc9IckiEP&kg^?&o@@Uy4cbG
z*xks=+I`cYCCIFhL7m!{H~$N1akx<acxB~(5cZZ)b*)>{aIoO+?(PsExCDZ`dvJI6
zAR8yR1$TD{E`i|g!9BP`aQ3%yZufcn-qWMMzrYxqvDdSnIcL?ZD)E|59C>>B?3%uU
zXfK-o*^CE<Qz^B(+wnQ?hHJWUM&h^t5edjm4jbzgd!FXNA9hgo$a8e=^t*8W-#GFJ
zbRZ&YcPah4It1QYJ8=rUQeY`nCE>KZpJ<ornanR3KJbs=q^sXDR+_&R5FN$RsXw8k
zOaQwCI17ew^rMQIIG5Y{Y72j0r3}K6<kVtV26VB*dr#2JY(z+kCRcHl*0jhy@4#>(
z!prZBWNzEqnTKlxGcy36WK`?!if8Fwl|l>sIzBPV`wR2H$)sQAPZ57?FfT;ZT(3Fo
z9RtVLIb%n5=Arh-63DmeawkZ%{4OWQ8r>g^&Np@H;!v+SRKFAguTr!fju))8M@ck_
zaWbu%%fE9nP8u1=K-%!G$PMwtJ?oTB5yy@2AYb)+Iu+YGBMrc+nVsiT%AUxInd|OO
z_rB#&xjo8`ia5P~xmILECQ6eOe$`@P`IciZoS7uv8`E;5@6phJ2y@`&1$r{y3ZP&^
z>okqZ5q*gr)$|oej&O;2ao;lfACD-6FA$&}BYh6@lEMgiQ7H^o=Zx@i&b{A6Sr|e3
zV|)VKydbW&|L^%WP>NE?QR#ipy7vbs%>xwdcWP33bNv-VYE^B0SS0gNA`3^@NtfO5
zUzhq(yQ4;gFVcRwX7@moWsYGQ%ZT^3l>AXHyS<+Qn|3zDaqN#fmYd7PX3$+OfuYA0
zRAp%acR6f6-=F0Ou8f-8N;?C~L5grPTW<hD24}fzO$L_*ZNAJT*QT0XB@CsQV$13<
zY4uho>4a%yr7Z$$`k)65TONP%3oQ)HD{C~DwK52F9$)6HuAA@+eBRwWb$|JAbGz7Y
ztPyUv)dEB;D7y|*FOM~H2)d8S$Y&t1Up{Q}lK4)^x$GnCm4&NUAcSN%bfGD?<DT{Q
z5UhNfBE!5rwkVcKe#X>r`1uL_VRuHm^{k_%$GB9Lma_9&;T>S7C>V@^|H2dGcQQpT
z;l>l%$L>G)Wy?Z+G%=+(9E?y@6p!XT<Q!KbI+u|tSI*By6r`#WUfZo%rE?GTot}7B
z5fQKKNvT1q{|3Vo!2>WIYop_q*mAo^%>gr&s-0`{*RKo&_!@=47+BiP=XRt7tR{&E
zw?5|ArL@%T09PW}CQw)i9e;E{{*OcO&kHY2<Q}coU?{ht?<sEYg4`+(*Z|MIqmj^G
zR^t==8R6CLUzo|dm_U)FGf;?t<>IFp+4UUx33s&Q*3i>L$gGQ7va3=lo;s@fut4`)
zisJqphdkN-wIZEKZyy7c%Qa(Z@A2qJH03W)>!j-MZJx|;0k&Y>vge(|teN2CH{|E2
z!dmn^sUc*GQd~A|xMpW77S!)&{VzchZolmoC#3Y-zp9B^+1gr$lT}$+&{?bzjQBcP
z!#9?Pdwj>>P%qP<SmUnus8NS=y#}Q%7|T+?d5|Z8cI*p2o;_eOAFxfed!PWLc^z4c
zxjfYA3NGWSU)9Y)!Gjo;zQDNAbpCX?n2&&QNcvezpB%2?iH~+92No0}&h9J}>Mmll
z*^}D7(E&}dLJWo(*OM!eZk3~_hUFj%<I7`(I&tXE_4nXN&0f9zgu_J0L4))bTp)8<
z9Z8Mup4#m_R%YpK8P|UA-|h5&fA@4$J&*?LnEyVu{WU-Q8)TXrF7ntIk@WJtKHupH
zqQG%g-XCzPT&y(7A>!yunJqcE8PAFkZ$ACTHeV@^t?6e*h};f7oXsf)sPUpbZjGvs
zWuuD;b$VVvjbh4x1~AnFTku(7A-t^ObdKLuz2cBwg+h8@{OcHI($giZ!fj0V3yDk%
z7`^1?6Rn6|t39!JvWE3kk4UFaMu_8lCms){Dv&eSIQ5ka=W>Gqy87+W!npWadxjq0
zQ;LS4nV5BYI4;GqMb3B$s&Hf6RSpD8NITS9EWR@X@3H{nrMq=TG<}%V##q0XTg&B+
zC9(4^T_&7`ZOG;|c9SvJofw)KSpkIol`ztY60d87Mh)F0gI12|NNo0AzeSh321?1w
ze18z+WsZOS%tV&PR;@;o*-w<7K<}Sym@!0F#mh}EE3CaZkhq!u(MSb`QvzIH_)4K3
z`TzYUe0G4ox;rawXud0;z)tq#7K2N>T9Aos+f{j*izArOY0Bit_9-__55cG$xr3D{
z22H0K*G;ubo)rU-bWzclQ$Iv%`E2s6fe2uu?y*pk{%&^OdJmI?MDeou3zI6@gfP>W
z)Nw=pdETu+NqYZ#eFfkyuC+-`UJvLNaiavy+K2`Dkh6}UqF;Qo)6nT$&gc8tJKf~t
zQNZx4^NbIZn%XNRHk%2nB6)^1;PVx?uQG+M$zFNVyF0gK*FuQ#^u!vzon=Iir#n`x
zGj98nZjtLbdWrD#hQph~ZHk^#F21h-mA3tEkh~$i)KC~RhS;W<vF1ka{c&2X?envj
zM0lbLD5~CPapRj^Orx{eJN25gul;!+G?rVYWiq(6e@w~*yxits=xBUqlL@|N!@D^$
zQvsc(>i_!s0Z%<93fOP`k7G<n0etfmSOWh)%rRX%G*(_NpU(<2U`}j6A9D3aXUl_{
zq=1<G!<Bq}9EZYSu(ZRB72nu|II+)s^{ZFl#Fqofln<~)W(xk-ev$SIDu9N46eqCb
zoO||R8Wf~|Z8l!kn~!b!Im-$^vN0a2baOvTfs2L5AEE!2_#}xW68`B<*XnE)vRt`O
z<Zi>^x6bFM>q!mup>}I^h?lD<F-?85!s8kq<y4lB)@#dBG^&Si-pQ+ad>-d^5-XH_
z2_hAf?gfjAs6`v~N4R-|hh9I~IAaJca9Dmkp4PlP?IS=``lRaXiju8@`i@l<B><}S
zF}W`V+zn>vrREKDKX+%X?K~Ia9M2}E{xLAI!XRvz$E;f^_H3&h05}l~3G3pEy#zRO
zlqt-ho&G*B%AZr)e+=^`Zt?wRiFSYkhHOF|0oMPUcOesl-}QZk&&px@sMB7WQbljJ
zrE=J)BqK-2Df52b-jPA2s3ri&;Yv{Teou`>BqVO}_LgFVbqQPG=##xKeo<FDLuPrO
z8S)#5Kbk|71L+0KLG`HVKe$AzLg$-G`o;!<m3{AeqKDD4xV@YV`MNc4{-QAF-Y>)&
zU%8nLu{Xtr2r+j$(=I245wpA&E9(I0Q6mV-4uFq<&cEQXeH(#NLc%Mjrew7qEMG9+
zJeIrHtuLLEb1dE}BR8EtUqN^<!7uN+;is(DTHj^ndEN<0z&H2ctKI0_4)MNo-CZ}&
zvDx461=OxJztyKfo<C8v5#Yi6O+wqipEvTXi+`j#M4Xowp-SH-51Wzc=3yU)FQDtk
zZY_z`>QANJ`jzUe_%U<8J5X(CAIReW4Ss(eYqD(s%i!7OVb_fNj~nJcTH4L8e_$Ys
zM|Wxf(wN{x;8JD?`FzXjl#OFdZ}IV_C<t)jlTkm6IyIG(c9UVP3R!jNCdWV-RF27r
zuM4Why<z7te~GP6>15V#>wRvfD?a6#vC7oqhST$OVlZR%ldZDt4UScQu2nWG*hXm*
z@Uvr7naP@u`7e-;_!mm-bTT8N<$4F?e2hoW8zanPlfokUM*A!Y?4W8xhrC2uY3!SC
zmPUms%rVgH4%fdvZUxz(9d<>~xwv32H!J1=-4Li4c;o5$+gp0jO5IPbu^`nsZ_n%y
z!U-=fcQo2g>WO4z84YR{l$z~7e&u5P`t==rv?K2E%W1RN<yoCcb$|P%vR9*TU7$Ul
z*3V67h4$k`U8{#HNZ?iz!^hWYf~&c~%iB7e8`h#)Mh;nV@<iX7`Og&^r4;!4*CfM~
zGN(WOo1Y*dXxIQ7qmfY;cUY}fXKH+;uFm(!j`6b0Przo<6o46KEocHP-gMDV+Y;A*
z<%oF7Y|d*M76G*WjMZ;gQeUO_vmyJiLM0=PkeK0Yf3n|lAFlWL=gR>{;BAwOHNdK{
z2U>tSj;8Bo!wa}E8i4e!bjQxke~p+FftiY$qy6pq^_Rh+VIWV<HI|2mnehTG!E-VF
zuYsfy_^g78+UH%>zdlv!B`4rAk-LoZ3GLwp97~OO2e8NYw+aD}iWx+Wq8g=1-m{^9
z@d^0H+Mi?#g%d_Ib}T2`p&#v^jF>cqY~yc$Rx?|03oS!23TF>Mrl;ph<L`oIcTJ&?
z@r^YCA}s+Pg9kbv2dJbnvZ-QJdf{+2xm#a+hN<7)1{BG>ohQc~{qQ*|N4fH~7X~zu
zNq(t%@B<$7`>bgu5DEbsD@oiA$B?q7b3y9mDdX^0hP+yx@#ebIhZ{am$86opv!YZ2
zwJrNtl6(#bq!v1l<CG62itPr{)YtglHLsS|mjl9$Ph(U%&--pSXYDwAZX>tOfnRm<
zAeFuO;^)gJ3CcC7)oYQb-BVd;WTfC$g>n@mcya$xe?<<_K0!DfI-*20o&2l|2Bf6Y
z0Go^uapg!3)RHNhTJKxyX=;rxzmd#5M?i?U2Y_j+D+MsOs+PQSzz_mIyIfK<W$;8@
zbjXRwyrH;O-5&RDFdr|R<z;t;|I5_&dKB1$A!-Amhz&cE0K#$}U?>%Wvti9rClT;e
z@(q6n7;@jrSb#ckl0U2aVm|{~_WrOA77|Qe-}~B|mMX2f{6VjaXgQIOWV%`*!)Ls}
z0zG9$*1wo`BgHoq)nn9Ag|}=Pm2&VNf^=Cp8WH(lvIoEo9>hv2So(MJKbNEwBEzMW
zf}mj_(5a<Vq$A1bjLE{K*dRl}eME7&K@=a^0!#7YKS8Rokul`biJ~bxYjG=_fY%Hx
z(hsY%58>PS8|L^+Cgn7%SLWyE^_%VKS!lJ12Ae&32o<raH~TyyhubgiPB&!39Pq@|
zv(pvnE#9F^z#p3PmvI*v%td7~rX2LPYV9rhT)bH-2jDe6$B|YjtVS1-4q0`tVk}to
zv@1OAo>~D2`Qd5C^GBmvD)vBJH^5vi8Xx8HjGq-#Eel%ZyinS`ex-m?OpGpHosmeR
z4H$SxLCMD0k6W;$c0V_FJOW7VR>SY!2?zZOsHofYy8flnwh76o@0k}f!9S^E9WuUR
z?@|$BS};>wgf5?+Am#xWd%AzLvRPZKx}K51<yEJsTwuNE5Is`(y0M9~9Jdit?CZl{
z^-_uKbBt>ZJrHplpn)@;ZlMD&ZvMPbWs}F}#v~AHsXPiSBe3XvW8(hJBDI-nXm7Ip
zEvB1N`$1RM)sAbSE#>YWf%BK2&nx*#SxNF;`N*R*G!-NUM8s!t5ehQ&I5;Wj5o*YU
z9$gkz8^@wAB!V)42?;8o6?Eyan|4U(Aw-H5NCf^!!4Vp;S&+u0k;?_h`wI<w2%pxh
z&6c4f$uFP7gp<TzoJW%A1=2Y_V0co7DGf+LLIYr}c=MV&p<)LTQ!2E904#~$jq+DM
zHwM3(X-VJ%BkisLd0_#&$$%2UWZ_4#P3C@zk;=dSH>GNHN53a*FArPH1$LX&EYoq*
z6vcWxDsl;P6|ve+hp?_ETRwC}m)aq1{5s!oip)pGQ(dB+v%TeTHF%xVoO7}i)U$n+
z_bmp<ugi}?&<3SfI4n<}Il{iMB}OBuenA_`xfAeewo|rA;<YAPI_6=JY>KL6=qivz
zqSRs#FYZB<SYp=oy;(Fw05FJYh~@&x7ud@?aS+*a3{>w=dV2A)F|fybCg>p3`Er~f
zYvu*KpfS&*G70hFx5!M-6YyLsL+~-D%S)EyU*EO8xbt9|WW5XM3jWH?b?}5VJ>~Ox
z3X#Jik><Nq-j@Jy)l{`U!s*&5qR^Ei+TK2HS%a8<v$2GVx|?#z$Cyg%tXtlsj?oUU
zd!dqeF!B%%LyyZoU#U^wCf{4=h!w+9^t%)M><-;0Q&5}wQzFaL3pL-S(JvSHpTo46
zSw7V<%Z+tcHOWA&vW{A=)MEjWc>Nd;leC-F3E<qv#}^NVT{#FPJekW2z75$>o<cIO
z3kf{(TtA_ubGfOiqoUlA`OiZ`1!)%knd%~i{GK|COG9=tM>t3Glj5Hh`czkU4!^&S
zSDYH$`z(dzT$p?uI9)O%=|nj3s2;$1z6xWK+kO?#4ud0GoGcF=$S@PO1TMgvC5s$t
zS6&l(BDnUQ!B|?RlQ<1;IHK)N4S=g1)9vXhEEjY;n9951Nag+e5R*xxU7W3aSl8zt
zPR;lrW55!aX)Y~WlGLCR_cCVgL_UDkab1jtY;($&v`|qq#cD(xqbB*i643-?$W&qi
z_Gl*<dAa7}<9L=CA!1MAaL7c%o@i|-PDt=m%dBt@@N%Ndvl-E@x_@k@xGT<adj-h~
zCqWw!GwiK=Z*bw?>%OVfk@nd!PZF{<eW-R!TYg9*b1>C_f-yi-3wRvX^nHGIL3iSz
z9i*e?JoBG-0M`higm=XdXs5|(i<d*!U1hpd$~_}19?2^<Pc=;P9R#Ba%yqUWo^=ZY
z$S++wz3{Hv0RH<MHz}&rSmEC@rOGTu@f@50z&WFAjAzBU);2%5SQLjs$TPT22R^Ht
zS{2lfn%uc8aWX-4JF_id(XX7VD@e9pL{ZcVt-rq5VtKsP+~FoHw;j*WJs2Q7edk)5
z1=O64&<{n_FJt7n7u$uIPc+Ih$KM3FldcJWDJACp{Yw7px}buL2oo9Z7^gK5RBXYp
z`&lbWjo}BFlhEp|nSrSPTrDE+p@E3FkAS?B(l(w&o&pUKQj5%m_MH75pxFl!qN5qn
zrL>^0v=#P&1*3ex88vZ~IvZN{XKx16&o9LMYTIW3Cv3|RiiCp-?JF>R1B7zV0IOL2
zm|tnk-|fVg$$j{*n~MrkI^<8c2{S63-;26Vem9VVBx_$~5KKhOr|f+`I_#vN+vk`k
z(9yA1?{d4}OP0PEO%{btu7Ou@LL4(vAwt3#y*K8mg-dbXzqGsd#H#+hVPeXzayQ>^
ztUiIdkr&jD2qPq+kX|rA_?Uh<)N;f(()EO+uQW;8jV`tEO^2zOu4<L?4SH-Q?78`a
zroW*H{v3j8OBT(NAmMpbAs+ojeH&9z!#g@cxOPU(N6UEzR=?Zi*oCU=OMZtASqve;
z1>9Wu3@o7zK%6HLg)}w(`3GXr&uV=;*Y7YIBSUO7AATQOXu|gZV`kx`@ZFV9N-zg!
zOkWtQx^t-Koj$6*Uz2Pp0iy8@p~HJMv(1^(wQ7B#b?t#+>;9s6U1gGFMofL*x)|@f
zw^oZEX>)A5kl$w^tob!(z)z5H&})#Y)!DCA5(!I;^9+5+PNIWk@VW~e-d*a5U|7BL
zjQB;3nF`$f|8W3ha3G{NnYDdKnQr4LC1(JmB}4Ov(dd6YdJ4)Z!1&!+R$SI}b<||(
z57i^R&K!%Yu~3E=*u3r?m8)iuZ3k#4sfOE&0zf<t>tiDVJSqMI9Fl*RB&zq2U703H
z-Ht+g-46)#&%&7xC?vUJG?CkmnLZ?D@}IE34OoDpaCQwZzEc>}Ka=rfYB8u4-Uw+P
ztYWTzH@_HOfeu_WsDWrrtE_11jm5?ET2;GzwJIKpKilA*sgnek$QR!WxcnI3wRBRr
z^m-Sd(*B#mW;}7I*>Dog#s8MtOwURE-SEhJ=}?ia{KetxQSYWX{KXCN-ydTVE8kwS
zvZ#$n6q}k(pDd9)?qb%+tTQuVfjHbNl!HtC{TQ?S+d%Vj%xV=H)5=v}LD(+Rf+CGD
z5V(3qL}W1gd-OkSOb&mtgFIxY_#8#@{GPDZUnKH~`F(Hpa}dIKvsvMi$gp<O_~t(x
z9{w;@A_Q>P0wF(JwT9yCw58g-<gOR)m{4RFWk88hIg%lBdj!(d6U?@`0tY{`z)c5+
z+Bh!iyQ^pYv|rr8!Q(anJK3?e4N14VALdkCj`Zaybb8|eBTT+KM83y6x7Basl9~pG
zg}29)jI6b%D~liRR!vG=CYv6jr~+i+A(+;O{O%2${hvwHGk;f>)k)K1%bQdF)i?ax
zu@DdZ{C#nZP9;J4-+ug$!}DLiC1Zd_5PlM;_G4I*EQkcUE@tAcUmPO@E`(+r@srV_
zoy_1hnBH6~+&jk0(=<Ox)I^s(i`_>f<cGBHDKm|{Meb(`e;U1%;pl?+$I@QtRmZpL
zx{J+*@WI$DPvKbR&+6X0()I5>H+-P3iAReI5(jYT^AURAvmD5z^Bg6+z=(d})7G-%
zYO+CP&??<PAGlpx1iBrJ&a16M!P?lX?DT4lfrVUjvsJb>Up1xy9pGJl^=pAV@51OJ
zmD*b&L};nf)h&}iWaM(NSH`wyRHNgCJlF_$vfTLy;<mk*Ruq1)948<>UDp>cWuhYk
zmR@SS0btR`gHhk7N}u|sxP%mFEj>2;S45jP&dkNk$UC3x8T383B0f}R@-Kcbv@DUC
z_=p)qae7=+0vNWZki1oLKU-NGZts%#7#dSJ94sZ|n42{pYUjG{r9l%(>HrmA7+f2L
z;{|Z_xQU~bQSfdDhqJ_U<?|5|7_mr235Zn*9$xWETd)p^o&8HkEoHmTiS)An`F2x5
z76tw(>~BXnQ2%-((a0fz`Ec*FT+3e6q|P^PmMXci1c>w~0=y8FHD6tal%@VOi2vJT
z7r~&|d^X26+u#x{4Ixx^%N-gB5khpp%wfM@u2mM)^{3n$GBRpuJWGXJ+u1r89nV|U
zRxR7~(*WP%*b&CM>Kg_~drb<<TfMvmcsbHY1?7os3YL6$NC)dMg(UYwGqsjoK_h~R
z<eZsl-H$4Jhm&P^Hc)|TiGq6ySEmn>$3@vI3Jy}qB)i9$P+JUaTXXKj{m|>A-IQ?+
z%yE#(8usWz84q;y4@_$#r)9lhL75aR$<`_xD(q_lx(kx)db#SJ3u*MDqP6ywQWMDv
znRU1K`Lq%U^>D|oM`ap(w`%#aEW8TbYfYn)K#p4hHe*MlZ)@uB6Z#{TROajHdNY=C
zt#~zx^A;y}>$%bcgJvD(cP_=jM+;3#li3P1wS1O6pD|4c&=zekx7GN-&tj0%51n@^
z@FPEp%nrac?Gyd{CP&?egF0{LV2?&AC6~)2V-J$tp++~I>0L*e?Z0vkVjd9v4%ivM
z@!p1qy*=~m$O?_2g8D%mxLS6ny0s0mZqweO9T4<O&;7gK|NBWPJ;G(;Ih+;6R*WhC
zRZ)M0hyX;4ain23Wl25K0Pp^fZ{wz31Ck5!e;3?;n;a4lAtulCll;!B>JmMm%Ck6M
z$m{2AM9NP2-?kiq+X5C?<J3RYwb{tb;yl7RIXZza35h#*Xg(w`)_*Q1V#hC^0h2_%
zmz0f6>#1nzV|1Wsh`bXJU-$DF>O9jmTCYA|oQlr~Ro*Kxpt7M(@kFZaNX?;M)HH-{
zAp+Z!DaA8PX38#9A`B9nJg3gI5npB>if4bqNa<zh$!5-=FLG+YbBUYA#D>6aR@{%I
zw!WNsi<Ad+hBVg7vQ>Hj64rU2AI8w_8s}Yd*f5Svc9HEI33|M1E`uFvFDomO{W+hi
zaIVr^ogC4kLBLUHNpPNO=qsmmTWejO>JQdeR#&wcznTJSz|s7u3?^-{(*A_Rj}cu5
zd2+fd6q1vC?FK9YL7i3BBNuu5R0ee>IZp-Z=ls_qtGl&*jT(7@Yk^xe#27rI{l1S)
zHiKc&zPsUKfPbA{*-tqyPR<AtFDlSLblo*2Er*_{t$9(xrtRhcB5E|&buiRug9x;9
ziJ&;hp*xR{m*qr;zG@lTdldX{kLRxkEs{V%2Hzp=y=7YWSB3t6bKl?R;J^DijKJNG
zB2IktKHKAkyWw{=-`JKos|JgbW}a>9fu-s4N-|7k!GIU(iFw0|OMLW_JWiTHBA;D@
z^4GJ-xbn=i@eEhxtxv)Y^xCztzM11quReU<$8;)J&V9vt_gnl)%VmT`m+0|<Qw(ek
z&<0V#!-kwWC6p#Lly|JhF!boHJTG!#l1Rl24Tj{pRf+R#6I>N^=dHA$tu44ceV?pZ
zq?dxrT65Ecoy>30Ds>BbmA#Dc9Xe_wgRQ;RGp(^3H5vmdtqf*gTC0E?dqn4CqXxiA
z?7d&TP|YkbJEt$;;y?H3ej3fsaQFu7QE*!?P7v@=pU_1SJ2j>-SIyNI1)uH9b~_cs
z8u*NVxZd}%1pDU~$@=-eKkP-U0&rLQxuU3;k1_R}?CHR?q!)6FxHFK2N43u9w*A31
zW*4&ap*W1hlgyJ`g6gu#Xj<sJW40P&%<kI|T*GoB<RqYGA7q$Rc-%pynfltug58~W
zSRyFmX0ReDxxTy4SRsHpx5g|XSxuW&zlNW*GfbAcs9=zjEa2<uhjYc?n-edF5*4LV
zWq-S9<fh!#FP(fcNVhh!gDcSRP+QtfZz?TIE31K{eg6;&6=g6rsa>JDUqD03HKbGZ
zcFB5inmM0cyQmO3<!QH|vQ4<V^w}4g!=a1P>X-tN<)8H&nv(4GD?ZW3?7t_i|Aq_x
zd2E1=KNEc@;K@y$UOl6-$(jqO<NQUpM+FJmjj#0b8LE%{FGIkzRB^ZdF?%Cm{w?`N
zcRPKK`Z7mNo5B9R$HjAAEUd8P=)x<R)05lF4HPkc0MzI)&Q}G_^M?{jnhPBw6sJcT
zY)|y1gly5|9RKt$Upli_2?H>S_t)xTP*6z$;Ky6~cCyPSC@<3a*9Cjha$OW#mKL$<
zaw9H==bCU0W!u-w6pb@9I!_mpJb+cSPZwW^OdioE9#;dR;6T})y;TfTzxIe@mHXXq
z?xlOWNmA8|MNtWFP$9Mwwg8Y=<1631D5`L#7^&}Okwy09)JU=r0Ov~TxAPt5M^*m0
z)vG%f6ui6;Vu*DMMf~&QJY}y5{O__fsig<M(jv~N<Op#mx2)8e#3wjcTD=`GY#0~~
z`9w?n9>}74&Y^F4SG$a1ozs+!LwxkXXN}9Urh?|wMYH53`zYzOOr%vP<(FBcePhaq
z`1;6i+j8U4UmGT_DkJ`U{dKX{J5o!d;5(JIxfr$NzHp=;2D+WI!v-}%O9~Z=ze6n&
zI)}AQtMnwW1j^Q&Ed)hjYP$9G!W!386FWr<wp|`1|9QJh*+WwW^0e0EFm?X3Zv1^;
z{0l{jD}=-9=$6uSraRRzhtXP*ZgTV2f%WNptLC=OpCsrdz^mzBep}p;3M3l(Lk)jT
zFS-s6-ml3LyGz2^hg&OfgPy)WV5USKZdW2uI9-iF5cX%KL;N#hg?E2xm_&9)^-5rf
zTi=C?)Xh~Rq9FgI;k6JzWuUQ&o$hc>!sq)UVlKchHH=+>z`yKKI7aAzE)g4*(P=Me
z#Fu@l8*(<o-im&4wjt}qI;O!GST!rU%{M)^r*Z%UuF4s>&9Vo2da1Xk_1IKs<*kY)
zCm823R@YVxq6&MaPo3oTM|A>3i4fn@i5LNUuL7!Il9(iZ!sX7IT3!&Sx?Mp5aiD^n
zh&O*KS+gxf?B4wPjT1mR5Q_4)YIABdF+=Efb%b{iet*QjijP_6o6{7O$(6EBP@qnR
z|4s(8h?Fj&{)BFw^evd6_*5-mQf@_IoReF1ydZ4&n)3^3@V2fZZ?cN#>ZaxF7nSAA
z!JRUSqUT%@Y1afE3mWXPfLG55ud#Tio%@~rtDJv6sJI)rq+=d8o(cx7`_siQRhk$>
z9R{qedL8m|-rkVO3|$iRbaYY&kJwkKk&W*z2cPdLl)cOHI!g+ZR=KrEV#6GhcvfZ3
z5vUav6xO|Gzmo<|1;|8^LhjXVpywq@-Ku$xYJsn%zFADqM|NjS?yF)p88dcb)IUG_
z-fF8QcqJ)K%Tj4Kx<oUo%6{_M!Ju~xiO84FGaMco3THFFg1);`9ai1GQdenYk9K1J
zEe-v_M^f9A6!+13vu&%hjcOx6EVU=ofg&bqELkfw0q8Pi2>)~(yJp5VahL6$d)hT|
zG0oxJHaI>A>fV-bi3w&hIs(AhSn|U{G82lzc{i`YL90&Nqb9v&D@870!7=kE+hz#D
zHVx-pAgo<lHp#&WwaazMmI=AdOV2XLr7+Pvgwzg25MxT*`4D-%MyYri?rpvrO+P_;
zJVBWX9)2M1p~>;tNP}Ts!Hzwf+P3<Y_9JFkPi2#WMoml+^hSJbQOl5T7)oSjk!}KI
z9M2AVevogMv|SR{!z)#d{BA$^-h_3{yCUkFooYxtqXt!ZR==%!?D~pvZJa162e1Bw
z;Nr(9oTY=i4Zj;8eyiAH{c>itmBbsE^#xxVgMsQZI~31Su01Q3s)rE(THF<9{-(0w
z7##KbD7%U?UWnAc21_E9r%@$y#=ksMWB@FfTgXwFaJtbCvxNz&AGGTceUFM#P*{>=
z=JYQ}N$*y!h&DW2tu!pnXHNA6Kg%l+k+Nz;nelg$^G<Zf5oNV0Bb>7-B~0#m?tIRI
zyE63VrdzuCzL*d*h&%JoQV@6*EMm;AYtjQq=mLuVZ17s-_<aOI*?3Mjx^un&FG{gi
zl@1pWIg2x2W0><tRM!T?B$om9+3dUKEJuL!&<apq9@UwR;_$g1#Ce-qSYQs}q@`7<
z6d|YsbQBO+`~g?b5yI_gF5-{m16`UU_AcxZ&}RlCfCR+9xu5C)629h%<Bsczs7SoP
zjzHd8@L@@gC18rm@SE&S&;k6aQ?5d%&z;6G?(uF*X;-Ndur}!h)9R1?0;J%6tm%H`
zs{(>Efpjg>M*!P^-z7y8Rp2|M;yZ1eE!C_sn5!`q<^vShZa<SiP|Fj*9&B*^#Mqdc
z6OyZ5moM))AS?HOeXzj3TW+#Z1u!r#x0fYXK!2qi{1zE9IO7qu<N4Gbh@S61{rxk*
z`E5(JWrJlqph_)3WbIJutOnvg*#!grlTx37qjFJhiG_lBA;qfm1b|?3KC*-xrFNEm
z5fgvHo>x<$A3e@i)4LN>SHl$apyDtZ-KQmRt+HJAf&If68Fz#m5#65KndR@nuvUP<
z{0-n*Q0hl}hJKUPg2QFAK-EaEhq+2h%;!os&1W-e?uwF+d1g^%W@e@gh?^QG8a>Vp
zdO0bko~b@mtpN5SBnO4Buz|&Z(;A-#C9FWepNw8W%(OQgVi)p+59=ra)&$~RdtKoD
z6al5*ADxryg0=-Uv=PB1{SB?o2SPM-Q$TG~^1G%b=jDa*G2o{Ez;OEPVhmVab%+r1
zK)qOZ0LBlMCzd4A(?=j%`zY`U?vCM+naFa{lV-~+YTq8vL^aJLdvyWCSZ@$~H=wW@
z1@jzu0LC|!?)eUiKoXboc|~rIFGZUJ`T(4doo|85=I8x}1z`9R1-t;J_>Lze+$<L=
z)4ly(0^<h((@O0vv%2MlE&l<0i_%F-fu}Ged<t8@pD2GLGn42M;KSx^9s%)1#pHW1
z7mPM`$O#ZHRJMq9hnX$fgRpp;q^f{Ut<AE^ZTnNi(C=1VY2ym`rpo<<9z{L{3cX!~
z^Fz$WKNzHBU!C++ctYNRj5q3Fwk9pf_Ad8+k#G`gHS%q{>UQFMv<0`|n~CPc2n_|0
z=J;L>y8`F0Vv%r?X8L}Tj`aq9h<0kqAqwg4OSpK1lSZK*?2IHhybEFH7ocNSAm%YS
ziQXs3Mm@-3vlrUWbeVJFtTytZVA*NK`)Of4*!F7k`3_HcrB^W4M-ZovA8$Qu-D{t?
zDfYV$L^Z@R%o*_wsVD34QbWnI$5u82ecuz@**-uUtGx6Jg4|I(1e%pd8Y4c&5R{!W
zcp83x!LBeu0&$mc31FIdDfYdG-7&AqBVpG4HpeIF`o+frP!xykL#!)Ko#(xL_Kexb
zuO`es^<2+&m||HL%?Y~vl)JUw%kL<Fyd&7i91Z0pcz3c?U>qZaVD=-h?g-E%WBlsm
z%)U*v`qO((lW(ro$&<$CXf7a(i9}5?0eg>JMjU1vPKhk$-)B%mpbe-c$@Cu#ZEcE1
z+FbM9hBAax1$4O)3oO>0am`SO4Ng+I{3&=f0~8qOe+w@L-EA=Z9?a>T6v`tQT}Ytj
z;u=~DFs{;A1M}Q3V)F*#!eb?R(jw6|f81kz4>gXDS-~hQl+k?k<)l+JQRb(r4fbUk
zOo$Z-7gl>DJzP{vWW?#s;+@UV!;(r<=-~$dDA}H*8*D<ZUVeQ+h{4}6<|u#lp%)Wl
zH0y=tm+$D4$<F5e9uMa4>-442+8NOBemYgtX`YY;qECL^DUs^CB}eTKR6hF~<pn}z
zJqslbxWG|M5(uI@W^;|opLes`$8H^M<OL@eD=Kc?r6pt-ir}X(UDmfbI4631Flrst
zkNR!8m~XZuv2n&+Nw~`Nf_kL1B`LI^{8V|XHlHnz=YYNG`C?YDRyGm|UZZaU<~wHv
z-TI_WBE{1BjLT7zZTr{jhM*uF&*(I#wHQ(REAfA}hjg%T7gXjm1d5Fez`LiK2Iz3A
z(TN3?YS`du&)h6}Bd&m?CwhTa0%b4A(Me3-5dgj~Vj~|`djOQ^SZ)TdzUi4p46FgU
z-JaU)iUJIkzj2h;OILP>6S1???r5<u&w=a-kk>09R28*i=M^)83}LJ6USBrMR#HzB
z$4*kv&U`~@x0HsYu7^d#N)HUptt@h%h_dfBt2?e_U_3#bH0|^r8-C5bRDp3;Z;NYQ
zCKInmC}Rck*`|5)+z}FMu`~v7e)UuJ)Lu$Jq|H&uq3&f02XxWNXz|d%(8;MS6cKtd
z_tIdPy`hmgzLL;S`+>zRK?8fREd)d^iC(5@m9v%2T<}@euuh~y7k2<94B>IT{+$6@
zUIppMAeEL;U0M$-5^H(EP|Pq>uWQqnjyN%+dSmgl$7>UguN;1lt{`}>WmDN(qL<=s
zxxj8Ydjy8TikhzSiy&B|pFN&|V4s@<qjF({5xO7n%Ui%IZW&uVOTII73jE71u+&i{
zeuZQTX*Bmheh%C>23m~Meo+^4Ojw~j$1S!?97bMF;4sb`o<PqBu*u0D4^0j|z}jMk
zUWW~)0ex?d+QkWiMF*Xfk#T#F%}y-oA?q^12zCrIsgPb8>$0yl{_^7}Yj<TM;3A;)
z&GA0d=d5+)4=2-hPYh(yF^~rE4#bi=@UhbD(YyQ_i)azP+y@y(OAJQi7ksj{9$-^|
zk0u=tSk;Wj8@K_yLrf%+cm^Q<*oCKg;z_&16aEJL#G*=Hmu>o><&D<<sQ--Gfb<Ro
zgh|8{EQ25PNEPFV`*H-DKuQkjG$#e!hhR(r)$`VUk}W({wJ#z(pFaC@;OxMNiy@_r
z5y{qDw?d8{fdhuAQ^m7bFz=HkN&VEeBt&~Tp044Z1%V8J<hykrR=UZG>NXi%(ham;
z{B>E#+27!O2`5qI^phGk2pzlyQpu-LM5JV(<O|agdr<%*ab_CaZotL2`erw%wJkw!
zDr=G7cZY?ts9v70YgMwdohBJ<OubK;ptg(Y_buSn@)Nra@wCahLsa<0F&~FKHeWIg
z7)D|20g%Z$zXQU~KC|~Y+ZTLPL;q~&-38^|MySzW;GE{o`}V~2n*!i6bRMdeTQ`{w
z2#)zCz+o^<w&~KI^a0A7?n#$N4dpkVp%Kh!!Z00#s^1QBvr3f?No?~?Sbsu^o{wGL
z)!l}~24iO!N;w_`?P)ytHzB-MdjjjsaZS6IoPLPDjLb9;Ir;EBPXFxjD}hxX7?O8l
z4f?mHnkAJ9kzM&u8-2~bLwb~{J@%olDJQenb_|1sHLa2_vz+&bXOpmFo*14c;LmVT
zTpdb?d~74P^}wBFUa2arj$pHdyAhW*AtXPYb>GY8y&0yAG1-6K7~#T@((ONRj?|S(
z+JOAc$oX|om9a)qFE|;yRDJh!rT7I`F))m&GYq?vrh1F%RZ)2hrWL{Kg;Lf|gbY^|
z&$?&dc_v!-Pe@lFN2A;rEm2#Ix=soqv<-r`lmqC&coIsB`^1_eBZ$h63Z!5d)WnoN
z1Y2V+tLOF_m%cAL4RdB^*Hq_2O6FPK)ph7BQnf&dhPq{@iK?)R8E!7`HOdX30_S-S
zQ+?uMrh^LOD7_QglQm`3s6^upb3?>dEZoN77Df5&-&?awj7MVKIBYA|yJy#`GaW|i
zU&@f0UWKT3l(t%2zQL|MV%z9U`h|4-o%@NFdVh?xGrG|cEG&ZEU#XxxV(7fe$Oxr?
z1FTx|&MUxuhet;EV4d0nl$co2i4=y<-I|9wNSS6AZr*IX803#JsZB?=;WS8GK2oyO
z!l9>38j*$kjBxrfO$3BC=^vj^CX9j=p8mePd_YW|c#%fugKi(EIS?!kJ^yOTT`K#b
z>N1qj9<AIYql}Q%QDt5)Ppv|ok_WT9u$L`8i-6m<^b8NpDgZ-^hR#Lk>7>D02aqr#
z-Y{H)l~DYjF9+wlUV!%bON*fK>dUIjWSLvvd*nmuC?bex?ObypS#OgJX48@0Xlxu>
z))+(JvKQp|z3<VaqbJHu4dxa2a<twLAf>{&>x=5Qd7LZMt@3PLNkfAd4SeL)i3Zvu
zH3WIo9&)FmJe~m!aarU*UhlwB*lXDE!_iMX>ydc8xsY_S>M!K+U)}*^#O4o)&zi09
zj5XqASW!P_yl-%dwJ)aI`8+_~c@bj(HiQ$VwWfh3l%zX>iHg>Gdn;!(A6h@4i1Bg!
zAb`3Em5AZyn%|q-;RZbaLO_2_uvI17(*x0>_LVzq1{ReQ&S@1NAppOl-_0tGnlZ$<
zp<jhSBF-xb5>n~Q&Dpc+IQ{W&pP__!ZG_eoJDk>51Z3i)!ipOA%byB{DZj><E2%O@
zc5c<yr3);}M?G$8g*F3NQZU5>H4%H}HCGPrHfF7Uz!6S*VLhBJ7foSnzJK7SS;&))
zf`(YQ*m}UH#TV!WNsE#{=7p);^A%pUTMauzKQ37qyoMe}mPYpm(RLPq{3nE?vVXc@
z-roB`WHoIe5_Dff=s4YoIX|2f0!ym^9xS><S(D$FswdLQNkV>CY&g=f@KDn&?$vO+
z(E`k=`8SMxv7#d#ko9l=7qIvT92Iro9iSjnqh$$rp#F_ZFE<i|hc2s=rUxqxof!4=
z_dA2PGs8fVloc?EUCMA2yNXQxa6hL*JtB%CXV$<{l_so<b8R$x69ET}t>H53E#W#5
zS;ZjDjV|4RDtLodSDsY%Sa_XEE@u#hq^n|bMOs+Cx~iufSGp>4(U)>a<{8GJ5rE|#
zMhQlstlI%h7p!-S?B#6_@Oz<5<>HO%1+|fPe*=~zKU~M_iab+U^o?~&<0|hXMv%83
z0pm?fPI_#sU@Bx4h8fv+CfI!II_I`-HV2lYp|7tIwAo6-4AaHmt!D(jJHiB9yb3N;
zxn7DyLX{yL4qT@fj+PSf=GMT9j+|UW?cm$OvvM$Tf4_&MTK8Pu?vE-1+;ElScek~_
zSkCqaQ3)t6R_Wr}VTfKXRO?r4qaSV;=ZbGT20bC0^Z=F}?A4R#(Mx9?*9w5{T8~Hu
z-v^_`zeTdPmX&B&P|v<)QmangZ=+|M(drlan_=vz>wNt&Q09e#JK)w)aqj)g1qB5j
zlA$8FB1*#8FIH@%vbUnEFO&fMHkGV(!h9lA8hC}6jV>)3e*?2`#bk<SR}O3n%$+CT
z4h#%9+UzmJY`=~HseUgD336q_Gb>Qfl8_)%8s9nf#@7bDqc6O}2yk%ndupPglARz#
z12HEv9QJ<P*?e__YV9&Ch_WtIOJ@+|j$(2*!BFj&zEi%9)x^X{Q)0Fn3*}&o>LdRh
zVU(9EEE+X0lfD>hRNUzF3RgkIE7v+HH>e@=T1x$!5J8giS*DVWV+%vHr`-W{uT$%$
zimnK-KRb1Iir7t*ig>+>iq&am*##{=!gQ5&bZyoRs#v}}zRO00dtKNZ6|&Zf!A$N!
z6k$-WpF(6Utj`SwmL!>iEN`tzfz-)>XoC-Pg<&{>@R)i(GFGgzkSBd^YG;b&t)2~X
z#H_5SyGxBvby!((T<h<@7{x1S#X@CUH(n?S6SVo^aIIZX%m4He_VYRsxT52}BK|1}
z<Ip80S<@bba(gP)Dwpd9ezhqxNuG@2y2|#IOLO=YgE?dwpTHL6n)FY)GvRGuA?3v%
zW1x3HKBFo^7&e0%*}6Sv;WlA47M!Y*b(Fa=E@m4bT`oiwy6>5j*thd913b)%dmo_!
z@sL2-^^L$3Tyc~@77)UT7@)OZ)kDw7$f&2Ju8w3n<SA$BBS4buN{EQhU)w`Ae3h~h
zs9^S*)9j@Q-)xQuY3pU<qVGueC>prTq>UiH3BW5yGwjYiO>cN%5wt#<i*Le^ldFJu
z8*}C>yi^8za$j5Ajy0eRO)Ss@as3b+3~h<%1N}slqLWw8YDlV?lgbrcgE=UqF1DhZ
zw=|ymfdr{A#Bz+lyn_rWgcHw&z`}Zf60Y6j!F*$#PG<Mfx_Q+Z@#Q=K%!!R`It_@j
z@De^sMC5Md2>N^*l5Mj>`)qs(WcPAa50sD@M<6j&l2x_Kp<%Ig0fCKtywNl}F!;|v
zB`iN`=wZyu5}_jHFbyxeG&DD-nXl4KWg9@UAgd2#($`=(^6WCfg=lDJ%RT?WHQ2;s
z(-Jn$;>wX>Ubw|J0ZEQ3b7y|Co2d9}c<zD<pGn}yE#@i8Cz>2g=CNPn8S|GcIJrOi
za*I0vf3NVw-rHWcK0KbxhbP|X&OU8-cW!;>VQK+*uvkm3^d4&cBapX=<8G9c1#h^O
zeGM#2E8#L3PfH6Ri+lS*h2sVijEn%iV!&kgVa0!Y0SH9V&Hh0t6W}0xf;mHq)SodT
z#5ShB02q8+lQ*(H+2jVloBit6(b`Xmh}rzS4zEf0*~t$7cR=pb9L(FTh1}?GiN^;)
zxcqpXocA~4J=^bT!1rHB@(Oqw9`%jxLF!&LV|kQ)HB^&l_Z_ksOvaA2a04^n(f70|
z$ls_-HH{a&r?c5k5@*nfNZPAv7I=PXiu-0eeW$Fn_OccBZk?J9bmIQfp5J>{8n3OW
z1hXz8@OV?lnf@+|rERF3Px!GkY{SL)oDk}oZW6o^u<@a$!x)A2l}Wm$`z&{Fbhgz0
zg`z{2fDmniwe`m^qOvRtZMOR!F&}e>*20LF_QO+h46fD6!&639v0UA3FlzAyN&!97
zcAOZed10-k7=qxHROAtlSACX!=lq0A2e(0?tw}r@W0S4b*0~N24Nt0Gx&%GxwMHRK
z*yC)nZjOoB`sCqRH|@-?pipZifHf-nw64d;*C<SUw3zIilP#l_q?Qm0BiuSmYJPQA
z&}(aZOQ>RfJ__=m`4^-8Z+i(+{?O58C>XW&X(xm31t0%}MX1l8XSz(}0lTVGK$lak
za32f}u}%4HbztGL-Q#?L8p4f~?p##69talY3!sEeqQ!DTi&W&<@Z|*DXEND%&uP!a
z>MVe~`XdMo>TmO0Ds@mx5V|hL#)_suIF80K_5p0N2D%r<8C}?dxcw`e|Mmjl!}zSw
zW=Pj+;Qz44a8}AVG|MK(WMb?Bz$Ogokh+7`$W+pDGeJ|>QAWD%(+fqGh-4MUQ8$1!
zWtq5I+!QcDiz~UHG`%p5gxH;auzMnwO{a{Aa({ExhM$lqtWVjCb=2ro;8`Jx2Ahla
z)~o}liiuNQ&xIdWX)T*Ab;&w*vfnLVVPmc+_4@2*c|7;SKUM)oQr_~8#E{5GV@j~L
zKnIxXG6Zf3>!qJDk_DEY2tabQQ<!ez-FaMbc$j&DmZR8bzDxCHR1}+w<?2xYjJv)4
z1oZ-7tWR)-eyhzb6Zua#4-l%K<{GVL9hXwegy9wso}lai3qQW5cWC`0L-Sv!Zw=a9
z%kmPK&eFH~!me=wG%Hqw3(oH_r>0o9g>)J%<I9)L3t3?2KbE$aB>!HNWq#xPr3-pe
zvMS4C-HRb_ZKK=V)U=e!aJ7#ND^S3}lx!(*$YMe0VHeXIaG6F5=CPa-QD6TNL?ke4
zU7raj)uFT}&g-aH(4Ee1@==Q73jDGGp5|KAGikAeknst8x#$HUf}lyzR+Vw8jP&A8
zLVbiYN4!4~v$K34oF>505MiFToPJwPo-L6Ii25FxdLa>3OW<PrHCu|MpcZ$5IedY}
z(D%y|xH!X}B8gHPX@`RHiZ*+^2ib(R(G8r2=*RVV4xnVEX`faZkGiK}fg0u~uyqrO
zhH@h)HRyn+3)q)V2IxFTJp&P=dcJ31%%k)-z2Bx=-dq9>hGoAi_OpGeb$H^gr#y@+
zA!gVg72@ihNKHCEnF4FsO1Q+IH%qu~%9br42YjS7ZCX(9_^svVvR{lY*4Z65MY5G%
zzn>RA^`DWFRTp&Om24Bg4ZVXx6TkHBp+I2@pCMR>1)ScCzoCL%BnWAgC#Uyex8aSe
zES29rx|FWiAfJ+BV0Pk?2@Rwj>^rd<2}H6}4RyV*ZTTU!MSO>u<O!(bG<&ZnqK)G5
zajMAeDErzmtLc051YX^v);N@4>4vwpDVOnobX%gFcLL6xoFf!gtJ9JKS8$1{NJw^2
zlQ10~`4e~rTj`(oA;O>Vd~#8d-c8@byym~Mvl*0v4=ACYqyw~=jQj6V`EfRN^0RUU
zw;qFgI*t1g0+iNRaY96!+JUtsKH<YZ{A7{?w8%H?ubRRrXrY{rjafRezObm!4`S#G
zz?!fqH1~H;_;@UhS~X8SlP!;O%klOAR^INpJHLgmw8{fLHgeR*4^OOpj`YkE+&2wB
zRYUwsRhf#hc4y<^`-U5s@4YY?<biF|gNxeH(koTggo0M5^bbYoQ90C2N%+cj0J?-t
z@#fZcaeq{;JK9Dv;SU?xa<zTm17%P17@=-77~6b(_L;upv%2Ey?IJ0a4<)}+oCN=(
zr<Xt6`Q#vKhXDzH@rB;oUz}LkCCc|JsMw@hs^32--ikDbL)Ad1r{mo6Fpl<a#E0+u
zGaV^+wt>1$v`Q7Sc3`m_G5f9OYT3%(C3{3W%{E$*_zOE<>WLCeLQpB-W5QJQIOmQ>
z6Z9kZg#>2uc~pO}L!}8Cq#hy-xc4+DiMv<lpy``4=nKWkmaV%o%Hy1<ov|vxZnA2$
zX?#kkhl0g={Swtz@(TE_^q#8;31Pjm6ka?T4)6N@R0SEV1~hp$POOT2s9-IARZmea
zuKN6kQ-Z1K$|YGyRG=Az(Wh^kl}G}Xb+kQ|B?ORUx?5wPNxl3rCzr@$$Lc)L+#%v<
zfXDZ5_crh}RiQn%8R(<c)v>c3VV_hzZ;lrgjW{HKwy`P8v;yG}*x462#eJAZ2$(Lt
z#ldgn2$Z!><<d|T*XaeE92kh9jj{Pb(rpNssv(DY>jL;ze4W5Gj4-<eFFtE^55)m5
zorW{=gNh3TPxhKi9?<>zXjA00GZ1RUo=0iw+f9q9h&&5tOIrD}_=1XsRqytp;K+aL
z4zR@n+3L#WPWLlf!O~2om74(0d{{I@3^ogu%4uoEwHz6}4D$$y9N0AL3<6~}j0}nT
z$ScB$3|QHp7W&zpKT*W={t)WB(V|v1x(ZZw1r6i(dULR^vwax~9$x8ng<R>&6%7_w
zMqW!(W0i9mPM{8T+CDUz+W=WlzCKTbp%?UTa)!wfD$*I`cCREcUpeR%j39%IyiEq8
z(FYi%C2v6aeC&B6uHP&(mfwB$q8fna)F+Tigny5~I>eTYRbj*+w}4h0+?ZBq&XDB^
zBlfhgO~Gf;qzD=gByW%a_mxi~Bw8ZRia&9f$gQiUeP)$|os_4^xE82flv6=SEK}3d
zR@X}-IeG#F`WZ5afoYqxa$&gPX$nvfF=x8bsK)Fx=ZK?W?gSH48{yU63&=0<tZd3T
zqH;%?7ag#navkb9sB#@R=KUk_p@*D-vYi^e;RKRNiNjwI5?hgG``m!f?Z7Dh9%9Bh
zy@L@d9TR$o?7&FstcR)fCOTDt|MZ=apQGVdzx5p*>l)sPOQeV&QyZ7YZ;f<1bU*k0
z;GP{O-bq)d@n@(MVge9ellsiV8&C8cwk)ZU-ASZ9j4c=*4R!Tm`3&>&?%vxzp9p20
zq8HzbMb_+Yy!ObW#k>z;;q~9!!|8~t2-joT%u&odKVF>CB0O^0`GR@RJ1`g`$b@OE
z35$c>CwHAxPXnd<5Ll(&$wdv1Bd}_~kE|U0C^cleX0c16S#_FxU`KeCFmxCYyG8f(
zurM>g31)&z#{xFGfLL=QQUWcOXoo)h$DN+{AHjiO0P!6*PCtv2Htq=WP%Cit+;SMY
zpHT&XHM$%)PZyV@&MLwKltrBD4Ri4!qRxN>7Wv(s<qmqf7MtVqUPriNUQaSQ+d^8e
zC<DymtY*)^!@{d|eKI+dfSRctP;ZkV{LbE3#4GX2;LG#YB@f&zwm!h}b$uDf-VePk
zJ|>YCIj^&IYUO(uLxqpe4S{~qnmLj;+D88`Eb_4OzBf7yqZXk!YS;ZK$)=iqw@xMd
zY-ng`=k9gNggM|NY<7CVHq2O~<Z04rTcgB7a)|DHxtGk!H<*!mO<V@m;`a}_DY_4b
zv^5Di4gwxqHLB9D`9xM$UL#X^AnJ7aRw<zW3aULV5d#STNi<_k^PWNVE+v}0w{y?6
zd$Zq8+uW`nblsIRe8GOrSBr+fx@ra*PRwT+=q~6t{ExRD^l2vp4xixd){F%w8_zU9
z&UuPHyjCi-bwT@$1|sS(EF})mKb*6w2>86R>&9$53cb3YJ)aTo9oX?OJgvOL__cN3
zhl1K&|33Tqe_3}<QiM)0ZY4<q<^F_*p{x?kgF{gNsluHUSt)Fwp^;L)46~;cZ#@qb
zcnfQO*FxB2q4-C-Zyc`&1+f$=XrO$oBp3+>wb%O8e)k?FUyQQqJUOK%&?60|?~4S|
z1LC}B_B$RmY{1JK+Q??#6EA5pw=(LT#Ntv8>~kmzaXCYT*u0`?L|fJ);B~Ertq;T+
z!(c|5%bD&oN0O+iALZI2-AY#hbf|RcdzhjR{~ud#8I@(!f9pzjcXxLy-HkkScPbsy
zA>Aq6-67o|C@mr-0@97r-DmOcvHxeEv)@mQfevx--0NPynDd(7%#vI9qu1Zyyqdj&
z;?qcyC{0ERqUM-G+s0cs+a8zs5$<T+TK@=6ao8olBCf728gs-8{qxtG{U0uIg|b7^
zoaSfpBWF@%C-=vF0`>%ubxUfk5#F(4h7?1YgJ_6hh$49&jCHI+$pH`Q`%ly%$QDIJ
zn%jl62$x-u@F#?WRIhl#%!MbsXdj3^9#P&d1d2~x@z>CLm_WDIQZ1F1`1po*HJeRi
z4o{|mKfVMXmcJke>pI)T4;?M$sf40h@qKkWup6eXV2(jO3`cUWsrbZv_N!xE51mir
z;#{^WW7J(`s$OzX%MLPWGi^WOqGC!~@y%TLKxYr_Lh25($EM;7hM8fT!m-;^^Ac0@
zHHB|~JBrR8p;@_1aQzr`B%uC!xds<i=IJ5sRdITCFsGD4%0j1t-VqbZIgRu$@*3}7
zRxp<vqf{RZobJUt*>EGY?}RHXnX^o&(Vk)#dJqJG^M~L@J`Wi}(7bZ?fcb94W8F_4
z1yXOUqR$DhM7tBkd)f*L$i}>a((dBpMd6;6zH(lHC3&uo!_WLM?pU3l?6G-H)=SzI
z!X+<So<u)H#y9ZCGthKT^6za*ej>pwh&O3-KGClUVJ`^??vAO#G%?>zj<vsA=9udu
zX}{4O)Ujk@lVV9TK<zzkcpt*}M#J$<f(q5VKsD;uva9wK`%7ZNI~ZU3{Y1}6v2e4Z
z%af<ea?Yw2^{KZQlB)X|xIoAC#O>{Qs5zd+cN%Dk#6zwat%!s?>ccGjyje7FJky6V
zUu(B+G&p_JV<PrT&`)>W3B@=-%w6O8>8&;rTY(KZnl=T-_oEB}mXl`9X9`i}hq$@n
zRVX!Mj06g!ZZ|0^D=WN{BQ&_~dk9q@N2k}YIEMduTsov}`g4EC;lokpSeKHww@%FO
z*;S?DZmZN`6b>)1BiiPNzqh*vw>@HUKytOJt<?Aj0)O;tqLTSn_!z=QEt`S9*Dj6F
zvFK(&;FH3pcZvJmT8L)nu_!{|e3Qm<XB3YWs!<RciQmKF@Wge@Cr?i15_Ga0;qrB`
zQi1TrC79%Kd@>ex#%o!+;4<p-7xd3x;$}1hEUcV-w$eWc$7Nr8gr*8bbAdEgcg%NR
zkJ<2wsw3boNZ25dVbVls<a}4#^|>)G2YAZ5(|0j(0{0<pttlcq`mVKiIq7~8?W?x~
zR?8l$kd^<sO$|h({jsErXwL;@)11o0a#~RmhdC8yT!UNw_8~@CqGV9Jsu0!EJds-Z
z>O0S5#gcXO!$3zva)n|5R@hgM+ERxaQ`+y?B|P?AcHz8a8LwIqhpq~XynfSr0AgEE
zfAJm0#?Qa?eU9yq9gJ$qI~HA5NkOEYARaPFC$nQ2>=qu>wneEV2*L(RIO#vNgtHCn
z1?l{)$qO-kiE%A7c`avuI|i;u)51kI`Iz6>Ms{HQL#I9dOAuvJFIT<oLuX_&nmLSo
zx^?WF2GSOHZIHAD?|L0tcExTx!jxwaq2(`KuB6z=v@guK5;OMXqxk40)=3T2d<_Fd
zBjZ2O*7cAK_b?>osj*5BPzl9>hnbsJA(}bpus{#WV5<~6TN=fzIVf(C5dWR2(7A05
zkf0o-Z5Oo{-m$6O;LM@<!ysA};s=<<_*@9-^K_QgbStO|pN8_cSGut>_cy6o3VHlP
z6l%p8WI+)qeu9(kldlB^-R{dhT(jmE>g%4ddc5pBY0SLO)juMw*014cG=IjaSz_)9
zw~Vwyf8ilmAg~^p(z)cCFd`)4HOOPltJ&$&4l?iR+jz5!f5c0R((<7p{<`iRFQZJ&
z)AbWIMnoq3>f?2x+mAPLTMm2hu8b=#ogRnGs3E)(=qj=>9b_o3(&2*<l%{g#x|W*6
z*K!~=Ipg2wWQLBvS?uI2&u5{xx@I?EBsc@=)X@Jm&0y?9you*5Rx*tqXL9B{wyZVQ
zApD(-|IqusR$Yq_J-(MPvaI(Hsq14!Ap4ueW(GTLxUa%3<SWa8?{a6)Q+8@Yj|Su6
zQQi~{@?jX*^gV04oGpqGNkwRE%jFeS)w7q){#b5V8QeuSIi_igr8F^iY%mah@_ND*
z6RVl48m95ilqCXUjC&58{ya4|Pxm9)O0vOYQ7g^vcB}Dh(~7O7!I;afWqdaA6&0%u
z<0;->N6VsFVyjYLBZMm7%|KTflYP@+2t;w_L7GzP@?g@O3y{|3&o?hLq~D^;TE}x8
zM#oI@XBFNoOrkLYuEkG)l3b9*cMj(6lEfm5gBu)<`n{wdE@Wuc&B}U#v-ZY_I<D{8
zXYWf8rnDqaRO(mGrM7$Bk}qCRh>;D~D|WlLp_+n^7e%f5`kX*{b`rcN8IXW!(q=uu
z${Wy~dyIkbX<){ioG!QTQ1Hf}HWxtXmK(GBLj|1Fk59)rI64u4uev1L$>UrAk0VBi
z@nK?6nfVm0I59Pi^ia0v*(=p?<uuW3gHWI&qaa(%hJO2xuaH$jcXT}JzLXsNMFC9+
zdtvd{`VtG}lZikN_zqDM+3=Wl-Zv!}y=m%M2Yo2-f=4y}$M*I=JrCt5mVCDWFH}Ld
zOwE^6&0uQ8?kQsV#BQq=c0o;309&tj!*Krs`_9<dxZu+rg(W_X6|IqJ5KKxi=O`<8
zB8Sq9*OIB!0L_RW^+svt8Ia8_K;98<*<fOub20MC7PrF=iX}DNv=*KfGx{lXNifYv
z>XNYo7}W#hCI}-ss)2EQBkSy{ec+!|ivgy?{l$K9%<$4=Rh6Y6Mi9)1MZ0Veym^Q`
z)EuiG(sf&;=I_41=P&ljXJR9?Xe9ix?f+~kfY=IiXsOXE=4X$v(9>nQ7dDpVpe^dD
zG%Q=9pZI$PSeT^m@;#!4x3PEg4-XHYQD{$;pMcsbOfQy1;O(B{uTx+Z#fL|mb3&rY
zDTs%J#+8Xd8Fs_<VDd2vXR8(C%38}6+eM#@x*EwmqTE4arWA7(yU=OYHjK->`wiw8
zUXrd6>@5ERKIX!jb5(G})HK%%%$CK?@)Y97=3*Cd8TzuOU=!g7ng%8FTEH@ERY|Lg
zJvO_%f}yyOl5KwmDdZ`+xel03Bey`6VGdWV8fGSb!B4D#L<9S}$WV8SQoknLG(q&v
zR(x;kHs&-e@*&<6mDM?J=x5tUR)gel&kg4;y~wj+0n}!dB(rqXcVOA}B`WOWaYjGS
zAlXtI*J}TxBqduGTsgiRv!2~NR``cn0uDnw_+PJmTF*V9`*mlgT0`h9_cX4ZNW!4h
zf7nvHUkeo(OS;L%+fmJ4<^>pLkdl&-v6s}i$V{@>$<k~2Z0L;<3J!2~d){l0_1XzD
z1se&V(uAF!*xySL4g~0rilT&NDh{_(B?`qwng32o>>IXRrZOhQZ~H#RNL8=1Zj@@+
z>|U9dI$!xIb+uvIs||(%2$+g^CRqzKRN1x$jo4QT`^SZz7$v91s`y>{^rcr?Qq)w6
z$MNfZt99;gd~+f^w_MKA*&syhTe%*{Uf08tD|$Z_FvR|q4aSHxTg8xnm3I|6<Yi$w
zDLbQ97fBLKb{{j)ZJsjwYWR&9Uoxa@G#an6o24jZP+PIhe|v`+P^90k5dFRRGxDxn
z5120OyKiPBz!K&B!C3H9AK7eMeeSA+Vw<1j7ua}8s>!@V?a_ml_jnc8H@?T-QI`SO
zKe9@k*ygzPfR38msk2n`JSFqLI2x=%#0PM^xNdKO75(66f2de1;gC}^{G1d`DL+k%
zLZI>A`Fa9oOh8oTWj?DAGkV~+1>2q4W-UUSA2~$2H5<<m-srHB*wBJ3Ms=#CE>EWs
zhlv;zvy*aA;DsX(!K+2*79#uI?UxheBRnb4!mR-8fppEuemzR$N6KK^L$IR%DJbJd
z?$8C^`SqL}m?IMzjHD^9K$=<|<Cyj(#$mN3ymx0>iDOQXVeXCmM1!GSg+UB!(h-K*
zt>VB$<b`#~cgHeRN;h#PoJoF0t75Uxfm0CkYfA4E6#=Dy9-~>rRe?&(gcD&W2BE=A
zve0jHB;$h`GeIa0Kw-Yum8YB_gfdhWP5^%mCHe$T4aFi>7j@FLxK(>`v%>Z-)q%<9
zif@sqgzOZUfm3b5K)G;0h_c0L+F&=&j@-8F+ul5J_YCw}!u`4u1l^v;D>aO?+7UBw
z{LY-62}<`qVz*nLI^kp(U4ga~F$Ug(i!~W+kUy~_U1)LiTqOT65+s#-6$33T6iA9;
z_D~Pl%i+=sG;-gnjGtydH-(Ly)R{F`$lw}ph1*mmHHzT>jlzeDoK~O}C_cqGA`>Pl
z;)-3C$m3wLn<+zc;#%UU+A36bDmfvobC5|bjHLxLC0btS=pgKB?9Kbg>-^;bA3I<u
z8JXH3HPfB%bP7ANCiTC+Mh;ncpHJ;h@TN0xRT@AAWQ;AHyuyDjAKPO2-3ObLAa@_B
z^g*t0Hyd(#TzIN3ezP|IwtuRbJCzDEpx4-`I309c6szN_W^y{x(BG|?Tn>DokF|56
zH;!aOWhKKwhHA!|A5)#<wu*1hv}c%mTX8m6%?Zg4bFKU3b&fQ5Fwe2Zl^)}H2qSU%
zT0xCO^yH6x&^4&0Id!L>b|=Z*k2<throKwn>g#!@^PwEaT5jiFnlspF*XT+=xp#aw
z%WbBVg`5miTmw*@xvP7dqc*;=yp37P=Hy4j<aj2=l<NDd57@Io2<XjIZj+^-5mkr}
z|78AHMN;<;>dJ-lGIzq^!sC{<L~Qgs4ND*2x1?KxCKmWnn>@GK-byEZH07-~A_r5&
z9>?yA+rX-)r}=Kju(8BO8p}X7>Hj~p1A8#Epl8uATE~WTRy}G~Qu^=I3(&?Hh(hLn
zjg<er*D`51WFOIxWp4U?L@*4pd2Z;uyL`msm{QKN)rgvpWTjyph+0lJywN3JrKrf@
zA9`#Dy$b^S*Q4J1BXOHcv02r)*;)4>K6F`Tw=C1}@(fG@rf)&g>5TBnuFd=)7}n=%
z!A5IPso9`e?_onUa<0gstwL2+7G5HX=jV4vRaR<V3AQUnj;SZm`NDkKa;VWxO7DYS
zfTq_DhCV?OC6BrzOMD%h(%Aft_Q3o4XxXAFWJ-H7$~`az|4P>K8F2EkBHLT+<U|`>
z!HgnJ`?PqBA?*o|mRPt+Vq-?X{mKnM@Xzs`x+fSvl|tXwHcg5eWNUugrR3l6rlnTi
z$Rru3E{sI^WQp|y5LYVhaVoCNW5aS$q3hoPLAahSBO5(!fg;kNOzdW}cGf1N!%9+4
zabS>n?JuwldSSdH-lEP7o#vCU>7L)GUmOF?w0F$RdX>KY8lRlMd5i5ZMZ->Y90#!w
zCF+oF;SkxPd+ZMo)KVF5(fmXxpqeMF_m_3@S&A}^Km4fzf1IhXZ<;ty3H<CCK?3lv
z5`U#0($yS#BoZG4s-)F9kv`7&!JBIUh(AYAxiMk=g~zs2sa6cHlEFz;cD9*|9YgS<
z@6-@18fj=|D~Y&j0zz%0X<LIAr$7)63e@1}K^C-z`JddeCc6kxttWi*+9H#b%;_>R
z;M#@|tb8qjm^h;AHC>=IynL>`QV4wPGUdPC^MxK!4};DsGDoFDPn~qY8JQ!N$8H7Y
zes7PgII7JSTQ<nmWt+i-w&Ez76$REfKmE+uPj*YiyIxfForq5a9SsT%!fR%@+pQ-n
z|BvIReMNu$|L8m^X~K&0PWo>^wRpAqTr&CZok+92SKjX2B`4@Pi_Ytgk)~Sfd7qlW
zOT<1?<<VeYo~#xmBV*58iS=y=_&K|EQ4uP~N-!CR!(&*qz4WktS0^J`7%E}aRfIoj
zRb%J6gr~wmSPzbgY;Yepgn61Q6@btU?R5Xo=kME77LPta#gHEmBzDsA{Dj;2mtClD
zHTB;m=K$EvhgpAjJ{b^?Ss&0SZnxS$I%T}r$u?d&9jdLeO}Emcp6F6q@~aXn5$Mnk
zG;9CDIL(Xp)s3cl1SRy}B?%FH68V48_S{NRt|N8JE?MV;{h3P#HhP5Aa;ss<84BDt
zC>Z$A?+DRvUkiHtBjh%u&C@y!lQk_>5DwZXlqbU&j3&5(sNxbiC;W<?=FKLbO_bKo
z`W<t2-4&#aZ$~EKD0n%mF$YsJ>!sIxh`)uX{IL&Yr%1>INzxz=6}I*QFxu)~e&7>Q
zGy=3P7~^QV`X?gfn|4`L*o1glTkfGqOUD6y0EvtO@wM*K+oDy9Qg5m(UM#11ZxDKL
z>HP}NlaW4+f#vU2$W_Fk(CZB=%4119%@U7@0E())>2HYNELdLw%!OaG!Dx(<Ebig?
z2xHH*hurg7KLW*Gqfw_jlSIQ+WqIBPQuohji_Q_GjaQQ7@xW{{Tq_;r+yh(`l_Uu_
zEQATZaPNpN`&a$1jeZX*JT>On_UTgMvpiAiMGu7)?S@mB#HzH}FKmdR-Ng+Ty`hZ^
z!_?Jna$X-OxNr%Evzw;_dWx!kYLeH^lvWKJ5LWoc3lO|h!u~ssj!f;3F1{8uP`oFF
zy*DU2nSSw9106LX=7*l=Bl^Kz8kB3Rq%5}yX4mJNTy}lMPdl{gWhxGVP}_TwO)$I1
z#?K<9#$HJy0wvkzl!XY5l|*jOtw~_yAb@&cvAqG=(v)fuY#ZmC5XZpuPgrrm0niLY
zclIZzO2e?K?>=Gnb5c0TO|q+s$T84c962VWZ8sP^xCtVD7+T)mYRX$IyVrIZhZPJN
zR205FGciq{0+SJf9&}7-k5>aL;eKTKiRt}g8rAUnMCX6+fFW{**yO%xDLnX<xfy>Y
zb5gPVpCxoX<^IfMl{`mx!D=SEfR38hM*`LpPVe-4k<&xm@?>&V`23s)Sfl6q{0f7X
zs%<av+j?cpj{}E%^RGV~7QQ7zVbfh(JMR0w!6s$5B7Al#L4=<Umx8MefAw>=l3W|<
z?c1>3>iZqYP-H{OhjF8=g<FP>-lt?kyS@z>-tG`q6zai=>FLbwd)_yT!%jcmI9h!|
z`Y+WUUAVX;j#q=x2}l<RO-=Z~&CjV;p(v}ghNjFw{#@v1CvA$y1Up2dszxkpWeaxJ
zvj!D*=UnRN-$Jj0U=TEeBlzD|xK(5iDnesB%)|@hgndC79Mzpek}@N#rNHHtpv20|
z^Z=$76S5$JcL1UmESq56f&+t7dF-)2Et7lm>J6f$6=M<{G47NkmKSiFY<^ig2rR)l
z#P8_*HukfBFOSEdM7mGx`!5>ubzp<~f&&u=<b#3pbO?CIN*q0!Mx@Xi0#C&ggHkT4
z>3U3Wa$LulxSZVbZ4ZZsKTSbke^mwEQ}NAs-hR!C>^mim7{n1{{rBYvky<IuE}9Iv
zfUwd(pz!T@8e1_~ew<y!5}ayB`6D6E72<bBIy0&|wwF<|MAz`vw@tM#sHzjBa;OU=
zwkDCMRKuWId^QQV4Qo>sVfd-{WCMG$5<HmIK$x=F_jqMsgk30~wn#?#kn|F?4vr}}
z+1Bo=kpuL>*-82hQ;rBpX+)O0$)CaXHJHM>m^LEW!lvf6|FZF;6@%`5ShMNslLcaq
zq6fHhMvI@2F+%6m{licG!;SmDVLWd?A>D;d*NuR}iN3<mCo{IFMUs9>R%%Uao2Mn}
zv?H0O1Q{XhVw<%1UzctVe<)M8c($alHd}Y1oTW9MSAxjvOyNF9Ogf1Nv-m3ETj&K<
z9=Z65&O`FEhjW3B-aE<kJl{=K&66YOvGxt-+<ZNRc>$I|2@MUs<!j}Je0LRHSn-v~
zz*2dC!4@3~*XF3zjbZ7^i=jACZ~#a%jWw<Fpx55KLg7GdIrU9UB@oj0o@YQgF8a0_
z<bC?3-(w>{wwk`zr$vVjxA(C%9T{2r&EPPa^m3U>%-HW5qJWsEd{;$REY8gk*|QRE
z@5bzadp0lQLQf}_iR|>TbBknCveDFsJsJ#PAk6k$M;_uy=9LVy@`uOQ3$~35tI}zM
z3`WaTEoSnQTO2OF%@b=&a(Wg$OG<U^$xJCG*K_}^k@@uDyiVwdcuP%^BSpRqACm6L
zUMzKtm)Ii9)|ltctyitR`OL?3S!BGm+?xt)G}STrlBs+-`K7K}RTo?5DX!Y%&Jv}4
z|F_o^@&i-e7Mi0K13pT}l2U5W&b-%3>2y5N_TF4g&)g+M66Xx+)&)4WU<b`;xYby1
zrHKiUyQ#q-(ddtJ{r7tfGi-xqs{`Qss080Dq9*Lrs?ucIQShNjWuW0R|2mM8C`X;>
zF3t$11bwp}D3UOXhaj-focfKF8s?TP<^i<B?b$XJQa9zzd$baILfl*N&wsiz#d2T?
z<)Oh|7TFo;iG}(^@)QSl$Wow!e8@a1An==doBU?*tp7M$Tc~~h-mtSUuy-5`ATYK_
zI1tMTBMB#^iDaELw32P<{W?SlCA2S$+a<YUw>Rhse15WBOLQR&lJ+7G8bKUj$M?Oh
zXG(#5S{9#$JvF%ymR#-@ri$9le<ngqehJx`>xGNajXi3y9ERH-7KsVKyXG3AXiq4t
zGx=rB>o{}+wOQuX+hqE^l&@P!9=aYMZ-dQcP{?jiinyzU`&7oV9U~4^NIsG^5BW@f
zHF7OF0Mh}T`FzxLahdnQs!pdq@7YPY?^IFFAX0Zphu1;L0N(q>)}Mj_n_|c{h1zn?
zmBev9I?U*_a-Q<$0$%a(PcEJo2_nnUraSVAf&8u04~se7eqZ!JH-%#X2q8iIJ0l<U
zmB5nzFXBH^S>#<r<nbeJG6@O!_xk#&i6;ZAiX_Zn-inKha$gCaaj)~GbO<<=q*$||
zhUn_{#Df?Fi`N;fC9l7-Y;E^mpn~#lg!51l<A<ra3X4==ErMnWiwJKBx{){1?bTED
zW!awvjoU>J3%=DXK`|cmA5PjxA#e0s+0(~r2gcH9<PxvFB|}*=^{U1!-c)7jr&<i-
z0%t1&hcd4T&q`waVoTGRVUU)Vh{Nj2SQk?iQi>vf^)Yb<T{U%|7=GnP(d(RuR!6iW
z3e+N*t?v&3shWR)g~}7A;f&oyo4n0HGbt((J2_|^HwZwsre{<s(p_Xp{QfoI(ibjn
zx5TcEJv?<K^yPQi-kz<!$wy2^M2;>TT(B|Ff)2#OS{W14)YCkxo1j0y`CLFC)X_r4
z8t=^hoPnI)r?lc=5IlX!gfF;ODI19Yt8^8GWSRmF&DKpq3NRj1&irIhg4ydH=&Tl^
zGoE^3l;L-1R(9GyvuL&Nnl)-ylOLB@`TUn!4x2LZXOmB1Bp)3g7J=6UI)x~y7x87_
ztOXbtbPu?9`IxE`JWB|^@tM6TR-YvrqAJ3=dU3X{Gro&~_oewTEI)>fw<_JL%MeXd
zxdf`UYP&YuVajikhirN9u9UN+Eyx~}OT<{11<|bPgy%S3YQ4Z@@fWe1a+Ib{J_)TV
z4opPmwa}QWw)@R{<gg!~7!U1Uq^|>MS4Fi)Emm*b_b_{aFyjD9tfDo@8R5(aSYGw$
zXfK!p9ehgt$?cXTyp@U{w?1d#+%Y}#%s+RBsdL|f1YN0u<~%t#){v0DgkCBx@pj0`
zS@Na)0WGXuDLX%Ml|b|i9gVqx>1eCt@)%Bn!Mm$}3-3)_fm>3e-ngeiHD9z!-^uj1
z<qbxo$N>$p0EMs@Csnm6IAO0#dGD@mhbEAaEOSt_^dD1Ne|zo3prVX#ypx4BzKCg0
z*l4(+&)eoxv;$G6lJ#Ca$5Q*+%`6d|8lU1JU)ke)mn>Z@Il;IV%x4j0)-H8a-uPrF
z<ju@})jy}DKmBF1-L`01tDfp7myme;6Uz0uJ3a+-_5Jic8ufkU<%7aOfJT=IHSGzt
z%I-W1Q>~q<Q4R#Xa@R<_h(qp&R=)bL$m}q+2C@%VTP|A*R}nwOeqvpgvecowPd~O<
zS|%qG7$S$w`Ag~b-^VaI{ZRf&sb_I)+USNM6-a}@@`;P(kB@+rDseRmbgWyMh<qwV
zcIi3F?wT2PvePQL8K+!Vyrk!TMmyq~v0Lq(A)D!rN?US^k3GsMPD5F-Y{!iehp-EN
z$^1%>LpNkjHdl}c$D`%Lg-hT&%hiXJw;>0(C!P+)m`PppRWk1DVf<$}9un;Cucf62
zMrP8u#^8gB<?uR>1a(gDFq#L$P{<zQ=o@z9#^Bjp>bw|o`6r~VSN`sJ6S@%c)MZrt
z{XM0)+_|t9j8fx==EbG%<9?l9>p?cw9?<OLrOP&dyN~S}&>HJW@%ycIjAV;CdeyKu
zUi>7QHy+EQ_Rq+G5eJ1@9)T<-S532mo$NO|{EA&Ae%~cz;DY}%>G_{Y`u}2<KtQ!O
z?sGq4d;a{dY0?(xXbyh;fglNauwD$WOWQ|G{Sq<$5Jms&h53C$>nT|gp{kM24AT1u
zmi3x^V)jhW;KoVo#O7^<p2zlbITqK#f|aY42zuKU$4kEl(>5Bmv7htCEx1T>oj7_P
znA0T^h2A=^Mvs^#f2j^p6y$`4#06}|3C!01n<}H``eX6h`g?{11Os8paqUacOHLjG
zEfuZWy^d)+2)VfQ1}Jv1nO^@>9kIx*&OCWDr2}u+fdjkxp0cYMUC1e@15xg`&*DFT
zyL4`s8^2VVn|>f33gj|rCX9Y!s8$?>fky?pA$Ya!IkC+TXg>qljk|Sw5(00x2^3|y
z-_y=B8n?rhY5clHNTW-7VjvD8n7sGO0-cNZN9i0?SmW5s28fg;;r#H|0(alDQHJ3U
zRH}L{wO(sfGcUp{Yz!#IU0L(qN8iywIa+Hgt9e#>nDE9Fo~*=Nj1TuVDqjFKk(&NQ
zlIgdHkJQ=GZ~T-jn{)cHOukg+*lgW@b-4KI{r664KmIX9-t-ZL9I$?Fvex26b*$9X
zlUoZLs88~K`lC2pwqmoT0{Cym-~Y;z+(n$rZ?d(XFEuk<cRGEL$OUWJfz*6;;%Zho
zb8uM@!>j&XSo*2zLrV^gTz#~|okU^{I7Fprs45WEZsJgmsq;x+FSD-GxA=hCrk6V}
z>wmvk;q7L30}DsxM}_OvcT2}(ea4O}F2Q)TFeAI%Rh*-4q~lDKya?#fM%@;cSfhRb
z>j4IpA$`lTa>Od{DT`Piz<!IE|GSJy08#D%`w>^-_5Ed2WuI2{5QVRAh^(cPGm2Gw
zZ`jE{&wyTKk>;^({+NU;n}<SjwE-+{VtgwrR1$@E=_W+CZxN_8((?YxYo41r@h9v`
zW8RkpOKwu;LP{)<4jLyeD(|60|L;+kVVu9==}h+Apu=_Bwuoq`KlJwD_Wa%5#eP+1
zq1hMplfC(RiWd<`I!-Gh0cbQHFEosPi(eXfnX|YgvPwYEc^U7CvAkfuu8XyDgYehN
zUvHQuRYQGCnmJH$fA`)Uu)OekML<o|JzmJOL2Y{P+1L4Ig#X3Kc??&Q%a~^CfB-wl
zkmn_-C61qTgag-hrjK5MZf%L{Kt)h5Xx@)A1VO=PrA&mBni+g)np-9wiYHS7mUTzX
zMDbd}Yvg#K%bx+GC6y2ajCkf+FWCYAm@p|uf|S!w%gHW(a{gsxGX?62bj9L&ZK|xn
zG4aJ+@3G-x=8ko$WOaxlT9ou==jD;>Ky%OfZB~9W(RLVTt>eylQBPPR;zkp2f2+oa
zoYb8C^J~<u#pfTr=+&uE{Ji+F&Z&<c1IOi1c_i-J^gU|9pK^+MnbG`bVejXKiRdro
zNiR@QC9wj?(#l^D4#c>>GvOc2{3oD(#Nb1dcAcc?3b)WtzlcG_+vp`35bhx>s~*H;
z(^3@ud@XOPuJ%~}+69@6c;K6UecrbWKt)qLoSnG1`bX)9-+i6-nG`6hM~9dvre7ND
zd1K1eBcu00O-G&itVq2|h9x78fA)%za4ytyY;vvEJvD_W_}~Ae6r6T6)x*`M6KEuL
zN=wp_zmCRJpu-E%Mcr^5j6P++4QKs$!O91y%V$5coqVF^Up4MY0{{7^-$G@n#GLdz
zcPCN_RmR^*2BhsNzhY20mKRDLYS!xSFI!yQDer94xNW^YJ9<p#+;KTiI=|F+#ZCKi
zRC$H#dcqF(E**_}_%?Uk$b<q6g|Yq%^}`(2{FzxjNAuzN<E&;{nw_gyDbPqOXua;*
z$pR(%aj-3-`np5gDZ(aqt7U(0tQqs=zbV-bGs+(yd10)ovmHB`=+D^cKZDpZ1>ggY
zuC%qu^Da|No1<^^yLGMo<b!Tr|En1=oGn@$m-R_vMW4HEg*w5YgF~jo6f>q&0alXg
z7563>q82kB#8ZgnnX6d|Ij?<GgdO5vqVPMBsIXU+@0+XEnd#Wgt4=<DSw>`}=w<W1
zlLYF!SAi+J^=H%6l9a}i)Rvt9J#sKtP0(UH6>RK(IhQfgRi5J-Eg1g(XMu$DrNY)1
zhFo{-p7S3<q03|^!qH*8@W~g6R8kdSWzh@-$lTTZcl>_>D^zQ?jD$Czult@;^^6{f
z#$B2ykZw_6V^^po6ywVESn;(6O;W5Ig_l0_gpX|WydH)kVtCmSg8D)N>*HO`kDH~0
zp7l(@U3oI`jhef&C1$QcWkbgA>e20!-%U6tm9Ztvra3HRI*`Bn;bfdQsWKj?z5&qG
z0Qs{z(9Mkz`$ka0QKukP^~g@5*|iYgd65fvW01{u?Do7is1S|)2pOs7b$Dwydn%4S
z*6y0Y9{8G((#;OP00ri0H{*JmXKNWe4)Tk1X9j7Q8L%HN8ml~>x_9TMOD`$pDPz+a
z#>X|hbO>!4zFaYt>w;*C!)7xnVuwE&KUQbeHzZf!tmQn(_(P9$VPTDH^vJot$sYF|
z7d+n2c$XJwuDls%AMFa?qwgE;d%p9^`!;xvgTI2utdi_W(_O-2)2S!1G}t9CgWn*B
z!BT6OIrb<a_~}7IF}1szjS0WE<eEYnkHR<Wef?8>j%v45y_IswL<>9BU#1UHI=^j;
zL&bJoQUUcUh2Inra7$UWo;MJxc-`1v257X$;c}|s|Chu#2ZBi33|*hpNs<z-`2O$C
zmud>fB8;j$;Mx@u0gp@2l=8Ejh(6HaJHA8ZhMSeF)~c2TZ`4nnwVnH+ij<U}2`b-6
zkYQW4B)94!I@1}oTB=e=&w<jxSnEw}89M|njM)@a=UUj1^0&D+n^rPf3dgs*F{1o(
zpV&7p4wrPaOi^R|9G3+@2pxX$B_Ff?t#4$f56$raRPAMs<79;ROCLfwF@Fz>Pw{YY
zaBybafwlXSpa77&V^Y1EUeiz&DQz{w{9c%Wl)Mj&7}EVxorfzu*)NgDS_}T5)S&vB
z$Cw9UA{X_0jY7STXE0#)YJdqb1Q2J`@Oq|nZraEL|1}H`$f&*#tcCWb)`H(sNqk+b
z6<zZmcgUpBv2aVh*x#C_*#oL&u%={C*Ro(oS!6{Qe#Yl8uEBmNq<3n*^s7JxfnKGe
zeL<g|MG^5t^8$2*VO#uZl4UBoIiJ4q4f*@6hY?hHi3j0ke2g_o!+)KJ$ciw;PI#Q&
ztOZK=lBUFcxsM-{rXAnDwKOQn9z6re!vm&XHY^S?JKh!-KO(Ln6%E6Gdu9;Jr-$Ts
z82+Ocmqrwq(X#1=bS@6iZQX6lxdt@kAm5$<1}MSz5vccFMDuZFu@X|F1By=86Z(~n
ziiS*H_aBT>f&t_?w%U=>x93h!j0S<_lV_7-XR2%2iW4qvNmq{^O4avodL*B`P{a%r
zU(j`7llcPatouYMd=J*S`G+Zjf&#C-S{>;PT2O)?kVW-L-HrVQP()c&1LN&$EJ-=W
zyIr91&ODZT{e6RzVH@Ua4ytVK@yC3%9?*US|9!nyRnWYxU=Bc0zQkBK?QBGuDQ|YJ
zzOmcmmUi7*6Fi<L42<RQYZ)~u5wIL*;KO0DORqS#4XkboyWe!c@J_9Hi0ss2QioC3
zpHH^AzBzQ?PUWwHQ1x1&W_O@MtP$5p1~2Ek_L7zCd2QlD8XLD-1bn%=jl+f#JV6e$
z&v#k1ub2B4n{W#rPeT?ljK=TaT2I=x1da5kYjX`we)y7iK9X_tTqR4Q%{*@y!9LMb
zaKN9Dz4<1Dbh1HWOe!rJ#`^F$Jm6-Nrl_Tf=l6HCWmLP|QQD$mx+?$sf5XZN!(wNB
zA(8)$5{h0j4{%HTKP#1=R5Q5;alPk<42*MKhKQ;<4S@ix;@wOcb5ey`vFw1ts4czN
zCfcjd1X%Bl{Z3_4Es2gFEPYBNzJSydl_XCI(+FfnO{^(LiGZ(DpUtG6M&8fiy(T}z
zmu9ywB)63(7z+XDC1#KC;5vS>M4$HnB1Ag$i~zd=FcGtXRZqdM5I58l$VH-;Una2|
z&rf0yQTx^2o@dP`MObC3SB<0KGOHrLdG!F)0oaH=wrwx*jzQ8+LqCV=d+*){SJ_6O
zlYN(FYBPJmo&)t5kIY@QL0jFT6EKN&x35aBh7BT6&s@<Jaa!%;qfG(2f~;a<Lm5ei
z9Pe{sH_D|4&E?=1Uf>$g^HX`WV0<9P>cf7_g8#)f{j9sk2fFIZSKikE$f22^v}aQA
z+3u>-oW)z1LcspFsh5+qeyj7kiPt(@)@v8@cGH8jeqGEU-`6L=s3Tp%oP{{Qw3oa6
zbbnmlU@;tC+k^_~#pFE5-l&~L4X(zD5+jDLUwHbu_g?0_o2%0^5kbCf?|j1yppY?e
zt%$*OB(6YJ&<=-!`+UE%VY`s!tl56)=<2jmlI(3HRzr8(@4qGF^*NGr0IsRgJm>Ku
zyry}4I8z0*w;ZX`*eaP#vT?-&bjL}_`2?iPn}s>=Fw|2Q>S4Pg-oPZ`!dhN;j`VJp
zUH2t3D?gYt960PmW;5N#a}E6QL%2d&KD?a7?DTDYl&CxmLF$foTU1|f#Q3nzvMi9Q
zuKR`~&wAo}kz#_VjGPmN${3b5iq(hS3#bq#x8L#_3u~GFARyxQE@o6cQ!3(9YE;2=
zG7QvpJ;}(e!W6nPk-#5m%C&TcrQYS?t|Qze>X{j(aNLgfsXAJ3m%H@}?+*x_yq!wE
zWT%3FQ9h5(TWg^}nEAkScGY>QUox>@y1k8wg@6B`T?x%&$0O)PHIm5b+dwApzbTw|
zXUgJlImxLC>GV`p-l%J({(WmbU4kokEX;D*`)BZA9Swl@NuI@g=b^}D=2|Ax{?Svp
zqsZe)NHYPc7c*mU@b4SUQl{u9YLA8jM(XJN9!lL)96TX)trBKvW@btmIEgIWeKy*q
zYq8^RXa-0BezaQR{P50Gv3Be`{?jqDa-Vl%*~o#22p#;9lJfufB0i01d4W^8GOU`2
z1~`C0ga-QX*xIj|INRAE1JDbI(M8}Yh*I*mdk%OcQ0f0r;%0#1TLDw7YHo!C)xJ(H
z7SQdu2`2l+aZ6wckpk??rL-A;t$vZTohCvR6h*q(2#89<EdZ*P6c`j;gl8-I6TAbE
zHdSXm(znhZUCmgbIh-L$sbBx5`AdjnKy`10=+g!ODnKGXx&-+!VIG%7xNQ5(c#LzR
z+Zwqs-%jy1i>gZoMgei1K>9~cv?8856-Z;fdTI3fQHXM_dbS*t&m|GNunZE88Yz06
zzLVq7Z&tSRlL%|lCfh9|*1eC&d^_)0%OZy_?sAxhdGmUum169GB4@doVj9HQYZx$c
zf~a|^gH<rhVn}>{3>MqAb0xHnptE)d3-7;aslz79_t(euxx>JE=V(~6q5OJf>L~kz
z5F6GuK9q=h)q599ZpzSA4ur9R+t6DoW0ttTDj=Jk&%F1&3vf|pt4prIjRxLdLBhUz
zH6&g9=L0Fhg9?klax*Sbq?IpOmsVd5dQ^69d57Fw`M$V=x%%4pI?(;5g@7j7{jvQc
z>FST7fAMnoQgXD&C?AXL=cb%y`WYyuBKsc-@Y1N==l%RtsxE%wSURmz(L#UlfnS!1
zFbQO$B^6r8?UD(ldR1ar+MsCMuH|#V#LaaSh9kxpKh|Z(nC`D|8o(@IKVdM%rLmLb
zyUh*8|BYqb_RS9i3KXAwWlD6Iw-9c5%5?&ZWm3o8DZh25v09gI&%xqZ8+4(k%+-{(
zLO5}Ra-IhmYxQ=rS{TbGKPuast2+Qx9?yC@x{%k;$;nLiyVd9TM1*R+>wWDHBjd`;
zl+{|}(f~1EO5)Vq=-DytD~CNxyTA$uD@-odr^>7Gc(+efUiz~XyApe>%m411+)7yM
zz<zo5{*0;!uT^w<K3r64UDWD!R)wn-PJfc}L>7GjhVDg^KF@Bq@Ubuoikl_R)Y53o
zzW>G55`Tba>YGEBq!9L0OIT>HJ~ZV1JduE>_=DZ&Z}$NKlW!kT9w+nDnhl)(1e{kV
zZ8$vT6y*lKMmY5N)QW~+4#}%G%xpeS%B_3<>=Iz^@PR%zfyjlj1nd7iS0zwEvGgYg
z9U*X7P9WquE7^gVlS^|e1VgkUDN-v6s0!UDTDUY{;c)dNbkS9EmPLplcE21b;otoZ
zEqOsVK%<gR_fTl)^k+ya1Dc6qtn{TnmggxnCbp+s4JS!~)q-#71xn%waa&~m3Ox9K
zpd>AhBN2f2h<P?~8N?f0YNSFefri3-FX(wx^OzTK_ipjC8I1o)Kf=GfL%v|@vN29=
z0LrLJI?y`Z7Wo%OU<_pPw{i<AhSe`2Qi8|b=P9QR(3d-GTQN3A(OL;fULukXorqep
ziz#)y`2lxKFV>K;4KAAyO=NDHqB2iIrMnFJ#V6@=iOO~$^;*R6e*`y7%inzfKq#+%
zh&%rapcJb2MXwpB$VExZHm;oHfFN7?u$Sn4SClS2sZ)p3+pUq*Rbiilw()ch)pzAD
z?olgmpGyR6o5f4`l@9?xD<^&){ahTV3(2UFI5l#5QfRdw{m)xXoEh_^3vALesx^5u
zn>$?br1m#dGz*W>G;bCVYR*V1?NRWIm-cH%4NI%m%~kk?Bt}~Y&_XKLzCI_p^$Tn*
zogQx>8|&Iu24Y*~{>rhp`TS}?W;sU7^DE0p*D1e;j8Ou*=+JTNlNNic=f9e56xA4i
zyGM9S3+m5yQx7y#4=V~X|M(h^9ZL%n-dze=2g<baA2M}?+(~$P3;%BxfXGbJK=M*B
zwo^N{a}9RSQk(h{i|Swc>eueHg@4X0<8R$Yx&qkr-U>4@d;e+qXER1|PudY^ggyQE
zb#tgu@q^0P(QN8J0*}&~vT_^L%z3WAN@uJb)D}AJn*qxW6w_1-Q<yhb1{`RFYeg#A
zl1R66tO)K{AoZt=a%q%-fa(3!-{JEh6-2Mjk7RUMl%;d^9GjGRU*kOA!XR|oZhWKJ
z@}{{x)d!hkaoY`N>3;=3D9;uIMTtme-5PvW&=_%Coc(T0sI==<xU$f6So7=9dWs%F
zCC?tj5ry73Z2eXuwm=l0Bc5-${QG!uy&ACC*|OJW6{EVe(mFAj=N8q0G&H89#oTLT
zqv|C+OlC73FidVB-B@q5B;7B@8mo}L7*KUbz5Rb)Qi-F8mb^i2K$@R*R~p@p%#HJh
zAw4`uxZAWJ+G>uHA^;2{U6>-m2%4SpeM;MrBTl+27o8RwOUuxSjLDq|Sh5knOz_WM
zY4V$Xc@@WLHJF_Q%cix7Ulp?sy8?{YKacyK>vGS3KY^x9Qa{SWa}TrX2X@Y@trcfz
zjv?+v)VXdDMuWgG-&%-RR(?Q|qzerj1dH{`<#0WM!g?^lmltmW1m>|Q-5qrut4o@z
z;pg=SLEg`4pgH~iT<Bp8pqygs<a-kajGt-iJ1Q^~2bh)8zH}jspWta{S^ffsQYL-a
z!8qo{9y!nmrk=_|PjQII{7M!rlU>ZN<E7k_2qNzjI`01TqQWV3Oy~q68Yc%Ts2}C-
zL_c^i54Ai1Pjd>u?kX#ft2g38_j8Boyt`diYq(i42j19QSP3^2BLo4&CW}yvuj+R-
zA6R{>GR7dENh`{8xhCbV71KO*EH8}~XZO)$2(~J(J?&deOjy{UY4I=hkTt&TMrk9a
zEGD6dGu^|hH5?3?=7H-;55be4xtiOlo`?sXrx`7d{`8~qw8UfILPWYoE%}ow&FQ%H
z<|ZYp8r|4?QyPf{3WHTwX3n(QZ{KP)ZhRgqHZB6dMe5_a+${Bm+^7yc6$9Mi2&dor
zt@IDvQx9Qq&Bt{lqxt#u_T(d#3j~HGu`Dq`lx^UH;&)|xC~Lnh(|MA+YN8=H+36E$
zmCwyQZqW>PF*xaeZqC1LCsy9+$#7*qmiZOVP@SAEpOG&DO-BA0tDWBBtz<;EkNr-W
zs0#LdC5Gr;*>3)|tzz)HN<!HegGy@|pL;VBBhTD0f`&A7D9-L4P%ov(z#etaOsqOh
zk5g{x3_SFX)j#h1xD5LOo4iJ3lTrD>j;HUMn<w*^az_W9+uM+n#~~7apM5%k7Kgv4
zr2l4neEqJasB)8Xi_xLF37{3OuUAjcrxsfRo>ul}<h!y3?v-+K?6zL5T_Jy4@9}+X
z_G!5@W`Up^E980=>Cvo)d|WM%nDW>hvQ{w@N__i-^8cydq^T2M8*nt%q(o?%Q=mb%
z)cm`pbo&Ax)zx>wwCWw$u60khih|xI#z;as!U||>$9$5UK&+NRlBe3|yexHvWQVn0
zT`NkTkEohtqCD!R&QvB%)+%WCUVrk+3l56wBAGPG81D@;{foLm;x^ZC54Rh{v}%ZL
zR-y0<;gO=}&i72PE0ErDhEanKH4xFLJP-bT4wPNfsXE3ACZI8s{uRT~t7|7G9SVyG
zErIwEJTA+E0WJ+$c*1YXKrsYANWA)eVqz|Gt31~uOSS@K5~*~hcuIV4%FS2g{Ve0*
zViGF5&pG}W>!=<3!PoBcMyUw&%3!pJM*xQjpGlJ*0IoV)xB4n}#(sLK6`4%1^O&VG
zNx$IIy5!3H3`WTB-B5PY$who1#>VD~`WZUT7_azzQq2Ms7=j4qog%&#R_S*=|J110
zvWCDO{O9ys9%Z}4y^7LFSgYQ9J4@+B`NiJ=WFjV0i^$bVQ-6|O0~;f4cFDmF=!%m8
z10`nV?Ih^!p)5I`L!X&CBOxEKy`iDZ6!vc9m&=$KWg-UScBUeGByg;tBq)n9A&e(k
zYhPPa*zQ4jbyMs)azDQqnbsa~mxnHLPpIX$)(}cirdvgUqgFTd>TeWUOyn1M?u>J0
z^%b^jLix-Eg$Hs0GQxoRGp~YedTr@C_vgs(N7AIV=S>4=PxtT<RSa%VAv8zj2cLFA
zD^LjnxT4Ve@Vz+-b#k3{#WYW*yqIq;{KpOhakiJ;g)-gqhA+qDWxV=B+eWWdlNidn
zeW^#!Dl3Y+Qycud@_jB{rw76cEqm7jw407FR?rrnj?yhgjM?ry^Ak}UX+SH1h1h*l
zlx^%Cw$M^)D^pcsoW)-oDUGTEi|NDB{qwu=F{}I4f;GOPJV)VmT~{OT<?>2TF7JCy
zeBU<iUcyS;!cSF7gh3*{58s_mP6YxfYIWDa(E#Rw-%1c9ffCs2rBQTYkE4sJjlrbU
zMh^tOcj_e-anI$s)@3^rD((YFqdnJAH42j8w=4eg?t1ycn2l)pV3f3i;8O8t=Px7r
zH($oJGNLg{%9DYfH%}H6C9<gwf7;dXYW#DC5iTtcfuTbkABt`uncGi(?q0X5Jcqz(
z2`G|EncfD^zQWVQ^pf};4*eHwCN|mLP7jNVJN+?s?uUbI^y7watR;`4Tya4OPFB3d
zhovNxZ(A=@q{xAnM#9@Cb%3Wbl5y_Zh-L|nNpz4XAcTsQ`K|#fwO{!5A4#wCpC2OT
z#PT@aej0(7-(zlkGZuVyF=Y!L_*oQ=u78asR8w6mMQdKWb(SQbYR{9%QEagJ66v+h
z4T`o)saL@Pavz=2g(THjO>?P~aHxu0rg)%z3+Be;TWWsy^+0@HZH+iXK63;nOi4uR
zhx;Ul&+h5zFa^aYTicR8pW#2_Pg{!7-<C=gaJK#77xSy`t-Ywf@NED`VTnb8h&GM8
z#S~(9U<JMqasS)t5(KC}Is$gTl=Zx-hF~JEt7IH1E~={Q)PIp7Di2@6TV8^9v?xTz
z@VU!<zsRuQGZ3w2@z}gIv<hQbX|YS6o{{wuTp)P)gbH}EJ|f$ME-2-HvE$CJ+&ILL
z=4Hc#ygro@ycxlSZqQ`yLA<&3KY8kMqTdy0n#DUkciG08OzzNs{|$dOHN~EB*3~Fx
zM@|#S9CrpP^YJ1!A4Vx2-A@AyNglc(oKGJXXYR3GpIfQxYDVd=j#frL^cCV+^i>E|
zOO{&)=`<dWZ9S~|?`}JBgsF{o*#`1L*Vk6{-bm438!gf!S|UShl>6+h`0^Zg9`DdU
z-V`T*IEb&t#JLXvWr_JQxx8k<5h9$jIw&geOG@kwKj%%B_NU~QL5IkWk1KCtG}^3j
zmWB%5<#DUWI5NK3Ke)=4qgnozildjqQoZBTQP(8k_8tQm`j~*X0RD7$U2eclznoLw
z;oMGgol!1p)y~>=4&1|!840QIc0vY>TK#PcI`cRA=IczS|8J)OdBPm=f_4JUYf8mt
z$Vzd7XjN<s119q4zK1g<7ko?)arPjRbHNV>xDg0W1&*mf_DMBCwjp;~;TXlT6GfXt
zKfuxfCLCkZN6~>sE?ya_{|Y=D81IlYZ-IM20yG5W>{nDomJIB$({CKhRr7yT%g&9<
zZW6Txp;tzF14(Rd*;+X@%U%*W%yv*LdXa1UM6QsI+dO)*>gWCzPDC#AZ33t{(s%9w
zuu~q3&F>Zs@s!U(Np38SS^+3e2gxUnm@j30TLj{?0!?L-@X-<%$I#6gqrgUXmwirG
zGE;U~qaXnDOcOqAR~qeGxijCn^thsyXpcqond<BH(2}bD+iSpRI!H6S$T+}<Kx09c
z2RD)(mZDQKptUvxvupO3=ZCUPlF2rGhK2F2Cv;CPL0;~jq{W4>a%+OF=Fb*89V;PY
z&}byupjlDF1VP9!=xH~5Q4V^&JYH|7XrQJfdMhLkik$nnm&o{BCjqDqrD0%l%L{1F
zQW$pZE7<j9OtF&9lYBP_3k#d7nv47DR-R{*3uuJ2ADqH=Rf^9)I>OLN{->zorFm64
z<;-LCDu}?LF@Gkh;9Ip#e0}-U6IDp)w`v4P^ne89!{bS=<2&z=<Ei*m$6#*!_2rfj
z^uEU`$7+>UniV@kv1jq#9$)tv;;0kH75V$4&ZKUkRuYB0WZsD3UU5xbamyyPu%#mv
zh0N^^%tUYTkLC!`Y7>&=RrYGWT-*)B0(ncpXj=ur&{T+U7Sv0@AIYI9$*&GklB<lP
zzavHPG)3X`%=Y!g5Ht}qMka==_)bb8V3a-iInDnVSWRx)|D-BB+f*L2`n9Y#{=0;>
zKDPLm?X&ZHc@AG3R~*d@5?;pHp~cd4!FOBo7#y@$jg&F5VXurEo{m>kcs=V9jJ^7$
z=8gYu$eg0%=Q5YLO@6T5y1@{;IB9#Ge<q?^(@1Chzu)06WoTol_z;d);RwtJsnj@f
z5q%oJs*Tw2px7#I#(haMH7g%jmiE_ghYC^z0L^HbdV?zO<Hw3dFVUJ6w}iz(LkS6}
zWsq83wSg896xsol6jpaaN9{BlVOiXTutQ<uokMK6_AcKqSo2&+7&n2FDvl6_unQ~)
zp(I>bf4jQ5EwD{5^h=C6;Ei4<h&>g&K)$F&aN7(07_|eiAQMTHt|V06rrmSk+J`2s
z=w{RXOi3-A_e!Kiv*M9Fu1vt`aHmc1hR!0R?doW`+8}hQLy0d9Vk<mvP%^J^g0R4X
zC3Sz?BXBMXdqAdf!L@MDSYaxsLCxo*jMFcq1$y6uQdVRXSm*qV3NvBzq?9r9369Zt
zmmZ6FID=7EpI4b5zI=pIn`?|CKexO`Gz68))UbhL2VG?^bT#-7>wdV|dwx28=Ea~M
z8$rcThC9k!90aJ_)LX*V$ep0>?0;^vKmw)++*az+3y2@(K02FAdJSlrI5{8lT8<>1
zyx&+HepAGad2xAHrAR><zB>U-YOn~St^XM!5=DJn#p-x9km4U&pt~4Xcd;fvCisZD
zY%NaLVmh$Yd5WDx3JumA6(Ew?9i_qKcO0`}eQ*4M_1H<|*Ba%&2pw8xJDcwzKdhda
z;$7Gu%DjX!(VlyInxBJ~<Yg*U%M7)QuMbv7ZU<@En9C<GEh{s-DI=@$&anAiOAKhv
z38(lf%|%A82M+SGt2E;xU#ZFO;wVb%gHj8SXOzKNx_6>5@~I4gGgL$g`}Q|XV}^*w
z@89y6%rJ$?$a&PKKBe&J4q9U$wI5(?Y-r8Tv2yipth$ktQcs$PLu_P~$Hr$rNzM=$
zk3)J!@y|BiG4!FhpWh2|1UBI|I#wkR&tUS-J*FF<*~TCc3wYQ@4=r~I0hspkzsgh*
zrV;8K&9Cg+NdHr_i!;$Ap28aYeVnB!5WjUkS?do+)M3t{@Y$m>nxB3NM&7TKYyp66
zEz(*LHz97VJWKAi1{LgUC<5d1t0HK&qS0AKKr2g}P!n<~Po`0(O!a`x)eaLOWSgp0
zsUAjARIeqYpqq@i7iYi-A)#vz6@c2x1hSv0tHV2Hywlhv@mmN+NX(wI@5@pzgtycd
zPl8&wF#1ajFY~}Qo{k|_p3&7f5%lQyq;~epe+wE>i1=pzmfBQ_2v{)PHj89pQbqxV
ztq_&tI0)1zcVsFbc<|Zt{jZ2vIJyb=GN(FNG8=IJyy~$4JViz0AW?zWm4$}@<z{zH
z7ml57O3fIyv$e%&r#$;8hztY%B!WUajRcX`M;T^cY4atM<L!%ExPmzJ62NZWQ2(+0
z0uX9--6likcp}^;1^+CE+gH4E^7B;*SECj|I0nFgKCPX_zC8Vrf&hr706g=^b-i9B
z*ZNV|{|{Sl9hOxWb$ct_A>Frtf;32XcXxL;f^^-4fOI!VcekW;BPHG4-Susr_dM~Q
z@B9Uba&fuVUVE-N$M_BSn;M^{SH8Ezn|jT%`5Ji#5ff`NtVT1-L<jm;L#D~aGT`dC
z2ONs=4pk5ik>l+?K3jKT4;VZj7FNGpm9weOvH~pg)!c9O4nD+3cz?v-130z<pjgyQ
z#yo*ob)z8e<il>)7tpV?tgE(%$Na0bi9`Rr{i)z{Hr6kM9Q5c9g!+Tk3h`#TAEcJ<
zKWNt|!$N6Gc~3xh&hKItbFW)0-B#nT-EYAz<*T{xIw0`!aTB~K=CfqVr$(e-d{%j|
z=iI95v{7}uYJOZD&QKZp^z!wJv}o07z?Q_@8}m#CA-PyNXYp+a-^Pc92<b7@iS-e-
zX6Bho(S`Z8If12IQ`J@EdLXeVrR1WNhN#!hex0fIX(O`e!^ecI^2CLF6oT#i_Gc=d
zxdox1$JdzQ_Ln%X;3~T3r<>N&{k>cO{A_V)4n(zN;;)<d!<c%DnV88*c>laLG}D~&
z9(qQn<l<rXoR!n_Im&}2wpTd#gHo(nC(l(b$Gv9IV?u)H5|`c<SCAkWx-~@a72tOE
zc#%%isa!~u`Cdr-ka4nWWlPbKH<4mEm;OJ6O+8v=Lg4oG-ek~g>oSjtpkMY?<>{Gg
zHIW;#cFYZAkzcO7JgX(Q2?io4#M_^NbxmB)P_SsC$k!R85c!3B`xeb%*k<L~CzMJM
zMz!iJH;cb?`&U94+R~VO%oV}+jA{jBD3t^TAPip+6lpE-90xZBCKZ><u{5Z-=&+Z-
zj(n9q!5!+dmp1tQEdz?kfh*ErJV;9jiy?)`Z!8-MX#<F6^+6dZ!e?{uUW^KBnfxID
z$$8d|L5AtsHEFPTP-GWW5dTyi`ewTyD5NN|A(Mg_D|(m;jf*kY09!k~F8X33AB3lX
zY0Vi6QzMEmJBpxIBiMWr08E{(<`|L)_kPEsj<F;L&wsE79~`+N<?_)ag_sm`s<>U2
zHvrt>X}CPm=@tNqB>PGcFGdK5V)c3U6uRMPRO{gBd3nC!GttN+3Vwh_s{>~7gyV5Q
z!iXdpRWV#CdWb*7N4bpBPgT8|VTPAUFjzWNsY!YnK*mV$PY2YQpI@yyNo!oK0rC92
zA%O~UVL<QCph<#_Mp!+7O%r2JsGYd{1Ute16~kzuzt!f4G|ei&LdE3)5^%tvr5E<?
zJ|dSiTThT+MI?#WWwdA-YqdO_NjNT}AQ@Ik@HdyGuZiDqGAu5|55d<^&l~gh0HuGo
znBr!4+#l#sV2Bf6@o}TH0RKsY%t<(0)i>Z6R-Q$N$dCid?Vf6DpaEdQmXb+1UE_cn
ze)pQk!nW9x&gEh`rtlgSv;FJqS!vCvVlOnTq4XZ!v4OqpzFP<UXb=a^#W?Wp!y-&O
zy#@xS(!fOdyE<H;nYHT(I1$2igMB;6$m!rszC8P@GSg`?t`Ckoaf6t7k)ffuykTwc
z@|q36Y^v$H<uTo7htf@WWo8efVF%xdZ^1(kFtihR^&(QMIzxl2gM5{a4BjXGNP<si
zQY(?2UVaX^Uh;-?Q|DoxmF8%QTWS@vkFqX6d&`##i3PJr&G-46^xS?=^miH)(ta2A
z%$9?hPYWrIJ=4Z~JC9N9FW{&54`3p#gPMWi?<Kb;>5BnWNVvkwKJ^M~YD$HxIcHEk
zI`LTE&*)oA-5G1uH7es0K^!~7pquJ!mvk(cC5g4%iCk(*3TW5X)B5%G6Q17w!bhsa
z6&o!J;R*;&ZgFJ4qV;GOvAa8OnURy~A)5Ve6J|`w06D&+tby3cm5MSgFXfVivzrVP
z%!x2_=VoV9@5t>{k4+ly6*HDg9eNeyzNXod#Ml*yLJ5b*?~yoxskf=25)XO}<;j+R
zn8&O7-tjdg-=>#^oSz4`-y{dmkX$@acsg$Oe4=}N#D_lzT^tK!FE!7Mo$NnJV_r}G
zDf5;YcY1Q?vsu{gO9_f22(EWo-pkJtT53s}61-VT6e-9^Sxn;H%O!nh7leZ1o#8N%
z#^H?YAIJ7gN%i~ay2e~}92bm0*DJN^W00jsAoYmvyHc^bV@9tdd}z$L*%wh&>*ZP8
z^z4q<&K8nJ#zV{(8hCq~9+0>_a&hJ&LE;lSM)<?wcvS^%w8V1S-WEMpEp(KT{asJf
zH!Ymz^xJlche(^k;f*wZ=w~AW_O~_Jj#Y46Xp0VB-8<T7H;^w`Dm5N#5&oGgGidbX
zqVUNEyRdN;XB%p%k312sZOmtEWQZ7^Bab!x97%XVlA7=Toyo$WQ(^!`0&|ciF|T;B
zwj#aG&$v<c4W#hR(8%EN{-RfvJ?|S)vLRmJv}wtfmGOlxzx|QiCznB$J)J8Ndz)of
zc>R`dp<}4+H3;b8)Iu4IL{bZsgSSaRIK^^QLJ$1Gy|6xL<|s8QsX%xKns%qq6YJuS
z0A^tF31rVSbmcH7_JDo@LN~RRFIP{6P$+14KbWDf$O$3i!P-Ry^`(;Z250Zzh-gJQ
zwYd5r*8mTN2w6jCo+k|{YQz>0^Cbg%vR-UtEY0oLK?^ur<C%BEOAcjzck~*!@*M$w
zQmGkH;_?KA7pjjG?@fVbUAI>{Li;)9H;#9bj%zUI(*oWYuW3h<KrTsXv^ay9U=QtV
zr9G4pbn*8!R-;~|aN{5=3BV@Ql|y!&1qum7YfLCYau6#qSbk&Ru-xK?j^@h__f-f@
zC0rY4&5wA-EQ%)_x4c1A`otYz4QT2n`!Fs60bH6S^4Pua%D)2YJw@U1tAiN|%J|~m
zx4<jC9P$^|Ce<`|I5e+15UUEBGWVCUJ4TgNidlHoM!oxPfBfr->-f(889370bJ{^(
zc#JuHQ{ozX(vX5A*wcl_pB>V_DKcWKvjAO8%g_AfuUy?)55>ikK!Pbb0jh8lcO$}0
zb+kr^T!BRbqcHzkV(IxZ)v^@l$Y4icF>=Fu?AX7F_TK`Iyvg~n2Q#v7M&+C<ofrF8
z=to`I#Dyu@avhlDgFp)?hH8vsCoN~`VBMQ-(Vjlwp^U?E+Z++~DlP=_j$c2pm>F)C
zszkT-=e2asccnmsafDN)xmVNTFALPJns*sZ%7a(x{K3x~?x7D%s}XMH%)z}O5N~@N
zDCv#BhK+09kFO|--Y&qZUP37M%Z+~(Fo!(UInuCmSB9GzV4d-ykvx**+HtGBTq@;e
zle7%1dgCGLFdk=}6WW5tH<I67*xj;FCm$e}Z;kQ2w0x6tO`nBr!U;{Wm-}|T@w_^v
zBsZNQR=JD%^owT}=v=GJ4SmHuw(xnNjgt>TA<*&s0gD1R?tbwtqky=zy0gmxd4`%`
zEYi@gW8BZTUA}J;yRFeIDY-<opxDlnAIR$>y`ksBZYb{@c_iLTnnle2{<&XN8_rm7
zHa@MU4YcFp7<!RhtePX)l>xF!zwa^FGU;mYO3jxo*_XS{ixzfR5xDXy-J0!Zv?y|b
z*ji#p`tnm()&o{orY-$NSb1ZNPZ*__Yo|H`-C+RcR!4S_M5h*R{EMN`VB7h=aVSmV
zoxq>{rG=goHqsTn5>BM`36GOG^~&zSoe+0!kRAj0CW1@5<#u4~CT<}5O8My1NG03K
z<?jSq3w=&1&Z|{_+g_fFa+UA-33oq^muT@hfNY$_N5bu`n;)#p^vP8oeMGk5!?TT5
zAu9)~W`Caj3%@_zU&%<LJth+2vQ6}!gXJ5)<VCK8ukOq|8|h_%(OnRu-bxm4^m%A#
z4-BnH)3)eyhpxPkN7&_vnqKE7jOdaPwE#0hUE9OBRIAbo_^&_0O#KEq`{#fOg~s7t
zKkDWY5!Yj_Rbu7}<X!S&nTM8s;|}QS^|8&Gi@65Px#Q-n;2eIyQu1f77TlGRXn$m!
zl+FMaOEmDKZb3wd<qYpRFy!yfh5~h`-Y((6HRka}*e~+0_nf_*Bkp6k4YK;(8!qCo
z2U_m)BidZo^SQbh>pDpZ#YSG$^jQlu)kU*ga|zof-!4_JlWO^|f@OdV?M{U5j%iX6
zf;TRbh)eaxpE$S`#NzwLH`2{sAPn{BmLGl#Hwxwl?*-7@vg%f974@UU3LSp)fq;^}
zn721u+jAI71B)bGAPiaqHWvn$msdyxGob8sG`FNBREIt&C{AFT%4WwZtlSA}!~+R)
z3k@_^CrRNf!LCurrvl26#z-vWQD86^Lmw)M+F3MFqklIPbd5QQlGUjaLU5<3xH{N*
z;N=$WO)se1??)jMn*9&eKnKKk{J^TQ1hS*a^51vNLOS3Wb65wZrt>if@NlGopsNwj
zXG%3bb4yJ<0i9?In_1jIa-?&F(XTnC0Ud}~%2nr)KyBIN80q;2@e+4~XcPU>57~%U
zeG83VI$t_Dd&C0-3Lt=^1;ENV&PVeW!b!dm%%eiXN0g|1G`{MU`U0-k>t11V89Yia
zYlV$|OqFj<#L;*KNa5L5Ve9@hpR{Q;e|u{-jL3jC;mg?nUOgaN?6zw$m-Ab{Rd==y
z5ERi(Kg-|I>E$>0gQ2j1J%m*ksO<0!PW%w*ev60nRp&y_hXgW$po{O{GI7Gitu4Qm
zyvQRd{|wqIo^J*a%{Ez>!8Y=R-HbhM{?~%xm^bqF(-DE{^1iUqG>b7j`D<J8m%P2q
z+*15kXpV}N1|&Di9jK0w<zVm~^i4_!tdSw0tD9dhHs81fR(2<`gC_J{mpz}aRm(nM
z_bj3sRPQ2A`6%i;GHm968(5V{Km(pHmnjQ`Er1NHH`+KV3z2RxcOipo8|_di5Mfe?
zkCHZ?nG-~~A{|cr^(wRY)cbRp!t90**eO(CvWd2tX&~Si**5rsBOC2{Q8&e>A)@)+
z*#9KG#C-S-JxD$B(m~wEw#d)fmjuVkO=2J*q)`o=vX(O>X!a`K%tQ(8Xl{LP;L~1`
zE`GlNw;2l~3n1WUzEOTl47V3@JJ7X1h5bdKLk;hEp(^T)MWrhgx$7`ik{-kiK`rg#
zC?K^hI|76k7UIg>GXztJALbNtX6tG@<>@ojAP|~G;kww{m4y{Hu`vF+h1Q~<e!|ws
zl1a|*Am|v-ejiyJ;gyKM>lWEU@G=qSVl$!zoxSCQkH&Cjm@+8WE*LmTf!?2|QIIL1
zQBVg`;%JSyL)oWa?^g_Cc`>M)p;sc7n%qmuGo>x&+$NfzOl{?%6h=Y#=oN<zVYe2v
zG0<QW`llGd1(~*?Paf7;=$HJ%;bb4XTZ4rX2K_|+1&<T@dnvhSy!g}?mDQ(5M^t(z
zT2ffxyK!Uk8%*5jaTJ+ee{w!C=QkHw{kdiSy2)ttm~EY8TiN8vEhj*94P<NMC)aCy
zh6VE3!+C=9Zzk#b0f~X#R$#wIsilT{_pcUi3?HL9^Y%B3AoOzWJu~i>%cBO~YtoIa
zn-1`7_w&XT0XE9&Oj)W1*?GNH;_PRD%xrp1WxTUvLE*9Jc+N%7)viX(gM=YO&eHT_
z{-<SpWT3L=qi)yO%J;L2vu5b_{5Pp7nwL_0%S*$d={8fP$;m74A6T-Far9j8l3eTR
zcXy<1keHho=wvxQ@cXfyHDiBnpJcH>m@X=#^~y7#>nKRvpO#s~d2+Xx0gN@l!_5Z%
zk#1<-Hb{?WV(1oJhhH1K1tfY`Sy2ZEZ+V^@#gw4hS0O^3BV3Lw$Ac#&RNv1*|7XS}
z{t05wYf#EUX>EDkT?P+bSSa2~Q#lCL`!*0mIh=WGJJs&blR9n&35gLvq$)zE?A)k~
zL`|cl*9T#)36(;6Q)c7B!Pz4tV7-qIJ@KtJiX*Kp1RR}K#|u~X2xF*B(v6+QAmqhI
zzeeB4L8$~9c&MO9STO!3C6Ax7_<AK#6tjFAdSGp3X*B@_1SsBDzWTMpIts*y%D+56
z)v>n_b~Zx8qZl8W!(j0@!hM?{fm>Po=77kg+VYJN7E6!^k`Xz;B?rb3?LLnb4SEm_
z!tPDT`4dSl77YTUWfB>n@9c*a=>S-N@8GE29l7F|&ZN5eb76}pG8i+&d6F|=5d~Kz
z@#v~lVB!^{!cP5FO4Sv(#<I5YNBG5|IB=o*0h)&xc~1+pp*;=cT`3UQFmITpZh;<I
zm1|~ig#tkz$96?ypP7S0LKsmGZY&XFMv~QvSpfNghyUbB;cR&#Dbg)k!fIOy<zZM>
ztsx3`E`;5A^+&ybG}z#k8DY|PAil9_q^CmM!Bp1E%Eyz{N4mtrEy6a2UN?H`Yi0Hp
zMM6jeoq-uFyOoW&e@qdyv6IRZEA#ocGORA6A^P&qN0`b#1XlA#uwXWIA@49MryJqL
z0T&60Jmu2Cl6@l(Q7;Z$A~1WNz!ZX9`}N}Rs0{V8%Iv<7)LiKcg^;`Ti-A=cN>^pY
z`DD2a-8CF-Oyy3B8p#pwI8UHj1@HL1z4Te=%>JLUk|qB>#Z44?oD2G6x14GCYC7O|
zE8dfDjI2ZZBd&3(@9X#iBs_&Bb9-i>j`;@C%Xjv15ZkgRcK^>f)duP@=ptT*!Q_+>
zdfhybcZPa_7Q83W5oeOKNIa8UaEe5&lGrgc=I**2%21P{4ZZHjT*GQEJeQ~~5n+Kr
z7UQN$OM!|pRdyx!Jy?plqrr>~>+ytz)<ZS*`c<Ylo;_&8i*{b#JlZp+D6%=PPINTU
zGrVFVxHS2IvWw9n^H^?vpi1K5>tG>bM0*fIf$vj5Si+M>1n5u2lmr3@uACOAD_n1N
zP%JhZK97p~c<mj_X;8M?=^=)eC)FAuHC(eT!(}%I;czb8KB>9)GRA#9n3h^OTWVdX
z_$2DT8bSis7#Qd+gO0Ni$`a@2{n49VJOzL6tS?m^s=qrx8Y2|W<g8h1-NnmWw|390
z3Wc68(Osi|pdfr)IPyqM@9vf;DHe-sgyvj9_&KbBCMtsOzQ6Y*S39NnwmXE?n<SZ+
zd&5Src+%)?@BXy6ZrHKWl~ShYJhVgN3~b`}beD`|9z>I~PM6Of`)B7aK5RbzvKhNX
zC7<O7ZN@uGpJcg4hIRd=@f{j8>&`t6Rg!!w0qN82d)7<cvL8RpOCO&-_%)h$+Xcs@
zRh$$J<c~90M@pW87imZh4S&HSmY8VWwnitn$`k4Iefq@kq2}KbDi<1j$-R`9S0obX
z*8y3lj^?CjVK4Cnuoin&$a#DHYK^-2rGJyTLZ1t{!P|ZjkepHyhG?1z&eA#a<PR4W
z=J&G4K>A_}h~>kS?bwlnaX?hyA0HCDp?)w2GD5lZR3VY8p?xrnc5yi7M@QXP9f(*s
zWd(O&j_^}{@B6WcEmMWF3E8-gMg0P&|4Wd2L(DEBpqecBU=US8q65xEG-^xHt4u0e
zA}<9C;lX{db`6=U+pgLO_<802+-$C=`ZDksGuie)ftHa^OiWxp&nC!(usKN}1$L$R
z)?w&xjO>Wl8x?)WJNNV50P0tZLUaPgMZO-=6GQL5Cn)W45|>4X6i#Q?IY0z#q7rep
z_6x6>&uRDfx(G{=HF@uo;}{Q)S~O^!K;<T22NCNGJHVJmn!(UMvz&^IogS`Ke!6A9
zxP&>+5sQ%k$~pdvy*<Y)jDWV>*=KIrYoqkOBqjUjw+G7j{fF|_;XC*7gZGmP*uk5i
zEc`n4SpwPu@<}k|_YGFV?%+7pP!SS#*!g;!RQV<lrqQ@r^|u~U+!J{Zxp)IG^3(+e
zEe5gUYaqu`PSG|@trA>I5vQhN1hAIm{%<f<;!R^kU?jE4D;LT12Ti{v4Bisu*_~tr
z<Hn9)qvu1bC*NUpeW}<lNbS5Ep;>AaCPit#hPXpS5t58OKSsG?iWYa<VjTM+K<8~{
z6V}3I|0A%qEg;4PEWr$|tE4Lu4jPu#Wu}(fZ<db-u!W>2!U#JmB1E#|E0C~{hv-Fx
z<zEkG)tb~pK>9Vww_19;odLkKWvaHwe>y5OLM()td{pxM;^}a1a)_9Sr^S4s2SPyh
zqcV2&OdrJcxnloHEa4+>pM;v;(M12%Kkc5Sk-a6F?bYoYDqG#Sz3etGoyW(0W4RwF
zA5XE*xrzm!O*Qz5xHxN>5pmpECN?QDXVs10r}EU9jyLV3Br_A3<@m<{49Lm(;5&;M
zI#aAZz+G!QJ^xnmEb9F@^p|y-pHs1R%ZEmO)j@rEp&^kd#oA5s1d)(1jk6wQK_UT)
zV#_`EMAP`hSa7sjDYxln*wm$(>Y$;r@c8E^ot9IMt4$qt!q!73M4oGP?9dj>@a9$z
zM~yAERL07)#=2_r;J_6txKGTt>Ra0>g(}pauMW(%Zf;qz`TFyTxW3$RNuz}3&uRQ7
z4Q-v!P?x57YJ9p*&Fl*e;=q%A&XDgvSxawW?+1FGJw9!)HwyLts{+6TlN0N`D`UFz
zyj8yg6pZIWS}VCKQ-D^m-&VD4S{n;MnoU|UPW@V4#vXK*r;wU@4M^}SQSiJ)qLh4%
zD<Ywrtq@Im0@$VsI{PpbLGI{_^aPBJ7vUx#Hhu*p+Y&kBo8AHSJjwXoXffh(ZSImd
z)Ywm2=b~b7$l=j?9G=OAIH5G&3=Y^|AkVZiq{&hFp^h@56|C0*&rQQ@cs-RJDHL{8
z?<EbkfjWI=K%xxPk2jWa(%3c;EkFaFSy3P0u*ASX7)wz?(gU7!viBc#^f{>B2y+*@
ztk0(wf@k5^Nsr${7<<!L%a44JzxrIf9cgvcA5DflQbfRJYs?ZNxJ0+AiAYK%ZUX{+
zFa#m5Ka<Z{eI@<c7rWE`@>Ee@NKSmN>v|_I1C5TmV-AQVzUO4o6*1)?W+}T#ky$_9
zo-vya<2p+fok09CZwB}=`Lw;Agcv7RBsaz8h%HRrOf+vaffRFLrbUuf?>EbD!s0L#
zI@;d??V=fF7aUqmW$s>q)z=(yVQe%oAiWq0wpZx@L~ycz#Yj;^BtWG3Bh))1+qZw@
zEn%+*aD$uG``<txE7Rvo<?9snhXzt5tz=o+b%*KO<)%%DUjT@K>d-Mai4qE<mJLl-
zulvgY`0`!8UM%ch!6(hHFSbx*u4&+)1SeNT@D0lFhug@e<-%W|A+T}QHE<o%mkovA
zpW=%sPXKdAeopk$qfm14Y+lFdDcbMiW)FGR0-$L)b<1hMNK|)O5XFANw-A@BOI|&t
zuPTF7lko*VInn@6HIM_xs{~Gl2B4UeY$j>uUZ#58sTkYLvRC;HI+XX3xLkbBU_3vL
zdatTVR7b6Hp8e_6`@)g9#F6-YTD=8EF`RnB+S92Jy^ql+Qi3Y7)Y^Hj6IVEdZ<Tdc
zfvd^^!UK=3cc#>_<U$AY4UtLq+v_(_vhz(fmk~i-YctAf0w0Mvr+PAYi5}bXqgG6?
zs_p$^VNjgTxO7nYy4?51_?~E-Fy|wMuW##UZB}F-WtYY5THcZtrP8&{4){r-c(0uN
z9(5_S6oht}Dx~4uitFz8Eri`-vsCUMTeH2!#v&PQSWfBrK7buQ#c3)%M}D@>v+^B_
z*PCz8l*NF8zLRds{&=#l2g(GMR&&!RQnfddbSjR3jg`@jfLs=_9kO9UNzR0Ee4+QG
z`{<BoN|sv9baH!S?lHVu*6Us!3B5|7u!rOAIa}~5BF_D+QR11ih*Zc}W;$nbvbL>B
zWC-hZN9xG;*`~+QG1;ylGsftd!Nxd=pCYxkiROfHb<Ivy480m*Rq2sF5u*D&Nw~a=
z#o>vV2R{QZQH{(xq3Au79mg5_FwZ^L^-p{=0{S1`Ll3>a2mX;xJAmX(PbdkP&3~Z7
zssUOaK(J!7C`?hfC>VGIj8?M%Y;E^j4Rb|Le&{O0t-<(j({B|Zcw_ty;qrJc-|idy
zGDfgI_(Te3u;}KG@O`)fdb&DFV&Ms9fDdLoJ9OhW-D<1p2)_GK-OMf^hx~|!8HSHJ
z@C}tj7Q_HPj4K-S6@5@rot!RG665mWf2Z6OH+_K9s|>Vk=Z=PvMwT2J*Xel68vsx&
zDVKJ_K@)O>I3ebnbNqcJJUG3xpNOIyfbh)z1C8?~=#7a3TE~xIoX}lds|ct$!d&?%
zKgQm59W)5iGr)aUK2iqK>rH(zXSZy$^_%($=u08SM=6`IW@w|nW)1%|=z>w@#(gJ1
zJVoY#-?|LpLnE2YVwI3DOnme+I5A^ah`CD}vHBb^9_G{5Gh}xodGa5oa=Da?;!A%6
z;g#7Gg>_W{67WdY<E&B3+)VI2FzV`;L_js~y9DOd+c;KzP}#LF?tPeu(P5O;-!N6b
zb0r~G$|E5*`l83tZ7`x;R-PhDu7PR|$|a}VZ;s18som^PQ?C;u&|@5_;9);PZU6Ls
zx({4@D4u}rw3jtlLu&yh-LG+>hG%KZjBM`2kXDNeh6+T-;S9V$epI;*)R#(hKpZ$!
z8>l3&-*X|xEi6IE=Srf{?#<(a>85cJY>+!zod8pOft%U*3>av6EH8|7B6CbfeZ|Ax
zF8aJOKAZIzBuRT*yA3gDXntk0YY%0IE9B_(;C5Z1&!oO`7d2^o59qPWhckGzJVv`a
zW1aA@8w5jeb_f{=81#KvBiHcWT)=zFPX~SBfoo18ISpF@dSdGz_h?fTicWCMVvzq5
zV@krk^$`G9e^ipa-&QpVd`Agq3<r^bPq90A$gwj9Z(*{wL-NbcU0hwiPm#X<6zVw7
z`e<qeirdN%Q3q=O`xW0ypaKo@Pyjt|1vS!7YSr@`I%3Q6t{?)kx+S0Xcz7pPov>_x
zVv+QYvm)cGYVs2e_JmL7*qH5l<m!%7uNWQDCz@WAef-bc!Sf}Qr@e1zddB0al3mBE
zG=<?1Dhkq7`$LT)2{SR(Drjtnuu1r4IB%Ou#bt1e4GRT)&H^LbU+(IP<dPXjz>l~6
zp(?HTAsG8@WTK6CI{qQ>Zl>O<n@xH?_~SM}KVA4{X~#5;sD*)zS9N!`iuKvl^pAM2
zo;RMJ?h;HMhv9}O34Ow;dB74r3xdYd?5~rJrAOJ$q0Q;t3nblw>0*=hlND3dth(Qy
z$BfS|TEbTzj4x`uP-49t-KuO++LikAZ&n&Exa<kqH%77#DjLqYq(k#HL*+(6YZ+ek
zN=jyIg5A(r1RHc%suje`8d|QSLlg6@K0YWV2xb?}8U*dARf<RRTO0T>2&EGmFygJa
z7?C`tlf8%$T@~%7HEdfn;g32E<<HFv#`A$a44=*JcLC0ca*gTxcE>OF>9!v+3rsyf
zOtKD8|9i}A5xm8~kQnSCg8ljS98)=*)o!`eYqKkeH;Q?Gra}(%2;(geR~)kCYQfNm
z+#6&<X$=11{{l?er>3HRJgVi0B8OihoZE6s3Kp4D&HLA9@jSS4;9+_T_;sTS*P93A
zm|yvKyjje;h8;dzX!U5wkHPKWJ|&q=gjQ6jeNO0O4-4;bMScuw{^H5phq?ioOM<2u
z_6~WFOm?E+Jp4`hd!Lslw&ugRtR~wae0}|h?nfe21(*vE=AQh=sST7<t+rQ%_?BN<
zz#YyEq;@91VucPM+Js?4ZteOCGYg|<=kNpzkJ*RFhvp(ssQPKXAWNc0#1N@Wvrdvg
z3KtOs58e9?TfUnsa$Ogudl4ED`*i==W`3;4E{4JO6fzp#xLCcrVI9VFw-=b}Gk?$_
zoa}Ib-op>*>tKWiV<D#5>p9foA8!FB;ru6tZN`RG*lF{ytBywq2m<W++zoGxK7;_N
zu&<B&5J+JTk8GMd&fd)tA_m}%@E6<-`FRw=3&7i9wb>G*dl`7}c1m}=9mtb4CqLsS
zaqwdJ84U7FBR17|hlb<>B;BlYKwRL7M}Z%<I(*QkA$3tYha^m20Cvz$q-+5e4$xJF
z>X&9H%v-+?sd{_JSPt2^;Uf|0?_{Ld(M?N338aOR2=+pbf&o<;){8xHTz5;56=H7}
z5NK489?Fi=2xBn3aUaMSav*b?9pbbRCw-^h2=1v~gE7&I^6&YdMuH3!LV>3wY^Ez^
zHoS^1zSy-Rr)YqbFUX!5RCy5NIMCp)!M@hJd=in&6gmA>4%FT!)ZARBMs=!bIWs*G
zXr^|=wLSa;QpiqCWSmitmVC&8CC)(<GJRz<-~R^XPt4Bp7d91K^;#QdA2&HRvxr2V
zJ@z>x!NUx}E41z+^Lih7pD!=xIZP5OUvn(3!6=$(A$PnUH`}L<X70g!gD|<D?AL`+
zR}PpaclG=)G9Y0E2L{27#xKBGkQlF$-qYh7)o8BZeAZ^2O9DPxczUF3ic-<PI>2Y>
z5X=z+dJE<QVJDq}?fK;omy^eaUrnF>n1~{^cS=e^<2u6iPl58cxA3}v$V1gKggI)0
zGnXmrYr|L#D0xampPTf^R&}?!Ooc$3XD9KI)YQ2P%>MCPyY(9xQ*U~6q{LJ7DxfQg
zv-5MWW~{~C%Biy^*SOuFi?q`r2n8Gak?G<ga+s6%)1EO0EtIh%rGw8ync4%1e6F0B
zi^c;Oe%o|^Zn9I53j6(+0!ZUeN`DJ2*$2@n*`Stl9@!;t{(HUnKU(_WkPg7UtKqi2
zzU~CNWVM*cb=6~q0-p!GiL1*xCbR{b@RQ0OU0o)6e~DaUggY275IGrW6r(Sa@Iw2)
z9G-oc!z_|Q>&Jyc;k{$PlD$`-%Sg-cDw_D+*WaX1eANE%5@+${0Xg@(3r}GfBB~Ru
zp5OzFE)j9*k=lzBj0OR-VlN*2WwTE0O>z0=LUr<R-^>ro%(0Ic>PCk65cYlE$4w}@
zVfX}5K_vCn6p6c=I&swvH+h>n&6*mkc~;9=J$GfbjlG9IiS{G6egOd<#%I1{_uS50
z!6AF^Po4MH^`X1Tif3AEGjQJL7nGmdTTe^}5!gAM3wG~Pl_TRP%|2v@2L^<Bj7rzH
zf^hgAD?@uD={43(s*hwpzJySu+^WycF2pdVWi2i=)hLlQJ)a7(A`*QN{UVb;^2_b!
z=)y{}N1i%^XF;k(m9O}W(}?pz`g5JF&7WtZX4|DHG#v*S*GC7o5^)bH;ie{M8;nrz
zq&ru2=Dbx5S#cS@YzWHe4}Kl_GQ6r%*A_<Z$J+Fv;dup`Xm<;>pGYtKerr`^gk@cV
zX!8*M)*6c8KztE%a3ZqXEBkL*_?N_rum|pSIG=m7cP{--G2{>qLILfx^1JR&q`_sX
z5&-N$rQZ2JPpbmdVLIJMs~jc4{aRL=_nS*-!p1x1LZF>u@1Ht!X}52Q@Gl6r+Br1Y
zWJdE|#=}E{-#iG8W%ulD!6VQ=I{Aq%i^vCR1;M$o!KKn_d8V9~qn{t2uJ!geQ!e}H
z`^eK~rNVM!g+lT-N)-H3+KeC)6R+DVMwCqcwW%|OA(PCK?EKmqe^1qj<b2)_b$EI6
zxlwlR<IkIAn(KEsL?M9BP&&$G;&hGf-1QxoC?dA&-qQ^a6jK^{i3WzTDwl`VpeZWf
z9#3CbZ2|cN|6Yv}1;vtcYA6axMEC_Xj`}VbwzSr>DHF>xr=PN*3dr|;uvlbArQdh1
zgsl(+o@tTTk3|E`SNq9@Kqs;FS`!=K&Xj#h9Lq>AE6TdrHIRXKbD3izp-$$U3(fHV
zC@W3k#zr4m=;89Y_3;b#)3XPwhrw``%W{=cCa>x>ue(#&3^Phe7!|(oJQ%X<@~qGK
zS~M+ZL_&Uu_r>$Ghmw?z(%Qm(ABh(?MRSq27cBJ!=kB!Hk~D4v4^D15mSW7miy6lT
zB@Aw8AyU|j{Pz$4ssB0QoKBY0bx6I7dWeUfQC!xCN|HeG-{F)C<-q{$LWCmI^UGNU
zK&vms{pvhX_6qCX0~8{`lxBMqc~jLUz#aB~3`;+9(NDP$SXdNp?5IOphzDD@Ijg6N
zl_o}_?xDb8g8C;Nx|_Of%*U+YH+o)=JKrvvyr03>OA~s@ZTDt+u~*&S1{>AE0JYuX
zWD6(iW%X?}tA(V>a%y)#Gj3S3dSF7VYr&-4PTGbs&UuvtxMA5ckI~4G`UK|3+;rQg
z=t;a;<hvEa9-8@MIz?6zF4j;U=G>?r3`LA_CAp@(StA*IFeMFDUh~mB$8Hb_se@Cq
zW51=$kt+fBY!7)`zle)eg0WJGs-U;j9+*Xb7E6uYVt;Cvss(H~$~jb8bjE6lx#IAp
z`DrRIG<Ip?A*s+J`(pI-#aL&z{^#^Bl84p#ayRDP^?Wbfoc-1X8J32mrRn%=5fR96
z(;zkf+nZ$%*8P<ge0F7vhs?~Af`QmpZWQiUrF#pf(2i>cDc9HPN~W^FmjYQGwdtx4
zQl89!bXn<*Y7Yg0seytqC6vpaVs7<9J#Bw#Q94enP0l908x#uBc@1YpqUmtQn3z_5
zuD5sf`6sV&5z~+DT6^b1v~j7_8|9mkwNsRTS7LEGPyroU&XMW0+JB$w|L636{m=zx
zJy#uw&uq?b+n_b%2nbOCse7w2dpkKM)C9QW>@klaO*|tuRtR8i88l>vG_z%}SYi9H
zH&JE}IA7Cx20!_m<|%rA$j1JkJ@J>#3M`Cn@5vR4vdG1Ld>EgJFEfYZ{5g&O<*4eF
zR1m)x`Q(h3d=XJtPld4uwcUP+i6!f*6<YSV4Qp)mH9>YThyFG{HLdiyu!|VcFGma&
z(d#-neAv{e8;1|W#l^p)#E63K0gUqSrEWIUr~Y%*P|}b``nr=p4&7z-U}kJcsOtTR
zPk3zPSH!81B(|}VPXK=Q;=-{|5Zb7*=sYNhh<$aZ$2qW+0iRf?)ZRR&q(YYOEo1d|
z4wvybX@zJ2+joeH#N6>58>cShGoli}f9CW@1H~`fu9C=wkP&`bQF`BIc`=t;O#lMs
z%%dBQ?Pvi<+pp><f@s8Pj7QbxZFifO_x(;L=~pXCZ6l8(O+K6DLkda*Q$LkWsovW&
zR@J!hh>Q6;Ud!=L0oDxC1p2X$l;85_TXgwG(q7G<AP*c&;XiM(*En{PG?M>x3acRk
z_tJakfSLPAIbHWdMmDc2mCbek)7rCRf8WKr5|IPc;C_|jTwCI>2HKc5Yfz<UlN?#}
z2BF5!sl@XpD8~~lHFahtyPeVPIQsHW1e8+L|Gv&a#J~<AK-G*<{r_(eFjKH^ET5&z
z;cRrgt~ZVC6Y%`cs;uqxJOC7dI2Z^(^lX-xR?3ZMI$0pM6KS{9B^Z9<e7u-Et+Fj_
z+z8x6)vGKmg&P5j=ieG${%ugy%e|D7>mqI!@}T8hN^$E|^xCC3+?7t@>9^E8TQ5#i
z#Z4YzSI+4+mj&$k2NGTWP3Pz`BeUYi#gbxUF%#Uy`V}W7>xMH>+(?E^!J+lWCiTgv
zkGWlgV|?TN<HzDEC%nGoM(J#SSjUO%Zn$c)#SWqI3NtyhsH4H_T+&J<3fPg<f{z_X
zOQ}LCCHQ@Gsvdy*L`LWvChZ$!xrc5H4Q6Zg`>C+5gV-lBIA3*uvANG^Pp-22r*1|w
zN!}_{3jO?M#cQg*%pZsUTLM;#S{?$U!;&DV@awU4);DCKkRSG+_^gvs2OfyeduEI+
z=cl^wtbstO6em4-Aa>j85peTHs+OHw4A%2KZRR{a{`uJG^*0M(6+7YZ`sqe%Z-Anv
z-*B7{6_|g^+^n4B1K1!>x`lYCU17~u@772_#*5l|5s_GL6Kc$=Rw6o8riDIml^rS>
zv3cKxsw?Agb5OBC_(p->w~`05gT4uh4-qIVu<ITuT_u#Apz3X}#xM7DQ<!>6bo5@P
zd4e-OMX1ww?oJO$KHY0L?bNV;kInsi(R@MjvgmMLF;oTVakl^37XBf|*w>Pt0$t{6
z!O<q5767yj2yQp#L}++_P5?Zq-dr|7W~#z5t|Kru`Szp3<G4voVp${xfRwUQ=2`z!
znfUR4B#mO*r5kL!<m*l8ZL6kUW?NLB_ffh_)<4j!Qck8QaicbAjPB?%%ADzMrl}Ei
zbK%t|A5w_kPSk60TCk}aXqHJyQb(reJ*RtfNitgzy5Vj2A%dE8qRu**m~G!X;SKPn
zHml+~{(7J&rigeld$WQQ9opFi`2fGn3tJuZcyss2Y#K&EPU8wInQb!tQD)*qjp_l)
zkS(qtD>y~6`JjB4s-pNO;C&kEW(*Q_R_Q6}l;xS0Ch>yuE265WwC2n7j-~LhIMruk
zvE2M|e}!0gK_od&9GMxx(JO4%Vqnziz*=X8lLr#PDp9@Th>QhH<ULf?(YT#FJfjE8
zt!2!XapBveBl%Acj^=aK*7(4Ct^6kfxr)dy3nce#JXGp=PK%lH{%e@x!0OXvvlNSG
zPZAFiNn0Ahy<32KBAwyP;5Pq@-Bqx^xsOoDX(Jq4b+c|Vf{5et`N3I+qnRAjGD|YT
zc?~z;uO&56E!X`1j~46vPnIC>yAP>YUN)teHt6TJ`t}1a0ceFPL(={*CTHv;rl<-W
zDJjl3$EJp`vY)Ebxg?r1>TM2k0aE&2ve{y-SMxJrbc|#k+JB6;j&R*KwaVK+DUh*W
zbtnHgaQ}M!tnP~H4<|7`RvqpbqP>O|402UqoeX^g$)~{W799%Gx&!v5(hOs5Uig3&
z;DS66N?*KWAQX5%NL@iQaxA&@@6Oam150*^D#(uU3jf(f$izDF3aFN~HLRS<HV#@L
z5o=KtPwsQRPaZr~2OqDf#W-%`Xfo9<C>H3J(uj$0!-es9{JdPH>s<}Nkd{uXqR1sa
z00vx*X{X_ky4O}O6?GP|8Dj=xya(_O@a?ub_u?F6qJ`7;kX}wrifT82!o#8q=IQ+T
zUW$+IcZT#MpG$cn@5`bVsf<7AMWchRH;jXVfx-SM1S8gtbPJ52HDH6dzct#(RVYm5
zeR8^VV{ar%WxY%0KFE?yPgBkjeXrxw8aswc&thkr9f){z36v4V8lTz4E^P|0BZ!a2
zG0So2BfJ)WczWFLnSR7HW2~AU6`m*jI5@9y_D8ZH9h=W|T&o}=P!LdKqwxAnk-OBM
z9W~gLPDn9(Hm0)LsTL?D>SF#h3d$t*S(E`T>`d=xdJxsUx~<@uOSF^CSNRcx=Jqmu
zzU-DY97<EwpU%g`?7ckR>Hb`WfTbsQu{`?zCsN(v>wZ!an(Z`Pf3wRDgT~@$fH8Gs
zR?MDM>9vICozR=j>x1%|YEnD&<QsofYwjQ|N(u@VFP+wTIaKb|e87q`cM#`aAx12@
zRma6r7N_{{2Km7ZD3A76r+8QrL{W2P|F3=XKf6c=>biz}6o_aUq<Y77NA^cw)fi6%
zi<<Z@sw`E36bk%~|4jw?DDQ){W{Yc8b-m@`&IN#&*@<*+sB*^`g8gI-AohWX9J}+N
zrSV7bIf)2@;Z`11EucJm7FQI*qX+-HyP}feTwSaSAU-Z)OWy_buTk3O<~j2eGpw2p
zHf5_hsOxu~jnLRke$hr|+EO9^JS%Nop4nE=i8t9_r=L+@aj)bx$X)%XyyW|L{Q8%d
z`xuWsIpe`H{^|VD502CO3X>$r(T!uxw}uoWDOl-gFoA%$uEvV$d&uf~+5-!b=c=>G
zpRJ%&`JHJJD+dxU;~cB<5WOa4jnj6peD~4sUjvdkSw;LQT+g>t?}h=K>jcM6VaN`6
z{^jv=s*1dlkaF9pAS`vtK1s7?opdQ@8mXw#4@isWNXEIJbxLGWerfTxkPcTdQr{!y
z=FC#}Lg^0y%&TQ@Pk*$n`{np<z_IxgP%lf;w?y?I#P-gA*7uEF_F`iNt5%F|L{jWe
z_6=H2s{#<GSW#ZDGZDG&rkq5QH5yx4R376zG2#Kqq%h#l!kfbIONhjKH!qxE^1+Qz
z*o4xZ)vlbE!mf2%q@ry<ONzeEe1Bv7o;X;mx?7CC{V-pS_l#+e9!)Z?hwvbjzFxlF
z8yg__Bbt~IaFvTmihCm#4@L59Z{KPAEHfxdP3C|6xsHJNQtH%;XL%SFM`i&9RwlK-
zt;}qDoRGqxaIX`A4?-;{zufm63cfqLKPXEncHNgui22X1qj?Vk*vU?{xn=*Z{XbI^
za$f-9cA-8}Ow|r&o6K)JKnj<O1J095WHQ*Gai%4B^0)u`T&&PwtD}lnEN0t~`V6pG
zjvDM%+LDkTfvH_mrGw0{Q_O#kKqid0D{VZRbG6{%K0>L4{l1t_6l!zwJ2w{=FGP^9
zrn8z+bD`<wTaapXu<OoalaqZvh?e{2#kK(9#YIdwC5Lf!&TOgF9WPa%-8L<JniPZL
z&uHt7kmd7Hj%08<mn-&__Co@8gM$(4idcRG63o}N1X?0bK70v?fC=-5YnqB=L4RuZ
zLTk~Qc<LG;adq?I_ThwAm6}H|Np0G~&|BZYAi@BmusAFR<zW*8l@FtFx8-8EbTGvm
z;sjs@10CSVN*XziJaG$7tPOjhLq1(`vT9b?t-uX*FkUP6p%5O^G+DlqV$QvXlblEa
zvNDVD&hF#=vR8fFaofTOvHv=C4ZulheQ65^E>26ft=GQ8;cYDsw&=-h@y0EyoK@b&
z_fA|T<D#RmNkaH6ROH;FvMEsUa!+pde`x5b-#5eWxBdFY2rZ(CmJ`WSzbVlSuE$Nz
zgs|v(NxfsJx3kM1qKCDZ>4ZAzj<j-}1F$W*kZ&oA{f8fZ!HO6!>oa*(Um18qMFZhG
z)bxh*t&W^?-s0Z>9y!4zZ<`P!NInnz$At0s_TWzzTm|&3Xy>-coQ7g*Mu3y7=U9gC
zL+S{29ngQ(skg?HNoES<OYkiEwBJj5vc0u+CKitx&_aNv5%QNG@O3*#V!s_rkl75<
zyJScDUSB^sX8d*0{pY3*rrPOvvqKzy3UkBF*Q~cN*N-%<?ziz~wQ>2y)07Q*hMNYE
zD)B|+LkV}gYe7b_E&}UP-&zo}_rm0Eb`2dnv?N0PHLa<{W5M4&dHJYyIO8{Uay1(0
zVx|cCQ~O^65nCAtnkm^-$g{_Dy2t~u!cCvaZ06(Uy{U>NG@=%c3Jkiw_iknheV9Bh
z(-PX@Xp{wP-UDNdS!*Fq)y_7bsmnj!gxn<Rd6tSF>N7c7V>86LQ^?g{_ziBtn+zpB
z>?LR|PSusfYSuALBD1Ljcdrdf-_wQSc>ifDqcKMJyp$v`O3YKL-8{<&KzPGuo#eUF
z3A}iH3#D)<s1HRml%`8k)vFTn9Ow9rVI(+bfy<HCV^z?~7UqNq4oB!f^V~2`(8TfM
zo%8)hBq2T%d5JPWYj8Z8KQjp8<#)RUUU`Qi6hSH;HFe%Vm&p~X-#MQ&D%L1FgQsjh
zt<gvfWsHnfb%PB9yPn)CiR9$xkPHw7s(7-@e$pBM7{vZwW}xzYMi)QcD}US|<N0UV
z{yjMUdZxk5WWRxnY6_zf47FE!LUjsYg%tpFkp^l3UW%&unB;%o#tF;dZkOm@=@UiA
z4;nytS8K!X3~J8taypycj^KxZi<wM`c)YYZ!0V;W=R-nh?7FaA_1VPb(2(5=lw~wn
zu({NzyD&n;o+RDmQ284oFs|lBKL2201}|26zC1ly=NC@&%scR#<5G{SFNztFcwac3
ze`_g(f`1xO*#I|Hucta*c*iBqr$(U7bRU*96wf_i_TtSalY}P0syyB9?R9=tG$<p6
zJ7xcJUsBi#82od-^)oeChjg1Qu<~KF5|%#Wr!o_07@EVysm0bUHHUDzI8PARbrTx_
z*skm%rK7mmSnD#8J5(al@)@C*hs?wQ%a5v#th%-WZ^z3T{Gbkhq;ruuu<E{3h^PAa
z@we%CrjO~Qn%-QIQQy=k4hnY7@pzYrye~?FmTi@yKg`E)q?bWzku%V7+e=$e3bO+P
zcub`IS#zo#V31u-DQVPxwL<Az(2(F%oMvlzjG1pgs-Jaq6Um6EXq)e3pEaK;C;Z5=
zkldp4bMV)Pz{n{jZ}g5whe;XL`|Abk`4*1U`0AQMw~Ni>7WF8=2cWw@H6THedHBOD
z!-0ZlCVnLc#PgSK`gNfPVIhjV5I&J4iPQgIJK&%C_D2A^uj1ukJT*(;5)f}luI&Mk
z9rpl`cn*orV>F*rTADpajKaS+?eA!(T<NuT@_h3iI;is=8!Vf;&7AMe5te;I(^f|R
zoZsJ^koDojny=W*#M-?q2Mi*P8CRRdeyR|6VBw~lZ!44WC}uuyGM9*Rc=qdZx*ZwU
zOluOA8QgOSXIdJ^W;YH_Re7I-v2Xc@JcWIHlo<CNSrdPFxt$))tD2}-^X5R%6NC)4
zOy71ofgvaS?#z#%%FXfe8MYw2u2mnLxIM}u3b(DqV&U;%qZ-d^U7CbU5LP#&Zm}m7
z4|KJn$s*B|QHiCO&O-iJVu5k?ZZ4t%sSXb32BKY?T^cGwugqx~uy;(&VUWLke*Z^=
zSkB2>mTzfzuwk610JnsNc{<bQUIWa^Jlszld`YXtss4+FgGK>I*{#{R?l;?t@nmP9
zvudl%5`~j)duMSIo$4V{I#;PG1~Q4RFtA%C*un{RvXtb${;r$(X?`@HZR9Y@VS0b~
znZRZ?N(PlXN@TkD1FIeWug~MmW@!fb^)_T@#k%R{9d8h2GHqwXn@N1EcLH%#2Zx4Q
zc**Bo1}Na5(0rNhpKi>?&$l=pZK+egq1091`<xpQ6+Z*Dt7+bW;Lm+3pj`V<-12sK
ztNrg5t9%Apa7!wkTx)FGzs-(+F35kC68y;mZ%Ab50mOE=Czsta_s`>>2fK?vvMK;#
ziR#H##Vzhv<OT5Y2+epcp$V@HM&BSLJi)osp-&(GK0)9hB*^UTHz9%6sxo!o(*b>L
z!wXj2%Y7)6if+#KTy@C#3MFYX%+A@-yg;B51q|$m7Mp6#qKQh1;A*6pp;8UywO*~F
z!J&=*G~X|_znbl0d`^hVTASFGML%j0o=3yqFx7naTfCpQ%7Usp#okCbo|HQRM!oB~
z39uwiHSf&P4JwLN^uRa11wh#B<Hht%^qNg*9jeLF^HciaD5v{i`%6SL$o)EMR*Zy4
zkQ+Hf{*b?l9`6FUSvy-_Zs2G?e?u-hFtcweTK0Z+L#4(c{)9DH@xfrc_pSH@07)Q&
zOFyec+>~o{h%^Ps<9|JP-Y+2zS8p+syWHT^_l?J*P;!(Nv1%S?FT~fg2gg1?=(!AP
zk2KiX8N?r1-|tayeLYA>k=rBYtt)Tp?b-Sk6AlDz$mfWD1O!F6YB<zy9|ne1o_F}L
z%buYTtLrwbtH^oH@Ozk&of?)knX#+*jZXEcN0N;Se4Ynh8}t|I4V9f=OdI%;+oW^_
zK!o6YnGtRZEcBE1yWY`UPec{*2?<DSE$Ot~tsZDupcaM&U$%RBT;6fE==>L~n;hyJ
zqDD0o&*K;wSu6eT3Gly965O|JK4r7jCWx;TDkU+_wEzsb_LdgzUt<6Mi-`dKwZz!H
z)tsA8QVkh^={M3dqZHYQdijKeDLCz7hjM6u;G98amxmuOQT&HXpEFh3BRq$xm%wtF
z4~D=!T5qTa4aHOzjcr4ngwkMJCACI`S-t|qw#oY0GBrK9;mU5p&#R(+ph+)bR;2*Z
zlQ=+QZ?saXVF0I~I2|oUQa2Hd4kPjBI|R@N=?ZO$U{EPwA?CNaNcg1qj;GO$Qikp7
zkGQcqWkMJh=B<Xo#IMjELmx#*ypu6L$Cx77A;3tg%lYEDQJpK_1f@WykjgnNq^9iv
zwV=Lrd1zr^@_YDWxsf)jwYvCPu1+F#P<3?-0mk_0uYwr_EQF#;22CnbT^IgLPs1F5
z!`&Y_t|Zngn0423aoN`ojNHtJaS3Ak(rOFLTLp~Y8@vambmsHSTmVFl*I|0^Qt9DZ
zR<oMCMmdSTZC3iE-OkKpeTsO&?V`o$q?HaB0bbvcz?gN@iF<-B0fOPCb~?kau24wr
zQc2mzmA@Ru;RFX8Bnsv*P>tH&_ggq5>0b339=c<`A0;qe2Oq1y-}BMypEDIL358<k
znI3nuYy!?{ry*L^{V9pj)-cZ13EaO+Opxqr@0CJ7=Ks~4|NAffb8TS9N3*Zc1nh>7
zV(lfR&Z#a7t#J!G?mC?jp(-9XORnM7NJ@EwPQ$f!18HixamCVc!ap)UptinPlPSs4
zX%rW9%sjT+a4??>e37PBL@8(A7a43T^mtcBL!trH3rxwMl{J`J%#IdPfvyslhpKp(
zc7u7en$kRyYKo5QK9<v$(Ql{4wNJkYM5FUAwxAl01~gaNWC2XZ?i1qFTR&ZIZ(Dt~
ztQU}P#@7TTvTi)u)r&C<3cu_K*>pl?y9@`#aR80j<duAMwSK*zBds5<jJ4d#6;!My
zXHAIDK`bxAy%vCbxa5O4I7r%A)qFW|E75{WakbAunKFO!9cxaxiAkYCmrk=~cJ|DK
zf}-@&+)QMdjC}Ic+YU%U!KNj2I@NcT-sPP#_UbK09!Xl&J(~Z>yJI8G@W;-BR75Dt
z9Z<+26MPHI8$fr9<Vo0lIH^h2))k((a(}wK-LO?%ty3)1Gzo1?g-eJLLt~#uS8x7P
zfbO>9F(#$e@R2POs<L}MOTfPrD*n=PdcwsEaf;&QKb`bCxPB#m%d@zN|NTt=zrO;1
z0rRU)V#h7Kh(5J7>cqXQ$JV|?fH#+Fw36w3hHD1xp|-CHl335_2HC2Zlk1$9;$mEQ
z4*o<L-S1Fl&>$1Uwhn!mzaD=yn;@5yKBt2VAo0|fN@_u@UX54d`~-Y5T%wM;E9ldq
z70u})o6*hYUc12Hh7-==nbf}_5fODUF1Fm*-&{CTMBn{=x3Jzao}TWU4q>4>4@x6h
zL8~G{yZu!?i4OE?_x)jGxp?>SBLlK+b6^^Ffu7m!K}9%vxBpOJwNoVVc;9WUg>JD{
zLom0`(%11xn89y`4mZcL?l(Ft-r{QQZOOWxjP)r%s&2~e5dD!RR~B}%Eo4)@kr)Nf
z3pZq{;OA7CQe*2p7OQHF@WcvQiA$52a#BxJ!vBx4w*bm|>)M9}MLLu&De0E(lJ3qM
z1f;vWLAs>7ySqcWySr1k>5lJro^#IgzW=w*H^Vq`<2X8d_S$P*>k0~-K((1S2DU`t
z{4;NHd0+nbBol?vN$KuWBEah5a>WYpT&a)Z8LSgAn}0AFKw%pv39x^@mH`Zb?<y#<
z5t5jt|E`-9w=gh14J!nChnauB+5aB!|A!Yl3JcPuVO@3+L?4;Fc-`$(+i**|qj6&=
z#t97RgU`o5qbfTiifqw=y=QZ$Wzv1d>y+ZxQdig#hC55p?&g>2tk1a3P8AvNksW(!
zcEj1sOVN@S$cz5iCJmCxKpxVr6l+}X@*#im0l1DmM+h@aZuP1pJ@0P9Un27)aLVw2
z<tH)^8s9xb>Iz<2#95=N_9VIKj{?Ns;1%&5$ASbH+(PQ;S7g55dpAXPzM07~nhkWk
zTGqc^udJt#`Fdm<o1ZQYa#xBJLx*aYrI|fj)6(7OLNh~fhrL(5A?R>eB%r_2ZT!G3
zb+;6uH=m{Avsl{EkJIG>LRqA{0|DF9QMEd9%<-*bscHws*B2RE_AM@qD$}u1mdTp&
zCFbixR*URg9u0Lsl~?3sMP7nhQ13B$6w3~jtpS>Ml?)r%p)|<_-Y3g;2YCtA2es2b
z6~`UegZ5qN>rA$DnZ_t`ZML~vE5-JbMmR5~aQ?1#gdcr+GH8g0fBq*d`9Eze|NdkD
zE{YVQFo}tW%7>E$gPxuVg>4rnt3r>KP4RFA(ljy&z^18pc@r1?Vg2jg3LK!|Xaj4s
z2CeYX%`6#PxN<IzAIDCf&<6Jb#v`GQzk{@=y&*7E5+tR{zwoV(f7T)g7WJR+KaZ!%
zs*`zanT-C3JMc&S_=*6aiM%Qz4%Nw)S1(0+G4jqH_XfzQpGY1{ipkk&<RQVp1~8d{
z=tai2A8@&PIzh0sYpzeAIcFr2$UsOLIcJ>OL#Ga-K6^hzP4l_=N`bsg8((M5Zk$DN
zw%A93`1SQ7G|~~wYn%E5VQ>n~CM(3Ehu{Ab^MJ56IQon4W!rT;kLk16;{b^z<RSrF
z4ZzmhK2PgDYG*Z_qv6Qbu;x-)MkRATY`sjawa|1=EFOYGHk<5{TKBRlVAQWJ6!cA~
z1{4^u(bwI}#E-zJD|hpuP-FJ1(^`^niOTLowy>kGo~xy$?XyPnj4a#g&Xkf1xUn1k
z_L{V5iL~KAu1}92|CSZqR*~@3e{6C7-=6pX{f<BBYiCb;7hr4>Po?E0d3EEqVsE`8
z$Js+w_lL*z<)egwt74pKteb<EfmZ(ehomxR0Q3k+QQe$#3u%a)3$JLun0f>s$9(pY
zOyeZO;aQ4D`6`zUP%6lupK6lbRY#loOq+~Aa-^p2wFCHEAxd7T%2QC-OdtG~&SGSb
zsMAHLQJqQk<o2m4#6FWV0={>gs3oJ$<$&c7M3%Wh{?|yr<V$$tV)ux#L=5L+tr7BM
zFDBPEv500F+0mS|P5|cDW0y<8@o4~+oY?AJw7(eagQ>M@-OQq{us>%H=3ZklOFoEG
z&Cb|9m_OST?Q>9x4=!(;lb}}FnFzyn<a~X)@+j|nve`>QdN}WwLlE?qwT3=Q0!Zh=
z!|?_+O9px#&cK>Sm+nusE}=Jniz9|KqsMYRo<%f&C`RAj5&z5CD>M?y`1B{!sUCqe
zE-vFCw8+!!u(Q_x+-{Qk@$1Q=0<6p;;lC^3|8<Q}-hFuk5i|JJr2#2RB9ce)`y_d2
zRuL3x-=t5L(?NZ+;4AK6Y)o!xqUGUXu=UtW2zQCsJlr^E4eg>vMp}BvkXVr%13=wZ
z9hqytjt{3&3CnPXY`7%^kEImI`j_hQVAZ|A)60f+YN{O+4Aa?><@(~*3A`s3jj%@z
zVhvU!hHK@`$k6s+@w{H;f2_9*|7`z$OXB#o*5p$%pN`i`CMc;8&@00q4FfX*%=Sc~
z!wHUuOga<!Pr{|fDs_@`$3wl}is!&a6WCK21tnW$dU!l)=>N3V@mlR3`Wf91^?l3b
zvf5-*kPGCTRaO7)h~>fS(J?y$Kdtm3<xVY91>l?-mOYb*eE(446g<pd*?yxzZn5|=
zSXv88Tn{Ka6izTo<yy638}mAL&h7Qnoj&pY=pAo2Mv}ItF}|ldXA5aE06qxkX407m
z_x~wSqezsHM^>=Nsv!2?q>TUJx%p2|QC=AT7^$j4T*5k;Vz0!c=9Wqfmvmf-oMnaQ
z<RuyQF*UnXddG`Ybu7u^b19Ac-7IFv+piu8fDC6viak{3p>gv)Jjh<&UQd^?PyiR=
z;&dcw64qi8&Dp~fHO)pwTdt^&HLG?7uyO><`dwH1Btq$D0X71_*zbY;A!FozO>_-B
z=1Yx?4>n;cruiS70S}y2uh$QGs%`ZIoVl1W?ss!#yi0YG9eA4dRT{J;E9p93ZLUbl
zalm79b&23l3wX}y>0PUG#hiy%$zamo0E#P4H5ZQR)&!k;VbJoX0n&L+<Pky2%?T~D
zn`8YRCole2?NkClT6K`_ruA;iwHQ%iAd)v(7#?Gx$a3`#gO)$*4wy)B(jFZ1nllvA
zfeI9h1PLVGW!Y^1E}cm4zJ*htX?1bGlyWY4eOZ?6{?SzAcp}&iw#gUH<8GDt(_8k>
zH^T#Hxl1C$d;af3_#6M@$AY}6-<PYqC5z5`5!Yp`a<X%~mMwWGe{4XnE$9!Z-9NId
zEm%k2S*)aU%+brWmU{7b*JL#-GT+zcS0B~r8<20$Nuax)T&S81&aX2~Itai-U_kHA
z92yUND2xG|rar^`Ak8br@F>y{RU!S7vi<sw)gIUlzH%#+F}I$Epmz%i-l|t}TDE|B
z*_y8J!x@Jqzxp^`9USymsn;Cf4>ALG^t;I{RpL>HN*h43!h!zn&2rgHiK=}UEz%8n
zhO@~Gh5p-Y5kyoLi(;=*R`WAJkaFaxPDAGU?10Pj;;b^l_~RYms+`nw{C&&G*kv?1
z34m|2QGLi|Z>BDGv+ld71CE_hT=qw*rjv~NypRGDFxy|ZLO&f%cB3bB**IfQ-YnPG
zP5dNX{GHF+Uhj1oE^B8)P-Y(dFW~HrHsX84p{*unQN@?j<#|Czx2?vFHC><+2L`8@
zG4`COHJARF`g_K66^VbZ`!b(MP}Zk~?)!yY|9{>ZMwH6ABH7pRwWLmUo@cA4rpLA#
zrD@`iacChJUl$dA5>r+y<%SP*d=}g^!u&vMEGzakPcRx|aX$8^LhOyw^q1vT@M|iU
z;oe%(Bj}WM^dfcs*s)_+op=8kRPdO(uav=ptT0i`1*@F1{X@ICMl@y8=^J0fiWBq`
z^yLYCTM8>-jTN#pt5sNiao3YitJaQ`SD7`>^{H;aaGZ?*rz#T8&JX)(B}oMoiUQ9F
zvO$y7<vKgvUN{FI9f34#Lj#D4rs5w$_5!3t>wv{DqhVNcrm|tj5fzE+wE`y9(g_jz
zv6@!%OrUV+Ojf-~&Pz=Ca2~A|BBBA{hLvk9M;rdBjIL{~hP;_|b}7UugWC%T+RWHs
zNrGInYL_<x5*I#LwHfs8CX^7!4hnICye@(tT;w~;grFVdE8{EgCPirzS%%?4k?_Im
zUeMNy26RvNM$S4Ju0y_nEP<Hno)*zC!&PFP@n?A#--~hgBUsbs;c#+QQlcKOU0B=f
zWVyxq?3?Y_dNK>*-IXG)Nca~L9L5eje!{fIBr)%IVnBw^p%$-Oz4S=XMkhSIn)Q;&
z`6_kZqucAtX(lyT2)RiTVP*mq^)l)W?4MsfBK(gP-3vzslphHp{?9KTIH}IW_y|*a
zS%D*bp{r|!!iI~zQIXCP+MhG3y3XaPl-rx*Xm-(ox%@J$)#MSdiyv3@PPwcItq;Cc
z0<54CT}yWrdvvABl^8OYbAavxS)*yi0*eW9VAQptS+2jg0ebe)W}#t&#7O`0AG8HC
zAu^p@R_UsZfpEB-;h#Y=>t5?WR<s;8b>gTVgqanbx$FyFm#5BEL+oZ0YLQw|$&2AB
z>6-_F#?rzWN{r)%e|p~n^_o4~UBW*L$QCwd$)eVP)TV5#UPMr9oXF>ZYDR0lIo$3V
zOFNoR{ms8zf5dlh3hQiu->`1k*H5q<6;C!l6O-!sTs@o$<VjG}havc($d_akxN14W
zYp}vxABVFgpFe^f!|=@n>P(^BfT*$;H$J-Nf#8l;x)8n>jZjkt9K<ex#SkcXK-WUM
zR9+vO01dh&s^e^od?%K9@gd6Z{U`Sc-!jb;wdR>0<sfMW;aePx_q0r?zaXR$aXE!S
zwkq!kATNA+zCwQH|JPR(GvtxiC~29xiqzj9{rQ9c{;S*?teB7TQl^{y$*elV9JfFX
z*VL#=^Oo!DT6PlK5<(xUOqg2F^Kj&c1~ZE*g2AtA%hQ)N*x_PLCAKZ#(^cF~lE}hP
z-V#-4z+!B*u8#e7w#1`x1G2#Sat0&t`}a5DIJ=?xB_-@<I8)c~&Up-ZPF8GBIGpXM
zy@%I><g8*gr}74u>)&pR53G18el9jEw}DASHLQ<{j<UZcl(4JsUXP6MTMwD<;-O^g
zG3CNksnG6f4cam&TQXh%EXb!sMxhMFvQi>y$uy@OmD&agqxCwoIN99yhzRV5xlh{^
z_{F=x=s;cfIY5bR{r8Xgl24bym_;RZS1`^7c=%k&XS)fGqmAF!o*#)){nPGt6S8^N
zZ33|YV>9=1k?k((<)}eZmTz~~q1FWs-_E|-X?4{;;atA1R+AFuhbO#+zRKXr+P$W6
zIyUUX)LD>gsZU>l;fxwjib)lqdm#Qs_mR-&LV~|N#_8z$$A}kJ^JT;SoXBK{bvNQ`
z^>$3aeo=OWd!sAKh?7y*7kbdXG7{}=L-e0_kKcfZAHvFjW9m2gUw84}m-4^-=5w%j
z*)?e)tL~5TC7=dw$bpZeF6Wrv)hfj_n~scsFU`tizJjqaw3I<(*)<wXV}zahrAp?O
zaI9NCPv|8|`s{RP-H>nXFoH#$i<-K|>`;uUs!dr?Jx@o}+j6<L=#OjC9>v(5Yu-?-
z+6e%My!PNz-*~sOZU`D}b5!D`r;J%7JUI+Si3aH!bppRK&1Wng4G`LDWfmHJvR`?b
zX>9ESiiYEZJ$>$Sfx=J75nRgwO#fnhSr#M-2Xfbzb%IWilL>sK(6V&=TkudEFHtR`
z1*QA_R%o|Ze3|Ba5HeqFus&{j#odzA`aPp<?t_bM7wr21J7LN)&3(*%t@~C)+WG3d
z59=OA^k~C@r_}B4Vu|TGHUX#?8b8IG;Q>uRh(*mnjdhl6&~V7tj$gi<FVXv3W)<oI
z8*>qJW!>sW6rUJ}J6<I|RKrWgf0pfWd$}ijI`hMqPNpEpnAq59izjZ9Bi15=_!t!5
zNm+n(el@M~dttMa3#=Bf?d*Ol#W8TS=EkGvEs~BUwuv!zFuC5HbMp2w=I^s4Mp=((
z{t~UHBMy|||MLqC*4a}tr?H|3iKuuq{R|Qbxe<HKt3GZ4fBJpQh3qGvm6gKctCQ96
zRu{XA;Y3=oCOa{uU22Uz<q~&iFpq;z3L5zhxZ)I2qmz`J+#As3V=)gPMmEzis(s~r
zELYRToL+_Au-}_k=(cUlZ2q_|f{)bN@x0g1GT)f#U<xC|wW(qTjG)3E%XO0KFzN*n
zywT|h8aF!9V$_#XiQf=Ih|6&L`Yao4rK;0?cP^pM9*(J~6r;~iTSfzjcQoo^A4z&s
z`nd+%pRN?`1+%{glkh!C@k9U|$+0X*%QDRr28ZIlU+78w5!Z0!s@G<f(zv^BnP?al
zIu)OE1f(unJUvL-ABG};91q?XsQWw;hXNToB8AaN+2H`yh~)`RX!xUutJUlPd6*v-
ziC=o*T&#lFc7x9GZIkcRYi24!otKJK$Ahfet8y$fclDLzI_z0JFU0ZL`g|dufD<Hy
zfP$R*)v8KwclL65T;Yg-XDlcYB4~)8Qq$X5FCNpnwYc%M#ZhJ?103L9@s0TJXD=wy
z`$h-$e0F^5uS)yh3+msW#CdyPZX@TM<{1IW(laISwuaIoowx<x#KmDvnFS*=nV9cV
z2e;o4eTg6Rd>&;;QzN%`%G0=UeT&7(oDrbVKiXw&ck!Ki=W0|#9L{K*Z#|f8T?vx}
zV@9!|oQUFcY<Xr38G0JYtTMjyG<=BYQ~X$iDrmRog8}M#m#u!En)UOwa76ahxdT|+
zpeiIxRW_MpIy8=Y7AkHYWYjgQ3Z&T9X=#Ayc*-QmbqRoIF%Hkgu{0-)Yt>O7)#ev6
zNraLDwhMOd3$aO#tb(wS7#ircab88ff{JFUi}7lYhs1NQSJfVF1wP2FsOyts5taCh
zs+JAkfoseez=YOz(U&_5sKmJb#tRXbV~Q23@q<G{nCIo{+sJTZq7h2qNbc8bPRPJM
zrAitw%LVsp>!Oe6v((m_VkLRKRR+8L^J7auCHuprxK-$EF?OYW5h6a9>18VWWT<q?
z3>evU*?{+tW+dv}&PJfgwE05aFwy$}HS34N^HuhayDq@*WJRwt5EY;d{_1!YxeeQt
zC;IysDTp(lF$Lwd>Hey-e|I|nrKtY#Ek7*e8i;FSv9i9$5rpf1!SFbnV13XhV@>yn
z1yQ6FO2onU$gW3e*WJfpTtpxw;o9&5JO5MoUeo`?M53cb@J6M*cV7EaulxPA{ulQ~
z^~S2A*PYJ+oeCQ^zo2A@q!cSEyf2cBaC~J_q?5&NB?#t9DX9{4G3qnm2BTDjb&)IG
z8VFRi54+kuLe$}$!mxDv-Xg2P&F}i!ug6*DONd8$JH5Vm7pj8LPORT-_o3I7ophF<
z$=}_;qAK}fs=M7V6U{iNN_SG5@j6!?B>FeRKGnH_9r{1_DWm@C!@2GY_cUM5AK5Cc
zL5NaJrUqhDtBv6d^3Gfr7<Mm3LTYBF06{stpxF!q%&l2U`QdyXVY|Ml(Zr!s=Nl%|
z8Mz2DCe-BewV}Mwn}JKaJt<s(?3F)E+gSL0ko5kbY5x<lwkxFzv36iL;nyoxEjJ5s
zw!Mj<9cQ#Uef;Ev;Xm#X3ZhRK0uXc2urbbe@=E{;)Ewkvc@&8}d$j8|nG@;S!$gPN
zu&T1tSFRDn_3z@@1=Dj*fm9Mm{m)l1*8BKUmiShS0#X4QIk3pHd*D&1{=xCZFDK{6
z=ESHfiPh#KOrwKFNC+plCR=yS&3@EU2!@5KJ(b2}RV+zQkDxH#$l<{xZoWpHLu_%P
zdTz?nc4SHD;?XQq5Vx@4wAg2Wx4Z733deoL*{khd^+CtWf*y@WLUd>%muy4XRx5sM
z_2vY$&<;{JR-fp@cS>Y146FNAcjmS&a$Ws1Pz8s1AiO(p){f^UQMhpl{W;923QGwj
z0;0Qa{BCwU{}_-{+#F<gFma}@UX$;d3PdV3lxPc?y8&S4rXTVQzEzy5%#XL=R+pnH
z`=eaCh1x{~=T{HvVin~OPOrz2BwqIilb!lx7HUo%;@OX(-l}tob;S_F@rylmIocf~
zZU-n6S$<5;;n%>c(M1gn{h1168w_jfdbgj^&}kW2G53Ll+xbHL`ca#kc0ixUt6iC0
zwKxMc!($IqZ@q^xgxoNm6yjr{^Sj)bCU;h#slnH5SnBO0W-3;VY|EI6^SUu7&}eYX
zQl=(19KGl|<mLs#RNeZL{&kugcPJAtlE-&Fy(<4c-~a829D;?|Kid63Qm5m>GWCEg
z67GaSl4qn-NQLrR^6T~%1JBJ64N1$6vbsRV+v#!r!!g~$0hjX|b$T99rN-9sT9H|>
zg#_+O$IH9h`H~AKcyu+1bpZ)=>fW0`0>|yb>Pd^JU&Hj8Q9nPb*ETy@y#b>4e9K#o
z=MFJ#u1?IlhT>Q^H<EO~Kqw`VGxbN><PG6wO3`yHtf^_q-^+Ewk;C*(u>Bk-?3S&Q
z{(~`+Ax;K)Qzpu^;tk*G0J5t#jfT=pf|yGR3lo3yD~7BoU8@3nP8^w(944u@ZSV(L
z%^&>)M}4?OFMiM@7SgBnD=XgC60?)K0SpoN)4b4VAyk&>Y_cHZ#ip8OHANNoB@+>;
zq>uNl@D^$}JoFXc334au1^fo)vuR#|zi8jm7`Hb;Q4#TO2Nk!xiR|f#B3-up^^xAX
z7oBd?-a^5_A-3y9$19vCf6F3e^gZ`g6x8wDBGLG+Gr&V}{tBwf(ATh#6nJq1s}q4)
zOik{rIS#URs#K`EJ^podX?@-!hB==Lmn0f=`W#M|ZGrgj)w2suGrVqcboiVm;#Auu
zUoK_0POn{;j6VG}JyKY`+j>0CYrT#?|M`E^-hch*bGo;$>x7M6#y8FPDTs%+d?80Q
zd)@9LsGt*QBqL%`7-!_`gCM`v3jR6jaU$?Gl=br+Eq4lA*M0Ao@3JXDLTnDc`}<&$
zITI64x)BH2@J9NnzSLlP3`@rqd)zQUA>&OyqRxm3<ol4XOwR=5t7@C{T+b(<C4b`h
z2*atyY8o2|fb?g_GZrx2hKwF<P=J2GZ&u^;rzGx0c&i}Z>SlG2vbk8rM@Kow-)BI-
zkzk!TjnM^ih?8d@P!Bk(EPS3AVl|fohb4PE?aT0;wgI);_@gT?WqPz;&z88BNQn6G
zB0Ce4x6_(DU%JqbSnha0XB0-l9OMUGm-J&#w&y}o)`b}q8&H+0H#DTMf<7I=_j(2I
zq;G&UHCYMb^57d>bFaZ>TQ8+MdU^!EN%!Qe#(Ml+6So{)K7Niz-?Hirw9lK($1~e8
zD7U|(w>;RAndLc9h}Tz|h%g<ct|T!`R~CZkJ8A(q2wp)^@^rq%U<8Z)2Z9Z}kmIg?
zNkh5Vy%gEoy^*SHHz$L1uV?hmVS4g(`CRA5@IP8vJfyep5g7DGaRL*ee*Mt2Wc>Bf
za)bc=Gsgfx>S@*U4nYOAH2Wf53^o^tpkp!p-4SDx$X(E%f59N^{`brLzX1#s%kQS%
zBll0rccTo@+6vd{IFXc&jv>OYOx0o+ceJ6nT<#M-UMGAQ;KZocG3jl%B-m+ZQZ#AU
za(tX!MZZN~-+a*H09xGSj>Tx-Sl+jxMn%lW3R;Zzg`zO?u`M)E_}l$~Edde*E2`S(
zaG3J++34cw_8<K_XUO~`HuYE=xKK*znSUZ&Mo-Ls9PUGbsId2l8EaW?D6!w`zyZej
zASn2*-tJ!N*K<(-LP8_Vuo3zGEBiau>On1Dcpwr>Xi-5oHd+r#<XbR+>TWJ+$SDqj
zn;j4Z!$8#_S_trF_mY-mJ;r0a!|@XqmkMZt0%V+LeH9~1OhBD6!6=ETL%M%poa45S
zf9>4tz$A?Jox1Mw;aG);f3$n8O@bqLVl{NFZ4{tbXjLvZ+U#6bWRg*`maEt7*Z>Yv
z*R*N+05lR}xvHhs%=~5D7jU4O;AI1n5;yPn;UzwwYRVw;eGh_~*EZ|IcWSJ1Z*nb_
zpzU}tecA94kLdzCNZjYF(gi%&?Q2Zt1mcEbiefDA-KIhVD2QMh=Vm|WQhj2^y9hRj
z3br>KiC>W}DDMW`Y>a<)-I&f^@L^-WSE{fSTWI5~#iuhvZM@S%-7#8kf0HjW>eUeA
z$D}82OE@n|aPOP(0}-+D(Xz&w{0nrX70q=O!e3*r5;G*)nMy6m(98eOqW$Lz!cU1>
z37SG;!7TH)o83RNTzn^46JM$0V9`T?Ah?@oyIoePGl{Lfoe_8$gG{m)>XcJt$+N(v
zol}zREqArs%39B0B{9CNTS{imL~Y{wXhAUJU;Hr?Eg7Iv$g^+1_s>bzY20)5>q3zI
zt}mzjGmR7b98m7e6e}yZQ-JgVGmBQ|AMtOkG}_;S>n%KqU>t*sQtF@po|^ct0%>Im
z``>DBynx`~d37UfLXZSW9;VI}pm%Wiuz6U3dp5!}q~gSyQp#of2uHT?OU}LhjkXsE
z^qs1gIa<rB#6JZ);b4i;1Lx`EuHNFt`8)rpN%XxwyrT4Z?Hauo_v@FTMnwhUxCqM>
zG{cU2!|$600@6ILUvKS7(MqM$Oi{NKRRz#V+jkut*F7w0G!2oL$ybeLr{ZZLwCtlb
zkWM_N@HOF9TkLI2(x)}*#C6;3AKTBm7KD3EEoLgLdiFDx15g_=Z`UhmGb5D}?B6)C
z6Q)|nkZi?eA0YEa$+XhZs4HB2icUl1zRMJ4_bXAe7KKY)7TaL|u+Y8=&9Si6u<qV<
zzeb+^%abvvFA~%O8TFPTTiTXT-0!`}=2r46$E6f%?(&_wQBqGG_I(nr1@1_N)J6{~
z;@$cy4HnA~&FM4mfDfXMhw_DXo|e$vr$d^T&bObq9}3A2&VWq&i_W~WpIRh;_W3>`
zl7!72Sk!dTH(~!UaElQlLH32=h&8=ntg>L0%=6EZ>!9hatqr7eBb|af=0GacIYruU
zze}Zu%aN<peW4>FzOh)w8POAy^xK|;FkPg+YXANF%_q-=vL#W<BlESDpq5ix{Y^po
z8?fqVd7~m61t1^vbpx{qaT;RCDmZhx+KUTY|2Q;0FV#K%ab^E>t=jJFv_CfBJ3j%b
z;7lD5HFYgZD=uGSAFU`+rYVXcE8&xkR$NGWxc0c0qM{5@mHf71IB#WKIMiLvPN)tZ
z@gU7f<Yll@0^Na3VMa^iu&97r5G{XgO`0ufqNzqah;>$3?~5g+L<k#@Ri}y`Wc2+F
z5GQosDreRri5X*Uz@odxz3b`z(xd~!>(TgdM0{?RBXZ=i)U3SwP6fVjftONAr*?o$
zt!3+@?V#zp8@&tUns4Q9?3ryQpb@ilzfr$J$gOEaI|uYWwqIMJvW+=<pQ;T=w`1O7
zzem1-J$IO_qC3Ft5Z`eAI4}siv24}W#iA9$+4cvw=a8Aeq7@qXz4cipGT$?rCm(ZY
zzshhtnU-r2_Q~?T7sWw%eEG|hZPYmM-|@=;Yy|{}z`qx@H;pSgA|_I0$QjJy>F-Qt
zhbT^M_A)b7O$e2Y)#?6zXM{;mo^`^|!{0m#{SzE&y#k{)X6X~AYDA>}prxi+7tDn;
zZ4w^EcnwMfJlT|0JGVu~W`YA;@Y^1KspXk&x*BLkP1k`}OV-Q|wU-JaaTW7>=+-RO
z*MhE(Pq0`=KFYUJlSFm#9%5;51E@9HI$C|Zi4hK6%Lo+D)@r^qqFUWh#8TaCKxDNK
zI(1jCkZyQL++3Mt$Z@V*t&uESZE7%qb7v{#m?G*~F~^xqV&<vHa!mp3)Fy+%l;Z}i
zi#zcwD!5+9^L+b$fA&#cS*w1#Zx^0MlG^f+yDZ-IhFK_tNnd=sUo&$y63G97235iz
zCD4tTijT8osqotCn3}emT?b(H-P&D36BO#GW?A&r068-vn;4dux3ttf?BRvecH=Gq
zLiNokd`A*OaAJWc0q_<fzF;Du6}>kChgw!dx;D}8Bv7Kpn#Z}fzoxLN>un6n+y71^
znFFb{ah`9jehd#ZKMPr|K)Y^#s+MH;-4EJdkhedM59hF30n!)iKKfE^hVG~8Bdkd}
zfgNF9FpPQJY9a3q$@w0x?Y4JfE*ab#aZ}h5@CGuKp>O{_Ru?jqbseSMrv0$Af4bh8
zBPf;i?8w+LYa-X`m(f%ENQp-o7!KxhtO;2UGfv-%<rSUO_kEkrz{{v3R$}O<Ej*7$
zbefZ2ESL8%1rZzs^jbaYgVi?*Wa?lTOL^Gw>lHi{3iCh54>M46oxS&u0h6lYR2fCq
zJMX93U4=Rsqz=ay!6VW)7w(K(0zgKUMT#qpQHDjQqO5wX9GYdq@7SNi=q{0|eeTz$
z4n!jpSwj|PTxTMdY|)x|EoUbUE&g)KZF7@3v;c<f4pE}AqON}AaJcBU!T&~MYr5S-
zGNk!>hW2u%NFQz9`=Tn^<>ijV`*1q6bKpgi%_y(dt>K}mP)Wwb&9zw|NY8@(7TAeY
zD!4R(`aZAXNm7e99IkuT0KF=@9NaJNn?%tSk{I%>xYMjq%w^)fdPC_qOooN)erR2L
zh984cI-S;C7{80r9nEr4;oa4oE$GX*zqZL<w&OW<%PSG4eZ!=^$aVs2F;eUoTCa@t
zCbJ0N_EQw=sieKz8G*W8HTsp|wNwQFIP+TiN*?b`lv9#m#riLq#_h8~Tm@JIy8+8-
z3;XdZ2rIIG-OqkM{wuE}yhw^gvVQ=+qxka(zj3s1Oa=1#Ctt>A_4)3K;j(_5lcWng
zjCJ)c%EcOsibvAyH<NoVLyl&&Aq$%=KHfvK-Xk*FI}W(VXIq8Sca7`19;!bB9uf*z
zfo%ibhPMg~kKM?uD>i}66!P^fR_Vwotil);g|N<%&cz!8A1oMkFZbS;YMF?ZX)L)I
za4VDo=AO~p(T}p6u*OrLpNtxSXZ_oYyTh-vor^IpR<T>@vfRC$;?LGe?D8OU=fgaa
zTH_C&oQ&~`isaC@zZ+08!IqJy@AebiZrrZVE=Na<IjxfJDZbK+>lSJJ92+>#FH&qP
zo)5Km0&FaIc&L*dcVoAtlM8QmlZ?LO-9JYwre-*2+rEHWw!lv3cX{4si(gt$v?v=i
zTcysgcZX*yVz%zLK~@u;6sZjsf)s{6-5)>}fSUbRuK3ZsvOUX>eHrmVC)W0zpFih!
zNg^A2pJ~tg8Mq<KqxpG#I$ka|gUK)aTaFkvQ8L0+bsY5jBQ9)S1_f}KjGArMToxPe
zwrFV7Hbe;Jc=hNb_cH_#(g6X7e<WXhXcFr0##c|KFJHEf2QhW-{Ftz2%2EOYKiTRZ
z&rRVTe9+mUk^bw{Cs8Tx_Xs@jX$1bMO!Gs+l7Hi^<5jT2_YSj}mcQ&qXWvXg6p**N
zZonkg-+L^qvL-bx)9^DL4-d>6<dAR3xLwnP!Q)KS89VjPaq1{ME~J10r2nu!l3T*|
zD%1?tnXu!$W&#<!hMOw|w8pW-P;XpE|MmjFCA|d=wlp6G5#Nxs+xL1vacR^uzr*)7
zo139$e>xz^KzT)*k-KQo??3Y_(>m-Ph+ZP@sM4o3o|9T1cbMcn`^1Tcq8MC40ECBy
zeyt#NgA$XM(a^}{xnGXJS6oX7;+fh!&r6uWN1N^?%HWvB39HK-rNZHSa^402VdrOu
z>&Ay$RU=K5lBI^h?<mihn9&kHk?Iwsw4#pYlzA&+h&JL*<_e?1J%Hrcb{Xni_<_y>
zq9RNneYyLGx<A8u1U&tW6Cy^B=Qz)&)@R2Lm%b;#4>Hm$-v&S8F4e+D9(O{(2;|*(
zkjmRsrVokoH?MnLMMJBC;=)D3tIu{~=z#v@9z(|Su?(4QB`P#ux`x569=AGDYLEoj
zANOt8ncZ3jS|c{VeQ>z*pjnFF^{@iwM<8)=Ylu2c&pQG06LO?(EnZT_zY2AxIcgBt
zM7{p8@8Ev}dH(%56$*skA|&NeU8(~sF!2R6ofhe4w;4AdjgJYQYEHw2vb$b|4Yf5?
zeuGkr(u>#aUV%LcMpNcJ^TuATUpznS#vp>Z2sVkohQ;A~nh3l;j_bvq9T_T~)Yo<Z
z=Cq%sYP!g9NI8d<XkmqUauDDMe9slzP<UR46PjixHr=yt7HnI-j(J!bluaaf?IzQ$
z1H+0`5Mo(I4VIf-+rSo7Ngl(1tZR@}A;nfxTl<a2V>1g!1il&ibaUYAP7y{;Dc4#f
z-(6=Qk6Y~cWGq^y4xh(<OPDAv;87viAF=sNo$hsIjGQ|5<UI4Wc5#uy{+L!<wDtG`
zYk+d5u4g)WNxf~!Yvr9fAe-r7TVL??LzpSAnrO^R?C|nxXsjq&#}pj?+Cl@$<$Ip_
z0Z#|RcT>@}*V`Aqt0{*~3%@7J>jN~VeGv}Xy4^JPC&?sQb#29dDnAM2w>0mzk@+lJ
zXjKuC2fMO`(Ynaft)0V>0tHscNAK>9+^lZ`cs-<|YLI%8cRZZ#ZTSF1D_l`_w7XE=
zem6Xhbofv7JeCJ?)9T59o>%1G`{%!RApaeig(Z(-b9)cjAF(%2t)xlWfF$LNyPGFS
zdG&O{X27V&{A@fs6a-8&K_>P10IEbS!#U^;@pc7^Z4B;^tT-;W=CDxys4MwBODw)~
zWHI}h1kuXHE-PbMjU!%`%_m+wm=F=ch5)*h0GZ3;l8b;!b#(|Ww=(!CMj6^P+Yq7y
zse<@|yLMwdXAfSBz=dCaqYNo^>V>dhwjHWz*ZL=}A5Sm8T^e3rUKE#vHEu`z6jZNE
zG{CiWy;h5{4~q9yud_ccvwg6(s5f=+KXu!LI%L3WCUtiKYx8UOf0-@(&|Yz;VQ6?x
z8?2dNIR%fol&K1%Ibkb<bhhf$FB?k*md+7TKiiIUBsx`u0HIXg-N3B?yw1J^9aD;H
zp`ouW6afbr9)|FCm9ZylZdUJlJ!|8b_+Vim1CbxkXB1@^VYU*n?6-SvPWr)|G-You
z9>JcB&ZN*5%Z9Mfr#r!Ck|lh*S!wWe44b2D7JsZnR_tlnPnbP!)$`g<mZ8t_#+-Hm
znx@gj2N2SO?q7Q?U_b2*xpv{YgT?lbudUR(C($p_Y(^@&>Tb6wx~1wR)@@&C@py1L
z^&^WyKh!xhTbOnVAtzPo`teYB$COvk7Vm~K<;z@Wi=&bxv(4Ye%*0Y^OHUehqyBM(
z55->*R5=sNwM+mWt{z!H_>{$Lj&C7|OMhZgHN?>E#@-|(E3eUJjoWMnQno}{{fu}!
zrfbT#);bHB%L$_l)13Y;p3d26O<mzmqmu86i$%F<^*gK~2F;mZI=pF-sq(jntAYBS
zsWv6r-tEp7=c;C*T63oZIS=z$J%rWy>;T4t1-$06y9>Z1Vp9c=|N7|2V|+x3mp)`3
zdl<+fBLhIn<_&g*wyA(yXL6T}Ka@=(5zL_w&`Xo-X)tg%Ue7EpRl{zF)dsA!#`5-B
zdGNbk0Y~$Jj#ne{Nw5QM8#$4CPysEGe5YTl!SDvrb!h7uUY@L{&D5Bp5ud?0bfF%6
z+KjV>x}8BEFw7VA`T>V@>roaFtn0Q<A_(g3Hv>kZ8N8N^-)Lk<OgH-IJI>TZ{CmUy
zyjljy{CcmhG}9mh>;Hb&{>vxlC}$^&P~&~;8;e(5kYn6fVS<~Vy<9Nky?>ZuGt8Dj
z+xh+Y75U3Fw4ewGobu5B0@>7cAx~BVMd@)mpi}Pu%KKa^I!#;89{+HSD0rpU6Q`OC
zOqL%CQa`GDRC#sSQ(eD@vgthzs-^4gk>$?Hjztexs%7jtRdL+l!T9CVph}NwI9lDM
zNBOZ+*i^@4`RedUunE^{<h9A-&N>n$*BD-uOr}<_c~UrK2G7|5q|l?e;m)b+>pJo?
z7;Wl6+*DD!Ma$JRh;AGx(g=8kCk#}ewBxyO_fr#&v%YP-r{6V4<`s=yLHW`o{pk3?
zXDhPYy1xPm9?%NvULV2y*01LnXG4Wn?TcY;Zf~RLhFqtS`NYUIL)U(b50eL6jPn=?
z&zI+CQkV~8IQ**NPT+K~-8!x~F`l6~?*}<5eVX1KJ~5sj16=Jcx1(=UWYTtGuO7;`
zb0PG5cSF4UQ9E}APcx5CfB7BlJ2oAuh5dTyC2<=<V*?B!zCAey0^v#A6rcN!)TMm9
zCyhI2Ra1O3ES8_5{My!>yF;HJ(cumj+?Aw?8BKn&-I=UujlFym0k0X33ea*E8)Ezg
zor`foz&Xc6BqIJ35Oz-%bvx<4Gvba9WTG7$2UXHBXDp)COD=c2{t~Wa=74@r6iW*~
zcnjv;@i>#wrMNMx2rRl04>*o9(H+L%3xVeu-AW)iG*6D!MX@w|pXy!>;c$fIFE&q@
zAJ3L;Aem#izwmf!2hYaL1ZZ%`8Qe}ekdc7w6iA@S_ci+^Phm0_fq{C4glpIbY*W55
z9E4^$9|If%OvvvB^5r|TZ7s0+{jL&~G};MT;Pv&~<@<W)dq^A0^Ao6J1u5#wQqXaK
zf(}(b%_TB^@vEy6Dv~7h)K{`Y!sV6*O{wE@ZL<VAk*kxECnlpfs-u1`73ej<uA(R%
z`O8rt!Q8k?Tf(N>I7fdFt|bHPts$j-cxoR&HlRD&vZl193THM?Bt3w9z8F68-S<_S
zDX;P6&bwc2t`paibljvn{bg3+f*vsaCEx2c0Bd5S3wU^Yc$hed0cMy^RTmNm=&3Ip
z!h9K{Q<B~#Ak@@$tYPZ=XuU+BH(Jq}m)9r2hUWQRj&m}$Oa(}85AuO-a=w&i)Ar&K
zW=AF8=cLnYqr;tKHvWd@*H$X1Wcguit3C7aWhl;b)~#G+=XegP$1Os%@!8cEY<%I7
zB?O&gt)lA1Hc_1L_}6xfQ2iZ(9w$ChnFI!CPVyhR<=s^Nxdo%B$J5d~S0Xg;csq?Q
z*R_COk$LLF+hhl$uBVLbn5_xx1Vqf`?HhK9GQ6biFLBdTQ`b_>J1S+SFd>e+Km;CT
z@CP>c%W&j2*QWFR495lYC5M;GNCKnOP5<Pa5_{h|+>0}#n7NQ1QE)qS#qSoTMxd{e
zT+th90XxDs2pT}>HCU2qIi-oi*sUh(#i*2p396ZeE;815c_mmEDq>iD;y!?BK0?CU
z=o(2Hx(5Ifiu_az28Vn)P-@6nmjdR=ZNWjQIM*_&Y~%S(%uuTVB6;eYQqBF&U~;Mq
z)hUB1(@{lpxlawqd|_jxD`4bI;ZS1e8L>DW=d2IwUKJS@xtufg&j6y{$X2ra2^m|j
zuP1m2;SW?Ca^ASE{$rKwmzEn1%y^nfa;%lQ;dEvQbM9yLrl~;B=aq#jp=YCOGHHan
zBWgDVe1csb%@Aic3%H<1mP(fz&}1-YI&Y16FHxmO3$oCUoz3mYI50eVLBsj&XBKql
z7Z>Q$rhc5ujf}gT!WEVQ2*as3r#YI6y-vEeT$zvh*SRECz~C_e!6@}t&6bOYIU7`p
zk9e>D51SF6FJB3b#&FL8$id5>YWni>b|C4?r$F%UfVXM%B6E%Vo2xHQMK^N|X3(jj
z`{?y1$L^p!bg%@>AEztz2dLfn8wZtaEzUu5e&_?qNOMoim)C*lPs-=zBP6TZ_}Yv*
zZ4#Vhc+LTIxJ-s>+Zf<KMj1=@Bj18#79XFtb-gW9znssv7J6H%hcyI^6R6){U#4;l
zZGqIs0Z?PC&erSvyJ|HT26mow6xuKuvB!NL@;G<Sm?qp0qtWUjU^_~K5h?(a7nc*#
z>*EA$iRUU*!DiK~a+-BtHm4+Kz5Yl5clZ(PD6Nu@FkmvZ?lJ$K(tI|i_U@>N1lwh)
zfh}0y3YpWmWR^b72JjjPNX5Ip{91-o<^ga|DsA)d!*Q^qQbIGkb5*cUD^lOJu6bLt
zI)oh4hi<J>$oQn<y@MLZ=i2)=Pv_x>oL!i%?to%EGZ2KSQw#Tk)Os0#Zo47hY68}N
z2O3Ucu?t!Jd~r5JoxVX>?|f|F@^?*s=W~}~9~|WC@DD}#4QwnWWIInw)bo<0crc|k
zJ8Dx+!M0pv1pAYh$L7<P-8=5n+*))Eld)J>?^`~b6nPkx$94SXlwnxnO5<oYc3~hY
z?ZH1AV2S$=$b#%OP&6DaqkVgW)?M=LH%m>?1-U#dL`!!%!YevXh108p%@}}$Lf$)O
zL0L&Z6j_r{nOVK)$5NGRJWCr`@qSJgpM12BlDCV-KN%)eq)ktXFc21#lCV3=8k$+u
zfyyniKlZRyp+K7JqZisg<FL_dJIn|J@dWlt|F|r~R%@)wpFsDZ2TzN6a`AjylCcbp
zTs)r`w4Jvmc(+ESe9?P6TM}p$SzE6ww1_@D=sA+8i)uN;Dn>4vch!yT$!O35Z|QXL
zwUs>Wvv4bGs$(()xNRS`yaR`gDDPWSTi3^h#>qfdi)u#GSj9^Y6<wm9#j4%mwyq);
z_!=Yf*JtK2Icjqo4(lR-$7Ma|9gcev1%AU~+Oxh|XZ99|WFWy?zwF6`FKCP$x@`jl
zo!gO#EPtMmlKmh1q$`q<mw#XY@guyuz-S||9}CYiRuW!W_X^KJUU|--1{>eqoJJJ4
zu8pj9)OTJVZgu^xr>^*MKl2c9tR;p3kwK4QKaJ13LX0O@2V%=(u$>@$-bzG3ga8*B
zm25AW%|b=Ex>DgYlglV!nh&|cU2RA>NU&|(uGM6z*W7vioc=f^G0#uQsUqnIof05a
zbz=itCNw&zvWNP9*@m=(k1)T^QFmX);e4cNAE&2w4`bCr{8yer$ehn@FJ{}_KEd@}
zx|@uUZW)?X>~clh+~<{=xW|0uwAMWdE*5j+9%zDvZ*z^zY?d@FkZX{JsyY^3;0`YT
zLMio9AZz9&lMs}?JSgFCi~|^2BVHfl`j^z>4w_S|dwgpPCMO*COVGC6TCuA+yBtSz
z?3cV7h<*bMl9)WMJA_O%zt=OiPOQ}F=#4_1chRYDe%pLI)uQHfWQI$)y(5l)TNzJx
zons{1V0|;*4JaUKbm*WBNx^IQIz4|OL2~l2J@;ZIg=!1*f8DVEZ7p25@~(CHMd95y
zF=;<wKLdce?}b?K>qA0V&`)6SIB2w7j5fYC+^|3Nz&JO&f@+x{BspTVAs~hM1`4hS
zQoS|Qg>4`&_wM~0iM?CEb8<(|s@A|X&T2)+N!NQ!-LAdrUVE2cdEK|K0<RhAV6gR-
z9tkg~&R)JrRi|zv*(^4|Q<$SJp%%}1f9yKsFh5S66R#8B>41waYf=oFyKcvxs-Z}w
z{M$N7r1RNf!BJP?j1o+OVTv`8N@QN9%zfEjXG&Zi-yV?N>U_%K!05DGFS0t;nZ?*w
zZO)A2KS1%Ytu~=DU{U!^vB(|pR_OAFJ%ao#opMExpa83p6%DY>6X#ZcxLlS3$wjvL
z$h>F4kC_iHs7<*qWk3Qq{nqHnyL@RyOz?3<9LF%VkH_Ov<HZnO7}jQ(+c#ksgQ(KR
z{qfr%A9nSwd(NhL`|7O<vFo8*?-a`~+co3A7wShN$kQt`wRxVw?<Ht*nO*@Vkqu&#
zT!G9d;6=l?XG6xT2+UW=T#@-#4<bhdE7z$>giOsP{0}|J`3IQq!8+9>pB4S4<;`Ys
zW>Q}{Ovf~TpvJ!jY3P&-UElHw%qEZ#oxF;4x|+CQ6)T0yY4nrn2JSxH3E?#tp+@lD
zqqslb98lz@_tx0(tmwFFCk`&XP$lBBQfciAZo!0MZ-20KFwD&}38iGUxX=fbI$w=u
zrH&xAg*4(Cxky&kb=(;}qsN%W)vV`1N4~wrzOx^vOq;+G{S!DWwULtNiM6R-s}G2=
zp{lf6>TO-hs*iZi*UHp@N7kv)=#~9$yF^yR+UJ7Bv0=sW=PvkQuCXK9UGT~Lc<!NL
ziIQw<M1aKg(W2+=Du%L6M=N=30bpcnXuhz)e|s7&{P2gk#8e3nGYAO_xL*N<6F&gj
zDZ6{4J4~j7eW-f&XTpFyS`9IuqMK*tq2i=xKFAg)=!q;Lq)+Y)$e)la>Z+~v1_Tfy
z`7D2}do2fK&Xf7Kkugq?zYM%Qvw4a50ph;RSjAMJ42BbUX686NeOG_`uj$_VqtC(1
zlyY&v@;_3os4yS`bULN$WPP&|Ax9E_!VMo9b$Z^%iJfr4U{^>9m!b3J^sSJJ5!^GI
ztJ?b#@kr?Att0P?UfNW;L4R+1AcmAN8gaMN_TnftKYG&){sAFAF$Bc;l&Yh4W(q8w
zW=>6-PY|+Yqk9<|{~-{V-aZ|Lr+4mAT0B=~GEc;g<o>B9U(iLWj&+bF6cG9W^9NEm
zZMt59TX1u49*&|$$6BQ4b8T-pk6ZFYL&Xh;GItP_ax7FU2gd}9@Xb$B-jR$7x~0R(
zZhv_CFq#dTPm0-d$1{w5zkioHaHFE<DDVJ#G3&#;Xs>KucdV}aW_Q?a8TMb^Q1Ir?
zj~3KlzDV3?ASqyRvA#e+kebmgO;u*e2W?s;awlz#5>v(Li}OlmcrIbvudQ};xP!QM
z&tb%w<}h{avmB1i8Qr`eYO=U*Dq8z!(hj*CnRQRKQWBxMd~CL9x){~(@d=YXB{pw>
zP{dwFC&0{roBOE}*KEG;ZVOeqd*siaUakn{>&tPW0?{|4e>|%YR0j5*)awsFUkl~b
z4#nR42{f*-iVPw0$e#qL%A4KG4kX_5D!ASRWNEyMi&04X#Aa(G{VPke+{bA8eV$gx
zGLM<pkNw9A_*+eS6X;89X}I~OD0avxkK3P;VeF6QCm9(#%)LNbO52<4=KCPi&OYOS
ztzOgR`ce8G$MfI~%Z^q51f2+J8cmwiN@ZFIE44j6YG6kmkSP4vh(?W}g`QAt8e-sh
zI5P64*}+Z#&SX}h`^S4CiLg<bijclUpHiB-c%RZzZ)-UfOe`)uV6-BAQ7aLR?`UuW
zFMRFR=vS0oKHsTQLlCGkN#l%`Exp+5((K%t)6B!Iz)#~!XdM#?JNq_QdUr|<bwc66
zX8OGc(B4>|5elbDgi$Ts{6VC{8|{64+5cMlmXZo5X6k4>nT0pi%@QMD#vx86#$1#f
zf@8X~?N+1GGEE4^;-rgbJ`7cMMqQ0$`Q`+o^~@I&a7LOelux(TcidjK^4NOV_E3F`
z@AT(F(ikPkv$0qtVCC<1#(%+I{^MJ@7udw3RJ<zGcCUJ(jqA>D5i+%g48bNADQUP*
zxriPp<@CIwrz;H`>W`8=jv-c8ClAm`Z!JaQ`qKk*rt5m0sLS`ITuIG!hC1E4=9+$D
zL25QURGB9cShW|?)~_^81B+fcOc5$!<xzVj6HLM~VSLJ=M~B}o8xY7^HU~DAEu;QP
zOl47!H($w9Xm4-=k;0$J1<K05Qg2+}LpJVWBqUyFfX;qW^`X#c8T%#2l;eCTvsK8{
zEK9zyaXk}2c05&zwAm4Z2V#;a7nDM&<sX~v8Hs2S2gV}GMM{*L#wTad+)kJ35W5x^
zLh3C*q*{%@umlshws6dwPMXJb@sre1D&rJUrb(OdW5WDEr;nvgy8z6*eg{U~gX6<t
z|Gw5tNa!P~EYOzomABA|Hm)UrFc*rIiSQ()*$+PpooQ2p8ys68pvDm>`FU+Wdv^!E
zU1(d51Oh(>My0t+zRyzV<1<qrOD4GXvVi%+aavz~_Hxd-rm(Sv2&k)sVe7-T=Kum}
z%utN`s02*2aN*s0OGFqpL#s2z7S@Tn5=v*Ujx9J6^ZM=ARN9bO$v+Qo*N6Y&MUVoy
z%_!_YPV;{OyFb7t$`oPRxDm2Jr(c`tS}am+dGg>QVo?89)kb}43=py$!;RT^F|2cx
z&no4<>3G!T%wjFw*Un5@f4;mud%Doq<sF8?{AKHS+2ug@IifQRTboENYKz%HRqjy1
z@$tlxXeqetM5HdjTUDUqJIZOxIgAlN2iE-s88GDFzg!Rr3lJ62Ik$FT*!0k~(_M=N
zVA^IPLp5$c(r(~$O!<xT%{%I(vkVsxkE&DNT-oOV+o(uj8`WVNhpd=i$d#WCG~^^A
zcpVuU_((!>49rF{_WPx1A_)OA&2_M-<Qf5Q=(I+3wDFyVl92F>MFNpTuE5~FV&(h&
zW?8&^O-q~YX<5X^r;CM({t3rMEe=><SVxr^UtDNW_3#!jnJUs*r+9ZywkctTL7;fH
z1}f9I@|}BreC`@t!bBsJw7E#oaji0vT|6}UoDYOY0<gI@qc^3n*6k#sWvCKC18KT9
zune0IqrBsAq#@MW@BnJIl{Hxqq<{-|+37m6RLUN_=T3`gfee{F*b<ODq{!8o+OENF
z2PYW(Sx$#O`5de#4`~x%efUS<+ZVPF)*eL7E3IZWhp|th|Ae6AWyDyD3CagmB$7WI
z6zZNC8JX;jCUMiz$V$AnbpPvOPRU=7vb8lHQN`qI9u3pG!9!V0*3t>*BJ-Jx9Fq+d
zOHyuxvxu|JwVc9XeiTuhKWQLWZKc)^p7ps+cN>Y?ic?c0hH9P8-))&S>HT6Fi_PQO
zLmn|}yAem6E+*bA0B95G?1RH%MJ?T~l^Ib48pS`2+fPeR!V64KMOckwwF1W0G^r-t
zj{r<j-^f)PQQb@f6;7?Pz@Y03Mfr;D2M{jGK_Ec&7BbmiJ=9Lze?6<JO%yxIdAbE+
zEb>JuNe1A^x0((4`pflHi7eFULV-)xgQSL=03ib*&(VUUys?Y;z32iU{VD27YM=sm
z2%w;6eB*f;-r=FW$vEQe2z)yRb2Yluv4QdUYHXL?P>FV21gvj=h~;{v-Z*kU*FqVk
zCpGS8+iNF4Zfdu!0IjTSD=t^>{qe7O+E%b?drdgczcf3~@ATa4rgQ-Js{c&+`=5Rv
zD+IasfT+uTxDyn;@6q}OqJU|uU5!m+%-A;!mMNO?-Xd|FQ?m}|HCoEJ=olF55W!12
zaL0@p(~xtQnwaHwL^kho?M~L(I>XtDsm<of0Kiw2Tb_-kAPpcvg-qs@!F-2>)(-eC
zHC?HtOhL}ZcNGQA<<v*k91Xdc#1ymNs)<rRt5KQzy1eO)BGyOlFCFWxnojNqZ5Eo#
z<;40Pisiu#o5Khlk*DG3TX33FbCp@bLzDHg#l4JYv?g6!89rxXHilzCUYxeOtHw{k
z19&=$he6kAGKU-2BN}R#T#~yeB8*aynJC~|?Gd|GK9-hNNFlge-PQc1r^>lXH0W!M
zjnX}1qCEnzI7G{~547$rBTk)}?&-wD#bH(Qy6MNA0es8Pyk3pEhvH0rIqG=K`T#Qc
zQ-IHyQFt*m=<wv+U`;sIWz`|PAC_K-r6z0tZihb<upNVmZX|Ke_{1@Va9EyExbvZ+
z6|g@Is9g2y3s2k980_Y`Z3O-Wn2U?(`|1f4+NFViEPbh%e5!8^6_X<0KmIu@{4Zq_
zsH+A+-X~l2A97$}ZZ-Ff2!vNK!6L`^7XKn}@E&1fDUeW){#JGJlkfij2z$$@EW55<
zSV~Hi25Ce<8l)RRx|Hq?rMtU3q(oX$>8^|J?(UNA?!CC5Z-0A_vES!*42M7T7mIbS
zIp;BtFlRMBxhnaz)rcQJmO*paxa%w2kGD{nUlt_FHF|!q@1w5X^cq{eFaG(8($}ua
zsXw8g?nNW~_Frj&4@?<Jl2ZC4Kq=ex`d2RP2TG+(CTpKp*`4~<%liv+emfIuq0*8-
zN@Ka@u+8BzCQa|e$_Sc@=44K`w?4V?FvWJ7^r=RJXL^e~_A3*Jlt6sbk)A`pr~*9+
z`1au4+wwS|lY9w5_Z~26zPnZuD|&CfMC8?{crv|yNSF4=1{#zp?YW~?<oNXE8?b;U
z&5=at*zQ*0i~O=J%4-^rJf)pRHm@n!WPbf(YrH3nTvE>0qDClSy8Rq)!JYE+=RSwS
z+O7cm-QUaghoe@f1K7t9F@^o$Uz?HqcQr??v{EV5wAG2dDBgbt!ohe+^KjGWP3&2r
zdu7GEGZ}t(|I2Cf@qycGO8auPZg)QKGuf;CIc}hM7fl{=!^Sl3uz#Aq^UPp-r+1|c
z+lC(g^4ST@kn4Q4Zq&1yk?(9{@PqV-jaD*l+^B!jKR}<Jb$Dmi`YD@wl-L7FkqBS2
zhHlhncb8##p`^+x;nzrFiRNshRSI%2*|G!tQ4b{_sS=Z*G$R$Eh36wL*XKnqC<?dR
zLd}{Cto!8bKh5}|UXFl3d~*@5Wk3ce6ho;FNJ)aV8WKGw;aXzv?Y`U{e?@f4-c47V
zZCr>nEwX+?DN2V!lA?=4UlIIkD)Z1d>&F=Di^Ndwqlb$m?V4;8ntpB90leyWTyl_7
zi~2u!hAVDaT&9y@73WtL`9OaPqWxNE?QZxo(iY2v*Rr@8S31Y*Fthou(#bhUR_1L{
zrc24ZfcebIa!bh6)c0qlQE$AJwXA=7K0=l&Grm(Ha<AJSfL@jRD_1<dG{)cvcK2m@
z**CRnwzY=K6XA~PC9Ntyh};+6FQ144Q2MYiM2_$nXg0$ZIe>Ss?w(414d{L+l-X9E
znTzbtx=8N!QNmXMCg-ccXCpuTG3<BNe`@pN7!6(MU0NWyoPRa|x?}Lm<A4o$=*nC@
z3+DX`kPK9<F<y_|x=!r(MJl0>OYYsN93s}9ZydhM6!@CxQ)b>IA%EIPO!$17IU%+^
zA}KHMji$XLgDQJTGb3L^h6Ioo@1g{|UlPU&`w@HYVaQvUsuTybD`X=bUCb#-rCInc
zXuZWoLh8(}lO_FH`yPkk&m}x~@94$Y`@hJUP?zX<D+bxF;&h3O6|qW;DR1|?400Yn
zVs1?hN9S$VzGsGMtsU{dznyTU-gVlgqxr!S%)jUpKMuR`eyCJm_~-9TQ5fL3f@5Qg
zEFt<3AJGZ0sQM40gP(8Ixg8W5FWv~-wjv5oDP-MunX1oKQ9#b*9#|`#%R9v#AwE(9
z_0mar$8E11c#i#qIe4F2olnrqWT!qW!m13RlGkWO9H;u2o;}yLTf~g4JZyJNE7S4X
z_@iSa)q<FJXSnoJU2J;zK~EsvH+k<29?tLXB+5a>lGo4TFFxd@?~AwGkSBa+wEn3W
zWJ0J?+8O|e2OQVy?fjyb2mTz9?yC*sbjhA6tB^sA!$b>Gxl6r6tyF%jGfe&JCn8LG
zl)&l(5`Dop<iB>E8!64vXui?ht^p?OCx4nkAswd*r*HLgv%fNPaM}AaDXQ3+yq(&?
zI(jJ0U`zlExf1&=jhaFKcJ!;GHaA_1z?adUHs9piMc1rck2^$2@yd{=n!deB_Ck%z
zC-;rtFez;%6Ll$ZbG8#~k(Xs%PO{w7yIK2ojFr!~lFz1Yr}&ca$9G&_qH_m2Brc;|
zW)hCtr_SSO-5-Z`Ymt}f82clQ|FX%iW}q=81I<h4WZ{nWWU(cxyte4CpGwFiA8rRZ
z+TP`qPbnv2rg0Ml;}C6|YI0A3q0Dc3zz--tBmEtu4Y>nn%$nCC)sOBk9o9ECv?=M8
zI}U7T_}&;Y*EufPBbVxE{d(?r+(F@Xsrkc7NVaXpIKh0y!kQ77s`xd(M<I5LdUFtL
zxXE{UP*1dSpjFpZ%7I_hCEIN99>Kc?an#>)Q-<5)ZX_U@Y=C38U_IG%G}vQDfx+d%
z4%u;$C)vYA>&V^Z#;(|^<6>-v^FgQ=@1I|k6k5<-6O|$L$LD`#8h`o$3-GJv(_PuR
zA3PWR@7CQR0?La9#@_@X$$AI%5skh&QKtgT<o?ay$UU+UH33IA-hh^HGwTRATkhu!
z2-aVpjbRsPnG4<DknetDy^19IT-Y6zlQI2*ir29SN56B+5iVZ%$?PJ{qb$?;wENg!
zAuYB8a9?&OnA3sTT~P5l%a93`oW{xYl1o0mwr3$-RJ^==ByL7Jw34Zcej3m*C@D4S
zoE&%Ysz6j`W>neq(cdjOcOeyWu?kAe9!N4wI2QDM%acaztoW+50x7Jh)-&h7)FAVC
zn27)jz7JEvb#_VxYxuI3d=JxP&#~79`=EGJ*#!HH3~>EahP{g$C4_b-g9)Y3W@Wz8
zg19!oTiUMKh;Qdt)(iI>earr*f5ho~ry6Hp92eAfKkzL_9O9)MVhzO2GYBLI0#QDn
z`_aE8TV?6rCRyHCbrD)Hse5BonrsAL^5BR>UIaNBr+xq6<ty+_D7EsbdI@Z!a_~L2
zed2Q0|GR9I3H9iqT-uW=gP!xE-IwELyzPj1GGlyP+jg9Dfrj4~_^2ZFy_V%byD+W8
z0xe_dQ<zUEl|RgvnNfu8cd9)1hn_^5(1T;L2}M<GK%*dxKrT5c|H*QBdgN(m(`n}p
zy$FcgnVyeZek{Jv6H9Hf_iO5IX3TocLSTS$ww$_e^?G8mlF+Gb?d0~9(0p@98K<L{
zt=^tl_wH&vsG#CSVm8bZ51y1@3fo4qdOVbVl=7RReqw=pf@aPk<&uHaZORs(c;NEK
zT2V@_DuZ##;$5KVTy4DV2O^_e4JnZq1?6UFfP?M}h#210u`+?FymYN2%$f)==7Mgx
z9{ddI24|x3nE_~rjY%P2#|R9K1e%vyYQYTW`YH55>%Flm>2OUc+AF><E!9i2fC3F8
zpQAETTh8=rGnIhgazvCb3~!@lBcUG5UPOcZOH`?87S;!@rS9BAm1Q;_j4z+c)Uc4i
zlD4gjiQfMf2>v&`pTryKLo?tnh&Al;HUIRibJ|>W=9u0`GH7}z%aJ#EPXf2#m@9*j
zKRc37{Nm4g5KJ51?<|^pAa;it<St$xl1}qHz$nt0-N}T4kpjYl2xR(UY|)!YA|-P3
zVL7vqQIIX`8(~;m`&$v~Aoifu_Kymaa~akM!WU#3v1P%xxpQI@xmiFGN-_9hWM9eA
z9yuHcJZ~@JWPE2=HMbQgXYj&%(89jfZg|{R%36(rUkch_W+X*~ah5*zVU^g+>FAvT
zZGO%IvP!HWw$N<1_Sdq+p$X<=oqnW3{dy)2Aw?nOzod47;<ufd8eYcu#8tCA=L5ns
z`3`putV+`aw+KpppY>`$SCxK&D$%x`OZ@mih@~IZ8R<ygG-kF^gU)gM5e5u>pj+JQ
z?zaoFRJZyhvDVq+UL@YqM3ByA{Syal@JG_UpIV010DO%P9}cbkTRZ<V5VIf>JhEdA
zpeB0%l~K=vC<b3u;eDCjc`K=7>KS*C1H(Dfop{alQfm%dbxowL+A{K>?AXZ8y%Dyt
zsR~m`AJw8DV%L)X37toDyEbeTE3f&di?e8$@b?NiC|9D>X@RE6k*y|WRZPnxwOS7{
zp<HKDmgMAo;_yk<Wd4H1mTTeM#Pu1QyWCS{YfoZR7qg-cZE#H@$rX}S0CI={OYsoL
zY?6!pYOt6Fo8J<rQj}41R#owk;&AFFKZ|)>FG(TEiF9wMLx}u^CKVuvydy}c6pzP&
zDt3^iHhSsIn`{%(E3VE|KcICl`s5)D;3xQ*nI6&6qLkOWli8V)2_gL5+w-;qAltxh
z#l~FTo#Jg87ue3`vi#7+H}vy@<T@VE8snAdof^ScnV0L2z74GzlpbdKLwlgr?bt%E
zDJ9sRzXnHoFsTCd59%Pm1AB!dz;mV&|Cd9jW`sJWI{$3mZ`$p&2BR`)Sy-RzJx{Pc
znCk62RoNCDec9UZM-=w1#}+HreiPeqsiBnp1=X)QeWT7TM7MLr2PWRj`JZV?SwZEF
zmN*ah%C(KV#q2AWON5{_bLkd5=GA#k<-J)Ri0q@A_W&no^ZZ!1t^aBTo*Vds<yrRf
zQm61{$W)uqfhw=^AwJ$qxpU)OVx$UWp=Z!7@~*%xkI9-x9Ypp>wRj^4w&R<Y&3Rv+
znj>Nfb>W37)y87NoYmY;<SRQ28W-|e^CK!yK%WqS(W+K(IRWBG-Xo865AgJejXrsQ
zZsUzwVlmQ<J_hci^Qo^Kr~PYFhrm`w&no7^Rm_m<l15&8cZz+rsGb%WH3bL-5?rk!
z<GY-RL29bZ)yY7;2|&T$+nO4Bk<VIts2;bCTUPwkJw?$%kK7N5mHpId_-iv|81HqS
z9G}PM%r?*O8Mr52++1!E-l*cvSHt)C_KHcN&uhOmRQJ%0u)v(&_d}#+R-@ALjUAKQ
z+#`K29j1STx&?#Urd%r`0XKH@#lCksu2|hK>@4HdW4(>uZs@GA7dox5`KT_P7pJk0
z-@(yG<?J=z4hs>jxaHE($3^6q8>w2)7zX1gU^0fB&l%CI7AT0;JBTE^JLq+YTU5)L
zUe9H-HHp5erf>WuD9<2Fo|_UsjT`qyN|s$9HGhPHYSm*6g$8(Or*LG3Gsrz)(}e-b
z@44+=#aP8qhM2uAKB$N`i1VJnqs+l2W;V!fEt#DGZhwx_Axc!vSH^0?eD-^XK&l<j
zc`Sy0XHENZ3@m&4=M*P8qu=YWW#uqe7wY_AnXB7_PW*7keB&4^hOm*QXyY{>gl1G>
zqIj{WOV-hE{-1LehSj?z{`T`k<Nr5P`EQSW-dos$&FT;5ug#Zl4kQg^iQo?po3;;}
zm%DX2fZ4_s)8uay(JB1pnq#)_CX=I@qFG^383ElX7bes6u2V#eO2ueZ;WWYICAz+L
z(J(uctj&5oIB-Q;`xrmJcaI4<9+<VCZBNz^JVN8Wggp1Z4$76iTm|fVMJ~A#=gkzO
z=pmnqiZUvj@wH}dyDzJrH@5W?9pnpjwMW;R(ZX{X#+nimjZW1p{^W{&Ij)xn!|%@M
z_^-3zs^`Bs#owq!(kVBpSJZLs6_oczl5nyi(F#LvL^k7Fg7d=;k4Cq4*gYG|CNutR
z6_;Bg`!c}!LKPk`t9_tgPD+fzZ`_$>&U8Cuc_*H2zZ_C<MgSH=uaF#?u0=d<ODr^>
z1!LOuInr#Gal6&jjz1SwOwH^{{ZD6u_D<W=Ii3io>Tfh`tW+<@-LxY!739MRbFUEu
zeHTTQA8GKCW41aaDp<Fc7%%Hi)^1D;8>1XgMLHnS4p*=CL<hsBqE9F7IEuC-#-V(5
zxwhBER9UNQ<7xdYH~EFHKl0025Pqr{%+5kpY2wo^QC;4023jRhY9xcuvvM5vRA>%r
z$R5(N?gU|r;={Jd37wP+PXN8%UUwWQUm4ARU?pg4RryP^C8-^H1A^(+EbFm*ch6@_
z<w&O~pe@*{jN@9PR?PY&q3NXoaRf26t@@F1{DBhAMET^~UhbpfJI>?1w#48@uOb5O
zV8?>vCVAs(6kc}boe$ivywASS52&<XMq0y&J*C>R#4UJkz7Rzap)4dNgEuDfQS%*V
zbm^oM!qq+0-qNd;oZUcz<%`=xsN){(!TzvZqm44kG&A5)mDq9LMq6agHy0;v{p#uY
zj(r8~CiITDEBN!@;E`5@n#Kp{8TqcZdTaM~E)>PzZ-8$_*8DuYNNGW=SwdCd`8Nlv
zLbU(_)ihLC(KsH9b-HBa&9cbP8=jfEhgS9q9WZ(aYsuTi{&A(6BZxV5xdG;D65z+x
zPGbm(g<kY3S=}G&^5=DJ_5YTjZPAWzsp4P4!t^tvnF-529Mu9}POgoM+8>jBe%Xhz
zjEb>;Rk;kFh;i+>i>_p@ukq@GWV7oKuTe&utQ&s?EJSyhtf0FFASNLAuRS!Bj0LAk
zoa;y=X121=F(e{KO0_4SeA#^TEz^{)>Ndj~usM@y-hPn|gkUuqJ-a<1Z-~)Y2B~1O
zoge=jj{Ww8+56W|a3gZvaRhF3;7mDh%M9E%tPf2dcx2Y!CKz`o-h%K-53_}xt`=V?
z0uHZbM;RFsCwHzC)Ftob+brR*!|16Pr+pCX3bi-?8lmx#y1~?hjleLV$~=5EnKe_L
zJ;o_=!=8|w&xjHzfW6;K0#^Hb*DtuZe*yR3#(No$9dcaUJpwpNrc#e8wjYq_YHgQ9
zCUUj>_J3A|TyH2eUtB1+Ocs%8(C`UgR3IMRR|+$EW)Xt=OW?lF2Q<Te=kv&K41;T=
z>ke-m4`0k}*4z)4*cZJGY_2lDehC?cG6RwrzIzr=&8k0E&xZIdTcSw+_{m`H^SgX3
zWexx{Ut%JZ9bgja!i_DS4))=gRDJpE@iAO3*%TdbT1=jyk17`ah1lSb$IQ%1t67IH
zs%{_;4&U8yKz<Rm`*G0=N?y&T8sSuoQF0ZTVl5*1wt$^Yn;43Xa_x;{6-Fb_WiMw0
zo|IVYkCe;HCVn{qArCEcmbpQj&x_sZEC5{&DA|A29P%-UMA)r@0DCpzpNmVSnXk>~
zC^k6>?HOr5`83DKy0mk96}MlMm1L-~F2}3)N5NDM@~PjeYGbb&zWC2mKZ3Em!0Tf4
zR~gqvguvk5cK4pEZ;lgtg}yggBFB5XXEuw;cy*ksHlMoZGMG&u)v$_bY0rQL_st*q
z`F9GzwZxFjn)<5<ukJKvTe)eYK7X7x8>6uja}DvA-kS9%!YQJW&)dhxgt4m&)$vOm
z3JHY*)zGOI;~EeM6hEznxMx6XMN*=Ioqbt$){z2|L|0hpav-sp-p$_0C3E68<U{l7
z#OUZ~f@mcN2eZvjE?j;kw^zZAhT|@TPD31RUfGi7pAkLuGrvLPSWWRr{O2WpIiH?5
zHqx8}MnyUV<Lb`#6@`o)Wg$LrKK4(d?WiAtwwM{EO;!J@Qu`ijf0kUvOg%pyYAiQ@
zK0Om9_8LQmgw+i_&~J*^+~0wFI5Q9t8C%7FtQaPSLX~PU9Q(~b7IFfdZ<?FNdq-OI
z^8V7epdx$keZ`38J#cQuPs&ueXlJr2(kjZ|b|XEU$YWyH?l!fs%PF#XAM@DEx8STP
ziPO?#Dys4j%jf7<XiM%4iv$a?Y}!XoRRXd&mkKywWhAuL8T@!Hj#LFhEg`4>0}Or%
z2X%en42|!$O`JNgL<+PmyMoZ-ZFa@paf(We?sKS8VZ$$i{^r^SpT2(YisKNwBu`FQ
zeO#ld@yT?E1Wsv#7wB<ph<h`28X&~4I%T1w5e$rOMyf68p~<_xf9xeXuL1$=-#N#i
zwr2-1@x2}PX$9Mp0N6b`($s_{C-vI&n=cJOQah+}-##q7oBXa)^wzuV&2n5hRDbf<
zj)(h~Zb#lI3-w8T`~A<zscez|c>}!Ap&H5GMXxygjaT*zHd&iiXfXEJ`f`1G?Nzhm
zR{DVpk{@JZ)6MU<K4sf{MWT`F+=jBrpXk><2lZpx$bF`KnTQ3g-U(FKuqP+b-TnQR
zq8`;|Uz>;c=YR~Y2cz?O1mWj>6>pfs?@UCbZU|}6sv+XpxO&$%+<0uXOZ?%&=tmN|
zHe)$Gy9pr{T&qmK<h#5k^7m_&eb#6e<63~oLE`uTY|Vf-X3rW9n%p-wYD6a#!7f1Y
z?AX%(He9NP{*D#e7k*QBD0My05#&DU^kRy(5*|sS$h<bqQ)R|5>KAJ{SWpD!doaNG
zG86q&0%s}BL!!-jliV^&`n8afMVq)J4ZhFilbk$Gna61b?iKJO<HRkThc;UN&$asv
zdzF(O68L|7xSoKo|MyS&9s1Ej16&u8t3U&c2w8tFtB>WO1*}aO)~Uqh802zEaCV^$
z>-6I+8s<sW{<He4nyH5{$Sb3^^K%^X+4qybkVO;l-za|8OETSFMF)7;+#zjrB>p#{
z<t5KhKhf$1iViv@71|ywzN<(RWG4_hj)Yc9Sun=F_yv6|R{e9mgUuSUDN8U7Thg8=
zn&-^;a7NE<?a&%xGe(3$rK5C`MD7RmL~A?++hsb`!^wPQYrjGX=*2M^W)j{8Jq?)O
zlq8Pd1YT2?m?gs7Zi-mThyUasJ$4&R#^gE_X_2y=+?fckS$nB|L<Y}vf8?u-VCxQg
zYyK+p(={+OuIu2GI}!A44mxip4wvV`foeQ52YX6r3~kuqt*i%HBoaO?p37t$6g*sD
z)JHMzR*qPxM)>Icx$By-^dDYvUJ2~COB{E9;m(M^aa6BKqV9Jj5!N?EUQQYb?d<z5
z*ECW}v6_-DTa#Ii@PH0FusQm1PcbyEKZPCs*ZuQ0)HRJ(u*DYgFi=0x_V6-Zzym#A
zAYf2Vb8Exvv+BnEj?TKZI0&=4)TDZ-)Ep7deCtRV(j3W;&FO=Jt0PnAq})l!;iABL
zC#B5ga_N90Bkpn1^BnR}2d7!DRFKvh!VT`SXaao?hTX@t#<{9OG$4liRH!zC7D5f~
z@rPXiuyi?BFRy+=KLTaW`eQGg`vyB=EIYJ?<1@gZCnw*t9c0oQ`Khq-Qr+R8Nqqau
z>%8$5Mp*w^%*}Xj-{@vkIr+XnmF4E@&wz}?H?1=qWmR*EmM_PKKl3B>E5iJzF$17Q
z>3a9Y!u}j@_hA3h1hmUG-wv_WG-{xgzF7;V%(#i=WK40BSn=@f2#F<X02=ex@8K_C
zlgWiCpjLc#)EWpb^6N0;Lk&Mv8JfC(J;E?|p2{U|sXk3B(F3|hua;OIG<_BQI5%=`
zZYzzPvvXqJ0tt_nGR$?P!nzMz49`*ChyDtr`+&nUA-G|cJG>B>?RvmOIbnW0U20Xm
zRMIJ=<Z}LA_m7mdCOn~!kORSHhuBMqaq8FR5G=OwU-HKv3xV>gR(2psbkQUE)vF{*
zZwcY$>;#Rg)1jYGjKMp1h?KY`2I4<(d0KXPK1->}@&Q(|7nvxuRXtQOPSoxR=b_az
zDyruyg@*EHczwq-U{@(|{CI8r3~1F5J?I;!gS=q|YPKolk5t~p{eMrU_sNGRsi1W-
zF8N>1^P_NctzER<19c=H4#gtMz!%O_v#bSshO#bR;<{R^nbWR-8aRl5oNz5_kNXsf
zhU>nj>qk|L@-7Jya=5Me8Nbam9Ek`yDLI~^cn4zfuGhx6|4a#5YSta|880DnW^V|i
zViH-xBK@6k^+Jr1i9m{~3`S}y5k0-h;=Gq(-(HkYb;x(1GNdu`?3-hXU9F~5hn|Ma
zMDn^qPuH_Q+Aco>+#jz#|7B66jjWLd$dW{KGGtPgXXM#7Ej>RVKeyix2H~+fY4r7R
zmgol(11m`>ry8SU1gFP)1#6j4kR+>`xOBlLelMVqUjyDY(<LpD=5ubqH1FB^waZo%
zaaTm}YgC2K<E(5b&|h_QdAXd2eHj<?kOG7{vL9@$alJ24g1MyjXA>DMG<qI-FA1*c
zNeS;a($jq`ogXWMl+ysQsk=Xxc)8j8(fCkm+35EJ|3jjC>OdpbD7AX*PqApR;3zuM
z3|_1_IjZ}girKWmX2QrlMPhz$vdS<+XunqS-|n6A8{ZKRy?3lKKImue+k|MH{%~#D
zoG*tAHEaT#b<U+V>=z5u)kEZbHW0^^ZnR1h6`nK{d=2p91c(wr-JFfArMjMdnT=f&
zu$GvAa@X;`B1N6iK0j{rW(Br?|80<e`gd=?H#!Sv;d-w9nl8`J-176?i&jfV^YT^u
zj5V^z3bg5cPb0w>CceEujQ7h<F9->$P>IaX$Y1yx!vQk~$+HhjKz#P6rZa?RfWVeW
zr_Iv&m_Ww0hPN<hdr6ythh9XfIU#Xdy_QUv7UVLw4zPC>>8MSXlCW47b%bD1C=}qT
z8o_PEbyPDzG4LM63tKjmt*VuagcKxC*N*uw)Gw6e<*gKG5dv)YJZPf`B8R*k{W)X{
zS1+=8IOCd#;n-x#Y%lLieewVH0#IRAtIk;$GA|1Lkv8xLdo3DfcZ&OXfc>S}?7D!3
z#kP7<49xm^jUNIoP%o&`lGW-plP7Eq|MY)pKD))c`E{eWl$%d{r1Z`YZ=Jd0dh?cV
zryGsE%G_4>_F^neI_WLL(vVW4tz1O(n_Vh-f>hbQc(oU=e~LyD9h0XjS-dWMD#KA4
zuvaVz8utT}|2}lSg@T&uS+lw;6DrN7BCMyL4aGB8XTx5gt14{D^C-RlvR|V~zb>7m
z0O25=m)@Q!t_gQVqa@*d!161gJ^zRR=S=Nh&n!!wI5f?5`Rn`9s^-k}Pa-E`mpn1i
zWYUa-0EPv-1{XZQ5)^E3$q3aPFU)wH&#LBmT~sR)y3r?1#NhG}8FZ*Z4rSR9(#*Np
z?!e!{e%?Ae2iR-qj(RQ7lv!4nk^FDC`5rEbkCr^?WmVxSqljFxqpF4A3Ar4`op~;?
z+mGwb5CGK$XDwUv9Pm*VyfTD1y~XWI5Wlaib~yAW>H`LggMqPVuJ>pr$i$EtGH!*%
zXIXV{yWZfk+cyCsmFJ}K;R`+dCg|UY9xZD=Y+hro(+D@n9YL(uwvTUf(=<@|ZChWy
zZwkjZciLlGez-*;VzPWAtA>P+%lywR#(jf&t$o2XuA=1sKXp6*SO4?P+spm1c+s@^
z!L!ZYe(*H3nQpO0oliUKyGYD8B%wA*%(XY>XHHAJcIhfiqe&%IKA&=#KE`>S{Z>Mi
zxtW^4ev5oN7w6n-N%SM^O_9=lPS{~L^u_+{*OBxO>%xf=L?F;}3X}zXjUe4@_{EF;
zN|lOkzwz<*bBMZ?QlDXCUjlpxQZYjV%gLtrhwGl3`~7Me(g@qtF<HV?!|}dvZQ7o-
zOqQu*r<`=($9~FF61rSYc0iUJX={bc_Rz=_?CS3xu^N`4>2hSG!wo7E9eq%(IOcH~
zNvfyaGiZNMD4V@QJ&PdtdK@O|IIsKh@j+z#cbvv071hrEtRe$nuIuWRHM>%=gWd!+
zWjwuoNbu{G^I`tSXFy)tHC^gJx<0WsB8rly#e1CQ-J02%`LuaRH$!|ol~Q7+T5k8R
zuL)o&AVE1+smLmlevunX&zF-X%^5bhwt4MFXbJg^#qwNVclDh)mCMq^cGRq;Rjf&~
zUI_X-)M$yho|x!!M%7OiBv8d{;;;}naEU!QxcFRTQmFT|t$Pref}l<}yCNyox5MqW
zN7qpgu{CJO8;;n%wobLcnT_8$a4i4}&A_8~i3~hE>b!jkX&f0wh%1}+YK)t$r$cj}
zn;PK;ey9SngM-*%<ckr~5~~c9BzA5^I;n^bux<{iU_m-jfJ;hG1o>XHw#P;2v%Ibv
zPB&kXfl&&2%nO?*8~U1!^f<o^8$%SA^HD$xkpfZRm|(VQC5znpeBwbwjkj%nT{%U-
zaXIg{5<W!%Y0Y2$!#*y-$oo$%7Wf2v6-xI`O6qR_;IUA%ImYfJHLPZ#EoMQ7v~wCn
zd;XKb(0qn`bi<apnzH-L&6fKPB$moUgauXB(3u(xLTC#4j4xa5!COIe^WHp1Ih29$
zTA81HPlu+K|0HFWpV@g+PMHW^K29i8@I2Iza#1EqWOv(EZe9PDt71x#7H@gQd-Quw
zQ?BOrpxyzv&B3|fOan$|M!5GVjYT*fs*hSY>Z6JOmGbT}`a(`Pwz6h+M|OgV)rvnJ
z>7YXO;(IX7YvOW^(?c}>gx0AaFF5WXNlp}7yKTPf2+9=sb?AQCLn-}b<~>&o$7(&i
zimSNJhJ>q>uN9*7_Q}ReyR9f1CG76b0#zlk;+7t~3N_#Zx^*<EE<(U%&{=0sF>f15
zgw>{E)&3?RLp+9Hby2US*<0!EhQ=S#<Kj`e0AAfvNiSK4(mD2m*MDh!Vne({@@Wrj
z)$ftT_nP*+)}}N=3?&X5a+J)m*4}(oIBTRql4@8x5X0_++@rqJZe%<;TG<Msg?3(^
zFfhPWqV}%Nx?Cd_I*Wpi{NmRy#jaK{3_joRJk=Nv@1HF^^Y2Zj;_nsS4Q9YES_NYh
zJ)Td=tYD1<q;SJTx<1daw;bQ95Uc(wUx=FX73A>ZSrdLG@xN;E+cRT-xb6zSzg@W6
zz`=3Yw+Xg<Dm*Q}d1NWSo9W`XG|KW>yYMM%WMKW&;<c!khEJidae-4rz|b9eQ_rD{
zR4mLIrQwh$<D|frAB}Y?h7L`m#Ic%_JxX)f?0t>PI~x*BCXLJ3j9a;xqT-i@#>`jL
zptN}n9OIkI3}pc08}+9dPIC5lPtvXj)$-(4-)ecMyNMML#iREgDQ#b`^iY)&v<&NH
zAobwnoKyYZ(<`EuVOg+fsH%qT*ZGBAy_q~mAU@K?mZvBw=}hq0;WaWTUcyfO_L~Mm
zOC^X>1L`gtSRpMp6F62x)5?3OZHk5I#p+Gp!?{PuKrC%W_wMBq4D!r)o+`;J$ZXj9
zQM3B5Z_j<73m=}Ro}F<Ug+V<R(oM@BN$V%Gfemp11l-Z*DASHBqYt--W2vUVeWgin
z#|N2ZYx?u%wOC}Gk*(7n^Lv)4v^slld`sHo@A{Q>fMG-r1VHGLh1{h^;mFR;*{0cN
zy>rEB(g?LIP8X`aNBxqdmfK=3#H>bAxYVYHdw$J$f{=vPV-;@{>0Ves;3MxudR&V?
zi8eG{Xj&pqwLkI4QKMOOB7@tX?MS$M6dR7g;J0W{JiNZU&P8T6PQ^Y2g54Y2nll$#
zN^4+ef5s!vcbS|TQ}JfxYk$mBEfXY3N0Vgt<NF!-=~>Lbw-W&F{+QskHYCKcc<$06
zm|rya++3|rIeG3|WG10nb4t=079~pDqV_YPw$2c1IFV{AczZ|}iWL!`f{tLvFXEQz
zitEJtPi*TnVKEKdt+h=xUe$N&uXM~N-^3PtS2Bc-quE?(M_k@3XT91j#*DMQ96b1v
zFZrzYd5PSx$!Gh%VhJWh<J)6VoGf<P=jcbm%lDUrHK#I3ly%}5;YX8r*SllNn&>U`
zyhOd^U4YDP5w0-_(dOrU{%SI_079rvYXMdR=zYW7r8?dio{yS0Q;#^>1;;@bu4h4u
zAE6A(j?M3PXt-yqhZHN(gaUAxbOnF^c79*pPQmB_QFKBXio6xM*wSqt|3#Ij5<SE(
z%ssYYqrWXj3HI+!6gHUQ$cE~JL$UeXU}FYl^Aq<!iYtLL#G{uu?tf^b|I$0Wkq}Tm
zCHOMRjM-OW`S@YP0xi|TJ=pOaX9D@%VqBuyRKYtS)vlVqr06b|gPSvn23Z6}jUHqn
zj*`Oi3%*%a%Fk;dt{xuS9eOR}y>wF{m*k6BufKe2zb!HlKDJ;qma@sE+VQR~fXxmb
zh8gYKAI@O2J1C%CEY#Naohc6&q)mF$F5mec<4|I9bgcMnm|NEeE$~$m5P@pAXi-8g
z0~Rrw9(F{AQrUo#ACB7rfj#_1_M4+l{2iC2LM3V6kxZCkpHIJgPtyI!f!;Bf(ze|d
zW^<@Il!*<0$zwO@GAB>Xvrx&GUh_<)VKz+LgK6!c-kuRph(VA~uj!!<W-~2~fx>D!
zIthPf^Jdtzo9qc%A~^=K!lWzre(Zn0o^M}6g^R4oTbW+#1-}Wu>qw)X&1TA0=1;R0
zCbQvB{%ki@zJGPk&Y)frFu#51L@rHe6`2qou!WJQdRpP7Ty$Dvs%)TrR{vi1#$4w;
zqSL*P&RgKUBMb*g(CSrg$>yYrAh6D~H>Fwhm{X@!sxEqN`v(`q6d6TsYEq?!XOqBb
z?$2%mE6N@+`c#%4Fzg8yXP#}%Rs4c`#iJu+WMt5FrLvkUq6wZ}y$y*&TgqgJwksVR
z7WI+K%Rt>S!rLK!db9a0sqaR_e0w5EY@7gqL(bY%4OoT6V;M$b=%CnJvg|C;b;*iv
zRdqC?INB?Bh>UVPG2l*?v7whqbbXo)P&uL9dP51+|NQ&w33%2b&C_=#9c6{4$}Q$p
zjmbr8k-jC5XAOET+db?Xo$8otA$a@so&jr{`@+(x=F%p033Rx=_6sH)mj_hw3a$=x
zxW)Au-~lxh5<r4OYK^Ch#tNz__xsr_JA8A*w!eug5-re*o~M}qsX!dca%<sk&XOL8
z##{ajbTkY}FpVc!;<fw1G2dKF2q78j^Je9m`aQd}OHnRjK`gHGUvprzcMQpJKQVyE
zW2?})KIx7)Z5gN38vRF=ZxgDPp~_5nP}7#K<j_wkj?kZCiXk}FZc;!H13b&p&F2V!
zXPFPHOm@+4Re72NMNsHteT-8_kXJmnUK4XdTOXFJfk0@L@&J9$Em-G&;2}R5l8W53
z;nPjD|6rZ}^0%Qtvy;7Phup%ob4h`U8ZEtNF!5o1O1)4F`gC}V{eD8n3GMWVwo0`Q
zbJihbI|eeEJMTfs!j+Pc0xsN_@w_9fv8Mz349e8WoPf*nd1zp?xCjZkurIZu40|1L
zUEXEsVT}D<i_gXZoI>w!L-QHts;69d4uCM<O9tn*b>Ulhg4Q({XDIztjKF+^u0<=v
z%>ZgdxJ*ic;75~=o|ulASX!Ey%Z<^~a^pFL4iGx<y6n|!F%nE@X5-(kL;-RRSz|@&
zn;&B+W>m`BiMD^PvOQ8OvfwiRsffDFPI!^0M0b|6H^1F^KBYcVn55`*5x!UMu~oVD
z&jCfLzzVu@$re<Vv;NCTJ^fcy3G53Ttrzb1A|d;I&X;sGCpNs+7J)GFhRB#>h}r0J
zxAVUUlv_=%T@(@fwl&=sZbBS9Mdmust@dVBX@k2kj_7zpW@_XeV0Xr;FfrU8-3_#N
zz+ie~{tgb}am@H05plez(ECo|w}Mj&1!+LkkGlbO8M`f=j$jP`z}J3bW5;VKS0|Fp
zgWL>ETKUx$g>6&XNy14|t>qS9HoNv_WFkd72a>j<OOI%&8qbfr*bEG@s{P9;Z+8Rm
zx7^tY0(zYGWcdLKBx;a0(EDTec+J9Y`&(!r@u2TOXE0pjO{NAXN~%k*moL2h(X8q7
zV~u5+Y|Go9!ZN_-r6mZB&A>RO(o!z$iG}SwiXj%JjhAi7ANlrg;0bvY+Z9wtRc!>}
zhsAzJUh7GF?H2pJJUuSkuA<=-O-ZJrype}fHjc|fIuovPO*|;eBA-i5+e^OTteljd
z1bjmw5{9AhQvLm!bEVxxq`k#`5hJ5@t*!d9_d~JS6&VBXat(6AVZX%-?9ZyLq%g8r
z8;!Q2qa>s{aywMKxjchlzY6)m%BoU5=od&NA$4qUd%<M}qR-lcGZ5=u656c&FfGsv
zc5rD_s52VBaJg9xrIP3v7J>Fa%yTEYp(_X@qm(LH*~z!{6Ntk=zO(l~8{~JZ`QS7v
z-NHPy6^*?rOP%ku=XtIOzcV?|)t~sXW+B@3;Bx9^=M(5H@Dn}Vi~X}}{Qkd_M#Nm8
zb0+sQ>X+XLzXXmNEf2f&>+U>~OhhinHjcyhz1HLf&d8hbC3lZSoN%3g=;GGgRLs@c
zTT1V{R!e`<Q#IV;R9iFrhRw<?2fBV_hS+PTRYX^&SH7C{JL%!e__@tQPRu`hef*;Z
z3NWG;9l>zzgJW0#gOK3?9%<__X{GgbOd`AU3AMYc%-3{kTS(!EHQ8fdkCuI(@6E*c
zEH?OB=4*a&BzE09`xr5=5mJLB6YhSSJYDmd7fm<tr3<OlIF7-PB|1WgSI25+AhC$7
z4g)1Oh?@XvR6Hi9610W7*C4T>E(-4F2i8H+pM~n#jz<Kvs^|?4w|Nr%Nqq@ZI{mn0
z9*?|mLIGR4ECEUzo{w<PQNI2Bn*Sa4+q71y5Lw&!)0c=!1odEp97!DckD}_o58K|~
z$<s%^M0qR$1qbIIhx6fniRIjH+_cF@i!IZ%d|c)6cMX|pO7t|~+L>F-LCvBz{P`oD
z<n-st2W1R}T}Zi5q=mjI^P>R0_}%rU%-Nx(@W792y=@Kd2m&fvh#D*KgdpR)?e_}{
z>vq1iOW{&f$!m?LNFj2<DjAM#hkdt<QgrHhJKzuAGClF5)-%nzts^j8D&%9=G8L9o
zOi%w-Q}mVPV*+|dG7~?me9f9+s*f7jcgOI;8fBEmLsh?KGui@^6b2#_$>F_$6X^;V
zUJ-VS8<Z$1xD9j$lUX3Q7SuV?l2~hEc#V&ILQ-eckpoF%OfGhI6hg^d8zKqxwX3SE
z#PrH`v&2Kj3g<g`V35JY;OJlF<Ub2`T5{iP-AGVpw9At|V|DI2t!=IYLg4o2z@JM$
zlUN1$zn|?tAAJx4)TuqYfOc`T-r~Vedic;RMwYK36)mr@ORpCDlO9QRdX%N2Mr-ST
zE+mXJ?!!-4TU><5?6XhziTK{Y7ARGt>eVoQ{YJP@ul~8gh0mgKhT8A7On0^|vd`Hh
zt6d&Zy<={=a&fmSyd83MO;h?ny4MsjaVI%36cr$C_!WEKpWk^t)>^NHD-rN(eFxEH
z=X!!ZXhyatL5|4kuU0taBhx%((3V<U*A^NV!v74=MWu$t@b;(Vr(O<qhF8`+xCoVz
z8Qw~M&{_Rx@4w#o8{BpgrWsOADThyuz)=a0iERr6wW&Jg45AK&y>>eLi~-_^_pWYs
z@|BpF$Z&FHtn~l`oW_LzkcW=%s@X5i<NZ@wS0n9@#;q#(7q?&uEzsa{eRZWTI`Yoq
zKO$>CF@ZH~(supt$$x33#YiF$rmG%FAo~|^1m4<3I}E=AL&i3rZ*WeTmoBR$uvroE
z{}C2aU!zq}ts@#mwklZ}kMb_Dys1*L5ju^}I~u(0Fq=nAZ9Z(OFENQ58#_M{uIGVU
z7e%ruFJA!6+O;EHcJKh|FY&Rjk8U;kU>;oT=VR!-fEncHuSmFSJbBTXdOhu40&*Lf
zzjh+)>mE!`81s;ck<1{N&HX`#){+)mvwY9g+h_aKQC3;dVLGAU=X51m2GqnmK?^0Z
z4Xhj&d{4sHAd{DqaWy!c=RAqw^2z^fC%-?x{c){RSc5ZA^{znm7cQgMrv7)_n1iEm
ze8N^52bqs^bZKx)?(A?d@bQB}6=#z-K!L-wsOI0zW+RskDB^m@t1RQi#CLvvK|Y}|
z{W$b?%g0^&(@p$mjnUA{;xS=@!}twL2-ZGv6Hf)o+w19T3<^U0AJ)A87w{=Y@_@Kb
zcLzJe5OFq#YMX1(jeCPxfkk~St&Ey@x(K^2<6N&(J?(f?k{dng_=zd0iSU+z+seeN
zx@+-dEw^p@M9nksJ^eFpD*3(E-9$oXXDIZWpdv*{GQW>pJV#zc#%YB22`7|?bto@a
zQf8Fjv0ETXO;wy@NpnBpF_>h-_0?FXf3RG_;)puB!ZTIMaNm}mf|e8TTsk5MTM%o{
z1nX}XT}9@djZcD|M8kW#kQp$o6}8cg5N=sbzO-lkuwlgWO?!qVcp>$p8TQH0#>B;k
zK3D?j9EV;?DDdTqYhXN)jc9loLQ5`+jgRVELC#VE-vZ0Is+B2r^z+aOo{bQKwNB2X
zQAr1;8HNfq3Gd6JmC@5O(yt9Ofg)kQo}kWU7VoXcKEac}bW8%!q-Z6F*RNIeQ@tzw
z9A`hh7vFclmvdQKxkT+#)^0BQBiZy7>CNdM&!G8VYse0v?xy)^{^O~wYnBM4mN##U
z$LmfRvYe(uCAI=U?7(st@l!U46Vs=GA9&px>rXv4Y4mITPTana5E}JeKKowF%*xk-
z87-hj2ta>8b-FfmWja0(di5Eyi>$s-r)D;5(7D{w;#?BDxB&mMV&#15zQ*QKQ6+gU
zffg-Xnd$|eL6(y;d3;aofHw>X&LgfhWuQX=WElJQ`Z8^5g<&LF$gkCF78pA9=j@$+
zvuUcT&C=`}t80<w8ALzEx1}jiM8X(@#0f^n%4D2QUl9dLr+XX=2Of!8v`lG>iX{@d
z*?<4`^wqBhJcEN+k0}<s`&S%JpbX7Q{UrLl_g(VV63TVxpVjdw?&#Z*emVVar?c1J
zXBXCQokHPe7{c3oZg1NjP8~YzMm<cfV7jr-ebi@Yv~Lz(eb9g5@l~_SvR+rcsWTrE
zi^_B5LMSD~1dXHZrWSPGj^Xmf9YV;pNq~|g#k9JaHUO5dn57@f(AekGF}8)AY_$s%
z>>V^zAK)NA?2w!$NCz~;NJ<fDwGS}0ppfB!HT<rhS6J%n`*RnGkmqALiHx>6XgMU7
zyKRAUOb%wG#-7)df<T^JUjhu|1uLOowotgx21CD8gX1qmJ2g;m<N>(x_Lnh{`{Oce
zeZt0T3QdH^fmZ$eoAIpl3d<#)=qBO&ehePbZ4|f87Z_D$Oc~C$#QeZRX1acgj)z#?
z|JiiNT34JW!wpBQ(NoD1lYqvnMi8H|Vf*oyCsyyf*X}rCGQoR<mn-ho26-ILY+(ax
zu`KYFhuFL0M&TYvpuLCAnbKClsk3Ex=BLiPSxRQrGt~5W#Vez{$)k{Z3&&zZw@)JU
zC+1a4m5Xrc_>W%2PtM!3QGU%G@0TNFp-0Ik2_2__>Etce=gmPb(s|a{#A?lK3F3y|
zjhh4tAu&}FeN#e|5I*K7UoS5PX!w^kemFMBNCYm9dwCtJ9TZ#VhTI=ol->158Vy4+
z7V2h>E|(TBUuNG#;2VPt-wyIaY%jIs9MCV#yE`^gD}EUf3=GF-R_F;$eEZ5fQ&clg
zo)Y<bCpn;_S$wG8bTSJW%J^vLyyHEBuJI5Nr9~4j2&~9TC29O}JjA1qP9u?p$!g6l
z1}&#_vFIuTU?3Dq8Cd7xN6LIPEZ}1N&Tl-LKTt6Lwfs{(6WBvVF#qv!Z#;?hFY-)p
zKZ>a9`Lz(RfZN)$W{9iX$4hy+A<_^Mt91|aXJz^lzd+j^YIj!A3Sp?;yS@I$I7_Uj
zy`I>rX4gZ57I5QZQf>MKxq5E)0wbB!YQdgsZ(0|?IE#|l44OM1+XApxtR)*Znujno
zLQs@ts<eUqpPaf(O@RJCyIo>fT5R?@$LdtH!>TVC9l2?8kX@JrS=-Y;TyBr<N2$1h
zC_v|s)qdGtZxZKnDe1Cw?tH`4`pUa{sY(RAFCOo84(2QJf*cJ`cP7Sk$8zjm;2Zf+
zK-<!2xs8N;&U}+@b1}vqjGm1~Xfwp?a(82j^c-c>f1F=568ipTru{b&_TJK=1m;^a
z_@_@k57-rJDO$g0{!5F6kmj|2d#E#4vuG?bdxGWG35P+d-5T$zhuIsBDX5bvA<1f5
z`R;Laj#0zOno*G>hQhM5-neSWEQr-kzV4E<ybt9ZQK2BO1>&5WAS^4(klJ+ii4z!k
z;p_*#jN1=&a8xon?v0zr*g=<;FcGJ@{O5gt;T=d+h8P)<+t8`yl<{0L{(IC9JSh9=
zM05<twY~C6AUpD-*Mv}dy{o**2;BS5rO4}@FZbHeT)kc?<)Wvnbg=<rN#9!aQu$|`
z=c*qv289aQXo2IRc*_90=w&hsAS9d)VA5)*2AE8Y6X$9>wS?ekm3;J5es7?b4IYu)
zB#<9L!SWSd?Xf^@LoibW8*=eOllA~Tnex&=r<d&sL-Tqo0L3r_^Oy4rVuE%)HM%W|
zMy)pDVeWKvLtvuPbj=v;zaF08Wbn&`#F*qO2>Y`y`u|eIKNRD~-dqtX@u=;m@voJ4
z{6eE}RyLN-=goaMu0O}dTm%`tjr-gZ>ahLcGkuCWc1s-9c)N!Wh3}%{{jx45;#*$t
zSA14FvlU}~rOHJwKoAY0oLBoN7icOfJX#f<c9>VNS}!#hxen@wu$i?MB$rG!$ULt*
zg2_b3YxzMCyQn|X;-%ikx3A>hW3Y>EV<BLH{gvl5Ox#I>^;ep86z8H{%!dN(Xz!PV
z(2>j^jGQTE@;+meqUQu%zKD`Dt1$c3_KN>IlSMMr>(?#DQ49nmfYVQ?tf31`C~^-6
zT8IZ?|3NZpV6R}m*SPn^|2r$IofL{Hw!(ZdUBh;p({a6Tc<Og1fi_b9%Ifnxo=e-T
zSMEqZ&QKS@mRR*qQhmk=1dCB~jBS1N)JPgigPaEv!j^MEh-;(8lYRzEUn8)4oOQle
z?$W*SEi@pRK&Km(TdlfE?{E>_U6DoCuWA7b7Bc%u0S4_iSveqN1W7~JJJeb|-m}SW
zNC^g3IhBG8_4cVXle2iGd=U8{5=KNg&7F&Ryy#wyLhO+jk*mhKQ>;X^<dLtHWKm1b
zC|S8P;D+w&b=Ju30l8Up1d2nQ{_nBx0~H$Tu(jQyDR`=WWGJh40TqI<lYVuAGyYmF
zG*`-Vp+G#4SNeLPF765nEnRo6x*mwru_Af&gdD24Vo0Ddd%ir~xuCDmcmC*O3vIps
zeu!`Jp)P-6SZ!`lvvFsO0;>G9Dib|Hrii70T2o72*_Elm3(uDUlJuu(DV%UQyA}fe
zh|HfJJoPi^aO!XrHk%%s%iJ=QIXfr!)AI{c1!nS};b9_fo#T~em7wiTCR!cTE-q3v
zU3Zel_f{^SPwiWAWnTVVh>QEBvIL!(gSn~IQu>q@ET#cnTWbN@NMWtTZMMPhy6qFn
zffS4q8eBV-X6s@wVEs~Zq%b(~nqxR1js~M|Z8b-Fr!v=evzcIGm-A070<4zj0(E=j
zPSr=5ogbXciVIS0-O^+f7bwIQ*h9J79J{oKn`5I!bX{4E?zW}w^)6FZUw6?DyCw>L
zVQI}sg9U$(TBXk+*$n(z7x!?wH>P|(`|CH0&SekaO7@5gkQtqb9%{$08}uvvXH0sk
zhX3OqfnnIaTaS+5ISVBz?Uns)M$H8^CK>b0(<`RwDl@OO1VSsK>nXC)38i01?)Q>9
zb!AvAMz$)#Mv?AN@wMd*C+{8?MH#gw`}<^s&JYX?Ydc1_M%90o<pD63>VOf140PWH
z_oO|N+Gsg18n2-!R|@(EGrM;(`v4f96A>h;V3ixE5w$&B^24OMMyz&=r6R^q&K)oN
zI>5>iW#z&EwlR8NE@i;?I*)tWy%~%~#~TXiis`CCjkon;l@Mm;h3V3^;3bb3VjPdh
z1%FVWIjmdvuf>ibn16aAd5HUsO3QqPeRHzEMnz|Q*yzPm_+@hg^sCs$X@vlRz$3jI
zTZNoffJPtZ+3(3bj{dDQPYqy}QNSF7IsV1Rt>W>Z-VVf0^&VJE<Ry2n6*p;YjgsT=
z-=*KVsr}G1nVMX_J!B?nxPR{_OG4Qd9!<9Yov~%5vRfj3{)w2e`jO0Jd>I!t2x|>g
z1N2DHss>Nxn*F=KkN^QjiTNWGCH&=ug||%LZ|z3}bCVyA8p)~}lo4xA#-tysu(dCg
zwVTkW2QMyq@r?&0zwdaQMYoZy4=7mqn6d~$(%E&|5qMt&bT@n%N_lvBLJ@h|Gd|u(
z*4uDbHRL1$Zo^w*kk3h2vrC_pyK;ZC?H4r_>xSB!-XYh1+YIBKkX5WC9RfSR@pyNI
z-7lV@R%??8CEV9y*AXN816bZtwY&{?`ElpQu2g7)uu}O%dh#2deqF}MFfC!)LG?K|
z0(%s>0P{>jxA^RUinkb83*W{2XrrwEhUy_YXySi$*y!q#EFaoY@s$dfhoh>i4{W2k
z90$>eoF_1j9rp6NPB!!$pnOe$h}wQN)zu87KV_`GB2~zHwlc90^ubq8{v(l~{P)<W
zrd04v^Jj*)8d87Ta6BN^?6ZA&A-JH@n?6GIXkXlPwCaa?C_(E?j2QWgyuE%jb<{Ea
zPA>=zx_YSyM{>|~q&N*_$wPg<EnWJHNUG~##Pum*D}FJebdrcj(8eIoMnXW2hI7+#
z7#NJj=%KkCYL145a)r$c)xI*Bt8E{4^$647O)h#iLDklx_AT7v920Ru+wB;f2RZd6
zL~4`%8BCE;Y_I=z$2#?n|JHwrCVNPCU#-5z!wh`n2isGlo0xaeqo#-6F!V!RQ=j$t
zl0@5S{2T3ecYx`V8jDHRsgr%tC(`#rZyI+%i`n3slG^yW{4SLjg^6J?Za+>dng5Z8
z(0uL>NU^a02mx~E=2$AFc~?ub(1eM-L$OWrh62yWhS!!28%K9q-F?VK?eeRoo<}<P
z>u+0s$%k9>kC4?UF=IVFAJ2M@VcSywh7%%SSgF%?6R;A97MX_ZRcm|GP}VDu@0g@y
zIwdz+xP!&~NS!rd#CZ)A<@1Y=7wQac4|4}TdEajw42@e(wga{7N0ka+%GW1B_E=)+
zps1o$L4P!oiPE~3;;8AcXO$_8jGR0;%wy)fnPp~e2#0!P)ZX=Nt+y~U4pvM+8d_D=
z;%rM-*n`gxUN#(+=!eexj$#1?X||Ex-QzA4Tvmw{qt8EqcV_)ac`nvVQG)|Q@{rK(
zg9yT%Ab8eQO!FDJJURV$kh5=aQ<@2(z&t$FiGBJHH{3;A;DVlBGzPQLG^5#fAd1}l
zX|qAgVwoDAn>N_h86Kfi9@=Lo%WsZ072%=?J|17k#bD`I!xDv3(i1v8-&nvxV)5!>
zTK{^wS~kU0?bl$>m$7XYKre&wxb6shNhlbTQ>e9%zUqbyx)q|*CET8AN;BrCp#UAh
zJT`2>p~UDf7{Mna*m(=~_i3&#;x~|?NY@9Fcu~6Qv?3vJY0Y#C!TzWKeGwFIH-+3u
zq}A5=C5-kP6Z@VndrUh%_(3?wtXD4vIhgWr3YLn_oi$WHt`DW4I9fy~j{rRULngb^
z%-46W&?j(r-O<m@rgd8Vaa=;lO~!`>a)Liv85`XUC`m#UYx9M@^)?_SIc$S@K3(Ei
zj2}n+fi}<r-KViTOtPo7qGI3e!I=z%U6hd6Pf+84fW)Ti#nDMpysrDZaxT550F=r+
z6($D*{CwrP4&Ehn6u;l`q{P)tkBh+x+EpPFFgER{67i^ys!=I~O>eBp1;93}eFuI=
zi(Nh^D_tUPniD>5RukNyAJ}8Lx$!IsF4``yh#5&_ohD=oPP4}a6nKv2Z7YH2N*JeY
zSe_c|Rao}aVi!^Q|03)gqw8waZeumJ+1R!lv$1WXvD3zG8Z}0nG`4NCvDMhv@xH6?
z`OZCm?sv}^`Pb33W3Rm)%x5AdDpU(lk*!=DZ7G<_|1*@JKwWv@#iLl2!~QSL^WPm6
zn`&#R0c(MqF=lzKSeg9!M-m_Vmex|xh87dcwDMEp5TyBb?dJDe<n+ppjRVsb&do6W
zNN)wW=Xf$#yF;L1X*=`XUoaDp+y!_Al3t7E>>E}tBi~d6w4c@YV*_qY9%e0Oob}Wj
zEZlb3{7D94Xzlx>L)cVW;eVNO2<=a$lHO|`lu~?2294uiVpv(&Ra2B2zBc|INsbfu
zF~|~Hrf4rIJ(TB)?hkraA`xHaUWB>VAs(>W29$mh{rD4@RO$-=uKDF*D)eS*s`ue)
z;mhR?{lJgDa74UahVZvnKz`Y9$r}-^D<f2Pe})27(K{Nj<Jw_=y=`OUqM+`(!&U8s
z^)_|DawCM+8{6{s6~tl?*a8bgXnmgl4)IxiX3#bBZ>4_Ag$Qc7)c9wM{NJ~Y5H`d^
z)>i!Q*p{rS@#~_y{n^5ak1sBK2_z_`@7>$tOB$Vs7$(yrkg%-sj!P$|pD-BoI(7DV
zjh`FITnxPTCP={y20pz58U$f4Q(t7vVDoE$4W#RgvGBN72Np`LT4VOtr`^FMzx5Aj
z?MT^TSioDx4#lV~5H_URb_s(D+>Ws~o+Pl~amHc=OVOuKxF2rSvA%O53ia%Uz&k9p
zYLYlNmTHI$E8d@}%XIra?$@|?zC9E|uQm=C8$X=Xnh&jc?<M&g_*kNPo<(0Y0s-jj
zyOvAFev(mTKxJs!#H}OU;fuvL)oQp`5w;XpMg<&;B$&Tu+{S|6o@vVUEHQVUx<L`4
zl4rd~x9RMt0Luzi&1T{sFfK}(+znjvo{+xrA1Af{4+`s#K8>$ba(U0yefmt@Y`H8!
z+|@whChCpm<Y#iBO!ygNa8SwA&Xswuu`5x;qPgpE&J`8@`or>2+h*{~{=02YY0!T0
zs+H7R)=~7G;t{9VjI#MD25=dEZg07r9qw>)v&^}DC!YLq(piY=?|75hU1`X2d+vCr
z$doQHA32$AK9*@~HV4{ljYURDW#!GcQUX&lm*E_4ti#dQ$EIz$G`k)q3P5dyqnOKd
zw|P4H`%0qzX@A^?FTUK#h?^35_kv8S7&xPxoOv=Bi2<-mNCbIcaTuk~30MeYzjMv}
z4Y4KyC6IZUUt6%#v48j`*5)a2x@L=ORC|{*!rOz}=+{I9B#;b9qAk<aqv8_I0UJg?
zY{)|=(GaYe)<5{+|M~%v6CmJc^2iKa&E@}6ow0Ub?|pWRzYO(jUsS3te;&1NRi#oB
z@RAvf1VcV`vsN3*fF&17w)qwsnB9IVh29Kz<>P9X-lqO|w<X}&f1+bvc9K`+o&y~G
zm)=*qn7Tq%Ngvj()JJ7K3}<KW<9!)Su(s=h=K%gs<mZU8J=O=4S{7DNyF)~5HmX~G
zCTqL`lZ2{uKNqNWCWbGKuSDNm3RTD$vnC(Vuijpp^ncF>WroH%PLsQVrcF*%6VJOL
zJwA-UJ`Suj4M0&mO!e&y^k!RUg#jUw4>u}zm4FS}%yzusiwb%y-UqMsA5MaU32Dg^
zzs*Tli3<&<eruorTS}x-rlQT-68J+xfk!f9h+@{E7yu6WUxmp-yUsv0|6^@yDfU-a
z8$3z*7w-7qe=w7Sv9Gk=Z0M2GxNla)0Bm5`|5PNaxr*$x>+C+0XKDC1X~0L<(TM4J
z0ooEt1Mo}On%B^_PO-h~by81tr7=pAKNVrfR_aF3W^QS^yZmNIv?rl?La@P(|5VOS
zC|rw>@7-}v*C(_enw4wBVp)Q3p*)EU<C{IPfJ7luAIirExY3&dQb4Cz68r8%ituJf
zg#zt%IV|Qs6P)5iVi~|akwxf6(I0^k{G^b5@MEQcn@*dB12jFGrR|pS^Z`xHd%Y`&
z1-PmPehC72PyNlV$39^xH?K4*%1Y^#=PUCN$#NZkWb;2!`W0#=d=ua|bPYq*Z~6}0
z5N%%RXahf;+TbkK;mo>o{6hrm=L~^J7`a;GTKI2)r3ETt+7;n4vY8@#uG<)kU3tTu
z9$3n^9k74T0XGjd)r|HIZu|^E*ZgW1%|zl=G7KNu8c|MMj%WN_%gq@rx<m*cjhVtd
zZs3}syhTwOdlyp1ww|6<s`grAj(e8M!~XeFvia0DrQiR|q&(oC&Zb_ogG2Zpdkv|^
z1^~b0GYCr#PHVnpI7Gaym)B!(0I3)3=UC7nO1kR48z&G;fYqF91XxDGzH7hWu34~Z
zfN4h|l0GNZBec>i8;nnH=DI_>Pd{})m<bB39vKr_??XH2vr?}j1i%2y4F&U)wRgoH
zv&kpiizO!^0X?DbRqG<44foAa;81{gEt9{1QKtYG!D<#iz@SJ>m=cM3^@|u?N+$tI
zHxvef<DNoZcpd$J8IaKFp$NCFSk5M&RgXTyPA}RYcHuV>{XGSEH=QJL?vgZDZ2K(D
zrcVreKgcD}nwZDfzor>dE_j?GT#NXS=}w7+tD~Wc{YjRTFG2`dWL`<*wU6Ej`E}b_
zHcX8h=l1I<LG=D!%j@~<7>OpjaXrPS^ssWLkw&<?^)y;$Miw}?SHbMn{t0+q0d9y2
z-icf${gK@T)83Bu1@nHuzSipTj){;MfDNGGe_y!}(T88>lt|l5le3|5k0SCU491z4
ziS#VAaUE%uy8tGSX2zFICP4gk;4dR#M(tDyIh^(RRJcHxc-wjT^@ExDvUzjTkUumI
zsdvY=fFy-RG0R4bF2;eF9|{<eB`-F#O1MR6nJjpKIBaqnfKx}5yxuGSUmzGLNXhl5
zlG`_^A~X_61k;MHt;sQ|b3x}F9gyd+aU+D&HUV!|URNFqgK~NZ5|*_@(DuNO(a}ar
zDg)Ytv|}K!(7`we@gO7sw#~YeJEg4qcm7;PfKxwD2*64iX5B8Yv{gqT<c?k|`zc21
z&24)$efU+z0ejUN!CnR`iPoek7x#k2=(InAjL>y60Q*8Zr`4C*hKsNr+=+o_e7ZSi
zuZ25ZoPdHFAwa2nr>B}e7ln2FyCU#)cwO?HJ<*HSYNPMr{Jc1vu{S|c7Q3WWQzJ%6
zP+%uff{5PEHsiRPbR0*aMA(wRMX$rx1DA$e8PKUK1*dxUi9y^wwL0mw5dcoWy<Eo-
zkO+o*R!(Mmtq@%R2zYR|;sNriJXDkzdHoCvhTb^O{~K}n_t2OqhK(p2Vyf5GrfRu?
zgs%^GA9?ZwBC`+%!^DGH2o{_d!wuEvG#B$`hsIRjQV^$&HCXpvvcRuI*X&Yg!<p<Y
zhvyk*-*N)-+3}}N-KjhgiTQU1i{v&~d*M-I%SUMw;f<7}Tb9BGzXMOSME)=DN8hPt
zr>rVVr)Bobs#DDAL!sS%#xemE0l{Cz4$`5n>y1^~B_?5;!V=)*&~aF>^!<xkOD%4;
zAZ-F$;4+8co;*ekp^@)MsAR<NGGpKf2w26KiTD}2ggng_8@9D;=S0rCQ)8THwl(68
zrDd{P$75)g!ff^y%7hqoI2<m|MHkA<w*`h)L8|{noy87ees{%$8NrkOw}4kf6)`VQ
zW({Id=zU}Vci*E^pc)q+pN+Bzeesk<IPl~9;Gutl(!_i@IGWlJQ^FFPdi!<R8f?*E
zj#@Od+v2;*Uk8_YDkIFNST}5ArHejY>^c@x(;xx^n52`_RjDRsu}cwb`}w%~z&qxe
zVrJv~FJWy!Es2ud0<<eD4Rt$R;I*aJ`Ph@n80HysN{$4J%%VlW*cUnm0k8?(#&D3S
zW%)57D5y{JQ72pAtpg*iF>srh*R2R~<^9K`m**-XL9g(p9Vx!@SCTg8TrBe;y;o{P
z{_}=KwJ#C-OM(XyPLaKknx!Q{v!s0JnFrw#H&`pbo2vejHn#3pCEGw#Kx8t}#ahco
zGCCF^K9LMiO49y*dIQvOTL0}ZS0LHFmXB5H--z^wMffwrbfYOrJm@$wYs9(Nu?EH?
zw(G;sGQIrXljM(<kG!&N&iM5bI;V{IVIU)b&0TdE>}kB=H9^uQreEXMy_=DDiU@Gl
z`^$AVNQX)NRTJs~17tA+-Hbp;o6!^y(UdH}5O|;kTpqEMGWA?$qzHj!Q94_|zk>wa
zf%NI3a_DmVt%FWssO0ropQ59CNUvI_w<<YZKkPfV`GvpJibg^NDH2n4e1izRV?f~Z
z#4|y0z0WV7HE6WGzS4rgCQAj9)=zQ<7o$f%z%4s<Kw|DYumC87pGtyK6}Weh1=?Ve
z`X~XBkq?`I-csEAv?<C@*<nLLBDA|7c&3O--C%XsSg!T^SFhn*UcsUj*#w9qG|^w)
ziVHV=+bsN=21UQMO1|7;fIk~dGJT~0t-At8G(>7i`KArC$NB~8i8WftE*OKMc*^Jb
zk4Cni=BFf{cBg!oZ7)>-c9V^+&{R+XX3f*P<7hk*Nd=w4wl?28yY`z&!5L<*$XN~!
zj{e-|x^}fcdwV9`!H`Dx-@akjAP*9=3ZlG6w};vLDnIM~kut)<#(d;Wzr~IEr*gL5
zY^-v&0Ahj!bjK;z-?Adsj)5=Y4=jzF7n)~H8@>uJSGubJ1Z#m1OT>%13FPu>|DlqJ
z@3%}d{9~E$t!L3W67b~0@Y)RUDvCx)zo1GG3Ka#v%rUOl+s>)hm`VNp$N#dyreNwN
zymj7|(~@Ru7;B9P5fAdA-!jedNL7owKEJT0lJCwGn>x9F2H%SJAD3!KJ{h9<irfcu
z_1;eq>U2F4-Wv)+L_j}DE2*9@r-cs7@ka$bC;Y<{%WxA6vZfD>VlTv|TMuYbVq!IH
zH`@<$y5EkVVuf<Go`H6Dy<VdV*`oxg^(0m8W<$-c(zoX%cvTI_WuJQ;&J<V)sslUT
zW%54XyFFfXp<g7I18f3y#aG(gTr#WOfo&WH8xywvKYnmWzdW8j@fIqT{HIc$LC+%K
z-i!U&{mU5{o^rh(EYE31^7Mw5J9~6jb5eJS$?7|hI(L}qrX?_dzHmFucpTvR3I0@c
z^?sGUO4$sBRW8?0OF+bmJ)Db}!s_Ik0qm+(Jz;jtEZp8+X@OowfJqJTSViVm7pg*5
zhsqrR5mlnV7HRHKDTXQ7^xdgwz(q<*jqW>EB9sxC?)!CLe0>9=PkrmDLIWp~UK?TJ
z2R&cHV6uWp)%JALd;P{1y?3%X+){riq?z@E!X`21edBL(TGzW8&JBE$tLbFcb0OA%
zkv1PkPu#fy1D2A9Yb~6W9Xt;;0#=0RAEfTjRwHP0{V{L;OkgqolJ-<nN1u<f!=$%U
zCsdf?KgcRs_#6f_ge;mwMjFFW>u{|Ky|B2wy^-X^UHv+%iQY-Z9L~g0rIC6Id|}^D
za#$cLS(C9;=4eo68vxsc{_wR9xZDyS@cmu#u}{zyo7p(<k^260c~^6%KMJeE;a^|R
z0(3T`Kx&4dc4PqTkUD=>hsd08o-^11Uyl!wiM^$0Q%NdrEtcJi=56=JP>9+>Ux2<{
z;D(yUQ3zew`xGe2_mB%goJ@7e(nmMqQvrd~o;4N~P2Sr_#VCn6)s!R##V_x?k3x87
zg}Yzm00`veD&ierIe8PyDd<`B!Lpq`GnK~cmv09gS&##EvM%%R?XF9}021DY?^uXw
zAEl`4`z6wkl1w_k)b=h4dm4)^1OV-t<=AZh*Si*-+V@bS!(U-jw*x&JLGH=~TJXin
zEtiLd40Qc5T!2O5t<AasOOd#g9m&eiP;loohpKEvoG|<{QiGz1x^y^PGA$8+Euwyg
zUv;@Oaap{)+C?}2uG9|xw6pFNS7IdWIkdHRqS=@ZEK5yUm<M9#ZO%QeCdP=2UUKBu
zFh#1uH#N_R{HGhkc!QXj@!g<JKNY0NFzF}Ek%!W%s@s}r+=URBqPzy94qxxxINs?Z
zx#Ta<Us(;{gg|TV0FyjP=j*BLY?K#PrQ|X>W~4imnxb)6V89jT!EUu4Wk=LhIy<z-
z$wC-5EXRLF6tQwl04tE~0%2%mfT<N71z|LmR{x;z1+GpE`@X{R4KNtm0iN+~s7EuP
z)he8_6#0v`Xalo}hXFk0?H{B1_3q~_ghJP{pFWTSazM3iB719;6G^CTjNvhN^IMxX
z^s42n-gnY~7fXRRuTT+t(5=sbZ5ZA_H&%EKVDV?zW;Uc=O8D+Jg%NLcr>33@1(#XA
z54a5g=KqZ+gp*N~(#GNHfg|R*{<ERrw|B~kI@dbJ&#av`_c$qR_g!GEAb++DOgI$@
zo+3wm63=rk0I0d`mGrw*GN>Ppf<w8!+K$-hau1yfqyM|6iWvfdZ>f9q?MN1)V3;*!
z>k<6(WU&tj&5l|%x9jh-4woStJ&dJ_kgdKi!4R8$ulh>cpkw0!yi_=46dm#Out?r1
zCDU39FoCak+5oS{h-qelDCiX+hIKxeeM>v}K%<&y)K0|XK=(#WhEg_<WW4AE#hsx9
z&^jVNLIEam197|?k>~4eyO-O2rt@yAjgHL#5psa(rH2nNN7M1SGWicY7KsbFzTA)}
z3JTqd!2+8f1YJYMug?<dnujH7^C3p2B=RbFV`(9@x=%&q4{#%~a}8jIe1<m!%nnPN
zna+UoZn`2xJ2hINDGON{J+8$s<}@Af40e6Y=KDkkm?HS`tKBm|Y$oW`sT=WlqtwgD
zXlzb9<^#~@NI%0p8ly~_K|X<+dY!bjmvr<7^CJiRQ!vnW#NuoYw|Z!Z`BYKstcxX(
zDzTW;CuV$pB8IOH#-X-@a=`O^p(DjM-{VQ{ns=fP726r~oa&p<h*gcd^&9MVvM*-N
z(`3`qaw<b!9BZxO<YHBNjK^q|#a<R_^6j<DT<oxkKq<0yW`A#*HA=x=#9?o)FSXE*
zb(`vu`B5;xsZE}pgq82lBil!I$+6XsttQvLhIdb93sMMM51L(4Llwbq=$tYy{CXeK
z-Yz8cdOK&c*391dRO%8cMTt0p1~`i8JGBhlB(1_yB(3O7(*y#y(taMuN2Z(8x~ZM7
z3M@UNDU5xk4nK9=fiDjkZjhpJ$VKNfOL@CD+kX&YaZeVQe|rIZ$P`-`o1*j@6wJr7
zucHX6M8EcIBh8o+Mjh&T%|FQB!I<nq?R(mH8z5GHTB8?w4-0nFWXB?;k{9#sZokuF
z+>1H;x_!Uuakw<oHC(#KzJ1#ETJej8<0l~OJiOMyq|=9DNh8v4<hwd5MrG37!ImKG
z(0LsP5{@6AS{?j;I>_g;_0VZ`-+=uAF|E^f#<CF)4%5d{<uLvH9#zUY{Ehh&?j1tz
zou*BWe(p>LnynKIl{-*GqjL5VzIzeh>xSPAHm%BISx{axZH}z}McdaN713W8c~t;#
zXRbzm4|@fVCcy6RE#lNl4^Jh%7h^vRcy0d1BgXrd7Q5-Vl65~QXd!E};f=@Y8J!|`
zQ{&CvaYeK1)5nu6#)J`h@{4CsYms8V5i(kc`{*e8-XyLC2nX<1(m!M^Av&c{l!2na
z`U?w*y|hfC3&JQmt~}qm?eJBjwK&H{WsklEDHRpB0;5#b?>hmCx2QA;ydUVwoj42{
zV%Y*7?X@YSm-iQ-z^M1a-$(Z=0AY(3g3#k$ayp-DfwqS~;#>#jU`W1Un`iW*Ib{Po
zYHNrlaJNH;!DZ43W;g%sGx=S>qg3aQ7gWoCgk@rD7{kUjFps$b#6}%qGDc$uF(!{@
z?BgT?RA=WAc~$C9;5d2xjP~U@h%}IbEqx=`=2}Dj@^v#Ex=Y(Ssk4UY>)u&ZsjtUY
zMH3XOd!-e(p>FYu6|iLV{0lTSO9-AnzE?!&^O<F)xwadyGdV^mGOPMg{*ewwO`RK1
znPUN5nVrRh8NmB=`d6}Yk-E4zkvyzIk&?nYaC_VkYzSbFAcIzT;C}q=opI1v_tTv+
zMb0OaAym9Q4-)6?(068%-OqzWS}!kkpc{8SA?FsUgoTP-JxuIRH@uE&#=*K(Mi0B=
zKnA4Cl(3KWU;_>#(uofs6b5W|NdcFNFuILkU^w43KqmB!rT(^Cn}s}E7@_Qq+5~vT
zekMwYCk?yox5vua%FfvoARyRIxL^+tpZ{CxprKp?nijOGAa^w@C_gVdN#5V6<n@+M
z<$W|qGPJN8F#u|@22Ea>E6<(`y^WfB%DANy5E6>FS6rVD^en)G9<2iv7*uZtKA&9}
z-MgUZH#upJOZbiyf^ej`xKc4fCNg;$Q?uxHw2aN3fY(*h2>d^$bAQL#fB(!-ZH<n8
zxG$S3)6PRT)%!xebiL1!?tJW~H>`g`e0L2>PCj{H*GV1MRy6Ir<FOg1k9Or6gbUg3
zI|IyQS3hTb`u+mI@){L{>v$`--9$tOu=e6mc9?7#TmV-iQsvVS7WZm*JEdeE044$y
zw%{Q$(3z!nw9zxD$rOODiVmB~CG_+js8<7cIbX&L6D7lt^gf+fVdwVtqLO)<0>r`l
zGk4PqqqNJVlYRg~Cl)$XTd#94{q<v$BO{nbsTa>SwG*y5VXbSEzFk4TcjD*LmD<kr
z(+PkcG7s!!^7x2$;X-5XD5o%78{y>SdG5;#0;bPR+WP=XY(7ccb^yKSN7De<N<pr-
z5%nLtxiaq~B{jB==r=7yA;MY#yhK9;UlQ>IVNNT)fg+fghY9n3?!|Af&3P<1f24uM
z-~@*6`1y!p=sY(G^oCu8;@hpl0Lb~4F{ic4LXg~XeKaM^vAg*2WyuksHi3(>S*g{;
zclwn+Zt1pmmy@yG<m6x_Z<)=`fHv*jIO)O%&f=&vVNw+RMu(5(o<)@b1HESBOLf*N
z+b!GjReG9bS`%Z^|5D2WMOiSU(y(o(c8ul5PI|sn^M9sHKzy7NY<Y+0Vr@;2_x@Y*
zQ_Qh=@^?6()Vkyf<Yy_5Ftl&!rLx{e`lt(l#8wQ^p0e62KY80a_>fI#7<nj`AOP`q
z%u1<3H1aUOLt{y;V!cJus$VkgyFHwyXl1Qk_h||4>uQZ6^^Up6Z2CJk0F?IO=ZV=L
zz)o<migmQn;Z#H1L6S@(waDoIvDMCdYhnUrQqW??%{%}&W-#v+2R!d^PXH4T$cdK6
z*>fk;nHca^57`|W>wIO_KTi6p<|%OrB@f!nO?Y_;753<SzUSU5h$(nJ{i!@5Ic0gI
z22VViP@Qjm50)_q@bEs(Qsxau9zoL+Vu)F9nWGlbYQK)wEO#ptpURO2q%V<!aeZY9
zzbN{zLfk@!R-JQ!a%_{Rq{x8a)XwkH%`8NzKkxxlSUQ2H+lJ}+iXy%(L=*7FL<j+B
zDUc@nO4$%|4QF{>iDRwsVgrX9l;mNp;l6RBC7*DKE?|Hal-jEMT+>0is6AG#yDTzu
z5TA;CpiQk`pMwjq(x&037K99~vS(LY3=|?SQ~QCm7x8c3aKHuSwv|=%f8Na+@$TyZ
z%z@}byG8LI1UGpVI($|N&K->xo#Gq4Dkht)a#wt;g)Z7@T{bM&Vd=aLUO`1D?PN*l
zBCT$<rWxQQ?W<W6J1?wH^I>CB;py-HZj4ki3^A7a4YNoTAr!}VsRzD!c_Vaqbclax
zH^5b5Co&kIx|QpyRa=RBHeN)l<eUPr#>ar!nkAJ(@&<n&3attpE#14OEFC_h=p&Xc
zgNsK%WPk{?+cIrIHuG?l+(5gC+j0Zk?Nt3yBbFOroAiJDF?y^H={}v;@12G-?!z9?
z%AIaO@xykGrrmw_q>6-8<u?V6-utV3NvCb`<>N;#r}BaB5NUw*gptA!c7W3?iy>D-
zFMK5L+zt>2h90g^+3#VT8WnF=>`}{gwy`&olt`zBLlVr8{tkl(2@o<Mv6ueS^V5Ue
zVEr?C!H`i)p}&*Jhmh`t3jKj`(S1<_%KflpBhkX?8YW5ysLsi`5136fl0<zX80r6V
z7yrNb<KMp)eA5+V0{-Ze^6i$QF9tBR^r+t-?E}Cfh7Lja6f4%ynt6|&#aSoGjjSWS
zg)?7@!|a2bp!_XPrW}r-4dP~Z_h=$;5bKYrY~U0#mIGwcp(2x%GHs-ctpz>(a-lRu
zFfPHiQ&`U^h9bp8r10&Q$oSWRh`a!1QW)^|$U3Y-f3!QPY;yp-eFKN~FLbzfO0`go
zttf=2pL}JCrAf^B(dd6WZEgf7h$>^nhjUDqY2VT!WlUTmP>y|_8%btCCi+=>vJQ4u
z4Gdg1Ae#Bp;$D2aRM835u8?*)IpT5h8_V)yiDC-to!!DXfDF0USjPDJ$h38h76=_#
z`mOO%EmZL$R{ib)|6H!=i?f<wlK9{VQWT(zX{}QPpsYO#^tWV7%9>fmsRGy2M*^R^
zpYPQamT~_+kxTT;{SbB9R4#{=`d`Q2P{~C8o^PiZHwn4zBB?#>{FZIW|C&%LtP#?i
z`svnl60GRp8tBseXL|hC0`^6|yEU+3hhn94-qcRr>~$NH9_5ZGe@Fi!zu<D#zWMeV
zv!>H)y|>PGA-y>1aSy$@;@G>~4dB^;y;IHdpA=Za@;_n?@DYxsb`OyWx`5Vs2KEEg
z)kCCW6oDA02fTH=Mw&c$qt6(G?$O_+IQ|u*9D+{)<OYx$wmyI={Vwx}`XU%r#G*cN
z0-ry4gANZ;$n8D<ht#*H-S(SL>4Ll{3aLf;W&@WkPmE<Y1_`9$93U$J|7L2lgMV}z
zgGBC_MyHQJdR4)C>FB*>+7bEGm`iWIq<FJ0N)5!7yMp(ZhTTy{_vuy-o>Z0w+Ozk;
z=rcZFjRP7`R~s{2!LOsOY(IqdRV;mSZC4kfWp96}Wz+*m#d}&nK^JrDSE|D^u05Q|
zgApIl^%VW{pOe|n*I-si<Iiv?iK(eSeKY-2(qs!{=kwY7I4HJG43x0Z+Tl@&yKBsc
z+Lso7slx-k<ED1*-&4W=9S{CW<bm4raeQs9lZwmis|Q=pQZqe+QfAC`m11zE)n#%~
zg}oM~_3TLxHU?Xq<%oZ?N{*^xn%lFOlwh2tg2@6C{L@_>JqtCXN(Uf8EbzJ6n(UFS
ze-A+Lj*r(#bg4f7nkV}aLI6%xiQasWB>*s)Oc366I6txClxQlDWDJl6m9SpF(*Wtj
zU92q!qY%xSa!D75gEpO`mBmHmb-Ok~2spn)3x=BR%SXtZF1NU&20qbWwKqDg9d=vi
z{$kxB6XJVGo+j8fwSD!a1z5X)nWTmJ3KwI!mNFc|A2oTU`K@fPSilw^Yfql42e3{x
z04}?spiPzb6FMC@VDx^Et_yW6q$}MR|8Z=6x*T>~_?bTKpqlFWk#@_o|04Jl*D<o;
zsm<}&+e`Jl{<56Zh~1#;QZJ29Mq!E+`3LRR7_@T{OoQDw>tyw>V|>E`fba8Far8P%
z8f;W{<DUZ=EBgF`Vht-Rn!wk2zgcBzX}Dv+{h#aC@v2)3(}>V)J`?Qd^^<G=)y+4(
z0CTym=QN>&o9k=Lz&~D0aS3i&p3;|O;u&3<)Kw04hLYp7PcUXzk1`8#fo9bGYKCTE
zax=dD*L;*iw>thJN#=fBIYBY^SbrNga6f%BwqI?PtJxCU@Jd1b=S*-=_2G$1MIt$;
z@>~DdpO<G9R{avV<@k2A4Rs+!{SG+A#q~UV8S`Zf?*kG+I21M9i%~tLbQ7_I$sWJr
z=no2i?RmZx2J>RBVR5TO^f9N}dYR5M6r&dlvIRUHC9D~BtVF#&?>pN^8)G(rmn$FM
zt5$r6Q~&vZHyhPO38;8c`1xl#0#AN_l(D)xsj9hoqsE>5`gG^f{OSn~sFW54x>>&E
zSgqe=&GK3^$mo%*m9PRaYpB)|)gaIZwbg5CU`kX*N0+c!ptPuZm8wtm?aT4%BDyEZ
zlz>7uJEf`=5VamP&vos3-lDtnxX$p#7&j}2Q)0$W8U&ndU73sWtf#hU8LccKy*6z7
zDw_1dlHa?<U$I!-0Sp9$Z|mxEu?6x$Hhd053%x%si+=#;li^H$0$VRp?7{}t^3+|>
zrpdD>vx7cQ_uZzbOEw>B3)6Sx??^cL6&eG(Q9Xbm0=~W>ykT73JGA)jRBX2;p7PIz
zzlR<dO|KpST_hTGs{gR)a{YX%WRSRF`pPD<E^VC7Q>l$hG!Gvd37y@w1JQS$&(g+E
z@N~MY2OGlGuhj(iWKG?7Q^qcqIiXq07n5VchB+fJj?z9h9HZjeC&a@XC`wgdI*t1B
zgiVHH4TciL?-gqD-bcH)eduyzN*%8&obUFQddJ#^UA^M<a%j0&RMhP<Vrk?F=we(*
z1dlBjwMvm+!|?o;73^K!d=MtPz}Et6bBhN?Tb5`RF8daJok#jQU9W`7+bH@ubJx%m
ziV>$&uFCd|+BgcZMii4rJUPyBIg(kTl|kTFUch~!En$znhH^_d@P+TsL?&(Ogf7$(
z`z-+b<lrqe*v(6<w|WTYW;(UXwJRQ;R{b&^czxJS&;fw|K=7;A>1xx|4RecagPumE
z!SZfJz2#VH^L9AVbfux-w*>Jivf56+`_ro(@{?@J!RP{Uw;W-=#&n+tSX5G>$iKa4
zPU|Q|Gg`at(-CFhmKR>#m_3fL;@w%HOoG|t^?r`;u)v5fF~1v(+N}-tTvb`LEr{EG
z0b8NJUOEDm#ASke(fc%C>jRQw=i{GDvxwK1o2lsDzLf@hV7%#KN!2=+Z4)AsL<Ecs
zR%yq+X1x;$0-#u#ZpO!)ptr{Xk+an%nG^<13Hrh0``n<n9$<`*CTa4%+A({E6+GI>
zjRIbOywjY=&f8f9;e%%aNpvDLm^#xwxSB6hY7XnI^K%ikWGXpAh;h<XslBC#M?+&Q
zZR75*!+}E2r^6x%Fr|%E_WN1xlRZ*+hS-NL0}rpypxv^nP65uq;hGPDZf<Tui@B%j
z{Hen)Rb6ksQ+#ecu=$&`IPVgwKl-{FH93(iR2ni`U4UWut;P~g&6mLf+m&+hQLq64
z^P1W|mAXD1qdY?riFYEd1<es`%d|@0v5i->ML0)21222vmuk!-4&La8=s)+6zu^nw
z)lTin#LEFs^d>fQyIv)po0?YTca?G15;rA(*X#Xix@b`6!OPeEksI0gc@jZS>>!SN
z=J{s2Tr?5zi~H0A<AvXKYf1p~P=(w6Xg)w^$+^<bsRE#lP+~H9EcsgSews0M-Q&%a
zD9#Wyzp|14)sgk0QYH@tF9@8wF2w=>i<;vKMt<_<S*L=_2^{6Rb<$U)<U>CE8)Z0w
zKyXWzfX67K;L(wxGCutY`P3tTXQ7e->@;hpwB14>v^|f2R2L=kA$uPs(BX_^`>74V
zJ^w{j$F*MKK}<W`VwLfQ59v4axOUHd9R|r#omJg<ZUB6_Tua_>pc9QF;*Hl?w;-{a
z%sOgG!xN$ctgyOrT8D3Lfa}-I<<Wdu+QG`z;$Urb^5QdqeRf2%`Udz&!PzE)MER{o
z<vzkr&A6?71;P3-kDP$+SDitzRM*Qa-^E4nMh>chozQ2J_A!?Jghuj*e2UO+mOpB8
z7pGI_qt!4D5+%udR-gE>1Pd*f_%T-rqlB+XLS6elEV&{39U3{i*8)tBLS#p+96`X`
zc83VS+J+CJ7Bi>Zgj3)IDWx8#F|gL$(`MYa$@L^o52$NAf2KFdo3nJvANExyb@v&Q
zt!@Mc;I|6=+}UT;D>4>vvHBNv=|O<909=bUu_AN*)mT$gnP0)@cHE3W+Xu9rl-cQz
z%?3_DOr^rhphxrp6JDdUR}>icwMmmjezCnp>qWE>3;g+by4v(pKku9UfDWdt6V!U^
zk?X3Xn)ikNC-2L=Rd?fLBb-8<(X_?L&F0`scuXz8rqrvgca$$4Efp&g2xvY1a)v=Q
zZYh7bpH{KcVfblab>{*W5J(O6hI!r|TXwroWrTsW3^Uviih-JQGXQow=4=NUB|o6i
zygY+bw58L|y1=b~Bl)BEtH(Ql{9xV<QTV{BjfJ}NMYK_kI&PXGoy|HP8ps(9C7lAa
z;byyo{_M&b<YC?aY;$>Vka6A^<WBQ7B-eOB(OEMpKAN*@9cUamgl9m-<XgDE<}_O<
zupLu*MHu`!Axw(KadER9rTsFabM-lFda=Uzb~>?9Wqlp~KDQGjK=T^)QTAwlbBwgS
zhgsM>pcA2Gj;8nSfUJk~k6baHLirEGtJ<qtD+&@kx|&-vd&3%RpSsP@H(vXAN7Ogd
zMb_jVm(a2;0n-E_4sWy<R0kdxos}>1xPbhOAw#M=fZk-{*@S^<43{qF9V90@V!dW5
z#Xjnl%tJ*0;1U-&+=PI-Oceo}+~7udRoi}S<(!bHM(0;>a%VJjIN64tB4W60Ic3a+
z$8Vl<Rcl{1FZRRoBp;*T19$I8Psi5MIWyDb!=_HU^EvJ(jCfI0%8P7X6T`OAi<t1Q
z1oAx|@|)taXS_PMDy|`BBlSJ!3i-Fc#5Y*EkkgE#kkmzxc{0=GXkq;Uh*GmW0Xv1u
zaYci@vR<0LGm$l`Ztu1pQyfVmI=W@=-6?I3cQ4f%${^7Za7j4CU=L#`VEP!x*>=4^
zXMEgsHAs|`c4$or<sjQLF`X}YwPc<!@Ga5n;c|NiUjJ&J?FWL`eb|V_b-g?iAF2PP
z`0x$obYX~))l#*o_n!=Fa@bruINdj(l12jvn_tx8OA1vOa;=Ehx<{If>H(|c3?KcI
z-Au2a+lLvhkj72_8zr;^u_KH2TX_NVLF8fcA~ybSsw+1qEeAP$hcg?XY1PqGs7t7P
zgS_3MrSnI?%sB{#nU5+Q@l$MUS^*n9A2GMxH<@PQadx}n3?Uy5QtvIuDu;a!VW%}d
z);7`4*enSof?Dl#Pdr=PnjB;!dO1^HBeIy#6~yEB)!3d^?5o6;T@bj8=;jR@<&ajS
z0ap>DFWegPnB19J-XGPb=?AK=NmbGJ-hY)L^?`v8TM|hpzmRfU>)4GKl&8*1v6MA+
zkyA-jFIL9kJi55o5RW%h)A{)xg1MxS7!RnBAE%a#*cR~m=>x{x?~aOd@4x@f`_@>Z
zgk-$$!MI8^`p5v=t;VFk;<6b45;v;jkaew_{g7gE69~8&0=8UJzF!qSW?6jXxIJB-
z<NQ@p)WP+v_DQc$I?k-XJR<@7U_=h@4|$;eH$Y7i>))Adw)9mTsf38jW{X`R>9hm1
zLByp_f0%OcB8esp)A7(zu9cXqXXLYwlMt!nJH)D#cyJJjLB-~>)vacnKs}ZMa`quh
z1G)PcqhS{{KjiqG_b9M-e3G_VE><DC+PT82lKkOqWVAYiGzWwo2??e!scgIhdYi^9
z@$b$T5{e6q+<<#s)tsLFB1HcCVZVy3EvnB|_Ie^!qc;^?;u}?e#xtOOIuv?5=)|Kf
z_FL#+>+Ils-u{vt0IZ!7l|(?Jw0ys=`IPnXSyH+AuWC3_1qz@bN-xwU_-Gn=O}3QO
zZZWd@*&_SA69GmvDjc-=Mf)%Eb$B@h5ja?|Y@@FC78mg@=i2z@K*>t!deGvNrNbk~
zla`4AO4@Qi7zQ<W;O?RdOQDajPk`3niq!1Gq<$57(B!9F8G<Uum}_U{D}t%Wz4tD+
zB}70JG`f)>i5ZlWca>OGvjn4%rh;>#?^V#n(ZT#)(Wq83fMoIot}(wg4FilK|3%5S
zB7=^VFDI3AkIF@Eigi}Yy*^UjP;00pw>mWv>mViuUe4oX$IWwWW$0QeEvIA}y(_Se
zbcXQB>FU9ULg{6*fx1X;_LA8BvWW9_l${{ox^x(_juV=aQCQjA?FgRR6TF9@7KmkU
zsj(;_KTs+bSTqc5B<%|;dQz^q9o$`(UV?wTFuxN7<}CDbmdwaL`Dj}rA4V*48hTsV
zW~Fpl;Nr7MAfUne#>*86&Uwr#QPXB4!j#p6+TQfRM9;p8E;G5OF2cz{s?7<4QcEXP
z5cf?KUIyV>#I%M1=_h9{2&QH~(Q&Os0!53O^tnS{(@6McepD1I_e6VHwm5Cua)Zva
zU5_Wv6F1D}k!NGdu%XB<X5k*mCYTz$wHHPJPNdtW|4edPz$zMeXu~5|?jzUPN=H_t
zG`5tKI1FrG0v#PW!&eb#I`OY&P<^xvG#YFnfmX0R!6haT&@|ZM@6%O3LQ27Y&`=C^
zad8FRdxCvs@IY2&)>hWmeI)8r(7fvFhU<oNbDXO;_xj2_ocWF2ZsR4vB?6(DZze#x
z%tv3rXgjz>g)KuBaf<z{-|JsTOmyi})`W8!v)8i$8BIK5dkm5UwP1<=C^TVn%U97a
z&6sB)0CjXjHrN0n4xql!w2EbY2L|5AEyL>Mu%0%E6R0%+pRhkG?83QZ`F=Mkcwyrh
z4`Z<Gcn-{5RUvTQ$8C`W)n-Xwzha`$Ew}r)C;j5a)mQ$LQpt^f)EmsgcHVp7{%5=f
zW>E6MJQzP-;FxaJ1UdlR`E<;9ywXqx#{j>Zo8zOFYaE1XB^Hba0jE*JcRaRba2@w;
zN_Md`z^~AP83h6bVy$}*DOkOH%Mc$ift9Tna?b&l>x<jfJw~@5<Cb@|F%(m5I|NKv
z#_2R1Bb28K+g*q-_G?)MRI1H5SpeYB9-Qb2QE9RPjgJYU#&?>NVw2f=beZHoK!?!t
zd<yT}dV_O8X@rO!Xi$;qF^G7^8Q<0FX*-0^@q$>{c?^FXw*}H<Rk;7SQeCnf_qX$z
z_QI&EGC#pBm_TUX#qMdX#9u?mK;a`ToUO1N#=M8vSIA5(YPS+_aEY`)Mm-qnNRhpC
zY9^gG4jyFfxK%q|YpJumWO0D}gj~(Hln(h8r(Iq14#JF09=ga8uRX@o#2@N3&Khfj
zdk16OXVEyQV}S*Nr9vbk=^_N4bovGBf|Ebzp0?FNN(we2>qJI|ETA6Fvq_+7*Vp_7
z@MjK#SW*50@ht(QC8Tc-dr=^~#1--kHRVaW5_v$*vj{w4Md+?o;13*^&>_0Ha`=I=
z7cq*n2we$pmk+N|-azn7c4*)n8tvs&l@?vuLVu3d`ZjPevSoNK?o`ETcv<PL&B3Bz
zLe;T{X@1EfF)VVY;T!BVvutv;$@Ewn?|00tO5O5gI1Nq23+MCsd&1%-gLhmhgVQC8
z!NcQ$xRHDLtV2A>$pPS92^s3uGq&Tod>2eb3IH&D_ArB|tuJ-iywgwo(dO6ZbCPIx
zgX>)?B8Zp1w%04fUZ0aq$Y<V~ZzkKDH-mOv`$|6I(8FI8ytWI8*I8@ZnjCRX((O9S
z9itds-7X&0I(O<0h$$TOwAmcYYaj4IZG=_yN}wf!_qN!mYps+L;gIPLhM|6p<LV1m
zI?5B@sr*pN^yU9<g^I;%A*V`MW9xkm&^m*uEHvFdFrRcA&!m@q65jGPNdvHF=M<0M
zVPWIaC+o$(mfGx4?X^VCoEvG`CBLSaXsnC4A!t^`p(c>ptG|r9=?Q%csaqTJ?(aI4
zWU#{BH6b?H>TWVtzlwZwZK7UF*fQ2HdFTy+;W4J&jFzkxbi3R{uGt=)eW=ogo*6Ay
z?rfQaFw_iQ&Pw*vG3%?aVq>oi*e|y>hdVzuM9cF)tV{nUaxJ<lM%6LV>V^uD_#VPI
z_;JeiBP1?517T>w_7Jgq?4ST5gjo3w=$RXi{Yz#8m8mM6#3PgtUIIiUkeX9Gi=`rj
zkw73b3MB-Ymu<9VKRmB(ToFQG2CHAdBxZ+@S%IV(FQf%Su?1NwHvHfX>j+s`sma!4
zXQ1Z@e~!c1Xa~XA{S$tWw{4fRRv3W*l}8*R0`izU+50)N2GgoveIAw#S!~<nslgpA
zy#nh(_6j`=2Qj+x9%3(;8#KYY`Kf3b!5%f-m5t-49z-eJFqY|1Hs*L6n(%F+mG%_R
zi3W-1*_K`?H9om;8`Mu(wimPHD0L`us->F>D7h;|PT!xqh7j(E-OE>_(Kw9tarm6p
zKro+%{1|kw4n!$zHbfSh^ha8lOq}-wJPzqa%F(DG@;|KAl`F6Z=FlkVDb`atj%^bV
z9HhY74l`)mzRFKs{VuW9aKZY)`fxgb1&*eyTO(lgYU9g<{FO-Bo#)&(@kLVlMfbFm
zcC{%ia#Ols_P633UwLSFh4HzX1!!{7CGXvuFfSq@<8HkUpPN5AnR(-Ad3-$NYk|*a
zqe4!hCy<NKM<*ePc9NN!Et`@dD58<5!$&LGwF!LkiZN2UZ?_tj_wA69DYi}$*Fsx2
zVGjg*i#l?e4fjXM;+)n$wM*8g@69ybxjUM~CHp?{x-MRd`S?WgPConW8c{nlRMEUW
z{%IyJStId+0a!7xRacs>BrHpluC#4%hG^;DTjQLZzk-6yQ>Q&nB#22^eBK*W=n*sx
z6)EU(grZjH?L&1@2EcBa7VLe&;wX_pJLn40bG#_m4fF-A-$JQFec`iKPJz2UU2FN$
zC8M!yL{@(c-}aajurO8LV@cdGSqbaW0*!X-_G!rmlgk6~<{J>CynhFJI6%Dl{+quW
zGw|i=$GDEcU}h$!m-8=uRX?>$=G#%mx`#GjmInKSnzc*TF667G9r;1n_a-KKPGOld
zcfltiE9a1S-DS<8Iqm)otv&^mDUv2CnTjD7+0blsj2v=wnf_!c8-Y~p9vO1Vlq!Lc
zK7<#8@PlYX5TuYfw?R)yzm(`dJNFo@1P&k+M7z<qgin5kZDhKOADNCnLyjkr2=TDD
z`xbLo^pvn^v-)U_)xDVdMY7EYDTc{%p5tn0|LfJFnoF7RR^DB7)=0+cGt*3>^R^(i
z8w8W)V=6M#$Qkt_(K5cl$&P721>~#`lOJeF!2&zJK594mNS$C<@cTD1sax3N2mW9J
z00?SX8e^+gSB#bX1%+HF0-*>=4)q{IEhQx-*oRtn{c<%#M(Cc!VEAJw;?oxnxU%x3
ztpSo#Et5?sfKDqN_iY(c;zP5A|Dnnc@*+NnB40?8djOa!_PglN&iz0@Kme1fT|Mv}
zQ4H4vx&B=(q$(CgGGwYcVkK~tJ>yCwtHkPMAPTdEFToL$WK>MN3bldI<2F-(89x7}
zu3XbE8+jF0J#Et7KxrKG%8f!QwB$kotJrlF8a{y^?h1x-Xb61uwXAdPKovM7M_Gb}
zz=jaGJ0^%wf#u{n&tG|Of(}#nGs9&3@A3wNcsvHhh0tw9nQn0B@oRD@mY^`CRP!jw
zcbu&4qO*-B;lpuUJiE#U02WH?e$9ci+5)<tVL6`r?g(w}I2rPj`oUy2;`XK{lo(D9
z0c+<&0fU0#lRaQtXfD{GhXBfypB5Dw;~^LzeJOVF^yE1Mek(DY{Od-1xi_wf5fqVy
zBaV>!)BPDEPtag2L4gSF{$L!@01BfC89W=CCKT?oZk-i<*-%y-y9o?}j46a6Bykhx
zpPCRPqR$YY8Yr8!tfs!o3VM_EKGY)xqotv24tcX=$OQ&=Za-73Oc%<a=`&SZc@#5{
zm=kUtCWvqti<TI7b#8M*n{)6~%3f4@KJfx89IBUp4iC%3sJYO#F!B^T73A(MB5hVT
z+3E)OD9umPWUR`r`3Z=~RMXqoBnELV7{EYYO;`KCxBE`)SST^1fmvLFc)y?iqm;}T
z|M{2ZvXZA`cck(s`TFGb8cxgepy2dB49N!}B!Uv{HA_QN<afMZ&euEIF&Hy9L9tyV
z@2lp67O|N&>~wxKi9-WbqvqTX&}7@i4{MI(tw1<4sQ3rfq`jr;mwE@(e(h0ak)^sK
z^|E&nR-J=J6U|ccJN>9$EZPRzfC#t>d-D4%5+rosxPlK3uebBtXx|HouV|!Cd4wQW
z`T{}e3R?GxVsjj^l9P+<36b;G(Y1@qvc1}|J6^7fEj&nJ!=FjhkhX7;3bL;uIQwqT
zkpA!^YcVg*ZAGDP$@}F>XmrZ!-jI)WND?rwbNfB0=!fE7{d+t^^LN1j>PDi*k<$l`
zEBi7nBOu5(&8e9pc@as81Drgb8_LH~m`nBTv7~uv{m^p4-ihV!d_yUm_(UsZVLyAy
z_gr%S&2JLD)c>Q2><(J8;9y*S_*AES&iL=s*;4OE9~9Ygx6I-a(*11CbPjz|>5K_|
zyVza`c*t&@*Ij|@4iKtlCr@VRxMyjast>^#K4&zc2nf!SYgs$DT&U{loM^?hF9;mp
zLcRw@#{(*dgrQj&|HJJ0gnHWihKBcX|AZ9-H-L(|-?Ww?!%_#^0kMw!2bS~Ct-;*=
ztkdLEEaif=xv8JAKcADmnuLd+(Y$$e^dx$WgYu6zdEK5<U7pjG<RMd0Q%=btN@E5$
zC>E0EpAZ<@Z6)IBX9b6(S9hC&dx@N?lF(?IJ(cBf6CBwT3!RfNchE-Peq4N1ZX!)O
zPLD6s3K;2pqc2q)44j}VJNx+YNmeQma&;B#*U3lI_vlhlE9j9w{#jq`k9Y)n_kALH
z9s7f*kwmAL5iJ-#hAaC`bW;$23OSvSUoW?QEE^0^ltwS_qo6KJgTF!q?@;k|V<bV5
zk5Q)?6hQhxzVr>*LnI^xt<>AX@W|4?a6wR+IQGeSGeWsg^%fEZ97+qM+a>VrMAEf&
z0h%H$vfghe1k2M#)sawJ2x1NlA7HZp*fzo*;wSsi?eTmJ9$IBlJc-~r%pt^5n0>_C
zFMk-^T7bqVHHM`?207^ul+F|>h@BpM2qH=d<fKshgnI~=FIVMw5G5Z`JK^yqsUc{j
zLnCGuf2kK{d`3qlq-eeit@(vnJrfe8%YEM`B9nfJ%dBS;Jn13|X#mN{Ep~=b^D=;E
z968^9zhX~V$NIvG*zM^{a|m$E)nGn{HlT<ZbU@^lU<P|k%iv7IftZ7L4yn2?l0u)M
z$2-8!o8WJ|5KJj1N2J_#8WuD4LW9DvJhpP*xoivWAjJ%bc_SaX5eA)tb6pM5pY)*m
z%!SO$Sqxh>IC53jLt)?5+ZUC>EZ)s3u+EDuAFh(NhoW9$iTS6KK~lmq9t42aWgx76
zS-d=4aXBopp}drhCNl}^kW(+12Uk`Cc)w$Oti!t}0p|^&)}x=9AGD+mL?H;#L>^4k
z2Z77Qleqy{2coD#ze#H$q6HQ?ks;OIKys}OVt;!%!rBeL8L^u54{5BnEhqZ9BSah;
z{~kJuf8~YD1M!q5TW2^dxyP%27*}xY6cD5C(wH4I^RqhG74x*;)aUuENs30*;k>&}
znbnuZV?07d^tE1uh*_^bayke?_YA4hbkCjEVF<y2WHNcUYnGEV)B2KF<Yt`nm+S7J
z-kpsgPhzmG&-LEO&I_4U8A<{P<Jn*{<h0sA=l6_^kyAozTjauYFSbo=TwJ@KSM37^
zpZVb&W^FR}%q_xS1uxp?KFw)aWI`SerK=(4BsUv(H<I0B@fdF{zplAYxgs1F+qEB<
zZ47m_k~cfylcgU}_q3sI3|)%EP6%$>hXYm)sZ29}xFCk8suoK%%igQApIjOY)Jai$
z<L*9&J(HQ^>>`~FJPYApLe_?%2#wu_f=F-QEiN4zsv4tq8OQ46VMsw;2rZF^v&T<_
z+F%8D;cwy}<oaJn-FN0g73_XD@LaYW^^yt5Tyy_hq-Oh|3H&aocwov~mTWFYem(l>
zj638Z!<G(LK`dF(QBD)2$H+4T>AkjzIB#8+yi`)QkOJ|F!9enLS=H~8L&G*-A5J?|
z=cSwXO|H0heNB*5S6r&jmAniGKj#C{9$qO31skJB5Yc@_9$%NMDf(fc0^d6HCc5Hk
zeJ1!`&vW<u1}bLm#CC+MSDi@Da!yvY<F@*&*D$F3Z;y&wzGY-A`Q02AauOfiW;&a_
z@-AZNH>?lepHUe3mCMWWhNDgWagP7>d1&pEI&78Fugww6S=&tMu1A_-lYS(lgNi-w
zHitLdv<Z9K$ykt1X1#@6@S?(~j}nI~+6%k68=sq&8Jp<ciC;QXQQ~4RRQsRPVjvct
z_KWyZv~i8hr_&|*x5@Kj8L|m;sDl36WqJKsVKg3|4ObZ+=oL<H$GNG^f{OJv)((w{
zsa$#Aw?_w>GT8~?<nvCC)xSM7xEH#;Irm;o#9GT{?jyX^>#FO4j=tmO+Ws(|e)WZZ
zC)21=bvsl2zNlgU3!VDY;p<1saqLy;Tf8#>d5hJZ$v)Dw??iDtb}6xeKVtx7+H}>A
zQ`wK2@IAG&sv{jujAsrczE*TP``S2F`3DJ3`4xT7^o}^joE#b*mhN#JvvLEgRbd*7
zvZAP^O-@OQzo#`>3`-p57AUO0KoJF1CT$)T+4Vh5-`lfluB`g+VC3g*T3ZxIfA>aN
zZ7NS1I(>gR23I2-bSFGtZF97X7PHUhxsybK<%D~ds+8$qhwE`t<@(_7-efM3a&z!~
zMGS6pNI^p1H<xZzpw(x%=~Pm|^ETs!F*Sy#$&gPqm-}prU@5ps@rsPIQD7qw4j~Lb
zoo=Dn8UC}3<juhpnyE7TSI~wZYd=}NR3<V57lR&E1+Xg>pwyQEI}HV#>-=z)fOz%`
zq;dTKgo$R&mlFbSsz&<R3uM%3f2ZI0aCu?)4nOJ#wg)8#+J8)!vdJpgGP1I=_6H*>
z{KRiZH#13_#fAROw2WwN29aYEL$KDX%LKMuP@2K@pAzdv|18P0+RJCbR&L`Aeg!o-
zz!PKy=+^g{T%@0}J<!!pi9AO(P7iH#_(kB8JCD$O_L4v@!f<3q;P2leQ1IND=r`f2
zK^ez%v>|jS{@mDKi~NGbPPbU?&zB_G9^s7@W&<zMAV3xdyux@OVNRo%GgCIh$YnYo
zW1WT{c3V_%e2hcvPrR0+*b0qV<DX{bXro>w{d&P{aO~C@h%X-hho-ZPit3Hl_zWp6
zAe~A{H%NydlG0t0(%m(Hf4WOLrMsIMq+7Z}y1Vlp?p^mId|=HQ&UxRxpZz?)4Wv$v
znFv@Yp2K(2)cpOrwp9IyOh}Ai#=AiOjKS{C(hVAZ3y=6Y*(>>qzkAKSnjdNOKSJHo
zI={K^!T*&FBfBeH1f{&?%noZvmKW@Xa8_f)#rc=#lC2cIH|3pThp+k(?%xr#^(S=q
zHMV90a}}Hbj-ijOP5S8QQGI1wHRywpB?cnBgk|NC6iDmD3q9%6^;R1BLP(Ibj<Nq3
z*Dy$A1e(4XdtjF7HglL<`u2gWhmO`GMHila-(ZijU$bMSqS&dUWVo+~4n9jRC|nCx
zdQ#MSu6`d4wf5U|0D@5BZSVgK3oR^}a?wdHn1C9Jg-i@%Jl>qd$C^?_pGBPm^Y7FJ
zqwI?7=hFL;PLei44fCU$muQh^;IY>CI>IwDtg!I<L&RgZLb@*R_FgdS!vmfNIz`j^
zYfXD}CR&v%cxhhP_1W(<wc^>}Zn3g&8GH3tmef4PEN)emy44h<<jTsUysY{5q|(Wg
z7|&%gZq2{ms`ReCAw7%0bkL73OBb7gl~+GdR$L`f&HC6e;Zw(-jS9zqf~H!$Ime!B
zV<Le$QDlq>8w)>%={{iG%+qBNzWA23YIGf?9QaSYcEXJEJCIP^^74V+#Xz4M#Od{h
zre{+VYqY25)_Rv2_r<j1ExQh)6PB|(6^7L!C5Imu+f#S9L#(CZO{2<-mVu)O%lP{f
z`9Eh*8RNJ6Xk<&o3p0A_dQb?lVtHC!nNif#Xij|Gul7ZYbQ-&F7;B<SHpBvD0GM_D
zamUx2yN^p9adCz}y5C&vC~SI}nVD(*_=5o5a@n=lJ2rT|7hLtQ+&%nP@!+u_%eA}P
zp@XDb--KigGvo_?!k)dJwnCIh^q6Z+@s@olrlD@Arfl(LDemWJMfR7Gs_w>I8;pZU
z9r|KhlX)p`Xj(fO_48peDeu`DSDRH&Dx~@sHOY)VS#vgeQCObNU2Imz@ap#N+RX>L
zg$}H=rz$=@kmqS8)4n|~U+E4`$+_klrzNy56@Ji^ljSu9RKd_a@7&a_7yIg7Gm=C@
zG3=7__>*PcJSWfggvpDDA3i^kA|{@D&^~azkefMuf_|ypp?`X|8-+&KL?20D6pZ9X
z0CW#}PMo7A3$FhdCIuuyJH9CZD19;P8T<$j1+?V4+l_`CttwgnYOF(7z*N9#nqdYD
z=}l)OGF=P$9W~*PcU0cE5;pM}Sn9R&1Kr6`rS^m2FkCz#yjETfm0WXJ3KK(Z7wU99
zrUty3A=}44J73_o1=#Wf>J@O<);U5x%Y;po>$j(}K-05~1K0=?-DrJFs!ve{lo;5U
zseW@O2hw=#<^Kv1+f$A!QenprMC8HOYOp3cDe}={3aCcB9{hU!mEHI!&W1gZ030TK
z+b)m}R=w<_@RQh7rwG7SY!QJ?p<Op?B3Kur`(lSfs)>9MoYL{Y;be2(<CyJ`SbSsF
z48&Pt>lIJ%;?C2}Dm1=XHLF5o)eic^1udQviu|*9V8iDS<weZ&GfwclpNg1~0Uj%V
z^CMXW^RX9F8pDp;@BB1~3%)OMYeCA|GXm}*ncLEPv!}H{Ece-~3nT3^8Kf~Q^ku_f
zgj&u(aSz@bho28PH{{zwRsAvnca9^e+{!rYntH}Ib$@!Jcdii$bVUs<qD&GoNst&B
z2-U3evh7ausjI6f4~7Eq$j3z6*N7slQZWPAqjxX}<2t*osgrAAazln+*-13>%G*fT
z47H6LlOhbQ15wntS!GOt!vC`!fMI4o)sUIdOc+^-{)5=r5q`Y~OdE(V8oDvd$Iyai
zfMBl19-Ytgf+HrrUR9iIfbhmW)nHJH;qjlxH#7Eh2<R>Dvv5}B`fLTA$I&}Hgtj9H
zCs*#sV1cH~wl0A>%hB1zFLemT^5|SM3zNVU!3TSb?*l1h{<&e%R%Khd&%C6j`v0`I
z&4H1;^!S$W%p!M-f}&{2y_Z8Oc7fam)9&VH?dM_KT$_&<$O9*NgK)M1dE~c`7nV^q
zDg)RCgcsQ{Kui5dI+)Yew##iK8T-2gTbivKdoiMWGCN19CQTh#2!KzE@!#*t_y$M#
ze&ao0LB?3`Cnq8mql?cQG?@%%HRF76=8@7E+TJxvnaT_PmEt^7CQ-N{d@wiOcTk}A
z_#oypCi%;my8>AkYwik06>W0inc%Q@wo&>CPWRC1zv}5-RXRewPU6QUr$25lyZiff
z+i7+sd&jMjP?`l7n4a5~<L_@)R`ae0iiyS_aip3{6cO*#@MIZ<lKNKsL}0NDjGy+m
zFr$Vw2d(FXJNn#D`9G%dAX-dNQ-~bMy6Y7u%#y=nYDe=-H`pNg)qec?Nyp7(B$zx^
z>|%%DTqtV{8$;!RT-eDB&##o}ANNxCl3VOk#Yo?O|B**gr*<j5)D-9RWKp2+t=CU7
zL<Ox=t%Mt-wVpIAziN2Y9+cd6^wRt3D$^zucYX=8ej(OTt3|TCk?%nS@iZNXPQ{*G
zp4Ban7hiBdRn;;pL{IcD>n~wF44GM@hJ7;Z%w;CD1`Iy8$EOyNP%SA6S;^Tbl>lx_
zljj}k+Pn9}P3kh?k_4L2q!K?}I{ad{F%wycQL%mJG;eI-+DT0Lz4Me733VWNSgX%D
z#Fim=`eG!Dq`Wcv%&+OIuIh;8YAna~oU8>n`|`YyHhgbo`Q~@706C>yM0ho*iZ_`f
zUhw7TOe7ciy%m~T$C=}A%U=TO{_j#!I19X{OE!i@aX_yBs4jpUuzcr-SeM-PxJN*i
zi)#tO%n1h%#vmAu>`fNp)*{Ic$dExmI5A+6(G(A5Efysdkktpd78#JR2T)nm#Q+Nk
zJ)d^R^H3mU(tyQDQd;jb@*1FjZEbCFUw#6E&_yW)K$vFzTNEvAsFk=7TO+VvNpA@r
zn~ur!@E;<NmW$;1<odrWIB(s|X?wup&ZRIi5V{U_1hN@Q<RI#41wv-z;BEB!>u3q4
zq<oLTyCgcR=cYI8gu>a*;J*x9;-74mThvNen4L@bu5!a!!sf96agi5H${4CNXG8DT
z#t*rr7W?RYITFQ&R*r+7RuYN%<p~5Lxv^mi@XEGZ?J$aEH|`Aqv9Q;Qhc@G6YnctI
zZBp{zt_6l_<>Ud2<GU~9<A6T*2c$}K8*Z-i@y*{VYQox*htGpq3j7gHAdjm6`L}b)
zCQ=~M1PMg^q;p#B;5(rt_UN;K5sngb-sAV2ok*A#pWnceK-xxuz<PwsajC=;NA+!z
zwpoxS0}5oRkKC&)fqMUsuTif$dpi$9b+PP%XhK?AY5Q0YT+0g3Jcs64y?yRw1bLTM
zu&#VV!XhVi!kdeLU=YU%KAv{{jI{m|;Kn*F7`pnVy2Cu^;Exv{=pe4hn{vy=<@F8y
z9gKYLnMK3%d%tw7_%V^7o2yi2)$1fg%c*rJ&Axj$pvLn8=&W>Bn)WTGk{uO}Xc8RN
z22NqEC+aG{ScDhqr1f+v_4QLZ>4F`d@>V^j^W)NR7L=^Da~eBPJcp(N(sFn_3}2PA
zrRKO#it$@!fmYwAisV*4WqyXfWcEP~9Cv4QYAtsE%JDxO%JFYR$gQNON^kU<a*lsy
zsTA277dY8eg!&S-shc$uGrzlz<W(OZpnI4Ly1Le?>>Sy9EyaE#U#~279X$8)(qFsi
zJ1eb2SF_xWxc0u7=#-Ox%gVfM(6*6fICrR$BxcKwkfTItNOj#WPT?|=uwH$zR{Q;t
z%cUes_^w+kyEM1tcM(|5`(_y=V!-H5*77rzXIbIt;n&%MBTc1_#<(365oJ}fyuT8P
zba$uIlLknC*||S?bx*QoS<ADxhHy_Rfyu!+u{s!V%B9}?1H;#(9Oi@|VGP!+m;Bu7
z^cwQmQB$0)$iLi(4!!aE!ykvxWuB2Zu)6F-q<dViyV&L-X1Ul9cS?Q`_Bv*%K#mYJ
z^f6<vsX_aVr|XH~akFf2XmQnYQv8TQ^y58#h>8}waq5rvwUUg3B7VP<%_LL&j^Ds(
zDzoMs^{C!$GkUs2I=wuc;E2Y?y^Cm(QBK>h!4ycZ!8^n)`0r(4O^3^1IMt&&Heq@~
zxC;xG1EQ71S+cI;V8HvD!_xRor22ZWaVNx+({xM6v3W`OTq>kW$ouP?=QFR{y57~?
zG`g6^B7KaEY1|XrR+W_)w4Ot}F(FCPT2Z;u0~yR<IRGt?KR2SO^+l2@U;@3(>7@Op
z7Xcqd<30QL_wsO#K|?=Z0#I-$_1hs}yU}#MM=+H}mdV~@Nohqc5I&RK%HXxrPhc5G
zg+x$SnKkJ{zQL>s!a-AIaBPOUDeTNnyAF}_ng(=)FQ9d#kjW@eM48chm1?GggS&sh
z6cRDJU}RF5K&{)Pxke_dcHLL_oLx>(8@yoxQC?1m?q)M;-RS##s#^MJwHqU*8{vI8
z(3d!Wcw~mY-~<Bz_~KgjW(9mo0?Ty9tD=;h$dyEswkbJSX&2z^kWlG3oU0BbfC5|E
zCS~L}tXDfS*pg_P4{L^~>rI!l#R9wAKmpQnzl^r2KMJ_)OOSq9QI`ifOPAG=@Y-%-
z`Ex-2SV|fWME=()Fahl35bCDD%hN(*TPcuFqlHta8Y+$ItU>9ukcQ!CkswB#G9YOP
zFtwewcEix^88j0n#bAS2>RXzKW4w5u_kekWJcaLVPL6b&oAd-A=)=XA>VaeQuDokG
zeg*u~#sOWbm31jUS7G}82}*EZf&bH>{6s+Uoz47ffH|KxMd{kpa_F77=?0*L7N*&C
z@XeSV4P#xO1E1ws5Q_q0vlL-*57c&af*Mx=J3sO8&3&3r`FMV2_*~Ebn<Vxtk2oVn
zT*7^QT$PXg%Ut_n8V5Rz1Y}O2@g;swl5(6X_fTyGuzqV2bzv7u+(J!-(Y6X{2eW73
zNA%4r5{b;iDSv7vw_HrKbZ*Ai1wkS3J@i|blboLh#W3w$WiT|w=GpYbW1zChp^g6a
z{UT<rOP7~7FdaKJm78PX^?YnuxV7WkGc8hos%TyrbCqCBB2u+Z?_qN@&tx^Wz`V0f
zwm=saJy$K-N?e^1<)J$J?=id&Y7-0ft=o;r2bqpeE#x7q@Kj$+&V%rbi{Ot3%(Ei(
z(ZZqEe$1Z`*YxTczM>Hfo~XafsPje-4t;JDR_{&Q&d3rV;W|BSQS%)ny?Hnhd(Olm
zqxooxAtTyD!W)!4-4RyNQWD2oli={Xnk*p;nnBfZ{8zMD$Sdi5!GS1VKjF;FlHRVD
z{-2${Kv7jya-Qvxp&pT+#`$<3g$7+@KH+fVNr#T8*O94UFf~k%TaL%}FrSDM^|sX+
zeAHZc=Oq*Q2Qv%uy;(jD$MaGt%M{emC0~TMoyv2?LBh@R)AWqLvYz(kS|<d`ecKd4
zQr?-`x1<ZSWb#<QuR~SCy`vP(Bpc;a8=AcDRZH%#txQ`~P@Y*u@?dV_LXZCnemn|o
z4%9(E_jQ%Ny(7?L;i)&8sm&EGyJU_P<DMsLRa8%N{b$Yo!rZhev#yqyJoF!$YULK9
zWhWiOnq^aZB9!Rin9dB1oy5T=Hl3xA?XwM+=*Oz{@MIBm1ithXzr6<~y_>1hZ+vvF
z;S*8}44G;8z{UUR=CYPvU^E)#4n7zc(&kzGyhT@XZ&Iw^E}I@0DUG@sHo9o)$KjM7
zHvMi36wn0t&PU`z_6}w%39iiEJAexkMY%u-fn;dt$*zbn1C$W>nJx?xo;hx-(sa}^
z6l3#Dj{@dXltDOejgM72@NK1V!Fi-}10c(IgzS*faT5l&|7QVY2*^q&xlnzHc-!^<
z1@Jq{&^CPCC~+V@@B!0`SB-*;aOpvk(rd^4g09gd+7a{X5)_>UAHnrE3s$VD9HwD5
zAj*48N)p&r9CezHx@DjY9r!~m0s5>ff~d#OO(acnC%oe==<|Iz2!|tW^>B(Vu~Rcv
zwA=7L{Fj(AVt*io+8AANWa!Lyr_gSgWxoCruvRx=6Yv})OeHxdYaGHA$=;y}!vf<V
z#eyq$DAwsXY}nSxII@LPQO=VX6vDdJILAYpo!f@VtANcZK?rhketw>NS*!?J3c=uZ
zML<6j^16N(E_^*_4E$FS`ZbISG?qXAb5wsW*(R|LVZ&Ks4JTm5uU&E+V$NL&tFc*Z
zZ~*g^PsK(E0s~+^&U;Ba-5w|SvMN_ZLv;znDn<&wM|No?6t)f+q}-1;JfFO6W7YGZ
z+z1=b$@kO6`PNV|=nYN=I<R>D_+j;!aF)AA$OZ7FP+|x|ZbsYm&BFk>Az|{mI*(6W
z`;g+9(QL@7h3J7hZ#WPW`=B-#Jt;|Xt2FGRpw@hX+l(lq*^(fp*IIQS+_3J>&d8Ew
zmmyw;kU&~icroIeBAr`1nIiW^C|wv;nNX|pjS!okFL$NsJyR{m^+}eP)y(EOQU%kB
zu9aWeNp18ma?g)CwJxK4_}IkWd@~go7L@wGEXt3g<r{;GZ(zpJXI{YrHG|U{4<35l
z301uf78|B;Zy%iGO}c>X6$BnFvZQtk?@TCN>EYLL;OjEE=%b;tbA80!Uf3+HWdXcb
zy;k|xiOBIBJhh+%0XJI+xJ0eZzbtIt6`mM(&uta)F73PJef>!XxprMqOelPr$Rh6Y
z#LFXH_&{{ntxw;|LYc}@7dH`KO@IPleuVGyyoYGJ%8^_>nKf$bUICMK4RtWIR5R^`
zuS#IqacFg%`1z63VeGf+LvQ&i(71pf$iFTxhCT5VmYMKyzTd<kq@5k(v>j-&`ly;X
zcZ&avwL)+(CRm4I%F_tlCFV(8yg<i3)o%A?Mr@Suv;Wc@vnur*ZAxj<`SpR~#89Ai
zUM!(hv+YYu%2e7Z<<a#x@6>K&kLt^~dGTrQBE8+5WP+C?R&8#<o(!mR9JRQ3deiKb
zP4uG_HcRDs#;mbrgx#}ltP9o7)JjkUNvDmU(pp-gkzRAyslSoge>iIo>BleFDo<Iy
z)cXf56K|%9jp)Ps=l~;-e!Dgo`0snd#ltWLt}>#7U#Jobr5dct?CHG|y+Al8d_K~f
z6)1~};jG|D-&7^(f&UUQ`NR?WCpeUY#i!69wuV;W8h=^TzIy!J;lZ803ccz8g>3vX
z@EsSe0O<mOotsB0+fba#?=uzs&h*`%Cd%R5H;#-Q9N6VE__=GO1)}NXk}2VY-OQ@o
zK|zAj!-S;y90L>n`C2#-nvp-2vRm2kUoUz|>^A?ZJk^ijMG|u>o63iD0N4Wou`ZD1
z`vk!X_U9f(L;gNN>d@7!IX8=91!?4L$Fen__KfdcL3?Toe43r*L2oBXJ^^ycz_BLK
za$qGF{|H|Bjg(%8tE_MLrVM7tJc9<A565SoD7EV0)BLZxT9h?;Lv2mLXXTU+I1-9_
zLlbyo`PqW{x)8z|Fo-xLtk)m!E+l<4QPlTk^rw+=h0l#R`>$pV#tuo^|7cmisQHtO
zY!g>?$a<VcSS~O4QWH2!Y|$-d$cc3}JKZ&AQ8t;cBmVB&pirCo&lOlrj);!whIbP(
zCDL}_|6PxUaO_t85(tanZm$<+`Dz=(H?)Sb*O0dw3r+-i9ln<7_wJN0OBeKcd(TFJ
zCF(xXixq@opbPE9wHav9dAo<v^%D7+B&f)Hk54(4aP#^W%Y%d)|G*_N^a2nC)GWUL
ziG+*7AWrB^lc>Gph15q6hgy|$4~gM|SVQOq+m!l%4N72;-{_}6sF;=}3+(xtK)-+6
zA+w7lJLBOIrJx@8@!zf@A8BIe!c(maBRg77bRIV~QjB_IWy_+#Ec}en-?rvoy^)+q
zZQoz^>Swi+CCW5(sOWf%GCWH-$EYAAznB+Ywj;BAQI-3m^5qOW`x_AY5eZBtVgB<d
zRn{Y$2bC`v+Vx3`BZ=r(q>drKX3fHq!Jm9!t^a&a8q$7=jUqlC!@2{;AFcjbOW(B`
zg6G`b{>Fbh_Dn-HayWJ9GySV!u%*mI(hfskt<Xb^?&==)s;*ldIe7j0JUw8|23330
zl|SnX45!ov|C`#Um{Gnn=BUk9!6*UD6yvHh|BfVl%4D!OsZDt7-s6{QSL;qah&=rK
zswp^X?LlZtZp8Y|(?qh_Jd4DG_@X)(tOtzjKF{|#XSZ?QtNhEKjiD1seWF$G%Jjt8
z{D6YwFqF73yM2Akh=1KqMgnO(FV~P1Fr(C+Zh~`i+|D`kA;seTYFZf9_2xJ%7tSTX
z1-~puVF*~HmSa*3KUshfQ9Xggw(4Ik&uTcM?Efj2;Ly!*Eztzvj>YktWuU?s*-v_`
z1%u!is%Vof)j+zX+iF{2&BWhfyu6SRWt^%A8M!QFea?)*8<1Rb=>AFw9*jf_OPq?r
zvHv}B(Fn^f%LF+jAx^W|4ZFny&;gZt0P=_w$W|*>I-XWZT3T9F9fAsk;E3}Z^Az>E
zoXt)JFBR{6%z+T`RI`(Lq}Xf}>qHKEFB=aJgyyp4phAVix_3eya`asT2ZfDpOSX=M
z9HiA`0|LuB(P(z6-8eRt?b`u=P*u9X6Vgi!reqgcO<{8^k!RaOhyC)#nE@9ezip@H
zG#ieY%Ee5hp#EYjND(D){Ha>5Ch?W@CW*AAxiH@9W*7VK-Pjs%UylEMH<#uBHN1cU
z@iz%W=Oj%JjL0{kCFYg$x_NL#9|FYM0(RhdSf|aTfBUK4q~rM&Z#TPdkM?`A({9ZF
z_;~=B(fX_?kIy#xR7%y$Ka4E$Hwg6FsBPmrzTT|=$WPBbO?_yoVsb-s&P%q2jNs0a
zM!c1mtfPOJdpKC9BKu4BpnYR}%AtlSCfVbzU%^~v@9u6=!AC3_Y}T4coD=330$JRr
z-`i{)bKB@4jo>_Pz2Zr7kdr~BzOXF1kIjPnk2^=u$8_+xbkxxTcc6M%e}3&~4bnY$
z7HD?wP(WT9x>!x*U(=DlM34S%u!+pF`zEL)PvMjBsY-P$vzKyR)6=gDLr$*+t>0}`
zjnlnR!q`Gy!mTsnaV?PcayYc;mLiRY<~=d@;Qe+h0#6V3Q#A5I{ZMwkESkm3(<I3U
zZfi|UsNFlIV9c2nUK|(yPDBnP2}G$0W@G$4#oUtOE7ybBn;#_Y0FZ+gbInxxbf86+
zfWnX0Z55lDZ+K+W(~#gy)Kes5;`9S#q;@x{HDqHitenwro4rK49dI;ipcE{^_J33n
zADY!WAi>adKr1r2aN6oV*VPFbZSv`0II}}W&Tk=!w9-UO^J4yXTL+|YFFwq1WnGws
z5%5@1mJU}z6h-?w9&LM^8g`GNaE#EurSySXIknpLp+MzZ(h<fId^wy{j-KL0Ws^)+
z+Uva5YWHz-@ya!-STyV`NJ2voS<MGzsq1~#Q({!$3Bv9)$tUoB!;AV7Q8laF<OTVm
zAbe-i&Zo1(0Ltwawkm}~`wujOK$1*Nplk2{J4(9H=mGAlG!CSoQ?rNw<OXEwi!>I6
z8xI~VGh4wtvuM5au;M*RL*8c~7?Ya{_qPQP!~zP$IiLf%fmF4z*#>0-*bxiBbO<1p
zff@Azme_?8iHx&_-Py`4oS&Fbsygq!C^A6AJ;CQC@7D3NVTiU089|(QjP(6_;lyyi
z^R5V|EIJO7xER$9FRtSMW23tGB+#-$b-J82D{7}J2m?3m>U@aRs;t^5R{;gdD(iG~
zZ>kqr&!s^@jRf0>9m&jQnwEi4?zVc$l!kSs_0^5fUJAS;Mv=CQ4dp*r@o6%f&p5pM
zl(Qy1I+-dxp>BUf<Wsr6tJfB=^OB_XsHV;C?+XTdNFr+UkZiCE%rURMo~jdqG1E2N
zc>2O$v<#*RvL+ga#tIEBE_(<2#nlcv3eQs}Kd1>rJMp{bL<rric#t26ZRt2A#4glC
z9qoQiXwqD)W88tRqe9>#Q|?*X*y0Q1hqhOrM;~p!B|$&aqPOW2z|zWOW8*v<Y%$Jg
zjVSyFX5>7t&N<jve_?)tE!-!i2<2S{VF&?T4BzGL#(W#m%*P$51!(SD`Ys&T^Hqq~
zrYB4b3!>#a_)c(V%GE<U4k~r$am3C}$;EccFNq4Z{|?zn>1IbNZBJ-R_$GqEwkJ(5
zRVzJw^<%34b7h+{OvzdK<`^J+<DncFoL8tW7w=t8*8UjJ6<;vM3Jm=a+2REQ6Oqo>
z#sEODVBWR-hVQIU-JXbQnzm!J$26+?ayg%)+Yul!)jU;`U+)nNPcSj9p>fvBMj@pj
zH>{iNS=q#F|9aTqV@*ZTsWJ)4h_942NX5rVL%^am8@+E~4GOY`%<$Fk4l7d7|Dc~S
zopeKoKWm4{W0gHAOd<2Wyx->}g*)}cMlCJM)Rf{#cyD%xAUfS>Po60JDHd?$@H|Z)
z5tbx(GU2;OI6hqBd98ara4S`6sE;OIY+duqW^wNeo{oflBzvb-9&K+TxRKfaqXtVU
zvnmz8=mRLxji$J$NHL&q{W}r&dCvZ4e!`jc?-V*6x@L;%{`r4wysziYyMno&qe%Ib
z5in-@P0l2GXwejHY7h)nQ4ZhOPh(4wV!LswT1OgwNizBO=fAjA1(7Fb_IK6E^8x1&
zc+9s>;71E;AVG(b_)F>}2#K;So6>j2HLvB$0%ZIMq+p5Pm;&RqNBAdN%uW}Em)-cR
zF_50tcY0NSo3fRM*VYvb%-i8%`w5%p_-bqc2Z)qbp-&oFtav+y>_D#g0;3cOat0*i
zwyJ<8+7Ra)e(IN;A@n;uZo(Ty5t!u7svop>J=_XSloMpQ#{fDO0E>TqQ5~aZEL)57
zqyAz*pY@wF6&?hD*s{V>aIB~5)uM#1H8G3vM`|L?Te?dauT(s5Kp{@2Yn{s13A;d`
zX8UV<i5oh-WdIf4voXTh8|!1>m-vz7LaFx4K*NajT*P80$X-e^_FY=lSa;Csx6@#h
z5?QDWJXObKRA><~Pzpk+rWekzdI^0GzQV!kaGgl4i+XPX!dHF?+4Oha4j^fUaJ@3A
zAP`B+Yb)0kjOA8`Ih*{y=JJ330sdX|X<`iw#rNl&3y!_gs?QU0Hb-(<tlhB_t`SiD
z){RT)1&`<EUZ`HT_{+yLbT)0=k)4LtI6*8>+r!__=PsCya1?Qcas!UmBaGI(M1CWL
z?s_0QjtX(h{h|uYGwt|jEuxW5ba-U5zonJiir;g9UYm^lgNiYrdc3u|()VjeTbJKQ
zbYxuI68ziYgCupyZnmah&(Lj^&^spI8f7)rFI$*yr0d7jP}!85I||pL@80>1E>lxR
zry+UltooKIoj`P7tM<Iu<6<g@9PFIVRcsqr)%C1!0|i~LI!Lj=w_E_!cdM~O=RI5=
zx+PC5^7a?s_eYHemW$6D(PFg4^rpRYG7frddfrL<u8_Nji!mpU<<<Mgi$TU|=gaZL
zI?KdIAE9X<#{;+5_s@ge{#c&F!$u)I74?h0<a0h+@%^u?+yxb&jY|}|X2a9>bWzn7
z3|nh}CiFC33oE5;tBU^!Mi)eTu9qzPU~GpZ&rePCz(rG)Pb*V(OS*I!>?#DFDPOro
zLj+_z0W51xDY?qTqLRr!wzkoe#fSe|LvOp>eyfgp1)^3oMG4aCv@-0BhB95>u)l~0
z=y57V9)uBO{<HTUOU8>95Go>^bs$ktfqxZ+ug|nC^>dS9s*v|TjiaShVTNkAwo_Nj
zZmaZNNl@T4yaZmGzRBMirZh=H)FtA*CrrC*{(~Y4)lbFZ-{G0427m{Qwb%l{FXygQ
zC?hVhU9=AxsCSmDFQ9!fnY!r05_t>rO5?RN!;dW|^)zcml0&4Tk76|T-qu8H?p~Cd
z=0<Sh22m@7Ig^1)n3$NHYlx+6mVtyrTcB5L+!3I-16g>EhjFeW3Q=!E(Gx)R4gT*i
zpG+3LJ9caGh!D0909Y-INj69f?H+s59{@8+fsT{>B0OcDWqg3ecfBBBI-^l|_i#BT
z^78J=Y%mCeB-Ipq5}d5=UvQ>tzdui@J}LuQakU?2hAdm<p_A~)!LJ=G3Io~ONeuKR
z0&1jnfVHq-L^R#l8R}#WrxZD@c~mO8tzfMVqkUg>7rc{psvGDu%v&uiqUO^TL>@{*
z>cqK?OefK!3M9%U78OSfO5{~t7vNvou4Z)D@c>YCJO`9PJg^25iBLOYq%Oby^wya8
zg{`-1!~1G~y3S^ir_^l8`*kif1;(19aO1ThbHKC{G+@1^t+QcFlxzEM(Br7=G<S%;
zbeTP&dXfCy?RSi$CMPB-%=kP@Q@4GYd8y~DACqAd&|4wXG>%lVuUcgMTI?I!4lsCi
zSSx{UjFMGntG$xFS9MOI;1vmBL2GFJEWBZvUejOF?$>%#=+0z+E?=kNa)K^v({kA2
z1ZuaU4{?7FZ1Pg`VvSmKoQ1bI1Qw7E8^!2DB9fMSqbt|#aERJ4*9H6Lfj*4Ot%4Ku
zZ`BuVyhv&5^00P*hCq6AtEchSVS|AvW>*kJ+oYAkoibX1pWfwyNi&nRDLeh0E^n;7
zOEm7Ge{ZDF<&x7Kdd0V8cl+G%41UvzFg!`Y3Wns2hZ8pxy2D<}z(50bV{01tF!;;M
z$-76rzjq#7Gd%gm&jD~)LKkwFE=R;`9!q1JO55e#1@(YwpPT3I{b$`bS<QV*q!q~f
z1;|UStJ^MDYy!ID%x)@Yh)<!7DH2-6?%V8xdyc54?3@)Uxp05^O;?AbP<IKJYP6X4
zUEab<=N;d|(A3peA1;daC!;;YyzmfxO=2qE&Zr-1lBs%<aL*(QJ%$rZ?dpYe&Z}kl
z04sn>2v$qgeZ@K>MP@$+#~*G8;2x%iTI!}GIxz@M-z2I$Rar_Qu6@3f0J)!mwGl57
ztW*Va;gCqKj91z`1{u`;nI|GzQy&8PZyuR>#@ur(SdzjU7((1vkUof5y)49R5sC7m
z8#iMak_%^dSQ~v-fRhM~=#8K9YquSUqq&??`8Ez1fIsO@eDSHF+A-v;|8RD#gt#WO
z_dC8!)s7EK+)x&T$L0!um|6i>U{>Yex<3`l1F+_sQ$WAJ+x9|r?Aq0gX+<_7IJp|M
zmnin$Y$GbZ2G3{3i;Rum%n=^M%^?oR2rGiskM7Hq^1di@7x3?8axib&Bh`s*VFi~u
zE|DfX+BN+{S~*a6swGKg1IuwQbW>KJ<N~<7!f;SqxOok3ZcDE#EYcThV9a39=3dRS
zYDg7f8J~b_gSDvGE^CMzs^U}=e<9bjJaAun9?byvUxw@a*|(A%zSgUaC)5)x_iykz
z9voDzbIm-qM6WZDxE(NCI8XXLsZFa1n?I8{O$bx{(>>#}!ssaN`c9#;qhry^G%tO8
zId#HCYNO0~vd?U>d436|#!Dk|8N^L0qCdM>rPk_-8wHQVPTzsZ{a~By?0g8^vP13!
z(p~K!y6na+hCc~vU#-ds-X0ms31_{&_;QpACF=6{@B-I8ltr)yeJNiy%<UI$?gOm3
zJ@6RB?#bVI(~vM#p;+%f&Z%jroq2wa+8B8r^Q$Q;3P~>OBAVnhQ&noT^<8lvJ8`eq
zZLmReL`ZS;JmS8%6UwZJB8u;V@a<IEFxdbfgQ8s5qNA7!Kh1V!P6)uc(L&L7RN`s(
z9^~nh^tpem<<Hlm?N;}BKVm9%Dis;gA6c(7uAtf@bm5q_uw-MB={BVdwsr-Xa(tXE
zl}x<dPWvI4(m`6W_czD)S)FjxTnpdSG(X8jYgPDW>D6}8?UijSSdDW+R~-aYme}~T
zlHb|B3^cL7I~QmSXWA0&!@+~EL|=1B@COD@xoO5~L4!j@-rHOYo}L5@|9XxFH4JXk
z=G_R8PJ(=r7Yprft9n-ln=7Y3&1R`~T$SkOXY5=H%T666zBe<L%}2o7;7_!XF<E8+
zt%@O&Nw|uMeK~sN=Jc5S%#<W_X@!GOh#GqiY)7k=R5kQJOnwvD3#Y!)g=?fO0pn<r
zfQ;zTn=*KcKy5f*iPnD5PdNgeIjM0o`i&SO13Ubw%yWQzD|#D<1l*4T+P>VI#6?<>
zBoT1lg@^>f^%|gvqAOZY7RX}^1>u6fBq1DDE1*JN;*#|~7t7VSfI8p*yzY;svhkxv
ztO5{*6#0M$;cYS#g~w*>$G?U(4|o-JdlUIn043m0J6aXH=FkzGQWwwU8S5XI01X3M
zURdWunSQHHEqo@Im2BCbYo|^1Onk_ryE$04K_A4H!EKeKueYHKcWR^yD#~J6r*<l_
z9PZ;l3J(vr`12Emc=0Ia{pVz*18Eyuc_ep&S&1u_f7GJW(mDg=QhWPZn!TDIyzz%f
zk<}foe~N9st#SZZqEA0>ObOc3LKrs|oP`uyg<{6tc`dnIkW3q_ZOW;>tm!o<I?oe|
zCvdhZUrq14(<ExghxalRug3`;vR$@q_7k!{s`vHWeQ$g^bo!6z(E;ij=r+D#csAoF
zz(VS=&-<>fqJ+n5t#LKL`QLw*YkEy~83JA@N~~$T9}1AGz@Ts-h?UcNJh*<9ZO=IE
zCly^`^fv0ur;O#Opo!dG$Z*pi49~H*86Wu%tJX}7bEuJTZKa5uzP#%1=g&c(&>Gq|
z;~bHkrjPH|H=<vV9%zW$1l;ZM<<t8GsFgp8JSvVHY`X}Q*r6_46<b=Dd2FY6uQ2Ip
zn!{RS$)KlyNLLW_CCYV3#++hJaU9wlj6?X{Sf)Wep;ZPN*5sZ=zjjV_iMjguJ%HUW
zU^DC+_Ov#>-NuUBfurEpYS_<<vYyk6tO*tIS42U?Q);MCv6=d8O)*xOg=<2Na&W`p
zd~y2F&v%c<@70Rz{8Q%4rn`0ItO+DL-`1M`(|(-Jqe@)%qeFbTUG8WfW)g2|PS4!C
z7+>}^e`ZGM)s2QFA9%g!p{floHi@hGo{wZbRL1&631X12=h^LiY}e0;%n;od7s?hY
zaVWR5(&N(ryKk5hsohDb834gHHA$wuP6e9(q=f;$34Jh;sp>h9voDQ{%l(EAAuyEx
znLHD&V%_+Lxg73D+D8>!n1SS*3my8!I_st*zzgT6e%jDm35~*JO&xM3v{|UbJ|d#0
zQ%EoJXJ0pU`SR5XG*Q`pt&P!5KaDb@5#hzDv>dK};!~v{12SY=CkGySv+2GB+a2Ow
zRj~k0_ylgi2No%f1*-*s;fve>12+J2HV=@i$7@W7tQSKY*MQu4x4?gklR$PYyBa|T
z;$7-e_hUTZbABXM#4<4Fx8|o|nE3%|G{UP&nK$K<MOM_bu&^l8Qz(7x3LYZiH6KpS
zv8Ky}p9uQp`*hFO;V(J=1`Wm4|6iQNv=?Efg&CaaCHJguo(&UtgMG#P5OsT{!{>pc
zOAMMUf2e59*6rmF-0%DXrXsPf6VoEanLC5Yy#s?601iB>02p9!Qca28uZOc*8!!lB
zMTtDSA7xhCbTjY1iWTLjkjC%EmY3ZZ36ob-O$%wWZg014D2RVWo)=PRB#jAJf!-Ts
zFos^igl<&Fm)%HSl0IDwqkBG{EnuzuNB4O#9D{71<CM6iP@Oi@Ei>y=%&(lfAOCVR
z^arG6)oa*|Yux4gNxQ?pnFk_2?BzwLBfll)ACq?$stR{EJm0jA+*>GW{iU~EC<^ni
zPnbNcG&DJ!ER5i>F7mmMba4sc{h@K5>PQdl0lnymYU}X1H#z?j=DlcD7~8PU*Rh*B
z#gW%LdEKJeU;t&h-%k7B+slvk`y1~nsjDVZpm_LhZn*N|w?m4aWZ<Q7#6*an^AcRH
z5Mw$T7`yo!LBz-9!C@T1q+gRS8`yRRjjMDzN_#Lqi%gq+y&_V{*<QlbIiX$!7a)Av
zaNFkaFsaKOy!>5cm7dyfiVD9}K9;)Z=nc$$OAOBzob)>24p}xVqj^WEjm7!(zvl%U
zQ)bQWC8>KwQi{jVcw2QF_d^3Oow-Fk+Dh)Is408vY=eS__JhLrnuC^Z=pN_i#pfd$
z{e^~E^+vqQ=kB+8p0?4PMMk~2+-bef7tGqC&C6c1Qx`<u#0%9zGFZJBSnwJ0h<5aI
zaV$#$AY8Au88@nwnf(4W=1q5=)({Au<&ieuVXj`166b8OXdplq<6^W-t}v42hNmh4
zTH2#V3N3IFA<v+=Q9P|Y(`J}Rc-Y<RWGHI&lg=#=x0Mty6irG_#Mal1OJdfMv0UVb
z$ca4P8dtYrLkO%#ags5!2Q>S7fK;2fN`V~qVz$OH;4J-FP4J9Nt6PUF+<o@S`l|J1
ztC4|wTbIaFa!gcuBfO1;k`QpskKzi<w$B+t$GH>E3$>Beg&@wf-!jYm^$mQk)G&)b
zjp4VF^DsLf%tQhp_H3|4{Cp!Y28ivtMz6<fAYW4hxw0#;pX6Z~;sptw!r%uEzsR|O
zr%n=zfO#BLp<r`oCwv+Vq8(W@+WwSS!qaWRW6=YU(By$0jP{8WV~k9YfChgN7~Oxd
zUak7xCE-(&pb&M3BbnYGgb_JEewq$uWG6g4dUI&J(&K~?Q2^qBvqv+VoUb+SGr<`%
zmrh5>kv$<(&e$uG)p#k|CX5J_kRTp6NQ;O-@51`eJ^D{U*{eK%pPMuaj_T`W5mc$B
z$J!vB1#r(D)i#sLJ+`#3mEFW2_{=3iOZyw@l4OBkndA25cT+`nT)DP*dQP?aWRo6{
zP=-fYf<e=&f;qdG^leMB3jnAKUTOL+_tSntdap-@*ls7VXAxb#38eq|W~!xII=;kx
zx%<=RC%>?D?j%yUR{fSinqQAkdcc%~VXwAt)QsxmW{UZJas=hCS@?E=t7HGyJK@YL
z+rx=9k|=&au4rGLwNxW`w-XuOd0Gv>!?gGB@6T8xYabO-E=L9STa4KXO%c&HED96H
zfei@fCyNX024&i6@;ezC3R!-c?x*t>g{J2V-^2!aT?H>9vR1stnh(eiKFudyG}>i}
z-7No->LY7B5Y0M27kp&q$bvSTOdt}dDmplDDbQ;fub=fjkqVWk>iagsd<xm^`B$)`
zg%^f?;`&ArTKl&9K5=)GP<lT`vZw7#>X}=UH2sx#6jIy~T3gw})Lk#5ipp&h+hIRF
z8~9<>NtqbYnnXP6ogs)+7Bh~R$ITF9yZt^`xoeMAir{M;(oqRMsO>!+xXod1!C@nD
z2G$BV0Xz(y20t|+z6kx~!2A@#ahK(m^|XoQ;B9k|x_QmVNDLdwxNH}aLO4}t01Y>{
zkJW4vW5Ig~N}_`ZL}6EVT@vbxdCc<;oLUzU)Tz)3Gnj@Ix5SRJKN7ek<W-m8WXnZ`
zp_vVt1QxIFLAA6J{nFtq{3BHWQx$c?II~VJuSa$pD(J&}L+-<VY2{ZqOz>I@0Oezl
zFM2fP$-G1={FY_=W(9GPI2i=gc%p!(lj*4A;n)**@B#~{9pnU#n>Hlc^@;2#;$1Mt
zZSf!)aD96ZDN#FysYk0&=rtCES$F8}V#mr*cX$Ytse=vb^Ss>6slEa)WuSWij=CR!
z2=c*Q$3R4M$1o7d>MAFgyyYSbb&X7J<K`pYuKKDu<IY6-Psah@?IM^J=(IwsmbRQI
zw)~cl$fHWy^l!EpX~k_0@iaJr*J)dhSisd9L+AtIa4L6c5GI{Fes^mu{os#uRN#9r
zdFXos3%`RgxD%WZ1(UfZ^Q^Afxp#E7ePmZ%OtDfukuB+%cYww>98OVuOL4x6w#@AM
z#C}iqE}GSd-U~Rxbl?`?-bBc|xfe9pE_?0}ifqOJlpT}HU{e8O!6esK553n>5i8c_
zS<qcNW=7Fw=3<fKO4@ehM*V1RX|J;Hs{gi!ykmmM#ZL~YQF<YI;~SyGJzMcda?O^y
z)KQTv`U>>Tm*+8Ar|nI9lafnB9%3#L1{Cb8-`pGh5t;YWl-!|Ze}xp1kvm?5`waz*
zA^h&5nZEpKtBDg7ZDx4g4?9`i(ElFW@YUUpmw#g5%vK%v6AIZcyJ%ewb50NImLf?<
zQW*4%D{*)@wbqV}rrC{}f-0S;HFoa-7YFy-6M>U=r?f$r8OS~zP%Sqg#wn%NJLwsr
zp*re2NS`u1s1GYk8rA>Y%v0|=ltdXxend+DvxSzIW&$3Wd^ca<xPQxz72iUh3T-xa
zqr<26j>M1FuaDk1fXZq$6=(d)bf_dnH|M9mXE9{9L+?%MU|%0;eCO(|<KB{!51^TC
zYf@R?{yBkd4F$@D#V4LeqL%nJd%m7s>1PlDFa8ZFhNfh(++lIWr^#p~{8HXWQB#%U
z!as}^;ehcRT%aQjX9cBV`r&c=`5Oa*z;gbq|Jn(Ze<ppcK*V~!#C;3#oYmu-)=cY(
z?=Zg?ziHOrM2@sZlNrr}Mu{quU~3J-_(aZvES?KkAa}t(7%!t0*XYXt^1G#Lg7?-2
zngOJL<czpSt>;sH=0Bv5Tvit9`#7-koG$aSKxU`Bs!bKax2LhU{Z!ErBgy|Qfre}s
z>gXBh=Dw&=Jm+yRzGE<~!hTEiJthA-I}8b9e<lDrDB^Wk9sDP+y0}mwg8?7Zq=+M=
z^ZirwBuJm_KN+|n4#-Ef-1A@K?h}jmM~$mLHv3~e>s_PHt^lv2mFU+cJ6DA7b2MPW
zxv-*Rzp5s!SQEWhyO{mg<ZnH+{!cXxXs}JgkuDLC$J=AC@mxvbY`5d3W*OH5duXF0
zsWjV-5Ma{yuJ-3S00Vcp73YG3LDB$p{50nGoL3DZ_P@A{M40=EXstF~*cjXIMAubi
zx6-C%ZCL}nOb3Bh)s9;-Hpc#QiW*v7@PGs!4R`vX3#gwd5Z8Kym?LSHH1=pq<cVDu
ziaqrS+pU%XO)+S=wOqS?U2?>!Sff&(*qz{TMLoZ%NpskeYt}FjBXjm110Q4xf5>E`
zj>E{VIAaE?w0!=<IY?i<K`||O12{e|q`DZeANN(gMh@4Ow`Twsr5Swd&01$5ACo18
zjWCi`m0C4tZ~21jMKjPG#blDh7IdEk$g>}F2$b&7zH!FfV61qI8C-q%FVAQz!<PpS
z9h%H>y(GZVGvjcTB;VHj;$k}A=HK<oe*s=hD0y0jH))c(O{u9Y)<{NL7q+7^n!Zva
z+jYX*N};sS#_v?MOtzlPr=o0x>sQDkiN8pAT`za#_@A5)R_!9nX@v18^;i1je;L?D
z7(N`<LOX}BgqscmqH+AVmxx^@u^<`e&kwa+F>NI;Z$3DWaQiaXH8Vt$`2=I85HoHE
zAPS;<`uHY)rP+yI<N|FnZp-o7T0|<dc$3@KNih}Imhb73$@AZTk9Bix2&cyee#M>Z
z3YP>tsobBt1z2%s-0xZoq|h97l(5&Y7}f#nn06z}S9doqPdn!l=Hj|ZAZ23rIFJLJ
z1?${xEf>#LvxN^QC;2q8sC!8|3>JVX@GQ<x2EqEDUTe}sUi3EmE*Bk*O%X;SkL6}g
z<I3PayPsfU-ipDAt5iWg(NcB9l&w#}1363V1|$`=i(h8pT?G}re2*6P^l*mUZ{cMD
z)lli|ZYSm)L7~w$*WjKfgK52md8^tAQWzq={_pZ<3!&Oxa_dD`K}1d+QW-S^UwqRK
zr%#Pa{H<ZIn^I#IVpLV7`OSxfEdY~{pSyTMWha)y5v>GF0*tV$@!{RaSC@pJid2!;
zwkl^gz5Il?GUojXWxX$UaH^xP;P6*95kRK^@F6cu^^AKge4-`HT`(K2$3Ia9&yffF
zss}DCVILW7Ykzv&NnA-OwpWoR(H~P39VC2di2fVgWYqmPOt4;BmB|>y+Ra$i?tSMt
zK5S8#LAwt|Na!wn>Hlb2Qv_dR2BfZxj-5vTnb-Ok!OE-IPKBRuH;9X*3@~8ShB}g(
zcCEm;7;671F{^vs@SHmyP~&5rF<@}TMx~RB34>J8@w0EQ$pPx)b2V{$%PVX-G7{5c
zKMr8|8o<Ng5YWIe=kw-SG!b+tx50vhI~#q`bPyjLdH^jz`E*~}MufJ3^nMQ<)ksD7
zjo~z1#;%aO)pVxPV2olOj`-_*MWlT*fV2G%%j!`Lyk=?Qc#2?*k!^J;%lbOqKwFy@
zkaSpSu+jM=!qy>$x*`iOigk=LJ&)>VKh*x{GYlc)F!`DwJdJ9dX)nVTEuCwmgq%!4
zLafSRzb5K=K0rs0FKYao8Lw#SM@8qu2hldaN9^{{cZPr6S53z|DEuN;Zb{~korTp$
z3`8DJ1C2rE0?$e|=v}RYjjO(;><a~hmAp&#fuCM-g$yhDymMNlh8L{OBhu!bC;#_`
zEkiCz4*A|EuMtZJg>YP3eVOlb&0UK3>XsXh>A{}gUlokqU*_@;@Utmcqkt;KQ0W33
zp}NQ`{f3hS)wa`!CZsnHT2jvxruBlu#U;7|{8*qEhwT($q@TMu9~7{ckk^)bBM)YR
zm)3YLM#NUWgg)t!{xiE_^Cx#dtOhab;J4$y)1vu<@h$l+StKRLH-2&hS1o+_V)5Z*
z{3jj3nszLqciAcm1KF8EM>dbqbz9`Ez)^Hsi43|wSrLL+H|_J8Z4X-#bC;yf_NKSL
z9Jg`)TJfq>4M2Zf6eW>j)Ij1x@It8^DZm=qD$GLKEh%N#4aA~uVI|3^Q`}r~s@*Lv
zJojklygCuN*wHvk^?h7j`E5(U@$+8<XN`N89lpz6+`x&KKD`cx9?;mrtY12j!AC$c
z264A>6$gu@Vmt&P9TE!FX_RZZ-zW7oo9S>1Kxq0MB)x>@wiGKxmys}#);*-+I&wzj
z*&Tu!bx&Zz=1@V%UjC)aZ5-E|t_d)0>YHD4;I>6W|D_5T?M+ZVt-h#D0KP%{DVB?7
z*ivp<*K1$vd;#$?7jT!X!4!!*N6KK;F=3M#xV55xhpb5tUsm8QfQ?|CUmQ0~WumCO
z=4HO1+I~idnwZqRB=nMt_>Lio;Qky5J<sz-p&DaF(l|iCwcn;>3$&81;)QLe5wV|k
z^5kZ<>uq@Z7cF>uXm3TTl3Q@hdR#{C+g36@i7IjeLrg9%CBlGEXvs-9YsXu#uOT*%
zkqJD+4M1ijq-vLny}QQ*yB8Qi85tQT&7<$U2NURxh<;Weea@t25JiNLD-58qwozV_
z<OE?S>Ck|1;GFXbKz3mJ`VXLVID7g5o5p2)0(Sx)@NNRa5j?D4PIc``6Erx`6a@MX
z|C=V1=#3783W>{`Ul>7ppvbqvIyhA#CEFVGP%^UKI@GsV5bWLry;GKjJh-1!@^JQ_
zv*b_$+tZ6wurART1uy@8?h2N^3rBnp`bN{~G>vya4nnJ>2KbKi#l2?0@#|@|<8hlU
z`-|o0SA;#<EjbrNS`Ru2?o8E)GecJO_AFo-YZL&(wjgN>e=gc4sckL*U`x3Is!zC_
zBp<=)mE{3$SQZx)z3r$lGAo0lTVzp9Yt@+U;se@wp2>g9-hf-cs6j=6??2)DU8+TZ
z1z9i9+k@0B^IDPN9Qy;d+QmvDNq?7BmFYp&B0y63(Kp9vaqzeH4NPAk;HE8N`8vC7
z#)f{U370-?wlD+X<W<@}&)-Rqcba!gtZ<<ImmB_AXY?Pjy+6$opOfRQmhz48MKYYb
z$7Eum?`NsULt^O+WF9-raSUF<z|8Xt>?`WlnIQCzolGXA?bOx0uHc8=o7L$Xul<TR
zomv1>cH|#SK3HIl8*so_QKTG)F?S>+z5s8M`sQvUI;OYV>*)$lt}3?0!+383l;LSf
zUvFoSs(O0tGQ0JcbL=;=dt;lR1c8C>zgyE48(u8C=l+cyai3!=@YPY84S@3ug2Xnc
zm7=$tir*GPcx=n0018e2VSoKa2KGtwM&Hi4+mh$yaqE99?z1L!E=Q?72Tqdc_erz;
za$y2VKr&01ym5>4`+FCs1mg_p?409e&RZ`77I;Tf^yH3x3=(nYC5Obz?X-bKwj8@a
zuj5zUY+casxs=_(Wmb9!@t3YN(Fo?d2le*vLWx42SwLiJ4>1s~mhr1maE^Knyq;kf
z)}}#6M2X)^qBxxFuMSekP#i+XQ)-Fd;zydE-xthfD5Uk;bB#YblBZi2k5%pO2+vPH
z7zpBM;*Nky=1aFg`VqB2+(DwbV81q+{uV2yB~?p^dS)vfw>7p=|EZ|vOBfK52^p9c
zsmN1I6tn(Rolu6({&t|zIp^A+dI_1&e0e%RQVDVRD-VPh|8KUlq^h9=<I|6c-=QXP
zDQ{=O?==(Ky>4cby~m{>U!)?5dj!Z&)~dJ<jJQn&fO?qi4Maf@?u#PGhFa{0RbtUx
zWCp^iLMoSZ+4E-nH(Ir?rtcK1@c<z5c#>Pt`g2^eOlJ(gQTLl_J@rITms|$F+OX|^
zJ_LwrhTj-(<eH48xx3ro*2;0u0aY@L63E;FHCRT`hkrcw(cr4k2a}C24=19p*-B2X
zwnFq9o8T0-q*3@?gl%)_w{VZyftYb|Ob4pj>V#}YJ$1VN@I=3W^S=?V>O(-e3M;4s
zxVYsj6>dF%^1Yk?5VR%nxvU#04q$Y!4|U!A4^?j&6;-(Q55v$P-Q6V((w$P$9m7Zn
zO83wujdV(bfOJZCmvl<Eba(x?=bYzV>;32&KDh4Lv#<O5W%O_Kl4gwpBwCyLvmHj9
zPBe+Ey0m`@*TrAzKequ9H(bBXl40kc0ggRtID`oW2m*?plE&U|7<>aH^*{&$(`mHp
zqqQ%;EgDUi654;kGY7blKt8vmgGI8Ig6)*fkR}RI&oF%94N4KEj{9BG$@aLKoE-no
z1&;89%Vj|FVs*aR2fzCzeAdY1^<QCv*wXsIPy(a#Pv^qvaOtDybAb|<$Yx+TVTB^S
z-5sOG!v9NuwlZJm3N<M)5>U-fyts@ItWE4iUG=%mKf-M;Z2ji?3wF8)2`XKA9=iCR
zVrUn$k=S(#;jL-3gQY0hEIN#v?w1CniuOPD!vb!5nfk3mV(s>V#aB4~TI1r`J#mCl
z1Bbi6d$A3@H(Si!E}a6UPC4~n>+3;nqZsL*7ru>)p?w!~Mn`ADVN@QZeJTEqwK0x3
zYPWBe^9dzX%YE0dBq;<DL=7G}pAKhCo<Lzt<j$E&{kR<Qq12;_>~==eV{|Yph3^!`
zY(VGYR+hJ?cD7x1*ZR*)kN;n(=FJE)Fb}7%n@qAM`Ib+y&!aL0hsRY9rtH&7u0P;&
zG+ipeDW2!I-Yrn#_+2O)iTaCg&D)v%xWn(*hfu!$eb>y>AM4<E7K#-1-KuWKuXDG<
zEZQQviB5?;mZx!k);~$#Bav@g#QLybA6g!}lOxq_4Q7ckz;Um_>+-j5-E?1fR_wsu
zD(szI-<;Lo<{6&Q6b9h2>5gU;!kYKVpbeL~y^n12K5dtU$T&zT3Pdg4L?TYktujs}
zZ(-(lcxFoU5n}wf*ei<M<+)i-=f)H^QZDsA)K;gAthu;etTtauYj|)c7D({%w!Glf
zv=Ir6bqUbsmEJq4)6Hz9WD!1exua@O?EvCU-d^~fwHMz0_G1ys<Zbd9!?~6lE~~#t
z-L1?e!3|f}Fa6wp@BDcniEyg^MRdP%?AS^vh4%AOj#%CEffAFR4K+i+Y{Zm6Hnuz!
zSfYH!w(7e=r~H$e$idJ2A2i9XZ~aS>d;t<s#ed_>@ddE`%=l-V@xw6RF~z9e0(y<w
z<qjwhiSZ#cU>FjZB+&`8Yvi$$G+f9Jgcd40qjo+53Yy3*^rY`x4JG}WtXo%m*?t6%
zNPM$oz8qoiJJ)~KRGLoD%<sui!ZIO5{sP1JN4OhxF-UYmvXbB>%vwGc(0i#I7OG~e
z6lJ=R{3n2S14abipHh!@3wxO#CSC~#Q$Vk#XdX84M?AoR8Yh|Si@To~S%6s}Oco!J
zaDmM7rp9vELF3?5bs?Z>PZ)52@7INkYr{feU*fYC-3;CVp~M&;7BUU$ZHl>EPJYdU
zzy3!2r&uL-{;ZD}wY^Z4pP~u~TN$C(z-QBpxLHY-nfQ+_PKmkj%&R{0tA#^@yCT;!
znsG(KF-r~hl~M6I-|Zi+_VZb5r>llwZGt9^aB7?6T+6`#PW>l0qL)US?k~`nZGo2a
zHdQfAafea_cPj~Sf9PijmoI4lcCNJ?kA7?pKBVL}M|M3~*7xEc`u%%0>lMiAr&)2&
zZq+yB?e8KYI}IVKQ@B><CEeP1Y4rQgl^P&Rg3(FWOZi`$oTXd+!bb<oe+ByC+EVgF
zC>DKC0hFqBZ%jiY_6k{){&6<^l7Ulq>{&2WgF7jtL5}Q=R6+dcq;T<HYDt5ZUbt=A
z8`RB!ltZp((JFHVULH7-pqUu))sb7NW6C{QSqipguG4G&L<F?JC$Row2M4j+6u{HO
zf-sk%W(2sEL~zuV9qkdc@PZ?FP2*x<pm4Rf0(LVKmNmt8=_F?_>oQ^O)P}M<<!~=M
z$mQ}k{+qnzu5I3(uYw|JQ;pT3yW<=`KK5)ru5OyAr?8<-1Od&9?-bY!p!+?Tk$3i2
z4MdD@yZH?`>&8;mJGbGk1x?jp0W?}Vl#+eRAmitB1iHVnst8w5*J4xM7wx$I`bVIh
zqHIK&w5mSI_lIFtB>V8k?lhVg_UzPi^Ma0SKk$dwTR-|y)Qn}p9>t|8W>v|6&`-aI
zH@v8n5O_)_jY2-2wHfv~=W%?z^!<jq@*l5h{AR$6*tdkKzHQz3*+TLibRpG>)ByHe
zTf?1qXa@eO^e7jm4vRBL1wT?<+A#@nQ=Q#W?KjW@ZLcF$Q6!9$wyzV?2wA}1A)^FK
zT70>c-g~Yf=W22kyNFMIf#jzrUVpfB^P}v$mcq`BqzT1PP$a%N7|fVdo^>Z8!mO8>
z3B|fsS`})nJ^J}A+_v4Sxacla>K_Zs?f=FjHd5`EM2cJ-rg))*vpHVl;6!!MzfVZ`
zvVaL>A6Drgz5bB`f3jY{mq9py_eZ9T$t2*cN`j!Js&p0-aHML_60+Fjz3tb4vHzXA
z;9^|th<di+2So9aG@qfsFae0y`@44=e%g=^TCKnWy>tyk|5UU<gCeL>z@E&Zr|Ojw
zrJX1+Ra#oTgka>+2I)`?d;?h$|Is&{rKrW;+i~12ty+_S{eZd{g}QJHo6_&BZ710<
zks(@O6WPK*02bmIo`#Obs{`9vg>YuagZNX?^xf`{r}z_K>DCQQi}I#gRIJAk^WDpI
z{|nfCO1NTLfxZeyehZ)?2iFz`FHJs<s6d@^_Y|52Qc_Yv>2yiDd+w(!*51jO1B}>A
z><&U%%&H>iUl&+8>x!gc1QPsvh<Z|ty3@boo!09tHCv1jtj>yCSi0T8*pxwdk`GqW
z&a>e@C&0whVb!&ly(}vjdM*i2ig`m5NnX(4+^dV<bND>o%u4+B^l8orLLi@?s95)S
z0TR$7`>Km2=}YcE0euMS_|;{8K%GOP2Viybp8!g%@wF*P)2ecrDbo0C4Mh5;^K8bm
zMEgD{LdKo}fN<-JnaGPe7aUN-bQ;YmG@dPPl0Mx{F7pFk<h2V@9-$LE$Q6CN@XLBW
zBwcf>a*?PqZc)?`&|-FSVQO6pY3}1DSShV(T4@&X)deeH*Jrqi*zW$`PQSOGJxD5a
zP5RItQy)YAfjSgun$fX}g+0vHBTVk?Ni=rjyHKh(GX8)GA4E+evM_%5h4(tn5^^CU
zf7Z+@nPAVf8gk(Jv-v2cd5S;kQawM)@70pq&V4xXNC<8j)_J||^X~2O2YzkAv0Xjg
z{%=c|0E{FG1DP>)w;(2XJ#N6_NtCni<!3dg#%9eMf{d6{(+rlC?9BWa!e~CKaYFp0
z!=<X>4Oyl1{aNUBQv72phGWAXPWW&Ak~i5tk61c^5kXRcg%Zar0f%}2+STF&5IEwo
zmsFYqQunFK6FB1JU^6z52dwN!?FpI3x$MR@cdLW>_$&wKLUy7I%x`lXx10d@Qu&XU
zz1c)dl9+-Al;fRW$_+P)B7x|W0|l-U-qg+Ub((EMSVaqaN<I{20c9w*w00Xdsz?U*
zk&V}%=lMb@Ovnw$IHWJNaGUi81cH6(*UkbpWxSXo&n~<{7d~|!&KG;1tl3<g|KXTe
zkx22Eh*(OI2lA_dk(EE9J|PLNrk9xstM><RFH~;^7cp1ev@{;D`aOND7U|4bWOT!w
z&KVVKV)QtejZwtpYWcixyp?yQg9rJ^)aKQiRAx#6BTZk8hK#XP@^b0^)=M6-`m={A
z%$a8uY5}2t>+{&j!nE#3Hw=ib-Dz!J%{}qDaof2is&^aBTehM-;%)2&;(Tt@h$qQC
zGy}vn?rFzGY4^y_I!V45;(Bf1m!E0;ke}K-zr?apAxJPLh3_!7j;c}3t%K?30nRU-
zFRs@VA?F9G5z*&Lzt@+46A;$M2MBv|%Np=>xSrL$jGmy(kFc<?m;`WdiUQR{kh0-h
ziMI#60(uR0ndJwtYRP)yu)8e>C8cj{0f~e*{2hqN1>loNl6e(n{D6cyo0pH>T)QrT
z9!CJbOxQ&qr>{+&q)nYlI}#Fu3GldQAHQ{!>Q+`bQXCEeVK1_0whPhtFdW??N@YVp
z=gj;9RZo4{Q*b3cF~40#e-2B5{v<4Nur>XVKP((FRjvyG=hP;6egsIH8b$klF<|Ed
zo)sh#rSJL_CKG`w@MccIKW7uW%GHcrSD_gu8b*xQltA~Fc$puNlhXIdFD#r+DB%@c
zOMh)I4Ze=^FS3Fsf&yCviLMuCc1lu`>|&?4_o79L?EW?CzL*q+FsW!pU4cF0SQj@d
z?0#4KD<C!w_iqL)mkNbba-o*8BBTuc(c~i=6M6h~>*VOpO^RBV=ubp35S~RKq^soJ
zM~e#BzXtIDC9Ho`ny6bT-9i&^@^A$tKFRHCs)_;uZa>xYZD3D7zFBv-$*E#t%yhvi
zUX5Y+@Li_lR;={8xF4^x<)8iBib36s%uJ8BD4$xVzi|Pg`!B}C-xd>ot~|Fq$WwDM
z@$r>`cEUCqbT8$gbY-twEQLj4L+(a^;JZpi&-g)?OL{dn1aUjd7sKNMEq*T9xHzv^
z6Am~QZb$ehe+UJzCrCFng~Mw9h^+ho+sY1S-)!{O@EB%c{jl;|0HUr@+z?e(^EFDX
zYPK?opvD~o-iE9Lxsgj8SEC?Ji`PG~iN7r78=_cAav`E|=G$f5^$*_<EUWrTz5sDf
zqHk8mivsD|_@1wI!zes@RT(q2`eCY$U51$p-nrQzwB6-p&%c5Un7pzpcb{5(yux7S
zjH_}YD12rH;_8MyURET3r>9Kt&TCzGDe&Uz&ivJxQt`YQ{ShP1zKzVck8sfXe4}X4
z_NS@ZsS_6LdAlOT5X?o9AMWx*;RU@nR3@mz+~v0RruJL<sS4Czo0;ZFdEW1MBRcQl
z!0f}Ug4UOcH4!Vjp!>j1q+GZ~4-?z9y;NZw(q9PN5eX9H%r8R@$S4v#lmj(7*VH4S
z&y8ZW)pJ|OAqX}cp^G1R?uA`fd@Da(n7NhzUlzc&D^D%ZCx<`!<!D)tw^TJL`geFy
zrjA^#Wy`6+Tgu?;qO9+d**?i5+c<M4k(lCdG#+n~uAXm;JH|G|{(D2L+6$OzLs+o3
zFK6Bv8p_Rcd~<OnbZF`gGfpWYbpu!1k*M67CI#yy46j7l6dI6b5?+5fMr?yTCgXSc
z-so*If+3t#C&GwgpNoQi-!T)#sxF*^;=Ybc8XG2Vt+vlIeXvX9UsE|r;=q^z6Qx`Y
z-w`$WpHo8VKNwSVjSx`mbl?oX9-*l3+~9~wIIStZG6t2<cUI1Z>}2#)JGmziE3%mQ
zg+Gm?=<{mRn)olKD={qpyY|51I;+f|Y)lJ16H?J>TiLLb{c_952fZ0E%tj@$nLYUz
zvBFV(+{eBG-TKx_D@avXoCsNx)^mUX(ZT-3e`7a<10zQdq__B15k(X8FnE8t=ji-h
z5<3X<NCXIRabV`gL~o6%#i}0uAi#rxcB)qBRh$)}O*|lh!JF<@tU@jIpC@9`L3rbX
zgGI4UJM;1Vr_of&1V&XI0^5n40QwM<a34asV;Bi-Ds0a26+mq`Tu(ypUwjJS6$G^q
zBT+ylV`?D8Ov?ME7cGxbmKWs`PLI(nQ@1+=W;NLUdUMBp1n6EqoQ=*R<x(p>5oAG!
z5iX8M;xk&c!a&<om@|PH(u&HHv+Zn)xmUn}xpaIjR8$l1kEH`3cDQ1oVD2eSsOs&S
z^SEeSN#+2gSL6=j0k=7(>>V4U_c?F@wp;kKl2v@aqL}$$?9U&;XoOD4&BW@SHdlXV
ziZv6(R`8%FqgkTDyPtrPD6gwdILxpe_SQ2g#FXub&uS&R>Cjyg7(ygW3LE5?dhYos
zGQoIpH`!@&pj<I&O#JmJb<e%-5a?}+w4RQfier{<e0ezRyEpHPz<+Xduo-3R4q=@Y
zmSpX(s#d|N+Sz=bV6VdOJO+3U3GUxVj`_XrlIA+!&aV&Wmk<Mc*nL~djEX;9^zI@(
zVYu}DL!h^zIS&or>W-&ACc`%C7~t#OirSKU3nl%A#s?xE#_HHC4sE?2nW1_&7_bkM
z-d#St??4+$Gqz==xO=|EqW;<tx9s>n88Mb~Sd(ySG>@qHP3y1Qx4JqZ0&iN*1=cJn
zF?nF*?2{|n+FlxVZdcR$4u$&tqkzrkO?{+o=EEAvGVMRKln#>>Og5uORZv%j0gHtd
zzx?_0Bu!8U_0S<$Wgh=?3|#Ky(>-kS76#?@sfo$}ndWd=i7+(tlQ6CZ-lcoG>5H++
zsCCj3Z`H{VW6C*05GuIi+PSWE2Gn(zfV1LG>Hh;^S<Hitr>1g91l&~feeruH?Iw`q
zfoW0%`lPB->Dyg|PU`7lOq;;ZX2}z$%FeU9Yu#xG<DJ6FYzm6>to!sq^x<a7vz4nT
ztAG2kk5IN%uqAAs2<?tkSIoX@e9JX~Y!x|6_V;&-vs4~mM21U)x$9=I&+Pa-l@ClU
za(h^l)fC;{Np`a*I;POzNIIbt@s*7&7|?cM)wbzg_~;w!8gJf=Tz9f>z*@UMv2UR6
z38&3Av+Wdj);nsYOt<mZK2k6lAeY?*Muc7$bjBR^x+K*+Ee;f$KHv$aNftJ}3L2_J
zPV8p#eC5VWf~AK@vhB)RR;i~d2=bs1#{3<p{<CQY9Rbk)GE_Hh<sQ=jzeLE;g6|Nd
z<UtEDGI%6pI6rNm`}0Ewjy2bXR{MQ~nDx{+&&_8=AjpnXn?nzJ9q(}$m)P_Q>-lTP
zSyhK`CtF-2&F|5or`UV9E>%oI$2qLyI6p+=C$!Xx|MbXN!CZpCdESnmNEc8S_o?V1
zv_}M^C!T|BwF|Y~%s$>A<VfRu&2HO_U||j7dh5Zv2lRxh2TsD^wIA0n{V2gH)Vyb`
z>DD#z;bqWS-^=n+8OoRk%+;DVD-J6#W%3%srSiaT@1YDqg`xc?Bc=D^(3@aT$zamf
zl}0-INhjFKj6j818u)~xM%wv-r7R)tItL76N(a$17vm%`hyw~o4Mq!5IXFJOyhnk_
zLiOhZ0E<<}io3>APE`Ld4R$MuInMrp<u{t@vQsl}-8TCo2@_D|&1VE)2*!n2zN1iD
zbRsYQP{7UKbSwf&8eiZcHFST+>ur~Q&`GZVsZTP`57#`SLA^Nn0@Z~Dnd6)Q4hw*e
z$re3pzK?976o)i7aHMcBFiHhvE)$aPT>h$6qd;|GCXRr{K&)x);R2<UFY`JGIs>({
z5%LlWiH85OJt<ySBf>L1wCZ=qvnly`+!}Gh)#vvv!^au!$Vj=YgJ4pdAzh%64?aNH
zMX_qO7#TO`Ll4I8TOI~${4G)c@YnHG4}U@RI>)CE9lq~R+h{=`kg0UFzq}?Z9;_rh
zUO{-6R$QKjHeOsFm7;nCYFOr5)WB}b0@@D;<PL1^3ODKQTn;HmM|SZaY7|yIvRde-
z_Zn(zYg=ADkOt6A$l1LcfhyXxq`=G?h1n=xpb0c^9_;mjI~#Zh)gq0>l+?eAyp*e3
zpa|pCuc7&PWwYAa#1xiP0&=?S%i}Syw_B_!Sc_~%or8@*W3PyJ63AL7%j8E&R1ETE
zQrr7|xRpjMV>B_~+$t7*&Xrk7V|b(lwcTX<?l2EGuD~nh8Wm$|O7>dt0+eZNjm3+F
z!(hi!J9zAxGpvfz;^sTVQZp03k^Ql^gkXj`N+h*g#_Vv?Zn%DN6(0LKe;O}`GNEC=
z8`!XZaa2r|b0Gtl(gvmACtS+B1=;WM*wLV>$wIFFiqvDJQvULAruA~8w{J_x7OIgl
z%CKxc(fKJDy6m{GX#&3r?dfv6k2kS;=Uj*oA5XJIF|6BcCF-@vU6;p}?R0nMUC*SR
zmEetmrwAVggEKfj(8pyJH&-EW19`${$vE4JIaou7knmPms<L?PR%QpISdAd^C(dp)
zj~~o_I|3$^S0P?e%id>|qD}%=%gCb{GWP-jt8vKlgUpIPkxmJJ@L@Jktj<2CJ4>30
zrX6{PebE^pILcl7azOAF@ym#U0#}tgFCQO4Nz1DlA7kpQtMF2JFhM}Eh4Oir|0;5<
zKmXu+y6QD5hWhft&&*72${rScRX$zxm#$TzP5Udt5*R>BOWPYKidQ7|Z<HANl|0s6
zWGQ!ewKGZa!zgXLwvai(`1h)aoJwo{0*gJ{S76M}f=Ovu+2;FKvt)-wCA;~1XUN)9
zd1If^18$7O+&c#7<s|CvF9V~;PZjg)e=uY07cVWP$G+o7cyp$89+<wm1vR3BUrxv=
z1U+6|j1vhIJqOGYzbhpsUJxMkryXr{_#EDDx_R7p9AgtRldBGmg3h9>*Yvdy8gHg(
zXtl#%Bk^9<+^gza4TVq)<CFTJIsSh<48RND2!1;2a0ErIVg^Ox&TIWb?5(~q?K>~q
zFn#pFZZFvh)+wL3NG|F*E+gf$`MZBSZ4Wl|#-_=q7s#S#trH3DW)ZaZV;V~n_2O&E
ztw6L&(;3c%SO7T`HmS)JMFmq<8w0qdP&@i4j!kvFVO%#e;|V8L16ApfCXGB+ygW8=
ztKKzS;=?t-g_#YW7pB{3&`(H>Bm;W+4b(4H*}1{0%YXETfE$xCuuw)!UKDafTQ1%Q
zVTwUm`Yh#9(?u=^Y%|p-A~wwm&cQ<m)x!2*QhpoAJa23I<z#N3d2#u^I`S;=qS7zz
zI>}Rma=RXAssE(?wIbBtWBb#?)XQDW^P%@4nMGHnM$t-^sDXGdGaCVS!}0xLiqjUB
zfdYE)mN6w@pr(nKn=u&cm!)HJki%v)fYo(+xl$?+fkPQM$7Ey1B;|{;J?%635j=2`
z>Anh!S9kuv3pP{xi*I)eq`)e&3w3&C3OdW-nQi{ksAwZ1ti%Pn=Wud@jq9$Mc7$9H
z=%TGPK4$rfd=#348>eo)*%X1Pgky?cYx8QbB+&J{yb$tvs7sw+ZgPelFv2^}g_+gm
zq9o8NQXy9;eG%x#cVSxAHb_+G^lj|eg02EJf#*Ujmc@AEdq93hq)35G;DTKSBa?9@
zbY69v)BDb07UWsR0(N$c9$50oqcZ$uhK~B(zU8<+P@&S-1St5fwV9BbLwzf^`pY_n
z=pRAe_?^$7T{%l6&FeXx^CM`Th>E{b4nttCyQ#GjEdct)Q5U{7nZfCXv2O*%VJtAu
zsNkO9WX9hIraa}64ApDrLGZ*g_WRps5sTwLzc!^PSBVbvcY%4)D|iW5=|qC){7R4a
z1nf3anXW_UeVk<ckq2=dEPe#}Bs}zS^v4%a+a4^bdG`@Q+6oH1cVC0^s{M}&PJ0k;
zT5pex`K2D#+w*d6r?w{@a{b%|UOSCrc`X^tJH!duqbc$P+L`eWAc#a44fY#yk*9wm
zN#$T)a2FcJ``lnQ&_WjS-J!R})}wlC&IKdlwVVWOyW%#A#C#w^c4SVAh^N|sWVUZM
zaX*D}r4~qVAYF~W@6b5pJKj56`W&Tpo{Y%*Z|q8D%XCtLptf!M48mTkCy?>I*}b!6
z=ZmexhYZvK@4Za;8<l9|i6KW6ms7is<4^Iqj%Ge!q(aq(+C7!0dy<Ba8U6@pIO`##
zSBp9-NkhUQ<eZ6C>v`5sVX_@)0Sqd2JqfCxc+3<Z%JV5OehrxedHdx@Z4F6n>Dm?8
z37irQb{?D_P7W1j!@M~}!0EwDb#{XF{`zW--d~vwdVBoP72cva0u9lc++65h2)J9C
zVX=E~Sg88k(pK?(DM?)3bHFqCboGf~0Ph4NK8q6R98<hd=yiEqEMNo|9Q)0l)dZW>
zwGzqsLw%e3aWjNf8_}pp#s6|a9ZNRXq#VJ`VtY64><&iW+mndj=xt~FIZNZaP2xrG
z%EBHtZ~Vp_dgFn$DV^sCpGO`d&FP9yd-wFtf8ZI+!AJW(*A2yTw({o5CLd<4gNkhg
zIrVJw=+WxF2V@)>wj(|MNt!ZT=Njo{GLW+-%Br}lU+c<|YIOHFfFCp-tIW1+mla=w
zFXY}q^$?Ds3`Qe2#J}jFEPOkg`*(n}^~vWP%gK5Dsf}Nu{``xWO)<2Z!w=rfbV{4U
zuVX#ge}5$vp%QoCPM`CHfMKlPm-=P43{DCoeZ5-j_-T)GyW$nk$mUQgi-+6yGn%e{
zu?#a9ufp#I{AuGmN|0z~fg_K6`eSYpU@*j_5YclY^<3^IH97cVi;W?Rs}cAIXtXND
z(Habpw$1=dMVAqQfwU&9u6YtLhg-}wj+REPP)CWw8YW1OINq@-3Eo`}y<8Jpzxunc
zh6+mh=B|5`WvY1P>yhrNq21Wah$Q9F_%6)M%oCL+;_CidZJw^`=FRqNV(L~GKtxHz
zW?A6*%GxA1WBsJABp<4`+RvU^)JVgud{<Sxo}6Db@$^a&N9tJ7&#2`hc~n7`<b;R8
z-TO=S9V^D#i>k}J8H_lx=SXwc$%b~=f;m_Oj}!)D34eSk`j`P)Zp(n7?<QvgU0nTd
z_#Sygwi;f*fCZ3k8n1I-C9Z&CQU!P#>=JD$5%bO*05UrM9MB_`XtSt)4Y42-_sy{F
z@yOhUs&m5CMX@lNO9t>JN80vX!jwOo3ZD~gQX07KqC{4cEr*cER=((=u<NP%wQo6g
zI`>U)2t*twJ43T_gG~joRhH^)WjD}K*UvE##09bd@1pqT8W6x|5&X6Bnm<B9dc&c$
zr#fSvLc}Bv%^<P2e!VgLWU4^P)KHD;m&1lsb++P)T_Url+|(2e>cV8V+fTBHT}Khk
zEeB<(kwYeia<zbCt<J2}b&pz~N4I*{g;1Lo&KDJ2rgft_tGMr`y#lEXdz&a^{6ESv
zO2+CWWJKDbJ_$JB=KKv;{H^t-C_g5>zC*T%8m?!+bTUqU$%%=Vmk&`tHYGS>6kumY
z30=AFF$*PdEg&-p@4r+kIaTPpy}oO_l%RHaeR*CHrx%|rVT$;Az4>Do)vB6^@EUs2
zEf$k3R!yl@VMKiPCv1zq9_g`l+50cx7e+F=%I)9yMFebiq1{k2!ZVU!g6?viQUStF
zYeKNeT$w&?L~y}wn7~@!eb!zE2Q#w*S>?=VkovpmH&7j~xDlWoh?ZUFr+#gr07Vv%
z-)T#ES0rFB8QGuZOeoU98C@hj(hem@Pm-o)qA*hpo$m&OMI+OO7yJgt{noJd-ra`?
zs2(fxXd>f|*vJuQfB=F4^-`2)_q@w2dLEM(17SsR^B$<jh?o~5+j0O$KBGP2MLRb_
zU=A4@?qqDmbxY93&;g|fy+KUBb%_@>$qjY}EpRO+1`K12H_YSpTdjB=8z2PL2G>IX
zxmWcYUxQ``?WH94^m*&~{hJbJDBI69xKodA=Mj)FRsTDc82@T3M~TRmLXCnht>U-q
z(YkjclgBlSG*_d<34NA^b{!Jpn*AM!=eOeP?tCfz;^SGN70Jr>?d{Zf9;aqQoELAR
zB<3ru*uq~kd^=%yS}4sQ6$ZkD-Bq2j*faL(`7HjBRsCv>BvpZZpyM=^$tq(?TDhsQ
zeD7%Uk;2O!Z)ePoaG|!>i0^j0BhR9N&8Fcily5Ck#kyDBH>}v>BLk;>R*-1RsM&b?
zQ|VxdwvHQz^$kb0GN8+Gt+srVvUrLA(!`M<@+Bw&kE<%U{`^P`+~J)0{FphB(*@K|
z8m0G~b))gs|D*?}k1`R`r~#UeYC#)D=EChLnA`Or0yoUW&edNNM(_6>Y*&xWh28%w
zwtOi_)$PuPcK5D|8ZPqqn3hHp_0fPf7$rcfpaCj95SW!F!S-WvPZ`yqpb0{2DxrbN
z;CC)&CNHC2(K9Bx4gvLw;pM7j^z5D;u2IDg{tpg3xRga>zLaA;lfM#K+ATQN;T*GY
zFOFhYFC`sLULy$$VmqkInUGV`8V%)UqEM7)t+`kR?ImuW8ams%mK4o~>lzl}dc9D0
z*skH&j=-imJTuLS^MwA#yWQP&5L9S2IE67faG7_cNxN=k9$SM!c`M|R#9t?ugPUQD
z7Rl#B3DH_7D-WK26`3;rTbBN$2&`fd7*9H{ad0}<VLXK`UUd4>_Y&TL$)StWkKCZM
zJLM<~1A(4sw^Ulm*D6>?Db!W6&?=LE21C}3Y494)JJ$+?;L0Q;;RsTVDyna1vXM9}
ze?d)jBz3_<xRBvQI1K2f#``#g!nk~{&KaaL`eo+l#*N044D~r6J+CAhCx=Cl(b$U@
zCJU}%GSYUk3cwAvU;>}xT{TYU<sb)>Ss`G{_GV!OyL@4t-5{8252Kzm$ogI$PYG1q
ztFs>KbBwPo=o|&@jbEBHH%;e22xV&wvQti;X1y=^Ie0<fQA&hern4)&u3~0GW{ikz
z&Q~?m^vOKdh>%$Uyq?@J<N;E_&~D+Ql*Ycc8B@$(n=Vb<-Y8or&}zFcWelavXu%Hl
zt>+=C1MCYRa2dZrwd~>K7q)jLpG$=?3>uz*`B4!uvHYT#y)dpGRouS;LK1{!X766t
zKe-*HJUkpfmt*N=Vn}SM&zooh3Wu^}Lyj8~{R#KJGBwJLJQ_3t`O*<S;BNv1BRzt!
zW9k?bT3N4K0zva?M-8pkKU>hDEB)3ySwR|e_UB~l`tz=P8!x>G8dkR7Ek7=itNykG
za6MT^=M?r;(@J@~Gt+dy1i&^HAQC+F1q$u&>f>x&-w7scC1iP|S?}|9Y>%0DC}`i*
zN8;^Jzl+K<_=OQ%#pJEK<)BW&UZu9FU!ujVZWI~$HslTQW)klhJ)?@2Z=o~@%;B;v
z3GYo4+xCUn=u@(TXy(oXY0gi@+|tXn1c%SVM1xE9Lp-D%zWEkYn>N*Lf2<w5q%tg3
zdxG8Et5zm|u-&X*sx2AYqB7*3l3;bR*_CPy?TE$qe+W_Q)opE*44S&iMp)I%w$?G6
zJ3}I|s+#GLeEpV70;cz$b7#>WhOIxxpLad{WE%NsvwUWSwouhIL~u5QW`l^ySG#BN
zfL|xO#Fcv^gBcJ_etfoH^c2jc$X2dD-APb2fFOH!)$uBrYcZmHV-qCP3MHfP90UO+
zIG6>B`|jt6V;{~^bNyA&=4%Slh?C@_(lYVuAaWi*X{$abzyU&H%^}~c=20oUQd>iw
zrE@;)%}vuSFe6r3C`MB%EcZL1!jlKb(>A&OIYb48UOudo#AM{J5&piGQn@r(c}_Ul
z-a2(*C9aMr#-5VHqwZa?u|Gf0)z2MM?+2s!?h(&7lmBS1{i>h*bJLw-PN$su+O}TQ
z%L!E+Gx8}2t?p+!w+BHLi<o!}h*}D9>AR9-Yh-eEB&z|!?(l4<=X5!OhRNgi$KZya
zw+4TI+1DGMz$n$GS9&mcJ5#JyVo5))H)3R3@O-m=&9P~$3}E=gD|Gy<>b;p)_y*6i
zh>NSSW7C+|;x89Qgy1A<UZS*;X_tdYwQ|nR`9%_rfAMF+2&D-eCVMYqUJtyeAf3Jm
zkflIxiyo%PSk|WT)qBY|s4otWU>p04;d88MPoX;H5Ri^vUrQ?G99&ER$zhw-zWO&P
zq3nCWd77`fX<UXD0U>>7<!_=bhzA`K#M=~ks%l8G8-K@iB~_X!!i8eMEenfjHx!lU
zz#K||<&k`X+7DDOh~|Xplxu#;7pAO}5AB99Tlby_+=bg})2_-YIa_scWNfYkUf<k4
z0tgyBAl$#@7vv|WcrW57&Ai$-yxxt5r&Z#Zc7&c04<%&t$;UsG3_=WFT5c)7AsD4B
ze>1Fa#ZJVZx(0BCsDO|X<0rC+I6x^-8&to!uGZ9a#8b*ga4>?&E$9ixW=eK3jJ=|_
zuj?$k5ek+P!<)DutI;LNU;aGVT_WE7I5VTQ=nNlxoGCmqUa;bde9MUX?0J2F@4!3p
z=Luj6_Y)w<kEqd(Oh&p+&|~S+AFb9*h3~6z4CG`-o}uKjrHdLW<*7K+o8E6H1hIC5
zd_b}XOs!E2q4TJ9pQ5{`CCDf$lUQ^kbduqQ1~Liu9veH+2%A3d|8q8U%Cff$7}5Q>
z5raiIP~0>&@(*m$S_Uja%VN=cN#bVd9LIT$!0hMrp&pR>dD+@f;7Y-vxnC2@XDvo2
z#*4>c-fC6yT&*JImV9uBK)cIo+>3uoqLc!ANIr~+hPwNHZ5A8)Xd8M&yYKrv$^{(0
z>E=l|YRe5VT!#BkVQxGnB~b0BoskIl#eRaLw1^AQfMi!xcW}93dMf=dYn{b{10Kuh
zpyRQfcvKp38VPuF<k9aI>W^RaxKn%{;Zud@sfh`G?p@EPr^Y49xK)x^sl;Wrdwaln
z-{{jUU306?ZyNu28WZ;T9_}7Vz+2V#7KLv0+tEqs$e(QS(?GkG8XCpcNUHgA=g#L_
z%j&+P1DWFHuatR;+T~x&PZu^*<qb1OPe(Y6Z5VZgy;PNKq+P;(6478`!Ma+yNyw_1
z()mkGZ97(i(!YJifS1jY6?-64b0PgJ4OZIv6f1NqEuSU;_Xc)ucKG6LjSCwc=Q+25
znD$}R4vQqJmsRG&femL3+b)!VvZ@@>yCi>oRJ%l4T?f;8qy^t}7|HPayU-aAC2EiD
z5lEG>K7yU?t}Q;c^U%@-c_TN+h~=G$)rFE~mr+qxN#j`0uuid-#kMqE9g%5?kv4a5
zwT(yBJ3Dq_f-pvk+Zld>2pT7|@J_HDsm3(b`Tj?>tLf8DqgoZJ*@(e9tB;A6_+us;
zr8vl02rUvLz$}BqM$BKwxkQ(VjR3t_mLg?~L3nGRX39cX5h5HBrEV0EeCX}S1d}P^
zQNwhl(PCR|^fE{)YxXaPQI^liai@EthQffZk8Q^B(cXj{^+4=?8+Ue~9B8(TkX<KO
z2Mpxr3hxfb@EX!6g<P0&l1%GbBQe;N^CU*6bM>5FeRpKZ2T2eZzR22q`4i?U8HbO0
zJhA!C?(wLy#~D%9?D6IpZ9jSAWlr})bG6w(O=vWWmi;+f3Zubi8S3eOf+B+%rj%^)
zk3dd@<x<LT4WBIeqyVWE0O{8@vRXeeTVPc8-riTU`bxmlb6n+XrSAy%g?ux|p(h)T
z2{mCvC-wNOl(MtXXNi$gCe}B-AHi0(Znq~Z%0_C-m1PcbNOrRoZjQf!81o@Eq;ioC
zARwzmCRd>FfrSj{Q6Fe>m9XCVvef;X?kg-BAv@lqh{5p9Y%bOlU^#*(gL`%q$nI(H
z!Cfl(ja$CGNd`AScIkH;0W1k@@45wLz~HPuWw3n(zz$}H+1+ygD+eUo2@rL)Oo%$h
zR`9aD#^Mffhz2e<1AuTyg)Bxfk!eHK$&MsLb3>{Dum>w&dxxbE8`#JlZbCCJTt6bB
z&Yrh#0cIs_UiR|?f3%6Jr%E9IvIL9lc{qjRGzs3FIc`a`n+4Vb3zlY`5LQ!Re@~}k
zx<~>NSu!hR<R3rt$ShCHt6+;`*V^cY8OSwx+my+|g6pJvzPEWEa{~|862nl$7SJjJ
zyF0#ji|_qL)WymsoWyv4bTA~42k-oNF<HtDkIWpD!j58-OWBcab}s6fn&atuyq1m7
z^cmJt+$<!R4YFlvW{K!oq?%$mp!$JYhRO{^tHQ=S(kp3=dIZ~Zzrg!b6_r3<5vlz<
zo=F1@<p}6v9buO5pJ7({LQbXlhwQC&6%<;z?Lk-k?C#z17|T;F(^JnqT)K!O!T>R!
zn#E|>F5=p!l32{}_UE~-EVTT%)WNb2%n#@SU%r$Cuz{+>oh^COju$op4efthoQhjX
zrjmg1LXp2F!2HILl3&xPgzdeBax>BA&1&GS06l0JHnze6xncIn_A-tKEtHZZxTkq%
zIxL~x7Ii>)evg~H#r;+gc;p*@_*@t|vvO>2mU0gA%yJHPue%S~)WV#Ttgatiy`cfP
zJkVd2i7o@Suk9_lf$@h-zV~H8P4z<6WvQ;-AU*pIB|P3DuE_2=Qae_yylN;4r*@_G
zpz2c>7=Ec^W&N0A!|W{>XNrI|^s8@Br(Q*%Oq4Zv%rDyL5xd*o@t886mF?sO352tz
z@aZ9&u^lVybQwRQ;vNGhG`g9_slP=iX)IO;9tTMbsZVST9wi3z?65C*IM0Rr8>45R
zA`*@F4lhZAL^95F4iv$J6=@S{i=stj)L>`4(BfXn)F_Rn^Q9qGCJq}l2taT_s$#?$
z)hmQx5wWZR;y!#UW+>n<kH~w67oCToxUAz2IB37?OZT|t3LpnNE|_rIGfxI+J*sPk
z8n);ABO;(!El%jd%656HJA_CrZ1jY(pwi9i70G^NvAd5lByRXTT?kQMK4k*ES1PiC
zbpph?!C&JSC_Q}0V)9Amf%1|`DN<CkZKGgR{2cS4cqjGN?uiF3)M!FBgI*o))#8UI
z$rI>H4g{mY4M;w!(_M`_3P^^YLpSGETDJzQDR1_G_>1m;QvEE-g|(jjD%JeHWjG{^
zFjFmF0`8UTa-iFs=}Hb3+o0zPkNZN#mFeqzP|rCcAxTcKv`FC>nQx8P!l>?!%dLRw
z?|R}*pHPm4V%^=h0Tedt$1iaWa^FZciTr)qfxEq~Y(fNrGtGKqG)URVJRvr=Gu*Y0
zwC~jkV?d|wU^!PBsc)+&k+71C>`~2zW5@F|WK!qt8svV17wJ*t;{C{)9c-DAhXCb2
zTi`k%@1#5%``mPe&~q?bIzS<2hs7pf0_Y(Sg3LR%yb;FL>-szFR)i`YQyA@(RaFVu
zR!2UaV27cpIz)9h!HpHn=tcD(vKTP*KTua%xh;qpEeD~&ojn9^{fw6`U1@}D){cu^
z8a`EO$tAtj20EGgOvU>yr*zQ<t*V~@5GE9ml);YqJh?9TojpWHvN{@R9G?A&bd`u~
zP9<Q2@mj4t8xwCc(wwbN+Qbf!r|UGlir&@AWQ=)S{&5$L35V+N>?rqXi2uHcE0+5P
z-!ESJ<N%-Q?55swTz*S$W$?2qqz92PLwI1kL6Z^JkgDV*jY+&{5Cz+PD4~?SUGq%(
z2YU*AFhWY_(IUp+;Cf0IMQ~>%$AL+2pMAh9H9iZ;MmXcPY={*?B5Tq8c&bJ}d%#8y
z8`dC+dOa+bAyo$DJO74d(~xhZ2@5>;3r%DE_I9SHuJH@bm+<voy5@B)CLK1AN-<u?
zHc$OUJ1}=C-#g}D?r>R<eVc{lDLsu>*Z5qy%BL%O+^J~=tCE>eHXwwG(JaVuYkM#D
z#QWt*(?8w5zV<e`ckcIBv2cG16r$85mK6@O!T&+WgdL4YEn#zNZJ<HvWroS~Tguqi
zOE-sSFC2&2=0>MX6eV>|Yf~xp^B5GrZ%3H2KRjJkK2>i@Eg80zSW1mFQ<mZm5hzC)
zuX<c@3dB%a_`Uj%JP)z%l1xVg#_6N+C9|R@%YEq5ct!lQ!(WFPP4@T#&q%?ue8SF;
zq&x~@6Z0Q9zM6=y-ivGww2izut%!0NT&hN|vPWAeTC|_<bKxxffqydodToyvYrTJQ
zdlC}0y2`;a;lvxUHYAG#i^rLJwhrj{HrIe*MT@F8K|=Jx`(;MHj<p~6@M|OpmElM4
zdnQiF>K}<cY(FO-$od7LXH;IXS=0uaef5Ar;LRT(HO>m34-ccEceT+mpD9t;_o<QT
zNW}FZk%_7AAz$K+xcb+v(!Wx+&v8fwR?-A_p?MkqaIYouts;Xs$F2*tt1weA-rM{f
z%tt~4G&!Yyt0n@r*gX1=yr7i8`-eyJN8~w#rO-`%yrbr&uqB~Kv!O<a@tL<|q(dtI
z%SLstS>(Xjr9eo@Z-hg_iqk<5o6-xQS*CdeEPmBVy(yhz1k4*;onqf;J{|2@iE?Mm
zwI&f|iu-}_7|kfVU%q8SmnS`0qT?2_5Yb7(3_afma*~cq)+wLtt6hbgE(c;Xyc^)c
z9sev+xVK>F20)388z#W52{zbja{Y`XY>ylD%F(8`1L#eO(CJz+L-gG=*%?j>;tR>O
zFfuB0*c~Mn02nH!?mlZRfcJ)xjceCKm#b!oK2OEm4GnY}MY`;1o%qw*fe~D^z=0}+
z_h$1Rd&t>+`>PMdBW(2E9q9mL9Y)Lmx$MEh4MczuSM+1kVAVP_At{MInnz%6{XP_p
z!eI<&&@@ZpAV%3i5__!R40AxX3=O<h?aOr2E7U*5TS+wz2*Sg4LRY*UD8BBoU358P
z?tQmP6PUrsMp*}mk-a*$I>$Iejr$$DoHKgFDBpz6tGahM-nx?IpNTk+V>)DoJrAaw
zvlF=%hyd*QIxfH*Y>bkBgSlucZ)WT`!=ZCI=~o>^=@J<ipTo`w#gy;oT)$M}OToUl
zTr{E3t0=`hC#Qs6Dz{OmFIj(buwuKD3Ki~jeJ~lC0q!IIbPN~_MWO+AD@QhQAqnZ{
zx+aSY&q;m*N$JI7a7Sx?2q_`rzsn1%U5J_L-FWyjKv%Lb<;p26DYPtFepl+aOBPh`
z#;LAl_Y`fTDF-Yqn~nsrJKS&?4HnpSa(#6BD$#?n0tTr$t#!nB6xTbD3Fnh3MTo|;
zRuENfJlFjbh~b7z$MN!X^|aet1BgAtLMx2fu&>LGf-o;qlSXVl2#MU}ioeKzI)x+>
z3R`PXS>W)%-<-JC?B%BftjNg)TTgTjW-ZcJubA%}sXI#$vN=htwR?0@U@{3U3phS|
zlJfXt8{VcwK$t+-6DY7aX9wF7YP@&w7q1WG+|6EwMUqPGpPj|l8Gm5W<ZuC>e|eKw
zGaUsJ5g&oi!oa@Dl*$~zrEWtww3!=toO=qacv!IID0ejLv&}Zc974!BoRoyO&}M+s
zti?au5^3A#$EE`P=dWn6g5jsKPf$F>Z7;v16?@Y&)gM@M68ZR@r;fPYGz*yK=w|(m
z>vZE%`qZNRdEaTiGGN3kRLSx)XBGK$JE9Gype@S6F{oA|8&0HWpi%#KIPSR(m)e<(
z^q+e9^kpX<wxH1=@2<wC8&@^+Arf2Ee4{{>_1bYVn_JjZS&WmwMXVlztAT2S>h_qV
z6Op`)%QVZ2R9VILoz@x)bQ{J-0f|WK?JyW@HC33(U)rU+-9rg4OFUi&%sfjU*c){;
zdAN;~m%smZqu-2yJ1{KIPPO?>y!6Z&{k*uKVi?Ekn<=<uifF;!6O8t9n;H0iGeH$+
zy$dO5DF!w9{hC>jgj+q#hwtkrjg!A%01{DIxi?)t&^>A8Zpl8=#=Dw_mUS2<)fPdo
z?N9$3<+w22G=e@VP%b^zl6!Bi1^H|N3I@KC&=cLd&`XTDB5p06ax`%Cz-B8^r@~~R
z)&bN`(5)6=NCAS4e<w~A2EXd}>QlMwxS?DLZfkuS3i|J#10RQ}fa;eF6!_k_zt;FY
z2AikG7afmzr9tZR%<EcLgz2B*&PTyDvFXd6UMx7LPW$i#RX#0|EUSGPOZB{c8BoHd
zZ^F|<g3FRIM%!3=^-=a8Nes`=dJ0nsPiaEqXn))Ab3QQ~t9ds%6^JP9CvGdjo2L8!
z{^Z{kG;7N>M$2oT1&2Y7S|PyB_8ZxHo}oLHi?6;GvyvfhD~8k<o%EBLFiv3=!igg`
z<EreGs(5C88ni(gi6`8hw&+n#>jM(zSh&zRJB7cAuVam|e9gjGbr-o8?Ez#&t6rN0
z*Jg!ez^zRU%oGs$eyRRfPl@*AFNh2xCIWUq7&U;VhY4yuRQ&CO6<zmUGNd-^G;_F5
zs*Zi0^n$E4W5d_my5)?u+vb}S><V=)R%q97UeNt}e*dO+j%7}9oZ7XX%aPB2FOmDd
zbFM+z-r4C2rS$zfi(GePRE(ql^AaRY@!v;Y0zNO%|G47OzdS3+#(3`0nf^IR#8DC}
z0$Pw^_L@y))hPbf0MqT{QL}_;k|Rt1iJp)QWW}s7w&Lf9#{}Ib8AJgHZ!{>n?BZ(W
zHccRmLGa@dkFWUqxIcI2$zfJXRV@1TWpVmziQ;?eGV=ePh&7a}Q~Qt^t+nYU&Gq_i
zvwryT*rXP+x<>JjQX)318St|uvV~uJKEXk6^jUC4iaP5)pR47`nAe-_42#@t<Jv|G
zMx<H(Hhmq$hW&;D5zpk0=@<#<?+X@bF>c9lIBnB+*pKJasghpx$0F3L%)+j>t(E*n
zGBwVCsuS8a2Z{W9wu1PZ4b2I7IZ!RbgAAPrPX1`#H&a98Rbc1$H8)LO#*TFllWtZA
zAWbhJr5!%cfu+aCq+%(`pDI3pzzUq(#IEp1Zo=~Vy5E{ZG=iQOs(c=QIW|BTfNY$v
zM(8F7_@n9VxB(8>Rn<lQciib@|9Re*;Q=lUZg1Qx5)II*5T2B%-~Zot5GH|)k{W@J
zRM&02cJ)oSIJ)E2KjT*^)yoS~Yj>iEO~VSpV&tC3duJi)*VDDGKRMOzsgz=iN(2s?
zxv&V<`Gw04sK`;Bw`(J$9gWvI9$Y4nf<yja4lfxtJ5hZ~gCUwZzQ|?9hhNKBeSIkW
zmf?R{zb|ZzTzmT}jrfMMr#slYizk}c3IVXAneK{PC=y4)$0wjqG*qUcC_WRuWivkb
zicM>z*5P|L-a)`NT4kxq9YvmwX5ao92@mUV1)nV<s$99%8qO#dYv`mKA;ro^x0El6
zj2J12Wnec$g(W00hCje;!00Gr<QyZ6JDXfstv4OT#3u<}AUPp{QLPv!0=kvsdT6xJ
zW(d9;rC^bWq40?9X`aMYjkQtBVhzR#n83>6;d+;m-;PIXIqS8Wd<utiw##Y=AA8#7
zgiJXVhp_tdRA;916*=*(HYz)~RbC{_e;l5|_wHl%v>J#e#~tK#bMz5Y@$j(iyVgIv
z&D6iMw6gtRc9`2@SOUOY+WvnpAUjZa3}!4qA(V2cW>Q5!K3I%A!gpzuV(_;I>5tAz
zbsh>dQvTC|B;&5NVe&_TK(y;lC^pu5?f!&pjQA`B=^VvLtY;*V1ZZ!>3NVNREJp53
zXTHMwr+{e=TM#hi5*Qe)rh1XZUs*-@kd;HLt>S_lYYCL))_^;Mnq-cNIFEAn38ddi
zOC~uia3Ac=g*jL8EdhCUm4$1v^MG_*Mq7vI9TQUf%OAh<>YHCJ$+Ya<MDGUsWgt58
z@?p^)BemnZo%_?iZi&`bz4qEAL>u>eY5Jy4G7{1tl_HpxW(!HrX1-^wo)QH5Qc6D;
zsdT=ys*ln*bo8<V0=Rf;uNDS*Ut#&|qJl8Uk8<iDc>J1v7YcaisAr=*?>I1b9dld5
z{YgM^ZSaa8(3n-^k@dd6OJQZSyykXz@zUT{9^y&SU;XWCA8r25@DKyi`J4p&T!CZ%
z@5}$+PzL@XB@g2+2i;AxTzGr40xl2(%&L&;i5_+iPJm$f-<TLi(L3PvZ>-OxGh{Zm
z=<ef^mXyHd7ZZ^71`#PG#w%}}Uh0gG)`+K`?z>($RBL^%`t`-<n%`$KxibI`WNvwR
zX|pWgX0p;q=hnm|KdcwUdPcCFJAc~L1oFP8_gf|vSs**tAqncDC_ft(1eBddLa4N$
zY83#%p<VWIhs@6b&*f+hwVx3h;+BxBa^B9+hxGUyZI=6UYN8#bozE85$U^S*oLt5<
zPv`Hsfp=RZ$@$5jU-vdj9>>ZQs%bb#e55EC1rONw_o|R7I$@hdoIllruYbNI;0&xr
zlh=0I#h_0*<;+YNoD-TX<Y+Qlh~l%WdE6{|Lyvin``aZBRrkXs%5acv3Q0<5ey)Jw
z%h>GTVzFjW_X*cG*W1@cy#BDDkHeXS?fITF8&TYn$+HE^9JZjeSs5Boh+i=Nn-2du
z?*E;r06aH0caHn%DyP?tO(h5i7xyCoVt`-%zccqo?nMmv5ikxLVOL=ElC=_hf8$b2
zr7xbPHtG71Xd}cMP~mLZHTH(-%*Z<h4B<SKX6i0AqkBAEVX^%<y;5lK%T)I$*D9bv
z&yk<uVTLQ<L%NYsoTsB3^82Wq;24e2N+m+v9|a1*M(0Zns;i#-;`G75e=w(yV8g8c
z>-%+=CEHIx?ls-rei7Z@S+w$%kQ-<o{Q8*iZ%#`Ukqn_E5T#$nVEwNK|1K=i%cMv`
z1O~Zeed8Tkpp^B)*;$WE`6V`~tS%K`Qp8B*q}HX@aY6KZEQh1;c}F?{;u5p~HA2_7
zV#+6qVrs$&ok_W9DEnh>SBZ_A=pumU&R++f<^d&Rt8*>3?aD`ipg*sV`egNyzGqO=
z3)G*e6?V9c?v%oDGrT+$Jss@EMh)OkR4>(xokqnGx@1>aM{ZSqzX)nlMv)ko+Q03^
z4|E{**U_{>fNug8yORz1((RcpE%b2PQNi!Oqx=89oG^A!zTCIKoskr2K;V`Ol>GkN
zi58li8IRUFvdg&=edPtLCaCIvbJ^~Z8jA)PNdNCgA|VQkTbrDI2K5{C^a?T0JCZ#G
zX$neQbZ*3NXk<yb6(;f{djxdojezTsJE_^ExajDlj0QsKM;KIXk;%kZMT=ba!1;1E
zf56yeHeQ#J#CC7uv*mEk6rCp1_a3gjM;!9s8&c8Jd&-y?)WpQ_cS>WXK~_b?I+9q(
z9eU75;TZ&hU$ZyOL`5cwh9)dK_=;IScd(KCzW<ww)uKiwRf(D@y6-#NFNja+yewIw
z4Z}ga))yNsr(~mV3lP)1%mbFH)?8E4Y>n*$=E{Eq)81wLLkRjv050S9o+U+?xt~fl
z1*JB~a<N|nB9ELi_)pcw@c#&V>#!)_wp(~+7&@h;JES`WL`rEyx}-a$W9V+A1SAxc
z4(S|_ZbZ5pq`Tp}dEV#uyx)HJ-rx5p2gfl3*L~e*taF`f8DYM0YGj4>G%vRNT)Dr*
z+rCqC3RYl17|eDsIXw_f6B^G=Tg^`;iyt^!DQ}iMRgl)g4?uwiNSLB~q;EiwZ<Cxi
zh0o>SEQk!@eh~<v{`VQ@RGH#i`e@$7nN8*mSIC37A%zM;7n+1{W5jCW6;oJ8$m!ya
zsLv|pKI-IVo!8ZGWttrF0H7qLvbmAk=l(<WeSO{3v0=T`*-{0=>{%|fYQ`^0DWa+7
z>dJ#NDL)hCWs$m_5QG6W>e2a_Ir{0fT*eM?-so>{4?^y01LQuK%>!tb#=gq3bbm$p
z&wDHJ5l_yv8+|XM72I>^O;7sy{y`6?s*1{UJIcstL^?Th_cwhXQ}-%8{r!X&n$S#q
zll5}};YjzGH+@f)L8gDA#5dcSX?%ef+Dp<WmdzoV52!xJA=QoFtluNdHPGc5Dk72l
zrLvoAgAZqGl+u2bYZAY*kVM1*v!Qf$J<oBmz+=^-dMQ@OX0kW!Gu<pC3NdgefA<da
zXg!Jln1kmXj&$=}ok&`U0ps1>PXu;Dw;uJ&^hAhC`4`l`Qv$Enx*6&>2zhk>P%QEn
z6*ur#wmt}FpGU!odK+xU3zMWxw;vR<r6kZHmee=fs{2x#I)9kSJXsAgM+i;U;iL}1
z)A*i(B*m)fJmz|p^I}$sRpFNtn&d{NyL=4F8h!@3s^?6kzzuw$m2mI-L%>VxkL^0X
zsDtrb`7)mHxKZ91J3K4wA_`{D(WPvjMYD1YjvsZ1H&%}X%HJ4(=G~K5`sNQs3`!27
zMD<zfmoElPvh%guoe_8{-%zw!IG~|^|Gc6qZ*tV~-@Y*627|rx&KJE8!R%kCW&e}a
zI{Cr)8wD-Gq{HrqAW4}OZ&!iz%(zsE#e-}r*LfSGLhqKjxm})A)PRs^bc2WO4|ZNj
z-`aFW=9R_4E4V#(WEpdv?mLD2ERkD`0c$-QhOJ*GnzTMf%7_gTauyVH(R>cKJ`F%%
zul0`jNSV#s2hp&xy>O^OKE9ZXQAsQvHzIA9YkcVq6?7X!BzPfJ$EVQOoy7PIB_2*2
zmxe%iWIwYSSl;H=vEaV@dinDwQ~!&7w$*IXk(r<Df<DuTjLMBvw6AeWvDRu&_a>h4
z+!^~<^CEb|O#QjdFjN|<5?U-G7mlHbqrSK1O=F06R+XUU8*4rIwF;6{4pl%`)Tz-7
z_s+sQwC<?nLx<LM##j?PhS5`xnjW83Dg0v+qd!a=e!OOpLHaBp1_i2T<vOkSaVVCr
z{GrV{D&IuQa5**0QKdIIf(G7oOFqO>mfawqbg6t}`qnI(#d?SYN%)Nmrol;uVvz|h
z(=T>~Y7}e4bT+Aab{4As*SmuA0z$x|q5d8}(j73@Wdn@7LW-{NBmR^A!u=oxzs;y!
zf3x-yPow*@H77d{2i>knjv&Lm2FT;#Ze&0Cni^DiiSQ20TaT@&+ls}f?|#j8!ocNl
zH_v{9n}>-#(Y9%w8bWvF)3fhOv$*vi-!wYZ%fE?IjVY4<kSa#2+pDDX2yMMnn{6~(
z>z46Yf98v_c&0FB$ty8e%TTML{+V|2ez#oT7JX-$V1DQNN_F8(aXD5bdT#->;_^kn
zgv)%r&jN}^zK&X6@Ux@1OzwbZw9dRW*0bf=1lE6YY_hmLBXd*W?LkV!f+`{j4YSVY
z6o(tB`>U%#uYzp7%8Ts(Uam$Fng>ITwpGlY0W)3^xG6Wh-pTeo<S9%`O8T4IQ@4}p
zhNQhv53Q~vdOf>3VPxl7w-h6g!i0GI@<uWt-{wv8u7LDgIcT^9FCLJAzMGytHJ1NN
z#`^P2g?71S#&+<eoqHP)<84%6CVWQq?`i{w2dWaE0bt!YLz%}qL!B+2x~(Qn>F2>N
zSvu`qT)(JJv(g<``pO(HzHBawdTDXz=YPP1XN}&udM^2*mf_tkd|7QezYwr_yM+)i
zuXo*BX`QRC4cC>VAPe~$2rp1%@AGQD_#A>6W$LSm8yjZP<aV-Je%QJe<^8fMFB`ZH
z_*DLr7Rkfqkg*2CItTZ5BEwP7UPOFqa0b7S(Qu~dkab2Pgwu^xdualL$qb4W_EZ%A
z+8BVoj)cck3so40xNTLaKOfr^Fbzjs^hDlT?`I<w7o%iQR3p57u_eU?lv66Nsc2Y_
zKHE$L1dAybT8&HycH`!k`+etFq4w8>=PP9(#Qs92+DeLp_ladOD*Nv)C4sOa0v4?i
zO^y?d3YSn;4-TOz1p+|<+MlRop^|Ul&icMfJsDL5lG@TQqEPhcRR9ta@k_6>iKPP$
zK-$z9mahyvL*T39sYhG9-RpN49CppDJ2SXv)Ku_s<P;Unuw|RT>?yCJF4@fshxa*4
z0Llob$B_|}W?AN3#?wO2KaWTyT*9Xb89YtVnA$CdY|E|6G<X^Ykg?(7eC#W^_SSh$
zYo;99y_h@#wiqd1nDs08Ipkn^Fo(T1Nz}0GN~Z1482Y#UKyN9AV!OD<X5DmYKs)dG
zy_xAfWzCc<TYPfFinG^Od=>!gzH^nPEX~ou0sG?CKl<UZ8UWG(K0mm1R_w+@yZI>N
zcV1Wn;1i&Uf6702R9psVyvnA;-P6cMctE@3MnV8rS)b<Bjq+qzr7_a>!~hz=qR2H@
z#OvfDzb$+RxqE2B<G1Y%18RyO8p&=<2d}f<P__JM$J>1_FNBSrS9?<<9kB@kF#(?Q
z4Fw`|V_#3fhl;Y%RPxRC3}<hkfzog)PNI@QW<we(X;4$<?)-DdEKw{;rC6rfE@X%9
zKg0_}3TU|3|FeO~L5HNLr&qWi8mM*xavATL-W6bI$NcU}$bI`yzCa-dZx!OjUk)fa
z<{N_WH%Z>9y(ju|JLFX_XpLeRk}u2%de_yjV%vqpc~y8Y_;_y7>yd|E*ZHZ?Az(k&
z`(I}skYV2+LJ3<(KT`S<ZE>nZV4hH2oXGCn?{ojY_u5FTj@>D%N%91p!|4|@)gAO=
z-{n=Q&LTsB`(8zRj6mY(P%7=8YAbeK^?pjmSc;gSNP<3;SPlhd_Gc21ogE0C_9jzg
z3ebMEEy~as@a23Yw%8se1FEMo?V<h@KKR2ub9_9G_o=`Tm6}Y*)1&qWl$h0Z&~=y;
z1^q-H>Iuw;BEvr6`z?+1<1P;g>>thAxH(}=HhvjgM6zA@jB);VOmCd17671(Alstv
zeQjrX(J@GY?=rkn)1z5t0XP0UO(yQbP6OE|n7&X&W5py1s<~z}Dl<sxYy!W_$M~<l
z<-eO46rel#Mh66U#3%Q4m0DYE+(RS;?u|G9l*8Yld%A65r!Oh}UH0vX*meMsLX0ip
zn!r0>SW%hNy#D2@1ossGLP`1wt8X+U)0uR3HzIyf?pz~#^yuHJ2=4(F3qYZ2dOCeJ
zu$!RS!p1<QmAmv9#NimXYMzPn4$ALvL6}+V;6QU+Mj2LMCV#D2FMhv1G#h8kg!(L8
z2qGOxoTJTieJaoPDFTn)E%SXw+G<D){9+%=<KA_@vuQg1cx_FtnY5Oc#23-Rv_8lC
zlJ$K@zhmt~7m;0s<A=Z!52D@FG)sRXgYX~#P7l337++L7=!$|9zBBDOUgH%w$rt~)
zYi`~Xx1U(Hns@~`b8*62f2HmLkCh4{hO6TA;wFt&`hPZm|9y!6L;;*!s9t(1^;*}f
zc&>U5LBA#49|Rp;7`+C(b@AE1v3D^ABrEfGsa7f(L4&QjrCfw;NLD`y{Ttr8=I7O~
z#KK}F7MITWbpFu-C~#<4LDi>34*D$MK2OF3CxTqz>JFQkb{6?xHj@!L35d_2;1jz+
zbZhY!H`8Za0#NaMC@(t}amdvH)qda@Mcm@$q>;lSihw^%rzTOn?T=D~-&H%}lCP%6
ze3S|u*E_o=eSe_j=Od7rEHp9o)G-OA=HtDBB4P9b7}J@zxc_O>&iI<q?{6O2MxWU2
zf?anLjXF^>lz09}iTUaIPnPn4yGGXk;BC*0W1Mbkr;p)o@ZR5Sv@>SM1E4lO0BS!x
zi_gxwNoCBZ`5(RW|Ka$YpqY$uzd{VM+8-u_{TA+VR!CM?sXRx~ReXl(aHdwMRn9}B
zC*XSJV0`@(W-U_w%_&)Fx#2K$Fmwly*jvXBx4HC>)MJX&fJPFfj_mwl-Z7!iesS_T
z1RV`xQ2Uco$QN(AwWY#xwH;Y;GzZ)gdA9X~SsP`csgBbI)}O;GMWul{@0M}Lzsqz~
zs-+t8IRfl+ZC2Gm`sUtpzFzNYm&2}SMAIMwIq8i)EP_r$ed-7Y*!maMs#R!Pi;Td`
zK5R(^y$<WX#zYxO7g*{%1?&Xr@E#tT&?&r_NHd40n^O~m^E!UIJMau<6|=6oR_6ZH
ztrNYvvC*l|gRn?bL&WfMn#{8kR3ZPBnf|$1|4LK;%MaoFkZa_JQ1li`$W}s7l_HO3
z*$Yy3BFh*0glats0mcJUuYXsJ;Y_viXp!(CzH319PjQPe8whI9&<9W*Oe>yE$uUG0
zxO!E!UP|PLrxC7xQzR+JK|4FZ+g}cS=YIk1Pe~^7%oKc<!I7oEHPY9?PryM<suZw7
z@1OD&o2-7)eJpY6>e$qGYU!(5S?06#&WFd_>{<vfZ1>GhxC9-p<yZ#_VfxArfL3Xg
zR)KY4`P=;VAKNEO>CDxj`9Ua6ZbC-0yiEmsuedyJFSxVbjQ?i(Y02og9*&2EH8^R<
z<20{VNdxw}z?fe>g9q?yHSs@FQ>(T+5Lgm4TJP1f<B18Fzq#Q1rR%^PXXfb%$`YPg
z`JTPqNQAHu_?g%kvWb%_+x0&)&HsLsqXYqWOBWJ1@Gn%R<}W4(+%A11oj%c|6;Fch
z&_f2UJy89#Xf=!`gV!iMS@9=wxY2M_aF6mFnI<#gtWr(f?7jSRa)cgc*TEu3TiN;{
z7{2a%mVvuvv?7uLWT~??F$%;?Y=2(5qEWrn_otCwCxSxN#PaDOV{fUb+d-J5Bp<ev
zc23cWXw{3&{e@ggj;_`eO1~{LC=G^TlJfG?7pgEI-pWvk8~Tg+j&gr@*jYCwouRki
z9Q0^l!k2-n<4F`(E;L#E%9k%n3g*pAwsI4H5LA8F?zKBfINkd4`9Tv0UM$7L!GpzL
zbrD{B+dx?gR6R>FwcZcxws26Cpp4jQbTi_139C^LpzJ8nGo~j#*b^NY#}4_pMW)}A
z^dD^Ge}85FeVAf6j%2ggJ0lkw_=lhSH3Ml}#GyR5(fEv<EB3=^49gpWb1io5Qh9P%
zS<9Z9Wvq|Z=1XGx21dx`1hbp0dW|T{jz>Iv8SOxSBAzMPp#(x~MqK!z+ON@!*8Tl(
z5o$e^LbOF8sFK#;j<x=T)Wk_7G2Gd3>Ar(Bxi#cya`1?rf2w*}&C&Lei>o5EU&SE5
zZ^;2t-fSH9(z}|R$NRiW@~ce${jbBf?>f?GfGtAuTD$QrLF1p`5`)rGk11cr-ATT7
zY;n5REnatAGvrgLBS2XEaA}|Jjc&*tm0OR1th1b~`DyEjXu5I0`=Ihy0_|*UF<0>C
z&u^j4D3&5<US`VwnFIaLB>Uel?%5l-=fxSpF?%GkXRLOi;`Rh(B%3bsWcQFq<;R`u
zCO=G+9iQR?)P>vas>LybLY<m*X;0WAxi^QT{?>3mg_}v`su5?eOVW@(BuFVIhkn5O
zkdJ%7;tY*jW!o}9bC_I|*xMK;L;>6jA8IN_fSNxag=;=k1s@U^$c`e0txDlWQX{`?
zT!Oh*ttAy(O{l4X7?fkjd6oNF-B9<YR{&J~cWs=tG*7JSsj|4DDWPVvbk;-4za~#E
zih&vSzlRe&-S|0e+O-`|O!pQvuf56UcVwa~;72R{QFuyU@Rt|8_Z?RtL$V8d+eyIU
zql9PebGaw>`agx-e;>#HdQkpYs3;9W@m$|1>mST+<6ScIU(qtm>!bG}WFsK7wnS4g
za-KQ~QL?X3PXE3m<ed63g!$cn`PmQkK9KW*)uW%=p0n|;^berq)LxdU!e<+h#STD3
zYzCT|4B9-p<(D6sYP(VGeyFiaEu(>JAO=xgpll!GAa1kuKHrB1JiptRVB?cdl9Yd)
zQAfWk^RI7M>0S(Q^t>)qqpKN9v9{ezNv4;$Jea8>RSk}co<OUxeoupEg{0iPYp<I?
zBXf1LOIN?ozc1CysDzRiE+CSzDQ`b)EpF&N)ZeYNVG<zH^YO2#g$XN{#z4yIGcX1C
z^ZTRTcyXY{kov#AS-@|k*aGD7452-SZBBC*R=>$2n`?*%lZ&0!wuPu1b}yGjox^pz
zf_uI!hpI|Aq7ug-{m@NyP@L>@8(C-)sFji^J4GY(h)Tscc$GVkK;w9j3F?W}_jFpy
z3wdVW^8PvmCV|*3w_5c6dJRe=F<`L{Jd~q2xa-t&DzPuwwaa<Izw^t(J=nNO_-pOa
z3KaUbEaa#^0HlT#a=)7h#S-3+0`o?J-?Pp8Wq^qY0_0cd22khTWL##XgC_}CaB~$&
z(*u-)E{FzcxW+)cQ#LHC3pE3%aW7LanKl`>NCO&NN&%~;ewUk}4E>`{Jd*OnqtYV$
z3=}CufRP~6_2HY$|L?`ofq1q(U^I}5ZOI5swNOM3`UtF+91z;73?rkXJ+PiFsQ<*$
z%2?3rjq?jmkyz_KDYux{bq}l6SYa7*ReA;24TLtYhzR%7Kb{Qmt@P!*>eoGZu_eb9
zFVUAsx19e);;FAO<(BvrdS`Py_7$Sn_)+m_A}<8h0H$xyTkF$+heh7VPk~f?@TpuF
zf{ZQvVK7b4HQ{|XcNIGifU_Gt0>Bo8zQSH#49NT9D_<k5_)}Tm_+yc;_cTk)aU?IY
zt=cW>X$!tcHt2VlDDv@m@U-iJ6guY(VfQq|iB(A!Yuwr7Hw{txEC;6=<VmyIQvl7@
z9d(XR!F3W#KK1|i4Fb+D=N)8*vZbdqothc-w}B@f#xd%=pci)U<EcRJQcIBTP;k$U
z<1Z#EUg#%ivu<r?3P=t>2ww0ahrT_If*&ks-q>89AiYm%yn8{1gh#-Jra;&2aJ4f&
z)fsZ9h5tlEh?Dp#bD|ajOyPXkC^+Ge7VlH#x79h1e~YmUoYpBNArPL3yG}fz_^eX9
z+#sYGpr+UPk;gVMJ-EN8O$11(Iw4Ph@&l~KiCE?Qfee+MY=wVZrkjc~rOzlXbNhRK
zXMA75$KjG5waW}@%a2f^Z4y+Vm>k!2b$ora67~`ETkLD-9es{Z%5?@$KJEYiiSR&%
zVakENYg6+&`O#k>1yvgjD+xI9r|mxWd@=b0{`&{DPlzujn7&voL%^nK!NpQ%1IAo-
zdP3--Xc{mDtEZORsfFL3jhl03^u5T{=*3i!e%CB0UWZO32Os$Unp#RCU1Fi}WNm^H
z=8<jU&}J9Hpm>EFB^^Zugw78u<<5#sp_9<jqMh!TS+qXuN#2fcZ=7^AlTV^lApk5G
zL_xGVnE4$0rFSS0V-B#SqexbY%M)`X5wD$Q$qk3Lct+OYM27LnT6`sF-EZD<!*9;{
zf2h8`);ilsLIxh<q_6EOjk7Eu2K$RQVbE#5mH&5F_kRyJ5@Aq)T!Y~F2A##WROj`)
zzTHPHoAdY1UxdLI4g!vN0HlhD#G{S120>GWpHGoe@g*Bh7rp}_W652<^DFAbbjlK;
z)%B!B{Z~1J!*<l&gK)MlZxS`xlV|#-Rg323g09{V*XvJ|&%(_iu77*w=|$O&Xwa)=
z>aCLr-hPT_0Jz;0-<78{1YhXmUZ0?liEIV}pq)p%-!-YWCSXGabt_$q>+>^Hh%ios
z2O_#%oxQt{!TTyR15mSeyOP;#OjA6PsD!jLNOF_U!YGFWUwX?K{bd<5?izhjg&9eZ
zq#Q8?SIXMsa;0QZ4wj7Y_kx`KR6D;`QK0#EH3P)lD`|M&LE6RM%rA|@(lEdj>(_c;
z(h49Y5dcu>J2=|zAQr=!;Q52_fAJCi=Wzx|prT+Qua{`EXT=O5^##ek4U>*{82&8{
zx^h$Nc%GO@UAFcMmwKJR8Y1%*tWzBilm6-H!iws#!3K=EQ_;?;^G?)pdk68S<-Nd-
zAGs=1Z|H2*A{1SDg$hzRiibmK0PL3P6`T^xVg4>R@L38+T|ggQ0`uu-kty)j$hZim
zm2msz{*pK;yZ-OmOu$Flc2Oouy*i&Hqk&v_J-9unh9J%6Rh6MQi$Z3vhRM#LKCA^*
zqeN~!wysq}fCn$fZcahaiN@{Vr;k}@l;PLVXBy?Tl_9Ln#F}>_*~r_Y?O3TC<E|=S
z@BunO2b9J-g$r=>ng(<(RtPH$0fuHU^Sc6%>m=l-+$IqL2dwm#0WVpDu|<zH%qR3L
zRs3m(0KmZTH-|F>;$9AZM(>FwPa2xGuIXSl_bLDYiK=5JP7kPKNhjC%_{0xY1Ov|v
z0%eCA#sBRJkJ1O+wKTQ62DGRqI!5Sw2Nf!qieo#Q8HCRBvC!$bcd<M*H3|x=q3<Lt
zQ2|sf!fve}=GknOgy(e``(sMb58-2$CeBeJqCwjXlE})Jm+ZAy6w_`4m1CMgL*qQK
zsIKv?S%+$8nJ3E4BP!4b164fe_jvy4)v0pp)po3i0;g*cVkp-6V(d#6^yTxZ(+?fN
zM-Hz!7OJ+-?}7f%oK;45pum0pLujSRS{@Ne4XY_sk3C>0wF(TdbDgopfC2B)WVXlD
z&<|G|^zops_S2EA#umtpsFS3>+kREr1uH5puH|t$2d6{J1QeZ6E7$u{bjB1A)_tOP
z<aK!k1rQH(<VnyRQ3ixJW;_5UpVO$V{kzM+vltv0(Tp4at{l~o19-#VeEzhG2O$6L
z&{jZ({)w2ci9QNpdP1`UWkKaTc5}66#iRL#pS@BtQ&I>kR~LZgmFGq3g&nN%OnWS=
zGnQHRtscjD>!Kw9!yY_w2CS+CKdJKPh%#|KJ??+K3^>9N!MNi0#e~au(f>u+!9^9s
zx?|I~yHI0(2$0SHMbj(bzM+Owm2NH7pU#H!Xo+2jbApZ+&~zv>?z`AQ>F(Nk)6abH
z7Jr}PWyri9&iW?|+YfbI$(s}hez|ochA`Jx8~sR644tUoUd8U;{vg-cZ3*$##u`&q
zZ!7-mNulHr0K&5I1_~}$6(F}JcnRj)Y?IfaT1Hz+M6CLMTTa^n1~{?Ud`GQcs?x5Z
z9dj~S!1#Cl&$|3<+P^`RerXbGC1f6G?-Y%?F+}iUyyA$8cu*ueH-{9D+L3P#*M5cd
zRO;(y**0%1iChOFpL*?8+Fq?ELBlV+*$nG{M9h_DJjbHIH<}l`HmNj^!87oiX(yEE
zEs=Tmj;f0q!X+uOS<AtXvRu31(0W~Pbz%ijXDY`mt?B=Li~*du)74gZ^PXE-MO{r=
z6TrS-NM-i}LDq1}J~lBY$T3${MkW-NF6C4N>>WsC@Z06sOqHq5zqKa$=M%Q6{n5*B
zfGsG2g^zf1keG|VF<)rq*{C6vvT!HMHq)<cm?W~z=lbml*g^w@^0nA6?d)cQ08tHl
zLB!xM68ASqOxlTt2W^j#y~UxrhaTnU+SHMg?(!+IB}iRbcChQNt`f)ekJNq=T)TBk
z!ZYk@o{l3g8&vA-vAiEs5n@QUB({d#(nN)_BlgQs)@V04`+x;#Or}+OV9@9P2>ttK
z1>II(VJeEoKu7@AN^s<Lurpe8m?<=?6L`)V<VVE7>1c$dS5F0NM*ANB*fojG<zz_n
zi5c%Sk;GF4zbFz;2^0J<NNmU2Tl7Lj`rC{5(ZH5&1P<ll%g{nOiW8>OO?nV}<7j`W
zPI7sZN#6Upz%1-YXG(7jqObi;$7`o6rz1N$Ov(|G5pV*?Y>fRgIef$>#m~F}7TLap
zqeXzw&_79msvdfBdb-lbm>s;gXPb={Ne2wFE->`!9jX76XyYw<ZjqZKzV<qARR-lT
zIV&k)HBnR0Mpg;fPUA?keJudO-M>Bw;A0A=dR{KLcQp*%U7t#`_>aw~SrY&ctWpy0
zq@n}2d0|Ms`fR#Q54r5;_lAAB?Zq%a9|stMOny0NQhw+HNCac$!!B7OEZ!i6d|>2`
z&NLX{eSxxv=;tAD_fZ%zO<)8}z_Qt)?>_UP8w1(rz6SkirvTW5R7GiX&c~C9=?^PV
z+*c{pVFxCTV%`EjvPN07!rJM?r8tY@+h3%g&_gO;<>>HTCvA^XUpqnS3=M1gk)$I%
zPCpCrP&&J8jC+5lp@gCA&F>f)>w4P6w{Aumx9d7!9-a?P3BU94#dQ>fZjaGs#30`k
z8tQxoPnIj2Mxk!>NZkJ+rs5i-iu{X*S5pkgK)GpE(kM4!xF+JUR}A!`9GoK)RRH*#
z&*O*k!5PUmQ2=oB(kx3Vdo?l21-}0IriTFgn>DZy6C&n&19Y@Y7;O(XHdNZGI{$e}
zaAXM6!!%$hYK%6r|CUA?9(9VAX}B!921qqU?nl?40P5V&fBL8q0qT6cwG=;lbWRIf
z{wrolKxty(W%^B0QB8nao{5GnyOiF2VaLCI$qW%iZ|<%(RvV-3=kV@>*!z|WEMaY0
z^8vUvju>QrZlL&dvNQ@2eE56@!H88`hCdVXOBRI4%#`;-o5!b|G>+Wm)#;|2IWHBq
zQl<){x<u4xj^=pmiQ=saec{s?h>z<|$E#V(Pj!(N+<jqTUx(Eho`usv-WT_00~~th
zs8|)~-AWig$E%k*zYXi-S+PU&#07b~MV-IJpt7f<8vLvEV!~ahkr#LIib9<32!j65
z*L@I^3M#PqBE8VaV7?&wYTU>BayWDeHZ^iQmgmHj$@Y!``JhFBXmfZdZ2Wso_O4DT
z(~sIp#E&OaWAdpWVYg}kQ<|N6g|6`l`PnFXz|n$gSOQ=)%Of@5VM!hL-?a}A?wOX?
z8vvF~d!y^_bc%WyH3tVaHxJJWKpd+8v_kkR4}0YcSYY@Ec_A?8MQEn2cpDbEFg$@)
zwi%%>>_5-|kjo8PyxL>QMLuc^6yZOGF}NW_QX|Dw%_Jc&@gZR~{c@cDNzp=g;i~7x
z8R%G=p$DyVP=3-jED`s{<-M70ztk0Bjy9B80Ft{VryHcUTmf{gjv~ZI8?}-c8jem!
z`XM6B>h%u~GZv>S7jyf2>B%rpxsBp{LpT1%yHrw<>Jo$vX=q-$f0chK42y96cR>s)
zgMv|!Ahv*O5rShSLc=ny<@_|%QdJ%Oay>~Sjh9&|s)Pxf2JQxS<owDn%-u2&i>i~_
zx`uUl{6Y<D$|O!3*eR;Un_i<0RLZK#duufNBjMT2rNA#u!-i=Ku}^Fnze+YXJ4&>#
z9nXHaxh&$+V^nz6V6Luhc#kZ>(DUdGesA}=<c&@<;3Pp7r^VL)c>QzZfxwXD`~|cQ
zIKAS9lwWB#IGb^13=Mk8(##E)eYPB{m5RV!DdbIl68HOeR`_=$qJpSO!{a~E%}`K_
zSJe#-J{Yx{HjR$!-3nLmTXh4_$`P*ITkZ;??K@3A_6tAP{n_m*l#Ms@u1brL@<LOq
z5xO&7MyTKYL>HlLm@M4;sseqsx-W$1WI>jouvAxnwBCuGhc2hhWq(!L4^Qpg^@pcE
z(O($2)F<WbX}@w`|Lv;dXtlf`!2_<+^4GXCf!ZHm@q{Wzao0N|0;24_&nn;|hn+!N
z><J041Fee~AwUvm6|&%Y*5+r;=T?I0b@4~(j-)K5<yc1aPdSIU->q8R{jENLpU$+>
zjrq=}8rDjLplK5FFTdPKVPhxerps;#8zCTqE2bx61E^Yk8c|orVQcaHcd^HL)bWCy
zZPU^FXW!7eNx=Iu)mpT&`}F^=XPy{Yj@Kt<y}iBh3a7v2ImQs4z2Uw&E&(iV=T<FJ
zr<MRo&hvL~ictUO8B0k7_;aZOVf}FvJ>=63rW@H6=?QGUztqSFyQ*~{<ZBX8;H@LZ
zvuMg(Mk!4-{H&7gkb@X#i>+V`_JM)maaK;^;(};7H9TC4Fj;m^#15MGt;^cSDd5P>
zns@tESU9Vd@h6sMLkdb0CjrcK<MDn`3V$=056&CuiYiqp)>v^lP1ZaKD8WK%a}d3+
zj)@pIt>shsA-Oe7SR3U6X(?7)mTLct8vQ&8z3s8g4?{KkCycAf^(S1OgE{T(Imd4N
z_eg-J7D*5R*UsM6^wlWsw~$nCGPF8RLm`SB{=#Nw<vGPp)O=+u_NOcheabK!VXb?k
z{maXT%ZY=9A=^7X7GpR!?CO=Fi(S{-cTgG}8c1$iF_!8tU`hU;>?&ahRe`f?tgfz3
zzinesd$s<Z7IJL+m;`J=dfo&2h4QjB-->Ylb94WR*z2goiTOhWc_@=kSN=Z6S)t|}
z+VIl!N(CWc$N|8EAYUBIrNz!pn!84y(-?GiLk$OTe0Gx9YrQQ3Xmfrqq+BeHOroBD
zh(v%r8`Zd4oY8p(YYO|N=~OcPVieK!K9J-MMlgyrzdie*76N-FgWlL$|LGIUTm&Pd
zq5Sz)Hl##HIxycxD)7>-JK965iVplXWu5X};coy>j$GAG0^GLLJQFB^@$g)&kkQS@
z`|wu|4PA5>$CkpvHP-2AU3*Z)3|><?th;l~v*@Y)xs2R*jMrUxZTAfDU{oy2X-=Xd
zjiG?8F9Rk`o5H>ZfMQ}1A~4HKrOM;`D@!>Pbh_o+RBP=O&&>0L@tj!1DX}O{{P(Pl
zCJv+w2h>_;dTlT!(*6@?fw=nbrj;hqztEW8zU8NBaK*gT2Q0@gFEKALvt4WwMoR^V
z6;3$z>LZw`Oria$0+OejcT;?B7Y8sP6nG}=p=$I&$|G2hylBRU91IcFbkxa8KMZF+
z7cKr{`M%ynR)kFappAf=ZFXmPZB_bG;80npICrt~U*Vo2;Klo^DWhun#4iUKqd%0$
zX<2nl*{TIC`adnl=r+4`6~XH4H3S{05X}d}gS^hKdI+!6(%X9%TS^sd)1iHRfI`EB
z*jLx4pH<N986Hx-o3ytQ079kJ!2PUba(j2pqgDJk`2Ek<X<3+~H8%W$WUVo`pWjoV
zJa4%!2#Im{%Pr=kWz^w<45%7}Id3>G?|eU929|S*JeDEh8v}L#L(4~8#0_HvyJ-aS
z!IdpXONz=FwRc2|rcPho<-f131@z?Pgw962kyl$a0s8vOKgat23br{y0pA#rD+SJf
z&&h52-Ed_#6f^6Q36PN?UsxHqadHo8H=<0>@%EKq7Fyv<9a966SAQr0Z@fez9oe57
zVq593Yvkap^3c4ls}+1)eh8yd{d7uGE^J89?5)`P7INLp*2q^6+Bj5)4wgj#GyoxX
ziOLdsc$)TRqg=`4sL^{?j)^~Bi4g+mb7%r>xgBHdTr%w1?2NtpIL711<SVgzZ@l2u
z(xV#CqvLDUvVyzg;!KQrS%8@eflOR&yDTL=yzm`!h0!a>xBQR?FG3$Qb;)`Z(MKOF
zGSMH{=j?M(0*!JdQ;ou}b#ZO~ipAynw|>`MPa(C(%Q@%1UQdWtA(QDILyG<I-FMKU
z-yGO|!7NNI-x-+Qi1<7-AUL~RPjwo1m1*Vn_%Uuxmb!3t{~qC5|E)^=gcHri7>_AL
z%>J{71IQ`H_(!mU5o6fQ6#fr+Ed(`<MKdHgtcmD?^+>+A9@Ma_@#|XlFep0rp333y
z0Frxi^S8f0c>cY7LvkeSii@+e%jFj{5QBu>`3b82Y-u-fv^q)x<$zhC#hu(W+xk{O
zykCH>zDeQl=&0t3$c!{dpc1Gq5<9=i@s&Te7)lJtet?h&&o!igmEzY|*^QCwDzn!E
z*_z4#Q8_^=$5{H>WhWbUhqu~tSA}@ca`~Ce<Yb+yTrB(y;n}wv%_7raS&GVzdM77N
zZ#PKuxdx>9&RO*1j&aCpS2=k$6@l_hH*tVNYhRd<%QsAt5QMJz2Gr4SeKlRsl^?Gw
z2BLxZT=YIdA)G@z<yI{Y70)@H{_toR!MCtnZ9V{DsD$!Ld{N)1&L)abDA)H^lu4B)
zET+)S+-?e-%Hcioo|K=rYJvL`;xCTYG{7q!=Vf$qEI0%YOb>a;&V4KN&O;96sN{L*
zw<Ac%`p=DFxA`9V@9Y8sn&0`x3v}AIk=uL6vnLy;7jcE0<p=Nkcq4yHc`r>4<T?vy
zHBzjdb-Xz|hp)-4MJ&dtnf@7RFVS5&e}m1QvA;!&vMA8&^@EiA_kfS24TYB#(S`J9
z^j=E9z@+{3y@P8aBXL8n2P|^R&pAOJR8IS`RqEjQ>t{T03!dEGDKBy2JN#m+r8~B7
zmwv?TqB8`0a{?^g*sj3<@*~kZ_=!Cwy*UCF%%Ps5B-Oe?rvhMQpZV6>s`~mNT5V!C
zn*wD&Elevj-X<qteup1DL#go~(!E&;!@9`i|7b_S_X3+;A8pBnh#p*?I59d36rkY}
zrg$oZM>hgkz?{W=V*=@Yk>0apK-VqMZZ7Xohhj!H+gce~P3w4^2Ws(N6CvF{!Jzxz
z9%MKAHH<Hwm4?Bt#+u<NOGMwzF=x*Po_zAVcm8<$F!a5<qSxc4*QGv-<%f6Zl&?y(
zR)S?ITWB`zfYeIJ{t<qCIwkD)_&~^TlDzZ*NdUJ)^!Ipq{%MUqem05}Uck12_3=`|
zY?YZj_3=|xO&7Pa1bf5;)V)hJwE(hPj?eXp!9(sDJoOa7a~ib7OCAAcbXOZ7e1c&A
z(x6m_?}tH4<CW)v>r`bj_I|tV<i*kIz$Wk+<ox#YC;+w)2-t@Ie03$NXEeYD2qMs}
zhMQoAM&UElJiISdL$q(&=dZuNn9BvETl}Vw2krs=vS!DrV$Ha8YmSW#*xoeP1<uy%
zgb83-bOpdtjqh#wGkS~k8k`3QZ}n+tOHwK3XMS3II9zNoc;B!VD>B_xeZ-1jC|Rys
zhn2#v->-2`<_~+gMg|lv5qc?ZtpItll~7W9|A+hADBlj2cc|FzkIq{W_vmppeTnqB
zfX|HPqfVVYrt9B73BA;g9DB;sbw`JQ&^qow&qH!T0<}_fi3W=<Uoyi*X_eVP2jIN&
z;-VDLIeaZCDY;BA$`qFEcZ13;#GzDyy$DdF{E~wusYMTurcIs$dJW|s^Q!e`m0F!9
zAzQD3MFRjZt#t&N`q#$Bd4^_I$tXQVLCDx%W?_>WK=R$$I4noLNYHO`^Q7m{C?V#x
zogN)9>{|mol!Ff29u4k(aUBBvsqJbS`^B$MpYRi0)49z$AHuev7wr;ky{n$hut&f0
zq7^|MfEmXDGZgg$)?<1*1Vkwg05xO69!zY9A-`3@Koj*>?%Sief#T(UJtj=9_gyHx
zjrI$TnrlVZfg7t@fiM7IY7+heh@2k(^$47LLH6#2`VU7}QLD+4bY+Djy`Iy`92Y}u
z^qZASO1~@cQRdM{z}?dS>Z-RQly1VnR@vJ6g>KDE=TEEE^Gm=Av-$4`Uh^sqRpDBk
z3F$Cx$NtfsUX>{-AZ$aaxn(F=2lpiWl6`PZ<hObU*^1E-+0zuv?h<Qp*;0ag26<zJ
z-Z_@}ui;(Tklg}=*zRH%PSLPCSC8S<^M;&X50Uqi<12UTfWU^iZUP&E=mDcrCQkYL
zZNq9yRVS=a7lF*Li|$Ll#`-T$N7o!m9*GQn+H31>PyQbFxIB+T9Zo`{fMwFUZjXW;
zhx2M)7kJzqW5-3^Uai%KVr>N+cmmw><T47_9=Zqn^_t3EJ<z3=e-k`x+$qR-5~+pz
z(Yq>2e9?lz$i5JHlsL}!0Cby`b5Qw;hr-c3lR<NRn)q-fKvnFjohYJImp2d$1MET(
z@U%;aMxvh3)-8_VmZZeQPO;!qoo2&E;L_v*fT@8C45kH5+yp!lk=)x}q+u_*tojyc
zU~Rn=Z1oauzJ%ZM8Ilbr9ScXD3S@Gu2}N$97}PoXbehC%u?!a<G;C2=ju*uIy`^jL
zTE4^E;{1X&B0`pQ?<&}~)au)$c>?TARK|HAT5boJkQ!K^2`{z`0o`v}O$aXH*T|*&
zER)si$jf^VG8o{g<AfD^`ArZ8za<e3oyO*ZWd^O?8*z5)mLnxR0r#r%@U+iJZxBM^
z8&d{q6Ww>fYDUm~y3IsLXU{F5wl!&FG?nfs95BS}oYFc{DyernY^1=_|5SA2O0YDG
zKlB=MMsmCrai6?sfYraLAI?lE4XmtwysW8u{>}&vwXOC4c_b8+%<`JW$E5BvK)FKz
zAoRn<*)1e5Izi`PaSEe9{mC^ydAC$~@97|1(oe#e!CKYZ$@`tU)xBmZXW&t^!$#MW
zyn?hq<rsR1s+7@aj;)}Z0;O@M$N}fCKZQ=JQko6Sc5^xagH4cFNR4#OZ^?JdXtHsi
z)6MpyWAh3UZN|89n^C`Z3>kZgZqsDf!^v_VV?fI;X9ogyT8@b<-e87B$U>7#I#M7^
zY5AJlNP@z<jyRMQJ%(6?UrK@pqCobX>mWgvqMc`MHxAX_bPjHC=8>P!AGRd`tc1<n
z1wHRIutn!+y~=}k5{cCWwA>LsSu_VkU_r_#@WRlZUL&}H$E|u;plEu9iU%OklJ+WG
zA$#{%Cqm)|cY=Lx^K_I<Km-J_HY>&SF7Vt5Ru1UHRsdLw<aO{`51`>fzdKlhsb7la
zJ;tGs&Mj?3qu*D5z~3OAfPmMFq7d!Dhdsh^AUbiu!2o(D6Gy>@Uy^F^rS;A{Px1qx
ziX#c=M!c`-Tc{1p-ZLu$M6yRv6-NKck7=%PuY{V;C`kG6#m;h!7bqqTqVEm_^Fi>E
ze+>P#7G)5_0mK|>I1`Ws4g*E!m(sPgw&&`~;Zo+Br)cM?fG?NJ?vsv{rA8L8uKp&4
z2e7o5Fk9BT1`t#f1xM=i)QsXkjCcj8c#KMCHa;$zj(`6YCT^@=QT{NIB<^cIzQI+v
z{99c9Hie~2Ww=~6j-sfj56aj9B#-xT32J8Cl`)*wf0zPL%hZ$mI_bf638+Kr$@yV&
z%@^G5>89s^5AK5h6(?7oX>0D0X`U9id`MxPp7S{MuNx;<8-NkSp(}_Myx~hfQa0)k
z1D_`U{b_yY$PN!S@|&<2Oye@x;0EjEHm5l%5sUjZ`@6j4J8u_MxMMZ=de`DcR3s7a
z$N3?b(|%qsy9YmWycw|6QIKCjsW%ryhI^)R*uIPd<FQ9hF){R>HHg`&X0CIYb+~4q
zo+$YHcK<8`^fjCe7XX+m7)VqR_itY)3yCuz;UAF<LX%nvSx70jgcOXn$FBm;J5BCy
zcFWN{3_=Dl$%WNVrk5K=f<JwJA0@D597XazwY_s@_5K-!Y{MEWf!qE06ne2nshWO@
zx)cLlGAq%oDHYb?>u-OfFWK7NwA>4;e2-7hNqPz8P}~<izmEVeJ!3~JzYxh-`T&|7
zjxN->)qqiNdr7}u)FO1b|0nYNl{^mv>LY*I=IPQ*tUH=ECC4*8q1j<tA1jq%qHI(P
z63x_yrpkh4!G}1G8CAV396_%?eP0|bDs8@=8A?i~#3$!3G%H{a4PRSleH4Evhgr1p
z>0?iP6PCkq6qKAJ;xPDFX+igU3+XU(tfNgW-_|Xga<;b7N&Hc?m2$1lvnLjwEnYv6
zr%Z!009DF%>-%C8C+fWKTiTW>cE|Z>{k-iC_v+Rgw)<)aXt_|Z0$g`briEe)q<%eW
z7OI6hUfAx%@&L;S)^231DIWniay#8&rNwljXhySK*7<Z$qT}dGt@cl{T=`VJ_hu`6
ziDb{hT_G!cc<@C?LK1`jWUEXxF}2PgGmfeiz(wizGoe<?mI7eRl6U^5dU_}BbyUn=
zNSb9~c!YyAy5Pflkk#lcJ{~yAGgq3Njl9YjA~r@srbBPSSo3#pOi~jVa!T9et4$`U
zTi~yUTfQe|bBQ4n!W}KDAPFNIoh8!ok=~X2f_00pL-FhBjjcZE369LG;bxAq{bF|H
z_mv?3Igfex{Rbe`zzt4L@?PASh?{k2bCn~a6EbyUYXk$X@nAe$dGolGzmsM9EIN^i
z(NrrD(G~iZ`rey^>3YSjxNoF6sy_=jEQ$gya!8!iHd?jVGR0b!Z<PCU`Q%II0U)ad
z_`BU|>E*CFV9aqZy)Pt!kTBEE6Neuy4&N@9er|pAI<@)z!&5SR)RIXpF)vu1PPHxa
z8{InNJ8p~q%x}TrOlOChOs~UfUcVtxNZ9oBn#dT2`C?x-9M0E_yxdxJ8ELs}X%Wl5
zJZSM=+}-<mDQp3|bPwT-eX~>g!3fF3KiRVsaI-UwnQSvD0ap+kXPJ(YaV^grlQHw5
zm$+yts}O2Q)qGOI^ba8|0a#A%0yCmqlp+MvbQp_tVpLL7`tDhM)jPZnj>4RBfeWM<
zL07TPv;@--Bl(uSs)22g&w~e((V(oz^JuGCFQuq%UsA2F1xi_ec{4b6jKmj!$gwHv
zul(IjGsmgIinvqJh}7KCh-!L*alY&H1LP-WMU#=N><c66E|Q7(ExwBJ6y65F3~Q$A
zsIcHWZZu0BBo4`AF`xwzn_>*D+4yXCYUE4|vZ*g{iu|{`MkbpRd1Wys%;@edWYTme
zUGgm{JBCYG@&<wDvUG4j%}Q0qahgs0WVFR|J=wY{z<1RgQa_#f3ocyJ1$+9JM~bpU
z+zDSE1lv4I3wZADP_?q=8!~sxrgtrDU?X8eb<AtDPLa28x%jq87W4Z&eU#<nHw&vv
zask|I&WNJQ4cYm(GYekz>|c!*D68G~^s8$Io>yO{%N82htn+8=_CNMHF26q)2`_PJ
zRME-uN+2vf?NUWJ-~C<{Q@mIUc3H|4N)V^*uC(bdLa&>ib1CHGMJQMEbc&Uemp`sS
z1Wb;559NGHf0yjViDO)Z<1KjMSrx<XC;ZxqUm=|PJhRFY&YqO8`U4^xtO#PsKhlPo
zWsZl7`Tn-*4FAX@q~%d#iD`-Xi42nyhV(P8D)$Qr($jU=svFxfmTW-!;bgsPFMpJ@
zz$LV2^y;_))zPfQN^+)(nfHTafk4R2CrIz4rXfL<rjteY?PzYppaD6h_gn#~@LL<W
zCg!I_?Kw@preX}zeaGDhMJ53w^+G~hg${C-pT}><`Z=ouv>g>dx$9o~>0Am>y*c9`
zdmu-CP!o<q%0Vnk11o^)dDX6gZD>XJqQs_Yxq-76OF$KB0jZ>2;6&!U28^SaH-S}_
z(m`%8TBH=n$GqTYxWs~^6S||2e-lW^gC0bWNy6I;UGR8cFz%d|4@j75Dqv~WfgC}s
z`|6{q!1yEGKhZK^qaK$2Y7|=J?a3jsFAr$ZcYOoNdR6ufvwo-s2Ep;dX)$vN==Pv;
z@EzEQLkGMasvUO?-v9e!^Q+~Y>gSU+18^^{gaD|p1s5dAsl1XCkn%e};7KnY>e620
z%^O&QW%C*cM^-?`RTc@!A-D#l9$7T6JLtQc(dV!vTseZZCkv&c2=^W~o5G`l>?@R-
z&wk54OI<OOnGPT_@W@4#i*<{ENK^uuxl`d!+2?S6xJGw7w9cfbjK=&0pPhg|g?uFE
zYu=ZD6oINZ$16&~!O1!YL*t9N@CgKOs6P;qBi-(nIT>v2rkVZF;afnPid%dRS&t6P
z|D}ndITI6(j&hJ7&)yJ>gh9OA=M&mM7&d&?p5-5Uf8D8N4Kyu`IgZh&WPV6gA`~Mc
z50(m3d3-s1DiwAt`}K|3<eBFhTxMi9+rG1pzk`iG*u$1VM<3k9<TlElzxs5r__DFm
zLSVy@wAR%|wn;YbTl0i?gPo+*4%m3uFu$yLkRXWWg{H+@3`{-eh<o!>T?e5#sABYD
zu!HmmVvvvZ(CiBO7_-k5NM!q{W!RxLVyVEd`Jl;c;Eyl6X6z3ySJ<mRt%e`<b0__k
zN&FQ7^^_aG69_!2O!P!96i}hq9E9lnl<|e#{;|`Kne?CZmnEs+DL|<9Yk90q_~YZR
znZb8^Tj7syNYYVNrKW&$>eS<j5a#`O-|Hk7@#D~C_>Bn-@2!PW+uTq*!y;wxxoKO^
zVeRfsHB$3$e*vBp&C;=rG%9Y}wBc-V@e=<2Pt3ot@y?lOY3~NiY#oIW{f!Yw6$C-u
zLeDCqG|Kd5-l*|l`qzSV8qIi<UJ#4<)LZPS8wIFw{dnto`z+K3O!cN3&g(y)!vg96
ze@3LD(m7)F&j%^hKdb0~s2t$Kv(nT1zW*v}DC~ePfYjGhMK9-F7zXmx4S`PGYO&dq
zfk`4VwFC48d3uW3hO!BR!HtK5;5{n_1{_plKUZ6-<xv9jM@(H92~+>+ckgu)7$ZUg
zLP?7NH+R=#Fm7DgF*Q+Rrw&FV7t?9+v_K7nit0bnLwX6r@HlI;<TNx=)@;I0I1K}Z
z#7G?TlH7K9BfbWxew9hjsc>ATY#w}>$^86?NXgNt08nd)ef|;X?yZtq#!-`)W&`+s
zN01X+ZULE35$?V0{w*6s1YttFrO^S8IxWKkpbXs}XjqI{&i%{`NjCLvraa`xwvN-x
zE2coR9UNq=1hwH6Wf!b5Ibv1Ge@nn9onhs8v_ET8POsNyN>ox~#E8XMFAh|<YrPth
z+O#s!Ie=N5kX|MA@sYw<fg&kH-cIn1vhsIebjn>Eg;|PD6zGlM7Wp?`tKKUwLbLJ+
zVCrN9BcP3;OltPwA(RndjZC9%W&$7#4wh~(-e5w-0-(=I$<7>bPQtojv@(V3joEy)
z8pU9=7*QJSiDg4xW+9jkxV>G^6JXkS)(d}Z2JV}L4{v<ih34$op>hACQ~I?3Jzh4$
zn$br$E3~#cCUHyR_gP}8+;eU?yQrkSiy*9F-&8j}Kkz=8Ga~uT>Uf&HD0y0u`O=W!
z`Nw|Cc{<x9*TBvRiIFc>%@2ih0`B%;@i=-U35kWbTU;T!d(TV8dHtsvxP29#(}TUF
z*z_xD{EX^6I1LX?4oBQDni;<k-z>hQ>H4dngP?;@fJDNp!qG}9K%y{@dq11;<%=Y+
z0xex*^L=!>k{%tWAS?k$AQSLrPfuP?=Z}X9Ex#Aerz}pss9*V_m;aStcYdL%l#eaO
zG!kt1w_(BRR<oQvdgrud+|45$sMwbsJ5oKaH>@yF;tax8Y-lciYxI6&xK>L1%d3Mx
zuZEDnx{WwLV)BliipP8qN541Fb*JR#Fd7}pP4L_YacD-a5nkY~eliF@G)9s049PzR
z<Y`3v0C=AWv_}EHTqu25{<G3V`cfc!xO7Pt+>S--7hQYzbAH8kZq!R)8ZpJ-k56s%
z@%3D#6DEk8xJSLD9%KM!qP1Imd;azV9)kkst!<3RNiVJ9W+PAsr5!|#$@n9J2R$O5
z0~>>VMi9tpC8<PJjUNXLfy@L^_<MeSdbB+;dVLBAC*T+zBM;>32YBuo<Ke_UARGZZ
zDKJ8L1RSRBKW`YbqyYI%`a2mC|5gA8^rW9AG6SQm{+VFY%bX1uZx2SxL7z^uo+vW2
zZ!{VF@be)~{DHHFP^$_qgH%ADi~;y+fj&pp7)5_=Nd&1N9N7<~u;p8FVMt(tS_s*6
z=(=ffK(N90^-A>y;AO0jP#H52i=*M2ONw~}S`Ix+kr}8ae}^;_q-2yB%w7`c*7$*;
zb`zK{)bfJ(;9i!DAT0lMEqc%%#3$MtXAM`OA7^m9e(8x@1BuGDZhZi{Hbn`NGFhwS
zW!D!uL4e3Z`mXx2`||1b*Es&EVEEzk2rw~3d}O`B0@W_G<5%o_@Pi~#O$NYQ5`VQa
zAlML`1huFlPD7_L!ze+6iO@Cgll8u?yaQl-9xp_QN&%QZn0Nj#U}f|B^Fx)EuH+ju
z({nSu>kPp+Fn9Jv{gNYj$tsl0Y6x%VaLN#tMd7tgZRR$uSK(_UAK*#|5=Ft%FVSmA
zTo_W|Y$2EXg10%GWdk|-CWfS0Z`_5}cr4=v#%SXXG_q^m{A@FLPGUmRHilI~Ug2T{
zYo7!M{U9&Cx;kDPayDX=rl~h<blbPScqu53WwYGcLlRqc9Og-~dkGkHDX4slyeIO7
z2M|xA9^*tS`d{XVCmXeNkVJ|h9q6v#0u~rMK<~}F#eX*OqGBA-YqBa30#2`Vtjlk0
z6kcV<&54LA;`&z#Kq(6VPo#`fon;XAp0yrq&`t}MBR{c%FN(JUpKSulu95@O8?df(
zZ#=be(y?A(P*lA}+Si}}SOvLL4CtZr{s&1MP*xOQD<CcB;Jr#eLV={1B<uNIZxp<Q
z&S9=mfff-*agbHv&7~*N;3lG7x=G$P=q17??XF-4VBwV@XhdAN3?kYC7ca*fE?-ZQ
zge|2p9+5@gD$;xK{{@!eg*4Z+`6^;TOp#kg&B*>`@58W{C-MP5?cm+_Ive4%v4A?`
zYwWMGzzuj_taLQbO757{0r9M4Fs68iTK2q~($;g@IlvRkg&E5W^bI}!<)p%!qd|0K
zhBR~p8=#E{vjWOZx;Ug0je9wfu=}uR{lBjKU#-yJ_PN#ChwXZL1=0RES;J$#Y$~@}
zy6Rza*r{8o3O_dL=`z&$MO(N>uc>m;K&0~b13_1H0(A57hG)a?I-T2Q#xHk7zrRzM
zhMG^M+~7rxeO7;M9EQev5BFj;yHV43Q`^H}grkDVAL;Q^h1CLR7Pk+cFIO{O_}#o|
zWxTV58^WI7&FlA$U_1nv!>+(a&s+~b^PiWq{ngkeE)7Wh8gO$y0iyR~Yrm~QB6kWF
zj(Ml(F9dXX3X7B?od(kA*W@X!sab!OzK@g0@nY^R*c>gYDoOGdyYBKXnjCM=2vQ5L
zzoO-A1BAUi{YL8J`0c8=iaP#u>As2<pmnY{!zrll3@_8+3!#9CCY9B9Q24akXSJJZ
z&e$T8HoFmUQWW|x)y&QP7&|+f)IIg!V2!yXKA$Cyp>SQKn*aqt6J^+2&eR5rs9_jo
z&J)N;v7w&=oS%;_7m=0Y@uVr(63up#Mj-t^G<{`QR9)Nl9tH+!1?g1jMp8mh>6UH~
z1VLK5gaMHbk(4g!?jF*c?vM@z=?<md#q%BCfAyGg)?WL%&V(;_X!TNF^Z+?mF?*iJ
z@p`TKfpo%CF8EViguu)?LF8MBfcE6{z9V<_c?qvhEPE=p{(Ce81L!u~)-lKv*A8||
zggri6zc;8Q->v*XAKC8LcZPoc1-6I?;ElDFSRx<EU=M5Of8>bk!0pa!q|0dUKw0QT
zV7rFrZN9jC4}nfuFWr8P<^4`Ixjp%?z1Oyl_f@LnxRN0GKt4-+(1!$0!}t2qbTP<>
zrsTM$1oRKYnZRz!eNStkKjZxmAXyty{9x+whv1*E>yA!<U|B#_@RG-?Iic&5@i_BL
zD)B(1qp3c=$PRZvd+Y<jz;+qa2Hg2_j<3kK2>U+(D6mH8k>j#gaKOLR3xcXpGj6k#
z%3?sH1ae?vkySIsJI>kGJ?pbi2u2#2{>1e!?<e^xg!mp}`^N8c(LXZDk_izwh-GZM
zo^td^z6gnUaa(KF5m8Du&vN#KU{=UZJ&UocnIf?+Ks>2t;`LV^UyKbL?{^M0*;f_v
zU33QUZPg61DHX)vZ0T2|>l<n_RAOaH67M@iOlUgpea-biTf(7MtcQIWY60y^(m)s&
z_xo9@exjAJu5@{tN!anDFA_vuDu@yCfvo~|f;GBt{4a_p-ekPTm!?f>nM@S(a-E4O
zXXO4*>8E%&9)gawe+4ct--LF2p(1{rGhiLzH;7?awDYhSNp`-1ZvE!K1>bw4^TSHz
z=ZGKw$eFn=j**x&_mUF`v_n1qMrR@&*CkCZO<M>`MJ`0(M5zzWxMdeqYcPMWkmC>2
zD1RcjCd0)io$K;j3bs*TIxQT_-|_v!L4iyqDc(l8&A7OBpqzMUk2;mt1D-~%%ZVyf
z2ojp>x*O@XdN&xVhpuXz3DA;p$P@LFiUy&d5MG6B#Jk_4{Y*F)(9jdHOdd$E8*HAL
zu7lYr7LfUThR9o|yOtf1(PZ^axd?0?(HRs%CkxX(>%7Sn$$B)oS94>V8c*B?`w~x=
zmq~HLZLc}CrtGI{O&#+D1Y8$|=l@%(ZvR@lN8=P$k!$|sFpAtytl$+Xp)X!IWkUb<
z`p4hKORbn|OR`aV<t6%Fj-qBM_O{al*?P`AM%E^4(x<$o(=M5!qU!I=3-+;^F1w5B
zt;UZ`NBPZ@)_%+l!-vHZiUx|FR3oYpATVcu?9yxRt4hcIa6`dqAn8%~c4^hpxT|{F
z|33S28v;S_*|w5*&v&wLceP~~?Up#2p8w{;?D;-?FZ4nfWLGQ3q&3=GYz#T<n%bL*
z-+DVzx_5ql2^s^OA|j~?4sIHK79GqL4hO&eK4|*ro6bLPt-LWpabN_es2viK3ZXNx
zhl`}sn-HLn-!}*hA}XIjVqxOi#`t~d8KM?SY@>&bQwESKo-wf)VZ(_y$pkJ>x4|$y
zAOk9ykYpPI(Sj;U2vJ1NLmrNTJ}7`f$6XUw#pFA@Q#}i-6mLcv6u@NIP>_X<d|Y%O
z9Y;UMv3uX%iXdSX>3PHb37K?1+$n}zRT&!}%Pa$d{5!_VOsgnwmsr}q)n&qH8XZ(q
zpS5Hn8xiN%OQsku$`GUhla@pTf5UBf^hE%s5YPVlX@|d66w=ALpQG+RSz;$v;sQz-
zgOOJb#cA<li1~L2BW`#c?5ueC1Iqi{_x9Yk{Y7bO+fg5zsV8^YPpp78EOw(9KNo@l
zZwFgyWp3B@)v!z1?y;cD_!rtMp<{=f{2*kbE{uKLL>}?=5t&+Q`&?bm|IY%zPUO4W
zSY9TvhWz%<DpU`;%!3VJ0d5hi;J)JTD6zA|UW@m2%7~|?)XAU>MQG^}9_`gVdc41%
z#K65s8~Eu7REsxDW)ky|2m`|(ixhVimxA5&FK(znn<xMG*Q7H>?Nqdpjx;-eOvJT+
zelr0Kxb3%^0-jsEb-jV#F@C2H6&DIm`Gg3j(77D~7PJV<FFW)<%a`{kxMGNMm>5vI
z%{vwrKUy(K3oO+PFVTVXXftLnC&?XYKo<Jti)Ja$9?XxIBik#vLevr0z%QU}L(qqg
z!wr8y?L`S$GxyvJ=y~-Gef@3uFk<fLQHUtM$^Ckr$32m$Ot?Jb;z)NA2&Yt&duZji
z6oZKoMpHd~{d&F3@<Mm7XvjMjQ7_bXIKAZ)GLziP*YahZDb|o{ZQnL*$E`RJ6kXWN
zpdE^$6JY5GYHst=W<4nvvmT`wp^!Dg^J{V4a2aTB@=dQKD$p!c`@LbD9iC0eHZDo|
zW%4_<rjHe)rss#J<|!lWN2^E2e@2?0Rc|cLJ?2_VPr;b09mzbO*jMX2QRhvSF>K%j
z_>zHl#kC0PXBE!Hg6y9rIb+Jr*XDuxuhM(rQh8Iw)E9-~M2hexw@Z<mT04d_LPBGP
zi$Un>`uyxt;~P@2b?wEp(Gj1?4^-IV(K9U-5iv_u;r+i4+LF&|4OA!#aK^#MR(%fw
z_l{?t)QEW%Jgo|P(5&>TtVJvSyp|s{;7&0rjghszP7{-m>B;)ofps=)QRw;gk$1-@
zMOOW(*4(=J7zIkfn<lLnQPW1r?P+yYR>$j4cWQ*bHo7RRiOF4lNa^uc*;@2EPGOwh
ztVYO(BTm@U=n@DiRUHQHi0iLDCxghQ_qy3OzMUL4Ko-!&2MW)>kqy~McI{Z1QVCR~
zLbEvokY3@uWubww{O|k@=wS~dJMUmY%%@MC{Q1~dA$X>7ICIhd-q<L{rv_ed#b1)k
zM*=AZpn@QBVJXYV#$O(-ZeakMS&BFi#+-$QlQ}=GEK<T;lFT*Z9#mu}*G#6b)&S)a
zPzkBdPy7(`uOJ$ds@34Ek|8p1)W?vphVo&nE>9qC?z!xHZ@W%WKAQ>9#3vqhtP}DO
z=nn%j%E3Mz4E24FVYQmbA)YjiWB*EtNN5?o)SCfW>bU>@P#F$QLHSRx{60aBQjeiH
z)Iu$N#W$6lM(4yf3O@xGs=jD+2}P!DD~Cm}ZX`L?RvUGu&4n&7zHo4mh^ZP94ogfm
z48^8I9I&9;Vz|R<j_3knjfi`~_trCc3fc-qA<-<`7|406ZXRU*t-_vA-9zzLi3l;&
z5DRb4RAfKw<{EuVNd+AZKfhEQfJd{~EXbl;f7>E|th>Qa<sO5c;X?J+`ANBo#xy!6
zjoa?5xjq&CW3j_7B2x+~R0?R@F_Uo2xIa%x=2m`N><@o2O66S{Pr4_FA0<ER3lS;M
z%{%;lL^BNEK{C*@c-cO<q>^<X7CR5dbmKl>es&*|Z{gn2$}H?52Nr%~yT!z6D#BsE
zu!pNkI5`MEg-Qe0?g4Z$b}T`sl(YSguI*+59LwNz#D})aZ?2JKCw0E^QoL%cn&11@
z^3*&#vmdfdv<HgRj^D5hkG6)Wj4tNUet#84N;ocD$m^guwYi($r@*U-ayV;P2v&)h
z`K~QR^7KML@7X~Qn<VbNc={S;vA(Sl-O#Z&a`{LEp4#7x9M4OObGubI$C;TRd81*y
zHRfxhKdrJEyH@=3yJ6@@o5M5KCQSdWd&lBYUX`e9%`Oc+%}Jd31bp8AXfSm)a=uPD
zN!sY;C`?i?|D>=SL4713+3ReH52EQa(f*a`6<{=--GBG69caOKOyWyh|3PaQ%^ZR;
zW71JWfGK#n#3ZApB{$?W?`Rdi#;kI>^>g?%2EWl&DZyVsQm6BCf%CA~*&3!k!<~Dz
zO0yv5<@IIWanlm}BAnyZc;36GKTRkON%S?Iy4pN96P~Db&<NZf(1T+Y!mgbYAX1Ct
z^+EY5e#o)?&!$%SX3=apT3AunV$N(1Xybf_ToF)%joX4=v929}H{9=BKu3cL(eBN~
z!S7oDxUHMsL{8x$tX`sfsQCSpCf+*p4tPItlZTc$X(x;>PJ-Pe56vqz6^gBN+8Bza
zhkoRfeS1ZS1DpA!S%K%pQ(LTwCOg;W>w9=(HB+96wWtY$O3X2CZ+<5jD}LTyE=Re!
z8keH8V2<TKhplQNWkwSKs8CSpNmFX-2p^vUf&+tTL7#ulha>Qwuh_DmD+(#jO$5;Y
zu&TkV)3Af_rR7|CsH{eX)1~0xd-9UcbF|-e=QjR|Q7CSY^sTzI-5SjfN_<Xo9HR3D
z|HKfUh{MqS4HKn9{{bi<2>t4`U>USd37RHf+xwC9pcX-W$V$!AnoPnmCPR59ee}N&
zrZic>CGV>3$K?Wdtm(i;-8BZuXoJ&6sIAAZFEWx%GKy3W@6AcNkfBT1ZlY);Y-4Qk
zx!-LJeSve|jp__u4wXaJprT18y7u;Z>r>6~(!}jYk*)mfd&9z7u&U>x2{TwaHL%qh
zc~8$qkMbW_l*Rm9A3GTv{%sZIGXF*j^Ek#LVFLCAiLwMHQmbOHWB<a|2T!C`gYc<y
z7=^JrTS8BbV=Q-G7Br91e6MUjEH&%+g!Q1zMW`AVDyKh?MH?2;yBpkM6yP4#T4SY@
z9@N#(nA6l&xC^g#`ePUWkWHt;b=>B1H%43E(Pm%cE)RNmJl>_4O0RXXRd~kx{mUb&
zV*FD2s^0|~hpj(04+u!q1LyX#!?nEWsPC}`5y~Y8gb92&8-HP!C*OBQEx}?F^`7M!
z#l0ULmwQdU1Q#sxrnfeQ<nBanqV`LB_H$EZQQ|Mx%i6B2JJ$yhKVhFK1(jC<iuYj5
zH%9!1$x{*^tmvrx&d--xFbT~qz8^H<*|srE%Z`>zQqV2hKbLL0*}6_y)~ZVt&K*y6
z>)RJxh}WbTqxEs7@OoY*q4Z72H}c2!+KT}B+DMsj**Y=N&<Y+W3s``&Il&KPg?g(u
z_WopEmG)s<dvn#hccZ>HUwd>)C9pTrn{)f>Q(iUat*2k_HoFMtPD$ACEqhypDqS~f
zOhGzk^rWdX!CEk`6y{Q&%d^L990EHGZZQD3B3g*VC*wxRxzt*fKjRFk-sH$t4@BBl
z=?hx421}H)l%UN6Lf!a5ELk$Uwt=IRBt$4eqEqj#g-qD>90gs<JQT&<u4p~#A;cvB
zR95Vz{rq*Q)Fd)&a;8!XhvdBhP6z|3Ep%*`gX`gd#SU}RtG0Y$f)OE&Lc53HFM=fa
z99Jc+9R@z8dmr`M;0j`~h;wlP^<W@<HM%TxA98C)<q?(_w8z!ZL+@eiv=RBJF=!V_
zt`#Oo0kLKA*)QM3lHn{&Y)yNwepeE<wc*b&g`+#!@;Ui$IfQndHGKFjU-4DQhNvKd
zi242^7Q`4!uSbcic=IK6%Y+Qy`IFgcRJAh85-8>2s`Q)!TD{W7EmoGWZ0=$Y=#3sm
zP!b5`6tm{$Fl^MIvJbHd2y*VogNh4c@+A(`^sLGD-28Cbzqf%JXIG4x(YGOSeiz8*
z@;QeQHNsd-N=}LwUU#sxS{J|{mBp_PUdg%_qd7Pz5ze%1PnCSlihbjbxCTtxtJM{y
zY#@z#w1R}FyrXyQMf>ydixEekFS}N{tifmlfTa!4$i~pKDd9%pw}DGF`}HjlXgr|q
z!Q;e9Q`E3t7*`RePSNn4z8mTs@E+PMNjH^V5|E?mdu41gx}P^_Gq268!Cyp*J1u&~
zx#H1E^e9_gR?DtT4Fdkta@Nt@t?!oqD)UAI@gSk4UQjdXfR*~@=5ZnS7EPaMgNqKg
z{TW6JJAc!qhsl(Fae9mC_gHyN=B8xt{1*cm@UJi7D^Ro32BWxN9UZQpV2Q<UPQ6xc
z$rqC)4(0<lUY~6(-|erZR^{6!EQ|f+D&jS5ZF4o_e&SrGCH-R*eu%AUl}Algs%_UK
z$sYIom^khr@ZRh@z3;nyqoCEr<wh6vnN9=U=jw$U_XSC_?G`ltT^%1y)Ryq-Z%ye}
z{Zu#X{XChQn>&KT^>(%tIdV9X=~3-gR#n5z#ukWWF*b9oNAk5V?#;`7_K&3L$~D#^
zv}NKqmLrKgm6(qOrw#HZDP&3TVt0}l<x2grMYCqPU1mA_eySoQ0yk3KXxgQ?XKYD8
zY;9$ga?;c8xBTc;#53<vtQZF-d~~Z4`kEQcA%X1CKYHJR>bsxJA>L#~0@DoBrOLBJ
zVSJFO%wr(Jd~Ko5^O%kIiU{^uL9T=6z(12f7-MZ4sP<k!#)~e~hJ4-^4h2HmZz?GC
z+avsvP3AEP-68e}6EcJ>s~P|#A`^-IBquGZ6JI>YM|&hkhb{FW4ZR5LfW_|%<$*Hv
zohwVV5dFUa-eG#bf13UtN`wX?17Ak5xAw<!!*swB=|lf{TjCr%IjrkM5E6O|n&J{J
zoHtXQ2l=P#xdvi7qyBcO<=&9X;6Wp8K$a6t;#Z^5-9r_l^T6-w4z!1!0sW-4{{X7>
zUk47?r(dWQMET0zLen&m%>e1^r&kyNEq9;Da4vpC==Gf9K?xp7L?kq$UXgog6;NQ~
zK!phDY5nx^O+Y{@ZTJ9ykHsQr#TTo~K!0D@AJZf7$yJq=Be8-%r9SsKdW&}Np}8a*
zgRPNz$#MDQV=+I>Gi55fP8ZDj`H*uWDr~h1SkVr(;54DYIR>c#oyFn!M>x(2`MZ<{
zWB63h1A}0K{&3teB>yBbNa82!C7x4}a|ToN@Fvq9|G;I8<)zYC=p{E5O*}q3JZCXa
zh4vgn?*j@Rx`hg8{Y+xMnGS}LV%PwYSI<=fHhEW3%tYa2{!_RGFGYHWd=~DK;9uR@
z%Mr5GZrLQngTF<-S-U=5S=VLCwMwK^NqV`gal1c*Sj7e7@YCxO@tclc^t~U(EZ_+G
z#H(y`Blw@7P*vDIg7&TO!R$lx$fqr384ndw@1RmEK42STMVhNKY)_G&tVNxz>BR}Y
z?to>~eVw4#YayDA%Uf$6g?iGK&_*1ZyN3nXbxeHi>+EU&w(2yW=5J3^$ZFudeVfBA
ztI)D{bK`ZjktsQuYL@`$qEQK>v=?y9b#4o~Rxyir^C#=czw`C@Z3!elFx=5a9>;yW
z`s{3D(v5Dz86tbkKX;mtdxG|9@`8)ZL}(aeD9Y~q*L$#<%}w8`LGeo)#{IRoyf`07
zBB+vvBz?B{1ywgX2{y)K7#v$y&>}{`e<074-!(PaTA$0|TQrM};dHJ-G|r#d#5U4v
z!v#<=#7%j#qm!2sPb#Yow5X_Q)AUh)89fleSS(s6&I)3(EJA7xO*0=L3U#(Ri1W}6
zI%=G9Ow>cLaCc|GbXLEfcnE5<=&sr)3$zJ~Du^wYME09sB^bJr%Z1xBr8xNa(tmy~
z#R@$tghopAdA-6gvl><6*>z3P>L0rt?$laAdf^oOtda@P+Ff_m9pND(CBM`39~u6i
z95g>Dg_=W^ZN%Y3<--sBDILFKAQrwaTPU7A0{QvYyq|n}OK<WzN9u>?^N?yoVPhLU
zZtE%b3KQrs8zT_u@)1CVrhf}1iYrVqtIe`Be%IjqsJ{WP(^oN)X4?>GP&IzZN{9xr
z_jof`Kr-W;nXKmbr(CRBr4d{<2e&upZJ)&R{5?v5etF%v$#ply0w-D7CG~2|X$H}#
zNXJ8&eUIDQR16{mH{r0GZsFxGldU*(Ya+inqkB7oSoC3?4uXsB*`%66tZ=`#1#Qr;
zI~8|ipTy(dk8r2fg*T(XGe43g71kxZ)65x6{tyXVW@e<80U+ql@evD(57)0O{tsV9
zWxFHUG}|1RK>aU6^zd1sD@^W2k`h=6vTF*3+$>Yw%NmZ2P!wxMXnwrmmoHmVO_C;d
zv|M4TzHRuSIhwiR(NMh~g301LC@;KI8pIs9l>IDD!=jzT=?&NMuu$}?h9$1dX4(WE
zhkC!hlbs^buJxR&SF}U99ZHnHa(<fUE0(o>NWoX~#B<*LCS9b}o5noyto$I&;pLIt
zJn82*%Sw~D%~;Pq!<Nl|yl)LnU)I(9aj{ckUul9AQ+>gE@XwXSSt?f%E<=;l5malQ
zvG?!v^Jbq2FR$ower$n~o3?evgynK(5Pn4v-{mWTkIZW;K~)7V)_A9(($hbGvYhl#
zn*Y_^y1!O{PXaeTCLw_<Rwbq|CF@!#45#>>FOc(r8SN*}kodX5b3;!nid2v9$8HSm
z(^y__O7!!Rd<V=W`{Rkn=Il3)jbXIPc!s{P`28jEgw{*WWYkuXlywHZHJ!)OPHc41
z(e?RKGM$mN+cM$9fd+;b(2yUFnjF1qPOLKVcheEg+`7+2Gx3BN6VUlj)^$;9<Ki}+
z&-5@MdNsjgR;WxPiuR2|6&^xaE{j^9nP>piDR^4lTYl&WMf)kSQXFvh!S1~xthcjY
zY;Jh?TNu|#4m@-(kEli6qa#xHUv`)gLxu&3b_{h#Y9gq<Cp>3p?3f6#IKM+YI$XJC
zH2LhEwWx#zcmEb$my%q8P3~KMC^~wJ!h<LMV(E#YnGl}$-`@$QF_yMC6(d3z{^ssr
z!Zh6-4Y%fUerQj*1IlKzfB_)z4<j2}7%J3&5Z$JS&`j(f$>3>Jo}IGTsmAY9*b}=t
zyM;epL@InqKVes~%}nm2A16H>XRY!)UN`H0-g>S+xvSe9?+;l9rJ55DaP`OY(Dfph
zlSqrZwc3$bqWk&i)A!F>H67AD&6%=o;;uitJl5SOE*e_>OF66<<8BqOW+Df?1v&3y
zhapo_>U;oN-zUX;xHqS8ndYme>(=*%Sjlxx7S$yEyvdEqt*z3gDLBAsDOc<~oKUxG
z_XuNEtuHpWSYS8*YO=H73fy^m{HNbY|1&#JI_{%4cTU<%P<!7!a{jyzSNL%O-$-uQ
zb<ZEK{ht9>ge_Hp`hq_Nc$DDowx>j8|Fzw)zP+E^R>Ld!az1bv?bykJxg53qxx`Rk
zmQA~G8^-GvK|z69u32>P(1VXaP%|r%H>k#X^87n%I{w|H$4pM4Q&Z>%OT}DUOv2$(
z%zN;pa&0j(23${a`^~(`EoNF7nXh=se?$+xeXFZ@PZdeyU9MPww!i8b&+_%<uz0$a
zmlb(oJNraKxZ|;f7|7C%gH0q=dwY=URWT;_L=I-CE6e?ckK78Z<)!1E?d=|Kw)+-M
zFI6MF$NiqiuOt`#4qbk#OITj^rd?J5R&U4Al-*;&3BhNayN!*)BhBV)FN8-^v7;%9
z&V;Sh9ln7ELd31}sUiQzEZFA&9<DSrL@qE|An%k{PTsI)N;d7e+sma}#U74H9f+<E
zM>QCG9$ytr=S>S59?J09{0DuEe<y@CrW{NEDB10*-VN6xh@fTwoN3F}^Fz~APT`u%
zeTFXLPdF0)y6!L94qDvUO%}$H(a6#5>2il4ZKqjH$fsh2DEx|C{G!UWRtU^+=!`gW
zW#FuoxX0#Cv8Osg=KBa(oy{UQ0rnJr11{G@X9m4XoCvkSpM@^d+mZRD;;5#>TDQlW
zY&H%C_A|@nUrl`B-pokaCoJM$j_v?0Mzg{~B`9=tAF%4YNm(MUJF=<5E>CZPRz0t8
zoz=j+4Q8?vw~L^wq$g`Ce^mt38B$&}h$e(<d{;);D@|pcPJnH&))pDJeKd8aI{-cz
z9B+=Z5aB0c(K^K$z8|407OeG206HWM%&0ZvJAoV;my<Dsi~<G7uBHo$dlWGy;2tdR
z3{);1R&*N?+W*Zo$=GRoA~Zcup2^#jk@vPE$;b1Thiib{_a(uQN6^^&>D;4$GPh`9
zSGf<QTsdeK=p=zllW`5+eq_0Dh0loDTBYA^N?rTke%*K<N<H!t<oQ$H8Dq}{6JwpR
zfiM5Q{xY$}|0{BPr2f6hoB$Q>sm1uBmCWjH;jEQNi>#n^ycl<zkC&H88>#JxIZ!uh
z;ZZqCvbq`i5ctO(B$~+=xjMJiAmqys2RA+=8B`yaXpE`(ETdnle~*o@e<E+P%ajt`
zvv^%?<Y*=0s+i4Irnq(uo$6imb0HQ!deL~*RAF8=QnSj?ruG@#9wfXF;=b5x0_?>8
zyL95Q?5K71^F1r)t%6y6*CK5&U-I2I<}*6_!(zdqiNwr;>Gl*?+&;vGn*%WAd%sux
z+a6kh-{m}?@v5#b&wn&xN&POGnF}TS^fzX`1J5Wi%aoAX_BKb_1H&M?@9eP+W!rT?
z@;Is<JDi96_wIbdco-{H9M6YevUARJb@q9WUVSbfJ*c&9bojz*Aqm+xqci5u!c2<N
z(dwq4kG2bqH99r*>kt*}!jy^79ewlieBVAH!l8w}T@2wiZG<gB2TAKoO(Lv?ujfSY
zmT%MsnOG(2+-kOC2o2%AmI1K0kuDDev=4+EJWt=d?w-kX?U>k=#gU%LHQScIZN8~j
z78>}p9VI@q1f!&KRV&8lZ=j~FP!=~qL9Rri(_ioqA{&5y=DQu?bXOGux;3mbSum78
zsjbmCn*%E8i%&QCIz=ea#9yR-q^a#N_FjY+Kn6B5gJ_--q!tL{NgP*OjlawSfkzfb
zLD9h(P~7fBz2jPj#;C6*%>DJ#S%`p9@9{Xah?kE~*tIGw4;f7gg%-;QC)^4DrYX4b
zf2j^tlg4$&w^^W@*9u9I+mpi17OuHeG#Gm=r$Q@dT{C>RM5VfDLNzLVkE&ulw#Z|D
z7U#0c4RksgdA!h-nQB;IF8N-!&mSAN8C-8k_vA_Uyrj<twFjKInLPcY9%_*}#;GXJ
zVO%=bNWDli?;+yIT22(tX337*Z`yc7XvUuC#ntixOs8FWNQ3}-^}ijoV#JRc&b6iA
z(ipwOVRfL&?p)SCK_UAVPcbF2-$-<KdPAp<R@B<Q`E?|#_LQqcv5VnsKPLBi%Q-%G
z+N+Wfru2b4qn5FJ5D8Ko;<EjHto+SIKd{Q-o~0N{@LD_s;RFS_k>t8Q#LkLx;}RcQ
zCs|%-zo_}Ujltc)#*Z|21}$%VzBk@qC-ER;N(xn3XjFa5&dyG-;KGyAiH&e@SDiFV
zi%VqMQQ&Fma948bR(7F_F%sqPfuy!a5GiNV&Qn1{&iE-6)ejY+w|`gMuE~7ww0Cz%
z`J+KtSTeOADXIpjSJ>!jXKH7`8bL@yl`&f%)h+fN*4+N>{QR=n<5*FZbe*VtTVC<M
zn_AiF`2Q%%Y3T1&^MA47<}Ng~L!a9K0^lD^%C|>&Ig}$oCrPrxF|!-=?Bq5FFVFfh
zqoBk&?lR<?pThwkluyW__zs@Jq_w%T{mo}0p!uM-AE1EA1P=W-SESA58sS1?ZXqnf
z5&14#%<CaL(vE^iO0R9V4y>aUIK5AlQ0@*@(`Cjg05QT!^*Q_5-rn76b|vrk7X;#b
z@`*q0|JVtWp<<10w$7hbE&lp1NDa<<-98>18e*uclX#E#zW(HA?(R)XFu=mn`J&Iq
zKRiAAZ}9^(&o(!gR<cl<gt?Ti6sT`L{Emf~?R9sBin?A+C(HJZlhw3RK*92K`tQ{l
z#yO|kU5(SzTb}5VTkXxe-LJW9Qb6S=TUpFp$t#6e#E=MAga%~f4E8r1a#V7PA?JKC
z;^6yaK|)F@D@Ms;!+5#K<$WSN?lQ&sKx5luSkDYvg4+)FjJi%{Wtn*r)n1B|DLhtH
zs}OM4@?Q{<VpB9^bsx;p6s2?jc>XDUk4Nv15U^!^JBB#TRB4+95!^p_G^cOhskQyf
z+m@rk(<u4&5_Cz*wRml?Qg)ZgTcZ37^cwk+D=ST{X|;>uJ6tv^%9_c9v$`x%i`X1I
z3ySlRvdKx790qYc`p}rqU5y&?e>~skCa{PY6QM;(X#b^P@1Mn)U49NMh`eBUhAREJ
zk8mo$?VP7{-|@8E2#x#rJVm;}uN(Iow-q6|<+?eQHgboO-D66~|6ltmBUaC+)Hy3E
z`e=w(+sK0fHekk77;&isbI+~qy3|v!WyN~qp~w%XBb~Q$+nY~H^FERz8p&;MVp!Sg
zDJ4j}1hGq)!;<CgzzZu6V9ea?Na4_hk$F3s9N=;_eCg-rs44G{7yqBY#S06+WFg0_
zUJ|dBK+-W##%=9i&+mS|^emHoU+3e$&Aby7oZvqGsnp|g5o51i{>DF9Vfo6b!p=Xa
z-GoeeB7=6{?xru<`lU{mrQ>FFJc5gSN%Z>iJmnlc=MbE)l<!AsS-XM;A^%Ca;2ow=
z4t;dfFTw@|fSLMATPJ$(oA-3L=f<(b@!R5wYL5*zUjIQRm9h16iaY0EElT;HG@cgp
zbxpHe*AvBV;?$=;vu`dAk23#6eQ9UWRK;@rU$&WU3#Z5nK8GpYvpJ!3Z$e{%a@k3Z
z|DcZ`k7banrWA1?T{x$+0dMOc7B717y&_*6Ry&Zdn51)3?4ERc3Dyiwb{dkaBK15z
zyP)f)gXNQdO?TI<2p*M}4Yqh$;c>Cv@BR5ayO78`7WhB`4xkti!V=2m-ex`R3=E*`
zZH>j-qZAnmTuOO+3%djnC3@xnE88KI4!iq@m?+A9kJsdg8}EzY<qH@;SqQEz>bIhr
zX)Py@R8sUKTKOiZ2Ai`0x*<!T9{xh|y70JruEm)>HcIIR(1nK~sf&zsqGQ74D|0~o
z`)}X(meT4$<Z-HF5BW~aDjtuRKyDCl*7ZEY3Kz6|MzdF|BD_ieel$xe*c0{nZ#yRK
zr%#w@^w7N?7}UHXIXV24$9-E5<o;AV)7AagF0Y~Z5DC>IAKW)v22G{G_4MT^q~QVZ
zN(a~Drt&m^m9}jp%zd%w;Df`^vkg_iz$lqyE@XaopeQqLd%oa(Jorh4Ba=ck0_i@x
zZ28+PN=zC!7BWDq*7!Gm>z9;!+~|H?;Jl3TPaCXp%N#MhqP1^1)%u#`-WB}du5CJ4
zq{4Pi;Lh<{0Ax%9HfNEx&kk!BMug%F&E^i6pTL&mpadXT5@=WOryn84H_^qYrCV=s
z$lA)QUFxvnbX+~<S4^?a)X(CN&Zl6Wf6V(Ut0l>HR%+e`3OJ7U<yu!MQH$$KS!&TS
zVC6m1qPfM+ohsF57rpp#(>tuT61m&{g6Gz#+)<4B-+`F9nE551R?lW|OfFJ7Lao3;
z$t69WOK?UyoGV{q0kjaVe8=08jg&L<Yl-igkWh|nhm3rG805+e1PiB6XjGW)%pLsO
z{#)D9@aM4?U7)Q6b-~@%qy_)omwRwi|If&{p_1E%To?(ChD7fuYQDh(AWMfB%u^-F
zFS*(bw7doNKk$`!#ROB3i9i2(Zn&d=_TQ7gC%bi?w$;-yk&_e--(iEiyNW723Yw(`
z&nkA@d2E1+H{720-O$51V$S<^+^_+W-?8e731^VxBV4<ECY3ZQ&ZbDNb_(v9AH+K7
zKp)c{?1?2oEN}@7$>wOg!p?srz;7>9V$|{t^meWU_c_O4aUx-&s1|I&DZ>64pzE|o
zE8%Rr`oJUww84mHau8YshEThnHUM^tU$kB!aJc5%^)VFP!OmKPNM`@x&~8KQSJbXb
zuB@ThRAFKjr{h2RJbc>y9=>@j-;4Cpwj0l{49{a)VHwlKy1Y)$t(IC_deQNhNew1C
zOB&;2%3yKVfoYxNeumz~)o{TR;M&TQZAT{JR7j_ROX<7F<yw^Zq75s)^wr;u^osYc
zWAR0eJ%^R;?e`+!_A#j|lD&Tjpf3k@eja$fCiUL?E$8=)_z0Dn4U!$A;ZA}(Aie!+
z#5)oo#mbU2gj3)p37l(uWu4{;hKz+u=1q?rfxpDt<4foUTN?)I!^{$`&&lhG!Z+sw
zX=2Ho4ZmVcS1yq6Bt(t%q`KFwm9+JC1-bSKij=Q!4E4$~mprFVt?=k;5Kn&xl(PJ0
zQ`KuhJaF4Z*_?U}J{9B2dziU6fgSWZw?bt+64w0vcaYr(h1dB=TF{R~tH*YU{o$BK
zUSBl$!pS+5GsC}j&;~o|<|?P5D1_F+`M&YI`u@ACD4jt8`|q{%2iPTc2h4qYLSdOr
ztuYyqC+&M<bs&Npa0S)8oQq@aHr}ZC*wXao)uX}mvX2SLJ#`m<H<+e<Xl{$X>u2yU
z8^M22CzuX7sts}I2fOa^<$o~e|4+Lwuk(Lwq6%E7oX<I5(O_BIKC#o^$hU<`v*g>B
zc3dM(3Gy-CL~+tf711xfl6sXg>hzyBf^V-j?gzM6u{?QI79rI)Zd{FcJmLp#;0!*l
zs{BL>GZ?GWn4RBNik}-VFRxNvE2JS{HTDT2e70A=rpPGCO|F+qZ~Id3^1H9dIBKTC
zj2h|IE{yxxD7=J7S*V>#n@s-)l@sopd0mEhGaQLA;BI`_VDWJ^gh}zU4tKYn`NhEs
zV?GRpCzxd#7C^Z{Y2Q4m^ja>iRwey8AJ{LzL07GCg|kQe9p38rI9qLH1AVqr6LM?|
z$JTJN&5>v@ji|j+g}H+j(@u{3hRJV#G&J<@Kv-qAt$xm5N3!LlcWGG3)H~scB52mc
z>oF*Af1#qT56>&6a(#bj!wa9mN;;@=(A;_oC)u=$xy49iB?PTg`IB6+%>DtGhF$N;
zn;=cKowu8gwv)?CZJK6uMgkWjnXkatTv<S4uh0^~g6w*oUa^<9k#5VdPci;kZ$2L4
zy&aLJdf2f-3n{l<tl~V7WsSi9gA0udyKIes4@4d#36ASZM<CANMrO1k13fexss(X1
zB9<G_m*bk5Os-u#|J>bT&X*h{=Q7QoGqxzQqRMx3J=n%-J*@&{W)Z0>zzB$HL8uYm
zjHq8+M|EB8ReuIqQGIf}PYc-L?wT=<S0Nf7FMO?*>My(LgPwb)V++{lR8#CNUIzyT
zVzJM+gyA<|#y{0e`l!ve<}BGe0V>xD<RLE4tJCRT%w>-G>d`Y(D}<5$X%|1?-tWU%
z!Zk3pUBO}nY{(k-f7KBdYKCzzuS{y=Xlw`!l-61&^g(Gewe9w)Tfm}3_!?w`<{cSj
zP$&|REDG3OC0!D}7H?G2gXc=Y0iF>(RjN6R@$t_}npfp!IPc4I-s^n*=}dkI5Ad+U
zUWG$>4NjBiXrdkSZpJhVK5oqMGS;u}zH$TN#Ae?>n@6r-c4W53T3d0eitXV*f_?Mh
z=cjUDx%KdG^#?v1i>Kal%S@UUn!e2tln+u!fmsN=k>bJd#%Ue^eL~GNF8kwv#bv5A
z@tvlbf^Jf0FpZe!K}hZN-yBux*O36q8B<pglOlXF2VC5Wf|D`H%UNijxC);wU+$Z0
zN4<O?NB+q=avRU0)&zF^`}TV2HV_hlmTR!d4uE~n9FN4e2qoZ<8R^0NE#ifBAR&s<
z>hYUpj3;K*S&cHyJ5JTM&R=iA@K$BMfKV0YPEYaNDnLK8%gY*zGb&1X&+Rd`B5`m1
z{=hS>b|FZ1imTlx@)e%z@<JF_>&)AY-H8IC1~J@>A2r?#n%iaa#vfP~Kl%6ij^)Vp
zb!<(Q^fRTD@!1~4bLf*)Da9t;gs7e}ee>WTpJFU_93j531Ml<W8?<^7+!hmto0TA1
z8_k4-1Q{T){|=1CpK>}I(YUZM@jq2^p+3T&ILHwXNSV0SqA`HD7@rOuoaeITb_>Ep
zBQ}#55>tSpC`W{~X`WH;*l{t4i!a%<X;_Hv2u}JWhu0==QM{Zk?`%0itmfizMqp5X
zzk9b*;^k;1?Cy0?&tVOuSn*pol?vHUVzbWI8x15ZG_d7r6|nYTp!FL8CsN`KHgT~r
z>AipFUzcd9yrWw9PTSU(%E^304yLRnF<<TCX(YHhoAn8|7>2)}>SioNpil}A`|znP
z_oKuM1vTJSguA?=aQkaX>q@&}aSF(fw9@j6FJ#1R+M|zpNuuDfnd?2bTGp&)zd6e)
zEAt-DJqPJ<JV+pJ36;qdPIUw4KrteobMRf6eYafxm7PtV@+RN=*kdtZ%*)$m6Qf8J
z;`u5^O;CA`=aNqA&7JOfBKz(r>4LUVy-w{1_}jIJlLexUZMZ}RJ-DUf{mn=aXUV|0
z#?p^zf=)RG(1@tbZ_{}0q=AJ3y4f%SX3C-`gvxM}e)Rh<3Au$jZN+A6SK>+i?g?g5
z=#M@HZSuF*k0tJ0Zj64?@1OYPspl!j0kXYNX*&jgIq)S4>>_@O?MGnft(Xh64+t38
zhrZ|}dA=gNCdxoFhPdl21}QXagFC?d%DS#z0}Ijg3%7isR=-d0mo<Z;$Li9{ZEVvA
zxoMGcfK;7$tK<w{lseZu4lyc<dFxPbW05suVp6Js#HwIibs%nja<V%&&F`)qWqq2V
zTGIA$ZrO5$qi#OShMCL+)D&_)+pQj#oRdsSdXkkiZCWFc*~03VWoVog(myox9Xwlv
zqA%I?Yr}z|$qht@$rSwpWWTvUL0Pxu<OR#-8Ra_tOotY-g$D$(o3@vS<(?Al^e9W2
zi>y$1vY5O`NYmuD+p!tcS$OhsiPmUqOpk<YS993S$oPk>jf1t<sr}t5e9IrLrGPgw
zX!uV_5|8!b*v-7PvgML(4ebHT0Euz!?J{aOdN_dgii>_=8wzM4hDjF3RDSc(&izJ6
zy;yC9?S-Z8c9@qQLx9r%Pr@fRmW7hUp$sUmO-qr*#cjx9TJQT?JRDkaf|k)8wdTe0
zZd!KP(iL-RAld;<uVGA8AjjJyD@}|a$mL!LzqCfZ!sD|t^)2(gv^^?mped9g&E;2J
z<@a;@Hf`jEiYFGIlF!#<3*^BxkGuR)QEHq}#(%#UDeR)Q2`mslU0@jhhJEp0e{7vi
z!{b!{pDN<sq+bgL#W$lM_hKZ-I0|DhC3+2DPR3d+Npgn@9Ft!eF9HT4-gE16>Q(3K
zloC9@@PlH$zCBoOR}6VxN0uEEJH&~*BMg)uI%;~<@hKF*h$U`$bM@DB4Zy@UPVCa1
z8YEE_{2#B}zxpU{O#q%xg;HRq*}L|w+8_)>QV=FBu>bt}j|m?!QA*vYi%C*v!q^7_
zpTe#|A9EWSCx-vE0{;-bL%GK1uiOwA=MC8yYs?05Ej;mus5t!G0Q277O?=6Td>_gn
zi=x$`EZNr1NHTFoG-Xkfu|;d0l1KfOoFt*9@xserHzrP;vJC?Xzm}ftQ1&HAp5OfW
zE}eguWt$??)CdE00jzp?3S2D{wv$qu1cyCs*j>i6ws(;##G5P^V6D8}`T5W5sF!J6
z_g$(K9D$nw1j`A_yQ{!b#qN3!R=ipmKyN~XX;zHQPBCvcd2YF0qf??}S@hoL$lBKb
z-IqDLrojoZw67Ze+fPE?-vA_bKj#+%>%b<A9da-6_rO~;NWtEFB<C=^16sR7Lzxv_
z7Brw$35JSBnN=TO^nV+Zd0WEZRc@xF4s+kNaVYzj-+#BP1{=UPWvB+7E1!~WF3+z9
zd8JOpXx-l9v65zcjyqW&LiP@i-RY(~RxdlAf95Ln_L#NLxK1nlnDt=!c9GF6Z$Kzr
zuZhz<b-cT%(D|8M4zjxXp@o>5{YJ5#c`#-^VM7@3Vp_TIU)nu37I7VT)3@3kSnX0I
zo_tbJzygNDe>4ZLjG59d1Rb7{xF{u>47cJO*fo5pC=w^}88ZjCOc8>6I%if9FbLVO
z=3Y-*m!=v-ilxcXFN4Lfll|zJ8l^;EBcB>}qQ)6T$tlHubyE`I{M<n0{55ALs&~JY
zcP6Z~MTv6FVr#Vq5u*n`bJJ0C<9J3vaNyS^3r{G9Sf$uZ1PoifpZS2_hY#o^8bMf?
z;`9l0j0lT6e0o1fX$mZzhfU6Z%{}Q1>s2N#z8NQ<tjy-EndQO!xEP{US@qMQj@b&m
zXI7e`JtUAn*c-}cL?Ls?(~T9xcoBa^^WG;8#wA8!oPe+K3IM@G+hH<v>%{fwb=VO3
z-P-WF933h!ZfnH~WtIQ6+z-<BIlsrun}uOfqnnn-p|4TdjI+Mx=?2;m?YQpO7r<Cd
zovRQpS!FRun2ZhKp}vfHvRbI3_n19@%_oI<nSm6roei!*xgNdbc6+1P0>mUIGiFhv
zpf({NM2(<p4g1;rtdzo%Z#(RvEHLknN9H{6nXRCVccr{D3eh*X{`j|z5w->7Czhp6
zI%eM$lV0B0;dEm!Pc}O#q1)&FKz>q8sUmh7Yu#hGEfHb?R3%CPW+$~3v0f-#`uq3q
z<EmEHmWTMY6LD<X-_Eu9&IqUQ=8M4MA6>|k<e7Uf?s~m@25SQIKn(x&&B?e5dMZ<>
z9~Q-Vl%_v{$?ngpAJt+dR(F7F-$1o^0c{FqtvXPm-o0JiYgLg&uLwo%U4ToYTjNhr
zYX05>+cx6qZ?pnJQ+8Mr5r5BO)s<Nm>8=&-;wv)wqGRHu$eH>ui$1hnh2sx+{BZGt
zjg?%?pv#MdwztDp&IyP=FHl_&pYbw0`C2)=S*YmtXPh9;aa5YDd5GHGrLr=yQX4-j
zmU(HM6=(VZRvE1Ppzzb6%yW%Bf@rl#MBPeYvH)vU&SXJe`1pgvX%G_Z8zq%iX*pGl
zwnQ<J+1WNc-l|EVG1|-Y<NFPVKJGVxL8B-j;Ki(w^qstW4092E2`ZUpQJQ59Y+7Vs
z3HeacP%)9fsSo&UE0fE!muGrD+@*%#Y_c<M^aop0I{(&##~>J$`sm#Q%fEW<^{$EG
z`HXZc%Lpr5OXMPFL~RS-kFPwUfd0sKgLB?H+6RE=uE&j_5S87an+)so00^lgQP@7i
z6iCseES^D4t5x!6Q0N29j*8l8fakRv4GJM=Dpm`m%J2R4t>u7#;&$}zzLE<31V*r6
z5SKts7zq)xWha6ZX&ZkB%7_MpqUrCSm1uNm5>Lzg-V035NuO$$|3N@Xkc4aoHsl)s
zQ>~4=BJ$`Ve;3@#JuOdpW4g+X6k>jaHM6ckN01W~W+69KqR)gOf`3eyvQs}s#;7sK
z1lH`(P!GY8Umy~R>!4=lUGc_O)3{CARx1gN<v?vTV(WtO9;BFl<1J*+IZt_iGEusD
zQVZvo<8guBlh@K#w9_msEIDW!-D8M008NG(zMLep1nFiPi(IN$5~U^wab3Nq+9gW=
zHCc@_7U;;l5mJ5F+lUi|4?-y(*=JB$KToFu%3bu&+cGpBU9VdTB|*Z~6VYSYtRvMd
zw1byiPU{0-LO)OofIupzR=+{S9G(D`O~Yb~mwGksbt$D;FI~j}f)<6AsQo^>Z?Ed1
z@d;6F+WXS8K4tJ{1P4=|Rpk#YbG3M$V&9M-8M;JSUY67sYJjsmXX!$mFJ~T*Z8uq^
zMOXI!Wbti?zO%hNo32(Mdheh*$JDKybIEx>zLZ^P-=acH>m5_ra{K@oOf+<(K=7Xh
zCd&+N|MjZ6xk2@V2Ap;E_D1n8Q#DrY%pE(iZXs$NVLpeQinF<A;@7dFCXWNs*|6(u
zUmFPxqKY(>UP?~nA99uNc9UMut!*xfeWcU!`@8&Hhy?A-&^0S3_DQ7dN=xv}L`kuR
z44@!C;8nlRZQWRAyb7Y_6Do)$c(8CyScufOd-mVr+Uh&?I_rd<w{%Boj#!!vds>86
zU`EgYerGR-XEw7i_=hBP#*gQ(2@5m?HLWue5ffAID&8_|2Zbt;B+nm452cEPO?C`&
z>Y%b7fW+6+?dh;d5TGQ3>qc}jB(i_dkNI~upt@rFyG^RVA^-jU3XC59@-jKX?Fx+L
zVH~dJ40NMNG9G^opSs#Q1=63Vj|8$vIrQEt?fpB46NswXaKiFwZ0wA^4>4e})GDxh
zGH4clc)-CtGm}ti-u7_|@42QgII_p$5*=Uw)--pbx&cvw-*)4keuDeu8}MZh$YN{<
z0dC$Z>2N5oF|S#th=kKXmNg@u524u@-9t9k)8C_7!%MRg@|_Y|0dg#=ENp{)OIV8W
z$G^G%CH~BOg(=Lw{4T~?E5eGe5D+wvY@LXp=-(h7KUQ}>+}iw+98D%r4_#+Zsyv=q
zhGGg6B~niaFjX|t2!S~|A-L>0zOd9|6VpJSyUD{8a)u_eeJs$b+UQxAjl)Pd@&Q(?
zMsmj|+*{Ph`U0HVpP{cSQn2NB^K*7OHrZA|toy{;29IVn-1h>Og>`>_B2o?x))c)M
zdF7CFaCCSVF^DO3LXG=#?6l*NR$K&4zN~qhl#{Pt%Vm`K+Yzk{qc-tu!#0?E!aCv1
zKd4ZK4KVeprE@Db2bSAYe`|pHym?e&O82T5*kmipJ8bo>ZgEt?9F@_=qb%P(-TXdT
zx=rs<!AGv7uO=Vp>)k~f_NdSBMhwX!ND^}=2dOaU+*UE+m{RlPAITqW%aD`9DboI7
zVIc1!MX6Qq!h6^+Zq8%nvdsbjhAaM&jP^D&$I+~pA7#o~<Be)-&HFXZ?j6}L7{ybU
z7Kd?$4k_doUd<Z>h!M~3E8&yJD9=QV+y%ZYrz5zRW_kseE_nBU*1Y&}XXkb=LUOeY
z<dGT=oYtz6=4uwjzXVF+$fc$&{-wZKXSH;GsHxge<dXlWs0{mOf0J*}mA4kMRN0!&
zf}*^w%2dnN8wasnLO<4%8auRIUR2sBj}9|^1!j))@OoixJD7wiv5l&};nO045^X>a
z4DM~6gTWe$GedI~T{NSwsPpS9IFPbJoj)tw|MJwL=tM0B$PQob%+`b<2x!=Qbm8{(
zD+uf8zm$7}0PQCvX8wJ?<eLX{&YJALfHq*vPA1zl%jjA~)Exn9By?~7z`tRp1SKYr
z_>aC}w|343>8zHv%xzHgA^;J>+1>ul4R|9EOay;RKP<Go=|p@%3gtkL?f@>H7s@X+
z#t_}_pSS;u+Rh^sJeJG)Nmls0n@v<qOe|;1{im^sGJ0|&{oRHAh9*jqo0hJPUQ5dm
z+{}xXw&j?#-qBjcI*<4}ah-Te;{xeD(0A@(Y1D$mObQWKo3575e+)cJT18qB#fq_o
zJok4`F~d&4>=Nta;CZdN{k5-FRqH|peJ2J`ouUR$d2Ybl{!xJ#v@aJvZTct_)g49}
z#v^1&kG1XzxiGw-b#!N?Xn({*+<yqx2wJ7lv|lCQ_p_XXG{PzvDtu)=Twq6Z`>;~i
zLeb>IqoeoyD#5Tr5+=0Ye0*<b!1Zy2Opfn%NnI?a&TikUd5~|{$#5w4TyG9>T)d?x
z_U})2H3-Aa|M$u3wQw=y*V#>KOr`l#xayAD5vP!qQ^>{^T*@tGf44<1d~t}`ccCMd
zVMrd49Ne9^+`x#CD4gF<P`!y#_$}NYpZ&CjlXT75WOEetMP}B1!)7r>lwAfRSzzYw
zO13UatCvB*-A5B0*LGm}Mgs0L<s)tiM)75=K-w}haS5ijt~pEjcC%hh?e6b(=DA?`
zld$$XAKjgvFKe5EjKmZH0UCqA#>Q<to26RaO|8F2!)5%j!De9QW^Ihc&?;mCT5M2G
zOX)Qt+-(x{bf#H*c%o-j#z0-fuWA54n|gOx&9F{7<4g}zWu-w|PwVGOFh+qqjZ_dK
z^?rJ=Ls~#U1opThbkc=uwf!+eFF*wh!&l_xv$5gX_ySvyzp5P2B(ur~CEUl$2h;Kv
zYpnv4;SY;Ta9GO}v8OgHf6l}PVoJ?Bk1I`pktHMAx6S`AezDOCdiYxiOSTdWuE9u7
z5I~WcCgv4-&TT+A2lL9Q?PA@DQQ+h;e52nVBx2T!$G=`Hi!rT02%SRguPXu-Jp4PM
zSaoC&>MHj`1r;-pa1zVX6K0Hq-AqwOT>e}BIuW<L40i3ONa;u_tEQ<&-6F!l9@%Oh
zx=l<vb`kwY@jfU~W@M<K-Mp-cT9l;2Fv!gP@kp0v-^hI>u)Bljw#e`G*rQ_8B-@Xu
zk4+nCUN&m?8_1_?F&G@_XQDB2$>`3OZ!Y(lireil=$@8;l^6IA$gJ;o=oNmOzEgV#
zPv@XubRwVM<xRH%^FkK(G7EZ;?fgBzRH6A_)pi9o#L0|l&knALLL^k~5|<M-Cw=K#
z&t|RTtj{|M#BFdgNZBr#?whRY+qY;cq9qgAd=H^3gzFH9o!o~>(6F00>;mbGL-g8y
zc~oh`It7ndVkr!YkaTeW)yG!$5Sh+Q;J6~m{~Q%hh4mk`O*K#i$ko61iAtEV<Lel_
zB!Y%>Pf3o^)&~I5A~R72IOo9cO394ty8wirusJhr7goI+wizk@6W+z`6V~@Ra2Z~x
zQ@xq}Pi#VR{?E=+)2Z0vokZ?OH0~Z*I<#n2UIx+y*wQBEjMT4J`BaPI#VDtzN;17*
zA%&kh9H2$vQ=*#9M#Ewxkuq~@%|#y@##QN3I^Mq*0b>P5VwYa++hNW(e+zbcOWo63
z@}Gz=EyFX;mft}^Y??=PZAK?b37P!SI~10SqHz~4rE1SF2hyLZ{Di18?=JWPM4AU<
z!a01^%{YoGse2Limhh#DNFlJp9!S7tsN>|%7v6~*t>l#Z-Wn{jDWtvAAkfGT19Q4O
zT;TgVjo*yU79s7JcnCAq9Lly)SF+_t^#wK90`<ef+w9{DE&R%-lmj_4UGW0D6@M8k
z{0pJrnM6$9T>bX%3}j51>VN?Mursz@|7+auy-g;Q$@5r=b=MYj6rtG+7Vl!K&*e_&
z5j{TT=vO8c62Y9;uo;(BC~m)TyY#-QN(S=Ih=xVz2zN(r{|0L2QJ@wfW&S?HBq1pS
zBw!Wg?$?l?yy*|L;C5lA1Qzwr&_#p>YBBFa&=oLlKbYrhq85aD7d~F@ntlby?e;w#
z|G%_!aJl98^LS*WCh-7;*2I$ivLzeOJ?ZTae=;^pQfdrVJHZ!E)_z3QDTO-ccZTHK
z;6jS?*eG%6T+SCzXK8b*gpXU~#X*zd2ZPxlFO6>O5CP=m@Srfpa&ezsBS=ZS8$Y`L
zg?Qb9t&NL><N`*rBs-rEc$d^6fcP8&3dqL$J>`zQEtJneodM<J9#<~B$u#j<6+!)h
zTyD<h?5CLf1_p96Uj=>sRJ6`-Kui3}SN72hn9u-}A=HCR6>7NwBTB0dt_>m8^Nu%<
zn9$0osD+&~K-;2-T&ZGqCG7tQ`|7ADytUn7fT2OUL!_h|qzx1)Nu?b?O1ev6Xh}gp
zK)MA<=^k1-r5gc}Zjibgf9I@wzO(N7?qd0eh08rV-uHQ*Kzf9E_HF3Veu112;fu&$
zL6CITFR|7Heh9Umfis6!&Qg;!bL)jYs@zg{?>S@XE^HcIyd2mU`KHOH5nb@6;#dL-
zQEcZn`BA-m9J`V5WVfi?>QmlpK5yxZEdf62P9m#SJHJ}8s(dCbX$_>ErAMJa=`pVl
zQ>I&)i@&v=GaMSsyf@IcZI}v7Y@^Ybg7Np??E_LIYo?fU((OdeN>F7fXof9N+mxB_
zWH&42@r8Oz=dqeoU#nVxYvq?hyAlEF?@Db4KH86Kamc@G^YiETt=>ehIGN{<Kn@Xm
z7muCe&*R>W$Sna|=PG_9_R5#m&JB?v^{~a|TJNjaOqacTcjkOjP)(lw+Y6vs+Qy;$
zo7$PzS4nvK7el&tC>wuJ3Ep_9sF@S0=aTRWBvOq*gqxPy0F3ix&FTHP%$AT!u^x7c
z(@67Sa%Ja&w}2b{{8Xy1P|0->I9^d&RFmIwqIT-ShJ)-sjw&k{i?fR>S8l$ogfWhS
zkk35z&GG!IL~<_n>}PcQ>y#~3cg55!8J-KmeV#EtXPp%;cd<TSB@p&5tHtD<2XKna
zqzrmGj$%;RCtxi!3J3%&FHy-vtsrf*b?s}U$ym5iaHt3HI`H=oYkH9^s)`nx445?U
zKB~~+!-b$V-Q7W=(-`rmkb9z2>D7LRp;QIUu@rbTUPr?{c&?Bk$U8f1MRe-BJLuuY
ztjiN;t!@~YuZc>Z=j&Bq<Y6ox0M;oEKKsd4?Ru)fE(fBm73;_Y&|WwmV#Tp*E@Sv!
zw2`h@HdLJBt3+rndec!K41Wh;U^}<m@AtCsfi4>CAuG|AX@n5!qUCemJSZX`nyhtr
z^lP|{O5=#y7!^YR^t`V~Os~EHeI4-`3c`W+kXCC~J~Y{FIfPMH;Lj;AWgh{C2ulfN
zM4QYz;(564En=0Uc|Dyn7sx7U<4)X|ZSt+pHLtcQ*6rTI79Va5mYnw&mqfScA2|YZ
ziG=W&A*`LElBPfMZx$o72khLD31OA;M3V-9Mf=Uf{HYY=rVJmx6PM6{f~J8I!i&!g
z;*=WpHcJK)I_>Om7E3^%DlYaus?b<&L{rcZk6Jhz#M!;%gkp>gNE|ATKE)87P=;q!
zSWT%t$ddr}jG0|HKcltotCEj<PcAL}eyTIMQR4$SBXC9FZ}6Q4coxfu6*{t?DA5o>
ze<>R!KQ+Mo?G9$NeJ)=69ppY3*N4&{kFf?h3p`;MTInxB8k@B`ZZ6LNN~BbJ0j|G$
zvPq6avk#}RtRS<eA0Z^OK)o_b-FmV;lfxuy2=}KIv7xE7U1*x5GKfVqCj%Y~i`)ZI
z&eR1I`CVY8E=SCCq2#Q>$W+b<38uHZ^VC_=&JG$}3c4tAS*#n_iuQ4Kj54F1&vQE@
zUhwy1Ok_Pf4^>Sr1i3OF@8oxYG7J1ti{0s4#&M*0iMp?rviqa6Rg31}%bk^h6X75t
z^Ntnu(KBZm$R^kL<LB_r#X7|lG0jkJGT7Q{M4=(|*Ll@NDwZC@4x@+(>A!~_%nQne
zH7W8>nAB_B?`_xo=I^=LB$#n%2o*e|HJd-kF|xmJG4G%6ur=vgZ}7z8M?skNuy2k6
zDu7Wh4HW0rB|XPvWSoo7+DRsPZ!Mr??{84<TO<_!D1+%#vNxq2wc?ytysMRlS1E8m
z2Uzv-ddm72)eWcRkdBMrTF-^C2<QYSJ=Ic`u$R{(G{jD{b~mevG{AVu9FN3eu}@%o
zJrNz}G7+4e?B-iH)%GW-faVyrVBD4ct2FRs{>RogG^>;9Sy!IFO3^pabWo)%A%nRf
z{-fLA{WeF_X0gOVD&dxHrZGTaWl$TQ#|k<Cx!f|I#&>$#<+|UfsCOUrMD^eW@Cn3!
zJ;-|AhpSR$y`seCuaBh0<rZ)oorPB~!Tg=^Of3Z5{D*{j${|z2)<9uYb)E!98>OD|
z{tt~{wsN0M0Q%m7AT$Q-CRjO)vhNM+r<SB7m_MvT6c#mE73Rxp)O4`gJG16UbPjy5
z!%PmJX2)!$=70*;wcUlcSF7E5Z!J%24dl$-<BoZlW|X+aK2fC@o`k$&@)|GD?1e|g
zwrnZD{AojbjP^daP`AKrPpBw-)C}goi@2%34mU;^sA?DRG^lc}=~upuV-KT<^juCb
zjwI}TRDTXfB=CqviKOGfUbFYgJitBR<6SS~F!XPM!~AD)=l6u?JvO=gDF^$2f1@ui
zY>5Il9jjRQZd2ASZsxab@ju8#nitc(;pVyAs2<T$l$>t?IM4}~iG3i<$HJRSaR=_S
zGdl&ZKCRj|;m!YA#qU{XKWpDL^{E?9=|2^U=Tc<N8#M<n{lI56$?HF75q6$1H@et%
z*N8^oN!{Bt@(}wE3UyB}Qm~JB;s`N~gGAItK%R#Pg@y%L*nY>>ht7}83b-zo#SO10
zIgK?+4D>naEsKc_ktqq%LsFZpsQR%T1;w;I@WnFz*M*FLS;&48kyu3gsvGWAhWph3
z+Ds|4QL|-EWK`60MfzppPsl~eL!GsT-gF-0<KI6zZwU$ueYp7*J}Sxp>&1f!-%sk2
zVOjy2-s++x-&z>z6FjLgO%41`5i5QefLs|mODgBiB7|OIKTBLft4fKD*WPP0!J{tG
znV>|{3fpoL@gfbjEYokmcNf@GL@r8?3j6LXc%AB_RuOKwPTupa7=ID&DQ>u!%Fol|
z=hd<(YYhHj7`+F7SZhO2S-IQ3k$>W11PqVry$M|J@F5Z34rmxEEMKb9j0;$%0l^T#
zqPBla-I|y0dMM~esQPR;*E$YyQ!NWuk)N<we~HkyV|7#Oy#xZ~2#}9Lt)8pMXE(S^
z*FHUnc<eu=69I6ZsWmhBw-)4vof6;5xm|=R$WEC#*rP$4@g?A(xFs62V954AK-=~&
zPDmLSmZs<CPtc!I^}JLJsN~by9dQkphLw7Z6xt1bP57E&Hi-=?_YjmBys9ps!?vAv
zD&5W&hPA@YR%*v1_F7*16If4u3UjA8gzPv=5HH3|nMVNXr+i=VIiaIdy1wrViR$*p
zfGye`RfRUq<ORrH!9Au%JME4Qu{?Lifu!%y0IT5;dDUyxy>=CTE-WrSBP@-&9(TNn
zd8=akIfNXb#-L96>sIydq`R}F1IKti5tY8N#Hy{TmuIiy`V!BFQdjxG_~ghp)n5zU
zT@>e=;~qL`+?=&Sw>hsRffpUkpSW(k&C+lsXNBIuf4SYP(@nbflXedUsx5SyrS;=c
z25-;M^X@hL4&Iu1WiDu$+eEK%8j`fft&hLw9yximUe@*K*vxvy4}Bn2;cBS5__1ZW
z;9GcScteys_nB1VfxPX!YwXjx-%^%%Orfx*KkOeCws(Je&fQ3liF>-(WPW*6{n(O$
z5igvUdOkO>KQ)P(I!P?^=_YU!1(q764NMVk=!>$!W%o%Y8L&^g`z)I_Fc)R5#DXki
zX537zmY)yLLm+OrNxRBDFgX->j2%Qc7F~u)SUB<SGfryi4VO%m2A65oE@mPGc@Y}h
zbob;VDyt*ZBU)}rgv(@<4=67>Dj?4Th)W<1UoaqqqzDWMH>(U~AIg$U!_rY^Xx9}s
z@7ql5^2v}1z9D)<o}D?817%MhofCVEvYS<g1|lvyG7iSBNEP3`*(!8n!bNvPLrz0C
z_F~7e8H;Jz3pMGf#%dIXauRYH7%#<+;g5&t0pf(FH#j&yv5MoTgn)M734{k>*>Xj2
zN{gUv4$xMROe40VANVZ9VMPv6!_Etnkr;pGVn!z{xyERkBM1!3dF-vlfXirG)mK3j
zH<WNZ*C=i~yxDjB_pF9nMB-Y@40sRUPFhlZA&0AqBi$Kjk-mN0mZjCGWmYt34vc*u
zFEL6fecmoDp>23w4yD0dK;~35mZZ2?ZaGQF(v;V+l=ufpP74ag<o)O|a&CfDT@mYP
z4UJcsUMMw_#RSUuIHKFcGC$ga!F4fM%;jQVQ^HsoW1rAdY$akZno#4it#erVN$7P0
zqyaKlkZNCO-EM12Th*8i84X5l1o#UUwS-%#8Nz->`Ip@7W=?pU4(}2+R+y-g-vgya
z{dx#S#3S!`N}T(X<ujAhn|dbHCrXKn7L}fKO}bcbX1%(UQpa;}6uj|(a*C>SvSxn)
z<N9Mp{0z^77Y}w$XI(w{oifI%_p2AMC-L#oV^AkI<E`d&%#)oy{z;9RUKcF1oIt&l
zpr=l68Au7VA`bq|U1JS-{MTISeF?0}>i*Al!{bJs@-^=Te6IFa#C0yfw=o}uMoes#
zq^5L_z9CG^w&{Vp)#=9vP^XfHAsKJ)yTpC8chYb`o|O2G=ucVaI@}^(q9R|lO;7Y<
zy%g;7SurIl=Q?as_JraY3_L0lXKi1oviP>3k&ANfe6_1M`e|dI+~#*l$pz910MHo?
z*``zN$0P<}+E^@|B>Q$7QPs?ksT~MGM1Hc9ua3{R6^YCaOg{&<B=1{@)@sxuxWdiC
zHJj0Fs@zAl3h}wKm)3-m;rIy4*QQRm@nf12s1!=^6?TCsUs@S#b1dIfoQHB0Gr+Rl
z$1eeDA4MS}gFAavs3<iFuj4g)dGvm|2ji(C`S5U^Qs+B@_<lPcJ)*#=&G)`)k#(@@
z)pvK=g}G*JjHOy63qQX|F4n|NxCc9!94DvQ$crG<vu$WP{rw{h@cLG2Jfx3NU#<6j
zQ|b%YsPS^aqJ%%!y5O!^gx0??rRhu`b*JB40YeAt-hg^{sXH>b5+duOu*tw|vpdp!
zqi^KgEb!p)iVaw5i*IUvI<T2+Hz1n~jA|sfTWH-^p!9ALsXEuEsPv9^P}mw+qK8bW
zhM=^o4qEnovY4mti?!9wi;)X=J`$QfZUCm}M?fz<GmWP_ZLH*P*%{nMBT0b1-XYnE
zVfyQ5%nN+{$Uou&9HSuL8SB(36Fc>ED_}}xJ5#@-MaoE6i%QyCVrl9J#{=LH86v5p
zlI_Aj&&3<&I@WJ!D=cU>K-22{g2`gGrtE<7*;})k$|M5vq*EOFqquL+NP^r$35p5J
zd5BXzD{kvLrfQfP$E%#4mDd?#>?c*K17puCvUbs9>yLC_i1)Fgm`~unQf;x74qMHQ
z<$7k~;dWYh>y`w((LZUCLwm2Ec@JjyrQefpfNdE#=Q2gF&-vkP#!avnMV7lt^+25h
zc&J2|-)p>`6?GcAI&F3baqrhPaV0Og0?D}ha&dSnTpf3x-gO(RIUI2EewNlTc$jhJ
zsQ!t4`-`f3`-hV%e$((P@6>;hrWZqC_|g)AIu$Cr&#kAQTuQYxOUk_v%fY#&Hn5E#
z6?j%hCSH@1T%8;$Bnevpox;fl%)9E@M2^s%eAvi{8pTZ1^M(#n4&$yhptIkvU9h6g
zy;lh%l0$v_Rlt{uGAtK<m@t<oHZ5Vg#&5KiuO;}HlQ=1wr(w?zE7!zDkS`)eCbo?;
zibbaXJ_Y)5S(2(zl`*R(+bCRpC+mY$CUB#OPN3~obUm3ei$Rl50_uM(*h{k9d-4IW
z(tK<<YB@3xJ!Z!F3{2;lAsY}=xNC)}8pNMj1>iuRR~CvTN>*+yfm_$=nIP&QM*Mjw
zJ6X^LnheDB4U?%a2iZE9-<E>2*2AOZ6Y{6YMYnq9sIUE5pUjs&7@}kXnvM^lhFGCk
zk>^9+qeap6KruzAHb(P4V9ncMyH$^O6p!E9`Z}RAR(^{9)tu-Ro8T3Fo5EAfp*-`U
zjTzWgGng7MoB9cq8!DnTVZ@{IV#_osqn)5bmX!mvx09JT_$=&7An-O$zsNs7tG$7h
zn~5JC20jhNk4Y$-#r|?fdtXfCpO7bjN>GC(7rv%~g73?_$!*x+vNc)u#pXxDPm%-!
z4Hc{#zkMvb#lA!VEJurMKj+KbBpd!N#Aq9=W+%bePAf2x`(DxK$s|~Rig@*8h6Igr
z6vSGz9je<M0#SK?1lGAM=I31th*4(?(WJ1inSaz1P$C1D;;MXrvv4~Gv6N`!6O}!~
z@b;}>GAiZVQZvQN1B~`bJ__*)W}jfVM7ktk%aK{@X*Rh9pw<+Kv3US|*AxHkQ~I*&
z709DK3G(L)1i;Cu4q&xPh`JtcS2gaPa{z;ONPuwnRJrWi*a8Df#UU@IbK{hCkl=K;
z;julV^0-ZdE3*9V{HrGg<ubZ^mx#x0%*Ts5>1lt|Ez>J37#h7F7hGOi#_AN+Sx+1S
zR7V^`UtT=;Bx{y=m|j`nz`<hT%O=|(j6^Tjj9$LTF2TBf_6ePvF*~7u5Oep|P~yl3
z2_4{MmZODwe*SyF)2;Kleh3C1q<GYq#^kvLUZe_;DzuNtEHq_cY@Y10{+>F#L+o@0
zHn_^&z<Do`uw2FN?B6H7PqkK^w-S1)OoZngU$E@bKE1ZGc?G&CbV2*f{)oQBdZ*_(
zF53;L%rlkG{!YXuEz=#>HD=HN^WR5$#&kUT)@8<i6e14KQqB6gUaBQh@b(AEXQ@US
zVv_dG@%PRb&gVL)K&I-b$Zov5MUb}ipz3wAp7+(#pnzXx#4U4&AE>8(D+bX$8SVsx
z#bNOI$bq&jmCK(%co?Vvh-H<cP0$_5gqusJ@IKwoOFEx~poYzW3_f|duT!rw8Us^(
zjW7xvzU$1s6*@mVhFVHQeF>MtBeM!uDAd@(j*E!hXnB5oN>qS?*a1+I88mFt>)VO}
zlw_#NS~sT58WjI%h^)9;C3IzIZ&DhR0qHm>nO8`Jpv!|(T7C4F+y$~uCdeZ5UPJ+7
zHaZU58Vaw>Q&f4CAB16cq+z<C%+G?VPgRl@)<UcdQ!yB_L4EJi9oF%%<d*9S=&UJV
zG`mWf#G?i08%Oztde#vF;l+eJ=){$I&#Z4ZheTqO;*8tpLDNqFRm_BcbOnhG5op=u
zSR*54_`dNDYDc9dGTdaxaGB=wNjoL^gq3m>1ru|TR7M?4#H+U-NgScxp~KgOTcIQb
zRrcm1wn>EhLBERyw-fQRcUz5n_1STF&UHbl^lLIUGIxk|jj?p+WxiUD*2^DMAyuvT
zqiTAnkbEP2Sj;`EiNyz*ZavG*cDtw*5yw9PdhG1Qi)M)}gMdUsoi;U)bhv9OyspqY
zW+}vk1mZpHdukAwM)ijm5Swf8YW20Tvk#ftxtHE7r9do4fh7+)&1&~4#F2e%-ErUn
zYr7QgE3DKqyJi$NociZtRuskIi*ArO+op3<Xq!-Ko1IE~giKfI47@o*(#A?!<k&5d
zEZ6{$ty#z@!`fc?h5}%vmN@unOZpCsBGOEcMsTji{vP<xc@?L{qgrEL2V5agNl-l1
z#BfQH=@z0f4h42NXxJRe=e2r6k8K-_k-n^u-Zt7+`#P5u0FV)fpMs!EBK*S#q`$}q
zH37+xf1QcuVC-)O$bn##_mg`C5*v8(MfxIE>kiF-{9N#B{i^iGd`~2Am{xC&pG``P
zqhXNu<!v_(WX%SqX=R0a=f&PvFyW!sIuf7efuT|mG;~7@0%K`;pAcsGE0n{wWTFje
z_L#$`v9EQ^W<FK?8Q6x-_&L4l6S7;4v#^>7dRcCj*fQvMl4|_OQI>q9*+&fc83b96
z<4RGm3d*ZW3>>|7)}#{HGuan-DlBCETcRXW&cseR^5NCR`nL%g6C!VJW+@7)2g3n%
zZ)me_xzHi70m*n<d206y>wXsVEIQg8p{P=-nu{I(3X`yYKbovhhr)MuQDRrUE+(g_
z`QFxXzqCZYL~RSsQ@=PlS@Bqv_dn%7wCWuO=RHRo^?rl>=1os5f@-18`774g<Tn!>
zo^jnTMjfFnU%)_v*NnM!ZG+{c10}v#`uZ$u>S=}JV5;QWQ+nZ2DxtUSF-D$Z058->
zg>7+T3EwTt50Wl*TXvi%)@ONa=qxpYTS_eho0Rzw)G|_xCszFCEqt!dd5y(&?`*ok
zz5GEgU)3Tv1@#tW#?ZgILIF$`ySHGA)JFmCI6jayL<hFdGg@bjVKe^UTQ}FcH&50^
zz=R<6meRBk1VOq&0t@x)$H(~AX8io1j{1<ZJ3jF5=re8=aC+7hXenl4V9hItaEmjB
z48bNbQc9H9x3*ajr<fFSujLTy2rd2ya^V-!1Z)HP42B0tc}H+kGWE5j?H7!$p6J3~
zFA~62S&cD~KH>EU2rvHl`Wy0u40v%+HR`Nc!a8p*>YtWMN9@$_IBvULe<kwt4se3W
zr{JOt=odJfq&w1un#xlzQz}S>>F^y46*0DWAWkTEo}`?i+R-nTES4>0TcvcSz<A@4
zFpMwMD<nSkuDdJ<bv{_>iIGi<Au&%HC3He^FQlv#$U8IbgU?ZsdWEtS<lZ=I0)<L7
z!6H_Z@ZuLWJWxww8E=TopQ$SAEq@Z+ZFAyr<qE=6qBXKYj8d-sVAL>(n5X+-!O;h2
zYGL%$CEiba)gW=RoEGJR>uhv;NkOyGJBWaS7co_wOY)V-lG@5ETwVGn5Us4;9SOuC
zC!3NV5<=twL?TU8&$m`!+^-TfSO<l-bK2Lw)OJDYaEfvTAFf$?Y}~H^eACg>JxZ-n
zuV7Zxn}!N`UIG1RCCmw@c2~R$%hQj{d^g`;pgR_OpDt>A6-ybHIk04~?U$bxT4raT
zZIueum+K7azyI|yYmeAh;+%=AQ*HJdO#k)G2KuHf5J%$BzNouDydOK)q;5y3>NS`s
z9C-4<Q-<^2;gO5Re)n@DJwe~vTG(JlgJmJJg-}#n+~Znoy`NKNdHjYUH-j*nICr~w
znrY7&%m>t2R0Ujm%J;Rk9UAbJMDd(99HQnuOH`~T=3LEq&wVAS;}eGck;j`j=lULZ
z>4&(oLGZ=d`4F#fIneo)>Zd*edKF%*M^w`HvUxG5(%ko_Egcgt+pZaoqwa1fVyB}t
z(BY!!2i&Sd+1Pl`<~mWlpK6i=(X#O1{gZTNXPdvrd9fPieO-(AEI6g$ZjU64q>AfM
z%rUv#`BwbG5=o9%sIU9|CjrJ(ePf*%&9!bRL)dB67Qy!~3PXG#nB)~I6;jusq-ZDc
zmZ~r=Gh&hcs>FZ=Roku&MWqu3R!R*zPqrc|5dFfV7-f)t^V;XC^7~u*_Y=U2>fc}9
zx0+N`77F*0)gnDW@mG9lE<K&TIRq;3+b_0rAh@g#K9Ta^p-nziev+bOEW8j-{8{a7
zPEYK|=;wi+E~-o8T(K8dMEoaPE*hq_mNw}J{y}2$n@rKsY0t#Dw`2m7OyCYV-J|zG
zh61Z>t-OHcODTFrENA!Kf;%aqHdK@4UGU{-)Sf>)wM-F>txQsTLhCtL6*q}p?cUkr
z1&=k&c<umBh0)u^(HA%gT5D0?>|<C=Mv%T9^NCWDX&$F9(|2!%>C!|@c!*8TkfqWH
zqgB~a!c<E*a3g1N-Xq}LamSE%_w|R-W_HK?P^_<BNFAW_7Tjx!B_N)=Vj4lpl5ZQA
z6r5nZH5yJ47?3m0OgPxWQ!F_w8!F)z;1lGvP?PekF0FR|Ai*E_iL2N20MCnusZ(>b
zNR9;W*9ttuf!Rs?b`Ai@8MQD_89#(CR?2^%lN5BDC?L72AG3hqwzVviXBpb1{G&bo
z`+*@eQEyeRX<rB=)d*f1{Iyv4=VLLjV5n?0)Y#v7_GJVFO;qYQPjrx7UgOXvOU5|l
zDra8tMCPFi<PK-NkaohI?{6xN&s)@YS{%A4k6WAzf0@^b))3~Ar^Eq=$!|vy)X6>7
zZl1rMruBAnutJa(-YDc>SHPaJ0hX1~+jRYU!4tWx`L_*!e(QY}`{5I%$Y_DXiA2Si
zMx`8@@1pA+Tbz4RXQx`l%-Xf+ERrVXD)rq!7N1{B5f2jajXg!Olj_&}-1VJS%MyzV
z_Re@2@)DuNL>L+8_b?o{>X$_%9*6x7R1KR`abxJr1uz(%d>*&{*dWaO)V7&Y(BTaY
zTPd-oMV-D5>Ww+_dSLkN$iKE8kYYba1v#U}<cJI&e$bQrQpa%;(m@UR>I;Md<U;sX
zV_;On`KRdbK;yRg1!D~bQ~=Rp<^KVga~>G32L^0HZUr5GzkJPwqJC7NYNqU?#MS;%
z%OvqZgsxr-1hCJa!z+|{KG8#_ljC2rk&#_yZOQB=c>DKuoB|DI;@mJD+S~ptAJ$0N
zcyVR3%kl4X$JbB3M*AIa#>xWOkD3q`t^IBzWzPvI8PV}>ngZ}>QNFX=83A+N_IQ7?
z`>No3c+H}pVBMvIXH>cumY6W1AVP?)a10fLZuztt<LvX=NP~Oe(aly0Ml4Ckc-L;9
zl|dB4b0)JVvM4CB=no&N-g%hG_z}H>6Lv(I>i7Qa9s6^B2_z(wEumYE7ZUmftsC<_
zlWRPl!019_zsP1l!A|O|q~m6?q`1Dm{?fz`%^&(Vb+Uskz$7(!5J9=4P~&dCFps~U
z#`7WY&w3okK1;bVR1rCbHfw`U=fNhu=C0VXb9HrEzT}6a^_kT1EE%qcshIK^X3}HM
zDrpi*EOnDiE@Gj+T;BZ@;TiI#enroK1&Nk`c#fr*B+SKN2A;w$uE}Y%vvLe#ITmQw
z2l~@8<+(;rSG}BG`lFb9d^_z#2r(RL%CrYi?lw^2qJA3(Q$RU|m7;s7$WjdOdx{9G
zCjp4YC(lK9H{usBPo;yyb!|}XL>Y-}Wti^gs~qH$XS6qdKugA+$g(?jVEBq;jCa^r
zjdJOT%TzbKASGijwM~O236~+6tP<=);e&MBp0BckWYHRe3FNVIHH>i@)I?dNhKo~B
z>ztyiyO<1Uc+>B6Mk;YD=;poPup3o6e>?Nl`N9s5DQd_!r4oxqvOuOt=Ug;R6(V3-
zV^>YJa8Rhd613Z7oDs@HYDtrmJDqO!VtaR=nDv{?43%q)V`3kgX+L`am#)zQK44bB
z0KaxFi0RO4=^HDZZ%^Dduc6=d<R_OeqR#T3(QlWBmx%aK9-xrz^-S$i-mWKv_&uR7
zm{`zM+00QzO@lWX^>Wf5i-{N|Zxl<?;|KvPpUB98KkqL0b-y(D<9}DJJq!uVW}sDz
zRWsYXZ@x#lnS_i2uM7khuP+m0lv);9tcMLvtnQ`I8<%bLr~>H+Z%AIq1683*r)QMx
z{?nG#{?D8dLY_ljBaTa{la&gpT#%2X1}@`FdxwM5F#yW_{2KYSMsHEdLCXV`=7*)+
z;69ckF%Ywp^7B_5p0yLp?tQ=KD_&~x+N8gn^#wa{fI>g(*S)A^_CF9X%Y7V(z2@Ux
z<j!gozZptujYxA4nzJm~KpSfK{Gu}kG`a_}*s&!9(oQW)Z4#ZK>ypbAi3l>^WE%8T
zZ?ThcS8-?gma-26kmnpL?;^3ZVr6#c?}Ma-X|MH+$JnzzJ`N4kT1UbTRmO@4^aeiI
zKsO)C-@VD>3z)j-T)en@$L(1}%<`C5sL*d<d#XSIu_}18xE>#n8dddLLE>V|k5{kt
z-}U|974=5~Xk%%+dI4YV518f$0VyfQ?cV;o-e}JdEaX#^_ZGkCp$|M(s9lx5fXPe!
z!Oa%7f!3IWl!Y}uUSf=`_%tDtUuOGo%3XosI8&iv@adN8(wc3{`6!v;yZd-NK^0$%
zNmIiD)(2A7pD<hbiVz8|5O)nqTjKar3G&`JeYlt;98#-!FbTDN$NmaUtw^&MKkHfE
z@6IRC&WHZk$W{5-rk(GE=+l+G824~ko<_*~CfSaYiTrxB3dy}Ylpk}_`0mDwoJQiy
z+^3{m-ef|0D3`8Sy6U~!q15&nVwO^w84N0VUKC<N&}O~^B%jGh$pVeCmO^c9mQ19!
zA+(Gcyn|HSYNN`IJ9^}t27Om~!pu@hoM?SFvaH%WXz?*vyeKSDZ_r3&eStk{YvqJN
z7)J~$mu=K>6$n>nrX=O&CO!*>U}gQH+^+%T*euz|#MH;vgSYz}EF{Ak+QH%B68LNW
zzNi?P=*`IIT|iJMxaPi$_V=;~k%M2lr8Z!mG)*hS1r-`xrEeXbrnmNRshCcSs~EM*
z$<^24P1^aEexDlT$xG!q|Jj@|`^{Npttc^Sc;{(pMuLdW&Q{_5kuT+$=3encb7d>D
z0mtJ-Hf)^X+dnHShTgUEqWp<`daAru@UpVSeZ4zS_S>e>_O6(dO6nUOs!0ybPigA{
zxvYDO{`YGr^qcWgq~_35#7L2e0wEJ*S7%w&pG5|P`;(4)$20tCQ+*6xA6fmO9;D=C
z8LBO>Lc~8w1s1+54MuRX3qGra6UMR-;uNf@V(l$qdzVBRi0{2;vuBd!3`ptEbVu?J
zYz>5CxxPwshnh->+OXVt0bT8x8p3>kZS(weRvoh|a^pkFiZ+>Y5>~q(0uT5p$K1gg
zNrwGh$>&%kq@)i48j=Ctz@hYqsBD0)4w#qdfC?jinPC4^F0gQSXSSA@U9*5mVlY*a
zffDoQ+FC21zTyI9oj>sQXal$+u1Qbq^Po&e%D>C6$S{CL#$$jB0t*p;jyJ!qh_DXu
z0~%U&DRB2lFH8BZ4?^VkfG2#YavA-<PR7H<b}aO_;}SSsAIu)TEjaIRdq~1%^N>`}
zX<5pF;#YG-;h`7}ytfbP8S;^@KgMyw*6!N`SF0ecGqp_OxdQU3Xa2KpJUTtdXm0J;
zq==2;GvCCj?`-l@i{_7+N{xf5yG#$PeZ*ZV+D3BL7E(ogx*wfg1hNK_&{S?dnsSZD
z8Z66m#-tHl49ZnHTpE-<KAbJhz>_*-pj~)C^mvxvwU{^vj|1&8J8zRpq^jlj56Vw>
z<~PZLP)Uf{*i`JUS2%zEB=JPQWYeb_!nS33WA?0MkU%b0lJU{vZZqi@J?)F@rb`EQ
ziEB}u!+X%=dxlU#MqVo|I>l5yij2u{*~318jlEW!#7?Lq8%xw8+Mi<%!~@3xudpc6
zI2S&xZ%q0wpRW>S)b0!!uzw=Ktckk4rBPV!C_(e{^FjXtn8Vg;H{sbrAf=FOmkOoB
zkm$3Fi;Hkyia0szwt5B_Jz(j-<fi#&xf8-<fNALxcqV=?fG5>i;F)n_cMR)h&<f<A
z@0<Q&@yDw$yD|<L6e`TtS12xo3^HWSv~gdbFy!uKeowl;%TbPj|DnJw-eIQ<sn9H9
z&4?4txyVfR(w<tRr)XD=^0QER5M;#p7lNcVOGjzrfs`RFRP*cf($(qc(w|KnL}A3+
z+SlHy8fClGw%JSE^y4YT5?(sK7rQ)Tq%3Y)8edKQO?pSP8#Yls%G`{OS`Rffq6Tg_
zMEB&YBb)=<=Ou>9r;B*BM_@o*Wv-)|TPC6XyV6hd-Pir~OMYJ!%&pmtR@v|}(p>}h
z9Mn`VFl5?%w3P4MA4$u-wSfHto&AG!D!z+2kH+wFoZ`B9Ve>n8DQ@`38E`s{$rTgT
z&>jaLoUtYThQ<GgM+i0jK3hxt@Xz&wG(sx?t`MSK+tZb8EmJCOOAi=`<=#L6179f2
zmG~UyIQ91rvM}H2n~VsRdASUvc-A8$c2$3WdHgQORg^DdK%g~j5d|YQ7p2(F)U33}
zRLk+RE9?1w2QuRn)17Wt-&~%Y?fW<qYn4CchGivJn;YUitlAuUc;Eb)GufW}lGfIf
z&2fCU6+GjZU3YhgKaBKIp7OcOxM$It)a4KsxqhszR)cO%C|=XG7;TZ;!BOFloyy_M
zkLV91MGXBS>-2e`QuHFw4-y}t6&e10w9@!{Jsis^VQ(f%rNJ&4zPC4~eqM(Y%9!Z!
zeD<fwd;{m~5p!+(S1_ylu#(ZdloOAQNO0Xsz+qL^vehH+u0z^iJx<rSPu-Ugeexmr
z=R_L;`BJNKu?UgGRtg&086F>&4L^DG-(7191=~*`HV{PgqTTAvWaoZ|XY|=I{8P&O
zeJh#n`+}YM*fh7ydLHN=a8Ku4Uw{0#Y$R~|Ml&iPSS}=Wxkt(Rz#7WT7gnRpE{A+6
z^li($rVXfiC*|;yKJ|@nd`cxIkRLRg36Eh2+S)|q(@hxGqR+>^6)JRA7p?w`_Py@c
z&+dG4FC_u3{q7f1_K&lkp7VOYBP1dF+)OZA)sH7K7fVn*_YyOhfV$WA__wfDqZsY)
zF+R+k1pJ#)BJggy{2ISV^LYvHz}^m|e97*}zgp?xEPk|+#Wiw0=#xCj{rzJ5&2oo{
z33Q*S)FOtMj61w2-Mck_^NY56eZ+g#81w9_9*%Q!dr^8lgSqMTH1Uk*G0jW$yKXJ=
zf(g2O1JpKlU&ww}9w_5n>@<|$kv;9~9M`emf4~Blg<gI#@2Kx|H@z*{w89aDC_LI!
z`i&k!8U-5NMmZ~6{D142f8NY}OpPKP&1#jbrxAX@7E}Cp=DoRaJIL*d%RujlIy^zK
zwX_1o1d7e?8}b(CEzF-j-?X$Cp8fb8<=xib>KJ`ATh`_Z-Kt(7N#|Oa()LGQU)g|~
z&%2pQ=-b0uhw;*bRWT;=J+Xo^bOXyVxs}>%F1z)?7yg1?{<?3oeQxmfer3rHl^RNM
zYAv=q6C8NhN{~n6oo|-d{F*3%aMyFv1L}5-Oo_cQb)j}lWF}qer-&_DzSmZO#FcKM
zaCLFHmc;p%e_>AGQx<EQqU`49gHk3Lk6#&+p|I83B>_bScdT$4cj4T?-8pg6KPN(>
zzd|eqGUF+>Q)QO*qk&x7_iM?3Z@#*_=Uu|IH%FewKQz9)dU4Sh=Yr#L?Y{Fkrp((e
z5#?xsA%W<hTawq87fTO3RYP^omTZ~@8ZwGpx>fC@Q8<QR_o~!6Z6FS}TK-j&{Bs+7
zLIDKV@qDA#?%UjHo8#NJ?8O1i#8p;PP@bkUqb-5X`+q+YqF7`v4|$7r$+;}q9?L+4
zl(?%HLR|tSP6$%8`}V#JV?c`!wLH3zJ;(>wb0f>JvJ<>ZboT2WkK;=}(dmj|6+*1m
zu3m-iL5x+NSzWzl8X+wt$3tU_-Bd~Sp%HU2jY9JVisqZ*^uXRnUtUYo=PEfG-B7%~
zTjCe7Hu1VX-Eso@Lw+y$0XIE|qX+ZmST~C0geiCtB$&T;v8xWa?i4O?^Jp(abpSoV
z1C2_sv~AASa$)6k-mzsp&GK|%oa4cI0+v=Vay^ZFsljn8_2IW~(iUo;oFTGodF7_0
zQ2hOz$R<l%N(V9+?)NskecHk20^O`ePZ>CRgT(pr9?FNhd^GM9T}rYwI=y_m9yG}B
zJ4sl&6vkQ|Tei}F|8E`sF%=1jRzq*UgHgR7&d{=q()WP@We*nIzf1kM=K*aYvx=4$
z=51L{%&PpwSmaw@Qs!w~e_`O#yT9K=QGRjgu82)u%C5kErHKMe(5CMZeat+6+mu6^
z7>du_2-Q2AotxLzC3@&R-h0fGMnfeZ(UaCw4@n6>(4}C%c^H{;j#Xh9PtY6RD{Bx&
zp7$`GTdX{tg)P@{kXv->ip5abiW&Gbax{^r8V|NJ*zKiLqn8HCysY$desh>%yFgO$
zE(xlX)LCqs-9fP1wCPvtm{RI!wx^!&()U|nFdNx?9f$2wUsYKKPqcG$B%W-VD++Bq
zb~tY*zRznKzMHAtXnEG~0CVW9(Pr*l8oJx{gRh#sFx_fXnGmAO%Zu_QW0pJC(Eg;G
z2;;~}Oh|q+a1}$-bBs9eT8Unpud>GH9^*Gv1vc%pd&AGB8q)8Nh_v|aH=Y?0Mn-!z
zzc?fJ|D*DU`{g&L(~N7YGf{i2_8lb=drF7ye;zvqOeoWI<5CdyBESYs_igUXin9Id
zblhGp%MTHpdmov>zcW(+H6YvVT;aC+3~nPPG2h{=6qG*$ety6?BxImFbX8p3=aWi<
zf=PFsr@3N#ee9=Zbb|vN*(c8?%EGvOV^~SG8g;nc`ufn!7Yt^w;lLW=MZYsV?I=6T
z<F<T=ra%KtMq8cp#@^Y=eK?K--Syaggvf%^qjJ-8m9OL4Zb!q9fF~09&sEu~lI*kZ
zBRk)t1MN0)%rtS3H-BEK$}P|J#933k65~#lGv2o`(pXBDw;z^uDg63}G=)>~FoB1y
zA{Xo3QlN|5ue;V|blSu|XG@Ppe<ljFqJ0cReq@&ABi<z%1y|1t`F-v<uLY@dPwbRa
zmn4yTM8FS_I8A)%@w&H}I1Cx{A{VLo%0huOpP=!^d*?Cn=76Kc5vI8`smH5B_y2g~
zKmqr!ef&?=_4g;3ArR{n#brF7XSzz?%7sx}2Nvv38gH<03`?}I_{E$->ld^hN$bm4
zt&|l0qU!0r7c_Z%(6o2UP%}8O>breB5m7`HW2K+t_5y2vZq`%l+r<+_Q&YFzJLt98
z`|)Zcs68U{<L*=z@@!uj1d&5}Vjtje>8`cyEd}nK%qUlAZ2L8O&Z%foOs=_i9iJ{>
zU!ElC&dmc4{xCbsb*TBsPtI6&XR_kvCP-(5W9tiPi>@Www3}TNRGLz2U08*MF`j)R
z2-XngM-y`$Z@%gJfXb;yL5&j~eV?>zSqVP&MdOR=<9dQru^;GT@wb2ygD{%;;XWYz
zeNZw=kUw0**DH)6EI;b|6xB;Q>$U-!)vjj`O5rFt0{t#L#)<U7Z9(FC*%{9I+Zty5
z4#3#ZyaCtp^?+F-VinVz2d`YuUbO8$mh^8y@n4H=c_+OH6$jn%73X4Q=hbMB$#tuH
zdiwiUKfmozL950ltrNMvAw7RZL4yhEGaO8+_V3m`Gu_v1(9%5`%?WFg>Ekr9r=8^J
zmzOUR8*zNcxQ|(3?Ww#7LIq?CC|)UK7i)U_mKL#b`s#DNBp(Z5)1QXbWMGQ!@!2nB
z#l~bU(x<N)+zdrOuo*(ah*`3ic9_TAx$mhvM9MxvOs`M8o*++&I=hS%cdIjKAKLP+
zK&<tPjHauzZpXP^{UJr3E|{fH#&_wNHs>WvZxU-&e8H@;8m5zeRuC&Vo*=@5qo!67
zco{7}7``7Rbl`R2x03SV&u=NOgT?sQdV80Llr;j64-N!YBqKCa-~Ra428!^9H`jgy
zMh1x#{MmQNx2DeP1IdZ66Kl}ivJ+}fKPhzn=S%nB+ZNz?TT?5rZ)}VXoQtHYWszxq
z<saP-O|b>6cwW)**B#xt-L=D6*vSv>MJ5*Qh1Mo?oc?@{4ziT031goTF}wVJLz|Iq
zd1G*22Ls0~$BO9qiRsz{%LeJ|1eB+oaj;(BfI18Uc9|K_jk@pLlo<~b?tbDSV4GEv
z{n|*$IgfYpl-cggvbgMVN9Z4Py~gPxoq%zI?z??*i=T99hKiD0UcXyLlbI++k4mFY
z&XE`R6o1%`)~ATvjx0K!3n4B~SEk)wU+8OKjIQP=!SzDvX}-~iNP7HSYGt>fzGu%<
zNYORtn)mA@!{-XU;Cp<bCD^wi$^bYD@eZzi_rGqogh(U~56MH-Xi;5sbcqLdS&Z=(
zE(sP;SD6Zg(6|o=9>Nt~kSB*k#()1)%*sr*=OLGHcvHVC_E4d3p&*X7C$2oQgIx)y
zL1+qE^1#yY$ni&M2CdW|HrkgZ$^eckACF}Y5$K9k3Y@DKFg)8E%dSF_Uwt-eB-wJI
zcG@sfoD_5}Xv#6zB$P{~d6Ak7i(|KlcRe%>M3<X>f5u6SnS>PIC22X?-YH$-BWj^_
z&3?4dR5qj9A(y&L7$h37LdMPMXIat@PzNK&h1PD4*VMF7_1wS5&vlxo$u6tIla51n
z9pA3dlo4K3Z72}E((qrW?Z5B2M+^YYwMh_vWuvXGP(l94;Zx(!_9&?+fqfynG%<@r
zezpO|?@8jC39oe%zeTxw+Ah_ayYEciCHF5!i0F*nBi`MbIuQ0#1KJufvSt#ivDs(W
zcl;kdtZ#H6-G6YRL<I#FqlndB0WjHdB9Ro+z!^2=y3C26hL2=xOxBAKKaErj{K#A>
zi($%zvCi=a&8B%yOI*^kiWnPb^~(9kvFCokV+7f#BJiga-Ar-S<9!?Z=GZ`#a?uh%
z94c6xx@1Y~UtO%lBT>I4h!NFX9+3Q+5|AL=Jzc0@?BQ^Pcj0fl(!){{&<3OzU;xYX
zRWcL3I$Mzi$SCKLa-R+|=im+;r6d^b#_TWP`u}<f{$5+{2xJ1E?BYm5TF>Bzhb)Sz
zPp%6}(tiajMTr|T((H~=U3(jI*1TA}4q<UUSCB?K9lqD9XL?MHT@|i)yJ&B<7q522
zMfP6C4l~jlG>E&_seWfmOIPf0KQL$v*LK0>VU4bxcD8F4%#8h9Y2W(Ntn;OZP@0?|
zluJhxzQ625(q)P)AIt^i;o28!lO6~z%`b%W=7{0uxMIzEs>b*4ywB!|itw4#+@~wb
z+RoQE<A0smBM791C@6kEYq`6um~Q+GBb*{*f#(C|O2B<}qkH_}0z}HGSnbKkCa;4s
z$87$`#(U?7fP)`NKGYo}kYT^F_r}cgv@S#QE6rq0bWx#4hSKM>?U~KfjH8IHNO5ep
z7WI*_(4U;<Zd?`){3~jId%}O}mH#Xb|9vPQG1@I6sE;@;$w*2`t7a{1Ne?v#)*j9$
z=c}A;|1pvlxK^395f#6%HXOHFkAJ*b?-j6~xl!v`-N2n9g;&#Gc==B7u*2Sxn87k!
zXJ;*mmJPc+(rw5YUQ!_%`Pn&V{p`sB(HFH2{6Wch*u{wna=LP34~cqTrAWtWQz9f7
z1J423MW5F;oR!FHts|!A9LE<{(CxWi^K%|eP0gx6*1}eLYByiKa>i@QdCQg#y3&<r
zM4iFYNFmjLIqwy}TkXC>v(BW7IeARqTpA{3st6ObE#?^~5WkBft5R11?g4JBwxApo
zn?k~G;@*0`Ro0FSFF(aT9$<S-K?*hQ%6*jeblUi;NSD;@kA(pmaU3-=`m>)5ZXKU`
zY>dGFXkq^QVh4i-ao;lI6qRg&^{6c}H=OYz=sM5LyPY^tmdQ0|WPM1R!pCm9#?`wx
z#V(KgVT-A=&<`&A-1St+=}%vp&Yxa@#p|>W!xBxwwz*ee9kO3sU&zcDD@p!@A?GD6
zALfY(R1Q%nQ2lll_{p>K*|J=o=)!<ceCUjjIJlIPc3kS8-u?J1&p?v;XmcHDU8Zgy
zHjg{mEKQuNtQ4$kV=wczpeh|{E=I$Bj_I^%s`#tYD3F@#g=3o4D%GRy_KV#-0B}y{
zAdd}(s%`Ivj#XTEY-@pg4-UOXW7iBVKFU|mBXK)h;w34#rc_Q}IUFcoI^)D4mxycC
ze>9|M|DKzNU~yd6abN!nLp#M8|5u{)6t1;?h2RRoZK81c-GJ5J+MBW5cC#XR<fN{@
zra~9sHT`?<{XchQLI5)Q9yYjJZ0)}?bgzGHDBAU~(LX^f!NUTznP2$by&O#9d+rw3
zsOSEDYg*Thdpgzg#9VjsYK@v4m#%LVZMcb`rTu338Mo*=E$;<$ahGpMOXfW0Mdz_M
zC?#<_CJl!=C8@N5RW20=F5IkOMpbH^MD2Z}uX=N`>)IAV4Dq?j0a!G4jh%PewF(CA
z)RZurDm2~6f?Cfl1CRYx)kA=Fm`@Jj<55jA2*k+N?Kd@tB!B{NrfdKuSu96i3_a<p
z)=_`2Phgld-D2HuFi0WY!9@Js-b1kp%U2K`oC9;d*Wj_;TMFK=TZtI|J=$i{J|W<1
zPG~;fhqBw4SFR@9-flToy234{{kt{uPL^C~uYBCMmdkT(08vOaTsoN33nib`zCj|&
z_qyn@C5|<=$dN1^5&RZ+sYR;V!gz99zegXePnKhvZjR%!y^8GGVQGv@F1mxY7hij)
zW#8m~{RDV@;h5p{ud^Q{?uIB>wk{B}sNRrUY2nC9Tc6xd&5{@TUQ<^7cWaLqLna1Z
z2Tt`gc$(`K?tXOJ{`uVSQ>f%biJ7HO2e0CR@d%T3v=109nJ0eNiHR#l^~`ai6|z(S
zvo4nOXNUmCHyPEM9F0-$z>OCS(KgwhokDS3`=uL}Bt8)Wogd;N@5~EGE|B*9TAX>2
z67;DR$2VY#{~69nozs0pN7(36Zjx3z$4dLPB9hH_#^&DA>!#twpthW>;kna=K|g*I
zH{U=)^9s+_1inPJ<;pgo(bgnaqpTf}?)##8{@br`e?{Qx#lAh1*zIRD$&W`@J#h!S
zBo&jDJ~z;_Gh;G0Ry$3(o=kh5nAqwRoc{l<%9G*5`uUTshP1khcQ7W+4;nG${UROh
zil?G!#)o2(W{phvr&1bQ!@Uy+Qij7SgI34kld<w9g^in&q)g@Tr{X*`&3sUCml}Ho
zmGd{u9sEvTj<>$moa_>#$|<1|{krW0K<miwYcm%fVd%1>0dIqOW!N&7%g26m-c1D;
z+3&Ej=k1;|*60O5bJ{bo3Wx49s^mqYPFH`p6g@Ah=FiYB|K3H-qg*EM`fJDuz~<Dh
zGMJJb69ycJrM5#mz9$FX^jImWLM0V5*1s7krW<8?y$(XC=iMIL<iFa~k?KP0lLs}O
zX`?_g&wtk}|KlhV1tG&Jl|{X#wd_CsfZ8s(Qm32ux=KZDk?Ubnin#9Y3(P!R%C$w_
z4(+PG{aHX2=AR+Ks5bjhb!RMwxvF*ZF5%5h46C?M(XqZ)B`tT;q|T_A{I#~-jk^Zs
zglc2ZM#p(;csc8BuMK)|S~1b|H_wg9DaytyIa%;YZK1$ieO%^P)`yM|#1wMQN1jTq
z92LEQAN|68|Fdo|<MAkbDE+LVSNp~W;<S|z8b+)jGxaTFmyX{;?Jh3O`(0uivVE3E
zDZFYy1)AHu_C}v=hNAC<!8czTFGx7u19Jc*8d~9TP8S+d6zBhYl!o4nQrl91k9vRP
zi8p^;vV|==iFi`sZtgE#UYr`<AUgL^7EJZ>S6|R5tB=9!5ST}Z(-8aN$5(A$qOgmM
zc$oG^>I3!q6bdQ5*O=My%7ugV_;Q82Mo-n=j)yqAh)Ujm7BYQU*?BXdoDbJt0?IyC
z60&b{_t!-3%VR~-Cc#c;vld**$)HZ#@-x1cA5b%6l_Ec3G2!~=Dyc7av);Q`V?8&G
zV%wCM$ZNH%8MW50zUNhrrcq2&_uh4_?-7t8NZ5(h`h-t-g*(?ZdgjSAdrup){;>h+
zrYJvG?Hk`8UjaGRV;P@ziWDdRQ0Epu^isQ}R)pmaaQ>}19@j;^obvxW*ZhtK+;L$*
zeoVl|SQ$LV{kC?sH@+<Z%Qr?IxD+5jrj5Bi)elGDu@DR>uq+G_tf_9$;+o}9$jzI>
zTjnl&r85GhgHGrziJ-*@6EaQmf9yO)g+(xZ3+PW0!FVDDtbx7&Di^}1LIULTLBW(C
z|9z8gwarl8i!_MO3D`VH2{fteN%VSG>NJgkK8>4ydjWhO6rmfJ$N}YFpl(wzV+C?^
zTMm*mmH)Lme#4vG=9(<jdy=W|)i|W0DTyo*>y05)EcGdRLvMWf4=UPNz6X+)pNN%H
zKC)^yNg#%;Cs)hbufsw6W|0}G!?_sOx<q_?lu3MNY+aFl=znlbm6IW7KZ(ghA8_#n
zrm38==>#V7-X$epA7-T8zIeenZ!?Fzl;SGa*Ly0hbLzjM^bQKDDU_M&`XzsY&_~=>
zDPk^K6>QXM>W|4&f|A{!29sZ+QOAq9P()^<!WLv7d7kQ(relJ+aA`!JUxi>350;h^
zH1ud(*WwF(cW{dM>Oz<I6^JaNV++VvUpbv)W>^8>?IR|5$nXLnPK48LVW7x@$)^c|
z-t%^_<^`>-tt|!CdFBA@JtAHqfnsjWH|kEZ0BtY_Lt%3lkje*603yJ!w7KGk=gI7E
z*N@S<`TF0`5bHL3OTGhgN|Xl_g61m!d_S;%5w)0w+T}11g#b%TY|vTIa}9EAL-V_L
zYk7RORXzyi_Ba}ThI*N~Q3hz2OEtB?pXn~E?)&n9iiBdoJ-yr2LJr_#Fq%7_@GsAe
z3nz1P%xk2C@L%j?z!iQhi1979Qh3M}?Y^vo@Y`(DwI!pmRGNJ#>mKe;nzK8|hMM&3
zJuzqKm^@<(ZHn9RMOEi~puH_t`U*RmYfpZ74+|VJqHvk6NPN*<350;z5st7e9gf@a
zeQnl&IU9eA(GjF6M*yWjg=~ZD7m=;0@;le3(tm8RD1)7C(AEVlFhyyE6Qreaid@%P
z>UUv0%UrMLgBBp=OP7c40*7)hKYtdIWSpm&tm>ciEb#dgMo!*hUA?+oJ#59Wd9p>6
z1OQE8y~|kqSNd@u>WRPiidlcgnXYzybh<NZ+jD=PlH<;zK^L}b$l-T1#V2r}SOCL>
z$9i)C;9~oSxbZJm3nt*-cxY6H`)pTEEcfiaGo(0qbNjHgCn14+1O@!yQ~WG=e_ng(
zC{gASae8-t9EX2Wx_s5r<Ke0zl^VmM7m}s%90ZBgCwsTo_;3}MN`-w2udcTLPFVe!
z+pa-v4%IHmGVmzY$>Cz##~*CI_X575R-J2d7h{J`%ql^Ml}Z2N^tmsXgxsf+0`;|w
zAK~}!r5T!<+ePV;ve!rI<UiM_Hrpg8_P~H|@4Xua{#XgBvfim{7i}P~tBoBFq+bc)
z*o9`KfM`#_A$!wOV^0|sZ78v=mLo&DJ6>FLQ%Ln8?y(1^M!}n4QVw~fNqpdd>l8oh
zthXmO38}sG#?t5SH%Y>kZ3%q&b+R-7(K*@1*{M&m$)*7dkpRJQGsaDhF^0Zmmid$l
zyzmjInKMg(-uaRXww`o*i~Xw)P-R|ps9gkFhyXCiUkN^`RaFYR_qSsWkVnjK{F;lL
z-~XlmffhVd!_2(SZ}(z$EBbBtYiWFs=L+kC5L|P!GYs-Yy9QdM{5Bg@r@C%G-|u@=
zynFD%rQ<ND@w%?cnU}~5q+j)K&`ex68dZFechUHISE{S&_RaA{V;ai)P!>pGHVEa^
z@4>MoD#D($5}sWa$AjUGkQXnl%hp`je4C`c^n8+OzjDCqtAaHoaQZ8Y(tNv;iE3Id
zP=gi^9>anUh^<6J*&$_BVq@=^FPs5CVrjFaJi|i$s!FBvcg&0Ap?X(3(Ui^wR}=fP
z=5s8GBNc4#wsCKADd*p_%jEH4n*ry$Z0jmdW$Him79071<@p$d)V@^P16CFPhp@Md
zs`8D$KH&gT0!m9tBaI-9bV+xElpLhHq`Nz$yFsNpr4cx!bW2G}=Un{$^UTci=9yP6
zbs?_%a9^>%dw+Jh#@7Kp9RC0H$o|_b+5LdAO$=aLpHNUCU@wS3_`&+$ULK7YZYMPQ
z!=%5grEL6=;V)R*L*M(|eW9V=*QQ>S#@9UC=c3nZnJ8|oe``ijZfQ#{R6~k;f8!|J
zF6cf&=EVA2Wu5;-HsPz!{nlXI!{gSlHhqEO8$Jr}J6Kf0?|H}`GI^w}7h3V=s8<I%
ztE*b2-&k05yr>(E=YO-O3Yi>hU4AejVmd-6aOkj#Oso@)%Q-%1nEZJB(QA$9AQnJ%
zDdf`oC=D!^%tkBcP-5RL&DGSb4aEIZ!p#mFGb94@5QwJ1uCGF-%4!(hgwgpP_VN8C
zO|HA+est5XK7B?Zv*G!*@^95~4c@x+5$3Qw%AHwOtVJ7qt06*~aG~=|$3F%8(YZ5~
zjy?azELkJvacT9Ue^Pe}Dj6?pvf$x_68Zr@c{gCI;QqMcq5S_mS!xD9P*~Lo)?S=C
zH$4{cH`u<NU`}jrydH)~k4oV8G?mJeVBv9pdv*6`3&|%+{o6n$7sc!SIWOb@=F~O?
z1mn>5kdd3J{LjwPN55%qUX9BDy9a#Y>pyRO9<=?09+bee$%>Et->x){+rhg7nLfs<
z?>*|8LGPPLckiCzWZ#~>QBufyVGyrUoHd>;_&<iD51nh@MaHt;9QP65_{&vyrl%AO
zYv5lFhiQ;YhjN;IASx)p?JAI&clPUZsAc9cfFK}OS5U$BASrb9h|-+%An)<e%cZX%
zF9$AR{9mrkP-`pF5Qw<~>!D^3ca+kKLg<sl=EnSk_l2oW##b68AH^uV`4*j2{r<Q{
zrx-Zo39rCc>gxWx=Vd>{PoY$U2uz?Skr?*>pHA~|IODH@-anskS}i-4g(2$xCgLaL
zL4w~W$Un*OW^+oT`MIfcM^FEl74oP)f6h4TN|wMUj92hnQg?I_(PS9ym|i!f=+B<7
zzCjpki|l=*lET&cF6`MoKQ!a2@984T&bK#}lC|-45Mc8YrvYzrPpoeWz9Sel_=8$k
zZ{C;t5Y%6^3UUFpoW73_nzvql;;jM51bM3j<MY(BA>tKSdbG5$6^Nt!W6uDE22$1W
z9=ZKi;Wksc4+@TKxF-;n(1Ca)8MntwetmGmMfbk$#X!btUi;0CG@79d$8D^tABGM0
z6O?WrC{?DBx_+wZp-JJ%5k+b#{Oz=}>MKPi<ELqubeV42_Atti4kWGT^t#CsOUcb$
z8QQDLWSWqtA3fh*l(W5QbdY$&i`@^Gyfi>-j29eH_;`190`9+Kh^955(+J-Ce;)<M
zHX`1DC3jZeRbLpMQN0XzwV&*156bS0Cbd}%$1%qewHq|{<92QCyg4LA=uEaa9BgQP
zg%jjDt(=cv;JQOCKwvqhgxE@rYYcB|bu^`Gizi}~EAWd!-D<qe^A2h4l;m@^{b3Ti
zo9Z6dczf#D|4f!+g(L7WOY|3>wYcuH)@nD#BDJDwA$<uyyV<vhO+!mFrf&8XZ$&`H
z7j@zKUa}DK8W{O#KRQDSfm14iwK~^&LGO1p4d=_F&B36u-Wzattc>nadQi-A>ggxv
zHyVtm)T-$#MRUQ>-ytaRa3<FCv_?ZKQOz~!$EDXCz4Kp&<*OR?qH%wDY`fE4=n4p*
zH)k=39I|nd>l8)Q>a@&F@74a#V*29A-cjx&N9B$(1x{4%Xu~xBV?F)9_nuhn5~W;-
zj#SBO!7)0pNU{<5@j7m5h@3fiTvQmK>}R_^wBrEbX{7jdef+k_r<nHl-k<pG4)00P
z+VyBhG-hINyFPLE7iiK$mO=2<Z?ZHeh`U~ms<rFoML{KuPI1PDPBKzj#3;u3D8zD%
z<xO2;vz4DFT8JWeAhF|!9+xhi5v-pZ9uZd)6!l`WzX?kQJKy>Rw#U5$eG>NDOefJ~
zmHKpgC6@b>+Zoc8CQ|z>hhZ<cqJy9IElQ=LNYeJZx5QESn>jSvO>C;=0{|oHyCP=|
z#+R>jx|n=yhVki$Jcj+GnQoSWH=)TfVoGRUd?(FPWwokB#XeV>D3Qrx^apYqamPOe
zB0X>S_nVsow%Z}(&EpJ?FP{?!mz(}(hmUx{);}vA(YbE*-{=3o-4y<}-$cFuFXw{C
zbNSc-64YDlrP3MK!Hm?D;k`-nzjKWp`C<CYBQo0~IoDe-<rbELm2ub@w|7V`irZKQ
zyjBZWE$ZY|(p2-Cv8;o@j)(Z|?0P;@zNvXGEj>>cbaZIX_M&oA_%X@<>@nDx!rAZp
z<}ViYeAUYv4HeFAgtt6*3pw2`wC_Dc7IFcb8V<y?JItvOI9RB_wiqViThZZtu8s=a
z3QnjCIEtG+f9Hr6(zQsmMf@O3&07-Ut@|<l_+b@#yFW?YN4tXvvVxJx26~ntBXhG;
zr6wV$I1_gfFYNA)dn72(yAdc&uxMW(Kyn+r*#}7I{(ltC??hm?)&~QL_I0(Rw~36%
zAd`2hSQBRBH``>3wdwO?iWv2`Qf2CNdG?beq=$Sd)@4qPUXvtv2LF9-#H5fG7oHW*
zYVNJ0!USrr?r&9*6F?lY)SV~hVMBt-YBTZ|i+ApqsE!TwuDAZDRd-X4gAopM|0~Ot
zCPPg0x<VH3LtYa~gX(6tVP+)y3S;|*LK`;oWN;GGhu|@S0Xw!*UD=QKD!6MoBJRh#
z#f|e6@Js2o_^Yl(I}P_w=9)g@Ao`*Q;=^|S*+kLhmctKz{D<ROcrlkkLgYD8F_3{@
zNj~Yaou9iUs@dSLHI2~;!fQtV&)Y*Nf#3DNI*C|8k61RvEPh5bw<Ldv(tuS2d<wsb
zu`k|2MV6~y+ot=os%XR5R_z3ML^G!;OB9RQx){Sg#5Xm*nsB}12DQvL(WIwg;??8#
zs{uF3S8{3vz7_P>=3V#wxo|$G(xnT+JHS}_v`Ls!e4IjKMR!Xml+0kH3v<hMm+W|a
z{HN%fZo}Z4K(}QQwn^Dv2$qT0%D~O!q3~0;{R@=RfUYX;@3J5Pm(7#`lZAwVYSGV`
zW;U;zRKXS$mgXH>x4%DX?EXYH-DJolOysy&A9w%;1KM+9d|exj3f&BI#~0FQ@ZWHB
zEp9j>9F1gMXTlBFOc2>#{Z@KaO7?w{4meDpd>Rt9Qu4qos%q-hQxSct;hSI|wjSzQ
zduW!&mggY8k(D0SvTzW6McDOstXg}I?ng1#GE?*(-_{!#Nz(9$mu!VQ9Um4)O#ZHW
zjFUd6DlztcvlX@AUQ8Y=hFE4nI`xYV#1>-yNs_kHWREGe5F_VAjO?rM+fpipFrGg4
zH;3z1yqRdNmsuAL($%=}Ni^BeoG}mt|8p;?q^I?byx-OqtC)=F?-td=t~g`;bvE+}
zuf_0)f5m%*FLxs@t9%ZGr$X9HBgu0WM=G>2<XDvtcXu@qeplM%mbl}-Uf937+1iEB
zDt%I={NTD`+YC>i6zos%-*7RqpA;w=umx@j58vMeE!9Qbnt6Q)!&?U(k*jf1XA)qh
z#4uir<u!?EJ(8j|)yN4xOKNPtn{(3RIrBL?gV23XN5?@GcNQS*ukYA{O&Y1irFUB}
zv&g~Fs}7EcO8@~>97xgOe7I)@e$Ic}#Uh`}E~Xu~n?LqFKe)6#AOvE4C<gpJ%+Tue
z<qtJzyu<bB77$-VpXxDI0jBquj#*4hGglO;5m;z*K}(Vyp!kY{Y-ui{%`bfB2G2_y
zttN-Un+B79%qsx%p6UQ6^74dP3kwUSBDq)e`pukbr7B2Iv{dL)^E-gI!+!hr?Q?U>
z$A|kXWoL8YtBMMy#qgscxJM<yXfi=GQ0;);*ovEmZUHEfz8zV}@`>2lg|aE#fRyUH
zUMz%7_JX+C5VXt&f<G`u_C&9>oW-Q$sy@qm_scyk)}X_?O{WitKa$ME(|*9Yw2CO#
z-dt!f7*8{OYglik4FwDjkH>Fq7dm=xRQPSSfS5pT*ZXI1{S4SDw%1pdg)qqkP@YnH
z4JkM2dk*(E=LmOEZckkgl~`xb$@!e5&0O}T-h5v;umfLEex3`CDC0j&q#CdPq~|(p
z4N2xrKNoX8vYkin90mbs48RA~{VR59d2_bQ=J-dBWaqwU1_n+%Yp)A)J;t8=mrks0
z`Yh4UL6~&)pzxu6J)F_*`}G*q5%BfdT>fe;Vy?^nt}(|l{q(8qj{XotBTCKASG?nX
ztG>S@t+W0~rvmIaR6J`}WKYV-=j9T?!a>8bZ$Gmp^6C>GN|-rn^10&Y(|}D563Scm
z?u;ouzg1ON><-*bbLzR{HVIK#K&;rr-tZ6U3Ml3)dQZIU{m=o-d79_$rx(SZZUBt#
z=~>!Q>XXLr%3{NOS~?VVL*VhgB)d$fuIy&E-OF<a^u;I&VulV32k?x2<v=#9eXqJf
z{V??~TcV6W1eBV)*?cLi+V_P<-b0KEkJ(cxSFGWUphy5*2w0A>(4Jgo03Hfo=bT9J
z*>Yzda6wDA@-G?yfAF*ZvCjpB^42FN+xupaPQgQ?ggn1~TvbxhkTybW*l=d-iE2>7
zjMlHrFL996VF|j`548g4D^`3iGb8i4{+5+joyI7b$sH{6AGIQM+?_`7y?^R8RP!l|
z`Pl&&3FA$U=5<R{9Z+8E?@9}b*4>~IeY%x|-{rYxbQU~v8V^H==PT88;RCkwcic-(
z33~*Ucj62N*qE>yS?7$toU;bYS(cpoj3EiRQmZq)ajk$pb!MH~P6FLbZryLs$%MS0
zqC<OI|Hc%lod0Bx&v!^~Dqi-j6c!ivfQ-w*5wk!v5*87wvhMnUDMW`MW~VLSP=f8U
zGXcMD_6bw06Jzm3;hw1NpOTRLhFMw9nchp*vVJEY6_sw!bDX4<Qz`}oLG!8Y2hduo
zx}sXrR=6EGFG<9xlKsR7>=p$2?R(2C2aEe+o)jRTf^tzYSu_4Au8h7JPRWA$<SgwQ
zpB@bE5_`+uB<XEwT=_A<^B@r2S6MLI^9;kNv)tTSn)tUY<AuD$;c-;Z{wDE8%3h_J
z^V6m4RI=*QKUa_o?2NJa2kxcqDz^rEK5u-^^i;hHJwy($qZk2O5WDW6zf?Zson`V+
z<h=A}#!{P!e=_bl_Ha=$07dX@Zf@?Om#z1Qm7cH(Bh0g)IO^Ec8~iV`vfn?JlD=$P
zfq7ojr0A{VaFHPsbK3+0!+oi9#ysv#zDrDBjG7Hmhvx&ZjAAEU;&@=?n@2tD2+sYB
z3R%`Jjmp~&>-%*kb<bSac2)uNQ6~SGxB)2HSc$<6`L?=Kf3EpGszN70!UfUu@=!gS
zdx)-0b+WBsJE0v1Rx<;dVC(cRVgh2Q#8&seE&XmE*eFx(3BK&cR4p3l8GkhqjC>NN
z#9vLH?VQb9sWZ<cZkcW_<U#({_JBp6-Stla9Ao`g!9zsP(uoPn5j#rTnFU`)VT9`V
ziDP4F+a;Dxp=C1f>=yr_(e+*yLf(n_byyA>$wsHp-D=;1xaaBlT1XFZM}MPq_a-<C
zqTUZCAhoJ+e(?wUIW>h+VK`}3%V_k`%=|t4+7%GKMz!rbGN{$XnE#Z2D-uz_T)h=t
z{XLo9W1ii#wp$zl1{j>Swh#Y}p}q_F(aNi=Zp6rTMsfei;i9@)ANQZZBy~{wri|SY
zqTPYD)XMu{yx9|icUP2(?z)tOhC^$*cBz2C80X*NETRu<Z9;gsKT>mBN3}LO34IyF
zL-)DA3A#TUlmln~(eh<y824n$6M$smhi?ZUycF^r(J!hm15Z5?ogG4VSQU*>Myx7O
zzKV0ki+Hl^d~(KR+`&SFmyyVof%SNo`$EYbfc^0@BHt#GW(;5#1sQ{{q{rwrq6XSz
zPvEWqV8%<buwsgB5wRNw4y|<j)#>LssZb)bGycKD!trngV8?hF&sFW->I}%1M`k$!
zRbTV{^=3jNE*9O1=qUG=59qNJ*-Fn^0`Uw+`ZNqa{UOHbM>~>mDaeG^aTjt#{!bZ2
zBqRrTBZ15ChYiHan~^-Nyeqy8dIE!hydUdq^oeSA@tTCAI6?}^jP%}b+Jn87bhmkr
z?pFqCLPbs2>dvg(Nzj}UL6VALD%h=`^;X`j5uk7Ia8KRyP;PuK50`>0a@}}#8e3UA
zdr8pHWCHDnhr}M}Bh{6tH%Zm?eDYC5w4PD97kKU(&VbEzxmaLIGlyPN<it!^>(O$T
z5D8_->vE}BDVxr5N<)=g*1CZn(?<xGm)@yE)IRUaSSouD9Ru$3H}s|)i3CH$;a5o@
zwR`(>&BvFL014?u@P@ciyuUHUBh)FTKZ=-6Zlkm~Yobkq6%I`((4N>mD;w0A`&ufa
z93@bPqb33FgMo;>2Kd%|tGtqNxo@H4Iib-;^s4j~YIa?L{_A)>?zsZFs}dqcKl~xQ
z+0|dSd~ZtO5uTNd6^7)(9?zJB&c?-p`!&w=ChuSNDthS@kc3AETI7@{aEED~3ZLY$
zWmCR?D}*btpFGG9$Fi!dq($F`RqdqIS=kD7Ml`hawtVzG?mG9pE9E;a5-gTbrH6mn
zkM#7^xWRb(ieA{^li?f3kA8VP?-cTnG&#N9H{+ym^)gn;)kR4s`KNId1#%Nv4VT$0
zeIBOBCR)lhd{_d4MLmt0AFSb7%G9e`4*8G>dQp0Ax5J`Ku?$+>elYFE-tEqHNw@~D
zz;-e-ZnX8B)H0T%xR>df5LjlIFmWQt0TZ~4jl9|(zAlq;RPr90LCzYIr@mF$u&g3}
zWTkUme97_opM!%#_bA`AUl<M@10ZRlf2E2F0nBd$q~2MIEh*0Bf}6hwPyru+>LDT<
z!x6&0U{uR)-EVTuxZ@Xqhy~+*%p6R55WL=Oy4V7gu^v)^9W+)d_SR<Tw(U?{QQdS$
zekH4FhEI6M3NIG9T6ArRA&Ev?kvdhcGi9T_^7Duy+ARSGE#8?^$}O!)V9&m`)a>Kq
z51wc_n-H)dohr+As*;GP37%LL&VAYVPvPx1W?L$qdb#Hhyf}Tx0<2l)14&emMYu%j
z?cq-v_D4a8<Mu>LU8~S>#{M`3`wi0mQts8C-&+nFi|H4sU8+*aEY5x0H5q*WDYGBI
zpsex5wwQvk&#QUEe}ByFyvIok7kgReXASs?-v%hnS?FtaJXmCKzmr-KbCf<<z#%_i
zSMLdSziRt^<4oaF7p)rp4!%0W#G{yQRRaBO+3ydQ5UG6<dA&Td&hI?2DE1?%>>{Y4
zb*JLn(jWh_E+&{^Gh-MB;L<yI-=EK#r66u8;_z{vuzm;#43{KfCld6i^`DUnItg9J
zzlNDI^ol{?Ie6E7@khHE(tX@v3^nV_;h_@lCT~_HhV9XGWYHsDg~0<D)~o7qS<Y8~
zUmoYF)Ac}bvEAGi2li0iU>+F^X#YC4^ESX~?{@J9islB>$r2U)p#1*q4r-i+=#?u6
zC#BOVYy1)>-eB<RZY*In9+CXS1%`V(A&4cZxBCl)`2#vZdX9Q^?EXt;UlbRksAS(1
zM9EtE;4I~%R^oeklNUZx_hSI2M4*z9sEI8L8xB-9SBGcS@zEG>Hj!T3eNgg=(ZG%}
zo_OByvCBUeKiq8=2O9~Kluu1$*d@)P_mrC1mz^R0k|pBNmdG)Fd|tY(z<lb_gHICu
z&r3!=R5b~UoN3cvoW{SXk|74QjpZ!y@u(#dt>WI$AIiDg6u3{%zI-0-JU!3p{7za<
z;XbZ|Dy~e9vy^}?EEIOy;FFjvL~C>kPNQPj2g9i_%HJQgcTwF+zkLDc5GR82!Bg=E
z@MBFd5ajT0hcq!TAvYRfM<1T~pj>8nj}qNYtcl}hwiQIFjb<HhX2c@UO4@_hOAJMZ
zM=0tn^O~bH29i!T`jIp7V4hJ8K<^X`rurj=-iLM=jN=4Ox4iH~s_JD0f)>Y<ireiL
zf~TSNz61@@iFBsb9M>xqr!Tk2I{s<&W)d{!{5jhx_UexFy&Un8Ot0Ydo8#s9V<<|o
z9_D?byYnJd)-aps-6mvtzV6>~rmTUYb-iBi2Z+N;eERLqj&ehMGOf8{LiABPvCfKF
zeSEfuQ{!<H@A5TYjakyao~N6CD7xRG9{1?i&zGoEtfr=$>!M~GD+wDagbJJth|*3j
z=hHWN)E<vK-WmJSYBmt5Auz~VL>)=j^nFGtigO%~QfEF!BKFVFp0z~sSzq+K6yKD_
zWM<C~*MEo|TU#bPFOUd!VoDF284Uy8W*b9JXg<2%jsvc~asyoN;Ij;J)L*dVueE}2
ztxHM6hsVY=8-9!X#Wn{S)7ytu<uX^o{Lyj>jbIk!Lba`b9RKp>SX)x9L0fd@!>3nr
z3$>hwEq#0*O&HLvO~N$%!W$gUu}scV>m7^Tl#WCGoQi!<vE|Ht-3@c8t2N`X?kAOz
zI&T3Dc)l~6l&>n^tc%(=-7La1_SSf;MB(XNr>%vRp`fDt{p@;Rx=w#&we3QEWix55
zB6U2%yv;edXlgWQCn27kNHP|^?}zc_R5tQE?;+M!deKMEi{ko%QRdP!`gIJN@K8N5
z-M)$Bwf)&|Sp4tjB?;+9{IYt*o#>zSABgdl=za@Pd0)#@RefOvled@a68xxahvOia
zci;^pvZJ3vZ4BO{ur7>7w|XvQ`w*9Ng)$y*@`Kx2QE>s5@5^1lQ@SD=5$21J7O^4s
zlN<yK)f<G1nX|>5r&d(;Q+XXfl_;0!w&DN~r6+<lFIlu0dzwy%y*)<GfIvnGr2})F
zGxVIS<N#w9#)Cod5l-j*Oyz9#Hcq8p$YQeFHhE<y!lCHX1+S=bptLQ%$?YU|i!S~F
zoSl^#45iRx8%=61>fJqd$-oWMiwyLyTeRhfWEP+30n38l%`YVqoxwm|n<gYE)4DzA
zgz^R1U=q9B!hXh*N6z4U=5>qVTXFld_rdrP)0bk#_-`&jZH!2NG^r721brD6KQC(>
z6Y&u)D1oA1=DZYSseCoG#KHsSr98|t00O~gfmU`EbO9mi$su!kn-dU2)O^T1|FU;V
zq&4b8-Oe-G&vOdhg<rZVeJvJv{^~6vC^pIpt*79OF0kJM<zFw<ylWwoJSpvUtNzH9
zAJev7j+QIynblV2B!WAGF>F8zj!d&Bn5*$>-7(vAj9<~SMP%_zoGfl|wDD2kPXZNi
zKCVs;+zOu%bu>mlz?%!{?BHNdOuXJ76>0aT<#!;xMmQDvVwhU~*9Z&k{?}@$x({$~
zueDT)rODqdJ|p)gd}QV;d{uiKR;rrmtM5Oh{>`SR>DhH#oB^~Or%JFTsOdwguHpZq
zhf&2QnFWvuo-KU2xofb=<exjXbB_Z;G{%$rdv6NnZJTo$lPmV_8KK;Nnfi}L5U-np
zu5{Z^!FmwA*J1yf*TxEiGW<~V+#asy?}ZWLQTWgnGl)9g{K=7IIOFxr!Aa6n%Js*<
zXOcz^8U6lwl!RcEKZJXXb`avp;rAU?OQ*BNxED>Bkw+L3V@DbWXMt@7xC0^*68^~8
zII*h(p|XmK$hCH+)KGgu>}R&~wWU{DX90oj&^X3bT%FuN=};mwQ7UYTJ>E1<vGOWW
z_TU{IY-l2b`Uj_>{-Xe__ev7R)Muf^RB@=&@V4;j8s6ja?Co}DP^1&ykcCtPo!OI<
znp1c|OX!+LrZi1lMOJ>Z?~Dk1<tK}0OEu;(40+Lk&;;!6N*(8H{-fuw@a9y$?s^m9
zahRH!!Q1HWXR4`r+b%XLXI>?t+Ln5W?5hT6$z$J5QaezKn|0wQ2d*{Bf6uKAvj$av
zLDt2t&D6;hFX_X2um@rO2-uwC9IvAD-}lvw2L@_V$G(P%@kFnqx{4qsMP$9JeMX|7
zF=bh5kgINe&W*||ikIMS9(dzIBu1S_^%A8H-re2Zh#=;WD(u6X(T}8Ira)XlFzJte
z=kz9OY$%Bl89JxWaR;{I>E3s1EW$83@$F4;)0~np@VKWx&zuU$LlVs-9LvmNHwm6+
zfg;hyWz)a+I9zexCasjk$jq1E&1(<sYlb3e6G{oVqj2DgC}OL^j9C<k3-I18+3uB>
z`b;|3@~nXaBU(#ak(2bD47x9LggiRm`~pm+H6@0qe;4rd;H{|ep8<+PNCNYup6A()
zat;!rWPh_7)==Ub6P`y-E}Aam#t2e+5u*Kwz^h=64ZPT8^EQGBx6FyVFpctQ*IH1+
zWf`2G01eVhz>7lK^@>3}{+lP;>9}Ixb*{b?sG_P0Um1>|b3B7_3+_7saXfD*_r`X-
zjQ6HWn@?2WhC<>_0$fdNLq~VGsgL&Wd^$j#SeerDicsPPFiDmBU)rr;h|2K4{074j
zblE)}OD!^$v4*_7+APf!MjUI7CdCM?w7dVz^Da-QQ88OLG7ea|;fjxbDtwwXSf$r|
zzpBIGzWFPaFxGa+a1T#YR3}2?(@>eB!Dz+_?-?&JyL2M`ew>oXPI&YIH>$O^g@|=I
z7oo)hI9kP-EG80mN~9I9+7;Q#&bXhqRROnw+VZ87|F33EJ>KE7x>yz3llE?rHKLI^
zz;LdbOHV5)`lw^RP``E{Xsx#Y)ja#L_J6ZPchp5sTfV#7mA-de4>ay;K@eOw9oJm3
z`<wZrqlc}iX?A%e*WGw@+2kwl{V?^L<=tAF1L0hkvWAr7ORr13ye>?y%=lwtU};gN
z{`j}@b@(e3x;;gqdo?#+?T38S6h7xYIGao~>f^<BFaFqZNP}^*4pITC0d_$=N@q#_
z;L&usrffjN=jji49QqD}G{W6#F>Wanl;uiiQoAt3=#l2<VI0u6o}lu6!bv(q#6O%y
zHW@LHg0b}xB(;Ic=rO>Q8;6p_G;hJ1c1L+aA>TSg5{6ArRy`l2n~1NNg-wNRF<F4C
z9e}M(l8<WDY#x_-7r&)AoK5Ta#w%_-u!<k6BRHXu&sf+Wtqg&By-0>VEszfO2Gz;J
zA-3Xbh(x=t$C_PTB#}w3;BIhU83>FK`8~`=M&R}7?av}NwZDkwKRhPEf3*Gm-RQ&?
zhLy&14@4zKzQUWenQS{i{qK5G2xr28L{|Rb^k{#pqt$4an?}CAeYE2p3yu7#4(eP^
zg{-eq1JS#*Hv6oXbHPR$yq?Wz5pT4HBq4pGatCt(pbRl#Y|dBF3%%u(>>gZ#l%Ln-
z&f}Qihbk=NJL4*<*6@G8pGXb|h~HC7C-GvN$2Oh;qP5Zj(H>Qa+*IGhI!1HWUb8g4
z^cYWAKM+{U4H#v67#^lBV!y)-@1#Vt!Iir#(vDF-2`j8`XO7B<Jgin6vw>EFKrObA
z)8~XmaM-jddhf-39AU-;od^c%x%77$2kIG$a!=}Tl-@WLA0R)$r>Gc;`QJWaHcSw)
zGnB7Ogu1M^5V{kP?J2vgiS;7aZ84G&gG}R6D#TT;OE3z|1N%s(>WA@HW05qgQ!hZO
z6K%4#jP88tXHQ4fc#UKWsgf6TswD14%J>J^fJ^VmqsC0pjY5U$xV!4<rm`(iKUhRc
z`&*4&1t8vU(i@uUr`-^jT~gDYx|+UBx*`6I5QP$Q_YG@1wGRWdr{J@wbyhY_v$KWG
zmbOvmR!W_Hn8^;~)&@;_TPcOz(#G=EW}px?aIU~Se!d5^lrygiMDh02G4Oy|c3-za
zv3e!iECHa-zF?Ci;&m_;zAwI;)XKi0@F`aJ5ztVsV|uSDSYRGKpB=&4-l)|od+?%-
z5vVAN_ea?*q?3@gClMPtYFoS*<5t;ceZ_TxL-9PZVo$|K^oXIw<v?XY*TXP3NQ_cS
z!_|J+!`Vu)qm2}GbH}Uy-$=Y&3}X9PwD{}*kMe?oiEMR=t(lYwiAOOWua28Ax~`0`
z>^*Uvc@|P!?eC-;eUGC%Fj(>njI3@?!B18qo%zf!vFmxOe3zGr4oY-Ob*>6LKKOvJ
z+s_FvkZi86RZv=V!UB>8a_x(WlKftBouggqfkJ&NXoYgKlf^roy`GZ$;_cwm&0k@s
zpTQmK+*!>aCf^>5S>e=`Q?ls$L-ftCRuh2dYE8y-Pm?y=PPHrDN;qdPSC$$U>3@al
zNn9#zN{=XR_<ZKBBRQZIIYN3XWj!AHk}5~L=F?}vi%{2HFTWdYk7ei=;SCe)-XRjN
zja;sYNyXvr!8?l#*@MQ8)#NO%@RUB?pAkcHUXZPbAqng6k8Vo|&je#Yg=&zZ*Gb04
zW7kU%PPeuX`sO=Z^r|TtN&Of4hTX*d_DiF(pp~i+9M`31G*WGd=0aa)x#d0+nXbJ1
zIXh-5RV9Qe@(?f4t!K7!Su(!zJPQd4t$ks@QKh;7qhAA)^DW$!4FzVy=Op3|Lk6DY
z$*SV@@;OZqj|>Sw-a;?6U|#lxPtK}-!&1+%d|utD{{`an9m&9!=ngp8{7mWWrqwa;
zFwJ`|P23-aBQ?H#%9VLLn<N?am(5axXp}o*TUEe{%(Ft8bfd8F<my*X;3JsL#_}Kn
zcof2l@)9&uzJ)n4$*t2@;I0T7g3nxWIXYbshTtse+f1#!yS}aX%?soMyZgz_^ObAI
z|D01L-4<tk=weSm-xl56pv`#Jc?agRo6j>;Hy+-q{N>F;dC?hjU*t*;$#1}463Qr0
zva^)^b(%w{E+SI+NzcRH4O_S?#ot)YNRKDCV)bVnUCE&j?u$1(e)&1{qJ-p0<aDZ>
z8zLf^$2u9mCt4Mbs1-udHx8!DQ<e+SY{nQ{lZO(%el5)*l<FF!T5=n|Nd>StM+-e3
z_)=miPRQ!a^n^gJlCTJS{F*83bE*G2y2TgzO->R`YJ|L4$y7Q9!|#|#4rt+*fhWqD
zIebC7K6{c88gsZAddYnO>&9Z|;n)+#hf9Y!csxkG!D!(D)8@o<7kId~jP%JW3%HMR
z`8&#KLPu`OBl)WrhiKteiC{pbE@G7Azi`JuZ>9N7DmTby3N73>&|V1WqVPudf=W3I
ztuy?JXSbm8PSq}c_R%5R#KT3olg7FMNTMPq?PKq%5f+!VQdvLGM)C|XhO-9RE4h~_
z<P6AiGt?4Q<iyJ8C@NPJt{S5+xlxkB1wHGoR&L4NR*Ge8ow?1LbPUsQdN&;GkBAgR
zI?<?9vE>*&?}p0)eIG&pM@0#`>6q!-t^`-EDR3@OxaF0(!;Kb#tM@$j6?r%juAUo3
z)Nn_n8o3fr(JQ*T+$n|7(?E11TwE8Zfg+u9Ep_X|{}LjTL=w5F#)fOg-h*!|^nS17
zyifd0FZZ<+5qbD99&Yk3Uh$Ox+RsA!+}~0ner5i7_3Ovg<UV&NA(WQcB^mUlN%}<Z
ze1=+tO!$h-QF|kUCnVzH^4_)25O2U5<r0xk#ZH=Y1wtXP-#WOg;RWJd>OvgH$D<@@
zOMzy>CyB0XODptxno6mRkr9MzqBKGCOjOZ*gpIF15E9f`HhvtH=G-HcaPqmkhmMFJ
zw2b6R8|1P_JUVBYJk}@QwK(q=t*$d`C!(P-<5fOH_r23_p%+g&?GDOA)9K;X6CmPP
zGZqY*7sYZsdOY*C+$0S?%)fN5UYU~GFjfib)6$9V>uiiwB%+H_jpQqvtiH?nJuzM*
zwt`r@EY!Bqj}#+WgCA$t_*zBlef|bS4@%ccMj#?Z?UO56{pH%hs6F)EG?iXPQ1D$w
z8ccN|EW#hRR+u?5;!gQ5Yg9m=89gq6K}5IK(vJeypN(+R?ti|T7r5s-D1VLO(er&h
zlHdxKe<J2`MyY5LSY&*k8-VA>^+~Nv{|=H`oi&8D(r$iuWY8Q{EvfPV!!;#O1}q_S
z1@La|K4yP`hB_2jSYI2~@WF2@#JKmJZSnm(Do!+!-$%U5^!RI_x_v(;w3OxyN>?4d
z906V@ZEe{te|&1IR_D)ry<qvE=lJUb_JpLN^p3lKzT18OnD7o3ObYYt4F0nPD$}uj
zlFR+{D7?`d8L3y8vzl#hI>ph8ni}63OVRj2cH`b}G>W)?gIjf3U=$aKkS{5K-PV4Y
z1$m)6f#P1h!tKLcgQ8(JZrq3@PrR<}J%7R5kRgWO`_<J74Er;^kNo)Itx4!Cu|Agn
z&hoR4C||bAlaS(|e1+5AXKS0pCL9k$l2;r58G4Cq*Cq<vOujS?xwOSjBC@BpIc^{Q
zmyYNT{WT$t3vUE<n8b;B^jU=Py>4wlme<QKe7ANE-J0cxDs_Bo-eWMzXzXqJBmi}n
zc`4ti+62whR*d~i&Z`rxIxZT|#!4NH_Gx`y=ujydHjyd{k$Ain?}yrRm)z{gWYb=c
zfB43xkY+{w?1xJn_HmvBW2sx9xD|Ch+6=6QVO`N%Qew#iV&IVlLyP0I<P>4RZhrOh
zxE=}<b@a!poJ8%7f#Dy=YbDO(<VPclSN9@w@>=~IM05aYgld=Y8|1XVSVcZLcX?a#
z7GFhQej_%@!J|!s7?f?urm#$ZXul#6?AKW*8Ur=%O8M3T%pc=WX{1YRe*(=9Ftf_8
zE8L6y@Rte?!i&G0ku%@EvPm#IpzjsSV-kGQEu-Ma(=3(7$eE$XfBxPRSMu%Us*I_&
z$oU%{SqgU8(;=FHi(~Nm31%QZe9(z<>y*LIp*T2&S}ngJ*}T?|nO4fgf~p?I{MgeT
zjUJE1y}G&0KKJ{E-8`kF1iTz4$D6Dw5Ceu9Zq1_=>MGIw)+}m_d#T}EG#z#PJ{feZ
zQm<J@>hEWxAv@g6HXPMFF%zP)u;vqGe$j}cNIvY>;inQ292I{B&|B_?B}Hv?KH`K=
zVhc>>y%RJS@08u(sJ$yEP(N`eQXTGlRr{h}dWq7n*y@Yoy1q`N`4p`CCAQD4<E0Sr
zc_`<d+MQ)ZByR*&$neOn;QQM{C3FZ0miQ}=1gm?IYJbb`XDTt;HhKndvO)(n_b1jJ
z!DsdveWprZm&D)*dHMj%Rdv8R?J9B9;9-Ay?Spc6*)hK!Vbp{6fEo#Uo*dWPVig(W
zQa|1;Ez+&?kdSn2%)t3o73DOz3-(_3=kjb$xy6r)At#zJ6CsAb_*PPkdoX7$kjE3a
z*ruy1_ulwtQ!|feeSK)F+3;_lh)9I0RBD;apPv$hj;8Rf=8S)ZnZE4I*u_D}CdKY)
zXx6fC!7k3otFFoB@~bh|_AQYQ9iyf43qaQA$u3)xQWE-MlX;QXXKI>aWi~o!&*;3j
zY0)m0$dFW?@&1>!b@em`3+oe+^!nQ7&W5--wa#Lag9rIjn~TG)%FQ3#G&^36*(6pU
zv~9%o$My!}Ae=hA8<X;{92=}W3Pt8A!)5ZF(Kv+3uC;jUPo?xW+0a@x`WMO#c=$oZ
zl`H2YXwSwGull<`LIqWqKIpos^e5Ca6&^H5U{=C~wMN_v=n@fly4~k6M~T_jlCmgB
zKhpyzS#@XhDPJ4|vRW^i@IS3zM(@ST753^YYU>=?FeS3~Tw-c53Oe!+<Rx@@Z$lzB
z*{y}b5(eOvCoaCz&!CmP_n;K9hX)_a<2lLAT`{x1a>mAnc_G4laB@8^UQ>#Jhm0$u
zD$KdImqDRATmfFE2$IoBkj1eVmXP8LQQ<uI2{^-^$4)zR{~n#?I6t3OmBL$lgOnJe
z#fK}<>3P(KJe5W0+_2=5?qunyHv;zr)+bVzUK=kf`<|IM044r7k46^MD$*O3{P2`1
zz%}5op63*ng>^e>oE217S;#t2VM=&)S0>$mlYr3@T5l+qASk}SF7UzKV-H6TeCTq^
z`>wo%rnRsAFB*$kYVQvzlE>%!9PhzUG1(AbV;7@w=|pA1AE*MLq5}h5VA1!kJ%Rnj
zez?)2dlAeT#deO4WX4zy8&{7i$yArDmYVC1Gm`1Tz2FFJPy;~4p5zWWhGMM}IyC70
z>%DZFX4DdcXZv<3NI8`qz*kxdS1Ehzc}tRiHM{-WQCN3uV?Shj{Me^CxE^xp+9-|g
zad~(E{^j(r5Vimck32SP({tcH`79i1zb2;iL$gZn)%XNeB9r#pRX*HHv^iwNsT1Ld
zF5x2)$$l14-yCam>zmw^3z~BI-aI;{_4(u-?7;Gva(1q>eerLcS3BW%|AkUY<nX-=
z7mKJD_WC}s+vXZxb#j@{g-Xy1r{i-WL?c+lURQBA>G?uuCdpu*S7d{;7W<Hw=|Rz?
z3pf(sqnxAo^n_7Ca~$ewH+6oBUE>Q>B_Ubs<^fm>YPh^aae=p>=S5&YnWvmv*1XbJ
z&&T`Apklia%wMjeYZ8w~VJ-F2{bX@&Gds6vEWtq<l{yw?J=-MH2X9o1;LsAh?sk3K
z*K+VSK=zDH0T_xQ9XRfWUi+1pM`C}|y)Y3lA0GHShil0WO{DyXiM>_%1@}Y-zMdM=
zYqdZ!ZZG}$w{4X&Bvwa<k!IHdl&+RM?6N}*$+Nh<C|COneI3l44xhbw_ZjIbeNgP%
zq1&E83Jd1AN8Qr*uBa6%gpQ-M(h7~i;S64dFPib&7klLg<Bj%K9SB2?I1jI8vzEH0
zRX)!m=tg$9qH%O*vD1CG&qwc{Ww|>f`B%UzLi^DJLSO(5xi~n#^_?znQslb-73uJ-
zrp6isThcx;6e}XgWs&d%Pq&6D{%5vj5`QhY6igIt)%f@AiY=351V}-!OkribBU<EK
z<8zS<Vyr@`9QnZjDzU&-cpGjWD^xwc+iAYm9sm{gI^2K~L}9fuy1_^xei6wFf8XGj
zYW}7X5oi$@&`QR}2sTc6gZycOHD^#Kgn#4d8UOKlWwIMEIl-#a)l6T!&{`FtM5<k8
z;6d=!6-2vl6SRoe-Goj={DkY``+7{!@M$U|qd1JdcAQZTv+{Y51c5%aK=HOrf@-yl
zKIlAmFwP3hqzut`<tJ{F%h&v<rVz)Wf-FAUyAUJJ3w^1X8f974^o$H-y4QHAvY@+I
zJ5B#w8>1Q-#Lq9W{-hrEp3fSRkI)47s0h*;<<5}^W_N|U3SQI2G3f-Kf(3@oPm42C
zuGO$ygIRdEkb+Krth%H!4~Zm!RvwMlfJ5J{7f%y$$vp10>Tp~RapklH5-N}DMilq6
zt2kt4O4M0pL-NS+D8%>FcQX!Z-y*|Z1y7uI#=HJaZu?LV?cn6fV1i-^#M({<Evb#y
zwGPa!&nx-R!BdcNW3Y4_e@i!RSaP_PVu*h+2I^P@b7;1d@-g=K=0sD0Sb~ffL&tW~
z4q}^vA-#`-<dtgp<PqMnB^sOt6mgl*lN5SPTcKJKdcBe<V7%Jpu%abfHbcGS))7L&
zp1$G#oL!`tjaj?Oh#QjZQiz@n&A<;oo=?ES_Cs3Oo-i)rvYvO#L%CAiDv&Vf;?Xgs
z-)nh;Szg~j?T^R12oq;O!wtH;%x_nod6Znw4Cph@LxMb99}5-r!b2?<S9ArrCa}gi
zC>?;NUIjdWus8BT*nAVl#UBX*EF_2M5Ku<4g@ig|1fT-ZEypNhyY3UNl#3RWuL+si
z?!bqFaMxfpN2zD`!E5~q1l^6iL7^M^BFsFh?+>()KlN{kQE6pj9lT^#T@H?1JbL(e
z>~5L*Y0Lw4yw%eDrK;6nJeNG5PlCuRbRHy$s>6gXbnA%v2e>BHl|WT2%;1h@11z8V
zCX(nnD-<7t1;S`o8w9$eU37B;FA5gKCwKG%9XVoA)Pd;C>Px=|+J9Y9Nt}*}U)`$?
zLB|h<O}n-y<QO1T((F%Uyda!2SsYU78(x=4EEVH1bFCB^Exkv_Z=A(Eecsv1@S1xd
zx*R%fT=qlsz87{fH}KfO3SCx;3COd}Az>eKe0=V`KmH;*Xs1@5vupZ-8cSblsd;d+
zP#{xlC+#V)L(EA2UpBBp!gtwB=$?}pqw?C@oLwh;cM{!bz7~?-C69A>(7<7@+-^}#
zH8Kz*)Vy}s@-J}Pc$fFH8ro;C+I{4CrNF}_bU5zPNV6FpX4==(Ir_RmDXGJbpp}*)
z0;CqG;><W6RVE9t<o{f^0RJ8i)Sr$-D$g{IYFODL8j*u~;MA4c>YcIn??=8k8csAZ
z4b$2Txe6IbhZLE+_`b@LV<V_*-I-j8gGzB3?UQpYZe(PTytdSOz-cl)0e25?BMC+O
zA*N=cQUoI!vciZbgIve`g|6^lYNtMCiFduL@C4WHtI1^m*{OJaoyGt2pg8D3QqXM1
zE6RWsT`E~dCf1Zo%37-An$ps9Oz<+}yR>N~pU%2R+Lx3%YqbXPgo|!DG=W|BWM;zG
zj9HZE8~#wa0||)Rql+AC4}qSYpV}9Nb=zuh!U`&QzY@iz4xf(ki6+>U47K9=>BmgD
zJBgE6B%V1@;z}owZ>1H;EWvlPmxUU+N4X6#4P^4zLrT8QI#l-x;4!GddD!&Wn@TtU
zXHz$Q_knF-9GZYx@|IbPC`um6F_C~9+xRFhwbCu{&>Mj{Cc`!WzYD#tXh0S;)Ngh=
zxnOu;)`im`5k3w$%p9<YDj)eNmUyXj|75a3VWolBZbXlOlCV@uKY-z5$)#cHcX-8W
zJ=O&Y@qLHNi{w7nv%k;wu`AnX*xRI;Salm;J^yC6%wIQ=d+|1dz0gXTka>L40gNOA
zj@oZoUZX@gn?|@IK+pwtmN=$PQSwY!g&@>bK$LFDW-U1i9Rrp2z8%Uj0296%_({{q
zE)H)QbTw!|3o~5IQ6wS{Gwo`#-gzJ=Fn>8!Fuc&^`ZM7<V|wyKXqA$ddv%+#wB-*G
zKMUd+9<zHz&JfE)O&&uF%gn=J#{;7LWFEz*0gY3EfkS?tpXEI*OeqydXfDe`e_;e9
zc&sR_r-XXUY$SgZ$}R&9kCx{lu}J8@{nN2pUtrEuP{{V3qKfEjD9*&<^v0DKe)end
zh6W2_2g$!pqN0=P(r-WmSsw`JY8=FM*}|bv<n0m(OZgkraRm$yE<=#atD&DvbY;Ex
zO2o~gKACsD+=f=6_r_3<@Rr0e{6Dk)08WY8V6-GsjeRliiY>FGYV%Qh&6VJs&QVK>
zezHHmM~IEwIDNWZ53?R{x>jDH6&$z_A40!k(p!%^Ei^Ee(9usl@1grD%!Hz4&{<wr
z*GAR}7Ic&VPz9+-FEgVz0AbaZ143m~bMNcx!9ddPK<cdO!OU(^lc->;x7=Fny<i~^
zN-l#!o&s3$i`oASnw?6ITeG$d)+8W`*y*7p+LF;h`GdmMH_0KgZ=uz3fgIF2v};7#
z+~~7Gy2#z~LKCyYdkETH@k#Ge1!B!;<528DkHoQSXpKJg^JG+E|G{*6E1(hwUgHHt
z6_RVOVD!h8gx2p|xK6oL>eLOb@Kg3vV-i$S7J1->C%t==i_;RLcH;9mqqkb9|0UE|
z7`^)yH7Twllj_j)nnz1&9U+d!x|&2Wf=1L?Q{P!#I^whGR{od3&qxZiqEwcJVLQ?z
z1Z9?sjqExmFBSDBAVkb``?k_wsLD)q>;XlwkDU4oUY$f9Pxnaz^D#vik@k0t-a2qo
z)~^sYvwU(aEC~uGNRns|;quP=FYYuZRnZOq$+I2?2{h_DOzGLLQOH|w^e({+-76jr
z>3^{R;Fxi7d0*KoYkehWwxZY{q$C*SX$Fb&JbLH*#<ppRzy(SL^%Y*I@^CmF<2XYn
zX9l7VHi);C`}|Nqnd1>i-PTM*)ZR#`+sHe9@CE|R$)!rn`Hk<>Sn4<a<gI1h=jlLu
zqVs}FEf*NYchf;^qn~G$#lS$_?I<Z|@e<{nrcxI)fb7JZK|Oq4XipV-$F?KHy~t<}
z8ox{cjy^KUXc7xfUW-Pkpk<1_i?-I$WbS**3I9`6Kzy{0fF3L$N897{LcZOO<-hy}
z50bOFBU4DR^T+F<{gni{PmE=)3nO7RVsEUGy>D$MvYit4rzn!0viNRwT<%x9bV&79
zM)1-+4GtGp2Qq5_qiP4!fF_KtZ8GZ2HeYNsL{pyVNqz{BgndZfc=yuJb;Viq{^t5S
zG!7lfbay;Y;v+dU@2`VKs_z>VC86<_hWhdQGYh3k?R3hXjIc)R5`f@$UCdpo_dfQ!
zxRudE?+p7d2Sh2JW22AkeKh>+i$2yoUQT<hZi60=_ab<QKo8f(VNm0*gux{Qtc)Q2
zmh}Y!^%6UUpXJ;Kyq3x6XUhxn(djiQaEOt=M$>C^Fo-x&EJZ_TzndkiDP##)luVh1
z#a5iae!h@DoGRBG8+X?6Gu8DX&Xj&qjO{FKvA^!2Cba#dsR=AH_e!ZijlUA!G@VHl
zH>$f!<FaNDXfx^w{$BA8jcWS1fO%?go=_Xfd({uwj^;XmC^sq_tpSrz@eLP+Szey>
zH1fQ$HZ_B5ZXs?NPA<=Uuo^WQCJoD@C8fH}qgxLz?fw!VU!c&TO@F{g<q`*|d1uO_
z@ArB74fFxkVF%EpH&*n|sB%167V>mh(Y~%S;GR^`G3yXgIMgaps?pUj6cJ=sd~wn*
zF-uH}!AGD;qZ*E=K1qtJj^8OS_=)><Mk^GtwtD-m%7G0?LIfqZ4f1`5^4m~tVbDHD
z^)=-eTy&b0XEPhZjnyv1pe4;WlF_5_Fc{e5L*0#Rj(LBBqA>Y9FuP?Vn%|D`_4Uu7
zIkzh_p#m{&(jxY4aDpUL$#@_b?1NR;-hU$?b@XJ+!{!yjp4P0?!JE%f@&L6+Y#wJj
zp-dk_XZS-ZqF>R7;<rmElw;)_Q;*>Ce#O%OFV+gU?=l^v1GqLo;7lQh>H%M^#IndW
ztXt-W>xaYEhg1Rgw*tU<Dtup<%+~nj`c(MHe+n5!c>wwkkdun&?eYXI^WIjk@Inl|
z3I*4w@Q*)Fa(sW4;)2M|w}10bOI)=&<r|7waT|K6^PUu^L4MqpLi%d}@0hlw|1F!u
zMI^7+gG9ATGd$j~RKyJT^_{G)FVSpwnt=U$_`mzhIy&at^T|K9r<Bi3$C~~%zzIzw
z^w+&{CJrC_VJrHA=gV)5hGiYNHV-_}C?YOB!X!WRXeL37uP-ix&;i*U;n_e(fcuA&
z9;|s21x22R>?_2>HqDlHCfy>z2k9@OdZJ+Z9UmlI^XW~a9h2_l+os6ghkLZM-GmVS
zqajzhIDKNp$KtG^&dZUC*h;K^$|Fvs<p$fV@f}H~iOT}}C1*^xi#fHi+_OhODTQ`M
z^=qpLf1%t%WV-(&x1JB2DRSbYCBB+Y);!*(esWyyHfpgV{!DS9(e;@Iwa~cSzp%|6
zpimIJNEROheJCi$mlv^H**|3o3`t&<zoO20Y2BbOqXFu;>5l$t9dc`%BwCl+)MBQV
zj#Cd@UvFmzX@@7~gj2@U4!i|iOV%%(8yAYhJNEPo1`+GYOHUvdNn8OdA#G0_nZD|`
zgA<r7O6=jUNzf-vb9KOL#U^2Ne-9Axk9J)oAv%)U*YpQU!-P?%NIjIvMrLDjZ0<Qh
zuPBRYmMP(xQ7t4`RIDLkI|K8X1HF0jIrdthNl#*M!{ryR9XPy|IG+LgM!$8ydrjsA
zGz6)E1h;_-m1i~o#7ZLiM*66*wam^#5#vfrZ^#KFH$^TYCoWEO#!{IMM@HmO&fDZ>
zmg9X@T@;eBVtn4#HDtXFIn7>9hc=+ap<iZ-Y4nR*gx!hmBG8jzTZCs+Ea3+wU4sya
z?1InSyT7)PaM~?+O?LZn^rLQ9uxRY+CFzQrir#>Hya{T*tmITkln97eBkIgjubJwk
zwFc5D>X<+Vz5a%_z7WX*3JAokxC8yowB<8*GgE*5%Oi>X7QYDezgrB#pN?9n5~QCS
zqk>g=Zg)~j7w)ZX7EXA0g)+H`FLwzwKC8<brqQYF-gM2{dfZR!<7SsIq}J%Px~|{|
zoH-g=Ka0r>JD9}vchfL**YDLmDTYNG-94S{tn9L5g~pILit*Y~(jCQNuUlqx+1e}W
z$)-=gNT3IL%cBnuEfA@g(<|n$dx$AXsMH*O7YC|F>pw?gUwpYP#5q@QDuMl&=?+=2
zRSa11W<+IePlb2h{~SXu3c>M<;Igt?tMXr8(x!aTwnh(ThoDzw@{>4MZ?iL_aDM8d
zE77xIc&ZP=Ey0qUr2|s5H6u83A2?sqIhZza;2(nfSy@o^c9y3?owv*aKh~?a)PD<*
zpA){0G+AzS7xnQ`ocGtE_HP1)eAr#<?+<x&m~-OCc$#{-RaH4MWp<APL(7FJ2759S
z+mE1mcgNA{PXMHLxC3x&Qh{yAV}-SOhjO|(U#`{YPxZj5<bp@bk{>zlKr?fga(5+q
z*1pBCa~yiz`@qj;s)@)x9ZE>P4=3WMjgnDBFT)gE$N|?GKkn)=ri~H4<@4u-G2Wxa
z1=B2I3l@nF67er+Wa9F=j-}RaUoc3>B)$k24VA+arO16wqpU{sPBG2mOFzfm{&Ksj
zxt3+7(GiExnDeyf#k^(f<-F&&y4kkFe*Oe7swZZ#Y3mW_IB7jO!>I95w+^qhb>AaA
z>iC4KqP{ov>5w>c=u?4~#Jr^Xtx}Bi0%6pJM7V72q)5|fMQI<=>#h>HcMou=FBlF!
zWg}&sF;Hq%WT7hE&Yr`p7wTcZf0&elUlDtS=rN*SaRh5r>6QAAsx=OA^njwC!m{Rv
zAW-8V+nAnpKyGC<f@-#*nYZ?GJeCbX?Y<GDy^_z3=n00ox{dU1z&RvX+k|ptQ20;u
zPc<*z33KMXScS0#CHDSB)O5sLCqBDX;m;u1m79o(^%llfbL1zz4DTicNI6pTB-5?R
zMA|q|_4jF%6Nk853d3BOAA3X#h!NH?)(@9jb{A|r-X&J^JMFwyV^g5Kiwf@DFo=-k
zSD_gTLf2o{ue6chpLsRJo^gd+QtNdaZvyLVGEb?-Sj!kFQTg@y=o1RIZEMBHbslW;
zYDhTi<o0p_CE4B$6%idt_LWvTm&=;Im^|EEOA7S=;_0lTqW;3IPcw8$*U(6ZbR!@N
z!T>{eOGrw0cZj4&N`o}g9U>_R64FQt3@J#y$KSp0{mZpjI>PXsbDr4yvxnm6{dq6T
z^{{Es-1`?kM&(N8#1oc1@)mS3WIB2qpIPLVb9v-JUD58fq`}yQO6akl%8*MgX0cKp
zSW+ZFw<N68!v$1^yQ?@_rQ#86)4wQ~-VcU;$9%R{QF)8u5%`>p7oTSCT5<iUQZSJj
znkEKdKur>vGzhg9331;Y2u2aQh$=MLl+Q5;fg^3dW%J;Cn5os`b*bd*RTC5MQXUn0
ze^?|?4QJcW#yTz!?GNG7=qUiK{$+qw^UvL7FLwy_fybYnJg2scA6F!a9v;8k%j1Fr
z$&%L&KZo{E$FW)<+ky(CG4RfJlYU0~gjyl??&k7YF*c|%OF+#z1@K({r+E@y)!r`7
z8Eq1O5=DJZtq}bmyMetb;JkSsbCFXp^>vzCCRdPpX1w=-7<p2+Uh+U_Ybn9n^(H-p
z7!pUSFG=_MSysh85~EzvuQr!OV!NQ|A7|)J>m+aWEJ#Ci6<b0Cn2*rYeqXv&m$Ow@
zIInk*Ry#>;mE9{XHpBl8iW?O>*_P0`Aaay#Jg86bC4}aOrQae&YZ+JdNc9H^iG_bI
z1CMd2?(2DMhM&nF2llg&6y=kT)_zE2_TKbW={Lj&MdM_w1J@EuFX`>SD&>uqPVKF0
z7x?JT6e`9c5fC49oB-eGG*Bj&jBnpvu8KsL-@k<qm_2yML`DVO^-1YmY=WaF@tAj!
z?77V6joeFT-NH`w+us?{UnmR_^L721re=xjS?+5h+d0qHYB-c^KC3I7;v@NxQ(RGZ
zeAg4IR@ad)mtl>06aSJ%Kw>jqkSj-Ic!ZB6(w60j8{s#8mSD?E;ly9>z2v}eZ=gvg
z=$sqKuu&K{6UPc;TdHF^ro@2$IaI+HHEQ3wbQf)!lq@Bqy*xFvl2Pk_h0oRx*$@|Y
zF51u+M9~b@`|z{CUDlfsQ<UaZdaK;sri8Z+KTcoD*Xo;=dU@9FJ617te%|s763OwT
zh1J8K#uk=Q*MFhsc;PA<QYWl7X-&1zDESbPhIF5#STdb=4fRMM9xw;g#&zCFciH-v
zN0fjTwqWrt7$G<mS}2O61bt1|vR8uZyN0Y;(d_(ly9BU~r-kN@CB$BfN0b02ZP^Z`
z@@H43A%=31AKzsAN-2<)pyNcKP>}?^AxpUBjJqMJ!tn>UyNhS9dPCjhRr&tr=9K0+
zvE7pR^7Uo!bSrEFa)~^wKD9*kF{!&2guB9kzSG4~jb5NgnVmT!#)N_4m;cCPVP8Zi
zjs5o$z0b}!ukh&ueihl~ag%QHJ9Jpkjb{e&p~w*`?@x7V<G9N4Vc(pKtolEN)kaV+
z0>|uAWk-5k34#W!{LtA3TXiaF{}y!}m^`{hVbH_9^!f^>UcCI>9iF{0h_*N<z!!~-
zRg>gi*+P{e%wNZT17h7Vd_-oSx3&GIMf@hR`wQcak5uF|ZSZJ{?Wr_W80s`<`O81;
z;6TQye(jxsD1TNI#N_d^nv<6-`^lTFo4kg$gc#+nOJ{dRn%}voQ&pxNh_8C-GFHQc
zi;Z^g3fAcHOJF+z!2FUB1Qjy%T(F)gRv}-f+7%aY?=}m|81n;Y>JF0Bo`1qo%qPg%
zQa~U}Mp%ul{6rFdfy*M9ey6OHr#vF^-OI{DRu4S9@QZj`(QOad%pDjXCHSx@x1k@O
zJhJV-qmFw?az74#yISWn!J#sPAkb;zrTbe}ZP+nH%z+zt|JU}tbU>SH4YzR{eE*Yj
zn(+Jem`JTBKi}i}OlFtrvUct=Ui3|#d<px-%w~9h*DNVH#pjv@UmLIT<DnHfdRd>r
zNk3OY+QF*3I^Nwbv%olSMKL*qW|%sfT7`YSpH|Lt9Qe2RyGwt(BJbyKgqe~H<;D~p
z=bG(=>)%M6<vAa_TD0QRGtNzP7m<!s3}Sa^nx*i4+j5Ib7%cAl9sJa>=%yLjxLv?U
z1dEqNR}7hm-SCIfc=pe|%9<H!SyojSK9Jo~V=P(m=Fh^l>w;+z%NKt}x_Q^EcqFFe
z_YgCth1AE^yL~tW-`@-EdzDvQTn&H%8+kn2cG=5XTfYt>a{-sQILfiS4Z}YALZ2~R
zS(6SL+iQfL&fM#b0b&reLQ|osco#6XFJS%G*w+q2>W^OfRjIB49bY~iGPJO_5w~e*
zs!uJ-<w2ypU-L9Zp(w=AI{DF>4rQzjAnbL>Sw&;|%zZa5Fk*FAiSyle1Oj7Ck#s29
z=;_%c4s)ls0rLPV=RQu_GAPvNVA7Eiwrdc@+&8QUlsT(I0@&}^|9SMl;H6@LEB+j&
z`mS|r=y~N$=<MGJ93*7{7m?g58<qkV4SqAtr~cKd9YeM!fvf|*x5$h<!q$rW($teR
zz+f(~y6E@at~AGn@b$yBO4l$|727gM9-Ov)AdtD01@J^|R?{wju{%&?Bkusyw>sOd
z|8vYDXjtk8c5meha!taSrysX+n2NBYmEWFQ2L6bQtrj}ryZ>iF$Gi9qW>HgDC!2#`
zJ}Y9$ttjGbUf{!s<UY|L^s+L9IvM`DYdk(?Mfu^)8pa%SGDxlkAYr?ErFpF0q_LOY
z{Wld~eg~daqdwe|Be(FG)XH|Uhfm^o{H`GXQP5K<z(qq!%5q=W{v9nu<o9jKo29Gf
zO|=ydM+5aiQ}haijqoc4u}?g@#)aW{UKsju?})qsW0yLdLHnaDcJ7nM_-;}a*er5a
zKb?P{>@R5b6{&@K?p5p>U6$T9CR0yT=^F+nK7EIc%>JtBwV^7L5)Bi=b1#MA#}_>x
zk#ZhpUEu<vGDF>Ne;J4LIALd<BpH%6S0gbMqAM@Kd~ua^=7a=r<5Pq*KFzmKlnD5v
z-*r4za=%iDYJKltIzL>KH@Y{?w%<ykR{NdS>sR#WG!nF}daz(gKmI_Jy_k;Obr4My
z(_2X)DXQhoIh)jUe78A79cQ;QG4Q}jp3OO2Jb;26P!q3+r<spS$i60gpKj>HP*$g4
z4;vrc_y%q`dOI@dJy<D-l^S5f#Vp$d<0_OsDP&mEg-B9iO;F1LMRmodT@(R*qTPvv
z*U=*a1ohqr_lW|4V|(wWj#K|4PdXsHs`FcuI+4nc`(SB$yf~dMo8R`uzPT?v#`n>;
zy)U`V=v2IBY-;L)d@F0Bg6?BnXku`V6v{;A7zEHZ?o^{Mu29)M4j0pOYD~CUJDCQm
zXF1*?vuTo&kx|)qfgIW*+gu;k1Ap1Z{Uxz)#e9$CSN#xD)`0>S#tK6vb;?{C$I(Sc
z%3;!v5DBLLJzh$_gZ(O)0sz3qlrSC&1{@y&TP`XiMAtDMdGYgW6_DoCuKUXE0si{K
z`1p8JmAR@;Gw#So%5q^mJ)h-7GO{6pvO}QL$QJiJV3)R;#9^{Fn>m}5F)jNpMdcm(
zR1-Udlym6?`IlGhk7f=$FA6nk15}_dSlU`W597;X7sd^yU3ojy{s{afDsC3g=nW?i
zCv=1Uuto$7k62iw>++%0>^JBlR($2bSrl4J9<4P-m9BvPNpGFF+Hh7cjT`5(ir;`c
zfx#)!qq4Zz8eXmRx#Ox$HBBcTwO@Yia@F|FijQI+$uR$h+07qM{!80aCp}p=D<#0a
z?S*t~yrX{kA=7uD5L$a|hlSD`X}!zo2oAL#ND&i?ki8Yfi;lO%aNDtOoQ<p!CENy9
z(~Wk2&v){+!b>NrGN0H^7ZI-;KN)t~q?kcl-CYR~c~;d_);(K5CE;mC`#F;a|Dr=0
zV|I&i8iIUB-ueWEF{LCl72w+mDD2zc7JVt40Ujci1Zt5Iu<V`!y_S7hf?Ui2_+lLM
zXz-Kf6<<!^ZuUZ6hph+fmzSdS6;QMTV`K0!Fm_PK4I;BKBXeAg1C&FqMIE8nMhx2v
zd(xbIz}%S9)-`7FY;#I13l^LZ?Qgj!H}Ni5AEJu$gL;P&b^x|)^r>p2LdAXXv0m9!
zWaV1MtE<5ZX{fWlbre2rij8&CPjl_n(BIp7Qke6D&F7l#X2cA^b*dGa`&lrZs@S2>
zH|V54y(xxt!!Ym)f!3VL@*lop3bZ|gMTP~wMdi%H*#scr@Nl_(4vKAGi18sbr6xZ4
zUr6k9KY<JlA$<VB&h6nZ5N89lJ&Q7;zcv+mnSNpWDUOYfo{68{`HVa+jn4TtWxA9!
z*tzjjKpGo8gpnZw#o}hBT<d8J9U<mqFQ`VbGSpwj`>+Z7gnARCYz)MZpl+~P2HPKy
zV?8#W4fI|%Zy*EwLosvS-0WZAuUmI$F)@^yJHcv9q%0Fx>#bER6O&GQfns|1<O_W>
zndAg_0ij-po=l0;l>7reLCJgcXSy3_8h!68*C4Gp5pb7G4DFxT>YcKms}A_I4Ct5Z
zY*<Kx&a7pJ7Pt+yboz&o*S``%>7E;eT_=n=wJ?O;pV1TV-`6)NPz$%<ZLE;$?%p&>
zdiX}D<yD(z9xRE4ot|CbM1GCVd|J?xtzYRRARP7kAfm@MPpr(anV8{~3s<coPJ3qQ
zFNiu#`CLX)Q%Vp&O5L-3wVbufncp=Mxa8mb9|rHQnKfhl{_H+iB;~|FsKm_BRpWq%
zm=8sV(dF-!{u3M3+~Y!8X#$Hq%{%{z5#J%f8v9v+0dHwj?+h<U!B_d~t&H6!>byBq
z-|}#!B%PXwL2MlFeh^^o*tT0}v7TdMybsbR2B;j%k0*dRLU-TbGz%^ljMc$f{mU{*
zij3HYJnm4)ekLCSj|!6E<53Q7s>t^^j|{DU@p68B0-ShKas^-WV3<OUY<d1Ej(r|!
z%^~xg4S?nHHMH0qsy6HhR(3PZ&ZZ0HMYyP~=mR$F#~6Zy<iP3Ze2zSM@x?S5Ze?Rl
zpS(ps%BLOlYN`-@JXcC0U3<O$2}DnEO;MV|Eo?UYwe~6se;|<Oy#I0fSP3db7#*&+
zvFb)z_cgZTZZ!5!-G0_LPz63U7GB)?A`uk7SIYo-$OX5FLA&RtliWKHs~bNK;sy`|
zt+7yEkjwVC%9R=t0zOj3;A487dIq)|-|DSL<N1?;u@w*r?oaUr0=*C`#$2~qd4&Vi
zNx_j3gu$L+vYcAPci{1kcc7LRreQ%RY5@~qc}9c<2?_kE)S24tP&crUP-mmv6+AB%
zi?Sz)I0!ub!Z2gWd4}CI22UzoT-$lLW7HAtI33}9V%8I!KWP4)Tas?w1QmFLYXCPm
z&T2(jbI$|^TkXBl)uD(~NVA<*XB0gb)h^c*skd>v*Ios$RL0NGsA=@qPZqc+E}!E-
z1doTW_(L8Kbmj(mJ@gziV||_Yamh&<((Lb(O-MGJFR}{fU3{6j24y|QKtR5i{uX%v
zsRw~8BpqaRN$Yx?x}YQ-M_ZXU#ryq&MzAFH`jVFg5wLc#zc7HYK{GP?UB<bp5`AX?
zQbd3<TlxD-k}OJvOJ2Zs5x{w#1*Ayp{xnkYuzwiVe}z)ZkMgg2`~g{VSI6fzJu1>Q
zovnSs0=c_u4h#!tey)ir_x5rpZ*;oFl|_EHG&wfz<M$6>&RwwH?;de%dAmbMZj>ws
z0wWc@mW$`tZ)G4#NuIC<8_%{QaXNR!QeBs7I<%H3pJd1x;`Z-ZG4|(N$9$w2K;hZ{
zR)8u5#9>@p3oa0{+*3GE3u6lD23Z$_KIPlLIt^;*{qxt?cbx<Yt6-;0NYbN*yk3Dc
z%Se+Nq4l*kxta;_nYd!4hYq_ljZ+uDa<4)<IX=AN%^rcd@fun@u6!HmXXjOEJ7L&d
z%4@x3C47F^Y_lv~zm%1bx#)ns>k}oY4J38<Gg_n`&g{)lP?&o3!l2Om{j6mB%_NxI
zkP`oGzY<UBabCOki3N(CC*VK<6s!!Td0R&$9<|c(FT#pTv~x<wt3C{opnobVR7l6X
zDr<v_aKhK=%07#Ps0zB?zNsW@X1R0A<c5W5f)H>P2g9?uLNJE^WO|vnj|)YPIEij@
z&q_ckHKZ?Qw2EhysQIJnH#%9{MT#Tb%!VOlE6XRl*|6&Fhx?l}k0|>K?&c*6|IJpw
zEDE<MG?LOw52=v)D%zdVl+>>r-*Aq$qLgrs7GK16e1w(MlkH`!%OP^?T8|a#$f&nh
zzglj!;)SgJ_4R6uKxGaIVJ;VP?;~%O7c+5hy;ib7|1y!e{wVvz=*{(Bj4K|=$$Q7u
zqr%ldElaM*yFO}D`;8qd^c97J6rwq-%`}kXN-8hh75+w);reux=HyQ39tS}rl1L(e
zf)(HMoNXa|hlY4V_4<W{79NTB1x4}o0(~INxlR73!<^dqiBg5Bt`j{t9`*i3l>1=B
zT$bjYPpTp`-QA|0Kg2ojLM2W8T`@ck4P+gD>~?$kY?AIfJ8I?$e}enlzQk4hDS;Rx
z-tXtB!}y4x?(|KAFkrUpd2L<Vm(9K3j7UPdtUl%!)a-f5KTIg#Fkf3BmlAwU38^n2
zrkC`&!&~(xj)OO$|H8@Y9jtE-8Dr4;godTPGd(9kaI4eG`;r9(uQrtU{aF5xpxZCy
zf553_pPd}@cV2HxD+ri3QSvHeg8AA&*fC`|4H+idEh!1pClolDHHNlQ<;&S-*hC78
zN)=DynI+mrWN_>otMnd24(1GOc%a7SkHuje9VrsP6=zt=3vX@^J~d#rdEp)QYPu-K
z<xF6Bz6T^`^H9kI`XiP9X#dmhg(!Lcj@1Vf{SLTc)gM2f2pS9LKL>X_>S;d(m3`#Y
zXT5cz=MwNc(X^qdw`(cYc6<-t%R@ffeIKGY>-mm#bmB7H-UAofiGH&}Ir%&MWof;S
zmxcxrn3YYWROkztd@d4robnZm?E~~R)}uwsPkmrS3EOIc*tg@5aS`&Ym=to>bO>;r
z@=%qv&84^o@^+fp(e1uzv)j|W2vths5Ej4pzAH@H&=HNTvdBA_Iy<>v#|`T-+%7SJ
z-@j!ej_klWNbI;Zf`-e_cB$t1q4vLHS9l`lz=3YUY(LE$MkNftG;HboAgXb1Y-IR~
zD(8ckR+9S7AIiS7XsS)Ch>=!HiPj{VbRC%n8(KS2OjaHg?=Q=~4$qN%bYBfYW+!RH
z&s0rJRF^8h2&tfh<X(dN+^cPBUHGt2d9bN?BQkaU8%uoA_1E8m^Je$Hk+&p(BQO6M
zNUCI&T-w4x%=fN-%KcCQf4;h7+aH+~iqKI@_HK}L(YyAVE*das2-r$<Tz_f2DuPeP
z+;)F$adM~EsWGRSEZWeh01e0!XXUvBikMw+C#7zGt2UN+FXEhA5YC)HD8DB2aLbfH
zx+iZBTEH@!@%P9+p{nbhW@sKyvM!F64aur-9e@$<d<lS3v8)Q<d_jtq>e*}8SKyo}
z2ClDpl1Edfl{c~qX|?QcD`cQItARJ;g_@$&4dAkMgc)l|MpX&7ghdjkkHIgJ2e<f?
zpi@LJqWZBEP3_o)o_;_7<lQUp{Zijmvx5+R$WMMW*~~sxCP&X;`I?cnPYg6+UqV0;
zi^;6t>D#2Is^|WMhPuG&j@Mwn{TYtVZ8RBO>_%g6Mc^1hh}-e8A92M3&?%;WMu1Tb
zd91E!%8Oq3`(a+bv~8MK2EcH6YjA}q(7q%f*(*MNk1cs~&``YT6aTcz*K2e6pHXRl
zrBecrQXu>ve#x@o@GaA`#2>V1KfzYxYPar^z#9(gAJa0Y<>BSP(#6j01cC=vevm7a
zbZO5#AtQ5DI%kHqs^Il=caioziu3*Jfbu2OuvxRgb=9xDG8%h@d8jWYelNZgP$N1`
zdG6Rt>V1(AhFvpZ5-=-=kI8FP&KLTkZbJOc@V6YbMfb#}%3_a6(Mo}Z(UI=;=Oiv>
zOGNBEo?00^EEbkE&ZZcY2ZzlK<*u7E2mP*?xtu)^Zw6Sa+p%j=rZ`wRbYU-;=pRo$
zdbZY2kh4Ku+4eRwj>hYQw@br+7itgW8>Q|1-rePB^0#)ZukVF$_$`!MSxk#={iy+M
zhjKlE>C5Rz&_n0bj_#T4b#7ih<w)z1`Q496Ze{*H0RuAkE^$q>1B~>)9mZ_m;f1Vy
ze6VuRz}?3QEg6$t31?Iht$t+vw=9)uJlW2QdFk+HGc^uSQm1|^QOlu%<t#QiL>*RN
z0_P|GFedWg;9w(IKkv0FqwI`jS+diPLZEMdk*iS!v52WJ0u6M`VzV<sEPj`ljNqTF
z@TEfufwEw{+?riq9$9p{K!Y0*4NeF2hx2lqc6lOqC2G`NoNNm-H>nd)U3Q+spFlcb
zumOvXv8n?@m|<9<@S_neYm!i0L3|Y>G&i?j@0c1!-pD-Aw7_y&*wJbzqmwH5Gt(yj
z@zaPPWiZV6IX}wmwqwU@@cqW`Xt})k#h-s935bJFU&6?kh?|}-WRm#!CFMwi2P$|&
z7LEXqtgJUXa!mX)_KVBJk5tbrKWqliWT9_s$lilC%%a-GPVVA62M*$9sC^xJ2$n)h
zWuNzAgjVdVx<B$GQqxE(`DP_moBj$f`L6$+wBbfMHbD)o8Uu(PU40ilbBgvL5YrIp
zdjzIb&(tMP(3U-azDB#)?~h6<N1b>tG<s#iFnS9jI+)WmYq`3(xMq)p^r^p#IesO+
z3+?yWL4P>+-H5qXyfAJj&Ss{JMR8teztLJ~KT>b0cKfx(kL*#O#;I=Ydw~T}|BNJv
z-SNud@+d3vx07~zv$Ge8GLvHn<Plk$-2}}=1M=MV2&a}i|9suiq^zuhvdy9Iv<tqf
zD%uN`WiVIaf0;+;okT#H@;J40Z2T#4!>kdf67+;n^zO|a#gx`&^NQQI-mn-clk=Ih
z2ZJai_;jAXChvQXi`m{C#vilM?iOhOM%pJtua?E*A_*q(3^ce#P5fTJnXwHc%atXn
zjRd(BJ~Q3j{yE~)6@628rFr|S=5M!Y?b_TY>b_??W4Kx~3*<>IJSSmg)>&qHEm*1|
ziRS%_tXF6OjDMeyMl-?a1RbE3!((HG0ZDXH15PPD1YrQqDH3}PjuEWt6fl|=TBl02
zgTH`rY%RKsBE=xbIe{{twS_7Y+ZH&7aZju4UAdM7;3ovw0dLYm)qr$Q`>Q;D9&p&E
z1lyMogd2shE5P4stN?Dpmb{(lERt&}a0bnQ#`5VM)sg%9_9Kqqf2d7S{o>a~WiMz&
zU3%a*a*~*JCN>;}pSYYv%`3Xl=MG$moFn9Nhy$<`-kg5|Yb2tPfZLNVpN4bK%+Nv^
zq3&3T3iGvQLtQsV5`tezc(|h-E5=b^KN~_Fahkd+tJ`7?8ecJZ{?3&HSco?hVmsf!
zaaIjKYQSb^p9g1N!W#f9p;EB!*jf7vB&Y~sZ8X-kijz^Wf$f`U*D6)B$gp1%)^=tr
zbQGjnben$87GY#kzeeO1c2BNxVbZVE0tj(Y**Q<7JDiL_@z@t(g1Tuy3rvP!y0Sx5
zons%*IUbGrF)`SFCB_E^f=*thnd<44Eg>aWxhH}y$s`u*TJnfN5QEr-c~U{px<tib
zRuY7-Nb})=tQ#Ev=(b#3eiPBHRB*2pwawL1Hi5$9*`^}cF|K+ue!?kjpbJUkoMo0u
zCBqK6Ow~I76`vzPlAXN1_%Y8f!|zyQ1mRJ7v6ekncNOLQwa)*!EDsd1II0Wz_n5ap
zW+{z$epfl_XUlEO4*IO4I|yg)jVKSf0}QM(Zd0n~R~F9`Zo{B})EJW6aIJrhIHw!S
z4Vtsth&I8<qR0}A(WoA?YjyE+(|5}x{<jQUJmMnk=0uRRp195!=dQ1aU*Vm}8Wr;`
zuJ{IZ@-n&YTVyLPL@ZZghG0#pY=H)Xs)>FxUGHojHrwLh6Km~BX84@%O772cinh*@
z$fS11D6J{xNO+7qva$|Xl?60}=2S9n?4^TfcgOf(O0=tT5@cUkbsA+fjSZf?H-|{M
zjghtG`oHBf^6VL*H^2$>2fIUWSKLOqk=J`sWYLgY6>WICZw2(%``!#h6G<!3l<idM
z)rFGr7%2n6Tng!H{mlLAk4T)?!A=F!s`Sy1kSI^6eIh<8_+HHFs>GQ#JJBZ4h-v!r
zeCPp#9ik<7KineJBzv%e)|TqM)eS?c-T}2)VGPmjqv<f!LuV-ZY?c0ZnfVk;?NyLC
z&`n`>A!3Y?`eQA3(ef3&@;iVN`1np;dVypth%%&%F`rSW+O%7G!>yoaZ@HOig1wix
z0<0v*&|~JRUS{%z{5e|odqPdvcgz!(?dQ!MU6K=cy@$572p~%u8X8kmg=@=5*y}d9
z)Ff7(U@|rm+GGwD@;Y1;g5nQ>$hrf*mAQXb2hmlKS`^G@EL*r??zt518fVXS^X(}C
zk)%*2f5#WT#6QBxo&}l-U7r@icWCp1zThLzcJ#-JVMd>?OcEn>1}!JnjRT9i3x4?m
zXD9~qmjI@01ct&4n%)WilokDVEt@<o9o;`8oHeO7Sf`&30vHmiAOqWaS;k(=9@4#2
z9dokyIzCc8+l5vvV>d{d+nAL)d*tK4f8eQ2lla*wm!`izMq63NMya`y)(afgf^QGe
zVa=(yJkU};_{(-4u;sn*G47|%z_3bzBDYCLaKui^%6)P+z3ZEulxg3cGxeDop&!l?
zyvS^HAZM>%&wEx+C-I~DtxgW9cX2JV5p;&exvgBm`y19lzjnp^zy0>G6T!WSN=u|`
z1Vsq5;#%W(`R-NO?MC?y=<|XX?e@H;M_`<H5!O1&)9`)4RuGT*P#jYv6HMMi8IgFt
z|8&kEvT!-IwJLrgTo;)_{v_{B-0=vfPoH4Dhx-CL4$2NqOz@tTx;566tk3H441obd
zjzmWEh4`I_oVx%0C?q^KxN9fky~_JRt+zBCSmAGFL($<a?(nRI2#`Y1DFpou3fFgQ
z(q2RHl*l`>SLi1q#O)7(t%j1x82`es9)IPXaSJ{JGr;1h2gJkuBy2b}w~?JNuFtz}
zbj*<+iAcNT#m?hVNBZMmBon4Bjp>x|c6lLoP(NrYQqiKB2r@tMPYvdTXUXAFVOu*E
zFdTz)B&8~YxUQsgft$&odnTOE%(x>vU=ftEjUHj!fc>#QiXdYfFI0<}nx-hr70N`2
zhMd;+U<zUm;hp#?Ea1XO<ik)F>KA=cav$p{4L*yF0NWd#9e=s9JN2=6uLi_#m63R_
zzm{#1!#?UA&y3Z3w6`4xC|a0sOoT(ctE~JVk>r_z9_mCE4mzFqZ20CYLPvnbqQF;0
zcD)mk@_81$eY4n3b5ARdGWFi;2xvYIh#zX|8Ba3Z**Qsuu*RnN5XqE~<M}nx7;{vP
zp|o-f9m&$F0Tk>GAS~&6GKxv(ElWlB0@~r)alKzTU3QOVGTRwhmy)IEQWT-f<oH2T
zzkbEJ;!ly~)&4Axbpu0s;C9w2;@L{f@vZn@D~}EZwqtQG#JN-A75Ab=vsMJ#5NxH%
z_;!M2$oxdaG)47sBWSO_roVefk@iF%XK<(EV{3XU&0YDeMt7I7H}f<veJ!>#)ITq%
zI`D|e^XAVFZeNUax^BW|y`L^l3+$g{ZN|m_RTo@n*B6|m)T<~rH7?!?S|wKBPrg*H
zYRAkHSK~1yQ$aqs$nPHNHBY=W^)H~yh|DGVqWT8yv$I!BokE??A^mHLC(2!)vv&_S
zum7ut331RkbQ%HPHjSu*ado=Z2Y_KmG$XJjk1iB3s3%{aulW77!29u_YW2nMPmA4X
z`T%vimCfS75?fU09`);$gA%9ZSad|f^NzO{NNOxkWgpPh(nI@k$r}VH)<))V`OyZ5
z%gV~D^7CaqIEs${9R9>9zXUUAd<wo`&!h8)d9@*VE!kg$sI^F5fB^Wy@&Qmh&{$TK
zak0%smR!P8k#Of000||5`Dg8PFdqttSqkejEyBpC@>=mJ6>eRS&L1~&tyj$wm==U4
zM%crAfDrzsXiCoLi^q3v`xhklyvMnIM+G1RPUYhgC>p_gn95WOnB)cZE?V<*D0lV1
zuBbUT04t9ey?nI@Zrrjlx&Y4Un{A*w%pe|z%^x<eI<?g5l8v!ZAmb2;f=V~-?loXZ
znrEO1wSzpl@EjgW;^2;-rRwx{+Wbm!@GNRXS+;YRe(P5Toq;br#=p>_3!X+%jdTDQ
zFjjbM?H8u7pcw#ZoReWM+64AgSOw$FSGGLmyh<RPVInP1)iq<>FduBeLcpEpj#el-
zfNCoeyTePbtebywcdjPyiTJPst%PT4?q<&w0}?NWzzy&n^`Azr=U_r2=FcH+A4eEz
zCCX3ZtVOqJ>Td|$$sL*TGUBngY9BMC!8NU~)~F%`yBn_GS3tJrU(r1Fp`y-|)46Q^
zH=2$}u$S(w@H<@V4I61KQOAf40~@o*u7`WS!G)l(cYg!F3v<0x>l?pXY}zNd5OP=q
zakh61;ur9X4HzSd_lc3m;aU6Kz-24;N7Ire&Yw3Tf((hv7LKNT46p!^kBbYbKTW!b
zm>D?eY=A}s_fzk(C;K1Z9ofzuDpB*mR;S*-k$v}sMg8t1N`*_l;-pOfyV7iB>2^xI
zWzKOT>&aM}tBy4g+a{;XNu*I-43Hd8Y$prb&DU~)d4*8S&NPNQlqnesfmfB$(T`LB
zCakZbjy@H>3J;pjCpWB%ev9?5x>7%6;}#oxS-Aq8*yd)p76nk^Wvhr@&GTuwq~GU=
z;&8<2p^g=@sEoukaR&xgR4Z<ccrxsg{jv<>sAC4|;GXfYlvn%&eInw?-gR#^Y)u2U
z3ZK!fpJd=ONt4r|l>cpFm&EN$#2b8;<wN$=dgqO~UhUp&C3z?KcG_NHh~YaOF*P}^
zOMC*FC*-f);H$y4Bm+Mt0EiPUJc<NKAjX*8X&86lMY1v6$<W#^kw#43B>`{HEX(6)
zDXYqbjdS#;jOr}rnt=yE_uWs0(M@KHB(0qaWg7l^8ikKOKDrO#&>eyO#(CCBR1wWV
z`n*N>H7GjgUvvP$HGAjzw8}IqJ+D73)NzoE4scp0gp_H4reE@sC7M|vOx#T^b=DzY
zd7(;QQ2(USewM@X`yfy?742w;ICox<l|LesQ`RNQhyQgIWirD+UV6$Qbdn`#8>u)n
zct(cE_U=$0Gb2<pNN$EBszIHK6%qQ!VB>uiSMstnz{oE=w$w^`L8<$`M`2B9&WP#a
zNpKzW3|igPW22`(t*FgGjNtR_PKKFV{?8Z9V(#07Muj)oT-8A|3$BbC1^a&9HR1Lx
zJD7M>%n6Bqhd?3kG)CRW!05Enk(K}G)1sp{*a(uWI6Eptp=YC8NiWDO@zy<={eU`0
zlr7UQRMhW}ZRMN8awdUdA_Jzytglxv>HW>TJ3@3bVIX>b@)DF#5yVn<8)dm#(7#a<
zHlHh|=8sDknd7{OX8Qa2y)oVf?8ws9Clpus;hofd6*fW<*|2=akeU~!51gpIi|P0~
zb2VOZc%=L3KdNcP_vA|tkdaUN<nI_e+sLM*Hm?o6TaI<I$Tl-$C%cteGa6%@dRg6g
z?>9)?oK`53pDsM$GaF(jM1?-apcN%F$O|=aY_i|)iuW^H5*<$<41QyeIhOm$G+0&p
z!q`ar%RcFJT=s%&S2|<DSct0}@OIN~{7ku(?)pV6fk#1TwI$Bev@K5=GV=^-go2ne
zZCBo*WxOC#IV20bNPEZAR;Jvp0fF@3^v~BVl;dGEC;n3rfM}cDWr--8$qDj4j4iv<
zU;gW~&nI5jm$jr0Z-4$-U+!w@@FPX1=Er27NAP&|P7%2YyHM9Nkm}1JStqu8F-Hq<
zUQdlwqF^|l5aQ~m-i2O}q;WNaj(L`rmpqG|5{=Fb+=ezlJQkB-Pe?5`&CroKEFzEb
ze58~z1PSd5>k-oP<6xjS($v4|eMd1&3v0wn%e)4gQ9}5LyzvZy(s3&ME1UB6^gLVN
z1AcxuMqdpjDuY(>F=ih!WE8f42LuUWWT;!TEkEF`-jVgS_65SPwuuyw{yw+SEBg(`
zFvB=`(8b_2NYe)FQPP-6kbX{B#J?hXnqBW(7uYG`i5$SRF*Jzx<z4tW1qnmI{H6jv
zALi={mS=-;7asa<d9?uNM*X4=)IXP4m(s!6?rCkQUXu5;(MZv||MZRt)Trc*9hPBL
ziGF&;9O&djUQ4b~UMXd_k?m7}F>l)9Dq-?oW@bw6STP8z<h&!`#uFt5E8LxwA3L0q
zl+xG$8Q%(xOLh&YsdTqPFCj-w;8i&`*C_E6?FpMZp68Y;&r2LOesA2Y02-eUU|&X_
z+JnI-XNBT}GpU#*cSmo(?2SLz$9E4;*b;V}mxYZ*$UCqrFr8z03whrhHWzDSnqyJ7
zF$$FpgH4qq{t4Se!G!n2GmDt$DEp;zP`6W1TgiSHz+gY;J|FDslehc6ZZ6xDg@1<Y
z7~U_6i{Q>dc27-kZRFsYVb>m+ZR^kibOZKnw)!9#UsAD<(~@2*12Ogi059T+f0hfF
zVs*Sidh7%CHFDW&cexNMF=VFT{BlOCE)b0|1Cprvu3)zFW(!5nN&5VyW^obEv(}fV
zz~FYXGqRPE?;7l&)7)M3>E1@lMbM%-EQmlPzpSyKWd{?`h~{&nAmkh6idzVHsp?-c
zUFELTjQoBK(F^V%5L$uZo|%so=XX)Z4q7JCzvvWIdj;}9KdutO+C6vkj3-5((tG{<
z{Edu4zOvP$&d4v(=e!<^K<!KCIoXzx_gfEiy$|o&8^aSj#D$@ATpBwE+uQ*2CMGLN
zNU{m>z>#M;UOW_T+o6;4L2PBvT9Cjk{%8oHOo^Rc%w3}odmdKl@5ZYgj&D^k9^1?&
zId<7r3x5!Yqk-7K$J=2n)4Se@9e19ez0(J_v!fb`X@((1{3z>fWn`n(8xqJQ%9@`>
zA3bV`$XCSz(jYE{{Ltp37`FOw-y$1A3Q3&AUa-iHmo;Gcx5`#s!rP3slsf;)=H9Pd
zkl1D(cHjP9oFxYF0942`kGphJV=oYzmTK-dhb$I&O$y$LJVbrzETPTXj6W7$ppjSV
zB`3y#fjHy(T|LYt8oNv_NB)GiRk?_lC^V$_{CABS$K#W`>Z$G^GLQX7{`g4(x>eB~
z+-;txt>^Nx4;S2>wUE{(JtGu-)HoRdY9aenv-9p}VVDHM%9dbDEcN0lBF>l4G2czh
zbwgpVj{)vb7aZPl{~3tpi&cytTIwy6nQL7t8TEaq%Z3?p&kF7%<T=P>{yA>#h|6w+
zapMN<xavcDLb{>;Rff%KEsRDEEgnA}Bd3v=rgYt&s#IuvjBJd5h1Pe+F)S%C2sS~V
zDdRd}XiAbT(oq6@)tpQ4&FhcGyG-<fR|TmD35v6zmO_Km&dL11$QbeIL*&Mr)!Th%
zJ<)9cQu+esb+cbt<KHKVsfUJHr?bUKT4c7Q8;xqgevNV;f{QBf_!^B2R?zcH9qBYR
zq#Ml5pA=K==IFNFVmu!;Oe47DzUE2gkfRw06<+e$Y;6QQ*DyYujq2E6zCZMBDVgi_
z)-mO|h2?_}m@7#jSl)E&t$w}e<TQtdugH&{pL0u+v#IAgE|pgJehf;a;FtG!-w(zE
z3`iRzTIy2Pj_Jvyp<Q^4tVauA&p|vy=UoyyT>oZIGc0&B51%6s@H?2LfSl&O)Iv!a
zkA1!niM=7T<0Iqb0g#=*s?c4DHW??cP$&C<sEK+Kt!yyVtXx1<lb;Ofv3PnUKl=>S
zI`+XkH^k~3>{sanJU{ZT8RJir3=v&L;c3tZoWDIaj(zgB%+Gf-xDFRm!hx=!z67))
zzJW^TB2&j@+Lh_`v{2p6yykU-Vd2PvAmC2?x|>Jg{Ph6LyvXQ(IZ$q@O=e;RQ0-q)
zIBd}8;=oq5^#S}o!PW4ui*5v*(ygT!@&Qn%BB=z0?T2y0hXCE1`b_OSghRW6V$(DJ
zo!8Nl-4X(Nu%REZ6V|=8y-js;ht*DTiCAceQWM`}eqes?8sEx%)D*|lrYt^(^0+DR
z?t18~9Sk9dGRdiXJr5VH<K0VDvlu<Mg2?$w(CB(t7VxM!`E@cbu;MT7NbOG&ajn09
z(l<Jm9%_E{aQc>4cs-;xvF^HQ&cwk;BL1O(HQttao9%ce>}5d3_)I2hJhh0L{!4Gq
zUI}C245Sn4>{!sMzDjmK$*qE|o9eD*K1EhxAN#mkUOPR0!hdP4qJ}2?`t+_Au^Rhc
z=qa0Ij!%j7(T`6~1@J>t`P(F83nRP+FqHn1kgcHaK|Ro|`JU=gfIym&#I3Mw48p@2
z@y<`?n^oFzDPJzyU$v%7IZ_^n`A^L@RQn$+Vjl~YDLg&gzQ1WG+WF2x7#AROUuM!s
zQe$R6#pr=0lsL;-oFQH*&3XJH<auxpx1kZ3T-Ry744f(BI+0@VxN$@=La9b!%2$;2
zZ^Qe!68Mv$>(VCXZ)~6=rRUHG=r+X6>H4T$tzYOeo%`A+s?<%_*qVIBEc1H3H|zn_
z)&Hbo3Yf^rxswvSPhB|AsrACrx;!d<?`gPR2X?)zlG%>-II-yM#Nnmi-LEp(628a!
z`|hKn+-kk~({?@g1|0QF&vjanuoZtnLw_$G;@_|Lz4*J*-W=$0DfhHKs>Y1?7;#3R
zBn$2N*aB%so}rY}gtxj4pd>b@kU|*rRCT#oeJVl}bG%jqbeER8U&N3=D?kenAM&A+
z={WRP9J(1NZjm;jLM<P5O2}A#S=b;?mWs8$KY}kgDc=eM#f1wmZW#vYYf2_E<gb%o
zJw7L3#|ohc8I)QG&G%Xc)-#*B(M^^xzP*=a3N*>O1-B$BDd^P<MnXY<7Hn5$;rAGi
z4Q*lZsO7l(V;o%hf5L;HVo?q*+V<&h9yw)|6Vfv?NK(t%pHU3cw;UCi_^FF0Ti8zK
zeu58N3^ZKxM#n)<Lahun8Fdqnljn|&u;*k)qGB1SnFtBcBwy4()5}terV#5`&`13i
zJj3izL~(_=Am)&<cs+h}?b|jbOxJUu(~J@Gr_fpxx73P{&5U`81(j2@fR>rZJD|G~
zAV_IL$nV-fC-N44_UgStKWfnyAS3*Gbr*Vf>}d0D5S1Tq)3#$3ZeKv572=o}v`)26
z>us7>akfC~R6?`rhV~BvZYfw~UyUa8dNLr~R$@B`dSIDYEa><LNrTjqIyv|b?@W!|
zj;%6;Fy>{^{5J+IQFLjL)l`1Mn<|XLJI3$SdGP~Q$mUNYSy}s+?NR?3<{8Qu&d1mD
zyI3htgekr{cAG_;chFa0GO5$eLlaJ9y6DTQz_~u8q*Z1(ihIV4c&QHZzMgE;prfPL
z?>_4%B+#=Itl2S<^3}ZU>GWLonDfCrK`_fhDMA2z_ZuR+VPg6n7qNDTNTpi9Nawci
z2qKW$I@g$~peEOF3=i8RNGw0GmDV|bY#(bl5x>mt3K@zhOZ6n6gY!!N&471xy!7T6
zd35x6wj$7SNEEktrHU)S^v|S3TH?;6*W05e7cem|5YBsl)k0hRORrn_*ll}TV`xDL
zwO@?+2qHUmvgi;Op%>P{v0LM{@^v)I?Tu(EQ8XUP+YUilsLv)|)I3GjO18a)dqkmD
z?C2JmgPYpAMm@Z0D@xi#;?gMlYU16y^FEp>xioj{e`}&(B9NnawA`Rd;_Y<+xszh6
z49XG^Uiaa92I(M;YYMhhFw$Rw^~v)7-6GF8ArL1FC06=mun>{|G!lAT(H#Xvc>G@4
z_~~3n7&2^5z*-PI?pg#1zh}{ZFWHG|8gMb2yj&I*O&Q-2ht5>?%h-3XIMN~u*Erd4
z+@RTMxe-TKrySwCk7Z4jEV356@B%f43|qoQosyu^u$c#>OSv<U;sFuvEj(P?3WbF^
zxjdf(EtE&O;XYFO5!T5!nLL*yxzK5DPv0F(^bOF#0!hU!Mjd)MGW(5z-L55!)(VFn
zx$ICaPudt)VXRd$_>>;bu&IcRmW6MH$2Dze2YtEmml|gyH0@%$LlXy?{ln9jBqO0V
zxVdO1{wIq4v`tz5#8TWYp$7Dj4jcY6Cr*ePDB}f<eEg!moL++Iiz7QSa3=kjwpiy!
zQYduiO7RNg5FZZb=x-o>&F>Hsj;S0rA5U6={_`sQQV{OT+kQ|FYruU>$A_w6xFsN6
zq$;YZN{BQ?o4qJ$pVubOKiBhtiNc3`U<vTKNrat3haU@aZs*kdTT=c2>vl^cgW?yY
zT7#jgNqDk#$nyW61@IbbQ4)(sEB>TNI^vH-KSxKwz{uJm%8SptH#!u;oWWIiv+HJ&
zKF(#?S#qrSN-Gh4;*WgRr_j#);u&u7Onl(k1b3UKt{V)sJ`I5dAZxD2q1s-N4=95`
zftn-}!XYWlvHrEar_*Io;@9Hu$|5Jucj|w=?sYx%Q>1T$EY_zz!$0=K;U2Z~YDs1n
z>r;m8enfQIss%jD@pJnCq9p=u3Rd1#Y%kR&mOXJ14A5D=wi5h{$E3|4=hc4whr579
z<rjkgz{mHn$#HL<W#wY~lrrRvJQ9b?+j$jUu>n@qzjw6H13$0|L(IO<wzw8Y`XPQ-
zb7IAH7}n&zbex_fT_Kmx3mPcVHp2c-o*6G&!RT@^yDfqPWN&{@u$4QY<KWG;Cz;^_
z%U%`**!eGG^;T2=^};W7-*GCvH%?8JWFvzwI`wp3#73GUJA<V9nHp0WtvD5ED3Dyv
zLR)=jCS*gy+I@p=fDAi9;I6-4EYc6wl9NgaAG509t8ZBkH$TA}QG<<s(ycKOeU7>j
zY6c<IZ9z|J*2IH+gto0wDcM6cdzXbEG++fP{vI#dKp^gmg)gu&;!W#i30PZcV?)A-
z_4vX#1XJ@1>^TgFano0BB%?85Z0Y4uK3_IhebLrzt};vmW6ESU7x>M3kT%5!48IZ_
z_Sitm@&T~c64r8idp4O6^oj|b^v-CtSeO9<nxWPSR6_AL=U_pZ=%r<$BRymXl^qct
z3fIEws60LjwLpAKlBQCdE*prrxdh<wOeb=VC|j0C2jJ&;snd2c<W3TxV+{~v@EmLj
z$+yQyu9EP+*X)+8V`4x6W9|J)^zQt7|Im;GvT#dq{%2s}L$PuE8iE8VqxkvH#9uNr
z4sDd5rQ9!`mxT0hYfx$sQx|%Ii4+ZzoxMIFP4$0QDOJm1T#N>*E<wwQ&`VnmN@2$}
zYP8|61$yk06|hq*A$&=Dv%oo^5QV){-t;(_pYnG8h={WAcD_6ACx}dG%Nj2MZGsTo
zYiB(BSxb=l)ej~n%m?tRGHCK;;(y4fg*Yx2>0I_mSCXhf0+6LLQGn=Rn_P&PY?uN*
z=5a+xs)~JHrnUV;0*m<<T!?+ALbZ6TGS;-UKF6s;LtU0ustwL*l0PD$-52)wzDOWs
zuFf~&T)^f>^U!GC6L<W0ije2;Xe`w7SwCDN@*a0doJ^Y!Ai2vcK_?(prU){Jx_v*%
ziP_DKPVjNR{^^_}kDH&VCh-a8EyynP=(eK}?QWqkA_w4WbfcNez68h#u0#H-g8sE=
znILXCO|Dn?Bj%{U5`wtm_Vm<IKA7oQblN}=zMw_;)A)XQrrV0e{`I{$*;_xC6w4$<
z7xwl$(Dz3A{e)$QMHB!bJS9gFG&GoS<E6!bGwIZ?9Iq9uCJJlGwBPD2yG;hI)9u7i
ze9IF`W;hoaY7`6)=aox2)&+gbr^pmF#>{78K7HX>=H<iYQ~$OwXqv)Cb2=69uccxf
zD2Sz2_q~6T-5Bb)Flwikkbm1y<91KHiVtzZ5f*x|JmJ^x3y?76L)D7a)9GuYB+Qcy
zYFj33=#Z0rYWdBN*LojeKZoomL2cK?{s0^jElXHtO-w5d%xmc#^kWD#>4uvo%z!Ml
ze`=?I@aqtSnHXihu@>}1p>aEaogWStavY!lEu);gJ&1W$w$pq}yeR4S$MJ+8jO(&6
zUtAVYr8b}759kqCu{BV?mFE^o^@d42Yd@wNxDhiV^(gu$#!RTAe01<h=UkB*o=bZ?
z@ulVa3zbKbvMmv9hX4|+(sKFcv*)LQ@hWCyNIHc`HPG4(lPfqNTEm!Q)Cr?4y)hgY
z8)N7Kuhz?q9P&{9_PC*O=vF(jpI97Fb5Ihc^<YTGTYqC)XDI-X7HN<EGZ1^qdydr~
z@_n!9pQ0ZsNr_t9iff`;Yfd~_1QYg_%b~i&9sWRvVk7X`Sl4z;g#YLZ>omv#-7CZ<
zH!qN?Y|IMxIR!D7`H1cZlNPybyYrv0ZdzFn`ta9|D<M&DUAmpujIRbr)W3lyFMfO+
zOlmcFj#$btnGS)yluR~Jcl_kj)es452hB1g+8gW+?TYZ`U^>Ss%S|OA{rSS}qBp`3
z@7PjqdIIx=7uBV5PM!7-Nz&~x#nDYfC#QX|-<h9$dwTtHnp3AGV&BJ?I5LsOH4c%z
ze*?#THh|GKV;uX~xIy!WHG3RZho;d6Ws6&}3S>H!y%JWQl7N9Xz507(z&{ks4Y?Ha
zXTpPj?oO%F)SX-CWoCWfBU`2V@aA%v#bUVvHs%#K_`X%kyBn)J`KQxpzL>jP-=Zi*
z;NuVf{_%fhOfu8AvbTZnY9wdtU>UyH0bH8V-EQu$H~bXv0kX$`XSwe=<HfE9qbLWN
zFpmPRsKF_d843NqUSDtD?-rjk8%tN53LQPOFRRP#UL<hm^$N1MC@58E$8Qa(`r50a
z(S2a#87ib9D-8J#8dcop8ogITg^%w;;@AID<5|R{<@a{7L&9<$l71%z>7DmqVEUjO
zB8y4Dr_<6d8aQg6Zg9%XxPv{qwNW$8x-Jfy)nvEw&1aR4;<tzoV(zbg|M;Y2Dw`Kf
zZB|C%a~)nPGYqfzWY4s7#_Nj8@fynWk4$tG)rITEb-%8ULF~~IUgvF&)5kRPD}M#=
za(dU$#V;=K_j2>G#3S<d6MM{zQOFe&QVL|4+k5o@c!GC-Qh4Rl=YfuRrp?0v<#Em-
z1qyytKlD#9Mfgkp+!IWTwSY}ZKkpThbtgVC9Zkm|T511z?<DywFA&>%4u+23d&4lX
z5U-QZk5>lt9%WPI6H?;{2wZP1*yhn94S-thDRAjo$hM>*$Ap5S2RdP~6#hHp<*QtN
zl4?xljkgpl)~o(!2?Yc3((yqp+q*UQ2%+;m*jMMbGc0M}gI%Id2oK&nR1-<G%BVzd
zhr;%xc6t!N0#0qEI41FJGT&seP!Xq3@837>ePsELjPHCc`UVxDdfRV=*Anq!NiyC2
z^uABh6S!X^_XW1Mym@?bwXUcqhB96i1Uq9=jD}_Y&Wbwv**rO)jFiGty$r-&j^{Vn
z;zI_RqzXa37}|&%JoEXg5I?*JS>Yi+Qgg?>Fj=_}L;pFgn)+r3TXM_urCKDhtyGkg
z1jXDUCA@5C)XrCRw=y#1xvtJe#NrPH#jd%^rD*<Czv3q9kw(&({;y!m2_%EF;ftu4
zrvrl0lJ$SJbxA3Cn6<~n9{k(qp~NPux?5?g>!v}6MV!^_Iy$X}&A4(C&zJcZWSW<i
z6N|gi8y~_KWJ()HJY>U2B=d^~UM5hJ$KZ|Zv-Fm0HAKv-+JrWTSAKyMVWLQQwhW5L
zBd%CJq)%pNFl~R{juu)^{o*<cp^@ArTTTD4k)5Ii`64=aoht6cZo8%mI5`aw`KU5A
zPkZAw!9DXoO?SvMy2nLqdGN`wur%M>6XC_?8u{8+<~h7Po;VDp7KdJOwF2>cZsYHT
z*Z+!(960k3k$w59iH6L2TrwQ-%jSPy=l}kQiR=Nfny+@h6(G;{Mbc(*o&-nJ^gsqE
zk;~XE!{~4;?vFq84*i=%{`>D`iJA?X7+DB`c>Ve2m#8BTY)T;q9cJZyw+Q@$L3XY&
zmCAAS7cTvhuCHcb9Pj^Bt5=L+B<!|`lw5}wZ}b{_r9lu1_UK)B)NX?qL&w7J#&do@
z%2v3*c_V1M4TiRRiWF3mE#Ye%K+ni(rI9CHUT(in7K-b~YsXBdi7q|foaaiGbG%2B
z$0r@li-L-o{%!c}exUDqgy;Gp&mH9Iq?5x)oGD+)?93+P81j+MxkBsSZ+p=}FdS!R
zW?7QB>h466;IOe+W%+z7Lro6qHmxqxs=WJL&%(If2EyC4AYaDN@lr4S(L~n&y+tMv
z)em~AUQ0xOZ9cvt^w#L?lW)Fa)lhZc8R@R`5n`y2k$BUi$Ccf^rAWFpGa}i+`=)%>
z(+Tf?O5J~N+yB0l0v57=RUpXGMsWi<D-7?vP+{xA3JORo3~Qu)maryk%l|no&Cr5-
zj+X5&<Uf#ajF}tGR_rTYS@dQsbs3x5%%hjcq3nQwttSC5WK~B0&=V7;$0GM$uU8wd
zUzV%zj_0$!JXgW)kIFv&n^Q!-5k-HF6UJ>=!c(x&YOMt@0~!34UC@hq7F-mWhl8T2
zj~54Z8Xq;^S>^JPhj8{m!ee?uP_NNE_RGE`Dq}Ux$vn(-=bf`0FSroScdxz#e2OyN
z8jGi5tr{O0sh*U<y;~Y1JN%{;J?j>DpLy#Jo1Z#~l1Fa-{}~Zp2#MOaB*j7<m*n`f
zQ^zNkO87#~)nI7-6=dv6^g2u|*c$$%iyx#5GLpeOn+XBKC0S@A-KgIGs(SzLmjE6=
zLJX9D)X~=f<pv%6cSRBoA;-%yGV<9@(1^SJ8lI|{Ez{WCb-w>^a1Km1D-blMW2M6A
zVy0u1!vyMnHp7xjZANi{b)VA=iP2UtAE?lh;m#MkfZuJ97`Z>QcpAC;JP?Px4Oe99
z?XR48yS2iz4AU93@^@;C6~+=C8Rj&Vy2`wUI(Lyka1bwrqNwP!zt9l=LCs^~aB<;r
zTFmx-`WdAX)RhQ{0dCv8l=582zHIfv!c<D?hU=Zu8;bW*&Hk-%xq)q|Jgr61YU1U|
zP91d$(UhWgJ8BJL^eU^M#+1H$k{x)<pYdM!<uab!7lX0x|E^Fov=sD^%|xxFr}B1B
zO-$J982=w(Zy6L<x2<g>jR&_RXmEFTCpZLmf(CbY4-h25-QC?Cg1fuBGz5nL-^#P)
z-Ez)%{y;TVbal_U=9ojTJ0dpImedUn_+)MbTgWF<lC@vXmid<Jr~&FUbwowSvV1cR
z7v(#SGL@JxcRrjYRoPCJ;-@{T=L+)1KdnuFf7`!)eS8a|kWYK7n8;!>JPB~M@X^Nq
zTHA}Cc((ZEYCb*!&WDp*hs>{k|F@5}AQ$j_<Lhp0AS2rSfnt5h3+q{nvj;|~ug{Us
z>AfR1_C(4SbYQuxK>XRApMnMr5wI8MtM4>3Ej9#1-DW560Pk0U2zy;TLI)IY6`#*T
zJzZ^@Uw4#@tl_VYD2opz8AXRpllJ%oxXHCUimVJjfRF41|NAQXr6c%w?fK5^xF)M{
z!v%DMpFQTM@JPIA*8~t>(t)@TLA65X<IM@~R4=M_Va*9Ooi^W(R+lj}ydZ}!>uoxh
z{L;SZ{73hU|2UawGN2%6QDv83LyusC0v6mbVWt;gzy@(&<MQHj4xSAag_T#C&XT<M
zlj+n!=D~D_w>z7=1Bb3hak|x`0`YK(q39faV_cOwh+M?02haa-9ethr()KS_%fBr$
z5|9#J2HJZ-Ewty$^|Y#G8qD-en-Ub9P(T4b!A|D>WlF*ch`fdY$;azI5Q@KyZXPGp
zl;S4xeo3@0ap=MUN;CmMOx9?*>-T4|51}a(IP_4An@9ycytWw;05Xg8E?H#3pd0GB
z1gLlyw}VJkH^+vuM;$HlCspOw3#`_TbChzg!xa=0dBAP6=Y>O_FwOf%--N1=jRpuL
zMRa*%<pne&KPjP4HQ6847E45m?V9N@G%^=7pEVfr+{}sq(2OKKpCw`KhC_t;?{ONP
z9g#o34dtAzT=j7l<F{-=sjWtWB;|YmX7$kUK+@8q@s86{=rxADR5(Ga$*hi2Uy#hL
zmsmL0Tr<H*#j@{G(a47L0JqV6twqOU_q!}G-(t-%9Seto=6tFl8;YVS=n{i7=r0uk
zVY?)~EHi*@_HPfb0SE)tzuoih6A-2bHEwr__q_f-5DZiQT!Buy87%bq^9N)ifwT<3
zy~@p|%y1*&zg;wvjDCRo$NpEDMZe^Fk|DP7;0_9TjlK6y_EL@dpjhMfA)ev>Qo|1*
z*X_^^pUjU?(#@d;V7ie?W*H5<$>uT36?{>=8|e;|ff*vd38S=YEUw)jQ%-$Lt-Lu_
zqMK=z+;*LJf?LhQ>5wN%C3Pf&Br&*9xL^CnmeGTRx)zqq3zvc{&G6Q|T@&n{PGz_*
z<>6|M{2(<0nlQY7d~<Ccdhwj2!IeupL=)43J6l9=pI0e0#VqlC=d=@<JX`hIBKloU
zD2b9(SgPH`>B@n?BY5z;??ADprr6%_66k1H0wALROvqEv!tvkcyP2T&(U@dDM((a?
zlCl3Z?O)KJC`h%LW{skhwVT9NUp#6#fjX#}D_J@MDv^=QbN81h9_P)UmYZ3xmw`Nf
z0U^Ki;ZAgIa0+8*pok<xyCLjt+qH#YW5R#f6Ch|1F&Il>H+1kM0D1~(hzDpZngaf4
zPW{cWeDD+C0|0-yplhqGMms=;Oa*Aq6)T}w2>#m}0bw~%`=y}IvCzOrS*S|VslZ1N
zM{$={+l=2KLE>k*U6jC|c&NEUCewC_I$)xIfu+1SeyT<xqvJv*Od}tpIEJYLAWv*Y
zA%PXm6!nS3K>XzHE_RTS+G31<5OuaIK(!n6o37SJv=dF>8{nL|w)BfQZ~oK*x73*%
z6(TXpdgrverao|z#?Zlol-<fzjy5h$H#(Bp9E_>4u?DppY>B2)ZF!3H1-lK6XGj#a
zj&b_lQ~D<X+&>SfdX8LTN<5A_rBIZ2PG|MWWqyPpT7c$Z$O$x6WIbG^15#a`uk#ho
zVYcoyi~gTC<$&stoLAvEaCx%;&=1xk@d(-BaTH$^+$7%6zsx|406ZYbaIWiqTwI+>
z>w1<V1xU!<$xjz5Am}<N!4}W@k0jJj1)`cx@!uK0-wFQQD~*8In!QMw$%SwUu;x+7
zc?P6JL7$#ZG~eqf+OYiN+5fw9f@TeR&uL?YhHp7nB5s!MoFKdi5M=bt8+$s5aPEQO
z`1fZP_AlT3%t7xttUdycKW>CRX>f^-cM8*ID0bWRRi}<L$iKa(ki`eP`>GE9y^)iN
z(((DC&xQRFdq~75^UVkhUS}?P8fsgi7u(<tih*C`0!lUS^*;i5jXMOH(7-l6ylh57
zJ}O)KL2uJC$AdQ|$^#O#>Lpk2Sn}eQ8z(+$2EU_ff4)g<c|ORrEi5aWZ8swTZsb-)
zwLpnKH!@%eJ2a&G`yW||Jij4$dj#Fnk)jELoIvYux?2UGuYETdDboycwQa}p{>&Nm
zA7nUArn{Pcl8`v=h#l3TMibd^>lMzG;Qn%DdgIAI&B8<&s2<Q@eL2C<k^eKQ3d=kl
zlYVV~j3I@iJgDPkSs1{>5pN068ahVEv^mpac@kVd9K5b`+LinCV{3{abIYKWYF6_Y
z+HrpX)+P@d%8etnTC3jhNx<vZ5M7VN4!A>k!E=HT!>9n`@sRWx*3M113EV-~@~|)a
zo~ugCD#0P#)}?#O?YDhscGE^RsWN+DZoCh(p{HaXQOBy`f7~#{duUvjeVFD))A?)w
zne$KccHhAGMEtHUTL+L7ngs|~alQU{aVhY8fCYrLzXLEAM!#$<^bBpV|Ju6(^vPy~
zposajw824iz;&7N530A%y$OPvF(6ZfwJD5^`sO*`DB;2ze<rxdNf#OAMAK%9&_*1K
zidR7SO0RodfGX%zK)20vqs#W#gQV>Fknhyba#U{uqeo50j7&#m7r`BwG8#Ip<C46l
zm>6!}-^2ub2l#jXbER{1F|6ZN4>9)OR@7Y`=XpGi-=%b=QQtGn@KHG1uRj%Ew5U*Z
z=q70W$&a77A1nbP^s<jzq4>*<j1$CAYxa^9v97&`7NMxE=F@qG7Hf}HO%iay+n>?L
zGOVN8N@)lY=Ydfi>9rPYc!%GREl=lATCE(;164HmTxoxdj=vq6M#Jc;V~vgO*d}>e
zc+gwi4JLqXp<Q%xj42f9JB;<LGaL&Y=Rqkynw{x8p55X+w{gvPSsj)q8I1u9K=6-W
z_3d>Yg$h~t3QKpM4AGnZ$1Rl;B?Ktjh%)2?0v00OhAM-X+$4YUzz&X%#ehOZT=GzB
z){ZxQlh<GSih&IopF@4DU7@pEU!ke|TI}-R@X}Q&+W2LEV0X1{ol@GSup-XgMJ}_-
zi3eU{)7j`|Fi6+e8<yQ7Pe0G&wEB|+&se-sPPmd>%a`vj8CLVMz6z4{4$OnA`e*OO
zmnXO)SmgP*Ve=i)c|hy#15&TQb0dU>5$+#nz5bZ<oUpD9-&XSh6riAN`PkuEs#=S=
zKE#8^oD(<Nap(L%xzY@<zJL)M@qi-Dl76097hx0Jh0RFLnqVH^jEj=*lB5#GGyS+`
zU%@gR3$(L*GwkaXx(puoTOxrUS6W>^0I?sMH>b;&F6>7c#(2eYt<WG6FuGRlPkdo|
z7xsf6f4NRy380d881<~qA2bT1d`>YU=hJ*z(${@>CEXCF4_-E26DIwkvpc;Qnur5&
z=!da)!Z$6YLs8HH$ANC&GW(#il0H)o8?bvwpGtJdZT@-Z|K0Az2*IlWw1d3mG>2j8
zYrrK{_VMxgc*%|Nul8}s&rcp;1w0U!LZraLDxYX{qzRsvZaN}`Z3n39Uvpd2s%QKH
z|E^TfXdw{FRxlwaCxfThe~gnW^FSubD5vRx*RMxBHs8m&t%K>i+}-Z?fU=z_C~p0|
zjq@T*K${4cl{r^rFZ33!j}Ved`uMb@^c29&o#Ku<J8wIN1wyDuEp0~#F{w7RI74qR
zK0loxkSg}Sp4QHK5rsO|3q_r;HYZiI)g-A^uq`ZX@k2LFB*%(+-t8k)LUYfyx>6i*
zC<ZGpI7PHtr;aFglz-YI=6`l#_TR)(2&BC>pOG%sYZr9_x5KDte5vT{tEJcB`0LxN
zkybes$D$BL@hz}U)O}9|n|)7|oo%p&U2{(KS*>%VL?HKmP8;q-X>R6t&&!e8aZeH|
zk5W_UyvD=1=4NC+Y~yk}5E3maN^BQIq{S1}|BpEMpFy2@xDXX+9cXZy0Z;|ODgDjn
z;TW8$|Id~9gblf7So02M{qg1|(?Zi<eJvNUuwAr5*D$z$Wcc?6zj38on)=7b=Q1r-
zxlNe&c$_JeBz2aF!nk@x;8w*gIreF2l{d~Hz3Rm8<6NTAGAwr=4?>)G8MrVegUUi=
zH<2O8e?ndz@R~2NJhu~kb0gZJakNJW=}KXLe2F!flN&mw)6Cf*{#2>A`2;g;TdoBY
z;t9>><@V(~t9>SK8qDpOHk&1k?tQJaE4J@(V+Dx*0y3QL6^f=~xl_5dN7FK_RJ1aJ
z`_kn2<u)>ao0j2K^VMRW?+eui4z{LlK=W_>?S^-}4M8bv++u)34VH5xx3~D~`2T?O
zPi7s9q)uU*!u*mO**BwLrgU@s=z14~fRbH;_mZ?#Qp$4JxZzOj?)cmUblzlh5fMsp
zEjQV{T#5^A7;S5FGNk{%rTlvwf`R7$GZGIeKb#iu4qnu9C6Iv&Jm5a2T}Rfh^eJ=S
z=wGjtAHd_7SVBiB*$I&)9ePna(V%2g!2^rwwzt@XLX;E0NqL_(TJ$OXSIg=tW<nI4
zYU#1ytG4haOoam9_xxbjCl}84i-2Yr0c%Kmu4bQh?WS23+|{OPLQm)6T`uJouduot
z%}9or5J!Vkw6}tKzLmQ6!Rf%{as89o>7@MFkAUxz%C&Q0b0wKR)3G37lK7=8;m12>
zjts9C_j@rcP0b-34yP{&_2$a5Z)1Zj0jbE4>GL1t&DR4&?rg>%afVc4ddCuXFo8;U
zCt_nX-zPB`n6OCreM5RJ{G74Wn>#bZ#sJHKKhLEfJG|FzGWWK5soEv^#s0y{;TUJF
z6c_fV_i>A_PaA{FVPkC@-if4fdpBoZx0-P6PZd5d1!csUAGtwp2^kKFad>_I@qCNn
z`U4HAGr+C;rSO9E3V^{(0T@Bj*+QI|KV#~%l=F%J8w>Ox)+-QZbHnBOuS=g3?l)fV
zl#s!hU9K(>5@|8_sgh6#y|RwYYudkI5{pao{?>9=plYL#+yqP{)*$;V5~dJ<#s4)J
ziZKf)V!yFi)9RdM0({kOYv+>19C43d5+#^q1S5&CI?^ll1Ing*C6HVAXuP1?b7v&i
z4G)L)=etxlrTz+Polp4a8Z9cXUJ*uf(*<7${5g43gBoccc<+9V1eR{25)r;bC)ezD
z<ymujI_qb%N(u~e^U;76-=S8n2|hmVly0S2?H!II#e)a*uZs0Mp|*bhhA-Dm<(o*e
zURe!*_tEgeRLJ0QO#Ca^)n)Ogfj|64=sAs4Bl(izBF6_;KTWuCEg`%7`1b<?fZ`d1
zZ%{lS;^ojrbE*YmMQ3IJqMCDao-03O_6ZsC0bX!UYv}Cre_XY1{fmh0mdy*pV4g+}
z+pflQ0c@E>{+sc*e>oBUVj>VWDypUltXiENkB-*Go7!a>=oD8%s#919z_^h(47mMh
zCgn6jOUdObLE!U|hhlk33L1(mhgDUS7n6lbf?7MQMLnrR9*Q||5U&^WdHfrl`fL1F
z!slF}iNoBxp1`~x3o%ghU8yc=n4x$;RiEK^c^dACtGruf&?M&WIW3HFQ}^e<u0_4a
zR$C&t-j~7L(~%6XVm#^^hXp_hJ$pePwY^<&yY;-~%@zx9;{NzpKAz1W8JgW<nJ7eG
zWq};()Ci;HYKn$>+8%eg_Cbz(dQ{9veun}*V;rmTpG6_|M{~PTb({cWD%2^pdxSAl
zzGJ6JmFJN{J?}+L4Chwco%n6D#$vwJqr-E6{{YB=p&3OR0=?Y7Sgz?eYli}a3|Fy-
z0&o^$|7Gd^w5I%f0?7b0{oelzh8rH+q;Ey=iBW?QJeI7RKWOAo!_PqRV1x5;(LUvs
z=Y$T=<O|4z!=}#0zsP>)&GEFja#9lxWBYEFT<WJ>R5;}#t5pqOMoPy8CtREu$LCB)
zxj~CrF5+*{h7n!6LmoetbzMymbK1SOxEk2^rZL_@#!M&B%S4(vlkAk0tf=cBRokxX
z0Hbq{$8QiD!gjMmrQwI`imfh^rt)Q&$C9#r89a>qkWoew_NE7j4nX~ryacZ(1pY@e
zGQj5%m3sZAS=A+F^b;#|-7aeA&xsZaC$>9`NnQ6C!S}zb8No1<cL37v9*^c*>q15V
z-pYJiY)sxqPHTc1Ey4GhJ(0&H&X;wmWlnBEUw>4uk<TyxNdK8iP)>Nul=W{3oqyhc
zf2DW+^%b!waSCYNASlA1B(vtRz#_?LEmP%sDO6x=<PT_pR{aJ+<{)rX7SF%0l!X?d
zjcR9zvNlLMy@V~tu-ox16EVAr8*trG-KY6@95b}XRcOhJxG-#r)I!-pkTD}`v@et6
z1VdLw6k;?ti?w^2d7wNnl4glPfKk?&@`5OjDFt426u7LXNba9ZE)1UAQKC_{il!yn
zr<ABR$9ZKAuO{Xu#K0t_YV}iQnI<#p`g;`LNU~g?KkeO(Fybm7jKioX#=NHoa`<28
zD;G8@i>E%>!2IP*44}Q`ZAf}~7+9sq8c$~@LXN`5uGQ!FG=)daJqA-04aO?)$a)qP
zDRxtEAehVmvy@VtS`(=UIiBZIkzjJ>RyS%j#S0~$aQ39szo(Y|nOFbU75wXuLV56F
zv0X@wgFHqeW6VxbZCv4l-d0<KKHE|%`%$gY;+y)gP;({4xH-TKU8)1j5vu*ovCHv{
zq{90YmVPNkdKuG|`VqNSmtt)+St;#w7VdGb1V-edA|Rb<lA%mXoCa+e+SLj^R?P;o
z3xQF*^3Qx<b>*7##3Uhx&-m8APN>q*JI^fWmu`ITnS#)DN!Do|YyG*PmkNwu9)j;A
ziRiNdwn1cW+YIFs_;N=^l!EH+@Vl@~;ywVwJse;hijrzOc^xDT2C|viR<8<er8xEv
zc2F#KS&5V}UmdrD{(5rA;Gp18TaKC!MvJ2l4ls4Soek05wz7Ad53Xt2?p=Nzmp(qI
zQK_SwAis?v=Rn>=<9K;op{?XPj8{+UJBgNQ!9VGQFGNCImH!`(^S>r84L~A75PdlD
z+;eG2su-CU;n>zH?JX&Gkcsnl2x;ePzfPcN9**FM-u`;MVp_Tyc?`OQL|55UfkmGi
zOD#WB5DhU_S|U!|>Uffdxm#X#P__Oh(IoGUyO-{1W5L)OCk@}Gi9A7>dcO5G5~<t$
zk(kN5<psuRggzxPQ9<4-<{W-g?dADyGEQcOfVOnyb&QjN$LV7_?W`-a+8a~i&^JUC
zlmiB&!$nHPl4zRMJ>meivQ3wpTMXP5OTX%O0B}UQnUO=dGu26%41}^38Q3K6KOOj-
zzG)LW<P*k^Fkh<#+~A{w%rVTCvQl&a<n`y=U!%EQPl*6|hgH4~-BvmpWs1x7)<m5j
z!UQ~r2Qs9R30e)1@9}DUp6G#GJ@RiC2<&zDt82&BLtT?`K$#kJ8{#$}GV-56;J>fy
zUj`c8?=g8;I`9fcMJhTZ(>3EU*81}Y^i{&bTK{|Z-Qe#6XvT=Y49-;-fOYNegjXpq
zIal~v?NMJ(RvxR+0A#sOq(p(L#bF@o4_9HDHH0=6nDw%sxWDLeB|-C*uI3xalo=mq
z8_Iozd7|bHaxmXOy1mz^{h2<CN00GlI;Qk$oY#ZIpw%UNV}O9i_|w(?BXgJI1oxLG
z7s~<251br`=;`vDH>a8;Iyxt;m5Tmf=F9Mp*L7Ko7R<WMM8!6n55hr8t6INol3B#g
zR?MbZCw8Y-n;+^-WZbK}s{@EtRoPy!an!N|U+E-SO?Q!kSAjHf0MM_Amb!;&y;wy|
zFiy!q^w%klNrR`;-UqTX(z5C=Kk;rh5M^-<NCF3lCP=E`q|E`o<W;ou2J<-IjJ)9Q
z7*W@?Jzm0Y*Z%SJ2=kSO*>}0r!h4tk@Q0G;0!k16eMSAp+WTKi2o;`!rXw_YU%j3w
zS2BHSY`VLwba%dY+>a;>ReS7J=bX9F+)V<5E)R*QQ0$f#YFT2V@wDZXL4p@T6hNA+
zV-*rw8m9Z_V~zKFqWr+UksBoQ;<B!j)!GQ1R##Z!RA-+VKPD6)ip}&X@nv*6j-ysV
zxVbYvwHP5M=Yb+8H_RPH$H?>F^^zDs+~BYd2XR*gnT|V6Z}_GT9AA9q^!^q0qGz)q
zz(=xOq~!5KRv2=s{Z6auiB2sGj@s;01CCy=%vZ8krYnuR85p&3(}1N;QIG*x$jw2I
zca2f2%Xv~Ein0u%KzkV4?rP0-7j#pUrs1f0;kR!_;HFg_ymUwT5aI_@guM+ufCl7H
z=i|+-l!#BGulGqQR=Cy_=<a*}+-_L|FF*d*q0m9^TC6ggfMPkZKdzgxavEcQtd`=#
z3wQS_lz+VJW}VPDk)z^Wkznjhle6timveU?&_Y!}AmBAr{_()PU#;_?o<o}T`abQj
z{@`!J%Ky6=2h!xCK=MwH7RwEor+poA>4xHG{t%+r*GA=P&rIO{z@6#fSlJoIVmW(@
zuoGwmKU-c3R<L8YuHV~R0(}<gWNI{tNR^0g@Vm5e-V+_Pouj!Sd5_g%xga1xg7Bd=
zFPubqqdtD9a8%1VwuGQqZRs8Bbzio@xq&h=s!cRENcx4c0v213gZoc@7$`33vH^*F
zG^OyPK1`%y;Xn}p;Eb__!d-KBTxTv1m8}MJ@-6i}iNDI<DF7wWcTUeY9wfs9M7&2c
z3b9w-r)$9M`x={5QO38l4_Wt{f%wzztrz35zLHU0+x;jl&$m$7y9Dlp8F9|Xi@}Bt
z0)Wa_l*jd8)ZmRry~R7j)nrLLj<`Yq!!}Pt=P`_Y0KS^gE@i=&U&2rP<Bj=$`D5MC
z={0jQZG%3A3V1tU_$Y5pW_w`kc_H2Zz8d`L!Sl_gnLHMUZZQ9NF4|kBs~~U1mb22M
zj5cwC90zJE@*1=BTnF#}ybTP){p9;65uv~^xQhs6Lh<>m{*4#UoI~}>_w>IrmuLtj
zA6(ZCxOY1-K;r@{6gWd<lrOkg$S``wYweyPcA91SK5w~e;u>t~eJ*9zv@FNV^vWw(
zaaOZ(VoTIoyv&7WYTmO`;AtpX*$Zb?P156d&Uv#4W**jzT;Cb$(2NID&_QKXj{?+f
zpZKcyXr+D*o9lj9vVMST<!UtiaDW?jr;iSRdI+gqR=M~euJ=>f14ZRpGu<g7930uv
z`-**@rOjV`kdhJ}4RbtP7OOt-*a_(az!1&Z3WCncYpP^H&x@>7pf>BrmkLQTro^?B
z^+D7DWL}1^>z;4QgeL(B<dF+6cSa^pmmZvksplgX3gqcOv_&udWnw<Eg9_wmKhSbe
zZ>ILWKI^;m^eR_o=zhb#b}Xzmx-n|_h=%~PDIVd_zh7l!6c0#WO#?KB!nThmLq|EG
z04hSfQfEB*zdY>!G+%k5LXsgtmcq4a1>fX~nu>Ab7Rnuw7OMA0n6)P~3SH0EL}6@Y
zFQ_>>H=vCf^}8v*p?aZ1Z2eAioP9Q*qe)bhtnCn7vf_BpDxuKeH*N6xGF1Pz-a0h`
zh2U5J;uDpQvlzH;=wjs$ScbiJ-1a?CaIY5KEX>p)>l=!9>BoK-&{pUJONIF>tWLa3
zw8aQU&lz{IO89kMr*4p3hbd8UBtblH<$8;HYk(+Sq**<k^7f<C3C=tJMrrI;w}s;w
zn~n2=)^lmS;6}k$r>P(Wj~=I)f^Pc@9n=>~6n_D4#4pbt`ae9IxD#k4frgC-Aiv_S
zq-1F|LlLn*!6*|?qj}}S=N1D9aO}RH3GtNciHv5x8cog*a8o18V1x-kb4?l-e2wY)
z4Uulw1Gz=xBe-_}E=0t>ok=z$%nCt9omN#RDDVW+XC>;<6zKcd!(%i(Hovv;<`MwB
z)}UyJ^WF^pawX-?mq$9Dv%1|ZN`d>V7v7)1ysd2Kfr@&rp?K=Ww-`Af-yyy@2<3Ud
z_381^@(OYAb^qrpB@5J$eJB}AfDY&XaiTurfwD}CYI;Le&3ce*=TDQiL+JJZBHhNo
zQ}0g32iNZJxpv0$yonB#XAVGJxabcty!{OX##w>flqSCI6iHx<)70Hwv2yzL^WD%)
z<`#>xS-qvUC}=7kz0%PpF+y6M0hKBPp$#&ok8QIRe0_&n3jTTRkSF%JtPAfqOND-=
zFRq)z{iP&sx~m?#Tv`!jqQGn0i`Pnn3sCYM++YlZHF;<<zMY+T0>cq;ng25O=5O1f
zoCbdol^)72_Gm!+?YR?aH9YtgzgNt{mSn+XI1b}+6WUQsZM1%Wlv)N**P&EQ?#~9`
zQb5k$Awbbot}}^aOs93MfimMrm!VNn+GP@n5i4R0N%Al+7G+Y?i~x(~2hgV&Kq6el
zm<`B8e2qku-}3I2^L;%l$RQya6&3RNNf7`{7qVF<1pbVo&u#@X<N*OOk^>kra281f
zJf6P)aqF@#6mY?Qp!@HQq|bIdqh~YNble)+G9tgn>Lc)bFr*vFHW_(iF)j9t4D*y3
zXpPUJ{LU)GXZJhZrR-=nrUIEm_|bYKCyrz!n^i$Y+PoHMwtalrf$RU;QlMRf(Er!o
zTtu7QJImldJug;I&v!@lewr<Y@31F+-@mOB39ne)L^r<r_yeD}x04KhwmA2^_G^b>
zF#EIGCpI8OYqzj=x2{cslb@pwr`b+|9B5cQ{z=n+q4_htQnrDXTQYqooymO333iX5
zQKV4SU3Y%svD9^S1!H33J-izhDv&VsE)Qr&YEurlcNicb2c*y(f>-uNbb&Ea50aPm
zyA)EO6>kX$2Lf8lf!;f@ZwS|5Zg2y6k8R9Lf!xf_LZv>%TyZXuHl>L6k9HIo7#oiZ
z;Vaq71wC5PNW2r(-143h2NPbo%nZi3<FSMLG0v#GcM~bL^i%%OHh{neM$1+Rsff}Y
zF}sCR(#z9^h`$;2*_!286Cs&?jEC?c%!ek?(B3+?>X33<@m?-e;Fu7{xN0{FSS246
zGU(P6`8@BY#I8C<zwiTt=Yi5P<?i#vEvtKpW?#+p7<BFY(_bD!4}qbZql>M>gyIsU
z&;eLT!t$T6W(<F?Bmb{H;3tSbyA$ab5Q9CR6Z}$pw%Y8G6M*gfF^>;Lkae|?$Gve?
zKG|&H8es>T9SUfv&C_aT@h~Q6I`=vJ@VpJQx}O)(EEau7cpV*%e}t)3QQM@~t>J>G
zDwBvDg55X=ZZ3#$*UsmH4qpX<<fJ;BMB3QZ;Dq1NW-TU2SQs}GZM!z+yY<n#?c5%9
zfv~73WS1xeTGAq$I_7SbIzG)lIz_Ul&f7I3EJf;!&sXm=$SPVZl5W8Y87C+=x>Q@m
zWrpGyO{t_prN8#=^g$AlG%T~K-lDy{Ee9D^Ek^~L##a{(a|1TZY_Sqiqcf6-TXg5C
z!9Ynas=>E2LeH~2X}N*ls_4`OtVaxgozLuUDrDmM^9@}67kiDSL<Rl&9kdn_$<H9!
z4@_qA)b&>81n#(=6`?5g28lt%YQdRrGlH^wN^9CZrKln#Uz7EVf20FVOQA2|SC+`g
z1%8jXulCWVo4F-Sp_DLC`;xz<rerI~{<Wy`U%8Jze&9a(B=tqlSsk`9ZjrERagLfC
z%+_V5@AOWt_f%r$kDX|8NyfD2QRz6}r!}C1L(&cuP&_WKZEf4-M+fjk*D|L9fNl))
zg^E7#49}c+6KuIEt#hJ|bm9xn<Wdd8mp?6I)uIM-G@Jz7ISk;Oh-+hhXK8`had<Zl
zQhz*%#(k>}cp+wiPtM!ur1;mPtOH-oI$zIoq6Oz;1<%_#V>hLj5Iwhot<LT-Ov-dm
z3MzCXaw}-e#69;XToFYw2|I-YYYy*s%DX4Rc?vBEQ#rWJuMZ8C$1LaR0yDhl`Mz`f
zkoMzRZEJpT-T(biT!<23wqB;0zujyXB+q-6Mf@FmT3uh*odqLrtdJUK?er&J^yGB@
zqvwW~R2npCj1RnQ5=6iYx=bg#^GSkuy=AIz$InNL-QC3^SMaMDn6E31YF+S!qItx*
z<a!-+L@dYi5nV+L^Z@-Sk2VYIrqB;DheRBU8V3z6FAN_3qf%brM=+JRkO!lBtjw=W
z&ZM0mqq55sOS|jt8Lg+|qw)_{L#B7U8{bjIGI%+e7Cu-Zjps3WHW1F2F?Yie1#P&>
zW!+V3ox~E^TUfK#N5B{PbpdQ>fZGizPfz_R58mtL;XwZKCHx|vz-&D2^GtHv(QF+s
zCej1ITTihvx&P>NO#-m8QKk7k$wkFDU`%0|I)mgZ1UfO4m>)bEB#98{UtXZ0fG{w2
z+l96wFlMSa%yUdo4<8solf~9orkHisc<a%w^u^RM*Zm;gO$6I$*hmb&GihXSyd=?_
z5QtZ0vN;?V!2D~xUUFP8p%k9KT+e(2)X$gPC)H-tc_Dg3-fcdCbFGw0e#arWMPN+{
zd3oK32YrLP<vPRkE^m>y$3o|V^1K_KSbNXjqpnEEHLO`lk%qH|qu8Zg5@#I=h)p5t
zC<*ObyFFQx&X(kxoV)Ox++~c4(fj)Nov+2WF7yh2x!OO<w%=i6y0m33U(;$F9~Gm8
zwbaY+>s|l#!}@Ms+X3dYKNJKs9L<2jd_q|Uhn)dpUY?u62L(c8M~BG3!NJW{N%Q*m
zNo>@x$FrX~W(9)yJ;^8&Q1@VsW=<BytU}&EKgk5jBog{Cf^zAjoVVGVT*1bR?JNDL
zjswW%>A;#Y_-i+gP8KR4;;E3m>h0Okg?)>Ld&|qg1<gt}=D=I5UJ1{^C4PYx9A9G*
zKSbj)p3KU$q08?fCKOMjb*n=j>21;T;-0m_OY@CoTG#UJne4H&0RmZd_8HJ)5v!aJ
zV3}OQugroN&zgEOzFj>k7RnPO%E=qK4zyV#;WC^$Xb=C}uKvple?@X)$IX6$t?Od<
ztH&3JPJ5$t#jc1Ot7T(T)KhzEFVSCVmEB&?77}o|TKsU2JU6WK=H2{IR3Sp>;A*4a
ziD1bnz9b5Mp5?n8wb`88Gtw-nSf|A%B|5G&!HvE*(6WGLArFCv&3awe%}EYG$-((W
zBxqmmFU4GaHR<(`qid3a-z=owQl^Vv4oc|i@ZP>c(EfG5L$-;0N{H~hJZg$WW;$K@
z9lG*LV69o%XPw*jGf`mAQU*Wr#d+)ZX|UKT;+l~?+P5E&>CK_2(SFQEeF0cysGYHn
zdn`S;`iJpldet=v4053ayt_|^yC3%+sOSBeXub?#K$q#!3uYLv%~h{-h({P^D`s!{
zn9WrXKq?3lY$@yCo+EohGV-j*XQaU*;k@;FUq-xgAf1=3;n53A%J*QLhtaR}PK;Ov
zqpC_t?CJn7c}pKrr@6;hZ4)W!ZX-IRCpgg4`Y!+3WW*xd#hU(fxxaIRsp@iQBr#Qo
z=jL%5cEGw+b*82pX$nBw!BQzpAYs!4G3XC;Oz}XR$?+im%Q5_y4g2Q-wZRxn*Jj9T
za*<2MB|7Y)MFIp>UXTnmUuAJ$gLj)*S!%x29QfU5gN@5(0?a_hA_7kP{BAL?pN4C-
zcAJ$zLGoQS7ssWrOmE}3Lyd3TVksOq`hO96%KD6ZC>LrqFcU7G%wl<$Q(~pguNB*c
z+eLJRVg9;Ob6Iayg)|v<0NLUAk49CIZ8{5pTux&)eCy3~q01vG0rV-o8sG&Vp&B9L
z&Q}#5>w1|BjGj+T7b$(W3b(?IG$xl%{(<uRga;A4W`THjuDMyiu*=|2E@P3Dn&o?6
z942115)idk*?As#!*J4-#5`qnk-#IC6bSpkf!ViIYs#?`+%k^q8vVWAGSaF_mqex9
zBNLfKwb5;1zA#^AZvj$U;E@bUtSh-|z5NJ~;2?22OF6~8wCNAmdds)iXj$S|3QaUq
zp6X?H87=4`=-0opp0xw+3HCT)O&eu3bJHfYE+mAi0FTejs97NV&xY|IJ3?Fp`n4B}
z<W{Mtx}R_TokWYny1uDmXMo=BrA?;|TSZYoEraXx8zyoy3F^zh4Or{()psX#oA3PJ
zv+q_y-JV{)RtGU&>iOz26g-?TB7MWdy-zeapX9R_@v#Z5_2Ycb0)Et}>AF@a5ej15
zKiOU~Co>ZHl|oYbSiQ$4$N}MtLjYT_{+(q(9U161V3nL$6}$1P$f+@Kxvb}PFb0W<
zhk>C@Oj@O9EZ9eNS*1NP&1rW|Drb_LBy|HV0TcmzB12fE@8WH=KsYGP-!H%yq3uc@
zQKy4$1GfgJ&H@|KwzHhjU;SzyU9l|bZUC9bDwFqMdp7|RL1ei=PNGQvRiMryzR%Q6
z>43ipe&pr(7bJGh8I=GN$WD=BU0tWsCIkt0)b6%y=<zpHBBL&g_6{Z{eF+{J##^55
z(!HO^tk)-rZ|W_`d&bjWxI-^|zC2!QaJVfzotR(1FD<NA8wEL`{j%*$E+1%qR(sK(
zVH!*aNH#gICpqt-B!ymP8SKuoQvQ$AQbHNSS=kmQjo?<Dus^nU9PNA}aspGGw%R*e
zC)XM9dFAIfcq~K-!6%WQJoIkzUGaW^*hUw=%;BSY-q+va>jOl1Th~Wjbpx0`Dwb6$
z(%SShX7d}`bl}!g6pG4|HB0!-&2cC&Fd_=<hSd?px|E?@!OON=(q!hDyuTMJ^cx{i
zvM_>cOxXP@z7X(;b=8@#^M<Px=lHL;??LN{S=(_Et~48pW9<%Zc|^b<$xb%erznWX
zEz~z98XZkL-5=9752CFN0Jyph(N=Z5y&vd{&6s!nz1=P~ZEbaJ?(DJ`grzi`zF3Wo
z<~z^L5zlY1AsNp*O~){jPi!y%l>;JlBHK>dcdjQ1_a0@&{3;1%piiIZm^_?lo_v}u
zpGQ*f2%nx^QKjRWCP2SX6Ogc5;NE7&*d-#nHIImWbtP8FXjLp2<uB5_5~%y)KVBtu
zv7btPaa35$<l6!1^STB<5S6?+Pk-|)fBj9-&S9h3DCBNk=5wIyc{k@qR%^%WGP9Ki
z7TcH2ji<M-_3ItR_-IQ_)%UL0QQWV4iJQ8jxw*Y}?}DB)2H#t5+0PqCdW*LMhxor;
zl7F7OLC~**L_uQpPhAl+c)mCBO}&;2DLpnG1~L|iWb(%6t*<b0R`_IXobwIRnR*yS
zTB?@L&|+pxdhkKejPrOzmHtDqq}jT>)tjX3*ud~GbeS&Rn5D(!w%xF!CY7$1k9CZ7
z={}o`14iDf_i9DhvWU3QzzY>9x+|IWP=li4bxN}L$_?u-=~EoB5K5avxZ8%wcQ5SO
z;w#0?sY-paGA$EB-yfK&ifgmQhnpv_j!hyk^kq8l`hz~1$NV7qGU_|U>zcB4IiPHC
zX^6E4lEnAi^n2Ih%K7mQlEZbo@1ts%$eyt%>)e+*p^c6qhP&W0*IKgy7RZfWM~MW^
z81r&1@t!#9?I#s=Yo*pAEM3LWH&VfSBasy?Igi))+`oO(Rro>i>=)POm`~}6Ab<yO
zyB10^h{s7cFX-BWHReNs!&qSwU{HR8naKLOH&NMP8NvJfOOg;@g~{9Fz1Oqzd&lq_
zN#t7-wbBA3hWq130^1FdiG}6H1xTK=@Sfv2io@BkyhIW}=Su*RCw$dRoIC7)aU}oR
zwZZK_#<G*D%|V=7NG|nw@wyX2qWftM8BaDQ!v-Ou7{Pk>hp8;gRs2YU>Q_7-e5VH_
zQ{eV;1sNK#Uo0BE=W+<D<l}J}?djvi`5N|}Xk&fOrgQFULb8i>B2PwK?vUZR7xn1k
zhFd=SA1r|Fh-3HQ500CWVUSdcU9YF7Qn03DZubyeG~zeA9e8i=e`qo2Y-Wx$$PLGR
z*b0b*@orJr94C0%3CDLkTgny^#la6Q-x)ahd~>W4Ef%HQa=7Sp>yJ5Ts2hQMQVwl7
zz4x<W)E~D@zO6Sz9D?U~r(Ha&=jGWHz>Q=7__53Ga$aU+{RUsq-Li-CW=3IVkw0H!
zbhqRH8t!=M<$COBUX@_IV4-+t<<jw~;n{h1#f7S!faflnC==Fbpr8folvcN~bNFOZ
zIDH_d^Re{S0hNe@S@J-v$$jSa00QNn&%Vd%dB;910(zgVG<Z7S`4m;q+Lwn)M+5UV
zknasFl4CG?kL$*}B_RURudtWD79TKh@sO(QKGPlE__qpP<hP&OtTb4I+J1#Axd}Gf
zq=!B0G(fa8DHB6=peatwFzR~<uVDDo$z}m+URADOJxZu4uIsN*|Le8rmjlv4aAVZ{
z^<vUsB#Mc_rr214!(qFh7=}Wl-I%fz!Lp&p#ygK!c5}^2`N8C+?AtNv$!us+1LL2q
zwgp+tPO?<P=z9e*-b^k5d@d+QI=SQ=vze>{5r>*4ptSjRtGLi9hCY=?FDQGbm@1C$
zT?b=8<dZ9}GjELxe<4>mJRPKZ&#*}Fto=GD0da1h7IHxN<6(&879?cs5Y)-SRi~)+
zH?4*`+;PF%B%BluU=9!Y_Y>EOOT&aDAtZ-t(|W`EstL2qNCOf*gL8XWAzU#{p#v}i
z=Cr0ytZLs%dFnXWrrUBpY0)o(?W1g5{Tqk}K-y{ZVMrU>r`T*fX)_Qt$zy!PFQ>(R
zXK|1+J|vBsh|4{L|76|$WHGW&>_(SUy^^Enq_WV@iYdiNaJEA)?`z1|B6|L<scG_U
z_K0PzsXscn1sWU9GYg4n0^uz=8@Y@C+I0f85#t5DmI_`C$!H!Kl4F=$Bce+pJ*x=Q
zYFD9rq<@VG^c7!s`i@XB-~Ai3%>2{!F<=zh9=gmN&`Z$Yh&TMW*Cgo&YCE^j(Ge4>
z2j%4CWO>{imsk_AfBUc4@9&0oON^LC8oJpL+A(444$%8E(O=H-QzF~!$N7DY$+H$T
z5E0O4pX;ZIsNHR`z4KvfExLT<THzg3v|vRrw`1RSF25%f<z_#%{DwCt#=5(t0pnQ9
zH>iHs4fdYhp>EurtXLox+<x{mjI%@j+3vgSDwhih1b?^obvSl_yD-veiHJi6Y1`#0
z{Z{4BO=qn$T7P6~?;3n&vs}~vrqMQ{u|iQ8IvA1L7nkoLZ`>LZoTz`?JtxF$kK_ep
zCK#a9$|jj8*V!6ZxhFCkg(fm!_YaHxU!NH<cwA<w&$oT|!eQ3EW*|0v89Q8uR!=HB
zrbXA=oX$o@`-ny=+kR}%GxwCHu8DM&X;woz+wRx*s^W(~wP9aq?dnfdl&G;i10IZH
zq^v)YFV==Ft=4BhL?c1Wt%J-Rtj>K<(Zgr;$f+-%Hz1|#&Qiird=N~W!2B+wR?kMR
z+%DWVr)Fl=hFVlg+ucMjrekesoRwF1ZMu+<I?bL7J+d#;OMW((o_cX8ysib%xAYpH
z4=1Srjm`_~xkej$oC}hx4Ifr$1h+uv>tKa;AY{t)mg4{L|9fjffWon7;0r_eJk^$P
z)yuU{@<;~uM&vH~aT8)PVOeex_hK}u-U>Q_4u4_8cdd`|sAgGyqdQ$T2@wu7=8lu=
zrEWSHlQCmd=sy^(Ge4Sa_XM#r#Wd<JC2%LY?7!Kt9!RKFDZapZqxr_%fX`!cUhc$9
z>_IwfETfkpN)1Y(OX<teVgM9O&sCDsDP=jWjNifOw?N#<CAQMghpRJjfhaz$hF#PG
z8twUZ$7)A2f;wBG)rDKGFILZwRBkU1Y(kveu03IxI24P_U6a`mYc1kJp(s52`{2~T
zt;-$L`{PolBz|53MX_`Fw`^2UdTt+@ey^IF6&(P5MqN?p56#3UJARNJUltXgpN4R8
zJ?r7!)GOFTG6jb0Zc6S2v>PgCQ$n6hhkkGY9>ep%cYqVIRza;iZLq*_2dhP;+c>?K
zjr2SeYkERddHU9rP_HTr=R%-W2$Ejg=fYfjJXUf9)^o?tHj)I&F|fM_wmxO62PR@p
z_zB!8cd?*v%HbG=aB)|xRl!-VEy?8ia*;uRpE(jqNdNM((liAkl>pSK@xn==Y%sHI
zIU1FeB>p`k^AGpupN<g(Pxa}_fUi`r2y|KnlRCp42(Q9Tt)dJD3ob)o%=aOj(Q%Ty
z-c+u`u21#B;v?rqD^ggd?~f@g?{*2awE}Oz-=z}GlE$VNtKx)uO~&WNZm~|)$vdv|
zk4FFg{k>2lpno_WM4r58K3|C1ffZ$Ht-hazp++HvmhRgi*Q45H$=B1@igHzalKI8p
zgu@&5d)L{@Z!kzZb})EaB0zxuLl<Vq4*wWwGzk27BK7nf&q+VlgHQKx=iR`gUBA5%
zYQFZ?_fuUFY=Jy-#|qzVzK=J7<8?$xgY}HTtIvb;&w1D1Vy@dmu`}hqRQD3Xj^$cf
zLU+DxiE<y&h(tMUG&FQgg)T=zkWKHe6ei9kwHju73NwFvp}2av82DALZTgmltp6kj
z$S-wKW-{Gq!v~!%^$(Pm?w{PO+n$tIxlV&O8oM^TsYj2JsL^|=4j0L9_Y!X|E(fOk
zNTd*tiR})e-T_34uML%uKVy0r{0kLQmx$EpIvItk83izVIu|vN#aO*9W#9M+Xha$7
zYfG3$MGDqW_TP6?3K{0~Kk^}fJ1>0n=}C0aj}fu!fV&s}OMQ3yZP;?53Li?yc=yk$
zaMQh|>fX`YY?y#&f-Du^;<QaPVn|!NCgpOmf9~}EaODP3K{{>52Rk0bVO|*WPY6~9
zhGF*G5RD`xi)eMFavd2}t+tqpmxL+!-G3(<y7-im2)YRVJXfMsx|ymDB5Ztpu?!&*
zhr0XJ_F0tR`Y<~Xu-J@3cW1R4(XWjgC>dM>^?saoqi#>*`u0D)Thqhm7tA-YB+>mT
zH;3a{v>Un_!z|u-L-9e{8X6wnjVurXPH#Y~Yk&UkIPzo@^Q$Ki5?s0@Vf34(>b0n#
zayt%#>G$%fQnl)ea2jl*L)7({yIeQIIy1%Vb36h4Jc^S{dIb?E4FV*k!5~92)|tbc
zj|gD?AF<ZWe6MBN>LYz~N`8R>RQHJx;Ubl~j{37_izyR=(f&<MlZy%+@Ks*uuRJVQ
zPTmLOPKUG9LmqnWE#+EQ2P1|46}mT*jg^GMaa*Wyik2(2BCU#43{O6zh;PKa#j3&m
z5jZEAR0>4{i8T4a&L`U+0FAz;hNm7pnBn`=?SssRJMpGYdk74apd2Ak8+q3&`<Asi
z&C2*-2wh&pDM)30-hhpgo<8zf=$93i5%EgcS#6Q`M=4@|McvO*5Yd-dza<L)>+Jq*
zM}yw@Z};ypQjKS*-X`NAx|w_)!g!GZEpb>#vHq?`o>qH~n@YQHl=Np%==kQWrI;WT
z2>VMN3SiiJp0><9B47eqX&*BjufOCvw<x+ZyyoZ3{GbaehX!{Dp0$*RMTdS3S3TYS
zLe2%6{H%aAxkZ~4HJ9U})b@yQ0cw+y+BTAV=KoX>Tx%8qr}P=wEt7)hz^(wKnZzN7
z%5E_TzA|2UqD6*#9!<uC8A&+$;Onzt*)0s``)chk_`<}h?#+6-ZXL2S;Ef69u&oHb
zImsm!(DTth_dF>MQZ4<)WJ_2Ze$7&6^$zlExwzM4Ih`S|aAZPT^IbQ_g?wtlPt*F3
zZ`?0!GWgyDJ~NcSZ>AJo+U>3Bd@H8UB5%T6P2hCVW=zN5FcuyV=e%hyu~!5v>40dq
z$gcyW4-3pvfvIfY@0<_hA6TKKrsBjvWuTs){&#>PLY2pj#XW$XPm<~Cv;r-%n^KSO
zpBQ}L#CuF))EA)AtRVxK+=v-IkLFas<H$Z1>)W`<Gz5&=@ml9heOVkF{yB=H<I}@q
zP4)Gow6vk1!Hb>rTy-;K6`=xjxKlOXZmsi)T9elUi%z=-_Ia(ojD~3jpL?0ta3Z6v
zN#Hk!W)H(-2~t3z5v92JHsPTOh^?u%>TZxZ@Erjv$2`aO3!uB0<NPYl0={o^zY+ly
zYL>o+giHd1`oPG4M*W(6aF81)oXhoe@x!V3cM|c<DM=z{X<TbuTwFNk<5~M*A{2B^
z+w~x5I260qPmMOZ9M66%%2k^X07{yZcP@9~BwHAIBAsG9Q*ff-b*ZZWa(S>|X%!Br
zGQyokgpI+x#ly4YpTBe)cn&%EoGTIObsllp5`5qd3{9m!?o0In`ekJc7z5TND#bko
zvxbgjO8;|_px}?kuaI&5lFx{`gsJeQd^?O*-K`j+P+oHpNgi{yqFewHzOx_Ike+(T
z?lt%OJI$O73+SoS=zv%FEWf+jiN3vDzeYzh`xe1Zb{4nxJJ*wi(f2S+0;cD4BM{mR
zVlXBi^2yMQo_#146$mez`(PIO)1{)znCGXM!S@P&6eL!9v=4aM`Xh<l7dbQJFeIDQ
zkZgF(4*d@=!R<Ap(~xi`<(j0p#!jbH9Sk|O?Nrxic-<W<eJkM&NX9_X9bIjp`Ge>i
z*)H=>V=C*F9q%DyDJ2nuXZ($^&YjnQREE!UTceaj;t{oSoVzU^)x7IDmp;;3Ij`?a
z7vgYiEd6LYM>4mbCu5aMqpkFA@5&&HNu6RvTGZHMc6IoxB7{a!4D4{i9CM+K$u);`
zUHotiCYSqkAFScm&YfLtWeY0It}E>R*2Bfjc#%NZ`!+eR@_u1ytgn8IpxNirF7XM)
zBp%y!zPGHVZ_)(3wUY4qv+qMmHh|KP)>ZZ(7*M>q{1_A}vIeLO{FI6dhsmJ=8Y~NX
zINW~td3)@|tEX|8PvrzlLg`_-_y83mCi7`B%hz;7V{|I~ww{dNI~xGb=JQrhu$ny`
zq6wfrHDVJt{ENepwgM1sQe$@BAI+7F2J$(cjHJhaE+Z2lpLh=t;~@xQ37~&wC*i)j
zfY*@$6eV@5J*~mo0iXy}W0+$;-~rtp^>hp%aq0;r+gg-#A_v|w>a;i^0Q{x>xS*9y
zW}=F0=aYE@CN9BQK<kPl)ul3&WKFNrr(JdSTtlPP<&3YzCCkp|_x+`sf6JgMPx2VL
z)9tC|$wsE{3xdAaWw2^n#K1!<xMM_b+MTn{eIkP&P|cMYV9=;EX*kJN&PhdgJA2*=
zB}#K()bGFtBrLcK-NOIphnN8*otV&Br1&U~BakQt*{`l2MzCpDvaN`DK_Qn4+8@C!
z4-IY*`XF?ti}*smX(2sOt{u3uzgn6Y#ecHE;$Lx}zi@kIn#99Spyx?H-5JA`KLTo_
z5`Z5lZk>rHg8f*Z$u0XDbWi|`pN5qW0hy~r#}FQlSmyT8|6CEBhPcb?`(jzaIP%aj
zN}C&)JashAJ>PHiyWDQqN>g=$cOlHn)2KUYYt&G@4D^OKE{teM(pI|bn<|&``_^S=
z^Nu|OLN1_<mk@Ax^T<&;gyApd1K|ppa8)i{qe}PR;J?LoI4@qPaNpPvLR^+lXJIzm
zA>;Dmay!VgdfoX30fcLTfNn;XsT_eyQC4lZYwu-;=Fbm0GeGFR$_5P07wf+RU1%;Y
z*V~WsvDU#xn>lrb2Ggjr$#z0pxmf#r;B|#P&_wKH$HEH-?JA_Gu97<Sm+7^)s1{LT
zg2P+<<O>fE9|0O)Z!i+B*wX2?<NoLnVO(Q<L~Z$Krr<`y;T0vKn>%E%*Les4@LE5f
z=kdC?%XU*&!3p<Bq}PT8$fQA#(6A6ZP*0>lCmPaZv15NYw$5cmk-QTcP}SsY-SSM6
zgSMBCqc`53%6}V8`&|h4o%M3vkR_zqpx;GS_F+k6fSF#Gk$y3g>$#4N62P`P_{l4=
zT^+uy**0IMRUaNHdwU8;gxFfBkcpBnlh_dM=1$@|!1GrT?|*Q^F9iLk^Vvr)Wpq&G
z5z3qnS2bS(j4!vHDB6@d^zxrR(@wa#3FgtgfR~U`m*`ePV}<Y(8oqM~ygc@qn~%g}
zMm$9k)OM9?`wK=X<_S0!tUrp#&fjhIMqW$^`eReSeS^bMnw}&z^usWtrlW{|tm#gn
zkm85m0zw!K>Ee-?KfIgl<}CF~eLUsyP16P18ZrGC)P@f&E|v4y`|fYw`zQ$%x+3nr
z%Opw!uWm+{sMe20a;3D_elr9gI+bnsuT&py^EiD0rE#H)Zm)HHHv1_iL9y<_vFXr=
zuv82EhG^VH?`!Z(;lqNc`kDb|Q;lYH{{po7isz3G6B!-=!B<3c$?zf|5jq?#7-d>3
zI?%h;y7YOa0c7fc4Gr{cXM4Hp4(i_VEm3OHgFC#*)J83y@Zr9JSKp!LOT%2i3W>*%
zlCUY}hG-ZL<q2DLMYwJd(Hv8-lYa&ENvfma?Uft@au`|!l2gp{3x4uIxo4rFWuiq{
z4LIh_O9>3R!_NDArU_y>Z{PLzM}3ft!WX%;+iyFR{ub2$;P#xglhWLuMn{fL)w#pr
z0ED`+^Wj+9{>d)@xU*Ep)K1LIJav+Tg3o2(OeWy_(zaj%^6agvV=MiNpRVT)DLxEP
z@OXE-KeS<;!hZ#w;c?m+>wil;{pC^XaCfd-+O%rF(q`qx*z($fcxa>D>|la+rmenW
zCNGKp3@DqzEAZc7t#)|xI%3dc{<CQJZ_d0han^$6q72gI$Py-HobMO?&W;2*C8`R{
z^edkyxd)GD{{=&~u4mMcKx@MNMLlTZ8fj-=6>$>>c(yOZKJ<P&6OFd0<0Ber@^^u#
zyEZ|G>{E@ES7kbBHZ(Xl;7*+4C&SC_?W<4K5}<72-VDK0i#cps$d-WMGr$>i0~za~
z0~v#LAB4RWh4gJyl^6oc>{Wf&WdYZUuw=3!$J;a7cT@cw&R7>U3+<y9<cO1V94ex@
z5<K@02UHwc6e-CQVh}u2PqfA(w?a*x$0}^8`Ko}6|5(Z3d-FkMq}>zH1$rcZn*9wV
zTT%#4wx8rP5$|Sx78m1G!bQZ>{(p46WmMH`7%vE$1_5aS0YRlxknUExySux)8$?NI
zke2S=lpu{DAq~<E!ls#*yJqgpoqO(D#}7x>IbZzqKF{;ZVzWAw+s7Ar?Q?KU+HYz-
ze%$Rh=1e%o$Qf<`(2))o$XinMSnBHLIdIcdJUFIS#Ni&yi=KBPj%8!USzxcVp+FYu
zIPIv??CHqKz24MU?tWA$)6-dgL_i4+&{J<6SJhud9ZaZ%HuN5Tm5hIh=6CbY$R76$
zCqdH?(q;m++HM%>79mtTn9UVI$V0H+vdi&3`*11tO?r!V|6TOGZ&o}x25w6P@I2pB
zh;Q&Cd=508ATsd90X3cMA1Wh0B=Dc1wm}6UJNB<IK=5-Ih&qADXY<nyZ@H5+I{c}(
zvfoZM1G$Tg&HLf>3^XsSkVAsvaQ%X`MyfW|4FE^K&z8=h!EEoXo*>R<69hrZWX+BA
z15;T3G+UBUt2UHj==fJ5hRa-{(5LPo_?<G}-rrxyW=d!|ZC@b$sjO6)^T&7Da5@F0
z6`p^{h7|t4!j1p;)S#7xGnj~m_^6gNpCb$Qx@`8`E_qbuq(}dGIe7Kgp$l<JsQ9~@
zq-q&@O$Q2M?G2*=5wFS5o|%enA7VaJRQVbP!KUgZTGbpLcE^F6`MR@I$Nre%1qPFf
z84j6N9H7>id#_+nRkOrfftVU7tGc_Mtl&s0-%y?sI2e@L;r#P+w{0Ud6wP^i@<&{Q
zV&nHKzE^ZVAs$EkeGyeD;4k@O>uB%do&miwBFgg4+THbwxh^4+AF)xU$JO;CMpI3V
z1(vWdBG+oeplo8<e$BA9FpL<UTXj-9CMs4_X>{Fo+Ab7ayO>iXbj#I?S+Q|H08-;k
z<qdjM%a^Nu@K0b`{&f!;7gh39icJk`Kjk91_Sw%e<N9Aw5?1PuVx4dG*SM%-&({sV
zrxCGXRXZSpfGHW^cOnx~qS*Bux<UXTnQjH4ei2*7tc~C0!^cvW$K&~2WBxohwfcEy
z1u|y*oewuQSD@QAv|=coz1Z#RHHZR&=bg_lAuO(z%KHm?t5!jgN<>c>hD%dd0N_g<
zR>&2f(__%q;-N`Otm|lb#an^qLYLUu%9uq-b_;TTPu2`)GpjAywoTBp-P7!}@jX}x
zpUI%hf9WRm>sNH0CVMlk$toR*ad*=DreIygPidJ43({?1o9ZTYXkSy6==8nL1)W?6
z11E@__d_Jy{&(CN!De_*lM~c8GZPS-bb^)z?&Q)f)Pl*uqUarH5xLf=ojJ&Jt{AJB
zce2Mp*M;zMa$)nzod@G?uY*ieFA1Fl75kom#DyjY3!zM2HVv$YRQAR;;MO>x98FLU
zV#yAeO_RXVFFt9#0KH#g7kkD7>EIPz@MbRh_HFR1h}Hk{{{7E?3Cuy`zf8N^>_6QY
z>Y4x+MZpr!`FD7W?WMjQAM<4&Z@J%$Uk_=2qh=<r_2Wk4IPgLXgW?EtS!m6*r$6e-
zG{z}gabsS8*Ye()d&_bf$Mq~a=C8#2Y(5&H;%I^|1{mU-WjxD(^kMZtXrtcSe_yLp
zoV<$rspU_e&#R|nA?ORu^+22~<B=6vUQ)oWG9EkUI(yAUK|sK@UwXmq(R69z$=@$)
zUgIhaWrULPv8F0EQg!nhCWhnLmY|4Rp#4%iR4rBq5FH<rCWMa`q2m6~XGPGxMV*ST
z$l49RdVpvBWHU9f+QWgPEj6wN9h7iJ2d1T++{QPM30Xhr*8M=Ov$*Bl=&m+WqNKb0
z)B0QHf_WiNt?cUf?uQ5x0t$6+hC|UGYBsbhO4YI?!XF6<<~>KYvR`lq>ao0er4Txa
zrd9IjxW6=(pa>IMegMaG5mSr_yTkz8gM4w|tk-p2QV2>2j4|v|VF&LMg&zGL1-heB
zE<O6V0PN*UARmnTcMGMr7po~7)gnZ40bSR@N(GWyVR>TV&wfe|>n@XVKEutjSJI`!
z5J;PtR(Id>-TCGXaymzp$eBRk?G*j0%VLRW7`h?|FdndMb92SG-z!JK?mWbuLywME
z%JjK9G4cdae7WF9!+osK`;wy`R7j|=tw!O|tb&ilgR!U8|6q>o62ba8*XYt9XcL)3
zBXwFaTzI(IXSozJY*br}87l*)X^3PT2|Ri#xMHzEBZT1;ZVIk?K~h0gH3SK7LVyG*
z6PUf(n^axBByrezrBy!RL0uAwPCUE!gPP#U?3?Tsw(7k=y(ozF9+E$8j6B~S7t7$Z
zYaZ)JLs@cx9n2%d5Hi)avTXmLA$UaMCprNaF~Zh0sPdgx73Q(~>7L^$aIn(V1@ubK
z|9h?ny#-w5nN&Fn7VYC|r?CW+kI^fCr(k(mYxlK=7r$1Q+j^nqHUy7dFcFPGuf{s)
z(a-N1I#IaVw5rm+FAwKUu=bMLJu?$tP&OTlrS3o6Z;qcP|Eb#^`#HY3-rvudY%P{M
z%yx9$+c}h`8Tnm*N-%yV_~R>Pd0ju{&)jj!k%5uWQWf8tGHAKDs_a}}0x1pADFI!+
z5mu5jC<KN-UhT%2sLHu(W71t+g9Bc#*;-<^*M?`N9cr;{em)$TJb1macE;%pPK@kn
z(G9$?qXi;_t~*>zP=;u+vY$AzDmiT4{jBTwx8k2Bk7Mmz2ZNPo9nM?R)f)o;7CoGV
zDs*@6@6DE4ot3(p2H>MT4E%Z0M;#9`O8zX|=rlxEM_<Wj%Njz~EvfrbAF(bQ$^S6d
zqh9Sbl52O2dRN%ZDGqRGqfO*X%rtvAd0?k9l^Y=t^|d6BeG%e&?(l(5JkW|y3_e)y
zBvL*(9=^CXM<_sWHR%Eg9p6#bT{|HmAp*3e!M9$Yi8q52G8nYC^jWoy;LyI2h)^8}
z#QF+qY(q@y7xtHV{qr;IK_o!)#oy5kqDGMSM6mti#}DUo(|AVB>Q{9_k5O@}BxyXn
zCd6b;8kM@~HAlmupa5<@iT<|2&SjG$CjY}CSAxLhi>8mR%#s6U4JiC6lUW3M`CfS-
zeY+Y|zHw4bUUaV8<@{bh@f22KWVu@JiCNI*?j4c{HqyBVS2bbu)nqrxQoa)60sU%2
zDd?<Fx3Hab!>l5rUHm@*hKLcug&)BOc6)if-F-jIGHmc{JdKhl6ER{YMEgro0CLe3
zU{;jIJK-iP7Le<Bv|5iO<N5|zq8osBvN`4BaXVhOEYDv^b~1xpwPW660<6yBI-hd^
z9zO-Ez3mXG{pOFG2;OcF;XHDvAc80$4B@*o3+Z{hJ_oGyc&XuD>Z*U2wvey(C2t(-
zx;94?mmKOUwuLOuR5Y+UQK1>y9_*+rA%<hhz%SJ7+sPRCsbQa<o}KqD)ZVu&70Q@j
zZD1cC1r?;}w@6~Pl&oYlX2=rm%rFseja}$#9>{f!P{fbwlcC*S_OmDY@v}$9{I#}e
zr^{$W!-gQChc5GwlmES6h_RiAD^f@>YEAK^)l|G2fp%3WI_&O<$5QDlLFY+NTIIlf
zB=+jf7&m@@K06~z><f?x;Bhn}^Z7^KDnY+{JFC~3spESA0p8xT;HG?$#494CA#gTq
zswLyVz=MR*Urwu92{PX$^*xA+WYdk^P&YRz_}m#;{uDI7<Z;#vgYLv0*b6r)Cr;Vp
zk}4m?c>Y=9y!t%ew}s;gN;v1Frzl$bb7MHrCkqM+T6Xb&(!Am&B4X5p60Duc)Ff<1
z?Y^F8eK+5*1zKhg2BvgxgCBInG|U!{m9F+jLjsQ$NI7i{3*xR>wgf>|752x1@BedP
z{x2enz|Ww3bQ&&W?&AwxG@($^R_UET?}G(|a4mS9dU}wNd|>NZx@E_7;eG+4Lw8y;
z!LEfv!aa|Y*ZQ={tM7vdnf^jTF7`uwp5@V6&oByPl&LKDi-bIL+>8z|_>{#*!%VTD
z(Oi9nl-+)@C)>dKnqXrM!98wj*2cC;o}ruO6`o|?>N~m*zKh6R7t@dioM!(e2<+%o
zhaegN8Q4GAt=3=@f`m;zFufMVcrO}}<K;UTXS$3S&YNWxW^H^46$J}qvEj0})%ID5
zqtEv;h8Z5@y$;`xFGrLi4Sv@`=KNCaQ6U0bDxozyMw^;qlGXV%GIH>D_RX>O$V#-A
zY)yLT>v`EGv=M1~tTHthzlx1i@tQSxI19JD=lY0$8e&Y@|5$`*=tog?cbd+oFf$2W
zrf<{Sonzl)eOWDdg9d+^_;#Ll;=@kbFb0?lCMs5Ot8ZHfU?7RJy0izKXS2iWoI1n$
zh_Co+1m|#}DIzp9^hrJqCR8=v8AY+D22Qy~)fTTu{%aesb%kcoP*eBB60<k&(h^F{
zDme6l?QL#HX&Wbx&*R9nVEjO8)kk{(1Zd}hzSV4dWYFe>!<z=|)+^+xL?xJt?F$t;
zT$=5kY)wO0WW1V7su#6Qf{RRawP4R0DO#Q{24$7!>M$z>w6)VXW8HxhP%0$`kmr{L
z>@!ZnaJ-Cs$gZg%m7CLtdt==$WF!2sqF~tZc)6pc6a|~yw17Ndtw`z3e~?LFoUIx_
zXjzk5PV&zsjq}g2k6bo?RR_V&wx1dJ@7~4sbaDJG%=m)4_HgR}uQCPK*C*QOq$2WK
zg>pEiKotqKnefEBR9`XIu90e@-*%3*#WGM}$YOl#mu<$?rpgvHW?j#94XxOaznKVX
z$+A2-YlAPZ_bjG!_21uGz>c4-2-V)uGS#DHa#MD0ZD@GVh9%)+_}-LG+N}M|m!RhM
z)m+}C6r-B=p}XdZ90A&v3^(@Z4h8otuA~<#I=x?@S$yU@zZvGY$MWD99hN_R%ALCa
z>M7RIFc>VULTC6`>yX~X1gXCbeARXEa6y2uuq`nm`Q)i?@iK{l=CNZ}l89C&4X;E?
zW!^+^#HqB^$8B%N@bC|}W-CX$uE)}xirLWM@&3b&j6cP$Btg>K7+l5srKy<;oq3M8
z2SJHq!bWthdS{45SXyD_v^M~Az~}PYn=W1S)G>XTk+TCrw(Bd*eb)zNT_65&jDWV$
zwx~h9=@jCyAke^71OAARRuOJOVm4I~gk<e3G_f?HYWOyj-V;R~0G-C$VJbO19(03d
zN0C;&YTQxV67nOdNhc7rLB#ocK{O|R)9^wL^CDmP3I13t^4y4Ia-KX~?%->}iWNsW
z1c;XN6i6MHqXkBRryUCby4c1hCN1-R0^==V%KxWtf-q?BcR%*gWXCQPzUI}B3eXpG
zbD7-5czY^^r`x`h(~2bbS&~Nn1;P_X>V+%mi;B}P^YL6I{}F99sW_QFDq-b(n+b5=
z-9q7#3Gty)6a-CpA1d_i%!4r=BWIj$hGU5OSysWO-Zc8?qkK*Btj*&OK^cPzmaEtK
zLjUB)6EtaGgYR)|-fI8zk(ui<@a=#(8h7V<*!7{lcIp0CNle@9cP!ANiM8Z`^6b+}
z#JqLD`KUCly6i>Oy4`5uIHNqUN#O4aB(!;Y!%}3G*6k@p20_UwGW6EKp_hG}J0ssq
zhj33$QMSW0>dyPG-F5szGGl2N0mHiCIZ$$_4R`Ekm=M=zU!ZbW<98zun|Am|s3d?|
z?!XoQn2;<!yj(Mx%}S%SzQ|c3z*cMtxk&mNj4~vwM+U|P?ghNhWoKfUE>Q}R<5Nyd
z{j2n{HUCUkA%R?AsA##Bim4oo{zX>>i4f2PWQ<LmbsJ@$GtPhhtO-l!jRT*o*!B__
zI&-yN_w%44`KS>z`H^xhXARQi`2H$xRIAd@(!p%SDQC*{B<dfc3gRReqkVBPlci>N
zkE!k>E()UvxWoDNYmtH}%FB!SrzdkhLKSVUjl;)}(JId3;)s!zc=G9QacBoA!0E{~
z07`tImd{48x$O~XKmV69NSKgXv7`0wbT3=r!OJn>NZM>TdL?D8AlPc^?zQDVSC3-5
z8v?*wnl3T6`dLK!?`bK-<lgyHt^Ns{{agHXiG`C_ySK=Brp=d!-?gNhg=NU`#%1iA
z4lx<6zOFC1xhZ-LQb9;y>Y~T*7KHQ7FBCo+m7Xl3$`%QwY7cL!9dfSx-wH10yl2q4
z9Lu6k<LA-i6$}3U1L!p7pT(WIprE8Q@0XBLum~xb+2wIiJ(q^aq`XMONj^R%QJM}7
zu2#hO`IGPkJN55<bB&T*wdtD6fJYg#8~7DiL^)Q??A*!af|BE3X-K`5*t&p5JMB`o
z@LG>{iC#wL*Td4108E<3qu(;vx3GcUA91JPHgrAmX`R*OIT<oPIjv_N%4t9y%wa7L
z>~FMNXs+;FkD_S?m`=WDlXBt5#ZkZsj>R=efM+8$?jc~h@$0GAdLLS)jiV%T1QPzU
z{E%XkoyC^A>12ZbJldCkkqWO=OB93RRNctSz$>7}?RUdbWu5AjAoS2Y%wPKql>7dr
zK7ZlNj!cUyHINr_3Mv$u-(McDP?(QpqG?y`DTvwM{>Fc*1ASsTRlZDQd4oYGf0~av
z;&XdxS8S@Wy$&$<qk2q94u|EJioWII!ZZfUHJ>vxpZQ!KDCMKmhy2$^0Ect(3A&Lm
zoU?Dc!c5rbshO;D39c2EfvyK$vJ*2`){yHNx#q_;X&XGMq2!3Gyxx%cmt_k<fei4+
zt!~{!C&pj)XJ@-}Mb0+JI>><XHE{m>0b9UPW_!D%%j$ceRM{QCS|FoS!u<NDVk(*!
zwHfzhD6`=AB!YF=SJXkob|<GiN>MqvhU$6NQy;i?+>_<zAc$a#VWqDZ=D*$7kAlim
z{)LUt;XP1wTa1C&z=@O2*+OT4KZ{SQRzxGuX>@SW9sl@dy>p>3#<{Zh3w0qVbvepV
zF*}N&8Gd^DL9+|+=^z^$i#p_Z{n8lP7lQcR(t8}-Rwb*x!!1#lL&@<Y#3W)iH6ba~
zEYVIrvS=502R?&`^MpI%;HHaT?rRFd51~AR_flN4iL7#TLG<oQe|PpiLSq)9Glc}4
z)(BeaqKAHXv%M7beF0bum|_1~m$L7KNIf`Du?mGW-6F6Z#Bdz+_4ReZf#Nx-i^IWX
z0!O9P9s^l&=(<nQAuaN|-b|&uejUD$_ev_ix&c6ui|U@v%=x76SpMxyh7e`pG&ybj
zh}|5)_?Q);NzH2wmQYrgQ{v$aE{8GABBcLdHULHT$`KT{%zP}iTJBs5fDC2}ycRx5
z8$l10yXiL%L(j%w5u|7n+GDC-Y(XM1Ur9auLYF(qM~Z}K@1P_X4^nHVMP2Jn4tg)-
zGDZ&4%X#$${mwpon|347>9ETnrj}Q|O!IiV%A~LRyGoP%NR}%#HsA=(aJ!Pp()n+2
z<VQ8T@_(gN5Q*r)e$HYAp~VDxYtHu)VTg)X-gnJco@fLp%dlnk9)kdZ9jU^P*#;R-
zw{0Gqe3m_k0vSpw{zpG$a#g8v1tA?iG`*u4>o$QjOq~HMUl#WUea^n23EofpbTi(#
zx*dIwjdXhOMQA?_hsi^ao=eOwVGBl@!$wbSs}Wj?TYF>ABGGa^I~<K}^zre~?a|TY
z9${T5%nP)06#U#I&YNd<<aEs)iUx-OeH3;NxI69|@+~f_+McEgz*D@K_u84D6vkm;
z0a37D<n)p&2MqwS;;9BVb4>5LKNEMa>C)MvRd0!tCR<4{k;Cg2y`QgBq5b}O!|+TI
zfH9^TEn(M#uM>o8jQjD8LSoC+H-&OtA`<j^A9Q|xp}}VPbJ*<J{d*$P1)=QK>Tnd`
z$DmJ<cebOz5*B8av`Pk#KYu3a2GZ48bT#w_wCxw0;caHVJH<9;N94OeQ?3D<+bGdC
zPV(M*uHv>;IbR&j9><d`Px!yFWW;P?)g)x=8GIq*ekoy#v8#bFE*(zH+(->VYPpXZ
zi;1f023z$>*LBcplbHAPM$?EB1k7J~(o@ugU-4=qV3I^8tbF^1lCMd?@q6zXr$aWm
zqU$RRS}3otW{b;MGO5mF{@O*)pXy<Mby2?%>fAiyzorqo6Tdzt<L0!@(c+#U+s%gR
z_Ir++Rzboj1@-S|TlCsw17Z9xoVSq&b5`m&H!itdf*AR4zhh}Px%*QvE`<Mz4E%#O
zl9q^Et}&e8r*Px=H`2x~lx7o&>R892Fqj25M7@G4jalp=>p)(J(?Q<ihttpW0i~9@
zbSW>zktRPM{`OmY4ou0D$d_h>I1~HtZQ;yfY17vcWZ}$gb}9wTWjS8*e1@Y@Fb`1>
zj0!?0*m;oH)uSYp^}3)(^Gx2#QX4wIx9>^#R;kxU&Di02y8fJ)xO8v@uuv@*i=r9m
zMDS#6I*qowN&7>|44@kXx!7|6HIc7`0j9?&)rS%m>D{4^!#U{rBbGRRw6U_SH=-$T
zP$V@qm1qD@_WTTq;ppmgL)}A9>z+aVJKSVvs-2XTF$5Hsg@RsoUNhDg&p;)Z5O+VH
zWbRHj4TOm2)_nP4YcwNWgFR(9nIX02dtb0Ll%dQZinsLHf4>F7^11$g*RR!GkuEem
z4;xTFS?vS0A{eQN4-&1XJCxF2c_;rI&MhNQmN~HpQ06-Oy_u*jyfa%K7eaZs!f`7F
ztnxDoW{s143hCKOt7sk+3paQN)dCucwvhD@*;g9Om>CUni;*hDcd(_`mC4GYiNFbX
zW+Jg#TJW{9s8B0yXA~-%OwRFB1*TUp3GusmF&9YVl<N2+WOwRyKQV0>YUrz=fCskj
zi9EXKMqi-*G9M%4?I~o^`DJwNyz;CE8S9^4c;fl?=tQ2_X9ZDX4Zd2HVrjrMz+3<R
zydu*^AMq<H@O4l;*~Q7-`9HGV!MD%%Y3C}m)lh&?-)O#8bE_yjD_x2jI+!^5=Fi1B
z)?kJPzh*3AyMH*vkN)8tW|WqVCLr~(##pQ{c+==6mCFP48pVS7SB);)WAdl*I7f}+
zPJD$k9Th7Y@Kwr(*zP=49hW4Z@TzKkx>H@8K&GsGT$Eu7xn8TQQ}p=DL@;|xOYe4N
zYr$T~Q$9ZYX7z%YlHfmoFom5U7kfXEJ}op}E5K>xhZ1J{$m>5!Q4ptpQCZdhmVtNN
z+p!h`e|Yx~Gegw2YZl)^9*JJ+$>Y?I{z%rf{BM29<25j%Y*xL;BCm&*D0N1$+1lO0
z45IR5%s%rR&kdm_D@=vC7<Rg@8M(~+p!*$qW6xTBa6_E<pnxx*Rm)1LxLSFi@2&_m
zWd|!PLh|eDeGsJg|FE)V6DR`aP!y2b@u<JCZ(l|%oZo;j1YiVoJiY$05Dr%3uV`tb
zggeUu#JG6Lc5hQ6<ZJVu|C(Ytx!LbTOFI>;Hj`OFtO3iK<pL*Kr9rAeM7vouq$>)q
zz^FTzn{%_RwQ76$GO-auFo?&+cx_T@*XLJ`4`h7Pd9!|^#^g;b8R_WAn~l}6tOIqU
zha|e11K#|IgZbT;ux-IXUH493<CzhIPlu$Bx>M{;K`Ffnfn`<s%e?LM4H=6BC5={7
z;c`EFwoIK+hf|mk@L^Fsmxl}U_}&(opCvYV%%2em#ve*5y+MNG@4UI#m!idgGtuei
zbGh+)xm28gb_{6#@f~8goz@W#mOB=}qBjrvrUt>>{lV)XvgKc1uc-$(*vp2%fw1-e
zbg-EcK-~7NMPB>6kLw!`BL#LklkloOe8-fa3vyMGyR&?g#`1vCx?cFoXrwqc4l%we
ztN755TCpY{FcHs>3+gPG98Yos5{wjhad7g2KwR2$GNu*lyDQf%!M|-+np@Be8f;<N
z*tD$nusxagcQvDxN;ld7{KK;$l497SdZO#DdWi-G;wNKF+z#2?8mW`WJh^OH;n5x0
zQ4|o{r4qi*b4;DVQ(%%<<^j5-Lw)PUDjhWx5~Ls`7gC|)*QKz2;dZ2ZiG&Y<Cw|Gj
zN*W@c^-u6|X7@KkI@*b^O#~{qd`UE$n6bh!U7R);=>ylJ0y7Y6&8Dj7i%lps-@S4j
zV$oFW^`EuQvs8+6BPLTe5j!Y;mhqA-nrOCGtJuv}Jzd&80T(ldXrR_coPV(^Cks<^
z2swtZzk9(!N^Ghq@y(FO`PDxu&h$W{{nteL0gBj<@6_&NHo(3mvV;(2kHsf8pzFGO
zw;scC@F!-WO0xz6YV=Hr9CX%Nj*Y(#N=!Fq9kQ!l&Hc$x6Iuy)^lNtA#T`jw>qYAZ
zPd0p(0pQ)Xk|c4m@gielLY^q%ry|HEw$>4#Y@v{hePQ%}|F5)RM@x2Ep*i{c;&0k|
zm9@WTvs=j9^X2FJ=46NIuu9<mH=)3cw#u=!Ic&4KqtvOmcihDkIF(Pxv`4`bc5~r&
zfRsSkXp0lWD0Ih~#dC4<sN^_wM)B%!)<C!yS<~Y7pUlJ$9_)H|w4`^4nQ*P#0S_%5
zaP%&(Q&_k}lIb6K{MlzWj?PRHtkU-t-#jCxBGg+`-CJrUPGfyd;_H+EocZ@|@@>N+
zyT0K<jSB>&4p-j*)mxGv3{=FIi#m)qBBvc+$Mi06z7ggjY_9dL-=g66iA%+Vt#`li
z+z1bg()l&Vh;Ug^qWZJr5<KN|wk1}!;S#rp>{UjOUAJQ%LHb+oz$b&40Zn{O?FrNS
z8FXxdkjlu`^X-GV`KNE$K2ihBS-fV(b?>jna%G4*OeV=IfIVuc&vX4&D*@eQP`CEm
z4E-{bp0=P)Os{$UN5pn&#X57uDP!^Ook=VG)U~m2V3iZz_<;k8_Qv*ZbV@#{dsi8=
zYNpj>BiS$2LK|$ZEE=O3stmZmu1E;IMU$2JSSyUqzA68q2JRsxHVU2^ZblQucD)06
znVookcZiy#cFg&76%RUMUob&N^jLpT3p3O*xfd5<e;|~Th6-IBPc>G;`=*47Zunl;
zKl)i~5+5xMH1|fIT_ptaL6&RH?lpvxUuGlTJx<S{*w+M8;!N6l`c>+-_|ndI$SI}m
z+Y3)t6a*4A-9OwOw?LMGF6HV;#}>^3h=~mAyh0Uz7STSMAiew3@fFH|hr5SLs=vb&
zT=p;9FJ^QCjrV_-Yv_cv$7hj%d!EB)2D}aQJ89zk`p=;*lS<H-{z#xrJL*55X3H(V
zWFP0>cyU+{FhloZ?pW4ZK5qW;Wc=CXHje}y4f6V1R8iIqGP?i6oe<a|{EH2~N)zsS
zzY3sl!vQQcQt`2&giL)bLzf3^y=7|Dm*0B>SRALuGei5LOyCY9@dWRlQAH=`zJ4u^
zT5rcdbW*9Ah;`gGeh{dNuOA=~oNke3I7NMWUaC^;Xl;rxlDWII>XCVQ_TcAKJh?=t
z8ywbA$ayqxHA@DTMECvrd_hQMU&}2lBmI+}-(eA9X#{k4im)`WL6-P&sqIP>2oS1G
z_%N9b8d;F!9w)VyvrfkNB)MG&Meg^jfvwlOJR?VpqXIYR(_2H1{1bunj;?<OL=*6U
zkL0*IU@!7lF%rhnhaLU=Ws}1<TB)mrk<8%pxL^&|3~Kj(dg3ECHWz!mKoQAZPWvfU
z-bow^0%GE0%?I>(Gp;rHp(C@c4C>Qt_i?wCzq|#|O;52`YX*EPvKnb1*r?5~SKq2X
z8HURUPol+J)yh<Xtoo{Vs`KW3z!Oagx(h}+eY^C4$gnB`{0b?6V)bW&U88Y0hJmHy
zA1-O!+VW_2q{8-C)>?U0r!##96+{ZI7+G8d26oJf<orPwphW>|%{bqCNJ2=b-)(Iw
zZF%M|+8!6k1qclQ4OCEYOdN6w0*%1ILKAkGMwLBLm(zyi6(~A9Tz5nof9)ZDN2Q_5
zh!D2m1@saU5@2E6mm{Oqy?h_J7?Di7FxxnJDgIwZ1;~e7FBc}C%h#1o_uhDTCpX7R
zw&z>CC2BGl0S|P9>{t}`fio2a_v<^xMwf1|WiPT5InTvFqvvW}H7}40OX7v^Z?p|f
z=F<^yx~y(tx3dyo4a)fmbGpXT3oLrj+z(C%+I29lE_LvL2OW#0ef9d92ILs@#NeY*
zz&qR|h8X_jcy07#KpO5{c2-e*|3CZ^WPE)qXtjs5O#9Ed{dN6gSDztRKLc?jMx!L#
zjYKS!PBy%|Zu_#z+g=AQvz5G@w`JQaWY&Kr?5H4|kmHWjA#<9Ooa!<4@p>yOR>MH@
z>79q@{CP4yUkg^xY7d0IPm@b+J8;>yZfEsuDtmQWb*NXn#oXmn#mZ*NJmOG22o4TO
z8W&5-Zqbp>lI_yc*cV@bJ=CnYCn~ma`bqip#GTR0z#*;GI^MS)YroXGP9n@k{0jmR
z7A+E4kH88$(zpmCWhiWm7|V&L%<?K`(XJn4`|9U3>Jr_Hs!zNa%c$Hppd`-tCwu)~
zOnA8c)B;>Ul*<jje>K|afbPK%yWdB6NJ~J2Srb!0aE$hZ6SSQIeM;6k00&m_BXIKH
zy<{J#zQ&s%$Rx5~YVD7?<i$ckI0@c>gV!e?;6}g-x?lsEkO&<}r|^HDi?0}PEPsik
za0}$q+Rhmot1#Ta;2$D!J|kC$lL^`eAz|E|%U(Rpd33wpyb*tIgH*!;BQnbMOvatq
z<zQY)pFC8z%L9E~ve^gTgdKu@&urnk>L!NO0)W=M_h-;-wM_;YmfyzK)6c&=#OuuR
zo2awvs+|1gu}1>$PsQ@Zw~lwscUO*^X9xBIAGb0DmuT@<)XV<fG{8WH1oem<Li)gm
z^minjtiu<un9uk5wUL2cXtoO}SQ`Vklipk^D2`vRvjx6@2be}B*`BFV-!2pl&)4FU
zP*=E~RC|}?zvG-|o=nQTz}e9KtXI0doI`_toE>5|yl1$`{mA23AbYTxI{87@AA3Wq
zoQ@0UBvv^7rL%x+QB6%Yfow}12mEZ8X>jCrq7?bsO!nu`kbwIOA}*%1L~sGD@wzi>
z)%I0t(e8<L7nzi(6k{bZXpFIGH#4uh040SQdS)0IS{>qI1nACQzqB4oqI=~mPyWsn
zRkz(Uv5ic^TQZk4gXme<bi5I0EOl2huOn8k)U_gtU;qGwv-6%C_%B4?<O+D67#8Vv
zvXFB-2I@qX_>VB+PTi<Lhwa|{_b!5J0-3M1N<)Y}uS3;Iot($llf5U-_EQH{1VaW?
z?XYYd$35M*Vne^eJ~evth^v-7Lo4{fpv93)APgB7DE>0j=-YCG*gm&PE<goUTX3Rk
zuhhdNkC7hBNyT93`3Sv@=m_L0v%=Ut6^`1Y48#IfQ%IR!+uwm2OP%O(5<~#4sh3j=
zb_R=BUZ1&o9=>-s#1I8s_-*dADu&CU3?A>8NzLgg+;FN?qEBHUn?wY4&@GH<qJk9N
z8e#UKLPrkRx;kUblC4kIdZ{__oNHCvkL~qBkR{mUBJaFk)7xAgaF{<(cg9CGx)7|i
zdm*;D_lu||c^o>c2^QkPE1*7mSrP##Q-oLnx^Xs^<*rzg<h>Vi`{J(*5!cuM&i?uK
z#w@1|>AQ)lk|J8kerhU0G*}T`Bs|mXRyzH#T0**_dq%diea1>Dfr=`^X13C-k43LF
zNr2X)lkL3Ej|ih2R8vc)rxPiz`U2hih*JG`T*!4=>dm(gUz#gKxTe7S?L0;MiyZ=V
z)ZtsBS$b5UkR1s~y+&zmXDjvGw~Lta00h`5(Ig0vo}Z&Yp`mZz-jtZXufN>x6(%7m
zDd{|w^!#KQf27<p5mRrwF`Vi@UHhSeK_DpuGMLOTb^qtX%f8!5DQB9JfQK6BtO3I;
z*2bjRC+<_Ri7gh?c^W@Vl5R!`|3hwqO4OhfCO0VcSKM$Woe|lSZ)2&7?B+Y&zwuKt
zxH60mG{@ubzpq1W<lC`#rN~KecA`^*Q7@$^Ii5tV0i#V;UwHdrc_LhQI=e$bP1b7H
zrv%cXLXVkPES;8y$dc>YdORnW#}0<asr@Tv+%2mfE4x5veDP0bQRgEi<OuHtKR43D
zHBI2&(&N|5WuMaa!rv1APZq$*ltSk?>Yl&CHP3h!d#{SZ+a6#&KF{eYbg}L9CE@(^
zYKW&hE^Vq<j@42znv_Ri_*cCZBIqa>Fpo{f7~*G0cE*E0Mq90m@;Wz$Un;vNn2_@s
zz#8{oxh`i?wV!$YoRs0%#krP0OF0S-9FhS+ro$<iPwp8TE|>3f<8d(Rv%V<2YEZDu
z?Q!(-Tk)-X;&znd<mQu-omft%+EC<2tsP~!0NLSOO@ZC-3I%*jNeuF6hM(I&;~AR8
z=g|Xjz4;Q36$el&sQbb~)X6f2&=)1d!2cQvoE!u{I<R$TJ-4Z5vggBK+c^R5){2N3
z$7hA9`p(g*64T`x2^sBHQ`GkDT!}#99uA~WE?g2Yd!@+e^~cEefX7Ztyr%W>OdbRv
zuJT}DF{6U~$58)7fXd;(95CH#9479R|2%i5!tIrdNuRZ30w?|1*&p9+&c)l6R}<p9
zl(&D=b4N#W<mB=ymDv2LpOcmcnUD5Sx%%36x|TjB7&)}2(<C5ojYJu<EB$ODe?K(1
zD7|#t+v0L&u_h5SH16ePGyTPSu><d&&Y8q}Cr(L41)I{h@rZk~L^(}qpgtOk(fD#f
zdVlXt;PI_|=5q=PiD=8Y*w4(&-{JR2u!8p{x}hV9e+ksT1KUr)SyW<_%jODvZ-kU^
zoBM;yv!teM<*2{_R7L}P!{h9b*ZC4!@U6=aml-i%W`*>cQoJw9(mPTWLS<6_!rYEH
zjiyU1x26x58fqL?@TEUGH&MgR)T%0hRl>x$-R6r=gCaB4Do;?mR|EKaWMsGf=d0t<
z2!$8$^b?72LKlBEfhY|nwOrRK5ERm5P7Zle9orkl5V3^!^@-bmElvcSZ5Zf^5d-G1
z^kL!H3%pqxh3o?>C3Pe7JkhYsMt~8fzd!m_C{t}VOj!KS{qT>-<-vT`x*3jl2f6j{
zvi{LC;C6G~5Qq~`Tv1UFKmuaWuwWz%?Q8y<oq`$vv%ZAvlpM8lFlJ_-6A=v@K7d66
zsEsieIig*2;98BgN7ARPLPj9VBR$?alKjh)bn{;nyZrs^FM(5{7l>G!t+FMf$9B&)
zhwCO+H^cg)#2-LWe3c`q2W}ERO`8hE=6`z6bKM~gHa2TGQQ;R*cns1^2ErlQ-(3oV
zYvVsm7#@FM2EtZghi_ZMhzS=-dGJSOt3(?WiiE{W-!-o9vWKEQT?i0n-JOi<Fri8Q
zizF;$?39>F#ZbVIbswg-zeCZ#w9MxDcwHwI#ISHOz}YtRPfq8FFKR^tPL@P}An7v+
zC*_i!%RTCNkvT)gV{CM+)$gWJUKFn}2r6d4)@TSPMtu(wK1gVldi1}C&u-S59d6im
z^d48O^=`rhftSH><i)i>QUyN<mdX=_1JWQ8Omjm#l-{~=YJdTpxlB;l-R-mB@t=xQ
z-Q736>jJ+yC3?JkV7wub_MQRL|50z=XE$HseYLe(=4`MrORGr6bNB>Q4lCTB*#hD|
z^>H-mlUly3!nM2?p~RbN;ljEvAG4>_y_IqAA8KoI6<}y38?&k3Ix2LUwu<z@jMqC@
z*)H<JZ80AunX9u<_(8n`jgfEP8PxeEofzxsLFCFy1{j7>x?2DOuYw$G+pm#1bB6BQ
z8RCgLB`P9VI8=gj4`7++`FrUL%!H+mx0i>pm{Y%gt^WP_#rE5xgYu-|59E62+d$*f
z1i_og40m99*qJKE=8B_Qq6zHz4W<r{UEnq!`ZonIe+CjLJU&WAm?{9#@9#X}Ajslq
zkzB@)IM|za>0Srp>Fg10A3&>N^J(KCzt`Cu;i2FucUzFdLHjc>@6McM{`_bDgs|t+
zG6)1N&hUT7h1fG*2t{~bxSou#pAB?Q6&Lr`n2e3nF4<|H#g-bH=63;)Y)B_ofMH#m
znUdheb76RgZ{ltBW-;k4sa^q#S8c^61Cq7G_kUS_jb+w=$i#G2!_SgG5!=HsY|=KW
z+L=F0mVF#zT>TA3T4I8ls@iP{n=?ea0$IBeKLED6U*fiZ_qg=h;ca1ABJ}d`Y$IP1
z^SVR))eVPr61?}nYL8{Rd1VZn8Sy$8|7B04^Cri0oHbv)<=TEkmM&XE_Sk5h#oVBj
z4wc^#qi%YODj$DZBYMN+&Fic_Vm-jYe~?3Z!Z?4#n_<ACFEkaiht!G-2j$?xJKT2O
zy$kora#DLpW-x!jDLdb}J(w3{;2&jiTpSWm;r=_|<M^Mai!+!eCkea$U)aQAi{#ZB
z3JeS;Wo$bc#CLeFP?kqHEVbw$Wsi~$lle$bgz~})=&0|p@EFDxyb9ED2OJX|8Gi>F
zKU-xGCZ=A8_xp?<!!VUZO^q+qa+&K(QBQglQZ<6{(VHh})b7cV<J4l5tg#EX>{pJc
zjEmjkoN(|uxq#(F01|?X84NHqNA@bcHbU|E@<mMA&oY4?%*K+gYnwdM+e671;3(~7
zk|89<!GUiU@-o^QNw2oc=!ijL$zeRo@>mWUPGxba14{X9QimAv?E~rVjMOJ%SjuR@
zcpNEL@Tl5i0VW;$f9aTlSZPicU5s%E-2M8|TQnu<_$dOur^>5E$$tw$@qv9Ft~SAV
zGoFdgwsa#ra!!_P{Vurx0^wHUMK2JC4-1STLfLvj*dRhkK~NYJhEXub=v=D{71yTb
z%qm}mXA)SV%Cc_JWV1+2YgczG`pTS%aNZxlPD1qio5YWIOrcotuDK~;IsVGWUJN!|
zh1aRjMmuLEo%@^hV}!@XXRqG&L-g8;05o+uySLooHe&Y<J)Zo1$Yl3Oo$}l5%cCV5
z^JCv%fnIoMy#SS)yVt^B^*}|UdM=OHc}*Rc#d5i=$a)(AOYDs(7A;qgIyvq-+fQIQ
ztY>X_*v=^yohpiQ+)r^{-}Jx#XDey(Q=02T0ScC^(O1mPKlx^<ZT&GO>EV(O_bgVp
zZ*hVGaiHDi=i39d^D=iZfrTmqbiDnkdwZxmy;@&3&uXAP8>;gvL0uj`8~nBb`+B>`
zhi;_<TJcR#76MAXH0BdHA`GnBM>3n&dLw#)#-#fLym^D*^9i}x48<Z@;u7F#nkoW$
zqm8_`ywtfSRfNa&?WjPo5UStd?bOunen|3!S1odABn}cT$w**jfk4vMaw^&-B>2T`
z+F!pn+66P8?539pTL$vz5-l>hK>ZEi?K+cCX=uzqyJbLgSzW;60~l!ANtk3!lfPl0
z-cH`?NIZAyFo{m3S*u*F{Kuie)`(~3=?lZ`-KpX_t73j1t%&%=IS|zLUzkN?AUjY+
zx<aXuaKlG4Mx#5W3DrWWQUv|}GIMo3Seu7r3iy;J7<c)+jtI_<x?4x?pexH>7fTZ^
zmOtAtUlVofdf|7##suokY@GTTs$P7Hx7|626vb15(#7;rzHqm1*bncdL-RZg3>`%0
zJsHwxtI#qD?i1e(6{8Fe`??z(C<`6tm&tXvJ6_glEtxSpi^4N6QDY$Je0*4gVJ_>8
zW|l?}v*&+5O4VuNHOM7HYLCFqn9og8HT;5uj-Wq0TNxLe$<w>>mHB}7ttk(rSU%lw
zX98dC(&zAx>*ewpxol<{dTHs9KNp#hy>XFM6!6i|`#hHF!cMi>{M%!2jDz<;w_(q(
z!euKhM2T9hvVuGbj9FM){vjRcS@|m5>8s5-@h)Yz>Q03n{V$)Zv^j!%Qb*BMaf7OL
z^K<B}nG&l&dmMpTo+s6+_$%(N&djfr2|&`q4vyIP3@b2d)QP4x1`=GRBogsx<V~kI
zVmGM^P!--pZ`wnn{_+F>&3~1=u?aqC>6r{s3BFYVwv-Ut`C2EgG&Ym)-yg03*|O-W
z|0Z5`f4+|9yKc+p)MsEdjsQ!r)BiY*2Ax5>z+xOg_y_ds{{g|{w_8Nd?O8y3wp2E^
zw;#=f8ha;JgG>YV39yF&P`tgZU+et-Y9ryK2ZN=$G%M4AvtS;t9DJ(HYRdp1?H%<v
z`}^Ul>+x2IwZ#j?>B*|F`kn9&$Tk<xe4~6X=Pf|k`ks719Lf8$Ld~oFs;<nT`rqu{
z|15Kc9su|jW_G*RdZ5y4Y4RT&D=JA)GdHG>J+=SeXhB4oROGD>4-piSq5v?!4?!kg
zB6QRG{AKm1TH{r|De)lZmdaOi?RZX8m4*7rAY0zSrM6MWy`(zFk8)&3vn-$GrWgbG
zM#wuvQ<;&4i}m!m1F|U`hDn*R+QTbI?erjWW4h0^Hcvs81S;7Rax&-h?$kp8eaf*y
zmpJV>dm=BTLC1S7n0PD;&tSU@WRxBULu0hN9tFv)r=`PQOOZELxLRSdn?#!*+fkIg
zX|!_@%w|yc^~B0WZPuz&ZT3=W&%q%01^p4Hq*%Yxd)U`P2MAg$_%CQ=-@gRO)$hA`
z{!^-OHH@4^dTgRxj_9~Hr%$4DNf<@WL~nAqb$9T(sYe5oF?=Zbn4SIo_I?P<VgCnt
zBF4y&P{i@7DC7{Na7-0->HBFF91sGl)ae@4_!c4p@PZjUYH3Nn+qtq1A353fs)Oq-
zlP{L%tTfQX6Vrd1TVolCQG}6!2tl8&)Pm#FWKtMGpzmFx9QYuC4I*H1Px&WatJx7d
z`mW+P*j9#dL?kw7y@_iaN%vw;NCk#-iAvI96A?^O&QM^>;+6EN*Z(GsllI;w#W4d|
z(&zV}QS#Z9J%H(-i>x5DC8_aR{GNEgOdw%{pY$4F&c~=^%Ci+Z1CsdTij}pDn!jXN
zI{G9{g?y_+P$!K3DAjA{i6UJZmV_&WDn8c+i4OL}v?&q;i@;F-xy55JAadDMAz86K
zZWbp|?1}ZC$DSNoY<>B9p}}VRMRHB2?fg68pkv_WXJ3Cw#H`oaC+X7ZK*s0<_Pbb9
z2fZP7^uc<^e>+NZT>)KFD`2C4w|VBv7ABCjl^o-gjb27((*+_as%`OCLHyhQ_BIB6
z{{X6o4syh}l{K3u*k^SH4wv)-{^DT9lDzdW7fqc^e67bPy2d{1b6l6+$)P=XdCip@
zm7n_K=qN&Qm<R+eTI3|h7!WZjtG`&omfV?+TxZmpYnfu?e=zIeWMqGdXpoX*%6jnd
z8bN9Li$aYG{~LW`v-)&Ga$L_-iEFQNci2IMKwhEn^SMsFhxmN|d#FAJ0QI5g<Hc-~
zd6Z(Vkv$Rq_f$Z&Wm%%UY2>4r89D1Ppji4V#72<!nA7Sry6tCWBcQNdg{>6K-OfjB
zqO+?&a8TJG6akq(#E+L(%JVGdZZ1x33Flwz{#*TM*u6s48(t{7)&omhQ^aM|HZ|<=
z**?$En2jSfH+M}!RuD)Y428n9e4GpkS&AK;HU^}`I=s(TJn~73WIu#_p;_8bD2W|8
z1$Sv0pBE$Mw~=RIDPqykPn46kgpSwj=mxz1HdB%J-TG4BS?W-&GQ;DZw8})Ewa7r*
z+Auw2G*~I*;f3t}aFnw%R%zl}bzJ_jdGalkXw^}xR%rjGPZ4w4Z*kd9V}Q~LJMHC5
z#C%^D&?kPu1-$bw_&uv>=MS&0t~|}BmV=hK#}e&(UG|dgUrCFn${#!ep0az?Dehib
zB0`3g2p@O-9eOfoMYh|l0~!>~I&(}9pV;Zz2oIoN2m|jX#<^nfHliZFfXkL_`N<cW
z1K*7}t~m}Nrg^co{A?n59WWKHbHU9SDC(}Yv)s|fIA6hmkItgsKIeVd;<OPy#R-(H
zpug9T09?5!v}JaS&1^o2il{)~fCX$8W>c5<2(<MEVDEJ4`r!I!nJ_4LK_mouUR}?B
zy(RwQ9?0Am+W}yPgdbB-#~B^$qzK=c)(ekKBcEv~G(GE5mEqYz4A6L$^c_a|p#Rb<
z3ah^bGzjnIEybLyt}g`OQSf2(Na$COu=7B8oRf>$UB|mCF&e%Bbvckw%+m4A1R<k;
zz3Q(+vB-_Vz^j4p{kd@pj*P)Gy|G}{^IpcvFy0B{+zci@_?(s@jxDylKhSE-78sT{
z3EAy`8&8ol7OKyCdG{N%<ZMD?ROlkSS?_A%^h(IbSjc{VcPsIrlWFSm^C5E~W3SzA
z^?e2+1tFJ9y$<mGWl>;h{EYG1#_|<Xd!0pb??fJQv&%MR9`0KX+1TPH{TzxPf||UJ
z(4OU-xq(^@#+e|3Huv;gwQ`QFEvuYtFU<ZnQciIM-dN>VkTpre<7L+;v1_rtT=q+Z
z{%EUoN^Sg%c=FtUfJZ)UY8r$#7>D*_^=aWpt8_<$fzS?;%WMq{UiXaf6^GeMEjY41
zY;q2*u$ug1{v0^dKUe)kuhZ@O_y#6G0J`9DjI3M@v;VCf<KHp4S!OBbdU9<<^Rd`4
zlYyhUb4`Dib-SvYe;>@XG0BpcY#wjN7S3wV+)Ickmm;oEKlz~~>Pn8pDfc0YN<w4w
zgliaDNr3G^b;mHuSoKaUud+A~xL!{{^o>_*I;&8l2KB0(v~(a)VPG`DIM${Vgv_{p
zSsmov&GyJ)zvPCy;_|t)H6mB)0j`?)ou)^;w<_(}h}N@}m|(NyeHzp9g{y2UZ5|Hg
zk-z8?sTQdMxASGe?OeMvT`o?m+st|gb`SOLI{>dXORVQY7H8Y<N0(lyyvYt(Zhlu_
zc?C+`>-5Fg2~uk!xeVv*?_6kdU0k0c2_X6&jm7-`+q3z^>GveVp`+;3Pt?M<-VrI&
zQs1pbxLz_2K0*c~Q87tR9)<g8*W(U%7iQtwJf!MUH^%b&>PJamnf>$~<Z;2-A^*uh
zAHLoj2~|<peI<oejRzE}%QYs1oD7NS9WT_H0&}fY$TFf>f-#GT7HZaHnkx0C<}m)g
z;8VJ9ak)F6{Q-!SZI}KOLi!v}`-Na?4f`kCAc)1Kl+IP8WO6~B4wg=fstkQXLmave
zACI%2_oZ%rJ+O|xzPY$(@t$v)>RX19q^s=o?%Q!p{lp~3g$J(aS}pnO*?GH#Oq4X$
z?<HpOt<;`JEgUpX4qI6|wco-!5Gc2oMr18{d{Ms@Y0nAB)jB@U;}fu3;QU3u*sysx
zdmI!AyO_OaS#S6ql><`MDgx3sARV!@-N&6V7p;FlesEZuN5}G_Rj>d32^X1P1Yb~I
z<R^f#mIs*)ug}4)i#NN;u%3-X=&!xXxjg)m)?xxN)WdJ5f%0<@iyb#%!<armNsoNE
zPwrJ<c$`?+_&oW~vHPn&hJ)D%{kJ=n%!o_EX^RSs5ripGSRGDIPCc#)coX;MfJd#3
zhD<VPlm?o`k?|4gdu<XFMt^hWMijg|5u;VgN2Edep8qkOHKPqj!u+}C&2BNJLN-6~
zemusGqb^*89s~Zw=_LSdO}Vn)<J053ik6cDDqVvf0Jp)DbqZj5GKr=x*>c0;5{y=%
zLD5+5CpA3vJpXmf!vDqrTj!TsYD8!HbMsNxiNYA&$hv^8Fa~GY$8G#K{%3-wIsAC8
z649mc?C%GKK&U5&Eho)28@o-4CjNeox|haY8e5|655eoYv;Enitk*l!w<k!o3@G}U
zJ={qlVRSIe1x_<p%CcRRT&0sNQOWN5=ZfG&9qRJwyyslvJ=Yc{l}Ij)hxW&qbNgPc
zU5wQPnGIRKa?pR&+ZfAatJ|%cn){w)<-WcYPv&MT3AMbsGU-rkkFi^7y~tagow<}r
zWmMnRld{m);q_N9`xbP#P%!^2^z1lvWMSeR4(eM7sEr&lUvd-_T0BI^hOPMGZ6w_{
z%;u+sQ-!y%^1cnT6AwezE|Q80*|9w4^}B11N@o+u+=OZ<_Q%*5t&4B_ZP>Kb3*ev<
z2I0O1K>;F+8SsljOx(_WBeH`Kz7wdB0GfSiX(w?jO4BKPNfz)k24?n`olL!*?7~-y
zjbN^&ZM{${`P1xVGFNYf=lggM6HM*jcLUxG&&e5Xpn8-7!<6-4t_I;Nr`?{)DY#Vy
zu_**cB+a@GRkZ4BY#=qZS3^(A+mWlS3<n5rlk!QXu&L>{X(Ql!JIW^7GseJ2bXaQr
zP45q|u;Q4-GgDW~<rWv3jH040{(_ikfR$)2s8;C<5b<391Oq)BTr<7kezj~x6r_rN
z`@b(^@KlCxCmJiTjNcq<dK|uEPmG4K%n9`KTYV;X-u&$F-a2V=ugrSbtknnQsN<aX
zO>8=9;wJrE{w{mW^PTAr%!rvnLRq88G=zbrNGhh_hxO%5oyg%Cq1~coIJO|Gx#?##
z&q<nB)Iaz{q$H6<^reM-;TS#a&FZPjyvk$Ev{CgSS-iVxCrXPCD#eV|?slu4K7CAi
zNrt$NZqGRE4{|gHJXfYVd`L2W@g|Sv-%Lm}x!Ew`ok`X-)6%MsxGnLcECVhnP$WOO
z+`s@MgG);6;V#eb5Npdx%&j~F+zyh!)iGKuIhKYqJo4#kdzGO|yu0122HM`&BbY>A
zXtt7k`iY$;CyWnHGC>|`s&_b(FGKn~x0+WlyV5{eI_ZToqIP4fX-H)HMQ2V=CJ%Dj
z`{Abj{G6|Z@A`|rl{~^awT-N6f0Yn59ECo&O@(=oiyYbh{QR~ee?cq)O9U({bC!zZ
zMGiK<zM<XLG!k>#FQPsjcRCJ=F9z|5Q|VNY_hze(o4l1yn!v`iZvb56{LYL&F5UO0
z(c)k77`xpv8g#a={P}>ifCnl(i6cXoV8<&F0Qs0i_3|;bueulV*@oXzopfCLlFisA
zL-fr<0tp8L5<Q6WQKKwBP+BD#zaJfy-hER8A5lc4hUjag@;?>fzF=vM6fAmambQ%S
z0P0&ivtzlodAVAPY#N2nd@jZ9O)ESul2PSAzvDd2B(jb7QM5em_(Svxz`;Q}BuzAt
z3K3F=_wKL4{=sUThi-JBDIrq9)yMhQYK;a$ZWxP{bQtu*{X)Uchw`?sLb(u~-2cj`
zD@{oI`}E!jH|{I=|NlRcT#(w31{VuA6qVgjO;r(I>yE2@tp@JY3FgcF?Z8a`*$x{T
zdDnKVEt{%{>oOa)Srb!ZT+1oR1nb{PVGtSxQCYvTymyq*FZO>=VQe5?+#3)+POI1(
zT_*l6y1z~4V+g@Xlc?eOYI>-7FDEb9dPZQ_y1tkBc!$#kQtVxKmUQm_u)A4TDhl}P
z41JXnT`jMg9$z|KC_fg<OX)C=L+SA5aojC*Z@;ZXv3I(C{T@@m`^CwEgSzi^A;IWZ
zBn@-zuS8Puq3R(I_n7%r(^P$)L$mc0&ZgG5kxY<cg<H_`7``Kf)uoSs@^!MI{KNYY
zzGIG!^KqFdVllNZ{cfhc;ojf%vlx$pdQ&HC7q+^KbF4io4(F#r-4Cam<aEU>Hf1=1
zK#GUxK0(L5PU{H>44Lfszxevb=sMSK?Z&okJ88_uY1G(kgEqFYg2rZJw6Tp9yRmIM
zY3%RW2k-gD+55a>tRHJUWAsP5?q}ZDob!U4TwRnLoZ&9Zk#mWWb8rVa6he@K*gh7;
zPQ9r;%p&sP?<ZiRvjwnt_!#%v;>kjKKmZ#kj6vmS$NOlm9O*DxCq6(P3`#-G0u-qI
zfM6h@M<(0_z{r^~Q9Ej^u{1Jr0Rs)R&=7})%;PXHJpjABcZQ7%ku`eh&FAU%FKWR3
z)wt-(z$W0*5fK@Q4D=Klk*t-6Z-bv}04U!*B+*wQlTY}UIMtLem1d#cr}b^c9v~oa
zgmV!96f-=uqFMV<?)gx-9#?k!054F1N_{E{LhQT${pl43Nkyz2%qcFSDm`Kl3w#N5
zhGz*UU(a51&zDFgENW$2^XtG+OY@0zViP*7)cmc+*0YprrJ^2Z){74c!^%wMlxps&
zwb2vStcc9Ail5^<YaU<-0{=dZ+a0nE)F>_U&HbVJpy~_&yvuS!{>fTlp(od;bj#CK
zJO(sH)o-jR0|X$KV<F2$)m0lH&yb0b7u6^O+d@`*RS5d0lV@M}4|HFSVA#V7kLM=E
zreqmcJIB39<y7|O%YGtT6WqjQ6c&<^^Y-P?T#2Yy)|A}u%}%TMe4ifH80~xoCW9=p
zI1mzW80r5wZl3p2+0PpsiW8Hy8`o*F2m_)540Gcyk~^!prpY!h0l;2QxB(0FErAxc
zOk3sSWH8?X&zh^AfXD;&^J!co@nK2FXPJ_NT9g;>AG~C*v><I-@^eL7@MXpXXE7Q(
zVk3&<r=#yE4BhDQ11d$xjkY&HY}sHryDgp4{ik+sa}$+UvzJGL-g^G_bpm>xXbcsq
zSa!&@A`H`hYB5NJ4;#Rwz8~PSj9Co>2qm(#vY@Be!??*GTK4ZBa!N?n6lLV-(IEWR
zQhZ!ycBDAF$x*3Rr$Ri%w1v3hmGKR?zU65j@{gy>u^M(J?>;@&H=5YasaBIj$|^v5
z-MXs)L7bD5isAXUXZ5h$yB^Hw@oN4&R_1uM3dKlDY24GBh1rSIkwg~A2bLsAry3=J
z`xCwyP4zZP&WF6~A<c)}{BI9buL$}cW(D><8fzhoxx{bFd@H!j21g%KSbsyFYnES|
z<|w;EadC2b$3FNySHle_dj#eCX#zgO?+j4y50>W-+*Y~M1F;Zr=lVfVB&mYDRh@L)
zIPUu-T()iX2WQoYrEDp2Vs!}2`UlviOOaI(7>y664O?G;_>-mT8n;TJl%JoMd2T=;
z6RomPt&^qM(7Sau8szA1QPCkryVWBMs4ruTYl0z5Wx$AotL?j+5^6nj0Uc<I+Wx*u
z>wtSK9AfPEU1i+NdCM_!H#}$aAgr>C>b$n@-Mo=-(-C7D7^guImGnAKI_hop1UKGz
zfjyrc16c^^6Gf?`3IeTIvp2%py|n;BA~hsn6eL*LoMPZg(AT=JvI+UWtMPx-XBK!_
zi`CbcS;3xlPx)08^$5?M(fkikZRn&Wf|EZs{Y}Q_R%XG586MWiLNXE=wF10wui<Kt
zfaYzorQQ)QAY)m%g`!`+YSFnHV-)hxya0Emx7D#tx1JKqttiO=+adbAV~9c8XcICu
zO+V=>TNE>}^X0BIdtdZ<BGBgvQ^Gk3U$5*yx7rx=$lvO;KOsrj>gvJa(}L8vvV}h8
z`lIQThYpwdLQLSn_0aSp!ZL=qdUKeq{9t+@RJOM@erl;6bhk{a=q%o8`^&xYz?CUk
zT>(0p_!5!VDoMrT!^|-Vj+r9D&8(r`im1xg08zkI-ER&d53w)n%d<NYE=#yRn2G4m
zjM{Igb#WWZHc!tLKhaepiQWF|OSv3JGhtjd*5IE%4`%#Mb<w1w!cV*jGSm8OE0hY+
zwlvTKAwf?2*p#o0d=r9jH}wR1stEKQK1{`u3$~c#pdhL$*Hhh-rTSyAD+=NJ1F)IC
z7$op!{r#hh0aS=KwWs6Cw%9QeDT424yjun7iAG54v5!<`kI!e`z#7xu_V;~W8SQX5
z1|;Zq)8k4g?=zc#lH8=2{bipg1y95XLZPe8?lv8v%bbAnewv+AP2I{0g22~~f!I&L
z{*VVA&fP4xS<jL$+gCUJunh}QKY+LY^cFA?cDY*uQqW+)hle5*IX4?`uj4emO{f$b
z!;&wLu93utm=&(1`X0YX6KDyl&7)xGlolj)aNooVIL#98`aZ>y@o!1!raj)6AAHJB
z4Ok9uaq%!j3TBz^rvp()aN3`wK62#F)ksHOi`x^nw_1HGK{fJLK~xs`dJ}Ow`#Zd^
z2@%IxyNMk^tze-SnK~}J@b>&}>uS8dwc)o7)?rSF_Rg3<oIfV0Krsp+Lk@{j#j2*s
zCXZ(v_5vZl>rRsKMtPO37BOfVC6d0j3hGLK+O=6Ej}9u5dABiB-entOx-Hk30srZ5
zRO~?3-Ica>x#s1cqBoD>=D*jm|9DbIIZ>5M%A!l(5pV-fGD>7u)VldqBjv>V3*61J
zUAjtx^=<s~`sv5Yp^go|w+vBlv;ntL#Hj7=UYd#g5Qty$OpSJLzGUqW1(s3=zv~fL
z_bOT=x7u=QWLZA3)>{qO$rRk35H#_;rcF!AKUMH|JWx5n%fj71ovD$DQTE+^{d(zg
zLOX`bn3Y_nz4@2hmDzkH+VfS6GW^nt_{LU0*LaitxUf>)*UvIHM?Ylch9sYEQ7I}M
zdRWbo&lacsAFf6h+E>G;3vKRSy_jf~c_ft7SnfWQ2V$c!1^?NlAf6iqKrV8r##2qw
z?_EbsdR9ogwq5kC#fxl;5mV;r=Z9a$)sn8@gBt;b49mWw*&*?n%I6U^{fxxlrxRVF
zEPARzYHnWwTmJ05Em@%sS_ra;1Gm(sWyM)Rr4;{2Ahc<eSf!ykKy$c<LV~Qf*cnvV
zBqe=q=KAy+hRp^8#xulU<fe>s$F@cyaYw@3^kaeV9nVa}EjJ?tekhvgPGB~D2?QBt
z%%1_Agna#G$9<Olf39Gd6Y-rnAC4#)8RDy#XIBO5%3?-Jzi9}YkI;jA9$ONPup{KN
zAHq$EsW21#M+>f$y(;EfbZWOMvDD9i*(R}J$d_{6&GN$js-mroV@)bO1JfcEq=N4$
zI2x18g`o(@MCR(mMHHo!2JUo9Gh+-@q3Hu#-Q*W*UPt>!!)(09>mp>FeaO=c3}S7b
z0xOOY5aS|Nn5eC|o7Z0fijv$y-4Ly0J1+P>kHF`NkOGcL4#TLfXQ@6IX-v>%>hrNC
zk>`zS$9lNyOR6XWHV(w$wScSj7qu>VP=?T6J>qCO(dE-Q!&V0Ra^5r&Yb+FjjdlYw
zK@yV{T>`(`SO~vUozt!N9ziK>MW?Qf(VhgPl$;nU;=weXz&>Un*1;qdiQkLHy|9!@
z1J~Se{62Re#p9h@_1K4E-^0U?96G}Z3A5V-@{qw86K7*T0hMm#U)OthT&Dv^1{ml`
z?VHE^F!()zu*@=o2-?3vdf*bFXMnbCRb1`5Uf|g1LF2BmLWDRvZ6tEITYV5bQ$w~;
z8K^Phz#Fd+-NbvxSYRwki3#K_y!onDMMdCx*dKiIb_qyk{Doy3g}XVUUB!L$#-s)N
zvxwg?&}p=hU!$?^JpL+O6f;KbV>oKQOlbreVABK;*%%&=mA59NUvT9x`nURymLJ3D
z6nsCNk+|VgoMQmhF7LApR2nxIphs+Dmt<ITT2JGCsN^axMi1cUN@lG`q%gdkW4XP1
zH^;65fA1}VOq>d;wdk=^8BT^3x{jrJM#i2~P@H`*@JV4|!f~<%OXr@G2`@!v3aa&8
z#oXFt3(hlS4B~VKH8p84K?yhm{HKsqnuG+g6%6;MQ|h-DnfHvo_f^UVGE))W=q~S<
zFh_-U{aInnfF)}&TDP*&URyDza6dgpj=xqIa!SHGgkEkCAl|66T_x;zO%tTicMSxH
z4}kzV#{6xU_kY9W{{+sGQU2J7Yd(!FOe<EEnWFc6-*A2>TwWiV14NVM>IuHDA@*Ls
z6LMm+Z`nlHzA)g?aDqxJICC66?~+I;;XU^g9gRg-4;HFW2^)IQya!1NDWVy~HY;ma
z2>9KLSXA$tBf2{oQy;`se~zLzGHiAbhx@LQuOA!0{=Jv>V$p*1%NP2_%Wx*=rk(hO
zQ_f__oprx)oC_eyXaHE;LZ|JDHTOkQfbOX;`iYb_LrGqRQ=Fl_?^b^nwV2Ca9yZu8
z#In|?S{h}1Ky*8AVJZ~?QdEMbekGk-^pqEZzbmB|WImk*Ay7oKA~kL~G-Z8hv(41w
z>s;or)i2<~4c{jaK%ML}NGjLN#CKOm<o;&qH}Vl*^zEc8mV%L2+CfI)2@&sxS!Jn_
zyoH}sxa9R&Jy+cMcJ-hkycv%dwBSR?WXs8%4J}K~3J)k>6cnx;`E6TId4%pa0uQEs
zMc~^Q5CGRU`ZTQ#OC5?bP|I?Mn`8e8{Q5)ZeK6jU1wwUQx24JEDm2{sBc$7YCL9!I
z>&6PVt@q8XLPR)}y~2tVx@OKmAbkl)CAU9sJEFW^tX;L|n-Q!w&7*)}bQl+=y);x=
zHQ_sF{Eu5A!)hA#ek|*fcVOXy6W?lXZ2A4Gk{IL^%4M9)jMUNbyjO5)=|QC^La1Ot
z`t@G=Q9wt#&+PbR-)a+?-?ca@Zt*9G`wb{AfV~r5qOsU#u(BMyi^l?L^AdKy-yCGd
zC{(J?Rm$8r_v*IE_3j6jtxT?`A8}k7$Z2R+GP?DYlikkSrW|--uJ)2JeflS@YO&SN
zmy!fK$-#sc<+F9ge>wFVHBt?UX3f8BA>k472t_2bj3dB(6*Tl+1OdSumnXHi?zXtc
z72-LnG{VWua$_D^x$*DIitg5&HXrtz-{`yYp(-(dl<k6`!8lC)=Uz}Z$y_{X(D%+*
ziY+7Qz>jkTxK@>g0#5Z24Cf+Uz#^2oKwdW-@T8<q_5w@^H(HJ>)FLgQ|M?BERLrq8
zXiNRB`=W7z#8wnxRac?r>WgRabKqpr{B8YXLH=SyQU?iYQ=^?qw1t8==E3}U(NUhO
z`iTIUA8|INSo0gNCAvPG-e}rot4F3)`zSyZ7G8A&UW<D)e*d4k+&3>zn<1I-?@_D+
zK7Tyevsv5<#_nW<X=@TboTh7l0XxR3WsAU`ESN+*v@kmCIbwze`8$)6@Ude7>#}V5
zLM7=<zOM+Pf$VcYj~BC_RcrG0sy-QFCDrD6G%jI+9cP!cH;UqZ%tG|~P*|5rLWxjs
zC8OMwZg2ti7(2<v@RP6RlJJFeTqonAg#!3O+vB2FM#GZWq+TP@g;{y<T%z^84{i^;
zu;4dehD}1!OrD_^i$hfbe!QRQKMo@ZuL8{aj^8>V*W7Igjtw%P^PEeV#%Qap0sI;-
znPU!bH)^&7YyhjNH%;{9{s}Lc1_46gBioPiDW1L7;tzr}5hX{B>$U)`Lb}QO$YrD=
zM({M|B>S0_0y;jyrRmAm1vGzLi5CMeJf#>LAHOSUQwuY8c(=YiZpPQDF$S~?kHynn
zj%tM`ea}kvkdBC3h1r^-jq;+AaAY$|{)rp1+Z7(Q=Ed3R%Z>Bf-AJ=iYkjK+5Kxib
z9`lsM?w?z;`bNWc=~zLB)0y{Yf;XRa%5QdukAbx8c-{D-{v?*t0Vp)>m*}QaExL$(
z)LQ$==x29U4MbCOO)B%zn$3w^^)kH_%AhgaL`EEswXS#kUUMNZNOc%*LlD`*o<&BL
zfyHWhzX-T!u`1h3d}DwNTG!3G5du?8n`Mi}LMyEVe0by0*V4mAEwFF%cC@qgI%zhP
z3U_cEU9CZ@cUngN)t4XkqPy?R>&4&<xdkHCD8(QuF$q}dATUFbAi2NIz?WoA6}th9
zThp{;%B(_~q9qQDe>&b`d03Y7N-+_vs%HwxsNz7Xiz1voAG481U(k+#Mu6^3BX?oo
z#{tPODdT&4s-Rmnj^{rSc%m=Z*_qiuN<4>;XEIj3DU?*NmXSxbwa#<>WSbyLR^Zc%
zBA3A`@%QkgnY5uMXez)05@Iy1*bQ#%$2O4B*t_1IX`KG?YA=I-yQlTMB;xfuss>p`
zuE($LL14Ul^THd++>D6Oivkk%5^~>(8wovPf~iPi!h3W0iNzZ~Z)DedTuFz}W%HFH
zO};CVfCrh>x?e+!;8TVECU07CaVJg6fy33P<3uijU55=;ZEeBE_2HCu4K;y8h;GnA
z*}?MN5I_a`B3bv<QInLbA2Kla;&z{!(L_|_X(a6H^h{stqU?9n(cZIFaLmwNoaH*2
zh$T*-9F%Bj@bp=&#jcpYGqI?lP+)OSu~Xf=ocvXur%P7M4By*R<)2PN{-$1YDp79V
zNTv+26VW6bLpxTQzfg}T$QT+7IE(iKtY#H}?5s-lPgsZ?WJ*CUST+#F6;^DOBuF9p
zI2g#F1_svf2ZA|K(j?MCejMO@sBN95&Zt@vVas)q$LT593KTVL91twZb<-hjoBMdd
zN;89jdpdW-_=~sza;4!Z2CCsv=G)67f0aW%4&W8`;y`%T;OgDTX`x|NXbZFt`Crgu
z=d%g8Mw<2r1%((}?>4%W0GG;O=ap#b3r`<YU8ihxn+N*~<GJ$$IuI6}HxMd87o_b!
zeTYT&fYKCvHbd{pF)ChZf>{tWrP!oEkYlZrX|$u!F7&B9ZOo5(u8d(jPwm{YioQh1
zVjOqgib6pJaSEc=Y5!ixf~h`iSdKgn{=OH!qxF1D%+vF}{AhWeI>R8J${VReQ^$(O
z+S31d2Zew@WUveF57>_L@Ol27$7@U<yIaK{Cj<JMB-6^*m(NZ_!BeO#(@Uo!fEXWa
zze>~JC32feLgvS8%t=iR`oU$hP(K{VZKZB&-eV0&t_-h_l0Z02AH%%H7lhV>7VD}s
zmz$?UtSY0jEmo@t2{u2<#4sYcAGQ-wtYCs$lveH;&UI+;k<*nRsnN-x0kV}cV7~|Z
zmwh`bnMg+Elxmfr>xnipKD$_?BZ4U^*xTv~?Vnz}cL<j9Vu%Uf<{P%pJCclsqKUuu
zT2<mG!&R_Rkqqc^BP<*|-3kZ)x~=FgpTul59rcg;5+IHM43T9*+Y$?llEhn~;H-<(
z8AJp{xVU^wvm1~w^8au!cFFQib6YX{n#kqmm|AZR`L1rPRcmNN(ByW9ljUNqBh$nH
zF~>?ADD^Ew$FUT!;D-VlO1?jQOW;K&5s~MI!uhv{Z8Wmm)2j!mW(>26ni^+I>xEA-
z(oxP^R2?_nlPYZHyAk78aUazJ!~eK@7pV=fjt%|ZuMNYBa{n{5)w~a1f}tP@Ns|HO
z&Z0(=l7K_2$yxaaO@@=PIXXe&WdYWdFCA``G8vyWq-BFX0rV7ehtu6Q9m6lcHW=u_
za!;Qe3u3eVlyK*LJ@NO<e#myjkX(iDS2@K7EL#Sd^o*EJ5XJO)ox32S(jMH<T#*W-
z`Y>^GOE|=?aL*s~jY!TJ*O(Au4$AmLu6gs_hXD6~7J@9qIbe^~3;$e`$5qzwCYZI{
z(*gkIzYmY8NkM>uWwQ$|SSnV$D*W!b)lE<3*T+leI<nSH#QUl72P?XAvM(~0T@&Ay
zE$({^jlf<In&mZlU#GW<t-3rf{jL#Y;YJ^CPV=VKb<lZy!76-m1C`_oh^UqAOBD+g
zDT*F5^!ex#{l0?2A#eR*W4!mJHis)H#R*1(wvZk2_&^WgBfjx%z{rn*tOIkX^*jwo
zc(vccJH)7SDMhoI_pI`5!|Qn2>x(84(I{Nc>$OGij|OV7{i+LWoptem-|J_wV8mek
zh430xDfKxSvgXOcm>R*K&W$ULmG@`7;qXuq6rd&=@=RKQ>&48;*=-d74l!!3N#r{Z
z3(WlB1gTcIblC~zs`zKMib6+78OCjO_Gc&G;7@~Ar!9Upp9vL{(Sy}xNxZf0PQZJG
zL{O-k<hIQIC|W7se7wQrVq*}tR;tikDy1~T!Tj|(epD>#wpdN%l2yUOzpK2Fu*#m|
zp_t=~{wDOjdA&WW{zu?d3IPnDDws(n->66((^SDkA<Hou|IjId;>9+yydr%!OI7tw
z^T+_DtHgPk`fs+KUB~u6Z^2q<Xx;tS^0jY|9Wl+n5if0p-|YBrJX&2{tyl0#;4-{n
zu9OwfV_AiILiW$7f;GPNBlC`Bs^P)oHW-#Mw;oUmn~$S10|7guC=9U?HyheF8cI0j
znJv4|dol4cVgi%xx4Om@^RX04-ZvVX(^&Fd6+-(!0F8W;!?H?+egXR6QlCc`O;87E
zN9w%VeMujr{+!vibhc$9QRDg<_Swb{%e55%QaPIReD7^@9Xh?fp8g#eyweX%C?%><
za2b;`LBa82#2B$q1+b_i@OogTbkB7k^KiHPrcr>)COQDCcRi~`M8C%JPF`TJ(tv19
z_%;BBl~^FLY5oqf=;O}%cN7$$_fYL7;Sm@m<7QMtmX63xUp#oP#;x_B6#n6d|I!Jr
zV>u!B$A04vX5x_uDPoCAR-2k3$8rDkHvHH7A-07X3t56eH3Jh1a8#@P{3hLmJYz_L
z&#xIoy|<-OIt9MGUlW`YUhltuJ*nUFvkd{HXF8N?is|9)?GKFoqEgbv8)8)P>gKpf
z-L}nW15Usn{JC>my}|CnxG)uix#7-dAe`+bUEHXcQ$dR-Vt=%|#Y!wkk2l|_sW>^Q
z2g3P~varH+BSCqkduS}mBFiCHk2l1=aui#fO0Yl52^i<L@g=zoeHvyh*aC0aCn<Hs
zh@N4pSuY3YyVGWQPvO)Tr$&@sBH?}sE1vx;ZKKoFw*H1d1~_`ckkf^N)#i%#kevMZ
zht2!elcmN8;}7e8ZGa2HMWeEzx7#X?`-Jw|HO=N-SD;}K6S;GQOf2s@HZLbKf?|#u
zT|yHtjvAg6#TUfSv&**fcu@3PZeri4wRa*(9Ft*zejWRQvDQT43pD*kR3zE_Kb-4-
z;Q&x9d?cZnc(Lm5Qog>(`6-gBqQIBZK7J`$(OYa<rG{L`#s+k0U;fQwwQ?t4i#I-B
z;_l61)NuFOFzQ>k#TtQ4#6ni_Y=K#nM#1rDtPt4(C2j;zwkO_4Pl1WHRtbeo!gQOP
zXBYIdQ@!VF77|+!tFESoLz9oZV^X0q?_XFLL3V$sas3t^B5?X#U(gqf1dLe7;cTc?
zftwR>yVft#n~y%OY<x1P-+o!4S#^V2l~{4lC)Lu=_4Sqic{<V%D<*^ydzQOBz@$7%
zQpDoO8TD|LU|hS4+Jsbxx-CIf^(73+QB3yd2iwxMw&FylL6tQl8BCGqs)Ou&s$#Zz
zl$g#QG(Cp_l1YBw2J6eMF+r|g2h#L$_e~5W+w!{iWF;EgHxf~EAgCj}cLEnBlb<sP
zAw)8ZCrya|p`ib(#9J!lm`agIN2%h8wO>WUwYrtMAzK1zF=5+-se;qlznV3OQdw<I
z-d<KlkQ}<z``(sMxi*%$SddsW!`<~78?5%EH7qIg@wuSSI@aL>KQs%#p*>z~hAi<V
zFr;9v;@9egD2sCETkG67x}IH|biT`NiRs$VZe}jjoEuwp^u3Z9qW#nkMs?cP&6zH;
zdtK5^|J36A%w^Cz7a`)4QDm^pn!#isUPhdq7X@T&g+g#2xsZv92n3|uuOlr$V$%CU
z_Jmk$oGe}(1H^e4Eu-?+3=H2heKkz&I$0u+Wvi>SdYNpqQde7v^WmWFC)d+IrrZ7Z
zStvirv0m9SI`!;m!XQ!@(S8M2;?AggWQbnJ3K!LdCDOBuMjXvj3BqE9$(Y#WLfM(9
zHz^=Nf~h6~D6js$fd8K-Aar<Ol_vhJ<PW0b<2V9V-663iD-BbBx-V+sc?VoOSK$ki
zwHxw?j#TA*uIUYP>xR)Z(RLXf4PC_s8h?^!t1BwA9`1VgXu#>#+Z6&3-6m5BfYe`X
z?b`33&L8ThZRN^!Jpo5F<J|4;U{aws?~{67y-q9a0PoM`ElVI1B?sk=y>t}!=gUDD
zLlpV9pg4m$#OhL%j%U-Z2)49|=2}v--$5a=Ef};kT^+u`h!;qKHm11l&ETT$TnQ1O
z$08(dD|19V1v;fHb-W&2jkW-Wf!#l00;Smzx^)vs^WHaCZtP|pz>?Yj78)?}3u9}C
zH*%*Bw#K5Bw9xf9)g#L0|Cn-n-j3u)v}<!i;+A^#x_kbH0i?Oo!A?lglIu5pdeN0#
z2>a0aZ5H8!4}*FI@#1Tq<o0d~iRmY;A<WB<aE6HPNw=Y8yN*>gx`1ITrvkS~q#A7w
zmRfPqxxw#a&hP&`E&8whDY*iqvd|`wZu~nlVv6K)fQ-d*j@eV3X#SRkPw*1L{r3E(
zQQW9r;5XYVk5NQ)eLG~k(|*;2E83ZkppcQiiQmtOZ)-xesjfBfI9kCPLBn|;jkl%Q
z(s(q^C-fUe0#Hfv8sDVrK^~mhe3M}eYBPO=Q*rKnh=8IhEb6ue?WAYu?DV-@*7sU&
z!Si}(n(~zzLd+EnmG}UA|M&vd0N<H4^6MjVf%g@D84R=gZekA+@8iYG9VvzLAJb(X
zG)9f7d_&<(KV3t9?n-Ppeftl|vNDJ+KbPw#i<mp@1$@L6kUF?3uxDRi8qua8tlnPB
z#v5!5PZqbQzvy0ppF*scpwLp;UCS_rhJN1+w<<Zj_?~j7*xzflSH|8vK3XCD^iKx?
z5U>G0Lts~`=evc&y}kPsO6kO@?MiAu1l!&v$tcUy)wu1YwJYlu<<6~uXPxtikrt(`
zWh%k-aZ2>vnp#IzvDqrP8RB@30Zm9lFU8?f#&}hrF%{*2Sby`Jvn0xQ8-?hsZ9XrJ
zR)zBE&h6eJB0Ra)fztllWH;;DG>(W~vYdjWN#im(TsK#&SP^g!pd{f>|0DkVuX5!N
z3q6)`8CKlF6^t*(Uz^Nwqur|MMaMF}9uoVy_nxm^pISqoAE)YuEMP|kZTH#jd^s>d
z*zWeNvF#bRMbrmHvw|61OKCxKDG?s?3o9R6n1h-Xlx|y7dWHrBF<-8~ed~C`<yK*k
zKAGsEfNnW@^6d-nwCm6b7OTEDaGTm!km(=MdY6mN@OC3`kicL{fAm7DIb6H!6Pq~<
z8>kd&;N#o1e|>la33w<@wfx46^dUzuI6v3bcCKa~9=tGla;@Y;^#O!|G57A(UT$aH
ztnnW;it~Ep*?vqEDqEu}#pUh6qHsE#q&hNHwrR3K!(I9*Y1&eJmL>N^Kqg58E!Jk5
z@NOh&S>p@0e1RY7{{FKooUQH4Kp<KPnlVyAYBWjI{h#3-<M`yk_e4BR4mY#=e?$tt
z_kKX77k0>duc}I)hlHR{OqV6H3W}rR%AbO4bFcF{Z=-fVOTDjH98F&iF0GbA25(pA
z*KuhS#e%j6kYbVoai*twuPKnJ(P2Sbei7KxzkoT!zpjwKf9Vwj&8TM{ywk*M`gl2M
zksZuhrfRw^?1QLoDx>}1UI6=szJzo^T!o(xyS=vznxlW&gm{+<v}1pL#fwjS53C+=
zIc{7x=6|bV3E5^KNqpxE!iwhEDn9YN5OjqHHVKcfOqslX&m2N-;|pJDo>2wJyp2DM
zrX7*|kcoi==tdNo;t_&;n!bmX0lT873~MZBQ5X_He!w2qlSQ(-mZp4*MB8f%!PS1c
zO1t+3jv_ayL^)D){&4D8SEV{MLqfcBY_y3)2;t?JATE%;-6Q(8@O*kaJ>c)S+g%yt
z{M@#UT4=x6@Fdl}xsRR6O{ZXA9PVaT3=E5W$D)-+Q<)HA#mFH6#tDao(*4#dk4>a0
zyYv>)Fr))>xO;PX9lSxyg?Cy1<!`1a(+mB3DDxFlu9y3Kp5>3_S&<PWkW|vz6O~5J
z(SUxfXR*WYJNl~3&=#*Nd-x9)u<7?m{E|<R8oRv+*?GX<QvU(!<_?Bqj#!hfA0FCw
zn(#A@xywJ0N&YovBv*ltB(=F#rmL9Ck|A}FVunxbUcKh8-0m%Z`>+>s^=`3Lhy0*)
z*8ebGnqMAh$v*UloZr74m#PQ0I}<Ee3oz@Let5d=YSYc)8lWmTat-}D9$_?*I?et%
z5h=FZ0Q=siWKE9S!_m~)Ihm~z%hZ#L8DivguJX?GWMRvu*6W<r-L>$cvGyj}FwMr!
z9D5zw_58M*>iXL8LbeNt3C0j~TLNUqv@_*izTU2nKdEyxn1pR@9-&B6J%yd8;+gv;
zc7F>D-pXOTcDyYCwG+3avr*y41tSm^<~jK@jAUHo_>UldNJ>P5mCXv!0RPxGn!id0
zwhtey1tr#`o)rO*3k6M#n$1UN7IgYvx8*&%ond_Hi}=F|3Dl@UC?ha+>8rP@CrFis
zu-h=mQ_37X<M4gaNVmahDp0^+f;J!h*u9$~)38ooFV!w`xGG3;JYTwQ$52j&HoPlH
z^w1tMkF{G~?^O60DFhf3TD+-MV0^|5dv}*5xs+y&^nbrUV)_`dB-E<xzi@Z1*MTTh
zLz%e!35X1zO$6d|XJJnBcEt5Xu5c+GLAT+st3@sc^KNZ1?_pzWcO9h+XTC(rwPiar
zC3xySrO~v-%l+)5OcH*W3mJV3_#MhNQz_G?i{DX{nZm#HzUHuh{;tR8``v8>_P1cF
z`#`wa>lf8TJ6lgh_EL7m0IXO_+=Ql?BE=BBiVV~kVe>z&b7za6`=);3j$~Th`r)$i
zY{!U}@p;XC$0J(`n{njPcm2HGlyA;{@Ju<UQ#-=vy|9P33%AW^RF%t`JLz}AfoBVz
z1VO=t-!1LkJhFQ?sfC6i&2ip!e9*7R_ELH=q~hiF16nR@8jwX+b;MQaZwTTX&glZ0
zTu+yeB9O*Y)SkzhY(Jk&f98HwgDTP7<w5GL{+Kj~uhK~rYQNPt?8bzIn+OENOcG4Q
z#z8orWklE2^<sy}JAAAEHffSV@1+OODY*Sn!C^JOA___33~s^;$C#7;9hkuUF4o}-
zGEm5N?+%s-2{>PIJD2PEAtSvh+1OC!_a_T|YiZVFG2ZytBSbXH`B+%koPMRlZct|+
zQKpj#FHJ!J{MN1$`aK|hN!`_1->04gAc=VA^=I^7gcLw~-f-q9k=#VL%O!-*$7v43
zqLEtnc7i~C$9F5+y>=Ziek0#>WCnh-y(}&sw3pJ%HH{u0R|@-5NJJ5KxgpE};IL2$
z78&38+ONfxdn+&E`T~CU==0rM1f6+fq#o%aPNTN-!DLS;K+jtMFNOoDB2~Jzp97x^
z+m<5wqfYOiueMqnfHJ7p)K#2q^&ORaD_%Ze{k5w;s~9LBlr&X7tump<BqkQqZLM+&
z$E5DYDhn|7ecsS>6|;N#fPl?@T=@Z!4dh3Ob78xdkxDs?*~Anyvm@Yw><-*2l2-Ad
zKrVID&75<FU?J-ca`6;LymotSv7y0BvxszwuJtTD5U-j8ao!kVvrR45wc>?oOjU#M
zcOuWfo?Oa{DX9Q=+)?KF;UO@z14+f6`LqT*ok8(==+W_M`UD&AZz~eoqg|W*?s_;@
z=JVKR0ZhC^$Uc4EX8QAz;&<&yQULUHmji*vK_Xt7N`i;IMJh^snPKyrE~Gh_HDUs~
zN~t+#CrysO@Q&wbl;~Pk&eH+pw=#mgH`};&)CoX@)Alo^C#WnRuN??Pc~Y%Y%o@R(
zA29H;bO$XiE?fDEZW~SQqWS6Qkx8YR)nam&FO9$|7_j;>zOya$GY;2ziO;C4E>wmB
zuBpEI2?o|?DKGH(>hX;r#D3`~kSGo#dCs&SNaH>QS~dqBe(<U<mV#g_ix1rWGykzn
zr{O|#PuB$F{*>3|%*+nV(PU9r#rQ*v2|yCQl_W*Qe+Uolu}@w2KV{+@y@wWXeZToD
zMVvSZM|`%y?r!U>HJHRMAh_;5z!rgMI2_A7+Q$~SfPFV;sR6kpR9%YOCHs$rwtqe3
zhAh`3vL9wSs>sUrJL~$h8|+35kW!<$Vxs$p3`sWbJ>@344Rq+#5=xCvr=Tpdy|D|=
zPm#_yJFc{5O+km4yTwz$+r$hFZH^y!5&<An-ip>S(96>-3DlKVz?<n0f9E(>IaaV{
zsS=;9E|B8<sLAg>5t1cLV&6O?_&DQ)z%o}&IqWmPvIUFJsEKG}R-l9rsfvF)CC{qw
zRZe<yY&DJ3<hm&G+%g%uIbQb>UkQeYa`pA6nBTq<(^<>#JC1hRB662jW#L`iJRiET
zES9uw1KaJcz|2@E@0o6=@gMm-Np-r-k-#eA6CJx(16N*`qSGEVpbfPdJ3rz!v*rOs
z5h3}Z4o9Qy=?Q7(N=1;9r3DPJgQg{qI*)N;(m*SmSolU^c1sJ@n#e@B-JxaZVLrD%
z<VR$&_uwPZhZ#$DoHtMu9%Dok*G-0g2pt#`r&6fVV{I5@auW6b9Z~(y6R|iT+=vG2
zOJRjBKL6B@ZR#DT*3J8K#0<<ly~DI)ptGjltYDTgDyNM^Ve1Ja<?U+UsTGe%78f({
z_vqWbUKvi&3z99GU>?)EA7(Rr|J8J$_52$b&Z!5C{f9VAc0!j_K9$vAw^>FuAfVB4
zQzDK>;a8Pz{dL~-9HzbJ9Jdv?>Ab!3Y>`na7~3pm9|EJS@fP)~Fs%|Xq$>OCAQdYG
zEw?4@n#;l}K6OTa!FVZ!Wch~@-EI7@qiLQa=@uCK%t#eYl=ru+I$Ib}>L+}T)Q^;F
zKs3PivR#rCG-e03JsNS52p@ekFIK4%q=Wp%y#pJm&kU*S%P;zJCySe>iYl)C@L){)
zR-c6y?mtfpR+ZeH=18wCs+MLS2fJlS+-V9{TdCH~X`tQ2d9Lc`t$}TjXTZZ`K9|FF
z(PNvo5Gd)aX13(=`Dfkej>`t{8=PPMda{*sS5n>T*^R@>@#l5?J_7Vd|AI?@`$9?k
zsWOWcO|)#StIjF~O^uGEtNAP2hUC5`-_-_ulkW=yxW-2A;?)QJ2iX`DXHubmUBfx1
z`m4<nJ}{`(fbS1DzUFs9?B1MUkTj2`#1Mp;g{MVK7h7+KohdJEKc!P<6`-(MKtV9+
zyFEooea}$kZ$8J|XkQ%dyKLZR9e6hZaS(h}@$g6_VTq7HiwGJ~0BmmU!@n*@#m-}w
zcUNVlPm}~hDOQx`X(T(uDPQfu0xL?K>4G1ziP-u{ymq&_8cc$yq>))UpSPtejeeGE
zQq{6&rgIG_P=E-j`{TqoFx+=A1eG#A;X)K^9B)P3xIdVs>tgmxfjXl}3^_tM|3og=
zA$YIjdoYB>Ni5cQLgPyiR^`H>2DxRxjHDuL1^9R6^Z<4!;`!1`t62LsfLdBa;Roic
zaU|HIulwAf72N%SN1)6!*C%f717b!dW5IBWfwXHJC+p#{6B^^8*Slrqe_(X}*F}KN
z1r3~uAOV*dsaU4<6;R}Z(}u9?`x;OA$%+g;bsGF{GXc#mf=Sp9`GoH$o}|FMYxvKl
z`+X6@O(!)9-j58c?pM>1m400<7Zo}Q+Dh73bc7DWEFcorUr6U}>r#!UzIe9P%d&R{
znH2V!TUKbFZJ;ptocco`pfd1z$9@*OyLGktZI*K5T>gIU2bILeUaCGjtozVD-=^be
z8Q`qcE4p47`L#i*!sogS4qoIclZBM{izV}$1R$gs5&}S=g6it8@<Mw7wAtXB<Zwt?
z4-Hw9S2!~2$u5mxmAX70+nGJm3Mr(u?BJTH67{eX0oM)x+xg8;N8@#U%H%XsA;07k
zRGMoZ1(WTq1iTus-d>5`o)$aa9_=6B2l?{~WOVNm$sgM+=Cx-sY()Jk$H{gHI|;|j
zwlOdNN5a8>PfutOw95rHujxGZHM;$jd3qAo-U3-^G-#4|REE$*NNdom^!Z6NE9&l7
zg90<ZzASL$xSccet^oyHP`d6bfm0YP#9_ByWtL~*I#rD1dKeF(joNQDr38Mzc7hcB
zyfxbtOV=3ENsMc=`BAowS)|4qJufr1EjI)T`|BdK#?`&fogvym5Vme`$J-y{jW+2K
z@19WAAsRhr=SeBgGCd27u}105Z$zmVuP-9Nlocuj^{ba65od3h=$pD5Hu%{wvm*;+
zHG~bE6)CyS;~SIv3x)d`G?P9?#Z==y9Ic}9x!fjBDoV<nV1F$`!HV{@Wwt~VV4;PM
zr=Nt(u%Q7`3&{I6GqHVKnV`~AcN;Y}0&VCa1?6RXT8gV0{q?J1$F@r_VjPJR1rDJ_
z7!-ng{_ZugMXL5HJiO%{HrWT7CdU68-vcWJF<g*RCUVn_rc3V3H9+YQEN}l_KK|}N
zHGcw&ro{AE5PffNKBQjf=jTn}GP>-;!%bwu^TwTuYBZ4$%<^5LS%K+M=Y=*RR{N$P
zW_uTgHCqU4*f<s;6@fWyqaz^5VrczgvF1ojIpdvxn=24eetKs!e@?XV@`OGzU;jOK
zD)<P;#9ADK&wV1oc9j?q_t*0IeVaeszr3H=6@{C8JlbyBjb=pXV_xY`RS*Eu9CQ>0
ztfh-pK~DV6gCV!4^@4i<8-iwqch-QM1$QIUjXBoE3>bm}Vx3xZ{%VuJM~*V}ZCXWS
zfcoU%&xMTyV$thfw)J3Qf?G&r9d&Xtqy_oP9CZ$ZazXGZ8Nf}%YdjSHr9)QZSQFqM
z2MzyvxyqcoQ6E+wBULFU99RAp*<h=&-XBF6mrW?z4?~-?n&7=rlN#{vinJ^_JY;42
zH*v=r&+g0dl<uzRn7rqC(VXjJCzIxS-fN9gDtarin=+GASE7$mo_t~@os+7O%!MQ!
zduba+Qu27|2w3`-GlS8)!&K+YX!`O603$Zjjqp=PyXXFFeoPB_A1Dm_Xm)ygJBEe5
z$m;xM?1!+Iz=o^UvhPz0(hFbt>Jv4lW&_PTP^Mtp!O|?ioEe?)T$*#}dF##!<QXw=
zcJS9t1RrDxM*%Y-rf+lVqVsZ`K!^Z<u2A`i<4CU7{QxrwG+0-5zgG_|)()Y2E8C%E
zPiqbYY&0M;x6L#*LAb0ts=E#_cqfYN@7&L*LfrTWf$;N})T@HyC^v*HCe&Qs4o%ZV
z_u#FEz(q3HN4mVPT$iBlN|%9SP>x08z=2Nr5hy`%*;L|eI<4G4q3mvs;z;b4W@&kx
zRGl^y<v{MN8!v9ww$1m&ICnYsi<)H&|M%GzL!clp^hl&6uHQiC;zm8nj*hJC=moCE
z&gug!_cwvis{P#;BNMs6?^I^b*az23V3A}J0;@UgiwOzop!i?Tpt!N%)1}2=c<p6|
z!PzEGR!+#NV7EVCZhY=yM7@Fh+WWH|=1A36S6J{xfamhA=eqZHUnmAB*gZH&Wq~tW
zEfh+NHL%?gCsnVGkr^R{h8(o+cb6v*I7*FFTe2=ZJ6`I5AI%9yQcZBjrk3=PFm148
z<4yotWh9xMWTwDTK*(%q8$M?*0Z#yMMGa!UmRd|!=ac43J9xUfe2k<4ZBej$)Q}9s
zgwX8()*DD^?>DOrg!E8fA@HYEgr@*4*2?!>&d3?i-5`e1ieUct-;sy22+pV9Fe%rl
zImpF@4T`E;-3IQS!hxL)449GVWrd&Y_DrK4)lxWC^%rC0M`K9uBc+X~;evm&eC%Zd
zne;?9ZGP`I+Drqc$rVCywZyzs+e0Cni_KJ9-;aJLN*t9sefqYr5PHR!3R*0~3iwiM
zG)Od<S-;Uzs_Ee{tKquh+X|)H+l&4L{dmm*2-sadEbj~f1R#31fIeR)q{_{gyL3Qz
zXB<+`yY9UGE2R$OQb5ZOX}!7Dci*}^=q|W1TK;ozCz`1Hb43L(M|0Rs3O1taE<bMf
zF<hA(0Ns|4=ZqVe^^6usmZc>A1k|+VCu=r6U4fc|ryTB85I0Aeey{1G>wy4Slfo%3
zpd*@ugwIvMXBLSq5j7Lnq}$JT@yLoZ85mSALA#z@ghF6!hTXHZsk3Exl}}bsUTINO
zxmDsBqL6+d<VN)ND^iWZB)}>5pQj_`1`+}TqqHaw>1^{`UAVS;(CTq6`=3rYoi{Yr
z*LOSRJVaCwqX@q~2^FZLrqJ;~oJobX`78?jtI7G>7Z4u}xfGw%UTR#TvoDk?^8Q3+
z_kNMJh^OzDBKaVeiu(CX`bJ|IW)?emm($p(MoW{<^A0@#5lB;TBKm49>ioGJAk}va
zBLO;lY*Pl;URdSI1GV2iT8Is^-J);r)C#||nyU?SX>OU1a^`E4cCNCa3GZM&x$GdH
z$m5AE$u8O+;xNH$$IxSHx32_}3KQQSP;u2|MGFeAhNSC}-L%HHU4?xL;)GVr(d8C?
zyyZrWtAP3Xm8a}wgHB)bcHjFrhHw{qzH_6}2sH!^Xk^hJis@k`5gZv0#zks30+7U;
zs`g3=-hfs^GNE&!`iFgpLo+XY+x|+kTrAWket3+(CKUYP1pXsWnJUd6pSQ}6nabSU
z`EV<0NKlsWsCLay_1tsa?yuQkw8;G`02R-cNcn>MxS;A=>{g0=-`0L2EEZpW*0+b=
zdSb@ZC;Aco?~3<UJiz+*+&rU8<cH}-aMRImwL$1V%MMzfW@A7nS}`(kM6_<alW7hv
zjs0`+<v>jyuF*887`QO&zM6YqMZx5@T($(66g#uF9Z(9MHT8f-(`N-$Da#nOV*LLa
zP_8&9c;fXUu{GIWTW-0O$mfm*z~BPddo7MG_|R>2xA$l^t4hub*ylT~e$!rBLG}p_
zH`(vO*$(X7>a*=2dZ5eJo&YyDDEW$1OWYnhU<%$9Ze-r60oG`buS8_M;g!jvK6Qw2
zV{=1V5UnJDFFyoBA}aQZEt%(?t0~}|I^PU9Sz+zGBY*`Mp$k+%Iyoz|5$ShNt@pct
zva0FvThzWsjIbn!QNBTIcZc$@tv)dDOk}I{HOk0LUT=n`blFJaClSED7f92i^h=_K
zBoRmj%q=;8Cq<lOX`wbW?WNrdYsse^T|4IhJiPqG+cz=DHKq!3K^j+mQX$~Pch#DH
z{M$MI*U}x(A2>qh%Mb;#?>}ob+h!x2lO0E=zuu#>G_Z7A;E)CXvVp@tnt>{VDOlqE
zkz1(=wTD|C``#Y^V}}UduGUM?jsuoSk#5`IGBpzTQW~nkx_Ya)P@!s3yfi@OI^XX;
z7cRxfR%U<9)%(t1hvYbS*1tHAg51D;HY0gFD|q2mJSo#3a26&3L<Knt1ULQs#GNdS
zd{sVP?pu&gHBzxx0?VFvYHXp^4$5LJ-0L2NYJBIqK|Qs%rlUhK%a?tQOy5z__h(s?
zoAxy_Pyp9U%vLmQ`+kE>@Hg3z+M$HmaWBJ4w3jErU@rX)E4{Z@SPDY^H_}aTz+dvE
z9Bp`oH0s1BlOTE4$?Y&Y#Fg=oI5e%W41ZiyaEf8ci;Li&<UU_i?2G<ap71+D9y{CK
z7*X1Mlb(z7UjokTdBD0^va@f9!>b_VXm-01Wpb~Ob|VS$tUsz}-7n9;ZI4V8+Sydn
zX2T@wa+-85@!!1+n9-eUZU51$|KK`O^tw{y0z%;({VtmgE)ZHpGj&4gN2^)-mm0?-
zCi4wGge4`y1C<7~CxmFSG8?0~_{3XvOMPbe>BecJ?JiNu((&j{F})1KB@NdlU7W#g
z_amBst21<lK&vzg$y{aQRyUHwvYYbt**kni`n#_9PbGbI4${VJLcG8_j{%l7b+SDD
z7S#v|uVu2W73$`jx#?B(AJb1#mQN-bn~|O_qG^9%3ffD0*anucb(2-ByHSYBf?aQi
zZI^Q;GE=!^9>6~`4{zp=s~vy!ou@UpRibGb%3X@_5cwvEm7zCXPjOM8O<^>i0Ys)R
z4(F9G4<edPy=uZO&ir(+X2JV;8LMMYXIv2kvhg)<e*~c+hlUED=bO<rhf<tJ@R456
z-^7Vo(%9f$84#eyvXTM|Ijbz=9?46)wE_Auvw45$`qj<JqU~GTei#-{BFnIaf57s>
z;WY(6Kb9j;QFnkhTD=X?I7e{xX3Yfrg&ga9Lj_=2@K5T(|9mK=1Q-Ph8~Pbs@4><6
zOY_H_qUK@ciTF5v90BB!3EyU?K3|$ftXJ0oM?)YA7{QtB$l{e~j`miWC_Ufmxknd3
zpmPaGLaG&{8mmmk3`bXSe9oZQBNQE4uIDm-yzU%013`(@+1U*gt95x*G&R_Kw-{&L
zY+vT}^{*<Rz;fS4!GlF}-70U)qgq?7>WUDw9rvHy=qrr)tpX<cgDc<{TyJ#RQ-C#}
zMRwUiW19<=5O5MeDnI4nHj%T@+loSX5H#3JIab+{QJj5uzVD)pAwVUYlL}}za9fIw
ze^``ZpW{y7GL|h+$pm!pGFkgHm)RN!TrgH6)CsBL85$BnVVHxdW8?FB0$}9CPi#=W
zy5?X|jX1EZCE2VuPe)L5wEJ%v$ZbCZ_M*xvU?w#q2N+;-q_!Pz4{xuG7(zHt@&8qw
z|DjU+=R*)BEqSJyU57#2l@|V4o`tfEEM+I9?e}jEnF0h=W)T{EXlW|ty6)m@PW%2p
zxK_KvCZgq8hk5}iUuU$Q#)(Q)nT}cvK-21DV$yUE_7~)|s4e25>>de0<u-kGcY&<|
zG!Q-qGee-r$c-S%TQ*#=6kWo(TLY>x3auCofcMwZd$|qdNt>^DSZw&v&$!CE-QAv!
zb)UW8dcgZ>9JvbMJ?2#3fwXdZt!?qT<5+!m#`QNqC$jZ&=VaizYD@Nb>tcGgvSRk2
zIe1lQDrLOLDIaaGB`tTiT5r`RZj8S2K&9;ZEC+0)q#*O9e^3}S2BO$36FR)TFj8d)
zjD{Y*rvjo-cLJ$)j1|6Lz^fNyu5p>Te)YPAZm%#e+;tnG&6z&z?NlwrN6!o}ADvlm
zl~v6YB*N-m<!2cMst=Xm7^cTK!h7}%$Ti8eN2z6=H3n=JyJYEqcdSkRI#^uaW<H(k
zFV+qKbB{5%Tcp+rO2bLx2?RKCs;Q|m_g9Yu`v8OvNI}>8Ap5D{rTSCL+25=gpA7ii
z+5afrUs$Pg&J&BS1#z0qEc?E_u<{XyF$tw}_AwWjp{!tlZ8Py!TLsArkJOPw3ZbHL
z<W1Us;W=xT>Z<nlF@PA<^xZcg@BqOJbwNGKhZmeLUNK~Mn*|Mre6CYLwvG2e#v|A}
zX$l3>#!HE9`1hZ%HdCz;+}9G8wOh{2%^!O=i0DL~*{cTWxw+-RdW~lf24Emo30U)O
z2+bUg|K+1$g8<AwdrK>q$em3fBxXwlMzXW1r(omwjBOQia3ENS-H&JZ-!jTIUND8T
zfx+V4$!4ul-Z_<S<9S{LGQfoU0x*S9l9;mnM0g^GH2Py2jW@eh*?)}1%(sY4j<K<f
z%0pBl?il&1h=zRXEQO3wp}^S46qjvEP5pP}K0HS32d4lK=0NcModDqaBkaBzeDC_c
zUf0F_fCB6Gq<Rz3<k1_6c%O;n2J6@MM*zNm;_qx^fbkxsG9F-c-Nr@xjM-e^oUj5A
zm{+J*-Jc!sKBj1%uOxO~r5J{v`|wO?ri@M5EW@59>$I;#_`YQLnq;eHjm@riyWJvN
z3!i@imYWU=KuS(VD&J%<UG{q*IN&t6-nA7iQ_#$Ko)qe6?skKO0{?SpM(1I^h5daf
zYJEjf$j2mS_cQ(*8^HTs1pL}UWV6H<gydnoSw0HOZ1Am1Lkq_jD>m_`M832@nLYtu
z<)Gd$F)(%tpaGVFFn@`czfq3ivLJU$3F?IQL9Ww5>#<HtDQ#ST&Biyz^$-DO=Gi7c
zA=hdxKpn4JTt@eLIbSLm;Be3gOuH%!inn}M!dVR>0SP;GmR9-JizH)#LN$V$%gz3>
zJ2@R({OOt^oMFAq%`n^VyIDtYR0d2S2)d3u=PE2mN*d7C{+ofXY>p+o&>I*jMrccw
zMS<-R1CV|NM3W%lJr-3E2gU&c?~$xiT-!ujVu7ElR;Ng8LH7IHp)qT3K8L>q1hRvh
zUdA(qFpocexGqpMq=>vV0T!7jw-cO<8i@DXXiOy<@+(U0Lmyi{*ezkZlaX@vl4oQi
zW=lU^A9VSz`A@y^GQA&EAi4UBWyesk?Xukk7iCKRz3P67Z!r21*o<^@k06!yMHu-i
zcAKTpoN*L#h|HeQuDu8o>RXF<gy`K*53w#Jvyj|h4h#lMOr6?Xnj$SVqIoL<tVkM3
zT4{ZElV>D(br4rP^j4%l@d^;;L$vO*-{0<|vk${?V!gf=L}WX$_EmW7qW6o7+u8L-
zUPKV&`+!fm0jM%D$<gE?@oj*~r7H*3hfp*enAmb{6+}S8hE~E*h4>aP7Di59DdpIZ
z2k_y=i^^V{WWW-Mfuw3`T$`%yH*+~4AbiRWw|4)XBRMJ*f}}V=QkOk&F!;CN6c$ny
z+xfU$iCHJ_1eUUtrf|uyV=eqU3YJX9uJdE>bvG%6y_3muLnr0nC*Bz6N1MedfAi66
zK*-T%FEA7Fb%G%vo|uc%d!f3AuF31l+k2s79j?4g+_qztydaKEzmD*zA#(QG@9nhD
zWuCvv%?_A?%+T}SxW105k5~IS<Za*3zdY)vRt{kz_J7mUshj{P6{o#*Uu{>sDlzQ9
zael2K$q9}}US|%7Chp10Hvnfnc$rJ;^Y6NV7>A`yOU*mC=azFse1&GpT)-E_7sKv?
zKQnuZQYkFk$Na<oSQi2mF)A4_<qj@Z%dbVtG?HNb>z)MlQdgNr=^=v!@Km;U$bYct
z^-2&(XhVxU4l=<0`Vv%cd;9V2&QCLv;PdF~upRIZ-Y{Zod4PULLWn3ec$+;N0r3nU
zkT?R}OAShEhQWSeHrN+bnZZyhgT~1`82c7T*V#truBiZANdGtL;rJj%w|C?P7<i=d
z)R%_88_$4fadp|)&n~Ce<Kd7jGOzuAe7#jvmEHUOEg?vErywN_(y0>C-6^SbgLF62
zT}pR%$EHiVTco@DzkGi0d-!z>&KY~(`(EpcIX@E>aC09skK3>Hdy(sHVlHT#vizGm
zzmo;&;iofxPne{bDKGDyR@)zMs3c9)eJjK}Rg#<}+p3t;Y@I15jj{*O2Y7tWcMzKW
z`||cKygkY9KuwE9p}$*I%%J_aLx~oMcD;7D?}t?s(BxtGx-|{aWlJ&){dB?hrouHM
zCux$_)lS!wM=f~!#tsA7aZjnG_&d{)!Ey6g$gH08X5AuklAwUviz4HX;6K*6IM1<A
z_B0cArQglr)--me%eBx~Ct7JAE({n7^5f~IqNvyOQ$TP!GqD(P(^oybp4_?WD+3pk
zG{Bf9jvu&>=~k=Zq_XX{AOxQXEK+NwX^vAVQQDu11a~+l9vTG^AoNhTEzAJVA97mu
zY5(wKh<We&xLD2Y@u<0KX{9yYV67C5Qe14CM|QuCwSA5f(DM&0v1Wq*bu2++MwL}e
zM3;*q@evQ^kpnR@4KEh)4N&+Y2SWCe8rR|clI-TZy`D}#Us=>Y+waK#COQ1JE@e8;
zfKe<i9htOvt5t;kFxsoQM9<e^Hr0PurmJ)O%$)9?`-?ZBtI=LisJQ)k^f93wjkq84
zaFFg@B<XZnFC86v7Y7WlBioOU@P1yr&KhFfflhNOCS4?~T_K>_XGt~)%XBjGZgRW?
z!-}_8dVI3!ieumQrWi1e&&NzdTn?r1C<>qTFXQovxM0CjyFZ%gS>Hd^dIl-Tmi@x@
z@cOZj%o%a}jA473$a)GXU(bojyf(IBS#UTlF*K7?h3VapM6C3Oj$*r?Ob(3XJOt`Z
zQycB0bZ->2m+DI2bMx>8oXmpTsAe@28_DE959=FrD4q(&3He60vCe1~Fv76&6nc!i
zN5+%TX<$NlLkLXbCHBNGTLP@-g`sux*_{Q-f8DJ(XCeWaK{wht4}BuC*Gnm38eL=&
z&shwMuhKv!|G!_4H$Uin>R0RxPuBPa;1L^F?ISVRP5(5xynPe4-vioCaJB-sPUGqA
zHt8q$jSWN-A5EvZp|O{EFL3DhzzylA%_L+O=(tPG;HA|QUS=2Skgd>k_ouw_!5Y0c
zY$8DA=_>fibe`u;`<6IjUrur90c{Kg%x!wc(RtVhxwMn@8(T8jH1B6EnVt0ZsgtMO
zfcL7Cd9I%@vK+!JR<!jBUO7*^p~-M(lloMkJi$Je1i?NDSv{?_N*51KBYCncfKa&I
z8b%VZHjbgDwLO!@>D>u9m{d-(Kscc0MY-G?q|#EC8?H13fXnBHav?Mg6pH3e7ZTHW
z^ugJ-a=RczxxCKb)_C@X97Lmq`w6Y&@roGt*(KgUcUIT@TCTD<>4&gLuB67sa&Z~r
z%C~x9T#yDB2IlQaT#?(Mz7{vYJ`U=F*TBdI0|RdY7pu%V(qQ`+_E4c@`YO!Zy-Y$@
zkYuLmn;Ng{I~<(1B1eBxx#R$<$%W=uVJr#Y5A`&|5J@J^JQ2KExqWAmRLJWVrFc*3
zwG>Q~fzIrT|A`zHI1mC~gh~8!7lw-f4s~p+&^_Nz8pkmdp>dYAiypu;ImF?7YZqB0
zRVh7&PU-_xA5m5r&$p3j)GOX+TQJTK#wK@w9^=%0UanYZw{XL#RpObinforJO8Y2N
zr<vC-Z&lfRw#aI1ZQ*;og4xNZTBGnV7-+>7pv7bYH6;<4Fm&%ttAe%9%?(2B*U9|m
z2h?`ERyJezn_E+&z1e(kD&_BzyXmrIX3E_DJyNm1O5!VY)n2iiHnCW!Cj@K0mKfcD
zLviq&eORcaT$NR9(aHagD&S_PI78g=@c!>l_dRBk;PUITpV0^|nO-ZQL8O@3m-t=O
z@=X_o!SE)Y7YdkG%@Mv$6<pA2(x0GF2#q#QK4)^b#b8!>M8De`CdhPg8Wd?rD4^P1
z)_RlHXvO5ypZQ>SzQO#jYa94D$ypqW5J2haxeR<1L&vgO$SMqH6s4*p04`wwI%S*8
zS_E-j_Pgsu=t+ul$qzWR0s1iTgn%<{Ld^fAr%Z=)IK#3NfH3?pk0tRrQ~z>;kIGW%
zYB0M4Upgho!ZvWNT66@Fr93hXnH{q3rxn!P6Y)0}=4-O`_CP_=t&s3e?cyEJ6y<h@
z;v^IVJLk*9ZY-B<_J6quIsXyAp%ZZ(X-YcpM;Jf$SJ4@LMohOn0~7F9ZCpQ!$Ovh!
zDgoRE5X#CWoB7AT#Al8@<L_@35i{zjJQ}V^Q%n-QK#}Ui-s*>90y7%_9EUt<IE{}8
z1-a?jeDf@5$7;!hlU}89TAT!e4*4+gc;&aS4jksI1mWLmP5M#>{}@95z5WZ(S@0ch
zeq$l+O)DL{M<2cRgX1|ptfwlbd@n1p+S-h9++x@nP08v0-1;(--#{b`*EMO%F-Kuu
z_cW$ZBT1RVJg@KZn&@=i)C7QUP@0_=PiC}mFZNAEYR%T2SuggCvA?(O?U!rs_S_#Q
zhL1WmdbK^b5y1JKt&Fz~dsUuYT)Ny{Dz7i?`OgwH7Kk*y>lF#}Ub6Co=VLYf4v(k(
zqD1@axJZ&>McbIrao5iH#3{#q6ObCdv|K1S(0|Fk_UsLtL+i8`t}N}|^r551cyI@p
z5_q3qe?;E6vp{fV5$X6K3LPu4Efge|f}M76&GL=yKcZb6HZ%c2^r|^f``xZ&MR?Cr
zL0qD($NIIpi9eBZzB*Xk_S51wQ|i#Vxw_fHvtGw|&lI1Z8broK02AXU(q#nos-T?&
zse&#0UvIoOtbv~}unhebxSAEYdZcBc{d9AN=d=th`S_tVZZ0C;U|Y2m+2u-u-SejN
ztu}(?%^xBpY8keTUZTrSvfr1R#H&qQ$F%_Mkhfjx)Vw^|A10?~%;eFP9(1zdqypgu
z+8R3Kq-KvMqwkQobT4arzdLR~LSn|}nMks?$JB67S9CoQOh|V3;W^C{h`=b?zfXy6
zuHcyJ6pKE95(uGBuNFz;m{8USjc{O?YmSRj`RD22=zb$|X1Y(j8RY0iE8s5Ny<vs7
z{d{Y`Gf`|H&jDCJHc;inpU!%TsMq@~z&hygo2qJwxmxyfO1q`jK+6$)t!>hOl?LML
zsAz@$E_A%yJTos{v`R$V%oZTABmh<eIphrn0vMQ0Prqs?(GU9Dqb1w@X44cxZJ{|`
zqT2l+YQD|7?Z&{-DV6gUOri$}`Sl>FbU#BkCtW{S%5wh0v+?5%))!!>l@qFcl_&(S
z-MCPC9+f7D0%}$gup1oz;d8Z7d5##q{*p6w^jI?aP?~O;)%oD$z0+KOq=W8!%Y5-5
zdtj>Cm+2gv3I)o?XF=X~Rwd%G)UKcMH?lzKw{v>4Ne!GWXQQqcv?ptRvhDlU3Nxh*
zgL{*GJI)^nS%i5H{>)?*?OMNMtbG}z(<qyH+`9c<ED(##^UL?p9p-w6r-{&xU<{^q
zmszs_uZ{rzqZ-tS-Q55@yO(dTmrWs%kQ&0-lyl&O8_viK<ud)vBM*Z;Q5gH4xY(Qc
zQ!Ggi?QX2x@R#{28(V9qrcdpeo(pe?lhZ$lkUI&21;F1I`^T#<pXVguSANJBQ%`jH
zsNtSG-80o#jkr;$%!M<s!cI(79eplVn|=EB27?^VZY}=$LeI<ceGn?@^F3uQXZwjS
zlI5b<wV+<11ydw5G(YC1xm#u3KWh}pwY*g1t{%AyYS>rLPe|j@0EG<G85Rj`l#D*^
z+A7OEz6WSSV(Y<Qddu52qcg0!XG&GwA1ddMrxvU8p-_nS^eZ7&vANk+hYP$*?$xA}
z;)JZ0DFdsWHD2vF`La~EJ&WunNvsx01I)zqW~Wx1jqs!_h_5S~Y}A|LJ)b=WY|D0M
z$7o@POm@!n3Eb9+<JdELFEp{L#&6s47*2i|io=hl{lu~sWMRG8_JL2^NZSbH=Unw#
zJ#dDetxB;w=?uoW?P>P7!#~+8#xD9Yz{6L*4)Y|IYe%rdS733K<-Ix)pupQ|8pLUN
zG1o9o=y+Pj0IFp_o21$G^D+CL<@w9C|APPtufrk}ekuFinE09vS`)!V2soSUw_V1F
z=V#UyLOzf)GVZGv=%#NM_%2?odU<RJB;Pe=S?j@=%}g2Y$UCl1#nZRmJUX}8)AuJm
zkJ0>Sqe;)qkown10UR}HAOuUyE2uk1VsJFC3ITZM9w>BoAzFo)d{%Ls#RG3dTQ#~?
zTE4n>B}gIu1eTS9wP^mWYoX`g&uKid&ZYfcL%TD~4Z<{nC)V{vDvkvTADMZheml>I
zJ3&MZ-(K~bX9~d&s3MMP)j9mOPaDZZ#OI%*%k!?Mper;w$UnnR$ck=0cIu+68RANO
zT8YfD+^wax<o27)m?9SWyIQ}TU97Z7eyLLYCltmW^F+K7D~Y8uq)@5aKUPs1+%|Q~
z_Q}bD7rUE~jh@C2U@m$In2vO?%mjVX3RXruFRv%8ht|{QnQYSnd+W8hU{p&^HY)qV
zUCtz8(H(;7Jl`C;Uundy{sh2Hn4EilnE#{&-n>wZuq_k0(cno;l<J0&oBm8@@AF|b
z+z~5(Dfb6s71oZ#rIx2{BUS8a0p4ykzOdP8gm9hJy#W;2O+fQL=(}2|q~mhEj8FQQ
z-hTU_CJpPb@iO~NS6MpVpz>d@gfMh87XG{b^~^E>d0F5eKUsa&R460nvxcrjq!Fb2
zi0I+bb&B9Qo8PhuW<Ewn9naC$`?>NRv^5^Nl2CMXyr({^915O^bd;jaym5@FE{hVp
zL8D1OkC4BudQ3<(7!{J|Xqi0oO;lc2726SiTjgFti+v(=UMZs$UDWe1KIQwAi8t`F
zM@)YZ8;@Fcm{9t(pHd76aO&<)Dh&nOY0vXB?b6eM5zlZ-QYr+a1#hc2UbZiBh7U1l
zIYK9h30cG24dAgFj$C^jnhEE41IJIm173@fx^Ovp&&wX<6Il#4ym2hhkwO_Ez;rD#
zwWrgpC|_@Nns!cE<bquK(2ZuL?VRosH9_7F71`{ZAISfLW-4ooJtJ+`WcqDAzW)x!
zn6K`$hsNceYdigu3|BJbTz5lYGGH)dH!M>4a3OiFek|y4vDConH7fK-GZiSh=|&$1
zkEws(r;Mc9H@4oF{k_(PQ$^cWRas5ej@vB}Zb@nMKHkz$fSO#)Ka&1mE->MPztYbx
z0?2C-kH#*h+y&0BS>0T#@{nRk&``G28mLTnxNE|`Ga8R!e0_6wSdUv{rBfw>E+kO@
z%2Cjb#S|vZE<dX?4`5BtpJ<ifKws}mWJRkHe#RD@yUN^t(A8k8v7y5=vIm5*`@zlT
ziP*lp73OM3wqQO0PTW(p7r|7xA&QB+<@2m|ad>2HXiJ{ric1{(kwPth=id-1&Lc|O
z`b+iC;8f3|K7n94NoQC@7R<rGE4~&@MefAn27Eb2Eq+j?pv{y!8!q#5T^*0L&LRlr
zcPR~aMu7`Up*UehC`=*%k@s0xl_yDscS>s3D2!z&t{T68`zsGo;K!YYv5d>D`+g18
z-&L_R9^9Tw^nz!LL4ZN>bP<r4QLBtP3S6S>)Y27M2BJq~W+nLf0#HHSG_YIAPXsoj
z8MU|Y$mV*I4}a#`T1kcy>Dxax-&?`_6;<9{R)5z?9HN0Un|owuqpQ7}R+GW!2!+{+
zyRm6(`18vru^H-rOMqS=ykUaksl6NF;sbLkJ9rVN{oG(6C#D}v%qNoP#PY-=f!gH+
z*GH9oM2Q5rH`*6v?Cb{S#pNQ2;I^HM&n$!K>lwP=n@+nARXu;|>`b*z8E<c|fHR=w
zgl0EsU@;I$aH-ef6KH#w099i1<5eG-z!MZ<-}f!fO5-7hQ!5ExK|3w7N4*J&R;$Cw
z0#_drzANQ3B)2TRqARM4opoVQ7_1}9VE9})BvS%09tB1cTMyG{JZOZ_$Ye!xC9%KY
zE-~kLpZt2we06jg>hnn>Go4Rhu4$TH`+X}d-Fa|+-iP52G5&nVQ%-?_zAd+zuxFTH
zj!17}R1rqtNe;Vn%0<p-t(A4!?0$><8p6!KU1jV8VCx%Gbm`pn+k%TS>3dx;Xhs6}
zICX!F6T$8VR_n8ZwivW0w7Xk3$v>sXm`UCa7@AE}zMPiCa~1D(Aousjf)Cq5PVHFC
zqAdt5N{}y;t`{s>T3l$jOxT=ih@X#jGgrqQ{tU-V4jituZSfP9DtSR33<butnD9OY
zvG4?JgPAmtQ_<}y!QyF#V78qc)-U7puoc6op@!c)*jx5A1BfN=hQ6twbNznVBvQR2
z5=l)~!*K-ZH`HymVHYbi#idYZs*E5qPC!>BHX!4$d%H?|#5|k=q&rUt`0tW_3F~#$
zG*3id^<$LZ-26rIx}SMHpUWlKsG)|^(NXBsQoe7xxjIbs(et|0-%~c%Uezn8jkT>5
z{d@1W%3tm^LQl;VeLkHQt5U!`9@NzFsMY0m*bi{w@*+8I8Vc?swcm0R97;9VS0&@L
zy4VP%mz)!Q%?$gbN(CTLZ%M^s{zY<l!1@G`g!#&elP1S=KLsCm@?->GUyWd%4(^H&
zeELC)@agkW3P*DP{#(t82aW0_V5#wTI(#Ruf<4R+_w0t`GfsU8FA8Ptgq0`jc&{@o
z_%uhc6z{@dY9DfS4Num<Y7{)lELo@`MI%zHnv?UTTTgb3*CbquySDM1WAkqRQoSyG
zW7%E)-P1X+ubqtN!@X<bhN*Wd{$YD0n!3y}i*K_P?+S(=NMrm;c5olD8IAug@1kD8
z{$^Ls>CEP5Vw@-mbMUCYf1c+<W1$Isj%eTIX}VZZ#uh=xP)4uR?+hfN6T+#?&Qh;%
z7Nzq-a$*Sf4W<yS^m+Ln14k4amd@APA<A|4c8rVlx1EHO(s9fZ-<W-(k+ajFDQ9Xr
zh=V*ed~_9>DLGv~ZtwP@n82yD`c$E%t)18NHfb10QV~A$yS`hNN6J&MZ=eAAX_*M1
zO<VpX^zcJqx7B`o4q7|;h1beAYUwom=PV)57I6fZZ^tAu>016TUbgU~{i{y;0gaB%
zdQ_spN$i}a#HdolxS><i9|?4;Nj#H*y%C0J^2#raN(^|=0fgZPBt%u)y;6$(+6P8Q
zGR;h!+8-x{ApTtTIMC6X*%iG)Cn6pp{@{GfV`pjJ=j)`!m*tezE&m_67`xqQ0#Mh6
znS|G+Do6l$ez~8|{1i<Q!t$pJDIEZ-?c^K9(J1o(*w1cj_;;+>CDN6=#-IB=A2*f9
zVVAS#siwx4=tnJ+St4+<09N|aaqnb_&YRj82Yj?DVR9_2AHOrmGrO)11s@)&@uU)N
zrXX_GYe0OFsmZny`2mA++suvsIt?j^#H6cscEPSTm_f}49nj&|0t&y`U!h5H6=rx}
z?__VvQIK-~Dxq8R*d{60eBFnZ9eur$+j!32VNE&qKS-|sw^J`5hb82Ew=u+Q`DI4h
zt3y~@L*#%+Q@7HW+<?R0yF(N&Io_M=sUr#7_E57OGvjNRyZw*1U-3F`w}z9xzALOk
z$<IicLmxZa@+7tohl!;TZ>sB+P9jpGCPyu$;J>CFcH(34XhI1hj!`GtHNNXc(<OF{
zqm*n}Y_3OOH<52C?qgnlL=!sb4Q(IhzN6)Gy_#y{%uQ9(AYn%R!>C0FqM?qe24imj
znq7ofFu2!VJPO9@=Gtj=>|fuX?w9L=G6)M#r$+6O=|B-H%BfI=>B`__&7kEX2C3TA
zt<s95r`h!|t?;paS+#_EQA}=SNcm1W2-e^G0*&gOWy%u=Q<OZ^<0RHeul*VhEKRzR
zNO$0A^3!>wo{rMFg{nAIt!0@S6}ud&U26gZ*}ty1C}D5!h7vG`@?$tG!<#8V;5~i1
zqlYKBP`C9mK?p{&ZgpMw+V<2`zb0CE^m4T?aLozcKQ&P&At}jbmWlRS6%P)@?){Po
z0h`oXA6Y9<ham*PuoOt`H`uB=?8c}seu)?H=b#j8fQh9O;a6e<Dt7`bYW2^xdwN4s
zq<z-|7@&3_%W(f7Qs&8j?;igfb@X%&OjcE)b<sk@D91{PnK{G$8&DgIkz=8@H2%>9
z#CSa(k)~hyW4cpitw`d;m3-tg9?u^2nBS|$)J$m9(!#Iac0aovt8m33zljosrdNeM
zeoT4ySt#p`N&O;P<8>sS5k*T0J~nf3wN<9zf=5RP7TvGNzVa;VeDC_KOsEo}|NWH-
z2T=YhNyY%>)5kE}Q+Tou@|>2?vfQ0M*6mxuAA=F}s^xyDGWyv)R_&QMOsg<$^%8kS
z$#D4)N3PwP8y{fawXme3;PcG{ij&pCiey((idlKAzm)5U;Ht4O`>pFK831H?f@z*j
zSk2Dxk)CG~VRn`5`7z26)+*)8!hBu;PO+kGbrOl<^t#fZPt=R}5p#=18c7Dj<lM2Z
z<ae5g4`xtT-tK>-qL#7muesnwwyoPFoR1D-O!h#(a4SiU1-+u3#7$-F^6Y^(%4?3c
zH#J`fYjPhKJR5Pi$m2taH#XwD>Ik@HNcCQ^VZ-!_$wnIq7wm83JBa4b*ld?~!<SUg
zl<u$Y#>VchrjOtUi?}EH^vQiD0Qi7W#y|64Ln&|qeiDP+AGZrfl8*X1^Oe3B$xu__
zzt#p*(plW$!#OCNzL%NhEbyJ~TJ;bs_N`$;#Oo!)&YG{P%B6%0S&vONKgme=cv(F8
zMco#1F0RWox0HsY`Qlzs<6k%J4>vqIg+2N3vVt9AnA2e>tGQ~9OLtBis$sACYs=9n
z4`?~T-Olo`7tJGK>y!ucV-?}{1*#)2FbE!2pT8kq5o-2+N+tm>t<T?=)}t~(H~#t%
zX|;N{XXN6E!`vO8Ee9Wk@IWy(?5c2@h+iSrmo1&9hD|)@|7QVAN@aRCaE7g=_`p`Z
z*T1zB`k*tACo|dJYn&av75@iKNaU(jGI6u{*u81vjxzFHmP+xBcD*Hx$>`K*zC+uC
zHtQ?-4lUM&<k~l2lJdX)d}yI%QVChz>SXM_UpquGpm?43F!5h1)6w_K!?wi;<Gu<W
z`T|Y__>MP2)@`Tjr+r9r;d{(0@cnuv`oTtaSMw%ds7eA)g54-E^RvowEaS&Xqn8)W
z^6bX@iTS57p|%bP2>bTM7~}2^9?zb{nhEXS?j_RUzPF&Mvx;EofP#OK>1oy-uV&WF
zD@xiUSBMFA|74_|+_V`4uK6etM0kIqih@T)`&m7Bb*_{I)La0&M|d5x-{0H$c%^E)
z+NDxW=GR>D5t4IZTPhZxQ(R_n!-2UTK9A?4<NBo~Vz%*FJMK^-d{6d7y2UT@$~MEN
zS>8&*9R!zLubJ9nEU82@BpbKunGFBokeZNManUf9q}Gwpna~pE|NZK|K*;&@LjsEu
z2<SJ1DKFry-ObVr`q8=I(3rIQCb3!1Di|=(*8zJ{OF^=nFn;kPS|h8q?q`HwOHz2^
z)ArABl?IGB3pK18L1YKqms5}+&)&%EESXdT>EsX5!gOgVWc8Nk^xHZ#-~^=&-u>0-
z?YvssX*ALPy;sU*hJksZ&Jm<uW-Vqpnt%LZ>2?upb?vfi>$V3A8F%Y=L8HwCLpXK5
z@tto1%Oqbn=78)Q1VfD2qlR^@EA??H>A#7>Z#<~xW$%SrrDw~;g@js|V;D5-V$Wr(
zmh<mgjFTC9EgB0fP|2Sl!qD^N<XH7@g29we=J+ll;w;`Y<$QbJ>v*R0JX?zk9+x4Q
z>^erOhDKX9eXnPwHBDV1!!kYWn|UTjqq;#t$iH^ApUAUe_hvZO_R9~s_ZevZj7A8d
zV+>x*n+m8m&qUU*)@3PDq2Bp7oN(~L%JPl)Ygr*Mos%-dW6)mgZ-d-2r~ayw0CfJL
z#NN`tPF-yK%vTvgPji8wyElZq8hhGOFcf>T-prHfxZQ5H8KBbLA~%=OD#(ggdwsCj
zx)?br=WH+#v-ZOr@Z@XX)RL$a<v6Wat{(N3);F;@Sy({bC7Tkme)cWZAKiE7hR^uc
z>vwzZcgW{%q|o_1ih{2Xs*<MH20v#?u`*vop;->F{Vl15n_!K4VoJ3PpT%fmoHRmf
zteU=NHYFl8_0cGD_73QP022A7#OBi2RVuAYikp^vz!AM;y4AVA$z81-iZmX=yk7~y
zI+mc}0njK;VlVd#SgB-ndaU1gOyTJn+|^zuK6uafdO1du$JZbLME~4Gh|?1BDc+rV
zk<%LWNKGOB1>cRSf<208IAf;Yu}>C--<}Y(N-^P--ofnNv=1<%`v`^f$+y8G@4*Ip
z^u}2?Ch`62w;m--fJMu6wdsqL29_)y80FY9a8!9R)$ifz2tJXg!h*f*cb;C5{XR}Q
z0dh_JPd6>+86=Z)Tes)eX-T=LO+52V7!~Z#Oj)l3I6coo@Dg?Rl(S-g;XckbIbxun
z>bl$h*3Wv1#-R*M3j%JT)2+l#^BiQ?(R6dQ(Z)g@7#f)~FqPFd^XBAi=R0CCoMS->
z7eg)RrH;ETr7-(4TZZ4$1LwetnDqQH+U^Cj#S;ZZxn4Uy*UFdP0tK_*ng7>3>I?at
zk1mE4DlOj&(cGIH%t->@<AI)5zO-#2g2sIeBkATgEeP2TSFe>UJho%7Sla2H$knC7
zd{=O<hgGD;eBhQt+?^S{;c$G^!L@p0GcHOkouZ8Y_($6Nb_0po^}>caUO75N>Hg`S
zGtE}4CfND+CoZdYfhX&}(>@|_$O;rbe25nPL-#?~7Dl&u=6bbQXpe&c&ncA`M`W0O
zbU1>L;8%m~O|7vMjAq@tq9^F~<DZ=Uz4VzKMt*N6|9Ot{zaeKLq3CZtU`y0=Qd#{f
zNi53iXh(aEPSoTSL0Z%Q5thO3J0`k#vhzyidOszJ8QsQBKpIPDn3C-2+S$;#*G+f*
z$N6IjLa)il<^h5=NOtho^Qz6Z)=MNLmX?lVh5;n+=97&uQvUOB6$An$th_)e1@2>a
zL9dQ2o;NEv&XfcdGNLW?viTna8%}ofD;IOcQ1plK>6uzsp2s%bd0|>~g+Lm>OGvZg
zQGRA6PK%4$o<?;ZRDQY%q`@}BTR&}x4hiSK!T&2v0H+e@oT#&`^x9`1P=jcL*;v0(
zjC64t0<qmEt7X5jjKj{>Yuw@x^L10$jjQ?gb{t@W!A8GGU}rPQBnPBcUn~P-9E?B~
zpG?n*Hq{~P&J>R5*pRR=ANw(|YlWy%1%E%N*Rcyh_!!i*9u`{?T=4ub8Khbo^ISIc
zyUX?R#}o1nh|@j=1rg9Y^$o^*_QQ&PLwS0+n&_Fgp}b&>0XwuBDHPZk=dM<W`70~n
zwfrAWOJwx?&$oOR@-PQZ8FEw_xo3h-Qh70$k^fTjLSgp1L2u+y!k~W)5<?Y+6eqoU
zlI0x$ikxdmSl8j3C=6(}D8?+sR-4C5CI(Gx(;VsJHR6S1W`QErEQ^F0ODXrz=eo}y
zDS`ej@cx*K9&h&MRPy-t&RkEJfsvu!GK{!@fuugmz3o)29`hhj6lqMzT7^kwKt>xq
zU&(gKa&{C$t;KDaRcz~>Wxh*2v22Mr*`@x|!&<Fj#|x|`2Z=I|p`ZN(Lsiy#_8p~e
zds8Po8I_P^^uK4uZ&n`%HNxoCCyO8;YbwfJJ%3)|T1-2mdn6UW+z(3>O+RWb$EQiF
zp!v{3f7_%jKlVZ!ApAX|B;=7B<%CtW7b;%Qse%b-T_R}ZLg@{zk(LIH>gCArRy_I6
zzNv5aicPCc9eKB(eE7{8O(l68J5TH|BCv&x1po=Y(UhfNPL-Dti~ad{Me=4llfwP6
z7KOP*zY~QkX_{TYP#C);zL-Bly=AU^;uK8T`F#dTNqD*Cb0t|(g<8~QvYY^jb}5N#
zte9HlEagWJ+l=60HPsBp(m!>;Z{&Z0n-(}AN(KWap_LA7Lxb;z)|NYb%Dh=d4*&Sw
zz2bg|Qh-*|{i@m}b{MOeX2aCb5u($4_Mq+1>#WF5L%W7^Lmdl+CScS9dU()>HL!nI
z+r5E=T(9|s0gk;@aG0DZ3g`<}{ZiMVW?p|BZrCe=1mM`|fb<P6RNeyoY;ia2p)WCe
zRbe3X&2W^MD1~eLMCaYl1FK}A)i7~V1tF8SQzg57C+l~yFRQe&s$?G1?4`k;H7(OH
zEzMX9`_F%$kVlkU2g+JQ+W&5kUboqPFGr%9W#eK#U(RYNDA?tCTs)p5UkqUV>oL#2
z_KkC=H4jF0fM%Dr`$+gxo}PN3C_{Zo4Qz|3sC;pLk_y>tr_-EOhGO;jj47rurkbwe
za#wlWSbl@}pYLH{nw-%>-NYm7iX_?rW>sC{5vMRDkCXp4M^q<BdQZdq=^Uo;`MCZ5
z^iQg8f6gqEPFjINwtDHrmi*BPllxf+*K_6w#5ZZ((j7YL^AG9o|0*mH13GGFHCZfE
z7|Hh}a%3r__g9t+6PB}Klqp*9X*i%k74H~OQ1xzS3si<Kg<CN&;yu;F99sHcS6O$V
zGhDJHP)g=)4{J?=)!&U*7{ar-RA@0q?7regx^>5q>c2l5vf@ASu2Q{YPV=Zk09>jM
zWE^?{(|pG+ag@<8T&o9x72!C0v+#n?OM#!eOVI1z1XfW2XCB$}eWp}n-^=+({Ldc`
z7jTM6-l}m5i?ryuuPY6~oO88Uj0yjJGEqaYsQfGf5+gIE^X=9kGV*A*(H)galQXQT
zwdK}?$n(CG&<H2u3{}rFxq;<6TAWzD!u!M7nH78MLn1TU>gKo)m$VeMW{EDzv!(~A
z{<hf6m1Ac%k?cGW4^`9}Pe!=b_if8{siz9DM)NP4Tit#Vo7?ZZDV1mZJ&?OdoYh%A
z`=WR~Y2>K>rt9h>TU5@R3s;iE!Ob<teAJ-6^}0@%BY<(S=_Qb*@?`3@?c+JQ*|cKN
z#ur4(jx1@=8hkF&vz@R0^RLBH)H&=Z|D^wX^`RL$D=X_{%Qr#dZ#f@Byt`oVc=RN2
zS;}8c6?~TepjceSSHp0~ZKka5lB^aV*A*Hs6Rt1NgcPn)+%uQmUTCJ`XE%+~bjoR+
zG3U%1w>KF7KH6}x4mF&NnW^d&p66rV@B4jtG8b@wuG<&r@c9w;*K~cEGNG|pai8>?
zDJ20_OKw8#_c6(|2qU^lQh}QE-1Br}rbI7L(?g=AKh#@^A>+%lE^hcDNkggCNyHlH
zx!E<MXmI}Rtom_%y3x4aFX1Vn*#B{vs^xg{5H8wH;I|sfh{?PWTz&Ov_0QF!rFvr(
zVTNnbE4%g#r1iQ*;`?RI%R$OwsZZ{KOVs4C+8=*PU=6<f$WXK7Of{+?{!#;l)FN;M
zm-p?Rib%`jee+b#2)SwT^a}yJ6|keC_pxZu$x>o`X5=s@{8gQg0KoG-rkV39kD-A4
z#Yr_<TX5LxPyH~I9bw&mxt`&+wDYUe2ZV|N{@rMY{l{{y`h7iB#tv&Ejc60We?F!B
zzr37{J~GRlDY{DWL9IGr9((R-#gwBGH1%ydMO;1^=;fwl;8wn4DsW!e?M>skr4u{1
znkSBpIgDdBw30K~l%RRJ>;jJ>IU1n`A`H;H^!2M$K-RBIXN1Q+qJ769$RdL}Ek%Y%
zq9-mz3Rcne{k+LStY3^|HRMO<+<5W0oq0MRydR%KGb-k|2a#h>;(Im?19~RclR^6T
z+w|4*k%_uhZ-e$i=SlO^4161&Tkowrb=j*Ozq0_vv`?svaM~#!u+bXy<;i%K=#f2p
zl4Chw*?t@r5tOF1hb81-D?s?DDxIt-kxn+v;_iZBCn0~^jYj%#CS`XGqvv(lsiqe|
z(`Nh~S7{L2Ckm!~>3F(0u~7M3C|%@b3oy+#pCE?Rnm<C(tf0qt`Y~24e$33$dw#@g
zcz$}gTj@XE7#B}vaVQYwIbVQ5ZLLZf1~hXHAC5}TDWT_2?eoRdmAo(pD@0Z{pO3Vy
z13W#G<|b!$?1Q!bH3RmBeTxpI(R-e1ZLu@UTnS}?(YowbuEp5C4e5=TC#fkD9_c7b
zAh)?*Hc9)Icf}Ia@c#1X&_9^wjPYisJmPKDd!FM3*NgCljKRRccoee%S@ShFa8`t%
z)Wi>Om$yHFrEe(&uBTz232&y9J}sBUSYu4LHO+?Ck0aszEEp8q2H<R@T>Or;s<GIW
zm;J!>D%ItE%}<cJ08oWaj!)E10uhP0_4(qUMQOL|QQU8S!UIxru%)Rfr~BCuaHCn8
z0@X)mtt_Q~55zI+?wLrHrRp@`fmKoc%9VOBp2VUX$PhR^M#ALLWtI)Y#ksDRij?_w
zz*RiC<qd<p*Y+!TG+hRCj@3=3bq0`Ee1Hwi#%WQIca&PZc{xkJzGPnt$oIGRCl<5&
zVa&d<sWBtlx7*#<Dg&fLH}f>;nzkRPF;%A!%qwY!TFX^m@tjDeH6O8cn@xwJNM{_V
zJRba@t=ARsOPMQtB%+8#@Osv1Mqc1aIMz)ad6E2z+bj-8Dgm30%eFZsPG4bGD4Tce
zVyNyb78u)-)?$WY1rcv_^haA+ET=55$~-E@dm#H&?U%dMGP#WoNB{Hw3iN(8eblxX
z5yhUX+p{+X)%(46J)E6Zb<(isaR4J~i!PFX!e_d;Hi#EY41+bQEm)E<@!w09z0wFd
zymjMnO7oIbUJ&(w|1lWVoL!gS+_`<eIjz<z>2U1J{Usk$q=bC+)CWe!N6q%nKwpJx
zEm|MxzC0G1Z*JoIKF}h(eeH75vHx(p;=Kt{k#w=~_GeXzLwdt2v&_|WkuJX)a|I$t
zW_-)T%5=vhX0RUbLPuA@!`a&YmP)5t6@DD`+6qLOb*4MNb>sX$DGyYwQ3FwaW*Zk3
zrX5+Oha1OKL9{^h;dbUw&cM`QA_(-oA<s6;n<*7_G4EXuB8q3v!9XFzSEzj(7nb*$
zd<iQc{y{=dJs}{G9BUJTprLA3p?UtSyOCmkRfvNagSRQ^r9YWxR-nP|w(S<h{ha^d
zheEbN@cGy4BQa1D1~OW_4cR_r9ZAJA97&lQ!+NU*=XChicO3so7mf6wt9r~VY^i7K
zX20BWXQBr?3Ez3=Cu##nxD%fPi~f^)vnC2<nj9(Rvff3lxd#|`GOxHwX}21(uaGrD
zh%h8Fz1k4Cmh{V5JWLT?q|#ovOZ`&_9fj{hz}|AR`kXdbNm+Tr<|5!yK_JKf*@xSK
zB}zE&Rj{FKA`@zZLvPa7gWKI#lIApNoVOzmSFP;0ZsG#C@q<ug!~Tt5&kGKd7fz1b
zJpuYtiwEu-`B6_{;N7K97@!cwNJ(@5!M9x*h&*n2)wQLq+4PRT;pU_pcga~#y}#kJ
z9~l;nYn*usH&wvq*UA_PWN?2d9JQGIF0Rlh_PlMdzOkv-yyD;w6*PpSN&dt)I5AkN
z$wN-zo9V?g*q3LV{C74>z6qNqD`Ymi5+dLhBiE$EN9p-kt9b@bZR@{yDBc~|{5Hq%
zGiK7e>^e(xWu`_>Z5to31Cn>={qjo?ELUZ>-=dClt(N#pF^@zFs-?Zf{{3gepdB*>
z4<51&3MGWi-g(5Ti~&{2J1-9N-pr8s+F}g*ZL!ewEv|J4h~MRsTjZ#J4+DFYzdy@W
zfM5qEmvK(;_oI9xW|%tbp3b<8XX_(e0{B8J+5S%<^>H<wzkZr*nOF^`ZX<@wG7-nN
z^MA&5|F>v8iX2Mg&Z85$*-l7CI*Eo!R8Htn!9Y-4ml-XIoK%edBQyygq3wFWT{p=X
z|7NFo`B$`*_|KJ_CR=Jwy5A+N;jb6F`~ngTzYi6;sIiqObpL?EEd`vqX1noeQ^Io0
z4PE*&5uGOP@s>syu9Y3dBS+nSebX{HI<0cu(O0KntB~jI;(Tu6)$#TEcPz=5r3UUb
z`4ljXD@&1sGf*p`?Yf-RqFot@?tMPuV?Ju6VOTj_<k}%jr|hAhH6V7cP(W}&feI?Y
zp|24|*wU0Y0f8jK(8iyHsmU_jtX$aS*fh|HlRTo+xH(Zt(;nSp3Z08=Wz3?XZJPzw
zUI0%U2kqx7C@>`cRh5fzMijp*pa1=R$iUmr<5@*O>`v6v{0wCUVrZu+Oom^r8X7mX
zCh6|MCNN*agTz!2)EFYrEg?|<z+<=m;qZPLV4}<U&Qec0;W@WPOLfLjh#;af1UoRR
zUhOYy&A#@L2`>5VPTD_z3vgoGIF3hP%ofwAw9>&?@t6(W{2pdPw7>JWoJq6S9>iJp
z0rYv=>+>a#JoMs+5UXq$>#Q97$bTKHpU6YKyVucz&viX6+b=(68;Dp}6rLK}*0vmS
z+Cvom$5|T#U@<9wwRuz$y3CUGN9W?VKTIS{4zchGX8J5UuM!tD6snlJJU>vLv@Ha0
zf95-#f_NLz&eX5E#(Pgkl($Wg&Pl)H)9}5x*^J=ceWSEi?%@U_op(kid{s;Um#Dqb
zl_RmfKld|;kQ{nNPUy05OWMQ7I0maUA0Kt7GE&IueYHsliZpc!=Vq?B{=JjXckV8q
zT64eIFxN}(p0KbfZ`a$*pxtQ(d-1R}UF`-Sn#_XzT!=8krbd`2^7qj;z6ZAS<fOQ?
zm&fnn38o=EMaETk{VT1W9F=xlPY>+u*A}~qiBG!OV!hrk>YXgJ?N~TzZA)IC5Z-CG
zHoXsVL@7{wxyO9dR`BWmy1}q-d2pP~y*)veCz7Yvzfg%DV7OO`<Nq|+W}`7`9{6@E
zxJEvi+7c~S{A}_R>7`YIj-_#%%9;I9d=|#68K<9q?K1HKn6IpsT+Q9bZ_Ud!Y<qD2
zBw2EpepkEvn%+b|`>EZg3zkUNeY}*~v3E?0E&A!~VBV?!%~mM7h_HZbjh0MK)IeLv
zG&sFbpz;k^acFaZ59oC*_5O!{ow8dZYY|-hX7dVop_}D`_$pPv25E*MKyPFNINv~w
zKCRa2Q!Nr%u7!7VD&A~&KG=Fb+HTEUr{}`!p32p=R3{+9v8QoHgsf8hb7uQozanld
zCZ!9Q1(%aJ=&Q5PF)SSRjFd;0dOOIC@2|v&U4Hx=PV8;j!p+B|59}eQ=)jphG4dmx
zElxz(sK6d7rJKLN7pjI8DexHOk#)aNxZ%&>pD#+QG(ZD#g6jtf#y2P2hwrj6D}r=7
z&hz0I{zD2G1$Ahc9)jemkVZ;4QFWE6N16G@UIyI*m7%8N&ICrfCCHm#Y3gng7yVx-
z;3zcgpi>j6rx>+dnk#tnEvJ~{=#Nkc>>gs~i$7rrK{ZW#;B)3Il<-iAQ)fCi4v#fn
zv25GPY6Ul%W##7<NszRzg5ZrLrHxl!c3bs0K(lOS9Hs9zn)F$zxU|9}=a8v&Dx`h|
z@ft>|-qdEg@Ma7J=o})ntj3=+KtD`?#;@NwK*Q%q(>Dm)oy>kXXv^u@mOGSuoH<*5
z{TcEKP4`<*><PPQGXBHwSlmm7pDd%sh%V=k3?{e#+Vzo(zhck?+3^Xo52xZAe!cN;
zp}K>#;;=$^=>1xFd;@#>DSX&RZ9wWrmv4(m)gIWFG}<46%C8>o&{@b8jaZ51Ws&@0
zVPfTdo7VbjHT;!5^M`3SGPxWxjfnF}IbHr!fAPlnZ|f04qsgiSB2#WVKgDXyj5}Tj
z;m3EZqN87is*6C#cl|x?A8z)Ug;dKfKJO~F+gV=God&8k`sc=iDy4!U(yUA1TAiH2
zC~)Zwi(L+gav>$vKVDw+55bBTCI1Kuh2c3}s)gJ<ab&fq8fQN{5Zrb4-!3NZH7Mm;
z&3s2mAMVSY^zy|}G;~{IFeejQySJ*0buaeo(U5R+yt1h;x%X&eF<+-bmjkj1Fe_GS
ztI5r^E|zP1IDc3s&Bw87=q%DEQxT7Q%>?3*Lr<ra5Ht-EBt=J`&qwJqLxOTcyCR=4
z8qOqXbln(Tbku7K3_?EE%KCPrF{{e%R+jJLIwGuA=f6@db|S0M6Uw)nn{NN}oy=Mk
zG{80H+w$#{nz_(sQ!n!5>I-iIUdv9GS-Kqm8SxA!+)|z0h6J=E>X;~wVV4#?XD!RF
zYulZZyliQlXJWwl*LvYM6%<DQvU@d>Vebx_eq4$EY?)ZgT@eTQ5`AO(v4n6Kpa$K?
zvCANuHy4R8H}(62UD%&+8~4=)*17wG6JbXg_}UNCpU2i#!(8D6)rf8>ENCu%eNf1|
z|0~UD+~zk!pYtQ(8)P!ac$h6V^=25MH&`iB@sna2a>||<2@GF>AZRDVtTf{RIU5?m
ztGM=3Ep{zc+gT!(juP!epiqzLS{|Mlu`mPZv7XaDzV#=$cCCEW3L^=Hg{+g|1U6?y
zzJm&bp1$&>P)pjHt*`b$Y*OHCii(5<BiYiWrE7!Lwz=Hn@$PV7>Qs!49F%j|y1bh<
zqOVWM*#JmamkJXZ3tP<r!r+ra9X_QsIsc{&dAF0!-n2$JCXE@-$E&xM+Abet2RuUB
z5}i_bltu^*gQY+UXi&;coTkwJ&!7V{QN*U3z5p#$z9whqwNz{0SY77>T1Dvb2V6cA
z=%3x|YL5d3E*tp8_14xMT>+v^8UySVhLMg113wpLHB(c#ZnhHZIyf4^yx*#cE!pbg
z3w>Ej)j`^lBa(_31|<3#Oc0=)jcRG5t-wCJ0LSkyj;Gp{DJpZD^JPoZ$ndLw%@Bm4
z2Qa}IgI3lwn7hxQ$fiT^3mtw4OfI_(2F8MpsP?<B=V@%_jb+1$64TUMj0#CT=X~#v
z=-r%aVvM7%dnQ-9GsKTp+8g&^<T}d5Ft7QYnDsfcXYho&)OoH&;Mml8zQ(8_TnXG9
zHN%R{T+OnyuM)c!yk}gs=rvAE=DA6Smud89*5je(J-J?8e}5?vTOtY3JB91UvMhV=
zr|q)-5)Fh5&{qnQTwreXx`5Au<JVE%?s|GQB>QmU$F<h0Rh;me+~sg_wn#*~@p`e!
zqZw@|;r3)qNHEy1G<f>J31)AK6~x_Aro3a#(yV>hb(2ZnP^2e=D9F>eAaYt+bS%_P
zV}<LmX187@ASLRi5N>-uGPvAZa!f|NHhuNZOu3{C3hWpl0&Y8kj40<r%wQO1<R0C`
z-Ta*-b<T{0t+UqNG$`d=(eiJ$5F5~wr0lFKlvUcsb?3%#Zz5A>c+3=02P(H9WCI7j
z&7vif`>CkG7$I<^ZbWdEa354W7y>z7Ayy{Yzm8gOShkPCaI9ca^Z95ALRuGOXSq1@
z1J*~C*k1aUM=fBmYj%mM!YW;^F@yJt&n!<-(bgiJdptCZnMtU!r;SUuYh_#fkh$%r
zs^!$(7fz=YZuFzfwxC8Gc;b`0oX30n%aIr=L15>jsc;8m=+8fmKy>=NUC83`OS~kT
z+bt7^<56>^s)s@Pwy|wB)GAr`yRSFSpB7FZud#^2nak}Go!ermzx`^im(nprQEyaS
z{Z`U4H1zO)pXn{C4`?7X-+JU@zRpwIUT!YVmW+SEV9Hlp{2lqUL&-HNsKsa+{ek;R
zJvL=2-ztN^&evPIsYEds-*Z~je6kmmSl_H^P|jzxNU0PfkNxkblyGaCls95MjC&%n
z<Q%#%^stDz$xwM9-aPgPW?5-S$E}HTMn++8d!L_7yHL2)ry!6Ato*!tQD6$dQr|Tn
z!hMW`;IEB~Gvc$oX<u@o*Dc<wr!xF2==$G(M&82JCSpKKr8G)`+yC)Gq#GQ8<GEo<
zj9%N2I?DhWSm`*J&97JhcbrYPVmXJtGbQ0BuM5X&{L2bFHn`CmskkNG=eE{4o{QdB
zr)4?v4!YJib(Vf^oe#pk30EYsQVnSa^<NDkv2vR6Lxm(jX5Y;*o21b0$%8}vaUQp@
z@=&)`8Z=I!5kz5esgV&I%gB|C#YTX&K=OJ8r~wBx624ctK1A{{ar?KJ#p`+GcW;=s
zZZ`bqQT;3Lllf85?;R^E>2<bTWo+#AoN39KEZr`Ka_AYSs+-DLiB{f|Uvgcl4ytiy
zs#R(kMeErrCg-~I`$y+Px1~DE*jVL3SO;6LX(aHP#dEKO;OW&+IE<A#hcBW4j0V>l
z{!)>Ot?&b%QwhylNtv_7>x1LP+3;Mcuqh+W$B-IX8k_q=t27&5EU(Ah<=#dJbz?PS
zqU31RKfd<A`z(H_?C~F@8Cn(kJ~}9zZv@rfbPci^wnTM?B<eYR)W!xIZxQ`&xOQIV
zn#43v?ID3hcs7{jpUc9zxL(@9CE%jwW*(}Itrk~qo*}K@ZA|u;gp19M9Y%O;>c0OH
zH=RZ7a!!XIS9*#-lQdQJL$yRtF9$NtS$mL6uu+nZ1~w%aj$t2*akT2NKkq3M8Ff@B
z=04DR39>L1v&p9Z>D-JI*QEizxgzcRxdd0sUll)XpKI+$(;f(JW{UBy|Ia3X4d<gI
zE3t&bt{aiav~{e-&so)~Svz%{$Z1Wj#aL(DYJGcV;|6qCd6{3K%0Ww1GIs_BxU>1U
z=N?64yO4|Bzrq8%iIJ_LN3#!E@Gx)<HaF|29N+2;q{Rv}Cef4R?{By39v|au%*EPc
z1*Pw=Q2PfdZqZ}-kj{RJF#rwu|NeDH36ElMAFxpV8LQv)fgfg}`J7ba!Dn+=YXvka
zSVB+gtZ>wJWv6x(!XUC}Mmwd-)sYt7l%4P$3oTS0j#IM})o=d&`N@!x)M{pI=g~*P
zXI)p)mSR;AoI$6Fuik|8&eF*>UYU)Ki;u_WhY_!Cnw>a6@ktMsCZ7;o524g3Y)5%^
zeoTo%CI6PsN3}U>qy|}PomrwzTL#Jq)@@&v{K`#~%hHI9hmAbwB{hc<27WZq=lG#1
zYmM@KRSSf?B6d`^!B2)>fw-{Z{`X0R=P`E1DIZtQw}$5Hr_5MepJHb!-L49r!ZF#|
ze#1f8uAv0v$<SdTSKb`fvn2ki`&C@B0Nyq`kN4{qmH0N#kGLFA%Byr&n~@$3>u`8L
zEgcxfG!aSmR%<}*+jm>AW*JUavMn<O<88!?#ay&<Jf~wgfX!R~tBbhIyE!aPl+`%c
z=Bj3(Wk>tpdau7aFsdbvwNCg_fLG<?b&@_hy^s^Sg{IK8rPHWqIw1d8f)m*x^n3#*
z>kCY8XLU?hjtQp=V-Z09aC_XAp4YAIjE3yNVq&XLx`&}&SD98>pq^N*p?}u14ruks
zpqI)zbU#MG=ohWBhz&n1y(!B8NIO*SZ3aDE{|KU<^_ikFB-|}uiw}UmO$)PH642!6
z6D!9FN9?oexmcu~cqmz;gP@c*$6^#L%7ahktTU+jKG{a|Zx+gC%X!*dEq$7F7%m*p
zy*@dTyTh(J?qNR&x}VGRV8B*P|KBZe=GXBvC`CSB)|D%wSgl&FaC%|popq|y;h=7&
zWbx7xo(zb)7?%d&YleplH``wp>jo~7@EFhDQ)2{m0nHXTXgvwBs9kLLORc2SzVsMI
zYQ&@*kPZ&5dt)Lu1MJDeo&N^zpn0igpfH)Psv6zfiSp^4aZ-o<KgyOF&9jde`MEpx
z`pEx&4489r02N`oS-ZZQFEE%6DTR%_cmCRTT6`K@zj9WTduSZh9sctpBQQ?4`e=<H
zH?q?pAQSx=cUUb2{rN+%c3qX8P&KCc-AO;?Lf0yQYXP7(p5RRfVl18gcEYMc_%hlu
z!uD|FCR;7BBXW!RYzN1E*rV-m5NX}B$LqF2wXq-m%_f(4W$6u$rJ7qo$nVNEHr2j2
zbhbwjJwm<zlOf<4{4{^4YSQz}5iOB?{Hb|Jv0bGYy4IWsddLiT-!^;Uwyw+}o$y3g
zVNKnQr#i1O8y@WNo^B}ly&1QIUey(=jx0Ghp*O~tDfRoPsaq}Yp}Ch!_N|-#f8Pu(
zGjuv%D)2#HwYI9x1Ff5LfVSoqQ?bX!n-p1esB-BHLs~CrDiW8}cM|0~L+zZ&be$VL
z3#-pY)J}KPensETTL9AWg#A^jj4E_h;q({cB0Zv`M9AC6A!avdk9i;u05cd$I8=wv
z1DJ<~lKItE>3Rb6=LOTnUB<K|`RGDJMN)tYX8Uw2+wX5t%Hq|ACQiBwSli(AYiV;U
zSJj-K6Nj?qEMSxQ_rsKdrudE>y>+{tIwc&0#BOSa_9ITZqyZnWo+Is4E(_F(?JnLK
zD_dx=F}IB|f(WD=7L)KR;YT1-Fhm>y)h;2r|97JNLuww6COkj*=F=tnuM4t|L8Ir}
zWOI(!BPV6Lxd@Y<PgHZ~bFCP*h&mla3iHo`bBAy>Un#xrzKd3V->wUlSqq_AlJq9K
zC28Mhs<!JGY<<6=ZCx1Utu{zEbe7>v8WeBeI-)N4_p=myg5`DQRO(3WfxD(!5s1`4
z)WGn0&Gjk{JBrg%5~{pLa$vx-QnL_*;qu6&{!u9w6}{$wP(4(SSrn0N$kUHO2B}&s
zrwZ2IY?)@(>aTxr)30j2Ft4*l>l=3i!hsb;nl&71apKPqIiY#m0V>($?NF_HXaQI5
z(}#0aZATTWOUs*!kH^T{p8x!6eM&ICjCTc(3DwirqtowAOED1<z&4&q;-T^l`&_S_
z3#ZB9Bc^(DbMmF1?$aI>*DvF)I{ukfE9Jg5di)u^oq4T4$6|b%Vlu~dDs#|q@ImLa
z>~Q$J;<$EueYLp6SPE8Zzm7`qlo4e2;rf6)#f=O%iuQVn4<2hiH?y#%7jN$bZL3g2
z_}iWZWpLy~g9h&Y?}{}_6Yl#y?%Qa@rqRAquIixw%t~7A@UlhIqt4}<o!cJ-V7~7Q
zL|H%AR;Bs;;^{%@euXo6AcrCi2jgoA`B<~yEu>7<H>8%!E|uSp`&t{%s0$eLwBoWY
zFn~L->e-W@&FLaP=ELDmYd~wRiHzTdh0>Vsg=k|{%-FPAv(2hUS(-2AJ@Q~7MHS1x
zo38-bA6{P7gd%i`qxmLDmzj;?YwC5wkzveEKMlo$T)&X*w%}t#lbxk>WQhv4Jk~kg
zuK;m<tyvFQTwJcc@c=}G*zxZ&3UJb}VW<&*)f*Byt+?Y{eRs#UF-JotAbRI(z1@CN
zrcD~(lIzh|V7UY+Fz7NMM#3F{$3kKZVkP6fPd7{VxTu~uUh`5}(Mgayl>R@$-ZCnv
zZfgUj6cA|%rArh9l<qF+kS=MYyHh~Aq`OPHySux)ySwk=yze>pe&c@MdmN0x82;Mq
zwf0<dJ@X0GVs4dnsZ;Dm5JnF+B$)aUh6XlM%R8(m0YS~p9cDm~@tJ*V+K|0Pd1YOB
z9k;TNrN&G6e8m+7^h8gP^n>^Y<D^m10Ax{L-Yoo))3WRZQx$_H;9l?Rlr35H@$we&
zZpBBv+z89LP!5b5?G|)f{>a&Gd@^cd*)ot+{xE|)I+S~?)lH!R7#tHKBAD0}@uX4?
zN)vnjlnKX<6+_;o#F5-#eSbdmgLLTsU*g68`#&kf=%JAQ*~yiwTgeC8lT<fdQ%(1)
zU2pZ7CrXf<PrNIK%ZRN8r)rmMqA7V36l(ET4XySHZYAO;@^Fq%x)|w=bGYnT3_@|h
zP)0;Q^JeDaLDjfv{ZsX(Yr$KZ;xS{7nl@?kp$7^IaN(21Hk$IA9Yh3c9tH!vGuQUZ
z5~pkEEbnrpF-cjk@f}FHyy3jsrMaHOJY<24;IBi`(i2=MI=x^llruBknK%<0oYsy+
zb&wZ=&{HEGnERrAj#hXnDRsD}=7%(AN<T-BW)`U98QOe!(0^e$4`!I>d)Zvj4Ni8l
z--$~mlv%5`hm0m+l8tN*_nUL1O)rf5d?G1b0Frv3uL_GFqZpfR^(mjq^QF=oBO~4>
zWV8B}4#2@(HjilXWx63x@t!cv|CfQq|Jf#T2ct6Io0Bx&bQaFb7YOV6b=6q80Fjv=
z04=z2rLj+woA+nb$!deRy5BQI=)Tw8Y3B#yhE8|iDnq%0=~XSOolI&Bg(vn?r}^o#
z^UwSBnh^JMm4@W^OX_pE`pp+7<uMioLUutli8$a5wZgRH9RgL8p%k0t@WuU^IGAm?
z@VGIPH3kSQ7q|W7>tVCyu5tUJ;i-)`Wb*gt7D@#rm`-%y$BM(qzsRMNmwyv;n8dMD
z0$<a(sl9-Z+rQe3^J)~f@B`1_lFgphQl0;r;Z7spS86{AhzJbs_8o>1eH0F-I6$?*
zcyB_o@E~w=B1B+XHxM)?VfDu~2nUC8e+lDkKuMo0#=Jir!Dy{!EAER?6e28cibm1S
zm_YlUS0Bb>;Wk0!4K~KsT&#nw*guYyg$V;-+Iw_nXxZ=`yMOjZh7WDu9l1PHuY=i3
z6<N&f$4A39?zlN`-W|gc5gxpUR!~lLby<#5IZreb$s!U}YrINA)Jtrer!Yc=8;nKQ
zXy03?hI~?=!Rjz;Ql~{2xSB+l^gbIyUh6E$eRvIC(<}_YB*9$7A2-2lvfq75U9@>o
zgzd+n9L#JvyrjQkG2w4I$iH3PxgF3^uokUbe<9wR1az5Q9iHC~`uV?E+ZZ~CC6XHT
zsQMay?D7mSHhg;oT+J3wHG}aSYz4Iqe<f~<YMYKl4BsVC1`-zS+tS-Bdi?Zgs+)09
zPLyfchl*v$mRAF8K$KX9bTv+$^_N`NG(4NV-KmpsN$1_z5QL2#+@!^#SXaEa|Ll3_
z4kPnkD)%?!)sh*Ac@n5zeI?txLORp#dLw}QPR(Fp)&pD<)2o>qTDo=4^zYB%K#mC(
z*Fy$J^<y(M@u`GyN`cM~on{k2jD5}OgNX&Qpbpw;n)svJZVrt1x9yPFn`l18mTx+V
zP92&L+`U8zSi){mY=ZohSl2$?CJx>x5pbZcn_yT#zs0u4;CTe%zVV#Yv&>*-e}Yw1
zgR#!)r(0o2Gj@^re+~LbrPN_Yv5reMu!{~VO~pfkm`Uj(xIBI%)r-et>cw9>KQ{M@
zW+xwlj+ym%zYKqW_J#9pE^^7CMl{tJXBE1|G6vQXn6oX9%Z8&L_r4|Koy(Oq`6w<E
zVkfxZTBigrhfwCSHUS)dX@98j*8`b$*W)oC&FzI1&uk?-)QkEJo<xU#5}E>Sv_J|E
zw8gjb6;L}3LrE)y=4hozM3EUyl=JypsEud8aUySq|FM4&QV;lZc1r~Zjg&|%3hJ@7
zbNw69@n0-68ea?!Cc3|V<)0S<(=-*pF%;BZ*rL$dbhhNBBH~qZsNo8Q8(iJ>ZP$(L
zbtyhMfZVyopy>1+Blm32D^^OnJ~3~`tiz6K>$L{u$aJ`es;K0{zvlnfFb%x#jjt&`
z#?B9;cbajUE-H=vI7U7{qJ}x;tG*^e6J~KyZe+jjC$W}PkPzq*sU%M0xc)BmeIs*-
z^DKA5iMY(+d}RBvwKsBh>ypT48$nZl@chN*YNc5b0aGb|i==*nW~<%O%t*b1VOO+R
znF<rhVU5s}gJ6#~#i&zkD6>%%9)r>D)NR96ef1$LqLbbOsZ4O#Vv$ns%qjOj_Dzsz
z!b$GT2F{nUYBdJa=KCeN9IK197$Ct2IgWv{v)M!9liD8+_i*Qr4*XeguL%x2Z!a<h
zS47CDe{z(uG<Xyd@qMbkQg=Z;-W&iDD~bDyv6uVxSzP{IjjDg>X9)%hkbRt5oWCzE
zR-+;^7`D0|c350l!NPyDh5~?1gjJ+$4T->S<_Bw%b{Bu5G8iP*5B4s=bKp)nVMaCf
z-QW1YHhPW<Q^n)=4;zR$*K?R!(ziBeo9ELRrK}=5w8vJCFqa2s>FX@&zEJA1s&$5P
z_f|Hu^faz_RHqI@yAvmC2q&IMqfLc?9f~D{M@A`^@rawI#+GggSDEhiG)_(+^0s|o
zO6zL_?oOiajYWC!<F$5tTtbe<B7k(T87P#m3*2-Zznl-%o{w&g{9V=mF9*9&3n=mg
z6zfm1j066aCbBsB=Byroqn##>13SpI<acu>n%hFFZumatwTI1~GxnF(bO+z+0pKF4
z2jc`>r#e!20F|TO3W;4I8TyKMBQ2i)^JE_K5J3}>h>apeQ18?dl;?uFAte|1T;L#4
z!_W73+M6=Pig4&5I)lnG8GI#LfxWcx+xsB3s==Ztl1WVVb_oxazm9idLN=J)@vj90
z*c1buL1C?p!?ul3qgbNk-~L1lj<DO_S|i_{TL87o_?%|PD>p9^ld-HF7Y)l2xj}YW
zm3xZd`VbDxkDaPzf_JW!N^~v~Aji(zmv;;i&<lYXR;7FOc{j<$8CC2O_~xL{=g5M1
z5ml;gL6CH=8?1;W%dLSd_r7si<ovt42Rys#lMx}^JLk_2Wgl(Ts{b)e@);|XSWd9&
zzrIjlTXdcC=Z|6ylp}mQ9!uMi7o}#m>IKXj&&Nt#qzLlF=3j1P><C&|IHL(Nw4U_S
z#j3^IoEsqAUIcK!8Mq_cF*+k1;nKq~1ZIwVo&><_`k+tm&6P`2%70I;IY2E=o>F1L
zCkf>iAeB<x@0!u4gxzl<hF#cD{_I+hxZ`h6K@7s`=1X<$nOS6^xBb@pnE&1qDMEw`
zOOFS0Bu5|5PxFNDo*vmaMpPpyyeTRAFCCVxZ$VBLfC{!dR(#l=#~2x-_@JYj8_uY$
zMtt{1JswPo`6vF&Qk^b3sRK%`UMkr##)`wi>}<kh;YF@o!u=nU)~oJWvf(9Q0`axx
z1al=^(ikXUz3k5%Hbz>XRY-P)h-URf!KJgh6GJRj+QKTIRJUa*4Qf_SGTa(CZ*<TX
z9lBoKa2mWwW>tGO*FIT2c1XwILcgQ6W7JKf7yEC=+HcbD^zS0JNAu@@>;vf9f3iT9
zpgo3T4$It*9SpbX?@Mt?_euF?(1Ev?+4p7{$}P&m&#8Pq{YX~bXffJd$^aJ%+nHoE
zYjc9t@FYU){NN*5Ak#6u6>NR)he}f2qa^?)uIZizs@(I^Ver?^t_;4rtw6?zoCX{v
z-;yWg(7jR#Tgg<q0`yAKG97p~^s)Ayw!d~at#FE9kd9eD3EfhOXOmGFe`|6%fAEUp
z$r45>rNO<jzTQxG7=FGo20N30Ftcd_mJj}J^YKcLA9jZa(;NPJBgX727x>Ib(W28&
zIoe&_jZQnK@NHP?a7FuNdg9h@#MWZOhjnFoFG7Yg+?gVtcCumBw4NTS;LbK@BG`&B
zhB)-4hqCfx82Ym0Nx&_t*w*?$+2WSd<u(RdVl<JBt=K^Nz7pZfxD&?_I~~ZyawEPz
zTxfoQ#ayB<dF_-DZOeLBF|=wev-w|~c1Wdu!RQ<d;vYUe)~We%6@anLIBX7ohBW|3
z>3_%svK?9TsLEsa{&Jt%>k4A3rKii27msQkC$Jt7BOR1@JMDbLfdh#4NU6ATHww-(
zwM@Z9mwX8%v}HAb2>6{dM@|n-n}Ss8JIt@w7I|cn=?f#0Yy-Y`N659RSc;KWS*&4D
z>W`S4Vr|8Qf9?HV2fB;Lq-CpIYFKh(!DQi8h<N`sqYs%Bau9h84VCbjawF<f1!tPm
z#oO>HrL+UXsiwe_fUZ?sK0xv-?BaXzS5+M6F!in56AB!LHUw_VW@nq%<7lYI)QTwe
z1g0azPGTmEp2|eMUXKq0fN}aWuZWRvUVALD9HnWqra|b|31A5P*%C<Xw+5EHmx>H^
ztFP$r-Bg>9Ym8$F6x1>wj49-AbB)v8nJ^65|G`l0nXje~?;BcWXGZ6GHxT?$Hs5fW
z<U(j=t_m7xd3pMxkW}qcWowe`BkzP}n9bKeEq}t=&x~g8RT79VdA&Sd5Z?ev2Q-sA
zV<y#4IXyA6yRXKpOnVFq4zSL7Mg$(p;$YFqJ)Fso`{_*C&%~w~D8Gedi6ep`%FCKf
zI-L;izNnR@VKUJV%LDj7vc&$yT?zf<0U(3l1Zwu_xdF$rs-m)p6-GWscrxPYY4~<P
zpuN9<({y>ZbvI5ykJO*=mZPNLTp9pk8;vzmZ7r~SQzl(VIj)`x@M9Ipbo549ZioH|
z#?q)#HKneN-o5qgyJXbuv^}?tck|4*20o69(VYS!a?6EvpRhl3FZI~3^3$lhan#bp
z{djq+ubyS4<M?0BK&h87uU|qBa!ln%;g^FfIFs!ex(lII{EK?mQym%!mJUBpWF%^h
z8r^`)Wy!bSN|sBfD+dU$Lr4B7=E*+?%qUTc%--FbV!cj3YD0YV(PYrGEAF(toynm+
zZz=EV2O%RU@M4L(9@R%tNQ^kZ?oEFDPlw0Zs4?w>61GAiu2@9|3hCI*XO6#{AK@tX
zg`|DdkrNQ-r&ns7+sI6Rn{yn+YwLtKMSFI%Ie4*}Xm>vIbVu_nRZx>&Io(*<!tkDd
zPeva9qoQ|v<PZj?{?-KnHFz>Noj@}0(|j~LWI#vCts7&xk$&?K*PX?<G*ShHtfOf?
z@hi>~{)a6uriJy^Qr#W+Ul)e?8m}7O{AcL({{jO46DzPx;J!M14@~0lnjQ_iz(86m
zS7!Hxdd(<-Vq>Yrn%1*d*jBPrB%ids`KOBrhxJ6e+6aUwMjsG|L#cE8K3?<6S7<rr
z)ReIdE<<hn>D4zEoj0!qk*0L;80os%pw70OMyiS^v!Lhch7`xq{@ezCbx_#Br7~~(
z)natGJK_4cxM6@(qWHYHQmDX-p((B`Rvu(}zFSHYxFsIlPH$C-W7J)GFGf;gcgoJg
z|1*>C?(!v=Kc9c4Mh`5iAN9sk_f6V<|Kl0c;(GE*mVQ&}&Sage9H+K3RN6b_bo4`V
zaAK|8wg8zpWpLAwJF{<KNE2$}&szZQ0ueXPvW!Nw(~(UcMpeS~!_|0u#i{C!C$cf~
z`SA9UBC`UFYRSe-GPuf0M9wGsUDX@v{UjcH#)SUwpKb)GgW0!VZ1WLu7~k;BGX`_A
zUUaR~ekbhPNwkea7msHhM{}OB+vrB<{)j2l?pV~2#i8gt`$I6nLGMz)NzSwux-#H4
zQxW~4EwExQZ8j%bYGbo5Vw>BkSeT2UlhC7;giw?z*(M6TqhrLT?py}Fu%;8O!BeGF
z5HO3EZA4tJCJXu&n{6g3n!WxyrCvq3vkZ8%T<)(NDer`@MQfDb>13`&KhEbX?Yiub
z{*a3pDr(&A0Jpv5fOH|1FK&%$;)IMseiIKAw}?#DUo?5kKUAFmfd@A;(`2q$)D{Q`
zZiuh^Jm1O$gW89FTdb829XMW;v-IdN!pr?3O(NIUi{f7Ee1<%fE56qvU|tWGq(g^r
z9sZiadB7X(YY;?)2GdER2=)5w(}()I!|9(l_Fbf9i3$Tc24ATEc|^y9g0dA1YVq3L
z(LrxPTVa#|Q@VP?cwVJz0|6MzNe|*7Z9uHTF|QaeCfqOc?2>gKk#txX3==ZxpwOy^
zw%lYVm3Gc0jvRDS$Vr=Qd*+;6ew%hYhY^cKMZ}?y6U?M6Y=aEv@+`RPc}qoyPU68!
zRQw~H>+=__y~*u&J->5-o|AFPV|$+RFbV+JaJ~o}|KdtL&uCwHSeuOn1l-ig*HVpP
z90O#4Uix6@>2a@%<BGrTD9I`q9w<`2GvbVwz06aV?*64*K|J>R@oXcy4){2QwSJ3z
ztw0No)9TY^U_C2PMaIIMNX8}5G-WlpP)cUsFzjfGnn+Ql6X!10@Al*pe`^XO8};?q
z)jYcOnp!>c>AWQAk$gcQHj_kC$j!Jcet$gHjhZOw1OMuqLnivLSvsVDez)G_=F^&j
zVCL9ydm(VWPlrQY*YEX5PPU8It&xUM&}6z8J39%aCKox4(+1#TPZHmbisJiP`OAaz
zF%axJTg5iZZOxp^_Ocw71Ill;hnUO5m&Z5g1ca~0yNo?Ha<rv3D(|Gu)>v|!A5gF<
z$4Z6r3Tld0(EwL01SBH>n)|kq-o@_opyLe&*$rNPH8Y+3*>6$wB4vyP#<zXG@c)Kn
ze#7uW6_O&)5pYD!Jv}C391q-Ha8A%j991PD)%PSaBeO3H?)VuIi@BZlbYqZ9a4mS5
zHI@j8v)fY24hK+M^1g5J=Kb-nKH%zcCZ<IQ#u2F=m`o2ApSnG(IMp4<WUTMTBzc$?
zHQRpMtsPAC1-5urP%XIZhD~57R4-fqYBGu?PEEARveab#FQ-%J8jRR`Pqg_sC^wph
z^)8?fny`#e2Y3b+lF+I56@_+iES<~lkoR7pr)32F3J`M{D%poy4t|`6EZ6zkLpa@V
zWk@6N;jkN-;|+^@tTE?KJJW=jdoOem^1X>#1)^Qj6CH(#>Mzd#`P51WSi>0GRzvIb
zY5guv4b;iHUG19m7xH&V?Nmj00kVS<e5gqhEe@aH*HIy>=86i#24nhQamLBc3aav%
zw^LpR4{xEypHdrUKo?!=1e{V+p8bV`=P1IoC%zF@6BB@4>^|G5J7N#Rk-6rFU2IRK
zQd*VGyW*!K0A@$>SALC0CS#qxKt2bCT;z{8<j#v`(-5jQkqBeib5OC-*2@*S;x@aa
z>RJBpWN&dC6M#FwXezXsX#yGa%a{_qAW06&S8hJ-x;KY)3qVF2P@-KBdL)*nd2Noc
z*?$6{?6#$X(JpGWU4@$(;1(OUx-xq5`#}uSth62b-#m~1{m4sMh6kgm4J~b?`-7P^
zg+4jQlNMO@{+N}HB#s^hji_^Up5lAI_(33xw>YfHYSs`SSjJ@?_r_x@j}6BCAwVmt
zZI~UkxHwB?wBo145ixwn9Qa7~*evlOnuZz%+;lw8hPFb*p26)DddQ}CF9<$2W5k?G
zIVL;1E<+*=a>g&EDjN4{sd&Cz`ATjyxvg>4n<n<3I1?}K+QWG3<skNu#yjeUjg=16
zn`UTqtbTz_^%G7+{QaNg`5dnyfLPY)bQe%Sp($bUfV-2^)vwqW<b+PcRYh+>3)bN>
zMYuZ-g?O4zhMp=<m7~uvn_zqRdJ#52I0DSk*soJC@)VxMtYK4yU^x3Mx(9ZX2tWXp
zKUXP(lQ;3~>G2|Xu7S<<K``(bCJDoKJG90~ZBxcy!}Rc<1>=Ms^<+|+{*YK2#+~hc
zjK(uE8LB4vVEBr@J1Jfuvf6()p}<+eo{TdP$1=XmU1FJHvodGaaO_AK^;zVlZ&z>-
zuE9rfx%R<_yR9hmB3lzrLq4`z>k$IIFvt!t7Fwy85((;$D*V6|&i&#&5U$~@b#*mj
zCV;|vwUn$f<?cWE&1ToD<F(P0eEw3d|K#LfZYk?HkRjv^OJuA-&wZ_+@uMgZ?Je7F
zxUZ;5<*^H&^=ni3XJ0{2rLr^yMiBK&w>veZH(gnPO4EdU0O1JQU9#N^?VHo%D_)p-
zZ(Q4hS5pj$j+1%WLmx9d7wcK*9I|B9@p0aR*(nBjGVX)1pw}dD29O^zNw{7M**;Qc
z_C97)I@!q0PHR$G!G`lagawvT?^E?C_@g;xKhoj-Q$qoIjSrupR{JTbL(S?n+d6{;
z07D|)@)v!&EP1+R#n*z;rVm(CUHE6q)oTfcJ)K(ThX>!%<U^Fv=t>}ejayPZrSs?G
z;-33zU|N_vJ~|w<gkKBk+l4Sgxjoh(ayV`ySNIy-^-`*NWMY=7P-IC!-3Tr3WG|p#
zr7Nz4^1Zg-qcY;D`_=KsuF!<?<FZ;6wH5S4jl5*kU!y0&C%6b(8u7}Pv6e?-j7zG4
zO*VPR0hpOnuqI3DR7`CtqRlJEYm+MJ9itb!JT>0R)*;bw54W)v3pQcl^OHqaK(0PN
zT}<TMG+F+`ADIG;Lf}T|_G+m=g6I9tu6O7f(~Ek?j9%NO4lOB4y_w-klYPef6<7jR
zYxD2d=61Z{o3Kdqc^J{<$!GoXTe1)1)96yNxed0axf$Ks|9chqPZJEhnnHCyu-VY&
zlHKHsbQ>vn>IfLla8d#`xtCy&7BcR(=1d8641EudqDZ4O39NDzHIlC?x#Z>dxyob0
z1a;P%;wNt`cExI`ItgIvA@BLQhAjfbd_xkr=<pS(`yqbgx=z6Xg8(su;SU6}p}M3V
z*6;v$hvoalAAm&)W`iC8w9b*Is@VHctGjlh2}Sk-k@$5_#@sh%tVR$y{@iDJtixAp
zFA0(+NWDaSgFZ;7K;n9nDfG~ptx#)wfVPK~87=xqz+o5x>-K<wcn9(*LowXZ5CIQe
zYF5+k`ncQnQv=sX`5Bq|*Ah!s009l5#+5gi5ftl7THj@cbA^b9xR);hA@y*v=KfJg
z^-1p^8jyNWVx9ZQ|H9%T@-asq8&p6s7VfP>SsE&+`G1xraYL~Pa!0AwpUae?oy-=S
zs)fCKQ%rfKhyb8SuQ&!-%KgzySKr}_Rs=OjyG4@V51jxl_-v5|Yc#cs%$gBpbQv;R
z-V<_nIEHUxHVn~Q!uaJXbw8Tf6{(be|L#a-q{>Xoa{{!haz8cnzyBD!&b35_f5Qpl
znmNGCL&=f8wvd*Y(=-9Y2fsJ9-+)!ODO0|{JZbe?wXWzj-O|6?2M09aDUT}lz9b`u
zp_tSKF42s7>*a}T(w}3RZD~q<z33npekUg{I?mU{2QgruF4{KIbTp&*D0`|fXe4$R
z8cVw=Nl_Ll4C5!SN1lwA@cLk0!XEMJgkc{r;xVlUZg4AS9b9&Cj0_ZAKxQky6K6UP
z$9`{qZdYSn&(Dc<Sw)Wu70CBbIuX!QC=r@>6A7xiHM|HXpDVU8LXG-VvKFb3U%gL@
z4rKV1l)my`_}4G@id=ykA*{=7^>BGE-vb5^*(G3e%mxADi?nLyM+h(ZS>1-CK{mDN
z8d<g}>Gbv!QU{6ITs{4w8DptFa%GDF1~S2O`qH>BZ9afYN`b2CS{^MZR~-N&Lk#37
z>Y$+TvX2ZjT*ZMy(a$AWh5`av#NYjss7kuvb*LCTS?kABU@7)5VjZa@ce_H#dnmGS
z8Q?1vu0Y*|5k=s(|FzB+yk8;vDK%<_SUT1uIEze;SYt4R_hK|zC73GzaqrTz85wY*
zP&C-zzR91&5uq9hpu+z3_yDo98nJ^poRdYTO1Ki6om*zyW&q?`%v(t0E21{xZQMv~
z4bG0vj-+!amc4g<mw6U^rNkC(FJ$r0-a+7Lxm#gg_ezx77-=_4>+dVZza#z73d$=u
z3H+T7zpR~)1p!U|%KD|y8`O!qBS(udaZ_BUd|4_6jlj!;fG_4LvMg7%;I_oHm7p$Q
zaK9P9_+aCEvFU{#D3Xz&vC5G?;tu5BtCe$TlpqoNb~rNtno=Q)YZ^U(s49%R5DjQ{
zQV=R$Cv?8jPt@&4xH?eOrCB#2dx5Ywbyh3aZ8A{%yj{5)QOxz5uS200zZ(t3Z^?yA
zjK!-eHZ}54X-K|}o4k6?gffiu+ipBA>3V0e=-_ujBAQL}Fx=v~#^%lVLuMLkk?T?T
zs2=fT+|mTGBg~5Zjq6>|RT~jz%_VR?%xW%s+h92mztx5+#5~S_Hyw_(&a!AVq(qq4
z){d}h|IiG*^MrlU`!?ir)jwH1KR+oZ{CJ)|9J3hbZ4O4VvzCG>2s&}Bz&mOpcyn>N
ztIOLL^>U!)=V%8yQ=dS<J&5I?0oICZG1>{e9hHhKkC^ESG~^(+)Y}i8BeW&!)O-&^
zKO3;%<0O<Y87{By&n8FZ`r?NGC*q5i*CnB34lSkpb>#6Zs<|YR$r>g68(I%CQRi*I
zz+9-)o;odVdkp~NtU{-E=@})+8<U^VnU~$o<GU?Oo(I6sKT?Y@{LGQ2$FeOr+}DF_
zJ8kTb*W)lbJC&N8cy4`EhcPZI<jH|F&EgrZE_ZQK-`=q4tsvObkz|OsdsOjw9zPtL
zHR%d6<JL`7GpIp9!^TaV(L5OauGZAB3PR9N+-S}cA2FHD4gV@0%Jc(stDvf0cq&Y(
zDI-o`S>nGO`Gg)~s0Ne$a5r_!Kh6D}u5O9anBEfJ0uMSMVa(%7wluwH+j-p;a3#^1
zp-^y|$3>GRiQI+elJI&H&$$>9<i<l6!Q$y~k7dccRHEi!XU5>9Xt2}nv<Ct5Bc8-$
zvUu#%`B4x9W=~;rre!J8B+%4N^Dj6wfp)7Io}Bq))=2(*uS+xj_ZHL=7AB<9;^nw(
z;&dgIZ9}^^upKus9Oc<8;wZ6NpVWc-R)imnXJpP+AAR)qftz9s$DBC(;Lk5YbeM_3
zVbAU=78lXzsdII>-)E303W)Nb?M%z3&->|uwkO3^2y-F}_w&GrlWy>{=IOey-IC!%
ziu}&Uc`N57PA;GgeAwd-O~uzu$$&m;^&T<FO@`8rAoCEyFVPx!$Ho6W4@<BAIMrlM
zYopUB$g<7#-hy(`sR*0Ru2Z5TTf+Na01wDCCLt}3R2YIG{POA_uff1_9m@k451>qi
zNF)-A5}H;TP5H-6g)}{$0?SAjk0G}ws-p+RDhZWk%GO1xd74IFeCOXSjQ~@H{9=|a
zD9E6Qfm@8W_5E%2?<rtvCSvU4HBVaw2r)@c<*>hS80nWE@5ZbM{4>6}CWAP4SnNc*
zB=kDlAJKeX)4)&ziuXW>0OM-06;Eg4eM^J4$oDltu4f4IVow{nIN<CO+>Xl)!j@Ja
zculx+6)tvH_Rv5U*8FHb&hy{3jtuET@=!V}Go_N5>=2ys*(yt@;3fJUK2<ESOfyb$
zRNG1yo*n;{chK&nLc2#STy`QR)pF#cGZ8H6h$BOJtXeJh$%5U;5UbXgTve9INl=L5
zalXroh$`I(zyAujm<VFc!GRr;HT=tf&h`g+w>?^8HyIX?=(duNTHPVlffJX_Q@f1A
zAz66&Q60@K8}6UI@m%%p8~b+D+zb7W%ZOp?l>cQ3z~$5;04;l7Z$Jx6Im&aP5pTMF
zhTs~#MQ*#n7$nsan4`cq-a-@0THHl?n4?DYqFpsT9U|XDp1`t+^?u6H8%U{Eh>Cb)
zQB)uG=(GxNr@A#Dg(w(UyiY(1Z_>9l6)r)mIfz?l$>&d_`bnv&i{R=*w62A339r}O
zYf>pVnDtt#(XUA0h|$14T@8>2taHafZSQ7^=sl@)!D?&bZoR)jK#$Y+q42Qkb%`J?
zzN>E3w0E`w(V@v-WEO=U4V$4C-5iPvFH?gHeYQeW{8)jN3N<eZsdQC>$S2v=|0#iY
z*QJLn>mfY;f;mR)Hs6>L`dP!#5#V{%TBdjN+59{Gm7;^od6<HD>E%j_XCU)2x6)0w
zJO625BxV(vM6)5oGQcahXD3RU7S4lR*AR;=`)of<so{Xc$jT$Csu;l7G&P=m^JL(#
z$zY*nSjPKdIcO-->A!Twgl6!*0f0y><@U1NG*OvBM@qH5{zTQJok3PL?-S-hOJTCx
zCC^Kq@@EU0^8*$8HcVPVHnSu|K-u_)RCk+J^-;E-`H&eFDDqOjhPS_juBf&ofR7;e
zC{TF-3`imF82ln!T5MMH?<+1>YZFWwRbw7Z8Y;mX@1I-VRx)nb#q&8^@bza&1RYVt
z8Xy7Stj1p$R}qKtny9sLwB=8r9f2#eRo;+q@weNpLo_7}M>oquRTIb~X`#0h{JbQc
zL1V6%L?$Gr-F`4TOEl`ZVrH9SW~~n~Q|f0qN6OW>qw-YB*bymHC@6k<y?!iVtl!F+
zM?+%2wEY@P#50s}(yEA#EPOT)1U%B!KY9Hg@RL`IMlC^1QP)2sMRp&q8VpAbr83kE
zqYt}aB)Jx>vH#Hw{`rlkG{gC}iB%ZB#5L#MZ=#UX36)X-)^yEG^UFUawT;D9os*rY
z=|p`twa~kpcjjwsH$KNr7x%Y%yISwAH(xFJz+PLHsDDC-H$Vlo_`@Ow`H_99z9dPz
zLp;$H{CEy(<77ZO+rRP12quqy=Hn<ks$Qn2t3@mAqs8++^HsluK#Y*p1-m+M12|ta
zDq=`?IBUQ484(gP`x5V&w-PZ=LS5vI3p_8EG%!7^eUONk#f>A88J<{DQ7m;LyMZN?
zcUSnd%gO0R7&1S<0WiUF1tSDxn#S#~AIt<r5})~e?&w7cSbi^KNi8oXsLzj%zAb4s
zIhr9clXtPHzi$k|4pUaqI}|}Fvi_S5^nY<1o)2kFc;C5!dWSGixlB!vcEO@8>Jufw
zrob5hcY8KuvOPv#0?DrwcjASf&`b590y0H4m#I(s3;h1{O4rA1=I040<iTVCq6e8^
zjkU3B0*PGSuf7^x(R*9Ou2r_bkKt}=NTrXY3*W2Ab7n=E(s4Ol&$nN5f3sn`Gk#^S
z&2aMb1<;1+m3MzT%xfQ@8%xz2=zbGV|8f4zX@-`7(bzZNtVeRSk#fn*Vp*J&bhjCh
zk(e6_R8nuRXNOox&0@#vEp_kjwi4MuZGL>@?;T>kg{ie=_*i0%_G<CzHwa$FiTO_T
zb5F%=qH(g}CQGR1{{2$aN|{I#e&D|n;QsDd{*U(rkN%9Hrt+!9o$QT60;c7eXLMX_
z)R=??ebVO^%p5afBralLD@pt8W!<tN;);@!@TbR^lG!7X?Sc8?8@cNZ*&~Ze5k6F3
zjDH&wpDPMV*M$^H7*Cq1cB=NZ$lv#r*!Rn$hkB$Di3>?m<v#<@532A#DWXtOi&vJe
z3eVO%XobHC7iU{qTXHTF^<zpm0zw5$VC2<U<sTe2iy2F=T$-%Z{cJBV!D&}#gCIhz
z%a8!F+LXvCmJenpeACj*@4)T&(Q9rXULXaBGwYm<MJS3Yz#=UoxO?aaI=JA`y?ibF
z=vZN$O8|!3-{o%?W-HdP_>xwDF87f|n)L;6IYIvqjsO3g(s%@-U^Dh3TKt}P)%HtC
z+VY2OLA7_O<b2ek(~!SNBwF4G0*(D}Lmm3w)Lz{kX%w+`7`H9l{w&6uqsMdEd#egd
z-R1NUV}p^OZ4om+y!SIG4lrL4(@r;Xock61ySL<onLj*#VK$Mshj7=MZ96k`5vxzT
z);O1^&=C_DYaAdJx85>eGji2#v&TEtK{r#+d(w2<Tln4;GuJg8kL$@;;b&WG!_wnE
z9jIwv$-M>Q{LF>FHB*LLhH!EAKhuqLR!illtEHWqfM=K8?8S&k^4)I|VDTeg0rl;4
z>Ty*or)T16Xwk;~tF-X~TUnVOF4@<YJ!o<;&hGq>kOC;EFW$5t^E>DG-f|0gWNUdX
zd94to!9j_HX1K%f@L0Zt^dn~3KWu+IGFjXov{kT)DLz>|noB&KldJk8p|B;JUB7(s
zF8XT4hx7Pu;L5=uA`l;AcSBdu9rZg5HspW(Ma;c?yaE9hyyO&4oC_eGZwV%eKTuEU
z0<7Jez&d<+)~LXR8DK6D;&B;()OfE+Gj*`<ETc($WP7;(OT%pf9pfiT`^pNZwjeRR
zQ509m9U)T92~$L~=mnR{?4r=l{Ip&n9Z56XXQ)8#LBZxge%;K=48=HtgYGbh{#fl+
z-ZD6`p(L^83EmA`O29JIa{lpIf4l$XY<*Z^ohs4?jYaq%9JbW&E%h@^W20{rOL!mq
zd0y3|lFsQ3^zVcbXX?x}0yEAvrC-8{W7#-OSIicz_l*uMP_Bc~D|jYumZuZ3VXboC
zvuA_@_lLMJ5==T5AL)-$1<6hs|Lg1d`what4a172{~-QhIP^7q)xv+M{zh6F!f#N3
zK+f%9qLvm-ykREQ6Gz=m?f#3R0t6ej$0y$9eWe$(6;^FDP2F4>vU2YwgMA9s=s4~5
zr7x;uJ`kpm#eMc*dX-`c_t!&s_@W%!6ZaPC4Vuw;z@UPRIz&$GLCw#(D`uG@NTmL&
z38%)fDE3!s1i0)u<R<b3yY8ucZ;QmfdzX|N%0c)tQ<hB@B!)&roFuip@|gONRGBcC
zNXIn<h)15>+)YiMHxh}z%@DGbN=YHyo=lRcW0KrR_)^q4$OCb(>RMpw1AgOa1M7ft
zW~#py3*x{0iXjF{UxLQRSw#z-<!0uDQMxl9nz3ew&pb7jNQyPIsE0LO?=W4xz!z6P
zoEJT3{b!IJ4vXm{EHO7h++3x&Cn8Q@Pu&pb9&an}KIiXEZiCZrk>#)I^#?z<5K10=
zH+}uD2dDv3AEr|yX;;$YBn}++t0g-j-H@`|LVUq@>*u?9tB}zEA4Z0Uf7>qOVtT2_
z1H>k4c8H0rx~`R`qwVk#=)cw@DJO@>ZT47GrPBi+s?Cz2thRdhKx%-4vCNyNYi*|t
zNlHbFcPvnCK6UQhizUUGqR{61F(+ph1~oN5Cy#Cpu!J8r1BM;rsk`LzUwi)wdXrU0
zulX8g@SbX3ug~ufM6kW{2ud<UFwCtejF8`N6gzPx^C&9VLgkBQrE1@T(1dADsH6%?
zBAdb2YMt$q2SmAZiL7q<xGb6YC59UET@9x_I>!f#nt9V1)Eq=5CPFF>X^HkA<gnQG
zVAjAVh<TB!Y92fu@>GXXnI2;O&1O8STL322_jX@YRxYsH5`MY=v{G8vjZCUvY)qI;
zF28WN*@)Kci=t9y3dCc>+?_1QJJXL`3lqZkkR`3jDETOx#eK5c!Jw|L3sd^iBZTu7
z4wFVbpK`yX3*S@4tU+!|ANGH&Klh^0_jb4w*uL+5esGt&dGIY0&QGGd++*3~ybX@(
z0O<<Y<=6}bq3%9-Y@^>h1NjZ6KQj^}_8YZ)JY|FG3V}^=abnt^)xDRl)o7}Yp}^g^
zdI-fhn38C5-V~FM=+y=F;jb-c3FCW=>4%zHWH9m?O`WvFQg(4l-Kwo4U?lg%<?4hk
z5wj;+{j}8^L79g8;T9>yNvpCO^)qpwjCBF8=}Ie7j$EJSb^q_*T1Dzt3tK-2;^`!K
ziyAgP*^b_feWuZ<#Jo$itN&~qgHTSVq1ZQ&yHCu`9v;Fe;=WfT32kTHNuQzMvA83a
zeCPu-8<&NG^H!s68-XvLPSTf^8r$V{Rc&@f$DAe$em0MAWzMBnLKa;X&R!5cc1DGc
zmrzz0=Al;%<9>jK<T@0Ue)6SsGZ7??7xf=PSRBA1Or05svg%C3k3tyogm+P4{HcF1
zRH=-qQIuM6ULt%C7?W*D98OxOtZ6Y!^X8tPpTM7~dAUDRYl4WQ#}!p8QsRjU$MgN^
z@!_-Oy6`<W1GH+Pss$UmjFV9S*bqN;`=VNKz=VPdI@_NWDo>&4SEgw6?kW-vqu5rT
z2<$}=pAXWGe**oF--BI9f`oY})8#+HkCW0=s!WySUQqk|k3)kHmI&BIa^kL=_H*O0
z(Bca_L{ZT}4RH}5gI2j6xz4nhH+Zg0$RplVr0oj!<ao%fwDgA_y#v)G94<48I}+Y}
z>y8IspXpd9e6q_gUgm3$2&yY$P@-xMv8b`bE9tT~8U9^?32))}Z~cnRHM{Mu$N`A%
zezh2YBwOI>LJkfUrlzr9b1GN+@XfFUzH;Bdgh~iQv5^1?Ywmlacs$f~;x|_U<(g7x
z!>P&e_c-+C#2yNTzvF)shrIp7c3!2y70pVAG!V*H>FNTpIdBG|aQ8y|`l1Uu-all)
zqt^p5t5{%@zv2FqhFv21HpgqkjNnS3YU6_{s<@)R#m70AUEi`7&b~aDf0>)?2C;R!
zfP^LZi0%w~wWZH*DKPi-<VJPLG?ew>OjTN=P-l>5844+Vw$9FMFjq=gTnA=x32>50
z<vP<oxd^^e*Mcd91=-p3*mQcYgQfD3$t9Ee!`BzMp6=0wwie*47Ho0ya`hyc)@0uP
zZ^t(=HzA4?@b?eoa#nRJfYx^#bsP6jnP!iy(C%WMuKofK;^i;leIm4gP7Hy=sdi;w
z$%x?P*_-e4F~&S#WCBrc7^x=$YypNn#(LhLeruU2jZRZB(9+tp-x@l*c~3`N-}!u2
zLy&|{z>$&*3H~CD948S9gvHmhiPutFthV^^UE#+!`D_p9Se=}XO;FvJ(U0xS>PP&B
zjP76OcTgr=%#^)=@PXN#k<GUhdsQ!JyWu|~|4><~i#MDryQzPfoy1!rT=a;qG){hw
zTI2|9R-=ywMQJ~k_HtQfAQ+B+J|tKf*WHtUQE6IX8`Lu&-uf}<Oi{4;Y>k-N&JIgB
z@vLHlJ>l~n$}up}RCPXtuRiW}S?01P+<Ue8`}8#w4x=$d+#zT7+uV2bMw4O+RDUSZ
z?QM#SNnlAngBbc8l`87``_n#VZx0;P`jbvvRh2urGkf;WuJ`sive`)ev2;R+p;NO4
z1Q>%Yf4z!-bua%p5JTrsu2&Jb_NO)PgN^5_?LSBn-Vz8|;>ZhL6sXR?=P8i4dLsLC
zar&RSUL_K|V-)k0%Ul1$o94W)4YN-Sa_n6V7FZEsr*XFO)fXTGu*f(1a&v>idJ<K_
zf72#c3CQwo4MtfM{Js2^lu%@%dX0?UW5LO8jPId;p@Z*+sPq&N7}8`oWVjlSINu-t
z`dOppP0;wHe*Xy+=gDu(d}}AA*d8oa?bxugOAguBZ{s55gFfME-8DT%Iy4<F<N<m|
zSfZ?qEaXe1^L63FFjUtX2Nm9*&>%yxo+f=Fsi>)Pk|5AYzud6Rf)<5aZn3YZ5#nQV
zL-NX6rm+BbODfXGF-{Mw2G*ZWJwJ^L5U`T!P_HkhG^})>4JZD1(+CIb?pC@zxiY&w
z)(>iAieP+@17!0dP;}~0n60-_&bLRPv*WArFgTF16i8@33ec!jzAkKfbo}<g5lN^a
zE5{@v>}U*f`Nh90#Q$t)34Mr`Z7G?pf=g~q*E*55mo)BqsC*d6=koY;YCg)Mt*nKh
zbLTmK%$pxce^qX>*9keSPRO9cC`dR9=X{lj{M?z<SapW=McXxxX8K8GiLJN4Tl;Nq
zBv15Qph)GH2LfS>zWjA2>2N@8^a{8pTU-}2*tF-Hshx}B(!EC%_??DcC5nZmYK15}
z_|xFG{j4vlmZkMvrG&v!QnH7qqQEIA0?p>l<~yShR4V>=nXynunWC4;$#WF}d|w*+
zHyX}=V@V{I5>UR0PIi4om^%LMgpQto#wP?&0;#u$iOw2db14f-yfZNAu&m&;<ISrK
zowsu@P7LPt_IgZMt*3P$W)_Y1$S}3rjG)+<bF{BA_M$Wk8LB!M%0;9(-i8xCymBvg
z{Bfr5=`?cHYBICKZo7~Dtw5P!pF+x2W_euM(vmiUN=1IJvVJjFmRwU=ehH(sbr9E=
z+Xo5H9d1jxk8-#C?((25z6EJ^bA=v}A;tb=^||Nx-)a+RMEpUCzFt1oTtk*y>mr9h
zRMssEZeDJoHW=>)63dM*9nopGCwC~Qxo)>T$iSZ(0z}b2kwjd~TJ_d@fjy~?%^29d
z(pf+u4JGBkgt`=MwL~XKVP`UdMEd>3;(WzZb>+X>gmnQJ3VCZN#Lr(v5t?lDt{7H?
z29;Q*3NKo=lJ{!hRwhq9VA0wJ;>gcN#ORxzs9iDjE*7>APGjkUdO{cKx_DF*u3bFA
zdhCp?KIqHeg^{QY*V%QJ2)uLT`(?jecSIW;R{i;U4n%T$Zw@35|B=Iv9iSMP^M(_P
z35Dp3%1pg2F)!QAmui9R3VusEtg}!f%Tp){QA5x<)L9ePy!S~K6D_W4l%C^$A|uHq
zWw|))5(?%%Wf%p6rdJ`mhRA#u<~juT<<&06^23IHn_~3RJ6L!s70)vJuWi=#VW{09
zaP3ot<R+DNXD@du_vb(V7|rnYrwMBFkVxh%2>1fXbP+Sj`Da4Ap&wsvVaVuYT$?{8
zcr?$H>3gp)pj~6Q!xcKTOcrU-7{4S4;8mRN`Cm&JAF8mZ>@wHW7x^W?7lKb@J?Y9e
zx794uvnaOEMw@qQ!oj4b*OYuLa*w2h9!}wT2WAKm54Rh;Lp&?ooepWf5Fh=6aCf|?
zG?)}TC}h?t9I9Mh){rBl^YMLpiD3*3mEO6b-lq(_s#k(x$7=iQ83+a3eI0>Ub4!+Z
zX9njFh<Ox542|?_7R~sjqUQ9FtPG7dk^<QZLx}{I$Yh$I%vo6!+&ZFR3$9MM^Uic!
z&;iBpT~`B4okJvciyx1o1p-*j-y3pE1Xr(zlNW^3@{!Ct=;Mag9wl+Sy=i({L{4_u
z^yKj@oXnOY`jX2`A8Z?2{^WRR-BxS!HEUhcL|q*j+2g%<0<$?QQlRfD^?YWsOB4j3
zkE2w|h%!q?E1Q<r$yrJLmNF#?xQhwH=%YT+pH;q7T4QB!I9<5%G&`cMzTeDSUMQBp
z=B+WuuL|)b)+T>dRW|kx$`7yBeebrjwtW!Iesh4g@8d;x>w7C{Y5wKq7Ej(&Y7C1u
zL?-@{j6kXU=6H`6G#jcgI{asP192PkB6In%`9USDmnQNd;al7=n*;IW-;8MlBomDT
zxfzL$=dFfTd{>ETM=2Ew^UpwI`2W4|6PghN?Jy^}3s;+<zi-DOUoQ9s7x8Jw{pJiE
z*q*w4HjKckUo~rw)k<K4H9u;=k71iaK59yac;gv_TI>mdiU!<#pv&}P(Y0S;!jQ(0
zuhN%_$rO89n2Lx})bm8PAz;x8zpBYaR8{!<{#x%1IP77l_VJ7ZcUM#(4*VCDN<Jb~
z`6yeGGg^(Dgli$37K23^axnhH64pp_wu1VK{xYFN*yG^>j(dU11vBg#ra-O6-DM|f
z53Me4cOPmOvjy)ya%c3zXvy)c(aCVSkRIdK@e01d+$5m5GN(!rDVb_N#nM5j)mCd>
z?2=Y*zT?~s9PuBVjMfPD0al=*`iZGZ*=T4GQJ=nkgP+W(HrC_c#{bfKt=bHuPWOgn
z`mnYT>5a=4ItWSqU_VuWDQHQP1nWVb46P3@zk&g!)AIFICksypbM|-}luiOEEHNR;
zv9z3aYU|F~jXCM+dptCWZ1gxrlRsx=o~Yf_&r7F>i#6Z5&+Fb-5ddx{7YBMP=Yib)
zs5y!y#KQzQ|MO1u$lrV*|GgDPn=FCt(AP8GwD*pnOM3wt_ZlOH^S<2lz^l=P!+hDj
zA|x<_u*^`70yslL<>_`fl5eD5XYfv>`{1}fOe7rhj46!Km++@)_SM08Nyxgc@GhE=
zR|WpHAqcMsLEFVX$Q%~StY2{%Nx_n$*;emzWO;87As8Z2_R8-^?P>>J4u$_d10x1x
ztwz)Uztz=(1@Ob~^pg2C!V0zItj(TMEEK{_D*rBARWv&5U{&9MxX-cSReKjs7s^Dn
zlO2}z_OfTDC|y!KwK~l9#2&>>rMad<KQsG_8^pLuaOJLAjC<J^P2Mtb8(`SnXh0-Y
zSBWkxH)8;?=X7+l9If&D`#>CN^#LOj5bRe=zSt|bHGwcVCduXQ3mkewtCO~Fw=6ZD
zhR)?x{fD6)6&li<2Kd@`s)N1P%>>si?C*_$^=r1mNC*!EDOd3s8vdcCj+FZ_zrs~(
zu>T!~ScqHDbchm0oknECxBL9$<Yc(IL|u*3lAi_Ye=Ny=dOQ>tm@KiAmXozpWXue9
zev0?TBYf^}B$!M@e!4#7qa#Tq!E7%sK!XeBuQL6OJ4(ljb>W(~9wu{8!nKlRVvC7%
z<fyY1Ws3w>bc{DOhZ)T{<!4(r>}c&j-Tv<DN|8N~@KiO2Xunt+5D!Dt)El??WVdFD
zju@@9A(L~r{B*#fV>1ZCKk(>_zF<-Pkli!X?5Ik7(x32y-Jn{MzzT8f6~?-YLW_zD
zv5t?CSwkd5`4Zp3)e;72y;I7!j}|r7w{JOS@6~#O@f5z+gAuGo)3D=7=@+PU!M8`@
zHhWOTnw|u^lQzU*OsMY|(2kbAf6OU@>x+6@T>ga!-QLrn<2YKl{sq3Xm@@Bdn!~6`
zy$9G)r?618P-fxp#m#U1wv+$pZY-F$A>Kfno9Y(wa0@OEP`cRcO%^QM*4f;7zL5_X
z9{d^fKMK=-+PLl&ID7WZ-MR+H4xWeG5x1_$wp2fd3*nBfJ{t6uG07d0+KbgJ@iV>#
z$K22Q!6&d%d#EhB>U_X)n~P#6E0zLYy3Vt=E|HDSfvM;MTH5LNZ(2@?@INuUVzB=K
z_6TuMg~hC!p^>K*yr>igAw#(EX=%xD=1}s|vc=Du`(qb#yt99-ow4fK0t-G$_aQAU
znZ;%g3#qdAT+nkZV;>HZI3TBM<y@DP^08B(3RwUY5pO|5Nddx;9Nwm#r#3oXBpgp`
zzk5mZKIThX$OHfVzIfFriBRpgj;UeAw2gx~jElDiAzZZITm^Qfu~?wQ<Cs%joYo%c
z9Zn~JKDZv;^^v8g{E14Llv4r5$VgTBGscr(Ah(te^>TgX@6W+zTa2#F7{qN7f`o2_
z%R(IYK{!mWaBwIirQd#}mEVBJ>5>p^hVcO4P+P8Swso$7|7EFe-|l_QNK2QN>AaDu
z{{PlMdq}}#NnW*$q@RYUW_&n1a>L1s_Uc^Iy@PxGZoTy?;4AY((MebLaCb(h(^eur
z;0WP`^-39tsSo17)t~&r0&!L<GKQiOFJrj4P+4ttPE8l5?=}nI<d!?W3Q}F5pXDU{
z{P(i^zV^sYmhjT*$X`Mjqw9_M_@dcB`L5eA;-Ll^@029=Gua=Eoh2^F-bn2@C6w_n
zk_Pv0$cZBlzcJ+G)C6sXQDazSA*_XW@i|rriCXS)QAC!>@Ho<N=}q_DK@QCEvZsYq
zhOa|lsFqCtV50IRvfg5s>G$o0pkJ;w&%;#T=$&aBJ-NP6E*JJ(a??jELc0IJ_6u<=
zyo4_xh0|eYbiz*4no6md5cTEjP%bPQb>sS>{wr65oqAkGsNia5-=;=C=B6Qp-kP#i
z6I6Q){{T0Z76X{V)e*|qn_jgRc%Ztp)S-3k`LRoJ1{E7kQCC}T_!nQ4$h-~u<`FI2
zbsoJC1^z9_b(fx6$*4eI>&a)l!!h|>8a}_<!M2ev71BM(n3a{K_Z267fhKU(ft_Vf
zng624-&M!!D}9yjgazsVtBT)7n}euUP4Wj$&FHaD*?2T7t9(-#3BPF6NOIMUO9b45
z=Mt>l-pox}TK{{+3H3<;tMBCD5YMi?<znhA+gI_ZL%AZM*wqCs5PPNdZ9{&vAVDIK
z1`9y(87BwjPx;E~VL@eN0ltx`{p}AL&as%OG$|q}E}2Rr@+aXR_};SXj4<WlztpM;
z37W&^W;-u*|C%y%H^4<uA9%r|=MNdPSZJZ4U7j8i@-=CAJU+;}kDsRNW;C)nmfHlT
z%b7|9ot+K20)O`YX!~nTtZ(@SYpI{I8EIiBj5fteewq~fyWC=jQ+e(<UNy2_UtFK)
z9NYh1Y4?K?{w3bI6-Vm)Uczg(f@IFDr$<4*$FIu5G4<|8+=kX-JwK4abKEc~UM(MK
zxIu!M`7h{?9S<}y!Ouvy(G+uzym%v*@2bFVFHb;-U9)~NS+lI9-`#2$$*cYOk8Z=5
zTdxGSmU4(u$fcJvdGE6&ewOj~6!)ehM`x_=Efgcw<4k!R4e_NUa3S<=^ck`Xtq>a;
zZ!zc+J^qR(N)<3NpZJ4-^iKW_^x44@!s61D=hJqw!c#Er%cl}9$RSteXc=1O0F16C
zG!%t`m!$L4P^%|2NPC!tH&3jPk8Q9vGl{x=o_R7DiAz^qoNx(`Yzs$ob(AhF)<o_Z
zhC@Y0d}_P26H7K~c!^7<9PII~&^kEVtL<_HNEVrW+xIK1^>bD3zthNlY7Hl$^eygO
z2HjS%?d54;hiLM~AKo!ry^KKEGDspde?#$l@ONgjQWh@h{P-(0(;o<hh74^5j*UI!
zB*&-_C4|RW85zyNcx=|!@0Zo=+I}J0xLWSdl)+|+$G-&KX8Vt^M=$NJ;1wSTa`oar
zE8^`oyGtoNo++jawXOGItbQ<b0B0dw1tAi!&sgQ|5xA8Qq52d@QpmkB`$^eaP_^i9
zh>T~k>mIdixlu1VKUtsxY<=hkxx7R@pkA2MPYkg7P8O;mJUtw_X)e^-P;71uC3oRe
zk;*wlMf`TU+=pMNv+LSHHR%t@q3oFFfw5|Jy*b@5zrVJ)MmU1YMdV+m_A=WMjiyGN
z$W!S02+5i=&15*<&7MQp@ioF_<g)Lm@lGNXZQRlgxJEB9Qq<{ub=qQRwHZucIqo8G
zYX;)w(Wjh(6MKEN3PZVyWLb^urZ@&8-fW4)ag*9tG0c@_ONFagamE+B6RCMSef*E3
zm0^W*<6Q@{0?w=&_4Z#l?LWCUI9>g*NqG>4hTUdPUGsSpq*7s+dDvEIGWQ32j*6Ue
zY^-4jRO@lZ5m%B5z1z%(uLr4|O6h`#7gh2F%I1I~#7+CPwSAa+%>@gOvP^%dGyW*?
zy4enD8&}{VN43UMDwX#|#nZgY)dIy}SVH>tJvWST!GdKU6?5&lecT7u5Se5+hS?|Z
zkO(%p>umrGZpG4YPQ{P678@>++)`zwhSR<^T59C5KA2_DyNZmZ(AD%Vv}n3N4WQO7
z1N<F>k+dDx+Ng%hS)<(&9v7ckxjI{udjqxqhpD$<h$>vyhUt_pX&ky+N<w<*8oH#r
zq#LBWyStI@mR35XOS(b2zUB9xz0dgxGwWH;y6-Cy=Sv?&TI|;Pt#C+<ovWC&>cs`!
zX6qzh{Hp+f0Dw~n%YfPr=d)F<!DzOhD1|~|KG(p;b95;>{PS?OR5?CR^RZ5J_}$C&
zC3Zye{Nr?2|8$|0%jn^{e+)Eb;a9eK=DCC455vN*SVt<Uhq{jsfa3&s0CP`R{0j$w
zWfHp2cc}=B(+WJtIqNDG5nNJdzk1&nPnX=c=j+KMN5!Y-nPOSYP*~nWBJfkzBOWO&
z0Ig=89;@M2R)!Spq@78LOD;n5;}?7p$24fNN@l@b8=vyY&?|hsoK+MqzoX3jXsa?Q
z<!m$kg0v>9_w^Sy!-BJ-1TXNT`L~;=*n#JUuLrTr1y*c>{gxffH5`@Kp=e^UaBS+6
zpzX&0Av2;xnkhw8p=P*ZI0`kUs=67)XditugzlqxurIQeN#2nh^gY=;bR#p9l9zrN
z!NjO|<=q0(Y(cLUghZLCUlCzw6&V2bjCi3Y*-z=6EpoSSKrkxK<qLPBo0#J$#<+7%
zABO0me3t|CMfxY7WP<<TG66PXh6hvPzZ*v}$bIPuyz&F+5^1OlXUv?}znSS6%cGF-
zOhlhgafX@Fj56Dj@pC~lXv9RVT9oJMCLE-j|J5b0;P0*pQ7*Lk&hr)Q)zFaPb%$`#
zAvyF?+RSoaj@qN4eVl1J#(SY7YSZe}=gDn%5>#irv$bSC=tGspavzz*_#E@@{$(WK
z*<T!3*rL7Pdw9{4PeWTSB!%Pb05+mC`_D{;568~@{v-GG#@S<!C!FN($g=umK|;gF
zwX~aepAq9kTNVyTwb$Acx+BO1zIGP>e39gvZ)`t1-)<%7uXs8nvvsMXdmVi1e6Rkg
zGF@fwCuWMQX~$x9GZBrTUhj@N-CnxDitJ8=1LelyiZ%m1XDK~&Ev=wF73%$^HaJ^Y
zhtJF2ZN(-3F{`U3G@;qZQWGMb&JNMsE9No48|q+tr2pbcsIL9?K#PF7vs^v=P)d%s
zu+twoiL?{C^EX$xG^@S5xjT#4ilMvbtsZZt%eC04*$=~qY1L0VN_4T@uY_68u#&0a
zN9u;3tma<wq!8P924#!z^TmMk&HT#;tPeCd68(8KObq0s^6gh~e@iLCHPiHU%&7|L
z-yRoxaH&ELw}XyFM6<5WI?frSvl`Vx4YT8V3n>oY4d%aW;gcqMFB<rD)sRZ_PN?CZ
zl2ZvB&cLtWKk9UQd%%OcQL?`kT9%oW%B8V={y-u&-cY~V>G2gF?AKh$Y4-E|imB!C
zw<7TPF33aiv=&&Qn)U2jU&fE>iflC;NWp$7D^keitjceq*jd#dLHUY=K%Q<O9GA|n
zse}>L5&aI3JmRB}XT1CNdj5t>m9$hrgFPGkyBb(^1<|vwbG)y>@L<dS#=aMV$H+gd
zyG!P|kT{zRvf&Aso8T$;K2xftj1bG^#&Q<>9^V&^-Ao~$_P`)xy>>&sWdo}E@_k*k
z&%bbK1=L;7*O{w$ur+5Z#ZEZ|SmLIt<g)p>1!I@`c{aQYDZr|l&G@xbE>sx|00+i_
zO1|Jq#lv9V^U^G8i_AQ0Lpb{Fh#Lz0+B}xLHtGF%SX$zk<M-;JM_Nvn9}*<BTasxv
zQ=Hg@IQl#$%i2|x#rKtlOHM82@HeTjZ)YH!-uO+vC#YPb51JPRJtka6upe1;AGsnF
zsin62ks&Lt{}aIPRDXYW`XHIc_anSW4<7lvU$fvNoR9PoB*%Rld3UZ{^~3ZD^R@`+
zOPkx8OKO)$;2bZK?0W(Aeb~=<zX$(=%!#7~+@7jlw!2C^E*8sL&5dn~++w!6v*&%f
zH_UL}7`fFE?Z@*rb14*(cHZDwsuV1pkbwi5mg#rLOEu3JlNpNXI^6cF?rOBcHGEks
zO~-%kzW|j)52V<3!vAm06+sv6-AAPx2${YfZIknJblgu8IPZ+QWAk`xZ!(M4aBSmf
zF^CLZy9tD!r1yInyz}d;dHuzsf7k^yA#{8jz2}iZ>bycxz%#m7|I+K<NSBP;#pdfL
za=6TW?bA?&HlAuO#7Zi7q+vDp{nh&;;0@~uxW4xKd#f*5lt>HfE|lDJUEt$+5O=?N
zKT50rOqR-LABukEW`eEF>U(Px$qy5dI9$w5D#3x<R}?P#(cSk6e&W^!G1b6Bp-hN4
zL9O+@$l%5}&QR}C2Y-OqF2@(5%8A~GuMWOS_N`odmKUcBm7dz+J1bqJEHUoyC(v4P
zyq{ckMiY)_sa$$we<MsPMt=UhJ7|&SbqE8Mn=_v~wM~wxaYQ65aSs6A*Rg2+sIkIO
z-Fhp`yU6n5M<WEeqzL;6>+s!z0%AOuS8tdU+~V=^+J;PW>iT8NtmVTW`a~f_$;FEc
z(IrTJ*S~lm$$ToR(Lbb{P61oI()?eb;Jl=wa5-Fxp{uAf)I<fsy`ZqJS)5Eyo|wTx
zC=V?NRFgq$<K_ga-_skRzh{Gzp61k6&7VwT7lY)BJLOk6*%XuITS~L1&AEd=q~Rh}
zXUzN3L`qr37v)%*&GDS|y!k?k;~y)t`4Nl}p9rw)Arj+x8waVkvC@N+bg{0)*o%!U
zTm0p?34wI<^a&_sYvsG(!P||u*Omw@Y)G)FNhshtT@RE=X6XA0WSKkx(UoI`u|g<C
zL$VR|)Aem%_bXmvj?B!7-1vt<U^0tb&=!~Wd^j(9Ifw8?FYM7nZqh<E5q-f7Un=er
z<>|iq_&Xlb?47b2vtfApK)?H~(2~Kaw#XW8=b!OLZF)~HcN-9HeU)1pmj?^N@;w8u
z(<a^{*H(qK)v9+Y+{M;V<x2`VpR^#n+~YXMr_#NM?fMDJ)ey=XH4>779G*CF8}3X?
zek%V;AD_80WqycH!_Ojo+5Kn)G@`P7DwVSdD@tw?6@Prfn@kHr?9Jf(gYLlyQa&d*
z03Rx4n4~P9mr0ScJm@~oc3W4Tzc%oF*d>{c<%Gm({P*MeDX@3%7jzcspl0=Td8sBc
zO^2%+pe%4fnIiX<{^4e`fEQu`>O#ixzK>x=+wgj?sG6Z?yW><C+FP3Y=zxIA(=9e=
zzl~pIr-xY4M!hY7|JrE*UGpg&j`ruMNk(`MhwTpDf0wjMtgy(hn_hh|yw=*E-7g2J
z)cWN5wP;X!<0&OJSL{)@1?ARr1UC<+p(ZLE;4d0{)J{ifaUAivy5A_3l*i4`1Umnd
z1%S{4IpX7OM5C>hXj1s~3)|K(5-C8dAxgmce4P-b3=0o;@MqEb<Ge=rz2L~RmhKai
zdVU~|dX=<zvHEqV+AfCh-w(CJ(VT(WUL}!6oY&rTD%NK+gZJ(1#vz)g&jE<KY}V>9
z6ZP59vlS5NEK`ln-1~Q#0jfV!y`S!+#~tunKFXH_YAlbRzap}H@)U-9<^K?E?jp8%
zkjvnBJC+1iZ!nRO5}!q-3<<P<+LL>A(C`*$?tch<3~=l4t(b8z-kbAA=6Qq7Dwa8}
zDAjFymjih$-*Q1_=AFU+iB}@i{=$-PaQyeV9e)Dj1b+R-x-rwHR+XIZZgXu@|0eOY
z(}NEdtJ`GuJz{wFv|UdW9U&72#0?@!RPj9+!WG|5=6R6+{Ys_e>;6Rfk1Y)YABY~h
z0H#?{vr0!8Me>@Y3&nQKLKv@n5ExXBnV(d$(`?OYKJkM%lm#o^y`1$Hn4=$WPf-DK
ztX>CR&r_`$He)Dbq)1(EfwlTiF;@_!JlN1=1<2CJXVxx*Q{cSHQO?@@2)f5IE8{10
z2KycI#wMBA2KNeh-P&}f*P_nRg1FU~;nXgg#EqgFnF^o9``ho<y`b^B<-w_vjuyl>
zQXmH$kB=&pPW!-dC7h;!K1$@${1$e;WIHDeeBxZQj&>TpSBl=xG~SfGlDI}`no-qr
zo%&6;X{yx_PGdscwo-KL>LeKOlEs8TD>zv1lLJk`Ob?c9XcaLPRHpo{Aa<gt>q);?
z9HlS(32a#VKD)^pn}7_cK_UTzIVtQAL+GBO6Bz&V?N-Kl*YR^akLYi_o&O#vuRMrx
zPWgJiz1+?3z$Xui@tBTY0ZU-eu97g-_SYgwc6=7yGE}jmtJ~9+<7aKeiaSwYlZ2jk
zYXY7b4mcxhRj<P}7?rD*@&CZ$;&=goQk|UnTQHa9c5gX(w6HVW?3Miq=D{i-db7P^
z;0Ljj(@07wT<`O(J}+KRKg%~C$cgrk_`(1vg?5;42yK@hqp<2X;uhM=+|~&Wy>!4N
zHY7As&5n)Du}DIDId-!7zoHXF3ZP2zQ#XN}XkQr>E;HEQY4_^&1vS3k^KUCKc}w0c
z>lOsTZbXN&O*!?x{+eL$z6ij8*ZT2Z_%aApCYj?^KRerHGTM}!AJ_j^=AgT(R|Fm`
zF2Q*EFb`TqHUN!{C9zAsUVQO+3i|;B)L}ajS2fu_TtQtK%@|IS`MUB*uDmWl!$xfs
zd%rYg8Ll^!qT(~l1c4-KQX-Q`sLUhp;+q6*dtS4|{Qd3U^aBX=YTO-Cn5frFF7-tR
z*XGK}ci^fXVKozRNq9R4l*_&^FzYHlY}&To_b2*a|3Ynv%;Gf{(0#t^>1`Q=VMnLm
zov7c1YE--p>>VqPWca+~_xHP@^|~Y9_BKV`F=;<SG4r#+LCelxaz40j+e~LI5{byq
z{la$h4^1(R@LxCkzav8J%gox%)b9AYqzqNYj2Ca<AAw~d$UEg<yU}m<j!jh<lCY4)
z@Ei+~XKH(18Qw|l`uv@inyS{}NmgQz3z|rvmW5I1Vp^$s-2at{_;0^Qn+^Mj{Sb+i
z!YHREjPrdm{*hkcOI5;#!;z;l@4Qsd_KIBxGbkaOt2G*pV->92D4+RJbVl9DslLKO
zH|p$@{0XqIt0jZw6Oegu0@CKl!=Eip03j6QZc^}W6_)pgQ>{#?kFKraQCgf7WJ<iW
zH4tiB`QB<LJzj^$em#mDZ3QM?<rfo~DHVfy6;e<S9wmsjT#QJo2+lOZc3}v30iZnZ
z&TOTf3O~*R%bwr_IJi-{P*UVHnJ_yyEKe>Xap_A@uk=0~<1Q@?O+`eZ<o-89V(GzD
zB%`fmuk)_bZPB1^>FAMQafaw58$z&P8R<7fjF8>&RMY`AAfii93bXHx{OukQmsMXk
zEhbAmj&l1h5S|xR1f9cbCl1J3#5pV{kFTitNu!WehVJg5b3dl_-S9J=L3ih2(>0)1
zIWk07u|OM^=1WER21L1Zlq?;#mI6JzzkZhWC}!ZU8wxl4Hd_H>rQr5SNZwJ-k(V2&
z<W{N_N6=}t+y5Df7xCMAe`<b&(YR5qS*}uiL9ItW8$iv~8j8?yvP4OKU|IaB!!oQ&
z@!m3e;-N@|wA|<AX0iWCr~QmKd#IM8%)31$eN8}#^pKZXuf5SyPDU1X?!Z%Y5%0f1
zygybLI*j=5$4BcOx6!I@V$!NYtYT>_#O^!Upki5^d@G0-d6M5V`~NR9p2{cl_QL_D
zI<Dmbc959icnVavs{FS}VIH9nF<C<XIrySI-Kw;%R~({msn-ZF%NpSTd&+am*z!N1
zGsPVx;Mzfn(AMhdX;|K#E=<g`Q0_*g>SWPZQ<~TM6o*F}Ik@DyyP;dX{p*l}@?y<x
z$fw=wj<+a(clA}PJOKv?^v3ELHnr0@$O!9|T?zlxW1|JS;)rVF>(fL;r=!D9LJ)qn
zh|*#ceA5PqA$L?_B~~6=6+rpGPM<0R=zGu~6v1UizD=DHkw2b>gKW*K>$Ap4pf~b`
zDI87|zx3Zr%J*T_64o<Fd+ffg!`*rFwT&rP@lv@Q!-u!H!Y#VE^u{|UyBAReueA8_
zwEAhkv%Kzx-(A)E><o2R;IY|R!h}m?b~<pI>}CLqF@&3by*#Frcm9I;CY-Eg{&_A!
zit;MRX^-nZPB-KB7ncL8m?;hPg9Nw@Mk7v5_fK+*_!i|A_00j=rM#V7?3S=IluA!p
zLN}RinAp3-^KRkRj8|d}gzyn%4wZiSmvgChi+>7MhgT9_$J`KW4r~Xb{j-vnUk#3e
zx}L~S0dmN=;QdjzHP{*i`Cdo^{trkEg1B_a3o)-R?&^4|p+P46E9}$X`QfzG;n<K?
z8OVqHL;1(kho8pcN0IOWsUv~~^zEp)ba@plR^jwT7R&i!h%PcNDmPTv1o1v$n{{sI
z`tNz{NMzppiqE`GYo35ihd#M3HZBY--Ebukr6#te8=!_2`7;endFCL}gpBQpY=L?D
z9*GEh+{7>*aRp8gW5@#S@_2!dd@Bx^+Bp}uzmK3OO3LaR7pdSWEPpfIF92WiFFB!s
z026*7E7tdLOq=;I8!0%TDGQHqOM;;x_S02Z22isCTx=@gJw-BfAc|K&vN_e`=SQK^
z64fYXDwm${@JWs#Z%)4&4bVVp18|>9B|`}R{u>FPN>duu<KV!&u~acH5>f#qQ6#}S
zEL}B4p!@w?zwPL=*b345cU~%;R874p%sMEQ(~Xv@dc@%QD*(u9|LaVGUtIbrn%i<l
zd}maX+7wJ_q}}CtGk3UbX@HB2e)(wN3M#=0x4FOAqM*#cxr`6%5zm^GDg26r<{U>D
zUq^>V&xnA$TVy2v4Z4h(wfnxnvu0+m)Hou{M0^gtBG!8LeRbqW&s^-GdrA4>p)Ob^
zQ5?7MD-J!IAFwj>QcG5_opk#=Z2~}CC;;ES`ow-0iN1b1??D7%TSp^+KfU~&<&ToC
zc6)t(V5m!2F6FkOEyuCn=q9yVs48E}Vm{^)8I0j@n9LCj9{vEyNpd^PqUGd|o#o4_
z*>V1<R;#ZkSgN#9WKD~fa{nh$(SIUVo09vl@5^KPOKJ?Cz%B5^2PTEzn0|@wmm!k5
zi*zl_w6VDR<bPcSVCjqES<P}I!EEv}QZ5Rb%BYE0XS)UMZwP3c#i{x#wQkGSkB>LU
zTf%P-82|Krneuw2U#UrR0XZz5T`C_K+Iyi&tNzy^yQLq_gfr=fLgV0?x1tDz#iq!>
z+6=C+;74M;$|}r?;^gmj0cYjXOS<B>G&5W`8`3sMbLIUfW_Q7i4p0PW7I|MAM|62L
z<7(!e?LbZE*wi%>Mq+mXTwYMw;}%fqC+$EZIX};In?Hij_y19s{WC_GAOU*gfqf*8
zD4ArE`zVuB%EV77_Te94PFD;dJEVfksP;Xl1B5bF$s&hBk0YT)w&dNJ5yR=iy1W;B
z`@U373l-1lM{90fyw$af^+znxeO^1!?pfKvqeu@8!v{Lm)Y9K!YRM>X|Esdrithbg
zXL)}rD#sLE#1c(3xFzSar&}UDHX`rwc1_NTZiTy;c9t4Ruy0Pq&cEfgKf&#5(X|tk
z^|rn>M7%TC7YC>=J{RTQ50g7x(F(dvrYGhJ4ajHFX?UW6qS|fmtC!Kpz2p2hey3p*
z{h&_RD565F$Q<r#w7lx(wa+1+OuoNhsvC-~oiLlND<p<Dr*gd5(Tr(HYTKYI%}`%_
z(IdGN)pQYFE39V0AMCl$B^Zb>Podwup))@}{lJts)^&CGmL9C6vOk%!r*SOg+qm81
zu+A0pwhh0#9mF0|H<QBpG4|_Tqtj7|c0aGi>l0SQHG-V663jXIQ_}3(3et{JN9U7k
z<`5V_{L<cLGHWyTis8=Cj)NX1g&*RIpZmnO7IseZ*88ti&3p;pkuKD})0nI|4Y_>Y
zpP(xhXu(_vza3qA@?=<l@p@-GHRL~8s{3FR;Zz}^_oZq!+fKc#%Hibd$sXBQ|Ka4D
zL%*pt8C4vX*yp>@9~o(LsPn_lrr~3><k+{kS$3<9WYR>qL*g(6mg|NvYNTp4ZF@vt
zNGedVw{fWi;S-5px03*-#yGgnLdTU5J0!>jWsn@4gMwZDt};y3mQi{tc}R3nXW)4`
zS~@U(sDPg`UpiBwwHqKK;lZ*qn$r5Xdx?DU_cT#W0DP^5pHDujf>J?&V*J5F{xh{Y
zeE4a9r#5W@HXfB9vhlR9xM<qPNM};;Za;5X7~L<*6@S$zM(G&*A+aW#-Q83J$pD`o
zCa-pKz_ZCY`Kb_-CI5w^Rqln85e+$}-&4uQBDbKw(9)$$zH!pzo?=wtGfV>)I=W%p
z!iO-_7f^sszVYlxTu~*(mK;&+L^>~TKb`pF$9Bn}6mnf5K!xi`v;=cHZ0Nbfnk`nn
z8EK+w`uqEr8F%Tb6bPl6Z@b;-GTu??M-{)}&*UiCJqNs^)VA>8Evr{@-}}#+!?3*G
zc;Jf}jsub?IHxb`O0?OEN?Pjq#$1j1eWoN_g%sudT#;KG^E_!9Pmw`J@X7u1ld#S?
zy?s0=6&WbTcr3b_55&98CLva9{od*mZ_<VYd-Kikh(hnl8v5F|HbL;n8U1O_uf5Qu
zeMKcb(yMqw`R&sW*x|FhUj|KQd3h=Q17VGd>KyzdETb%FdIN}O+NT~EbPIRD9UB&1
z9@l#z9~|=%ME5@{>tAhclENbr6AtA?&&4Xoud^+PVQ6_x03+Oa&`rKqsWh572j-A|
zqlI!&KHvXfcz`n76p70=;<}+9;LYzZbC@bo32@0^h<P59hoV)Bo~A6`Qdn~fyaB`V
z<wKPZB2sNg7Gnm|^NX9v;li)`*7>t*;9jDO#D6HzLW@a8#02b{%(EWGCz?VV`vGfX
z^8X_w8yQ6Mekdm1faDGwe|3JN+hgs~e6mZ9+-lIhDU^DVajL&3OgqbD)!j6w95>XK
zbKjGL>M0CoFEuAvCfBI>L$<0>drKdTdc30A<`h9K4)j)4{un;Kw$9xNJZ9}4xjRhp
zZ;G=x)3q-#&!ZUX+dwB(VHQ_T+8KtG`G!D0_Hg~^%UJrht-;ZJKoNJf)v>wKrAP8E
zl$F-?fMNYsn3euO-{YLykth@vC7K8|IM)X`OX9DU&$8S^zeL`}2toeT*IKACJ4>F*
z<5Ycb1Q>$26njHUR7Wd=+?=)Cpok{;Gx7okPeK#LjXxFIuo2xa?7T|lawcb{!eOf(
zW2tjpsow=Hg@br}IN-1C*25wII>-?W;q7)6%od(=JWSsIaB*jlU-NetMl9FgDEAkP
z#?A4JPd&WKALgmI&#OIU72&fhR?H5MxVkwP1q<2d3lbGoaH33eZuB)(pO*2tRQ9P{
zz$%PYkV^F_<`?QgA*!2$=EQo&@-Z!TmD@!6a}YAD6xEBN)*u$Lc*RU{*(w0{lO{`0
zQEH157W^r+oR*M@%-=e1-9d+7#o&I9uCiIG=Y{lh0KHY!Y=pLAS5Xd`3*g^lxbPj$
zhXoAm2~Jgbln@gYA)C3fc(#Z8%N=H8#fV_g%_kM@q(K7w2n7{t8a%z~_$w;x_N{ph
zgTd*zg}DjK3Of}_>SDEc<)`Y56U|1i6UuZ2h%%L_Vs(53>kVRtuYQRyNk-@e_3jur
zktVo(HkHS2wTvwCFy=i=Tz!MwR#K9!MzuD)+lEhzCECA>LEi)Z+zT$GNC&gr6*U1x
zuc$y?b+G5<c12{kB!0e3SquVw{=2S_jWC-tK96NJC0XHUzsSfE-azHMH-LpW%<Xgb
zGq7lWrx<^X`VNC_4zQI_;jLDFv=UnUe$#x(Nkia=%(+akkG@G!45NxNT8?T8iyjK3
zLJiCG*B{EPcwLOyQbj}+ym=z};IRqr<;J2MITr##`lJ~TDCEeJeQ18gcVwU5OYQnn
zpcFHNBtAO05@wGdWGNignJu)~8$g0Jit1$+gV>FS(t?#*X8A;mVC0DZQO*EGt_HEG
zrrS?pHPG+c10CvV-U1^Yy7YTIyU5@M$DN@W>!VZyUqaV-1E2tg4LHceO9#`10TbUq
zh1XaKClG4)>02f88)Fqs8g2s;68v_0pY$_FOEakdvrr+C2LR@m?~aA9d|dD2?Hm!&
z=Qj2cTCVLUMDoBmwB3wy{Y!@1j?H9VF8!B<_gMg5(#X|0c>^|A{!WmrI*bCX^0P*V
z2hV&qp1ftbKJ%l?c1Iu8!5sZQqqgPo#BrU3$djgmD!AeQ%qh1NY;HA(<`r^%PnQR&
z!tM?*T=o-_E}u;-@o`4QGw<aSbIN}_R~q`hHpTWnKll!48S}e}E2sple)AenaRhwe
z1!oUWBUtcCoTIpTKuHJClM9+qu*gu8zu1n{_qKDR4B4xzmiXCI&hH#HpcVG#)S!*S
zzX6iDqg#^&YpyDfog<sp%Q(!l%^a}^t0Cl$sv<<-2UiM%ZGQ~yxTwy-#OSh+ipk5o
z@P(Ck8(#X%CH9N&^w1xHC(*v)AF6K!LAXT|X$)g2XV<5{GrCFI<GeL`rXaH8;@iyk
z7m59iIcPmdFCL@i*sbtz8XGAaeXJB)+Lm6+qL+<*;FieI{=T$spN_r;BTEPFS^NI#
z?7Ypm;efC<SEj9y%>0qxJi%J4!mm)Tl%YW@SKkt)Q2?<5*2j64aK-oHn@RBq-@xI9
zlB&G3fBV&=k({_p8%Hd(?Ks7X+2i5`!`#i?GOQV5!^m1zog!;2K+|T_d^DQSoU7q-
zk|u;8XDxO$n;+%J>v8BQ-$lQR;t6CuDtMnsB!5aa5r<O}kQ6U(S@)DpgP(t%v)l<L
z=2ifmV~1;4&rOP(Fxj?m@-rM)nAC219-ITH1cCZS!FzFas`{J-BvsIi7Q0sN_pc{A
zY%DGnj-WKl=$LUpv>^tN;EkDbi{Wsr0wuI^X@(iXM94~ngi0S&2_kNVvpK^YEQckd
ziHa54SjKRWm7_~CT11`EQe<C35U@v(=I5GEn*VtXJ6um~Am73^IQkj0&AkIZ8#JSY
z85t3!1zP<L39{x#Km1h5y)4X8@Cpp*ih0JvH8+Q|Qq=yDS6HKQ6UJte!@GbW!+Er<
zXWC=)A6;k5>GTfZN|Ye#X;cY267Q!98vNc~8Mc;;oVW?&8!Y2OMwD!qW()~HW#iJt
z;?DSM)?s6)H_AWPySzf%XHqS{%c?K`11krEJ8}=(#rGwur(dF{+nk1DNRJGWY=yK-
zU!=v|+H02Bi7v#3+fTe1iGP17eWx(~AFl3?Ih@dgIO_>!#PDg$@Rnf9Ya1UI8teW$
zxb1BcHgES&C+peoULFr~zx}iVC*}e1T>(HC#zl6fV-H*8;7VaEl>Fdw+gBp=*IVh*
zN4(@mK*mPL6^E1zM16{52|cWp`h4nerthKbKQ;f4ksC0MNkn$uuX9Tjbr3iP%OpGX
z1|ctW>qzNpvLJmomVy#`(`CXA(I&r_4-TKG`yCyLAI^@EGT%`8V*cB3{p8Vh8iWon
zoscJ&9z_(VK%fAXvBjooHA{ou1EI;s%x!XG0b;^JUta{>bH3wm<Q;f-g@<jQE|2EM
zhA&@TPc9(*#An>Y?B-0Anp~WcT5x}x9&p&5ef!qXW&bUSKGCsXaWnuh!orie&wXd3
z=7vY^B}bY4vGE%wLHY8duIw2PW|rwVasy64>R3wp+Yj~^8Y=!D5H|IgIrH`-u*ITQ
zE4{;CBup%R*31UO57IVnncu@Lg}qt-67%8360gb<93{4He>iMD^YjY?TvLjo3mo{Q
zgR&x9%B<$RHZ0W5r%0F48HZoD>f$+GzIF-cx=a5`2tx;TTn-USoQ3%mG0-)bkZ_2K
z=K`z#Sc3Emhf|Q;1Jiq?(yWvOx_@xGH(MH}oYhk*F1y4Hh(k&}GpY)FLjf-g!ozuW
ztpYx64gjc#)bM;2`LKzoVk{>X>{bnYyyYPqjXJr8B%r8-jPE0Rlew%;$KLh6R(GxL
zr%j7NGU-4Nb&wP}clYZN1Ku`8vemSs=+OiR^<mt}krCIjbPx^UR-$wo@b*?VYswxw
zua1QM{^#Dj-j9YnGUFM=jU3;|B<F*<8}tQ@sXeqT2-#yEi=~wz__POrxT>_V7I8Mr
zH=M-hyz4IFBjBsZ40^3{0^Gw8ZJq{xZ-9g29c%JuAw*OAF!(3SNc?~bvg{BwXCMJU
zmC`jD;?H>W>!n3S((?{t$Dj9M;stMi8&9>nSWnZal9$6qBFI(gb>O3ueX__9&OkoU
z53D)ddQblxne-=I?G*q7h(n5?3B?QW3F+Ba2v3TlKfg$bOH&8(e#wGJo?_Ay&E^pk
zYOR&o6?rlebxQ1y8x2N)8@lheBPmo8N{w$46@{?YC&Ax>Z78tMU+yAvQxEy&D!Og{
zqoIMKG$e;su}kNhO)KLwLGtTtjJ3JlqrgjJ?UDPDZj7xASzHe7og%VdDiE2>A+uOu
zLWAf>3)hG2?d<-h*c@CE9FD!KKpBlB(0K}z=Kesv$J#CJ54gmXEqZ;0!{n`btrWy4
zK79DP(drC-j;1V~&piGgt4c4@A8I)Dn!j5DM~rk~xS8r*my#U5vS0YMFxQj|JMYm2
zT2mhG(#^N?mGXbyNu{e^X!9ar#_kQ1?@p!^4a`>SoKaq2;{yo>j*?SUK+0s{#>pDv
zp&`tL)ttk2RSh^KXlBaB!0$8^kot>EhTCjW0ugnNemc!^>ju?HW9F-vN!~hsq2uRW
zX6NUpmkx*X1e@E}16$8=UYpLDLk`X60@4pXJRkhU^(h(u*6#O0+A5T>FRBb+$2{t;
z)bmTFQ1PJ9U_)%2si;Kz@(08|5k`&$U&l3%_<uSg=q!@hP5?Z}gr9Ag29sG};mOF*
z;oly(aMLfya5Ha7lXUBlWmvRP(W&_NT@vYc+6jEjI}UT^8AaNhl~Y-_QkplJXlSe8
zCbAIjK<sTb7Y(fc987w00x^LDIaoA`nRm=|^b5F*!O%{1s**=kS6N{;*p{lXWP&K5
z;t9i5U>+nO;<CWhZMu;iYX%h_7DPvt4B%r4N+YMuZHU9^7P{xJgl&4Rxhw{^&zWYN
z@mI=AE);oUrvH3LtL0;A6|0sLMWRrL^bL@mBM)F-CQ1*+znA~UTo1tOk*oZ<C%?ZJ
zDr81{O;hp^f1bIeR9VBJKR9B*4k-k5YAH?<%aY4IMV<^CpyW{`*@C|y@_69Cw<%u{
zhpW^i7ZZoV!LkZ}rdEw!fT(TeT}PD3i<`XNuGqu#Md`(t@c>Fb_4m3({7iE1DNCdB
ztcqJJAeGw#1_obLD*yO=*`wpP^bfT0571&S!fbag_f3J}PsPMi6W9M3Nw}(D)dNUq
zP;1N{vp;#%Z#`rkQMV)D@FmsMZ4r^$uXRKs`i>$EY^9i{oC8_{{{-rR4h(7=sTdM^
zrv55}8DBioM1bZ91}-IJV{I1|gCxBH%4pjr^rRZKaupHS5^)-p8N6cD8+kx1<IL@o
zu9j^oDekg#f>bo5Uz6qDTh5fT@HQjMiadId6f<|f^dI^gTC!INhKqT#T=mNPpIt<i
z_Vt89iq72r19$GP%*4%?yPZqlX<t1ayLX#{CB%TiHMq9>>QXKH_Cpl!@K~}h1nZ$~
z|Bvc(1d8Q!fBD(d>r1*nYk#N@Arh5^(30MqU(A^FPuc%{yD?qhfU27hj#%dN3elbx
z+=hVr5$K@!+7~OY+y*+z;Qpr;misv{2@go^z28V8Nx|R5qt^uihdfTM7HJ(FzpRmZ
zt#9&=m9pvik0&eUvAVJhHIew#{c#_Lie?OryI&S!8_wM0!t83ZfY}s2D|+Gmajq|R
zE^U?3uz3Pzk#~n1)Qf2_wd)bWXxCFF{_Yxux-u2Fs>VD~$-^&fD%UPpmF_f{SK-^k
z1W&&g_=Iyz9Cm5T#TM%h7R4hT0};TQx#_TUt__d)rb@ut!@Hfi4yW*g%i%=2gysq3
z1zP_HU$?@GoNr-rWANAD%*6H$>c5EDIX9Ki-s_uIkF|wLAEaBGR3ANCQ7;#H0$Pu_
z`}{;^7iU!ZI&3)sl`hBXRNKegs5z}t1yYh?2;I@?GGO1}GD~IAGrZXBn>m#vwy-(`
zG`WWfNuSsL#ReQde@4V&0z>3Y{wc}cT}NDnd5iyE0a_YM<cko%e?(N2e=#J+bAyp@
z`w+?Bbr8)7#ba%S-A2ltDwN<ZRcf%tVWCRwj(7tv7Nj(nR$>`wKl~FW^ZVcF0zA}A
z0v}K4E)oJAlmz3XOG_1V)#n{2v-ohDL8cZGvE;&J?G-FI%B_Hal)q=;v+5ea6K`nq
zV6r%`YSsZ#vwq;!aRysvT=A^Ci|u0!f0n7UUZ|Qjr8WTW2Ip4`KETZU$v+IYwA?5D
zef)^Ul*s9F6J&69la(~mHA?z!ive1d@82?q?{-_|K9$1en3Vl*YId0VmTiQ9PJ)!{
z`&iC`ibYKeaAw!$P2a)}wMBd8`I$7oi=-tXL^N7Vi?;rXNesIaS1JCgRyN&cwpgPl
zvJZ(LjUi~S$t>TD+zx&`aQ`rWy6K4hT>Mo`ov<A0?%e+Mr->Rl2sFEi!c6G4KapLL
z!4^L$L9?>&{Mx{;DTd2)K+Ith7ANw!6Dy36Xfk^KZ_NDm-t1>uennjelh7g-mIl}R
zz=>2+e&@L6VBB`Fhn{ug`C%4z`E*nAT$uzcu0<Zk8fB1|epsVjR_=dn6eHD0K$G>J
zxsR>DSqCe6ZEf2fWk$2$gQFnJL;&o>w&NZxK@{on;FEf;t!Ztw4&KeqX3yKHe`&7k
zmWVOIgwWs%DCNa+qZ<m{@Na(>-_USbGdk}X7NY;q04i(ig%8;|C9_axjTF@IJ97@e
z{|tv%`(Lz}UBa*rhV}2^SNIDT8@VbuLd1!oyj)v$n&91pJm*;E2Ia!-2R<nTg`n4B
z+2~wre&>-F*wFD5mplh`Y+pCGZ=~?+6pbvYu;vMJK}6i0K+caoE*y#-*4sqJZYXU|
zp1C*8K_nK(ZNin~?if0+x8^&pF2sEE!v5d^ubPP$!R7ZoByHT~lRirK#f!}IzpgXW
zdNY5^e-dmf&ifEfBK?uU;VlkRaj0Pb?uL*tqqvn}o;#<qa@}CDKbV#bZlmv||MUcx
zQsl^11y3LOg9a`=KxF~i_(fZ%Bqn%3asr$|7k8e>(d?SX=z<cp!vxIuE}f4d>eL%d
z<I>r#YLch^)RYn`QPOIU&30=HRg%Ef9sHX2^=!4x9l8ToSN@IK=oHXlr*@N#21ac%
zfD|QFVwCkksIP!2zMraSlMvZ5NDV5yjDdq$nLf?zdhZu6k+re|_e<&=QK7_h(P678
z)mkF0Vo55!HiZ*+d{vWVkt?Vr)HsM8Yw2jQmK^VjCBC24pu4IPB|fQ&?rAK>M*z;0
z5IM6XGBtkM8SkUxNDO&{j|KLEb3;_zZ@_v!+i8Xz#wIoV2O!<mwn^gKms{-ZbzCy8
z#Em#$YEN3Di<l}sC?U2bPGN9NDl&kip7ms<l(5&h$w*nw4G^=1l17JBfg-6&K9bQ7
zN6Lv8VU#4&r%LX)9|C46%z*pSTwl?f>buMQSLB1$wy(b{F8lab9)O@XogXGzpulz4
z72AFzQ-)6QBi#<yYBt5jTfX>uiNy<5-l9*8v8f8#{LpKP7SU(P|H5ZnWEb~;rc-v-
z7JBikh^7$Uu*PE{>p2Ty6?0#Ys-;<ep9NU4pY_+S|4~l)15qfw=Ng8$Hhn*BK3D=3
zcEMGbMs|__*NfQADC3_kka)4g*9kODq0>nt+=>oHlmo$BWZpwaf6SzvOx7l|l!5ee
z40JAq2Kn`B{j~sC9-J>f)7)v!_eQ!hIT?#JHu3y=C(@htwf_r&v=Ia-3cH9+=G0N1
zV{_#Cs6+Ln2fLl;dc`L%dar*Xk01@t{W1oG3W#A`Ue}UzjWnm>xbJR|f*AR%zY@?p
zLBK7(FBh>L<~iY^<*GO$0#gce3E)I;-tpS3B$ly(?ssr@U6Afv=@kH{zkC2u26}_Z
zv~ogQg2JKb)A4;Tx#^yNzE>-of2WN1<6V|Op$6lNLHj+z&=20vyyc7v1%E2D_|Uos
z5Q87o>gf{{B=+=r5GK~v;3tGVg^PS&3@*s|sAPtO&|tu8XK*IImYq<RvU}eOl<VfG
zA9Dw~aObzYA5#y&EjL^ciIjMY!lgysreoAf7Dk;k-dNsnC@HHVP$XV}h9nVHzY=hW
zB_MTwR$7(idhGx}@yjc~D~h&e<GvLXi!f1ASmB`u<Ao=vhIj>|;1vadDkX5Vl1jGl
zmmaS7CG1GJ`6LbijE@rdph&bblbE;uR`mmICoKSuS|Y3^En?P$8$3pJqy`?~_CHwU
zH#PCB92UX~Gd1ZCCG(Wl(fxhAbvE!F^+%LcEExp3x!}FoHyCjbquoexpVT9@IwOe>
zO9BGOozqJ3#m%D=FF?c!r|pC~YN~trX574HA|LYm4mf@C_3P0Z_Ik~wcG!NpPuedk
zuCH-7<D>$vWiqz~452Hrjdfvj462+TLAKUA;_V739rXdKvp{tYiSH3W{WUrpn41&@
zGS*vc$<<k6D$<WWPXhQKF5aZj|DMx0itpYYQ07VTHsWs8+UT9bTZ-sTK~DQAD0n3I
zon(ZcV`+#9c_u>7b7nTzz1HrrRyF16-9X7rO*Gpq@ddzs8SJ)xx%Cr&VSHF4;^Lr?
z<##6^@&@XDG{(k)&Llmhgox}r%mWf}ZMgmaq6P1v(O_!3GeR2LOCriBGLp7xg>E%r
zd^<kwvj-c9o{bSA(E%RvtIQMgT$in%`JTRtXI7>YRn?1?n=KB41{9h)89&=St~;;K
zKCn**eB+=LxP82H>wA6(SLzM)zb#YN*Zl6EJNoTqbi#f-WoU9TzVmo<3(ECYJk2tk
zBGdcMCvFZNoCJ@e9kVU$s+RSUp3nrPb^G1WdMnDJ49>sw1C8B^_ESeZ22s&WbRb+w
zyhYxH;vr(r{SagQJ?extqnPNmdKrut>JJ`k^oc-sNz?Hl5#6OLOh02(3K-`ytoY@p
z740AADM~F4e71eM`NHL1togzRYBMUuCXO0iA$dsBeSMv-zIxnN$caFV1^(l8y{cqu
z{>K&Nt;Ptf_}Gs&n3cf&PONS`j?~Ff|89_4Vi_>W97Rxw9swHIitEs0I^EXqY>f0Z
zrpcot8R8b~Dn$+2JunF!Y*@%CMq^250Ih7^BrSVr!tmXEGKc#otH4JiN8R`8a6AcP
zEcBQOM#toS&-oR9r`0xkfZ{&3%BZknc@j%v)2Fx*1&-Awm+$o~k*3WU)jv{-<I5p9
zZJnBRM)fvYXRP|2?&7A{Ut(hXK!zb18{7WbQ=g_a|3Th!HF^R>`D!rdvBIxqkuKXm
zkh57UN$a68%x>~_3_f0b3%gr>vA|bs?)owi(l5_Z?8Pq}K5<iY^ly(sHs1Y6i?$vX
zGI|C!0gn4Ai*iAbXHN8j5N-bx@M~P;lTN?t_0#bIf?BxRo|@bbA-L0)&2UREYTw;e
ztn{00wElgyYSzq-9-O_ytAI!{klSye_L6KNAMI2ChI`>B|3CR4A~C(IZA0YlK*;iI
z+=tkoI!+uwNwL?LyR#9b&c0i$QRheJo=1(^9M}-_VrWwq@4bDDjn7d(eSwK`^B0fh
zvMPCZ2ZwC+hC-_!V1IA4=o_viBL$M)D1?NrFNqX{8&%Ltzx)Rxdk+nj0T0lHYWF>F
z*FSJ$_+7H@V>CY<Tg#zN^!4@J_hYQIMCN{z_Hi4*ELj<}x_f!`f>Kf4onC2lELgVL
z^HT9RdTuBNvXrRyXn<T@yH+QOb-N<;K7VMt-;MOh5~Ww@Sla8`2ey{;WAAG>d=qgd
ze_8dJ<8lU_*eS}0_FHp1%v(a$w4+gUuitD|Ko;$AxqUVRH3ivgRPY?$%ZXXFu$k9k
zLybxgD9TTpLRhk99|B|gC%fP615GQv;IE>Rn+IP2^48A)hQcyfO%F%AIV4EEC9qO*
zo#H*5WB4DQ!`lvLIQ@v!<_iH;m|DP)xVh4;rO_GrXxNm&?^j0YprBCJ!Rzf{F_Xr`
zNWeNoWv777KzT9EHK`8vcU_iJWgo}YqRckak#u4Qt$E+fYl!yaH2DF-5!Vc}wA<fx
z#{H*SV7R1HYfcINFN5$CfLaZuQ)7Owky!e{1DKMvd18JrKnZ&OoqC{jERs2^`1L*Y
zrqlw<{aZvbm+38_uL~H_5&uc4awPGI0LwCUL`M9oS$3-4=>&U8LEO`gKikmxXV4jH
zy7m^J(8I}K+TL}z`$7c|yn|cVtd!-E%Ing|osxA8#>DY%<rtC!(t4UE^*)R;#Hi%)
z+J{6qG1>esG8xQ_Np_}19)u;yj2baxP*|3Q4Gv;OR*QWV`d-!Uy#?>YmV`=u;`6c@
zxiqD5{Y;-eMM#*SNKmb0+O>nFXcWNUKKgp|oyaWc^cic762)8=szeMDUZqT~S$;hn
z)3KdtA~j5=IHb-3FAhRJ;k4NQSoeWNnSv8W_}Nf$b9WLg9c%o>`grjvo+-oQ!m?4j
z$rv#ntCP*Ch`{qELN&bebeZqcBSiQ|1lxKWXCA|;?`81hg>f?->%&Fe``4FqclRUu
zXQ2V|4pX+e4Hy_tczqM7|Emp|!{YVefbx0L4@hQ=&K&?Tr~UnI_KP9xF4CKv?8gj{
zAo25k+vC|tJRqz4!GPamy-1WuvC_QT7sj!=(ir|tA`(0QtGO{Y>#u~_mRi6J8ZN!H
zV{8yox?oV5f}?bd*ham~tg#sGO;&SB+=TfDdu0P*^bQUAGL4o00z4mJRS(1s)xLuG
zxV330C;98DEQrf59`DZQgOlUK=<^#)=;<DBae0OR>MK#180(2R<t=Bq&DtMrB!lh!
z8+75NZ59c!SKI>bAB}d|zhr$nkh59AV%d3xuo#TZ@$)c{Cd$;n&ZnC0>GNu|hD-OE
zmFWKD=T`Zm=dG_4Cvrvs@(^Mt1ji%!*HMIoA)&%?m%l*Dls?$Zh2zToF|3Xk)dTE}
zO=o{^1|tbdSwmThii-nmbS&DXQ^%9&aV(K!AHvf>DnkF(R<${-a0>>N#KN&@=V@)7
zI-}r);u;*wP+Z<ZOOC5VEx*JYh)UbL#ZJ0&(V|P>VuSu7(mL3WB*jSy!yO;gBq>hK
z87wv?X0=@Z@*B8=ZGVR?46KF5<1#QdNXLWXgQ7jxJ0*d#NiUP92n0-%(rFiZ-9)T+
z1JUL@xmC}rPMyDP?JXWHZsa>&R6;x~){NS+NOu!aYNgTKmm4ih;D8XO(xXj^e(#NL
zAD34p#5Ji1wIX`?=>iF{M9EGj)ns!R`yW$IQ2%eV2=ET-?e>Ar%9d1^X+PsfoU0|z
z4hVQaS|C|di<JsMB)iBvO6|<R5I*U*bj@aSt-{y{2)@gY81h%l+|zg)ONIqXya}fM
z=Asch`4w1t5io<cYY0-9kTDFFy|VfFl$)5LD<H;lyItumni3+4n@t&sv;Q51qx>QN
zkkRr#z}B)mX{c^YSUfQ5_ywrFQfgtr0f{+!zWI=&d=9TGHBSj8iy&iVYSh*5j_9rJ
zArgK9uJK83NuOd!C2txZCsw<g8;+o&m3UGKbjRc;99DUpGXzi{EzO?1kwYS&dQDbY
zl{)+iu+`8&9;`2UQ!eFdS0d^apG=Yp-d8a9;!*B#J3ppIK8QCGA65!?v;!_R>XWR=
zmW7V4N1b+tlX4Xd3DX9o5I-rcO_O>9!N=7PmNw?|QKRDKbFaQbvzxhECLWIR>Cms}
zL(=<vfx7F{#|p}}`7WdK^*3P*toEQ*YAO+<Qo9-NKD2-nnhPTdCsEQscay=|usQIq
zq^zmh3IgGmR=`)VT-3Z!PCED2X!h}xoj%vBG<>AGt_)DUc?FqF1E7awG4E32thgr@
zz5FZSRPA^Ab(_kT7_KVG#0k1cz~0XFdo7QYdy2%RJ|(*Ob+_S1ZZ%hypAh>m{%4ik
zbths*l{8xVxbIWF@$^xhBAg+XWd~pb#<XqfcewrXJ$vhQY>uruE^~mHGt0Y%ZsR&w
zHPm7+oJLteGAZeF#iDtDE+zf3MR~nSU2J|Kc&_gykadqk?(6bpfo<%(3y~%^L9v#U
z$x8x)THILGLcLe6IU=CXY%Wr(=2r2cZSqXFUGF3Wen`oR+}NuboDHEz%0wNPfC!g7
z5dU%JL;CBZ6Vrc2Uy21JCT)}6?H~R29nr$~L!RU&F3%eJm;Dpjk9WR(o>RXsFL!n>
z3-eF`7+W;Z64Yo*yhKaee&fA^M$ToGgurcG(qE=bmNhE<{Dny~w)0*#2;dv;M2clB
zr<y7SY_9C)E6|@EGX1GGserw0FeX9Xb_UyBc>o5Izq=n{Q$a)Uzm%N}+`-CT5VGY|
z&6;T#w?`@z5ho3@=dIjGvgTmLOezcg#Zjfx&6;^`bZAjgCs-~Duf%nSs61`PBvU-k
zgj3RgeOfNv$k(Rlxt<_~Ec%rbXq1r@E^&-(u(O^m8F~T6;ESB-s6sQnr&cOx7@0A=
z0D|85G^)Q++s-3Nr`x|UEX>0bluY_Nl>UCRjGHQf&Fgmk4Y?#=#YmMDhTlinK<gGm
z?2k#z<Q&8Ox+d&RYnx&}E{Q}UT(P1M(LK?>R}ri|XrPzJ<PZmA@5>I-8$&{kFN-4D
zqE+PvI@_a_KhD~2C16QV$<Ou|B>9KQgTd3nu@M!t1Em0pSMaCV2iW_g@|Y+Uq%wPD
zWyS>D&gPmz=Za+k>)<saX}vGegyZL+7J;)?STdi#${*-?HA|8WK=JYIr^mHD=wcyZ
z$&Z;l_IA`s4v%(Y7ERIhKEGQXjUD(4y>wa}cO<`LMoItffA=wqV@6w@wBch#Mri^z
z^e1ha<4~c(4>{8es}wO+`UqMxJbJBhzc}#><!f}Og$e%L>`*g4QlFYAZ~Q#-VNtPC
zJdb17rT_`+@ia;Pxdi;;)@z5ypc}nqUc+Ahzg+e3S^^T{Mq2MGeF@Vwxi&|w<P?Vg
zOU&}X{Y2^nVqpS!SZ~7r1@e%v#U1qxbze%t;4-FMZncYr^4=eoQr+*Sic>-V*z$@P
z!CdOe9YGcPM;uiEyB5$x(TgE`f6b(P@w*NR$OsSCu>%jJcgz^B@{*GX;ID`F7!5*g
z7!MAdt$`gD^|t(K!)-8}&N?IfQ+j1japibZG%_bojs7!~8in+ht3s2IxWOaV6%$Sb
zVpXI3Jt4#>l%ga~V8(%|?KzWRFP5wMi>JOJ9||%<O=V>inGEExp6U(f)HeSVIyO_6
z+dbO~fE7FYpC6wAb*xjrXDjbSw)L~nht*cT4+e#jL#zTo!C-8sO=qc-==ilGX%=>|
z9<_w<jcXAbeL|_V`DH?i7!an4{#Qz-;w%^coB1xs#d9-0RJg9|uNDu=EWa{x+&2@1
zACM{|9<Le&#iL2bdF#Np-fp?miz*$ofCh7aq*-lQ9SU{kl=y(A2T~3W%8Hz&L)hh3
z*9B7ik!{!Q-9L>SDzx)U6)2_Iw;o;k|H5TiP~+3>x*yE4H)6yV@F_V!^x5HI(wFyp
zwMG(kM6KEU>BVSLqw`E9HXTuM+CbsD7QT)zVL%IDN9^b2*miGB)+sk=U9;089ZO}!
zK+lMjFsgqwxof2mRRZFTzW_@eko&V*bf$@(+spu|D5`j#Id59jkDMh+(s&7?w#6W#
zM~V1(REKLx++annLJWwIsI3@&B=pLfSjLA>5EeL}pkf55Wz=%XNdEDNPX@ve|4iTs
zrDK;Z<po)MjH!v&9zPzIRsSmsiinu=Fn!8GfHn>8cRX;B>ew)D69RJhK4{&Iv-SjG
zQA!>}q{MrbzXH?gOz|SW$<c=M1Z^>Lp%;QJQ-jM=Te<V$jG2~ky`Ts-<iF&m2#ctH
ziI8r$uJ22MMr5@Pv0v|q#IIj>keeqw_f2SwO-70G=~p1-^K!p<@7IpuL>4cggof&T
zs=*L?nG59vfyK0z9S-6G|ET9aO5sQ~TJHStn&5qo{&rmJCVsO%trA?2WO2R@<^9%0
zZZ>%fq~1YyV#F}9l`Er>9*Ke^KP55I`ymo&(Ps(tP+}64p6h7T&?&VyuIFylZQpYK
zCze?q3uw59js27m@nagv9*G%Wv!w>7M1eWZ`n-_&<E(2vAc_a}wcQ^B@swIViyzBA
za$-JAAKZBC40%slBF+HMWcYJ4v}TX~tAhCglW>*(X{dU20ViAL+`!?($6D=b?V<A?
z0M$jj>nxO3{!$5(O6k2q9af=>l^B>!ErX;&+>xbu{wK$0P8c3pcRl*J_xf1-AJO^h
z(-F-XY6FR@@_dbd%D-Pk6_GSU2j`UHhr4_2o?fP}1RZIdQD1*(JNK9K9o#{<n}JXF
zZro;4Mc=l!$-j;yi2?Ca85kDb>TZW}d+*Tn(qN}X&4J+(0R6rMwsu;obW~ATMH8?a
z1KltA@{BsEO@m<$^r{=xK$Pqh*H4d86O&&SSYH=v20vrN98W>dPr<utbLdV<<7r?H
z%~4bOj+NA+BfKjWhQAJZ1!(l&ah-rkPzS(F8)OV=Z@vh~_yObxGBblun*{3hHFJCh
zkTQ{xL)$8H<Oc};d>rO4cq(r{(0h#`O_@DSe}|?aHbn_f1-;CetJwK&Fi|!Ke~Yv+
z*foQ(M;?7)Lql2u&|h9Q32H@sVf`{zz=C4gVlZ~=0Zf%!0IU%)_BtN^*{<tG-9;qw
zRzDgcFuq=BtulCddoXefFgKvMl#jnW-60ZD7@O4#$^B7ZHU~L1X~`|QE_o<?v<_fG
ze<(MQwBoqgviK46e>n=w4;L%;&}xEswrxk09Oh0Oe&UIgNgd5Vzrg>;)msK+wXV_H
zbT?AcEe(RC(xEic-CZx;-5^MpbcZz3oq|Yrr=)a9pU1uS+UI=VZ%}wAbI#|EF|L8h
zZ4>KflaKs|PJuhSxk~E|ecPieOxuHFoAjm+C=><7Bz)N<e9pqiVGhXs3HIWClRt+m
zf7k|j>Tn->he{vgkf4Pgx5G?-JXAYdLdB3okYFC@iZ7GC?t=#q`JRBe0dia?b^Gnb
zU-)ovS{TiR2RaJ>D?_xgd41nPDZGtCsWT(`lojQOUe!Jua5EXA_%$$lqkg>OJdz;f
zd;MzpVZn`f(9XP3l>Y*!ac}5Do3S}&-=M&Hc@<8M?;{tmVZV5sQEzwGV}GtRE<6H|
zL5a82pOYMGnSQ#=5oHUw{YE+A%l2}0{jG$$GyFbD8m;IIJwt%$Yg}Eb-fz`Z#ZBk+
zFZsdv6}jW5N{XB1w~CJYp0C~lciF)NaYo_DObiVDXR*ab{SSG&x^5p+vY;;MRE&WJ
zruYoXU5y1%Lij-bz!Q;I(Qqb`FOHY-HSG=xn55y`8tKFR<>1Msb9JMpee*FZcGvXy
z`GuzuS2E=qNDKdxaWb=~Wun8izyf(<hf<0oeTfA#ZE!1qMY{n>`vr!kIPqRz1a8Ur
zS^I27rQ80b#9m5iqvV&a`=2~(NI(86Gr_(?c~(98MP`skP^Z4%<C~FG4(g`#{gjv*
z;G%{XN_>DDdwYQ<|G_<q<^1?~rBx(27j<k-9FCilsw9D`VfH0RX;Jn`_4||Oop(M&
z{@aJw>6a{}NU7)50Y3~u5;;pkT=QtQ5P1a^Zwyobs#06y12CAt)B10?h@?^?@|iw5
zg7FgL5PNGGW4XE>pI-R)SJbPE{mXWpiH1z+6J2(?j*gx!n566m{?(ECK81dCf1wiW
z70jXB94q%j$3B(=p%&?1F@y8FWNzH!(vNtfiw+<KF53_PZli@?0c`q@p^U^zr`<j8
zZDf{&lsd$KAVgS@EBmw}VPPAJ?j+LgIlE?v&#>8LDX;|wSv21C2}o{hq+ur_qSTTN
zvW+oLyLU5|LfZ8ZYHUh8MCszZ$ZBFd|MqH=2yzT^I+vwW!MqUd){9*{)&9MoJnbMg
zn7B+;rr&%^Wp8dQDA>$T<$Eib*8kaJRS}R6yjhhoIJavOh`Lx173S(r7Z&PMla(L&
ze<~h}*(kR)7qQ=q1a*mR+GUBbC!bMsV}q{lZy##yEfC@`g-otij#k&WWFq3yO0!??
z{H$6kQ*!~O^<A2v*qKKCD=1=~m&WAjt01CER%mkN(}zmQTc^8*KzPzWM=bVh0k9hz
zCFV%KoN3OL9y9v;4;q2OQMkL<5V1WoB9K83^A9E|mZanR0}5H2?tNxGoOpg$<>$SS
zH2WpG_uA};lfQhTWL3_>#;2<UgSNd81A0xpMQaqjD?e;9E8*a1eS;s1EKEB%TB^Z&
z0F9YWGs^xG*pyGDh!Mowb#gk-Ko*o_m*-w3*pFdARLhJ~vw5_n{EJ>bfvs`~(I*w_
z5F%!NQ^so7_i=f7CMGg<uHi+{#eZI#Dk&{AcS~*S5_OA-Z`LmrDG1)ec?&*@iT6k1
zol#4_)8*yg9e(dwfM%T}T!#A9A@;8Fy=gKM13)*;7K~~<f-P?F_B23RZzTH7A||w5
zj0zBmh$WZkvw|cIsu|0)T6S!R=~P9S3UtpqGg0H%*OlKb$R(%f6tbAn=86@_bA`0z
z7dlq!%oXY**Lh*!@=B>%ZI`%3kek1BF;Rb?@<<xJs?hp|)4(??zi{UG=hQIA38vI*
z*EQc*X!0JfrwiQMznvn*=JZ}R#i>VsZ>(xNkBnJncRSW06mYfitV0e1>-3^3L;?%p
zd+nTze^~%gdxA7g?=Dl0(M!mqzJGJJQ7!#Q%jF^dqgIpb6Bu^Ct~h!cm4`Ql(-YwM
zznLGgNl4==k&==J#Wo(WAl&bau<ShcU^<E|^d>9oxN5N>TKW0uNju>AX)bK)`tH>C
zT}z*q_hZ?>azpf;t>l`VJOV{w2BT_*Voe_Gdyh}PKT?YCKi<KrT(}5-jDHX#`EA;$
zt)$vKnyCztjQrGzOaFpqlCj!P=n?)@*Zk-X?swQAS(G|nS9oyu0Fnt0jI+-FGYj*7
z1E*GTzCrQz#X|=c+0YLyhC7uI)G1kf_*{8XU*D;Gu`xcq7e+X?&<0n!9IQ2`K(Xen
zZ&~E~GKzmEBvnD^ci@=NO?LpSqc^?mY0pSnoR4;a`L=2~%pY4edra_(cRxxc{GnV7
z#j3BCqe$sBcHnXI!<dXWQ0Gl0Z1`aA-uvpGCVvoy8|?}Dcq}r8NRK0dFWbcO;Df=7
z&*23%uRRS3{Cse)*lx1|Aim$<<sc>c!xldyJZFb>saBYMEsLk#qFY%#d|nX1gcL`9
zL-n&P9xtC1_e{{M%nAG`F8)HI%@w{i4=InU2DN0JEfDY2NJK>#?gc6eBdLUvitxUK
zu;*@KqMsUjtjU<H>%j-d4v#FS_>ru4J1ah8>2DCA*%;HA#M>p{*x+vBdJ`M}c%8UC
z`^6No47sYkd4BcV&H4<-1TW8^5epF!ad%NDoVA0QH{DO_J=7E?x$;OW7HUA`S1rZ~
zgZdqbbo-s*V)(tMs})elKaV9oS8fJ{8j>r@joB7$q-uHg1S3xtr098;a*^%oBr-Tb
zR)8iYRj<0#xgA2S+B0@BCZzBDNkz%57$-dpC-~kRspc>4>><BxjU)QzuLe2Eoq&iE
zQ_}+UuFQ*A1X0{&;->SVH7J1N6CYU}_Y5j|;&P*cTP|z5&c+3nV`mPz`!w?`23@v{
z?|@I|3GWW@ZuD#9!JE5TUaX}Eguw}Zj79w((8qg(P*}E#5!-(^=eD!(4Zff#F9Bkd
z$wMB5ofnDz1hRE~{v8<<Ux<;I96EFPvwEOCfT?*3?5^7fQDSh{ID(w&(gIL75@}_k
z9vK$z@wSPNeM^lH4Y}sI?UpsvR_eopraXbLdGG!NHk-ru`O)pn$Q;&?4+i1ecZ8(N
zgP9+#%SS>s`e|5y_vCrtIQ;UH&V+vkiht!RQkk_VGh7>E*oJs(PYB!~aw-(A0gS9a
zG7?OaAd6<+u?DN!r;-l|r~I^pHjX$T4fg{$%~oeiu1b`MIMS>dy**IhJyiGJBX0C~
zo$w*=L?IV%;>%}HP{wx&k4)w1X&OpI#=6f^;Z#kwP5Ex(LLY8@t8`~kH^?=WDNi+8
zICkL}P$?lv6|Th!uW1yn<%&A&hWqE2peTalwsW4SHsmwn_X?5bDk_9GL+JdR#u*h!
zB%w)#|Nhr@)lB)y<<tGaZm>uY2_@p)(D_c&DF}O`B#@w@cP}@fpbI-iHZjxcvk|~(
z*-55&%+*(`Gi$cEvBS)Cklg-cJ_7N~PMexX9F#$mA7Aq=_vQ2NZjUAo8{}xG9rG2?
zV)XrHViMmjB4#;r;ittK{h8+V^I+lRjE%WEJ(7S+x6je59aMFVroEQWWD~SAtaa+;
zA7}?<Rf-&b&sXXsSE9Zi?F?tefp%GasFHuf=eC+6>TTQB{i7vsx0K7QlQ;caJe|u^
zv(`)U4+Yj^LWzyu9DE@ZO9?5&AMhDEW2C#nLi<H?aL|xMCxo7`sE%@=CggfU&o5*Y
zKkg`~YKwl$2mam3w?O7}jJNlJAQT%}1Ww%Io7K!YiehE!F+aeJgjs-y18J#c_FA4y
zm3~B`oQerJBwdP<_>eF59h&&Rqm_{@jA@^i89X(CHi(i%DAB20`jo`V@0?Ax|DmkB
z5(k6HNKy6$bPsJ3tUX}r%I%OuaSwPA=TS0Zy8<SnM6nI7vTZaMHe3r9@^D~eHgg5z
z@k!@1{*=pTF+3_hQgOChpT#15`&7TaLqA@gF>Biw3sn$`(dq4Rd6Ue`HiTd-?C!hj
zv0KRa5|jf6y!y?KrX!2+^yfecJ^N$)ld?cx2t*($M1zElUi^ehM?HmElS)sg{lrDT
z{ddFC-ewdVcX+6&OSps%^R>)0wZ^A|@-Na9D3uG<yYayqc1||MV3~>%*CQXaRZ*}P
zEOB+bqOFC0ji{gc&a2_%uvDSxR_u~gz6*+6^AE7RnM`2!Z|D)iikOJF*icst_uqYw
z`wpZPY~t%-nB)WM7~lb6ho<X-`xw}DFhj-xk<y*vw*6{Shj+Zo@FsIwv(~RO?CwV0
zZ9A9Nc1df0fY^9#W33=UwXN|77OYI2<!YO3%ZZ=>SRX)+P8a3d^j-w5e;C0R+0;AV
zCbxq{6>Vc8D{aqP(f1v~sS_}rkmFhChcXfTg<^-f5pN0?IDr2n9zycZlgb_-Dr2z3
z-I&U^!8=p#Rv_Hhn#U9%_)v}HvPaPYiN$e0s`1`w7tE47?hv>h6Pb*J(Covt;W29T
zQONk>M>_K~KY4y(KTx^|g1y=va=cKeGnN11`Vr~#g*<=Q;<p)+3A+{FubVwEdU$Cx
z5n%~s%o(YCEM}Q)e09Y`kvSvsDbib?J9z>yR=qfXw_bWLx*00-yIW)OpZ8HGHtq>n
zde+6oIK52x*0!uU_Bh5`w@@I*A4M1+J9EV-=5_A(7_p4lw<Lz`*m*ZyolzV0sxgz}
zY@bky^R^@Tc?wgn`}clgoS(q;fd&8ct{VS%KJM>#CnSq=J8${N?_cuY`SntdarMut
zAX`yr3EUsm8Lz$!m(@w-VSUx5iBVY)htO2@Rimb-nja3r$PvfJQ6#N~Ol8u;w&rOj
z6oyVLxdhU9Tlz#Gw-Zw269BwMbM#b<9ZVa6Qf3={PB|G#2eMg+CuhL7Rk;MJG+Pfy
zcijW#eMG6mry!KCYD`y03(8vN`UTS7OL-Eg`th;Zj7m<tR0gV)Rqfa5R&fkBo50?U
zT@;9k(I2d;#JjoaZZ8%dDfskfYu(DzFAT>}Mm>_^Bkd)i@nKSbXMrSv)|a1sb4?#S
zTo;jCg{Y_>wi6kY^Xy883tz(QTU7EXsWT%G)dS6M`VX-777wf3n}`-^0!L3DYCVP?
zwm^nUIS6JW4S((3bSKD=HHcc7+kM<&@e9az36r9{c}v(|!BsX+!0|S3F5{E1Q~Mv>
zK{;^gG@EYxC87>F%>xEzDR7YKIS3%|h_HVGTKv(>_8v>kg^0-BH^g(r>sZoVQO56<
zi4asUX6K(x=6<g8r`<-|(xHN0PaWf>h$l%{E!5z-g(!3X(8P4moax^~1w1nb$^N50
z1RgpNk?oU|k=cN${_HO5&)Xg$Wo7Yw<)ogsJAXRVgRt7f5Wf&=WKb66jo=5_bC&80
zwGtLooe(O|@Fi_IFfF#XMi(gv4H=E3qH&a@w_beo_)_Gxa&dY0>})UC`1zlwGu1y4
z%FGu~;8UBj-^p$b?nVfzSk^X4_7ra>C`3*ssSn)f3F<{d_-E|C)-XEtqolBxygp#?
zz7;x0YCXBr0Q8^}y({i2jhfSU!~^loD|We3WD`*n^m36pU8py!@Tl@FlSQqtpgOtp
zahPMh^8LD4^w3vDl|XYwL{OhH+t==cx&r|5y7=-YS&%gs+ho+C*8(m{W;@to;~R^I
z_;zJal=d5MpR=)j8r5P2oV0#K8vb%X!A}ZT+O#1q1}>4GEt^(buQ^79Ztgv{1>OBk
z<VVs-TVsLcV#|}#==EZ~03W>uq~+l)|6NZ)ca7n*tL>U}vpyWZOY_Y!-L5^;5KhlN
z-%(GzJA6_JXX4(QNxITuWn{IH<O?v+!G??IhVDrKvOEMSV&8Icj%;5>>_XyMH7Vh0
z5MggGjl1lzn>qQ%UA!mF7U*7wzf#ggaxrVwhXd0D>iz))cLsi~ey-b6WpiQOxn&Kj
zCYn@0EqxmtJ$aFMOwuWolPsgwwfjFQk?a9q(Xo9UzSVUI(JPTnUy#>QqqBx(W0nyW
zE5A35TSfTZWXdSt2aER$dYLVf`2tWrrT&ym$B0?MvK<ni$1+**dw!g}G9vXDqD^^U
z_W;lgue9bsc+Rf$MW~FEAxPJ_PMge~L?bSq0#bCjcse=64&Bz_lPF9irdmJ%mR=&2
zr^zsTg#X3MCQ5^ZO^t9{5_sSqAPfNm+_D+je)`@sm=50Sd}LfY+1N^7z@M_pY*bw$
zS=wH4TTA*yBu}$E%3<x#CS1}6L~(cE`jm(SA$07)+=46eY6!u$E1F2Y0aBQPF4=N5
ztKD14c8t6+$%5L_%~^X{%d5kq5`_BKPrtxeys2`7X@K?f3TapS{~jh0U2L5hN8U;^
zw+|OTg{I2R*Z&04RvQG!yd5hQ_^Q;pWZ$S%XBM$E3sJRNnx#Y7F%Q4rwmD0oh3ms;
z*~Td+v{0>X+f?Nr7=OI!<TjLyr9_A{iwf{b1B_2s!}mESgrC%Xc=hRWsA9hm`<mA?
zvc;JH_pyw%gM(x~?B%~bm_IAo+j>iNM{v=PRs%mZyxITZl7S%q2dj)O)5u0|rt?X=
zM;;rW?+rPSC8?=x&Ae$np>utnk@Ghf4UL~4E;6u8Rh5#Rtv~}2+}K6DcQ?XiNkX@M
z3?uLgta5J9Cu!yf*;9aI`rxFtx!P3HKrv{IY{F$NaeaqgXc_bNnZ`pt{5FFMjf|uC
z=S1A+4|5Hw6_VGBR|P*_vn8Vvp5yx|WM$w^x{Qc>1}jeA?vhUyG+4bUIOB*=$>4jd
z)Z){hXLtSb+t6xt1Z8=feJ4)L;F|RA(Qr#%F=cxi(ibcRbJsm8x%5sO>>b~WX6Lkp
z?_2h$gNkbli+;3H2&VczL4LA}zF|=-p!%S*UGsaoAh-0ck7I<kuiq(!%Qm}cKW;fJ
zkR#?aLn{?eoy2>fqI}h{6(i7Qq}XY3G6$x!$~1ZL>)s<zp~3{4Osf7&)gwfjNvi~Y
zI%qk2Gvz5<CsC{|-w&tvWSH&~BOGEeFv7}wB$5>%B;s#NQkc6oPQhw1g@qpvNHjeI
zaSb6QO8KD-B2D)13R2+HF7lW&nK6{Vi><?6*Fr}5c2gbgROOm6c#a^Vpnn$+OLHJS
zvIi#)Iz5w^7E;CBS54z^DL9?4W4v2v`?VTUMnXx*_%-NcUb4PxMnxD{8HW0p7?5_A
z2#TPoq6>SZMsknsxbQE3dO+@N=2#HtE7nykmxE~rUj7|l<MpUvh8Hk?XqDk~e?cpg
zGaFidThTn=Nymw4=f#9aL?aA0^XeIT@_WA5hcxJQGX+;gpz$^u`isy&&oxBg(o3uV
zp2!d5S|x}G5SsThg|?w&jp539_ZI}?d<BkQE@pFAOFEA5qm|Z1N&+2gv$3q-kEnVY
zA|uoYQX3Fhgr2`<oyi#rN3|6vY4uofL2z!?11!%F4ho+Dl+Fo9>MP#$KYw{}LHTlz
z0~E|N!pzD<5Gqiw{im^j7ha#!pE==yl-kS;Ye9Y6q%xMEU>l?6FSzK-F?NaBqahV+
zGazlXnyLM`)`4+%{)3s{o&8J4J!?Xi_oT@0rQ*sQf<c7@I(*MRhOHLrMe6nU6H;lS
zebxk=MuaC~Guq)P>;Zi^SXDCUBIkQ1?|ZTTtrsVgpjHW3bq`+Lln(;FRj>1k%b?}c
z!*^X~<L*)$&8QMF4vze+Ra;Jc(|)GUEZWO?pZT&^-Fq2Qe<4%W|H`H9kv3r8T5dit
zUkxL3RGi53wkM-ptycdrb`!2iG*gbpHo7D#H@F;+L(6QGQ{qckI3_rSd8(SkWxKe4
z_2Tv0nX(Vqen+XdPO>2=cRq3OZ69hr80$wIeL0RbJHU@AlFITEJ=mky?PkN_!IqK|
zlB=xv@@S;s_w%UCH7}@STCII{zwg=ig5;g3R2G>{I1h2${mL8-sc)iguuam~>o0Qi
z`^=g~t56@1TnUqDKD<|%-H{ZrTg8-NFprst9?>df<O|*tn2hy_TFkxNi|QTn`J*VV
zx$7;M;AkeFRQbhY$kc>IS*?Y6g=(XV!Kp!9Am61=wdP|4Ssn}9qBmuDy-ejI)?TT&
zY#BEdnY=~n<UMe-C({uFCX*rmkvKdax&AE{PFa!Dbzz*3;GgxNj{!Q8k_gdc@3b1E
zg-MFEs?^7Ld12>*pC0cK3*Owr#sth~A!3pxR>Iu0;m}IIPr0QsJ#N)78(5f4K@|ML
z#ev~2`~ao}r`WFkipa`~#$Kf`I>kSOHNIh+t^i^AB-I8|DL+C1(;tZrP$+Vl4!`J3
zozmyXya!>aj{H=kYt=&KeLB3!8@@z#UA`#TG=f2}eXSDCV(3BY@3+?G!`6#P!5518
z-GuJ;@`o+=dRdBdrpn=Jc*9n~Yyz&^j|O`8WK1XpGcmB5jLJ!OZ>}oa`K;b@s?WK6
zs_=(%oCPL=>36uYCciEaYa9x&bP*W8!+Ke9V7(|rlh9Lv*P>nh;p>ljpgdX)5UJ$?
zR`7^<52CB?TO^O~FqKzBSQHwAIdSXkU<zL9PhcCrvIJkncE19qhWj2cUm=eRjrQ3#
z-(nqrg0v<z-L8#G7x=LYeqSh{In*3=AY1Br?BMPAwEK}gtxaImC&B*PnW#|td-bZ8
zPiLmhlxg;Hc6LlrB`s<b9^NfG$#X;{b|f4`(37Zic%R$sj>ty2V_Hy8TR%M{1g?Sz
zAIoLlFv2eT{7`fjlc^gi{O?8hjVaE@&1#<O&;mZEA_$Kc3Js}oj9(<9hp}LzY&Wxb
zJqQVqC|J2nk&v=+#0;RTppm8i)25{gr{nskI<K~QAIzQbs#bWWjtZQVG?D+rt*hKW
zslW{={74&>I?zG*%_gm7Qy21*okIVzHG>AhaNRUz2Su!8{(PqSB$-vt$aQksEH(U*
zCD}B5Nbno<m<mZAJ>&<=-6KB^^vP;F8OQkb<vn%lyQ?gFeNToyjJ+^HFaxjD-N-I?
zcW)zKH+f87s6kz@noU+?yW!->yu|$+OFx}9VsHByi-y$4Ec~OQWtAHSUb8C8wzwK2
z!76Yw?ETqpe30h6nQ7i&Q=)HnqFB~s=_!1p*Zon4l{u)1?|CkJQ}ye<Hz(UZoj;;(
zM6kBQ(`OlU4-=O|N|%V&G%Dh}FS+tQHINp_P+AB=#CV3av9qvP-8Q`W&!;IYv%*I@
zj@5Q*owakEc%#)6nT+A%x&SR0ElN77RW6O9H&zi~@uD;l^<<O2I$8!PlaW~cL_qSQ
zczp{TC57dHO`##HJgL;etQW^9__Tu#9COQk-n-<54;YjT;yR23-Ou-(&-s8v`fBb*
zpTcRn_Z%qXrht&+IGbzR?Z|jOhH>ImhDue_W@PMb30!RhyO*{Kj2^FXSzl?{u!sh{
zp?Qhrr0SG4;x2JHH<f;P0R<xp#-2v|^V9wQ=|*pNRDI_L<PS*Kc|T?F@(QrHm0Osz
z;;_V$8I%VU;bKBS97M>DD+(*tGdLiJW3y+h3Qe$o5HYF?V?*9!4wB-;Y(eIeBJzpf
zSkdTJqrJ*W?(3Kbt8^W{b%a6Q(cZF)+%j&$kz5=EF|<;gy8B(XDLEb`FlBkjm-iXG
zrP5Kl4#%eMiyHJOK0kfN**Cy^eoJ*5BZ3fwP<lcibP?=M5QO>kr<b=CaWvbk%5C0H
zhVhq4&l{pSUNoxL#1O$A3bYyS?oZGl6QO3|Xfq2#^mzRgTwNJm2UE+SIn~F|-c~F9
zXE4WX=R`Z2uMzv!!bLXcGxG_Pu%6N>HQ6-SI)>R`EwG7jj+PVprWCBwXqFot|1-%0
zPe80Rn513C+T9Ho$u)aXCN-Etbf52j$Wcp{6h}$id0sJpShGYI0mB_`B?ZLxN7tfM
zXcp}@;-Mlu;3qiWM&tPMSr<*PQE6)cD+5tE{L!9+RTE9R5h(zAm2bZPBCJ<Cg<7h%
zpW=e~r2x0sa`|F{^b1lilni5j-0wpiu99{BJeiHWjnV(~Wx`#*ug8az<UzBibEfyF
zWUrT>=1wvuV$Gx8r}b=6LvcF&O&v1N`sQ?w2u_RzD(rBZ2kYG*WA(0=Yu;P<BTgZu
zDmLzZPt7uyJY{Pqf`4+vMkZ4{srsNS7_P;fm#l(@=FS%1x>H6#*QoujJBz11uKrx7
zAm{)u`<?adT*m5BzE}#a@@KJF*hr?nbV>tO)}v|wsXM^Yyj`WbKEfI6q)LpDC}0`F
zJ1#=B|1EM`?wibd^Lgai7?1SG<;~8BT9X4Z&*En9VDO6(&yYNcs9STNgZml|>iyGo
z#P7`uuSnh|?X9tq6%)SUyrAZvE++7ZktrI$^YV`SVuz{6Qq<XgTouXGof~$1jS_$1
zrztYZ^5vEN&gaBu86%1IS)+c2x<}gFCFa#wgI00kNTIX0nqwstKL^^a=}B95>G&BN
zYUs!IjP^R?@7I0#`L8|r-|qRM@G9Ta4!4|lZkNoBv(dG8mg#5$^S(S*{Qy6hfQY7>
z)*EfKwY}ZIIE$jR=Upy?OAPI!g2_`(JEYZ#q-^rf&l#>Ey#2l0JjK*zJ)Q#%WCikS
za4x{TFMGzL^Y@u5v;94`3)Z;7mbLV9=`=2gmoq0|TJd+NJOx5`yp^D{4e(;V5C_M7
zv!28F-g^VPNrVK6+4J0el6z}DN>UXoM=Lr|KLwF_gA1g4baK}K?Gv)of~fvJiIZm0
zY@iv>&M({h9Qlp)?5B>~U&NN}57Zp(?P)ny#!Wp&iSXD@)rP$V{j9nziNKvBEtAX~
z2mUg_ga@D-yN>kDj0_Owl~?T%er_{*B<-;vA+CiAcnxO<@lJ}6e4Nv4cBJh##g>n#
z1;)RzNeKd78f_JeEyv(!lu^FV@2*pm*<$dH$U&!@!2iu?(IcFd+GSgI0NtP4&>uNN
z)6@{48S|!^zU@Jg=f0~lUI7_zUhrV!>C?!}WWPL~D^?2RLc=>9@S2*MJ?;Cj4Awez
z;IQ4t`xwUGxTS?MQGW0c2o1w1Qc%Tu_Taq{LeO?kH@^=9SHv4uUfEALZ#=s}I`?9x
z8Oi0(<&*-lD2OmNKP=nU4ZSlMkFqjAI?mVg+%GIa*b4b7o{CJN*e{Iu?C*mjYVte|
z0u?qK@mJnGeh0}_FzV0VK5Ob#dcWHTgDv=*Z1i8Q(ZkI&r36&#y4Zr&<<q3*T*~K@
zi-Yp2Yo9mXzJSc&3>sePkj;+1dHoIX7)>m0`_-c;{1`}s^4cxxzbCp0-RWV~!KmzX
z>xYCcuMiVE)VZJU*E{ad`<${lR)v1;q;U3qRV$sd)GY{?K&!2f6d89td!h&`lK=T4
z=($e9TnD?V3CI;|_sXkY_oHbNT`oIeaP)@ij&vF3msi$C+`y?B;v3-IK-A9UyAND`
zhB&cqwz_`v?C*ly{hk_>S9rIBHP(S~#>ybp7B%z&wxM?K&za4JRfZf0%akx4FbQAC
zBw1}Kppts_3Fzd6CWp?(%{t)`3?d_aBsg>Xw+UeQVDJOK5i9oB#CR@!pVvixp8YY_
z8+%`sOZ{xDlaNejSt?tGy-+>3Off=kJlvSB(#H%0x2^k=v4l;Xj%Ailt&ZZID(eDP
zmPT?ktDKm`bZoHYhU?@ATMqA;0~D)%;;k%1RG7R>3^kt`&7ARhk%y{*1~=Y`i88=$
ztCizrTZZh;V}Bj)1}}cnNj@L5?t1q;+kASwO8UeXImY|sobac$T}zJfX%E8xT2`>X
z`C@Fw`Ly9VzFmoP_-H)s$mM11Hyw`?<vOo1JIBiB1jk*!52IL{Q=72a6!!}=V)50g
zWt(_+WXFE=5qc_Ocgyx2SUWzn2a_S|d)5^?AvQ^)R8f#~r{}sT;E9a&8sr|n^~P^@
zd$QG7BT%}m?4V!ThzjqeU-q$08l#$PU<qA*7lwfYH}ur4R)LU|CO9X=c)V6ux%Z)2
z{~!c9M$PuSTJFbbmzrcoIa<jMS;UhT5r(#Rlsd^u<oA4X-}M>;ZpP@YiI1$;_jDSQ
z9R<ILIBz!93K9|M%Jzr8&DY6_K~<^zI)P%P`gK#4Oo!PtuNlPH$P^2y&0tb}lqIfo
z)bYCANkOI}36B_kccZfrKau+MbZ;LR4x9Hjyo7>F86N3@-N(K9R6I#g?F=i%W~F5&
zRu`<-+j|u2VK;T*rpe{KvAkA5xPTFt32b1F7pU0ulwL0ixg4kzVxh$hG8$Ip!aPd?
z{PD6tA-}j;?f&YT$1Z8r_*HHg84hON&-Uewp2*T^N|ZtJzE;SQDm?Fq$OxR&J5CYW
z>CpE`Z`X6s)J2ixDTASlYm5h_>q0G>fF&^06k;4aj<w0jTRB^*yqnm|)G!D>T6<+a
z#B{lq_S6$<LJ#;7+w#uVKbgumbz7cl&q#(x{vF}E+=slv64kG-0sK-pL}OKdLZqd2
zP?>Pi1hkSbqSWXD0(_e&H&Pj;NzYl3Q!nUK`q208Cw-uZM6r-2v_vAW)@_i@4brek
z21juf-fp-|^j!~}VdPR@+lOOqd_tmEk<G@)EoByVhNmP)q=r7K!s6-8J-<7wx*-di
z4}%EyKQ#l(byfMd(GNIF7=hoMQF-Y%=9>=6$_wOR1xT{D0MZ1*(LeTup9bb!=+)sI
zu1=K0!%$^mhpdB~`$1{3w~(0cX2#P9=h!$vNTXRg9EsTg4UfH7e<)VmW?N7<)O;wf
z-a1C)VOg0x=;b<iFvWf$$8>#z4`QNtJ;<R0VyLRbSP`z)pWuS1*keWTC>oMDJ{C&7
zZFqx#CvwSv+H{;RNLaA7wE8DFinoF1Y2D(hQ^P<_$RS(CS^~p*Q&{iya`T-zIg(jE
zI+4*j-SJY_G01=lJ%ZRf1A?39Ng8Ceb1yFnBt=QDUI;coS8mM<2|hnKQKl*i1qqMG
zc#j|~2N?+yo+rVFHwoPzah^OfUb7t~G#%#%VqI<N$-9#F_^?(^bwl5fko>zrrNIrD
zDUUw!c|4peK-O=4@F<vcbFTdPrT8TFaCN2V_q0sm_1zpEs}}mp^)aDj@T&Sd96w5W
zD>R&1IJp1c9Vij<(egh2qnZ2!;sMUdiNcd1{~s0P&8u4W`e7U&J<ZaOrnc{-q8%fl
z9jpMf>L~XI+Zf)79w&o<K&|=Y{*<n$eGc%r-2d(wp{;HE7QGds1>@Q(fkq+G+42?;
z$PaAk^n;95;aJ<J>A0xR1)aavnMBB>PLxV?Jl!5I?dIBdqI-A7TEVPd9V#4a<I?el
z_<lGFrh|FNdLif{ifKX>ltaV3jt&z)B2?F+BcDYbwzpSOyfML6QFhUo6x>-IgjIu2
zeELe{l@+&%9t0cD2U0Dv?rG(tG`MUpc=MDQiMk~->t|m?)qad@9Q{3kf3Y67XX-}Q
z6e@r*BZ$S-39+J7R*WW)caE{I*^bg%Ed<h#$%v`nmj3~jk>k#V|80Ht$Kb=&*0A%A
z=YjiO?Xjccd{b*i4pH8BVJ}6ExqDx{N2G+wP%)*PforRIfzXSNABcoRZUSv2_7FnM
z{*zCf!mK}O<5!hTmE%LNYp6deRaxtC3ljxuxEC2X82y_Rg(&C(ng})*A+X|o5Sn&y
z`xcgqI&b1839C5&N;8-4FRxA7P(gGPNo%cTez&<!h|2MWxr@W`7YOzgS~~}o&Rxe#
zLCX3-<DB6QEIW@sV*X7i81A4n;Mm@Z^6x^=V?;VOG2#yo9f7MqTe9y=q6@Jgs0@69
zX8qIx6)*<I0K0gaKbCzc@dJ60{E#3hp*bJu7BN^6d#u^RoNPtHK<6;Qv@NP_Y6yIB
zmy6a_9?vE}R2?8|eXgdt)0U#VQt;Vpl<9-qUHZJxJr-!P@8_U4UWvCs#`=X}HM@kN
z9^|qGbsIE@-2f+Qvb?IJOpaQc7l?{aps;4H7d!nr)f3STU3sBl2_bSJVm5iO2*K+Q
zhbJwOPDCVOX9V*8>rMloh_F|PKfQc2%w#t}fvM_`^wJMYI*ge86}-`{3EE#J6N)=!
zTl+>RvC+pXAZ7CS(F28%v1SULckzBy<~!;QVvFr7CQJuH$o1pR8mc;aa|Ob+!pKX1
zA<)#|0zk)BsTT|9A>`%1r#QTX!4%!CfQ7yf2#$T%Px3LQ>uigRLnbUw>T+$?#LVdW
zYo8Eo0f}rSJMODsIFp_T7<W@eq#>t-wEyTFDAeHAPw#e8<R_#u*;b|aLncC=n(5Vc
z2B@>+X+IEZzrdq|+jltWaLCst9NEiO?Q%FXYIJAIq+h)?4_=wub@bRf8=NuCIc)cP
zzFa5z?8Cb15qa!&w@dE9fEDQ&jvmlpdv)Rj`ILsvY2FeXr(tj~jeDA9bHdi?-WJ5C
z)quvLH!vd9bH-J@@%+@|l>!O!jSBukT>M!e;K<r8FWz{KN~Xq$gXsqtoF%1mCDnSZ
z61rwd*tKNVw&>$td*-TrZ#mauygoI~jqkVdm=HLom-#^XnFlxc)$ljdgtSBs8Q1`&
z*`0%Z?HVByy=gwP9m@QX2GWvCZSuIQkw}i8bE{2!Ne`VoL-CZWFa&v6%z|h3{5mRI
z*earv7C)4Ch)c1Ej9xZH4^SDDxi(3PdTU^9>ef?gK<!$me8$vSTi+&-^FMV3*lO$u
zH=muJTY}>PteS&BZL|UZy4*A<b+Pca9_$k;eI%D2O~gP}U4JZ;%79#$J{$fs*5QCC
zCb*5Vhrkt2RnyQd6Vx~o@!4F_X2%^1kfNw}v!G~s^Lxo$y3GsBx!o^h5zbf_;qbAH
zg~0AKH{HV11AxcdSfgp#pV7mV;Z^~cgah5e`No+Su@}!9<Ovvdvh%6Ny|{%LRN}Q3
zB_aRee}N8n7}qqo<%=yUhBn@{9;^YEaldYbLU0?11fdFhFQirTF8B+Ux{;zJ4fH+t
zL4XK$jZu`RfdD4CppI<&5`DZCtqCd8>;_g93RurZ?K}A2(=J>0Oz^5v3J7S+Mn5FG
ztxNLW5O<t&phA%(M)j`~k#Lgqou~gQr}lgOS2O_L?BN6enK@000r4o)N+ONe>=(?s
z#URgO(=nP}!Z#IqsMZ}}?OKsKTgJ%X0rAu@_cz32<k;@A1y&c$tN#p>V|(D%>jezD
zkjr##t5a2f6Sm))=}AYqjR~#Mp|M96N4pcWs$~m$7$Imo2@r_~DF~M0Z<aZFoV-h!
z-2l-pzf~$pgbvI8^b2&@Xuk1Es^(dBksf@oo@@Acdw!7oc$we<&1Uuo86!`dr7&<T
zfFLQ0UF5gy5r`#~c<_Vwos)e*lsbMw&>3(R6vwE~snu9GsVta3BGa0%_cNw#u}6fv
zP8*Z)od-VO!JALK*+#vRiZ$H|s7elJ;TNYB7Ug<T(j6RWk(k8t_-*j=iq**sGPGIb
z=30G+i}jQuZv=E@ljqgSVi%f;HT#@*YcQ_)uSzfP>lp3_mcoNu)wsIV>bR5K?U<8r
z@(g6+PD#b`83c%CRU`N*VSv8EW28KsVbpZg^sgsZD>xtLRFV-bR2VamP<DkhnjHqC
zICYj(IYo-fJ90)+@$3xQT?u1Zb)%2Cl42jUhSw;i>$4~vO7+mnipIldAUQ@V1MdKO
zt_yG=+(~(QZE;a5m5VXj!N9kuQ5&el>aSQBLcin*LK>rJgi8=Owm>CT0%ZHE%oM9}
z@JQZ2;hI>#R49Pp89CH(CH3%0ip3yQqXOvPL8mxD=-#$5TvhI05gU8hIv*wzt2I>x
zujP1)@}lQ=dEs!nYTkmOa@CYG%{`9z?!2VivV)YtfaOCu$gQ~q#UK`fnYfp{;LWmA
zJB&!UX}(KK2%+;vPlz*+)JRB>2ioxt-|2Ve4YT(ooiy)~o#Z6F7w_BYUqaNe-)JHo
z@+H+VMdQ_~H9>Ycn&)s=(&%)^hAG|wVRF%-7^E*}TO#B$XtOw+kGlXC(0ZgCZiAOs
z=wgI3*xHk812_k7Fl4*C3N2ukpEr7<jNimVx!v<Q{R;kVLH0qj4ddU&Q3Gebw*S@S
z`3~Ty7Bzg@T&ZWHev*9^;BvUUNg7z=^s^IZzCkRkb}LZ`_h}R^waXF)5hLU2_}+EW
zXZR*c=${w1KQavgG;gKgy$!}prH@7%+bACALTR>BeEF~QMPdBcRw~JR1Z<LTtOo{0
z#PmKI85b+KMf%cR<cUllr<rH;NGxQ`W9(mCyvrk(Wx2!C!!ly07D5eB6#)>-=@TN?
z<4dI8vS~qI)X4I$APjb9_j7MpM1vi=&s!tNTARm!vumrT!0^GxaR+zxO;mF4^@4mU
zu@iIV^($<q?d%3=)sj9=wvYRmzuZIGidKx_d6pJtN-a9ORjz;Zt85Mte1BPZiXWJI
zl*OZ%Mzp3lfrx-3-ed?5){Yr-F%)pDkKN{p6SXcH<%5h4l0UfhY8XjE#P~>B8*F1#
zYl<#vR0X3*P`EYKFS#-mnvg3BrIl-*#7aj-ZWDc0DoiRg_`WjRd^<qeIOb)1+}KTA
zD7Bss?5{?V+3aIfq8E3X2v<{;Q$r5T$J-#`a}@vdUfcQ=zO5oM33>z`He_L`&7)4;
z!kC*4DM!G0ZyoPlg-&yPeD|AaW`It+_X?rVZE;q@;v>~<_jK_DVGNeQ@T|u(KP8j*
zjG%PW1{fWFkXOuBldRCkg^;)<Ai`9?B#>4iE-L~v**f)C2t0ba!af|CUL{buBilm>
z?HFm_tkPOpyM0_iW=1_&st2ymu9;$sdZm?M+%qrR{)C&}1yIIZK!C7F2y-OF&?B1L
zfQChIZ-hU<4AZQG9>j!`0Li&w>h79~qcdg)UO*dw72||-Mj?B-CfIT{K_?6SnOpP~
zS;T!F>$9^K0ybj?C8hshjfG_5VyQBHRXB<0n1w`i>GxVNx#0rLcVPXBR{yRIwsY0#
z6M#HfbxH(_FrtGwVit3_0emU#yCfvLwHGPOY$WPf$D3N*U^N;}YHjKvg^OrzMAC~j
zp6)PFq1)PU<G$A7Ql5HH=Eo&LJlpSIXy0)g`j^P5^Rh+xZ1Wc)hiDV3lq-_k-xSW+
zIasj`_!j|ACCuwuOW&;37xGHBAeCg?HiWjzO>yg|pKYzU4amDYU`D~qCJ&Hq({xq8
zVBEj}k<L|%iLnqnrv>sZQh2>mihIqZS~|i1PMSdf!L8iRXEisC>U+Bv5LC#_)8oY}
zHU<5F-7KeY^WEtKo2ka|6!866GBB-f#8v1Vsns;es@rY<HZmDh)qb5{SlcfN-(729
z09ghjziFp|mRr-Y-~%L?T~a#R>zZ2XEjVbtRjYmL?lWe(Ixy8seC&MY3nA-vXWRPm
z93yZ9>g4_2nYl`pyRVwcwwK|*+a2;f?+ZT)B;;1grSIKbt@)z<d?`OKKSXFav6Jz!
zdo4iyeV#bE-d+H-L{!;r#D_`-9v&_szxViORIF`p9_Hl7Z*c31CML{738jCu+<7tz
zZWD=l8-6+suQ<_Zk39F&k*@S}joV%Kbvz*mq?-7E)?N7XM!lVm?<IG%02Pu>re9^z
z!?*IG_Cxz~l6TTs#+zfmsjT;FO<2^kzM2llwM*Jox$`WLXuk7k4%|K2UTu_uEFZe?
zH>j0}bDxEQ`)15x4a$MFtJ#Hf=easnXLHI&_{bU&Gj{inQf;PE+*eTxn{uirMs++t
zbB`M#&dj9^mD_n@<Gnw9>S*OJ+Mlr*ClBq!-bxkFLGxCLh&O;J-;*_v7lWKP;kYyG
zp&QKBv<}m9+Jl<!Y8FLH8ZHYWOf0|SI2_JZ@$}tOQ5<wlk6`suVd|uB2qF&ZhU4Qh
z$O<x!&Xu4ZpwjQ!E;USa2O+k*1h^v@X*`G=S*aZzS{E-UAX!kVAz&tA1}H`Y2AWOx
z<R|wZonIdwyzfBS-3}wL9o6~H>)Jh}3*s!Em7?pK)~Ne**nilnty5QHpYQW$3A6(#
zRaMb{g0~F0po0(qUfkD;_^$<G)fj$hw2_qZvfnqk40`$3Xr>HA1h9@9Z!qr}mtQ($
zm8eyu-;hAputm8l*qb8tZHSVPtPhYfJHpOcP8E54Ce6U4P#g3>zQ=k3Xk7fO_2_la
zf?oR!dv<ms06SWB`U(EG{ial=LeM{~UZBS~tdaNmL?99R2k~#ToqkR=M%op&Q3zLX
zXj|8T3R8O6(E73mmHP{<iO~Byz?HoGPW>6KZ}-<P&y_u&-6n85D=yTUD*R3%gbUsh
zRI+<$8W8r6)`U_dsO~2}_VZa<xQTFrq`fwUSxi6?db&GgE7NYQ%q_!`>fPW~g1&hV
z$&~WxBHS&(rI*Kwq0abnia6UO^y!-4sz{EPt>+DK&{Fkp2b+!5p0i+nk(|!~=n;zq
zKc~~C{@F}I;Ua}W!D*Rid5AU-yUZ=@1!2BU`r#pRj_uDmLz}W~#dMwhtkM5yTU6RD
zz4*zU5}c=&G9V6nYxA9y!375CF#ZYdH0gXm=`F#(M{OULf6!(K0U6%o``i$Xh*G8P
z!ki_@d^Ph+u2;&bRRwEr2w>3%KgE?m$pIp<<QI0=Nug{}S@m+Ym_30cnQiuo{efsr
zmF8M4VXIpH3}zI-$=QwarTmr(GBY{%LWw`hlD>rh>|S{{Ati6fpHH@N|8Rlvl6O3}
zLUgW@PgL7+v3)kB3263GlladUf_!diHgQ>hFD%dpi{0!`p4yq*I=1WOyPfEG-rHVX
znLa<+<XUt1b^L0-<9befGQ$?tMRLNW9DM^j47>t8<Rt3K92&1*4^5~DNq8|fFsZpT
zs8>k-+3JGZ`~bB{{80Fj{jh_0nJ)ZSV2D$<CC(MG-7rL-`TII9;y@V&Bn!;m{J@HP
zJHUzjX%lZV)35D-RK5eNyL$-7k7QE}YV%gm=dO?bf@|~}z!U{{nxJ6e>3QrVt#+D}
zsnfUwlyg@q!~0a6&yda$X8&0rNU_TXc;l@44~*t|mYam)q8Xq+zZsw?o;QJxI|l4M
zf-fVy1poG^AKn9Xu19fERhfCFX$NjkBZeP73AFS3cbBm(#-r+W9L6Nft5@CY3g920
z!C{bJ6N1s=oY7e&G~U14YaFb-r5karaJQ<a9@f5!1(o&|&z5_dE>Z_P--ii_tpN$v
zRf+DC&J!H@Hr9jb4?is)+>1Z9MZNXBX{d82pg`f}vdAG&tkT%PhgpTGG)aH4R-K!5
z6^ed$BZ`RO-uEd<@{IMgARhK)wjvAajP>BT7t;=GO68)7%ZW_Q4%VWOh4QKcTMDXc
zH~Rk7vZO-kD46B~=5(z^jPZYEWT3eb{)p>u{m^>SVO>JK+?HT^8kg%x@~@=dAGt5M
z>-zV(JfZ$CP&387j~?S$bDkvk+Y?x`YwD&m3;5{fUMwlooD6<SUxNQgO{Md`d-8V=
zFdX}*ar)HNVC!B*<O!v1(oAm#*12D~%;lBR-;`@CM!Ic8E|sg3HBZwe>&>S^)2aAJ
zRV}gj{o-@z5u=v6Td+njHN#!NB>Dc7BBFt^ZT=(KlW1$Q@Xien{^R}&mU=U}D8jDt
z%7=ky&spFSK<iQ0w=1Pt*X`8vUG?A7C+I6!T%>rr(t~11hsk6wKk{FnLXIMK_;f&{
zqR8gJzo)JN41<^f$e*frIaCw#0Xi~QF_QoOG5^jT_9gsD-KWgG$lmVd(q}2UspS_+
zL}D3w%3>58U7B-)&yHogXmV6osVFjaI5z(8LHq(MfY0e%EK(G;#ELQsFQIU<Zp>`r
zq?4eyQg`Z}jC72lT%GuY?SXQ|vpl*;b|^=NZRW?4@AHb-NgK5~Oy^Te0w{3D-kjcl
zn_DcAME7|Lus*GF#|FAERMm>Ij?RZODV`0Y6tPRV+FE9nn!NNYa4ANoip$|EOKI1W
zP?OgD);?f8mj_yP`Lnglll+$1dL9x(daykG&tAnpP`L7xXRX2ZixGnqCo~Rb0IL>2
z+dzk?qE#>ems*8>=W6vQ>Qvz}oa(67WAFkmw7UAYpEZc@0<KSVhNuXh?_kmcC=JyA
zes1=FOR8$Os+I&8mI@usm7xSZnvb1t6ofJrylxJbI4oj=%2YNmtQFI_h9m&NI&VZa
zk8`Yke8F71?+91TYIRw<Kl04CYAwOtvIQ3;(a$&Qmg0~T?02j*`~1C|15{{$UD%P*
z{B<KOQ`BC^{cFq9by-Gz!N-%IDQs>NdlOdB6b-6ruf_8vO_5R*QNf0sZl=^J5ah(z
z_iyxG>?dp6nABx<2=Td&h>g7~CL2eJTZyIIXGv7Ul^LZa&@sTvv%Td&Yy5S4p*x+A
z_=D8A=BiA+aXg0f{6?sDfydiaR2-Oo2FzjC96_Kv$6l1otU?m(Nq3O=R8eyupF~Gh
z|6+i;o^<WUEl-ZW&jLz#*KiV3{|iT3loDrF?j&S`gMB)=^a4lRreoPI;T%9V-!$_L
z``=Zh@IkVfuMPlkDLadCbe?5lAi?12bAtQtnE^BcdT=qMQ;o&q5!3shCK_qlQ!2H0
zi`4Zd+#-8d_Tr?P%8|TaijaWj;_dJd$1wo}%W7sApgNb3Fm%Bmva_V;eSk5F0O;pF
zW!^KdvmANBH)j9c_s3?y#gHqEdJb9bU@d|H)%P63-dZqf?{WB?mavtp*M2VALL=6u
z-SIGFuE1L-c%2$V=BvRnuu|PE22ud>o}X?mFMc*T$S>AONj{k@@1!i1820vb>&j6!
zC+m1~=%nnx=uJD%GUJ~PSVd*JB;T#b>yb@#_>eJb&!(CUpb;20GZrMnR-F+I&Q(U>
zW=ca>#X?)cA|M9i{$YNB!6N_TzYp?%O&$N-b`(F6k5=2iRCe51ouI*X%Yjv|J><o+
z+5g^jld)YB`FE0dEM@wmG6_LY{EZhgk4vl4F)w#=j*i&9lH>C|II1MpIvmefQ5;c7
zFDDD_Ho}t?-dO)qY_#L6w}@*|=KGbjg_b6agQd~FI4|?l=$dApT?hW3R~<VRfJ@z;
z*mG?4{%$lM*Z0FKp3`IcXwla?uhAR2Gqw25YJz0adEcmT$xeW!PeNy{!?jFTyVLd&
zk)4ch6C0!B4l$2X4CaS09|=lwS3kjF;iqih1DvtuyT@V!?O6Uo^{*vDS+2*4PEV&|
zkHQ<DK1~~fmJM7c_4^Ril>#j?=`JR^0}Q6%w6|Z|_&EM|tNiCp{6B4pC}3vm%7B%5
zK5$~sC>~bowodZxW*CR>BK`A0z;{wSB9XjfJSjN)R<FU~L;snL6r6zp9v3=XUJZH%
zi<Oygx-!ewt+Y>R{q9N=evK%(E#J;zk-EGGbOKggS51rxeS5oi&u{+-;J%lD@p~{_
zp0l3WJQQKHld0tV`+HHWAZK`0vRHmWqF<oT`d(0>Ei>L1A<;L2F{C}`F051AC&AWU
zK=7Qz{DI3bPzibuqG8d8WT^QS8Eg6Rc6Z1tnZr1v-mjD2;r?39YNZ8oratvPi#CZ#
zyHu0&>fNsZSn)M)<2E+AEc^m_IZ<SHx8JWjh!@+J#4dIZZ;Qdn`RnHZ_nm!r<63fd
z75I9u%evftF5Kvq-{cwQ>h-ScLF+eJzTuXFrP8&B5b(~Krium1$?e74#`?+IPU3#J
z$zYP9|KE-1pC}A!-XW@0is=gNlu`aeW{|Y_2Dq_58}W18dK<mJ2O@asi_Q1{?}ZL_
zTxU^Zeq79uXK`$PJKfo7CMYwZ(bWBk$oS}m0zHP=Qv8?0T;IpbQJu0j!RKWbem1Az
zx|E>fPJK2y#hvamSqeO)qlK;9Oo8Y(6`IN{mFz{o2><;F5tNX#evCk*%kkX4MDwCV
zI=Qc#S)2)#oq|EHIZ=17F6KEfuMU!=a)a&N<pEChSZX0ZESlZl;?eWlp53_#Jv!lf
z4uSsk+7>41@Xd+Y^-AD-Cm>JPYW~>%xau?aAw_R^^U$L4Cy<GcNqF77{9I=_m0@%(
z=`&PHwcgJ;T4Q|I0GK0DS)RkxtGRHs{0^d>dTPnO9;cJj_@_^IoP7P+%=3plCeT%A
zXoDO9FDPR<w_u00Ybh+Pv<u#2m+5!<PELn0#p}Q=Z!&CwAV)5hVfx5!dPKTdc{sp!
z$(H}Q(eR(*9ECOjbYT-<uCWI~k~n6bK<wm4u$7f9+!6m@6(oFsJI;4%RHojVx#mK1
zp_T;>q0UFMBa4fwi{Fh)!R2vn%F9B!hJT)VP~{h!l*VKRDe1%Z0uawFC#ZPj(i24%
z7QDUp+|O#KX->C3JEl~4Uj2C+pRS!~{qyhb@ZKIqB1)@7AzOB1HeEXhokc5gFWvg}
z-t`jaULfj6tt=o;@VVOtql&xxW4o2FqTA(W@YpNwZszwl!e%x{KMM(#vQ8s6X@9Hz
zSS-&B6+ILhEk>uVZ<}w}JH6yS%bbF>=M}7&o!y?-tzVgu1<)J=TUVKGq9tw;As;2W
zS}hYEhz7IwXDi6=uC;kwjtJd5#@l)7A__JzrWLcV9;|!U{xh2k$q)u8>h?;L5E57(
zeG!BG$-=(_NTTGYU7_b(5D-AL$rt!vW6s|q2wrX-3|hH$L%@_YB4veIws<%A|9e#<
z!X1uD;#0dKaLG8F0$cKNRl|rEPB)G^4wpW?R;OI6g$N4WeU|`;!ASyrlaCQNcQNCT
z^PPHG3L>P3)1aFC4l7mED7tvVEePp9FLhyQgS~O)^w+f6-1ZnO<MC!#h6Q4&Ev^S9
z#fOzYa>(A1sC*?{BY@dSWRORpePGf}2bDC35tn$=t(gEiS%!lA?Z|b#S#+4>XEo*r
zo|9ZLO;!RJrvTOgike?PSuQWeB%v7?L*4Kp8$RpMu1YaxVo5CJ`ao#VwNtL%+AkZE
zJ62lyde7^^?Dh$7-U<iSB@+kQ9#S^dNuBmBGE5@t*WlW}WskTx)Rf#0nAYi`lK*-A
zz;TKOSEg2BRrV50q6>&P%|K}U`u}|)gZMqj3m{84b$qV}V?4mSf6{rCsrXswIOlvv
z^Zoi#B+nYjUcpw>=WVAU93GebcqabDt0ju~HV;xd#aOsx7Eae`9=nm<xWHvP9qxbL
z_<P7*Avar}qmZC01a+B{dn|9t%1JIepG~JKb!qo5#-b)z8cnL*Z}gjp{|{ql9aUw!
zwR^g|r6rXv=?(!&>24H|?hXM-K|~2@l#njzSkg#$v*_+_&ds~`yW<<@JL3%gTEoFu
zvext5&z#r1=I`RG4}E8EBsXT>^+vTLGV6eSfWM78-304z7J%g5%+{KH{o5HNOqQ`G
zpWSp=N#1CDY3aoT{ZpEqNu$xIDq5zhPsJL!(f6K0zWj%E=V=~&oGKyk?CEuB1CZ;`
zhO24^zMA{!Cv0eMvRoEFS}cCJk0!G3<rn$q<0(i|ZZV9?|5Noh#6~ri|2R_VU}3r8
z7C<WRC{Xc!g4}F*Rr3<KfBqKuT;<^E>b8}mi&*@SRcLm4ED`Y_rwM~%r5?~?^)4EN
z$Ww}}pt{_L(l`A@A<x+~v!qec6fVnMf9j*~N}>~JtT_qe%kjc(k{DNYKU?f5Zwms~
zx&UmVf*(6Qn6QIeBpf+ty?@V$)9$BQx!;HKNClIVbzV|_j`;Nf)>l3k2?Pu%GCw^I
z6q(dNl*ZoAa*07{=fY1myKx_MRy<H_9O`>+UH+^kzO!HaIE~znLAgsEMPGd*Ee?B2
z%YabQvY_l(r4{<?od{1rvr__!rgsb<9%K>Tkl%gxRBdinG2J%Cg9f1y%psrY<v}Tg
z`sPc2A3fpTzYGHzieR9F-SL9stjBnL(7)mrKEc~k{fx*te`B2--Ck`JW)VhE_)yoO
zSQt4AOz-ev6snyw^tZhlj7SzQLP-s{ztRAbu#`JcQ+V2){wgXnsZr;gx+q+?auP`k
z96O)9C0XWwk4hIQ4oF!!?7e_88K2KY@W+{BX!ElITNW78i~tiMJjqz0b}Xg$=F|GC
z^_1t#%*-ZF9{w&hK-G&38t20IT?P*yY}{gEX@hWSkGF=N`oVsGDIm@G%tPx&y<(ch
z2)2~DOl)7{^F(p+X%A!a>G`bkb<2ax$Op_U?DxPE!DDYAs<P!BNfRJ%375n$*>bWl
znJ`gIOFVFS4Epd-B?w4ZeR=h6#jcZv&!2Y5P4+|f%;1dR>}F@IJ0i`-3MZX7sb!qn
zh~I1Gydtuh8GBi*H@(?-vO8A)EA7>nMW1b)sm3TOKb{On_AK;8At$pBY!Rl@GQ>-w
z(qf*&WuL?IwfZF|jR!9{xOjVrkn`3GORI;&s0w0dwT<1>!mt`eXy@G>nI=QcyOhOl
z29j53D&J2^qj-wlPtnJ$ZSGh0RU=0)ih}&Xyxviv!|nV(iMqlx-Na5T|GJ+4eI`8C
z;QI*fyHgT??S#9JW7!z|8Hn9q_m$Sq|LP_FI&hw%RM~u`6|7rJD<Wk6yjk@ej|SN&
z)e26(rv9gL8u_1o&BSQyH!Tn8$X1B>Gc8al%U(1E3Yp7;0J~N@#?b}>EVDt;6ct)o
z!RKNjMlj|}6Gc6EG_W6Oj9xRyatB$^9QJWvonEd42?;CiobUG&4aQw7KUmlxPlkf4
z0w!0dL~Si(((ZUIkz%Z-v{RJh_19%*@IRxL`~GI2FHhUJ|2rZ&+`$?LnN%nmQGvlU
zQ;ae5c)n@Q%&+3*`<Yju;{1NRcID6Z*$^ao5<eboQ-zbAi^HWY0~q=tTkbyjo=5nu
zx3z)})1SeoJmhzA<fYrm@MMLE&GmJI`*xTCc|!`{Td&wzi=u#vNmOn-e)8?NOTd$e
z>Twi5*qUgVKug}gb`bG?R=`dV>1@|aZ<n-Dcm~>Xzn!b^(4PaAwPTnH(b@ZOC*fV}
zT2HXge-&t|7-2^e2q-(Qe+`veXljp5DEdI&^*_qIe<By?2!p`af_u;Zw=xK>dx)$!
zpOWRX;cRy7dqtL@F3R4?NRQ!Z9&a8Gjo#`%K)k>b;(vbyHy4%R%_P-lZ~onW-q{)|
z3dLJ_tBf_ADRGt8-qJ}x-iDwS49UwNH5}ZY_*Le~_kx#dvH}JL@UTwCG>))ElT>Oc
zG03|o7vZJ|4CwGts7-MB+O-Y8F5LyY01{~Z=As1lR~GA|NY~0r73c|KQMVWkO__OE
zihZ37`+`hoF!F30(Seyr;kYCwnch|y)#=5t+zxtD^l|m94#UPu99jmrdGcS=qTijT
z7$C;zY8ZKS$PbDTrw6+*Rf*ubRn4+~c{;SV)}wGDaA-fhoC`c|0Yx3NK^GtSQq4?c
zPB(Vbgw}4pMk687D5k7u!@?Jtx#PtGk;|<&0kg$kSW)_A+RERL&j0uX|EsABL?uvz
z+l0JE)a;6366xy7fD~xje-oa5R(8UERllRGe6>T>1y#ZGJP^qvl8I;NEUB6(wlhOp
zXIUD{H5{M)eKBdgE+lwXccjMf$rEmKEJ2{cW4m3QkkS90W5iE!$lCGHRC&-OA;!Ar
zv3w{O10iXLn+2S;T&4~u`Y$Wfyci0Kv5K5Dokp_RkI}l!C3Ot3_gAtR4k^4^ocwIZ
zuvjnNePI0Tb0~pqX&G7*(~Q^%unPMPD(v-?VhdY@8VuQNTRJa?v$wUtQR_V29--_U
zcBF%D>evW}WYYNHZ;R0AbMcM%Fk3`DFWvW`;%%&dlU#d8XARa7P50Hm<*9!aTK~NE
z{{4Z;s{m~(-X9-P!|`xY*zT=fZqku+dw91{Cp?-$CDpr$TI*}qrpqUz#(yN6ytRgd
z66FOwp4Og1a3kR#tqPTI1ad}=z#hYGt=zb~bJODk!uL2rL9x`9z)kq<OFX3~%;dY3
z3Z1Y0SAV_h=TPw$|EnsC2%|mi3cBGh&z~$C2|8y%b>Ai%*LXT*t4a=8EOnis&^eAO
zxih_{;TU+B1npQn!x49)SiEC1F>8Tn%py-p%Ffg9A8n}hZ&w|U<8nehFS>CfXM%`>
zORy-ve9eprUb$HBw?|`L)ZEC@YgN%^;vygXzu$-d{wx$TaqvKKzrOw%$EEy|PS-u^
zpxI@Z*mzT@>!1ayJYL}Bp?b(09KOBZaDp;gQXsj~NmTG!svk?e9>^F*<RWfNbCH)x
zuQ}H*R~Zh6Wjo;!D314LE8!iS9Rx}YC2&bZ^jIB#2gZ=ucjJ$0oV>`&tj10Ra)kIT
zD(G?IjeL~+*iOeyT0^4qGhV;?s)mm3w{|IdP}ilr^y8wqD+5slCT6+@TH>+fcV+3n
z9_gyEweY~-krfT%lN;K4MbhCG_@KgG{t@kK6D@={=S%Ym0nv2_KvN0{)Xw1jT>ld)
zxY@g$^S@sX|N0kkQoLucCFgM?t?`%KTb~M`nCRIXK2!g&IV^hyr#^qZ)wo-?ky_=L
zWpwL1X?PuDcKC#UNSTl+U*2?kk&yXxEq`*IUNLhK1^5}W5YeM=_CtyFetJ5cxn(cf
z^wDz%{k7^GmB7wN$$kIpJxZ?MjR6REf$InlczK)lmyi?!Fwd?&$laj@Gie?(pbL_O
zz76}6Cl{ysT$!yan(prAggZ_^uP3>{!B}XTXQgYR^kQgtx-;l)&-AykpM(e{>@7tB
z8`STV8W9@BZWpt^GOH{|ePw57p7xzvXo{PUQc6REE%i4fI{XI3O8xvF&C0(%hW~ND
ze8PEVsK?$X=&1_TecJQ+`?9!6CkCS_nyeuSZ=Xa~b)Nm5t0AvTZaigx0Ted^>$g#p
zBsw}!S6#a(u3e*ucWTt@o!8xSc!Qp+FIv2@mAcp`JqzI!=mJy^sl}^}vq?fK6h&re
zg=I|bhsvSuu8>C~gq$C61o`C4#v1(T8Lr8(R*%9#c(aM0Y{1l@5?Acq%gS4z^ZFs0
zSwD8=a~RB<FI&S>0oZ87Mp#SuWmZ@p-FAdQr?S-Y;M6Y9{?JL5SK~xSqZ=TIj1>Z9
zD~kPHmxW|2qsHgHcm@^x$4CEP#l-)7A^wz6wcraSBnbX`t&!&*!GJB+abg@j6bPKE
z8KT~%*0=Az`O^E|6h{G-AlBfitNLR<k}dP1DcN7CVHwTu?eX?F9^nt@_3c+wI@+KJ
zd=-yRH<RpR_xl;Z78}$mxp)DcKfT&mA3O_q*vqQ6bJ=>hXE#=ak|ziLC@WiY2j|+S
zTZB3L*xHtXoc@%v4emZHF7uZRmTyxUy0FHNSGuHfs4LJk=O95&YWN@6($(8Pfnc1i
z47-+_0i{B%<QAf?G8^j@O(P=<rKm7?cBIb%mrGk=mvYh-_r(IfQxDpnP^g#eed=oO
zS|4|tSXX)Aq-^ZJJSqQiV*HQC4F@cf<}2tl1gjCTCBE^lI%=Y1+~ZR$`Br8$#$7eb
zWddC_5)e6mlo(101*iFxeBG~;g-YxipX4eP8#x6Aq&{4znjt+@fv<rMJ}4A@Z6ism
zxXAT<dafxIV>W!}lL5lZk0aOxapOeUNwDz80NMfW8ifF*#Nt7x$BQ<|;wgutiD-XX
zA+fTgoRH-!2F+5+hMVvvAE^u>pYy8M#d_t;+`37`o!(XUcHepJPG_CxY7a}hLgR7*
zQrxf_@3&g7>}$I2McNw<1h?8FMt_AHLCCIeo$9L~4p~Up8b9&;Z_a$&O|EXE{y%OQ
zITyGbg&iR>9f*FJkhpS8Sh}wolyf>e^y~eNaTBxc?`oRzoVIz#)o$P4bnL5D%^)>E
zDLqwU$yHF0oTxxm!f!4KmzXXAC9xCX6ajD=8eSO&h-7G=u6kOV(v`UtrhuK_Dq4eD
z3~KLpLyHJ^2LZ<`Y;(8&=ntvm5ISQ$MgT=lG(hC8DwwT~uKEI3zvctcA|`43enZ3@
z)v}@1!~L~H0+^;{``t($5FeGiH60YA#A<YCe&U%P0si=mVvQdtuamj9J>Uz}bRmT_
zZAbsx`S2gD%TriIy8`rGyu9!D50<6P+duYZ&(b~*ifm9T2p737LmQJ455fDCHY^Qg
zU9C=#1$*`SU#&%l8`Oh?U;A$8=^2+%E5hfGFi6&l8TJIUtpc&gNOZFk%h%w_z@<(7
zQc9(bU-1nY3uoTnhth5H6D|3<l0lK=diOt|UPp|F{d{cOqq0*GVJJYix7t)~I^>cj
z8tDkQ%Mmkn-?s%=UtJ=`(8TOGZ9_}iw1(&8dW^~#1m3w;>iAlAoD(nBpH9FT^<h<b
zpKV3XwCe<_9u!Ny*zS%Sz3Lg!Dj{!1A->u?SJ4${<`-4Jk~?Jlzc(ZQb$*C{W!e&A
z0wh1^@NbkA*~QJ**dh;&M)YIl-@gjFMHV+`-s%ywq%PIpOz0~Y7(YmNKlk(dMZ!PC
zhM*A_fgVYyPQ{lci0rp^!iX=21Q*YUtgNzAs2+x;RbY_Zkl-hL8IKwWNnqXoG*CzX
zD(UUnUX8uY*t4+f^ZxI5H&_0>!UD=bb`{S?Ez(`deOJ;${}!d#B2AFU`(i#i37f5N
z@M0(<V0$n=7`gJsKKmldGcin8ZNEwTq4E5BN14-6y~UxuA^qo<0A=OfcVzEi56B!4
z(yjg^<m3-N$MM;b988aEQa6PyjbUQ{k16T@c#Qn}GEu}uS>pkyC3TXO&sMuSjC<4H
z2TAuELPc2gR<qIBx^ut)g|?RT8)}o!YzV(#o1&WaH?xi(!wcQtWygCxQSs#%;BcB-
zj$fG|*OmG=`RsJyHM`#SAfVv~sqQ!6fK}uaiUgk*?T#y*0-<ELsH6g18J;+DrZeAb
zixsI~{3ZPMvi0T!BcGjh+#NAyJnCxOLH5_=V+xZ+M^^KL4<}U2zZ|HNVN?v6*yWMn
zJmxL)idfu~s%?*kw2imslgHl?H`8~;hOxU33{1F0<^%#Z1Ge9-)8a1mTCZ$?a!>zw
z&CMTXI|6y30YT=9`XdhcJJAk>zeL3U=Y2%L1V_lM7m^ZD-@~$~SsAhMSt?AJ$Gd95
zkC=^<Ri9`f#zwJtx>z6UmtBg2LPF9()!umTOMC`twuKJ&NmTqr8`s^W=3tKW8?23f
zgg(3^ky(1TN@mSS@FZ>ZvFGfK9VKLgOW|v<9-*sSoR`f8WyDf5$Y0{J-qk_2d~=-;
za^DW~B}?iAfMVk%jpmRY|N7zxcwoa9N*cKmdF@6r;)^6dp5Na*0lTZd7S@GYTL))h
z7)wF2h)t#Kg}2UYEUtHF*Eb`m_@|Q_p5v^9CnEmBje1f^RNHmbcuR!^M75KD`Y{bi
z(P15)Z7iV?vbf$~EoHROxGk=6Yc2Aw8!J-pp@0?O;E7UW6+pz{TM~Myq79UGMHh{$
zG2=J<DpA(5;biUqui7?H3^uGya;+DCZ()zh=UNEeQ>9XFwT9TZZ?e88JN|Yn4?W|C
z4MAx%uK{J$rbH*}pgYm7RF>-3XyxEb#s!(?&Bf8sG$D~zrPmoRYed0?9YM|EU1hNA
zc+&4ExN^sbI~&+5M*-Isw()MBYN`2VNGh+i7aOccp|e+^4!>qAHYz4=zEYp6Ir#C(
zqdnXgFbb7%5GPA0nlszQR37c0KrRb`sbXmI*+?dR#%J>-xTKcG|8tdbfF|@}vhZ!z
z{`?u-pxn&h4jp#T&3<K+1W!;&(fzNRQ+48{EA(J1{(RJ01Il22t+eaa&prpgqNZU_
zf11%Rb_l$^^;|iT2TGWu*ZxHROK3o;=w$aZjY6Vn5xv`igbF=p0|S(3N|A2~Y=$=7
zKHZ<5n*H!o04A;W4<JCwWeZS<c`V$=HMIeiZQdeW#Uf;OFBceIhu)PoF%&Je97npl
ztv&i<d2FXR3_n*H@%o(GdV@WqA9j623kM)sF#+Ukt%3}ZrdcL_m4@01rDU#9+v=%#
z6R^&9aN3Tqf<WxH?E!p#?cobqdH!9?y#R|M4PJ9uXo6)Z(ib*go{FZ?<|?N#z0?b^
zuv-4exA5#G(;3G<=DCO}CSV$Oy7@hplxYOC+}_7MjF?A9k#Y&%?S+1+4pwj~B4-4D
zZNeW;zs<-2z#6iDO7l3VK_C%t6s&*P=$sw*vJ+E;Hennc#asd)jWx(+8ZiTVYqDRN
zG(WSQ$UFkfPP?cAF8z^wmvylRegC=oB(9w4Y#D`f_@{$@wz_9!O~A>P1!Whk5^<VG
zQ)u6NK|Q}K36n`r=&oj{F!^@zd~;c^1&Ig5e)o0EyUMavxHYgKF7Bx`O8FL3p@uwE
z0AtCgSnsEJ;*O3EDvR+x-oepFiZMrvgHcA_E_qYS-aTb8xSff*t-YG@)~8j#cO>+c
zjl$E;YHo@NDaB&gJa)~ebcfB42zD`slHPlbFt65K;HMOF;l(u~a^vnUgbPk<nd+}%
zi3E>?gE~X>lKM1aXTM<&EW4QLgsSUqk$WOzI)^{8yqxny(<1f%UBum=qQAM3v39_a
zx;*@3pjav_ol-9T14*{MTJ3&==Li2w)38mx!O;Y}ny&TB6uv=L4Cars{J?%A?1$f8
zgi2Hzco8RiT1kx<oUhAWQGGJ1#nD4>+vI!Y{EgH6L)JRtaQT~jX;N9+nJW1|O<J%m
zLA`_vOE@?x8KUz%cfcC=O*th7b+TBWaX!K79p_~WBeg}u5?RFxi3~O=kKCLuPpmGh
zYD@tO!SP-V)Vnu2$BG0l5Tw78%nSjnhm=9RY>`ou-k%I)gC92u8|OJ(L4K{v;_eH=
zo`)uqDPLL6<gC7r)|LT_`7p>Ulma|&&iOw8@H+y=VHjICAZ#vt<*xy+_z#3?+i9@9
zrCF@^Vua<p9`=U9v>Qos47mW8#R57J3(oh|mc(3_WhFM`e@qco@$&V?UomU{5qgVF
zK{Hd?MQh+qyOX3!^auS5LVAFx@s2ZmSPEmh9CSdQAO7wzSU-unwmoC^#_&={nw*rU
zY-sFfvg?3AIli#eX$d+RpLfVfHX9;1h7c9R*P3TR>99TValTxwtax=KS=<%{<H={0
zdf7jp=_9T$5{(m|)>4QFO7S1fY{dvUZzUQ>w-Ok#hCWI2l-(o>@CtD*X&ia^I3OEH
zvok7>us1a7{OdgW0&y8!8&zNr^-pv^>d@kx68FE4s<NH-W`bqUE$xE5#M)9>ENR$v
z9Tc%~T5FF0TXDaLLzovs`5&UWS!R5wzeHT%m!l+E$`WyyYgEBrsMiN_!klgEqQ0NQ
zB$K2&?**QmJ0IAVAF#$(*32--K=}TtL@g$kQSZ8im*7dvYClzGYQratX#>rYLbhan
z3@bGt>Uj@#2Gkm60O711{gA54x>dK<ybJ56C$X37{+vK+5kKoc>#(3NjEi{Hq#&G`
z5Zmv}3fLZ>Ln-b5s}vEx#>+R<hi<`a2w+7#Eqj;PwNstPT<qbHvZ?c`lr3X#b{@>n
z<Gs)Oa-7$chrBk{M$}aVIx4@U&=4M&Wmm$&@1Cy2>T-e?vmwqix-IFkPv%AYdVvAq
z1E&4;D*ja_raP2%%C0Skkt)ct*WmQ1>v*L@Tfg=ofhngY(Pp@HyNR;t^s^NI+4Tg*
z0e*17f>#%P?$1Gd_(EmR(&Qu??~^yp>mjnA)qy80llBmlg{_cK)hvHMbfuJp=j}e%
zC&u}{lB``OYMkrFq?G<s%2=|#y@^D9aj1-rZJ6#%dK!kixk{9J719AsTBU~k3FM--
zMoHh2<#`+z<(-=Caqzo4Bq0qhd0j#)F}4crW*l|1#SRT;&`IX``eZ7cPjM2i6S{F}
z0V6whF}8(H<oM5|5sU)v+^zvSwAH5EFjW9hQpqL+Q~A#k3N(IZr(S!5;4Dt&p(6f^
zDE8)X-DIX3=CC$ElB$W-6-dSX`|Gh_EVJmrvl-4t6am~KR}_ayf&cC-gXkDk8>*Xc
zOy&js-pb<fv91;H`g)3Pr8*78ollt`dDKPn)!zEXdbaH0RR*tt85g5;mp0bCn|8W3
z9buHbqh7V}^DpmrOhH7LTfr8|SV7L9%buR-8`oP;-QRd5oHQBjN#;`E?vzY4wr?Jt
z`2;J=5yV5s;h1UEj>K8=$B&N$S=#>YTA=-OX!;Zaf1Z!PE*tad;+xmSycZY&VU6cI
zN!F^akJLI{wbpjYbj2LGJxdaMbj?un1SNB)g`B*~hmq|QaN!E&uDws^*ARV^@Niw|
z`$=-Tf-be++~UqUmnJn=9F>6G9@qZ-`eqDU^wZoTkyUYg;bD_cKa_0FBO{GTOW4bo
zQ4$gr1Yh86_H^;Z3M8tSFedO~x02Q3*GqS-Rc>u(3NiyS1{}^H+U1|SKl9SLel^Q@
z`uW_8kd6OT598v%2$TxK-yHjiU#wf%sfr!$XGYk3e+{72Y^O?S2HqgGdn<_`>iIj%
zEWlIJX!LZNrumo6@sk$9`R+8WU?%b~*wysgrerB5?fz)`FKZB}UgQ+ub|ER`DZl{8
z_gh|tn8R&T_w5$qWc?Ysb8_!mRx-EY=~5Wm^)@rHgfGDkjOiZtAphpfJG&5%l_RvT
zOki_HwJWAa;c|_*Zb&@+n|ZSMG79A9$-Kwf9n!^|Ib#Z|f0xp-UJpHI&=d}+WaZI`
zw+A<*KJd2^m3@6~j$7$&sR|rH-sjW4K;Wh)91j~s<Hxq2S<QZ~u1pE7MbIjIgNt=a
z^mP9h9>O)wH;sIgr(cZgt7OJfLR{$Yl_@VA^Mst<KLcKc&DXIU^wdWl$K33LQX<}E
zb9AY8t=7rn1-O^@K8ZMT>!HF_mX>j~d4mMO$ap7;6*gaQg>ND7SWaRC7py<4Gi21E
zv-Xvv?GeW}I;%o%1@WEHPKS3za9ndm{ULbt;UMGbG}H8J*+imt6;@;I1H@Li00$Jn
z`MQ3-SdD<7`&=lc@j|NL`f^#S|Idm7<ISoN(-T7QeLR<gVJp75I7pmm4B=0~dcqY_
z@4m-G*;f#el8}w?v<G<Ub){+OwK(E%4r9pp22?M>Dwj3jJuW}Jwr#6$q^0hAfe*fF
z#sg6`N~yJM3?}3S@H?`}+@*gNp`gA8P0?dD2+Va~NWZS>{kMi56o5MR^ZUW&?GC#O
zg$M?ZARNxosx?MwB*9Zs!BbaRJ=Yw1_dVeHGvX^f+gYh(H${U2a@$aG&5SZ?-14+!
zcrY1X^~gw$v&8Stk2u0!twe4el>=bw4+ggO;E&*vlmBlR*hfTUD&3;1x7DFZ(+!(+
zx#P%3p0NT;LI#**9^0gpMF{p$@xCZ^s`KKWtJ+x9PRXs_T1JLa66v;Z5VaNPKE*ge
zQ9Tj`gOp+DBvYadDSh{3fRM^RNdim|JtW81aQC^A1GVK$vIQSL=p!p<kR!}gm3txQ
z*)FgI<jysvSS<+ix$URc-f;_bCl!W%`k0Ick1UvPJ3V?3ZZU{;|3rp)@$S}z&u*Rx
zn){5A2+k-&e6nTX9_n`?o}5f1_~6UzccQfEllScfEmb?t)wla!>j&VkBH#~N>rGVH
zF~Rpx1gk8B*k6J>dfPe?IN6t(H1fp(Zsj}3n^Txv>2Alb1@8T(*_T>e2s7m<T(Ayc
z{Ss5Rp~EpAAV5i%+dc-W8mu!#`|3c**p!t2H*G(t9XBxLS-XBg=C^8%ujwmojAfQl
z`r)2q_LAM@DpU=V0_?YQ=<XjM?gr(FjZ>L(F|<AY$f!xV61eZY$zgBq{vmeXeXN|d
z<ANJa*Wx&7J)sEQ>tq9Yp=+l{3^0if<6cNHZ|*71j!Mus1`$4|K?+L+Wik@)>k+#y
zii$?z66IzQo*w*=Pi|dgOriAYV+&!5TPL{k7F~*D3QJ%;n-m`Rew<8*ALHvZk-qOD
z7xd+KHSCHV%-o-+rK-iLM=(cA4!jr;+B4#FGPjA5_kf!;E+yVjPR@3BR`TG@nt9s^
zHfMVCjawk*<iX!upUylHHu9DiGhqVNr|JRf?d()mus_vFIm!#1g+%4^9uG|}CNPs!
z`vd{Z`J7+PfNuH7H65`ePc}e!73pOS3_A41W56|$bF?H~0({{+_2i8MAisx{eC>`E
z|K=6G4Gl4i4z-~kh~x#2uTa+~vj4bE0#hFiY^YdUzXKc^k3RS!#{A6z{<jDuEdgVn
zjY*o{fHP5i5{_l)m#0L~aZ>Oe5{|3<XYy2^kX@O_N_D&LfDI@hLI-`{Rn@h=&7&?}
zBN9WLR>+s{-tzigTNAl$nbGdHMPMnV6vRFmvt0W#R3Z{l!kwsHiyBS-Tx@W$7O>DE
zG1sk1k^06{sN##q?fBoKDqzNbM1<Pc_oPDFSO!72eR?|jq&1W+g+XN?Glrsg`TC9f
z-evB)@mB%;Q4U+FLf~0_L{sKooFeGjjazy4fo-X7gFFOQECK^(yx551l6)H{RvI3O
zu9`B!mOC-$vEsP46Z5H7=;Ot$9o%$b5mva^$LwD_>Y^}9xPYW!r@^zwbXmcrCvOZ+
zit!~f6Dy%IPGYX%?LM&ON+G*k2%l{yD=-%Nt-kx?yuuF(XtZcZ9IfmR%Nl@xCLN6H
zk<V1mQyz%~CCg0*u`$W<KzDa02zgWkIRhBI&ENrZY$BTW>^<N$yX#4LjmW-cEB*6W
z&&tWd$#mR`ISydoKQ59A@+QvyEgDcWARPA5qT`(MXZxi*>rMK{T-EwzrVP3pF`N<E
z-EC#FI(@$*U6p_%G>u38fMBd8z^2}O!Nd4;+0b?BxAj)j1((&uzO7yvb<}<fRATK9
zqB{$N>-UJ(Z!bm@2fGxrYfuvNXkji4YfXEBM#fuUWax+askqHakva_()a$ZgSX=s+
zw<k}s4-^6@hBh*d(6pgs4pmWGDU~3lf#v<E!%FATa|OdF2t6pYSAVx8i3{Rf?v%za
z&j;M2&XZ!CJ`evL5^%qjH8%&jHU&@^CSNF?T+ZRG&D78<OtON~2q2ZR8gw79S@=jl
zLk7+uR<pz6*VM)Mhad1d|7<T&@p)JILkNtvWOojujwRhNFVQ_YcLI}kMJ@96wVS9u
z=p_%DWT6~6Gkl^!(12S2!}}aL^xALKnRthfo&0i#csL*YQ=C8KZcCRbX=$vvg$+|>
z=aig9SZuMwA%Ku?RPPPA+Brx6;Zpj$;0vdRoi>qKH3mHuXR);wOVsi|_VKLcsmI2;
znYA1lbsBja{}7I9rziDAU0=kG%ycH|!if>HyAM0GggTH0DnroO%D{$z;AR&U_f|$g
zh4r5=ic79Z2I_0sCpwd+b)KA5+^BQ&wfQ&*>HdNl$H%J?NLY$ZCsB>`<h=;ljAdq_
z3gPpHGtq(`)g$zY?Gt|5nv1|xZ#~6<e0Tkh7g$nd93dh2e$SppGPTGpD<m=$=yE&9
zy`r-`76fN|d*WlwZvz#T?X$wIPt~_C2TpoKm(nsD&BZ$&gPyBq`cqzo3xY|bcLX2^
z)N6v}&TUT2&h)z3CGUQkd=Tp1aO6$J4tr2-Xa7x5v!&SV?3LJ<U;Kj$9zq*1lF?U7
zy31%5lIkYUXPnlh^oNcDH;d{b!tw1OrT?#=s#))$?DyYx!IR8dN+yzkQ)85MJzl@w
z_2<(CTLgp!E@J3oaz~{qs*Pny)0<w<_-1-Oc}U@@D-gCei?R66YVI$YPg-s<)ny>A
zWEyBH*24ch;l!z7!5yr2f2wY>VZAtW%MAPlRcN3IEzL-K^)<55<=IHROoI{sfGqG5
z1JkA*kDoWv-h2wHL3{-moj$q&SnaXGhO4o$A_On;K9Y%4`3X_+4Lh%&eP=AQPfJ<w
zl3fs3DF3e62h(BLJ*>V?J`x1;roeYG$d}VRr~U;_VB>9t?>X_AUb(JXVZ!)MAS&av
z-%W+}`7Xxx?4+_|7bd-8L@rXTY<nN?qe3Hx4VS_z(U_tj<Q`%_Xql+HRb0b`wo8t~
zve<O|icE&v&To^*B~pWWsHwppr+;_)_+~-FN?&-V);#PQ33>ART}hnF$HYKlcnmkM
zV40$Qhe(29XcxH{Wp}Q*tN=RnJV@V(=i=q<=^AZMh9b2R>Xhwuyxcclp`tgIZ{9GC
z?j$_djx>YK^{|^PN#hBVi6W9;>6Zp^TN%f9A`1_OJsX#^pucs>;*R+oUw8iPIkq(s
zn}Hk-C2|;$s-m8=6d=f+IQ(r$e$;LLS`HHLL(NBM1PVvAmu~9gnw%mv1*&-XGb8Qi
zZO3b!9%6sm8zDSFz~}u+ThRjUV&4BQ`rufRSTzu}#twZK0lfsx=(S{3gR^A<PW>BD
ztp0_~9)9Duq30-yc-3I1e<yT>h<;i7ynC<BGK21LHGX^G=&6+2$!+Uw0&06OBm|XB
zuUau+!G96fM=iW7lDK9FEGsmbEmQe-XSan_S`TL{k_t^octGv)P6li6QeDh(sd3i9
zqfp!%XR2cDhbP~dpP-)4QmEE9URL$(?&N2rE%&6J%7CzW(D*Ii+erXXA@emYvu0_D
z|C_n09@@SQxQn#+#ZJy&O7IU`E%oW;#v^RqxJNei?z*T*y2<sd{7&=Ph);SWiN^Cp
z+*usG!mzG4ZtLB?H}FN;??LJr=Gmjeg}~ag+F0-NbE8`)4!rLE{;!d@(a-VmgZJq0
zDHw83$5;?sRg?LvsP*xc{r-G;{=LX%OGl)KNr8jyFv2D{v76jAcr@R?7i1H^`jQEX
z2p{d|T16yj9t&O{#uM3%m65T?IoeYqbO6hoN4Ig#LnW2}DQhfkkIx<IM$$9eKF0-i
z7GKBykCu=ot%^D5>gp<;lA&1-fATThiUmr6F7oFr`yM0YvNop_mZ!-j^F0*-Hx4Nl
zMezZ5pSOItL~eR(7e(9F(C(JytL<4<&~_?IUO`g!u^K%~ad&)LboOIYAD1J4N<o&&
zRO1=8?Gaqxm#0|S)@&@bQ9jihzn~YsG_k2ac>l%XxEv<%klJ&HH4z7SRW(mp;ItL}
z3P2Y2e|RUL<{v&(dQh0CY3N>HVqknQoR^W2_>_qP2gio?48^Q972ZS-me7q5J~9fn
z17=+_`q{HcR9)%!78v<*M7T225{R7l^=up*ePaPCbuT6~pf^nJDjtIk&`D$MChsGy
zadfvqp+o`LG6_Z`Na1KqT>>g5g#Ug-;g&Q<D$M3jtRWM=8;>OkrbRFkuWecQynA6a
zliT()w_et1RBvSF+oxd~j7UqQ+ecmR@RDj0uCF%sdiHD5t`FPvY?j+;&Zf*?j8S-u
zsZJW{=d(4I)eG(vZEugM7Zx_1f0j^Hm1}UnE>|w(aT&IPSW&;@Df4zq-kx!b(rcK2
z+7JCyfAPZ0Ic~;Y$nW9PekZH`xvXJ>TVXDipd5CAf`^~t{q*-Be3daVbbQ9+6X>qB
z#G7{&UX?N~aLs+{%0HF;f{RorZaJ9(FY0%yKU0L3#tIP(Y(3mS#jas5esOe}!ZsG`
z5#}4Z$0>kr!w8y0tdGlUJC(WMy~BD2=#C>6CsNMXNBByj0o4+3IX)APjFy8CwAmIx
zo~RE|*MJC_ubd`0ZCc^oW*iiR;HKN?S$+-h6yw|Dg?muo`?XWuRIYmAPh9`~k?vGi
z!vBfrci5;DR1;P%krgiS0vhljqEC^z^BW5-?E~TF8vqgE$ycdP766fh`ez^l;3q(0
zC<W0I3I4vD20MLYz}}h{Bd|`A1F!XgH6scerw!=zKNi4H=J@6ea4q9*fKMB+XAAuA
zuD5|4uLXtx={A6?nJClQW|E4O0=7#(t=(Ro$iOfx1an;AT{r@r$Yvl_;M`%nKy$Tz
zSN3eDv}I--cFL`0-o0{K^T>9Evf6%LZ|>`oU9fA+g%xO!2-yzAn|{KJy0Iyy@G8xN
zcR!Wz_<4u1LdpAXLh%=<5&@J=<%RV~dxYMS@kOd7ejxD$US~ke^BT$EBz!E$j^%tu
zWz2mq>!#G(zND-H;S)s<$4qkvBW0Q|__BwMOGe+>b(?+b=3^fNO4?sle8qs7oxSS@
zYM1@R(WU~H!0IQ}L9f($O?SOh>;a)jPN%g$UkMjpM~8wJH?bZB2Vqq%uWYdYNQean
zlmc$`%kdJ!Su*k~a>W61ol;PIwl_+;az9<G92Bk0NCPp;%%>M5VH|G*7yY{U$I48l
z8V(+bKOO$G7UdB#vJW4;V)Tb5gQjPgt!4omf7d6}s~e<Q*g4p?-u+Jcz`6e;URe!n
zuxmU9Au+ri<a8GUtorlR-X8orX|SZQcTeQUCeF3V{eHcNhz?;{SKK&!gDiz~BA%UU
zPHeLs7S2oxZlB&iAH7rXR<*`94x~zU_-p4}n>4n%Y!aRZpcAv5BF&!AH}(qt&gJd%
zXncS5j1L>Uhsro$^g8T#f!-8NY&FQA$d4r*%&Hn@2UP{#q6A>Jg9N$&s>}4@(01jj
zK)aOEPi{dW&2u^8oV%ur(nrIJGSFsDXROLv{pP9^M_>ux0UCM{iofXv_)5<cGu5lu
zV@zT`mLEeO?rR=*D_d`(E)Y*pHqovQz=wDG`ktj>P$6#;sWF3uPnAlztv2iuHWSSO
znIvWCVcAO>>N8~;HY^L-TjnuVj1o8jsWJqe@Nb(w1f0?iVP4A>O%BPLrF{@!U0^T5
z9--gFpK~EA>Nsp`988{FOn2L#n>J!J9n};((5|0JmTva6KQuBYcbvGd^{6c+iOnSL
z^WZ>QtaF~YAEIdej#l%s^V$ty_*HuF25$fzcC6NEWmgpaj&YxHUEi~MC~R3*7V5Sd
zC!4#c+jVjR=1sGAwKok2+e<}m<>mX}-mSSA`SM})HO#`hQ%+C+xmH-uqkgmR9`8ix
zB?FRI`CxsymnROBMwjkG2>T+2Y94Vs?$8g_X2mfVa4W(y*zC_XN6TKerp|{C2ZA4Q
z5=?Sm#}~g$g8DST(P3F28++QH*gE!tYFYHCo!6`{ULQ_fvZUf5TT@WHwqbnecag5;
z`N8#$f)xw1YLmBSp2JzFg;e+cMt?>SU!dklF@}pR&!;jZ`{ue3+XUa-U?fXMoVe0r
z)EA_3rAQ@>&V|{7yn0Urs~CzK(5jp&_Lc{Ph=)8r!(EpIeLGrAZNp}tfV;+kcc`A%
zRL^jI2a`e+6^Vv2`dR;=L7AOd>&rCRdbjH$GYD1r?w}%L6Gp*Qb0_v-8`rdreRB#4
zdb6<Xb5fuh60cmK%yN4CW@>9xVO;FNI{nE;+@~pzcM}@h={~34OwX@YL|l(u?$OcP
z05julq8b$~i)w1L;e*(n#ZL+k4C5GP^oo|PQHMF5@{eW3CdFd@!Y%0z3&QIjXmHTg
zAD=k?cmvKI3s<eUd7M9jLDl{JMT<THN_cVb170JrW;p!nrpBvh%SVQ_Vu4$eF6i3I
z%NR(w0vO4#f!OG1qV2Y=FO)BLj=HF@(-?z9ii8dtM`_IIQ`ohMTDvc8(wzT1CxRuD
zOx#KggyEO~h+h`2Xv6~iC%eV?%KX1+4rRP1h7h$cc1jvbgp~!c!X?636APn^NrL#{
zG~*zI69B(m1Qzf4T&EHI%P4(zIitk2gZ?bjV3PB5<QTv~zsSlg<fVJtydw9wITFSd
z@1R-n@e<q!)OpiUWijTSZMmBfI6?#qHzk-=Vv2b1NuPLtb3aNS<+7)<Vevis+%Oyk
zzE$|10?H5bI0<1@$^|)um%o&S_IP-W^OWYqV5^FO5Iiqw`~}mwm$jzM`=!_u%^~(^
za)A+=V6hawJYvRZ7=sA_p^$3%ju8E(w;nL?vUzFgDS`^8xb9E$#B@nK+PbI8%~@xt
zTh})>-`_9}CCG?}bI9bpV1xBjn{foJV6A8PC!~FtDB%fsiSKBG^oqzF8eiXF;2d^Q
z#^(sU=QD<@)2W*NA=$K?>U(?Cr4>sl!U9);XcJ!}%HNe57pO=Ct1GE(xj$EDTZa&W
zokkqwR@8F0=V0rf&luFt0(Vk=)%mlFF~hetK|XT<&2snyw%r~7JRVLz4a6qYcR8I!
zrwk&v0)|<AziYXgOvA%A(gU5>WX#i7K8BqjQ+j3!E7xPS4)`>k`JN8TP8Z-FM%v+*
zFC=z#3D*QX+!F<*PS-9}I4VQf(C_$Qv3pPCmjlJyCV*^Vi%Y`iXsAln-m)CY%mSA|
z{~A*W8_s8DP!2rdq{b1?e&3OES%jZaWlJikq3GD9`5yP2*M~K*tC3vqWl4oG__utv
z)&J~;O9CYo@vf3FnI5_r7=GHx4p*R;q%VzIq>$E_8c>8i|GxPKtecc`s5mWO-i_u;
z?|{kzOQh5(aYv-n#T#H8knVYd-0sV}DBt<XpGL%(81;-(_$gST1U*b&XU6EzBH%_(
z@AGD{Lp<n9Pr(F7QkW65d76%1ClMz<ffaehjYQ%>C1rHFF}2?1q3U(_eJ+uu5kam$
zL&@U-bZ=!c$FK9oq;4@F(V&7pfK<PSb-Mmnbmx&$c8L=<5=OV3U%-RklVa`5xANYd
zjY||l_SK!S?B$0PFZ6SCx$pc8T10(z(u)OOoV|Jpk4~Hr>8H3eF@1z7lAQ3(AR1}e
z+GOO1-EzZnw8Gx3!@zd=vjQZJ=KZR8VXcE|eI6m>vFjw)m+;e>L85+<#oES7cL`rG
z-J;Zx`;cuW6QC}68Ep>|$adBKcQ1t2OI+4(hxw=+1Cwi{>n^TRbASBkoAG*kRac_x
zLhBbb81q(uY`<;R;o(TM4tn*j^z31MC_A)q!EWr?T!XHG!ApKD0WpvH)RkoJnhvcj
zXiW(&t=bKUQid`OU?;L+#!5Ky!R>kpFt)c^;&MNEPEYDMnSuprPO-0levFWkw8#RJ
z*?9FqBy0%-rY$q`)kY^h;-Vh_R7#g&i{o-Ehkkub+bmmB9U+b`2Kk@XW-t4-jr}xv
z66D<+TO;;{c`2o^H4z~~F6(_$7dY3mfYVX@EYo>Y4|x&MGm$%A=%Viio4P_9#{Hzz
zW#C9z>qDkN9@c(<83~qQ8Zwpdr~arU4CYmO*szErR@Vm<n52e<0VjeovOu{_`1f$g
zX%&qRn|a2*M;&N(CDHLeU>n=5tZ?D!3mr<aTsPV?ZEzLLO}+}Bjj7uv=PRgn;=}0G
z3SR8b3pt9w!qHXR^HQ-nJjZ}c=VxX>zGC7GV?%CQ(b+@bu<B+NqWf4Nlq(FKVjgpu
zT`<He6fsPjdExz&t>n+GlKqduol06iq*A*@!Kb=yU^@~+O4KNNdV@kSaXz>I45JDp
zHpvEPIN^knL4gf&bTTwBJ-UM7c_ny^-c|XEW!RmtfF_Xmj=$4ROhkSX_LHL73Awma
zFfw*EjPjR1Xm~L<sMKb2PKYT4W$Wl_xolX0jVi)N(y6j@%IMT$f&ET9(zcmIy!oq@
zCi@@ULiU?O>X6dlizv^NmnX>VVc6HrYpG1Pmt&$n=Crer-)NaBsVKI^ZNn~>SFR7r
z!hm}2C>G6IBA}}Wd?(sZ(_J!PF6pb^vdM399K*FI1SttiAiB^6My6fM+%d)q5P7{o
z(JP90%VWZ$8u|hTl4DM{bZ=_K`8gWtW$Z+7&bzTGD==UX{;taP;$vd(SMB#dBzcRF
znyL?)Q;ImvEAYF0hGnjKMNbq9I62a_lx)5SnUHQSr1`F)9YC2&lQy`Lh?I(aCt)BW
zK6^RHp9zTUs}fjx?l_Yh5`j@cJm8e7iz1N)6-FBw{|(P4CDO~?O0^1g2fC36&+}MY
z>rtTtTiQm&vO3QO_ml`%1|OK%hwDkIMFDPyo&-jwhHa-m&pVb4h|Q-~dyaxKrpK2z
ze|c@?j2&QuUW9W<DE3!vRqoW=3`b}<66Q#0TAeecFNa{zkMHHvm0mEF>I?H8c?snv
zuo<~}zdlOW<Cm63_dpdG-eaJ{UJ(Sr6Bvjy&^tbti+;5Yk_j#6gLeK%h%EZ!*mJI+
ziO^L?Db&6Q6S`FV^Fuz~cCyW}a*L45nF(&0$&C5j2yd{5y5JFWpq~fzJ^4t&-i=r*
z#z4e1dj|k9Wbf&4S8!JH=atRq@m4n3a8M!)T+{MCsWhXXD%-YJ>rd8R;MBi=XI5*&
z$S@|Y8Ft$lOGWierPl0&zDFsJR3TsfE&09&vs4WKGD@{hZ8SEs<M(aC+CaOp{MVr-
zaNW}v4U*JpkhOEbOKki-<6`#Sth7zZf#k7tAmnVC@q<b4;eYS{jYOC*$aC2u{)IKC
z``t;#L)HA^tCHz9jRbL+$fP1d*l75Vp`M8-Nnh0vwk4_Cz8pS-tuhT$B61a`S^T+Y
zPhRqA<xlhzZd<;&*9*UEEs>RCMH}7@rbohLO2lh~hc&~MG|nH82IUjl5vY+oWTU_a
zjy-1n(*Fcp#0Z+hU)}(Y4<`~Yf&MfWxxnw{+y?MTCF1Guuw7!oAmrQJ0Gr8)R7<Q+
z83+Qam>hi~m#N!OO3%BmQ{O2RwGo=4g7i}i_WQ2rM;I%K>`JG@BD`*6T%qSaA{p?U
zV)VJ*K65cM%F!3?IK|;dVm3sp+ZSSyQ^JI>p1ORerd*wO(p2xysO)CY;;V!o2}5rO
z#W?Ld32VPIE*95=2vnN4p-#t(!Pr^|2-SwOvIF6G{-6VmOe!!RXH>r0QjQWJTesJd
zrK9)60^=PvpT_oUA=KYGvb$5~cj4clKv&zl_%PM?{Y{pY%h!42p*L}IUMBbAWDeaO
zQ!O+(gXhw$dR6(V9SN-ZTKpc$DAI+%g0!BR@bDd*2I70=l?v}`yhOK2PVC3C&IRK0
z7AsH*U+%7Mm$hN3BP3vd9gyZF<si-XCJjuPded()`QGvJs13U(x>!@Cfly!AxAy)9
ztbLAaHhBxYo%lH=C)lu4Lb_V5@1LuNhXO0zS?YZ@^U7tHwJIh;Y=by5g24B3>H3Dp
zPS1UuPmNY7J9QLz^KxvqH5Oj5USEwW7pA+|aS`&8%7r1>*Jeq_scYFy+}#it0Drq*
z6K^m}bFJf3s0zQP{Un>%>|WL2Eg7cg3*cU)w-8d_-m7s5xaalrem-4LJ>=o*z^pzv
zTg|FEnn(OI%{@}gL8V_ruRDzDenD)nG8Q}<M@{bY1drwJNM4w~+;8cUpYH$>a2tX(
z^IEh+LZ!j`1Vl)sOCvd!Hd7pT9yn|@YrSVKglr%AcG&}lFYMYoR+P*J@A%dI+dd!`
zXtDbFKTr08C%157yA5`V-KN?^&V5;%iM{PU*+U%tULsVyn#e#}$lC3}<o7f{*~3?|
z1Yv$~+2|f;WrIHe4IN9*SC%=4FiSTkHn!^dxt-IcW+VpY+xKNVzx&NKkfbBdMhuC`
zzxVqOVx=>~(NtWd!)VOey<)-Wb6*gY6`3p-!B-V<6fluRZ#zZ!APN_bI1$kfK~LNs
zsqEcC*q;=5D%Cut9hg<fg%JJ90Q6ga7pmj1-)S}^$uKlhg+Tx-s4iU>v!hHA>Mc(m
zRhXnsAJ~$T-!D72ev^Ih!1LP$^mnRfqIOd#XBgf)L>mU%QTrgzUUm9>Ym`w{Ka06`
z6Z0hIJ*;3i34-?6f=|#1$jf`Cp<mmLHwtDyN_CshO%bNhcuF#xO~TtNi9%3u57bF=
zm%l`IV0<F+{EgBxAN9->tAS2O@%xX9+}}6yCUdH+j4<lYs(po_rpcm!DneFE^(nn8
zvuN35oF0;%uTn+@(k=$|aPlbe;MtFV^jn$DH(qdedrjSAU?ba2m3`xLpv<Q<iNF}i
z8)iV}Z2{FT_1FmsJ=c1*ge!eVR{uNOas)h!6Z%CqGPbn?;^u?q8@(i{5W*Q2b5n|y
z#1<C{ztYv03lS@~9(yz6G>(q>Oij1Py+NIDMHH`zXT}I?_YWO$v_MLdg(=%cmHY*j
zyFGlTOP}?L3`DhPH1<Ys?3b8u0Rmw}kOvS0vWdw&D1s@!AE0nidhRHtkiu{E4u7>D
z2zC>&<2OCofwr{NY(}T5qzfk(W1rT4Ruz&!*j?5(bkn?w<?Rpe0Y`gh!XzMIaw-j>
zgRhWcM>m8w1?gB1Cc&~D^H|ZeIQoU!rx)D&HzF5v_*G9B(jUuL!tp?=6tMD7Hqlud
z+qvA~T`g1wVbM<zb7`guV5t;z`<yIJF_Dt_fQZ;qncZR86yZaab}DwK@e-P9`G%XL
zA?3N+-z)&K&@+ssRlEC-NZB@vLsDxV`_pfg#3(gO84UfcVQg(l)LmBqFE^z-#ZWlm
zm+ge!!@V~{qkR1PS<lM4v|Rg_@De<SOJO-{esJsV4;KM-H~T<Lq(4oQgSH6x8K>~3
zu=Gt<DnU@91x2rZB-dEXHuWT27wP2~Yu8xeE}Va5wNt341X8F|qr<c_{`5d_(}kc4
zVzD0LsFS>W@_2P|-|Eh(>|6j|dgesYuo!cqL2G3Rm6yf#x9!W7yS-cg3v3pBfu|hJ
z+ZY7xpUiOIB_TSJlvgE?u}9E%fSLJMnLe-T8%;3|lIUv3`5i_W-rr4G8jQ8#A9}Zd
zWv}8rEY=pz%jm3TwkD8xBi~{(HMs)gxwPJ#pC4|A+!`@VuSzi&$5ywB$15hQ?Kp}R
zO!9{zH|zKZOfFoR3wR#c%m*CQK>drN4vwRf;LSotCs4(01DhNj$lLh#2_bClWtL=3
zsn<7md1R2NYgmo7J6NmV<CtulO-^YpJ^e4l3k{5vNni%xkub^@jm9wdA)K6sGm|)7
z5rcQ|U&u80Vb~3}wjrtp$)4+UaWG_Gt=nx!exMdgnLUmF7Kew`#*>F`Hm@8J1!FdA
zx#F=k68dovaamw7y1ikqT7NNiq}}oDI9TT|jvA$`)uG}v3z5OlAiPGXt?m%Q7Ei@b
zyg~j{&4sY$e1;0|8Gr0}{D@6;>whshweQFWqU{$HVI%=C)xj{}?o11ev!}fIy>Ozu
zy#%)wfeG8<gnHdZVcYmeBQRe$Zo~-fAHGx)=<3AJqnH!pOq~%F7q&%ny7s$;p9|vp
zae;jW+pfU!_4(Qxnm~BKTqZ}SY)wTrhX75A#VqX?l<@RH+rTN5Xg9`nlfCYO<c_V}
zRIi=<?S{N<1<|^dClxI3%IQZAn1lCd>FTyqamR30h}1{G_pP&vn$^mG3gDNsdm-<(
zsWUZ@i*~hZ_#wJ|%&?bnGw>Ia63|5cvrjOMX+j=vb|wlmnRkHJRTDlwo}OVkijf`#
zO5n-XMtvLXRsi;-A|c0|H&<5y342vucv7_@0SqrEj%P#fF&FCqfTm)wP^0VSpru7P
z)|JA9+<9I962zDY&5GJSbLw)5acbd@gK0F-Kf<~IruxTTIP~pX&_ka$wYvmNTP9w?
zLEcBh?+cpEyDbJGSM0fNhj}b5g;7<9|I9Eryj4plr}iik0zv8x$#7(f@6pzOjtC+L
z25iEZ(?DDY_yZeZ@$P|?%>qM{bM<rv9_Q*|_<T0=z4I-k0mjT+>iPCsy1Dtvv;FT?
zK|{k1pE$X!{Fu2ZJ6Ru965f+V0XS+>7wwJvr6=vePu>MIZuq;%&w<v?5epJ^lsKyx
z=oBeTIJF~ZdoA&B1!m~_?T$7r@pG5k_8Z)T!uYT5+BHhYiPw2_Mos!$T|@neA=<(l
zn4~;IxgL|OiLi~PUm_0($h)6VYx{DC)3!VIZIj*o%2@PC94lQC%^FCOGRJ5&AUp`6
zZJ)Ah;cKrLNH~~#3ls8@1oQDnj+BI(Vf<J2Y`F7)K!=(dKBX!=2UeZiwutEU<|R^A
zpk|Lw)yN2!9{yLIVtrZe885exll6YSxU^?Nfde2Il=OVg&8X^`&~|b;nLS#82Bu<4
zXM;}%_jK(M1C8sm!m;b*){jUP67UaKDo^}U&Z*)TJZW2Ci~PxKd2KoFWO#*3bFq?#
zVZ?R!uV&jL>aOnL4*~Lt<Ip|~E2R=IAH!q`Bu-WVujhA71KaDYzDC$O-AJg;eH-@r
z)$g>u>B?Hk2T-{j-WMB*cr_!a+T+w++uZTtctCq8O7)2yMDBXoEeU51SXTgz;lQ9~
zzWN3Id?2jw6kj-HD!%~iVyF|PI80aO;yrRWlF3x2&{tCw#CSRqsrjqgSW}o2{4ZO|
zE97lDd|;=cWp!=RG(Q1@gM%Ll)>Rtuu(hm*(<Pe~Ws-B`3m?QKlbaH53f{4VOmSV_
z$RF*TZ9r~&^9hE!1R~@~k0&}jIi$b@a{?8k;qLm({Gd*qaYJk##^`7I>iOqlDvViE
z1SIo3);Tc<mIMSX+y!t%BE2LoBw&}x0}~6$l5QYB5Fto)FKbME=_VU^RdEA(D0bLk
zejB8l>eRJu3dk9tLP$FEV?73D*-i!OKcYQ`naRVyvEj3Nr5$h@w7}&=+5YUK?{LOW
zpa)C~f~B~OSoNflPx+Zd1_F#puFU`8=`7f)>e^_%Hk<B{kPt+=yQC~Sq)}Qzk?w{~
z!wU$~(v5UT=LV!xx<M&vB&E;vJLftVe?V}pHRtn;G46q~f{u3d=^=lJO?0l4xG=<f
zjfK!K1`Uu7_p@O+B#Z1=zyEXKE6!3y$DGw;UgWQI-a*})YT|vgT2F_>nS`zC@JxB^
z`a7KpKkMN)Fv!(kgIboa{KF}IlBjBaa^kECq?5zADabG2SqAAG`VBE;!(NqF&uI#k
zQwtOgeQl0bJKjC|EgK-dqGADI?Fi|knU6<cr*-oO;1Mo+t*2!=^gNuuUa#r8jDAay
zmjtz_7^L}GLlANT<{@hCb^n8Wn*s_vpXdWYgGWRzVBX4=_7Oy05MMV>2yID`K!z1<
zJW|dBZug-k(5sARuoh?zHv5&t+VOkA3pby|UWjRDQ)RgY_SdD*jn;GxRY(v0%{P++
z>=0`=&kKh!<xk@>k7`8w^vU+SD8Y`Di*Oeuaf>d%br)HmTnN0hyU(htE3t4QQ*6^k
zdMUn*sjgQWw{qG1LQ<1C#nFutIU`yQ+OBQM%yxh8fns{hJg%kr$$p&s*|A^3qsV5~
zvZWZ=X4id`sea2`ftgo^v(9Vifld!i4OWGjo#%D6pq8--uuvXtd+*?X+|d3>a_OG`
z{m8nlst3i@lchyGgrjxoF-eOQg%{G|kZVP>rw0w;^Z(v-drDwk?P40(W-Q{N$Z}hL
zWA}Pgt5W(kjnB2sLmzIgDs+P7kLmW$d3{!X>~>k{J7vxk*M1gyttH7mqb$6Q7wfk8
zr#B=u%aiX<<iGn163&hivv?S3QWa6G700zAu)5^{9V!EoO{!!0E-KWxU!T;wdUY1)
z@91biP;e&$A47T_=G*$pqvR!6?gQSXugXpaiSusV+|_>l9;z_#q#jy-0Y=}=u`j+M
zj%6)vKI|Z7KPvL6H@f1c_MOo08mk$pX^dVPa@#KZbg^T7*e<>l>`B8#Tabj`=9j-0
zYP|}}`8X?|gjbYksAN7x8YQJcNBA^i@q!!>ggIT$!sqborI0EGr|0SKpQ#&I%xTC1
z487O7KdQRWVQ^X45xo(XGfal9@=Gu&!j^IDL9%czMO??tCPi$^{j;dWKy*Cf6Udk$
z61v$h3%w8rUtgXBg&||e2@OrD!3BFQiyU+FH!d<J)9-GMCS3v_WfhG@9pwuYhEYeK
zPVGGQY_B!ySY>wr%Srhi5!!_D>z~PAZuQ#~>2kA-Ik3EenR^tySO}6F;s13T`d9+j
zod7Mz4n-qXm^%%=c!=>4Q*#>>NI~v@I25yJBEwCt1vIIwqp7<nPT^gW6v{4qW_^U(
z^~OYzY|xhd1<pa*Z60XFwto8?gKIl5z&FkD=#c`QA<vw}+Z(9GSPFyPg%H-9MvFst
z0Ea4Vd%O*nAvFaLbXd{aZQ>^L?`G#Lvm(P;-0rjB4^%zOK>LbgLB2WV6hlk{+~Jc?
zdg?wolU@Dnm<V!?dSht<Huveg>wV#A5%Vl@{L1KSc!AOB&u3^p3=~m~cU?4zqmUu=
zx2rUIjQNqnccPq;WiZxD@AURa*MC~-HTF7%3wYwdvR<RK=SD-Rl`8+r5MA%W@${Z`
zG^^Yl4Lsh)yhI}scd|`yh=%>e)ni8N=HUq#Yv5I53COjwKf%{+Te@-p@2&I$0I&Wp
znOR63l}HLY%)GyOkF$YF8u`BI4%<*Ups%ccnXPW$AL9{b>?;py_dJK))TiYNkYP9$
zWokptz<;Fdx-{^hePP*dCX0g%I$SE2&bAARE>U#L^-#&t|0}E^@?bQ7aY7Q&a00&7
z%9(3U9XYI>_LQLG3H^{6+!M$v+|BUEJ6&?C|DjtS-1cU$=UMva!_CySZ>!%eR~8$K
zLL}ULI_y7u|3F7ViA0AkzHLb^JB=y}WR||YRb+Cu9q5z5Q%YxRXC8U_ZIFwqKUKrG
zotxGt&yCq<fb|m3c%+qiW>O@b(mrq@%@=w^=J=}4mh;=lQ&j>p=lNhi>zsuaQLpQU
zBqOdWVTl0Fe{!G-9I(BNjEuZ&Hp)p{Tt?Sx3zUWDtb#f_qh6wtej2WGweg4ozN@y(
zo12qyX)4!N+}{Ux*uBi@5ByXDYGZ_rwp!ls`%a7(D4UznR?OaAtVG{Sa4<tHu@UVJ
zJx>r1MBshk26K3=Vqe}%M5dM3gpin2?L=|^c}(5oQ@#&b`ImA1hfBeO7$({HZt@Nl
zGqQvodgY6<*7lu`dYZFRgACHs_erbNM?dPUEuO$Xlv<pWAw|@+T#9Z1J3y*koiAj_
zqu5Q_U+Y3h=(AXze`=pJ<ez!5U?;^YA8gO_S<KV*J}_K~89HiSd)oWVzl+$I6B8md
zMi1wuLd0irAeYZ!eNU>C{4M$7h%u1tjf>q9D8K28e<S&P*9fIwE;?Z+wp1b*3h#)s
zHu94U9dj!0jon=`bPlCvyK=D+?aMdz_=ImLDorPkm$t!MWYT*C^zTfY)Ct2%JNVy+
zGI<WwAhFgq2mL;b-@n6Pw~oQ1Rfi>CH=H=puBBuKp^exsx4c2U$+kw=-%!{#iUpU2
zJ|&QXxRHktGK6%Af(EsQwQ3knLCw%w-v;T`efaUbWo67y67HVcAZY}4bHU2^2N#{9
zZQBK@Eo>H}Pu&iN3VM5xfa{p1yhe_RZWV5GTwGE<&J>!2QHXco*Uqava0>9$B*M+5
z2F=<E7NKjczotKZC8sFx1Fx%%=QjO7!nAX)OjK;FxsN-bmCbpDMx{Hfg91CV6Fjks
zX_z}N0SdpPZU;J=rLX6YSLBx#p_gT+xG$&3r?n2;dc3GFs3PrhgD^qVlWe!vaYNy;
zR*P||qU)SFdEH{Ap>0|U@;X|lP$YB)BYm|wq~<3+mBkIRU46OfQjJi^h}DS17VNoW
zRw$25Noe@K^E`g#gQD_O*-jUO-A0S764rM97R$r%GVpO9Z}H`=k;AKvhwov}r+Klu
zqW+9ZQ9;kOeCzn9&RzG{-L|qri(T(yi@b~n`fc%UPBk9R6_`Px{L?MR3f``@d(%PY
z{?A4%xi9>-(hElW&b>bDBt@dHDr9(JK+e~Pe&77=&UY*54r6-#>=s5On`b*01yjw(
zvN*|nWg^jU1S&0}eEAhZ@jVemMt7HvKD{)47}+v!_(2>q{psM<tNXcvsQWY=>s4KL
z)8W1SinJ3-K_O!8Cx-23_3iz)>qd9gYAsV3wEDbD-0j~>FtCeddICzx923g=L}Jxt
z`S{rR6!L>YvFL<{tesK5RQ*(;duT%X&>?-|*zcV?hZMe-^bmO1`v|}OOQXggiNH{A
zUazMZ60z$m2z_tM0lee#J}A?1m>Z#6Jqc9e*$>}&3d7#~*Gk3!i1;S|K`&X=fbqxY
z)zC}Up9j!8OZ5gL!^a_1uaIhX_E}CSxojxJ=>MfEe#T_T&azVJ+o+&FqH<$>P2tb+
zpem%=ffk_Z&OOYY@cVy1puRM0`u0R8`(WbO^c@<9)W|99VWbwA<JA9SBuu2I(^F^=
zMj#`X>8{gsmFDc3WAtO#QQ#tl8~4D$%_)_L-#0egmIQ*i>Rx+Qq4A~SX}WMy|Azyk
z>G0Fb?+;4CmWHsf&s4LFUm;M1h64#C!-hHbYOS#uq_gPB?CTVOkw(jXGzYs;S$n-a
zMOme^?LXwJJwa5E6gMJL)ERJj2h;eTmr8s<E{$9IAH|+}&QF6=Dn-orq*#2|cXgI?
zJ8KkGI{yz3s>(c`Cv-}9>{a!uce1l1gy6xCy0}|U)oolH(sf_gI<%1IOIm>Pm*ual
zTd0ysSTCSWVuvf!cgLE)F+~2PUN~4fo4B821(o-a3Q+;<w=d{{%ur9>GLIu1Y@=WG
z`OUU{l>yN4mM2}_@zM*{)bj%L=Ze+=_h`hXb@UgTO)MT>cwqJ4S{cwNGnyeWXNG!U
z-nHZpd^ehhEhnFze!Zb^QatTYLv{KLg!gpi|3qPpaqlZ0#~{o01mCS$blsxxu85zG
zz0k<{A%nkB_@q>}q4vq95NeIq(}M<6g|=rXm5)9zOT9D1i+wfSeF-*_Q^1Hd=czS4
zpIH{k9E0pQr(7V}^LkrvYv3~d=6%&4NznT!v`F*!R|I$UE=y@Mc=YLYpMX7uh{AY_
zVxS#ivY-8P!}tbc8;WVF-vliIZ3rPvk3jac2`0%`RDNthKZBsvkD!zc+v6qCP;{Qd
zixeGM*4{i`uo&;-#-nypa<K_yQj*)RsKFX~r3crm(*^tN)`JmI*@|m#9rLptGyz-E
z;zlQ{A6L0@fWY9bd^`4HVe)LouIvFRDH++<+)uHGG$DT<@}FL`sk7N)53}PbrDztD
zadDU~Xcm4__qlpdJ)@NMLt1NG`fypVUnswxB;7^Z#ii%r<$asnZSaYBjTN~!uIqTR
zU#`FYg!G#)`Cxk<RCZoxqtJRUDFsk?Y>(}4ulTCT$>P%4bS17&i&R$1KC%rNe0!r^
zRy~=ZQ}8RKx^bkyHX+x)^}JZd)#TE{u;y#z-p!RmyW_`<M)hz#8SvZh9V#q2D*nB|
z_Q_^2D>gT$+K&)98RGo7tqxiso8)G*QZ(yPrmAwq4_UHpukw=wNKLHNcnNnO%ikT|
z$J1vJoI7rf_)V7BE}_&;s2rd4SRhgFK0k5dr2-e4Oq9w=cAhL5LG-5{^4l&w#(5jK
z%$OKi@yUPfDc(ZGj6Gs#UAR>jEYK<RkuK>h`y`gQ1yrK#B%6LxgFbqqt-gYk!+eIl
zSjR*oN84|X6&3WM^RPxiMsX$RfCQ<LhD$4%hQmehq#)zi^a6(P{*JkckL?}lyk0C|
zf1@9cc~p0LgJMQC#S(T-f>ZsKKOSq%NfXw9yBgjA04|0#;c%JGzY!O6sv=0vW%+jx
zXPD3sIEdF3#xKwsea_HzM3@Vhs3vtE$EZLk1oXtKK@^m43{d+g!kxf&jO88DNoDZ<
z#|rVT{iR3UBSNa0`AQSGUTec43$H=(;};>79wZUv$8gpgK)-oE;Yf!jl*3L)XZvo<
zIrzqRHdmH_vw?850QD`H6W-HbT(YE7jJ~A9tva#5gVK`uG)_0krH2CFJ%B92dEd5h
z8GHa>2Cs14j(?x5*|i5%zpIei?@C>AUfo5D9pY%vA)CLCf3OStHBRJ|-N{e#UTNbv
ziYi2smWne66A|@2c7Tb|RZeIT>-%Eh!iQi7pRufGDm7#G(34l`6Cqq25yW!FGjv)E
z=iWk;J&@Gr9^3WpdU+(boweio)q~O*ndnEHOe;*@qRAo-ImtbTD4RszSFHkV!}k}L
z?SotrP<@?_<4yyI`JvHnqT`$FuxxS02*4X=o#dTCwYpDtooCITpVcQL_1-s29N~p_
z|9%aqbNDsmM;}!)jaEs89AOE#(@7zb>F1uAkaN;M#(wNe9y@;nTurRU`kx&i-Xib?
zqN1E<|8`+{tiGO2qn53FM(AWv5lQnjamjl%>itL*dL(*r20Fgj*A$z@&u{UOz4Nsl
zk>AxHY4Lhbm-1a%aM!m02d^1U5tHoKcRFG&#VU2l^jC&j8btTc=)mUMW@p(;qg@zH
zfto*-_Q+@80em&8=v~~REIbfnaP*E;XOiw(!Q1qkD(R4Uu7Bw?axFUPE8yi`bn|dZ
z9Lo)t+RCaY=v7KfCGkVcc>*@a{T6tUlv)Z5fyd1}Zsl#GUx(^AjSOYZKXZPZ<9p=1
zS+)So!cTJA=&Og#@4dnNT2f@-Hhkyl>0&*?E(=c=>$mzKO>`qvs;N@Oud{P?)_y@0
zh%xip@7C(?g#u649G>xC(fuRe+dAUUsf+buW#Cy#d3k(#eS5MnbTezZ@Xkcgb;V%C
z)iLO|W=W?Xcw4<S?3VR2WrKrKNG`*t7zWn!Csd^jI_z;}BrE?wzcJRFA4rGSFUJx^
z!|@v@P>_wl?bQ^xXG-zkEtJn1en$?$sgTV)cd#Lxx@H!R59)V3At)_41^N|PoWS6r
z=6AiY(vpOdso3+PlF<>`Lq5Fcrq`Vn`8>N%NRFHTp<V|$#KcSEv8J$9nC?nB_?oSI
zG}s;i7xjO}Ti!u^8K4zrTKT<JWK|x2^*`}01umHTk2=upuCwrx=cl759(G;=g}{pY
z8lFufvF=N7DnHbn0TKK?d^V}X$gK?`6zZ%)bj)}X1EbFFib$SqntkBsGFwe=rZF~(
zM=mXXTUUu_%P^HidksS#@*D%9zUs$q2JBNU`9qlQERbS^MtS@{JP|%$YgA6*lu~L<
zYZ&tbN%Hgw<h`+Pd~z9q_>uH^+m$)Kgaf37`0VV&(R%_VN5R&+CAyjDhOYD8PgHOn
z?2fk(+?o>yc=OmLT=fJ0u%Wzk8^b_!9L5ffEWXGpBqS<)=q#JYV-j#(b3>t&`b>9W
zkc0<AZI1a%sUtmRWa_v$yBwTS$huGF#Pp`OfWGLy73_Cr+Yp>`m@kdS0Y2iq*mlo$
z@+(m)I<!n@y{iAN^ao`9vU#-%b+1U@XiEjBa`|{;aOv4bukYGhkC(%+D?3Yr+=*zW
zL50?!s_E8q3_0S+d3oy$RO#4eqgio!n=!l`Tyc!#h*t?;Z<(d{7Cj-;B7qgWx3Y_&
zU!m13^K}LBA*qH#hU9yBC#NDCwXFtYk*yY!-xLA)V)j2No{G?kw%M6**?kArb?{RH
ztIWFcC+dwFG&HTNmzZ;u^r$v`Uzlkqp;w$Dc!pdo*Qt1Z_Ct1|$TS+yVv|y&GN{MC
zwS3|S@?fcCLSJ*_;oe~5Dvgq@e`D-0C3IbmD&5)EuE<ULyQ;cT?VtFB_0hQdKR%bU
zDZaiX)}4N4t8u1}%I^<e4*j{0C618x<Et!?k3&ht<L*thEVdKKnnE_YnC~?0p1S6G
zoUC8GO!Y|NR!M76rt!AtowM*jera|eQmtqzer@2SY&(P|WLp!4wp1Ficrix3E0_g8
zqR$tQo!&~3(t)i4D>}+lg;9;9G{))%g}YsISx^|;&CD7c{lWaQ8?|z#hs9$P=Jaoq
z_t8>yVS5aonfmQ*zP`+7g|j!(n-qE}!UX5a^tI=tFJk-i-(^^e!W;0VV}XBo!D2mD
zH>dLf{up-TBg2kfVj3jsA4a%nWQ2~PhdJHF`Lg4L3%-mJVoSur1n~s)tq$uVDJUd5
zB!}dwO$(m31NE{JVbPGA5P4+$ZauOJa&(~MaP#o(<@xXk)b#oOaNKD!Lx_Vb9I+*a
zt`9mc1#**giYM#>c(JO3|DrlVhLdkl>MqTaX3fPbsFVGGS8<-tcx(%yvDEA~Q{cp<
zfLH*~a}l*!zXQX7ejI6Xsh`rAa&YvmI?%DqNZ@SXX_!|HH&4ly+xC@({e|LanIVf^
zT&xCJYF$)IL-(I(T#7=t2RU2AsY#Ay;FREi=RVCjRbdds&*s1~i^CfwnD2+hEsDi{
z+MO!tH5{+%sKUA<5A(DJDF9YA9;U$TX;mcV5^!jHtq~#89FldH%pGpU%b8)T-7Q31
zlRsbilUR#Wi8-^lv9fpOwefz~5P_t|cbp?GbMq<~5K+49awlh5`vdvz4#EP0*0V7-
z<+N~aAs%}}k(f8?w-u+}J1s<V7muh=dFoNV<=N_+CJ;S-O_^GWA2r~)Ulx*=X-$#Z
z;)3USP<jhf`3Ki8!Uq+CS@PZP$<goHI>XB=e*J3hVz_Ny`~4~RG_#Mu$?+f+ZMJvK
z7J;NV=;bK0(r7b?>tT>0AhngqI`jZWk)>znylw2pPL(QozNl37NBYc<$DPs!Z{>(R
zc|HCqBYl)d@Hk4SqV(zZfS%oUlrD3+8SYf<0Wy)_QvEZERSZ5u#Odjq03oE!W9b*p
zYZ;mZ27e!UJ@_hQW0p(zMI>wzNo$tP-lfk^86%Koy3+dh*K_*S&zZ`ss$kfS;ihEa
zs>5`v3~o`_(F4G%si|;g8F>9~r;)Z2<Go3rPK@h(qFjIPX=%%%{1FK^b`zh|vNS2F
zZzu9THooniGX{*O%69BFs29<AJd29lw+9(_M_o>)WmR4HXzSqvr_b)(X1?1CK&nDw
zZ(2mImrV_P;@Z$5;P;$n7&5ClgQjPI=SzF&r0rTEEN{xMOLJ>vs*m0A74DYNox1d_
zxQ~>e^H;xRb%WiR87XFB1o<v-d-~bEPcT_ch^G*yrX!|!B9cpfoVZvnwsC&2v>>W<
z8OrX|z>AuUh0Le*yWwS2DM0zB?l+g;pUoI){D&GwYWy_)dN2MZpghz1d=CgD5@hX?
z&~;)wm+1eBYRt!MGhg#F;6-Sy-HN1hinTICC$Ko`6gw&La(Vn4Flc(70V<PIw^bLk
z6Oa36K4|C_{w(9e$uH}cI00&jrZ3+%*k_Yb2!ejfcrjR($3LHakQ2<<U~{o*=Sums
zggP00b&2jm-ZRPM^`?-X{n<HA9F#o!n-HVrr;hF7uCesLyvLk1-=9z5k4dl-rz&YO
z=!U{*-<1qPbS!Q6#Dpl;p}zDIQt~^pp1yh2w>xR;E+gIEV>gTi-r*g5a&1tdd0Z*b
zU3@@UsCB~4rnfb{4&AtE;Fs@AI7bJo9f1a{)w_g<-3F&8mfs@Ejk<5tQl2Opk5{Jr
zvubuR_0RpC-6NBw6!?H-x_}=k+^%q~_mbiMmd8YtFN#B^BtpLS0Uo%*e$N3xM`#w|
z<u3kE#6An9*?iF#Q3CH|xHUFq+p0DkkR|7eQa*YaTkg8O6Bjj8X#zQET9U|9OUPdl
z6o<0hG9L?BEoD)Dq9rh7Yrsob>y6d{b33*sw28SGvco?j@}fgA47?n9#X}A*O}sOa
z;KIkI3O6R1axpq=H!7H}sF)9_<Z;{TrRBWhVh<GV2>W)i>=~$@e0<EyHC<c3s2<oN
z-1+yb-D4%eBm=S~0Lw6F+v(NEBQty=1DCggIJ}9!QS4{dzqyFznpl4?5Ti&5ogB#0
z_M7s|1Dp!XBRrS9;J?1k;Hy7h4W9Bd7tu#QPq5ic&`FHh`gmW9D@v%d;=M-8P4eGj
z%Nz*5*gmEYBBn1p;{HtSS-d?F@tevc`|99nMBfk_%KnWL4{s#8;|qA%TP+P#eiHNk
z*gf~qZtQ}~PTy;-6SN)$t=T%|mAQs9No(Ekrt`P=g6^%JXy-jB+<(bk8us{xqw;nS
z+w^3XpPxBV6mhCObpLF$HPQ}`D4~W|<PBy01c944JndHC&z%ILfAJ0gtqkT1At=Qx
zsCbMzH{Mqmw(eEqi~Xi^nNSwapOW62>W*bpw64tf{$KD>7CN)t^1%r}ku|FhMw>KX
zK!oT&7<`Ev%mA}!!@P+A!rpj^g5u=hVj{nh%l<@TrVR{fbu3UNgpgBkHm+oWmcQ=B
z=zCR@A|sj{OFg^0yRqpE<$<9AQCVv>lNg&9MC&rC#%l6=0dn0i`}3>2)JunpmaBC#
z3F62EUOVv=0n1LlK$Nk^{)mPq3o&aTp>t@^A$~YK(>tj9?jZ}K+n7a#jmKSCy3H9s
zUx!*|kq)4vydk%MI+q|0fq>*0&u$Rgk#ThUYvYU$LwJQP!zX%wwP__9JCW&6cWIr0
zeQL4&%K4&`s$(Xv=pVLh4(r*gi}riYO+3&V%wc?wQ{Df31|6%C(<c6fA*AILAnK_e
z86XeyZ}ha#(_K`Sh5Fv*M(c>!ksXM9p=gFVCs4S!+#YJ_7;Kb(9N_J_1!aasF9;+w
zvkAH>|Ad30x$zbe#y|-6WsG2R$C&G40KnX)KpN>MUODmB?BAfMV%_V&r3bgz&=Pda
z#OWZ>{Bm3B2l}^Z<tvqe%WOZ1%k?-7`WqQ%ri%7EPX!TjgeVsCIJ$_NzU6+fos#^p
zL%tnVaxc}tjRp_pszBwEE85Pish_aA{fvo#??%Fh5E8C-ZvyaYtCwZBTIsx*xhzku
z+<&@T`drXLNqL^5P*v37T@gwpZ`?bm<;!Fp>);P$S^dE1W#g8%j>~R=FNXbi3fN6}
z&mfa|7bZ=k`wubaqg5AYsgDb#v}o)X0?@APm_z)9S5_j6m)ah`P?&rY{reiNKEyfo
zXz|auM~kh&L!Xu9=RR7MXF%*FB^)aw%LhA*jzP>rO_!^B=VgpIH`k9@U%w(S(Kp`T
zOsI0A3EdU5*(}Hy;Slc0u;VMOo-Zhd$@S8<G*74~wEmI#+#o5sSGlv36?l5qx|vdT
z+A4l?Ydt3ID>K$7XtgfhL4gaS5W^}GcYZ}&d$A5ne1m}g1XekM_1U?UF1({AJ`X8(
z1-44_dv#*2yzy8s9!GNhUGlx&uZh?g1!beQr`adEE}%@~#|vHI{^Eo!v>xGfNE4?0
zf+A5=o892;MJ-?nEc?9m#pNM=uf!D{GX}%0NkeEQ#)%*p+<fH&HWx3odw8LPG-R+S
z#^&lTbgTuN`X%d-#}IZ4{Zt3lsLvzv(qAL&3$PY@+C9=rUK;PCj%DM8tWnt8Pr&*R
z-_Qho2mYOx3h}`3Qn}G~d%nQxKt%We$$rKNa+B?y+)MOljN4qoa(Cm`FxP$t|5)R9
z-O&qwA5V1{_#p`Y*^Av`d~l5qz}+>S_y8YyGUQ*tqmAO8OR_!pEJb9fripK9ruygH
zXNsG}Gh8qIcHi4dc|I~)1Szz@yXCa$m3$40ssA{|_M_|hI4LT~DTsroaR(&s2!;>`
zNk-!{aqg};JmCsXCk)T(o0PHScdM#T6dhF_6mY8F;~>4$&R&g){k}4Sgn79@>+OK>
z)6yp$1>v-)dmYIbG~jYtY&BEPaa~|n@Zj0_=vej(Qr~T|bl-6U*QOJURMso6NEooc
z+&h=vy4Zv-gvxA7snLF$o?1#+ugK>1xZFt6;On6#Pi=My@DlX7I*n)f{a(yvCm#H+
zkevIs)xk3`?=dEQ+G@&W_KmbNm_@R}hDo@`%s%{E^=nXy?APpk+|!+tyV?7;pm~2M
zsrglO>es0jS$VO`c)lWE2%W*hOXvMae(de$M0Ik58m$G6$`TqEO_!A2-6)rNrRVxV
z*V<R>7Bf+bLUx8<uFiB6_iHpPcDklIf;N)So%fn&+r5@t3Zh3Up4z@8k%80t=@)Rd
z=}M{4wLgCGkx=*hdmXvt04KVG&R{+Sxlb<)BiTb9e^ExG8<Vo(Y?@fmy;7=gDE;hx
zX2RTwyqhjl^wu!DR$*Fou>I=}1<*B&2($f|5a`OR#FBrN>L!+au2iZ&1xXhVCjMFB
zf}R`Hm6i@foyjjM22g(7xH#+4rhoJ2*_AriYT-kvwy>~Yr?)@OX2^+6BE<dRbX2K4
zfT+4Ch5;Em?mV&m$12RAf+2CVPk_nrP4AWfg8bmSk;eup6-Bf2wmO<{@t1fI0iIp0
z1O31YL(I<xL#qA&2lqEW`$JT1B?zEqcR3OmK0KE5$*KqGvp9XF?T)slmK*)uopB}y
zWi}(e<UmU<tRjxAm&7g?n~z~9uASPUL1?3Y0pc<(wnQ^DKCB$VcLwa6otQy(y#rqT
z6UQtC=n)1kMa=jJC@;<rdYt72m?sSn-2eUzT{uW@-SYy9J1E^{e3(4a5`c$I&0Zq>
zXe8h5OGN6Y$^kB;8K&)9x!$MPNGgiOR`4Ry5GXZ*9IFquJ290Zq&aSb8P=twJ$F%F
zJ=iND*ZVC7>nQ#D@&^qZ$!9S{u|Oj++Y4Jn3lrAK10vh{r~M~hBsz6dqFjGDK6)`g
z+%3Wkj9)Tzn6(aBE;Xmq@~f|<+K)TFUIX*oS1?2csk0-p|Kia~E~1yyI1TY)mmjE#
z{0w>>IpuqXqg90YnZ5Z2Q;RIl*eS(ppS$erJMOr1cJ<ZkqFaly)C?*gaG{*HzWc4Z
z7~n$Q_F_9Z`RAAL)NjW8wV7sJPb0-Bpo*1rzL_ZW;sW$fPs_zTMqbzaK-k@n=O@}=
zF8KMi!a`IHYM-0%Oop2@O<0DuPuJUJ1Cx2AxOLwOUfa#M`V(1OjoYhW{)0b^SD~GF
zzlmfubt>A}sG{GGa4-g%sif}Rkea+RQ-X_~FU6^rm!}8`-a7TzhOn}f2!KLQva?#f
z2VMM{*%?bsT2HdPrgq|l?dtk`+3h$$jW%LBv>(!BV_8U}z|+h@ud}V~i(At^JDs?p
zR@SMlXir=rH)vugkD{njzFni$BNa`?V_0g|$+ataMM}}VHNvT(qpwzb)hIo{_xV@>
z4joBem)`iAzTgxU<0n>!`!ZcQtSMk||H`z4DYMH&6&E2ZcbRp&5NdCFeI-A<#v(e=
zWUg53m=RmBGcvZb)@{!?oB*07fvnHZAE+CQj2%|#n&DXM@N;4wu{z6%9`wvkx>U=f
z@`*^0TphKj1TS0}$%7qQt}fWi|J`}{IdEUldLKReD*j8R4WtkrAX?P&tG}c(<Z3rX
zrPvFnE3X!Oy&%|TFg6r}Ouz#czlE7KoyU+0Ycbvw+pFwSPq|tKo|K@+#^AQb`m-Sk
zq}x=nX3(VD^(o_b%^zV*{0015jp`SfWG~0a9ZM^ojWdzT?kL6%NU2CR9)Ch<f$E-D
zKzm0ZQs)^sI9fl3O_D`jK_|sCbuAx%*Tp@`Gt;4ju6;caKY&7*-#lzj^huO>76DkI
zg-GWMkQl9&PIjJQRJk+qIZJ5TOtJc9_PMS`zS=#sC=YfN@*~l2aiA$WVd?%VYALzI
z)bO;z?{@GU<!43zB*pKw-cK|o|FY|EzaM6ox1P^SJCQ->m6ut3-KtUvZ|bwmN7F<n
zmV>67T&#j!*xS(Lc3ZMMm>yPCFH6W{$iv*Lwm)9)*Q=^)iKdgEI^11u71qEGMSF*K
zKZ$srez+*|iM%=D>{n%Lj(&A}oK5>@;#=cCnU+u)4%wwS$1cn9de-1(L~4KZEdZR$
zhsfqG{Qs=c(OF0~Mx;((<4v_R8oFI9H%GNF5A-y!Z0w9XAC~*QFHnxd8*K6(7R-fJ
z{QF3UJ+;q*?Yg^IjFfyr_q<px)-IpCxUIEf2j_=ic+5+wPwT1EiCmnyhX;-&^R<KW
zy5Wm-PWb~?#=|Lm$De8#iQUNut~aTCjAHT8?B>mW2p4nK)rV)TE|pYpX~r>Wp&s6y
z#+lpom43X{7dr^==WiAqD~R&l0|ktR(eK>`+3H=JE>S*`JrMe{8Yrkan$a=M^t<Aq
zJ@dDxU;y6Rt;}}8Q?Yh=5|5G4VanOP$Nk;p7$xoT%QWSFZ<lz#tx`xEAY6Zr2yjHi
z)?XP11;Kc*VD4_93pr@fPuu19*ki{;sIo(PuF6!_1!P*;c+Nm_X%utKMNLgjArYeS
zzhz|mi^=t#UBg)O0(+jmF$MKVh*f?-Iz`hbXR~XY(8C6UEh^~MyRfwk<@OvG>kMDZ
z)|p062a{#zFJe-^!k(!A*Vz*c`oy#`G4!}*qym(Mv60Yina+{zaH?>{bDm^I(KRk1
zJyas`K*NF+y6xD=C#@Lri9~|jMYeN)KXrEVFdM5TnEe6-R|@RNwr4^Ln&%=xuAmvA
znp;pZ`{*lI0|}etA8Y7$uRdm~ByiHp*fQjf3Re2VvZzi2w;#A`xr6wtO^8Z)8ps?g
z$m4G7-*^)&)L4=%yc`ORi%mIde#cNo8FpgJq{eIM?YtFr`HNIX-z4z9gSdUcwz-iz
zR7_^Y8MyL`fG22X`E{0}{#?_urGVeEGa&;l|9H~DE<!?q-TP1VYCdG!w%@us1~T{s
z6%Eyea8ljoq2jALmZest26pZ&gmd$5?rb)&YnQ%`0BopgmU}k}pFwyKd?Fh2Ki&=*
zxD5PFuK&sC3>9ln<ebckXKd=qd*v>Y^01MkmC`r0Y#5Ie6j}wX$YNyiZc~mTxH;zO
z5=8?!lSiHTyj(YTa!NuQ6R#;L_+HAa(zxmh0#~sokH>)(vIhA9F@yBENEl<XxwXLZ
z7IVM5q)%0g8fEl*LC!H2{i26m`fmBUc$&{!&+GID@iCpAQbya^Xse2_)LrsiEj}V-
ze}TN@NAB5evc`L40MgoH#YeDz&9?h6hlDwCe)(})!{*8dIwQfcmPTK^&H5r{qch@q
z1K^2izu{e>lx#ttJLvTIE_(HQ2Mi%+7r=frIBDS4zx%k{*6YxqD%!dC@uL@L1d_4T
zj&hDx?6*2XMd%5HlJCA_=2Wx`5U){6bAk^-=0`v&+F5Mm$CHtrRkEc>6BRyEOTjRv
zrqDdP2Df%=re{zr<Ot_Cr+C&X6Gds>92!!^$0dz}t_TGJ^}wyURd~hQNx?<(=IF)3
zSC8GY`U3qEoS0+}$aWYNNxFZG4H-du_UIK1cynp&N_Ga0KlR_uC*XED%rn3yTx@06
z;>iU#mrW>&md*xUA?|r}A$^1#H7rmVBX0xIphCGM1vS(!Ix7C}a{d(zd^O@XfW5;i
z<%$o*Vh$incDKL>{RK|MI=W<`@AADhI_vuQQ*7b2-$|`|DFBC|5iNd<x$Ak;xRr6T
zjDM|<qTv2Sag7k^)x1|7A?*2fBoK<4P`7_%fnE1jTo!75Np$)!`NaPjb=Fc~-v!py
zU%#6*T5zPkQL8>xh9g#itW$P0lmVy~JHT94%#$JV5G18_IE~sCAK9f<7GH5kRUJ5p
zl-tG*vD+FP3${+)Pcs$AvZQ|N+`-yFO#q1#4LlmL$>s0i9z%ddUd*JLp`?Vhf;yZd
z{%BTpRN_M4>!b&y#`PZ^NX9q^icm1eErxA!&G@7ZJ5jkwwDvxc{iHc&6<<rX8Ts#>
zuROOK%knoso$k=J+G+v-jrW-+&&{Trwws&5ELz`s=K=ns*&hVhNdm88W2Z|6PysEw
zQiWS5qqFW;+XCGum>tUX3K>7Q_jmn6<J1e<uYaEEd*QFX;yYMx+J&AE44fQlZUNr%
zja1k=Q#G<m@e&vDQF8C-owD`e;QCRhTsK7uVaJN!-v}N5jk!Dg#L{lq`my*`1j-MI
z%blsjtDO?QsVx6xI@a3mASyg*IRE>^j~#Mwj1c}<sxnHv^$7F=6;7=ggRR3?hXtwF
z{6bu2skY=?lWPzEhmYXZ$Yh~(-z`%=PSnjI;~bYde0DrNycU>8K(dvfQ&%MEQ#W34
zXIunjUS70Ego(stZmA@{R-m)EwWv2{{GtmY;bs0yuI#VvcQ{)szyv*i3o>s_>1eA<
zy^l(s31xlQCAL;b=FQa|JoW-Pxg7+&m}nVs2yZR39Y?jYziftyGw(OsUCnF*7dhJR
zt(*w1;~kz2>X1@!7Ax+z-`~}k79`j-1pX<1^Ub34y0;QJACD4@06U0<gN&cY4{b+(
zZi0OD4Wxj@27L$9aRmB|ABJ))F+u4omK7LJwu(Irr%tJ2(PhS92lO7-efiWf<Gkf}
zyK6Kz_z(6HEp;d2#Yy6C?=T2+k3(@n{!PvE7a$S)JP;!-kam|R9uT5<72Op=R%6}n
z&J)zU%e>vqo$0J^H(yiK`ezj=W#uuq0cbe!3sB@mdfZoN8BQSn`Wb=Y>!Vb=wl}cZ
zT|u*4d}Ad7?4eC)Qst?k*M6<;-X^&F+;B@ofRp2wA<86xe_v?67Tx(Tit~M%gG}%d
zSp9y+^|aH(?COVMP#Z^Yv63f4meinzxDD^CacpuH#}t$tS@q*1ApW!oqVnLKNZeGa
z;W$a&0JwXSa%#B;ZH5QrJR*5*-Mr{|wo?SWYkCHb{fw+<(zO3Vp6r!>v|T<g>s^RY
z>&VJ)lGbxwrWiIw13yCr-iOg1SNLkUZ_giFL{#LFA)=OY+KBk6zzmr+4;s(khP6+M
zjkse}v#%o}A~ecjB9RUwa>eLOO37KC7YFH{{#cf?mDv(H|4`wV6m(^`Rjz{=N9DQb
zr7oh7UF@x4tcU0ssJJd|Zg3&cb#G05#Cy0p3tG>mw)5gpik|P7{@KULGJ%)l<;jrr
z|9f+yD4__A3d3L^cbi*~@aWO_dGiFyQ<kLEzOkFl684-bvw~V2?3NL*;J47>Zbee;
z&ry1U*I#~|{S4J%hAM7GhQ4UX7zT~oQAC#L#7TMy+Ha-vNmtn{pn!c|f$GR|^ZSp9
zT)@ZRbUvpSX0K?}sV1wHTWcV|6E3;6*idXrpRc-VcEZ1=VY?`=p;&=_q-@_4$w%+T
zqLyrhMv52;<Nu^zc)|)uPf|%+8o@Zh3+Gl9oN5G89HX?YqL1vQ{Zn|d!`k#t4mpH~
z-qBR|KKk<I-gEb#2Y{~(260RMMfhF(Okj-L?NF+hFxYryzRGBbE%@%G@^L2>O)<9W
zE8!*07qC2btW9sQrByrGno+;r>$(U&s2NV(z3w7q=jc>#(#YGIZ7<kn6}B%2k)W9X
zH<JhZ6NS;9w`Vgp7YhJK#E2(HVlDIrVPry)psTowq7|DL=>?&l|NhZsVXoKgT81b0
z&%QOuT9}sOI%|Sa^!NDa&sy)q3&Hjq@gZ39)~e?u?2w+50;Pknx9Uw|eEe+4dD%F}
zOBQ{UM9w$WFXcXeF%JK{xs~asl7|8v7U1B<%b~q6qko7VH=fYUmX`f+CJ;qp$%D~n
zZz}LYeES$+OCaUNZD`d6hhn1zg$^zx{E`S`{g-mH{m)KpI?PZ0<a@0q40_kGJT`JV
z`rB`Yrrjp-*xYv4fsEmnIo}I1aeT^=8qB)tk60hRMx5>;h*l7)B2ngcvtXy0k9PRZ
z_5?(ni$^7|hcYgP)1M}}ZcA8v+s%oNM>c<`5;%<D-+~YF>x$EWNA!)g`0jXqgwxxf
zv?rBC8im@JCI2xWToyIoCUY}0swHd1u{9U3KRIhpXvN3Oy$8^`3E45L{R?8M#7g<?
z9x6M{e|7@zZyj$T2gZJkTR9s=S&`Z6s=6f9(h+h)_w3G1E^AysI&xcd2o>8(bE>iO
z8^XxDGsI5X`Hc5-+i=%*t~xl9B*2B9>cMMy7o<!LjoTt`zBB$;93kJsO+|+9X|R>5
zfUwEX_{79wAVi||cYgoBn^U=F^bsi|J8+$?RVp(arWzPGY@}u}KC$p&?08_tkby4d
zR<4rTW@bvQJGAV@?KJwK-qtg2ac_=qS5;g&c~a{coYwnwF&dMBm(3I!lXmd6-F%<b
zYLXGhadKU7+moZ2^s%cmoDU=xL>@}9y|~P_l6Ua1zjN>TD5}f{QQi(}D=)3D-IlZ2
zysa%(GBH)0DCh?Kvic9bauq6lr@1oG<wcajiJ8}5AtAdr<CjTBiJyhPa%Ll&-=}LC
z45#ZUq8{dyKIB*Ew`%lNVVmj)2ikfL`uUR1{i80D|32GOhK=9bw706S^>o7z#cErA
zhCDw;mK^p%82igUI37{Q?d{k5g`-y8{|$NbkIi&i`Y_&g4Hu!lM%l&1Eqx*rNv9*n
z9ONp?u@SLZFc^kW|DUlLyo77k!^B%0*g$J3(IfO8JQVmq_`|=@cLSblaesF^$6<m>
z$!CV?rHma9Q5atan-o>Q<onp^;zEoP=+G91nqkoB$T&zzR=dQ5>i18taxB8PUKc>~
zd=?wqA)wrSZ2^vzKO<s2E<7zhM;#cfmoLp%T=;&C!;x*CXCI$pC-cD?j<AzPsCs?4
z&KMg}Q5@@;`%SlRkrI%pr8<hv)ISw4HC+>fi!b^0Q|I3?j+OvTy1tyB<3~PH%L9oS
zI}osqe+#Vg2G!}$zVZHHdjtQi$Ws`%-n3i(BzN9xXMnvTZ^`kRrE)Zw$a0zd{m8!n
zZ{jNgmlt{r2o=!W{)QW7G9S0=&@n!O)l-`%B=_Y{;Dmj(iTZtNNfQPO?W$L4XyqHM
z)u%iI!XGE)G^u+Y_QRjIZOdvte@<R`fW|My4CKP>=sl;s;jhNrXgFRJpL?KnqW9F8
z8lenpDYdq-^SBx0LME9Wn_pP5ctZb-h&GCH5r@p>g&c}5Zd$F!mE3ad*#2{2(6tMU
zRhtWz{Rm3jSkLAZ(bM#`>;=PPktX}AS3jp)=ZA`qx;0d%;KPhU1<qef6_a*Q@#CEE
zyRezJ-j!==;VpU;F0M?M=tyV$z`eHB@)0Wj*sQ3?nd+f~p%iQjO;SCrV<LK|wdCOT
zBZrUk#UFo+kLVS}SBNjJ?-S6E+Wf98zj$uv+w2?<*2mWO3=7%2jb#xZzgs6*b`(Y*
zDFl{V`cB02E3UOR^*<u%Jjq)b3o@lMmDd>XsO*dvq0?FC<ez()e2j!F)mc3XCj?60
z@ehCbAw6ORMV5(O9uM041aVaqJ92U$lCsj*=gP^dW_M;bpPlLG>KfZs)5#{kg>@Z&
zmL($*((nVvaWm3FKZ);pJ<iytDqiwK!<{@{iekzySY%JtS#|?ZhFN%21Hw6r-3>aM
zPvo&hR!1hha1q3j(e|v8u6w#ac)A)toNkhy-_$MG=F+@~nlpoxx!=}=wf3L^??WPv
zc53+Z)qfgog*iO{XoXnwf`QMXs|$TIuEQEo*b33u%3e#L@EFAoN%rA*I97v6)1l)b
zUKc++Cg)@@M68i?tV~ZB%$&GKT>4Tkpk*|<d3F>3n&ngCxN;4y-Zq@I^T(mozlIgO
z!1FEN`fa}R+YNn>_{b}+o??b;dW2i!d(0iIYf1Z6)O88-fp|xoz{g4O^h_`a4B&j_
zR{c&54fR)LKQ8w0r3f~+Buuts^f;e;KL!4^eSM@pJ8Z+(<APgVPNJZKrcT~C1Z}}y
zi|q6=EeY3J0U~6>!s5?V&S@!54i#{P_a#PT$sJ07rMiTTtD*JBu33ig8pF6#)1CrN
z<##GiL8xgG=RrW!Gu>;qz1h#Lj~HH4!efLC)}&)RUqzWtH>nAdOVJ4TzA8V;SjXIK
znXTDR0B{R7>#z81$rlL5G?IyE`?tI%{|4lQe1ApmDJo-`Pm40BFi%r?^QqSA!Qc{=
zbfiJ;=f9L9KT##5O}58}DAvYe)7`jU+bj&M2({hKt6R-jluQ1pwQ}qu!{Y$<yE>UK
zueYiPvHMxdFiy=B^^_zc(PPJ@@9!8t&I{a2f=iAS@v^j(HiK57_3-<j&s#I*YfI*@
z`lqk8wRS)u(?Z|%FD9z^g(}l3v74iJE<?QCsLweI-0F1&tsB1dX$nmUe}8uz*{UUR
z+$Yre_4n3X?17l`R~grtw*bZA>$mw3BWFW?f>-DMOebQ#@)Z~pwU2q;8pfX+Uo0i0
z2$skh`1$Y@+<xI7)GSmv_SV!(Zu}q+6G-`CQcen^hjK_x1P?QYTPSY23}^D@vbJy6
zy!v*e?OOW@cWV)aupOi1-K@GSjaI4e-)C*35bEK#=$o`sT@xaYqO@f_i^&r+!3VWu
z&&4R7Grr*D$Q1kkEPz=LK<Q-8!zJ4@mvbn@{%C&(=#mfgpR@tH^}%{}w5rstGr0G=
zRc-r%-TKRiWNMfYiK^r!gWe-3kw-zaG{c%fhYt5OahHB9vKk1-B|w(SWGQdQL&c0q
zjsmNcCFyMgWitYkmm}mIw{eT8fg571UnIPQQb{9^P2SYZM!rM;mHrn5b_V@SU|)G7
zE|l>L%CyVg=6`p}c!e(>;I#mR<9GoRl%1Ih)rg#1^;i+eu%Yq-dv;gkb%Sumb&J#1
z$Uo<QrG+B$@K_E;fW(#rMRI5%_S_`aNDel}w?K9Z`>{z^x2M#SYaE@Vi#n&EMz;ue
z$}$yBen$C}(KtwxEV-ZIwQB4dVe(AJ9mc9dxa|~|)?|>CAXI^xW=zuQv6bFyfK458
zUHh&ZTgZ1zls0vi`6$Jzyvm@-=?7RpXu-xty8E{(pI<dG`X__oioAInk~cm0z|cw=
z(_<T11twgNcfq5{yj{yN*`{h6*j+cU)degPrrF3Q{WLlgiqL<miWdnhxzX{^<85a7
z0?Si7kYv$L=H3`z>?x5QV*Z&OLSoI+?Je+$q3S$Z!*|>=zm_X`C+;X<t&}^q+)Z*F
z<J`x3{Lg=0pCa(A+WGF@a5t=?*zv76717E&e!|(b;t6No_wTKDU<&w0m011&*}es;
zeCo-`s6y*e04`gc#^yb#lxwMS>bM=<fFCD&?fsqoHD|p*959IG=n1#J;bNIcS)_N$
zXZyIC;ib@b&|Ob`@R7T1b`Nv4*7W&_7W`txC?|wq3oCfh2EE}nCt~Qj%Gm!A(u%@;
zNi|Azqz1h}iIveitom8w$!0>)+_%B7$MuoYCFW*umfT@gvboP|Q~ycc7pK~BF@*Hd
zPi1w`Itl(NA?c33R8wNnci&1^pYJnJ;%MraQ88#nv2UUANWej_Hmc5xoY#@oef;n%
zljbNYxc8Kz?*PZpp^;BME!?T@r)8=8m?Zg1#O6p@{`ulbw69?Q$8<m6?51DX_a2MC
zy2(Bhkd*-0*lQ^#SYP&624kX|NmYVz;5sFPuH0lPSh936p*h(AhS;2y3Ca=&V%eEk
z7q_QU(&y?cNQBH<V4!B<uQnLAdKS6tPJ4XAa`7EtEe!hudoK4D*u?Con%(S#^biZS
zt!LWjDZ+$w5?HzgrV}TdxlcoRDKDr!EA;Bi@vyDtYgiv|M5_a#Uo7>3!v#cMFT0BU
z4?L1FW?$$oY3Oq{WglKGE`<9zjlqTB^F`n1Cyn%dE)x%}8?cn7pk=)Z5p-94Bu*eV
zM$YNxPq&e(BeA_9>OeN~l?vQ`5j=GEezY>9jOAS@GQ++T>-49uAC&-p;tU*LL^|Db
zr|4G*k=>&N7p~S^_lj5)I)%fhLSKSN5yz@e_iN(KK$(P$=pZ-7>MPpJ^^-te6yyy<
z9*CKknQa#U=f*obeM_+L>2$);*_HWa^{)=knh9_X`KTT-5i0#8u2fDr3GO%g4<}|v
zLscUGXte(oxj`OiTO~hGPrzKe!W6jq@;)Pc$#?y8L}^MDRmM+To{#{SOcL>HdEmgo
z*>`}Y4u2%!LtVoW5-<_pLR`+4hC0xsMps%^f#t?};?_FSimGTtHxQsGtM<$-%b133
zkOIR|{??n4{>e{ZLs;q3!ed0v!54}<7v9yS@q^>@w5!`dwc&D2&ndk%B~R(Wt%RYR
z5LQ^vR{GBd7kC7|NY6b)HC_rUh1q_4D1AE8gLZJ+qyIl7JwIu=+$|>5B42g1489Q9
zR0U&3IV<1Y;z8UuhUl6;TbpV2pDXr5Rt`Fl&|i}q_EmVKC0{J#7{9rSqEYc^JiafK
z+P0#McTvoYXv7vuPvkV$pjiB9p^h*5vK(i}1h_F;s4mrjcw%1D6W<}Zm7ynht|Vdq
zSW3X;t%<-<GkQ<dg3O}?1@9?;JW>KlxvvsWYW^6Y*NidqQpzgRzIzlmFo8VXKBh=|
zS;ix*SM!z3N2>fwr}q6I_pvwtDaUzm*V%lVj1snG9Qr{<6-k|^jHY>t^bh{HIYeK3
zebBMX3$9_j6d3Sp^rFykQ-K&9O6Y?D@OtPJ(%hwF3O}~C2ltGWn#b8;EiAidqe+C~
zr)}3C39XuZqoNkI%TU7HaLz7n(h8NSd>(z7{aaG;%sxGju>dtJnD_A~9CJ-F+$Um$
zh2JG{5~V22k*K>C7(EzrJha_J4IdKb(X#Bt)`2=^{;Zq@hohG#8H1=l8PD|>1H!E<
zl%OKZ4!T-3^w=Ew!okn7qc6=*j`iU7Uy-9_A_c|;<=)^WL#)XSaM)@XOGL0Y==4R9
zz*jeFIzwqd^p}<VC<6r%;@i*8^PTCjzNGQ$Ex{XNNnme?r_X_Ss02TnQAKI+<syI|
zUM1Lk>0}em(+P%2wZ{~<B9SW;BzMP2?V1gVHq2nFJ1s@S!ua7S1l|8#i=O+fwxWi=
z2=NeI1%eYao|8bSgu<%pKLLCgwWu)?Q8}0Y9V4!giZE~si|RR&Lbp&QJ^n8fT^c&F
zNjGxeD)ReT1A@cIbDNQAYb7vGG&7lQJs4az$1&Ed<}{Oc&kt8JTeV41{<Jx*st*0h
z@VmS0W0TJO=N6Pz$@~<g-k!K?y$t2i0P|t#z$5on7@6^)%rvT}u)Hxf>@_}8@DrL7
zIMr(n=1hHmO><BC0@K}i{JLK_T!}uRMtcG5&-ayS^+$|<Z&HXgHmFkwC84-)kqsqf
zQLW!HOT7$xYQjjXXyq4KHGWM}GcXozD-n0#K=(nSs|Wr%J+8sZV8|`+x+6h{w2Q~r
zVZ|1}=_-Us1aU5GFrB7YoO-(OpGGdlgkR964*s&$?vN3_;19IZg91$Bn2;?X;P~$+
z;A56i`|};Djahq>>#N;u!Ir`N_lF5m_iya_#v9#-mbJ~_4})>oMY#a|7{L<AA8V@y
zDEo(o=*ZNXo)f9|&`0w!yGCv4(BIR;iWRetLV;`<cw>?O#R_-+aGc&NnU9596y$~;
z`C(z?>wh}JQxEU|A%16`c~GNf<;&pZ7JB>PSJ(PPkSG0c`|otp_OKOaa<)p)#<fqo
zOr4w`1jSy)&?U7Fa{XC}NMIkdmY$<=v(B{>cYJfe&hs8vwTL_}%H>y-yQY4+L3SVY
zmM&9Gfrn~BdZ`KRw7jp34>EisJWfgjw>#NH`~Jz0-SiPYPZE`IdEa@<lc^ajqxL&D
zrscv+m3$jv$WT_Xv)zmi%^1D(mBTs^K(y~c%BY*eS*8ULTH=(+XO_5;&(i*&$z*`!
zO*kjN?dr(W!_r3{>qJ6S5DK^n1UMGF6dJ!r=@_d12g_4~-iCK^g?jmb8ReS?d<=c&
zAXcvQv+-zfL>~*v(Gi~TA|a!tb|DywBYtmQk1KLVp}9l4C_S%KAZxIhWOVBf;-wS;
zSwG8ceFs?{h|M@@{puCI`=HtXVeBoVvW(VsVR&f?DFJCgK&1ugR5}HuJEWvL1YWvJ
zy1S9??(UZE4w3FNU1#mR*IxU4V}FA`FdT&Qna@4%dEHkW5c%w6#rDn;$4|ic{eGX{
z)@=>!THMYyDKx<OrBQ}n7J9cFO2R#Z7;1x{;k*^2VWs42e>gX8*s=c39&mbk^?t!w
zVQ)~-gc7mi-61lxaHB!J!NBjYETxXXkK0y;=~xJ>41N7!)!=-0eDiYK{8)aKi4i&^
zj*PeS#>`a4<?N{u5hguu8w<v@HxjU?=IH9=NgFuRqN&ZU>1!7=S2>|kMrm$JAYe1F
zG5Sh3eg#C!MVZswZdtD3(|YNQMB<(~_^khWy<EN!2%oN85)5uXJi(j>#Ht|5WY-+d
z+191I5ftWwXuXVYWo^BY&S%@@{mf}Gn?Uxg$EeG)3gTh=v``7JfwgEp@MEeRx7D@S
zD5<Nu1Sh6jF(<}=KpIrq=!bzGal}9v4#Q9TH8=A8I<G4ba9tX%@bg_HUQ%$KJ>GWR
zTz-!^-YQ!Sf*js<B0vWDh+eEjh=dTdgP-mMdhGHG1c4i}r)$wMGbx@d*uNjs`Enor
zD5bN*gGpGSXLex`Mo2u@Y@@dOTW^!XhU1Q!mFGIA8{r9)_9<Kl=gm@x>M;2bYBs=$
z1@2ye!(ra{<8GEREMFx1<?SP<7$<euxB!E(&UE2>`}IKefHxnt)ejvJuYWq2*E7!S
zL})A9A2u4BG&<xy?=R&}RsWg5_2EVuYCeyPY3n9k@p4u^FO9*{MzvOkpNx_;uuopi
z6e~e1&~!349%=y_+4yB%5J{2pkDuCRq_npPuIa_YRV+k@z?p!VFkgZYX3Q;|cawPq
z#;KDZ9UEF}xw0>WttNQ@OY`02@hvFBkdCKmiV3nCKJI!QILkCXbw!}F?7KV{<qt0x
zq>+q;KPj@m*)92Me}WcQoD4OA_Y2B@58}uXI)h|5yn}Jbj>R}@Uch`@8lsZ9Pk+MZ
zCN_d%BpO9H=@i6h5O{}L15+3vrgxhB+r^Q>;OdNqP{3ea0nium#Cq=mZ4yX;b2KIW
zt;7`p8bn#X3)n{iUx~k>_LmDVd0Ine_68Xm=@9b!yD$7;7g0LXHe{w;Ne+cYyuQIm
znF78?X#|Yhdh6pw=`&pp*dj;y)!n1M#I}0NEwmoFImeWN4d;mXg3ow!GFiwPW*aZJ
z;h|V{Gb0F*%W}|y4JIpCjn|&9!31QnuMif83#kvG%xB>o5>bHxzom>tGRXkdlJm5g
z6QY3CWr`(mc;Cvs;X!f+Od03PuvT18c-eTw$W>}i(OlRG>n8WNzaKkUhn2%Z+z6IY
zWMI-AT^&}}H@^{VD-xqJ60nTZ06S`OtI2kbAWnKV)2~Tl5Abb)^$hW3PCL{xNvomy
zKx;c<lK(^oE~6=Km%q#in-sWakFjle*qVf<IpaI1@RXNa(9@eiHbmvoKSuVkl9!pX
ztALH4om~xrLZqX+7Ehwgb2MXT9_Loy!RC2?!CThR0{0{HDX9>+`Da2+0rm_tjmbJ1
zfq#niJ5f^UDIW<}Vd`~$cDo+EVDj#DrYT2UZ4t~MHnRz7SVK5dUotc~r~2o~(qRnF
z;2@%rW|I}y(vX&e*Ot($*r;ziT;x0Z(|hcgTZt{1Gqd4j!krh`)neuimq%YGi7%bq
z&&SF9gv>>@^<rk*6lBdQpVL(kGBipdQuh}Oy~y%hoNT+j_`08;RTIIhs(rQZ<Lr8~
zMKPV}yZHRTwy(7a6>oQk2#M6#F2=)=*HH}w61_k02gg}K>r*qAb0s~+Xz{|h<IYFz
zOSRTvJJ;ivM@uT@nNolgiZ?H$z23OiR<B_2Q33dfWVl={QOqOb{ak#jAfcEiGmUmu
zm-XdLJBPJwC$-59Pl74cvV;Gb7R<?UBkdmv<?Ph&6#-T_6(yZ<xI74K10%_**@Udf
zN{@Y!IfaTvdLO>coNP!fHB3ia=p(Z6?EOY;|MspoWTtWOhRP)e1O*iu4kwzW0-rX7
zLF^8ZM{V-xzmr>>#D;K&L^t#jRVvszDZQId%^d8P@>((+bFgH<3Vx@mga40IEzs<0
z*pYdF-3bqoZOcD58geIqHQoYI5t$<=NZaqoeD3}dQVIicXezLP9-QC43rn<YHdz4X
zYbvnkyW|8T+EO!bBLe+dK%UKQ`y5pYaWH9eOy>EU^<DnZ2vK3pCue{zVOzn4ByhE9
zj@hjD05{B0Ge0>Pq$Yk8rkAxt9D88M?kc+uV^V)>O{YK{WuI&l928^>`$x9K$yLB7
z>DZst<;~P;ox_pZt(9#w<hLIV)Oo$ZaQlz6C;^*>7&%DM?s)?>FF$>i4}%K_?cE6H
zAE6dhwN40cC)h@y?`Y;l8rXnhnsxgKz)-%%e94x*`k+;192t@LyeF58so&OXM+D+$
zdnz5T1qV1}Un!wK6Am`!`bq?t96^WX`D~+E#?F>cv!QI3$+2nkhvb|AjH<W*51Sw<
z!Z9FJK0Bk88S`-5pAt`|Y@A$m*VkIB6(3EZj_RE|>_K5;-Gp=cn+G7hcaU<amj%am
z=Bvwzpl%8-tbtfMIZ@KS9!%_Tm^6W#%m@~1?P`R%3~%7{X(z0QMx6Z0M<U@@%}waK
zITisDiT9gk%ipT(SkHozmwS$f#`B@+MvwSx#3RM_t%oJ4cygz!_w0gq6yvSv$fS>L
z%3ZIT&F)yn<*jeyF~;B&$0`F$fxa1e&;2i20%Z^jTKLyO=Y;B%_!i*lja>)ZABEAA
z!JXmRVVy<0ExY~x7eG;7EK+8SW#*yE#DppRG?B~Ilo$U=JZad6?gg8QPNjF1Lw?~I
zwzxL+;$MV29ub4Lipv>eB8lj@LGt%0X>mYP$aH?BnMOxmv1QA+x9K1Gw<gB8oP=Y~
zSaK$M!d=HtG;e-?vR&EOuX~q`q;Tf_d|<`vYQBgZ`;)ffSc-|hyUFTK8a+`{qhaXN
zw!bLZ&w?;czt2q~10Qq<oCXzjF)F0so*qhb%21dJS<Tf1*b+VxeYxy>zJR5b5impB
zzT#2la*%DU?@T(H(^==XlR|Y)&qjQ}&}MAAja$Hdoc|K_JGY5^0e}xzk9OvDI)DsB
z5w7u&VwN~*O+w;+2h#>a@g#^S%%<n%^z@&tg-GSxffH>))_f77-rzoUr?N-qBeh^A
ze|6-zI$yE(bR94HR;qp{7i%Oe?+7vQqNXxK9jVxAK&mG#(+3GCo#h4WB5c`CZjyyO
zQ}iv0N~;~D4WEVy5l(~MYE@t>4rziCLnBWDP2SjpI2W%ZOk?gCJ}?ei0HdpA)sA3_
zT9^Gn9<(V~`GE#7Z=mR`0lFwdADAS(nFDd^2LH}s>LuUm!rV$&2()vcz*UpY!Sy^=
zj&MnOM!reQDJ%O7+Pm1V9iJN4w(rutF`jjKAy{eoW02ssSthqB-hw!0g29OXEg~wk
z;$+EKI$aN7<}MsfH<{=M_uJ-y5UDOyff*wnm9knL25)VDCv3wJZ43o%JsCIZgpwd?
zyas?Q)^Y=#s}O>8KY+iegfu=q-Yvk){1H-hhiQAWEXTxb3sIg22t*c1{E1B*tq*M9
zvn8D<I2st@HIglXR~p~n76Y`NjYGWQ*e8qUywVz8K)39dqyXpnjm~)!=#am&Q?A}X
z@=@AYPHc~-u%L7yZX!R%q3|t!kFf9_XT$7pFV3|yhhb?skE=J;mQ&;+ZW6iOll7rl
zZDAuQ+dOj60S&EsJU@35oo_NswTRwqP~&)iT~ZWGsvZlo#Li|!rgys!+XH<<Dj>Gz
zmmbq7%;y(L00K}qWYHc;{2&P&NrFPtetHtb<qTuQXX$(*-;Pk}+=D;i-rlN9`#WXG
z(p&QSkRk)Xe6pEl-ig)vc0yFC-||n#BSZ(-RkHiuFfu*FYBa!`zuGOWo<ql@+sPJ8
zr*A8a*WF+mX`n|cX;u(&?FjTD8MX7M2_6;5C8tT$y+1}d!v>&w@X3JzYuIi~cM{?$
zb6gh(iNTpJbuX-?=0+;J`TW#lYTK4K&@B?(69kOsWs=1}w>6fjrGD8O?9vBxEoJ+?
zFWXc5^X2q!gwx+Rk>xj=NDfHq4?us0(r3+6-h^8@%?~{K`_pYXT!-Naf!>h-{n}&|
zh;`3ET5?+nSu$~3Y18+*<-(X{p^e_4xaVAbLPS2grA@{*e<I~hZABl)MIYIULzL!l
zV`B&j_*<gcN|f*UPD<VgNLsx#1HM*iA7G#><-Hbiz$dpARpt%}gYLC!lY7Z}*H4M3
zhWZ{1?)*0L_`+<K$(bQzoy$XcV0-s<`Za0Wl7m9cpk4evFyA4BHTV{R4#)?U7NP=0
zM*+myx#;hTR9Wtwc6YEc#G_W7^(<yHI`KXLda<s>CYy6<GH~|*Lxt)94JJ}1j8&ns
zmqak7%AMOyg~mE@n&&4)ZE(tpQ#cJNFu%QOjfdhB4LsgZFh;vP_%V=+c->yWuK(O>
zY+uyu%9qPwjWtu!{+^L0>;7BNWp+=MU(#2O)IybW<5ftBmHsotQT;GZkq@R*`N^<b
zaoX<XFK0n&Uw*2~{@chGj^VtQI_Np>X69C-s1XcM0xeH02VlR$rcQts%2b$PaosK!
z(sT+Jfn1!6z407oNKmkFnM;V%Uk+upG=aq^_qG3>GnBgSvVrmwqYfJ4Cu$ldjE*w+
zEY6U@>oo!$T7q!?r+eGX{8Yfi+CXQhCqaag69l(+z5KG8>U@8FN@onyV>v~HsK3ZK
zg+h+H7zfA(6Nsdg^e}=Jum{QnBD#+AW_*ZuF9gEJ5P(xO%sG-31Db1{hf_FAGF%dK
z8;ukD+8+(1ASf*M>NCo<4=^xjFv;>Op2V={A*k6GGsVEo=CU!vwih)58W%W_6u!cl
zN_8IUL^BHzTM^L}z@=m%Lfbs2g4yWMK`(>2@6xm_zd_OHkO84x1{iS@iVhshL?Pd>
z=Po**^H+KD!hFiK^mhLpmU+^}{j9dnNfh?a7Ia1Cj_EO-U+g1OOS<JO$}LV{w`Sak
zD>_t{haa1W0=hKE0k`^onaYURbg{1d7Tn*CjF27{fhSk#Qk_>Q`zJrNQ#{KNedATo
zm(=v8frLI9D*UoR5<n5T{9Xz&S}qr#t(u(M!&Jdqg*w9vNWi4J_L><G*pI}?ClJ6o
z^{iRL+P;!$3qaRNh5xuSvH6MMn6hn9v%wDQrE)uNKpLf37TmTc$k~G3UeFSpmqP8K
z69*o?rN3=aVbUhHxg49!xLN>%{pcpGDiD4HXB!3PTjMMwo>31SO2$?|p86VyGb(u<
zHkNNuW<C(0MwTkW{D#kB*d~M-0m)R>FyCOxy0{Oje^nYSY&lPI?---<_+6uHv%&1?
z4@DBll|RCAjJ{05#J43Hx*9_)!_v3#E|<VA=5BwihbyM#d=u|VY}5DZT$k@+@cvtA
zv*9Qe3~n!P=Y!35vdY~jMwPq2QDN&Mxn=}N6V~cwsxR6cVal7mF4(t2bY{A5F;3De
zXS#Wa+!!ZZW?z7jPz^iL!Hn&4Md3Q<cHMLS`?8w7U8CO&y)Vq5!eoK&=t(LZ?Odc+
zIM&IW4BKZ;7-f7NJq|-rrz2vJRfbzxaMAtyA@{**LVVgpMj`zxRsCKr_HAz34X5)I
zTMl;$Sp*1uAH?M}c#o|BN5r;^Zq0r88MD&Is$ack3a&PeQMOc<NyA+F&}_P`^Lf;k
z24<PW)n3#EwNAz*fPF!e0<;arA&Z$_(6=t<`4(_@URxUWdb-~=YGp_%pL3)W{)&Ci
zEuutnloWewqBGC>@O?MK-xO0^0i8QZl5t65b;{HLU)thOz2S>?%gT3*7qkpD0blZc
z%04Yr=@^tUv~93{cao-HmCEI{Y}FK59>uK>W<RM^p51OTHEw&ZbGs29JMZRr<__;h
zc)8pNWumG6((H0;Z$5VaBkXFn6)%FfAfRd9Jm+Mqoh=e)D8OvsPcpK7GJk7`SnrRL
zpS1k%`wW^f6-cXacSYNfgVBgutONH(DK=7tSLxXp>fc^Xqr#@qL6FP$yl0eyL<MMr
z1>UsS3&6kUtF$NYSrFJ;iVQ#)^0KIN!J2CaM((NHT&_}x%E2&i%t$p_bUOp=gXm`&
zBjyt6A*s~zjCwFN0lvY*!76q4g+F2s;LY?7KG7vb4uDSC*Om10@w%IaO*XhqGGkvz
z2iWhxo1w&n<0dRxRNiFKgCRpjbg_JAAhUQe%f2+sLDkd=utpt=m2B(l8la-A6c)fU
zcXl65vP*#>#`Ao@F8UT9V$1@+Kn>yZ?eC$0YZf-MH`9`D^!AzD#tMcaXY8-zC$QD1
zfU!qZO$q$(i$Ursdp)$;u~>3*3?lzUbai-QG{v-RiplL4HPQr8na`LtE1X40_&<jO
zvS7J0wt$$rG-Oy6UhC_m%Lz`Z$Gnt27SjkjAFLrgWUJim*+iB`$`IgE(PsgrVS~YX
zH7T!4*-b`h*EUUVu)=c)UBy@Id@(x4T?CU7cR*f79PS7ch2aT?Mw{_AUuiqatO}#L
zJ+tfYcD+CC{WMKm&{t*c^Ygi~sNUAsU@Dy#Nxi_JYq;k*PQZnKErZi&2l8?y;3%_F
zky*CLjg*VDb7PA9Uc3wCm)rMZjn5i0Y+;%%NB(Sa)O@vUGXA{WRUN($O_5B9+CjIR
z5B?T4MSBsx3QMUt{7OsrNH%Hy`-UYjL&P&*EWbmrBn3vZYc$t<Z8=m*R;}4Bv-RK0
z4WCg~jZ>97K0o2T{cMpco@Ctdg|@s7j5G}|o-bk5@s`_U8cYFINV!G6_Qcu#UrTRz
zt=ZUqUIoDpV+F4*cIDtq%BY^d<CA@p)q=c$WpVSV$K~&~>eC7#W9y8qbkqBQ$SeI(
z0*6=x4vDx-d<r;}S)fHJ>RBI5M;%P%N7kPNzU7i-QL7MIaZGl#uuiR6WF0hKbwrPj
zj#kY}{@BNn#)(t{yyOgv>gtB_!^YbVN)L9-m9?9F#lL!mAimG?q?X>S$6h{SY`N}~
zxw>k_S9iOkX1~lBee>?eEGs(E?dWu_)Gxyi7R8ouv;OvnbqQv39#!-x=2s`J-5}?4
z^E^r@MT%4eE@7%o({a3nY&g07dsg4(QlFV>r=L01V$x#TK`QD=^~2TG$=ya31NYjP
zfvOd1yhVEhp3oymTq!I5c3efTUQ_kUxh+-*(Qw}U$37XzlzywGguRJGO~rH`+1``l
zcBnsvH1rl8GblxQ4y}#P9^B6>``c^~{a*GDXGR5}xdlP7_JP}I*-NsN-*rzq;qbPf
zf!LGcD}o<&8EVw&2YDCAOi@zjE_a7T-5Uoq<rafC(mq&|<LCskYc<y(yvy2a$U_MI
zx}MdyXvZ!<yG0bJAoaSj+k6+_jYuxK9awD2A`CU^SPXuJg`-osY!iu()<J`@T?buB
z>X%ptKKSHUvmcCtoxQ}FcR3gh*Xymo(Cn8RAHn92XwYttklry<zZBPk)0rj#lGpT*
zwG>#XYgR10jm{+4ZOW>Gz(vO?B@Hsa^8~|0Z~s6|hhM%wuYe`O6o@#S`*uSr1|!tC
zLIbtk%+))Dk>vI)>NfI*T1TFRM`4cYefGQx*1&Bt8sj&=7O%R6Usfz_xh)(XA*9x4
zBJ)n*HktX#q^8I4s<~I5q`mB}9f@qtY*Ie?%^INna@Pwam0Zq$D}B>n`3}A50RE!N
ztGYFgb7mnP2y$iVRmub|juvJOIeIHMUkv3LSLU&E<c%<sb+JTrvBNXH%6k3*8;Irb
zEJD};Um7h|u1y8th1I<S+0DP8R0LA1I=3J6fTRhCnv~&oQj}A8h<6jdSu`=nSo2%d
z*C&EfY04TNb%7AKs_pU9=xJcO#;9<)tMjAT`P1mKGgMN}!2Eh>I3NHLWPm9Qt10C5
zBP+863S0T+9X0Mun^(!3pl|Tj(!mXxUaFtvn4n^~nC<Xl*!=+n5#&}I=i1t5SIurl
zg@uhqF55C37XVk0vUl{YEi5W*Y~jjrceEN}HhnCY0>hbor*1C&@{I_t;c_W_{}f^n
zV6L_jrsdJhk9YUO8!_+w5;F1vq20}vuilQLQLrh!8hlsdjK;vQVr~6i&uHpOj^u?t
zT3!m|Q1{f5dX>f#1Rd$k3SX$~)@?x66SB-w8Tiw>?$q4eE`PL8NqXtU$y~fJpO5Y)
zaha(x7c9S1_nZ?hZnNfgxT}t!3nfjFKHJ#|;xx0rzWlcMn9BB3=x}kcF}-f9dnbp~
zlkmg&1~i3x`bc_zT&$EBohX)mTgJxMcWd{oS<m;CU3Pg9?PJaT_>M9mMCd){l%Beh
z84+1Qr&6RW??v?6<?0#)e-%86<<H9GE(TKA+?-Mt$daFZj*=hZ3GkFY44b2afaP6W
z`=PL>4YDVuT8(I7=#e4tHqrW(!Nkv`T4$v@73J~91GyDw?YkDoIRg4Vmvlf3g*v|I
z?uGlrP}xE_6760OR%3FZ98|3fUjNDOn#24ocpqxMR~|#txQlXr_)CE(Kq@=8xou0Z
zs|cGR(+w}yL(^~bm^A>4ybx+kYVO2}L6HRBTOpp`53g&GIo~FxV{fY=*IS2wEG~S9
z&mP##8^mN44$^DqhpAFW1&PvalmnX)%ii=@{$(-R(mDjBl-Io`^#2*aj3-D2qjjh!
z`VUue#y8tIHM!waYwiLozU6Q|l(7o*@1e$QsTdn5$k|^oP&FNn7Yn$OCmvM{U<E(S
zyPI7c&j?1-6t^XPWhj<NiVM_yl^v%X4OLdTVNEkyteWj0EZ@i!GTFv%!!h3kA!%mM
zj9#UO_GD-PTtI?|%`6?@2eg^#Rx6)a4|4blC^{<B$}qoO(nkh;EzifKFye{(=tLCS
z`S2d-0&UIoc~?Ymv?w%!Pur#n6o#3T%F2kWf3)Ep6_-)V0`&=dr{K~dfS6U{jcKW+
z6LGiG%t(zFmfi2nT|fc@h1x0t>?4TnK}O~LRZjyw^)oe_0}fjqyeGBZP9Rew$L?y$
zQIx(;SVeE=*%@csiLa&mPEw~nh&OjeHZVf!YMk{}|5Yge(2Xz1yzPj$lwTtI;Yq9c
zJZjIfxw|;??b+5Kdiifjl8+lbFIc2=aWGKT-@Rs=`%IG<p8Wx~{8zS<W2OauA{%-D
zLddNH)lXQP?WjbUuEsfxn-s^XD`<?M*$I!EORAH|rN?o55aqAdSMpPwDPnStBxL%K
zacnCWC1eu*Pi?^npNo_a4yqRH+|DCgLQPiqzTc~8m?3+=7q;s_xj8>>ygm3d>G`|m
ziBGl(brdg=(qp0c!#A%gi<DPrhQ4{6tR(koDB15|X%g02+zJwHNAQvDPEz06#mD%R
zhwi3EdSwQymbIn1izvMppOpAy><T8wfg@i@1Oqmd=z>uMq-HuMUJ9v$3_HA{!vXar
z`9Mw`&DqdSAqF~AaBhWIG#;@Ps#fME4N8o*`^W%Jur#eqd~vaHQ3U3TCO|ZcPNDZ0
z=K@^I$#qE+L{@%sC+r2wL^3HYXDl5`lCriMz`1e(cniu*)V&q*Rm<Hmj(6Q~uCO-y
zmam;QP=jn6@D?-XMtKX%ETYCn%nri879!H`3CRFB>o|yp#*H-QN>Op`r=<%l5+c*P
zb2bGU-ndnbG*qlZz|*JTCByroc;t5uiA%-Q0+X9QvnN0qwyu=EqPX<=jG>nVrjQdL
zLo_RO(;?xoE(H-5i+X7sDTDD8SoEOWf&5@86`NmrertIjRzAb*J~uex7{G=stJDJ{
zf_SB)?wYOb>-Hi{;yu3)?^V3RzVP4`HRagUk#IY(A~^1P0Akw-R}p5VARUO4;E#7x
z1_O`*P1LnmDFe6+#^VepE@bz=G8B_&V2K&bX&6H)g(FKR<jG+I0vxtgAi;}FxxKuh
z@4K$igS{9Ub}3vyySNMV>>q@`U8a1%Am$*?QLHvqEb!prIpu*n1?H2fd)GW?0Kx8|
z6lV^GixprF^8mRvLkLk?J;*@8Nv)zMyn^oe?)b=<cppQksjMfsylGaLMCBUJlLL<h
zgNTi8{}e-IwH|jgZ`WTgt~tc~^AmHAU)f8DQG((xV<*M)YA3DUc=XZ7Ik)b?O8%Y#
zCpOUqPv3$|HrSh>%U`XTzc{+XU3?mXj5MbJ%w@`{sTUSH?&uC?(NndS+P!BtK7Zhu
zE_)g=5+><&qEb$}pF<oEQ+*o3?z)SFwL&7PWs;3+zNz%>T=Jd$5aEkMVe&kuw(5S7
zV0V9d8wk`%MW5@xi8PFJpGiHi4j+`_=MiCRpWSuiyTA$y2Syz#Z5X^<t_ekGde3ku
zs&`eWSFVto*kwN}Y}6+}@}5i=@<I{@Z@RRl;q%bTg9ZES3Wp=86H*-Q@~iz`$`?sl
zLE)N4p7=@pY!*ZJ?^>cKSP*1p>^c#ekb0#j^SEoDTLidYXYeJ<qznYN0V7zfGyEI7
zyCdNaCjTS?j8yO84j|J@pI)MtJ|BfY{zUIV-PMOqggBr|$KZkZqz@^O6{uj=u9rGl
z(LQE;5*LI_7b5Il%-Z^5MG-SwVml+IEHes`zLyg)BTA08gk=(c&Wu6Ld)mIa3=|d8
zp5)X{;B%5<VV9+OR>Jf;Jaw*5$dGnQK5mD@bl~WQQ*&JfkA%6byZ&P4o$jl#gV#1y
zCdK&&x=Y3&37r)L5YM-`HP&BzsA=`f$NMN35O-<<fQ_3|3cOHO*_(A~qzSl4_<gLF
z@)xx56mtVFarVJ@g`wSrkvlV8MG0zd46<*&F5AB0Uks1`?H%!#7ro714DFT>8rW)s
z2ank?k<&K!S5GF!I^4IIc!An^tOKH>K&i60LR@fiof|G?)$#vyz`M%7@N8SQ4DRpm
zz+IP_B12AVL%c9)B}n-6#<Z0EI@BGi*^T{%j*hkW`)=0JMhmuX7~%$Rz&W!YRCph7
zyPdGx70pICE>&=#I<COfPIvtEKp2d1+UXF~2%$<C6EBOTBM*@Fbb~QE-~TWm^$28-
zt27#A^Q}@^@wvoD-72n`y%kwO$FeM51Po3SHN)=8`GFYP?pXH5A){(Hy@U^VA5}+9
z-=GHk9PjCx&%h8^M90dRW`6()8HPkg*5w|(dS>3j$)ILxnkbF4Q+KdnI;o64DrrNz
zMKPJwMjxFFJ1hBb;T^uhmsLhw6o2DAS#O3<wLj$qM5Mi@vk^&7ET@;A=eNGX&y7U^
z5m3K|vd-~0!?UmEmi)hoMvt7HcZQ)b!BS&NLM)pV=|Wg=O|zWNhKbA12vLVGPuROO
z+F_3jo|Lt>f>KNeP!X!EZgk#yYBm4qj!iJ7<+R<|4iff1HM$$1?asyg$(r$sU(176
zR0VncjuKvS{;50KVx#9~y*c;0sdBMDE~4VQz^hHqDx+=~`CxEbfgG<lctHok_I)4N
zJ|?g;${zN0j`B39m6<4f<{>InhC4q@c$Lb0W+(WYnHm!ZQf8pud<%QpqD0EW)_=bd
zE}<vqqN+VxY1G2>20webnPso@xv&v|uX-z=@nv-(xzA79jL>|85n=OXVqDasV4p~=
zotn3-O(Q8R#`o96T$gimHHu4(JHhwxU6AzMI-lIMCl*fw3Z`#Lf$^dYpQ&Gdyj8c1
zA*%bL)rIsUwoTjPr5=7C1tE7$JZ?>TwDY%F25&rukm2h$CJ0RuYTu!Mm$D5$9tKAT
zA_czitfCofa<)}Se)$X%-1kN4@YCqllN5OhGRT<z@mF?i_)_WSm7QVlzkG`rih#jP
zrf`c&6*Ydst3^L_xDG<6)Urfvwi{BD@BQwYPmc(CG%qi@_-kBzstFvHkq;N0KNHiN
zm57nRlLIqxY}DCxZ6u?iHyK*ZQ&D~TJqD8`j=x`5RIN_TO%^O`*ofzr$qbka6DmhW
z54ng3zH&IF#eGO`NUM6!z&8A%hudavN@1gwsZ-)p#rM`b3ufT-pk-#ilcK6eqdqQI
zp%1a9X0p^mvER!irL;eYeHB6wgD-1;l*kz34BzeLfsL5<4y98+GW@6~j2>IRjN)qv
zGXo%*5i8tZTyMA(T~`Q<$?)D=u`PYKRT@H{I9xVI_C`YLVvuUGl2^hzjN^lExv%h<
zuU4p~eR_0cTXJLPcg>Lv<%pl8SNGBr?WC{`d3)$rZ=>fe&FrWoPVg$IDNRlocVUaT
zP~}(J|9Aw{(3F>YTRNe{N;RLBUg@O~s~j4D1sC9eRXM5B_-TBJSpiwI{LjZm{u2cr
z0^9LAe#Csycr2dz+1H%o4qP6G?iVhSMN!l-m~`LE&qP3$S#Wn);Qlw4^S*q{QwDjv
zDY;veR{p0i!7J08>y!mDtG@WGFVqa&R|VYEf9!wf@F);XFB?k_X0V}I?lzPm^B7v!
ztgY7R$9|~GlqQOetoz!(b0Y~8tY`;ZPOUs!sG7W{mrGk`X3aI{IN02qT1sCe#mYrd
zBki^$H3vH~){#$Yv^fpZ|6;~Jp<tZMi;LG^J6`HcSmWTPj$nWX6ws$bUrp<|95PRr
zi}%JIJ{q^Dy*##ZzVRO&?BeIO%E%ksMP8tUH+?af`yY!IkL3i2R47V+Ll-6gCajs<
zoB>sM=3fotKF7{b0%Z*^Cng?7P4EtVXJ=U%5)LM#P4FHV`n;tY2YIy<EBD_2@*pCy
zcPD8iRqDrO1$pkRu-f;<8yY8NN7I3Gx9mH;AI4MpbM#(P!jEN;hanahSdmuu7c>sV
zYKpRMR~8S}e2MPx4u?!#HLIp_3e7IC`^%au+qq;PMy%iC^o_FbHl01&ZN8eiSRNpQ
z$DhpIlFyT&ldp{DaAVXgB=e}Y7(NQb`NT&`1ewe`IM{w^<g>DKiJ+e5QPQdvXR7j~
zSufAHne;x#r;FMnmAL0dMTg1RhGF_AQLF1gR=C7&7Y<w)YA?@MKUqVFxEOh9)A(^2
zF>WFHgn!B)DR`)IS!k!n<o;~tg$8P3{A=(RjW0LdM~HXIO<qT@&vFa>;A-*vm$mpu
zU5I{d1$YW$7R;3O2L0IOy!{8PHk)D%=M$2?F(n?NDJxI+JNy$&ptqsAFp{N?IEK%a
zP7(oTrs{9~02!$Goc-zlR+xX38v#~$aD}t<#m6hQXpZqMc=-~R@VyWtbqZ64&rC_E
zMg*obC-p{knmb|kbkQ^<gwFGB(Y6x=;SGG7F1!_Q=XISj30EfMOzooh{b#T{dtoTw
z<7uc)<xG<9U2@_Vsxr$oL60X_Ps}FHE7zEJk(jkR^0Or`W^-xOIdvj~M<m{|F%@a2
zeSO`pR8kPJ%>T%=n`Ak4#k;@VPvak+aR^yft)#$L3MI<ZzTb+mmca-FnP<EHWPf&x
zE7uo95WM}2KS}0`hF|v_#9*b3!mkSom|<qyXR3quG+!hXX35#8rYu@35RY*rN6@>X
z3ld<(WV@YzSMa-X>F#&mATKRn)W5ATY#Q=Zv8n<gwvW$v9cW6Q1rz;Y0`h1Q>9Ozh
zIAY?EBzDVOARcdFwJZx+0-Ugkr~kHN$1!&oIiqebCS?gbD@bb`r5znFrd93cbWNXO
zQlq`>$n~7^83+T;H>12Ry4N>5M)K9B7%cT)1#cfwA$lpOJ%%!$@d3+K2a-SebeE>}
z#cnGGBj%mwS+BT0!~uf&`(js4pwf_&-KV}@NFjH#LSPtUtoy|N0Bycn(v3oWMX_>?
zvYd)ZUj${ZN1&{QDB{hjgv7v<<J^2xS=nLh0`AkF;I)DE+J)abFFbx#@PKP6@xWm4
zQ@rizI{CvRxAom_P1|)-G<C#UdWZq6r3Ai*P{Owbwy0D$Hc=Ig(${py+Y_tps$cR(
z+G~g#33SR6Rz@-sZ%Ih}|K&9L=V?X`?mYA)UJpkf7r{ORN9XCQmg7f@g_=;{o=Du$
zrIcT!kQe5pI!!s11P;(DYNoFPxck74jq{-KufTsx4>5$~pJ(FVANiev$;GZFzG?=E
z&B^N33tXS@jSJ<Rnek2x)Ul)51A07t$u-B7>ELVmU<rfz1fF`KZ>pFXdXN%|l+E&L
z4Vz@@gu#>{5Bb!#`x`;ER;?wGR2Nf35)U5Po80fhV%}&L^SHPtZ~ypCvlyK+(4Bzc
zCxhTv?+B46mE`&Ng{rm7E+RR;1xAQW@U6?mkChV#)FvbhOnSBK(v6C9zVpwKn2!Tt
zPG^8tB4F7R`6<mKRXM}$tjdJS2SH!6;rGMgPxza|FR#hRAL0gcCo5I!qi$r?u-Yud
zH5&{j6lAx?V(Fc$tkwjXPr~^N(}pP!x_;VcZ(ei#=PC2gPJx0tg2wQ->*e9w=0`xb
z$OM{aiSPm*pk$?CX#(b`VzSQbfi!QtjRbPc=2&=9aD&+wfJXqM`BFnQZW~zOJ)k0b
z`!LG?pDhP0^(vC5A3BTi98<Fjb`)8bXeHCd<w9k?i#_iJ(8b`+RNbf1P4p)6h(6s8
zrgI;Usjry-dM`l2J@B_66lfY#1V@o+fBiBsYq>7ucIUJe^FlpsvetlO#~oXxg+~nd
zyNo4%7;#(7@c%JK$QSoTkB@&cCP=s4CV2$ZTJxZ==VZYoo(U||_BfEHRNgIn7CVhC
zd)Kp2BJcD2v)h)Lx3e*{IH80w^z&sr4d?O?mEj61?!-^KS=e3#tr9WU>Q5Jx55$N#
zzoTGKogT6iT6EhX-y0ng&aXE(?f&#X4w^qaXLcJ5vK{%aw&@=m6^aCLT(Xa|8!XDq
zq}T__dnc2XW!yip(S5a(zBx<X!T6IPvhIyiC7EoIa%pT2Ft=1r+!1?TL4zZv08#CL
zhbz`JGBgbOnfN>l%!GcBx}A@+=CuDZ{6{nQ_gCmiua-1c?8WP!NA&{U$ZP7R>8Q(c
z8GQ<$aFk1r2T4m`$6k!ggOam$9u*1$4nz|I=1eh_p7G$;<3`z04#fUvs5U2BZiKYT
zGb@a`uoa>M*c?`E>EcnFC&CeXW^yTv_1WvbAK$9OB$wA=#<Q3Da5;|TrFqQMhKdcX
z`Qx}<sxsIgy~F?0P1P<=ilV+M{Rbj7zWJ%E)jX*I-FTcrK$Psr;=2!9cQAmS`+lB$
zBygHmOsH}&Ctb3YWm7ENWC+!o*tu8)ofHT^^_z=~*wWGZb$YWW4yFH{YgA98R#M+B
zQ-!^D{%@Tu>np2ylf3aZPq>(RO~6chyXjlK3|iPgbBTsAAjEmCHRhNDlxxA}PG`uv
zQXs+j3x&4E6|j1w-B^VKJuga?$)w-ki+9k9KPaF|g_HH~t(Fu1*(!I%3utKx&=Cxb
z_xwncwL7Zz#YHQmIj)Ay?rVSbsUqLm_Mz)3`Yn}L-tvH1(CY8#&`~~GEOqt$pCA=L
z&ChA1*r|Nuc<HuS@n<3bi0Gt$8xmRcSSs)SeX6?)qGD05CCilZ85Z^DQ}|`)JQbA*
z`gG04<VpD~#PMQPC$heEO(V5GED{oEx0mV{UIh*Ia9=8^Yv0<q91=<X?Hhz^xq=rm
z)*BgH(d6YK8;wA~n%DV$(QTdL&GFP*?^wG2EM=|!ocU@TQ5S@RMk`(yEA@hR+?TS9
z5MPV~?Q-%k5}bTvNZ@4f7gm(8|Iu&N;XvzZELU3Qor1ebRfyOvvH>3)rHCcU7+gE6
zfNS(SJE#5Qu?%l#hT>q*flCR<tX2#%WPt{u|9J)Dzn0e5RwJAVA2|xrgg}e;#;3pq
zHaA>cqA$Io0&;#>=~B*u&1YwfHJpHxFh~B1h4gDCXsy6;4|(>&$_MeTk1~*%`|j?<
zMNWLWkS_HZeWTq2zjbDdw825+o*Iih<w{wLM?aOXk1&r6Mk&MgP=6LMo>*S(l5qS!
z{m%RHNFi-J7yV3e5}aPIF+-v3teSUMQJ6D@zWbMFNY8s99~;BZED!14o5+X7m-b~T
z{lq>=^@P^SQ(>z&(`d~n@BMix)#d?_*+A2!4FGFq1zFTncEYlP{7Z!cH!Oi$?7rdQ
z5A?eLP71-`+{DxtfbVKi1hO7{#*3ZL%`O+;QKH~<|MSKDzu!&n0+HnlH2-?Bc7LEP
ztmh9CUb%Zq9(8OS$B<z6aN$B*u9rf2aiA<vq?E%>i)-QDA_yib0^AO}!M6jnEZrjv
zE?D(^@K0g;H31@Y7PSGFjKak76xdKEw;r?$UIwj#0O9dKxjdXmvb?v}D7biptU3a)
zp#*Z9b-y9?#uKjvQ1Q1>P%r4r18Tm+F(}r)MQ*hJW=Kr0-?Gf@35SGsyN+tV1Xy|b
zpO7r=ZI^lMos^#G!hQ$c<XQ3Aw^@v_I_qESs#LNeofrD{2Z>G|__zeDt`1|kC)?Y9
zyvm%;cCN9bU{dkzF}M-S^&u~$+teAA#|yTu|KmOTkGEQk6QZiBdJaN?K@?unZmR2X
z24xrrMJCTjw8c_w{!=egHlRkP^s)NI{LdcXKl;rG4tPSgq4Y?~3%t*|Bs&ANs0DQ@
z{Qa>U^+D1SJz>MS`r*^EuOCm55*ETNf30gdVJ3mWTfQGtJ-3d)D{#abeQxF&-w5Yc
zXA4u7sj--zwKUTm2}1#)M^aRTpB9W#V#NJPdhC6_N!Vb`)la5R$}*;-;IPIl#OBnP
zvlzE~@pKS)X`aI;7yqb0Sl)eUi@c07ltMJ6wthTA?`>n#bm4fmy$7aicj&gFsO$yZ
z95&~>W@@zz;b171R8iTS(=AQxp2FhcK2}wc@+NrZegNOU6F&1*8=g~EM(=D-Fig1J
zhH&D>zwaIa$#hWgSON22p^qC_d2(GX+U^0OTP^@zHC!}BMEuGm?5fJrr+mS3a{k4@
z<{C_Y+-SSNy1xInMf|)5?}sr=L1T^FMfIG&80gMBSg<$XzmyQC&uR@~L15j4)N2T)
znq6Vdmy7Ys_CJX)XL%#O7+<hFKp^Z}svO<^3`#F$jY^}Ewiey0^fK3;(EyF~#OzTp
zqv1iL`=9MHJz1Tt0rC_V)0R00trbMIx147~Xo(+a*)!&W=YId~k%4EeB5iE=>u<lP
zZ5#4rXRjQMlGmDB)j`$3YtP4*;=R_v(k1?Bt5}G+<Z6lcQ(Id+z^yz=Y(0cZaqn?6
z%A2myVarvuJK$k592y0kHHk|<r|jV@L59GVbE&S-Dhd^H$ozZ56AujqVI7Soz`!h5
zthPbrk5;3DIl%X+)qhnpuc!UJ9R1f)k*gvD`?UzQ>VZ5jh<(s24b3KwaJ3q#lm4Bj
z$JzjkmU=Pt{ikqX8InHAKMGEtNpJkz8b7qRnMSGBcrMS0<Heuuv`Z*Eol;?77nHU8
z5zOYKfUSrGiun?KYVpK}5^Z<pe3=wD+-PN-fkDdKhbsyAa(-}sMuT7WXpJvvD2%u~
zwvU^Cd|)xTccL>`<2nwDKT{PJMpP`Y5#W3@UF0CA3sJJc2pH9Y@ZlKG9KrzYc`W@;
z)hJQv#Qfem=_Jzh>7ve3?FLp^9$O#nt4Ucw(`AhZIxjwlv2;2>X+F$W7l4T+Vk4;f
zrc5<3ymdLbQ#jC0_14pIg&wcb5$~PC(}X0a-NkfMRvtDcSpcf%@8x$4<2JAV`%?W~
zw=4KUrJoY?YE|h#)1{ztU`k0{zF+1v%ggug*Q3|Cp4-DLUB^T&lNG`rD}mt#!uhQB
zTGS~@OacLo;aIv^!#1H&k$w`}OS^fK8@(j!W`s%t8g_<ud^|$C?iao#U5zxUVHf-4
z|KSBdke-)15GdKOD1wbf_%S>4^Vc8&@^roSMD@f1S0KTBLEGKfXFr<yl5SagX<fD;
zS+~8r`3i3W0lyCaB#a<Z7>64Cr6)mU-|I7QPbf-eeyVb<v)f{fCbZ4`6#VIgA3SBZ
zuBnN3w%y;`UAwIIrWv1nk%Z?uy?d#bOtyH5$nK^pOji^GIYXA4-n+XI5BJd!;?*{i
z;(`_=EpUN<Do4G_0!!+9gBI-#j?e$L&Hi6sod114GcS)W6mY-I5ZFyhn@V!t63FlD
z)>(5-W%;MRumOkU>k-;?kgr)IkyLEw{<MUN%8j}k%D(A^=Gqd*6=b|Wwdk{O?6f8E
z>6>STD$7nq%}$g@9<`@<)VY0=rFIQ=Y$8>fTK#-wZ#tjbvt3W5p#I_1Hcq|f?yf-2
z^?TxH@N+iVE7iLF*lvvzx#(B??kqdMlTp6k7P!dal<=KN?f9A%s%#IVKEK-tmo%QQ
z`tw+?m*uB!(Vv;-KEQ(SK3y1-d^AdvM%>ifXt(C7sz@`(&x>-nSfGuo(d=Tge72{W
zSdjG1qF$lajMM0UUli;g0wN-9RV9vnuWmP+L7pYh5EjsiP47IejwdI>G(WkQ*rBeD
z9iTo7C~2;VxV@6)J?PeUSN%w@zCC&PYfBhE1MN2%-W1<GQyd|iRFg>=W9w9&cFlAp
zUHD;K1HAjgG<m{q-+;AW{!Fv0D1&iD2LR>(G!1-~x*61NbU$(`CQldP3@0bh-ZJ9l
z6%hy@wAvN4Jn5n~jbm>gP`+YvYtT|zE{Z!-k*OBYdA!FicS@J4^VcNSahtWlJO<e}
zf@{Z<vOAxT=0~hmXQ|h({$q8RKT^a#&ns9*=0I^OQq+gF?2f_cHNz4+ChZ$Q@L>6`
zqV)gTE&u2)#D!Pq^jVC-S<}n<{H46uihB`$3J2nkO`0zS*y*x-{W&<YcI@@YYqr67
zH|IAyE^<Qeb>6;vZ@k&}5m-G!LHI3>ZvF%YYFHHY)<i)bb*_`{44qqo&Vs(fp7d(x
zbICXPCVxJQn$FnG;$^dokGKhAL8-1MWd`%AzojggWtWRb!MNRgO*P)Zsaz^~6XxFA
zXb{+-+ui|HkXd`|wcI97O%-4=j<C>(ew*lXyhLe<|9I^UaF~JljB@?py<Rf(4i#%q
zmj0fVsm|2?KrZsw5Ptf;=K+(d_FLTj&G;-a-vE3QmKEM_7u4`kLr6J!8@(f4zD@l9
zzIo!Il%7cz1DTa06S9h%J2-#atVV|z={9b}WG@<GX9Fl`CUKOEmr$-|O4;*`Mj;Ki
z`;glkqiSQCNI80y&mVuP?@Z>)b$>bRz_hIy^KLzp!AP-{+0+YH`2+p@wFj*liVu0W
z+`$t|6+x`4?`oNHu)xlIP|Gm<osJjzaK&;A3X70Sbq$BKxOEf4@?Pqc0q>#=Tvg$a
zLTo<vjKzoX^9iWXb%mjHMYSnwBbe;S*oUZ;0bs}O!e+V2+im|ZiNvZbz@q6i-RkO>
zk!`-k(v*o~%Qc0V$Aj=xjk(`kAO)A<KK2@=6gek~z)-2JnWwU-e|UGW!2Y?|DMioe
zzZ`D=pZll<6B!=DYOPl?SE0l|j>fznql_23fH)fm|5}xNx(LZ9r-EmHZv5eVgtjz#
z%({1LlJ^RCBF%%^ANe*f&CBkz8-XGXp~@_wkyW`6ez1`f%tY5MT3ei~BkXYcFg1xC
z;vcodD&7tDUx>lrog0o!p*p!<8N!x|7JqjkMlRMO!R;vjp2#Gq?Qtajl8r8y?~x-e
zjKgbzmpe}79$37jJG??;Bn?Ec$CwNR?f|z4hRrU*D=WiEj!a=OwCLqoqNJfjai1>s
z>&Zm9w_eGGdHm^buc?)cn;KNlWZJfS&FE>qFf|(GPbxv@BD<U^T@Y=1kRspRpYZP;
z*Z*9j2n~4CDde#v%Wce>{^|Mx1<Bntf6aK^X@nn;UJQO7@|wQkoW}<e=&wHw9{l62
zPnuHbgv99{kChRi)kLK>i|e`A`LiY1upR{Adc6LwD?ZzdT<l_xo!PtSgLeYi%oe}p
zNpa4-dms9(-O9k4lF7pz^Z<&Y%B}9uM7FIx%@42m<gws^@O!Q*)?%elD=zF4#w##1
zCw4iPJAa^Co-Y4$FgKBN>)vYr%|iKJZ+NYa29zatjj1B#^9!5tOmF+A^GTKh_lMn1
z+qA~(R1*{@&`6aVeDAk4>le^DP^yN4_)=2;*PmYIej-$;^_TFU*x>p14)R~7O#iG+
zVz7`DZlekB=Z4(d23-LZE+@Qrr`wyvRT}qWpMHgP7;%3=hU*P^o9l;u<~vsn<-5)I
z(K)T8SOQb64n6(yvYI4Wk2`g~=0_KlnB5EFtslVr7>Y;NmvwoBqh6*%fGC|>JUXXp
zbXuiZ-yz+}Pv%ivO6eYuFo8*H#3=%QdC0sAT#|P8i%<cL#4(c96gS-$^~@*cqN_Ug
zn1QC>e&z?23oIi5ElU1K`A!`#n~y1L7aV-8HhqnUSVx4vwN(%(P^YY+QLe9Pzkko^
z<?BqVL=a6+3a-_*y+dj7oVe5x*YH+P0Qzh|&sXqAXbRL0+Kc>X(L9-z%o+E;*C6af
zn3{>>Nm0fQTtYEj=bDI?M>XEHw6TP9PHV@d80kvGtsk?2qR0k-5$-d>nFj?q8ZhX@
zDgu+aF_>Ldr&g^`7RW_|-BYF7uq5_Vu!KzMp98F#<Mz7ojh>x~rk2emT~BJ-Zo``l
z;(b_gxqx&*3Z14}y23ie+Xj^qaT@O@-r7sXMt>fRy~(7X4IVj?DihP~|G{HAPI0<m
z=GvgJRyzW*j2)8)wkL7o@1kwrZ|QDJJ$TWp3=Q{L(G*Xr&Y%aa<&4}KzGqg*vF{1-
z*53t}qTPs3i_8kX-H%e_)5ro%!43jjbCA*?w7cFUnDDN3>Ew<7KRr(es5`K_fgk*V
zB!vI8$%7vy!xGkfAvYlq6_wYXSEu>9@#N2L2kOPT(F@NM?!caK#*F1=)O?o<xT)%I
zd`oVQ8JO32cO3+=M(k$D6APC3lSj4q&mv9vri+~gRLtqJK~I|jhxFyG!X25*n{jZL
z)kD41;5z^8<|gNly+X;@l3z=_RH{wAMyeCn0?%fNF;S>!8p$r?!0RI9ke0`&I8wg?
zKb9L#swqUmg$wKu_is(Z*%iL1tNa4+p~m>A6}OQQPWX!J6=9WOXhX3K53_)?vj)TI
z8`Q9@5KyaJPM=;H)t#IWvX0Gyf;D~7N+;ZZH>me32X(OZM=dVrRyoD$>s2FI%i=!U
zIxUpn%eY#Br#Jum&Wab1LMC>1f>O(l<Xj$|z2(;?=?1CZxvO@M_dJCfs{AIU7*zIQ
zAbm02ak)R|Ip3RHLGK{s?-bU)zCyM;h{NIb54!$XAIuX40h2>YvbEY2A>VWwrhTR5
zE3?nE>lOCHO`dKKfcai^X$mT^`slmYA1|@^ta6BE(>Mz~^SIX?b3NJfeHW*(u9d%z
zFNr}@j=Z`&Sr6I+rW4*J&hH^8T|Z#vD~d|qVSdzb)2z<RP@>0Ha<MIt#JAWu)L6f{
zOS1G*JMjLgP9y6<C~HBiY=)wuq_0?>Z0&dFviEv*sp@Tob@@B<z)dnVLilezR*NIl
z^8Hq7*uHIk=(l%i;BAzbPsYIeZA0=OSi%1+QvdZY*nR?ou`kaTZ8=$Wpr4~m7i(7t
zL5Lgz&_^<SHL482&%eDK&IJF*y>Zdt-9Vb4{^x6lsk}nL_(<KN3Ix_1UqD4Bb$*@1
z(9C+1Ta)V10y6q6XXpgX7b@!LF{a;SWW{qUGydsuSU-#x5XQRxGl<==oB46RLYUE~
zJnuOw!?F{lvWn&o-~0&Ga%&vjIBE%)4Af99fBkWM`csSKbxMAX^BK(a&x155dkj5D
zO|_3GOF}F@cide~Rc)9R9Pzlt1Q&3A3zMmUlce(k^uQE_lE6SszU2Rb^Z$47>M!PL
zBDz{wRAGE=Wt2f(?09ja{f2IEk7j!UMJMJREJ&2HH9nf>X?VP)8|W85UYyMrJpbMK
ztK)~ImVA(7a1ffy<u5(+R~oKAZ8c{R+mIBz+M1kQA~knLczz~fBc%8w{55XZ{Z-yp
z?ZmFID|ph%pTu_;(%l&Nc@bDPF5uO)B_ZcTYnjCJX}0R*C^K4S+R&gj70|7w@I9<P
zgx<wFQi+$|1=t>hwar)G@|B6>Xc=sVy-s`a4RS9+KI6mr{XcH@d^jq#7AYzT4S~R8
zMr5JZGEXArUEmSwL3|ppx0o0gC8JkTv%x=MalrTo)8v2di2xSZ05)CborkN7sCytr
z`S`&UcbsMPb-nzwQH%Y)!Pwd8&RAchDyxGTbMx<y%tuQHwy&AdFcuP)twm$se&pMX
z*i2DzyV=72+by>7Foy3A(B5cjp%ux?!#BG%3i3bZm(XCYr#TR!Iz?qNDZGhPfif9?
ze${-P*+N!hRIk+xEo(^nM8fTYv-+!Zb>$IzOm%g5n2KvJd-F*DfvP*BJGd?6<t<xx
zBUE{X-vU4~zq@Hy$^(GYNKSC7Al>>lQ-Z*I%(2kYaU?E!7%@K{8$tzd0r@7ZcqUzl
z8!*4q?E-9HF(VM%9nbH|<oSit3M8nF1EVg&*q=bV9u)H>rt?0keWSdW6i2<p_oUtd
z5J)Ehzwif|>MHAd&pPbB(a}8MNgwrixAHUr!iB1>jSLLNkjTCZ{FPZ=^6>Tjore5Q
z0yY8%dudPid8~6tq)yPp58tSH4V}Y)pcao#MdyM33XIrCKLgs_z|AQcwGLJtkX8GX
z>HoZ_J$D_7XP(NJ%lgMGH2ayST$ac<aO5M4C1E!=pA$aBV)nG&=oueMVCgI&16i3s
zNX0Oe#`}LacN@2-0J%VEbL^)SuB}NBR^w}q{1Q(loa>#y)pF7iZzgVQro#~$Aq#bH
zmEUy_5!J^jb*VWEcB~YqpTf;9U&qkd4FVn26QUCm7Be$DK6;b7<ckI0{uyNPh2gC~
zD;OpxvlG@i)@Y6|Fke6$7~F7Ke@30QiR!7@h{pAF4@u-*aPzQUc1~b*|BgXo4L_@>
z{jA+zz~PwH7=$ho0Idhwz{vKWKd>K=)_%bBh}iBFF1(;K+n?%BBDl!U3Lx*IeG@^W
zH%WcLOWsPs`Tgqn1E9101LP9M>6bH29&o||+roU2L*V7hMWmiaDGokP&uq-!4B9Qp
z0O4pZBepx3p&tt8jTd-_eGLX5e?cx?RoByQ{3s>hbv#ay`2MSeN-Ft~(O>$BGvqIw
z4!BFHOMP)uc!aKNGZAD1C4-E0bn`#Z+H%S=u)ceBB+QS1AZ`@`%&%JHwVwJY!#T7@
zX|fqTftpSd-#?UqnZhl@YD6Ki4Wz=os-bwGqYDR;L6kH7@HwouYmLkQL)Uvp!yUeB
zzllhg2nI1kjUJ4agy_B35M>aZ=q-9D2%-hiqD3!JhUnd>(WCboMDM+w$KLz*o^|%S
z&VP}$u*}T&dG2yupQ~{L_%A|qvdCWE?FpV^Lz)jjzgYts7QcybX`yq$kH&c9A2I(7
zro)$F)d;PAg2XA!RnyUe=M9I7A!kpAyJOvp%aB(`1iaQ3^Md}YwJ*SCyoVMu;$1tS
z`j<|~$PJuUPFFYP$ht|zM7kZ07}5t!*Enp<5dE=RvRoF^U+(yk5cA@P+Rvly_*fFm
zox3q{lFiaduVM6y`MdKdj?22*TIO4sY`kM{8fBK(<w%W$k4Ap<8xeBYb;}oe=bf)L
zEZTo>${b>T+m03DGqKW_E`)<Ad1HJ`{uX!9KGj|i+oSEdqKM7v<+Da*uzuL5|N0bd
z&Q)yj>_3Mu?PEUxO9?H?;r&85w5h|ur~!W^a57<}lQzKc|5si=kAesf@_4xyn}?hZ
z2puzCuL*@>YWLwjLs3(c=jSNAGW#p-S!Uk)`MJU6?6X&&j@pUn4Bv3i#@>+mcV$?P
zKjB>tljprYlIJ~;wiU(pQNf#j3&tE!&*L$d0Uieb?{C&ZmM7xGKL!*}FH!LPn@MEZ
zq?3}N_hA|@gs4PP^oM>ZOr~&o?+aADbcl~8@vi8}==DX~4mEVT`Pt52a+JoklXpj%
z%Wf1I7OpozbarO3I!i@I5^-H?g+;80T0xfjoXREZ!VEaV%1kaK;GL8MrwaAmpO)?E
zo6&SlB|RVg&-TdgF^w-vOGB_r)F4<1d>M_45?D}2Nb(r|=cNDNRgf8`@Lz6gtzxJL
zZ1Cp#J3)K8&sC<h6^4Qp<ZWdBky)nXGsjl!GqR&kKU{z)!2m@tMr2S#q;7n8D92`h
z{QwqK%E^e{s*v<3&!p!G@9~H{@5R57G&s3n{ns{0PA!c4oP6|eRgDJ^RxJEPJ`aw!
z$b$cXTB79z33#m5>2}@kS!ZiZXhoGGh~%Fx^0DCPW7R7+ugK9*L*?$5u!<zT$<6RR
z{XtwSM?^vMuRAp|o6T>x@m*wz2mpwj&B@G9MY-~TT5l;9k{kR#gJ)lvh0EQhTyQ44
z=h%bYyL;njw*R+i`Oin`zx#OM3<9%x!Qt|Q@hV%t8l6H`5E}9X&(`-5&R=sVIm09e
zru%-)-{=D=pWE3VMcUZIpC(fU9&JyGsnR~_CAn>4=BRutw8j21@I5k5J6F*cEMSfU
zYgZV_ud2+HAqojUs@TWtT<F18J#^3HYN*e){jvY!$0IeLo!Bc2b5~oXuBediPG4)C
zdUAzs>~L^T#O~^$&p0a*nEmBUCr?ry&UMEcK<Nc8*T1mD9)*)nPqOLV+&++k067ND
zY~l}*2blXKMkNYKpTJ%{pJoi`CZ#vW$c;fc9LYd88q$z(&+E29w>e4H8jZ;1fMDT`
z#&5mP$q)Yf6CtgPHk_kCK=AwkV-naF|6HHWE8QUz9$}Ltz+3>pR|H7TvDAvRI3+^t
zD*wMP-TzTYIx^UuOeO;16OZGlT3|Lx`58gIO3lrMr~&8L4{aL*zo@V%jV2OKv;bT_
z0l#Q`f15^4Js~P@k28bfc=zaahojAs<0Z7YB|)zZQHd4D>B{HF8)^%(^&V|_#;iH(
zfBYXB5vu9hEC(X+v20a$zP)?-g2w;SiIDQ-G!0zpbW{5tCAv{qR-~XMFhMLR&|$>c
z5x;ZZc*;|yYKU?(Kubq=gt<C6geY>0%-aNf?z~MDhC`6PZ{i$ztp4+8C_@qTuBAsC
z;I<tC0^jvPF<cfGvJ73%iT}Mm`%n4vJQi}an?_gIB$~k7+hjo|Uhj`V^k=R?+-0`i
zPzIYx@Gp&D5G8AlZPRF|xJ@+`vmr>AppW&Jg`$Cd&(}}V!^S{3Ihbe)`$}p*&C!Ye
zZ#mLo{cKqi9`|Ha!qkgseXafe@3&v_V>{M|M`0hMhF|pudm0E|U$|4HCULPQnhhk_
z+wJ4RgDIR71b2UbV^d@%*bBV}J1Ex7^rp9*;}_P{^xQpbPihTG1v-pO(=ICTsfXV8
zyGz#F0YDriGvEVEunV2&qpi#814P?UAuiDN%hxjtL<#)wedB)~p|2U}HjO+<Ujn?r
zjd4V+YkaGB`zd(*M-WA~j50}bkvP@fs0N+kINU3xP(8ej2mV6o+e^6sNAP;%$x?X@
zvx6OC5h5K{jWC|7?E7Ip{BgaDMsv%ML*{4~XnPEQb6F#TQ~sGG$ga0WWTtuL6vsTv
z;62%TR^$671-@rk6*Tz^w`^->a{iKd<+Z6Icz8nV;VSuaA9bVS?{$pRZO{j4mG7O-
z$>^#iABa;mqs@skLKHCwq+g+Rk~zqKUrG4YgI-TCfpy4^e`kRTVe|N}w>UaYSTVx=
zr2x_8B=Y}mwkI%W+%35XVffgT#YFN;_y~#ipK%Ony{p?9)=$oJeb8@!7N>`eEo;fp
zD<M)KPK?t6c~4P^(JMmq#V^Bl5oY(Y_ePdhh;gG%bVCvStqD*|NMzca-%6JXTuM={
zsKheOIJakz84}WKFN70n;=I8}oA)#tv~9{S36S`=F-l`=v@wPay^xR9w;iTm3w*pW
zLN|toN#L^0n~)Pheoac_d1?LCrZ#`@DOZ{Ao6_#&cgrcwsKAYVPEo#<r?)%gyZOt^
zccR(is73jSp#M27g+-xnA+8{UGX=W7EdF1o;uZa1mJEUUS0K8}0jGjYWGTb{TrmG<
zhQ6+#KY_GzU%2lwCZFb(lfYV!{EAGW=c8YE<EL%QlI65{<b*|CZCPv(`oa5`(&`>E
z`_%8j23fM5s4or=?MIOqUSq-N9d!Ma<9@yrqGrd<I1#Y%Sei91WRr`_YnKb6RCL@b
zovEZXs-d_14tAStuz#b>yz46J>b@4^>f`I6^*<FW<4EE81!p7Y6&EU=X_)@e;*XWD
zkmRXV&n>+@C7}WP-4QdK^c@|>!c?bZ%wsjqdf_1jW4+|?lZM#!kTUW#{->W24ud>J
z1%$Pc1HBNw-ihJeeu-8Htk{y|UfsVH7QQ}EaL4<<cEDe`&?UpVa+EV?LR<k#ETGnf
z?T66KAQsE;BMQ?IBD9i5E&1in%|J#&>jG|upfp=As#IRTLY<$sq2D=|^JhjFx7&Wp
zu$v8rzk#Vl`Fo#!dxv%e$U1OprQ{5U6#QU9r!htCitdG*E=~o(Ke%YGKK$#H0i#;B
z9Lqsrfgn&Z(K_!8p4NE(eMMo=o9y<LPW+71vfyxwo&}`U$oG9;A%O8}jfHA9Z+<3A
zi}?p$36(0IATm+ZkfY<NV#Zq9aJ(n_Sw7o&ci6k$|JGueD?TUiX#VdfAnW=4Sl?3s
zRF!X;`55=V&;PxcW`R@VVFD#%ERHV@dB~38VjV${KRw43=^zb?lIL30$N<1jZG_JE
zuL<4$PG62#Hs|3ZxKRM3oksFc4|Sd8n{Cks1T^F)`gPYLc<6<JWJ^k|z>a*xUUtNY
zce75~A&F~-lFni>+YofpJ<w%3Cm)vy1M%Ec60ejIh%5G=zt*z_wx`j4*M1L7;C#+u
z$&+y`A{*0Hz8A;LS!q73%PK`AKyy8_!yu-!@VWqEH36j!K2gK*dS2-K%Lpw*{P_V)
zAnIefj2t`h)1MgRakv)6k?rZ9KB0-$;`fmcZvTn77t%K{pyaE1mV`bMDDJe75_~*@
zs|&KN1C%8A#P}KW;DzdrQ&+)4e|9uS*2LvvD3gr{C8<x;LdMcMzclDAW`xU(@HPR;
zN5F=Wp>ow=W3GkJh6Z_T;Bd9QQFmV-HlO2=3`5{~cC-}re2KTcB!&I%a+1yxbm1d}
zEfn9tf*N)$>MBpT>F>=HwFAaQQK{khj4Y+aj5A!F-`Rr(Kr>PgxjUagbB$wCop64(
zFV*^Vzdx;{vX@}@$iefs6sX{zr+C=h=V{mwI-QXx;bvz(f7!(t!Q;nic{K1_A*t4E
z&@fdt;(07)Gum}9-bIZ=eQFXvPdwOzwzK@!Isrv~1tut4r=tr^lET(gIk9%9M;Lei
z&<qoCq$oCrPQ0-mI<47wO5Y!Ke(0aQ+#f+s^Xd$=avYqtsY;aI#5ZICm3}T_^2rCQ
z-^3;L=lH!SyVP@Nmb*3D<Al2#Silg1yy1K`$xvc}q1>bgb|=4JOyvd|mVD>!sSFwm
z#XIz6iqn{;YQ9Q#!cd^uGBA;5MG9U71Ja3YjtNMjEO*(}{d6pzl@EY*C|a*q7ogR)
z$;l^)D{q?+`NnD5PxeX&q+AkGDVzy6C&k3jP;^AR90n0k%y(HbK)_64X!+DYjQ=r4
z>0c1K*(!QAJ%97zBc=e-ep%9eyJEc8(FVkd6<(ZA{*8-|mzv;h1YbLX*phc?f?ofC
zlmw@NpX&Aw0t~ceB}2e}8fX%fhUw%qUaaE?rU*&rAbf;T2QaTc=yY{|nJ)_Ycne^z
zW&l2VXS+IVJO?$Hg5SdpbCAT<$7>Jbr<Flwd#}3|Fze?%W7Dpbx0*fBMsUASYRh*3
z8ps@Aa*XABNTzHmM!vzf(;^c?pV(^;{z?DSawN}SjG=8380ud60KT1>8MN_A%3}|}
z_LIWKOZ1e^0OY!*e)ns{ol(*093WmfcWVZD^;2w>o;(@=KCvnfwJzHcU=+MTARZ%(
z#;kvOXHuA$nkKD;ylnRZWPXi}$zKd9AG;Y1sp*{AFuukC40qx&i4!((8w?2b-P|(?
zb1nun{}B*7`AZRqwCCh41MfYJ+L`<n6bIMDhc(EWfymFapcE|JBg?K(q970BmQN3}
zt(!puBd>dTvi+*3%i=)<&uYRD*S)4pCqO$*nJ74j{Mhott4e!b{6^2`+Q}UhGiKmT
z{2(+43Sv=@5KllO3Xj;9sKI=HHCgZWD!l>6DX#=jwy=&P&TjA2E5P7uFir_5(5u&Y
z&dn_kjxRxa9_Ma5fL69xtqpdF`l@Pf4VaCd_owhBUCY-wZ}*l4iy@WCJ6tiADZ_Z!
z=Y-vX|Jgi_Z|Gr8P>9KB*Q*YpBJHXi5QdyU88%7n^g`%c=ImnQ9pGwPm;UFS9U!1K
zxSUFHgr!{R&8#EA9kXfXyg;a{C+yiPv-5+M>}$R$kubeMMH-*#NcL5+hW6^m)=Gt)
z=><phwL%>f49RwAceLJ>XeNR^Z(xZ1Fs-{IzK^l|G)(*IV|A`~6nr#rq8fPZQZq%o
ze(h+4nX|L{QBLuXsErVgNX_69hgO*n)8}~QtUcj=?^~}w6X`*DN_3$=b9p9A9P+ZG
z8k@3Q0jM#*Cypzo@FrUOiS0F#NnVCJbC0BB2bW*BOi*~4<%zjTT8NNb(-ME?EmzXL
zGz>6_#o8gi>cqIl*r4dn27ziHk^<LMzhPZG@l5VmO<xzA%q$es$vTxc;~w*wM4)x<
z$@^Zcbg;ay9^A@&>upxI-}39ll=GhO9EZBpye|j518v{+gG-Dv+uGIfxZZ2)sn_nn
zRd2S^9lKfwnTa9(X(S(JidUWlrX?IH_8rd*QM6nGnXdGx2{I@2W4vq54XU9}_7BJ|
zq>fU~e)J=Z9`OH4PwhZ0x~;})=yF_A{G?dW!JiXac=pKp8i3oST-6=4s|=(1%~0YL
z`}MddW)J>wX>a{`r2qay#r?cP@Q88l`s_(*_x)GB@YM?ad2}0d<IOczJ3=bWox_uG
z!2IFS$h*lU0kO&(W3XTs#oU@S3SO&3+m|Q$`224P@y9=*Y^woh2YU2&Uwva9>Vg@&
z!hRCKh!TdJ-oJR;3C1aLEJkmFG9yzWFGPTn_d+D9buaX8;lZI(qI*;R2Akwx>tsF8
z`NMh;QP|*eY~kl}=`03+!3;e#+Bh&9l^qx?;cZ!D_e0>HT^*KPio=vHPu>rhyq_rf
z1i{49M%D4{R6h_fpI8HlzpM&%32;u}D&9gm7>~Co1#3Q`>Su4uk}xOszJ`v;g@NOu
z3mQXFJna8E<o63P`Nz);!PS)Y4NFQshZI=7xekk$s1LzdjGmc_^#xICypW+GqFWS`
zq~{?G{ab7*3yy#V3HL_0m*_of%zgbDJN)%Xs6DCZYnxyQhlU$|j8O-WI2i^1?i;cS
zhKQ|K7-sxBA}bN{zCPAi2I4!8HuUn>R?jSZl;l^R4d*JIl9<#S0*2^&^Lm@^1Hj_I
zuB#La!dXEPEr~6IWS&=TZMG9N>Vl$8%HsehU;%#cM+1pNDa@>JFz|e4crfY2`xfJl
zl>X64zbzQgV0>9HU%Co7%a=i7;>&>SHe`8V8TK_>8?*z2mY8DhhuLa23%mk!N-<1d
z{#+?}by7e@uPQOwr85eJu^c3?zRfv{tV9)wUGDGt+)fL7#RO&_5}-X^23kT#(%2(@
zDjZ1YDVW<a3J`gzgKt6gm*&dBSvfvGd1e?aL6JQ!arwZyQUK?27s%SBNIln&Go>Qq
zI{ZCmLkUo#V(<?fozRnV-s=wH6);nN6%eS)_NRvT6wDi61vLM5_`YQh!;*YbTHU7R
zna(ileYtI573f-g^qNZR&7DogIYPiY?00~4CzLvOXj9bZM?^-Nh{uPfu|C`Kqz8|l
zu(vIR4eH0-+ybX~(O;@It>OchNt5`%Kq}H8Lo%9WmbR(lC3^?_3s|Y0Fi9j-qz004
zL`L=f_1mi12uM*q#6HDujluxC>~*2|B~{yZklj{qP;ND;(%s!1x<h)Jg6Pd@*mXRY
ztm#2voW2cO6P~Q8nsc}OmtTB)7h^TV-i~@YZV-?f?5!2oKWLgaxLJ6c*R)%I7(oS-
zc6Dzo>kPSI5>NSz$kKnfc~)dd1^dIRyLp0MzEZYE9EP3T5nt^L#@(qMYcPU6D3%O1
zf0cT<7;mBDWYe~F41)H<iJEJFX<Ftbi}+dxNZc`BX!xwB2+w%kX{DRmCp!Ehpu7Xb
zP?#*7ce62|w>|2>Hrx+j796X?qvB#!Kmnb6_W07A9C{E(9YXkT&2H)0aw5UNj_6Sm
z=H%k*|LT<Z2@<%95cOE|3{w^^X`3Z%(B+*nm$FovzgR<GguT7rhGDl;bl@B_Lnj#K
zvWapDN&5@mpkb42X^hc~q<LLd>l{da8yLeo+bsFvnmTqGs#AAyI+HDj1^gHH>j`+T
z?3{Qg-TpRCNc2;^x)1T|2p{z#?@%G-`X*g`uyeeg+ukimxF1FmTfI9myIN_|;}36`
z>x`i%gv<ui>L5GHV;G2P6dtE_qT|O2Z%=J-G0+&Ll{zp)mJn3#4RISFMHBb>Wt#cC
zuUy9oSjT7sj>Bby3?>KOJVK!g{6grMucR9TD@cvNwfY$uy)0K7z)AS-Exd3>BDR=G
ztsr4h|I^%02mHOqSW-C&PYF(=*jqVKUrPNT-QxBygn|pG`y!Eh#+aL4ADpkA?k%M&
zAjn<^m}CYIe^`N_u>^Ng+m}Awdn^|z36KZW{b@+v512n2qKtpeaK19_50fS{-v`sE
zpRW3o`kxcc(B{BcPl(ArHCC#;1%we=@z_LNwV*J9-xb=?TuZv!`Ls>9Yd2p4+`9OT
z?bM{`ljj7IHH*0fdxrMm$ibajek$UlVF*^(P<1J02OC3Si0m?S%t%Lq3ZC-}8YVt$
zn*pbie>!&dzJ1*>8d+)iSbWQgc4{6_76@XV&gs%`pw~mKA)z@cCuQY|3G5=@t%D5$
z9tw^xcA-9}AwA7zJP-Rg@KodoC@_?S(Ccvp&-8gJ5oXUFP_h{+-~OHGMh9A+`a>1f
zy-^qDHGy0wmZCmbn++WxgF0RQC)_hJ4H;n`tRpGvU$S=#I9Xf)qsEL3nu5t!&%d@E
zrj488u%YSNean8Z?L&(bO7)y9HV`5-;wPPq{6q?)U$P%&RZM)@?(2$(r7yXG?x$0>
z{mPE<03fLB{Usg^`VN=JO3qgR#u?jjAFsL}5_*FNaWSdiyINnPrq-!7=>%hGs5@5Z
zkcwj2RltQoA)4+V?s|;sl{oDuY-me^jsZ77AF;rs6}0y3s5O1D?wqYqKc4lqtm^%e
z3G!xWTqaF-+LxA54Lxgpc>8~D$~ZV97tp(<=2g-c@;USCTg~^!xcye|-&bS{m|II;
zhR=c3(SqM!qAzVpw!G~@=ij%ioJb$qvw>LxgC9fci>CvS4c?%OvsT=>HNR_lnjk^}
zzY0sa@)ggMN%&Ai>dT;oxMn4x^-aJhx;b|GslL8U!aeAd$z{WHW<O`!^st@3wT0_y
z#$k5F<o4!=U*MOr$|i$WyQt-h;~9)IWpy&ic05Fl$~;cFO`n(kooKL-=IarfE`mu-
zfz@?C=wXu?8GV95V4qyF;@X!Yq4pls_8;1wr}sVZ3b?Z&odybHKRGE<e2K=|+MI_P
zk^aZJb)3n03ljDmg$E(q5t;o20VD~3Pvn|Ik*cuLnQskSWSFh9h(7H0XUzB6#A%@z
zT|*n~YzzzxOnAG%dix|`kxp^41_8b2r$P;JH8_`z8#BAozvfaoZMwZyV*VaNtAj+}
z%ynYRt#6lW#`;-@JfeUQjB-Np7aI>Z9MN>yuH$CQ3dUdO^#_?~X?jypw|$B+o2{Dn
zX2up8iLPd%Kv)Zaz{YHYuA%J?Ty=^}d)vnI4{g-Mxku?j?cAuMVeo5}2N?x7xeNDy
z!`>}=9~<F?5#lx)#Ex!GbhAXdrmY3}5~0NS(m*lI{L8PM^~9lQ9(k}66r*`9MV2WU
z{*jQskX!79Z8r0iZLszqT!t?Xg3J!HYALQ`NlWwEQK7y>pR{A!6%c=qHUt4flm!F9
zaI%5ae$>tPO$k8s$^G3yn0sq67B<7sEwqSN)BllPY_?j{1{-(R!9C_ZrXW+xQHNMK
z{EasNuoU{u`ZA%`(d!XeTP7N$YPzKkcoz-QWQ98gFnIow!)BznD+qmzqV0K4X4i`S
zMtU=H`plj}{5KMw7L2C*t*P*OFIs4{=AX+ulU{t~hfASg*deWjmJlF#bGG-}e3F*y
zDnigbZU20&A1gi@hRMLsgeD;&!XXvS)<Ix#KiO&P3-b(wIYSnzmplTz&X^XPDr^04
zcIgc?#pbp*w#B=+wl#r7+jRP+Zi+*IM`Au52cnyj1;kdcsOk6SZ#Rf=&L(%7EgCp+
zV!Vt_1B{&k&B56=^4W5|wAJ|mex~NHJ!c^X(=<P4{ng$BSg1jI3ORwaGa3l}wsioB
zTY~J9J=|);X(Qm1do!#cYv8d2hIchE=N2?AS{#CJMDOkVh@X7@iCbCJAr<e;43}pw
z)~Vx)Hcmt-vB4Ni1=uR1s+PvQsw~G96A0(md3AX2vsHc_YhO7=lp9)&C#ob@gC=-<
zOs|DBF5Jm2V3f?th>5RL?irvg^&F!!ZP0Jz^+zWN2q5pRkAAU$qt_$rH*@nI6O8GK
z8>z?I@fm!PQA1oh;a`lE=*`clO!bpBDJ$+fBS+Rhyd-u)8!N51V!{2BrEv<{XxZKH
zx9Vq+7h9Fv?dM2$4CdT@yM5>$N=py;*>(0sXGJF$DZZ9IIc`0NV!7Ef-7o33&@)RO
zyZw_X<q<MAS_3`HgHKDA=~KN{8T~{V{3m55nojP$_K73~60_O5&UMylyr%Mnh0+@M
zquz&`zx|+oyEL++xnk)BI?Cgd7ZpuvXmh)bm;J+Nkw1%Tncx&*zXu}_>ndK_=i+?9
z^I*t!11-+|bST!06&hEqR2XSP`Vk>h=}_)%gr{1LgT-TPWOU?DyvxIb*mbhZe-;{P
z*Hl}f!sGbwZSFrK?<<hk5%jG=?a_&CxfJnX-jw=f&B?rO;gSd59vTkZb9%YVy9vON
zQYWziK%XRzkRIJzcb2)No*MfQ5cS3$t$c$qE1X4?zrTxku_m#D4vd{t7tRxhif`)Y
z>wVxM9JpcX=|g<=dswE<>o|}25sWmp?<#J6RcF;ARetVRt4p_Vd|)FQm-beuJlCdB
zaC$hjT~fDg2s%<{%~k{s0!f{{rM;a;SkXo7j`PGqJOVvGBAe^*G`VXVd*r@1q+Ws@
ziOevK9L>cmyQA8BFszLK-jzg0HB<VaV<X0675?^IA&cngE97J+NNiumxfBk)e^2(B
zwQ(ZFwaV-Afng_ku}~>fs4C<%$@U0_Vyz34Sjbly3;BwRS!3*#x)L8-p;Uip*b^G^
zR{c-T;+Xcz7*|nQDsL<ilSEHKiHp(?TfOUv-@6yCf1@@e)Da``y*RLdqRX=$w&T-%
z%M!wnnteMMQqV$E3zv}Bc{6SXZOxcfTd~du{o;tRPO0#+{YJp(^Ud5o8yy)w!Jg#v
z{D<{rkl@Tn!Kw1#Kx>E6<FFOLPO}IDX*JExRb`v-CTV)ByBsv@gkS<x4v2RupMfP!
zxI?EEj1IIJmaN7K6W1WG5I@0!K88yNnk#bk$lgx}^UH4@#IKrzZsrOC9<Pfd5X<KM
z6TQCRwP7?=-9FX(!rTaqfL6y+kU0UjfSNV3ibYUNXc9;?qg`&6E>B6{#ag%Z(&5tR
zl6Dsyx!6{n*i?Sn>9`w(4Kt2`zO^}JZ%s$NMqZ5!gQS`GUj+{r*)|>o$k)nPWle`7
z`)u#C73wYCR2>+3iP3(RnTn#3B0+ws1MQE|{!;rRdexGeRT3!!(kW~<9PL;w`<_Os
zjz(Zp4hKB>c&!6*?Le-_)ZA4lG-fL{^U#y`m>6oJ!ofy=V1|6!Rs_~WN(<iTr$Ln7
z2xiaL2CU6d8}jaV4D_EdAIA|WvnfZiVEnX41_%G%1|T-G9xQc^ZMRa?-X*E}2}p*b
zJW<}4)T860T+&Dsn??_HQ*Vg!%A9SlJbuA^IaY0!Hs~Yb`Ii=cNUi!6;@YBJp)V5m
z0W;FD!l|PBG5ge>8CkbzoYXf?#3orlI}#0UWN@zIx)}Iq17k;{&gu1xHq`$ZaP&MH
zG^kF%cCXy)!B8j{uJwaLEiH9Yd%eZmdbh*gqm2v2zt~<+;6!!yFYnuet*C%@M;fQ4
zdd<JVH7=VOL^rOr^6p$i-_2~Q61wjtWO)cIm`S9A)YQv%0Vz8!^wqQ5GpdE-XxlJt
zwdH?iRR|517tQ^6ia1N0iD%<XE+gwV*O#P#e_giU_hvp=X0{B%GqVXj`q}YYop?a#
zw*K8qL+2~-!5bPgnPG;`X9)gCk(|Z|;=0F&^`m&18^)-44qg|lS=O9=riFF^#!!Oo
zJdL$-1%XtI2HwPG;_FmJ<UeCbHPHNY04!!$wp@H1GZ7+QDy99{bq?kBk}Hm~!;5%#
zooor&k9#dSj~6aRBG%gGGaG>el8`EoNZw<}?cQNAHQ#KDs{JfMA%@I56U<m9rBGsY
z3$djw4rCqdd8c}w`++r>AjEtyLfEQ^#6ino7Ko=)qx0lQH#kxU1|t?G9R^YbvJI9S
z?q7JoN!qw6*0jvOdU@ysgnM?OO0SU8WZN%0R^!Aj!*0-btySadiZ-|BbvxSC-~aJ8
zy2|xP^9`hS1ucjMzmFE596$u3p4NM`7^>!`P`phWj00o)yY&ps{b0tDT3Mk))3k|J
zTX{p<d0no7uVLq_uP?=Z{CIBm-Z84~!hbv1=Q?&XT<GsRzHjEk-P(;<)*wmB2+JT}
z=fP4^9Uje5Mn72Fc?c2Cf3vspCa}{gKk8wdkvi{7ZjB*$0e?agL=I!6uDIYse44Er
zpTEHLD<Jg#=aMb^+u8tTB>xw3eos5m|8}8eaa#apKI5X7?n<fZJbOGIFSOf1>NZ1V
z)b0ISY@yljhibJ=#lxDK%eE`GNrA8YJj>s59{7F3-!sIYkr~9~2(=k{r0vD4AUVbS
zh~8u?=Jm-u)ImSGQxIa-G<HKeEw@q088kQ;O3Gmv@KPrs`40T66EXzy0&*%Q{*5uk
z$u?WA1<ofx{W&G%7BahE(O){k-0#F@1+3xuLg)$UZe=%O`nCEkXq{)J*ts!-GmnS!
z8?~d>3(5R>D$6PY2*)_fY~_*8n_~)zonVevQzjjRyoT**$#2aka7%u7!pVRJ7jKPE
z`ee>*h2D2GNT0h4k?cm=QSxaXNxV;Z?PPhb8c)#_nP17&Azn3Ii^)-9k)&zoVyO(x
zfFd-j;$yGhh?CgR$0~n7Ui~D<5%okUE7IK@e5XLrR~?q#O3t-3=g&$cLr;^zoWC16
zm|6&AZfXa{Xu2M?h4To{Juv!7q-GDEd7el5+U%=a<a3tNkKmdvew_wok@zKzPW_zO
z_QaOKfL&Z146x&ML+!vpyzI`r4%)G1{&|X~V^Y{ey;ZCiz~C_rqhp0TD`ylJ4xXfg
zUwAQ7iTm<?NvzN%@SpaDgSh#YUbcYE-KN{d{Su$+o^t%DA0qz9N=hS!%i}TI;BTcb
zTEoS7HsG*JgTx^PUPCv<`iJ#hp#^%+_+9N(HPs;J${J2M+xh)VJRNT7X!dHZpw7<G
zWGK4$aNChEXFdBakTOI_d>wIQpcHycn!%d&%xN}UjMT4Pg;-3>Ai%o`$SQ0;y{v!i
zN)-?r@*oOUwJwEv4!YRY!mH}iDJzqwPMBCpg_cFTVD5~_4e$~R4s!0}9lr_pL}^t_
zd!~o&d*1u``^0XmX_ntM1+1BdHZ#e!^LN-r`#D`7{mPxa2p7b?ZqUZC&n$corRcCy
zT=uNmuNF{Zh^{yYwylq8M%_zk6(ty9yL394?@FUP842_^g3eEsRNQtr_{V$PUzSa2
ze~;*LN@4}Jme--CY;QtqKstnRHF;}?c5`UyQ<(X*n|9M5$;DKsp2xe7kYCaHAPDfc
ziDCBscazwQp6Z(MHtgZ3D4ONRfs4<CB_N_tb`|b0l}U@V)EIo~pzS1j8eSBnQD}E6
z|J0=H?BMD1JrtISwn5JZA$%Fwq`mRD_uKqYY??a$l$9!`@mJ{jER7#cj=6JfX8%?S
zf4`#prpp7w#7%GMU@$$lYT>4ZjjH+RclE%|X`((-=NP5kJzGj+=J*fFs{eUqkxmXZ
z?Yc#5uuDMUliW$Z?i7@xYm)-&blPQYa#4vyUwiQkk`?AVUX<6$M2kqkO7TVB0J74`
zm^tWnY}A&Rf_d~OJAT%I-uc`5r$F^LIB|}u>WmyA1>YxB1H-rd)nZhUHX9wEqDk+w
z1;`-Lmd8XA@|PliZ@TjRRjw%Ff={@X;)g)gQ=lUSKW)TgphJ1$4-@tRj>*|@V%(7Z
zR@qcXbs+h;_)};$=%mEi2y&pW{vG@sD`Rh4OXn}V1>(~*X4WwoEFxCbF<WMN&(0nL
z1Ai+NsUcrg+<sws22Q@X53+5oV?7`Y*|wtYXx;u$?k^KoM72*w#gbY3^0D@4hZ_XL
zx?KK8z{LZ+#|Jy@ek|zaC+=pq*v#uTXFo3pnc>0Bs<6N)-EtpAG%vRE#{xu4bZSXB
zn@&-+a;!Dt-8f@)PB_$-p$h-rRCZTtCyoUgk^8$!ni$~S*r3lspMTD2zu0wp{~;TT
z+CDg|$Zl&M`ER|W25%3gk|Epm**>y&szjbR$Qf^Jr@j_wVTHd-{T&ZQ{<f^&=We1#
zuV=ESXbZAwmrb{F-kPA=J{@G8F47Lt4Z4MXn%*qFK2_7uF_~~N_^Ly>5mb4%hZ<t%
zsB{@#{rve^->qq?o8))c-{GPOuGii=9^hTFN!GZ}hEj#V8jdcZ@aZ1J@iZ1G;zeBG
zAy|RGP!{%lQTNE(rmeEf*WNZimF78x(?_Uj&(PmWf%rgNyd1bZrkEN|+<TL8eUU(o
zARUP6asill_E_2}r|i-$Nce2fDVWJQF2AZbwOdw2^Z8a6Kg+dcpAW;&cyGWn72c|d
zTjqXG5o%iJ<Qi84YX5A5$=1^Y{tYghS+~0==p06sQWltFPwM?BB}DOmGv`1D2vQ8;
zu;~1DmXi>x%~Rj|=Z2@rf0Or=P7VsXj%=sZ4X?Is-Tx%1NfoHiGC0@k@7qYkX_U2N
z%ekswM|mAC`580+X;awkPZ8M<E7>`<7@J{ua&1Q$UrqN<&;-<fAy>xR-EM28*(Rs2
z&wI@vGhQxkSS9IB2T@X2L}7E@b%OoI<UfIv)FjlqnRA3iTip;Cx&=Qp9CUkfFi_df
zixtr*rI4^5V21|*wvaeI)IyE7yzd#`*ew@Bl$+gDbg1yvh@P@t4G_<MBfv}G_O+wK
z589sFCV-FG%D)v(-u!!WKHwu;BE^kdg5`^O7nV;nf$ifOlk{lY(?Yeo;L~ChrT}ED
zRFQvQ+`QJNdhzq%aO({p#Fa|@m`PBS7J1Lg)cAJWWJv(tg6EnC=WG2w00JfXgMxW6
zLkCaz18<yZ7Oufv$%_48i96Ve+CE`M4Z<EJX%({b*N}0+sVs~E)HiwE-Ncpb&L5#V
zKhb|AEC6|vwG#H&{>L=WgOG9xfct@G&<Y=g@)B$1?)!j!;}knb3u_&YH&jw$<M|KO
z!0BA2G-?0v4PG!GjMsI~f31i3x45dSW^zwE?5C>&$Ru$Bd2ou)KiSQL4d##o^9BZd
z?Ds)%xzv*L?i@?S$@>yrf$W5*`@(Mv@CjOFt4ZZ<%Z@T^Rjd=ki`r6V6yD6zt_*`^
ztF?DBDsV;jnaf#C>Ay|dg}~~epZ*IfarB7V&=HX>R;=OBF1Dd+BR3rhOxGyNQRn%8
zay0fJ**&Y=vlv8|TH6sT<y9hV|6~L*+t<U<5FzRzc{z$HOm(y$yqW$m!>Oc}RYUXj
zODDoZ#hZVY6$${T!W-r9Yj+8s22K+6?hQ*{nBNEGyB7}%yvs|y-_XV^_k5$iu;`PC
zA_E3hy+wUcyG0attS|*^R88>%gc$2)=7_P3U2PXHc?v@l@#l;O%%C9SKQF<$IyMxj
zJ}+%L7m)WP)%EO@#Em#S!42JANjR2~Le`N}=Gej(&A}jEkEkwqOmMRg+ba-nN77S<
zw;ra#)1ZR?T`^{%-A9xHN5G!U&Y+CorH}lVfz&>N>HN1hHV4R$CLzjAjdi>#+1-Mr
zJ-Fl8Z{D9LzeMwYv@Rj}($dXD<B*MhA@!cP=E)8G=+tSvG;dg-uTAhUJHA`oJAEbW
z7QloOCY~*fWIy#AE9%j9#V(4O_ujlk<=$RNHF2*toJKpChRd<B{DNQoW?1P=He2R^
zRU8J=?LY`d+-9pe-fuO5?6H+76~gbYVg}KF7!>Ge-)>dfkdN;Nu(rKcdjHuKFKt8-
z(L`k+!IHs`$y`rcnT%7XF5fYDw!hqZwM`+~C1NNiQm3(Q%ON8xnvA%Z-^9(5lXv{A
zLtj}=PbG_eSmlA~$EuJfIJ={Ep^fm!1S*Nxo6`faV4wBH%rktS|F8gpDC2ag;ds%|
zWtqsD<EJq;v;h`2N^Am3!9mjUjd3?sx6ruH+df9%hwKAtj++EY=bdyEC7XYg(Em>U
zOAN9)kFR}yVyQh;22&$z4HAlU1aaunkvwi+Ac$AiqW4}1sHtgWF<fpj1~}0cnt9+H
zgJ+n5s`;z#S`I7Z5Bv6P-J31<a%E9;{MFr#We}S~&%v>^I+Bn^QyCiDL52QK#fL$u
zRDcu0Tx*cK=zX)AWQzi?0RVbLUfZtQmaYh-vxJO5@X#R!wMQSRBB;wtey@K0V9>7q
z-KrEb5GWpK1y(x<PHPHFP!*!S>-3l_Jb9<0u7&&*B@e)Q?5iXlyderH%Oa3-nEcut
zA5!{ug7NzDltInoOU$2vFGX{$#jTwGJOX^Fw-QKkj}(AlKPr`}kYDziMTx_kzVwyv
zan(!}@>5ZtQcVn<kE3-p@&j89sv#a}C~LUsd`fh63m`wV4fI1?P`{9NDg%Rz0VrB7
zyU&Ro{f~{-KSEMV0l_-%iA)rm##Gd_{0srS<?#gc@^34X($|7+h+sct5O~<fT(?-v
zxhl5^SHoU5=Vwum1YGy@^P9AlMjI%%y-Gs$;Db>V>H<o?6>q?6h^jYP{PIxRU@f2D
zc-e*Oo`&GNEuh3uUF<A)AHW^I_aw`~P=f?N6q21Q{?RsLH%f;Amib?7sBnzbGJAMO
zxNApvTZn|bLHy$R-9FynhH%C26mKoXC3WrJ?MYP=V_J3q_c%vA<GUQ`f%T~h=8%Ui
zLq<0I?S>31)uRfe>F!Pbr9QMqibvySyEH{&HV0cPWTg&hArIq+<P1k2VOk#1!g6Jn
zjg{_%xxz2;xK|UN06kx1aSi>jQ`sBqmu-ASXtgVGqEo*xHpwQknfy-Ym~-AK^)Zyj
zb#Z_kW&*D(dsd)fxUBt4{37iu4E<zhbi8C^^nrPc->YXHXds3&ZIRDq^@RB(8=6KO
z$BJ!1V=+uG=8+6)|0c|<n>wmh&(UM$dfqG;%hm203D-za$iC;xS|B^ET>febC?C$0
zk)#^pUY9V;6j9CMJJY#b+}}U38XscaK40-x|Ith|jpMFRrV*$s`*lu4AkB}a=R*#}
zXzKIxcT#!#<uVV3{g2IzvMy*cdVCyzm<y{URU-^H?eMkv<w^p%i3kFiP7iw6Vvxk7
z&cvv%xoO_Kz4M_#6I#&$#}w;yhCN#OO)HG>AuXXEFR($Pg+GfwJiL%X*hDFQhxGVH
zd@|K}4=xir%r$Cz9v&XrN}F%Gz{;hb?*r!#e8_0-1%){>euuom<fJZ|@5)8e)Aj)w
zlau+`-XeIj1On-pSN^in;|3C{Rf(Ltr^<${(no_#dub(yck!yCTn2b?-(Q<kV~>ZM
zN_<J*UT;M`O7ur^sxTDxL?Qo5pr&a1K(bBqGlG_v$^nF(phNwJN(VltuY<)>%gv-{
zVQ+(On)9!*8|d+FLak$D3c0Z4WP#oZLVaVFTwL<0x+lC#jIgP|Ry~E1`4%8a69ZyL
zkPB1@eC?c|b3QlWU}gt121iA@oOXdMiz>&($;eqC;|OSE8WE9>Y*bkBzS@_HuE4@%
z{w^UOj2rq>HPqp`N@z#AQiQk_O%L;s-woJ$=E&~S#lJU<=xZB0SzTg2N`ks}z6;Bj
z)qZ&)hKVbpmUp?}kDK*rYVdIDR-;HuLN1=Q!XvSXgPuS=zV(U)cjV!y!#!A<u<x^Q
zQl5`lsz}_$5B0D_&*M>zp~gy(ONPFv_ZsK=OYdUi3nr#$@-I_L2o47R&;VZeCkGmN
zF&zjw1K>HwwBgXDi$qW8_=7J9Rw+090*%a!#8Qf_CA<jZ(!|)PFq^>4n{;)zT}+KN
zZ<n}=hbQ)*4uawMeX+x8FWyo}pMmnHa2t-iLTo9`F7BX*tou;)Jct+jlw~;wN|Xy?
zdG^}p#`E_i;WFVmwP{u*2pk#`$HH1x@u_2grFX$xbsYLQPll;4iV8v%jUEJRn}~YQ
zqHSAb;e@&dtn5p(ogVWV@M!Wj=Ib>Hf>GM5f@;+nygY)-JnGfZ|B{K@)4t7D5w4%^
z&(_fEqY7r9<(4PqzqBf4kBd@zm(w+$SOW<fkda5DLlIBzEe1NaJPce3fEJ7x>KxzM
zm`aV#Ugd9)oS34)?$=Fp9Q^XpC?dHjp7Krx*p4c6^<vclH=3{r@)+Hh3owc9v?^nn
zv*fP=!$n&5n$vE3yAyb4Tfgu*r1EiY0Tm+Kp=^VBhTMEmSg$iixw-D6$y;I=Z^CVi
z;^SbKbc{}n(>ZNvfMSYWp2=0B_c~Tfb=@smL)sgiqj)a<`MQ>LF5|dVgbm1wqwUsb
z>J^UEy%8#*Gpw=y-i`TBd%Nrq|A2`8b4H?_NY33%7JvHB=M`LPj#>1dy-fIMEq|%t
zZj!)}x5H}=1zYTI7uu->jv*}OnPs0%11ZgG3pY*X?{+c0(-mpmUkZCK5;uhfQqKS{
zT&~0yibuUHO&$;c%@W@Lmor_%Q<Q#UmxOy+2SJ>P+WEWoA_(h-SZuqtbL>O4H-1uD
z@%qo+(K|j!l7<UR4Fkob$0ygu3>~n4%aZYRRTx*7WxMcE`bOCH5bE5##3b(Va=%dE
ziCV*_=V~ANhP;cC=Al*vg?23)7G*qyY)23#zpQwE(XXE`6<4WO0xKcvQ~2%q0a7oq
zd~09Z&QvQiZMC40rgZs$op`zXv4#0yLAi{QI=Fi>@cTA9$fREtLNiMO&BU#S6@Zw9
z9!~qE(k7k!Z3Pc>QGA)2urzPz5FlH@{;P>XIU1<?h)G_3cw+G8J|Z$mSdxoA{Keku
zdQ3i|u8w<9;zp3@nHrq8W1H)Nb1IsbF#Kg=U2Btn_6_WDK?izNWlimwac4j8CVZ;K
zA@^Lf%s9k+<vKlx;)|=)Qx7s^a6v6bM#Jg$e!$HR@U-T|*9eP@22Cl&g`iP+i%+wX
z4p&1Qz)tYwg{;pedi})cy|E}N<Q2&7gDfH7vR}ri1*tn5vMt7EzhOCl^QD`68?=82
zPGXXcqU#bZ%vbu&_qLr1xZYQssgWjd2bVsEjn6nwTcca{KxJDW+huTndp7hm-*N((
zexx#Uskj3c?{x6JS){=O1G4?A!vO*qmc(Umq^p{!O0GMmLB-bWJ=8Euw_P`Lgpqs-
zCQj&g4*C|<19(OfzuQX#Q--itj&=OiGJG8uh=k`3id!b*NXq|m14cpK+e^eDy5gl0
zpe~z$B@i%rDZ0-H#j5O9@~y-6ACQBiHC=xQHa$Wsy-emasC1>10XS4<n8?1donfr;
zK~~3aG}~_{OwkzVxvy8z$BN#azC-Q~A`MjJ9vSwtX;Qxch?~imPMU9Q$bMLi&O|yz
z)6FOMx6$6QdgRA{YFKqB0d?A@>&wTkrhQCCAreLiJ1#r@`vQBylL>0_MUZ8MNkcz_
z2zFGusV~Vd@T3KjgUkoLm;}!7ckSq1uls>HJ+S(2pJ|LHBS6wgOW%h3hg+V$pc1Nw
z)c1RBhIsWJ5G#*}=`FAA66|6$gIm?JL119a>&L;KfV#*ku&Pn3uOqkGcy-Vd%rcmd
z`qoZ!7^8{P=C+?KuL5xe^L0<nXo2(mwg5#xeKg#$AIAM9A(HYts82t7TQV%_$=l-p
zV8+lC$wB*j;OwIE_c>(Cai;%hl*ef1ihY0l__6#x3%p^YFzzzeJ9h8kG=0xhuk=!2
zm+-27w!&yfcYleQ5GJ8Mqw?yN#RK-l68vn5sYm)AD?laF6|F<%jrd2Sop%qb0$KlW
ztN|oi4|%l7IA%>1P{ch-;<Q}la^H3<@J4Xrlwmwl#wfHgOtL{Jp8T`{w@4p_!uqb+
zg)`>hppy8K{v`g`rU_1{q5l1ua5~w^<E=Hy`)VFp_<y?%CH3g}D3`HvCDaNo37w?W
ze2mQp&m|n1Hc-dzPmFRvG6ccJd>u$kW=cIFp>L}zI1&2pG0#X2)$g7;S|2u<Te<qs
zo1nZSbaMof?cWbFT4t&(6MH;=d7rIg#%78Q-rfA^HN3akRg&(kvUBZ-RP?WOXBQ^v
zN<poQ@#%La0W%{i9BilRs$`gHUPmd?H^Q4cvBM?(JpS5ciK?z+C+LFS!3Ms!1PWQq
z2d8MxuulmyxPA^<Mh`}R@piEO;YCzOoTR9hn<MoiXQc<#h1mtHCfHY~SuQzkyrsJ5
z6ec`z{_7+E2NM`c9<k^qOTC#-rh^&RD9r-0C<>+4h?vl#os)@A!*rp1oTm8){X_2y
zJU)(<;X+M|#TbF`iBz{m6o2W9V1`fI@6W2<pz&hWBa^<yuB3Auz&O)VB+>U8EwA03
zZh`qkhJ{EvGup9#qs-)ULe;yp@mkLKitjj-K(=X`mfxvLY?<cmZq@NtiVP207n|?!
z%k5XKfU2Hasnpi;@ZDn=g%T|dl}^mlkoO5m5pxLGZ?(!dfM{-LHZy0Uyno9s^sxD7
z5v*Zda-;ShpZfjQYFv-r$tq4S2;nXX%q^>B{FN7iEM>{E_$k~JP?qpd)mrQ){z%>x
z6|dj(ZrmViYTTC8w^TMIu08iyqqqrustgRGWmgS({@6z1#s%Ny2%j=tZ-2;WI5KX>
zMTGGJB|V|k=-h;%`s#D7Q-^3}dPU*-s4u6+=FXc$);gb{3Uv3ztY-`Hjq(b;Y5>my
zP_v88g6i`shXWC0q7I0^WP+a6nkT%vfBrt44(LhNvt~m{^5vGPs^J}L#p%8AhDQIc
z468(sw?WxCL=<2PUm^CEM$a|lbeH2&a5wuKJWk`tqrY5bF+5hnbQL2<&jt!!`#hJL
z+X7$*!BZ1pA@U6sDySp~b)aSxBT}|dI$sQ;l;3PDi!pZ6a!&2VH58kyoGWuM@rG`6
zlMTH#Y{NuxjZVmVfH*wE(7uMUcav-#=yoMzdCD|!7$P^Y-Rup{_7$-4v?^|o{v>Ka
zEAG$V(GZ&E9KA01q8&dc&ilFY1K5DdQhB~UZG0yPg1=Pr0}+Yb;}ozC-ZFbXy3{t-
z$LToB4zr&a5CZ`fgKrsr;gh*5!f$AbkmZFI!!vPM;$<)IQp{Hzoc^67vsmOX+FBH^
z(+jR8$w0fto)-Sx)f1rj---5;h9B#&=WCRt9!WH5gGuAwwB}g0$s+Hl*r-D(<-mqC
z?$lRR=R1kJxsy(u6vIiG<wWG2PYcitnbaN+z8d%&Jd5_R-9~vQ(82Wy8CRh+0yU7E
z|E>paKqr<uED(Nq^2~5^QFLPzEAHDYgjZ3J=XqHD-;mRE8G-h=9f>sP`TQ>_owdU4
zQUdDq<lrp&y=hzV^YaZjnMdK}jUxGDN*Q!Cus4KU7_S4^?<LDKqLah4W`J7)U<N13
zpi<So@1>R=j_R&&tPccT)`h4}sUG=1>Pvj*Y%6rES~}rZv?atPlqA5X(fj7}E^l!t
zoocgoo%2?=SUiqk2m00Xy_0E^(*Srq7F#d=A?Eu|6RQa}draEn7uf1LS9UB~<zF9V
zmgyFhnPIlvE3`JmZ;gz!5S5{WsbynLGa4R@M!u#Zi$x-0SsC9anbIz)POlg~jP-=5
z6`}{6RPSq>ltM%`N7ZA}4Y;p0IL37_LVI6g+>c39q*Ix#v`%uR`r$^iU`l->RA{z|
z|H-dfdDZb4;ug!(H1#4Fl^g&mJG(w<<!a%rvYF*AZJ03vSR9Ra!ViC{0}=oMbKQ!Q
zGEv+XakqX?97~y44x#Y6A11$nbp?>i4>3n|8rw7NfaU#Bg<$@aug-1fT%xaxa<Xl`
zGf|w_p5#+;#g(T|MOv^34I0?OmI5Ib!OEFSQLY_)`;7$QFEAg=esgX8ME%A0aq#o5
z@Vp;+KQd6Lvbw|Dxf3DoX&p6(<6g&s`eXiIeD8F0I!o^I`Q|S+Iw=V&I~B*yYnq==
zmMbhOs0#oWi#4I4#IipW21}mB;e#|s7xyK0I0s8y5s%|xz{J-pi#iRnfxU04EdHbv
zOf9+NGX7{ZyR%IZz{nGwNu?QD9H++sXxnT?lHwe7a(p)zGzuq1*&YSD%-9G0N@-9)
z#w#Ye+E}CgYZ?GCms6<+Sn3A`jDmQ%1k=m>?}VVFx$EwM3v(r#PC&XJq406jrB?!f
za?X1b!gGVhK@m^@)CDC*RQO9~RwlSTH<Qi@M(Z#Opy@N6YUKMa1V0I=xI_eb)AQZ}
zEgA>w)J~%QF_CrqHoLBItHD+WS~3uW&THy_au>ODihng(|3hzl)?Zas)vRb^W9f}f
z638PeW_4u|!rZeC^|2p#X*0vk2}UEPIKh{8a9RH|AtLKC_0@N024xA!W~vDU|6rUO
z&SWyL`;Ac|gONOM;H-L1qiP!<yZ$4rT<~a6?6>$Y&aJy7^<$5%c-5U!qk%~8Q|44N
z&zh^q8cOFRci`@6kM2o~ICh=k(`R_Wi~syS{tI!m>&y7ymM<W3y@57=iA@`(%#eJK
z;~zZ`?=IqF(5emcL-?(=DmKQfWB6N-`dbwww#G{k4VN}v6`RGL_pJ=YZ`&Z%Hq4eO
z#A~GnMaU}=G%hru8#CCSMg9|0F84)VE{7bi9TBYh#exC$waWS#)|}S&d@=p>3++F9
z!b!KU4e1Z#tY#>E$dnJby?!PMTvly*4hV#fT<+#p(zJEQHnsbAxlD&UOc#AUr&C3L
z{*jVz%v<P<o&JZ*<hS>YK0J_nUSmRhpIP!~#jJMnC4~2R+aM1jX@sZ#>!|gwBCMo{
zwDpyGMm;1VF`~Uj{LrF1e%tde^X&F(CpIE$^9!%uqnAA+#VRtC>fiRzc%GSBG?DW<
zO4EO@Hc(g+NDk&t2>5DGqe$iZo5bOza`sTFy!!RkM&@A2Ys)ctKsjMb%s8WfEyYoZ
zea-{UWqz9(*K;i9Z)eRF*3)B?x|b71kxuY))PY%9|4664r77a}csN~WvkjHA**#Ex
z{iWyIxYg->MDN!yISt2Xd99Go0vmamWIahc4@sd_qW>JbsDcogB%p;qss2rEV5L7i
z;XQ3##pwBR+io_SFID8nLlGYF-m^$t^x9A&HfMCyBz{%SbS3hIDTq?v0Asa?PyvfX
zf-#;F(Da!qivg$B(~D-H<j_j2%!nZ6c+f_7FMZn|SMVR?6p&l;dRXgvmnk>>DldRk
zxy_wto~lc5F@S*o#s+uz`0;%b!9?e3UH_3veHKD$eownTsgi4nP*}+E!-gNM!(sCK
znch^*E_{f^S%(3E*}48*VEJ;p0#>QIQCw>8l-7RioQGK@4JY5%o|ZScE{CnrhGMi;
z@3@Q1!rxcFEeu~L%GS(~JA)SY0kf7HYI!P*;cMsFB^aQ(_XqwX<zr}bNMMe5_DYi6
zbFam|y^A<zwLfLk6UmNUHvq~+DRPHij*aT&h<Dz7yUQy)Da-2-F6&t-|7sS(Jd(_A
zq1!Zjq}F)XrIa}9o*3`mk;Ly@(UX|e!Xh*GjW>m0N`5Ziyc}&lH)@*t32L7k^3n>U
z(Lz<xCcIsa{<&{t#T1OR6KQl;N4qLAC>G+05hod!T7=ZD=hT*2%Gob5yfW^96zWk%
zv3#!TLd7yodcD~HTijS10Us-3#!7u*C!I_pHifCGk@Qgfcw@5vF9*JK^1TcGq|6Fa
z(iA)LB+efGyx%0w74zQ7kpVv?Ta%_;(dNp^dcUk|Rf=oB(Zfyip@@2TyO=NlV0lDh
zz)+ag5m~}k*F&fG|Il=nQBl5M7ghu*rAs;l6o&4S8bVT~hHjDW?jBk|N-064aX`8|
zrMr>t4(a#i_h0WPKe!eyS<f?b&OUoz8|#ngTwLx;{f&?+d(I7a8H3|hQjXeBtP_-;
z1YwU~MEib}5XK`wA~=*&OosY0l)_l6qM>aZIkv}x>c;Z)w*anh#_ah*zU$@0;cu(_
z`XHXLi)Bqu1tQf(bhmgG(bjLkGugeD6bPf91g6xrSQ$#t>HJ_e`ykaOzbr%LJ$BjL
z%Yhu*bUlASF0PtI>!b25gm7OqXtEM1Q@81Qzb>8nr?4B+)8@7>8co-a$`#R!ooE$Q
zD6QNAUcQM~$KS)$c9QH~alxUS4pK`<NYLubfXzJ;2i2OrQ>1JICegWZZ?2f^E`jnj
z2L8F#(-*bpu?lZ)&~{{I1DrKrYxRFa$l`kOh}^}6L$GJ?p(dM89%iP+qBF*X@4{Hm
z3||(Z45V=7s=qcHOfRNZDq}GM{Sz7lM48nTML3<>KSC|N!B^?N%I_<ogq`uhXRI(p
z^VY$~fz*7&TCI8Pw+!<u!b$KS--0kXMLadcy#Vo?5r!xBzh-ID^V<HwQ52G&bO@rk
zIoFXkZ*PLjhjF0bA`PjcS$d#LOe5-9z)mZ-?b*xS-(4@vQon~_QEZof5&~T&GW=HB
zrx)O56-+Fj*OW7%Q8*<(eZJ(^$d{rkVW*p|cL+Ei@&rW&*#{uQJc^HpEi*c}ey_Su
z!IdTovGSMFE%z;#D(^GdrWwC6WG(nj{;Y5IXPQQwA~U?+xbKlaYY7c3Oor<7+wVSB
zwKqyg;IY1xUMe5m&TI6RA4||sF92R6wjpQJ#D1}{!@uLSNF|9<N6sexX7gY45If61
z^(~MOy^O~4ES%5A6C4KP+{oRcKoN4j;M$jWO&{dzo1dldAi0$ke9_a{BJb5wN`ChS
ze@J-0Kut%>vm}_VT$G84)}jT;+c!X>!dCxf-AvZtcRK$+cz({apdWQv(<;PsNIZpb
zR2*UE*o5I`O%vd#z_=u`3^(G)Qd8!!v0}^3CqQg6k1lJmj#pbqwNkzO`~)dmdb;=k
zG%$o^n|6y&B)>#|DMkI8&iI4oTUK^Gqe{R;s)>L0T@gvoiYA;m-^L8@ytd7ByJ~CA
z>)nazO6F|YEYTZyR~gNcqSP}hyn#!Wg;^lgu7&)Ht+XhcXK)-&KjfTIRcb=1*Xq_h
zM>d9q29oe4xP<MtZsY4MVUzO@m0DonM&#>9u;^9JaF1}Kk19G-d>hR48YKGVVT6IC
zZMCDrRA0I1<Eg7EaKRN8^x$}6Lc37Fnv5v%mc)5>1n(E0S|*C0ZZ%s@W<B`ngkZS=
zHXibbhDjaAC}V%J#JK?K1*!B~3vCdEYA{CURO|D`!E50>=pfeLwNexTN>3MQX&Ccu
zJR>_8A@<(lB+jPxXt@aQWWu=;?gkQ*RHNw~8-C(7ZIPp~8|sp_O2eLpZt^HB>wdvh
zS9gwJLd2u5d)S0tg}f4;_?xo>mKw^P93zy5;U_R^Rkbzm%pfUjoa?&zw~T44=W?4D
zpwM<aj4?eAT&aH8)T|N7_<dzQV&!}mn_U8_UIb0?<rNCGmiTo7V$g4qfh438&eoW=
z%N-Xzj2ZBp$|j+R#|RR-Y26RY9S2=UI&=6b$5B$a4+9q$pC9}h{A4@iyJ?&QTIHXf
z&ZD=5`Q*I!ss~|K_E+}LB(_GCg!-rRS3#c3Y5(u5g|k4&%EKgMf+V!3`qF65gY#}O
zUY9h^^L?ZH?+@4_s`SQ2!L{czaxb!VZ2K@@SP}Kz9d+?=63-P4i0}y<+2krom7(Ro
z>8R5zVrGy+dbz~*xW8~;3%%Qts#ciV&tB=I+uX9p<Gj2nW{UK)zAUvO&|OvNBL4Nq
z0v=;2$sbvpH`dXLzCTRmJ#X~AaHB8X)DkqF0a8Q9_SvKg@j<}UOvS=6*(#++^C`W*
zAI=nlhPD>}ns%0JU<t}dJ>Fr}PJ=U(LRysb@5~5enJz6*7RFej+2qr}>NatM0QM{t
z9{t+-_4tf|4`jWeNxeP-@!fMoP5EgrE%wHUs;-lJwm)ypIAl<*FXBw5w3{>BtM-x&
z9_usN|Mb8@X7U9!?G$~QI9q#w2A?mTGh{1r*!*N5Vhxv08n$Bit#bK@{24RbeD%(2
z9V}r%m)%#hmZ!%*p0&&Y9d!RF#$yvI!|wtz%+i4*_ZuEMGMx&0uXslvW{o0?w%Wgd
z=T>f1HFa2d02*QH0QZ)e{Cf<G4pY~S6=<VBSHrOvxjAfON-O?NvkZzyQt7<vfx<Lu
zl!OuQC4x%i=<J9CM^-fEg6auI6uggUJc0$Uc|>VMn?y^(XMUS0%amjE-7r*!wTHbS
zqE*ptpm%7bGF0OKi>%DO`2-ji_K!k8jW!JZ29S^fFl$t_f*@42Ycep$7Tf*vN1?Jo
zHM9q<zVizci(DW7*mV|4f~TUTz)A3|Y*Notz(}pBJ(_`|UQq|za$wVT9&343mFQ!p
z0u1Vp!jGeKZ?V@);c3*AtN3jlUT!!(`I0~p^795W_~<#N9%2);cKlMXWW+ti%(I2H
zC{LiA>4{7`yu_!RQ~-(lF<=fjD0fUlS%$8(Wl7H+@PM%JunI{uvaBC6MnE&J%$Gd(
zlkEtec|lrX*^=(EpSrgzRB8okg&5OY-#@sx9g$+cWt9$53VOtUe|PasEr#;jSh;x*
z2Vf-NO&jJl!dRZ|6_!JBI26RthmdT=m>Ghm!;NJm(kX%v@BWm9HvdUE=O~fEjy9$I
zblQ#PABMkYR9jXS7x{yX%v|{K!A8dHucWDJhs)bGY|B-Vd0PSgyMK6H$2PZDF8;^1
zW}8ibiM{vF++Slbys?FW-A1Q^B!+9F)NzA$x*~ohBLyc0IZW(R&bp7*k82Q=uW&Z<
zpn((T3!?ThPa_(Q@cL4YgITg~3F5bVu|#1;EKoBhSK6#xsQXrtQ-{!`iTUK%!b*pk
zkc|h&HWA0vGTIJ@&I!)-j$ULyL=UHV*%}Xr5-+7_8y^_9=k2>FqANsTi=KajiOFf`
zAT|s;2KHE4SmnxA%0-S=qk>EyMUOduzdb{n*#!I7fdIuU9<aXjQ2Lfz!e{38{hnai
zK`v@;=bIE)1X~SRmvX7c(_Btn>)jQ{T8#@5MGZSD!FIac`~>^Up6qo2FfH=GmDCAl
z$q3wiCyL05z-~9i34G8;qfJ{(PM%|CJY;Sq?$Kk-dAE6|g+pEouQ%d&Mo|b=s#l6$
z?{djRx(|t?jvsf$a2R)fD$e?f9J-HYHObuBXT{$HO$=Kl4Cp%<5E``bpCiDY;X|&g
zJ~wP3nDiMtz7@T{R30^daS`gQ4i}gY(0~zykg!fDe&v3Bo_nIpI03#G=~8#5S<cgh
z-OJ40!<9hLs=2aqZ2<X-1}350QTuv32fTe1+c_W}u7n3GS_+kPN4zrz9D<6hK$MHi
zlMP#nCq~#`iCU8(D_!Z9SJ57C01O}v8xj{V3&<5k068dPEou<XxE&*K2+}D8>uo=S
zqlOW|hg2u-@UQlJKbF~n3#y@xPOO6`$OmcianXg*Ktd$uc7q3}sVoCzJq@9d*83Yx
zhuCw!CU+FoAQie8`R4IHcutfV$ezhR%pfS@kJ$d$))lL!n~ujy<!PI+8WO**{tg#;
zboR?vEx(KES6pKqRM%New_@;(SN`Vum;HnBLdzfN1!M8b>OBF|t46b?#5v$2?n6!0
zvpweWRg6DF0(4@rHUmS%s;x&eFHq_)hsi(omto?W`}2uz3p~qt#a1;Kb$x*y;5W%^
z@SJwGYVmHb_RZxgp1ZV4!S;lra(#ruHsmmT)jt|pLdIZK_P{2zG?;`Ne&JXu{7+;G
zSz=a;#>vkBtF^`{?JbF!$K})aegiEtlg2S3KZU*!imEi)G@Pk-X?;}ZeD4>QjQsVd
zO9Dog>N01AeiKW(_^JUO%0U2Qn*07?7JiYlHf8PyTLvAUPE$ELHD9zh0B=jhk2Y=4
zfitlMSQmlT)A7s?J>g@eV>z=%0@~K3nw&>iD&H8`G**uPG9u$9>AIviu+au=0<p#S
z6o%vlKC8=#)bI%@#Dyc~pE^bGQ-}Lkbm`RQt<8h3@Xr3HkSdxc0F<nvn+tTqo~`{h
zldru?#OF_rDpqsokMbVg@=o-@2|S9;V4Q%9^a4H$equU#(yN;uNw*!M!z@*tbnqdI
zpxkqD24?K3<!p1xXnsRnXTa;SlcAbf1j3>nG|T3DVTstQs_U7qvzU2RC)T9NH&?G=
zS~A`pf`7~r2CpZ!ig+Y<*IhmlgU#UAR}oK_Vo97sJ<yo0eM`L%3LoEdI-zk*-aG)w
zK<(#X)G~g-?;(6HT9lQghxml#&9RM46^}Oe%cCU~7s;4UdG-0^PwnQYYj22!oe+GP
z=IMDGkL|+Ti&>7wl{e_#uBX4Prptoj#Q^V-YrU~r^p#2I<p<hjT?nW2wWM|s&-Evv
zpiKQP_5_ztS~rn3nCluoyW1wUw%smdex|mkFWl6L2LF;eT=eH*jt+?CBVv+JDH$K*
zQTuz*3{_f}8b7ui`C;LFxt3tj60jQdh;u&n5^G&}ybMtROaD6=D<w@|#wBGq$T(t6
zRj59mcfDWu0o2bXKy2jSRH`Vwj=d1hDQkPWL*wot+@_3F>t331#8uKtGnDDyQ1rIG
zwN$1(TY%o+cQf;^7uB#=xfLq>VZN89QCihQ!i+ZK1+p%t{Wq-@^DfMI-wQcem#=Od
z9Vr7<wQZYj3rVip@o(`Dvb+|34_HjIrBkmx@`o6^k%tx)(r><X&!=g)Q_vjz8=1tf
zU0Jim>n89Y!f<<x93UWCc?|CHX^Fpjkbb?j;sk^mYvJkRXBLfRYIiz0#vZT#HQ21j
z8!@Q3zy5?5_b-Kw0QG1~Z_}_)WfZfSfCKrQ8|E^C?j{N%lxCLZqMZAcs_QbTDdBWX
zCUsy^NG0yZX%$0z+=8tKnMD@mcLBLKfkgay>{p7`q(*XHr0t#<2~dBqItp2DvTG*r
z9v#<VPu08se&0^Tpf`iz>fTB_2Oze~rFT!C-h;`SoOz@)7Yr;og<=K?d2LfAmNChW
zVs5d_sP<@;WtCyBG_?MvMppMWq|Az!q10L%R=1(7UaU9NPsxPc5GUe3hH7};)a?p5
zQ)=q!TE9``lK{p+ZMtm6q90f_vgW_DQs+fG<D5N6ZpVj8=?k|_*MP(-f+SBKAf+CE
z*11_Te2?=ZvG=Z~_NwFTD~Lq@aixyt)s1*R%oP*~GDy{ChLT^B_C>Eg#IM0rJZjxW
zjwBoBkrn_bsCz3k+n5xZmb}s-_~jwK7IijrMUm(Z{&s6Lm-!3viD@<10lY2$Z4yIN
zA0o2ZqQ&>aW)dWY-{M!GaDqYpm1kx?!6+I(ucdWq4Qy+P-{$Bc;8KiB9a+Rxe#8+i
zjhE|W%9RhLn)*b?wRD~QqF{a42sF+!TYVg0{l|EPB`63*e~kVrTEaO-_!v$q>Qi6(
z4{;cNB{|zEBFTwarjkM?_9Z%IfR@Dt8?HCm`Ha!7*~$B*U4gAQJ5^i#pO*_A-s9%E
zY3pcS!;JKTl~P>E7}uLMH_8tm_rVgG`b~J#Q8|t9H-%&Ue73P0K|M)cSEr=Y96vq~
z5x=g<D9R3bmY+RS-cw&qbD8?|A~NPpfSqduWW9(toX4srbUqQ43ZiC~L9spgv6A)}
z_6WUQIQ?+7q=1QhnO3=T3zr}1s8Z^Fi?OnzSxqbM#nt5od>6~4q>EKI+@wk`{3Ncu
zcj-0GD-Z<O!qL^#bZK|$=#Vmk<j8uTr`*+|1^QK$anfF(@@!8TN0|ET#H`JKN>EpI
zs}b8&mO(P4WjBIW)Z=q(RNF2Zxhr>jT5!=&qLcc+t5UZL+zB1k6q2YPF4mbhPu6p3
z-hNqxyw;m?oM35aU*cxMtQ-FdJrP}zZ78l^o5_0ql+yvU@AWX)KEDV43d@*pGuga$
z-6Ej*gUT9R!K&P!CEblC;+#O5n6SK=X$?F)n^q}aK(H>@O#D;JDK49XUnch1!lTD_
zyUY2AaoMauXw*4DHgy`yxVLc8`_a@6vLu3%jF7>!ZZ|~%yOk%+b1Nsw!d=qt6bz=5
z$8Ff~HU-QeVgHV{v7~hmtI!-+6*bFOU-%cpeuAuzXuS$6CfxP=O>aM?hr<+zeJgLM
zll{fEfc49@T_48%lsP_>+yTykiR1SMY!sa<!k<YV!>)vZ$x+0AxRSa%!Z{B=jZ62S
z7I8QWo`~{L2k!<8u7s=&fH27gcA{(%vRlITZP2Ob+J4=0>I@%pnqPrcmVn*KCbXPU
z@oV`w*@WXfR-&h(dN56*qYy+m0M55<&@;b>1?UCF-yA}NU(J_KVH)Yem1UyIx6oAL
zaSv!I&@P--grjplu;JKcS=peL7F}z(gA><$OX^~+r8XJ6;vl#_3Px~Hvxc})kkbS}
z?~NQDdOf$^EnG|b`L0Cqk@kg2iR7)2*;I=c-R&BuPJW|hKf^tm`EE`bfWqrjt-3#c
zE6c|l4PZt+5i|6qdfH8r&5`a*GaN**DeH;a%68r@XSR61x{MFP!$sfvA<I%9po6vf
z2Ww+>iB@TrKiq96+%Rw~{MpfguB#d#i1FsX@#f7pY%BP|cHU^9%yNDsR9GvR1!_R#
zh;VE^08GMLB3(wdg^ttx08BX(VQaA?n>(J^=A`ZkjUgtBr9w9D7hlc_)EUV}hJsU@
zcc*#Oaa9J>*UetgunGRAVn<qAb17EP)I6Ut>P!4Me5Ii;W#anNV6b%2w@G>*HGDjR
zZ32IVX1(mk2y%NRG6j0j*GK8{SX6qZENz+MsyW(^j|`U3Ol*Vxhz-Y2kBlS=7cp(=
zfXJcq!T@hx?GrcLI=jCT^WTg98G5RxwMD#c3i?s|4|CqOYTjiue9S-%QvxcQdTr`L
zYY&p7DrC>>Q}kLbmEkMqwLUHY?80xhK4>_vKKDwO>Fj4VQi~}oG3w^=mjWt?Tx;2C
zmQ+Ng7T;(`@LuRVmeCS$1J*R8Ei%x2=d~kO)~UlJf((UVsMPVEJ>E=W*SlQ9D8}|%
zlR)#B!&)wZS(f7%Z$Y0Ugp(t=qV^c}d)m{KU}#cKSOE>L3+^&{ZfV6rsh=|xUoCI^
zOL+Iu3kw<dIb`*FOdUAVM3(Mci5VMEvg5Jq+TtX2)q+93{m!XT^3WON|2=Q4Da9M~
ztlgE?Nh-flf#loy7XX9H2Z>A4;voxV2lrOepFO5KGhIZ-uOen!kSxa(Q#k3boOkXU
zy{~O<_wSHOnTU9+bya1fOC=dGLI_Ww<Ws_644=sarv(S1chkqYA{?Du_-96VU3J85
z51TWGz~$Kvcz=I^RekelupG6BBq<@`HN-&Nb7=ha@gqatk^~7Spkk{0Ja-sZ)9C37
zA&!!`fV!nDM^B0_2@VM~t$*;DKiN7?&dV^@3n_i3-r*d9j~7Ko>ocE9w9#dp@rm@A
z%>NX^N9tHW;MJfxNYI#wb(gfD|8f!QZ~&ug9g)){(Wb!+<1N(CFOJg{4PfDxEiUiL
zsQC0g>S}h8$+Rk7v1}BY^Odv1$OClH-(3^;A?<o>i{9rmVs73LO2gvTXz-@w4C?m&
z?Af$E)jQA@Ka$r%u3TK08^AfLS}ZHhLE(SIIQZ3+3f1$pJ9>QdQKkWrgX|;BLJ(Sk
zECmNo(nWXcX_o^Ze!T5?fx(40%;dc3ejJ}7ya8J|r}y=eVle3-I^N^C`xJ;3T?Xu>
zYwzpn9ovBR5O;Jo|97DJ7l&^txDC9P6+EBt4kZXoJq$NC`p95~?R|-WoImpqfV);0
zENy%+-o{euu}eZwnH`*Zr&(+Cn|!ngepco@w>oT3Ol7bX3{fFJc!I^>Jk2^u@i7oG
zo6cjII4iWaDA@9tHjutd!XBi9GAr{8jAv|AJp)?s+1dh^sb$4Hy@2>U_Fz4AsH6_?
zGw3yGm@?_<eVI5BX8xg{yi#<6#iMy*6C?$dd1i_-P)mOgkWz2k0b5w(TJF%*Bbe&A
z=3U^c+WLAgW(^%Wn@|)g-~o5_)b{<>>4J8xXoO)I^feIi52J7rc9=bDG~<nN#RGS8
z-Ulo_H5`qc6@&3<Ro-}C!7X}E)*_wO-n^3N_^_RD=8|{5SECu_xye^z)&?G==*bZ8
z)&etCr=Xg1Kpz&d7Ym9mWqM&&t9M&3+ANGUJ8I1Fw9#zz?*j4#y`U@juwssvRU+2g
zFe~$kK}^@5J4;kK0aeC!tY1G1N$->qp|kB0hT+!oe8SRsADnh3FyfpI6@60jk`{0G
zU1tR3iu{Nt`;+v2wt5oZCLf^Z9W~!Fue(3Q+1EcU>)}8}0akqmAF-*FWNW=7d=gpQ
zG?ldnI8@uwLi_KMC3993tU2`V?Es(9JpOD&j{?{DNP=}9l<E?=1dh~Lmefc)Z;eQ~
zzA_ZWcZK4_c(;bJQ=@-=gueRvF<czTSgIf<q_6AnmSP;-&`>&@j+*X^6Qy1x0EG!d
zlcZKNicEtZGc1rrP`07{ifghs0zOnJcedZ~mdM%qgX@B3j)4dZA?hr?vhe$DxU1*+
zYQ(%M$uYSv+7#rlw~$5<?=bigQPNcB_g_s)#cXHvj*;!eNL*%}_=h1{c*~e{(sAxc
zpF)Yqo@Y_KQm+hV_SqO>Sd!9WR!6Qo5c;=-H~cc^Lb$(Om2pY&WH{~h>#KM%X*A)p
zFZ1xuX8&r`!9Qp5u}B;r@Nkg6rx&-(=+?45J8V_0nt!EN{<Y|nc)KZHuJGk*d?R;J
zhVPvws2X~Y(JBTjvI5Hnh_E)0|BZReK2qaoW;%m6ZW2pB{YPRh(AvQBbn~xO1Ds{#
zNCJ(m!-~x%PLa{-FX}Ur%>J?-xfEVR3%+`K7A?LcPO+<G4H;jBVbjbLJ*6%TQX{Sa
z;*;Rz<{IGpPG6RY^e>hhwa1Aw>RZtfL~|bQ9PSdsdGLppTG?kgN6bNwF(Yapz$v_W
zCHm50^ND?jp(Hy%Uu>H!Ps^v0Sxnk>Su!_ch(e<l=0e}P-F`wH#9?ilHt78bxn2-g
zPPH0|!*<tYXV^j*fHUJ3SZ7|MwJ%4P8h1%&EA5DOGDWUD?7ywm(ZvZDv$mqn+xfW`
zC=G(8#@tWBFb1;zI^TnY)AO{dl_bo<kWU+VLVvuQbHSMt1JiX0^~g~WYc^~L6w3G%
z>dIzf%8Ls5w*;bkA_2k@Uvz>3`H<#fa8r-OtLT)}Esu_5BIZNBv<&z?xVaWQxJu0w
zLZr-KAzjO@05Zq5V%_Pqmn}LLricGp*(y0x=eo8`tU9s-1t_fN@Z|T{Fn%+~WXWm@
zqOtF6p&c$&ZM3N%D_GM|(WH15&@8?eKb&L~X)nYcLV~Nr*OUhgJf!?~nb+Jq2jG;e
z@OQ>)<VP4+<6j|YGYl<0x=p8=<Rd%x0C&tdaz8Vhigvq~=rjth4Lb?3V#K?Z{^v>p
zYXq)N1Gc9ETMMF@4Mz5otZ89F#{^?<sFUiGdH;J$0>n<%JGQ@v%9#$UJ#kM|m)WZ}
z_Nej8v_0Lla96)5Om=Pd>$jLfFjaWf=vK$;71Rh_N3dB0EfP8Z`}<9G^e3T3_h&iW
z`JZ~~)BLi44)=>FZcwx-`utL9U#{;9W~`rLcaU@4H$GHt9aXq9(FEr~ZJ}nn#A92S
z<yn>2EO?1l2;9$YOYRqq{_^3;#)!Gsct=k(`MMA*R+wjlKUi}5J<PZj8t*o4X!R{C
z*?Z4cVsNMT`mOfJ2%cgO*!ZfDtZMcu;3HzRQj|pc{~cu^5v_~>q4=VyB6lU2HmH^%
zlz2lBHNpwFl}9M#dJn+WFTdjUWzi;(1;<MNg%s;LW3de?YTw@N=-+?vvBE{)0h66M
zY`abU1m~ag=8$+&alm0)gwMSETNqO(BZb)v9$zfK;<7AU?ZG?@%|e8@gOGg1vJw0(
zmdNS5otjH0oXPZ_@MT@l6_yw*jOKjUDwu?IhR)oKI45RS4;<Hq@XB_z&?XHS)@h^(
z0SU02XegFV6q;{2QDom+ATJadpsGUUx3|)Cwau){5q}&w^<pJ{Ry;Y-`sc~GIASqZ
znM(PHx-|ffysQg%n^!WcnWX5vB{Ob(+s@;(IaK~jeDwmLUu23+XkZ<wkBx<c%x$WG
zmAYzj0@PXYEnbNvu)z?=i}!bvZWdNlCeKz+{3+J<m|VP@-6$=;leE@3xdyVI`OD%*
zkZ+DK8kC)e`QSKeS#9mR!1Bz#+Savo-i3T7q7_f80}0ZK+Z@)S9$PT0Yy-8i*qLOR
z3qE7FXHwgrTymS{GV@l5ce3|WD%H;9FINcWw*b?VnJyz5oj{vVDN7)0RL;((oOBxh
z!avNq?`X*0_u+jpoR+UQ<ZElmwYK^?Z1RbBx{?9;^Ax3qWa#m52YanlVf6+l3oL5A
z=a%Y(0yeJ;%7Kd|0aU`Gc(LcMIg1rve^i)vn!A-~VT@=rjm+-}5lhFLP`D}nFAAf-
zQ8tfxiR6Ad{)!TQM)hVc`PY>2$K$#%q5b!z*OG!*Rvda?b%f7<o^1Ya*OfEaNQ(eV
zZc+hT<EKm@ag{)@6!m54s?d(f3$=9Z#B+A*5pVZOWr9h$KmCq;|CjJoUG=8+?D%X_
zqjz~3W4vsQ!Fel3o1jv*L%l0sP<WwwzgDoU-{Oh$UWIdqtoYT9N<Zcxoj(kT7ufY$
zj=3q&|A`w^&HuSw*dE7quWYYeN8QkAKWf9qEu<f|nGav9`CfD97;E-3B?Gy)Z!f;0
ztZY{<UaP9I>Y9*wH(=$aC`VtOKIW&0(#8fl6dMoik-u3+aook>dZg2~>Pu;jS4%wj
z%dwew-TSXn9%Jy$*-7@YwH=jlUj+=M@mM{!cJ#ejp{mFtN_7N-4U-jHT0xsj^+VSC
z+zDO^Xv4W8PnF^BPFQSbZ1sl{#Zw=^j8FxRwAmG+hkoPB63%Wt*oNN+{Dmq%D0cjn
zS_3M6MsNOrfULF#o7)R)nzw>j6Uh%*OKSqeE7*q*us~WU8Ukb8W$NiIKp+&(3VjEd
zFBd>UW+32=_t&bj*@Im;=K2#?^1crhs^725_ipR`HPFJj2~Mm|p<~un^Xm|&AxecI
z$L`NGGaM+b<S;GFZ$bd`SiX!mM$!8<I<TWu+onc}45g@M`-;+f>9fankPM9Ev3RbM
z6SAQ`Q-Bzp%|PVc2DM!6Q74&mDBgKn799jWA~34c{ZdehJem2&3VX>E<Z46WRe&Hn
z`akF8L6hDLE2?CLxo6kRkKrWBLVMiUSZHyGy&4;i!2PRpXy#G#a#I9rt*j%or?RGR
z?BN)&R|*v=wd*a<>vC#fAa$kvt1$T{DAvZ_g!ZT-<g!oy&PZ1$vEP!)5jKxR0(;-N
zOkcG#TlH(U@1)N2Vqfe1p~%Htwsl|+nj84mxNLkowb7H=vhpzMAz;=HTVt3+*IO=#
z=34=m%VaLIXhu!gfa1hfiEq1qhdG1zCt7oZ`qB9l;{kY$BaNVnwt}n8ArSJl55n-w
zLj@xS^XTS8sqa+D2RxVjQ9lTbPdX~h4L2r1G~?8@+!lx`!iytb;hxu?Z2q@c1S>si
z9qwcM#Vj#$RrW4R)n$&Eh#7JB?h^RpNr>}_EG3&TJX~kYvv-fH-p<=e<_H(&#m&DA
zU>vyGuY*H^N)k8763X~(;&*-2XoMcw5HZ)Nl~wcK-q=*7ebCTVlj$sf%6Xn7ck`y@
zJhztw6TMq~*#+&5wV;`DkPK#kpEa|)yDl}5Vh<cKkxU+8n*VN$*aamH9s;tD^10+~
zkPvX-?fobY*`+_-jGeC}QEt`-Nu)KkkgnTf%&y1S1Rlt7k6bcKYgz(#rm#*((2*0x
zTx5El8TfU#y|gN&4^28_IQjtA<236y>y5T!?4b98p~O262EF}vI^YrpYJ0dM*;-Z_
zQo9l-A7nvZ3adRd`TG3o2gWn5s2JL=3qtrBfa$0AbnTjG%7ePQ_NcKq!sg}-jeU&^
z7o4Oj5vk!Ujm}u=ZZh4KTlxI3x4*3^Y~;8S%3B?<f^rxhixh3@*|nfp09+6{@dm*9
zI8|PcYaS`^<54~N+HScSPhBS4@m-(8`gL@1wjx%l^CT$*KWonr7s`Zno2}#)R!uFd
zII*nF<bHTf4hxZ-9TkB8qVm|Jng3qxW8e+UkYt+QpcD3VSnvHFKY98JM(QGj7U0x+
zt_qbIu6_tL7+-q=-(i%toXjBKZ5R1Akd7uP7r=nIO27Di#+X9Lg&U(B+;B@o7;%;1
z70c*x!{-iv6}*?g(VB|Kur=sQ8M0wNV0HfznVdkiP|nOv@KgF6fx-)l%6@MML*9^p
zNvBKnJvN#f;H_jn$9imAAi=BipHqTY4}#NtEB})Mn#aIr-1%(ud-WZnR>Wad*0S14
z{~B?+?V^&aqz1?ZE}n%@<a$8NpYh~QBRl}ZRX+I7b;U>bw5hXgo%SH!MZ6hUAIT^F
zv*u*HQf@v<0J*87fQ0=tnx_)lu=?H4yT3{9&))tmzZOUaN}d)Q18EgPzdt?MNXgA-
z$?}y|gG6o6O|>*v?#R&Y;W&a?r*9r`$<1gET1^_J<KPW(e$h#++u;2Lv`e{MY`<VT
z2q&TjIyF16!SSC>(+P>5UJ7{6w!43qUX^WTHG(U<VA*D$KGT%nbL|(4F4_qYg<rt@
zto91-f&IoUUYKnLA`PwVRw5ZMqi&RjkeHqag9x;YiP<;O8tjs|4uOQSPd4!u3-uCv
zW%9xBWWGXe+u{!@xf9*hyV!y1#o=SO7n2DaiO<J$T@!k3UN3B%-~DKZ$x@3T)$JPD
zenUjc$sqaQWPa|)4sVI1OZVfO&FV4uU{u1XxYW;ASX@Vcx}%X59<gVfzUXeXP%rB;
z$J?dVjii^Rw*N?Yol}D-=5zBselky?)-&;xjV>d7$;g)X&3#Ytm5{y@A0l0YTUz25
zDP|b0;4Y%AoG9$jf#>}W3?XYW!+Y}J5Gmp@(w>D<=vh3c6o#c#o&>GEnLjVw|7Tmf
z#gOu7tB`1-n7UeCS{99iT+hqrxQB87N9rQt<4RX$@=UknYS`Y`dtX7+$G<!y9CWMm
z7Qx9qb#FS}Ty)rR{+@=x7UZI4XD_3Yq7mzfk%7THFfv|E*?M5X8WTsQ_@eN|9iOO`
zD?WG^2!Q!du!V;#TF|Je<77Pj#ad#nE-*{p)`tj7bXU8dtpA7$P}>Idc3n$ntXuSD
zZUNXvhXiWcLwYmS2XA@FWDZy_KN0+qOuYroYcKJ2Gg~Dx!>rStu~z&&g<N#2VWMdN
zt}hOLDHyT9AquU7Q*Nzy6h_qZ&X0#<!6CB9F_#Vvs>L9`P@Tx0`?iuzzt$;#bTIRc
zGu8=8JP9(U>-mDG?Y#K_tjg=sVUOb*BV;;@{0Nzw_i+r%ol^{jUGw8IYcqM^ruaQ5
zeDbH-yKVz|B&Poh|2{ydbHoiRf{{9L#QP4Ur#pliGNY*+x@l&PE~EGs4(<HbR8%sy
z(2I_mlMS`(LN8rG7r+?UX|e@IcfsjRxQn*v?JHBcL2_dY8TPJW79BAwn9d3KWzbV_
zxKOn45Dc1v>!9#6q6A}#b;13b7esx|dow(|$w!{F7xw%Dni2QFa{um=tDq|ovhqzf
zpm5X^%J{`*IpqIm0d$N^IgJz(L;l&v@N8;;@9lvplhgwk|Ce4Ys%q>JO;zb1)wUC7
zPtx0UDfu@hv~8U)@2>Dy%)ZwG5n4mr=?`XH4RN$|fYf9%nV77FV_3N)q@7E%&l`vs
zgf5@op89+QX8M8p{o<;HGE-*Du1frPYeCD2?njVKBSX6~EKF~3x!J51;-_zYGhO~7
z{Xc6*0%#uZD6`8_UC`jh@wd!}yktw15>NU5`29h-IS*KjEO4KL+sECvaXnA#xQ#AC
z&R=tR=krw(6uvhMXSt$k1(qfGU+ldi0A-W~?4|`{$sB5B*@g+m92Y?Tm$K7liI^*(
zz&&&vRO|U2#=@C2yPSHlA*vBB=KdR^REUF_O(w;!LnE!i+ruBuUuZ(99*=7o)8Ika
z#~P4RhAHRGaq~fA->{6CphuyYW5SE!k$r6c`)14RHsc-~P*yNS>b3RZH%X@lE%th<
znDApZmc#Y>a17cf_>Ah>eVXd(Mk489fjG}*i(-Dk{_^M~g?S-!C{1SDGDe8r|FNfr
zJ9h|=GaR<#pEs_z<I7jO{pRJo!@z%x8})P{>>W|I)LaZ&E$#7#aV^TyLxNc1;?C-a
zOzO_%eju0(12|Y$UzmqcPkMjdsIvM_9-CC#S#82R(A^$}FS3y$d<V8PvK5I&=ri(5
z+4IcsjdoH~sGc0zN?cM$EWnj_MKI8^MOWDSDc35sf0(HTKeE*+Tj|Pc`px`IBXpM#
zSc{{0^vYkCvgfeGUki$Pgz=YJe2h}AOpJo#ys}XR{N^Ga+fU52N?~N!NN_RfZ;b9V
zlowvOl$L{10;A5y4jN=tXw&%NT6(L}P_bW8PtF*#GT|_?2zbRL9A*~X9bpvO0Naa%
zFENc|%V4k6>|b>qSk;2AiiKo8+;QplI1?z8*#7s+iDo*GR^{q+2F(dwQIgP5TwJ^t
zD^uX%q%-b{jv|Y(@+ONVKJRn`1U-Y2)cVhp>nzhZ-Reg(ZdulHkmu4So^#d1;oY3H
zv!|phGr)6AcT6j4BaB8(YhW!*n<cw>Lye4Q89BoVt&zTuKx5k54SzT3F;C5c`>KP&
zct|E<d;t#o&b3Z)`>y_tGNheska(vQz5n?msdfQe+(OcoaL&j(;$cgjMB0~QouLl7
zEBg=gcYXIPklILMQ4H*BXuKn<lC$BT6v?Qren3;<h>8t5QIbi=wU2}8)_;=V2#18?
z?!Br-n?D-m-hZ25A<9BAw9pv%yUDMV+n+$Pf1=!u0^C-;JJvbQ>mbw-)D2bwn(d~4
zA|U@FW>KY1!mdNzR&1>VI7Fj2pr3yAYtvo`*H)Wyg!48|dG#2*oX^(+p(!#lGo}^l
zsjg1kYYIo!(~2;DKpy2f_nLLZ&1S?a=81%{E4^v}#ux3CQi@bg({OtqR*5}0Y(NPf
z3m>Bc(GeMF(C&<|arBq|onh__X?K8hfOV-z&#g?QLU<2T@vTK^8>5AR{}z3o*^g$Q
z*=o=_>u6bY{WeSxH;42D?x{JKF-_k4Y1`81B<LV)E@Em#Zao<i#GWWm6f3P>qLcE@
z35}5`EXuy7@0HPg(6TA@!}mM2{}>u*$5LGV9?C=~;@Qf*|I`Qa>?`*)PR-46f2{jS
zfl6TiZ&~V4WZ;qP^#COs2j3i%g&5M5k$)<bR#g=K`lyX*>3vYzC2cHP*3U~utEP6e
zI2&BC`^VJ|ZE6NwbqdwjKBlXyf6S1-KbAMtqq>+=+1lchj#<D<ecF-w*)QZ#$z*5L
z-bLryn%Y;cEh(hG)StgQ&mF@=F_rW_GUKw_D4GMBy&P07nuh>(-*%pjC`amJIbNP9
zbqU@}QN+?7Y{gLK#8Rg<H0>&Xe)#}%W}b*~trQQIBK1Oa&>W-4bAu036~r7Tz_}Zj
zA$~?!o35dxaYSA}g7`ncJeB75jsM+(_6H_dJbvuJgfNXQ!Xikxg_8<4@Rdj2wRQy+
zQ>UA+a{A|W!lUgo(EF=)ysTZP_^J}^>2DP!0VzYv1rQ=Fl#U`5uvSH8GwLviw4&#o
z4g3vUfs_&Y>0(4NjspTRh7o!&|8W8Lnmp)_P1tzQI2#N$>pZDT2x(|B!k&pkV>A(}
zqOTpL5VazXDmF1RFZ)bUgl}B;7n8MGZ1(8`ya6Rb`d{<kuYo9v-`Xnjcj`WrN_>=O
zTqp%>OM%2E&a=J|I9;WA%vn!M@f=b|D-F>xCY+=)e>DDE#@u*<sZn`{WeR+L_ato0
z(Pxf2%9~ZrL>-15q?3lA8-m=j5rGj^oMafR2t8Gzx&TsrSjXqL>XzEHADpfZt6yNa
zJ%W7+a)!bL#{_il5y;`z&A3GB(Zi_CcrQBNX__pbK7P1|rS1Rgk-*QzmdsFpwGW03
zby_W#nfT&AwJfdHf1SJ7i9bsb43Hx^XE}mXM=civzzE(AOhxaYCps7s;u{N<sQh8H
z^G+r&d7+cd(6={y!amqs?YGmj2jUimh+zJr`nFoyPQh|Um}Yx*z}=%0oK+O}PTnz_
z&>b~14y=v!US0RKgn@cls8l+{;0;D$6=EM1^Dl6tIT!skWYv2AzO(`xnGB{+XdlOj
z<z_$`8=fFxZE{*9^~f4G5zTOZbk(4T`#p31PyVHtJDfe_NrT-6xm|%|Ps3;74SID>
zTgyRRXW^wpjwn@92UGk8-Y;(XM*0%l_yauA(sJZr^d01QiHr94f$q&pLVK$w&nFt~
zZ;jqOkzY$r+Sd78;{9jbS-s5oB+8!aJt{9^l_erHGS`n7BF(si=x>1YroEZ)W}SI4
z5|Z;V@+dDUuR|{qJCAs`=x9tuS-`&A_?avC{V}O{m%&k@E{cXDm(mGW<O;Baa{O@D
zC2T_PNG@z2F55-ops3P*N0`Xb8YYBFH%GPdwY^?_gI&p3Ce%tVrq*k91luuM^ma>m
z0PF3(*OCuv9k<_b#iZWOA7wj{r3Kusp+-Av13p^2i8F7vJHPYG2uo4XRLsKaFhr8n
zm*_Q*G*b(hkj#C^4yh2MfznG_qm4J@j<5cBn#6{hEyV3N)P7TAYO==;5lFBx3kF6K
zA;)VVC=3qV;$A)af{6tgZY-U}tM1FP-{XJrqfvnHjd1A05YsA``4)e7;KSr4L{}?Y
z&#P1m96gFZ4sb7y22~X0uaXgk$B0xyZ|^SX2YlywoAcZI9A&9zk(V`>63&vR2k=Jw
zRFEhHr&tbIwc0C*Itz&FK04@|mDG7(uq0l5c9CB+gz%k^Ly#hf#$mID8v|0ujZ+J#
zrUlSqZ93hP8~2_#xNsL+z2%RR^@P?zQL_Rf5;pXIl1rBxg$c!Jt;^@JGV4Eo9pwf^
z<}9HHiJKMu1|CRL1msDCz=tq%UxS2~?=64Yd^|3D#c9(0x%|X{2^He}a~b-IWKt{R
zb-&x)aTW`be0Er^L}@eG8jZvR8cD}Wa<@I~V@QX%?v%+41kw=hCPfy*x^Hp={qe$|
zzSl&jQEMrR6X$&sThaUb4B!~yWyPJ6itn58g?;?3iKOszzPc@ZiN&zbdl)I=RY$4C
zZggMFB5>$wsu$-;^$}bzvSHT4RiJUv<I18_)hm9%++$LltW=Q0tooxOk8rzj!;EKS
z+A!QkX{g)xJ^Z<|xLsQq76EQ3%cWKqg3GTbyhzAeb2aweqW?dWKG$3}zZ_9lVOm6_
zWdoZEk=uJWTxv$^*YmnhM{U{TP`D~cW_>b6c{@-rzt%X2S2SZ%nwpwQe>6zhRU$tk
zX**_Km;LD>*QHN1BxUm5pdMnfT8(yy$BGY0u&LXtv~M$BfYA1KgO#`3(9oPaolCPr
zVH}CA$h5048(K8kwLV)O0p+H^<?no|ynPL(8%n;WVJDMP#dhv4cu0%yoTafDvgTNR
zly-aP+qL(^B)1mz_5V!QxG3+Z${4Rj#ywpq;-R+zAN`3hrM<(av}X0@6gbu+{_$k}
zKHtG&#-yb!OXd!su%7EW)NU^fGm7XA;NFSGtNo|wle}TbNH~9FrL`pW*RTSXgq=pU
zPglwnxeqJKfKSmql)1EXZ{Q5Ow9dWw{O~}_Hj5(Zd%dS{U-Iwm{vN8{&2Lfl=8J8j
zCmHt?bzQA>cqB=wlneK31xxe0ikF=P1W4cbCc95~PAa_qa1e_3FrCDrB3I51%uUQx
zEheMVL+WYkk3#0lj&qVm${hYygzqN(ySr|f&@tZlE*ZLBLb2eQAz};q1C!OwGQ1oj
zgoz787(fl)xyx0b53_0brC6p0Tb<M%D4``A2=!oH<qXpD!IBq5IK5``S*4KNR;qmO
z1TkpY)?`xB;}=l>xVrW;V2z@Qd^v!Sclx71&Z{-4ya0L#tmSBQ95Cxk%h2Srojj1r
z(n^feWc0$fp1j<|?2AS$h43bp#1kLNRkEnj^2%R-4eO798<H35BA{<C$R1DQEtFEz
z2UIT#$NvlQc|O5x7>qFy<%~sCShi=-@;uQNztk>4VZoq>j40zPul3hI=FvIrFStJP
zPq4-d4&#w{=hnkDHb-w+bLpoesR9TEU=)FCT%yH%Kq>kjEhQydr2dgy;whe=n)2+F
zFA$Bcd&v?Xi~d`BvgVpGTcpXgK^lZI0;O<y>ZdjXB`+HK3W=2-3`G(8qrG3d!wR18
z4ymd@dT9KblIxVhjkGU-Xs`)QblgT#ouI(Z11s!m|KO4U_a4u~S>j@1rH3*IWdU3C
zW~lT$dAFwW5MlAWu%P;*!(YTTTG1ZkvQOgL5aQUvG5JaJh)|Ja>lJ;cVHLobluayG
z38iK<0aLdt$lqPR$POZue|S3=VCa}N`GUI!e)EQS_K4d2-(xR4Ss`H&nOZ#o47tyD
zX2@K&5Sj$hQG=%xw@mwK3RoBH2&?_jYtV>M%x-%$`uEM=fY5=$W%~SJJu(gJ7^6!<
z3s<0<5@ox<gtc?xKjXm;Z^HHKYQD~f+yTR+0H|)RQd;W2T9>`Ao3Cf=UeWEp7}7v)
zF}%Q?6J=ON&>^{mlnQ*otjpB_Z2-}PwsWYp6p)K$$^XRFm!EZ!ShNWG_A#8tg$~A&
zu6Lgr1vjMEZWjoajZ#8(P5IcSWuM1jpgw1cwXaew)}0k|d0xbrc)OEF;i_<u_Kx>>
zfLoRt7bz!4s!TpyV6{LE<|-wFbcch`{Ht3^ODhzOXt<OV^ep}J)wB{4u4CXMCWDm1
z@J`QW(Y&$<;u=48zPC5oyxb8yeb6JlL3wdD8QjzQB!JTYCA+UrRJ{ymjC(clT^p6d
z<Rmg4DHow2(m5_Oaz_hq<Wwh)_7(lkp#zBo1KR4cHDQ|UoLf!U9{MO@r36|K3)vAZ
z@+?R0%Ao`<8&j#8aOIj}*IyLEN<Bh(m2FVWpu|`Zz)y_iSU=WzuS%}hEct_eLP<qn
z-VLh~Lbx+t;Ciso$sR`DOw{zyc$Zc@kv)6J;Gjnn>|ZSf5f(tIGTFvui(ho#px^;)
zkBv3H%AoNY9l)T(KDEbrX*he}L~*5Obs2t#C6TO3URg!hXSWTS6nVqb@WY6+%u6u}
z+dt^*UD)I3iGOySG7O~;=pcC-&<2W84luHq;G&r0;&YE^3VRIGm#ZW*nAH!*fqL}0
zd~-gH7`_gieR^h=hv$2<b=Y0E&xc1Iq?9<^yIQ*WkKe4v^zW}UDuBrf9P4GQMIv|J
zZ^YqlZ4M|;V((Ghwx^h-r`V_t1dzg|I|?!d*OQG$qGNEU&{N%(!#dp)V0g~Ra_1y*
zqT%NOm(=SDJ%3P~UVm@GWv&>gh}%Y@(XG$_*8om(N{g`rMm$!3eL$NTKV}PfU8xw!
zU<LtuJM$&AwOTz`#5M<!_YZ;KNb?^KgP~<9L>Wd(CfgU0DRvZoc6b+6Tje~F9G`vx
z`foy#a1-kfv2<r+3gl#2F5Qbz(VG^M4&ZgnP<J<?>uiMx3jq`=89`9Xmf+lmH#mk4
zkaN#hA8&b;_HNsi>WYSaAr5f;qiJe}PuOt4p-TtT`3)2|XqIzprfnjH$?Ergst;NS
zR9yD`(2<5*CH^+_rv&nI1GE<=#8REoF&ieJ$uRz_sBdHI6p1s#c};qf)JfMpFrvPE
zmxoNs#~SC^gQtn6UGQ<B=+{KCCG~;eb`MgdVA){~tY>9*NT5jNX8u5Kbs@JzSfiK?
z^O7|DeYnSQQAu=t+xJg=PK3SB7@-Ci^ZnLjS#rSB5g~SczFrwJ`OeDLCw4$jn9AFP
zh>M1p;PB%wRt%n;{%>^0O0#oVhS>$m8FO#r&S(1O4!#d0LO3==GIpGD)9Q6zE%k4Z
zV*5^qJTELGTF%Q3?--KUzaq6<cl(TtN00OLq)kJ|9%cWXrS~emQE`zkxUJ!qa+A4U
zE4Bw(ld{!rX8owL_?2LgN<~`~){R++RvG2IF@LHZREJy2O3-pXm11|iEJVwuLHlrX
z_)Sspn##tvGz-5){aC**FRkDjg#XIP5qloB`yj11*2$6>F&Z^fb<$z<4hp41x8`9*
zdOH2o3R+w)wUoz_4dW^zWSyq^Revl8)oYhRSeMo<C^ku*62B1fa54dRhaLtE=6V>(
z=+D;^#FoLMx9rxwLO_AnJ^XRgj#tI2e=NbD?>6Tz6-3tJa`Gvu`l*!Dwv~4Hjyxn)
z3V~9Jc1{N%a`qY9R>Xyc*toOqtsh69c!#68r8X)d<7(N|dbMF|?}R4{)2h#U^&fAK
z7t)mtQRiFRuc1`6`AnCYHm-NXbg!b+CKghGGX-ub6v|B^F2p^aFsly<67&YARi&=)
z2lic2i$+__X`dV&O5nZarnr4kF9u^^?<3(uOaY}&9O)=j(x@uE8o}i;wYcQA`{hg^
z%uU&1fi7__)$(8#RH%bf(Vj&gU4Zxw2{))v*>eZqKM3A5Ct;38V6eGF?-w=R5f74#
z1(X12`}U{ao^O@Er3H<G+1A`fMIIcHT6f(^5T&<#SLK5by($%#o<Q~+i5AjBHag;V
z0KO6#llN)7*rnDl&+E`u>5Rlh$~4nMp7s~(*74F#aL|PfJQ-5{-c$>s6m&mp#vh_6
zYxuvE33nQ1PM~%foV<VfjCN?xg|6;YbpSo2+<sZFg)P(*`81gJ!}w1^?dVNcTwQ6C
zizGNqapL2X3&vqY={-m8Nt>Y!IhpNg9i;M%&O@0z`Qm}bhpVGMT#k1cbM1LrYUDnS
zZ&oP+RW0tXgG?w0F_+;?;q#?yyxcMQp5)li*h*yBtU-h+1J$uq=p9oB+qtQ5m&;NF
zTch1R6%=~#`nS*`NfTfb)M}kD;vQQJk_ovKu}nu&wmC<7F>rMy=~B|t&P6RvM-B93
z*;xzNKOY~n!*rzJx{7^-{#ooD8bn`@0*oW_s|AI;GSMQHv?D+JJ7w+d_x2p-XE#3X
zTto}meBSs$7~99TNbY{M#S)8xSs+k?ob#8RN|WAhJO7(WiD@J2WYJ$Q=0d`_uCRHZ
zI&bF7Bz@LD#j9;PYb2A3w=&t%WVGYsyP~Z+O@`gq?P2Wm)|740f(5j>9eD)>v1UKk
zZ_YMjXK&Up25v6L(>->^3)wYS3$T)HmXUd_3&NHbh9@TLPw_U`!|+YJGx6+H5yu~B
zZByYGX~j{7QT8B~CD*PKEhP<unJ?7W7z^MfI^4{?nR|+z$W=N|SLsRgxahak4|r7C
zUS;VT6GUk_kVWq3ZI@OY3)Q|cLg@)+M)x-~vsAAhOt&sPd#m&;chIjJD}-^U<nLGi
zN}blq9zqd*DE)A>vjF%A=NSe_eoL!rt+xBm&n&?#+{2Ux;#c2n8|`P1XPy)tfTl!;
z9aQydCvCHzaSVfRQy)P@wkz#IP}FdR7%1OAl{CUsFSmx##7m*c>D>rwf6V@c`@Y<m
zo**%EXrT0r<*H6W2zEj+TyD%me#?4Q+uJM0h&}8Q<z>KZ3Nv9g2`$A}?>3-!=>N*2
zef{Y4)&Tqhm^VZ#14BEEYiE!d6p}NefvgU|E%M0Oa_s?>_X_TrQOaMr-vSqk?T(iu
zR0m@=y{^mUW&m1u4@9FeU=bcWUh5<56u#+#q4v3Vqe5)5p0|W+muU~0g@2<u3P+4=
z-9vXb4!cM&i2gMfFkFe^DhdfnXAW*n|IQu8$fde8ByV9n1PyTllgyAa<H%N2W5yXd
zr#p%U8~b7djVNlNa`%f0wk&>nBjJtOt_p1CIz(O==ZI-&Gr*TtPRN(*OE)wh3xHhG
zoM)!8G*yB;z|XOnUBpfaQ!4`izp3<W6B>3}4fB=lA7guK>1*5FNc_E~BdQCDG8~}0
zHk7XYxi~O*CVw)_3%~yd?{?amdo*ZqnEI5++HrjGuSd%*`~K)ri?B51YS>!HeA=~h
z<l<&q*%_oK#>L;0dk$YJC0n7)Z^T`iXzvL$=-~A6D@KY3$JS@Dzc@H`&oE|D*mPiY
z#(L0&Y^ceC+R9CrtNWLwZ}IuLk-1W3j>@sZ8m#J``<g$_X9{7?J5NQmWRIV7(_!%V
zH{BM0n0L>eE|@AeKO6H$b&y|rr{N&RDQcsWK+2vuk_=oEMa*@72+r&9=u#h1ey;d-
zOj#wduX~oVGhYYWcih>*9!Wy>qP1H~NGMbJ=Kiei{;qbrtj}Tgm(wncW>EHwIJ}#C
z?CR{fcNT8?Cj;v$&EMf#FKnjEWQ%oPGsJJnQX3!xY>r%-eTe&A*{qG6_mR{S_mbTT
z2n8aGUFf9&{byGsJxg=<|F%5|*+<&9oi@x=ECH}xb))A3SMm42i0=ShD*8QetP_D#
zB!T>;I9Nb{Nid!sXs{15Y&p8v+-o6ux8;W>qy(7d`M=S?|29XkJU)2%e<Pd3Q_z32
zh9`^nEdPYZon01W(b7`Uhfp7e@!v<R-oBb>PfBFctk-x+OPf=<*w}r!bd4oM5O##o
z2<^AB<~aAF1~(D4?HFI~2t4U%h}uRf{%&x?<vZ-)cOjrE?R}b%Eho_#w(qO=$|7om
zZmFmNQU+J27S2QB-=b20C3ic|_g?#{xgcIQwSS2&^B7~sW=E;)`Ka<vVKjwf-KU)&
z@w^d_n=R;qe>K@qcRt6k9MF3GbQ*jdu`G4eH)dt|L-g1nM4cfmPyA{AkFmFas&ZSy
zg#{N#N{5tmhje#?fOI!VcXta&H%NDvl9JL$w}gb0bc2-K$=+w5yZ0Xd9se0)4Oxgo
z*7tpLzVnSI-l#Q5M>fpho`I0=GEL}a-^vR8BVZhLl~CWj?EMRHSKwHnG!{pL*<}-7
zU?(^~S5w>p_rX!tpsjV}1{7P{gTdAz@+VcH2>g%t=dr|=T?L!*()0b;%iqV(Dh2r;
zxetH6SY&eyyGiH2yk|OOl})Qn=(K|@G}P<c4W*>RAa?Y2*^#4WgTa}&QKoz;ZBXs&
zgpjYlm|4Xb;fB-Aj_^Ob)?YC+53o)RGeOt<u*5hr!g;3?;U?@|-L^m1b$889y2?N=
zd-YhfuqMZ6Lit%Gkh;VYvXgzZdY3t8XLgJaZ<6)I^}WZL+?ji*cp)%Wo;TjI3ZyX!
zdZO*fLCuN`RzNG&c%?R{W|A5SCzO}Xz4QLJiTSte-@7_~-|Q_=UjnG;pV#zJ611VR
z-Ku=t-ao2KFGh6y_d<0bmX?;&194=!?En${UE5`MdT?C<<&PEi=Q{i2&jC*86PVVw
zL+`Hp0yhWgq2L;{8X1*<jr}4BWp4vszjf;Src7dhAx{cC+#Rhh*pgUIGVIX+Z%Mme
zTsoSE7T2c6$;?ZD_?_VCKo|OPo-fcM!iKUt@bw!+hjVvGjD`dX1217qifYuz!<z2*
zs*4SmR!i<ID6vAFaBn*CYY+@w!K3MtHsGDdv0dv5u>^|QMqPeCW<@$ZZPPhy8h*=P
zA~s+W0See98-jbi#!qQzO^R2bE>iGp+Pb~i+de6N_D67ej4w9OkLus$YL#UNG#$W1
z_{T~9e}4#m{>N*BEC8F~y-Ml+jkabf6JcfEZu2-NJ9N#vX6I5p=c{fM8J{ybF?Ji7
zy@=9@kwZ<{WDQKY_@uV0q8yqQb|v@HUX0_*qnAjE)^k;tRwP;Md)}4v$qW(G!)f)h
zp}w1`#(|1<^-g{IAyUNX;U#EfNQ<y{g0W{~#T&J#ixTb0^8#9%R;8MZj}PNGO$a2Y
z^~~*;3szbWV*E$V+e9IQ;fs&JaJ9aAN3SzZa9P;!t@oFUj!t`@TawNne)S829WSsk
zY^3^%=FgZBi{1n%-2B`tfSvva@9l3l<!{&h;|J-U>rCDI4>9#wEi5iS4g74|+-r3)
zKeS}9nsP!VP?!b-45t*(IjOdC^S+aV9z^YI+2dT9W`Y&s>&+8E0MAc8^Sw&Ac)}x$
z!Y#8Knu!n|64D7O{1Pm>VH?L#!bF};%%NVw#K2cT8-*AO!5stAYw%d?7?9Zm7>*SR
zXu1ShKKecwpYQ?WUbG8f9AX}XjK?Omr=S2<|30d}%RU$(&LwDczeApZ!O>a$+tq+a
z?037kG7VtZ+KbA*h6FYME#99^^!OY8{)(-J@HoHfsJh?8q2Nt<qmowbl6~M<Z)ay~
zH86C#`8M};!`avKVF-s^bO}=j+uTf+mr|=$Q9qe4YASz=M5V4(w8g`H#LRPENIv0g
zxq87uBj@(!lmb~<*;jA=G}#x2WR+STY4#Iun)cXM93Q5YUH4kdGF2sKs*h~?uloWv
zX$YV%H9gnt4W$2|UNJ!T6S`$w-eJ+H4Sda3-k|Jr#98P0R8%A(+<KeRao6jzF_(R?
z*r<-9>l}B9`xvuWWS^A3HycjPeFWxu2wfIZ)dGY!oyPM5vcLB5IL0p!>N1)uMsIA}
zJTIn!T6dmcT>C$61-LD-#Sk9Xx(F0(MtVL&#Vj?K(TJI<%(hAfefe7VQ97Ccq)yV@
zRv9)P1otR&_f9H6)Jle>wn$8Zk$pE^U(PL<&Pk^lmTacoc*&B+JEvDI!2VJ~6xsp)
z?p#@{?N{FXS58u{qgEb=b;Bp<a2QYxKVQvYpUD8xDh+5bY*7W8z@I&TUvKk!1e?}A
z^4?~97(4D9w_B)U-S0?ccK~m$^swbfKy$NI2S;qCQd^z>Y?u{6P2e%qvq(M>xdCXU
zvd6(1kXSDP4tXWbPTOT*fL?mBH)n;>=W;#Q@mtj^GV4DV{>B{g`y|GXt3}#?pclFh
zjFt38GPso)^jbCaSax2*qP$=c1K5z(@i(=t|GHN&@|~a&Y>v%Xp?H7e5U~nvN7;_q
zVFMJf^xpUSKxM8;B{NJ3s)ti9R%NQE;_&x9kgq-qICX5Psj8sYFvf<oj`5p~p0#?g
zmnj!_eYWLsIgNgH`b(`+M>8J;M|x~u9gN6sedsYk{Ynm7w*KZZ0ddQ0pKtOb!S3Xd
z5b9NA<4Wc+@Sy_8pR64K@g}T__s!oW(hT{)z8N(G4RK|!{n}8n4F(z-hYI(u_-)WC
z+!h!)yDdt$RpR$NKmH?M=+uV>Sh6x;>>fQ1XOu5S5G(_>j2Ef!i+BGy3s1A21UEfS
z56-?2aA$Sj{Q?&85jc_k|2;$gb7T@9!8z}n;}Fwpi7A>ct9nKwOjl^eS#wrS7s=fq
zd0$g3e#m-V<E>s*H;2J+3|roQ%D`C@$!+{jQ>^{|Qs$te!$)U`10x+qZ^Lx90*~F^
zpus6=@f_R>rA6!f>37KWXD1Vqn8&k}UHZ+$ue{xp*ymRsvXmt0h>4IW0t})ua>WK-
zCx9^4Sq%Z^ghh{ec|(6@c<&d_ZK`bVJV4fGzX0vna)5!Mu5!moc6*GXE+g!78b(_r
zAV3PjcRIS}9Re=%X7WmSml4;WK{Jpzc9;u+_))m=lY0^IYEA6FmI67dXpWCpHCAhC
zHxP~b^oEv$pd&$u4_aJY+$gwfX_<O8O7azMzj}28qPuoO0fZcvJt@06m9JS#RkO~;
z=Qg<{9^FYW0g<+5*20NH_~St<ASb)$XbVB0K-k=f?~JElpB0BDO;<F_l7_Nl+zYcX
zd-uf5E^Y0r26zexi9!IoIg?1E3^HA!pZHK3WL}`vGV|z#{VJ^p#DIb4P&!`WP0@tB
zWx&(*?B+EDyZg^W4ihOrtY2~1uJ+9V0Cdsga?n%%bx8ho9`vA_PnSr!9<U1D7W;E1
zV|NtG=7=LYY2WUahn1DqMu<+yWp5WNJh?D&s;PSodaI<Mw~FJQoRnljyOX$Xv)uaa
zvpZDY%vYIPmn6p@Py)41W_J3*P<H-w6+#~`_dg|LL3gcwGnQ3Jpj>XyCU(~t=A#sX
zvLjMwjF12!2_G690?Od@a|&(+^B@|tJfP#S3nZ0PUi#hDf-Z{s*OOzhKX3`;q<mfV
z*8rVQV~rvIXL}h0wPXAMAQ-NTg|qmQ+aM);tvi$#amJT7VdcNM-hbZY9K-z-V^c)f
zo299LE4Kz8k+-?+=LgSbXN%p|S+;exa+^+M3?y9LR?=>t^1ZiMn0{b;K?W}d|2jj1
z@V@MRrK98CBj>V!-u?2>W^}GdPG-4(F=SA4z-Tt%=hFPpMi7c6kbOEd{+K0AsbLS1
zh-tWD?s)Jy#)ik#p{VVAL9%)f#~P1B_L$-abdik%0tCR7LfE}7t=UZmPyrn!w4B3c
zff^3WxeBxv%K#%_7l6o|=non7+jYJJMvrAEq3bwI@k@{lyn`oj=t0}9>3*&Znm%n*
zg<1SwWq?$~^W!_A+tP$n#$RXQo1c(Z+k{{yu{)j{Bss7M1b+dNKc%ctHl+-luZ`o1
zwRits@j1?<A2jIey6|O#R;EZ$#~u%AOxwb4PW<iPzw)`UT1TlFIU?#t%ueCpd#Gxx
zHKVj@yOtzDx2Zs@0jUU5s?Oxq4g{t7pg^0-GtB_4ztv-lCwfH&HxQykb-WfjV>u5j
zTD$44jiYn@IbFK=SO*}x5N@Yo{KxgV@iYK9x0VCam9>>(S$R1qgMA7CWQAtYL8T0c
z{c~p@fGA~!>q`O_KR;C1;YmRE1wz~efNhl-5aM@{{h=uOSVlqd0%LUdF8CVY$Lw*;
zt6($#vG;`}^Fx$b<{S}gO#ENF;Lq(6E98szbF!B%FPwn{(&}@oRD(JcyrN82b2jSg
zu)R=^Hq-87>2OxDkNRuv*{3Fp+k=CP^h^SlByS3^DH<c^JN)eXX+VgZ#(}gN+2!Oc
zK|RA9V$~|N)LGc1Q7%($w7v9@77j2n{3NkDZ+74_Y*9ReX6!+*{ZWnNHFY@%`y;fr
zlwey3LCYj1CYE3`XmPzf1s4MMEC4`(X}o{_cTvYv`DaoB+v=pdaK676hJO?|2*E<z
zqpr-L5@B?a3$Qn2zNbR96K%fOyH`kHzL5oLxdi&_eOtzSBG5bNK!YYm+|=|7*0Jl6
zUWJTo67JljGPM|DYZN}t;PWTqy9o-GObDqgg~vU^H)_Hw*64tmS+bp7{gKjREA7jH
zam9;mVRR|~Cn<8|Pr`-&`w9Pl-VkFzcwA`J#Icc&?yl>5+0zD|uG1qyBjan~eIPC=
zG0;Zzlr5=Xn^UwXZSxJ%<Je+qf;3o%mo%<$X!i$mUDk95(g~KSS4#=^SUCc*GsSE_
zTO8{2QRGj4N6UJ`8!bm_<x)LMW5La>!(R6AuQRps+pqgFIv#WMFVE7e4bRfR<*yNU
z&Jsp@8ZrL+y!)$A{MS{S#y$~Cs?<&oam)7Aa?i`-tJo64EcNyA;;^38Juo{pZ-7gs
z4<o(*?D8oIu}Ik*QlwC=*l}mXctH}1S?-Pq88$Q9mO5H?^_bSS!&=r@EKf-pyjr1S
zH+v4#I&YEmnnGqZ1ZNl#g1QySDq~Wc?omYmyx^pXD<G|H4jQsu-j7G%|MC)ogLgi(
z_O)s8ZBx?s)pYO5Lsc_g{iIW!#UGK*xfpY1eh6d$j7LT5bstOw;v}VrpM(6xYn_(v
zOv^C}!*mwqYR8z!EG`XzO;ecnUW|$aJ|ikIJKZf0=_f^ON(&BnLN5T^eN4G-%C9;D
zlD~|3Qli8Qal!rcTDRZGY$pFJIsIcF{&Cj+fBsFtI0o2rwyM^^eZ3pehvHez6IEBA
zNLmJDA<8rux`0U~Cw)cDLEkQfm}knAu_(~X-OuS1DX=q7HU?)9M7l&A4;FFCw?^80
z5MWwErV$+VXDy~Ir|0z3&nBvL<927GP9=aY#^5t(>$U4Q5%f?0$J=KP@4RE=XVYQ;
z#c=$t!SQM0$yXV#w?8z0)@P68&l4Mz=SSgxihEk5Xbx$=Wz5?eW=7O@`>tAArF&qk
zftqU5s2jInC+H+FSE0C@A2Ym(B=UkKF7)?(+iQWF5{XK^5f4d9cUF0(035v<ZEvDi
zoq4TS5|S^`;V7W&6bQZ=y_))WRQ`Km^Y{0t^3a*Qi~5VB*$O0bTvi+!%~d5e+o9O*
z2i}49&Y^^@vn=`7T8&l5J)sjH&-}LFUT44!)olSJH%d&?;Q-lz@xfm8Vo2~?gf;%W
za>3~<;08_<aXf3(<n;S==(53ew{&gCxJwsf*oP7Rn`E)hR;0!Myb7UEe~#x4e6GI=
zgu%X0fjTxsyC-=f`+=izJ^HOf$pIhH5K%y;D@*YOsJLgQzwLeV=7@>)&jQ?84s|tb
zB8TtH{ol^C!l0Ky{-W56x2eIWSXe1{j0;aLvpS_<HT^Yd%c0zKh;$}D^!~FP8^b%Y
z0Ah@B=WgLq#<M>H>Hl&Uuf=op+DJhyq{-K8PXh48>QUQ8Y%9K2*n`B9fMj4B)I7+n
z?1EuF*`p&lllMudZ9%X9@Yk<&$Pcs{c=w0`?#n$KA_v`Sy<W?H!V!ur?oSfHF2^vc
zk@rZa;PU$%9br0KDJUn0e)lYCx%Z$zpNO4NWfNrm?nNPr%(sVk_gd>R{SyB_-&J`y
z=Uo$Dm)jvGnMy5Hjl8@J6_G5e&z4ELk>XHx$M{0!YHy*K93SS*E_t5&!l!i;qowjT
zN>pkZk1Adr_I{<L<;+e?CwruFl;jTwc82K)toO|`Jm03=PLsR%X54lbQ$}0Z7o8;N
zOdcx?eFACiCp2r<V@+j{kb~BwT0IE)lm$a6_0MAa?`!g(??dErj3rxM*{z8noFTM?
z6zF?etJ*js<I~o6KQGEZM?nbSu-QjP(b{`1Bey9j0Qi7Lv%l{T4w%C*<+;8v!g;U3
zc%=nt6gi)<83+;DVCGBW5Fs+adA<$qytzPQxB1|qH`!FH6iR^JE<j0EgKM#jVIVtp
zOZyh;e+=t^^56sBmTM!N1?mNT8P!)+6=Tpe!VYjd)Twy?exLNzAf@J&W8-$#CC>-e
zex3dh;jqGUPy|5PX{PBS?V0J_z2AGxaqn-oz7@$=D(d^`YRoClRrQwX^4hJi_;^NX
zHBZvluqkh#p@%6L0C6rD*)htwYW;9phKmMSWp_ZDPlf+>1l_H}vX%LN+m!!0eg5^W
zhzN1Je`5_qoTRj2+{9_QHLR?H&8*SM?7hjbbD&d!HYZ%=$aNZjx)CocA(8aE&b}zu
zz0@c?8kg08wFCloOO*oau64ZT;4GinJlfZ=Mqyo9o@MHhgmD(nDv#flpo=EYG_Ewn
zsP_}B+a!?Ck$#eehR1+gG5=scf%_+p>Hl#|{{4e?3FnB0l9s=+n;>l&Cm|>_VRRB-
zt!&)a&Q(ZL%Kv>)&0yILi_q(&Dj8m|*yIp5tMvAKiQCC?I!5s0^L)Nvf-cZMF*xiv
zvi*9{D9UVgBR(v7TaL#3>?x~F!(%TyhO|C~{n;De9jaamm)Pn|zRDWlwp=?VOTBvW
zf-4`9Er2Ir^!tOEEx-)#0;T?A6PNKE;a$*fI{!6A<_!T3sIWj{KHHPU_fgIWq*)$!
z-K!@+E4}nlF8rhCAVo?^_jBIib%mHa(wes;crFv-I35TydE$r{K%UY4Zs$>G{!z5&
z24GwHgdfQRGi_dOF@les=+pF0@JmZT4Xmv|5{isp<oeC<*`Gg>${B}m1>ODK$2PSA
zH4Tj-fZ1RC3LRVrlW#TUqSq%T9w2x=D(j#AG8<a|U{w`oqnY9!|ELC|vNv1F|J(4m
z2~22BZ+oiwH6JkxGXUDN|F=HbABjSNh(Wr|dzA=m@!bffbCEO(+4J9}1-(uPF`XNO
zuDQcslh8aG>&12b?y+6gqg*z<GRNPaowm86aOG$)=$2KFS)OFE6l!HUD^V8^U+byn
zGU|IYw?~O1s;Xh@c?D2qs^sBus93J|nOj&sg!Y~`J6U1U>*2iLo5&Zd;}%CS7Ov24
zpwC<OD3)Mj^bD}G*6^hY5QDGK#Yc<MIB3B3vc0_&w`~00)htBZKmY`x!_nb-D{Q_5
z!dNafIb^W_zq_Gz$ryWdcTo3Y3D~*;Riw{tj;CILtzbEkC%y~BG^M!ixt+Gm7Z2d6
zGGgrskZ|>&@j0x=XZ-+psY{!Bi_Z)g+@$8}dG1gs+-zt2wfDz77D(3$23^7Mr%y}^
zi*p_|#{T?U@>m40gghkId|UL|JWDf(nw*HqXZV%<{W+x=lWu`(x0&UT>w-n`W}P`I
z762>XX7N3upNKkY+!Iq$z;Tr9@o@;|0}=W8-;*DygeMS0N~ohbeMvEEOtc9!7-S10
z#-=GlRh(Hi3<trxoaxOJJ4HaSP4e7I^E8qwu!mTOAH$2J0$uC^+u<KyQ}m1cIpHoa
zFZp`@=ss7NSq&cBg8Dtv+cDhpQuE9uSm1a<QZ^x+<J@qJr(6uokLPZVF%(o3JMnK8
zkkz1;eHyjsdDG9Pixu2&ahyusgCWBZu0@VdYiQ{9aFKZIGWB*VdTpQ&O)m29=<Xr=
zq(v$NjBQ7wNq{1{hxRBjcH~v2a0S*@_&-9HqiBvU@7bO<9HCl8O$ni6f@u*RG&JfD
zn}tLj2^_}AlI1qe($JU7@k`-InGVoCw;yid$)7_VB{S^PvFM?BWOQz4&z1B>jM$XO
zrIVWCS9keuWjO6$KF{_jk}*6HC21B8I)D8L1O8$}3y9|8Pt-DPsuW({a+&d_^h*&(
zLcaaz*036OQiB`L7StyM;6Ka9Jr8EegEAjp2mj;4f<5|CE=TK94QTj3%M6JCd>iYA
z1g4+o?kk$(Gh<!<IFtR8g%%rJ_dX9L&%>ZBNbP7IL-afctxYFkpY3@)oTe}s#WgC?
zZz{H$8Yhhy1LRwI+ZD$ZR!ttCueYU3)`;V)%GEy#E_PGXwZ>?Vjkw-252rI1pxY<I
zf2CunE9rHhO5dX7WoiO?t?b@Sb1ZiD+OM3jouQ|S7jJP9+=KLjhO<{7t^Cz1G_M5E
z;a&ths-_?l@~9ku&YtD&bScRYKZw)+N_)aFQUPAlI%Hl~7f4*~Qt>4f4Iw}+yS;^B
z#Q;_G|6Qg5Z_v%>M(HU<0+N}E!&nlr(z{=)tBRWxPn?UHqwu{cWNc4Oc6ZW<$|_Lc
zm9wK^#k2ng_^W()C=WQ4kD|T@#TE2-O8N2eq%Oa&T}e(wQLywnIL%|T1~C-gbZV!z
z1w0h?wE}$HFti^1TdQThJD<ph4P0XZ41<WV?FG|$C7^591NTTmDlemz7gB;ANcj74
z0^W;=8mx8(0LG|C?x0<}N#a%JwNe88KSw;j6o{9TH-?fGJNE=H?xJg@*w!$pW|fWm
z63eu0OX*r-l1uKdnA9%zsMHSuTVDo%vr;z-5g`Fn@>3D}d1+qjQI_?d^_s+3$N;6a
zO8G{YP|>c<6#IA<?cf)h<hR9a-qD=j0uOgQWj#uX9%rPMljN)sImvXH-pB+Rl1a&)
z#u}g5^1yzp(Qv03??-_KhIooKyVj1W`Qvdyxo8IEOm%o>9{*#40xg{bwgSwT!of-n
zVd1QB<ogSVf$2xd7<3rLS!B*>Rlw?0@HrOtsrU7H%5Esa?{c$4fe_d#w#3k=QfM>6
z8Rv{kDF9{dvQnO+d8o+1!y8rwrr4U$FTLO7``%4>&jvihF7JzCc$#!Q-AOp_b7kZ*
z%#00Fz45F47GVhCD4G|$bUk{C)#Ft9k|w~ahh%s>gZr|1zy0<@e*-FZiY8@1FdR7-
zmI5^+EGZOv7>Pa5&nyKPqn=-PH}$}sjY!b*v2OK`>TCl(v?w#S#}JFTs?5LQ@G5BX
z0G|atSsb8v;Pe|Oi{j4$5)KAR19!Q`Fo>!ec%dBsn)p|<BZngZ7Ww^jRSWIK#a#V;
zeur9R9?UW|;%|i*Aq$H*S~@xh0oml~5?Q87OyO!>6^ytoPV@1%_4bBOXX-3`;eJi#
zr!Ian<UhZ^Y8!i>-|TsJLhEvW;evI8#e*)D2*g~zf3<x0?df2ack+y3sW4?$vAW%o
zk@coGMzBipc3+t*ughtSL2<p$BMs<C)B|&yumxA`OB!LTe#5UQFs+7V3|s8lkF6bS
ziy~!v2vCbc2|@V-2~i*m9xBI<#FsO$3Qog>QW$klOW5cU{-{ue1klY_+I<ADBTT{l
z{M!btKr>%%)qu2ml2E|3YL>qXl!2So=?BIeT@?&~)!s<0VaMM5()Z0SlV7O{m@kZB
z2LIM1eX=(w8cKUHrEmk8t5in7W@Q5Me*#Z@q=iItT6F|}iv=YoZKlM6=#LHtFL57u
z=URofiqacOQuJsu1de{@%VJvhfg`)}LjS>TspOFaYBUh-BV)sImxKi7dkML{H6bFz
z9TKx5rR>QG{|>SL+QI+%mY4~CzrAJtBVn+o;Da^xi%kktY+5S_V{<_jnGzjCJr%dB
zb^QmPnY`2b?9qGyw8vI;u(NBr4Jfn=8>|hp*Dy|7W43OO$Sk*BfAYu48$JAfrtp@{
zxkNYi=5%O=HtXlBY&^;DpT&2ll`njT2IBT_&q}MD6q7>`M|@s`jL5P$bNFmkXCuWT
zUWfy3V`bn2vgPZ4EwTT1j>{s{38b(b%hi=5Q=s-EPaoh3+;qb6m`S4p)w{qz0-FW#
zNdQ=R`MX+mq58fr6uU!}?Q#6e5X7;Uy+3Kqq}IEX0TEJFoeaMppIJw#6;Jw>_o}Or
z#0hVb1_PeT)+vspD(1J}e4Ui4Gn+_&_);7L>7a}6Zg252c{cSQj2IaJcij#wuKOsM
zjz<vM|MC)&e{6c=FjZ)QT2Y>^<O#aqdL++oyzKcbPJXZ7Q5zn+F>FY#B2n|rN~=dL
z@C|JnXZ%pCZTCHo!^WrsZS%{+TQC~DwEA2zygWP1{OlHvepJsXs%;DeO+1(Yr`#sd
z=QL_9`#%_L#6HJ6tRsX5r9APOB&l&O)xG=vncXyU=XkSujp7X|Lq}Pe%o+x!x~^^|
zEiG+eJfWtrYJP0->n4D$G9%UMdl4!`*`%%pYb>rATb0Q*gSy*|mgZr%P)B2LDx{B<
z*zogRP`mR`@}0H~Q6e*@4NHm3kuBoR@4kAR0uR~Go8kO|Q*&_Naw{Pto3e}I_b^d*
z?W2f+9I(cbLp^sa->ulEd}K(khn$H57pf_`-WhfYKe54`a~I$S1D7$b({PNTzvn@y
z*3c&vPn0_G_iu($mhR88T~zmGU74@RVj$au-tG1Zg#_w)n(A1ZWe7ff`7s}>lnk}a
zw_j(N`^ist`1vBZD~@ev>r2*~8#c$150B?2ozgS7ALX)4$QxQ*ry8GM+~4e_P2ia3
z5k#_TKm;C^eD8bG-5}*`!^rTzBs<%KdpevyVq-v<fK-IqBRd&P!je;MA7qB@9xVO5
za5RY~TRog+KAOt9{0OHid!!<=TTW9OduX?NH-7*;)3ZnSSD;0cdG_e)K;}!$z@W;z
zUlX+8dwbAJ@Nl>rEBai~=jL1yjFFYVR8@gn9}wB2C4Wr;MKr1?^MZlyZWO_aEWh6#
z_15z&+acexybeowoPXv-^XyW)05#td(3BPrqB{2jY5T{_uW*kr@sN*X%k@@sWx&>l
z*P_vCP98{TzsUr0_+`RDu<*TUQ9?jzBp1lzW;asSyPy21Truu=G&tJ@O!xN}j_r+>
zGcu1JR_h%=pl80zyuQB!=w*PC!S(D{zo-8S5CSxABZvkPJ$?*=cfe~Z$P>tfPd`e>
zgA>|mJLm@HivRMPC}kQ$hAT)*w#zf!mgk+?N1yUZ#1K{Ld2C`)y$o{$;!pBA=42CN
zKtjZ1Ch6^b65}NKJu&*1i!)$$OE*0Mq(0WYiKoBR2RQ>!O4%a@#|7y6l>!CLvUkWi
zD1!5UX#wC<rpdSm<}XbZD{#KWy=VpY4liOBpRe)|0tkyIC}1!b7&P)%#Qfe1pl3|H
zMd96l73wmU#cyhrI9&Z8)FOJXkorn^?MAeHJVwO8-xuRrr1=r%vVl0K_F3NTAh#{o
z&$R^%EyWrL@EFU7%(&)mfFAR~`N06j4a9p~0Hm?|b&FfA9%l{FrnT4K%(8u^0rwDd
zZ}YlpFQOC0!--eN&EPjcPR;FfPlN{WLcD$o0?ljn5H{v_V7g~&4WiJ)k>7(d7~ue&
zSfh|LVf;=Pe{TbNNr6!y?!3KFtz3NR47*M2T|pO5r7lU}AXcF+-yy0I*p7U^1T^i6
z$9?n8f=h_e=3Z{W|8hQDn%@~>RuFng0J+*s(Wf1@Ga}rD_K0)=!#`zEKF`g1ytuJT
z%mwBdK#`5447iBbe!;kTt~mN=pinngs$9%-_gF`p@ktT^SM?u}7-4@{nr(5crdmA9
zWn5s~GQ#c8UuWribDp1ym}}Sh08)*`L2cT5n!X@&-f#wt8`^VC4o;`7*We%&oIwb*
z!ndKHUE|4MvxvuIw@lQkG59je&At^{sol7|Fh&USb9rw2DAP&9_8<@3n-VZumdb6P
z5X@caYUSo(>g2;agRbi~+5VU|+(B~)lAO}hQ^(@5-lZ4or~)NqiHJUYH*H{~$8}G*
z#hZ5c)czru8)^%70E5f8%74%H5&YVxf8ITSsF5!?`mUS5FO}JNda_WqQBVI-MXgWc
z_=^ixTc7>+_20Cb#rT4+I`)A8XJrYcd`8J5P=XzZpcgazxH<e*o8SEdc%(U>D((eC
zd<}Fu-TgVDtq7J*gcqynyomGkyO)ZAdiwLL-}kq?XZMi2Dy$7~Tis3x_bkvO^O>f;
zH?G|{#<bO*oU~*K6&><kl<xQ_&ZTNUBES*OWOa^G{S4?uc>C(31bze7smp>U&pmZk
z|7!d6fxSY6muh-CuUpmBQO`SJ!I;{qgI9g+=aaJXX|Eo-@35{|a8vBR8$^nJpvtfh
z>`F}<{e<ndu=ATEYCBAQ?y4Z2XzWqE{;wAtPQcrd{V}cCs``kRU8-7J9DpFfL9YZ{
zB9l7zR4R{L-kx^M6fWi<VUZ{Y#Y<3-yx8T)jYxzGDKS@4Q8F)>EELYIN{p|fK_{Wf
z4P#N333|_(eeaijaJbHUt8L}ub<=m&-m${Ha#h*ZX9f)57QFcPJS3xs^DcDj2gX{C
zu&40ygvXd~1ky`vUMG3ZRA<YJwK#h=7EtO(6KGO<UJe(yd3#qWZMHu99$nqRi1rzo
z6ts3GcnhWv1n1|Usk+~uD<X0L_)r)~pYkp$D{|Hp2JRLbMu?xk;nKWHP$0H89dFs2
zK0tx@5K+oe`pgVFVhnAEkYQpQ$O#EcOaQaEwyPgBZ2K_-#Dy+Yd1)uW4-dC)fDl_D
z8QU<4R+R<7eBX}esagp5cA%b2xbN5YBazy3SB19u)}@*v5%Nsuh*K%dkaQ_%%y;&E
z1Zz{gCmj1I67a%bh&`{3rx6UFHgrB({+<)ES>t$M(|tr1&DG@&YG7oo0>+Uuu$U5y
zblW^#EHcb}NG1d7tTn%z8Uj7-m^#bGCC3OheNt}O^-$O!cZ)-$dCKOP>wss^dULU7
zh8jmoU&YVhJ0(&G;0P!|aJZ(Y@b@QRwc&$<h1tYMD%lLc`=UU1E#&RmE$_`ekUo2>
zm;I4N)zJJU8YL1b&#x*`)EB60;+}e8Cc8XdT8SKC<YXf#?!*eD56z<~aBpDaULkAM
zsBbZJr(=H^IY|z)tQvBT_lFaOjg!@#fxjRZNf7Q)rV^)KUjp(KhMVY`Ij_AVWyOqu
zgiJSrp%}p{iAY>SRKudz0=B*YS*fT+gjPZN)@e$PGo+9ooOoXg$Q)yt?N^0@;Gzv3
zR1dn?If0?o({+hA@4QBV2$UV_mk9@cdtW{86Y-#dcZD-eAWU{mVua_PyYqFdX3Yb^
z(jci9hAt|GHYij|Wy1KJNTgG}pJ@7&75K4cVV4r<ySWv=BUW;(wW3apYJm8f^!hfu
z{r%qHh&PI<sfh`$m;_7Mui}xFJ>i8rV?7kHI;8F%%kupMsZV{8FIaVOWYa?nHo4NK
zzV@k~55CEl<AaTBDli>b2oR7KTVZi(8Na*prpjyL9wSFjTffPPP90hT(gFd|ioOem
z3cP;5z0Z~pFNq=>*6uovyAUPDuMA1M@x`FsLO!X!To^}BVurs80ms$|*TdbJ?hN7;
z_0n9}Pn%G^Tl_Omua{Ix_7->s^7!AnegqeN;ePBGA(Kib7ftzm@o@&>J9_(EP1F5y
zccH$#*5}q$_tLOG%)6qpBdOz%b->F{11296z>gKA7Nc%OE9kYxGV89@7C45n=)3EK
z&FqWEJh&cGaejDA?OhikV@-Oly&#V_v^6*iYPu9bcG?4AhtHN63xeLi?4ORQ>mjWN
zwCVjixa1L=FA1e*3SQI2vUz=_Bre|#_X;}l30dc#Ut&ezz~f>bd2vPeaM;Ev7@{>K
zCv5^b&bmre4$CdyiJ2;XsdXl|(NN&|U9FOqV{&qO9d>l>Da!G;xZV3sh1<SphH<^U
zpl8Lt4>Z(ad|Rqjt=U{Sp|ceC23bEXuJ&mZFT*Kh-=nwP=yX8D490W3O|!dvEH`Bi
z)#2S2nq-)To|JjQ+r`1P7I>Td(nNZC9japPz4iq9<PoEbA)|uGF`GB?=ymm3@1r<(
zL}x@VOEH82)y4&)hiWf9lSBlWH1|iQ(vU=W8W5aVH+;z)(710AEw)P0tB}x&25&s?
z5KCZ+7)k7!;F(tJ<Z;rAgI}Pb)_}(7w}uA<d?a-faQ1B=P(;@~>EEz1190YuX)0*O
z5KC*CDU1C#B13$u=c+=uv1>9>r7~SZfb)uktV+YzsB>vrk9X>)e&o<heM*hwEamOE
z>6xg%Ox?v3I%O^G&rc69O<#W6t-gT>r%gkf&bVU1vfv-~mpPJ#B#6}2nhs%q8Wtlz
z&a<7<b<4EsYOVU-^jms!25VGUViwK)L<y$=PxoFlCLj@i(EeVfQftN`EB9J^McBOs
zAYPGC1A<<~Ue}!TeqGr!OQ4iP(s?S+paoe5Na&#^pmfOJ7^;HLAJcK!kRVeaZApHi
zFn00f`pLZ4-LI3&{q;-yGczTWs~O7jY{O;jo?j6s(k|JxUmZ`wnZq$no^~fs>mys^
zIVv^KK0j=`0wVMQa>??qC0Bc!$xs-h_ku<!;v)I-e);-BG+$h}%KUr_Z8yI?dn&XF
zgEl>~MSXh_Tnl?9-iR#+&2*S%%kkmcq8P$=SPwY+MRGD~Xr?Un1DB7!>mJznwBK-{
zB>d3~Ab))c7)if?EVuh8XN&c=5j&vbvPvlI?tM?cHO|YZw5l4+H<Jg(mk+<~1jC_|
z$YorbiQwOX(O&byZ&o|KYvS3qTSn<2%PKO0+qsTY^!<JDYMy$_K~4w0YZK?5_Pjat
zQEoKU&iee8b|E}7W5`!B=!vTUAjver)+9%HT1@;gdE00;un{yNet<-<v6yf5UHr5G
zB#u$;Dj|y{jy$kMrAaSzafdik>O~;-HVjQ8&)4WM37M;Sdu4c>#Lh@Y0?F%nGBm?!
zXFDX$ChJi&caYjbbWAWThCK3OQ$Y(J70fECc5IgrC})&dk!*w7zIBBl_Oj3EkcHK3
zQZ(=2;rb*DJTtUy`~Fr6g3S7II&`4gHi1gTy9uTP%tm0A*oGpdjsvkb<mPHVu77-$
z@c=NO9~dWq?1oyqeL3>n{G$>#*2Z-L_8QDmRyY4=eX_t7;t<WL`^J@4k*NgBG~yrG
zJ5~H_OFM3nU&JC=b!~a&Fb>@9vm;LoK*paH5g1`~uH2-3WeJjXsyX?TblZWpty)lN
zI9Z?bM0+sIN%IUVj(4nhpUI7_7gP{5rAV}cBsGaJ+6uQcC>}94g<vLt0|RbBB_}e%
z3z6m%qS7s=t`8j<50?;xPgu;`0gRAR59<lZ6(vYEorLi<AaSBe4jDxL=?)7OBmmH~
z$tJye5_U%gWJ#%{q-UI##USk4u3|sVGoGQxm*wqjWwdJjYES8A=ZmX=DHTRspx-Fh
z6DB{LDNjk|aj`_zgpYK4?2aY9j5{6nfLa>6Wsj6@YIvo7Vbvr*c<cWQ@LYXNL%f)5
zP+w)w++{sT5_yQl8a3qKpq8q|B2=?5R&nxe8(NFLyfAjhreMkOczF*L6z)uVEtFBA
zhA#%z-vK#+BmTzc#>k}x35cU8eA{RyBE!sD4c0WkWnhzgF_YKrJwWCn@^SVo9Q!^S
zYh&R(9U$uk^~dX0(|+#AzpZ*Gi5zbd`;9oyl<AG}LT{+9$|ySMIlc5;2EUz2`>Wjr
zi1#Z?xJcYM3Kq8F2)&f>+e5lAt?LYjQ&OQ1ct$5i7g`U7q3~LvvYwuC`54o=qLh^n
zKNopC9BY3a`3?1dn=1$>h;Z4PzVAgFCn(h+6?)#CbWlWLtXTN|^wmoPh!P&uwNyH}
z9IORu1S9eOH9PA3&D9apQoOYA$KJqi!)dI9svs0ubUJYU@W+kpb?K0J9~_vdPG~j-
z7#S8i^eo20nM+HuF`=;WyGfm-iC=}XU=3lit>e%-q1WX&zi*`b2f_>i3||^do?YCn
z?pT3X`7E&sP(m)BRp`&=x+=)nNv(4Rr6asjXmqL|9QMy*dH6jv)<_2kDsK(5|MZta
z3T=})P}j=F3_;hfL7B)=Z5#~`zZU+gB7@$MBpBxM6c!0J2p0KcppIYOiht`)urss>
zQmGBR(!_LfQ0<F!q{kv5YWtj+#_xS7k*5^ieMx>a3%nJm65FIHfm7l;HX>+;YACe=
z`szrG@~3p_Rcz7;zVLD-OgD64G8;zoE*H|dZw$KOPRCy~lXDL4&d<+(HOE~PDuIN*
zQjDDbKKxtw>s)kMuoq>soGT_yoRDodu6!A25{vPY!m)2WUv^+_dR9_}Z{71m_zt*s
zIVHZN5#jw{R;f`d606lH?eCme%Ya_>(mJp6I?9AA8}RNhaMfMjog^k|`q;=!ZC3NX
zs||0Zg&VFkmDr8g^2S*w;24eLe5T39Y7JVWTT7#K6Nqz6qfr>>uLf3{sA9EpKMKr3
zp?-DH58K`%a!g?a7;zI->FaQZ=RSSXEwA1M!IgbrqDIVp3bN}g?zcQ?q51NuI6wOg
z5<_YUg1xKpQ74EB`kgcjZZT+z5tM$;TuPSTe?tP27J_@ldkUv`M)4+@l#p7j3#4Ex
z$XJYg_-H?YQ)E7e5i6!t@;9B%KA=-<F;$f2L~pHkV%5}eLhu>*B6nzB%Lsy6?M&;k
z1r05!(~DebCYtLRK!|An0GLgL(_e;&{J$`c!fVpd(i!IeIv)NyEQtM0QDbpc8tk%j
zd3x3&g;wD*>_tA8s4ILY8C;`n8(I^1!F5c;QlU(h+4G7GN$=~~W=cfw2lm&kXHV>8
zsGvO5i=)8MdWV!slK*)D9+!eZS|vqyejCu>8|N*WetBy-mz;eC<p1v*inTiAgD+#>
z#Y_q@3=PCyahj3}bix(DL3>gP2XZQj6T?`bh><$lQCrSdSfKwJU~lQ{qLb1n6PGLb
z>PqFpNo9hBh)2x~AokyGUR=Us4S?3j`*+Cr2&gCwr;J~r!=L+6qtGh935$OLNBG7^
z2ehU3Yw};kvK7Bh`$5eJFgUu4<Tr!p{?rBKh?v{~>xC(LS~7KFUcv&sA{QVtVidta
z3lQc*7E_~c(|p%!m&Ss(0D~U)eL%N{pPII#R^$C??pHiZV(F)LsH}tRWztP|pm!rD
z-Um`yxzg7VX!t6OSn`}DEK7hhW*@AM!ed7_f7x7(z1U#Gs+vNnkVS7VU<yk?LcNkF
zFOL-bsl2*fOUUDmDREigc(7-V0Ja63%ae(ppfkmC+Yq@CAn?pO1ASlBCCr7Z3T3r~
zR4{-KCP{*-hp7CpX;er@35{Mja2M3{J08{g9RmEpo2i4sd}1Cy70OPYM@-;ydR`sc
zVD0?GC@Tf1{WoRtdCvzh?w|OqCMwJ4N^8H94u%;-d>`<0=;xWM=tJNn$)r&ILOB#G
zHRl5IMF6cG_}JZWy|`q6J7E!@zrS;gFXKVn699*X(#@TOHZL^Xr4j0-1_H&;5FLJH
z68t&A0YhAT)eX+rqTD5JP~h~-fr`9p%)c(!DQpjF9UU%~*m{4tO}iD0wTlq=I8cJ|
zILZd36opawMTQ(s361IcG_bhAmoPUF^En1RL21d}uf2h6YJBc+x7J>&{5GVEX&{`3
zK69G0go}8vbb+2uwB8WTAr43C9c_2h)F-pCNiWEN_bb#Wp~;XgzMDepXc%^pC3Q5E
z^rsnkr(JEb?Ac-h#76=92=Q4iD<YXSib(539FSFlMPv_A7#tT7zpx}IW1VhS(MNe~
zc?ZxcOyLf0$Wd7Kd$@;ovTiZVctZ0CTHb2=;;VCkzccQeNz1i${cSan8|rt37s(Al
zA13N-4i`<!*2(@8qo2X>7%Hk^T3l7eR%eNAsHkH=o~z<13AeM;^R1Gm&~>6n&!Ij8
zKDt#tLuCREUM}X6O(P=nY;`RQhw5H_ge!%mqgSxn>1H*W&`xxG>VsSMokS4!CBHSp
zFH!s_Js4Z5#WKpEm_GX;*<PIGX=&kAn+A1{e1<l|O3D$LBM$`sLTSA$xQ=`Ip4=~&
z?)0(q$r=*8d2z$ORs9Q$jHn^q{4hPP!#)!)O}3xE-~Wag#LY6AJM5qC6EwB@C8#ic
zxyA$fine92Yv19Q8y9q~?_na1RldGKMlT=gSaeg7{k;Kwzc-%GQ*x&?!{2Z-=JTIE
z()*4ZHia53eYDf!iQLd1rySNjQ7@J!JWB&?<3qb9s^E4%xJsOb=ymty$c8Ug;g(bD
zK<kd-m<Za*p5q^xER;m22r)l@)Up?x=h)kj#QXD{!Z#dGfA_*QiLLH`nzE?WWh95{
zSJ!U9J!g!S=43RCk~uf*{%D?Akh)AY_^EhK_dHTd=tw!U11Dv>C$}`XGlTndZ_(1Y
zT<!EKYW76s)_RF6v-Q^Wy;-)Z*X3gS_nfN{m^ukJdpZ7@S2YXD<;mXPp$<TnyF^;!
z*YSCUzyUHww)A80L>UfB%RtVZtWZSf3)(}Nv&nc~LIZz6pG7R9lZ{^7X$?XS40+Ii
z>13<<$#-HG4m!&o1*LYDumY>`t|m6Yy6gcYv@>A{sRHd;jeS9{PY`wtvJwqD^P|}w
zvQ{u>*nrV0Zu5tvWbt4T(`B(bmat(GIUGExV5oKQFVwe;XV9|6WDw-=5{|@*l4qkZ
zBXyG$5br0w{hi3kW=t&>-dKm(TdXb+JLa<Rp3^y+RtzI<^ROwL3zCxKiM@HVCH?h1
zZ0Yl4X`Bzw;uexD4xag2m@kYUi6OlLjR50u=t_RgZ%6D_v$$f6N?X9NNg<>#+!fFm
ze5fi>;wbCHr+$3&Abe+pc9R(<3maEa=kX2Yhj)B@LY)AR55Q1hlR6m8dVP}~rU=XM
zO7@HtIwN3ju{76jvJcimuLNhbG~om{HHwITLFhVHLvux%d7K!!HQy3#P3l8vQp$z2
z7)?+ZE7}=(SF&({hDBCq92`2@nGl4OR{AUz7-{yLi*?2q)h`(N#M2oW11{gfom9Yg
z;=o`Q!38d3)CuYIP#4tucsRqatzqA6R<kNu6UA*4k(Ibqw~XsivTHe%7FA{=q*|hH
zDR$)$7GQ0r-)w~%*K$Nw=Kd;cPxxFVT!0nC@dQ^4l_kvEW|TFI%_NIgsWU;1a4v>c
zVW}ewVm$u#xy8bGCy@9yzhyw963v&qlZU)9Z5uD17?pbGbp>L1*>GrLLA~ZUEW3pk
zX`|KDTFxPn?V6kk_0x49ax#%zZom4;c|nUH(E|Ff1XNfC7A>(%rAPQ;^mG{d7f;W$
zmh28LYJGVv(rN9O)^N*Jwj;wnz}h<-iQ#dMb5^*16Lc8iV%BjIGf~|VO?O9HxWi$l
ztBa88?E_FR#*v7DJy^7#q+7<~ou86+^`fn^;MSw7Jou_n)AY5dHu6E6S|Xw7lW~;Q
zJ(!Vjnv)d?_ZS9+#lvpZyAz4b*VPVNs--D>Z`v^-vNeURaQ69f;Dop(bII|86K%bc
zG~su`z}h4SdM>-I4dQNd0tce>;#<LDF?!C1am%vN;|j?QJyYLKy2-oYuh|@{(r4E{
zBA6<uzbY&V?q8(b7a<eE+AJ$=1wA##u#tGPwf**aNOM&MQQYqF{--gSDH0|=6hq&h
zzJK&uI^hOTOZ4uaj8J@WgnO`>ratvz*32U4gL4;Ojgv$u@R``ne(6W=_wq+QpZqpT
z0Ey^n{WjvZ%`Clu+3z><mD3k{<wx8<nZ#BVDq(reN7{WT`rxU4I?$lZO%A7pOr8`N
ziMwUMJd1Tobo3Xh)Y`QI(7>#f<RHdQITd15FOG;j($A|>pl|Y`HfJx9E%>8z#3y28
z9%d!Ae0)`|T3j(wstBczEJ7Jco8dK}D>@h-93oD7Y^10u8A&0|6EyA8bqX{8e%iX=
zLNu$uo*-EQ%fHuv5N)taNL@j7*|$4LODc*o04G?em>d}uP6aUvd-PXNSDSLH@ZFOW
zTDG=}Ow22SOz7B&eTWeqiV~6IYmG((GVnhqZWp*f>j71-N-TEDY(2PE;=K5U+&bh=
zP}y?8F6_)Yi&>yvOP@@rCU?dsk`!K$&(KL0SYeisJu{$JiPKj}qSzFF+GAJ#GQ`)Y
z)zwHIw39M+$jc|Wi0R6tS__~x@D#2nK0x$h3#CRB)V<S;iD&Xxf`8!Oe1|W^eNh(1
zHh_o^zl1yBNqiMDD|-1|F3Wd4-m(Y^y`X{!%w6Yf6W=5uPjZ!#%AWP;z-_c^e25Nv
zTc6&FN<Dbk)Tb)OqvWGzE7g?|ggL9rj^!k#<dkd_`Z?aHIR8%Sg}TQP&tAnRow$#R
z#5BKY_JVK2zN(mv6T`LQVo1TfQLSUU0|-f_hVGqIe>qdNj#|HXNyH!({X)}nXyI4N
zHiuAaDjajFo12^j=_dO&U3S(VyV_Qw46PO8>>Tu6jCk_S!FVolu7cK1ryta@1s|iz
zoG<OMNU(a;8VjtJa8$1JVz*U`yd>&INwRoVoiHP(3XqsavZfu9Tu4=Y)zfp{Jh^#Z
z8Ykn>Pk_Ikq-+dbty~Zog){AB2HIOyxqvL$T*l`C7kkN|^Q^%3h`QPw-JN-BIFoly
zkslR)ZVL>rs;m>Ht~krKEeQi~O#H0wYW&p7)kZFys@El)#Qt$G1|$mQxY|&174(F^
zj+1zuN=#+130b~cF=H_&&9p>Ci4ow*xU<V-@(!d{)%f{3ks@xP&jlV};&~%>?n6Fz
z7>G)Hut(bPSld@v&OGg_?kGB}XS2(~3%a|TwsG?~IGeC_tGC>}wps7nbgKyGI0R4v
zGU7R&V$~26mL_wGBwi?-V|g3Uk%hKJFooq@i$h|3z1FV%wvLNZ%23Uf+6Who_;t$W
zM{ox1%&oq~Jq%fkRyDQXW9)`yR0(gwNsh!0ljHnsU%5B^24(g2*KGL&`&Lv)*f7Iq
z=N)wPyx$0%rS>?4gHm5q7MVn*Fr;<(?vA^2g$q|W?Ur?Jbgy7`!rx$Fhp1G_UCr4?
z9QK9Tn|ziESyX93yIyiGwMJ1WSh0tZMcg#HIp3MI{6fgPw3?S`i^OQZR;BvZe&i0Q
z&cZJ^6Z2Le1>IUilBK1k89J=C<Jr!Ty@=}_IirYphVw&=NA0t3a-r}yP%h9u=J5PT
zZd3BNMQlM2F!4%x?!XAdS+OfRj;GZl27y-ys-UmX1&c`*!}%MKL$jN*X_@k7-pW__
zve_~WN=9--MhsZD-X^;tq^b^p*`?$yKmzl)v{W!aKXvDO9t&MS?DDN950uRGVI>;7
z?*-@sDg6h`12IDjna#gS9}dmR#`My~!im3qDcNTF?iM3X&op$HIh_MXD&|WYXIx-G
zze+$V8G1mUb1g%}8NeD326lIa#nQLAO%%Zji;*f8$|TjWeSV_jSH`W`dc(FWW_k1(
zu>ZPSgH>9)yU<>N#9i1#a;J7&XeL#NoZ$o**CCwRqT%uXa4YEMjw{TMGBE<(PqZ^Q
zT4*$#LQS+dHBY7_iQoFfr3(ldWND9JVZ*A;gOe~5$^raf{%?135UsNXW!+DcVwhoj
z?pwoYlie?U&g0(H3$j6>IroCDooJ9u$9h(fR}g-;57d=LnxVYVhWdPNm|@*XU#p+t
zUHpq0USK6+5(?>iF)V%XuGQQ)tFv`D%8=-t!;ibOF%;Q%+?S8SoY<pKn;1|!qF@~4
zF!M|DrH{(>(%ObkczKtOL?3(5ZQyP@+0f_*(TOeEdBZi6-?VbLfP|<;yMMw6S#Ma}
zUWOAtHT>r;!xSZ^ll@K{6VtGZbPY#_3I)|`wLGa(H`jE|>lZEv9r`kh5;Q?r!`@&G
z(vY=r#@rop{H|QX6T=T^cIuwn7ag*mp`|bv&y{n$yMIr|BDOtxdi6z(a)Ur|$wmEF
zi~gSbN59=Ig6GM8@om>S5B_MxWdW1s&FN8S)QxcFj8Pk%8vxVjDpw>jBt%ScL2g0T
zad#?f;ciY$lK}5%{1VT^_nAY`^mJEOCai&BI3`!&gtV$0-XSGw55|ix6VgP)!vht-
z%rOvi)Fiadkn0)oR86y}zltAZG#`IYzCcuoQkC*cSBJKd7!bB7xaKC24vZ84Jp~I-
zB|WT-OD#`&CE|e&PzNST0Ne7p9;S{vQhw|sp;$NRiD{j%<QbS?HKot!w(#r&h!K4S
zsRYRkG!Ef_SvjMeHGs5TVm6j#cui`s|E+fNcw9)8z>)tTLPgxxIY|s#2@$z6aPO$#
zW~apFW8^brJaT{}sH+hrMqT6U8Ir1BIC<+o&J&312AaR`V3T}TnjDOG+-7i4lT7Z>
zgORYp6_Qf0<jX7{C)9PmefJo_UTKDBFmUdA$TsBl92Td<Tkw;S{E$FmQ6gj=o2rlR
zzw|xDiYt<gA%fmshzRIG6iJVCe+G?QdSvm5BIV)-7{I0UrM)r?>~vY1^-p+T^BWL`
zn?^?6yd5<UEnW+KVQT-C7mJo(a~+{_Rqyh25CfmrL;@j63B3x8CL<G@VQ4%TgGN?T
zgN#J==p!);3wvYD5G~vbgoYOfHa#`iP_IqX4@AhI6<*}xojK~b*>Z@z+lr{&`AS!f
zgiL#1^?5r*B8LfJP>}C6ZwL#1EVp^3{B}JHX*Go7qgy~e6Q1FlCgME5Qj{d5c%^V*
z)Z%)8PW*fb$q9lQ-G<x8s#KRk+3RBtRLCcd2{e3BiK{uYsz`>}fzmAp%^UgU6e$c?
z%&=z6tr}Zo8>KcY2NpjUZK6IErH9bD@|ptT1Ea$)_#e7+)}H)h8$jVhzh13}>JnUk
z-w6e`Wl<p#!ZxBPfQbTm99duR-FtW(e-RNL9sZ%n&&V2pn#x+Np`ET)$x_#^e80*%
zW`d|iIlIsk-uq)p;eyMORcbYen0g2apZ7WMYr(#ONe7K<386eRbp$w!F4^2wrcKm<
z0;*%c$Zpdz$~(+364rPN(C{tvO|NmKh*;vGG1=N;_eEG!7}(airn~NNvE6e6w|a&6
zVria;Uc+y9mIpv~CAj;oxJwhCzn>!3|IAFirB<z{o7RI^i$*)bg8t%W2i)2u(>4Dx
zzE~U~qhsSYyltz2h+Q8CMn<q`(LFUbaAAc=kj12uCP}eNge@JsMkgQ}$a&Gj_TRt4
z6@F|gdqq9*b=s>ISI#z1b+89$;Te?C50U!6feDO!sZ3+h4kTPBqj2PhM7-M2nV8i?
z40YeiV@8<n1j0Sep7_l~84n6A21nZ1_}GpCoCFa(@HT*?fJPY8lM`RP`j9>6+;qud
z{f#(B&6*SS&AnFL7QF8s)(4_=bQRi>+!K?PTt-?l_k;+y6A&Q2#&bOnYJ4)o?OzO)
zS$f03!5r{wcP3V?XbF*Y3pl^DzTKnA#K<e8HI9v~>iF%svnI8kvD?}JK>0$0^D%2g
zbZ6)%0J}m1pXB>Bp)d*UEt<mSh-k8;Ij&R#alBFY)3D{C6eb+ta{uYcwQeIkLfP!M
zZ)B{oLf`<S;oaaudY{P>cn2us3^}b;EW;By`%Iredw77g6!+SYK0Xa6kWYopbj<JL
z-GL#lzDZdJfN`C*I`q;$eZRqS%El3P@>UwL@bos^p^J`L|Dt^ZjQu0%a!x*Ue70cY
z_-XhlKe9Dqk7!KxqnQP#slCoW)@Q(JfWL{LhDG<z#S?roRAq?kek?{)g^c?+XHER6
zSt9Dd{~_zGqpFPBZg1)Cl193b?%1@1Y-vfQLmH&PO?M-WbZkmO1nExcP`ac<8j=1M
z&o|CE@AIDj9Et<RX5II?V$R<jR3<7^BI<c*-f_>qd#9bxj7qBlXrU_Dyn*eRUMj>y
zrwe-R7_B<+qD1RfDhItzNkAFYIx|?Pm8+gmd#MCwr}OB$WiPL;s_QqAV`PTu@W(h`
z)RkO43xt!F*ACJ|a$?bw5qssoN7Jd*Gx6@zM$~Oqa!0C&N@s9u-e+EYoK+@pxBWQy
zSzkP~R<~lGs~ZLn_4q9k_)#GbWawBTQQapsCn#z*xk&>!3H7$7Q0D%UlIBg2y>~`F
z&mg)}UDLPcseGI%^+ZE5r!9LWS-o3B$;GpXL)gPze0(tiFfM0+nll@Sqi&oCme$!X
zpBjvkVV!COvN#`iV0xKyMh+@sV5t2-wODJeWGPmPCc~LZkS=5ND8ib4H{2T><N;Eh
zMnanJV2u!*DMt|vXImP3q${8ZnahpVX+}pI6GQ5H0H*{eI_Nu3)7NX>vQbzsty7Ok
zqvNs|f+mG>dQ*6JsyMo5K;)T?Avmy*jj$`2p4Ku$-e-S+A@Kd)D3CMn8xqy|+fzbe
zbpZ7_>p|gz83UH`$Hyd41sfiR&pkXpZa2bZ;?2N-o104=tV?fo6;{rpx!8k!%c2^n
zE4*PUZwZijYUD*mY1h<|FH1msBerl`XE8*$$B!$N@sCLb$sWJhGOO(P-50QJlTV}=
zL+V#;S<g)PWgTC76Z`(HIsPHut_DcGB!re+6cka$y5Q{K(J<a-&bxefYR;CbCFr~*
ziCQ*M2j@_)@}O@B%2G34T~;@LRgtybJ&k-Q?lo6Wxia)dBBi`+ar4yazfL29;uG|V
z6}93?4m=@6x)OAtZobbZGJpYeug?O+#!YkEH%o}TfwV|kc(b1*Qa$T@N7(YodRWM_
z&vrT(-nBSb+&DH10yz@+r`3sIdTe8C8hB=0S0q20Q_<|5R*-3A+Wt?r9k%(yyxEj-
zK0_^LfhM5o+v@FoOKT02;`e6K7#VkhiSfAt!^WefT4iiOp&HcFB$d6%Ju@*<lh3TE
zvofsmS4>iQTy)yQqdc(r3VYod1%<V?eRze1jt2>9Ul}F|dy|Tm2}AI6EWTheEdnnM
zu{}nfpBc7!2ORtfi#m<oXI5G;o%c%`v^d{2zLa5wij^fY1wmpls&gRsY2-@^a-mar
z`Q-A{89Km8S~19}T8OFjw#|J{Q-?4gPKK^*LehN@>Rp~D89%WKg{34q$!*GC;}W81
z4_KDaa}IBylAzL;>$!^T=V>y;x!1@VTF$Qn{t&(0Oxb#ixBt!GiyS2=`G@pKZwSy#
z%69-oE0E-nDbG!2vLI_VKUi*fNT5w>C^TGVmbMu`j^pV#7jP(Pw6kae^1tfAUDSWY
zE#`I{3a9J|J1khR*TvCD8Y}@A2D8hFQ{YV&_v^fYpdV0_x?(l;r`1D~QW^e0;7M;<
zTP#%OMsqF+nZq@HIzwt1hNIDmtrAoFeD;oPK19#*wQpP78%ZDq^cC7ZqO0agD9wKv
z^PBAOAU|w1Uk+{l%+^cR2E^4i<R&7nFQa6MS&;sbic^>6Q*e82EW&w#-}!GNoUR&7
z4C?FOag$*o>zKz(*R_qA=4|7Qvn|{5N5`~cogRa_+285Idjqn#f&ZG1Bv7F5E}^TM
zI29`vHj`jPJ|Py9zSCAe4!J&@RI0!b#rRp~s$I@&US}(*yu2n$2pm2Ff~j(|Roe`t
zh-W($H4eSDvntcxm1@I7EY(Z}XtoZw|LFgmSV%Ov7H6AIetxSt{XOYKyGTCg#k9KZ
zmQB)sDpkgd2!oCQ|5cWk3Z$+oYZt}{hN#}&-YP*6!*|tOSidx+v3l8OE9p{@Aw{?w
zM(mWivS_V@Xis2Ker1}WZirb?ofWDT^fJ$CP;n@Blf<NS!MvD_00lV==zGP+SPTdt
zD<}}pjo!#9dS`6z_`9UnI{G){SO+A!<@)8riaxWoUh^X>jo>UDVs*cizH_$~f-b~-
zDId$3_IY!lOYp%Gb40iOvGch2GFtknmQ{eECW}h;OnK|*v4{fkt7*EEM-q_C^6*YT
zhrxPBd~61Amc=5@Yaaw`q2|bgYH;sxJ3qv7eMMbaL!GRyqFCj%&b)rn*(IppSo!mT
zbmvjt0py@`yMtK=cFdx+2g|8O2SZE4_WNslgd<EKlx_6dN>oCwM#vbspxZC(7Av#I
z^3Yy3)V5@ZRZBHaBa7=>pb&QC{7f(q&1KlA=JT@l1=fxvdJnVoPb<k7L%2EtBxS^#
zQ3w(4n=@@tIQ{?vR?ws4c<oHHSBl{M>7P~D5w%tB6NU2F>NtpQK@wPo-FDp-;$iGD
zlzwqblbPtg>%R_5C;AV3Xe%?nc`bJw4(C0&4zkS3tf&P{mc?*`9$;hl3iN86Uqw$)
zblknK)o(X?IfWHFu&RXzyE?nrxvLxbcilL4O`ud1Iayd=^=WDxPl}aP`aWHZy}*7h
z1?pR*P7+JuvYNoqlP`vRfA4t_M*GfWu|?RUnwA>ZFNWYVL&6LLTt%{PMS0mNl|J^}
zq(WN!ygA*5AGj+vy3l9^Qo)c7NoUL^U9VUtfyR=1vUc~t^VNhMOG?KNdWN-d#v=M)
zpFd<}Rcc+&H93GZncp_0eoSnHQTW4$?MuAqVYn0uQ=J#ZK5FO$1`#uB1qh0jW{qza
z&FiMHAn15%xVMg~Yh}{|ioqV-x16sM@q#P$CU#b97Sd2cqJEIhh<+V*Ib~z85P3mH
zfyQW(v4`H;bTp(^e99bMfrXSm74Q0m`$(S_DUR+^Jy+r<mGDI1n)jQz%Gb5kQ5C1R
zOM2Ryr08A$>icOrKEdr@dBp#yw~i-9(vg}P#Gs0>C(q>ydQQR&cX{bCmOp%WDPg#m
zZ&68MZdO~1X1A0$)1_B!WD9o>&PK(fh02Ff{{0ki%@#3jwhiO6AdSAM!QHWFYA@ie
zUKhngeo^u1`cB)}7rfc#$V525;46n-GX3S>@VtjFe1K=QEc~Xsk@UZIpDb1-+q0tW
z9S~vB`jz^t@9qB;vi(QTCc}ytcwf%^{PvHN-Km@n&20_=IwDv;pCfmijhTmz(K*Li
z(cnp@Fy*CG){7rmkeE<Qm3wjWXw+2u_Y%lvd+2IvMIlN%<cL{D^vBL#zQM<bY^alD
z@yb=P`t_4932Xk67^tHjpsi^SfX4q>oIP9BpGzJwLpF>35Ky60X}?+5N=I5`M7qZ&
zM!M2*b0~fSavie*p*P)^kHWFeV!O&w4imV-1t37{!d>@^B_RZ`wv-$-TkR-Fzyh?&
zKu~(fxKxMW)!Iwq!^lF1j{Qb-r55L5+ASRt*ep9ubrEEt89>=9=`=pER|0{hBdo)Q
z@^fSI*m}P~G3O6ZP}t~2zK?j=FOSLVXPvz$p|F<JnR{D6*H4s+SW}HD&E9AWAN2Uv
zYB3;_f;0&@@Huro7-Xp3bBzv*z2Lux2ugRd1UM}<z52(L(p%VNvO`%`VYjpZ?Au01
z1aSKqymB6z9Yr1H!4?R?>T;YL+)c`1K>++1S@SnRi$3e^;vie_1n|k_)uQfEH>B=A
zBVhH80uy;8M|G2|4A=Zf^Fl{pN9U(eP@OR!ld1RUUk=a)qTjTknP(55y?_oMKpHqG
zHYt?9|9X2!iF}_OE{uG8Zs^3MD4pt$?P6Gv)?AUyci6Ov>mcyXtQoW<->oGa<fR7x
zpBDt3KXvhBfF&qpn9-?3-b4Uk7&5nZfrBCQasQUZKh~|X!l~Z2ODhcO7oI~&guHH6
ziMtce%GnsDpzMJ!T*ihV4O5xI<8)5hK6N>KE>Sd{;y3qq_44sFD8r=OhPs@N-`K8d
z7%KNPa@biI<Tm$i&oVkI4v_t1B4sZ2uX!yO8wpSU4*IKN(9tTtMozk}PAtRHJlh#B
zH5Q~rFEe`kmNa@c@=Yp--d}ATl3GMNE6D4D+>yKmGHofPH(7?Aq*xm=HmN#{u@fSb
zw=_qmIgL12tsr#qxz4}r5{y_h*%&oZae<c&9V4GVhAOns<X`RV>_jgu8R<EGhNNKb
zFK8v=Y8TB-d_%i4)=A47sog2Hvj+vm!7IrJRC1A57_Cj*t6rNpwV(n$0eh4zN<W!*
zs!V5Mt4~l*Xc)AARcy<ie2WV@<bjO!Xs4G|gB%RV*H@{gmBRzuL63%TBM@OrAbc7!
zN5V4*8b{SLkaI@EFWe8}${|Kz>as1-#YWDh`v<$K<xYJ6%@Hj;RuUk)(pfa{{_+K1
z=R|E9Kz}$(MMldcnEMtqOi6_?s<dU#Af0tQstg_puP@x={#&!!i9X@W7u+gQ1`|0|
zmb`-GB*i~RJjD09`bEDVzb`<y@qAJ-R)e0p`A?(T$w9q>fk%(;Asg|Z1<QnpC-J<I
z?uP|BX;){0!W8Tc3RujOL!W`H4x4@#ldvj@nuKwxP=qc%_jA76c1psJgjTLJlgwvz
z_1@XaKa$frNuuS~j5}8F^-MB4an|6xh_7Vn%8!~dG9r7VQLj*=GN6NYKkJIU03N;<
zC>)5wTBGmXEEqeUM!hVduR?yYhQr6Bz^_R!LgE~iI+e2@!h#4>(D^#yJX5b%W0DW4
zZfB7dxAcg(yh4kp^V#d{^9Wx~fuu*dKM{PN5N&RI(i9$~U%ih6A@VYhEKaV=YZU^B
zF$9k(a1CU+flf^*y?Q#QNTWmqJhwK5@W5X99sC4R+?j=WY59h7h8bn6m|{o!Df7Ol
zl3`FIyAC`8h!r9M-#uCztF=4{m?`wsvPCAfxB8l{{%yub2O(a(46ML8=_kKPFgAq;
zB-+gDvd7u$uC9&GDA3Wkwb>MA95{Ttm^17H<26EubdkXS>L_5ukhKBYUm@fb-UC!b
zqvG})()Sj~7U)~Q$Tw;S&Jl@?TYP-SnK<Hm3Zi9RzRdL6B4`tyacbx7jUo5Zw05^2
zY91hj6|1T?^udaTNaLOeO=?$^wBtRTM&HxW<eHm>#ZViTRqEgTT@p@NJ4WC_yx8Bf
zjru*eH^Ahu?Uz8HY=(TUr${Kw`B7(pUd)K~;kW4np{f5x>P|uaDY{Oz@*!Cp=~~}1
z_pl}#x?UNy%U-8@s(g^<Ey0MlOdXH@#yMb*@>E;%uD~1q_Ty0sshuDEoThnq9{}vw
zeIP8ES(23P`HDzk0AD517ZWjwhogsk4bLC5yvrg+mC0JpIw~8AVb{DfjFu&sX??N8
zsKq1lyysV0Hk91ju>4$%x<<X8HbTckp0;z%n%BLIIn|$(;4-ZP=s%l6#YQ5+(eYfl
zLvA-LYoG3Dt4v4wj1_utv+Gu}KC=3y_}A`_!ppwd&vUx~5=1npKqEr2$m}YpoU&hu
z{=7)#{LgP|0k;Q@9Ox1F5swQ1t*S<4S}yh;@ZyqbHf8?nw*|f?PKFT?Jj3uvxxcOd
zgc;UIV7%s0gNr%0IZ~%Z#SX3vpG1NTO%`A0PI>GlL4qu<PXL=^wy5niL<j!$%jXF?
zjYPYGM;RYMq)poX)4he#=>U~brq#aaXdrz&y+=@5o;MjmHZgu!Gq*zqE;b@j(fv(R
zzh_6D`^0`B1j;zbUW$J<B|fI4L;x#XhFyzE9l~BASV0v4dqe@4H*A1aEGcw5C~FA_
z#e9bN*liz;cDC(x3SPGIXfSro@ZcgUrXl<fsumetf-am9Q20G(g;qn4%o+@nR#}kN
z?-=$2Txa@;Krs0HKWZHu@e(ZPdW^upB#|C8s_vWEAo^Rl4=g(X=)-^4VIAnlIV!TU
zp%hK$seO5dh|EnQ1WC<7zsk#j=A<O1!f>vw@JqTyBU+cYz>kFMas|X@VDixXLYIPU
z9(fymC)+Kq3B(i217DYV+UecRB^kH`cw*C3Ott3`21n|6nRXj#0ljpJlS2p2h_edr
zM*)eSh~~Uw<)8aR)4sF0bi|>MOPg?SgKD1#a6ajBy|>AE0VYxNOZqdZ+haKm_8HJi
z5;`n25Y$_J*Cf2vnJ3*wg2c8ZUbj`^mIpW>5WQ|8z|}FG<)y+KHWTc?gtuqoF(izl
zzgf8iuDz`BFZ|YH0t4@TeM?^Euva&4Q@A1x1gstWIR5Vu0jv)F-7=h6oqii2mFd?U
zR*dMa#yCl%NxNn}FxpH-(*HUIsuBU@h$6kbZNb|fcb|@Q?lCbAV+%`a6~jhtpnsu~
zed5nXn0N-S;OmWKsWbWUY=-cGas@4pB2}HHea7Tj$MrD&QJWs)!dGkd7aN~m9SPp<
zR}dzqmA(vI@m8DA&OaNV4f|0m=l|(`KsxC`O5D9Hb8xj-mAUFCQy`_zPFhc4w2Y!6
zg#)fKQH9<}O@1sTjvKdOGv8(HZZ5`{<Rt#u>$G@&#9yO8dBkppOUIGpI?Pr8R^Euw
z7mt61qkJL%XW`4<Gtnh=l!&fK3%J4^9fl@kJ5~9Ii*Yx>lUI6H42m)MU|o|)t<wBl
zf-3m)G*Cp=J2)FlIWCFp(Q~KlIX3?yO-|WFMJE{>+ytX15+F!%Pd0wzqH}==vjl|~
z?=#XC2epLFX~3&s<1$frm#p}5L+0E+;ny1CiX&u$%EIKYkJz36S|`FJ@UNbLkXA)C
zL^C&_82RoUB&gI`8rAg4<}4%v$DIHos3_z8bknL;P2p7Ac=`lhI`10XY6b4>e$ni+
z$J)Dc|8WJ-1O_GD-CA5vveG7A-oAC~i6Y^!(YujIE{Z=24c$+($k*QjB?KU+MghO%
zsNWpAWY`hh9&7UyI8{TVl<mh^c2~i1+WSlk!rdRYh(Tc~XXeOHVIt2yxKM2w*`HzL
z9_7VxYf*GOUJ8KpW8=6Mr{&SC|6gw!7^BRaY34hF^+=YI$W)H#z6b(^h5)Eq?szvP
za3TV6wXdAtEpHr27QveH;pqrJSU-_(fHjn|Gi-=peb5D6V*6rWCy(iJ(w>Re-R7T|
zaADd%bffpL!<u#r+$kt$Hb;B#_+i2)<3$}5L1k6-ZsV4&?U<kTrlUjSrrd_vhWwp!
zB$^}sA)lTgm<O)5tMO}H;NW_Ob&Bp-;KRNuQ;@qY6stxnw7g#$FM8tgW|+QHa%HpJ
z7zB7f6?HA*6V&bj31BMX(+gQq-i|2>cTMxUx`P7<B<mYMP<)YS|FZ}}s5HDi+Chke
zK=KM{c=}h@S`c}ZFGxMo%GeHaM`9{@jJdMxK0(@V1tC3AyWkT4!K?Gc$X(8sG=!!F
z;WzLyF9e&P8hYIGqJo_E*V(`z%v&CwC7Xwj%V&kp?)^B_Zm^jZ;fT%Nf;CAGqqhR}
zBf~{_J)$l3;Mz%}_+gWr>WbtID1!j@M;`p6LY2W6qr`cgGvron_4it_1SjoEW&6cq
zWJB+1@{nm$nyGiRe{5u@p&>f49{xD@1g_!Cz#pvWmK*W5p*T^C&^?q%U2AAa0BCgk
zfFr!Spxn7X>;pL45kR&NbXE!<;3nxETZy;9S=MFxGxMtCaG`!*qq8Z}|9FpFk-?JO
zOsb}<BEm9#HJ(ON?4TKM;PT+792T{IAILJS+`S$8UzI!<FWM&G8-pd9!ba6s+$7f?
zLVLtrkKhbk+X_7mqRXXZHlOoB3mp?2t2#Lag|HuWeR-FHVgC+W2p3wd(=GHw&KYp9
z2;Ym9y*Fxk?@G9)7|YH_l4I%cwY$1qOfK6+(Y0~hPR{$@Z#SRw6rZX+%;s&Kt0hKT
z8YgqSX=S(F+a$YB_H%<6nY!Xx0_(pQyXnvC^E&Br6HL4|59*UIzjfWco5%~AU{1RV
z_?1dt06{vWw3Jz=ttRM11vEn_OBR1=2zu-O=HmsYfb3;);iU`GWZ69s;ts&#P@vXs
zmYvsX^yxy|`^4X&)+D?_CQ-i5EtpgJ%Q6N(zu2cq{+K=jgB358fyg90goH~AT@z(&
zaK)r@Zy+Xv`HXIP0$Qrj*Lg#Fvnn{z!lyz1&d4bC1^l5okV6rxnG1S<rQvsv{_{8U
zZ|MxtJlc)O?mLiI10UL6%;*)<0vN7-<GYNf&)<9~@5e(?6E4YQDMEzl2`c(s%<2`<
z7nn9U2j?#~>A<xUnIk<-lXZvX-Y}Y#WpVze1I!(~L>Bh?7kLxN7>ShhkgOqP8)px$
zDl(89TmyX@R=48wn+t|R@IH7UkhPW*7y@fin|1zX!>)^mFN(!XPkJTKZ_hlz8{#<r
ziIC?={|_2Tp9sKuRmSoH=}LW%%Nfaj*#BzxpWy4a=&DzC|CZ<3G8V5}{7PBic_jPl
zt79%`X__%JCDpy}Vm*_g!GI2ixPYFcRNdniS(K&3-hXwT`xPp24o!TrcadQoHF{+H
zjW~UvLT7~OC>dp=2QPYPTJv_r>6@q%78w{;98M=sv+EE7et^FJJWh2a>19DQF8p|Z
z|Kgxtf_wM)#b5JPzryzMJO{O@luzs8N1R2ml=~-Cm%Bm^-OYlhON2ZPfe(oJ6NU7=
z^bnI@UI!Yj^&kBN#UA|b1cSVZC>fwD_!Fiizi@EE|MZ!i0(dulY^=gxCe@D<)wBCk
zn=cV9xk(ebO^{8>U^&j+7*p^-;A5L|{t9dT@Q1W^D>m?YuT+(z1MD&e91Duci)7Y#
z2dEhO<ES|tFwoB)-Kpp4?9fmT=zJC2$XLY4tD=9vYfe?qnhGn*2tHg2pk9ga1ipcB
z@goUa!=P_?A9gW>&rcd6iCEZ&N<d}L*Y?9t_JzKi=8=3Bi%5C>v<Teg_qg-}F{nCR
zGu78lv@e2OZShezM1{PJeD;dB_3`NOwXXo8m|XrZ!P&oGX{Yq>0{;&Sz_b@!F!cRy
zgk66`O1p&;k1$?3M<`OA6R;ixIae)U!%pyWWe>t)5#a(Sy%Zr|L0$W!!u}q2|Jn!K
zIBKDnzpGnbVgpuvEAV2Q9WFH^-bs;`>~GQ;R0vu07MK{J0)t~ODuS2kEAP|Y>C*Ta
zeuZ~-{l)*&`q+ui$Opzn-q4>qIs<u^%}y)tA}AeM>1qnDws>XLRWxT=QhUC>V<YGZ
zBT2aIlyJGfx+?mn&-jst=h;#Cm_eIc>9gFxc4eag@AfhSK|O3LeVLVn^CD4eQ6&~I
zO<~%&{(ZJb-6Hisz)&Y2OQFU;QJC^iDI9x}=)~$fJE`GUX2(EvzYwC9u*Jsmfg8Ba
zw6c2kkIZ+L_+Ed1%DNRQX2>kY&s9w;00Wz-)L8l}MrIA&)hE>*QgQ;0+1U>DIn4Lu
z*2pg<cyr{dXUws;OClvqqxk)d{G#YP3oF8j#K9=vmYq{3o>KUwc6Xy3Z3(?pua@^=
z6i0PJqq2g<M@3yQ^u_KoYdnQ@f;b5}2Ua7dK=8axnVTR=U+*&1oAyOzuNTMF2>{_Z
z5y;8SI7rx&@|`MrF=bvSLox>;V0y*8<k3?lJVxCvUaLQ2e|flEGI9duQgJ3lqJGIG
z<+q@|5KzjYHE8v!6>tUbyrT8;r|6f*d_96#1>a1wc+JCBIsHllLHAPZ4RWnCfdSjt
zXVmu1+qe1ufK8!za#K_zL3dDYO>`$)?(lkx2Pc@S<s*;jOP+<aXaGci^`a5CN@|};
z3*#}JycfhX$`$%t$JHWG_L6ey(>>ztG*6q9+(FKZw(Z}P{BU_lj9aJdwhjVj5!~L?
zLj&Anv}$CVG@113r{Is7S-jfFUbY1vkj_rQc)_}=JSqAnSufa*3%v|Fo0@SzMVE{D
zo34gyv;`=F%>P<9%tw}AD?*XqFOfZxHHa=j8JSoIb(#d3S;^>siUI;Q%bb<kA!hMA
zb)ckDdt+qwST6NK){2r}@(G&`+J*cEjG^{AIUoi7p9d5=MxfvJEo9|&A5F;RDo;sl
zWApi#srpK!f7OQ8#@E%45!J?Lkw!@@VR$?Ro+kGigDY*a5*OM8iJwkw7y@ak-ZGMo
z`Cd|-k1>8);o9W|%UO4qn?D)X%6<ZD_%_R8asqZm%+N`~_H>a%9k1jK3!k#BIO=_u
zDjyGzj$^2{3M09ABlP^JMvp~FE*vMh#)U}?y~Q&v!lVr|Wme+g`7@e83DL!pcf_<2
z9EfSAxxlWruigG&gRl|I7XF0DOEP;3U?WttAjajOTb(j@M1+JTzkc=V7PTz!{_Im7
zdnBlomcLG|15cb;$p<C}WVC)_7;wx_$F+5+j@*KPRxTV*N_0WfvZq3}Q@!r|cV9d1
zzH56T5g02Zcp54SY|4?p{6er6Cw_=8?s@30-5T{A(Ul`Y4f5mBk^b$n-#fJ8DDZZZ
zwq%$(!WsqT^hG<|B$||q5ZrwJ__KT@e`P*a8n-l->DLyB-2dA|Nijvkqd8!E07={&
zi<<&+37Zy&oi3{)K*g19m3)5UT!Wn0<w)lPahZvSEnP3z*IW(JCp7R~UJNz?R%QEH
zc-=NQmCwGj<enkhhDsoB_(+1Kcrjb>H>D4(xThwdJL?6VlQn!}wp?ofTkCsQ<xNDT
zS-=g~DKfd16<EL*bBeIPePmtc96hedy&Hhip|db})S1S@aezR$%q<QFeAofao&S)S
zyzosag&o!4e1D2V%wJ2kh28B)KkyzUgBzsI+sO%p%zYr_j93W3zFY(!h-qf1U2fjD
z2Qx?HAbWh&i(Oz(%!-<VW=jAB7Yh2Zr0-vCs~Y-aMDdQOeXv_fa`fr^67SR3{Q&~1
z#v%oj*#Va^-b>JAmVZz}dxPik@CfIZ98l)yeY~6TNst}Ey5-S(qIwqayXq@)zxe<5
zM9>k#G^8RwZ7?a9KsUfza*(ljLQMS1MBiq@&93&T<hKdN0TZvcxgx((jC=w=sqQzR
z?Outd<<6h=>gfHN7QaX2wNy!blF#}PG_9y@1A}CxxtZ2*d_~QI3xlsHSSs=(AFs~i
z_%tf@2?FVl?(b0BdWK5{u?6$-9E#_^noSkG^O+ymlB(H}nw$bG$lL@p!7*)oycYsX
zo5r731BF9g2Id<Gw)a=LX+z(n_s&6rF&P}I_kQpOuS^cl%U0eJp41oh;)8-Ap_J9*
z&tWr5bnAQmHnmuMT9@cf620A;R5q=sA=S13lZ{5TxbeqZMxxtCL|h;JBRRTR0FIqA
z`$l{e_Rd!|mOT&&N}r5w8;|AD04k+%nuGnXlnSr($bDlf>C-<v?cb(FXmUo)kg?hv
zh$n+IxdebM+^~e&{l&Z~vo{j5O4Fy8?iH`=OJpc5fF4GQ3xinz2YnRC2i<L;dzS%s
z*OS2+{Fbd#@{K&r&CTZp4(=HfZ36XZ80aBc2L}gy1C_LNV?!yNuV`mRYGBAzD)K$i
zC<z~>@6NScz_;=lrils)7!V_Q&4I32TB_K`M$rGHjCS|{m}fmJw>VYtNN&A2@MM?I
zI$=n4`x)}CdkADOu{s-x9SrVFd|vj5k9=7m-}SMEvkW&1mApHjwp5;Y>MP{9WCLH#
z_1P~UZhiiKh!cIviyq(08rXh+Un9V7ft>gwgwb8z;+=jo05KHrIrC_j52+BbA}Yl^
zeYK6Qd<>*J1dmqgrX9uUJXRdDf7{=)_7r;ZQA&YZebuuTgk+>`BXpOIW(qqwI@;CJ
z%C$MRUCi-a)DO6vQo4O0Gx5q7U_b2e{d<tFP)0K4S``7t>Iq3r`3Y6ugen^nrD|V6
zt)T4#cdWW46!XtZ<h&PaAg@KvX+DTjTtDtBI6n`uC!L&WFoqPkWxK@VR!63Dj0vLO
zu=5~6*B?CzM78@yEdSl}iah)5oG&fQV)A`cN5t0Ai!VX<=A4aKE__%)Y{>$1qwnU3
zEb@2Fcxmix^EnNx>5{i~{);%2SKiaitsTP1T*676;gBN&m0s~9;^<d}QEGgXCj&c?
zVjiyLQYdCagn>=+@XH2Ug@&*c)iY0gMfyJ6?SH4gnB+IU3uAnTcE*O^U8F&>R?mj&
zHMSJvDbv)Ws?>NQ2E3zmGjhbOC&13xe^DrSg;(K_a-Zs-GClp1srE5;y}0ETEn~?I
z{{t{ORBmF|P)o;uZ4*yLko472@|o1ipI52UeqQ<d4a)vG0Bgw%hRLb(I{#dngps-1
zMRJf|%-DW)5>`%mxy&=m2BfG7+33gEi#^_2Tp!I7!E02;6+lMV>d}o;bQGhe5I9B3
zm93!2D3GaW7!=KNM4oT9fP%F9XAtCxzoHccU)h;jwolpWv8XMB<_8@>yF-`XFFL4L
zHi3Zgj;%T?vtF@6o(l9@IzkB3c)U8GM(wS%20q*mS3*95ob8IDfyqXo(J};O##&*#
z6uZvB$k0$eby8cl6%nZu(1myZ7Wwd<oZT!7Gx{@T5%}DMS6H2;4p0kLK(2}d-urXP
z{r*SpOJUPw406(A0m04HeI$V(&uQez9=#(0(P80Prz4b&)U#>|yai{WPA;xnW(gUl
zXGM3uH<<l6Hi8cZ++EBUiSfSpSX}rSM5HQ{$bu8inzul~(W6)lTzu1RE_7=n+B5kk
z*p|?o@@<r-Zuthg15#I8L_Q}C%9$vKUnZGdWMfI6{%`=?-wnWWtwn3Oj`fWm&12=_
z_mb)S$HD)u9ftNqS_t%(d(7Bxuq$<Rzu4*IpZ<mU*37HBdo}uirlzQGU-{P*vJ?_8
zk7lAEWa*4Qz+A3aW#|%QTBntZQI}ue-V$Nt?^0PW6$aZEY)9Q(PvpxL&mwt~onim)
z7yJ|StsP26#r5pUgcp8*&pbqTWcBmhu>`SdUZ+R`mk3-CPurBlV{&IXqEb2BBJi^m
zdR09`DhFsd2=OGh)R#)i2J>RuLyO>9N>S}74ZI!c``dr0oR(}7=$#Xr)`I-BYA+|1
zu=;8`_J^z;RS>s@SF?FQZblx?p`#`$QQBXbtdQ;g*t1<pUH2Eme<>bL!k=-saGG+O
zP&JP)`?;aE8JZ!cQTJFp7}uI*?k>F!G=adaRk@JqCuxXP1<Fx^N(7q*&VTF4&@h}S
zV>6!KG3V&(@pNR%*_{q4bz}m8Xw1`mX4al&AnEwoz`MLmC5yyEYlf_boVOjg>~VcF
z{_kXEJ0F{uAt45xag~}80jk?9hUSgV9&3VA#I!8it};1p@9j9<b<rB*{Kez9uErK7
zzO&Ey?z-Wh5eISlzVMOP8IZ1IuBXUrG<RUy{m}|ssh==dPkeY;O#wvl%ajz#<<d|C
zD`u!cg1ZiTBc1r4AGHkhYBu$74Y?{DWO|-PfTD8&mTr~gjoVzXsGBm^&5L6cdeiv?
zf$Y)*`*2Z5{^wYZI93zYr5#-fK#dt!Cl_+TYpVgNS>@M@c@(BsL4`Jy@TkvdB*sk!
zXeYzvkPMRQ^)|xFoR6D~G!Xz3`94#CU8u@S@VwA*`CIBMvadTNN8UkgQYL|gfsbW;
zCe@`G$(2GT9V0$tCon4KlJW##VMw0qd_8$mD|4vhVAV}EG?eaF))2fZ)7rC5n9-N{
zCPGX`{ZVUQw{O#vm-~pbv87t#pm)fHiIuC|Jss{e5%%q}M@5g}Lz*<19k)1X3(Nll
zX0^%89bhjm5pU>a^&#r>=%X41OQ`4h-(l@z7yK+b?Oeqa^82GsG+ICWl3}Oft}Bqu
z+<2jN9qHRKI-!`Vhxqz-;`vg&<+$z-oBYi1uoAq}n-?rOMN6$A*2?dsPR*WP*0f*S
z{gRw!LogB4+uEV!vF`Jx=wBAqEhi+pl%KuBz5GTefm4L3yM+5cU;D%Sh;BQi$>BAF
z=o9wJK1YcV%tjLB8O8xdnQV8>Z)Mu$oL$?Wd#2=*naR%`#Z~<-bQEgsbWkSqk!oG2
z7J~6eqW~yCDOdk<Ub;m0R^q@X+0$y`jDK67*{uXv|E~BtJ!`8NDIP~$m0wwIQB~7f
zizM3AfQAsdM3Y{D9oy3SAJmSCCf~nLH<DCO`|2zchgx>BlV9G^i3^AZYCJ>!<}n?Q
zQE%1vaIO5TiKu{W3#wsHe$Ab)o;8YwNA<inS%Y-y=D<jEnv1@R)w&8K+;F<|kqn_N
z{QbjJ^!`(gP7bYvh*q}$`_nJqe`ajzpD5<p>q)}MFk64mDdd*+won^MG2kNu9rneC
z4bY=;tD^|lvJ-|0bUh%}9%CYQ@sCRsMk|V2ok?;)ZTa7O;oF(8d&uE$=h2{GXFrWl
zrmKG2B{qq@Z^38S<edMb-M{4kKipep4ajV?36*7po+J+Az?>fL@3uB}Tx;L|F|c)z
zaeS=iiw6h=@uOSVlQ^{*A+9kNV0uNsu8G~B;x;B4{{Fp2T#9<eW@i%!;?qo(l#h`Q
zlFP0im&&5CsP3G9zS!(m`s51TuF&xSh@AN&>)<YfY>^6j+aMyURDz|>oeophTT?Lw
zwS0u*wXv>lU8v-7CsI*6RMo(N6$3CJn@dK#Z9^VVWWP`rbpibYQ_AAoSqTLH87$FL
zu-v96leCTm-+FKDxLB>?y9h2C$UBq{jh9Yn&7a$|nkK%+jL3e3KsE@{8Jl{ohfbO2
z!=Om1<u)4z9lHh5HV`va8}~n+OPeyoAHN_W%cW|<g{Gh225nV#zxa3ul+&XwxR=B&
z5C>;K`u@B2d6lkR`g}s}?YgcEm<1zm<0vGqup40P5;d7smw$8oTq|1=Lz$%T#<aFs
z+3Hb1#$xGpF3hS(rn*A2k*if>BKg|I>Zx|l<dH?#m8cMVJFvso^$b5RqcQ~%wZmtd
z7A7CAltFV<WSQqwmf~KF!FmktOA^rZR-Qj^B`QnND8@j<Yz5PS6w3MuMWZuE;{IAT
z1onIQ_h-9U0Ilh%?K^um4nCWGR(~5tFHITgq`Rw&I(&nd?0$OdW8*pH&hLLYbZ)DS
zy6I4|J(s#8#%t#B_7F|^FO&)$aR1%Cw`iO=87PgsI&-t(e~(M>dYS7Y*F{lGQg~<L
z%p$JLQ26^Rj+;v!Vk|PnYCV%&NYOH)k6sE-UM8h_aB?-Ubn`{F0)n*i7Q^MQ*IUcu
zQf?a9u!{4F>S265>1?JIF9t(S{RH#eWKlQQ-DMTrx7q4DELlPl)`{r~%vI$@(>KQg
zr>MnBp(f@@A0W@#v^bnwx8dWj0^EDT2n}e58Du@T8t2QCnXSGLKT9k{Pfa81%YX6V
ziSqsKC-%kGpLv3JgZ>JsxjjU92%!u^Q&F-xEfPNgF-{<tEQ&(V)pj&&xaDk4`ma9e
ziA7yDi*iCb3JPkZ9Pxxf)Hm7EoR5mXKM&0dc|`wHE_mj=ow3(T`D)w}BJI1C_~YF>
z6@hn6QoryK=*5^?31B$<A)lpPXMeXG%Hv~>kXa^DzP4Q1+L)o!j1-`rN=mARuzwgn
z&qs^Mv%+tNWTdq%OW*%9w{hL`)LD+ue-?YhjPH%lCi8k#mAko@E>*?h<D*&;!frGy
zA-mY>b7ALTGAW{LfoZwCKUZZh0qAK~4|HiiQ5R%~-%Q!zXBcLVbJVob$I?<L?*~ma
z6e`8z#%}M6H*<g7esP#&{ELVpV~}A=!gf+rAoKL*^V^=Hy(%dgWZhuxyOg~uB>5<Q
zfT}DSjh>;xp^rA4G(Wup=kCStV4-H-7-0~gny`(M4VI?(yyEm;Pl3{0Ya8YlvGxxC
z=Zh>d2chQ^XpzK~qtWQjVbir?Mtv0WaNTVMtVPsv0}6T9>A7gqW*p6PLFH#g8D!N;
zEMX^$`+?syG}i@`i=kP+0v~QnfgGD+_!jW<^3`d5@P7^M^Mk%(!e;6Sta_3KxY!(2
zHd`<lHzEa`=x4--t$?B(b*#S90iA)c5JG7L=zf%ssF@qm@@xuztI_8X%C-WwQ}P+Q
zA4>~WQ1ESsQ`m-m^3|I>MQR`UUS43<8w<v2>}OaG)*$mD$AVHn8*D9A40w#<A?dhE
z4L41jar_ZB6t%4MU&)Ia*{V2ByMC;nXXkwmrw!20AUhN;%5}tlt8d7o$swqmsFjJA
z_!@^EYBxP!`C2RU;|$-IJ?8}NsbX8X8^T0N9C}K-0aN;E3iJLLUS)`8ChbD4``%Q<
z1xD-y2vD~*(OtGVQ|i4DO*Ftcr(U4T8^toOC#sbFy}cuTX#TC4hb@8<Xyo5(BVUId
z{_Gl}!Zyd+ZXm6$(^z1%pvxVLIJU>HHMlX7SD7NHL94)C=q$`JHv;%m)Qw$iH^e33
zi3f{6^qLj0_6oj*N>+6xWHTK?y>)r^kK0bOk0H4N5%|=+8k)JnnSek$iZvcAUmc$w
zPXbdu9d30tCVaN=Lz$zdi7c7D@%_8b`p{XLn;k5k);A>NXBydZzW&EqLYBR)+`2G`
zq~{bWiR~g#!C8^~x;hIXdYP~tYtNNLz-yX0H*`krK$=nl2p^6`W?u$Qvgv=LHRQ`>
zjfs_g`#Pa$$`Mtlv7padg|@Gll|uEaEAhXXHJU`Q)>gvJqrg>-GldJ&>5}IDT*+%s
zjhB-{-?@0N&$}dE8|Qlb$I&`2T`_WZGC&AsPRidVASRO$Zga1j`}b%2US>I(>3^+m
zxz??2UEN(2Ue#}J9Xz?g^$i8D>yVW@5^^NeEy~%7-y=t}BM;xI+x%`#KIgHTN!O1N
zjNQytjk%<z4!j+L$voU&*S^{4?cTCMpPwQ?yty==5B52D*pa@luaCog5u{w9OF{_>
zYin(IQNL`cmSTUvW*-??bnPx127cG8Y+BX-8V6DUcKctL_UW0h&BQ6~?cK!Iv~RSh
zpR;6A;u9dW(Fj)2d=6!J(#K3+6#<$P(ObIEPA*9TG-U0M5dthW&l<HAaP8QgYq52x
zS!)J~ZSwmOFDE1Hg;_gk+(vo-uBG3FnT@LQw8?Ypa1+RLVsoUSKQB>LX0?~4#$z{y
zlG1O~md9YGCCEe3RnflwmEXp~idTAKbh|+%Q#I}gnyI|yp06J<!r&Aw-#C#XH))}3
zMs*e#f7DFq)RDdp$iJk!e5sC1%$1)Ip;1Ol2>ECv=Ro+#cBboY6OcsAuSR3)dgMiu
zvn1lZX|0`6w)7`-T+Ej;0N4^sem;Kw?^2it7Ao<|=xJ5*sJthQpTk@?Y8v|R^$U-P
zu8e?RM(H)~6yz=OW(ZM_NVt6|5U*&?N;RaR_cZw4;nbVVT@5scs9Mip4HrSfgAwb8
zpOd0gxno&&fG@<LpUY&Ac)GJ5cLyHY=f4F!_O)yMu3U9eN87POCr=%l((o~swUbN8
zRi9EpgXKGVXtC$p)m)||{&fWIhC{zJWT^?x(y)}R%JUJ)ze$2C7pmI`CpxHmyenJO
z1J>+!lrzNnGHt*OakfCW5vv`W7*SzmT0bv5tG$u`ElL!Lp#4ByFmThye!3*q+qBdB
zYd?3{yqRAoZM7Etjq)@D{vDz@Hi2KN%K@5Aa%`s`THyP|7S9^GDT(H0-%)pajb1g_
z8GuNsEH%{-l<OF>jMD00{`z}ba6##x7%>b+aW4>e5QfWkOVerd<2PF0R99$GmBh-G
zBsFG<nFX|&^<hFNn~H`Wwb5Gn)KN3}J5(26#zgx~GxnQP*8z6LVI%&Gkb@zqsoV}L
z^>qxw``{oh;=KIyC@Hxu$FnPhGxKvdqs4zc*=SJ%>^(n%eeD=4w9AvcSHmf!Bw&Ns
zXZ}mS@dvsohTFZQ5at>Zzf)!9qgL`6>50%n_1~Kvr+heR2a%&wyXpKg)KBMubK}qV
z9qeq41*rKPnn!n2Ih<%)rA&kK2<>l}j4=@n>hq%{@@lp+Z=%U|V$ab)8wMp?+ESb+
zi2jO}yG|mW&`u9K`CVPbzB}x|{IF)>{lRTKgn+~Ia8th0N2!r|4K`n@7psmrXE41K
z<^N%H{{PbEE4eq<=Vm_DI$v8{Eg?rAU3Y)Z{j6^kD-Bj+*H&gATmM-?c=Q1QTB}o<
z=fBy&ZSV$uxnS<FsA~p4q?r+{rH%c51lVc2l8U+D0^yu&@%-2bos!7Ebh>IdB)ETR
z@DYEcE*Q;v^7>6lnf4Iv-&Bs$49B^3ySpJ4lr}65X1uD$<$IKEPSLwPo_2kC$E9iI
zSiRNUy7HNhW88q5GJ1m!Zuc$HT$7dw-v%npB5jV%a@9?Sp)eWX8FIh-A@)|80W8m$
zdD7LmFkp=39W?LtGwGD_w{Y}Uu*=*_;SdMUg$8<q2UgN1PcPoG&mC7*xK)>yUN2}a
z1(I3FCKs}<eUwl>`SC?dw2D$byjV<_>%ayIYo>FW#k>pb;5k|B4I&IJZyH?T3hfsR
zQgRgD_-ZUQg?osXazen_WEAL+>K&xPa3Ahx5_YxM2}9{+D}Y4D5-3|D(`)(>(a9FD
z9UCMhROrXh#og!Wzc`#?=nu`L(q4-|=8RQ6)T!D?{WPh)_61XcI$mT0-IhAX%&afi
zm3=0?ay9?c33N??GM}Ep0B_l|tZ+Dn#LKcQBplKfYh)zeZ|KQ=FpvjF4;}Y<nrK+$
z;9AjXKMRySu&LV#tF@ipywhUM$vX3$zrrafkxRdeDBikQjb*<Or{rK;+-c;_Kgn;{
z(I$9GqQ!PByQfpUO=ge_;5i}u6GzMXaQ0(^rxIYx0Lu={cA}SlaTcUv!-=EP1%$;P
z=lWJOA}pNnUyZ+_WiTs+Fup3Mkh*@H{F~2>+Hf2fJcB+vnk)=wp!(A!cBM|qD6x*x
zcM6sT(#Jjoyptb--Z4gpA@U@eA=%>**D;jiS1c%p=r9a^3;Bz*V6(3L(^N;eW8Dpe
zJ<dwW{U)CATryl*I-2;X5KSsdQ@S)qbopI+a6FSN*b^Zp+Hqv2Zy+0gzc!I`0kf5r
zmIf#9b9qJ0yt({Kp@>_u@{G3ww$q6f*Sp%cuVKR2CJr+A<2jX`6`@U|UGlHi6pGiX
z<UFc2?-GfQR4pk6r2p3yiwMKwVML8lByjn698~t@>|!2k@6N&krTb_7h)?6y-$gS6
z&k2qp_(I4jTt{5S1Rd|`#cx`=conokL)|U0veWu*-}6C+BVM~NfFN}N57w1}`UT6L
zwEOS328lgeY$TyHZrKjaL7G|b!5c9<EFN7A;NV^OIzIiEr(*e^#9`B2zwoMjfW}@i
zcQ!LGfsC|LAl4tW{8bI{KQd;z+4jH45n@{McSO#8t+oUHi0n5{h%k+g^Y8+d#6;8I
z1&P(m3{o!FSZeVkPd}sjl5`0Uc#X$-urlVT{=?N)GX3dRx`w@Fv*P<ksg8;E(4(e@
zPo5^+-fy__E`LpS4D^>Ty3}&xdz}s;{sJ14PvrbbdM67g#8h9X7}nm6AXl+J-p8{t
zu3eM?GK7e;s;-|nL|s|Y?fjt0AqFJGf|jVMP4drY2!=JF!>8s@QifBLa@pEXtwCz4
z-1*cMZi(VQBcc)@b6m5h^z==DWhC?`;*i>AI^ww#yb`j<M}n;q0rgHLCuaULY*Xh2
zbq@A<MjRU~K%bPso(AnYGlNbB6Y?iUw=wFKh=Q9|?Tc(_ROfKaE(qt%L1}KAQ*d!V
z0LFCzlKjb|pSaJS6u&HE-%lyR>%)CO^-wCyrN=8H9>$IesTDZ4F0&Trb=`4<PfYu1
zlT%|>Hwi<2Whzroc_U8AkqD+@|H&0`DewYR2lb2jyaiOI*Q|bpt6seH)>-ghPzjvj
z=U=-e@I0c!*QI2$ru|58`aD^`#!-n*)$YUFI)OoBPMh_@kj%Fox;5|=LGY*7Ps=o8
z#g@4j225|{R$*c_i#OCMCiz}e4QYt1|9Wl^o?n`Q1)9J(iG#{{k_0CX@hPtms^kL(
z$Kcy1hguI;<AKOCt-%5@X-)Ul-}c$(e|B$%Qhw1Wi>yZpej@FO8f?x7H1$&8!b#Po
zK8ioc4J;xMP5AHDwiy3$CEQ9=p88~ctwc3#m+cP7a9<pCGH|rvQfSMl)S|7L-~Y{e
z2Eghp@#_1RnSYbr`^jz1{vC6kR_enfe(Nx%wnuGs)fa#|lHS|6lGW9XLo<M6*I22s
z2?*x3U~{=jMXOMq)`L2~lX=ti3o!6ZU#Z{MK}(eWC{D^b$V(E1CoOT!4x6ngP5#<N
zRr~HWE*lLEJ09F^`P;8#&%WbE`R*>UxOR{24DWNt<{y|-0lF0*6_0k<)7c{zBK)2g
z<I70b7+ykG>xB>V-bYEM1hViW#!{5e)90Z}1-4)3reOaWD{Z5ww<e3GbQ&$?l}u~>
zMv`u~b9J?1r!hu5v}sRT6}dB!*+?Po)I7jUV(XYd_Q&sZw3NF>F;>7}HN|&vxvp4V
z;#da3dtijYDd=izs5WBmpQ>81u}9px(r2n~?zWLY7XqEB%;3oN#T`CgIdWTPa#pT;
z<~+icvAv|KceLop!8rb9NyGi=N$r**o#+uKMv$nHK#%`p1`!H6mPr-ZuL->N8e!xO
zn!sexF>t~5tNVd%@hT+)J*@8v%uzGA-{pMk$EgxlMl$s~L0L|*4?|_?_2fJb&ifOB
z4%z>(g}ZhWQs!Bwi#R|2pMCWWlY+VF_a`o2!M36V>l!r0Q+g(mDd{tE-=)go(;%XQ
z&4OG1&&fI;)t0a374tmtmeWW$Ebrfv+jwGDIZm7~e_0e#b2@R*&|Q~|>(7nxJA{5_
zeDjI1yg}(Q;eH?dA{lNs`={gU6f}pskk4<ils=?bs*~*oqv5@u6?W+e5qC|nob<%R
z!s_8KP0%OudW8`Kp)2h{Y12c>AVNL`mc}TcDJ)DO#-Ps|CA2JzuXGIPe%`;H*}v2k
zpD4z}<?jAWPii&ccQYiQj_+n>rd&%ssP+>M`fe6o>llchyS+OKywqWlj&$Kc9ZGDK
z6v=$}G1C1-*)C{H<;3dd*?b4Y8CY3``@rsdPhn!HdODo>Nn}c^cLb0UU}}EDmLyI)
z`tcc1`Du~)9{TRkpC(#kk6ZjS_+pos^*MD*>_lqyQf=!3a4FKyZVK1rM#JBlGvpGr
zCcKi1mIPay?^mI#ezIrS8{)Y+H2&wW+>nUbUfv87o?#~Bo@7VF%9;S%=*gqdJT3|F
zDOX-e#1WH&DlI1F75hME3GOVjvV>>RLVu;URNB&iZ9jr!0LN?VNU*9ESRlz*a)_5L
zm)rg8cYkVX5SnzX<ayRAFi&nT*q^EE{nVxQ_*C|=(I;v)3y<2kzp0OaX1+ATUblcq
z@Y4qRecS!Pt0i%!CM0>JOZ+JyQ|>66PtEh_8=q`E3>$~=sNFw2f`-m**~cC>vCEqN
zTEH$v@uE^@e8`7Wb_8qRce;d}u~!o&bW6>CQ~C+H`QzLrp3YS=GfAY?_fwni2l%+~
zCUmC>x>#i$QTgmRN7Lz9SbQHIOw40u$A49-tGE1LPG(ukQf0=qUt#CoG=tOo2$eoR
zi)$dw_4@OL-sNKUYNM>s^}9NAO{=EbCE`)WR2^Ytm&JsSm%_Jji`43A?{VG$wiXYL
zyzhMbHwdg6TBTOeAhFmtuwOCZWzOl|N&<^w3@ago@Y*FFiTA9s{?z6Vck_$Q6Y=^X
z)6=)%)Jto{_CwXis2GJ_S$KP=zszG~3h44WI9OD;h?;2hvr%C<y%Y)Y*%T4^`gACw
zZ^2-vWaBj}2I;CHU@(6_1WWR?umOh8;AE5paW808i&GP107|t3JRbILE&Q19&)=jG
zM;~QlJ3=&if{xE`#<po;xYT0l>@#$E5cM$(*aAC=#U}(L)QJ$6=ocx-uLNfg0D4J>
zkpKSnOhfC1X6c&kIzrN>vFEa-nIPCy_&nQ;6Nvr&9%OW$n;nWL&rs}Le+H=d$P-{>
zs<mndOUD*_SE51l#|)gC36Sqbor+clq;Q#x(~ryg9xR71F>gHfXLuv$>EdAeV<k#Y
zLVmO4ddWWal8?A8ZJHi^=vPN$A}6CAx-(;7zA^**N@g~!;fsF`uN@t$fkoK5RXaW(
zS-pBl-BL#wvxm;@L&O&#4CLp!)+~b!vbX$wwDs14OrnW0wliw^R#2&B_#?S-3nTmE
zdpd#a*R$(BiQ%|Xy@cncJ$?vhPs)(ZkziUqPhEjXE6bes+O0LkQ|>nw;Smy^#2ZJ_
zV6kt>ICf3K-!u3JsC*3@YzQ)QYK}Tdcd)bM^KfI%j*6so^P$ElY$TgJ+6Q-z2DPU+
z8)v5@Ji<5hLeV*a_gB?*^3n*TyH1Y)xX8o`Y{o>`Yy<(rXnTazo^B$U+sz}BM?w#X
z(A+$78E4sWWZRuO?u59e{8gggQ!4;6luEq9b&xNb_6OcH>Aj+4dWHB!7bEH8xMPqB
zGgYu*sY2rY8U?b{_UohJZs>1jx9V5ZJ0;5R=sFb<m%?3EXUT^xuyDacj*UBFF@kZ)
zoRfz4?)s4to?P|X$n<}%bDkpLp;wxk(p!yAR22T*|7EN^sGgeZFO;={lfrcPV_Nz|
zo}ld$mwVAz#NGaMiRH9-8|FF9B|h71Zt0<Cy%qVE2XlC@J9|33wV`)^@FSPF^2RdW
z_=?45zXCILj*lJ1mkL><SrUjDi~{WMXvB7iSV$*Z=yrNbJ_QQd`1607i^Uw2t7K@B
zk0#w?1Qc5BivDv1>xg2LN}`t&rZ;pNwZ*uIX~reA$))NXq%EG}w4NKVQ5X_xc`+GL
zd5sK^J99@q|CMCtjnJmY%}nw>P_TdrQ?ufG-j{gH1{bFA-plj(@Q$g1Pd=mRvOaiK
zpX_Cx(Ow#jIIjX3%qI!T?a8**bfF`CrJx7k_zP~`{gr`yu;#Wym5cz(avCf0(7p2z
z3h+zx{>~!j+Y}e`!4P>W;-HV8>uZ_qh8zMBRwmQ}r5Wdw^kgFe=jewRq~b~-9naV@
z7LK0Cq$k>+0E<a=z1Gere+Fm=gXnJ8@f_(V?=9+qW^0*I9;O|3h9RhpUwe#3$RALT
z65j~|eAF$~StXG!AA>)Eg>Al<(gOPyHfrtv?U@^j>EhPd$>R*=g$vySg7qyhj5?JN
zHido(J&R<?efm_Ch0yM)DiDxZ$1%-ap}i*GjMU+&z+3adUPdQia{~DhyLm4G$R;`;
zm&(rjeD3$-I%q&5#=ph$h!pw?Z5{nF*o*&#ZoqG11LnZ^m?y-P<Deb;9PvY--g&+I
z^RYgvOf9uV?}R8~J6R$R{W<k0kxVBd1|fWA$@APA&2EJ5FTg|Vzrn8WN5`kCI_bI&
z;j^inap0x-Ac7vMvea@MenUPypoo$XK1uukQTEnhU2fajupl5vx6&XXE!`y`4N5mi
zNOyNjNr{wncXxM7cXxL;eA9FG+H0?U*7@Fj{`?Ww1@oEDoMYVMj?}uHui<$j?Khut
zc#2djsc0Zrf6T!3<ZxjOj9^v3(>0om=X~NEOmx59*;K}kR`~&WIl}5>`pkzdXxIq?
z_I9$dg>&l~anytH)h+!XT+bDe&lA|_n}ml<+e{y=x|yJ6p-zj%3mJN<R^woBDf!W=
zHlbUZjVJOl&*<PyzNwWGrRXAbULdiUYk%eSbRuSDdPZTo2wD4jN{Qy(OEe2=`UIN_
z<PKeK2j%JX5Uj;XiY}J$;&lC3bnj<j&f162u7z8K=iSN)XNHxu5oa4)gF?Jq>TG{~
z$vzM{!`g0#%N+fjD2Uyi7?V6T{*sk6jv|k^^@|d#P|f9>fj}1S?aW|OqAu0$SE#$k
zy3c}eX2!HB28IS5Wuf>?lEMNt_rSHekw!B!DCLUjgzRXxwRW!va@N%4BXqgT6~f7B
zKgZ?U&&^9t*`nvW3j9ZyZAg^QQ!GI;EvpmCHz`6nMPK9&Os^M&G(^Z|#uF2m)Yd&~
z%f!}N>eT&yFTD91oog%1GYPO=t_Cr}xWDV?#U(rH-1k24JsO4+y=QO=q$5TWfGfoO
z5pFQZpYnD9Dqe`wt`7yEED^AZ_jBs141`)BRVCG7%gI}_l)4bwY&DjP&42MOD^SaW
zMU=KV-1kk8A);Iysqu@$W<EIt!b8hy>2c%M*WJ&({NRH-$zry*Qow_WX!!zd)q%=w
z5ZKMJ%9#F}dztP}yY=h8`J~}`ur<WT<SJu*Qg*3OFQ@kG_r`P4>xrv;+M4^xAkASL
zHqeO7ng4<}{WBrvH~PdN1b8w*MuN>rImh!hD(i^!)L$#nDgxQVSaZ4F^z>1|OzjgL
z%`RB>uR6o|@J`6dU`AB8z86_fvl)`2V0h-{@@4QB*1;U1Y~zJ&amrXqA_zzb29^uN
zt)v1_>p`R#L=*+U1BTeVxshKoses&PZqJ{)JEWhSUN>;U+=b`H!4%~cZ^K6f+^jEx
zbQ(VeN%?_<d(*Td>qo>J_)eayoFOO96D7XexBlEISi#xJgN(h2r8z2}Qw7qTV4QMk
zh%M~N&W&RA7Jz79Pq+CZ@6($QR)4+k)=pnW^AEo2Kwvt*-G-Ejis(Y2L{uV0^=Cf0
z-4k97<vNe&7nSpaq8Nn|UC7WCT{SUUS5XN^@g?$l*FH~EZf^#M+N?nK0m<XIl?EqF
zhwCa(ojMd19;(h^h(ZSHwiZG3QZ7qqaU3J~*824~{*SQ7rrd8yISBvU?UugJe#oud
za^b)TtHlWr15=Ve(@-G({INg3=p`hOlG1|maA&f9oRhW=)k%upSvM%?{{_`cNrpyo
zBG-I@Sv($HL7&-?9+&-HuzFT@FyP^Y<*?!gHq14-eQ1kt@S>l+nKj{9FQt{)pZ**x
zKCn>dM}PEm)!dn8Fa&v;mPu|nGlbovWYHYbASd{FbdmS*fB@o~80T|=FxpC#?HMLH
zz{CT+<HiA1svGhGXwQNwKJo%pf@wqq@sbeH5V5u4rx5|`!!%*C3~%@FlDmWI56gXZ
zWj|A7w5)``IwhMeaPoYC(pS+)!?eFz@KZgdTlf+{6ZZx<Tb*Vl%1|}vv(+zO#~EY=
z2h&SqOT`o1bFDxbT!om;iAz@A$1w$W7N~T3;|+6Ef%lVHtnKMm>6&NAiX4xD67>}m
zc|5DvhVhoYTsBs4m07S*4Bb56_ft}v%uHBIi|Dqtf6<bl&_V&Wh$Q`U>rTq-1le04
z#OATu$&9(BJM}^44d1ldj582#|F=&3Uo&g|@szwAiH?uk{j%~>iuQ>G%S4@6l*@h;
zaqmq-W9tDsi80X#8JX?!RkdUh0?LkGaU*0$s`@KuX~B#Ec=D~b)eSDy^>tr2&yL$i
zP!nKj+xM>eqQ*L=OI3vA=}^u#G121=hfg0v@yUer=txP5G;;?v?d$Dd3n2`K{(N87
z$o-gz_gk6S{56!5>QeWUHBn4MWe3N>1ofCavr?tb94pY%SuvQQpa?Dh!Uik{vY#wx
zMIHB3T3@gC%h5m=AE;bcF-_|DcoCC#zU|9q^qsEGegFmV0)wa&WUIC4)s<<w!sti6
z*|qtzuGf$SnN4G5c57dxZ-$o`?M=R3-OXnT7>_4@JVCa2HIWlRckIxP7zea)hzv%u
z5sr^HB&AN}?ggb(^Ysp1Ewy<{L$5zii+3mYhCD0kJy|Osa0)9TbGyR|fTx$lp2)L$
z8eZw>{{F&Ckm-Ismuc!>X8hk@dA;eW?gKpWP5@&XQEbhRM)UJr0XjX^!@*@g5Gsnz
ztOy!0k{^qIUio){^S?j1faj1}?kXf7LdPSGji;p1-EKQ9D6#V#wA?c73+a3gdMV1q
zDu}Cuf;jC)J3+rDv^juXZ_n`t+HVp+$Qu#2t7<THgX_lqVj^xe!6rzu4R^J@@jjM8
z7?Sk(;VyCA@|`q{)!~P4J!pbo^?vR)4a!;nI`wS%inzP?!Y$ia!Gg5u)Ki18`s8;H
zIIJZEC1FP$P0J*7wz;FT(T8SrSbGD6smvp#Ab?gsKu$&q8|dg4+1^PDVB+Zw6gTtc
z6y4oyeVk?yJy6h4tO!jRoYII9^LV|w9*OSq3mMYq-nCu+3)^^1WsaH>iqRAbpxh7R
zKI;izs5b}#YzA0$VZ!)NFvx>xQWSB&0shW6VHiK%1g0jRzwoWO*@P_B6GK`mx8=`V
zX87e&XG@7?(U7{BQelyd!nPPNAT`e?TLS;1pFHz1F;_iF@y}}{yapBzjPXYMKwl<-
zbI5FYzCF?w&K<~VILtp@pfYB}iWm0(JnC`)1X#aU)Q>If{8;pghf_jN-!C#PO_4#Z
z3!k$_P8q5ez9ZT3RgdyfbMhN11#7KazNej{#6t;K^3LbTdV#cR2-bp{LhH5M47NF3
zr{;yKM&u(-+ruEu9oj}*mMR7?zB_g&j1@O}DY66i_pd`lD534WSztQ)ur+%^mGgyz
z#XD3CGJ`=0((#k!E4YMu<oWn6r`11|u0mMtm%b1+V~H7mXA^ku<asMW7kIfIaW&vN
z((I86{3vuv?alUk>WLDyECuSE=ocz-A<aZ{-vZ8fzIwQW6{AQ96%Zm6e>s9oUR`Bo
zaIt4tkYVg}SW+O$+;}Mz@CzZK0e7~Rk#STgTY8#EzIFcNcdXCKH|Jw$B(eEo38_wJ
zSL5%OYq!dbxJMkq0r{OIPcj>eYW}DE2T!<$Tvz*Avp*zr{E#G;Kr$v_aseFpg8{oo
z7`US@11b~6Z-tCeMUh7Re{{wvzZp&!@tUp>lgbltQ>_%oKA7wFM=)9Wl&@I#Y-cPt
z1Rpo<*(NhR{fRUmkA`H@eiM>t41!{L778X6>`0M@*Dz(~>>Q-E{8H14UfHdrcsxCQ
z7u;)ybJSR_>ewFFUF$*iHylOicJS8$NM0sIUbFh^ClI8Is+GPqD_;?2VJ+WwH*gbQ
z?xX75UHatJd)!-XE7LR!kkx^Lj=|2Dbt0R@Of`oH2C*-VGp6{l-p(Y@ai5{l`w<9@
zQQ^2GK(gT{?ps6qUx5Qer;83+(Xq^wr*1COCGX{&c%$6{`-Au(MF`@)4DOA-QE!?a
zt}uylelC8m*2ba@l|YP=n$OkmtIO~&5qirQUcZfPaO}=D2iBuEs7gV)&Cbb_!4daz
z!rR$*k69WJltG@}%#>-eG;g)rzXk>1cY13pp+Im6cLb%uz9rHYKrKm4QAYU$MNJe0
z4NoMDQbHVL8KeIFVbS0Se*JsgK$gb$dI$IBdeZXpa;B%m#6-Phrw^E*le8yVYy@d<
z3h_Ih&yki1DekDM#WGM(Am8)D!a@)eLHT<Aj}Ix_aPu}YaE-Tle&V6<a)me}02_s-
z-S~7EihsZ(&U?U2q3?ronP4-YrOZOWnwPn>Hp91I#Ac3g%DAm*yf7;vEPR5GI0*jT
zjWdgFC;3eA*jc(JQ&KxJB49sTN#>X>eR=CG?wQx7Xmx+h?Vg2IpM1Ui+{gX4_gzfk
z(MY@SA;KWbjDwbrhzoWI;ZZ6k^-JXrr5@?#oAGr2c;ks6jhG;3Mq<axIxS{X(fzOL
z+SYv)<*W&JFOaLuh4!X!nQgKzrgMQ&PM>D6bfUWLUg6|{?wx&_`$GZrUQ}ZGI3F8q
zk0>?UVB`<kf2;$)AMtR7k=)~ZH;{(}C~B)@_BbQJG~Eyor~8-#?vI+E9L2uDji(92
z6a2V((;i@7aO}v<42+E)KbHlu63+bXTK>7pZ;>J8vaaFMek!l5`13RfIoWf!nJq19
zax7cM(n8!LaSu0H2J?`VF`6zIml$@648&uP$;qhHr{_^w&gj$0rP!VJmTIsl+)1c!
zfKYHFPb|8MCX_n+D0}%4H6R{Da^6aG&6A_Bmx`f79A%lh-af3Cz*i@9`6bv*c$Cq=
zJjm0U&MQf^v$Q0|^kscyh!1->OG;t86Z<Szh7cLwHoWmpVSjeRKIo_!`Lz6#N$DQC
z%zAlx{llGJUEsV+u|E<=M;&)5`Rk1OZ{IyOVOuEcq;U5x8B7#CDKM3*?|x12htfdp
zqn1SH&Gie)!u`3si0B|e2=Ln{lw+iad}R>>Fv(%I-&Z~ZP(04PuAQLV_?~sGBV`QY
z<`NJ_p`JOm`y*$AgtSgc3b?=7pMS0?3giyy@uV7r+&;$MtL)^%b=#S<>&uqsy6|bc
zlVqGLzSBE86FdrY@bLJ|MzRF>4hd6=(H>9uW=Nk+x|zwq1SF-K@M<jQ6WRF?-wC1K
z(*u5k2SK8H9W32iwi3rx&4uD#AapnWRtCrn+rlYMFvxW^UdRZH1l+!4Y2KOs51f<T
z>g8S90%i=2kIuZf2adm*6S)v?_sfZB)vz*-n&8n!P3H}VwLD%koWbKGk}Q82z#A{i
z9835%k`)p;TV;2u2lIj>agwCzBSF;71$bmV#yWkJ+3e+qkDh0@)4G#a2Vs0jqq`FV
z=D4HS3RRbgG26LvXFCCJf^s`n?O6;z!d+=dTeD{*=a`yhq>Zz=G(jk0q{$-x^W7#Q
z_X0@X$b7BMjl}9sh1ndd*{#Plb|1jUChy0{gQlNtEug;nFbpEtFiM4aU;g>r{&k`M
zI4}BQTKoxNw(rjx*Q+&ClTOb9yF&W=sa6EWvPy+fwvvy?W2o)xPBYy1OX=#?(XgeG
z3Fbn~+Q06j90N=-{6pB~*m|#foAvO?S8vF)ISYp|d<Vy}J{5*VXVKeJvUl#TVv<)2
zUeASPWt$Ah%0*0;7CVQIPID!&Eq*l_?kO^MHjQO%oVX1RQ0s7RXGE%RYU$D5yZcNx
z`v(OF3OjzK>}YpGg*Pqp2)o(Z%&HpV4K1k87Mdhe<FlB5l`*9`wse(Zg2%xa!rCk5
z+3JQH>4ilzT<_I<YwUSsL$WLdX)}2E5|yuu{c0hbP*Qibg}NAO+Z%p8DXN^J?mM=+
zbG7*5vy9t6)`L+v1&I32fxu%F_*O@ik+c*#jHC(Mg878O1o8r9eu^IH-yO1ltS$c_
zigE)x|3aOURI6;kdzYDam&fhFJBuz7hI^CyC{e72&$<QozPBIH;I<1Q3-sXB>+MY$
zL|=sQ!2o0J#)_xM&i%UhSjTD;C5l4jPyFrvTY-*jmy|IWPZTA3VxA!%Y4#B~@P+OT
zk|@mwqY6jx+$w@M2l}&rvgl6XT@}I8pPxWs>{YGXbLBYOy_m>dOS!#B9Iq)Ivz0Sj
zobP=|j#OH;{dqFON=(Y?wJ!#8ncVw(-!IhFO0uo(Q(Db@?@;x6mj7Bbrvd6K!bz<3
zydo6HNu)onoRz79^D-q0Ph83=^ZNEf;Gg%6D3=#-k{bmAM4gvTGCUXKzF|`tV(2)m
zrmtwVni4nxsAI`nmi|BPr9YPJziuodOsEo9c0d%Q6(n#kXnrDpOf2`koGT6}s_p7e
zlWauSxOU%h4%08xg{ukWAm{3ILkZAu&@jcWilSy}cWm{3mS__igT$E=o8&~!&#QHB
zP-><7Vjw9|#G*+>`{nZLkicb50n}sx+-U9AtxgpeJ7+p!_{SRnIW4CFMUeI<v;8x|
zb@K|XW;9LwG1U&Wo43(gAwB{=N$#_%4eXAV`z|^+QS7&(TS<FqI4o&vf(*&-I2p}P
zgZmlrA63>X@8K{o*ta7@xT9Za)-j7}xyQv>CF-~8rK6Mxw0yQKiID#O#|@ox3<bc)
zZmkb|L3su-bQ-d|Xt;LI0;h$S`{|q4$=b-rK<J?nxTq>Vx5H?Dptyzi0BP4bK(Y^p
zYZSoo6B{4Y*{e16!D$Vskj{|k6w(=59uF*3t%z*%{J<n4Bq#><p9jWY)){>8>VO3Z
zz;Jrl?aqq7gbtmcxg8YyJkqP>)G--{FY9c>Aji1Z`?Vb$F01YFQH|ejhziObkY2JF
z$qESz-?mj4Z=?WshGfnbm7ChFdk`C+49sRuM9wRXQ4w+0Q=8ltR%`P4D$9fgwJ&yw
z=oOAKLvZWN!@~dY0w|o*o~uH_dXE|4snlLtCiHYFv7`6nsJYslo|vHB-(|a~zChSM
zwc_+ACuA@d^mQEN9I;7Ts{g7pU9wAJ7Hm~oy~7m<z;VvjkW}T^E7->(naj%|6_0Hh
z5y3|9n6G<Au_YFHY8y|c39Cm7fqMMqAUDGM&kM|p-xH`<g#eLuEpC}tt^$6=95J6V
z)?4bITl#P7sc<r=$NRCAyCZsH8nckhE_B3!crHni83}s3c!_X|pEBHcsr!{CdlSW$
zo*MP6_0R%SntQrqu#MEI!HB+357ps$@_`48uaWDIE0W!?QUcK*&QJRxsOXTE(3XH1
zKp+f~?O0*8^RLY9wK+9ecnBTz_-Ba@NimPt?LWW9Usy>*V+^PI_jwOaW>9rcn$02D
zn_iB3LpzVCvj>*y^4PA0kgOsad_TN;4s&g|ul&$xe(#8Uu@m{QwLW9;V0*LSEODDE
zxL|wKNHQqt;L^8wzcq*$ORc`$S2Lw4)cm9=N~d(&Wml&w(SRcQ)f3h+GL_=)|8al(
z<v!OzQEuk#NsC0Y*G3H2DlQF-fJgPU)2OS-aj?KnwE7x+dOAw+5g0f;!?*I)wELDh
z!Djx;9O4L>4{B%GC%D&GYSv+~H+xK4m(1d2d!3zt^NQayspP|KOzM3OrCp@tZdEe&
zK1BT86@`jXhbFu3nCl<`?KWRDEYFw{74iMsZ?D%<fnB54K6X9%cM=6CS6-i@hEb8R
zteu5Vg5B1ZY%Ffh1RAyx?3rI<Gwv#l#z<7lKe!dQ-b^>tzN0uKG?|3THIjRfj2V-*
z83mHd4BYJX<W49Ag5lJV0*!o-gL9jmyeZcmnu|ZB&wxg$3QbG|x%kqWBopV43*;{g
z^0y!K*t{xDe6rqHl)cv+FH@vtgqCe|soyznmpl3a7)ublvkLUf3kHM%-g$a#mUJ3h
zgPqJCO^Y059v*BVuOl1Mmw+Z<zif}quU@gAYYrM_#e67<4u`QrQ0iyK<NYkHe@4F<
zq*`_KOL~K$P}&9|Df+q|VNfG(*=J^+8xcQSnO|-tJ74xGHAt0TQEOlAgdWd}I@d7b
zc>Oufd=?|)_TnOK%Kjjs!q(_8KoiK6DIPAAl%d-fCkrJi^d(l;LOq^X95<#$n?bWZ
zTN~*p8}(WG>#U7Tn%kvSEEtV&yyep(|K*!HlvjN<FZbSIK_QJ6RAmHpJLGY`Y0#v&
zWgw>BvHY|`H1%&+=#O*qFRx)v4wNkAI&?a@6rSxM2i@6GPI|(+Ejc}RVC4@x9~h+D
zjhKD=W$5@A%~QEB0_V9AP?}X3E9hK!TJ9EBd!O&-yUUT{m1IW<l~w?~cbZADUAbCb
z{O$z$Nj%mM(*UvM%BlPD>f$hH3x>#_aG~6zV(?h&a#2*hSA>_LZ*#39T4EV8%?C^S
z7RxeyfU%Zmo?_G`Mz8|o+>$F3xbb%==H{LA<3EH-x_=jE)c<C%BqvRob+tGqQ#aGp
z7=AhZ#y<Wf3?f#<i=BrnpYxM$h}+w}l!GRee0)6H@b&~eyAD)o&I@_$gJzG*#W04f
z`x6nLlHHcB5SWLvFoC%$)TolKiFQs!*j-#Udz6m(ZoIWsIPKD#PMg6#F0%qml6N;o
zyYg`XvWr{7p)q*i;3Myrxx|9}&$lM2(6h8))HmYWH(Sr;3)qH(sV&&Y6YZksFwT-3
z)5dJ0gQ<O<C<s{ce9dcxk1V6V9kWE7lGcSynUtVZ1yq?&Y=KUO7hX+=z8c|0AAK{k
z;UFI06ZCd|F~Vxj52%Fd#|IwUhYRyO6{hM=7D@<`I8sUNvoAj84I?5<2D3^(sx6K(
z9F_*Maab}A$Pk*)4zAZD-IENLgJ+&n`QqOZJ9MPiI=&PMrAeK$m|Duumg-NJp!+Or
z9O~=-;0ioa7B`%3P9crw6zQ}4lpS{E`5G@(`S;f2VPC>J*W1~Q=*QHm%;)pDlQ?zP
zBSX~{Hxp>5+UzV@G|vpRwrbp4b#v^DN-Wrd0d)eN#g8AaxEWG<Fy7`HhX$G^yME^?
z8q4DpRo{3MKBJj=dJ>HT|NmcH|7CMaLQ4H)H#%K+oh?ZXKdc{We?PX~X@g=75R+^=
z4M4t96f#KoB$+M}GhDp{qhhMBWu_{x=l1=jHOS}DFQKn#QEfQyqjpYxqi(O|*eau>
zb+Dn|7p%OkCh4$rz~<LrwLIR&eXadMDq%a-ZttZLFlLkvGOCD*MBEU678<$M9rLk2
zTG%SkA&hrc>})g8v07b5F_dtBH-F)LTDLSNCz-i-h_2!Gk)@kKPx!>+rt`MR{kXN3
z4BljR+!?Rpi&O;{eHW|c9MF`Ja9%3Y^W5kech<9yEO}TA3nJhAKEii8(1TTo_Otn0
ze!Au1_M{0z?9Ply+#RY3H9FrI&Q?6@e6uL*^v+zR=3S32PBKR=9Nj^zWVA%<n$^el
zuebK6Tge+Pn%04dKUiGg))i$E8}Tyr{`36{>*Wi(-&pPQ6~QZ@+%rpN=naj2a?tUT
zhmb`hW=6a@RVu4lkWsQ*(DrI@<N5iRyK3XCYRM;gJzCEkO%3`aQQ;Z6xNeRu<@&w0
zUOEg`0D*8EkFiWS<_ZUsU>#sFF%j}y$a}I{AZ3alvLq|nH-VN2Fp7BHt~7|bRGf2C
zh#VZrm0V`H@S&RS8Ab6q9e2Lsif=D@91x}|(PPsyB+Jy=^2`+{W=K+a;hG;GyY)L;
z3x!)Sx!wNgU~wW%)GrEf0{Z2P#yZ8q`Mk=gcjzoPd&OlJDrKbn{)C#6JkHmKH`XRt
z`d5<oW8TmLoJZV*Al+<izP7#_IdIFcamyPT0*f3Q#PVp$Rp6R<N7<RzSCb^eq2G@;
zI5B8g`NaUL4553yMEG?*&*kt@JQeTln;tCUaA(Neo6VS8Yq8M(%X*6oKP%F~*4oeN
z42ItxlR1-$F!=m=7v=tut2Ex!v#74m;3cc^#B8s0V<)p~P1vG4Zm3;-CvIowWOt$F
zg0l;&8PB651w6(D+3gNR`+CXKqwsi&00TYk0TcecV&Uw7U~}~YEhx0M0m^c|SL{dA
z)|;&xT8f1=H>4t~s0hhr|BBf%B~19Kv%)sF@-1INg#UmXq1(717<kB7=0+1ip%Ggi
zH4=b>)no>Bb_3TULu)-&^oF{Pm&-wQDcg9wr*O|@h^MgsZip}Q!td@NMSypAdPG9Y
zO@Dw$W6P=Y<v08W+dj(93hT^=$0>8C$wK&_x{>HI=cC<(*}P95IUrq~q~p-uQP76q
zD99cTwxwEkoYm?2^RgUTp!xC_p_Vj7ZmODPH2h!I^(B9bu1E~!>3msccxQ%_ul}JN
zew8_yeHi@cg2P>|i4y%mqXR07iJ}nb(KBUuOR>~D(PlF03VPT5$<lRCp8Hpt$k9W3
zTh@TmYalsUhVMb1zj}T(RkgAJTD*5$;j)k$YWY53W4`-atL)+BUgMIx&=Ako%HF)N
zrp4}D!^3ycK6Q#^OZH!P@`;?2EUy{bEE4&vvO+wEVyX(pzFX0>GC(z4b6v>O^IWGS
zFP|ptfDj}B&ixV{tk$foL|kse*F1H;s+HdD7B>o(clN6c;&AvL4Kb6&^v!GJ){l>m
zY92><inI}U`h>*cDv(G`#}|d1<@%B7|CdwMx8(`0KowJOB{4Zm8gck5Vw~Hrxyt-V
zfVhX1%wH_J7$}D0AmE(OS#%6m3Cz}vEGzja7PBh9!(Hd|nJi@LQ&^Y#$WIesKO&4>
z`)kVE``2j%e2xz4^%0+sd0p-}87C%7zCy6!(?jBZ-$9BXvRxU<fEYd}d#*U9QFZRJ
z+UwI;&Iy@DXBBE;zWaB{oOWnaC-T+)hYO7WwA5^EHdcU;`6c!Ml_Ed&<_F>79Ie`J
z%(I%+m)Nvg1R3<rzxH3fm0#?cf~U6~8I#%iG?5p0k&HrdwZD_jVROF4d>XrRt)31u
zayGODZ0-(i^^)b2;@NQJi!~!y!{{^;8uFl!J4j)@@JK)*ipQxUEC0XtegE^WdIptG
z=!TJ@QVzMfMBl8Z*dv?yGVc8%ryU-Y!D8bj$>ZDuft5l&+QtT`=m?8DibJ28(ORwZ
zGBVDGj@a53(ao03hpWu=I1g9s=2%_nLgvac<qO~${P|eA9N^gK#&@pQ5@}}b#%j0f
za-1(ZI=JAl`RxlZprYvzhyAc`SM7t%r^CTnyg*~iHa)_?AJ3+?_TJ?XSb!TOBNEre
z?@mTtno1Q|TliUMcZX`KHYgv>!cm-3%3Z?{5obab)enk9lJ8h}61s-V00UaCww;N5
zAMQAQ`ei8$f-Xvb!ov(fBrDm6TO=$`PQx&sfN7ywJfHg;3e+wPlPQ>;{9^A49pN2i
z(ug&DyC0E@&eX}qli!M$d>wDTF*R&LWeA1F@46)v+_|5BLcTaU60@5W+hDWFO0IL5
zl6kegv@|ey?us5s1Z5on3R;_OJ+V)I{}X`!_Xhl*4<Z68n${<FN~X9GD}~U#ry!QC
z%oOv+=3rI`^i0#KgPr|O^*O4Q*BzXgaUtWh7f|GqZ&SUVOY~$<(F)^_x7X#zTiB14
zd?uixp)pHFQEO{NBgs+Hgc}~>Q@eU?&Z_O3eva=@GZV+=;OqJ69Vzf5Uvy<^aZ6Ql
z`L;U|2)*t{8oI731ZU=W+=_<79AMC52niMY{Kc1;XL54#OwI#8iCD~j;y_pgV<GL1
zw0EW{$L!D?TF*3#U}NJ}Sbh#)bbOAUdCx$>BQ%+7ruJeWUWq?;T93l@!7f~yBQw_h
zX+e5-!AO0cjL^zA5&*1XX0z-wjnhU&L7rlAHA>(?D87CB&`$dHSfjEpWrz=XG)rpI
z&13tm&x`sUFdh{_nahuvQV%4NZ|a%aO6pJVr|ak4TJEv#GayEh=7(xnVO|GG<^OSo
zOXLsN(uycSKa@=PPQ%&Z@YhP8-scguYWr9nB3N%uTg&fbMluPJehLvJ8?avdEh&4&
zS0j-!UDX-4BO+Wzs~y~6SYd|20->`g9PFJ?ovi83p<SnC^8M$?R^nD#H^EtnOD3;O
z{)4_+Cz)r}c%A}-+f$<^#`Ua;6KGmU3MBNYW0|^T43ct{F0^bTg1`Lg!QWboqV}<E
zf#`*ET=~kU7vL{km>Xy583plW2+4A3i>+J1en=~x73p?864L&*l7PbXMnAk?Rnp_h
zjLjg;2gw@HO6z}@ty$Vrk4g1%?|%iKkla2F@xuR<?x?Shc5`@oAi&{fKYM?S&xK&`
zOP_|%)fY{FKStk-@zDpyI_#wZ$x(Uo?~d#VPYmw=vVpD8k8cp<en2oB-QNhP0@Tav
zn%UEeu$<(;GqX~y()wezG~J`k&*DlGHFzs3*@sNJIEs%T2v->UX8U65K&!^(K$zKD
zj{uY<!7={TOPT=0>gmtCGirn4!&KYeGM*#_5P&P^LB)8$8l#MgcsLa#GLL+l^nf$l
zG_JYj*jkuDSfoMTZep*i4##8~-}S@Q`=D+&*2s&S&3K|pGm{Yr6^x<}h-oKlXWQQ;
zMk&yM@8Giq>)^1aJ861(WO@?3TZ5s&G3iU-;7ue0LsfNf?6T~IRCi~Noao{1`rzUi
zCx7H|O)RS0eY;b;b^h?gG_*dl5tAJTY&9dS!uT}bvn}Qv>Aw4NcRAPAv#?%{(&G$C
zvOI7rM+%McA8$;4SW=ngB@e_g9`lpcXSAB?RNEolC$u+cdJaf_1viFslVTzSJEZK_
zVCE5QJW)pP`Rj02Y{xp}8Pqixlw@8lA<xp?c4F5Xo^Nx{^;hLHF4iy@2EEIcnF{_|
zYI^sB(rLWh;7g}^-l>H8CZMwtW^{#ItoB!#;a9lU2JFc?eO-1{&z44gxXI?7f`&i6
zNs7kxz9W->9B*n=(t14*{UHR8*+P@I(lEk(OQ8m=bV@RCAtY|rl1nMs2Drb?t>Fr1
zm!1frhN%RC`bU0d+Z=!M`T;2dncFV1?I(XE+SILN<d5%GCw&+JEd>JK-SPId|H|CO
zW%Cn)@oYr^^iF+R+9DWCb+2Po3Xu5nN2B+^K^y^Zh`#7UrnHZj+ao?8W{m(dc{hOQ
zO1y53*<|t?f%A5%JYm`BI-BYAJCF@0Ebj&k!&=!Uu|A!k5kvh~1J;YKTyIZ1gwQ}n
z_2#^O+oPXaPlAJFj|(^?Jwn{2_ZVto>AmwH=HRAkN)wSx<cu=7Y$i|`$S<l3`kn5J
zNss#9g<CidB+!Fq0#jZgeO9MaJ$FlGYAQ{VzchJ&d?0=QtSwm~A3^R{y`EV3akY-{
zu(i>uP-tYxoo`1SN}BIu32EhABJXEOR_ky^eF_Nmc*{k%9kRv%)`s=FeZlkf$Qz*I
z$wl88NW|5y7N&ptfa#vGaFkfE;(Soa&ACT|dUBXWa)iX55};TlUmuw^U^=HLI*}*F
z7vC2>TAm$~B{g=18Kb$YBk;wY3OTN-p+pBru1??__a<zHDU#RA9UENzG64>$O`%R`
zHVZPIiACSNe_Hl2M71k8!<ub2l6%p3Yuv=q`W7SRP-*<&wBHR73H0|SYZ&m%;lX9v
z_k`2*U{6@zzlS*%t6XMp1_7p5O$S5^TklVc@t~}ANH0Lp?P<$VWe$^d53#>c=S;}k
zO%}pv8kM$k^04yC;N9!#QXqb+x-3QrJhr@XNZJJ>Ox(E%4ZtX;Rj+wR)hMov6(#zQ
zB_<+<1a8)Hi%MKzqVUl@@bCxiMgYwd2;GQsEvi`!>dA;&J!1p5yz5p#0;UvnD?v6U
zs6COVa8J}BCH0e8YzTsMN?hXERW!h>0}ZyThp1qHn>N!*Dm=O8EFv%|q5It2;~7X!
zvK*2@lKeO4AkimCB*mx-LoYrGE~Ds8@1v6Dd!6B{r8RoP$3o@LDA7ufS4%=+KWZP0
zr^q>%pHVz<^fj=WZ1%*Fbux*_geFizRxc>(s5Jnq7`mU_9P}IJBiV_YY4lBRQ_6=f
zPfUC<-}rcOt&zoC8_qDQYmvnE$*c8Ntr55sL_#+hGf>VVbToo4@Nu=-bnQpqX@YxQ
zzii80A2o#Y`6nOasX_Nc0Q+Hn>snnu`5{gE*5ZV&Z=y6k7~kkjs`X@uZ?Au#ovLp2
zWiq8JaCbb%Yq(BZI!*A~UC51${4z8MW=Q>4^Qq?jG1$YN5M;oS!hB@E&JG6gwd{M{
z?hjlq{reI^m0*KJLS{9PCKv`ey2(~GFkc1s!$C$_V_bmv<xznb_#G0QlD&bEKMwp5
zpdjZt!ZrkMAh@|4tg5h;`Y8VUjr_;^L<}h`EZkr_@X>goj-^ngoLn$7yj@Wq(h{ND
z;X6=HaGB%<1g8HU9s@6(()f|Q0x$*>`dg456C%JFm&N+@r8-^R%@<^z2U}jZM<<Af
zGa1a=wXpYCdzixoDo*#V!|ihm4c~Kv0lL?l2KPdt&^XFww|d3#cZ2DRo&e?F%60hr
zR&kmBK@>KSa0+N{?^%iR+VYDqhvj58e#7(IznB)XriVAEv=}6JP3SqmA6#y5w-X^!
z;3uBL<!chgyUURH+GIjp#-&JWNes}GWc=~c20glwR!xSwbu=3t3)7M)a$|nBr24D&
zhbtC`V)^9Td{?fAvTXjE-$3U$Gkh5#B5oUBPQ|MU0TdUG)5CA<2G>s73`aYRLK$#a
zWUnGp?<b2`94n_rZH}k3U@msn;LS#x>~~f(fL_ew#5{)|8XD;*J3|<BgVW{%IR>A3
zLNbM2j__?NbFQm!{PxV42#wSjOwG}v3;hxX8E7BK!qJf$a&_u~Y@`nGON<5+IKD(D
za@vV2Oh&OL%kbX#TF|(`9=f9aal-urqYeFS^EJq1_X4GB9W=XCaIS#%x#{pdjPH>>
zYKA;<O3k8C-~9=r)OqwR+4H~YO?pTmU5*Zj?m<G3W{QHWPKjmo^-xOl6P;VLS-b%E
zP!z!MiAYaN&7q4mi$6DYpN^uuSsgmCzkjMcxA~mvlepDmn+)GLnez!JCdPm^wwJ_K
z7K2zj_*qfBNpT46ex728vS_jlMdA<l71fJfB<GtASJp3YS&Qd-f?5)<dMO^%%~559
z%W}UZ$Foi>5w_?K;Gnu>*J-pQR*E2F$G_Ue3=QKu%zk`WkS|WUxDm6PSsk!A`*L@%
z1_0<yp)|*^h~4wM!#!<$RbG5ITL!bGq9Hd>zg!#y2YzrTAlw=IEa8tNZDS47#UF<>
zrM{L`XBS9U(_#=hvz&p5z38wc_4Lqv#O9b58>2*BVU7)k0Xh2p`{r|TykXY&`{W-f
zp&B$FFWIcuMR6E(pjs~^-k(B&EE<QpgQ_`LIwj=~BHn+kfj^>pI-*8y6MgwX1SKoB
ztfg6gR#bnO1`<^wmI&t}NTa%|`rt_}fQn)a*P`Kblw{SP!?Aw>DRN4nxC#6z&vVq+
z&e9@Z!mCgzeuhK9RctF7nVgt&u}HgGq}EoNKSifzu{hLwVZyfHkbLXF-f?w!kad4Q
zu6~8W=m9zbaIT8tvf+TqLGCqSTPY4pPlD)hL0qJgCSBsk@VqtE$e2QVSY}hh1(<iG
zg70_X&Ue<TFpl3-+4a+sG3$PC-$hPp!?S9fgOfgd1nh)21vue|sdgq5XG7^{LsK#>
zExO<O>90;+LQK+)D~~LPu$|5*aB%IyG#MmrWW_NB4sv9V!#ONA5ET!iXs24;3N?6n
z3IIHE+XuP8>9h3|&s7i9chefA41_@R91KWJnp95H>5>l0-lR`Z1n$?Lok1c(+A7)m
zzn(n4=rthjLbw4JNoX<qTMGzAF#HwEK-o{P@ERBpiFVZIdIRfAh`-tLpkGNrLshP(
z6OE+Ehb|*V04?^@;!mxj)rH~GyNk0DtDj4@+6bL~n#6T3yOcE&<aBAORWyJv<^Ge6
zzGYRzC#2(jty=SDIROaArATzE3huO_%hPJU_Z^rd%TG|tSE??$@_16Qh$#$2HJln0
z?k3pF2;&_M#@BKr=8sWFODWPAqo|WCIIq7dqh4;jx5mbIHni8+15Hd;Fx^WO_Pw_3
z<lQ0UORslMvlZNV8eA?N^{I8Xdz)GQgsj_H@#4R7KBeCSIC{rO!g8^3pD~%N|6?tI
z3l05VVy2*eloD~un{bWBxS}6u^#6M16a6P2yk3egmv-OF%jN}-P9EQEoo)BL8G92u
zN^t&!yE|^B{PL+Qq#TPd|MUrfG=QL-`W%@i&Hf-r%Io-a<Fd29ClXsP3z-jL6kK9J
z8N<LO<a|@E$)8Q--;o*0Kan@t)}F{^MpZ7Slp!8D7Jym`eGb6%%q92!MVpe(g8OB)
zGkMlM;nQ{l9f2PAC)~i+`{d2l(NWe~Xul+(MfUjf7tR2*km$eKjEA`0N}k^8mu^}y
zdVH8#b_tn^cEm<!bG|IVSj=1rSa*Xa4M#7$2HMmNI*mG^m14j(mroG-b6?a7qyXVC
z@If!jX;bVxw}ShZ7w~^?Y;#C|&o*PUs<B!w4-a&#jK7c(0L;K}lBSaVvYQam{t2N3
z-q7?!#frhV-rX6r_qS(5<B99fu+1*_J#QTDPoxKz1ObziPG(n}h@%rNRnV1^m<KA9
zcUBm1LBQDNf+Z;3d{J^AZ7=g`v7aKQU)3}*Nmg~sxZA7RKn9z|=zF4jRHZS8q?J`q
z=44%!e)aonE>mUumyzufFC$KYuWczx!mm4lqs83%9FKd39_d|8)l!nGcQwNBFP7NO
z+JecC{P+jaL>Cl(e1HORCCDrEa;+<t;qfA<?jGl!9E5ueqJJ*HD)~ocWl@kLr#yex
zi}atZkto#WAR@r0ZgCDuUyS5@io>SDcPXeeG@q@QhFq*(d^sG#c568Ba~hJCCt_M{
zxo@sCJqAG5or6tI^5tJH12J`{NUmf%tfvcQg$D&WhjzgX^Cur4KVU(m(<p!3RYoNK
z!s-VG@v}`l_JJ@6J?0T6$-L$`T$3_;Q{7YD3-<986t}sJm)66v3{EsqS0g~e{8j;S
zF%Vq>w#^{F3fCB)#RLt0;>^ZdDd*QI6;uJNa#&HpWr5Hf5eQmps4eTVL<WcWzg@b2
zJ?;Noq<{XHmk?~jne?!A^<I$;x{zEPk|lS6tgzP002@09i9&EbXEF^QXK6IxQu(yr
z)>0%Re2i!9tguXNGEF@{{mJds{8%<9h_za=8M!~XIgn_3wNMIOwAK~HsZHV1Ab*H_
zHKDu$Za-HZgX(cwz9j*+CA{Qh^4p30;!ulZ<$}@Lhc<YP@w!SMEc#HlL7k|zcnfda
zd8<w^`@8I@*9Nfl??!UrfJm`Xza9maf~b%q;cMp|5ML<Q0(;cSKF7fSOYi*OZxcTa
zWNAkx$mzpxqsty^O`~iKzXn<a`^0ZpwuS1iGk_F13t&j!0_@vie55&hCBUc+AZukj
zVvu=fgh~1jx3`z*Nz#iH(#u)#^R^wRr5oBlFi|Cu#+T~oeD^4GrI$ZV%F^k-+1b-I
z&kr1yl*%3~x`Rr7)@EqF7JzLWpygF8WPBUFcX_B9?N)7X7LGsK*q~#<07gXTZt14G
zqcSK*O*db70S+!vsj}(|_J2=E5-t4k`xX%0@GIwg1&D>AlK%Ui^3OBlZ?->5Fuzr)
z0JYMtpUJ>a4p2W#>oh2Li%ncUTse%MZa5Pk4)NiR=ZqYtH(qIag7QORuIeDPzoY&c
z_Ur!gVl{)~(@Q5x`uFb)V+Y$zCSO$;MG9Er(b17xpXT~(7bw!cU@=-vd5&bMxUIir
z^;r0Rt!rj1y09;<e@>n51-L<yD8Jp`S?v7;p%}Z)oQ5Aw0r@52`w~vSS9OXf?E5Ko
zSeN}3q5wycu**5BLluLK4~=cqz?BmPAN}F6ULXW_4x5l3a4fqC=~G`bLCT}~yEg#b
zW~Ln6u<l6fZk>Q>p&7qb#*OFb(@J#n;NI6S@t{5Og~f(0ZfNUGShry8LIh%JPEG)@
zGAK`dD+>b;*?BM3dte{$fz_)g_Z`a8txlcpjJ~&k>i%&1ZiJjHBj(wxUe$u`2qEV3
zT$V2*l=$P58RC^A+z3?53thq;wzL_cu2&7bI7Q__+`NMLjhYsUmt-JyfO*ECUzw-O
zP5yky5c}rlMn!JG^s&g-9n;O2YR;H7l&dTY*yjE7ZTw3O;Kc|<^wzIN9vhg*(6#FJ
zM^6rIeoFor%h)LcYC~Oop|@R&CXLQWyQ7_AV8EZoYL+mb;^mX4M6sGIxKrFL?pLPe
zhM55k?|pt)53$;PUe&G}b|W=1eSOlqhfs2Uw?6|xC*m1R16PKg*{CyKC}F#uD$Th9
zh5f{#Vz{$zkT<D*u#s>hujOuZknuWfnEtG?cL9z6EsvQhFW~4*HJBF|GdfzrJ)8!&
zrimjzN&l?1e+8bRi4wL#est}r3WL1a-ojjy=hb?ojZDkl3A?1gMY6~HmVA;DF7rJ_
zA5EuDnIT?W0OwX8zOg^5yu-f@sb1{M_Ce|^x?4F!>*FF0$MZ_~2oLNXLft`5c@A*#
zoBhG%)%EG*H;e7Jm3sYZGXU^3j@p+%2v*#p$J4{{+p~wJyTg9$2Lb(6aD;N&?V>!Q
z!SzSeri1RJqX2ui3fNAn0;Fe!SgX=_!VpNQH(x7P<^aRhh9B)>n_6kz;u)?HcQmRM
z0WV7Hvv~`&nt2%mZ*G9KyF8=WOc~ikv6fbC*g_A8e;iBTurUP5$mbw}`M97`>YPmm
z+q>p>XwH|%fq1rT5LWaN(4NPra5~zzka@2?El?~*u(e!UnZ(_*fv9}|BVd`BV^BeZ
zz>e^@Op}PyN1FSj<697w`^^S+Rgo5^AI<@S8BN47NYsh73!9(nm4lUGU-9iRoJ+>0
zQRYV}_xI<yoc5Wmwjd<HJmZ0a2ZoBEsJRd41y=oS7mq;HGXk_wH}L8^Z0zbLEKUpF
zJWi?~0jVt)%zihu0Oeh0lQ3!d?x>mV3^}a*Zug!R&5TH41^gnJx^6@&h*|`BV6OV1
zzuv?RZR7q<GI`XgYgrFzLMX5rcn{^(gfInlYoZ<x^3PV{GGCC|h}tY|^vA?^2mabF
zX-8rjGSX{U7&7Dn9B(Um>wspE$(WOK-2+=u4{TzKC2ij%eVH#%Dc7(<iR1xRM%?^v
z(LkY0A?H{LzJIpA6WuW1eqsy4rqE~Iil7qqK=zSgaIj67i@o0Pfs#-55U?JhV{}qa
z9)W4S0(np@N0l{0<}+n}2JA|yK++v%z|NL>Hpmr5*Kqv)Of`i13<_<ni34O2pSg$K
zxo;yQ5G@OPeB-M?pJ$Thz0qsqDNK8DekBn@w=m(ctm$Xt?}BiPd_J<xH$3FkYXl5b
zhux@Knq61jQ(gk`c#RX@M3vC0B_19<<1%k_k6j1@au+VxRRm_!C9QZC+V?&sQ<^Su
zYlsyF=fGOpp5<<~M>l56KvGn>#`1HVpiV-@%ipo1R6ji`xW^0Pt<!}`yS~0kUUB!e
z_ghQjNtLV_9F9KrcVTi|QO#Ga%o7fssfJs!T6EgNcLrVb00Tr<GN?cORYWT2zww=D
zz+9ny6ZoX`ZD91ZW#ei+<ScJj=Ue11wd?j*5Zxlpp?j|%@cMiira;(;2q>sz)P>s{
zVIID`El{szJg@pi-9?gj<W!?**({~c&-axTo4`_29zqv;Lg>37&;sDm=U$fL`d9IS
zj7eu;R6k(FSL6>~yeZSW)FbihP!}L38Us<k@v^DLdatcQEr$e`{Ek+az4lO->==^X
zb9FBLHt4Gw(+%b5x|ha|li^POq@;co0nr_h@b>u#*KGH1$D7c0%diixPa4s5q2_H$
zMh7dWG)>$^0)EL|ygqnlmjq6NB5o_%l;o~!x^w@d%UAj0e^cN5*NMu{4pD=x>IW`3
zWiz@fvloP>5Akc81DO|2J7j$HVTWr5&6hzvP5e+I-Mnt|_qb`19%}Ma_3myjOr8gL
zY*~+lu%YqAFiBE$kEBanJII+tr+()?ff8zfkxELyzOv9~Uo0QGg>!O!YHVEd3&nNr
zlnJoef*IiJXzeUhW5YJVp9i#nTG~yGI(d(_6(gG#h}|<Boj1j~Z!B+j=$wxQkQdzV
z%nQTlt^n2Z{+_XCN|yI2Iyue;oGQ0fKGqILNopiYlDzEeC@;98)vnM{vL_4TUT@Z2
zYO^&-V8_@)_GL*4IG+s=@l_v>feDNI2W_^gvYk-A#oo}-8~J*Bb4DsP+#6CTW~l2G
zg-QRr!+PXKmGLEDdXrAW#a9q2o<FOF?Eq}7gg5$;aS`29l*qN*ud$*Ask}R%clpAh
zBe+|tk`rohw5_Rh=p6e<=|0}yDEf=evAQ3>a)@j1=d)TB(x16rX=K#GQa==L25N(J
zt!8%^c-<(Y_Aqqkp*Cn;U`GNIyl%e{8N@~d$?`i#)Q6CJjt;)B!<#|3#S;Q1m`?W7
zx0*BUk+8A`kXQVTt9TeGl0rstGw`;Ih~WoKkrGxL)fq56$v$vZZm|GH@;rw&7{*+V
z`lx;<AiNgmo@jB>%8o{vMmyypi^@~bkwfqZ2--H}@kxqYmEV5mKJ9q}1CRe><7Jf0
zMIQ+aFfP@q7*JLyk(1$3a8k#z$o)PSI#*+r4*rY#={~`U?j`Gf+WJe4_Px!L6a}Ym
zDAZU{@oukz&+QLl`q+LuVV_>(bJ_;ny)%4)&t)&a>3>@yDm-_US%#PDU2fRlV34we
zcv9gkTn8`|TrT^4SpCl>HDa|#!1T11I%I*3rh-XX)L@y{Tmfl<73&NK&@U>yP6Wx@
z2iTShXzqa5ShQuSBZ;3^Da3A0UcCYy{3_ykJq^OQy=(5z!cGO?)aEO4EkG+LQgA}R
zqQ&9nXfd8*oU?q2<8`ZRB{A1`S9A^KJ{5g4<6a{9dG+4&E=l2L-&7Y>U0p=~Q#MF0
z+rp*62-%ya$zY~yzPI-2bg{yxM-Lp)T5ZC%z*iwMSyt_Ih?9STW$<%JW|daX5f>N)
z#(SL^uz^5E@**2NcbW!5PTQOX!5Qgpx^}TBxQ)>TEyvSfMvJ6`>U>vjj?R`>5er}&
zym`IM1RoayPR!}$UYPn*u}<8i1f65OzMP|hJVxK{(w4az5Z;E6^eoDf;kUD={(}?a
zaS``Obtoa<;;<DWz1%j>;Y!aO715BgZv9918clOxN%MS-;(u+0VNo5o@cEvqN(JxO
z>`iw)mf-QZBBq9T%jtk8yyjYA_5D3pM3Z#H1qfo&S@S3Se%DBa0*f=ms|M;Jlc$=i
zdEI0)F|pxSOkD=!MP__YJ<hW?z@?n)+oyEb`~5y(ekL^dx$aV>nK#9(JICejQh0~6
zV%<bQdzrs@!wJqJ=3?)N8H<+c$sp43h%V-F_Ps}$ncATSI>@sK?tL7Wv$?8ME+;R^
zKkB-R(V|;fYH93heX%wZ_~<%2^VNn|$2Tlle|^!Zx&}`_66GUWNfJ({OdRPCG*4Wa
zjCL|St^*kMtX7p=nPgsW_&t~U`UCN9gCSdA-#fhdw&i)H*KNPzn53%HK<P4Xn^U1?
zP*Ye&<gffK%QmSpUgj<SeEU~qLUYbHWrl_^J|>1P1xhr8++f~i^E8q!a)t)&T1?gb
zEQX0aSr~@>EDulP%~&pP`)6@%`A~zpij@-@P1vWJy$6s$(YV)mh^UUck`=hO9|erb
z?(qD!;2oIPJ4xPp@0k@mdm%G1I&Y~WtRjo5tS(v%*p~8&mdL!onKFHK48L_%%{!k4
zUM>*2*oqM80LgsDvdI_MNgvMghXFC{$^=!5#u;$+TGzKXEh^WIQo*TiHM#~vCJL;o
zw_%XopUz)cewKFNZ{~D7)-F%(<Y}kl$^dU$!OpeO1__?U%?e|HI}}~AWu|1tlE29@
z_yfgW5I7PIi6;<Rpq8;d)E_mOa58EEW9h6nRt?a8awIZY#08<xC}ius{C@Uj(EL#7
zqCkUV6H)k`)%>F<KCi+u#Q!<V<)_-M{7bknL9Q-k)WiJJ+n>h|)+ao8B7#>Y-(IhH
zzLsgR<J9lNu=zIh{vx9b=>(50slt+Wqt|cq0c}dU!Or*WI>b}}Vr#;&2s@v#ZQ~VP
zou%tnUAG4PmqWPuCfW_1kmEI7w<}Kan^%$KQjAJ0{7cWXjq1H9>W2%f7mbTfy^o}N
zVZO;Zm8Cf~x4lALN?e7rv;y^S{ea6g(C}gp6;+M3HR3$z0BgFnTe=Vz_FMUhD~xYb
z$2`zZvj-Mdjo)zD1{=s%GHCx)>ipYHN`$hPJ)k_-A5Vg3S=iR*2l03?+24$=l1a7j
zRk#iqc<B0fxlTfpQfPVUS`=%VyFY3ea^7_Da+boR_$qx5(o~?gUMg;Sd1`UhdC#Q`
zSNtBUmZdWJBywSZE#}xJT*rM7tKKx1$Jx5JRbiS6&q~O8YnsYpCX3>o@ehCJtEw%{
zyS-wV;+p-9rEs?fNV}UIRnWH>C#rv9s<EER`Mhwo4X0-guimNar*5<42Y++AAz<Hm
zQ65~IT-3bS@9de_0I9>EVd)j-@%(Bm2(z~`5zy8fUfMz|Sow?WbrA&Tn_@%O3h^I?
zS|!lJ3b#Fvo~S@S21c0KVA?k*eHv?2B19dX(D}U&lY`&Wt`!<PRw9W)RR)%!RERSA
zOo~~XLotDmTfVD(M-Q@obOdPX^nx&ma43tPL2yRnD?tYctRgxdBb<DtQ#Knr@Nt~*
zv4Yi|u^KdH^HvVT=a=#_AtB+=*feBA@hVR*p$mMi1)QYi_%p<RP<?*I1xY<eC{(g9
zUC3#xx=J*?4}?7$wj;tAs%<=R&p+Rsqf4{5Vtgcvl%d=2r*r5)ihx>aNF1)UF=+cf
zHG@OsiWpceuitSxzyD}P7gsc!)W;qWO*DuHh6Em^68ruLH0?-UgR7mvk#sn>sUT{Y
zTKFnZ4+)h*dFUq#b;LsX7ed%N7$iKypr54`E@co~2<l*-POPS~BUBNq?AP>m#<Pk|
zw+lg^mGi_>Zx!?vKS@07)k5Ymt_)~2!d#RYbyI4PRYSS73tG(!C~F6&@#B8`>qYWF
z70;l(n}J3rCK{BqclCjTK%v)Zd(XHMXOOhAF()nNR1aD{oX=?_U9ZvhQC~3|qI@QS
zYBC|mA(Ga2u=Y-28}T%OS&|y5`u^bV(_s130adCu=j!P8?gsNhQvO|k(aC+w7M^>+
z`E8LY@pR|auj3jGT(gE&D0bl0LcTf6Pa^iW-M73kq(Z)`65pIVE0|xHpyEdd8ziU4
zuB=BUnz}3Y45=(IjCDmm(*b&z{_K%-o{q3y+UFf3G3u-Of{2aB=egPU_cV2}j>O9S
z4YG|9*fD8?iH){X8oXFeF-N87h5h?8lOLWZas>-feyZx-@22(}1$2?p;nBs)uib@V
z+^btl3c*H;4IN31U22OrXN6%Pmb%oh<b<G{dz8)MFhjAh;O>|850~jYMy$E}!+N5|
zJ9AH<K$8JFinc5{O?Rx1zi#~!<qK6&<8?8|0Ece+$FHF;iZEsi#?*VUfy?l$Kp4wc
zKD4#8oY@=zCy6Hi3UYLIjXlb;+54FE`~`NWG6XWiiS}TqqeB?a|3}wbhE*ALTf>4N
zB_+}g0!j)<m$ZPCbazN>x?8$MT2hei?(XjH?v73KEuZtf*Y(CZ|MVAo?|a>I%{k_X
z>-2v)tG1@Tr|><`luHv0`br<A7sSH}jOzWf*4r5TUT`{D3x{*GdR!#{((REuVi|E~
zav7{h3dSU4rI}B8JjPPsps*swMU88Z)@k)ikWaPxq7Rl$k{_PpRwpvExKgN95yt6f
zdjwp59+-$KkjGLMJ^9r)@BrgYr)~U1a0L#d<`?T#8TQYUS;+4-T;PLFR#~#?qzt*P
z=gpZ&8kIuBBJRT6O8{f?2SyeJ0-{D7gNAWK2!^DIx%W?ap;F*w_aQXM{EbYc<8TX$
zFDC{0JF$k!joCnR(xY#<A0=M?7!}c!?YI!p@Lco9ZO@E3A#d*SAca1w!fYUYH4{wx
z6$%5jtNdO&woq5o!KS@MM5^Yr>Ci*$l;H8D(9;zReAyuENYq%*Gr;-d+>3a_*qaML
zcW3iAC_8Tf;uvW$$7H%O<Lf0S9oxD>ufiJ;+CF%upP#H^F!iUgdSe;g$(%>%w6T#l
z9}MPSmEZGF!?mHZM@_MO(}DsWbr6`F-~agXQQ5j-sS-d3@`8dsM&C3kttI>ytSMkN
z>}r23B)(s>=SG?#qK=wFwZ3Qq?bIQ4$>iRt$6J}lJj5IRV4K2}X0CJR1i6r4SVwB&
zRfKimm7)=GAbguyr00*q^QdP^x<m&zynBCiiCfQ$5!om7cn{@B$v7iMbD6K>FFNmR
zBGxzhA3LsGg&y_AtBVm=oA1=S*f`w!$OZ<n1y(v70#f}se9+2bw?g;AYIL9>RM8nS
zc=T7IU6(i)0rZoN75+@oa^ZlR-TeZOhlZaHBzyDyk%FqcJ2}To9*rTI_P65>iz4Tb
zM;yLgsdv>O7uM^+(<5wexQY)AB42oBrhFuigtRQB2I1BfYkn`C`yD^bJur}Ma%X2=
z0ymupRP)2nwHwLP_=acGj9P^Yd{Rc^Bu}?Js~7m2vDhzp+uCGhd3}_6J?QrXvWEsn
z{)Ws6)&y03;;GYN%$P0;Rh!t%Dkwfbt4IPWGKAgC$R9C3L?2@ujPTKyfJlD2@WW=T
z>e70B!QXdL*ooecsa%2YbrV&CjD=x-<RM{E`|%&U^4;xbO1JeC2R*lZ2cVH*%q@Nf
z$wL=bF^G@)VNgt430azdn<o^6f+_apZ%|;4YX4hw5psL^x?dCm^m+HKFVRm2%f|8#
z6$qfS29dF9)7KymaeEGDY`6%c-*JEP!~QG=Str&iI{`jfYpbtjt>;n8^V9ZAT~0M2
z!!s~aoT6^v7k$B{Cr2zj7!Neq@Szpa14qfyP~OoapyORX90KQqf2SA*s=?e%ytZrK
zgF<k~C}S~b*NQY&ihcj?qWS?7X%WVfdvYGDHWx+`T$Nfgb$ELDP&?Qhx&}Rvw=5}$
zIPAmt*&>iV$4;!9&Hu<c&%YEN_TXnN{0#bHH~89saeq(mm6(vj^IGG=O9sP<oJq?&
z)MBGMJd<Q|!!>|%1H2i<3v9-`BAxl3o8Vi2@)u)f{UQiE`Enht(p<}QGB@gjV>g%Z
zoRB5*Oa~zLxZEZsX5Xmn6eEx$<E_`{n!>=MPkDS%k-eni9&3W6orm-oA_gfdTD~!A
z-Sh)!O!#W576}XmHD~xf?6Cw!a+^dGPWGYAXJRt;AH6tGv><O`m}r<o8AQY*tS3OI
zlr=kHyFcZY1E5&b-fZ|2A{Md`Opi#~C(zxc_TAVgx<THf-dWMRphf8w!^bR0PHF-#
z&Wh*O8&v3o+`W$~5#$aJH@##!h0xj;j{tJ34D*;^7UomV_i)<rVD#~f>DgLB37!Xw
zf$TFWyG4ST6DyiWJ5cHB*7L0|huC0vn)>{F!&sDyH6Px;MhN{9ru2~n0v12lczN^Q
zc7f%7foB^%bnu=Szewyy1=l2op!v0O6k$xGG9P6W;H?Fpk=l5uVFX1&IK#TE1w_4k
z+<-#SGk<<5+1?JgSpANUHJdKX`x<@xWsz0xRpDYfaxOYimMr()y9?JgTJk?`j3Ns+
z$KHMH^%6MNgkl=rZUsxHz(-Xh!e}(-k5@-N(v7z76WWGMrTo_*;pYHs)Q^c>l4RjW
zcnFoIV?t_`YaM~{YA`Ph#sT`tOv`Yrkk4$$YcP@fl@0SI;?3MP4G<+B7LdOXCB?;8
zw2AuOE>LjC9Oee%a=v@w3X{SB(qP>UiC>@7Zl?XZ`ByZ24hRtlhGKg77<WP**5$MT
z{$oC&$1~5eA9&LI=cFN_Ht^&Z9_2_Pt9acr@Jkj7chTeo+=wm5hNxYz8sJ?e5SklR
z0Zty17A4m0U4Jz$F)^YGVE03m^jqm{e=r{J^<r7Rz5*!TpSc~~*AF01#+2rAD3xzP
za}F*#?cM*dXNisf+!28GL>%q>tLk?fx_~KrZgjj9{R7-=Ce0t~ofHU8R{8G=BB2@x
z<Tu}fAdrby*~4p>zw2$$xpQG#TFQoVT3p*O;h&Bu-;#sBzyPH0k6rc{d@6bE;Z!Ic
z7bEw#9A7-z$XXW-A`j^snkTWX27J-C&#X>mJjGTkJWyNC#Js(DP)})_80#-&BI~?p
z`<$Cz9Uxu8YI`oWwlq;cy=ICvl73e*oTBwW^lc^HV*afc=?Rpuvz+LMHDIQp4w>&=
z3p{;fVMijzDd`R;&St%In5mnim%j^CZ8UU)r*6$CEJO#MucWR8xh;UC)}uV9_T_CA
z)A>fAOZBzvCrlLDn4T?!fL8*WRVfStm#nI@8H1su(}ah|-+4+DsSI<-j9jQ?Q;BBC
zRaLVFX&&5taZz|*<GfSd9?E#(yJTB+TV_458aN&2;%8+T$YQB|=x$Pq*QLw+<%Y2z
z`p~+_x%pGL(gm0ZDD*#$WA>p3#BS-z8*2t`7}>_&`lWH;1a`*m0C<NemWBs~fWC5S
zyQ4T%{@$;@m4d;y3ufMflA|;!4M6VVrWXd7a+{3uX=?9^7h9AHl;yu~1ncS*u;_Hg
ztLV0cmA_scf+c?GVDd+6aJ|&IpA&>sRk{3;6;3muDgE*#5q;Q*1;*%gRpzM*PRDG2
zoWDu`W5-3i8%{VL{W*u~Tcc9430=X&HVPYG|Ie>vpGqAHMX<`8mDI!>$0`Qj>?F|0
zzlVyZsD3mm<(y`TJPi+yqm1bRzP3iCtKZZ}9ECYlTX7`wJ&f8S+OWRIxQL{(;&gq#
zXnW;|{7JPodoa}B_LXd0>7qa-rsYqX=JywZi64tUjCi|Qze91~?&GTt@hS5$^Xgn9
z8TrBP>mR~=p?_9vGsHHdy4^L9sj=NbVS2=+PoBQjtV}o;KBgPElUu&#x-G$=+(1B2
z&#F3!Q)Lx|A39<gex;ggerkw$#MUO$YNM4F=O1CWQMZ#?P)OQ<(-Vuu<4-{UR%shv
zoBlVeYJxePERKg+&rC1MV`7R;YLRbORFHq47zj$Ka<CZx8?5KF7{FZ~q>zX!YszvW
z;27E+JsorFjU5U>jaiUsW$VSF>C*Vl@xQzPUZ<*N4WS#Lfzg+T_CimvafyEXR|HMu
zhXL$@$PoJ1p;F+n-yt;W69t?Ml-<dPm8gmo452TqBPnli;S3Y<GPf%`nvT@Fu9N7+
z1Hi%~yGVnRDa_!MPYd_d!v<iJ+m9dITSnccYJF9xKF+SG#Ak8w?=IAoRJ+%1Q~eL`
zx=^~!M#{r|8`@UdLQ}^LbY*rVYam)boM69ve;*^{WMyT!;+Nu#QFquIss*{aQ}w1h
zG_cD1HoVbuub+*+f_nmN#{+Hl^F+H#^|TH8W8Q2GaXfZsh<a``K_9vgO5*XbXzs#b
zUN$}y7WL*TGcbG7q^cSoE+%SO!tC^)pc8}>>@9q}0e;{T`LYF@@o>QW^Rb)%zH<eQ
zID%MgrdYJQ(d8+IPyBY5QaRJ?PN0+2ws-iQ+3#De(2yTuf<CELJr%xHID#83&p*0H
z1Q5bT`00*Vsbv|NkDF#pE)RCt7#m9y(!S&3BM?_k<SLmXJO>nwjD#4mTMP^=w5bQ{
zZ-jG5s&>SBJgRMLgvF$@ylMTGs77k5)QW;Bx+|!2L&p?F#9_z#v(xwb)LrOgj^V@c
z!>^@3jc|<^m8Ir6N9#hSdQ_e2jm~97jkB1NeFQ+4Wo}X$no#5}6ijwF7OYKvqu0|Y
zH(uE+q>E|Bu~t$8_!37O8H|lY<t%BZyk}$!5bp%v_^QZ$EPYI^Ril3CKL$6<!Y6Yu
zF)i<I92(3v=KLqTzp%#0-q3YoYScicjzduoZL>XQN6>!7-K$$FA_<X0jKz5hmcCV^
z_KaN#99H(~@E3`O=l^(3!mHYcKS;jrjlrVas|3SD{X#XkWY&l8n4cP_p!X;k(S&@^
zh{o#31?2KxSfJDjZAttk`r~nyf1pyJRA8*p8R$;N>zmH~@LQ!Ra%y@`R4x$#F5&De
z8F)=yd{Pi%Zw2v`Q_1mRn2zzmB1?Jt<>M^ih^-eR`*mPA<PBp|AC%JlHA<~NQ!CMK
zqK8(@_a3+QpRd-JBAeo3q~!1adQa+jF%{xG1~sPmE7@8===FoNVT{1k>I`nN#sTWF
z$z`+N>0B-)G5<$6kapo^zpn(zp+4!1&Br!r8rN<(N*<4Y?7lw?3cUIYr#~uCN*sxL
z(6ZMg7*<pqVVL~NQE=!>4c|6nVo9i11C%B$;j=!?XTPHT8jKkLF$^F%=RYvZV}Yn=
znq%^R8$Gic78^&n#)MT9ld#}0<~k9j)KkvI(DlTGP^8Ya`@m&`^1u}HvakMIzW+Tz
zMQ#o8a#wz^g?JZUt-_?o(@{B9?s`^OM_G#xQ1m#Bezb_}yG}@3TzqfCTe!PCE1&zB
z=G9WEB9H^#rQ^D;?qn{{r=z}R4Ak!e&x=-zE;%!R603egpk^H8x51@MQMkSY?uH`x
zaex0g8llLV2yEs@s}^|aF&9I^ihKNwBXZ%!Zl{$wj&TJW_f1Umwg)AZIgX#-w$k!A
zcY)0Yr76ge|A;1_>6Nmxb5aUqJA9;b$!H#CJHw*gSh6=^fETl(V!XfDXXF3b1*=yp
z(#CT>_DHa*lTBbV)Vyw9j(83`O0^Q>PIWtWiF&}j_zUT5%DAT4A|$018YND`951PS
z#OidcsbsH6)W9R&MGiaAobi=D{~KgRv|KlGmaktkgN+3>4$@=WwuM=mTuM8b#374a
z6n1~wf$s9bzx?2jx0_*W`OpIyH;ei${(x!)C%aja+1`d+2Iyh6>fG7mEV{Y$*A(98
z9jVaY^Xaw28M)%X{WFw_T@Z>o3=s#;#vb$+AzSrWNUG<WRg<4WDG>*h;>Dd;eGr@_
z67{$Kv=R)!_YfYpJbUWXNN+(@+wlhGR64yK`-I+?t5f4!axU{zFBO(5@}Xevfq)8m
zk`Us`A*w;O#gfI;h%fX-MH(a|1Ab12vt!qK@{mm@kSjmGEFLT8up-V~>EL*%aFh{l
z(DtUxS*|XnOTGWbBFwOiq0BtL+;rDUZSe1+F?KN3^4yh1qMw%QP*#5|;WKF;EaR53
zBRjQYqB0~DJD5EOx!%BkTz~J<s`-}svJC7tJ%Ug5qG-y5-4YvxNS&`<=tr6F8pt%A
zcM=QTD*4c;Bn~VaG-AuPdJt|9q<+m>L?yC$W-4i2w>wJs>Awv07P7w$#+>^nfFGQ`
z)A#k@FY*l@pcCrE!pc~uq#PB<hn4+1rX8Apxn}bZV?Ki`Kh4!TnTsWe{_5lr$6*GG
zr?A=(4mr%55B1Mz+*}dPmo_Gn)t`i*Sgw*11=Gw1@W;7y;RU3mAwf%>VR=NcnkUZv
znsb+|(e<J%6x|y!X_Bo$I<Be<L+-ggjMbnUR1{ZBL`Boa-^|H~=<cnyC50Jbw;Hb=
z<NqW&LG<YteIZq-;lGmxd^e_OBNZVV;)^!^0jR_eJhG|RT%MYVR0+<TU(g+PbsrnS
zc<^Aqm&NaiMto29k0JU`h5*L*9M)m$Gb`D#kwJ$#GRymG#`>L9UiGGW?=*$2GdtRY
za{qHz=n@$S)}GtB`eT{sY*g}wfPb-`cc}rW71{3HS>z-8od<!6^x=Ucjl^TJV?=VD
zrqr`Z5V~>#sQe*$Y84a&oTEMU2`>afLryrJvZQ?D5Ugf~m#g#=B|_NCWnWp&Vr-fV
z=uSKIzZG~s)^4Bq8+nh$lwvE%TxF>t0$}iH-c*XGrp6#N{@R2%h6&a)*WfF>545^C
zMiYgJzG=}7dt31rH!e2hK$F<&%7rd~^0S#NR3`c@l@y|(EK@wy=Wl9ZV6+f7>KV4~
zl#!4>bEE+3qT;Usfs5G8njmHY3F#%ETUfU@=eG_N6=V!_3VL7l(WJ!$qj)%j0QlgP
zc_GpbSN|IoXz6`kPAV;~AgUj2BQHd;4n%gE>+clgIxE46kVUNxvW|p>Q8@Om3l8AQ
zpXCSDEW1i!ODn@_2c;_d5Wmd>Zf^O~)$3$pZ4UX!jh~8X>%S(;6v7MA!l9o90MM1z
zNkU{YYw*3;W&kST*S4Rub09`Snp?h>*29p<@4S{#5oY~Y$ou8G({>aj*k78+7>65A
znGG^yION)j!5!mjvMC9HDg41p_8a`4L>j^7RVJS#|DGS&wiO1UQ;X!VH6lZj84K>}
z5H3wFb*m2R_+^4(CNjp;fyhJtWt0v0ymVe?OB^io4}%fO^AZSOyf17f<;6vhQ1s|P
zvH{_2*)PQmvzSQYHHHc|5Na2@Fc<vmxcMHk#m*Cjk(x&h`<X1i!SQf2sqYh|SeP7g
z;T(CYb%|o|`e95BprL%yZGB-h4rqUg`XXE1pG1Vx3*M+^dg+@O{kUO?YL$8NbE~G^
zo!5QO05@3w^c+A4L8Ow}z4|4^D1VH=ypiVhl($G7)qu3h-2+W^e%XB_PU=x<%fYh}
z`|*_t)NJdiHZotNw!afI3gZ!t;M04G>Pt#$-`Nzr-2NBQrogl?YyNL|2b5ad<#PjZ
zyB8gErCCqC!@)Q?&~40%B(zfh4;z%+ML2G!>NrRv7|fi{nhU)Z6%9}AWbZ|7N|nvK
z!+3!sj%Re+sx8+nFI8lC`lY{jPy{af1kEZ#?ueFC-S@S2ckWBMgA3!tv1bnM6g2K<
zcMc%^)7_rSk~-uCPLCs4(*q;k?fE4C=v1U$mB#G4pKlXisKqK(i(+}JEwuWOuqS`e
z?FgE+Wa17SPxu!XF^DH%?%w#oV}pq@9-5~kKq?@9Jpl|w=HAJyh&A7|*9&uP<AsJQ
ztB@s~jlMe-8e9zUwIIPxPh@?xxKnP=uqV~>;`>_MXn$DN1<WxqrwZ$$k&#1k^*Z#q
z*AYw9X6TAf$B3O`N)hK}Cvp&G6{&jZQT25Qp0pN>$!R~A6R-twajEqlveUe?n3eks
zQbj@Pqe~&1bwvP8#iJMeXL_aFRk$}N9(0@uii)yZv;Ki&jg2Nka~&=yGD;Xx?Nn$P
zQrcoFd&7yNFy$6BDYH{D%nq=10Py13J_bR-%Go|Y_v;Ue6(ALaNN39+HU=k_=m$ZH
zn~vi8J4!`;jn|Tj?V3}66d_F*&AFN<hxNh*iz{TGrh~J{-)lCxR*hpz(WEO)iQ>)x
zf6y)JpfP>;+8E^mWd>S&%_>FMl3YJM%w1tcnMyYE>91eX@#3nEYFJH%6}gE!ikvrl
zqpC0Dsr-W0vXda+EHkj_loMnu`op=1^6b8iOw;~B%~<(iUP`O!c2t#}q}6y_p5M40
z>}z)$c=$UHncGBJX}*_JyE?`7`Vb%A^1fofIWlEuFwuY~^3)GQbVJ)2;LYdg8mu>g
zNnP3hxNDYhdh*o>oa_b14L9U3^a~AP{S-O^P!)_}a+FgHOidA%n)L53_x`-A*0F&o
zVtpEoD_mR!o^tugAN@SYroSrw4EZ6v8|;{g41hQVQ4<HE?hn>iivIi_cNmMB(MWhY
z2O@z$t>et1wy=zA`Nc`=H9m_7-Cc#-3r(hc304$b#!b@U<|*Ts@!Co}0iTy77juU`
zTD-htUV7{pPN1VNkc8r792*`x?cEf^K~D^+9oOCoy+xDvFl_yco$`~HrD&g(I-Cdy
zHS?w8Y%p35V7u)ByD1G)!A=sfmY<F#S}-n|{6BPz%t)V1SKraaOVI!1HhcR#@T1f~
z+}$7Om%5_+Cumwjb3R!vaOZhF?_v)eZo`}%Pv*JMA4^3ho!a!&%}!N#^FQIO>zcP|
zr1A56W}7j%S89}qOguh<s)L}hfSqRjgOy5XM`Ymr{T$_JWnT2vYxqLhUIU+ehlS8s
zN*oTqg5z2n&-z0a(yH$YgCp-(MKn4gQ(gaDq8wzY4hQ#5QbN*D(Jt4uolE{XuAa9N
zLn8%V4>NuF$I%3$+=L$I_24(2+v@t<aKO_cOF@=ln_W%h>HzobqLzZ)nQq-gx@$5o
zI=(BLpHT;i@aRVE^u)wc*}J5NEL0v<Urja(?JyRwV|fs<WWdhTlAB{BF?yt10lv~w
z0~_ZvZfTuDDi^t4%cDqyY93*zB0mJywyQII@Cl9EW@xs|0LwAXZvdSHClyd_-J{Jj
z-7&0M{N|F%fffk(!Cd&E1U)M00n%XpF5v0O?#2ZL4!jc)?}Iz^no?5vP>~?u)y%>_
z6~%2L$s@Jh9)1NF2lNW>5S`Z6r<y(PS<Y^B#amIFc!Jm{ePiz#MNmMj^}C=OITGH{
zz+L!(aq4}0d{6fjZ`=rLur)Hm;LUFb{1U~ndCp4Fsqu?m(g@!zO#2-S#u<z^G=!sR
zdzyV9^X1X7VkG}@?JzmW`JF<nHy*Z>FxXg1sri9<CMp&DqJQs2jkcL7#@HKGrQT7r
zR=tIG8^XeU34XQYK`P})Lo3$sba(K5cl6J89Vg%hRiUi~$8n|H@7+a4%kr!72vVk=
zuX^Sa(sk|&j7Da(M7*xW?LK(OOY-)$@l_fP9vcl%iuNcx|7O_$%<5t{L~420my@Jm
zD8|8u(^A_sOimWYpI-gm213-({x#E6a39BLm(~FLXN@YJ{XKdnRx%xyIR$jigtVx3
zeBaea<Na(Vx<<1*7ZW@*pUVefxQSM;Wc#$=G-&c*rle%i*kmwkvFw+K{$k2t2v)GC
zN;}+}B}uA0(;U2&ed_Ud7h1qdBT_ckXm_55g=Z^9;Z1*qObU1Zqivg-j+*uV#Fqai
zx!O-xBwCEoVdRT)a((1lru7cv8ZP7hWP}kq)#l6haPML$apVE+YIH5v@9VI-ZsOz$
zr;UaZFiUokfzmvaZfo*$S@Et6DbiDJ*^D^DnZMIcoQMr~M-RE58<453V&Ggatt(z-
zdXqS}c5Ku1ze5t$f71?KAuK#XEbk~v8$yv%Rax6uQ7>Y{!+7bmySg>S7Aj4%7(pC*
z%cY)&SL<>?VK$i?WHz(hC$#v@X_4C@Bv*~~?}zMY+3VjuAhFNyxUr(s9x;`j;q%2d
z{B2^OAPARre|)Eyf?R_&;F$|_%1}NTM`B07=c~VU`5CC?f3e$vele$8;(f*CPe09I
z=n3zRcJXy*8#K1^8{HIqh1RG-w2H@Yo3xUYwkd?OAaS~r6>WHg>r8||UepNHMwUeo
zDyO~$hv!LzC}K|$4g$LZXrynNAbA=~9jAWif9EiZW%G@C)>=^lc?!~P_W)Jb0O^PN
zl<j72w4@=f=WFmXRBIqP3*QUhz_cCjfqaqyrprvgV&$v+kf)M@lBUHREYtd8=P)CK
zV!^26*J?h7)Z6TNnkY_srAE^+iml?lAfgUdzZ<yY;QSXVts!s!`KYo^wbiv;j!U@&
z|4i&>S#dom^6PuwCu1gO452&++ZE+(ZU8*Ro}Gz+tx@nlqtFq?hVlL5`vg;OcY(*t
z39avR9fT>~97>M}L?K{yGN$TReMCqeXT}MAZSCB{IRt67y};uUc@=bEe?%U04Fue1
zO$-;m^Uq!-OFKwa#m^*Ue(m~olWs1xnj8|$^IjeBPhty@bZz60HL2vYa<v?KK6_1;
z>UHE^28+C7ebvSGGT!#Gq80&U0Z>L!h#W;9CeYD#NWO{6JCUr7%xh*!7vNtHqWmT#
zkm#6C_kx;``b{BRV|}cUKCyU}8PfF}W{IVA;mbb9|6WK(MuVj3UY0aOb17<KMfv10
ztbUli!sDO;DA;$2wl`6b+O{D&e4}{!EM*39A&SL~NT>2W$-Axwa!*K&uV^5NXR~4}
zol0JKH<=I-KQAkCXPw<kFbx(KS0d&lum8o8t;EUJIAly>5lJr{1<k69>V%>x@+owg
zM;Lysqus!$?-yB#eZ&r%;N62=?EyhHp7WKNujg%-t{us`l0U0Tw*@bY8%jCz+p`Nv
z1c4mKad_NXXKdaYt`EB7e|g!@`VBfF+#Tr{S7TRDzTaQd(vLG3cEcB9_L?w!CRwrK
z^}GKDTr_Z?+-mQNjswffp1x6&|1^gyH~$-BeyJY=1=6y~Z10~XU=ePC?)3@?-Gi-Y
zTTCaEyHOEH-L1b(LBLwg_E{BHJ0%@nMuJ;KdCP~MVy!SRa4`zz*tkHYBqB`a)&%ib
zabV#I1Z7?R+y0QqQmvvQAu56)MG0u+?FM+2V6~H56an3XM709**vT5I1kd?v9!b~d
zfQO~E={OujLxd@N26W5;=6puAV6LLJ&?peLeb5pnNA|@s&Hn*$(WhrZkKYP;^Np%s
ziUU1pimY@p7_V&dp*M<hLhJRwgTTY7@RnnWs(hj<Rxm${Jm#ClGjN<`jfSm@{k##6
zUO?8Y7YcjzKqE8nrcPS|uk~COL2PUgUSuE5ux=wYQ+Kj|j+SB<w28~4O$0C)wpe$2
z=(i>3oUelAyB$B^QDU+rsOUJo_TgO0;MV%vyYTaPXz~gTs;dyfcpM^T9D6iv^ou|C
zRU9>Z19xI<#7gjsnJR_KcmRJf9n&*Irnd0uSeMMOc9~YqQof8LIb-!nU1P%jH+md0
zI9d{){V8I=Zyy_p>M-YaF!SCXLH84ny}xJBI7D)udl_;EttpnR!+m;2IGm#6m;cad
ztxCX;waxy|wg}APx={J5#+|jm$HN2ydX*Rq8|ePl7Si)w$GNQCCw;$ee$|#Fi#`h2
zseCSHcwzJsGSx35ywN?~%oMAGn|nTlP|tg=8xAW3hrRVp^U*GuZpd*G@8!dV3lNUn
z9@1;J4!8T$n=)>id{k21)$B#9hkrN}NF(v$9Z7y`J;VlKpR}uxNDEC1=emV@aFrzT
zme^X5zkr?NQqL#bc_ow6Lu3~=U4_DErOMA$>Bl1SUNz)mQbCq2<qo+lRfGza8zx4J
z$l6#59?`h4{TG!~Ow!tE94$FaX9KX%P~^TA9GxmuV>QD6hVL@1W&%P9qBF}o$Je-F
zh!(3ArDjs?$0dyUF=E$Ir2f+3YF@1J<{?}E{)Tim3{Y{U^O&ACc_>(QW-zrpjK3x5
zc$C@to6zgC_NzcWg0TG~{SpaX<HPxYBy=5!HoE57j231J)iMWBjyPi1eW|5?zmN3g
ze*?=)R6+rl@{vAiKY;&@rBinp{=B~Be5FJ~Q6u|qzc<pGJjC7|@IplL6wvZAJk6Vz
zu)){Yr!eQic=^6UWl8b0S?SRMLXht#UuATrusPeDwVr+M-SVkB(@pE2UM_t6(A{76
z7D}xdA7#`9ltMdc260$311L<5-omj+cr9JFG>=dBCsjh95PKEaw!IDVX*Yj+^PUs=
z@B5WG(!QLJ4v}{GS~?ys$gD@LbOb{or8L{<Qt{!hv<YdY$Gs!4Jb{1ieSjbN^*m^Q
z6F90Pqz^}X?nG6)oZ%UT#Wc2)aT4bA24C5&)r>sv^)B5T54p_Wy@~zh659Xmq4oQG
zjOBemr72y+3vq>BpIr>Wf(!!I47vuiMqKUPjJvC;jES#ldDllN51IjAcVu{s_U5M~
zFvg8(S2bIp+WZZsce5Y!3%7kp^UiId{B~;h=kL=-Oyr=)+WKBB1J*}X5F7&Ih(x0`
zLjz4XWMm>>OANU<31x!y3x;EA-pvts3Zf&FC?u21e3j_ovQ&?Dfj#58RByB)v|u(p
z_%*HMeYFTstN+voCOgv1qrHcz9@lCU3<^7p*r_Dw@6I>?&9$0jWCvfd=e#hOF<Yvn
ze~8t6CFGiaE^s{f+#98swiStO3f)twXRh9%fR=m92E=V<?ES*7Q#=42@h&KVDkSIw
ze&W>|^T0#z%SVJj+90neM&8e@ZU7Jr6hCrB7uPA7X*jUk9ReFCV%MZgI+1j1eZQMm
zKj?$JPZ8pmz_u~#>su@}InD$SFB-FIOh7{R>s5~`MIyj+0#1yZ1ASFTQuzQ!78@ik
zdwr;osY;)g^v<F4E!R~eNzC(PAfdPB>rPYhMr?9}9)ntX-&_6`7nd&O5Bc9RGpVRs
zRS%i&IqTU{JwH!bnr;*UM!#M>npO0zK;c!OWHDZblFPNTr4<M|9kq+{#{L2CKuphs
zG&4CR%Wgl0B*HdE;YPA(z}`d--A2d68c8)M)j(mzmz-d}x=<|&gt4Vk`S+?y%>Cw#
zf4ChfZ0hY2&6Ml;MVuZ)_A?g5XQ)3iPJ|EHyHLG}WXCkOctfXCyKPX2&9!|8k_Cs5
zxT2yP^|xpiuHabMvPY{Xg0X1cJdG2-*=rzMm3A;4O&83sh<W&LIeK6vR~q`lPhEA}
zee=s>e|pQ48Ssm8Q7cW9WmLIEsb{yUF4bC6!UEwR<th>D_T00oL!uKo@z@;!*e!z^
z^pt6P-dhJc9&C|}O_M#4-$nzEz?YCc=WGI``m`j^XzgSIzxVssyuVO!N(q=Ls;~|y
zBgxZPh10coaMw=8#Q7|yeH?LDWBp%NIu<o~lCx$pwq(|88p@!($6k*Mu@#109HnD=
zF<j6X3+Zt9FM93Irv@h~lh?pqw!%13Xwf<D!7mpRl5^7lRBT8;;(<zO#Qyj-5;Yym
z{8zr12zysrtL`laclvwe9`VYd(-#QpL9PYLMFb?4&4INXET0iy!zjK_c0ccv2X+LQ
z#RDLIVCD5u{ec6ZSZ8cR@IYpEoL3%zp5{T+oLldMS#6_WVG`|5y>mY-^y9}=9w$iR
z64`rCkZG1{wbE>~-XZY$Awyk~(EXo~>h$k-j}k2Ut~c3T8t)O?#<nTow|d1@0)UeR
z>fTs2ouu~wJ<&F+JWYx-;&Dyhn?`Co^dI-Mi*C$FG$b=JhztKHcuz}UPUGz;M-xOu
zF1thm?tA6+u_}a<S`B%}0@lNXY-WZsy+MUZ#jn@)R?Ya`Z%%XjaKQQq=-u3%l^d4-
z={Fjd50+zQ9{DvQOX%VmwVfChmQq=$R=Z}&QcsY$W%@d4#;KVFk!O&p>2=v&+DuAs
zzwWHDVu9w}L@2U5P#pcWK<o36+`P#<rjheKPnb<4`Y@;Qf4VgMuLW<f1FX~X-Wv8Q
zY@<b!HU_i}K@cRer_~jD++56`=Qa@kSr$a#a%Q7aE;q8_+<NN<t9`9)V32f}8@h0`
zx?&_1E^RzSOSj&1G5n??dCvkJkU26xDD;!4IE2v13>s8KFLA>v5d;J0ZPQSi)LoY6
z?g%RO8${sGeuFZ)-X-`{c+z59wEcirA0`FqC0C&uFH#D<sVlC0)a2>;rcM6yJ8Yf#
zQQUW18zNE)8hU&bl_rr6AR7@u=RJL)_rZ`JH@SCwIhce8>)y!h`<RGk%C|n6C{WT!
zo(n66L*}v30svGK79tLp{ED(ef+j_g8X0z?_k}#NUt1K6e$_ue9hTqqjz^Ng$zDNX
zkh(1#yE^+I1Czl7z4}9Ap+>^|7@1CZ1cichEdRC-<-Ec!NZE2UDW<d&GlsM-9a;4m
zaJ3QFfSE%%HHPF2cA#<P=kJnc_<4(*0WW*9#p~J6zb1F*uS7KBBW-ac5E@|JO=u$3
zjeC8vnEgp{V)Sf&!oT|N!@z@2O9dtdCpK4X?XL=JlK1Iv;G+t?({mH#;U>tTDiI2T
zvgm^Hm~Y`;BYu&2^A-bROGFOtT_)@RGR3R*?Y-ywOYVjWJ|*I8H)B`VmXjS;9@Z)B
z<vV2uhoi}oROF4^aqW_j^9fmj6vq194Duw(ZCs9;Pi6)l7KB3=;bJ)9+L1{-A^BH5
z_9<xF;Zr8^92SwvcNiRmMAoe^-;1iI^Z6}T0l190Y*|U^baS<qOy`t=vT-&$b!_5F
zvpDGc0tOZi5rgcr_y7No(fro(f$J<QcLT|ezF|%S&%0f5@#lF#*0x5cX1i}4Pmx!M
z8jv|nP2-9o>P>x-C9K}i`~i~`HYy!q^~PtXMR|#x)CEpW9!=<oyD{W$9e03E7<~}B
z>Pr(YNww;;b-PRqYhKJnzM}Z?1Hxd=Hifv<PR<*ql+YCEp{Y)qoSGI7eb+8J>c5aW
zO%im$aN1fz%eUm4NZ+3yX|g-el42sbjBW3xYqB@Zy%olpJek=7f5uD73N+a>id26s
z{@7?=f4iMEP8I)EIwzTh&Ewck;k;_K&t+a>3gaSxQ05C=@hnZ_0&zp8HU^#(?2^~?
zr{gc7JVZVbAP0*=3%tp|_gQu@vZ9uIC{3U-I;YmO8IbXmUbbN#2fLW3U<U0M8O(`Q
z+^NGEW!ma_J+DXB)b8gOj1w-O26W-ANA-$g>zND7%G=nQHN%&&o7w>y%cl6x!^T<O
z%x?=}_^ddDI4zycOOl~vZQ}grQTEq!CN;#6Q130mm7DJUjkJLugGHANLgVt}MeJ|w
zK($nf$HcA=RhReX5RxpKe>>C=t7mj?pon*5Pgn2`_g<l-Qbs9|SUQr)V^3-?P6Lvs
z9A2yi@!QPgJ~Tavf4=DvFl+es$VVIMl<%-^vmmc@N2hZu8AA4qbXz?^`g9uTuaup%
zcLC(V$^6%4kz3ENQ@=gVgMmCL(c)_lzHCqAte7ZAo`}&HL-lON=ozThvS)Lu3mF+!
ztS47x{3}5td$nywg+ks{kSkB7g?RnnGF~L$G>)mBpzEoPaA`+60uHjG)z{kN`U9NT
zrW-4{$xUDm$*Ilz4)9*f!M?wlR2a`r46f?!rVIjc6N8@W3u$N8$puFhD@gy_>HN<z
zB`;loU={qSIq|IV%wF}w+RkcuZu%v|=O7uPyJK3JTd7te%6VL5b@cSUf8JcScg9ws
zIZngM@>{cVsk2=5(`@F&{$!rV-Vc=Q?4p{~*k{Lv-0{6J@e%3!qZzgPrzgo~5&;g6
zBlGZ%2FxJdr<=Y^U)>K{45jVJ9$b_xWp`|~nO?V2RDpW;c6vW0oB8D=fOTSCGs5$A
zjbuoPvp_dAkrn*OSleRSW|gN~zSN&_<1rsRxZ>ZeC|uotle)^vy{ddT_qa75gYF2l
zFo_DR3)Y=^jK$N|`PKxdAz|RT`3!s7Mn2xzd!SQ%!q<^k1H6sl0_E}WbY08UvPA5<
z^M1O5$f=0^?0?TZDY26i6>c5#P}I_2bI0@355LS?&O6+uUD59aMb!k~h&f;ztB$~`
z39md-EViC)TGgOTp7IOUMIox=KWslV9JfW&an49MpzE?qaD|)_mEaVx!n3p-=d#p4
z*o>}bdxg;^%8%VCmAEQZ!EOL^wA5z<9_A)!(+NY7$l~STMNrvD;}(NeUcj|^PN?oI
zh#OMksjcVJZ!%l(1i$K8`nMw%GcTqB3p~JeTk01|1TUL@4(^n-XVstU0IbVZ#52>B
z+PTe_q&J_hF7{~zZY5~*bAi_jLJs0_53_BhI#EN&ytNRY=tZzzO5q}L?1e1PQRU@T
ziC2>=NSJO%xme0p*?x9@DzZ4s>HvzE<IDP)reaqtIYsEnHhXFWsoP1DQ}YJ&R=8SX
z5zltfrUyGfGZa!81eJz!PWo=NLwHultbUSn(@2P3WT6w%+_N)QKESw8HicI(l2QIh
zu1r6CCVp;X^3pk+h;p+T5QP7-Wcu8`^@x1xk(RPa#oh^)S95Q=nl6Z#Kaj=#aE@2f
zv7W$8_00Aubyj}ii7lQMFFf6XiEk)<di3P~D7xTS`%w5q%l!TYX|PnTzZe#U8Ok`z
zcyn^OS8;(QnUjs%WNIkfIKrnQaVr?^urRG5+v~MJH0L>a^cjT-(tPXZzLe!j=8_p+
z&(um-P`haB0X$T5w!O~^-TAEABzv0AH%LcDW}><6P<6KlC0!R5w*Ov?l#srQ%<9={
z4`oXIUGhCL(z50AU`r-LDVYcl^UfzRx4t-W@WGC;F;<B)EwOO91a8{Um0L1?$t%7O
z_f~X&UOMrdu`i+6t(lPF*}{S@oTpG98=4Um8+Ia&;`q~mMa+$R;KqVfHbIMTrjP|N
zqNo86fkU*arw!!X>DyZt6gBgly85x0y=-iGx+*DX%WcD53WaTCwp1*#$QJc}cXt{n
z@bG7C7xQV(c_X|)hY(V$r|$K5RiI<k29<fmnOSZCM{cg6<_&FFd4wG0POYGG@20n}
zQF4_QzFGUCh)j<H*Hu<NrJOhUOevOm8w)SEkYbg&<v4JFIm!iynG>!2xC^TTI#0H1
z%%`B+mklIwHGT3OMHef&H#%7;_b}xoKG<x06-Nrwgt*?L0_W#-7OJCI(A;g}rN{+g
z|HR9>0;yksG-T;K?yF4o%%BgpEHp>vY=|pb@vnNn90$!yC7!{e1@@n=T6sfjH^3hm
zm-V`vX!Fe%5Km|-EO$NXHR3oF$vxv}xs^WJmAX7B^H?|kM?oiAzsNmI_-MFE{VJvj
z8hD{L*}0Ar@vT~*6H(y#c4Pu9A8;aga0EK}@q1D-R*fya^xZ7V`>3bvihVAMP}yh6
zV&ThpT!xhva9f<T$*!jaur*Ua#8AaKWqb9>1ME=L|9$zF`k~NQeAQt@t|m)YJNIt)
z7ZP}~@tU`&P`k}jQ5a`mu%~XUv)=agtldfNo>>DI*JhkRE*<f$%|<vOg0qTNmD|g_
zf)f5e^^q1W9=tAAFn#eYyfu_C%K!$FwQ`+Ec9+igY~_k0kdM6>N&AiB?O-xnN68bK
z_@5f@gZbukk4vbuXpkK?)a|fL{BB>R<u|x8iuVl+DLbLnHBAgF<SLrqtxq5R6~woC
zpNl+v%3I~(<lM@Rz$@=xprql?0c%`Ur0refxQJ--8jWWtfthj%++|gnSs)9_ytlL*
zhV*VUa6iB)vwdNv`MT;7koOfkIyjb{4eq)*nxoLTHh-G6rfaX}6tZkv!zPzFchTjO
z$IV~Wf~pdLH<-f1kpk$%Rhiii(`q^DWsqJJO5SvOxl4c-Ik@)$0}sPB-arMPnnZ8;
zwe}~jQj7aYG5Ai`M}qK~3|DRU)YpDa2yysn;w4^>2MC^8hi-n3jJ2yjx&p#Cna=7|
z6}OnG-Ev5dv~boBCS66JDGzVwMm(zthC}ze<27w1fGd5+Z)9&V(=!yE8=;(fR*DP!
zp^cuj=ZBqk2h*(^Ob(5=RVQB0_rkBlb(`MXtFlLuv-ZC1ALdx6E?xukvnhVlCyUDK
z9GOF@4x30Cj2ON>y}}VeO3ugWn&zrcjf838@`)V$JQSBtK*6C@5Jn1~#h%uQtI9l3
znoD&w=qX(%@+8|8ugxZ9v-y-8%&4B^6czcG{XaMLhLrA-1I?y6qJrHY-|oi<@o}q~
zrNalmF0R|}>fyfC;X9o6VLZpCMBDBgggvs}PLEfF;>Pt{t%Mr8Y$b^a{LnhluI1v!
zasgZh^Tw0_$dafqW0So$8^-ciXD9vTMM`2Kok>;!%zyo>&oY*&JNvlX!Qz#9<+voB
z61ctB!;!`1hd$-EOc#_A;`UWt!}r5ZG`Y&VE;-CdpD8W8LU5)Wd_#WL#?Ve+q*P5&
z%3V8t+-fhc<(mGuD=r{e)fSd}r-t*MBmuJOcA2w#w)L`v)MYbjs=poc+_IWry|wbJ
zi9TI@VW(|k^1hCK(g9fSGqBGAQG^_xA%4OBh|yXNm{L8;JZtU&1D;>i<F)INyS@OO
zn{0Qjk<WY-^6B(-39i>O^r7$<r~pI$1&-%^H1YiT{Ivb?{-r1;)7`DJ$3{~h%_EgM
zv<Ck@*EJdL!o4^!bMBv=_+iQ9F56p!?&>dP1ZT#lO?;GK16_ISPCW-*IBHrRPm#}T
z5anIP&{4<5T=~iVJGoEC|E_QLdDBcFX>r}l5{9i3O%oVkY_OPjTJZ?{s>7asIE%w>
z!x~>*`+2VV@N!39?P{ZZi)ne_+LLa9ru&ZQGK_Wjy$AP~ML#?alCBwz<@2~tL^=N|
zxW9D=-wthZq7O0F?(}1&0iCJX|8vnXVh4rPo22nkiOQABhaxDYx}P^{l9?KFZ^gLO
zk%^l4Haab<3E%<ePLJ=gO5L)rjlii;nek{tA;^n-JP0J^{{X-PSm-Q*49`*8mb1p;
ze8q^eZCfH&-HWAU_o2qK&f9XW<&o@78adY+Cj$n9|0*u}K4*5@ii^*wqy&GuEo+w8
zJ$86jBsiAr6+(W>%7=a)H1*F3iCdxTU|+DPXjcmUEMD>o>3Y#_g#3U=Z{hZ0M<ozA
z`dkT5djK{u@ci%MZ&-3%z#g{K)rHG2(@-3=WWNC{>6~d9?MJ>J9)P~>$PKhqQXMJ&
zSktfRQKH`}X%;3{p3dZIs=RG2K|K|NAD(B|YxgtrZgIRZeIW6v&bS7eE#-nk>?ZbP
zJ-{H=STn?gW3na0t2?Z6eAy_ET9tdX&zx?m6&g^~e7C=Mk@oC9CbjqP-oi{EEdqWc
zmF=RjM^gTjGx(g-K6Qw{#Qc~3UAbMEJE7`_JHSxbZcIvxnVw&E<08FU)L~BMmEy)l
z-Qlbz&Smk@MC*wthXyZLt$nw~zC%>Oy%K0|z%;g~wglnD-}_@e_-8tq2{0z#auTav
zyC^CSwWKJ~S6Dgd5Xg}B4+{?5eM&ZZZSuR^uJ~i{b5i4;v=zayMZp9wP|#=jJe5PQ
zN^y<bBM|KmJm;fgS+V}l!Sl2C=XjLEx}Cr7tdB7dpGvm5e={xpg*I8q%o|3l?IqiB
z<8E3i`arKrNL?+{>qsJpRLx74I9VzNS;B>z4JPhIrNR$3SDvmXHHp5rm7chpcn+)`
znHMMOEa`wMzun$)>l4lOo|mfWBBi|{wAwH2KV$Fb1K98pG4-^Z4`y)dW2Kno!N|%-
ziYcL|LcG;=h2f4}t>vC!J8NfOE;4W;>AZIT1cqo6OqT*DZ*Ebt%);@5Qe*DHV6ZWb
zBv$~OcF;+bbMw-Eo~b5woc-fCs^4-_yAcukNYb(h7?|GyCO;(^;M*bhlJXds#k>??
zf6yj$Cr%)n&MI`>ZRHGa9!ch^Aoxv3U{wn<j97OF?ml5-{3)T#yTi|P$fsgi8%io#
z4!SY&k5A?CP4rEdSxgL?+y5Gh_O%Xt=cB3_9GUBUm8N{_o7UFyo^bKi;*{y3KfR9A
z1y~a#z`6QE`?cs%vfZ%6PKuNEm=rN-HE3S2hcCJbt>}rC_ggg}by-sqID`Tbl=AqW
zzCMBBk?~rcuBB5N)uvGs)C&}vgYA(=#0p6LUbg=uN%%$&_|E4KoMqIhH0-&7fKRHs
zCjAl@unybH`iDD0+FMX)Xm7}?zbh6pnWen``HvFuq460sc7I(ie&73Q$ftF?&Q5Ds
zY0dh}97~DH65*+uW{za~oXy{AJEG0^J(ANZ_b6bjcemWs$%hg;_6v;ze-7sms2WdE
z;kwPBUG0wM(Pd9%IDG&01YlD#0mm4-aFfways{&(PdUF_Sb1XadT#slf1kG(KBB;`
zm+ym?%hsgkrTJK?n81>Q3KfteCReTnRE%@9aB(~n=QM>h0yF;>*b`@zQ!upXFpkRW
zlx?y|KL&TNPT{uHGNss$|5zuF<X4xN(b4hJK(#-6y1$>$b{mr6Ip*Xsq4>QZ5T{0+
z70USgHyZaBXQ$RNr~iDEkD4$r%WM5?3CFpa5&6o)9(_{U-#p;1!33Xp!s}{MzZO79
z@Kx=3E5&jC0|S<cFgycex%m;2NJ2LKdnIL`!YL}TkB;Cu7Qb>*R@L)2Q`*WkZ2-SO
z4rotursk^nJP(sZW35W+_w)A~L))1<kWRY?DRS|BRx?JwRKP4FlEoe!3=L<Js9#!v
zhSNW@63deo#R=6fit;Z8Ou@Y4s35`=8o|`czup#?u|(k{5Rp3-bUpq9rk9*|<<`rT
zk-XnHTZ|EKjxL5|77jK<MR#Yfjbeo<R|&~$qaa4uq#KWukXf4i0<qXB<pm&_PXs6l
z$6qjlQZh6gmQS7TgN>0+lS|H&3o40#(52B@Te-p7>R!+JP87eVNbdn~&|uQ+66=Y~
z2=0ee2PM(H${!bxh0a$`tD9$)%V`|d`g$=XZU2KLY$&B^)ZMQ|lpboyDI*f)CESuM
zt0i>q#^bq}@8y%k=bGPgwAt?`5X&~azf0Eno)!B|&AT@Gh0#55o#Tu9XS*IjzyCK{
zCVyG(Eq3|_C-#n)jqfT90@2Tq{2APo_}BJ-v34{usTs{h7cwj~JH%zvEd;_14Cl^j
zr<S#x*Ze;~CWF15dBfM>Z*TqNZ#k~(a&Joh)13V8C|jJpk7Z3tg6COi*7)5)jWYIi
zsv9i`9NWoOQcg@v%jcvSF=(4m)i`tCjC#7v5r~z=6r5FQaeWY#%a$&%N@NlbO9I)!
z(dW?<)XMj6ZSYq$XFAX75~Pp2Kdhl$xHF_qA6r9plF!+m^KMUuP~uo`*aWct3*sw6
z32UjynVu;-#T;f%a?D?2-|q%AqQn!^de2_bZa^NX9Q*5_6Di&uuhbb<dTLsg(WBo}
zE<SGmX_~=Z2U<Ua&A1OLT7hyDgL|334`8QW>KlkxjplX|U{6!PeVS*0brGG?%alCD
zlI4{z8fUMS)qA^^=4zSZF#VyBzzu+5fB!Wby?-%vBStk?85kGQ;quISbX>M#%@L{|
z+~hgUb*V+y;(TTZJR)^<#wcsSVJynMJ3ha?o@;_GH;fgEUtVQ+P~zYz2?B!Qa&<iM
zpPr0XqeByL>`j(#ap6Y{et<HRer$iG9Tt*}xjNYoKMk&KdoX+9L}2uUp?mBK`<FP_
z>Jl0Jrw^(K`+Vcna=%30p@DBb|2?OQ<*wakV@w3IJk-PVG;?L)aWN9UuzKDMB_41Y
z#~v0*8yAhe7<oK9-ROs-<}F_?i(dSbYlXbxAX2uk5KW$VJTwuIYPp*#;dadwb!U8A
zb(4HQnZvJi`Aw>0Js>6tQvZGJ)7jiD5JIR}QHRRyn^rrH5N<rYUxtbv0r-)gU~#+@
zF*Mb-|AYTTC+f?yCz8GEljo-s{~e^VMb6^5CCy2Tmh~G|!o>&=p6i2Iju6OT0>6jr
z-r)k@PRKK>h-y<Io7?czakEnOgJYnRdCR<33GPnI;PXPuqsE%WdhJxO_YrdsQN`@T
zMnFLBG!aKp=Q-Q|TOwoFXLW_@x8kzKHCE<bz0XU$qyav<vqi%=ECXO%*2fvIp}v{c
zq=HG;0fws{aG4<8T7^?SLnmcUUKW&Sw~TBY!8p+*f8=7Q(UinoWpJIJQ;<~}2XNca
zoL}M|olasrE6GqO05N`eScNjEuRwq6gfRbSlxY1$RVo7KOo^yGxh{15r3(1ul~_*$
zZLW=;LrHrS$}2IWj$C%-$^IXxq8rsb7W7{ij_fKXybcXW>4?~Vydr&AML^LpcU%SC
zo-&?p_+F;%)?W#Oi=g0)bVEgC62y2k?UsL#rh1j-X=z59699$CeFtD0ixh1_F%CgK
zao`}&t4fWwf0oBzSMHag$UDR@lTf4I4v|$KEi7MGlQTW8<F@3!$^xK;ykL3Z`t?uz
z@KNvFewDXx!k!PCcKA-z=yPYyaGzqV#`pj=2$|*|9yS3cxAyj#SFgmV`Z92oRMT9x
z6Eoro(0IUTM-XA)7xQ!>@XRWD+MN3c?#Zt^scsN(RTC@Yj6VE31fLAL!}J&qK;;~X
z#Nqy~Iw6LAm61PcvluC7#d8GOd&XpUFyJr{Ei=5_2uR3Ccep%R^+ku!{lq&tD=&3E
zOzNl#ma~-qeE0)}@rw<b%ECHzLQTu#QI&7f9Umkhxc^!72#83EU(tnXn>)^$2;9%L
z%WsI4t|`7}Oje72$z!vu>geeR%>80hHQ45wF6%}-7?;5K;GY>!Y>q4rG0??}J(@u0
zyOg_2d%jI$-%cH`X}&5Nx84GH&RV-m{1Pt)WsTb~_Pt4>%V5i#U9rreMp>gw4dh6L
zX1e0A36I0@vB&E{Tq86nNqarJJ43lT4JX_j?|-kO|N1vw2@L1ui@R}XivXKWbmFh)
zV!lQ;&)q+g+}jB*To}A`MNQX(?hNiSV0+?HcOzrdHwQX2Pw#(iWZIu$Y~@_i)$rB>
zXEVAIUUCF`nw%bzg3Y82+^fg6!21bteij<sh{VJQZZnAMe%|!-erX5w)(;v5JMt1L
z3gbzFi2s^-KNG@Ab<&uty<jAf;Q4k`RpMc(_u?Ii=>1uwm-}oe-%hd}a*mc!tkO6#
zU7_5~{Fm26C=*etkr^&x>C_}Nnv?6yLng^KVq;RMfuctJ&dILw8lcsh04J_FBVNo%
z)vwUZ*F-*?0hXc468J{v5-Z-gfpq}w^CS2RL@b_k;4&A9fk*Iqx=2G$bgFnsLG)bg
z)KqZ1JHiLmT$&+lat3!fDn1phLmYfnNH-~qFD3~F_AMIU<!F1Bh-iPgx9S?4?<=v7
zG}9+}zn<~gv4TRy>2RxZ%^r`eyc%}1>N}dBA5UtCVGA0(NnDwi1NAhO^*%zr1VJai
zRHM7EzVs(QY1!83*=Sna-<UB-3)?T~XsR=*bEmsF(0RTiFCfx`)&PiDI&jR)1Vb=8
z!)!(#(>_|7J&(BQz>|q*C-`*5x<gfufMYTSJ~N5SGe$6W>Hll*yW^V7nzmg<MOjot
ziWCu$b)|}e^kM;|i}Vh%(gF!pLKj7<6brpXL8;Pv2_O)nAVh)Cdw@tM^bjE5iSF|~
ztG>JMyZ_5y{0O=4`^=fS&&*uc%s9}A9Xd7KeG(B02H4p5&$Ap8qNzdIT%X6a2QfpK
zPNAp8Z7$eb>N5)Rv((c!Nji@(53VZ+d1+aCi~!j&rkbCMG`YM?@ol9?LqC~q6vs`W
zk;rQ6uqPzZa`gn@zHOl_<&JPc8F_PO**I4~zp84d*AuQ+uFWdsrMTwDmusD^0@``0
z;K`1^*%PIOLZG0Ia>1*lq|u$}&emmT!F3!587^LuCwrB1q)kBp`X*$`{`;8sm;Kid
zf@IlkzqHxTs^yO6oz2>p1!7W@0f%|3-o3Bgar-DigcV(09i_FIx+d+qGTMTlI{W#-
z*kZd_@9|}5{qEMU%Zkj^W=X<UTGZtu@Zq%q&S14xl*V=^Rv)9<Ybqwk4!6)TNsZAE
zw>0zG=<o4j@lUK@ZoJ?*Z~<4`=mLR29KQOT+xs1p8zcN|f@_+YYR3*fdZzgMG%$U~
zamd24$Kk_-cf)BL*aAP%QQFPjw7A)0;o-w~uf~7M2X%=o*qI68OdJwR!&65ZUdq=)
zIPmDuGh|cV1E;X;;RD8_+dW#e7QGweo)~X4-lNm96uxqLMw-Q9^plnQX+(wyCS9Gj
zGbm#bzh6O+-G0<OtQ7&O$lb>w>xWu!>D;BmS@t$m*9YyiLNA?pVmipxe4DiO6mGI*
z8x&+u64~0Me!`$T9^ovznt%a4u%%s=Khye0-k&_<J^v&hxVhliE~A#dVn4pR1ArB$
zQca%;dVvL+BH9r`q{pd_wuU<aaE#9+H^&7Z;l`Q;J$|8N!ixZ{2oi|j*hT*M6s4su
z^*~lj0-10HG<8@5=|JIOw=VVe$7YfmhU>doFQk$I)I~_2KNOs`B?ZFm&N{a?8iJW%
zZV)TORqi(omRM_JsuJ#ZhI#_K6trZ<uYdj$*E`d}i9^q87OQjbIf<^x{9!?JAnhcy
zU;V-XggQ8MlxSSlW`LD=g{gEa2=MmR14|R~vbzx+lOjfOx7CB6<E%5Y_Ds9e(nt`q
z)x^d^?@wC?nkJGtA8ab1!z!<GZj==7n`@AkB+5Wr_@we%oOw5vihb+fa1@a#F(mPL
zG0Oi*9@DF+6ZZY2%`U{An-?f<YGX8(+yNzBY5+>n4r4{lmRVz}S~v0_1;~2_yScrY
zdS-lcy}-3exnmrOXB3lUbMMxAoeS`{8eKVru|zrO=_iAteMBj_827=Mh(LL}7?0ez
zDYJK6cTM-6>$|k<Ir1t0+}{&nzt&8I@y%)ujbX)weoxwj`@=*G+=c%1O8>N%06GTv
zI1Q=uLZ+H}#c;XKfFEgt`z%qDFoTy{(j=4WS69}bN!U%r*u2)&^AsG0TM<WBxhJdk
zwikrFZ@g5ouG?AR5!!KGM(l~~?XNv9XQG@09f*_No$>TU#HVyfyZQ<GY5U8Bn}M?`
zl(_w_ljwzgl@>?K{T2R}JbO_V^TNo45Tj?5?G#oie=15PQr<BTlEyLwGJ4T~XAs*+
zFlJ|F(B=MJt2=Gw4;^a$97Q9ky*Gx==ev(|KhJO5w@+|X`NmZpY=g0ut;m_n98980
zB{2!ITVoHgXPW_?Nsm@Z$#yIelAo=o!_vN2R`Z71_4%XMxI1n-SG}kB%ng~RPuGGT
zoT>hMM7uVDiVsnlTkT_UjC*WR#Zb|GRZbJf_6mVVa3$H8%9mTZQ!QM@@b1f3Z@ZKX
zO0bL@3IgTr0f*1@xls!>a#CWSoEUtfr%<eJxWKGxe_@)1K`k0$EX8mRg2yf{(OnCT
z)-^1{W_7c@T?8ElYsS;KN?93pTN`Vf9YS$Rmj(8pyGzn!;{&~+mS#5uXarEC%Q?*o
z(Vw3ob|&v|HWs@(W|*sXiaVDyu|VEvRkjigNX0pXNA$m1S@=XkYZl5MX`6upM6A^k
zpmdvUZ8tOk<q>c!!|Khfv<Y!Ai_9<4<wPW%3>%<iKW8IowT_f%h5KI4LAF#YIF)9_
z=yvh5c<=`+jLm{noH|_{zkOz^m;Jiy>jc+=!ofS?0xGZvMzZ8b3y0;lu7+7%mvVea
zdnL=SMJ!QiNd5#m;hqs2q|pYnjTJ~%AF^e%BvKehMOHeW-VvS$Q{U*N2}i3lQP9Xw
zliZST=4*wot<v3&x@F@N)LeBt(7p|tMeIY%#sYXWuCcB~UFm?HLcq{l`d#qa&pLD$
zNB-q*j2U^ZelNH5(!!Xh`!>E$5qL-4X`{}5C*Ce8;?FsksLhKFI#!s5wN5Gsq%wDA
z<=FL0^iqm9kcLHXPTpHsxkG9uM@&fP&AWOoi$wImEQX4G`L>x5ON8eS#XHE09`$Xy
zzvZiXfER7UEADM1==QxcC6tR7q>Fwmi?Y#~U!}8P*A#%-@(+4Uu<=7*0n1rn-7^=1
zwuY?MZLvjK8EApoMeDO?dc!gdPX;}59f~$t!_Je6KU{#G<ptIW*+3dzS%T-t_t!#T
zw){!K<KLZO@>LCO=CYjTN#?w}ljpU&Am&mpeVQ$6Px?3J=ttpDLW`<#W|BgnahU(H
zu*iajxs|p}#e*Z>i=iLahwRrU%X$)H3cA)aAGCc`cigq>+U`_fQdb7fVE0-98H_H@
zKFt#P6^{Cy!B%MemgL}~7w4G@R))O2mALv}XADe0Ew(fR(}}^nMkboD=z$BF$+wXo
z&1%N7B;B`XGW8G8fIhSSVXet@55knz1<Sf`Ul4Rrvc<m)KBGeMt-@OG+!f$G5qnUs
zHN(yWZgNuqLI!hX;wB(Y;k@#9V@@89>Q$FhQs~I?GGQzai&oY-9#(i$CorM_G)gQr
zQeFpEl@%15&gFpQ#n3%|eAdcv3?d?vW~3+m6gl#!D+}7&YE{AA-Jj)RVd(zeYwk?U
zZ5&d!27waMB3|Y#QH&P6rbc<O)^WSV*l&plx+#@a_Q}%mKQ2f+p&XbK-gy1))oL%_
zJl}=;2aesOlh4t&<SoB0b`WmG&UMqEU;s+BF4q6YvspNutue?$Yl|y}@bYsfGg)p_
zTA25wwOZ;hKE}D12<+FR6BIlOQTTlhM3)${*6!$RI}0Sg%DV14d}@{;5}60wdCbvV
z(H1z}uN2|d8i?Jgu??rW;og!3mXkob&ar(<RGJa%qLia>@d~dolP;tdj0e(|J|lb0
zM2lUFS6LA*{NRFfmj=~&?q*ZtBbQo2t(0>GWW$B3CZF1WUIKZZ)0aGT0w23Ru2epE
z#*hO+BrxDFXekRt3KJN{@m_Cnm2_txyurOjwX*@n>YeqL*J-k70_U{{mh>^vxSb1;
z9G@n(=nxgoJ0VGVv%ZQ7w}hmnJ~dW;zHGf*i`x<C=pS$_GH)Sey;*_#m^FLqFdvh}
zhqwk;63@wYHC}9NQ$RNc?4fKhbV_W(%X3+N1s)^JST+csq#urT^!Hi`fygF;1RdT)
zLGwrmPTqhW^d`qTxw9orRC3^r38}{lKsH4QyJ;uQ{_+iq*kLei!tl$QxLl|^|8~9V
znnWa>9Pzzo&JyS{jDaH*3I#WU#p_UHj@F!n7U?_H;%1^b_LLJ=AI*1kL4jGcd;#=^
zO`h02L5>S$txQ7+Uo6+6vQ8J$I*rWYg4EzmoKu<b#}!s1Z^0MTBR^rpVazjnFho=X
z)Q1*oPHfc=+CjK2&kb+6CWoUvhe0k`zYhtfmC2%L=?I<-Bx4^GL8fu42+{IPnd=>!
z*!|ctkSH#**VX$}=^)wk12U&t-Ed4w?Sq2CZ_{B_tE=nzo<x|>>S|f^6Wyx0e3iPo
zE1A{N@1JS!d0UzrLGrc|qkBwBg#6XB_GBH6%=c>k@Gy0SxY-pO^DWD6n()}1-gbSG
zNR*J4ks}Z1?T)$Ct;H1-Xtegy|AwF7A1+H#ym$P{pIOhK{Swrl))znz#?Z?VoaR+4
z=3Ty+mm!>8OzoG0rf34yF%=1Pt_Eo{cjMNx9pHS2+SkfroLz2a%&fOEbR4xowqF@M
z)xmw(h6WA0XHgeoE<VauYIlzjOK@4wS2^Z<sw3ASNkmHe@F?#>zKtqhSMxjKe(z!T
zcC39;d|R59i$YuVY8y74Pr{+p_5q_`D-(&YYT(W=@wh`cn`Hi^ycy47f0jr@lU0mK
zcK*|cTGw0JV@~o(sEu9|3Rh}pXbbw_{i1*%qtDNm>hw6cg~TgYR|_3IT%g1!hgXS>
zF>!V?fb_kWMad@cJ=As49p{`&c&MztE)=XAq{hl9k=5C~mM(ldU(uisZDQ$d5RQr2
zI*MnSV_(WH?#cF>+^5A8+@M1G*jg9gbS-8lWVO)~5OMC_%`NU*4i~Dce0t#SqgC72
zLJBwvGTtN$73_J>==YvshNhQC?(K_5RXV&9IucxhRn;&a{T0V-p$Nr2kOJ2WB{Dib
z46u|K8??P5fb;{*1YH++$>OEMvBR^QJCngGY)$B#sf#u|9o~rx9>>VN3-)Z8y<qb*
z@V)acR<%CYwRsNP%nAC7m_5cB)%{|3qenQ{j89@{aLv9M?^+U6l!7DQj%bOpnxxQw
z-s|eqpxb}x%OJ(UhIo^ev^ZAU6~W%7(7E0$D;{!b6CExvNvU*lCT65HdQK+E9Hnfh
zn)xLCQm&qnL%k2T{djHI^k{3LxCh2^Eu-ccteSsbDPLrGP1;`RgBbBt$7R{@;K4c&
zpfYu=rgQAts+N4vy+?zmZ&~j^CK}$%A2F3&VylK)Grq=6vRbI(*_T)&zN0)apAL}Q
zvGkiF@7DJdVUZGqgjDvff+O^)?Y)+Ww%+SDdLRVvj}h{7)|eV{#p^+<?wak@O!e)s
zlWJF|dqh3SB%7!YL3Ymh{$y|esW86jskyQeK4VAGl1rX0_;QeM!sy+R;qZ2FD8Cwh
zO7Z*`Pg92HV?X}4$y|lu0T*g<nI%sQB*&W&-=8*{q^@pqu@evY9i_m>W30Wjb5+ZZ
zH@N4v4uUOL&p4&nOm2d%2v?-Q^D)b*XU0BDtITDLusA%~X)8Aj%Eygnq0)&}W~dP#
zf_Gn(WE3w<V!m6)KaI;$XQWMOsgsbYkz$kwW!61|EBZ|71{Cd_u3~d+gZA{u?ES2H
zgMz6Sy2v>6d|8$cb4+&nQ?v9i-#oWVIhifEykPAJiCB7cADrK#W`Re@Q@KuRy|c7K
za;{V=mj}AdwrFl3Ty}2?XCT&zJ(j-hsiPXzidiTX*R1@U7Jhr4W}!4|jBP1KqatgF
zHD);d8XoQ-k^J1HvMODZHe?f`GaZ4h@+jVItKU&AV{z)2KAfWxi<C_s*FuTpG_@2<
z^@Fygqt6d?V0lR?$uPdvuuZ+>aIry_6tR|jQAghs6$`Cx`kiXw=DvWU^F}YJUCdR4
z&Ht5B2+l~SG6M5Xquj15LzgeA;`V77Sp<ZyKX&~>&M2DvdbTX9nib!x)i1}Zj3#B5
zD=>^_rH!bySvJi+V@y0PmV@fiGlf}K>!Cwuhw^-mi45S4P4{dgI0l7M4FE^1#?9>s
z;mqUKA+47te})oYu*I*sW@~P-jy*p)XQ#cQ-5T4f*+zkZZ14EF7?M`mJJ_~K+KWm&
zrrwVJev-qrCNu6(fWu1=kGKKmSDkgn-B?dS^CbWaSc~%^$-AyD_4w&b7EGIFPcj0r
zjYUki^VN??CGRzh@Z}rScg`coL&UpRX6GKCj;YO?#@Ao&P>1=g&im}TSl0!~TqX-Z
z?`=0SO;6Rgt>ZISL=#nUy||}cQlk>W;83|eGx;oYa?W>MfCcl3`Kt#gpSxdG?Y`KV
zY_?z0Opat<NQY0_XBq`M;<PFV%$wBE^1%thdpbS4sWhjWHvUA}##FkNkabA+#1oYB
zRe$MnYF<BpvuDe8R1Sx~qqC3zy(G2s^>mp}utN~kDKTa6W^d_@@)*XbXBY+<)$^yJ
z7OoaSgh~SF3!A@vb0okr&kv*P-X#Shc(-0tkJrcjmLJlb9|e&V<DT?Z@MVm&>vBQU
zHqpJn+q6}8{b*){m5%hax+G&b3OC?-ccHxcG)YN%b_cXLu%9d1q~n(b^INn7Zw9ma
zgkL1W{7HkQSUuyiG7Z()W;!o9LNmRpJ{hnd^GLsMb{;W?b6&eqB*YFH3aKBiI)ssJ
zbfKj6Z|DCi2461p-)lZE`@{v?+PUe)ci9X;7`ce6bk?)Yo;Uc~7S}4Ps_nMhVSdjh
zgf|Eyfa04ZII^0$sZ(K?pCvKW-<^^_($2VUmMhhDNT>zhL<x{~O~Npl_l)vYjA)e=
zKi3R<mUTr1*a0{BI9{&2&VQ`8dNfJmc+`FK{Ky0iFk+`$g}Kc3t+O9Jo}N(WXj9Lh
z=pog__1Ozh6!adV7J|ZXwYV>t1q%l@A!A2j$9y&;uM)+}t8JQqIO4->i3M{vq2uQk
zQeCIc-+TpBB}skQ1WcoT^Bhz$m~7nKJAsd>N_5i>x3xyV>=msp-cGJ*p}m!v?4r`D
zJ3lykHGd>IQg-|E>=i)THv805tEToJD*4x-8d$<b9gD`mI-dxyJCeI=*~IGhPVB`c
zkXyYl`iI<3eeZEo74PfT9csj-_!u{=F3tWr#CG?P&w`DHS3ATy6RVdLlcAB4@URZ^
zT3yWa#0ubT*ko$@xo7}-7(T-llM+YHkd%{ImiKQx9t5f8j<}Y~O*iIyg$V;V#(QgB
zhz`xs-No9SiY^J_!29{NF#Qi7dZiGme0im3-;myGcC@2Rr}55wRR+3Ux!q39J_%RY
zy*|H^9(~<um#MA}$-|1TesaoC{EDDa1p!dP_b7#7ihZFJ=*=arS1_fN((7F{)ti>1
zPbId>?cbFtw>$HkBY2=wnyZlB5>0(v5iWMEcKD<0dbZ|kz#XV&QFJ_YY^HR^wtx&m
zHGNvoR4`76C`Vn8ZERRx4x<~2xOJ225ZAjDjw8lWD*8LA90A5GuZF<?1~b_AgP}My
zy#jKHwcMQ30NW@xdMS|8sQ50gpON7T_YOlqXngsG%xP>;_{I!%zT$fMJ)^*pwj9Z&
z5{#_6J`F7M_<{y&KvkPieJu-rn3Z<A)$G`mg6Lc70ck*Fm;t#kM_#AP@a;4Nx6?XE
zFiVrqukN|^Sq4v$E9)z5OHn~6ks@<+JA0hzJ;UBQA*MMuCUP-I5oC%XXVf3be27Zd
z)G~E@WlZH?2Pne2n-8fq6b?2k3N$Jbs@M>D#3u_8im;h=IILOtx12}4H3=_mGn{uq
zF^Ql=yN;zY9P`4=&bc;X(^O}KLZVzT-D@A6=7^ER9oYv0f<l85F>6}1VB~Ryt7!4e
zaM$-w?2A5ZnE>9b)T-&yl0y}j+q_t>iL3hh){y<w&J&}|c$m)ys+|$!;Wa3{_PJDT
zFu;m_hJ(DN)U<dbhAciU2NE<ht_YGz*lN7Bu@4mGDX+5<am$k`wWbK=>@dc*Qi9lZ
zP@Q9*qklVG`E~+CgExcOTyE~+Ho7hItR3n-kaMP3+#GOyu$}Ex&Ie4SMbAMk!reoX
zWDivo?}M?OPt1#D)vfg3C_%j5;<qo2Tu4RJE_x#oIB6WjoMg1F*a0{1p$S&PPMk;h
z6pzeTZ<lR-2;0<74pzvLy@qWcnv^KKO?NAb^|}>@1jwvv2O|Fm!0rh{ytL=l_{}!+
z)TadjLMgd}iqWnr$DU|>5L=0wFCI+r)>&@%zt(}e$QbB@5$De7?ob6a07IZ%w`(Ve
z>ZJf;^2TS^!n#i5)YU$GXA3LlgQi<jO!)8Z9sc;_6m*I<?FW!=9r9^+ftFRnkh1Cs
zv+SS;fS9)DgRGY$v%iruubIL4qn9wdpE0MO6K8^0r}g=!K^mymC(qu@y$*n1LnfAR
z4X-6{8$86e#MSblp#(pBj2rN{^lYYcs!LW1`>V3h$vc}xp2Q_Qh)bTS?2+Kek<!h9
z>bZ^AxX7gBGrn#!Y+k8PGt`aKxUMgZCnSD3ZXZX>oBO@;IlR<$`lVT9gF`xtgU*0A
zn4!LUWP~1z)+zunE-^pvpgH$x>eB`()ay4a#(L}b+-0wNm9prB6kc<9N{vuxxLBv8
zTbT8ZF#rvPj>9R%3+jMhDW}h8c?_;7pYsW|1~scltCIm-DXhIC0gf}+G{9FbpCNUC
zG+0p|#cjsk{`k_?Gf0#%UdH>L=x0XG80ytueb({h_X@U>ylV=gi_{EJI;fTwq{&4^
zs;VgQi4e|o3d4*j7hAx7sdn)2hjxr@=xpXE^6AUgI-|!4sUbMMz)us9fl@o&O;Dr`
zZ`MCw?kU|Vk#o6)Ai8}D-AhSavRH@J-C!W3IgNyPbnI7DKfCd^?>N`3Q|L!qX9Ciw
zOBKfCc;#F4`8d<>Ny_<dug%gPCiHXc5=o<Evtc1z9|Jlo0JFFCA>?SU7STu9mhNLX
zrZek4i{b%9Hr;b}alhtGkPN+uqTQ!z6u#OhULSS~8aOy%!o9JOb&f^wAx46WjU8P@
z_z1=gkI&lNOAc3hC%3T~u-XFmR}}XU7ua|S&>UWcnBD;23myH9cP;$GF!vqH`>k&r
z7;BpGF6nBvF&dI*(mWpG%UQ>@j|Ck)u*=!uJ%Q|=VMG^f)qiNH(qrqoLw=0e4RFn;
z9#Rq)E1$D12lyonOEyPsMD6lHU$OR8Opwo6>AsilP-^7gKDC5n02~?1`UzLNO45$)
zKzZ>SkVnk%8VXAiJ+#pUFHkb@SoUL)Ll@=t?iKWf#W?>EGY|j1yxg=~wXZR)z!<im
zsl~3(>c5{5J&W8-jUU-vR3=Kgd25nc{UHVDFRs%I(!toieXva!i_AIs7jXx<npv<6
zhbyRsZZs5{Jt}{TBPDznjr@7S{Kt9Z7P#o|$77Cu{D7W@7c=6Fet_Wc%6gi{o8My!
zZk(Iqks8lO)==}F7ZzqR|2&s}Y?k&bKw5a~#p+Z%W5>X8%9GqW=QNjBaqP%a3!ISU
zHa^TH-1MJYe3#|>`aH~h0ipDyPOmvsZ(!+CUAUJpM2|<#z3}NGvO>;ZV0*9Oz%FqH
z65g`f^^WFEotOE5@8-dPVaM`>=nOTw0d`M!)vqgAprKwahuIzjcd=C-15_T4OMA_s
z^k1Ag&L-OLmiyha$cr82FVisxMAfati?6&ve8nTPWkKabVj`Vs%Frx$rE>S(xkIIA
zm!HPHS{vxqP^0Sse9Tb0hqOV^hm6)XI$y~Q7LTd6d8;Bm6$`|I)u|>ofQd74dZzAJ
z3SV2Ge38O4uLhlQm{_NoexN^m12?ZF%Ihrpetvn-+<)u3|2B?;kEVWN6VykRi9%t%
zu|~YFBg9gp0eke?fY0g?iApKXypqTUsCQ1rZ(m%YVa^x(VIB^LkpQ+nLMILy-vkAE
zSYg4$#W9{64fZGhNgRdeXl_i6&|V2CxD_uRc!yHMu{Dmmj-;^8z{{Z%u~#}qGT(dZ
zl^+J9a_<GXnmpt^DmPN+Z%Jx5lc18Ec@<{niQ+}DkY~Sqm2<-u)2#rH9oTh<{QU-T
zx=O%}*VY;E1u!Mk904R2l0ehD#qM6boN4&mtw;Z?s0a6h(6|#9C3JLcZ${J3vSBmy
zJa1LjZJymJA%cu<Jto=k;xjF_Q{zKSb%jEJwCD>!944Q<(#{7<xdkI}Jw~}ga_`G9
zqByOg-xw9XIFQv><qwWP`;Kk{VyGq!=o~6UVOY%oG}=26G&75K-v!jfMY%iIb$<OP
z!GUiVHB7#zP>CmUx7G(<0v_+}K)4M?gP7>Z<B!@rSdU54S9Zl`JTfP)Kxdbt3RKEk
zVLK+-{PDQJBNqg=W=Z(X=NW3sb}N$yM>ExCa`$B2?N!?NyW;96yleC0wHK}sq`oRj
z^wXoY8~cG4x!oKCgd)=c>m<h&Znd6VKLO-g=L{e@s#93ftNw?=VKXLH5u%-dtN4uE
z6iZgh{DCI+HzwL}lJjgs%?$@@KIQujw$v1ItsTInAYNL&zP>X6S5l^64P9N#KU8=8
ze}DYDUs~$r1sA{2)a1`o#e!VZS|iw|8AIQ(0~{wt9F5vSzO&4~Q-;s%zx*;X(jVIs
zxGzBPgaH6~5@%6>r2RY+v=qu{5k2{hIelN&_m`Wj#opfsgE?wKuHSE9rV*hZlcAK-
z+>JAhjWx7R5MInfeG44^{dF4*sHvADA64HydG=!oV0$7xya2`~aV~kVw>OT0>9_%+
z%xel!2yE|r8HfeyZ^Zt?nqSn(oAXp`r^90?9~m)xURu%wE-)++u&|8WvAUG<oK8R?
zGERvg{BV@~Wvzb|!;8ZQso>sD=|(*N{3iPAcfM3;#eYPac3u8AU;1S~^8DF`LGQtC
zmjA2feG!3M5!f0g7#OqmofiG0+&;@tG-EKMC*%h?{tzwx_qs2_fjDa5xX&zqzrjCV
zPk!kL_43BZm9ig75B~8gru;x0C|0H3e_Y7-uNp5+`TG@{<=LWK|9$L#Ska#!eZB+4
zAvhvlt@ywH?|<+3(R=^?T4aTK9^F4K<U8GxX99=tJwnT;A7h1oxC|(!IWe|?bhMB7
zSK_EP0OAN2i_HI5I<=MxbZW^_8*<}ci30`1VOs0#_I*zGzY_Z&_T<GT6)-Vak`ed6
zv6cWqMj(!fh;+zz(*05W{`!}nCicyS{xq?_D&$WS`^Hdzn%Fm|`p;ABTXFn6#r`Uf
zpQqS2Zplwa@f)|~r=$3-IDR^cf0f72SnM0m|7R@rO&Ity7W+mVKhqt5jc2L6G{?vX
YI#lBQBsLum?*l)|3K}=^uUS0!e}E!YfB*mh

literal 0
HcmV?d00001

diff --git a/examples/pplm/imgs/wooly.png b/examples/pplm/imgs/wooly.png
new file mode 100644
index 0000000000000000000000000000000000000000..190d3afd49f1795245772a5d8b81a50b821d17b4
GIT binary patch
literal 679776
zcmZs?2RxPU|37{s6opelb}FN+BxG;dLRPjEg=6m>IVGd)O;#l#d#@xbMMCz@9>+d7
zzw17#&-?p-{IADDblk_euh;b&&)4(yyq{^PDV!mpC4nI5jN)xsO$Z`qgCNoc3L@~z
zcoj1<1f62FmXXo0wy=Pp+s|TR9%@ACQ@5_7MdyPh&q{}A$h^FI#Xp+DfW0U0jpXkX
z4LKWB1^%_?D1Ny}IXr5^7?#I!l%%17_*BkMIPu%79M0d7(zhSGkg~%dQG-qJ>P8+p
z&5k;C@7d!qqn=ncc6ga<Ry7Q@g<qhd>^w2X{vs{5X?@d~V63ki0p)-4sv7Iw`*0HR
z()wVgqN2uek0a0-N?)@4*{P(iKats?LLPx_dZ?>cXgJO{rZ@51&tpmXYdz=wtEt5<
zBj58R5Ix5}9U1m1)@k**wou^az?&~NIqHtuo;vZ&uM*07PBF%HY*JoUXknH0)mp>G
z957@nJjgAOwR3D|m9lYG5k~uw-Q?rtKVkB4{qv%7#6@|mU-v^fliF!dXut_##?*OE
z#6Zp~_>%kIb=7Sv3!=Rt;gz);ClaupZ(`1zIq8r8Tat(!|8Dvz=KusRL2Lt+5JWwW
z=PgAs0PQjpUB<`zf?sJ$%nr$!G6|i0C&^_6sivP#CQ2C>t-$~AWN4h5hA1L{LM1Rx
z^2BHW;mN>4iVLr$_#)3{k<ClWQ_y{UO3y6#g^napI>U<nRJ*@pMv4YPo8+#Pi^jQ4
zQfIO%{E;A{jF4|a`~-(*e1Z`1Cwk2Y^$3234Du4Y2K!8G+4Cop%r)$-8H0}tQEy^c
zJHIO+dYW`1ZjBzFf~83^8etR^W{)7hO!G|o#21|`3J!&ZmpWN=^D-+JrSDVi`8TCo
zv9l^sTRwA?-^<v2qy0li`}#W)7yQ0}1-TKKuuR!E-Bx@CjQLN>B+=gjep@+HRY6AK
zcCzJ}1sNgVqkg;5oPHW$X4c93HJ(|8{=tnL8Yi0Gr<E@*d{Qx;E!2B|T8H>?@K(c9
zv)&@Y0i`9yr7KGmu4gff3ZFP8Uf73Mm>QntCf5yr{MqQ!nTZ^GQg??M;Zgk6fQSZU
zQ`|V++Q_cRuHpf{RA2*(9>V*SLl8p{iS)GzTC*2xPj6p-N741n_><nGt2yC1<vNi=
z@@|0a73-*qEDB8`Yo2+bdA|9J^JfxNelicedeSyOT|3RXad$&&!|Qrtd@7APr?gFy
zUa4NlK;*zRlGN8~6B+^<j_fnhH=;$N^&M&|o(k!Ts(9zQm+szLzq5YpRANv@P>Aen
z*&^BEPpZ$Xug1QUa!HrhK)-E$qx<$8e^8FnRL^cnN(s@go4+nI$UOY+u_@$<IDEJd
z8C_P4;g8*n>x_LLTf@w8pZW9|qKKCb>}JUiUhs-5aol#uV#(gVU8;0Ou}ir_aY#u)
ziCU>#kzCm?+N@pZ%cqW|xcO)gme5yyv0JaYUWu`Isk-Wfb3f#0jk^{t!nJdoF}6Q;
zCw7d}Po>VpS54|pqvA!m49$smP48J`zUy39D^gj1J^A9SYMTLv&RdoFS3RUTm!q0h
zOm%p@5?CxfEFaFCo)MnGn;|?O`#eo4RVk_c+}*&d_iiecs+20+&C1D|{Sce?^Y!*^
zhJvp;5qBiAMsvjr8hcKD8_56u_IlT=2kUyb-c$XIXVveOzj3$N7+pstADVx@z(sHI
zgTQ;+*9@6P`92E7Kd<*a>$~;-Mg>W2yZ7Cmcl&oUt1bh7%CQSwsbG}pR^?GI&{~bD
zOY|)~@;wwk>?UF*Iz<F*b>{6;?x!A4N1po8S<|hDy$f3@U@dSpthIINU+DiV{Q2c)
zg%?cm-&BTGU6VHBz4>n5I3Hh@_%;4*LU3YVeAvf%8+F&~n5VaX4S)4c+Ven@Q{_<2
zQ{{3Uef&}SqTHq|*|^!@TxE_~S5VCJyYY8ZV!kC>Gd%M7VPKVMbMAo_65XlOC(u;(
z-Sv59%d@HE>6(u>3eOFc4s|(kRottHTSj=XVWcs5#P6P@$;32kDV%S$YNb+m8TG99
zL2k!atB1OChhH;A2^rJ+Kb@<Lue5QPcezrzU!hYe?$qmguEM3dd4+q0c)0cFy8=~%
zp?N$}cSd5-RQHdfT4H`%P3|6^cfG`R1UA2Jt)MgZY&Na8J?0+I-J8S=sQ2<T7qq5L
zGEI(6OE8%-Y0xsC{y{!`)?MVb0K34OYczCAOv5yy<zLN48eO6;vR)*-cxh38vABP1
zn7R6*S#xk$beI`kchGAY0_h|f>SocVh345NkBOTTEaGu~C@f!-z-vx()}E^$`5mve
zUwy~_Nch*)eaTIg7pf{Aoi?0;WmmUdlAR1~hrMeQ4%T<GgZzWaWDU)&CMz#L()89u
zel&M<-7tNwrX=;8w~uAZ^xf_8t5zycdVc1*WcS8Nv>cu8dSP^7D&c2BU&3nfLGjo7
z%J<nUcaS}!B@cVG!!5q5C*4WfOY(AzadMjuM^4zh9U`$kGxn=k)#ct%S=+MOkopqy
z5^8MyUaFL__2^Ivo=7$N{6_hl`jX?XK22sFW_Fqp+sf)~;veKi5~r8+YjYjD=bBJW
z_vDD=X5>O!4kqtO&>rcnPxXwxsU^6Qb46j>Z2QVG-5|0UEBjW~T5bLl-VOnSW#`AH
zkLARIj5YdphQ?kfd-kcw!lwM!d7Al-CNt)DRr>5pja=@9d3U8I?lmJqLc#|at@H1`
z;d*!I&cFTpiEwUJxM$e@hDE$ZLTjqvuHfe3r=z!T@01n2S?^uXnrb>X%I2BWnV6V3
zlca1g9{VCTA?hOA>X8lBNjJ~#u@PNn{%=a57>9)xo{ahz$*y7}PEUPuw%+d#Cb}AU
znGydedMoqeLNCV>$pWqDoa2Ywp^w#vy0s<|X}aPjK2jUfZE{U52cgPwj@(^FYL%j1
zSN7XwyFaJ4>Jk);u5K+bRLpuVe|BkI)Wrz-#M$YM2>JC+&Q{)Ax|dR}igepd+APOD
z*tm!BpO<W9C>Akt>soas|8Sv3=>84onbgDD#T}0wBzXY?BmFv^2QMw(Rd0%Q(|xn1
z=KcsiCN9xY@u!T}yjhO24&EO0?vIuFBF`m~@Vpp`*pWbcY`X_atZzh=NKmn%)1uPi
z)Hzgds?w>Zr8y*5V$4TE+u}HQs2`9XMei2R#?0mw`wbrT@sLoe)=||M?^o}JbtX+A
z!^XBdJ-&+h#O>Od)veWbV$LxQFzt&Ic_$luG5Ex-<K46W<8W{bJ)k++(<33x8Ce(Q
zq^b>o6SKu_O;rf;WQ8FAAPCwAFZut5AQxT;S}=tm;dlt5wok6QBMN>&WUj0ri^Z)|
zdOO45l~eY&^_?IHj}!OhgqjDa3?g<`RFxy1I*osVg8TLO9a#t>are~LbJjF-W43oh
zKCrg2WOnwjw`9fvHND`JINMkE-=2}f7L)J<sGs5@KJk3*NN`+6$4$L_<c5XFHB_f3
zc~InsmGkpyEAvd=3mS@hK00j@;#-U5^p(F#!+(|bC$`cH=ydsQD{5w0_YQY_x#WiJ
zP4HVk<?6$DXg^otdBh~c6mitLr(AW~wx@1yWBJ6wvnQ7Mza+5KDJDD%+E=9aFzBr=
zwRufT?7H{PYWKN`gP`?AzcdM|-`(92ShL@Ab1(e_ezbMQ<=ymK<}J4Pz0-}o@4LU%
zIVMrMy+Ovw+JsTuPYOEOdUqo1MPjK>h<y2|CQ^k(2}#>w_<1$+z%8fO@7EDh6}!JT
zw7Dp<e-L{`oI&DCV$sWIB*=jykIjAutI`=A+Q>trXR>tUoio)6*(l6dU(ZhW>a61X
z{aXe}>*qrr)Tilnjqd#1OVFLw$BM67yxB&63EO+tLPbs?@#WQl`MN^Z#*JO77J^I7
zyxQD3zpjxI$tm~nxkL>e84T)gx-M|iiqg0X>Y)qS&<HXP4+SkBIbzXMzdk*A7T|Wa
z+25>zw~)8cy@I!JQRMQIn_p=kh9OUj%?fb})0i699n@i_ZmKj7OwSy0OhqPs=9Qn;
zB!04UCPFFcMaLbFlSSu08r#>KW}XZUc54@@;XZfC-SEXN_j~C-19L5@Puu#An>*Bz
zkA<vu?97PhQG`y~oDO*^GNr-ObW&QRP20?#|6M|(vTH%RjOE3_JoM}%mCLOzX7`KZ
z&*UK;`5b+2v79*nyCm)?R&f0MtGgSSzdm#Spu6Pw#4T}6ZC#Paf68yl?}Z@#m;knV
z)|K{Y;qZq!D*3YDChLTKGkTK8bE9l0C8=+_wN8aMil^)wrw6*Z|EkR$CO?O2d!iUY
z`a_Me1<g#C$(2qp+}imhlRTZz|4Tz&5e?nE!{-a9Y}TLf`yC0r*6xb<LZ}|J^z1!n
zRb%SIq9{u_DnB7keY?zW<H_$iuElG(_PdV1_@XUNv#dX;|MvQoZ&e<?%-jNf@fwRn
z^)K1(@>WmZ(!uW(R2RBGxnf9aik1kS?Gwez#4{#v|9m|xOvO{6)Fx_-dGut3N0!Z!
zIj3IsbW0IcDi?E+P|s@{Dhj>(R*2Bx#P{wVv*gkv-&&qVm1&eDRSl_(Vq(b0PWVPi
z@kkbaSdwfEc}P4$R_1NJt@{A#zwS9|BKJ|<FS~5xB6r9$;x9}On<fO;D;g#O>y)>-
zTdQdIe&x_#F_E@k&E{NZ7xIJ9{<Qeo#6{0XG^V~dn0TDaRw5F9aV^=zVNk#heenC9
z9{cpC@oHjkFQo+Ib`}G>?BTYF5)Iw#?K|%_rRSro4BmdNh}^rh*)G{nGI<h@mnwc6
z4{=&aODT8FQ&843M%-<FQIaf9>|%&!hn2Im49Q14{Uz<s;=x+AkFGhNl+)v!lzTuq
zcwWJSDv$j~dM#y)RDqdXKbL8ql~n?v45ep6O|nx=CBFHbAgM%j^Gg-i>uQhaey7S<
z^mM$-<CHGe&d9vyFMl=Ef6{%>Hm>WsAdz6|QgOQlGqZ;GT7I?Ay2H-oiCsy9kgMYR
z?44&Bl&=w6l;5s&cYD@WYRv6l)vpv8W>p@vmNZwC`N?QYCi_!rQ(l8;fbeaCkC*OJ
zj8A{FsYm6EwZAjS`zb7+AC-3Pkr;88{`t{&_IGW6K1L@#3T=(O<0avIHvIGQuWrx2
zyTnl;ypj78FC^O*wVXcx=&!k38FeuE(lzg$_64F%p#|3k8!0rg4OMuu@$xe+`BS2=
zy{_6wua7g^cA96(?PpB+%RU-WO}ETTddB=({1N&=-;7RcI#ckCTE<C7?FF+br^YMk
zl_Od$H-rqY-PF3pY*_b5?``897Sj&`raaG)cWNJM5ldD^iA!q>jNPHD%ht&*eMC39
z8kUeR^;I>qRP5d*O9qqqmj>lS`w7l``Db=LRQ7Ka*nY|o;=g$(u--1}F-i`(FWX!*
z|6M>C$<J)7hU}ettRt?-C-AYWk)W!%vnQ2SK~Dd*!(<tu-iN;Jl>6FrwkA&D)8(aF
zk24Kug=~3Ms@Z!idqY_oFF!korcz1Qf8=njec;XD5&OnV@(jXYX*nL^Y-&G^tA+($
z6dG3;m$E`WP(54tVwTIW?BsIw;3bi!yv!Mn2&eg;$9`fN??<jg3cSv$3Sc-JVyb$%
zTI-9CbcFm&JVmc%irUi}D-%!Q@Xa#^g6pF+qwcDQF0X$16p&>J7i&t$^M&sypR=EF
zO<&87DQW-ovDHJG-0T{4(cv4$cakI-Po&JqBzdpK(Q4!p7|=~4%#5D+hm95~SqygP
zw9nGCOQk#t&?U5vv3$bqLGdC(`|Q=!&nE)4y5s9Y8u?g>CN6sDx0(+cyr8i-S8Dqr
zwCl6~r-Ls-?q1G04#Px0_&@AcU{13$5%Lh=rS~<R;(mXU?yJy2M-pM!q~4X+eHGK`
z;}OM#&WbYk_LZw#=KCed?_H4dy&E(Zovtwx_&$`D<^fqS*PG(b#SQM8=4GEIFIqZo
zrYmR+F4a(vA$H$RhCaSF>HGWoSAh(jnfBPl^Z*pGLX>~6bt{GEeyp+`<Xx1pdGd1n
zBV!?H;z87jEn;a*|42{U+50kcA3m|xI=yXJI%_Cl9q250f+%CWN<6{F<*n+kr=*V^
ziY^!N-!htT8~K5#c_)5-TK#Utn1;;$em^OpF>3>vGq2H9kEIKD_2ar&^VfHZL3hfh
zLX4*na+Y899p$0ux%b_^9IjMhrSsv&&esAaVz<8Dzg{GiOLN~oSL)Y;U0OHISZi*R
zb8pi=*{`h4<6EpPNg9RRWHTRCcU*b#)P~mQMlIbk?FdniG+O_s%2RL2Umf2zbUBIQ
zByVkMN4?He`+SnlE{*SUtvs>iiC)x`t`z~n@V?YF?xbLXK<h>5QH5^-@BMGdw-_H0
z2W3b<T^Tf@E}<u@;tHfvw?cNSIG=o8LbCalSD%7ExLcA#<gQ$TmcLV2i*d=-6Q%(d
zG+e2T^Pfek{F)+c<ZH2d;LRn=B&VOmH`E{LY@){IE?j%k!1sH2=#s?2<ab{;5v5lj
z9%rl4ee94b6n?(S>vL&)(NTKg$D8S%?fB56)AfA<pNM|Xex_4O+7KUpFrI14`80dy
z7lCpd6Zu2S>gE2mC7H(^r_vpSxHNpD`2%xa>F}8dliZ;Vi8W1-+=@Jb%3|OqC^`I=
zA5nMfg4J)s!xTCNzcZ(Hty<??6(j~atTb}Yg-#a+BkhXx9LXlxW1AnI4B7Q5AIf2*
z_aEwPODfC16_Rnr)}exnVc>{a@R<;;ucYi@ssrjLjRF~GxY(q94s$5A4au479&APr
zTUfXG>xZSixqlTTlii54mnacsvm!TlD3;<z&pOOiO>GJiTo9pS8w#%G=5zECx_h!~
z&Z*_H(e5NZbor%#sfZZWfW&1=O(W`_AVY!mrv^Met75axr_itZ>DWz#d^Y2st$4|1
zn6M)@`GT_aBo)6SD4dDQo_ckkd3kQ*m3-!{gd2Da`D9np@rtwbN;dA@{K6K@_bk07
zL*8!8A)@(Zw5{wu_43OTUzGEomjYA^DauM|yZ>4`l!!_kto!Y8^v20t*nH*FnG7_(
z>v@kFfgI&&fnQcrv8JuNcQuw>R6ZfJOZwbS2N+8Q*4Qbh(_MGSe-`*l%oX{@@iUFZ
zllBWvY^B|#743-(-^(-E%&9FJls<mkWXnGBlfYY?tUWrq(my0*OSEA>Huf<dz{R*l
zJio9o=P3KhTU^4gzdld<YG~ig!??=U(|<4c?|1terKYyOFHD!}9XOX&8EUhz<n&sL
zu}m&3ysJ4n8eG7P^_TB?b<7a!A`pF#TwKwZ`9*_#fyzCX{hhyW{y*PqLhbqfzTm~j
z=-2;ZAn0YUB?%j~QTE25rL1hJK}rG7Lj)pB-ohEZ`tRQVz6buUmh`{BaZez{*CYZ(
z6VckWHPeqk=;}(uIYrpUz3Wq`C*dHyO7Pz|{{1;uQRu>-pZ&dP$xM9DA-=EfR3i|f
zE$R;|TeUB9ycvE}N|e{Z#^V@q>hHV$``y$nDC+OM?Mh0w?P$IAcD>+mBWu_bfw;ph
zQPD0Y@ZanI{_KxArv`Jz@ERA(>0Mn!AV%zVJiwtGEr{;Khbxk>=K4G4_EuKx`5a#c
z{zb)i&!P83fz{zvZj#&4VqT=jIeO$|hs8Qp8gj>7de`@;)SwpQel<oZTXXOLI|Tll
zuj7U3hoEUQ<9~Zt`j7dM-<QR`7{SgZ{9?U0b1xG{X-QW0dU(F{Qu+7EHup$6vzbSh
z3y^-;fvoh=<%UuCXC}zs<VFF1!HGW?`oBK=Z(F9zI4_ee9z3v+{p7swBeoNddZWYC
z#lqrQJKBpv87!|~-T}ScpXHu3?<}06Hpci)*{?gq<KFHE&9~ZDV^HAs^ZEo#2q1c4
zbo&VJ|32#w@$a*QJkWc2XcX!}@t%?FC-e1Q{gvzMt|?!8L;Eb}SXf>g`r60*XE&{@
zuIVeT72HOz`iG?5=lO!ydg=+F0YP+ov*{lj4LM<4A~*NHEZ2{fIU7q<`W#8XccWzK
z1Z-#%@z3!Z+N!GdE9nt;3i|y}11F+yosMaL7>7V)e*c|^LOt60&iS7$l6Cjt7OSC~
zSH{mxpYRS0zB^SDAtWB{F=NelyfyzYN?FE;)~R(Nb0K=Nkmj+cuR>Q`Z{IQ+CF99e
zE+B@WhM`kzfBiX`zHxf$$D)yF80xoks_`@<>0f(FI@EBPx?vtyaBx`-S@F|`eV*mx
zawmn78PQ7OPd%>Su9J&3#@;S)8a~?h(WJO#l)A8ZrTj@d>@s<_+_l``vpD|es6w~4
z_L+Q9sCc#df;mRgp?)jxo6k`+r-`s<|M)uS;E$L82%^v;O>=i;8^4^xpBSR&Mk{T%
z)trTY>!18<*|mBkoCa)ef_KL%+=PT}$FtVBpxna<{8}yGKgRvNlO_?0`t6z~)-6Wm
zT43PAABv(0lQ)r*)xMf$G|PDI_~XERwkl^-t-FgdMi4Hs$Kql^Fr#`1uV!-xlJuR8
zp-lL*<Lb-thNtIU{2NVqX-+3&iJ|d||5=m1HZL!&1A%z>@FCU{{a8(=@?iWbi#cC{
zH=^j&|8fXr@3Ozq{yr`WukQRFpBg3`%P!#ZCWWdkjN7jNImmIDtbVzw5xerC^D5kl
z&b|_a(Y+pi5G2I<zl5q#Fmg0&Df`KOcOz-o;vy5C<=+jZ)AR(sFPCwh*ZdN$r3uHt
z0+PUe9s^E4LAUAe`3Z7Sg|?U@zS{^Sbv*zHJ*|7LQlSL{2U702?~7Z&tSjGvbzD#W
z^BDU4mdV^=6sow`@96PLkp?|}``^!O@Esu=JOfU}KHf`P!v;<vZ*7{~%4OUk`{$k4
zCYlp;@!@NeV>dmOs^Oi1JPR-InITGZXobx0R{Xz`SBW8inajz4%ZYtS)yTu{ODy|`
zbKQ;+mef~cy18}ZVp@!j4nYNqKvjSMO7>ckV^0734#=h-x0NJ+BLh?){`*27B<Yww
zJGH+B2v3Z~afsRv^enO+`qq!FBDrso&A>;F#6Nc38OHMv<V`bi|Ifz!Gq=_zBo4VI
z*Q6UceuH4B3j>)5P8dqrju#Mj^2AbYrgl2#3cl*bp`G;)xBBzyC0ysXQ@b9dDcA)5
z_-h{s8B1kY%kWV-K!iDsvZ9)yL78+ehtLHt{bc90jEc@{xKi+(t2VU<*k-w`UstUK
zb~7Rbe14B+7|%e^aM=Ic`}>86`p5FJ{7EL*>pOZYulqbVK(f-wQ)OYH4J(kZY^6jL
zfr3kgxXu`nB*+tS^NnG?c%q!F3TqS(i-6QWUm5?Aoo)mSv4cq}G};%p7t&ng=1PBj
z(FlTLE<EJe^(7hM6I-@3Mo#c&qc=FRfVt&Z|94Ezzm}24^0KExc<S54Vg(g5rqyzL
zLs4(E#j88Ta>!o>wl_}{*47=c!exKXaiz#shJ?NQf!~qnj^1Aj`P0MxShQ>VjZ8=!
z=RXGm?z-G@9W$BkCYPaGz;);Ur0{{TO;6?GLEyOv^d30wImYvHFTb|pGQnI8;QCtr
z<3i3~zaH%sAd(u8b~77KKxX}JwFy8%{<ixoFO{L_z&4Kb;nkhAk%Oa*1&{`1Wj8&k
z)tw>;|K8U?c1ijplEG8I8t0Wre~*%lW>YXiHK86?BIA!;dRYC~ejRzZHJthS4X%`i
z#RFrv$x2>;{LTMUZ)iE9lbhNqvrRYd^X-8%1Bt0aEXU_|<20ql3dmXDKNy;@x<fve
z*WY&^MSF4ny^#h1mE7KW=PG@@n?7gO1aO&F4kTy!%%W3cc=o?KuLY1>9y|Ej1K5;$
zvL!<xT`s#&i--FI$Rzh4tMGrGT1`YCx@P<RI>lm9sK)V@C=^O(u`P7iqRm>|&jnO<
zyK<A!gME!GXnSU6X4TFYxZv@ke?DOk3XEXfU+dd{>1nFsaE@zov&J)q!lOK{6xz<l
zz!x{XUEIMO`uaI}xPWEp!?g1DxHf}urwvPXBZioH{u6Bf!4ppO$K^0C&To%@$5vBT
zt};zr8ywqN#Px)Od2pY9(GxU!nJYPwN|WKZ^Z{5y6*`<rpeAMDUJeW<oT3Vm`!|~#
zpPGJfY##bN;6KMn2>1J0#`CT300iooLABlIOHerxjR$>^`3LaEt$r-}?OtRt7e1^@
zb)wJz8AXjiJZTJNt3*bz(Ba|`7#d4~aB$gjy0sI(^18p;ofh$ZUbOa$7}Xz}K|;3H
z7yE{!swt}ozHkw)5<LNB1oZ~STwFVEpo*@Vv+TlXCHU%bjV^%IypM+Z%}mp9_wqlm
zIc_EDy{qmQ`h{s+7513fky*6v95~1a=xP)SPS3*@o^g(Qjdx`~*$7}nSzc@IE%E&i
zNR0F2oQU>1j1Mw`wF8=?@ZC7v<M7>~r8ti(_P>uak^gJuf@Yleq<HZcvVim|N%p_e
zL(CKWN=xWgc}GXRH|@nYn2er7y<-p)6M8kcNQ<5B@1_UJO8g&HCF8Is*ku{)z``kA
zmImq6{>z})g{K@R7!68Xf|1_@zEYp&#lQAoUUVN<nGm>CFb${WR;B+P$+&9nSpOxK
zg)NDLNH0zVqGM(Zo9cBsWa%4zdsJIm9SWG1QQh%>VNBM044{xtTye26&B)NOl1=F0
z1<*r(s1+Pd*T5Z;MF1l#{4s(d4sDFqh!cdQ<FMqhdkD-8;z6cK|KPhTYcLPy@CXz}
z<JC#)mFsw|)<WKUs<N`xhg}`XQ@1CEDSKvoL94gq%k#9PwsQ`RcEeWZ2_97|Wq;H&
zF|J$00@wSC`CR^jdm%#(tA3D}504%+{@C>K*U|-rpw!^gaas6l7RhyxhH8?I%~<$2
zu#KJoMgp$?T*h=x^P95}ec2@z;BrwX8>LLW{!>)LGXY#jT%@eG<%mzlZ|8`-#}9ip
z0s(t@Wvk;CJL1UPUTdGN!@FzlMS9622UYA+%lrj{K{0v7DAmYYKDI~y0A&uYrx1{;
zK1Nf^(8@d9d82@d7q=WEh?tqsO8%JZ$7r=ojyJ*LGH|U1?tc(tNtJ}Xd}yo&v~+S>
zR+QrwegBQTrnc%B{K~nxdbs9|X2G7#GP$wAlQzq!{4%@MnuH}+(y@y35seq<U0S11
zsv4D`BHH;f{c*?f!V$7x`poA|MSAdSIh>E}xpD_@Dcs{qgWaEW&ZNwnpcVNwDHl$e
z!y$i5Zo_D{y(0-nuH?8Wulw1eQ|kD`ckeW{;e3rke5BMDy9XS=Zf6iZY~d00f70ni
z!Vj%e4o+KJS5e8Bbq_1JKqHb09W%ev$i=h$&0=y_tcW8OZ~>CkfkUBA5hUYN05|>Z
zzG@;&`Eqd?BOisspunO4E?xOsha<a|zgZu{PF|V=<xi|MC)dQunT~qLSXe%M@cmlF
z+-kEu2jCRwY>S7>+c?;jcrj&^WwkX29l4dQAC=1q+U6aLcIP>F2Ez48igP2uE$#}K
z--y6LH(<m8aw!pi^W$!&SObZJCY1_=YXBufEg-?db{=MpKaBtwvUeU-QV?8mUi=V5
zPy08?``c)amlzjdDT+?<vchfFpQtIqZ_&AW*eo&4C?osHac`%7j<K0TF<=ypMl)U7
z2q{L={kJCWytJE}upbDgxXt|zE?VHIZxP1+HD#I1H?r}<EJCkdOSrQo-4a(G*9L~9
zr#eosLPmjYYxm#;Oe0$Y2X+kZ9!dV)<I*T~ol0X{qv;wC0s*6jVGE+?7ngY$#Se1p
zXBd@(>>atACpwTUEHLnyW$aMpFcdj1N_!Mj>uz0rGrNmPCQf=VV7<fwAT^KqIi0_O
zT|oz|IHiFD<_{)DH*Yr70pFu1{&RB-;O2aB=;oFQ8PqWHFE}XWNp1)Lz_+gIXMzKB
zFXS@|0GN{;cm7W{bfOQT{URn{EpmHE?oKC+vCPq;I+PrusrYRmxzgcqVXT*pjY*>5
zP1>+Kva$tk>-_&Jgg*<Sgh1%b|2@D4xG@nrb6nExy?YjgIg(G;7y*p0eKpA+=d2yT
z4<N}qKCirBCs>mb29Zx&<UIla2%;ktI6{P;@BZ3EbBG{!@6N+TY8s|JW|%<<Ne3v0
z5?3tUd&jx~kiNw7wY%Hz1-mf~dce85J5|A^DG`$EpLgb7!D^Kuh3Kg!pD<hdrEagn
zy%wmmc7A9+Y2LkoY)tu>-w`Lv<kC_ac>Ov(0GvAm@a=%*PyYe81V~d-Q{Z)K2#>y+
zl21$mUSlWJ@Wf+57WmUj`%hhrKh4CQLajd6fd`EnDRq<%6W|GXJOOYsLQmi`Xt7`a
z_yCYO>djry+RJ7Yn5EgoL6nW=TX;n?<y{YoFBBPuL`mqp)da8tkv&FE^h8Sb1zni@
zrJa})<D>NT5Tw8s%&@q%PmV}3-55Ge#FHuuoKYS0;h5`+W-aUP!!J<rT}KBVE7u!s
ztwmHX4j5{#)dFOZh!HS5htI7<Q3kk)^9TnIkG<+7lT~HUWKcaHB9i9xE8gmOi9SR9
z-ttuc6YLg$BiJ<qhJ+(sfw>&c#0}P0<8;#`R^eWc=QUhh@`p3=$pSzz5j^hK8YU1b
zPch1KttR!phx=QL!<l$c`WsOB9F@&-k583*^zifC;#%+i#~%gYPoQqcEmANKhc2J>
z$JZCX1#BCk_aQB<Y&q&-RNMWtNpoBfS?ClsH2yo=)RBa}o@X<@sf{D6^Dt`&dvLHL
zAz$IckwdAXW;S3L6cuU-XQSf-P4wav%szbhD{ZA6W`tN?kJ#1Mcz890+{Tou<vJmX
zu;YTqM-3Hn;mVF3m5>pDf-NkduCu?RtWa*OE@|NL2fJfxI_49eKL%ZY>l7e=Kr`VK
zK@=`S!Y0B5f-cD$_;KO+s4Y4LCK!`^+8FA47QATsb*dOFl!-3~sK3=F##rp&!kKsL
zRsann55xxoOMGg1mYfe)wJ09diXD9%9rfGW1U^qLIOON7gDiX+VOD)2oq{%xn#}FI
zidyXA)Jcde3bw}X-Mokgz!SzWa1-?mWmR79k>It|$~<z!5gl(gHXLBU{@SsUlih$B
z2aBu8DgEAn3^>}-t!g&yc}OlXTtlfcE8c_lsWrv|tO_7}>ObD9!N|}kU;N=KDuMid
z-NlfTLwNSo%@Yggkoi3e@v-H(k$_1M;N#-QxF_Ac$D_*CYvdw+*#qIy)Ef|l)HV1p
z!+6p=(q-U*zLCj_P}a6UwkHUG*-sS!$g?Odc38pHbcOC&(>>yNUwz^#x0$FAko~C<
z5^f;<tz1)`805xdkDOTqjpT^WKt)MveVp;v2gUj;kIyuu-15bP)O3pXwwNxlz@3TP
z-VR6nEt(q8qb-Fj*_LB)g-am7^2<d+^W5H5H+Sz%-~ZseF;<%_`?bMA3GB@sJv#$#
z?9q<xK{8y8PfB@h=l%cS@$noSND(AW%K63s4J*Kfq61Mox`2NAr{g|fs0<R`Qs6PP
znvM!=+uDD`aIXN{u$_$_#Z`}KxGk4ME3V)*q`a3A>SzTS=^>)NQwMpu3qIc{0Eh;H
zhJ8I8eSLiyiwq;kN3Dv4@nWU@tIG_9CO$I^0GqrP>)LT*(uLLn;H<!gglo5~EU4pZ
z=;(>H89@YMd0UXjDdI2^q?!&vpL*c<FaA&;FCe5Qjy4vRTHHJ610?X?-m2dV3KiA}
z$|XC2`Xh@Vk3T^HO-E7j+QMXylasZX7F5O|R*>cYF7hfp0A<j2<8ef6Ba8H0f=@)5
z?SLHTF8Wmr2$r8OFgcM!qwIwCM>S-ZP3V6HbZvYG4WRs|sk>{pK7$4nac930gdAu#
zxZ}9uJ*$2WwQiJ3G7*{sZ%1i<tBCSqhA5xIem)JNKDXC{3h{k;G@m*FzrD?LZBlV%
z(ed`XzJ(>#F?GCd-+Dg;;`^e}G~?9<I*^h;v(-DlQyH>xNApb?Nbpnp_vR{FPrzjs
zh1&D>y@0q*8wRQdjA}3khJUs@?lX(-n^V&K#?56<0%3URKZwu8r$+o~yI@Y`=FV?_
zazJcLs?EmOAM{L#5GL69`C!Is)e{SHIvhPp>(-5S=;p_CVM2Uks2<L42DWII{mOYP
zKpG7f6<f#VU6>p157Z9LH72hTLF23w_n%ub$lnXsVIzx;YW5|Ve{cY5E)yc`E>!h{
z9#{%xB_b&z*AY;7uEhd2PHd=HePq0^Nei7b-t*3NQb@7~w2d%Pn!D9Ev$IuwAb~d%
zq4idKbg#;?9($k<m<yex_uNdNx4AH$BR`b~3OE>hm|S9^3Qha!7%^|uvzyp8>xKZ-
zNFJC(oIpmE^?TO(o(EksuCc}0$COP|;|;Fm0vR1}K*7E^!fz~S1?CgJe-RTNHv-9k
z{VZBZNc1?FGS<)Y5r(bOO4k)ozE+&7Eb#TK_1(CaRp>}En$@xC6998c&zTu`a?!RJ
zVKrh<B=4pvgO1HVNxV%ALZKGZ*EY;uCda=$X+uVR=;u)fWxo5;<oAWJ;^f#FPp=S`
zA|rfB*^>_V^e0+*AgCOarCcU7R#UPPW;$0E?N6@IirbVA*{Q4j#Nba+$^Za8b8rab
zGq=M-_XDFFERzQZ2d%&>Kq!1&Zwi6=L3)IP?-7RSk^q|PuWQt4ep4GD^Jh^aMGAs=
zD8BF=u8DS_6F>qGEBP`(S9(3GK{-)E+<O;>EJ<scp2<0WZc7()D6P~2W=iUXMc`FD
z2Ffa6Ti54{m~v%aW}|@#?0Y3<86h@12C^ed*F+?<YCk4QVG&t#TwKi`YO-V0LKl$g
zZ;e2jK}=*pNDw3tL>3I9!@vBt)R@}bJYW~$c7l8l%kdj9U<L5lBQ%2KdfIbmuX?WO
zN*u8VsFG@wz1K%35FA^>7NFt9v28@7z)#Pxz?7-JuV1NwYU{Si@PThRhEpu!(BUv&
z!uLz}Oji$D$1_#^YE1h~Qk~zBx47RFdZe%KuE*9+)6dl&0H`)*`;R-ozw?=pD};(q
zo44kapm8Y6Uz<cp6_8u^&)$Jq6UUYBjWBQ;rvN0~QTCp>=K+pfe(9y%64$gk@2YG9
z@K2*@q?a=|f@G2cVkw6#^dk;a<NOc(ZXj;5PO(L`J|m7op3@LN)A1<h-2L-No2g}h
zsGx=n`vZJvO*};nOk==U!&pw%(CF|$3LQ^r_yNoxPny-2dzhH=T_6h?la&SJ!o~tt
ztfCi7Fi=Wav9iUTDrXbef{9Jd1wjFd3z(Os&D7aG=Nv(TVPoCC=XvoDeUGC0W~SPR
zkG}ATuZ`7rjUC4l3(G9p!^OwrD5rxwC5@i+AHiOcNBQG7>@fY(IS_qW2?PocSo)t}
zN=qJrBj#op8R!K;UJZZdab-!((>uFRb3YNIh+Zv8djwx~31AYy=j+T1{aBRJ5QTFs
zBPk%-3pYS9u{S>vH8#KR>M<SEe&@pb(cO>q?gXddZ2H$xM3l@%jSF@*#U<E+58;AY
zS<rh?+3bs5dD4~JB*0p5V|7Sui0!~`)E{7FWq&MgF~O79)x}C=D9i1MDd+kZ@?#)(
zI(Y2GsUwYZR(AmHi9D(w)*Pq=h}_x9`?$QiAMQ&W4xvzkOWkQ>HK~0Inffb5t6GT<
z&)~@3zw{P+EM-|9X!1QrV4K7Zbvxnsb->Khv%zKnGZY(jPtlujxN*eST$`N&+h5Wo
z`B*lzub`p*l^hSUkFw{(6p;;`WE~6h-*C_u&|&NRINuKfXXReNYmgf>tP+~yJHCQ5
zo~_$}hG%yPp>dgP=m80h?|awkL$E)X3#%;9dI`pD<W)vmG!DnjEZ*pSUibCO=6-lX
z*j*;<6!zw%bDD{!p3uk&2YWJpGlx*}(TGo4cPOjv!+Fn~NJkQx+*PKbtvXde-X5=4
zl$RA7nF7v2YEpm*cdbq$tkEhY-E{{DA(;iOwCm3Cz69nS<yOAa7tA}Cen{H81>QLn
z9e#r_Dr2|DW6@ch4D#<~P42W))j>jf#_vaOw}&7!#Lwb;Q@#XfndP24fjA^61d!0~
zZP0$XrD6SGK9S`$+$}p)1#}Fx&ul29l>gqrj0{^4yT`TN-;jfWf@$vDL%*XvEQWc<
zWh7)MV4U(c<8iYLPX)|(%6Bqu14e#sUhWfXR&rp^{I;;bzsXJV2g**0GJ^5wGa-2{
z**)kCQ)+J^aALd#K@l$*1(!m2$>4oN;=i94)!V0KuiLBwv%OqZp@bBJ@WieMH_SP_
zU3vQPW&#AK8Kbvc>pPw6V-`Tge1(xuxBp)R+pBZ%#!Ck=T_cPhTHfwRGC`YZQXHR;
zo{%`lw?t)_k;e#(A89tY@zeSwcyWfXi>>C2uO36ci>3k75mDPb$Lth#*$=aYy|feW
zdXX@C4nFnPza36>ADqlT5USvf=TLecw|CUO!PQ`P;6KJ!5VWIg<`5Bac=j`>;ihyo
z+$Tbgn}Fd=ASV`_xbJGd=ehN?$IvXvB1ZZLJ!*K!Pj&p6V4)HCfRH9op&MdP6Fp%Y
z+P%dw-M{RbHdm(>VgvX!r-*>5UMqh2+yl(u2g-M<jV8Wagps|==>09f{c^zfAYpaF
z%5}zKzn#^i5%S8`Lgg*b!w?h-9vuGf{)b*lU$&m;fu~bDfB}M@b8xMtNV|c_nh-r8
zswtlUJUR?$)0z}XWkwF|l-VtCK5G|`sY-ei{1CR3a@RUqXuRYUTd`4C9~~4+UX$TW
z9UHYpdJsYXg(1-LAwc9e$kMsjJ|O`p7dlxflK@c$!Uf1y5wFwP!vy9w?2I?eo;GI;
zh2=jkHQ;eP7^RG~Els_@cetbj69#_<7qKk2hxXv{0E<Amx+i@h6M#t9&<U@V8mGZT
zz4aeS?~hR=rO%P<c23-LVu9@ad+^7p$Y!HKm)UR#4<#Ekjc3tdBi<-y@eU6zAx1bO
zBnwR3FSBp}gM;%$2k3<}*Oz{7xPb8cytS+*HB;wz6^!nZ|F{{wzvar@HR$=p9drN;
zG=GCM5o&u>V^<DHa9lTpsbY9VDNC99Ajy#sbmE`OD+@%b1!+#F$|f5#SKvD43YZDY
z8=7cQ93_*3^%x%L4ZUM-1?Ehh{?W<eDkSS!9eLN%s89G~l)CfqO_Bjrf(FFVN)!*;
zx{in;Q%l?&0)^``E>(@(bz@;D?}Jem<2sBf$eHA$;UQ_s-Z9scgpHf6{F;g1%EBO#
zb4YJ5+H=ra0}>oRIK}fE@tpQ$O!KbB;vo1BhchSgi9E(4@4SP#hZVYjCXgKeIcGL>
zzpAEy5(0e|57O6XX%@YwBmKR1z9bb>W$*o{^p_6fm|Guk)XIVz)w(zy32<PwPb+u0
zg5BAoBzzpu8v_KjA&<r|)rRokl||qp4DH~>{-MMl4Uc|8DisKz*0*Ejqm-@PpmT8T
zu6d;u-Q*m|>5+{Agvmg5qyvd>&bR{!a~h6qvBm!~FT|BX?Q`$sEguO&;$CbugMs)l
z_i}JV$xO&zRWhcGQD|Mauj%myv}P|$a_6D54%_XxGsKj)0GDzjB&fpjqvxg1d^?Al
z)b5i+Lsy`X`;%e$5L*`<mEpD`&cbQ+!VzHRU)}9wQZ9qj2BGvgeT6&3sKOO8*ZbJb
zcQX1n(dFmO=Xl{U-zAo#JnRDYz+$DyM$2=!^FeX<C7b?`*M{6_KZuilfjss9jm~;j
z8)66(MMb8D7epy~JL81VeNX}lC1zuiSsF*p3)AqRa8=<tr8z;Z3Lyq$cHw&6oD`V&
z`T7Cw*t8g3D0wFvtsOo!v6~P9pb%&nRxEGbOF>;Jm|5&QMNjftD^k$B1lOG~v!BLo
z5%z!0MgqvU1-H#oz!AyHMt6{xc^-%XifeXurpSPkDhBvE;EXzbQiK3j$}R6mt6zph
zLV$<SZxSo_pvT}oOZJo1=5m~1eYcs==JJ{h7nMYcoGi>5{V)Aov!0M9rNM7F%5IP*
zVa$SLNNzM-1dP4EcjvWRR|0EWUP=1`Ut=K@)%tdW@D*Io2Zxi#YO1bodsfYR4wH%H
zPD%S4YNAFL<?Mxf_AX{PgTvg2s3dMMQ_2{fnJy$_Rw{s}EwA}ajE-3Si^<TObb044
zbtr^5nBT8tbenJ7|FqE!_7LTzo%>inx<M-&^yW_Ffv-Uhw$~t96f8;X#ZSHr_XwDt
zN2Tz{7Smn6?^Hdq0+6#ozAF{t4v5_ZOqHD$3AJ5^vv9rwK!)m0CZ)CU-6kmt06`iZ
zKB`NpOYNSUtKO@vKdOY4XK~oLBcn<~-znlKcziH2_K4}jcPno?(5t!v3=Syt(S>!@
zygu6ZSa(`FZqTIS)KBXY*luk1lhfroFogrlchIOQFJIxCF-kL+SL<<ufiARG=6Xz|
z*l`#^Ua?f1l;80DClyJb#dYL$D{LP#<=SfxfH!Fv&c@JZz9wE$Qx#@>ql3&RJp*Y_
zaE!+VHnq}jsGz4QgVg$+U~#~cV~LYv#95&S))LcOEJ8-sM-!s_P=hr)SHB~95J3^y
z6`r)@w6rt@PU;rliTV<<;pkKV-4zc6qP_bUlcc{draRs-G8u~l6c`m^3#6nrv3I@2
z2RPaaEhesX*hhM*e@H9{K$PZ<4jwpVVaM(r_xnoZwUYReFoEUoqlh;S>p~W}eSJyx
z!T`2}?ce+XS<qjA?X3?(vBxn{Zfx}rHBM1_j~*NrG=G8{mul?h)XSCnp~e}?-S+^_
z3;NY2`9CHZ$@w<xlfAYC*M2t-5lznHZvO##5p-7_t*~ARH>j5$Q@AX~lahu#0;Q(y
zq8Jhv!p}v3RtC4m<FkU5y1R83!?J>31A8Z`=%hta3Z&l#&F%PFysqQv1(^Jd0PS|~
zJpYc>Z)csw+EbH&)Oq>!UXF%im8c7l2f)3X4iLk%zJWCdU8wbn0S0DMx2rLCbr(0l
za%5X9#Dg?0W8=Lb`dCX3m)B)$jngj7m7C4N_Qhl*euXaX6_h~XLWw8_*d$y}oEA)9
zhx<3mKdi>BU)}T;v)jHDT;?PpexkUt&Gl5nMSTfm%M@UonMq+U_lgblD*ZLZ#8Spo
zrJB*TzFW<6N#Np)%+0j!yhVv47itNKU=&?g!M3xHPGeh(`5k?edUR6UJcfhjkVs<5
zGM#u9$Sz#;_A%s#xXH`8F)~g7tlJh=wIcWI0y=f_ZlX{?Fom4i&rYL8@VL5N)1<VP
z01ANKsYB+bw)u5HvDCeQXR4&EOc^EhFSd@*KeFcLz_7MyyfWlZli2WGnN)^!2(Mw{
zx2x`=2AmpfT(N1yL$EY1q_yUi1#`S1Cx^;OC=&FXKyU<@t#=YH>mFPg;pVW?!;LpV
z6W+=k{E9FtV6)00G30Ju-Y5+i{VED2R4Yy-wzixmV#}r!_Vwrnq`B|W9z4SN)Uv<*
zn{qRZ+}g0-q#j|~QBHt-X}2?soOSebiUNKs!*^bdNXR<e1cLai`Drsz5hS&viseH-
zj6lHC*N$)_ucJ~WA!!Y7%9x)U6HHE)e$KJ!ErL3DD!Cm|U1RJWPx&sT+vIQoQ{LPP
z=vTjE4NzY{l8jRqh{@z|#Qm}UrLub$=;g{6B}mwqiYOuV#_3I$@7Z%Wu>Pr_sfxbH
zxt^)rwaT`6Vl4?MF7jOdWUGH<Eqzi6vl+=S9up7Ol0J8Vc>QKsus9N1Jv6$5b(?WD
zWhb<=<6%yt;;&_BK(GfMRv(Iv;`EOzcO2ezUNiIRyw-KiYz)XjY?ZqKrm8$0k~b<(
zLV&iz9f&+`;#GRfBr2R-OYg$CPtQ^F5p@MuS2xiU7wXfHf2KB$>}@GKaCMP(90a=9
zso~7M{2pN|EzdGC)^fK4qMDY4p!SVW92Y(_*^hj4?Rigk-nq}&Fy#vdeyzy8D$pfe
ziY4EuyN-$46#9MK#sv?9hJZfWg%=$M4-e|G5g3ixZpZ_sf4?kGH2dw}`?S*Z<UeD<
z#r^B>jGtot^V*tD*0+h78541z{(5Dzk$GnW6CWQ2`493oO>GoKgCCBPu(S1hesZ$j
zJ0oskC=_<f@30#?E90vMSAG_A{{8Ka;r?eMi$D?xKmtx`6I)H8B5qsl>nl$AA6duW
zUZSDsZNF~X&bgexFaz7I%1p{cWzv8#vhM8Fd+Y~Yn(o)QUL<edm|vv)R(ZJUAA#eN
z29F35C;+Pnn50UE*_c8<kjr0f)<n)`QpzN**5IZn{V;k;eg){6uV&@d20M{}7^pHv
zmWCrqt`XFzk`zU>wk$E??FpHjfb405qCCnwCq7oUQ<Zq|1Ly_#WdQa?cJkNm^-Imy
z3l|QB%tV1PDH=6ny7l>sq4?er$Vq%G3!~%Df%6kwN<MAh15?Y>Wd$0uxNeG!$EI?z
z(D_wUn_$zIJ`@0{2V3I45;ck>4mrGRt=6JPA1(l8pzk5bs+FXT_gR9XkVz^>o1opq
z>8Ac_lIVq{5NdZw#Iya4NP4qZNu_6EaB326WXM|2m9=};853>yOO4|WX)&h8Z5(_B
zoC0X23JJ6l*|R;XBK`9W&_LXTeP5JSdFWu25LL?8M@D0mnnbsQ76YnjL3wLr{yUWd
zf{P(wh};=ry)r}M&gVJoX2q|^)Y81gV9MpX9pF@kGjB<U_R%r_8M!!-vra~-FZG=o
zx*~aJIFZTsw|CJG6Ca56*P`H4T^%HigaJ^r<hB7&MgNlWy>es!A2r7O7-`jc<tnVE
zzrAkHVQQ1FoFEQX3Q1rOBLGmsY?~ZVV?(fD1l7MN7KYkgJ*vj8U>S=}{dA<)j66K1
zApb~ls2lIwomD#mT$FkvZZZnW(ysd)Zb{#v521YLb+3ZkCvECg)9vzRH;}U9iF4VP
zv6{cRgI7(d0p?<zjmZU4ywR^UuVf+vRXD!M7J(u+Xak;6ftJ*S!`0_F1|ukiRVHM{
zopG#N(IB}<T6^p5f>vkYysfETk06+<Dz7E;u%QjRtNWPWz^x2WPgmh;Mfn5Rvxfb_
zNg={%rS9hle&v8tOG@;%fw*3+R3>Fpo5*#&w3C%zL|!ZwP8JUfo(wjvJ~2?h?`QV{
zNE&3lOREgIS9YX_H_XK%lpCj4b<mYPeUFYVZ1s@}K=$8&*EJv%rG_0Vy%}*zL<9mJ
zL9c(Nn35PiU<sxG($u9{Fr@eG*a;RXw|2sc?cM_<MWrnVXsPSGfzIOK{M}#Sm!yRf
zqR{o&Rct+$g=KOKi^1L-uCd{-+qXF8!|Uw4>CEcA-|WLOoJry+EoKkCgahuTZF%G4
zWa5kU$oP5%=B)I?3jRqG|B8;ux|W$K=>sVrD3LzQA=K>cZ&J<W2qqn);;mt*Q8tNa
zUF;5RMko*A#6pYGZc9K!8}g;*mEiS)pz<39CsvesNuhxl{%$LHzJflIh#<cKBDHy9
z*7r7^WoxLtr{=)Vbta}u!LNE}mginNi}M9nYQ0Hnjx{z>#GF9`)$0`Kcj%0W$4FQS
zr<8bf*@Q?h1t;_9#C50@;kS3V_`iu>=`Zr%5I8@3M5Q(jnxBSn6&)UL&b&P8aSO3}
z12wx^=8rSf3IceueXB+vc$AC5jN_~((Rd+9uKpI$O-})J2a5a_H=t34508Lk0_7^o
z=*LiWYBQ%}eiKvSu_wT><|XnbSvr;DL}!R1-VZQ+Qu9+Mc~owo4jz+tO9V5CEhf_Q
z@<GXfa{P%|Uj_2Tkt0?hL6|9FvgWU&ongF&JPsC+Ua5sor{T8+woyv`Is*K)A6-!b
zTUM*qoh*-^rfV!a8h;>~kkB8dkg!Idq<@4gu9nM`X!TWDu5&_Gu&&lFx6!C^{Q2;J
zZw%G<jTHem|50@|#?)A`XmE@w2}2Cb?(C~6dA-1y@$v|HIQ4{c)}g&H4+)bwuin-h
zgZ#dI3^FQJy{BG%j_np%^TfO6$fiLcNC{DQB=-+gyjY>INMYC*?aS>-vEu1XcsHKn
zdy#&~+u}T-@7-t0l>L|Wx<lir^Qc9AJ-l6AU8#z^3q2>61skWul+oLZ4qb36{xG)%
z*;ByIeoFD;)Uoylhk9%GTmb1a6LL^PmoeihD|<Y^sXYn-TqRCRB8OA(uFsB9bJ^(f
zbBNYHhMyk+4|J>)10oY${rRC+K(#V3@0coP!WEMghhS8f>jlEhS;pq-;ghU>4(P{U
z-Wue8weuB&nG8xKh8}(xr6_^tk1DGHGeUMtip*N77tD36_ytIljB!gImqe&=LhsZ<
zhU)dQ<+ZKF{sNQwaG;V0m!fy<Piy}EfRhUV#^GCnuI33Kh6mz+T;TaEXWOrpm5c#N
zP^_~tAE5}xQ~p(DsbdMOw2wH`0M%T`F3r66KCGu?T_tB7i#48G_w07KfD>u%u!DC3
zcE{IJpL_P%uqMeBYEc#X!@dCfx9MhcyX^rZYLu7Pv=>AW&F~>aL`k<TQmpfUua8bN
zOkeg@Z|`mG6E?L3)Lcww?lysz^rg130N*j^;?zJWoHFrPCyG+F+Ox>kFeLHFkzB=1
zM^Qo|n7MlSlb)kL!D9%&=2q#*LV{0Y4tKDJepr?rn(=Shtu7?)Ifuw9m9dBQJ!rG<
zX+T>HKX?M3%2=wCAa%WBn+C==k?5zs^;pgw08nvDh?0u-a!AQHYm%oUWT@==0&SA2
za8fpo;N>3IMEn;VHCA3Tk$`92BKfu4Crr9_z4KaIBS`D_f$|?@5eRZbsBjD5Sqd4?
zuP=F1FipDh?$u39Igh4HMb4i2BU=O$=OK_Q+J&=Xr60>_c~*E5^k=Yl=^`UJ6D>(*
z8E)?uto=00wY(!8v<w=TF0qpzi;eKT<F{&l3ua8WIV7Jn0m<FW!lTGjM)FQmG?D>%
zC$%0(yFjPWa1STSKwiu89)cf*`GkCu3LazvkM~gBsYze0Te_B>Y&T|TR$s!975Te4
z1iM)daMq09UN81atNS=*mMM~?KeyG$lRkC9*R5Zm+pAfcIM1s#G#4i*lzwV;%eL>)
zR@lW=Sg~>3rF*>)3~QZwuyXy65jC8xCo}^@TsgQ@gy9yqk@bFDU70;c2q)n({A@z+
z;EC0;(n6l;NNG3R1ap@0Oh{ABD<9s1>G%X)#}}Q~I0EN?>{rXNdJ{l4%ECvySdr8C
z?e^Z972;xb6GnE#QF(pK?}>*W<qO}LpL3$GIXVs51GULBpd!n+QaaqCEBQ>kE&oY3
zk6=<$>WPT*#Pl-zZTCWL{WSNtLU=M@Xuo0wNMiW~lb=!(7N@S%Z|&#sT9U9!7}RDE
zKp!2!z-oAKIdTIAtF?QFwqEycmk%E7M<5WOjb%fmgGNWB(r1h{SQMy0R3&uT>7GA4
zr(q&S)(y1kH>GcT4B@D!$<b7f=wUDp1}b9rBv3=QW>+ODo`a`#glmT_YB;(c10}@b
zKk5cBOTmO(vYi51!78{yH59aji}{a|9JyD34@%c_c*ijk+n?;gpW_h1TQxUs6M(r_
zk1MSWHD}%FTFIe?vaBgGvW{i{kK<SAQ#bb`4|&~+@=P~uzos$8&1;*GLgVo#*@`v4
z*&EEyfwO26dmd3pwj9m)Vboq<sHyE<5k1xP2~Uz2u#rX=bcEdZ#`#Td-{aOnQWy#7
z@UT;<N#Q3#OCc~tJ9QWIPVzR=?U~<UrTNi^ELahgJ8nm^&Y1M5RCs7^btp&+(_OZX
zwj(aoLX{pr@B$ul^G@q0uc+P$j_#dGs#a|L@%pJ&9+>_EDPh45|Bl|c&#0h;65d*g
z5p|2ZsNF2%pIivZUD0dy$qJT#2<v@@3|#>}G7nlZ%{-Q*<2025c()^{$)sS!G!#e(
z3u|`(ns<9>+;$NcqGeo!BGhmwei5S;kM-xHPX)Po&TE_-XX;-@ytf9ox*yiV;^6R5
z`;E8v02q<XXII+lqx(qq|EPNNcqsS(|9_AaMax9zSW{ujRw?_ElA^2?5ra0$zAs}f
zaZ)JD$eu_Y*~-2RQj^3KF>0*Y(pX|J7{l**4d?y+{H{OF>5p!@xvtmpTpo}6Q{(De
zk>K&p-`k(+6rVp?71lcq!JaK)&Np-mT<9__WcXJFXIj>y`5VNoQ;|g(n=|(!!MkM#
z%QBUO_)3ssmfXTjfN90CaNigi=;;JQ6TG%A9nu5BKi&6Crw?(Sbcx6}^_@QzEMg=?
z*p4X_>33K=oW0^a>CJ#9L`@4z#kw{o82t{MH2q`K0aztvhf<5XYULkOU_Tam#B8W3
zL$c2CY`1hx;(N?~6>}Z0ZN)DX&W9H<giMc2)w#;LOt|}u9Wmx&S`BbD<F?WCy1^*M
z``GUwL+qvdULjo+ihf+EKt)g}A}{Zf;ybtfXe{ypQc{RAHTwR-FE3LJga1Vg0o%uj
zF^_2AS|pxST56IJgtlmE0Ge>A%A)q`Z~Kef{#w#PzH%*Lo6_|nME2gwgKY9JbQ5u3
zKqg(OYpSJ_mcj07g1jyYJ}j!X?>!vZ>sJ9{)R`Efi{3fAR1jM4^nH^FgVZ-a-9G8h
z5(lt^<U1fl4JLo~&}U<medkME+q+)i+ZUgnO?m#(b{y%TfxC3szL0cGC}K{jpFE`A
zdTjddMd#A=nvMRumtb$abMC1O{6;&f=)#cO5whU3q;~Br=$9XTy5&cMY>k{Rmrg9d
zM)?jE(AmxIgQT4fkv?3nT0dv_P`KAixYFonA$@M{JzL&)P8M_p8ZvBs@#yo;NFnOZ
z8Ug%Z%Dkf+@9E+NMB`?6rO4G8!`Wj!qU_rHeg2SrNUl_9)?L$BaJ3HQWKskIZdPwH
zzw+ApBmeo$A?zM2?2PiD9vN(5pS^dWA9xU{in0W&+LLn&gs2bxgY09wL+{3Ed#yKt
z_inRR`==t{&s1IX!R(r8SNEf^(3ZgKM3L!@9`}aYNZPHse?mT=Xqad93r0>+`l0T+
z9LOA<G**B^3|xk->ioRm&FZJpf<eK8`@BYlV=OS520ER}@R5zf-+Xt)(5dYD#3JDd
z5+%&1k{^9PrCs}%gRQ3l>_kMiw*Z|6an?@MNTfZ<-6yQrMa}b?-4wqs11&vC&u6qT
za>#exGa?rJzqwtJzMOU(dgc^Fli??w$FirPU0km>3^(dOi_?}(ylN8_mb#bjJ17Ag
zP%pXv>afV!;sg`1_Na%b;ekcYGRfB9+&o2B=?*fr<py^G!6~Sxl2UjjD1a(Rcr9Bn
zmN3~V7zX;cy^4^CP6=`iT(HQOloJOv(r-U-0Q9EoOvu#)?9Ue3xf484!aI*ZnD+X^
zyp(?CVxWmsQGak9xj#kw<=SZfO2(?k>89XS(i){@9$UNeOH|L>K~)alLQs)?dGQe5
zWjScspUgH5g^yrzKi0}B0nsVE|6aJ(PoG{19(D@|Rt2-yP7q%s#?{LKE5>Z!vh+!}
zAR~#GFaK0YoRa{ZCudyDY><tn^F&N=1(~nvxghBjHt6Hp9fiO9&z$zG?#+vAU6H`k
z)%!R0zbkipV<Z2r;ZwubAJHH(ftJ%jyY=GpnQQY)X(jSr7BBx^a8SCMe=V_$RDwE@
z3o%U$9GU?6F*!Y&01CR_w&&Pt5Sg&RtjA-N*4HhP8Hp6O5tiBJv#70T`-}zaK&aF`
z^)-haiWI}iE$YdU7EZO+hjjO~f^$^~syw;BzpA8OEfSc~r_Brg9pm)44ZUOOAYCg+
zCfro7lFQwufJ4DZ4C2KEs=ScP#eZL8zpULkf-PzeTuZo}r`Sf!>nYz<?*?mM)J9#?
z;zXCr7v^oR%JXcqlGI@O@WL8}^ydVvd|t`aMmanIdfF!33G^ssK2E4{lp9FIxt~4_
zMZ=veW%137p|hbaG+Mvz>C?HixuxIkdlsbeqE`dw`1*fOmQrjc87pIJD+H#fNB?F=
z#By37dpIhx-&qvuMQW=dT9X3Gg=YB+i`Uku{VqX|MaY;EoT`y>1}S)+&fqT3s6tny
z8xUexrmU983-#qv3`$Ib_G{=FV_kx*6d<of7X<nDXK+Tn&w4}Zrx(+nl>|e1Q*OWb
z{dahV(1J^L5df7sazPU8wXwep{^87C+Tp-v%wuqBCYyMx>O6yvIQzUgv$_jQdUGa=
z5jlKk&vQHS=fg`oMiY=;1R69fl<R9=<|d10zHCAHW+=GG_p4rD|5OyrGL8I^821fg
zt%#oI-(Oz5Ol{58r$xLQ1!IU3Z^p0)Dgerc*ofoYS>LP)`*P}d6NFz;jjX<I*%p)(
zrWT~9eXQqt^`^)zjA!&^=RE~OGSMr(n=<8v2=$a4TKT$6EU41F_K3%oFSM%ZxG)l#
zJT!wyZbtpthw<9ph$a9OlAH<*q+Pi#AMd`W;h|ry%WdG1YBHb{exYN?_5_1Aw(b-X
zS(m-+7ngv3CrfOoR_pg2F4(Fin!yGC)NE=nV*v`pla%>|$?vz-{oH6{#a8Qz|2Bxx
z=10j966%K8RO4?Dvb^<%R&&p_L0`~+ybWBfV`v~!L2Y%p=>Hqn#<qCg4%|9rkhz6d
z_j&QM&=S&0xVi3>FqcYa-bOwHrExqLVp&9%81W7cI+o$F@1~k{{+$mF`fMYtZ-oez
zP5p6x>XAO@3fI5{@YOy(K*0sy*e1lqIEV^)Cy|P?OXS=>ocKX<m+KvYxko8?JGWX-
zy_71)*Y8&RE^A9S$&_x*b;gRA*H74Y-gD_6xjx#9LZ=J~bPH*!AwA{`MvM6Gzd3L%
zS!xDlj>=cyR}&<gp8ZQI6@4h}yae|2Fi)Z^t9`icje|M>n2}e-ZkAf?aaGrHP&J^?
zDPbQQYyEnUUuBjtFyRhM@(zULg|`lbl!bRXaUUqqJPY!usmu`V%VpKkph4s-9G+iU
zRrRCAJGkG*y&Z{oq}o>fH8Q~B`CkmjNYp;&z-EhzHs$_-^ll2vxAtAxk`#EB_An@T
zzET-+AJ#j_7al-#N2md(d$SeD*@PuK7j~6$Ovz2&O2ZFYwJ!~U$Q&zhiCYa40E6u)
z*Y4FLbX0eoSFGK7$@{L~UDK@<RuOSgUUq@XR!*WhJsiqAD&HN?`SiPq^u`KqOw|<~
zYX9!}YEG$+ioU<M2xR}jP)4eWH>%a#F-nOdTn>G2gMzBdR#z=KhyY{LwkFJsFVFt$
zeTPxEcFD-^^fuF@V5>@cC%gRDjQrnJ)p|^eMrFm9sZvDiQ*y?l%3F8uVtmd2$XuCN
z443X=;73&Ag|M&t>??81bY|OQeImBSymqi;&ew|(-^K;yQb)Sh84OzW>G@APch36<
zg8esx@M%|)APZ0dR^Y;}Ar*phTObDCBDqp?=_;na8<YTGj`e~&nblR+z^-+8E&*L(
z$rm$lB~LwQ2r*@7#IO{pSZolXWhmqMW9MAx^+j)-O_7iYM12%cyFlV#2CWezf8I}X
z(Wl&jsyjHy8fU(fgJje*&NN*v8=9jbC)3_lOMt0}{@S=;l$80mZBn|B@aZ?ri4`K9
zen`xl&uwR)#e{HE;`vQnSW(YgVRs?u+8My-Rffi$=z@O4jbh)H<roy|jziGc$^n-r
zh?=;_n#f!6wj_h4bUsG62jTx&gRy`3MsR7avu<vEe5iq<t&$lJ(C35F0>lX3!n<&O
zd$~#|Rdp>c9x=m-Z7Rbnfm5OAt5s>TQ_Q%j8-BvGhs;A5febu0h&kd_e}q$(9QxI_
z)c@MwKG&;p^a@8j@kVW}T!(i?oxx!3oRTkP=-EtN1|iMXxv_<cGEa`kPtdU9nyG^;
z_K_gv`!bbUFG4r@yU>}`WBVaXN__&BanEz#h%mK`JsO|!3vwNNxVtW23(3X$jAwk?
zUcR}8u~{DL<SI%u3}WJ#Cvv{uw)PdC8-9nr9z4AmUTYEY#8iICpFy)tQ~Yr+-prQP
zPK8hMj})_}%ClzE1h#TTEc^k_PNTa2$zfTAz;qoXk)R(X>YcP+IP4Vh=*$oZ{UGs*
z>Aoj$IDw>LwbqeaNz@C%ytK~Ocn}$BFy{YuG2nC)OKNKeyRxuOL~%c_WK*>Z#ypih
z_t6Kl;Hy>zA~EoX=T572b1P4#@Pc&1L$60=@`Ey=HFQFYB<k2&H)g8{x#CloJqJ^9
z*9dF$*Bl#b!6xQKhh1Padx+ZKk2=Tx+$rrG>sT*^r?f95@VAXBfl~_FE6qU=Hy^p@
z%NNDqLHUXjQ+mQfj^b)OS9N0Jea7c6Q+q@$B9$H~hG#g9ML21Y(;$Gjo-<-G6<TE3
zB8gPr-}4XypRhY)aBkv>87!S|tC8Um+$@<@0_)fsn@!PFmo-tMc9F|Dyi&v@s$<!Q
za&A!-U!;9nNa~j586Pfh2|P!ko8<{)1@As9+f5rS$exr5H*g7dVpKmbSz7@g(uDNd
zg6Y32DoIQzmwNA5N!K{X$j^%KAd{{U)U124mAKh+FP;z0sijY?q0d;maFH4BWH33d
z(<()N!hPr;J<*<oN1(E!mD5V7lwe_BxmI#{b>^&d&RDotg@Q<a-<P>!5?_uP`=_0Q
zi#eG>gzJgvT-H7yg;>ZS8eZ~Wt?8TQ4-DcFDn`prfi7nMiIc58tO)f)=qL1#-ZH5C
zLSYgqNxEalu3bdT1Un%WJEJIPx<$Yn!>JJQ&Q?|1Xz7$wSZ7{6uJUSKM-XBJ)j175
z*5ALWLlw2EX}O@$uk8!ZRJA1k4+vA%(jS+Omy=tY9X}E5oNNg1*f5wfvN(PF=u!v#
z{n8)Tc(^t`yHG7F<O`o#l*ilyQ`(v5#?WrBj^}?e?QND3^~Ok^;^~Jxe;L18RGU8n
zrJid^EFQphJS$GgkI3n`lCPh}ejaq=IYXa8o1T84=VUVW#gYWHp^$s$j|9=Zr_ZsV
zi$^rZOn4iwsfuQ5n*ayO)%;#uOs~qfnHGpG6(j|IU+4;bfxS8DU7Z#nFUYmKt9tVC
z%W;*n9*4gixaeVSQa`OS=V|YV*Bu`yZ^?34QHia&s;N2aOx{<6{(-|r(wnTfs3H>!
zkG7LTsXvytcVZ?Ir|{_PS}HR>UGfvvEjM#fm?I<N@41x`JQN;2(HTA>1_pF2ci9W(
zK%(|{`h1EjgG5#I?k4nHNTyr8?vr18cCxR1lEVBF-SXg&dfRG^%Ia#j(JTcwJHLGQ
zDRsv}gZqkmybxjHyMNu&(KJe5|DQYKOK;}>6!j}{3f*NDj^A&U%5@#HS6Dt|w|j+>
zh@MkJs-UBF1rg`Po}N02w79zO#d9Y3;Y)%cFo7xqXpDpeioIBVTf4WxKN76C>gitU
zeUerL&%pl<@0;5DaNBq;G>${Al_7baXESN+{i2XUM1H45ez5#rp{`}F8W8`AzN^_~
z0N(0q=mSQ`yQoVOciFI`yooE(jI+eY;}yHQQwvo!<!%xq?d%mUQ)Ob#Q0MV{SGn)y
z+9*X_aDr1?eygNtWnlPEjc?EH_mtffHdq?z02Q6aXZ55<ja<4W^%y5|*agYKxM{MB
zr!|)p57FE22{*xtzjfJf+F#=d`+X`(lwxebC8&?d#{FU=%2;Y#sYUtpYfZW|zQNeQ
z_c_Tuo%3ZepK7evFp9QN|H3paaCOIo7ZCbexn8%c4XPyJ#P+?X8U?60>r*rdiD7RX
zG~ebh=qkR+3uA>f4WkC_HPGTA)1s{~`R{(_h;}hzvLZeR6A-bI>YFj<b{B7T?77fC
zS<?}r`!AP-xN~2MKJy#`c0_L!v!slmrx4o7>f6qF!5|MBqb5b`DA7@ERW>}HP`7kE
z*NL@vMAvFKrk{+8vb?<uGYs!qb%7hJ6;v&7+Oi&<**}A~1Ecct-Ywz;-ycOpT31e&
z{`R=<dRtoP^NAdr+uSeRlRh7=<O6?))o6kvT&h0%Ui-jSU<;vR5@uhUwYZ%dB%#^E
zP2;Wm&ovCc*Om<F*DD-4FOlb^(K0e*K6WfkQg7^$l4IL$;41;ypD#q%GNSUqsEWML
zzb59d?DxfFZI(9e9=G#iKiNIrblwS%HZq1CbC#|7yNcMtdnA6BNR>{i5yX63Xj!*R
z;%e%8O^fE6T<B<%!?ltt{=#3Fe)F3@8FsBr#~CKJj2z9HbDw8RZGSf}c$mIN>^$uQ
z`lI#QD%Gv#hiUzid1sktb&icQ_W{ki(lyms8}}#dy@YHr*B=>w4!6uYPxUJhRt~CX
z6ii1{+(NC|=Y%*UXG~*klBei%N^;Ufx!l&77Vq95wXG7j5=IY?S9H)0#FALTxzmV-
zfNm~+!nOf1+nvVw#>z1n&~yh+0V0xgUcCxg=)b1EuIpI_i&|l~TO2%jN!iR^M0U3`
zWl~+8w|ywABX-KC;kq{_{D)6I;$gR=Q#^7^FilIz;AR4d!Ny0Nh$<4Me|0X@7~7nh
zj+Ry_<9ytFN@@mfKzC7aELD9z{9S^kH)W?$zS^|UEtL&b^T<z_65nRX{oegkRd45V
zy9ig!ITe=_Z5q*AM3M8EGVj7QknWvRS~AbSE3Cugk+2Mr75`2ru6sL;Ybej`_oYz=
z^TM1+sZx}KE0_JRzP_7G$TqU)j_^;t{+PWJGwd5k$o@0BysG>c^^YsJXz&wR)4fZK
z7I-2}s?MMMsStdloL7A!ZNZ0X)9>|b;Y%S|hRpWsDPNXO)4X?ePsi$<%C;+WS}wBV
zZb_IVzd!<d(rp^4vZT3Oh>;BK)4tO8qd%M}9NTImzle|g$>~l($TBtRnQH&92d}us
zWO2HR1&p~rys%lIlGn!j&uls1@x#PkL7i7tPA*trUp8*S#Qa3>bdf%wG~}iW5#3MV
zO8V{6+jSf7w?2J~LD6=HlkPlm@5cY_yqr-@sJR*_(jV4{dpPJi?`OlaR}@}w893wQ
z3$F{pZsDPk-~F7~hP-HYQ1<UOQQ|jf_N}onJeFn_stCIT5%gYkkGaV(*|Kb2*#GK?
z;U<*Q`Nn45+e@9ln@)X=N6d1M1JIl%MXv6HVRG0#$53rSGOFOPN)A9PcFX?5?u@VO
z;ST3k7P&u%3l{da;z}6$Hbqe2)3+V25;*iGS)`k18^)%Z_c2y8-Qapqtad}Lh>>;T
zu}Q@UPF3f_ky)QfiJf7!Y)vyf#;HVNg9Rxrc&QAkt?KCy47C3GC>MTGGOk%*=X;)C
zFT`fo{%Ec+)-d8(l)wcwyWQsv*lt`l*>WNri`?4|>vGZmSP)xu5*8xLPkkDv&Mp3w
z&tbT%tMpT&p&->yoJNYr@Pf1M#X(ns?q0o70F5=Z__MRDAuLYI-})&cK$8AX((&`l
zmR&-~3kG$=`U?i(@M7y&SO0Z}iB^}Np25i<ZIsClYu#iYmOD)@>#`+aMuz|GwDZbZ
zR7xkZfqB+-w)p95iP4rb#f{B5=<I-z?dKE|ZgXy*+jHRMTeXr9k!s_j%<{=6ky1V0
zm5Sx;7jyZ!)i1Zqugu`dd5S1(L_E)ljC-%!S6!G&OE-kiTc`h~HT-Da9N)3<N6J*J
z{3l0Fc;cOEUl8J*ZW<RM^qBew>C>+B>L=&Q^ag94!0d%CReBF#F;<CdGr`MrkT|T;
zKu~LP03>^$xJ8bSCW(0IIc_?B;coe<;qjT#8Lz$VOeK)iRV^K#Xh*nx;y^wccf=DU
zTR9d=HD?z$u2=X4!<$1XEZ1W9)kdpqu?A0zhh7CCOeldmA7h0<?5O&g>Z&(W$E!Og
zEpnrW@L|mhxi#nVN`LEmAq}&*00oBWD|E^3_guxm!b8?0@kM56bvt6+b&tsk`2p36
z<bWM6M)8>j#vj<AK%lIX)+wls$G#4$YKc~i6nRUHokdRQ%B?qur+pvqUnukQ<DSYf
zn!O7)1(JIA&r04c8J{NOR3BQ>8h;D!IUrSgHKLTqBk02n53UG(-g>mJ|Lq<!{hIMY
zOa0b=StLKOk0I$HBzK6O`MgB%^3Ty5%Nj)i<1<hg(STqifm0{yV6W)-lC(OZL1SCQ
zy9JFPL&Wn}5wzg5v58UjZfAK88E$yd1cp2^(3Wp2Ey&thdr%<qT-pnIT9j)IKM!@^
zTh*bN+ksV{TZsq48*)w6*@E&ctn!I{thF`ubpp%s{Z{HW1U{8yM+r#`4)1>}eH-;t
zb#PemkiF89%3Qpq)sBbkn=PF1r`gI|oGdl@q_Kt8-HSJb+hjh|I06#N=va}$#_!+n
zJ}ZJqo`62j&r68Ic9m0#BP{TpHH-$6ex5uR6P9?^J;v_2JWU6dRcIM8OOL2I?$?i_
zOUbO&aN9);S%{^elP18v269)7erVT5LmM2?0L+74#;}NBY89II0;3si!he(m%#u{l
zC(G(W?mSfJvlPGrPs~ob0Eud!h}k-n^1pMh;L;(DTa8RBD?hBpA7_V9+h?j%;_8JZ
z6e=m$Z3kQ{n`wFjf$Q}@*Tv6#Y$4@4o325THhc2XHTjIfuJ9?S5Cv&Ma-Hm%?uFRd
zo_DBECpUf-Ix!Xp<#uFKnn%<ma`IeEFxgw-p3GI_bE)FbCF8RFl<hj#tA{sb+x(gh
z6*#e&kG+{E;NIcQPQ3n9duAb+?q0%(_yArw0kLeEdyJb@8efhZ)VGF3K0o?%drB*J
zNWixm3OA%R9c9t78#(hqVw0k#_N#HX^7taHv304iNFxpF2sJe|xOe@UNM<dlIaa?Q
z*Wwv0N>$@j5<}!bTb}h<=|3ZR2K4hD%+(S%x;pNbEK<&Dbrh}j;5}7yLrQIqD#MC%
z5kJ8;Vg>5T+8%Q07XGMkTHD53)Rp6m7|@VLZ|}KeVZS)*KJRyshmR)9@2V*c@f4VJ
zT+N^twuPpLxbENww5oG!T=Yje!=q}u7_p5}n9OYvQcgtVNOXi2w$5>+*?sbKj{6U)
zWv<QKtNYJm!fEF=?(5ea2H~b9Vy?z}q;aJa=%Gg<ITppv0<(+5I4|~mDMvggKdO6@
z(F=@!Gilu12g7gRkIDUd!};`7IGN&ca<n00YT5hzVE6vwxfS0Xdi{I$r{ZL2h@)Vq
zE^fu$u*Jp9<(EpL&8(N0a`<ENzTtK(LVQrknwy8VoKAcf+c1?@HtsH*NzgF#V@#7Z
z>C0oUgPniW>2azOmQIju8UHA1b5z1g0BLe4c51Br*>Fogm<te>#l0tkd$VI*ptnmb
zXa4feVcJ)vO=>nfC<#r?&F7Y^!9URtiO9yAEfQ_iMh(aE|DBHT)0${Ir%wopkjFY=
z5{O4oQ2IgoWpF$$?(G=A&HVCTMO5QFfnb5PbWK+r2MzZ>C;sNNELT1@_7sGJYmnfb
zHs-*xtNWOCQ!qYS*^{RmF!~J{zKEl{QrF||<ZSjH2u~8dv~GhX-IWWYCXx9$uzOl4
z_lDnm*iwU~OGe9?la0+f>Os;EW`1)nQvED%ekYNYf|MO}gI8uBCl;Eyoc_GJSd$Af
zS)8z}LFh30OdokDxHd4s<fWgmxop+*r@1ID>F{MZ%L&6GE&KejQdiFIiN^Y`kPzF+
z2DZ;L(q!+FckJKpf3vt^(okZZoTY<gS77Jp_H#KVGYuP%i;(ALdHH=2yeQ=@y>0>y
z`+S~N+rgb6235}~Xp_M%#Bz6Q^J28_s1EPQo>XuucQy5gqNQb_>8HYrhq{(DHgC||
zI8i7csKXrK7F_-naF?2fK7L(6=8RIMRKwu~s>+<HLRm}P<?hMZy3KjJ{ad)Ky9b7H
zLfpHfLrO=k2dd5ds@K$#m+I!hdda=)Yxc<taregF;B{6-UQNKtY;hM!Kf=GqLs&1U
zY#DXSA@E9){DQw`Tv@eYod_%!_UYjERX^~!9E|0>o}it9L4B%bT?tzbatLBpRR0cS
z+-Q<OrPh%OdwEH@e>_<NRL59kb9yh)6l3m2FOjz|q--pnTrrtBDNimf18tw4@OrCa
zFqj!ZU$rNkMRA~S&@MX@=6_42msq`+L_6yhjy*b)=XKKWt!MhvccXsboGaQQDOwe_
zIZhuleBhU-BIkBV0f~=R@dJATAT!L!x)<E{!Q0P-_h$8q5he+T^iFr`H%OHY;933q
z_dJi-ze~47{Pj{g8U1<K`GI?Y6{*l&Q%)sF83V5bcH3{;z)Ts=2u)>C*z<zVhQlWM
z`|i~In2u)=F;2g-8PhDJlPWprD6&a~VbK)%#Is>QtwHk;&i{S!fZpLo<ucE3zs~r*
zI`1uY#W!6%g=K6x4^%Wb{Vnplq9T7fuF;lO)c&73={TaE86OtmMCBU%iuQR*GRtV$
z0HoOa3PQE~L4C~=9~PZ!K;xbR<_$|X%6V$3)U!^3GcG)|UxxDq8!80oxf-52+l55*
z>Yqy;;#9Xaik-~!EcL(Z`RZl)Zu*lcayyC2@Dtv2NQE-uruuDiwyldnAF-bgqZET0
zBJ(gOT66i$)TV`qvlKJq(cM}aEzQe!IV#^3i{^<9$?9!cw0X$G&pidCI)^&!^x{Z`
zFmt9QP~Czv4$rrYTs$^?)5}3CgNV$TS2LtmBU*28tdTSR*mMTl36qLx!fN3fb<Jn0
zc=#LALRfT?TDo{Qx1Q+~)<Ebh%&wm;K1Y(osA(c$#tywvc_g%u4Gu~+%qE^P26y=c
zPIzNE1zgU{*U0^LTX>Pt$)?1u?NieNms1Q8q67+5B8?7=CLkzC&=aPE<j4lC`$L4m
zd}mII9ZGETQ0i|;VIi=_R(@GQ^@59b2AVS_EOou;EtWz)N?>(ycmz5@I;R`7_v)0_
zyaW&R&rPZYryf&_&UsS@ez<qR3HQD-%{xCn42w}Pwan0}bg4J>by9W2m$h(|tauBb
zcNQY@2$Bc$EuRh5x)_enWHf%mlspZ4_O6pmX8XF^*dn5D(~2)XtMzV_tU4cddabMB
z^N7FZZfyR-bW2!xOW~<lJ~O#Ykq7HKazV!Yd5xtY;V(7QIBFC6MxO_>g6R*}{LwzK
zXsi$+^@*4K5`KQc`XYIdS`w_ZFnMS)qMdf<<mllH+|Nt9#PH}(cgiUUg$9J=PXo>~
zF0yQf@Gf$95(STPKDdMUvh8*|3L~a{wC43>_K)kuX8?a+31nVsqw)SRUPoIa+~Dq!
z>2rf~Xnv9t3rxx4Ie0|x=gcv+5n2CGA*OH4U-$I~2G#io(>pI6>Fd8xoU@~orDudX
zze4>HJ(BMH@^9K0&99F{e`03wCE$c*;jJ_KkUBD0C9EKpmY{v=c%qFNpCQ>6>#3k(
zJ*)Fp?;FYJ#7E18pZlKcSx1F}?!?4r>X8)-h^|rFHmjgS#cSK1ikJz#Ft}AZ704aI
zfii+dHJ9^}uNUf4JPM29v^V+k51+|5o2$0(_E9MgmAr#J$L}SSb-b)S7+b=dwMU{y
z{B)4HBOd+TKVapC@D4Qvj1)hnDTUoPqD;7t7n8f=){o%pydKV$+*j~m-P%1U*q8N^
zHqcWLQ#hK(wEnRMad?2bd2PcOlF_UTCu);&@LaQOVe_lYc%mP+Lal<5Z<9!!3|<nO
z7VP5=sDs>)_{%7)*<mexwAm*xNRyw+Y?7mfxH~WG2IHTW4GX+I!;waL(ujjYL%=xI
zbzi-cgQj|#0no;Q+B?D`uhdoHO7VV<+@FE{bLAdXsxaYqeda6lVxA+kfU&v)#mU2a
znHuYP(Dm6|Irskm6xYK~8c$*l*x6p1YP_veU$C{<y<=qLH?1<^lcfe(Zy((9|M#oN
zixhT|98U0g)V?<zbK^!GXlzDb0ZI#?$8dJkEl+e*+Aa`l0V!8u2=%E8lQ9|wEJz(}
zQzC*DSCU4=?zn7pCg!aC(hrq~MBOT0547}Nytb|m4~!hVmfcc{POVX&@?-?9rDrha
zf3L|5Aue4HZ1WCBWrZ^R2llnSGCaH5YVxS{1D(I`U0d-I-gbdo=wv`H`tS2(xh}R}
zcq%n*)CY4tNp^bg(M`N$FMGA{L2U1j>A;s6W(<;{NdBmBCC3xYwqW@}wimHJqvkX7
z3xAH%x4S2`58d{AOyBM@zH;G}J&_!ZRgxITIW)us=uyY$?Jb~8`>3A$&_ro@=68h@
z!*8mj|1AOKU4Ufm;#vK{>Fj<wNdi;nS@1u|uK^$fBEgLQ6%7yt#@d6*z-euGdS}WX
z2Y1yXoEhff)mIn!^9n_KH6#<l3Ju-+PZY@@St|<~GgiH#4y{CW=c5@^W-0n+Oyui9
zsBE}bu0H=C46#{{$Noh)2W7G*;}l0oWNR4BA#@B~HKdTkInWVCL_<9hr3127(1YdL
zYG4aR8&2vq1R#992hy#UrUL=;N;6)Z-(PuZoRWW<Vj`wI8N4ZOTCz9D%#7-lq%76O
z1sNFs;P9ygb^{$tw2yFw`Q;#CDNN2lQ^fsHjfJ&hD~yo4l<fp#vq2Zda=fE+#QpKu
zCLQkkb&6--+ZI_p`8wNh2%V&UdR`=F4ws>A*I%fy#lyYe-6tw#Ul)GdZLRY+=f5|W
z<fqIk;?9%qR5B%YWN0^&%2uM<gXSq=_qsb~92;Wf65oZ@3|#s{dG+T3f4y4#&>f<j
z(3AOv($stnt&BSX0U1=A*-FI|t9pkgU1mkd(z(?7z$prclJ>fe8#*`DEF4W)%o6u2
zz%k?*x#!GY;ht4*RX-8pq!&6u0UD$`-*#^5I`W>-zeb)f_ugeN#j-eF%!}bxUR(T9
zzg!UCGQys&yHOVqi@AJfzK&k9<2rZml^wf6QcWU{%PdMa`6o+z;>;mFmI5e+I1(oh
z#3m6uw%kYX0X?ZcXUB-Pf+m?1Kk2?~vgN#B`)UN^T=|e^m9-*VdEI{v4s#@^d(M}+
z<)ALFEQdHRO{xd~k@I2PA~(F}+rE;$h7VJjW~i5<6>2DiVCr=B@wRrA)RG$W@EttQ
zGaH^-wTS$96*#=`-uSK6tg%SGwN_Nv2QPi<*t_-?s5U%^j3|4nSS6fjjU<hyS88~r
zG^M&%_F*iFo@iU}8c?@max%=G4Y#*=$t$bZ`7i@S3#`9fJ<~VT?Ac@fAS~iq$IRq#
zpXT^;$FRbM=kzBcSRb{*B9Az)y(N*P@2Uff;h+N;$ZC@WfXdJ{QlYcu56xic^VZMo
z{g2ZNB%sjxYYB}X?=}paL;OsO&C7-9xH&#fRI~}=2`sK&GET+=K7d5-KXhHm525}B
zY}<)|Xfq4<ZO9N}4MD@Pa@%6Z8$`%>fyb01*380QLuN<cw18(Ir?;J&?5O`87MB(F
zfgiw`;mdW0v$B=_w@W)!suT21c9DlNfdk#y>ELbW`_?na^d)=*rHhTt5jTZXF}JrI
zD8CRpMjjl^UrYa<81!Y*wd4KN2V5bwy01^Km{;MAsLobm{e`lYcW-zlZt->1zW36a
z$O-Azy%Hg&U(Ee1adY0-_=w|t*k$9<D9l6QKIzX}59bi6Dpyn4f6cg(IId{M;OnfW
z<^F!!_YpfMKj@EZQE0pRA(~M3>KnE;R4TmM<BEY9<q`&ts6#>oq4N15KR@HoeB#^^
zqDf&QiX0Iy|2u(JiSy;FzA>U^=IfB{%X#ylK_gc$dNR%8daiK*#v(i{;vHUdVwMgu
zujI~@CB3icgGs3u^E#bsxqvc;JR#5H_-E5tDe>Dc)2<cED{sQDuzh?+whXjZ4L8f}
zsL2q#L1<^DHF5|+kplJLW6f*{i&R*2VO-6Jr6Eh94gNW@%<NoFZZmT;Ev>Q`(@Z0S
zzG#{+g?SKFZwuN6)S}i?4o%lx4`!7txfaC49r3vM8(#BlkF$2}pgi`<tbis%qy4(T
zSuxdwdyk0z;<50Hnt=!nY6JdPj+Nc;Rm*{eSlw=E@^?5DJp5c0;$}sOwN<<lnkkUK
zc5jOFh{?-X+Rk`qcdxhqUolz%^^clRR+w5e@<OS8ml~A%*p!a19~Z1v8*-qp!t4j{
zb&BJ<;up!IJz$WC(l(g&u>`>?XwJm#j$twXBBp=!`;iySt1JSCIe?hNdU}j4d})X`
zdG=^lM>9kDGRlB8$>Ge-kvF8B>AQGDARL7k*i<vCu>JyhXiK+iL4TY?Jz(NkCfau(
ztz2l^Won8_Or@Z+87y8=I2_&zq1EKyyQ!2;IwM$ugn%78bAG5OZQWU>4@is}e#bQ{
z3ocL2riJHouJb3K8Je7}tji!ydr6g5`kz`XHg7}hGIoAMrW+ul$b%n2(6X-H-ATxf
zomDBi8qZwm6CR&my&zv{&(<?NySWul9eb74x7Lz%uhoOAU{Hg5cF}vqJsU*-?`+v(
ze7Z>VXdjfpdlnP&|KPu-j!bHE|3-Z1z=uQ7tU&A^dE0_4Mcz{gm^rey6>qXw&H`#y
z^?^=&RfWD6(QW=~1K?lQBZ`qhKMnH%PYhz)9HrhawaLik7?A3<7EA7k<$pMas1cC#
z5;5q()duXmLQ8J0m;Km$2$c%Da_;+7R$1it{{ZQcW<*Mj^JYjC1mRXQEfU)z^t|z&
z&jPw)SiW0!Nla&}MnR5^%Q(GnZ>kP=MNY`9Z*{{rDua%w>_N%liZs99^SlpwjUJ^=
zjE0X=nT%z6YHF4m!Rg%K-R5(D-<pzl`3fomrT(u&%mNeNW4(rs=Zr;TbU)4Ts220O
z%eCYwvh%-pNnc}Aa!33{pb&!CIuaoXE=i3|uKVPml>=29g}<?nkg+vN((5D`XUlc0
zFjV{#7|6>4)>@VlP4_=wE~-qD^S(-2=h*pMZ#&1X$jO;yD3*puqN=~yzwR<Tix*N1
z4IKjbI)d8Eeb0JVhDuMbeL*B}qY@8!T%KrCXnrTV+;VoX^oKvLzOgK|81?$`otsn+
zB`{Ef*AMe5xn(`rrKWQHc&H@zM{mC!i8?%z8HIP7*<<}c;7o=xA$q1+W#V$xHKRW&
zTgrJ41kW;m-_^<STymXfdw#7t1K-Kq^a#%*`7P$8KTi03=5d-*lsB##XB4~R)>P*4
z>XAi%0_;k>Zl2$FcaztOZ>+iEVjM+k6gjIk*@6y-PbRk(BC2rKxSR9Uv5j}v<<VfO
zOYd9JK-F8dviP-C0HUy3O-IZK8~k;zx~M_dinx$DA{1odJlt4d1KpPn<&nH<<ZxVo
zSV}iD@l6<#@bs~O+oD?%vJpFLztnp#;9P{RDG_aSKLwb2x?<sR`+&Y<l`k+gFSqF0
z#si_!W>{3v=RoI<3~LAJY`xNZ<a+ZzMON+t)PZkr#3=fha8XXNC<ml{HzO%$Cc16;
zy0BIq?}5U1<Q!3F&L=T}!9~-=j_1rx$=@H%`6`z%tuQ_Vl~T`|vebjQYu&eT&-$t?
zFMayw%H^ozSUuY4M<&ky*8}%H3UkZK8Sd4zj0|`&AfNY%cTsZzCm>MTtgTmN5gD>{
zCKn19zdjuWiHB0KE|$pYCC(aJL5EkfK2-!!qD&R{T<-(5FY;(<U~q_SoP}O+l~yC9
z@`tL5t4*<WZzsvEucDB^v&4-5OhA<~{CQrnEWcWt`ri)#;(`GE8*Dl`&O+N`ed8I&
z4{XcL2k-EHvt8a<A&z*iEUAW9rPNrdh<)LWU37|<*iz^6_MU{@0nf5QFGplByjtD^
zIsQ^=AC6jI=)%p1)wJflaM(l4)gqQbkQzFBPtYgJ^ET<4WZO*3W4{t^2|1plE>iOV
zO_C`XYsF=I5?=VpYuT=2T_L^KJng9MLqz5Z0g@4_TO+LiPeg^pbUK%>oobpIHQbQP
z2U=;KCiKgF;}QU}0}S+7e<>)`VOpivMO1g4waa|vG`eO;DTi}pA5{E8V8D^uI-+^P
zlr@o}?M_A&Qe8f=zo6jtyN_p7M6cvq;8nQBLZK)cd@qi#>5ct9NrR48juDl2WsXap
z@#UZ#^eGbJpWu`m)aC7Vj~$=6ScF<G5MFjO7{`Ia*FEN^NSE!`t0%uZ#=NCGI2zqV
z8*^T#8jBf34hfXmZ{PyDrPQz1K8s{rNi=imk$ylGJK@@h5R~?_fZ1mdgkS2Ql7o+l
z*fd;GK8l3Rs0wfuBSp`tpJQe$kV>ufbS6ya5JMn05<qqR|Lq*Ok7|)>mZy76nAKAQ
z!#h@+Q9zgfgq$+4ba)#pb6+7#!D`SkDF3T@KK?(5N2=`L@X&zY9uSm3mG41Tc&gHc
z%6GZ}Pt*r4``F#<e`Dn?8E40N7B2w3gY*FOpT|czAM&Xmz#Y;n?2<bmqCaqgKL2rL
zrBwQ(n*oN-be%Af>Z$RT6CxDtp^in|c;WZmDJO$0YF|RA!$UkExRA)H8nowX1mKZA
zBI7n5ZlDKVI2{7HxX$Ty@-qVC^P@4qDU%S^T+PjqyFMP|=bzCd$Z#)k?3_bb?Efp~
z0pVXxM%pQKXegdv&03m!LHRf^uae9}5a=Kn3C|Lr!EfK|JSMWe`UJu_c7w&la><-B
z&Za&3k6tUpW$O<$DX~<X*c+}}VvZ33rXUR+>A4$X347z2)X8G%jBnq-jvOs)LG15W
z>dp`NKL^CLFE16%6+xv6THR6ZTxUGE-bKw;9P9tVXZyKhV8#=j*a_*zXY~hyq%}<H
zXQE$`|MR}tlf`{SW_q;D9@;*enDwa@pk)XC8SqTN{^w-My$yK4utX4f8te9v4Qrle
z4QG?;Q#;nm!#BCG0UAM}Dl!a+V~x>I!(-S&0*@4e$t!_)kh7Q#+MA<esk=`04oSqe
zR)e9e(vjIC?iLIn$*HX{A+CWvvVJOIQ@WHG<)G_Pcdn}*Ypa1RdtR=ZvzSEDavz^E
zmWO!-##WjC*z9+FG=$&()x7qxTshU|&E)Ui_8QqVlEJfK7W#T|1c8x_iVO?RA-kO}
z7Q(C$22rtS6w4{p#6m5IBb_)_<R)9qE0LQy{cLkw9MB4n()addf3<(;{-4hZ71SC=
zj(ByMrhywMDzV5TMVKAZAO<}AEE!oOEOZE3a$i`V_n)DktDq`yG&AEWGmvpS2>^*)
z(#&VKa}C!S$DMye^CQJSEjx)df(DvSkWe@y#g7`|bzy_*kLo)jFQ7(jr$Q&ra=%b-
zR*Np%E#6~OJuKUf8+4lrtsk^}@x`*BpAe*mckkb^qQ~u;7-Sx?JH}Dts)Z&=j?hYD
z;Hg6ouTq$Aq}W_ew#CJ!--`Cu1R1K6xr?rYb-X@?bJA34=}Qj9!=LJXp`1QB_O2nI
zPpNWMnN07<rA}jes-!DSU|OG8`}T=XlSukQ(mgV5u&?32!90t|3=I_VU~39;kw*4%
z#(@|WF!~^LKgktW%{oQ^cAGu|+<k;c$x7m2?7~KF9D;o-FO6cv^Jg0H#Cr&B+7t;Y
zdK<#$!!bxz0?<CJu;V6-M?v)WBX7KJ^BHs78s_?um2=HaV#{G&BYIr;X&-f!DhCof
zYC?A`CZM&IN$Rn}JI_Clbj0_(zo_(=k`d;k+1%biALwH2Br<SG4F8<TOrs<JFb{Wj
zg|@+DD|R|0q%sID-i=MXEGO@C!*E-^VpTq=MW%n3MxD2+eX3fW*V7sq*OfY|_1qFA
zAih25XCc*3#Py!ZuR`EHQd<l8tk&@*fQ6R*>H01VrRCO}q{{@eS_a`qbHHhnFFr72
zOvJ5%mW<q>ER%TzGJ^u3KIUsPI#b2}-5jXl4%T(LeiFd^Wv|-k(0XB1kqc}dRs<U&
zes=6H2+@%dYw%lrm7>3_$eZykjlQYkvT0`us)~4QBAqtRipVN|@g6*|+0n2XE@%%v
zc4uJF<u<Ma6ej9yb*$4}aQ#`}mo9gIStWH_PX_acEtqJ}rDL&rj3Ki;KIuBMWKs|B
zUV@BgYS@SIwR|~z`P$lK=8P2wk*Id-`-|#6LOolR@Um*6jr%%nu)}Y8WZ4qiHw3B=
zh&T=aYz3rO2=mx`Cct@993fJv$HqPR7JSrTtKd8%S{oAXvlQx}Vcrm<pio8#g!!wz
zdC`hFf<({2%XTylp-|N~t^KaDa&)OD6t+Htv5#aO4OhsBgqzVd_fuHv)ToyiU#kXg
zNfl6Y9A5zj=KBRewf+$)zxx+Z9>25{KOL?OI`~YXugQC>Ycewa05lcxEQJ+Ov?M?E
z7mvG6Kx$_+PIX8@DQHlmYwRTcY5DUx20jLCzir})$b};^!y;_W-ET1z1;MQRPyF0;
zJEZU*N=NQ?3i1}pk!^pCEc@AyuS{DfbkWk=Nqs18Vjg%YKrgYzdx^WZ6`A-Tp0zQA
zvEJ=(mh>NbR4`-EO1=^yg8uvi`2tenZIZ~mssQy}{Qop(@LOH8U4lC$8^8(zaGgT!
zqWhjPM4;C?kY!OY)Cp3BXPBgkiue?y@BcM&V|0#v#o^F9a4V}14rl3wMu+bfLhDYs
zwzd^wr{qhF{&qpQA;$()oTVO_$;rR5Y{}w$=&=*m8)LdXeEOdufvNqR;ZZ+TGAI0g
zSJYS#+&<ITgq|Vmxk+Nh=Ot^JOXPRQ1+dhC|1vS){2vWW(RY?Fy(i=V>ac@mFG8z{
zj6|faiUfkwdN#vCF}&lG$fX=^)r@+<65H+m7qT$q3HkY^NDieO5B#+~IpO~*^LD%<
zv85k{`_mmFDaceJT5EQ|@p+k5kKm^9KPo!z@}#l)zq=E{j2R#Yvi;%5DA1JXTHeoD
z6Ct_-GPQr*)7FpSk=jK$NGUQWWJ;WW&t@)LgQ^j5unu}0AT~Xd#wQ_Eiq<352wRJa
z5<UmsP2CduduViVq$p=8RFPi9kuyJIKOTYPev!ah+~kOb%2eIzkl$1>?jaft!jL-i
z;FhJWyr7kEMr?a9CIbQa<sn3=dF1<#Up*}pB()n@247sNdsmOx?XK%DD<O!&1jI9^
zzuk_>7_FP4$U|s#Gw{kUk4i8G_Vzvh-x-Ra9?=zo>#PR*3OSy34pb#vKLV`i;(<Lt
z;_sMf0b>P996irftXK#SUFf4B^Zyjk7q}(xE&hm*(M3G<s*%q2l^6A=SqN;w{=)os
zr^-1_yYL*iWpTR1lN_8B=6BbJ;5L#^H(g$f^Fpa!;VQ<L6N2rdkB2-hAV2Tn2o;1Q
z_oVUIc5*w-fdKwAcfQR7OuZp!Ls9pqOxI2O7sm7_-tF+^2)5N8CwgCf@8Gf0NHlj*
z%$;31C4#8-ef|5w_s6q`02E!+J!a%ob|%d60fk{C5Y(S<;eJAdTWBQmf>uq(IAkWs
zolT!%AcK#X#4$x;$8@3pC;+=|s~6nLYq=zq3MATqax0bv6Zo12?6V{8ONSsS%WvzO
zql2<t^krz+_lXz==`x<wfpd^-7%mZo6O|Cv{n>}tid&+b(vMb2N_Ad?e;#5TXu+>P
znTsX^1twSW@d^B?WAYbrIk7&RC?yLxpIKRRh`2s6pOx>N*_f?!F?VQTl24H(^xAEl
zH!iOw3;Dq#P{xJ2x*P8~uJD#u0x#01vUA}Oy}nGEQ=xCSuk?G%)z!h}*@dicr1rJy
zgW<N$;ngNvJeD#cLiFgDa@jjw%qegBrOG}=)gWbZmyKro>H-7@Mw@aJNK6&_*#&8z
zv#x<F9-hR%rr+{j_V`<G+--r}v;X@}q8#pFAu=6B{3d<%+#kV=Y+AQwKwbFIad4(m
zHAPM<RsJ`KGedVm&auT!@8oCy+EtkOgV_4Llrz9ra5K2CWje!vutu?3wJNXnf3-fo
z3Jx2yohiikuk`4a+*mpOs87FVb7QurtRCsOv=oto&1TGdyDlw|!&-_0*KvUmX2IY;
z=AW`0d1M~0xs}F+2J41ls6RHu&5O3mZ3GM`)ReT$Rdoc|V>|WL6+0f}SbCY7&4!wY
z@*8nu^}{niE;BEcA4rSYal1DkOzfI3kg{(pV@)kuA1N`Kd|Xk=W;*;Jxtq7I{tW~F
z7*J*OjY215lAdICIA>Fib2fJ2b23!Z3?%6TFsZb@#D+HA!$vWse-i96Y_tIy9uO{T
zEw2Z$b@=UBd^GZ9HEHs2&QdBZ&y^isD7lj8DzJ_yi)$W^+R|`+N;2FV;YENKv3#u#
z>aaNc5Hd9&W+@o61-I^MN0iCl!FhYgkic>qk@ru*IMMhL;TP0|sj(8PF4k<Vtc%ML
z!4)uVnK8E(Lf4PM&newT21hhaLGtCRf%M4N=fl=*FB|_7);zwTNcdkEzPOkFiKZ-z
zUOoo80%X)@47@a33u*4MXd%(}t7v$+1;~nZdiK%kT=^XqHNg2+pW}vK6qdb`P-8P^
z=>*d4pE9mF+0XMa;9ycYwvxl*j|6ymJ3J?BTf;O637g^!^^Bvd^`qa?@BHN8RlW3u
zsmf^<%REASR+6rZ7g>N?8gY#zgr!z<vpegv;s!ee{EbZUwW({;U7yv*LZSB&n7Kxi
zlSQ*-U}MQ&=|O+G5c(-gv*7z9O6UkW1S-85tNg{^HB7#s4X$v%&Ly6!>bc~q){1S%
z-DGL-i2}cCnYfFd;|tTCD!O|LHZGNc`183MhA=ml6=s0ef?^lsmfSOq0&otjB?6+p
z1iV25PH0M2m@6mbD20WqcK&Yw>#47E{(1AVd7b;Qy3pIUX@Ma)FEZIO(5X9SN0MZ?
z4Ae~2%Jfmu#>N@_dZjE{C%1QjJ+Ze;HTtYEeH#7EKe$HvN6^!|7cDZ%Wz~@yw5y1q
zL1S51J|(d+3_)$ceW79)K2cGQe&EFV(h(8iz~nYyj~$w+sEnfek>wR~_c?JnFwUOr
zV9k1MlJLA8Fqo7hHaj^P4E>n{t=Cv8>z!Z08~d11)*bFy{lb;;j5S}^aWrG&G(+a6
zed%>84pB$N{%<P{>;Q@4jIR)k%LWl4;L_{R=S5)aA@h8tl3C?HWMCG;7Wo@4>393V
zmD3S+Uwb)^!!<JUXZQycD*E&*+05ii>yhVAPnLeVDgt{kc4&%9tR||v{(E0T>;rSF
z#5;_oG=F%q?s`E1**L5N-+1bX+oH+GBUVALkx3Cv67m69qyVodS1$9&OzWx7ISQ6L
zjb&Jh+F>%P<fuIxyf_HzR6vx-v#2gGG8t-qXN0gxij$&kq~JB!E-qvzeY$>f@PD!9
zgc&$4lIq?5y@>0hCtZcx!b-T4Av(ASMPuTpn%N5YKWJ0Wn*8z6C~q4g?#LXq`ZUt1
zg8zR3M*ZL}1KFh7@GK3Zb6O7t6FIj&Vk<KF&4MW!*N0l7hqr$^?t&xf*-tB+=XO^+
z#^T`JMSZ-b{O7)kMxRSqa~qc9dk?P!Pq*+3bapB-epyJDTY2hG>{Ir)nDZ;eAWpP)
z&ZV&3sE8^_16DH=QjTYl&JmNQs2ar6+ccGdcaC_%byvV}{a*CwKT#d)OezVkn5GTZ
zb>5DV1mXaQD*sRHprlcn?4=4xI*|TgDTtN-SB2t!)92lg^93jaa3X-&-+1HRdS$M;
znxsq!5NZz+hQkwn>Ex_U6zAR7hm-cqyAd4#U^H>bwlUC44tL{Vh0m_U4!e7C`eo&6
z-0o|l!pNnQ^_id~71pdZ^X+yiboDA%>T$nNIYu|QF;j=kS2U2Xo7a!+*RQWc#zv9W
z`puM-&)K%1cZFA<-8Va@q=~@&;C2z5F#Qpe6YKLsq~peN&=2QwnO0K+nRr})W^DSH
zkC*DSUey|0MEL)cQn=PghrkfS6G>*;8ezb-LLH8Jf4jF6W^=Zhm|NH<`+)TTM$yAC
z><6S(Ayj5JZ0VaOuR};)A033V#rRI<3-A8@46;gq5!R^k&?ZYi(_-SPMb-M=n)6T8
zw0w1iSmq<M#<cLFwSN-4))3UW(<gz$(4)xi;vsx?6&y0fhU?0)@j|3o0Tbl3LRson
zUb%Xo-zG}OdxOh!w5AI<K4G%`f`V3uS_^$-nCa*zD@F`nNy+imZl7*{GK>Qex7)3d
z&l;RLzPC51W>Uo{XaX58p(tm_^pm$*E3wL9Qdwh`>~^0u=<lCjtwZkf?t72rz;)q6
zqh8y7P;z`BpH=2;{1$P0eg3u7M%NUDD$?)#mn0+1(VzgZeaJ5mGAxYJNkM9P3R$e;
zZO?N|H2T|XBn=PToqRt9_Jn`Z|Jf5WJ77=TxaGzEN*>9s5(9Q@xy|;+fRhi!?;ON2
zY*tc#^McY5aXTOexMsB0MF%{1!`yR2C|8`mu-);sXiR0gj~;cvdAvt3_CaQe#u`re
zY|se^WK*=<$8AB&7jAU)&t~93CqmOV#mKIl!hDI6@hhQ#gJ+sA@I}aM83!J-k`fkO
z8TK4=z>iGDhh<iByc!QuPW<D<o%AAe9Fe++izl`afDvi8?gwJYjt4(8^e8SY{9CkK
zQ>`3glU+C5LQRkXp!M=QFlUDO`I)A5N6=9xIZQ}u*hiMh%h`I(vsM5}c9jcC*3J&j
zmk`bZ-#7P7e?=f+LdB@|&bZ&caDWiAo&n$Na6h)p$aeN_7*8^wPK4pIAplKL<3-f)
zC?Q0(KTY4y5(Ru#PhK%O>#tEREAIl%fi@ca3`a7;OSyybU4*m-Y)z-_QUeaa^l<6i
z_q>uEKDSoW&RHo{?&un9GJPEO4DrS9Nt@@w5?a-F45$bNKOU#&E_E&a_8W`S%~L#p
zFlOo#5C6hR>p56ovc$eH@>*752Kx5}dsDgQJg^)CuCSQ3)A;oFy+8B($b*f^t^f4v
z(K|esnJs(}(759K{H{bVE~0L5JAjL|eR?g^Z9<Rx9L$go8Y95!!QhrpO&wYD?|B>0
zE2o1Gd%8h8ylgrS6hjyVTA}XqW!WgF9$HOr*Mb*T<&=v(F3fphuX?%tZ2D)<hdj0$
z&#N>CPrv0j@tL>#<94NlSedg%JiFd-ZoRUT@nhh7D0?+W+~a6-^1+GQ^(UhLV!ZAi
z%X;m>h81sl>cBj?{2^qla^K_lcAjmUMbADIRg7z_b*Mi7;AblhhNn0JIIx;fa9NYt
z(Yb~fAwk<~mfr6}n>q8`ZDuMXMPu!|s%A#m`@GJ7XQoJneM?JOQU--VE5d_AqJ?Ri
zF}ks`w=8aDcif~k=ThYS{3qw6vc-IU@bSM-(XCV$u>i0jpoz*EbAD4-*4F_1H`?%R
zrJrWJ*kPzD+giasL&aa<-keL8X&)~>i^pmu^RckOw|t;TSRb@i68}Ccb#V9^j~N?E
z>+&Q2T3OfGU$@8oNX7S;O~ht9dhXW0t33a#t7l?tpIu#`u!5P=(lYbU1*n0Z`6iyW
z8r0S@y1++CcQoKT{(2>1z^$FZ6jfN=7lLE1(5IK!T->qR1Anq(?Q2(OG{ryd-lndq
z%CKx=a#T;4{m_7zJlILoGtb6tc@2iB!Ao0O(VSAy<&Hel1lZo@OFA>73IOr-UsAIZ
zaIBfqQ>sMUenKk<KRP2PzPi_0-4pYNGVb}?tS|!oYt(yFJe?ZYFx~k(*Kof$Tt7;7
z0*63PP7<G*^IbIzo?Xi5r_Gpb8i+%m0jLtKb-qiwAl*elgW<`@Ar+>HB$V_UJ{$f%
zQSGUkKAP?%+8%UE3Fmf^&GCP47yeM8`OF=TV?N75MWQhbrh?w~Pu2r7TQnzzvR@d@
zA|rvK=es*rUU_OFB;^?;y25Z`<E)PHlc6gdsCu@Y2C`A;KbYpPsr~nl)@e=%{0l$0
zfQv20pgBFJ@$OlQpXT)5(h|aw1bq+t(pA~A!Qu0_pF<A(AX1a@ymXTKymT)_Pmzm$
z9u`!D>6joEpwhcJK~&}2YG1wE@?xp{9EEAH?5k2+x4N{tSyN3L^_SE$1DMivy6L)m
z9`DrM?`Af5re5fK#^y4CsA<a1opH_7+@0lod2M4fILClRchpWmS~lvdy=PtJ)!zPa
zLdX1Nw9D$_fDb;T+O3?q_4ax5BWm>XF_>Nh%z5|<FJNY!HzOi)_TEUIIA=W6S-IC_
zJxHEsO~fktgMH}bQhm5O<=#yRW_g!}elK}iBuuJU^}!(X2NhLTP6j@r!L#EF1hrak
z|8FKze!vA22T}`J(?t^w<+4CM>OV7!36%ddx+t;g%Xet2$G8l*Py-sT53@N-a{dFK
zCBg@B4I-nb2RY+!y=`^d&!XbBsBgak2VI1>=72NIazOoYg;3(ByI_~pBkJIGrR$g^
zn&s5V0d8dW!fQvi4zliQ)XSw4Hqgk}cepud%;wZ&I>G4RqX3O562z&JS^Oy-N^8vA
zJHZQUh2e?)f=7cgQZ`5Tw^ftx@MYdJkt&rQ96rG5&2^UsuJ?C?QDbyQbkF;QFEf0M
zl{(Bo_heszOA>oIZK+1YxDG7AkHHdL(tGA~lU=dB3t`x$^43Z$xLjpnzIFU-lV3iS
zSJO|QPEAWoxUHJ^01({M;^MNhtH<b4x(A&It)p?8jP|wY?#Z#Z!;Czz?v7NGI&rtO
zRKsIn<_R1Uwboc{t&kb6N}c&Lw6IaHdRED^PC0OiwqQzfCoSyQ1mrr_6LQ<#1|Y2q
z4OP2PA$hO)Kub2W=ly6!By3N3szcm6b6O52eZJtSk}~w*g=yyU<-Ez61vZq_!NK7-
zQu_6h4m^_Yp6fG0X#v5K-J<Y9`38=LvwMK`X1|Z}ca6SVCi)~z7s0T5mQt-jV@B_q
zZwxN!AOBn)tQAX*33Kn~^H{@$-0c|00cH6mS#wplNw&g_|M}MzTBMcv5zbj(a|cxz
za;oX9>G_5<O|W2k?sB8cdw=Kn<WLRO%Av?em`4>IX|JGBUE#mfB4HJbtn8Gf7{+R%
zbKx83TQ~4!9u(8cQj7nOt1l0Pa&7+~N{S*K?Us;GqSNWMl&#Q0A(E7}QYrhs&$Nis
zLRp$q){2y}XJ1CtnUGYnHP)D<F+*eqW6bZmpF!vMz5jUM^PabOp8L7)Yx!KC>vP$^
z&q=~Rm6i{f!TGTY4!H+U+lx_8HPI6mxFBY(jZavH5U=A~kq^b;sJUp~(;Jmth!5WH
z$#~`l(pp#m6LliT*h}K1(X)^35~Agqls%(iDCwgzk3#MGWSmQ1CzOE&LpPb@dOzM8
zU!uv^ow3EH<YUjw<$Y>dt{Us5YC37#&2~F2yT8c8p{&d8+0}cM-34u8SJI=Mc5O&H
zpcT^Vua=P1HORGJrid5I6Yjcan9~+4nJ27eXEQUHVPy(BNUGU(QTfnPq$($(TM=l{
zR$KtSV<+pdWlWJm@9Y|M&&z#3*42?_C7=vlsTHj;!zo&LTkH*oPYH#<S@6)sR9S}k
z>n-o{*Ezd^qP46*EX|j^407H8$Y<p}t@H1g9(fxfceXs^7C?2^=!<)!_Cq^ed~oEg
zB1iUc(*c1Ofd_ATRdk#0iQ(M#RdPCh3DZkc!UZ7t-#ETD8kiYTG5&f{dD|SN>d6o0
z{{*VB-BkIgLq5|JE#sx>)zpawB7YOQKw@HR*RJJyO@uI}ECT)#c-gVitSaH9b{8v5
zguJ=S9ZW@&tl#AUKX#6;4QDbBw1`QOD-X}ZzLJ7Z1#k}1U?|XAI&d>rnVFCjE@0;S
zk)j&e`nu>4*5CH1Wp`BnVhCO1ZhLsS^~Y2ZWNqI3Szp;j^u_Dh6&bt9vi$fxEaU@i
z0ymmyja3giq3VN|si)T74qe4MPG4_4HAv?k=Lo~s+1ky${uxZ__GmX>xl16fVQcte
zfjH^?@Y$032UXVGWa?Cz3Uxfmb!Ff@cY@qr*!S^rcF&gOpqZu{4H$b>;$UL%&ifLU
z|BOFb95-^#p5?7!prqVQ(+-vt`&p;J9hbIh$~mr^G?4sfgt(EEh`m;mwlMY?;rVYa
zYT-9Kcl$c)p(H=;c7LzmlGJxEwuxSyKzl-d@q*6@@T5l_of4}4Xt+;pXS11~b1>`=
z(^NyVxaC{;YJxF)k<nYfWutGufqE%?SHWZV$nj@ZOxUr`jted={cL`andcsTW?B0x
zG5M2CMU@YS{*3#^T!Pza)YqCHp(hZ>y0j}OQ0neT5cKHv43D$px2Gmvfn0a0At5C6
z@z0Jo=VrN+vZwt?AudA@k5{ZQ3yXMZ3sRIBR!!q>?5Y&~_%zNCo4@Y-`JS2WHM45;
z10aoumqH?qq4B|6=H3;Db2ILRNo%3G*$AZ)F|_8dzf#v}SXrehG6AGclZ?^~criVA
zFV92vW@4-Q*G-@?W~`M*yE*)UAX_SBn6;VK)9~hUM6B_n*4<<3Yh13@bu4<^gOb(j
zDcc-^ZKgp0^lotEnAUqO4NxB@d^#xeA>UZ023h0m<iT@LlY@&om?+of#uib}05X@>
zY;ht{6|#(`Y`g9RD9UK{*hE=bGl)=(y3;>+{LO@!`pCT^_M5M4KZbSh!(k~adw}#`
zQt%v^ubj(gY4>iwl(tnNrr2wX&cCT%eEOmb5JQBXu&Sv&rCjeBBEvY~0jN>^NRXap
z_$}!N4m(+YqVM@|FbV#WlT2cwHPdYM=y7NMAf~X5WBI`T4YNDeJdF%IRP(`US@1b8
zYp4%A?zAxYTt*JvGg@6lYir0~>=S*K)ITMx9z&bDN$2l<o91AP;!TFNn{2p{sfo$#
zw$og`ZhgqbHXzk04zaVqz(Cya=cZ!d`AqDRwGYrZbom_&Qy){|k|?3mTglXetQ?4~
z3dl31ma)tPC7;IfthP&}=!f~AW_N@~2UniYk&}fVJTdq@1D%ki4@Y%lTkkhed(dS7
zbN*KQkNI7|W&k0QkP;5}bT`e2kRbK|^wfVG<fDr340}q*hfVm11-c;K2VMdD{F}r~
z-e)QI<z}5M-+X!F+lQMeTqisX?D8_*>)ftV`geZBWc@1TYg;u&*$vc3o15_UtC%4!
zrfT>T9@h5;ANIJ0U7l&LU*%U=733ga8rD<wK4zOZyIl+4x5dj<!6|TNQSaS!94{<V
zR=->G5`{xcU)AWwy0t;6An>)yy98q)P2D)3DKEiS79Qi-mq%=nAmZ5YpLjdVbow>{
zm`fBPlvb;g#u>ZtYfK9b!A>9<&ir{R5gIq}$ye$_PCy{N8;Q9ypD%qFIJY0YCzk2(
zk!2+hZX`rN&@UlOLl8`_S^Axyb!Z*p)xBBVD!3IOM5oz3#Or86x1x7<-$3W_gtc}?
zk2muvWowQS?40A|wYCn_S{50LR6BLF?ip2SeML^+?AUampi9)g^#Cdo?+`2e2)T9X
zzI?JDTl(3jYu@GZ)=j!=n(Rw1pM1jMQU>LdU}UEx&hoD$6Vi=#fsuN*wPv$4QL*w|
znUQ;LZQ@@WSBHeYNWEx%r59+~)X-Txs3{*0*ctv-V4wRrEh+9P1A~I8v-gA7X!tpk
z{+AHy57K7q5=pn(v;NST8o<xW$$tIyOI6p76EYpdu4Esg-MD%4CcZRiu0IIELH&FD
z{#co6Fd4>+W#(sJiy&S(341eeH<^gdC`+Q7SUsmmuKg7hX}#lqId3M=BfBP_PAF{R
zbPj^a2)`qYH}M#|>r$&-*eO%^_uA8?HGXy5Ki99yyt=i%GDUk$YoWDGSjDm}_V+e#
zJ{gRPrt1Y2Dg91|+=y52GXJtC%H0a`q5TD0^Y=T}&4!p^X<UGd(u)M}uFc4p=LyS<
zIVw#~w?7;f{}!t39Q=ND(Geo6idYV;VA<1a&BHE$$rxRyoAJYM@7T}g&VsnQi-oIF
zsjcQElha|a!CP`t3@daaL%havbPqtF7&;Cj-5<YGr&_oqJMNF$yMtezQV(Odv^10L
zfY*zy>b^4iWlz3M3YYgNs=|rMd|kFmD#}*|3}xxNa|!+0PjFrUjZ6fIVJhUqRVo5R
z(3^AE5vie#R#&Km+ZoJtM4lUe8@1@&)m%wFTy-s6vlq&JnZ-qyH%{KUDyr-Imte@J
zu)QblfOK!?HL_<(@ks{z<7Bmdv{T@$w{rA(3SB8Y*8BNwxX{~V&w;g$^}HkQ*BFX7
zT_ay=FLA?l<wU+RYy4%G^u{Lp$vd5W1Mk%mZSMIl!$rSnnf8*(6D~Y{L`eS0oqV&e
z<>|SOP&SQQ>CR}lUsF_iB?@M9g-G|D*O1VO4wzi|%SDa7rG_*3<P-f*iK>9);CnkE
zU0YjveiKPtxk_rcjy9(OG`kDGpsc-FT*t)TpR-FlTUd6m1q(?_N=hb)&j;b&phXDv
zCY-`$rrGsWNOxSm`#c>98sF};82tc|#?w`_`iuF<BN5mLlCRqg<Lkj3@#c5Q!V!)&
z)tTCSde>I6XY4a*Ci70D@g_i;);>I#GYCL2TOsk?NFm2@iMrXn>UMpl22LsoBGA%^
z)LpQbuOCLx`~DmdnUW0Z&D(sry&(fF&ywg%6#93sAedZmFZiFNlMLwab_(p&`sFY!
zSN?+7FKX9)y|f`eNm~u=rg&~sC^t70Cs8VngUOc1cfQzjzangnm}gKB8nkh=LbFlZ
zV*Lc0vsNrxT0Cs{gOd;ifV&*2@xd)hqxVflwsx67{gn26khQ%_M=AAPV_I*;9YykH
zx~lo6saVt5L0wtDSN*hyXQIh1F;cpN<KXdm1}u_nug=>?_)jwdky|#~-BfX@FJ4$=
zzM%_5lQ|%JPoNxZl0&rv7)uWS6Q?epm}q>c_F%_Vc{Bb(nB;Vknk@U(ygFhVY>2^-
zG+9)8sqoCR2Jmim6wR<;7-vNhBa)!}x0msZxk(_lnYy)Zf^)n>D#K`J(<{Md=>~Fw
zgO_s=d&Fps5dLg3$wTgbacsjkkZm5QT7`S-t`#=1>-5_9&mRD;5E~iU+)SETmnXa@
z|87kLi0)Frof#fIyX<@s<P&m!j-zCN6_e@x>|GmDYz>ey13tJnFgV(?QsRM6c^YM?
z{WHxsr(fc7-M~aaWMph?EEjBJjK#tG7ma9FUe2C^G4_G489#r!I(X@)58l-LbQ0NH
zU>(&DJWib13ZyQO#J~o+9=qz{rQ)<gt`GocIo8Sw%X_`ec}j<Rm8z0~#L%g4wgIBn
zM@PrjekU6q6)%k-+|YRW;f!?(*Twcp?+-N%n^ND&2?wt_W^o~t8F=o{zBlfJTN}oE
zV!G_&3o2(naY~fySBboTqD<@=Cz>}@N{6unrVRAf6Hq(gtf7JLqWVKZd!F1GJN+|s
z59_eg2Y>=f&j%ca-ZIF2yBN(@(No^%3r60{`t2QI|M00w))}BG-cZkAC$dhYjY7{F
zXZpv~8R*j9#?g2cv0}-fsvOL|hG^hg%ySib%BTLXZ{EBi8EDP+%gvpZYR3zOMq^<z
zgu}<5->BqDq^7;tcKOVC4DQ1Dx1l0aRS!B^$KpM9lC1LYw0t}@S^Ff>D<WIP!Xa1g
z*Xt;HjOdt#2x~Cx*N&~Lu2*(0d9Iq03~T?C{tZgCw>7Y3=&=}~@y(vN&BG6h%JT*P
zHCBN0!?cf~<+P%n-N=7XbjWZX8=nam-C9>h%dc9ILp-a?ucxB)f(XW4^zxHUF6Epy
zX<0v+w`W%9VP_;0U*6psF!RItb-&;%aHS3@J?_XjVKGiXWS~Oqp^-nciTbfc&wbbu
zWq|y${}5?7J2q{Nq<O})SjwGU?0t18C$iFAhzoJW3Lb*EYMm3PzJ#l$Lw~5piBJ7D
zkjZ$+)zp>7DZf6t?a<R}H#B1C^8UR_r86gc|CXf4OULaC)tiXlMF>_lzc60x1O}g8
z^FrxZVi!4~`izhkQSG0^*6mI!-WF|y<MpUWyHQ0TZYnosNo>z_b{ZvkOEPJ6uY6$W
zkQaM0Yzh?S+b{Ks5-#orowJyIcN7#9xU2Jo4Wt0A9d~7{_X-%MYr5J$Z!uMXKkm$c
z8C1U<za>dfIga*aT^TvDeh(S?AkeTlR@Cf^>53&+iKXZJnAc+F%q^sz0XM^0Tz)Jg
z&&BT>fOqwn{4sIF-w{Rm&7IBYU)a~w-H%gok$?HtRic4u9#c(mI=?YstMSw53NR9S
zS)2Ltbrnywuk3!<y7howp^wZ5IseiXUuKz9ELNhbJ)OA_eHH1;Z2s!S9>-IzOEy#`
zeKx_6JaP3+!zWizCeAxwtvgj1^P<i8@#eOOK~?JUdxNS(3Yqe&M7TQdl3I$goQP^_
zg&p@Kl+q`XE2APcG^g59<f0VPI$hKnez7|a#1{2HqT6z2wiDb&`ylHQWPAsP`R}u1
z7;6hU{Le@G(<PF{&AM51i#5<6_N<Lo2^f!{V)6d|f}4@HPi+;~!|P&;5Md=QzsCfI
zlEk02T?TF=<b?TBlnQ1Wy*k^p5%y2ys>jZa^2_2pZl{{s(T`~4wW=J_%hLg)QJS%7
z4@F-6=%;E9wOyKF&L`hNGIvvYF{zd|LkGb&>*0Mj{a2K^v0{lzw-Wc&QOw4j43nil
zhtc7lJ$=V{*1DZM?&#QdsfhL12S4c=th>#1hh}BAdCW?G=YZLoBXPsNb-zlsr7o}j
z_u-PbSj>+0FH@2Zp8WN&rgkK@=<&-Wy0OsNC+xv0Kr`K?`hzr~#(qeKnrY1DU?<1A
zJq5BPW{XArcDgAbU*pZs2Sr_h|IJ%ueZ{4Otv^cysMOhIdhr3B{FAxo3T((QnH9Gw
z2d=S%h}_BT<SF3XXh0va*=gcUFM)!K@+&-bN~|$`gwt!k|6wkNy24J>)tsr8jN7%m
zzC69xRcQMxb4VctBs*VIo6(AFSPSu8sC;CiNi(_Fo_lI~I$GVaDf3!PtawybX#Uh2
zFQ+(VpM;|5$Az}`c42udV(~V&Ke%S0>G7QwasLdehUjITW%nTy+Oy{u!E5Zi40U3<
z&4`=)^yB2P>{kzWQo;`wTUC}REftmm1ldhZ-f;AqL&HcYe~RQpBuBUK3`lsHbiId0
zh?FDA^8yU3p>_V{;q5FrhCu^jVknD>{E=fkw@JMLS#90A)`m6m)o6Gu%nx9RSC^l#
zIv(pQ@A2ZrHb?FYK|Wln7N{+ddZM4UMZY}5H+&?cKr_tsOOBq%l`bA``gf#l?~NFx
zQ#(cu7RR1E#BW;XB*Z$_=oefy!-P(G7k|xc!Nre9MA>dNex(Y6Wwt>!5y6TLoC6<z
zZnBBG`AAV#r8rv7{p#DHU7-~&wAnqGNuZhS$xvdXc0u$+4V*bF!^q6iWhi%vR&`#B
zv3UA&E9+)4r4gE_M1RgWnTq_HG8VejR>bbl58P;$eekU$A>!V?0v<dx^Kkmi!Kncf
z#2ht`0=ypt|6308xdcPYIrqg0UH$=+p#Ybwi?Ky*p0InWYAXslhKIr0o<YGN1d>Wt
zh6vV1+Gx{TS<1tvK{Klk<~;eOu@k{%?ODd3cE6!R>4dkcXCY1o;dW0;i^oUbv$pZ$
z$mFv8FE6$=WC5nc0!v(f293P*=5y+0K4@(G8A|TIA=QH@4jaWx&OPoEcx}G*)Q+4q
zz68PI6^`-py|Mhk@ynQLxy?7>v~-8}s-&R6QF_r8k}HW(+yhizR~EP1uKUK~&r*r>
zu`rR)@?Rw){#N(cP3_VH=91E(Fr^{lX!7W!WW367*^0;a+>$))ohv7P!SX47U#3hM
z@7b7{PEQdXpP`0{aHblLyLO=-fs6oBk9q2}8LO!{UIkyK%jtMFvE9SNJit#U1uy0~
z$wwUs${j`d)_#M;#2H$XSj)1l(!m;<flo(Aeaq?Rd}den=!OTfe038KZ>fx^(224X
zTB%CJbG1$vOK(ru)dmMOId~0IfgS;`RM1(Ft6NT)fFP(V$l%~_L&iR(Tr_g?QOsfn
zj*y?5{Xz}pvINSZzUICUp1Ut69C|nb5?gk2(*dd*<K;L1V_diIqEfrXm(h0<xy_P;
zhxa}&zWhkl1lCtq5Bm2G75BjtMGg6koQ5*n>OG%#3OcA-L^8}6>P}s%o~l?BY4|#a
zDq`DxqOR4F*w(r|UfAbJE-fFNysU~Y2L<YXPi{}jnVhA&Uf9zmT5><8t4Ry5R5wl>
zK((-Y3b`B|_u3W;?dML{1C|t1dty<J_<#w;_{;34u#k-dv)hwQA2M|Jga*Od;5kM0
zV{fBzM0#7Z1>-E%{c2C!(Z{}0Kh*EEY@HgYAnS*5f!OrScdw9WF~pV9V}oE93nFG_
z`+rUj0c!Zgq(KUolPO+Oh0+lu(oAz^+|Pa=&m4r)Xdwdmw9{*qmko}=_6IwFUNN@_
z2?>pABXWx%ApxXpGFQBH5eVP-lqd8HG1-D%6(@UNr8(peM!+A3VJz#7l*3)?W&QPb
z6rZevYfiay()!9@KpJw@yU^sFPOuTmPTxey%eU?RkmpJ~P(`|4c_Ch|-T!_tK3|n-
zESO%j!m+LDzDY*<Qp#|h+Gp#cL2c(Q^9L|0>*|vUOQB0)!H}Z($v-XC#^2{pv=qJJ
zmY#Zvu9~3xqSsFgU;6M&@T}Jn`pXThoZ`#uXb6?Q*V8B3+b#uhM!yVEINrOD^i7NJ
z>t1rJhf!SY*bjfmr}Ptx(i}G9RRy`Ed_@mm=P4n@XmU>Sel#xHejxeTa&v;;4h{VJ
zJmGaufal~6gfHQ3s%v_RrO#x;@cPufpEH+OSp;afHnHy&MS#5(1PNVQs6C5RclJ|o
zq-MuKN{7!h-3qBD;Lk2w1EX<^ewD=RE0f6ZVW%X|QZYVqj;?`Lci92~g0S_FWBJ7N
zj_$R>I6bM>_|efDH9xnC>;-%>8BTr#h++X?XM1q{rbCVsCK-w3XW*!M9cR25e2Ra^
zT<cRTJ=0@S#8>!*m_U;WxAO|#nC<Lx<*&jPH3xOX?mw_@lYo!8+29+s&*KM=Mv9+)
z{^o8~u5)>aIyAhPDkVP9hVJiYx-MaN*WSL!y?ox4t#OAx<XAJs!Rt*inPpey+V0!)
z@Dn>|<M&&xcgBkworR{VSW_pbXO6h%(1?3DoxgsX7E6<=vUiXEuZiL?Ipyfi6@5QW
z^amrtDryU+6C0-$Y*n3#g@xU({_JsU^CY?I*q76oFT1!^>NX+G^ntp<=CDKAIZPB#
z7-ac-dV1dV-$LLZR5ZF>ASk%9n{>e7MH}hlfqzv}%TIE|#l#>6Y9kG#$U<|F>Yx~Y
zxg;QqL7|}&G`Al{>n2jX?AJQMxBx#in?nRRq`!hho7Y1ld5-~b-RyPWvP@Z|H>W>F
zMWSSbb>CY<qE${xMX_;znKYOs7_tE=SaW0-y(FjC@5_5OXLbi>_NSN@Zas@2jelRv
zosnTF#~pz7zBN5C@&i1BC>rNA>HOMFcRnPow&dSZop<o^p3KJnHE~3?f`>9f*~#0%
zY9(0b^mr#$WQm)p+-UTi_ym-OlUa?Pn$0l!7N$R%mv`1{zs)P1aROpw{U5WF-{ns2
zUC&{jf+`9%`83%tzp%1rwy08?Xi4W)+*>IhPP3EyXyQi3dox@Q<}2?lecYqt3Zg*%
zB_YVPoJ^+4h1Xn!`sd7XnLpk-1e2y99;?q{($0~r*=b)%AD+MMUp>P7hycI?ubDq*
z9Ma83*Fv1GAidPm>zIxajYt+6k}A8QB^5AhF5(nyS6e%&nB`qevm?~T)?`709#m$u
zZizb_aiju~A#W<+QrN^eooNrf1_I(N>ev{GyNBhQD~dt(r}3}8(6D$dGD7&+6L!bY
zkbPK16Kj36J=FKUgM5Baq-N=T@nrCGO=^Q}6VEeXJm|oWwhT5$C3lw1){Su88JNUB
zn&aKHZz7EoD)d5FY3^QCSG|Ce_{G(pUipDNYg%@uxOdyfi#P~-T=>Mt(VatASi3rj
zkAFz20Tfe^y;vm4m+F|8ty~P=)vZ`=&CNzno2O4+svM>n;0UiAyhCa}K9%5zxP5y<
zEBjTUSy{zC|H`S?W7md+{!-Lg>a`vTe@z52p7qvVEcBXqRB}HI?`?DHvLBA%H9gq5
z2uCIEBYGwW$HW9|ys^!{cW4aFyD_l&_uEV6OP~+;yMBM(`ejLq`K0s4O+g#48}EHY
zlg2+_$NV%YGVYeQO_ABLq_~wboN_V6E$s3=?ZmPb+x}&(#y`{+P~4cHt5+9RIkva(
z-*fC7j=VrD#w<P-?GRkrU+xpAPFoYztg?=An;1-;5-v6~?i;(F%8@B|aT_@=lek<W
z6BPqxLBhWcYpc51IAlx;4#Y_>l_5o<YSS$!u>S8WyHk$%36(A)BVpiKRD1X(mebDr
zz>mLPyH<23zN}{g4C($(&McI@-pk6C*&0tJ{+Nhv!z54>W{IlVNu;0nw6afO7omHp
zjj4ZNwTTJ4QhYhLx?e{6P7Nit?%engeE6u9=JVA#w(;dXI7K<WB~L}b+t<GX95UKr
z%<*8KG7wtyh0>o?dtvNBkm-B-i4%rWJ&RXo;R>BmNK8EBy76@Rj2dfmoHf1eh~mIl
ze+3?Anc#ZWQKXx6m*1=5$akgixO&D;frKKymIi8}YDw6!>DfYy`bNLH@|PT$b)BaU
z)jT7PH*PGx$e~OBh}ybg^l-G5bJD3Xhi`eA>6^aOTz4=JN}FCxanYyVK2l=ybi5>u
z^I2uI!7gyN#x;#jRZSqLH)!nHv!_8@oXzyBOkkX+v^DXGm<r*an7KS}c>s+XL6xva
zDDy2nueC)<xvrdE)Yb<Okljd!9SJYAO81zC=6YleBLU%~M@B;Q8kE3Isd)^4CuRTA
zbNlW?nPwcK|8D7%Uvj`cSpq!XVN*q;eJXQgan<a(Rh&;j92z$=667O#Tuhr}7stDA
z3kMoR$8NA}boju)I8&ANizdzo1h``zE=7U5f3~z<wEU$3gvM97*;>oNK?N)>z!7LK
zM5FWzN;ThidX#7~E@3CXG=pENCZIwy>Oxr$!WKGp>b0vsUX1GxZ#!vu_0j>^@CRO9
zFb*Nx?DL+MRthC!GwXb0uI*T`do8OvDV`d3%((B1zBMVsH{17|(&ub}tV>|5LZl9C
zJRG(yE5jw@fIOYpeZ3&i43M~O6yFX?u4cqD9_R^$=d4yDu!K-G1x;X@*u^?b9ta|g
z&Gw8mREClAc?P9{xUUb(pFK5!Mc>Sx;ntn~x<bv&+!>F;_$F=R*D7ZdnF&0Es{iNn
z49uH{h@lBVJzPmKPtTtzDR!Iyp(=abBR#4~v(|651wxy{v_iSR#y_!EUnAW|!CeQy
z1$f*4)aF(nfynmkDr}yTCBwNyRnSHd=arYOrHuFkccLdIJ%{|%$=qR<r2R^fBSMe;
zcT+X0fZd6sZUR`-`pOQc`&a#q`3hW1hVyE7>^q7-y)ovG;fKduh&C;hnMcLDWXe2`
ziW@RydhZoI{Cbi>a5=>Po_&0eB(M=8ql+^9XLZ_YNYa&dV5}meQS8+m`Q_o_43@I3
zY9W2XiNkeszxt_^wMgq}@vDJgMs$fh6eX_y3b6d6NcSM|Idr4F*A~zdfl16sVQkCS
zDlUp=Cv$&Hl#!E80pqAi`YRjSb={@T%pzNIdhL$N+QjHp0ZZK6cdV-jc{`G_?@iMC
zHgKw8o3-7r*#B#j{#()2k4RZBs}2(78WU6wJoqciG@w}ATx$7yV5Rd?38*lz=$!@g
zJ{9L-q2xh!Ja}0EM<AcG4LpHmR1!_lnFYJ{B1mURi$bIbz+4ea6$bNHtUsqPb4*ET
z`S?c`B!O#1*G2|zRNmU)chXNO%`bItZT=toLPRM}$I~e)#`(Hys~>&ZNjJTuL<sx7
zmqIU76Nuv~n<ccql1>8NP@n;Q&^~O6O5E0F@nyu-z4!~0)_py$VO1Rd>`+i>u(nIi
z+9qMHfsl8KYqq*b?L=>xgV(ua(pk&0K~?pnJy{zU$B7zQ79SvRjQ73~-+9<kpRbnf
ztd`KwlY-B$ps{xEXRQz=HMuamtY53>O?@!0{#8`Wy*SZa*AjSn#ypkzgw^m%<AK!q
z=2_*WCm%pF<%rC{ub<*ZChWW7<~niYSRXjmR;KE0wQoo^+IU6c_`@{v-x%*f`-n8;
zMZA@I7&83Rc`G}2!<74Pq0T@AXk3JgZYUeeg_OTGh#qh~BF#?seg7*y+^Ads&u^1g
z8qaA)Wr%dzl=VDeqsgolFw64p^Z3naZ@0VFo>N*muC~s<iSk$YIT4+AEWL~PjwQiS
zyIvD-bJ_(vRQ^(;_*oYPre5j3{y*se{DUg)0m{S)`h-`h){~WcPe!|``U@##(E1;b
zsH&)_WG^;x-`Hv_LzOP-8LZa7>8@X=q+D7#>lI$T&7B=EeRt3}p6Q1V(=IVl3Yd*K
z^d?1hsiO4tXhvDpB8?47A4)Ka1iJ3^+7_!@XcPWbVnOyYZ&{&SeNZBo-aMe4&{!4t
zt^xEMQo5;J7?RFB^L|-H&xKvPqbJWS)eH}_7@Z{A!{9y7iCv*u+Eq`Qe}!(i-K64W
zU$pZed+H_Q1ct<`C?{o)gz8F3+z@B#U;v)%lB?rHF)`XijvdPIv9(n({3=f54J!Ts
zoaSJ^TXG~NA`@~fk=t0nagD4H3FvJ3yTJqoFy1t?Rm|*$!X7Y+;ct)109@%L`fBM>
ziR*NFfc6gJKIf6xeGTp;9RZ_~1Ujdb*$80&H>qy{pM@OLMDuOW9WA2gw{GoFo&1-F
z=$|rTH;*@}_A)D@OlkSW4+mV^$1kMQ)CE>HPkii#gVdF*^e&`!!n>nVJx+OZsl8GK
zEhy_+G3zEh=!9bFfEK>)kxP?3BSbS>JKJJKVF&Z}RVgbzH^p!Xi<LXWg)P3{EeAsT
zbM4TS$4gzG=zT-+L1<VHS!7j)gP#z-`!j;4uZ_yb?G@T&FDjGq!96HMoY|H0QtD6}
z3q$XRTNP@WN7N9C0zjzdR8C}9KVqlyR<twJ;n&>kQi|9MJdOaD*7%6JviuDwyelNK
z1i;ZaS?0J$8EYoFMH%0Jzlhow1|UqeEnNSi6=~W<#d~u*`_4)ItF|j<pR;s&>X_Ou
z*2e=xSXMP5HJjP-`&!9TS|JYZ`o610C3H4WsE-xt=(rxIO?=_|@htPw<$m^ci%ZVj
z3fqi$#%z#sc20_g!~a4n(wY*?FYL$~508<y67msJBpR2Ip7@SX1}pz^Km1I*$zICP
zIfc~O#`LNT4bghp;J>Mxkt0CoV<?3$%ZZQ+j|1r83Yp!CiihnzOf^)+{yDk#tN83;
z8EV3V!&|Npae2ZP84eR~-r&PF<eeg>k%iR&xX)n?`@$?qR6LLcLa{TWV}7%6B5;c5
z4%#QacF9Od-j2N~%Rh-m(DnxGN=WF}%h<8aObYIG$-5!6SvYmX1QANto^G-gZjy)>
z|Nj1ErQ~z*FfntAs$&v6&sOKwCwF3H4CkwZjkw|6;H4!Qt_0O^3*G4{n1VIswDvpu
zLb+~0@B^J>lj6%1@E<YP1EX8ja;Md9CAjs_9I4{xj@Xk1Idn>{pOBs%rTZd9^Tqi1
zDzo?71dcXy9y4qGxDQ&Lm^ty^rW$G$Ipf68L(z{ZgfjY5=mBBVcb{nA;Lv1V%?vuI
z8NNKQc>m+R6`xo;-~uzo*-*-I&-Bm)@Xk6)lP^;?%lJvNJE9fDBi<XJebh6>CqjAv
zW#<)e5%$^n*MdWlNCsPYLfJ1vVu|2sL7Y?qTuWdBglt$rhMIM6Q{hR#K%SUwyJK1=
z?FZCG&qZ-ZDj5M#UvOeJl?rMEQ5+6r{Rx{chMXjaAh_g+W6nyU%Lm5Q(wIPoI!f9O
zL{nAPv6gYQ;mtcr*A_JZH12o*+cZc-Y;>-)XwY6dY*ChyPRP1LcQkS&iO<}mt6s2p
zcRqMJf^&*{-QB08sw>f+J(JLBg_HIQ4fRFElyzD&<1=T*)s7|GwM}-<GS>yn=1a#1
z&gQ!hd8rpsj^YcXOJ@SuLBmI+GfHDmx_gtdR#P}kf67F2o(mq=k?!F5e8R~9bqPBH
z^LOWvt0^SPkY!6vgq420U^6_#mT)4JBG?(reSulBGmtZ^(*vqjFN=Fdj;bC21_lg=
zg05?j)Pmc~{m{1dO+qt$XwZtQiyOHor3!fgOj8_Tfk-GIG={HL4kA)!OmJ)Br+(8i
z9Tb5@c9ns*<}E_`y4z|a3jMiDUHEX(jb&pytf7a70b(7U`-{sple;jvg5-?aGe21B
zNsoPW7b`wL=j|<F17PY9^aV>)iZdqrfN?GR^|2{*GwgU&wBlTP?&~8h=GEIcTvx@Z
z$0O=Jq$19&-V#b~+W|`NAL`?3x0pe|s+0SAk8L=6jXq|QS!UZ>zsgH;t>MuDDgItD
zThWW8k+|AiTl@4K&dyT!ZCfg(V(b*AzVm{C$S>+paoxaBC)bY?_dNSCP85u_Q}1GA
zG&AGutf*n1n9QR=5{D_aLfyh!7M=BzKQRg99a#3Vvf!)_hz&qaKeGH!$eAh^yR%k1
zWM_(htHS6|iA1?BUAGPUOk})pZW(sm07*|@+p(0C{ll&D&Rt$F%=OB;5Nn2q=0{FU
zTQxmjIM!~Mo<#x5)DVxUw#s*R0t><@tzESG+H<zn+YS3Pu0xbQe7&~O(@4OkuP>c#
zU52K-R);QCAQ}qochM&oUUK|9TBi);^Jm7JPYP|$;DRV-jkKAR(yy15c;WFGJJ)jE
znF|{Yf;aB>*49?0Pz5F5`2`MN-UR_Ug>38iUX-oB2?{=U_a#u;4qa25l5Dcq`R<TR
z(z+*J(Zj%Z+qv`8rGIsEimSEt-lRD=LeGLFYW|hBHyEx7$HMT4KU8dBi)=*GX);M(
z>dt!(QJO=kA^_g?XQ(H5GmpI2<P$j(Bq3%#TM4|DK!evP_%<BCCj!5Mw-k#Ze6`Dw
z!v<%()n37Fz2MTV+Ho;ZdLwQ$7eJW=UwPoR6y7E$^j8Gg39{Qz5&b<Q45gv$O^M6z
z7z;g1|6n0>Me>=?)FoY)eS7|-RBWo3E3ln?|In3Kc+YGD?yR@=yTG$D)GvOL6B(BW
z<Vtng|7fxg8~gwvU_oVf_w|wH*{!xD!KxIaKDv0632~1=`kV!QGrO^OG<`>($K+Sw
zg4vYV%yPz`OS?m*(7P#UBMfxqJZu7mi(uPiXv3(G5aWd5OQJ_C#NRl-Z@61DBEAxz
z0%f8gxX0jZ-v{!Ev`pL?<6)5~E$S?w-jVWu|GsL9($f6Pb8b@zoXj@$k-vaKfKgLq
zF_&78Hiz!kUA*F0gs;4rCAizhvQ=YZMT>9_DNmXkJ;zY7Lg{k3F-s2R**K|vbgyl!
zO*og^Sp5dS9{3cI4YP3S2Ba=>FDgkCuW5g_m_l{@$9;B`S*!{KL6foDKh>%WSKQ^p
z<(+c(1;ec@Ta-Z@zgXe9MZ1c$qT|W996r@#3Dl3*6;Z5RigOy?_~9$N+y}of%qKPI
z(LF|#14454+%lgpqyKv27_#BhSm{R0@aV4MjEsb4b?ZeXZJe<N?v$=9%<_)Luib6Z
z_X?HB%0qe@8o09~oy4re7?Xe@C2qnuB3}QFjukK}@WBk=A?zQ%l8n=eLi9@m14O14
z%R9-5`^ie3O7n$~OZ!Le{A?|!q7yE&@2fESPG>V@Ck{JPV0gOO3Z)DFUNNka4O%aN
zdIl?Ue=U1!^j%-4|5;n{Tb{8Elz6f2xl)&^K;!eq>6I|GFo$bt5o8yie4d(*W~ooO
zv6XJnfAU|}1qrC3z>}4Xk~lLI^_*lVm1Mn?HqHT{(M*M#7!0M{Ll5)XwH{`t<C@4o
z2RoM3+f2ck3Z=U)(k?_r%-7P3bm9`|V>{9s*wk;kPDTFwkaHeaa<iD0MeiSAflR}?
zqvcZH4i73&b__4lPHr=IC@9oEGH$zv%ltL-MSUH$YM<ss^gXiaf<3d_8`K`LB1o<s
zEI(G&4o;K$440(Sx>^8+uRN@BV+}FoO!2HKIFVejIWltT`P=OR$^PT0A@!Dfg@zWK
zU=WDAUs2V}HSKG_`f~has5Y2OKt!5~@o}XR+k1h0QHDZ`?QLexqnR){Df<mzzU|)E
zKd;5L(d@yk0JtcJ31oVCe*x<IshkD-Cs71+G>|AWeWgF-qc)Nw4kuXZ0r>U;C>)`a
ztC*7iqS@m6{{;TP`ao_1G6DyY<zr@Y1=#sGtu1|L3HcNlk(Q_3b^Vh{0!rV-?A`4=
zNH);VxLmL-bmK9~^#nNKzsA)Dd~8)mTI!P9Rkmdnp>ZgY4ec{u86U$KYBR^#j^pF~
zv7K#;amT9JHHkegCf&)Vd#UBwB=eP;S&ysmO>LERq2bk%0kiwoSE%|d#~qLfpMKD&
z0QEF+@TiI75CHqVmaeVykH2zz#!g7kzueUR0Jb4rRQ52NS@xWlbU%`~ZI?`<(YC}M
zQk6Ze5pvhpj&P<02+@TAsoEY4gjD=iyTEl(+n_`sIfHYC4sS`vuDM)A&%{a7Y$wzl
zP{b`PSf%R&skSFzqrEcLR#JjAqqn9d5;#IcTuly-WRyIZac3+jX|N>LY|BvCPSp62
zhxH6ZKlUpHF;E=i8EHztyn*fvO&q<I?TL0z?tEEMxk5dia8jr+eOzg>nJK6cLZIh=
zR>3GX-LrlrSto^BLJOu09e2&}FUAp8jm-$)dM{4|H8ECy=qkQ>lRE>Pt4UksDt8_q
zx=*b~L0~Dy9b-9deaxd`#m3?}Qz~3tfgR+2tH6f3wnV7EB*~1v)J~@h6TA+-8i4L4
zDL8<KlWpf)0wUJ|#P*vdPd<lf5=3Ig5gh0d2n1QAcL<V74&Fe}H4-e+)$Wqe)i;~i
zk)GNLsmz-<h79Mojb-@{fOC{zkgX*G*7G4|Ry!>`>@l|KyeP^B9LmBQJ#koz2oTtc
zEtx-#1kI)t<va>#@6Xp?pRVT>R9V&0DWR}>QZ7f|(&7s#y~mV){>n2qhaWn_UObaP
zCl_ZFhaI7^Ss^n=7{W&B?JCAuepR7zuf8LBdZSH@>>gEt9z>oj5Z*#sQ3LIK_XB77
zaii4`v_2CN-*NwN5Y~wl9GU_%ojPYhuS?SWU`fmLSVW|4>hzNvk2ES9+jQ>%C1)4L
zt0l1mgbbnZfzPY#6#2f$jDKR9OoOZq$N{D@)U^rC#KF?@A<Kuno{-yq1>BSReJeBN
z!!5$;BBy{l5VZ_ftq%YgY*NcT{vfz={=VAlwxz!NQ|$1PD8t8j^mL%J8A(%tvr9O<
zA;JDw8e!}}kG+kJjX_A<`xIaX{EQzT*~9scsoEBCxP@5I$7@(3Y=bi2GyZ&N@&|HM
z5a*KiSAV{<QhGoD^(Atx#NAEg{E98CcO{tK!{BNEd&K~TS~J%31(3GMP@$&;W0z|r
z_rh?w&^_q2e?k4OV|Y*WsX70N(+f{?I4;!b*$z!PKtj+4!HdWjAb8;`9~VIYMxv&z
z=-Z&Gw=#dg``kt|XX}q>tj7^<GnwHP;#y$^oN(LS+$sz7;|RL(`wDLVdmh-Pf47GR
zKtU7YfYg9fkWiR#N7|uiWh%!AV}r_xAERMZMXBU|(a2ed$IEx9<EZzT{ZMfPjPg%}
z%ozl>ZWoo;Hc8?!eP8v?I5F<eKQgY{9h2OM#5@d~YW+)vYUKZUI-J2d^vpebG+PdG
zn>$zXF&z(KMeh@fWBV8wWCU0z^b3P!ocnoFFce*KFu5<27k2?vbqHKd2_EE7x_{83
zyKLRL(75W|G%JIu)8#rKA5vAPy%AU3HpiHqY=VAFkR^e>LR2XzJs*6~+14j$nkX&j
zM}x=UErJ&)NdU9_G!!(CKksZF{eV%r>VEGy)$!mfZ`j5N82XH~%_Y7iFJbK80Tooj
zPUy7f{g~dn5Y#y{)_12Px)ijK(YiG_1Q3;b9!ZH%Xpk~7c|{j*1y;o&RhA2`vj;_E
z2CuzJXQ0LOxXv-G<8<L1lou@IkY>L73~-%fEZ{D$mW<dg`JocK*3dVj2JmiYgfG<-
zLA?OUK=v%eoaX{VtX&TD*)S4GSV@2N(k%dHm<gcXFl=eX@W@;jiopD%K1}W;S$CG~
zI%|)$R}NdoT&Y7eQ_2h}KO#03o}A(@WC~5i(Dbu{z&57+GxUp+lb!q|TKWMmI_&6B
z$KQ-x2k7dz<T-?eka-ft8oOEIxMN3xW}tlz<*K{1);R9b0n#_298iCg)_5~J^jOFC
z2H7d49=^W}ds6w|Pnz#me5zi3@k|bPf;8JxdR}8mlF9j?gdEepe|=8+p>N+Z4zrv1
z{)ZDW!+JnYm)e9us}AIjcY~iq?83003&q@NeFI-9;=B(hWL`uq8mAN|+Lw0n;n)E8
zCSq+`?Sgzx^@;*90<n(ViYU;P5<LJH3tNN(twTOlSbv~Oi*)2{50}Lq;d;jAAljXG
z(@7^@nE>r9>>GAC!~Ote6TU`hwoqmr#%U$KOy<>E{%PX&-;AuaRaFKK0Q8@jnu1s-
z6pZ5oLgo$>&h0sNk2)=IWW@Rms*HT3c;gWy7}4@@26&9`W*i%McLS%fhZl`?XmngT
z#*a&_2YN>(<B-Uey3x^!WSb-CwKIWPG5;F(_^*P}rzOLhp@#3;_=1(%JapK7d;)TT
zq~LE9cwVTm!i@*e=pHmXN!elCq|>D{Y%pCYZ7Mw<v=o(;c&)cfqv3X7%Oha%bprWb
zjUU7_z++CnPQG_#QFUKl_F{y5fVbB=j0Kt0<SXn|pdiq@GHozP$>vg0!<iDP!$n<u
zGs$<faC-k+ssN`ItviQEJ*ewY$tv7qEUZ&4vBEBA+ms+16~hgnyM(=krHs?6>KlCn
z<r&O<%RcG|$$zKatdrG7YL$~kdC4gljOZ~7{fNAu3AtE21bQDYY<BT~VQyv#$X|&n
z4b)t1mr)gl1wbL9R4RY9%YY1AXp;=Nf0jo(BeNEz3K+NOzJ*OZY`&uuid*&C4XzM(
z-t7EK;)-}wg|nVl;Yqk7Q#!=IT-B9!EUWf(h908GNDS7+1QQJz_oZqOqZS%0m?d+-
zxgK=jB8GhiF4YA~B-@j`5-~<xleT~*G}j=;+$#8Q%RBsokXau@1fK6}7>_Yq5Q+61
z3B>mOiOm&^4+h|kr~ZSz^4ZoA7Nopat~jw^06KRT|1G)i4$S$QI#LA<>6X(gKJq#p
zX%o}2odvi%_CYoWWvIL!PWCRW0S5T&_(+V=P<nnIea6+N^3LU46r-=A^iRCVj0n!-
zRBOD!2`ee#S3o&aB%QG@&d7BY>H@AwpNC>V13irnZ8i|kbv9RfZoL7g8qPSV<)i5-
z*xFPO#OVhVlGG#pEt$GyHk#Ydo#YxU#W9RCn(ucH(2hS(G4X0D6SZ7+ZPBQP`3_Bi
zI0VOeHN7--uH2jomzKB47cqxq+wR6wkF&UY3KABd-LTE5?Igw@g$QIRhdtt$?%ua#
zez5YVyJmxV;gm8~djE(aLyTvN3F56F$NI7h7Y*IF&9o2j%(KhpPpa7}L^((EOHoQ3
z@|HV+=is^aV*CVl%$2)JshsnzkW`eOUqXo3i8y~CsmUd;6;sEDD3|}548^<>YCMO&
z=mlg!E6s;6gHZ;;>64YHhr`d?ln;08vah`KY1?Hjcdt0WQ8)~d7HA{>gSyYPPV;&y
z;YEYO;);ZG-1Wk+Nd<7@%E!w7`0w2^+lk$xDgDbzIK|WoX=LsbLTrzUXP?K3@UHwE
zl(ai3C!_TImkcNw(;_1=r*A#K#e`5ErxW?EU%z%C6Q$^7Y6!s(Y-y=5xb%wXxyOw~
zbq9}kJf7zX#B2C8{^d+8e;ulxtzDa}z&Cu8Y=I+OTseObQqQ8&h%A~?eY3Nm^YsK_
z3>B~#_cNciTx@h7^QBFcB?lkJ1t?=dVkGiO7-OKx8;%06d3&DYN6^nl4&D-a_X3nL
zi5COVS@A2E%tL0@7kytJy?~Mx(AC#}+P3ZpoL@m&b*(x2KSxBNjSF}t-WGbLSa#BB
zk>`$GHA5_BmqPO0{dt22%=X(wz#olJNSr%^dWR5i0b>T4dN+Cfp_Cfp$87AuGzQ>P
zACp&yfp3PM58x%<%4bD`jtlBCb<wf@mh3G3sq7hs!Ik^2p4)EAEiU{X?VrMCdAofY
z;qlNzdGz=YZj>LU(D}+&Lk>EihDBSwju>7q2ksYVW-=7WzziDTJD|uin0%wWOv=b4
zW2j#*Fi;Zgg;etV#2uWxx>PiG`=cPo>i>T6%v6LD9<8LT2g&DXB2axm;fluqWIE<e
zR!NHSjKqnou-$dPdnP?Bg*wu6;aCCWnS3Mv4@i}>;Sp6{Y<n1ll?}e}2F8TPS)91<
zWt3xq-8!W>w1LvH3;S4?p2`T&3JaPDvAnfkwQ0e!2XR35B}5On4ui?y<YcpMVgNl7
zz9zi|H-K_n)IHNIos&FgYvElf*qsL92;g@d08r_b!%FNJ7;$XXHPiHt^Yp8sP0=eq
zdjNDE9+p8$ThQ?ig{a5C2!RbLU+ErXcUWp@+SHMTRc`IW{>D@{+Q+3d7Gtez17J^Z
zA_)uox*C1h{RW_x1Sw4EO^j6?3BrN;*KobixO3Z*u^yM23BY!f9N-n=$Fnn0XmxGk
zkNL}{RVALkM-wg2R0qZukk#jf2+fcWoUqClZH*@%Isvpx4L_x~<w%nrgqZY=bM{~&
zSmnZ2D8GRW2VnUy(#XUzLjKRmrO?&05NjPfj6tLZyp~&rtN_;>w?iPl1E|bU)xhbs
z_@Z1v7Gx+Py$AURBi$-HQAJb0PN?%SAH1-!P<uMYsYXS~N15o>wcLPy8>C0L3s-$6
zKl^M79MDs|BXNao#5x1CLx%ZNb%8j65SE;_{d%}fKLgetG#S)Rxm;U3cW+Bnqa$#*
z7C=mS`=MvJB0@F!!0^6$%UFH_#`}^z4J6C^!B2J`bR>fhnJCw>2OJX;uhgwD_6mlA
zrg~YXPgtNL*EPc2=ncKnWeLY92*Pb+adq?p(-+~8r7f@A<LltSl#w)(0$B>guI?MP
zwi>&Gf<!2Fp^FFM_S5Sa%b&Y$+7&={Xzli)#%2>~ZIlp1z!t-hJ0bU)KgV{f0qi5S
zklbu_L;YQkre&&m*$v{S-D(f#i?#%A%{Ms7B8I;^ZX+YZxKj{e(625$^ZjLMmI;fl
z*7O}>Q!y+GTA0xSm0|$p-)c)nJObGzcIahcX2aaWpRgKR2QT3PjvuKd2;owNUU2)E
z=HJ!B@yOAY&YUsR44wT@TR${Kp3w6uv#aPv8$E^ipmQhCK^H(Y5|na|KZo)T)}+^L
zn^p?AYtiLe=RGiHVcIJ>xQHBAQwP+F(gS>itAAk(_uvZpL=aa>EDst}nzSjlX#}sM
z;*Cg;y`^~}*TL!8DP2HX4UdmP(FhvY`O~y+`8dq_R0raEJ>dEOLC?Pwpj`<f!YZ?%
zg*v9SU+Ettx(4ynup7zefgP(ftb7e*LvQ*?DsgtI2m1zT8A|^oR*CdJ`4o2N;`nay
z|9U}>O*|GEl0-oUY@4K+NnZ`n8c=Zws!w^sl2T<sm5l23fN3B{C?*Eu17X%i8_hC^
zvkQ6_{U$2rKMU}e^*=-<nzz5_8{$iyezC2tt}N!c`JI~=SG@qac$J#Embfdm>=--U
zFKy$cp-qC-j${AlaU4WIs{mOzb>d@<+mX;1Y;Rjx5dlfB3U{C%sKjD$G*EHOQ-h|!
z=QQ_Sa3U$Gl1{uCCo0g8bqsv2lY{ND2!$@dM+p5GeXy+ePcun7L1<VfkApnO++Xc=
zwBadx@E`Bup%DWE(8rLuao5aWAZa2kzNlKxn3h2u8%a3;;=iW$)Z`%@bG_2GVCHTJ
zzUP*Q?hv7p=T#H2HO%iX^4(ZuE??%yA;(((wrrJ1G-eLX{N@q?u#i#EA~g471Q}+F
z>Ve`gG1u^j)&82mj7!$j7vn6|q)5PG*WjbRbTgtM`8%|vVip<mcba~Q7W#5@KSb=!
zLk4q4d;{<<<*PX-tjy1~#(R`taT10ypuc9eWWnr7{7vJdW(wkRHuAVZ7`Rcpj9(3U
zDF@&#M%jYpXwKXUeBL>+p;~Lzx&4ca9bHvT2j?bk2v|O(X;=K>D5&u#qBNuTL_3$S
zGs;)q5(0{dG|u{#&sZLh<O6Vkmhcb<3lA$Sl%S&MGXZ0BqW>GB%YGw1DqIj|NT}r8
zMo@I#o$E=$?jQA=R}|P{Kyn*83%U#+e>~x;(qxLR+~YO?Wql_?kd<THXM8o%3xpzS
zN)ruKSY&xW8f;MoO3@A@y($;An}`JgeL2TQ9eI)OoO#=T+f-8)1I@$(TdsAZ7Y;ax
zHf=NjIc{uecwwnSxt_b_B0`((GX88im_hy960pH_#4p!dR4sFekNT^m6H6mb<lwUg
zlj&-U<K})!Zz#$FpfAW{o%weV_5bv2(WtrM{H{Plvr)S=%}C^LUwH|Yy*|9(4Eh1{
zNKs^@EiM<z%9vCU2LczqYTBn6488}P+H)QYC@P0Mdf#=N*(z{d;u0o&UwsWGT1~NT
zR`fGHY=yk|ilG#CuE3g?j+LiM9z!J#pWXca)pTpCnl5WAFM#3tMi=7#wuOJH9(jp+
zaY2iZddT24@Xdj!_W5}k3YQIDUo5<cx^i<s;A=?#LE|?d#O6)v$igJ%f44$nT0H_&
zm;mC{KD~BR_Lu2HdUL5d90<h9Zn?NN*VwWAm#4;JEK%8y$(kkxZ-6YXHC|#7&B*Nx
z@?YI#*wA=$bQ&SWAtd~qZgxDFvj+x+{`T>sCI|`?D|hI2*bx9@VZeeRp{s+$GmyAK
zAq0A?W$UGv94;2dndvx7-u=%8kq`GS(-3#Y-{bci`pnm>c@`^tj@0-KX$^p}kr1L8
zMcwt4_je?naAiQ2iqq=|<Ny{Vm9lWbL(GH*Zj&KMfAsOf3t5cQRC=Bgfofsbc5K4s
z>N-p2+01IoVx742QL^90g>S@?@$g^wiQbMywhZT9M_bVjAX$O+-xGc&rp6E3o^Q$5
zp({5Kyq)Hj*7)2rw#&%mpt+XS5|jB19_N;hWDdY*U2H_NVn7lIv8yKFHXRUJl$@CO
z0*iE<=hPsIkiLIIAcSay^7TL?uMDzV_)3}kdZYAxv~7qcP^W}*>~qITh=Q7QGh%}`
z^xiqaJ2V^&UqUC1V6gBIjLib!z)brgCUzcEqQgX4IzXa66=INQ{Po~$PP3a-^X>NH
zs*_+AHmYZ0n*$POzqdsq1q%fIE|+8X*?+Lv8MCjY8R55C;j~YIz0p{deG7F1x~a2v
zRm<dOr`>>)%zUw}BLaQL;1n$T9_0(mftYanf^qavRf<zj;X)t6a%hp3g~=EQfG>~E
z7l;FqudOU!e%$Z@2pd~k#uvVl8CI7#NjKptuk_OnQHUh#*z|~4<T9qGe1tk}guY<F
zWD43FYb*05K>YQ9$53GRT{Nbl+9x4@x}e7b&N6hA&INn8;QwK1GxTc{#l)Z;pa){)
zubE25dYZN=xaz)@W^>>HLEoFQ$qvi=%U9aP@NXL$$1^9?IaOLwG|FCH^;3G$!J+;k
zS`W*&dNwV$YhILOQjTUM3&Fk+lvrr<CXTT1Pq^N_7*-B8_uIk^Z-4%vlg|lEk%rB}
z>ql%3f7ZL@6B9ZtE6B_bmn(j8yd%zL3sT*-h7Wf3K;A=tx+?mw&g;JEzVf3bbI0QC
z(8Dn95)iUjNPh`}Em)rk??wRs529|A)WQjUb4C!`{(S)}aZ#@xE{=mn7LY95%~4|=
zL?;Z^`n1v2$E>YugJI;l%YoHkxt79v>BDUR>;F=`dVJcUu%r@#f#aCpLQbjVS+v(e
zYd6_RBjl$R%vDB#YT4f2ZG2Se?BpI0*qz$6_Z%*@!dy=?i<o&%jm=Pg6x4T~_gf*7
z|FD$8hs%XzL~F~d!i$%XFF`q|Ui`9N%Vrd3Zws7}z(w6{1MTMCp9}U=Z;jGYjg%(T
z6rh%JZ0E|E58#yTRB_cDOAyxsGqpzMH?pIp<_P>{q!(ARo#zLn%&`bQC=vd>)bEJm
zY<x|%QfeoZ1RAxQZF$)s$OG+!(52c@Z2r)|tN)Fi*=uz8NANY+%hW%`l-ChD<#B4^
zLYN8t7i>WF1+ONbwMxr10BA-PK93U%ehr*_I!YzFcmSQZ=^~t-B|7m4&b1pv=^<K#
zP_cQ$gJ6zzCOvgq=YiVA#EwES$-4=ruHAzW1u_~V$(0HMam_*S4SB2A&Ai<WL(?L$
z6vo?B0P-{ldA7_I1l$_J5SaS(#vEA(ZWY;Z!TcC)59h<At&Udi=mnqUkJ-TP(dD>d
z%aaO?NR%<N*kl7#eo>w^_d6pT(N@P#bGGA-HBG?8AX(V3fPmKc>M9KOxsUKiNCcDG
z9s<sGXyKP_eF)er%5s{g@=pOu@HdnTmZ)N{_k>1E%M^f94EhW6+egqx86xJ%YB#%(
z{{$GB5P=<E>^Zd7ECc{g<1&}TSuFJc6e1~j4UF4b_FDxw;!rzJqATqk_J!+F4(}n2
zzCiY8b~$VZE@1iZk9$aI@rgnL2+#y41uuavMM;RB=6)>pTcNRAVeaU~Y(+Pf@MGH?
zgjro4n7DiSoCo5E=FtT}lVJ4xq+Vd*^w8$H1Qnh!*H%=~su8vvz#xoh`5RIrZw9eQ
zOj9A+Fc+CM#d6HkdYirq1<o;=6$TsZX1uVmBKunS>lfgz-iK+vA5o(=sAiq1oGrxi
z8hqq}g#@8<h528A!&$ccCnhC?{(k1GUfQXN?};toYEclZA%X#%3q}r{JBA~uq927~
zY^S*chqfv8-EX|47^LXtwodTbRuG=T6i84h9k6B``QZ;fDoih&{t?LgLj*4|4L(AG
z@!oPQvxYlE1*{XNe%`iD<^%%clzIe_!a%jmT+k}sX@!oXq`Lqv)ob2r<9bv7hm~Jp
zk_PD1>YN2(Pp>AZFB0LT+yq`La9q1)W%ez&a9pNBIDM|x3guNxC);Il5LCm<Bd{1-
z^#>O=JZhiNA|?QP1IbidOac%HP6(2Qi%Qi0&*205FysVfxq!*A;N-@EE(rQjqR)OC
z7ORD`=Wz^!e&#~@iU#0KQqjhnD5WNvAZ=50JJ4;JTQ>1Een1$|Nf8En>!A*T;8g^_
z&Sf}e`-gl%G4?;#FADmr!X5-)_eauj*k%-?^gHs0K%4@SF`U15aKuTdKC;Wtkrz(P
z+Z6cCX7cl2IY2!c?BX3>I)|P}A{}e=LK9_WS;|(t`pY>GV8OPfI8!hVWJy6-1Kl{Z
z2>mhp{1EP7BQ*E?@KN(JQ4Xce67Oum%o*Bzk-Kem0PUgrHg83pL{%ZNkKi2d2H$(W
z=BD8u7926LmKBQ82pja1S;&ifAI@Q4lx5%-Jl3|x>sexH7Xm<fS1$dEo~F9*xBY#5
zJO^J9g*ujMI}2>$`SB1EiVTCU;XWb7Vu<en#NCJ~WQw7Z?R)}U{pb8)1j?UzHP0Y-
zDdp-yg%*Y9?Mm}y4tR*Ct>_Lz!wc-0!h#LkB#SALKt=$~08L(SE1|drZ)J4^MdWjb
zqI3WMIkk*z3<!bNB*ZhFJHLPy0|#swelwC`0V2d)jm5)i$k+G(I}s%S5{Y^@c~}A#
z1Na`!#u9sZC?(h%98;K65rO_PHe>#Fg2xVp8N`)#(MSonZwIvXLERw|J1$<Pg5chG
zN-*d{k&PC<SP3R<aC+_HcUg)z0PumTK*I#V%evJRe-1N?AW_9#`5*s||BRKiUh|kQ
zbMpi<ptN}r#W2L0QS=K&@BB{b{eN4T3Vk4A#0OtTy=-z|d*V)7f=X)=Sk77P`40vs
z*Gfnmwsy#;Y|m-}Kdu)r^UG(N|6>bzY=(FZk);?6ZRUzDY6ayYo8Jw|_sM?6S}b3O
zARJ0m5Is+IEzM^)yq!%UP{QA9gIr<+#sO&&R&2u3b`(MX=K@DRM;ILk3YQp%4?!0`
zUUUAtrYq}sl>1U#wFy?g0Wcx^ni8Ml<Aob8p^lDD_t2Oj!mwxVnGnGWuKh_NqNZ9g
zVoDguaioBFnvN9UY3%~M!uQh>@!3n5rlB#wcj8dhL6OhPs14C%?!Dl9ly%6_W85!l
zG8|9dlG=~u5<pskL$)6qytHnvA9V-<3&{FlUNbyP_E<&cmlq>2B=Hgt3x&EvClOgZ
zl6UEf&GtxyLub^NqBn-=AIgTYQ3JB93aR~ieDNS=7M{WoQR_}_d1p^hCG+1cxKwi<
z<c6B%05ZSLt4Cl08*AntuOvnM1W)c7JB;3ft-$H>n+vLYf2Oy=kAA9#9#HnH(T*wf
z)jIpxX{lp~%_#0Uh1AjHPP2K;)l3Amc{i<uz$@HC8fSKhSJgux2m1X8tVfKW=jtox
z%$T~_e2j7%)m2z>&}9S*E;B~kS?%!$n1DDyF0?8pzIX~fKdgsj0u{^W@XJ@WDE@_+
zy6|{vm~s{ePr(?j^WoO9Bi88PS+)Thi7Mel80jQDAKs|TdGPC*U<iRBuUg5|;9T^5
z9|ZjTcR*rAt7_rM<ciw$2&CS+7W+JlM;`zKj#P`_s=5_U8GfbyaKt}6vC(yQB?$Up
zD~f{R2m@dKdY&XYaQxmaNWJ1*(iZiDZueA3)t-6u8_FY4go4apv+!ipd0WGIZvj88
zem|Il-nX`4T5xOUoN8*tkBRBs$jn7t;E7RTObV|*Ks@>Ue6ln5EI1Dr^AB^OE_Pvr
z`xZkKL<Y%dv@#$6_Nn$5aTT`{RBR7#5zDLBUTU^`9CrLh<QpcmTY>$O64NGWpueVZ
z*lJ%GHx8mkg1-5@^#k}3@PCYu0mqJg2g7hs*Xw?O7LgBtUyQBlhtGiKd}aj8sxic!
z(*+nz?-s}+XXY?1unRg_#}oS3W61Vh0Bxd@+EOwbX;av2L)eoO6Ker|`|4E)VN7WR
zpOIs^ZV4u8qx{vYJS9@qeUgi(B!140h%3&(D7V+DQ5c~|MdT50h-y6E!hkd_DPXWd
zNz)gjq&@2?@%Hm$LOV6pEAaoZ_2uzUuJ8Y&Jt-+#MQO7XjuvDYg*HWGN!AuE_9bK;
z6>(b75h}}QRm3RSmm!r9Dlr%gStboLSqEcge%JksI_L9!{q8?Hee^N&+|P4A*YdvJ
z*ZW#F<14q!TsAIEF>GKCN_9k`+*vtO;g*AQxfTwhy<KssD9(KqwAjM`->-P_qgl>2
zJWvXr$EM+*tPhydj+c%?7`>YjTE2M9_`!#zq3|9Rq7m7k+9CFB9bg)?e=SK!+E468
zp|UbGQed@e-T9?|Ue_fz>0UUTp5DJn1QW%$ohK0eTWW|tR-Cfn=`K4s?h!RxfnqYe
z0MB|AjJ0?Kygsq5wctF;9kEgsPV>6>F&OnzkBY7{XBSn5{#cD_rvLB5^gqt)ZI|qg
z1KNs!lg4&V+T>W$XCK>=-q@!$4fk6c9V`1N6Py=%QiSOVfgL;jYH@chjr*=R?rj{5
zGO&%phblTD+5E!cp;2$B7y7sTz<;eDxh(Y5yCtwKdZ^c@<fH-POD<jy84=k$g{~gR
z=^+}30e&-#)f@A-iIOGUyNpDP;mPeZx1i75VqE8{Mn*#SktZoOdnERNGX}5K^(%uy
ziF^z(v1zimHqHU2;2q6q%8ddh%ig7i*L8oLU0%p>X@Qu7H8FkDiO$~??RS6Q12>mz
zHT@$>uz6W%WwCG4UJVXC;&S%nY!%#L*R|V@cA13y0Qnb7y-(7UbVWg6UF{<9*0LSb
zi2-qF0AQ?bg+Iva0F3Yr87QE#BV}In`WBp@ibk0>FmjxeT@4XKYUZ`koHt*4o|-G=
znh%CHv$U7E9|tR03C#d0CkKe=rd(cN>pKf(!Xt)eeo>7blxRC=m-mG8CaIn0pfl`#
zAFeY_(1wURe5!QeN;_9zM!*eW!-m$gMJyT_-t)C*;UZfx12(DqPEb|&Z>G`tSt5-6
z*KElc>;U*|D4OxZSl+R&Y(>6?fD5p&(y63Fk`m|m$E^W+R`)!&v3e*{D1>u}jEC=4
zMpd#9tLE#9$5SR=pY5u_PUtc*BUP_PWFs|Y*6==yq?mr~uDn?}>5((4_REX8Xvy?m
zq~k5~J-+Q4Ts<z_4RXzrdKLvkxd3PASjxqr)Iu<nfS%l&_Imy-A)HiApf+Dd=6g@i
z*>%Pp${tXOyN!}*2o$032d2TI4};ZNXOe<|#G>|gc)I~`+8iyeUl5rNdN2jw7*q4i
za^WC5RnDEYK6?fpm~lvuTR*>wv~Og;=b@XwUM9Iu>PJP#yrcm5wCyP($9Bi3AK-<Z
zavPantut9{lh->n$*ono<%o<)0SQtlaKa`y6z)YtG&(ls`v=n=_J~fTM%Wz7vn7#K
z>8eP`<5i)*?Pc~kBqBD5muC1(r(lc%i1e$UAi4pnTR@cq=ET03T`h15kYGn32@m<-
zP7rt37Q^U^qHWT9gSud!$;bu45}-EW0C!YhSesQ8_-di39dG#Kl^QKTeq>*~Au#Du
zb3VGRRg5%zMkJ0Opcer42^Y>qX^RZhQuPD8g2#^{gUX1aDoqYW()>v#AJr|ll@jQb
zMus$hBapbzP>=M~N-1+Drnh$j{t;rS^sj>wVo)rDtn|p&DR1y@H$jl^aLz3bl*!Or
z(dY<pJCf_{`BOU&@elISOc8%=kh(-?5w6Syi>0%S`u(;vnu=4qm(0es-JvOgp@7FD
z8{?AC9HxQ{cy>gM*{*9<6z+zUX=pc-8WW-msc1x3uI!Y`Wn}AGx}np+$->$c?_qi;
zst<dBSpeEzo08&+-i>cX)Ff)nS_`9%b+93GAd1lTBP$FVz~<WDvZ$CSG-}ZU*o^fR
zgf^*!(vJ`a=&bgJf*@9o+W*cq$YS?JtzaLY+*^q4mJDNj1S@1C49bF{v!387r9d)_
zPNF;LF~ZLgGig~S9O;vP0^KI3O09Y8@g%%i0>TKF+;mrg4qwW;6Uvi?c|yHNEK+|)
z7SR+}Ngw1*yc(Upy87_e!TE4Xk&*ehHS8ZiM1idYID>R9ajL71Hi~-^O+7p8&$&Y(
za{{v`dn%!8p#Z`?QZk8x&wR(4{MsuCcvN0|-re#Z!1~2sqC^%#FBn}4ge|aa@EQsd
z+EK+nKE=b^b!MNq@fYwmF~)4l6qEL3!7t_kUr$rK>w)747HZr@Zoi6Q2(LL{VxEV~
zdoW4A{PG?Krnh!TZ~FD*?EKcPP&B<4ARn?ZTI#%Zt=58QD#(J9pNpA;69_r*292J0
zwP-U?<WqHq4>4QPHqGq-07(N+a|{AIr}*4DH_KC)4l0(wbEfKBLLH)ioQR^lJZNg=
zZ<m31@MgqtCI=z;Vnt(~7~}WI?@cr|Vwsm7_Hdv8CU{uzuUSfRC*(pF;M(qbLuLbP
zz3Y;ZJMRUy_hNCh@uIG?AqQRsP5-^JV7}061J1SJj=vl;29=~b+BjCoG_!WC0uk*9
zIDyc50==M3P{H_!$j8(M-{4Z{1*JzsfV*+7Q&j{#&U~;9t-<Z@ZGwY^zf0(UX#<T^
z7}uGdJxJ&0a=((HHw7Ojevj!x<V7LhX=;4Z2+}z)HW0Z*%CbhTfT@SR(gN!yQr&M%
zyR2x2me%_8`20wSJHUOyr9HnoI-Mh%TiNi`1l(K*6~iFpqM~MD?TQs=TIc<vlnSBD
z`O39Nc-JQT<}xk>l&|Sr+KA>(NQgadi@~{V&t;H4b)98^F97Ey7YYhvDENYM&d6YD
z9c%h%OM>`r+l=6;_R0f~<0<4(VrKG15Z)v+IMDF}qqM<<myEw4=B3I-Tfd+Zz+TXJ
zBS6uG$o3@`S0Nz`tn^cGz(4hkA#=<hHVcfmwvFiSqIFkiPssFGw2d&l0%bs8o8=HY
z%Mt7J_k*T&2^5Q?_O8(Yu+HeEf*9#YFmN6~so0BfZ%88vm4YMefE~oVEL|p4Gyp9D
z@Eaj2SyGS;hUnhicLr_(Uaxq^6N9krSnT<0m4KlK*afi<Y+YwGq6p}NB>xxmsBPn4
zr)UC!G*^oYBWoKb{U7`po)8iqzht5n8HOgsv+;Fsq|g<cW!t(%QVe!2YzFJSyormp
zrMCb^yZi}=u_o~c!v$^EW1K)<2<L8!;`!tNWVoLfiXVjS2+o&(=~0KpK((S0gnOVh
zH3Y-DfJ1Glnwmgg?$ChX0Az(5H=_1{k`2IB#g-aQPWCQTIFW8rLTy_EG{Q-s9Byw3
zQw;#M>4Hbiqv~H7D{(l&7A{zukP2X_iT)UlM-t`vf_rwcT4C-}C%J9yBlCT6qB4W$
zr8P`w!*6Bo?mrn1mh)qrw={fLo38LC7=yTQ%W=OXkS`hRw?OEa+B~Fdjd9kz^(K>z
zp7?=xytIRr_pe*e4-<dsgiNZzN5Hm=02qhI(>~JqejiH+m!I^gd0hRuI7IRx6`K;R
zcTqpLP?xfHhg{bBWaR`CN&WQ#^Cl~vwK{vPd0Z>4IaIa>Ebj`B$w2@+d!?T35wxQj
z-<%P@4rmyE3*j&hH(-arCz;bI1Y~b>w-A&;pm9fuDvGjL&8t{XK?HP~ZUcTyKwQ8R
z-v_m&8r0-q39&C&1!g&Rhv9S@3O7ULRy{;PU?n9Ly0F_X_4N<xhm#47sZpAHd8X-e
z5>ATEWFWttKj;~oeA#8+n|Pz#y{#(~@S$lzphTZ4c4t=2K;^hU8Ro)M7_g74r5(<!
zp=@`RGToV{0wECGC(de&<pHH22)fIHFnG-eF=?bM1nf7Am{ir{Ew`u|t22TQoVnK9
zDY2PrzkEUN7{q5VR*gA5hU{Zt1h7oP7gAtbT~-;HV`exUPj4y+AbN;A3Z5v&C;Vde
z<|T9vb(FdZZ^=bD(?-Wss?Oj?=_7->!dqBx8{765f^5NO-{O7<)!Lx)DH;8S$%h_(
z_|CFH35>%CM6sxX^H(@nfd~m_1>QmF3<zX(jUdr1VC=ubI^dtBBY%#Hq;6|a+fKr7
zPu!U_lajWXA;g?fltSS=nQf&}w#&>@>!7cewXc@_d>vcpnK}nxJM`v-B;l>#MCm{P
zltC3L>mK2xI--sj#4UFi8G`eoZ_Rs61EF`2v0DD7<^mAsBBHd>h}DRR96&?7>=sIV
zyN@Ea8W@o6VBzGJ-1*df+~EPgs>0P*HMZ|X7(5@_QI!A6yxWTIaLQW<H){s*RM>0L
zPf%L9fw-LbS_C_2%v9bk`L6+?_^J4|jU^302yPi4g{(<HMMWP0D@|$L1q6x4*Vz1E
zX7aX$#P5rsh(;&X+3^WC-;Z;LCBqoy1)E&rVboJ8X=U2%kjnEmhwg&~Dyt}B0a*aZ
zVk0A8E;J0`a1%R_^E{>vd%m;pgEZZ*d`u7%Kgw<UbmwZ}^xv0?xFz8DJw-?8jwMim
zzxBK%8Ns3#oOxv_xvhcIDu<UDcFfauN*|Pdw0nBP99)_0ld;BB@RV?^uC6HP*W!Zc
zpSy+JJxwX{G8iBv`|1yz4rCEX1(O4w`SxQZIta0UiWU9Zf$bDp+rFRxpUK;|m9;2x
zGkbAViBdGXbgAv0r9a0=Mqeu%*jg*{O4fZHoIFphBDV8<xsy1ZNlS~979F7wD2u(_
z1-AA`2it!QuFRV=rL_@sp}=k|i?2PJHSQV94lAnJN_A&?dfDnsR={bt_heEB>x&m}
zA@a5FO`8kL5?~D59<g}jK5DeQ^_*u{|3&>Yx2*lidhVI@r<Dmr7`#|!)jhb@$+iaz
z&+;Mc^2z0<x~s%Oe+BN?Q9=KKzH#DyJprb1sS&l+LA7kvld*HmJ2N|#e+ey%@lWJC
zRNa{Nt(><3?ARp>PUcV-;?%UwtVr<mbl@q$ZZy9#Y;eEs74uAw216q34s%dzWai#9
z7=CgH?E$kXfg7q54wqNAPL7WkNgXNWjZw=y0cwk<X4!3rFhKqVJxm{cZ+?-W?Gx%&
z^#$i!!jX}yy?ADf69s_^xQ;0e(Vjc|T&kd+=%&r<+bNqNewfG4X5%d#pY~rRO+H%g
zx7M4o$Kc13_3f`Db;7cKm^S37tB1J%S}1vC|N74-8w=OX9%IDm{3WC+zFciR>+BAZ
z_j4j2s?9$5FxJT|^|guVHSYcU2?m+?b06}q(EB^jDXvyZSj^d>!vUt(3OFmxr+~`u
zq4#cRXh{CM$_Z<Un5r5<a}E#{Wu<EW^b}+Tc!qb4hb@QPI=rrM*az+yc5B;iA(+-Q
z3T96^T=p#iNrdzSem*sBR&`Ez@BAMfsEERRp>mi_sluE@--u&9obzp^n%3%R&GQY<
z_byt9@0FTkU-^`NT+APRX7h%I{x`>YbTG1ke~a%g6^WC8yzN}J-wH8f5{;iZ{g`IM
zNyi6voAF2yHD#kPfi5n-(n5(oLdj^O_APzF;RDGX+9J=5bOHAjDqLO{lR#r1ghWK<
z)q6&b8UrK%Ep$*o^@K<bTUomQ76gjw02u64me7A+cLu82+fhe57*l3Zd_ah9o}zBC
z8M_M&ewBG7A}5`w;#^`#efXV;6wN#W{gl<^_<<P+<SuZ7De*O*x8hWfOI7A(%2#48
ztoab|MZcEQ9$nsf6G^{J){BYJ>3p7z-aGfp!=pd>Bnpex&xnaH?~^tn+47yo-SB#P
z#Q|&@fAVJ{m)8)JPKZi4L>+-9+9*V&uE@r=$oX2RlrfTi?$u>4G%8JtbgpHl=#}zM
zqCNCSQ-NoN3(RPgZ<rJl!ZjbVCA5lS+*jb5hv;^VrM7=_4t6-d2p5thlbA@ooM)R-
zv_LiwGOwNNoz!d|wa<rizH!N9QwJ0tg7q+evvxX#v{;dLXDm`n8z-+|5mDN^7~CNk
zKA)TKzpDY5@)YMOyfjtWLwJRS*#nOM3H7k#hThB5--jCUzLavh4RnDAiw#ka<qpET
zwhAc1-4Q=C(d=S);*CFp{!?{9kDzkFGnm)XC6+bEJ|V8gk^K}h?xhR2(&H4)Ue;xB
z0F}QRaAUiy8@=%R!ym2Bfl&2m+#VQ}Ze8=uHzfw1>JM`cNjOpO<5eOwKzb8A7d}(_
zTtQ%fxN*{Fi^oG;>`=5wZ%mUu%xj{?UoDdp;WFC^e0uF5k1D-Uadx@>(1jJo?Z?_&
z4&|}@W!2|CYw$se#815K$uxYGR!!v-+G7a1>`>(rTgQTG&%Lixf`q^jo2(-EX(xNZ
z*7LdP8!E;wqk>Qq3jrnUoxUl>c4<-(nY*A+ruAsNp})U>dvZ#td5*hh0a*~M50HC3
zTRv)4n8ACSqFaETKO%Ye1FZ-<DuH`mChn^!dR=0&UVZMX<Ezx$wI7~^ZEneky7~Rf
zHug5@9Z97>_4Gc`Nml<>R0UTBbLOn#YT5^`tSj^nYUrAUv&VyL!42e!ugJ`)`OJI+
zl?g8+??rhG?D#j;Ko(xDRh!y~=5|=>?R1@7nMdXM;d;Jcr9HJOtp#PfniAl{e`x@>
z(Rh+b?n&G@_ZnP?w3l0qyaDw^OW?!{{4iNY@tO-dF%Tl2Y$OFIwV5|>pL+)aB}+ZM
z$6wTxlz_RVZ~UpcfL}1JKq&QDh{F+gNkvH?(36yw2I$H>ogM2hodpfFw<6T=e0JO8
z-#bG89k}wlkl`L-a?mX&%?E!qDxqGLZjhHe$9`>SjPLPm%W*)%n+;&_#DL{TcW&O%
z3;DU*>@Ku5E<L&Wsx4W8MZ8hSMrUc`ckVjw+otp;rK^nLA(KT)XJ_LE!p!09fi4ex
zS(;;kwfT2;pLo;l1zkN>$Hb$~_4d+vKloG4sw;1*l7$twLe?R&$!%7h^iV}k%~_+4
z^m(}Tsr$P*5RAFzaez3r6M)etc9x1XDa>R6cyPTU2I>hjXA12pq+-h@pikr-+)yIk
zi4HQ(r?RANc9wKgP3`<8O?Y465Hc^8mJ808gIYnFV*~j5$Otwdr-X2eH|bqYxL}uD
z2y_a<aq>@P>KEk*S`<4OFg`vOH!|>(Ed<vkzA5wP)FqYsntw-}$mrl$OQP?ZS!Z*s
z&AS6Hp#L=x|Ne9X5iT(8fZBbl93})YetM>wI@1XS^$?k}RB$%Ca+<bmv*0uUX%41a
zh1k4GrgL>tl;us0nDnqk+v`!TeJlvfV=VAzaQ1mnFN8f$+!H?ng-D<s{Ac&r_m{a?
zWu&zveXrTMW(w{A=ZM;UjiIh~(AUs=hqmo$G;)#QS?B$p{?Z%TWQ{gHLT1^jxFc=O
zpnNbN_iLe<dL*tfyVrX=+CbbR*>k84aAI4B<N>{H@orJrR}NgqO!W^RY3K^7sl2tn
zo~S>eHG`_#|64lBVjx6di9rcd6LP)4lK1=EX5l7+P*A}+s-_;_ix4ZR@5~G%5W!8o
znwO`AQjZy<b<Zfh1S{zNaG27$(6w@3NRoL@x0233PC<yVqVe#{1pvX_e=}-u&D(*U
z0!);WVDdjiD}ZmIT7lgYs7@2h;O)pm^H4jp7C<r#Fw`#eKKdb^UPu%SyE?073#{$7
z<IY_^b?{!QoaO?qN~zI0s&|)KqB0V=FGHlNQ3`JVDdt67vt_TxD)cGMn{R($(+ke$
zJF56i0v!qLCMtg6>{XxGd(K$DBdMf4NSc&k{+|iYV5GC4?E?*D#MoW%S-JXZ_cKml
z&P7Jb5L(->3&kaf>b<PrUEVYH`hj%xbAbtLQiri!wb&db_^rt|RJ{35?jyiwR8Zxw
z%(b6QgfX)_6GY*74F84PI<Rd2U7cBooeo0~;d_0K&Y*;FM4fZe&PCmzS<Dn*ah;h9
zw`Q9$d{H--SodJC&ITKne+`9`<H!PEhnG1Ch44E84}P9tBRn1$i5eIB!z?==+#rR6
zRSro9%F@%fr_r|2#G7eLwE2*B7;FH@M%S4YH&rW9VWVd2|24QF+)ZZ}mhHysuNd{M
z^9(OK0Jo%O&s<3)|FY89NnlyQA8?ms{^Fs$=_L(BTMsd<_!o1GQSZPLcTclXxPl!r
z`|?-Wugvh8Z*mhH=*JR?=0e`N_AI8I72cXA-7rNOV{o6uHA!FO7o3PfH%amV%msl&
zZ<T5Kam`hl0a8%Aeh&$FOW=tYWnPMM+e%Jm0KFIt^TbcSonrg}3qC)VrMeqljoiP^
zmjn|ehN#Ny)Kb^o2{$7mkO3QXISSm$qsE#K^r7P;%5J~Thg?6n7q~-udreK1Ord}~
zuQw1{4gP&bxaV2^+{n6(iy^bI+NMHOg1i_P@-j50>bp$O1X4SZal8$HLg{-2S=}J%
z4wv9yv&L$B`@wiL+yv(5lx=rfxRn+ACOAO}nvr71to{7N4r@=oqp=7rjV}qM<F`c<
zZeqcJ;=byYS9&i(L`v_#a8=(8VUFO8l*Hmsb);R$IY|b;crLp;g1bnTWX^&_7>tpG
z$Lf8CLOr0%-WJHM>KwB3qFtwvZAgSMF2j!S4zF)Lk7aTKwCT2~(rBS?=Ee!(CbY1(
z^WCLAHu~0?MqUPtsj8k{81(DP9>N0Be|K+clx>+jA7o;2DX05JUdNXi4*4$!jC85t
zmh`a-U>yAYTx%Oa$PbO^Q=PaX*T(jOi0oZzMUaFs=lJSdinFV6xR$$J-QEFU@Hc+1
ztnKwL0MtQjD1X{3io`D&yyYu)F4D73{Dk4DUX>JU14fnjNyH3n_Je7UJHOw2n`;$2
z>wJc}lZ2QDK#$_DG`iZ1!tN-MgGcby_$i80{QT{6i}5-5-3g+6DUa1Ah8$LDSPtxR
z_t+E{B%OgW13k%WwCc6KlAt@j{9I|JD%bemE&lo2JMd2c2!v;YvAKEEo@SwNEtoM(
zx1Ph3mk;C0*v!H?Q?V_dz-z;boN)gNHdBl-%Dq8K%`_;d-=q%xKx^3N^($V}*!cDr
zf)Qfo@D(984Jec}ZeQ;qycC?D1v<A0-EBL9K3$qh%f3I7O5U%hdHEM%Ix5{R3XHg!
zAJ67au3c+#WhG%M;9c*ylt{7`NLho^?M;@zg#;p(E7r)-+M02EYfQY+|N0+}0|pTw
zf-m6PZde}yMxXl?5E}F1?pQq-_vK>UO4DH~b7PLLXBqHqx8&@HSNgy=-cD%wEl8al
z_7iIUO9C(T+CeTPkN{j~#cOY+>kx`H#YlC_Np{g*72u%}b~pVYM(Yl7gA^tl{GpcI
zJ!nD=E|?n2eH!g0JZdYtDK$ie8q16qH`U%VrNobLC&Rm4n8r)q0k0b(>ik(DT&yp4
zMlMJgH(u*#HomI)&>yk|^8sYmuS-|#<DG4Gt-M-QrLXK~arGU7CO&D+!P&X=hqYXb
zYdVAltf#rUs}CF6)h`*D{g@~s$1X8O74?A|paCHzHbmq#Nb}A!ErX4Od!B_1=U&Yh
z<dz2x9N>$2tp0Iy<k-=Rd>DXg4qdQ+L-F4(&$Av_?ClJ2wMa4GKsNB`a;!_WZsbn`
zB??>i`0Xo-JW9dsq(RLW9Lw)^O8D0hgPPt`<AAQ($1m7={17xui{%<z=@Fj}2!vIv
zTi%0WA89cZ2Z6NF*?;yW$sN0|IM?<R>zhB=+4})EW?#Y{gl-jMG>6_2Rk%RT&gwjK
z(Bp-C`2ak<mGTgSgecv;O1@T}O&5#b+X7N<d;(vN?QFM$ApHFq)x!c$rUVQR|75ou
zh1S7w_9cFyoe*x^3UiPZ5I#KK`^QBw>l|}Z`I%4x&p^3>+BOs{00(=)R#vXR4Ln4k
z8E(y9-sE6ECWAW>`-^)1iTl~Xa)FMFtEIo>vq}eHqhNMt`L+3C(%MifYmKcrkPv-d
z*wohAmNU}NvJ-t~7aDiv=kopSkwehuXsvA*EK5`h`-fcne$NOb{X-rDG>`MPcgDPG
z=G%Z(?i_j6V38CN@%Zs$kY!(WI~iZ2)!=e4Z{u8#_?o-}v|%q3oxDazuBUgvD`F;I
zs*MKW%?khuWT4V2Z3LVNv#AmX6d?S|mlHfW;0WfO?B6l)`kYq-$Pn9;7jCUs_dK@z
z<4OAH=o^!*o}r{;^VGY$fMjJo?-m6GuywIZJ*R?&6KTrL_j$ZA`>AlJQ++p6h9g72
zol4hK9H)-ib4%w2@34xGBHrva+=Dfpw%r$>1?ALY(2l<pJjG~au@WYyy%l7>|M)?k
z-UztR<?(4JSGUxLNhcUyKh&qjb~zRBm3K!<`XZ!Q_T=ADC&cqXXMSJ0;J^8I-amp#
zsAbzvDvqgx#r_!@BMPc(fau0Qv~xW$Xaudtl42Ri`D!f=+$Ss(zoBQW8?6XYX)tBm
zzw1s-q0~un_6No|v^Dpw<;c8D1j0+%q>Qs)h5L?yYh!JTvI912v7WjI2{#3lQg7Vs
zw)3;Tl!!d0tC8>mX=!Obpj+4-KBA6Aqk59-)^QKKOVzEu$aymO;gkWVb_m`+saNEO
z9#DH}xDY6qKOKxHuW|4c#lkrOYyWp^ZEcRd({;GE(%;2=N*ZPCD-?VFb{~GIaIGo_
zWT>3Cqy!$%`JZ=a`lRq8QqAXI`b!H$HjpZdxclnd-p1XZT|ZDDZa9d7PV4I*8@jj6
ztug|q#MpRhxZe61o)~O6H7R*-*l+SQez-W!Y>#0kp;_|4naaon#%N%yC3Z_Oo>xel
z$tqC%ino(doNXPh80L0-DZBg{zng$`5%65^5iU?RZkg7j=ZGpY++~OfnxzD?56Wy2
z=^daU_P_S8eUm-1BxN=yGOB|YXgqEC2C=`|Cc;N8oP_bbPmbL{;kb38a?A7T_!qGc
zds>YW2It2fo41T27p-Xpm~h0sUa0}sr__STqwcvr&@U%utbr69A};I#0EsF0;^nB?
z{kvDbiSIA*DKPBni5X@kS##6}6K;E~R#766MEaUasiF6~&HUOd5j18kS=q8&?gBq=
zcru_rY+Sw|AHOCc#XG}0I2e?OfD@$nao8^xAxHD&!fc>0T{DdMF|C^$`qK@WtDl!b
zZw&wdEzagqpd7^t2$I0r&)>C%hIQOR`~{3mC1Ovq+;`7*bs`yRFaYWSn)~WMProyF
zLG!@w4d_6mIQESkjfdRkd1*kw{FKuj&a-Y|51bfq?`Qhc&QW6ZdW9a->ypQ1Co8J9
z{J1Q3MEe*+Wj(j+3oPSzQ#PxUdzGK2TrQA)N$8l9OQC69{?%FD8$7%rh>wC-8|Hz6
z*}45w6j5MdZTen;-i_`EZ@<;O#RJcWc}30qIWY=@--k{}xm9pcRy+N{{?`D&GT4Fk
zSlv8}Y3QE_1!~AcfGpSd?{k(H^F&e2@6{ZM94Q5fo|I);Wh7Rv4xwgSOZh~dr=SOg
z8t}0!L_Ox@<5suPCM?_0%Mg(paq8!#uZu75xj`ci6ks;<`tjq(Vxx){bUpk0@^f#n
zoNm|3Q$54MykAH}FLTcOhTzZ1z`$)6{gzonzarZ2vq+Y8_hx5F7RT^#6xq9xMfR*4
z*bgoi5^ZoFMX!-dgE9=S7Y1Q{`7sFqUirfZM{KqK&KJ2i8<&-^<~FN-Bk({gl@Og9
z!k6V9lL700Z%y!D-xsop6d&mY*0bCbBrSZbN;27ipdAd2@#^0NeW=V0+ANur;*BF6
zlj9&Oh~C~68eaI<snt5ed&2C`+eS~WS^2QWle18kb&qkShlnP#e6Ikr|M;=Osh+gZ
zeWJI!yH_46KU>b>x_jW`_WHp0c1AzZ83w08L;Z~aw(byG+dourqFB8S)mUhm7$^MK
zUx0-~qZIysR!GH;f3r}f4KOCSXoavgOcNBC&8yL8WB0w1HgP}xi@jfQRyE%FYSD%_
zR!4h-gJ1FOL^mypeZA;w@bm<%6Uu&e8T?#z?eOjq(l_kJrovZ+P+y-cA)&b?Mz$c8
zF|R%uv<(}@-#@n0dxtc2$dD<1W3WqH%gdn{0&Hq919?t6OJP|1Omc6A1t^W&AQX?o
z>SSk_^R$4D0WBnuQ^P9T{gj=$Vo}h2n<nR<1Sq!sgYS_FS$AhvwNiauT^%687SHT^
zON)H1VcaMZiD6z;Glk(tD_V<}uN3-;=+OG#PA|W;@^Jl25t-DbFOI2c$jyy0+Fh^2
z&|3V4QayJ1iH>VJF!!{VeL6}|w5fxXX1JPfw3qWHX$s@ZM)7Z|)0~@Xa`qQQe$uii
zSeNag32%x$=z+OKqB*o3N*3hI2jcsomInMMI7Rrod`5Ug1b@no&oG85Bn|56!-Hy<
z<U}YJcE-_$*-%3B2o16NXM6+>V}k%qMB024{=el=KWSB?&>ncT6%QS@V0QwSWe`{6
z#IJ*Js}xj)(9eE7HTkg&=ekeX5C4tfK$?XMxwby>lj;jCMEH+rtvI@rrMvo4yDGd~
z7{hsS_+e$gqB(bW7+h(A!iV2-!|OJ=JS3AdzV%+9*g-|gfgF^q9e9V=9@12w`Xst-
zc1=7Xc=^f9rzC<rDNl1rCId0CUnQGyq(43B_HhBZDvoqP()_;DWItlSGPtOT6@sV}
zuz~o9pVu!OUyw^<@#zsYA;aEsA2kwcA;E=exX$E?U4SC&j;{r43Idl?&6Ie8HW;l`
z-rh=IRmpyW4fnJ*q5TGcVCG{`br^;*5%OeL*V14>L99B$#NRmOQ4fg4M}hV9(LG7_
zM%k}5uQZMdf{4JPMRMchqEh(R|E$jRT?uu|$)cf(Livlah8%kG)dbW?g<`?BBdb3Z
zvE4^{OXu938}njTc~9}2yLm557H+be15<s^Z*9DxvJ_f9Hu+`Hz$ET{S4gHRB~na2
zPRK+Dx&*z+$gvaNrBtKd5iiusNvKIX{E&X~E~Cc9pgxu6(6Ka<2v;)j#FlkP*rQ|{
zd2Dzdp4s%&z8PA5BmoGBN{-_*`s1g$;=>>&0K~IL#U-%|qRfuA+5!>_fNf>Go!iwa
zi|N1EpQMii%4tD{7X9fTM5~Mrz88pPYrEN$BpRmTWBG)w_zMsEfOdjB)c_P-po*yq
z72i}qMGox>K5D@D1Y>2vfm4GLe^;EbyA)!z$#em(XUog^xJ_6OK_A?nVJtx|h!4Px
z>j}&9FQNH_#!drGiz+X61Yob>K4ta$X5A~fC(}jTr6O*AerE5v`ALYIuw{4(YcN08
z20z^fWyh;k0o`TEd4$KB!RA^&IC64#16eIwR@o^iDP(KQQ(SW91Dh&;Hltoz_0)uO
zWfbxSJsk@?JP;DEld~OA5E+@sDE=rr;^(|j4mTC*W<YWw+?sG%z;=4N{ufr}AQFC<
zESMY^a08XT%WNhY-J1UXdO~D=o1yf=Ev&FNc7SRn@3&@3Q}T54I1~CzOY-|bUZKcd
z^3NXN#e|UJgGCGbS-N>HFx3h*U;TGwk)ii0<ozQVIt`9i={<LqJf~K~%gIsENqie%
z7JIdL<=!$|F#mnYX?vPR97AIst3~6^uRh$-__?ptD!!{~>=0O`-sE%PPwhPm*3Es}
za)8n2I!KXI3uK}8X41=>@}-eQ?m=A3S7UAZOSR|T6*YU$g!djdkt1%XrLD~im<gnh
zSlEb8ED7Igv{A7lH*Y(4`ce>|1EO&peox0dbx<Ct6yvPjU;2Wvw|8WOz*juY-iLS+
zN@&Vd0DrwOgf2PMLDU8l8%%uJ%8LEZPNRt_z&*-)*m5b^3a56d#zEizaNu%=I>uD_
z)Ska$Q)UT9fzY+NF64y4c8|uVrl+ShL))|+24-)zx^uNyP-amMG!&LSz$u%3&yL}*
z+%mF=^IYT#iG_&i%v0K|(^{hj1yX4r{Rb0l><H%0L3tRMeO)2huNyfvZ5K!(VBl)Q
zWIsu)aZ&0x$H)l6(+|`o6VDhNH~t`*3-aE)JpLe_OMo(m%R#D*@+4y?3d*dn-}M^I
zvs2Jw&c9ZUF}Ht>AwMspNO^1qvcLZut-_Zz<Hx3+sg}DY)bB_ltqw`bJ@F%PVD`+m
z1AkOU_u8ammB)7_!zy!cv8MeGK%ilf$(Ckylmx6a4p0U?5nlrLSCPrV@i$;vWtyz~
zH9a*#Gs(w<uVFaLiX+;qtoGDd>|cF^=MD{K@`2yG-|!0K-@d)5^~Q~D;TPbTgrK&;
z3=RDGb`<G|P&8Jew&M^3x#(6$ZVf!Z$6rPuEG?9ud+xZqkD8|6gMeod5qRDZpSMNk
zJAWF_ujd04z%*&)nYA#B6kY0DoX2(FIegAc+XC1MBFw5*Vj&#^8vG9jNxAPy;z5T^
zfUMvM=<HIz1AoO(ZTMan$3Ly%OKP5uFj=!_C`*Bzxl&m$ywRsfEXS^{zP^4VGq$``
zT#eV1)vmw3VA_vu>lPfGAu{_JqgP?!VEc=)dIx2{a~iBim&l-H^L9)*pF5p}p{eYp
zR#^(=J;S_%43aG6XaFf^5?ArwKTCeew)J*Slj~7HU>3daIR<9{*_iCwn=B{u$d*jL
zkuNd^CP%o={3w|}`lNa!D4lV|&efu|zOs8|)W$S2A9TYYfQ0~q0q32#QISF{fR|7B
zH!az;!7)c&2|m9Ku`FOwzd1{NoFoZ_&`xYy!qDV%(!P9#e^=9cTh()CRuww!e3c6p
zoXR3!R)>CD5rXsWg8jtaDG+QIylKBt9csuP<Q3x~@)cw70=0>Bsq1SMDqhHm2mUfq
zo_qpzJo!=0S$Qu+Xy#gpwgG}07!TI$yhj0V&rUf>o7k~;xfN9f#PFLpoPFVvmV85T
zsTdGSf+1|R=#6{(3NwEBBFk7efXB6yEt0pvg)ROv^RT@O?WvBpS)Q+Oy2a+KCAhNq
zLyQ_IbfI2PfU<4>;2~(Mu*SoIQXrYxYr+R|nh!L-9t+|u^FOk;MJ3VN8)4q>9WIZ=
z5M1q;M7K8`uQ+3Mc`)(PO+%z`qe=c4l$J;4F$@nyu@TN#6gX{BtFJtXESZ<l@eS>c
znjK9e+sO}2zHp(*499Fjcs&};)?l|WZTXQ>sh=(o5`u709s-~cWb5(t>RDn;E?ZEE
z7YW}QAQsZG3*@GN61$7NjeLIR8m$_HPuY>EgXTYjhTng$9cnbdw({o>*GFHP6X^-<
zyMM>P&AZZL;ox2aO5jO&lHl&k+!hYB|2<X&WWeXH7j_g_T;J`4s1gGz#eOW{rvgK@
zl>?n(<YBsPa#B)~ew%`l=&mP$Vur~s>JgS=D>t-giEc^#?BRkOfQ6K3Z+ExtWPSlV
z3Jyid&3s>^v8<%zM1t3%o}^j&wR<bKV3dtApDw0W^0?NVBq_RCd;Q!P{A05_qC1^A
zYQ{s}C}3&Jv}IM*ApOP`uvvQ6^G>sOeWT8xk%Qw=sbcX|{`54Bd^2AJA7fES`0CJz
z>dO{NKr~p4szzdsR_CG2^Z$9&ol-c-WYmgWk8(9!;7tgqanDo4Lp7_kJ|_QgvsJT1
z6QvJ<W!@eZ{aD0}w_o`Ehdz~#z-juoH3>_dZ&gRT^2==19F0;NF6!Og+g=jC?=>r}
zgg|(4^B3#F$aVS{sk?XUz7GvnYu1PTQ^QSd>Ik)^e1Qm{Ag|8i6WL8Mxw1yb|NVk3
ztb5zTZ>6u9LZ*%S3QHzH56TX+PJsW4189lb9o4_9q!hcEM8BjFrm$iy&pClYV|epv
zVq!D`pG(EaDn4VcZOMH4a1?JcAu}Wc`V9GN#GopP>HOp!0AmS8OzS4XIZ-{NQvuQi
zB9cIYCXBypB^!vS=zB`*G=X3QufhMU5F>>fulYbJ*o&f|kop%A3GP2KPZC{by)_$a
zu^*;5x-dztsN7W~;xBc>+Y3MIEz#_{8oqcfSe~|uz><@#Mm0J^B@{X=u^gZo`eLx+
z2YqEttEFmlYeYA{voU7eVA8K5M6#^wBvuoJT05WnlTxr4xp+_>Ch)x8b%ka@5`HL>
zT4YOdN?;KDm!a74oO)m9xvyMQqKD3tGbmY1kB4aUx(}zaK~O4!g9Ku5P-Kb!twk2%
zXt`}ST@-G90t_O9@3n_axE)0rcCJ{nFow3nR45TsJ114z<%17v$D9F%_c9u>%9A?6
z3l$%I`4l&pT$E3iW?t0s!|EW625?waTA$+6Zv%x<k<B?Nyg60J^F8sriCRINLg(~T
z0rVSTWvU%X>2~B^yLGZ4X>QD<jaiEud(aQgWXLCz{5J?&5_<)id$qbz_T(>U;>BXR
ze31#M2jZpRaFlZWO0ZNTUP#5qO4!a~%mbWMGU~$*AyHiO02_60grFGx09&1T=IOP5
zuus(Uhkn-l%)jJ2ozds(rv*}NM#j#Ce*umUvgVz6akjFhciwR2*VJV^JvO{<ehk5@
zgwQ%ioAvbA;8B9@>gW=SFUli@sH9wwntzSzjbQAwwR4Od9TrTU=3ho#wXepwyrw7i
z!<lttCzJKXr7g5{87-F0K>e?Pmvr>?3Ubj@%$rZ@7PK?FK{5u6PdLhy9x~V*%d2^y
zT7QU1ts6YVF-Bw=re)^guJKDCnC}YWKu~J2-i1+1{*=xI`xN}?w=KR=yIu>|2<>(l
zH$4A^f5|{5Fz%v5V_H^m=0+JX(gsCuw@<wvli6nV%Kj9F0zw?SsmQxOI1u$a$M2<*
zO4s$$NL)_@kAcAxKbL8CmSBmMWo6{0R`E{v!qxej*$hZynZRk*zKWdp!R^V_@)c~e
zMWK*5DGY1odCB__Itp%N=)$#iQW0QaTk`2Ca^1fvS=~c7D=o<N`&*@;3c5X}>Pb8&
z*DK&_Fw}0a6CGM8g!3`LC_^yvtQ!O~G9MVGCZYcH|E_^dtV7HbOq+9sV{b8sQ#Q32
zsuoag-tcmNY-~BL+92w&O1`Cp+$L*AVsf&1|C|(u&*QeM^Gl>OOYdof#o4*-tHM%|
zRpiCt&qxX@s%=!$bMS#9gO^pI@L!NRk@1Gu#<`eHAb9edS7y{M^24%bx>R4DW6^LR
zcePnSTZaK>EZLa`ZLr#T+uNfUST0kFP{3aqCbE7ie)o4BYjxJvyon~U;`L0G-k6!S
zyZRQ_GOREX3dR<4)5uFjNPikTKf%z>TSk!$!U#U7W&TdU^>j?19*Xj4xvM@n%9nz=
zMgmZQ(GRo_To%PGsl<4-(2)M@C6o=2ylL5XOPVymaS`6{_x0r9Zv%zSdX9xcs>s16
z#3^cVSpR0cCg<4j$~xk{npopk(c^h_b~!k@{)wVCw?~gQbZHcEQDb}C@xXVP^V1u6
z+UXw_(N?m(8j~qLM+@x~d;+z_a~?fnfK|cCpUd&L1F#3m0#157h_`Z3lqNgk@OAuf
z7u5x3k0!O{B~dZIjsiKm7Zi3V6SinD#eMQ+f~U#ehUPC&AEV}sT8}>rhr|=&;4h$K
z2kjJtmM_g4zXCB#*{S7*bx+gEb)QF{G}9#$O<w7^ZVx-_yIIz1-d&|x4{2-ckM3<(
z1y)r8XTQxt?{SP>u+#2gVgT0_O^Ffp>Q^oAc|((YV4*Rb-|?5AXpsE<tG-(DT1yIQ
zvSW+K%dgcYD-`7>LkMDD1KQ%4sroCaw0ceHAjXe?DI?SLh@W96ePJkf(5qU8LHXbt
zyI0a*yWOdr2KtdL%Q<UtWlKQTJ{8eB_9UI-YGITQ_&{9{ITvsT1)E!H*dbcmcV*lG
zhl^MMrs>QFj9>xvI-gB%Eat^1%__KW6Y3zUnQv(oE7W)61nY#FUl3vW8o=9=1hr2O
z=v&&PRhPbEKfkHd&o#RipZ~nN`%T=e;qX!-?U*o`bY6lpe;4&6&D255_rgQOFDm^!
zW<QKj{z^%?j}P`$RdiJ~v{BCyvE11iuak5&cOZUr!WoO*4BqvQ+~Y0=3K2l;9?aF5
zf23*99$-zZ(C;eF!@0|@xOpSdm*{3fa@(_|e;Jlbin#0kzRfq#qggAKv2LLeEp~X^
zV#_wG9Cv!^lVhH%=a6pPjCM9-*57z;w*945j$i5LLd)A+u1p}m^kf%1DdsJ!*TL-D
zSC7fhQT6bRlKJ8Xp{C8_G?MZ%S^_Vx0FC_G;D~&YX-`!UkgKN_Yj?fk0(=#Y5o`pB
zepx&QPv%dTDChMK4;t&0(xPUZ2eu7(Je)boLNM+-p{V6Ah#`ESnWE8b4AT3zHUg@O
z^3?>QiNS{cHu=}YK=ai{Y*Rx#&k1-V$m72>UGPvQji*_l&<eTXgMbKydRp&@Ix#w#
z_;tilcry{{VZIAT#F*%Sk5RHjw3klhr#RE}#;MKURYY=J<f0R=R+%ky7%Xzl)9C0M
zd&=ET3I5Rj^FKvgJ!|H9<QlOAuf~PLzFUvIZ2!ktmifFiZ&VZ2#x?d`ekJ|W8=fFe
z2JvC!ABmKGkY%qoeT%W#_64%pPe1n+wOw<z4t1ETO4uGIkMCKxF)YHVf5_}h=U`z0
za6)~=W)ak-q7=LzwcdE~kynykUD>@S<C=<qGaxg*Igpr3pKkRzO6B<9)6!B3K0N8A
z$V_EbvxwG3FZ0E(MhWYIk;mBwpy&mf1X4!~K#g%syM{qQG~9@Yu3tU%y?MOAYtoe#
zhp+qu{tH3RS?NOJ+cWU*Z3!V7pIK~3ZQ<7;qKl~EW*~SAl8#cHqIRA->T~%g;BG{|
zwb7Drjo}p-AqBVu=M(LbIYDe)w8~sgZqsh~%YN(5H4ag;0aD$9hGO>iE*a0LS#PtQ
zJ@4VN;N?3%Z3;Wy9mG7O?V1F$S#9R1(lFlj#WaPv3KTAc*Y#=#%_^#Td5N6Mmo^O6
zix!8xZ-4P=WJ2Ccd2(vx^rU7MPj6!4?O2M(@Q+r9;TPE&1ZFPl@{y_2S|d?VFYo9W
z>(lPA34KKDO@Nyck|;>q{Y$oz@6uWwl?yQEDF|}ExbmMTS_y&w04DM(XZq&ZT|HcL
zj0Xu!n}_4pU&oYIxbaqbtQMJ%OL_2>bsKUB0#dBC^Y*OEM%9t5<e8Y)(^65v6Gcou
z4+N>OI*bY(DOsezd^Q>ACDr>#w*EYDw38r~K)nc`q5h#@vh?eGZl`M6inycI?mhn2
zw^el<7nS!YdA`E;E<R+w%aqJ}B_q_LGRHPR3sYNdSts}DWb7|)-NCLN%j9(a&tE}>
z1d=pCTVjvl-tJL$bq!6DoH_qCd8d1E_ulS|$*&v)e~rJ5wI8mLq!8a)pg?KC!xw26
z5vWOBv=#OTSUl8PY5LHpOC!=d=CzeVzmpnPhVGHtjJmQ9x$U6bL`>CRb8kGlrw;;}
zB+sa23F^<F{(f%c6!ZUZXnaDJO5+AdjfV+mq<1Dr3_?tiWybat)zSYHn(%i`)@iey
z@>BbD@otw@4?Q&9`$6e=`r(KonexF!a`~;92i5!<S?u>G$hKzVEE{9(C6xOVJQU)h
zB4NoiZAJZvLtVP?pP-)MqWPjynh_JxJnKSDq&O5Y#)XnZp{J#GCZ*TI$tH(R)QXR6
z$YCc_;6A9$DY#Z87abTJ%p;P5kLEmaAIfDB#ih>RpK{ZvLDqkpc=&5uXLMK$1sS5D
zaEy;oSlFzsN@t_H_}>)*o^>D4C2-@=+_TnfQQtoY8n51vSbW8Nv+`OJv~p{tt-Y7|
z03ZOEDE`R5K2M^B=!EL+3pTjtK2oXgzL+e2OmTMQz~7^ReQIDn%4}=a9Z|#*c?3A=
zdGE8B233=D)Z)^~z>eymgn|I5aooegm#P>Pz5VJ%<{0CFVv!H3s{BWIM{u>iPTIb<
zuDE7jed9Z}P%N?pdt?4ZdD|~BIcLi|WtUtwD09rJ-Fx*@RAbTx@GD15H?6m_LeiB?
zIki=@-nTtGxu>I;T?~n4nKoP@6El|gYI#SM{V)jB#g+yTumrSam5}JAgKbF|JGh1(
z-u64p3M|0rs`-xEM4-+5_a@H#-QDYLm9y!#o<m!Jrfqg5OUK0&jrq9|_59$4?6|9p
zw+z}dry_klhkczZH!xK9bp||?sk~EP8y`SEPo80P-^Eh1W+iyO0+Tm1@xHr_dex1B
zV#Vg*U;Y5o-G#`QEhxnP4v+dr&9)-`N1$!v+p32CZ+lQB2CT-qM&~Iu=wY-a!pmG1
zoJL6^E(TjiAJfA%+ohYm=d7I0eDwSRK^PaZEi~ruhzRdS-Eily=pL4p#pXm!7ibb|
zJ3f#l{<C&FoA(Vtk!@=?Jj(qE9SJbX%qcBca`M$0juxlC;~J{tM<k0L*3%o){R!V4
z__VFONBpqV5~Jk4jumampvN<^n*>v1FSqyMFl-hhEcPUC-j}r&XKw+OLZFZ#a06yv
z&2WMtT}@gFbz^KR8v|cM4+01qA!+}z&5Y`Hn5TmaG<<A)XXFZ^DrYS(MgSH^S8a8t
zLGP0GEl|<9>fPt=nPJSU2*skl04I0ZrkESyNiBzRMm5K67x2<x$JpgQw1&%lxkrNo
zeFtgFb$WfEs+&<7yxu<G#P55JC;%`BYI0mwaIcNUp^|Lf%l;j@Y5b2T3EWOnm{rDe
zM5*8oG}yru-Z646YS?Mhb^qlr!Fit}vW2t*ukCoT=nWu+8`=9?Y;w{D?Wnt{#X@(r
zF38<GvIK&Iw(FNPv6^1&f-F;0L_72ng;%zWQX_H16)UG=IfQ_f>$Zq#2;R;wfs%*e
zDcI{o^I_WhYri;rIcTP$QAQU81@&~ODP#Bd&mU#&Co)DcOSl9=YI2;$lS8&le$+{e
zO>Zm{a#4j<Zd~UvD)o!rG1i|Imjk=9x_h;&cXOLwhGyh9tTn-OJpZ4!bp;t2O30HO
z$%L>Uwa;>geo_YV{JA{&(;Rl*ujW(zNV{zZn&}t=LRdm>z%rBLCVzG1Uw#k}w}1?c
ztbdo;s}pJ@M263@f%*wjBY=mV4{FKx%{E_{HywE+H0BKU#x3w^-{Yk=`8}V4Q0q3U
zYuEl764G#WwYXZfGEG|=K0zWz<DRA<95nSf0lvUxWC|FC9!%<|@&6o6Xq|`w@~2Tj
z9)t!&pdA6}d8tQ1WMaMDA6ciF>zm$htVY1&2GZZ}JAXQUS3;pP(65#m)*bzPOa4C4
zWF43r6e~howp*k(%AeM%<Lo=QeeSuirh>m)`i7E%2)0=CaJ&3rRS7Md8od07k0x(V
zw&P25`zFN8&!*Xf9`0nIqMA?0@5A`J=C4g^cSWpcd2;2y_X_<jix6i(SDF_RWJa=h
zf(NP5%`mHJWs})Ls@dj)%?K841A#2mv%_8*&~t2dnpq3c4&ZSg<_0yj@Tq+Iejn|7
z*0Fx^>5Ydc;3Z7C>Nz(CwY2P0zHg^+<8&_Msu>-Wje=5wn+EkiyeF?nEPRFmz9%P^
zK%eN>slROcw)_^CbPiu`G8^Z<R)E5UDD{Fl0%KJH#cR^SMN(V#7Qu1r89K%=-xQd?
zwP}7KnY7i9O+P6d62HO3l~O2wY9Ot8I?&b$qX6A5OSPW!4-GX*y(_^Wbg>))^lve3
zOt<Qw{5bFQo}k9Rn}e0r!w#R%CM-YFr}~wbHMqXz^sA+Y+ByP%1pWD!TB{JMOcY>8
zO94A9R*!`8a|fplJR4dP79<Y4r|8qd7eXOvS($Av&}?3odV&s52{1d^8?TSU?*r3f
zSNIxnt;-S%=h(^0FT3}$+DFW}*e3f%V2VGL<HvL#__)aA?6FKI&kQ%m-VxtH3@djN
z`8x7hfBhtCWw)94kCP~xTs_{G&+r4M9b#v2<tec6<6Uf_FFC4GNeTiSQ#&v~;J7>s
zUcTiRJ6j50N~Nb-3*tn?Iujh*ChXX{H(uJql|vkUw0?0O)N&_$AK-93AfAcKBEB5)
zp`HdiI*ROQfLp=KgBIW_^LGJAu7BdS%1K~Ik~mN=Pi9RyVY*2r3)Rd&6w4^NyZGVl
z=u56Fx-1piR-Zt=6F@eCjD|-biOAY%lD#(O4Ev3$?(Ol^NL3y#34qvUow32g9`S?Y
z#c$S`sWajGBwvaj0bLfLf`hrw8!imBSD}IAxbv8ID0E6Ea(2fUVbi6co&C4<O3gtP
zTtqHW7@916f4a--EVeJfA3F2Ev(?!_v#R6nq)LMH3Ab79n40GR5wPygPKm|^?>w7d
z&o@gyXI_dMmK~A2T9ubB5-Q!sOMd%<)>|-jQ)d7_P=7W$x!y>{hCik=oB#m-?dST1
z(_@K@J|Il$y{kaHZ=&0`XPrZCP34rxfSgf2i9b0y#xSxjPiV`_l96Bb!#1AdJJ^SZ
zmnVcIbMKQP_=58&@$9K+`oj?fz5n~gf&ZL|1d*&wFvFHb0r%4bJ#LiWG3x6&P}Z?3
z$OF>Jf0mB-HwGkcN~+l>oTr2YeBw0#ZA#VyQ+!odPn*fape}RoVY$!j_3ZV|X2MnD
zBay}i<?rI}=T%y(`6-QIQb)mc8v9*LOLaz^4J__S(F)$a8b70QGL=4ZL{W<MxPH3a
zQ|rV{%(oMq6`ckF#4lMEXcTWHTuxANPb?*D{}kXMX+$%Mrx54c63Sx_ELcx7ZDpEG
zqE{(odzHiy7{ukuhIz0wXKGi^=mhY+gI9X2{@R-bnzU%-{6F;V)o-LD_diA%o-)O1
z50KE3(?Vkq{T1VsgGK~2oa9ZRRPO(-GK!dCALX2b{Vz4NN&UA~l-ajho`hUJmm{7A
zL*679;)Zn@-rIHjDPKnl-Bp}vN28puzO3i(`-c79OU4VxgzayptD=We121+?obIq{
z_lGYjvpCdMYfFk(xHm1s-fhorWUw1JJNR#E88+Kiz%UdD4<X?cF;Om}22DfLH=>7D
zN<tg$awK$d%uFuPS_1VlpaD^zv;-jxF3X~Cq}Fe>ptFA`p!4z|#@u=Q=NMojaceCV
zRq5(3J97hr&Co7=u2O{HN!|lqp#9_B!{yp4vbr}IH$tzbFEg}F{e0H-YIa__d{@tQ
z7M<a@+OY1)IsBGX%MLD{Gjh3*_onWXmMDCR`+Sr20(=TtvyKmA!dtxJZ+!n^e(fy4
zND5JF(=rf;%$AR$8X3EMkj$r1;A{qusi^*sL&*wwaFkv_oi8Ua;?fO3E=2rGpf0ZI
z>PdTkP&sb)@mL-AufsPPqy^R;kt+9!>((9H-e^*G$;vbQ%R1UeIW`Zwr1RiMF!sjP
zsy4_aUNtVg-?<9Be{oyA8|Kxlz(QRF&I@>3Bq{}Mljw%PaiY=Rg+nkqPayi;Y0${1
z3ooq!3Z)VD&s=DQmoQuSL0Cj8?nda@m<MG)&uJ{!LQBIMJF2waOHh|=r3WXvamJ+{
zw6T?>&C;t2m&iQo^MeS^-UzWAA(z~e8vO%0gLkrB&htTtG^5l`F7cBY@)9Fcf-{Lh
zE9iA3o5i{hJC2p25pm=dGYaJ?kLCb|nZFo(1MZj>6df&Q<6xS|9?%{exY=tG4)4YI
zZFd#Ptg+(L<R6RXxh*(Ii&T*<az*tt<|!GhZ4e}915zF9yPd^(<;eNCTR`Tov@v((
zw<F~}SW=LeRPkzXsikn_Ug+3FowP6tYIGA_^3NT#GF!NH>&T=v$nyQWnjg*ZjgiY5
z*a201U_897nr?+zh@g#HUJl|1S?gTi7iz=Ph02C$%_VXsX8AcyUb2o?%ga*Lq~Lek
zKBF=&nCfuolsh|_@>4+UPTd3H$syDxhp*w#G3a(si!~ef%i-%(7NBq-wgnib;Bq>1
z7be5TP!teSe{pw`uLSz}xY^@}R#0hkTe#IiFEgdjo6!Ohzo(4n>s1y5*PbHhQ?E5g
zgA^!bg-?XAv8}@TFDy;bEnDa0Cxr5zPM@dteZXR!+?{Epd|&<of+#&o(i^Kerq&1d
z7NbJ=e{a*wLHpO+2Z8JRy+YO*lQj4Pir8omc@)J%9FI8@>5b^HE&8BXB2+tWx>?dz
z0z>(jaz2zPXo-COd0~&zxZ7UK=2B~h0yY7s^Me4`LbY8NYofi1+1<A}R26i<-F7rK
zAnYoajNc*vys^?Rbrw)Q&|?fHqUZH{<(~OZFAV&!NKfs4)LKjdXhXi2*+LET?UHq_
zKM@=JouEV&ULa$!AfWG>-u&{RDR!@LT);H?K$<hwy$+>WUg2%V@*~fxVO|0ZKVCmK
z25Qs|iC$P-j>tl+^jSDl9n@xKow?HM_bA;Ie`tf7nYD&B@!R7qAsBLB`4U6`AYv>G
zjp+)>NMF__ZrHb_8aIJ)@j-gSDn}nO%X9hP{q)~vm@Rz6dXY28DRR6YskD-9(6|p?
z9~i3qbm=+3JbQmqeD#%A2RpqwAxTfZ^7C*#G$~hX+snFS4a0!zJ>7r@t0<xWSB;i#
zC=y0M%iRjbo&f}hWSa$-MYd$k+(0{!sr*PneVqm2_7)i!3F)1``JH3#FPe~OU>3n*
z^XIx#hodTvG~tD@k&#3Yz92#vl|9F?i;cP!)JcS${=9879hqRQjwR8ozD|AKBPRI%
znLk6?vrniIwcxEs)M#42-!7=mgNHKv3sjyzYRxHI(hrYl5?X~F1#OSrIDP%AwGxX?
zdf9u1$50*ATDs9kO*}!7A65eyva!XAq5I&D)B|7vjk{?iR3cK;9W5C3FNw#Z1k}0h
z3B}J#dj^pl{J$&M4kgPDo*UmHaxpIONcR1_<+EVM7V0`TZ73A^r)l5k)w#9ZtQ~QX
zV@bTi$e}XdH?eid=2-Hl_oT-9WbgcP?VffiMpfUnBRR!q+@E7It;PYxey1*J)s2=H
z4*O}HL`9Z>$L{Wmv%dERFMtQCj~K$BSe!~qZtop~x+utIKb&roXS;+ef%f_!$rNV!
zNdtjCwZRPGzNMwL45Npv{?SIL(qp|YZp)FfRWJ*ZUKnzMCE|2VO!r{-e|8x{+nQJ`
z4W-`6Q$iDa<TQ!N2k0YqM5UcUWLsLI1>5VZ$Cy(Tsh_^+0rp#2kV5MC<cN<Z^d*Pd
zL|?69IPAi&KaLT(3R0g(i{A=L3Rt?NRW|I$NN5ZN1_VQoASAt13x(O=8NT=Dzxd$+
zTz-3wAoAY-n6AVc_RpLj1I(&mr89KUG0T>uBwT$1QU-AWYmeDr{0x{k#|E1Hg;C8K
zNubJt`8F0GTTsu&;R?j{|AMmM5#}g`TmJxM=wJr0wfn~D5mo>cp)TyK)>>?*kd%UN
zkAhz~VZq!DE-{1~_m<ccQ1;M(%!kei5R>)u(~3?k?bt<`T|+FKZgm%=`bcE)BS~b+
zcul;Uy>^Y(!JL<<V_fNU(!OGe;{nI#8Cx~_U&}S0cgD*lZ_IgQf6#Ut{>S+bT=iR%
zS=BWj%qwXb=FU>_jBR&fBfKYzZ3FTb%2uuwa>%`aF(JiJ@xtxa&C0z`pcNIj<f!ng
zaet%Wf*OX{k?UwkQb*qRn;4}*ake+=8FZNmKVU{?(%q4sj!n>f0kZpXTVH-$YiRpB
z$0W3U8dGjKzgciEus=JGhx)K}g|EySNDx;mmD1sTqi__pKy2?)VX<OZ3@Q9>g9mMh
z=yH(Lf`uyusX{ncnT#S}2yh`zI|lg;+5Gx`%6SL(H>VvPWDCYUL2v-*4Pqs=Yu_j2
z-5t68s+8>aIBP{m3U7;pth!6(*r8pszsGtHseN>^QIeitqasO#QB||HwdD+}Nl1H&
z!lQD3ZlaUj&<`g(saPs{R#o@V3FZ4GFbyYrPANheV4eZ@4D)E&5}2JBwGlB|@Agg|
zcucQ{=@I}%C`#9QVkiA8(0f7Nek>V-uVGZzR@5nc?|;Hzd!>b|go<Rs2n{1mW4}BK
zr%jjMQ}PN5*g~0|EB7*SR$f1YQSN}<tZ|T$E8CTGdgMd%o|>wOji6hw_C=26<UBiu
z4(&O#9AUkXN^E7KIo4tG+!>d%yR}%TSbK78?K^6&Vg)&8IQX^zM`M^C6HZM%=9w)L
zULVZ8j<=Hwi`fx=Mu}Oeih8<@o$l64XU)y=)asg?1Mh~Fc{&@EnxUk_PPe1wiXQ}&
z2#@uV?<mP#oX0whn<hfPCov|O0Fhz+Mn&BZ)ZME@z+5CIaksO#fYh37KRX`uWuOxR
zQ5V(e_u<W`B_#E9Kp>4)nc-oGmvQ+EQB3V+vfemkWr*9IStUea89V+1$NXS|O*~`=
zFf!5UltZ)Kta0>9AQVsl55CzAAerky!j&Hq+N|cNJU{$A9}^20Dn-IrqS^XI%Z?bb
z4OVZ-eCb)*>YI?|CG16i9c(f0jC34L?M;YhD^KoLVkEn_n=5l+N{yYuc_Ghu={>1w
z5pZH&lIWm^j?~2<>HFGQu~NR4r~iU4R)=OZ=#NFr4d$>vh$ZkPH{AOaY#{WyNI2RK
z#%ShxY7L}<aKti@1v3OzB~J}zo{cV^SN~g^0x+qv4RSY5uK=8)hx^0f<jPEQ7Eq;z
z+45*-5Z6g<-9~fi*Bb+_3^ZUT1IPVrQ*06H;+#kX_0r@cJ~3JA+^L8Z6F>_0#<PRQ
z4TNoswj_+O{~uRh9uM{UzCS2bBBi86=@e3Q=vcB$n|4deQr0>oWhd(}N<}Fuq_VXu
zvXrdDjIC^;IE4md8%tx!I>s3DyPx;yoX_|7{L$;Y&MTUEzn|xR?)$p0>$+`xbzdad
z_#UPS`IzT1blP!Qj3NryoQ{ecW;{Nq-;sq+t&(;x`R;HyXHA&6q2b>4DRjc<NqtQw
zemTES(Q2EQVv)qT<usQgU456G%UW`L<29bC?EA-5OEfr5LU4ylbMz(K6fSonhoVud
zJ9A~=rP$8i<?`~1D~O}U`@5Z+Kh7uj19en&t!-^FlBtihL0$)8lQ2KqYHia1gl{-a
zCv*btvDM41;ra%_1egM(TLC80c5&bV=l)gip0RwW!Y*=-J<x*C3}+154PfMeC`HU%
z*jI~SU!^{T%>;Io?8@IhChqPDIPrz9U>5Byb9(<$t3v!0hDoQUu8|dYc9z+7%u4J+
z)WP_PM=3qVOBI>y_zCM~hITOdlIBMK0bCZlNDo-|@WA?>o3y-JqxfNo&<nY)?-7+<
z&J%*JAEt*S=On%ufKRf2C^`A05;NjQEutK@H4)0^n(8L}zvWKM&iCD+d!J_etAW7O
z_L?4I8cPWtkD{>pRIlTk=HutNADOao#L=;wWVQO-n+4CzMh-EO)21|6823GS5khDX
zf{Y{tJok_R4M-Qc0ddXISF77c{w}o-S*nD!X~4>6BE0tQST3}k!1M+9^tHuDAZqV}
zAT?|C6;Dsv0t4_Yq~pBk$XfE>l_9R$CRSGMu8D41zmr)$dbETnbKytZ4zq2SDxzH^
zCI(5Q``WL=yJ;$?5~m=lLknV+VPB8Qit)W8dpCwkjI-|s-ZkL~>4VUyW0}4+F{Krk
zrIs388AwfgIZR38gO_iqtyum}ankZnh9}V|PBKGRMy0GKd1+z_d(dmsBm@4vfRs`E
zHVSv(2QAQV`2L+|-BD4h8*KtVJn~8`Ygz3fZcMh#D4(YLCm%iB5YI?f6EnFna6KAb
z1H80<U@8FGs)f<+o=dm*l7MkUuEP-upx^9-=_>TCWHfTOSkqwh_5Rd>>L*>oI0<W>
z!r{~gkxti<5rc(GdPrg~RCunT&~W*sRd#{B>H5itb6(_=cD^qwC#E)f&T^E4mXkJK
zCkGbSv4WdCYG(hY6o9gcmoBsNC04fmJQDx(j52@dqjEA2ZTjb2p1|V;w+#7sYPR8(
z>04>BPrn5%dH4!9xjR+LaNZ<uI=?NH{lUcNkL@;=FDpy?LYJ9NJSC9da*A3TE)ju8
z*)u~BCfL60`531+(&V$UkCSt22tUXot)Y)W>179&1c)41+B-Sratsxj*X>lDtIxgr
zC8L-JUjFz<P|%%83nyv25^8uHUQ)Z>9dL1%O`>;fg;@F)%)_y!>ic@c+uNo>i3rDv
zuVNS8uF>2zSn-eH3o)L5C~ippEQw>6DEcjq|Gi>(!oI3z%8elPN6z&Tde9|CUMTrW
z-{VL-He{k7Z!v<_E7%_{c|7H-{d!5>ISlKEj@8mfqd$7yC-`!ZR88|-5Mtu_C_J<h
zv{SkIlKreL^U~^8%BBc(Zw0b6GgEoS4d3w~;@x1q<jZK)#DM%u+<1AN(80#xGdu#u
z>_S5pQyj|ay~wQPa$9S?+x$PYy4p%xb@-YtIYPq2*(&}D-lV`&v?wZwF1fh$41DgC
z0yib;$1XxHS|Wq~#kA4n-h}7ilk~ZG2b`+Cyq*#G!~BbCbQ){@IpylP?;<ua#TVv4
zz{aUNuU{Ya>5j1c{@_7o=AgHiOjD_yS5iYyPaR83R;MMQ88QsGL&Cpa@JhN+YA<@^
z$Eam`9qUq4vkfKWr@Ix}jV*3SX4`y9r&B2Gg*vK|!>4-c)>{<3%(CAB>bZI_>=Ae9
z^bbN)Zp0bU8<y$Xo<yX$q=ODur}Ki@nDbc0ZXb5=vvn60s|x>qcm2oM*4yOBI-xbG
zKUFntWuj_2-HU!mrzr0mpnEXtC-L)ycjS4_E;oOxaDRWoVOzX1x2=bP=l-A@<XIMy
zKlRP0e*}bE|1y6W^+BYgtZKtMMwc5rNx1F74FV_$x}=C?8U~L7V`qLXvY=xIN|ea(
ztY?7B2)-WB?{tiBreFX)(##j{76Y6=4deO1tRZd`%u)dUPQ5=cA6a^=o_Nkiw1)s2
z*rU1WuLWHLx&`qKO#cgS*>-^@uXPsRZ|YGAT2AO7khP**3lom3_p@}hsR_*XS;%4F
zXI+KmC)l6p>l?z>J?zPfWd{wD>eibwK0l>g4$uL|1=IumoF8POP(WVdpCKE&so{{6
zW8q?EqfZc}zah}!Le8|8l&~8+L_i}e?aJ7srl0wZiqW{U-M2_OGA>EJ&G@(mio*(~
z+>5!Knx4<{QNFh88vOi*#~<vg(&(^BhxY5u*HcJkF;O3L@(2EVvhZ12&WEHo^bRDT
zwB2!i2c(h2;)vvA{_2#J6ucE@9h|^*bcR-@R}_EVc^w%p(%`?-jA{u!?1r(SK@|ol
zE0iQcbA^iF1%YK0fJlGdU5PpdXr|!3z-aXK7got%`tP*xXO^ygAK%D{8!_snT8qi0
zRK^)=wUH(M81+`~PftXkz;@;-_lIU_MH`^@L$rVJA~E)OwEvy)l1%lpt*M94HcXuc
zcPN*!TeUE35-B3&^mKN^x@Cdi9-63~G5rd9QDFYH^+|w&NeC%7t;;g?X&&_{5JgKP
zX_0>{_MpsqHZ%t@TKhs3-fg}zq<u2BpRG#+rejFZdu(e!SSLc+0j{RJWV6AOcK-F_
zfZ`<j<!i{J03bdhAiL0aMo+6Re;zsD0Z)iQ-!a378!7;kwVSK%y_m8MNQ9V<R&Gjb
zTNIHJ_`2~X%%k$mUg!b!BQO+coL!I3tRvej-6RJ7a&{{`ygz8sFxQ7MkBSQ)w1}_l
zRNIQZrQ?rJ_YfFa5k-;c;bhUSMJv5B^Ls9n`YKl;v~YFw)sGBSJA(Mt?E(*J(tjN1
ztAo7hb5jHN7xJR95>#jEFMatv5(ckpGIvRRP_Imsh`jFhPNlR9<DjovN%Lj&T_NIL
zn%<*neA4q10?ev?>WTgAZ0;Rs9UT(hDj=OYIfs`{(y8;{D6Y_3rwcatmN@7AOK%ur
zhEm9F!(7EOoyem9q+?QW@e^sBCVYFJ!)!rwW+swi$;K<y#V6$<NVFht%P$8WjpP6F
zBZgMWzk<<x>d3q@`6M7|G`*~mP$SE9_5PcNn|h5F-`9=YNSXAr&-*K$?Li*e-=oQU
zB8Y{S@#biiNUq}b37(Xj6)y2T=z~8Y!|_`{mrh-Fq`|<9h8|&V^yQbq*?|0Cy+33V
z=;Z6ZJ#r7NZkUL5#2aj!v!S##oK#?23baW-NYx_}(^Vf@v75HOfC$zT!nq$ApW{78
zjs680;A;8=p>J(-As!l|;yEqRh!$@r200PsfYps}u9gT~#q^R}VwUX5dC^k_LxD}d
zcv%7$CL0-<2T7p8EySe2cpTx5#7e`xU%<d<yGoQs9OZN~x43BTQl!7z81ps}gy?W%
z>O=Et>RTU!=eK9^0PNjiN?BF^X1m&-Q*3(AI21Y~BbRTe#3o7p54Q1S_pNhYda6&$
zK-c0{uOfuN5>Rqdk3E?+$^UCZbK~yo?ZB<9SF3a7%}Hx<L*NFmt!j0v9RFDppEV>e
z@KL)DNUj1Wj-CA{4)%SDhxpN#rv%+E-BDF&N93dOcH5bt9M${{m4VZmk5SA2z_2xB
zXNTU7vT<~qZRN#bYYZd5124{6Dl0OXSM#>>5_9j_pRLw&bI5lI<Lg07ed(SlP$B}|
z40z~Z;j-xqDHmfV09*w^aS-_wXe=%{^s;jR;Hm>w%ePQihggSfY)D0>&E@5ahZC-T
zJhD{MmODS}5@s;topa_T(~0<3GpK<JXZ!C%MTxH|=!l(A$*d3LFeK~+T{9GJ!S6jc
zwv_?>m^4FHGFOSYfDd&V?^?H-<$H&)O8%M<F14)CGY>{D;gbI@&cgts>yi<5tYh(Y
z0@@9-LW;vo#knN@>awUTc4JQ~Ln<riI~EApM*y)~>h@2zZjNd~UXnxb<{T0E%Fs5;
zUbUJ%7t-JPZ{EE)xplcDxBxh;$lumgmAY;JvmCR6;tuDlU5qf8hI2nq(~~4}*R8CK
zor?KoZtixTAh>a9DubFf@TPL_)QGhas&p#tC-@d2y!^>-9&`+PaTVXOiIeWmn$q*e
z!GL9MQ@*}7qWi?i3C)d|I}~VL2Qhou<lF#`c_dP#&w^4Gdcp1wO}+#nl^=LV-4Vo?
zLzFn;AJ%%L>M5R9QqbV!fM!tQzOuoFpb9Zt_rLN+)_j-eTG<3kq6WwOw~3&o%Fv}i
zZ;}4(;?zbtGz+e%)rJ}5Do8dCkT%sn!?YGF<HgF7;Z-_7k;TsbP~L}*v}WjU9p7Z)
z(}cuF{rh}}QX~#QbFjL?|L9rUm+s@J4Gum{SIX~4<}EyFy3Gh*{=uqfm(-)L1FddM
zq-Ka*e-)xidisFB1k)RI%L#v$JbI-K(wYy<;N4(eS1giwa=9d%#WtwQe^B;1FG!cp
zv0<NN-QZjq5^~~Z`p?spe5W*9R~IB#HfPrft?)UIuBf4R7-Ip%`YZ-)U?slKvYEll
zag-DucbFF{pMM>|wibBU<l;p4pNSL0AWh5;=D>kWW-t94mz!hqD<7JIa_H(D1O6e>
zX^{dRk&}~rPadB11r_v`_73kyx#mA)5ZlR|GRj;x6ENs;3N;KdCq{J9*h#)XBnixE
zA7^`XtqXqpntGMW`NNxj0rv*Ha1o6#GICLL*9y$)55NYTbJRkSHM<`M+}z%_vdAi~
zwCIQD#KAB6Px7CcJlrMY+VR9DrQl7BM8#WO>b2lBvhDC~O!EKF0|)m>4kQcuZ%5sf
z3oLU!$WpeAhMBSmljA=oSlQzZKrV>Capa`?+|*foBn%X;SGUcSoqf`%To?zaPuK>-
zNI6}%UQU=k1D-1G(uocgU0DlgrR1hJ4Hh98e+R9aA%iJg%9mvDxs%Yo#LVu%g%lVX
zt)He|^_<IW)&`qiZMNlz0pHLgiGAQi85$lAZf^J*?$po^y5x~fG@z1sAse&}hfrgf
zqQtH0xfF*x{bCP`Fu(g!{)}n@;r{QOC5yT4&c%f0t8UST!eY3)@t5hYNFqT#wN!qY
z#E3dmmb88Gy{4D$o9{+m()Z}>f7kvefFv#jp99Ho?jUjb1c}Qp55<`!zHiJ<4TYBA
zAq#UF*+ky6G>(kvpUd5?(~E3*Cd-SWM~@a*^SB$`9gGQBwOM|3K9tots;3h7W#-t{
z3@4OjrZ?zBrLZ@pE0L!fIV>e7w(TOn)+K{uy))LShoAUa-)P6{0C`<6Qh2W?Y=ujL
zp^gna+zSprrfG#ClVOffXy8b0SW<^tfGqV)jSccm9;-SnUv3Nx(9^=a$<SyAJ=TNd
zxbLk_?Lj=ssLKY?Q>NLiZ!0F6tcx)dRrscqE$(^LMVoN{F<at}UeWde>ocIFi#H8M
zCOAF}jk$Uo5*sl51ZOwQ1q+$JFj%^R`F|8Gqw+?8XC*`1=kCi8JN>u)q^W)0OFw<z
zwo*ptl;W2_@dv9_>z<3u*yt$Nl#G7lbbE{|<$Nfv0l#8H8ha32q<3|9;6X^uRDNa_
z93&pFdS}Li@nDc<{<xI9>~ysL@R@<cJE?jqL`^Eec;|Uijxwa?KGm2s7A*0IK9YC}
z4Qb-Of;rtib}oMw9_~}HoCjz{*2{{x+x1nCLP^}whV_+TLl-qOec5kmWy5(cP*ssV
zHE^U0MhqLGai->A%m{YMb(m`}+6b^xw=V`hEv6m|QQ4AEWP8o=&DBYE6%ay>Zv|5e
zXqo7vs&c~yeK79;KP%>%&Zfq(&l8KBqb~mezn~;|rC~<uts9?P)D+vCh@^WGKhu{y
zcKxRBmn-tX=9<Hs8<Tztqj5iKu~9*RnYB+mmu`>aV_FmEUi5$U4s0AIVWLuE;Ay42
z{Xd8*S2GkZP<p66oiA`c1<B1ivZiHjk*mtSQ_M#4Omkda^(-K?xd&n%RA>@Ax;&_0
z8@8F}Zwvc}jTKGzM*X5ApQAqC-#FRWB-t~N`E6NH+{ZD-RATo5iXBn;+c_H`x71Va
z5}lV`_P<MAMC+d@_}zN5Y`iezn?!Q{S3U^6iQO0?+ZfPL=!VDgmB)cgMGbz8qOMqT
z1aBXSDO08cm;6fCA(l+@hVl-@Az^A&W^T(OndW!?0vc3leO76VMnPVc`O(sU#<mh3
z2WpRyBEF2&8f2ZZ^QnB{K8V!1QLkT)az7NjzEi95tPh$XQtu3HRS@k??tJ{HJY&AD
zW2fQSZweVfJ4enKlYMk|KAtE3;M&-*OY!uMFZL%!#Bx6MvEF>A>DXf3!LM~p^I)2O
zNs(1^h0>;RTI_UlW&E(u7{PM2Q<5V#7GU~NN|XO;5BAeai7e?Tuz#V=dO_Y(GS9%&
zK>wAj>zm^vpC=kRA~6aFMYvOdWCH6LHDEV)Q=8idSuUD7tc<zu5p8+PJC%8zAVXV+
z#LemQyfm<L<mH!YPZooyQtt|uTEtNZYLEU6YA2lQxtg#ULB8uWW9HPJbk(=6FSB7{
ziN-ztSBfzS%A5|qwpeeSOL?^u?Iuoxmk;>F_W<Drb|xUZuO)&lVQ$_^(S1^D-u{Ig
zvd<N_jVvpb>}u3-7;ra_ouCEgv&PFGg^au83vWr`^aPY3jI}~BWF38dsaOq{IBcJu
zlQ}m=8Jk_kZBt<0I4d99V4kC(BI~N>SC9c|!y{sd*D8vZUXeFpW{s1!nKI&&xc?X7
z8eAv(zyB`Dh()u1;he$0c)l4Mj)<`CqSm7`%8N~eLLC*|XLrFpJbsuljPfFS9(mit
zHpG(-2U$c^ZKiHAPUtDZq=5H`jxRKW%WihU)D=X{D2ZEHNL5h`(M4ur9wSaIe*vLh
zHzuL14M)9i(`=_<VKfrvJ*Tp?KM~OZT7(e#B)<!02$&}x_HO9!bd6<t4I5E8xV=la
zHt$Axbb&_M4sdQsR?kc8^O!Wf=x||@cLju>0h^ELD~aYZvKQ$S)Q=uz6V%bv*iBP`
zh~)QWtn0gh(yZ9cYv4H-vi{iKO87SEg3-|svhZ-jAvFUz9rXE#H`}eM`Iy}Cz&pX9
z(z??f2`LW}*a0Lg<ShuC6a=J!VZ?J*{@pwKJqo<--FdHxQT<i}ehF;;FV;8gJhtTd
zaGXS;HHu(n+zs1F8;2#G@Zi}5O~G!kIz-w8!2Z(Bl%PWgJY|YG_B4UN1k9hGroEw$
zOdo{~E0p^)BS+PrB=CplU^L8k6!z%?-Ak01D~HZjMjbsn7!fjnG|~~_s&AKP<u!Ud
zBXp32`62&-!MZ<^@M5p}x!#21dD<JqMmRm~V6BkxR<edWI^$kjGkiLS<5IjG-?9zt
z=5T}@i8(;O5YmB)*gHtm9V;x&x2M9v-5bU+K~+1LPeR(^;vFHRd*53z%TeH(jgMD1
zgTamu9-9LY$N1?YMB__+cCmB@sS^-o4J~7bi0i~r=&#hF>cBR&wJjkf7kbB;k?lUZ
zw`<QHE$xxm1C!Sh4f|E0znTQ+t@^+P=mEpBqs>7t9lQvugYpWBgu_bJG2{O5Krs#5
zU`<qB#dBLCfDhfHF4`-gFgTXr%0vzX;TAMn(Fx)O69u(lM@SP^^gSIC!zJJ6&qZTm
z#m54?P9N)PnTOQ*(A)laF~Ub@uL5NX#YM8ApFYN!9nROBx8`~cj>{(dbuiC<(y6)2
z?0bH#AgrKlbT1rl^J86~Bw!jKPi73-g31E7%{ao*-8MVoY7e45^jxcYhQ<b8?ZkRi
zm8qv45`I>ucQ|U)ENu#8SU~TYxilXfAhRG>l~{V$5;GkJp=$Px2ynDA*QG4v{mrX1
zTU`_{+}pQ$#0a-hl5w~FJR<6+BTx7hZS9hZ;nZQ*`a`wBM;LRUDGL6~h#Lgw^OhS2
zpX=aNwO!sRG}I!_Q;E$A!;HeOeamO-EGMa?A7_?!6xSE?Rc%`w!;a^iwHAzN&dv;c
zt~G3wQwmj&lQdyt-t-fZv&tH!ol~@$;U66hV%=7Y)NEa@-p+}0P+;w4=}c^7Uc;^S
z-Jxl9jHY6A#)>rG@235N@J^<<Pu6yZNprij-(wZDIc<_pZBIz}o@n|QsNb(&BQ{QW
zIBRYyT~O|1i_7ogE2L&U$Qw`_uNE|_2$X~fe>D~Q`TF394$Q$<I9KQnbGCFdLaUuN
z1O)`l!judoEosYJ{NxrQ+dzLIiTV6-#K&ik0D-|)*68E9aa!&BrpB9ZPG$+>?5J;*
zm{>pIW#g=oravBrA1oa_BR2Aiv9}i}uqEV`-R(O@RL~E<mg(xfMG}7KqW$`e-`+T-
z8ctTOj<&XT#On=i)0I>96Q3sazbG%>&@>;tFCuf;NY+&F`lUMt#)W2j_07nwr;`DK
z#uLZbrQWNK8uE?o7dtQDTvsK8V+lGHGhx()ZB*@L8`m9-!;lDh{a2LcvAlrp0K;Ct
zSkA74gzUeNJvP<V<n|X(Tp=fORtJqcbe|SGwES7z5e@AosL<tYCzvKX{ng_+E6#3M
z{!0tUD{^wssyp3}tbt?WX!^n(3kNe<?gCK5T##7-IcBOX7=dnIt2maXWmPy?`HE#A
z4ft-0nC}-ECL?=if_N6K(;1^3;AbsU>F`UUg*@LUz$e)?IeA(CYK`JhlCsLaL<gYA
zwd@67njoJ4F*wK9tk?xssl7iSh4+&0g&RQcBBo{5+`W4i{ETT2nwpZi(&C`J^LxzB
zEql5iL$47OP3@b)42pOeRvus5sij>TkcVjZbQFx&fl+h#gI0p3z+|eVde3o8!0<ui
zB*NC{Nx*Q%N=C5dg1xhyw;P7yzg?UdT`kK?xZ9e3r8h1)($K(R!`TvJ)NpuHMn23R
zW9*&Ye@1NSCYM72b5>$<%Qg!T9w9P$9k2>?#I&`ano(V8p;Y?i10m(mTSxw73#5py
zV_Lz$(F;VN8~OOs=nS@HfPhEub>AxG>o=bhbjTe9(gXe*4lTdr+#iyY%W4K0<DQB0
zbMpT{sf#TBJ;e89tfmVcZsZ!pDH>*dI}9AKl*Tc<x@^_WgGnKt>0SW=+hF(&ZY~3(
zGEv3wuZ%B(lH7RPDFjEu0QR9mOmPLqnm*Fv9)!7+bQ>a)yrTNKNM_UPrKTTvXun7j
zhjVG(CO{{u5_K`x^IT&Pn%gg3OK-e+gZJ3-H&4MxUICvUo&{RT`b#fp{;7k@0=_;Q
z`Sj16aVp*k)(ud|Ipy!bN*L=K3I&fC07+W&|GM1BIAKF+F{tb7(;-EMk`#*pryUZl
zgf>NRl>2zb;irxHRCmu|)UrHuAWh^Y|F`I}Pm1or;ugxB$uWu<yR9LJb#s(b9TdtJ
zl75Vjj_hy_y?qW?-N^jK(G6}HDjhFqMEIl`>tV|P+Xvbnl)7XpY$vdXmrk0~V!A8}
zAvG>!Gqx}b0(c-rMXg42s%hWYFoQ7d+iuJ|fWO9Ad~gg#1_vurF3<f%MAHcG<P(DV
zI;n_wgo4)P?^=Bm{!*@b)u^$HR*R+K@nTt%C!dyd^MzRb8(pqNGn=6I(J+ne3vo`c
zzYR>NK^K2oEsfadLy|dQNE8jBegif<(gDqwDH2ei#`XHDz@7If;>nmbXHMcEk2>WB
z{VwArO?5)vCSUTu=fj-^9*w|6&|uzyosjCp8=SLNhxE;E++mOBTqe4aoR1+#Hf%JH
zcYJf8zm9C%&oSv4AB4PA(0eA1O=0F=$ws*RN4z8_Uq=SX#GuYWD@41y;ve?jCxW=D
zx<XGp!kulmU5{=x;#IBarUvxw@i0l|E_Q*-gG(_ih5wW0Fvc5RQsf_sxaU2tGeG8Z
z@O$G~fT_&vb9nN0+I5w&4P4fH{Z9hKC$SA;Q+|fXy5<M*cRh!p?fMckdw{qOB+X24
zgUC^>vI?ZII9oerHbR@`m!jf7un*>Kf*Awq!LG$%>M#DMOcXI6QeEl$5_;Ckad$Al
zBihAkbc!JrWvf$Pyrc6AqKaT~kvsF4f+*TBZ+eu;wFthH#^)0f^ld@r0fr~~Fj^en
zx6rH)DgMpW&VM7(ZLwXQoAJ+*9=;G4XkCb5dGJvN<%7*(hS2megewKaF)D&7I!5i0
z1>|~zD?M~iFEZKN2`~6M7e>@vAMnxW|Jy<1B^#qUR>5sl`42KQI(xvSXvv*)Dng^R
ziNVmD^`NiUXYFk;^Jl3tcR)AN=iu|Xl0#BD8NBi=Xt05$$XpH&*XK}XGy7PDL(6b3
zB3EJ~mX2P(iQ5>)=<0S6iUGc-975h9B<6<fhkk3L<Bc<LXEg3E1Pyde(0bhTkpLSo
zk>Vbz7^aAh=OB*Nzg$q?(wlFX$Cd^QxAEbq=dBBD5#3nDR3d(2`?C6xJcMP4fCo^W
zf8NPG1??RPQGwJ9p{S<WzmBl~E<6s{tB++yk7!=nSA{_7p${MG$7eU;h9W0--<kr>
zxH-&0NbTsmUT}ucx47`c$X;x(;K)3Ws|o<aSuXQ!*s!1ju&<p)sKSItbk9M7sH`oD
zPZIX|W9&C8H$V;@fHZAGl;qtgftDrY$6(K1Q)^4^8(obYRgvk5Sb>{Bi5}2g09&Vz
zyL{Jb3|KHubxmHiX(Z13>M^itgBPp9d$D}wz!Y4k)j!^G>`D<tJ20bzxfaDX6p;%l
zv}KJ<yksN$REuU%knUaQ#HH>nh0r3W*6A)@F-DzQ50hb#J30XyH-?VD>U;cm!2jZU
zo9mW@g)&RjX#OEvo!{0G8%-jpxeG)0VAS~KHSg*|Jdx4hor67^-QzRM+mb+ErO2H~
zD+4QuUrwxKnK+#9>;Gj8D09jeq=+No@NFc!R-ojuhI0-jk8dPC88G%T!prYp^@r~P
zPTnxkCdd@c6NzJp;!qLz9(~z6o<?vU+irkr0)1vkQodlc1_XZ4zh%0mE_lx9a{fdS
z*7{y#GQ0>f`O@GF4LPf|*knnGmSSQEr0M8|n+CbfT)V4Pk_&S`TnF<#EHfK;zq(}0
z6%q&N+q_9HA^YtYc2{O>8JLylrK6__@>zGwx9j<E;}(dLsfy`9f&I_>4UaI%L7wbk
zKn=jG3Tq(2^DN1z-e3&jDKjH@Ie}Dd-^hMI0Z|&UURN7YzQCqm%__%YyX(~Db-drR
zBA;q*2);nh0B;5(XM1uCP6p>rZDbf=9|9g0oU?c)%xDE^5Qb~waH@}xrwqvA<BXu@
z=H)CH&6ai$*K!$1*4g31LNPqW16Ya23G+GR--MUQdW=t&>Hpq@^M5gA?)tx&^)z6L
zkZ=^FbmnS)-GaLt0iOm9JSCrku%AbbyRskRDAEoG<9SSa92+_q*jEVIajdHx<p$Kb
z0yOdqU4lAwLF=!?A1OWcWVh$y7#QVLDb8Q>Ye!>G|D~ujn5EI*mMtLt9`$HJgpu<e
zX~|%9Mxwc??d%hQe}nlKnz{qSJ*x^mDcD`0v^a(jM>Y)zsxCZ-Ul_BC3vt7?4yajX
z%sQALz4{x53Sz&k<-Npla{8~cOErKLF%OrrfGc?Fe4f4X4RkA?i<du1T@-WH7p6OW
z&zmcTCg9^hpOF1#naz3c$+HPSz>WFs%KSf%2Ep~ID@VQfabcf+f2?DjD{%R5uxnLC
zpk~h$8c@^dAA%E&L|bJ@_V+u>wdWDf0WL47OOC%<O<k(N&(8dMA)!BsFV%35R4NFp
zrHQpP#O=Z!&ohN>V>&vRSt9Q%g#^(nde|Rtrt)t72t+r@!D#$NMT>AVCWDt>0*VH-
zM<anbnauV4B_qJJ<PG7$o%(m@`Z~jr$2v(#?R^cH>y0ppZ;v6QBi6vkPBWajC)0rI
z1)lx3DY5+YZ0u>`@`LtmWn&42C97>Mt|arc`G_P!2Q88Al#~b_;dXII3Y!-wWqh9U
z-}^aR3>z2;yRqLBq78%<n}cmO7v!(s4?}~x+}68mK#rrt6ahh)@r|E{qe7vb^R;;n
z=4gOo$(Rr0TvIKLg6etcChzmG4|cjz{v8I?x#HiCgJ$GH-T*3lcoq$498z)*r{q5~
zLEP)5zhB8n6-I_alteTB*NSVw9(T4M&^z3WCZ9vqsHH$0gxr-^3~}nID?eZ^p73vX
zPQP4rKU)@rLRk8Ry@YhEngMHUEJxaf+Y<{Tk-OXnF2CI0GY|PpIU5p2t1Sy6+7dW0
zWvDEL1Ect`C>^&M_iNZwC<g2$=6ilN1J?Bl&>8TVy~g{hFpmEBOg5e-0<0g<zm$hu
z08gDjobq+RNr#BlmvqDB7!!}n3Y4z&1<NSxn&TB<?zP}{T}z$w=b010oUYm+YDl?p
z>>jD1PzE1;a>|+~35H40DsWg~hU1%NMjBFhZ(p+v6T!fzf$$CrAD;XdjU9o{RSOrF
zJy5U1%F%KCm4jVVpx>Zf{PzqSuf~RTz|Xhc$I5Z)vQ@9Lx_L@1fm|iuhiTv-brU{G
zbQ>QyG2%JDr#9q5)LU71-b;zGh4(C~4*lz~RARcoj^&5Z;^+Zo>1ldjjlgA*ip27%
z|H(*9k*p7`lI5r+!1G_t8=El6f}X`>C3@fa)`Oaq;SkS8hOWPqM<4^-xJZp)_zl(;
z`God`>&Vsi-w+A0i8m?hzkT-#yMb4K=L^~YU|#_Yw?Kp2ZnPDFeu7lX&}Q%51Nyvw
z1ytl#OnL?xkxfTm1DMkO&lQ66?FraZ07=ui_+~tsG==5}G1h5<>JN>H#956OE@aH?
zh9#-YTCB-~CP0XO<t7(%mwD`>$<C3ZX6F(@md9tzOHGuot-t+q`_{86K84;QSMFXL
zY&^Z{u!z?4b%!ko;=WF|L*Iouy-{%vm9#R+CLi8d`#M0ydDZcQk9!_=ew;p07qVpR
z<LK*vHAT~Pqw;lHb}d`xa?`YW%Z_QLTOCd|yL0|x;r#1Ko%=SK@s6jOOs5WS(QPnq
zm>6KIXS6AQ0=PC6W7wQP-ktgw7J2M~7|yOqs=-AFokW;G%?oe&IlPH}sw^MkYP?G5
z%{!muInPGd8^5i#@7`zHA}(b-GNng^%l4wK_UZmnylxN+va4bn?rv0S#=39UW^{M&
z+avwtTFpvhbwl2zVmCF7t>f;13bwC>s^Zsv)rSvz^g;HqK?W`5PkbxHJ>1bUMjtY-
zYmVlF<_K-%4f%r-V!gIDI~B6_VX`0jR&pHQ6Cf|V9bgN_l&4-DD(M^aHcZmrzOROH
ze^ZHwf}^pN&k?eyvU0~m;YCjlq<;JPV&HOkcz%9<7OERzbtXN0y$%;*x)W~L&i5i)
zm&x;&OxL1-1%Rr4=5mK<81QH3#rt?`ir2J#XQAa^f!b9^|0F_0)yD=Mq2Kjb^@)$S
z_hgzWq^lGz%(m?4D35qHx;cv~DwZq~VIySMH#0K`B$el1-SlpSzc#~5XZA4r{8C!y
zh9Gq%apv^DMlEjp$C~K~4wpI4K=ZdOFfj0?rp-An_Gxh?oHuasUCbHC0g$vwiuVW;
zoQs5URQb?+2y~j#OM77=Raz*vo;6=3p?1Em?v4N`Bo%J`&Y7D3aaQO0RLCj#nWulx
zjt&nG4Xu2RUa)A|4>FFrxpE2AiqFHRgz8D{lF&81n3GIg1sZDt$#(K@{^p!K<}b6S
zb<ki7_{tD1*_U%>gJiL&xVWUGWJe}%3%MJKilG%@1KKk6L=T7>_^mMZtEA6TbnCJ~
zj}4Ks{3U&<3*V+Q0}=8X8nC~83!C-?U$`VtM9@~o+NQ%)P)s{9+EJgT@x%)L;wAk|
z0b@g{kPV_l4p7(AriSbjDcrd^(;afpzzJy8JK=5nA8nA9zU?{@G+l|CTk-ET6SX9j
z4`;Es=O};>pV1(pv3FT=IOk^~Tl(h-h&k~$`!WpW>eA`-(FDLxTU!ZxZZ)#2i+Ot(
z7xV4o$M|C8mC7Qq;({;7CC8Pboa^h{u^xEUxU5IJ;ggC->~zj<oui6JjMWBJC(~RT
z3ai0rc&c^QyN@w3WA^k+z8Gs3I`lse(tLfUO?BnEySwXbZ&}>#;Zyk3)z!5V6>BF3
z0-)CBJd=e-h*qw(p>r^{Y$ksILigr2IB?DYdm5?|M1_K?1pa?k+W3sV;1#das1xy%
z&E>GIG`9P;ySJKhM}jJCm9a(Uj%F0%{kanifBdy^i|d4rCDVK*ebpZ4lwrn+y{ckB
zdVcyG+j?qV2p2N^$M3kk8rTxStKSvk&O*5vCnuNsO$>T6Ch!^yf@bpvF7suie4TkA
zi_UNs{ORqz3%n`uL{rPk(S{P|^l89<FXSW@Kipv{FMc)KITsB2fgC<I*7VH$H!Z$p
ziB?xZ3Fs7j&sss8h^iht$GcZC@Ij*34Cnl~EqS|8)Wo#sT1~gOy33&zWgl_4n7xt5
z(5QJMqX?Vzt|%q8)6i82SopOH@vh8|6XE*8zcS9GJIaekWNj)j1@f-h`}TtL;$qz^
zCtHla{R{(E?AHMeS^e#wWEl^7UY|*yHaJjW_lthwYt(WQ47g}>nzNn^mI`@vF=w}8
z01{KcveP}DPi{SKzXTU^-GhE!4MNHHmk0$Y*8j3Hr2RqVDg^#~y9L}V7*}!*yP?C;
zwG_4a6)MB52|#d$g#II?VYDIH{JWj%kX+^=TnnLI0D#)Q+2_v|T16r6A08*SgBPR^
ziT0T?*{zHp?d%*=KqG{LbhwS#*#2A3+kft`{|Hjz9>MdexskNwp48{hjsDJ+NuIly
zYm*53^NU(rigeWIu<l&YfeP5t>kT^Zb&Xc{QpMzS-1u&JI<eW&a?O9MK6)6w68n}R
zfe%($o`2m78@0Rp*+5QrAxga7@62d(MpsK=uiu{nHDD_aOFwfpF75uPidsw|A&|v@
zfI21{um5XZNG-Zx?7aTxwM2+4?b!9&Cl*f7F1d4Vqfc*c!G-Lz*Yw#nTXzU%TtIM=
z`@UJ=1DlO!yYNC}#m`UHzy4l#=aZx_YrfRZu1pMf4lXLnz-8=nf(z2`Th873D(Es#
zOPRZ_ffN5kLCbYl@S+wGDZM(oGno4gXbvD^9xMvx%uX?y#j#xB5I8Epyfx&soLGM@
za5P9^NQSSY?hn7^B@P9)?$VHhs7Ect(UN+GkF8D=qS~awZPYxNpc7$nZ7E;ICDlj5
z9(>Ggm5Nd($`_07_hfQ79GW$GIkT&?vvW=vT8#lew$JMEJ~8B(f5;a)sy&2>*9uBK
zHFf7;=S!rcEu?!^J>;3g7sSe+nelrU1WX<{4dMsujXeeLO=yM;Yr3zd3V}?itUA6f
zUj+6%%DT}SdJ8X;gJ$!R?i<s(x}wiwk@5xHmCmmsA|m#w*z4%%NU!0n#WC)c_KGO<
z&p5b^c75F4+ykM=hWD%wRLsve1g{tO2$&tuekQv)7j%;|-O9A&(e*kiyg-i&nS#oB
zJHPM$Jn0Kk&R|Lek8vP}uf>KnpNXu7()HK+qY&HkQI(R!=P<hdim6B23Y@y_jp)UG
z8pq~~i=g+_V(>db)lO@5Cf~ML#CBB@2nDD3IDMKtXo&{U1rIWu=YZoyMkNQ^R0P;R
z$>m+3#q$xFo8{{3$*Gc@weJc%i;_xzQP+tG9@3{thFF`W2Wu+-cV<1WRwM`~d)D3r
zoW15#zGpT{NF45MDi-nKcUQwcuyRL~O4vP2D#|;&j$@<aY}`nE3EUZC{d^k7WPAnw
z_mM*z5`xUP#WRrMQzYIAzvn~wf`2+}O&?3aC4t2a$p;VIt|ZRyJy3z(2*?F`azqrC
zDSAl)lg9V)>&?l&NkrVo*Jf$a&zd)3xfIidw@KdKyt%S=hk)QdLEN5JDf0*MaHSfX
zNp*;K^37z3M5?30XN<GcQ+a|>R<)<KMTW0r+P&?3_Dmp<u+<lCz>q|Ahs#3x<ddA)
z-Jf#?DdE=LF`9*|)lS>Ws;E^7o{)0w-EsFBMwE)m^_bQJ06s;S&Lx-@C*$nYtgWp4
zXpr>4DtX*{jjhFpr{5>aLLgAf*sg(2*^ug-|3U`$1NYX*mB|Z!Xqryf&TcqnnYNV3
z0kk`8*GjeMM6QB_tBSelY7yPX53PI)ei$b8I}`Oeh{Bi$bElKwBFpseN4_$b*u{#9
ziXKEMfGaqIiYoTcTxh=*Q*g7mI1CzNC{rcPB@lxB{}pXB&Yr9FT$wL`Q@`WkpoNmA
zc)8H0{52)H)l3JeTe7ev$275~{BsajLPTDa5i)#MC-3wS5tko{Whl37mSuE9I=1IG
zBw8XmfKI-Jbqh?!)4gMHtr<6Rkr#!+v#bSNmG{^5j3pCSh<{}S&F|WG_>t$OUF&S$
zcK1VJ-VfW6_fCPJDBKo%%<vD_eIAg{tu=oT0dF<SiI*oYin&sE61&8^gFLW=6gqtn
zm6TokxeY|ItaRso$eerUt)Jvs@ycFJIMF&f+XG<BD8NWe-fQ|^vtlsy>Uk{B)8Dwl
z&U9<^O|cl9Wfzu<M(|$s=3DFWJHlcvVR_!dX;-=hW>k$)`GK9<$+k&G3m>_Jb0+Ca
z@;R0^4i{gO|K4hD11`iMmX9uSr_Uwi*^aO>6rgt90`>?zRZsz-YcPMQ9iY4dCDS#`
z{nJeP7xXO*zh2^lka6mjmwx6#6_w`WJvo>uCnp%H0)kF2QBfH<<UsBK%2Z&Wn941(
zTZzwU(ZDqhyE)~uw}Q8a3_qZPe=B=#eE6BYCJcB4Bo5sR;-3OZkUovUxRdtadAKI!
zlMjJwXMPf*5{8FE@LNB88JZZq<Kr3Q%m%%AbvDlGYgC<BE%TUul<bXiTOzd2{ATPe
zAlc<T#0)Jg9gKiR^i;sp{yLwNh6<0bz?>Tc2dmTDVc*+Pk8jO3(&RDM>hx6pWLGRI
z;b>yZV$oeMMG--FHGjg+C`l8;i@R}cpAcp)7}4|BTd*Hivj=cQ{CDKDRM=lFWk^or
zWg-9oq~-n`{%){Dk6-`M@B4okk@4b^R~P6>9oGYgUihltE+$&d@21kR=nZjuepH!8
z2F$S)tVB}0#!->fmXbi0FaKEE#M(}?xjf)!>bdSNVtYL_SZgx5o1@v)Ywc}nH<#Nr
z_TGqcnc93dnAXwJap#~~9*1{TJ@G@T6z79T-jtNIrdXZyoR*w;whjj*+gA_)ZlLX(
zuB?3C+|Mz@XpdCB(iQs<cI>8)B1$I8fq~gZ^ToWo7D_8zd@2T&zzqesRfJ9aaO|kL
zT1>*IBM_cde+R7eO$0&bR3A1A;Cd8Y<${TV6jhv{ag2y!a#E6de{F-@h+YOMt*0kN
z%4Fog_Nx=lVWz`jYFR)^A_lP3jeX3o-;E2cq^epsKw^EiLhfOaR_Up(UR6s`DcjZR
z_03JG`>G^*UX{|@Mn3EH0aXFk+cJAq?Fkc<CH#;(T*j$dvI7l3Z&E~O%d9frvlw!3
zK)!ow;9}=+f_jd3au(g^yS3?sg=<(?woxB~oxb#+=F5MsBkLDT-l66wammeX`UqQR
z)E5nQC0I?K;lmmG{QJ^-1_qjJ^<e^>txP+;Us_s623>O$*6RWGq9MU9`w-a&rM)O$
zu8@08fUM4}H-PRl0a*d;=z~4c!SA9ZeD9Txa5%{vAE?>)Q;R$LjUvcySyaz6i5L}9
z|KVlUbVilkV4q>0#RK<o@7XZC1jpiSho=I;|50X2+C1LsLDNc9MpXPR!QOTNDFkRC
z2Qb7%KeKMSddZH!Vx0-FaZS9fG#ZCqc$#X(N!0c#3{lMz-qUBYQR;~WuT_X0hju5j
z@<=FzqqLYxhjGn2A>+gLJGGb%LfAsdA};=SeQ(zT1_9;OzCargQ4na;fbSh&?e2La
zM(u6FQ~KpJD47wP>rOD%yuBITLJhni7q-07Ie1!Cu6uHl?wB^?k44^he!)2Ami<+=
zwbtB+--mk4BHRa^*WELi9{4KD)q9rD3Hou{-BXXN<~&EHx=R)mSMLc_#!{M)sp=AI
z!BMb_N0-6*e%%BX0RJ~V=oiu5@mW}*e549D?zvGbPU<256KT%`?20(^^-?+7wIiYM
z{XiUq%Ub&T-L6;Kuf=AwDI@|r7pp$R6YHxXTVJ1L`E$yaumMI`Pn#Bp(?GD-2-!W8
zGUEU^K#5DQ=1zecV3R03%OAeDTQ7K)<?P=E#BNPLoo&<ix+{-wrHvdqFf_^?%7bqm
z7G`t<sc19t%CqFs^R-c@p)pcQr%D~}BaQc%N*(V=in27_hG!l5%2;vYl%|$IWx?lu
zqM>BY++~X}y7BwL2yE*=@!RK{hxI!@$vMZJh5FCh!OCA7l6)LqxiU_xW0AVijc2O8
z51+m!;Cq8sl$=jJwJrAO%Yhub(`kRPdu?Rg7KNCuG$zZG0;Xg(7<J2LAyagPlx}`F
zoFo*jW(T=4;nZXhzrq4|fb)F^pn{xS_soowm5t$6S}-6%kR&Il;3M+cGh>B@Mz+D~
zw!pF~E$wkT&(K(==xFiC6JaQa4oJ)5gQVWgu-naUkc2{2h*GEv)TlD6x36G_(0#8*
zdo6<7oCEIZ?>t@Nt!6)?8CuEOZDG?#36-mR_19yA4=dgooW;ysAdxQa7`lDdu%!iq
zG{0Ul)synBy7_8C9fwo<6x;kaWajQ_kgUOqpaM{DQ+u|zXgEl$#HVJs*iH`Gn)HoX
zA>(4JfT>{@`#X(MuZJS?D}j{;sfoxCSn2}O6b3lWJ&`V--SQs({*YfoA#Ar(^OnUT
z*HJ2qm39W?^;(QEP!2&>1OHlbLtJq!!{ARyr{vFx7m?Edfymn|2Pg|uZq4X^JKrUL
zWSR*ccWhF2eI5~^u|6TBzCZBBy$OFZVSNo$vSyptP#=kW;1vh-=E2XAzs%FJH_ww~
zHdb#urfgsjUrhH?EX@JFVb(hMW<~Uk<Kx337x=Q)!`HC$rL!qmhi6Tt6Id-^bvCIR
z^hY+C{e+D;T-t9?J?l@^m?md+q6W7SY`x(E`eWZiQH2hB)i-O3MRBBFNIP^+&Fw->
zGdFxVjIPjiIk^35O>}i0n^_>E1yf0EE9i7ck0F<bw1zuA`(B1M-w8if%e(X-Jkn_e
z?XIYK&tQ4IS5g&@mknEa*yc`x&kLxeH=kd1U!vHM$pF%eI=3M+!YZn~9Im}|xcDQ&
z^>JD$0v4iM%C}YzA2Zzl0BurOc3wM#+A+jDIdSxH;@A&msYy%GA4_zun@2<e?ua{b
zVDBk_&F^@;gpgMfRORvAUX}gg2)esS5W}3q2wVWFA^l*MSFy-KIJqIonr%!fF3V6#
z#E`TGT9N0Ivl@EL0Jk|BJ81P)X(6*<f1ozl!eC&MN)nCUZ$EHeb8e~<Y8_}T9Hq<S
z6~~!*SeA~;dVTyFoP+9Gg1-csdT*B8E<;%NPPi+h76FZL4LlzM)(1?-R*~xlVBqyY
z<%RgfWJ)UN<Syd5=VS32v@n83$r9>D{76h`i^!Jjb=iYQ)|B?Pv(^TTL`6mFalXwG
z*{!E5z)04STP7qscZ{gfJw%+VBoZdMrx)EnyS}$l9n5Ex&c)@bV9zZpP}LdDlCJ9;
zDjM0E)LM6wq&AZ!9Sm@2FbbFMvYv#3OM7MVBSf|-s}yJQb#AS%Kc6g;fn|mRIh(Nl
zjKUL<J@3Xf4X_qx#Kbk~*8SDBT^DdTi!Hwsfkqzidm@wD%|NH99_k%P4*`LKr=<}R
z-dzrE&ytkepw-nIl;l|8oUguL6swkyu!%0B)Sn;st;*POzI}8XEv%QllBl9?6ny&)
zyt~b7uJmGfS9|UWK?7fAkLmFA1V*rn=uhJg!k{4`p@==69Wo+vdMqfOMVXsZj#L@H
zcUr~UHogv>%AtQa#J8AKi9&sZipZ>^{54s3M=c+O!bT9pnY}T20|64S*cZ<P_B={R
zMoT4<+B<9zSK5o+F>+YK|9v+RtkJ-SjMt!h@<CMxDEDL-G7niFrM7{AmA)TfH$Wzg
z6OR~fdwz&G#|D%JZ5eTKbOGq&x8mZrmlsNz*qB~m7OIxS^U)z8kID+7q!DD_Q@Gi*
zjPd?H_I*M!KRs5~yk_vckHe2g^t%JCuyzxPf(DMVnLiwSm&qG#76110+u7cT2w_>P
zzo~Vvf~=j*RpDErbr;?`ijX66hu>#(wtMEgYT;mL@p*)!+r@h7w4$VLdiGN+Y=8S2
zrb=z?#lP@CcVT8?!mHMPr)KueGr{4vL~#xu2zoo?;^k;=d1tbXI{G2OvcTFA@Xv%a
z7Fw=#+_c|AN^r44M0R#2%sri&RUs>T>OxE!sl>2Y{lhS+@~s}JxZ{au&TN<FIdKz6
ziASvGcQ<FJGUaOzKBp1jz_|bNgA3={BazLZE8bf%8(PYAP`^_p7oO0xvCos$b6hL2
ze(3Lxh>*!3RI~`6ZMsgN#Ed2fw|K7b*NPD#L0Vq$d3YoG`nVEy@1i)Vt<^W*H8eOn
z8qy>cs=w?P{HZ#0_ZXixE~MQKqG-uBj4AbASV2sXbyKoMnnm71tHi*mE{+I}e!r_+
z+V{k5X%%4XbbZm;gNv~`9YyV(_htq8B{?p_g~<Kp+{e`OE(8Ej-Jmd<o|XZ82K1oG
zHss~TYaq=az1hAO(pQzMHAZ-q*>{TBbEM(XQSQk2Fsa7QQ!wT%M>R%=tvN9idPG07
z$z^*OWW-e(oS8;!0Jhx5Aljnn%<*=20F@f*;ZBhRMd2T1LjSZ}&W^ft5LIgxb$Nl<
zH|PD~;Om1{t}vVKqR<FM0xV`|mA^I@kBe#L^+*n&+Vvvjj3ZcM?9#MWX{-EVK07b_
z!>mbf_2?tYDhP`<+bMQ|OsmIvcGNrExX*`sa{KqKc~m~MTFRK5ZrI-o-6eqe;ruQ7
zwZCSiAq=sb14Y}#sdA66QA<jcm3F@_!qNxuyd{VO^^wp9KwQuxr@UlJIN+tWI=1wN
zI_3_Ro}ifu{$R|NUJ-t`qY7FAPgsAxq|;Mhi8(<C3)eTEB`?{^%_0ORC2rVIOehfk
z!`AA+Lj}R}sw<M4WM|gO|Mv*8E<M<<BQ+PwzI*nSh@9iyBM+B^xZLlu?@2v38IWxx
zBP+}9{S&oPR!bSzmv}0l6uirvKCc$ut4e_*6o35pDc@zbdZV@uLG^a{TGY+}fTOdr
z&D?M7)vG4_+|z+d{*bgCLb1>r5mTnsbBVSn;E6}v^@IkE?fh)U4;r`CZ8N${ndU67
ziJ;@;%qvt&(JhX`niLx|mdDEkWqa7=-3;ay*<=xUnnr2wBQQuF8X6ydk2iWMEEVns
z&>Xam|6Au#fHofRaZG+)O?6)d=XZ~%vKq7isM{+~U}=#g6n>6ayQBw&gAkX^3k?F}
ze33g)GP%jQ$_7)ZfqvphCaZUAg{MG`k)~9L=~@x84s-;xwbN{EGIkG;-txKyetuwk
zuL%jR|JSIGyAzYhbKAkCYi@|Xjp#~_s%HMh(i7@%gN~cr8oti=VfWRpC3JO)Y{_K*
zT_p8W7witsRE|=C0qME+^hT)+H?5ciYy`mXuN2${S9OJfAWkT>ReqhaGU3@3{S~ZV
z?$gz;ii-34Yl3h-&Hw+Fts1KWXjn816{@h(e*114=Lc{kZIPu`kSWAutlxNm7n6-}
z&CuX33EB+}50KA@2-21)^*5XE5TKy+ZJ}&I=__;-J_e<J3!XQq_$k{ga+1lmoqIpP
zK75)!xjoy-a!u{~>OJGmKkk0?Y`@a$kR9c62sL*qI6?&9Ee{j;MnyI|mS|H%i)?T7
zdXY{{PV)eI?AzK|LTM#Xlai8cq-VeJ?y$G}PWH8vH|{|<sPLRDE@OP)>+?fVGCMvQ
zwPj*WPRSn3jx!hengzvwHoEWud78+*Dju~FHj(-)R-)l`3-TUFkbLbGc9lQPy9~I<
zT3{B<&qBXzRf&Q0hM?IrG-a{&IjYT-Z7K6%#5ZW)PGM+;#kCkOe_Qr*5qZbMTIDqb
zvOl~X9c{mD1{R$yAxIWaY0~mvgEf*BXUQfX=m2ax)sB1g6vGPFU#lfh>Ok72Ae$+&
zmB##_&eF8UU8eR<xcgjg0W)aVTBpzlDPlF$g5tvt%PE=1^ppz{tA2Mmu_wZ~*VQWm
zwZdryceSD6I=wgUewrjXqQBx@5u64ub2&$-;9vmSh3SGJ&mdJKLJ2X2wys5v$R8T#
zCH;=e+KyiP6#9cuLowGfRlphNvw{h`X_QL);I8vtKK#Y1wx7lbr+xQs1R0jKq@>%l
zoZK{%N%Un+8#r8W*|iRg4Nf5Vd6vxAc0h2AT7JH3*#vIxu84w4Sgcxp(KZBIw!R<x
zNT<f9Vqi8%&}$n;diwk4cQa{2S7Fmn{GI2YV<Z?I7y8;kjs0-Cd#p%>y5@L&iMO7s
z)M4e?iF>5#=Hc<m!8kYD)d$E<0~*PARzO=${|UX0f<)AOde+PXH?Zdta$RJr;~p&#
z8BO@JEYd`T`&9Q=NgZ6Xc2UT;T6AX=$ZBPPxn9pZRulSx+y}5jtU&<M<$UYA@7a%*
z0l%C8lsr_iF~wvt?z_t@qFqR`nT*s^{U_PMTGS}>!ihKchMPf;H3}-)NptatjZ~M*
zyX70?<WLoia$G7rWV`4DEJp=0{phvW35KsJJSp%=+O-b47d<^a9rVn&c3#sJkQQj7
z6mLUlpq$*No-rd8+WGhxP@G+9uyXz3Eyv2QufNfeOt3<}@6YBo4W{)@sKiMbzG{0Q
zP$i{lGat}FcKS7w>#6;qX|uhie{*k7;@&Ab?l0_-@cy8yBkf-`Gqk3^U(erTm6T2g
zH9n|pQ|Oi2+Bsib$*i^2XFaJI-a%SjtWHg9jt}boHKU2gOomI;VQyFco30`LJA)V0
zuXkBPCaaon;jaZdO#Fn#UmJ06>j;$jY49V>-1*PRY7R^1iB<JoPp_s|CeT|NQO3Z(
zbkvaNH6vFCCF4fAzdpy$2UYwmzts$NR>`*BVCS7{QS!g7e>{U-AkB3=T!5~=6CLjE
zdF*a*!UQ(K$DldFXjtxv8%}Bimfi|lp{cB~A8;pw1G_<VQ~f^|b7Ji~GVez<%%!H>
zS@NttLy#S;P%zNHNv*cDhrZEM8IlBtdzqByu);r1xS03SHt(;VNP}MfzpF*Hb}v0U
zyR|5|pv}8>Zx21w5vR37C`wH9d@{je-WRur_3dJ|(Lhpizr5h->(Yl{THctS73yEZ
zSpm})z6{1$$Q2J31tp&R(2;r6*XPQ9Itya&nxK9@JB{Q0rM*s<bR{j!Z0cT}$xLWS
zv63_q1XJ_}Pe1UsLy)DEaIM|il2@sxg2$ufh!i;f40t%GND2<NiEo$r+;o#R0>zb#
zES>{E>DLRBo7b>&-oJ(jTsIU<DU3UH36cwNCV^07HW@^JrFvYOv@T`l?gc~h`;aPZ
zbt@F%*4JNE=Febm39WX(GafmHIyU{<o)q=A+4GmCTmOeW00S6AQ}d>*TrBhoblWF?
zfy3FqEA1ZO<DxcJz4Epr&7LJQ+oX2Ym;8y_b1}~`Sz^d%B0BW#J|4i2wf7{70d0U<
z6QN(shxPlbo5DUO-1Rd?X=%7$&W@As`S&bY?s)PYCFnhcLQL6HI(#ZMd)XGH2GnDL
z(dsxmrJVn*DuqM&sILax-w$uY>DN5@)!KTRL#oaEnaA?y^}Nw3Be482!pdLAH+oEQ
z(wZ4?$w5P?AkO|=CX?0F%1ONbLh-rCx4v&0{v-vh_9KG2nHrqxJ}hx!iwXFmB(ZyJ
zdqjesyg;?XlFM!+=Hh!34C<Q3x_(RY$hd!=()g$j<%h?mytIFx4Em3u2i5k|)&Fof
z=lXVZs3Y`rqiEQK|3ictwJd?`*4$qwUUhO<Aw_lj!IygRCRyX(pnERKd4enPc}zx?
zM)+WP!Zy;{*pZ<M4k-j0A6bi4{md3h#s7Ph%+5la1}}Gx$aSRbz{&`P=OQ9d5*R`2
z5X-X2P9d71W$-3|!2p+CdjF<Ej0E#zxL;*t0?dmk^ZDWIN--@w9uH<)iPx3%Vi2M$
z{Z5#DzSGOnA!`cBA@qnw&<Qzk0QHl<GPTZ$AE-C>R#v-Ed(mq3zIrNke)h}*Y966@
z_{YSKyfi6eIQ;*u5SM?P>jM_2q_CElxt?=FONG~>YVAMa%UJaW{Z*2Irj*_3ztt}f
zP*p|(e}0MgU)UnCCkyy-X{u;<P-Oqy(0A&`$ADjhgo|T)4gZ3K%{fWr5tgN`x$=dI
z)dzg)?Wtwsz9_V8t>Hi|cp7+rJJ(9<TXBO9p_&Tea`_;>lo3ZToz2Op0$RhAZH^Te
zhGu(QV;V7H9b4f7&^Ralc1KOd_4_C1H$c+AqkEn|BRgu#A*fZx=v4i7+IyYR+>f5Z
z{w$vp#L@TFI|+JRAALQsK6AB^&(i<||F?iRfiPe%we;^p(KL+XvT!cm0-tiv&qS!A
z4;?oG$VT_v?t;uW``uDZk`4Z}U~Nm0UW^-YmN^=%EhZnQ|8jTy#lmp{<Z_#zJvw;;
z@UQsi&=tbe5My)7_(NY-wU~hgV=ThRM+sfTm0aLM&5*`?DHBt)%F)eo-O0C!<H{73
zD7~efrRs>z0<{GKq-}}G&~^AXvjfPgYg@BQd40~A;ds<$(qO))SBHQl%lsgn`W7nk
z=+ZCG%AOW*Ym7aWegEp!BK;f``m{<(9Us9`ep0BsuR85sW)zg`;VTZ3GulSL1A%Tc
zZDZ@`5Z$!G{;te_4+s9sr~2c^q39b?GP(hi11G*-%GMHvkZZpIMUdCfcZa|7L;()Z
z#GY}I-8kGaW?~lstCJMZ^h`C!b$*v=c-0uJjRq?{Gw)+>o3I-vg?N6G_y&5TV5vXS
zP;9ggu1DD_61K)i-69*|m}hs+j4PmNa80(W3uOKmZ+V!tV=MPtFPcTm5lZ21l#`ne
zm;<-k(1}oJP)u;U=b#V^r^phyVPtL)Uzvzt-}tKs4L1zuS=D%1@jqD_Te+XaG8&A;
zplLHbIcmVQIZ*MO`131%wfG5+IgGc9N~i2pi`gp}q4qkVmbo}K4HZHN87*dwPlS`g
z!g*9R-YE_4XQ#r}9NYFC-UQ$jkrs<Csr^x?T>G?F#BXFac<cT$7PCNw<>$C(itO+Z
zjG&-lkrhcJ4USLU?RhU9<)GkiUf!J-+fWYUr2<LjF@%oKnPvoElpFrHd`{npR(tVT
zj0zb-_hrq5y0nGL!zbEZHEZJ4gmL$zhvB=^uCW@~iiee=zW=pY5$Bva$Y}kit0_z?
z)=$e+FkA7hQt~uAh)))}wgn<vDJpxWp{rN_MG_j+#^-LVicm$#c6c}fF`vn<$~C<a
z(*TnpiGcETXWGntgRMsK9s@R(==@+ParZg#T%}_(e+qbvj#REL?GyRaR%2jpCNGM(
zUVQ8|_ca8yUt(hOU37M$tk+7b8<#Cax4}_rKA%U?nSD2^jjAvhhVXUZP+Ktm3<{y)
z;RtwP`s2Mgo|FcI-fz47)rW(qEYnp-=RH`=*APm=*~<$1g8g@`YiFFeB~9#|aPikE
z^L4;!g;r@Lecs|5y=mOj)*vf@A^L&`KrcMPlC4f~1@0O0@f}_juF*93S?^8DMY=rZ
z*45$wgL6h6hux>umsiJkm)5c*UeqPrm%!!ce0NSA&Xb=C;MmCb+TN*&Z@oERv-zY1
z^ygXjGiv*vKr2{R8`ORPm|@uyyo?tUKoEp*t;vqg#@`IjpSdJS+yM`;UlRS#IY}|@
z$MEs%v9v%+-(I?i=J}bOZjJ+EL3=C0xOL^+t#Z$%eJ>ILct2Bb-068i%Hld4@4)}Z
z)|G%mo&SGblx)(LQlgZ|ol1>sI>=T?NscfoN69&Ij4(<Y%SO3x(nh(4lpzeFWX6$|
zkYk9ULX#PcF~<Dg-_h<?zyJ5yXIJ~|<JtMl=ll7*->>T}1}WVG?a2C+%#F{83Vqv?
zfP4H348%TC7{X7@K3LT$a-b8sVijILC%Sgjmd?x>yT0#$kz2yvP02k>Z?O}kBy9?g
zdSejwnm=AKHb946<^=o_O6uZ%Y=Hl@YuDBnFHB%gGjK>FDE6Y~za0oaeZ2&P>t3g~
zV<o5N(lh8}xpjomS-+X%leVhpaz3>V3#qcXCHr8iZUORTecG8bqCzCwY0w;K?a&3P
z0;y?NVOX7L6sYZrcv{=8GJ`~ITXJ@}pSS^`^ONT`%xMDY6dtzeP7w1nqQTe-=`?{1
zrI7)Q^!ykkn(&&~Xb_094Px+6c72Tr$<R}JflCQ`s)iPfr-*dYCdO{BPdG|8SH98n
znxzHs<nlPD8?<@eB7Ge|8rN4^N};H~4nQOT?>y0b6&qjw^+Brn$?FN1+tHm{3pbn~
zrHTA9N}V7LQxPNRxk=`&XuNf#DZI5y&fgg3#F?)G$EIWbMM=^G0~b$l@3Hw}N-A{I
z<Tkig(@DJ=O6fhGJzn3hUze1GiLPdH^VI;K(v*zUK6sUn><MX1YsuA3LgeZ-8ju3#
z(=s*fLVg2f#?6vm1?IMk{1Ww;UdQ_11&XjS-M<UA#{d-)l{iCIr+pU9d+TKS_?&vC
zO91?W6}g@GNpC%f#xb8pkZWkWB<y+$oH;miK2z!G{aTEC^(#>juL$$m-P5@+8i(21
z0*=;43%(sZ6`2oT36cwOVfq%h^N}(tp|N|5Dg!tQqP&H=YH6|4oUv!sQpGFRI|?Ye
zyHkc%jKwX1hIe%0TEP62)^-|=h5irG9<!h#yDy>h>VaV&bwV-<q`@|$^o=J-Df1lI
z8dmk(6}DdZgXk@|(dp(IFF|Kt8(k90nDQ`l#J@oCkga%m+5}RCzGO!Bm1_TD;6^0)
zW8yH?&b*S8cK_Z!&I>Yonl|~j8P&I^X7BgQ>~X#C%Xmygc~cQ1yKya}ws!p2VqAq)
zT440<1UAo$-<hJ19hB(W$`trevF*cOy>cSUW0qtJq==hHHD#+82E1RCWJAH2jA$QV
zMrle$RVAhM%Cqz|AW)#OMDD1uv;R@sHO{ot&9&4`^I!&Dz<2RPLCJjm8K`$d@T>vq
zov!{vnDiAbcAXUsm!zBnFcv5p9Q|jZHk&D{<(8HCm2C7x{iO|@+&E7Z6x@713Vl&b
z&6aiWLd7z2xlNPWBNK*WO4`dxo|=PTjQiXkT2@`JL(I_clD@RDnuA#upqG#+)@p@0
z6Ao(KGrn5GoYdY{%m!^+(r5iHzXP<fAY64&M^F%|a-LN5!-5svMfiS=c72uBmUqlm
zgt|I<(2>h1mrBV)wrgaOW3(m^+UHV?;97J)6`o+Y=I_p{B=%m*5)gRjR=1QaD&-AT
zxn5=C;}Ph>IWdFZWQc*LLPAj>uS9g{3Uj(k)*n77@lLs})~TT6oSq&pVgU#c4sT)s
zu8g)>p*#4s#Z`xCvD=GjMc97vbGN<uGE_dW<LOzvv3TWfUtq5YUR9+#GE>c8DurI{
z418uNcyZLJ_?9MJo|i6BLsG?+?%}!7P2)$;w=vjNlE3+Mn4^Z7d+LU0x*PyX0-8EY
z@>Y^v*@3Sa>KWBH8)8p?8O2t_P-v_=^WPH^mNkFBnTl4c=fmVy;wfzvp}YK2r|DMA
zjRKa6U)TFT9}h|K_lnq_*N^wzh->ti!wPsGOmW8gYn(r)ZeeSa{Ji{Xu5K5qJkSr(
zfr_;sN9D{GHWAI^MnMp|NzBO91N1E1MbH9{_yhxAHVSU!VD*|p+|<V?ytF4rfSbEg
zuB3u-d9)&vfyd5d=;?_pl+@zI;q;?R&vT$M;eDJ1CGnBq9w}0#LVjvgyLZ&O)rxhl
ze7GY3{Rj~>bms?SEnhs72yZ@=GtuijO!dU$@jS+a8Kt@tdK}Cs)u{$&4#Wc*vqsib
z1r@2KshLbSavZ93je`{9j?mJ&tEH0KEu$DWY32;0dYL%Gua04Ms$A#1u2gZr{!ufk
z==TeU{CubwS*!hlIw@IRb80!am2-UW*BL;C{i_(k@Zl;&rk{pBy<(y^{AP}*EsPX9
zy)x#7N|I1#4&1m^wjY8S0TpK(Cjd4qIwEnR#e?CxC?n((lz{2T@HGbas_~#E1v+x9
zllN711J~kExqZ#p6cxqwm64gl=kbU#G9<1X{3RoPp@wERP-wAcx2jM);VSB{RHE<1
z<gKw!d~NMzfbcCw>B_!`P$AE7W=+qR{duGj6sw)R^xgzyrrK;!u-XQ0fZ_?kO-y$B
z$yOGY(s>dQDY@Xgw1{=gJW<PH+T;Ru_8<>!V#^M?j;ahW%f#W(_ztis{`7e0z+E|$
z6#5I5e^o+h>>O9?LLMWxUazuMmC4P1do%7zQ}&@ij=u5sN-O%!Q4Bz2o_^dHD&?aS
zl-#rZP1(l^wbl1S58C?;wOUk>+cPE>Tz@X89ZaqrWPknIs3vofPy3?9izL(!`Of9!
zfVswV=0NL$qh5VzHRr3Zn3w^6?4!H>xSNSuNC03i*8Tgse9cU7sPW|$w12fjh11$b
z#Y<?tLH1=D2}c@xlv#4Dz;o)nLS|k<f8(h6Z6{w@kz`H}O*UnFT|a>NHdD`gb1F~l
zCJ4Un+;hOoY%R)tbe)~0TPQ(TRTX2DOsB`J{oz<b7OdzhvV-=f3B}}8VU2Cfar;wp
zi_V-+$F8@C`y_<4B<uI2^<ER6OTpbe{PzLW{(VJzq2KrmuSI_Rg7T@BiF|KiUl{9C
z;Siht0jl7`*8&v^%cl5}?|Wkasr$xVC>HUX3F_uJ=+&p-zci!a%vt8(;n5l5Mge;F
zV*kE?<O>gJcj0uhy;xxbGFG@dJDjO{O0PDIJ%Q|ubXtqRWj7}}Yn;VeCA)D4uLblt
zoCJ4*)fZKO5oZ~lOk4uRHI9YO7%iXDrW)ej=@Juq6dD3?9fR1S7>UF7nM(20>QT>V
zYkwZXdHnE<gOSnhsv!oD2%q|?3#gng`67;{7&F}o(79%H3YNZ2KAIheQ!9^N;$Sx#
z8#I49!)U_h8zi`&whJtYRHQlx3q!~SK&lWzK}P1EH%i^2V`jP#DlU?JQ6a~v56QZi
zZ*HXiX>n`i-c7OFTWrwIN!y8V=`%_X6Gc8M3ngJ!^$ac=Vr&WctpF1eD9+2sM;fh5
z3k6(mn|q<PwZ_W3R9#a7bVr$54Ld4y*L0sP4+Jsmbl8OG;0V*3?Okz|K_^Kb1CFuC
zjY6;Us!;a$pR+^_m>(C?pbNR>Q#)u(A`*??>3EMv<?VBVkDm^+3ofh_f1#5c1Xhr1
z>-V@0$1+%8&G@YyR}?rmOUhBJ-rA!G(GPbJ*_sAS0A%~{G^iLp0K5s|__FA$4<)UX
zt1E0Q(a7OVf5L-UYUV)8!Ydjt<9>^^@apDaOM1Y(`byuP5<6V{TyEx*LkcNZzH3t_
zIp-0>2VZhEhEFh2NADyuC2m%!vVB_|7W!RdIK+T-r<%LX@fUi#I}Z13nKQ=upo#50
zD!lY^8!M9XpCapDa<2UFeZwDuE1)XW=zo6nai>|5P&9Q#OiK3bZH@+a03lf3Z8RZV
zb26n?T~Ax4MHpW~6H)~kx%{6%vke&hxZlI})};zGt>1dkkhps5Qs0EZk}n7P++}21
z($fM*%@POii%2?ty_=-HIzDO$Yr-ydYGgc`G}|sOPyDY4Qu@G>Uc@##IrE=bJonqu
zdt(DSJI5wEAxHgB2F?Vz=6a-&Pd(rAu;hM^EHR~|WO9reRCw|>3IHnfMb4`Wh*0uq
zqkw+nq^DHW>q))-MWOe1AAP_FUi{{cWTPY&0$_JW*e21he*>Ky6x_>vtGdr7yJw7?
zeSy5U0j=7-x@l3Tva3=*6D3`U@ZlR4v3^94G6L93*slW(1qvp3^U_WeaoqZ8#0Sby
zcPNuTn$bd9C=*$&5*kO}oK>M9DO-7>e7#P~L4+wi$f@Qj^2od){07fCX`TVE4LwO!
zwM`h?6<aBR%v--G_HLxX!NLa>JI=03!&)cS4#Rbs2SDy;CMx6|V70xX!aJ5Z+`BMG
zb8w473Wy&b3uSnpFBxTw)9G{r!%$lC8G?GEk(dp8Uyl0nFMUJZXCJ#8jHM}<1+OnV
zBN&TLk22qd()U1VzxE_=YZr$@O$K;6T1qZoowjIJ0Mb7GW1zqx-1@Ei3|};woWso=
zSQdSRoL4Mac;2i=g(pS1KodFTFtF}j-ggWTkC>KzOgnJ%o=qTogux<Wr_%H3{VJjM
zUYw7R93UZ%*hI2AA2w-M`?eP@f62rl@zeSQiqUYHI`I<aWN;Sk;01P|bH7I951$eF
zg9QPt01hd)@j;gp^`FL~SHz5!?7pw4dcalTXjedc3D#9qAWJjl;@-N??QotFg$aF8
zfL)wzk89MnIlW(29xzCyy0p<zf^9Et6trMDVmBMXxfQd$UwbV1!UK^}hr0!kW}u4K
zIU0WCH?-2x)8*qpM1|3X;+T`ogz8r8M}JGf*n|*r`7I7#Dv!*k9p}@l&u;FUuwQq&
z0AR85%CazvQE@q9j6`qK!-;-JaB#xE8E=?Xxj)<&Pl$k&zxJ$7nlA0&fs9dyxS=}Q
zWLDnO;3pPA8Re2E{%0F!N-DIMXPGY_sM=NX_-Kwae9iDZ%c0^fqqP0_L`~a1ns|QC
zwS6g?-gACcra9-Hm+WS#|0(S8Fto(Jz6V!SrljERXnMz7%0UK-2p^a`I{}*1f^(c*
z04jWFE?B3%XskrE3rNHK>)WyFS<Q=POX_;Y5P>H$Z}NEc$A;x)e~56Pr!&31f%(X(
zJLH-}+~G=*VdYTf7X_ZN`R{RQ-cI`&fwup6Yt3K~%>LC7jW3oFpnhKCpDn?&9Ui-Y
zifwWvm;7$NV$}DJi8{RRhAjhu2uBy|lp14>vEA^SN@qt>jmr5oGNTi{q(uSRcbme*
z4T_&vKsbc7!=BeX>#;WcfDU-ir^BvX#W~M&%pSL9_vZ!d2NRDzTDWGHyz_PMxD(zQ
zG(XxH+b7>y7>lfgO<meIC*9>Kzz+NSkM0%=zAAt6Anpg5QjN`U+dGldDxraU|BBsy
zthTMfa?LnxD1&IOZ;{kNhP*3eEs+}mKZ!qeYDc<m+VA&7MIkLHYnZ7`JdaChj_;v#
zcfaV9jWw*lsC{pU==#g9Q9lO+=?%Z0t{O#+Z}1e*r?4haUz!~RL0R(C)8zuO?(AQ{
z@3`~BL=2Eg+zMWq*<-l6ww&YFy=_;QRTO+&`zb`|m}$*^BS+J9r_kwlu4c!iW|djQ
znU_3nRD1@9pdz9PA0T}vTd)a=9N1Me0rDM|V%_zL*7wk?S{tg8yu=OeP6Z9OebGKo
zlss!Y-S$;P)bXwdDu0|#cAx`AmXkw$x1(hN&zq^{^5&$WrC{+sN^=Q3cktGV>Hnpa
zu<S{BlZRx__6$78HXz}nLfF}NrL31FNlJZN5InY1nth=nfg{~4ZyjpFvPR0#9ykFw
zrM<=@oYHyCpO!#Cpy&8O3Hho|5F(?oBRk3~d<^RceP{OWip<OHIEz3{CCBg!gZ2a2
z*#xL<v@XsS!9PH&|GN+dh&)fdw?>UHiALi&(kqJ<S6A~YC`9+%4u2mAsslHiA~csS
zSq4!Hvn(+pbqxUw$SxHk4Ybo;d6BOiAMLL5iHz)z64{{=I{!s0u}t!VmF?Vh_XOIg
zWq%9ODB=>9bc?)N_1cZd=OFu5Cpd1ilurulTeWL+MyDZ;rN2!=KChQL$|<tm%=(`V
zjrg82D@T^RB{H`;TOzgjRKE6SJ`*k&7Ust(1!Z11hYP#@PMmit(^(_V>!A<T6a_${
zSR|++VT2;0+jS*XTX8)k7qu)1q~~`Ul*?UuapY4$;G00CSYYX@zsOkwA8ras;i&J}
z-XYRY>$ba)!`ZK8wZMHd@%#mz&8kw!S1PhrVFy^QaUW&M9*uS=mi(>7L$YFf2C&rq
zK5#pi{8^f!H*TzkYPcgIjCt0ZuB5soJetF%c@|Ep{Lj~i2DPAk!9QQ$qAE8f>_nhO
zhdGS~6uLejRNaU~%p<V9vt%a}r%<EXa_I^o-EZULMzmL{iGF_1D>&EvO6yatJEl5_
zt2ti<F_P;RJG)<i>jLfvI+P-iUKKTh^eWw$_cb)DbBX*J!BO?UQV+k2H89v4?2Fm1
z9n_13b|kUFO=2k}lfmdrIND@=inI`4_nLxHz3Q=-w;&Nod03)Uu{(o+Y~ItPOjfSF
z+Mc70NB*#m?-}dv{>pHxrsuLS3q!nF-Y2|-=RBH6Tqr=Z!8V7K#BMw$wl{`TTUk}m
zBbGUjwN9bCGkhZdr;8I9(1bdLA<>e(HjrZv9d-4`I7`2O^$H71$>2BWr$4V9&O7_O
z<Z&TFoZdgw^|`4*r#+|SaeGb-)(|(JsgdzPJER-hoqLcVFpj8bLD_^zt=zpU_k_qT
ziuix6*+rT7x&SuG^Hb0D;FaGUMILEm=0#>X$>#z@DYINAqExc0dE2G#r6CgQDGs_(
zg~CeiC;d>;y!nb@Mj%GDf@YWMA&K4@(%2&&-aKIn#hQ)n3kP;3VyXyh?j~V6`)%uA
z>eqX0n(SP#s;@m<?~D1gsQ0h3rC-t01Bz~zgO+R9gGA_p>HK8SRy!LkFMa5JKt>8>
z{#$y$ce}@9Bb<Go3=I%?YENLQ=`!(?^6L<dR#1HaeoiGdph3-@*6boYZhfp{eNvuB
zXVG-7S#rr^o+47n^7*CU9_N46b=A!*r!ITG&IR~EkGwOg0%JaWUwf7<?`5Yx;UpO6
zJmN$anp><eTS-VRdnDjTTa{H4<|~v#<Crtqze^H?6`YEEZCa~EG*!?+>@v62H|*Um
z`IM?=Njh=PIiX@*@C~A3hNKe+S^r@Q&5c2`5AcX$Av`9Pt;L&dEp@M66^n^<-y4=X
z)I($aW3v12V|z#=+hS8ocGq{?0**U3V+p}nh~=3NMr>Jnm0Tpy?(#I6q%OM49E<MW
z7dN6{I69R1Nv(YuFsTUpY@lZY1IdGpp5&^HB3?;Ei4PeM06_L+Dh&=Vnh^;YUS0KB
zFp;_36tNq@qLwEvX+%;flCYEqU*T@2c*dr4q6s~f!uF_bav~I?rWW$C-d4-*X-ene
zFP*8f!woLiLC@8QfKLoC?u9w;{{ZRkpW0tCK3trsDq6+(oFOTyOcz2F`4P4TBzo?o
ztI9;{|J-%*(H`n1o7+z#xQL;2!X*gj%<GPQ7iR@q3^Xh?QDVloPx_ZU{v$^Y`CH7i
zq0#&{Rsvz?_AmCjYy7pOH*P3kWCKDhqY#C4z6Qkn@<mHF!~p}YL8|`l$R`Q7#2|98
zurTnP<IRTsC8cyFj_L7vb0~Pac1rx+{D3IYWfOlPSDT__@+^rlPR3f`aOJKyUc25S
zREY3s>X;1NmM=1_)yP3nT;3BBP&M3LBO+H9)jea!02h}zMt_PUmJ#Idt%j<59dcOj
z0Z8J-Cx?%>U2QFVuHctS^eWe0L!NAV5@3V0<+tpxmc!INtWGwogeMs6{jW&RK&%it
z&>x;)7R+W2KOm}U%m-F6n1RVaDSp`l0-m`2v1uOadRJz0MEU)#-Zet9RoqyjZW3>6
zD2nF4M}g^SBHN%3JgW15z&nw>fQVIrO5Hanj~LxAtco;U>JnAqmSYZ+WG(K1ilM5~
zT2~?Ca2}%2skE;crs^z<maqcNl8WV>%J565Td71FT5TY?(^v4)f$8b0N<5h>KA-3+
zx=UKxpfW(BZA?F4ru)NK%yGSE&sWNh!WI!XyKQE*X4~k5BvmOY<eV!iw$hcrG};c@
zH;35TVYpOPcUXT*7x-@pI$2=h7dWJyZd9xhXcM{nodO{;Dk|!znuJ}q_JAe;hcaI8
zk?>P}i~WKnk0lh{M@$vXjS4Ey6?C7S*IyCSehu<PQpDoczORZPza018cc**EGFIe(
z`3Qj(08^k2l@Bj}+4m&^5i^5Tx_#xZTb9+#+51fd8tmVz_xv+8g^@pC6XmQLS`e=+
zRMh1nk=*nk8<GVK#He^%CqJc(287DqKK@*pF`c@-gtcNkK0}()L!?c#dlO?)KId*5
zfcr<<N&ZZFRnJp9EGZpuaJd&_f24Fjvtuk89Qxe!@}8CbLcG`0llKb6p3dtx=qE}>
zjd0?Pda7L`Xonc{3XTGq>lyyH{Hqp#zk#%n(f?<&jO7`QhTDqS`5I-sLl~5&=8?NQ
zVY`Ux2+P&I`nSjsm2zQUc9t?sTX57^F~xaNDRHJyJNZ<(($nYS5=!a1y8M(DR~f0d
zBcDRevl*2UpZGPYeodKt9GhAkjfd|^KF(@l6kx4gN~Fi%J3E&aH6y&Yj0~bw`Q~>)
zMhA6DZ{#MrL~1Kk257jWW|Itx<D+B<ngNvs@yl*YytUH3T0T!-o<xxixdiGx&QTtk
zXDm^8ZGBn_ps}MqtlW~76yBFo28ZxG{6p@98JVCOy2O>(wUgdMN-#sz-AXfB9>4ZT
z8M<*z@zaEbxs1q13wNqHzPuq`vr_E*xMjWly20m%3!g~plAJBy<(*=Br*cY+dI+x1
z1JD`f?X<tkHVKd~^EEk0c#9W%4Dh&|f6cJ}?n)0xQat22a(|e}^uV?)Uvj}~l^it&
z7tET_etrG?(&w!F{U;+Ubs8UsQ)+{}_h%EFOp8uQi))#=f&wJ#8KVa=o~z8A1|lSO
zuvx^j|E#gS%f(B<(~tgT=YE%JaQ=@=3Nn1>jw&jb9H{)Pk>Y^*ExER^GVOjg?e6Zp
zxZ1D$dNl87hy<DZ;L}Y@0h+qjRVhzm$;em?huur)YW3OztJ2b`-%oI@!1o%t86+%w
zO`$)~Fe0HFxt3BJ2ZDw<&nvA|5>n>omjw4=JtrMiWwH>ojvv|ql32t!j_7Vajp`G2
z{JRc|Ke&iPWvkYX3-g*gn6pXwT?heS%!4wYvv-9U?MSV>s3^QkHLr<Ch<kwed&3-!
z_mB$=9Zgxri+l8%Md5BEMR!2#6tftQph^k1P5DKGr&YZOUxnLBY=<+RjC5b<Z<#uY
zWwY6HT-Bg|y>&~*mkthdiX;!U(9CYw&+egYFA~vF7OIo%t4<P`q**3FEa>cgd_oj`
zUMajmBGWEqonZ^tJ3Y=@JgVZgCuLk=`31AmlaYi3Dr>lt=sKT980DXW!$HGLxa~y*
zT%z-CHZJy(le4X8dH=0H-D`Gxxu@F{Ya%H%)b3w$x>E4esc?GUwNcX)hr7kerg~?D
zCH$3+1ME78uC=%8oOb;5=ch4HgV=JT@Ed0ekco|Y4-aHKw+x;>$Dp4rwZr-!klH)(
zxv#0AIIeBn`^?nrmG_$9{Em2(_+bMClk87vP7ouC?sAeZZ<`&4NQ#XDvj}X)oEew@
zCI6jwhB+YfdGRfzmf@ke!^0x+${BYWJ8`##`JO4U7t8H*G(f^V*xFlNrWslBy5{q^
z^O{<F4O6#`3Dkt)PG1Kt6dbH#5nKxr$?{Hs)P7Kn*2~<jbRBdR=u$Rc5d*~`dB@X?
zja7v|?TDfzKBjR?d9P@$gN5?ZKP7&%^&KEYFQMUT=zSB1jcs0PR;||2ZQ6Z5VtN1(
zSiY(l6)M8@lM29{^dX2BBD6&}nViXOhco;t^;BzJ<6?EdwTs1c&pe=DQ_c8@H5fX?
zlcMsd>BDC`pcvx6SCPIKoB&RW;^$w4A#UxS0N$c0ZPIVPokCn<5%sML6r~dQ1sq>(
zS-L<a0i9_pEK_2{_;?}x_vX6{>>C&JDD{ys-KaU+4f}q5wcH{~GBj?;rMCk)!x5kR
zvvAvu8dNN_NOC5}DD)kSnL^BT5@AS0*6CnU$LrDag3-)eFFHlQZcC+VV!(HT_0At8
zX0U=9pe4(@cYpYm3%TIa5FbV||2(I_493dVm1A0RO4r>yUm)TYRLNJ;rCINvX8I3P
zE5PSrcL~m)HtiaYkJPYocrfF7Q-is?0jlE)O6N7#*|#$%yGpuUj_LU}dD$pMYItnQ
z%=nkpV1U{!rvz**uaDMqpo5N|1WzO4-bNszjsKW)J*k`!ZReQ5D$w8-8+d1?#@9Jf
zh(vBa*E{5QP{1GRKJ6FW@j9=U>@Tjtiol8MxYyaN++}*qyh`=8i83{gL#TY8*Oli<
zG|;6ooHx-cU?r+l6I0azm2~)N-wc%AiYZTecnBMyE-JfGvZhLvev>S5Mk?{<gz<@W
zBZe;`i@<9|cl1|J5%zsgzGhG?#c!d*{MzBR(<d0MnsPK<U(L~qyVg7H);o1&*Os!4
zvJ$7**TDj9YhaMEpa4?BS;j{VYQ}{o5r?HFe?EcVr_mO8J8IPFWc9{lDx+;$T`rU*
z&uFyLMh~akZwjNGPLDX536$M`A{67Ll|ezg09l=XG|MF#-otA7bd-@0uEDlJozh9?
zJD5HkT<#llW4M><FSWErqry!M)hbP6@BI0r0*Au(jB+y7bCfSvI{Y&0?B|l*#R?ml
zyPw$7VlAx=EMKJ37b5w{Mq89{k#M^xqON;XmYMIt;C8Wt8T0J<FZ6cB(b$)9pIw}L
z$em2Cb4P3+%lY6*W+h*4iJ2n1EX;^PAPU6<dppM6d65NO)Bp<Pck*#UeyR@AE?OT*
zt&}|0(jx)&CUIi$*_fct7MKl!OwXKFL8}PN7r}`L<AI@q1$gsGZ_wzBHj!R*Lc+S4
zbpWiDlF&*KZ7R|#lNB-5FRix{b#6s3vD;{KATjlO*W2@-I^-G%)f=f$kF7kAB7@c3
zH_eKAyiXKTRb&*Nv18+yJxaUJ52*5NUvw7j4`0J`3phA6=u*v_@$SyPo&LCq*h!xu
zYF*S!ez-O}YVRErV^KcO6}BnygHKv`|CfhCd>yjPb5sXRY0p1mYz=SEA}~bdeMnf$
zXhFOqD&JW;vpQvE>KbE?Kbx~JTe+5Y=1{(cgp$>ojkJlpn?<<znZVkvdES_2UW!u>
z;rcsnE<WQzK8-+~8l@-CYmK&zS-V+nQe5_><iQK`(7zcAEm7~?d%4Rk62{!jJ7rm?
za2oT0LlR|ox;uo#Xr#GdqnM#)+GJ_oW%C_`56LziFO6sm`v~h=9<q)L6rFV2451o)
z9ULNk&g%RY^!4v{0=EpK@YfTze*pOT@*K?wDTMS?{s?2-lBROgNJz+0*rR0U^`-i4
zOWN*;%C)k1?Kwy%vnOr6&i*OU9vJ=;fM@%}nr=)mjl8Z91t(k1$7kJ#WpU|#AoKJ<
z|5weqGr9auU*gsB%*!-kIo@hsNj1ZKVcff($K-KY3%1kvvs3R~qG;PDu^;_VVU=Hk
zCkq?54P!lUxFjB)O?9sb=tlw!kGU8RDD!kB=9;Q|o&*Fs9756M1gLT5dp3BNDz_Gv
zlmF!GOl(pZK8QozRXS8*h3ZEmlVro>M-<AMpGlr*;pXMtg5A}(FERDi5E?X|SpbV!
zPX+$FE{i_|PspTJUL85y=)3*={o(7U0X4pLedFCN!)HCU92e&O|3PHLW`@$IMogny
zF|3=tFXgpU=NIpiW*!GU4!3j|5y2cQKE{>5e<v6_+7sj3I{HK~Kc!^1=RYV!_XSxA
zC$r}z!^aja10V?(C}Mxv{I2&u^O@THJ%=ojSPgwWzE=r$TGp9AW+6ewHiAKU^BMc`
zV6{p>WnX-2>F@;YE5mHC-zct?CE^;lf2D7Grd7I8BWFd8Z-HWFDknyyXN!SnhNC%N
zFP8IJl)Q^*idxe$d8zwVUC*utCW@pw+Hr)o18SE5s%uE&_Y}ZSrHepmj0dAMMTnKn
zTzgT)k{~4UIWWf5Ya%Rq#X)I)uUWjZ_l0Pq$tO8brmo%N+LX{K!f-sOqrMY``QmaY
z&`4NR&Pi<_uYXJ7#|Gx}{uQxa=!(aKpCt)8iy}X@m|tnpJw)HyVWYrGVpGMa5ur=3
zF+`}}3@pmRKXmVCIrWx5?ktfH{V!)FtJA@^t`X;>v@DQb@ymgG26A5IlIL1z*@88U
z@l)v&6bH-5Xs3^R>oUw6Gc9h%LI~BE+bMSi)(#0n70N=$#M8pH2UA>8cT-Q&g*p{s
zVz?QrdDyJ$NTWc_3s{=(Ej&N-Gv_9d%!n_?1!vU*EIfX5?Q}R;rk)J^+vAB>XX_<W
zm8~A4GXADNv{O&8--eUA9k?;9*^V*nV@g!L3Gm%-IhdySI8jH_sjAhLL8uJu8Nb+d
zxs^#q%BVZo(3qU4QPTwtr=ipng82u=z4vpQUC1B)7K|B<a-y*V6zIQP4<Pne*j=w3
z;`T<l9X})`5bRCK9D#$o;AZk7=my_wR0K&C?l;_1l_>IXcTwfU<zL3`-gw6P{avB3
z!swHHCs}M2Ht9x*NE5(Yoyr8=*Aum7gU>{Ek66B;(U*03gLrIW*oNRw&K#Oo5GylB
zrIS5{L9l>xkRX0}-dl}Cr|8GBxTm^@nfqBeg-Znm#ND<S8uPt`NWRDqPYrjm3q3bx
zqgL+VT%b<gMym+^T$&{0L=F0ic$M@ChV>x<snp@vzs8?@(zcgk4t#c9*62H}qL<uy
zvb9xfw|_NjTS(P;agnshk;7Dx^$FGwQ76@13=gpfm2AGwc%yPuY?mX;#0+d)E!>u9
z)V(khk!nTUbRJDw0ZFQhaU^_@RQ%U17;|GgZN&V^&pF))CZ<`^f7|QByAK}#22llH
zQ&{zEM3!{S807)h{P%k+Lb{ION=I{xuzQ?X{i>KKi5YSUeoXiLJtuERCtP8CPXgOg
zV2ML;d2ee|;Sr1R(0)1M0|`6vT@G6r?B`IX;n^A$yNOzqli_jnON`t#gm&o!x+&vx
z%*6A}q5>9oDg<MgnjzoH(jOKPByx<*j~ncssjpN>w03L9$?rMpdp^gdwOV;~680!_
zbZDy4CHzeQcf$rlV>XAX@_XuiL-k;*-PeQt+}@`cS*C5D_-NxRJhnfDU~U=G_?W?D
z$M)ZBq@9cRG`7UJLrA4{MPHDZC({nAiG(&CQIp6V2h)lBz6;#I0{!}-27SUy7Ojma
zl0O5ayk0#}6%ZH)60CZeP1g{M$@au1AIgJt3L%1@qc*!RmZfyK3<3-wB>?lwTE6tA
zc;zO-n7p%_T>nPBmgz>d%Z2EgLdwL5cO3(jPT`tb&YW~1(rSon$K1f>hLURW#73sL
zhLX1vgc+J#Pj)~$=HZkZ-YPopf<W+@@~dhnP!T%uncW|jt4NohWAS2Zb4UtQzBHAT
zZAW9{MwyQ_h&v5uRR+>-e7uQNSRD#bQC%90O#|`YpphB)xYP-6u?mL@cN{HmzxiT9
z6su($AN(m-dBq<QYoajbpJqy!4X$&O+QYVgHxx4*<Fj6SXrx@2dkVmq_>##iZNGs&
zu%LA!sW~uc908aCUi;qVgAwq{h~T}ftyNu66}!%(<(Xr+V=oID7J(nI(h~4hBx=f%
zOl|*EgWY@Wr%7w2J-m=}R_vbB9js-%nui<wU){Yw#Can58|m|SJy!mjR-vsxSw&vy
zgK}<5N$xA}d{bLl%?jsO*My!FNwnLij^kTYS)vk#GjsD7%B!13r^gj!Ij4b&;D_gW
za_sow1TdXwfhigehsTdk(Rr!$RZDwM-xk4)o;~jpEwam)g8*KWDv!!0h1bkfl@u&7
z4eIjH=wMfP4BrNfZEDhmtk>gx!a+*1MZi<=56YLmARnm!6@T77?#0q-RaZ5fTFOOp
zHQdoShh9^Y8aT&{AeRNxPvgLRGKe2qCScyE_GbKYU_HB<{n>0jN1xEarPN%OvVj^D
z&|8Qojf9Cgp!U_@ba`*&lNXNUOlvT5c_$Jm0cqkJ1X~_FlF;r2<!O|~@^|L~#kbR)
z1n&{vFAUS0)h-`znOnAAh4=so(=1-8JreY)EYf?OeV-fjZ<~uZ5l9dO=3RgBx@mOY
zQi~_rk;t;)w0BZ7Xw2TA^#Xp?j?5B%ubp!)k2VKUAU{0vw>%ix<h28rHR9YXmjzh2
z?h7GE`1E?TM=xh8H0T?7H(NOz(gm;HDqm__#aFN6wx`xYmO7EkWv=zNoQa|+uzk5W
z<7bpz<myXp<T-62flQ^vq6_#P78G3#4W|%i0O(DW=12l}o#giR#tpWL+!#4s77`L>
zP6)T{>9Ab$#PPL4NWR41iVba>2LNgw8M}t^usv&n^LcitetvwQ>i1ap%_bn_*$}ks
zHwmR??uvZVFHv5VH9oKJx<s~Ot&@j3n#&ZNd}*Fhm7jgJ3azZuL~ej2E9!xt?J|K*
zNC-h>J`i~`JiP)3%-oMpk=x=c*la)R-d<-<zZpTfPn-)N=hL=b63hxkd4Kh-CMnmx
z1&hR6-#CLU8XY6^lX%X)91|6Qj7CgXjWGY1VEAdseyrH`DPC~cOUsh&zYx%W*JR@P
zhAK>en&>Kn134%9#qJ=}_53(5k#4I^(aTHJnYkN<p`CrIc#hF9+BYMcQ7yZQjy=ib
z4)lC$nmQge5ku(dfLg8<i!fQF^ATypok;ib37yI?WWXf4Dqu^nx7C{Kwg>(~r5&3e
zmeVH$>Jhh0j$7(gZ1u!qr>lV}k@&6Ufg`Txw13yu^1hU<6Bwn=K(ixmz6_UFxkvor
z6SLA>o8NTnq?s)d2s-qIK=Ir!MT|379pR8APnG9-?(MrKS#{Nl*NJAw@V@dkwarN{
zl<>y!464C6S9+lsKhEP+EBY6dLU096`NL<X%A!0P$=>$QhEr$K=v|&wfw2mnY6e}>
z2&)sfGuWs!*ABZGiEOBsLB=Wm@PI{)aT~XBWniLdbV5&s8Y7`(H&y*F@doDZS9Hb8
zTB1&*)t4RZypO)f-LmCG8oI?DVd5-%ozIT6=AF15)2#xXKtazd!H&+qzIrfnke5g-
zB=y6!2AA0}ENuy`-zEI5Z!0A_6792qt9S@TBi7S<tD-K#*=Vy%Kp<i3f3EzV06fJ1
zL7)OqG}sin(13<IBufsD-HuL((@9M=XoydMgCG;@wvl<+C5THZ(ELjYJ`gQtC|qP$
z0yB46p?bCEOrKw4*%htfS^Qo?<a4x;kQgN6Ku%osysE0((Q=y&Elpps<RWGV2CpqF
zaL%!#cDT3S>08Fde@g=i<c#dZ&}|5K^o)je*jDbcog^YpZ+_9<KKMLag_K6t9<XT4
zywn&cJ6?4qYD?$S!3*!qa9{cUJsDz&p5El3{e`jeU+`)Ds7!6qou@KdvQ<>Z4%RWf
zDa|2`?xRa+SzVYV0%|ZM?z`I|)guT=6h?wzClU<InM$oTQKjed#A1%u`$<kF@UU$x
zl@I7y(S_kt={ga!Lc9SCr_<P2d@#4?_OT`L4u^0!9QE9WGvX$K!5?S~2Urc|`0QSW
zLbv2b)Jc+!qxnmf$ON@^XZ=b)5I*Q>;~|M#$#7T1r8Ck&e$&-}AeOZ#?#K7P)Yp@U
z+BPxK{7#w9blob|A{8Ue;b8*JrmFgu<7~%r2OtK8hc~N{u<Oj%{Hf1c{@+3(KLz*g
zO11Wp{pG9KXXH#hU7mQVTuXG9f%nmD@zw@!saa*sDUPlvX*1G}CZfMMm|ZK|r}KL4
za7yaB<U6H<+uxbxHHA#Yoq|#Wl6rSXWf9Rwo}Sd0_rBg!?bOy<$Lc5GYD%LIpD%si
zLxhC4Gre@84dnYcBHr0Mi?KvN{Hmekl7;Nz82Oi^UY?&mT_BPx5-L_btKY`bKsMbB
zY`u0AYPY{lu+7~bu$IO^U=N8fS9bL`OBlrv9>bwxZM_dg=BUS}4&`Gp9gVKE$!N(|
z`g84;c0Q#wakPprd{c76tl5eQl1<AnkdzGIaavn<NTF5SU;B01m$3)7edx%e^LQa4
z-Bg!u>?0f$m(3}8Jnjqxk<6yOVPNgp91Pyivx2(ZM1{5ChYlZt+kbr4Zcnuil3gWw
zZUj)=Ojnz`r3X^xd^Pc@IaR5`Ws(}Qv6_o;Prv}DmVyMHhK<zLiW;T9x-H*0L&(>V
zeMC}eS4F1F%Vgh4#RgR^KSC%PBn$!Vx}FzjWz6Q)_1}?}O7O~kOAgRF-N?N({rt*u
z4a>G6_4{rRx!382#*6XC90Y;L;XhUa3WuH@QR|TTF7@0{Jb9z>7m$;In-%bA$P8Z5
z<sfmSsl<v^rPEUK2QIF`dszb{h!*KWBfYiQulg8jxMKr~)HoTNr&4Mf>FYIrzA!JW
zQ#GWlp+QU)qjH2Op=7&eixY_DdG_lj2Scf%GId77lP{k_^VdUJGFn;1MTqVl5WMvn
zl9sG=Aw#-FYWsbM%+xQ6Kke5QKdvb<U*e?uoWF)_|I~#&rXQ+Sr#lq+_8h$))BV<o
zF2#you~ypN&uzhXcM8#6k)J3S=J&Pcoa=+;I->jNcI%)-6T0sPQuP#T)-y%Q#7#<4
zQqo$urlsMiKAK)Qb{Kj%R!E<kPLdH#%hloLgH5jh?5uG7kHir3^qd?y))3?MVdh)?
zUv|ekG&*7bW-&~VL>t{8VD3s=6l4qnqYw6X;H`0Irk)EZ4bNy2+sgwmk7i35&LJYa
zD*ERTZF1_8a#j+?W*rnu#pwo+WNrS1b%5y{K4DtW!YD);lP8<vG5af+Fy59>XCu0+
zP6ImVAm&KuQHsP8>X!lVKn=CI;O<hG^!{`%KA7V}tsVZJ)dPXS50CW>bSDEk!osX`
zwlED-8joIEHa)^BIRiXKRhW?MYP)!irY^;Xvz{9czi!sROBvxWDZIJb$gLYDWgCWW
zNRCpJYGM&m9Fu!;qFC_k$Yu&$yTYP)(O>PEiGmO*JnLp17;W;;|I?;`SgKV9Y*=n`
zQQ`=Vu2(;P`Q2Pd<L<g00`p}eQ?!m%5a$5S_nCq@=R35u*z~mKVCEAMMGQ+RYCuw;
z67xll$=&kuwWmd-7gkcEf~>EG>!FRQHN1R9C<;^$58#vb-(C6ZP7@5~=Q_orr1a-s
zV)V``h~a1JK8%5<Z!|-7for#RvQaJ~qXN~^is_IMzyG&L!aEQXgF*8QyYZ9^KUHQc
z41h!4o;{l&1j3WU3=;T63Teaz9)@|35Hj<#Obj$s<@qCrn?5CnB+_p-ZrQA1U4EeC
zahRez_QZ#aMtijnR63JC+>Dj&^-naFcyfBAHRDsGi|`9?ca~eOx@f-B-Kp9Th%~lw
zL3h&nhNj9M4X@~Zku4PKb={27*;g@f`uHo=@v>+-M4?!K%^X6^B38Y1|IGJ&Eb@DB
zEj7m&<RO-M82o@J?BoR8RF?%xoZ-jihUvd??{IpV0N{#;weIs2r=wKgc)99R!D|Fa
zs;n3&9dK-ou}PANdc_4);<)!pCCVGdrz`dCIS~Vlw$t?+F`!)%!_VbJykRT^)n{<K
zWy&fgt<nu;G=}D8w5gpiO+^aEHGuU}Mn+0!ZixCVR#fZ-xn{c0R}BT*4G$(m`JZ41
z0f!x_)@SHT%deifg1n7-XQgOwE3D(APCmR2S>6^?@AV8>hB#ixXUM0&JcpEzZGwOy
z5O`$%p-EqHeDA@Go_6n~^f;xz)^rvRE}=b9Pak@6Vt@PQm&r3b0YLUS3K8JA1UenI
zVmL`uAmtM3YfP=OK?F6&iu<(=q-X;(>K_wsRBKb7<d($XN;L?_Uf*RfIuy&L=vrHf
zCE>L1wgYksAjL?FO2SakDp{{2J9Q6?`k$$_G};^N?Ko+j@Y#s!uHhO6g3s4?T%cL;
zdG;5O0YiWYv%MyNdasCCbeuz?CFGF(T=C6s8XSI-6nZoVsg%ojn1sX6$fptQt={pa
zR~@ER{<Qi!v@bBg-`{_;SRK~zM)j9BEVFl8^J#7fQ+*?z#x7C9jEbtKE+D1OP}Pku
zz3GLnTE+9T&F<B!&!E$ct}M*BK0NbSDZV|Xd;Q(^Htolg9=a?1kMpGHrl|GC)Zdnp
z6(`-a6QUkWN{WfS(udbr-#8%s*N4=z{B)DWNJ!376I%%e<V%@tr#wL&43jq(Zt}FZ
ztFZs-4#{*f;)iR@TqhXLw?rroANR0Mf|c1Dr&EYjxMIvaZohK5H13&Rq$xIaXKZE9
z@|YV>K#1Md(=$?oy&Vma2;jz1R1H0=s?uN9B39b;{AS~0ek?br-!~8b)V&;fyh<*?
zNHyo9RScJX*3<jf-!YVV&s=aTPKWQ;QH%TFyHelnHrvN7`F?+NqX4VIgkglCm+J&R
zOW%DUN5UjSsDtXbwtNpln6@j<C`#nFtmX9JUeEQh@O;QiY!46h(0nS$rH?~*-H{m}
z7njj0ec-+|`SL|FKpPXW_o2CJKX2q`y_h)k%nCw@HU=}u4?coBsJx>utmFu;bR<Mn
zcQzAdn{>(Ie6YyWj&ssyTDWImY4r7mE$ioX=vpVYTL{&tkh#?HTuDS_lR5-*iQ3EO
zdEG{@GHSA)U$sM36SUzX09k0BqX`@FYuPODZjy^|@3tDYjQzqNWiS=^Lk0+({e?#l
zmb<)@v<=wUJM1{Jz2ijNjTc6%k_%6=zJC2$^f7KAF#7=nax3{OlT<@K#w9@D$a>#3
zcS#)FRbl^Xo5>-~v{02C!U6lw5QZ|bEi+(8dD<U`k^ibnBeo@Z22H1N1qEv2oryiF
zGHz&3TuwWLae;)-BzNgRn)b=W!(?sE^D{-QX7_V5xZ_S}MgKkl2t9}$bwX2JbLjq4
z2czbs0I9UGMF6XM&e7(j(E{T)1EdqonX%l!zU;LpMl>0@RzSb^*le(9xr=?cZZZ^Y
zz4crRU9}&?wWsSLXzAJe)@rcp^ZkB^w^Y^N<_6sgjvd<c&)9HH_F7|=BoVVG5+Fn)
zS6!@JjUk{A70;6XE639b68nPqy#jr3TARm?JLMJrwRy@-YscFb%~kTeM`Z2wwo?ah
zMPEe@tMyguAPXhq+kxEl?LdY?%0472K_K7i_Kv-Y){zfBnpXrp`aD11k(bpuQ*m{9
zQ3O^ovFB5-8$sc(R3|Yn^gGbmcCyR)(o3kdTylD=<G9?u>_i*=br64dOzcSGe32vu
zStfk|ZvV=ALH>5ReLIwk82nq!VWnaQKM89w5tg6pEOwJ@BFKJj?W@-PKTA>c7%X8b
z>edG_;tp2V-aH*x>-NwVcIAqGd{FN^dru&eVXRMRl1a2q5{eljVX9@jjeE^3Z+4Dv
zT0)~gayfn%h^;RBW=AH?hExr643uFa$$HhOQb)Tdk_N3*`pEs<Cp*I-haZg$&LCd`
zYi7^1h!pjl+q74;Pv6}O3mlQYB_M$1CUM{Q4V#@~=2gfLHhacxL+`}6BONVNPcB&y
zx*q~sF*ZyHj!-2_o)bTN`-8na3ga{ef+^$hpr6lDBo<6%%LVeCvp9rfUQu|cZE!ei
zYx26{rtAk;qvIC1eHlwDktq{JuOP8w(3~R|9t%2$uiDFARE*fD;SA_#kl+24>Nj8d
z>z|A9xr;fm8-qVQh3geE;H_sK6iVchGqX)LAkocp^Gfv<hX!qkESP?TjlpMd8lTXm
zmkfjsFXUO9A9szE^d083NJb5F&dhGN^%a22wg_7#H93Y+U4TOQbd%U{-c2$RO$tdP
z807d{Sm!2>K{t;$sYcB58N7sqABs6J31y2JJ33-LJv~oPop34o;u20GCm+qfqM3RM
zrCs^yZJkGy6x>;E^-p)!_6bQCrVCkg7V4&aDEazk&Xq=OTevAhThLfWp6>hYX=RVw
zHKSEc&1TwwGMB}O$U-)LkWUJ|)dag(^a$jG!w~RKnOJQtXIoqI-}kx)%~041!A9{0
zaiHXn9O=aR>}@^5f0;?Y2so5h@;>+@xwJCkLx1uc9QlvADjJ5icPv0Ns!v!gu;@mD
zImu#ja$+2g7=Fen!XYS7Eetx5i|@YX`Pa+L&hgXM7ZYRjS1Qy+?S`fnk^KET)au_w
z8Bn<uleGJI*R^&tK`5zD_tG<@QjsYGT=^pA$c4U7b=aNT6eLcl94|V!?2-QO@7%8@
z*A6$7?v1UPpEhMXYdG*K)aK$F2V5Oa(jiNwXx00PnTW8SQ|r@z-2{V<Vzrm~{07Tu
zptMxiL1#5lGM9X4q_5S-2(re{PgO0mGh-;}e~|-YIR@h2b{+{{7u<Z;+<>+aA)Rnx
z7$$@r+(FHF5oNh%ms6Q!FgDo`TA(nu10T9eK<65!N(y^)R%YRA=*x-r_OP5&8+k#2
z$Q%oq<#B_dv9gH)kj4N(INuUW|NYR5KOkysAoVXRpST25A;N>!3Kv~i^4|48pw7u7
zm8&Pjch=QETpfAN$u#gBqB$KD6pFoXIC7XFv}iwZIG6LpfdzL**G_)4qWG`3sb|4n
zg#Jglztn@rcy#6Vb@f!Ua}t%-Ys_2;&6;aOP0n?ER>JJlqCo4z;)t;}YHA}|r?AUN
zC=gvsK#9UV+fC(daAk9yY**YH5-tO+Qt-sSO?m;0<9a0$lkCxN-@6#-9y1oM@G-k}
zo<UN@{X-S>)e6*^`Q0uM!;Gkw-ZS=E{PeMl6mP{`h*J;KL~^sId2^pUkciK$IMcl^
zHaBPB+Z&K0^<}{^Sxl&eWI=!D**@Q=k|tk9&ff3UKOHsn)`uv`?}H4^j%;_>CX@W)
z5SXryK^9$y-s?qvQF9UKuHjN1++1}0CKw4mR`*D%rEpf89pAKM{AQ)XMYz*dskYCw
zIo9Krmjx0VeYF61$Kuf22!C6_-0A%N#CwDT@X7n4?+^38aD61n@}*GR-po`8i)j*8
zC}|{bGz<dM3qR%2-{z+eiiAntzHYMnpb7DdM}eKZm>6G)4f6yOZaz*~7j7*qk4_ZI
zP?QK(makuvlFygqi4z_e?Vm_-hBRmk;v#c~OmqTa6<>(DOHHuU2HDQB=-$0(D=Fks
z_rxEj(6u0zOm(h*S&<6m24s%3yzDfWWmV6p3#{Kwh*TKk)MO2FS`lpDVw7{L^q!r%
z02=6{N9QQ(&yo-uPw~nY_&|S<#SQ!UBzhZiM%!-u=_4W8_Jg|Xw`%I_H&eU1Pt<AS
z);ERMkcodgK*y1y5CX})3;S|53HdZXJ<yYJimyA=Gc;TQ8mQg`5m_WLf;z(48{5mb
z36TWyAt63VPp>KJZ#`%<#jA|b!vl5@p>2XirrYJ~LZC9*(Wlg_7@eH)vZ_S;q>RO?
zLt-%DV1R+`q;XmiOZnG@m=23IyDEgmgU8C;Oi;W&?yqMZe~nd6Oa%cm8fXg0#{1z3
zb_JkEW7^qU=hr75gb{qhiNpBTPii?|LpcXrV=tX}Bl3%T+Rc!+6ouy0x%DT;Cy`+o
z_18$f*Ql%I6I?3bm$`&M`<1im-gV?uqc5xZOlzY9aJs(>xx2(74n2HOM$wiL(0`pL
z8y^zx8^;*}O%L3(ON+P+g-U+)W)mKJc$mh@2!%(VuqEY`nnW-tE6+l?bQ=;eZ<5F3
zp9p^6et7@%F<|P12e$ES3#hcCV20wk?DJpW2EPdJ&B0|>d{nzaYYsTk;`&P~>x9Mz
z`~5jbeqWdD9@QUN>P7zWDWHNgW3OR0KIM&Uruv;ZcLBC3dz|XW86VG^b0zT*0qVDK
zaVU;ElBB)=;EP}NYjTtrm;G+%p4*URoff^db4|muC1g9)Q@_sqz}qoGA+?z>at3Dp
z<cX4JNFFwf&kYO`&pu|=u4N{-_$5J@kJg9b;m?^2u(tR|Dy~*2S;JS%(^l^S(x-zz
z-J>jqUirf}21&FWzUni2#KpObEL5_pGUtfjmIACc=en`Mih>&LqD_5S(yL0_rZNX0
z5oO>!+h6zP%!4fHdz{nHxrdmAripbrT?{*hS+?lt@nANWFkcCBE18RL5Wf^*>%+E`
zQ_8&x{?I_lC!_|$s^ItX{wBCFxTp)BIqmFE^Cw}h$|(Wg1cQWdow>kgBE}<U)@%wk
zignt8<S)X|AvfX<w`yK9*LR^8Y3Y&gnmyt)G&Bx5<sZE*a{@AwZVW@{;t+`1CgO$;
z)!M(?e)OxkWrx{4I!A}Nzk2h0AEU6)zmL|DYnw2hC87W^u~AVwY)OA!-Us86APOvu
zK(cDsY$L4DK%;@s5*l_mLqPxz^B_V?dNQUWg4XYj3Rr+Jr0T5*5W=qf=smanpSZ^y
z>Gf4Vo5^Yd*BPti`CcH~r1-m2)=O?mOK;A8o!iT#KZZ0T(Nm#$$<Yya-XOaFuubC3
zF~z76(`)C5+P~Z*<lWryh^-k#D4m~lEE&J7hO-oWm&3RGPYw(QaAtS?iWm<m&-=c>
z6}Wha;h`+~19=}cOm81TLj1A<4&T+@+yhV+GVlJuz>b%(hjmXe+2^NPE#sXG+^=hF
zuO@Vr<kx5_(MK8H`1TUr@~sp)W!qjj^CM&SrdBFKe^qO5F65bk?bWFG9BM-ZzC3}M
zw}CM4yPMa?s2&ZfRLu|Qn9Qc(Ro^u2Lh0P`o4Zz=e{RRF7|Y+-6xEFvX!CP$VvKxq
zsl(zCRH-U?Fr4t(LPq4rpI*9}bH5tj=EM#T6IXB-L<)oaTdPOaawHYyQ+rcRrA=1`
zQ>!AZN;0pXzCZ4}gldj*7>!TK*R=FF8JaErm)`Oi{M#NRLx1aOiRRMwIICSknF3~}
z&dUU1Pu`A(EW7HeW|p)hG6o2&ZxGW%66m!RMDY=2VXX?!hax*iPp36s_YDkM_-xiT
z=3xmYuuzt#D*oW|y;>2BKsLS5pS0>#Zwu!L#XOLkTNLfFcfDNe!pImsoAar6!SUl0
zL`n-g*(JSF`2J4wvECF(p;1mt9X!a!luN9%!S)7qaK`d(zV3H-FP+!2eMT)CTd-%B
zy=P3&h8)eeW`uC|B{j)3LX}zq)!*&dEqB6d?L4msf(pi<{uJbO)AUVn;udg~eBAJ~
zmO+Qik&M=ORZ<d<a7HKwNmUx9z>)Po*9!!UQ4~Td%#910cRg<6rMLGd=V+>)^oT1^
z=SP%zR##(-&zoAGIxpKTy6=EJ`N)ly2@PtL@Gea2HL1kN7~COtJ80iU2{AUNss{rw
zQ*<E#yW=I@O%hMWKrR4dgIyLf>L6dDrkU0@@#J<9qAtF;<w%snb$h1<IrU0hhq`h`
zbH!)s={0PZiq9_O+?FTS8v?9!=blNfv|7*|;4-}TBJEv7FvA<uDq-vfZ(ta`+8DY9
zev`qXin^rqq8-(`55{?~HB=$eSOw;NB!#;oc~1kZY1!LHdZR`?D``c*4(v3iJ#l~+
z>y)mgl2gN98<FALVF{`mF{&4Tsn$+b1xaW(grXxRvjCm*q}8CaS{!|W&b3zNrBC>9
zAm<Qyv$8rr?SjGKjq~W=zYTN#_w=-rTPe3U<}me(QO>3hR%^U+i}s%?@B@?MCtRDT
zz<F8>&gx=wjE0%|6uXCiI7#2y|N1LiNT2Z9P;lTu&E=VAE3dibun#);pMe%YZ1U;H
z4eyocOUIkLhs<%h5SG3@vr_RDliN51-g<%Hi?n(edW+$H)}Z1St6xYk?2lQpMZQj3
z;`8nZZOwM1>^vZP!$SvQm!>p}*oD{Q4}=t$TdpypF>(*ZO)RZY!U!=-Hq&p$)32vF
ztW^x5NZ1JmOGgjZvnDVaW?xRh#IQkoi_id@Rl5ic49<xq_C9S4CNGdvC9)f{!OC{c
zp3!xMFvnCM-l;IBR&k2RZfU7wscFhL$uChbw^J8p3h7noa6W?cbkKc=ID%kFgJDg)
zoK&IE%0l`m@yRsi0#TEPi3O=PoB63x_5azhX~>2>`l?`h{`n{EH&@<$y?(~zwU=h9
zQQOlAe5c3}Hx5_H`V!=IPrQ6z_5Pk6mdvd=-lnMTn`+uvySbTopuJ7$ifPx3(@8_G
z3YpBx^ObIZefBN}U;9*Wo99m+NIr<;%Zc?CuOaX)3pScyyO-+Vp6)+-x5w++G0o-u
zqlNd=APUB3N9z$)p`z4|3tNy7u=)?UbPhYuen26F!sZIBEnX=@a>qY;jP310zJ#7B
zQjF2vh6`B9SA9yMkqLE06uQujvGYNJv;}6Mqj>>lt8EKU6m8G%?p|V1k9zrBXEAZ(
z>2Ys7GQ3Aj3>;l$m7u(w<@_^?rna^gS1*%~;A<+_sslFZRA~`GF!KU^1yllHt;c|c
zM$|=2-g+4G{lgQS21Cedp4n$UmycuO4wl<a?I_;G{E7oC;KXfewfZxet6*V^u^eF8
zd`*a3Ytpa$JH1+TloMsHj|O0;BDZHkH_F1$CD0V*=_v>o%~6nKF(358sCOhEBXG6^
z%~4&5aId~IOTm|a+@DllmiSWEE~R(sGS%!xLC@_QnIc$xaw~PQ+7a-wgu3(?a?KvQ
zRdi5bZ#n+;o55^K373zNt{RJNxKO(PH<B~`IzFC9?a(%<45&^r-vgsX#(LBX?Wz#>
zrni`hE^_NkakkT(y=UAb5-w0d-@znQ&bRQ;2mR)a>Sg{4@bFN39?*jurGIBq6oXc%
zI>JJ`rFdSz{n29V?AbhNUFH96HehT>LdS6KZkuksxGydWFyhC!yv%{T&MIKz&3zq{
zRkQZcWXiy+u3{$0+?|i^EAXIC<?l3oL>u})y1qOf>b3p<Bbm05v}q(tWr@&aEg6bb
zvQ?6-bqumES;j8sl#@z@86lLC5F-0Blwu4<i)0<yrVKNT7-P2I{c)b}^PJ~AuiyPo
z=e*AA)tR~HzVGY0-q-tjza_jh0UdI^P4hT+fiV}=12R>~)Tv5^h>$P-jdHkiv*BMt
zy@cYBl`QHEBgC1v2(I?ZctrrX;J^=HL(@X4R$8X1Bl=hMHT0KS8Hj`x2s?^lABDLN
zV4Dsz+{Cuh3wr4`tRlZGAIFWsj@uS*$h%tk>SoU10%iPr`kwQIpqRPvt5(!1lhb4Q
z`k?LH#8MG#>~x2sbkl_u>bTH&ptYoC^=(4-*GmQhoXu|3*Y7Z$&`;8;Yhq<MwDWlT
z5Mwli<GOYUAa@c1A$VmgE<cWBFWv$l^W6a~1bfCr<G{3y!DFG&Z&Ij9pd>ocY}&0t
z3VW741pntd^q(X7?etsH{{36qPAZ#ivh?Rgzi~T%!+b+{<Y9f;?CpHR>AW*H+rV47
zuFmC1`kLgOdNu(y$HKmVzR;X6MmI_~<1+5^hk(z30pd-26nn<p5T}xv3L9%18l`tm
zk~uM70qN+m1+9;V+5bLP+zJdMAp(KRT3lusY4<5Lf@q=6vG$58ox~$hhL_%sUya>?
z?X$L7T3s_P(NTIbh{6;~;UE(pX=-7!f5U;a6}W><y5pd4l6qZz_);&Jr2Z}2x+s!Z
zda$IB-p)QoX2!EB*~hG8IR>)@8iBHF{xJ84+i|3;cCStXJInj5L@AJqwGf<r-v%tR
z|43yo$_fnRK_kKCI_?p=G?OKC&tW%W*K&Z<7SN)2bas(Oo|+f{=JjjJ2dIK*!7JkB
z%h8qh77q5W)T{$iE&sUWknb5?R7kUa<W5x1(7DeXeo;#Zl@d&Dt<Z3WEp0z0z1=v#
znnCP;(EPm@dwPIZy@o<N&m?jf^j!2Ks&7}`IQ2_x;KW?y`d(W=jRnVRTa|XE<qUjQ
zWhS>agDyxb3^T9DZxU-e|K+9yW14lHY|$c9L<t}n<8#lTKERg62`0MkL-=>CwqnOG
z#t8Rn{|wtvdc^jzdFC#@uyN37G6YQk`j)e32U7z-AhfJ>RdUu|R&jhDO*y_{KxF#n
zp!ztMWS-2ei6majG=9LUe!Ie$q6f;c_#{?X04EA*RbO~7VsgfOi1{uUY;suOUqA{#
zB}u1S<^o#Cgv6Xc&WUz1`fMcNusAs7fvGAR>T)t!2kyL&ZG;VuiL~s<hBxC6k^8GI
zze<{P4&?uLa@r)!ZV;}!yKjgqMs)sBpEfsnZnIRe&O5z>4N~<rZWcC;%ywYp{PI!f
zj)-BrH1dON_5Owak@h0Hi2RO4&!$AY=aqWzm&`E4&as~tc14Pdj_KJgbbU(P8%#tJ
zj^2r~`J;LEt^a|Bwk72r>*t?{9d{BsDDGQQ?anC6A0GWvF-wuf7K=?)(5<<K&UwR#
zJ?!Rn*i7p925jB+>Tei(R~KgDwqwy<<cB^<bk6+y({{^q8QY?Sc6ZR1mfo419Ml}R
zOana*4f$cG^EayXsZxaIcI@t7x%8)d(tp15IEM~@WseWvmNR^2?DL&d1%3A(f}c9^
zf!H%phEwsDL(00iDE`Mf*#DY4z*7DrF%Hq5llT~{1dbR)wL#EQa*rQkTK9NvIu+b^
z-&!Qr%VywJ#^Bac2XXY_Ium1Ky~3ax%msHZ^bhrT{EQAn{s}|@fcW+J`1mtMa7|L?
zD@v0qO=T*D&z2!v^V)sQ$##4$wO5k!&2eM5OkdYHPJiG@<lxtkwUgh6>Q0}6L4F>t
z1+-(!-9PW1h{lY*Pps_?eBdh?*&zcg0cnz-xtsp;+^qaltc{{BYP*rk;MEk7%-82k
zq9*f^j9qeHw;VepZc)s=n@=jeQ<<-#t49_3O-1j$tdK;ASA9uR#Bu{2@Xul{ve!pB
z&4mB~42*mrE*WWc51s~9+NfwWK1N9X^m=SQB*j}ssKJT6-HB$kMh~g=$p+J%fKS(U
z07L?co-Rg+qZ^c9^!0h}lIx{J&3nEyjmsQNzq6`St?59Oy6D3~GhbR~#@aEvQW^9n
z1A<P)tzNMMFFbdr_wgq1w4c08SU2%%Ee4wtD3F23dj>e(?14Oh!SBt?_PALora)UT
z)Y^dyWZx5DG_A9$N;aHAFx28xRuq3;MfhCWRafUMM<5elM*I(|rXZP#OuXM5tnRhU
z4cd4_wVjF)Jjt6;xm%aKxzD7b@U{Nbu;1b8P@A4zVSS(2L?uzLRw=DlhGwe*yprQc
zsNI<&-^EjDRiiz?81}npw6!3sa|P*+KfOD^vH4e3FlrVAV}Jim*3&5>?HT2_p{vE7
z21(;{TS~xgLlE8(v0nl~<p*Z>`uSmYoiv_iMDwvj1FL3V=dDqohJMMHzT*%z_|^IY
za8tYT9ttLh{Hh({AlqNH_q~=8qZn3X`ji;tIkI?<vz-^!17q4K#K+~&pbO@1H3LCW
zsr)234_`~$o4$uRb<Gnnh=mu$&mcf0wDB_d)6FkbXS7w0?z90<U0%)w8j%1$A|A}%
z{4Y1A{~RWVJ>VWrZ9IV0>pVI#>lzvwsv~dkrQykQ{d<!#I~u0Z3ANgV%bZj$vpNqN
zzWPFyBRFD4tm-IW=$PAzpY;uMnZ>%W646Rw7u&}Qza5BS8XX-@A}<_O7ERG9zi{~O
z1;oSNVA+~|E;}I?Bfyb#N)-;)Q0M>-%uW^R7e?4a25@MS>|ECcdw;{%a)3Wb!IpLh
zH@9(wkS5vD-8AUp0n-zOBIKH_k>6A~dT<i}){g?(Cz>vYvvu0VyDf0L?B*RUtML5a
zGyez`JXy!`jh|NR31%B<t*w#1kJPlicgqcUpx;e=JG#!o?GQjigmx{OfM)jJYovFz
zqRv?#5qD0g&Q$aok8)&s(J|<pKg&C%Iq`KJc%YE&pKhv8MT>eqjO(Gu@-i6ib<g8S
zn*B>j$>-)PvgLfOQjAL8C&IOIn)57uf_lv^mc@AXJ9f?k&aOuJ=8{HVM&IbiU|!*l
z{dxyMlhfnltw4JC^XFb6&!B`Ek`25HzYX$6aVzHqsiqHxtqLL><>lKyKK%wr)x_mU
zE$Eqa3bHA%u#&AKrg>2mU$Vie*nko_UgV)?AA~|)>bhYJI0n59_W(AKMJV6pSF2A+
z3x08$#h4&Ds?N(%^AYm4a&(Ck^VqrvW7RcV?;{t;P2-)NYfbmfmlK84AX6V2Dxohk
z{z>fUUVkI`QCsX6iy@Wch1e0F&Iiy~E{O`Z1k=xWlQlB_&2DC7crH0T*KoBX`ox$u
zDy=@_<Dt$Aky~BsbH;L(LoNH~qNA!;_~RAtSoI}C9w`CrJdb<tl>&(#vyDio+k!dV
z;0z3J7s6xg4T4{cfX}7g1#1{EnB*6$6-C8Ay=fXpmbK?(IS^$UNG_WcDy#N75sM=Z
z<_*{Xso^W8N%D!OvUu`JzK2**UC@L+mF5vKEGA-Dj^z$r!0CJeSw}#N=I-t3E>3Zz
z8mevl190?1_x=ifayQW7P3`J*_lToU*Rdtbin8=^K;ImkO=rC2x__1;kiiLmw5p+_
zNC+V=xMce;c01HI_kOf$!Is7MwF(euOUr%T8rx9wbiX8-AChUb*zJ}Ckm=o5BR{4j
zWKdMJ+t+5+u8KWuh$~lpkQw(mDT6P#o5Y5jYquZ6FHVmSk@F^5H5dssc>~v8DWIrM
z!}tx+7WmOZV9T<+yw^J0Fn(O=)cKPq0#kAUTsh?=#b6HAtsS#z(szX%)C$WtppXRG
zmga<+K)}4&aj34z5Smme8A$K<r4?Pva_%VjyiY<LOyKLW=cn0jF}hr`A$Q!dNY5!l
zgv9XuW76I^`_5e^<0NnCw;`{0ED{$9F!Pw01wg_1uhm3bXK~X+SKX}RnA0rP@p|m{
z0a!aNp*D9lI;^q{>fg10sehkn^;p!9SJ|bpA=QIx4qjxm%Bl5BDwls%HrwKyMNEIv
zwjgaTtD|MeLI2>pZvGXC;+<{SNIv&r2++?Th>ZX1eKr68yssLmxw6S0|CnqW*$v_C
zwZ7@PrxsFZW60RwwAR)mJSyQqZs1AgqLClrXz&~^9l)vbH9uVCUoMkWqoz^QFHwxx
zOIL4Z<s<gdaN9kyk!QOgY}zaE+yW*7d@YL@{{yo38-r7D9Pw?J+7eS%wOZ2=6=tvh
zI3=YKd7;P*@v+hNMw!j_2I|*W`W7E7Gf;QdSDnkEeTebC+|v1C8_vn&k_*d}-iL=K
zSdsSQ%UZlB$B;4K&gW(n6EhL0qZJg(g{<tOp}3*{Jm%`z=bCnS#O5Isf-o5wk`-NR
zdK&r3IXw=;<CW#Rb@;=`%67vH`i)|jRM%L0M7>Q?uERyq2O_bHZGCNdj+SniF<=p{
zdpt+A6W?m-Wa%VzC)3OYH;%gwzfyTKhMl1iqKDr!?Ao)O*>_Z*C$KYaSzP?gr2h{h
zmveWu;yeA+=76zx;b&5%2C%0;RBaHvi&uSQ>(}&KBhw_R5Ymw@abI@?e;o0lq}@H8
zR`ZvMTfjTf^NHy0svocu&X8Botp#7RSe6OgMc(?ZOs+I&Ab*FO{6O{#t&-MC$}Y&I
z(!wGd_PIV5#iuamm%-s;?v>ZpA#O@TY8rOo&z_XLv>j{)M(Du4968O?2dn~vji+8A
z(j29%22*15k)?Xa9=AA-pUV*^PuGrBE?<JsP*GNWmiL*Mu$9>NG_$8y{mDmG^-D%o
z6RUm9fT>N!mNQwRjJ>T@p`o|l304w4(maXf3T!qyGpZif_C8!BHI4K6?yLR@ti`U`
zlsTx0Djj$gVUqS|3HjV)Zn{8VRPl>@b*p0Xr|{r4NAHCyGKnLu(l5Q4t(0_BY&-vs
zrQ&mh?eig&&<dd-k#X+f(&xN5(u^7>yheaYiR-CDJp$au=Dv7%C0um}tT4J$;kLjF
zPNrXWW!AskstewV`UrM37=h}g^L;p99^XhltW-md;ndYuiMdF(yau)uK?{dEKKJ#7
z@#sq9ZtY`Fr0-<r#@v(YFOohR?Fg}gMJ6FgncEsn#U}I9R-{7G2Jb>MIaG17I&*X<
zjP);5GMYxuHa18edaTF@$!_hLs0(EZjf`7JB$AFKj7uD~CFo(qVf<&F0EFYYl&s$W
z?Jl4<Wg>$#ohYBqC)&8W45=@yR%~?v6Bb87A~IWr*&n~(GH-uej_+6^kW?&;IEc6U
z+@HN4&PXyOWjim(ob8H#pQvhAFLRrCKr}@{;kjBXc2mUp@n~!P16(5}T<<a}E=S+(
zlL$$<gn-fj4-d8z=kaK-xb*^1O4xBIS3s!Ut%2)%Rbsm2>WAmXAR28%A3h-!7`67;
z{Rj-1%*_^s!t$q0<p4eCOO0$FQQ4io5DIzxjy%&<XMOi@^HeRzelwk(G;9KV*DdoN
z1U&+0l?Sj_9lzo03Q3lW@(hf#`O#gn(L&YZptbtnOT!BURq%2m-Jq)ormWIjs5;ha
zWJ%ua+`(^2oom;swdah^`8GVSRuh%+v)-J5&~;}0xYG){(Wa7P)+a>W4~Na0fPlQ$
zkwL`n_B~-PnhbZG^7!dkYhTL5;tZG<v}g&tbe=dSJywRfkO>Q$wz8r&nkVy@-i6!l
zO(I&|+y+K9hwE_ze*%ljM(zN&K01_52lg^EGcRJ!vn!#q@vqvgP4UKvXOpU<!s%Iz
zkV54F(RRUmVt&L>`ZMYK`Y6Xg!foR3>VSn=h&1RL+XdyQf!{Jh@-sAHqT*hc3qQGe
z{C!aViQ!%NP$tT6RTi&A5Fu=c;lw0{vBo<yg!C8(1D)Fm;9ur`s1cQEk3vI<_YxAK
zg+LoMGaG2*C0x57h)@9lEW_9s9CRlM+&YJKX9bP_`!M4ljJV3nUv)rW@BQWu)^g<B
z8`V??#K?B|T<EN(I=(=vcH}hGZjCSdJ$NWC_^6#{otrb^`Og?Vef=W6v!&b6)k~oZ
zJdSq~UhT(`fm?qaLhQ)Bvh40zepQ>{cZ}*amp~g|<}jR77<*eR2aMfZ-R)f5O4=v?
z@|$XJa^C?eO+V(VFDBOpG*Gia)5wuSd0bcT`&S(rO%!rj%FQIhyQ@Rrm~~BFh`%r&
zu)s<B716L1W1^|3t4B#pe2|MqrDSrW4xjvaDFo%6nfo1=d6Cd#=nBQ=k?HMlSThj}
zO@mVth)@(Y+_ieYs7@cBDeM(42nE1J9E?S<3s~92KhqFOe!jk01adhABN&~YQYW>7
z+i6+coT1|vBX>yr-1eeDwf}v{-op>s3+$h{zC$YLL$Vjos2x)?Q)_sfvRYNU%y{HB
zxv6&LO-HQW?&crZiEhI2%7%CU<mul=Zq?uPgKft#Xef#>!5e?>zqRvDUGl?n__!V(
z7jVn$;_7V(aU}JN`!s3_^h>tUl3hjv&Ki{D?O<BS|8~19>e?9$_S!gi`9wn@D7Pe4
zOG@rqt%63%@7;5X+9??jzP#KM<wOPU@Fyc3+N51xqq_l$2=X9;4L<j`UNJdZAkOp}
zNaN54gq(@#Xs>(s+zidtg0#1}?MWhgb_XnkIlZ>_hg|-+ASy4lpKAv<+im5*CI&`o
zwPvT|dS%IsK$)o+UIvvHBLp@{EGkb&$OJ<C!X6?Ij0hc+Z{+z22ZW0+284~FOhA;r
z9X79U+|nx~By?{yZ!QQkH`zI0QByyo&hLDYH8>o7q<)e{<bFxkvxouGtlD;hdK-3~
z2}tMQTJ7EUSs?zRXeqnl>_0)V?S}q=@6N{G&bTI>b_uT#c@g357RxFS`)(t;|FD0h
zulc$=s_a{%KcB6R8RD|B{<)v<s@+RTYrqosl=D>?;O4zi^m&jypIB=<9(Q?jTBe$4
zvc$#uD0NDmSCZoMOw1jH8O!r`$P3D&)6-|~4|Hfw^2$5`KOH(e&lutYA$o9pLnqFM
zuD6H=_frXw(xU`i5LPZ>^}^p&RaM!tYm08kh!rA=E{Z;s0tkT4<63HJA9?pIY5J39
zK7uitH^Q2l2gbQ-B`Q+EgyR4^cXF&XL&P2#*S#=2(=+Y2#fm`QJyu41ICi>6V75Rj
zU~%dOeK-;VQ^`3Z7{MANjhz<EkVr|uW(f{5$*qc1<FhmjQ!zgEut&|*P+h{pLXv3O
zJO_{oIy)yMJiFk^hNMlQ{`^UQTw9gSC3DxqIjo%I>l)4<+_3+K$e>uU&6yqXu2uuO
z3eC6Qd%!5k(=c27=0Uqpn<+Izxo%=lw=33xB{^q4<aI6p_3YJ^kA`6G3Oa3Ubip?0
zGzYD#=R`1~YJ#a#ic_UT4Yk5NdgBcZ+fPfdfJ=H%bY0hF_syGR{A~3#YNmYp<aVa=
zQ(HR|uA?N#53?o&Hr~2&6yq%Sn+wG!Upvd*@}ozefS`Ui7<;LX5t##TMm^8HGwQfY
z#la)rJo_9$_Ld}u>%7RQ(z}f}6VZx1qjhOA&&3sjeBIHBGKZLTTgx!E!#<lPhC-lH
zD?LUEjffw{w(qhLy;~T1>|$%O?|Q>xDudD6I)R3}iNvIYgydv7p~7fmT)Sl!Q2X%{
z(33|->W-{5$CsBpt>$}1CcqA>H+uNavKALz?MLQohiK}GWCjTKAQX0t|6^YvEu|07
zYi)o3Q$p_cQ>7`V#6IQrrcSaURgOI27G$;*`HAz)b$RI{klH*;FfU*91T7W$(VQzH
z53xDaTXN_lPn^8eVFTCIZ=GHsD|xfmy}+7_X;d1xHA+YXtCo7Ft`2$D_mtJos8`sW
zpIm?@0IZ1N%PRs}X4~a>5x8M;NFElP`kmofW+pYx>xUa22lQn*Bcyt#%N|pr$rdO7
z^an86;uCM?ROR9_U}mUBW<gNZB7B97jbF^$#zbWIGMN}I0K8{sx$eP!8%o>J-*WqM
zPj{9_Hn$x;>dfCGKl<y}g2fPhv*har{`D1G<X;o5!i?MxJvpukOS+plwfiBc`~iO+
zM+tYsFyLHV{%w&06x=rVMJ6eI@UewGRw&QiR#Rl)r>`uU3`#`leBN)Ojkf;ws2m6J
z3&koPoKg8vIl;Tr-^%g7zeJt{6ITF?`y3ne-?;@88_oxhZByXRP*?`|1v%k#j>p`)
zeTFq{u~cNK&zTWF)USw|VP`@TCByFN0nvRs=LI&Y&3*5g3eN@fXt~x2n*Hx1<CX_R
zB@$mGPdts+4iYM%XDeTFgM%9g`!qO#5f+Ns5!US6gS~FtrzV~@jH)r~^1Z&dt6*xK
zq=~5u%W*4Hf&Y6zpG4sR7Be#A8lwXCZnl(B;h8;6`2rlrg=ek>0PlpK3*ppMVg30g
z*X>6bZ0=u})tBAwQR&ChFJw<_IN5mm{Ho;7$%>XICosf2ZPqC=4HQ11=V7a(7E}|S
zsYab!2X&vtvA@;3zb4mBy>&!DU%vN-QcP7BeH|DoK*A~@*AptMqc9K9bl7ag&*?Rh
zYz7~bkcK#rotZgpqUl$U8zm)QQ(|T4NP;Vl5rpw6&iuaCq!%J&prLD-J&bX|#b8Ez
zRN_b%u!ey|sxZp9WpYx=J-B4cKeoQ<!XdSu26>ed<XZkLd<3VQ9xZQ6-914ztVF(*
z8GJo%h4_Wt<b_(D(VWBKX^zzFx{a)kiV6w&dgHBv`3AfrcLIz?%dakS%4JYgM4J=u
z-o&dN97vrbuOQX|mFi|seP+yhCTL|NA;&1r?-K#tt%&Pb{ReAFpy7EkhY?&R=g(>M
ztYSVd9vfBYYGogeuC1u3*mg+vtNpo(Ihq#~`$6wW_=1lya3HTLVazW_3nu^@$Cu3^
z!|%EY%3G`Ez+(d^twO0gk;zZN*vXXBt(QXDGKT+xBD9ng&fgVI5yLqc$y4PA3gX&&
z4~jq@H$kqxVs##GCm61mJ*uy-4?wM<p`lk1_9B;#%^#s0+S%hWKDFd4zy#|7Fj?iQ
z^s`JdNS4*Uv%L$9+Iz*+UI3<pUt^74^&>1HZX5(<q3TuX|3NUj&)!JRc#16LS|R$I
zeTuNL<Bsvq=$}gYkq1Qu?M;{iA(j>v7V-wXQV^Z0mM}4>K`#8TJk^U!N4<&G$2m`J
zEXu5^h!jfnZ&S9i_YbVe-i}&5aKUHY*)I0H1U5GGe1BZ!fq@bU3P_^k*RS2~w=yDk
z+X7!aSuVY(R~!`tD{x3Eo&kc>k;9+}>fI|}Sy?GyOpOingJ6FrVaLXp<uZnk$VIz*
z4E2yNK63@lsGjH61qE(mia@cSl5P4c986h$ZP4yGqt$lOMGrDOI<YWb#FQlquINQ5
z&u)tSBYm)Q%Y)FdiQlFkj2Yu}l3m8JuP?{7e}0v<TG71;9^Wdo4KCTX@Dl)tnW}Z_
zd-Fuv|8X3l&Pss%L!q{=R)O}blF!#Pq7v+@0#L{ft#+uT)}RErPZlE$YCG_bCMfmp
zjXyv*rlD7{sQXgcB{WjN7^+i{%4vtVm8%tqJ2u{Ww<%3u1q+4GH3Hb4?)2zo0BU@f
zkB{GWZ{D*=q9?xfWfRCx2f$ybF3*rn%UtFQ=4*lCA6D;7op&5bZ~1ZVewdvK1bRb4
zwgXRH7@iXUrg6rmVlZy2>P5u2@wrR)HnzfkS5bp78{wWA6kAEEQ-<m1Zk&k1-k{EE
z74cY`2V;53b!WsB)cb?S>ZHJu{5&@VgHMGROM4&R=z#5q#P998D}V_I<>eT@n*38x
z4(rQ<i{6@B+Z&9xKeo<tds=<LB0D=uZf-7GP2-=)u!mGLsr&f*llawsFK@1%zAHiA
zj8)kj8{zyuLlt!~i99m?PK=O39+@Nns`ODwIkj-0MBB)IqU|G&_Bmx3@moqOW$Bh3
zVRu#%d4AC7Bv!)=mFb|VdOw`kiteuaDKZ{RVo%U-z9IY3X8=q6jVa>HMg+B?nnEw1
zWe09sLiF!7#SR!8NzXm2SNE{Ft)_!^_vBCyIBwFKjVS>4oIR~=dGF+;>J>!qctBdS
zHcsCcxC8e#bU{#uh2BmXiO1qy(Gx)5@#I-<5>&btCz~;7(=Q14T&dAWroD2@L6Zxl
zQ-CTWzdYe*^u~Xjoo<U8$xlqa9>K1g!>>90bythMR&*dR@43Lfta<6g4avhzVZI{m
zs1*J56i@z2WLCL42&8L|y^5gI4|loDZHtlXS7Xxw=yUCGca`!?*QW!b$?Bp7JLTDw
z%GHtT4yd%jl(Fg^3B@8%^<$7`4n;XtH41(>8Ux;MlpR|eDpV>l7zjyq5_fHYZ9dWv
zW%yjn*EKaYcI;X-hB<6fq@&S|vl&-)ll4P+5P-EXT#*^yTXj@c1i}xv;)81IOEMpK
z#2S~ULaOJoS`|W5PgZ9H8nAf2k%Yc#pB#|Je7-V7Z=WJ@ML-Glk1Z%P@<+dE0{xD%
z)u$O}&s@|}@b_asd9FurKzM)U;T#d3AQ?XaOt_%Qq#F<AHhdCnWs-Mws*G^&eUJjD
zLF$li+p@z8-ABg#ng`oLf2Q(pkM7ProBkbLDu9!oFAdtAhn)PrabLe$tqe8McI|3d
z)?UdZx<G-$WM`Y6-Jl?<`qlM|CV($^^`P14iN1VH+@TZtN?&C+PkTg|pPzqK+Z$&>
z9xIT(kR=j=a!t-{5Pu{NhVz?Q;ph@|ri$oh7#esREDI$bs<GF)8pw8pdnv;`G+bIe
zU5PTUSDZqN7^~~B)v^N496v!Vl9(97mypkEu7F!^rfDTu?G;o!jo}PxYGT9kC98h|
z8l{7Rr;KebS9#r#75>vFF$M7CxD9cIVY<pacCN9k)A(>&0n(zFKi>{bq^(wxQdgv>
za?c}zoZGW0Bi~p-U6jWIC<4>|fk?Z!9|65Kgm=PR#osCD&lV&i=4)wA&uw1jZ7I|%
zqh09IBLh8o)!w0MX0FBgtonhJ`LEdSMU`e?IFr5j_oAhQ!CD4j2KMxlJ9ugjHXUgH
z_$=^Ey>^ewSEQ3xI8_ty6F#iLU({QF9mlf)Q#rGGsI<u5Iz?9a><U;a+!Fi#02AGS
zJmdRe=*SEqG((#9ybrw_7bCQDaLfZ89A^tWjC%PE*Ux-uU}%~$OhK7|F$S0rQ|_d(
zK904-o=A#xwp;a&qg4cf*Ts-5n-=+UuyHgL92On%mRICz<nISl#+Toa4S$koJzG)Z
zk#s}al_9S|fQFH1&NBE1>rgp;ni+OysuY5&+@(5?SQOr7%_|Eu0O`(JNUNAaU?ILl
za)^OwA{`&cHbEemMQ&EiCF}B4-HFo@gUUP!$R!oL_~6BaChCsp>vyalTuEuqpf^1$
zwth}u_3*BAfE*C@aE`QQUw&hHk3|b0kb7kAib5k#0z>BUp>q!H^k(GQkYj6`<R@?{
zqRI3IZb?I+zA`hK@_haX`AP)$XQ>08Qhx{7m(FL$NCjw};&dh#CQ&Fc!vzsTkFr6h
z3FzckR#n|fZmlEU1sH{COL>IA^B<cOyT1$$U!NgT>bkSVo^6Xx2<EVMY-mqPT<ZJ3
zQhn@2%LmPQT0|kr-ypgM;GA3T5xiVS=x<X+W~(-CE;&7S+$C_!ww*xO0y9I<JlSb9
z4)iU?@?<CkuE~jI;MdZbELO56p6N75TlfpM%*b$Fdls=>sigMsCf8wfPB41?Ujgfo
z_=nBUnpHEiO%r(hOW@H}Js}$d;@R~DTpT!igQmlPoI<Kf?Y%YOxVm_a$fbPJP>AHj
z(EU}(Ob-}sz8Wq|nah)L4|H<Z8s{+tf^q<7gVGE3iVRtNL3MPr$P(hbBDj5Q$r$$g
z{&d`T?KX-IZ61`aO8{B~+Q%`C%b*nxGSPo*736RLF11{H?y7@0q<)q6qxiX~T%N9y
z*%tzNZcJ6=z5G^`ah>hSRpYeRo;Sh4r%B3X`TF?y^faJtS9u=(mTW0a$nLG?@zRo>
zBJ)+zmyX3J5>=_fo2zfT<pd#0wBO&7FtOgMRWH{=@i9t|b|{|U?g^{tYZv^uRTguZ
zK#u-ISAAJ?8J*L3sF|cy;RQi6V1Bd{)=tvn%vpTQwetYJ(0qgAd$2Aa(Ha~;pK!$!
zb>eLg>+U6}9NAFT5)#t5e8RY8rD*TP7AVOp#EBJ>=GH)lyFezUcvU(l^6_*DK9v0P
z#VV`BdKJ&tp)*y^!wZDwHk{KMh8j~Ka27fX8e*7yA$nBEjuBW2l0U~Nf~XH***voF
zyKs@Zq%+L(lNa0-a;ZW-H>QqJ;dD|DLu_a`b_a$aNk4j%<dUYGO#iXLv5WSF5n}m$
zlgnUfLO!pMd3DGpRbh5tZDfoy+@UV=1yoL#7IbN~FG1uF@pT8hi;`6<DEbbSTY6&K
zy$L(F>mOB%zj*k^Vd>3395UWzoDrdX8-qE8lL!TxVDJqe@8!xpA0&JhM%b@ydsdbr
zPQO(&O%s9qbCI{}y<7sP*j&oq)e54C+BM6OVJ?AP@)P-%2xoJ*4Ebf?+;)7^*D?~H
zhT#f%?n1jc3@C&z+9E!W=bP)%e=LbYE>F;5th~B?EVbO}s;L*NN>AZ*#z3#h==Y}a
zjPYu}dq*e2&JTAU&_KzmC6x=`2}d!4Bssp^-Yor}Y@*=ERgFu|F&y+&cg4{c+46!q
z;V2VejZ9<S=NP`9AR`9Y6lD^BvSXH_Km_OVfJGw=FGxy)3zgZ^U}KIJHwL+JTBCDo
zD~4wd^RTj`rC=qF5@B->ZJq|c**`YD<J?L)Rp=XeNO@0Mb8_NCql%S^hu0#J(o7)G
zD~>8Zp?-1skXIX-eHaQhCFNTlV3$Z>re|h`R6N`&ZzHcqKi*GVO7)h@0_@UzA>W^%
z^Bcb@<syC8r**Q0<X2!N1iZU(`>_84i}x^&vH@?@6)bkYm=BL*9um&T5oRXC*iWo=
z%7cHy#Jj18Hepg`cu$L_IaN<+Jx}%J92b8Rf%1P`7fku>l?QNo#Ea;l?H;m!6#&r1
z(+E5Blbv~Sd#X~6Y?H0CueqSf2Tg4Y<2>H2cvpD)nbt=;vM*a~Gik|&`&b<o_tb2Y
zh>>6xHn@%cdwpZQ1x~61J1Y+|n0qmkoCB>DOBFIP?Q_HWe!W4xiba+6UK&L60LQx2
z)@zGZn2_-0Bkni8lBvH6Inz^PKPf{#!x2vCwc$ZQ^ZN2vqoVkZoF!X!ssWDWdVN6{
zBnxC_|7*npHqGFsv5)Y{Xjz&Aw6%|;0!D+VIC-HX_CxI|<%}s-RFpIR8On8CkneB2
zH^>`4r(8xZBgBTcy4Dr6FOHn(yBDBC5P1>-^YA3C5gVyrm#%KM2|)Q%HTpI>A&#qG
za-{XX$zRN6#QrEC^Ot+zfg7LZzK#!12OilRV`^6%b0$M{Ry6|J{ja2GWMsgWjf+Ak
zs0<=iKRAjW4@^PF+sz*sQlfaa@%Ni57w>$vBWZ8+d`4|=$kd88N&=vGWk(kPI7mH-
zsLx>OUgwEG-K%j8-GGZJ`Pro^rTq4UqeehgoeSim9mqS(^xT%lyN0SipoB_S8F)=+
z-01RqnG)*wfyZQ#4S|T*?s4SFe?qh_+lqP`R;5U9-W1`5z3+%nbI<9&jJjhtcxb77
z=63RF603s6uSP!Io))S8zD~8d2Z?gsL9<&<b|x)038EDVB6O5$1SO<Ato(BG(@-1l
z-krbn3>>>T4SSkfOlo8aMkC^O>avq)D5|TuhQM##w&Co(cc-s?%@b3Q;^=`;1?7hU
zL?hhTxTKfMFesr?V-}-+Z*q0{*lx{8T6LUNpl&*$Sbj`$BL8{Qoo&g@8|=Z7>|LX9
zcqh(b!sC)`XwBA38zI9j6T!z9d=os0Ar%bS)j7(6ydaS7(WZV=U`0M2Yf7#Y+{G9r
zV|NFt6}Hmh@i%~ufr&ihOt3ma?*?XZ8_oVVFcqHiN0&c{^F$TdBI5qKR+6v3K1%Vd
zK@NM_m9ns#ToL;S1Ick8-|JuvpN)L$TMA$qNQti-lmT4=^9`Eb>GDbQHzR$`gZnnR
zH7%c@mXz^*JpF@bW}=2-g#L$~O9c@N>|a~s&lPc_+$#^neJejwd^kZ*G+ce){kFrh
zyWkdH!*)59VXa(8099nc{XjUD*#FhO0~GY=p#iAqA@XhKaSrOv1KaOCv3HQ?+_Wyl
zy9TCQMXS6mnwU!!k;3sNe#<R@AWb_b+sUExxGJ)_kuRvL^fsn9KlxCl5LmC=8b2TD
zBeSFb#S!B9U6x%TpdRy6e;Wpd(EodNp1WffT+&vslbD&ZPi_ULlVXP+_9$P>sYT*d
zt70O<L7zZY@*OGAiAl!JIwJ`9B=u~nI^pOU<DM&#Fm=6igQ}BCe6v`*Q?7oN`as2e
zk6O)jRK36!OR7P~MomgDHah#V)#2)R;g#v~>=UKiPj9lXe5CcYCU3qkw_T<s9JMy9
znWS1(7FDBfiRt>3n;qJx=Gq;4Vil48=}UAM_I|LU+4EJ9MnJS+{-SZq*RRe3z;Fw#
zqkvI~UtIO<UO1^Hxj5wZ8J`v19H=?kn7i3B%eWr5<&d-qZ^5~@!RGu5gm!Xz+L-t2
z{E*j{2jwCtk3*<~2q-p$L{C_Gf3J04?NX>6UY%os>-}_<iT5gI`7X89m0jdAR4@7V
zS!NIZq3a`2SSE~yahE6mW4mj33@$!mL4nxq<{w}{eWa=;=f24UagfSpx>~Ucy`mY<
z7&^Zvs5IZk?*47as8l(AV`9FZfBj3fq%jsBI3e9<KHUbOy}4Uu7rZNv&VPEt=C|sJ
z8B@<&H)6)PS(yH(NPWi$eOwr-<H!rXi*)G$4cv2Ah&A~vN~Jb=ENZ3jZD0lA>SxZL
zVQ!02VkhIOZkgLQT8<Gq*Ppd`e{}zO5y-$tW5=)dnCJ*X@2e6_xD-(J3IT+t{lNBT
zx7h5T32oLH$oqO=YB#2hY44mACTwK9TK<-ylWavZA%84Jp6bFP&H>x*jEP|5VX=py
zYl@bN)FCZ$9duaZde21Ig{+dtFRlQCIH%#~(R#9uK2kvjIk59s9v>IF?w=({>8bbC
z7udIpYUjx4kuqsr9rw>o?^GYH>I_2hIQe!W@m@QzKXce;)<j)IeGBL4;f=nL^V)WQ
zBpw*}oSYsXANt=o8*v4`zbbvnxxBXj!PQcr)W6HRK400R0a?5&N#Xe&_iZwdjlK{D
zDyEF8{deLFc!2=SFZ|%b_1cP!$@{kH_l;JG(^aYbrRpbCJimi12r)oRj=-?GaZ6X1
zZ43ZLBvx|e*DzfTK$QRn%H60ZeS~iPab_l=w`A{4co?!E!s`VfE6^x=ffN=&oyHDD
z$I82l1mfG2l)>oXKIYYqoC>7t>ewR_s#D{|+@HETB$CG*YwAfp%=MjfWmA-lp_?31
z2%0HN5k%T7GUt7b#s<Lk?Zl6{^hS|)mk5jNHr97J5`EG!)Wv`rnUe4y3$x~9uDy4Y
zI=!O)e$jJCfL&<p(@a|R$s_(ILfi>4;bu79<zsg9RrUAr!m6=_(-%$*#HXkad_;Ep
zUXJzV2%?qZgJ^>2FT1uzK1PQz-jxIS7`wsjgYD4FXa-CIt)>2cYt1PsKgH*t)VSj2
z1JozYQ&UXpco}B?7N~#9{T)O*>~XTy$Wwywvdo@f-?0+QoK!S04@>c*!|+!^!p9#;
z#FB`e(4}p9Q}=V5Ll6Frvj@0+iIa(p?8!*3&yf8wt&)-POg<^rH7)@DLW%n|a5XmZ
zDTPMp9v9jGqO0iuIRq`5Qc~x2C@B|EAKw+i=DF1wQ^CE#MjvnuAwyA)oy%WQ=GMF9
z>KMX8s%#A-_(uBQ8$|i>BjY#oRZjUFj6b-h=i!NNeF<@1CLvEOx6;DwMoMBlt1t(J
z<mKHt>0#y)!fAApH4R$7+d0A!;ivs8%I+5@<e=!y!_~#gtZEg}3cqlSn&?tsM67+C
zewTE3v^_m#zXbI*Pb4jI`W9`W%irl^aQ4OVFiYIA^9{w4Mab-+>%XQ35`Z9ypEKas
zz58wn9yM=3-UUbjq*_PH_iK6TX1UBykWvSfu>@vgVFdk!+qTTiKlTK@X~|y)cE#}6
zYK0&v9Wf1>kh)b<l<W~G^;=7D=%LA>y{H_0P6&``Y+@B4cG9ZP-!pwqm)-f|D#}qR
z-(i<~9fNka?(ds1t9K)L=W&1Z;BnNiBLetlcttmtuqVROAMhLwyLhufwY2YQ<R$*?
zIq$8gt$2~EWLc$)jer&m?pZ^hSXG6>rK-VU2}{M9w)*jC8{QM-L*HdVGS>)K@7TV(
zC$9vf*^Ra0=bD&oA2LT=+uj!=a_P-9Q!bUyn@*C6tbKN7QD-MrUAZ;*L+)u(pUIm2
zRGWp`7rM%3R9Oexe#FJ*zi@6Kq(D6h(=sAo&|3i+X2carYmhle$%*IO$#O1v*-QeG
zpwt+-fa_c5ru$}`3FSv7Y)sQ8;4hm`0BxQ<OoO(*IS02=vjyBG1S&IgenT&5j&+l|
z4Iq$3rJh<hr+$lTwIWU}*%ufTBh;MP|K(NMZ|)y>Vy(r^0WZ%@SLip+4Je4Vq7$ay
zL}pTK<KKj~mz=t2a0_|+&J0U9PrG<Q){@X19E#kh-tIE=GzeBAV}Bhq>WhwSo5_A%
z5s5*E?YOiH8?k)l`uS_k;SP(oPAjDx54Y)3&QLN~33GB}!|zynXje^4jfY8=TtQs>
zhonpE8io^@@?Zjc1vqebk6Cv7S5L@q14LGc$Oxv~)6~+%{Z%Vjz^=6o0@_~?qvm;>
zjt5N8`F;1!^yi=r5~@gBy1Ib#DC2$ow8C;;nDJpySCZ#NpyJ{Z!aA}~^lp%Y5`iP+
zT$(*()j%ncE^^?<u*LrVsB0#I3$;V&``d57jWL>yB=+k*(%y*~>hJHry~Ew#_k^p4
z--A}EH%2%?4RRlG&#9U)%%3Sr+MskKGE!q=2fL6gx|?$8fbz!nlCOBPxWLl48x{BK
zeozmE;d*Kb@l}O)71dpHQ&Xm^unf6bmcTE*br}HR=u^19%95XIqN(I)zqve~3=8K`
z0g#_6*ts4)E`YQnu-Knn2fD?3L8(fh(1*7}c$2PZSAV1J(5nGpXW3&bIWQnu0#XL;
z^Sxq<?r|+Lo!a3*BLeDH#SuEki!&zg#5YH|mU?S>)YNK{Ew;EuM8K5%g#a5y3EUx7
z0DE!K7IX!r{T&**st2lcMBHeV&E%fBW8iQ3_i{S{4p8Kh)mD?&OGFjjM;BG%Eik<M
z4@5f8r5HJxAR>%1Tt{eECJKL|w$?v#2G$_~s7wyr%ai}`?vAD>{^>MRs!zBkjo>YH
z_>{oEa{sHuGP&@+tX-<rTTj;<8~gaob-Jjk<kfBDxh_Mmv9oVJ!F@J)d6>TP@5m22
zAd3Uw%^VSw8G3+y*z+;ZBix_G;Y?0`ejy1cY&#{b7tg_$#v>njc6;W#1Ek&kByX-}
z!gJ~FRf+fKH^DCq{zGdZvBS@FTmUvoHHmlef?DC2R<GDTQPo13R`T;E?7zP@U@LIp
zy6#5ucY&H{O>L*@qsqz^wU;>qvtObR`6}<-zPBFr5sojL!W|KJ8o|s}Zo)TG$q&(v
zZ<(UWg|`7@Qx;|aW$_ptoZW;u4Kt%TWfQ%w3**710gBQy{}rhJ`<L!71|$ghvNVJC
zBquWup$J>G6NOBnl+dJw$Yl!F2;)A*!=JU2Ao3VyxdoSO&@|UNWmk$$ygsm4!^0ad
z;lXpN38G6IDlF2`5DkH-jq5R-@Z41H2Z!f>ufpTt2mZ1>GgIWm8jj6Qk0<ETDRk46
zKCx^-)$1`Fs}jQMh(w|@yB~riSRR|N{Q54x_ip;NnU(mOpxd|7w)Ba8{$`|UyZ||Q
zP`hj^J@QTrP`4XM+`-HO<&LT^7jI&nbxRHN;<4)7D)akbhy)ZAFw$XSocWmlGc?#w
z6!J+eN1+V6@xzlqS~!CWijZula*Pl}LioduC?wbU%BM&5ifd-LV7KoX_~~6$l1#m)
zL<9|0NYxOri#EiCf6MNTI$|oe*?*UW^Vr!;61TSZXQf+a@Bf5{PJq^5w^LFQtTJnG
z2mLAk(e1Kq>+31<t0%o2DsPRIMHP|z>2P|G7PZ^Yte;<LM+Q!HOVU$7>+x?<@YXFV
zCNbd~TtNbcq~U;T^CvjIBwTC$R0(LC;8Qn6jTe>m*`{Hd+dDq_3^2i74JBNQE#rg#
z=nHu0fDV<%-iDnG(AuOYl#A_289S&q9D07_?+pXP^J^`9*EW+hj*k7+JR9?k)~bHA
zT6&dX#GQZPm+<h!lv4(M+nVH(TB#Klh`e^H?CQy9SK{6tCN>_)PRWvvsaff48&J+&
z08mj{%2nk?Jp%)SjkJT?$A{|JbV*4tM+sn=E&x2&lIc}d*&FxD-+Y(dSOn%!JTkEB
zZi2xf_)9su=1|5N!`C)#vdTtQcS5O=-e0pQBtLfmB+ZyOP_x`!EOAn~{Pj?5>P2z=
zVzt8NHb}#)S4_oOsviCu{%>ih{}snOy8T((zR$tg+qLMXR0s=s<RCjD^_2UL7<4?`
z*LfkuC>fNuYdERFu_&*lg|Gc9h1%Q*G(5LLh`u8S=>JAT_7%L$)=NcB%=dsO7T+AI
z4*K;g40Wj54zb_eKhXXdZu8uvd5g!^x!xE@c7Nn;5Rt)US0FWIlcvj>Ts=hN+6PY#
zo9orov6dH)iG~6qRMOr&BYhx@g)EbcgoP*rqBP3pHbMRSTE&876kEmS0|b9BS(_Z!
zSrpr&fcewUuV9S|>qlJi{R}|+J>7Ck4(`nYnnT96c#`e=+kWkz3?-1ut02Z6Kw{Z+
zJ#D{3hR-WF^6rpa>7&z~cDg)rM6`#Nx1298!p<^At~}TMFrjO1-z7>u9*dq2UMV)=
z$##&Pg))b!*9NgMLI8<_p(ZK@o*?q5RSipz6R}%cF|#rUa&O1Ry^YHSj0(;|`mrc!
z$wM*%P;1iR*w|EdFRA&!mi&$vgW86A>|~bjCb&!f=B;^L8v_l*s%=E^FnhJcARA6#
z|GgfLjMt_bose^KdwKsoV22&p;rt5#>n$r}CD6CiQE7L{mlxe6W+Kw`)&Zd)SDEYT
zWsTVHytosZoS6Fv)bU$xtw3l)DnSiv)+koF9wj`6u#;mjkR>T;YFg6RK%rB*1i$JM
z0BKFVYYK$K^aG|_;O&&hmlzg|jQ@#@-n16}7PHrIwvh;=2w8>Mm5(^E;iTQcDsd|d
zT-De(0|3>Ris(1)7n(sM0+yhir){NH7)NHj1JC9Et=Jwg_T_sc`O1~cQok!Ht%tV1
zz#2U==~Gu)Nq2_4M2l|BeO{@>=SurN`bq4f?GS4GS-o@RvHQ}Xwe@K7xpo4r&V0Xh
zbrNWM9hH=XX;0t(Z+z(o+u5K5{$&b~v49(&u}Y-hb7drL@4ckGjhsA2MUCg{W|Cfb
zO7lve`RsT#X<SXlC3V7)iF&4Lg#-PNQ_)AK^89?(S|K18d>JO4-(8>mVtEJTanG|f
z@SYqPdivWB6uZa;lt#$K`gQ|c6eWC96-}XJ>~i$4iOxv=Wc1z*wY4JGbt}}-zw{$-
zb5y5QO5B6@a)3()C`cj~kE2vEvRX$qYQpUCRx|puQo-bhm+FH3BqeQc)UHWh`*!q?
zc{w)F6Lm@f+$4Y2Up&nrUv|~R`5Vb1`E5`d@^`xs0Cl{%!gF<6nef5IzlJ{_L#5Rm
z0NKjS4`eJn*l;%90<EMYnyMK3{$mJ7Z*pjtj+l}y&2}|V(4E!8=|n7q=Yn}P(EeYj
zkACcVpA{TO{=c6M{An2U3ZRQ$_KNnpJWq*x=d<B5=%3Do<ME=9n_4`2=M*UHxQCYK
zn{M36ev|tqc&->wGvsDBfW%FkwVddNBG7)$nc&ae9MJuX^8+_mxb=-xCMF5A-9(!%
zY-c&AHjPh`KZVX)i!WNaNPz#Vi@bYs=cq_k>ews2#>UsLUk5A%lru-iDhou~b?%PR
zXw-XkDG-!KsCR0MzOuNb3b0>_)mFxs<XHoifKKefnW(;!B`n9>=f}&qchG`{--hk|
z+xODEH8O4!tU+Bh+U+qRAbthxYyF}KEMtTYC*8HSs~~*GEa-XbUv*H+DEJ1C3U(*0
z*{(Ry`p%8;u(rN<0OxZ3QPcRv9hd`VnVONrXUgDLXgKrB4<>&|{tF@yiOsXFj;R>8
zIj*%EW~J9Lf<cQR=UCb9x;}CIX$ECga-1D0fW5-4x;Ane@ua#i_c3E<&NL5LV`Ua^
z(C^532cGahmzD&XZZQLEAded|m<~^wK5E<Cho?%n6X@KXlXHLjUM>PKu2x8&5eNW3
z=Q2W4d+0=b6BL^iT>k6K;cY7<bT#flXv62(#W~gy-;*<A)mH}WWzL>?U)*2-cqECJ
zB4DS_&ogQkA)sYQcOoy}G41(rq`IItVh)rT91g{vjpKtU)pEeBiOp*}?rE1F+#{Nt
zt^enx=b!xTbW1J+Q37<SlHTQyMb;oKWE@~FlaEj18rJ!w7@43nceiZ?sjDB_P$&LN
zV7-B;mB+0*gpzv<X=o4V*ga(1UWk8iIA5c6OIda{KYVNh)b!LfW<PUH*W8QD*1BTq
zsNmq=zHgW===FJB){rQAhQ43w5x)Dop>i~oRxxL-wVa14b3W&)q11Fkt0Ag`UX2az
z6EIVr{~`lA9KF|XB@TW43ft9Qoq7N4jrRj#jQ{yR`$Bfg;kmRin<TaI-r<G_BCJP!
z{nt#C6A>^g!-CfV=;20@-O^d;T?=N}RQN-#r(t8_{qm}w-Gvhjtf=`uE70QBbM+4#
zew-%uHE3{8{GKPIz<O2Dq}9?Q!KArm?<8<sE9LWOwzM9%I<MG3ttNm*h67G*Jvi!;
zhs0%y5PL<TH|7flpE7&56#YHEv%wg4&V}n>oKzx1D#yXw;>_x8X0k{GRPjTbu0ldx
zSJpt~4_zt0k*q$@%BuCTt74;P!(cEh>XsB9lTAw+U1@d{t@M{9U*iGW0A>~k`%O&W
zQE%V8mv*Ln3vm2hXiNmxlWIex#y9~cj8zqw`8^!|=K=|emwCRVC>+n`CD-=;{LJ_k
z(|A$~%0IL6lY0&oee{AL*@AGk^CBB<rLdvo#FmLLQb7_yq%+xwgp=*uJ>gw-r$jRu
zWF)a}1&bwqJMT=ZIYBNx4pK##{Jd$?31J3s&G`3<O#s6lV)-ODytGJxT731%<kqoP
zJI@z+3!K8hD48lqBf`B?>EpR+Z(m>ElUqyL&&76e4Ld)GuiI=9wBf98dOXVX^#o^p
z<wh7wPN3s&60=fNcVZqnu(fA`h$P*rG5)iZQvs--t)95~NgqX6BeaJ)Oj`i+Lybu*
zV<F(Ii^bNiJu9e~PL&0E_ol{YPacFTs%SwCqaq_4=}_(a2J{(9l~&D%B}}*K`iU6M
zk1s4R7!2Q*)@C$HH~EjRK_TAZjvEsRWBUtb_0wCgEQEEm_S8j|VFcZEF@TFuMSFb=
zH{?H1m&aJ<*v1G8lB&)n=~E`rNo11ame2ba3cwGdX?dQ1vGw+#`D)1Q$<N-@^2@|&
zckN((#rOw75#`j>q<LXnPcU^Z-<#e4>E4d2Rg?P%2vbY0?Q^BxenEp}AfEka;jRB_
zZ+e3EX0Uejn{S`0R%i`(m9$s4H<^1LwXjs!zd6wloetjGiaC#SqrtrD%jf)Fe|1WE
zE=0^9z^(T;!}>FBxdYe8^Xzuxt*E7=w2|?@=+?vih&#P0*+T67z)o80z=8JIAO9jv
z_uhOz(8rmQfa&xlp$t~-!h$IPGrGFEq9eeRXQ8nRtHeA!SF)FVGI3M~4BQ{YU;#1c
z#mZ=qB~vg3BZCEW*pBS?x7V9^u0||<_2<ZYhr&B(u0)0SnL$u@b_P2VOAc&!L0bq{
zyCUg()aldp9%yW;+G61dy*B%|%7Qr8o}38>x9)qpcTVm}Vk0Zb^V26pDY&${u7Poy
zq9H2Q=KA?=zr`1R>r3l>Fm*qhTZ2jT_G!zn$Fd3hm!2Adh{cNi-COab;F;qHap%xo
z%iYje9->;kZc5j-$DH;#ql?_tTFPV9YrI6W0!qskcl~GYW3^Wjh@7A3(t<F3&Vi4&
ztnt!W^Hlz8-kHxj!9?I$TYozg`)7LRY2~yzPVn<}lfT%O?eh&yt~tsUAYGtLIX(E3
z9FqQCH!bChQv1WQ_nh(YFz`$MIs9e!gno=vO|)ORrW$$XzoxyxIJeLM%!4&~ZcfvG
zAodk~DWK=1A}ffpJ11jphsfki<&0$HneFwqN+>8@y@VKq8{!;oRYzBA$EcNzi5*2t
z=<WOlhM}jbdxGQIZd~bqvWy!Jc3{<ld#za5OfF7dN*S~a)K?(%?8eTJu43z8IW9>Z
z%Gd?wRmmKr;TDHp=LWU5>i(F9pR|{@=WjeLjBDe3TaXwW4jwzzJ0y2uS5oay+{Mr4
zY(giuKHziYJ9GAFC$_LE>e#G502A3Wuk`(}5%QP*MwpD%R$gJB*f=66`j@@e4UY21
z3u{noHw+quvU-~5B;ep%E^$c}oj(OXe?1Map;Gzwoc9vB@K7U1SVNxYCziEP^ERUl
z4}a0Ju&($)fBM%k9vFUur~P4cVlTa&Kl;w(^O@(C)mctj_VzIxv%lgYn3ILs*piaD
zAAhhVL4J18Y&(NoD6}~^PWa;!+T-WqAD?cooZ}e`7MQMOGX({3P_M>e+J<omEvV!2
z_Ihf6|MGZ`c_l`C*TI}UfkG0W)ZhYjN20tB*KC+uWHD|a5rcp6nLUUya-`VyzTs_0
zQ+u^)<ZMMf1C{-6;@bH71RVcDRJ0QpGqdkhjT0_&GE89J>K^U1Dn+C^$+O#~BIc5~
z?)Kf)(q%#dzj@lP!=B+@<9v}&Y=6l_&*Fl3UISw;c+9eC`U2MsCuqkm(265=X&miw
znV9-i4}$r>SMV~N{^kU?LxorCkXMf=fk{ZYi&HsmRIYq3s8avH9aX+kUU`%=#RNhH
zYoE_?yc=)CpvIT-R8!|etoH{&mp!e4Cf6=@5kuVfPj>PrzxN*J+TD0dK-(mrk(%C4
zXUSuF<2{vC@h3{yqo7CufSxl<lJc#R+oOSUsZc$z^!z77Zihb@$O<0v!W@Mej{)3O
zU&B+I>Ot|^OcP)><ZQ5*JRc~`-!5B8g$(DhBLUkrO0p;=$z#%}>P(_y345D9O)ZdE
zH&bn=m2(n8`#j{1nU6*FK=qy)c3K{fCh(yq<LB<Pt^%{Vm6M!Z!U}4iw^&g-5QjRP
zwQ`u1aZ&J;#id9q)=Eu!UD(lDwM+4a_NFi-+3O;^$EJ1YTm|u>xCa;+ENy2k<=^wl
z)AyZ^emeEryxza}6wqdqLD`4BZ&t<FCc^M`XQd8YEmVC!A9Bu9Sln!Fg0O6`N?!wc
zhN6{~9gyS_p6uB(;_jQqf^T=SRGP8f?uh*~W}@wxVCdi>RkMB5U+bhz5Q-AAyg~y%
zQ*NZ3$qUs6p?6?({ZIWGkFlZCKdX<o3_ZlHH1AtL$w*CIk*fjpISu)%pyVZhQ4=Wf
z;U9BAM`L3#VC17RHJLXW%?W>O^;@s>>QmwB$>*2zH#5d+`R+E{V^H@=R7>)hRA_44
z5!_J57=+44W+(GWgU-bpPIlekl&&w}^3Az1S}vU0CM-UF_eA(O4(N6q1t3WfCVL*@
z;EI-cYYpu$h$fXg+laSpx!<aa6=JwSHJPb{*+7uxruIt&%9i=!r4weS<tkr#ZNLw1
z<=)U4(~F|%S04R8zP>x2?e%>>sC3Zk)T32YhayJJ8Z}NgMQhK{R;ayd#2)R@(_ydL
zr!BQpvBl~Pg0xjliPZ*aMJOpjM1IdZ&gkd+$M5-rCNI4De)5d_zV7R~?#cD2kwtUW
ziyG9#;S0Yk)(*s8is^_fwrm4QmV7(%CbDQssU+~s71<tpw6-JsM1hq75;g7*m);T1
z{~PQ8M$oCpsnh+6zv%L`!gJB-1J6oq!+=O4NLWBXZJk@mZlkvj&+bS4NvgM(0znkT
zH8rJ{A~;%QlS~?E60bhGV@F%WKB$w)G$d^0(puf3m0$z};6k*L2V}EyTfcELN1MF}
zEub&gl)UarEyqlcHk*hfa@MFcjyf}5(6rG5qZ*dZs}OBNaB9hkO|8+x;j|iMh^`Qa
zqF^<OwH;P#l239~C8yl!QNE}Vw;5)Qc+p*c4y@RT5ZeUI$WhiAu=|Cd9JnY2dvv^O
zKH0sU{ukvqiT{W?rOI}>t|jpVlSS5nTL%rSyXE7{32PcLgo>1z{pN9LjB517UzVpg
zMPVAr6J2Bmx8LV?=AFf}*?s&lZ@TlszER2cZdTjuK3;2ofo+$kV$`IH<n9Hhe(<B%
zWj@Wr2>QzJAf%{nHVCRC1KlUc-U0rwz67`;Bka{vlGnkC1L&k2tcP5-`M4_vl9(TH
zvAJ7OIyV6b($Ca6=1Zn?wn?Wj2KITdnhg4TqnNHA8tQc2*alc(1ortdAT#%TWuo*<
zglCBz9c#cb{Col#sqP_Xjc%F%0RX==@q#LBr5I;X#!=7{8IQz9nQl+O$xDc2A7$NO
zZU-%;+0D_;C1#QLH8AS_oGP8U&`slvGCI}EfuvgnKS@*QcWD~4oU`ioYtZ&CGaQ@x
zrSdT@195)=*uw?_)ik+$V}lh}U%!#JKz4fTj_E-;fIhd$a+(4PNFa@*PU`YY@gXWL
zgf92-azvt4!sv*BO#Y!+yAo@K-C3D7{9mfGt7OmG4m@}-0vnfCYLp@}YMKua$7(=~
zxc}s#3s5q87q^+8N70VC-BS47rya3rRM|12w!HTN1#rjy@ECwKxWBVgh3y!cXk%Bo
zN8yHq;?doH%`G~6ZwtQLF%Yy;JU;2JtHnA<;#WL#a3V67V18P@%#Z<+>VHT1IUuZk
zJ2G9#f!<Qp@Ntzm(0ipNCoSt>l(~ETagEufve|uKBLCMT0g~}>0j2yg>LRT61@wW&
zj0tL9mlW8%1pM~TL?-gqbX1^Nx6Gmi!i|Ne+uip146L4mz(B<iI15mNUudrzFI`~F
zPI9sG9Dx-+!d$UaOUsU?@1aK=qxu604(;boVuGfdJHCdu9T<N8ExIfxHoZJ%FM8y}
z`13~#+PMX0s_#s$bh4eJj*qCUXD~LaA993wS!@rjgkwZ>T|`DyPc#8Xr0s~?Hf2j%
z0h|i(dBdDkhgEN`6gy4Yud4lKWZvpnrRm)Oa-)MM=|Yn?m~+2;;)rM0?aRW>Sq^YH
zsrNa!H>@tJ==<A?K7!OG`p@RK(q*5x(AHT|M6rt7je6D_NC)w{aHQxv-UNaxKrzQ9
zEy->}ZUvC#y&jpBmF;WCHjD!P6>9^;d<+>O*r8-I&@nO7J28V4Ue9X>aO{2HDl(3n
zn~vOH1tdSKDw5Zs2>E*rh{X5I|6o6!$rDK6bve#wGkYL9lYew%1o$L4f1F&p!y4HG
zhtX51E824k21{D}>c59s#N92A5gE{<<iw^l<+zVus(&(oAJD5nkL*h>h=RY|YVKV>
zY_g!;(m2nVyhTRGYVQcp(Ti4kcZ)aqnhbTmhGU*bBGU5R_55}Ih%Z4BNA}(ogxybK
z7qrNCuA2mEH)<vb8l|zzXtC+Pq;@x-L8%?IxNiP&-PP2#aH)gLK=kU%<~d8>!=2af
z?INK{aSa9n4gWU+V4?uOEPAqbI;LxJ->lEH7Mm6<$#==M4nRKuP+^cD{Bwl%dl>4=
z5H4Y8l9o$?P?wz{eeQ@d@kce9A`-VCMEEapp1>0L0fEMw@$vAhV|!cD3FLahz!=vK
zmNz{|W(;)k1Ev?>ss0W-B?>Yb((1iZC|vejSety}h3Ipo@SLq@|44M$pO%Zi5NQ9(
z?^T3(Tw*!t+5k-%b`!J?>|}g0qFfquY@S_xwX7W2&0c8sDVl8fRK~(Z9)O;FW9ZcY
zfSYh_B|X^{NdK?nS|&MOBR&v-`tCOk<duL+!+yc0*PmF|eZb@pu7=iukYDg1=L$3X
z@@1yfi>}_kRR@5|+b&zs3;@H8cogu-#|EHdl~57qms7RgF%eWE{CnhyIb}tUK^@di
zfnj_EF<?mxb`vAWYisW7-)-|!@IcRE=cIp+%Zux%_Lc+MexTG<F3n6G7V;hfwP%Nh
z?Y6=t!#f7p5_3VFjA~dCk+9yqp?iVIc=o!+DXgfaC}2rgGV2A)IJ3AWe^=03fu+S&
zxfc*AQ{&^~62C{KlS`567MZd^4VtjCukqAZf{!c9ka+us+vmy>ZakY_&o!i!{xd(|
zd#}W)CV8rA=JC^477l{fH%qSBbkF3xS!;SMh!8u7dGt`f#^CP5V5uU-du!pDM+-w&
z0ALP^K|I~T13=1>$<eL?0C83wA_VuXu&~(~qsW)Kh}5A24)GjE2lDLn@yTcpz7#sC
z$2W1}RatI(R$@H71Qjzx+cwM#>rLlxVQ3@e)>==1_RPMEbFQ$zXs(unpdP4xTBzn0
zX`G(>rd|HTOtaqP8hb!uTg-|3=@i42b*s3v-Xr0*OV6+zAfh65u{EhhW%>_c7QC+;
zALh&!w|rmKB)T^Btao60bx~2r#}bD}Tf{p8_Z6hDjKo}<#>_EMKQlq0wv6%zAS!g(
zTjJ#wsn)lr7egR?H(Yxz+mMm>%S_$3;Bx!AzZ<uGo&PdvRHQz2!>jihgnO?IHHtp_
zEqLzW8>NFH&h~3ujxk>MXNalh?5h)o`JX%n-+H`-WQS7M!#_XRo#c^+eM%=I4qsLV
zjEk0iyQ=FJGuuHBP9~egL&gCnkgf&FUn;tz^HHSXj){j4qn3q^WLq}R>atBZl~KS`
z@op4rY+Zpv$W(A;hAD5+$LtCzvB|?@p9klQTgz#<2adsRYZq)?1zthjHn`APwKgPN
z)#dqdM4Wr!8F~K0HQSU<Jw9;Tbi1e_I(%Z$UH*{x{qzF6{LF_QQOJb;$;J_*pTW^H
z9F%VL>Sr`&q)NcNRsIyM88(p<n>;7W#Aru-_o+50(;p(q)#wVsRK@}+^Z^>e`l)-N
z|B^D}<S&y~a||pqkivme{0|Yw%LZ`?;;Kioz3JyfwyI>AxHY}qbZ+jlf!tmU5ug>*
zc8QN<nNNiFH^^ydIi3^AvcGkfUzGrn#eies>4*r7-v@~6v;iW4{lMd7i5>TW^ULFA
z=v*90MOsl~-y5GuGp1&^=Pz81ypg#>Q2m=GaGh`XsK6&5Ae3;32zQK4wd$`}I>HhH
z)%F&0quU({V^zb|(cYy0s5<{)P^0y>+-_P0M^OEO{lN(Ax;{ZoifLzT^Y_H%Ms=^0
z<T&eIP*g>Tzv!f=#8@`<pcmbJ-51xaA{@-N8n~p}@4uak_vudrm5s<U5(<c7UujP+
zx&VAuS2eR2M}&6X-=pvlM0&Kg2;95?y`$ogP_VZ~eo`|xt?y>fa?esq#<FJD?p?iG
zj1yjA&l-SO@tR~@iiwFM2UKmi9O1LCQ>79MxCc~s89xXU*oa0G<t-t+Vg`r9T&`KN
zl4C)duI(@S*WntwOl<pmE^WH-EpP|IgpF7uo5oNQ<r2=;x!OVw!8&E|{K@5Du}t)^
zrZM%&>WVYxFp+2&8r+?S(w(f2(bDiqpDWOFznM6_azW9*wT$i>h*Sf)b8_b5+OZRl
zP#O5J#nr;lTt1jND(Xu!>r_p<eQR5J0^^$%V`md*BG6XeO!bO}IF0&C91xD&K+s&m
zPi}o6qT7v8Q}{HLZ35vuC7<)swygbl1rcgSWcOU1xoj}HkGpJKm|^%QmoD|sQ{2DD
zX4f;=PtST)SLYs^z!Q%bE#TvO?@phT|Frji5r&Nt5Eo&91PwjUBVdJM1JQXXQhvt-
z;#gbVVj{<Hgtk!4C7s_p#Wi>S(ykzowLSh7up5Hcl0oUz^7RaFbgfCb+b34?`<*ar
zv6@+_9z)?;;(Q`F<zAh+Jo|RPBHSw}23YkbCo&p)^$e$ujLiuGH)Ft)S72GG=UrHc
z=<(BJA1rZP)HdDT=e`<fW@iy+!thi5X011k0jo%4Vm^<(I(>tE1Xd`$5c)hc!n)P_
zw$1L6Ux4f%=%}mAZ50*QRtD~kGy{`%P!_2jtiXItB+REu@&yZtS)0;Y5GR>|HZs+R
zQM>Nz+w8suF1(&g-i|6qPHDK*>H3*K^=JSa5~J3+wDZ4_E&tBZ7hrP&GX(4w`qs<F
zz`zdZ8Z;3JqlbT7u^C71<P%Ef5w96d#EyF;d;m3vbp3e9up|&epiVTJfLqMoVeaoW
zCf1}2%rUl`!m}yG(G!C9gn~U1r|ecDU9=uiHfcB}&vD91;fGz@CpA}eTyznRqzYW6
zlLiWB;h;9*UHhq79^@Q)&fXWUD0!vfBY-vJ5Kud)I4s#Mw2javLQtw#1{aSrlUZi+
z0ist6f8i)}oc()Yt@)esMh^F4j1m7VtmWE&>?}&Z(w~~2dyjT{J^bX-)@GsoHrwi_
z=rVC-_Z{WI0Q3cop}e&4<$o7p_0nU6gVxS{wbmPkho1rHgey<<@z`(%`a9UvJBT_w
z4i#g|6EiJX!2JN!>`2^#YlH=}IlSV8<oX(Qkln^jU)}VTqH*1fp$a}9#<XgjwY7ID
z?$6>o)~;{RJ77SAw_3o+(AOH&-PQO?d%`c7|2wV8PP}kMDeDN|w}7THrFTl86QbMP
znvjY~-+tHUS5zg?gVa>lq62_r^P?I!HFXBeov(w0zm7uVc2F2(%>!T&a6eM?e&H55
z4Xgpj3FsKotJ#YxoM|LK*!_V5Xy*QJo+I8j2ej{=2KIlIpLgn3T{Ji9=e-w?D*dSs
zAOUxfZQKXB)bVG);Y1tYqIx*g;cCH6_Dj4k1_ewN+YezS6gwPHUs(10pWY*yI6WRJ
z@&ff$oj3QwDsF3ITqVrxoZFv5<g87qO5z$h+7L$69~lI{)*Q)~)rroAd<nAhp9LUq
z4Df0yZP6Hu7iHp%*Q4@-f4Yvlpu>Uf7r$`?QuW)<=I99vj|STD6OWFU)$d<vycRsq
z3sXTN9Vf%>7b9~(5jxOKESB6kpKL#+o$)`9Z?lgJaJau!kUI8d{~<kKQr5O$+X7Lj
zn#Z6ByYSDq`%_a&C>MHV>X0pYX(!|~gCFF}#`)`NL;Sq5FxlcyT<S`~-P8e)cYA18
zl!fR@TiAJWt@7r&Wllk!36j@6XbMrlPPKq*Qu!MsaED4G@k0<bDN8uS*3u?gEzV}h
zt^}$V0w*lR<yl-aY4XOoUGr33-8H6j;fq1g{cVK2sn#q(t*3k(C9K!O*hY{Ra2(y+
z8Q>_g7J$F13M+hpl?Yq9Ko;vsV~_eTRlK1{*Xl3fluy*s@j@LHlju|UU6z!pj8bid
z%d`GX5rrO+hO*Q@o%U@_XwIX>CYBR&@lQoI9qZo|o6G|nI*f^8gr34@+KV%G^8S*7
zMAt*2I?GS`Mn#I#t2k)=c<haH$#{1_SKhF9<YY)NsIDg~GtE3Rj%@{gQ7b8arTYSX
zeSb_J#sGT&KqG|1#AM=N{KxyxhL(0jOx}$9Jwy8}T&4$C;H=9Z4O-Xl0FERqaI_#M
zW*p)A3anGR7rQA_kjo3Vzb*|a1*kIf_aK&2Il*G<9~JNmV?k|FaT!*V5!dS`LvFx3
zU9>=15R|~@%N{;)4NDGobkSu%^5E-3oEEMwy-yY(0+Fv(eq01aI_I!a$ltb~HZVL_
z$Lu*%$L<V*8jE3c`huQ30w#;FBvPP;crxWO+98V&t`pT4fe$zCE0~H{Aw7K6h~y?@
z&{l%MvJHIwjqcGy+g>7PWZB!nJHv&7#%+SjURj`~{}@+<@7-pcAZ~?Fe6%kxbTyPZ
zy--yTI8GROo|{i*?x<F%s|E0;?x%6pP=jopJNOuu?Z*|P!>=amOJdW3Xc-NB-gse*
zYQ&4K(gPo}YGPO4^qfhcpvKMN`?7lZm+J&4Iy*YUTW0*%rQ!s9OzHRzC18{p3?kIm
zwWmARSIcSz2K4#`8=<AXC8pC$>cpdSB})oQ)((nFunzYod20=fEJL#<)ZzQOv(C9+
zkJxv@8~TG>Mwi?qMM28(FR=oJDPcw(uy=oU$gvLMy8$np>~K`%aP9aP)A>kK&5F;R
zraSjdhDQB1X~oI8M$f^19PDyWAMWd*%|AgLvrat7Jm}&KZIVC#wXD`+^@HLTaE&|#
zafceqH!MM+l7BM^m7k%jxB@6SfFuU^nXSNfw^-?K)~?t(Yk^_r`&(mWb6>zL-Nsm?
zC5h)K!$p@jdNRG{T5s+^Sfgls*UR~pEL`bmWm`DKrY4;frGa~t$dlEHLoTef%uv$>
z4>Wn{dBP=|weF~O&UX#B2dqp1ldM1tS<;!6&bWhaO*UlQC6pcz%xASqNQ+6f*X6OF
zx@%!)*vEkT_7MyzU{q3kENJ-NKn6*60KE6l=x6~+D>;@oeA7$lt>902X;+qiU<Z?f
zLDsN|#`R+T&Y(O?{q|_y_v(A)0xJ*#J?4f$mWo>}eHY69ri1``Ds!9l4A+PFA0kg<
zIie-K6_1iT(XtlP<px2cl9I+Ij=*!*q1fe6u4h$lxsuZi$tv@Q2Rh&GmQ%)(dy_M0
z!aZ!0Hq*Xk6$bWR&9&6Q>dDL+ZIO$e^obCMSt*fk=G*V?`S#OI!#k-auQ%%syu>kI
zK}ip76Q{PIJp{#<4Z1?2pu(%bVEv8*2xu9PAz`{X(kHOXzZ~%0IAXJp57Z(ZT);Mh
z1$NWe)Bb}PPE$ckU)MaLrR`)sM9cqs^kk1u6rYH@E6?Y=!<CU9N6^c+;<}II7+6Cn
zNN|5G>?>{WX4KYKj7y~-iYw>JHZLVyfI^X76#Qa_sQn`GTI2=;p}MYaG1y9c?9)aV
zxZ#5ETeMoE&tJ>+HhCQrwZkvER?rzWB^8Z`q<npzB>|md0vBC;7alh?xHf&<*CAK3
z8J01|Wq1Evbs0o`nVq2`I3c~B6AsgZ3cd5K5QJu>x>IHh<32tVICrF@tjiAo(8ziQ
z!vViFfE6%|wjdTDEL9~mG6_0~VGiwqgF}a*&_?L0*M%}ekdi;i#drF}PA|uxcT{!z
z^@z^(qL2!|06xR%U`Ax((w+z|ivW*YYaPy-)sA~FCS)5kc@#H$pN4kaY^h)~`wD!_
zU)O^hECX0v)vz$Z(b(WEG9h5#tnWh0&rpRn$m^#q%<un*!C=D{4lknaV?$%QO8Zx6
z?Fm^K=RN*-A@LO?p-6?VlLQtgA988A>*mV`K#f_!NR!#W#?7X6#s(EK&%jkZAwBv$
zQ|gK2uJh+$PUsQUP9K|z5F;ej^^3QTLforaErD<ca1l4nEL5#SSNqORc+!HlAEXPb
z4yWGnWGW{yjyJJI3I%JDC=W$^)7VKUfs4Wqdc3w6Ov#s#Y6gme_+il&=G#sC(0G@T
zbC*XJLS^P&^`Qp`*6NS`y>!4g-~F+vWcT0&i@5nO6S8oTG|?s~NY1LBpHl-v@-2{X
zJ|cJe<$OJNo?clz-+$@oyAW$1H{%ywH9`L)m)l?@aRvTRt~EU1An>xmNcd$hau~bs
zXS>bSwJC5Myuw4;5`6=k#`^;)R&^kkZ8gw<2tijmAO2aqi3paFhO>l&L(F#IXvUBL
z)oCmzn>}cu{Ahc&Jil1PXv^D`g@|_fac>(uSxHY~jAU_o2Txf>p#6FkyK1JO(c=!T
z6aCO#515(kx8S>6&OXmj4Zm0D?0tl<1?#Zu8@P9ZGf9pb>lO~H8Haa?I|fgldXBdJ
zZ$>0G+_1-k#EcAjA{sk)6Nlg0K0~lqg9Z)SM@>a)*?)ohOV0R&jn=8<C5yA5qE+q#
zTze@L@%7`9*CB6FC{`hN$G#F-Yw+B3{MqCVV=*Ypwpb6hB6GKfic_IXA_vh)boVeA
z&kYM3k`uTMnG`rgwA>u5LTPP=AuuhYU9<v=bP`+S<4w_VhJTCC)tDQgLh9}`inY(A
zE;_fhizvdLmo3abYVmkNe<3nh#l5fS_=p~}9~y(=gvi%{3=yxZbmmsYNu5o{7}5iU
zizr+^=v1Xv0+*(q`eS9MWHCh-otKeF){*}uYZ}hul51hJ&TF@J&nt<^T>GK%W@<uq
z?G2Z`8;CJ_*qr)dKVXf1G;akYjrV$dMe|})|3i)cUlMX-C9uG10WO^vKkZvq_u*<j
z`d?ly_h^bs?M=oMBc4LhL;<TwaON9m0o+V_v^s%S;4_&n9s#Yo>vb6!{8CoowcB8p
z-OiRK*A^y1=fI38;5iG;CEf$kt@WF)O)l!oX9ng5j)JZNl$Cn^CFRDy%9`6<ai#3r
zD_xXj;LG-7no{3PTQ>3bLGQm`n6%Gg``&Jp9c%Nzx@})WJ$t!WEH>rOtbFlu`k#e>
z{dKdgrHU}0AR14^R}@bNA_GN#Id-j)T=)iP!T#X?qGkbZ?Z-`@J9p2Ny8&**R;|di
zV}i804E;|z22-am!;ia0FE~6rAGHQ7)c_iC0jhYacfA}I5pAl_-(`~pycna}X9G(~
zDl$JYoMs>&_}pjw+C=wUZvJ`gh)^cUHKtNf5~>`))cJxfS?&p3z<6r!*I*wLoT}XO
zck7#A<wsaGgY76u0Uv5elFbuG!hFXJUhQ-vB%;|q<1tOmne$M+_y{*RJUsj$ON_|a
z@v+aZc-rskcIQ40k=8-u&53c-%A-_!6+G+0m3oH_A527UN9%O1qv4xWQ7u2}H)ScA
z#HLu;EoxRZgJU9t7#P#-bx`4&n!mj?KE9k{^4~VnWBt94kZ&QI0{XcJ!8Iy8R+Ss{
zzq@&eftKuSY{A}1>XixV%hP$eYmNJNag8+Ei8Ll;R|Sl-fMpYRFEx9_Fzc@-wM~O1
zM!O3H?~wk{L1*dRxN`y1y~*bXE}A}&!MX&$t$GLSCg}@jjz2lln@rE?;DK2m8f>??
zxkmCAp5vO`e4&>i%nK-SMnYHGj8&t{OJ8)O!$HXi`TA#QEUqhdRb!o$MpU1q<`Ud@
z2I#cO7m8Twi{VA?j;f)<sf|Cq5^qpc35Rv(v=gaVFGj_nNni8<p5mUk!ij-f4nJ1i
z$55asY!9;eKEL$v`~lNL&${PtkdptKns+*XFP#E#A>-$3E|n7Juap80(y#2J_dxde
z2DBDhA9?%s?fKWsEu$%YZ@QkI{e9jjqm3E~+}9Q|xg3SLeTRXz=GoY}v!33OhdT}w
za;xW(84Y@VWPv#%(SNg)tB<0<&FW#LlAl%jXKYHsfIHmuC|np*N^OTsCGSr0>&&A?
z6iKyCd#Pt&)~7{YXUCX_+xE>Gxv(yxBhpBQgyI8Fhtr%oS`s=bRuUZ}6N>=}oO<@E
z6f>R{o0$_dvb83O9*}UOR&L7?mcwAe*imvN$nI;(6W!`P<7?W}7`wc-wqcgS{$4m|
z>cvS8jvDRYvSy}}(lypo3>i}8()h#xZ9=2)e>0>~uyFv1P5GK$A3q~f*iv?{_tK??
zuR^G;{k{7X=081W|6aIwX7T@XMLQ41UBN1v3J*P-2Gk_+Ba%x?5JMa6tTS<`%D#Hn
zDkg(SdiQ9Jnb+%_uf2hQ2#`UVK|Y+1*T?0(pM3|uO*{vV#SA3VQz8|a$^6gAVyBv>
zl^VT(^6bfMpj+JVn6hu+8&|e`r7bKtau$8aQm&)z8^b+;JVkw%q4&19lXx<bhNABW
zlF}M+>j*)w6uMs@t=20XF_+!s6@4FF;nL0UMv70Oo4&u$Mz2jo9_vUuR|dZ?V1SfI
zN0@I%5`ixYOt?tXJC*WfB?~O3&1Ar>!i4>T#7A%4C=2(zrCf5u2|YnQ{LE^p^XX10
zl6uT1DG6odBOoI`k=M5L_(9fz+REGBBmduU?|ApiH{eRv+C|vklFT2f9AbL-2Mt6+
z$g^h!;bRfN3ioet2G=ehcSY&cqscV7HT<e}S|NQP<%x)q+_1u_CWq`cU6g*Opdep{
z6_SMNcudJ>0cHN(tXbV{0u6TkENP=u%B3cbKy)@&aCF622QN^@bB&QN*T)F!d#QFq
zt^TyXeIwv`W8|&Q&FYGXCR5x7I#4&U^^1mcTHYh|bbB>`Oxmjjt^-iRIk>v|B6`@F
zW8VIF1nW==uX4qGjRZTf@Ov5M%fNhoaRNipZoscb)JJs=y#3F(516jt-_e7%)v(F0
zfTHPN4avp89T)Jx(O7ane#9kB&)!aXTd3~;XdyRDl-TP8Ds+~Ux?Y-yWaaV?XWr?U
zz}fbVPar}9L9_<QpNe6F38)`{(Y&&Ph{O~Q4J5G$5JU<<RE;HkQ0LMK=Cfkq&pqrD
zOj##13UonxinO=4|JZ6=ix1@S5m9urd<wYQ!wu8GdWy-i5K^>XntvT{d&~D)8wWk=
z^LEBFY=WAKZ}@#>4-o2LLY*{wQ$X1Jn|HJ4GoOPY1=Q4yE%~T&hBOfvAn&S5RP^9Y
zaoermMYt9a5TN{(67(i~FVM-%__yFR*GE439d?n$#0TVRVD2-}|4bC1%P#@aH+ZQi
zZYWl0pL@A&w`2H`P`Xv0)Uk8AYmgrE$!z+cCw3e!s=N=SAm+@W+AfI6y=xZ)G7|u`
zxoPDMXt;1wWfk0mAjdd={v6NFUCFI=h)rz>tZxXMcUwzi<)omo@J!t>S0UmeKAgfb
zoYh4<!@)B?0;+)4L}VJv!VIN*DOv`Bp~Vi5$aXt8k6(p9v8<CTNpK_(EuIgsC)Nu{
z56cfYXWgMF-_;_jhuf-tQ^d<M<|~$-h*EoLy9)qekySv^dmE>tfd8_fonjcX`{Q9e
zVPQ6}7!+Vz9>pNkyBN1|sV+yzr9fl2RN+Q7<mpmLE}RG|hf50F2I%~9t$PwyZ*m}o
zMEEsf{o(5oh~dtf;WYiJ^Rlh0sZ#rA&rr7y<R-zdY}u=@f}e{OPrAi!N|;VZ0%R~w
zG9NVS5ZNc8C{oy@q<6+isXd5n%BDZIYRoNr0k8Q~Te!(ZR(A}gC1i@JpD`Iyw~ib%
z@o!=5RNoib#{d!KDDpPD(ra%>WY#c8IXYD1f|+*)kDbO2Jk<hoMhLK~05AYW^Tl+x
zbxs6%KyZmm!ruuC2OzWS_${%q*qu6Z#3N|Z*;yYbc|5s2LtUOA>Fn!jmq5midYNyR
zp@N3c1ojP}*p~hJg-W*RLicJ%x#}&wc>}sx-->!4^=!w$l4|DA6PM=F(_OUs79G&O
zf~)BnLnj1S;oBXo<QhaxdeFFBJYnUDMc)4ZEM}r@xDLut<`I1?_|%H~a)^pF&uJY`
zlGPjK<))CU8A)0@Uod5<siu!95DR$ENinr2d+9^!sI47)j-B<P{`?teGrdr*3yOyS
zqfUgjCi3${MQb>A9HaLKYX}g6%ehU7Ff?RlDd?K=%;ZQ=kYwcQ_gRd(RGYo<Kmtlj
z66UUfI*b#{@YJL>xReWD_7q_4LllUn_ezw=u&>X4tzsADs&+99;Y;s42azf_ET6Ud
zM>#G!;oKz@YXHu^L}tu8kOxIgx|STqN9d>8T#v+E3TP*wkS3%<_TBzn6g5_nQ7evG
z#ik2h?<y3Ilp`n4Bu-%6*u#P<mH8yl>1i#Op0a*jd2&U9oCsRBhl8{g?9laNcwE3>
zt)Dd5qOTau+IaS`)GxU&2|^m7#b><>_0D&mYc+k33CN6E+mCw7#B`~j;7h@CmCo6F
zvclaYR;k_)HXXmwF#`C*bOd6ZLdd*qZ`nB6Ty~SeTtpKhOzZ7+b3PR(;h|>Tt;MmE
z0pt$3ta!wh(~sn?Y-yImMk@=MwIX{ZUR$KH`kjTdY`S<^A{X;ws4L6$F+l^-EKw7$
zB~Q?j86Pa~U@~Gf>nb9V#2SNE&e_)!@+oet<d|q*adfy#du{29f~NMd`a^x|i)Bm7
zWee<*d~pW5lofiEDqBu>&=JnMyvLc9MMw3Mdo5DgOi)1q&THVd!Zr6$d*i|R^pVy~
zQ>Tm)<mM7P=Q|&Qarb)&s&DR>Su*27a?mtAf2<S|9I^YmV&w3H5|NlL(&xPJz^Gw4
z!=sT7Iyyd<Zr=c9#Po4U-Z81=7N7`2ZSBFV`~K7ZOfX&-IC$2>eDefvS=axrEz+5C
z=PkOVvNCEd@yN7-G{@r?TbO>jqbtW+KCte_Y^k-pO-Ys4ixd&h^q7MxXY%Egvx>a1
zimt-5n}%rh^kl}?;Q4nQPX(h1>^_8{^5l%Mu)j;af4eS_eI{79#(pDciVd1KMWJ@y
zNVq>z3yy{)QkNGFS!LLU(U>iXQWYN*>IyoV+1soYQ{PEZCF6@bgP2t@ooMx*vY!eI
z%S2pv_U@VNxu{tyv=}m-cqgcfa-5u3EPA0_p8RmeWXe<voN&)`m3$jCXGbb<^^ek7
zePjPkVTaO2(>1*s=cI*H>_-%QaR=3ox9~(<^8M6>@^3x_%wr!HH_n>_j%mlPDi@xL
z`?G<7c}rb&DP$PY9&w_~V7J!u7UtAWORQ+VVO{!ZQhj)p8uV*?yO;*4jRo~mha1L2
z`%xGdD1Jp|PMYp*Uoje?{itTarPNio#NbMChj1IlsqOEL(W*HyDfGtXyxq43FLObz
z{*H|LF2;>~g_xyguZ~LY)Q(R8pYTINqZqel?H$;$^gGp=4#;3sSjwGon+nh1@|XnD
z!+;5D8by(J5jf3jopv~IAsZu@QUaH4@M?(<RS6uGz{~lB1)GD|MmNl6!Y`No!T!@B
z(NucqV)*pxAIJBk4GRuNnv);qtWJV;Rv{eIT|QaG2!<Bj#&|Bcah<-~AySyD-zG8>
zM$ntQekDV&6261VP~|^r1uR*YuVx@}E3ZEjME1d^ZrQ>ui5i8WS3e=scimIxKZj{~
z^27=l4}dlptm%Jfj$fCT<AIpL4p!35Yrs`ZH20V;4Jj!$eJKC(B3mR^`?s^&QDX0D
z=UT$iEYf7;W2?``AV%phb$B7g3kLv|PA}CxY(4wHH*DFE*z3FuRj19s;BC}7>ih9k
z)!{eX%w&cbu&Ua7i?nn*zRPKVTgP+|KI|$Ukc+KkQ&Fk>pKDIHRYfI{-Z%YRi9jZt
znOf;<oVX4eI{0rbNTEB5VFGbJQ5AGHO&c?TJ|34s(-|`-CZb1ggQ>H}wD=q*Z>K{@
zU5&&QbRB?5N5-JoV`FqEZ3voSMpxj%QNTPzYxn~E%>E6w1TyD%*ap`agjr*N&S3SQ
zZ^fuj^xMwgzuH)z7C0^^pzp{?W_6`X<@G7?OXUdoJS(od;U$!B9Z1;ii*tYBZj||A
zk4Wc;B@K*SyC8=tpx5`+1w1RSoe4DQS)h!&p5N}Mi+h9}t9Z9`-8l?|eBq)e=>>A-
zEp0>?D*IU6DAu75HJdpoKM|OuH$?V3?T8unQNC+YxT)YnEo>4h$lN72)C?L^9h8jb
zOeW?tDIQIEfe*=}x7Op-IvK-|KOLwsZ^Mps!k*XzJG+TXBQ$Vzc1(Br)wMW2o%K_j
zdZM`_nUc8zw9fyG+AiCTN+s;<15(Y=6GB&82p{^xN1YTycKLC*ZJow008_P_#e!8O
zS~qV2g~sn4Bu`XK1?PwETJU-v3EYo4gVhDKcKt&q^=0T8R^)BGN`@k5-v=cK+{C!^
z43qtb0}kGw=C~A^EpgobKs5GTch<^}88B+vk)J8J+rvQd7~rCxAN~CK^WCg^(qLj|
z;he?Rwa60D%+H`W_orfC?M)HUB!++pB8TMz7koaHEqp4JRJiC_kTZno3sM*xn_*6{
z+P={~YR^L6zSoE3Cq~ZqdDJG;8do}LneMuhnJX<F&fuA+lPEkIb3wq%&~LGNK6WyR
zQm!$U2ENSIBA6$^N%r@6eqdwI>Fo*6ySjh)LbXkXJo~C!Jdsd10cdRyj<+Ea+VJ$|
zk1m@rLfG--BAoL136Clf<{ZrWK>I2%Bqt!GuZ$QEMyggG+H<qtSfufNE&I<P!uls{
zod7}X-DbkKwixS#VDB66sZEF<<g~(O3n@LU7@Cch;jaqNt!b}2Gn)u@XF%X1mKPkq
zqOGeC_kqDCFvJ&2;09Zak3H4^w$sZIBy_bsry0Y-wh|(ZR8OF<mc95sFrEbv6}sk_
zR92C;&-cU;Vjw#fYO9x3j~}c8kE)kCx!eJ<<zC|{*Jm-v9U41wD1!&V06ka#%%HEI
z|7St~h!e-nwWEtQ5Qf%44A~w;9X|C#A%%gE>uqh3Gt}~wxweLGMb3f#L;%F1&&@&o
zbgsY)w_=|gjocY_l@m}#=?A#T#%U<Qu-}0P%&h54<TXrCmqa`Z4(rP)uM6r9>Z9Tj
zavsrnlubiN$cOojZO%7ExprhT%uXw9k%@a^T`@ctkc%lLb=62VKF{+Pm16b_zV5LZ
zTC?XTMA(W^(cyF8i5(*HLqW${_ZDy>@TJi-b><FEw0>JZQzu?PI5)PK5u^C(i|=oT
zcjgToJ+;zb_TmFNlE{dFUz+65@X&R3xs3FN8{Qt$CHVPK#_1W9&*(AFzOl&=I(C$~
z<=<J+!)@XAL@cor$<m6{Sdm%aiqH1U9~O^-ve@4^S2B9`m9&_@#d-Z%=1~lSgtF%$
z1x3aywHt;oTo;rAlIs<JuE}>j?21M3mn9nluaQlm^<l<fvwYzauxst&AU4Nn(XVX3
z?e4BMCDY>IuWS14p-p%62b7l(HVUR<Ro5ZiV>Uv?t+9C1Xn-{)f*o;OavBibY@XDb
zSH?~grFQu_P;M1u1`9azgar$!7^+^B<g>|>vO1%>Gvqz>i{}O9#)*2Y?rF@@(XmX=
z2JrhI1792IQ~R_kurnr@X}>GmW4mb&+_jQ9b2?B64&3}}CL8AcMYS_m!q-LVX_R`6
z1RPg0$`*@Z!T*Z_tdDL$4}qndjwkwTIolUKB4{qg@m{c)oG)3M#==H}M2sbay;p}<
zS{|Eas0^ET|6l-icCZj+ap?+#3Pwr%9vPMD$@~xY*=AQFz7_DVmwzH1$v7M;B@jzH
zRDXuN%}35dZvfj>c)VbLP3??+QNsS4ZJzN;orf&-kx~<Zo_19SGtR+(KTzQIM=pOK
z<=SB1W{reYeV4a^8E5Lr8j<t>c1w#qq41}udO+(Zgvi-PjNd?I3$zr0ZDp<amn|9Y
zkFa;?pgOQng9+Ln1NFP^PbWIJtHcow1^d=$duLdvewHT3LnVcFNHwk~gcc71x||D_
z@sx#CExhbQLY-NH4>jWA#&)K>n0of_LV{&&=1x(EN=KdGc&p&oL-KZAlP)5h>hN}<
zIjiZ2+!usIDaT9eKVCh(?G2Z~Q`X9q9v=O+s(+7(`u<8k<M>oRs7KC0_LFOk!65MT
zj=Y+Y8?g?xU64S%dyVB^M#UU+Wv+B~JVYrtJ9FIPTWr0^7UhNYE`+5@FV?=^4AuZP
z@#}B=e@taBuJq3Ld8v5m?kcob&^PM^BwFFBKEIc!<U600lhP)!GnAB?3<{?&syE}s
z!vWaVm@0yF;;URAo7!QFw=8QhIM1X)_Jl@a=5U{OKGxA`eY3ZZ4ex_8o-Xdb@)93*
zgr5C=DrpR2;%?lUplg8^qZ@3tF%2>ML*i;C+zf>~M;!?6MvJzvgfAr&7#YaA!sbvr
z=lZ*>?17l#6SuFB<lBI4eZ6{<LMu&qviidpH`}b{;7*Jh8)yk1v{_9eg06iEVhgL@
z4YUaO%?C3BD4uNaJUMf=Kk~?3>>|3p=0<;$wSdsUUQ4-(0k!kzvgP>n9>juP1$%2|
z!1VZt16CN(2b|YWN?DhxGGSeBy8VD_Rj9Ued!UWE0OpkpfiZ#=n<E_Oli`<EXaF<v
z^HlH>LFwYpdhg~cFefX`<MZL+B>E<uPZ_kP+_7E?2cS}n4)B=ov;j5*PV<snMmTRe
zSsI@juokjFXMFd!m@S8P+{SRKzvov9Bw65<@Cm+DP>U>2@DZu71FR5j{>)L6mA0sm
z498u_2ag56$stjfG!FbFC`pX=*O>#c-IKGcbkPlC{S6OU_69kd8~Z>eH`9y2D9|m2
zQBQdJmU~JnI(@=Qy!R4d3dQeojM!J*dT1UJxN)i<-(6-D_^dY}8c|X18KQxm<puS^
z-QCf9o*eE?;SnQ526}PIFU(pI#B%egz1oSqi#enfxwsPyO+lR^IcmFh)Z{us&#lcr
z$jm9Q2BkQJ?Jb^M&giht>ZY8=Z+#G7e<tDnlr|9(v7KeG9KuH*6Wc|Gb7Q7~6T^Lx
zi@B^XkC)n8#Cn;APTP7M`=?A6nolQk?6(<!crbc0ld@MsKHrc*q-QY9H2P90?XzS5
zOd+rvGW!I?L*TVjOn3Ce{5*t?(D_o$*4LKbKfU$!6kjyXCMB6DlIVTVMvc;L2Uo*C
zLH+sJZnsuyN|eZ!o@ZMxw_#p5y$kE4;YO-X$b)$Cr5@o!4uMw#s#;lzD<8DcE6XoW
zTy>5)#xhezMW3+)f1XT=vXLXK%pYHBe`8xWF3H-jLM+Q2yIEYH%UM`UCmq}-L(Fu-
z1$8=o5o|_gL?Yp@WS(y~L$_!N)f*-_y5fN9D+qM+7~Us3XpGjDcSj!QyBmPEL+3rB
zX<#CwW~pd(^I0nN!Tsg9^?n*P{oGhy79PDAt`0aUyTms)LD4Zm0ng)aFM;3cKim<b
zuQeHdD==R(b7c+NWA|V1{6mpjesH?;APTIq_!{Ymte6&Yq_IJD0@svCo-bP(_3r1j
z)*^R&Z0f<%5cL)JsA}o2a@SXND$@14fX%x*;9G&z{co2BsG!AdgV_doqjVw{vHauP
zU|EUoP9MvfJoG@9dO8WasiF3mp0~vWHTlj!MjqRz+7i`XW}C5eU<i@;%EBwY#LCW<
z>EBv@C-{*ZPcS`&nyM{7G)g4cd=#}(Dc@!W)n0x;D`tfDdc=ayrT0l~|6{&1<%>_f
z(a65VYVk9S9$Kh;l#kdP6sYgo#jP>&CW2Xs&Y5h~-#D=MbK^+ONbkt!4*u&hcmmhb
zL&ux7ruyF!Vjkf_!j^jQ1Bp1$iS1dynwH4UwF<<G7TnikHiYQu;!VPd%I`w^SUkbC
z$}9Jxs!y)m;{e_K{<upY9X3FhuU;8y<3SrZJLIsUJ;4MVH!fH#E%2&VwXsA-;E2G`
zqeQ{CjdSZDsDDjv2=sY_-|*1uUDZXKg6<#bJiB-aS<hS^)IaxBDh#!_K+Gr_+v=tE
zsP|EUh1XbP9!mOh4n?}^f{Uzx<w25L=FFJ9VWE2h$N9@GCm9Tti=j@9VVv7GNjH!m
z^QK7!4_?l^pTfGPdP0>*EJu^}sMyu-2|f*ayoa0VVPQw0o%6|G`w0F>8D;Xm*#60D
z{$jFtBEhG?QJ#TjD*P8qBh<M!+w-38=;3plnF)lBaUtEy2G|4wDi`y~$_h_ICH=hT
zrOq^m^K+f8d=DILsQ~_Jq4b<fTi*=N3Z_g|aSc1?+<`Lh$Mo)KP1IZ~UH|x{!me=C
z7j`_{9yBhkU1>%&g%Rn=tS>Emho>cPJ{M71Hry%3cs?C7OxT?T{65%%?Cn{Hc)A(y
z>uGa>v)#Ny<p(C}JY&j4dNK^hZ=fS`_^AQfcY?~`(!roV=IiNv(vT$UFUD-*)ViVq
zi9hX>&>5CJ2|2NHG?6BFRv7ry0Q&5?z_rI~!e1{?rrOd!%MY;n)B%KAr4uB+@86};
z;lGiG5=sQ++XN9+gCW7YSJb$^ma_7Ed6|}QOI`1q(o04~v;%af-Evv_FFo=ueScrS
zS(>`)=uM<yRbF(>0{4ay50VSi0s9<OVQCHNockVxQ-It9l^UIFcv)D|qdRGjDu+!D
zgbq|IiuBNBo*_j#Y5}fq#%zARmNBV)sm58WewsQEE*||RAo9u9jKC%1=mpO@FH0$w
zI-nu(*IKnY1?2}XEZ_9dea`JxN;%Vk$Bu`?PngV({zb>K%i+Jjy9~|A$l03a@)J=e
zOF911Cq#x6Fw`o6i+#~Q>m8WT!CEd$jVPx~Y0O@;ugUY*K0RkhL|28kRWiu`&I_#E
z*cW0MBmVl^<%^iJxbK~Pfe@g-+4L?x_@j~<*TA@{d}N1wT(D4Df_HjDD(0IbTskx6
z6##?2ldnI^|1f(_q$7`Iaka`7egxABG4Jly)y;=!=T}8}3)Lw7CKxJ&W&_>k7XZEo
zx`$o4WFc&cT$IQL64v_*?$a;S)Q=ob9-*t0OTffjAcN}b=pjsw^Sjgc)Ae<E=d$8m
z5~k$xm$nNl5IqP7tOblD_0Vcc(T!gEw&>HzHBVD*Dwe$P6&3=<4VAiCtQ*2QM0^L{
ze^?$r@1ojvxfMGCW~^d$rLq*OtBIktvqm4LVhC)hxRl=Zj)gA9AJI@|>NGgncp)JU
zRCMg^BWB2T^b`hLqpuAZx&<!&kNsHhaqg*;an(q_EI6I#)dtSJD_)LsZ+iCVOV431
zm)H7m`Un5Xc?ZwS24lj&@B%L5>N}5m?hijGq~rKOg)d7XYTemXQX+~7?91Wm93^Ry
zq?pcgD;&@|g14DE1NC*jdl$ltbS9X<5S**L7Wqi(K~9XIVgm1ExR%@gaJ2+82RV*Z
zvW9~$gjYG1`bM=XIiYNZFl2YTq>tF0-(G1OonP#fkaO~z-sT<L!jP6LD8i}obk~3l
zIPV^XInbCBlUfcaF8-g|0_pmrhm!B~#pIxLRT9G}4ur$xSK94cJG_bo6$=(u)Too*
z%^uGWq~%hHLBn3$9LzUZLpPW5WFu9R>h&3!BEI)TkC}}!lNk<BY4TT_z(SjU!zLG>
zH5?c4ok!yg#vS2p-nHdt7ndPBH_O){pDR+Znfi?i*rPJ<BL8b=*tcQ1v(d~D1wy6g
z%~(>_9owM+56<UQsnihiOrVUY%0b4c$^Y1}RkA-j?VMjjpT+E-UjR)IeX4L`Hdov~
z>5qCH_I0jS%F+_Zt}GiKdobJcsjR`(VlWCqDM5pFr@?`d^>4%<(q7vyN1VXFdUe_5
z30b)+o&R}&Xt@lNBdFr}sbcP{bGxD|ttL9ej_b&#@|$p@jTlp;Fq@|WkqNm+-F4Gy
zDx)?PcT(h8QZ?J;tB^iZXze)NO-H$ReK<1VnN?=3p{k-*=G(6Sa<=JesM!n{j8vZ2
zUdoXw($};jZO(sHLEK$Fk|=ddDn~dgcc@t$R*<6K#oT_`7=-ae$hA!=x=8MY5~&m}
zv*mLB!(oLCl<p(d7|X>@SgGJAYm45NRYzoORhAV0_9&w$zA1&o8jCrT_m|7TpIc}S
zvuNP_H6&btHsdJpr7L&%THwVO!qa2!i*$Szjy?Qkwv6(AwVWVLBy&@<UBIkW{2KC^
zJR^H^_jvqTZoqmUdKdJ+9uM-@W<x{{Nzg#>tK@afs(ePbid_PZ8qo=f)pzy#p_z%l
z+)^1_c*EKS<>YM_7l0q(@?yD;g@4{1af#d)*fQF)r2ycr3t|=Smwc?xLr_P2OXP+Z
zA5=_&=xpV(e48_lB}rk__QJ9&&+)5!`CQ1&|I}I+>>6c<&>w9-ehPh3Da%7ib5l(^
z)rv)W=0<gj^jEt!z%o-Ilr^x>+OvR-eHK9s>3v-5CA?72ZWl}fEdEz_-C$qJfZJL9
zrqU^DrhYO(vf+ZF3Q)CvTJX+GpHo1TdzvUY1rHE|1H0B;6WWRQ7~GVh!*gQoMnZWa
zkU$QgbP9AYfSmnffV~c1dVtPwgab5J8-sE9pKwNMe-3?Zn!X&qV9^h*>b85DZ714e
zLy-)tE8-@S;Xo>!ao5YkN$rO;<lFb~ZV2!xq8@I4My4-;yC`}fz4fhNtS4^63-d2=
zmm_gEH)$B><{Q7vEBEKzIEhs>SH?uzSc<%Rb(`W-pHZZ%rm&=*)tcM;ZvEZaA@)|D
zZf1Wyjj?7w*F5IxM91>ak6Iv~O)gvXqn<Od4;)>1AVmytY|DdWYmOtxtRhc!Wxc`1
zMgKB*pAb+O-5wtSO~A_Ks5a2UMO!c@0Dilxe+9>7w0sW<BTw~PS&Zr?F}`r&t~Szt
zD<A4c4qiF!1BHb*>ZvJBUcg+Cav>E1c%K<X1R<KVgAUTi<+jPm6NZvPY3Kh~`Wr`f
zQv7y$f;q$Oje0$RaUYP9wmJo(zN86$bt^}twNLAxtjVz2ANx^9f!U5wc@r+ZeD)Cb
z(B5&T$7tf<PWx>ikFi)rO=5eu5G<DB9rqN5>w3v9hmX;Qxh&DUT<8vNRFyqrLiPKQ
z#vlhOoob>3cEp$^#&uztP}nbM-bY6cu+s+1^2Hsj6dGE{B}4hoe?y0B^l?ArA!%y5
z)Yw#hLW66EKd{HHv7o-;%D=ON#s9Nn(zE+J**1lL7sebf%Do>KlnECD2A!Ea-HH3n
zm9T4ClxxaIJnk_KGF~qH2J&jI?rAiX8<4O87YjW0wKUXLvEB~nn1eORRwscQiWPp$
z(J}F+Sy#TGr&HS|<ty<B2}`T6PWR48k%9&U+wx{??zx+`tS7!wR)gX?l?QeX_Sn3b
zx=CJKJ;Tz{AmsCO)AIw{+r!0k#0qmMSsGP?LF)Uy_We$?31LIKe0m8pz25B!b-1v1
zzz$<icR9t^=P}7LiI}btKkfw7S^yBQAu^yP0y1?d)+eagjjE?GHC&Uh^Wof|fWAbN
zH{!MOo30BM<j2a=5A8r7gayxV9qFxW*Hl*SeGgF?`%pnyEIucCc`r&a2=2`1^`Gk5
zM|>d+Y6X1cZ<>7$%+kb+4?8O>u=vfZ?>zj&r$dP9CXx5W-=tV$;d#J7_w2`ZN430{
zmJC0$AA5R!uMdFT$0A5^2z53Y2%5h`$>{)oZHOBJ(tP(1DujQdKH{&kYAp<`KLWu?
zi+=siPQD2N^#c+3yZ}y5$DDoDbR@xZ;#9C0r!SRI89GYPTM8Nr{1gt|i@W-l#JLYT
znZ$fU4PkLhtY<fJpGmjk-2LT!nUhMtc<ed>Gj{L`71+2-ZsT7u3*~i$YXF-_|1e*(
zAX`jTmg&uI5B}U^n)ZfOpBVx#<_ei09QfbP4qRz&c{yZ#JS4t1DMr#JIzGN@{I?gi
z4FRDC=6`<w=|_&u8hROdDX;9#D<?63j{8&zk`Yoc{skK~xpw}bit2c++d@S5cyoOs
zv&R;GeCm{`U2cAUexE-p;8TQ!t5=QqBm5;Za;7NNle})5{-TRXZ>KI<2~<<p%Xex8
zC*xV*g8wvtM>Gf`b-^s)R<j1XBdHs+3rz<Mk(nFZx6&*cB@Ts=-%kNpv)8;;AqsF-
zj$B`5OEO*myD`dv$sG8VoqkGx=G)Q`Hqi@!8C^_KAbhBZ7n1GOL=r!|>NORFtkjN3
zX6oI~<4TrK{ua*sdbr&+DvHAbK)Dj^{w=OFb})z#n0}?ey~EV30&}4y%8_Y1eA#Is
z+_@umTwyxxiQJOzl$*^e>i*Y|+wp|oGv7H6j(RZXifzX;96iX3td`i*%31380n19?
zz<OD*B=$f)fuKF5W<)pkL0DheSTf6Pw%z9L=`ksN;D8(0QeBGItx2a}sNc^;3r4c)
zW*Qntxp#)OoC)Y~xclUxcuXH<lGA@#BPxSwm^l{Io@){iCCi;E=*x4NjWUTD*UC0i
zQEw%DVc^3&VWIpfE|?|refhenzrVINXyJhg^U$&MbpC$5kPR_e42E6)MZ7EW5jU}j
z7aD^E^NZALWl)owr=bf+uj}m6uc-rNF9#r&+!NEKqMv|uKCGA(uPsynKJ-tm@?KwC
zV9ZRqQ9n5k&41^E{8C(e*Wg=}$mZ<ydVVP)t~I}>O(3a<^3;u|Ok;MC9)ttsm^02{
zDj2o~lY;2TXVuJW>O9kQl>8Q0=gdl?pC^+Ci@O+Z{e9)!M?|~@&CNnz-xBiyCFLr4
z#>$#Twlp=EYwTGZ)2)G-(lP#bE}RB8N&z|<Jl&Y%j_Go0nG{(pI549ShBc#m3+>m#
zwY^<RyB@FQe$`7mRLgC~{%?>C8YFI%OPrFGV1;yk18iaI?@L0dF@>>iu4i@<JW*lC
zPCV4ynssAWwniB#yidqCiucRS)Z)$I4d<5BMLBy!1cx)C$3Z3RRA0-4r?+q!a$s<z
zWryd`q7$^QYdHZnA6eOyTfiG$u46|5$cx~+`d<QHRx88$nnCZ{qJnp~201FAhc6g<
zuWz+601;YI@+*i&K;Nu-V!S*K6kDkS8w4Rn$1LN&8*^wQjq`0)N@`esH6e?A<D~Jp
zOi6F$l~8I*1?9d;rjodBcK92Ws5ZBI!H;L0&``@o2;4sg5^X5E=$O_3qp8`md=ZSh
zZr4O-r7&Wf^t5}vyS?sz*3WAcyVvCiCXzggeD7`3Xvm{5B~5IPO^rF%wRGNTpV!yW
zz<R82<vJcTSp#V@I3w>j>vYXGSO{hJLJOeVOMHK&@CRfXZQFC|h+xy$nb)`M{Xq;V
z^*aYWFyY{e7W1@=`pXRqJSolCVcbu(svF&$o7p$NG96yG(&9o`S<H2W%-+k1+&SX@
zOzj<R*f{QzlZt5*=K6A9_j&^IW*V||uu1S@q)lYtE$fWEk;6Th1Yy8-I68NCN8A#y
zSZWOjL_2cvUKrwcVF{wdhkAm9Kx+@Hi~W~ej1Vx>RSDZJwMnYyJaR-{^VDXYBDWXT
z2r2rA=ay6Ppyf?;P`nyYJA2rnq8c96Jh-3w;;z?K(tMJBjhE0&D@NNG`#fMICQ$^I
zOL^**&Q56_C5k%s;S4rOGo%M+r$U-(#DNq#RQ;PDC|ddwxe#KTv0}h|s3}?Mxqy64
zDq{g;p1o-dUUXD9x0rt`zi=(+zv{6bWn;<D&a$6ArQ5nT8JtSQJ>f|F1#VE-rL+3x
zsiqaciRwcSh0$!N9A!ISMK8*ZDFqFuz8=Pm4X-^$sIIR~atd18SG*7=48F1WU01gs
z3|3Sxdf3|dNBIe%aSMx)ozLFUj7uNhw+Z=@5lXk)B2BJSkm1DcJ(z|1%k<uNGA(6F
zNYS?2wxQp=xmg`9k8lY)Nl5k&+Ml=UtAnV{rDMu_|Jt8IUE85su-$Oi)t)ocEo;aE
zpOFH7r$uMjx$k$*DP>v8GW-k#E*pDz>>3L<V_prncGEXBN{u-Z>+Tj?udPnA><%dQ
zh9JH`alYpV3?kZ81D4`Y{Z-Z#h>N(gv4AD2;a0s0t>*y{e!$%`h{w5eUPu={b6TV^
z3YaXwGLN^n7jL>bwSu_#DEQGKN8*xd`7vrK#3O#$(sAPu2xP}1xdo4CR_97{ZbRSv
zYIv`%I)-#vV=RN=$qiFMM|RMY{Zs#cw@~j!)E?4;DU-txkydt2S|dT85ni573lppa
zx~ZWGpBWyH(d7yNud}4*fSHUCZ5;tCmx?ntP#7<$WP*bBljg*tvqkXZnO_Um`=eG(
zpvXmCou-qqdvYa<Ke5obhQBE!+a<wP^#c4=ny}N5dE13XyU|3$?N2{M*?2{A2+7Pp
zXEH0@wW`v_UwJY=Mnp_7(|Nkf)#OvgqJ;gcSj=So_M+|}gOL~GHDC$qh%C2qn+{Yz
z4{tgkQxI1)7EendT^%7_<u1YQ48}my*u*p#OT?mz&DIuRtI#a{#dJskm}$-bIN*gF
zB>CuPWK4ZWkWQt6RsAu&CO1uzjm&Q+#+(#Sg>-Y;$?PuBesOfPoVfAi0tuPMJKs=h
zK^q@0xyM~a-lHumgG5W&12=Ux1-w7I&9nNl&9{duz~Voc5cwDN;4zsxal5kc{Haqf
z%cSL&8>D$^R$5WM6I@+CzV@<_NEBhN^=ItZpn{K#es_x@Jc<dF0=I*hW!<a}$Z9E#
zR8?Y#R1nddUF9B|PWfEOwU{wD#q3U^S^1BUH|`pp7on2e8Gybxz2CTBK?>g8j;U`B
zT1v0TjsJgaeFsnz*!Oh+!2*i9iedpQC>^Cr4cO2Cf*@UrO0SAi0wkgQTv=TOC4ka`
zXh6DBLMLFO6N(V2!KFmN5FtPaA<6fWxbFOCzBi7$?x?eQdGEe^&pqedZo8!7k0XGq
z=S1&?Slo@;SWn^(F`9pzS8stQL5rB>5b&n%w2ClJ6kBh7uQ6(2E$B4kaaTR34KsmZ
zm7`V(bB2FD>&(K$s`$j}i>FKVoY$MWTO)5AW)wNiIuSRTeSTeZrIl$Sk>R`-E8MLW
zP=8VshT-O^W}=p*4|D%+ISkeS%US=zY7%T#b*t43u6}`wdj@hkecJ}SB(wdI^T?%$
z(}p|>H5iOrLe3!7)z;QNeBNE2f(vvTn9{B0iP5_WPSRZd4u5r~^Wv!+yVfRuxI2$q
z<X6R_;OOqkXPEjAS8e{e7p3{v?G10Mxn;8{V|8a&R5_Ry3nb8~7H~`;{bcH{U6GNc
zhNVn(fubi-wJ9!Z$An~I;_LRf+)WilX((1HX9Zuv+z;zZaSI!kC79a(**D^2K3yXS
zonl-(SdAwtjwl!QA=|P)D4|F1nTw}%-D$8%H^@4S(F4Q0oo7%`0;vyAw~A92mj7e%
z1GA|>-vnM&FFkd)7cs$FVw09bckn(6<yl>Oq4NxIrMVr>Bs`{STX&xXuRPV{d{Jp>
zsfih}YpCHrEtNhWesH^^W>jOBT<{~xIzs;2^t#IJh&r^@KT2QTC<yCI**ZN5asll*
zmyhAQpGT>0wQOXB^D@(oq1Wy|yUR=;&;sL{C;GVrj9V*A)`wascv#zFr5&E*uMHg_
z*49hMw4r50cW&MG@^YD5%D7xdBEgSt?AE(fGl<KUXJrgrYlmk{?w70|(b5^e)=r(w
z=Xyhf)52_B_2@^bmd89TJc@$yqEH8QGg|6PIyt1DKow_LCaO_V=rK`GQVwkCj_MRq
zsi+(~qZ;sn%1bD-&2*k%XrnrWROmr+P;Q1@E4;>+zI1QAahc)u?{^Orj{M_p+V4;b
zBN_r@?7)24`>C()*76YStCn?QW?7q#B)E>3Zd$8$4bW+1I9CcY>SU&$?taq&T27#s
zgVCQxXlc8G!4|8lE(jNpu-hLF*!sEM?LhM`dQ$TZP=|&w$++3jW5A3c;U5I-uWJP+
ztJ^lk^_-V%66mm;L%c-`q5=XoH{Swf<e@JJu}5wd;gx8uMlhYOZWl%TeaG3d0MtUt
zDpPTqe){148sBp25a+rrK+)ZT@?jK{vFQ?->3T!&D*wEL3623n-L2%81GZGH12*Mn
z10wl{!!kF6+yQU+cKgiV&!5nN-U>Gwq&ZZ%;hY`ZO>#>8kMZ2er%x)8>;_T>y7LAk
zYz5j$L`gHj_ad?HZHGaW@9l?GcqJ~aS)6c6#H=AliEgLXG{8K^4SN8{fujzB!4svP
z-W0bYew=@~X-K#Iy|RN*WAWhNQFM;Hg-B+l=9bqLGRlprnDhWVs!z!wyjTOfXW6=X
zhd#+WwGa3_q7uvF9rgp`{*j}Ysv&~kJgut2fzapKyTsg^U7rD&_9vw3K6y%r>&p_Q
zClmfUw2vU;VO&5-A)-bwdG)S_oGV}*Z5OX~MT=cg3AxUd?}M3#9Smv)l^++LFslN_
zE;&uD*JhN0uYO}_UGMkFFyfV3mKat6`Zk%)4GqYA?0s_K-2!*jU6Y*a0b7y2ZcF0h
zVJ8wn9hUR=Mb{v%2nEHMwI28Ko1ZKnDTOGVlMqvFIAy}khULU)M<jPXi)j<w>z(6Y
zNd-ycAVTR;qHxJ32edpUwPV|x*!Wq7tz)FJ2@7_R)o*xnFTAn$BGy3L%WJz=+|OYN
zv2<3W(tdDH3~k!|Z|BF9F<1bl`$oaoeJT?|n@R)i@lRX_)4~v{=2l-}6bABH)FkV?
z8MzIZt~oIm^EW-_rID7ghe(7w9AEAAd+gGRmFI;=$<o1Wl{oZWDeCg_a>c?e-VNey
zMT#kZAMk;dA(b>An}Rah0H|JF#x_?*Wwz0GDy}nLti4gVYB28bW6)Kg{gKvwu^lsu
zhAbqfoEHmP^KQ(>?2?SPin!g9QydX}w(8*G3Nb|b6SJrov3zw53>wo{RX<e|K=v6g
zExT<qVn`}D16I5oot{d5+4Q;$?Rbj2C8BeueLpI+BX8srw8K-aV%?x$R616joP>Us
zWhPO6;N-DmnRHH{V2icLr<oAYF*&^yr4kYf#fW0{rDtSOL2$&BpNn?Ixz$^DH}GGW
zI<%)mgou36@ZCj~q475I3Et;L9~VlxdpM$7$h|3eWH71XLbnB(9WQoG1!183`)vHY
z!Gr6fXIdfC<6Nh`kL4?SA4QDVrQOAPBrckCJoepCF@nF@GZ)1_7l)+r>?wI_dF9ID
z5XBCX0@7^_8^5$VcenO}DkK<g=V@}GD3l{7%C6A-dQ9}{G)n;`(dm$)blt``IzS;{
z*=L|MX=4K`9O(fe`dendkLuBfqa5z_+_XlALZ|O6F&tnwdYP*tG*a?lR$Xa^vhPTU
zqhk<$nF<|AmlSjIZKB?qUL)nS2$-tYH;fL=&b-)uE%R_`j{ndOF8yKzTAsF4E5!wU
z2O?F*Wn02bgXp}^`*t`u)3fAQ4gVqKbd``^O9o@kcRPV61-vO_M%RDzb)!grcAhq$
zUhwf}xkaH>cV=LnY5x<Z&DQos8>gep@c4}euf)YZ4$~w_`J<|x>~8m~t8}_6VG;p&
zjb0%rfaUh4$fSvVms5`RbSO)101k&5BmQo|PNFP#H|u<QVQ9jYQP){|KW}8j(zch$
zyThyHuumM_^#W|=>CJs3Z9jKjB>TMTTi+`lfbp1cmZ%L#=K;g36+)(dD6g3DJ`bo?
zX_AfT!%pPKob`7O(RE*_nj_3%R`P~{XJYVd^Ve&hEzat4DFZ2l3o*0FQCD!-F}}hU
z0J`|=_&}+*Vmbs8&t?SPXhq!hUwOtq8N0;_b`NI~xiClXkgBZJ)Sb`LHob^RyCmgI
z6Krw?+4zKU+vK01S+AXb=g_nclyj50I)95oX;=p%qPgLt7W#(~)8iWzgf1JAyN9Ht
zq(HbZN8cYe<rwq_7VZYLojq9k9IdggK+Upj8FHG?RW^kvtGN0|=}oHfn9<#77Pgpj
z%GSw;ymE^v>hN7lbOd(+7YAPZdqai_Sl6$EGDANwrH**-UV(P(zCkVII)V||Ise+R
z*ecLSD|v|Q>oW-^J^Q<z&8IH{rKroVB6Qe4lw&R;U*(Y*=`_9mID_g6R)RO7HuU!e
zu@~lVL$JS{lNDanlARlXCE$bf0fXhmCb@*tV9}GseZ{a6trI)z>k4GuFuteuN9iWe
zXX#8un;fZZm!aDBvkYvzpz<|OA#Y*nC=%$B?Q{Kt*h4K(qx9BGCrfyTx(-F|Q?kY>
z$=R$UGzl#>;!P^%`7z3HQr-OQU}CR~z0VNp)cTe5<lP%-gim>^+sfYtjI2vDBK<o+
zP!F|qV$RAlO$xOdtKUGT>ru%OoHnMJJmIlN$DRMxyQsN^06|gV!mY{0CBvQ+vZ6&2
zC8@sA)Tgx5>B)!6hQs_AxeJ;>E)FfVa5c^31k_!pr_v7a$*c@0ZT$Cs^yu^V<7UWU
zVIP?Q#`V>WJ=LB}C{WIrB%IiF+e^~n{K0!yxMTk23?IhdA*D<__`hc&hg(pLVYX(o
zmVE3Ye%_LB+tq2=Cb%H^yk4O=ZS|8_P+((}1O2p;g~W_J3wbZK-ooSd(1R^Ey>ViR
zZD*~qZA=f5xGu*58MGA_!5l=0CG@7m!Y>~2DObv{pQBOoZCCD=KCDNj21{e;YZpbF
zaK)La^y0o20~oUb%|@Uu;hfav$`F(0R2<l_A7Lr$1~Z61=^bQS7wwZ^?dp&AXn#g#
z)q^bUa&+*8>@&R2U#LXh9L#DP7#Ohmc+a^W-t5^P0;wsL6H2BLic^1Ioal9a>}aoY
z4KSv-M=xcM_1=;6*Fz3)Z`tD0zE8zHi4#-QEi;XIpf^DTFf9^~DwWfL%)DHFPnWE|
zL*n8Y6KP8RWeUxOaH$H}#5)5O2B+XS%RL?)O#l4m%KYVt0-I!dTi-?bV051H{lz^G
zUp+1(y`6hCj~#P`Kktim*|##<*;I$qPwnTTI_GN$vwebFOzX0w{Is@&+N@g)QJ)h-
zY}Tutv#7&c#gkY~z5@^M#9MVSkw!yH>s|A;UzqE+rfJ#5G=6jA=T4oVL|SwzSRiz)
z{7Kp2d6e96mvd1>UK%8OqijC#!`&&jMSd<y=`F@(fw~fx7TVaR{Dz_-Wh3htR8pXv
z2F}XWJ0^v@_Y&1cwkr;y0z0V8!{BK-e)Oisdc-7nV5&ewh>z!1B_XI2(W`!@H$-DZ
zla;<?#k3Xfc1y(5mEO{N2xF9sXoN6YnR+f@`+?SpLVf5+K}B?FUi;ar*!J<zI+MSK
z3%@iiFhypIhFP1^j*ocscw&VcJl*E&J)rAg5b6}DkUJKm#X^<Lad9vPqG<Ss9~)WL
zLo29oD(9RoUHshFa%VGl+Tbm@tL#QFt3~`+ZcO1eY^ujNU1<0lu(+4Ix$;@Eob!YK
zszyt<uc*|h$dZRlF{*)u8PrzBPcXf4MH^e<^gA_;-pxNEj)mn}SZ})z)nINC9?m--
zlZw6KvU6e*d$%`50q6UZ&aQrb(#|GeMmB42t%blB*|=w>;t{R3?+f<2W0D23u$$qA
zU%;y(5@T3#%9IwDG~#eT4*VXx_?RrpPx8>>SJF*+<7*UsQ<R*^-X$H!0)16^({XW_
zXjCVehA;_()nL%UC{Uz?I{47{h5v2UO^ZU#%;&iO^cxv|z=%`{v3-j+8n+1$?|NdA
zyu^$s*g!xvPif-uTwO&iwExyr^{ed8yf(6v1kAN^KK~vZ`o{KK(E=@++A-iA;Ac1P
zp=HFoK#*#uZt<C@_xu~=(yV_&eR?AtI8uVlJ>j|b?@Bg0?+@NAy$U-#`q|G%r`d0Y
zpXp?6l5cssyX5D+&6_c7fJk^RuY3jxY7o#9{GW5u6ew}UK~wVUi{?#|_izde+uryD
z-EPZyRhFViAmT<JEU1+>rLJZYJlcEX`r=M;Ie|wPq}@!*(d+d<l+3D4xUp*#;}fS3
z02+z=4_a5R%hRbI`~KGzdUGUeWC*Dmp!F6V+XOe5J~L3TD|-V7OzKHfbo+-J_1INH
znnH^_7nZuIyWl#^c;~ys#?V1tU#0Y!n`?g5*xx`Q-oa_xRT}W4uEkT@HXV!Sr3|u$
zY4?Mvh!5bbAdbw4{>FzVzVYu;vF3Q2+`9p-tW76KW`gJzyPPA{57e<&@YMD7^vd{n
z;NWMGwPg7Fun^|hW)MnhaWf69+JtF&HTzpOf|sq><7%hi1KS=&$tC74(`qTUvZol_
zL;TMyJ+Ci9Ish-7iX5L|smDE#GtF<l4j(tDh;|;Kp6od;`SThq@kd#aBh=3Zt)*v6
zq@5ah6rgC7PGMLrgh|RcKI~gBrKM2DZ4@UqzCfm1R`;dlRInX<f;TuzOG|1;ZvsFf
zZ{gNIfp8X?)wE*#ak;jOk%8MRmuiUQSM%nH^B%zlZ2tK$Fs&M)3?lwE>naKAMqHWM
z{@<Q(Cu4yVMNu!n$vkl#_FYo}FHn5^;_*v5PtG6HfSp51Ptt5uS%!Y#H9i286v!fQ
zRtd*|gD^x1%uoetpO&9{9KIiq`@{AEdtB<^d8=A+U|Xt4d1q{&p1#Z6I|yfkn;DTh
zlt{<Ih1uvqrnleE3nDPBS9a^+a3rqM&EALDac&xJpVl%iZd!+_U~Rn0>?RNh$J!h4
zO2Y*!Rkjzh=-4-UGHRmTgL^0RgNCwHryK1YZwLDji8&quWGkT<>A8G?_90j0O6iFW
z*CKsYUYVCW`61viNzU1YqqD<5WH;zGx{JK5!Xk7OF`E-I^Z4i`QjRQo*F}eXFqr*z
zVMVjvsXKkD{u>`>c15rJCn!G!Cc=ZA^2R`|(-sXsCq<8^Uyfc{D_Hh!4&9%S_HHk(
z9J}c*2r6KbBNX1wbhiw~lQ=~SgiSrjp#ph>ugY<eFwws7HyZ{v#eKoCh#6(k?|>u#
zYBjG)X!OfYb)?prykfO2@>C02-`_4h4=O8D(e!V9R6Ly28G9qZrhR3}9{Xgj7-o5W
zmf~ldbT!EKGVYHuRNuX!#*(_s%!VGZH%jGp>*t0VRb{_dpxulqFT<QxAk)ltrfELO
zn%N?oE*m#}x4xP(c^Lagc4t62rNe4iM;wdi)xnTQptlaxeW0g=`kE4=5^%as1%*H$
z3~uUT?AvvLN$0O+zlq1CG@AcA4EESQpSbqQS8-;o{PJI9bu8X)ICSi)usIJ61EPiL
z7GQ(O)CW+2SkqDWdF=B0WrwfFT}Kq0D8Nn$_O`#?qj^IqoAc7vewIvU;h(B?icD8%
z_Gq9KCI6=pguTuQpvZb0$?YSM;F}}LuQQ?)12hzb2w4#*?|GK+9QyudQy;G;eQ=Lb
zViL`t-ghtW@pnS^jjMWSG=YTIR?|qQw4A9>869Hi1)Hx*Rv{!EqPL#?ZXWr>Y8Tjp
z0CCH1atxTf#a;GU*MoHRP|s;A0qF_ek20X0h?6x5?e+hGp@rvp#cz*PQa0%h2Moij
zUx<^pL!z{$gZ%yd&;MO|LRQ7ZTs7aN`p=B2%!SwY=G_KNl@oYSz^{@I%>3e)!rW~B
z#)6fTa=E~kfFyjX*p9EDC~07k3r{#XIh^90GnRMQIh{`;&Uk*ba~)vKc2xPsw867V
zYWoBq8y*)*Nb5a_s27`ZTygQAg^5O(fo`%MNP?_llZ0PijrC9UxUE+7tWVH6eHhs>
z+3wcgH%+UXrC*@D40Wi4OdqPM3yKbYqKaN;$&CULqw7q#vrACG!72qX6En~+GCDYm
zJbanAFK?sDX#{TNaym0rF2tpl5}aw*tysb<_voKqg#8XE*^hq(ly}y?A7SN>p#`w~
z1XnIvc)tGHaLBF$f2=i-aJ!2WH2IjozpW?33#LKaJ&m7P#ok;{L_w1Vr`k6QPf;0-
zz~|0|G&1aXUhfz(^_CW-zBs0>TtMm34i|aZiDG3sd5Bhh4ekPJA*t?3R@#A4Vkn%Y
z?S5@)Jb2lCp!?wcuX4sCBR-F;h7r%h6e;wt%VckydO9+L4m$JZhoOoAp#>H5#VV5D
zhf`#h#Npo2!AMoC@Oa7kT44WFIBgm%3J~kmOS{M1wRIY&s=v)vCi<=7%NzLB-xpP4
z-Bdk`A8E5{mFNE%xf$B%!9_%YzgAe7_Ius`D7eGM{L^#bR&$%EreR^u)xbY(e29xJ
zNyVv+bO7~_?j)04{Y}pQIx77X_LSPu91>>*)WINmQ#4pRjcKH9Ll!wn{AiVl<_hb!
z7)LrlprV9PJKS#5ttp7w%E|=80_J|C9ARD`GqTQXo%oZ>z#D<8rn!BX_BG9R&lj+c
za?Ou%qV2%lk^0c@Q8(nRSt!RmiCwNMKjjBnnjZ87iQ;@#OrK}ou&xT2Te!gJx%19y
zOxs9@s)!Z#nqj^A!oKb_%|H`@IC9A0f3alqr^4N)7{blQr3ht;&h6AF6%F^5KX|9-
zwrTgtnGF`$Ne|G`JxZkYDAh=*(cig)Yz3+nz~btZ_^UpF2fSxTB^i+xpY9&$Sawca
zqH>2W9{Y?j$GK?9LaO&{*F))z;m~YU5$NH~>W}A<s8vVTcV`ck^_-J$HVX4Gbqki5
zdanHO-CWiIG4;}ks>;fEHCT>@qJqzWV3N9}#}NL$RU@%a^lYW9r5b@Cb|u#gD-xir
z@YO7FufEH;#iqn}?Jz%*W+9P&gRwz?O!sN7G`=9+m8azNwH%v&M)4MCr?OpO){dya
zfvGdL68eNYXdBn<mQ~F0efG)$qLzjLu6o1d=ljvjHpXA0CW*914D!srbXj8IRFv5Q
zshwdm&*O#hKDxIWKc+5zPVZ7^OK&dEWc9)D(5_z?yTt1#=w#)e7q*prh3kP~yq{Jn
z#Fn{Gl-ke%_8g513`BY~nSRL15l=A|#){mTq22(SMql+~GYUG+(i66?cE-xm<|zJw
z7o%2e#z-)oO(#@Mobbfl7Y(kdDnhgJPWhB-+@`PkDNHSMH2n`@j=cv~_#_qd-2wWX
z@N`l;fkM_c&(b#6L&08Pmqy_S00x{I^9}p>4=;`p^re_imKxXRcRDjD;F!@1@GStX
z&%U(EH2$|+osfa=hCc#H8HPsn`E&GSB27*5bq3R|llbqK59vA1FNv2Z8n0f{!(!S>
z3$4l->p%JxVqu9(?u8!bkgk<_g|LDIS1tefSvMQg<Dg*LnC2aUM?VGat9d?S!9o%0
zZz{fIXZm+}7;i7GgRxyKxUg@Q%h$#m#UGR30cQz$(ftDhPRh#O1z%hcB5%>_qh^fJ
z>jN%+Bj{nI{k7ZXxMR`Gj!?&Z;~(bMj&*;-TsC7j*h_Fb-#T57pYBe>So3*Rw3S;`
zn(%b~COJHlDfjqJCqP?pj-S*H69!d_A3!|DvC05OA^h@2?}vZPsA`p&&IlR>S0<mi
zpYifK@fj_#hSRaZe}B0f5yFSs(erN<wn7d4E?!>i)=t`jp}crDo+LATMMMr|6XDa*
z!ce)y2;((CCKFCNDW^*usxPA6Dt34(m6^QV%+L_Yx_tiQr%!ebpc_}kOajwq8XD{{
zMAi$)u%f6o8gS&IGp%B`sKv|2xfc#*oR##^^1kVE-xUTWNSBN&0_A(syvns}?eq*U
zFL^oz-CThc`I5Uau72i4*)*+K@)ew(`U?JN+mfPz(eAlK;f?1BeG47eV1DF*G0yWU
z+$M8RRJDA$Ew=IzY?i1s2IM9LoY#_}11x&&2l=$i^;~^wSrdFDi-_Hs-C2j0${wu<
z<-@I~pfv||h=i%ATP1V=v6_!etv6*<IdB1`=iL@FNQ2+!<(Ht-=-p?%F1f9w*(<o)
zNLtJmOdEcFxu7qFaH5pN;@FvqxRAsOi<2j%?VOZb?2>l(Zr}PbtM0Ppa`lav-ok^x
zVL)tV_q)Dd9chYP#3@_DfOMA$eWx1^kXKx)*q-;#lZ_2|Jv4Hj<F%gFLEH1?sJQfW
zRaX@bzLv`tsu54_n^#n>RwT^hHWg37>(XAGgkD=m6o9P~KY6Ag$KcW*?%xc_$I$fa
z%EivCBSVH=f^V^Y?X1cQ+Ih<#QPbfUo>WyfNIxfz{V2L){W(CYpn|-cq|LnELBzQ}
zFi)b#9V%VTIFbXOTo+F!)LG5&agyi<^PzQP@$tO5K=FlO#n0-S_{<yIe)_TDz98-N
zrIEj{+-<YvH#W1cn{5B<A!F_-R{kUh&u?(<%47|Uuj@$JILWn0+sLX18S=F7`%zj$
z!7q$IygFOuc;yK+F6W@vbZk;Wf-7qGuG0q)4U1^{Y$z&NlIw=|Zy;*E|7QhTN?9~o
z!yH$l_B;aDBTPyy<`nFOawfJ29P2o8szwsF6((0)mDSLFHj+SU%XW;ZB-zbTHL2wx
zn->kPxo`@#d<D(Xlt0C;$J(S22CRnr^NT69{wG28jvQ+N6BiAyr7FF4f?0a`P)4lW
z^`fp$2AL<Z)(#~E46k(|B#Hj1QGu>5E-vHj7t^ieN8F5N?hFT^d9M<fM}hSg<b|hs
z#^fOoY<F0NbBSpOJkvLobI;(a&-?)I_i%smmf#2GnghEtMs*N3|N45_Etniqxl&Fu
zw{dLf4~X3QoheB#eD+rIOcZUBM;uz@Jz1}6V%jA)A#4J+27p1aUXfO@NRQaM(o(#`
zqj>N<taZTaUGbzK_f02^^g2b>QxrraG$=ml#)l-)83OLx{cJQZipY4}Scg{)nr=rc
zb%AjZWL94O{rfNm+jKUQjA&TS&e>?BAx-tB7!GM3V^CrB4-Y)>mMGzNKkBrSqdaA<
zrTtEv7+cw9wSR6$LD2A-11t6epv)#{hkyxxIwkp*!F5w1ys;R}%4M>%{8EjQ(Tfq-
zc90XF_hE4h9soiigB$$2!>&6!J71{q8Uq({!@cBKJBVa>4eD>OuIbz_u&&aXD$#mD
zxI1)^i}3^r%cE=@^E;;ZQ?EHGpmx65D0s;3X!_^*M%@{w+kefpp}h2UvKBGdoQzte
z2b>+*LRkkkoHM}7?(<Q@i1F_Kc2T`c`{JV1Ewg>c22@3#0lm7&oh@bIRYMcq7dTS;
z4$IZxBzk`+`n>DaE<Esj^vlFh--4dG_>Xp~x{$1C>iyDTsrX8TCAJeN9<xvvr^l5?
zlHZCs*jvOjd{QVz;O1a;Qe_i9BV8h<DJm_~bG1;72I-Vx!^WOHg$J^)&ykfVI-fpF
z&Bgj^cHBGk;L~ny=sEycz+by>@wY0JG{8@Dq}*)=4=0}>?&5a!E}97?o02yFR_r(3
z4Eg*=h&1=tP3I8%ESY6l-;=#CS77SmDHSVAZxrsQ#WYPe#y@NMe($7{kNQttRAbQ2
zFp*OjyVA*5Wu`e$=FPI?>KU1(PrwQSXh=9nCb>mdkC_e_HeZL=*Ff1R(qqWD*A-2&
z^7!_oHdaV2`>0cgV*)+NSjoZLqu1ELVW&H&?GR5nndRm2P*#6@adEtAqx}?W_T4(m
zm7{y~pPxU=2{ZH(!K8F~s;T6rFZb-g+SRpPshw80%#3<c;=a??8JLdeEl~e4ggMM7
z8Ou3rDcm!J>{uNwLkF{`LkeULKSEQgmEPnfj^EUr*6Cy>>$>zUuP*1Hhj*kU3%uyq
z@jsPmX2(#a(r8EUmYo5(Ll#|b&bH^Cm`;b2+--zK%Y0*&MlD=WD@8ex)wQ4b(%I;x
zDczWi&222JK~W8-g`hU#1uEe;87X@xYXHIfHWCJEN?>xxegCT~;slDwsw$q+F^@&+
zB{j`>p<I&XTvy4eW8v2V?srYhE?%j;5V~>ICV~}L%Akshn=M3X4*3qRH%)2m7c+bu
z9szJcSy3|)``2O@<!vN_?L$G132AeC$Lk_3#c8C<#zm`nyEzE!q)9=p71)THb<?V5
zf#ouK0{tUEyONkmPr!|mO?6Q4=SpU_X-Ykn{@Y@kuNjhIs{V3r8E^$kO&SZGi_6=m
zEHnW_HIm^~KWL?Brk6Zx1h;Gdu)od@^13}}Wpp?F@E{}<9S~TmH+@n%s9CaNsZ%-&
z=a*nnmrDL<Wh8ch6+gS4VKXG_oR%TZYv%!EjJGS?pNdkt!*dpdxPK<R;&L=)t-Rp4
zx5#9-lgOSh7YkT}mh5e<xM(jy<TpeX;`GpgQ8X?5#>H2q+FWZk_5ieVlGWYKp0PLz
zsy(HPB5&O%g7Veg-s%-lb&xyigxdB9dPU@ndo$XjBYXY=huz(Ml)Y?m3=DW{T+4Z4
zv3`WbE=3*YTBZ(pC7@ElW~pe($%&3@u;@wj#SLV7PC&C<2<d@W2smD2Ew`o3UFVZ1
zXs-SM0B}dt>~l+GsP!f3Qt^#Nt15-^1WvsaESXXf)u_p$X;hAVpcivXQ*=KYyvPAp
z%UD^gY^!O1E0LQjwH$|GdM@HxDQ+cc<mia8;5|Dh9QAHglm6$MXx%HFH>acgy75{A
zTvFBfk30!4*Ng*#tp#$Z{o*GlCo7A>8C*qN<=4{B#kv<dmm!dVr?jKxr5e`VelzQ3
zQ31@7-^NMNHost_gz|!qbK&;M*ALUlf1F-&c3a9z+Wm^@5M1$6?UFA3!K+-8tMAM*
zOACr$Wx6=;GU~wLIc>l_c5;%)C6lPh>BJ=o*#FN%xK*V+XImth{#4+93<b#4owDeQ
zxIgE%RXBuZPlX=>3a`VM<wl-RRr;b^u!He_?52J#<sR+i-TJH}a4D~!j@Ia4q!yys
zBA@It788cubW(eTeQBu7AeEkA#k}vM6<s|GN@IkIb(>(8WR&egUQV&JuK0+@*4@Rm
zsS1(quvV%&YzbrPw+P*hN{m2DRF8!Y@2fX8+)S%!jY>8Q?2Jm*3bLYC9|D`F)aU<p
z5wQC|twHUe2?>Ra#l?=gxP%}Mr$zbe<_bXW8H8KB1?D4^(8e^8Q>DP&G-UMz2Mh#Z
z*Kw^*yXR}6+aW#ftGANn$&b`74MlMOR^EyZ1^hVhZxwbTTfh*ib_<slFZ!eF-^B`w
zLrB7%fnFaSZx|N9x}VTO+)c>}>nf0qmDd^}`D9wdV;(5u@4r!qdfvk1dRfW|owC60
zYbU&?NkSIC<beqR9&RoYGZQ{#mKQ}%5D)Kv`@`+y%;uArOjZ}^{CWQX#Kt0v<BHwO
zMiunMsV+qYMcj64XI_vP&3BpwYH-GAC#_Z<w$;;b&JojBLi>U?Qf+Q4D#PiMJxf8t
z{Ky>9SX3)!*1Z^;+wnfRT;QK;F!(sCmG^}7w=^5N^N3ZBt+@P(gf!U}YdTo-h7u`Z
z_-rMeC+S3=k(Q9u;^jN;Glgh4I`?k^xY12@+!^5Y8Ji9TA<wJ<foKaBjI<e!2vr1=
zP*c3(Qq|w;#Jv!R9(W2&*>@-s6}1q~?FAhxA~mW?#M$SYf7@NiJ>s+AR@`qdEgrXL
zd}qoY(aTFJzRNU!v;fb!4E(+d9%k>y?`f`rWffb%Dp5&JXGa1fnHDHuhYV`eq@$Pl
z=k|G?l2R|PGcLFX3-><-!<{XkxY(#reOlT9atx(vzSXc;@l0_*a~nL<J=)${ltt*M
z7aE?oa+JAg9g8p(y}pr{yZPbPE66l4eT9H8`SQlI*wK^e$ivnt{w*H(x7UUYJ66x(
zNU+E!+AEB*g5V79<JPcn3mzY3Sk&(S@tg-rODM)P`TiwtDJa1p$;*1{jyxSI0RvIw
zwByKv^s6vbwT2+NrWg(JAaLuklv#l8MQ!*Vh|)j!w-N-d=|1u6gx0%kaTJNRc{Xnq
z>}uyTpKWJTLTn2$;|lshr1_8s_iTpodoOPb)(336sj=vUG{@aZd95-(yD_S@_Yl?q
zAr?6awc%UK(60Hub!m4i;zn*dt4{3W3o!9o2B`G5$i(^1?a<Gx{LzHz>WT+W$lp%=
zvk(7<*^ei%04Iw0jC3nMW}cpGEFtONTR~&IRQs+=ePEtUVsbI5%$BF*Fp+{}WFwgY
zyynbp!1$8!eQ8z``ClMSx!0-r=xn*>w$5{gL}1IY%)a`h(Nmyth~NPGSzP(L>lq;~
z$`F0D<#O=o|H2ffYdA+91*WDf8BCK2)HqLI%Tj9fS;+g(L!7Sp0fiI>DCGixQ*}2@
zow>K$4s3zcKY&lF`y@A}g$$OqdUNYy**f#!u@rqv5k<fkn*pobMbqL7gOrQf0YR%0
z)n_8&qLdK^sFCT2%IoBed7Un8{?i8jI*kwIwLCZ>It`F1#xsdFfBoY%Z}a7Ho}Z}O
z!F#KPeN~wponu0Fp8);ro)i@zt%6Ac5=Vq?nF48Oxy8`Nb;<9~1l%NLo}*kMWbCay
zZafm3@OEtS&yw!g_M3T~hbKKnF87Y16i?MSkZHFxm7~9@m@^vHy=#@p%z~9GcINq%
zT3*vG;pO>&8r7?%Vogd7DD?dmPobow-q5Un_?)Tmcz+X7vqp@sZOI1JueW&|J8EJ&
zOD!qzr_ZMj9xD|mZUyAMip)v2epkWcQX0}6gKDLzw{VZ&1WwG5yBZk*AK74yxZ2L&
z!?Q->xfKR#gFbMkwF;lVo*IO5>8cwW-C{Q5qZX}4$m2pF+Tizo!n1rr*a#-IFa#Ae
zehtN0?eHD@xn?tFA`a0%UqWtMG2rEM{UMtN7NXKl5MM2Dd7)->+B9KM8T=3M*B3t7
z3z8gE@T@$O#I2|<-fO%Xi#snFH8yLGn<z2cxyoSd8>w=|6p*MWTflV$S-Pk*xnEOA
zkO<bXTwyYeCjaKGeYwCkhs4+UFer5&bqvd}Q|@^=8zY>MEbI;2da;NZNnm~|iyPc0
zOC^M$N|jP3rUtCqSs9+~WT#omq<@uWgY=SR3Dr|h)2$Ck&0yleU9@{wDa}^vy8@(3
zy+sAuS9Ae8wwSqJ4rrh6N@b(TXRV8e!}X}dWyTg-G5A1*v*(&yxnvB^TGaps&+u5G
zj<Z7#c#DSAHa0d&Nt?$dt5@sA)qBMy)AaG+KkZQ82eJI$^{_;mqE3R(*fsU(DDM(K
zZax)18|9Nf|DVYujQ`MsR_(;tO2H;dx#|hycGscWA$OAS36$<oa2=`QSuU3gXPDW-
z1W!j)D06Q#yggPByV=<D+yBHOEbWms>ntQRCnNiR`lNYXDsk{AO#hsna?{IM0bvW9
zVFZdxKY79l%%bxiVbo??BI(w6l~0~ODLE9=eYGU8YNVKHAJ{<_sJJ?Q60oSiYmowP
zB|G&m9JAp1aly8y{6e2#Xv$2V*S9ihb!zAhAG!m*S8a=Bi3woRr5$N8z3b8x-rrv|
zyvEzSzi@wZSs9iuwqNXqI5f8w#SXbm>l3)&E@zQsHyHwkH43YcHliXLz);d;nnDeG
z0&wOPj^+$!x{lYT+B{`C1slxsx#ceH>payT_k2Zv;=kTa%m6zlb-f1P`COT4&sj6U
zXLj1B7;9@QCwJ}Q0^)>87M}|SFz~k6`E5?L{gN{1m@N)QulIn-qn2b83J}X58i5g|
z!de;ziEb`SbjODt9H}LJ63EbtndO9Y0BRDk9N3%TTmbk}=GJkCN7fHT$NSV(FUJNm
zGQzzr)7|b#<y4UUx+Qc7m=i5}FqTxgt2*^gV&Asb6n2J_rXV_15j@+U|MJd_3Y@7o
z_8~mVZJqDkIJ~PXv$Xa73SXMR<(kvOfVbaiDci{%LHNJFgVUl7W!*RxEYiB|;-9{U
zre<tH#_Act5e@EdtHI1N@5YRvww3z>fgAB$pnl8zmvrCIv3`CN67>MCz}+hDisp&|
zz(1QlM9EyL=~u3N9qFK(XY-LTCvqaM+3i&7?nWmkY0MZ>*e$E<w$X@?<%<M=bI+f~
z{o5)?b`rq^y9F!OXZ8JNZozuVR&pnVY*WSx$ACq9g11GJtwh9b`~E=CU{n67K&{%A
z(7ZIox;bu~7Se`nKL<2$Em7|HHx=we@?2D>#1&<mtQOh=3EhX6qpV|}0ZA^Rn|A86
z>5DvnFR$p}Swe5^6iu&1-XXv|6W<ygL;*1++Out9lAFa>;WI?)!5D+_DroV{UnF+D
zqie8w5!3snnWU28!8i3Fwz6il2iC^Bf=*7JFbR=XhWu0+2>dy_VJf0P-pF-kbfKAG
zFr5z2kB=89Goifg;r^6zf!ASySg8N;dX#~v^q>tz*f`dfJHzB@o*hPW)J7r{3PfIi
z#V2ns_=S^l2@v|@rT<1Rj1r_LC0a>BF)n3z_w9*#JccvLStD~craOFO&W(tUYir(i
zyQ*eci6+gL36YizufpCgKbWH1t+UdZ@+<XSMjBbU6;1wgPYE2{rXT>9QYhMUHp4Vc
zFI3svleDMwbSt5J8XHQz^VG`oU}aK!3K4_+aC7Sw&t1WznnFEyH!9I&6YDLOGq~I?
zujgTa?efM%<elfKd4MG>q-*a2H8(ARbA|<O$l4kuC{LiY>7Pst21BYZ_)Hrg-KIj2
z;gS_-i-olbyCLDJ87-ZRE-|Z;(Ajjdf58x=0Ko8W(|v=S-!H2Y9~gwpD8@zM^I@i8
zjn6m9<4cbI)OEGN=ISXPTw|Wf>{kfRR%fu6B0Wy+Kh!yiz~KxH>sZ^G^ba38)Hxk!
zuE%<WT2y6U_*n7Q#3@!xZS9r3fXrCnd7HFXFWB>3)kEdNE&D%ng~gOitxMvVj)TB|
zq+)m}qm#x21ftgqw{9$!O<0S}qK^`+1-xlbW8!}rN>(J(A6t3UDU0sGWi}BaG>WC+
zu3&&#;WG0IZl5V?$J?~vdMnP2)3C_#tGV22AwIrM;nCane{O7&yX(|Hqg|m6KBk19
zmJA%^ge++eOXK+pc%PSNau;508lUDyr6`T<YTP+c>{ac4qv`ra^__<y-cN=?x<QMA
zraq{u;l$0-fP)Dcxr_GS_?^8xXb_^*g0#^s?=OUzhq1qL<K90*c>{x8S45P9Ql*mV
z8kxp6@9A~aX~M}~Yeja0*?;`3TT)aal@2WgLXip!5$UPVsA>vnBOOHxI(n47Km|XJ
z4%;=o@*%T^9slS3VO>Mn(rpMPI_`Aur<2HIPvZ<+t5Lo_$)~DMFv(h;vi%jb!Wj1;
zeNY8i)E8$e4LcQK0Bdl8!5dQ)m<ia;2A%>vaD=pZUXM@R7E2c;fp(>HPwYJE3|o9w
z@Klp%wJJ@I#$(K4nlD#jTvQ-tG!B9TirJwc!*CL4>wy7}6Ah$JL!<g7ol*|h2p}~?
zT9$V;X|wD1f$o3h{2<Vnlt>={(#m+B#NdTv@*xc}|K83fYyOy(Y?q|$iS7uS0vzo^
zWeClo(X(gC&gSwQPFRslURRNWFQt)0XZds-ZdmKf`yJNy6wElYDR@>DN9*W6P<9Jv
zP5RidUG_0iC-eb1ht&Cbh0&UTwB0h6sbVT}*lg2PwS8kFooBTJw)VA8yt%OXO1}~o
zU@lvA{fKY&6`sIw0=9l>e;-RbfDJ*=RIsbZXtlRpR83~@UnzI3RuIS+{5}F`9yg2H
z5iy=^p`Bb6)qY}asefr2=U9#oO}cQf8}N_9i@1&si*M(McvkR%=e3|S!<;c6prV)~
zzEP1;*XX^19f`z97tXKTV;2|=IRYNd$5ns34j|B=W}AvA_SHsy<hPNX<D6%^QfN*!
zrN=>llNzi&_U{g6Q~3S$9y7_rj9+XGj%C2?hRW@=^n!aEVFDf7MWq&g*kR9N%ms63
zYqYytVeCXMoXRJ9>5m6_d68dpX(@)_<=lC7uVW$8{tY1kwJ|A8V(W;$;jFlVA4G&d
z@NGl&->ci@igKo39SzVZc=Ap;ZTEi5s=nB55@QED&T3&1s=xrP`n*93!qCI$ijVLM
zwXbGxCvT!d4d2=WddE8aDgBX{LT0L!kcC4LIHO6?&C~77mU^mb$Lg0mNmlDb$@nXJ
z@*4WC0g@+QnlYjbxcXaG!5ACK&{wTI+!58~ViMvE6>I=kSb^)Xqgu#jb`g$60N;}?
zi$Y=$%8|tw!Ps#EH_8@X38#GrHQUCF&ED({N>DESDR&xcI%yDb57NY(ikKa3;kGEN
zIHUUB>eITfG9Ebdx=JN2l_kI7L+|w8&{*_+-E{vW`j~nto2O4Edylst;~<w=-U)X&
zV@{%|ot^qyFD?Ll_R>kH=y_zp{Cv<LZ*l>fNx?`L^3{IR(<tVhG0gUCgQeLT?!Zc>
zm+d(b;>J=cegjbNP;26xZ_rUTQ*2PlV1f*M){hc6xsnOTvY@3*xdpBL=Qq}g0CCxc
zLD<rJ^<7*xOJ-Zcvioz~U#3_IYAn%H?#Il?LQzQRfT#D-!x35Bf$b(5_#R6`yVSjI
zW89j1*MK)nrwn56^GY4O8`kvGqbB(zRa!_yM8ttz@4!iQqQC3>kkdUc798Pw`+JTu
zlTkKT)^kICQqIAM2KZ)6s{|#xCs6c(9_Cl3KaNn|Fn(`vrpjr&iWa3*jQOI$t0S@$
zHO{$^bh1)ehsOi1P8I+G%L7K#^-yRYg~lBxRL<pewUibtK1xsb&Fph(&TSG50sdy>
zLf?&P1}guvFotzu%tRP6YSKM2#r`u&IfT_)0G!3f;%DoNjQoG!fNQFAhP+?cKcut`
zHbrG_I)F+&OCwX+_^}sOwAsTbw)vQJMF3Fv=k2l(H@gB->SItK#w!=7FcE<!?_U1U
z@7{c0Nxu>3x&Ao4{hFC%bpNX0w7Iz9t#@i?wBj&(r1H@ca?j}R;BC?IVwe0M)nlqO
zPrV51h_3C6ZCcXb`(5;r`awC{+Nkd0b;PR*CL+J=r@POj+)&^kHQ!ETPs(l?eZ`im
zQW5yB#Wz+{E0hSIyp4vL@I=UTUK^5J-{LiKdTHH|n}$4pa@P}|SFt*hGfjU+N}ms}
z?CMTc(lgQ-d9m#i$hiaVf=BDH2}%4CC<Ha7;2=x-9&Mur48|r?2dXF1QwPG$$e_t0
zWVAJS{5bLQj9)jA9D(awjVXF4R^82#XQa_?>R8mx&Aa~l*hkMq+VZ^7Adn495m_Ml
zMX4pw=l|{E!Q9lEMyEbUHov2!G;2t9-i5<;;03nA5@i!-v&0=Nk8u!A)8sC}Bl_Jd
zA?N?X&I_La#*~|qcgsA(x4>C!Pod+c3@O9oqThch6gnm%*o{`AZXSBrYuoHV=?GfC
z_@w)g8<Ie$w(Y&&pZ+J&!!E6{g^UPSKsgZYao%1&nZ2HrK<i-&05(4=*yViW_pvh6
zcZSJ%1ns1b;f?VF5QyBEx5j}_9lp*pBdW)^egKWEk;|5|s=_CPz~n}mY1b=MJIS9r
zAR4YEP+i_ucR!kmfX&OKDHD$ba_`AC$~#cSIn2&S{@$8ZzU2x;qvu}%+cg#vC400H
zlA<X3vR_4d8|;eY>oW(}(86jk(F*vR+<h*j;fHJgJOVPl4=O503wE!2?0mj~#0nR6
z>v<Oi2poYUy^g%L>0Kog5x~yl>Untc6Zm+FREC&&3T2`%iy78N9?cmE!HQ(4xk`#T
zI7Xgk8h1~6w%tj$G6=nX0h)0_&1G35Hc7tu$C;sTU{Rp1Ar80oM1xZY$30HZMHqVO
z^}|6g3C%?vmS;|ISx5A)bcOG8(j43zsh$NNpi1pCI0h+tSQm??8E)oo$twb*hu<vL
zm7Q+^Tf*UT&(Hsoh0<&6G5_uf<cm2!rS6e=)k{3c^Qy#v20-l?#w<t5m8)(P@=h3)
z3;GbH=l<=X(siVrWpMl1ZRO2F4iE!Cqb~R&b2;lGq??5fb?z*@Z6T2YxFPQCf+pSc
zfosDn8GwmmhkIndoxw!<_jvTX<7G$SZ^f{<eFm=S@}kKp>rD8MwH!c8Ng;cQ6b3_w
zy1(GFj}Kw?18LG0;rB^gqD2WK`A`+0J$t9|L;e+^3+ksF-pci8feemb!{te7<MU@W
zx0KEq>Ov-t1S71v=~MbX>>8E`upAZ)mI~uRt+i5T$iG@CAjU}i{^UT=DsvQoS?M+Y
z$rL*C0gsLKVtXf6sa#lbX{GT#i$;e!?LQ)<HBzE%d}lPFX^PnfskYjdJ4Id0k2&FB
zlp8?5FNWl@rSidh3wTZo=iUza))Y!xrumZ&OGzEjjM_|Gxj9951cky#S5|3P3ccKl
zS%?*Rh-y72^wKBwP>_A{Vjuj@m}L&A^B+HdBAv><o|XDT8;6x{i&Bbgl*K4wS3ZC;
zYqc6c7f)VEi^u{<hl*!RYgq1av&rCB2!hN;R+q*l2Cyo+sSlzFB>F)s9CLl&9XK2=
zc{)?I3w|w4o~SPM-*GtqgWJQb2s7CCZxFVG@-tr}q5s5e?jV=t(baASQ3uE&D6Qz#
zR<}#WFpt*QNy+FJ_vOk?eL5#}u3YB5fAob$ltiO<Vc)i*u)1Bc*Yk<QUDAka^s5_*
zO;Zo#mA}s9$29SbQTrzna66A*{<y4T=a_9SqKm@<-OUd3U^{5`H9;|w1qyXKSRbn)
znABkz(G-G%j-k9?)vK(OQ}dK5=C)4VsZ-U#BV=>K<1JT#15VYk_Jct;Pp24gp0Q4o
z0SfWfTM`|08wj^oO^w?i($9?Bfijm)vB;wn>M??l>1scYWB2Cd+7h1ucW!@x$MG~H
z?sDP{5a;cNeliawhhz(Y2d7z|1j<N7;~l589WK>=$;Q|`HO}S<VRrGsHn%4&B3nIg
z#0@n~MB#0=vgIfns<LMo{~XP^+`CIKN&2MdYxF<wVr1da&g=wSD@`9GU4$DVBjmFV
z#E)wPY>m$Km%MK0sJEu#kXTwWp+;#RxMzo|-LuU$MV5LOJ07&atan&}z}nHoJmwsj
zYM5)j1lQ!z9Dqoi+pkvgGd#ic2$1BqQqv|&vZp|T9nPf9QCni60t?DiOcgax8)P4E
zAklf;zQM!zGrGFE51(`E?);U$E-gJmigCLR)oHHw(!a;ym;C`G8azu}%`2C^`2RZ&
z+ZVRI&^{-zoA7=YH+@Z|?btEOMIY9_QK1}WK<fQNo5Cv6o2FXYnx^`)#KjvdX2>G|
zzfoSKAuo*+_DVShFfux1(ygXgdZK;;7O8}qADhuC)IYfH-R5U@*qY|w7f}Cd&3vmg
zl%TBnT(dM)shVrKO0_EOzt9ot?ZdLFPWX`kroa@eiZ88q-?5FiAKx(R4)Y5F31EV(
zoTQu15;B5&B4!3ztGot4!NMycNY+u}tLY6NFo-l_+kA6JB6CM@<|3#Z*D07{=D*tj
zSvbn$h(q4=0?UWzWF}C*J07!y49kj|cpdQpFUVOgedS|?%z24HLiFPcdv_9yzJQ~k
zbp9Ri>4q6_IBihINe(E4<&0fSxW|F|2euVl33irx>#G3a>r0HK9vPps8{_0#KE%7b
z&xp`c?{^IEK8_Y%aeJY*_U_AsJLN-}haUb5J0PFfRfJoSeo^X=De}EN_n`BM$+`5Y
z5DZP9G3ugOe!t?#O}Ici021#yV?ha1k(%uu?YSR3QHve%+;|y;vMwe^(LBparai_?
zHbXi;{AkP?h1#v_zOgx7=TZiUQGxgw3>goy6#lYJ{dE;Y{v)x1V^D=#?F~8q5p-0k
zb8LSI<)s@V*j=U<yks{%Ry3V!oWX!jK#s?e*g7N!t`AN@zv=eBnXc?q6q}HKsP``7
z@KvwbXG%rSj5W9A>zj?Ago!#LyeA^ZH;+hsO%m7xo&+*$2?Piu=kY4xlSEU2!q7IR
zkq?1hH+A=l%(uykVph37uh<z!sFkKZz@_`7^o?AWRzT--uB5v+yHb*=iXr+EQpcFe
zaH##DzNPilT~gHXDrH?Gofd;-kRC+WZGnk+<G6N=>FEmb)Det1bp&D*IaNRV!?13Y
zb1U<Lq>4&`bAXO?J&3(XrWY;b-)(st!LQjMfb%SZ>81f2*i}Pd8z?9Ab5G6UsK_#-
zHFi7KLUPv~V8R7xRZ<BpRj6)!Y27Kdvw($UHONl?VAPn3CYm`Iv0d36D26J4b?G(L
z@!^bM24}3kyUcA{g7w_WQ{xrsXY^((DA{?5ofz}S&Z`S>7vG(3BU?259w?cGf^%&N
zlKpccjh=PlDf!Fn*JQuCDa6}~>~Hl?)OrGKM_RD<%H^!(z7is^G?FiR_M=4%_!xei
zTk$Gmjiw;wY45&6JYvRxH<K=}H&F}@P2UtkmpTHM1_oPDX?y|qMS-`&bFKrwm;#*y
z_c14jU}Sh?WTak>Z_5QU@}pvlfAvYxt9-U&Ct=84v9H6oqi?NXI7fqBG+etpHlM)j
z9s~vg4gi^5I%&rZ%}?lEPyz;31<IDld6K!yE9hTrJuBaRb9eci*u3`aDI}ibz$>}-
zJVa)QP3eR6cYJj{SgbS@k~Gon(P0}b@@KBnsi!TZ&@f?>)T5%vjqfJVbs#<OJ1&W?
zS0FFuBui{crH_dVE4rVua}v_sbb5?XfPiA1372COpJRS-l84#ma87rqt2MWZ8Ea|!
z2R)3*P*%FEhh|4+9>yp<9`cqt8@)uYLE3wi(@qzLy0(&0`I9zEpMkS=`9#Dl1=w~?
zd{eV52RCVaqasZeLF*{MoI(O9Uf`!aT>2LvHNav~R8P|Ak6kVcLiD-bakd}tjeX}M
z1&sp{U22`(BfwtQM%lfV@bAt-<VcVBtunb*U{k^(6ewSuoV?sNyrIKEo}qE(Q4{3I
z1E_uB`f$Vq^KWRnd`P%;!@bp$vJzD%oF;IEL@GNfXsI^f(INdYZZmBOih?H40c!C}
z%K^HV?ty1OA{q`7QL-Oyn-{14UYM|_SJRadQCg~ti~$PDJJYA7!s>uwfA8!Ug!@Xn
zpd41@f@T+k!GNEK6aG?)zngj}pnBfWzP5hH@KU{Fg$*N1EjqWttp(Iv2y_XBqLOxm
zi3-@7+Yx%ZpDGTI!dB*m<D;@t4*bl;e+h)6l!{rY^fvv4zV3djMz1svt^L8=vvxCt
zkay_8dQbI^3WqEXUu(J8f!p_I*5N$m;=@fhr|znWjuHIyp=4n@4@1BSwC&|G6vFMh
z5*k?+1J(UqL34%*5D5w&V=Y|#lFsryJWaFP#Z_4l0LKV`3P(E^3fqa-QHiu7YzfUt
zlfi5!Zm?$R#|wV^_NR;1dWSv;VnOvt_wk*-NW0ol7cTNpEoqD0+@e}od;|1|_9+Ox
zXkoyCV&drd6q(KURLLl{OS<%zdauZB6gQ@lkca{?*aDr28DI@T+U+8m2<N6RGL+6>
z?(gJlo`k=bmXwqn%{AWZ5U-~_SS4$JS~Qj*lI!s;dILGE5;qCcJb0*QHi~AR#02`O
zU614pFhn|lwHoImzxlwnPjROm!HF0$qO{ofe7W$A37pjkBFvc((4P`|q6;lNQ6_C!
zJA^z8?8`%YbEa3W4b!!p<er75Derek8$!Mb-CgJgz5|=CCw5>A1sUFbYuax_$VGjB
z)t5U;630d=O~9gs<A1ay8(M&>#f)1N)fcTN2caan903yh-tOlc+(*}Kh$}E@G2z)w
zffnZc1=5D+dOZX(tV(<_vthNcuQNn4HmoC&7E$9;Ua)?3g?TKO){ZTn1kV|7MceQn
z<h2+vk?x5Ju{|y!lb0=*!kXJ9UvCr)Tde@66QpZw^0yS8l7Iw?LZE@<`XO38y`OD5
zwBPU`X#S&<OC`fIN+%b?58Vlq2^Lu_hBe3e7mx3Ko<h0>#ut0tzMkKZH8g)j6sj=>
zrcn|UmGKYiON8aS><p!0byw7kZmb%nVDb*E$dMUgu5dO%@l0%J8`;H204P_;v>oBq
zLuA+4CB_20YCqQC<lHl&O}=qpSY@hDo>dFHjFnc3Jq*?VR3CT5^j4+ts4}@+MO9Tb
zh%;)dlaP;jA3#H1@dV_0&?cDH1;o;D=JatnhL?>VhY~aoj&Rxx4-^|%h68JVkzcu;
zNBfFNO_l=LnMaKuhH{3rbnsQc3`_d-m|ZF~6(`QKUFL4%@Q({)ytty#^hSY_)r_A^
z#ioRpwyeojU9%~jBq2CdJ!y*^x`<7Msn<Lc+R}JQb@<2c=3RXjAlLB3aH&bq^^h?P
zpbZQ?g!|8{I-L#^PKkPxoFwomOn5Ju>8q^#**Rv2f$(Zx#MLRDsoeu-nF#@x69Fc&
zT;ScpzDG~*pM~kwc?#Lm)_Cds@V+MXgEBj)R=(dbHa@9MM?E6nRgf9ZWo~26tjqR)
z{;)`iaGnnN=-bLZiz|d31Y-t&_(R3PES@Th!vs}xkzAp$4sxW#&u+4ZM6^m}rmkP;
zb?p{t*XEszAMSk6=SVA(EgqK(L5yhJ{z`(YxnL5<S`!(np0P1~6=X$5{X7UUp@7_)
z{w7P2fV3_)K`z=Qt;q3lT{S=}YMenPUC(<;dlxYBpbzYF2EHn)YvnV8*#*BipZIT?
z?FQGK{Hs=+dvxD4AO9J);b7w!tnX{u)~Oca8H?r;Z|q`|<%d6i#;-r%J8g0K!<>1-
z(W5)Wp?^FV`{`bDl~b~v8GC71Ug%_56PN+9X?ztJpBaYx3asP~@nDL0KHIZ4VJ0xn
z_#<H9-MPJUamY)Y!|_s;ClKodM_(lHOv!_{)w_Rt2G}Sa@zNrG?md<x+)UtqO^JEy
zJ$K9Gmvj&$NJ2f57NK%soWzYebA~A%zVq-)25S4&<4~|v;SRO^MgNbi?|^DDi@J?8
z6;Z@d6qPbqC?X(&h*DKV5d{mqs&pv|QbG%56vhThm2Lr%E+zD4qk~A38k7<tQX>Qi
zB=6h==l|b(-&)REx`st^zkAO;XP>?IkzMk(^|G+zm_Pz8<W_NQ!d1?yrs~zJeec#C
zO8ZTgr(#p~;_x>g8%t~VXd6UYT<_*lG*O;*rM+^bJS-Q4CYb4%96fq9rN_b=<|8Yk
zq~OTLVjrqiCk^2B(ojx+{0X??eP$opuYSx|>F||>iDF<D@avFO!_rufsuk%PBN)Jc
zENE-gP}R+B2j#zlSX`VRcTN3Ze>tP0Kg-NGK3Acd<8erI__k0#XW8D_S(3K!*v^=8
zMji|Pw74_50+{J27>Mb4jH-EZV8@q2@hZB5y1Y_a=45uxK7p&JPFc*kSS6noLrRmV
z*XNx}3jwSj5)%4{QePeRb;H4c+D4+8F-v$pUHoU{LD348=QZClQ_GX0dRxS^zStU5
zUv!Y5CAJ@k!s|>lW^ON)x8C{9p!hF5dP7vyenK@;`UI{{z7vaT3)FaXXwFwbAy>SF
z{RnO2&qY98;qxDbU~t@%Z1oYjyu}f;y4or;QC)YiZdJU*Zx~65Z<I<(5pRio_~k`P
zPp`eJT-xFLp9AxkGYmP7MP$70Whh1(pH;n?y`;;zVqJ{F#GUYfhH(GU7&b}bG~WcI
zwqv{N^NNZnSfFk5VLa2gRGYgEGq%M?+-oU}CAr<<nGLgzmbj~eYSGD!p0=e|SijHt
zy!C4JAW$@;DU(-NTRmfgp-IpC%P?}`{M%kH++ouv%S6}Q!!k4HPTp2%5~widtvfVd
z5<7YJ9^b#BMG3kS^XthGlP`4b1jjw9+eckhYP*q-^n`LMwho(Uza!p{i$DPZwPKZK
z0?eKDe`X4E4ITxzCpU0qqcmMWCH)>ZCe*_Z9<umL$ipgE4du-4BAquU#`c*>sE4Pg
z=dL50QM;7OC{;pNH+0<V7xcN3T$RNB?5f*1<~rSbBoa`@65eO!bImcUv;>9XkYa{6
z+Rfcz{8EuyZJs~*Pd9O-l`_!RclGB6OyWrPP=Q1kA?nnUN4=R&pK&et!aF|LhPA)$
zEg-VVGlEwuoZU*ME;*~)+pk3;lXW`0Tg~oO0G96TjoBq}N=~ysG0=o>rgMR&*>0)2
zoe<rv|5kHzgVm8d3mZF%K1C#^x4*ytQ4DXd<DqV)CT(SUE7W4bslyLbLaQw4`u;N0
za~;^!c;^{K7GPwQFt8(gfw9TxaEpedOF`o6k~cDL;&5VToK(haoqW~vM|&z5(9yvL
zU%`^>ebA-iHy<Y#WLUeK-^l4Esh_!~tCUK*(0u4mU!{70Q3O?sgEQ23`Kf4L?8mDk
z)6|2k`8dX#IFiUFQTsegzo~tGUj7yj50q=!Z44K^(2$=1O&d~d*!*Qi@2`{@;n1#P
z->|(&>U3f>OE~GW{m|_vUJ+g^g3kiZWp5IA*+Kf7^&-f(NEIC<KRsoNOkKv|?WKpV
z;*VZPuxcdQbBxwFX2?jgIYzDf>Bui7ow(^?_@qC>25sSkCKyH{VU?TUre-qiop>7v
z$zwzY6kxwCVkJL>37DWAGKKzm#d=J^eQn3&nN&rSZZI=2vVI9IC~w5CIX$*f7O0nz
zuBJuvQhzp$5CbXbb=>yoBu%uaH@D&QEBgz&iQMv_Qp))pZegX?pxBHsXoio+&BW=x
zX^QcV$1Ipe4}E8$P2vR=xu^Pqu3d2R<)3ZSP+1{MEVkkO%dc4G<aUr!vhJOcv6w}s
zxzM{gYs|t<UWyJll2Ye>eq2pSyKo{YDY%|get4boQv6AP#sQJ!t(iAX{hA|HQ(Uf%
zuj>7}<-RWHsh4_rd#c!ZMf%o(VXqhU$n>7@PMsO0(9Jr^z_fZiV*(rI4HIm%9~_;G
zX7n`)v{a)USw<OCJ#2#VT#?oydGoUfMXbag^_^OW@j2IGKI-?Bal4F{XRYpobYwum
zxq?wT6XS9JZvtoJP_C!`+R8$*ym8TL$G`LcsX}kmMje_O{vYFpv877ny!gNOj8*Pc
zju$Px4B&NF^paBMgTg}bh~jYo>zy<{TLmoR*~=nQJE}+U+hR;Y@ob+XPo(FIPlj{f
zsc;okm`GwT=#|+NHu)akA1Y>y56$;CwF4_~FOWQNpJPkSQ;Mwl>3FC6xV$tDp-?>S
zLZi*ieXo{A>lVrZnI3(&uAxN)I>g0{4zcuFkojmKy$gVO)bDzD2Rd<Hu75#y@ZIbr
z;S!DmQ=DTm9ZkkFAj3&h?mT-J0q-l#!)2<Y{d9u~FWo<zs5tb$Ugg-&wBBv5<aCH(
z6Uw`KUy^loW%68)i1C-0+uL0mBMl9;S3L^Z#YLNI$5u`=+LHeOTH{&FrPT*BPm5Pw
z8(27@Ov>_Wu&(ikEuKa-h#LPT@k5&~djp23KZLGZY#Gq%vnqT~ZvLmUEUi0uYEQ>7
zx};8pIWTn~H;Y4i<W%_fM|Qi4>81G2>7*nEOK5P+TdvFcs&p(jO%}00c^I1n805a0
z6Y~R63wM<D*+s}(v4fjL6GHb=baEq3jF1oY@u%$6b5l}V)lBj*jPw(MW>E5^#5M3V
zrn(7&C7*03m`~L7NwFVbRXrp3<o619csG69JBa?gl->WbU_6UV2$5tHM9C_m$LEcf
zw#E}%eAT;x{sbhQJR3JafMG%+ZnznL=D?Tz&``rh2J{bejrooH7+>t}CS<AulnLE2
z6sx`GvHw)Vf|W}c;Pi8^_HFWP<@HN`jx0(}S-4#F`B_=xVl@<4Z7J}ogP26-pOVTU
zU&)~u8de`l!hK;&DGJTkNo2t)-gocu5j&%qfC_nGVN|p^(M{43Tc<R1ou27JpC&)S
z*%%kQ<Yk-dKE8SuObPz^pRF^va?L^p(;wCP32aNL3pVZGQCEGlIlHO?;EhKy*?+y;
zlIc(~^=oJd`V}|+M}&Uk#I$4OwzYlZ|LpD``H{5Ba4W+b{<Ck&O~b@X9#`y_GSVX~
zi8roXaWS@|xt}<`_!~wDa0FGY+)bSJ#x4>3_9C-i0ZI9f#IuWrJ>re0HaP0e-`A5r
zH~jx>7mvj%-0wCGx{@$0heV>2>6a-rVeB$<l`kyLRtW5F0vXhrz9avZUl-(c`x?p;
z8NsuGdoH%<O6zIfVaC{$-oe9@C#o)hVm&mns;JytejweHAUL5PElgdtw()#~H(7tN
z4R!`kJ4=er8wYJ*mYEII&h5$;vCWp~h*!%?sk`%SZ}h#sIiD6XRr2%vI!v;?eAVg|
z9T|+`S+HVMp`Vtc&#vrr69k#T*%BwYAa}A{_rGmh_#!IM@FS`X_2{2NXZvhz<K$vA
z4Ci>GY!uV+Ey(CVW6;MW@|qwwBI)lt73Oi_PjBS^+d(F`qq(ANGZ@2NI)Rz|82-@p
zbBdPeSyZ8OhKp5)YaXUFyx$+V!}Ec|`UrO=o7|8<MXN)T<%cf%RfS*f>hdA*c@}|=
zxqdafXYRxy-32Q9LGpGi&GP%1WTm_T-HF8@UFk0-Jyi&-cqsZtDRz?1-JJ6-HV-%d
z!eUYpr}zdo)&T~iVvS4zebZ{dJ?{c~LTz?VZ8ZcKuih;~f$oMHLp%(vWEZ(eeJY~~
zJA+8beJaHNrd+EgFs^siO;-MgWos6#bcN@}@)3$0GCQIh=JFPLBWh<p!qx6Nz|zA|
z+B#r#QXN(ay5E#Gb>Px+J>FUejrem@zxW1?)>XemJjl|}doGcFN=1dfdG>4{B=qpE
zeRE3Aj*gD%TAOm)95Q73BG|ucCKSdTcGy6M`?U8E^*UH%_I?$(f2@&a#`{fjRJk4R
zCr@^R5d_<3zu##>-00HK9njdDWD>wQwP*`Ax+)Xa29(R|R8KNHWU<ON;NcSL>}L%B
zh;fVpk&dziSd$S~yc?pZS`0S=Ht|2Gw@%Lg#NZ*wZ9LavaLn75h5hM{&UxzMIKRw%
zM3xHS3Q^AS6CfsCjZjoML1J7Ncha|eAN}7-6e4x%*Gdnlx9jVtX{JyuV;Umv6@zis
zkSQ+d-nd9sR@c%OqPg7gosVQ9^2|h3VsIaJUy#RLp}fnK?7mo$=ida^o#98>pHHmE
z!7IKr^t<w$f;seB)lOjfXPi>v!H%kla0oCQqMPz`UT$x?3f+Xz99Hv1AwzFM;%Xmt
zwc{3w9xwYzubdbStiUuoJmZC?3*iQ8T?7{@Rk3rlXZNDxdb8{N>=%NzvjFO>*upZL
zKeeX-SgHpU7+8drob%EGe9+AxIz#@9Hm&J*-t5H5@zwS-noPRS$YlyDQ8^{FXYO@N
zb0)+n()}#&KdWnz$lO<Pzca3c%O_dc^?A81WNdrJ@KG%ONT1oSx(wIi3Zz#y^!T&m
zAp5zN;CwKJw%+lvrD!vaCX%yH7-iiM!<(zixxTPs1nhxPq4Z-akd;rX3yotCBq>_G
zShfD`H&Il=cjpf5^s_5JAQUuHfBo+_QJd#Ar~I@=pEFV-y1FheVw>&rR!S;IOTW5*
z-D-r30A9(!ttT(pfMXWy`0J)!edSM&uN&>zQfGL4@&fOR-=x=0j#Z&wx1>DYS(2S>
zg2mOlt6%Ng36hi0GM0#D_{Lv2nCB7gFhcpiAUG>7HXQR4!VIl11*2_pr>rU#F~G^C
zQ?FPsdOhRtl$qH@_n^Gsx&Z&5@Lub=#_u?+r$02gKUd*W#^K`$^NjMbB}u=Hi*ID*
zSC6NnfQDjYlD97CqB86oFWHniwz*jCYGv~eJ-T5fvtUCqe%Is5!H^gnbUu;x(Z9Cg
zRKEwi;mG~pc)Wcd?|#a)KUVeq8by6j5bV;^=eR7Zgn8p=9dwNC0Enp`M_HDgv6x%k
zz+!xu+0gZG!2bkuunl12VnUd2Alh6F>YX~2BL!=8zHD=^vwQ!`aZS>C)~nd6w|aNv
zJj!KjTSYP=AH5-8>J<MVh>x~meGlP#Kf2lF<#`zW+E%uMtg7K4dum&>l5<*`bWWdF
zcdla8y1G%Mz>Eog9wq|p_SkWtwb<T;c9ilCT@q(dwR{%fbnomS2E+wR=ucsCKpek<
zd1!r$-rvTtGPVk=;YU@M9@;<b%EuKW-&4=1nnuDkWBiFF`pp-Wo37A`dr!#;ToRsR
z6O*d9kTy7X?E=&Oi34Lr)v1}pbA*rH_sUW$uEMi@h_pWDStokRkdv=d{E~5>2smiz
z*fab(z(bE{#;Y21@OhKW_CtMQERKfv6$xyWr>R!d!$CrE4aL$08hEKZfA{v!kAdGR
z=)o>Z^m#-CEBXU<L!XcuRo*$B)>_=Q?+m7znPl!Tp^Uu=zmo=B%^jHB2A2fHBDoF&
zMUGS24W_?I=2JWVC!F0DWE8W>7h`!x$_Ac=Jm-21cJ_DOyD|<NGY;KiqQxlQhjyQ@
zN)b9hPj*$M<8>bA95jA&Rx_fd8!15$LnKG#Ig#qb6gdDGkS=%E<He-Iy<=I`D_1xA
z`BoJcZkW72^`rD{aP>$YU#B70y7ZaTd@=n|LJ9S@%N{jqOGmBjjcf84S7Zn?I~o;>
z6|g;jvd}SpW3e%Bx}$wfFTNly;O#GYy^QSCoQvby)>Y@An03|np98@R*aQF~nlEE;
zB#8H@Cl5xn72*5Mxn9{gYQDYc$L5PD`<6yZ&K*Y*v@_COv{<7ii|lp=H2k^h*X(Hg
z-;^FQSbE5)R&17fGw^raD$_OhyO&-Rn>9?G-7kt=xm@Tb9uzTbb#~*NEHXb=FL(a$
z^*1KJDJzhxJ9~YW(%1h;^p$0GZ*T0mfRuXG;}x$*8v-Rci>t43+~2i1=&b`GN?_>3
zg}?LN?Jfuz_&xbbjWkr1HoyM7fM_u<>*jO%woHL}i>q!%R_xJfxjo5T(=vNNey;4e
z2dLc1S}469)^E4?#AU>WY4AXCPWb`yl=3B~8VSxWM2F9B({K~y$5da-Rw2PD3<-g%
zw|vz;!3a>Sw0MvRcd`EgBnz07aDVl_DMZ49HZ+$GEttU>vdPKFGHI2$y;LA}K&4Se
zIRk%cfGgu?(5y}Pl@uHNIi)71B8X_U=6YRydY~xRop+1vLoOd(o$|vl*I`G$Nvy=)
z8kZ&xVa(O+O)(7%j+J<IGh#M{dijfE$rgCPs5`J_;b2F-uYg`Fy3o$J_Db=Ya2A*k
zdN2~_sxDuZ2C7csiH(Df?v>R`EEv^<Cyc9m&pRX`01jx)1L)GCu@+UUjyuj!CgJvJ
zem4)T+aO0LC#SPo#GZ8Vz3tc=o$@f1jNqCujRL|yAj)7q{`fWP^C#84)Se`V(dE<!
z#iSK%;XP7MK8*K#;<;d_`=B@+N60E;v*r7|DsJSwVxK2BVb%BY953+XeM&<pKp0~5
zYmyxVIaQnA_c_`fwocY{@>xHnJ+$buQR{vgBsD0j;j0n9bu0IZ8Dko`o8X1>DC`4e
zp1nsv($_sXmfvOY3m2!(`-%G5F)ke^&UFpBUU~d34B`&INBV_@qrF|NHkB?kp;moz
zBIs`0;T$04SdKF)$H85O+2037vv$Vp+amKWU%*Dep!Orf_N~OJNBN$6=^-5jbMRpm
z{8K6Y(Sct7ObiSs-q4A>^ea=}xcCI(9eZ_UXdZIsaEoSf#LuQZ=r4ngmS*kg0~ZLB
zD=QBt`*O&6YX;XtFY<I(Utl`$?A;?A3tT#5Kp@+&DmCY|r1Rwj`6*KBeRp!al~jFm
zgp7q8H2am4TO+1|3)7JFV&h!tA>U&{LC=mJ=Zh8EbdhuJerI?$jyEwY?#@jE)BJ!b
zta*Bdxwz_cYTa!3hyn+6<`D2udwIvB#j!WKxKU;&B>6n#p+l3c8E#t6en)%W%m8PY
z1yfuVzq7?H64{&aAQBXo+M4Bj>rKB+9CZxt2&o&pVHElTx^LVB|Lf|h>}LQtMrrv2
zVOfeA=x%cpJoC_3SN*6P1k3<s>dW4*olB|kpYaFjp3$?7>6YV*otB3>&WSZNppGif
zbQyw6p;L`%Sb&9bSVg&Qr*9@XY+ak7Z6ejs_@U@4(=t3>@8g{)vFV+L8v;MoC6DwZ
zvs#Gvi3EPS4i*>q%XM6-4H%`~w;hIw_=1P|{z*Lov9?TqNDL@Y-jq_k6`pWO8*eSk
zUdoH_mZ|zg9r4uc`}{BT8M>JrPxenf)cgrz+;tQXWqY7GyVRUvNBz#sk<k))miJeM
zk?-)mHqrealkNIW0A}e{+-~V+m|LuI1|dV9ZtniGX*!12nnn`UdQOvXOZNM4h)C^!
z6e$?PX6@l}p?g|nT>}YJ0L}Fz!|}`O$VXbpu4SA$Q@;|?fRB%~bamhOB-}eC?CPo_
zp@>PQZfQwcEuL!KfU#=W{0+U2X{v|Y`fE*8^4RCQz4or;l=tzj(G}&~>{(=>gng&6
zjn@=ZUwK=T4;;(u#>;;fT;LFm8kB5Fmha{R_GQJ_9QmQyye|%TkLZCZ-RkWzkr|_T
zGj8~+b8d!|$9uWY_0E@HNX)@tX11dN-SSnB8HhTx9;W$>+gQ}U0j}!P702d2sj|8{
z^{gfh<59oxfhp4I^c*8srHut7(r?UC#4!Hqjq@=}u_RNMD(mc%t6yjr8zM+-pwk~M
z_sb57SJG};vx$OJd)7wrG9>^7Ao5Wdw;}gg%BJXc$8Zk`L6keVUOT4MV~?0+Xa;d!
z$-Y|(8R0FS1Z|N6*YwAH1_o*E1Pv|L2;%6^pU|uRGlzsl^?9D0im`=n7t)rWS=y5a
z^vD+hwI$_`7L#aiLY(AZ_DoV|7bl^i^1|0d<sl5Ek#j^OS*=aqXQUZORQUUU6^+D*
z*EzOXy-3!L4$xY20#F>a9k(n<5@~QZ#t+Ae=a9-<;>0(_F--r?jOPnr6S5j4*sT!E
z6e3D5HQqkC9~yA<rM+f)(sp8#+mrpUNexSQtI+(B;yh%;Xkmtq=@`t19*F~{%Wm5u
z;5$zG$YCpDXz^+<^Kpp)^Wb5MvGQ^mYQNQ92CFM*-7l`H2ojdbD;`Xl3U(l###Nv{
z*p2QAv+}eqsk>e%1*-lrQtpxsdlkVdrJ;{U8|TaTGUIbnrvGF0r7`b5+ZVICJL;ur
zqWgAFz*$CvsF`IOlrt({ax??SmSm>nl^$5VtK6j1QawG1ce)M)6s}8&(ab3@(b4qD
zS~uKTjHofk#s0#Y%EL8QDQx%U1{h8a0~&X#_U(vQRLM^Iv@NE*S94>g<L>TepGX%0
zPPWXEX)&C$<I{CNQR-TR8?)0?FxT9;zbSdny6J(EnOq$2^<$5p6^t%37PLOEPrgU8
z0#eg-YCfw~z550H%_#LU*h8M_OvT-!<B+fUp^h0xLxfr~A7=SQ61ED}rN#jbD07g{
zu?~Drcb)gT`diZ`8NO^1{e@BM3=<Wpo*Sub>92D0@$uPvWSo-g;-cc*LIuBJngZIW
z=$RH_V%oObe7wnqM>VHB7BaE~2<&ZN9r#|CUv>PxyQ=B1Nk+;l(GjVs=F0c-`_jaP
z*XW|KMpvEET8Bo;EQPKEh29#b-8tbdv9l(oF4fdC-FQt*SCi;B2WqcWmYdBj*OUHT
z+G?Qj52m=b;qvx-=3}RCyYTf6X1ke6ofWI8U-EXI5XFiShJ!bVzOkh6*Ba?2zRG-Q
zwf>p)?L4B$j84y!9s+?-wMu?D;wxTzxd4>q{@nFLZOkM(HQ;qMH<FV<z3H|op}QS3
zg8pWc^G2->9l3Yrl(y__l$p+E*d@DgNNzH??=s>$4;H{bKbh-N`WAu1z=AQT1g)K0
zW`^P8pbj*;MFlzF{YkOX7?w%?FqEZ@i%AK|p<1MZ<BHf)WrDGa=zGH*t4GY@B97DY
z&nM5NIr0cicP*tpXqw0@Iv{fhU58jRpkbSwJ!4`<D-!Hp<y)8RV3@pmde7Je?terN
zY{(tY^Kio|?3O!u9E(!15NQ8aEdjEU;=^LP^VM%vudznA1!=yRxEiS@4YA1--OZO;
z+S*o>4xg~F>rvdfGozrIkeYR*nIsitbhNtsTDE+zP);pgi90FXj-qXlN`2z0(wV-X
z;lE_u<(Z5Q@6c(mMP_11Z2zA6$Q|?)W_Zab%^y^uJGWa{HbhaXxE{17kFH?h1%HO)
zttoWF%N!^zt3-N=m+wL_YOEwHmt@O9Q>o(7{N0Sh-Rmv=U9z1Uy01{!{{RDuU29eV
z2L4Wk>M!#NxnCcgEa?n>sTG)gvLgL14hFPf7CZvp4@5cFV-<R0-8-VP?EXv5oz-(?
zF&#GUK`W*%ukOCh%9S`y_zi{#jpU(edMJG!xB7#wOKHoMzz%G1Nsv-=o4LCLPGn!Z
zsz-bTyd=L`@_jS_Pas}p*NQfd-k|DCMsfXVePGm}048?dv@yyi-W5~zTh@;i2O=ZL
zFiChme+XgXwk+X>7x_Z^0kdqSZ@mM~*$c1C)se&2>d%~g%y<>UahFreDH_!$-*Uzh
z5|uK#=9TCfA*c)yT)i{DtbsFGU3yrRK4zFi|0G+z!yu6a&+O%o2pNx)+Q)<uo^vF#
zVlrd<iO)%Du7f4%4VUM^nKV6}Kl6c!V3=Aj>UHt(pr1sX!y-TCSq3KkNe0FHi*=Ah
z$siZ_l}?7$^pkY!zqF9-NqWQ#_oxLy7*=q8ID?6*jfpXvo5=WML6P9$IQa_fYXI;e
zbTYPOk5S9!HV-?!Lv)&q&ZQwzyXA|25Xu)4Lhqwb;}!*fKt1;9<*4u9tUqG71L|k)
z8Wz{OeH5=J-suC=&aL7}p%Qv;S4&H+uIJ*Lj+G4B{zZ|q(Bu85nP^H=bmqLzLXKJ;
zmat49%aEgnX@(n?PLqRxr|4~EF~{tGUmNBhMm4a69>*}EBUCPalHKY76L`&6eCLU+
zxXaBdcg?ze3uM*dw)w2zc_PIIcVx3(;^r>iG_FRE|F279;8sF-ncjf9!%IK=gWsjO
z$;ldL2e+(lL@GZX1WoMZdD?lgH_r*=DvXCYL8!pMM&*LSu@8&3+Ya1Gz3Z3(BQfwN
zD$+-0^QF?E%x26nDdm0wGYWIVhKNwfFb%h>Zw`2{y1vcI?61z6p?<LZ9s~-e{1Cer
z|IB4(Wi2(U7;!uyU2F}mwv0)-(sy8Hm!YR<ka3^zCKoI3zhXuEYOwrKl0n^b%R$HL
z4DE<=(ubzG6$FkthwF-BHieUA42!)6Clks;8TWpS1}jXgeijhO7vqBpNwc~QoE<h0
z_((k@d!Lz4`-c`no!I*qsH?Pm6)n~~-^FCYsM)Hce!!2Q1yA^pjuR~Y&Qr>V?FCZE
zLfjCL7>5&k;kF<=H`R7^b938qdEQ)@O+2U6L?ev45#SGxeZq+AxzM_>=T91$vWCTU
zeHrSVweEX)U@?*F3XN=ks|!ZS)ICmR2fpsRTHOCkpTkx<a~yzR0<P_kFgp1D7$4~E
z(<*V%S=6t`%58EZF%N_5yNHY|NLO`cfbJr@ySXr-NOZu7k_Wu&R)-g?iDxZjn#?&=
zPQ{Jdy7wj4_T<#D7^lo1HOMZdx>Dic8C@S@r6Lv7o&0!8pbDdB`oy`lOwiDaiV@Yo
z6OXE52PsFvARUu?f?188+`N5zjMg5@t&SY+V4z_}YZKInhoe!ks9R`8>M_?EO97(>
zBfX0p;)1X-mFenW@oMYB_@hjR9A#VYIFp9R!0P*wbcd6N8QJ<5lak7_$$)vl?%aK3
zps0+YF_h~zC06Cr{yWN?jH8^(a^m799Y3+-*c5P283*^2@h>(N_8EF_jJu*TKbvY$
z(4Z#XS?Hj8C!rj29Uy+2>jbYD%{;2^uB|^sHT0x^$Qo<2?Z~30Jui|t7j}=+Grru9
z?cj$MvEmQn$AJJGYCBCn9%OuUapT~d>_9`qFNbn!zgm@T@69Zj%}}59xmRX&@~&C_
zw8ECnxe^KjpmAaBVR0dQf$1bl;V*8u8z-Lz)T3o)6s^_mmS28>4_E!eo$W%sT&Zx0
z4xp>`A4(sn6DG>D$L=K2b$6Uw(7>4{)Jpjs+03>+(TQ=lK#8G}^;v}gwJ!kq5yO=M
z4Vie9@T7hId!H}^#}{U_orD=JuvDsVR@tA87sZtRWcs%x8y-lneRcj{Ul=4go?%bY
zB~ln*VuG{G?(F9$`oh1aQd1(2t1vxR8D0Ve`JiWTF2J+-EXFpN$|2Hv=-iDqwo2tD
zhumAdy*7nnsxz`!Q!9&mEJD(_>_fd0NmYiTX^t+kU2gqcWX>-8xbiLerY5E3Q<`yB
z{<Wf3EB0-9=A8Y9YflsBePo|-bR5th-He#o1*`OHCFw0bz%Y%_wvai1Oo&Low1b`j
zLr{wTGxUvGK7+4%E$hVr)N;I0wVQxl2(}Yi1VE)H)K9Kri~pNFpw=!@$qDQmT(Qoh
z52c~RMYPp4pkXP0-lyLkPLKc^Z+Lqv57WiTV$>uljI_fgNqTLIRhNGYL@SpugmWH?
zikdYt)*K{@4y0Rp+EhHa@4aBR!>-fLK=qD*ITop8ga<wTV_MD_AkcgEyK1llf8@4e
z;hA~AIZEk?nrsL)y`h2;Y7<_Kp-1<`^k)tpC^*egQ(S5LO5#^O1Vvw%pjT%Yctzu6
zpXTj3%BWj+)X<!dS?QF4dUJdNN9_LEJ9#>!tF=|G<^0K#Si!~x+>@&+g#;1cCt(^8
z8DCyyXN;C116$YKv6s5S-9GdU=(T@#*eGl;*kOrg>rwrK``z76+RyqqAowKg^D69)
zwT8VOmHYDtM$%%U;G96+&LC8>iIe?G{Jh<LPjQIbR6}-XgZa}ytUk>U!Talmtt))f
z-1}ZeIW8jG_qnmCfnAWe*Bi1InhAe`07?<s9QUiEC1x^p3cKA0e7-SwXu`;ak9U>^
zXH;Wtd*K5w*gf9hQ;{U0C|zZNsv5I_qzPq&vE`Ls<=wT_XXZGTU$^3XKz8oAl!6OG
zfA<Q+#J7HuxXTVj@OM0$Ag7jf<aqQtid@dZ1Qb&0FqU=*Z@;7CZvXPzVS9xfTLBmt
z80(CKap99v8A4Cx0XUMK@r#3})vI%{|6x@u=tvon>(l<*^$(6MCYDRiov0_B4U<tE
zM7_erG)_zE!kbRdpUn(Ts{g{)w5F>grI{u<VN^*O11v;;`5UF#FAkHc*Ro^(C8mGY
zq1Q+2o;_-oF^h9}x~uT;v%oh$GxeXcpovMicL%zTrj?MXC?{H!tSeE31kVBiEB|Ze
z4LX7-O{cDur50(r?8~q2rZiS(O6(6~UoUjSNJuqj@p@i#jM@lemg$i;fxy8^shk{%
z2v)~ut>5wNSytXxso~gSXZrlRys$3lx!4UPxe28A+#jP5uRqnR@PA2TQ#2WrpHo=Y
z`P}KB+CcN<O>F6|U1;Y<AZ99iFi_9N06#_2#G*i&Z5p$X$e7ntS=RlCxYvtRd!@R3
zF!lPGtmVNjW+Q}Uc;Tf}MPEoTBcXyL=y!8>d9aIXr#t!-34ZlDK6s$}*68wdzj-<A
zM4q*y>9*yeStxV7BakGLfZ~DrynW4&w(d^Sy?kppG*4E!UuL7$6fh*^Lk~x;`10+i
znM#xQrzw=CoDD(BE#~N;n>^D{0>Y5qN0={L_+(dJSl}HZWqog!RCjwc4(fOI`uGg3
zpR^vP4>!TY-rk3=dgAp7MH4k11y^~;>`1Q1S;ON7lPn5H=c6Zoz;rboI(KUK>7zDp
z0uD9?W^_8xR!56Zs<w|FH|CA?`&DfG<kkiq4vhcuMy=F(>!4?dqGq;|zDx%1ztAnc
zVnGo{!%Sc|nc;KR^4eee$J2q$h=G=olEYN!_OzzcBed~M;Q|N^{5);Y_bkHUtNw5v
z^n%xxA}Ba2$~1c#JdBmM|6MEI=s#?eq{m~Ju%07OC2oa&GMtX%I2!FNy-Q-h5u-Ha
z;Tf3HmSnC20-Y#J9v8Vqn>N=QO%tLkvL!a;E_RK<p~;!j_7#Mn^|G+)XxVHMwFkr!
z4`a;C10Anp*;KzC?0MpND_6%X=i;=sj$Px=-=`6Dp>ge=BeG~}sCJpW7Bi>FBEX=1
z{-CY<#c9KmQtx+qaC&b3#p=IYI_`pO82{0~VzmCU63R%nv`y?0T1&R<71Q#e8db|r
zfGJ*s{nu*jo&EMfkV}?iiG2RXgeYXdH1Y+F!=F{S$F?mrIQ-d@R}u-_)6yGRrhP6u
zrT~~Lyb%mvE*mARP@U7W^ji<oB`fWWJ6iv*&;!FOauEtQ=|a%MUX>5AnSmoU-S~JF
zP)8e{My`g^N?K~jx$NfVXo!n1{2Vuj7#IBdWVMw)SCP*^67%dBpqP04O|n<ByO^vY
zZ=QXlgbrRm!*JseKJ|(7QvVm9{D7B~$!L6v@4C!H)HDvUQyX7ds&pBA@u(AkF`Z*D
z0q{J89DaWOip2~&#(Gqu5{y_CV^##Q(YuxYa_Uu}bs6fR@^Qk>vfUlfAQ5Qs2iUhw
z`k%JF?Y1z*%JitkP^y*WEL{>)B(i(s6#m_84(;z7w3_;pCvooEK$|sJhSNtZMEAPt
z`zWSSSE&8F68F@Sj)Clvk?wm?$(iZ>X-O54%g3fXPPBsIx?_K>=gNX3nrX=;S%+~M
zgwVnj_z%oXDv)^nE<MlUF^k-M9XV(GRT*RXnm2PfQs?8lZD;~)3DE~O$;l~wIl1Y%
zl3Yw#M60(sc6F4zZSh6$li))2_C5`rmnf!$h{yc&rMJwlFSs~J&r_Lyu~f2mXC3v$
z9VHAV28Qj*sxeYx8kv42E5qxc?GB?+rug@B&>-!N{Xk=L2Han<cCN0jWWyYzqS+tQ
zfVyEYQ->&T$7rP{(}dr?e?=VV4jnE1ccrIPX6@2XYh5eBV_sd>-F?UBiu#er-7)c>
zr9@H$$vcWi843V07%-rGl7UT_PVI?^Rja?aQR@=lV(A2;L>u91mcTjV%#R)WSb>yX
zWzKQJawfjoGOg7iV5=yS<ba(Jf8ViuwR2E2zb;P{IAnwEg;9syeQV~d7o)o$x*wRi
zrf{hP%4vXvwH=Evvak{^{VS*0eOv{c;rE|>@5=);PBjY|HzQ_-6!?s;GtZ6t?ZFcM
zq8aW=d2k<7s(WazOa3ZlrWMp4@D{;?ao-CnI54VwE(C{c)`s~R1`x`FBrH4(9T0Q`
zMjPxel=vxiUk0K@%Vd%(!KSD4`PvdCb1}QA>M-i@!a}VLq&naQ)70908}3pS;U^I;
z|3Hy<J2!+f(7U|&YvB+^T8iS>?3XsQlsEdBpzHX!mE*?fXX<hg&=n=U+;Ar;?|_BQ
zso@CT*TyVAcn`1^$SkTQ^LA(l@M%W98aAvu$Qsmp3KGG3Kg)$fePH4|BYnfDc=KXz
z>SK??v3`doE-433m*#*bk~=Jq{&aX9R;B!wiO_U#xUR=kC_vpCm{9kQm?oEIjswW>
zkM=V(!*7UeuC*Kh<huj8j5#x{i5G*~$ux|XmG>mD&4wnSwB8IF+DI}1oj;!T)p$_M
z_)vj0L;8t*5X^CisC=0uwYD%r%)xf5alb~cP~CTf$vj-d&m-M$&y1U&J%fl-5<gb|
zOQGLT@m+D-GG)qUJvuFC&T^D+hWTVRH#c7&sLl1^`xX<xR&H#j&-Yu%bIU|@_jTT9
zFU&a86}9}#5x7yb$70=Wfc7xw&!}C)ImJmVc0~BGYPWX@wc$DLT2VV)w+T3PCK_|f
zy#HZU=DFaNAIoO218LWeB-EB>;GR=HN&M`KqCpImxWaLesi+4<?1yxQt?Gi;?nmOH
zd@+%fpmG&d)UE#1@AS%P(T6^(><2~%eyKodJw~-`<isEvejF0Q^j)Ydh}S4md5n!2
zFi)UfW7Z*zc2lVt4T`<v@0aD)7RytR3B_>O=<aV{lz8>L*G@5AQ=U`DTF#U?dfdln
zrCxXt_ov%hKElyM<rU63dHmL~+c2So1FM#;_&J^Jn6j>!jR$)_X1@NEjBHR_E43c5
zSWMx(cO#3o8)fH^a-LFu6zK@{9+sblo4CXdb1g+ik2~2@()nN<A5C)*S-NxYo7xh!
zGZ`p89)HC&$)IEgFw*&}QfAjyOLvB2I=1?A_J4p{LyxF8<i%vyL1(EM3J4I|H`bT8
zH3A0B=!pEQVD5*ymzU?Gg;S~D&@m(_>-FyIzRr-_1H8;3K|JvoAxt5`(|_jL`5#+D
z1#b68eI-)><{3>uL;(n5ypC&g$`Xs1GEg2&ub6zE-#Ol6IVaORI$6xg)RxCHC2EyR
z;73zLQs%H8<)Cy<q!7$Gnj>8kYG6>gv`(PzMy(c-3$#CSz>~~vfU?}MFyl+D0Rr_e
zhnZcsaoE)krKv(&mY3I;@{;Y8bJ#^#GbC$yASK1mxGqfR(Sdnzjl6qUA7a)Mu49Ot
zxdJsmB4jIGooDw9WTJXqR#588N&bFyeyg{}){}=O7+z%8CLel|!J7;^`jQIfiPUc{
zbXzz*Nt>>LKU~#?><mF<t~E}J=F)3P9yO_bs@CK<Ah-sMR@>*_gI|s@vhXR0y)X!*
zl`#+9g1*HGIHJjLBVrqqQ68!dB?OGG2<G8I&RU(ThU=0OruvL>yo22EseQb+-J-b?
z8#hn8LBd}My%rXpe(Wl9MySNaXEg@cgpEG0_h;~WMs>eYJ*Ro)F=tVckd|4rVc15P
z$Q3vBj6HU)@UYBnht6L1+;iC?7U01hpW`?wg6Y?9@g80DS#ZA6VuKPsi%6y>gJs?T
zo#<wp3cD~njyUTtV%6oSEAD2dKYU7Mxy-(QWeJA8UegI>gyrjnE!4G$bCEF;y&d~I
zUcU-4bC`YwAxOsvymq@P#3-9_<=0>oVpB<mg}y^;=c<5daBLg2WAmj%Ihw8XE-vBE
zNeuMsV-J|iboySKxrQKPG|Ijl$Zz}%D02W7s&#XBW7HlPopRUPYU>EPi9X(CE%UEa
zhhZEL2tWKlBq?(hZn-^mLNR+|wR07;-~DoT;0?69^&*-pqG>^VQcqvywps7)g2+kz
zno((-<FsIupLy`HI`;w+&k!{acl4x_im@kHZs@6T9E7dtT(xu&7?kt#-<tNi=D|zp
z>d6W8NTk1*2(DP9+J%W#=UJ*!Vf1txxhmbve5LcZGJA$4e5XO0Pwy7U9X;xOQ%kB2
zQeId?*Crn4Y%bn-8<0KzBc?@3R09#+$d~QocR@MY2+c5(+BMTPC?xqjlY+O$;n0#-
zU$MEmb-xdz;y6+nD={fJ8if&s`~tZM<%LkD|F{TEz)+$?n1ZVu2MI+xV%OO<KC3N6
z(Rjt#25Ze~!W?Uxx&+z*d|~A=z7ElILsVpbdc&)NFUx<daY3Vx^_M}W)xb2lB#zyr
z_h8NkiL{UQqJ<R^s{{DL;C%n*q^-O*J;UaWsau6*gJI1EtclGoCMvbYdH#~%XV3;q
z<+fLsG&)$$6Vv2_Hj|RK>3osjZx0?sy#&A)U?DBUX;h*<gvLmI{?4lO&;KR@fnMAo
z8^xR=`rD~jAUcyR*&;?s5rAw{eMMSFfk?PuDtM{K#TN(1@&rA7;>t&Ka6jHv!De&U
z`{u1}uiUlEpM8z$nc&MKGA%<(GwM~L{iW6r0q=$}QW1qJoB<=V^!|Cj&8le3HRzVH
z@ux}d+x(Bu?t$GjQNMY_k$Z1dPLD;2{aqH(^Sn+02XA>>xF~C*I?0L7bunVK7QY~p
zZ`xN$+b64GX71=aSb_uF`uaPo@!E|`&G;uD0FR9QuQvlYLYpn}OnZ43YL;6GtM-25
zpB~z6wPRr4aCcP?Fq5Kj-&?TvrWuDXcQ3O~nBI+?B_V|4>wFbJP|hHrBm}pr6ybtT
zjF!?FtS!50guA<+IrFA{yq^XHO)?=Aq`m`xx?lct{|&*AR^j7{52z_OesfZ+9>&nX
zfX_3;xbp=@MOi7OVy8I{fbYr*rB9``vA0*Fn+Lp9=tMfK7`wlFqtM&_U0%^#*XlR7
zJ&f9;*VkoSH#Cm!**=w`?uCUUnzHZ@L`8wY0luJ<q`I9sgG7^{R;+D&{JmuMBx{7e
zE>vlw2JD85FC3!=XhU~273jQ$3l1<WtdHO~>Z@5YNG^9YwwSYY|5^6lvigS={Y21G
zuW%9xp=}@(0t&t*PKwCmy7!%tDewrb8h|X-5XA6L@qz>aq=TSKfZS<Ebvulrt!V}Z
z%9t~V@oX*Lp|PonN^#0y-JZOqWcoNoj595Tt>npmO4Z1%?6^WH6wTrBr(oAuo#bxA
z9L+KnZzUw9OYptWNU(x*K*_rL_o6oP!Ah-Zzhix2qZ@LYINPKmtpGb5xiqo)citup
zMOzhikv)5PVOY3+#?IqF5~~PX_)!jQs4XPr`)~e=F{vkI?RMN6!*!sK5khwjNjgFj
z&&YW*I{Z>Lqn~X(9Z;X^<LFBphxwr?Xc->HEQIIfg2`)yEfXhVKX_CSvMBlrcjt*u
zJFB+Ea205yWPp0wEs5Hela6KNNIN7xBN{6r<E_I~r^lJr`Y<Rsa+EzIRgFtNX;<~G
zZ~G02C3>oF{wG|>W#ii#pYsYIS#18OMW2yzCXmjb>A*khmN@nQZ5RM->0o1rsV=w5
zwFiCL0Fhv`b^K*uY+h$l?bYS89S)C)pQk$ye;XWE#F>FbUTsegVFgIoU0t;{AXWL0
zrFnbUG%xwR!wYtU156t|3C8r@f8L8-)Nfdl&ndLYnCovQ+RIIFMgGdM!ae)c;5Ulw
zB0U98@#k6e0xJYuB4o6?M$k>Iy2|Vf7&og_E<rMR`mK!gZ_v_=e0msyuECZHy?<B=
zJ}^R<I$ZZPeRSf;JkY3jH0di7QS^j<Ak`Myt$xT?k1%h~f8t>TY0_0tt1`Wkjj>uV
zTgD&{xe;fK#EC!4VxnXrE%JPApQ(?`^)B=OnOPYez9BUK>pr3_lEK@3Twx7wha*S?
zKcaOOvo5b@eGi{llxO{^>0DtweIO`HPhR@D6>P4ckcIW~gH}*A9)+k4G`5Ah_uE;7
z?Vh3RX(6-Pa`2NcR$<N?`P`MP9?kXXnReK8kAOZ~vL6a0C_Tc)a$SOb{6TOvq1w{I
z4JDJG{5~R^GpIMF9^%Xgb|#fbecgr#R;c9x;2?K6)t-@trc&vkJI)psegcp1FTo#n
zE~XAsL2uKqU2Hu`vR=axJ-L|Oj|ZilUH4nO3iq$}><kwg%{g<dzcxXwVKw;dh}uV4
zQRJl?u1@oSnDu0WmWj#TjT*KBbJII_Y!9!=LCB9AP9*f{9vh~5a*w@Zhh#GV#BMO4
z1c$8vyR5w(z7lRRn?;KP_<L)A3SP_ufd+QK%9}S$+bq~|lf%7jg;`ZS_*O=$cA&nx
zjL9^ZP~NSXkM8s5Y(bTa>S*W)YIaoXV8r?MBfYE?t<uPOd4(I^zG5?Cx-6=6`+{SI
zv|!2fF^Uv4$B{8@O}|Rax~8p1OxMQ|n;mrWe8HuK@iYZ7pf8~tWk&A@$8|4!V;56t
z@=$AFW;%~Va%S@;p9(hBt~uT*LcRRI6m;wmHIP0w2^YQRJh9@sY*5^V&3?^h=bS#`
z*LioL8FL^n_#4e>g80GRNIJ;6S6_V)hnm>JY$u?HM!u$F$p_a>8Za6EL5=1R)C7NM
z3%qhOXT#$hY1#`%o4wS?1y=nUYRAz>HmF4Z+woYAg8-44hc-9e(Pm=JJP0F^^3!Gi
ztVlf#HHFR-Cg_va<xT%N{RUe8fK!W6<VR>VjJA?puFI5E(|lJ6{keO?eA$wBYqBM2
za&2R*jgPD5ZstH%$Bk@agjPszMuEy&m-R;rt{?uh^wp!08--RnbF;IvA`e%le^Y(i
zY%sao`EExC+!mCX9(uomcyLo)tEP${9d6&^XxtcI(l|d~=xpsYSK0$dsdCk|ZxV9<
zGg|Tm#MWx=!UJ{9X=aAKyzz?Ea!2Y<TRb^GY-@iot@XwR>EN+A%A$ek_6_J1`;Yd)
z&-;<%?ET!(ZT(_9Go-dxkDxH|Bw5>EqmlZPi6M{s@@g{P6&d<t@Uvc{Cl>n6V@TPq
zgzFG_2{exgF-qj9a<orj4s0Lv_**ggtv=uUbn}+WghrtOy(KVW8bnPm)jkWtWu^A9
zIRjo-*^ac8vhxK4Bh$l~j%}jleZVm=hq+b}L`P)z_6H<I=SHCSL7Yc&1EQYZeDX>%
z(D-dTj4rde9w=cFrZ&pS3sY~sKe$Gn$TVVX0M*Z~QNde_6#(1X2w(nI<7%rbLzovu
zWa0{)E5psX@YTmr|LO#_)9dX{DX=`?=1!vDkjX9s<P*%32W?=K@{9(~oc)l>UpheL
z`C?Eah`lm-@L;Di%4?nI*glFmUOn5U*<PmvaFj?VNC-Q9A2HVflPSrF64f6Z^$%pi
z0crLKXgG-uJ&(LT;<@qs36>)38d)OytYw$I_6h$$5EtwSc(w5zP(xW1A_bAk2%+_m
z&H;G={j7$uhSs=%tUzNyFuyMian3|UMu-<h>6A_yrnbLM8QtOSo2RQMZVy}yhCxV)
zjzoLHDPNc)VH*0y2km+{!5}o&1+of6-~t!WLbIvKRsiT?C0WCnFu6Dy2u^%HA&$9_
zUs6)iGp<Y02T}uA6(D6BoD+h`w>_wvPYrc7x6<vjUXH~L{`PI~G0%<$%xo1L9uDYt
zzF1p|WIhMR5lOrhHh1b#X!^OQf^6crWB=~OViEg`=hY+O7hI^(@^rdtjofQtD9|hJ
zZJ(^k@n!V>NGcNB{u|>|+?<3(#>m2)vfz;f=1McCu@zcgSF_xhs`^LyHT-(oyp_@?
zYK5EeHu`eX3Q0vQf4%VQ+2XErN}<=KV!{K(%IY5=+Xvl#XK50mEX&H8atJZn=+NG*
zNx*cF0OXokl1itNffoZ;c2`$^H^j;e^o9>Fkd(p6BqkKKfS_ouM=>{qF+`(Jg&Vdv
z*doi?4^3$+@9)mAmUfxyYjOKs(#pQJ7P03T-j~ui=*n<Ij0SIk;<?Av&Zx+^@36o9
zJ}zn>1RsQ?;k3!8`%`u!y-dh)=mggb(@!#oJRibjrUv8R3+zdI7I3%Za*3VGO-HA4
zrq3}^CzwmK<?*UxxY|_){TPWDVT6oM4g$S5w3_oCYRz$2um387B|KoX@su{ob0FWG
zIi9kfJKApreaJ6okR!&O?eD;pfxbXDT~&nPziGDV&;_X=0;grTaxqfqjDP9`>T241
zRVg#QGA1eq?hvf&3!uJ)@@;PtbPmpH1wHi!zOyi+4*b&-7a<dfewGapy0OreaUCf4
zy}7nD5zI;0xt`#qW4GhLtizo6Ijl0s*fSbHqs{J4PVfC8LvzS09JM<8&*dBhbJkG^
zYcOZ7LnG&<L7?A-#yEL8J#ZMXxt&oh9xIg7Snkv-EMH)wYGURQxF=x0YJyeWf-Ubi
zJ8w1qfO?5%)y=i_es72Z5jh>y<Q^yVSq$P4!BM^{vy0x(h0X=cAwk?l&MjD)KFij1
zn(c^YMbpU&T5TiflAT+D<Z~_a0fSSjOSueX(S!^ACWin~xehs>^yM&6L;ApJ%F(O$
z(TfNfuZ;95G)E>*@c18N4q=cW?3XqSo4fItigFnq`49HDnihxaqjsq^TsQ`kF?o3I
zAX*Bf6p*c5Vx$zYQ9EMdvx*s5m5M8H<BVNdo!+X+k3y5;XQwY01)vNm5;ceD_FV0W
z<B>f&l0sVmSlrlpGwSc4aQ1^A+Dn{Ow9JB0E)Ls3S};Y~e4q*i2MZb!Obq%NW`&oP
zyrtUOrlS&QCr&rZ+=uoMYBm2ij17=(ltlia&w~_w0N0V+Pr!ELcPvCfL+pmBGN_0e
zRv(uLE<)rLx=~7ahY$f_KqaRa2t?%{{W)r7Rzh<O@=T}Tlb9)PMk<5W1bdljDeSZJ
zwW47u;W-=ai3&1-x5G$l9|eKP=SQzNuT1%v_*Oxh-qA&8m@o4TJYu=7j2HnUILZCT
zOPDn8!NClmGD0+!Q@YDKcbZk#CFrh`Pqs0{6Myk_N6LZsVYHd&0bjqtyjKIaOZ;Zh
z*2gm#k2^`wF^=JCe%gKAELzn&;>1bAi96`?ss6?E?qTN91Tct^y~tr>os5PLSrb9m
z4lZ9-D1i3l+Yc<~t0We1)1<LToUdW*z(5U@mef?`F&M!xs*kJ6?61~*+?JwIul9aQ
zWX^)k+7jeyG7T=SS|(3}8tA%@a!&(v__?`OCLOe+3Y;&RQ8Kp8B-G#H;3`u7rWZDM
zWd&w<oc#9(HNE#n+miLRwEhLM<JyotpwiJlFhj`gITdi@Jjz;a@AdQS7?mSi?l9(9
zkg-VC8Qa&4ST?tb3+^AtEucTq00h=kd{{cLK&6AdMqLdoWICk!KJ?apF-uW)zM#Z7
zFSgGZ*;RViYB}~QrR|F>Gq>tgY$MZK_e+=P11yp&flz^W=2<s!FzLQepE32#nSQm<
zY!PW?uf%d`nW3T&T#jThm=3FBGsiQ&Z8&8u;F8o0xu@F<Iw|mEGRIoq7Ub(MV{nvs
zaC#L_G{S%8wOZ3w8*5i>Wa<{^t%C$vF|Sl?;PgEW%fo$VnigW*ICI?aGRYk!m(*|p
z;hyvOvwsO)bPC;JdvcGOv-OeGr$Sn$q`UU*eeQW(3gRJGI^7#CFRcG5<oKlL9sXLe
zD=)XlRu{4Tlnueub$(UN`%HLh*jpOOE(6{4v9Aw0rz@#MaqTV$-EP|(PW)W3V0Cil
z5X-@FXAsrDOpjrQ9m{=`G<bCQNyL&o6OG}NZoVSfy<4ax`ikuyx?kaP3yfuN$Idys
zsOVviJM|eYn*5p^Gd4&FGzSBT<{R(QMzde|T&eclt&{8g_0-~Tze7*_);YBr#(ntV
zU}H88PU@Tu3~~GWJ7RnD?wXW4FBCG~j%YXpCg0iXJjvzDas8b-ykcKwkjuz-=k5Su
zsK}m}<@Y$~C7x}v&hPmk3Pk!UeTN*slEI{1quI~$ZUKA~9^Mam@Jy=HrP1?OuQ+9Y
z@Ew@^P5iE$qD+ZXt7pC!O!Zl-!l<&X@Hf%O?GY=Nc!bI(CN+LvvEp~2%KXH;?SB<3
z7+MP=<MdLR@5U>G52AB&&OX@kJ@_y$A^z-{_(a;MU=dcOVvyjso9D<bpXH(w!K0YH
zMGWYtZ+$;HavIxNJ^X&scX}7a34Jp)7QBdDFGh-SewrzBLuymIoo>o5|2Pi-5qad{
z<iLeH+lx2uCdiMkM&Uc{3uDU;4wZc+URW*{i~C)IbU8~Cjtj;;<{rV|P_B@&&LmVo
z$t$8PN@lk5pW&--uZfr3D&C~Dh`!&<*p)9R<E>g0_mGQF2LMWOvAFos@1e&2PE4*;
zRO6`3*-w6-I`!36Xr088jF%4S@EgA#UOtQivUinL)DDRClwoWCnTt7b&%&3Z2zSLB
zNr4Xt*NoPO2iiDq^VE9&y8D)qVB&JWIc51zvT#gb41h-rM)Nv<L0RmMHp>ic%0AJh
zR9()kW*S+#)*9Ph$Ful$U*HrEj3^SWsJx-q&zrn*0vsgN@xBF*aaBgXHhhT3L7QCG
zfX1OvM%k2OrQ{ds!$X5^7zd-UxP>W+FQa}P_Jxg;tEBLH%Wdy)0rlgnk%9Gb&DKXf
zE>`<p0O9Sy4{?7rY4R=ZrAtm<IBqsxyi~~{?o@UOeYT;gG4hg+95ik`HfnJUIljj9
z10&As%i-mTONnQ|;_G{-x4K3DJnW%o#;|blKIMKD<w<3`G5+1g+Vh?zep7E)o>5FW
z&kVMTsq8J4O%a<fe3L~3tda2)mxcGE^-$%N4}6xRa$E^)VhrZD+!FiLC^RYpUVl71
z)7#bFua{yMqtlQB)elSl8DoWxp`ZNTOQl9txO2P@oiFVZ{yhA~7}~_WlMB<r$AC+z
zD(Sp?$mOW0+~HrBy!FQy@C<p$M~7o^>KlBntGuQRRPsOW_o;#}HFaUxXWi5#=|d6P
z%VuQQ9EDOrBPMPLNAoW;mW*!PwZm6MpYA$kFS;q99^-JDd7w~^skq`et3;k@{Bgp3
z)NpFzTcLfSQP$mul@aUd5oigLqFA1_|BgGoS?j)xrtdEVO7QS}>UeoEiCX!)WDPJ^
z<zo=MhSD+<A-AR-QmZFnzMU>I+TM1nn^ar1JOJ6T;TZ|{9raPqg+{YH`I5{1Ou=5k
zV@~n6SB2VCEhpwozj^hY-Z-yzp|L(q?<%okH|bs9GQJ&ZAD=cZxm71pK`_eFBX#y!
zMFu{tAox3;PdeJ-R5O{fPR`E%v4x&!=x@IQ0#^svDZ(lH;KwMc`%U3QnEUGft~YIU
z(Le13vE*CIr8n&gOT`<?mwR%$)|Hw&v$!cZj8n|B^fJCv0CfA9$}P(-tb(JhV9(sC
zl(~`bgzzNv6sy1n#bAy*yb|_un12m>NLX(gImAu9oHH&IfxSsCe@z^>-EgH+Uy7i=
z^Wmo(Ps?KW{`YbP!L{&2f{6K<GadHRW@CO$7}YZ$`=(a6dyn!!-?+T_MEm=J?L(zf
z>tuhG8D+V1*RH2R*FSLA)i1`JmcRX2@#;gQ>nFSN9r|33O7~#)G)PpKg{m(rUtmqe
zEZke&$oP;I^_DUvm6FRN*~+b^n@eDGBqK0wsD>IC!EFiz?IhV_Rn8cH-#ZSgA9F8`
zuSP6)0i4#l@4;|QNVAvD^hw5-#bCZEGs?rgscvB;H%={Oyv2~QQC5(Xep2?nv|h2&
z$+~{7D}L7`9Md`^9E%_7`u07=S%L31wX8Vl(wluJSTLHS*#Yjbn-(Sr&?7SD%ZQi~
zVgF!Qd_kbFw=CM3o`bBR&zvy~ow%xVh4Y}c+9;;)StSuhqMr0P?9l2`VltNfeQQ6L
zRyJ7t^vl&~D@1kMEek-#(;u<m(dcA8!N3dR8Np3GslPM0B;#*+*xwPC;kTaE51V^&
zzN!eF^_e7BeOAnEkJ_)=Tzz579amN7$i&tp4Tt5ftlG6b?C-<KW^V7|Rgd^+uXiTY
zR7JZ@v;VXG>WV3S3epV;VFAHgn5o{R2Wc4%D%;f>BJUc9&4r#U9h&^~Bd$FpWOi{}
zqpv*4njGX-&^T;3Rm+7bf4n^{Z;(_;Q7zy8EtbXPYVx7r8}AI_$Fi?`e7)@3^W;+d
z<?BYrlLB|KA6!@V6&PG4xBT<q8;(W}hC&6@Ek>Pn>unYFPC#!(hXV+E)E^&SYCDZK
zRJUo?grgn3UOp(oCco#G4PD|KmhhTj4lh}McS#29Rs7#sNtb-b#e@{HsN2J!ea(1A
z_aW7i|7sLVxQZjJR7mH$^|Zlok{f2eZM|qt(6d9Kuwzui=Cln>C3qjIPl>zR9p5{(
zKe?it<Qq$U$AVF@R8DH0o+B~`=MYAwD=u-`o)5VF$&7Tc`rhe$1BoX>&qVm|0m^xO
z%E&B|w_&HoJ>C}9vgk?zyT}I;Ag6Ph)3v{8cl$ZZFZHMh{xV0OtXexx>tHyE)EqSo
zZ057vRsC!`S*n%MP34JSQZ7S1<(?civ1r}p3Pe4ruVsA8snZjETBJAa%Pk$ziCu<~
z74$Wz2g4{^sw+$RViZerejYKKmr?B@8*hER@=E*zBoj@hpClz1m;0kyZI#}!RoOT}
zlk!q_%))>meX&+$u(m3k#TeFBLimTl!uxMK(Z2Pd8;_^C`fiUMXttO`CH96*KLdQe
zQmR`2UaXYeATlYqN%o&PXeRXQ_Egn2nlXT#kpdIR=p*9GxNPMnIF+xfEd%ah_P&Hz
z`*?~H6w%IzKe~yd*>j8Niv@+sS3#L-L4Q3T|LJ$)MO7Cgk!OC1v_CU|b8(Elnj6h_
za9Yo+@~Va0t-y$?Vb8#DSa94<3e+%_b*l#J`~}ef95|2#`p*N6*^93@6&g9-p{gW2
zndsI?=t*J!9atamT7pnCPTu&fn6J(rvn6;)$E<ez&IT=xEzK`{E}>Mv`q<sFfvJbw
z4>6;iq8@U;Vyi7NPcCy_xkSK;F@`Gqu&EkgTh8N0&J;tW8!96;(aaYkoCPoD_*EU3
zBNZA|2N|0WgV`(02rR!MJ2T;^vmc!9wdap1XWkK0_L{$}!{!8&AOI`=czpzR0Z(LC
zfK>DG#i5b^>y#Ny4m5^@xKrhcC<_a_3Xaqo4!kR=eS0&Od<N@GKdg#9k$D*KHfM0K
zaC>nK@3;R$)|-G+*>3;i+bk7%6D1)kQ=yV6*}PGP3@IdIN@bpsjN9<O<)l!iirUJQ
zWQxq=8wtr9iL%);#x{j*w)cPC&(=BL>-T@IbDh(5qP?H{UiVs`^;w^_?##cbVZ9;y
zf((ZqekvAPve@tDv9Q|MyXDM6RqNhwi(THY=m>Of6XuCAvdB{%QHRtJPRS8LFk;H0
z*n!xlp%qnl`-@zE@jcpt+ueG#(88~L(|izZZF#O{sp5sv@{iR9pT=00@y+D7$m=XQ
z{0daRh4%$f%VOIKZ&V#@?J2Dyh-39vpN?HQ@>pxg(jAx@>~VqnG2jmi|BU-V&^rAZ
zCZZ2|PB}$&y(+Qs?_cy#YJ)}*i5Ii0VzX2KhQ-;cM@>DC<`Ut)vu;f?%In%a6#N=j
z8*KsS)pe~xi6w*8!1DtXN<WFsvYlLjbhiG0{PMriS3}lKlT8P@ch1bdV?A)vFMaW{
zKjC`~J9m5=WP9NBFTg)=%ZgRPcKrzUu4fEGV{o+;)9g;C8<Di(*gW{JJM=#=a4JwE
z9Zc(|h>7OS9`t=X92=Iwa6={&caj5>E2MEDlPdS5_IVt4TtD-~KJM#Nb^rOFpR-E<
zamm{UP%2Os4^PM>rh3*W%L(^4pHAby*=%Pjki2GQBal{^SNAR!Se#GR^@<D>zHv}d
z1U;p7L)8e({1KSBH+i!&!h<h%2J-QG{tPe|Du)QVyg~;HgI=$O&$KC0AQ5r9)uWvK
zndzEJ3!Ck>i>kx#K+Wg*8JMQhV^Z7}d&36%Qr&1?i-7HpI{xglQ%>FS;49Spq~&+v
zrcMG*XefM<w_LZXO`K~Be!GcN7~<=lY4>VEv^F?Xv+^X3F+K%rD7nR-ni_}Kvwi|j
zEn*<>hE_dNQfp~J8Cr|;{O=-X5H_$m#Ovb0#1@D8abLvumqq@_=)4X-cF-aJn7b5^
zvxi>b4b`Q@*rfyh(6M@)`San)jZkZoCsDo?7xgDsf?i#Yl${<<2=As&oNh?Tbb4F9
zMl<SQo`<0geP?&zm(=w&({K*HKd)K#E4HuNDG2T=<lZkXze`yWjq8JR{vK@7%GSMD
z=nfbR-cbvqb^{*)3{-ZJ5-o)&YzWBWs4z`&U>V{kePCLV+$eS77CAf~T97hvV|<c;
zpgcPK#~=K6oJ6;H!ai72V?8N(zqj1R8OdhTeEMwCp8>{}Wvht38(2Qm<~(d)#g}^G
z@tFg>*+m9SUmu^RAt1;5S@p}=ACV18nY;jyI`D3D!LK@f6L@(iZ?2Ma3G>=hjoi4k
zM@qe$1J;&)4My18UHPka$COjqT8Ho{=gh8viy_snu>X4sHl0~eXIU`ddcl|c@i5$b
z(#Ii`dgJZ>txnRrLz`-I<0j`Cma&v3;sOIKYryqt>+c}E*!P&&d2zYCHp}k(Ig-nT
zi#oR=jA*pj3yTJsc0bISZ;ztn!qS~yz{zFVsZ`Pt+#9G)v~9a-P5@IeedZeW42-OB
z_Qv`x+SGdyesCSQ%FrU(>eYO&aeV`dQHL2p60JTw6H%BCXdZ?g6F&X=D!B0{6^|;i
zZ$z3cH(%)$BNi)#JDi(w&AXMHO&#o0%Az&<CovPsu8mO(xutXK{gb4ORjR6G2v1*p
z;O_KLKJ(d)0Xjbk@Fqd0YRUvarQJc!nf)@G9aY%JUZ(X-y7`>B%H^id6zN!}T3fA0
zygW8{fACFjq5rUi{H|twyCeC|4m~0ZWl-bC;k4;JopN1u0n@{|w_USgP4r4iqgMOz
zd2Kz4UgmI-*(nTrBu)CJX$+9vg6ExLX2(y5#X(jzztUtxCUBiHMES)L+q~3U#k3!J
zUU+@5yTg7kmHf%W8%K@HT!LPs4}!;oQ#$)C>2LI>A%5-o?(r_HHj&+sbn3yRnZq7F
zD+`YA)zG8kzc&3k^@~Z!e6<=++g-^>J#lTzn9f#>VZOBFu=44pa{pbN>b9p7)Q->p
ze8`9)9UV1mgNoA<Y)I@$>oIp{!8t67&<6=?(Y(AhtG&~~zKJ!7%o%+%FGEw`TG6ru
zd4K)<OrNI@<ql`A(`i%!om{ho7o*uu)7|Mkv2<!{_<Mq$s`M`K1)Q5dhfYd;%0Qm}
ztL)S?Hgx!eXiRD$rOVPS?81-L<^`3QzC?^ettN)_TG&qBdl4XtuUbIwEmX*t(gjj1
zGEL0)n2ROXnQEX)Z!Pt7vC1N(n(u=$%7RV)Y&S=*#&~_-K(9FWsY_Gn&!YC*cUmo^
z^apI#tIjgh(mHQ4)Pa9ZU&rSV6Zu|RnPmI?>+MjGTXW!zboN)(a=UfaoHQccrywg$
zbeQO0lfx1_I99N(j&Q^}5m^tGkAi=!rRfA1(T8i+66p^s_@kNv)s4a>=zDz+;U{6v
z;X0n)_C1lZ*h2Rw(A~JM?}l&5$uv~^Lzn2xECsga^DOKF+kunIX^ReAiumGeR*hhX
zZ*sEwd(b<_oP#4}v))G8T}bN8cLh`n|7Wu%U3TawIE6g7j@4gRW8V1WqMwDwBLK_9
zKFF|@(8@Xc?rTqlRcAkBzE5}B)2g7GL$SOmu2XhIddipT2yrW3+0K7YN-}eRR93F!
zLy_VOf!@I12;<iWWv9SPZ~aPN-nV#}1Lrn~f&2k50{Sn!R~NGbZ1W_2ozO@quOZ0o
zvChZ|<*9u>4pE$RthRWQ{olld)u10iTyX{~Ld`xtBvVw2#YKIC(xgp|z+qzBXcu!V
zPao{%i8Zf1kJccZqaVOEi&G17Go`CYU*Rm&AJ9?B;nBxq=L`N)(a;N!AFlSnYAbl-
zI62PJH{1z1YEQw3bX(iaQqug@byLS%96M!$iTT9a8*oo;SxoGOS-|3v#mnqCw^XPG
zh>K-(j46l>nAfay=6Q9o-jQbeBPaU(Wu*9HEoNGFS)Cgl3E2|W1L??6?xo#NAbS3}
zO@?$TH^eZsNyB28TUN1@8@@@%Wd?Q)i_HsuTp?JT17Sco(3;gh889OJJ+Aa^5%+b}
z!6jyBWMm{R1~E{^etZ7!Y(q=!qc!@m#vB*Tig>fbE2XnHP6$?;z4<&t!41dy-u?NK
zGAi{?j*waEa-e$jORLwGtCpv93wi#VI1@+>>o7L+^9r6g6g&YaM_04SF)7Td<Ihg~
z+7BMW2W#<0JJqMfu#r4v&dZLG+VvH1$$495>J%jV|LJDk&l1;A9R9t1R76Y2#>2}$
z2vk|0duW!mH{lhQJj=q9kiCiCfkxh%l~-_0Y+1W+{&$EnLuo&|{AYr(eY0^dvy#2c
zhqg8QTQ99|oSTq`UiIk<+q7Bpk{Tk!%shp!>%P6DHhxIel)zt&8Drtc!adb!2U-KP
znv8J&+N|o!z*BRFJuz0xLjXI~ngt#OcA$Sl?!~O*ueJDLzd$KA`#j2DUc0&PNQJDW
zy7n0rF4t*mX;smdx)A2|*bR|K7-E|t_o08er#kF;&t|XB6P*>4S|bc%>~n4=)v$Re
zyRkRE+p9~SwjmqaKtdVxKt0dPdaF_lH?CW0Y-I$J?7oq<X(zmlwSfQA8&*#A%{b9K
zzJ134f0`3V!OdsEe#f5jv|a={7SA^UVUm;zd8+k^<*7-Exmz2=zi|=5(sb3zX!W4G
zn~OW8!GE>y5Pi+ueYu}rx!i_<r}DC=Ve!u3`e$9{=3v>fKPuq~)SGG*>((rBrsdpg
z<-x=Ab6q)dfxTpU7NvJ=W!YfNe}zKByPWjzz27+aE4ycY#gO7|lK+Ml%veht_X8|m
z&g#^a*R)vcJb%a>T(Nr0KX3N+^&ne^-8<F=9eAGIj)=H$uv+DQ#Lj3U%~N0zUPs-U
zo(aF%y_u5A+Ao}zT0dVsOJk0Ds>v_(a9-7r>Gi5l!k#F7!IyE&pO}^|;g3b|;`*ye
z6DyX0j?GJ&T0>j&R=p9~amRgb(Vbn6tVX}#ZG7{4bRnE$4<!&R5U7M4_cid~q!I?9
zM$zJ-&H2;I9sn-JrF`U%Y!DL`=#t94MyVaN_%+u#=!5J3c`US%xL%K^*-V`-3BoT`
zTGfpF%Ip#8TUsVYQbZ;08_am2084WUO78kbLbOP;41v3wkVW=A@pRN-X1us}!QZ)D
zJ;`~!ww3b1UB)`G2BfWS<Vr)iTaQ;`Y)%7#u={LsW_0)K<NE9LuVWi4>A{I5lrarX
z=cD%H5IvrG;=aq1*Ot9`<Q3a;+e6|U-#Fg=$giQbMHl<wsVjpPT3TR^a^Co>khB0-
zLmJd+Ez&r?)oI%{F5~Ww!Ur>Z9A}zgu9DwhqJpJHW9vS)<{c15R;9aAlvqY^AxiOK
zF}6zbeq`q8mH|-mwf5kLN9?AHYJBHL7f)Q+jkB}82ir~^&D(Tizz%U(;fqR|7r{0D
zdIHC~GuCI@EIv0phLWT)G%X}TyHpia&{tj3m+_4~O1wmt&0K<9W;olqNDo9fjFzuL
zgIMskxo|-BLsJKJa(QLtLm?@ed+G3)Y6exJ!n4qQrnoEGwoAn0Xvp$N+mAbGU5x<4
zYRV1r+KOaFb9&y|Ajb?($OdFr`;Pc%UID+BmlMu*Uf(*4zMi}KKe#$IK!cR+?u#Wc
z6kY7-U^^Y8W2B7ch>Na;|EjXRNoVWE@wNB!F%_=Gta9n}7SFFgtDb(k_4cZ-?X(O4
z)9RMDL^OF=$Bp}lVffFjihXiQ9H#X1GqaSicy}5pSR9u>jj><6P)ol>1+G8;5F-g6
zDF3{{jucDn8BOmW1m9}gs>swt-lRo?Y|TZVtJVg-kttVcuzS!Ny?MF{IWt1xrVE%^
z*}5_(V7OIbF_lTOCaEi)Anmd26U3Rf`Qq(}R?2foti;7I`EsF+sA~?;YuJ<Wk-<z@
z3N-vI>(mz1Ogx?b){RPAe|B2=&{utt;Aa&vyUR~8n_T^$H6d1om-p$XSmG#<qPTqP
z><%J*t=F-Bc<x%87^@_6rJ<k})9OnwcQF{tTqo*#s5TlnGNHRx=1;dqv4jkI9@SYK
zu^XtgjvDgn(){8$<MW+R<=5kL6qoYa#BLTs-0Qv;>#d82*sPHlHJm(>3!$r&M{1|v
zk;i9p^ST<nsRQNj#|cuOoNe2_5|)q)yMO5Qp~RM9r#(egcU!3i)2s*Se%^sn7k2dr
zr%@Wm%4dz(4v{Bpb$XK4w+bcl6(MmdKoyEqjzsrd78ok_2<>FB9cBUWNn!(jeMh9{
zSpFlq+`6)yTt%(9A*UGnrA#JN<PC;04?jLb>sdK%xz%v$hN{%Vfeupi9-Jggc&Fdd
zmSHcRQ(44Z+N|6lRN^k)CnQw}hRo{LIkOw)8?|dZ<BJOvz1I071t?!`gWQlZtI$?l
z?LwMnol;53*7tvlc+dI7v;QHX#1h}JJdjkvguCfbaN*#9XFXWCExm%uV{M_l6J@Y|
z{GOnGqy4J@{bSww5M-*KxH3N^$nRXvc=%DUf94Fpd1`DAhYWzWg{G#Jo8rDy-0E~@
zSbS_!&^D!#uSz0!lGZn8iR{fes?%e&UiNz_l=Kme*`3w-a$hR@of2z7W4CM@CsVoq
zMDe!A0c={^XTLqA@=)jaaiDe#Gh>+tkuhDmblBZVop9Bdv$k)$a*Mj|HNSH+uL`S;
zgTd4V3O1>7C@VO6^yGfZ`$lI9S&v=V4qVMX`=d)ry|<IyxYMeoCa3$hT;PAb1?SR&
zu-W<%dkE!NbXzq6=MX=oD!*Jp>5p`qdIhIjWbUcmkVH}sEO?n<@6x-l2U1xgTg;*c
zR!$A`ka<iDOPES+mGX{~<5r~IY-*}^m43XWTeAq@AzHFJQ~d7qU_`4bYY;}q!|P6j
zVL|HV<pQ7pd9!c!o&78RYq{y_Vu1Si(u?Z`ZR>%-M*1>p4sl@*!QT-q=|5v#`)o5i
zmc?K-07W&Xm}Oh%(&xI$oS7t}5oNe7`Qv0-=+UyYIteLxm8#&-UVZa|;BoK0RklHG
zxI1p<bxi?%+Bn?l=Rf}>v5IB7PR&WrD7FVrwc=)mpcO(j`Wk%)t~9&Bz96Cz)gA(=
zcSxX=hNA&>f+SHCp#rg@S>$yv2mxX*rw&cc*T<s<!j?Vt*7h_ZUS7Gov>`p;?I|S4
z<l{3N7KX;k31vwY{vq#TSL8bst$nMdsJ0{$mznyOtm&(2!qK+98C1?OpI_xTw+(~1
zwOgVLRag3T?&-BNlU44W_00BnIw6<XY)m=s0L5oKfHz5ZOPKqvfPig?HRPNvq+>W2
zBf%|R=8krM;Vw42AVe#MB*Eg9HkU*I(M7f9E*NpACgE`0`r{9=ao`cRfO&#GG)Ud=
zZ+rC7Q>V8CJmyYbpy>p8$pm~_?Xbenow)uCG|T0e>6CP-I}oRH{e$)avlwg#CKWP~
zt}Z1QviBjai;m7rkc|~b;I9{@Q@=)Qq;dnSf=c~raQT*wPR19n9YM?zZyWHK!UYz-
z?TF1-k>$}IZ*znNH7{B5+E{{8BP^Baz7Gpq@WEP3tEf+5AnnR`$9ij(^G=krop1{>
zgkxAf{C9+cjjAIeUYd~)`~QJkV%o1FZSVeZv`w(g87J!3QFzyJ8)uv@mvALyj1e=c
ze+P*HxnovQ-dvL%auwIVYNBMt6%u4EdNBV1^Fvz!`Dg6AGHV^WvYGmC8Nity|4;4n
zljl&zbWJ;?DSQ#tOmL+4KxgF94*TvfcEj_%T$?0op*anm=9I<H2YusljS^JzVh9HH
z6d-{XDkv=^BPsc1$ku3BW(Tj?cbjnl?2jF4{P0$<1f-ST=D+yAhBjrtVLdBBASNhk
zJcc@&5vqFje5Fv?5xO<L>gaKd^vSENXrGXs`!QI~`YGl<Ag;dsLZc5L1o27Yq1IDz
ziOz{Dmh-8Cn@5A|5A+{ntIoh(=<4jBOxJurwtf>$i7D81ePW#**LWGSlspIoY>Tc?
zUl0y1;}#xcM1uQ(Xsd!BR&aI8VaML}lRKKHQc@#XEEjxsZEdBB>hl5UAU0(iD5g*=
zGHHu^FO!)Qka+AL|LmaH6&@3U(Kucw=yx9Hu^Zc1W!bl1oN9<>(JkMO*o;m1i5}xP
zz_uT_-RPU11PFO@Vq#8jeE~+@#MucftiEI}PB+Uw<kf{U77ja97O9|d#j>4pc<Yk_
z{N~|Ac1`f@k`vPZcAL86N(oYuWhe?OhMz->>YsKpLouIDXFz@}Z&U*1$bG)|akcA@
z5LVh=XLTn7!YAZy$x1nDt=5R9+YRW&E9hwjm-0dI@xdn|H%QN}z4Wy^1OAPjcarAB
zaB(Nc+k^mpMPO67%uAT#7h5PRq0K5gHtM1wZx~e;Psb_8$TjWHN2<qNkN-Kibv+1O
zbuY{O@Q25eZv6obzaC%`{7_MBqWzh$X}&K27aMIs$br#`9!gx;o_t6=lkZ+Kvv~)9
zw}_Oa1>DN+mbSiBB*KQPEc?=n+D7;<$xBbf&Z`?tllfkvSz^4}IFMB^g;b0IW1CHv
z@!dg6#K=p&Ix(I;sDQ$Y3Sc2@IF~u<9d>MmKZ6q`2lJ!D&)D5LQYVwgxp8vku9s@)
z;KED(E$XmwMBQKnG5oPH{Vg$PCVLzr$Vl$Ms&uRm>zP(g$7a$0ZfENVhd`bi6E4_F
zkdLS2D6wrfoXh@PR;!2SR9AyMVb1ywDkn}T3ceqCq^5>w&5ASTMiBDs7GZ)<QZV;K
z9)z5G5jwru7{{~ABbhnAgUAAe0UJ{*<XLa+a>G$cgooxm1wJ}N3E%?wq_XUg!MIP+
zj3SLVNJQp;_JBk3whDU7?GHq}9?nfKC>7$qdv4<4$(PAMtS7JV%tH+jdxRt~gkpR(
z(kFY-y;()?%@J_vtt3;8BmX}S4CICA*6-WS>vN;TmiI7PtR5bIR$TX6SDxZRIxDV3
z2{K3Cq|gk+WApN26sDq`QhUgt!oN$OQ%^^WDY8VvyQqP6dQ8AwA~WMqEb6aD0RP=$
z<lG}*&7I)gDs)0$BcJ%kN3QgIh@9D<SJs*W8hqOD^yj6YJgLt`9YWsdpNoP_UQ!IB
zqs19PfK@utbo?l?<p=}v+44g=8}He#yg@|Fh$GjdU3Rfj<}LM)P=d|_p_p@AdhoUB
zuIs5r#eM_^AcgJ3x2<<WlhQ$_x8KElx8jy}zOL)J)gk%-iTNej9Y%v8JXaWl$K*|7
za$3{|JE^qc(lcbrT3kQzu{Uk~c~Nh|jp|oc){$}6E0sDu3i+x&k2B_e0xo0ZHWZCq
z9&ht^Jx;LTv6&4%KVtJRuUq*atc?H3n$2T37N3{>cQ^Rl=Mdp<9Oj`p1_e!7Oi44{
zdV0fJ)%9GyS9$OW2v@ufG>fT=-{w=(g45;$^n%zNtKxk8H+K0B`gZVbAA}J^MrE(m
zydKYbTXBUJTy;JrqapbS+$>Mja$$}u0eUB0@`;=ma!4ZSnuE+1OP+Jfl@BUzSG?@(
ztNLioqM^<#a%*?V))d`qmm%u5i7VEL!}0hN{(To)LB1sR*o+>Mp(?@Kn?B%E;!ax*
z8hfulTG;N>)<aX9Z0sd@J!&6qY;B8Ic=<i$GWap(GVYTIcb04CH@+j-&}T}@vVqF%
zs|Tjd2H#}OHgkYgibd#lQJ;+KIr^Vm#{4g>K(UWF8+f7^y&gAD)=u?W@rg5;rzVNB
zwmd&#b=vFtszf#xJL!O_i7s<AlWkU@yn>41aa>GCk8m$ZaVHN>vW9OahM8oy+r-0=
z>&${b3ofd&cp-ekmdWFziH9qy#_ePcIQQL1XY^Y8q|YMBIs74ywLfXhuT<)(QdQ2E
z{f~1~GwSM<!XLGKGiY*6Z%b16sqE^`ty)ar#`!A2&SXo#MuP<9T@Mn7a!ltfM$%+a
zU9dI@D>!U2ynS-ETy&VBO)0gx;5d}~M+ouZ${rg52r~fyr$(=qz#=eSStx^{IKq)j
zUS%dsdSw-T6(2j`yT_HYNB+M@wxU<P4b{eu-0G<tb^KQemyfp*_!7e|TdSDX)rF~u
zwlLvJr<tzhZC+ssIhnv7nUOMtiP;7>pD`S@FcGj4eCdze7Vj@8aYDlbHUx|p<E^zr
zEb6b~2fCSKOZK0L9`iG3n*(pE?eZxa0Ci_}Nyqg5b($Od#X=F25|<S<q{QW9dCAl-
zi=Fr?lIct_MzCGkUJ4?P3AxZFk0p~BFDQ)#{_aeb``V=>JetcLQlS^bo<$d>My9F%
zL1~;V-v#b#vkCp&3Z1bWxH!h_+NhR>XZK73tUmQ$32llyi5+Q?$pIRpXatI0ESm}$
zgsf*lOt1jarEsWwe2-oiXUpciIH*n%`u73zrFMQ?$Ui|G{hL%-`yZv*4JjxqZ2jZJ
z`LQ)?L8~O@o9??K4&EvuwFg^`Up-S(lT_CwWv8-EpZC|v<7w87fvbx8z;1J&aSna=
z1;XLt7=PCt5yYGV&Q(tvJl1~;%7IN(AJ3q{HiBfBP+s-L#s5$^Il3*gqW9%Y>#lQV
z@o3UDh4VI>sLuy)l!iirut~EagGUZ6|19#cn|1O$DxE4MJ*S{Lqj4Xgx|%(;60yE!
zjO%YM1bv5(=FPS^945C;{|V^Maq30%=ie!oE6f~>j4*&(Up*%J6phonDSr{)05ATe
z;&#u=PA3QTPwd?f!O68)v)qwjD}!qJ1&B5`$(oJKyiCehvQ}UJ4F2xY@-2R0q&?)j
zLX6^5Tiv=3jvlUD-PCo0RXgpnDly51XN>xQT`&^|Qp|SBU9W9vTX1<wkR+Q!`6#TU
z6EHDyWtHs=I3i*aBJEMx?w5p2(A{>E3A<fQdUfm<+n5y|)R!jqnxlC?^<++_X4Qn8
z4!}9a6L<2rs*{_fYPaCOM<c04_wtFADGL?LR$KP@go809M-HB1GE>*hS>YIXvGeKT
zk!{OciHmxuekCRy{Z(99!tu+2MH#mJ&=No^F?V<G86K{oscjJI-q7Evn{JX-*+kZ7
zDevgNxpXPMIh+Hem-j*G6B!W^8A)HEklv!CBsqrtKw2Gtm>Fm{eA_!otK+IV<~3}C
zOddMz%h@;7oyp-3JC;kn+>fs=>G-;=QY=+PXb-;r-4HAyn@+4WQSuNxV|~M$x%N^L
z#&Hd(eJdeCu}#SoDisp<UqsLRs2SSeF65V?UNO-S2?9@t^B8L9h2t|1QoizQ!;X;3
zmbzyVMXA5iTn`UNSnCI|X62<MpKIJ$H*T^nRxeWi<E_^{Nh<^W?|(jH4%CN6*Yn(m
zIB%Ujc=*EBDT$Z<@xcW81y=dOLlCKBBYN2m_@W#H<SjUuc`ZEradgga>N``2QY^c8
z+@OmD)UOg(xs{>JS-V#!+ze|R6f;?DuVydLm~+Qm2O}#wyM;R%lDp&hMp@F9DZ{*X
z^X17<<A`Xr>jiw&W<7UUk0k@@2}a_I#PI!-{OlnC=$3etwTG3UNp?K6APfYpDqs$_
zEF0JC0w>!nS{y<NrKe3%lLAnFS1N*bwk9>2;}YmXbhs8BB}dyx70y`%QR;T2Gtw%c
zKdVq=tjlkHzV~Osm3^d_$-EBP{(`+`vhjSk)+h2WPMIUc-;V<`OMSFnA5m}8T#f#Q
zpP3B~sOl3%RA2LiYqZzIvDtSS9pn<PE)RjX7blUYsqnsVtqHt<a(r3<6RuBhJWM}a
z4Fl>1zGo4L&^EqMPholZ{9}8tga4jP(P4#|Zi-5D7n{_ybBY?S(h1{h3g`qt?Sg&h
zA!SorQhxExMN!6xk*i5GI@emKw*h+mytF{x(G_Z5NHS&0KLiWAt6)q*XZ=Xlv?#SB
z-#Tj=iK=XRxlA_BQ8%L*Y~uWW-w-d|@Yq!A_RsMgU=7Uo&w0!izy5@F(hJ{x@_(4H
zVm)atBI4qtHO!i+<=<pDB7Z*HRQ!FY_`fLIWj6*z?5;}JuF!RV+=9ipzg}pQLrvd#
z2oJp4@)xww{BrI30~ZvZ`sx$7xi!C&&*ZIQ9ZS<3e`SV#aOrBrhJCeJZ^u<;p*uTP
z9HyjionyFErk%gRJ<()l42l@Jr$BME@EI|E!g(xlgtJy{r7P0DrZ^YN@kHJAqgliQ
z)UF~(m!<@xHM!34gD=9Kd}~Jjo%6A-DZ2swp;La$ab189BfX;XUw`D0QT8+Kb+emo
zS=EJ$xgCdu-G8TVel^I6B7Ux$!#&`}xreEL;QIMlOW?KrDko`Wkn}DzmG>Nib^9dN
zEVA^t`r0{p7r~#>lKjD{I)R9_n8<U}Ur{r#PLt9p`#HcC;)O|dUBgPdjW8CM_;vN~
zUO3*MA94gL+1$v>oJk1sDdkH4JWjX3U*AU7w&5E0QoE2)1H_D6CN-fxZa{nLV2Pnv
zv&x>0*R&&GbGPj>rCT4~yFp3kAFed3ksAH*w%MciPqK8=G{rTCrO^C1_Ys@7Ez7ID
zBG_p+2$?NDwjqh@&Q*A{tJ+zes07I=JjOQe0Xr=1#Br;3pi>bSgL5SQqO$H8h?IqY
zt*2ZGzxZRJo#qN|OWURIH0X^N#@l$(k_r{tH3s?45ZQ$jQ!94js-00v9&9n6l*2kS
z@syFdt@=pVjv^=5efz&CI0CAf^DDYYK{HlDnDJxrcx_Ts+e%CRVw8fvR?%Mc3WW_$
z&1P&@aX5E92HZV%u#t0MBD7fQgOcx_yt9un6Of*28-EJR_|clJBoSlXPFGZtZ&n07
zTP<M0HP(me1b0J4BYS`I+l#F-D1YdFUb1H<fmxU-<z9PAE8iolVv26o>NIv$-}Z?o
zVGl!BV;3yp$}&;l9hXo&&weg=C^)Tlh%bearfN7$eaC5tehNmDM>*)~zEBVZBsTaK
zQa-EuUgj)9Z{}6JA;n!7(DvLjIGzm$H%5Ov)nJ{K!6I!kGt_KxX>6s*uCr~tWC=sa
z**ngkN|JtCvd3ZVCZmWN5qTzczepzkPcOga0Kc1dsRzB;&S!S*^8FvDma{PB!01E3
zOjyGGc*OTUu{+O!YaGc%&|3L``*+tEnoXY@8*b^6W5`utdq;AIO<~Y$hZ%4U9l)jW
zCM{m(I&WxAT=1Q4gv1ziNkCVf3%dwQtt=|qQAc@E66c>>AvlFD>I4Ep2yglfBVD?0
zV`W6Z_a65cs8B5bSA*PLpS#%>5WPMeqHax7XVz@)i4S{7-6*}_NDPV0?9PNMH|#Ss
zdQEdh$waT~p-q(j0n^-&2_IwS68L%eXtbp^c}gE3g9{0oFJ{T?uyMM$6Jur9ij=Cj
zDC384h;!}0wUXLX&TBU1<J(six2hSk#GtF^I_Ygmx?OV}qdwr7wOw)-S!xy2`>!nM
zW0?x2e~8eG$c21^-z-Z1Bb1_3UzY~|8aO%{i(8V_Y_9MrtPy3CGD7g{`^Kg<GbUnN
zoY(t9vf>pmLBF`C<=K{VOB~KU7BTbia%6escfaY%Y!^s&hCyUQw)ib{Loaf}_Vlq~
zHC0g}CUQjifQBE4^OVY;T!@lo!$nyO#-KSqybb;1$<(yzPQLsn?CD&3x7XwQx-h!j
z&lY?I@$@Z1O^(99%3GDu7<$q%CJ(r~19VJ1tmJ%K1`SMEf8&N_;;uTOO&BRQoXc)E
z=Z9wc%c<W}23|Gbmwiys_fbBaGf0r-5JsElb!h<~3^XQaU!Y<O;0Zk;ml>9-b^w0)
zX+=atXG3{>Y2vBJuFJHWd%gtf$Iq``N1lDkY3i&lEC8I&(hlCRr-rBD?oSV%K1B})
zrPJd1hu}Q^fz$+cclzc7lflJ$m8)<b^)>Hk)qE*sw?o+~%UNlwy5Tf^-mf-k7brI)
zA|Mt<$oKCISw--n3&Dp96#JSB?djt~gP3>i`7eaHRmC@h1RG>p#lXtEQvwugn-8!Z
z$Yj{z$R^CZ5dj*kRS)NOBo|BoJ$3?&`e~zTN{PwCls;2b?J2&+5$O{->s}m)cYirM
zQG1G<L!9o6VlJ@F-j_E$yE~c~%*R^*u)Dlvb&~2&96{oSy|;R*)Wc)Hxg!z+_#=3h
z9CQ1aV@`7E7=}J6V~*`uv=j7DMao!Sc#kKmD$KJw#LZ)KAlU9HTlOO|$3vQhRRxpm
zb}#6I2R^ZsiTa~(J#kSz?+1S^jRW6>TPuQtdEtZ&ha~+4=k-b~SjNSR4)+3aAf;pR
zduENtPsW>d<qPF}&IW6}CB!RWdH0|t$ALT(fC`mca%G#QfAq!ePbt4z^m7v65lwbF
z4;sZls_=Hj@IQUgxLTTjPi?#v%d8Ybn~Rz4axYcd7ff<AcUQ>ZgYraFezZTKgAYgK
zO2i|8ES%wILc6x71>)OBgHRUkQq58P<>{8%&|Pq9C~c9`u}F?gG6P6q`fSIlfrJPn
zYGnM58nsEFop_hMc$Xp|r{2T*4@)hNpLNU|^{`c+!nPEk|HLqnjIS$NWq1~Kv7ljL
zs>k15@d|$R2S2U%rfyiE`^$qJ*FU~{GajExt#E<8+x7qKT{v>0%KO}RCnvM-8x?$H
z^zC3G;T6M?_Is2%ZN}W8r&c{`gsGp(qo+~K9F51IISdq9*{0?qA;)>)MNO2P1zjqi
z$~xe<8!)Q!K?T>86UJ3PuHZsuhPH2I4tBG$3Ztz5KvyZROp4s?$(8`tBxS+`pQ|1=
z`2=Jc5fPB!-0hx!p6-YGB+#?Uh>Cs&kb570VItFb{e9;k^=ETrfYS~e`f-O%vj#Tm
zWP+_4o#XP^dxD1yjD6-@#pk(uJ(MS=XKCLj$;qcTx>t~?mgQAd7sc={gs%f*2eJ_e
z-!-j7?@s;@Yw#H$HAotyb>P|ophk9Yk!CBE3yUKge(n{-F0M_gdp|Y`?Q?u*pk07J
z8_4w^#UMhO8*(hzV5svAAa?LZNBZEqKo*805wTEt*Y3SLUMBalRZu*f&V|3#!+Lej
z6=BjN#UPHAjRP4H#PB#?k`^J;uT&xHT%4Ky(xLP5p<l7vVL4qpYSprV3UjP2`?->P
z{j^J<FO3!xLy{xk{{cH?t;J)vpl8R>8_)0col5)F1D+!i*lKvz6wF<jx-+aoo3bb}
zTZIjJSSai)dJ`kKAvCRxvu?5^zT6sAaN;jGt(pMCE;F)zs4St`5&<(lSn^`M6HHl#
z5*sQG#0@0IJG$g-*6g<n@!(4K>i8+OpD+62Zrnb06@4N+WX4OGygY~s557BgXE#z3
z`ui+TC7zHqclP9(&soc|YFiAlc-z98EW}8pv>wCMbaW(b2x^fL*A?-lSV%m1!#3bK
z`(BZX2K&ALprS?pSGU~}6A%=8Vr1l1g6QcXCgIwlZ(t+G=5$k5CqXQrU4J#jQ_y2e
zttZA><*<~Z9;1>@Ad@7FB3&WrM9eMrPR@<4wY-nP3Mm$z>jOOm^Vh`6>QB}?aH}Ie
zn$1}?AtkXZTTSQ}QE+`^j=db@*@;-CsQI~Rkt$KaB->jB>nBs%bjupN4L<~-$E#;6
z2fq2AzdBia>a#Up4k^22Yk9R?Sgh@3wcwSrTJ`q7TE}FkOFG8letNUvx(SYTb{92n
zq0)f}-A#7^n~-BCM^aS&7IF+XbuPDd2$-oT%-uwA_UdYUOSfB8SVB@~udX5<lP<6E
zladw!M{tX0`acGb1Gwb?XbR+H_un^2#Xw8TP~t+x<J9R0<=a$B{?{x`$5*S6Zqp~b
zZ811ag$*P0$J{%7f4}PAj8}EoGhw+aAvN9an6%jEDZ!LVC!+(gx>Msdc}I*=e{#3H
zdUonz!byjRr{ZU=0h1eqeIQ@>f1uknbV>>=L?7{Qag+IZA>_#F{{eAVAw;ZPrqhe-
znxWzY(c|2|KanV>-iSbF0^uSF5t1Mmdn4N-l2{&dj_~K_)-AQq86MW#yv`LWZzD||
z_LhE65QAkWU;-}kZ$=7&kd6SwCo$UFoDHPI9Sy8f?x)9nkD~@i$?8V$t_`;%q=Xe1
z;!K7u%9w-Ar(atV<V>!7cTP(gwkdn@$wr-+!?I_q^~02LJ=YThc~I`9lx0iBj?FLT
zL_{FLu63V#X2-5!6du*fcf}?1vu3b`^_b<RZn}95bblzt?sW2bT!C`g{zvOyOfBGS
zmhx0^pD|=gb=-EHOOQ>OFDvpG+dr9Y$5*HE2dQ!(gb%;R#C?)0j$x!zPNg|;1^lg<
zX>W5tzZ3sJbbk+%Fls4ja~EOreofZMd_9@_Y$z8hsJ=&aRH!?IeiprcY5{ujhzL~s
z@17TVbZujIC!R2N*RdTyavtmFgnFliL2-F=s<n1juc1FpI-2{&Un8pWe$;cJUn^!8
z8z>1$0D&bcDr=mmPZ2JM>OZ0q7Br_)hW71Q-^q#jwUWj1T;}k0+^yZ{-Ma38wC<q&
zQBJ=8f79D4=9j^0{mY!C3l%;n2jSD18gUbepd_zuAFWmbCpKEw*x+fI{(!NM{|zK@
zWO&uwQ@jvGsgu_leC}oEft0)Uxz4#{S-4O^9F5JB{F5-#gT<7A)#HECH<<VQ<2>FN
zF2|2s02iT!dTB3B*^8N>&~*Q0Wnu295rx;4(%;v~G0bHcyu5Zo+8jp>vRKdHG?b0h
zyLRHAd5bB%W_#83Am^&}JFv}Muw#(Ik>eM{I{E?%5{X8Ftgdwj(6W-Oxt>(|Ju~Xx
zhjX`w3_a_(thEHQ^3Cu%yKM!#LXyM6yluPWlIDZr*#&K(u~kb;=xT|4?GGrCK}#9Q
z)WedIQ$r37``i9IymA|0fWCoR7aL8ck(0n>b>iLJ0J4L|g%thpGJ~<q0obg~=!Yn+
z4lan5nZXIY`Eo73c&yy-iJsXR`ao{`+@le>*O^&Wqzz3rDY4n<kndjj?la-18g%l2
zz&n-tpyJW=aX(d5ILr3`%v|Yj71ckam$Ng-Z21mW*nYXETe1jaBCP0Zs9@$LQ0O_^
z2<?ja=yw6t<O|tU5tz~qLKztt@wS3dFug4#7?=}tuAr;bV&)P5M7+a~%YsWTKScj_
zTTEqiPeglldF@X9EMO!CH_k|~GowA+g<idWG5pI{^OjTq3tnBE8q4YZP4iU`bC0@4
z)?Nvz(Cu39xY<MOSUupAlYtw?sJZPpGNY)`;Ms_=lf-sL1eor+Q-hJY`J1t<Xb*PH
z(}4DrlQJIw{~#|DD}lhJ;VI@T9exp5F#oDqTxih$CdJ|AUbfqrxM48p*HE6`fx9y@
zq86&Ku5Vk{qZ`DLRb%k`ZqAuvhCmvJe`Q(p(d;(h#K*9t`k|j3KnV+UF0TuRV1I?b
z#M@ViCx^fJ$Bl)mlYJq}_uT_zRu2-g7wFY}KsrdI=}$e)oAh^78O(n>Yry3KQ!{-J
zF=Tq$Ic9}qbm-!0TQ#l5UggwY!MR-qyQAdFg6zOIE;kR9trqnoE64wyUyrYDYwqma
zqVbTqw3`RBGq5=L<%*%w`>P6wtt69K-%m-)tLQSll`8cE_D8r<1Zpduzh+8!AqY|h
z2nAVkdpO_-ujq;|7B)WDsq5W1lB#W#MeIoG2EEIYql~Bqcck>48PAr<$ks-w$X^-=
ze!<h<inU?e$(^U|Gnr7dVLy{8N<P_X_~6#C@mIU)K;&XnrXV=(l-i6t{lkZ;2`kbL
z@4^LoKOaVf++VeDtujQnj3i)EW=^`XpK%LFxy+^am5pQOW;=D^#`+2CXr-T}G1YA*
zcx^$eEgCAE{oi?z9r+b+dIW~gO@^$Z{&J&h<x|M#a&CyjmO^VP?7Z5ts;?=2@@oDp
zSj&mEL@|#Um}_P=3KlWaR@d`eeQ-Ee-O9F0UW7R$;HGyF9Tlpygq9R?a}bt~h(I(?
z#O%k7#1gk~haaYK;1=GgJ~Gpo<jMYWH(^pw{C>9@OwCOg>qIHeknGjD%EmO<ac}c;
z?5LkS&&@ae+gd+J<^dD6tV37j;@-*0bhaB(+m>NbdW5eK%pHFWp`kB-F)fk4dX*_O
zqft2hpWNVDa6|U~#tGQTt^I1~_8<9M+wCmdsY`rm6Sw*+Ovu_?CbE_Uw%P)nI!lx`
zs}50xIbI;{W9=LR^9cgx%a-#pud7CsNbGx1pAjQ_pzi&u&jz-<0VJxUO^s0MzQdyw
z+RU(~fi~0~nh<mp<G=;FCT#-`XzU7^M5$~<ghIRwn^cnh!c`Np=J-jSpj(j3|CiEE
zAz>L@zPU8v50!<J(u6AEf^T{`-|sk@8i<qh1|C1iAcK?{-{c95W@8xWqGhCZs<4S<
z{_Tl6Zt~o~#PBb63hg@E$5icj>55=|GQVL)=#t0rx#hwt!Fwjzo%|b#D$TF-U9}T;
zZO`|Gl=t&hak;zSBFeRKi^mpdqZ0IZ-Lq6-LO8V|e9)N=A%>yjg<9Y_^cwBGtbc-6
zx+BNt6NOPnW!aa=z}GD_>3Ukv{BT#&e0N9o#3w9D2lJ+_723Ln#dpjxmtdV1h4v?a
z7?gC&y9#L##6w!x7jk^@N;LT+RO3!*A)defs`_8Q2SxU0{PxWr0@)V4ImQ0Tr~LM0
z!e%_2{Ff!Lt7ep9arcYNomes&2mii~!;D_U03>D<{MtK!n5cXxHZVJDOL(@bg2REV
zVWdNmZ66sfZIHX6A3y+|5VSh?WStjQgKkvo*JkGCj8*DRzN3a9Qa5XQdfiprHr@y2
zQN^9E3ArNJ1d%0SX35AenH~UH`3qEnm2bk*fPx7(8L-)|;<vbCWyA~$8#Ag5<1m=n
z4l<V|SYqJls+Z)Ub8~ZJbP39<ewE%LS_c*Q)u`bAaYfg~zmpS3SO8-_JX}l&k8%9~
z#m9j25fyuZ&d6`nHngA7C0-<X@veD|>?x5v{>IPtR#`hzo27}bPsjU2Uq1hByaVN!
z7pSP$#P=9){3YJ|i?fR!^QdXgxG(<m&qxVO*$a9C4qQIjM&Nqj+^Ay``zA6;_*IcS
zAU&$sUADzC=YegQ>jGpGub2J@?{g*L5c71V1Q6+!-Y4LF+|y-W@M&$*C`${zGhY?M
zzDMLZ;DnxInEr(^BGG0j_cn1B6{-opd@0k74D!kvjpJBf;Atp-*$HHiu^Fu9#WA<A
zhiqllkKuHJ<gj;PJTWY;45~;Selwn6SlG-ISR10|`0*Yy`nLkf44)lj3UAW#MsivE
ziy?Umc1B>mo;r=tKZNNZ2_w`j9(6#z>~#$}_I)|q>bHW^BmZZ93?!Efbt`mFuGTbx
z>#3atIjv*;0)Ks#?9wA8ammNMPuhsb#DGw*S;-_p2OI_^Q>-w4Wcw0Zt*3tIqh2>(
z8Bdrc%Cunl`Dyv`vwz*s>ag6U>f%VzfH<pG@|D%gR@8itRYu{2fyz|}8BV#s$^5Rs
zRSTE_lFj!hkP&(S_UMWGn%+4XbHw*7?1ses*3|;|q{}j()SdB1nFTK+Td!v6JM&~P
zrg&u<1qkqkGmMcRqOY$u{Cv|ns4zR5FXU`IJsysE=uT~?J#E?T(qi+{H8oIo`)F;j
z>qdph39JnxxCjwmv;~-%0HelhTG2A}?yR&lM0G0O&xlxN#}zswSIx_t19I#;unwoN
zw5J{`&H5ffUtxEaW$p5bkU>Rg$>HU_f9kT#<rX%W#hLKssXumlmW5SrGRHAZH3Etf
z4q8=WCNxh$H{PsV@D5aO`*jeLV8|ljROcOSVjUWc^W25CXk@R>V^jwRk}v|5r0(#r
zJz<IILXN~GERMRn^0EDst_-8StwHE(z{B0fD5mzSpBF|=##-q~r_N24G^a(l9ACb2
z!hY@pbL{0kA=U}CLuqd&Ad4iK#??Wp>R3O;5*rI$^Nv%aAb2^$HG<~m-^=%qNu00D
z3|(*xxDwDia~n+dKT#d%gZ>y@#6?+-A3>qjb!`<4fcu<a?Qw)AdCRpnBUDqcYi3Jr
zFUB^!wIjYk?Le`a@Pm*Up%h+?|ItUD?T?bp{i_kpUANk>(n4H+^^?PY#&c?160!9c
zPROn>+XR}inlbX7%kd*S(t^Y7R~}ctsVM)ph^D(iH3uO`NcZ2N4h8rhW0{M$Wv*cr
z-uUmqZ?|Ro8wo-^n(yo~1vKU}A70F(1b$Zm;?*EaP_(Ij7Imn0<t?<ZC7iv*kU}D*
zMi~g+ZXI`YKepA(X!DO!TgO^YL6>Ca?2ftMD*-)m5R6HwgAB$%GWJE!^`r8Kd#mN(
z1*7!G$rpTo39pX^kB=Atej!EGCaj}kBo8nTP&3Kb$D~v+&q5N~euv|Nc~HH)zhz)?
zt`M44ZG}oOOnNt@0SG2_avr`7fN0~?h+6oKqzaTqFv#TYdCdCY-B*neUs#))6MiWb
zA>y(~6^$G*!JJv$!>_z-nd8hUT@6&16Bj3*4!V4^K7q}hWI&SKksZvjUH|QH6B#b0
zLnF7zA0aI9{MB#@KwN`6UJ|>o(qUsMXhr1b&!~GZrL(KqL7li86lv&jPy4BO_YtHk
zt82c<>Prxj@o(b;5SDrVg>wk$LDAK@t|=Y$?#Ex_MgQvWBpzdj!;UQjl3m@urxCSi
zy(36{-K-v1vs7ie))!@i1(JMDv&vvOI3z+9U#=yL>`to=hoAgt-J@<{G#2npq#MkT
zcabc?iG87s_T>;)OOAo+Eq|?3QMY?NHB(z=WITMJSadkx>!DL|rZmK7dI@;@c8yVI
zE6~i^ZdNyumz}Ee{#51rW-@p9Dy^5#-94gq6FNkQgTX!jhjiq|*;TZs^cLYh*?jQr
z-J)X{unJ+ex7cG_ah6&3H#x*(;Y168p19Y|s?vuz<w%e!XzUG=`#C6?Hr=Zbvy@+v
z#54sh@@?jY5ty-mvfi2p{!x`jy_sPpj9!|7z^_g`qud@ftOrDH17&x7x%`6pQlV{5
z^dcUJtTNCW(<1ZkiVkP<pyK=A3V1PUDn#(SOvxSVoFv01pHLSsZ<2;X(1ubrACCbs
z7k#W1#LVcfSxe{fhL`U<@8d|ASycX!7PwEd?D$bMl0G8#eK5OZUOVMy7YBcSRd8=j
zAR+Fcyh@fDM%ldFv&#A9_e2cG1SU<a8V!eXsGx7RN_pLc<8xa<<z!(xW6%xr!*e0C
z1!{7}+?`IH%Usxo9ntMQ=Q#5tNVhAg-A=EIh+Ep56+PHpBiFuI00uOT<`$sLdFC9F
zYSqU<^i$?SXE-$i(F)KJaBK*sO30^^2jNXmEWKG;r%jYJtkgFHUHhN~%do3A85@#0
zpdGS7p(p|)f<K;9{z?m1GCecb!zv~6;rxd1Liu{a4KQp{A|Go7w??%MhiInVP>=tv
zI2n!Q;1e=poePQ@e(S1BHnj_TY;C`#<9_iyQbUvkNju0k{cY5EK*EcqC?KjUyUrGt
znC9W1X=|ZzE<?fGz)XYLg7O3T*6xE5J?4OKCtcI){u%p~?z-aVzvl61WX?|<BpT>3
zcFWRt>OJP;-d@{;xLlYk;13h3G?sgXt^~d&i)3vEj!okdD#+9IL6Yzk%E55N_9{bd
zG3p;|go40mj#+@RZOH0_C=Rv*B8aDXt#;@WnLa^`6#WNLh<B}^cmlswwR8qlSi=uj
zzP79V{e_S}?Y-~jA?l8d>c79(z9cYCI<EON=4X2JG#jNji;<^sr?+JgB-jpbHm;Tk
z5Kc*c?N&X7MsEejP5aEk!bgjniD0!Vt2~NTe9ouc(k)=ZE!Nt*{x|QY*^~a%@LDuE
z6PLf$k%@${f>-xit@wQd(cAgyR=c4h^JfdyCMifa3Xhfx`_8^a4H$!uBi;Ej8dGg#
zLaZ;~P1zz;!B=_&bvwhh&!~c*@tak^%G<KQJpbsF^ETI%h3|~IDXe?m-C>T6HdqOQ
zLA21&7df|tUv>4gseuV1J`)#J-4mJPU~rZg=>E8H)iP9(w6ZN(Jaub-S4d^N(dO&;
zba(I{s8%w*5F17-x?evA)1`VNDC8aDyGnMjRtWA$O=FcL7oOZ?Y*Ql)<0cN3ci{=B
z|D%<1^pqptw*)lRsNq2EG7PvbkLp~ZCt5pnL!S9-mRbjej?q#wW4MgrXp;bnMv^6b
z%2r=PryMU?*k2MB3cHw7%+bgoa)oZc*zLIe!!6tPkft}*wk&pd<ZZRI#`Vxt5Yn+>
zQ4Q$8_+28l$wVNoS4^I=QFyG7$7+%Sb&q7~9-Wq8JAfLK!(qWrS|~tFA9h63-qyir
zd>*41wxc3}ISy(HgTSYgY+BL81-#1%HW`ekQ}o_%LSV+rQf1O;@4htVr#iYYmZ_HK
z<?F9r0+5><`%9%Lvh804w0>jJ;r!Ro-1rkbYlohqs$S)$l@ki_UsYe^mi67NIX+wX
z(5>rh&gZe4Txy1)y;7on<ut#Fh<|z#?S_b6vrVdA@@$eKIm5F?oe>eIqiLCRiN<vv
zb|TGHODk`&1ZtVGI6Hhgn-w_IM->U?C=aZ^YMlSK-!@)P|CL0-J*+FY)w;#i%4)64
z_B7+L=3>%@kdvl2ytPTbO6KK;rZ;{t)!s~Zn|rakwiNc)xsuuP<K{$8k9@epbaDLM
zfbs~cW_rZd9z-}7F|`8%Ng-#V!Uyq~fH?OV-{>#fO36=yc^^Gh1Tpx}-*s4+R$Q8U
z7G$<uAlH8~8$Y7U*HKfN&8Kw&$Qo$0!j>-NGKOvXGeO5;WQppN#~2<*J9&+Y&WqL=
zllh&C_Ipw){cVcaAwg1nF`;ZqJ709fT-eb;c7Bz(mKhQ8V+J4ere!q;E9G1L9ysAH
zy|v$fWw_(dQc}ApsVL>3>knm9X&&LB+#j-7oB+UbBI|hG=j$YjvOOHSp}1<Y3H_8x
z`DfqLAxAuZZsdZzug{qwc}*dKHxLS|jiPKR#ZY|5h&gLV4E-hbpS!gkbKC=JG&INQ
zALP~xWB?m?UvqdWQ6L9_{E&TzEjk5d&J6l8<fPBT3Rs4aH<s&Y{Ml=@&?Y&`czS~A
zRA-^R!qwqedj<3Ks|sCvvy1M29frv&Udx$ITF^~MlWbh~RedTahRC*c-OYvI{tj*=
zmwW}n`?m2ygS5!mE&KkHzYWHg!_$8uopM^cq;DNM`vJ*)7`y7>VX>lgbHC4zOxm`P
zlUURJ<JzG&SmpkxrV^yPXMQV*^JM@~2e(I{yO;@B@>>C~bAM`t`kQzg_v&t6u7XU}
zA0$(iBDjh0q`7@(K9?sckj1D+({STo;BH6kAu;Y9%Pa7=(Yh6XEaeszbEYrSsFPK0
zhhLltnZ1;)fCzE1wcsv9ry-FBOl4JNKgCpQmR9u?Q?Nx)ZH(g_<RtfWBHBUg@XLKL
zr;_r;=@>7de0WHO3Lf%X#R4s*Bvq^hSVE4Jeyj0xVTn8!;<eJp(e3*`e*u8r=nNqL
zkW_i_pxF`)3oE)XkO(%|EL$xSNf}`>h(Z4$QYMB_{O45lHG757^vH?AFVVm<vTvxi
zlk6PauAS(VLv>QyDn_8gRi4nVg0JI-jULic>x)ikc*A~N<Q`wj+6^!qF=<NAL`gI8
zYwF1Dixr!tB>mKc32}q8<Ns%fp)AN5(=>lI*tivV;C&5v8~&MPltU!>ADt$VLjole
z5p6;`R=H_)lJWAy8`jrB9WgaX*6-DT-R|aawDVl)(+j`HV?Evux2#Q6pl0Ru%P6*@
zq&&tU2r5bW&KNGUnUd}<M(NKDan$oCqR#i|qZ^tQlwPpOcxt0Sch?m<Cu{ej6fVLZ
zf8a%TbTD;vtyE!AjC$-5XfLtHoq^uBW{);VzBlDEmMznpym0D@!&m8x!7@a?oZ5I@
ztb^%p#jDq7CSLnX6|Ud1xLhCAJ*iGiqLD@Uhu9<+ffk>7Vsi|>Ey_bvdCzV+f@;2u
zkCC(HD9@wd;j0NlZpMjQO#}A}bWvn<`f;m}Bl*(|X&Y3Avj89&J<DAc2gxszzhz(o
zDbK*02cZvv>&@TRdYZjZ?wILcpJu;hp8P}o0-RvweU29|T8!$&S&#c-yi$Z@2H!>n
z!WZ*jqHwBeQ2s({Acf+{y%#x5&1vO33oMpn`TK*6$M{pvc1+*j$LM47A`yO%ndoLd
zjFbEu__m1dEdvI1&1E^!Uu|SX^);LSFTLm{XT7KbVxWp>f*hNEMsCXGO;1?+<#+Vv
zCRN|RvHI^OV|JLlH1Z-D1Ce|fi2UD9Vsp~cYp*o#e@Z}_58}G_uN5`;598IcYGhmE
zbu^pdZyr|^2Z8wX4sTMK05oAU7WdW({c8Br!^VaW>DU%F7lOt7C&J9)#mA^~!)HG;
z2R;mob3Wys(9{Ax{HyrTTCm&iUKih-j<LW8h`bBdx!9e+UyhNLPi*2rWAo=5zL#58
zp+Oo(HU5wtr#7;yVxKsZfI>J_U)C~A1dO>7zG!pQ?&VE6sIg131PL<uldUd>p=5i8
z(NQ^AcruYW_V@<lR?7qE;O#B5*Y5(C<amGA2?`FOZPZd)-HtqqeFw2ITHjmF#P-UQ
zfC7<o(4>JLyOM*;u&WoaNu&FGci<1P&6y>@Kap4=X!C^ctXF)dc99j=#Qtc#l_1VD
z!S@+WqX6<4-At1U?w`Ez0%&?lz*&@~=ee~M`=SvRyh)<NMYUU+T?;NunEnNwIB&Hu
z(+yFIV`uK`@<~=mIh)9jyK|6wEsd!GY&^Ci!+KlFmN+!%3hN9deSx?kVbnz?R%?PY
zPr+bcKv}k(C8O-f9Q)0PVYdU7w6<VNRvB|3v+u6nJgVV`hN{GYkfT&2OYR36l_d60
zTt!8h%A!3!Ly%#{wQiVrwG`;g$cWDC_!n=b(SS-(v5PRxFpoJQs(xj0_K3!^(M5u<
z)CqEYjm>GeQ#ttl@5aL}e<aNIhPhk+!HWHo3pXP*!^@ogO;um%Z*FSGZOjvy)d{D!
zYt)vY{r8c<DYIwoRt(~Sn{+=o!+YcrhjPP`Q286@QHl^0wchjFb5J!iNCZ-6Z4ZjJ
z%^mi8M%rj)rPV?9<D+!|k8Ahp2XRpxebb-Hoi%GX{|>u%&AbHS@#EjB^AFmacYu=5
z9X`fe;Fo{A&kRMED=g^d!PHzRQJ1UhYBn>(9GyRIEz_2UKzBM2Z|4~d*@jAs=LV(t
z8?N@d!nEXzT3Q7H{4iJ<3c@RU9dUG&jB6d<x!>0SGbg|aTWJ+}DerEa&|4J3LFZR=
zy0ap<GRPsOE`|&yFqcb8FlixL`FEW4AcOv4J?(7rbRZgu(;tkW?r$%A&}Ni}EEjB3
zmMAk8SBFhNyGz+=Zi^gQvABVO{@i|(nm@TNj5Hb`%Jk}#_LJ}pKipW~4tJ#X|IB%h
zxs|)qfF~s1G^=h?WT*wD`?^)-%ieGst6OlaW%S!lWxNT-pQka5<u>LSUIod+JSRyp
z`+}tr1q43!0ABKlY5%9#-raZmYb$<7c>;V{Z5Q@2sMmuMSNZDip0+MZAqf-oz?5u6
z<BIjs{LgUdsVCU!;7TJQsz!E0np&=)tjs3GGVvX?q-{2oeToHO%f&<LFjfL$(^jyD
za{qzSiGS1jyHV132PR^E{AVw$e^qRrQr;?`(5FjC>yzEaCduL<WK#2w|BCseDBt5*
z#3QX0I&*UEY8z5*<txa4;AERSk30&yA+rw6KB^vxEOVJJs7h>fvvya_X1wGHISE1#
z3~3NPYvKNG&gV3uO6F+LfRKq27Q3YO&-z@(hT#(}2e!7OUgrpV551`;jGgD!6#S_K
z>6-i8bQ;AF0x--&q}{s=(GyMwCsT9n6qG<4#ywF$!th5c;E!-cOrHdeJF7S0b7?cr
zYp+C|1NUPt2nYy8X`l~ZmCB6;I79Y{sI#p&7xO2yx6jxYyewhu%ba^S96vQ+oKOb+
zkBzVLV80y@XrC>q%{v~L`V|RNyW3a%_)Lr!Ru*M-%4UvetevX8aR#Z?e0^50|5xJU
z7_~Zm<5?gt@7bK{!jJ1#P-u-=u<y|5E2_<L2`FLE)^G@*wtJ;zxZQ?9z|`)67^tgt
z7`Nm~d-~;=FKTT&TTRi2UUMl8KLmKV=>ZCCl8Mg8a2_8pJuLK!D<Pgc<Q)p`zwuY8
z9K&@-V0k@Myoch$-knbU{If2xVc|z&Lz@h9!y<!1``1q!7p4D%J7I<Y?${-g*?5f$
zYzOB$8bc3{P{QvrRGF2nxpchVlr*5Ol%eM7E~bKQAPzlUyq>Z1P&v{Gd)7NCz_yh|
zR^a7WUJ!SPqtS#yjvn*t$4{)c;7<*0Hmt5(>mp(by_G<gkko{*yDjRaid%<_>e{e#
za)A5n)cSnI`2XSFtr=_M@F16-rWr(W%RH`c{il?C8CbqIUJ=*ro7GbaF-xp(Eueg2
z;xw+O7*&;&2j<~vgV9@Cb-A|rqY<dYb!&!Pm@zV9lc72ve5T)Qk#h7R>gWvh($@q$
zRT4nO3zX!sO;6?A5S9s|p*Aqkq)6ksbH3EXjGPhrI$t!Y&-BI7SsImG;aSBlr17Yf
zz5C#P7h*LML{lW$8q6@J9jX~sREpAf3L5uEl6f&|7U>nOn7t+PXIV2U`HBxrcUfvS
zExKT~qgAr{xkG4?>_#6^Lw6+l(x*vq@s05*03<LB0ti}|A9k2h)tRW-yl}BhWTVk5
zvdEUWFo(G-yX}wGio>@s`d+&na3QPSE=<jkfAPcgEo9~YA6su85B1*uj}MOOsK_~%
zQ;80ztkI(Eq--VomR(9xb}<n%s?*_6C|hMKOOkBa*U{n>k|IkCBN<{CqOp(pUa$8!
zpZmTazu)_h^QcaldB2uxd0x-!d2#Z-!XO}hG#I)53b1YNvTw)7o&~J1v?9lS7j$LZ
zZVj$dPQ0*zwcRm5j^)A4_hW5?0%N5OOejp$6<{Vt_I*F(|5FMt9J9YBKw%{8`<GF5
z!IeDN9uv5EZW$3E7^vtj?!{Lg`=udAGg2|fl(c7k;jnQvzIELkr9Bys`Avh<6ubEy
zFdbQ_5Y`9A&0gH~Z1tdv3b(wEZ4RPc@s|u*tJ3;ujUyMf>sIW<N-vR-5Ww4Nj>*ff
z(TXI*AGGnw`$HUY^nMOUJuL9b#MJW>TtOD$&u^ZqOAFg<TtUQgx<);Wvb#aJA@A~l
zZV$K^O36=BMYT!T2ng5CX^xr4ARok<$RA!Do2Vyu#j)8nr+<7;ONAD<Dii6qSFnCA
zCFu8PXMSBEHo(ctUml%Kw&Hmg;%I8k^IP2#3l_b+!*=6g7AQTj=R}MWF5_l`Ao2PN
z%>>gcw9oylt0PE2LTe?Nv1TZNKU_>n%~jokozBZhV&#pT(Q!j&B+bKvZ2ka$XL&zE
zpS^;n=aDTA|7uk+1i@q0ee+^D&*MDCPNX%PuZ&G){TpNeBPpQ`eLWelD?AixcVu%@
zf0*(E4PQu|xV4qY45p89;(8@LcKhroW>wsnuWUPZaFE@^0JH^@9j6dEasCL)nCGN%
zk1Z~IK&1jl{p`8dY2y`H#QzvBjYM%JXjAOCi#R~6XwucNA*ncwX7SZ1&W}gc34B`D
zm4K_x*JBo}5EY|_!Y5O=y>7MU=;)7y&_T}P&8YCf4A%L#JY0uy-4+Qtu;*!^&`hQ*
zr>Veq9BU}6`j&EK#tp9WLlJYLE01{t)v_!rdDH=f`Umsv+3O<pD}vq{zsuPBu4^?P
zc~OS*kpreYq{BOJH(dFJ|Km7_5-tRAKqgvNT!4fsTGXB>?P|(-u_kT0op8F(#AIb^
z1$B+X(|Ev;YZGG24!0HcXVaumfLbl~D-<4(UxJGfY2>`I(a}mKu_c3*TFY}6T$QGZ
zbd0PR($f|}p|_*P=bu`0cc0UqivYfC0IVH|uW}|jZ#`XjnOmM~<~&M$NzCJZju#oq
zaKEIr9~J|H0mO|2$CjK!ty9OGReo*2cLWz+9^8!U{h@#w{qpU-b0H~{I1=^?ENoIO
z#ze!hb;U)BFgCoOh9SR0u;tz+skbf8JGp<ZD<E@t{!ft?+*8CK`#dor8mSFWsfFWj
zvahiaZ|%(Q+m)z;K12{*>9V2&zY?CedtepAn`#j^1T~ie03GPUgEQrUB!O_6C*1@5
zhYu0i6~u)2?>+Rqey7@meX0X$!9d6o%c~t>9TML!5C<VbPt2_(iQ(ou5im~_MKh!o
z(CYa-#f(W4qcMWy=*;&{`3+ah-#980@#+)3*Ad7bni(XeJh+xUZONFng3|O^b!%{g
z!X@|rg{f7rFg>PwBKRX`b%;B3cFlG9-Q|pzl(ySEWRx>hF7UeI(tnRn$=Hk=ip$BW
zg7fjLqmfk>C+DRSghfA){ECf-yh{oCoI~4#tZRwr%w}9d2bhrDp%H=XEi<auYg<sf
zO_*1pxwL_1gbNA67G3q2dsx-NX!moMb1g`=SEjwY-zL+{dxLx~D`RaHzvgf}PU5I7
zQd_=-DOQr0Hy*Z!C30$6xIF)Q_^p2UEP-7#O=-8l4Kd7}zrup)i+B@Xh2^6?F{0>z
z;xhKs?nbpwWxE{kc7B!{>I$C!RHQ*%qp*xn??a5WKh9-iN8=!}+l(MvOvL8=@AqB8
z&mP9Y)y#h8HxIfU{r@fjyU2)YVoFOZAfto&%U*_Sfc?m0Wq!OMa=GqroZC<?TY!hH
z`Q}8wXrS$YTEx~$K?HH@WO52evxnL`9Ix%fNcd;>1O#xn6FS<KfRoLh1YLjb{iG1J
z^!g<xr>QSF@Ha?1zd_g%ToE4&r7WFIrh}+41Eky|*<>4J49G~a(DjgyE6P!N%1i+A
znC|kLSOH`Po^UUW5g}3p(JDjrsEY85=Ei=*YnP<My?WR)FK+2Mt2d)a?ac*9`6QyW
zr6B>8+4oy*b=<A9AE$}48RFu9y<+`1LAUeH!luH2=m3$Ww2sBP^qZTyVj+CtD{|uQ
zV?gajttchi_lI6EXZ?e?YdZXKa6unBm=`0;j}tAvy%57wgfja1kXD*hF2?sL5yLa9
z!_V~sYcJt?q6KaaPmsT`eduZjp%sEt;1F&5R$YX!ijhAcLVzhmub}rpO9xH`xDdc6
z4E>my9T4~6CL^HF<nP`eDs>*LuO)flhu~-c|HE^%5bH97pfLbLP=_}DK>Qfj!}gVf
z^%nDQb2Sd50$(WJd7Qp<A?IU=&gijAQX{2EnD&msewA@X$PE;h_`d7d-Oq+!%8axW
zQZ5U0A;p`Np=(gLeCxO~6ec|=?FW%I$exCBXu>{f`}aE*@k49&ib7aM?u)93y;IZK
z4YvdF>Wzk|u-`WpsI~DiW*HJ|d8CpBlzYJl*of(=BVqOo*5hv`Vc!!QVQny%WByZ;
zbC@WMX+ez%Z?;ni)<0ceTjFb4nd>Kg{;3-kK)Y;G!smVvHYf{Wkd7fo`pQ!TkPHmh
zeR&0S46fCC0#$0@i}~uD7v4kj%jam}7-8Iye23)=#CF{jcB?XR)FFHAZ^ezTE(3Ug
z9(p{Yyhz;$#7#|fX1Yy=)H`ZGo#bmIZA^FbXEwp)%JygP>@UKa+v(&aNuRt#xgPJ+
zV4MNAQSbFcPZRARGnBDjTqeZcin~4q&0x&b1&SxZ7qW)@`m6U>s=>%y01Fb?cD(}f
zhG!jS1iu8gZIFw?Q+(Cq!$68tt^wBu)M~MCmIFuc!2+0}t!vy)8}6VI668o14cI*^
zbeV@D9b=X}E=Jc1jt1wrd=MxDc_@o*c08I$NlKX1`;#o<6i)!RpM&0xIFeu6#&)kr
za~Fe+x_B9uimV6WaYpJUbr=Y>dQMdnfLR42!Bb(pFmi_`Z41qDee*9uLR_BZIw7I<
zk$+Da)5f-dDi8NC#+*kl9mY;7o!(}-UC7Kd@t>N*MLBz>jq0|n4*2b6G)vN~Th2uN
zkjc|MusrT|bauE+Pur@47GSQEWYyhYzd7zNbGB5fGD;{$pFyZbT2sYT=uEvmu37gb
z1%8-f*OP@t{4UrPux9v>#1ER#Q+>OPlrhhyhAaP}D}45KO5g#Z7gqc!FRtSnzg+ES
zMm#qhnDSuhv0(P1fGOe*PzXuj#$Xedqe*WaxkmyI?GJ{+e33g)_=2Rl2O&3DK5;gn
z7FK8&W)tBEc5rQ+!cM0;*C28e*=l)E<zLXhv41AGE~8`2XIXs%jxOJpT&04)-fwHN
zHFUc{LB@JXqxQor{7U5+R|rKBsJPms9<gU!UKt4VsM}JLGxA`C|8hMxiQ;q0#=WFf
zygX+07{(uE0Qo=-W7ae;o?jN0Z}tO-2<OQePSW<@VK6|i66OoHd!}p{NV?7KnqXL9
zT?=A4*kM}m!dj%e9V75bg{3IY9V5u$8ofc(*jNXf5LW26b1+Ej=UyTTNS5D_t>U2m
z2YnYH!lc61b%Ifd2iVh5hObbYZ1@p1lel&qHI%`fN@4Mt9>FjO@rMD<2$;#S|LLs8
zDhzRG4Eqfr>5a}pJxmjr3iG9c9d!o|eTC*ZTyaK+NW`b<{fI$CO;a;F0=0)<%u<tY
zZBu9axW)nG)x#a_HoknkyKg0l)5C`3RGE;o+m7IHsgmQL#J-w9|2GR)L4&@G_uF7n
z*OTA&C0>8sNwe9eyc+i}L1A29h4zgRvQcG%7pE-b@x{*WZ4zf<0P?fs!;K48`BrbJ
z*ivnj2t1o3|4(g@d(Mk%AZ|FY0T*L3Q3()m#NL8`0Cjxhj*<estYl1TeoY217-z$C
zOfVAW$Sj8%aI~Oxp{$+x^%*-)#V0l%U^f8*-!|nW2TB(p-|B*ndbpONt7E7!0NKWn
zx-};z3VVn4lT+31a3(Tz*Q!3@A1q5Fd*&J@x(-=5VQOHLA11_lUKHE&UN>T`weId?
zkNL9aZ5*&k9F5_+6|pDw<w*_-v<W-f4^$DgTV|eqzmgiL?NR>k7(HE-7Ywasb9-VV
zwS=CRcmYx&i-ADlr$DsSapR{Pn&fy$OH3^t7yb(pRF8bM)zsLL0b2wL>@G^?xSy$}
z5Wxd;SY~%D=G;((?-T|T8(cC%B~7R5#SNC)iT^P@GS|+oSDZ7{dNlF##+b$NtN@wz
zF-PRlU-Y08(m`kd(S|A_cBFqcmtQJt>@euu*V8F4(f;EYwokYgV_9+Nqd3MB?lrbO
zv2+JCKL-QZ2VzFSlI4MrA77dr^~FS=pb`p<ik89POXfr8X*U<Y6P!`Rg9h7UUjC<G
z=)P;c)wi+ZWjWRc;adDgewswqk0xg<t=`v8$3AuP^hCu;t_ivfw|GrD2PlxB21*h7
z+A(--JspAkL*d5}84&u%W&Y1D?~aA;9JZ`C)@FYl>kf@UC<V~GDikz-t=iz(&ejD5
z^*AdmaS-g;h-vn?2(z1^o0gcr(gx@C;ByeaRuJcV5A4@IaRkJQBP3xJ<YM5U1oT*+
z_qcdw!Up(f%TSR<Pb3x^sol9A-73zlT$C=b&rM(KXuph|fXNPzvad-~%1@TM%}r+^
zo>71A_>LB)V6b#U+(0G7N44YEd0-@9vZ$$sxt#|taO0FYJzm(^HCE(jDQ~7q0imGc
z?YfomdHPLAZbK)jnPTD?xMvD<tWN&H(8zxAB0}h4>W0;HghzzzR95Fa+4?(oAO_xo
z*QqUsF17<vY-|#KFY+Ao{j<#7RJaw9(3Z(HZwskBj8O|se-@fP33QjqD_dgPVd>Qu
z7eKnLqGQ2xd5{}dA;~ZSt~556CC!-rU|#z{!+pS&L;KixMB_(n&P#1E@94giY3>h&
zHxjbWX}7D*J-eW*zC;t>k?-G5f^s9ALxq+VkW;$RM7Zs1hmD~6uahOvm9D|NcxJsC
zSsH_2Z+?EPTRP)c$e;VW%w`{L-7;zl_0~6Ri216dBjsCAt&igS(1EG*jbb=ocj=?m
z-&!bSO#$77WgyZki(tkR1R(DMrl$@(Ii9zeS|kWJ@eKWzvzv-otsp(8ObTUu+52G^
za01$+|7AgukvfmYVvVwBzOv|ztb{?<`_{?iAN7|=yFi+fp#{8aXidaoR2`LEo+Oh4
zRNr5-)d5lbx7hB2UmteSc;GY>>-w3Oymy3~*|24`U{~MUOna;?@U;y0Ns|0&Ri;zt
zTp6!HAqI?!V-Wbti@FASR}rMRCJSSXB?}oE<v6vdd*9>n1>Dsf3Ts457uvH}<MFA1
z@6NLp0h|b7ewSMa;KUWHd3^BlD=P~t{Jg7XY1LLjJoDpXKBybCI{yDKGcxYhkT8Sz
z>qD%3y>%Ygn)V*VKc+H7^eUJC2`NW2XSJb(P{|fd)!n}wEeuS4WNityBDv`xNVuc7
z00_An%kw@_%*Tm`6p#Ll^ow%rL$$OP4<Zt}i7{|n63loU&_GUEP_sqdNiyx?us6_4
zQpd|lgIO$Z*_tR(bIVv9;hhg#C>mP#*=HjzNRpwQ?J`tG4T64*KGE?qL+mDo-@sI~
zf-W#T4otXrA&Dh&7kRK1*QZZ2(`@g!pu_u8m!G>LqyzYD5x=H}TDa|UaP)YlA5wUu
z;1SH3)3U;b9I*QHEAjidvvM9P0yuOKVpGstNCJU=Fqkj_RL@`Qe1!)zwaJQSb}jR<
z;%Z^+Xewsei%qZ^LsVl??WzXUC>kqraMa98%$aPaRQ^5}B0HvT+C7&K9Y3$&#5A=N
zcA{B1iMfEF4j{T`*Rx*d<#||A4iGCyRY$QkNuc;=9gN%?gVEjR#NZ!0qHY4M(5j`-
zjRvq|v?h1mxLIgv>+-cnI&Q@8`H}F)(GY8KfPCb#L`Q1W<rJ~x$DEJXqFpcU#uf}0
zaV#2Hc~dHSHDR;QrH$PJ*;h4W?PknzmnUs;ibZ#vVmFFLrg66ZJNc4Lb5;KE@WTAS
zxr0lz$k1!eC$N4-z=Go13u8hod=Dl-=|FfKi!e+>Xx>lMHtx-(2|z6(+)M2TGp0=I
z{Llfm8s?4FKl@BL<5l$jB}KqQO3k8mEaxMh^h@?eeS|JK!<-@>a147kpW5J*3wgRI
z;9Af;4=4nhTfqY(YxA^R5y+28h$Cx#K5egc5LPQFN%vnmc%?7lDEpYadT0e%Rgedw
z!|+30Rn|VIHshm#0@sBej+XquhyD)PNr&b?Hf*>`%Ii`?*l`JVWca}#h(Zd@-Hd2t
zb*%rIQ3g;)cqn~`+|1*cuTr;A^INjv_WUlq+#W1}mMZwfu)KVHz-ht>iTA!`{sspN
zjRWM@9XbXGfyItbmwXi_i@+qG)&MdyLivJ|kdZ<}Ru;T^cs~*k>Uy&2;L?2Qa43T@
z#tx!$QPxq8W7L~ra1KA2F~fxL$E_PXRPTExQ;-!~4&9~xEmRVhaBLzLVyz~{lZ2BK
zvCujJZ7Kt+Y?2PH0t<acOp$O}WQ%0e+ZQd%^9^<dT<%3$Qg*G@?GQ$5i3{m5Cli0v
zsH;I_PGR*>y$ZUoQF>^&nJ(3B4%?K2(FCt3Hu=j|9~)dIkHgJ=W?at`IMw(F`Y&&<
z*W-Mpu&nvlcX)dxuK3t2!!0<eEG!*{f?O0(L}IH}2s`2^s0s~DKMzq}yZlVqTr(m(
ztOo<P<SJ~$Kxzh3{RVduU;;2OP<PwKO~<&sA7Qp4;eQUu5W-G{O2#R2kD23NSEW53
zP*HH5Z*L*bb|drAC@+0G-pH8{zrO9uf&AGbz>@jf`@oL54^GLI<r9Juy3QM}6>%j(
zh<f|w5vYX|-mes4+%ITmqELYt2d~>TNL1}Zs-NUoHkF~5PdZXzE1`BDScRjuU=_+p
z{!&QDJag0Dw~m{A)u3Kny#wlRu=Dww=)-R@04tQ2BjPM7xM>1!CgK?=02Vbwp*lHg
zcJZ|zzj6udG=yQcE*aU4f%C0cZNlfh8V*Or1bo<<j!gs`<gX{*oHnWcl+1qHVoAh8
zJGwplVP7sn;LOHF|9P9yg{0FZ3bB;w^~4FNegBE;3-cS{Xlb56r(fTJ$W|T{cmg9{
z-=On4Qh1lC^q=x93iZ~-2OIrPZqBJNaydxthP?a^*1rv8j?A4!n;u+Y%$uBWbVkMX
z2`?a_8LUX^_Jr^c-K~ARQbt$h%a?js_ftfIA=o7}>@pm>X23SeL1?z1z%7*<Zso$l
z@Id)N3KD(!s}>ApR)q6tA!n~XmIH|QPQU)qQIAyF^Y2W1fr^*XiaK98aTuy7eV>xe
zPW|kVL1J6MhnLGlIiTWE79~`ktX<BopqMSVV{AYh&I|4|{+R3Tty4#D)KflPWQI7_
zA40r}9%%y=?i)onlFs_|f6U&-4@e5smm(v-0bNX!1E&vg`SeAfwW`>{7!WQ~Zd;Gz
zRwmFKqRdSl)omH1?MJ`W{Dd<N@?cRB7}eDuAwJbUelTK||FoSv;=Q49GQ2G)e65!(
z-pUK%A_(03|N8nZ0X`;eqnJLMNJ_sL9=2CGSPQD}pgjqFh~c#u_7Uki;f$&P)Itw2
z)QF_>I%J27fCbhb3>eJj0;q(T>Xo~`Hp#(;F-l_03%uEaPdV>Bqid!4V~ESg>E1fg
z2ru8=>M|3!dvT%%mE*Q)!i-q+nAUT%3-0_N1Eplk_36Lo-x|X`II;OrP!2eJ7$Xwo
zQSBi5$-o_3@}M&VVV7yN-;hTLOP+U*>Jo^MD7cG)c2HVOu!TF&1k!dOY4ZWQ>9;V?
z2xYTIM>L&a;sk#<VGl0M2a79q9zYGaNDSI{lew?|G7L`!*}^5pJD3tYne~b9!DG|M
zyWPSb0e9B@S*~5f7V{(A?E36(1ytuK=3|ee1C5KM@{<e2cHP!cqQxvg)z7K78D1wx
z&lfL**yrU48HzgvI~H-hEV4Md7_yU=v6)9J2ozhg$cJ@cT4CCa9sAIH7>}Y*Ai#kD
zIK0mMgiuSKU4g0-H+u26d$)Sp<F_;aK(WCitPP+IH+1~YhVPG=8oNnYXbwT@znBx`
z<?A|318;mBB_?lTsSXKJT}L<@gExqF&iVqrJmf5ahwlgoTt>W4XdRJY+=OOsSJgdQ
zdl()dg}`K&9&o`IR5)S0$#dRQg`wuRjyzi(-*4Y-J~YXg%PjX~yihf$uFu~jOCQ17
z{VmOl7m5L44G^|+$~XIXELa2au#HR?FNUdMxCuVdlX7`^1$3YsplaBi<uBImuO*4Y
zEoC=e6L?tX74smb%fI>=B3OaIvfFz7BgUl08)1%>OLJ)Vk0)Cj2{_UsB&Mcwm|Edy
zZ6H!igrEid^BJr>82vRfAFlO=#RKdj7ytgtUb|)o(X<d4TjVq~LIZaf@dSB?+mP(r
zKz(6o2F0i_dF!)CbpHwF4s=gt*-#Bz570%dE6OY)QD3*U?*xx|VGe&}gj+f9-TF0b
zko9YJ*JC8oc^xf`Nd_8HLB-GWnzkfOK1N`$v$^Twgx`VLiI80a4R)l#v&^~ElQJ)F
z4<O#zNa62@2@4j91Gk&x$oH^(X!3O1So@Ygv2#CkxPuXeI{4Z$gaOADNde3$O1X}i
zjua;LM+*C&d=WN+QdGZu3D-qdDPF7J%gH_ciF2I)*7k#!<sbf9{7;O;Zu+gW#oj{#
zaen3Y;q?dkhGh4f&dKdGt-k7TG;QYv@$I>e;lrP*7Wa&C#ta4i#eX98MTFOf=W@}%
zG~d>0Kcs#1YLnLAyBEWDUL}#3Xdr-~4QC?<8id-qfcm*Tp@6nys0KqEs=fV|tmum5
zPmm9auzJYS;bADW&^PhV9eUiM`VAbwv(ajxuvSV>Wf`oTuz2)*oLH1J?~%<(EhD}v
z)56|6yw(grOkM!z$k#0!m#3-UURxp3RX{||)x=y#lzug{orlCQ)PATn$)Kp!oto5W
zx$Rv_-j*K0O9RQVtIHKnmYay0mTeWg&X-a$kPzM4D>47kCtZXPQGP$dT836L6VC|&
zD113xOr~ydjJWW*6ei4wD5|L^wtV-aZ&CV4W6qbNuEbKx=zI7!aQR6Bd%1ArQ#(wr
z`oqIqD!l3!R}!|zmxy9e1#p~_wdTFaXjq&2m2F5_USGeu^Qb7avCdYKsKy+=Ec5S!
zD3G{888PP4@#KlOc%f2@0+1N3zY#R|RUsgkgI60ookN|C(0fPe;eqYo)$pr<wok6C
zYSR^QjgB%=nUTGdCPs6Iuaqk``9m@#-U8!(?qiFGo4+P;7QU%T_szUSXRIal<MQ`@
z@h*+Ts4F(HfE=B*_<${&UN!g3w@&{Ai=0d6#!^l*S_*RW@lfWRS&4;SHCC6q8^b>2
z7DG$261a07^exc4YwVt&24MgJzHdkd3z4X!_V*?^LNvMHT83Khag9NC5O8d|36@fc
zF&y}@Ve*whPlWMtTm)7MC<=_&vVuZJ8Iyoq0F~hQ1TmCdv7n_rZ38bFK=*8vCmwH9
zlVKkIGr1CxYE}*D0_0EV|Ey`?d5~~2=TI3*T##eBRWrKwE>CVx=SsuC8*tu6&g?a~
zvN;M$^)4AkttTJ1h+*SEWI8QBmStG$R_9Fii+|=bF*Y_9&kUqkov!OZc1uFW1UNqG
z!xtQ%)J=0!s&kG}f6KXJM=r`-Y(1g#+H-!TGgSpT647iTr1VL|OxKN*$joH02AB=9
zV-fp=Om<cus{1>nUb8(7i<@9(^MIPi!@k&HKi{;M=WyZXHS(xoi)`y{(bw)zS+RqX
zr|bS%6#KQrlch1$tfI*g!twC_18~rxOWdOI3=JUZR#uuCw^A3_eEp1P(M2_Ii6UNy
zO?rdzF9o?Q3jngg02%A(s3dl*M)z15m*0Z4^>*ZJhzy=xrw(__M)gVF7d3Gevout#
zQnu5gUp1SWc=qgN&2kTz>zLDF6{3efmr_0R)%R7J6YOW0Mi356v1o^qyRuh<hZ;}-
z3nwAv<%lIvC4BGEuNHgO)cW~On(j+BT&ToWf8%u1%Mq&oQp>>3yiL9c!iRS6q1p}?
z%8x1qY)DkVuAj+iU@$=HGFRCF45oGP12t|>Y|yN^GPg*v<q#~$d0Aa=-y%$+E<iV`
zcJF`*a%0~!{NaM!2}~Opt=Wm=wRu1^GtMHbKUiM4NVRNj(;O-*`ub*`i*3M7%~7Tu
zaly{|aG1TNn#Qe($6Yo1mo7R!tb2e5Fj6Nqr{Kc5M4kpQUeu^dJ)trS!&xbMr>TiN
zO@fh>WcKb~S%nrITcni|bIlmn$YHoKhF+_@w4Yvs?Q(c>aOcYrRH)r_;~Ng=gj)ST
zBr-S1vBujWBLvrLk6zJ?AIc+G_D*b0<J?U`nbxX0o=15L){OD|+t_SA*`q4Y@}2{j
z^fTN)zr~W6+hpl07U-Jf!0tu1so(OKaSJBhSL>7q53;UkC{<MC?~WNLT<>{w^5Ilv
z_WgVLOU%`7nMLR8yIFO2MM%QKpYM+1H2$X8+mA<_QqkZ^ogA~6V%244%{Tiy2i4q^
zMdkzCQEVzoL<KVD4|P1~poez6$XaQj1qcUxgPu|2!cw4?isLt+O%ET)fx5@;*jdZh
zYYrwkv=)ZP#v!4c>-%RBbD>4SKR0|Tjt8a?!^}TL{EqNYM01~LVeBABK!0Z4YVgur
z#q3<gAw(*DVwP^By8H+HEkvnP54j5m-%3EYBqTwN8zeS6G^4S+Z_<5VF30Tgs7xvG
zBP%yM*LXYS*ttrBp&!sliVtAZ>|4aTxyxx|5v~kx+ehWgiQ?3PY=fnsU`I;H|L!ex
z$@~HD0x#J$4XF!djHt8=XQ;bT<Y|-Z$G2%;j6Pk|()jmUFObXjt4t9v5hSMcA{X#9
zrg{vxTUIjFr%Z(+sqZ;gz+A~$&yl5{AR`%|J+lioHSpYyEsd4jEsax#3&&B7bcyia
z&JuTIeSLVmpGrlE3qZzI^Z}s`Da-=!0m-oE1}mt3Dj|-ri~rm54KXdS1r!@ueFbZR
zV+@a~otCw6=cY%@3OnJJHJ3XF`?c@P+S}glh_t5v<Eu=LT8KVB_o|c@3s*yWm)vrR
z>-GLoItD%9c;BQ5?M<%DfPrXU9^umsTj>x|4@9m|HgfseAjz}YXtK0Rt(5TwDX=`B
zA+wZr(SkA~ol0y?yPk5g=WqC9+u*xuTO0LYv-OnlJl;sDx^NWPsM?3AC%2;joIgbz
zlMhPsN)gmG)A5`a%#}`>zlzxpTttf*7^<N+MySgwb2gj&)OzVWMQ>w$-$GP^mRxi~
zmJ(Qp>dx=`bX%0Nl$T=Uk|~xF;{SXF&Byn^HM&^7xB3tcHxkeL3&_s;#jps$O6>a8
zWa$r;*EFHMxtZsXhgy)>N*_oL6B!RFMQMjSN7YtzQ~esa8?vwv@%grCgWpAc8?n<W
zgNK1YSY^8*^R?q)ltyU2P2(I|c{V*+fpXhwNanqF3+~P3q`rD^a%r?hItgn4;Xf0_
z2EBbRpgqME6(Keae_{mn+ruoAcEmH5p}p)oUZN_SUs<RrtM6F27O3I>P0q@{L{LCF
zz6nr&MLdkIw42HWZG*J;Ulsh_IdE?!cZn>QT)*C*L|5gdd4pc8K+_2bTg-~I4uTni
zXGlNi<p>0jw->*(OSCxR`KjKuZDY{gL<Uu_rM1a11Fp*mrojvKTA{Mn@&cS!$cL{W
zYn%&!s0v&|p77Q6D37NXs#4JKJDb`@Z4b!(@)#9~7B4NBG|r<|tf+<DwEx`G)|R{k
zerf^<-Qjm>9w|8<iBy5swuac1^k}xJ@}*#_;p2VU+bCZhDX&Ggpu%ag+9#%cIf8qJ
zCtk(F6WLYD5Gs2qR8}zSdS})-L*wea=Uh#}r;f-Du8(K}E&2EPX8$Fl!{QJ~5L49*
zb#(H(9*n$)QNpOU4pG)-ddf?#hD+5B^iRz5(9;0rB6wzQ0GoZktHr(5;~n%=%X^;N
zL~uu)JSrCvE<$Xebuc_+%r`h;l4#E-_e&gm#9J2w*I89{c8(JjNo}IyzZ4+;4b_IW
z^3aT~SXYQ{@a3MSW@1a)hVZa*L{bc<3xHIeRPL(ReB0w=)UeMPel{HqvHnqRIx1c4
zLZS-`E)-}!4A%Ho%odX~gA$?=A|g~K{21QV++K`)czylqr!`k{gq<oAwNTAiUnFPj
zs*{|9$K!pE$uTHEd;R#RO7ca<00_pxh%~tc%Fno?nqcQMzLHYt*B4g2*{jZLA#-xA
zr2Y4fF>a5Uly<lm6^1pMhFl>Fx;0LYfEto*aZdVhKCOfx_V8wxi${$);!Mm0nkLNX
z0>vrp<A8!T1+Dh9zf#Z;<SdhJZi@l32Re1<QPD;`GxcQkXXn<|9HuCkkTB*)O?DdX
z;*7nkHwC?`hYE^K`vT3T9-S7XT9@Vrt6aap_>A=@{D9Aq!>cw()@zcaEA6bixrg+L
zVe^xpOD`3D?HrI7S(_i5@qh$>#JH5t+zdLM*Oc)|Na&PN#Gk?mblg{1ixAZ%S%ej4
z^SO=;Y7u#ixQpcz1Z*va`nAu}9G#IAaMAKYXLzOMz0lLOP<jqB=^HCiSkb_I3pYs1
z7=IV8Yk;rwO#N?!I!IW!hI3g8=<>t<-^TWz^&X5L){oeUXi~~+yTN^lCDZM}7rS+Q
zYA7QGp^1-^rOX2mJMZB4!_o&8bP7Q-!v7E0{Hm*Yk@I|r>&TyoCRI~;16kIzmm~Rs
z*Jdv|J`vH~QPON3QewKaK&34{!SI*FGY}Yz@-8jQ&^Rx68_JwA=5DEGglJHlaYMMA
z+NYMOp_C5of>(9@)Hm$ll&n)=t+wvPc4xqqkBr2^#7O_1o`sxiVDeKO;ak<_kp)GO
z1Wa`)`7cE&r5?>4;XR2_jp02?nAwx+$&rbAbB*gcoNpU&axRtZTR<QLq3YKzmNdla
zyvQc5hwO3L)gn|ytpHWRsRy}82eC)YsEF-xf^H83kUF&XZ1V+KlDxjIXc(r~Vy0pw
zXc6JYxB$lt2AQa;sHprMiNC$KPuc_;s3bF=6LyIeY)+KTa|v#B>Y)g!7X<{^L$GMQ
zWrs)kn%^%dY_WW?9p@Vky}?T2lo#o<V4l&_>qhl1WJ`xdW=cgRGq2Jdl{ZuXH}(E(
z!f{)jdupv-2D!f?vIj&xLqVv4Mt$GrgjlV&X&Z*tP(Q>;4!JsWGXY(78Lb7xKH1^@
z6w7HaHKlLy>}%ki!`$vc(hDIMj)(PwG^MeGcz35PDU28U7R6JT^;>!t?I5cl^uBxE
z?b8REt)N?ZPgg0*?@w!Zqxk(Ba;NmmbrpeT{?sp!R(m@2=+iQ1PCkA)+_L#5m7Mf_
zvH1-@zwY<scy@^TUKlw)GU;--WRG_G<sq5&E?^JAGjpWH(Vib1-Bsp+Fs3e*UV$|q
zMKYy7ap4{&%wI#5NxEF+mnPbPSj~Ixw>fN>MA}6uM;z>OkJ_3loa&WQ07*-P@t(eT
z)O9_aH9O`;9djFlwC2V4jmXR6j*5c)5adUVy8J0&wRH}C8`*n4#m5fODiYD2yIcyj
zS6;~`KexXBc%Ns4+hLSCM&`GpTw21Kc8n_JsoQ^o*5e1k=?B$i8l>mxx*`P`xjjo>
z!hhMA-<iv$nH6lbHXtm3Zj2W4kRi;EP<#KpIwOYy)|I(ommUX^vf(bF=v+0G-cqNc
z_5e1fSfxg#Ka2FoTcd%eq)7@!=4``N_~W6`U`0O76#=G@rX8<Y+XZnp7{x-Aww~<X
z6<)kzUg8AnAvQHa&<#&9`)RY5C+u!fAy$zD8ki#467Hp_hICI(%qea-<wK1*9>E?I
zdPtO0f(2vo_y4a6s3_J2PtdEGH^b;bq#M67d4iE{am0Csw@T&(Z~)e2Nd*_w_6h!`
zwNKw8rzd!ipiEKptJsbnl5~Er%s5310>V)~Ezx?@qoM<fFBAJ=ru~w#c&5~v|6Sw|
zb36S$h~bov=IvJzl%iB##@8rt`B%%fQlCFidzAt~sx9F;c1q~hGEtY%$uqh}j&2!H
zO?i2))(6r?P!_6H=(#Z-((Cb4FIH3_jTv*y%%c*f=&IhegT*Yk7_Nug(>N<yA+`HQ
zDiaD23Z#~_zm!m0#~Ww(>+lKvd8vh!pL<f~muCGvcnu*5b0)q!oPhzJFt;)eiV@J3
zc(SP&jhz+cZ$}BExG0L_UO!d7Rd9N|z1q%4J3q8y?WzYluyBOhVhnZP(B7N`nR#m+
zbqCg**iNVK9(1k7hMlo*1{kc+yNC1(<~u<g(|GK=i9ZX}tjHN&+lOizHnY>jiqmkH
zbr4-fPv;$&`PQ|1Mbmu+tp1zaxT8Scg*eu;@YkKU|8e@X?J@1E<)*1+;kP;PX;~}t
zEUm`Tv2iqfM;il@oD^HGf0X6U!WE;I@R`{$>X>W(0*aOpPW$b|(&^?9?_EPv0=j0e
zvl#JOqsCk~(92y>Moo;ar7WeX8wM+XEfp$e?3v0kUUBK}+{(T)^fog(+94&Y;M3l=
z`KyoQ%nMKP1<<;l+IVZ%F3*gNjQru6$dkjDZLovhg8&Ppmpg8KOC2Vy$CDwVrkkif
zOo)mPII%IeL!zz^p{)=j!Q*?SyY@*Y*^0q@IU>Ay2CGU7{KJ8p@y~Nn-a0rCI9{f`
zLu5-FR$r+Y4&5*rYxz(eZa3eWL?(t^HK+05sy3q-@<CB_JO!eH|7`lqJor&-23Z=&
z(TX1QH&CL}PgwX|eaNe{%o(-i6lp!M`x*NpWZ@<V%NyxW6D8ziN$G%5M`oF{@+T#B
zz7jc89a8yhn?3WQ#_ZexInlM)1lBjV`j7(zfSRp<w+yX$wq8rIe26OoR@mf-OFK$(
z(0jw<<Se6y*;D_ct7vz(O(B*}rd&*O+Ac_5?-WX+TmzrfA6J-_riu$^9)>jQjdM~g
zU&Fh!4Y(>JWHy&50AnJB*zS*JwlkJkl7kgm6XTO{xO3*lS+~GD!f;Ig9d57B7-xls
zVHdq5-EL_r0Q(x@wV4Qm=@okgr|i~~)dmqS+oi`MMwWRw`@AB~SS47w+F&Q%mGLFS
zurCWRg7;~>L2YO^X|dITs<9+&;%l4T_Dwint(MvcNVVwgeTl6Nf&1AyKC`lb4oYo&
zsgIpTy-?QNpO&{$qwaPdEqm2xv;o)Xg^|aY)rF^;nyVMaa5y}KP?uvgrF)v!=*e!{
zlsnfoalI$7M28S8MuZ#vm(qW-oq4d)O)b-q&{XKC-1aXeDz|O^LT$-Q<j11C17W+c
zHyx_LqP%%)>KT=Z=d%%kcV^2VNAS<BA?BC*TYl@R6A`Rg_SBFmW9M=&Du9@mG5X%u
z5mkZVSo$Ji5=GZT_YNm34s~lwVL$b)<krCnK*J&<vAlq-JT#r7?G8*y(Gux02%E~u
zL<pwl_4`a9;M$Agm&^5z>7Xk#kJX7fF`u`QBwge#sxC!x0^n&EZbnAeOyK(yb@za!
zIkLwU3qaliR+~p_5KE)#Dpj$iZEYB0+IE!21S@!uf<o9ne5L~P;Sp!8g*J)RtRqXS
zUUz1*g!TV^`MYcsDOCpy3Q4OiZ4cWk!5FcX@nAZT!SIp;<zB73yDjgfZWw&oCZOwg
zSM8Ni6$N|GidTGsd-wp0We|}X)%rx^mY2z!xZ2<=i9t}Xz2vj=a~cX#TmyFT;wDL0
zVj|PiGq~hd-7x>&=ZN!0T!lU&YlT`7Vk&k}jbEjp1pccQ!l75tos32285w7)F!I~v
z`ZAKTpwx6LA-4AJd;plkdB&R`f!H?+Pw}?|8{O-yTU{<A$W^50k-*mzW}mIDkl$=A
zmXxJu)7~u+%VYmZSqDx|z8(SStdOO`msL{`z1JKj{M_e;SF;;rV3yP%xN7Ks!NdD8
zMQ9J`+?eS=V46=|gUH3Vifsug=X!EzzPSqFwdN2~>H`iIo{aTW$43Hq89eh)j`zx#
zR!fnXMsphPZp62fO(RPW5m2jrSJ&vB0Tj3C>(kl~XMl~xUicUShSc^>Eg0yw<ek2s
zj5t4$ZJN_RK=c)!p+ISrh0V9u><(t+2XYj@9NB4`hVY4JgZuPBGjQRF!TlSi0z`;J
ze~Zls#WGE{W3i@*NmiG_5~f-H8Mi`5?RpYNblKgMjZ?CCNnIfBdcR(XzH7nnZ&zfK
zum#sBZ_tzP!D=qN1PdLxDR*OdKY|55wxw}i*#-OL<e`JQ1~nFqmKNXp2uO1+Az=%&
zQ;keaaX)d)OCGt`N`wl($ryoM#lv|RzL*X;8LsX=oGhLM#8<~Z_DY(+lO49lL#v7j
z7Z;w!STlr-f*rrNie+1`PrjHgFd)W8C8%sc{ljXn0_&Z8;sEV$9KiJ7k@ZP4{_`cb
zH+Xk)wAowi=lDp5@tV>n`ULg>v%~}C-KeiRCg)9=H%XxhjNW8r8RZW+Q#BB2M@G&H
zK;_)_o<r~~Q`}8A9Ot{T?jxB?kXkC%LG2TyOa>hEIVMv8l(*M_v<l?AFjlL;Y8?EG
zE0#`TPW57_J2Tliaiy@Zuw{5Qao8VMWrcdEN<O!zMkNMUmzmTdf_BTwpX+6j@F~mn
zn$pZaAylEfE)vYCu6Q}JVx3o@AFsJjKhau<b|)wI=|8=DTB%B)V-&&1YG)~+gc3`t
zUxUOCv3gAegp9xf>b@QwGV~sQXs%lIQ)QO9ist*YFMC7kVU95Wf!{*s`ypP0C)_ci
zn`QC4|BgQsv#*bU$*+TdkDE6`&5)BO-)Eo>1e(zyqVw7^_Yj)f3UPvp2wd*q%LP{i
zd&Mxc#ouveVs7^P?B<)XjZ?uW{UxnrpkXh-T=p39&K|CXj-4xOPOy9h?2EDL@Z*~l
zMC$$c`nQZ>0F|#;AZh36)yEE%`RdJSf6-_%HW|r15jIN7OR3$?aI65<Gdd82g4)9{
zg4%9h`skVkXQOvtW!P>!LRbp53UjU40K&#o7`$4mM}5-P<qibvT%Pz!qdkJs3O)AS
z0@9^j%S;oNiny!t>X=D@``@OgH{d4o(e4|anXMo<0U`yZgUaLRhX6{OVSZpXTDRHb
zp_PpJ)$c&G;yuwPY>^(+>!3CL^zx3~Q>6z!B!xXxR^e-E_Pt`xF#=(^lcSfYpbVE1
zcw%ef0v;~J0Fn`mUhsB`dlTeG;S#~BrmIinuG9m<zGE<fHdzO&@5#7QYOH~9o7ibO
z39B9FDhS9aH8_`T!L1gTv9f?VR-`pX?I6UXQfqM$+b-~aW=CE=7Y{2`-jY5wXq>MP
zbsrHw#+PPzMpp#g^JKPGy!bB&sx@%ER@OLm0`G-+#$az)Ug1XolcT55u=Z-pfL02F
z`QzS(&>RNCGBrsQG&py^bdiNcu$K`szTu3ozo0xInf-zUn4zp7ot<G}BUh(apNFp2
zIwv_3XSC6!+qEmP$4Jmdr15axht2b!F$PG{{$OX+r}@1Hj61m=W#i1Nbb-1&%ZG-d
z<IB^iX?tm!-mqUqIx6$DMxQ`Bltk!F`Utc9zFH3#Kuj}q&4j915*=;y-g0IOE|kc+
z{J{PtKstNTzVjpLrVO=Tu@{PabAyZmRqu<8XXM$k9Xb4|lh-Y?>+E7J3*NCab_bs&
zwF2a!rB&1+H132H!`~aBnP}{JhU$J!+~gg_{ekxM;STL=OERwVQpuG*1ib@squ`0+
zM)Wy!dl>^`^GEh}CJOi^_`Q#~qfM@Mb*&~Y2<2c}NDPU73fd_U7<0wWNpe>KDM|Dz
zv2b(MD9tsU)gV8g=Mz;eVEFQj>9>D0QLv=%_WJ;aXL^B!d;eY1?aNhaE2Qvr<7Fnb
z!Lt&6klmiNMd$}mj6f$uLCa29n3$;6wg8>m)Qgq`-m)4Qu_C8;OO>;8kFv66YsS7c
zj|QWxL1_V+P3Sfc?C;K-c=m@TdSO%`>X;Tkn$9TzE}Wf!Oe(PBUMZx-DNZOC?Xn%=
zeciwKLYx{w+ZSH0cwLZwOa_}u-M(xHVUhrba?@CO5Ki`TjUP4Hk?_F2rqy7h`;a$q
z!P4q7;MQ&cb_bYxZSz}T>hTJKlF$X#|2$*VC?`yt9YG{>;e72;P|$H&i@<!vLXDvX
zqD9e(Vqqi~6%7-XNUsJ(`tl8pPO!Ja&IOw%3jK4x7_nPQ3Rxm76)-ckD1x&YJu_Jc
zPOE5D93S=Sd)mtpK!}5{Lhurj7?r5i7wT<^y!doj<>~5Kg_J;1T(5iHA|AEE9N~U3
z?x&#}UtUIhjEwe?m5b*5htXHZ6%(-lOYJP+po~jaz02jX=6C!*bxS-<FN{tg+4{Iy
zWWOr_ZBtTAuYlAT5XY`m5LT^FeC6X47RZ|VT4Au!t>c;m_S4`$RVHZN&4dhfCh0EZ
z1n|j~Few}!D5eD6aNRWoBMw1o@5B;GR$=WGUju78H<OB0D!aK`=NOE(3l_zpOj^Df
zU2BiXk(=I=S<?b-U=e~*$oS)g?T@wRG`{e#xirQu(wiV_WT8}PFu>5bl!O&B=a|x`
z)($|P<agLv%V{M9+@00h;=nO#mL$co2k1@lqbXP#izTC;^U*Voc`mAgGP%sd@UxQC
z0GxowR}B%wwc%HUcN3E6nm$op`e12(=DE0=cUN;Wg?0G<GXK{Xel&f>GGAd7pwv}c
zCZzICdPZN4L%Li3?q9Z@%4K!OL|&*QGKpjF{eN|LXVGl<`P)V#^tQ84C$2FK$;U<-
zeV-{r0}CVDhMar3^OWb<7Q{bSMIB5K15Zp=<8Gu`RBc(R@&<+>WhW{&wzIx1G!>%2
zyrnY|c$&nqb~g?;&#u2^!fU_Je0deo&T5`6aYPy<)2sSAaDsrr*EyMie7yi1haK(2
z(J^WS#@(uwwlpq!^MlNQ$w5+rKjrFM9CnsBx_@fLbW$HnlEVjbvHXe<+cHs*zf@u6
zbLTj<3?gH3nO25SMc|`@b?I!pY=Hj8#NYkz$p5B<q;bR01jG~62@Y4nC(&AJBW7{<
z*P||5$OQyeCov}SH(u6KC!j*ZpG#Q_EG!iZEMw;aV48GQ_n4(W=ymjPHtx%fpPy<x
zVv(Af%&Fr-$#H#WnXQg_9_PhnQgSSSrXFY1;H-^)j9<@4n$tFTW8pf%O6aRYLx$Q`
zAB4k*hHn3eRsDf4;=AxHcPhftas`3_aOdr0k9ivi9-Ro5gtdUBV7g*0E#%Sc$MR!T
zICs&x)twxT{b1XHEVA=HTzK;R3S01kdOAq?<j<u8*jXx$#;%yX4;yuV{j@U8z6R^i
ziH#f$DQh3+yS2DIN2Y|4amkRSUaNQKpi+SU>;mz2q!gq#BdFf?zcd-&fN}ZUx=Z$a
zvwh;&M2Qrx{k~d3<;ub3HFfXUya@PYUc)}0PS{>(39PD>nd&^T66BI19nuWN+eUm2
z#W2sMS}Mwwabb?JP89fag`I>(LhD$4R#8CiwaTw+XGv12Kk+7t_cG(6++lj|$cr1W
zC{U`<RgY2z$atw=EFkx@bDlvpmg9b|LnrYtuqQa!69H+l{YOY{_Sr8-ki(ZM;`+C!
z>J*D1lvmRdC5VNZ>MAOqqpo+FyPb}i%~R~Qe@u8Hwe%P9{cp*MXY(zK>*TGS>P@TO
z<x887Qc)RFUkBvPwVF}ZXBURw4wNIpnsXu0AbPj<JKd{m*ole+Z_{ob$0!-tK2+VG
z{n2<{>wRG<F~ad-l^pU2O4u;8;MZ8pQ$UPmF3i&4<ctq_Bg&#0%S_1C3N8z1O=00r
zozFDVC{kOiaadqiAZja7KaWhbO9rbXv;k7UrW8a!*(jDkYL97npLU)l^;>%u92VID
z@Wmlu{o7is-)-j%@2N~bxgCt-&XUh2;O~b=EVCi5X5X=?EbNXsO+&<b<~4>^x9{4U
zd2nB)o)tj@lJ12?F(+Z!q6riu@fX11y?r|8rJK`<I|p+*JBf?OF*;4bZ+p7O)z`$u
zT@{yy{gA~j^6B3k4|n8{_%PzU;)Qr^Yne2bnmCLEstlsbDLq1;i#`5^xsolN0<}KK
zf3#$ll5K}{^Jb`EP54;zCe@uH>4=SiCuCJ|4tYTW_d3hyk$;HgyJq=yLxhwW@r3(o
zTkFygOLFyTT{|gM3vm2a2=kXH%;%w`D7ZfmHerPLIyvv!MF^<}lLD)28EmU{;>BWx
z!9p^I6?}he`Z-VpSNq45*a<jMW7vsO+@si*fh`%XB7%)~7l<kWuJrn1!qe*6C!bmh
zOxKHfI{UnCw@jMZW~cHb2Cy`yqI{wg*p!rKn@c|1<VmRrO8P|4&9PQbK%xv#Cdw}e
z<O<=v$u>(V+WGeWJ^#Eoxga)9IM;tTMGVxAQ`4A5&cH)hUoah=gk0$(D?ve4P$XIj
zaoLH)Gbn3GYE56vK5+V0V$dx`Ksq<`P?OQ*bvD@V3Uu)pRP-1;B~yVKm&df`)@l{l
z9F4*N^~G2eB_WRrT(v=W2ix%hTNEsf>#1LbJVlLl2>*Z?PbWdlv@$G?-$6JfW9kh@
zEKAa@WcH(EwFO6GKFY4)vH%qaX-iPwm$g9<F}<9ng0abPaTIpwO$o5}ZtHs>K7`T#
z`k`cdtgt2C|IZH}b1@sUUJ0wLx2E;<V7Xn;>xt4pM!#*}FK-FZ?u|i9GJjYAk8<)a
z%Qy!o4Ik$-{s<!C=Y6`qNVsA#NTr-UN#|JnjzzYcVXR>uWIrd*TBu(|LRaq<lx~*g
z?cZEs1|Wk7l$yEg(x8}Czi4n~wgCetNGJ^_V#{fBuTH_-Ob#!>tSjDW{xPt|o45!^
zXN+Y@NDo{}oR$PE9Jwa(7LR!9GVxCpS*>%>paNAQi@j$JA3wnMu2H)*5aBMG^^gy@
z?h?;@E`PExv=dUc>dv~q0?+unIX--50F!3Biz>Um!q<<C#CKXN1@*Hb!+w}G6PARs
z_*g^k=5d?>SVsCOh4>AxyWvOB<%)qcw`<?Jo5u|v>*Fhx$lXq`?91`lwyz)YWMg+!
zTZo8q`{Ne!!A@kYFvY20Dnn*+#N8^6V`UCo;c)dfI3(|&l7b`xFgFJ*MwseHC@K*{
z47J%m@KH>x7*1?YyRMYjosW2ivrIIP6;juwE8ho4nZD3W@T%qr&R%><60rd`1=tV_
z`f{JAAKd=<6t);gH4(Wfyr8u2?(oQ!KCmqOCGw`w_HCI>E-#;Qv(~lUif5q42MOzp
z0@>mB73JbFHc$irUWl(F8>zuSVRU<Xqi;6%t3J6LRyjW<;h79jx$#zWNKeL4`HcN2
z970r;l|w^-6+HcVmX<Akg5HCQnuyFIQut#9fpamMp9Q2r;TAH7>Q28W9W?$RGZ|Bi
zvsVPca$&Z3Jvt(*27aQ&BtYGU&1?li@|1&pP*eAv;Ssg7^#NL_sMu9kw+2>b7=Zkd
zB{3zawJ?9d`J+4rB}y`uqtq7Z_>^kZ^~GppVi1mJ#y)A|Lb>^8*JTBzwWKR6b{&RA
zmCE#gP*LaSC^O^4t|MLmSj>ZNxmv|$JzZY*nIhX3!Wp=M5OP9qp}^i;j$-P-7GJfe
zT@we{tS(rEAuH5&;_B#I`KqHKTQL%<YEyAh2}s`z`Ucrn)`Cvayf|X~reWQ<A|&$Q
z*63Y{Jn@hq`BPuF4jSra<|$7vFF4;hSj@-9e5x&c?K=X4YNo$>SMjQE8D9xyhRpc%
zL?s~A4B!->#PJS5y;H|CDxot~s6p09&C%rk7K^=@MIEAlmVvm-GuG15KD(t*U1oqG
zR1~HlK(G@!ZbKlV(vjbfV)BG_-qf;GPJg0PtPQIVwgISbdWv2^XsxoMWU$E+xpRxr
z$ecF#*zm-6B2M>bTjY+|k1wTT?LJ(C)->u7G|}$$Jl$rvI6G65eGo80D!~B0v2}Vv
z6A#~7RvoVflu2dg6JMSSvX#*@Z1NCEvP?yD<C3Rk%{K%&qu6Bjsuu-q|3~0_&L?L&
zh##3wK^b*B(MHp0wNTEbuBIi8H~cuPpvAS>hDoJrFAuu0iktu3J^T(C?3+zW+CzmV
zyr9HRSaQtzK9=_iY!9{Y_F$KWvd5wlNRT(FWS+P-TOo}>(a>!LKozydI5!Ka6dbEH
zzn4nagAE>84(K(GZtB#dL#Bv5jhcoxJdrbonh5Hus6KQQ5|F=a#Bp^Uh}=yyrPByA
z4_BGC-&f-8JK!L_0m=*K;3*W3h5}Qs{}^mF?Z37<3<>}3y05x<0kK^RMU$JNg@sdS
zFOh>aKgD-+FNWVqn;SQ^gksZGurW<KOIJ1439P-s6~*w~`p+6h#eX%8j*mwr@iWy;
z1;7v+whl(-p!$^;RHj4<%9?K1!ADyQ*@F7x#?u_(`<~0*Y;J~dQ&-vdRUyEqYPM4Y
zo=4i(VQBcV-mZwC(S5T>O>W8=Y7F+(J5yhoLVhMV=0<mvi2=>k?3V@X1cs_Poux;&
zoP3HXQfueJoU<ZRcEv~eMzb;XGxpN3(HQ$-)a}t&&aw9TJ9LVJ&vC+d!y^PBXhvOO
zdxDXgZQ<%2t4Wp*$;hG~^2$O$SzD;+iXFBKLsJ+cJ*y^LxuHNWShg0<xpiSD7~)oD
z@^js8qFADytu4kD0xNr8&4G~bUh$k-?m)1>X6pb%4M^0#2uu1H!~-)-uI;ezWbo4{
zA7np`6(?>sdt|f~Vq?{&C_S$U|ME7WItDYnE`C?Uq5;+=y5i(h%aIMW*}zZ~^@{TS
zvKOG~{->zc%fbUsyW(Hk*pM6HTTL;Bm-LbGI8dIzYGcS`L^77ODDm^<5K`#<$=rv1
zhnc;ey8z*trNN8X(zo#^=AA6;c`kpiFx1Z5PYoE2yk)*2fk`soP%Ab8iq@@{VQVd5
zYvXwBCR<)$>YP%1=inWcAnMKkj}?#jBkg)`a_2S(XH(R;d`TD*s0;*^K~$!Pd%Hr4
z74-l90ISb7k21Jwv$0KTNxL5Q-SvzU#5d|JTf`G^4og?t<+}+T>ItTQ*l5f0JFK-N
zOK<742e6P><R0Ih9f4oaUnMvRNwcQx?mQ2?jH4q*38iJEig12yGlelMl41FTebrW#
z5%hC3C?Cu?Z&S0+`P1Hk`KxPKd2-gMx~sq(cn2Fzf3va`J`q4kYV~=_&ks#Rz(+2;
zbVLh=_&(_oYXj;cj@MT80-FE4!%XzvyR>mDU_`-19eK~`<BiIpyRTg=uj!gzmzsol
zU^~)~h;*bhWz1_pr!j#pC&Jq#6bu2`Gf}7zHTEM~mqb%@9nuIw)DNDj?M-*d5toSx
zJT-_upUyJ7a}cYm41cVtVvbdJ`u4-gq`ghB?cJPbu*60OTHZuc5;JlO-W}!~ipdU>
z^jO2%4q2)&*0VBuNfRn>)rT;f-dOPzMiXp=U-R~`QH|e9WM(act&E|7?Pamry3dxV
zWz*iW&k=)PQ`J6AtyD-YKZ>ec>6#`kxATJrmKN~=oOqaQ87l#7?KjTw`Z$qHN7+-}
zV7(&Sw72(q^@n84p0&<zZAFeRydQPM@uz5S)>8ESy+ER_0K!jVS5K$QEn(13g{Knf
z@Jfhds-sE$`w^u@tB&{Q$~^Nt%eDDRvNW%WjIjtD?XP@5Ods;fx&Fp-byE+DgJEI=
z$|lrEP(pTUN1TWYb8Bc$Fu+HaB2Ry4_F-l-3Q<k6O#E{>gUk>e!Xbgy48;FVUwh>H
zI|dwl5LA6RAXim{9?{J175VHShG^RmcfkbO#L*tCx^$k%LrHw!2K#FE`g|c9mUxJA
zrnerJF~bcZWUzlru9oQtDb>ljQ}LUhWC%5K9&!!+%EM6_MOh?Gh}&&-KHa)Pa+Z`N
zvU(yTQMf>tLy`(6?SPcEv*xfmSHy)LaF)9}(QV#z^Tgz448I&$d?Y?nCOUJLySaa~
zNt`My28%nlOu~>`vV)BAAVa&{V_?6a)GJ3-XeX{;bgY&qg0Fw@3K8S=O<I_L@)6?I
zw&Hy9G4e>%zg3Yl{b*d>4F$ol=k@}I&)dM}C>;jg9+WY3-G#-Vw&?|HS^u?20*Z%2
zivR-fny`_PwNyt!y4u!=lXqZlv-Fhy3Eh@{&9S}7{w*HCX`)~1*bsljloTE|dnlaI
z5jkBalcsnwLhZ1eeyUA+f<$yow{BM=W^{MaN6o*Ws?ia|u5rD}6CV%0^vW*W+oNX=
zdd5@z#|42<S$FL+Q495armU=<%rUQCQ{Q^mfHfF>v!zzNz&U$$ygfe3K~ytzab`I~
zjXC#Y0VqN+L%hnra*aOSeu8$LoqYHzp&IS`K#s}VV6reCeE(j-t1>^%B_oo`M9Ir@
z750df`SCH1u^u!wJ;60&0V+@Yy82^ErsXW+^hTw<KbI6$i!Sck{2QZ33veF)sy4s@
z`w9r^ky5!QlQQ^5-k045oDz6L2#8IM#cse*G3JDx+X~v7k9S$6rZ&}FshV9uDfED>
zc_DlHc+_!45#Gv+BUa_h^n>C+3JJZVwi)MZfEDraa8_{LNNQzGWZuwBK(hJ}S*L#O
zLOLWSpn9wLZMWd=jMUUr7%C%{JsfK(75c4%a(wyf38G2fX%(i%j^8i?K}nt;gGMn&
z6io4Vb5+~Jjyo;s(gh+U1;eDhalL%VcbKKV+bmF3V%QOw=XxUOj=0;9?)ZWocCki`
zUpab!A;`*NjB!HsjMoM^fVDk%p2b3%Zc97sG`*Fy+7hCUtCTgmtqvNE(UHh&Z3=L6
zLUhmX%z!`9R>6C*$$L`9qcy@&36)lEX@`Qchk}VH!iqu2n5PxLEyl+L<Xu)!4x4yF
zhZvhnzA*Dl3Myz2z|w^eBu1&3mrO_NlE&9Gx2g}F?+7tizIuvF!kj-)#`OCFRzY?<
za|}9hg;(Dj(w_MWBct$KSWCLrW4*%wSG5n`32ULL#*q11Y5yCJXoP&S7DeF0OL>dz
zE!i1FQqMe6b;Y=u800S{2Dx2lBOa?_lr>O-Avnk35fM&Y9Q8tF)@in+dF;%Y^TCQe
z?_Xe>;QO1WH1g9bt+l$jnfveW1RSHKvgH*9XpL?IqE>;3A5^40S2yD5+A^Gdi7zq~
zg!b01KnCZMk+U*yNlu9h?d-AmsDPIZTP9;7cGLQ>Kl(bT=a_yAEpVJ+AGb4Dz3S4s
z+Ym^>%z?O@{iuSF=saTVSB3Ix#xx{hzV=`R3!cACyG{n9Huf!^v2;C6`R5nFy@0|_
ziui$E{iuX5zS-jMAUdpvp!%;`ELL?c>R|#c;CKi8bN$Dw&#7Qge8oD}5-cAA1ha&+
z0;79V^)Ii%bp|gHcHR(@@VIt4A3y(3=83{}e_=R;lhT6skrgKNNxOb8y917=pj*or
zD`sJ#)?_I-9%Gb^9XoR{V($QHGU{UaQzRGZ{fFT0T-QNVVY}5!u5@4f5ib_EK@Sf1
zk1#`uep5I4RFUgtn>D9sp9AxhKi9}rB9&iR`#|q6#fX8^1)W2<p(AWqT;6}hg046d
zVtz+X<QLtRVq!<TkIc(+TGfMt-dJG*YLq|K4uM6+^6f)CH*w@b*s#c60qGR|Imhu(
zdsWsbnO$61F+inRU)KAUkBa%F?<6iF_q%Vwc>~45vqz|)WjWA%=`kD<fc=5$G1E!j
z?gW<{t~(d5xZV6qN1@Q(N;r;mxyD$KLBz7@#vsFt^N+Q;u;!zbV=+_9meD<l4L;*5
z_Zo^eoJ23{y9vUtLz!}xZ_}t=*@1PA{mxB6h;ca+)p8#41?yPo7sPqL!g7zH<*0$U
zs75e?lPwIXjr0g$%VRMaBjlYF{v<H2r%Qj#wGwKb-qAv?u4GReE^EG!5BXkG3}#J!
zur!9RamwH(d){#^iAXKX4q%^gXiRs#vtH1w+lToyW{R8qDdUT9=^m*4L^;$KQ(&9O
zpk;L+a@tHtHNN!ov`LYccl7@-PhG@vw9^6pgz(!4uw@8|o1wS>1b<7JKGTHI(}ZUR
z!G@Y3X9j1>K5;`k^5t|npR@3y;zwv#eRs0Cf?(O)0Ti*xV$`f=0oHcKK1z_!0@aMt
zWj38v%3jFBz)KnYj$Y%u{YA)SHuM3Am$8#H{kzu9=7>FxK4`QS&~iIx&LUuK(ZubJ
zzwV2}Y7i4bPmY`4*ExzV8mrrMlM#v=G=4Clj3AE`akoxUlB^Nc6GHS&D`(LT;d;Fg
z`hx3~cke$CQa3vuve~f`&LmJ4bUMe&($*t#WsY$Bgs5N?FTtsweB3nbA%W0@A+PL}
z@50e{RZbza1~N+is03tCdigv)c=oYMX&oWy3^niLGP{3|h1*vf0^l0XM^<pc_q6$e
z-zKmBKepaH9_qb+A0Kr@DspFOu_ULm6i#()VYEmolr{U(B4w8~W0XXksAH>y%9g~C
z?8{h^B9x_3)?^Z69gS^_`8{9nsr&Q!d>=o5-FN34&Ai{Q*K@g^*L6K<4Wyi^(*3t8
z5`%5Au97t+Dl6X_%1nc$b8yi2!W-J!NB6#u6tvfSkVCt<Ih8Ru0OK`gjt7yuW*h1|
zBd{HwNF>mOf*UG)kG{ry=)a3`N`diJ=)L1y;=A@tJ$u^ZQRQRc;I#|!t0N-N?5M2J
z+B~<q;vY;Mhmu^rKznWI)s1KeoVQc?+OT&XeHgWJde<1E{Q6b6$j?wX;SlZhrScAO
zRW?O&_c(vJNx`B^#hwi*6OhV!E&7jHW}T!}O)DEc;SRufcGbH=k~5TMlsJ<jdQdIE
zrX4jokmo*+d6o$b(_xNeUs44KpgVC7#YqA0WrOpR5;m4QrL`xo#1Ct^#dD)b<ksXh
zlgG*>(EA+0%TJZ$^}2IS--28Mq8ZfqA|IT=C5mju?7h)PR=@adJs8!)gDuofmrFxL
zL-8B1HXJdn6Cs@Pfgs;%+R=fYXTy>^b&Ft>Yi0`TbUe00*|GyGm>vD5nm;Q5A8a#N
zv)<#(59~!@nt&Qubk7gv1AzAM;JONV;d^%*p#-!6#?Pt{6)Is_Nfvi5@wc{O)tr^w
zE<>FrM5OR6kX7JuQxN_)%TbRX7L<U)HMOjt$L_m9eGeLLd=uD#A=LK{Zs@%O?Q~Zf
zeg0E}F173qG$JF)pvfD>b(=cln=u|L#TdHoSkaNqa1wGK4S_Zaen8g&xmk>J>5t0%
z=rb!LD@IoB2NdOnZWB~~ehG=-%bH4U1%lJdtMN&ZSA8ekjymkCYyxi7EgvYo!9w;B
z`zk(9CbZ)2no}6SeH0o=&>~@-y}f9<v4nH#hrE+CA7>W7+fz7?atDj*-O1^WVQBOj
z2Su<-2Jc8d2PUEy0ED{Lzlbxe2udxdYjMwCh7(81Yg&~$Lhcdz0hxCOcGh;(`dWgD
zXLCGLP_e&9=wI)%oO@3Vya3(8tsXDrFg~b~#FXtuK$|b4beS->Tt%O)XpT`PkKhJ(
zK`h$qXw1ojAkO@!`;mK3(c&@e$&9OD-KuhYFLzOxpsLjcoBNe?g4*eCGo?vb0Qf~v
zWq?mfzqY!z2Wm(Qu7#=!@2aN(Yy3{{mVRLRVf4R{9#D%37N~Wn@-NjB0zEa4AHvF=
zigdUx(!RtR7t0s~qxM>KYi38J!;6-;3TXYRl<v3}=NmOYRnN)fKr1Ji!7qvtdR3=<
zf;$k}{A)93UL$7CEIxtB%Ih~zZ*+i0q|j9E{<1}x<O~qxgy;7PkAa6k|3?aHThToz
zPg;dmSE#N0=_gkz3BfyPaq;^S7F2;{#}|Sr{*Z3b<i)AchFMoEymPoO7}b@Yjf$^o
z_VVSp;-C4##ly^dRUepxzqN@o5I8EK!12ug56*uz9Q&;W6=CnPWh@{_79wFE0xTdc
zalgw}3OK0cbHixs9;6cb&yx=*FNZ+Fi=9nb^1tRxKr<9CCx>~&2h6w!BOE;)OElHE
zUkSaX{K(9lt=!x%MY;1J(6NHqL-JFG`7IdbQyEWE6dBD}b+ubJTSlX*AG8*Rscg4d
z?^*{mjnGqZgb*PS1f#L}3-G_9$^Re^Z61N9I&xD-Z}y;D1{1ARtqdZ1vk>t^9?Yyf
zjyPXFFL&C2X(N*npbIZx8@9KZLgw_K^CwV6aks5lsMGC)!_*V2J0-)+yJa7k(}1WN
z0{^3_2a{~<cxp>L&<2UJd#Ce<?|BkJ$8}(z&Rf;Z@coOy$6gCs#yc`RZ*17=E@?lP
zPLU`-m-a?JeVdGhQQJ7*E-(amdI0xHeIJM*f7gmeo?dbrSJ_v{7OSn#e(U;sre(WV
zxI9bO-mmh-@hKEL{UCPs)dn9Yp&^R@N?{8vxgPvkQSBo%BJl_wjk>zrI(Gk}x5a|r
za?y?p{6{cBTp4*~ajHpe1K9Es=73PU+TJ{!pn$MU!M|X-L?Ig>XFBO7=En&7%7I@{
zCO<Y0+@7RY2*1}0wgRT?J!s`%RV{#c;qS$V?!w;Q7;h*42L@TQI)KU6D8LA-KB6V0
z#Ribu{x^;VmxfzCHg%0v9K%7`mDXOnj0n-ZJhPu&M;S?L^wTLS=)Std!tXC%Da}D4
zS(#Zm=!4G8g{(R5i7a5@)%fahD`O`exC6aKCc}t;S`fQ?x9T4l^)TXhc`>M*LAzt}
z%n!~ZeL032*w)2NN+ot58)ZD;tLdPr3hQ`v8#%b4p%?+AzhxQB+^cLK&cv*Fh|D9P
z9M%N++0xVQO$wrt<m7mzvSw3V^X4Jmv}+}zUXs=(n?X2$wHnmG%fR76KjU~T-e_Tt
zq2fCnyY}5s2mUy%b8L_6lcqGU@|`;UJm)lm>7GYlBl;b%Hh@Z&SE9P|?N3kuL706y
zY@>;Lth9yyOqFfw2n1}LO6+6>0DK1>HMRNYjs|LGQA#Lk=r8M$nNS2BmsJsynQ7!n
zhdz3^wQdeIL;!<*Mmbjx$O)e?>yy+GgIXKpSev#xvbTG~DB*<Q&T;Zj?|zIZrK17O
z{nvpNXX8_};C~+(o$*3E?BF8VB?Iokjo5OK_$=xYGZbnShn|cZ6Qb+ljqmLnPs6wl
z%9Lw>;dsv2E~*VQsX60pjlYawlS(S1lR+}pV>%n%d+F_pJ~FPCoAZEZL&JJ|r#?OD
zx}-_Jc>brp$gg?f4)vZh!vs|XhMTQ}fOiNxO;G%Ds?(&?Fc8JnD4*n^fh<7U(R*5g
zzZT%AV=lIAE4ni_YT!{wUg}O{9dl-jR_eh4Y3QIc>}De0UBuFlR~Bwi2hrz`&-6S&
z&GI)ksD~>p1MN5#F)#@9(xdD~@3oCsBKtz0yLgnf$_-Sg5^$t*=9^i>He&H&Ra~T`
zlw)DPD%=?Xi5HZCT)fwjQ6KNB<BL<lWwK{)5nf?hL&ntSd1Z><=)4lHG8(GSs0ieF
z6cwRFpjWi36{A}~<18Z|RPJtv2Q)nI4xc&x;L+F5t8?O=DqR2W|F>?8gXRov{BOSw
zOcpO24tCLXnRNvQKsCyN1^LeAQxZbE>TT|~1Ou`NXvKK=&+IemXURKLcN`LduPsQ$
zwMBU48*|R<T?3*^qwHOk#80WY6*KBsoBcL<$Roq?2{z_<zc&AGEFS0d%9dUMt`kHO
znw0O6-$Rr9J19m|BD;RUH1tP<4cC&MT7@<B9u2HYuX{t|?i00hwe_3cm;K|Dk8%kL
z*+Zk5cbQ0llsph;M}j2J50zdu;toNj7bR{Eozp4^N_E4?C9hjZ&REYo;n5Wiw4=;{
zc5<$^e|l1B7W1jg$;tVTzvrJ~qKqR)O)xvrgAG#P2;29o(Gx6?J-r4eQ&oD8Xud89
z{BX800s$9?-xow#GA;*<@lHewOEn+aHVG#A$RU#_cnxuLQYb${mm{5Pcu=P<=Qo=j
z6H@aD;9ep&eojt{G!wDF<@<a*>0tG7-TdT~Zl37%B%}?@S<L)l_GBOF-^k5?7tGUP
zhwn~dWvF(jP4D!v{Uzc=o0NG?WF9eRgRq^HIGvEG<$&3xPyzrr1%W1$g~XML2(UD$
zE<HO{GSA!+LP3i`1qCr0K?Pazd$Ae?^<^NGL7Q?hIQr}vH$vIcJ7AYce&^qGCcx4x
z(sZj_!(`{P7muJ(IW{<QCOh7$Lyn_P?j_1HT{y77=qP?!3W`NFJi9IR=uwV^1mw3h
z+1tw?O-lRvDP4p_rQ^bX3Fr6Z|DQ&#dKKV6$pSUiwmxKI1ug(!Ccbh%{Sg1?0P`>B
zzTpNg9%t5{nwi9yihI@Wm}NJ=YI8DDRE5;XMux+q0ybO8y-$>OoOP*!a6!f)<}d?M
z$cXcJ3T^lxt95n0mC?udoRUfIRu9@aN+0}1ET$T)RvaO$Ep^90rt(D=GQ!j~RBnhH
z4>yTR{uJ7NPVUPs6+}{NOg)NV+L=VCseZIGt+8WmB2sS$RXyJ%@`Hd{{ZZIA0U^q2
zkNu;))?WnnL!Sl}_@<FB=$YgEH({lBZov}e<-aan&}|>mP0iEoU+>ybwrMIsdYY7y
zac?0&rm0#HQ!=_ofus){e*y95_rl={WPhrmB4Htz>qFjg^Ac*8f`<xdKkr0D9Q!5;
z14`;H1l--V+q(HK-{1cGvwx@(9xgf_qSHjA28y8Uy!JWGk=lbKF@jY4EDnobRl}KV
zYRT>~owJ$Qmwk8QzbqptCBOnghSCrNRH`+D9#|jUkf|EYii_NC?77bc33<uOa*n|2
zPy|if8vEI6uOMc{Yz1vAQ8pD39cP}t>fet6+49G~yUCooD2g`CbTMhPvDCz=82Q=6
zZez$i0YB4&?1HDMOT7(HgyFh{sh#gzzU8$_m&>;nVYuA1ItEGhd^@AIm32AgSjLFy
zXhPa+d9XDq3*B4dW)=A|hBFu!%yPb|QIZ47(EorA9jJUQ$<zz2tt%AL4z$n-W};j;
zFTK?Vn{zt*G#}f0?C#C_JU$O>@2o8LbU3gd9rrC#*qEJdBg_V%zm2?_njE=b=zDbv
z<ansi<c_&S-tCzE+2&^&Xp?TZu4Cy`zoCGiog9XmMD!6sfeGZ^ZCIm-{7fAE=It^V
zuNbncX;*@PTL6TjEbZMTCfa{<M25gJW+n2){R*AVrShnl7)_wmPAz*8#4EW+D!*R9
zI5&!<wp4N=n2*GqBs-y+pPs(w2PjH}<}*m%Dv-fl7^Oj#miWAt3WoTknI%~rvd6XQ
zZ4}D*E!1|sDq|al(evA|o{9Eya0ZPM;Bk*ypMRJz_Gs8I!`sz)<=!aAf!vzIHkHs(
z+v>!~LAfWwv_L$~OKt#o_qIndy*7*mc|HwLa#B%A;849eKI=lfH4I<OJXOm3Vjw8|
z9?AnR7tVgj5j55iOj^CRLeF8}aDGn1yHPEfK2a{v;hec2wBK1tTkcqY;*H@XJ#P7I
zZMEEibkb~KJMBB7%sCoYVx_QOHln~Z-M}D)FX7ZzToh<;SE%Mx8k5fIbw>V(ek*C#
z?7Raov+BlY+fV+c@_x-jrjVgLRIRWyfSd-qIL2lrvNeBnivo=jfso|0q6&V>Z5=Gq
zs!#1`to5unvHYQXup)vi#s*j^+XcejjmMfiLNHIFWc*ZxeCqxDtaj!+BnNS}K<Yu<
zbYOegJ23X8%)jd`zM8jpqT<BJpV&E8b>T#S!iS0R7jkr+|CsYt4MYp&1wa06w3PdT
zOfsLX?7GTVFlBjWnfm7rpvz1_#nIf1cFIKm7nHdhT6cbM;tmGK|M0~J{|iVQYSRWM
zWlsfbjs|OKE(y`Q;6UsdVum$UT5dgXPfNHc8ckwq2v`p5vy;?9rG-q?6)z_#vWd_~
z;E#xkmUVo<BHfFq^++0OPfQ;wGdcBqvoks4>9EP8^*t__nVDi6mkrP$HfU>c!e+3w
zOJsa3jk`I1a#Bbki}t}J!Wt!{SHUmb`R<X}*Yb=%i}1yMb{Zw9*@S2Lcy&8r<$+Hb
zd*$~Tth9P~F0R$wNG>bbt+fPuS|0Z~x8BJl`b`;RBC1PxQ@K(ZfO9|f_ujexeU}++
zigr29?~nE~OIrMQl*+fUU^}}QxR0*PW+5aCc{?G?dOY!8fU@o$LDD4f2CsL3M12Hi
zTA{90Q;UC%2FvtG!V#Q6Y+{*pt9SU4H-0P5U@lz!fOLmmPG2sJG7>%o2+5KqLU*Cd
zT`J&A@UJik>$SFQ)WqXUns2zT+kwrz+}ZG0-f?J_UaZ+AWCsQAQH%NMI~q_Q;1`_T
zeP<#eubwkw{>a=^<BF(tz>sGzMd`eEmlGCe-{nArOn0vS=(_X#XXFgfzRKnp>(s)5
zcIGm6&SDro?FK~p+8P3syQ4u=4!=;p$vICDCCRTrJJAQNdBUMl<#M!4f;s)Zu8Y|k
zdvzb0Z_R%G55>zYT7k?_SKIy5R6bo!r+^KCCgNYjaHjEqYO~uIBS5udPB;k)hSoZf
zTavj!YJeKy0wy;AWe!Q>mp$p1eF7e!I*GYU*gh_G(v^QfvK*dXWVcYK`dea&knF&z
zKtd)XlWuH+Jwh8S@mrMin8HyIEwGrP=7TKxO~W<nh%_RQBO>%SQgdn>HhS*X<ck&P
zz?=t92fGDbg{NF;DR&;&Z{M2RpbE#y%v+h2gPdRIq1aRCGyXXfP!&A$G4u46xd(01
zka?ZSr@r8i$R?jVIr61l4eK9qb4j^0ct)R|2gM$1c1i3?V?<2o9&;UlVtOahj(zb;
zXCIXLsmEM^Ov$O9F}@){0;y0L`T=_F)<pov=w7nP-4{w_Xf?1Pb^EIbuMZwKVT}6k
zN=k=%6P$&Bn~XTgeNYPsCZ!ITgkZQqiEJ|8XLyGP-WCJb#7~5d+N~xaSRO(q>*lHM
zk?J^q^rfMfFuVU=0d0)#Iu{0BsdLcPY{T$iFY+?PvXu!QJGWp1dr7p5n94)~C(oJw
zi-Zk0+np0hnDbEI0N&fGU9A&NQS|9OSQPb;adCi2S1~1DY_IK>Ou=tG)XUX*_S?2d
zvGf`Oue2u!>ATZt&M@=i^y+56eP`_Ls?K)55;uS1ZU8r^@TrUeq8Wu~2<=L=-7}^i
z_D<{07k-3-LlP@1bdCM#Tmo)2@H2TBg2geHPoutlOG0aT45XTTPw1oqRut$H4>@h_
zq)s|~U05aq0|2+k7xtonD}rDP=3qFmn*)eV0G7`Km7wWx4Qw`>%)*0rns6@~pp`yM
z<ZfykkdKl;jtsD-!yL*^WRl~3v-%1moi+`Sz(nM~TnrB;7FVoDbYuWXHJGV4i$#w{
z|B%u^k-{n43oFcQkl`d|9(yuyhoh)Ks>xmpTNn*(Kwo#VJt|Be@@>Mwjvja&z16?f
z_}xB={Y1&jwU^87RGtB;Y6lg05q1mNWRWDxMTynTh5;zF4f-Nx#_LH^RwlujgAy;w
z1a%uAT{@_=W#PSGj2zowi3Mv1dM)&gfKu(ikURvDXVhh5;4h@_kRxXBC3NrT3%;m3
zzu*baScDS!2E=xQwLDbTr3RMFHRKGhIG_lRr(eG9Gvar0<>!9tP$DPG$F-Jaz$d$E
z`?s)64~G+yZ^wzc?q{RlX0Pvf_GsdzIvO~hB_td>?%fA>L%N6!az}JGSg394T59|s
z;&w<yzI&KH1ck&si6?czgiUoL)0a&!cAnPWdi3t(z=l7&T}L}cP%Mn_QZm0eb!91F
z3LZW$I1#~5Zyv!S{3OlES@TZ)zkEExGVX~cFA^gUami%NHz1}dBC3N-?wQf_8wD9`
z3|c3P3$ZqG@RAGGr(LgUfv^F%XHsg`&slEfCC(t|<G0!Q9jLwn6GEsjeVx_6bMWz#
zP@RXqlFUNwO49()syNl=9QFhbrnMiJep}LU*(5r`uBC(b(dzkUVo#DIGrV*5n6&l%
zIvE^qn*UA-Jw%NTb>xXEXK+0p#f9dWb?=O7Fy?v11<zqLWHGOwNX|aszg?J?IWnGw
zI{fYz9qEqLBZ*q9IayWo1WKeP62DA#NZ}PWT;iCsyhZ<_JQ>EoqSzI0qu{d+vJ6&F
z`x^ZfRX1GEk%A}Pkn85|C*eYRp{4J`h}~(uS*$^P6SM0Z$vf*t1=@g~IF+c=Bs!&o
z8FZj$fmB=~1!|Op*0M8tOz-wb*{@)$1pI{baoillv#1FvO)5QA@lNr4p#qiK0j7P?
zg4*#&@Tq$=a5w?Dx4x6>6^%)F%a~Ve%$sIVdBVzpj<}{r0|xi|?Xn+$rXv8+FUKGL
zbf#dr%1g4HwFCptqe!sxjzRL?-2V9XE4Nr%f$$Y|89cW+Sa3pco<NpV5gea2n5h_V
z3V#{h>X&i&a4}FV5IaPB%isqTDX6+hjw1Y=@Nta8r-4x!1+`!>DkF4(nIF0=omhS2
z!e;{kr4Q-7z6mt3zDzA&^v19IeO#{7XH(;8zUEUS2HZ8pCr3WFFBTucQxeiyYG}wy
z`E^gWi8VmJ^EHq&$a4u0oU9d*$->WDAP%~cV7v~%g~J8~G039Vme{NOOEK$_WV`1S
z+G5)Kp#Nsw`EDKvEYYAF&mU>evdWh68K^oNXBE74ZN|?)RLAnt)J6$U5M9OBFB#>H
zkO?JdDFiF<gGEpUdpiwswhEuA4wetU>UK#N!ZplEXg)3@Cqq`><%@;?yMV@MZ!ZAG
zCZTcq=Nq@<6z5SEtfQ<3u`>K-^_6{oTUAd%da(#Fn(PQ3X%mDH>~mgk-?huwc$V(n
z5Jf^0FXm9)$YxxnCB!{29R`~=J7VyU_JHC8;V8g(b38f+8$bUCKIeFfrP_b8;Qn|O
z_tXl>h*CwKb`?%+UNp8K8w?E~2*mFp0Z27q9`W_UFX!1`&VO!D+}bmt+Q#_09x6ss
zV^4rv!t@R+t&BFwPqtMIx6Q9or1`>MgBy<j)aOg6+MAoW<#phFAvKpQ6KxJT1Uk~e
zxs1c{(6X2M8#>URpqCBWAf|-eoLZqYCmOW)Y=esU%>DH}uwKY4Z5sl$1teS~0Lv+Q
z**06QdI`%<!e~|j*svgc^<tRa3gddl8p?Dks@uL7XN+b+BOw0u5k@emDfapUDuQmC
zKCi0*)TF-d^1><ZrX0U!L41zm$zncCApU+Yhq25t&!Aoa=heFNxbAgKqspgBTe>CS
z#J>e8S%@o=YYGZ77iB-qBY%R|nK(+;iOSm;x#0y<Y>alv?$Zo<3~%L8HNi5?RxOj^
zt2n^V;cO2Z)CwSbgB*w$iGbd7`CrRc!kCLge$){ksIs!(X23U%j&7ojVxlujY~*91
z>eaY;G)a&M0%Y0uQ29VI)=^pK)k|80I~ih(nyUW!WxoqE->Bv&nM~MfP@vYkjG`&r
zuREM+a7!7d%qk8YIuG=i5mSi-D?hmopEHvQ%wp}unZT1R#X=j%*=Ndj${+<$la~r#
zkss`*8Xsw$mV<dd^0x^oIb*xmUHS3USl3H*E+E*=B5q^1^5`Hay4^t3iF7SLmZW=c
zLBsP;geolKAg?TckO3SI60LH2ZC-_3ZWOeLx&1zE6Gq+neMVoke_@gvYLZTJPxHmz
zmTB89_IDFYKtQOvK-y@gg0>GX2#0@C<~9jA1wEcs*d%kvIlz`(>)w?5?f-<;S!IR(
zo)_y@)ikGjyCBCF!GdiaXQyl-9=!hk$m*dJqDX)PBh~;`omtdT8>(#YyQBBgR8DLK
zWUcQ`9OZZdK37@reviRgk#}G~A4WqGPp_=cyZ`r;9)@ke>__emHqzZ(_PaASeDCq`
z3Gn8N{kjqObHH1H?>><}gyv=>?kvUrED&)ROr9DiiQ2hC<W4+j%+H5W6#Tb2Kc?l8
zNXS`)p*x4%e2<lgX$PGTtp9}6Ejf+UHY%*_Jpd9mjFa_JhPv1$v`ey_*5KaAKIbjI
zs=^)>7(`nUvll|J?ZJ<utVw#>g(i*wSIP}4!`%a-#)ZEIk*-hPJmtFqx1V(G>P(+Q
zsqWxVGiu!fwf)gJiat6*&@#6a>czBHVA**1>kn|<$;V%&CM4^fvw3WN(#pTM=ebNt
zt2MzaI@(qf8>431O3$l1$%XWb#AhArhqBGTvWw@(wW^f8fw=rwDA#)v!5Z-g#Lz3*
zA6}u7JE3BFh!80EC4?Eb=2SZ6t6NYY;hK;Pf{uZs;7QLT<z#e7eQM<A2!Y-u-;nIU
zTK-(3f-b1%mh(C4x#h|UxK7MZP0UZVXl)@(6KDBgQ5~WzaV;aW0wi|@`Vd4I+K)@d
z_s7e(%$YAOreL?Ka&a-75ci22W;V>Br_0MTze|*X7F41DstV+pQ$5>T%D&9c7NH}A
z>uJ(uLS|eUO%T2iZ`i>**7q{d50$xQk|~I1c+?$Q3GgENex{7No7yW9-@!<DftfBE
z6dL|-j<szj@{;?nYfp{7%&9e{?u?~0EL9IDOx7De8LjqAZB^E5=tJI(lvGJRPMubH
z`&B=f_4<}eC~Y{~m)9ZME8MWYLP~f&0{D>M0M|mbNl-gHFHuR^Q<(sp9}}qFgLlsZ
zhmJLFnTUeUw!*AKe=%E;PNO!Iu|zjWwO(Uuz7$}HXOFIwp-j=w(`}3u#=(eWU|m_V
zqCp3aG0r9DYV|&0nWT0Ly^%fyw1Jiqu~aZMmFgeS{cwDkS8N^!b#}L$QG>zS(bg^}
zD(@&qK?m;GAJyhvOx$KHI46jBaKIdM-k%g=UIG^WRu~2JEB_8dXqGfyjkzVmuXWqi
z7n9XU_@<R?Gd)Cm1xdoosoe0_h1%q7e+3f7_f|R_BQML{=kOUf)^In$(^-d6TfsxG
zMhdMk+MFAuga1;AKXi1lj6SoQ`VHg1v~Vy|D&Ygr;I;y-F#T{RVj}wmZ$jq6jgrF#
z+%BP4*~p@7YPn3q++$cn>6%=hJE$+;K;aS=gN)Uo!f!7tM$L!JR2o1SYes{50-LV5
zjKpES2_dZCTL?TxB;wNsN#84MKbk#X-~16D@BS&Qw*KH&?l`2+@h-5Dej!i0G2Rrr
z7&JLJF0QJ!$RxuC!Uj8N{L1o><I_=Kf)jyR0Dh7_aD?U>vB?0~I0JhSl&uApHnzx%
zt%h)tdp$VxD>kyyne?mn9B>6}mBVDimCs7CmNR{dUrl1;2}NVzJRl#@m!=h8+~ty&
zj?)~+vku&^hC*FY0fc9b=w47qGmlFP+pmV<?)@O(Vg$|{Y<L=EF6WD^t9jUnBXbG~
zlD-J3U`Z<u5;yRfeCQaCkl(KY#=~;<q8xWmYV`zisJUK&@bN~w#IXD5Nv}rrpP{j*
z$Cp&^AMGnCSKYceSuLc;GwvW<#+oWjn<J=u=i{YUJUn~M(P%QNR4?p}WP)^v8TdvO
zwe@?@K9+A*W3jC@6?8}4tsK0J0uU3e%y>KvYcG4u??3tp&Zd62?zMceTwb7kIB*$u
z2M?wj`ZO8dn`?`j9|idA0zyhW{X>P??eF~5DoRd0hJsc8-gb2lOT!*hs_fmCYV|p^
z#R42tXjSpa{b~W5F<wLyArJA$SBu?1!krdr;7KEegTuZ7G`50iZEyq+8E$6(sq#D{
z7@GDbpwPW?-aGk)(iU)S-+m1dYg*l33=|F+9rEW~Lxmc^0HG9;Yca&U!oA2xP^6le
zlFh1cOF<tEH|_h|%VDxF&8zc$R}e+M*EZLyTtcwZ8t+SdJmhK;ooNvEz9$b9)@G)`
z$;TqRD}cnkLK5sTe`JkO(IBI<6;aXMbIf61K3PI*8P5i%I4V6U`Poao24FgZrvsAv
zc-UcZyYZ=V#ojD$UjCxfsw|DH)Ttw~cd=TaCA~Hsp#X&ty;jfwKbWP)%s{aSFE?*&
zi<tBzjw^wM2pyaC=Nod2EjU~Oy(cmQ@q9@}g4XaNV7_5o+k#f`beSH7bZuu1rv*2@
zns%VZ#sXb#*GW4lNXEf3)5Jriy&TyAh!ZVD93{l5sI5R^Bw9?kSM@|AB*#O@*O>N8
z()`P{K<qfE<A8vCPdk1$R#0SberUBXdBy4=>{VG|z|Mr`=AVIN)}9@XN(<v9=m{tS
zfufuEYr_Xpln>s8(;fW!yY^DGM{28--s+u3@JofCu<xQ`se8E+y{4LGh)Wi)FS1<B
zGY&m>&X;3ez%u<W0V{YfDKx~_xSrH`#|gue2VAz1{knOwJ;piAeBt`lICn%Vj!B85
zWi+z87=L*Xe+h}iN03R=pd0GTXDGCRlHa%LbGumf5w(H2a}cz5I~ES?P9|a{(qLD#
z7wSOYPCAz{hLIPr*z4Xvh)QYKg?0_xoAF<U^(7?O7`OarRgIZmsEqKg)h<cD)>4F2
zK&O=@2>3&4<-!u0>h(UDtgn8tA0-8k5=6sw+#hk(Ji~_t07|b#DIvQ%1L`ZZvE!>x
z##1uF7?$0Wc4&uC5P>UZ2OOzyLz-dl0B(GChPP4SB=-zl(U2tMiK|yqsSrGUsLYOD
z>uxQ%dLCXOBf2`8Gl8ft*JAF?HBW&b+vV1g?a(=?w>@YrnOM?F?!h)GQl~n2nZfYE
zz``*R)yN@7uhY$P)>G2%7<uk1L1WfYMx$#7<|OdjTd&3kXuPm%0X`@k<PAEfYGSPR
zq--sP*O+jE=lOiDeJ4TPB|Q~e*)GH5O>Mm@*^avn!#YGhLLDS>nt<8$E1M#B+Hw%W
z$}rmeFDHBS)g?Jy^92WJgkAe}2QQ!syEgGcZ4H*8l91m<*Ejksi-W!To`7^Gp=lu+
zw-R7{mhY~WoSWg4A-o6g@#<tv)Wn#_kSkKkUZ$+A5jf8DYL7GPusrEJX@)q1ab&w+
zDVww10cV@89gt&knQE5ZMCy?7n0$j-U*9r^#v)jP%RVRlP7lVT5d;<wY>G%a{rZo}
zYJhw@q<y(?-MXk5E2x<Nqr7MtzAWy<(L*Lw{GY-$rH0{CC-IWubmuv!$zX<uRlkGf
zuh@tK<K+~v&DjHq!oIiGO#m+nWE6CU<(Yd`-bR9hnBx=_aryel!suAi4F(`q)ul4z
zYV`<1y8k8!2f%dzO{`TQ8*PS)Ld_P=_mHLNucmV(KyrB(qCpNhPhN#(2GC)D+&h0o
zl4zl&wa~7PXJ#WikY|s!dCgSNz3=5f0kz7$2V^<YNUPvTAo|s`FZLe@mroAI;_6@p
zhEOfvHF;dAFs{#C-Ai&`7t3M|-22}OXjI>M*f=bJ6lEQ5Bbam9&qLb8mRh)3p;?H_
zWXHUY+)aI+k?WEP!3OP1pQ?!uq0<NY@;P)W{mWp?1!xTtD!f`D=P>E;C;5-JdzPSy
zTZ1`Ooq;+7ZBW;OH>!K&Q4V*xH@n*xz5V4AWr{;+zB4$%DGOB^fDVsG3`htRI+jqq
zST|}efuJ}f=_HMry#UV<ljPS<AeLv>v)&To%$dcqx4*i~_R37qk|JCgnuJ&E-O1sF
z0vM<DG3GQf*m$aWuWGzu^SzE=f~L!Cj$VAntqQI3OFe*|!G0GK$Re;63W7-+y5YML
zV9^mE0F`AJl<mK%$Z|phak{~B9AojIbvAM#`VAsAw3Dfs5|$`_HlnxSaS)oC?S8jP
zpvDj^Af~Q9jm{6i<ynKS*wP5yq#=_*4-O;)OJxvJP1mZ4fq@Mi{Hx1gxi#(E&WNe=
z#|?5UfUNR5v=;1*l`J7C&m8bm4=F*n6I8J47XZhb6+kw_&^Pt5vlt(+AqS@lPH{yf
z=SQ3Lk-F*WGGS}MXHhik!+AWzj!9*nn?<%+05)fg2b-Xh9Qswwt$0l@-QV3nY_3?P
zF*YrG+7Ym<19NYWZlgo=YL^;=(7zzQs+bGu>_rcS-?SX>4$U+6P=dto9D9>hW2Hy4
zVcI@k9tym`p;kk1d1)hA^Dr{jk3g1yl<D}3@edw|7zeD{>d|xUEpx_lt4@TCLP>XR
z=Hqt)&Q`}^AM@G&v<AxXdm-<ww5B#QK4blN={2ApdMqEZ*tjrQo2e}^_YMxHjb=0&
zX@Y%68842syT$Lzytf<a3S_VyO#@ki)G12u1N1O(UZ;^Oq=#{l7fqkaF{192(JeE-
z9A3S1WCAEedi9r<nZWEppp+?qG88V{d4?m`i7z2lIqjww1G?6F9Dm;RoDTr!<Pvx8
zwb`XlmtSot)#udx_QsE_5e5Qx#jWH@<%$G44M_P2vdN)NlkNBU%&zyTz`Zax7ep{(
z8+$V{=T8K(T|*uszogx&J3rTq#TvU5`EsLw{=gkAo3solm*vyM2Kh;h!0`?scuykn
zkhr)G!r;iW&H7I?#$&N@fERnEVXDceL!Oq2TSl1aKM^PwMYNQa9AOLvlwNg1Y*XCe
zX^hkBZ<Regf1+$>*1l6#SN+BLVyAk%{%6lbbF)d#lmZf(g6Dc6uEyaCymnp~Kvj@z
ziRZP2aW$<79*AJ6;gb2RJ6KEABIR|;s&TOEEci3<1UdTae9iX2ZcM35S8gBqgEQ(V
z*)1%Seq)QTmE6WtI$SlACmKf3J@hOHIK@3Wt&Xt#e?keTQTlBnoJ_<6l!trE;Z1o}
zF=}~|96!8j<<>@jl2&0M$(a-a!rc@9NgpaP<k~(%Ps5OY2;?Hr#~}m<45ALH3!rtW
zf=P}S&~peOT*v`a#u`_(y(PmYV3TQRUR|(r#57tWK{^6z4Op1~CZi?#9H|?$FWa&^
zY8q@{8Bi%9s^KXT1>wSXh<g~!k1}`kk+Ke(4^E2fU58x_61u<OrwY`-f;%qk4n!HI
zuQ{ZX_1jP2M$f-|al(JpOy0-1VVLXh2(ecqJjuTRjLiS6)jH07JaU+8&?BR+%_I(q
z=yg#5$VYJpWT0MsgFuQ4yB?nfmFUD)Y!oVul`B<@kq4gyaSMV}q4t)54ua;?TzPw0
zDO&6eqF1PUI2U5_23B+&G_nv~Evbo33SsA?ZjCuWE$L}?XuaA|nS|E!8WQe|TBbqP
zxo`V%=kF<>&T6!Y&)WxK0h*6y2>D8mM1IvQCd1@3S1<X9>}i83G}Z&<1A53k(i<%M
zOh;#Lyjp3%7Imf>3hf+EGTHc8drPsQ-1sFpR#8zcX@^fEKz!~qHuw4;pB3DsF(Td)
zjYM{``@o<kHP4T$-vH_jvlJvbUpz=uXfqXiu%;!&0&(a9epBPy5W0y*Jz6H}Q@-6D
zmdyjHg4n{EGkF~&t(@?f6l4N`kO$>c*t>XGMDelww3_hRsfs4FGfjoKa)0aX)JC)(
z>$g%^@&A@$Al|utyridZ^jx#EbWlLi1`koF4-EPmRlubP$tP;`GeX}pje_oK344+G
zd%m6Ki|Yt<;s`!#9N-ccMMOZ#!_*>e0!H`tF`u9c5;z3hCHmL*xx^qYdLYrb`}exE
zf*he;g>6Q)K^Wr}#-=zAWI+SY>j4C2tNRXAr!6!y0>@C8m@~B0{Y7v07D3y5_}nN!
zXvsTv;VaC;n%Yk0!6rADmd+0alpN3<gC%Fp7k<fwRvxLKjtTdhB&e$DM1A(qw5T`o
zW#9n_T4hJ7s!Vuo7B6ZV{YbKr-1|y#>m9pg*&U9Ur{!<LmC!Y=8rF$0Dd*6uUzDn@
z7;&KiK8V)GEA*%fs(!bUj?bHq2E)2&*M!MtjG;TfPQZw(0Rdm*?tjUThY~DyFH40R
z389g74efIxV*zVhUR*1XG<xEMQPhFkjkz+coHy3~qc<5&x4(%<qFs>CZ>t_hitMTH
z8#ywt8v)?V2r<DdVCzTJ<`d;8^3hu0?;e8n#m$Tt&1^F*@<X^~CCwv$;wo>OY(^r^
zvO<?o6B$iZIJdsTP|1oBOj8{a>SG!G59t6rc$uLe9bV`4FWf)YJ)Y3i;ROOSY~jfh
zrGP64>A%z6UpI_|bf6pqi~a)(o|LH5Mqf_^{Efa{1={Q_?OXErs}9EB82o$u20>B8
zpy9(BXbf<;jRh8Lz12tkx*+4}#D!ffEA+T#<gkJ1j4i6M6SYHpobg1gDp(BoKsCQ0
z$QoFAru3laiPD23h=!We2z`sa<S((6*kq`pG};|Fu$WPIKE%}+6&hD)0@?N=b0?M_
zrDIyco)@%LH96_dnOOUI#=u?E8kp5$1rN3rD~{Yl&mpA#H;z=ZE|4mdWO_xj>APfe
zqv^epFL_2*F?>rUQ1wPW+e5L6G!P)je=JtlWQF5FDHK=I@_)=5Yl5$#sCR4pAwkiL
ztJosL3;aFm_5Ys+clwA?!J%@y0dPS8r%V1ZP#*g42hn|%B9(P@3FbYI4ydBqETlf+
zQ*C=7EAK>>D*Be*PG9E$8-1j~KBqS5%Rob9JvoJk7+fplS(k*Q_(0cgY)(x~y^<^B
z>Twbd2|{a?MhFiHaeR>|t}eD&6TQgojA?-rq&8H|!8uu%f3Tf-1@%xw8!eJX0F7;h
z+s7HYaAr*MG`sOc_+`M;3o`hZ?={z;X}ekpq8oJgv9wBdw)E+Fm$q9gpqCutiuT~)
z>?tqS#h){Yeqmd+u$&MTFW=|2?$m_>2~V=ZspyB1<r8L@OE-qZ0m``V|JoPSIQ)R$
z9zhSW2A0zf8$v5)%rPik)0YH1Z`jTgxFkRXAu2`|9lwDQ`@E(v!M}WtHF^e~2+$sD
zCQwd_XxV<75WD6u;wo@VgN9>2jdZ|MMci-L9g5lfl_W-L1IEBcZGPz$_y@tAGEr(V
z>~Ke%e>^m8RpnoV8e|rR6`&jf8!6g6RT05#@%@HK)`}Q=-NM=3a>Cx3j*I9y?NMfH
zoS>BW*=Uo26rTe!rR|!A-0u~c<#lTIlIb)Gf~J;Um#?a+M!Q0FsnwHFQS|aTBz)?P
zhVCHustR`#lqP6Vn=q~dZR(J2@|o<lD6WM<Q74Y}2d6f{RsMtW-ylzJN^Z%G7f}J`
zG{^oX4^8!jS1;spt~NQ^gq2(E5Z+B9+o>c}6KT-jnU*kT{;QO%{AZOSu2j+aoZpup
zRae^kbmaDV#`E=JqK@8bg4$i1V9U*%H8D(od&H#iL~A&z>Sv1P+nA#0g;OS$$_bVO
zQ0O|dqWa<@t2oC@C{!ZSWc}Spb=ccL1Gj6rxAMZ<<`Bt|9gT=UXeVX?Nobe%aX5Q$
z6Euj5ep`$6p?#A4`;(3-`=#vQDLWFR7yqR>cCssqggK1^ie?5B@yW&z*L$Bfv<DtS
zXO`{E?oZ_M*90yWanW?K>=Ts<^pXkq6lpwm2{a$wR0M(MaSp>`3FE-0P>endJTUva
zocvzngH5!iz+)%JZ@_K%<ONF+S)$deWegQ(kJ=9i15~_Y2L|dvaeNA6QUNJ%nQTqu
z$MvG+a<I|?@&a8MMbW1k-Ejxkc|1d&n*R?K_fI5Az-ni-`2b<!5niX^WY+ZQg-RRU
zqgNlNMLUb5@pDLh+?vcx2Gd6T@#U&3WKIi4T^a{EQHRF>hyUtzcLqs^Ygoq{X0jJ9
zU~#l9Ig*eC@v;*Ris)4QNHlErga5#2=AztVBsxpmopfUuj@TWy$+#YAn^fTU&>EU@
z*pnl?unM5s4p#3?b>OU#z#yj)+Zo|`#+IBHbog~jz8g<X((&2icD2%<%ctWzLR6R$
zh|j|jP5rW=P_~4^X{#rLgz8}?dmig!qn=?d6hgA~$v@`ttdWR_uhXf}qoBcYqaAQ;
zfg9Qi*v?!>^?DCv))GP2_JgPLq-T!#Lek53jvOouyE%jX6VMxbhpit|d%O#=<|io~
ztb-n^b4A(#sD)HsErm{{>M7?w`P|Tn=h!2DiIeve>F<aCY_wrYS;qz(zs@70M}g{p
zey!7kYlqUV5E4Zs;?3jUvy4u+N$EBG{0$fOCPETWS6^C@E<vk~NqNc*KBHwr>VrNR
zOt$C%NqFb#X>|-*p^jLIO1kFhI>1GL`cv&TO&a#KJuHczW5*f$3YfzQ*L7)ai#NWG
zCE>|bbb+@Iy?}Y1Bcl}YB+MFECOM|%RFO!%bI=v;c$!pff&<Sa<QWB?sf)xs{DAVv
z2zwE(F(yoe)+YE1V^t<ZmD@4`3IM@kdsS<CPV(eHN<N4llU7R-v{AhlU}xoa;}w4y
z#HF(ReYem-=&mGh!J!Z1(d|o+w8dJ5@nVxo9~eh0E&G*q+&zivjq;=$ZCKnw`pYdk
z@1UH>A0osZ`L~}4g%4yFp|VM2K?tj92L__?{X!?<_96$INskO>8I+?H<z~+?PHo&l
z@w-O?$p%v$uy~94H?G7PMt`LKm>(b}jUs`jqz9=3b??w;+jz)lKnB->5a6Q$$#Dly
z8jLPme=$72q=??~uMLF-OE1JmUyf15M>)<1JLCkAQ~Ue?-}pmUMfzoI-Pm6bmTByG
zI6mW9HUp1y{nvKe_5_m|M^ZGIM0N~ki2%wF;qt=&D8t!!9k_9n-$;h9@{--UIAM(0
z1qIBM9*tRn%<OkGtxD*%@V|yy|HXSx&*){_%L<3c@!D^b)w{SYeP|KmrFhM*tPLMV
z-0dO1q#?7pBY{K?VMnNL>LFIIspvHE#XGRm=SQ!y9!S8YLtc#PIDVb-$#KYmSS`2j
z-F5NAyh5YvUT}EKeU~=b;3hjD6&;JC=n5vydWJVu??8`Zu=;ZoFgL=-KSBpGqIk-{
zY$`$6J^QU%<dFfrRVq+{_rUsNy{wK!1fX>rIv<$>Gl;CGJ9Yl7nUjkg`qS^03i7|V
z-<Hl~*z7%J08vH}O1=Dhd|e$OH-83EGWs@El=t(&-&He#iCtQj7V`9H)&?Wv;m`#f
zT<N&DT_b$2a85F(P&xXm$QKTuJgBorGM>~%X2HlNpz`cvuqHv4G~wn`@~8S$u$+r^
z`hF`flmg5%97!fY1Q>6Cr6*;8!j>VE8dFQzp*HPMJR%>Ws{+-%%hs@&1k@~06uy<V
z@yle~!#)FmasCGZSuNYD^9Vz(-DU&Uf*K7~aoz9B^q8UirgC>DG37FPY$5fhf3Gu7
z^O2evGe^*dV9wsXGI@^@(kSv1XapHC`NVzi^~5c(Rzytj<3w;}JWs|LLw7bJVsUwZ
zh|tOL#H+$~wp~+`_Vq}}m48pH`y!^uh=}+M@(hUguYVDYYrbNO)X;1KxoKcL%J17%
zYXm+g&R2<AkWpu+>cV_!s3@!r_7I0A_f%Gy+&zeK-$o4Ti@~-7%)0ftlAfNPM9(8v
z!o5F6$ou66qbEN&{NifGH%JQ{j9Kf_XPG1T9wEMho_3~{oCY%OIQ{IJ5*iD!ou&ke
zdrIjVXZ%9by}O$~5JHUk=Sr<)^Ly14R<l-R)rbvr3fe2S$q(*UdiuT0MSCL8^FKbe
zkJnO4W44_x_v&45j@Mpi>*OE%=j8T$eT(8h|J_k09Ws@)T`G$zDJ!V`x}V9eSRNaj
zO>k5?zPnli5ngCym&jH;qij>i#oMX7Q3*AaO*>Xx52sc#$T~M_l8~yt0jDoG&(*N%
zD+G>{$+8(|_aNtDDVND3s5Zv+O6fj=DpQn2DB78p7xwPs(yKkq`))QnC4Fo%M4m=I
za?g^U${fqyU)f@>l>e6mZn9x!X2z4ZMf~+<mE-baT%}i%TVr*`08Rs@o#j9kG133A
znk{*C<;lvn5?9hT&uKJjz*WfhE4l}M_0YIuoZS9ol&;Y>&ki!ZyN)E+{L&hqJ+~z_
zHT(ngVI#y!xzeznvk?E3vibg1lnnRq_xE3*ux(kaC7Pu0E@J_bXrx@;xkM;TJ}&x2
zl8PGuR2nB{Dg{tB02<QoB!#MbF|nH-Xi%9n;F>2u_9O7lKY`+Je(pKTgJ-~+)w>rY
z9JO}OnPE&xX75_<{SEVFD8&f``!U2G?&%k}9X5#b5cjnEfao~9U995H2S$+%A0cpv
z-XI)9->9bVO0J3#4+MOP)+IyJhA-ukCA$usq@j;<^Mm@A;(N^=n`ReFsx}_}!C09k
zr?=9te3=k1m%2Gkr?OZ4vLW60-flPvl}vVF?=gmRFq6+D#C7h+NA?JNgVfEgu}0sq
zJruYjURw|I!VQ?1GID~khGv2P<58WY0-Pg7AaHews2$HYnqx-;r9M-17=5nk;34}D
zaB~d7$#Z<5N1=dMMI^D#2N5ItXLD{8!L-O#(=a)*{5DAuz!Tp9i`;qZpm=rR<?^#K
z>To!i1L+2<of{|<-{5C<w$s_)&UT>m4uWeH^K1oR3=za4nv#Rdjwkbv?G%Y_qz(HM
zUa0;j#0KmMJQqs?D;HdO5u#E2^yra>wg40ccmSz;e=POqszIp+*{Yc!1l?z}#?{dp
z+9C2Msc%A``~xP5vFFJpzc^O(Vm34Sa`vmp%e14k`J`MSi(X5vitQr18dp)v-^~&i
z_TAk0=9?m0%0~p(%@0v(%VAz!_(6{nP@<a~>l_P|E^=MEZ+3-<_qmmX0N=ZZH5P)<
zX&Rf&yv>t{9081-xE}7U=6|!}fj_ZubcmLA2m(La3p&bN*fC#LjA5`ErY=nrsAcIt
zWBR^i{&Safg#%>ry7WL+>GI`Sb-EAl1&ajQ?(B}*o`zKmzy<1OtNLs`E?=D@DIp=T
z!XV2RUsex$*BkzFPxExToKba<uW7zmDs#C}If_yR51?mB(Y)a;0Ky%Ye~N42O7SzZ
z-vIqh!B;Ky^6CrW1>UTp?rDBGy~9Ti_i)GLrbw_7@Y0$r2CB@TH^;Q#oS;vwy&tJu
zL)2wdq(fD7-q7A%vDz_&$M<N=!Qb1z+*4GZMAYgviEU`S!)S)Q%ICmxfPuxB;tH!s
z+0Kr3sW5)1h$ISY4Ht)GZf=Z@UYTplF05>6PGmC&!ZseRU^z(VNZ}r9Io2K*dv!M9
z!UOdVeLz!ud8N<>r>Z5f$7&DF9z7aiPha4b`lr!u<p}y<719zxDe%=gE&3aQa&YtX
zhBYUWxx4vif-hE*KX*DV%2aE;p`1kDz}L8g!tLnbwdZu^XCbO*A{s#by&Q0lzLFLB
zM{xs!_-)N}cGH~rD9#BBOF~gc&wUl&SO*blW^u8LrqPnF^dQc5uaPa=u4}frp=;zA
z<w2(fz>1Ta142^l2~#PrWi+tkukV!Qdi*&Wep=b}T_Z;?J@qHfp!#BPHmB_l-2530
z>t02aQL^`E2t~!71IsY@slhg9g@kX6ib*w*KJzKL<=bezNlKC9N4ov~I=|T<tH@a5
zl6v9u8@Y^zi&vvoT_|q%SQwq%c=*0SD)FIuGil5E6oU9+-q?4PHn`0k4zJYB51ocn
z=$c5Ggx-cfE`9KF`K!{+jr;;^7C8fURiM5~q8-4n@QsI|q|0cfYe57(!s`I73LN2L
zV=-ROprD{f>diLO<wy{l-!r7$Sr0$vsVe$n_FH)#w!0L_D2j~K+C+vK7d$(lCHMGw
z7G>SP-@L>84xCB2U_sOIQTRHj{bX%r(+GLuvC@NG5e-c!`>#lnSFEo~`0Wkm|A&%h
zk#HgR0;7^)7o_k<NcdoRtgFzvxNkoG)>F8UqzLbtA?tjbE_iEF)-9qrnOs<sseFKH
z^HPC=?E&TB+#z+7Oge@mHD^X2b{64oa>>dR@=Z==wUO6K-Q2_cu?KGSrQ%FcMOl8X
zhwfVJV$s9ARsF=3#7+x0wAifXNF-%qM!nh@4)tMP2N>=g_5c{3=(+u<*v5}0GmL%J
z!EY!a6@lY)vSZ56Vt2PeLCtoNtu=&5ZZndoypu&jm}qu~E1vtEOW{KKf7Znc7uCl@
zyyA<6CoUksUK*&HQ~_V(;xo(ETi3-mzuZF+=WbpuUkE@x*p|d=rZ>fieIt0>@hnJ|
z8f(A)fFBb3qtEUhG=Af+&d%$Y*E$J{u`_RvM8?3+k?vmdz#-2sQ&3h^E{UPfHi$LL
zg9<Ku_b7HS{-(t0JM23ld-`GpG;z!4c%_6I@WJ=s#Ekml)gT>obg<`8&d@Gw*zf^G
zIw=?agujn3cZ&1Kt~y|D92NV`$D$H=CxcqASGN?XmSrPm&*5il?q>nan&8ha9IcQa
zqB#X=qI_(xA`X%1)`v<~@R^j60CZcwd5-O=#Onlh9)kEe?#f59Wb1H!a&5TrFC-Ik
z(A;bzV?TbtB}=1imNk{rwv#EDWw4ro=zO9BVKAZboPn6<24c?Tm-c8W(5DVN5rxDH
z7I~@Z2F(?(X=MLo(WtwHdzvkqTfRl*&<;~Vai@vsn~boZECu;GJN|L-@b!JTDd|a3
zh;RCMML(;qp933}JBY#<6%$9sOg{U=>xPdlWNR;eym+%Y3V^5M$$WWNik1rCT7`#)
zvuRx;K56&Y;U=4)a}PVcLFohlTrSpYircO(Eq81@yqO<QNKtRrFV$L&^XJu9)IR-l
zZMujZ{@iu1BIeVwAp+u4=yM1X;i!F=vfiNW`mmIZIIoAF@wpl~4Y8oV_)NXPn5Qpm
z_-tJ$qeR6+7_9iaRIdBW+~M4_9@TKpX$=r=_S}jh=gCCde1Kn?`_=T!OUM8C7oI(U
z98rm^b7Kd3<Lb_rR-wEwoEX+L6R;e_S|Twu2%0mqXDj)mh8Px|@Qbgt*~uT*DqkcD
z;GU5na+fdUJ@m#N`0;wiL3m^-F^EL8F>h|u#B(UK*FFg(a=rrY84ohf@Yn`u1>?lu
z?dJ(!I`93c$Yz{N8CjIqX*vy{ktlofVP2h?7&}y=yc|y}+;~{^ATyO1tG*6*z^d|j
zM8iq_O?Vsg)MBpULw%&i{-fv`5|F{k#yoxnl{-VLAG-m$3=W$s3Qa?_Q6%`_B&BrZ
z;V)$pb`fP}Xk)@R)}fMWY)@}88fo}j=6c&e^|7J%J&hZRD?iAjMF~lbqUPj&_~N-E
z^NI2z+Zt~ar?QsV&w>e>RdSc3*yOW!*FzD9IId!1GX0??x2L4hNg}dhBDIsCZoN7e
zeOMI~@|p|F&pm|vRo$1CSHKaWX=H&?W&3`47m3&(gJ;h(ye095RVz%$5WZKYn`yhl
z8gVsAXqdEQQPj$y;4|Tf!*nX7Vq;r#!>o5jlQotHwi}`OvC5e|P^j_JoX}s*7n_v#
z_c&Au=6rEc*a_mZ_48ktV(-1Vv56jTX(nT^%MQOz>SpcWX9{+Y|Mszedgyk|++~nV
z1`<5)sl)byxooqwIb7*dCab57oXYR`=c`H_So16jRbr4s;M7494X0SZ4G!rOq6iKV
z@-in@2lXK!hu3uG65L?&*=giBF+zO7_e4Q78`z&YAaEeMljOwI7LHw)v2KKGVH0F7
zEN`u004wqk#}C(OV~tnU(9l?>5lJbe-Py*6syOxH<&>t{>$Gb)H?-ILNU5ArH1i){
z=ma+3Je|8K6HJHUCmq{Yk|!;)i)dl`Aa^$+<ERVy<uaar(ferWdTh%)9B-p6M?d|n
zHatW!@KPMfE?j=-%d9u$m8uL{B(rVctbB0sJU8p1>e4xLA7yB%o2VW)PhIUIfDW+r
zZJvL@cHGFR>`v*JFzFlD*e?Dc^pS%Y0UfWI>4*tc*G0qNj2>B$KO5_YGy5rlHmDZ#
z@JWrD$miV1>n~UR0jU&gk;D$e9q6CW?$F-PdF7172@>(t?}u_g$6681%$uFy->3rX
zwLyi~vTwRfzD;~~x@rtn+qT2JSkk_4g-ysR&MW%B<G6H#!G0$9UTBwl2_yFT%Sn3d
zWue)?g6zD5dz)XLN{;R`1`OQiR>_w2YwWUMGT0)bqR=>$kqMP|>_!)jHN~I3$~5$l
z=%aIpajht_(|Sf-;&I*(;Y}vlwqgEGiyM|yi{I0jE5$x{oi#MOeli4u^H6HiHT9&f
zG-h|;@+aV#TPhv~cq&R@1*9Zy2m(`BN$r#^&4b75wn111Iu{0zIXoB7I_&Fx4b{D^
zLLOfuWV`y<&TQNzbiE-XbKb(#Lr4MwUoQ{DJHwJbs4os8I!R5>aM!;}<#<#k)lYv2
zK9jzj#W||RY^}JYGa7#pjmgouDyhVZ&vmo)4r6=VV@DPfv-d+5gO>W{hsx`>0dHoH
zN#lG_pYssA0?uQMYeJ*}_>d|-A|bR0l2t$o9GJ=&)tcJp8t|9b_(tf|dD=T!<GXz;
z=suGVAqjLdUzAH$IDPp8=eMVtj0E~TY|9Nn?&28++%5aCBxNH^KkW_}2QDNN?G<G_
zEuYk2EBI`^l<80Y&Pm}jV<UVxx2N!Kk35@{4?jKe?Rc_mSg9lp{o}3?^v!GLyy9R=
z*YC)BA}C3^-11zoA=eh_r?CS9iXxZn-1y=s1*aFODEDmK^>@!@Ps@Z0Tj~N4ASz|O
z(vqedBVj{jkJ9{$Kg$^@`wX*TXTeZZ=a&A9O8#>`yi%`vB$vQLw&Xu}AJAbqkL^(r
zRIx_$cPyNG3_LeY_%5>QH4U{H2Fcum%8s619e%rAd^i1(g9(6IgeYr$eYBmd6>ra%
zbt}LK(Q8`YN3xi)!)Gh*p&L|;0$TiE5nFRm+f%0Z*q$#Sb}#jRq`rY7Y&S22gls+t
zr*!xar>+hdJCH99SjT0}wP+;W%6<m%uqS|!Vy<CWa%HHj0>GkmPXfF@qo)%>Ge(h-
ze9<_h0;Wxi8b0@bT3QrkJY|%0B}9Ij1+0$u2^J<hN82qL9f$O5=ct(>{#CLv(P0f=
z-(94^9*Hg9hC>5AXXC`8l2a=Qv_|!nYj$qMkus+Z&X)yw3LWM}T{Q&4U2WdN>u}Sy
zL)#lN31gAla=zWV5;9>vPumZ&Tw~*Pv1xEHc%LkWrPWLZ5~hzBH`uJwC)y#1wp*Aa
zFl4gX`gWGl!{VuXB~f7sBoXUkA?sBh`p||I`Uye*cs6J{ajdcyB82gwLZr8BkTl8~
zEO&GWf_zb&0_6{M+ZPeCz4Gb{r84+m^$jkGlyz|sbAHlu&>eaXFNy!_m_}b-%VH$+
zUuEMUPAO9(a+!l2gs59Uvzzq#aIX!+*nvKUyZ|5<!|$)A6T5}R-%NGn5y`O3tLL9*
zk}i>1Ds5IX%!<Y8a7tA6xqE)g)$V*TM~E<v$cub5aB$!|xo2qzwK4G;y@e>wO5IdS
z)?y#<NwDkqzc@n3#mLl8sQ)>9sTEi2FPy)4l{f#0WiE@=xy5Zq8~bZBR7x@A<?v2O
zHV>~9F3~#}*E(K59Y*m0{%o?nc)dmhsV#B>J<hQV^>|Mo$>*|6ri3M`C(T1GKP!0N
zy|(8xF2pW&B%qqj`aaW^bbX?}GyCypKyH2vRcwTkK|vzRpuPf0f>xYNJsVb3Aa;SV
z_ygw3-BzfT#e$6cbi0{*(8-lc@k895jlO>P4wQgAOl2_n`T>kDyC6(6M8*45UPQ68
z<8^o1$BS|AXx!EAFJ<Rg+5YMx%=Y7WVHvy~goaM%Vhf8E4=8$|TwED2atcT~dtYo_
z98h^LUoOyHKH%SyvnpgS6201sHmrUPo9Aa=hmvQuE7cT7go;1!6KfIwX@44pVQtOV
zOoZT&w7fKtvy#-!yHLQWuk$0xL=MqCO$Yr}Y{m}Y(m$c{&jCvpJUmj6Tis{@V<!3H
zO(@!47GGx14;W{CwSgq-G$jMIen&@W8%8Z;6rsGO<D>O`Ii{<;`%CME<BpswEY&s}
zG{x+6hErDBCYwU+?~)}WBZa!IvPy{X2ADAtuJl1wC%mD@(Y_f<WnDSP;$8*6pVbg>
za5uioIVuON$x3RGVa-WI{7Ur!<2}{UI^DfsdumAwEwGsTXu%5+b8v2`OQ(Y`_T2Sw
z5Z=wd9~qc40_pqUlZL0mnIgN)bFT8mUY`)=*>Zp&dbrZCw{sq|a?FehJ@Re)nXq>y
z-{Ri+MuqmkG{K%-2p{=)1_s7KnR)Oaa}aFPziJ^`=we5wtbgL!?Ig53_EIFO4;MfJ
z`YaR;0g!j&sMbbYiBMaYYvvzQ<7z{IX|98uWWFa3OTZ7)iu#X>1L`|AOn(XeoKY6<
z6XB?+3e#H8ZXt2<EIYL!cf+=sfmC<7o=8=7*7f18w*w61TI82?Kd@)K;qXNG3bU}t
zEJw0O>L%ch3uIe@tqSVZq0vO)8I>ugy0Qkp3+-;qxy?z9tH-0yKI)~hMu@1<Ffe5I
zAgrW>`3_<?WcN)1P0B@Y_euB{N3I@^OoN||d68}pcU9XqQrzJL!P?Y4OQvx_B`*xe
zZ=ruY%Rb!~Q)LYpTn*uN!(+&OOk$qIe6?XRU$HUpT&Ul2hQ!8Pp&8LyILV7JBXvDs
z#lcEP&b1mVpB?fXJ5U2q!<tZ5+hN4}VYwS+fePK1fH!&68id?75Bo^seRiS0e{cdz
z<vGe>)rqNM&=qDz70h3PGj@a*|2-fjYbCSyi5Db<!)Mtsu%8J98>8@{a>u{?Xp4Zc
z{TSx|e;QWj61;WT#yV&L*fV3^xWV%0IZ+$iZEbq3EBQWK$K%Ggqk^vDPm!ICL&KS)
zFh77UVs6!VBz6gt^a>z>;A&XU3`jRU0Mrnx>k0&K`F=_yq%#A^L&Ced0>3OuLL_xb
zf@{@x%~se)EU0jc(h9(1-3D;~zMd#*Wl`8?39XifMp;ApL$o1;5fc&s4u|@l)4UZa
z_y(t_Z}1QY&SvRJG;FOiz30`dQY4Q&op0u_PIAbckE`d5K!gxt`JnM~;p<yg<((;}
zR(zI#VU$d{N?2L+Ue->|-JSHY!*J-+=Ki*;>Z}95s=IIy&Enih&Z!R(R8cBtu%_Fl
znfmf$JYZeDM-lS2e(D^C;NXr+O+LT0#Co4e7C`Z|(?;K5t!wVt<7kU>cA*CmeWdj&
zp*`#?jLN@H0ZFMRlJpHZT26cZt-U<x9rkW|BVt@HXV(pvM>XhRWlp(#^UL*y&uJ}D
zSrAsy+$u}dZgG)&&<)8<c+DGXE0p_GM0Dn1KE>OT#N<@5sJX*c%Kd+Qy?H#9X&*m+
zqaqcWEUl!bvV|#2B*aLYr4U(Lrb5<qtQ}5Kic0%JmQvQNWnUASGE!NN?37Go%bbqw
zIN#6bzBSMDe80cf&p-3bJmsAGzOL)Dyx;H7`x1ZoUHLf16TJZkyf{|Rk<<Lo^;boQ
z+tojO;q0YKdBCWx-DSazSje8QJgxG|bE8(>6*^ajTV9=DI6w_?<;=Qo0ACnEH5;eV
zvzj~JtTMXV3#%msLROVidc3|f0{39QT!#`UaJ;K@7;w~gN1vTyf?OmmKJ}at3>ntq
z@L)ilK2^--Sf6N}(uS2@1Z(C$VUeSFW&#pK$78bVOF1&Zi7-oqm*@+(Uv1F}vaZdu
zvRNFw10aQJ`04A5@>30r;mLnOmAB4?!;NEhi~C9;zaNH|z@=V-F8LCUY*WILeNBkz
zt99&E*DcmMSx2C-I;6xM_(sgq2}0EaO$FC$hZAMOtRv<ur_eo&n`Ec}d|kYvaYsvG
z_}o2N9tg~NZztD1Z*jO}iiIkX+e^zz9wxC*nuN&{<Y)U8S`}hmor*oF4#H8#3*(cp
zL#p|u97^0~fvXp87Y)=w>T!A$QcE%3GV_hgtks?q>kg&u@^I#T0O$kpm2@fMkYVpJ
z0bnhBW(?s?7RUN_paqsnNPD!r_yV}zY+qVp8Fq2yiljYIhGv=}bfejjbdt?}ZAeXN
zwRSMK+4w_Jm0me|XMfxon1F@J2@yTWpZ7FtS{KgaTXCGHs|0T}=wGOQz0WU3lTWwg
z7T}2kYEUH`+O|9P%Nc6LtrS+|Grg1%ktLAm6ge{kL7Fi2F@B<T+3G<tY!ukz+#_dk
z8t)sx4vUC=@NS2;2_w3vJ<jYeF2^G5g&SIJjwLqseai7F$`M=e<x`&xoXkR;Ib%I@
z-u8tId~~etYtHi?BZhWpg=!()_08?<skS=NaB|RtahjCEv)D&1dUJh5^K4BQyoAlL
zYt^G&fE*+O)n$ayTR{4>rinOq>YJ<d!T|`~8{Mnf^XdzX&zMBdjkAZ=KVI(d9u}Z^
z%*Dd_YDUFZGY-@ezBdyjx>Q(nBzNP6@ZfvIa!RWz{$Lk)?*aT{c+eu)TNs9mwerb6
zOdqdr)z~7?jb&TcSZ!Y{V3RN0P*7b{lWN#68+g@1Sr|itf-#zajNw@p-VQ(+TiST6
zy)n6z3SIDkefU-ht(6#FM^Xw!Ul>R3&-qbq52yVsZ*PM&ZrBSV*;^jP=0(+&4VFw=
zS%P|apkWyi_YpRV0?;H$+I9|NN^pA0)llbPDHZjKrd7_Ww39O0Snf1AB?cd$+g>%b
zBWvgVsm0+pp<6UN(cO&&T*nEBmy)}orJiAvd;oo<D3#ZAb+l#pQ$^8(jRzDSunXYt
zJ3^edy>qMp%jJ=4@FF~<di;7(dnV#N+y>A;S>;jF*X8=BKEs!>f4_fD>q5a>H!tc)
zrQwzI4>XN#!ruM_ZPWT1ZB0N`0=dH`2&f+q3+D7(Xf6!5+-9^<h$lbj*umvjz`LH-
z!<b3%TqW$Wb(Hojxkxv1vJ`PpQ=g$C9R*mLo4m&uQ|Rp&U?`1jci@Ph0DMp}cl66x
ztujq>b%+R+UY*NlS|->nn&hgDOv0INfUE8xD~rG@{PuklZTaeTaMIdBVV?*;v<_PM
zu%K_m?IieT=$n#0CWJ6)6J6}w44;;)A0qN4%l}GVi>ORqpi1@^ZI1c9>8j&<_Sgj)
zJ7*{aZCNNC^7WsIMCgBwV*0`|;jH0=9-Os(^stVSD{ly<L`GjETtuUQC0SAi^yxvY
zEGOG)zBc1C?Uys+jq3klE5V~O3d7k~3bCoa<BF4U0BbHkmIpD$R|mpt`|=CduuP{5
z0FvSfeb3k)s8COrr;>Ev)5+<8Q45GRyUuK5X~NcS0PBC-bO{U3?^W&)ZTmm7cza>5
zkNk=Rm7?)oa!P?Uk?vcf`%!xXd^&JiNMdwZv;&XTAgyN=D`#AKsVN>Cq=yzu4|?6T
z285Do5Cw0#gN$ci|3deAx1fAMSllvhBJ><_A0w@?yCjO@K;=Wb(0cKFEC~$|DhJQ3
za&_jRn{`&>XGK=f$3tOS(3xI%Zc1#yE$pg<gwn%wdB$mw(#rBI#nuIlWFY803UG-T
zPuX$g^L`B>Ge)H5B6pZ6gq>h-0ugmGfTcOsBiuN~f`^N{3Q_7k{5WtwS#)a>#Sy+U
zR6uoZ-$Y9YeXI+WE7WwP8p+opL$d0%&V4JnvP(@=Yv*{(XqDfC1Jk8BflqgEr_Jx#
z=Ea7$*rWxz04@qsesq9H;4;oqp>CLpA@Y;qor*+0dyxb9q`tt5#H|c<*%~P!vlIDO
zyo&}4g@`VR9@-S}LWyrrt$^ZMM9(W+o2P$G>D5=v#)M15M~&z@*rp^E<!2~rpVDdT
zTN9GH=G|<6A;b4O(m4{11E2U~9mHy9^9zqVj9hz;wmQdtnA`<x4tO$Vm<Cz+MI9UM
zPrTAIXZT#2JMeE-X}jVpnGreE+UDh#hsc36HSxfhBJuh_gH%O>tgxHz>&3Y<+HP50
zn$}7wX);+d1uU#G%@UonAIX#ozylYgLm2s$HV^0@c>PAv-Rl9l)S=B|_(LHh#13|Q
zTK{!Rxd1V;06Fu`#SuV7K=8`lb;i!sS(;wo$;5Zx|CT<1iw~NGYxdBJKfemGq)5JW
z2rmwJQT;PCiNI0FV`<SeyEMry?qcT10s^DsS7@uaQ>k^a@T+Jzuz2@w`sTB(W+0!U
zqexgJg;m>dQ1ZjT+&7SCV<_fYCAtR~u5s01GLNhYp}9Ft9|(iubkRdr{Cm}mu=8D`
zQ`zi6gvABY;%g7wDipn--tlj(NOFQds|FXS8t!Pmq*boDMzR|ir-p1`r39qOG^>!e
zIHNPl`rCX{OQSNA<vtOWYz=(yp#wC*b#PW7Jspce1;0S8U3Irx6>kxsF5_JeseJ*0
zgN%*h3$%{zJ4GR%W;`&GPeMH;+iVnlLH%{M!pV{(!IeLBjH#}IY*G6)$0}&+J@_$P
z$M&4=w0c}D-xuegWSTpSRSn^0f(6rbSW_nKcxmwscRiCIL^*E5NtERrKkWu4$J+Cs
z4hsGe3%{Y>DPi#1$5oR)18Mf@0i7oQ1$Bp%?>#2rP=2=Rlemi2hKvEZn(xYR%>T@O
zx_EdeK?uix<v0HH;xI8}cqZDl|ImemvO2O_ZvXOpE*q79w0Zq&Ate>hW>c@ZZ_0n5
z1iGXz>fCsltr)n440<n3Fy?RD0WrJEJ$3ANr0qw5M@N*bUlbl?SPh`MK9h_vW1gWf
zEFws>!E|s;03!K6Zs;PX@9`EgGCZ|D`P7j{qb|+^R`;ga<jNU&am+<+H|vP47)}Gg
z%to>-DL15e)YI(pN-j68HiP>QZP?25w5Gwn%ll9GM!B(X0xR|R>6qqnL1OrcQ|#MM
zJB|_^PZVk(A;Ysryd?vBM1ozO;ppVHn(Q5ca&sNSwp@Ie%r^R@d%o7~7yK?mR>dZr
z4O50jK;|wx$Ma~~R^b;r06T)q%ymMiVgUyApyO#bT0G4k%PkfueIg>ce4}^`*ng2k
zfcon(txj&KT)I1mL;kCs6oJz*{T<-Vay9Y{gqgXeI^|o|`4}jeDuzAUNfKp5<fEUh
zw<G6ucX#(;fr0uhH`ZQhN*K>&X}QhsQfqPmS9W`W0SH!gW{IPHh1-rl3(^Ld|I?>V
zZYRnNUcZi=se?Dc=Fa>|a=`s^-Oe5kb|s}*OBp`Pu9S{fi|EOO1x>G^-HEcwP*TMA
zvJ>vyl7b%$fFu9|^K8D3Mg^Tyg~)?#KZ^?mjWj#FhqHFb&{pEo)JoU`TWRb2k<w5q
zOPTjJM#l&oqILr)lXqb@Co~q;##Hz;??z?^=t;Rd1*i2Ay^nZuNK`N*Y9Q=zKblvT
zsetvyzzK`9UmdvjG#6!t_s|Ea=$cxfk`@mbV6!;3ta>az^$=&fb&w)N>n5taf^f;o
zl?I;(!9R1<(6<JYqKVp`nVWVtlVvt(60GIRrt4!*F3a<Mw-*4#`k%GVc9<C{E*p0$
zmYusCzHt=P+s~dJNZoP8e(-e6(&D+q$W5pLOrpEhdisIHW~hKt022o2av@m4Pwv5D
zcZUo!nbr+M)u+qC^DY@6doudMA$?vUpVss7NfNK43*hUNnkKCsNy-tf2Wm&t+!yL(
zcR2}xBUO3+VK@(UvA~mkL&(M2Bt@@gOy>Z*q-iLC$HS0prA5$NSHHQ0Zi%I;9;-kH
zmKT1K<jG1^JndM?ofc+vzUbRDm#|EmlE^DDby1pcf|#`cG}+UirfSty17}d~?bwo-
z+h4ClFfH%&9)=PKm|9EtKZpjrQs>EAj}=B6zs+eDH`<sPaek9gkvrmp?^K39>0I`;
zhVS-q(aRNRo;C1yN)k%ae_Y)UcTqx&1^JjaYZC3Lr=;Zn4sjJ<aP%iO8HJW09rgn7
zrd$01E`U{ws;c-5d6{(+fGmuMWmMx55~uqyPzIkXpR~*CIBu5Nx*YGI>{ph48tZbW
z+~kvht=q8!8G;3@KcF5o0~`rzM$iAJ&N;MIrPs`+ZHsM0Dh_GpJ7g&rIPn`sSvSnj
zOpCN;EaAsgOnB(@(5K;R89}(OqQ&6)9!x5;GB1QanLnsNPGxSsd<M+Qeer439p<=;
zu6wD7ccF%JH_!w-9>W9KyzFO9bVb2qKw^jR34kSSs259EhQofyclsBBhTm4^V>8+Y
z_s9*Le^=}^LVmCmi5SsDS<5}M-B!%~u5DejC9D*tNYwXIxN7~^F_2dIMZGvQLKCor
za=3g7d$fjS>G1&knVldU6S5UnSR8~df2GzO!tpQG<bd}eLrJ$?HRoL*AWV7!4?$W%
zZWD&<m(rP+7y0xT3HGi(^6)FL$FE<!u(&Q+PQ%gjjt-^p8=xwv2yghx2o+tB_A_1;
z@4{9jR;&dIAeLCTA+U#vfl-=^%8S6)zl#(+2v{1xRh7mo4p4<TS)f{5Kl8VUNQ-sZ
zarVJ-dVqY1bA#tevQ&Q=lKD1D^Fxp&@G?7AK3=tO-qxIT0>qF>n{UglMQ9v}XL0lG
zUg1#OlpbbT%Gt7!)^HaB=q^szSL#uvp<&zI6G0uEuqhQV0GFYONA@-L)C&h%a*Br_
z@(L!!k5A%lLX$80EUN~>ZXEWxeeB?c4T5R?$ngERn$q*RWS9at$opx=Us^1d)ir~X
zt4q3aEFeo44foAX_zO_z6$xb6)+HRuQ#2@gUuY+4a9z{Dt1(SvtBh%S0cs4%s)AZA
zV5SkkmWqA?%x>#Lr`Qh*kI!iq8}x2nEI=Fzg?*o}YGLh|mPs@ZLp?lLB2vu*a=rQ>
z9Y$w~I46v-mc^YGC=6RrTMt|Fu$9;&$up7_lp>YmfAXH@s%v@*2O<n!kFxNJXdA(%
zv!-I!(K;567CvOZ6}vwXewptxO64H;dy%<{pvpbS!;*S!p(v$at~d5}to;uB7x?j}
zq(x^7#Rk)H$z<9}q9Zwe7k0H(=x!4;7&XA%M{68rLvsjz7awk!IAwt>T8-dqi$;wV
zWLUVBR%j_Nu@!u=j>6a(ln=1&Iutq_(muMK^9keog=~#6U@=_9hgTGIa+M)C>^~7V
zfg;w@pBRM*h27KSy0lS-Is-CP(8O`R_{_>tWp>XC1ZxL4V>!ZKF`q&&WyjWp_}_qo
ze4uM%-j{%Cm?wWLteh_byGzdT9)q(3Kh#}q9!r$qH?O8$bg(X42`9Nhc6pNJl@upd
z|8LpHDX8ik3RWOkk`Tt%)Ci*7&T04i-`j|V<wmfNg7J3Bw!LQQ%uhV+Ru2~HZdPH@
zie!f)@I#E4SU#ZNhjd8S^F=<=SV<)b-V&bK&sjMf<dD{vb0N8&XZREA(Pzl5AGu%p
z`y{()h_}8@c-NnYb=0U=3#|2U5U`1u-ylO++zUY$MhO#qG@T3ynQEpEk-x_?umw_S
z@#LA-68kupSS-8<>KeAT-w#|-b8mdA>$eRci6AT7u||BgWGiq{L=Vb`zGVFo0Bpxh
z^dU+FZ2?@y72XSzOc72kiO=jl{=Go=*0KBXLPWVx*!gRafNyKo&?+setm5F2{{a0L
z0Q3OffwFG2G?$00VA=99dY)YF;f104Ik{~GR^@pRMx~}tvb;)8OwMbtR(N;0sxR!u
z*;WdJvOKTUU`7TQb*5c9SWhsPYih%fjt_nQHz8Oxtq|YzVG3%yp9)Z01aNKkbWHa|
zYvucm5Z2E($<>W%(ZU?80S7l*>*4-L^=b8ii?nGk`KV3(K|sy_wmZ7pl@jFqtcscy
zjas)@N0{UgM&@J^gk}IV$liMtLY@}+D*KRdG7vW8M!C>RN=9b0zQW$CbKkUO<ZCzF
zqOFNsC!R*hXUv0rbjFJ(S$B93?39~XS!#W2`o(a1SRG8_sDtkf2zn=N{r($aWpZZ&
z(D*v$-STv_FQd>l%eL5@lXKsAF88Feg@!La?T8u~SJnI*lg66D{jdCK=Pu~O42M>G
zZ*;=;d8)0vzh$XDl>9R09v~fJdPk?BP*=Xr>o!<xEA`C#P#A8ocXTl$O0MREybPgx
zRON}lCVXr2hz<vZQl}OU0>y|gql(>1*w>0|4Kq1pcQ@TPH;rG)lF!fP?ZAtYFoXMK
znO_SD153bhjX7?ZWwesDP3))Dp!dVcFE-bVYy--klf;_<0<9#6_mxZ=#rAQw8M5!#
zh5*39Pb+zeXmMy$O3Yg0CaAkWp{cF%%V+B&i@nA3e*e`Q*#G;QbEHKf6#akXdDUa_
zj?kh)-&3yNe>#SQaC~v(U-F-)slL&r^8D^R-cV?}h2;;#$+8RZF(Xqnuk05~OL&{T
zKrQh}pnpgx0yk`q#n}O}lDVTP<`OzShSvy5O+WG18sEbj2nc-3*b2E3Y5cVOC%HXS
zxc_G<Y>#!~XC<M8j*SII+f;{)KAzSi(DjF8Ztcsb80$<UGUrcSJYS^LU+_oZ#3|5|
z=m{*@)5_fR)UdbBKYfcvTJQI&eFD-HJ^0_r-|pm_LnT{$7c*L^jxGu&{nPn)7^@m?
z4g5pu7<d>(cM&^&=}IBV_->2h$XcO!T~z)4kvNh8N#!Y?>NGcaQ?~E%LgtNQ2b=u&
zrR9AKlB%e=tK=~AB{m%x9f)7FTbk=KuaJ1$_x|htHQ3pBQLFk>uQ&@p>e9(ol{Hm6
zwj4f^@r6Xeed1V2%g=Va!Ox^cI#|c9gx3=)@+QAwQSryqnyV7tdaMe5QGUUc$d`cd
zmkhT&li>t3OsS127ua4Mc%fuwRKNX$l^Jz*iF_`J`70hf1&flY1Q`Z~oyhQt#xe8v
zKQ0C9zz$HMiF;cDFXH=f=dwVfX$=Y#P6`B1AiG%{(qgd&^fBiZP;W4*0=sqHEK*?!
zzmJB4op<jx`d3z20zRw~iV1VSg6UgK*1fZUOx$a74ec(Tf34K7vBrmPZBdrSeJ*SI
zg}G=eIB(G|id0lK1*^RI4pPFOie-TK>q?<bn^opE8yiijz!$v(QXOI0noS_$EEl@O
z{MzVjNv{gdAzbqOhXnneZ4p3(8dAVZ#}-a}0+@#HjOJNm@aEeiKuA7rO_}MCI<w?I
zt0sgJwL0?^tItxLP1@b`!M~H!NPrw}Vhp07h~3b?`tG-Lu-AfqEv^8{Zc_LS0u{og
z2!X`IK5ql6y@Ii6L*avQVlwICU0$^12;@DVBosaR38Yy)w%ttjq54)-9(9Uy1LPly
zfdLrn-<l=HPkexVka-g35CMe*Hf&c;p@l;h@q(pM?uNJ6i$t;uJD*Q=z!MTiPqs`P
z#v1L=$OhIw32S;ufow(?@5A}fC0Hw^&&<3ktD8ktIqt*H`iE&F?BzQhq0CiO%8*RX
zgRkDfZJy!>I9eyG%L^d1rY~AN#d#1;%FZNQlVpT0X1fpWK6?u1yri2A5fIWXA+`QR
z&t><zkuXcc;SLv4&M(k?Y2Bc8<kExFnG}%?F9I6eUf05Q6JyAj^?updwCCEad~E5H
z701V5i)Ac`0suK*aSq|V(=PmGs>KUu<<mf5X%&-bQuq4*+P9!@5I@JE3%h#l7IfYy
zE?a%B9L`?pbXRQ3_7eWikar|kXO{Q}2raJ_Q$a{>_e__MWl+$k&J&D6a<pe(@dwdS
z&T+67{((YiV4+d1Yj;MX0(EGb)eAiOU`77vk!Oz*zDu)oCqS<BYe=eBcdIcKcxeeZ
z9zXOl-1$<m1?nmKmEFAyniZVnP1A4F&@43B?Ar}s8rYAXYSoqPg$&`XZ>D`^@XZce
z!k!7@y-{Oml>>WNe(=Pf&VICF5an){hWSYRL{Ih4ja2ht_wG3qMH0h78mkPJR8^GE
zHuvXRI;V$H@^i^9szLUn8gt{$A=nsA`}&{%M37QBI|dLxMTTzWnVXuhY~{CMO+f|v
z5DG!$s)$w-cpPE8#zsHs$1#YvUab`KVQNTbP01WdfGs`yMTdWwojqmGd_U$9n^5nR
z#((jTkA|Q&c}J?|U})|_f~X;UzA<dqTD91q);;jR4E5xJ6bmhpD2%KNh_C-s_>fQ)
zV#DQ%7x~ovsJvM|1!&&_rS!0q4y9)%z6S>t$}Rl{*u*w7z5E2I=YebI&^5=cwr|R$
z{~KPzhxfp1nC*^raq>rOy~V@kf*xZAzr=!D*mt2+ygWC23|J0QHKTd;cerJRzWl(R
zPiaz7ziHArly0zM$)1*P=aBCpOM#UMFbVD#qaMdtMdZFFM}|CB6=jc#GnllZ*ary!
z7kY5}tsfvmDxI&$ANS!o?}NfplF)RC%D+mk9fZ|o=eEbXBtoMPQlb!Knu3Exng&>t
z>SYD~+HmBjVl6Us(hm1O*SZDlL{oUG4tJd0-3v)2Xw0Osyf3&;|4+yvQ}*!O+w3>y
z7Mi*=F8wda&=N*g4&LcSw45kH+sL<NsZD-LWuVJoxF+g;ES1AR@xf_U^ObhsK;ubD
zH7xzq<de)*|JJP}v|K^S*}u%Bm3>30Q<GoIz87gvTQuKK=HVgR%dz`-98}IzeIj~R
ziVh8@y7Y@U-Etu!_NCLEAZ3J6s70e@lZd~;<k({M4h*1hj%!t-IAfjy(krF7GhAa(
zsWc6QgX9p-PbcaT&u1<>ZOsS46+|yEGY9e|g0;9av1jUlSta|~_zWhjat@Y~APk2Z
z*EssvT`T5U5eSc`N|NkDF^Y*XRQ5m~`?kteFoOFdWjr4$EOv~${jmLG<M|_adot`!
zdzt;oCAbXSr*g&}mdNEa=u(oFN5(`y8q>q@#E|?Tsnlq7L()v(VdM_mSg%30>t<KY
zZ?B#sSV8n3A*Hs>^KrOsv3g}Mig^Bw=H5FWP7Vbkc%+x)e)yCM@F~W4{`V*t&pTwP
zaJR<VL84T#Mtjp6F`}7gl;PAabO)6#ZwXzFdC_)2AOVd#=(a}u2Y!}ESr0wlSLzZ3
zVTAWV)pKpPLh~v>wmA{0Yeu1S-!}8b$C9sa4BEl6GP%Ed593VKNnZ=bO72P=!VnVD
z^PTwKThlphHp{N1K6-5>0_u=v3*Zn+0~3@BWP7IU$aaEk?UUZ7KK++m>oOvh9Zoc<
zkq>lTV(1{eXP9rpDg4yBJUdm_mKkz?Nj7cohwqzdrglC*?QWwZixrB+uzpFai3W%a
z8}kB#Q27%m7t$d<3&VmIv(4r9IKBT@aM2D^Qsa&}pc)PWX4`aK5T*%l{dU3o9mk5a
zukeky9lUcpiaVhWoH)4F6pCpaNDXKt`hl5C$`}vam&eXhPxdB}%@NZhXS2886qb)^
z$$racd8B&?OVg4WVQ%(ElWoaivP!XbVwXTAkRChzT|;MfE<{Ofs>DHd=hgE~kV96@
zd|g%cQJH1QXDT(ykrhvPw-5-fO!mhwW7XB9O3<8nTOueaeKQjTeT45_5-@OkR{A~9
zasW*0r0kU_COMb8(q6&z>bGYobLi40)8?F69mmL1xqds7$Slzrvn&s97!6#iH{iYj
zhw7BW*V-<Z3*^Q=5z5|IdHqdW8a0vyZ3=fI?GC87NzP{QM<{J#>wIGCqh+ygGAgg+
z;sO|{rS8k2q!0v)`P^d0Y6LY7(k_BnV-or5qchigK8FHMvaG`5Y&K2GH_z&yR`3=I
zIBYgO<Nv;UIDMuMT8Rt?4KCS*aI<73@7s$hR9Dnka!YEKM@NNKy~+p=9uN9)#_UX6
zi|Cp>doEi$=w;CFhkq?k18KGGwD(x)cjU9ym72ib9<Hui)B$}WBjm{F7fi8od*?Y`
zrU2WXH)Dq_{_q<qK(fP9>qMgNK`O`kWXyv4-XEM)&mHA_0}-@L2yL0+2emyX&HicC
zY!Q{EwEcyp{|*$cJbz=y$uas#I{F0PnJmh_a^a;3bVU_7eFLIhxWFxF#z;UK^4E+c
zQ68M4EoQ&y&(y8#A3PnF8SGf@I$*@17fgO;YZ$|Q3DaWAl0g%~<en?Z-worj@rKFI
zK>c}0L^zkXoMv966S-Lc^~}G97>#0SM6eapgH^`<G4-lCn=ty^_>52kfR}#_4>VSM
zQa;1&6oz`LQ`jHuyeY9il2pMFssG|?-uDM~15^YlZ6o&anfY}mVWuO#_OufP#<V2|
zqondsV%4LzPRn>lrq0Im>NT2xON@y$b&rgUD2FuFKoKZnt#cqkI`!+$>sIC*9{}%`
z>&YzXlkE`*9BOiA9o-rtl2G5rbqB_`ewfSIMZOa52?*DjQIVT2W+ud@!>9bD`!iae
zp5iILQpwwsuuwP2JAkz9BYb5>wg37ex<L5XAz`>7THvAFsXfv2eT(3`<BTsa!yGz3
zxxFy+L<vVc?2rPIAo2TudT71>*3WG?{|4ZF&l}*(2(4>dh6gF^Le><oueFSor|wWt
z{6;bsA+ds9?5j9^O1sC-o5@_hE_<dZWv3qpyTS^KeV#^7xrRBSW$C-p@AP~Q*e_Zo
zk-XWq`jGOtCUox2u4L0>Ye==esDk}#|B3ud-bHr7XokanqR9XIrx!-4YDC+P`VO7`
z?=#?5LS)KEdUG9HD=Yf_AFSaex{MH`544FMJE*e@W;WjZr^~Z=!JMcE?HG}}=L}s6
zzJIumOH)uu2cJhBM1_vq8Xj4?8bd`G5EyK7dZquffofO@*&p;P4=aiP!hN*UFYDl>
zHoeQa@ZO2BP6syJBCo*m`6?E6bGZ5E@_||ibQy_be=OD$Xdl`D{|3ME+YJg?jDk*q
z%CdDNLocGITD}<`KT=*u>(6Y3Js>AK;+g#jA$~^adpQ1}$oTO2aMjeIM%fLJ<BkK=
z#jI1SdJU*xcOd{8KEr+jWDErK5bHMRs`5b|Nuq~9z?9QTD8}GFaxbmRphovO7M4kY
zrIg8E(!*KRMUL$L4sjhZS)JrNg-SXfMP-9N;;2?gZASIEPg|c&(3y285Ko_(qP-70
z|N8!WbsxByjH4Zr3M1H{`BSsT7YMg<L(S+w-@7jTjp~!Xe_{BC&)s5#UPlOm<J(AR
z*J)83HGjV<)*e>v&XI?&f0E5n<eF*I1mQH@1;)D9f1lU}2ZUk>NoxgDC^Z0312&Rj
zH|l@(ySGt9;AZLw7LzkUrbpA_5r{11?PaEoInNpHePqpBHXQ{e%{Kohe~*(x7^z2k
zYq{LS>KS}Z9Bxw(!gNB1OD5<&!$sOg+fH$r9P<Z%t&$Hra!d1ND^*r2FV&jT#0G$2
zmu*|R1z<R}Y=zWrxJZq`zFnB~o8(^1{|%iBh+M;ZZyqQqj$xk&m<6g3_n&A!29yo*
z(eNwM!_wddF+qs>a}in16q7_cCXRr#gu?}G$#P;p`YwX?0{c8_yfROwe0@)W_+73-
zC_x4Fg}6+p0qYFs5A{12m^ZgX=J~@&{)EyGhk&O=2g^F$cQ@_8L1XX>I5wf5Cfou>
zQ8#P3l!%}3h6>s_5m+)(XaHZ=M~SxYs1{>6DBoeUsps?f%rD_q{c^0S-{7;#(=R3N
z>>+l2xS041XN@!M=MW?B$da#FbOVS3E<|DA<Zy3cW0P4kD?Xn5{=M+oYogg7-h=;W
z<CoX<DSw{gMj;U@28XWY$Jh56!$GaLdmQm~xh2kKnEX;!>{T3;wIa_M43ef=n8)9+
zaX>9rg7GsW{}S%B666B1wdrNTluWvwKsQ_l%JpSw5ZBNq7YM*rSspJFcY-*U89q9h
zCHrrg+d)#&(ok?{e-2U?D%C*~Rrq^tPn>-W1&ltnA2!jzXMX7nXoag-I9`^t1s0EC
zcj`3!%t(pas^Zie*i3jpHN5i#c{IV-!M~(z<~LAUuAe=gfH~ApwhnLu$S$aH;u-5^
zKdlSk{)`I=D8vqSPnog~05%@1v@TUuwA#n5M2nXY5j~Iw!k{jE30=<0#|RU*+5Rj=
zp6>WFgPN)nQ*VrPz4N_nXY?O7l|88w)2-GL4nQ|#=);{91aa-n8SP=$nG<OlQt>D&
zAWQrX2<+!aM!?$x9v}pYPR3LtV5ZV4iQ^by^poc8N4f{rZ~?|wo2nO&vvfaBE{2jH
zL^Sc?!=mfeEoMto_*35d6pXk1JecFE@W9V}%m)qKId@YZeJ1lx=X9)gS+D!ds8Uce
zz^ldkAFq^x;0f9|mQQdYm5Lg<+T**i^*y<H+vEm(4dIV~#M8@fk;08mNQY$J)r5a~
zEI6ooE}m6I4mhSgeA2bv$#rJBB7Dn@*dl<ut^%=r6&eaZZYNws84$J_{z_MO<&jsR
z;=<OiUa;YM#z^l@jm9YqMdj(_>XHhj5Exj#4t-9!8pcf4qscZJIlc^tJg9$aXHr3g
z34f(XM%)2rJR4i4*C(AP8yAHTQcR=N`00fM49H*;;2^)Zlz&v)K%rvWtdw)MbF|pk
z{X$(D3FH{rZ{P_QK}{J#@ctHQ6<0iHcJY(;Q*wIBT52zKkM(5cAMpG}Km5;Tujx2G
z(s#Q0{|}4<FgU*5EdzBB`2op7ny)H8JKtUeMI4z}J(kQvUXIXu>dTcOX)VC?aAbF$
zFE1G2V^h&1oNVS{2Gw@^a<{OXT#~iP49C1G6^cfYckhDDty}oqO4I_BP1sow`COx%
zp^q?eFob%E&~7ELCme!hZ!fCWJJCNl4nNbynmSp~_Q4hh(<yxMiZ~Pos^F+6wHSua
zX9jA(){=f&_VkV>lbkuY;NkZX_NgUzIx)!IvWmAi-g&<5*;Ctc{2OJW$b4lSeLRX{
zL^LNkGrP!e8v;X}Cr^)jZ_ifPihxJgP%X@y>Cdo_bTceei2W%20nYg)L1ucD$Wu^0
zRMk6SR|JXL<tD{3T1OO{|Ndc1)*%^jA*i^tZdo}VBn5B{PKWX+!Gq=8N+_JV4f@U1
zCA?xGBuSpJ-xfGhu%_4U8C3QIyDUEYk^<|;4@m(a!wzl887fEKkA^nNTw*=L!445c
z?iVOtBlPm8>!I-S6c`w<8*@HAiIzLx`|~Eg#&Yc}lAi7?c%)kp(-XVE=NL8A4o8QL
zWd?|nk<r7S%%UE-4t*&X809hH<nUz{0rIl};LDvXZPYwu4v21G)<DC3!Gl(|kYKb<
z9DLFEslh29&LhAZ&Yg`jv1Fg={T@=W{d9Lq*Dpvr_|4*RUxvW6+-+b&Dg4$a`r=%h
zGeLeV&U)W6p02Y1m%|4fx{LK^*I^gI2$IP;^JKN3+3Dh#;?5hr{${tP&8~mH8}-C=
z?i*~Sww*88l@r)@xku@8c7ct-zB7xV>p5iy<o!(uPz=}XR!a@S2OL!zC&5YuCZNbk
z=yptoq0_@qDxS~e@bBPB%1am&_x+oF5<Z_{#{^ieRdJ~Psc|w7;bD{I5po=A6zm&;
zql7<E^aoU;ilwPx&lRIH&ht0I7KQmMO6QWBr#xYF<ofqnEzi{2&P3xXdib#Vzlz3x
zmk||=3=leBR<HV#nvS`U8Ho3{vbnJl27e9ctO8~raprhKdARe}fMFPju?l(DU$uC&
zEwSm#R0rvhn_TY_|Mm1F=;PgQ0=EvMF^5LJ4M5k-a6J;gp?u0((P1Uzt|f!s0k?4g
zw%zaVffkfDI&3CX(#BuWF;y}`@e!m|hP6NBh9`FQlp8~3Vfe!08N8ny`)~HYfe7qc
z2R4nd|C%v0Z_J|Zc<UNNb~+IVxN#pYN8lwm=ZyMtKZ#J_M0%--OQNri*H0l&d3sAD
z@3AR3ArS4!3)_YcAnOVumR0a`K654$N*lWeto0^;<&0f`U%W(5UR;LjXg*CJhg;=j
z+%&q}W@o)A6C;$gPEYXZ#sdQR1VFnoJRQx7!WNZFGsf~bhh&J_{evw@q+KRKIH~l`
zyidQ#r%@ksm5*?|O6rP3AZ3+I%Rymql0=DiMZ=kHUPq=`z_n)Ct0IHL-&wPY9Sm7g
z)9I7igA>}WX&=EiulT9pEhRu8uU(|-;oR$}+UzhnxKTFU;p&d~eFAqK4(z!yP!FHj
z(4GENYy;O82lu3(lOU`>KxYh4`gUS7r?zHYv_AZuK3g+V#mHghrDvicrFbl9O#J-C
z)RxD??AQd%JTJ_9|5fWCN*g_VB%K@6t#f!7{$Hucw03ARnK>kd>pRH2HDy?I9M(x+
z`Q8$u%Z>tlSOrdl4``nN;Rg4n{wxrz0)}wtZhlNis6=>>M0izCaFE5lGj_Q1?PVh+
z$(b*hc`C*0XsRrMc-&_TfE|?3Ir2cw#>9}1&KN}aoY8vUd{5X<AW}yGNSak)HUqGu
z;2DQFtw^es?w|8OHBE2maOH4K$CoMb@T3B8je-#nQ|_ZwYHE@g3cV6dr~{mVq!5hS
zCP5l%JUJ!I;C;6jcE?T4m*kb?pz5#+yEz~`#C!lyBJ6*nUq(C>b7C*d1m>rayop&N
z?z&yYG5BX{Y9IuwS5|zsnvMe60dh*CRu<_``2+k6iwCjhoQA%00;X-9h-M3_T5g_)
z(0Ts-;^j`IGuhU9wz12v7v5GoU!En9&r-r|5q2xSDRcR;HgZ;0C1m7CYo1U;6{W^S
z3JBv83@Vh8oEJDM6^)rzy>4_CCmPJ^!`vnNIpla_?8z4TV~QZb*NiJ{x&dBZ_b!&k
z*B+2VKv-%{M3GA!6>QHA?DF-^kXT>c<1loAx3?Yd@094YIDd7(0BeF2J9)C@fvIlX
z5JmC0L<4>pZKQ1>wc`-p{?NkG1Mi<-msTAErpf5+NV=n|;Ef=;y&YK@I??|Z|B7-R
znOt7R7v~m>;0i%T6kMfYh-BrRV`RJH1NJomoYPA(2J$#@8;IHigQWr}`oOLQ7AMsy
z(E^9dPj#Bx4yPD@<RMVV(PyHM3=TFdZt_?y)g_z3pDy}Lrv5h8uUa8k4g{ntPdu#T
z#XFBR#@Ex94_Tk-^k>l(e5LqSU$>Jrz3TPj!pnyyS@=mdq)5qKV!=&`MGkAsADqd9
z;7n<SXBj(%tK;oLqXQqOIOrXM3hE}t#wpDKcg<+E9@YuvrOhA<Crpo~HojqL5xScI
znD&b;3(B#cOVqx*Wv9UA%wS$D&p7#p6ED+Y7{;fO$7wJkN3jS(76CKu^m(@(fQ!1&
zN;Rb+l;0MfHVM2#w#uimu%Z`c=$qZJ2j||e_-qa$i$lZDM8IsI2V}OL+&jbECaAQ)
z&rWb9sfjF4Z@tai>#SKwow@|9gUT}~_0C3_y<fJ^MwYZ74d_S~z*MesTO=gLKmp2A
z;1+XSQ%xgC?p(*2ZHLEbBk%sV8axdo$ORF-#Iec?X9&mV$Im#IWFzyn8R}lRv|?Z{
zacBYbJf)Mq%p;z<^@4K%dNGo{N^;!9wH@_M+j=xjNi<oH_Qo-wDnni9GyEAbauZuT
zI5XOc=hCWy6omGnPLN@z44GiyMAX1*0Ab6$!_U749weJ<)USf<(}1H{+;m3cSP3X!
z^W}kJ^7dO+K@xj+K@e0l%iS4yT*KO)mE9q_E+cknd-=vcqTZJ(r9An5z;+Xt&Wf<<
zO=xlb-s0F^J4TzZX0?7je(TR-e_Inb^&5F&UW?S%TJ?;fJ;3#{D{HOQB%jP&7%BvK
zEpN>kFiZl%Iygg}J2s4Nc^!uJfY+pY2p$U9cy|x~e6Z1Kqf01~bFIsHx)P3zlAA!m
zeXK=M2=}6-RdEDrdQp%(FwsN2&9BQEV!v>c543C5_7Ejj&yJvJ_Dli_({?~%FYNZg
z5fifLXmx=jTJgEx|Cf9n!J7l%_wiGTerVQAVaz$2^?{!I3ZWZcTaHFvX-c3ZdPNB`
z>|jtO-ZffLvMjy!^jBq|v?%s^1S)eLLO|LDs0?}WUQqS|_-}F8bFpEbi%Sduzg+H2
zUqqVE41Y&gXUAw=D09NdcS~wx#2|9`iQJQ8dvX+v--3=DM3tS~sWLEMBbp5a<%wLo
zqV^xVg$b9lQ*y>vQ;jmxG<kH_k~!LAN!IwQDMdUpFIygjrO-WLer{>_jZd76E!Q@s
zOEhu@@{`9|uj+jv=2ExQ-n|=U8>v{6{gm|x$Q_u_{O_Z~`G>_-hqGfWcr)q=u!S!@
z%;HwUtY&y(qyHzAER2@+Op((XGu~sqw=}`5+Zl~?z*H!|CF3|3Ba&_jDVNq=b4>Nd
z5fY4coD#el-}K?eIu(<Bq7Hxw$Jtu?#egFkFkAX&VNye>zOe8Rt*jkRl!APc($@sB
zX>iRmdn&n1jmS?+5Zh4AOWuVdc|gI~ElfRa0ojqiE1o*S6yx{cC0wEa8x_2n$}15`
z+IxwjbBgdGlD-!^TS4D{uXt5YY$vI4gbr#LuZ(KCDS~iDU0udd2CwdJNLsClUcxwH
z+G-<0mgpN%i}5Y#C+uO8`^-7j!X@Rcz01@G1($d@Iu8Jj`dL~~v*!f7>~weVz_>*w
zTC9+Zra3*j^i&4n5}Y8~Nuk0!k9ujQ{MPY2HVfV~nD&@nt;;hg4mv)vp2&11%ZjfS
z-Tbwbu~cex6IeXL$#y*ed_;^n^Cn95&JI3S_S@PE!~xxbQy@9+<yuTio>L)^B?Vi>
z;Y|o&=#*#`rv+^y340PFdvK$A|K-1LV|UtEk>oymVL-uHQ(JcPIbQu=PB-|4?%_>@
zOHP8Q#qYQgEfVXg3Ms&S`DJb?D%OKs2@gy7OaO+jPWd}PR}7QtM_zcC{ZK6((B40-
ziG#SbCUVB8S)2?teyp9(WZPhIB;Sw-<05ovYyz3~p(2^R9g8UFbhZabVf8dA)8yPv
zLS?HIYcU)_3KSW;L2LCU@Odjt3%@nMuCXB;YDC7WQ84GYIk^1AyEiF|SEos(F8-RP
zM5@{a4z66pt5SVh)3tJ5?X1wCE9}aBVtD@@9pJA$RscnmiuHusaC(O2(X7va!aSa7
zA-d94o@@c1w1PeGg^6pXX7eoUedsdjY?szTlj%v)I|TqJ$_aJf=d6vWHZALXx(C1?
zXu+C~lM}%07TbUI@DW`<jHcC2tInfx*dLgj-h?Rl58)Xp)ioI~`hSR7<<uVawf)M+
zyWVD_T7RHE?~pOpow|0oXPUjE9(_)e-y68^$}7ih=#e2pF=QZT^#(RZ-DbW$^eEre
zf?qrXtjTsR(V78;4WKWVuvT71^rtD*>@XKG_z5iz<HG_WY8GS!m#ReWf8e4FA4cd}
z8x=%|@*CUxL_0zmuh(44^MDlW0o?V9T(dN7W*SW7K<Tt|165sMH`nR{R4_BVzjV9`
z+E`r9^&t-8GBB`~14Ut#Yt>W+Nu=U{e~}wg@D`fkQhq(8JJlf@*RUH(5}r=!ACK(&
z-ceQp;)Y{bMmS!!H`@_JxF4Vx3d5K_!4|kfS!Ga8|1jLl_&WQWvHUz9;zNGwK%;N8
zeVRCLxwoedL^i0*99s^(iNac9%O~Bp3DmeO57DnEPNT-ILHdzlb*~Nb!{G&A_QGFb
zAy#mBznF>Lv^2y*MYMM4l;8WM50p$)FFKn!)6hg|Y|Pn?UIL5Fu*<{JUB?Dj4a~^k
zH?0=K$%)qE`!`mpmMBjoKKkrl2Q&Kivb*2r9-A|d9M1-OlUbjf1c>@KMf9jqBFKPx
z^<i>vvduH*j~oCyf_?nlmciUK7EvyqpSpbn2P(_N;W7h=9Z^uZ=SyL@f?{E4ZTs^r
z6D*zwtuSc-lU8saw_ncR{nJkPoD{B5)`)Ie@W)$mnxH4I(ZBpjg<<{@N-9C=*g*Bt
zV1dxl;hc18r@~7~Wqm3ARwb1_R1d+@liPHB(x<uUYp17!&9lT@RE&@&FAUc#*}eA2
zQaIb`P7YY()b{+t!{N@ARVD_nOA;*Z@Gvb&b1f|5Npr2orgl$Ms_n$~^SHz|IyWYz
zjUT)N>UQYG0@PQ9aMyd@_5jRdKkm}b$n33O(mVR_IOazud^=e@NPj%es^WQIB*CT?
zX3QfREEEb!yJDc5*kTLrz#kl{A=wwV^bII+X_dCG1EXmeg-))2CkpqPyd+(<NSt^K
zz2u17`9Z0XLZu~h-$29?2YomE4n`<8V{u>KK$<C(+iUM0qi!%zO%jN8(BOgl0CazD
z{pRHyIhg$f(@z|mL)H-5N&t>$okny8jbu|U0@Hxm54wlt)9C>9+I@IF^&bA6c$_s@
z^?IM}hL$_5N3)9Au37QH-_ABTzYU6Y{2ZA2MRG|ER|OoHpC|(SPCgiHRJx35hY*0n
zQwjtv8~3CG6N<Z^=rJZV6^oY(Zc@x($SiC(&({jLfO{~|2%mzuo$hWU`4sTu9a&HA
zzG*PlCAU*+K<7Zcug58ydEI>Fv~Rn+apMugiLz76Kgs5?w+$YM);{@`^Gy8`pW5Kq
z9AbX}Y(HBb!%sLJdvYlb8O#dLGi!TjcsD~@?a_1WUl!a1M*k1sAe@@TaZON&az;##
z)SW=)i~A1TQZC#dBGBy|3+yBCk2uz+tDi<;0t34?@3@hh?OZuQ%`Zw9EBCMTuKQmK
z@DBa7xE_#m&oaP{mDun6hVde4nPn(ZY*W-8m~kE29H)|DmXoc)WgxIJ7ZeZUF7sGr
zBrj5$O20z7$f<C(tygBwC$8-#M-%nQMgn)*ZJf*QOFaTc1B@c^@D&u64!&h-=U;2h
z0Vm+Yl;S9C1*te9<UBiDKyvo@?WtHu_R4d{rtzVM$V8Sfk-7Xjg<vJWEVhB_mF`K-
zR4i3<{hi;#2Si=NSTUrhLXXpE^BaUY=a$Aj0Di*4vCOf>r<(alS4f<leVw4iHsPp(
zd>J|`NB$XntPlMpxZ&1AXI^;(fk7l)@K%&u%)IBFO9)>_GEV2p8V6#jgx4ytV=Skh
z{s4O$%xXKZ^?3b7^E<T9y3b_yuAe*md;sOW^?WJ;(M9(doLIf)8Qxxm&R_WLhL2<v
zkw-{&edGR*y=ZN`d)1doDucl-)L^K!uVIpv+CJkUhKmH22LQflI<{C`-tJGh&nD_6
z(QlXVGYQ5#|IXl`(YQb0EF{iYgMC#V)l$L7ynDd8;!~3Go4BF7ajNP8s(mbXTSl?#
zW*(L#uBP+E7+!<RGv8GxLmWs;uQ)dquY6zAG_K8Ml%7v|sO29F3c3MRu6b^Wu0~8e
z^wq$zW1t%eB#%*LA<khWo7r}Dyg&Z+yMgVSIo?cEP>A$9n2Rj{CnngT>x5cPhLN*=
z@Og~(vC`mSGUYCUJ!BT<2EUC+3QcG;DIDneER!y;%4)FITIU$I9%3CLCp|^|ULUO=
z?)ZddUXX*G@N>~N_$%zy_J(NPhM-K`ebM!1hOm4<JOaiB5^?vY`+3=qeqpQ+o5~|A
zusc)etSD_h{F15yUHC9H(R{40T(r`0+TnR>pt|<&`R8Akn}BV>+B@U?!Dp{9pb2(_
zBO>GWe9j?o_`M68iZt8U40S+<Ylh*clp|;olL1G`3EC^LBs&7zjf`wW^D(1>*wdX;
z%RZIYCJc^UCWmogs<EDQBxvlr5)Pf|NGWKyi&Wc0H*ht`Mr1|7)ett^=dvDG0pWoH
z*yZyWnjiPB6>inMq7{>fYk;5>kSeI&)~!w3x=41fA`pA(5X!ck;$1Cn-dK|4Oxg4y
zPmB*tYk1zX9-i0;;8}7Ok^KJ9!Ryex@plo7^~uJbs@3$tlb7dyz=NKl?H=aq>;dD5
zs&k08WDq<TkL&k~w%xYM7Y%oXzcKU%8M{jZJAI|Hd=DJnWCVNwsGeX7_v07N6r@{X
z#0vu2FBLz;WvV&$vup<BNQ5&Pg>vd^g<~3uwt(B0T%~nc#M<=)M0|vwnb-RYDtw5p
zjnJtBTM&ijmRP&N5X4J#eXGfSMir_Fe8Z_0uf8n$!Oxg#xW%rbYCQOA-UsV{sCKY9
zkMoBbp4Yo(DSupT1eNS8KB65{13*@B24G^HtwsLeMv{LLzMVuFZPk1*yg7b!R6*ta
zz}z<^-;`ing$9R!QChOk^TZh+IeW_N?Vb+2Zs)v*#bc5;Vm1G=h%mJbLW4=^r(n=)
zlyg;m?YeA;F92yKg;J6%oFfKN<E*|}x<-5d-nm4vLH)diuTuYv5<dW`KU)Ew41FD$
z<JzlXg0pfvTivb^*^}s1ME)w=u@IrF%xsM`YFVbRwc=_M()6o%B3ar70z?Bia)Q}K
zR7EPevyN=9omKQgg37!?x(ED4`?c-W!1>o%zT%O4&z^c##UA?!SQFy~wGH22)s7aW
zTgAC_V9{O9!lFAVnO20ARG9inPX03K-`bj<7<T;#d$bgZY`>VG`_SaQ`_(hf-V88O
zKa%irDT|W)9L_gAyT6j6AiUy|0a$r8;XywXCujcTc0eKwefNNsGz-tVIRmn<G;>XJ
zt`90K0Xi*)rD-c6yV}_{6*|7-`|kavh~oC9>(D;T!W@2eq-|tUWdSeVv@ET1qO6rx
zMRx$Egbm(a@DqsE2aLkksm`gaulf-o`7uoiVC^7t^%`+;R(}qcjl}{kgD3nyisg5|
z8nt@()%VCT2F+%tc?Z4i5_yS5r|SpmfNK@#?vGnbcszmWG*(+9d}AUrATm4Sw?@um
zoa}z6(YXQEt=tGWY6RzIqW2gqy~2dPG$6ndSu&rE;&RB{`h(AmioU{C3wXRsbKl?*
z<mZPD!|Y$Sr(utl`ZBhcYVXQk8pUtrOWHi<A)Jr5$0AZK8<>3CVozAwCzK91I5GU;
zj+b!EY6H%IJmw1%ZF*+&`V1hmq4^O=5mK6%iBJT0Cj<{2g;4U&R%i%gdxO5&yq!Qd
z9}DBU)z5Q#MT5)9i1H#p=q_w%?7M)SS~tClH{mB{3ee+<s)1^fy4V{%FA^rdNqK!*
z-;j-JE!H<1q|r97N=mIy=P$cCaqD|lmZ3mhq`xe7AfZRQpxfi`eoV<-mgI&DuXld_
zVsh6u!xfYZp!7f3V1s>&lV)^a<o~07iMxcs4+6!-DzW5wLcEWe&wD;A)VP^z2oR>i
z%#xwY%*F#Nc)DtPUIf915cyXV+D*c$%7r%u#=>!K2UrzB@rG8ivn!D+kNb#53)~fR
zp&Kf1C*8^PmXaES{jmRhOOy+sQnWS%yG`raW}R%2z#@`{>Y&~=A&~O7a=|tmLZGJ(
z6E-@`{Y&9=0@OkH6*3TEHC-}bY_#BJc&)$~%s7rLphg7`1Do1(-W=w3csz)%ZREWL
zwJTJTwk$(ew(uRRey^SNTF5WOqtQLdH-y(8(T;<Cimw#*d_;P@)6_CEu?>c#F`DYT
ztt@trYY#g)CUFrD<JK&14(Q?CI$TO>Jh^c#JggnsbAxl$L_|)(9*>RGCKSBc&7>@7
z8Z};U@UIj?Cy0zJ4!973if^SrGh*Y{xSSpIkso<i(<~TMic^efI5|J{wne;^jbUry
zX#u?I65LLTjVLiV)beOBO|-^QE&6;iytRc6*t;JGGSx%%q9kw|4Rh7;iOf^#9V{bg
z>IivmhN~I=5qdZ?wViZY1`1$yJWk4%9FM~Rr2zv~VuqR-XShP$i3A2*<A;itT9J%;
zD4NhzWqDE_U_=lB0|Z~KI11@TskCtN0c6Wx7Tm%T<wSunj4>@e=f}YBfUwd2R?+nt
zHSpe{auZWh+jF60%bS&5wEuM*Lna6m1tWp82`s55ZaX9NnQf_K15eRU!He5aAv0<;
zZVi#S4NEE5e|rEkYWnt%Ee=)9e<+s@$~$J|lm|Hhv4FVNGz&bKAn-6!N|B;^Frqf7
z45to?Qs`4o71*O>gtT0DykK=t3mQyP##1XYd@`C61f=EVA{ards+*7ErwQ;&f^7;+
z>&l8BZ#SS;NaR5s#xZYnEM-(Sas00+Is!0ZL;5Tl^EE(-MDsxGUWLx_!N8I%-@N$z
z?H_jRI^+Wu58jvqT89gR2lH5m*5I21_@*IeoK2&(RyQM2X)7jE-A5x!M<}J7Nb*nU
zFbQ!zF@8BkG8?6ucN-|j$mSj7ELmuSY}_)a<;J^?f{mayY(a9wemm(j7&b7Q2{?T@
zQ*}hcUI55iipzX1jV(if(}A35&6LMC{rHsi6@M$J93c$;b-T&g?B0V$S_9v;4(d7)
zO4JM22m$&1A21?!*GlE)Tib*8m85I)<li`Uf8YQScNS!3-(9)I<XFrw!;eUccisA@
zFg!2JlfCiZYsdcW)zm4^uH)^M!!~whK@d#fdPHbf!)_|!i2u*NB37MEx!5#AjzGUW
z+U6}=UqDSp?D~DX`P)yYu4^u8XeD)B;y=B>%e~1QxDT%Apm>e821+H%(l|2cHHT1L
z%zlXrUOIWkrcz@O{N7tY`ocHUM^E@`yW0hC21LEdkbnSoO>>bcbm?U8``ORF<~sCA
zxfkb}8jYsw$ukB)BN>2fTv59{iAZ=27rw>2${LpO#5v^-5<0lR!tqW6o1iI5-aH-Y
zo1%!LuXh~4p9wV*Bgzy<Y!O!+^Vs;M>ueuPtsd3IQ_gt(Goy%`#dRDxoPESHUx)I*
zO3NDxf(xRUV#E4?OhXn)>3czUBAk!3k`~4)bEYcb3ozPIAiMfH>$&}Hi!=*6LrsFU
z51`Ep|0=o@S~;IpLPT&*nG`iN!MTU$sw73P@Edf;N(Lr3Np86u+>5yg@`MBQllbH?
zc@<9@fCZj!@Qi*NXi7BO|BK{SReXk*0XMZc!IyCTWdRy6k09D)>JCAsh;4g#lh$K$
zbZ$Il=l$cP>7RiB<?#q@VtHMaUl6R}4b2)l<`|K03osbu#y(j~FwECO1?F}v>V2=$
zWkvoDxIIy==a5@Wp978*eLGjUs50R07RSsXK7O8#PD(Yo1a1>oLDXv?fJr+>s`Y?!
zgrCekaOaI}gc)tvn>1GCl$ODoqm{Wwp8dw+LQ~&HUowqZ_Fc86nBd%fv+hHy(0xdE
z8<>*7-9kAXUO<d3`bOn{7_Ps594qvi6GJcqVd8o9MK0BZg&s0qjh|bBhwj(x^j)6W
z6@PTK|L?G$X7q+|@Y3<t*|%jbp~b*A`A{<s=rJS8K8|3gl2t#v5oVE4%RxE@CQvrU
zQ#s=*M&jsh%^vOdR&llxo>ECX_aUJEY#TrL%Nw2<zCaSaq`SjU;N=QA6?V$(!8FVs
zBqhSfx-HVDE@C+ItYb4cmwKB`Peh?5E!UqlwrevbQ)v9uYdO1S^IFxItA|p<!FuH5
z2}J!p>@Yt_T0<0r!GBvP$_YR)t|RQxIJlWnT4bRRK4QMgv3G}XUIO^fkraiR%gS?!
z<)4k1w_&$I-AV3Bg!EyBALZMd`53iUFJ+d2vwE6w+AM!x981pU70JcL-9OS@^WP_l
z#KTCp8zn@<SdW#L^{0~MN)&IeJLK*J>M#nB8~8o|(bKkVb&bz87{kA?nXm!(fwCWd
ze9--`TtBMe=yQGflaORYD&GBls>=>u96G`LUkV|+hUbfJl8K71oD&JAjVKfvEfgo>
zmSIQ83@m{dX?7yGp{EWw2-JJ6&H;b{q_N~2f(QZ->mx=x!^=#j>7R}kcbdUG`eXop
z5kglps{Tk0g>JOHf+$!#dC%H5B2yb&i^rV18w)B@U+w=B_J!6hUrs*$CYWzCHKA%3
z%eiT~oLQyK&mzr;^hgK6uP4IRMBorPl+o|Mns%p1lpRWzgSY@55~Q5DS#`kAXvbnv
zqtqX~k{u+tp(j%#1#?))Gpu&|n~Mf8;J}@ly2T??mXCAf;ngGZ40E_dZN=zS9>Ua*
zDR;tdUgb;ui);(~pkk%bJ|*qa<Z`Vr{(w*P)bTd@YJ5z_VAHJ^5s<>~YrdR94X$O`
z)dmNXm6w-!@lY2afYqH;`6K|-ZA~PG`1IO*2vRz;kw{?r1KIAGMaMce`Wr77shV=e
zf?(z-&A1<wz2?O=R4GoXkjls#iL~#zeA&3jERhe0mx&-4UnM7JzWl{^2FH{HyYKc@
zV<?O~j$>?$XfNxCp3hC0vhm7{(>UF!{;29)GJ5CWmZ@#sdC%We3R7I53&^*Z{8Y%X
zO1^4tO4o#%?#WvL=ypU&eo<P>|DV722{ghbRYPUU6S+NgClz*I^uz0e8~cXk4q+Kd
z#2xG*{!B=!mMv=C3YVlTlIF1KP*dqU&ajsd-Y-ZQnQPchz_eewlIW0B1!hUgKsoaY
zpv>S}E;UtE4@o74Djk@8@BW~+Mx+19)TS{ZKMQgslA1=ShA7_E*HavvpK-Faa;*Eu
zC47Duon8hm;zrK<@qhWogPz%5Kw9Ch($)WXvJ)SitvyfKh0_T)_X69dNln=MA3L7E
z(SRZ;`PmD*OF?x3Fv`<dZM7xSIxY8zg<OY3m_sKi__j2tY5h(k&-}N@gu$aOlSDO)
z1mE2r<++WINK!igx>gN-4UB(Q?Io`>LfDR)Kpz~iH-V<a{}L01XZr#`A;?!8LFOap
z5?0q3YAIn46x)VSA|5Dzzun01{8PiKW4*CmEL80@i&fl?YTjcy7{heyvH$XwIhJ$l
zx#G#}$L1=oqVq;PFoIF#o=whN9>TFO!R}bqqH-fCErEBo|5@hbEd|smZw+G`Af$_T
zvw?HJ09JNpC-#g|2~N-V9~h3ZEK^eb*o%uyl&e<hbr!bX;e>m>`+~jFU*Oo<&g5rI
zVp&jg0=2(@GJ<}4(Sc4bh;*Rgj3Tu#-W-9l1o%w$58K$6YW)ZbKzi~Y!82lxx$Bd2
z$s~%dKp>A(b0Fw^N{0#s+ji*?)Gi(z+W%lmr0zWCT$S@!?}l#jtCs;e?3>6<lf1sZ
zsEzwCuZaJ#9`aZR%88Iv9$zE0HGpv52DcCTk)-zzf4T=<OU99PeLV@;r^skXUi{M6
z!(J!3I)dME)k=Kb1ZYLS9c_crBN?IQ%L9(W8#~G(23CW#zLM8I5+qJfS!-rZ_DN+H
zTn7No0@<>X)2<NVltu&J;E9*ndm-%(Wv#JKbM_7q3j<QDRjwB9)zLHaV2h3`4xS2h
zle%8hq`&HLT3ev8?=rrC=_%JTR2IAwy~BX2zZMR-5+ih$q55+Su4@CiTog)e!OJX*
zN?dJMkfwT5s!)^b{Es}5nbx!Y9h0A_J~(-|U7Raf9b6~KaR0)3bl&RmQxCEyjf(-O
zbFKG#7aPx0qWrE2npG#d<cS7L__d*j`$dcQU{7{o)`e)?3l?Okld5QG4nEs<MbGR1
zpBG{m`%>|e!Pik5<M_qcqxKxzC1=ukhTa$b=qpY(rQ%rXSaxe+&?N_Tv{ruH%A`5;
z6ike>CE-!y(y=Jfx^+oTE-oF!O0|w-eX%&jjnh&8@s=7vXYp7-eJihAo%fkX3lz|{
zodEreOv;^D`MmS_fP=T0R%RQHBgssb$Y)6Gs}An*Hh>R66a}H-h0Lt9De+JKEtsQI
z>!fy1m%if3Aav<6la3c~D2SV|{9eawpL3k5{&R$<$ckT|d?SsTH80Lq{LySM+NWJA
z^iBba?12r91<y!>Ww>cl#=%mHe?|ClxKu(|-!!RMzDFAXyy<sSg+n%4?7?DepG_^v
z+T_o|wx)eTx?^Yz=CbhZ*fdh&L`NEw2YQYj7yhI$`YsO43GG;v;GEYwBtbqb>j>ok
zgtv2se-S<{_YZyP=+ue@DPq5s4}P@6a;)6WakuC9+|IIyAYk1l9+LOL0CP0r1uNPL
zGemRU0nb~HrV9vrC7gno%624~a_3GKS@GK++2;%(>+|I_x0U%PM`?#(R9+a+(hE_P
z!F0;Ke|n1G7`uwsl|Y~?s~a9m*Xtc0Xf5q@vaHlN3&Itf%CThN{QNB78^yV~yvL6C
z;J@;se-}4Ho0m}0C+lL5CGb`xtzj&VRB{)AIY#NEttVvjVfV`5KZQ<fKON^><u8Z+
zF~WgE>uO(U48NP`Qba|B<&ac0D+lWOz(2We(=!OtPhLD<Jc}9Yi!Qi{w@3^FIRTl|
zj<lA7L1o_lb@s93owz_>>PCM?;YtuK==O2)ZkscJCwn-6mJt@dvv+t!-4sMQxE3l<
zLtI|o+%0#@h7o#r-oW}9I?{9SLpz)dapcemK>s>JNm&smkc-Z=jHr{=*BgNIg^q0`
z@J+T$<)4-W<M_S@dzV!5J;+MB>wl3Q)x+b#Uv@Y^<%ur_qeGUegEO^cGc+R7!>ENA
z=?jmxvQ3a^2tC)~PfqS=J*Q?7S}f^st?Imf5EC!_q2omtDz+((g-i6B{GnA<R|Sqj
z!bH8#0<P5KbBPBS0+wGGs=L17PV=Od3r^P3PF}!<svBl8Z7R!xeneoJrC|qmh<OXd
z(zi9|m<rDtkD|IYL}Xs@31G@$AjHc^{zjN(xLONutpz1rSw)UC0+3+qprDBtpo&DQ
zJH`%*3yYIC23WGylkSf9{URJ_f@AS}%cn;D@a4sIw3Q|2ZE@V3Ue=zMAJa3V1qO{&
zVb!(vHmULz+XUZBn3|$WdGrVw8{KV`#yO?g4XIkjdJ=$HNL^I48pNhpWc@3vg0P)?
zP@&XFl{?6yYhu#}*XrVEu(XVMdXZHz$8@StlDq)WLscCM07@~j_4u}|-{+E-7d)If
zzkgOMmS2L2z1w-Mt69D?9K6*b&dPPEb!MwX&O0m#_x*JCFcb>E2zyH)o=H{{nYEFd
z%12qWWLF9H-B;VAMknZA4w6Lw3Dl;G@@>0E&I)}?$%hEtRSM9A#Xa+>L{bvU!bT*(
zFwEp^>j0ToX|coQ^wL)c(MHR_qRSUu{maz?{-$dT|5CTTEY(zF#tY0)E(SRuGG$LM
z-T0se_tk!cIx|y6WnKzFk_55fPy7$YV`5SJ=o@4te+Q2_1S7ZOvozZu2f^O~(oRuH
zX=F2sxHfOc>*A!RxgV!u2Xo)5n7pHJmveSWQjDuPbfw7k>{p9VD$SPH4qYxPp3^M<
zrKDS4*>ucWT6e>v>iLyo|A(wEkB55u|DWngSGQdEW9n9-xVIQvi=_}PB*|{bPMe6a
z#aM@Om8+yKQdvgXcam+aEmDM=z09DH7+bc%7=Eww9^L!>e17xCeOxi`_xrrhIj`mU
zdcK}V?vs$I#xDfZJN%zi6nA;I%SnI!NkP?M_bow*>q~R%6YQ@ij@}LY<=8Lt%N5nV
zs(KkQ@Mc0wX@74TwK)c87Oa9Ro$D3H9~nEvuP;)41~Jv`((U<txB;Z=6&}MEC1OrE
z16;`d6M@ZO5yenTnxf5pW+BtQto5ihjPAo%8~h$5dsnK3qo07_<hkBpW!*ZXxhwX7
zPxkm5^dy<7HOH^-(tvru8;t*Fu*p4=wK2LipxZb-E{udHsO8QqY4Ly(WW0z+9-?-?
zslC4$?&UDkDjpmre%m~R6B(AAX)WH)oV8X5jBOpR&4LjY+4}rs?Fk%z$_FRk>W!de
z;T!m1{fMbkN#v2}Fs~eo(fcgft0?}#<?r<T(EEYqxdLW8T%6ZW?RB6F&<EXD9<B+h
zhKu_0_DbmYKsg^GMVSoTKK$CX0*I%sx12Ar%P(4aQ@(Z!&5>|9%^Gpe0y&;WKPWY!
z5g`!Ofa+m9p*`fED`_kYFDEps4CiF}NaR6+X1VB*yq0OSR{9FC_){uJ&ah&8hn3i7
z+Z9NJTs-VcwFswW!am=mL*cVH4U&G*5wdGX`N>38ShFD=3{_poR`vOZuV2{Sw)t?%
ziyyY`Bgd}2m>xmPA)}$`I|Jp~;So&P!ryEUfc}=p#c=vNGmiQ|TtxMcR@MoC_+}hA
zT6l_YDKDy_IJhz!OLs`zZur;%?6Kk<kF+&j29Ru>R67Sb2vy~6{Ec#yZO_hw272G)
zQO9p%kbJ1C3qkXVhC0f!G{ud3ih+d`>aM0pzVCdxnR+z>)z$hQBXtLFDUM_KPgW6%
z7KZ07jPlv{8>kVYqL&J*<G#SmlS<eT0!We+SDr%BFba#jwg@6cMKtklQc1IiJ*aq>
z%86H+m&c998b<-0&V1NPrdSd$p0=?B?tsJjqjT}zLPIc}#I``-?zc(yRfnSloc<B0
zb@Hld#8Ii#m2n$UG=y&ia^?s$O*fug$wWzIUjfX}`&ikAbD6V+bcbDeFo>7b`f~-r
z6Ygni=FrKja@N)XE03+Qdauud@5t0GIz-t~O$I4z=~mHAb^+P?AAEB3BKb<m&dbMX
zb=2`$G!ZI-2<4ARL0Xam;gL}Hs^9sgo)&*{^E#wR(|#`5+d%Lo9l{b}U0AwMO*!yl
zjVZX$fazs+4Xu)Tu_Sq8sYCpA&R?o4y7KvY{?+KRw$yDEX9+I5t>Fbklwh0>K#DSv
z*y)Zs06tdCwrtu-*iP#JQ>N<k)Q8gsI}HMay;$;GCZW45O5XlPGRFqhsXm#Owc^;~
z0V}Yz<AY<$T3zA%3EN4gO@m)9a~c~a<eVcay2MHw*qKpA1@4QU(_cvU6BHqmS{$$Q
zj2SPQV+WLH!;{qnBfNF8O9?NjODW#A-~ZAKVKkwAB<7O(s|!<8-E=cj3-d^zfWUjl
z0LndOX!U%lHL)JhEoJika-?Mg?}AipDHGvtUwTV-h>3p2<;KX!A;sQT<Ca}X)-BWf
z@Iy~~TQn?q>POerVhlznqmL6A8-o>BOq5)dz#a21^cIM$!?nkR1wqg7{G=YXajV(E
zpCiCDInAuhO(Yf61;XmA=|>p#uQ5aMz5bEPn+8xn7jGRLoHc^FGWY7`b_is|!0rf~
zrsi3@tZ~w&DXIxa1d#vCdO^QR`AR!J2w`M%e9`F^mFLEZ6_PC#$A)0aA+bVo!+iy_
zS!VZ&RP~Z>Tw{RxeN{6oQQyi0F)>_?FzPm<ZySGcjhMJ-k`+x*!v*DNZ53iFoT4)R
zl4*HS-ZFjMZT*00W2+uq5cDT}E3IWkVQqfiO<B3B+qS5TYttlbP(@wSQ#al7*^z2v
zdiUH3=S~MwxJC=Y`4y((#mBuLmwuZwCj1sqk|_7?;17gd0cxjvsp&D$G+U}7duvCp
z|3k25w_1b_v4+(~ImfYcKSD#(^r))g;=TNumiV#pteGXmsG*w`zYRbnmcGZxc<mAu
z?Wn!<>KM7VM96Mywn?>|m?%^eOUerA$*l{EQld{76t4s{&;1$Q$39$j-M{k!PkI7P
zaIfDA80QndB4<ljd80Sv9;h98Q4afq@Ok98=i;y4{;(FdWbH=-j~aP8aXzr`l0(gx
z69KA72Zluzz|G5{O7SFvcyF>6KGkf!!8{QPnP_WlDiKHsRaH|{3!IX@c+I{45LpSf
zj0kc{gM;m2qlM3e-$U2}R$zNjFUq1{H@H{NzJ+bzY6>aI%$HkJD$f^v#^q{cP%r%k
z)*wle!PFVLjm*4QuB>rSj9(2{6@kHJr*Wz7DAj8Aln=@bgcDFkt=aJaC6DS!#761&
zvuZBIze3-oJzQ8&pjpA<FKr8=egJA7Mdpr36s+YhgIG?b3KqkiF3Y9NdslzPC;D!=
zWm@#BH<*o_tTj0KyI$*B*{0rSVi^vl1_W=Gzui`JlE66Ts*e_-4CK{Z^W)TOsU2{D
z`m&+HdkE}AmL)#VGCT7ESu0U_VhJh|ykwDO<7K~azAf9KCNcN?8}Z&-^y(&-TZzYj
zU3gRm_3iaP-7wOqh`czM(WL>sLH7Jf>2aTR>|9uQeQBYvKaT8+uJ!_Lz?DvyT<Zwf
z6H0-FS3*i6#?FV-5BJYBlRaq*P`x`j+7x%=>K!TkIb^fwrspf~DR02l^c<Efp>#B(
zG()UIUh+56k|C7vW`D&?U%$b*dUZx*MPBq=+0ueI`98lxk!p2lXld#E$};8*r_kwX
zp^JhQ$-xhX;8RwJTgo}vKe>;5qgy;|g?%$L^wZ?c6=gDTw0lo)?QCJUee)h&JQRU?
z>cvD+sOE0}m|}PgDMNl~TFYt&VoTXU`fGFqgt|Hy`T;DItf^}stT<=xa+;|}cY@Q7
zM04j%i#?V1Z@OM&vjN9G57}J5*}~%IsS1BfYD{$%#fzwp#5H1?)i~9WVULhaDd-ZE
z<S^RnGV+swEAa9wkcbkmb0Y=$b6A^#N7^+N*wYh}<?bS%CiTD!)2+-WRG^K2xn!l^
zac7jTc-WIeOBqtBDoe0fZMQ}ny~t}CoXl^j?7K>m3b15J3^|Za_^P&F8CjY&+&^<^
z`w@84HSs$KHiNw=(FJEGDQcZ!uy*$8a<))(!7>{fo|8gnnT;F5AK{O_Bt1N1P1#|<
zcrlPH9%hG@(?WaBo8Na1=<ps`Gi4!Vy>kdEFZs$X?8-)&Jt~<nj)wSaizA@KwSj2@
z(N;pWwJh8bca#E|;7&LByY3c9e&z{<c^_AkhFUJXJ&Bfae0z~>;M>ShbT;4Kz|gb6
znGR%4P}ts}Ml@Hhrd~neEY4QLiFzw68Te3Gpcz^{(>n~+*<esQfc4WJ9rhP~Phm5-
zfuX*X7Z(zjTa+%CuOwTe;z#3F0gS+yfUH<?z~-B=DeS@FgvnO0U?(l{n`2s=ndCbX
zjA|j+&tSnE4hC2k9D?Vb#I!drbj+s!<>$$JN0=76!#Jr@;<QQfPekao*G0it*6f0Y
zx_TaKNn>0JUQx<5Pfy7Q`WJpMu5&EPZ1+Xfl6C=qoNX^+1gRf>^BZ}~oF*;((fUb{
zy-53|CBz~<iDpelAzz|!i%)H)Hk(M-fp?&x9#tY(vUig(4{jd08;1&Gf0zVtn2tqZ
zZf1QmUmzsg6#-_^vBGd4-)I6D36fAu$PGQg7oED0)vt`qZkxol9AP_CUw$GP?suUn
zTIwq2**tcK^?h=*XL=av1$`c&AXucv8O335r15*IPmA%5=-fLnu#dXpJvDxw2nNdm
zO@F&l2{{;ouMw!*X~MqH$ZDs(K`s2g!{=<R)(;i#5ZqI!<G?swu-?-+rd2N12*U{4
zZhOXaP|`2Z-94er9QQP<M)nZbRc!nQRWg`v<8qUwDCv1$O*lh>xBPWTmnS~}v*1<>
z-t$)Dfbt_c7d1NMk<0QUG>?Ywb39Dj-K`$G`<|&zSgy?Mnk&ntw^)Wzswb>O;p(n?
z5zIq)Qjl&z>=+X_yivFB;hCyNIH0ESi+xsT=o4|u8)&LZ^hJCb@{RISyvJLgYXte<
zZ@4H4KVB{}uku)n!*@U>adsq(ad-XrONFw_+B=D3qRXCMgXf^)f~%>@=%(pfbu?s-
zfU1eq=jC9F)|~1olU?vTT+K2nTr{ZNA7aSLqRosj_es4$RU_~@+k<W@!W+416Wz<&
z?BAOe4p4&cw1!WUZ2(%P+AU?VfC)6kheT*kPqrM!lomrr@C1~#zZw^GPdmJ(DF)#L
zB?yjBB<R)~41}AOn<d6sS|s~dZL{)E_1y^II~YpQa|-MRhjA%I*e|l8z<`1*)0B{!
z*=>8g_9Twnj)hvVhm22{o8~LOJ8`Nx>*~F2T?tlYclqH4`7R^HxsiN+3v$EL@qAGe
zsFhpP-x`fDh*1In=(lW@n@eBvz$XZU+q<czfiQt<8SOQ~4rnUTdQ6BCQCuuDa<f38
z<Cp3o-8G|6P+Ap=%Jy`#(sCwWG*vZVQBL&S!A+d^UxrhBg3C?bSVf<6gxe%8Ivrbw
zGd^=3l&$Y6*Y^8ZBrjzPJ$7A~_0y<7Z(;3P`Tnx;ErZ@lYnQZQIC@O@q|z>nt*}m)
zpvlv19D2E(D3>mX<Mizu6mua6_OQ1LM0;`kr%_dn&23s%x^>C77SMp&jI8E@N1j@n
zcn(ho^d5_Kh(EIFXCCyME+RKt8l-*wSn=E38Nz8KvGp<!GJrQI1^e7iymnv~^;yO=
zH?S|n1inM5-tdwR#FJl|UT{*T^@oM0kRr;femo%!hNlps$~i*oM0|NmO-R^|w#P+@
zo!n79`mv*(Q6X(eq4>i&ngz?4N}4F2*t6Z|iWXD03G9DZv7|u0A3NjQY@@FJ%$#b2
zoRWTkr=+fK@Q)}*Aa4nBgd*h-iX1FuE2|M?K{x9X;rBEU>`8fZO(~;6!!>v(8rDdT
z3{%<DOg8HlkhJ_*-;7d@Jyyjy3^lpv+=&L+Tt1d96?j*3&)v*^U=1KQWQ;Vx<gD+9
zOz8>McVb1zg)2TTUd~Ot`5J@ICfU^L#~)Qq(&JQnIu2(-Y38<gZ_MCwq05O!nunlt
z18x~U?MP1`@rj|Ed;f3F;o~K<srwe(jS?%C7g60!G$AGDc-LOA-fa3dQqrpbnZ4J;
zX&=KSJrUbQg%i)>lwUf_#p)y-UlX()Xhr@b(m7(krZ=B=&kRz6{O}g~CW+aJ?zWUe
zqZ<}ROzwkCc<6I$=R%LjPkiVX`edjJZfmapJfLXvMR#397I?889QT#J$>vx;4Guj6
z5+^LQc`80VSx$GDe*m*>9?@{F!6uL~>Ji}qr;)tP#WUQMhv~+&y-<H)3v^2wCxE>U
zWjtivT^4?~7g@WWG&I~y^Fb40|1&zx@tM~}rlWF)J-WM_k2MLOyK<t8SV}npFG!$3
zPBhr+nK`HBb|~18UMvM>iKGe$<a7*eMX#?-4~tNsD7$CvLNe<uvD2}rLJgqK&~&~1
z;^i9m))7rgTBUVmk%?%tu<9V!Tbx&kVsHNVtOuMwI1#XuD5uZ5exz$JgmV7#X-_9}
z;EirNI$OyOn66<p)?+F=g`o)hobJRjJ-(>jpBKN!|6sAgN|!Ay!To4N<te(am=#NK
z0@}uerQ!OMuBfJ9YWRNCe$b?T@uvHg-;caLX!^16D?GTDdVhxW``B#7=HA&a3w{s7
zq7-ZE_bXhFwE~|?5i#JLO9!f|C>h4Hm<bgU-;#^^yC=O7Cg2Xqe%DUVZFieVr?#r?
z=51EXOedw8)f=5bqiQ2=&p+$gB&M1@qjHL;o*b>JMvBG2KTml8p<-;*Vc#zdLfYEr
z257R-vT;+<p?%TY`_N&jO$ken>15}M`uH0W0MSM&21?nmaw^By6{kZcD_Y;#SyrfT
z1(AfXz<xAI0!Xr2Kmg&G0@c#Sy3@SmL7!;8NJs<pJOIkW;WD3Zk;hFI|LntI4R*X8
zWV%aUw2HEj9RB8RbhMmO^}*agon2}TQH-^Ee~i@!ny(PEahmNuT@hgMgLQJwiF71(
z^9DRtd{d|W1Yq@(5-mq*G`zle!W!Xb_p<ET4V|D}=1u(xXRJ@VoO<YC-CY?cg3_+E
ztEX!mwr2juQOQ-x+lt-{(gf*=USA4E>r>j&N^2kHnL*eY5KM2AT$iea1LW=AmKov*
z1LCi)2d(ZZ@W=-j1yB}0ptazZL{-L%w8HYPrbH_9Ksp%Z%9<#>b+vf`vfK-%38ur8
zlIcrus~|{oy3R9o1g5%=4-Af-bjwu?dF!(AvSRHsl{>Fn-35=GH%fj7CT?x7N+4ds
zh`;cIjjos|cXrFPzz*G0Mv@dr2nZ73<6s-1jW~IuCj^o4${Wj@Nm9hmQ5%VdF`_Q3
zgFC@kFVQnzEXINn_RJQaBa{}8W6QivgYXVT+q2+ty@lWc0j@&(0h!cWj#A!KGmhZu
z*9FHHvDYwLIF-=PP3*LW3SRdYkRQ@5XfBWVZji^8EEhf62d&Y{&JTdN!j|e3b`RQ1
z7ZXD2GA<v+%W9fxw48Soim`YCE^_Fj=*5seV<<>2D3EW&A{(u)ezkmMUn2w@w&b>)
zLx*>&p9!AC2{-wxssSz~^Konr+!2VZM9b8-J0De@9at+y{CssIJ-!q1)NtjGF~3#1
zk-aej9cCsrzO16PP8uyTT&5#HrZUyyx29z9g9UrBI7)USee+7iVSajZJfAIV#<#NP
zIhk*Sc@6_t?_twdQA%qJ53Y;)V&U`=?u-jMf(-K9^HL#!yL@&N77zU6QF<!$`s6Pv
zd!V6bw)oUx{O!O1kX92EpgNyCB-OfQhsuqGrMgJ6NERs3Vi`g?7mrj-*Z=ktoFCNK
zoV|Aq4xfy>S7$u_;<$XWRUq@M@i8l1vy@rn+F@M=h^qc}L>6TLj0m+$kUEKt@?zs^
zN1d1Zdjtmp0De%(MJ4{F%sA8chX6&jWu;)PM9NDt<n6f!F}Is%y?3n>qAj01MCDP`
zVmTrrFr<0_Mz}#}8)Q{1;alm?^{SM<<NdR#`d*b1am0P>Qp0Og&%fLoJ4=mD+96{T
zW8uK81{tMZ)#m1w%e>ye4jPsFXC9H?MIJZ5d^K>?KeSGz0+qVK8$Ii8ZbEouX#&Gi
z_{2w@qO8++3D-ov5rlg|NJv7RG2oV7Dv^X?S>oM@iT)OY&L-$p1dSjKW1DJ;3BByb
zi|6%#-XyeqmDu-VO+Nw}UaCpx=h84<`MdbF5L?iZ3c|RQzUOeoLyzllfqXTvxivCc
zwY4kWsHK=Vv)E;LY<lsaDWlLEA=U!Cy;Kmx(U3ETSo@)J+qk#I&&y6Io|CEW5tLYQ
zMO-+6Seu({GnqiuFz`Lrlvw2k(D6&y2vfqJOR9aF@%6;+UnX4vX_oFNf7F;tIF7^3
z#m%nC1Prs$-QG4ZWg=owB5DK+Isl!6an(h4@VF?yk>8Xr_XZS?0IHY9f0B50o(geE
z)ANQ<Y-S;Y#kH{dVb80=1m9pGQ7|yn5<yn?o1mk9%nOYaVlWW{D!Ul2QIm#BDdoHH
zWu!2RPKM)cuK<fU%hHjA+H@hEzvt6#!6)=}Hi>8fk`*e=vds-S#hwIXU7j!YC$D>K
zg;ShPeB+GC$5L}dMPYl~LSVX=4f17e6F4V%lf6L<<CxC_CRv`CHJXKOC2dVKaM9hU
ze{pzij;EA6H^LZy!369SLSe(0=}=GLb`uf4FW}bQ{b#I23K}ggf&`iFIm<2M6$^Wi
z8G?y#_r{}>z4a0Xbk`kifmioF;t&SZ@dD~oh=OV<GM}vo5X~We)%SF8^n;ozTnb|$
zrx&{U%m#9nK)mJ9xg)t>#Div2PbuJ4rr($ker<U@fugg|eQcCXhrZUz?D@PqTT$a}
z#?hq%c!{(;UX6PHza3|_4?<o5!k`dICW9&pFtlsVmyXqxWuoOASNoYLE%S$btj<L2
znjpTX+O_AW{F62d`bVCS*YmN;lhFXo4CdgaR<Na7vGs!vMh=EN<05h(01D)!O6gl?
zrJmD6SN{9QqeaOR%@dSr(mW-e0X@@}2`B*wzS@)uX*GFEI!&&`6&9g8<}-zFfcr>K
zf&+D5B9;L0*cD=BqiN}g^^ZvqQvG>($=l7%P2l}}3SG(BiSNG8S*HD9N4HB@l%%<T
z)A=3YCwl>p6gMRr2&xS<!JY^lF^P-o88r55U)#4EWw@CVm3oroO`q&O@T6>3(R=$?
zhMCHRDMnLi+3eErv(m$Jb_D(9F6URP>$UIkKj>gEDsoSyJq^WTC7ehr1{HpW>~Hj~
zh@iM>Ek?t_#%CNyhhLTzI-LOV?r1gs_5&ztm@T`BEIJ*5f$31vii`n#K(}^Bn2%(}
zx2wpsZ->)83@K>8mKlvjlxWmrU|3#N6+V0->+OXp=kcE)l}iBR>#~AcR%^2eNFChU
z9Gbc=$<uCfXa14tg^0Z<IftFpjFR&b!9S)RSe{DW+`_7T6#wG!myxOIu0AX(>I)ba
zx<b&e()GF0L=iw!igtE8;k5wDb`ZSK=Z2ieC-Pwbb25bfuBI82*4GZyP_&FbsCo6I
z%CrQ~-5>e9T3A@{c{Pq#kJR*~%(4UW{(?){)&T?@q?GPs=Prid69Lf7Ua-|g3o7z=
z8l>BTMc_2{LP!D|caj3LAwq~5W(F9g3pKZ-Z}4hG#emRIC=8t5`fjVn{&z5rAX=q1
zZ28eVXh}@+w8Ys4mR^2yK3#jactsXH=VqNj%#)w7LLkn|h)|WYk-fP+s;MybT3RPc
zWawtdrvx$DumHwr%Sbl)i7#I+oGAf#z5t@s&lAVxaP$~BQkoN$TG;7)ht6){yum7;
zN)hOIQGw2_gw8SgcApf#-X<zsu4Wt>qH)AxiPJe(?4~RNGuS5d1SJVgsscp~Hupc)
zR?t>0b2jx5X9{uS#X43#4FmEd){&=Z@zUONTS`MCfK<D{zx;Pbca+g{G3{DNPbH28
zSvFSj@M3@9MhO@5RU#gzrtRqRs8XByb;RUv=bzE=YDgGF1x16Z6GbU7a)y$W%UdN)
zEYWdl1~X%2ZGGg0jJC!jEUJ5ZiK<z!r=|%)Wo`;U2TgV<_+KG|Y7ff^?U3S5Msq;5
z{$a2i=LTn(<NMtsjsG4KM%Y*zIxRW<iH_Y=_mj3qTlOd}4&FEFcldbyNe9$K!*xIl
zb5Ny)6+8>ub8@f~i<Ox}Sg>5MF#BtFeb&tnqq1oN0u6)B(K3{gS>JwC%sGWX$tM)v
zBS^WKygqRpgJV!4kj3UXJ@S}ToA~o5pN9VQZimfj$m!v{EUcP#cQ{))+h9O<`{7Pj
z%CQXe8ypq(U(PupB-89t^)`OJuqL6!389WqorMxGI2HcQ^%HD8ke3^+xZ10R7w@zp
z6j$b4tx5;Wbo}vw5~E^ybRq2ElFWh17XviM3V77Y(AN!(_l+99j>RXRDl_Z=A&!R5
zL6dKTsi%A@B}^z5>6Wdca}4=azzK!#0s>=mh11B(c-ue>@TJLh>-w#0*<zJcXe!)O
zu7MM|#0jfIZMg_=_N?85?mK#)MqezfCh`w2MDa!S22!F+?3snS*nAr6A%t&Hr;Jq5
z^QWUksWnv5Kk1ojJi?9ML>*W2Aly__b;@I4M77Zm!?zV$^VB|ghScZTR8w}k5#_8(
z7W|wvX}pxqCnUI8qLNd<x}$g}J{>Zo;)k4f7nll4n6@lt3;lkNFS;|ZTEcL=2{zyO
zRkoAr%C`YVT_*I@-9)ois>|3+N+s`J|La;Rso<;-!iDv1*593eAo#bT{qS`Kl|N`p
zDzMie_Ce$F_ICLDLi=IzSy_hwH?VnXGtEZO91Wiu=YLG?Q?U1kz0Gn9MJQ;#%Ea|4
zLa?n`y&o#`)DU;0Rv<`>MKTWgbhE@;d$kX9R8cKjOpq{<HKTG6m<NDwAY|pjGtjI`
zJKjn&eDID}!Z0^^_NfrweIWuhM?OA!I2L4k9x1=n`8Z}jBp`t13T4=?@pAe3B%Uoj
zA-M8928oJ>G-)w>-Hy?tX{g8dQ+gpyu-j12Q{qdmw%ulQFP0H-R6CrR5}ODh>+a3!
zG;({&`3?Q1@BIH5RXcP>b&}1Npd&ti!O)10k1p6HQ_FG0d*joUflaBT7N11-k>!`=
zx%xhUg7$I_ai(c;G5ub`Odk{mI7Ptbd$S9_Cy$yU*#orphsp-YmRih5TT42d-4~_}
z{!ck|)*p0u(!*($S()1VH|RshBXSS0fjhw!tC#*=*MF*6F<p#D2<9ljCcB#kzk{u_
zXA?bJiCH|0Sw2(d9fzYMPz)JtLbt~mTdc0V+)@SpKX;N5PCvjO>!+4U#d$}p9UKF?
z+}LZkaY0CE!Xml#TlOgrbaJYc{T+n-i3d_nVz|ENAIAr+?v0N<%0(}67O|d$@4dL-
zeN%Ms^jYe*ol~X2!cQqNw8&|eMUmQ&EQ%BQid&w)qn+Y4Ywc4GV!)}DTo!=z#7i8h
zPDY6z?8IA7iFe2-_EAQjZoa_qQcS*<H0%>Y`tY$1L!Wz4`y6v<uyUrM0sNdD;=xDe
z2_9PO3THW%B!Pk4Z2d)djccXV!$oV|{=A-rL_jGUhQFHfba)=NDVYGPE&hbU@`jt(
zil78TGjZQT{_QTshW96&87VfhKnG9}olEJOqlpfhF$gt|j!su5>~gRwD7=XY5BoB$
z)%t#E7Ao+i_pWJxNNmy#Y$ArQ7k<cYGYG&18GhVEAP<6ct5@0Mj=6!T^%2=JX$Me#
z4z-j<G$a0xo(UUo?QmgAAL%u`=1wC%TyKiGcxXW*-flRTVpEL_pb5;2uuxQpC9m2b
zM4k+B{VFPKDTk#%_cUCo^&`m`K7S@5o*$IrgP|BN)@X;MpF=~jXgYKnJ{v;ni7{a`
z7<3Sp)bO9OJ1}8hEmQLKyJM4~0^37G(~_jR4CiKJIfdK8?EClc4PGf7Atji)+=?a>
z35Lr7hxtDB)wyR)j!hR3OWq6NJM}C+tK5bXYrreenr)X`jm=kQ+z6FT$A2Oh^iMcm
z%K=ryA|(RIWZ%9F_-uL(O^|LV>uJXCi8|PEw$*LUM^3p-kxn(2U$yYWG861=WFpy(
z9)Gi3-aIP20De$nF5$m}w_2MI^p}Mt_>fnAc@K_qQ}i`})XZ~UXH5;cVR(InHcyqO
z?1)b50@J;DDqy&x6-?=HC(}#Vr8AlR$x~tkwnk6@N}`2dq>7u_O(iHxYVL>7`wg=h
zt(9|?mJs>(E;Bnhi`{_<l<LN50$wuh_+Z!4^0#}vX~nwr(jRC1otBKsG`#Psnk;Fj
zfRM2ZDlz4C8i7=}z8_$`3ST#uLKk@ZJ3tN0VNW1HY(9r({adoN;Ff%C73zl@Gb))4
z({G0{y-N=89<R75rO7az=(;|B@^EC>Rngq_$=KUqbIE*gSep&xA^c)^vpc|t_(WA?
z9}JUT)Ag6}5B}Vp%@>u|3g8K^`L>|#AESpSQ645xAPHT|0q+o{CrfnDI6GVaa71I&
zG6bwJ14lj2KH&{=J{?j#<ghU{L2ys6k*}~5UwCK#O@^peNPSEm%FfSmSg|YPU;5*v
z4N1-?-IIry4enPa`|5NaPqn&V(SJPExIIbE{Q6A{^A<(;_g`Wc08}s$5YRAKx(mbs
z94oqq^9+K9aVav$IM0`^!or{+g2RRIjQ-MWUyK;{@D^{;a%54d8HinF0T;m_q*!Eo
zTIj*@K{P5XZ`niU^FE6OgYCqE-*HH&MG@lg%9k!4uszD<jws=0I*3Xu(P7SHA}wv9
z;P@blVB<Yw+RJ}Ouhc6N40G2FlG!&S_KN|X=h-g?ls^joCvW8&s<f?4+s04%n~>Ik
z45;zfifLU1i{R=gi(tNJwHLiF*Fo|U4A?-+KZLdXjU!$^w<IdtlAoDNCz(MeHdVC|
zH)V!F+vB2}^wFc+R?C4ifhvbV&!h9EDHb_=Pm$TP4Mh4tvDNpP*AeTk<|XQ7eK}EI
zhq0DyZ7OUMk;)ZsU768fBt;n2;Q;wu-`AQV+GG6A!W8-!NG#b2CeS?)sV~$UO0dk(
ztZ5_9Sh2Ij;RX3$e<B>&DFk<#`xC}o8sSzpQ@K2}Q{v3>GOC4<A8ps-sZ|4={5bV&
zMV^nuNZ<xN_Nb3dsP^jz#IW(a7-vz7Q9_wW-Cu}b+=r-MC)K55)I<=~&pJ=B3B~-*
zSr!E$Yi0ZTAq-6Ewb|Qvsr52|C3EzIvEzmPXtC3>15;ZMust1-V+u-^VIJr6BwG+B
za~r0ygl7IK=tmc*!ttR!{`UY!x^1j~7%RNN_OGcjOtgfKghNXrWkxu<jgrkl_!Y|7
zD$^ES^cE}T0GfR@3)`HU7D(;>z>SmAMS-<7DbXYQj^S^qSPf_SeV8hZHh!?0kXse5
z{p@LQ=R)O~1sl+nIyP>+EhR5FZScNJe;qQT3|V7co0Z|_=Hh>LH<XUm)lnH46_A}Q
z2)elq<x6sJ@~tOwuhoGOa{RV4r`^$qx(Q%jg3~j}A6yk@RW_*;lRcYhf@R5`7h&kQ
zIU_5Q&S873zcGIY;$JidHpQ<MJ_&J>!s5k(cG%Y*BF5L&+%;SDdFH}Xj}_~jn>5h0
z+xh{R#bVUaP7ghw^H*9d>Vc=^sn3AWUrN%rYNrMR?r;}{{PREXMY4<uXLPA6g|yoW
zl^+(dQ=bXf?>xdEQ+og|Qqe{++|)rkKVkbV-=VzR{{aY9p3!z*j)x$RzAY#dS@B;R
z`GAq=6|$!98@DJ&Hc0pjTfsK%DTm4uN)DqRaHmiuf9lrLUx%ku07Utfo3i<+mK5-Y
zL_xbaLH|3a=l$nT<T4{?8^R4;wU{+tL_$jrTj(0J7yKF8749$*vA1#C(%HxER!zLB
za{rF2G<cTjuH|RddhWw(g!hBRYVU6LYus3<wm3vr$&)p@HkGRXkd>U5)bFl6*=N%r
z{XDp_E<?Y}&9G7Wc}bb0^^&F4DpH8B-tf@NKu<v>=AfG<Rel)}H2j~q2nUS1f*oLO
zkD{<8lOyDk49R9iuRn(lvJ6Uw^cWpbwuT}{u4B)2nl+=N7mXe_$AJ3x>%ymp78-$r
zSM%~zZ6qp>=Lhpd;YL#dlu%tVQpYlYvR1kccgpPsBKJbhu2LaqRfUk2`|MM#JxGOw
zeBWx4@^}i!JsRrV)V+N=g*3Fj<aZqgc{AEJ!Zutwj<GozoTsP^vR}_yLLP|B6~CE1
z<V<wk;R2Fj6-VctKh0rylOU<L9VxDyR5vgQ%4zNY!mdBBe2Q9U_VL99Wy5Q&@81V5
zbwWhp?lrfv8~|LJ=_1j+xS+~!Gms1mnDU;;lG&M0?_3vM12iw2M*;$R)MfTcv~q*9
zcP}jaR)htWivW7*Q}iO(0h~B}o2`Z6z3hot|1gZT(wJDyr_IcFutI=u5XU6ObNLx5
zw!;luJn6+q1;l3M3hHmR0a#>xaTezJ8`fs-T|pr@y&1ZPESC$Ua+eU%l_ZPbgP_tQ
z3nvz)%b+ZVuj$u@Q<jE1XAz%YU{>~z+*`)G;!;!6hMIG?ZiKMP>lN`a(4=salbhXn
zaY3go?nfVI295)yOSc=({ia=3l(cW*jzUN=+xkIVh4doFu5i}JWcFSz-s9=E{jq%S
z6l-gFR~Es_b_(#u<~6@ww{=@JhlAgvc6YT3l##FH_8(l0Dpc-kXd<ON#SR5$+eSqH
zVli+><x?1y?yMa+?^DA=BrDy{HfIj%VhXxhRK`Q}Y1)SQ`?i-5!dp=u?|aCI;@b`|
zA&_n$<+oKvG^4)qr$O`Nz|ut@Q=$YWjJm94ci`xw<Gm0L0bx(`5uZ(r-y_c$zCVoY
zJPE&4m6n!5lG9pUA_?3g+kNRX#%^%OuHT_N(ziZipHF3W=Z4`DjcmQv`)f-Kr=N`8
zEEbNjIJ^o1?ANtW_W!Hn(neH1YOHhF__v;_^USFv8<K>=vhKCTmuGV*4ky*m?2)Z*
zgu@Tv0X0y8WWccG^Tu_XU{;GItu{O>=f&~ok(jaNUIwV19mLMiM&%;-v=^i5fy}Kt
zJ&3&sX8-Rl45|Z|xhofiycJR#qo_-dmwY<Y|86cfeh@c>NYB@bd4fuYs_Pv(XQMyj
z1V4U>PS6{?I>Uhao(AI!8eq-K!YF>6h+&kNR3`&bo_2njGqeL<<sfOqJc`rR(%Zav
z6D_yU^beg{gDJ-c&0uv{;}aDrQYAm|x11-O(yOhXI8wzKl2x<_8)jWZ$MCQq>rZZ9
zToM(1(#^REYz;Kjs&e3L<8R^kVWJ_YMzW*;AwUMbOLps;HK#q<lrHei$c#X`4MkDg
zvn8Ggn}INtTc|VIvNNX+%!L<L>`YD+2;%&WvakN9;@`+C2UnVbMO6*+IL`>A)$`W$
zevCfeEjLS_nRYD8)n7pLKrvAT=^u+?1J4%!2VI!|?G|O6f6ZcrwRT&EIMHruZ>EER
z7YSnLDG^ln6#M)E_R)ZD5ig2uV)IxfXsC;n$-lX~I`ZuRBKgAEV*8-Fi*rXb>(E6Z
zTj^eBrF1@Zb;wpwC<76=Dc3rdZRCppjN~R9>4cCGeKpU8g;<%?G)?q0*e3|8Oh|P;
zOS9{DtiWfv<?c=lF;K~cEbBJq6_goNeGHdX{HKTu%y8bgUWChro0_gLW*$u8TnmkD
zEyx+tH34&u+S8j`jb~82ldks|$$P4+Llkuk%?9e)fa*%QDnx0(kuy7TlB&=nsXN%3
ztw<%04;{eYiD=!zKl=28(;Pb#8r(Cv<ixK4)M7p$GkxUPbEY@>$F{UXWvh4sC`+{W
z&-_fh*Kv?JRe=cBTzDiMF}a$63yd;I^q%08%T44&lN5U)R_~7H+iheZhWi;xKyH&p
zued}_BeT2S^9)X4uW=MQT~%-S>MR@jtH!S>@B}Gq4?PF18+ZG%2w3ZUQ5iAdsypF4
zRYpXsn%)XUCL^atNF8p}rK-)r{VY%*RiZj#nArUwny48LkRWi8QeP)P$fEU%wvS*J
z+krHTTD{qxJmp}HHvu6VvXtp*h}l?eIUCMnDHqNiC<Ktm0@z?`6}DrI?N&(33T1e(
zHL8+JU&lL9KZXHpL5@i70riL!U-5|O+Rud}pT5~|4PhFt9jlO3FZwJX>$oz$@+Znl
z_u5$D31qB<)T_D$70U$OC=|+adEi2~=$Bw=!Tsy>PuISMoD1AYK(b$)XL<A6#`@xr
zsabXP1z%Coz{gK9;H#_ZwXYcYXWSbHjxV%uW!l8izs873?%)>7)U!xQ8!|6)xEbu?
zyEl_J`xfd+j}SLf;Qd4(`v791DyUg#s9RcEF1bulz)YFelV8qC{7r+s6`H<s&7#jJ
zE^djsA}FvQD0P29N$qw2R$!cdhcikI$FIk%3+>@}+Shu60xVDr+}zy_hfOCYCH#8_
z6p(6*m?*HBz=?O~k0boA=4P_!XGFmU>!_em(t#2?ZUqI`rHRoUtrNsnN>+d@=VVH7
z2dBklD*;$nZB@pfkQ$ff_g8AdM9kF)7DsgK=A*0!VRZn3>|;@pB^?>xMTWfk50d%E
z+|<{pt|n{7?yG<V#D7BG@;e`qDY<L7|L`dEY_<Sp9W%~<ytn{z$%Z9st_>$j=SO6M
zlY_JjC^+!1c`_k)aQeK%GX_?as#GI;J}knSze{77pV7c@I-E)_Mru^&;!5Lk`RH(r
z0N3n&@cphM4$GFnhlIpHWSD-ox;Jko>ay@jfCHe^vn_4y7{Vn&v!3)jGm>_~GTJbW
z&~V!wuwV0g3D${O`xfl7V;!=t)#dm2$Jg2~_hb|rw>>*s0sT0GCkzeTv0BkuU|_Z9
z94eCu5vx%JyL2*tEl&PikO~e1&IbXn^|TN{dvR9;EnDm`5Pjiuw+=jN<-&n;P`3u;
z4J5eH3E}@X#lYY9eM)}ufu>ik$5Li`HAmib`-ievei~c~XbEs@nKjlHC5j=r_;zoa
z+3AVE5ksZGq|X{3w9cc0RRwsf5l#c;Jg86(>Ust;jpHhw6`erC1sthVtE8J#2WAH{
zfl}**+(K2MQy~G|5h8Md{SanN?;raH;R$9`Zj`5MbG2IOCyQKu#KMEo1jv-mQeDC#
z!DOv3zJl6LR|LP4;ZZOV$%h@X2JJigmm73iuCAcuAA^NL%eNBJro&<-*l0X^p)gWI
zA=KZCX1d91YNUOpWA*2<*3molTQ9n(PK4{&xhF>iN|oquuTIX1_BD6mi|oXb#Xr_w
zbj0}GqUQi){te2LaDWuI{*$47&E1!Ewvi~*XZMezpj-;rh)dtSZeUH{atrMuw8=1^
z%D@i99kmwOfjf&QrC|}EN-joY>h(wq8pATa3>AAn+tXm0@-<rmkSOun_(9c{!%yMQ
z-o%G2mVWntxnZ@v)iJSfSud<FJhSG$3eL}gfu%4Klq`%D9QwL9<#)H@CnFjk_6}}&
zTL%M7LuFvM$*1(=S}WBwE$YqLD;9sYJ-;lMIWzsN({C9ZJ#zPmpD>at6TIm^yg%4c
zA+d)l2nkRr3~zQNLo7u~OGrE5E!D0P;8iJCny7+Wb#OSvtOljOJVMa_;v&v-2_T&p
zWvN5p8{T7i{l7&mASEYfeDyr6=oOE$AlKa6Y%LpJNFTNepdz#YkeB9E02v7?hx%HL
zwYi$M#jp5j$3nrLO!gLoBICK~%J+wQoIL4s(>=0TIomUimf4hKX0>!TYunwrNihE}
zt^i0OegUZbs8!07+ca*8Jg5#1xNs~`{kZDYEt}cmL2QG#cRC|pA5%!o=V%ZE#FB0#
z(eV(tKk%uMf!q+7Nn7?-BW6<99dPANGYy}U_UZxU5-0?+j5I*m-^5=;sjeDeL*@@R
z*LGCcq^k&R*5ToW(I_L%1^(|Y!7mDw579Y~!wyMm3TKzYb?z&_bCEYRnEU!4&B)Xt
z9fH1YwM}Dov!cbH+7^Er!#f&=U@MaJUyM6_Q-^#zD?Zy16ZN%pO{=g-i=6wcLEO<0
zs}lAlR~0}H<n_d?kia_Pe?T=w6%pXz^E1X4t6naupymss=|TLXn}HwSnPFupK+2Mb
z=c$LJD-X2zJcUTV9x_IFdP$i2M84&x+jPn#UR+e}7=y=p8~?eaXUj#konXwRm#YA)
z^R}#BJLpL$U4CaGO3eQ+!r&@{4(Fv|wK+ZYh3vY`b|TQro`lv{yW_&0A8>7d1A>Wz
zW`>-7hw*eSh=!iHO_M=73~I7itr31MzW_E3ofmGm72~7`zDHPK>V6oWWmcTKb+yNZ
z3GLRVcnn=jUvDvs8;$t}mSkY_H;ZtM$}rNEB#TCa6v!Aw(#4hW;vzz6(yt$R7faBr
zx3}x+B912(F%4g>QuSj&`Z+F(w{Htsm!8iLu#a@6KlBh|!9RH@0FU#h(Kte=v3WIC
zl4WY6hsLolV#Pj0&BG|-PMY8#M1(Z|X<~!21;+1Chd~T9J>7&r%mOnO`tGhB0O^dz
zJTXs_+_5)SI+4O4gb_$a4=Pj<2a;}=Tfy<K_Dy~=+8CH7!&A?MswW0}09BCly~u|F
zlJH8(Q%S*_vSvvl-oLKG_$Ra4r*{5`OW(hh4X54J8)?m=7q}X>+c@AZwM%cgg$)oM
zrQDW_Q1=X3t2M;0A5X*hsG(Sf?*SERsF6g1F;Yky4{#vGXAGgo_krqzZC>)Zj`|1N
zbQXT@SkZqkkW!}F3t{;%@S(meYnO(QG}Hn|OWzI*mIw%BwaYfr%~V(O4b+H*%#fGP
zb|)lA2f5yBNRy%9r4p4{C0niQTfG^luebiqx|*?+Rw=x4fL9JJhmBl@z7B-mThe?+
z{V*d-Z-!ZieZ<9+yZ1xT0R;CUhaHfEz+bl^?MKbFA(aoFhHg&}dtr}`I0Cn~2P8?9
zI2zrq%)R`LblzagulCd05ZS8t6Tb+F6kwA8Z7G;(zQKV;IReKY4rAF|3!`r{w=$7v
z33^ErjKYzj7st`<TPfC51{A{CdPZ@?!J`SznGLkH^pGL>iOg<|c^G0NV2iNF7P%xP
zl34uStGPd;$LVWD@<_2OamqYSnh<jAq?oD-`^Nx406HYFzD(2?R3*XvA*R-bA*}Ku
z91Ez}8wcZ<_iN<8Au^Tw&H0=XjoQ5qw`4sMP~@CTabsc2#W3;RiP*+v_w51&mdpN8
z$_+QU)WGkQvYOnix$;-NhLb#~{Xic4dibTn;Wx^_TrZK<^|}|~4E6mYZ+U)K)8V0u
z@_N5pLGD|Lp4-?+p7fHDu<8)V2Qjj16f-Mk2(mv8m%jZ9mVz*r7y)wuHS2{g_S&48
z{<VG>TSO{bU=Q3qCWil7e0y_GswV<|+|%ZBUS}hlmNtJB&SHVY2o8+%dB7LdKUyr-
z^GOD4=7Y?2mHjt*i?&le9&~B|+;g^&+ENEuEyQUG@Pu7!*4M48sUuFywam6iBswOT
zpaqmj0(#Eq999&r8AC_%sz`2xgOea-r|V%TmH?It3^n?Eq>LNGH_B01;$h!)!#K_<
z0Y$1c{fLEe`09wlfyB=WAvC62(MdnW#n}1`1JtmkJG~}$?ZoTe_VZb8j0B+d(j~ff
zq8jLgs)<hF!Uuxr6XcE22AI<z=<3#PCz|FU*^HpYZ&hJ8?{;j3tmR_-?VH4sRsuJ0
z_1@rVwqgo5g^#N;V#+WTVNW?0<vw59Uy2S&7N!u#?d4FDGz>fdB>yp{UN<Bc>$%$=
zI#a>mM4u#XbSjI<+%lrD1>A9ra;@~c`g<ASX0@7Ws@?txKn-$?R{m)gG+VI7l}c+p
zIDrNCu<hck`tz-$PioH@+}ijSTAqL@0oaFjIjP5{Kqu_dhu=LV)_$@}z8+2E#AP=u
z%AhBmZGo&8%?G1n8vTqD?#2iqC;y>DQxFU^Sw8ix4|wkI2XoDcOxd=gN>U=HVn<l?
z*@30OjpQjq#i0yj-Qt{19;?XIa&xceJfk~0J<B{??;Fn-x$@YY>N0Za|2Ch9v^Y_U
z!$N^ck9F#GnR;$>P4C<I2FzP7#_1NG$m&XU99LLJXCMTjX;OmgKPP`4anE&y@S5vZ
zfpYRsqEL<Kbfr{`GAOf=wi&UOVcxJ@)ouqWxxHAzd?{+S4dNWgd~^9tK~6+Y*V2*P
z=1lpP#Da8F;O)e|7<w%WV0hU}OBbtxCr%@m{>f+k1c0Ck^Bj>Unzgs&QZ$d(TmR*L
zErd{z)RYms%GNKlsQWYnkm@z0m(xDg?PFI@1Ke{`Jf9cdV4oevT>PoTt{0>6_Ps<X
zc4N%NYg4{^GjDg<uP*Mg4;0L_Cr@8&PjIC8aVWjzI`I7<UUZh_fwQuvK*|yn?=@=Y
zySgK%@#}k2d1+l;IAuIM((w+?dMlPO;w9guj(mq%8eC9yIj_^B0Iu>5Z88Qh=uznm
zEkLK<ORP=BPac)On{-g2vW%3%jY~Z1jJZpi-_GMouEJ;_X40b|Tkz-jgEpYXxSXpR
z1hw-6Sx@g~p-usKzzHOgi^_kr(>ZIPxd$uBLGT8rmDO1LTDZnqeBxWKG|oPB7$TWt
z=Fvl)N90YQ>-HjXY+A4Od-8&M=l(5#=w~)OZ@pS0m|lpn5XIqjVxU)XGn-JX^<Jvn
zR#7)#GGoN+k>n8@JQie4-o@BuH_8#6of{Y6GC_afr=BGiU80YIiW_~qP;Yc!8n(3O
zAN=1{qJl!|AZPT@lo8XUMf;rj+19w|q>UZ+?c4427oL17_r>_L;Rf=W7W0dryuQEY
z|3b6oE!zDR;Xmd(ajKy7oT>3U@q2^Zr>=nWJ2Kt2uN$wM_ErN(f(Q%^w9MVN4h|y?
zCLBf!jRp$wAyX`~94*HZmd>YLL#W^}FNd*4iCaiV7L2g9LN}qMl}omM!~lJ409M#y
zF^@p!kBHuPnAuZuAg|-=^^)2bdp);tT<g?&W!vfJ{MoZc1#6u%_u%)W0Zg^ZBrH-j
z1%dISwQVW8_EbtfqpkH^ZL>!n2&Ov!r9!!jt1P(tZ3fiOKg|E!+{;0A$FhU|VgEW*
zjzP?<5gY3btEs{Q_$~XPl>+OA8n=fzqN;qL8|fRsuGCjX2@>dF*mr6Ty@ygIm#Q74
zgmMs=h3v&NB&12-?f4rWZ>oO2xL@p)VqkB0=9Uv=rV{Ob9bA|Y&{x*U3`dqBh(GIV
zdCiq=S2v&szw&J-s2y-QJ8o23tYc@mb7nVfZG|L0sL*C)&SPIZC96vU!=bz^tvZtE
z9T+EMt{oQ5(qd;Q>hX!aqK=6pM&{KA{0U?h6noJjgz5Q{7>0UZ_DvwhN}mx5zGUUi
z+qb*XOyc|%@e6EUi{F+q0b6rV^Z7$<0;ut#<gi|IN!>5Odm>|czgIb@c7NxE9SngG
z*4Wr;z0jV`e;tp9ndZysw9Z>KjoybKEm{~CqiZrc@O@SZ)$HFn^sLto$=ysaecaXF
zr54+Lwdj-OZnTv05%a{Kqw-H1u^v{u=Y^U%Xk8#c4S+`opmo(3D6$aDfLDULGy3Mp
zDL~KJ88!t1X)E~>6suel$=V$vof)q~Lo&DgsP;hT*i83}uVH5+6vd>PZo3XyCB08?
zv3T=6^rh<-_3&6XXYwKOeU*9uFu(UT@VgZHZDJV~7;SUi3oMQE%BT;Rd_&J5;WHMA
z)e%c(152=Y3QkT!$o(BI86HS0s;GxfVG#ua&~T1Ll9T!^ax7YJKMFNML;p#d5_J*e
zLp=#eZKU$6Xy_>O;XKT_WAQ<dC9HyyraC~pr8iTGsA8vf+zMMRyY=ki8%k_*pvs@2
zXJrC&FMKjgo1Xnp9s$cU!1J&z%3r};A{rq#7@Izb9|LKI$>lKT^p1?5N@HpGz)6St
zR`Xd&-Eo+VQ1_#j7+8!OrP6Y!A6Xc#;>GwMkVx+!JfU8KQ&Zg>79#XLPWHP8!;4dH
zP)Js|26{U0up}^XGa66l_CmSSPq+7M*{hb-eX8jU1>~_pFFOf~XrG@gU8$_kxNJ9L
z?sh;xU{yh44}(O0gtBP5j@9~|+g)KtU{DvgjR={4l>~U9w@L|7f9Tzl$?dpkwMJBs
zG7(E#uE^?eP(%x1j%d|k6fcQw3a@so^r+)DwH1QNG`Kr7ozuyeE1<;@HeTIdP1N>)
zdOsQ!ozKVZ?Q6RRG4Y)FdE-XOMP?`Xg`WjbquWDeomu$=m<PL8&sp~H;b-|c&nSw;
zOPUKUT{a&KWoF_dZhjs(62o`@%e07|yZypl6Xx`-&`|4%1&pEJM}oBh?YZ5jcuz4+
zuC8S}11AJQ1~ucf33i#W_>U}uMx$+$c41Yl28^%fx;5i&)wwhTv*EgoJk9ioTrW48
zxhiTS(bo!l;lKJ2pfI5Y{IQ1~<ajxJC5dH%)+iR|&vn~m9zN|}DJ4ni(6QJ_ehSrp
zC%!0tN(DlCFE1IAuF0D_{PC3`r(2&>_!^vy@X+!J6V3Lpy4KF{wSxhWEiR%`&uYLt
z=`iAz*p>*LUaPznh!-$tH$Gl9F*Hgm*yJgp8?>RhGvFSj3KMmZBELt>aF|6%Nep8=
zLSnlw6<P3OGgaDzMRfOvc4I&gxfOm4U#)+jA}0=4mg4ZO{J`vyq8|^z*;r7lv;&x-
z%OzHn@jA=#&V5TYfOsQA4;8k5(=z<@trhidUS48C9mqygXbCB}_evNiMDu*i*l;Mo
zKVRZB4E{lqi~4wPy5qM}n7Bg7YXIctwy3V^Gkh<B;8b-MNM#5X>F!)bf>f%|LUi39
z!PY9TAO1`*He+-=pO=emJzHedXoU3`jAnrv$lfMAwv#-$rU4l&#lS{n8vf}QhxLAx
zn9Qymd^W!bYXlm+Q6so4jIo9n<(Q$Dr83Rp;<95wY9mUycC!|AL?B{wU?5POyk^~u
zpXrw$F#I;3el95(zjxx{YRwgTisR5;a?*solp36LbSH$%k^hC!b^D5#c3#)Y)cCZQ
zh)$js;Hx5|YW+!dq}mL;q<aa}VL&yYQ!DXY$y?Ra+zO>W8B-H!m{J*7G=tbm5#wV-
zvp0rAfgk=@p9mBjnd$M|$ke1Oc=Yvo3z)23qEBq@+|RaLGx~b*!zIDlE_?tX-6rRK
zHgm@Fsd02sH#amv5+Y@V{-;(m9A+gdLg9*&WO%i55^&wXZM%=w5S>xXca&+ePfH#W
zv-MFbYOI%EW{6|r2`G9^tk3k!8uk7D?u)=0GN)I~0OSA5DRIP@FJb_;*n#NDT_jeN
zGK|tF9Ab1XpX`r;=!-6qTY@#XmKoLbW;%ZI76d<)p!@?rh<*i#K&Wjgp8-%5NOc|M
zv|p_G<XALC5OO@fTKvR%U%Af=GF}=gJoR)<!ZN*M`j=v@{bM9;W(aD5?v({)Sy`;@
z3uwt!O?U!iEbtkDV?S2IK@ut$zK`H4_+V&aBB`aam)5(Y!?M7_DM$u!K?IanPJ9%s
zQMlAj|HUj#v<sv00T`Kn*=@?#DY=#ey9JtaCwFIvmx+gopz=%5Sf+NtJ#ui(9Aco2
zIClq+;P)18;yar07jS^X()9{>se3p}(AK~_{z(NrnCIONvn-h$ekE>2p{*7pBJ4+J
zCvkW2K`vf6Du#$U@LV+14F<V;A^jO2H;6=g<1S*J%?G_kd-h%YXJ0PCPE_!v;qLG%
zgBL0{{(*j?=}*B|t^CSOZOUJktJ=a5;iyqw(V@s33!haP#k}olpWhC;4-<c(!pTa+
z&kLm@D5F9?8+GBsmH^V}8AcPH7XBmpe`5xC$=XBbu(vD^d08w>_)DrN3cYKdork{|
zm%t)UTK6-)C??ip?)S!EEpGm9r28LN`(qp<5-nFd&`=Z2knYPKrjy^yB1i5LP&Hze
zdAb$1heP`OukGCT)1e=VT`&iL*xTF=O6vGyuZa=!qgwLGLn#c)%$jM}5>&NRe+63-
z_~Ki^_5SCZt6Ckh>9(;gz9v?y5w|hdO+S_T+iXJp_n+B`7(+x1DTGgEVOc|${IIGk
zV}d8`#Q->uX3R(L;)P7!H5i(G4rP21ub+KNAAu4|vC@49=DGvvA3L?GNzvAHrvp9<
z&hQ98%^sH$`EQLYoXTym6T`l(EVJLQ5&n@?Q7f*zW!fovIx_`DlF(Quk;>r5jMvMw
zNe;3nK}wko(lT&+_#7pohPlOs)I;DX9s0cfza?(liBxXYIbS|iA{o0LOgp{{R1((o
z?a6W9n2hxvdE6}P0lvRhC3ly$9@SXgq1^{p#ZC+a@yi$?s~utbZZj3%OBI3q^66E%
zAyLLB2)Q~DlMk?3Gt|CXMF$-+7f)&QV3UYt!E^{N*hX$suwVw8FuOkaZ*``88`kUS
z-1NMPyWFi}_#%zqU>bD7_8-$uFxHx7v@=WgOKKXz))dw$qbNX<A@J>?KzmK@1Bm8)
zSOtzWM&N^!|HvcYB9dg*ZHBkY6)piz+|^_uz@#!l-MSUZTV+1G$?`e*oZ%~Hov_l=
z0j8uiBAbAJ%n)`|8+ptnrlEhp2@ICiHOQh`>}QmhmBKMD|15K3w9JLg9DhvrvZ$#`
zIlKvV7;qyZmGzoG;rcdj4^)KRED&x2A?+7kkIDvaaFlUsoA9yo7oP3o2Yzw@6`B#S
zmPCljgZQ~@&mbm%W9GL{j{!#B0JuZnvas*gyj-4W<6uz>;m9ZBjQe4_BPQN|>0)wv
zV7!|~ikl|<v4r4mH`<$}DiJGR5Y^oaw1?S!-jLI)Vb;&KH`;lPDNBw?VBC@~LHnZ1
z6RiswXyk!f&xlMjGKzxY(}$YijE@84P`;s0cR+%wP5^XupqT*N3_;g05ke3mf=I!y
zbeiRuM5{XXI8nZUrc?e2bi>_&E6;X}1^miFuyEPI{PGsoa~huPs_0qya}O$U&Q?F*
zhkMshkT6i_Vzx)hu}JtCf|gY#>(FGU`~&+s3Z~6hEtRaaG<``e5!%w>Z`d;NrdgWk
zt@}Mq%|zCz9Qj%QFtfu?pYGRAlFZV%b|Av;NM6T%^Y$rM)5y2UExbv7sPPltGL&?G
zQurYg4y+Z1LT@Bgo)b_^IhV|GbFk$Ur$FD&78Aow8C}I#$D&unIe6cEx+CUSKeMsy
zy&T%`q5NJN4i(hU(vL&fs`i$`sHYJ~21XGHf{^T?*M+~D(9FRE6k&F&qs`ZjxKqZI
zhE=*#5#och9YWEH-{mC+0?XyhXVZTI*jN>0EiXHdT;xJ($e18A0?I8Ql{dEhQV#jM
zH~o*pih}}RUlJ-4pw<?y&V%#%`@}>-%I5*O%fhM}gwmq!1_kKnvsUNGDtQ#nOsA1r
zrr}r!bl>u15|fR&W6ZSY+dxX**BICwzm}8+2557&qzHc%6eQqJNhZZS>TOBVAp2u!
zVKvc=28&6U(Cbo>`|b=`dr?>&!;f!zYwGYd`t(95J^(NFiTDjHaxHnBzA@9Oz}piP
zJaoQSce+_;Yi|cXPi!ea(sqKNe?#w{?nZCnvEr*E8M(E45I_gcAEmBv@`f%n0(^?Y
z7Wd>?$R!7+BGA4>Xm=yFx?!;gU>$u3qIF1kkxTDAQ~S2546W{RBq`wZ65-ln+hgl!
z`ibvM1P3>KoJ9o2FPl#Ua9cD`uKuNIYB!2I1OzI3rZMFWRM}*$Aia%jk2N^uuLbMd
z)7CaX7~8<HPpxX1CVrEqVr@tuwDUn{_yw7fJmKOgc^BO{s2d3mz|wC1h`ltN@W=mg
znr8o`y8iP+jlO|uWCDDzG}buIlI0f%a4=q=zyXUT(~p4R&({dA-hrZllUXvRY;92R
z#275@X^%-AjIcY1fwp^9LIqgmuh9!UDvuTVemZIeMJKG--0o<Fb;<@f=yT2&u-_>p
z-^#$y=#A|ql3O2^nZfb42v##w)qtl*#a;ODj}@>|REg&H*-T$C^bh(0+;H~k*GtY6
z0sN}C5pyZroQ_Vx3Sxqd5%8oMRzVag@C8dUXOSdIv4c_DBM1mCam)^zU*({tUoG{x
zvZ>}x3e6F-g{bNYnwZ_6rzln@okq)7wIY$Q!gtmUENE<Ij>@-ThjZruo;;G-?r63Z
z*1taOZ`Rq6Ck`bacvOn*P<!o&(X*QV)3qOmeO{1GGXoIMy@i%3>Wv05cMQOB@T04w
zE87yRBGE0lHxQ`f^z8rH2gPCxrWCgZ-NN~3j2(ZduZS$??srHV;K?`bBPHq`8b0l-
z&Dx>_liS+z*_b&chw;yAFU)FM!8wc*Vz5}UFhK4iSZ*1i(LV^F_P-aQ^%yN2d=dXp
zL0Dwuv+mJ=?W}uY)rh>wi1^@=;eEnxigIcz4Y|m+X+aFe@qbj;6G4;ZMg}#-=t#3X
zV2!O&yaE##+&`r<hpBqPUV3>7u~Kr!l<@I<$oNhgpvPY$pKw3PYzId!v&g!m!M*M4
zVCpV12kVy;D9C1RFf4w8DgdTD-Ne>k$LhBXRDm{U8bE=k{uQPLVBkAO-uLOq9Po&&
zyw*fGLtQH;6cc*ez*EmA#J~8a(#4Ny;1J_16^Gd-7|#bYpAji=Vo;`!HbU8;BOJC^
zd4*KxPAn)E>ENcen2@25hj4c|dU`K9R$yGQ{vYZdj*PXH9qUJ+m)Q0kUzb0#jqJv!
z(TUxmzotq5GWQ~NpC6~WUC3d1j(G(UZpvCG-6G#MxO~?oS-6Yp>7PN|7O3bW3V;8c
z{uei6*8vR#02gfelV7~A>#0OzG>j{!`uDJm7=a|L)Qp}VSJpOo-_+;IxT6}k5@56m
zYtMiVGev}<zs#$j{t$k;mkn83&=|zwIiZX))?ujQ2{ctA+rjOOd#P3Sd&_))S)wJG
zS%~Q_pW&%*#=ta!@W)l&>Fj~<>*)`w+b%|}PXT4Dx9?y7o1Z{@0*4H68^ga1e6;d_
z=k={VrD%2bVhQFMFh2MnF9rgTV4;8j*rWT(4QZ!(BU+}nqH)X_nAhdqqUCC0bF=`i
z5gx-5m8w<$cYjQss)gbdKyeKUad7z7Np<6^jwt1+L3wQ7{_y&oyTO2ZsWk1p<j0LM
z&SB<D^hstd&HpdY#icu0M`#rV`y7jf{~I_(Y+&dSLrmZA{qB$f4G0_Cj30mDl|1A@
z=xF3DT~clte#B6^__8k9NYiuW@I9F(<lAagxwz>#O0ia}O5j<MW+O_cESL8K+@6nB
zmF^&(Atv$#v*LJ)6$7jOAn@o>nc@H7j(vMzK)Y_5UVP4cj<MmNyw7BJoO0lpmB~L{
zRaKmSXQCBfDSZZuRm)l}^9ueSSzjIqb^iZ9Y0)ZKyXa17D@Ua^xvz?DM>!_RRja6w
zJI4%VE3u_=wInJ>Ora$=qeCi-klbgE#xNsdVwm|opYPG<yPwZ*{@a!DK3=cq>-l^<
z9?!>9;n#?=x~7TmHiB7A+dZIV%k}rIYqE<0kg~!IG-4wsX&Q5iPy(b4S~7gYYg$fP
zJ~nOpqi`%JN2l6+8M%jx9@(fd{D!B9wMAw+GM*0A^DJzdHWJ59r5gM;owox-GEY7Z
z$Q<4V^>54PSVPmRL7jK=*<iK@kEQWu%`@nMB+7TzVB9arLYPMr+_vS72d;sDljQBb
zypq`uN0#C7vMqa$XVJHWM>WMqoE)!BPc4$#u<^PT>lUlM1}L((#Ss<_|K;Qh&CGEb
znW80tY9Gu0fi$^K@lJJVTg;YjxE3oG{TH8F*t>M@oqHT}r2b!7IH2+WV9l#f2QUB1
z7~Ewu@=n+fE{W8fN{F?3wk!e(3w(6ZytApc%5yXpJ0q@QA7>A_e`<IRvNttCrPHk-
z){DHThW>TA_&X$vhj8>{B-z~v)M=KIS36d`Bg2;h#6heiv-VwGDwEzNrQjy!4|n#K
z=5zV&MgWwgC9~>3$~CGAY<eLdS!#nNLO|wswZSLayQFEad9|!<5^F~k(@u2S6c!vK
zr!Uo@=z<NFYeGy<CP8yt;qKEjp%C<1a974V>IcGj5xz<nr?&*nLKE7zZ0z8k$qdx~
z2L3kiqeoX^pWJxFrwpoO-9JZUXpa^{d^cVH(BlXFlY19NMNW9Y+szd7SN@lk%jG(u
z2sz4X)gZRF!Xe}3Gy;S9RxsC+lltE>H1N0^pDMY{6uRaOye2(mS4(?wLqE&$6}Z+h
zie+D_!uMMmpqjdjdxx5E^%jrXaJh#N6SufAdc8YL7HU)wQK5-u?LGzp+ORo`_-`th
zS))VILZ^?Yk)#V;l>aSIGgWr(+*KrD<6G<zgkh$673`>_u<3rMW*cQwYL~y+oxs1s
z0?lrDREtrR+{QU2Jg$+0cXe3Ciug-kE=2DsFmsvfy_u5>LOu-X(q`P>Zv?8W&`^}4
zg@y_gejTy*I7{FoNnT<lt<Y7=HGIopH*L{lBW)E*3vcp{zSrcfL=|kHDA&C)`5Tz)
z5SxOJK#T1e;Z_D&Jr9dH)#?7Jk(YXIAN#q5X7D@uIw{;ii@)gS?XU)vjGx#HzZU!_
zfS>x&rFKbyf7?Cu>ylAO1RV+W)wG^AW%{f0EB|qKuj{J|q8$s@Kcm{*kO{IrRQh-j
z4C6q>WA0@=AN53%l6Wk60y2)6cxWi%<LD{?vm)jay85|p=r>tsE=RyHy)o80?JMTN
z)6`oYjb*wSa?&=*^!zce@CU!Up7xwhn_Th)CWj&i4;NVUsLemXub9X%*9fqjXWE<s
z)U5@f@00{R3{%)WDeQ50v-+T_9$@Ht+oS}PFbEG7{Vn+37K5J>s@Vq?N17;3+vq9n
zJ!!}Lgv&_$^1B_IMLOfEyBen15ik!)?qedbz=v6;iJ4!jPJI|2U|!6a>w`QZby>(2
zmV$Gq&aP;9;>ToXMM%vOfFpqJ8iG(^PKlCDIjBtF{+sohq=j&lmcc}s?dk(X@8io%
zlEcfyQm3SleK^d2am)NpiE<ahnxG7>*YvbG=c+&~&H!}q{Wu7Z_4l!X2#6qEXz=i#
zPDywu9k*Fz_5me*O^Lzd*zkDsprASGylA?*y0oz8!`C#ADEE`O%j&dfZK0PV_6gVV
zO<gsf#5?XBn}(Wli|5s1`X0r_L4F`l2X{}|d;%qRC(Mi?jTKO^OC#*G2}o&3K{^iG
zCV?D|H2)A8c0Ks+^6M{O0_Oa$<|+1463K1h<|>=Yp?ymhHgHQkmPlr-9jQt4j^xOy
z6;DUUel)9ZB2OP(`&(N!rrd;29k%gI`$dJlY^cNsQ9c6h{HqL)KglB7EDtO^ng27f
z+O2Ws>eK{1X7G}7wbBDQK|xhw_XS$~#~%+!!ec8B$rxIBrP#v+k<vGYP)~iZI((zq
z#!W2BH$|Rvf5)--CwJ&k55+oV6w@j#EAr30?w4=zF5dK>YJI2Van+;#w*84|`GUwA
z4aRb6O9T0Ka&mhf^2aDp+CEs|yd-5_o7m5guk4-LqSA})v!Df*H@>1MXbgfK!{(F0
z)l@@|Kd8J8E)qn}IiMqr^;tz&iNb?*myFOnkfzb8eAL*^jh#5Tk879cFSceRJ|!z6
zw~fcEWOU`!zCO9{@>|A~4H@}pNdnv?PSFn*9U}X%0o33XoJ@lPpi27&__^EZDThb$
zN9bS|nL@}?v&7Kjpcvc4;BgFJ^<ryv#6AgFp|6oZj7Jc-0ZxJiAE{Y5{zW@<8vb5B
z-I8h=YjQg*s!K)@oWIp0EZXD6ZklbIOqJItioevSqwq%e8`aDuT35pSxZ*BNP0eY3
zDc9X1`9Yug-G8?1pYK^8d<GtmrHzQpa`OciXx95Pv<-{}@yy_{SY>sUyak{r$)C{2
zKFBDp1r2Zx?UDYnqz4(i@xHuzM)X%tVnclMif{996W#Z*f?#@@V{u~yQ_h3c+gCUK
zlTGDTnBN@$hy$Det<HI!ze$iu!)YF80$*XlCV>3SZec=Jz8gF(ATJYsGL}JkXkj^T
zM^XQynhCwwhkKjSLu0M2(q8n+x$KONHoZO=U)L5-XMavR+|G##yEW*zcC770Ur->s
zt1a>=O|&xp0F5jb$#g_^;4ql40Ba_z$_M1IW?xrILpkil3$f4tFck&|rFy$--}wP^
zGQ30s@9-@RymAOR&VR@q49?iEr<XD9*_Vlv*aRIl&C(1@Tg^doY+D4xZy8B~R)Ylu
z9|?h=x>;%iACPQ$qIoJ88k^Ch3<8py^BMoRN}M5AxV>X<%Zc;enOPth@HmZ+&Bsg-
z97aj1!X-eLTpVkYvww7t_Tu-4>xLMNh9Ev&E`xlxoe@Zpk#vV{NX0{)8qVE^6b(i@
zWx%NpcfNsvIVF1SDtQYhAq{bnE`BWJnxE-`i69s}3YBxg_1wjYANy&FBfq)&teW@c
zSLwDi?^i`q_l{R;t^@q+K-Nj=PxDCHk5$gQ6TVu7*HuQxh^B1Zt|WWHN&hwBU&){B
zdKb}ToEm8*P|M;V485sX0`3m>O$iVs9D5`kSLjnTWO1=*=Z3Tc%^~WAax%&5?5ZbM
z%n!9WfuYF(>G;2Er8YEm5avxo(i=7kbO93U#oNnRC!utRZ<Kg+squ^jBkR>1#HK&d
z=(t2#{DaB-$$1x!GcIxPkSLaIvhW)k%77Ft7+|*B4Mi`|s=JKqZWA^GOwI=JV~@jO
zA4WXl2Oq>6>dqI)s>CAGuMit@oxo6Ba2FE$Jz4PI5I$V?4Spg>415gsY9QPIxJ)6{
zWu}2>h24F4^D>1)ozehg+|+t`)l1VZv6TB}i+|AjJlMIx;4pPD2)22G)_kz(0YfJE
z!)5j5(?yFi$w`ZD?o~XLSvB2Li4@qY#tNtjeRgY}M!#FI@_);i81@!^3x^}P6XOB;
z!EF^9T?>v~QXXL$-y#6-le`3`3zC?0qQNj+tn2LooXn20W$?Qpvxu>{jab!A9;U5a
z4<h%b)tI6C$keH0@z~ze>@61QTOsh#>K_|CAhSl=@XK!2BJEs`+Iv(@X)hLCcCifr
z@n~DOVK2?o2o8brd&`i5smO(u>E~UE>Zs)kWZTyY6wg5<xmRE2v%J|8Y>8)BC#NT1
z$rn|#V{10D-Z5=S!5Qgo{>NLd5@+g=8_+=1*|RJuQ=CM-(JuP(!Rwy{ucMf76d@#I
zoK63r&XAO61P91Z+Z6XwRc-IV0^F-YO>cX1HskSAg(1s3C1-)YS)ae9?YlouScFd}
zq-(^(wCu2TgZn+u^y~!{BpkXRI@wTtYTWYS=sZ$rf#U5$5#C@6(e04T&PqR@$$HnY
zSu(x>a<LDG{3+;>xLnd54J{8ZF1XR$N#$gg@PtW^;)+J}kZd=Yt+62Phy34y8iy{(
z5Wp1eUxFG^du3+38R}6MaOB~ZpoV{}<{=6O<d5X@H+&rIU%;5KDyo-|ZBkcI%D~=R
zujP`nP-NC5gmDE56m^X$J-z}QTgxC5hzJ$X=9jMmo(pyt(C9^(z8}QW*@B>r=u#76
zRzab7*21-E=63?;Qi=G-O>UhC=9%g_H>pei)JWLK57%7Y9b5^k?UW?>vRCm5&!Y7R
zPa$TU85_oc1}J|@T#dADjKJk(EEkkcZMRw1v?rOxCXQVMy*I=Jx4~~s{L=QTWss&^
zq0Hs*sNyi;{e+e&b>KA$ciwyj6710Cgx(&aEz3h^9YMw>-NUwMlH$1!kQqu|m#wNZ
z7B#_svpHY=o<|w4Q`Z!J@yLlz+3Ht`sF>&JZ-o3~!R5Eywc@%Z#)@C(L+~0J)2y;T
z^XO8)hZP1A?q53rUrgLo%7(t4Hwpb_$tP0GzgfjT=YraS(n_V%%|10BISk{Lp@heQ
zU7@s{I|)0o{_der8i!NZ*~EU=Ee9)jzu_xy99}6o8IQ{{$yk6ai$%T_xIzn5_&M-R
zaGQqxD*O(G<(<#C74V<spFp=pYVTC;|IB=?7j5|NI@x=LVfpSe$qz=@wO}qRx9j=j
zjlbr|DWXZH{N(=ml3C+VPXOkp3sN1f0wy<`2+=gmg!tI2wVb5?48Z#$mRRb}Nu1Lv
zIpNo}aJ0Fe$IPVA59!g-%o?9Am8+{Gy*}+yZ$!e1z;lJ+@$*PN&`(3Sj4>wc{V9Z`
z&>S6Y{zcJFb_k@(G8=a&Qe?60D#-GRc$Vq8?6z0e1=zuYe!?Lfeth0HepmZz_riXU
z(NAD<W_GuVx?i{zLQKawRV^^p-;G{0-~E|-h{;uN9-aw#lm&%$hW)c+7XFk+b5x;e
zkTZEJump{nGM9wf$<tIcD}_Np#RP0V6jZko#_uKf_CbWNP3Yfp<S7IuqNho6LQC`!
zggp3UxCy!>?p-|&4%`lQg8kSpFe!weN32Z<r*V&8`L*NWRzChlff7{9<+BM<4;LJ9
zXFadNwMpA$<8$z&*`I-<T#rA_DZW2N-VF95hTnV6ch9{ge6cq{{$Z}9c~KFNpS{RI
zg9Fb4vDG-RLAcb*AuF?<_W~S_>WXFO!w9OIcg6bk^7O3L7=xCmL5JM2Ji4e%*X*V!
z%zq#9QQG$8;m7{F>+w#8BN4_s9pBrh>;X4Ij9ziaj1va55b4I6t2V`5gtQ8t1K@x-
zIH<Xds`0ay^w8@6bHj{tJ{J~~9&7$b$G5bx@nwXwVnWze0cRe#@83fzwDVLRaW$R0
z|Ec8Zz|oTCnnio8sV@yB)&cSzP}mTZm=B7-2Wqm#*-Nf~_vmCrW)NS=WV?;cCQ(w-
zu_4WG8%u9kf;Kwa#=G>{fcILwHuGA6(9KtD(n3IF2vSBuI;k%mWyY=Y23slCx{lqe
z7l}mQwMA&x9KFoCdtAb9uP~`P+;ZLM*4X#gW1F#F`nW7{uK8H}vFb^YTl%Eknez}`
zN&5|Un!9WLcR~C>-si+u-9&`-+{W1>t!vy!!N-Rti+~6;e}3XA*vj^703s5>{osyB
zu^U60Z+Mhq+cjc0fug}YhfReEO$xC#h5HsmyLY{=8-YNLDM-OkfpcVk%ALu?Arw`D
zx|Xon$9rz}1=6gHYReidn>xu-8Kx0#wr6r!vx00fNvwXV0C5+qJoKKDbJbHOl1Mu^
zzJdb*;Gmhm3CLD3hdDQ{Eh<u9bu(TMjJsz~1two&E^Tyq^+Ov7T|qMTzpt=(@r|br
zaG9&Tl@5H_A_IGDI;EcJkFzb1(e!D`lr}4~OM1NE$7X5g4rNMETccbA>bV}d&6CB!
z6-=s|KV_s{j#Ta=aMntEL4|S)DkxYpd)&(2BwV7P==o|_bcARZp&BKb9|mt4rcM|K
z$@o`|fyf31a}+?>>f=pXO_KH%B!y}HZ|Yyu%V27Y_bb7>41G-lM9b_^pdEv)ENxDo
z<#uSl9$^1Pi4MCCK{g*bWWoyVO@AiaY{+s{6!cHt8o@_16-j>8gL?OnJ^ViK%g3&@
z+6YJgmFgKUnOsyNIy5v34DDlnhegp=mee9}3;*zJsq$0MA!pk#CLB6ke@Sh+TwKJ%
z<B4Q}Vu9G8KT%zbgkiQDQQxYcAm9x`b3l3&f~_u(x-<$oDTQ@}=mKj(Q=_oRsTB;N
z0Z<9!rgocvvQO)Q`|%qe!SeC$>V_=0u;3NwQ~P$6COu#HZM<AxNLQ@2J#o`%bUS%e
zyc1*((fs+n^~9s_tmNOAVvanjVk%a=El1NIVi!*CUH&_rb~!p-zokMaX$!XQo&e9-
zhV=W@>3{&7*^qE1lUbe!_)~bn@^3&{&n7or_HKj8L|K!!@NT=mz@LQSb)ISVU8vW7
zC@W|`?0oFTfzbu<FNW46L@K2XH`!^b#j<uFy!9*IrSn>6vS{zT?31b$<xdqvj19}L
z6X#f`f@NpmOx{sH%Nw7KD_(d*K0~{oxp`lw)WOP4*Ed{60U^Ls-)=sT|5K$iz?YJE
zn@5?yU>=E;wk0dAp5(G7Z7Ps@8+czJPF8J&fmpF&N>w)19ovaa8O*~hwxgP(*zV3<
z<9X3(rvUscE=u&bgxCqO1{7ZP)0^<lMn~s(-_p-Gmam!6&+#9YushKp9R9(hhh7m6
zSR94hDnSlpHgRXZY3hH47988^G4Ge)gxg`a_5wXIZv1pCpyCBx!IDL$ol%h1PJ1yJ
zQl0pzyB|$$4G7bwg~D&!oBq#}Rys`WU*2?U^8uj0@DCq%)xpbPJ80b5D73{s{<|WH
zyHVQ8)n~h}rKYbpj`ZjelD*kgCS4qx7L-Y4!K!j=)U2lXViD><fx5)HeyV&d#-36@
zM^1|*f%)x;<M-kNaR@`R8sJ|6=S^ic%Fcj=J^Va-Vg2zdMP0E*;|m)oP?L~?Hx|w}
z6^BJD<*-JS6o4Lcbw4Us*|{bM*`koe;Loa|>IskWUpxbiqzB)glG@Gmu|5bmb*J*V
z&c3i93Y$O5Z4Yu7P6ef8-cNSGfoDQ<614TO+pXS=m~~|(vp=5chc`M9=UW;L^Jg;b
z1rj}&U}zDpcmbkolt>E<UNtzLKyKZ^nDtULCFWsFFVUUX7&~cZcGyGlOq(}DsDKRx
zck%iJ&cQ4<=7v&T6Q(SnFn6?mfff?<D+>*+cA>ZK%UWng#YbTuAgJ^Dbw{T8(0fTi
z81Oo|U*qFOzUO%OGwmMySLQ}>XA)Vpz&PC-sPU2oT<>T19d0iIo#1E9cGoBNu-5e5
z5^-M%&==q-$J8~!c)Kb=k`#PZ1$C=(54!kyM__m?F(n96V*Usat*2odeNo`+vE3k9
ztA_n=@FDHZJ6lon=U={b%|?WVLa)mDg;7mRH8-s3*6go%k?wSXTpF~k&QRCBt)5fo
z^~I0qV)9=EQU*z}&k&<+{1b>ys+_<u<Z@t=SwR8nH0@)P1HO1$OBittaas@TVjf{2
z+V9^(pUnparn;}<WCaW*Yhm+DYj&J5KjVtoTlU!@YL?u7&n5M<VPCi5|ImY+drg-4
zg<E1!cs8qn=moSzSCKH66pOr85ajRzm3BO={0`gL?oHX*pf7FyY-?fTbi2GiDfr!A
z5XwS3=<j1{98(rvBl>%O$zD-ERZIBWOiTX}m|-se06ZG(^#qMhoLjbOCQz=S2);NO
z8wXE+lXAAfs{`K&(A4^m!*s6IfY5+fEnXCq)u?ZjoJh>)@Y<1wvggYicbYkmURf9e
zvYs<&0`S29@}BP$=6%Kb*OQcv{9yQ_T_5@<%!mXF5P7<J(GZwOj2L=L;kbx!KgNU3
zvVDWk2=9!^w9irnbY@s{{<Fkm-suY*yQlyy4;L57^b&G7d}qz;3DpA(<>KywMROb7
z1>wQ`Tm7D`{QsX7&0tEGv+SuB=PU{bc9~#KGrX?iZNSjUg*%;}FMxs9I|Q83ZWMK)
zNkeldH?l~IKabRBnLp8Eo-5p!dvwzcv~ifDW`mi@LditZu-dFOAMBY$`sO$7XL=K5
z3zu9#$c&3U@2`NK{r%kM3~vXUl%|QAZ9<?^_D?w4B;+_>mRL6p|B3!~bJ3sZJOBq7
zQ#}YcK7Pvt{nE!eA5KpMMO>H$JFXxuJ(H&a0oEO*l25ffmmzKSLI)p5GYr+@`~!sV
zo~5}zlCtTj((j}^4l|EBj-v{5AZy~TMTJm4lr0i(u5@G{dlHBF@?0-V_elicdSTMF
z9dalJ==TAnA9Yo;9)qc8^GyekXFpC8$M6^jcBNLhyl8EI4fG}HI+vSt6xIsk5wMEH
zC?n*f?H#asvl3&OXWV#qw8XiB6Fdwh2H0A!KLNVWxBar~fM&SC8=)H_3C){6`wEiO
zS(u!Bl9iXa!|228K4)_C%lG1%g**M8eo~ArpL|rX=&NjjsP-3)GNIw8<Htsow|;!9
zb58#yrFTQEveOG`PK(i_4r#SmrGott?pfV+4sO><=BcX{K3g9iMLP8O^5A~^cYjaa
z{(0Wx!hxC2ro<;=b(vc%2iIk$<@0r=mEeqJ$4*uY@SMfYWq1m|r^16hWMN@Jm|U4#
zG^!PDFR4DSKC`wrPk)&!P&*ZxJUt^N(hsCJ>&5%dhUG1KbLEtv#danJjs!xI%h>_1
zOfvuSU)&AiZ}fC^b#I&xnm_p##}~djqi_0($Ko79EJ`2SNF^nS4gFFxnaS(-aL(UE
zDqo)bJi>@VSH~UMDT~)tPDy@Y*n9;N$H1=>LtjRj0OI(dbMSGycA51N5RYlfWY2?{
ziJg3Yr0pP%ZM2)Mu5OPMo#8+LGU?R5+UDiV@@h4K;-8cY6BTrBg>9)1hLda3$cFDh
zIO4$k+5SveLNWc22xikhTD0e93Y{Va=bb&UHuA;q&FeuRlBV=$*aIi?_1XY%n%yb9
z&5D3dZK?qQKmYN3>iq`BpvG(<n_KIL@S4GKTU*;sv_O+{?UGGBQIhHj&YAY51pIqN
z;?~CyT1JMTyoK!vN{6Oe%Lp=%llMnJxVqRDYQV_14{M=2rx`r`&iPEt4z)MX*sE=$
zX10K+&)xee3<&ttwdXE)#M#*N^UeTcs+OV==x7yFnDcmd>)xg)jZl1iFCiQFW$_^)
zh&34*Ai(;+cM5JLVUfS6mCqw}ZYgD-Ij}9fSE_lvw+;ATj#`KHHdJ96f8mi=hcB>3
ztDn+^-x2T8;1xRQ{&8u&)8Dy?+h?`Y64{jgsnu|L?F#`E$6JR~?w`WB25<3Br>v?Y
ze*;_v>0ASn>pVJSfp)kHbg6+kiQdEroc)*IqHt@ds&V!SlQFGlGR?mS5Oj&Z_c*m%
zqLA(no8AaSQky-e;AP9dplZVRIYE5e92>S}(Fgb`*?YFC`t#R8yZ)LnRYxmoyw?2U
zGiJh-@SJXO&&~&=B;KF@OhF*v+i%mPHT6pilj4pFOOout`vZvwMQ7?S;*9Btf{u+M
z!}oGoF=4mC+(P}uMVneS$jn1iWqL{#Y%0-_4;91jH#n1iZ%Cf#&B(|gxK9KOPP~RR
ztN`W_UddCc`a2Lj%z~KDv1_k*5TNhjp`pC3er@mzwn2HCdnbm9<GE*7$e)8!aKnK?
zPw<0N>Rg{TF5`TU*o~{hBCTuY?FSNsY|*IhdnJF_;>6jXB<mP~e>Xi$n%W${Tq;u?
z&gBuplj+J;rz-t(PFIU~{UDIfhuYyIbfzEsRDG%X>z6XKH)6#bpX8*R@Dv^lGRB=J
zdlJAExz=%Ak2wg~DKa-JJDqc@{y2$A$hps(-TVe7mhXKq+q<w5n<yjDw$PsQ!P};?
z!NQ@1xd?_Tfk<qecAB8}f@azpr>--@VPF|7#bk=7U@5-DETLmBytAY!sg3#Onm;)Q
zi$J1!;Id&(%XpT@X}|ctH-XSBo2hEi9xDclif^7qeu?H_E5?z0<4j%{38GD+_u9Tg
zN}ivmV#=?UbF}xYr&ePR<WceNYfd%(ZF$)3V4rYgt!+aJQ>5YoiNr*oL?~PT%<V*(
z1d017BV%YNr;+l3nO4~F-Vy?p=hiTF9UUFY430zW*a1N&i}%x05MeYmfCFO+=SQr`
zgJ$9zXGUmjELMU$V1(la$&tTa-kfaE@Wm6Z8r;-U>Pd$noWMBTwzrbYUiRe3%E^WG
znpGH<RKiK++RKl3^>Q2HOOW2HFlG;%ss51XNK(z+4J*;c!B)y}-MtpL>z~rkgTrTA
z3U`g0j6yAHQqH~ia+6ulotkzbvZp`tI@Ej?FM=QfBkK-JX_Xc;Y_J}a!{Hx<fd=e_
z7zA?|Lp%nPUIhygi=n-VU60qHp^`lZvhS&hoW|J<8uSx%B7|j2DfADVbwq$9dj*md
z52gnF8`}Ty$^<D8$qB^$n<GWWo}4AwX&(KNq;Iw<mhOXN(ObyCej8cEKqrz@n9&jr
zoDYVqEJ>365*3$hTQkvRyR_hcfxMFnVjA$xgRggHBy}Rq@oHpEAn)!2^0)I;s_r3w
zUE|tXBfX>IFgyaM=e95VsL+vbG9L<&qY+^5_1WwYNVHb=P~?c0xK?_DGK?6UQ1M;f
ziVZc3CCJ?45_Fu%<O_#aDQ}Z+*i|SqbDFyL;?Q^cbbS2y8WZB>X(Q+?_h3^M7^Zpe
zE7fqegH1wuuMJx*H6ue2Sve0_Bk`eaLj5+|R_b!s!Zd4Li=(e}T#B02J4S*0_l>7^
zJ#!_}wD~tdo0)Jd=z>`mr-E>*IoAe~gFU%}q^?x?OP75QG99V9aDzOf+Df?nA2FG!
zyT`3eUr_5U3PBR|cRLL{yVO#-8UuJXk=~y|9b!Y{Y(1{F`%j32;PGU>E(*eeJEFs`
z!0VeCf?h3Im~pTPPITdSxa*ocJG!&r8+7+*a@synyxb0O8F8w5sG%^u+rHe{;dSsL
z2qZ9}7G_%f`@N}<w(7VZSiWXHDLbp4`vm9pAUGyohOMOSR>8@c%!lpEI#U~_et$@g
zB9T;{gFMZ^fY8#l+mgq8*osGtMMc#pT_PR_xLe>j)H|_b=^!i1Q!ia!Lx&HjqiU|>
zd6QnvA4vAw>$@vc4{*-j4Z&@bbRCD0PRLqvqqB(Zn7*|Tu(}h(k$?qJanZYk6O?*x
zICe>X@}+TI$zYfc`MF=N3u{+gxcy>`lPb`x8M71w$aq%0Nzce|lXG1d<y&$7+c3;p
ze(>!p%%Y5gWMNa@7eCx9oF*~SO*1pG$jChLk}Ey{rfNM~d<*7<5@1KOG-@>@(Zp}H
zfQUEyCHXnm+3pzJr|gP<Onxby;!MAd+-M@_D)G7~=-XC?y_Y`~8++UTg3YJLD57t!
z_YP<NLJFQZz?FzEli2p;G$I=MD~PZe3OGr?EQO_#Yhe}B|1govd?`wt@f5)Ds1vYu
ztFilp9CWn9H(<8k=y|IO>aUw`cx}0$nUL3Ob3V?_#=;_+s0_~Qg+;E|BK~6QPe4j^
z!IFnydO;9O->bAO&!s~kYnus#I%tE-x3giX5}W?;IYyxSCD#G#A@Pm~5*08FY@fZu
z-OsV8fBxJ&xgDBlC21EO3%BpO%_~c#$1>Q?(*g0k)oERoFOvsrtKhSM=>sqwOhc8d
zD`N&WLHLTV`+70L`CjnO2)J8P?oJDfMvOq0@`hk3z5EmkGHSoCeh_A=>n3~q(9g*>
zh_2~2HdNd=6Mc3ZD4ynDh=+eT#_=z$D$V3suW2~>N%jElxvGn<m<I;L=O$;MtFnKr
z$rySX#jn;<L&46&MA`H7edj!z-`jIv=#zrIHDK*jCG@P7$}Zp1`fLJthYNIqLELPL
z15?}oirchm%}S_BLPIy2+=ISrTs;do4$U`mYod9pQ!n*>eHxO<ytP;#jCH>T0%BL3
z<id$;@UF?@m<f?`ZIIg^XG^3^33VzHN40^sP1BdJdD%aGOotV`c#k9st|DYXvGjf=
z_t?r3U{uIU@`c~wD=rnSr~^kX_W)P*5!%@zqKXZLkS%DpzMxg{Ry<l{YT0Lin%Ef`
z%Li*lA)$ep;7REq+#;Q`T%O$?b)A9=q`G_8qu(zcpE8-n7z<)OGy@4v#h${@M*HLi
zhDowYPhjN4ES_PK`G$F7(Njy^WPVM1!1B6@0}15fa#+up_2Uly)Vqf^_`s&j&JoB1
zjrKKZfW(Zv{9CiR^D6BpvBV0FG{^8-+^*n6`+HN>UVi_;@oDJvBE>;_<fcr%CrfGy
z0%HG!mqpS{CEJplmsv*eajMK}&sRpLiC8<VOWfc=7+wcclm_h#4L(dL?*&m%BjjVa
z<q~ZNYId2q1l%lqJ_AX@Uyxq3FI)Y9Xh0)%<{sbTTXX<yfF8XOB6VLp^1L3;lRubs
z-ZW;>jSE5dJsLi4KANS*zn8qLF9bftvP%0pAAg<k_4D&vR@W&6sbecn-C5@<BAXw|
zguzwS``9buZIkY@Xm>y}0vq)gVxu-v-ZzCUu*v}@QC<7^6Ygpr4kjIkpV|3(rZumc
z%UC}#m9c+5xcb20wz#syEx7o?v$L=Q_qo;EcV)lmQt{9q;MlOR0FwyZoZuyVF|}=S
z;s#R*Hf^#v#pwPcF`qgCIgsyFY@&=vKvH^r(hR{~POcnnvv8B?pCoDxh&=ZU+GYAv
zgXplT3b&8<9H}E8O9Z(5AskvGYT>!@c{dup7*ktx_(YQmq$Q)KF2(ej0RSaP%;=AX
z@nx&yV7~8O48r!Y#F2^bG#=KC#!|yW0A_rWg?mbJfir%1f<qfO&MGTzsohbH+^XtH
zoYm(V9u>c(dgH!v%&UTDq&BJ@I%eM@BYXn($|=<V9#qv(Sb@8gO=;(nL0{RP9*?el
zs%u0@PQ@zwG{h}Cy`6t@XcMvCt%WD(O6d&1Z-OTv3CG#$@a@qpkJZBO*O&7NdA#ou
zr8ag9xu`3Kr`@?}4|TlO1y}D_*T!}-quaxuda%|P&d%RmO6hkq6CrEJ$~AeqnfcJ!
zrc8lrfVbH!fm?5OIPbzwA93LL*Ds&qCpr&9Fv@L&0^Uxic(bW3sQAu$LQ&|Hj<&L)
zZL{7+(UWy4o+I!D`cNCMkQ4V(X@&%+3u=Ioz9)He@3KngCmWzA-}@K^&=Em~REK?}
zY2q+zk(-&+<&4en(E6$`)e%jx&WWY8qlGwwOKvfI;8tVI4(m@-oMrFz26x4<x>DHW
zX<iC@u(+&?!BN=LX#_d6WPa^w#uQvx>za<!uG7Ky)Ka*!M#XH%CZ!l<CPcf!*tU9Z
z#eZs%HE<64&swphav?z@PMIR&l(Dyio2+Pr<|z2GC(uaNJE`r_kV5fh%;4I;S^KGd
zUV*Hr6vo;>o*94OOCq@S(wexvGXr0sx%fWAL*nc9>{aj{&42wMWhAS=U?h*gjV?g1
zEs>)qjU4$>k6BQ=%x@`jR-6?+RmEW`ozG;IrQf9L4XQ14?0RPVl*U||KKiYiT^WD5
z&W-2K;`%%$PjmmemtTJoF3l<Dt8Rfa=WW+%e9x|3U67vGB8V#K9vD~>S^SWt)FHZ`
zqftM*39cIC*uc}$sd@e-(;!uDeE{2AB7)2;jNoYieL0EfnXL(#Bh6n*n>`*uj}R#8
zJ=5mHnuGoTM!R1_?dfESx3gYHyeE_&u$#3C%5G^baEldLg4^y!ZK6za{Z13e4jCMN
zhn``(j66GRe4PJHLK4FIXb&;bxg%sqZF5I9nv%bnErn&6mfMr95HfRQTU}MGGGN$3
zB~{O#KLqZ!q?*A9`W9pe&dz)siw<aAyt#Tj%d@Yj>$&TG;b+V(8Aie#VMTuNxn(hv
zL8*J;CO4+rUx0Mo<Fr-US1NxDKCDWN(X8>!%-fyHglJ{yS2*o}O~ZB@8rwB25)!7}
z4G;%vo;yA3ochM0P1J<!xqDaC=?bQLG+HR0KV(q&IA>x=*UK8Xpf?RYrI>oF?CpUu
z*q>RIM$t91X;Td~@MfSPapl0`a<{Y77Y{(XX{mUi^BD|SjC#EFuZjJasod$Qc_h_-
zq8hn!74`xo^TEC?&nB*^NgLk(4RX*!^%n~u6S!8OBM%|nzcA;PiK6N~2=d;#iiZYV
z>Whx~^A5wH8SuYlg3{dnkuw+mYTSk=S*nyZC;a*95b1pNkXs2h!qSroPQF%I#BpTk
z@LPFm@+BxIF!iW)UHAwmk*A^-y1a5f_p>oE6xr&km$97$I_;fH?Co1MF_5$y3O9&?
zU10b%haI@&D9t~5{OUZCOGEC(w_+*wu^`1sSS{Q-q6;Pl5Kt_MI-(b5^=S9znHDnU
zPhRZ!GBAEf^Zaq_Oozg$FD8@Tb}UI73K1-Qa;^OJG!l%hx4jO8vr<m(yJ`Zqp|Z5d
z3r%b<?di!-pBf#61GSZgZC)^<vTiksTgYT$79!vUC-TzfZfAU6)!&;uPg`AuuJ2rp
z>u@{oe-DfFH9sZ~Fpym%oT=Cu3#ow>YiOcOoz7jrj1+m=m!|~hx~^qwD7ncyL91#v
zh*@p^vw<;Jh9~>Yg+c1>hfhbuJeHEQT-t*hsLENubcUnEx^<eWG2)rq1#u9Qg>Kf#
z21q=u{?Nd8LOO@Gjdpt0I<`7ce5RtgVMyJXu3_RtX_keu$@h`Ip9{7w56ShSo+l!x
zc>3G4945Kn)SZoY5t5O=7bABB7KLEA4)hF3zYQ>~ZH)3B_;!@R>ccDk*KwNH^6WXq
zNJ-`g5M!o3*`D_N5>#4|S_?<FKsmKBXjk}y^Xo-P`n!{voy6MPe1u2@TR>KQPq}_x
zuX)t)6eM!2tj)r{?AtZ>=aI6lP#slwZx(&WSJstKQt|GF^B}@D!JgV_{T;&Eq6KuV
zeGt?StmkmDKcE0DKj!ammNcVAqu7zVh{hJ#4I>W^a);u!z(fZCwve|akzCT${VE}&
z>aT(OGX?+)CZ=J2`|mgA`V(6$#14|WgX}CE4-QyKD3I9DeSIOZC^G`H{qMMI;2i+#
zL)$_w4h{~g`Su^*__FLk#hdh_I?_+_Q*J{3mZwk};d=I>N5dqPP%0Mwc2<HsJ7s4?
zu}>xxkEGyJm~Amn(<*7F0Uz;#qbGuDJCde%0TkoqJoIbatF?oloq!ZrV=Uo@G%bNn
zA>~&YkmQF>4$}^fKbjDR9a2636biO($w(e1mYNM>aoEvJP0u(3J-2PJ<Qo7wpxG!X
z8aFwL&`BhtXpwblBeRFBMJ;w0!$AclT_?<EO(=Q^YfBa$9+-s*j^;~_NjgXBUdMxh
z%08RJ0B1cmDe82M&sAQNa1ctNociF(ykXx%?2Qn-`4}c)<^BYFJPSm`mr}9!>fZ$k
zmj)-HZM&{ASRhb7fSC9#jjYCD3L?i#z)$uY1k|`B_x`S<17DO4serG+G1C9-GXaWK
zNR!ts{WEkapmbnCcP@6t2LMaJ5dnY!iCZEI6fa%e0P4Vww}~I-PD;o;o2uPpGV=Pn
z*9=GI@}k3bbj}o7-h|vSz9HsBv9F+zfxnZO?A?(!9B^pxG)V5C7QmDyNdv0cndz3H
zF9-*ySU}pg@=rO8ak6@Vg#`ajXbEZ(P6P=6VncKlrLkeoB<cMjmliS%^o*61gl!f@
zl#;pEd;7pEre%}dgG{C-2)cZ_CYBIfPZPgqV1a*2uHvuV9q<o~JH?c$JZ@{vcgd#-
zSJ*nv3r^k7{KsfT=;O~n#@Xxg(}CV|V?fBkY6MiJ*zYP=Q(Ifh)-5W#(lMU-pbz$7
zA(xFw^FcNp^et+>F-H5is}f8AsJ-!(Hzp9;dmmsci3j3w+^s9MA!jk_cs^H#!YaQ2
zov@_+ipE>O25NMsWCwwUyIUt=oz=922#~YT^$2OQyqZm1tPY@hf+%7wZ(Q?Ofb%Tx
zW?*;F@x-h>%WNbprT3A$@T7k=QT<TP(dhw}n6MpHxr)9q#fXH5hW3oM0K4?wr&#lm
z8`*OnspMb-fCxk<59~&$;36%Ioo8WZPm&c6QFCBY`KI}8kz&pD7i|pzP|Xf#r3U$X
z55YEb+0Su)M?~+muVwWs@fJzcDkL=cgOG@4%rPChy!+_|B(1a1Fe=NxqgBB>3|c4s
z$P~{NE*kV(lXcEP@O!}#r^EMcV3a4xZ6UEf6^*y;{U^&!EK*uw_b(O}6GpFGE{<i&
z6Yq<Af}9mz5++14YRK*F%e{ZtVUfz5B*`W0LGxQBIMCm06#pEQQ?NsP*mu=pW~A6V
zGlwah>(F7x8HYVsoVGc{w-Z|}^j9-<4+%oGu>hRgfOnCt`#)fDFoyvgddMJ+>kvV$
zNjSwP3dFB@`hK90!ufX)i#U-3s|0??MAU!kpLma0k}A=?eTS&Uyq`&~+BT~c=(<5i
zQ<CB3T!PUO@zr3>(VvO%j@$|Lz{%aLO}QPJbj6evb;ZYqK`e)PZbxOR6IcI?Q!Mdv
zOuz+34$rq{7si?%gt6LX6w*SjxUA@wh#wuOC6cr~?YxyKFFX)+Na1i?N>>|mS4+2{
z^a}QuL3`$kd?Todb14~kx92|MVW+_QbP=`Qb#%HFc@>WIqNKdwf(erj%&R!XZ{G|o
zjW2E1VeTxTKXKs>)dvm{=8tgqq__j_V=*bue+kQn31&>pU6%9o!lE2R|M&gLiGObM
z7Ac0;z?w_2_N`bZy>WNs2X6`tes7a9?70D%hxWdrEwT`*1rrmCBN$B$>=4S>Ky4*4
zA=#zp0+a+D*oLG+Zex?xT!5z4w}u1E)gVAbr;IURp5dvAw^ts5<mIRM7uEAG$CIvm
z1184%#u1D1#PK6!JDdidv<#Y4lzS)d`ZT1JtsTT)>+X@-=+@#tq&&~5G>w+RW{wwS
zs+rDwSz`E(-_zIE_bX?KIQNulJOzv<Z4AFK91UI}R0sW2ON^F9)kob0Z<RBxuM6`e
zlCB_FrR8`2seBm0L?x~2T-lcxcA}v&bI>_bZQ*TV+>4xTS>V)fy_0N;W+?E(xh8E{
zp5!!PIaSZ)5Mns{nYE0kn^7#ci)hv*oeu((>HLKP#n&1$SRd%3KJwK**e5`8V^u#5
z1|@JGgk{gGk>eq$c)tq-H@6mU?rjFI{)I4!<JkdKR%`FZVUU^h@iH9j{y9}k4?`Ef
z5$KyD6Ucb8nzQ}8%<XT4p|cV2O=o6}%ZeWsa2=XHZo6&E4hLuhB6)&TRlNvjHlTg-
z1%U9~$|)V=F3SqWw=lHfb^0ZHf6gmq2j=J(Dh!6;dkG#v|GNuMk8PyPfZUJtxGi4z
z9dTCIB%BF?WUPhB#wNk{i=6Fjpt$o)CyaA1&_kFj3PW-`zI^%OZs55ve`XcTS4^YT
z8IYH>Sq<sRTdq4H!DB#O@ta(j%5O5wcOO0s8dsK-#MFef3C0}~FRQ;lh|7exrRJ>v
zj>yKwtm|nfmnVS^5QI_2#svCmvs{~!BQ{Uv1r(8bsAloyBN*@zGGY5BfKc|fXPPBb
zJdCXJlUk1S7$}ZWaZ+Bl4Ww6dxp9n9HXl+<wPdcmbEke=*p&v&Sz`T;aEXzFts~v=
z3F6g)m$C&mQPR1Cjnh2@+jfqKrFM@ePnPoN<njK@d;P}Gf?%bxNBv%5cgLs7e8VP)
zRtH;s?h$x~91)J;yb8AQzN4kzmLim$DzjmJwoOw)t~q_ZGIF+38(vxxZ@4)D+JH7y
zfd%g{E%o@ul#41xZx@nsyW$NEj8Mt|`M$(6z^ObsvgN}xG2(P*3UaeRq7gH{;t=Cg
zyp34U^z3<AXY#>gc5w*Vg(9p9z}>UZw)rVY_XKm#a4EyNe0L~<@yu^N%_!SfWLK}s
zC@})f!rt7hrBJ7lg7d+LTpi9t;o?((&8!Z}+`7!(E@hC9=usuzE`jr{Locglh0b&f
z_o~bO%Z=z3johPe0}^%JwJC)^xM(c7Cze!=etf~cW_L>*tTX99B&EN@4wnL5lg7RA
z-sjDt0!Pml?I4}AwXlH9jHL*`>pHnBFHgge`xj;V=bOf<E~vPE8hB6@?<9A{JwkE6
zKRr6!_x$Vk4E4M$TdBkd;*qoxJMHUW0`Fe3P@j3|<6v0j%~4)9gd=^<x%#18I7ACm
zz!gBJQLuFF>2N-*yz}bC^*J;yh(O~=#d6XLAoT&3oIDsx<#1@8Ya_~P5;AB1x4vk-
z3wP#oNK^0Lz|Sh^v1Qn-hZ(QiLv_F1_4+@CtjE}E0?RM&+mCkbf8N~O>Y+sPfk8@$
zHxry{O!2$FA#*P;J9B29yP4;F(ugPYy2NO`+?_nHA?#ezwV|TU^oc*ggs1suK*gTi
z{J8LlBrCq+UHVpzpibPkMtb;|TeFKZ`CYu>Pi)2a#-i2@?6U#x5CgKc>b*}{r%*&h
zfNaL0HoskSVslM}-{FSm-bX}(=faZev&30TC!Xnj`vPsE0RIS*qDD$iCAgI}PyvT-
z0;Gd&OG$%059LkG!?f{1qu9w>mqKqWI+pizPlNc(C35lDvlR7~F3DhV1#*@Yxhs`?
zqgt~-;ehV9wD?;&VUZ3CL9kLyHyAbsrZLceyZ<zO)|<6y7`qU{LzhBCY`0o1C*Eli
z0;gl%<|k@yr==Yz^?P2K2mhe5pU!9V6mWWqbJ{(>np*xRJp+TU{&shX7RKRA;NraM
z*F#KsxLh|bN91eJo!qKlrI)91Pa`~SM?MEwmx^H#Ra;ZD^;$<+{B5!6xeeMqNi}R;
z63053Yg7~j?4fHu>bms9=W<*qu0S>b!ka>%G>*@(V31611Y()#^;|T}^`Gq@(rxut
zQf%-+(bmd8T^BxvA+X(}Te1E;x_>;QVP+^*3mD<+M@!i?33xtxz<xNz(C>w)Ny{7g
zrX`fUtV!+i+a|~5EyY%%^vpdgS84BG7ZF_jSM|LhX>dTLOlU*WnA#wF4*(wEC}5Tg
z1({_@NvJ?TWgH4%P}!dat4I>0?9D_=1#P}wVUG469FzCIi;6~o|I?_dvK^0)FlyTP
zwG4=w-RnvUtv^*qbXk~YC)liJ#)kcPTP-}I3((M}AXv%oEg9NJtrs%uCkTQSA*J>V
zK*S3udU0aWZKp5`x18TB=mPEj)QAjz=tKDMrxHQ)-v*ng@5`VhoFZ5CfVR*FRZn=u
zgdw4gnl`>TmM4IP$(-@U$g~2N2x}^3*2_7_#w=x=%UHQrvr7V(Doi|%gT;ZhtZICJ
z&ELkn8>GiAMu*25K{J@TN{W{9B_yb*q@?7}&|Mai=Q_W9q0hn@6J7z;o)vQDm}*y5
z@h6)=hQp4dc6dQ&#&TkQ3Fh$JwJyzC(1!}i-`Yh5AsejWur)By_8a4WlLZn12u?G|
zPJu|?&r~m5Y_F&e`~I6cW~h>fku|H@kaAHAtj=I>5G~2HouiTheWTMAd>nebv_@VF
z$^){U(P;cLvvIofApwU!fsS(QE#P&LW$kSGj2o%=i_Pn4GFRJXy^LZ9NZ7b2XwoeN
zLy@G-^+_*NWrF6xDUx{3FRDH~{$Za9oFW4tCZE7jT1$vS5!fA9-WN=Od<#V?vJNgc
zKx2sJnAA)yrJ;)(P+z6SFEC@SfI?}N(+uHV56m6N7YR*8-1?D)=K|cvP+;{(F9^78
zB0CdPcJ&~z*Z!AkPc`MK;0EJdU=hetq)~jG#e??Q_<HChyjn?goqy(FI3c7S=Svi&
z0%o`thg)eUh2O=4*~Ff#vg4FQSxRA11m|-l9vaPL=Nk@y9~g|+`X$SKMq`MhPjXNu
z2YoU=aDrN^Q116376OP;u%$ftU_+DRu?L-xLrz%WF><U`R5c5zTCgkoPt~k&=+xd^
zQ1XKMIL`9!-LQ-G`py_tv7Zh?ACO=8DUVBeP~_EB{3I?<0xILK<k?skWhrC!!M*4V
zDTDt+w;i;xE91_z9wP)EMGD3<99qQ7u+=2%+qS+(F8<9~)rKurS#-OO$^e1#VYt|Q
z>)o0F6`P?Qt;Pp&+Yig+PwO60lyc>kFQ~95G2LhPAG_UodkN8^ttR}gd;`8lwk|7-
z5U-Rcp7nSCdqov^gC%<p<}U7#5{tYq4rjvK!=rE+hig%a1s|2eZR~(F$^yjoa6yl{
zs-^Abe{Hs$dr1h&+d`~_-oEj6=cD;>68L?c%D6Mo4GI;&DPhJep;x)!%^$97Dl~hR
zh*-YQTgYmhUsX@+T$nuTVHCgpNBC2+yWPk3wtCF2CmC5}D#~P4jaY78J#!3DA8z<n
z%3(s^*KPrX+n%{FC2ZO1v(6H-<ZNf>pN===QuJkqul^6rg!kam1O%=DR)O0gEI1o*
zyV9^20Ng$gN>^=j`{w0DIyyNhR_=<q&0f7dC}#TvM7OgZo@v9EFd=FzvS4junl^c_
za-gHB0oM>N__sAe!Q28~az@6PfJAn8g^}UmI#4r#3e2f^Y9@`>QE78HrvF!8k#C*^
zqV!S|=Rv5hFLgW9%1u+bsqC`phd6LNVD%L}fbBqhRMCB7U270uD<PbirelSiuzAwm
z&aAv>e0~6Z;kfyiFA-hS0zX)BS;g5YEPvJ+!f@LHz?WYpk~wv&p$<3VLU$^tT%bKn
zg_F&wFT4R12z-S;2)6W0HO~{s>Ozv%<Kq^Jwy|wbEVYlOrUclJXJkxo@V)Y1U;-?f
zHhZ0>5E9_l6N9PFRU*6A;HiliicLbz#;Rh1ONaEpkLw{@lFkJNZ_)@WQ3H%MdxuRf
zdJvf)`}ogz1=i7yR=U#~zGADv@V&R&a4y+}tzG_-m_gK(pLLSh6z>u;N)ZamUWkgH
zULPzJdiy}VMtJm+RN+mC{`M2`CVt(#!>#gpIK0_EL#aT0v$9sVW@ZLygf*AS-3Lqt
z27g>A&g-}!n7Kj<_SrwbBQB43*K+NoqrFOQo5vK>*mWN0NtPfGj-PC|$$s5$7_d-j
z+768f@@eB*a~t430exq!mVkujB{%uR+hW7W`%XUxUlF{ZMoMlaot#wS>?R`*G9br8
zsOzz}OQv{%O(dhftN>2=A}%V=bFcOGi5?JTwpxA@E{}_*k^f|R6i0`N0&ZT_Zl^sQ
zPKY51-kHjgY|izX-!ie5l)c4B>5%cHqp8a9Re`%!8<-&ymu=5IG7q_b)~5KV$(^kd
zKa$>Wa$8J2B%5#HPbXZtpdLmf;7f$qLSGyqqCFkx9W-OVu{Kty{AEhh4&irgx5T0l
zlWg=Q4v$O~(Wje77C9##htz;@ZO;bMM)py9g*540(?r8a*@=}UM6>eee9EuC3}3=j
zRPF>atRR4hb_5T`MWb!4nx<|<r!=q$_tTlRE-ms43=HN_0WZx}>$V_Z2<XnQ;`PqO
zgvJL_&I<*AQCvSNPK_|d`66&;%7dx<lKhI2;TVDo5{#{C!qs!C1nj;r^_26?n1rG{
zoVF0Rao??hwW)aag+(Te!&f93Np0*hvuX1F<1$D}KLt*Nx$88ulwLnSVkmjeL{K)&
zu@S|zCn5V<k1Y@<oil++-puz#=*cNLBAYGBP03`+p@#9_OC}2sfcHO$MYZ|4VsJ}h
z9NESu;yg9ZX4=!#NxWQlzwkRjC74Bvkbt!Bj52YNHz%ub&vZn@yJGiSoLU-gmd@87
z?*|;HkOp}_VZr5eiNIbab-SL2$A>+Dm9AywB;<0WiN4YhT(oOpvP(a3WPtI{W2HS2
zewi5y*jwX==P$XxEnPl&(54A_JwjCjUrrj^FYM*g^(;|@(YyrK3HY-MY<q?uKrExo
zJbdutcs)mFRs(phA>c@sK4Cr7;gxdM$K#Nj;Fa~fLR_U*AJ>jtJxCM*$nJEl+x)|D
z0K{Fkd)abHuUW6+tfit=d7Dai|6*>j16m6+5mG3Q$S!WxJ^w3(;<EV)M8=J!K~AmZ
zJ-KVM2@S=(k1iM-L;o=zjI^|~!lD4u#~%bbY*oyok&q~_G)rn{jUU1v8byYQ!Se%v
z%Z?J5G<5P)J!eY5p|?R1)qGkkD*qrtD;mj@SMu)8-#@Mh!dYXZw~&?wUp2iuvpStu
z|6CmqPVcnuK^=$s7|`LU<B%m>z>22zTIe0fo*_T85vX_DYs;j>)SisOL{#M`dTQ0x
z89%%~AdtEcd=>#N<HcDb0Ofnd1?_m>ZxM^CzJ~v|c|Fmh6liS@Y}597&JIc<7!36q
z!uY0aYslje)BF@<TW#j|)thBLl0rOmA`+gWJQhB(;&s_ST-fwUHc}hud(Fa?NX@qa
z#$tvt)7mH%>PKT7tp<3#4tC~%u!0L{%8NA>JG0NWZxtc2&OVEKUotVMB9Hb1P_@I;
zp7Gwe*Fnb6eQvBEkQH!r<SqKgFV=h$mT5^ga77R;&{;tc1byO#`l|&8XD2mrUNH0~
zC5lC<DZ$)2`0&5DK?Z8q2iU2;b;{H=2z)S!2auQHaLAF8tl*k=D?Z*QEJGSqZAyTe
zc7#$EBFI~2^CFI2vbiZ10w58{0NX84?!LYJCH7>o7L^%;9z+Eu@fBJ{e`^rYPcnic
zz<6~a+e<UAPq0w3PYMJ=F4@XI18`MOeUH>&=p=g~B}ZR_B8VfVNhP*en0~8lq>Sut
zI5n?3MBcRX20TD!8jX7)tIP}nPWdSShHVWDAd7?~;HH1no`<&@&EefGe`y)xt4?UR
z<^`|vDLeLa84S<Kco^bn9rNM-k9dqiKY~WSV6B&DVn6+Y5FvO<kb(X7evd%Du3Bna
ztBGVde6&tMr)$GhQLf(4a6A!AmBm^knDU7~96Rv8QR4;?{jzF8WEC&`5c~}8W);wg
zemXdYsoV$YpjR%pKsDhhYn0#yv8bGS`0mUVuX3Y0X%fA^$jrjxS!u2BXm4uVBRgP&
zxwNZ!2MynXmIkRk#j3Hk_GR5KhuhXjHbs)ki<8A2jaJCt716bh4pyZx{Wk`^2!HSh
zJ;U%5tQRXb&`Oon8FAYe;sqI28b!&}nC&?iO;V8!d3r3%e}>zrF}T~ztV=;!11$rH
zzJXvLxVJfAFtxOElW8VYgam>7-g1uK^uW@{K@YR9;%b>Bwd7tg(goVT*LphyUTafG
z7E$R?D2{>!0|P?ux~ZH}uyYacAb55yl6{f!!*IylkttP?Mks#q7RuX3s%F0uo%Dcd
zRhuOZYa{z!z6By-!cw9mMweQmC_<f?8fuc*Q`pdtZriukLDGYMSf9rmne3BU^c9g}
zQRp~fTUu0@rG{Fh;5?i{Qn1<7wJ28>Lc5g?<n8vS6^zSnZRtb51CP@!hO{_SQl?hh
z6Lrl;7%AlTFy3fnYoyEj5eRgZ4b;7|o~MH&a@<08FIw;esWYlmR>dV@hs)<;s|3<9
zjFlfktG+|Ect73wDBEQ%YP(s}mNE{)qg>dnvznst#{rNbD(Jle^nspPqbCcJ`7>N3
z%12E>ci*Q%3?x)C?aU=w1&%@mGLe8QQFUFyLWQco|4DFMRnN=+4?EDAtUh&R-eP(+
zN4Pm)kap%FGYv@9yi${Z(X-Yu@BXzPke?EOw5&~|?voS#h<D#S=exs!^7ufxuGJ=6
zoZk*@Zy$H#Z9tH8$1cG93J;gqnl7NcWu@emh%~<05PtCG#530k5R6oI&6dp?z9V>@
z$4QbTpKC(>BbYIit0otQyJ*NBv9OTtS&3V%7_vt3CBvkd7^d;D-i{j~IDkk_PDf}(
z)OX=AMjXi9yH8?LHB^+jE-s&USENl~`dFTdx&=5!;kdS%Gx-I2NT8`Y7ecuOCBpuT
ze+v?Jn~xwL5v_F#k1}(2fMWQ0csoP)_`Afidl-=_`GGmQYVQ&jlL`)A9P^RKj6D*v
zLxxvthVNe}E^seaMv<OJE4R1L{J6vU`6M^*GXInUswR|%ZQNW@`oE(*@Ms>c-(~&K
z7h~1^X_)IB^IRTW?}^pF%i*`-$f8UT5qG9GF%w{NrQpNKQRcJ|<{OLt0knWD7M1V|
zhEia$j$Z&x;#4u+PI@t)l}M)Z0Bk7w_<XU91P)B^`Qp-Z0&ZtD3Sdn;7h@K1?nvuK
z7{s@M=<gBr_3!{57F*(mDehxFXSdFCies{37Di&9dwA&A1?Nl^58SIh1x+!?Mm#B4
zFOuBrG+$_*JwLgLud=nkdD1ayJB(TXkAQ0k=?MIEQ;#@c6^%!dQ3n*-<^J-fXte-K
z{8!>=HQP4>WGo_CGadY8?r5L!%^iX{h|Sg!Nc+^l_?)QaLjRXR+NC~vk-`P)Ol{ch
zc}e0@@@`j&MV}5*{0BON2K9G`zc2aV*$H*uU<^B>mJ`*72Z9RxQkswKuY8{6tgy1V
zBqO8j8tnNR7i9em3=Rk4<j|uMm!B&!UZ1yi^_ov&lVL^Hm2h?g;Q}d!uZ+Dl01?KI
z!%NbEp~pGS#Ah|UrwO@MyG@Vf7n6cR>6=WUm)xjLFTf5;i?l2)iY)#!rx<g4Z2<MZ
zOjRpWsa!S;HtU^6j(mEWF;dBfFsdo~GXBLtPym;S>Xy+{+s11P=?h8a8&ZV&LqjXO
z_JWTLbf`f(ny9B_s%r~7ekG0LxD6L)a4M=UL))h6jk;uG2Yyf0XqlwL^<^80Hi30d
zOP={1?d8|o%Zje56s%j-J>RLA8H*7c!H5?fk3-H(910k{e)?GRz5n5AVq^IuS?i?c
z*KMGY8`@Sp;AX&smODXMaRFZYiBPb|<v$(i?d_F&A)?x>Jj<hV(?Ua`#HJ$IgSt&g
z$-zNB%KijTEcohv?ho~47xOS<4JF<^ik5>rU;)WUjwqWN*K<!0$tpaz_DaD2Y+&IQ
z5EV5Q1TkEs)tti3@7|rP+<}GEtC5+vxw5zmRT*1-gO3XoE>5>0QGkV*_n=H8Y}O0N
zk35dE1l1@ytbCvv88+l!h+qc@OKJUe7gEw}{h3`2BBhv6{BhAGI}y?r7$xE>ZRkLZ
zn1EyS%7Ks*MCKA>bbu>&w{|{oGPx2vq<_q(+&lrmw1`7$uM@s+sEgte#F)`UD;VV6
zjRJ72MPrV|&WW?ajgY<Vk85g<QN8v-52}ehgs>o6Z!%L|!zNfM(|;*gCzKKJg<@|}
zr*TEyPVbx#39Bz|enhBVAiV|sMsT&$F^BHd<mVeRX>j#0YpuFr@?v#)8}`<)08)A&
zf1H?b0o(07RQ{b)+EB3UnG1bzjh*1>=7Vw#0n;ZCN4giAWHRs>%2~2urX9Srq}j!p
zZhedY-i&Leu3ad+eUiZz%RcQIb~}rY;y$N<inmw0+E>n6-d$w9HC~6WD0f0-ddLq~
zfFRl-j1B#6@wY^%OaILO4Zs6Qnm7&bni<{6PUh$t!HblXbcay(m)62;7tGON?h(Ih
z790S56B7BAL80Iyg>n!)AmO81GpAure`cgu9xyf558)I}Bb)Sn>LWQWOzx*&OHJUx
zxh%Vhg&t%Ir|?c2jh?fWgy}1`!vPxMDn^r4cr4Pq45c3w_ngGv{kcq{fns<Sov#-(
zGYHUTM@SI&UQkd;tEk7;W7TK1j<)~f5^#%KECOXT<np0}ws?RYp2G&ZA=st47_xGS
zyFl(w7Yhx?{%4~BAwQtk^&BAVoiN8Iulw`ou9<^#`sRZMuLwRA8%@j&mu*f_pQ<IA
zdJuU*VQ0DjjtNRFiVX~+plGNOI@c!Pl<Gdyz$>zSVyvIf1`)iMW@14kGE#uzyhQx`
z)tPL_ivegbOr_igm{1i4DI<H{>vR5h>t(;rJ(=B*SL4YLKi5CX-348++CFXzBO1V1
zQq4lq+*r5@4s%gu*M%`=+r~SD2A1|4i=TUe;LEF#vVzou+;-Gl6MH;~GMTDg=(13x
zaQJ1Hw0-RdY({dQl@Q2))MVXS9TX3h0s5ut+!Pj!%%6e5p)@`1ssv%xvlBzDyqy~)
z`isO!t?No__p<}y<it-D1JYl@-Ii_g`jZ<EEDAEF^~&BRtnbZ}*AR@&g!>&Isq#bM
z@Xw>-w*>(u*7yq{#&NmtVcn8X^uU9s<J`SU<bZN%n~n53P;D#hDg(}n7nG%2cI4@l
zF<a4+1wr8DAgwYa;ZD4G;3J|zur#q2vMOMZ*2zI?{40SHx1Mx^njrX^%K1wYITAtS
z;8~AZ4;bsaE>jO#{h2|nrD#d>>iKwh_(HWY`-1{H47(My4m<kzO?B}!u{R9P6Jb>*
zP|<6Auh{%IS0S60zT(58O(JsBvq2n_?{Z8us8X&$NffYR0wo~4bf!02z|BeJHYE4v
z<H<@qjDhBSq5eijZz$5aU|3JA5z>q*U`gmIR|^F@wErsXsaY?rpk#Svz?n%c$nAO}
zsP!QPOfB{lZO&COg#?8Zlv-xB3yP^T4?_$yHVCP@A&~TnCAA^8Uyr)t{?=tqRoBl#
zMep2E=zRI@ek{!v>AQN{0CLyfIO4PFPr)BgN)Z%+MhE6o?}9eFd+ibT^UELK3A@~;
zn6U9c$hHRVf5p-gg|C#y?|o0gE{1*rWWlhJYksy4(bi1CX82OxZ@qXX@xDQiJVD|;
z^S2G0{p&@aOFU>2X{YnM_<}aD{2*OMt<vlin&ScZCnD#{c?tQV<*YgO3>!4Fp>8kt
z&n;7LDZe}cW2|#4CN{!flx&*ti8&@QjVbRJ7``zFN6E=)CYZ!X0`WzS2hrYD=9R7i
zO=(hGR{G_)&KA9=qJ^5z68Z-`O%DGjtBc1LrdPbf5Y3>BoWEXhKwii<PdXmsIQ1Dc
zysbDR+HSTdFZAamLhd@<Tf6Xr9mUUj4IS2R)02$pbFp6sBY7qDiCq{FXKMK_&m{UA
zeQ7WHX|+;{yN#XsZ30wav2&0$0k8*0+*kig)O~cZQ;B)hj90Pr-v2{0TZX5|lEr2u
zC@_MsD7rG~*)iG%DK=YIh*W*6b5P}(x}TlMw)u2N0l?h$z2^;ID)#2#w5)b3)wdeD
za9dY|90#8YUiXFXB*eZcU>q89$%_AvtuK#<djI|(gsws=xh*Jl3l(n3zI9VlS+a*g
zi!%1CS;mmt&CP8k%6?@DS;}rKBXuRANF&>*EHSodOpGzV^Lh{WbH9(r@BPQUkDIQU
z_xtsJo!2?f^L#!}x)zk@NS-G!@FkPCkINz7xT~D2rHDp)Dr-{3ehXvLqyrUT)4fOS
zhILy3V|bC9rj2dCguGH9-ly3FuL_TfqMKNq%!T#2wNCoG*RicN2=MH6)Luqtfu{9K
zJVHA6Q4|hrUx!QBQxm4@p>zgh4}HW{b1*Pv95J{1Ce6WK4ZBM0?F9O(0h06oeHO%T
zp-DtgeqkMJ?$5i3SlirSrMp#U-`7S~5#@6!vRLg%^(V4HT3WUrqnjdu*!&mHu)jzd
z0S#{OWWGFvO~p}Dg)E+Oo#1i#8LvF9e+;n+J*GM&_aRy~>0OcHf5lDu&|qnhM$DMA
zBreoqG$5deQwB&BRnF-Xk?|4NA&F{)eU=GQ7<7fRxkHMe?#Y0pqXB{B^FMR(Auuu#
zg)a{z6VNHckU<)Wk$lcDEeNol0v+=H)@xBc)N-!S@CC4azDaY2=NPRt37fhS(aB|Y
zr3Tns{Ar_Ih=m|MU#Q2NbbB}McBJ;4`Az#Pa?W`XP2B9Uikz043&|Ad$C|jCq^Q}x
zoLA5HiIunvyj{M)NKWkhvu1Pk+Fk<;1jQo7T5h+_$s~jIgJt#u_9VT?i~lu&HiL%O
zcEE{-u5pNBiT8zeFW2>Qe<ThY^G|z@q8k}w&_KIjj`Yf%#XVH=a(A|e*Rzl`{SB+-
zS9H9-KuCwr8`xoMorE}`Vl8`%jtkD=y8ikX(tWo7WZT<Smo9OrhUn;iw&5krk)FN|
zk}2nK+C{zXxbfPoS_iviXsqI?M<I9g*N~*|S^ky}F_P!mnD+Tvm`3Nh9fxx(hMO6+
z@RK*=wg$p!>3TuM@Hh%nk1g|TkMm+<;vG4k%6>@d=m+<qAvtJQnANhF&GhY|4=|EP
z+fsQ+U5kz?eAMJ4|Cdch<n2IfjdE#wzKYTsR2A%8p;Jh&>R-&({7Ta*uH$YEXkGv{
z1zw&1t;*nif&R<7wT`$VGU5eKo;S~^_x6x3SsMzrKv*^m2O~x|LpS9_o*N2VTIs-)
zw86^-J{qU>XXg%b7w*n=Kp<Cue#&GH0*>-hZc%~HexxTuW)c>Nnkr}Zhd)x!->m2p
zrTSq=QY)eW*5#uc+4}fxiLH^~Sa??DB+%SFre>EnM<*H%oVpMw?F&;!bP-FJuh*ob
z;5?60sxrf0BVV)JdFQqw5wg{Tr2iA5$dDGJ(-ZLCIHLp8KteF;hO-expVd>v+z?!m
z%59he%j1)AYobdTH2H*f^7w$&Mlc)P42l#t$~82e965m{0&)KTPSN1@jCwm001a?@
zSF&b+OYBX%pVA8Ih8K(=Je4u5+jp3A@y~he?BuquA<~{uK#<-UCsp^6&T{Q`X7j*$
zm>n?62r8|5t7Ww(yv~B^e;4|B1npn6xgaWKiNLCS=GPB|2^E5)P;AGxo5*>>JB*v@
zgwsAiRr#xJpFzlC2yBo8++MZ#3pgYLw>U{>UhV!Kx=Ot|?>K;1Ed`4;ztduekCi!x
zT}UYWm+M><Vr!nH1+F5M2etGyyaUaP=d<~Oa8DXnKzJua|7Yd>0Yg?e@4^432N}Km
z_D;0-z?Clz$`e(aaNSQv;SlWGJn-S%)2<vo_Z9vudVyZ$O}uQ%>Dn(`3z*)&%K+As
z=!XFLA-vbZMgvg|I6yThGC_4oL4i*;0Gc;qp;wRe16(3nXKx=J<OcPg2>`7|&AqvM
zYE~;zgSr7iV2exWWgm@K{ut4!`AD^TSNi@n>nI|cbva7=yyWL!$63v4<M&56q<UpP
z`<6zUUArV>xq&AU3^P<K5B~4q(A&zrA=!LYRsYEQoj(oq#Wg$cfo|*ytw2vcz%3-=
z85<xb_m0|)vWj44@WPeij`if=j>dH$-vR~+W1+ZnF&&BXdFJXd_HVj#V`g18x)mgO
z#XFFK4l=zGG$qIXD4CjAj<wjd%pSHU`bFjcNbJe3|0(Bfy4*$^oSm=v_E7dX=wQ@7
zkM+P?P)Dq;4RVkEU*LkeooNlmLU3r4n7@_-C*=r$Q3nPD1{)1&b)95Y-1h4h4|+=&
z!wB_pnQ~HflH`&mfOKd-kCp>+Z2jJTrdQqy2+J;p&KRNp*cK&j9f7O+MO-!&Gi9k>
z_~8_*Y02V(763CR#t2<P_!U~bi0co5UW$Abz_^V4P4_75YdkSdQygx#%>x4$&Rkcf
z`a)qm83texCCn8TutQ)EAVG&;OdQq>HzH3K$dd}_<%;|c@v3bZsqW?8S0+;WLz72(
za<*v6dZsivVgp<irRI)enyht~?Ur46D4v(UT-HtqNM3{UxOXEaF$>u^2slv@4+D&=
zK*WKugwg#1Z`AxJ4{a}yHf<7{d&rBs(+G>2ZT=a+EGYKiq69!b_ab_woh%2@YjHs|
zkLU*if)1R==b&pUC^pbX#~3r6(vZTVo~;<tc`FBMyxQ9mKMwH}HRzJ}BE3Fl0rM*R
z>B}a9DOOxf?h}0%ca1;oa27#sx*VSkUQ`bsxfE#HDTG-)SZf!1gBXxGum|NB&fyF*
zXkPWi3;YhWtdZ84btp&ukQ_lny+_0aLs-m8?|j7~LnVhW0umILc#AYGD9QtMJrk8}
zX2P<YxpA;75JTyn=mnXZ?YJO+)xuBcNi%w1L6*h)VK2Kz2QK6uw6cPO{)lG<f+tVH
z<fUNX@i^5fm#J0eM6OTJp*-hNcPXSNJ|CoW0FFv1Is9qVl`y*ys@Zu=@TDu^#{l(z
zJ#{GWFw+c>!pJt07F7Z_rM0v(Cdc|V<BEih@GzE?r`mD(j3t{$4sb(acsZJQ(Yslt
zsS+sA0f&(<!4rWxgOW>W7w)z0Oor6H7g;HxD=Zq$rr=^d<Yjk0K~wzgg%kr&j8>^*
z#HIm83_EVF9Kg3SOxHhh3jUWlP`B(y`lYBj$t1olP76i9)Y0o(4vwd7ZY#lhWc$OO
zOQ^Pey>`j3tC~wQ>Bab-V2*pSdMjV}MO226ZyVjU`lY?<AcYK=P_m7+;f?>aYWkD`
zCj<l@rUVwdUqxv3_`Na-St_DC!x@Ktckr^}YM2_KKTVJZ8c`Kel<T%6^Ol`Ql8BU}
z%X=$-F4GYY`iLH6fQkwf<m%LbY|r%3-UMwPfNACbi0{c>^C3(8Ovbg}0||7!HQVdx
z3k`7#1)SLyvrmFc%&km>pve@pb-e3aCU}<aKl%LaDoaaZj3x%425%O_tWE+qb$MWv
znj?u2E+udYk8}8f#CP+xSZ;g?g&_9*aYa355k5T`9ddJtAmKmR4Z``>J!q`!N~Baa
zfW%x<3ma|Rs5@-jp@YeDp32_{VSveu-Ubi_ckRI%YB$l4k06F?sq`>I+9&(!1a<=S
z5WS4zTjRTEkXjW<Z-F@lB{I=|N|Qvsy%4S4n0K!|&+(o&;eV|;99%SH{kw$fef^_7
zx3n4*652>(G~%0ljGGPY_cTfRyArx1&gEW-D>_6G_Y%FRo5Qn~lWlmmt^c;#`T~4|
z|1KTB4NfHznTPK|jgSC*LD(Q5_mx{&J}!+0nz2aaIzHwKRtL7wXVE&2X-;L3(;qS)
z3$2GfTYUi6CuMHOA$-D`?)UR)@C<uRzCi2>SVt#bYr<_k$bpgNgC_hhO#(0E6xr9T
zw~y}$I;w^daV>wrNE4t-U5EK1{7cj7x+JiQm$ESNO1?2Nv;LLG$20-=h63?D5=Wo@
z;qS&>CzoQ3a{jrjjL<!=2Rq<og6O*syh8{O0f_>drMUbTZ~doeMS5!oQXL4Xb%1T=
z636XPiOc@4<(zoGH5s))p=Ya%IIVf9+}LMR2d|=PzEZBo%O(*@{Zds`g}RXDE7qw{
zwjn&4O}d|=6Hlc72Dj9YgyoaxJ)O^#jm3$+wZQJ>P^nra7S7Y;;+h>bAvdJ^>6|&H
zk2MkjY($sd!QP~cF3HN2EUrQg#Gc>7sy%?m&YHHsdK2)MM7cgyj(sK}(Arm?ILM_h
zwn$fp#er%}xfv6K@8rhwyr4)-m6LdXuCjD!(Qxd@30BYdov^b?PM(KVvU&id7qR8B
z98b)*oZ=a;rO$+khX_6b?hKK_Vd~QMi5r(M{^dQ(=w;FuYy?%0HVM|1O&mWs{tP-5
z5DPz-Q7&->N|fE`7z8ySyp(5gUCu}kvNa9n0NDbAs*c+^#s4?Sa*4~D2n>BZkt701
z)@{s^c4G6shZ(acTK99q{PYe?d2}1SkAMyicII_z{cxMrzzrN;W~(lkMI2!5nD7m;
zp}(3uv0oO>xg@M-MDKX;c`!@(Hy4(ZP}<R4OGaybG)wYhO@VdiZ&XYR=bo=OzYqhN
z{{Aq>cIFpgrgUZepU~J+k{rknx_%LrAxj-`%`N)8IQXwUs^B?*fTNBT;uOeNS=)N7
z-o`0URBwkKo(HO=WvsHF^`IdqGM;UU+`FM1f#8T3{_bx|Sbc%z)881JFkfBBro?f|
zF}%)mhc!s|<|+?4PtLJ2=KnTI-YEnJd8*r+{ERGdeeDvpN}&1qUJ|g)s-=l_M!p-{
z{B;1zfrq_ei#jBE)7lF4^&xF-Os65&jjjHb$#>PbtChR!>O>BhKJ{*1;U9CQez>0L
zxAd}J=n996mP5b@ltu`NOgfS8L%0WQA<a_f(vH!eb5QFOF6oSXmcV%QDqPy1NQJqx
z`=GQ;vIxW}?vJ@k_3d^Z%jJ3_peT(<?}yjtf?dE^FFJ{tJ;8{~mU}{2hRleENl+Iv
zh#q+#?#`jJENZxA1AyoQZ4;1HZ3!>>Tcfq~<k3x{+F#3ZM_1WALT&qy#oEO5fxNn~
znZc%*s7NoQyp%Rg)57l;b0GDAvCq>v-l1is12F_a4COQlr<mn}3|#9h%YDKF4X1n(
zRE|$$GBYsk$OS9`9CeOh5H?P^6V(He7yi$yng#0?RiRz}VbkdMKkE+x(bZ6cm;`h!
zpDpUic0OU;L^E6zaXEjzp{A<~AUzZ}bQLI3tlKtf^OmFAAE#Us<(50lsuSeC5|72v
zr=#ujG5?bKTH4WDdpuIMyY)AmGKIM2FPiM?C2&n^8hbTauf67XszaK6fdfxL^F;Of
zjKbLg>-+je+W)_^4kmDs6aCm9VtIZ+5eVpcZCP-bGWU3VbP*Udr810fo|;;bv7#kR
z$j=}7z+GQdh+0L`Kz_d`avA&k0nR$IfSPQpewCaH+T`;P^g5xe!RZytnO>AeAyZ#}
zS=&Ck*t9aviP8XTiE+14T!>q<#QgORZ-Arxfw0uPOUSp!Yo4Kq%tK{FYR^c1Bzi~q
zEO7b{HaaenbTIXq>iA0|@psyF#slZyV0`3Zm1N^opm0sReIN0&(bipECG-opSA^!g
zHExfTHvhQ7MKhO)0?UW`TM>^wlYt~q6B)$~pG2HqM9FErp2?-iVAgbc%bc&yM!`wF
z{1#;9xX|FL@Tt=h4s{WfZBBdnW`_rurkzU6{BX0RE`R_2Z3y6Hia{lJvFjtyWvA0n
z<c?z<m-u0{x0m-ys>Ks{MW*Nrp$r(+;(@5FF9uS?NkfpS8rajcJe%v&^LPI4B!Q%m
zfOcSA$w27{#Jlt7I^}|cfj~IRZH%5}DKyZqQVR4;sLY&z&qY^&32HTZ{Ay?C&P4VY
z?#H@0J#j<|LAs1d=l+$@zmQy0?biQZ%Qzt^Zl@Ot?nHnbKsh$K5%J3i#eO2r+)5`K
z(cWDq?SYmPPAo14$YXDbjrCcd4k2>5q;d_dz4tJq=u(nJTIdy&qBA*D1$0oN{#CIc
zim5)RiXH1eYld+UJl(?zkgTpyzuIfE^Y`!;!<X7*vjKCAoP{Gh;i<koX26pE^v6En
zR+Q8xIjAPe2%jiE!4q16(X;<g?URP@HhU45!Im`Y8G*R!6=Qt1+}RUA_5)?uoNzcw
z<}OAYpmSXgmkmIyq#OB$kMXtKPqB_*n4=|}mGw}Mf986)vTr0#2U0O7)d}~@Gq5n=
zNVU#Gy+WWwujb<~m3h&f?X*uEC?0~hTsDQqa4Va`viK{MILnzg@AqN6A>lol;sXe6
zsH|xdWaVS5e{M?<Oew|_1Mzi-^7Yikf8)iG2#|jGT>7!jC`6ZcJiV$QFgLiSXTkwZ
zlfkZu_77}+BTW-n#X4tyA<fTcvx@>v0O$Shd^Bh%Wu~`5n8DPM6W~4ofNnZi?l>Y*
z)ZAg9GQN<7Ko0!qN#(c3Ft8dS^_(4nTWksuBxh5ji`qrdLza1taJg`Nxi4VV0~S?9
z?l*{_aU>a2L^SF~Ok9A#cw6@E%5n<QFi!Tizog1ag4lkRXZ@Dz`UMWTkM|44!29uy
zCODw89Bjdv5!%aqMbQZXcr=E3$;?)q02V(bx$Py)ofHMGQyYR)xB53CaSdD$8XU;#
z;_tS!=J9|B=_oNo9_UY<In#+e%kT2^T@WiyPrE^EMb)phgpkzPRE&wyT6$BD4C!pm
zl6xs>cnsR{PTK8DW$>iHgB*0Co}*}|_C51G`Os)@7@}K8&Dp=A#UI{SL@pR%F8;Lt
z!|WS>fAi2q@L?D`E3QjbJqlA#9jmW=I~PLBjvNYzPB>e)vQ@e5<Ep-FiOOO<a>>xb
zc0BFS(S=&o35_i{6^LE(fq#g;D?9v;o-q^y*^@Tn|KW`3X=;yu|7Q}_T!@n5)Ys0M
z%Sv@MzSd;aSGZijd;W*qrp;|F3B{h321k>L%>eczfpQie^J(GI9?oF<6&t{b<-QfA
z=tj4C7ftePU|CVt4*C{lK`uU4m<n>DU7xG$wPU2KyINkS>JE(|g|9K6hbk8;vjZ|s
z6@!=jXg|CceWyEwViN--g>i{@1TLd)9C*;K8)Y|o7Np}^N^GiO6aW;&V4<`;(HoU*
zCA6pevv;=ZC6tIlsolLxIL@9Kdc))7%O($X6E37o@q7S>hCyoK@U+%~Mqt7mn3o2#
zA6|u07YkMn6MJ-!5yH42|Lc;|3D-ZyRX!NCe+|8~K2#`&SHr~c41WpNQfE{CJMWZl
z-UzyV8qOb2g?|_wNW-V5h5Yb)X|Z<P9+I5XGM#m$^N;!eo(WGf`}Gq`*#l?TAtb}&
zw4t4TwnJrCsq^`ti`E2WDyjdMq)iF4Va}KJCIr+1!B0vixvF+bJsW?2rDx%kyppU^
zJ-AyA*)kTi@I_c*-2P-Y$%p`8J3BZMfcSUd&!{f<=%%@h4{`r8gCymRJ#F8aSDAV(
zL?{E6RE2biQ4J%A^(`i7nIiu@QbN^IpD4~K3PMB#^6g4!taIDGy2DE;IPyTPs(m7;
zi-cqn1MtE)`uq10l{aD{7&-Ld><`+S$;OXz$GOR9T4gu<J%J9WJGN$;w_^2o=wIl`
zMLnihaixLqnCF0wg{_jH6ig`X2>owiJ9aVb6Sy}zC8Hhr&brc^5;g|IR;_74#7s#u
z4syxrXIG52hn1sqNl!}C9BSPN0gtV+-%Le{>O~T1vwQsfXk!#cMQ{Kb+i)2_oO3|K
zMuollS7oTGu*wipIIQ0%P?4$}o<X1zjrRYwD*X0ctHTI#MTRlob#~TF;eK3!Iy_q>
z7`oa(QiCk=ONKu)Yr+i`wsR&LlF$0gWfITp?t=V$dvY*pqxYWCIfgZ;u*d_#dMt7x
z)1$?2ATTe41{;IX0UeYyZAgedzXVPSv<E3+h>%Oq;dSo*CD{Z$9YDzlsI&e?44&*O
z=Sxsnsk}`i@Y~X1+Ht4o^IKIqb&i*3IvsU`W(Ex>OA-U8Ig8euF-LY@)X%*ZvPS!l
zJ|A9AZtmT<*Zgczg~?nscf`&SpQv&L8A;067xEN2CP2dg!w%Q#^>D=x_!}Ujc1bA#
zK|W9?&K<&SPuIS71a3!<RXS8{XUEJ0f6^Y5@4RRkU<#ZVc{mwZg&bkwk{WK|9fD&b
z5)4HS(J<)DPQ0;Z>ByuqMFZ2>D53@B_DBi~OHL*vevF%cNbqF-y9Fm+IV2^C=pAxS
zJG|xhGaHXS{gXp77cJq@Osl1}cOQQ0*rrV7q;kILE;VwZDuNMyC1%}*4GSpny78#6
zZUWwxD0(7E+Ih>hwOXJN`M)c!lAo{oBSAy2D2M+ZZhA+1T7K4X`fDsNTJu%~-o!m0
zh|wUZTulddzVhL2$)fH<-XTR?&YfU6++;ZWT^VC-wkFVT1PP?eRVE)TjC!ct0IOI0
zFH~hME&Z2Iwrefc^~oQWZ~$P6@5vF!tEz1#_J*NCyC`)f1QX)X=B=1;bFF%R+rPsY
z^-1H99+JSfRt4_@NiK1F<Y>$Kp7N=A;-l}UI!Z1LFshc!-}x=*JdrA_gACzDRAM4S
zaFk`qp4&@F4Wk{2&hc96EuihIH72EC<m8_VmN<eOs?oJCXUy}q5wcNT+VVwr=*`_5
zK`{J<qXBP7*HO{X`%4{DCWTc-a=VPDJiAqJVz|i@TW@+6By4>&%GVN(kNa@RXYUOe
z?SmW;Hv`RPq+R2AOF_7<@T4)B2N7drq{4JjKtQT?+81&L`%a=x8xVEK+b0ehGk;Ai
zGtn3_cD&9iK<^ABm&m_LYMTK{1$38)n0|K4QOxv2Ik0LNz=%gw#{wa1(J&H=D3Fl1
zuX*MjsZ#D3Q|qTQ*wH*b><@rg8wV+pe<%@jyC6s$d(wa1;E98y&n4E+1e%?z*RO3L
znRhCF2Azbf(OF6^JWh4?*}Pwv)#vb>>!ORRrsS&>XjiJ$KZk1D+*Hues>AfQn0RYy
zJ%|=sT68b&H&$r&@tM7?JIKC;Baa?&Jk;UuWG=8eD{1tfr-ygqN=|n>z2E31vRl9=
zm1`<auIih7*`f<O0G`x1xx_f-#=Va7_eAz(Mq*f!$GbSF&5|`&fd?x(yP~k%63N8-
zQjMrP1del|-Xg=Cu0X`CV^>tw&5c=3TD063vP3N*Xkw4Sne^xBNk7VUOpKJrVqPaz
z$a%L&3eCO8Xd*rg-=Jw$5A)Zz`Z-Q&6Ud7s|4KQU77$ohi^3=+SaMk8Z3)lX8!^^v
zQ&1v4Vlw==@1%zyX||BVs9OTjr2AoY)EPTM;X+XgD>fvj&K@Y@A($F;2t_3Z)Cl7M
zsi9UX->D!>?Fruv$&_Yke+bIda0d(p?_pKh(laG>@n80yLT7e&?ta$Bqy$(Q=}hLV
zQL-BnnXcgWMY_P58JPC=cbG=fePtR}{y!0#b|5~4;$MVIb+_ah{5#$1)j2@63$w*g
zl~CxAuOkv8;!>epsHv^ehw$ItO^34)`fF3@4n4U3VmXP>(eTZCjpltLi9UE%(_c{)
zOCy{m8iz%?t+t36dVHQnIEenXr2Ld7(HmK}mnu&mV!S#yabU~Mlo5azf92MU<B?f-
z36*@)^*FGAZg=gLKW?Yu$3u1jABXo5zo*M${NOIt9KDg2^udy_I{`moDQ|@_Vy?{@
z*nK#RX$o&!Wj{*X(;}8b@Amu+?F#ULLJD-zI)}bXA&cMv+!6>ay#Rl%78z?0a_ya<
zz0AYPN95WZs&)4-6TZl*3zI>SQ2QLq+qz}#P)})-JN-8@QV>yvl;3Md$A|{AzMb-A
zymZ|))7jkF{+Z!qDfvJsOElFH$oTr(q~P=-yr_@vJf0T0<Vhl#wRU|=Dp@uZM(_8f
zTU<d2E)fClsB2m^sp7IKT!qkG<x>^x)qG6BNutv~*`GLU)Hm6IouMw|lF2^6R05@G
z_M#@#FL`}DEp{X5Vf?T`M^9R1JG(WjwlEdCW-HX{E?rC_P&iqvjncYdtTX-5<L-gC
z_E^iNTHZ7a*}4?bxB(Tiq3dW<*wobaqGVx7l{Ek#@PU50zyz`d{+=yKIRYDgIE|?8
z3{R~cVaoO1jEdr72V&pGZ;~#X*9QMk1U6adX0Z;0=d!8Pp%91q{vT|Tk9k)p8;0sh
zVPyAX+qPkIbEc#~O;$VKIHK+w>pVY!c6|TT>LO`G`uaR4Fh15vWTHIx=o1wmU^L>S
z<!dIhsHIshV`&G{3_<ezC!>EwFyWF{qF`!i>CgOzZ^wN@6t<|jZBI}jfUy$6yON{b
zx`S#5hf4s67|-;}RhxL^*&VbyI~|2wNEXhbIFWhot)kZ&p?M{)IZx+&O9LmkSoa&K
zxhU7UUY^!0Uuoc4ZI|q-{l3r*>vx?*Iw)~AR#QCmf%3@C2?-P?e)K>V@^6^9uKtGc
zXW{{u9l}xGJVGVM&hDBf+OruLH`Shqkh1dI)YK=(l<H=`W!b1}ODz^6_ZI1tFVkSe
zZM;(?MQzI)1*BRN2ihJq>fq?PvT+9ddU%JjTx{!i;i_9O3xw$%33IhPF6mqvhM3q2
zY}UOyUl94z8&uaR1;OsvDTAFi8+W6OY?l$O4jB}n@S&$LT6?KVx_cvZAI18Cx{R(n
z*7ByvqrD;dLEF1i|IIpG8JCeLR_7ns@gP7gmHvcSTa>WiS%!G`_6c9N?y&MeYzS$O
z%vWqilk`P80x#UTEx3R)=0Zhhr8Cm|nU%qD$ApRvuY=p@P;<N(HOD(>q3BUfj*_}V
zi8CQS0OK?t(SHH8b{rJ4Z3!xRN;PYKg<Z4T4R%d%ohfp|T=UjyS-A((QCoq24|p)W
zD~~*0E_l6NZz9kzu+wYu8M>|MKT5*x2~olAQ*^TuSOx01jS;?NP4J-|1ysgl49w&>
zM9ef|X;(BFh8o7c-b)DgKh8fZ6uTo8I^1=^o1jte&j44vA}M4V`!tvXR+kzXK?wZb
zV+$x3F2NPMV^`PaYs$Euzm*;aB>ZO~|26)vdBfJ`nL|voqsoP=-+}riK^AYR>Va3U
zal~~cRTUH$bX4aoBNTe(nURMPWG9vvNo3AXap%24j>X~*HPAWMF$Z#PibkgIDWjh3
zP#Oy?57AX*2<Xs;M=)|=CEcXLi&L}YP`|U}PhRI&UZ_}NpDp!T>ef=+0^ZktGM*z5
z86anc)OQlteK63|PSM!2yFSaFZ^uRzK4yRe)t|jY3s9PQ^UP2vhER@4oe>3D9p<qQ
zSak@juKHU@Zv(elonA*cV%5>V2WPa*Q~@g8)_fjZMRo~k97y7_=lqqHbU1y9f{%Lf
zXeq8d$STZMp0Jle`A$baHM`>jQ&V^up#VNtEb`@&)^#9;DuK?ta--I7EZ7uq0tKPi
zUzCdMBN-ei1VH`Tk}RNO>}mjYLcXg626FeXFT-c_wtct}gZx|5)g?Sfc{PN4d&}%o
zC0x?xwQt|plyXMp(=>9XJNkm5`{bC$-cwol{}JvWZXrG^*GwF$KCg~gdbPis?T&L*
z3SPE6zd#291DdWxeP5!;uPq7!O$Q8Eff>K>xdyhW#)4qY6sG%mDe`HD{~R`gZ&n}z
zy`bYa{2UatRc@vOz+KT-h?%`MAK_B^;Ky*?!JI(9jrD&1LfdP)WWznf<*xj$0^%;M
z#UO^(JegKFCx>uip}7Wb69CKn$QT}zbf@gi8be<T^o%g$%EhvLXsJwWH;1=ysP*kC
z1<F~$_sm5@i==8RlKiQqfJ5!`3n?TKK~<8-Xr1+gH=cQ*G?<+Y#0uZnSh2(lo?`V8
zMq8C(aBKMP<Av5ncBE60uW%}b1@z>K^-#m3)9T9g5=V@Q0_#1sgG02vZUT7A{_iLj
zcO5d)ewl_+AJ0P*L^gcwe?d7fz+o^xWn(WPArBMA<f=_}Lh}O3YY4@HZ97q53u8K3
zUwe&A0EOm`V9ip`xp&a_7WjZUaCINl-8B&nhy0nIZSZ*s=iDG9X}$vWTUY}4NO<^Z
zLmqdoj&MJWJbHdjjA(7;=20!z=6m}llrh;tdI3rO^+Rx*Tmr&K+S;PT1gkq;?`2Lv
z<1@I%^+w6qIsE3XYRuMg$v-heR_IX^yy160BqQGtnREWu>`&KJkn0w3vi44d*eCnA
zh>UbWnQYZ@BL+B#|AIGH;!%f6%U)bo*D7tX9u?UI1u{r~>ri9%eYHnB#@!Nk&;jkg
z$YNk!9dT)u$|dM?p+t}WiGfBV5>WtOqY7aOHsX+f!458JMylP(tI=ilX?!Kuo?n;)
zRppnY)H=*lR2DJJXlufmZry2jclDX_hvV-Rlp!2LK#0%1pKBdou;Q~-nMwbF*pgAA
zj-;?+rH@N;a8qD5N>?!In@aPB>duo!9Z8A4?1&DOa@I@kVe2FTD!fmhsEBHSmMN<e
zT8LcssLDAEulY2L@C*$nNd$-wtXQdu#luAcpZJk?296`ff)fX@mX4>R<KM#Vp?6{1
zHzhkKBg5nLAhtbxm7Zs%P$#V2)pb8Y6p<wk;)E}!+`6$;7MUw&;20PUH{bj}<E={|
z$K4|)25uO)D15?q3TO#)uC$WhU}2A#JEmg8v;>?S$89hB{5yojuF6dZDlIWAfes|G
z|K^^lgHQL4e3i=0VV#5&eM3S(aJ8JT0em00y1cOb+m}C0+Krog1MpF(8~qi7Yw~Xl
z7Qkd=z7IK&Z^B$Iz(F2K??kqPGlffSJ?Dljj?<8pj^-iVpAyb<AJxTz24E3`iFRYL
z$E=oafXEs+$n89;)hY>s@d;Lb_j}l5wh=e0F5OiW^~L%FLNOmzk*jgo(tlhnr}*C<
zOH{CR=@$NgayQm$Xa_9EF27EsE(XnwB2u?myz&P&jUF$qOOv2IkMYUA6n^Qzd7+?&
zTm%^pvnE2o?c5k4TC-r-a2{o0aH<-b@jOm%V?eSIofi#<5IRDkS1!&SSMw+ed)=$|
zXVg_xR9snXD}VVK9EL~gSnP=}D8pD<EOSxUp5q$UIEoKia`E<M0xuu^5QKKIiP7j2
z+5ojf{0J9++2rNYd~LeyXC)p@r?57mS%;2e9|!LW#S|)_sK;}yNtM4zuuTzy28b4f
z5B_(5Yb19U@w&gwa!dx>uI)_fj~xkCra$RvW_I4p5`Mk~5yv+}Gsb3Lz~}8SecTd=
zRYj_pJA0YZd<uDEzZ~9#Tdj2o8})GR7(=?|4C{YswA#Yl_%3OjeGz*qxL2RJIF~hZ
zNW@`q1=zzQDg?Px>jgFps!UFNFB5LOezvC(cqI*~UeR$K=Yyqc;abY8tJ-gC_2i8X
zfF{B0k1aY$U@Aok&Zyj(7k6jw%??8%p5O^D1*Nv<3e+<q3V^sdK#!|M{ei$>ATg}&
zQg<He{5{*0K3S%^CT5Bb(U{izm)~TFK=z-xxv#4(^ux#uur5h^*e=|eCA>~qZCyBC
z39C6n(W-${d6z0T^Mb~9y32b>Jp{{6{D$7@sLuLI|3y;ALXgH|fS|$s!=<~nF(Zd%
zgf>Tjv++MHKrQ~I{Y;zY49s#(AD{s!`oQsx7jnh4c09&43PZ%6jP#*E81y<qw|LTv
zwnJm3c|&j})x;&nlHNEruMH!YPWalr;eixluJ4~{EUGdkf~3g->rPGVE>aLYnmhf)
z<Ac?Man`Awn9>iBR>x<X@-X)Q6hCEQyqht^%81x0u$uRdd56k$Mnt_8U-(2*P1tOf
zXE`16Gd+3+eHIgTF(z1M;3E&AwWodQdX_k%bYhNKU6+g=ntWmd=7lb!5Aan5!4ER|
zi8b^e>lP+2>ZEVTp%@_Z`i6#{5js~+ybUSb=B;634Rh&7V-HQxNEr)rR<pks!juz%
z4||in2;J)OOhXT3Ylq1K2aBF&R`@x|i7QQ1S+ai-?TkI=?vgo^q8qWZ&D#sQu`)7f
z!pq`fWIe}c3(!npA{{G)r$`6PIM%&OirD!lT@&YX{2HW-6&+YniemxyCAsQ3=5`@K
zaI~&+3j-Cf50k1J_xu9fPWVMbmUJT6t3n?Q{;M~^C!tzyb~(xolylij|3-Lih#WJ)
zRT@(Z0K+KjZN(^&JtN<c$#&!Il?Z2Y?X@zP8er3Wc)9aG5{h*viC|KqOanfG6V2Vd
zDW7od<P8PfTtj^!oF5R&V3_c1p^VJfLj0p7bzoZq{_cQtH-^+xrL<yw_(ivF4H#ht
zgLE85B2-`yHVm;@p<ft}Y<RSW2e-uyv&u)Q*g5vg1Kh(oK`yz_t()*?3FkBqfMBdA
zgvcE=$N>+oG-3Fh!FMA<DpU`QOP4<<h(VR(k*I=x$69kggxkt3nI6H&`US(sz}ZYw
z_=&X!?cQ)a@Kc(}8I_8nZYSYPbg8}BcY<L1uDw8VmHFV>YO3;nLQSI!HRaO`xUX($
z;WC@}mdM4H;`g3k=*|S8L~EPwHkcarxy(f{E!pf>{f-QttB~Lm-gZ_+KU6U~d*Du+
zPi60){!MULr}`71p=VYl0vnsz>+cP=&FCcV{+p0(+OD&aC5<dUjWpTSlPr*(qU(;i
z{QGfK^nC$=5%-|k5h&!8T65-cDA%2L<uJxf-G0m>w`>5gJD$NPp<D+JesW{c=Fy~P
zh})p-HYSLi>KJ|o$3W=go8o(**A_IXIH}Xl@7W;&T)ZrC=ObI~sf3a>bwltCO{B{P
z<Cn;h90npaZ`|Fn{(lR+Ewv5?gaM84;i;uxnM_`%S;UZ7rs`k!ezY;%1Hb{>mrZ(z
z47f++nE0|AiYV6?5!2(_?J8DSegc_<mrc;$zy^_W?;JM{T!A1=W&UmKOXO()4Z&xN
z$o4xV{S9?8ah}+_i|Bn|)OGZA9gEdNjX+0Nug#jZ)yY(&xx)!w_oRAcTCSaekcA;h
zDujuX7jyK=<&IVv?JerIG@76NgXmY3eXUL!{~**o<Ddr&K~)<W0XJo?Oko`2$i6H(
z8RJp&htIj%j-6$>!#=|q>L0cMPR{Dy_w^LLt@}192BfQ1F8csbz{P41i%Biw|DYj-
z@M>jeP2NTZx62oF+3sGHZ@r++O(%V_kv-XiBnLqSMMXu<*pxMY^X;%(UjYk-TaPV)
z0bt9!(G92wG6^9|Lq-JSH^{jj<Aru#oPg|k>prafZt(ipR%JfmkcD`dw`r=7H21M(
zPyJEO^{d3Niyw<H+jw8ImoG^XSB5BD6AJ}|YjHT`Bj$qdv#7q|6Fv9A^me(@1=AAx
z1eRe&g-ps{QR1c;6Cb{t%g*U?vkJJcJE&rj)N)L22t{~Siz+uQ9Jbbmq<>%uYZ>JZ
z%)T4RlNqA1*<)WCQwx{aVB;W1NW`pS(DZ-+28>E8+DvF=f}j>zI4QKaNy5RyUB9z%
zo1?$w%E8GA9zeAA?}s<w5(BtsSB}!=Fgq-V*5$IS+4nq?RPw!lt;w@hh)=2Eu}RDl
z$uhcTitB+l>Fx>@_AF5ls1ES%9n}XRPuPp$5B@-bh1(gwo-P~kSz+k<Gg!88Vi|oF
zpc!;IrE;TcUL#6>J3^7OpEWf5aG@0#n2k&kKDCt}#fqdK21j2%0e^z!<=dM~BJ^yF
zRW3i&g7&^kbg9JhJ{*I6=-wfyK7Ir8J~+weTEB<s27yoOyH^wUAXW0Jtre8m%>g&R
z9Hob4dAXb~iCm-F#u1Zi8qV{#X(-+%LPejfd1QlfzM6Va7kMAd_AmklLm>@HkQhiE
z3%N>pL|U$x0r&mQIn#(oxJw2l{HpWw?Tn{nj90)yGqYfj_Rna6i*C2G(P4vbK5#i3
zUgpRsx0GU}1GxPsG;n#63d=wKkO5Al;NjD4O~O|53_;;_E-^#9UF%`i_i(uF>bCCa
zm_)O*9_)-O`IFDDh)89GEe*1do}JNNlfn5;qiy^}S_^9?ItshL_L+CvBa@DpB*U`J
z?QNUKuR)ya-JGx;e_dsQ<?Ov3hq(&AOJaQ4<q^*NDjS<(C?iwVR}gswhx?d`=glJz
zaY%DZm&h%9`!*;$n%#JPX8#fEw*n#eV${TVO};-2*kVuQ@uzGV))?A2VDqc$z+R;R
z!&}wvdg34VQePX2WqYRa1R6aGU47KT*r)yXVudu}!W3s4nGoCrZvV+m93EDWbjqdY
zX-~F0gs%(tnu1C&3QEJi;fYMAZ4p>f&dMnXYg8R20pVEs{)c5=1CBL_n8n}DAAS&s
zJ<H)#T*v_a@N0PMFTZWZOJ-v1hPoC!|M;;J(m8fgok(l`yS*$NfT5zBlI**Aa?J!(
zGk7v^q=P3<AE_H3$cVMBN(_2T8N7mdEKdo$+X<aA=6Oy2;Hs^CJL+pdDlwV$6!$o#
zo%iRC>^2nM@M~~G-2rzxj<`YT8a2~j9-#r4+x;%N^^j8EG4Gbt1sB$;C3aaxSSBiD
zyX-N8&LYPXy7$xn6WHVl@c)1H;Em;X>TgBdTWw;W^bh60Lpbv6#XHEoP<J7kDdxGm
zh6Dmpkoxs@Qaq6h+0plx98L(Ga>y=$*DTh}D>4S&K+Qm5T;PGaGu}QWC#6pp?^fFE
zxs(4FK&?%DK0qT}w}owkv#`!d<;uH+aj|(Nj4x)iLiq9ub(*j7z<1B<lpnjl4VNZ`
z?7q(VdjBGM;bYu;=k1v6GCXP~GA5>^6=JkMVkTC)RS3_592Q+&m-4Oh{Jpm!|J2({
z56B5KsI6zYH&7f$rb2J(T{Oa?4eFv1i;UCBfl0l78rMbRG>+hrkz`(>k2MeZ?r>(t
zyqy2t;VKct1-6CXJNdajuE_J}FAwomdLE^=Wrw6p3H5u5Bodr&D{XqPHt^RiH+7=B
zR>@A;lG*RZ7p!spyM@>9^Pm`P*=F;ap^NLh!<&3!t6mR$TBRE{x2wf)a*K@2mYbqh
zVqz`B*Ci_5Ad`?rP^cI(YT5FXO&NTT_eL!kb_lw)IXnE-n2zAYrR78s{wQ}BKOQvr
zOONUX3k1U3(d71Oo3@{>EMAdZw~(j%gtG|rsj<fi@ToP`p?%gISkj`ca`CQ!WcyNQ
z?m#6cn_jJ62)_!wlPA0jC>PTFtl6l#tasFs^VM(pgxUeKQF$a^(7x}2rQE%;l<%gP
zx^1wOv5(&C{iCe64(FL(ax9@F;hf~=S6-F&QbPs6sOfY~9w`6|SQ0WzWfb<r@uK0T
z`<^qD&@Jg&mNR77fVaK}fw28B{6*}lK;|By-^o#kzx|!n+i5{VFXd06tA0F$#7gIi
z)MdE7<&mOLkz_z>!i!rX;&owdE9^Hb#87;q>orm^v?Fs~Idc$f4slvG-ZR4IwTkz8
zI2{4U0QgQ$5=)yNEY8DeZ1|$n2S&leU~)~D0t5AIP95dKT!)MD!oG7G(yHB(TXn%B
zx}-<C#YcA&Zs>Tbt(nqMxf5$BtF{-mZ*h4dkLv`GG_lexY8@(M@dWEdlzg3I*6&BB
zjMxyv9NkyZ)=Y+K?QOJrV^3lSF07nP1AS)UK6#z4l2RiDSyd?4f8Asg&V;!tSOve(
z*6(P8{AZe*doA^(kg@IQRQ0_CPa`4I+A~!Mu+U+O7%S}mk**+-W@-NVu*TzSA6u4Q
zN`8NR<t5q6X{H_?I1Ko?*6Vo1rg<n3Mov9?Cy_ez1Yo$Z1PCay)&!oLqaBYS2kzA;
zOi2h_=IMl-UjChv0`mFY+FtM@I6jm~Z3AF_+b0dA?+H)LY1X)h-HJOhW~6#=pm%)Y
zE*_-^V#%@ZGAK!s(sdw_ARLP0GTq<_M}ENut1e#0)X?AKdoAx`On>d0s1w}`dykKN
ze}UTGd%x;?ab|B3Dvu$=;V6n9$peGXX?ue|BKec?ACgsPi@cTdYTiR=k$Uc;{JLsY
z--HeetR6ltPRlUQ^Gt_XO5)6Hy?xHXDO0zrW~$#?d?&fY)BQfr<6A=fdn8IXtj3vo
z<g|?ea{hIJNj?A+Yq6mM7)t~xkwK|Rc5e!Hce@yF@dua<)yuH?l{)wme@P_boWpXV
z4+!q31-xlm5c9=q7`+h&b#{VsoH0?luI|^Nf?N3itr}s{C85Dg0^8gWDUPAN4i+{c
zf_vgOJ+QN#z82|7Y3o&1tHifhuZw7!UKiIqbxQ_(Fh)`bpFE;XQC$tCx8MGrlMfKC
z57lQY!P|*Ihf(pzgp3uA5~dpTlJl)}l?zBM`RD;a_{9yK<jbX!2NUZ%$@i|_gB(d0
z9t+%@UQlk3#?a(<NDxf-jLv$<jRg?VpTpJj9=4lc8#Ti7HYuLG1vj?4rVX=58<YN^
zt^hzm-Z^WeTAukxuU*Kl%!|+()41&n28uE&vq&_V`?<fqDbxOI2^;~FTPD*>A{{fe
zq|+`~e@SGYBrl>TEgM+6^*Z=YUfBl=_YZ{Q!D_ZA^CMO4sxMB;DktJ5JfVAJ0bFi&
z36Fhdttg*=33?IL`aMa?&3HbX-n@_C?(9AW!xPPT!u7T+SQxOs!qJ>2f~E!s3YUM1
zRYO04pE*DnkX~}u-ZMJgpTmorHpi9;Gw5jK1hH#pzbc|TLd9q&wK6Zf8ljZwKyb#!
zKaf)c;a!+kVD^8O9olZ$r=XPgnqP4`YvY5^pnk}J%2%HXle`l;jqoR}RlDKH+80lg
zJgR?ccbjh0W!BAq1YsmWSe@=02$1hhgxsN5)Y4npdyG*306b$+G~?L%6N(Bg1=BQg
z*C+M5mbUs*VGeXbiD!bQ5|Rv~U;|fdVF6M6D&(EAXS%N|OxT;?3l-6yLslmNklBxK
z(o~k`_4^M@?ioN_lh$g(^Xm~2M*N88(I5C2&6uyLeuL+?IFitDZ#Xtiq2-zp4^lf+
z_Nl7{!9MqqAZ^X$H|fj|8{qlNYwmT(^!R8;wu!F<YnC#YxM1frbEz<1z@(b@`O1Qo
zv;6h-i|0QboowWNyq<5bl$DUF!qINrA-#2QGz)m$pQ~Ln9y#Q5=1<1ki^Y|1(82|^
zYY;S)p3y0Te$y9g>>N`gv2{W~WL8O7Hp<r5-=9-2gI$H519oMitmbQe+g+UnQTmjq
z=da94KbF+b!bNVV2mgbGt)(M-gdIZrD1n-*48LSuKmQc+mREa|Q?a>R_~s*M{$La!
zZFde0LDl2rbv5Rs=#2a?DkUW&{2QHv8t%U4^pMzybTYhmww^a8(O)v3FYwn?lJ9yb
zm9;p#8Yfh@NI#JQb?MSbrSzKc1&r=+sP`!0@I#!t0bxo|z?-KN&`kNudfA<>1SbJp
z#Z%=w3jyRoQANpWxf&A?z~I7Nb6COVga}_4EBv<?fFkIcJA(rMHq68K5pAg8Vk46^
z<!jATL~3WX!YiY_6p0(ELLQ@7^gomlRu1apHn=OqQHq*b0Mobdg$iHq8FYLqEUAp?
zr2Q@L`7>Rp_vQG-!fI<y>Qo(PiPaasM`?2zBzeTR**SA3A0bl#r^JMoiw9-Wmm><;
zGD)5859Yy%?#y4KfQH1|dN$w`(t^PNU;HQIFnKr%25NHGKdA1{4+L#5T!D8N@MhAx
za5#%I3%$83xAYYuAGAWgJ}LGufronHuWItV+F2ul_#KN7NNCc*_yXbjrfFa7B<oL_
zW*YkXyb%LzC}5%$;>Wg(+-4s6`Zh{ub;3ylL0zctYAF}GZp|+y#?T!Du>gl=)qJ1r
zz0FTqlJh^}2XBcF6-Su_gBil)7Ekasvb~AReIy^zG%;sT+zVYm3&(EBHc9l|9i`pd
z*bmUxHzoP2XEPNt8PMP@N<E=j-T@C>Y#4B6bWv2ed#F>F@(do4$W&4;m>Xef@=;`T
z^+oqRhkeNrC6}EY{vz)>gC<?mY!QZiu0zzis?)XfMig>gs9AzYMJekG`jdnXN(s^i
z5z`5)L+9}k9%?^mOA~eMa*oR9vB)0Dr`it+9{ss$cY_fqN85b*VWFjmL7zQmE#$Y)
z<A}1bNXe(4Aag&XX#8X=ANIbz`_boGs=SX6SKR?NfMD|k`ju7<GrbmXwUR^(Z4=#<
z&r+9$sfyy6VH52OWgSb%`T?#Yq8#NJ2zEq*EAxWX<doF;{drA=d1L9wKt)jz-oeg$
z?G_l&L2|5u>NcwDO$QtlU1fIY&<eG%rC5xf2no|W{SRSOwlI?O$&o{cZM3ssX^P90
zyb{f!MZ<}5Q|*VeqvGgKr~YtMf=%VarWce#E`{p6OO7L9`*1d-8}bHK3U4I0@lrgO
z*YFYcT0+on9(uN3)3)Ve$7*j}c&Z^i8BnO$;#;bs?dUEj3KPxh+9-jvI6cW8Bcrkh
zyT~!T`;Z=9O1Bl?(m(*ZU&Bac9?ULUB}<JSxNxQ1DzEbfOaX*WM5$JW->rXwfN<c{
z<0+YBf3vGK@ubA9%)Dkc?sXtg(c(EXj!Vrgz}4c70eaInwLpd`j4AnkrWNr|Qb%Bo
zj-Oa#Z0zpGJAS>LH>N{_Ub}XN|HLzNbbW;t;zo8!LUJ+kd!nv?sw4aJ)1NRC&v1Nq
zyuTcF1BMHs8dxz?dGmc?A66%D59V-U$A1sz1hN~P4v`Y{pueuywE;bPYNuYM6v`XN
zMm6HUzScT$B~Mwg`hDf6sRKyrJT!C<e$UXlB>N~cfs?TT10pvLDA2`ioDTH9^jm1l
zMwVBqd(=Ja`j%V**n4r7cM1eg_N>MOKR9ZJ_YhA-@-wTwAB5RZwp=0i);DpFHY<A0
zF=t%(YyREYcxKi2!|1E(-G|!C@^>6;p_vn@Cp)%3Z(of8X%PJuk(P5>k<KPJ*ky|i
zQsK$nUP>Yh;XL>9%c7_Hh*Pw*l(s%*YsF1eeJljRk?acABMngtZz2te(dbmB*1=1C
zMYg;@S9a*RK2E??)w=nBLj?cs1|swU6n*-afN2s3zVclu?N=pf7Gsokk$gO3NXZB(
z>ekHKR9^ow`>;TmS!&CE`|6BAUgvt~KA2Wg1{ssA;l1e987%P{VW9K~%I9*0FMjiF
zA4d=v2=_3Cm=BF4d%-G{s}HlylDX$lZzs;%np<{WhH7o@$K^@aihR@wRE}R8b{gJT
z{U?qi1+b5hu;_8?B9NEUbkMtY;;jYb=EDV_c0hJ)g<==}6|d;P9}!K0dx9M2MZ%!I
zAx1QLtHw0Vn#iaeRwsY>KY1=-d=ZhE_$lPt^gCTYyvneUo>^U+&1`G`a3+EX+$8yL
zuXjP{?((u+bIO=n2toG&8OtlWtVg#wk3|9v6pv=n(;ZKmkInDFZWQ;p7z)Fveht0(
zcO)AU$@I?+rC#x4tJ0*pCjwXl^9v02H)&t%1|zQJB9tYY&A^?>%+^f$3$i0EfOf;<
z>)ir-RN<7G+%h-66ubTb@6n%W)A!4hM^=x1qNfc;?m4E?&<BUulasaVFj}2s_&J2+
zLEd<7Ih;4hd91VELg6>n#)u!nVvq!d6D>=jDwg*9Use40Dp6QxZ$9WT!VhG23|fAy
ztIdP^X@aB!yQTw{0RhPHZ->Q`6H@6|&G_;CPkJ?o2J&f&>L|_D2|QHY=2~itzv(#A
zs?~xQj@`j%yRV_Uh;O(5whXo4Oi}PFgm-*0-3uz)QXguOus}Mm=+HE0W{ShE<1obp
z;a>}g7T@=Y!_fc=7JP)~WC&ko!l@L}dYF|P78v^lN_G?7+fgnmy49~BKlC;y@RgL%
z8<f?GV*UAXVQ03yu@tIr((?!Rw0-p7yx?mctpVy)uSq9wObnGg9kNvq-JM?P1ii}A
z)&v+j&Ap{?qUnw%8FkkwFnW!u9K6R6Lj$f{MSkn7w9mcpg3*hfN3`O!o*+Wj_r?~w
zZYL5S#Y1Eae}*{u`t0E}@E1ur7TYrdB;CQlF9U<vy^njm_g{BJF5+z_f3{xcO|Kt{
zxq-#jQ4O30Hj_q@zRckKFyV~qva_S|I5ZN=i!)KzR}J{tG(s4*H%_ShCxPNi6B#Sx
zvG$i!sw^=%PdIW=c-^YBQ=^PfY?u0MMW82HAiNw#>YND-Yu7lvoN3~~qKsn>PXHA&
z&l%WewRniCv@&dBVfy0yS^{l3v63|ql5Z8af9M$qu~Le&gZYSuQUC@BD9of3hnVL4
zw9X!>g{`l~fCCX+P{X`989)3ks}_D?cu&>k<&<hp<-|`v7v%ZX3-!`EXU;AiW$#y-
zpJkY}cN+0M?BF;$9neZ^8k`6211hZ{iuO!|Nk0AQAD%S*%rN<bYU|S6<(zUi@@9vR
z+2)72=nvXSpRmo3`{M%6Ni?Gm>*4j5+Q%;>VjPM6Xu&9o9h}vWY#^dv)C^w?fI0`$
zdwq@3&)}!KE3R1A-}e&Ojds`SH><idxi2`tF<405((jDO5h=q3ysu+<V-VUBxL;I`
zp`E339BwT!7w_%U#}`8n<Aj%2ob{eLtSPylozsB8h7mDq&M=f|$DIxg-simxR8k4J
zBUDnt_|C<}4#>E0%=<d*`<qrj{2;M*-6~mgLVXjoW|Nl}tZ^9xd5^cft%a=1qAEU4
zvgR@)F(d@5dg0-AQ^Wo$seO!ce%A3wUiwVsGZ6m$oA&pb2r1}*bzR8fm5}ld`0cmD
z6qPoIry^BOL*0`Rx>UcK_H{gJ5bgDSJ>X#0`W`UsVTAVRysG+s1hV{|z%QApJulsd
zXXm<11M*`>)@(c*(C~F3CbgD#30uI`U-J?LlvJrZ6S#|*fe!Q+yY+&oa8TvPz2Nc8
zmXG*ZH2=Mc8>Au$P?LI^cpsn~$vNTeL+(o%0;p=>Qzl~oHDT#A>!TX4(%XJQ>+|}2
z8f3?Tr^_jWB{ck4<!<zcz)%|D#!d~&%sa<P&Md6mmjT=Dv!Kw`U3f&ffa~`JFkhN;
zd3NdWA#PR)aa9||M1A(cBx3dcW-JaBX_%^raqU_6O928mcthwnDPPC}yQM?nEx#!C
z4q+_4fkI<4%Me{}SRUiMmP~pNan2uVwe6`@o=KT{ui6W)n@?XP@Fd|_@-?^Po*5@u
z9Xb(0S~b_?Q~Bh@UW>d=jW!6Mlm@%0IBPGwnxuVB4-fay*hA<$&DjaU&RxlqI&j^3
zY~ti|0I*ib4!9RKOIVuisXa11J>SA1$$dS*uehB3gN>u{$n-yi)wuCIbi#bsnHlNr
zoX$c=<$=PrQZLZN3HRZwI0eQ@6o(u*poJYUx*D@4hWBZ10W4$eKrV@VTAnxN?;G$7
zE{P@LhVYOiq@qN*s|>4+8u0Hurq}4<@2h+*jrF&1N`)x;D5;$09(DyxB3we#O4s2Y
z-$gb|j~$$R#x9)@v_{`Ne1@BP!qJsd;+V`ElUGI}I_8+z)h%bt?`hz~_VG(&qTZlt
zrqd8YC>O~_ranRr#yk&j-dn+3iu5C%k@N?&a`Sc|>TVu@B;mm}*$w)CLF8dfos9R%
zH46uo4<%8gz#)X{cxBefs<Y-6`^m)Nyp5*MZXM09bqx08QS}W9^7i(gUaE`dy~Drz
z$o-78L1^T0W&x;bA6@)%M*qdfkDM9Z!LWVdMDJpg{qvm+jbDgDFo+Mpexe-4P*&lQ
z!C9Pi)sk}_aO<P~8KI%3cICV5$b=nOB939GjSM_N*}(l?zR;AG5<6>+bjRTuj~9Y_
z3y2(u>jv*HnF5j&7_;yM8uE(i)xO%QLyOBu*f^S{{8x4%-v&GS>@&f}t9N<23Q*z}
zT-NG5|E;;V^DP#S)+6%*?A)C})ya$9Hgj3+N%zcJ_dehRsgBOyh0-3$es_K-+=|7A
z+)v+FOV2$NlYH}7*SraWui7JO%{G~(nf+7vh@RRvUbyDeAM&t!Qjd=7D5-;6P-^W`
zDz7)rRS_SEykPZ%_<l04K?WL-=Y$o+QU<?+;L-nF+68QS-64Jb65IS)DyO4L3};&8
zH~lD={`V;EIz9-!1dj_#bO?9A7Cg)_y9#la&J~mJ)&vf@BuE1N9F~rK2m{$k+OP0y
zaUqWo3YcPm6xE-a8Cow;nLbenkES{}fqcU!vvTL?;$ecisVG~mdT-rV{Op<jBt_fu
zphtw1E5QyH{<m5^GFf*QL^{~301cqlce+h2tQk$S>Qwxz{aP!c&Q96}d>R-?i(>yx
zefp%-PAxq7E8+mW7#cyORVL@MU?JZ<QDe-!{_~4>7F{VrGvr5@Z?dkzxQ<#SDwWY(
z>{*j3Rx!@AAeJ7BA^HmfmKnmWs6mWysqP|UDr2gfBH&d1A#a12XtBmqA|gDCeG6vr
zrCrVHAnDwMf|f1Kp(gpixyUIbZ4i5hP&rgg_X4@mMqhs$@Z<q>_h?`6wY`mU=5P(U
zq3TZs<2{wfuvvt8h}f~f^mpbgAJe$-Bq&@Cq!f&dFELCi^FBI4xF`MyP1c?LAzGOy
zM@6ZR`*6sI8jicQnUq<=rR3J2j49e$_ksoc8D*1fg|^$@E3y}Jb9I+sZRWPsPAO)r
z8OiM_|6#&oRwNiFIxDmYk+{u4ksk_BNl>CjM)e88(fL0>hcA3gen^8^(4Jkhaod?2
z-$Q;qE|z9%=ItYA*03IOuuah?^jSYa9rL;b5-qVrK0V8rg&<Vv2Xc>V&kX-&#Ggb+
zwWwlpAR>}SEeP)lm0;1ZcLx4paTxOaf1>|wXpCd;$a?=^FWw8~J6nw4QFlbOgwo$L
z<A>j8EY(HA6%~biP5)em5ttoq&4(Jbv~|bsoxCsA_+5=k)X{?Tz;jC2*jSL6=Sfre
zda|-ElRr=IzWvRqzzcJiBAQyN_IVt>zyH6;Br5e~`Npk;#xu7d-F2XF;9UX1Zpy01
zf3$yKuG+)b8Lz2U*%=(1g64Bo)R>9rVTX`95rbSQ4TgLHD7}AgkSy>^*5Mnhb4V32
zPFha4ai+q^W$$ZSWnPyfhRG=hiZgcTRc>itOi<;EH-GCRWTgXv>IV3Y2iMR^UUsv#
z9kXAMi3#)k_d8}wXM(o@sKqu<9Jp{f4d=!ivm+aDidUD(s7WR!#$L)sC3iS+w^}8v
z(gng+nwRI#e0L0jwH**;q_kH*xNbIR)njl0eaN^EFnybkF1(E8s;C(5tT8sCX38Oo
zG-|_?)V{zQ@_FS}-TDdAu9LopFi#kjNYRrI%fFez<6%N4Z~1zB6~q4NwQ&YecDF1+
zKArdTNI|;JD%toQ3^ADJV)){si1FA#Ej2Z@s3<i!9c!0p9ke<(>2nE+4&^~Agvl$x
zNw6!}rxOPTiR5s_p{2N)#fTFLeUpxNC>LH4Di<e#<$NG$$?#SW&{?ok?e$YHs$8as
zJnAO{lr>(g{Hk<g&fkJ*RaSY@Tf`FmGwc&uh8^%A+szD_-G1EiX5;zBD)gPZwi@Rx
zKFck<1YCeJ__kt+#8I@}qgwQ8i7~_0m<kv5HnD2}j~&i6a+=)1%+R(=4+FFWk>B;*
zhh2d}>F^>;_gA73F<N!P`V<j`hZvTQPR4h&2{#&~IpdE46-Hw>?Q$OY?NBCPBt<d_
z`jNvdPItXujs7o6n|&Z^M{X<cfv6E3_0@0b8!R}^^j3R8AXvulp^bMq`)AN%pbHh0
z@kgSW%Q$bY*HfEqBit?`0cL~>-4RjF;1jXzgp)<A0Fl3N;WytiE4<jbIGiQ3AZ;*r
z>ZF%l_mnZ8YO)zX1=yALnwxnAw@w0WrIXUIKF!{1G|6zxP7ArBMqNQ3Jak$e-TpNs
zV+OIRmlla;6hDHmH@t>Dc}CtdwfLfVAJWMXsl)fT?fW1WC|Fa@iPpXHZJ~~H8+G`0
zC~XF;$6=en3i=O1a>$i=D1Vsj>_jf1(4dYcO6&36IVtp5{wTNoI1mrja}4AqE}=y?
zqFhQ-gkaW+{60yi<APgq<tILiLOq6co&h;N5$Ostt1raEO`~;Fd43}Mgw6aEmXR)8
z_&x0O`NSrPlph2DDZb6e`ZS3*mU(jQ(kWl3j^5<=%qvKj{&bWy4k>ih!gqEiD|M1t
z%M9ej=_QLxC)vlZd0N#!WtL)KYPbhfaGNi4`x>98iy7^MpH#WZ{l-yQ`8#b+N<U;8
zAxgK^-dg9!XVJ3IvBuJJ?2n681xau&K^Zz8`Mt{Sb(MqADTN8PaF+o9Q;>F)+wP|y
z$LU?m<Xgi<atZ;SMe75+tAWAXLij50kB*Tx#fBafa9v?1nI5Qb8t3DA4b2tg^qD37
ztbnIsG%Sk+Yx~~hTYVEN_^gBN_A`A74|YVqqsyknv^(&m-5Q?tGA<zX-9Ppv5@z$(
z(Nu3phTEOk+J{w^apdD$i*fGOgu@J0mCo8&kWF47zC^&Gg`Gx`DI+WsvdG&}S877A
zOlR77RI3jj@GyWHX}9{f-Pt2MAD`)a6zl9R6mbzc98T5_1M&-5=S1|~&3t)dKkgab
z*17_QT(Oy2YpgJ|-KI*9rkeROD=YFk&Qugw)WxtJCIaeiEkgl61BHTCXku1mKrp*=
zi4c*3%aofSBK#Tf?dfK0!*lCQx=VRaF)GI>ZJ80BmZBntY{SfgLhQlKXZXYL^~p<!
zbcRVDWH7_KLXNr@bC`09z0zL6y*FIx7>)mdbDlHQxP4wq9iD8XW+W!+3B0U&+BSUp
zNDp<u1^4Q_4rh39fr>Xg+yqb_pzjcHPEy}N0+dLOBP+cXa3beQS=`U?KYi0yj{(Q#
zm)_-M3nB3CXa)6OU{7RdH!602OBF%u?;SFllyw+RQq|evuo+C)hX$)U{N84%>a4e1
zJpqL@4P{#Vd#2tXF%J9Wr2Jdye{|H*4wl_x0BP!#^3RbmOyYKe+?sVF+j;+eyx=Hv
z`~OjO<$+M<|Nj}|C?(mnZAzqrQPH<V$+^|WNJLVu(cM&zB}9gi{VsbHWi*wu%_QV1
zCFii+Y}7bLuCW!<%(%jk8N=`Wc}Bb6-}CQQJJ0icp7-&3z22|)ks04)$Y$qDftzUX
zCYo_)ze&wchw93eY)Bk>am#79r*3hL+BVyONVIS02mov*zNkPE+rO)PaF5D2xkld6
z@VEVEKgs(LjrR=#W}LfvKnKlZ?!+y;#f5@VSCZ_t+hN^*fBDY9P0~H=@1#Ai!*@x@
zFmLMhOhvx}oRJ>|%sx`;!}RpsJI}#4)c~#_&Ss?H+s!@pL`A}iR%GQ<@VDu`dBk2Y
zGuyo_fc-V0EdvXICHwYLIC%4mA4JH?8Lo8_0{{>3jWw^UKcffgH4}w+KJLN$%@jZE
z-yfC*lAsUI%x%Uqf$n{|gT6_sE57kn@j|bdJQO$r^;fTv(3GR}rP}a#ZLRGttPVCh
zrSl(mfyFI#P`bd`nx(YE-xIi2_?bpPrmp@Iv!%zmv`f5U0OMl;q2~m=_|~-46AuK7
zHe=2Nq#`X|`jdAEwX1|tjbl`q%c)hGSY|=@;>1W?`La#@&nIxe2j&oGJ?9{q1>h(L
zAjU5p@E>~1JOfIOUigtgRKilD5(0hT#y=2SNqxrvN-%nV0H}=s;mkn}o#9eg=;y3p
z@jDJ_J#j?YNnD7vJOnH_zD-unNogN=BwmB*-foUVCZT0#%6>VA87SeBDa@k8BohHE
z!CVC_P={*$g+_P<-+XdKtGN&kM*M4feF+cFq5P8G1$c15v{l$*M5??~p{DO6jh=&q
zQmUxW3>yBcHvgf^x8HdgC^+8%KI8aLQt)*)nV_jay;b0JzQSZ9%LL@}olaT@57d<h
z&SngFlHd^zq9Kp`Y!E#`!?3TsVGps`0rglK65OtxKHz}Z1UEAXb{Z*!V0<fx20rkp
za5Zerz)^&?!(Go_Hanp%U9lcSPc3>jH)x4{%WgQqXW26Gpy#6|V_%5TT`qZ9v#@3r
zcKo|b)nC@@9f2C<6tlErC|VN!l?29m#>o{{e`iC@jHB7TXhOsQR0wMj_u9Fo^XHJm
znae>rup06JE~8O~sZk)&tKx$eXd=15(-Uc$=ok5RD9%^F+^$$9WHstj*%SBzBvWE9
zr|go;c#KF->;#St8N1uK%Co<pUW(kPP{)0jjk>N)8uzy<FmfI=tE5Q0;c*MiD`yg3
z?$G6WQ0J+#6oZLH0N+B{1QM7}54^kmZ-1^e1U7Lf8dbY(I?4|#y=RXy5}u)>++jE%
z+o*JifFwv3b>UJ#Q7xcufoi8@FSGg@s(xpYtA3^#Q5DuUZ4j$vV;_;>dAy@d5J>nk
zoKhQsksh*#g8amT8x{Y2fp>G^<XD|gulQAh79JzKpr-&pwUZ?Eb>b%w%v(m`J-A%-
z)9&|Jx51{JV2|Z)0LBGC5BRr8HcRQ=1{(~5+)*9>N2O@+T-6QbFIeDaAam0zeonC)
za$r?i6QpM$#rY{Mh7{L-z$0M8K#z@wdPpK(AHCcaxCtA6;}SiL6Nt#w)_dmWbS1;;
zJ?)AltI4JgaE`yc7tKCyjBy8@>jXp(uC=*g?!Z5n%0-WP8$9MrRVcJT`)9U~ywMDk
z&g#TM!~cKbj;+|+{w3=yo{N8mTzr*ex0DX6Ne>9lv?Sl3o?C=N<fG%B`PZ~TeH7>7
z8$=ZaK~IqZ7#v&%j#LrdjVTG6e~f_n3tu;?<9ZDZ6}aTHWa#CdIwu=)b)@nnNS^UM
z&*bax?}S0ZQ#*~}vs5P3H6OYew^|>%w%iXVfq9;96xcHlJHHhjjd(};NCBWWC(DMm
z*F0HwJoW)Y#Pj0xBy%nnH4SZ1`$Fg$m{pwYtK-k3!$z(YRLlX$^doR+HKDAK$DD%A
zO%kNNLbU^2<fFbXLqA=B?S&@j<J5?S`4NpDQDvs_Bibudy!6wx=@rNAXm3dd_^5f^
zcJ|k<3Ugb%HM{~9vS4v%c>WYR<LHFV8T|waMK$W%)U$!MH^_&!4o@j$ddSs7KioF9
z>gXMK1YcZ1w34n0SD(G!?;vcggCsVH+_v=2+9^~zP={h?5!@7jR9`@bX*s=Etg4PN
z487g2ef#lBtEDR(61Z&K(6G_Yr+7|M!rCs4O=paIqcP%DdJEtxh|ZY4K$WrYv1me|
z7%sOdu4Qi0`C80jWDXGQD#j3Nj(?N6N1HYI5l%4Fwfx~}YrTYXwR;Y62z#-+Kg+v4
zkhUCx{8D6Dz!kEZu`4Aj4|xR6t0`b?@_h(-t1wzg-r+5VAe@FcSUT+ugcN9q9=jmU
z<);vk{6pSQVHFUC7Zm9dJ82FSQd-6DPFmtPSqLbX3T7;YE7M+-!O&cMTfh51PunKB
zYH6m|M0h4E-TLoPWHNW8sSyrJe>V^0Bg>ApLtZR9bN4D<Pdss)vsXD&>tD%M|48Wn
z5<K6*@?_Go8!T@X{#8?d2kkfT=+|mQ@i2RjjCIG;3K$_y^h?NCc67XBU>r3?So0yR
zwzLHb4845M?Am7h*~>zpKST5w$Ou-|dB!QxY1g_-@wr&=Ljy55zL^IKKJ+?F`OC<H
z551wBvJ_2*1<DS}hX1?==R3_ah(8OJNz3gg-FeXf_GzTld@{*i`+8TtNX9eE$uOhA
zkqzYs)V^B4Vdj_2JEmezczI^ycag>Q9cN6e4&u7|#$a9;_6>JUsq5)XJBV|{cx>ET
z>z>E2%{Ge8m!G}g)!Os%fJhLICZzA)HdCQ7*X9RYYo*|i*=>K_wx=a4t}FDDk7Dz%
z{kzSX8}ZKql{NAYV)o<s56)tXML}lNh41k>@299y+lkb!gYD?ypv#v6;CHKS>o#9!
zYpT&HRn3BqscX#>H~NHG(=`ypDx=u0MQEik^br9?sHEnp_#2Cf%%Hs>&k^>(Av{tl
z3?)oCljkkea8|ZifoO3JjV6J0#^6XO2&y2TV#nzbn|p4jbA}#{5iU}FHEOC2^H;9g
zGM})Y;t^KInyof4;rcBS=AE$+NA$~j207cF%4Iv=S=GDw!=T4^$?(&PWGY<2UF`JU
zBfaRK(y?kX?KKrUBJpTX{o1|Cx!A1+LEkIyrwExqi@^5fdef|Fq+6D;*b{vgDP!lo
zyy2R$N;~<V!m4&O<*fJD1{Gj-zjy+Sr82sFx%4<o(SbP<NzC*|xOqio6?!q?bt>%_
zbxvsehLPXJ<34^{7<ZfD&OZFd&xY8a=K_Lb^OK~Eb(kKiJ)kUp=m8oiBD6xB2oDlB
zDF7elg!&$?@T}nYoSD^1UCC>eHY;H2`f4S3j*rJd2S?`-lwueaPIXW0h-zSX+qlZq
z_7y7%YV6sI9+xQjBnQ?y5+x2pH}jg|CN)yZ<PB&_)YCD(L?dtI>kUpw!B#V_xg_;3
z<8^BewvVdOyKXkJx0OwYxI=G|4N)6tP7|ArYIKsLmhN@a)r9d~5nbz!(^HgJWLf~^
zc&53c=NQ&=m?e6tnd)`(ocE0dc>RQ_XE#q*%n}I@tmz#NLn#6p8mctWdk-7LdL$D9
z_V$Tt9-tYkj*NzXmV&~{>d59d#f-^+%0H~56(WR6bVh}0uHNfxBXQ;PiY%+<<^YMa
zi}jI+N<fm<zvt6_>+6N(?BY8;5UP6q*6yC!4Y=H?_XC{T%G_YOCcO(B_bB}%<K}76
z7Qe-xDz(1ZTRntTxVPIZjQu!xr&&$FAyTlxTmZSVN*Who8i*-JHI|>q8LJ`U=8!>L
zCkJmfddBo`7_D(vq1+l57qZ?=%)y4PHRNSayuyc`<XB_dd?VC2bUt*^&T-nN_T3($
z3LN~6f24;0duz^0IWtB|B<?PA>!iNnvkpSfJqd^!0agP4x=viaTv^bC1if?k=i$3t
zHjR|!iTSZ7*!LLoT3pftFor+ExOP50^7%-1?G_v@nb6kF;1qO_wR@PJ<<sj!c9{Fw
zT+^CT2=tT}GaUwgHXCXNxfxgG2lw{p7^g>Yk9OPys_dQ#wTlz71|6f}f~Z!U{_pUf
z6R8M8ck6T~k?{1$=ER1{g#{Oa6MG<pOu`dLkNeuZV{}7&_eW5pU!?;u_u{h4CcE{O
zT(fn1(a^*jl`_8^>#~yYYM|JUN6*{)t>%7pmZJXl^Qtxepw+513MiC=>?cN_{03i>
zE}NskC`*|=L|Gw;C7<Uzhq3OxU^S{k7n5A^Ej{H1oYq(WR+G*4r`wF(azx4K9-J?7
z8g#1jW>%~HOzK_kb5&uLK3a&k0V~=dkQohls@nCLqEhR~=JWqio3n=}nl+<@zxo*2
zc5GePQg@;GVE{~aiS<ln1!a}*j3*IM9}N-Ta<D87QT!igsz8>hy&OMuRN~yw{E`<h
z{;Q33;F?Eyoxh}%i=$JZLdoF6$!pnHr+g!3+eA{qjwan&4opq5)XphB&UXRqh`7LJ
z+dna4SLL5X%#GRlid+^I=_(L0<hSL*Rp+sAsyA>6)L{&$=pGlItG1{rtXi6wI$i0X
zVSWBPYV=lGx&r}Lai9IR_?;Q7jK0+Q@So&WfrK*dA3EKd-4Wt4MH@7{Ti(oON+d!~
zBeTyvTemCME+IgeVFB|U?CBPZbZPkLg#Ts0H9EUc^4_-fpvT7Hg5+=ZT!XaTa~KJy
z^$->2%Pidv!zq_wh=b_-%`!*z+Ykr^69I|3&hFH%QpUt~5!)slHqx}Z7r=-K4vz~n
z$Hc`hTzwphj0<aufgGJ3oC3>&1s2`2Wm31kxKIub`yN!aTY<=dpuXjayCt0{2Y*{U
zm%IX0PIF|CB;{Ala|ChRs8>oJK0yzHl)-IQ@pL1xPXj@`%2dX@$QZiA;B5y|;T^i6
zxeYx`*1Q=PWz(%MRcoH$)EF*C9(X^F8b$Cj8hlU)shTeQ&-r8QEM8d7vJC3NB(X-O
zKiM<-4D{Im&~!E+e~KR~aCO&fi6&Ol1v6vRr)y8$HenSypyzL6Po67ujP>r;Ph+nO
zq5UY=>0kk5X#$?JKLd`c&t*H!Xm_twTAfXxeXT4$j`0X|vH9zk2Px4=R(Aeo1>l+Z
zMMeyd%Avl3efQ0Xks`e`kwI^z<ISeg6B#F<5!=m33QqRCdDI9MivB}C4l4MI?!?X;
z_l`)-jfz>Hs7jA3M%{Zngqs<@$XarC3H}mkGV7C&rszV_@!NSNHm`CWmZedNg_;71
ziQiTnS-pv^_?|ixd@(6z1m-~Rn+yX&u6cbb0C5F2ugIP+F1UKYjtRIP=an=Z^jP?#
zx!H9|Y-)v~FB4H#M;`F5D_m*Hcw%f%c$pKhl}4%Cwi<i<i{`;f{D6wZ(w50SwlS^`
z<?^*&Sj%R-tt$S$6dC^F{j_9orWtNL?nNPwc_F0`y29|`_u0`yJ!0PWPx(BqdOAO%
zWHvmtC6^@34?`ni8Y#gGS&dcpfQ}jKA%{<kKT@M!?mv0+iL7t(=9u)#I&rArM>0R$
z#d<wrhMp+_))2$=#9NTR*OKS?6EbsA`bV_6p9tp3v=N7Qt_k4CB7Khdl$LLj@@TT2
z-^)9fJVXm^1mr`}n;$PWXpEeG4UpIpXdpYCJWAL3$&Bnpi2@rOfi9wj4D7I{U8GhO
zm(*_0>4w6;dZ<wW-07vil61>TCLBf<CeF7>%hLUbXnqNkTl*s45eriNVXy5gjpDhc
zQIir;K=+qh%tH2oQ$Drm?5DFo$K^DbiJT_Zbv3g6FEU(<@ClhInaFymdcncz46_-~
z(4nj67K_#u|7FrjKdcpRFMOg$fjPW&PB4Y9VPR~aS!+MBN=a}>_dREhOLdB7YFgjQ
z8--BJ4~;(_n`hHqz#zs76joq^CuZrF8&i{F9onPU{}#Hu>gn66+B+5R-ZW`j?RVBK
z`4a4oSeC-d-=4+aAz1br6Nv9IVjc>Di8kBdaJ&`GOwKg?C>;pBFfNJ4XKXO?esM_0
zga)V$;j$=co`55DEZ3|hLpdL-@y?oV8k@daZLZ$Xkrl2Spu9S}koDJg*zUus`G81u
zNmxox&lsQd-8MZ_LhaCz!5ZGh+c1bXW!SxyA{kf~Jy|GfREujgBh;g_RCF-9daFgG
zEc_3q)Mxe>tvQ7FD)#hqtHD@%X_>}K*ijjb0+77Q-WSC_eUm?W!uoiQB8j)Be1>F$
zh7k0|-XinW<$Xr<fm29bvZZsUV4gbm(NyaaY|+VUQQv+D>ES7;9M|#r2uZ?%;i9m{
z!BwK{*G>=VQf1_pOqdZp;WJ~_-z)S&ifrkOCI;KhZ5h@}?G7cL)3uSem;F8@nL(JP
zJ5%EBGEAY8=S4zS<-(5CY8Io&f;Y^8GWG)=ho5CKJc!P}-g~_I!UthjnP%LcA3@9{
zLtj26d?y)wofohQcl?Peo9-O_V4bfrH2p;+j*g!61242RAD?QC>wmSP)rGwe#NiF@
z>_gE;w8_tsiBT?>oxx;xPfc@%W|5B`Z^&)$M=XHLL^5&I#oBgdMLp5X^kPmwX=Zws
z+Wc)6%W$=(=I{eoc$kMJN6kQm+PHX8>iVr8Ldh2@$xBlSsDb`EB=YXr04wt&P^IAr
zMZH{p)+ET9x^Zs~2t;h)-fK?(|G1pr>Z<k*tFc5u8M?q#N^3o%5gtN^qSm*b##L`J
zKSo8QHCrDWv6Hc=y4<&K--=u#L=zh0o?2c^<#um0%!9xkT)C)<y@Ckz?7S?zZzNC5
zPj%M`xOfGvc#N6~TJ%@S#R$l+!qcf05#9P~CQfYjm=y7vUqqBusKcaGcTP<=jn1Ae
zdRz<4@~l*4Y_O;OlH*i-4v62S`)0~oNlVr#Q5s=2-%nOxrb&g4I|$%Y%q<uhpt!?@
zGz?|GO|BtrUFH2La5DmNfGqpv+{%;%`xR=8MPjS)6~!?7_4$kF<iCk*gc0N&gj6es
zkNSCX%6Cw0)9tm!l1z^gKqSQNGPygB9}B3MN`nK3Kkd8RF)(2*v-m1vBVf)~&t>YV
zSs6Q)N9_@vL(?@-sD_a}DSK!J6W+Vmvu64n0dmsWml0zNF)u>LA(l(NK>2W(HM&q>
z5p8b$A<`OKgBJ9hFJG}fBD`$!=Eab|%c1bxX5rPCXC7HS*{vIZK6Ruik;`HRi8DwR
z)7&^KBa*qd-9aiOl)I8Zh5i%m^-;2{ULb<k(px}^LkJ2C{erXRL&&}U*ECNTFl@3R
zhQ0)C)|%^3KmO9@?HdH+rt81=-Bc6R#=Z2w+dlRSIob3efxNFS!gCfj2K-_Pf_2`V
zH3~-A?H4kIN4_+pAiwJgbZAzb)-G<2Ssk);zG6*Rmap6gL5#f=MK?s*#flMKqxT+N
z67CpZYD!n9Z@LFE*x$L_QYIzenp-&)@)VBSTgr?k>eh|fi?9D>>`>bdb-~2Q2k|KX
zg>^nVTKriu8QeQLi?Tet4{{23r(pMe6d+U3(6o|NPrU8g)zUF5(wqohUq*KL$DHVg
z&B5jqTW|>Auel)d^)o;BrnsA2mdGoecMcmo4Kpo|>N`Nw3l(Z3*1N8^{HFnO?RD0I
zR{K~*hh^B@{*2~1@HA2y=h}qNayvM8hMxl3Rw&p2=>P3p*YyS8<L|4&y>9YJ+~5v-
z*=3fvG%MsV^qSI@RZiLTk7F&q`4LED7ysHyC~tXM;1UltMH>%{&c~!{lsb`a4O~3&
zVklac?bwq~Kepmec&FY~3qD4P%@d_BC;=DCT%E&I2WL_2I<za>Js;Jy@XGKJF6keD
zf<f(`$a3IYxX?$c?b(g^fk0!+32kFfTD>JQv<ncI$c?`}JgCrC;9@kU=tLN<pXEOv
z;gfY55OFV_!UpphgmtX><LkO`y6A5!r~?p0&(12b4V%UPIzY5`s37p>YjtA)46(;I
zO48j<KZ(@(n6|*?>!Xppe<X2hGCxio+pf^;DqKX{Z165TKEhsl8y@7M;-#8lFqx}S
zy!1O}!^td=>$`z_IP<f~5%~0dXv+}JIrbsL6bG+ENlL-lnS*1Mdm3GBp`S+=z(#nH
z0MEh+I)vr@$D7x=C5~JpXPit!q3@?^-q;04SrIF}qb#3MpRB-u*<u}tvR-5?u!XT{
zYbek7FT(i@P;<UB^3Q(z#dk5RO@onhS_UCAo#Xo_Oj5k|uJ^{ZXeVxKWS%BppaJ>I
zIEkbNkGl@!J$y-nYV{iyaS@9I^%43}!YoD<!%s;+JEYE~L9j+JaZR#<VBi{IIfN(-
zd_(;l{(X_&YbQG*kApqNN3P=IAHWs=nt#QQ)=i-Mfq4!r$WHM^O1|6dX%dpT`^Zkn
zw(*73LBqeZVIW0#x-vZdj&3hRg$v@(Yq41Y_uX_9Kq9XJOF)IQ7rQ=~>$^&PC=}s=
z9xJ{V&O5O|>1ei9dpSReWfh2=N5}X5r}kB+BX!BS-=6*17L66$*gG?6q6gb4`;I|R
zhZtwn<qR(*ml;7>;3Eq&@Y~9)d#>l~#=2LSJTi&-9(^derJ+%Lu=3^N)JVM7Az!_?
zkv!KV9NM`!{8j0*#%`b8-nLcg+iNG}y+F-?8d@Q_)?nMf%)V=5ldEZ$akEIrjX_$&
z*OaRTg-V&nV8k<MvPCjlUr*VM_9{(%IWAAfeFX~!*D8^KO@%RhQ$o*0c(>@vG)Ebo
z>^ss5%DvWF9?aafy66Ce0>FB&6~`j8+4wIH;cZ9R!;lPXhUAI(7a~^Vq^QsJqAuxY
zE$CPB^+ZbMipOcFKbK44f_W#gpdT&odkP{RHrrkVj3G+4&c=-~QhuyxnREKHBUjdJ
z+y-Mqz_Ft^0Wj2S?<d}jYJZT$gX_S3sPVb%1K#kt28{BvW7c&nI2Cm&$GTN_G5xfI
zECXlVQhKB=-EkA%!rvS-mT+nwX(pyBi&?Zc94jX%snB}B-hkE&N>rrwYU#*_cy#b#
zdQ<x>H^sINxPf6|2RDQFqAX7yFX=MSn0)I#a}|yOM3em3d!y(WQ0`R!#bHGq4t~q?
z$L(nmwDd>iRDKFB9~VmxA;#LbV*zbAzx3SS2kHpq#@$#enk<%&tU{Og^OIXO$LRT_
z-iW8lhgVQHI_GedH{1BL^7H2IRjva~iBnMLEK}AJahm=e7DyCGHmV&kb`*&Ii=KzG
zK1+Q2>;;T0j2nQ;s&hhYgy=H}FK4rCKAL<IB7pkOvY&G!MuTm%9y2dqadi&TdUMC=
zTe>;gCRkqxNFci9kq$ARiD<CcN_ae<S^}aZLmyg7%kt#?KV4RYBBlAGM|S|2QTO>+
zpn%U|uM%#~UMpTS8Y-AblOPeOS(qarh_75TkemSzj+a~(95jD3uW0Bi{dcP9w~jM_
zsO{Q$&V1(lHEA`mnJLnUve(*wQ26sv)lAZsc;&kZ9Oag4_~(H93?6rdR#?`f@NgBy
z4>aRE76M7bAcp3L^*kbrcE1Y^3~~|rYJ*vOOX9`;1hf*Wu8WtKHx{l}#Y2VA)3v&j
zNb7dwLE#OBWDu)#&nKfb4osbS<ew?%T)BU6OvD`>>nVlYcLfc>?z3317|}*dCQzNg
zVYZlW1+Rlj6wQBNSo2KMM(t?NE4_ss+CcD%C?Y*X6COvn`;m*Yo*Zk|^?vW|ytJko
zJmE5SA&6jLb^Hm?pVvch6c(1)Zlr=Z(l8{^mjk#h@XO<#RMBbHJ$Hq(P$aM4aP0Mo
zQ9a<OF>+$-76(ggpG2Yy^U1o}sjW*0Sm%DI-Ycxat}85d^*utJO)t{b-f`*>6LvmD
z&Ol;o{e<QhH^)gh`YU%*HjN$o%M9#!vTQjQum1fv$V8s?PDW%7)Ek?{d_;!}%4iv=
zNuIS#mg@}V9(a4B<qZV!xJKY0&0LvSmkG}!<MZpRV9Ol!$0ggo4580Nqc4q!IXgX5
z)DO01d1bQZZN$c3pBC>}1?svRk@tUtdI8mgU>D##Jin$j`P<(=m9I`Pu`6n(GfZ@Y
zfS`dowLjaXn>{~~2<aI{#)=QI+aIF$kM_2#u@kwlv?C2HeE}jMUzNOac;~KwjjZbp
z-r;c)&{OUf)SQ8K@X@NOgcZm>dQL&(nzWPlwlATFQ?ql3_GvF$f}Nj3rJqwh2O<_r
z+N6u~rn5V=n$s0tA(d<keE746{yqrP1--`y4TXSs$wdW$n{(7ln#)@4Z!XIk`Vz_+
zvKG(f9v}3(soy&Z#bVl=z1@pg+HEH13BEiuO{(`u$EN9jpu9#*s;}0QU(UI>OSsl=
zBM<}OuI+YX778#cAui#xk_dqJ?*gDJRJyA66;`8`^}_^*5!Hj=?KqVzzuPrix4E+q
zOCsHbHj86j(Y;pRsrd*r{9oQpR_7@;LG4WFm0;{$!Nv%F`r;3cz9M8*n!xrMTIOX1
zE;;>Z3q`oNWo?m#Op2o|0Q2|*!&Nx#{q=q?!;3CWlGr(9T(~K@;;@bDx>ykT7XD%I
zo+)^XIEAe`3oskY<>rB9I^>Ft6P1c>Fb)3I=K|szP{?TDnF-mhc;$8Xc3<Sq213zY
z<>_0T7q%sHx=ZJ$!fQUb{=UheDky#Z*KR3cww}CFpzh$U0uypqlOF+s-tR+=H3?E#
z!Zf#ABTILX(m?$Nk$BfVVxyjEY);$dlE&-Z(Mqd{HBw|H)9bE*(t!0eby1t@N2JGK
zGI6g{W47kkJ(y<5UHS7l+TKBEEw_l_^?JlzOd>U(w<vfvSW*2^YH!<)E6q~M_$;za
z$a)Q1lIbEv{h?pZ5f0S-_d0(~O*M#`w;u}fdzfCmf;KGRmkn*MnKtZ=`02DA@i#K9
zTVCoc{I=#1!?dj$@BOgA?PV8}fMNgnEY_ne57lA&PA5WgYqq6=^J>jt@_6{lu#HDl
z{TsA<BHiU+EmTDl^*CD)G%=8Cn6}CeAfE5GDXq!2JyB+j4;d5QlHwl%?CcX&X~hq9
z`|U~*uyL6cC=#LTJR348vLo0<0~vi^?QdLJIJDBlu%y29YgW7yh}#?^@!a9S@7xJv
z)`xbM1^?KTI;xp@b`V1O8uP!ru?n|3rGr$WDF!2!@1Om+^XajXsU+{tZn=U+Ig=4%
zQm{LBr5Y~FL2tQzs|md2NBFsX;>WmeHMa;&_h0=GiqowS&~n%3Tf-}}hne?P`HPt4
zgAj2;5)1Un9iW{J&1P8DHJ?0~Jl*WVwQSN&8(U1LT=OwTy|SJc#hPJ~h^(O2&tefr
z7A4*Te4{(ye=1(Cb-r~eI`E4Fm!XOkT|F{I{Q$h)2OB3Pi{Hz~ZsF1^EupTZIyl1Q
z@adc~V?;WWf%<P`1lbN|ZnWuw0?Rn`$L;@k(#*X{cxj||H?Q18_xfq%h7%u@ZXH0b
zeN_+~<3~E~?Dw?^3{qOF|5dxP{}46im3@v1^dFg*$?woj7BaA|+_WTzm;|@ELDl)q
zLvw06Sh6zB>;0XzU0D;a&h71i_D}D~qU)*MP6YY{(yN@;Lh6O75w1v7?#>1io?(~>
zlUFMsOQemIWIPW=BCD9BxaK`z5#e_8%s}UhCw9RNph4RDM<;)`FklIM3O4p|pl`S+
z8_!`V+|AS6`XI08hFo4Y-^S%3A!O*t#=wVJm!5KnZfDmMJB~PSsCfi=;sb~$ac8Vk
zCRQ6g7GqzOSW5cb<7w(pvG0sl#D?#TO=dE5U*?YA&eDC!$+F6FkzBz-*Qi~^E%YBU
z?t-Lnm1*ruk^xX%>v}u4WBXqLj^RChLi_sE2j0>-Y9ZIAX~;Xa9uzO~8^gloHc`;F
zM#Co^Z161proHSQD}BO_6De|OkTLWujllt*;uPc6Xm7YH!Usoepa$!q{dy=UaQO3K
z#`spP5>WZ!5~R+(>=^-I^tc1m&}Ph@y3EQH%VUkz+L2y9@1geGVvI+)2t{T^jX*tI
zgNQs^f*3l`Gy!J0QE{yUIS$5;wEhB0kOp!`fCBcYuq4>aO>Hk)JrExxJ<h4oq|=xe
zuT~j#?`c=r6Z^!)W&{P$v*Ys!^bSu14C1AD<<us;QH!dW9?ko4P$(Y_uCEt|)UAUq
z7{B^+AY_+K83Gd40{@1JvCl*;`=z(q+1<k64I$HxH<<<O@a>#!tY(!og;Iwr5?-2)
z3YJ-iUNOh$o|iMB`5F3R<8jH1U-#bskziR7hwtPrxO-s<cU23+^tuCT`=JgI-YS{M
z;aI6)JE0EAFF0D|fF{&X9cGBLm0-mo5Jf;4x|=m%M!-5sA*ce1bSXC@X<dl7iP9<N
zw^?e#&YYUN_?U)6a@9Jyz~(O{j2?#;IJon_#%fwx=;Zn^H=DiCri=*l#MsAr9(ZfX
z?>!V?H&lDIY(>HPYZcOwETO<^?YD(%Nn6WGf^ArKjY)_%G~h=%tG4cy4GS(n9d;G!
zu<(Pd<y%8;sW~I|OJ5@fF`u9`+Jw0=Orvrep6}g)PcW4cz%g7womCR@z=cv8Lo2~4
zj)TY>4F{Expb^WxR{|0}FBfu#D(F&@nwRWruU>=|S+d3+vaA@n6Swa~y>Q{~Zqn7J
z(^hVrWL6m)lnICGp_$Q)k3gU8heEX|<tRk2`N+n_R=g${W5=(W(0>Y!LL|CX)PJB4
zy{7(=%}cWPLzS*C(2>G`UIwy%+s_V&K>n|$vd{4R*IZ1R*iW)uBe<wacX0$KoL}0a
z-t0jL)$9zSbbgKW@ru7!W~n`KS!yt+)15Tr$}&?RNI`{-+HZM-hIHGwX5<f(OiQJd
zWIb(7h?%(#023!loXL}3&Sag_R~6PeBwZ{~idP8H-#9sdo0%X##8rxGh&85gq}_g>
zMJm12O=y#TgSm7=hZvz8k^1*mk)p}MdtwU(qBC@mkP&Y692?&wOxn-_U)2w18=}F`
z>oQU<?b<&R=^7C=r%f5lB*SUWIp_OONS;e4^f&?x4(nW?e|cu)Ux<qdt%z6!<Kk8a
zwU3Z{9v8Z_zRvA_Nf-`>su28S9(4>LZ4(%$P*pTJnPWI80g-;Vo?Oul?$18d3>D}m
z>_u6uRQ?5UBGZC-Q{QEZYiIVi^BnMiRh2N8tW1w%DqCbg4_!GGWO;XFWD^`*dk*~z
zU`ORsj)Su#ThC#@*WTvw`U?I)Lo1@;FCiJe#$W{JDM;Oos~vl-Ql)MA=L$$l2JW<K
zmx!oODe%dvqMPI>?_`+LiS)1Iv7X4aV>fa`n|qC5@B@jvilyi>xP4iDO*Ou)BpwQk
zI`4W%hWYT~wv$MY0JrP0!Ce=t@yrsT^bhch-~OswI-t$6lR_svZ8*?*0MAl?8<7XX
z_GBzkvm*#m9MY>SSR;^;l-4jMQA*evH1L}@io-2BGyXDJZ#cZutE`yMq{LY=h<t6l
z3ioqzFU!keBQ%!b@iZMbkqF{v-wC#5rKuZw!?(8+O=SM~v@d+WTTViECk?}b?kg5&
zljj;bcx@#xX6I)gg>V1r<3tgxvBAg($hy7N4#b5HYDeseN=?sK)<_wqkE=^U?C6nb
z4*CLWBQZDdofvoFilFV*5i}{&WyzA&4oiQ(s{Dpw9=|p}<(kI~ywP2s3YF@T$GbS9
zchzLs1U`M8Lhz;rE`g6&j*);_5~Z-Pu;4kg0z%!YbfAuF!Tt4XCYr#vrbnez{3=_y
zb*U-LEnA1FO%xB&IAtupld!44BRbSv{TwK&Bane$+Zu<37Yx-7-fjOcr-JeA9AXh0
z!6G;p2)iZzVQ?C;&I4K{t|7`qCTr|cuStCU&nt(mSo3ooVK}T8MWJ8YrpsGXIFf52
zMp(~)ynHN0dL)fr<!ETvtjM(X@jQa>4CB%CLGp}1%4i?DZe^yGMP#fMdxHtR&!9(;
zoir4$Y>|KLx3}1fuuwz~_|Ph*Ek;H@zy}%vE0UpPPMOszJ70wY?)Plb<h{b`I`~v)
z>qpwpsD5mDgXq!Tbc#K#iS;@^KczXFhH%JzReV=-9R`jPGT}Ivxh#03x}JEqTyy<q
zuy#6@QktZ2OoDu!AoWW1{60GSQ8Qh0;N4<Ex%r&4AFI+h`TpP>s2On_B(LQ@Q30#m
zB}_8Nr~GmHk_2w}b#kZ-I$#QQXeLCk+dlT33-2fuY2lAFP0Vq26VUuAD(0+eiAi7{
zKFgs94&kUsrY%QiaD`Dx6<W{$*T7sIx(rYeQYG;lG#uXzh46Cnx{IZTA=M$U1`Lej
zba7-@4~Kao8uK)Zy6|QlSu7s>wJn%6|91nwg(@}EOObTweJI@AM!i4YeyNGqrXNZE
zHgzA_X3ZHYaiib-Yirds$0~{4ECx5urB$0?it39i&3KK7YT{E6^O0)`)zU~Q3Vv4a
zTkt-WW|xG#<Vc)&*{6w4{)S3xmt_(6yG5jtiK*jjK2<^5TV)#i5b*_Jj#yEli|+`y
zSXW96!?x5|fuS!*oJHc4|H-IP$oKKNtF*sj?bfB;8`3CniqcTguKLuGlJj#j&k|<K
zHCk?Dr1#y*hLcd_cujBIi%M8)bdQjG0`N>Q!8Moho1l*P|7m7NPTljrq~Fc=x>l}O
zl4a|S1sL|yau^fga;fG2)c$u%YOcDdL-~L$!|N$!a29r%u_xJH6-;@m{&erRX8u8;
z3@U1?-3l=IS>zxntt>2Y3Pb!;@zQ2xtUn!0Gf&I$DoP;{`qkM|F@QP#u<a7ODU0*P
zgH6^iH-)LyUAgG*kla9X-XO5Rd)WQ@W@mYlVFLDYN8m_0c+9<VEVEI#dvg}R#*<04
z!tfh%*$UF}jk?zO>|3vL7_wmn89%FP#dnn|jFSl2WzMQ~V3sT$sda3Uv#R4c7uUjP
z3&we4RAS9Ea(V#@_lGX2vkUuZ4ssO$kWM3|)`wNq*mXddpabZU9Jv-nF95kKLT&1q
z3P~|SL?7oB+m50W3euA%#M7sB1n?f>Vb(>_4CP4p1f?5!<?WovX<$Z^Cmj*Lr_l$d
z#YNiJR*!p?zcK2yIZKU=AqdYP3E0zd;_-hH%IhEiWmkBVugEcvjj3l81S9EL*oKL>
zax^OUR!K;4Go9c}t5~pSFB;W==(B;m>|=UsxM+nJfF)8JsQUX8IIS!5Hhnn?nF>_#
zsnzXfEPj%$xs6c{dlU2rZo+yb4N7RMgitk!TuJj1xZGU`S0kJ+h0GZtPT_K$fR~zm
zi)U~wHoDRK*|bLL{8A=JZeE2%xOPeJIWW`R@fbGA&&jU&h<EksudXU;?Ax38ON<xh
z24A;w$JC@eiO4$|RcOQSzjgO;USsVhv8kmk{isw)Ar0*o$g0rc1Qhtk)DNMa`K^c+
z#`a&S`3A#u&1CJEGetnFF0o>`G*orDYh}6S*7_E;xUR*95LgG39*!YN;~(&-&{=&R
zyPKqW9XT<hSgBU?`fC9&izv~|;6E`byX4$=XG?#=>>+KGAJXK8(yVP<)(xO6M3kII
z=c*0F3H+NwDGj$+QIIRo&q)YgVzVWS@QoYWc}x5}-O}2g2gZ7Opn$gO=o^4>VUTu0
zztFso6x&Z;y6~m3!?Gxre+jGSV~MR%qUN|0Npyr&$tIqK>TCV&oltc3i)W_rv5Nh+
z9&zjaEbliHA*a_=uteWTt-#m@!sgA2pF8YquSoUsOYiVpTN)>;NMbY8jESYhq79^?
zsuFH}3c2w1?C8I*qwgjzs^|V5eLVqZH)MFidgcPP=NzJyrv7~OJWNfIf;ykViGYIi
zf<ks-12;OVH%YIQ;lqYpI`g(k>^pY;N`;~P#E<W4pIHtW*d4?mPoN!qXpP2$-|82#
zCmq20Z4a`uE8aG-uRS%>ga`i#Yz+UDvUp8M6S?y#?lL>bUz8pEJqR?G7T*|!DVwk#
zLH{2T5|UhzI6=Dg{3>gn2-%51_<M@#&BvL_ZD}pSxF2>RZWSymrEzj6=&jZ>OkN+b
zoo79@oEfHGxwsbMPv0I6i|?63yw7>=+isvT{-PEJw!l;BPm!i`I@m{uqA$gx^xgh!
zU<*F~UOh8CBjwG8idZFT&4m_L2&cp!8jAfqN2y8w1ZG3CMC*Q^ieg-qwu1Kfw>CD`
z%aNSnp*^1vH`WU-(BrNu{|<8_16ds>xn4}7jd;$UZgmpyLfYB7A4NQ!74F8TLJ_8y
zUbSHVG*is2AAOxM?_i%F{s4zoAs5TH;3qYU98}xIyn=T$CNu*-0BbaQI@4vnKY+K4
z%(!?Api$$n4PjRrE@bI?)m`m#Wre#{tThJpzZAw)uoh>JVJ<`_m^<@j%Yn}&__|@I
zV4gb`$||{Z$wgUdq@JUj<HAyVWimYPE}Ju5#eCE-r~g4o!^1of(xUk~(6wj&>ay?$
z()b3ncV+yS={eZl49HEF42IBG*c4aIV|CArSvbTVgpid(eBON$?>*KMZIS_S1Mp8F
zNwii;7XTSBXWp3#Q%f^K>)nyaS7SyvQ?{ai@{78t@pwX1>0pw%#l68E#PnO8udcR^
zi~a>YX>h*@j^@Zv@%n<@(HTAN8R3mO^@aE3QrwzrI4BO*WnS{JkdTS0TRY+0E!*cm
z3_i}@D}E4~Im!5!emI&Z_)ma2V9b{c0(@YQqeXY?jrcY%=OM|(z5+-N{bbe4OJTU{
zYznyT;4Zpd({_RLYr09tV+%`qzUekn`)1FwzgU}`7mn|yUcy+Fg+w%vyPlW<e4IQ^
z*{dcR_{sJmAy2bARq6^Jr;zw}o|PnXCc2I5U>xwMaCBX7YR98lm#<Y-K8*1xVs4&)
z;W`1$NbR0s6<Zq<u2x$66>oXsEidsLp{;@|dO}^K`!_hZF8UqJQax0*Ecj;cgt5}P
zY-z>{^_V)TAKu8Cg3mX8H6twb+4-uOX<>97OzR7wuGo>YvuH6ZT^a}*j9q3X(KmKa
z8Xfv$1=R-}O!xA@^2YJtA3ge^AuEF^APNZuo0Q1n*BSB=Z7>VW=&LToKI-{iS<ZPL
zloPucsxy;6tYX>puyq5xaFXaZ47~LIUxR1w!6)rf-Sembx;wF3`?Mz30Uq**964fH
zWT5vRq{*KoT$4|_GE06GfR^cYL`U)!XrsL&mXp;vdul~kSaNrxa6CfpP|R*1+N2?l
zq-9bP{2~vUfhUezv@Sij4Um-y-wttOZKB0-AsdGarw=dst`qL?vtHQ4mMgF(fSBC$
zfjbLXyn4jS1Zg(9+R(U7gt$W-H3j|D>G4LfZ0re8tja=?n7}lT%PtxHq0;*%F;#)l
z6nWRrfv+NJ<*UPh6L9XS3m)C~W+|pX6z_#v#G(WZLB+=l#xk5r!?&X$VI16$-ergu
z#x57F@iSmroibF@QWEAdLiQ2QAnE0AAH2JLJZ4TmOh4VDj1`dly55SRHSAAt8PCCB
zXij8LsHeWE1=awNUKG&qSo%sl0|AAv(H-z;gUpBgAdjHULf|LVHgUEFH8d7Cp*+$b
z=blHTZG_Cda+eXVu@BOM{kWyUd-2d!<?ZsX^Q^*Ez#}09ab=Dw3eiLBEu)|(8UGT3
ziQ&+t_?{N7sp^*7H3Eva5(h_mMLiz@7djCQt|DC|ap!K;y#)>#<JGtiznuAxhoO_s
zX9IWmSAutd_;@E!@+`GqH@J~vj{m|}s(FOM)1%2zSDQTu0W2F@hjsogN{ApYtEV0<
zG@;T@wd*Ph_0_&BDA%-8cfRbcq7RBNfDK?!eIK<JpcYi}e@eo1k2GbBIa{OUb>I+K
z2bLk>-xSaF*gMlDee(o9%Ks>>744YJ89UD~x6`+$U1p8#AWX6vi6Mfdy@ek@1(X>f
zS9ED!_i`Ij#ns5W*%xR=Ma+si$!Q>P&KJ+UV+A$SOiQ%S#1(DTy9Jlzho;j_eu$iL
zW5a9>1g;37cSGG+g4HYVVvlRq@oEph4*xVW$}h@LT!?d^<NU(x;22igNh0Q3`s>`i
zC(Ovp0?#Z|g|TIPgcrf{_g{A)b074@?cYV>uB5Iqrm<FVewOQ3daY3H`se5(;ip^5
z3j;iJ2xF+hfZjCzV187aU>&H}W+6g*?giF8K1c@+><m|dK|p#%Q?^=-5o;o?Sta=*
z197wkGi5a`DfS7jB$~cGNtRWX6|;slF31uO`xNRos6M_jHccrvnUg;j6#|5+$z5cd
zY$TpM0Hv>B`HJcVLYcMV8uwjDkiNW7CD&^DiMjJ67&Ft3Bcb36Z*ZUjmsoMcd$c%V
z_L1cxmB-}g@_;WOvb!mhqluJ8mEweX+o3^V{16ya#Txa%trt?ZxpZ7@P}`Qh0E7c`
z_V(#>dVX?I?kaynIDZOKW!|ba|64@6W4V8n5c~^v2uW&L?c>k89%aQ`gg?fp__pUT
z!}Jrk$3s4ii-0IU(P6A+i-H3@ms|(w8>~M&_a22c{Y&!_cs4l2F~9FTlOuWQqC<td
z{aC=N<9ygJ)J<DmRm~W;PzpT>ZQ`oj(eRH2E)>am3)5*0%<u#ds77p$Xqwn<@U)%F
z@E`;tnNVE=70U@a%RYLAK%N2vXUCclz@Mxo4uiRm^B}w8M%D7?!-D=5YOMn4v{6QH
z!OwV;@x9`qtjSi^N6R&WsRP0fchWg!S-6$aK$;Vo-{eZaKPtp9uamE!Uq0UBCs4^C
z<R0Dlr5`jO$P>wLY6^FR0GBZ|Ypn#@pLw~9Q@sLkQ_P>iicZmCo|jToK}qxr)|Ng@
zdbZ4KaBT6q{NEmWIN`HynPjr(NzH>$MfX+=gxFEJ8(_*PSADe|{}&GSU(LIfL$-B-
zi39DKS4*a^D+dnOb9}FIq#+=c!mU^VlTH|mLaP63;7!$&jU=VQ$jwDJPTSd5pcl=S
z#EC0YV)<ox?;oM@q6pFG4X6<2RwI-))@M-om;`Jres!Y7!Cp}h2j@kJQ{an(C#+)J
zTALq-EMfag{ggQVo>cyo#~CJZG*p?anD%lv&rWZcAz(H;pB^R=%K8sbV;2Em)7kPb
zG5)$k07AOmXD=SXdidnI&Ze39-5hg!f$c_9#aofA{IKLl=};*_lTrzDfJh!Qu@t-e
zhf9IfWyb}Qs1%j#U8b08v|>$^L+o7gmG07UOS^4dyj&Fu`_<(03j0`la&mgML_216
zRYG!<<xQUqv73TT$gG-Br*Jx3JmJ2JvM+gqlAr;)lEUetx$P_;IAr|cz(w)krt1;6
zec%J%h$A|x6$n%hx&nZ@MHs$ZHPhLGuXnwj-!~sKOYZZqKDTh-5%eA$X<1nsWWY-Y
zCxUlbOr_?GRiX*0&WXifPrHI|L_cll*Rxa0z{{fxn?yVtXszqi^T7&fg~nT$ckh5d
z2_H#pOs-Yn+vi<T$UQjz^>Ys@CnPABJ*IoAZ8vLv(Zd!0w&P&^OPOf|_eq4OWWtoz
z({gC%`i*587^DBSZg2B$yqy;hguxiMxf;EA0a&7ir><agq}ojwqnSOoKcU(9kVH4j
zWMiFF3)|m_BSQm{&iB2Ovgj3IO?X0Ig$yiL%sdf=#P|&6wmllVkPc18i_Z~(k|D0>
zV4>pZ0pue=vGSzwLCLL5KMNd&;qw%XaSV5gte$R{OlD-y@G7!0N|6GEJv`CfjGri<
zXj=Wwt7B!WWL0LY#@m7+P#(3`{y4ZB<}twQ5J@;co1ppAkG;IEsI-Iu_<@&VyG61p
zR71qKr84bh&+yoDr^Ak)lseRAj2lp~!;OZjIqX=*j>Ku`D2phH`lAN~7qc{#g9)9j
zM?Sybiu)k({kj|AYaJ8Mr1nC;v){YD=q`vB93Q8QMed2+<j~=}mxKL>_&C#obuTP=
zWLm9AidXR2A;pcj)^#!CQT0JC2EVqa4?Z`&KPsn#utEK*Y3?o$rURwP_rE4JtJGGL
z+_dam$ZY!~HP4>dHGUYI+Dph{+0$<3@ZR#B+H?fG=$t}|Ao!<RU(DGOiN(bIA%P<R
z8N8!FB$Ixc(15OU>Qe$3|1mY4x1!xMF23V06c~A@;p7?{o7FNeHV}j>X;x92zkGQu
z7xhPuOJ>HL0wQS{^;}9u6@RE|h|PZHB#e@zcg(zQoN4kboHs2Qwdq4@pD)skXya4p
zsl(iFt2#)qL4yrxbqm<2LNb0I2Jl6}qSQUVxI|AOW$M$u$&|iSp1Wk5pfhn43J_`U
z#xTNuPZBZEVcGimi9GP}Ig(<{M%Iu!cUH)*fCZ&L6lw`sC4|^cN7yS7vhhSjx>fUY
zy`t^t;YuWP`fr0qt*KV=s{^!Pw)UsSY-Z(mKvYyOZ0<3{2D1lcKT_#0@S1HKO=}3P
z(nS_Xnu+x?G_<>ucJVH1<o@J4x2Xtt;Sm(07BtcODO-_l7#{4qIUMck`oNpJr*D8W
zs;6|u)e(6eoJIiR`SQ6t(F~P4&)-T#2ii{Am+W}tmd2jmf}g_|#$3^`v2~R8>Zz31
zBAJsy(uAGrk-ZbvmPUdf&8@7grayD7Y0D{5W#N@gAD9L3Xb=m|iFk()t()0y+weD<
z1hq|ot1vvFd@Ou)BPVhkF?x@qd65xOXnYRIZs{`TJ%RVtbHh<>0e!X}c(YkuSL&IT
zi|hX4^HR@#;Ah)hpGAIbsGJ{sNM6zXaU~i|(@&t&zG)e7OG1{08QRP<cmW#+1&d*a
zLpnO?Hj9(LsktiIHUB(F5y)ki2|b6CTiJm;X*leW{HAzCV0y)!4hVX`!P2KfSN16q
zzm<%;fb@z#RFf<V(Oh`g2CyNB=CjguoMcJx3X<@o!>Q?m)~Ng&uSdb?%zjA_jh>H#
z4LWLW`Zi_l&<=jY_gV2pAtm*^DT{?@dh%0Ybi1M+d+AT93Nhga1C-Zi^XieY%#=8z
z)wbR^x@xkD#ao3QB;ziQYoF!a&E;5~fin6zSZ;yH7_>?_grl6hO5BKhcy!Kq9W>`M
z5j5YX4nnPL;t*t{T78z91p5kmoHA(_8f2px32<gz3f_e!wtfZH;|}e<Mg!qDvto`Y
z&;4GUY<m82Gk2G-4)W82W+Ja;(;JxDx>F*4%8~4X?`43rrAC!rLCO#K`xa|D7Z<Py
z+3IK}4@vmIek7RdeDcHR$KMwF3V63|w}X%J=npwVM*%g{*q6ce1T+no@SpGvHiFL$
z{(`zvPmlx2rVNe>+5z+PYNZ`86X^C>LE(I$U{nO6q)JJsHhpTOe)epJR8hnq3b`|~
z-cp4t4p2UV>v9UuQ(}3i=8WxT{K@I&AC*fBg4~}}K$kpy+ZvRvkAE>KeOMB1MfKrS
zXPAA+=8S@UA8#V{U_?5o27LylV5KsyeTR5+7;&D-fRQaMQX4kHJuJRfJ)zRuhrkDG
zw##6xJ5DSV5;fjk#;rY`(6R_eSxVv>E7AZjyXRP@1XoCQcrLlX{6}j*8i*(?i2^aS
z?}g>3l-q#w=W-^qkdhgB-{!<536|gP`8YtPJq`&^ZR%M>t8msKeOJ{S%@v5<W>#bS
zonRKVYHDlPVre!!J{_5l?<RHR`PcvXL&!|~-(Z+-hn!JM<%wJ*E~@A4O#L4vz4dgH
zxxplS+cUD3fXePxZ`I1Zwwnc$i3-WC^w0<(O+ogyKq09<aa;fqW-~}bl^<VnX$;Sr
z4mh8JX40m`1>4RGQoX4BUN{|TJ5)AL0#E%1BjzznO-s40T5}F;kg|H$Hvq8*M~j4i
zFhu)_dK0ct_U>dX8kQMBd68J5+%rI4s(4T$(k*g=jIvi{tsG2)QA=e_4CC(KaIlx=
zH35XRJ(ar|>rAMAoI$27$AR#crIpUYnkyzfy#>ULkJI4!{|=jqqegivoeRdc-8LM`
zE=~4v0?S`(7+tAxV$q8=cRu&e^6h{d0K@Li7Uw1COjdCl^KiF(W`&{RR)wkic<J03
zY2;}87&-k|Pe;i3#_X=CD)gmm@e6L|+HwCwve|AKa5*6%fifLYl^#LL#K(s=EV}M>
z4JMVN$M_KO!mjjrFgHb8-oSYRXPTh14a$IBoicpqjYdN$5o>ONSp$O|P>2e8O2D`4
z6*7d=JMvI+c-?J(3Gy{xs7S2YVR+o818NR_K}_Ux&_=w_XnIb`A)4H3ARpoy=D$^L
zI|*qt`!+~Z!SR9y8-MOH#7~=bT<Xj>;L9oQ8bN{HO%`~zJsh*Tu5A2KHFCXW;3l;i
zUo(z*fIxQHUU*^8%z~H^Mt7+f(<TDGQN$0PAsmIJNcYoF{SytUln?|1saz~uw@Qho
z(2w+!zkXg@Gmggjd+YI;_sP#0ro8M%7HQ6s!K0|JYvfAxjIFyBNoW;u@-_-OS{AUB
z={S(2UAx)QmMSB=22aza1_P0f=I(C7@U@}qUKg6<oPy%dyjBGD{xmuO3Nlwn54Fp}
zlfOw#Wo!!k&i4Wu%+S+AUaDQs8*zs*SV3VugEu4oa`T+hE6T;i6c^u9v&1|H+q8`C
ztj)T2X6cj)1AgH0p%EDCy;cig)fQI1&b3EY04??JrE<-%&P!Zr%B#hhUZ;DGX$4-9
z47TNuec^??v1`eS4pSbq*~Xb1cA4qAGf~J<W)Nzu><CA|Z(t8<E?{5P&*dV($euhw
zxCJ;H11nIdiQQBX5rjgf!qpjxQ%Yk7$vFxNKtxENpk(RxA_p2KrJ(UP#S-oKf{SG@
z%B0F*!^|NZLeBv0ro~S#(yI<Pm=0UA%AVnLNomH1^gb3YPZr*~-gAs#6%NegS?75*
z@}eZ#V#S$MWnTw!#k5^)YMT@FNq#f-wxrudXybS{iPP<*NG0^b0Mt8bp6l0qnapTq
z_e4NH0L|+71MG1Mr}aTlhZ{Xe5R=4Oa4rYMfXfD@yw8pTUIur=-D-}M(InExoV7{1
zXoVU0w(@RhdUc@V0@B;VJQRWdrnEM)cQmrkx;Puy2>h2N&Ud@2P%%3sH)?@@Koq<N
zK37#RHsT=4V95_j!YKrsT&d9NS(z2A%}OgpAJm-5!=131W%&=dvhnHbMl)9M4#9GW
zs>ip(SN6Jdmots@_#UHn4tyHv;HYjftJ440_)$<#-anNZ_8s<UwRadJNOriVJ<6f6
zspRv4T$9AVMwgC|{N7ZC9M6Q@bG`GcujDKj$Ox^{N4BRzQMI&|<;?>PDAo7BwTr`&
zl#R0py5MwP^QxJ0q^np<$Qk-_2R9cgq&V*qn9JqRK}!U(dg*u2gW%bX7wlea&B6LK
zz2<F``|2lx4gr-8%mxQciA>N{xSDetIP~-40SQe~Ck5mM8Z|r75Aq0=CQEk_e@$|2
zsK0-O1}xfj_HN0Vp2E9&#})0;EWURRzaPU6E=v6?pAl4JrhV5mb6%Uyzls~vs3_H_
zuSik<Zz=u<KlYW!#*4A|_@IIm9&)+{^F(QGNl%cd&PEJHLYngbAZa8(hYHa)HZ0JZ
zb?>N~x2kj0-4~l>sV*wN%Ax8nWz9neh4eHt7EbZ)-z}iM>VAP`Mj)lh$hz3nFcyTB
z1qrKw^&WmZ1Zspch;9m;irDyt66F3MT5*$t8`CEnvNfo>Mf8t4Gt@=b*X!Oehq5s*
z?{h@P$6jV3KLeId;Ga+#OC~PyXS7~1aA;rRH+zm?3twf1OJ9+`yo#!0_Gy6&*tu>2
zjkE2)P@FW<05pAE^`ukD`5MRQx|%;>{Kf17!>Knwu&~i?N@C4;au8@mZivV_iOHr@
zWbKl<YaECcd3zJTb-oPyv%ofQ5e#`(dXv;z!L9--`pHlA%6H8QN7X;P#<<e)SccZt
zcZm5h(>Z9n0_w-*#gnvTqpHg+ELb<A#QK2H=HA#ei}DSYi||Mv8Q}3^w02Xa3I|1B
zBv2@3<(xl`xMYT@emP+Vuq<q?5rsI#ier9v+1L+%(tCO>j*7pq-i7?6XJBG?+oU^X
zuEJq<7XN~(D%8W6uQjyiwRr(tS&6arqrD!vB>~_nA>V%ByM(+wI;kOGYr^v#WPYlD
zM7IEWw|s8}1p*bYMx*12XS<R^-*ErC$vOI8nYiBvKC-6M1B?1-i+UP`&}PP(e<Zl8
zHbhWfR-(o?W5uMf6wAKXW{gB3rW<65k8eODcwhd=ruLJ#;lGG1;m6TsCv0k-kZJBz
z?1>*M_#7)edtFtuKPAhm3kJeG(!K@mHE9l^E>W)Q`r8U3B?xy=zOQW{6(tQIFOuES
zkzGz>dSP3J>+xeXvjIvcm{D7g=0OPkN%D-Z_~#1L0g0VNjlUSv_WqQCIs#zVLZA=<
zzCk{EC-2UgqUkDJqdU<8;A|W_x@#zZmDc$8cyY5i;cLdX7I?Lzj9+=ds5ON@@z*{T
z%wPu~*ZX%s?%c%)uJsbap4=3moEu9&2QRyXmY2$%y5`hWles#qKv``FqC)pIh(}X6
zKTrg*^_i<#il>t$<;wj@a>4d#@sqti#b3t^53cJqHzxU6ev1yUbbJdVG>up@022(C
z$ELS~0JL;+rfYo2eV1^Z#@0jEAO~N*%NyFDWFU(Dsmp%o0&Bb_o8T~{bq@*jwT2Pt
z&7|}e+?mD`5T!Ntpt|F#KNiAG(}p`oL%C>{jI;b3-+3H9Tk}iA{|+RcXKN-VO@nP9
zNEwZ(I<hZty<k6xc3lGHMELL#%f@m1`|fu!Pe@W=IKY?WM;<K#wQu99ZrRYPv}j|o
z#R{yZdfOhAj=xRG;oVvDG>Kkx2{L;rn-<>!b!Xu!#cEhBM77mcRt)y`R<^kl6T@1t
zG4^U!%}DOL0ol^F-4OlPiH;9({7X+FS(5WEmYDRon}}iE*$mlwZ19KZH=$i6>g!}x
zE>aEGo@2s{+w1X1&KAEn`LbGxMC@q7B@>SmIMQkuF@kTUr?*rj7J#Prv@j2ET9#Gr
zd52C`F@LB5=abklOm#O_^QTB?Fh@q}L!3t_Gh<CP+;uU9fyOu=TM7%b;OeQy(IX9H
zyJt)(kb>6#taX)te+Lod*V&U~u)Dl-Ztnn+#J}&N%>euW>VNv_?Lf~SH}u0U5{ztQ
zAiLmqd@0LsZJM>F59aL#@38hngX44`s;xo)5AAkn%N0=Jacb7E&8an?9BoM)2X~0A
z-Ms@)gXrzPM#MS}kc*IqE!r@P8T>XqhaLf4-mP;(sthlPt?`cF4eFASUGpjer-Fhc
zVF89R>7vZ}>c+APLvcj0k*phr*WiIiiO(*|dJzgS3#c?p=aJ9AV$|SZMI}Ciy&GUv
zNBI~M`0vrLX4xd*GU(IViBmAoO3*?71_+0fot@;uVXmn3;!AXreoY#0grYkkdTP7$
zQ-~bq=f>p3abkT7S=($_;w{@Vo*Tikxx+hg)*oyQT+K7|-z>v2Z~p2jF%b8h?P8d2
zEoH<!#fHqZXrIAnDQL3mt!Wf2y9!Oj=)a<&P$0g35-ot~wbci6$EE<g(Fbj*C=Il0
zL}dTiKbD-^B$zUuJx?ZM6%BBn{PRj=8(`fKNpvveO80KqchDor$R2W~2T^hDOq6Yf
z)G&V;cgvjZN(L8Xu~LU2&w%Q}WWB0@7O{8%MrJ6W9(1S|3Z@*kCX^hoU%B2}&Eog3
z4Mj5#>M8F_N*fi1u32&ALE8*mtb@DJ2F!37Jq?Zjxl^Fgg-1o7D60;b7Ji6F#fJ{N
zq`)Ig4pG3t%gut!7X)%Nh10KA_TEWDHlS0Hmk*d*uQmhx#sen3^aaFd5^=-)O;}wi
zu9$|wm<)*PfM~;hi?=W;&<Jt;9_3UF{I4+kowoaXK=C<PMEKL#ei@bJXTc_cVTyU?
zBGii@>=_nU));#NsZ>54(!nwJK%P@Spt9SGzouXU{;$(xo3UXBvS(xNZ*UrZX`8QX
z&tv}3a}3asGXh$|P!Mvn{d;s%C)5nj;Imt{uVEV%ED7bGld@F+^3AoD)JrNmYDbYZ
zxtwjAuAKQafgUMzG?j+$6}J}ok?bs$!Yx3G@rp7d5c4Z|XBD_Hks!Ss1JN&DHdNOC
zzvP)Ot*6kED-0iPfUcglm%duOjj1P2s{gNEoQXB8T+5EO7KGn`AZ7_S;%o-4Rz_>$
ze?4tq>*U5%F8l5nI8g)n4gM_7Q3kWd`^fZpLIdkAC=t0WJ2!RpD}10<Etg6*?;;W7
zm8gUkGf+x~hF`$-upop6r@s&%Pynn$X3y%hL|2NUP1Skt##L3;**&c;b5lFjQ6;gD
zE&ez*G>k??|LDa2kE=HignIw}$6vOgG?q4nC@tD7$(9)0NQIWGNDNa^)C?+1mSJ#j
zw4hQH*;=$<3~6XExGhM9#9(YU35}(UA^Z0{$G!LS{{3Em+}y&vUeDKgp7S`5^El^u
zp=wP^gDsRy_YOO4TCIXr!7O+Peg+jmWZ^ld1lKh0;8Dm@%Kc$5;>W+y`13Nb^sV&S
z^#RMwWajsDDSHxJ+4V|Si9oYQ^HsPH0tMv70w4J`AtorX%%yfbWMYt2CV)De5i<vP
zK3SI%nf<dDAj#XjkMC-|v-0nIibt-0G=G%aKfMb&$*|SAWiKu!>ZF_fpu8wLh-+s4
zUn3F}nGoG)VF7Iok+HkQX;F5qeo;L8(7mk}HK}#>igJBMN<+D~%yK?;#DTQ`=vM#9
zeSjYqLGM+^{Q`kG2c){d)@~APdDvC=45ehz=RcM;rj%FD+zjpXB<*RRU6skH(8Vli
zIwip3!f|VV6Xv&T?rH2`z7TSk!SH@IRFJyU75+Psj%0vzB3%=wx(vB_h~>BcouoCd
zRS|;H0vVgxpEZIV{F>9ybL@3eNL&NhfV*LY{Zl`80NT{?G!X)Z^@Z%Qk3ia*&UW(%
zyX9lcwf0bJAI<9v&2G-lAjOLn`CG*8P!&-u-8X7{P97S!%wnnaP_jVH8J}=V#Vmlx
z)nfj<uoEy>@$53}*%pZp2|9P4+`?=|_KB@+;1?g(pX{35=a{>BMzFA15<)kq98bfU
zhZl=~pFFmY2eykn+XePy0#dd>;9%EEsH3xnERJ@*v+epC3f?CNGzjO33l3fifVNjS
zGL>MpuAjO^a83-Hc!UwcgSodw<MGa?R3SDGJvacP^Zz@wqFf*8;~Jj}73W|lxykJ5
z=Ca@?k^hb;Krc$@;oe*{$>ldfgl$rt`#AfHJ)cMm0;5qWPh{Wqk9N1!QcN}7KWa{`
zNr?*DZr1X04MsDD!FcxwPbw%6du=yWY<i?pg&PCi(;FCc0;4Bo6@Psarf~DRlQ_uN
z9yu4b-s#@(sVhgg^=haOd!e(gsPq%=U1I*M<iLY@u^5UD(1#$LMTxeb{kzF@9vl_Y
ztSVco1q)*T5~kXWJG?Wphm)c0I#CyoC0R721s@)7LB<V#o(a4zddu$M)`91~M90ap
zsLzEd9km!Oq(fQmgTIiNEL&0ZJ`|mTi<Bbw*`Ihj4zict`Ug1(II3mN!|KG;Ub@W1
zJy)St*wd*8D^=b!Z*0lwU*}bp0`|Mx)Z3+KbPS{ih~k<CiVIu67Ys>a;{oBIn=V`L
zRBM~w9_wUTJa&-vYHBXJeIz_D=@WMInYDt7=&kVY`^*8J^gV}-hlt{TW>QZr#0u9M
z-Zk2%RQt%;p{xdpk&YIIb2kolT5lSDRZf7()aqyoW|q4)^j%2rd)cvFP6d=+Q7{#<
zBl8xa5d8nOi{n8}s4yXQu;ccqp_A*n6HCmFwPLAU;bcYI>pBAuMyIs|LE6;RAAus+
zGiVP|T2#6UVa;D5)b_DHsLF(w%vW-G;Z~xzRNLm#O6MHCm5N6$4qN_CZ|SeB3Kn~u
z9(Y3fTJhTjz7m^$Tw79UG4ZpFmYa^%@}6R`JG-w}FOzHRF4*;??e0~-j}pocQVGvD
zUk;J%I`iykg2v@Aef?iEHXiSEUaXE^J^r?X=djdrTi6NKLeLiy59;KeO`lou&!|JP
zgJVL^309If{*`vgMd#t(WB2aRWo+dw4}G5U_nTRadDSr4jpFLBaEF#!s%6a}f}=TZ
zZG%KxLXOf|?nh?VQv7}<g(ll2_xjxQ?{3S|OElPX+{@8>mYU4P^q&;Ry1Z5dq<<&~
z?}g$|UjMP7-+Ip$n0LV&zDr-=VdSJJw?(l-)PtVCb*jd6ma1x6J3doA5?5=hx)Z}Z
z)SM#X?-rsqU>^r2u~!tEzD4Kf^@M~r1a2jUJsL5twdHhdXvKbuJRhDlDtOHAX5Mr4
zvaa0{bdR<Frr!~d=eo9bHyT+Noji9zHG-6i8|?{9bq%GklE%&{SN8uQFdXbFOLPr6
z#oC;{GNf~SXKq6GlypNQG@<nOGIVVlY9woDUvq67B;Xz})x_RF!W>J+IjJ^mb{ysc
z969o$%4_ad@t43<&S04Fidw#LE}xfd;_|pGNs?VXE?*^ym3u+k>fL)=d!1D2#JSl=
z{om_!*~3%q4g#aMLr(Bf3`aTf@aR1&F4<ojYZ^~%ZW5lmRX#NB?V}?mLH>(*iN*B(
zjH}yi`SZim=sl@rleI+S(3|TNzv}2wvq4A{L+Q3J)jrL=xZsq>$-R{Kk7C}Ew?oBI
zO*~4>AR{I^IhM1H_HE0RFTN9%3dPeqVlR{}mj5}^Ih)*n5?-jSo0FWlx$x)(42!B}
z&gin}-!q%UoGCjyGLOk-;^?0yPEk`4`F^GOIpI5VwB&x}(@F~0=ycbvQI>;pCA?St
zjNBUVLAvP3qbGkJU*Te`PqYkncT5ntX!Y8$GP6XIER}9vjjkHDb!Z?cRy`AG#NIb_
z$;?b!9@L@vDQZ8dK!SVQj5NEyv=?d39qddC%MSa#q~S=}oFU;@VcU5s_sz&yW_PhF
zeJ(5QePZ=yj}|{-;RWX8q+j@l_hKTLOm%SN&AH>O&AKU}p|HgsJgBvC5KG7Vc3#45
zJUiE$*<jS-je?iijJHp|2L2G`tZwke0{P1wXCiN8e(Jr^6v*AZ8O}jIr(EVddoD}*
z@`JvKL^ZY4fRT^}pgzjC&a~0HIhh8B&rTLeeYmx5uUbBfahAv5xaieE3pa?4mIHSK
zM8~QV0{6R=VM$rf*Z~$6??wi~#(wfcI@7k!{o2`33G19x#SB!|7iQ$mxAeb{P&*A^
z*}A@wr5Ce#Q}1DKT#g2>%8sV;18g(!M=q9C+V++3!DB@in3_3q!g3@x@_0DL4YrFQ
zkg}7Q2>Q?WSGx__A$D=K3hJ3zo@wwZ2264)*S2%e%I}SI1=}um)ey5#TJb~-3$XN&
zq;a1;dv=o?h};`Sx*n=o|I;QZv%e>@$F*bi#T5bkj)}a9k+F(aPOdbWz4(201x>La
z+qjl*R5m+N>niKzPvi{Vsv$%|uQ2V<`@$1S`CDs(auYe5is2*cHn7Sn6S~~EV|i%B
zn@xC``&wCt*cxPRb1hVE1J6x=c@(|JF_q#clGmGJV`*{ot2LTGrE$_aJxU||j>iT{
z`spnc-K0#<Gpyz5Psw+CeTf`B7<5(Q>X;DRetkGlMTrM7M6y{&waO65J`9bS*Bk&S
z9p@5~is=fz%9ZkYVXG20O-`-#Ssaj&%iR??({IE8I3#zxeceD)LivUE-t3qP!2UZ-
zJvYvj`|1sLv*_F%S(U@3R`3u543#Qqo&8knczngB#5hY-Dhc~lsZ-v4wxLmk5sqts
zXS?%zg8R!mO`NA%>=P<2{zr}+xk_GEm0dVF#|jMwv_IMHkma<B1G9NP%9>_-2_`ZM
z!Sf_~+ZS3MT6$GI%Lxu<iI}h_+rb_zEM`g2D9<%y_l+BzO*;e6g#GH=w8SamaUGdy
z5=TD!OElf~)brt3ON$ZXxf2R;Y3?hf4KPf@y{0Jg?W^WL^uOm1MvNL!uquf&UA^SM
zG)50qp?u9|*!pS6vtFLx@TaWB0+Ct}4q=hzgo~7GGWC+0?dviKAgxeNB5X+R8@@VW
zLH><ns<Mxkf2G@Sx#s~)LuS{COBvB7JVnzDdvCt>1*1z}e(CtdpUu9@Kt3L`+saBU
zw%c%V5*^2g9G(j8t<B8xNCTsIV`}RlqKv~|LdSKgh9TQ4zCh&M-rjHomVVbQ<Wqx|
z{o$10z9TJ(zR*^XxB8m^I~9>Af*EA=&cORBDq6j9CHHT`*oFl>zXq#ZTiDhglq=6!
z4DcLmHH90=?_pS-NUVmhm}bSQ5BWK7*G>I}t)%UMt-MtJ>9gU&Nf~9!TKM{BNqlc+
z9Go-S&Y&ni>6YQXwf4YK+8))ebyDi{-1VZ`?YFkoGN~H)=-U+jTTEuJW5Nt|TDkJ2
z>*eT+9o1!b38yfu&W(9(oSf{6f<m1$^I9Z^>2b$u$-yetoa4ZK=S7$D82+}AHy7M{
zt)C;A$P4<5*}xw{5wL2rSt}D~>0yAM{Kz*pfWh4shn8N>1~>&)s2GxAj%%iGymG&T
z8#McnU*)SuBcd%g<K8{e$>INT^}X+rv8-p$juo76Tc8_<yZ8t4wDjBtt^EmM)f3X?
zt2}6+ZJU1M3;3$*y2n8Z+S1)U3aZ|t*UPPPTpu5GwYr<>88p$@YC3sbVAVN3GnCax
zt>u?|%!QDK-XHy&stESs;E^Ni9ZN$)2(GE)s+zryf@xC&;`&xJqbfGmK1~`UaSCc*
z4#l4_S{0WwEIgFO=OQapx@e2yVcmG#)8@renYC-ViscaA4uxFUtfr$fM`3)Z6qLk6
z47(Y!m!yNG_mHK>HrC|p62%4{#T1jv7+KvzOkI01{o_tzfll8hFlDur)7%%UrDo4e
z9tCq{T5<f)rhsQ-G&8+Vae3l?F7L<uju!7G>G<z;4O1isZS<Y{0y&U#se9N~n@{Cm
z0RS|-AV<;*2?czJO|}nX9_wLZ+(#md_TF4)0)0#|awO68rvUD3_J0q2eO3Af>bOuX
zJ0@*|qv6HP<$QAn2@7GTECpMixVZB0FYpuPgC0NfJHGRedi%9?0MBeX`#WMILVuZm
z6;gy@pToj_HhyPa?pSVlXlv@{nFR`V4OR=V!3=&@5WA6oZpEd)-<<sBn5P7=s<cob
znEtp0kbvPjb2;(MvqgCB*bB5fn^V#>i23Vj@0he3yG}Z#FBEf!S>H!GcdDr!U*RPX
zz0}>7YheM}@p+S;se8MdDITI7`bOVl*FNejcb`&D{SVow?5#DQeDMayYo@l+b`|87
zy8#ATcG?Tvev$@HhU}<-02V3OU(Y44hoyii8xbF$teL*u+Q?pqKkP<)eV~7s%X|w&
zG(F1r^J+fe^5svJMu?@d1IQ!n2^NjNo$7Eg44{rN#gK?Fk&Q!x3fNN0!F4tH1AQ<0
z&(QpT7}xTqTYZ;djQi#)FiCN)JNJ9wtBE$>BUmdK{FCvV{>|Xd*VL7N1&gmidwd`f
zcfm`sC_8s;ib36kDBcA}+Xeo0y(85n#q>$-V}-Kp+6Si7C;iwxEIt3~acbGrGncaF
zU^kW`7PW_C%e^V>&YRUo_QDM)n3ZtSzqHu6k<46leVLdWR-BmGFD-wgWuUm5LhxE@
z>3Su)Xz1}jEBT2Qml>A&_>eYpQ%%0jLzm_=tc2n^=|;u$>`F1^#ZD2g7JL$|bGAIx
zr=Bs_R$0rZhMkzagWj3&q7~<9T(=uvJ9s86nLx*h_XegW6TWRAtx~W#8rHF@m3N49
z1Gum7s_!KaMHqA*EAy;Hum|Zp*X8sis~$oEDOUfxdY8Asd`#{BgZ}RljwS|Z+Blpn
z6Z+M}0FOnoZP7SoA>eY6*nBx!YWo=GPlCB$!3Pq%uWSdS!f5(ywO1DHOi}$+ZCqiL
zQagPnaB5enX8scTHxnYqZ|CV6!A%D?y|Z_U-5KaU=^@h9t-E^SjL-CR^~mgSBGf%&
zLdAW~PKFvGC!cl~W*^v3F_Jpch@L6T9!H^}rlv`#KhIF)rdMp93^@~2_X2m<Go?fZ
zBgf&^Kkl!Y3r>#XO_ilYM_bhMJMR#717KNgxH4*w)&NK*GonlJ?+d&rhu=z^g|+Qa
zxT7jo=lr~7;G(K&rJt#ZYsYL_W*k&1bsZZxyRuKJpe)2tS@AUvGbA&!?GCaueK_|w
zUa=S;hP@c<6ysTtK4nO5+=6FuoT$|AgLKiJ`u)6$Hd|NkFsI3&qo&a3N1JKb-G~Bl
z*|E@{ZV#efFN3YuKJ<Ng`PjwOUOzp<2~|J<K2r03q1INxJE20wl`>m|{n)1X_2#<0
zpVn4^`yT)q0Br{f`q?h3`9(v?QqH(K5>KQWknh$0@9flG3kx?d)-h!kWVl8~$4XsA
zX`W1h{24CGA6ZHv9INE!q<@gxRMdU2->fC!k_0bC0wY~GzT*3zGYu5uRN)rS-x^#r
z{plsjr^>HU_4c$aM|fYtQrGV0y318(l+T`h#nWE2Oa2C*e}T{_ia%Lnyp6W&@X|N-
z-&zQ7g&EZW8&yyhdFXKG1HW=CF}Luvi7D1>&QWMllZqgznp#Ch_?dKO=ylWy5?L~M
z4^h!uVVi9mM!ydq2ZOSd`u%NGzoD&IJU3{Sb*%!tT})dG{(0p1%*&HM5qioH5qC^*
zw^AhUe&tTea_0ECuf#r0%p7|~#hULM#&*{j2#h<Y9cuXa3KeY31Ab}ofkgh(IDqLG
zc!64e$gCc$XUq_6dIVT=?JI65mJ9$U+vG5wi_^YPDrQIq3E2kqEVHlh!YASz2M<7^
zq4C3R-j7bX_BOtimBgGLOhx*p9d6*n@qz+!?0Cpvd<<DUSJH!x%xb2YHdpa4%p?Ej
zFK;pV8~1;m$}-nHw3NC@?F%byF&vX&=);ULkjcqY^SAExE)Yr08U4*qnB~x@c|9{U
ziFH`g<E^!mYHCLWGcI&DA`HcYr4@M_h+Zi(=h?G&fR1Q9&$SDjK5|$p`;`0#YS>|^
zo|C~YvQH$EEYi{xmxlC>8q>|+fhW}ZjiDR5kMs9;{)2ve?$Bnbh_3G-dC`8n@ydyx
zZY%Fiqos`piwm}(VGLQtW*BI4KP_H}MyD&k^B0Q@Vq+Ro5$RlFjF#{Qw$_~AmXqt?
z%H9UV*b?2o!(s>hRMste_2Wxk73$xObG<zMMWO818`&}is2irJhCV3^{dv0z#ZBkz
zk83@XGWn8O6U$6$F^1KVBe&hM_z`TjAA1x9x2Uksu4NZ5Kr>~X_WS1|3<+wh&;@+)
zn2~1chScFyJoLG^s>za1F@NjI;CzA=+ZI>;EUeL*p(_3Voj{Vg(y#vBu}p)DXM(7V
zc$_*EIsrWK*SFRkP+Fy<Vx`pbDf;M7Bk?+8%R~1x=Y9?Dmi66*+Lvj`dG2cn-9Tbt
zT|XH6A^S_FtLOju6ti#|tEXJ08|*CTzLcU1*^)HR=+&;rY5Z5(i!ANS6sT0*#NjY+
z5A{;Qqws1bU6s9wO5g!kgWXZ6JQp(I3aC|82#bQkj*Q>!2NYgz>mj!X0jXC#gASQl
zqCQm+p;$9P1i_xlWf}d?Afrs~GlVSg$4J&Bp&}FW>Q*;@%I_JB>>2bkA29vQvobnv
zGSolHqc)^G?_u5a2`qD_&U~lt*iQXywzIDnqpmtz9(ZW!n=f&KnU|rOMMpab+C)7c
z1GK&|-+qI5olNi9(8``&1<zY9h}931+D?{M+;Nz-#!>PeQ+H95ksf0K=A~)dprEGq
zyd|KLd(l{kS=}7Rd?p@UXsDFJ%$|FQm1P=C_e3cBZ`ym)!s08G!F?Mlb@nPv*lt*I
zVqv`zvsAWjwSpP{mU~NQWi3zy`{4}d-dS&<?|_d6vVqvZIRpm*)iAr<?08Fk%j~%Q
z=gu%Tla>%w`&slY`NvL2w+3C$$AR<PQ`*?6I`D9=R`C|adRG<&4y;`(FIpjjc|-`E
z!C8E6-Zsi;n!BU_XlIQU3|$@gvp3MCmoqlHOSDc|QA!7jy$WT|dcBv+0t3@^y}t+@
zQWcw;6Y&aeg{gs@&l>(~Y%KTLvom>%$VBIHTiJxdI1V)APKn(%U0RYY^5I&p&z>vy
zH!fF<p04dFd0fbvt01`%6w=qzUCM@MDo3a!Eov=B+WrspYVt&n8O4M1znBsyJi4MO
zeyWP#GYecDQw5e5pGJEMxgl*EudtWOpKCvWPJ!+!C3}8-ga=su%)+DXkHKT!<fk$9
z<Okj7D1C2>lqvUn!<>E)B+>)l<+s(`UR}%gT|+=k8(lc^%HceSrtnJXj8s)D$+EVF
z<?>;9rsy?4>L2u71%u=M4c#^QTelOX8BYa*Kg2hdrMVQ=`8^EE>UE?<KP=2%ayE09
z>^^&zX6xr(KyeHmkUE~vU-bG{RR5V)bp)#h&xd}=5n6H7>vPJkY{yW|AQAwiqY^<t
zk<`!OHZ1uHr8fU`<TTuk#mim5!7>dU+u0m2UwlO3ZQ(09J3f`wWs7%d$`zY>$O+WL
zUV?X;b$j_yyLf}##FmC5VyEWGo?0#uKu7stKOr1|qCT-EsxK(8^u?nxv{h*zX0{r(
zFk?P4VUCZlkr`k__zv4XvsAO~n1Mp&&zc_91BE{?{!1aMr^7rZ*-yLoFvZ`T(aw&I
zr%^uPj<>G<&=j~N#Wa1JYO96P5`$IvK@r_r@>u)Nk+Hds%3A3NQo(dL*gQliavgK^
zGW13V^KyS)$>BT_NzX~@Xg-lbx6^L#x}Q@~%}RYgI;_x~DZt!`-c`MyWE%uELB;I-
z7fg<>oO(3e$t#T7mlg-fLM9Rgz0h9t>S|@OsZ8iQA>LFe3+RN5*U-jvLD5CDuC@;=
zNQf>7J1Q+Ff<+=vLA?RBz=_WCuXO-+vGfmL3}dzC&Ia}-0>Yrp6l-2|<eZ#&{bt19
zgf<q{s46hJ@rm?J@Uwk#oH!HIYvO9Pb7o!NMb=ozsQtKLN_RC#cW2LihzzAnKB1ke
z&fmH-h?x6eke~<7zAa^_A>ZAE6`JbOsg5eB>7}vsDQPD~0$8%wAC}wu`Fp`(h`;Hh
zrH1R!cs&zYS{4>pz9i>3Yy01ux*feI!Cf}d)yRN)4d7NXMI_Eh<$m4@e{-=*&$rfC
zSb%V#+}AcB1HTBlQ<QpmJhu|}u@k>7=8<}T<Nn03w*E`_ltg|>eAssCd$<9vl{REg
zn9oJ2STWiwNcO#0m$)R;X0N_U3;cdu>z?tcexnmat(MUlhHXZ-1hw;M*xXUs!%a?f
z1M=>)G+tT2TW>vLSrzLzFlTFR8+>O;Ek~zfz`kFdQ+~WeQ%P%XeB>%neG<wW@b~59
z<wd`UIA^3qqE#Nf0Qd}3T#Yt@{kpw3AC7AY5%Df<phx)=$FQ|9*Vw=$H`6o8CDiYO
z>+BD{7Ez&K0O;Hafu8GdF5Q*dH{4?q7B}uBYvfWmJ6EO>Qlb*QRZXqTwU}t&+A$J}
zZ+i!<$yjPrp^$8&eds693}a6=ylDGF<0ZUra8#9(U2gi~TuXy0t`4Sg<-QzD=P92$
z+&CT^y+^rnu6D%QX@S?}`+e5(@|>Al<v_@JStzY*(TRNl8?rdZAiLpb4h1#EpPA9!
z_vZ)lj=8gI4`Cj$`CA3=3mUrII>*(4JpVYl4PX>m)f{;gxCS$MWHR^k8yKDXNpK5*
z85r=GC1)qJt3Cx9qNT|%-Zo=+OHkJ6c&W;hY*!AyF5}bqaq4V8t3ojKYm;zLWOd^d
zxTkwi1C5V&Upp3s$VJ3Y6xIlgy&*@#Fbn&%Fhtp^Xhm9gkO3ewHgWvgKRQxphWq?X
zv^rb*3D&UllW+!a!T-#jnRpW)qp}!dNOB}}!hB5Hfz{nI!5A0U3kKZnEuS4iq+MvZ
z<?g6C8~AECxoJ50s^E69s<e@d)sJlq2gN)0+#Q=Bp)tGa5w^x@vf%kJ+yiib9Gp6D
zcTcA5(&mk11-#e323<YyEZN=zlC;}~v94FLMQ<7O{7(u3Y2m=b?xUJjB3oLq=O7ld
z95{RJ89a!keGJ|+8mFW;Gqd;Sq^J;i40ZR)fea2Cy5zS7Ep5^nSKgA*Ub&D3*r^t#
zrY%SP$7~*v<#J7+YG}c$$b!T|ZEWCV!JjS)^;|(`uKv+ae(gbJ8X`5~Ylknf!}|Gf
z(?Y!Q^HfV$Fngjq;~QjO)FYDWLU#g8huA6+rAMAgA5}jWCdwwJg-LA?Azkk_+^?P;
z>BKr<2^mOGn#SD7^B5IH2qpGl9Ty3?Q$1UyAR;+G0B24pofvm|(J@hhFN>S7DBrC3
zD-~hLQ{bbTR}st7&`wN{ttXTQWlbUa6f5A=f0GqyYn!|j;3Rqypi^X;h;sh@U*8T}
ztgCgM>D5=MJvZc_rskMn;vCxFHo8p|BdK#W$@PXSrqXk-0_D6kuTUiYqpHV@&|8Xn
z)g%OJE6R#`#6Fwske9#vfjyN*oi5oT>{u*J$B!V~2v@vs=T(g)rfFq@-zv`K`x_T~
zb__4m-pg6}-Dn^F_fsxmg-uEdPkEvV98OqnuIp6uI8mWIqyd*a8FEzCxM_+rW5V1e
z79ThVb>$tpQoFqKh;Rd?VFpAAjGWeO82d+N7Uq@t?Abn@8$$P8Q)}Wa1Qa=v>#Ytg
z6%d@*Wdt1}fGa~cz`lvH24Tu@U+D{UwF8BK;~Z7mY^tTn!G8Qr|LjI|&x3vxs3C~m
zP4*(~e%`nHl#^)kXy98Mx$C;;LF*NlIyK9DOJXWx{&!F7-xqn!jM*-=|1H5o%#7O6
z(pqlo+G#{6be~-q^?}LZ`o*k+5m{I68xm)K4nv(LH9%(>0vRnuAaOi)Ef5i%T|RQ+
zl33LGTE6pZHb1PP=|hX+z3eh8qAISd7vx(9-SiF+05++Ov4^-{STy%BX;mEFrir5t
z8O87!T33&0g6n%5henn8jH&41t{;$D?}q7TKSS7)C3ZmH{3-=ZtE?v6#U`e$Qzw?)
znT}6)dkc(jc{-$BZ{jC|X1^f93a(^f9Zh_Ln4z&Pw@|Q+DHuews82vUZ#IIqXy7JA
z<TcnPnT5kLA9+sxX5pqk<+UZe`ri;g_h}SL1a&gb*To2Q4ZO}PZMT>5*~gtuo`@*G
zbgky;S8#BdaS+By<40loE`}+~G_A$TkLUT7S=%*DwJ_<s`a?Q}HAM-=pY*Fq`yboM
zPmLLfQsgWTt%I{LfcHc6-Or0FSMPzT@g8E|ggy-7FV>~~2EihE)Lnd4nlKDS@l~?W
z@o{|EbJhFv!*{FD;c%t3-_QLvu7M6EPY;BuqgI(d*cRZel~A)#HiDJPDzvR+d%$0H
z`8X@xBRiSxKD6}Denp=PlhgU`>De&{$&`&^q$TcwO$zdR8xld|a8X!>tIi()QXY;^
zOiB85@aM^*)sjm_1sO4w<VU$_!YG2No~)VZNC-5-q8@I%(g0j^Ek9GZ81P}>!`y2G
z+L%W&IOrvL-ln&4x6k~@^<=W+JW1#6LgQ$*O_7y>CL+a&?v5MOz-gRdEr;s19lR0a
zv|O+Gjbf!sf)Q6~fM3@wLFcEJ{@S+=LF;fZgMzy}mRE1~P-k7RlWTyX(c-}Iq4wh7
zu&2FN!MQwwYzibquo6Wbpc$xMz<&d`#NY;z@&PnH_GWFyv?~(`gRBGlg@v~Q&I>@*
zl+h-VQ}dgI@cjIoj72nx)GMDX>DcF6ryfK{^Fo8XCYvFdR)<n*14!6nPf{o)QpKJ=
z`4qk>50>7th!Jz@xqHV=s9SwFB;-Qf1DS~umF7Pts)P|xDK8YQjG5iJnWdOaBt1DV
zhB6hMv|CJl!bFt&Abbc-)^ivZYrcou04INC^dAGeKH!;70#df;WJ6;?98@pHkId|N
zGWFS44`T!eb%@z)uE}?{nXdp%Ta|tcn0<|iOS8xu(3i*<FGw9X4kEfb{QCNQa;)RD
zz^|m^!KBfy!s3ZqSe#}9uo5ud?eIs}47tC7-q!%o6f5YoO0;kJv?pBTS_(NZ^r;Xp
zi_xf;yA`x7nF@6}%7lq{Df6xHmQyJTU!{(QFHpRbjjtv@3NHYN3E`Z6vZiD`xYMKu
z^QQ5wA3Zr-1xw&>eO8d$J3kee{qEE>+&>Jsqsmvq%&DLC4!9^*P?oFYk8_hjppIQ4
zkM7SSAC<eK1nhf%HtxKd1R3JP@nLYButUBErq-4%hemKH-&FT1VFr)QH_w-_{fauc
zlauB3^|g^V03ZhC@&apz8hY3R+Z(Llkr^hC6!4!SW*4Jz*J!WA2=AH38=WS{5jHI<
z1|b3YDz2wov$bfKfrY`$+#y~K-&o;zw|?Dzl*Ov41*V1-Pk*{<t5OR@AZ(qRU;iGl
z1}<wYxZt-EJG#F4I@5ENoD{?94Xg%TTT2hQg>4def(Y^)tRKIGZnU@mZE2^la6@bV
zX<WM<iin5+JTn<Y#>!8Dj|W|;4d=WPZz{@XyHQ!9`^s&y<*7Njxd3vP<{1>JG(QrI
zZ=oLmHwBC@6zyudkR$-g*ahDJDF!0G8Ebtij4#5()=!mz2yBV+b$EgcaK8X1Ys(Hq
z+Cj<ctCnh@Ps9xotX!n-9FmgMhIykO$GT2I<fbK`BXUZExT7;3qn=3D-a7mMG8FR{
z{S`ZGRY`xpD!4(zEjCDDtH4Qxt-v@yI0g>N{+JHra<!I_s}pjhpnX!BlKrH+CDZlC
znaPVvQRYgwp|OG;T^duMXXx0wa`Z1_(oK!tK*ePcnjsqR?3tQ%gH}3Q#1NJb$yuQF
zQC4Yc6f-Ms$sazIf{W9)wd*wSbMc1q54}T@5ys24!tZZd?e4^J66K~350|*Y;)E!M
z$j=u9^#&i{?}AMNL@C?jqeJ@hpEBTaQ(#s;W_?hV{B<2c$|X01)x9UKWnd)FF!Eo*
zihhZP!}zMobL*T#KdR70uM2p?c(XwNSE}!Nec`>Otpt(?>IM)FV9uf(oSzo(VCGaZ
zD{f!Qr!u7ED{rUj8n!IAmHq3lsD7NMlVXuaq!@)<e#t8x59vE_{(xzE^i{O3q=o_I
z7pM+m6D3Ib_g1d7Rn_`l^qZFQ;?admMgtWfc&II|JV$0_7T0m&_1E14z0wC2`J$a%
z#S=PZ66rf2acUqE*x<D^(Jz{N+<y{<1-NK~XD8pf!@cxPv^U}8gyh1-w9K60I)M^v
z=(6AsGk)KM4QJD1&!j8aW0v}ZC87fnbc6jj*KI=*ilV#on{oQDFX1SYN1M@O*;ib-
z3UJNRTZ~qTL|KcKAlM!DBBcEkkWxPgJo75WizO=81`v<`V%jP#r=1>rcJc|_O=~X7
z?ui|N<at}0LALh6!zcgSR=t$5cn%98u>z~{lbz{GMalaOZavRy+s`zvZEDeBNd8!J
zL{Pd`?Pg~GRp6ulw^*<Jjw$WL)Gx3o_E@`}>tf0q6}&E5wW0qvX(t#6Dp`kvxqSBQ
zVc4!h)C`|7@R1!EPbt>KL0O%&p9^s|qiVBUk>}V9dt#fSS+$W&`ow5SW-b?K$z_fy
z<1~HP5X+t_$y`A%nr4fLoC66J)xZZe5K?;s+e1xc`h=P9KpOJejo;$LN)9Mj{_Ihj
zH?7s$E49rd_l$<9f}|;U%T2^Ju~M~4#%<g{8cLY3_zLps>5kJ`T@V~^c06e8?MVYL
z195)ayZo&RbGpJf|05LsuH3(wdEhc(5FW!U7);F3prAG4t+oHB-6F!*`!dG3_CkXm
z)_i!yrLg6W(3E@of9i7!Ff6(`z&!Tn)+8v3*^^aez_~m%ut=s$L{&0-iFQg$#SGV3
z9vWD0*upm^1fo3%m&Y2riZ~`w0i^AHf&MVJ!<KVi0eo-f*<aT)#nl15UY09b*laWY
z_xDL_{O?|qfpWj(s9|3w(DCOeqGydaIs*-38@zt+&C%I**f#p++H!TulM_9t=kd#y
z8$y*6SWxQ+P%m+Fa}Nko>7qBF4P>;V%%Y<-czmeCRq8jya2LZip7=b!^A4bs3%TxG
zqaT?ALT}EXimN|3M3&y))T<<Rx_o!{&rY0!x54do)fZ4&JKBbD9(JSgVpGC7iVz%O
zk1D!bf4sMEHMAGI6;91}PyNJ{*ow76y|Q$xjZ8#W?*HK&W}M(XR<}PewO4ISqH$+z
zceHLm+CjU}G{sx@=1~pz5l<UV#r^tSgZHA&jWy(wmODB7aF_=UK}87ad4b(;F2iQ3
zM8wUNbi;JK?vrH!-N-$^=-*625BKKqL&yZfrjL|%hN1{hYu7z!VOj`SPSDZB4;@c!
z46jO?ek8MpFzEmdb-|?zEjG{qkalS=g*xwwFnGZ9<MWgcYFG<S+vd6e<DRMs#W_t*
z2t!_J98P_U6HeRK(>c7j{?bM!<sncJC&6rro(i+11+RY8axa#w{uQ!LO$Y!GI>d%g
z)?Bn#TG3Fo!uCEuDDmsAR+WR-I#JYyF;KxB)rT!!pBuWc9;^&qX3}}`E3nqRu-6`h
zm)p8eQtiZ!09Pp!ML)e^&q~|-Qx^p@ePH;`fAp@$F#U$~tN&Pmrxrd{jD*UI=tY+w
zhYC8XhhNS8J|nmaYAg}|deT?eDG#jt<VH%v8Y`;CXMeBVrD~Ej#~yz&JNq%Kr-bCe
z3WA%sV-L!hUytX5e-ds3uVe%N4#rSor9JJp8J<GR!Psdav!A(sb(DzBP7hkX2AN_Z
zrkts3>lzK|`tUnp$OnWRml+2S780$Y@O5eTG2P4h7D7_AMjzx4SE3Iev2<4R|Cqpk
zNg0mETAyrW?^xCn!X7#anyWhmd$b;KHLs#aikcO}dy4*SE|Pinr8IX;V({uvZA1BC
zVwyzPk8H_JpfG*9hacHnFC$btrn{9wVmzSmpU6d7CaUytR{0KW3=J1wlkRsCTu=%d
zj3*gh`UFQ!Sl8C#@!GE150G%o+IfMRvaoP!<)=NO?htVU!XnFihZellE`Qjd%YA=`
ztwlSI%8!~rRxz}gg0#jNKXlKn-EZM#>z-h|SR)nsj)~heS@L|noTQ<RV)@NXEmv~)
zPbdtkseLgV_tXP^|AFVpZBE*BvnBn!D(25gBg#^`$h&gQ`a3F^*Ry$m7$H>sz^^V)
z;^DD$;wN2OTaJ%HtJ#D87~QdU!(HP&h!qCXCw;W!F60cqOS(Fjc8>BlC++R@)?MM4
z<^ZX+1I)U>nzdrPe~s9-kDoA?>-z{Q76Dt22=b*n&>AJNb}+^>UEODiu#zO0?YLSA
zxmJ>8q&axmSe4623m!J&NDUr-9+X-%9weZON5M^zl>W!(${c}rHB2dG&-O(J9>AIv
zxMS-X-Uuprq$x?Nb9HEI3}3F&y<Y7{<oIe7Zi8)<c@|^Pg{r&1v_oFRR9?N`3swDJ
zv)&7i0pMC_L}xt8YhWwsh!$qxe~|-Ch*p)u*ShnpnOhlkfre0V+gdZLsoIHHj`zJN
zB%%*1e)eoLY)?%8F|U$Lw9*G7$UN(z7I8qNErH(fnUCfOx|BBNa@na%zz2QP+c+WI
zSm{9HAy)>axJW52)jRaeQpnF4<GvGtS7`^+mq94|cEHBrxG56f>>rOz$D?qrdKIzs
zB~Z8{T|#V+GnM{%^>xsf|N0u=I0IShkP{&QZ5?!L5Y_V^a0INk9cvNi{&&6L8}qB#
z4LH&fUT|#;?EaNS8=aK?=z1KnTFM4mWyCc_Y;XN1d$3ih<nPZ^Q`<Yz17tb4v7x^b
zNPF<D4>PtU?NQ3~6w~8^u;9v%Am}X52F&D}2$>N9Y;^vTq0vEm*8HJ~x``)}W?Pv1
zb|vfMJLb$GEXi7|S?~HxdvU<9#di{3Ui7wX47;q4ACA^U<^amGRBasUT@-x)Vtt-!
zf-H;a2_SqqXcrlvv2{r-cXNIKcsl<G>3nMML=UJKMaRyXte!x>a%kpJY9+6f-YZ}k
z48a&q*rJzWJK+x+@e^+d4N4z&Gzn{^o`Lj}rZI7HAQ`PkYwug@Uf*3WG{zdsMT^~?
zPRbz$<_XS20q=|wa{a)0krx`mXREpd9_$ZFckBI)3*p^!>fZ)2#c#S=iN))gt!&0|
zKXx+9Vnv$Cobc)^kcZA+gKt@Zn1`eZFl(q>r%7I5W6$#Pa2Op2Ikg)In3OLGFe3hN
zDzFnb8Yillg3nJpPj1FrG||2a%E_ogjX3Mh&oOIy?H~<J)F*?+5@d3XSaT+#UBLY|
zLwc0UKihC5-BZ_xHy^-7<Ik<o#daX24e+f~3q;ThVu<r=;E!e?ib}l7WNZjQ_TxJT
zN_3#3*sLW#mnOXHbgEbB!;c9fi}2~K=4sxnJnnL+t!_9XB&Jtfa(XFu^ZygQX|~PF
z7ghhcquESpU`K5C9lHAxquA5NF#M6Q%rY;r>+63?Ewmz}fTISo*x+@f;JgGs*<6&>
z<wkt`g!c?9*KJRG8eXjUFuS67(hu))$#HV-kvWGq=Q`=f9pi$cbg(|X_P>x+dG6l+
zqhYh(sb>Mca?!w`^S{^1YTl_R-d=f5d4a7%cqYimoNXM!tHr%DV(8L)RFM1^QgFxd
zO}$p=X5gxuIE=R$0~yjNq-NOB)7C?GHeb1~J)aE@8nOmW>+mmZWqqI&U9^glE(HDw
z$#v=LDGyHs)0UDFQCy0Cf2wnp_lO~|4SOuy3`0gh(hEC>X|7e(-^(ZJ5!#l&eZdI-
zxWsd<zSG-5&H_Mrj(Xbe#&N|Ack@39wIb#BAm!5X|M_d4-n%y?Yx{rjScWGJf^j5q
zP`fE4Xp8Fvl1vFhuc;Y3j(Y6w2R8r?(2|e3&b{2vspy~hy)$=|(0#qQzO$Ge_I)Y}
z=kVMM5+mowX!8Vh*1SP{M(2TuVHJ?=l<P0EJoG;84*joCVfJ3L^PakFQxNI=kt2-}
z<!(nmqHtv&rZPayTlmWdOz(M*fr%aw0D|z+Jpk5F{~(BXW_)nus<7_O$<r`&92ZWj
zVRWWM#wW*yTNn4e_2!D+DLD7R4=;|9?S$$-2v>6#;wyp7ge*0GYlGP`6p1{Rm1=WF
zNi~@qN#{r1F4T_J$>~NjKm5VW8=Ek!_~1Y6&X_5Xp^!!-gsnsK3eaVCwcl?)MoS^%
zZS0}b)q{nL*X`1QE%2R+F;Z|--v|Aqj?()leW|K-Qy|AFIRNgV7>WX=At1myW~>c>
z7%bTLub{NU!_RvZ7Klq6vy61HwD@E7<W3N$4l2KnRvnHP%KeAL{3lMWymRj^ePNLj
ze)nv3PN+=CS(efKonRiCrEG#w{wfHn{^$e{3G>=_q0*Nc+%Uc=e%=>LR#Y=02*bXS
zMH))qKrJ$69%=uV_O$=Rqi$Ojt(tswH8o%Z>+K-F?V^(x?6(JgZtL&f1yK0?v8@$|
zA}CKJh5bEcG|6h<zphlA6LDzK@qRQ6r?iL73nRvfv-eg@?XEE;7zoR5=;jzZna*j#
zo{6kHNc_hqnwfM8mxqH<P<#IHTY<@NY{e4%%5potPjyXRJT)^rS;p0;nmTnFly*p+
znYqXjy9q(i|IxoXl-mHFz2`+cY2%PV9(#&vz^{Fmjh%-TyQfk4;dEwJ1sj-0@bG`)
z;+B}nSbNZ-LOamN;tqv-wg0SG7gH$iNeHh_)5y(*l9DLac3Xqm+F}I^Go+zsBjcBe
z7!0EMwrT>=GpLg0Rn;3PHSl{<CR8gy0Ik2!CicJB{yUs<)Hr$}Fe&qC|1q!MwBENg
zkrqeo?TrDSX<-2_)s|DRASDbs(d=%d8I%L?3OJPD9yU>#BCfvk<IFyb*ZlpnA4c;a
zs@|#m3$$Ta`i!p6B8;R!EIY*uoUY?M2Nvrx<A4<A^_@lgvMbCKmZfMv62_dZ_@j9T
zzHW*98Z%jsM3g}k42cyOZX?E&0wOe;0j9%0&>N&OK8kJf;^uXOnqu*D`D3;rsCye-
zG$res>vjYk9$z^K@ClERoZ$60wE<fXtw+3$G1ca#x^Nt0@C0x7uX%p3jxgT`lEe<l
ztL_F(GL&cbveyB(MvEkIi<tciGU4O(Aia0-=8gVUq+!nUz&U;7vdknvCjlssNC=e^
z57R>hUhKp1vp*0MNlQLiYK%2o_MQcX6=j%Em{-dZm@`{OsdjO}@Xl>jft7%jDFwu+
zrMmdXOD))T&B2?$u;|t_mAd^xLOB^c3m{@*M7%66yqXa{4`5Pv!X%uzuiH@-d`A#-
zkxl3s);{#h;9|tNC;(gE5Im;_577D!VL7}(#!N>41MsR?^Ker|^*g4>qE>UI?B?a^
z*?7u|J!0H>$8dZFyY=fi<=)WsYH$z+*u4#;dg4x^#21Iwb~`mgM6hwwOD(*OQ|B&@
zmIh%)t~I9rDX=u(nCtlfSoCOM%CO+_B2=EHd1kN0u$}!zOP$6OJxGik{>gxuQXV05
z-QL9wmFP|ZZ3#;s?mi=84UNjN{}d1~hBcxDO_S!GYKmzxT@<O2lWZbyA_B~tm;N<4
zt#PU)I5&|hUu)w4i44|U0iO0%?`illyMBTrTo-%EjRF3xFt;42xtAyRO%)&mVL0$2
z<5w0H#8WD&_leySA*=l-@JTk_4@N8$q0HkI?FznetuMaNoV11fkPpOwW#-wIOGp(M
zti(A{NZxsXJ4TR7$3uOoy|Q+}zuC32=-VUkNtn%A@T~*)dA8(shx_K$%v>?nAWJnx
z@;#ED>_p#@UW)jHwdp>cfF@Z)(J>l8wo}`8Je@l5;c;eGULx-Hrxj)GQZ~Dn5EW2X
z{wcViBy?F49VT32jCiOcM8&qlmy*J?$X;@NXYkwUEts-m=QtIpe%!=K_($5zkn%mU
zwb)Ud=HNq=>0aF8q|^p_^rfBmu0%iVuTg*phu*T^E!Ig9RPPct0M1yf@QpT8f9WZ+
zbg0R*`=?OxjU|7&^cG{77R&4(`&Y)4a}3S^6<)NIqkszh>jfJM@k5j=!qR0}X*hju
zVui^^V8_qAf^JZ_l7teY<0f82Xp8&iBP<IcHhbWAgi6y8dKZo7#+1A1UdH)CR;+BI
zwh=|B=Jf<u_yx*LWF<~F&}@$JPQSD|P<camu*q9LUfo0uM4w=PO!Zziqyjb^+F%3b
z?&J}<EH$p2BGd+e9`D!GhPduPbvyYR>n27({|P?ImgGd^0dbs67im5W^3Q%nrnyp!
z4!iOQ(^!?9%#o<uP%OKD3<_p2M@V>$Md8_)e++GGu#WGS+WQi9Yr)b^HBUXpt%Jjr
zFGriim<qqs9{|x&V@I62sW$P$-#@6G!|eM%xBEy5tIGz&KWC6rL$xoYz4D!<mAl<}
zWL}LlpScd^)nIb%h2k$!;skR;E8$O{6{Gx4=v!(FwjY-mG&!NN96R{jeQxR0D)dcL
zt8Add?T&x`i=PX_PnT2vIR5MOjU%)X$`0g}-QXN|ugRkj>z7EF%iW>>y`kVZ+SfL{
z@vsqXIQe*}fvD-Oiy{_OXki0F{ZAv82+|4&qPv7kik^v7eAE;pIgC#K#f}=>)1bR7
zFMxV^h@1CF<ZWfhz|3|K!o6lwFQ9M5iVq(-vUm5vj7`GYDAYa}){1wG!N`I6Xl&L3
ziDhuFrn^TH6p?Ex)u8aH;BDg&x$6S);0Q=S{GodVIa74CTII&^Lb6dC%xxgHG24?h
zgw0Ol!ld%(Oa4ta%EjZ>bvStVPrVr$=BGD9Jr}5$AIAe{zXL%H!qiAx74@2Y(|<+j
z;5Kw`XefMJir-t0qj+i?!Ck2daOWCclp8xKHgIQIVx$KkzzLZ6Wi;Fi7Sm9P{2IEZ
zUKxSfeetE^8$YF{JKK#K=S*3V0SUUXE>Tq;<1B@i&af78WD7>y4DjC{$m`aHD&D*H
z0je1QkxY+4DNm8SPb?lh3@Ux%WhBf|l3{jXf;}8?R~<NSk5x1G&;dBJ)u;VBYat&6
zJ5TzUKzP9%ok5hiPw7rc^({DU4^a>x-yX#K(Aw{OVDDTmf4Tct&xW`q!seU>5w%uA
zsz>+k@D#gW(JS(&Mk}zT?|t#y^zPt@4<gt_=qL&ZT3;>fWr7e9^*+H*h(BAS1Y#fw
zjO{+^<`=rgFmU_iRDoWfi`XCB5o6eezv|9G*U}xja8J~Q{#(Y8%*@(hjNUeASwu*Q
zLbW#xnn_qK6*2U-y_%#LUfryYcaFQi@q6hqG_osr&h{Xbld(DrfY7UibV_zFY8qPt
z>c$5T*8UUZ+9Z8W0Yao{K*<Ls-pnBmT4o=!ifWEtk=ah5-YIxTDB^Ue|69aK$*eLS
zcH0a^T33i=Q1mOu8PP!il+o>^S(kwy6+o`+#!7&4srJ&p1%Xe<o;Kb!A(}M(P0-zy
z7<T}jw4hy+@aJ51GTL0NCxull;CM!Z0zkalm^;NDPNsj@9hAzXmhEwB;yYV$4j@(?
zuI4BpIjWJmtLw{OzIstsV$)KXM5|UUe|qEq2x4U+i`?JoyR2x7Xnfv*`N)4B-gfBl
z^T|o;;rjC_ejf9HSQWLxx=#adM#7iQrmWk0e$+!xt4831m2V*P?9N)xxuOp=4bZNr
zz9M2%><O%2!5TC-0ElpHtytHRrShAucn>Xo2&~o?d05U7CDeS0k=>jg9EjLox1p8x
znHiYr2bEBfcZOS;;-|Qa`y|rws4XtHV`yOoGE8*INb~*{A#*2<B)J^#J4_WbN;2~E
z@=#C!WsrXV)MFKNB1JSKR!FBXRKcAn#PO(-&}0QGpk%S6O|l>n_ZhPJ+6B?=ii!i=
z%jIqZk*TU^ypQRG4H#zjcv3BNYhW~>{^uPd{108&L^u_A@H`M$(AM%D!zjnOvztEz
zq7NdeM#R@|ZcTrq4_r17R_`~}JZDGmnMZR#K6G9N>Jwg|S0yNH-xhj+l;>FetUECO
zmMex;7ji#VGC%Z+GFo8821b5HcJXgW@x#Gbbl*C{Z-+yznLIlzX$$Owuf4sunG)ZI
zuIf88sHM?4Q|G<T=qhUD>a;f23HLndF(As+T8BdIlR{K4+ye9X?qHdQFhvwkZ{6^Z
zrysbKs(=QZbv&O8s;RFZVSTxa9u3~iU%*K;Ah8Whd&R9k?U=BtbT;sy4#yijs+)C~
z|2*j!Oi{j<jXy;8Tv35#wvSGD)X<KB7mBW{S`?M}A8A+J4ILs{mKGs%g-e8)HTe|R
z@V7|EaQ*u9jEz^+P%oGMic^GV?a!X&Z>BdS1Uh4sg9$zR4XOfD4dzkp7|8^!be&2F
zTavNKB<`VS5_@=f!G5+9QT@ag5W@&|QXcSbv<8Ks+ozp^%I!@XB#$4->pvrd_@}}H
zKP2l;1+k$blD1@Nv7X}fg**HBra*Xp0pVfYa$bcp;+aOX38ub=!qAzs7|?xt(as(4
z8;8EL(!YYO4M1TEG({iWoK#8ZleRb&>v~#uODS5ey0ltK6fln^ipGkxnY|68f1+;K
zcFzYHxb^DQi2&&UZ+nR0Zx*<RWSM7+=iiS3>eON&(8R?`>tWRK6-ji7;*&KY4+eA9
zF@4nFnHZ-kUI7XSc=YUV%a98oxsqRk?)W8v85_>4DYRQ^xdz(s5Ka10I;Hay={UM6
zjPX*ai4w4HTApGhy|P^EnLZ;lg7fTleN?VZ1AZk=Sab~$ycBD`^uL%UBo(c+3?Thj
zP$z>`4O~TKxZ732puZbjWx6A|#bJ*r>UJr2kyHQ^*jwiX9%SPQ*T@r4Lnf_HPk}Nj
zARIW8^NZ$Ohfre%^BV#(<G`W{Zu}Do&S|y++i730rH%uU3Nk$gmHEEAS^og8{FG2t
z16`g0=f5F_A-4l=>W+I&(8}f0$k4|cY6Z5%1Za|#qJiEruc}h4Igr>HJL^JuH02F#
zDz+T-6hJz`bN9;Hn_tpK($ZaZl67aG+<LQ-8uZOInVx9^o18T>$pl-+)9r+9zv0!1
z;u}uxnq+ZkvBLV{5g)`*(_UqWAQ4KlOsSWNuxj9EHlm)^$j`H?{a)+3iU5@&#kUFU
z7o&&JC&TFQit6ttV_fQFTFc+)i*L$E(cK7h3a6>@mG$Lrco;B)HETx1hkvV^$oyv$
z#E#!hJfq;W+@`K;MfvPf%X6%nO4$gK+vEy(^#dnsegO>xtpZ6wDfJ04b>(Xs_?5Qf
zn|!Dfr;vyyDL8a?g-KYUJ8>SdYyi^!?!C`^twdQ6b>eL{{E+p?o!2tZM%1<gTOw!J
zNmjfQ0v*jkKbO^pwZ-845a-`<1x_>WuP>A8gO9_?OVEEZVM%wM2Av?l!uj-P9I~G&
z-H#~yDUWKoxpTMbdi0GhPQ~jyolY)|cq9`EO6*oMNsmPBMRQy87anBiYz&%x6dgT0
z=YK*70+T@kq!B}1+4)5d7CO4wV6K1;B$vgyMY~$*1<|u*`Y9U2XR%h$u_F$jU-A@n
zOJPUzsBE;vG!ruZn>eY%mb0mkwJGEdeMyn0oS$8Pka81ny>R`r4a8UIzxfBtiI!`k
zs$S}6YM>^79@{Co>z{3qJ;>j=_Td)*AJ<0n>hW2SihqWd!_DZI*aF+Ib`27DVg`@W
zlIPv!k%RdzxlOgun%?}HNj)`CjOgfr=P#^pK(3rA`%QmExiXG&knAL;N_n`8rqjBn
z)|HBO>Kw%<F)Y`-dY+$$r;xd<(Ji&gRQuzbH89C)ZG#>;lI;Rp&z2MMq2>JS`YP#p
zOSoF;C)9Gwg>+mw0E>{WL00kxIWh9kKQ4r%O8PjlH2==*mF0fhVN+lJ6{Hy3i8aK&
zIX*l(I2QOe5N^+_Pt8;L>X}*L^D3~ghW=#W5j9bbVm7gVbE}ou5ojUofrxn!_Q4X=
z^U3H)r1~C~3{W=d_Cv<i)h#T0%$JQnWN<R((_-4!0|ciJWs|UuE_WeCqzy7O0&J%8
z(b_R8n*bzgRQ#?RYqV9E$QDUco`^Mj>Q4<FwOAM1oft4NPM_+38C?Y<EmQhWg2_i!
zF{S<Hg@z=~sGld#Ff1i@>h9zGWDV){9zJ-|_`H{aB5TW=?kTqEHX#Et_S{_=F&oe=
zkM9#_gt27)W=_c~SGH6iNoCgYedg1p8}2Lf=HWE?iLj#6>#C^=vVY(!3SY^rO20i%
z-1l}u(Nu462?_;43Oa3#K_s2p;I_AoU9rJwKPS<9hjyrGqzD>>bp#I`2Kdk@qe>_s
z7w)9eu$*sB(CifymoPI88wSzApleA8{RlryqcYhK>k~+uC`W%2d-^Ipyo0%jahqUz
z73GhjZ5@y=CRepb{*F3=5u=#a1o?J3Z9dlE6u$fH5yp_aQImJj6ofTJRr<`707|mL
zf4PIOwF=~QsSyH6;RlWyov`DSE-6^3i^?{{*_5H;@+nWb7dgD0qjQOKh{Ax<7%GuY
zL<OaiZUD`^FLHxOK_$`yR_i>OMe>!aq{dS~gqK{jA?bMSjx>h*>+ADa)&?E@YXev(
zPzeWn@lU`6<0~KP6WkD%w$z3}e-JRNC79{pvBE4M4cP~sql`ykIXP3)@I$m+{tUp8
zml#p&k-7~Za08L%xUO$se!a1lzmacT-0)`#%ABL4ds7(UU|TC5D)f<G9()X-y7}+x
zZjMb?{5Q0AIV%nPCBsbdrUaUbk)A__!K&LS6;YOi8dYgL`K9R1ksP2nY0LSs#yLtf
zX)VH6L%q<bEjD<SUXxFCwGyV{=t(dJQ7Ax(Sd@HaP9dEt45mr+`=z>aG+N^+!0#m?
zTTvY5zXUAdJ2`0r{8-ueN)e$8Wgdu(7k1?$aVl1{Q&_)oIS@AbL}t(IS1mQ@xbXEY
zm~4Y*u_w0}keysf=reap?H_|VBGiM0^`pCgZ>GaLm!DExj{51UHt$*BkvXs9wsmC}
z)s)FXo~5RCXmr|I2(1|>xP}f&exgxkN@L7Nd09Ahz7aE0gRjfnYA8~NGcqlr2-UjL
z-EF5mA-s8LJ4O6?6q@i*;=qf-2)iq%SVLV*->^R^Q(MO@#hz015O+KF-W47J<x0YC
zSe<e{7{iZklXDW~E|W$u{-6vR!u19%>&|RP4u$>qM6B!uG?gcdB#R$i{Yx2swww;B
zny|7AUsv|~#m(CVl5nPlY(VEoSj{~%<@k2$<Sv9pz2^Mdx&(GP`c!)++t*Q2n<;+2
zPHv3R$G1}|YyL)$xkdxzIJ`>!1{}^kUOXKP$`oe{fbynE1W?A((trl@U_*?RpMv9<
zQ1u4JM_jpBg?sKE{)Yy?03$QApn-mB_O%dc6+~%t3vu_BkXXqd+mdlpi=Rw|vcg_9
zQLR{E!Z{hB8HTG_6X8qQ%&c<T+o-p}6Q(gGN5jmPX|0AM1&PSLUh8sefY{vY%Hgcl
zgnBW~7W)1;gICdztor9i7-&Oc12w|v)I?2T<qF$oW$q%Dbvn>&1WKbF)DeXn^PR@j
zP)Kypj14hR=}$JJ7}$I}MT5v(|75q)9XX2&N6Gm8@D3C*L=-_A3x&;z%>TI5vLIE6
z8jj?Zotq=#&s71hZDAkdN_o^B3Rpz{I8;<<z#}$$`ka>&UyFJRu^;D{@7Lycpj%|b
z<%FkT{s$td21G3jpOFLo)0byMDS-9Y{UsEdfy0nTt(KXok5dlVIz%Ren&P>{b{E=h
zp}gxmcLL~^0)_DEQ)sC&1+(=?K45N#Ea6|g@tuc+7the=Dd!{PRR+W>d@U+et*5N5
z4`wa22o-9ThZpSehgmR2`l1-)2ouh(pF<FhQTs7Lo7-ggC+X>{`(hL)v0D`H@<?Vh
z(%zzmsGSlI*6kf}M#X{Vjo|wgZIrB9bv`z}O>r)IwdT`Lwtnb~c1XfpOD&%_pHF?^
zUa<g69^7o$0-}|&AP6ZivPGg!E*<69x~@<wkV@+mnQmi7QeV=i?NEbA7N0?ZLWpJS
z^HAhpFzD1R+CHzlKI7x`534L%PLx|$tojc{aqi;(h_v7N>Uk*QMbYr!;M#<6x43_%
zRBb_`l-ctxe+%3GXlPK1QN|uLUdE|5O2uUHJdLq<htTWXg@!N3VZnFk^1+%~SaXoL
zLlcC!x>n0#--yA55v#Rf2)74mvDmpghZ0Ux|3&Epp?2~geG>9hJ184Hr6cUmRTccq
z%ITi<L)_eZG2IxAIaQ`afCyvn5$bSs_BZK5XjFz8CV-Nn9o(jBxf1l~x@j6bE=5UJ
zgu)o~6Bk$K_?F%t<k2Bm*>K>WVO!?t)?M$~Ukvu}@o|k{8z2&qf5jy{{|cV}>3M0z
z%g(L3KwKV`wE=8+%q=tPH#?86N)TvtObxTr%;&i{2=+n3_2da&wWo~Uh2Ky48+G5j
zn4bROzwFvH2;2k4XHiiBX4(ik13H`1w2VU7up{J~4bb}}IryspNv$2F5%Y*aR17)g
zgeB=43R{k(WiI0~pp~Gky{B_()*A|Y|DiXZV5S5E=7HRyi)d>=apHCA#8Y0MF-@Pa
zdwa7^T2;gntk`bpn|@8zG*q1@sXKjJLtZINO@QU%&BB}+ntt7k7?)t$%uGAMZOzwQ
zd*wb<^3dLfG#sYy{2IYmq)?C)24c%*P)y<8q1HBnE}&OWvn^xRrmoOg_3<&<O&$lM
zaglN!2%?q$KdQbw5URcndyr|Jq>>^_TJ%)nDO<9Z(n5MFNx}%(W-LW^2G3JUl2l}G
zqrzCykaeVzWRDppOQ|elNrS-{zWaBk-tYU)AJ1EFbLN~m=eOMVbzj$Yuh9wy9U+@<
z6jV{=i-$te!_j$m<$H`vasNHE!3q3zIB35AK8hbk#Vg+nLfI<XP&Imexm7*(4o5{K
z@Vh|9f>Q;-=+1nXPa;F75f2i{e)g8vRe&P!*`II?&MDdp1E%)By#(F+P(x^WiA(Qi
zo538^fO}WSJP22ae+hH}q+u3hvqR^Psu0-d=~I66UfPR$!VaO8S<|6lG@hpxir>QW
z?`oZ|M&+89@ELe~+X+la-?1%*xkh+f=#LiqH<&#KwqM4nLt)pt(1t6-20}#L7eIxR
z^?Ep>E>fGI@*%Yvd!DrBp8fBD-%=2b)@EaSKu7{i*<9WuBL#4J@gKGM6PxIX9sHpM
z;~5>4=ypG3EoLhr^x>*AoC2b<*+)@+i;Q2pJ%QSVf&yYi7Eb#}dE9nrkIh6)L54v2
z<qn$2#x%2@TYz+*D|?w-FS6o>+Z8hBDy__2D7S_h78@=<m5DI{i}qT!$-qlYKn}-{
zO^v;hp>&w#=8$<oG!uqddwKb`+Ff4_vD+yE4$Zcj!RfeKSU)Ri&hqM%AK+(l&NxoM
zpbbW(vt*xj;TY@>P^@OTuU)ZJe0t?~Df!oIgq7+ej;<|tVA~n^n`#_hfvxR;cv;yH
zc}vO2V0_2}zd5uMS`%ID8D6nFd%FHu_xL~XJc9C{qz==#R@n9fa4CbiCNONc8Rx(P
zO+loJd^PRvcMBUskef(CNDpG4f6#&p`W6TSi2<Zn#Ei9l!rMExw%x-$V)swkNN<G3
z;WkJbVD4U6_Xh+n+6y!gfKwK5uAp1yKv!;gla^EVKE1vyhUw?K)DqZfL1?_iYd(_9
zZpWUD(`^Cl5m*W7E#dNiIEOSpM8KLS7GJM2IL;2nT>BzmDp$V!Z@`6^P(Grna$g_A
zkI5O8*D-hu6A<P0^F{>$JXQgE-lsUK!}F8R2OtM1x3bwkb_1FH-+H(;`aZ#g8Oq~q
zuA`~-099^xiyD1BdA;4yjdo9>0t)_KnS1{a4YakDry=qG;Ns|!@O~4yIyB9y36Ru}
zmE*Uvj@fj7TsRa5`4X+JRyemt01aZM6qf<9-)&Vz&P5ItI44<4s?oFk1Z&aDhcyOA
z{dFj0&RW*P3UkZ81LLKMztutTLDJoHXRQ$Zw2=qz0RamkL$Ctdl{<Jv2D*TCB77-(
zGLG}X5!9d{z28=II<y87V?=s<YlWc#)G(t(QfPpBI6?4m)0)xyV0rHbKs66^>oLM_
zG5TbF^NXRQ%GX%aLCUvBjuSkOQExuKvv8u<%ldCqR49cbZ4a!qS)|2-VAu^bS^9dy
z@jt_7MZ^`8x8OQy_!!JM(4>N*Bf`$nB2n_F7tfNqx~og{v{woYZ;-yQR#bcL$aeu9
zC*AF?gI}g!Rbun>f$CQc%V3)4#*;&BKPsXt*6sJ$<x%7%rw<8FE3f3DQ;v<5-b=Rr
zpcSxAS6O5uJ}w*@8c>2kJfw55yb>l6OJdG`VqT0muQ{K9YW<y-K6zCVh&&pv|8xHn
zgDp~LVEJtRgB7S$ds~^;50=+=!g37s#2XVh`{W!3?i&{2diRW#Ibur;x5j3W6%aD*
zw<5oO?_e)-r@7ob4^O8jgEUZaOc%iy#OOyse0iBhX7l^Q?1Ifd*Rh^JfluvClh7;t
z*22xWCA`~xIpt@%%8P$5#IWmESOgCUFg79=MwQ{v;x$d4)SEuItao)CYkjclLBV2j
zl6La3cI%4$G;LFo1yNYQ>n`y<hiQ&?RY2&s;{waqxKG2o9Nh`UkJKo-9C<>%pKIIL
zt{&}a&JO)VYxdqbbO~Emdqz~5xnS2Y_Qp$%SqO+)t)7x}Wa3V|rb<#soqU>8p&*y$
zo$M2X;#!bktW|EF&s(-_!&8(J;O64++Q?Tc*diw~sYQHo0FV$^dn~)k9S=}pVeexw
zW=k$kjOOe{mAlyQ!!oD8o`%96%|FQ-OGGZ}s_o{p7A%jgp9c=mp&~8i*>AzqoSgYh
zjZ?}D?WspE5vEW2cu%(yU2fR|MOIkPx2hk=-eSq>7Wwm*E0S)o(7uM1uUOyX+*T(I
zLAfHvjF{gAVY=or*hvUS6OV=43S|L`77+o!eoVw6@b-3}jJV+3Yp<gWbo_YgvnZ!i
zVm%%bZ#6PX0%rxnSfw<!i5LIy$>Hr0XbRBl@(d(3wTr2aHoP??R`-i%D0R={`d5s&
z+lIj7hr=hM@(RR@mxT9<orc2nEZaZj33vE;=pIVvVYIiMNeSsj^=ss2R+7|r8C4Vd
zI5X2h(_k&b)F2q681j*0E#rRcUW_12EQDF>KfeLll@~8s9QJlNYoWYr&zL{)8Nj=a
zWzoMJT?;k5tK?hi(-#8czy*1ie3;l7WaZrRs~FK>^vt<xKc|a(F3|A|(Z@LpWo-EV
zj#HeT@yxze7GlQp45RDvQ)?k$#|^X4e8sVvEp`mgEOZW!u|eK6DnX54T}lK$b8sjL
zD=Qdd{`QYmK(r1C9HG>vVxH}O0Qqa4UOHy<sABHeZ%P1K=`N`{wcewM6CzrB9C6#F
zG?Nes@gBL+FMGeMZ>g(RMf>=b4@sj!B%AdWzJGzX1=!RaPaRZ$+m$@Wd4NTzOj*kx
z>r{RFfX@5b3rpi%l8-s|>K@_I{R4=Xifs%Sa>Hz`f)V9LKA!x9vrEg=E)&Qc;?Jac
z_y3@1PT^DFeLp`2W)B7@?NM!n+E-HyX^%(&Xl?J>dsn2$%^S$qmm}VFQ`0v6b*u-e
z+mAsvN9FRbenC#-6OU6jF4}^UKMD(bn@<@WD?|`Zc#VV%gBgX%ddlG@?yYj`SZSPM
zhNub@3O}aK&a$QGS@Uytmt{U9KZ9ZSkhy~)%H(9p&*R*5gSQ%dE?cNyZYE$4Z>kqo
zYN7@tUxtJ1Z|SbnD`phY+C8b+!4SI$sgs4;As7F&J8c<Jjs?N%*U0p^*rj1wm(DBd
zGf;3J%`i}O@VlIu7K?$lhgJm6fjk_(AwGlfSE6pcnSW)Byh%?VEPsZlb5>beGZ-MZ
z6^Su>bn!<a?a=_RXy@H=Ec4a>?X+OIc;t`rV!z8_`<K%B+9G3-gfntz&5h4UZ6H5j
zX9;~ywq|s5z58j=^WJ4wheHwbKmk+jz(RfvRp<3~j2dhBZ!Hkm_BeRt51gmmTY?~{
z(q*?m-|VvdDn?Gim(GD#;_#!??TCsWNYz}66*tc=+`_e1t~i90!{B*y78UZyos4I^
z)(M!<#xt`N3n^bV<P3fdvsYn8jM|C)`Z)g24t}kd(7g`s=y1${2n<7gOD0WE8ooi(
zKA^kG*0O3j;X|cu%az@<CZS^dR*{zi11i-Zzf4N(m!CDMhJS2OVT~jI(M_ifz=Sep
z<Px-$i2)mX(sRih7DNpSkr3k{4^@pOe8#2>R@}fuk@QI@E8eGmQCB^V)&|O+bJ4=f
zjgWEYtFa<7&)@Bc$kPd~>vsM;@c1#Q8GMQc7N1}q+feawvHF_?K!YulvQiXrG(}YU
zOY5rEN=>Bx_=yFAz<%kt+8}ZDFJ!54SqVs6aOCD;n8S0CvwKhHj8~~UQPert&_6U9
z4{b!#F_#=5c8r$&_m0Y%6&8f|pqXv-Qt|;3O2?u_SC5Ds11=<oJeMrcmTf{hQCyFs
zSCVV(f7&vVDqW(~j2t#t#y~tTxj{4Nr0Vm99z}@9^tQJcEc+BPr;XsJUd~gZL`?64
z%h9!2x^XO;GR{f3EI6kS8bA8yk0h!{94h+@0>zl(zaJybFS;)?v~V5PG4((zp;Iwz
zN7Ia2{$*tQgp$pj>9w5E{XqAkB6Kpyw)pyF&!xAv00e^P%Bb>)&Mu(r!BG&63Sbjc
znOp$0oUUX1kdIi!U%!P(6KiS<`VT%u-_Ra_5R=e>@Nxo6>bPwoMr84jl)8}~Me%qF
zqJ_;0CxG;-F&&JVHz6*14%$%YycPcn7ev>KFehiqkD<32Bt$o{tsOOy{tJ3I-Envo
z+aD>v$Y~m~x#bikgY!FP(Sq@@Wulyx>rg79S_v^>WEPc`_>d7xc0B_i9PedNh2cGi
z`->QSG}m!}us@bV!3xt$E%yL0f)zIQOEBNGaz?+)ozB@lWR3uYlB9XP-BwZjaX%eo
z%_NEFo=(I+<QMd&aUG_8yHPS;tvS>)P^!T2L;oi(F_)aG4*%;BF^NTUNlp5zEyMoF
zwd|08Ps4ejl<_r%|MSgQWSoKm@EZrnbDap<|E&mzjRKd@aL<$4eB}WSYyzU3y>-E?
zY?(3R{6@<k3!RUP5qZE-ntvHt4YnjMNI{Of+;C4`K~Q?;kY33?0kb9CmH}Uu7ct+3
zHDPbHu<1_ss1<CR*O(=y5L?3t+JXQ%m<KO*9s_KR1$AR<-81zqoLM2a1L4&W{q+rr
zsa&Bv7;dK9htXMe$baTHYLC=#Z~$KrLjG3v?Y$5ihXNT5li6brxUP{FgiVT%8#Je{
zEV}dB43-+F;wvSM@ARRfneJMt>va4g@Blm$L*gKgl|&ZZ$Bfa+N6{(pe4<qOyu;(Z
z{_39If-gP@Y(7gfu;Bub1ZPgR|I%GR$N#B=7W!ZC{9ORj-G!AoN*s^gil-_P%U8fn
zB4@eR$PkpJ3xr7-$5e3kwk~r&%aU5HI?N5QoSARBH?bA{99(=_6bB+OjPyhLBZQ*J
zU@&D)ShejHERnXUK#6n|DjWBS4YFydVpK_mfdN(E^*Wh0Mt`8X9uE|7ZN*fW|97U;
zPCUST*5G>Muh5&wRa&6fy&}ri^R8|PIY4$hl!?<+3+k}-4kN|E=<t~OX8H%ihEs^j
zgyXdTl%nE7w?KP@{ea%~F`A}@>P0lb$}~+W^c;?K9APW68W^q*W0AtoBU=e^Q07jb
zYp12evHyW`DbPwfn$zR~-hY$Ao??&BZ>ZhXF^#=I52KE)Hqw0}tTcoBQb~(@`jA+u
z#Xn{87jObp0PseEd~SDZ--KubyAAPK;Yg^1O`3CL6Iw3Yywp^{9MC1Nbh%fhw><i1
zI-yAeMR`uTc9k`N+bK+{TH8{>x!{7b%c^RNs{CT|?zPM2wxDI9yMGy5Km|OUl1K$&
z0X<l&>*ZUR*+2!%jZz1)kTu&5IoKpHa$h-5@xc|RWC@f7z?u|Ih{<Z&^gc_GtLVCk
zkRlgcwGMF~oa&(YUH8Uxm80Q+?}+Ek&kGWqSd0#IzEx9>xrI>wN!|pPN%ya>sOv$|
z_C`&=lQX5DVJ)2vA8^UtF4%o3R<OnUM-Pr!#*MKmrIAoO&>JHTql@B7)-64D8q>-a
zEhZe?puf%>_+vI@FQe3#=VyPBqOQ7PQ@T<d2r<%1KRwU@1{%S19daN7HI7p6xhGL)
z*GYo=KxmCr8fkA3-&Mr1)aInLWzGxllqLGBAdE)!D;&wVPn@*)b)Lg4x^%%ca%vTM
zT{jMV%UIxY!t_#-JI`jDH^gFdr`>S%ftW)w_O{3v0}eszdz{OU*Mk_V<EFd`aWg>o
zUp>pj$e!nmr7#ZYQc&Uu<ZnST=aKxF1{_dyb}kEpMY7-dXp#T|jf;-sf3a*D!yduV
zLw+p!4DsYUP-eN7?(^8yWhBjHewf7Xd(SZkogIIzFbcmfA6EX>z9#MS_`7VR<ttSp
zkSDZnp~}`89OW=VkV<?%BZ27az)cWaoB}8UOoOk%zFmTVXK+aYZj4LYC)&h3edoZ$
zM7oWm52&c~atuOK<WNti!5~(D>2YigK{tfUMIAqC4$CA3PB(g~X|9%f{H!v8zM)zF
zHbxLajDRIlz~LV`2{g!~sSPk%;(!o=X93X}onH{nE~X~6@>2qk7g{<%0ZRG)<-fiH
zod*4L&(Hoj43J{p83VUdsVXQcjOos3(}t(BRM%eXD<psTm8A?y`>|TqYzfrww<0k;
zFc{(7Mr&G>7*=@0F^@>IXy5k*LvFc9=<o;H^h#D&2u}4Mi+g*W7kxz)86+@{x3h#?
zzSkDcrC*?(%PCazm0*ezj=)e=1w$xiaf*RREZM3et%3_d%Pu6ghl3`K>|Y|y>;j3Y
zSiSBOC+w22x~Dw0q%FEfmLeFVBEoj+an)L^w7G4?`eDpj27W#;Yk=s5Aa;hYq<6eJ
z!e94rCa`bDzI|74jj41fvpM2`^vU)PU(pf){n7+>7{Ku~1U4q{V%*I~LyYh72cUn(
zioj#+&9F)(zfdZ=h}d6)cm;6$AU%qze6hpz08d6o%t_=tj{X4A|0JWO7gX9{HU^YE
zG*KYDe@2^eL8ubDnJmq~l@Kk7`T4^W(tv;{oLGIp7P%RFZ<2~P?$O5X(g*XtOt-F9
z0TL_zDk@fe_2dVh`x$6>)VT*pGh=V1TUzy<<w*$qgRT`!Xa==x=gvWXETBFM{#zta
z1$t~1pgsIr!T#sZM{Q5PL8|}1XHoQ5Z)FX5oYH}+ZtE(tX6PRNQTv)(>EnKNoRR?u
z1YbnFM^=6&<sqQ1%E&B)Gnoq+-d0e~jCP^yqOe+~DZe=|_)X6HzwUQ9di)M~R0EfA
zs871~7b%mlFRX{-aCC}y7t7*QjQW;N&<4%`$OB+&v=RU#HykNexoRpF?Z0MPy%F&h
znDT}&9TqXhu$7+qFTMrC%~h-jKTo2zc9J&YX*Vu6^}WE~OtV1Y4X<E%gv9ooQ}It4
z*zlY8^bNOx)d0H0^2qu~2;7>QVgNNfivp=k#sq4IfkldGe~|{{P^hpz{~h28Z7cr8
z=S|((QM(Js9A6tnViLAPiFSqfy6AcJq8hHN{9!a220GxjmO}&<ig99h=Dwn-(Oa*)
zx#}PM^!bPD@unF333caxR%ZyHne=G5TG}2cFM+&wmDJ~VSJ15yeMlv97ixX+%q=XD
zShE&}a{p}ja77<H&KNEFYe9~SV?rtfPFa{U2YLX(<|6BM&gc4f39CcgLs-D*(?~!9
zhD)bM&4(ZYjtUZ(B=ck95Mo2d!gsje#G4{OWVjixyvG-!ERLeL+;FW1R@FxL0@Jfv
zbqkn59z24?PUlTMmQ2<HBKD^1ibjX;A!(h;0zt_dD50e_TukDQm1)t=hr8huS_Sd>
zmgW?F7Q#dJlybuH2h^-uq^(494PZss6N?$PdE6!UYOumSqw6IyFUlNgS5FaAZ@Ndf
zZ~er`q3UTOyCeT@^<v!}sHXvz<U_3%RpD%PLW)7iL-=WnJZxUlErjpbE)|X)21_X<
zS(TW?(n(VB5)yPSuowm7VssIYeu(Ld-47s%MAQIv`AKXw$@-y#4J`DDM&qbg1sOuS
zXrf1&<k!fp#E94?Pyml)vMBF^F|Q3UKAj7W9rgMD_du1y($U3Kb^`Ugx~l#<dglPU
zdvaRn2}gwNyex>-i>FM&jyum#mutXey@()#l?<G2rsE~7K->&2@NfhAmhCH(ve>f|
zYn;H5i2Fz>w5C{=1kNtNAwf4t8OF-6r=y025}|&XxFVjk5U~(!Pz|I6kPEw1P-aet
z&v4$(f}cXJ5=Qh|{02LD_RdYo-}Hr0XK0V^;aRDzgT9rNNZxZE{UIUK^Qb;q`X^v$
zVoe7%u^<>$BfKFt5BC_uY@7cJF~F8%PgH#oR)ZwN>jX1#Mb>56qMIpl%N@$;_pVH6
ziHYJ0&7m$C6Wy@C>-35R1rd<<OsymA$lvU7!w=<xFy<|;c75tGBPMpNiO)*#m+^y@
z%W#mg9CT&>)w7+5`AQtYBB&apPGuaNq|_}selS)%o)q`HuKZdT^PHY2udcm{>L?h(
zy>b((oZ)=+VD2quPXrnnp@1KGrxJq&s2u!jS{{(eR}jv7u7Ye^VN(4~ll&&ju&^P?
zYGrG1Bs{9S1*7<bi0hNrFnqq~COPz3N7{W)a881!s!&HZ4z$#=A-ZntP0atU&Aitq
zfm<a=e8DXjs_q-IqS3%2)_SW0t@wh}ol>y(?=;QLVV9Vc+{~(Qv<&ObdBoP!)WhD+
zACjBzLo4(f;mt~7R>w0m4pfsTu>w#Pr&LK$O<lKX4mwc4*W)PZ2!<VJdwvt|eIFi&
z8c3vdN$v$$0L%H%M{v(QA)VKJ>`XXI1z>CQ`H)^%+toE8ZrBf`T_{d6y$Hki0ES+H
z8KAlS;#*8+Kz-EU=N=&lV{lO!y;d@E%rO1c#DMXM9D)})gL9teB46fCTd-xj<&Tp>
zs!g=;SPY{BpKdVJM?w6Ld?n5Hd-`Z`r{ihadllENCO>4nf+$BHTnjmYa(d=AMo`C8
z$gf~GnfUQxS-(Cs$}QVo+_e#HLVspF<~>eeH@UHeK1dOl0?pfz#}A&bwpqE!HS&Ms
zV8$=3L1MTE!D$SJgMSW0M(C&KnqUA?S$z(xh{|RbV)=jVMAgS=^wjm)O>$U_U;fVe
z^^5G72r?myco3q_h#&*!BO)!n<?fc4xA0fZV87T=x#g*|1IPdzM>@~)jq9-|@>JIz
z?psXCSSkY0xCtZQ1fpE03am^7RC$_I(t1SvMZu41DhS0su0mI3*|D_FQTIO}sH#kZ
zPW+U8Nblzf`nRCnGD%qmAn#IaoRGoh-tVyYcQr%u95M~hm4nByAmDtQAfwa2g$Njx
zCa{bnACwtL=Uj#!glb4pL`s~0XL(RWgRA1`mj$#>@5bcaEv@uBsN%W>^Kk~S09ylx
z5=J(nB(L#Y6yaF9!H$kY5RbXX+XeJK{Lx5+s+^5gSx7SGe{$O)#Q!{Il>ZslaUp`?
zOC|l)x!@sjEn@oMTh{G#=WXY09&y>si(M_#TcBC+!>u$(-~zifnrDE(Lf<K+eWA9L
zaua+IklzN*B%SevY+G9->MIvihI)-dd1uW61Nr<V#%=6kUU`b!wyAR=4G_3*37(_%
zYbjfS!wNdoGYnEXp2})&Vha%#MA#*&YN|uAJ*@Apnrac|vJ6y<p{l;?mpu@D={HBK
zlzL%8VhcK5e^vj~iRwg6l3*VU$(jZUA^HdGSWkTib`~pVA4J)bUF}78!v<pbQB9%a
zjxKoMfg1CCA@VPcxbxj&et$QT_J{*HZqBLr0-Q}cBE5vAE-R>w;mD{B574f{gFXT}
z67h+!5M)w|EyL1Km0{#V{*e3J*6nCpc<+LhjWiyx?MRi{8{ms_Uj;49=z^y#(X%Zm
zIT;WEm?kqh_+jh9<wP5viE&x*#MC&~Z9Ad7KL18MhL_-IhSE*0cK)~t_SavP+YY^g
zxFJ5~HDQ4B1O&^<!4}+*SHSQxE6T=7S_|HGoyOp{vJ_8Upiqy1pa>cE^tWH^=HC}^
zd2nRp$1{Lt+<FAA86U+Y+c`c~jgr_k-EQ#Tp{E8sHIPI~-(<eNckr~zQE7P)l4Qg5
zI4~r(ucHE}<ol3mY0%h%(M$jTCh8=@dRHQlUUVD@{{#^Vl9FkeuDG&cSZWfG0v65~
z9%ndN*BawCEr8Wkj-`&r!U^2x&T`?Cg>T`#coFF2ftfXgkxqt|Pp9P1ytK}sVT~s5
zMZA33coKU~SZDM%CPbff>YauvP$UX|pcj{(U}4D?HUOP&Up3S<#-it+HoUkbtedq8
zxl~!^$RVeOC>>Q5Z4`(@8kPGD9$^>Tj3R-4JPz5|d<wGU2C>!4rT1Qbf<T2No8s9j
zmvXey2o7G;)KPEMhy#hxO8CM1Iq6m8NOe_=q!KCfJK<8c4({h6$5@=^0!SMgfCL^9
z6qO{D*smQ*?z;-?l>6BnJS}OIlx!|cY({Z;D~l|U*zNSPx&f#sl3Z8UYT2MYujj|4
zM0~LlExWhc7W$Dvmzev*oIV%28yE*%EjO)h;w@R-#qhS**$Lt21Xm5pPB2<qu-hey
zkhR5ndLfh~d7fw}<WHIDjQUMq>A`;*pH6EKBTP}OFj_fy@45dDL29pNayAb=uKcn9
z#Aa|915s_Se_29Cr7hSsCYhRZsMuGd|1hK9Da*Xf{<0D_W62Jcn$uDg`=e=%bC}l!
zYKjoj^^GU+pac}!gaMm7#R|*<SLG@HJQ03eXy|7BD(2<jqMPbZY2+??+m(8tyn*!I
zjWc$R+j2?#$8NX<bvxF~+8>MIPZZXRVgLK&-qq{ZbmlZchagZ>50#`d#-Qwd3Vb8h
zFL(bnIw_MxW#7JhASXi%pq5r1t%-sfVHP9u9UP=%(si!Aez-!xew(t8KW09m-TU?0
z!1G*XRLFGZ&xefFwsM3ngNA%S=WM+ZgeQ4dQ|5hTm<V>_xlCwb-$Y-K94C;c!IE5n
zYOXnK)->b)iN&2NU=4EZjX_l5m!IF2OS9;19cMmH;_amV{a2v1V$|g8>4=MC8a-DS
zG4yh#de`O^qD-A+u@>BS8q%fZW!M5qpG{6R6(+uh%{pA3h!z^69W81!UJ<4e&aL4d
z+q7i)IZ^coz5fu5yBO>v)x)4GU@>aSZRSh-W!MvQ;5T)0RrO1y;+=NjzH6vCioM-j
zlLgYkEVNBU#`;^2ROblfY7gg%QOio!35Yq`xA~T1{#~1e#>tr6p6ceg7hs;ad57X9
zpYLKUOZ{!=ndrYjh4dK(h|7$CI49ELWlaNScV*xX?1{mD==En_+<{(QdDvn>2_lIZ
zHq$*)=C>`Y@CBIuo+Do@s2ib>=I(HR9)}=iYJ{BjZ@58-^mIi@4>`6NcfB32<Blcz
zqPXQO6u~su^#aX~{*GiR`KUKASs6VS4;}tzQf;3j>z$(JGQ<w_mXCP9jqbyM#_0GV
zm!1*wfELfHWoG1TpEovP;T*X~p{cM}Nk*e$A~bxEZRk^|hC^@GPNF|8XFv16{O`3!
zT)P`KEa2vkD0+<p@H&YV66N(jUzUm=0uuHE0Z@aR0!N33BdlVvbED57dp!c#>v_Lq
zHsY;f?*Kg;C*`nMAIMsZmSRb{NKC=st>scD)J&l<<yY|CP~Ezxc5eKQR~>v-<v_I%
zP3`M?fy3EVc-kPgoz7!Bd0af%u#x9-rc8<KXALZ3EY*AbyuABJW*}DjSWwvq#8Mb$
zL_NWTn#>=bi6!^0EQqD<hxrr~)9U@h%2KSYl|7LtTU|bRm&BW3L0*Tl!q<LSY)-L)
zW>Q!+D;^yBcSsck1O^7qZZ!B11r1U2d`~o#SO;zv#3o64ELbu?&PQjmnmOjE*$0$;
zdJqNirfe(2K`ZF@hW*nPMoigvO<JwyBoVGvpONWz89f&4AjPRw)~H6j-2naz<WhTS
z@~wg#yjh)qIjJ)Hwe@PbAU8XF8+yO(hI102^$vnUeiPRlokVOb`<AvL-=(Y{KsG8l
z4XCxc9B7Y$jlKmImdP>HE)p0$;GqbqjwUPYAXz#t%}Uj27Mio1u@A|S>&!sf%$@mV
zjLP!xmcp0a>Di$*wCn~f1)$^q6eZ%3*YkU-*ohRNA#I5F8Ep%{4P|H3klE*RGmzO0
z=59wc$-;?JFGj+Mb=`x|8h`6>D=arC2$u0eSReY;U|PDqI_5a&wV^fKATJrZ9H<=V
zYkxH-2om5ExmImT*|h9`7Nf;iit<?K1F-AtS<;(>&LLW81hf{N>w^sfiwE3M%MsnC
z<8a=Yd~#_f;lw(wFb_19+w)!eUH9rl15y_{aF*KP(>qwwak|PCchz27)_vjz8zW`I
zI4y`KbN(%sC1Mm7^rKbpDzTj={2x>8z<DJHK(O_ZSz-PY?2>f$1X_^oc@6VH6>}5b
zf#zg_u`HIcz-bVpsw2J6HPmHK7b$>X)s-pR@>HmhQ!f8qZtkDa-d9CuL}$*Is2&PW
z`&RW0u}C-nCizJ5oaiMfg+xh|JE!*Pq3@%+L9efiryFXy){r`-4|uC27h-P(`^O7e
zOED5Zyxm6mHHnP}jT6;Xb}22Zy60Zef@EOU8A3G}(B9)e^QxOBZ)_=i5QLGqCp6W`
zR!)Ta>vwgHxw7=SI`Am6DeJsb?~ea7jT&{e^qg2)f}R{G7H(9P(sw!f^C_7<ZmUst
z>@(V<yhUb%X3v$?BfmNY@7~Fv^?l`^y>&wc*f!3?QmXWMWfn6EDF9jiDUWysDGV(w
z@RTC?s4=<KIfMG&<>vosXsMT5f%7~mB`=3q-WMhUN~MfVj{o}7NDQcNNyLO80GoWi
zr#p`<gXi-~Wf{MYokWj-GPyKNzW=g0XI#_RU&gpE=zptD?mBi@tn-S<{g0ypNKc|-
zJ0vNqp<3`f`e)XBd4(bdy%lDJKBW;$m&QLLB#{#pOKz&aR+oA2c2vYESy7Y{4d93%
zssr;AW4PF`BTL&B6JVk%r~e()$)>eg9KQg6!5ua<I8`+FuT?YA+RBtFREF~U=+vCv
zA!O^A$@L+3Gg959DaQo(H?ByDaC-9!FGOd8zChiFMw<rNfXXMse*L})+`1kr{f=w8
zt#^n}poi$TMQ6f8p`ZwD0b3KM`R{86XETJ>K}RO#;IXPvc8-ni8NRG%JQ9%guM5rd
z0WZ8f-`9OhEeHA}kQO_HQ;lVwX@|W&XBvFaG)5TJfJ^y$^c1nf$8LDDhSZy{x12;C
zvDA0DNCn9ZWmdNCjQ-XtY3|O*UbqDyH6b?~d=0&m*;aH?`svEFWkbj=AJKHB?o%BS
zAu7=aHN6my<ie_~!HisCHwPzt+RlRvA@Bl2MP}si{&<0+oupS16B|U7H@d+v%&CWD
zPRZvNQ>l!0uJ>o&!|H&OzbP6NRx|w*GWUXT#km3nwjt-{xrlfXbcC|&^3iaL+S(#)
zAOF7X_#TLWe~p=nv`Vg7nBu~h3PjPAWW_nr!>+mG*q)c3dEg|<7sS_=<U!_*QdK(H
znRq97CpLQ0>b!KjK%rc#-B{T?eeuV&|Bb1U^cQD#S-mGa+pQs|*+nAD*#~*6M-EC8
z<lsmIcQYwmcGS}$!SWqE&*+9|`QRU>k-HgYggSXRd_Q^drnYyFIwDrn=fF%paww;1
zoa(IqsXY8a$XKM!?7qj-m31=#c}j<t(scy^C%oo)Uh1FrZe3-(ka=(*f;(Hk+)6Hu
zEt+ss0A3vsIn0$w^S6}R#6zs~bk|V+xK^fh1qe5~`x?NOpxD18C+(XzuRJ)Uh0QJ&
zGh%FR-_`9QgB6k>?3E*|0Iqowpmr_RQuK8wZ3y6mV129;dJd~b;v<lfajtZx7=ojW
zb6qJ)eV$drq7C&}&+72h9Ir{Bdc^Bl)Z01M>Q9ASw65xOA`W3yuKUz?)>5U^%5S;h
zDHRx>GbE?J7aM)+!w?Cqz*;eQ%uJ`dh^Y9z5;h4?_DqXI)Uana*wEp44|vCai*K0~
zV6E`CRh0=dz7p*z>z9BB)Lk;+{7WG)9fMLgF^9Hb!vI^~MzZdS&_;-H<{Mgh?EE29
zQ$xcmRf-|Iu+5Z|5oK4UQs7eZRQ-f{tM3|$4uR?J(#=^&+wmqIe9mKvLvH}{L)8pI
z703@EkOPy*7C+PEg|Jo8b^D8PR*H>YTFSo2+F(K#Zp0&k^lHSbg>D99@ZPaA_Al!X
zh}`ipk<NCaxSgb?j_`{4)sxQvLu`AQGoNbp8~Ud35t0^oGRLMUkO+KP=<qyR{+zpb
zy-o?Q7fC`&o`m-Jmv2#-bW^3HZF{OZWQB@-B5=oEakJ)To7+DGZ^oYE&~A*&FX+q$
zQB1lDBwCt=#ir?h@qFMBI@FdsC+<eoT?&nzu#<SQ<7v&Bjunfb6|j2qK~q{YzM;;&
z)C<}Wr+Pmb6*E3|TRAlnGku_t(dN3AEBwQGKZmwpm9xqn^^0A1t@P$-?Yhbfh;j&K
z{Lwp&gpz2r1e0RFePweE7+SEry*%v1?=(2K0qSAbvch~3=4Kzle(kNp2gSv8)toOM
z${R!&Wj9fiEOhvoA<gv4_)=Fa7ER%5`j<`jLUs!L@<~}S-0>3s@)7<*p{<FSQQIM6
zJ{hnOfA`<#>p$+|whvR^)e2TS=T12`s=UJ>7h}z+3AWKRi!4Mm`f^Qu(fRXYiH$uV
z;sHefX`i(6f40oVVR_rBx8N#~ug$I)9aY=92V^YH#i!xk3ZjBX3YFHrrQsX(r3a?>
zOXCaQQl-hPHu(e)W5^X|3$zFkg18hSb`M{#Sx3uT7JoJn+(}Z2n;Xkl<U&>sPZy3g
z?cE%!)C>%JJZqH0s7`$Si5<e9@w60<otQVD?<TYKP~;Eh5de~H%4Ce%oCA|kKKjYs
zJNta<Y1|@^)YC8jk4KSX?lqIBH}U}FQoHvEOnL}zACPIYO^};lRojlDQ=?3IhxO_I
zUz$fO?ElRbrI?%vIb7F<&x0@!nsP8rMLuCwqWr}{I;wr%F#ymO`Ij6MiG2Z+%a{Sm
zCkpMwFsD42nhKW%amCeReFV5igqrwUHy!+3;y;^iS1FJ7*Lm8Fg!$m@|9~I%Z0q>n
za7+5t)3ip)mNFCyE{8&VIN36e!bcsYEkrki4tBJ+&xPm)G^B|4*qE80b{U+=;W>}b
zXCTkJj#crW$LC?7!_k5Dq_8b34~8;)!ele<KP*@2<gW+64B~D~7imcRMZ;PmgGG_s
zb`;Rd)v`7Awzp%b2!-z@+l2~O&;uv_A;$*g(2(Z900JE+c*~#z8$hRNY7w30Rq`Oq
zUgWTsoaI0@IEf*&uYX3GkMaV#=Z6jd#e$-t7GVBg&you32=i2imUT&4QMbt%Lc>tY
zU<IB7_E7=!A+RLB@BI^-(rMm*AE_RLE73Q=!V2mtlNE;7xlNu)_|0tZcB`DxHt)av
zgLue7*8Zj2yEQyF+@l;je#nxl#=*-I?9dvrMIUyjNP+i%&CspUl=7H2c!!3}aQ}2O
zqMf*Ks`q-$>;b{``C}Ebu7bM7dMr1hn*btpXsbZul*ty-G{%ZGdFMhpuiyLv77~pM
zF{uNT&Ez%PSCH2mhMysYA6xDEXse~Oe$$+^Itnr)1))1{t_!O;A=icrAx3a}n&)Uy
z0qaJE>mEe?2j%M4T`uI=((HFFC2r$oN;AM-XJFBo_D#$o9qP5934<9-p)So;V(w_*
ze8wtO05M!nPSQUSkrh^sGLr*?2dxw`y*o?NE8hb#7<uI!RyaTE-2~Vry8#d<^`N1|
zMvl0k6o(s1RVOj<BTNK8BJAvSxj3x7a_N0WIyUF`BC!WBXM{B{l_i5Q?2m8&^k@Y=
zw{K*Mp{6EE$BO3gIHwE%_XJCf{@#~cG%cjxbU^!q?kLYjH+bne3oV^{i#j!KW&V+t
z7G`qq;<tdgRyW-IKJ=yh7fLG*NEWD!h$igst1D<K7*_4qgy3e7Uy|>zA!V4Oi@W>`
zgbaWBH<n^%s_}4Y?oL6D_@LJtvxf0lL@LpBV`(iQUA2~MWoJpnm*({4=#Hw&Yam(K
z96Jq0wnCy!p|0xhh>MtlUk4O;rCp$_WMEpAHnwgtq5s!D$}ZgOos^8l#$8m5{Fs@x
zmbu}7tv}%6bT_RAgf<3%_i8*)2mTI0EfE5z-=WG7=^(c*cNBYQFi@Q6_Y#KlvkkKF
zRn77XSv>$O<h#f_l#oT`tbi>Or;w>H{zM4u!gO9?5SNNkbqVA#YqUm08`mQW@38Z~
z)TIM};%fM<xBSsZ3XsF)n!-8-eZvI}meIp-1Muv?khC0D0Hf?DT6*k?+O|C0+O*ZX
z9iblzJn0N)`FyH7gYp^pAE##m-qmF-+`>|Cz-G9<w4l{d@mn{j)qrid(7*8BIhSDG
zXnF~{i}h+3dBX%oRv4j)&GL%K1m$<PBWP6QA4G?_RqM+mf<c0{?o+PxjuhCL!CtgU
zH)?1eZV?Xpmb1v}<UIMos-OjjZ4(P(?vM?>LVtw6*Dj+;aW||)1;B`J(0z{@Y$b9V
zU~ou#!T!l*rVyPzdxxCLdD%o&3Vlr9fZArpOzzNc;nZ142eO*qC*UV>RC9z2$Qtz>
zMixNW<e__$5*0sr7LpAvqB}6@@9Pv8JDBuOq|?qRLLcuUWEp(Q5SiQ`_$ay=T)0ys
znS~9#G}5asQbN7sZ#yW;9>{mLC_<BQU@$@C7%X+@H%de;lud&_wm5r~L7z9_SfA~v
zaOMpSfDtz;@CHKINYJttlb*s|IAsW!!3<9y9@Izw5-c*s)(fGRgIC4v5oy7k!i_xi
zM%6>ue)D)PhW}PQBxI1>w?1C=72ZQ3FAf#O^OqbN;fC<Z{6oq%=lHr<fJ+nuFy95(
zF(wR<!+4eJi`-#BrSZ(SA5qU1UtUP8fbQOQXuD^9z(4!neERt}4S(!!u=9l0*v;Nl
z8u;94Sc%Xab+%f#b0<na^|d?@-JJ7cSUcB!Ca`g0F>Mr_jL;*!fT3+{?{Q=iH4Gdc
z!0H=Py}`fY#}8F~L8~xWV$y;Z8FxkhPH^ZZsPTfVQcm~JawM+B(U0{P?ZvjHj}>y;
z%5*b?J^;w@8lYN=0~ms#kNh4!B-j4O20P(ZTqs^sa8-Tv5yaT8Rd@3ERcE7UyLRqX
zIsfb3_YL9u=Z3rIqUXg0(WiRT7?r&AAW$^XL*b8?VhjV&KFEJlS>|4#(~~Y#YhueP
zRip%M1?o^@bl;!7smBhM0BTpqzP`m480}O&Pf6u2+vIYK=iPy{ZzwwT4{$Q5IZ|DC
z<5IV&$jb>q?JzzTsl|>rY^+N~F6UShY=(P~4S&!If)%S^VQ38x050A6;&Qiq#HrBT
zg0(B|#NIxovt=4_u@I6#Fa~{MlIa0h1C&9s0<$X7wi^oF=%77#dGvHRW@Ii2b4s#6
z{O)j6z3P+`)brgYl_A4k_~_eQU&vfV$SNQa=7&IMN;oayHSNj^DPtE){l9+uU!>|C
z_b-PCeXfzqpKs@N%O4(@E$8q~dpF!Zrxf?TUE;6Z?|T(oC7K7e4iob)yA9+%-mBoM
zygt|?V`F~LBAW%_OHI}g&n(&d`s$8<j~VRA+Iwo>a;v03SL3#!9ivk{>z^)=(b_xy
z?_H+K<~;@LPc6R?r@P(De)GNjH@)45ooXc9B|EKxkMq6NmbA3YKOc;%_{e1+eh^SB
z-94yhG0o%h`vXOAOxKf@H8(k@{!&=EpuxmgPZ54gQXC*#*ZD|V@l?dhF&}z`c&;~k
ztjiAd&Zw|Ku(3k&v5buh2-H;g5q^7G=j?g8k5>o0(+5I3;3ORG)L{?^ou?%^Y6Zdy
zJ^>Vme6CrG8L8f%l8@gnLQ#UDt2Cl91dDfz)Q*POk#2Y?*qL<*-6}3gD(rCVdY%W_
z(A3YU@1-401cZ$KwD-4O0MwSGN9d8~{`C?lyNgv>!grjPMUr9MisyxdO_x?0DfrFx
zoLe}G50@Oh9&w}E$3~Su1Se29d60}iu^A278bCb^-^@*mJc)W^S~ADX7YcvZi_?9+
zqua+2`tN4XZ7GM+oO?p>F*p<a?fVvchupx2qSakJpUh93NOx(Gw<jDd59ZH!$Fv3y
zK(uQ0>48Dvg3YPXO6)ooG;LCRJt`pnnW@5~x@vQl+Mv9$rS%g4!%baPW{n=5kfzbq
z{jIjB_-UZu+8_1ZzwK-KfQ!dHz!)ZvDgeUq3p4|8QNR}yZ?-@`RIW7|!0Ui08P2N>
zs)@u%&Az2p$*#h<Q>*x)=%b_~WGu5a_o5q7cF}-GWp3|Bukun#Osh0Xbk#fn8!})j
z54c>K=p#@*JO@xTcGEL6%S~jJ`gePA6EO}J?y)uvT>i6xK6uuY$Y#${yk+}7ZZ&g_
zD#Bxu9vc+Wa>+J9Rs+6=xCTC2$<Spqhrzx!<Yx1Bru^4*psu7v_9_nVqAty9p}I+k
zsj7+vc*~-@qxb;|<LmpG$dACYEx{$HyL>we5d#X1&0hpTh`H;G=+=b>D+*k@U*ahy
z5PnULs`VA0lYpUjwIQA=><Fe4WF5`rYeE>mpH0tWz-Gz<YJ=j3si>p?cOnxw?OWlO
ztpz4nr1ZWEn#zJrfMW8LFTR?*EHmmnxoI4)#oS5uFpIl(fuP%b@Vr9E{OEjP$afc;
z7@Py#Df2bh<CVE9>aTmMTUOq$9X32Z(*nqlVmH><lE!&jtG%3J(s9ZozgOGrI^=Da
zR<$&sG$77_rTw5eO$Gi)KS&$qZdt^*0hX~aQ2N-692-dsJR>3kBA%g#%lXr;Dj?&H
z2Js%4sI2RI#BQKvX}Bk(w16uYtDX&Cfx{hlgaT1yyC>xf%!E&IAS-gsUD8jIA{<Wv
zRUM!IJ+P_kD0o~q+WYGO36*q7r{g}*YIXhXUUj`JaxRuiIw10P!drmcooawIXr!X_
zGXt9F6-y1B1<9=fv=S1*9T&~~46pZj^rcU5QT>TlT}=xBuD~$%v6b4seY>4|x3iq9
zo|*0cZZmz?X4*%6?5>*rp4zUiLpb!SbN!?u&RdF{kl_kH)g|AtitDjXHflWhjr3lG
zn4@Zv+Wm$BmcS-04JcDHe<IB#OINVV<KaftF^a5*b#WKDRx9@AszvWQsT1Ds0k9Dl
zhn8?bF`ZEXUh{{5&+>koyoHDxd9V3DW;D?Z_v<5aPiyyJDK2jC@s8RPD`SAFr#X9j
zz-3VR`a`GD&R#$Bu%$3O>_@FHg+CS=O2IQzmw70v_;F$V7fb!yO0@nf8iOU;(OquF
zg?r%EU~YVaxKi;)nV0OcI`GI3R^4mO4XA|CMoDksoLGPvszKRkmDG#7X5o)X`|Zw<
zV%7JxygNf{>G?KF2pN;?o*gy`{vKR0e-!%5&84}{SfBwI?PzIdNtVy7+=_edj_Qs>
zOI`G$>NoaZDSd8NB~Ot^1_m8h<Kh7z)Vs{=f%!r_2?5usp=kIc#e}rf*df{Z#j&M_
zMlP3F#k6XDqfYMc-?P}NA!xk6X=98TT`9DkWm~JoF&mh}@%hhdUkT#o!9(FoU&d(B
z_J=QoM6CmY&XYP)=bi)qMhnMO1+e$BxxH5MF|N&)MQwL}4;)#VVBpxq38t+MY_(@Q
zzZV5Zdj{D>0m=P~bmp`sRK&pqTNN4{(6z0x*n@P$juz~ypQYOYM$(_W2v7CMHyl{q
z)mja!YA(e@N6G5U3^O#Q@2XKCM6B@sMTX~+rnSgWD3{!Ee6N5Z*+_CunA8~aKF0<B
ztwSrcHU8!&!N+p{(lUKod63pf|BEHu5C2D_F>QFXjc9ew*i7WMciCx?+iH#XQaNs|
zI+_*@5=dRYsKN$52YV7whT6YQdV2iAW)L0+X;aM{aExH_yQ9YZ1S$Ij9oj9D(kmzY
zt(tn)#v}WfsT%{^JnjJoP<FMtZ=atG7CF(6fV^OMGT(iCP6^SE(17!`eHG&OH{vIy
zIgfw7GdwMUqhvng`5i&mE@_De^9svbcaAM;Uc06wEvkUznwItY__vWx@|@iDuVsLS
zgRR!}0Y|i71jgeNI%p5x(5^J&P7_coA7e)ICX9cNJ1&v9<k<2s@I;2;(2-9LxzXH%
zHtMwNUO_-=Z}w8^d%z1Nvy5`L&s_SH%0>*GQ+o0i)d92-E;X_^WwbE}2ka$Y47x9v
z;b8J`B;V!E&0*L-+e>`3l{HilCxqz*buDU;okZXgTp!5Gi+mI!lG(WykS7I>zq{B;
z@@xSPe-v7WJk}@8!lH|PAAJj?|1agJKWZC|%dB&JetvJ&+5o>F#4JfD^@fAKMRg}+
zwu{@Z$p1AIrhf2++SJxXhF$)ylJgm=L79l5pHty7V}Zu4+yzVd>*{MRd|t{l>JxJ+
zSdJ?>v3k_|n69-%Wexnh_XT!w1>Uq(jRFV~?(&Oa-PX1sPGUMi3+ji*?Ij`t`7;x2
z1=KM~IiEDXGoMefC$E0U+pD_bKE-}7^RkR)<XrH;D5$`o6oJ7vxTFS1U1hrzi#%9B
zuk9Fa-MG^z{FvI*9pXZqBLH9R6@^QvMruH%mD9B=UXtVh70F3Aaa9cP(3cq~>IST=
z6Vw`)0TH+ugZZp7MeEgI8v+HnD|D!Z!?dmF(H?fx71D#KcuaQ?Hq(OtfW6{9Vn_rK
z(>eFw8_(J25r<Orh*jv6ZTqC<ShUK=Zcz@QTN>HMz;>xA%(I|<Y``)9Q#IiwPH9nj
z;4<}w<5n+-<4BR&JiAjMWEenJL=<mYwyTCi^5O6p?}{?~25i$|Ywp(zW{vZ53<^*(
z2H8)W_i0P6pXyyQBeVs`3)ixOl^35yJHDX7nUPWLbZ(V~vH3|Gh)uE=X7(H--NQfl
z#TQKj2cWaz-%BT6y6@zcy<B`I=LR4mZ^<zb7`-)tGPkP#{OfLPfxu0Cht*XCIhJ>Z
z_^IVwxlqH=)=RjCY%snoMJ#|%cT-X~P*%bMTR(t?O)H}aCXlP0B)NaVP?%DsQMvN@
zbWt;la;WQqR)%2`B8h(D=CY7N^5@Ln&ioxm0YofXOMaR~sK~Q0ON|rCJ!)}pLVe6T
zOg=t7cqe?CDPb8NJ0WjnCHxHjM${m5C?Tk<){j~2PD~E2K#e{D_WyF+@G*luVQ6?Z
z`}9$Ji5uDw$|=G_<B54CtAaC*eLKb25ViR`R-Qdil@lvzzbK;_Rlll#RFOb}FNQ@6
z(77j@RQ0<h2;;3dVbWFnUYFy~@w&oJ%OUBXw2FaXsbccK@BT+JU5mG6Awd&*!l5r<
zE)AEtf-MI2ElU?{-EXhvjx-crRcI$FZn|VKZXnqD=v&m(HQTvCHgC!gDnp$tZhUmV
z$eqk>XlB*gi&{)xHU&#Vn~>IVA2hMf<i8Z!)oV?+?jW|e>8C%>C(o^<tKa#~yZPQl
z_SbVPLv&h6_|Y7+iv~H2I@Z1>&7hR_bS7AcoCM(^j1|KX*qFs*uy{rs-eq#^C}S)%
zxQjNRc_cg<&;~=pZ%vPkq7CmE$SPoT#b3SuxR=cES*yv%AN2{1$Gj6nwoFGzu9jP$
zsn1c`7WAvz4N#f@u{cgpb_ta3mJ^O7gL9W__nG=jh0-j-!lxPp55zO#TgolJyy0oN
zKOmIfeq1aY?|+Y$Bmf84mA}q8@;0|X$^Zv6Y?_{c%&()w0l+2H?_(i1j+Z1^h<LS6
zBSB&=J|i!Ty9Zw*FoH^xl?placrfZg+W@7JhCi2^IG22Cc=RpzKv%Hk8&AV~N}r4)
zd#B7|%x<A2gLqT@+e`AJ#l2Sa;EuKD!tZ;isb=Bjp_3&C=46V`rOUq%$Z7J>z=qbO
z|NMW!PC_k>*JLF+MO`S*<~r*vdiDQ~vaC^tzI~^1nvD;uTTVvN`BoNwgYufl|K8{S
zQ}Nz~f0Fypi(rOv$4>`VMQY+h_QXuU`YX0h#aqx1#(C_)mT7$Ge<%xKjL0rjtZ*$(
z_|e?+3DWEKZkzh<k#0-TY~kEEYe!U{c=(hLd<xB@puqXE8w#5i;}a{fZ&Ed1d&QRb
zq}Tf<R8d3!f^{^+Blk16G%aTtuID;wj~--@1Zy<+aHJ3A1H#5Fc$E$hC=Q)@?qWcD
zszF|sPN+9FJw8?bKL7S}Aj&}CZ)CXX@JRel^yq+%ZA~^l=iT9ST5hqm=+L96*WR)Z
zSi_f181}oV;W)!z+Kla$otc5j@{6`q&mZoEg^cp9iKr6v$vecdzg#vf1->QsXvvaa
zodqpnu@ip%{<Fm`?-m0ER#C6QrFStHwZ%pkm(%9d*=|MGpYyv?4~kT({X}wkSOGG(
zUf<`ovq3~Acw{PBIUc&Yp=6`nKTtMo)rz|h48%YzncWbH)+Jk?Jq<yIRUz~xQlGG7
zr5pTM>N6qWy#3vO0LgGXpv44#vh%wg`$%6D_Wk-+NTSn~-8-Eud5>Q!)~#YCarp_h
zE3}<NrW5nudna;2k-;^siDU17428vTF#m1)8b7el7>uBTV$8y~TYl{0yn6pUZ&mKn
zj;UubOpM$9PQ3QHc=K03g=qIL7disxq*KMAiF*ckeM79Ke{$eTELQ8RJ>Z19j?aM?
zjhAj;*<-DSR{4D!ebU;gHuIkfk+DTk-!_}8(?{U4JH(13cZlgiSFv4Q<+rc|Xeg{p
zgB|46%6}g{`#w4%VwM@rXGUL3lD&1#tXTGk=FLkng`rkY2PgC!V)}ij3wofp5FEr*
zur=H}_3PKU`6e>tD~zkcI__ZF>DVOS02P5M$D4BQ@zf>s!`L2D>&NB?EGX`90nAlw
z^ipzmF=}NoxhupG`?8)#Ub<F{M1Y#2p#|-0D!5k_uoR%;m0y9H8dJ#F!u(2h!)reY
zoRe1F=H{e_&%a`V(5Jd#vC+=Ne6TMHlm9FJaj20LrC|HbjVJCYM9W?aoM}m~SZccR
zN+Ikbn`&g1KZdP_t|Bu!I8``U&@>jPo0=LZO(`RlEH@LqRHz^8(9Y5kkFg$&v?=+(
z7%NTOEgsW8l<X+jZBHG5;xv}i#Z$j4sIEBDM?fJv8G_^07M&VwmdOKd=Zx>Y7U3HW
zGg|Alub4bnzqve2euYKyMq3*J`Z0k*PA@yZoAW&om=2;gY@HHdUG{?S4m7AZV1m(#
zlo0H)P<cn+=o$@ZF{T=eZ-O=~!({>5?m95Oc4or3(w)*`v_6M|uzR<adSF*b`|mV4
z?SrbmMPC4t6nm+-uTQR!fpxUZ679IBl2aa)AS^>X#wZ&6`@Vc$f>V{jqAZ;He?8yS
zw8R~chHnphOuRjp>{Zf9eLo^)FkSwh<sNCAp<pxnaA5Sc-^ad+v31agH7KF24&I%L
zVfYcXqEBl&deoLxffoag`tFl<ShiP+sm3y4r-xsDwlWSFew~^y&-vwj@DamWT>RQY
zbL-t$q<>rj*03G<r1h@F_i7^^Ze3|1`$>=_4pz7E9HjoiC1$WYq>@P2=d9SeQiR`_
z<^=2=8#Lm0!+KNequ7N&l*mv!gWqDqyAlCp5$MPzK|K;Khb4Y{f=g|Sx21Kd7MY1?
zKy}j6$He@<OexrXA8DKp9e<jRq^^@D;*cDj{)02n@mLzZtADVkHfz6{83bW-We$23
zlaPDEGrWsG@1nnDojP=&HIw^-%u8NsFUFps{=`cY$>VA?(Z8L0Zm9PQUp_3YVgd=s
z&iRUR<@_YA!UOB8_oD`oxU_)++vjA$ZOm~Q+PNFoZDd0E6{z4i;Hei>T+3N?<%^6U
zg*JP{v%|;@7uOY=LX`x~m&C^eS&}U~7?djMISc*uLv&Lc4*Vu_rq!#a-8wkI0#Lbt
zyEC-L8!iHI7`8dSBvJf^zkhFd^|oOha}gnhL5#xUv-|XK#X#iL0o+vd*eb{bxGO_z
z(1mb4U%hmpieo>T0+7oGENZq9br!+>)P&_%VRG^TYzjuLBAC|?GaQBxkiWNw#oFG+
zPos5l=UfPNJNDBP(i<W{dB7`p46zBt7K(b_y*CC97U0w!8JtD5hj}`U_6NqS`@Z!$
zhyubQE?(_-eJje-lN2yR-Z&r){!eA{>1vlF=x;#3rl5L8g8pNm{H|4#_gARU$2k0a
zQ^Q5!4~{YeQH^eKp0>s%FBTw_-!TB-vtm=Q6Z9ZS#J>V&O8ZOq^$RcPsai=875TjT
z4SgI}n=>h<<O;CiCY?peq&jz_8Q9<yaH&}PS@_yIEM;)$&fvqOs-(h%-MKOboO$##
zHFfo3V9CQBSHslfa1X@9MaQfkt@5O)-D<vO$ddAYU|sM4j5{jT=k-z)fAxe8I0lGA
zXBgRN8&wP)NvH=j>aDNj(^M-M6W0mh`ms-(I{;G|Ls-Ep_{y|Fukyw=&wgMmMGon{
zM}gzSV{RPHwv50`foR-nHe`I%7Y&G^&Ti~M;Oe3MM&RG81bAY4<}+nzgV?4ImI}s`
zTyRab0;}AZ%3ZkgqivqIercE+bl`N4%<GPi9+(dvA6-0g2UJ>K_K*f`SytsS`lL4~
z<@AK5Aki~JrHkOA1Rwnu0CeqeyACAduF5zp>XI5pg*!W>hTr+2ZSFAKs6J>Z1D9Az
z*R9y}-!^0s&_4Ty(i?H-Nd)^kv;CLv$xuc36#e@^P%)JB2hQ#c!>vv$r>R8~_m<1S
z+HBFr`hU?S8mt*$ZeRuuKphw1hI&4bSXZ+RGF2bq)S;v43Hx=#f~cIFkV0fX`75SE
zu-7Emp-2?o@bwvBySJAZ^(fv$sE5_>TxFV<7ve6FOHhr$!pv<%YxFEcCr)FlPM;6j
z;t9t<U-5s(1f+yo*TKLdahm=POLJs$EXx6P!|9v#`$?h5(~#y<e$mpGj~?6e=$>yU
zoi_f(!mD@WM#7#|(DmN=q_*j|N|Ba^k!`<b4Y=Pn!sm#{pJhO5Cj|LIpQ>p;#7T~<
z(oGbTt+iZm9o`YZJ5pI=>&v?t5dpxY0L0GyuU@i4OSTT=3)4tPaqeH)&o75@IU_y`
zbNyISHiJ_EXk<+7#WK{iYmoZ<{5khrd=CvYY-yd=I>ON~{h7hedu)7$msVHa5nFuX
zV43sreEe1Rs2-#C0$Mt&_ComR3{`KgwN@9y*}Gyhj;;=dR7!5B<*Z!inK})@FwmQM
zDu>ca?Ro(gnV@I*0@X~vNY_?Bg6u~GrxF!Rz)@m9<UdZ_3?HOKov)o)x9GO;P7o}a
zqQ*^8>SGguHjiab{s629Fn#z;dds^^sWUFnXZj>8rp%ir4I7#DZE4$Z<}reB9O#86
zdMI86Lj=!hqDmTB3I4M4BQ1zgTyNU5EmtRT$q9ycH!OV;7WjW6n1wdzF5zieVI_@a
z+q7WUNSHY*wm`HV`ZpNiYlC(N6hgbe_Mr~gHttgDU@Kwy=t<qls{G=l9(MdWWR*^%
zRxM@#l(jI}*kZ84`Q_Q{(IEGA5wn$XaiKK}x#k=KlC%|~$C6<K7~qMOEe!}#FnhrH
zhQ&G+fhBJb*S!NvtMG8id%Y2FA+Q3>gRHZ*g&WlEWdf_ABtHN`4qO0CBs>8*Aw{YG
zJe%sYz!>eHul!fM>W%o+tAVIS;9%hBy74<So&~|XxrnA-vQh5vOo2106qm5I;nL`~
z)KN{-yGS<q`OWwx6vt$HX}hrv)RXl<jw9HIpPO<mG6V~4LitVn&}*~sSntjNz>N%t
zWE=&{Y#qAYD&_mshjXfHPf5>;#C-vZb+ypv0PfB3I~A{%q=`N^4uOu?1LGaFwblH%
z#Yf&<b;j8NARR`3^P#E2E=Wfhry7o5CkA#-UxKh?bH1nVbfp8-Y>kY}=jZzw1D8uW
z#dHfl%jL!251xeI_H+r_vwbAeZd@4T^7!E*@0J1+*<8m6@;yaMYq|bz1hb(nX*W}9
zlY{F<<xj{_<Ju0%*UFq}*px_~q>_ieIUAups#vigq;3JSd;;~$uC=4WsGg<JTuNAE
z5>SpicBEx|Ix`Tu9YU=Ly|AZC3BLKz4$$=$s<HgkEFQ_?QOUM!Ej?l^EWwe1rBn{Y
z=Fi@{beE<Wq4x!5kWWI5Ah0k}UDKU_VU4wEm1`<#OclcQM%-D#2|)BeZ;V)TC3(%A
zj}t3Qk+ThKTN0X4T`=ksY+;Lt?m8p74k+hxU1vrd&u(Jz=8<>dpEY-}A-oRy)q&ab
zC}7cPX(ckev{4*+)kaI#J@4(<VqBV38*kS+bt7Tiab;R^1zH1+2Q5%cjm;+5z7*ea
zT(o2AaCr32jzu554|D}IjK5$4*zOOlg<$1w8Mv|<wby%j?qBt%o2M2$RZiAGOvsj(
zLn|w5^j8Vu=KR3<BiWpw?$iVfetEhyS-F7Rb$p;Q5^S#oJKS?u_$lBeAYOdauy-*r
zU?(7zea1gH)92oFd8(o(g=1<H);H5PlCEM4uqV9-Faod{zI73PSy$Qj*`Dfh{ad?e
zwOHUM`h0Ly#-OGM#{ZI6e}R${?Nw4)E%IJ-DMAqWc^HdCBCC>~|IBXXUGra4<LSY%
zIU~9U4PJ#>ztPE~A44?EVB$9!BQ}HBa<ncnX%ppC$+~EGqMK%>(4svluQ{FMm1bwT
zQ99YFAUjWT&_tRBk#*An-3IRigA&(HuNFPI6+ZYh0_&1+?w|maBQYGnL+0(3rD@l|
z7uo7N$LcrQ&3O9M4?aE=hSzW31|GBDzE+x}(-HMMHky{}Z6%jmW6^75Re@9$F+Z{d
zZlM%T3-0m%svoGlFv*RUo`ky&Plx8IcNfYLPP9=h38kVtK@b8<x>cX7j{kUn2V@gZ
zxNh0Oqm)Q_l~*0A5sPrgO$7xf{jt7#yj3^d5~+iaRNvpnvcjatoNmA>hvCszQU|6S
zu)Gbav~!jG>yI@+7%HFtq2#%iz%T-CToc++DQ`tN03puc*6Qbn=2J-glH%e_{6qLx
zT;e0{cbd|;^UuQolmT^?6o*%9E!$8^X4v8qp|kV+wp6(VLRi$vTnPhBoLAio91r=|
z+6r*?J*b?5>N~FKGoMBu$y)fFJ(Eo}cO~wMBBC$H(Veky-=hBiF!kl}P_FO)4+<p*
zl@x`hPFiuwmUR^EmWm|Xv|FPx$vUK@gi5QuWXYZ^84OORBxPqrmZAn@IhHXoX6AR@
z&vZV&@ALZn@p+v*&pdO_eP7G_dSCBr$Lh}rx0K(<1W;rc5~^S7l7($(J)p)2u<m??
z$CQ7>X#uXuj1~2CLz$yUoI0>$Z~6crNA_FOH@*fN-Y*j4U*S03S63Gt0??*)f|<yJ
zq9FJvBH(ki`aU}#?SO+=zeBO&Bslu2G#4dN<mk<xTOj;1*z9!H$U>KO$AP2_X9iF{
z7|R6g#Gw<xWFQ!>Z~jO02!y+PL2<VC{3i4jWKY_qoOK|!!Rq72=SZ*bb*iBn7M(H`
zI(O9;k0?nTZ-?GvRL>2l2Oc>!^_q6pSqGy6556vGoh0h{kOMkom!T@|-?6U8D4VQs
zT^TjnxMyB|pF{@?r^@FxU+THYkE^S3>$Rs*##7IV47XQ#KkRtK3O>V~g&26YNG&_p
zO}<>x{l2JT=hD^=4TRQ>$AT_?z6%fs5#8}0Z)>@9)-qg`{$9WGyXpddOOMS{`h`rq
zQWA1e7;BA8O#e0yYqO^unE!$!bl+X#{t=B}Q(2Jj0zO;x^ytqIQ{@80oOEkKH8?Yi
zqyM)We^QNK0Fd^`f!H)|RsfA@XaFi^<#DUm!-J|G4GD$d!Y%eU>espAHaON44IcQi
zqByr-M62oI^C*AtH&)k0rqI%2eeUqe_p{}%pVlh`z5VZE!&Q)Ww;Av%XeyI~fM3#s
zj0NHFBCFzuAj8NR+b-_QZ{NOh_O#P4XJP?wKEwffc_C{=TSPuX5d}%L2%t54d5~1l
zJZE93^y=3vd=zPichFJB>!w4x=3wRT#-W_izTL0)29-r#fZ3bjEZ_Rj%gOM!!3N^y
zwDlguvt0`uA=jwgn|uHczsa%cz08kC{>iO`x*L$msf}_Aj7ii-5si}CpjPs;(f2we
zbK60LI`GF9r`g-#1_~ya{X6i?p$PUy{t49l?-qDeexg>Bh<dj0L81`1f#$LO^&X$D
z;vlJxeeAlyHz-7cwJ8@IokTa=M(F@IK=NmL{&Zgc88nC*FD%Qjx1I~6C+$W)Wujr|
zbfq$yl)+I`%d;E3Su@ojb;u?%4O*-5)DbnCja4`Um|d0Toa_Z6wMwfyl%CALVE{t`
zKB^r=x=O}!f9`MEs%x&Z^N&+d?wdh9!2hm?-v{?I;F%VvINcwO%z(FocrShrKet`M
z;z#KRvp?=2JFF*zFwn3%kXqMt5DD&XW<JcJ=d?h;nJBJc8tk|EHd_}JWBU$7cKxkB
zP=%u4D~H&19g6GmhCbdZSdO$tP9>hX+g~3`PyHr<*D*_^lo8QYQ(X(c@=+(;`ek89
zl55Q+tiT2XJiT0iNoQ3m*w@!G>w(|VcQYu7#mP0giMSjW?-og)iM)*_oTX?F%YAh&
zixC<beT+Ml-vz`~8m}`hCY8g_obqKN-~#}$u%RJK(7#!q_1xAw9;v_GgrrOady~PM
zx`O690q3LdIgS9pSHb!MPDGqelXwMslVDL`lLHd99eK~|A*(YzkRXYPzNdcBgy}u?
z<)wBhp(iS*S{ryHTll3PqRaVCT$?;DSpb?VrGZQazw_FJYNNiqdUnX$VojP+Bn(Nz
z8PqVmsVWWeYL?5aTH;vVs-Mp2j=u#%>&LzF5yK#i4ERuZMZcG1avb<4nV-wxiwqxN
zKZ}AvF_5|K8b9iygHAnMR(Nrwjx{M|Xu~J2KdNErg;yT?uNv?}pj}_jrO_Yuf756i
z85p|(6ZM-LlgBiqzrYz~&Hf;#KFsTx(Qj}go7pUKKZftBaA<>7Tmxw10g>zAgPq48
zq2i>CEDo)$lD^Vd6`U`CcE-}-CH_76>7b^<loLjJ3~|VE+xpA5vgg*g@^@pz9WF|1
zN4Hp!&)_Qf;N^XHYux=4O$yy62U~3;Vg{e0uIST{516L#)%ahN{8<dspczlzjAtdU
z#LT8>C;y#A5T9K+=|<qc=Fchd`{}&Zl4J0_Ew@oaOtlL;0KwRIrt~5dZm}x|wT}-5
zB-C%|h+@iHs&IKzmFqz?*pmf8>HbTM;_!VdduXQxiT;px(w(Tc9;ssNX=T*NJq%+D
zcbU8Z?C%i$k_<K4DinuZ&;r*8keS4iop|OFi})ge(1U&L3LqcE9nMt+qMN}NsJ_n%
z{w1|gwIPZhD3;&~yRftSY%+C8lt{Evb`Qvsx!lOYU?vUFv)ty&7C#XHZOj+F94aUW
zT!7Tk2x?fYLyCw{=Vj|h`3gCdQb3>&yT%&l_>_IUSqX+4eha8#`&E8rvi!uH9P76z
zz1l&$h?9J_ae7Of5EHCVgE_nTOgn~|j4)OHVIpHhlDZb7{ZDa77Zp*9cJ^3GTDAJn
zfUd(W;WG%7@09s1lphh5?$p@Nw@%-)ag@1ry2f|3$T~%jrwG5Y<wAL#%*YoWi<HSo
zYB<t|<$--hBP<96h$O&*nl)o8lHS6kn1X;_8JW@`?$KV_N4oFypp@Yz!gn1D+l%MH
zZw7A|#A{yo^k@5bC}{`>{|!3KpHnyu?Jj*OhK^PcFKRK{Ni}srJuXvhs%IE?2XFz9
zVC22C!O&);iaM0fd{G9qc=%%IdZb}l2(Z54FIb-8gQ52|byg8sE<`sZWFf344hP(B
zXZ5oxEkO#M`&jWSC|fMAz7G}=ab17?J*m(6rjVBPmnU#I@+!&$D%?;78lIfFixFMc
zeSsa?u#H3CwzS3iB&eoCowhCD31AYa-iTFEFO+o2+9vE(c`b@5pR(XRnpxMx>qqZ}
z&<M$vkS>?Vtyu>Y++0$&CC6&XZauOMq#oSza)cig{RMR?nSasBY;Np3EJ-L@D48nM
z(J9T>(6vy9!%tXHgI!q_Jx<#QbiwkgFGG7Fr8F@)v)y+JaC7v-_Pqr-+SCMIoH5W<
zeDZ6gN>eYBrTm>0cuL-@oad9Abe`F}*Ltvb$dB}zS20|fNBQcKB!)7%aSImffYe|C
zWjT6Th<FgNbKiP5TTpVgZj073Pr`9KokXQI2z|r!$hryjVonTNsw<$cXDCkVQ6P+D
z)A|ZDe&xXVp|QaRS|Pn<&e&#Yw1zsH5JaZu=*fZFban2lKg6-NcGsd4Fsy*yd0q%+
z@N`n_IrnTJIG}ylOh<SDku@5pntTCQyq7R<4`HC4(&#K<{R7hKn&@6=Oxe`6t-;j1
z_)OcLQu%+(aUYmT487?Af1YfGrZ6_oswU*wNfp0})S3a#S)h#8O5rQDQ}zrYwN>$}
z%YEmQ@VGDel(&H%7!h3W!Hn*`N_VV1RAKd<jK{3L(~>%ll-6iFiczL9#n7fZ$>&6V
zfv7q&g{K!%hmk--tN2fK+aW{mnF!=R;cCN4FP|>{+b4;Lig;4hKYRGv0wL_pa>?Yi
z>ZK>pZj_1=iK#>&R+aUTzK|j0wIhSYNCvy9X@aTP?+;H}4VbkX`7zU&<GHoLQ?Z;x
z-8^ytCz{QCF&tFooE^2%^^=6`l{>G`_<~_7P3KYhtE#wgMa?0(Bie{!da5_rT%p4?
zd5NW2MUPzN2EzTbo@_91<b<dXd3@@=!{J-)a)X}0x{FIbMW<X6NGB?nSj1}%ai|l|
zI>iP1oHFIzeW7FvX!y*xs3(ba49u}C0_{+6UD>$eWQ}j;avA)XyUguGE5JzU_ms>i
zdKgtgo{w~G>dj**ceDEXS<xc7yF?z)MIMnU4RahJ>|6&m8V5U$<!Ph);QG|mR}E*0
z+Xtn@G#ZMQ6jkXA_66`oz=4s(*$pt8<!ZpMup!svH@!l^eo!5HaHdpkC^RqcTIIhd
z@!W$jH5(8#it`}A{X^#(0zOdL3{H4;E*?h}E@|#5I}hvctZMK$k9!t)p{9@M=Ua+|
zuulre(ENDw*?XJC-vZN_R6Ha+{INakB`T@xO0l#`VEFC;DZqM{klF)s((qnY-Z}98
z!8z4$b$b>z+RIxVDUp<<-h*bSRxncLG2nzuRTZCP|E*qrTChxVM;}j=M3s3+1sA{b
z!wM8%;4D3B>{M2pD-0K=vWRKCa&N<{2K!cZET~W%42b9T42^%A8-Gc2VspJ5gkTOT
z2kR1!(cJhHZ*qdqPu7K3>H8OIDH&>#N1z_x>5ms)T6d!ez@x>Q<N(t$dLm$EI%eMn
zAh%M5F=^5k{&0o1O|Kd}q3<H@@!JcM9OA})MWVjOo3Q=9tmy_^7So{5?L1#|QMGN2
zB;>cn({KA&2nfoTI=I^oO)f83eCE{+t;KDkk)TRmXOb8=Gq02<$rVSFQG@q`46esX
zOOfsFwpSPC8^~iy_@r#J=2;jv4)c9K6lD86Y>>J}I`w;jQbIh7qbtcukx^xf`YKi@
z<mqgYo&`D*-PBP2dbUu}C;zHFbN+dpplP;KXeEp7$XA^1WkFIRm%^4TZX#cWS3}oN
zd<SrU%*Lbu*LN72@6PahCT7>0*Nb3Yf8tsC<_zh!<Z2_>Jvdu9I15@EQPh3I;1|lS
z`t-dkV8n`mmE?i4^|2k0t{!dYZ0Klr0t_BqR|q=3b?zQ#QX}PEel2#lNQ~HOXX>;>
z=p+1FhHzI+8K#aJ_Op~#x`y-<MYGlO#H{v1<;Lmf6i(6$XLP@qP=r6;il%}nJ%HjR
z<eRbI-?od~)BJT@<fP`xr^GW*%vH32FFL-P?+A5U=0Mk+PnuGLzB0CTh#T%vJ>{3=
z&YaUjF9gR~MP6qmgl$)W?!v-rR+j%jR}5LODMYFXZ7)!94QZ$!D2UhA`5w<Gg=Aq(
zx8vs$bU)Hdhw_$y^FZr<-g2oQmGZwyWKGBc;ir&tE%Tt!9_YB}xa8P{KvZPyG1ah~
zBD=K6Y<D*-OH#H-W}PKWskYGOTUT-%(463IgNQm&l`VLa6zP6WWE{&dR#^yh1rKIA
z5?84wX>3k>_+rSe=J3yPr{TrR!q$rE!edE0*LbSzKN3-9m*C|7@r|8Gq-L^pl21w3
z=<NWFMr-z}A)%E&PlkMMs@%fW6IA|*r96k>-k00Nnh*=Ld)%E#{cmOV;(q_n7VB0t
zNxJNR{Am%4SRoy6zYRAe;2($%|LXc{;&hNOyhvq9M!SUcMa-3d7U6)X7F>I&Vy15)
zME=Cf&PuRf&_#OOaiZ*OC)VEn{@8*8&lYJ^{i<Jc&tZ@B`KKRVH#l;cf}wNEz#Ev3
ztQqg@)KP$1+k7{q%1?67#z2>hlF*a8AX$m98{?c@)ZOb`08uC4S1q%qT{Yc)$rl5A
zjUh{k4`>ZO7T5zjIn7P%-;;AbFDdEr&bdMiCZ%Szv#6z<{BnR|0)YFA6WQQF!D(RR
z-oEA3u^#>JOpIqCvykivJOR1aKMe08&KdFlB%OoIZK~%1UX<$989NQ{2xl6zhDs8B
zSe5>8W*lsE*&_rH7V7h_;QT5}3X^ds5NVq(z|<RxgSO)d7u<z;{k1@4JY1`|Ss9qo
zr0pR`GN4FQX2aincGGvAy8Hp{9_d+sip4-4x5&Y&a*1m@pP%-7xnLY(u|eYkSgNoJ
zs%C}~*lHP$Rx-xA3-yV|v*pTHFAxadsTK`pMktOTy@tTpWTeaJvb;l5-f%$57zfp2
zZO8r8C+DV=EoC#5D*hM@_%=r-_U1u71)xF7>)hA$AGDR|MZvb+6IH^|D0K(3Km<wh
za;#IpJ?HO~Uw_RNR5IM5*oAU(yhIn9`*+ZPBzw5IxN-I&4o4%ugRD?Bx6Mx#UTSc!
zICgdE=+V;>^-%A{6LsIc-c`$SETXRQeMYgQ)nS{nwiif?Pw6a2VffvPhr{3G?u)&x
z&;B}-*|7cI`K0b1IuPsO?lPFtrBbkXb<vF|&n$Uo;i`DRFqXa=r(Th-=s^N{wIxFC
zM`LN^c{w&O_2*wWgp8n!sw;!MRg#vu5)MhdEM2Z2(CCw(7LXhM0jz9ZME1{x;(J8%
z&GW?M^)u-b6+wgKNa9({8ft}CCY!V%ff2w8@TAxa|Ev26SQGU(i<JN&#De_@I;AJ@
zs&0SaeeZdp;7+9fMTKEioQu5>GzHhekYwOcWRX=g4&wRe>ZFvMwp(W~b+-zsXEQMi
zF<c`OdQBNn`}ocOe7xIhGU`j`-~pBGMaBdYDApyQ8iKwyNMF6}ivDspHjU7AM~9yL
z(_@`rjg||b4N|;Bp9Tc3go4H11qq_b>K53@u19Ap>H-o49SFw%rVTOR8)5D?(pznQ
zuwJQpC(t0QHKzsI>Lwa_su$GzbY7BPNF)yh8COIltjPGp5)Ix|oI3?Wk|Vs7QAiG#
zI+`mdHLgbA83m-9m!Po%RbfP;=R>g{96~zehwt1=D{&sMr98-lW{6!Y)jBwChxf5S
z@CXm7(h^jzmRxc7LAB(ji~j&0c{PNB@>oVR1dTDC^wb6d;9`XhCT$nZ72uwHIXe%Z
zLt>W2Vh3DKAFH~PB^nJAGHo7<shS$jv7e+78Vpo`z7rQ;K+j8eqS)Zhxk}Z!>XwJZ
zO<h`rYw)7rxu(bf_{uPg$LxgE3E_N6msmr}(*{ttNvJ`1&aef2lzAoDoUw+SKx=a?
zGBNS1aI`ohOU&%Q99U^MT;qoAfboNwrwlA@L%*{8-tl%}U1@A%rlh>P4D$NOcEFYV
zQiG%p5{>^tI5)gS&}=Ye3a}SgVFT#fL@0#tM^uSW1|VyA4?`y37O(P|*&r+11u@+R
zlCL6pJS%H<rABD|&G&#C<B>{hv<DaFO?R1kA#5zz<8)Te+G1^5v*^t#81ZKZtV^h-
z9N88FwL@n0y3TNbL&o3oOFoOP#?ABCoPRNDx&^OZ-9DiTPSQMOY9U}DXHcR$YhV|;
zFxZtkJbzYt0rab&cj?$Bd8nA|IPu(jbs`8}PBu;K?Dxiso0HE-UU!lTqB9jS`-eG9
zsvky2j&d8^0vGWp9-HLB++mu=Ms)#?5J)Vo3e~l3xNbQko&6WWR|WRA7_{!($k8GX
zaC56Apkgu(MD?KPtOL3{=Lfnv$KieMS@(u#_6wpjIlBvzO`^x4D>BWkbcqLuZC1&W
zA$4~u|HPQ(y#}F5<*Q?hs@K6Uz8OfVAGhB#E6(Verk8ESU39(~ko5^O+mwez6|e0F
z&`>0PTGvKS-u}p=vo6#XRY5=tR-N)i-03r#(Tm_)X;?VabYqRsr<*->7n*e{RSOm$
zKgVAiJk`v+0Tbl?DpAx37YTKcL|@q25V0CkNHZ?{o`s@yqAZR^h8RZ5g1B_x=j27i
zl(z)H5V2X|OsLA^xVZfu08z`fAu1uK1PU|`zdaf)1n=moGGsFRG}Ry@lJTE2>S$*p
zT&-$xb#2&}9`P->%&qe^eM+OOOl?7lnymg((<Zm`ln3Ts&cT!8{Ale9`W#QsG20g}
zKU{9`;Fpmdq4OACSYeu`i&L>4SL?`C*&^pM{JIIczx*qDQqMVH@eh8xA@L31i{T?P
zjcdABTY?#Ep55mV4T#Q~du_Z_d@Pf2#s)k3MY2I{JL@u94S2tE@>@~722B%!ICZ>d
zLf1&8T?p&gEG(M!4z*a2WLx1y>ecX!5syenW#D6G-u?*URsb9t>aYoBN3O36vITG`
zM?wUvzX|o~s+f+Y@i1WOyx{0Pe*4Jot-EJ<YIRoe64J>mz(CmBlNg)^a|^@7Xe|;?
z&xrppPsd1g^4}X36>(0b0W8}MlOwMQXEz;r*Ixg7E!@{I&e=Xn2v`tI*7a0>ddOAw
za%RYS0g<T6&}HtJ6P_ft&O5CPyCpEHR-sX4nqZI{mryA$fv0NvC~}KA&`&N8QFC(W
zp{Vcx+aJHoG;y9@w*2|iuM33?j6<NFXO2pVpS)!)zlsTAUcPt<yUthPWDhbZHo$0r
z9=J!<wy1|Ke+%^wB$p*7c1I2C+uC{FD`Cb5_NE;wk;$uGqoZ`E%F(?dN4!hnG`M&e
z&q3B$O`D^lmVq~|e_vj1n&-gdkAZI`eCK5|9QeKVJm#6F;p3iDoWfNgy<2Q%2}dm%
zCG@5vl9uKI)ZDejH<Y??=a9Y<-MEi{>XJXfxcoe1@(&s0kns`%u?S|og?H#x9`{M&
zO&7UUvg}}MTb`GHkjZ%hVesBFnv`1FhUA^@-AUPE7fO?U^IETX3M$pP+Jf@}t?|W*
z+47AdSa5e!u(6d07%EH8-$y0GXv`D7V*Vo35Gakt33DGm_AC)a6i^5lls-RVH2&d;
z42P8s;AH(hdEDhf82ne5U=bGaJF}Cs^W5_H3~`gBy@AK~Wp7qW9*F9vJ5_vQxb=mP
z8|hW}X(r7d<d0SQPchW^vtakWUhev3=y8*0z(z%b$UU$?UYhb=rxf0oV)X|ZQ|<|L
zy-3mf9mOFF)z_J|Fw~}ePp)j=?6(G9wWA!xE#InkS<l}xdR4+6^^BOe<G0rs@mP5g
zTXFLCv)~WI48Tx5K@NuXoT&+$4YJN7?Y@WFFN52{$bp4s-R+U^Fmx&4%{lfKI^~`~
zIYA9Nl_jo3C?OU{6RTbECdlJsZtj46qZLF$5ui+qfE@BGMc_U(`L?2Z+j`iqhS!3w
z;g5t=?pn;24LEJi=2@h8NL^FafMXuJ2sqZR3!<QIz|2+jNRe2DpOL=J8p?0mlkWgk
z3-)#X?(S>x-iky9=ZPGqsAASTjI;LY&l(zuU(*%@tjV1KFPc1xd~89J++ok=RPyOT
zW&<TbO#fd0xuG@=nP<Z`aJjp^`Xg3s%Gd&gpW|m_Uu0K*{B&M;<~7-<peL`Bri6?g
zT=~HChHxR@Ugfw|U%YQO>W46}&t(^n;h`Do8gcq`$VNyO%28=<3YbP1pmuQ=|7~|h
zel}G-2`o*pB<^z_GmfwYP6vksP&CeLZ|J_%ql9^ZBX~L_Umq$sQbF$3IV<T7@3X=X
z;RC49M7O%C71pm5p9FkK2Ie8aMG&I-0O|LhhOF<{&K7I?ZolA5P;wcHg?A<?*(vOH
zdmA_wDioiZYizEomK?H9z!Iqoa6}QF?dS98H`0lK<O9u~fm|f@R<2Vo=)Lqh8uIGL
z!{^z2ZY7W2pnsXypmKawRW8!)mwvIN<I@bwwwv$pm`p&+1Lt2<K0hTINtu*iVe9D3
zI-szt@enFiRk~6aXF%bpLJkER9Z1iuondBknDYMvkYFkv|9h9$vtS@gho+<@h!R3p
zh3ZaI?gn|t{=)+LxonGis=7caZCQ9{RRhq{u0Vn25d9iRS(f;t-YWiM6PP5Q{e18t
z-NXd(c)y%Uc>66YFCzS(31(g*|1}KZ)AKQTfBIgRX6N0PlH9_JuXq(GZ(uy@x9`%b
zcTT=3>Vs9;7QW)c5028uY26IEY4THV#VG{VakeKJJuRQzLCC`C2bi2SC_PvQj$Kar
z^^ei~O#*+?DTrN`#LfON&h25-*eOQTGCl?A{p<sSTU)|<m1{0>w>?k+QeT_dxjC|V
z7yFs<>$$+n%-1yYZc3v1K}xS?qZLDYk=DXYD4U<DNCe6~+zH1YUG5gJoz@coM&wj8
z9!=ejh)c`LOu~yupZk+D@a{L~M~GxnYQLA;Lx%&2ei9#FN+BQ8Pzb~RgkcYZ`~D-5
z014rY_V12ov`4AKEWe6SHMNFUYuYj_H0oibFJL^IEfuiXpEzQ4j8QX`w;0nw<bRXQ
zUiN=`m6Ik>)$elpphs)Eq?5<9NY$bXfr?f4MWv-~<^_%Za!_!Qky{WV8tJZ~v$J99
z;Oc=xNAEEbyUwg>OO{a34H5K#tL~Cp=^>%fG_fsmS}6W;{Y{tCeEh_F{0*g}+u7Ir
zpGXmRq`J(_Pf&U)J}^qhK#CXzT|3gy3ByPw)TpwopEZ<u4sD!HaTqA;HPD^=RpSGj
z!dr2d5;^hM(?5<mO6fQzsWk<Xn8r2f6_u`#Qyt>*S>pi)#G{lRYP6{m-KeLy&&>TM
zFKCbnJ>+}ZDcF=Zx4NBK>}c^O3jZM>SzLw!j87LycpGQ5=n8-hio~}HV^!k9_L+YQ
z(ENoMi9S6yj;jl>*Qr9;qb*w03OcNt5{FK~gDmav#6+!m^h%kMzK-Mhau-0SKMJhP
z{5V$|vWiB1zxv<MnEJ~iopV%a{X{K{32nDBoDD0AUNMj$S)+7zDR44v5cp10#o=kK
zS(X8EmuIy68BaZ0L5nM$whfeg$CLWCe@Rikw<HDmZJ4a3@W3`I*u6|YF+Hw}6l(+C
z+QXZ5JcBUSWW+5XCvs|UN)sv(jXa<7mmfT(mmu{7Y8f$}BA*+_jZjxHL@0bhlALK?
z`34vxUomrA5Tmr#!S?|Eg!-#hMsq1joa2YX%lW)--;(BI2)(oQ)zwnG<e^N}v24;=
zNyGNpNEY#QU@OH*YERgj_Rt6P(8sJ%7CU9t^tc^8q+O&_$UWN8lg#ZXW~!l4LFlR)
zgODB=oruXP=7b_&=oBm}tK&&?P*)7_*|ZS1Ol>Rs-5MG&V9;5rp?Q@jjf9}&u?z+N
zH`z#jLkikJ(r8zaa#xiqcy+aNb*%U3ApErz)vD-tSrCbx@TT<x^XMGQ&DVb}0`0BX
zTp5hy2v9Q&q0~o84%hYZ0QsCp%fi0e!ilt;Wj8yWx8qcpj#Qk=hssYjCNH|U^JRoM
zsY5U!>`?Kne!bk2?4W<O`EU7+NjV#MOmhX<84i~w9+`nWzhSXfJe}jvC4WUk9hKHh
zhg;O0@<N#kYeW26eE*IR-H@feMB%T~IGr|OY)4a|%K2}ko^#wUMx2T0hI0fkP&^)I
zyi4H;%%HMmq>>ZQ2|5ME&cJ5ecZS7zkz_$C==N6XnHvw~m9yA=sAJCDaF}DksUR8*
z1~Bj=C~VIn>I9EAaSZL~PAu{>RVJ?V3K`f#9Q<4&t$!7|(^&v`f}jT7jLk+d0cH{<
zV&4M|?x>jSAiP>Ac?cmH-lJcskbI|fp?m}*cB2zSt(!W9^cO^qdUxA~778Wjn(LLM
zGobMO`CQlRh5F0rpwFC^c=TuC;glSazpkYYn~UrYdu!t)2d#gGoNY4qCVq`Yl>Op+
zZqeCk)^c!R@|#?{Z29k+e-dq>@)05)G6YY!3J9RE4Cj5RZmt;$HzlyLL6*75)nP)z
z6r^8DSgV}?QV!sMc?n|~4Gpxp=!Xq<ZK*mSxxQjBM9(3K9(>gPlN7g+#~bG+e(>bB
z`&4qKTL}D-q(Fuq2q)w&R^QE-Y7c$H(tWL1WdIdG6~2Q}VwHy;V94R&hz^2JNlpqd
z3eIY&RIy{P8fIR1M0qE=U|KI~gR7f;w~+Zn0q6Ith6F<)9VY&yuCKdnw|Bvnl?<qZ
z&)r7(OPWi5EQB2#a00^-RVe2kIr6({t^jYE!~3AmzF&t8ALJ~^eVrq<6(i}H`zK^<
zo`ZlMR&|}um!-NP8OGVgO{7g8UJ_k_8XHTe;s}I`^K)5c*>9nt*Q8iZXS-swn(QoS
zym_;Z9Tj=N8c*n354Y<_jrQByDO`meYT%rzir|-lb(Y|;P*!U!-MV=@9-GFqxa)PD
zuC|3PW(uXR{;HSs?$j&i64RhQv6&g(U^M*l9j2i@EZsDrG1MbQz^PA!tg5pW=0gE%
zGq(W?5vLcxXkEKGUD3Yp?!7h{bpNp(f>jG1Z4qrl(4jWaV~pf~0i4*1dMSVk2v$)*
zSG9yh+=E&ughhc^(MazO2h+fdN(hfxCMeptb}#u1u~_Lou)dZ~=?S5KUR?^e<CDZ7
zqPN$EKP4Fz51iPeu%S~|xux#`&eQ4JZZ2`4UHOMY*ReiGeFkrbFMsEsP!~WG)ukNA
z7m9|wrt`|hu#co(ljE}_$`)J~ezR55d=Z-ZQ@6o|BtAvSkW1hnYAKnMo7Mk6K^}^M
zl25dY#E=s$L-%T%(uXuNyjuVgFys!o(Sw*Jk3wpDXQMuN@@HTdE;B+MMULKUOhWII
zgDmqbBHo?4I5|qKS89j;9vGvG(EfbK1@_lP@13`O^c2F&>7XNdE$|`eD)vO68>P)#
zR&bERfvSN^6o7Tyzu>Y}Xpb6xR`Fh=8>pY@_!y<v-x-nvQ29IlRey|uzy^4dv%G_L
zi|<+u2`JwTe=H^*0)@3Nle*6WQ6&P#6!e{`a~3i67;)X=5fG87fY;}nhXQ8oke(0S
z13$tN3zIkYQm**T?A6pH+A6GZ4KCQW>?FEfwX3GIbEOiddXOPDy!7jfY8%uTI_Xh$
zK>)N#AdQbAWHH*FGykD_nusvG6TR%f3Yj3HJWSSZy_4F@=Sms=YEqJ&wc!u<^|d5H
z5x&vgk908ar~uzQ)?QK~s}mj3L@$CgM*5`FaxcZ=aiyPzOVM|L*GK@oB}Jibs&;P)
z$XOUcZ$7Zf=}edTKQLk{awf?1;RX-b=<#Tdbs!y<7RJB^nc$%6l|%H;cpD~ELaH9H
zupiJi?qVf&5hKUI-%NU3QxU>czBJ37>=MjW9r?rV^0w5-68AO9F2R|G&o>v0jDji2
zW53J06~07Ntw&a76Hv_3-1KT8S56A}x;!vZUOMn_BEx}{au&8C`PmBq4Injwop|oG
zrjD)LRmgx4U326GIQgqb<op6#v9nbd1gH1pN-fu)knUkR3Aqw(`2)GaG4vt61)q<g
zuI8^IK~b+O`DK+`CR2>&UPa5^_@Km}`1Y;F=mM}hES4^Sku%h&&nS-F+kNwe(#`H>
z?p><dX5<vjh}LK%m6$UfHWZAe&4o8YSBHrL8uNyf4Pqf3C0@gXu#DW_&LSAhTm;%m
zo318wAR)hnsQQMf|AY9F4v8(Dl~9p>0=Q!^^|vBOGS%HT!OMXTY&`hG;Lm*<V^Z4{
zN;?lIeE)m7<l-@Q68}#$&R%q*jkWXwyIV`;mx$VQh7VK6{yZ^<cAd)K2${W}43&B4
z?O3?aPQk3@mtvK*%Udq>vy`)CxF`X;vI*~b(FkPjlIi<jyS1~b<e~P9-yppTA2C#)
z)c`P0=;H~e?$8HDyooCuDZb^4tx|sk{kmz&YciTk=Ra6S2<~pD+fMt<Akrsc!RGPU
zv5026QMb}rH_}!w**Skt4v*Nyc0*qUE-gVP#<9qZSE~S`$Ptq4fwh7pykKES?L#yd
zB25nf?LGf(M7SjuCLRLAcJksl^rlyV^_ogq1qXB_h0AYusoV|z+MH66ArFF{@O9@%
zZn-}So1W!}6j3!yUf&cYxDhX<Pkr&cWv~D1ZK0W%N;EL=^k=!HkpAULUvHkq&PpCp
z*#`(_@)zJ99HM)zy5lM2e!E@@nj9$dl20sfkLwjAIHX<E^C+AnDJ7_6xsoPbG%T$`
zAHZe-H*B8g$7O0@Cm0T(V*~#)HR<G!bh2}~7N)$p-hgin^VL~1V|v0~S``-(WHap9
zRaaZ4t4PSWW&b<~>J%K~ZWU*R3d2hYzv+|37NNqRrw|>5yaOJ4Q_6!h5O{-KT`eq}
zg~Z>hOmRHuN$&F?RKv3wWFX_~egx;BoNYW)hO{A{3w^}Jqk6&2d1;MCL+Pr&<H}eZ
z4e?=#f%&E-8{DN%6jl6ng6yK4FS_21G-)kz%?#+luVbQZtRIc*5eU%N#4>C5(d^8y
zQuwQ@{pxIH;a?e|n=U)RBb-+_gcoY6KiI>@w2-og$s>iIFboTd|Il@$?I7U=^E*^E
zqvcTxCT`)*!8tPQT>2zHJ|V9)aHqV6E@C}p5x+Z#4Z&Gu940FovsVH7^A7UR-tl^V
zPj!ittV>`ct$ptLhu*nfZUSZUCH)ZPvCxxz#f);rm3egSKkj-w8Vhaa-pB1Kmzdy8
zx*D7Ia_B_vuJIIjOLPfx(z?L+Vu)=L^qXz+6m_w}4ox%()h0bca0q7uR<*MfkT#va
zDlS6Y9x<TP#{h{jctmh2-r?!%tD2<=zr`U$PJOHWh}m^B+fyVX9~$a+YA?a0U*##e
zch5~C(88?!#OfUm;#w)lqaKs2aZizybJ4hRSRR9zDT@H)mS68;#`7Z9mtM95>1$7?
zO;=4S)qpGShIqHh{txpINNZ}6*5bq)`fTKupwnDAo~HS+_0~p@jL(0AhbHbAd~<9(
z)FZ`sbFnB<B;$|ns~wP~5m-;<8YgI?1s(FqH>l0CZ<%x>)@VdPV!hq4{swrn#??SR
zH&YrPy6UJz?&fD-tO%-XmvSG}ojpshgRt22m~jH1^NYf93K&oO(Chodh|g5=C(V^D
ztP*7kW%yOpoEv00vG)QEIWBk6o2?bSP2@<3rL7Wjr-rErY=tX$6!l3$-I@Fuk#Sy1
zfxH0F8*m^XvY#a`{{|$qTu{k?SKW|vP59({=pm<!NYrpx9neRnvm4hX^DS+d48NSJ
zpbe9WI2;1f5-R@vdw}(LH~QmhU#J?I+yHH~fOpNb^djX7>^RP_X)7hA1}f{+EXF>k
zaS0Gu%a$#!wL&&&82XuFoAzXNdVM;ELb%DMJ^7Y+AHn*l{%7GIt_HA8TQo=iPT5QO
zcR%W(9X=pq{QpU&+J|kHS4_r&EdO-pmM;GpGbzh7V`y3-?ik|cX$_Rz5PgDV2%tTq
zHS~ghYh*YXY@Bj|{J2P4@>EV}b%c-Fq;E`ZoY6WYB{-pgd|X(gh%_~^d%9{Mvnly+
zg`A}M5g*EdftNOh6Jkp9My5R9xw3i{T7x$kWS8i22GZ%#W?c=1+n_W{s0&*KomZ+z
z6u|V6kP6g(7CSK7$O17Z=otat1MSH+4k6%QIb>A8CHg`8vhPUyjhVN?Ao>CIxVv#k
zsBo$xFD^H&6y-qOXM3~oR(gztnvV3#E5E9dadydW3{|2tw?Z{<@ckuwFE21K>K=Fp
z&0}t1rw4#IRYQUHJGI5&JTUa%{A_YIT8Fm8%<07y=QNx_V;36SZJvpo&DiQ+rb1|D
z22Q}9PpCg%33>!o1M200dj}?=Jz|yy8l$>jtVjiy97;NPKE`}nocz|@NNaL+>hLlb
zcw)yAk0qXQtSXQl+$_1Wm_@xRFM0i(p>!K00Ag~pSLuO%s)Cuou^<nvH~(o1zxuq<
zulNs2R7&|T8ig>~R!xan_GAfZ*)*|Glw$}TKKH_*qqeq|QuAYwmVwhfbM%C))7UqP
z$C)kUmAR+USAi&V=706PAPviKFv{2nXuaUgW}si{8TbxOBj|epYxzzO&mJ*F5LwIM
z-c7dOD&V)(^glt0gncO?wYM)rT?oooa*Z`8LC#QmWaWB5VQHuC-IhgdKSdA8zVlym
z&)TxU&^7o{i^*XP>L*4HExID_mb;pB=T!L`)qTT%U(-swa3Z*%nfRy2ha3Qn;W22K
zB#=m6Rs5wBJly1KWS19_-o#85-6V2uFyIs<0FDDw*OUcway8SLufT???IK-j8}gfb
z>oiyNuE;Zdp2O!=`Y@%~!`ew$vLBjeZi0C7XIQ$hsYz5Vg&}5+=*AtC+nZYg1B#nD
zw*To1z{pI<a`tMAi+{yc(|k&M0|W%~DpTnfhb5tSf!d<rS9}lEu7`0@7)A87$dP>H
z%;KPeKGLhFs02a;xCp8J`Wm5+PP&GV6M&scT$a$Do)jbvO|R~Y%-ih|nF?bWdtEXc
zTJS=PQ20>mQVvBqQ}mJ`)Wp;7dp|}0%tD(BFV8p)Oo4PI4doeuHUr;{Rpfs7YD=yr
zHdRigiEVVN@<R%Z+45mv%99=4JuGu1s{MKwV{x$0HxO7|AD&e&zB7Ju5A@XlR=qN8
zwH3So(QGJ4(jmHSR#iAAYky4(e0>7J==eO5S!VR-qz`|L|ARjl3Yo4*qltV+?$lHH
z40jIgy+aGVA9`KoY)7*eWD+S#9!Yd1kQkO3^ZcDKdNjXp{LM8<3qm*#4G?R<Twr>{
za)htq(Pkfn1q)^F6gzx{P#oZHTpThnsk7IK)-4Z9mL{kK1s39!px_O<qw!8hrzDWh
zUR~+GR8|)NOBm7X(x^-RgB%fEk6-DM4$S_-<N*X$q`74Y<Eh@Q=g!{u8WjEjHA_$}
zrtlt{fG!F*r#OF#dp<plws#8@%gj8sW-fE@%#R~M3Ki_v^nPy6Ic_e}P3TG5l#sm6
zk;i*@x=I;BALq34734sdbk-Qo=>>eOB<VHekFp`PZ7+=SD(5;I0(qegIX(PJbKcyR
zUbe+ZxAubMplQDc&{ewY?&R4a>awW}_5Ez-7up}0%m(|LXd58k)AGI%`EQtW%>oue
zA=~RqUvV2w^L~S*wh*4<tO|!iD+I!FE2L0xje|lwtN2_xqW<)3gQ&9*6oXUN-NUUP
zl20+i(`!+TNdB_)N%r7l<Vjkks%tgVBVD0s>R9LT&&46ri;ZjLknF_0*CKlbNlyu9
z)<+=T;bWT)bkd`}DbdZ9J|0^;TlWMpg$`<n!_sHS+0~R#L!9xhPT^!nup9pisH8Ib
z{YYirJgo*L`;3$Jh>3s)RB@^OuK;I%R(2Zpn;E-Je?Ha8&YlzY!Q$)et9CK=NP-0j
zc`w-H6q&0BsClBeSm!G{^3%$^t~uB~=(04GhXLl#UagY<cW*x#Oc-Q*v<%DM*ffYM
z-2!`%4YR}BjexhnLj<^6cp*A?3a^J0+ShcQUz0b1!;|@re~^64_0DH<7b7Ul&QT%{
zc_1QJI{vG4=f1|7kOYIscJTwW)sQN0#$nix{k#fXMfNv3#-jbwycZ!B>wci<S`1y&
zfM}beEncJ@j9?i;UFc~QC65%L4bVNO%geJ<J_2!G_y+*}pvi^|gfu=uNzjUoy7}=D
z-NBz5$yN?FZNX<?fne;)p@EdCC<<e)H*l&`V*V@4IaB^o*F>Md&3uI+)|EWPDjt8E
zRY!A0s{=>>BYfvSZD+)EaLn+V05d#_bKrnjYYH`!?`vADfvcBnCZ`3;bK8Lbe)wAu
z1@5ithay~&l4^t6NCmXgJ`q5<Hq1fj{Y+n|B7zzSl!Qkg@9ssgU2;%IL6<ttI|<8b
z+~d|J5!>A7c<sL0-q1Y&Z8YeQ^|Ck+-u24L+=9BSro-#-@Nv%37R`7wg<Hd?kLGC4
zwF6@dtZ8jK$+COjKirMM9!;~ARLwd1EHO8!BPv2v<>twKBm}#;?vm0N)Xfw<w)Ba;
z2OJav;hUUiujBlWHV%c#9xwc(GT(i`nYst$m0GSnD6Kb0q#5QLE$*`H=9U5FwxoDG
z9hKqXa+k(y43n$heTP7K_%yEYw<a$IS#9vjV9;-y=9mY65va=~zD3<fB3M4KsvYC#
zF*zrLAP2*=Z~=*$H_wK=?Njje@vHZC=$>0DjHxmT>9;b!2_%YO(4E42b|8NsC8uyA
zwlt(jbauwiX+9Q=I2<txnvxv2Qu*X~V7{9i>vfWRp^8HcvZ&?pbquR>?UJ{a&`<I7
z(4#i(*VRRi+Q4un;^OOYFTBT<{J;P=|J@g{HXC|y7T)5tPy@3`nTmu)xLslgkscmC
z?fi|r<dOC2U(BgJ+@9}ty{_Dk!(5;7ubfB}KmCCf?{&)a=Eu-3e4#oPdd&Gfi>|V6
z{lNzA#=h$ulsdshii5Q=Q7{A*46{Ih>BlMx0}#OwV8HxEfHXw-9I|P$2OlZm!nEFi
zz`XU;O>51efMw|R*LXhi6#J3ZLY_vkFp9e}QsDp9R3uG-w$#1;*((yxPH25G)^4v$
zy8%gjlKXA5MxRyak>Xj3AN1~^og^y&r9q)BPjvqKrTYW*9w19Zmbll(wib!Wn&TGf
zt%O<;@p>2?^;bh#0Jt3HP95N6h?*{^-MR*My%KE&TIA4Gry4%_+0-=w!qs=FS!2{`
zpdTtcBLW=`rG)6_K0QbdW!j@Xo-w=TGQ1NXgeqaNdtmq|`4HQ5(r@PE+*}?f<%!KZ
z(?P>-L2u!W9(E^{wgi0-*xhGv{X2C4M4kJmfkc2y)xnzWhc%O^PhYv4D2b6`V9GZ_
zUDl1o!974H)<)4pAbIvm#6^?SLsUP2r;jHr$C5YK>uMiRf*3g5VCUB92=ocV5q$$;
z&5I6nDFnOhuKX;SAAxI*di@$^A~A1<wgre6MrF-f@LMXmNOY2s_+j&1KYbeK`A}#5
z+p#{Ng<}8?06)ttwVI5&EDBtI?Ms3Ac>v9a{Q_So;0R=BuNikE?Q_VE?hmOeg$ax+
z!;0n(LcBqY8JtrIslC!nW>!y~+-Owj$8X7L2sk(Kc(j)<g2mzd`%bBMtckZ8zm74i
zzsTKCX$mYQ+cso(>EgdPGVB7z%<Mbl>t063M7As<31i8MIMeD<cTn|#w7<~q$^(&T
z%u)^~K|?cvk?=cM14}Wa<~BANFwUz(Vs?0(QFmDnNlJ?n*9^`!wV?64kXG4woODaN
zh}P}=I-SyQg?}ZFE_>A3$NsL7Q9|c@r<)uTq9JUaKsY;Vm=mHsxFc!y7Td&x_rCX+
zh2yeUDjRKmdp<uE)Gr+BemaY|dxWn+fo0m}#;|DAjdE$`Zv^apDkz$$P=(2rs4o(0
zjS~$UJnb-e)giPpcnU#@f~+|)+9y8!0F+$Cdo&#M;5M(nbRC;n7Q!*3tw0eX5Pp2;
zvVS<`^ca;-4QJ&zePf7Y%xS8QfUke{j}O~qlJ7uUl~@HoKOuJ=GZ4vA!xj9wr7HG!
zBoI3XA_<-TZ(QSnSv$NAsCdrjsJDM!g1$e=H!#$(Y7g3rV<~W(<uYBPhal-rX@DH*
zPsD!|?O%i0_?+Me%t@o3Jqh4A8$l2TDU-PQg*PnGkmBY=wjR@*!}8bv!YN8X&`p{v
z2GQGpzdIngnzHkAwi?wXG*+BN92{rdCY|)3_5)TH1f~*4s9O_0!1&iXF%%+@;tHE8
zawNwslp2Jq&JF34x`l-`f)kbf;w@s+<ZjP5qA7;Rlp|Sla=ZztAi=vw9SCOw9J1mA
zav;bN7_3as3#A%m4&~j1*e-p|oxQII17D1Wbk6zY;h-km4Lu;}yFLBk`xSA?JK&7x
zbvul4dhl!<%a>e0dLn3tgt(R1?cBGj7~}vifyxs?N;<M}1^2osDSr@K7CDNm>DT}l
zqv~9Gf>^d)r7>t3WwUc$>i&lN2s*0>kU1MGpYp7F&>Z%vlF#Qg@Lrcu9Df%r@no`S
zJ{(y$fIJntC#AtPY&^LVVD7pxum$k}|J-*xud(lG*fV&6wo`8XOeF9kY$W8KBpTAH
zxza>r&C0y~T2+|k9=3J|BzIVbj=_gLq)Sj|JUKt?yxwTKiGrf^7ER&c&)$^-xB_0;
zOL5Dv53Wx3I0B_*JpT9%A7VRn6{jNsMUa;SgCEHi2*4VSbMY~z!5*Mm=2U(fHW3QD
zmbov&I@H(DPnRKZZwHdY0S4?}0?M25M}N?W0RleCHT`Gbyzu~+q4nK<Bi6USFxsr`
zOX6eniGA)1RQh+jOYLjhQ<cQ4t(?$yoacwbfD;j|%>IwKz<k`F;0|BVLJmAml`xi1
z)`v<nYLdpPTtiygZ<u@R`vdFxiKcwYkJBSX13C(f#c)YslX@f15nt`|DyaVf(#pW4
zx_=PVOBLt1LBImEFXz+dv)VWy#cXoS$O)@z8)T-`$7lsoTIT{6iv5T)rqq*d(_59N
z)5M;m2pX<3u<4-F2XLk#!sdDmdKla)RGh(`<cbk&2YCHTmMmI{VIcnQ6^!M`7sBfp
zkEV_b<TZYi_ZMO?FRYO~{U?TnceFgV<_w@smI1-T#9CeH);(Fp6M^?idSN2f@b1w2
zDI5Ku{p+X8I%ALQ&(~^hy7cdpJd$=1(41h~;wU%rT2N5ONL?DvK@i%&jII6DkSV^B
z2Ki%{xyI+K`n1OCDfK1GS~SsDiH4rRwP3jMwzBA4ZDvDLyRQY7e`72<wt;c^UfaA&
zi*)kWC3BY*p}8i;v+Kb4<NNDR9+N|h0Oz1UkZ}A?P+lcz>BB%lSCCbMP^u#hDlYq@
z;I3E!LURJwNg58LKkkNxCA51`Oi$9rdo?LLdz^RlI4^Oy*-)_6E?-fqPgqB6F*G9B
zIjRiCAf&lfsk-*+R>=cR%4hA)pgxt`1Eu2ka^FzZmyCyl67yQX#JW1wmmk!ffPpCT
z8Q_0GC1B4?u~s3Sv#(3xfk^&MQJCQd`~qlUiO;Q;a;Ac`GGkUJL23yqYT+xQgiZ|5
zyKl$SXxT!+^Y%g{5Vgw6%c3R~xQJWkYFCVB)HOWBb1(Gv9F0M|us0r;`#uPj3aT@|
zK|vz}WCo_*0UPu)vLZBG=hsVC*9h%&jSz^8;*VRjvP?C0qjIz)Y?FLr;5o+d95~>0
zV{3~q4;0VHAXLFC2#&?A!h26Vs1ElyY*R$EDgm3raaKRDGVB;{;YgU)%6PmVvTCWJ
zI|9_4m0?{;w{zGpsX{}A$;+{h6_VJ?u04AtlbbK<#%^d(@-{UAUGjQgL&GI+^$1z`
zpvWqHhw~%QU=ib)m)Es)`FXhnvDHgNqYZKu7Z`hi$So->R~xhtp6=uRv}gdNp2=RD
z--5L0TF<Y!|G(w%L5Wq8ZSG~&3$d^u_;x=g2`qFX7ozqxO=7hxf0Naqo&u3o`IY~w
zb{abz(BkCVp%y0MZ@qeA=_YT(Qm{kElf?dAaDJgm3q=@<{aj4L%$WV~*O-P3YzhLB
zjqz8wEh$Ry83V~^LS6;TXX{vZ`#EsdQ2*M{iIHd32Q&tP%D2{Szh^CSSkkrm{d2gz
z<Bv7;;S`U;i#yqo*cH>7DTFjaO}9~5yn^<8lOP^^tA6MZ9Yu3nhlH+!_USU)HTHH&
zl>ocP=E};^k&{+R;I|}s-d}&m982Ccot!}9;RM6l1r=(qMw8F93^g2N=X{C!|I1V;
z1KNA&mv8HU!OXY-8y5T%;_c(oUMKHBt`VVp5cEz!EJa-C9QvZgL`I3Y<%LJ6RrEa2
zB9lMX>oWE$EykR+Mvq#N{VtSW;Ns{@!1BE>y;SAHU75tcqD&oqO$UsAjj0$;)J1(@
z^K6V}W3(X^lg+GQCnektH)-0kz{4q4G{-H}(qrXqsg~q4;Nr@x&4-|j)b+d21ce!P
zG?ouQi9LA-*gXOPXohKS)vC~yXUh;nx<*5>U|Acz0j1ZqC^U3DZP0G6t7<T+{Ct$-
zjU|t@i`d1j_?pT7bHT<gV~V37-0m9cUBf<zz8hHZ83Zs^OQdDxCuym|z|JqI+z9Wm
z3#gaU@$P6v;t*c+NG85q0Z||!MnpG$YQ%p07gM{V0p?1yivfxmhgCYTc)BfP;UaTU
zD8{BrVcZz_>)@ZztHF4_EKbmGaGO8F;SWY9f?@)HFqAUY8^G|40I}GFE&>6@=4?Ar
z{q52;BWe92g|*syRGt)`EE%Oodo<nn)8mfW?!)l7QNT@&W}Ouz%~g@QxJ2-|rinm@
zP$P|v*DhXl_%R!r8@}&wIelkCjhM0eUk*Fq6=GE(slN`gn@CVECYtHZyU!wH`4NsK
z=b?E1VyAH~#4O}R133!akB>bj@8SBxJY#q?`mzw6`3Qn!^2oY+Cnuoz*Y+3)KS0fQ
zU~ZsWZgVUYc+)j?#*$#%0a6ja48v3={E(?9i2)MYrM8W}7U#-#tq^eeH)B>!F5Icu
zOZBoaf~PoMEj=2Q-~ohD>EDIZm}FnLXMvgjk<$N_26KUu8YnRM^tPj~*>&X~X+YgJ
z+9v2}qaQJ%5}yklocr*1ZNJO9X#`QoKuKk{j67;UcRzsl_@O>)Arf69Jo!SknQl*e
zpZNP;@(#u5hPLD8jL^4UJDt!;8%Y!$f=u7Sug{WBq37gfi9iuOuLbo^!|7Z{ERCI}
z?Pk|?nzOet4e|2<H+3C-baqqirNoJUL0@+Sk$7lZAm(aJD+5Cs`2cVtW%VyB0YhuY
z9Z|5^?-&Mop=5cOzJ;IwPY6}Iev64M#;m<Nee2WWJr`p704PA4g1bKt04A6=)bA{-
znyMOk!X{)Rp^kK?X|Tw~!~}+4gj(DNW``&5j$p8b!!|9o{bOHO;2K8iUsg#Tk42&>
zD=h1uS6@Y4@JEvNJc0+N!3^D`fWk=8LFdNB*hp->M5o^xgGfgk4mSuoaynkXfEA;L
znj632kR?sCirtmT;)im@4E7atCQbedy@~27s|VmWKTcUg=N=)aPh~II7?;Ya^X|2E
z0J6!g55vDF7Qu1DCmg(Xe^P>`fcS%%f!-D_ox=vI5J(vU`}e^KZ`Xntq{9}1@H4Y-
z0$D8qq;^^m;kChBFc1ONuPJ21M3#IPYLF91WoGw_=zJ@ry@-D+0!J^$Q<HlItr$9b
zkDpg>go+rHdv3Eti<D9k_K_X(QT~krF4O!5xTh@*qKp#WROljIp;V!do#OJ*>GUm<
zUd^H@kkX~0`h5PqQFapE$RP^(oX|pY7YRs68zx`O?&?_kA8(z}NmU42E9nfMoPT=3
z2JO)L{)8*ZChXo0LjV}A>IXW2k&%%>gYlqjc8xYRWT-L{7Szkss?1^U|0;)03|w95
zl8?YyBPQefAXs4kamDeH4H;5c@E#bwjXDJ(07gp8n&7;WH=!W8sI%JI41C?+r6hPj
z-z$WEkI5w4+kDa%u{^)<RC|t*GS+v>9i$+5DBZ(`fv{tLOKZAClv0dZyJxTJBH-F~
z;g~BnE)f5oz2Wq0xZ!sAd}`&i9Od2g>%aHPJh4C9xzZ~1v3q4MXxi=s>5Hfjncs><
zx5BIU)p{GZw%<5hq13t;>(E?NLMN}z6FLWoJecjXXKdT`8fx(FDo}F<92*@VjT^a4
zc&4XrZfi}8#-GGxSpEy4d~$+~Lpt-gA3t=ok$cLIV*8kTni^Hd{ms(J*F*JsD5uHJ
zw2)}1(`PqeI$+KbX+*FjNKoRKsZ7U9m>>%LC3JX*MX($e@N-I-CyU&ji(j_@q=XU0
z9%%Z!1LzN?q~q^wX?h<hpX^>Xr5&{e=YZ%0ajj+UnKNGZE|r-rP2x0Dwh<<!_}`4?
z>e7-VkBJ{D-Y>l%B>V%S?y8?8JIW6fD!xELph}A<tN1%jD%eH$LY|vfh~s%+E11!*
zO*Q&<p;7^3_7^D7r@4v3Cm$UQTHrv3G2`@uAlUWc%hVNJwb}-tYVT+79b!_tfttrd
zjmBp5BYXXlUb5E~w^WKru0)X<T={{xYpUZG?A@D|^-3_wX{0P4CDUMI4xsF49FhiK
zfy@T|nHXYwY2a)x0EoDV?zv6Z2}S>1dxI}ksP`@*!d5aE>cv}?#X<J%QQsM)EC-K+
z=w;irm{&5LA1stS;3#$OyO6rEhq;yi<6RknK+mHLgBwFs{iJyxt5Cdw6gN!b2&lQW
zS<(`El`q&#>lQ1%IMk24Sxt+iLGgu=CegvK<3j4&@La2(O<icwY6HB@Bu2kC@2>A|
z_+0m-D3x}GCs?1yEbfN68C~v9G~fya0Y8VByIyTvK_F0=4ZIv9|4A<Stv>VI5PJu%
zHW7#FAV>H_9DhM8f;1tZJKeeiALryCd>=N?USM)=buC?%+=;8L;{BAxYXzVMvei|p
z3pUbsBfcgc0}$I_i4av-kYmL{OPRZsgZ!h1XJcvH;t!rY3*K}R?>qk>P#kARN=!`N
z_m`=~=Khg>86@=y1W+eUH#ND|S9)tGHh)v6&IyH^JMNJUVS5s?_B@q;;a+zg^0KPj
z4n-@NS_<=yF=<-|3w(%R!G8IxZ4myc^zLbU;nMkk>~rxM1LEVVfr+uj5J&}M`%R3k
z&fywCuS1o|NDp4cz}gUIip|d(ac3z|T0p!KkgYfHTTj62b#+h~<BTiet*b?OivYrY
z%w#_;Zs{GT=KSU>VAjZmd-v;n7zZGGBsC-e4Adyp>L3)e{)hVr5tp{BBfIp{n})Uv
zmKlbfEklc&ZD=b>I$A&MM&j(W0;nRMHD{?=?g%$xpLct0*E*mu-?>!hhYqyVEe#|1
zjNkqQDok@0zs2A`Sj<aRPAGWtL80KgyZl}KxX+0)$ieMs(L8AfznSSV$z4S!3H5sw
zqFhCe1g_r)<#&w))d#<uZl(<_L{4B;8j{yCyY@U_UDSl}m4{*SkBJFXpL$SIvitc*
zc$NH<x`vGt!XdSvy?XlorY>+01{JMwC5?T@*k?Nxv$NS|5J-AbSHul+-;>lctnJQo
z1J5cc!<}&D>%p$Lzcn-l4FQX+5Wp9-A+En*eNZ%z%#ZwQ!7;!RDPQr(`SIT1f&_@8
z@H{m`c=V=^yg?EK*eh@l2eSgf{Eo$vvCY9PnT^Q<P5TDgY;25c_Y)?+M-XziOb)zs
z{C&&>5hS8ZT%P1QUt=X-@lYQ_?crK+Kq%7kqKCaWzbyb(yY=CyWNFb!UE@TM558C?
zdZpDvB=v$gA)mGxsJ^AoccE?h8)-zr<M1g1_1Qv^1z?^cc<~L3Ra;qZU5qtCSlb!@
z{OwcOjHC4N7#?o}f{TEjhOXaLf9?-IHUTY!0IqP93qpO`0J<Vi#eMV_BwgFGPV5rF
zhW~-m_Pc=U_KTG$9kX8p95zyV%Un%M65!sSivwgFOW?l;{{;CyJL{sba}ZZdHAI-)
z`@jU1n+GQ)1f?>3OW>_k^y{!%S3KtT_Yc*9ya595q0iJh#j|UJE-Zm+-Hh@w$zgmB
zjRY2+otq^ueN0>aIOUk-ieexp^L!0qcCoe@K1Q6X<4GzQi_17bT$_82%LsauIT{D{
zGh$oyM`~^==TZI(P7wr#<GY7<QItw8z@Ve+iVBi=vKylq1c27a%x?{VtP)%UKrlKC
z>Cb3@q6iGU3doJDI)DYQh0_CZYz~Y7w@1h^SiJ>kDQ>pRo!BV6Q7HLKy<bAc;?ED!
zp=s2IsXSTkQt{VZ@G(ue0+xw@2o*CfW_Um$$n~_T<hfJTKQC!u2SOi!+=AtG<+Jbu
zP;{e;G#gFE``>tqSjx%=B!?U(r_dd)F@3h-u?(U+s|3c^tB9iUg9xIiy5dLdD-Wr~
zbC~x(<F(&dGURqbOG8G%>uV3?BHryX8AuNrj!Bvq<C8cnz->arnRb3jO}M4Xm8dZK
zX^8kRaCKB5fOsCJcXp@=+W#Tc!@T4nFNdiJ;Hyb;{|C@x?FznB1fk-GD)^^Ey3F!t
zIozW{Slr2jUIjDhXt&_&ppw}vt<_+<Qn*VQWeLzL6KAk)SMs9VU3q$)jYky%zYEsD
z<v1Ag;RD+99i91h+8Q!_i;`WZshLwZ$U6YoBftcDcRLV2m<oXM@h3FkvG2C{p$l3L
zat6%caK%rJef!}znR#n8|0<drL0Q}EdTZ;t2PFsTCDY`Rm-W=%!iPO7LtIomge$lk
zNo(jXtrbLV^8R4oM-7%ko8?2?2V+{lA*tIxQA;6SM-EN{L1RH^2T!n16y3<w-INLQ
zHcQ_5X*N28tGAf*!JRP}CX36vrptrcCj8JkXvl%7*yM3FO~?xRVQ!bt#O_O9H_5D2
z-L4q%^=I(;h1b80>JdN+$}?aOh@Jb5$|B0v$J;MuJBWFDK%u-e+0@F)P}LB>jc*bg
z`4Jc9walKpixkmhR}ErjuJI{`Wb5B&AHq;4oc#4}Dj;BGm<jT$efNQpV4<f>1R!f@
zKGrbthB6T&Cd|JyV!=xN^t>Iw(7SolXnJ=Uhd*4+g&m;5HRwpC)^B0*%HeRL#tr;x
z)WA(B!%hl;TaHj*86CUCpF&In_;;|nE3Y6W=6B}i7$y(?Ibb46P-^#UAvXhLjcjcQ
zB<>(YhupSRqz9wLV{{6<AJ#W4ZvNhU2(t=JKeTDF3|)5B5|gp7Yxnv3Xo&AbyXK7S
zgZTA67b3#(oIvyWCZKwoAQ}%8souo0l+Su7=SiBu*#j~_xmon-yaNh3BE&Z5-~wp8
z96C{KT=TBjV|^y=ciM!O`;#8pL)+0F;?5@dIOdG?+W`Tts4J)a+p`baw26fA`FE9X
zID87Q`S)?zYlgUCV`Z5NKs6@Y;0y9!gW8>aNV^X?T~sMS!*AiA6<`Rw5@--N>7z<i
z1M1an^hRdYd<9A-GVb9e1IWmJ{s#zdNN8)QM}iBHusvVr^lr*SA49w>6Tf7ghF9ih
zk!F_i4^Wf>jY%6%MyYs?huUtD{eoF4FtNhu;-rW0K~kdD8G!#{Vxr}zqq3Ubr*>K<
zqCG>}B1wD2GCVw9`9?~_Qd3tIBHwGGSqZX*VG#g5Fm9_#fSMI_DuWdWs~R|Gr&N<*
z%B~nR1Z|n&zN1OhE2AY8yMaOHR~=nD96oP8{^r6|7Fp^LjAv;$&E5vL!IkeO*da>M
zAo3vaEsBa_<9Nj!$h~N8f*I6=zg7xinN={S7UUzJ^Zt)T`3yKBP8hBnk!#bnQBt<Q
z0sE!Q|7%jjxOSs6p7FuPiQZOS8?T3AY@*lpJI(K%ttCu09h9k&Sru)1p?Ag571Oa^
zIq0kl;?hmOA0K#Vb1dZlk@Y6vQ19*k_y?gvQX$EbPN;-dOBfY7=ai*WwyYItvlAm`
zrqoF(>L8V6q!Y47!dQmWVmCQLvW&6}Mur(0X3Wh0{*2D^|6SMjzPg_4d3rSF^ZwlL
zdwJcj`@UN~A|u)nqx+Ya_@J6YU1)7$A8bEacF8Tqcr~o%k$Zw>;+tIkzeGF;Lg~$)
z%0&P~&rwrp_#))d%fNv5723m*!WE9td=h?EY(1sQXuaGl${a$|1$5W(pN)$RzQ%rO
z=ty)h&&3y)dHXJDLtdth9?7AFBuPzbMd0wC|7X5YDCJg{mU`>QKRba$@j9#p%@OL3
zqIuD2-gK(O>$RIjK^ydi^_a)&0m$&v<63?vA15&1Ot6)a_SH6;57r@1C-Qr4=MIIU
zR}48hpx+#t{OawS6$zK@m;u_%FPFiNMr}rKmt3XVU_~IeovEZNcFQ;W(AF~*k?zks
zE*Hf_?o)S$2}81<k14yMfqYCO)P`bX0Fp>4EU<YkQgdL%chqTf4oHA?`_aig<ie0D
z)oHP)po-eLksiMPk-_O4H>`ju>efAHvtb5aq1Za;oM#%K)Igx9{Nf`%Hd!knnn3@p
zXJ=^tRRxvuNfKD*&<yq={8EFJ6>;HaBc4s=(eb=nP*u5P8J*y3R8eWeI7lY}$&>5B
z{nAr=m5EJf^M%g|2h|0EY-8K#Qa?Hy^ObOzAzFb0ivUvsYQq}<dLLfKETC8}`_qzJ
z@+PX$;YqMC99#KwGhd(Y*u()LdH*jkUyuKevY;G>qV)xZ@3NjPi`Uw+(RA#i>k6&s
z>t5Fv7u28Jlmn=Df9)=&KQ(G$rrr0U&auC|Gm0goY_rKrof>(Bub_ElqkDlfRKm;f
zmf6=WAi_#U6&UmSqsdq?_U(~}_Aptx3TGnHbYV@}uPHX!Fy{mYnHeqOuKa0}f9V`^
zdF>v`M@|qx$=35}+&SMD_eVuCa)ovfa5-om?aNTM3tft3!IFYI4?BPE`L0jNx(qML
zyzAnFl)+@8!sGlyO`$5peQE6pX$Ez6c(oAN{RG}%#FqaOW4}RzxHQ7&Kex)tGfMMi
z<<i+{q0jGc2*C_+`XIR03*zvABZLZ}4ede-2Y^<%;QELt8+`|mu<Ev#ZIwv@s$LJL
z8a4lfRjsjTOf-f@3g+Pa+kTKMP2#JpJ~EOf*eA$j<S9L$g`I1XB%B8T&cN-hWq5kS
z6*KrJ9|Pc!p(Wrq3y1riSIRFfj2X6=d&%f@51o;*>&`eU<J4XEty6V=SdW9<M93#*
zA8I^(TVng=G?;*MzZk_&-|>g4bN>}L@^u;nPo|-Pav5OC?!`;M41(Typbu!|Wo+n!
z1~%bp$!AcPANnn%+Lu26Lub?D`uPQ&<;ftBN1IFx!$3sf^9rlYW|P(Rp~h=EqNH;a
z-F!?K;m2iHugj1@a!MJvuvZJ$)z&ViQ5Of<{QG!_F|z?2TJJp2V{d;_Gy#U7of>(5
zU%s0`3?VC`X9<V&eaj3xNSNLxvz`chGr{bt|Ma0k5ZAtCJ)l=9D7jx~Hc={D>Js{o
zcK|QOi!sOyM8`Ev!8KUkM{}y)x&ARa6>jfD<u#WJ7=QF<+hEAh4ofWu`oG~$VwnBr
z*x-Gioz&dIlv<@ddtQtDeIPuQLTlJPalG0$!RFoMyuDab*{e!-*sG1au@*D+eIe6V
zi5ImCHTD=s2edr$jTtsnc!~RX+X#lM?EGV^^xM;|DB|R)j9S7;^C3$yX>>WHSDBeK
zs9EIF@U?4b3tidef^K##40n>0#4Nqd$}@JcXRqoqh54ko%F?m#kWSm+TnEViQSpC6
zqi^Z{O3<O#68k2>)aFS2uWnR#d$&iaGRimtz#Dx?6WEt?hP!Q<s^8XSyenO2d0`Qb
zht6xtSlw=H^&2!JUdV!yIq=VJkD;?s5Tw2-=NI3^ESWpL*&@U*IkM(a6Yu(9W6j<C
zO&9iVDR(p84cjM{X9Qh({Sx*Sz|{9nu8B)S5lQ6k><+|bJiMw<q}e!5*zS64Ke}Rb
zIU3W7a$Rc@W+KL#>knyz-;}dSG+|50Q%M+$T$A8W+y=a=ziWq~)Cath5n}b#?;cCB
z-UMbj*_ZT`mG{Niu~Rfbk7Slh1R}g{39RKrzb<%c$WDb7v##X<0~hu|FeTi0^!ZLj
zFv^JO7b0xv&#E?{Ae{X2YAa7S?!HYrD$A;J2|3jSsxn808(6Z<@JX#H`tuv$H1_!>
zIwY@hl$rabZvBNdGB=3eAs^bG^MxAY=T8#jz-<9QhvpjtCMiHcg=`!*j1D26<HyYI
zAmnz^^Ch12M!*C`iOUxV7QyID0Z87U!uZo7L8f8@E5D8xX17Ra(yvRUvBi*#xVg|-
zqJYJ~u_;SR4>K2rF?<$&5OxgsM5N_y2Y@&+vX|#iXK^=$x0#@QKp@5N?u8*%CJF5T
z{61R!b0H|f7%U3rDQu)c5w=kwa=oAri2n~j{6Bg8<UXPwVJHiQ5+xn(J_ep;sJ`%Q
zA~$!L3AmfOYUsJ)ZAoZda&cFi{L281=h1NW=!=WbUxSebK87aDt*c#mJq>72^T(Ah
zL*qUyaD4X5)gxK~WhXAeeNT-{$uNgAa*iPTaCvUa3qgiV_SFgJNKPtdRI5&TxBU5c
z8?1x1y|)S7?WmzqE13EuoYb1LTE_F>&I+3bdC2qoEX5WlNw(qu&n>2qxjkll7rfO-
z!wp+L9#4ZJSfz2)05vIVg{CA-?YS~t>&$o?>Sj;IPYDw;`0fXUda&Qka(;P&)nfZk
z45Tg#ZS1dv@FJrFjLWTrE%4KXq<z}q&<qcArln6^-mkB3D6CGVmtPbz{p>@!7Z0M_
zVD>}G)9}Vb#yuPU5uy!?-^6_rVJi(fgY4s~^FY!0Xaf4Ek@x3co95oV8}OT$cT#?G
zNj*to$K@kk=rt?@w|a2AqN^LRbzA7>Zbfp8(+oDFczoHFBf(;cKw^VX1%j^Ozx1Vz
z8c`tk+@=bEyq5ls?+m<INa;0>kJjo8t&MFDo>=&!Cm0goQbC_1UyaO;!aI-rz%K<c
zxPnH0)G=c`={wL4S7_3sSXt0!SwNvPj28g@KqY6d@8A8O{`@x^UH@DaR&ATny|)^C
zTOvB~<09buU-@egJ8vPqtc$!R9UAgb_LMLTar_FgJZUG2-u2YJnxd7nVbf|KI(Kqf
zx2IHCXET@YRk#Q27@0oKUz3%g=x>BVh(pKGi*F3ZPa1zc{?-7AaO;%#&S|trSAxY-
z_viQjS}ANKC7MGVUu+Y7^@{Awr@&l^tD@#zKwr3fMX#=YUybVJwN|_$3N&TikvX=R
z_VN#C$@+QK2p1!Wkq+jt1F$WFEa@YDp{O(Pg&6={J_1_J!h80~hk~zyx$kO%PrIP-
zc@jUbw)8RnE?Odr;jw><z!XnlPq|_8u6e@xi?C2?J{he#YWPrl&_o8rkPJ|NO_;Ug
zne+pn*4$9P(22HyUO53-(Tz1*4x-P$91Db-z%6?@A6I$_f(LXZ2Kujo1`!9$AQ-FI
z!rIm2t{@zBr0+H`cEOV=To=~5AXhihe$A#tw^9WvZLhYe9ihDJ8ziIuxNa=qwUooZ
zNFqr(pzKT$^o0;~_o1|e#z_(lh1lL`3TM;qKip+Linb{9x(`Ek=BXy*^lF}O&G6CP
zq%VeKb(xO;-tYpA4*sCc`anyB7VjD?dAbWU5-_zNEmmj_|H3n^xbWZ^9|!9UAELuI
zU{a5urALs^P1a(P7xv<3%W<^9OQy#d3Hutm9a7wB|8iqVMJeBn!KpEz5cUvgRgZQ_
zUR}1`gdI}oJkHl9FainI^$v6%?fe2GV1&=K0R}KCh&Z`GWc6wWYWwh)o(T{fxP&G1
zo)kr$!}7ftV4-y5p=U?&MHGA&!=^_)lHcjV_i=Z~%jq}e7hMH`1rq-M&dAgm%R~yv
zI~?)GK3yQddUa`oTc7vkg~6klyoq-y6EPV+1zTwY_rjd8-#W|Ve14^(@lK|82}fQ5
ze!Jr#&pkH(&OVfwN=xRO6;A^Q`WG-K<R0P|sV1CRHO}9S6uz3^3x@wt(To!)46D>#
zTqbzmylJ+SYgA{&2<nwdB$r=1A@E>#q5(8W1xW}Bw5=NNpcx6<<AhkO^<V!tUe_9)
zzhr>+)<FI!JPzf3p5a2K_AW}j>osg|b}baRLw%q&exZETafyIHpS1X2&}`@qumnKf
z0OQV~b0`5_$y9dbhK+eA9E>xJKq(9SMW!b2P!a8mp?wh=W|MUfXFw`qojGjptM9+F
z211{{3y6c8LF@RRFF3B*r8;!9{_?+(6>dA><F9^)aB#)91JQA}QHXxz+w~a8;ePpp
zI;#PsDi%I`8pBV*7_MEc|0a2w%6nPLc7w@RzcFNi9s`|p)>6j{znE~){$(9yr@QEu
z^Jl^VQ|MRk;|V@<$)A5^1!JZnbE|wreXf5V>5!0m_iNnUj_LIGg(jOGoiFQk^}hS>
zhn4GS7CPde=Wl5kysTgNr%v?j#0ST!wZ~?4e`qAg^;$iMl8@Q;)=cEaNX!evn6z8^
z8nk=2#BCnzrhmS!B6j`Jr@(coURfz+id__iY*Me+t6VU}wM1F<#Jg`b@%C=KS~8hZ
z4lR(wZz&3%g+T;nJ;+-%8+QYwE~>JkqGDs$*%#QdQ#E)gq<;`P8{!<!YC(5E@fGOn
z90=c2*s}D+@dNEowVE<x)~(*SD?J?`&#goguJ?`_RnZgYm$Z;DlvQF`enufL2nUT5
zPAWGNUf$H{?}kdAWx2Opu41^T&N1EN-e(vS>fR15zG6u%GUfD;3gYDE%^;5c6Aq(U
z6$cauRfc9}W^p;wAD&d3C5M^Ke*4yQSx)ZuZQEL2E~Brvckxh%uQ)PPJ7WA>VRruk
z{mMz6fRw^_Z6&2YVuc(8u_P-X6?&Cwvy7z<D?vFme8a0gDi6P(pPeFb{ZSNzKkq#7
zT2_lZ2pa+%$>d2UBeaWS*-5ODu6<)8Br>!5`uqE@yv|k5hztk_m>It<*9RhdsslXD
zed%9C2=;F(D=T{!f;tLvY5{xOT;m)98b#2U`u_d9>B5C|e^-~6m$PS{bNw6ILEeUG
z*nqfms}in?^mkc{Jh5*YMTV}H_?bKB2!I!NJa{Db_nhCa*Islc6zSetAS}8m{#|`Z
zTz4^l!hBSk)NDRFVUApHMv6H>(H$XM7!-82)=rWyYaHHCW*-uuXyk5YY-~*8RA0(!
zNE!v4m^_;+7`%ex>u8+4#-&?Xmdhgq@0QR91t2jgV*f&1@^+4P7RY8ki>$@U#?Vtp
zP6D2RpA{^T1+s#aCKjIiUC<(!otO0&(BOH>w|=<wXDn`4*h5~7Yx%KWVk6(Ocg@|w
zu3QOC+ZKwO*;q9@pJ$G$|B$6pM(VAn-~qXZRz;8PSy&tg>3BX{O^Xh@>n9IhEF0vx
z$A*E`&`7@ta%3SsU-kw>p=y}*PL{f<#~a7n8SA0w59#NJ+X;lRB~wWdS^R=YW5}N@
z7o^kA1z(*zKZk@@mZX-pb5BB3t-=7<KF_j!M=A0HYn}d<4tlB^;&@dHL~gaISXlD<
zr=9gvH^N)CGbB63FAZoRp4dHWjij=DV1-%Fiww@-;Gp}>45f@}W<=WcjrJ;vv{2)s
zzP!9H4wGSJk(!SC9MOxX_Ho<UABx7WQMZ!{fKt{HqCaiT(gA4X1q`dzI8Vg4P?~x{
zx@bUZuZ$~u_39NA*T~!n!6ar;jo%XE;m1q%;1YR73_gol>-q-UH6^)OMMcHZ0=%b<
zl1GDE(l&kUX+Cd+W+>Y@hC7X(s%IeBDF*ZY>~N{xeH!hkq-FV}rSr!KJRT3R0B+q1
z7@GeYeZAR?`;=$yX1b*PUCF$EhC!?t3Im&v@Uqk}bQfc`+J=Frlgi+~dpz_=^57PC
zz+8vuj^P;aw&d!~#N<siSRUKo&(H5bt%+Ikr_V~1V+U{z#`}vSo!R^7{;t_3p<{^C
z&8lu{rT>%UhoR5l>-{z4wPTl7NbrsFsSR`D3cI9^o&Q*^K6<H9^jYVj=0y|1K}3Jc
zTI>NM)wUZW97S{Ifve{(^cMK3e+XE(tt=;}tStCO#dopOhyuI-xcOye81E0zJP@=(
zleMD?L@pt;)p?ZmOHUJ$VM7}5GdNqKcj1KOP&QsSZ<nB!#USTr(O7v*lJKzP-CSK=
z-_`$OzW7G`dGZUef68icG;uXPKO8d!T8K+V5cf}D18m8LgOXV6A0d}>ZDE?P_h0p{
zE+E8ONBA4Y2uPX7>OH8NZr(}X=~Z)!o|BE($vP7+Mny%{jKAssGk6G&FjRHJ0OFMO
z93*CPg|w~WJ&DJT?rOo`^rL%Ol}L3^b4J62c>8^jy!g&rV=Rh@x%^1fE~yJXB5WY+
z-Px#S9v=g1Nbbip8SW(OG2%rNJaikfSAP@pPF6PE>r|0DCbP*U{q5Vgj($x8z1XE_
zU;zM)eBWiypk{tm-N>bsu+0gQ+M8#@GU7^z^Za%)n`cXX8X4sSB8S1S<tO3`-Yjto
zo3H;n)#HuP>)9S`iXv%<Q)(H=XBCQ`S>)x&>AA9^<Dx!4KvZ%2R^_@Rwt+|%%r1?+
z4bvp+;mHugcy_HiUh9zpa$ipoZn394+L=7wmg$1}`hc*>n_>a>y_4r`d$u(G@EHIg
zBcpcHbJ|_Ml{zOCj^>>n3Q%~H7f~!_SFQ7Fy13S^RC^bR^FH+*6SUO;NdCQ<>W%+J
zEm&lqTB*7Tdw5}C!SL`}89Q9%!>q~r86#pvPDWx?-{%EI0nXwbsy}*MhL#q2;SY;<
zV`C|+8vt?s0QI%^>~qdpuF0)KBhJ45{{B`L=p~cqoklXWJ6Wk@o`4@bJBL@A?Gec0
z&&`dKPxmukPS&=T<Z){M-E8~%bU}9)O?Y|~F<}Q4;c#ktdU~8qI;GF;$ur>AMz=Ij
zqQ9RW##VnwlK%>K%TGt2padTemut33%d$MvZ225Fydsmyf|-I$7cr#h5o;SAgcw;m
z-hGOQLt`tutH2~%FNYTRia2IUdy#p&H^wiq)7MDtr_3lEbFAv%VgUFb==}Wjd#}V?
z`3v)X0n`#ttl8k34AZU$>l@KarmZhWaQ4ESHAWUmo?Z?oe~5wKd3Rbg>4Auw(R4xX
zw=sDxj~Jq%VVvf_G@s04U*&!uh84uaME=hAwN_xu;_AHIy$7L~P}+O%ajUJ-S`l*<
zZ#><A$de+^PT`ejZTU38!1z~z);BQe+Gm3Z0UeWjh@{C})yF2gq|CfTVwNIlGzbAu
z8LEY_=uEW|7{FJvO_W$a>!{5jCY-O~HB0peEz2bkzii-snl1i4?r@5V9#ypDFwZMq
zRUYgWxYS;6w>TIs6yC*ddMuK2+r&1Fmgi{S&23^$JKt8Ps~>NP=xzXU6qvvnx<SzK
zqC|SiCrK>gdQx3fi}-8lEb_%e_ah#!`J?^DeCMGS<W#KsNJJl%N{x^C3|@UFX%`Y}
zms}UmmCfuM;QdI5iFx~0i;Ba>;%378GaE44oB<ISWha_|O~Q&LoRrXX!HZ_JFPk4c
z%>d7&7^L$t8vGe5M>08{U(~{_spOZ#^PZTyW@i;Dbv*q*jaqo`p|8RNL6A<M%XrR(
zSKBf>*>}<NP*YPQ|Cp{ww;cYhqd<hv=j`F(A(%@U4pw-6dU}DRhTbvR)5F0M@i^y@
z%jl<^?*TJW#9=r)ZlUWScD~p+3}wD3m&z%CKQS?wg=qNW%y%7g95D)<kA|y9TA_9F
znDgUG)IB-i33!*wAaV(aw^&4mw)2dcRGQD3=4$>w_DcCCDmkg~MrUNqcf%J|KO3Z`
zc(XNcq+)IgWHOjgLb_b%;=BuUVQM)tqtnT=w(dWuDQy&@rA2BQuZF6_831SH9`@aq
z0b*r=Mn&?xO-|{S;+j&QB?!Nf^A?qff~9d=tEGrJEBvR0+X?(rKZ+jhV$iWfpUO$q
z$Ei#fMy21kX5I3kRX`g>ERYold%l1~a0V*`0gDxIcc^d>HT;MU@WPOr)R^d%*VfkZ
za~UlWgHNnP%=ZucsI`-4K7teApypZo@S5=WxVX5xek~fK?<^J6H`uM-uuW@SslJaw
zX6xbn8v&I&#Ef8dsGde9u*fwu#jS%6n~?`04t5lza#iZSd+A%dV^|SS<CQ!UdZ>Wm
zS8(R5*U;ukP&X;!yT&~3Pi6&++FcFpGb~*qjN{0H`;I&|9Ki2wpJAycB-sgj7py&l
z%0_8hXxkV&FnXd1BAz9>3D=t@YPt5DK=u(&rq8Ge4#&@8cl=Dc?5yF32q|D~>VGxg
z4^L{RQC>iKR(sIg^i#ofSo(T@*jA0|?c2AReGpFSIGmev?U1>y;r+D^E5#~&3U(>(
z@ro)W|EaWJ>zxK^&zn_^hjWy?9>p6(sbpLUs^6o?EWBn0C3{D;?YFCK(dd%vYVi)H
zwK90_9llQ+nn1dG#O<zz$q;s?jEWyEKkdQi#l&=W%DNDF_wkdk6ay5*qk^*;A?$f#
zpzl26f<7ZgZ4vv#KF&6%`fQCP!MKHck!4^eSnMFCuh3E)AUD^YmXuT68GW4%66@Ep
zqlJHd`q)8;Wbej&hFDCO<h!f^=KEDtRaM<<G46VJ4j9psYLKhVap;!iqR9D9cs~@g
zjXkE3KaO@3ysYJcSoz;)Q%x6g#=FrE=+~imj~TVm4gFWz$vz<_Vt)@!3SD0k2M28D
z9)H&l$-j|0)ONVWe${mAKmJDb1t0enU)nQds<!v@ZpCrtY~iiWm&r@1BRoB|qOBK{
zw41H#8?cz_Y4S-K=f?*bdl#!Az9<5g?-@o(llTwm5)tjH&5B+h55ojZ%To+9Q<ZQ7
zJb}3!=wxTB@p)i-K`Yd(Rvwo?w6jym&7rXC$+aw}Ze(Kiai}u#cymB&-N>oA>*W6U
z?S|Ohn@3WuQJtWS@%Qn5T|N}g4rp2n2x(Yv(CX*w>s#>fga~E_X+MN>2Te^)YfO{`
z0!d2)ltDA3P@?+XIyFHo=XT$qVy-SFR*N<*f@D-LyX5>E6Xp;46hd^R#ODYl&4+W9
zu6T40nK%0y+Zhpk>PABayy^P0b-7R0?6J8Kk9|^&aaDW}>9V-k<k?=bM^!xc(g_jI
zJlzI^Fj>3lt7I-HhC72xjq1f?4sx%NmE<}g5y>BSltSu!2g)$1yru+@Ob@Y^(($^H
zs(34y@xiS9vUHz<$IWGH)i>pJQ<}wML=zbwIA0>OF!)8YG49WzVX-zk<##GN2D&(|
zEx&J2Z`_Pmb<s>m73xvW2Zw_!^270+o&NF=>0TCQ5OSy^FN%8cpGMI4_MgI=()C_L
zn$qDb;fXx}qOn5T7EShIV?Kk6&;#dy)In3*+heCpXWrdBV0RUpmcHgx;g|WRmnyf1
z8(C9{;`(m~1`dpebLa+9YOYB=-zOR@J<40$oy-{MFZ6LFik-6TqOI>75NkPbfd8JX
zCbxSt|LcdYB7c88Pg?`rXoJ9^qi796=zD6QZ^TU)lgz!?F}(Hqbz~+J2>ACmD!SHk
z=IyKe$3$999yEQ=|3W?I?2Xw6$I{Pt{!bNFXi@c`GcvIkhRfbljc51aIqq;7;!8n#
z$js{Y&fDnWfG$K7IONICo{j3i*xn>^$|XJ49Xt*^YMu^dn6?-RwtIe+&7Y2q>g^Dg
zZYS-E4Pg`36S@-DS6~Ow7P36bv%7l&u*Tt7PA$%Dh8c=ioj#(_*xqmvt3sZpSWEhX
zc1XuQ=xL&h{>9iiL?it(=zZ>xQVuD%UCXA5w{(^)NN8VBS^VIE2MsafPmtHib@-8^
zT<7v&U8*MPYAq=qa6P_TT1<fpQRIC=Jnct#W&hW0bS*9}Vmo{zpA3NfX97!_S7_>4
zN!;@p3NI9-r>3S_Y$Prp6T(`b8cDtiqFTGx-3Pz$7LCVB<x*Ymp@F+qyv7?ybk$6>
z;TD1PT!wJGFOqmgiEycfGv!RX<9Lz*aY^&=i4GY+0cvV$RCaG~OcDRotShB`o)9Y<
zVBR}Azf((-wYw<HIefoZ{))?gi1&z7+!YP4xhQ$=_9qW<hF<r2>0h*WW8<ifYiII_
z$$f9E9_;oGanomJ7W%N7=@xOerNv>rUSyqyK6`Sz*nC1ydB1+bb#?m$UQPe6p-}{;
zr+#CB196p&-vJC>0X;lkiX|DonO{&(-AF2BElgPG9FB&7B{<C%hA&iWv-Qq<2A!VI
z7)UjcPhE2E#h`G(-D1BtI+eA?HU2sR(}}R3?k2GXob2gS%y$Y%=toomwJ5|$NRK9m
z=jP@@LYR@5VV9q(WZ60YLL{dg8!Rz-o)mW-0kGT2?S4nFrNN43bt#^MZ>GAK^@e<j
zNexBOY_X(WYemgNM=N}cA>}ShZu{`z!%+(?R}ZjXCAq+$prCbUHm^SA7<Vua#wlcp
z%^omFY4a$uw8$?Ke86CPiH3#?^$@M5r9o6ing=?U&xz|zMzTlBj!SCNXrqNwE(n1>
z8WlEd@U(3qK~C|xhrpPBfk~wR2BV=#*UQVDqPxUql3nlhvr~9IiFnx?DA5*#P&Elg
z#7E5notRu|0CCwos(z+R+Go-1`q|JC<o8@OPL)fI^gm!=V4(K;!~8B)UdggCsU9b>
z{5zSOUDIR5GM?r>jScZYbqtJ&7#QOSyAZLA;}CYlX*iJUhVn}Yfn8#*$RF+vsMp)7
zDekqmLQiX-yq6?_*~tAd_s_^&lo><dEpyz)+atW~#sXm6jv+3{qnEnM=d;8*`E>kr
z(NYp$u;@*#eDfn)_XFb|WMqV{ZTe3fokKjCbF>fOhO4!EPG*4u)a2=Km4a(u7na~t
z>B*cH_D-q8$MI{vr=V3akg?E+hzMhjC4gD+LQ4vvI8ql0JuaoQXVAaw1Zhd2(YEY}
z`i7$2+Id*|0oac_EyXf2F)_Ip5C>qP!^0u3$A=Wg+mO**`Kj-0E^nn+ct&lB`bCM;
zX+$E?>UbeHRBvcCV%a?Fs6e0~BhnV_+^d5PtXj86hi6Nh!A8gu+7Am9My}_!aR2CP
zNK(qU%EwR6ukqNl_^PyYtG*KA>AmXsnEeV(u{vH0m+_S?yXKp1?E<ys0M+Tu58dB?
z#Lr)II8wNmMtuvQ+);}VNXI&Sg#bJHpp(x~5MYN+e-&gD6_KteW(OERNFPUa5{rN#
zyjJ@?>bheKG8zd%)Lkt@rD;!Nr_xJ7LxpQM*Evb7vyt3EV=XaOr-;^p$=`^ATG|?C
zA+OZ?=R;Z~vAKO=eNOCVD~J8+fA`+oxMu6%!q-wCHU6#Ss0N5-Allys;s5le$EiWM
z%xTs>`!aBTiNjlo%18w`GrFaP>pdKgzVN)MhsE!Oi)UFLs8Ffu)V0`ganU%qrn5G)
z>j7T2axj%i?aSm;q2FF8`H$d%HbR>m{4a+G!M=z^g_=&w^z^jfP2r6chtl4f&g<^>
z3Acu%heXUSlO9}_Ftm|09>Y5>Y(O+uY}G>t(jBdK`doEI81?U2BM#PDLYpyVVS;6T
z>yee+s)~s-G?`)xJ{^|>7opP$*|q)mo;`2!!QLpgLi_2EB?KcI19_z1dbu9<49O9a
z+mB>t>z)EDTD{}XZwe4(FN&>3R*pM-qQIASUA_FRKl;+6ym%*&+}{G2L9m_L?<O1g
zt#?0dG<6I)lCiqP^R#{RpYb4DHNHT!H~t1mJRu=Eb`DTutK+K&MFC0)-y(IK7Cam?
z+I-&z^>8N!`R`FfX7TPP6ld?4gneIH8lxAvCA0TlMT8b4LuL!=-{$t64L(_?ikNqx
zNK?z|<L#7K<PcKP8_K$c%`ojb(d>8b+_|lp)%4Sya|tcm*@4~ygYRf+b6oV%vi(2`
zO94QcpEEra$wsGAy5nnvmk82CcCVxyTwwOpeTZ5yC=yx3V`2%RdCR850)F~~Z>_D(
z3iGm2EX_pF8Z;8MkC6v)2M>!Hj^9h{PD|Xhw}z6mg-)UV9`k__iwlE#B@ezsNdDWq
zzdxS0XKY0hT@=nzLCNAC_fvwd`@@G8no$IPN7{Iegs+j*mL1W6b0QhK&8j)CYE)(u
zh}9b(n{2eG=Kc6Z-^hNO_9i6b%9zaY_RxDJKU}b7-r|}xQZg8o3TIW%YQMq!OT1#&
zFr7SCV#|C-E|2%gve3sCuZm_$XMY5uqhW+nXN_8oXMcH|5hl5ccl}TsWBx86yzEk6
z8`E_w()Ek}B5s5`#e>al+@9sP3s3k3T)~Gfn*GD|!%7PmGdFu01%2J4Wm9S%g}&fZ
z48ExOt37qUNFDL9W6lLPZxQoru0>LAjfn?{g`Rxl`4uoX|HyOeGE$AxV$=kNnbi0N
zPEV15<t-a0fjEV-Bs6_6)-%5QcOBF>qJ>TEL<lKv<+UffcoCI?7m(>eLa0+hOyCVi
z+Ek936@Ys{p$H9)AHuX`>+;hTKZTO#aDwe!Kg%d6D0a|`P~ht^s>W!*C4&<NTtw7y
zNXlxU_FmN#4aeaG(nhiWns`L4f3gSZN}R{RV$HnqH=lm%u4?B%EMLcjcx&6!a%4XD
z<oL!6CoB=xF|^a{s~AXtL9Z=^&kZpTGV380J<^#<lN$8?uIS~-M&JHT%62<Uo@z$*
z8->6?xj>xiY$e6I)@<|9eJKE^%BkTaxkPk@pXV!7vyJeHhi=7!dM~E*TdDuL7V*2<
zS^pzfvb?aUsK_dxAJGlZeD&Sm_$vdsPYY9$Df7ScT$Uf*^XbqCcN-+&JJzOFZ)&Gd
zC?0|i7LOo|w*i%U4)!ZH1}#SPMUsz0wDfcyS9jLP3P>hw+98OQcC=jgwU<(`QT!q?
zgd2_0lp0Y*;qH3|URK97PiIGG?;)%uTx#SVa;F6^J!D_6*7LIajn9_0arjvEl!^2B
zG1MeI(<B~AAT@D+<oadw_TC9+j=V2oN61kZ+ov#kV%wp=olpgtrxES`3Qfn8I5Tv8
z=_CsarA0zj)kU-XDY~galleFM{WW?GoJ(b+)k|~K(&!J%Pl&v~B+L@fgi_%-q!`M!
zV*;aA+6)Mjw~T6j>h+~x_flDl<Zh?yoxHc4Z^J(i?O0aiH6;P3?Zt>C=FdgHAfpe~
ze+8sxri2!|BT=?)z^nLIv-4~H%*q3AIqMj&%+^*~sT(etlhXhBX?<8;kSWIKPB^_F
zrmYQf>X_JAfWKqr=VLs0m-s~gT4>!r07^^?KN#&V{*Qf2H8DPAg!Z9&8`AHtno*@|
zSgFf82V_4rxenmWS(a}*`>$*}GLS3j$O`KD&ThlUn<*|or}-tY4Yp?6N%CDi;9_Lc
z@3D89MU;avsyjPwwc&Dba4_r8W!Hvp9eS~TGVpKgtegc~=+`~8=!RZyl1T209ih)z
zsr*d(rh#NMw}|;oB^*+Q*%ewxL`puMQ83GYO4@K`MGh&|DSgc!={OzwihbrMvcw)_
ztBJ=JOJk|+w`NM3@mfl-G}<5!uI!;ErPS1jnixi;R%YM*g+aCgo7-w2ic~*Ep~&9r
zO*e)S&#PEWduR&O+9G?|UEe+3Ap762$Yn(dqu9rD2`{sk!bkwR1<qsRl9|-(*`GE^
zeP#v+J!Khp(d5pIw-~M_#Hdetps86Aest07AK9PJsfERTCk1C+ME$u28upRL4~P|q
zTM|n>67xNhjE=+kdW58$e>S}x{2w(sQyB?OT;F29D?fR?oYGnub0hCF6!h}PH1T$7
zrNbSeY}Ks4W%Mvb>g5g0by5>QN;hD1K7d1dBA!?}^oFj<-hv@6Vt3|Ab0&aOmf0t2
zcXM;Lw;(fT@jd>#N9;m8|9Ony2l@VjkDxcEr6mTe3V&E&_|1dos3=%IqyFma6}{wE
z?2hXmT^bM}IXD&uyM{e5?UA9cgE#~K`w6lEn|vzNL#woD7Al8#rW{?0Ai7zWu`Vdj
z-tWwGss9R63S}WQ$<B!LTYX$2;|tjEjPV0H|Aa)%m!$VlNZGN4ooKOgxww||eDlot
zXWw*OUH2QE*zIP%h7i%jb+vg%`{xR-CYOCEqq<}*)@<yCi<u2xrCGh<*|JZ*|A%vq
ziPYzwU51Fd{C@HRh955{n8T>d+VW1y2|5M)7CBaGmP=1C>t|o+Sw=2=W#aG>rQSfZ
z0LK?@*1!&vISv(Kvp9XS{r;9U$KUW}k+85n98BS5dV0Ez9Qp^%1i1@Qy~HNIOm#Dw
zW4w!{=ZTj8jP${9FN`E)x}1nY^L$xVVM)L=;uJuup1Yy=Mw{qdE`#Zi_RkB?M|LY(
z#C7NX4)81-&!js2@Cddu#jB<)pDba%AB{3CEFgZ9A9z&NXzZ|@^wY3YCPcRblRHJg
zyhA=IV&`KNfcaUX4a(dmr7Q|NAk_D1MF!XJmx_ry`uxPZzhP2{hi+qcfx3QicrUe`
z-ZUqA${o*)NR;whoL{O%m22g#`0b$Uc0X?QNUv)Y#}P;*MqL%2=MldkD|o9Yme8r}
zZ?QB$R-UEtUEficQd%xsweKC=uMOTmzk~|D&e)MieFW7am#dyZg*}N;?yHcyiQMtR
zDqEWIc{jGzi1Ad(arSL-&AqX7PJK5#fKaBbBsPhfCsmLO{uGwHjm(mF8*g8BZz`>)
zWiuhTj`iD%#3qZp^-h|P$qIhNthMjvI5;y-rJaPl(4@YpSTtxTU3@WjQ${t^^OU!k
zD-j%9MC+Ub_K7|6U%gNMwe);~l!t-%7TWx`C6LJ;7MDq>^t01)4#)M1k0%$PFsqs9
zr>-dRp(t*j+_2o2JYJQ+?_HEdoX)|GYXBzj!f0_#du5;6x8F_*Jz}JnOlN?aw9jt>
zBLen#Z6K?*k|en}W<tyC@nz9QNv}fBOl$PmXJatsJT_|agi?3uAmldHkz|PDKtdnT
z{`Jett&~SX!+MPWux#P}YdtSl;&-Cgi>EmcTUb12m!UNMhhjs^ut`het)v#?Ga}}d
zPy_ke)!?e$zG{4;NQ|weNZM2EZCtTUO4b&yrcbUG0DS9UffdtYaUO^>e5XVw{Z7<J
zbi&pTjqYv@{u8|BH`oo7LdIX-`7IlBf1|X<Anw{?5w}+KZxLj=C6g1{?VYzhy7y{W
zSXlBbYwyNs$V_s^q210qnXcb6PYW{LM=P5wgqh$P8W56(r4th*ccIB=EYA9r)%4Eh
zz|*Mri-|#Dxs~*!<s-1u`d<-?gN8#;3>=i4SuSD4qJ^sD<l%^DTOyQ}yFoTp4U^=H
zqQImUzexRze#E3M#dZwaoYMH0NY11Don?T{yN9?;pY0}DD1*o)7=QMDNNXq^gcwPm
zY-MSHgr8V5S4x;KS*@bz%L(27Nv3Yk=!#&o^IyNxk4x^d#n?#d+F}l?SPQD2uE}ko
zIjb*Z;cwS5N27L2HBej#wc1#Z_WK^Ihr~h+RMbuUKMp3>ZkXkg@!YzRxr5~Ovx@w~
z@BI1a->LDxGCkNK#Tk$vW#lW!KTaKvq8E4Vp^H>Fdkc)%kQs)B0f@LUFc@!x|6Up-
z7|EwQ80kIbJzg>@{<V^_1>=Fb+ugl-UjG#-pa4-IJ$XMS1`<6PMUdywvu4FUYWLaK
zO?5~o;5^K?4V1;LM7q$yQ$UqbE$vRCumsMwc{<Fc@b3F|iU7cjEeFK}`mU;(z>chF
z*A-6?J8=VXz3howlhW^s32jkN*C<f<y>qMy+|!|RIq9zEx?ItSI6b#)<U>&EL)N{@
z*|9f#4j2SZ&sH3_d@D@7d(2~p8Ove#ffi&8--x}-;P>7OE0BB=RAhX}Rx`e3Q;;nl
zZ5<?%3sMDb*rZfNfKiiqBw@9hl06zR1v!*+hp)ROYQ+PC$%fqBRv>Z-Z23pq*@6)=
zE0H%Y)E**K5mCvB{{JoQH4s@$gsSI)16tyGZwT<TO4>bkOBCmQrpuA#$ZMFF_CIvi
zK9Tls7X6fMOT<%WNm0Tj?Q8sxDAM<CYW~h6H_S!n{&=`@fe>@tNeorTy8z{HD8hMV
zI4Oe}QtZuVPuyaZNP04zI}bHnBnd*?hGqezRA(4C$JFLCIgiAf#Cb?$cRuT$U`Fle
z5=#*I4#TEcE<MNZ<1)B<{MZ=$gvZ^5WWi_*nO7crZ2&%>V2^_HLjV4o=jPEz#Mmh^
zpd9=?xOnUa#HqSZ^;BzA5Y{e^(#+PQQZAy;EH`|k(Xu>=L-MONr9BIFM@yRM(ZIrA
zHuPvC_F0%$PuGB`@&c;Wrj`q$@c;h>M_19tUUg|jPFDV2`#1fB{9t0}-SbKFiD(>C
zXr>p7hF&W}d2kA<Hxi!RY&L!vh@OJy;vOfgv8R#ZYCq-Hz&=LRG~52Y^{xEiEyJX`
zJM<g(ilodWoZj>)%z1>a{!6U-i?3Ls)$^F!vNS=IFk{%fd-Kr^m0b<+$xi!)FPjYs
zKBjfvzP^0)O^*8a6sKfK!uyu_8Izna70E7tE$*RSk9PhsiET6!wG9(lcs6#%5x>le
zL~@UP{0ftJrOlHjZ1b~z;`CG&iqrp@O9JXyO!RDRZ%0zjLlk7{(EA{W3VE@vZftDq
z{PX>Tj(wk_7JJUIe~@8LDuKH{%8#ijJl2HZB;?dv4u=zAv3Mg|(O&CWerVA^G8_gD
zJ&fr*yP%E@mm~-rbOlIe-`Xw38`T@qDxc*})w6|aD<pJTmWzAJyD4TYp`Mb#_j^ZO
zP?jU}`d9)WS;RENZ@n6}6*m)|=~0GSiSVBJw|>W<ayPPc49&%%a6{PB%!@QAxJHXI
zl$Bs!>h*75G~p>abc7BcJT$TVs_mj~tYM%Fa&z;c_L`0hjEoy%dhv-3XH}N#zTp@r
zwC?-t%AU1YunR}=XPAsg^uNRtCAaF${LO{47>n*^sgLT}iYDz{i58lM-)2(f;Y2u-
zpnx&Ix!p-wd0_=uSgpkQ^cGFJu}>PUs@eD$YGW<QS<YSqSu7BUFA>BWlDOY>qVt2y
ziIlvfU(ds*ob@^*(3$gaFe)?w_FNTbGlBAS&DDhzeDLPSOa4pB!^kL>*8wqE6Gmxv
zkz$paX6fo`k^hgp^N>0cKn2v`4es<VY{nLv$D^`o1}q|L3GSQUAd+k6W}xy>{xRf$
zsB~>JT*NIG|IfTb11gjWCWGi4mB<C#Gi}BVTcK5yE1fw2QQ5Vbj)lmh=Numd*^iQ2
zVnHg_T|h?#J~tp@{M=bI+-__-9@oWp^C2)%*Ay+u>4D$DE`VCjRt=OUb%08&PgFX!
z%DmFM=4C#{{(w+%<hUTQ3s_e7*B#Bkf;{vt99ITl5@O!F(&o?A6-9@eEl)NA@|3ah
ze{>9>FRE64-DyA&nF*?V_CuLEeid9Lb*U`|KO?KaW`zm01gfu~GO@U_^+wdk>^nf8
z&>>mFn3R>r$8x2eTCs}2)v<F|N_<-XvPUj;bywIxg&y}Spg}uN@z)9u2yt2gW&x$T
zER=<bZxFRx15+ujw1!z?$?;nCxsK7!R+21%nZpr@f3r|R&WGge^m5JmmX|21nN6YW
z$2-mCIjNu^T+zxMwc8sF1u(danJckLaT_LH>F?@g0t$`x@OPyFP=(?z=L-&gbKPNH
zhDNaO{vM>|wva5H6)$$wH$5punNgPHZs(CGBSt4c6yE`&SiPNIhjUnc5D?Y=BG-)e
ze*`V7nU#7<F;5OPBd*yU;R^w?m{9$gaPIdQAQ=+*ug7~9G+6OSU1po|<J5aTd0Q}&
zi<M3aK)`XT?NdD-G9xp&RqRu+Tj&?`WmuqkVT1BJg`#jIfQcH9x*d<#1htuuNMC^e
zh+Gq^uhJNq1_yj4(#dtmS0b8f5ionHMSV04M3QkY+D?^UtL_b&ZP^R>-`$OrodaWB
zVGZEJFh8?zP2JKSEA8N04))#Kn=?irSgYhjo~UoS*v%cz4IlzA)r&qWspxQv<;ii-
zg&~~xSAn#scHU;&G0|W%Np)Tud9jMDEH~@VPXT+;!rgWM=(|!MI4=4qZ;ROKy+abn
z&>M&<be!@w`6=fc>Y?MMBbDwwm0Yi<tE=li&%=5gAg`DY90=E!tsOH4To@HQ9C~Gm
zd{^^tC4lC7J`tj5;p>(n0s{kg8>gC3%tz46&~d1vK)eE@)L66pN@Gy3Bq@s$D-m;S
z0gC4IN9yixdH`V;coJ=rO!>pm!jftGZtLbcT0XH@Tj_DCNVE5=eD#aqZT7gV$+(fF
zwJvvu){kJ(ty;gsS8<ASeX#4=8Np`T^YuYU7O#uU>EVB2;*V9S%$5>?K~-d6DlqgP
z+Je}rp^#EwMcoLn(7gV$j60h6gVvIr!KN>!=sW7B3Y8nYXGn`+ZcxdIL*HIbxEQi?
zi`WC0jwN6Qxi$)VB@-H!>5G<EKq5J;K=XPn;yp5d;@U9d8mB&7ASgBXfMGCS)p=#q
zqP6e2v4zM_o!P@tpD4BSk!odh4N%oqKTx)LXh!){T0%E=0(4sPvNJn7zrpTJ#Bc4f
zP?DJfdJ!-`v!sgb_piNVTxYQb+Vlvc#)gK5F&ouZ>X3G^#=*4a#iJ>rW%(-4{g6>{
zJFHp+yS`t2=c&)9_W3LFp87vx<)J)QxdiSO#Jt(FA8W9WX3XI8`M8dvJH6IL9zKFV
zXJNUbY5peY`zK;Frx}M;{ghVgH?lBGB7l}IOU`ql18n-`0Shg%&BfB{-``Fs@_9h(
zVn7vNnap0AonSQ>WZ{~W%P+1%^w1jj!W55I&r~~#lG;p-l^7{kUf=Zt=2hL~)BXA<
z2UZd;IbY>HvrMU#^^QWzo6o)ZgC=2ak6d5m?|o4SL3e-pkXMd~NVDbwBzX#`A=#>d
zno|wUTQw~^8k(FdOcX~%Ch2Ot=Ylb^GvB|r|DJY_Lg6dv!(fcx#xzOjvSpE+qA4{n
zA=4%pEdw((F!8OdVC|!-c8&9su4H!wFnj@cW-YFd?>^lVjF$*94J8#@w(HTt{~!Qq
ze_6ceo)Cr;ntHOu;<L%=p!Y+}*-x^(Z>$D(;o=L1ttoyp(qAfFj51wp&qb<EcW{S!
z4`c_olS1JG@_c&dwGby?Y+;4Y-Q0u+^W`OlN1UgL=+Kv;9}p$p_c~Li^Ep+#C-*4&
zJL`C8bwB>`22T6z`Q8!3Jz|M_k@Xq4pkrOI--9#3#Jl9#w-1dZvdO8pTM0*44?xW;
z2cHf%RW<XYLDQeEX;*kGjdXQFGoV~K=E37bFA*q9sO+97w=K#Zn}?3g|DKF}2KDBS
zH~vI#{!zVEX6<0fJ{(_+T7-VFZz(6^4Y1G0!$b&{eSlj;aSCPyim>|{O8?IEkld{z
zcDtO*gStj5$1}a$UP!C!9!vj2Ld$31WhDBZW8u+ctQGrpTL^duVcW~JRoq?|A-deU
zRq-inw_>x%<Zuzci~OT!no-VsiIJbBdstoZixW$<wQUfPuLK=oh7d})E0I$@nNTcL
z>HL!x)RZwIZR~<>sR$4R4btAQxzc97kRO8W%zspEc07LT-Up@)PlUfv_?VNvKqDV+
zqg{WWLgaDA<vl1F!*l-<hUW%9v9c*@)W;kSw`U(S5UD!`9g<L+Gh0#@7FNHySe66i
zlo9?U88NRJna5Dp7m@5=`Xtf{S{^-5oX<~Fl!Tft*yBy>fvmC~?Jf0A9?pZpdrb^i
zO8WF89j8HnG39RL5sVip0v3ep3_<yU6gm9D?NJ#U8zb3XKs@Azqf_CMZUqY~G*;&e
zdp3nVdK-fGhj%kQ%m%oN@C{XA9jd?T$_g9RfVR+26JqU)p^|5Q!XHTU=q2;O&8?~d
zCXl@)cOH6&A-<AJrl@7(;|S-F)&GJfPzJ;K7DD}Mz5ZnKcL-4@CMGU+_&$E0+mA0%
zS#)~DK7lGm|M7r5>gHHWO;327W?eCjy@okDQZ1cX(GP-_R@z1384O+xIZQwf4G!M?
zw4Y%W$)j76C1VrnjVsS!BUBX-?RZN`O_yw)<Wy=S_J*_U-vdb$DP)6|*zw?+VNGm6
z0bJ!+zk|hJ_IO4`L_Qh5w;6A|R!aD;!123wUT;9{)^m@0#P~HH(uZp~QOQ~-F;lVd
zuByH!yDJ7rBl1Zi%}~tg4aZkwm}q7K?`eqNb~<@4-5Rm;D=%;?-r-PCdm_C&y?nYS
zs)@`nF(~8%K3JWzJECy>PbBX-$~J6P4hVHwZ4DuVj#ZpiaHR89rIl&7T19f{9l$=1
zj|u-V)mzHt7AQULpRib3Xn^NzNnphrsa-s`2g!%69VoNO(nK$0-)+=_#Rz>}riZdY
zOI-=dO?WI|2;StM6EhEF;>bb(c+nb-E;Tm0ia~uNRI_L+*6rszbnt)?5oTP^bt^vv
zEz>UjEFknTNZ&B9$x}X3b7sM7U4SsmY*UPmFOt|6zvix=DX{iX_S<DL@Zv9>%G3ae
zsO~#jwbP+8`gfS`_ZrBxm$h^6Vzpt{TwA7uw;oox3~<08iV^3hlQwv^--#<Zz7f7g
zqGg;`UWs`hYBz?XU!f_8QBc+t4sg4%^Y3n}DezQvJS6nq2${!V$kB2>8XCd0(UQiD
z5fs}*S>JYArGtVnlwz>iL=r|y@UXnckLV9?gP@Hb%Q9pJl<YIrMulPrB1~3gQBnKy
zqhOZLhyDPuqh^a<uIju5|G59$Yr$_1ctRMM-LULp5hrmhhQfuee)Fz@{p`9*dq~8O
z)il``1c!81PT8Yl(=1U~)gfufYe+nFXQPQqM*a|x0qV}8v>mHNyz7(U57NhxoOCT)
z94+iF_j}uUi?qy%y$c#*iEUP*b(2o;d{DM@w^%&j-eb)-^%3L)N;S=5_xMJsA`T~1
zO~~IxURUF3#w+XcgQ~;RxT?uN9}m$vCb5MLe2r1%1oaVzysik&``^|w(MqBghPYyj
zhay25N$hzKGVa;07LL*eK@E!|OS^Y9vGSl~+MSYGj_ZoLC^T&|JHUXug(#nY7~0AV
zhTm}<>IQ}OJ)&(&H@6joUVH8qQaGB)ayz5fZ`dK3OT+MN1>MvgBpJDs6uIGDX7his
z9>OJ;@k8vk#RBG-;dLkiIuiyBggHA|uab{LX(97Q;1Elw$*eu)NxYhpqHD<eCc2fo
zZPw_a&wUak_>My9FYlAk@L7wFH8JCPuDcltG>r)CA9Hz77YDN?k?hEkgCDfte_2H;
zR13-7LRU<FpDVc@>KFc2`M5gPJ}f{nGaB_xb@yRp4K%7FkF9id7!WH0Qf=8dG@|3Q
z#ttW#$Ma<Q&{Y+)Aj^-&3#fwmBD%6+0_%a&=W#Pg<_uI44g5WW{16O;N7|gSz?#iH
z<p5Z|Tf#F+o9BPH;)^Z|6cLB|d=?K&Eo_dEw5z7yNJ!wp7e%F*0~LF2JwWav#fB&@
z7P$5&cvv^=qqfpQ0#N|)d_FyDVi7lOAS}kzeQP^SIm}xH`sSk2WG`!Dqm<#QAgsRt
zu;AGRZ^o}+TBI4ubS^-Q0VApty`s599hWxYxr@N$a`AZi8wuo@TTtphOz20g@Vj3c
zQDcACZ#{C#I*yJu8fj><sa4IPov2*_i=x*|W}tl-+I|C$K$*&qOZ9IW8tH2$oMZ=j
zo(M_}7~ouVFKS)?-}{LTI;qbD41%LY-OM24ICI#Hugani0p6-Yd+Jo~(aTMEDBcCp
zbhUHzUT82bg^H3s8B2E_NG=(bN!n#-#1-*gG=U2f1$j<uJ`5so);_7P0DlpQ!a^F&
zYB~?Zn`2kIiaz!@+Q^%Q{jq9_1CQ>c`Y-3|)qYuGD$rcvi9_`*&{sk;IV#m<|025&
z(l@!`f7~r!ZxvIpg(d)vNE_(UXf>w%CD|rKe*F&GSsSQ)+I7j~zXwX}frFYDx6t1M
z?`$~)TL_#3nl&7-PF#isA)-0reO}>}P_xn+XTdX_)b-$-SAlw^b?rA_ggU7{td@nZ
z?lvmQLhp2kuQNJoJ64<WVFlq1l{Wx?GW!B*CKek23DPB5TQ+m8B}~uHY()Y}fpOr2
zd2vPgWu9*I+P(T%2q<DNSV?H2@#F?xI)a?*wS8QMaY~w(_6!|e*wT>F8AQEwV)!S}
zG!%g{+MkKa*kuTNm{M2#;=i0j%;F|($=Vy0z-PV%@qsQw@MMJdPs(hZ2I%=Bi38Q4
z1=B9htXnAcFV^p^4$R*`>TV8sdHRVaHi$IeK=YyV35Ue&do#QvS0+;MOyUyEn=Jo&
zzLOQNq<+E?Ei3HS^^`labpz7>!A*==m~f~f)LGFjRB7|0`p+nR%j+6SCD@AQ0y~;;
zL;1X$uH+nU<c0veA8BxdDQu<{iu-G{vG9WYz}-n0=l`}N1<KKPA^8iGlN^q4OlDX?
zdsusCww$ng1Q*OS5rYX`B1m&C)J634h(yIsB7D;CI`o?&c6#1b{Nys6L+>QUD=NQ4
z9OmxU?SY%4kOeo#ImX2tRkJI5e${SB@=lWKfO>2(2e;Vb!7mv<(5lN&MgbGc@)y7U
z451Rte?>es8hk3#4T`32UjZWnw9Ji)|HQ;xK-uTtWY>&mgFz?JtM71)1E_w+IVr>H
z+*8XBMj)Yca8B&np;Z5iegKpy*uCW%D3l2$*@eYUe``KhXs0vAyG0QllniCc{X@v=
zuc)d@7vvr*6f?K?cKcSMO^?ck*rw$(-$V<5rBF>nC8r9Kgf<b{a*o1Ua<Fmsx`vM1
z2<nQNit?@E#U_FUHNi06HAH#Z4{ExYv+L}VS$~Cgp$<4<s4lf=1;O}qh-XloqyOsS
zvz0--{Q}%Ur~*#*F#6_56%<Ku;Hi2jjtC7EgFS#RRgL!`pN<u#xQFqYp!I>e{N@k>
z6K!J*-<t_PLX6U1_3Jdet8Ci_?pI=sp1Kl}L(Jo4e3S*?<@fQzdGe$?jPjpq{zLF;
zN9czCSMvy<cz0H4$BQMl>^US3pulVC%-m+t2yg>OO)wSG#eeKM{k4b8pDzNw{L-FA
zP1=^dPML*SG%ZE>jjIsvVjz`G`TV5KPG0Ha7<AHy?4L9sgdI)T#VU!|u?TVV6G0{s
zX_*a{%ip~&cDa57%I0LLw15^q!WVHCz+E@xeKXWvOd!UyjZCkdt7crBI}gwUboYYF
zqa7Fy@Byv5dDR0BZcJq)w@{~YADPotgtHG(R(N)$zR6kgoY3e(=(z!MqT&$IL~-Cm
zBCX$shKAJ7?n&3l>o4vWQWnIAN^NX0OU2^{s#_b>1df8P0YOOnL6pAh8s+YYRcl6?
zs@2*^&o&{=>SE6o<ToerAQXDvobeWX|6OfdjXi!n(!7;+<b0ZBldyqFhqE7<e<T@N
zRHZsPZ;jd}RPy6c^z3LJeEB?2Vcs%pmgPg;B>4-T!<uDfQ#UbUYfZM3k|3`{UB*!~
zizKXT_I%g3<Jy2Vfwa?Q?X;Cz)EpFSb&oiY8cmgST}P2v=)*-2OPKuDp8^``bDbn@
z=a2(RI@A^PwQ`4uz0%Yt&>l48%O7v7q%VaD2Eg4?1Sf-3_lSK*BZIp*?>u#bUkiSp
zQAczR7gp`MBzCpXV=N*o+LC=a!Y$PPLq{i3>d;-Jm=wf6*ByPUm(eohMX-M;(%iF5
zAy_GVM7S8q*KvSgvgRg!zP~EWHGH*gM2KAo6m*-jh4vyArDxSmoiAX#h!Z!jME#LB
zuP~pTI8SUI%VK@bJBns2=pCyF3zfBN|L40f<#XQ^t=sYjiAvLPb?=suF;7B!ygGa<
zFlc+OoDgCK%v64aij&YT`dxhs(t5VzdQ&Ct2~XmB6MJVv6N+*<5*2sRwxMm&8j-q)
zXOcCe1xmKcl~l1j&4z6~O^fc5UO$cE5P$P>?}r(2elB+S@Hb+u)@ND7vjY-p-J_Dj
ztn<$sgnZ<oALnJ#Lu%zS`kUPc_R>w0z%8MVp-J_iy@k;tTkD(5&rAZhxWtZ@xf%_$
zPT6<e$b+K4R%WEK2w}&f3!aYQQttY_<)O$=Y>XjFlFn(5@^%L!uDE;EA7)Uw+6BOF
zKpHr#p3s^SZ6nob@MbYxefq@uJ4R|EkhDyxwSOUAKnto&m78LdTZ>MjpH{@F-K_W1
z*S<$%&JPf)XOe7eZEdZ{+9df;_7dCv=d)gb8sc^JfNQ3Ln9u(|t(5_#(&?#55f_kA
zha}GZMPr8o)lb$;vxrXYl>JsiC4^wSE>(JiA4whCtGQupzuQoa&U$lD3ih^!`}V7J
znggksH)@gd?AZo2)F|GCTuXEvW~V`}Hz7zuNwjW9BllA#!&d1*9wq?8Bsl&s05rX5
z{D9^0F%@C_D+GwpFe>06cI{N5<UhG3)5s~RHTP0*b!g8&KTBM7OQ=ct=gy&=n<zKc
z6t`R3*;5TM-;JVnBP-Owu8D@VDcZ4DWTY}l7(kOiWgceH|9G@*BUGq`ez?Tu&q%=Y
zCN02l?8%b)g8>aj8wRqti6YHtgRY#WHau6B_o11T)zQh8x^_js+ikFujIO8P972@c
zN81dj)eZm1Rh}>eQ~IXkVlR__>W$20<kYc2yh07`(e<WKm-uS`MRpm=McBP~%{A#h
zzmr#=+iHNAs~l(VJ)gufAB!fGG|v7GaAoZzcY4z=vhXWW3B4+G%5uxSD6_RRTk85m
zX{30^|6V8jkip4psSJ0Bq)-~o$?iMNq!W2K;?B!FZJ|1@|Np3Z6L_fC_kDcqm8G1D
zgg7ZwmZEH7IEPd!Dtoq;p+mM|(uhG?w9sNLYuaRQlzpodX3EqNldQ#1MuyQCjOBkn
zqx1QF|L<3?^Lm}undkjH_jBLZeO=cLsS|~dIp%u2Wd9Hx&o>o-IwP<T>YSs;$H|CJ
zw`>60Z(6Y5qQ}P>zVh`{;UR?euI+hje~0@-!nTzs15{kHU_x-u$9YmEUS$gLT`U@*
zj)N0fwN>bk(!~@33ZTNLJlSr%tCg)7TRr`sPu@?^bKZXBe-kKXUv&MPicy2;pK2+p
z8UXdSNrZL|O37;vhb$%W>iN}~Op4JuOro{h$g<&izxK&sa?$Bd!NmLx4Nwp}0?|<b
zqkmV+uF~hyixgaya?L9c0b)*8uF4pAqoAMUCTcP`t?i-K@NL<lx;P}zifu9Ot}!>Z
z=R;@-k`6?yUcmB(jT+mw8|v$bHy?{1#liRYZ`@=+Y(M3Bc8Uu(YzyhF8&6)&;0tn<
zs0Xu%W~NhoeNh7MOt1p19RE20VgQs4xI<OH6@w{Cm@kiwJ%<QxAQ-So(vJ#rZpVDZ
z-IcO!4Rg<zvFk?gQT{P8Uzajno;25frAnkn$J&gsrol<e_UL0T!X9;ITGAKYJ7^-9
z+-q;YSLVnI1|D<%pebZMvQezexp}vEKLn~zZZe`Y<tFm#L17JAM~7qmdg&eO=0?3O
z%Ne-%n~}3^v9V`)KVe@}E&TUVXPT@0Y^yc7+L)o+@+SeX+jb&IEE+HLV`5?==wjT%
z>&XM-;jW)Vhb^g+QynlVriPpa;0Dm!X8ds2@&B}#BamXmGFOfcMSrdVIT!z-rkK0`
zKXEaQ-#&C`L<q(M4`6Cgwk~EMbm2*)S4}_!y3RpzQwCjv?;QHleZc9U)9EVyky3gw
ziltgBFbo|gp*|0wp0*m!mTPwi#-&BWd6aBMBIny<=G%jGzuGb0xymll?R|Q;jCdK>
zXH9{J2|*~)!A&SJLHQ3zwrP~8!aQLYKIPMbpE~x)WN&paF_;-09k{8YdRr=gv%WE+
zeK(#W-Jwz9bPWUk84$-5-{I0xM<Di1+_3O~rj@@zE2r~^<h%ThAoutD?595^h{<O@
zj{1=H*;~Z!y&tIKDw>TomlmQs>wg3Uw+sKKT0v{VIKrs+N(hlJqfJekfZta#0-WBH
z$t>A)3EM-?SGL3u<&M3;FpRIbC0E^dzO;+^xcyCtVrhYOsM|Pxfy=DB$)L>o>$elC
z$Lc*~mlbXRQ75!_vww|j$WC94b^d(xYbfBEMR);Sp~PT;(gNWKf$MaiT8=~Gi2u)Z
zahWFDpc?~c3$jmToV!-cRrbeuiKtkHn5y_Yw}MgDVAW_|eEB46_u>L)Hzs>dN8K<c
z!P}l5bKH!>sL)tx=b?XyI{%3HfiK*L(y|V5G;AVGO<4wENKSq;d*>KByjp$hSl3Cj
z&?*89yy%7)yaxNE=tyTPMj2VlaY!6vD@C9+Rkq{_tJ97Z=cprQy7b7Cp7xMc!5!sk
zQm&a&C*0e&))GS9vf01tZ!!fl0_1;5V7g(1ml<xb@$W;c3VmLV;aZ|Qzzqg+p>Z_;
zih2?W$7hWy%~D?4U_$Aqjl-QF6hCr(*|J5yVlYBYzhT_j(BrOrXHi#_pqa^Fgl<#7
z^*aNq+`BWU7FFqPjuC$c_}`+%c!J<Hd(K~eSy)21)e8W7-G2EBI$c&MX9zHv=}1np
zPJ3y;H$;4fCLfXnuVOL=fMRa~@A?fu%x}{MBKBs@)u4kP==fVi-EhmQ9(aB5;<c-M
z=dEA1U>GPTlST}Ekk~kM=5-JAw?8GpBa!o*?63A_RRc(=2rHh%Bu&nQjFZD8e^Zq_
z50e{kqJd`+JelY%c*5Nf<BtlVeaEMN2ta9N*3OvTE*loGbxqv}{{Bw=J!@+FSC*VJ
zeBgM<aPce>JMCDW=M=K)R43zdW_tO1UWotcZQLiI0(9w!(9xe>aiA{E)#iQNG_Lbj
z{Z{z_z!j*Z0XJ?0uHM#D$Cf~>t^UVk=i>SJ2TR*Miow4jdWLPZKxWSY^9yjl_{5W-
ze+O=<x5rlG^YJbpYA0Y1923aj($oTYN+i*+2;{cJh3V@P1B@0<BnjzApY1M6`*_+U
za*lm2iyfmj(DW9xnwtTQtoAj_tU88aY+A=TOOChih*RIX%z){h{HVyG!<$;=a>3>L
z4ma9Z5$C6uq-W*=OQY#mANAK{TKAfQ@Sb3da?TPmtl=W{eMBbPSW_(u{}+^O*;;so
z&j_$c5NRM&Jr9YyH;k^2y8^`A*VN}-0riT~LE&?4z%xsJb609Z0dsEfAYR3))nkPz
z@M9$Es<$-%<8UE(I6?j)LUT`Wn%;K?G;?gsV==ZSomi+YGKi#BL}0ooG+-7{fcm4Z
z*p(T~Yn83P^JNVigtGDaiJnY(%ksxz@t`#EIPVy^x;WYVIJ8qZKV<p9)Z`aP<jzlj
zoL2byx*X^WYkZd)ACpoZ_k>kFV9jqNuQK_BShXvgh7FaB$62c|{YIgM`S|egaD^bQ
zvMJ^o=*ZSvmfycY?j4K}Q}^~i-0?(9-0*LUnspDukFF~9<hCah0@=_VWB^;xEXf#M
zt<*b;GwIW`OEFcFpV$QN`@xq6peU(ozLdd7#o3y+9%@1=<dRAejeajen)ESAxxrNV
zJ4_=iF`?TtL7+|UR^nu;ptJx*bLB4$Be-MPF5G(ur~cMaCx!`)=1~^x;#qMq0@-1B
zkE5+w-S(|bKyVpR9X?zvA31z(yHarE@30E9ro1muwSO-M67$&QoRD}o_U+rZjZ&)S
zEpU>!qr+S-4@i%JiRSM+GXvsc&Tm*K`5r2U@gxiyb~-=J60p@8+1u(ptMDsAM18n&
zq~Px$I&Uh1T9tkwf$3MJW!7=BdLOBEa^i6mgZ$~iw(1}CDQUgjG{GbhO(EMoOA15J
zPCTXzWH)OM_g&q5=?z*BirHI<*sawXQJDj~qnMJSSe~;*Bf5=ULXNq61+aGaT?Jw9
zBGO_}*$2dghgHCf<F4F|Egyuk74R|anwOz%Y|}tEokxE8YQ&JaSe<EGGd+x*$JlTc
z)BK)scJjOh?>?vynBML-#R+03Sxn^}O9$}N$0Z>Mon;zP41t&iclm@bE1s6N6Ezqd
z4vJnwt@%jydxS$b<UnDYUNdDaK>1`1A{P($c-jwBUdOo_U6=cc4amr9WuG*30&nr-
zkt)H|-+GUPS*1pPQ3rXhBVfKGz|vywr4syD)03Q1$qJ@9->f=As36B*HcIcl^23wy
zBfBp=eWA{o_fn&Y5KVDT@t$R3Y%hW&9DF%WiOnNZl@mJBNA%l7*dyzeA&vWFBdUc1
zqz$|=Nfv|)+0jP%O?QZ`HeTZ%qJ83Q;)`_J85HAMB4}^dfg*Sy(K2APRWh`)wJD$R
zb`;%2`83BKkX3+mD7S9(N%yYphPe8d#^Sb)l?!(V@|c4aJ#&zip~uz%HVC@297^(Z
zoFrgQF)EqfUJ*=qngTgl#FV)O|H`fk18)Q*!|3ECczVN2PUF-hPnKX9iP86Q6DbS-
zddyk_7-z2&Ys5UD#hVLup1jJAxWKzL|Ea*^)}sGo=6A1*3dDeljfJCEHOrB($SM+x
zJZd65ltBMjFpxlqAZwj+4*)|Xl@oW4B@;-KVJb}$Q8-9qJ{^*Jj?cF1SbmMjFjEoR
zI4PqHHM9ElJP~`Xtxkg32JeT4u^s>}A@7p27AVi!*LX@rtk|JLd2du6mGA@_b;A{f
zJ_8k&)Co`jz3(Jram!!*t1TEc+aeS_8P6>3y9GdZRrQas)PY6`zELqOj!ZE`YWe+E
zTJPa1i-%sGNWKqX5n6!$0(vQf18j~Lo!9coLUa{sfgs81`=3z40dnS+oC~S<bv%9c
zCfs@syD|H7HO%j{5!N}>`qM-FTgf9{I8Wcskz=@9|2U6aJsdz6(eln_qxjhPGbY(_
ze%Agh=-SjTXkyHMGBD~&OjR`leD<F}7{%mPxsK<3F(~1iV??jUFb=acYbv9+>Yo|h
zBk8U=yeoLXLUc<XLB$XgS~@D(B$w*S>V0(`-<>kh5WYw)i{7sPe8pQJ=CzRC)yy~L
zy`L0%$1F2v6Ia3RLyuDpv>b0h<QMsLlz%E@_}>?>Bf_od7CAFl2UKzWiV3Q{@)2%(
zA?TzCuM_-a?L4C3Ra;aroG9=(f2>Y0G2r}$^X<*jIoJGGxR_$G*!tO24;pLr^Kh*H
z3MZs@znS`JHO1&9t};L{u>Ony=A4$;U(U9rMMC!!%m&R~zXHU{*1po#Eh&nkP|pyQ
zZW!?`?;}-z*DJ{U5w~8h;9Wj0>Xd&7m3N$^81;rS(L~l;Hyq~XZ&6Sk9vQ5tt-oRl
z4$()yG)w<ntf8Ud)!f!|3Sdz(N*j!RQB5ITB(8?EIEa3b-c%z+J8uHlAbN_HL^&YD
z$h`um<Ay#zC%;YxSa>FT7YPwOBsHw~msR#$bW0ned}ooEGnJoX&}x)~hni$14L4xa
zuaGmEGEBW)llupskAk8VOG&m00bH-*23Lrpyjbhy_Nriaa91{CZlS=c7+dE=bfZ+o
z%9W(kClL8`*0n^(tQJT}sdrYdHH_NQb@5HkV560k0!8mnNF@W0RR-E8iR7cDS=BI-
zM@b4b#Q%NS^JR65<Z1-(K({?7&K)!9j6?=)Nq!GF<iWS~HCc>Fau4k>-Gd`2kWFG|
z|25@3wv`)nS54sD3*(Z|gl;w1-RRNiMbgbt834i!5m|UX2f~C8Dy|VG0!MK23v5{8
zbnCRjmj|J68EIJk4x*06=O29U(BS(R_NF5Rww+t&NF*rfaMXY97O@h|Ddi@SHCvMv
zvt7%$Djth9gx%Rvk7JT`J^6y(6Gw(ap}V_Ll{3NSk^Or^R+&uREy9KzEx~uJ)f*Gc
zm4U30-n=}cN_IRma}^~T@=!DT!M{47P5ZKa7?&b$wjxs~Kfg%VzqEyS#6P=vX1~0;
zm#(7mmc4SGdTr4y*L#~Y<YR+f9UKiw+j#R^dD@lc&wU2`fT}`PDg`j&mQKU$T1<I@
z*o6NHECM;ug4h5Cp=YX5wbphe>fHmDg~2>Z7<i2A-RQlFELku!80Ni^B?~D!G@<-=
z((J4Eu4QXlqrRiQ2bGI%>;^W{BDgK2kN9F3cseuUa8WfaQB&OjS6gZ+CfYY;JeAjC
zgu~@qi{@(xC3E>)AdVlup}rBx{!~;sQAm^VG^TW2+e-?cB>4<@{Tp^Xjk&uR@sYD#
zph5`%Q-9z$XsDUuJ#St50t=M4vFUu}xZPSqQt2%rLRsE1m$GtzuyeX>O)GbN1B<my
zIbqyF95qt%LZSflT0*aHa2zx!C<55fT$^PzW18akAe>T98-0>UcwmevF}Oa`Hy9qf
zAOf9u=&Lop=eY%&?p)an=sJ3?m;SGfg8VIGUB--o*2hMuk?&|ctqD7KTESo(Of7kD
z<PpTCK}QtSwt%>%7Z29Q`9c;6s@=Df7&Ai83#sd#yQ1(dMu+pYm^!X%%w2<qgXEUk
z+t+7#&B32>R3*yfSAJ*mShf~<F_YfO2RHV`JBx1fp06#)PZjuyt8OjGt6A`#??*`W
zxo1TsOqB+o)#(y})`qKEqOGL1k8D-D%x_5wUxSa|ep?{Xez|A!go~)%x_WcPLYI)L
zhC$2b8OjgI)wrHmhc&>rEKVMXwG6;ed7qIRCunC=<j3$4no895iRpNS40%0{6((vF
z+uC*W9qGh|Wdfdh<!&qukYVyy2&0~WP;jve39@5AIQzvOoLT6CU?Z9T7e*F_af8z)
zI$fL8iF|j#V?h5HYg)>0l(gXzy2nc~|Hx1E;O-R#^~;z`*8aftW|ajlyJ*brsf8m`
z>99H4t~$A9G?VL;oUuo14>&0j6<eA6JE$$Q;Sgy!vZbV~iI?Hz_>=Cklcp#wcFY=%
z)IQ$>L%d*hwm<lBAsv^|?3AdHN=M}Z<8B4urxp9+d3k^ICk?!ixFD>42ws+K&*hPp
z&|Uf#gQCm4RN$0p2qmn92edGJ$#|YpNt$_dn|TXv<wHt+`P}(#zPHOk_@2!%0-agu
zj$_okuzG1vXMg$h^a01o-)<o#2GFvVQ;Mgl>B*22F67T5rXm>v(N=_FZrUr4$qqDJ
z=v*oZ0B7eEqn_{F*U4GUw^pAHpe`1~B8N5FeJGYo4CrIB1&0ddk0i||W9%Kvj1HHI
zs-<v9iNp7Gmj$~MH&mZtHHFx!m&~$_66hHPeQHj`N^<6&j<u~uuiH468HEDfVE={*
zrYIMN7v!g}gX$Hx_Fo9cV0U^eQQXr0_P{Rju7>$EjgOJQ7%G2M#dd3zTt!&^*fp7X
zA+g6QI*&5(irhiOn-u2BkFz7FJWKxqtuE1N=AbC0+jGftU^&J~Q!dpwVBBj;Hkc$7
zRoz1T+7<tLvf;Oaw8Q`v({a$*&do%mMJnp-V=x?K^@^$<QKk7ng+yc-%WF_Vtjqzx
zcK^8cyXQaoDCOpL|Mw&Rrz)MSW*WK>8ZeYcX9TX5o7kjtw(7!0A{WtX2!B!+NT+&A
ziOg`!o7Dy3^^b+V>joUm)@i`Q-Q;vDy}4QWj)Ltcdzru7?9A-|<H$L+%1%qz{?aAc
z+yvKanl7TM^oXHRkQ{618`=FNo$bSC>areySKg3$G?@}tTv=3lnz*=y`L-VvMw^Fu
z<y&iuiHzuc*Tr{|iDAK*hUW@u!!OB1MSdUp9wu+o@YMivF4JJ?7;S%sA`RaOHj{6{
zkO`AY&LhNPzW-%~^5O%lgzUj3VzMzdv@CJ;HCxlSh1#ga2JFqC+6Nzz=^F;RoE-sc
z0E5!PJyl8KWP8=pPPvtk=?N}&5e#yNRa()_<#u$xn50sF8+ST|lu;j0E@`J_jTPXN
zA)HDywx@_Rf-eai;iI?7o(O$+xM-#YwRH{t&+lcI-cb35X2K?ay(pyAkLS6WQ#b2x
z9h}z~*7kdtY2$AP26<(-1=O!Y_xz@lJJ3B+sXQC!g8>pnKDKy@<j+d!Bid4!ibG>U
zl(>aS3TbZ-_n3f&%ZBjkeXXktD0#mlubnxZN9?&HA=~nXp&~qi`b|!21t|g;nUhP<
zf~oBEc*Z_6Hs85zc91BlE~LGr1leb{oxd>Wet`I7|7g*0v2e0LqO~*L9)ioCENh&k
zM5bOq1fziJ+`&}|j;63HMPveIt1<;Ba=GaQk#=rn8rbC)SlxX|nc?KMKN5C&zJ%A>
z+8ge;bE&Pg#MSMf*z-ifH#jr7#R|iWR~Ggwu6VC<5dBBc{$GSv4YFvOX5QY=8~Jtq
z<@1`G3_)Jbh4|@$bg}o68wK;Da~N+smJrFCzIYYacx}qrK=0XwRQU_}ItyvZksg&K
zHFY_NQB&9P0}D0sQ^D!!Vj7f=NMZHs9C%{Wn#v7L-_0CY+CKebI@^(rDDbdkz^DrH
zL_^}w6k1F$1l-Fxh>=*j@-44E3xIfE;8}U_7nfb5G|$+#icH$|PFpA%5<tiR!A+u|
z4%ImH$X{~`E&lvXRa-#U0isapT%~bw(r`HtWI@n(I7Ht6WG~OUeME!L6yob&dl)qD
z`5TZV?uq!v(}CDSGk#HU^e&A`@=)L35&DjIq+_##<MP~UR6AymkYC?jtwPY=q_4Os
z<EE$)<Uh`X2E{(&4V4o`(EYydFK)Z5tIK0C?4w`55z&Jh=PzfQC_V~O5C0+2ifaFG
zu~a%4jPS;b&cFqK*s$quVlA+3^NLO~s&KuFBC0|dU`9Pgc8jgniH{bS**XxjM^uSX
zzl(%a57EUlU`}hg%B(C<?!duX^hS1_MR}pw1il;0VoE2+GbT)bf(aarS3Bk3xU|`(
zXC8z$EzW;ECYW&iHVWYnkGw7fbgZ{jdo7e1QJqiaMQr`xwx%7^*I^oWPbskTsKNg_
zF9ta{TWJbuQB|?^z#lJ90)K)r6c8*d_*mqDku_dDuWAmleVn<v%$<A!{sH_KN=F~Y
zDe?CX)HghQNAXjhj4LSYY^egbJOLui*3j9T?-Tu6E=1KpoN{LSbSG`&#<H@y+4hZH
zL1(8XnxvD<z^ORcer3GImtUfl;cM23XzX{d*Ijk)BYm>QondgIoleAOXz5RN;097T
z2H?8b8Ht(#mUA`62B9G#A$Grfg>qw@iwBS@S=x&Wff^o9Y3Q**ibE5M?ggtW$TtE3
z1?_=j>#I)T-||)n*Ni}h76KSpZslTF;WuNGo&(x83+I=1VdwDd>yq_A1Y`b?ZHiu&
z0u3ZD&MwHt23}9pG3a&hmw_KuGRH+{+{a;Lw_xJmog)(3BgW&9obEX=l4x%!N@{fp
z*v7mA|CApXM2>C5uAG@|e<3WR$&z<&+Nh+vJ+<rb&2_o*aOsIqrOqEZ_sQhkrro`W
znc@hYR&aQiSw|DON2X3VYXxqLmBM{uMlABArF+Ig>yI_E`oJXDU?Wfs#K6U~*3;cE
z;I90%PT(B|MN(XD6Ylly?fEsNVWUrEy*s!MZtlenP^+a$?QbonTNafggBLI)$3VB(
z<E~iuP*2~f1-Wb|I6*;aENalhi1ei+yuJRnY4YFUv(EwcX#$<%3=?}#m~uD(ely<$
zRuGe$13=Y_J)-N%ix%r7Fd1W6Mf^SME@1|adiRZPSdR^XXcu{88#0iT19ElFkfbfT
z6QrOdsFu$62-v~YL;H<B!Rd3Tvyt0w;=2mh%Pm#1)f>4fz&)7FK9OrBVFzT)sI@^J
zmzY%5-TnNQe5uVQSgh`pym%(^u2NAIkTJU(V$?Ui$HD@v*s?$(d-MOas7TsQ&ueJ7
zTX<)Ee9}SDeGJjd;X4CX9H=gyXvq^%KNxU}+EmpN=T}*$8oLUUO3zt|FLc5Fk?oHf
z>EDKeN(p~0Esaw;QmYumIw42X!88C?jsDkY%0B^oEhY#%A`YR1KoXPKwxo31G_9|s
zyJ{NO*LV8hg-g^SAKrtDKgAqw_77GAs{;WfPWE!jccbr=-dQeWD}^MA&t?giN?Si6
zylrEoe}{J0=g*%jW@|>rW=Xh=xPt1#4l`9E0fLs&xyDMI*yudq+LfHZe^V2(!vFbj
zjQk)14$7M2e?e90<W~9Nj`$aQ=YAUIy2Nt~#q;EVxJPWNhGbug>wc5nyy=0QKA|yd
ze}iV?d)~sQb~k?~<Wm{HJ#xhXsm~g|8C!~X%e_0_hfhx`)>dMMEw=PJ#?#p(5IDnq
zgP@v?q)g1AsP^r?(eX&b8QU_wdTF&Ps+Tyd6NZh}yVb`ugys=Dg6}a_vUTaa4$L2-
zp#zI4Sx@f!gy_$G#0`Nplg8VpsTX1dWEgydHk}RJNTjPh3X|df^T_NEo6M|ELJRia
z70^kzAnO)z!oHgmK{aZVd+4fnmzyyBR>V&*|Adik<rrbJ_b$bU>6{FVLp$OzH+VCI
z&1LS%?%kCvG9LVwx*7#Ow!=X!2|GXI6#2EuYimR`j%fXqdC^#1IyoKmktM*@6k4x4
zt*(Eua}@#0UPzyO@*5c_msGM<LZW=nDIWYc?$ylf63EH4B-FMt=Ji<h(Eo|DY*~Q4
zRnilR{Ua)cIx+9Z5YSy(Do>J}Q&YIk_mxFoVDU-Wc^fdh+J2{>H#U&g1OeDS_oZ5T
z7@p$A)8S*^KWaB|v^Ui4TBs9KGGV2dN~cPcs3S7!*T%-~|MGfh5VMLn2ZbF46932C
zC$qGj9SvjF<$o*UBvIYHjV4Ltc=?7rKD}d~+Z%~<wApJ!r6YaQ3^u8Q-^8eDbjGgm
zZ2$aidC52ONfO#z_f<0Se{I>_i%Z)JC|aJjtX*DNLzZ~=+v6n^(U`3Hg4n(^T}Qya
zHLxO;&wRrk2yj*OKYM4N7Tq!&o@a*3&8+rstS#(T&S2V8@5uX35(YWj^gT71oFq$8
zmEp^EHGM8vVT{aX-btvsN}IFNcv%XgqE3>;z|k@~4q*l8DKWU<Ci?&(i-!z!aA5UP
z`thFdUT}3dAhs!s4@Xlc3?I<zbe8XtiIuRup>PvOEpQ(^bri%Dlo(&xy^Uawq1-=1
zILpc>{oPTS=TaM6&~~n?o*O=oE6%?={FdoXlXQw8_Lv|kRjdIYc{66}u$@71#@St(
z`Gjd&vUSNR0RRqj8RzWhXI7u4-Vw(-pWej}>6W5v9dHUM10@op<UdzBwD5cBL5q;%
zrG^KN$0)6t-%yf=7l`PHJE7s|zw95t!a#>~E6>mN;#P{}UI>-eTTb4dpF_Av>>D>(
zwz92%_IqB8@?#KY@79kWCV&%!Gye7YDcVrd%XEI*bLh5!!@<_JDL9^fby&K{<m_+J
zYIhH!lRv>K(QnJakOyDNFtVlKC@mKrWFVC(nDJLlxADVu_d))(s3YZhZoGB6dpJf6
zYUZdkBot7~TGMll?XrYEG|d;!^Id6ca-rfKV-M?i?hvZOI2|m_tk&G$;n$35;oTv;
z)`{AYmEzVhLoxEzZ-0|Z;oU0MWkwrdr4sGkuas(eca++(V9SMN5w7U=9duLgulqbJ
zXsgp1{qAal>v(RpLhRth6Qb5BgdAvb<!17`L?arQ!E@VUcHXG>Tzkrv9U-RrMB>n8
zCa(Xa|1C;fw|ylx+1~$qbKE@?RCjdVQv1H%!A7)G6DS!BgCPK8+T>gC^Ij-l_2G%A
z-ek;gXWb;$uK^{L#<6%FvQgl(TyO9M9NG&84h}#o96e47FGE8USZUzrvsp-0H&7fH
zd@*-WY3^}8bZ+{)+x(s=@(!uHv?Z>7cpdH`X*s6kuc5-y@9COmhKY|_d)=7P@3*nM
z?<O2wJHsV)h565zCYq?J^Yoo%rLqgyRp|l^&h<N#Cd2$(_4I&FB%Lt(x{Yv!A%UTK
ziA$L6J1BdI_Sb_Oxahi3&uyO}|6hYNGAbJvzQNeG$Y<ZnAb5=rzY?{5cw<vdz|X0!
z0IN0Sm#*nhC58E^Q$n0*!Q~&4l5~-+X_1$B3G*ymz|QxaCtafI49;I>>>lt2W?jMc
z0g#<&1zPxufWwH@31%fF_we13T|(|S!|~twD^3~1@<)8X4RuGMJR-;yWB^B}<Qm^{
z@#}T-mfy)36H~EmVEzrH@G+=W-LT>=*VVq`zOd&D-25eb$B=f~+`K)tsb%(bd0}yS
zas4QF#yRGJ0A(Y~(b-X#Y3#t5tyPyx)YvOmp}o{Hi+a<1^%BaKmnk;?<9yCkDd5;R
zP?~_={%1w*xi%bi!l7MndxH-(_f=e+cKH3+zDRWDEG$Mxjn9t9irkeNEgXpeOT@Ze
zG0$|)S=!3H)25%tt9&GPMCz32>z_MFWd{3?zqfk1ue610*kq&EDuChoInPrSWQWWJ
zczB2&{NBO9TpB-hu4r9i#5HV0X8Ud0@!SDK+_~4@4!7IN%macVU-b~p@0(|h%}86;
zP0%Y1EmlGkm!CZ~i6qGvzM1cqc=tY=>#WFU39_mj98iJ?eiIARywF7jOvydAX)VMB
z-Y=-C9@o5se=%Q`>Cv;xcAVb=J92Y*RQnxHDSQAyN1^0lNW;`m5Jq_RFIw>($Ejh*
z=`E9;MEB}Z@~PT5B47-deHP|+nIRokLb+8n@xu{kpfhvLzin$v)y6iCZo9m|*#1Br
zGfSL#Te|YcXA1=yW?8Eb$=rZHiK7O60mT3mMwKh`%Wezi`Gs=3`r=~^jKkTdlCfO1
zg~qJu)zZ>1P!)cJ&eR&d#-l<2R8e*rXqE=?GlK0>Ov)z&+!F^tWewyQhRB-X`X=Mk
zD$2xX?D_YEE#@uYWk`;!{623nbV-1+W!Mcu6IZV<vMZ);=|ncX7v9>_`ny5NW$1Ys
zVMPWA6IA`T%x^YeuDYv%dg@jnLY&?4c>YB(!8o>%-qu2^J|x<%WwdIhZj>+D@2F6k
z7zKN@*9Uc?c>DxbWnL~S4=EsGV}c^?+ti_`iTntcED<pWTHP3i1<o08+gKbJ(GVzc
zAr~DN#enLsy13gh<9SZ^)V#Zc;?0~~C9wGknI9;4eV1EZwfdpNJ8M|fgYT+Kju1L;
zcie}DN%ABA0kVQ4Zv0E$ncOI!mPen^rv0iFxd8U%-5{?daq>OpN3<eI|F`o?($1AP
zr?iK-W#dit1FRLHq3O`b))S2UDn8)em^EJ-IYrxJF_%$48M@G_6Y1eLNZ3ZtU>m}k
zChxMz5X)fqL*q-w*$<M;_uT5S-y8TFzqvd6J}CMP(q}s>0xSsjQU@-G9^Zg~XPZ8G
z*}OV%9revt&>81cMSW!o#!d1jY4xMb!`Hq?x9^0gH8FIqo~DM?uG;0QCR9_ZbhIPe
zA2a5)M3{d|Se04p0n5o;@(5@e4ngQOOnvc2IQ)*~@9rgOKwH+KEbL`|m`6($PEOf1
zDXOI)GL<joo5`bR5e6uS-2RDrR+EM683=!8q)15NUSC420aPP^^x@M1tpfCJ_t<&A
zI(-<QvqgONZWNRcGMmNY=ziuZP0p1K+T*Ka{_$nzd&I@AlORo%0yTbtn$Eozg7&Io
z<c2kB*!PAkLkL`!<yYTOXTJPslGsGpm9w86rn;_9fjt`xbCp*82|$UHiQ%DEA+PIq
zCTmfc1+#41L*;I?cg%TY_X4)u{QVl0hq#d5N;kxJU{QBxjdqME61BiYB|$ys8~iF`
zT``m}!R8FRqwZKFwFxqsnGI<OTHwwt)U>WRt0@~jp8wC1yPd+Oakz8VMOE<``@zNw
z(}Tk(GC;QjnH)*Ep5yz}U`8}(!@Vzx_=~E?A}rLMa?6(>yV*PIYcg}~Ejvu!aTlv*
z*W<*$)m_ElAV=l74K);oVZYVoQkQoFZPn~x2GS)QP`$x6%z6RWf$WZ6^`l6!6ot<l
zxiDWqA-@!)-0xd#vEh&82p@pGi^|p5*v~MFqMXGZ4$h~vw1`30<K94JT#oME?j>Yy
zGAMd=G}Kl0xkP?}kGy;S)?)q2_~%ee91TeWL)#%l#W?ZD#k(|F28v}c0P6gCP%W%z
zn8Tas()9!GFiBEdmyiJPLiq3kbeg8~x&u*<klrNE0Hs95^GBl8s5{^dK4_AL-8K2^
zDF)~8I;4LhBzpq2=1va3!3@?MTH#grWU0N>#3N-Q7)p7&TvSTuZ>?bjuagHW8;-6*
zCF6Bj9>ib#+=tKsh)n2A@QAixBHzcQ@f&l8hD#^@Xr7CWCes%JEfX#T2MYo|3R-wI
z_l%mgQwF|`Hs#s~f`lcvGChiV^+A_RO@3PXIYY=b(cA&B4!1v+o9QTejWN9<)v|T)
zxl2Ip+8tWaI@GKHXx{GTd#z*Rp&niA9v=HQUc$W>5Q2*DTw(tvQNbcg+#hn`_inZc
zP%F(>j{RLkJ7p%FZ)vLoOa*j+rLQZA^-U?uVi^B|@DapSAIfJI$ciPux2R6;dCWJ|
z1AL$q&t;mPd-A=*mcUopCQO*-KVdDm9cq6Z6Zb?2EO_)<`|ZrfHC{u$W2s;PIwdeW
z2&O;F0m@Ck{qIuGNQ__VF%ss-QJuJir@(O3ya2ge!)=fb<ZI>={o#)S{iCbds1BDO
z`J$jo*0TCXI-%o!Yt<+F+l2L{y?Z;nh5AcBnUg9<n+793MF@{r0jWRh_N5u!y@s-0
z(;6a+up83`|FY2fFzJlz2^4rU0uKJ*rH&l@>#13^h5^?<y1&ehII`Hmt+Yr^{<7OS
ziO#)Gpy4KD-k-<3kW?)0a=ELSkkb8O)^UO2F|$otx}JL;epr^rEk~D8C!;Q_8zxQ~
z%7>t|2LzDq+nEBablKeicy6isxpXPnBEKQT@Q-Lp+2wS5b?SlJ!s>Ls8Tr8eTN5E4
zg7_Ks#_?4AfBTErg7jSXW`Fl!DmL`+X<3;3$9#c<%95WSDusWSg8$lxLAPYGBf+j|
z14^S_E_t<;-ySAX9k^P|B3x8prgzb`w2^JzJvR%4j;wVK7(`h@PI#7e6i;>U4C6-H
zeG<gU@gvS9kti^a1e<PC^W<~mRWJR5V}&1xQGV;Vjrmr*!ef5Hh6LvJKf(v(LuV!p
zPcFfFXr5grBvoT1URUW{`p=)mktR!<Q)|{F`VB=gq%vArDCxKim>9-|x6P+P^ou{Y
z<JMg~m6On&y1Q0j%i?)=@4T|(<y+Z(!w|O4Ud%oPSU-93uGm`{O>WbvdJs=iK)%>|
zG9B6vmZ_{IfIiA{DdyFz9BjiL?NO$q*bgwNxUwmlm0=qSqt8&PG}`ynCXi_FDjOTl
znF0Huc}XRi9v7#A8WLQW9Zx)vIbuRI6OG=v*{cbNM8Q2i&wqp1Pd;k6rTBMWx-azi
z9i`@cWh?U+Z9irWq?gKF1=xxaN~UHqXZw;AlhBp7QRrf6X6N(p9(!@hyQ=GQ%7q)2
zH~bu&2kRyRZAWw1W+Jw7jI<&#3+$BKx2WB~e>>#aOl4d=AoFkeDgbXdNg|D<*#^)U
zeG1AOTg9S#XMWCnp>G_kdolbwsPX~9oUZw?P~;NCii}LoDy?6vWx<?*pAo~mE$j75
zTxW{?jP_Rg$d5Wn%9Vrz`87m&lJN4~J1Y#nc5qLwvFE9iYhYYc=gs%fljS;2dV(Q?
zkgcC{^FSHTa<c#MSQeD{=#g1c+5B%4>efr>^+4w_a=PPO#sptd?~d?tP>Y#*{6hZF
ze~Tupu(-@bOWdZtf>a;SS#Jm>bmUJfXtG9uzJs^}6<A8OyG&*0`X3<1$ETK>=zByh
z$0+3nA3hztedX47VZqMj<)iF5Jle_yE8+TYBM6pC?d=eG>f>fT+$u1p%34i~OfS0P
zqHh@AI;USLVhaNs1yc7N>c&}B=B?g3Nx7Pvo+|AL+nKY&_s^iW`0hmUxyt^PoSmKj
zvOA7Rhh{~$Y_2UWy@`d{6mykw0>oo8!5V9icn~9LaibTGI&Y?nX!NaVtjX;wS0~b|
z$FvAX1A|EORn@~GW;&c(@~0<1Rzd2ri)wT$WquxoxO|#GZkh8+1#r3LTL0I{cbvk|
zky}i-fz+~dzfS*gZUt^TKT(>DjPzJvoR2*Z4vqF!qDu7|t7b-m`@Ntk9Xj*2qHp{c
zFk+5km(Kn78B*7NOA}1Vvi~62oSLNqOt-gqZu|=BVdo|gX)G_dH*nUpW?z5-2comZ
zwil~8wFlSO7xtwS7InUdRUNXL+D6VgY(G=^zQLNaAC#dWb6cz^GN%w1(Lkx{*=30-
zutUc-sg2|RdRnA7*mZBrEG-r2i6brjFHjRGr|XCp$_}Lm1ebvvY|O)bAE5qIFJe34
ze)!3XWSy6mMkCi-+q3HH7Ak4)*}*z4<AZ}jVvoM+;@x-ZXWRP?D4KN+&Qe2c*&$H|
zh=B5T3MwYIw=^2@OM1CT_6B_9=?&9$`{J9B5iNK@EbU3V9>jDg4M<X}AVB+v_0?+N
zuZoqRVc%7<j<3z$zWfN9@$u1PU6yDKsNe|s$P8J@_^vR&2e9j`Dh;-5^#lwF`0K@~
z4?KjA@Do-}@Dk`&6o9Eo?;o5Qoc@N=RAlvr)|OwXRMY#gKy17YvL8?TUBJ8t)mWTF
zW}%O*CmNLV&Le`*erhe958cR4{-a}H$aHeBpNi3$hR9!j-G8;29IketKtG9KKBCGh
z?!@9fqS+w0YxwJwK*y-^Bwp1Ex++ySuFGuGKQ$;QY}bW6O|0K_b~Y(_M3^Ey#z5a>
zdvPm!Qs<muOlmZ9@NIOr?X8<L#R@>;BBcDt9{eHs3Pkk@rB+1DnU8$-y+AI&By~Td
zpCW(5*cNk@rKR*FI_xqdLee_?bvI2YA#!ns)VK1w|Jv@G5_V__sZ9eB?hg_ss(9bu
z%Vwp#UxK#>`v`m@Pxyw5-C+~;LT~<o)BeO`&395CQhb9EV7F4(_#x=Kc(yMk29xY~
zv1=$uc6^RQfZ9?tK-*Pl7p?(Q@E8Iw<cpA$ZLQKQzx}e>{{5~Nqc~zwr4K>ARLf<k
z?Lht5oPvX<!K~59K_QJj9lKjyFXr`N1GceBi5`cqtZ_gQRc5Q&EMk+X_<L+GE(*zp
z=|FRNbq(&%T|d(%R@)sFmGdr{sL|;+5W0D^oY%wJ1G6&v78f+IEEr$|N!91xg<1M_
zQgL=V{%>>OVp$y+IxLWSP;~kr7Bmu;typhQb`F;@SeNwQS$~4~A*t_tdmGK|mzK05
z^0%I<@)DKX#IxQ5hG*Srr#wHR%sfWy!kw>Lpr(T9Y3#Y*E=h~=E7hWX<eF>L<<8Z>
z8p%}C7EM^jE6o<zgFG}T05)pq@`~#qDXtDb*==(U2_l}}g#LW(6>#RXF$qUL{3SkU
zeKYIyH;B2KW-bo5qp>omiF3mns?5!_N#u~;$Dj?*bfX??j)yo#djuMXvHL#s!?22d
z?M&>cz`%~S+<Q#H#7Tm86y9yb1&z;ikHA9V2t2SoQvu|KAh$8KExdis+)YZ+?I6~R
zdekfxD5~C97U=GV6EjltS~DvKG^OTCQ3{JCPKNGArr>j2a_e9_KT+UM!Ig0_>z|u*
ztiN_H$hYn7TiLo*JI+6BJ^gFc`59}C-NA)z??@JI`Z5?-weveFKk6&-O$W*ZFl+wl
zz)+pRCrZ)sbGcx{=K`^MIL5nAkc$8cLr?co8!LQ6LzvF;6B)Zwx5V5=d;<Bxec3*y
zv<8^?$U!>K|KBV>7@Y*dv4Q!czjJ`~FI%qv#@+{b{wMDfTre0@UqDATstQ~##ka2T
ze8H&fG`6bFBlmV7!YSY&HulAn>BODRj+*bg7S*eNWK;9RE(vK^!!BJo<Tm*Y@0!VZ
z-tFi~_#7tDaOBh#?cGs^0uNNC@p7?uMEoa%I@ZB?>|ZFH-E)PUwn2M&;=^wC>TbT`
z=9^GL=vkOsaV|)72=>d)6H|9DpG6|9@cQg3yvjFZ)Me3#3eAgATeIu%Cl42?F$g<~
z6Hxb|>e$qtUkX>&-xLhAkEaK0TRG@LTPpRVzMaYj7nOyD*I>UFvD9N3h^T@_!yh|$
z&ZjQUBaHZk(!quB1>@)z@KOL=PuKLuXW+_*e}G8KSrYUkr};c&oMZ*P{}q^Qsi8vg
zVks<SM<eO^MMZ>i9a{sTkbG64`%jox4$zP1rxmbjK?5ee<7w%}Vlhv>&dBE3xBTIU
z_3pCHi<?;62$Huh+n*BG!EiUKrK0Z9#<%i5*8!QLN(G|;yX=GzsDKa+blq_QPTe4{
z2GQs5_Z#7KXM`hv`btx~wkEx7X`7wTnj6%@(j=t1%O!FtlA^yM(DMWst8HiF-o&m-
zrdM{`CNcM6FgF;Yqqv{)EirV0iGfH%K4a_4f=fp11<1(IQ;>-|pPv<FyZwxb6FXoE
ztSt%Qo6{NK!L5V}B-A+;1bfy0m^AA3z?M>Y8Y775jv+Fw^Hz(Le4*f_%EsEwkaLAf
z-mw%7p~NToZ9~k$5oX=Kjx;cuZ{=Dm21a$FVP1{bIdR^FDn}fM9!UF<f0HelZ2;L#
zg<~ZsAt;@H4l&3Ny5TcOCq$Lt-XMeP^74Z074Di%e^+`+=KC!*;#~^Md#^e5#@6+a
zz|W9buBW)cl!!0S+U@>D_}ZPxr<Fn4`s)wSiaT(x1#C9}eKxMVEqa{j7W_dB)6ZaM
zAM$_ho-A=duSmGx-pft1q+WVyLBPJH9Pxw9UVJ<sibn7Jf&9KOk5q51%)UoC8&P~d
z5Kjri*`LSiTp6mGyIUZZvvu?{Qgk+M+H1&s_KVLe7E({BDfV(3sOs@%*WKo>=1NyY
zmgxxC=Khd+COA{92bH_DG#|%bZr6HM6ENNx{Qc)=<Tq?E&h2>i*41`Z_B)>uPTeBR
zYAY-(Cha;gqk<;V{Y;6dMR>%}%zRo}KwV{b=!%N8?uwMje3VH^CwTrVv2|^@NOwy2
zc%r<YuCig42Oa2N{iS#rwUysBd>yv%0%rdK8;k<6tHD{nuS(!jrz6~QkFP7e5DgW2
zdXtaYfa)1zdyrowdm!(5I0Ghq$IAP+M3KSx<#?s$NoSi0ltqk8=2bjx-P@5;;1c+&
zhO9-X3qTr2aFA|y#=6F@Y@p(>Dy(AXz6u|1C-XpGB%}W2{8Kw8P%kE})`Xe{@CV0`
zG)guI<2Le*CpN0C-VWk-_Ta12opUqT#o@(p3y#bNyaEj5QQxEW&rejFEinnsE7j{S
zTCzrsNps{y68-dr^ruh1tQE@B7Ah7_5v1xUc4yM>c8_nWP{eqeRz}aFt4R)v<^5xL
zNI0r1m80y}+oZ7UcS4Ja`u3mPHpdh$Gw`{a!=f0h-s+ZLE&x0IY=mqg&lxPGT0L3(
ztf|HD=MVbN;0gQ;w6jM*^lQcl_b4s0?s%rvMG)|G)=|?69vE?`H(-gQ7yfktL)`?I
zOVXXy+RlMkHBZ!5_+C~>-e=r)PEb|NJ=uOQQ29pG)$+ied_HB7?Xf_WZkKFqzBKHb
zeVKP#1^(IM2+fV(#>ZP~Hz?F;Z8t2h>3(5?_5(4iSjZ`PU(KvR^DjQOwB*wsZ2qL-
zIZS&CRypc_hD-t?`nt-Xu+GV}*&s}}B_wZfc_tCuK!UdP<UNFoULMY5wk@2)`G)y(
z{wVuA+(OxhX~=6fVom;WdRfUm8A66f87PknuNR~oZl9$jY%Lc|l-)wB^^Qm>IGMo;
zhT!WuZ)E1ih|puM9X{zcvew(4TZ0{pxUVt4S6v8Dl7i>yscBk`w0uURH41^>_k#9j
z<KQc=&Qq&SXc-_es+C=6_>u<P`TG^dcWxv*A$Ff9{%}?W*chEhRbpnAuegMr8+83i
z=Kgc5b0s*zJp?l|9WWAXo>!!r8XV;ik7LG~3i8D|nE!pp>~j~^vz9D%WcjiF2W*UA
zY&2a9V_%j#X*;q@py08@t7C+bh}!O@^Bu66d75wC(Nl(NCbfi^{2@T0CUKgc#N-Q%
zY-N=%VWjjozaq>P*QTPhZU3KYp7?KDVs>j_dJdg^sJpikkPpa|k=2-}<KW3oSJ{iV
z2l6E>3f7A+2ju`ib6etm1RY!V<!5;xT#wdA{ofcFhwIP<hTem0UgH}<ApcvLd7d8S
z<p`CKILZ~@FNSfFG|c}50hKyu(pgU>DP!K9?3igF7Zv}vv(1uFSLV&wxkXwGutT5L
zA$I6(Gb>mS^+%Abins|Dl#$cV+l!5s0v&*0^q5udEc+~vlcc*njE|(yMbxz&-L-Z4
ztRj4~$JliV9P83dd$(ClqEd5rObu436llR_TX2C*kUYS2^%|E$qqwg<qqekaWjI6j
z9iXcPDUcQ<e!IU7Nrr%H)C-ZZAN{v;zWHsbZc+P3A+Z}Vge=Rqe4EuYyzb-CUbL(s
zXR!TG-b%TL=>i3nLFSlk-Swl3!=mnsrUDp;V`FrusTy35UaC-uYpKjwAZE>f@7KW!
z_O(9g2^w7d+q?$eEEvbbpjrsz$**s78rL!5ri)FBxm{Yf$G#6MKzY;5wCcDDlX^Tg
zSX5{3^E__fKhD}3tfn?DuHOp|g!+s0SVR7ph}?eWPbee66&FgZJ&mWn*<^#}_rNT5
z#z{{u^=!hAAp7>WN87~Tz1`h?$lvnwms!9%HlnNhA+l>IYxIZtl1#_?%&cE9(HH{y
z4C-pi&aaQwp)CCE@7f#}Blr29yqr#8rI!49&)awvpQ*{x#mvQ4{A-_u-K9ZmF`l~^
zdxmjJ7h6aJ2xIzKEEhFW!3_V~8HN0%Cwo8N-OF|O*4;OJ<`@zv<*IU>p~(&?ncdOO
z3kf|oZ-HWgJ;5e*m$~CRyz_k&%OH$3C9-TfB5JmrzSnum#1}fCSxMX<1y{nlpgL*h
z1)<ZLSlqYV68VQeI`xviR5w&+AWx~(klZI~HBjswB1gLF$ypfqDu4YY_;>r@1-xhi
z+y-WNQPtuv4i{@{S>eMt4;H%UEa&F({C^5u(^H(jHy_8Bf`uD>W|qmXxxd1H#hy$l
zWiCq?lkHlx``q)>ZS(svP7j^wN(peten0B&=#9g-g8z5jCNwd&UlH0ZcFDm-M5U*2
zYxmwP)%sPy(Pc-xGxE2u#d=3Dm?Npc6I8Is#gb3Y%)H}9_AaQHwV$f}RFLJ8LOSU1
zj8;<){4wpSgC%+J7u)V@TYbcE$I<qyPc4xFRBX#dDx&gQ5LM42tM^Wprte~CugDLH
zj^-dc__i=|(KLSv&w`QO;VLIMWg7<n=xc2qXmYIQ7raYl1Kze$%lc|VUQ0t#cYmOp
z=b0%Rgnw*weW8BEHKjRwfGOB0q5@i(qyDGsif8unPGTxnR%V_%Mg&^P@8=p{n<T8R
z-2*S0Uniko;ClDYCbV(|3m(7gaUzsv_7&zy9B7@r7+PEJmGP>8>tL@hqVDaso3mGD
z8;v$RycT3nq=OyK?~PvBr`Z1luO+pO$TJCIRsqxxm`{J5janMm7%kL&Nb56*+73k}
z;d0>0@;cmM*5i7Mc4?l-F!1SB9<omopOr~v`j9W~l-fUr-f>RXudRrBI!mpqtLFU-
z&}MVb_}9&NeJozvZ&~^EPo7%kgx2tRT~9-ORj<X(!_j`iI63m7iDGQ}%f7ZO+FY2w
zvYTDKB1}`<SsP94Bx8N%gc>4JIV9bHm%n|wA(sk5JUV#~EtJ1#qihy2o`HvdC{sK$
zCAbcg|5Ssxo=|^tu<vNca(6qELAbC<$L)L8WTz7pO5IOF@qPVHmgDdb0e9qYsR?7Y
zTdN9yv`pn8+a*g4{{;aGxP)R}+)?ZsT{e7DR3UJ;;nx=^{g?iTQdc9Gu#tFdDws;X
zfBXwsFg}07X!5{arp(mSdnoMklRCJ=b5Je^z3KeH18;kKL+gS$Ohk^9e>S}>xp14A
zA){S~x2-d>Wu}i0ZXki=wbV18sVpkV&YrjXohnwW@}(|JajtAp9+ZGHV$rIMV*)#d
z1O6%0%z(@dVgd~j70j9|Lcc{5kn7d(b3HIvvg_kJ$}L@gEfF#pr{@KS#LRXjF^oJ?
zFfW;>#sfMckNx89aC0omBe${%S-b+HGy;bmaokky#`1=ITy00eYS$E>?=aURy`22Z
z=D6H%^3a+sLXp9Ly=;8dGbnsQW^mU>;Qrnnych)GrgS?|v|Q6VJ`EN18}oKQb1xT3
zm~D;%_soFE-6z;r-3fB3HGJ6jzyjk1G>Kj`_qD+YRnTuvOxQ%a{PGnAsY-cse!$e~
z$sP5k^AQD?C2moP%u#(CtXF;UnEYj!j&Q@+?D=SfT4&_@nRnHqPJ#_?2OPG!=qA~`
zk7=x||KDDzh~(pHtB^`+p#}K)2MJ3F;6Q)06$>0WFb<ZG#)Pv$2%uj7A26{Q>w0>A
zEf`5ifsRE*H99VI^Q4I9i*AnbnZr8Gddw&7DB3~?T@iu|j{dQkN177KQ)S*693FXW
zL`BkmPQ(`L_Fns=^vEX6WZrDVj1iv<%I?XPg=04UhoO10h@Xgi_rXKzURGVgz)i_N
zeOSKz8oqCJm)x3Ed2C+UEfeM|)BGYjZ-J`<gJW2v^6zEcZ)RqSNd(SJM+d!5=GpY!
zYd@mAFfxA*-z`3$IiEY~@#u%yP*tkb&O>eU&zN<zIpFMBR<Vd)@|gloA^^YcpmP5o
z@qBI1+5f^zZZ!g2k&{KA9^c6VO}Sqh?(S*7?2x+zeCmcAoZ>9R|Fq%Z=5PVfsR(!R
z&BGXaOX4&_5c33@)p+8Fl@{}#?v%Ok!Qn?fUWKLU1nE)|C1l>h{XH=(Y9`S{`&$t?
z{&NI3CVLs&y8SXTF9PjD+g8W7y#!*m?L}AREUAKHgw8Bez$>?X<9cWJ@#8S(!aRH4
z1iDsNbNk-jW>aiVh;CMby6{Ly+Sm#2;RzT|8xx~6r@Zk5B6heBF<bf)>Sd$s6GJOU
zuk;Mo>8<y|{&*aBi!QQ32sUgF*Z8Rrcun4^S?oExVe9a1cE8?#>tUM?_!x0EL_s(Z
zIhN!B{Q={+{J(VoL;X|&UH8_isCL`r@Y_|H^%+uo-px|DZcntB7Y8eUbal*ZOLG=V
zJUVrg-FLKeBvcmW-8Y2iWkv>Oq<b{ap1@1z=AfB7KypRKcpFbZa;bJgC;F_%%hmR`
zS&pJ@4BlN<u@w9c?iSf*YE^#~s-i$4Au<*NPi)KO`n+bErPhtXS9L+uJj$bjo}SR&
z%GsJT?A2xE+zgh*feUw16fw1xGecn4a~9)&C-iD}b6`RjNeTF{)!lnDCUdC0P8kEy
z={H59>~AqImG@ZP<x*YWRQ`+x=xFKjoLJ&TO{xU|F=%8xLZ=5Ni1lxe(7?+b@-6!m
zjA~TX7AmK<XxLBpgOSG~Vz&J-`w9b0ZNx0Rl?M)TR=`IuFD7{`)E0&*6F;)&YQt_}
zLKEg=P*M>)(wp^!F|Xng&>^4i8+qFYP3*;FAHcuTP%9E%XjO|D(b-!Gca#c6_qGHE
z2E9d=7Kj%ZYN0F?24sV8#b}=B*#qCsg@%IoLAz~<g?{Xq2hYV~CZ!07UfB3l*Qe+N
zaZO3ZkC2UZl=EOX!Z~`ISX3dasVA6{{bOO6nKE<g@9z!8DZ<x;%<HT}0WsQ<cWN_?
zif{dY5jFP?k+%EkH_?)NWBD5zYevp0wS?c9<2fhIQk-cIMmQ#qg%T%O17$3mp~+LX
z7<qH^h4;vgR-#U8iHop<VGGkD)L1Hn<u@L?L+;U_fKb+rs)~aW5-0WAUJ`&4+@TD8
z1BYW%x>R!XahLeNf$zJnz=t=4tbe5i6R)SJG#?t4?~6?p79XZlw>}U#3j&KfMj4U3
zzPg+Ab`(=T7AwLR$~K0!w*CnnkBT#feP5tcf%Ze{S$fQ8slorguh|3wZ2vSM{@|Nv
zAyf!{>D*niW3f<wJJiZxqt(5~UMQ?sHv>!dnbC4qoC0jXAZIob>+EgFeA*4Hcz%on
zYaPzvx4(1+l2ur{XPh{^kX{>L6Kq2z?Kn6LzN|stS8%ZL9bY@yG-p)W5+8c#7VZ{V
z3WQyg%_J{pUXo#5bJREc7cgVN<MP>FdlJdZ?Sr>F@)whle)eqQlD37#FCKPg-$y@}
zhrbW~`SS|PB0N5hg~shf-zqmHT{hh`>V<}(Z^C3DF#MKsNAyc(Gk!?H0#=_Z`WKse
z`x^Dq9`lY5KPvxn*>7D5r|67!dg`_4P4Zb<<ZW5*2d)_%D!cnQ4BKRxu_54&^66)4
z&HtVzbIr@`rZ(nZZP{~nBISFhfw)b%)9?*_(QmN>?VB3b6-!CjoH1}K_PSrSaaVrt
zzp^d7%B*2-*Q#)=_U%8ic$L^9EPNDRd13IO*$+ZHt^J}Ox&2siFJd>j%V$c{->3h0
zownO@+g8$6xrcMzSr6Vf{0KkevVE$~z-DF1F~j&~=0T87P8zxSwu)K+FC!|Wd3kC+
zJh8FpT>ZPNT|;eqFh*-H_|l~`i+ui~xBcOlv2FZV{iRw!r9P`WW%SD__JER=g@V%J
zRDjhd-s-tOy}hxshb<aUJae)bjr9H7G5Hr-F?UsOU)6+`Q(21vt;K?7k|5w2H;gpp
z3YGb_{|2!!7@tp20!~=|*!sm6`YQgB_Ektw-k&yT_=NuNd&z8PA8`#o-@e|1olW}f
zf8?gIXn+tV7vYr1G&a!Tf*Pq!*2INy(=3|GMg;n+jPZQsu>vKM#EjPPilQ6&I`x`L
z_f_G_@Vvsmn#a`0_`ilK;{<08rkPWqIk8ImnX6XHBXC>ICgSHRD$U%lyEZG{Y#R9Y
zMNK1v!N6IZmz7}GSwykR24x<+>`wsdk5?(oN(nhweG3cn9UJ7*FC7##F*Ce*<U1F;
zQ1QzbRa+m7<p$4zGYNZqriig{r>ymf;D<xnCaEr6FN4wgpyxCi#JU@1wCW=4KFA%u
z{r+GR`1!|vB5K}BVVHZzk7coV|4uRYPBGteH`6xLlj(ON?n8@k_YP`spW?KZ*wBlJ
z+R2E%cRyR`rOC>En0s5Aj^4fUA&I#^zEoKKD7e;PN~{e3Ob}wsgqucAGiweUN=L(t
zu;FVSLU6RGjc$25B{Jc<y_uI)=HI+3d(K7f7Vz;?7LHi(jxiQts#fHLhlR!5@ZwC*
zKpn#-XW0h$=urW8cv4dYWv!JVNH3YR@I|1%(!FYdKGm*_z8ZRR%|TPE__?K27m{So
zLSD#POd^<>Ih3O=MqFuOaAQ@az)ow;mn|<u#OgnJd3^s8BT+4r9S*#&)zUle9#EC;
zVl~|<_MR#vamRY9V|jG_$oj(h0)fv(?~O0k#6Nc%+BirsD(?GWfxxNk^wjFAs;X-r
zFrmkEwmaX_!I+gZ1bqt4%POPqYs$+31zdaMH#VEClT$0+DwkdGk#`dR1LW+8!IgS|
zG&gXizSrT&lTd5iVtXXVW0Ag@xlRf@|NOr<%FFDd68|kyh!xDvaIeaArv9H-S&Cle
zbWX_HpP}gEpOt{i&g?-=8Pn|;Z;RlkqGVEuUwyoW%yr8~I&F3N=7#O{@kI>ksa`7I
zk1HZ8*)Kz$mUTTm0&G-%eJ!Tvl!Xa=*~$weJgTRAxgYDQo<FJvFY^P&FzEJbZ7_YK
zOaENHcbi|ZXWh@`5WnEXN4)T1z6wR%yyubkm04Jy&h4dGDx;4F9cnA9$OW2}g@#Q1
zwB?U`Q(wzTOzpW#v9bC2)vVZmt@q3%Aaf-%i$-MDXMy~vMkism+`sU(w#~1HQ4qDi
zjJy<oQ=%aYLG^A7&D!xJ30^MY{O^-{fArYG5%gA%{_xMTX`eD(oi@K)C-jk@N9Yjs
z`Q}x#mFx)AZH?uT`im94{X}=tlee{%v4W$%FbGDxh!-aPro77Lec4wb_L>PXEZ*|Z
zU?UuH>>R;eP3;3d#~QuZ>cLk-zOs9*TQ1FopUFpc(CL6Eyno`e?YC4DloqGQ*3SIw
zi_i7m_R*K|$LFjW-N+%D#m?c~#|`Fwwf&1M#W0B}lg$vI$h1ZMVXH}Lj1M@cqMt|4
z2=aXU|ERk1c&PLL|3R^BB(<v*QEcf#sZ3L@Hl-!esW3)|4jO8lBQr*}(w0=NLa0`k
z9F-ZlW>hvIN3L;?a!g_jCgT`pey{ha_WOJPp~tRcKJU-_{eB(K=kxV?D-okJkWX_P
z&8wNInv&*$XE!&r-cTL6;B_t}qCc~>9_!X3y3g4uI%+;$K!M!!OM&2KWCE3FOvg)U
z>77i37Z4OMVjQRUec1dIWii}TkFC5upW+BZO$}j$sh%ED6^pyIvN#K~m9T0=IUmTo
z$obqHXtChu<_W(5oF$%J^Z$&`(n+yY!X5=F%p_cIlp{IN>!dpn_v!Rq^z?ZKCo+fK
zDp*~w7^Z_WFp@DyFpS2}iF&`oK$aAwyQsbqHMsOClrix$Z5+|JtWIa@q<DE@70+SS
zQ#}K(F%|WV)UW)Dn+j?Pi`vUkh_?8hZS>qb(EK>NC%8S9RbgZnpjt%hnhJK5kIqks
zvYZcf-t@n_EGZ!6`eu@M2A_5|Ir3gEf3Z8XJMEG`3GYOU6W|w|Jc=g~8mlgZq73*_
zRSQ%g3&D<Gi<H(7|NcBM?;YygT_QbeRV*E}fLR{Oi@mh{z959{E8B(L(Mb;zs(8Y1
zk$1FWN`h;UWqw*35M0xv`gfF64Hnky&*q@~mJdz_Tka0i_O>u20NM4J>1Kx|Jz!Ri
zGDmbOC5sYXm5%%Qr&n!PTpqA=P5fq!;wFY5fbP`I^Y3AI3z_p$cFspW^KbY2ugYOu
ziZmiwzQcji8FRQ)ddzqQ;f-h4d4EoKUv<OG^DaNV6aPJ-5G}ZWoHiPuW8NT*DSVGg
z7lw>Ltenms$&vc^I<fw2VQxn+-n2^AiAaL$ha}>Z)j6J-+~QB&N*g^lhkq+=L9?oQ
z2FsGOD;mSrzhzdg!{d~F?ff+gk$C&0E3w{*tk=BoQG94=0#}tFINvNXFQ0rc6liJ)
zyRRy)^Tm6rr`XS#ZoCw&k$UN6HsWN$#h0^~yXdjH;Pg|V&c-TTh2gX11#EvCPR;Km
z;^IdZmiUlQ(~A=R^5xj1ux3NdSMMojxMD~{ZD>>~;Y!7+p5rEOK0Zv2%|NqplcI=F
zPsGkh7H>q&NNm`OSeB=e*~@<HhXIMk!KU=yM0E|$fgSio<d~X<=BUf$<m}-Fr0|D@
zWaHaAb!8W`>$2L}hoRe)4aX^pDQc?d=&~cI?iz5Kdz6Ff<9Zijfw>ukoSDn_o(K5v
z<TGWr^2P2EfdNNBG^szZu1FQ5x4oNH-)gygRp(dl*x56L<H3?aN=w%tMd-cvIwc){
zHukTPuC8L3y*Xv<SKs}L%cMrqa_Iy+Hlw@c>UOS_W_2#)CVD{Dp^O&6xha|tZ9bNV
z2rAd%(L7f357-U;OuxP8S#Ua2X}Mfm)T^8X%506Rli;-bd0FXIoVX|51z&8dI!*Q~
zo<8AfUB=SPhzrP5FK*m$d-gwlMwXj>`FVFb$KL<B+$`f^8#04RUGyMyo(96AedI6j
z%VssB6Y~>v6~Xd{8{b}jP+ru<BpqyI22;tQDd|wy{Yp=u7k|Qdd?->+ch@es6)`+0
zWSH4)FDbPc*M};af7x(brm^lPRY`}cB*oW`dmM;uTMl~mEDK6XdR?%y;<6k+^Z6L7
z)RSX|Q?){=n{0^J!MZ6W$7S3rl&u@`>fkquMU`vu*2;A&^DiGKa^LTMmy*q+?edbG
zOFd?8Rm<Q{4BeeXAnr<n^Zagch@YR|uS<*`_>a?$C!fWdQ$0C+EqjPcbgXwQaYLA`
zrILHtYoy)?7h}e2ZSsfiF79p)lR!K6(xLsmDr!dVTZHZirv4-`e~?aoV?B;_PyVT?
zY(D<G?hzAlue~jMlO6P+3YoB%_F61b8=9Vbs^^5s_~msAfmKMAg=vk26%je&SRqS9
z#H!tm3g=`TNZa3wn&3<i|Bpxr#H_zF4;wQ<o=%)r1M-)wbn}iT?xGZR3e9Jo{$aFJ
znF}^1PVKYpHmsYM*vh2+a^>G6Lw4*W>R;pjO($~sTdtB~1PMb+EOEZwd3Q!t<)fQs
zB8Wt~c1yfm8NS-&)pJe!*`(R>`A&-=0!-7w)vwggURN2-(||qxlU!N;i-VJ&7O0Je
zmO9mOo0NA-tfimfrmva}X3z#$T#TN$*#OsoI<T7tvy3Qip0H<Z+3J`>v+d}jX55dU
zyBqbVPC%~P6@{H2zU-O>zgz9n^5w}x@6FY|O;zi|DiPQDF;nnFXR9P1+!3=XGa%Q+
zaCBT+c)&R?tfx<MSNMuFJG`CI{~{QW=>vJ{+Hn>o?M8I3vOfQo5~v@ao*w7e^%l_3
z`4vz7W$V0bPr`8@U5l{jb&KtGYG6foT;>uXB7#uc%InOFl-xxWZcPl<-R3=c8%1O2
zf@1(@UeB|70*&b+N|LDu@{oL+-UgQwCyAH$2*(S)dSaN}P!08>sPxC0k<fi7InxG(
z?ls|%)#=k)J7@hif0a8smqmqV@-(A={-xc|Qh3Srk`v<Rdlxc4Zk@U->?=2aZ;Dck
z?b&QDnLj+W)^NsWX8-1f*rB#dKRFyvKnY=F*{tWvdn)P1{;Tvw(7$@|Cy8Q8-5-Dt
zoyt|ae?a5b=`B9_!;*jap>^X^$)mF}<D79yU{@}O->RTZ)YcH)qw2J8nQGA{PFacL
z-lO?slv7@)E_&3=Fn#x_NF8fMu~pop8ElVl!@(7TI-TH5IBV60<=y0Q$&2Y-*l~>a
zcj%sBt>cP%Y^qp9nhxJi3w#C`#ydzv>S7Oka&v57$x(s8wn{^o@SHnbUthofFIt1$
zjj2rQ`yU?p8bBb#;USk6E(!>%Iy*XT5+5N|WJf8v=uSk)%oTtyErZKDWudA`SgkDD
z+{`#_)flZg|7t=2+G={u(oH<Y)Htq^EXFLRx{=c@4&@JjE0P*fT=tM@eHqyiLY(j@
zh3?JOIwnU~J;#5=Z`J>?Ijc<|@S=*(g0}`3A|Za>@}CD?njF!sY`R}cmoBeYz4y3z
zUScATzRr>o)zQT-!`>=8c%55(bZ_^lpNOqUzm%t*lGyp`y)B&wR1{Wla97sx>*P5_
zb~^^F4YPY9idu5Kx!bK^CzQV7Dq)ZI8`61W4-TlWA6+G|pRwFK{(rXKr#H?!^S(q%
zhb$x|db(Z!gz9}!*{KaV!Pk#^@PL;2(d}yzG;gkhkZvrGqut6B{BUq(c({?=^yJ8&
zQ_q3uoM6@)+=LyPlFO{8+DrZ1xE?C5<-&XUbAnt!22exn+gmqb?p4Op(fPC~SCc~9
zEOcvb56E3H%=3rz)|Ml0DfENl%98PXZ9%Oeq&xKV(#!9(=(G2<XSxL50ntEsxU6hc
zeDT9V8+nXTde$y|C<8x78_Y)y!}kW(U7+yW|JX;z7ORl1$jKWP^VEi_q~oBeP*aok
zqk^gfc@H>qo6t&S@73NX6a5cKc^A791K#tnqDzJAEt|{%W}F_&k!)I(<p{a!$xGej
z+=8j6kOLhpkjDz@j&6y>>|&-!o_wF{Qx^hK8(9A`CaOoW`LF19&zhvaD^NYBf=@n~
zq*WFy$~3~NK0q($_kSLgnS|<KG{-wlaMUiph%>S3ayPalfj`%@RlwHUnY@++%{@&h
zjx%|Kj;rs9!*02zW>et92Na?rCfQA38AHBCLabFDGHxVw?{#9zBL7Q~=&AYFmFkSw
z<kp2UGTyI~FET0gO-`vd_%<Od1=iz7KtlIw5{-BD=g-OV2H1aJ-o>~^zvPcjm8lK1
zsdpSax5{qVy6j!m1r$qNs-kFwmSU%CkJHI5CAP!Wfj=V)-+p>m;vLikIXlFyeSPj4
z*vcUGK}s++H8{(M@=P##csLaAU0J?>KIDWSNT+W7!}+wKa`uMCW1Awzz=>acbL9$H
z!h7w>DFpH7`;q2<hmS3gqFUG*#_6a>2$WADko{M2URd~^@!*hI0jAoKw(Mn?Lqc3e
z`JwLboIjn6P>{}SWJl)w=gsV*sVO5D+Eqi%ES>Ax?fo2ImG+kao)g?6DA!!EN^AsS
z_V^p?S8V$zYBBVCVU$X=AamN2r*3{)xT&q7Zzs)S+V9L!h2RqVyY%gAfLBP1QgHP>
ze^3-xI$L6x$e?5?BXaA5+TLtmm0&`PeDt}(@R3cQzO4KNpbA_8hm7O?`i^oV_T+P4
zl@_Q@yRWm&O8HVU%_$F1evHm9jc81rqst3@+FAlZR^;~Izb~1?E@U5-ZN9m*!<URL
zPk?@}we>*TpR=&W7!)ssXO3P_|EE9`>`P%p$U~9jg5{KP<BJ0nny)h_w>cw@!48Qm
zY-fZ}N$WuCtTu{YgK{%9vYSbFpY9rsNBe1B?(C=U6KZnq-zK}><#X;NUMi&+2JQf)
zzxp?r@3QK$cbbhZmKNw5ikGTu3P2*af}n?0@A{U|y&Lj0oLjv!L9UF^=6!X$8hh+t
zu##+@!?T`T8QzQwgPTjhi7x>WG&L_8;F`!s-f-J+d>13aQmgp;8e+}I0V``K#$?FT
zsqhK8mxag|*xlrvwD_fK)eZ`d`D<c->skjVQGUAFyF!uV4IwJ`m&iNf^eOT4bn>Jz
z6o|pBj61|%U_f1?L;ia)%=kWpTJD`0$G_OnhmT7%5JHG8d^f}CMc3;Au%L$7v-c=s
zuB#N*c(h#)&>;rlTvO*TW-hmXs0KIH7YAz(LFKWQ_V(W{2Q>LkL?v6Tmu_AF^XWK6
zPI~!udtd9r4!Fappxx7PEwldmk9awdg$mLX(mmxmkRW72)CnnD73ga|zIqAUbkG!v
zDynMDV-|Pw8yoPc%ByECHwd%vjNv>hIUkrjAy`k~3pmqzl{d0Gpl*||fzmRz#S5vg
z`Il7++52fpC$5uwY`TNbn`>9x<lKkXYD6eE+>pxFs#pwW`(#0eUCIW6VKz@?o8pQS
zS<f{ISjS_s%Z>RwM%mdXXHVTBA5J+1{S2-KQ#6KJ*o)5ks}I#ert|mk#W(gAP*NFb
zA$|&86h>y8(m&+F+-L+xyyRYj%N5%_Rmp5o(Kk6i-0wz4Y3Z8fkb%~bP%^qL>onnq
zZ->cxuN51y>Mos+9to59XHwht?w$BWK6~jqGlja7%#up+&nf9G$LGAy4-m9(n{H>+
zI(qc+Hk!rpsoS1=&(ZNkJ|x;TUhr$gH~y4GiZ9sJEp9k~w}|CN`XX^e%BJ(t2}>8y
zd_oVe+tH+=gbgmiuVtl2923XRzS}tz=#r=Inz=%*?9Jg~gc||M1h^5`;?m~YUVGY}
z>G+FN9=1DGAquAol8%bIhi9X?qfvbT4%Z@`IM|s4XTbiPlSi@YQ%!Lc$5+?K1t%;%
zlV+~ZsbEeV%~|^Bmq$eB^B%&9v6{YyqoFLj-D2@bqhH;~vpdhZB7l3WgUuZ9K8J0n
zAt&H=AlC{ATk79T$mJPY<OqeEN(lInGf=^xb(d|uw`(MgKUFOg&<1n;prH<y5(JZ6
zu|R8hdjMElZOJCAnt)fk4M1brkUoFU%H{6UcGb{Gk0#F1o1?0C8sFXC)D%hkWw$g|
zAEuJhN2YD$+^}6#P)VJES3OCT!}@n=n^FyxTx3&)FLb68nj`P6oKqj<2+N$>YP7My
zC(63bas7f@)RdfGe{)GAB_CB+o&~XI>S^2LznaDKbN0Jh>ueVEO*E>EQ=&qzHN|ZX
z`+MENaRf_#>XR!UA02KTKN<=DNJ~RdM^Dj5;iL~sRgQItM&t2%dU)=Ynq8*n&mLOW
z<+m==hZNQMUfJP+GYU=9^>*am2IsX>_Y)FO>we-xtfM!uV9<6$u_7iiDp6oGyX;84
z18sel9|=Hv+BDgEvO@_#C^d_zQ@xq-J*5VxrWaafapPKAA2|h4H;&Hl>Yojeb}kob
zU%w5H=ikEYlbL{VS^S2;GFBTZlC;VduGKgbh}P#rqOefR<a3BbpOjP0HAaV@9b63=
zv+?K5UJq(=(!|8iPnUk8B}Px>;S^$DuzeH*%tg0CjD+rA<&*;NOeQ9Q>$z05w{pQT
z+P+Htu1R_Ku8}pFX{0Ej$vD*-hisIS{dMxX?g#+_E?bpWoVb4PZ5&o@o~w2@He#mE
z^R&f;_!SFgFh^KGf~D;*uHRp*pKK+wzT0{;Z}#L8Z6*bh{2B(DR)MS6UxtNq5`|P!
z+fT}v-PdM?w$u{nxiCGdw|z&IFv~kXtQ7~T0TW}<^qY=XJM2BC^A1H~0BL!($k{U#
ziSqpV*s9dY0v13u%fa~6#fJ^m2@4Tst&OL>Pnl3H)w+!8LS9zI+`IbT`lKf}ds5=j
zH-)kPM9u{dyV=0oYFKfj=E?jma`d1LehOTl?!XV{gK>o>mK4)#0LzGrx?GKp<U6l8
z6(<xXOHE&+8`m(0e3|bX4!x9nc=ISEdSYSoRA7sj!w+}j&<t&G+GOZ_e;GjJT&6C>
za)BAjL8X!_c2Rz^+W$JP&hXmGrE(+=>QD?3n*F#>s@T!gF<pR+%E?TvWYY%kkaa{|
z;r&+cIN+nxJIhPDMuCy`2N59Vo3yz5xU*qgdARsuvo#F03NGNyt>u0G>GLD-+V6*!
zw8hGmjgQ@ej2|u@6kCt^f8&)-ew`A;WgvNaM>kpF3))|$o5-{Ph>CU0l`s*Hq2%SU
zx#$N@Hh2p)yllC*IbE(X6>6pPruGBB!~-gEumc4?gB8P*wj#pQZ4H}IAT}bpGI^au
zHG}?g2PZt`#+H$0Kdmr5Ut-SoEMNMFQMC;s@*1YV0{eWg|3xx|5Lr63rBbiXM72ge
z4u%?kwnODv$#l-txCo_en%cAXn5a1)Fg$qdJ|DtYH}?o*xfJ|x9%0jOoeZ+p!M^PZ
z{+&>CU<&8G?9wGQ98L|8;N;+WYqhO|C6r!|Q!Hhhq+K9v{kADxDwS~VTYV}!5Al@f
zWY_%4X6D6J-?k+DlidOUr(GgIGE_iBq;3wdeq+a*^m7k1RpG-T0l>ZIYuDo7oVBdT
zhhebXz^ozjUaZ!6CvNgXSBsG$j4pbob+Ljb*16evAyE~N%2=v(juD?n+KL=KooU;z
zjd~_~cP607F0dF}(}sgy7YmkDDz@epB}I<+sXvi(lal{<*3ft6nvncjoAqb=g30<?
z=~oNFRES<irk-RnIr8(U-j_z3t%q`BdM77&7Dfc*G>Xa%;>z8vih(3Qh45>h-?xmk
z+Aq_rk?;1)_c``+WdT8yldb}mfTDA>rWl?2j#WabWcoMZhJnTD`p9uiBNZB<1K0L}
znUekQk!Rs!XDm9MCbT#cth#wSRj|`UQA>yyH={sWnfM4o?p0F^-z}AnF0QVwKr1d7
z7B3g5o>W4sXtZ8CvP6xjx`%naNg4kvDIzx#iiRP*(ArQpgWgs;>r1o#JfC1Ct{0$9
zk$Mtu3#mXVwwLTH{&?7Gk@LyreY8J&1(SEegL2lRu#OPd_nd;|WrK*|YWOXHdC}Uk
zA5Q~_THngj!KqC3!QFIWf|&348iFz0TF{&ixD8ioLDQ7Cr(o_KyA(~BW<M<#Z{i&_
z@R(IMqcN7*t=q*LL)uo{6wP2<p20G*H{+yTS&3zHY=t>wt+-qVS=KRfmFp2B*z4@G
zf$Ul~x_dJ+qK!9|_plIo7XGrJpy29D@Ao?oC20O^yaVqYZIO-1W2QusW{XMpSjEO`
z-u^y9rAumR4<E9{6suw&gdrTd90cX?Fwf>C9HLJl<Ez}KMEBcZylXO5;K;cyOiZ+i
z{E;-mS#Ck$xz#G?8r)XI5(e*vP5UxxDk6Vg<4tNp0y$%Bt-VTc@DrynqtJms{&>ZM
z2?sCB+RUwtb6@&e8U&l+R)l279<Rc$d2C>C)zlD~d)6ZPwbwB()NK3^q<kw%9x`~z
zXHW@!(Y0gMuI=x-J3*xslM!)_az7Eb(VNSnYEBc4ZB29xYpIS^35L3fYdV7}#%snw
ztsqsCQM&>(zpk40@o`WSELnqSqDT8{<)ijylqC4c(M8SvZ{Xsj@1yYn<iWwxLuiW<
z%FM42R3=KzEA2L)C^*$JwdM{{F+L+bdG9!d1$LDC6-ar{tSn&xW?6Q&2k~6N5*K{P
zb-V*U)1mxjh6yy_S$J0GpH_9eQ4H6*4mpIPq+{2~?bQS#k!YPQ*nZca@+MJpuB@hg
zz&jD$TyMkq3A1}anR;4}UWHYm=Iq9pkqIgF7qy`iD9M*vSv@^KRJP$*nIRBW_tTdz
zaqQqREto>eE?JYU3jF72P2Y+<%u-k=3gQ=mM16Wxt0RQQQ)#_rAG-XRUy_=3G81-n
zu#f^1{=Dhc>VV<IvwIbybEtFv46#Y+ZI<8}w`V()Q)->KaSy)Em&4-q#23(!^XlZ5
zW39;XJ+k25dps>P;3ONciC3NbpeO=Mxt^F+sd)i{-NGv`Rr@yspt72vH#+=w{D0#c
z-0GK0<WtIrgBr6{Jndtk$;_o!-g&vX_^aV#Jn5u#w8WP4huJ{fu%wcE`OGDRHPsn%
z{`6_f&IWqq+M3MS7-3KO7L-NH2EF8`h8ZUxBHvNC5+e1i*%X#P#4R?HQahEGr#$G-
z;R-2Fm{ttuV{_&0hK394Gdt}IlW=~)@(?AuO4ps}V~|^K;1}T84lOS;dyNP4NS!${
z$!5=c>!J1F__2k$`dPsAuK#2XM{Eq#5m~+wt>v1@2nT>REs9~6z_fZ@$4<Uz-FOBB
zMZ2Mf>z!Lfv&k=6MjDCA_G`66L~vV5eZ5hL#>Xh+9v>0BmV1?%p7AQE@bkk`v@^{t
zkLsW>?%S|qBKioT^LvHh<^^tuvd#MfFEYDWnlSY!Gb1BIR~Il<(|Z)T+jg@Z+8d;q
zSF|i&9Ftygs#D(?#}Pgx-<d%lSP1HfaQ>)VdFz#mR()6f64LFEhS;+%il;gL$<=<c
z$r*Z&85$WGeWL(pOm5h}tDhI~VH$b^ggVg*nCL6pdKb#MIeR4k^ge=)Z$IyCqD2eR
zP=lrm_miVDF7%Y9zY&ge$28TTGoXc((i@K^;%XFjGvWE!m+6|`aw)d1=cXzZ{>K$o
z{|<->MGQx;8LZ04rz?;eS8o6xQI{KAf!&iOW~zHCd3MjB*uX~7pzr`GG!7#@WLsb`
zQLhH3k2dIhymSYA){v97aBpx*DcAZt8`PVOs;2%}gktP{%Z}!fKTfnk2KN1J&2DHd
z)%-_R=~M8b@&Xlv4_+N<Qxl9_wYjV|$b|vMaB-EiNtYvm@7NPxyJ(+jaKnW#H4LWf
z{)bUEF>e{nj?O7pNkJ7>sh*VOxaLi6%GgK{=?t0Q2bk;kkzM;EVk<Inx3m`#pzcp=
zqB^(P^kLN0BjFH(7r<OvCC3*RkXhasWdAeQZ%r4JkaJxSLHB5{NU3GN>w#-~0pwqN
zuQk}txl0`VibJ{~bm+MhZwCg((zYvNo8i6_DP%?YhE2QM4<u+V9#%Xj+}@({D`vMR
zNP)}jvU*BaaXk)nI`BwezBqCtpnsrr`~(uPht#}F|9AM>m*HW71X`2ye$}_Qe6mvg
zuxx^*3W#;z17t4QJit-&o*aFO!k>6b-S>TZKyQ9Pap;$0+3O%G8|y*%1}tJbvyKwW
zQ{;5vTV>2;&(}i*i2H0L(#c72?XeFEu3?ckEUH(xt70i>#+DSv(OSGcBnmR87tpsp
zl`UH6p41d*svVc%t~o3|N)$R2T!rwJ9$Pk+;qtliLZ3_xCDv43ORRX_bbV5mKLLXV
zrJ1_2Fe(D`b?j9M&Sc=t&+}2{5B?A64Ys;$*JjH5&0N9hm(1_9wg5-Ul9Wg{>K>6P
z4j3o~IBlO{-=o;^;Jp+8{=>ynZ=EoDHeV4fylfP45H%A@e|-n;7))zHOw*v#cRakg
z4<gt#*)Nsc?1x)z6eR=BqCbOAbsX`xZ|S!nII)rtO;awlc0iseI!ZWC7M>a<a|54%
zJGv^fDdBBZo3r<eC@uRtMp*6{alx6#;np0idM)MtSsIO|tu_QQn*i11NExd-o>y?U
z4C_|jhpthjj)}n#;)k3)(Bf||O5AvfUk3H}U-gMI`Ry}T@Pw{{DneYw4OOuNbU6S3
zC^`dS;cjB%#kr?KOYyerlB;-#bn`z7WFt{HnqcgOUW#l4rlX*JqY+r!Oa{1ffyFAq
zwsWZ2^Q#3A0Hv$~wEmxB={fF9eg^Rq^BB7~c(nnMtNf^)Nm9~hi?n{h+V**Nv}R;f
zVV0w(MQSa6hE}2_rIbtkX-q!lJhN<6pBLGDv-1YUb)s(l6<BmkSN8~@q<RD1TA&=u
z)YAoIa{ayb%OYl(W}0%=qyubzf_#x@mtX1u{Iw?0cFQq;jemlXak8><Axf*HK~}}V
z5}c$;O%1qj(2(^%FFAj+<@tJqoDU|%FT9l~vw`m`F-+&U#gAYXu(~XmVvc-NzyQih
z++4sAH+zAD(Ji~<T*|7>-$YE1N)X^-C{2FYT#r8yx&!`XwbpIb8`OVAOL+n4-~!DF
zj5u3;!@JQ{K4l(p!d(pB)9eT|GuCb}evk_*_q_I3T*=GiCEwRDQHb%;D?$VgL~O>l
zk^cvx(@E|mtH7&{CHK@|{?qJ#A=iQp-QCTdiqzH3$e@kD=^Yw-k!%g!140J!d?QHF
zwuUleY*m~eApM%W{U;j~^yMw%s3tI7ZwX)GLFAhqH+5XLs<JQHufAI=J_mHDO59bj
z$fOu*@C(5wRyRI@I%3<Zyn1Cl)qpsPTqI#KzvWZwI{4j!c51+wP)L=q6NKhXA?c#l
z2!bYZK4O>W%?5BD-&mKS?OXKXwBaK_`4m^p-skn8U?@pOhM=%w?jz+QTzTJHlr;7_
zx$5a3fb*5*YcP0tN>T#i8#I;Gx}PQ{LL?l$0fm@{waJs8KPGp!r~v8=QCD~{bt)1m
zP-JdaP3CjKXTh_H1SkTzN9hYCwJe^EnGP^i_Q#tf{n*gJU7?u0khW$s_@N*r|G5fT
zdQA^=_POnPEMx!zoGO1luDsj$$hlYb;{uZHW4BY-mjj+UT_mEl9A*+TfQ#+C^WL+J
zvV?W0=VnYiu9?~YCN8u&`s^8TOw9$@4ZDC~Ry!M_aI32Ar}fT?M0940r)HeDJWco7
zC=(*Fy*SHow@{R55j=-~Mi#V7Tcnd*NzN7a!l4|;kde#`39Xnd@}8`P9>jiDY@H81
z2Zp4db*7$KGzhFGb1g=jPQaTOvX;CH8PsVtxGmE{^nO7r8?TS^a0Q)MbVm2x3WfTf
zvvMh$pWi68Rh`?t5j=uT0v`Po8ja780v$dL0rbc$g?!Fzoq0+t*fnn<mehwD5r#;W
z0=8U1(aD2C#Is2Ub&yPl6t$&PBl?BFytIjn6EBU8B}$R*)2i5G|APcV(q@`h$FT;f
z2<&vrqx+`>szrNIqxa~H&^Th1&AEY5`*fl{92~peNGP{Ka)O_7cP5aNmh&P!sx-?d
z(}USfnYdKQMN`_0193`|5NMuTptcq)m1;=x@lqA7rW9G>>?hh*Ym+qwj2m+Fwu1=9
z2vUfuMWOXyDJsoh1_$littOv^bi!@a=3@5t$z+;1WFna+35ncj#vctCFNZRNa3l8B
zjpo<J@<TvaeR&Le$Qc^ynu}K-$-O%EwKhB<A$-~7U&qWqA2`a^S(~tJ^0T!X-0>OG
z>D?N#XxwkAMJket*eG6*eheA2kVJx(xwxX@%yL;e($Po0llk>xhaKKKp|1q3#W~k=
zSTk+u^a`M_HVCd9JLoo?0T;$e61G6Jelo2;&3WbL)KjlkQ`&LjZE>5V388;hQHe}y
z`wmOwNcvqXBMX?SrJnBQ);(+8r6vQ2FfXP`tohH4h%7-r=M~@4W`R7CTAue++`wG=
z^4Ti#7K^YfpCfm_&H&|+59+8qRR4DJ=kDTOuxVg)+_+>@DnbW`CRGyQuX*?_crd9H
zGC0DuFfEwvTxk(^AT6JG&SS1HXweW)(%RN(!u+eJY6`S~-Gn>mQJ_xuDp$EpS@ZJL
zaFK7uDtW9eO8lBk=u4HUfY<AOKZkSzy0Ll8Q3^5@{stgExZ_<JuLcu9lBa^%`$1ei
zBsmel3!qJHPE9*CoyLq4h_?lkF*WsAwR+1Y&hVb5hdA+r4f{pU$~mXFa5rP|I`~8O
z<H8nmq8S1SDrUb_415p0<hJxK41Z*x(4X3QU#sff5(rlN7*itwOH1&0ixhV09a#o2
zAuJxAdharSTv0r9r)DP7&P=IT8Y=TTkEf8^Q+k9{qeNkav6~N=cWv_?ah}Gbr)rxl
zUjSkKCsOP7KGtkWi6FPeXj&)b7e&dvi#)Vi7i#Hhi2OHsAwEY*hKt=(jNuhoI(BXO
z=Yirfo}7E%Usw^d;VhZj@=6!JBO56PkoqGBxz~<dgk>H$5q0M~_#jVbma?v-c2Qa$
zlnX5qsC-eq(kAi36FQwG@)}U~vDX3NZT(fpL#AY;^Dt#!6NiLIdH``eZgZPr-pjoU
zraKAkOnjJKiTx!BXr@IuP0IT!HB)5yT=*>n?XtZO=N=;Kend?Wb!Asep&I}?Iso*Z
zs+q3}!RvhTINl{qOUaaZS0&HQCsaSgyxD7=<@n>!4`?2Ya>{j=0MB@38|IQ<upKSU
zT=(Q3ts5|G>1wdDe4-&?n|KjmC)&REeCA-{>vGFeC!)G3EzJ{4fvgpUwdWw1^x-5P
zpr7?yI2k$&#V$Flg3&}|ujowc_XdH_GPP0aVE!iUveDGK;1A61g2RgLFPu@f^T^3?
z(n9@FE7<u|&+a~OOuC<Fdaq^4PAks8O0zhleNds_X;)T5v%@C+?K8cKI3z<Y-=&s#
ziz;;G8HSPKgxYx3nrupYjnZ8J6kg2vL6BiZ=4}9#6py!=H;e)rbPBb8u^4lO<NJt^
z#IBGqQ|sD|)`~5XL5Ix;%`UAtvM{8@{KOsJKXM)TJpZZdny^$h*V2pwo}4mvLyU-S
zS=HtkqzY9NCmS`#yeA#bLAgz}O+sjyOj!(oBtVj*xLxuw2`dz0gl`8?`+!f9w-zC0
z`ZKV=q{+$mRV_~4Bc=*RLRI^M=+DpU!8EGX@q)9^Bx7QbvbQUkRB3&iJrgtk$5W~3
z`+<ZswFUAJmK0P1ZRnCAr7(zc9rBekR`_Y-#4JXt^sJ(#AV0b7sNlZ2k0oaut_n|y
zFusKo-`-$Xx%U`tkLV2%3|S3>53E3ayES4%b@Md;Dl|fC@j1F+cbVox)tM_fw5|0A
z5^#T0I%WPr4U8$6Q#DlHr`DD*7cn(;kM_Gh6y~Xz5<uGOlZXV+4ha>MYc4p?_{5Os
z?LT9tJt#uk5W;@k3CDSQoD4&;>d9k$hq2`+P}u@<YqU!+lpYfKxQx8L>ey0ui{?FB
z^eJ8s(x>LV$P}{V?+Oz~dc?VDmi@}d<>|aej^DEtU!{OwFBSLVs7daiuF?bjAtYqy
z<O%?0pk)Me{Wl2eWcl=BmiD8>EoCzAbT%Xm_2N>?^lmNn?^fwVbr-El=ViBbx7sTs
zGXY{NIZI9#&y7F3oKmRhYFLtENBrEx)YA7Sy+X5mM6C2&zhUtj-fgnkuixhd|2xiY
z2k2M;Hw~1sz+lSawdwUSjdQk9oXFl_xmIk9)BxsEZof|4)cS;=frWIhFHT;ei}9En
z*Em_1M~Gip^5^Z@vMY3VhnT-RYCM`IY7MIxjpU<f4dd!ASUCI^x1ejC!!+w9dArKH
z_+)JByX%@-<Rcybn{)w`WvJV1aTxQpr)6~WT+)YwOkvxmdWJIJT}V>iGw`>|ndODg
z5FppF79RpJv^r=}muW;N3U^|d{Ydj$qlF3oO)yIp&gdc?u30<HB`;p|pS5%9rnEwq
zw0cMiCaziGNYYC28i6=*=s47Uuu;83O78PyngQiky!{5m8jK1Tl468bjH(yMGOkEW
zT9yJ>g}E;A{!93_4@h+&;Gl2x#K+7NZ_dPbBXaN)?WnCu(zG%ldE+~`!bzX;QOx5;
z@=*qj)~#syN;7M-#?E#Sz_<jyJL{I790eb0Deem1R!LlM=e2|196S%pGRvJaxu?PJ
zzX>%W$&CzZ%<KG-OL)Xk7uc54yfV#jdlQuZZg+J1ka6BKPsEbrFL`%mUNBb<OReAu
zDOP53g-!ZPd6_7|)s_UcWT#pXddJ@~p+_5rZLWrQr8YWyl=3F{9Nb?zNRiMal=!6!
zT%p0=>QA&!(+kDYep(=_pivB5uD}&OJ3uMSzM<@&3Dkxx@8%PhBdRIhh_?b|M8a%*
z$gy^HY&l}TrxmBuv#~sfWH)@k_4DrkP!xGYQ4ibRSy#$>4JfG71Y`Lr?o=O>s+}L3
zvXpK&al&LPhpZ@@kGu2*lCWK>^1Ly=1QATC$vN99UmUC4k#2nH$z!UVCv@<_u%CaE
z=^UUjJPcJGpSu=P7ZR?K@J!=r;+Au$8cj8M_5nqa!Gdc-;ll%@+tq(9^xFWks{s79
z+LKTRi8aQx&ZU5fbIObLjuLL0Baw%UHC>^j!RLQgCH2Rupj=lK6bCag#yfDRVYXjw
z3!-=z3bZkCccS|`Uju}4uJC~NG2`hdJV2-)queE<S=``0VC&o-w+9(!%doE3+ZF3&
ziw21o;{II-1j4pfZlimaaYgi&(#V;sc!CoSIz^t{JR82|g5e;g=%~&brv<^TD@)t~
z8y_4*;dl|Ve%`G_rRa2fd939*JocRE{~kMQwyc!Hk5keMDvbJUA_x=M8bIsPo1^Ef
zZtB;2f5$Y=#hHGLG$Q{yZ}o~5;z0#0I`&n;ZQ;H&M7D%)1m#Q6Sb;xkJ_Pjx5e0%W
zt>0X?1EPfO``tU+N5QiF3>WSI+1{$u+$~TA5TrMLJj<oJG{QvV{U{3Kr_Ho#2v>gY
z;4NjF4C?FEI7#lcEk0dUM>+D*K1)!W!i}#35f9y%)EeIW6>gaUvTT}93628Lb|vWD
ze+>Q$FbM)KfS#KXH`Z|up$Y1gq%H_J?j2jiP-6Alh#r?q-Z4NdShrHF2ERKx_0=;u
zr&Ka_+c_F7N7MV3C;NMfUaFqcK+I6JE=3_JqI_MWVOY$%lX&kL!nfCtmtMYj`c^WI
zrO_CGnq{BW$xei8&3J8#w+3u5bL(82`^pYeTJLyGCto|oCFuToctD3t)9|*h<E@EG
zPpF+nAiyOYI`M;*1sGn2WX6F^cM&Sq$U;ao<DL1_)MHX^a92V^N#|KBv(Vzo&2o&I
z$i$n^BY(y$Yr6`p_*)8B2*VK>plzOM$I@ZEUqIbaEB?s@2_vx2*QYVie+pz|PoS)>
zDpP4y3;o0C+{Mwa%4A-0bpEg7w{Y*rd8>clBJL{gk|BrP8-^#2zWHrE(NO_z{~7xk
zrqHoA*BmvH)KWVOMNfP49(dmrnrhr?Wl(f5w-9n+FQY%~KhN)$fUwn1lU~VRKt)~e
zi|&A(7PYP??42-VE<JOC=s_uD`nAt;BgQx9+)MK)yjEaxsc3WwR_%Mc3hshsln*)f
z%<n#G(D}T=Jb+O)URooJX2|T-qz>+4!mBckzXJZVRw}g#r{Lx3L8=9QL>E9LP=}-V
zN4K3t6`OowZ3&Uk+7EJ!pUDwoEgp^wozq+zGu^jpp7Dk|4Iy#2;bhy7kAt=gZKwo~
zSvz;--pFoHIE4Q5`xsn^q}~!LSgsR4PP?6Bp5Tb+Bb8L-#yPfT_g*578WZZ|ZZSLl
zZW?DkXoD;bp0l0y^7LIj5bz@_bI>;T&I{Hdv(gVt>&4;`F^46s_SRb0NCvuFPZzS&
zQ9}f`f@n^1(E_*3*T6DUz3G-k*iGTZO`4-8vdFv699!%r$(=lBi60l<+S1F2M`vRL
zmKn8W6JERDDXS8JA;nXHk9*}1s#G-4J&CFyO5gD8V){ZD-iv2qUxdwJ9dq{mcZ{;^
zU^B`@$FfD6HVUZhpoY3&#4h;GgQTMaPF&LzGaKT9C9m;BEXDlxi#p~wy@D0M_{uIH
zXL3ZzE0+2s7(b+YqxvVTF}~ng6}lh_`Yf=ny;MC|d6|B?C@J*)`n`tqo(O-h-^%FS
zvbyzf8IY5uEWU0LH|}#Ft-$r=9LsZEuV2e5!^G+|?WoRFE&UDDEhw^SWF@M;8xMV;
zv%Xy6?HmsUEVaPjtTYIji*KVP8-Z`<qedV6APDwRy#DbW+##~-2gI=^^Un%El3VJ(
z($wp^Q>X6w6!a7rVGdU+%qJL%BJ<<_yI%2$`el^+PW4HeFJ$2-y|7eMQlZ$7UvdU^
zp{CkdROoP4r(m^PF49~6y2MRGv?PS)5yMf~qE+om0K0#7w}ITB@N7kWB*@h!E%$Z@
z+hvr<VRe<E9^aoCVvD2jvF&VjOI%Q2aYZ{M;oma&C;gsIUs`vF_Lj3=p0*Y(_4+$w
zq*S5#elOMFvpP2tvn1usOGy-ZpmVLNlhE*RsgEs`$D7>VdYxhyybc*O%g(Ngp@<0!
zG<Cy=SV#+0lqy86^zOusr4~70K2@cQKHr>_WSX5cwa97NB&Xck;xeWI5O&Q5{@!bj
zWS66p$ZqlXI=hi8Img_dQ=G6AgLv~f-!~9F9%W(TDikOGP@P_nE$<#~81WXM%HWPn
zBxiik)bH$MBg#$Rkd^v)#%_hv*0tv=^K2|8H*5T&1fjWB74m?+PNF5kUC+H~wPbta
zl7GG(mSsq0g6&(SKbQPwt4Uy$Ww-Ld$`o_Ca=+3D8~oXuxG`uMHGhRhc4FuZM6jrh
zZYjrqIP&poeE0UhCRrK#6(98Q)>JP7XpSJ(YE*XcX#N9r1sNpcy`60xh+gj*a2nTN
zI6LdVMX}5J?p)3x=OPnc#6R%m{@b6htRnz7JVLUu*u+bqJJr|l?(v874&-xORYyXU
zFpoI=!rSNS{PDDM)_5wfvj$k{ZLMq_Mk$A}BJ)IT@Izl@HuGa~*^zCHUW)xaM-4GA
z$)<|oyVz@GmtDEnZ_PeCKGvAI_0Msti<oDx*r?F;+TQg9E|WuIXrv@&Zry7#fEL9b
z^(<TnB({t}eE5zM?u0uJZ2kk$i0^+BDSKDcg_4?<UD>a_FUrW=aYuWJO|J)b@^9qX
z$IgU&pq8@SCT<KINIGXob!`zV*KKz*w(Anl!%kjFTu^4djL@5$OgvFp?#|77&u_|^
zlp?`|4co-%hL9jr()J|ZL@tP)Y0B6gtk~ZD$4+x|QQ!KlEFJvE;X0Mw<5M^0{|x4O
z)Gkt9sqq1GXL!$FE{`wUrQ{krm}TDA*jV@L_(;#owF{2Glhb>i7_7nLI(qalOJ^ia
z`93<9qKzebk(o64)OW*ljw!>O?bbmOzf>yizZ5(5p)p<N2L(c>Cn4S}0mhi?P;OXY
z@{a$Q+Aarixo|6<xZ9SS%PA8_fB^~|+jOk`9jhzYPQqq%GrC6Y8c3|6YqIH_D4Ut6
z{EoVMIZpEhtW!r<#Fw7T+MQL@kYVxfVMlz9xC=Qrc5=tvW(6yu_#AisZA+?2*g0}1
z`LW7VP+hv>4HF@y3g>uhp%r85^HzB@y^QahI~_&+=RdKIJ|y41|D{qF<Nvqy(D|w0
z5zWT`@!bI77%%XJs^f}@WwAE^WUf3rdzYfe-ie?;H@hd_<dmfplBUi<_eyu2oQ$5I
zy9)@kHt|4mvd|BK$-B8z4wV7joDU<da>y=Vn`43Y8+t}>RsQMy%%qBr?#Yn~3+oqM
zpdl;j4wl^Q=>9!0mq7lMF1lO=ja?$v$SIi;6qob-24syMHiMumH<=h<7~TYdB8GK$
zU3`NyH-z%NT|qMQ-3x6)l8!j{EoLk_{bQc!uHi5vR*>$Rd2RxbVW0vmQ2{Jv?GL!X
zNwi?queTD0gS|vsj<TO8e_^pzXMirfCbHlVhzvo{{cnF-b1Y{jw%Fb6j3`pj{L-|9
zBGa9XP6<I|bTUrr-II81a8M>a;(WrBCF{h$9jDns%NE#5CvO!D(Y)pk>{W=JX^3mF
zRfxlpXwuV$zj!IG^EQ8YHCU<-e3Y+2<_&I$iA2sQhLfs12ADhk^5<P|#99x#<qsN$
zn*{_Tk5u5I{WRS@>{z9MXL^2|2GM8B!kn(4ACh3G{`?o>t9F{GJ}NmT<AE$rt>{Rj
z7E7~a3ku0x^6b)-1!c?Yi9pUetfKzgbb&-zkY59xgFh$rN8Y3g*IB;&TF?OZcqZh^
z-L}3NT9p4KBu;9tIG1BG@IG%HmX}7Ap9zTsmThIYy1R2Pj&@qn(A#|3t8%|4WmlN0
z5WvSWL1x9}=n6rO>J*hlNS{wiARn36!5H2|p7vtDx5k{^aV>wiuQH<kJgOyj3-Fk|
za~Iy*%@w7{KZZ7P8nT#MW)`Po?(BZOSYD8Asd;?r`E$81n%KTXYERxm2Zg#-;$J!Q
z-kTsyBVd%9>iMzvxpd4{eAFUGbrZNR6NTs(A;AWnUrFWTg<49L-~oyUCO*2)u~Ecv
zmI+Vh#q)GgM3K^Iq7X;)@Y&+A?j*(M@W@uV3NS79_Jot~XT_ETe-`t%&UcKYhIsXf
zbFwG#fnWL>_sPuCc|4up;__43%*z?1@Y13OS|{8WBj|fDhyf1yZHeg`MXY*1!5N26
z=k(}sM*(HKv_GS&A-!_ZIaJGvY9i(7HHfJ6RZ$zAnt$q|a#p(|aQD=mp#zWkTB#jf
zn+i=fQeFrO`V2rTeY@la^^>Q{a9gaFR0TQ*7EcwlFcr=V&GmRUJo4?!XqUz(j!FJ-
zltqDVs{k3<TBQ@Ur7%HrpSF0UV&?kL_Z6bMqbD5a+u#q6-JO+OB2N<v%r^Wqmb#_V
zuIc6Ip7BjhV@mzafN&~mbX$cuc5xCowIF~G8Jj@!u#f9gWx&O7kV*VSIfSLq=9qr>
z*p|ZSx^Py|uybmf9G&k3q{nD7b)<{h>mdtsxyR0s_v`0jjTe=7tzLBUxklC>>$=2i
zp@TJWyeIEbXO?^0x_Z9=f$~VeeikuvQD*l`d*Uxx3ZLAH3Lr*aw8SKGOT~23=!dGa
z$9pMnd1Z;I!Bpm=)os2A$-_?mm6$SC{q}C+Amwr5+|tYR5*k%Fnk*L7C<DNmAO9|B
zHI&73xdjv<z$;5~4&d)D%bMSq14~(zoI&!f=xan*X1(F);bfGZPOPwmLL^)%5L8jX
ze>9UeZ^Vy!AB5WE@1x@`n4W8qf(f%isZwfMZxSG_A?wCGk_8s4fiFl!*jn#G-Mj@m
z_FN$)uM4``aXQ6>F+<f`^ai1!Uj%t1LS(*(hAwBcY8}<65{OL&1x=ZgTOkfBeA_v+
z@l)E!qtCAFN!3#GiD`2rSC)Avu9Te|38fT7VABB0vkXYJm$?>f)5ooH=kwbi-Bh6j
zEb3@sTqshdkQ}=9F8DcTgc>#9mr=%9qJKorrc~wK_&8ODN<5%8D+e_<yhfVk><>Lu
z-^xxLq4M0c!f_^+Qc5m3apym9nSKf2>Q7VRn*F{UXm6xF$v|+d4OIJ`*}0+#w|fm?
zUAOq=mbHvT1*JyDLs5>`-xZSTO#Xrj8mVup4EvbTsKsMtbAwqDy!J%3PiQ$LvQ0)7
zB|{q5Tl~g@j!D_d%9WhH)M4&pgN=A%q!un<sbCvI1h9Sw!{ufT*B#%qdz~tKTgVCG
zTj>@tNuU(sksM7}UWSv2P&O4(+lLMwVV|cPIb+>>REfm<zF_FzqFMsHw%~D4iTFBx
zD+7`%Vr^Kucv%(i`1PMFVngZqa|%8vsXy9Al5<OVl`kvgT9)M1vW25DgCi1zk!B`%
zOE%k2jY}Ph$YmVjbv!EJzEssGX1<+GLVH?mlwvR*WR_FqS}8cDBdHNyH=-`eyQw$$
zKo+QBgd6j{8FsfwWhx=+n?fIy&5|HADttikXLO~Mc`&?ix*@7!?cTmi_zbgbA$gkV
zbGm={y!c)8--~FnGgsecUfU9B>sU+pA%4m72F1+0-I8E$(NX~AAVrxK>65Xr16Qw}
zCAwR6Hgv?VsX@HIWz;*JNNl*VrCg&ZE)s7D;IfgpD?u=J9C>#&bUl=skq*(FW<zSe
zNfhproia@H`+}csy<5*cKE037SFBTQmOD3k_Eth$R^n`JhN2ZbVv#K>SVdJspj3hy
zICW6oHa5uV$YKQ9Qdk<Gz(jSIt)pBu4I%6TC?foOpsc7175AKJQ}}$<l;^x@aI4gc
zbvZ>h#l&vbaq3Rj0E@5qK4FewOAy=mr_6MZz7q;qKk-@=x&uSUoQHE~-vU*YOWeb<
zNtB-$PC-u|A9lObbhy{}3{bYK%h_iI4C6;{{#xmf(0axPJk`h`h0`^%T;aY)6Na>Y
zRlCo*k-c8}VnBoZH(Uo0EL$m%4{B!{5$pmGC;Dm>`d}2X?52<TKo$Q>^1(O)o5<3k
zHy_g}eSqq6Ig{oUH(&W=#>fSZH<~cM`jPtjRIv@EtTAXR;F=e^uxs)YQiB)Qif1)6
zvyDnT;pKyPk08Uov8<yq%bfdcF-{^^w-WaRWRR}geFHk{j??!UXnuqT>}Y}4ov7)A
z-U8HpVh-z+0)A{1Ec~VFP8KTr@;mbI7R0bY_h)aiJaH_`x|el8<0}qUDU~=s$*77T
zC}W;qEa#v8b>wTVwm6vYA+2&EiGzYsCNSaIe8Fj=Wn+j6R&k^Ai0}Z&q#|SjTCVrt
z)?wB^4K6XvYY9<Hx5D$#7u`mJiW2?m_{!oH)h5V#8@d%7@X9K+e9SZohaail#ZX$t
zQ!Gz>2#@dQp{e@aRa(wZhb)`FOYd*!qB89>5%J>(J?#U}1TDz+g;FRLOpNHYR8PPU
zQTslELPZSqGZvr0q`kFQ*{+`AatSm(Cpri%bt0divGvxKL_daw#0UPeUon+V?kNi1
zkuD5F2*uzt5`s%XX?4it2<`Oir@iDqyjy11V|G4B-n3ir$6IFNj{~`^oYLr|5GAza
z%Vf86z{@wJet**``0w={AXAr-h$jMGZ4dz0Q%itT@OAsB54{r7Cgn9NQeld^Gb#u9
ziG7M5yPX{x_A1=Cqn}>RJfGo{*vL%z*gVlzC(4z_2-l~EBBGM@d3xwNJYAN#VH!2*
z3?%k8oUGZgbI?aKS1Uw$I=|eRdn2t+?w;l7PUr_xJoU$JM;sGrEv&3$J%$T8NFUSu
zSX9(??EXru$Sc|_>$~fQeE!gPK<YEnk|UzOkzvY9pr+ZB3-bp_i2}uV!^}*^j6`$X
z!2sti1^p#VZsvolnmMVYE!EI44|cNlhc?!jQH<Q1Ci@Vd-~*m#$k!5D8my6&8Vo{g
zSMhE|ea_0Tt#()C2{E%ZWLmSfDN{%jrb}O7E-8P(l}k>7y~3gqndlmvPeikS$A8aB
ze>h#fn7{oKQ3AQzRj5nSY^#!ULUgO{K$TqF%~0C;-&ziH1V<gkWpX2rmaf5@;jWuo
z@Tl6H!!4X{W#;{>&E^0%*@9ru=o8C+QB~B`abuVPbT3YApuB<7@Y%MP;E^$BvdleR
z<h;9_udF=tbKay&0YFc*u>w!>)`_0;$Y1tvb(W9vX|JFBA4Msq77!7s`lNM<YV}T;
zWt+Kdfvk0jxpXC#_T`!j7Mx%zXo>ES-49-~A<Im|_r)p#Utg*2KGUr>j+QZDngMZW
z%g=e0=Ee-A<(y{egN8iNuA(Ze*TEc5WNwP7w4LcwQV$)R_q3fDw8_1j!i<=iFkLT;
z*?|~arY8GB=xB_sn3>!fh@5y!U2IWTD?JEyw=1&ChE0B9KZIs<4!ZRO^`yf12ml!B
zhNcf&6!jIVy_zmP$u=$7EThs6pED7I5R0#@Tvk#s${m%fR%eXhXuZ75j;_2ezW`?s
zReBmj)svJAAhr8#%_FUoo6wt3Qv;Y%eW>0%;_qY9#K}z_qcS-I5?6=mdB(`4m;}iM
zL)*v%xI7BVWPR-nqmP}bN2PsrIVc~re31?7iHrXk>IMSiNji0<VLA;zs<2^u^8us(
zG*vYqifotc>h<^ytZx|m0Ep%Mmnsaz&wCjwXPj*-3TQbH^4^r4#EZb<S6V#u7%wRE
zxjUXaRl~Ml#GImazckLQHJUR2<W4rt)WKuxxiZ|aSk}s5YV0!aF(dhK*^p09S=|@A
zUYXtUh}m+Rae8@U^z7FrlX3++X*4gz6#;gJW3sd7W0DOI?-0kFvFyYeqSt!<mtZ-p
z$VUVB-Eu(YFz?3!l;yGm=cDH+LKyAl_&qz2TSCcLz_Lqno>x6<Jwfqs{U=6fnmLq+
zq|w`&S%STMs-tJ74wm;jZv*Ob;cEb7PL7k$Ss8$W(NC0Nk;=O=_Ir)h(ie944GXjT
zLU$y{IWma)@Dz=a@NC{*tNC9h;Vtj&+BhE_05nog{qWNcZ1A21Q~d)&Qzuf>4#SnN
z?VtVUR6A+ha@llCuS$@Fy1LQqp!US0FK4KEaT|!ouT&@BS0AU5HV;l9`s6F8VH<<m
z4unakx9~{Yfpgpt+so)<s-oxr!7y>EywC%vfB|8)WkQ!?(sYG<{z_b)EIh6|*9IHd
z4c82O)U-?zeD(Tm{7zf+7tly4{0^$oC#FNZ7d88@gMH85o22ArlhOu4UhCG;y))=0
zhfvFa4W`=^7qLw?;(Kd%$B_yJE2<PY{)R6_k|*FAnl5H4V0p`h+pgZvUm?)+TvT$%
z1|6Wldnn|8sk$ozT_Qw*Rzl6Iab$P5Jzm)ZPh57KvYm?*^qIBHqOL@5M(B^T1!Q>}
z(lxpQZV6S7V17zU0;T|=1z{0DSOJ6owv9gl)9ZmXGW+unZVJ~E2f-Wt_DYS5vejGW
zRMuzSH_Q)WBfn@9LaMq!E%S7=b=!;$`VDW-TfsX8PdOdx1m64BVG2y-?qU({Hz=tg
z-}f_I>_>g$r+C<X&4KjU$YvG1@1Y*%hn7;3EkKxr#I`5HTx6}i;G5924cXgS>3(u2
zUjhyz9?nlqTMT^Y2TekyoRfZ7hw;YTY}2Z~$ypfb_l<|nknNtcN0{zs$j<VYJ$F01
zZNSbLm3%-<Y@WI(bA^yu<^Cs<+H|PphNg(<*7M^q_io-9I4gb%B2<!QRg5Abncg%y
z18`G%+bMvEwV#@Bf5x&1E#{4`;+H@6CPwG_qRx4|fKWr3JNO;qOMYE6%K_f!f{!jn
z7=Iahs0jaGQ(SB@7iak-p()d+33xPFS6F^z?J1^KWP`uOsc%okSbdAOW7ol+w-#7g
z89nt|M85oCLFIA!0mW756z718Y8H%=w|<tbKEP2w#v@rWA?vv^f&w5TwsYRkdq8=z
zD0ryekK_Q-Mds2l6G)%#`5J&W=odMVseFDCTU68US-_XP&7TNpN+-+|c5lq?rFx*A
zC9-wqz$)z$mGlL%#Tlo{Ob7N(ukaY%C97N|m{EH9PfZCH{}?Fn>Ea0;bHS5>cBJkw
z3kJ=VC81gUR^n)%EZF>?;tpW;;t_(ok=>^>B7qaVM7FSP6>?4|z?<4|fRaqyIK3(8
zgvaUJPvDdJ!(2g-MP+l>=o@C(_;dRTD4~s!EX@C1oW*I^{|U=m&8zLY=_^x9WArkH
zdnh#9*rms0)0K<-yLn%cCH*U+nX;&op8nwUKX8B{M>SYO5z^XBSHYt~1eG`4tZn+O
z7PP|$P{-Tyg*G`nctn``$v9;QN?xZzgN*zy3e`c3nfiiFjT^(-J!+$eYFX|&v4oi(
zvxg1zhyhrNt489q`s45@xxl4-V`Ysl{NbXD$~!X{H^^SUe-0X4W=SG<+Iuxt&U}zj
z-bZ#mp94S9yiD~7qnl;t29*t)>++yJ;QJ`LGyT#$zfQ?s8W;mmoCi=dWcC`cFPAeS
z%1Q$T!i;>Z`Z#5ZAnQ$HT9P48j_M1!<3GZxSI0Q&uHM+m$bO&9NLSo3i+2xXR0K8{
z!(1H><l8CKpR91wG^fZ{e}U^XF$=FM<!Yie$<~crCLRQsVY(pl`(yF-PR{E%1v-oV
zazXnY=R7m8DyL*7+G9@jLXBGmV2o2eFMp4hvHtj?F7F3b)^~=*_aPYd9}>X&6_`+n
z9ezsmiSDt}s2vFB^*}fEc&&Jg(-CC(#%#P=JtzLEywvr_5M=(b88e~5#s%3a`AIWx
zc+;I%F0wj}Whiwm4O2m+Nvdbl1Ld9VoKSGntlJ}A-=8;yFY_x-3_TsU{ArSRK&w^^
zaOOoj=qWJy0Sf5k>6|k2U3Rmq%7wpI#$eSmfeG55cahiVC@n`>VtCuk!ROBTUpO_{
zi@D&NRukTsA4~QJSPE6Z-OZ#r9;1`Lot8~T3oJc}*UK+kW6diT#2Q;3x?X#yfCd_H
zR}ON38B9R#3ckoHJ!V4VkiIgxW8a?@>h^ERj`O(%p99Z<0(p;2<`Dd=+sHOB*MaF^
zYA?T3bui@j_xC@A?VDv}r~14Z-S;h?l+MuSoZ?WOJp`Jr3H^QXqbiu?6x6My=lNNR
zuAE*eONlLc^{m0$np3O$6@1^I+Hj~7fT*W&@Q(2qv}J%sMlEvz8Lr+(q2DEevF><F
z-SFe-phCOuJI=knN><p*-SCwreNj`NHN6Fn&v7VJw)C{2Z&zG9u=%<gX7f%Gj<lRm
zGxT@APb>Vkcrs*kO8RV3zJ#XM<-ENeS2vtH6AM0gt@dl1XAWp2{(h-rfs07p4j-HH
zBA$6xkZZiLc7~p1jvdN3sPG<j@>b&NWR+U2-lv~`0FbPMV!WIp>%#ZkPUBYaw~+I{
z2hTeJhT3~u2o{X&2s-AHxjDNe>nmXy*yD#G3kgg&wP^jN$znE^=LUK=Th5>s0u(+;
z)6<iWlLUPk{0Z^)>jGMLUn^6x6$P7~H8N4<;ehyfwoJf-eSWIE=jr}ks~eapszqL1
zig#sZJ)FiCygc3Noh7SY4nGK5qM8k`6V@Nn4)pqRF(;1ZefcG9Wb99?E&I@&x<7>i
z!9nQkKzWpGn6{ynTEFs-v^(DsF>HOwv|s1lZU?>MhSR7?3eNlLeBkdwX?C+js2JfV
z>T~RKEO2DY32*+7tLu(yGF#d&9ajOd3kax-f{LJ0l@b<f5J5z-Q4|YBdKW?hTt!8}
zl_)9*SZEfCQiRZ43j!iVH1yCT9VwxNkZ;Z#)ZOp=aoOMQRnB|L%rnnCGslCPcR*UM
zR=LEFc8hu0acCo)5yMRUuJU14>phHdFt6Ny;Lwmh&c<U8BBfdPx1Soli_jII5NKcJ
z{@6lTap+=)sq$7;Z2@ZIG7Q#!*vwD<$qVGzb>^yQpyZnLbMBb$$vNVyI`F4zgs(~l
z;5n<yXXvwbb8Hk`#9wsM@NvKJ*Ggjt1>=ogO1QHo?D7_vNyDct;uZnoALwJRXLo4r
z$(M`fT*bh8_@Y<uz<RJyz`b+yLpjbHQ9dK5Sfif*0dL97N9isXGej9H$y7OekGB=L
zq`(^N-1lqw68O`Ib}lv}$YZ?YkWdi$b5POC;X`SK4RjUxCEwE#9E%wGtWKIjC%y-X
z+FI36p$=mPr9<+~W92@_U#~%A>Y7Jj>xjURux~+U55=W-W5z`6LuDog8lb?1I(wY<
zr+#I_w<Pdm`x(>Tjt#F-X&c<TmAUCh8ZQT-8!g$Txezl=LI6TE=wLMLJX8hYGG=k9
zC<Iw-qBH~|8@=VIA_>i0co`Ep=6}i7(j{=o6qHg@+yUEn(u+1(3dupYpm=A$*w^+3
z?)4&Y*#Bg&`nHBG7xj(0-S=Jq*PG<5#lB8wc^skTSAJxRgJaXSLT9NaE1d*N3tY~E
zdG7wAF{n$QTixj{;YX-AO_Wa%SF{WYP|v@rf9`1vAJj;l&~*~q|C!Q>bNi@>cWzLa
zC{3>knTbc7Uvx%415Y=3FSi>7(Tc<5O`z6hzxx1pR}77hpCk(2J3CASoO2tRE`U1q
z;+dj|#B#k=IQ4f&wBvGiodsq+Kq_&JoLGre1qfzluk|n;8*+@wmvbUA_VKJ>9|G+=
z)y_k|ZTBFHbOo5F^cyJ4vzET*@V9i_R@3K7a6BLZxRJRF`sjh=JNu*99h^ykw@9%`
zV<%}?@wqX8_dfM57FeYQbgVV>yWKE(b1qR$+E<8Jq{`a{E9%J^N_fXnfLZKpNJF76
zwuyr4k{b9MW6pEV)iXWm26fuv%i{B`;ijx?1OKS;@kfr{I3P{-p}C2}!9La_W%YKq
zKE`T8&qKDY^>JVO8A8TtoO?EolCBMph=_oea4<k4WXz8e7C(nK^1<SQ3jFD3@i>rU
z$+$d$e4zwt<*1o=N5Fg|d$UH@NjKKPCNEQ2Ozq&I;FQ>>_M{#6k5t;;-#Y8?lqBRa
zmkm&QCFdvi$a*7I>w>`7CuWxMp+OU61GF?R4zvGCxDkj^RZ$%RPqdwCsHFMmY-=65
z1Ly=pH}(AQI1cxh_=X7eJ+(en@b7<)_#WiD@t$Cr?|J@a+<tN-D&bk1Aun14Q&kJG
z&I_PqRz`WF=P2D6CX#Th)5^2P4O=YI2M%@l277;WYfEL&`(Jrnyy$TIF(rFq2R{G@
z>}<p8A(PR-tR+S_;a3v+I{!KPBIvTlG+&kAN}b!#Xb2i_<Rz}dIYwOTlk8#(u&zo^
zTdsi>BE9l%Q>K8$&&k41r)qEsBD+o`@K>#yvvAW}n?TPirT3>mB%+}oe_dY-<d{8a
zc?npFzgkyg&8TGwItu-a>dZcGCyvVa;~DQM&h3tOmb|Vcg%A7pur5_%XW=sX#X;3}
zjElu|;9;J@*q$qYZ~ob=jA{wWy3)RC9)W=O7`T7`48AY|R3jA#wxJBgpwvk}`R|}o
z_I3<(n>#vn^nT8L*b@&7QMdacoO^~Fm%al&W|iC?xG5*|p<{r9@2B71GIhcK%1dfq
zTHn??jCme&;)q7e{ts?36d9n#awe~a3b)ipl(1jHBi^@2e`A!3U-c^24|QqYj@`2i
ziD9tJ-NFj!6hVcS!i6z`c;UHT0vFnt2f__;)vsPR#KqVJqn7hBsQYT|4!LP_beMg`
zM#p|lJOK8{cNoh18-~CftL_xWlr?~{Et>6W4*Yz=;KvPhFr_Yf(BuQB8v#V6CJO!C
z<!3`mfwn0KZX~P3djWnW(ubWK<{jv^AK^4v$la^Zd)RNm+Es=*f?y1}r{N3@ZIgSX
zaa!sa=@}ANVDakie9ymqD1o+`kL57+kC&eEc!XS}>*O_knCsvV_n|i%P78~DJw|2e
z-G3>*Om`!MCmH{G8`=OQq%3AZ<B_g`^FE~Ic*GvTyO>k~^Yw5s74B|fukW6+K`meO
zkB0Tj)sN|YgvLMCd6$#y%3eQT#+KDQM`(fzmfy{=pRP1I)YTUY5<vpV;lV_?+r{t>
z;B02vMpHL<aUJ$8g0D!n624rUH~bYgPjdXgBwv+NZn1}-pWn*1i`h!cKWwerq}ks&
z2^BJ}Tl|KR{{boBv){7Tbh2O7uME2)m}b1NJ<|dOqrRk|W%nU5slts*cm{ub6)Mly
z#PgLVw@G0T(ks0jGN%jd!M<l74~eW7=P1OjXS<z`yGJT@8#7{0)Gq35Hm{G!8sJ0%
zK|3=0@EiMzj7T^l8c6yanJDiHr2d>F^8m}&@~er*mtPUu54UgS1<1B|Ev;_86em&Y
zW@N!%2?#5yT<k0^{Fa1|waiPjEDPE6Q^*oeYS4=XG=mR{B|0veP+pLUN80E^kP(<D
z@}lo)g9a~f7a&G0sD2<5zumZxc9)Ydc1SS3{AHYt6LZBj)#p79J?D=#vKdS2i^kws
zy2f#ba5YC-LD-01;zkY?KF48T@!gK_{-HVm#So*9jrv)1_NBw~Mn;h^3BoupBp+Us
zI~*G%k5#RUp0A&E;d*1#NyT!k3+G&bpQu`(&A2B$6YcKmX8|e`EL}qU^H0ZAJc>%T
z-&jMrGW(4DIKq%u&SA%G;G}hQ(Bw-HYak`-F_-UfovMb;F2a<gG7rg>F-=S0rpZJ7
z2JrvGGWF|U-XXLf@AV8HKf6`ckHN{^Hhfa9YU|6m8XQY8p%cf~*tD10MfW?Zd0fVL
zbtTqC;=u*;jRHFzEbA5t@)MSK{!@tpj-8M^=y_|+Jh1uh*_-apPhie@YbDGe0pk5=
z;P2w!5fMf?hnGUUvFvCJQGpDHEPqN+wl-9KbH<07#)I;_%#wcqPxQSaI&C`8v;9wL
z!VWcM)m#>3309Zb5FANcCb)YQ^1EwTj3l*^dbarWjsq-9T|H<el$=wkgLxhdXE;Q5
z6au;noa~DxjnTFaDIwJc)_V1RcF<4?5;3){#XR?IwzEVTv8GjbH(MTx?ELUxfoZB&
zKG-Oo&6>UEP2DX%aR1|Mdr=va3n=V&`Ja(jCk*hwJMD4<Dj^p$@q*^WEVStSu;{JV
z&^xv@Svuy9s()Y?4za|V7Bo5J!3X%metC*juBJ?!-J0x-8Yj<i5w?*W=jh~I0<Vy4
z68^=JR<t)_ob*ZWs?$)EB5(GRxo&CSF71@&V6A!xu5Z%{W2oW9V$!}2tGMui6YjQw
z;iJ*srhXRik*be4(iqTk>qO4{L=3H6f*j`(yuncqL?oJdZ}^wefT+7VdIc4wsgRRA
z)pfo?h+a4~^K3T@9wmSSba7L$Eo1A#T}6u66|jVe{oKPMKUco;I{TvE1DW&Z;_j15
zJwC3CpQ+33B+=<h`Kf&IdRd<rcXOI-J8sNOza+P#!`jo3qXdtPJ{u}{fv_>b3*QZQ
zgD?NAgYI0HVoy$@CO)A>pC6!PE3b*b#S%lf6lZg{=$DL7<qwRsnzsz`VT5^>E6K1=
z_$b~iI^7%&z%p?3EeHyJ5kI@z5%JTOg3gNGGzT>}FvoKV6Qt1lr_s6ta5Fx5!gT~s
z89kgIbG{i*tSsWX6q0cY!a5|$&N2Np+LJvmrZ#F9`Z~l-fQ}foEwZ!b&XAfJ4HpGr
z7UpX;hf&A^h2M{Ag8@JNsaXp}6HqlFa=~r=@>-+@hdMoz7aeOMN(t7v*_8FezojzE
zD+JYf%ngDC-f}4g>!KsW8WETS*%Xx=Bq5F?EsdE0bxd0d;;AcIQ%`Z;z{l~mro>9s
zWvHwSSmP>*4GQQw_Q{#Cz4GrAQu(cN)|SMrG|h2YDHBD~kKRDu?7*FxUC394nMNG<
zI2tRxBEQ}`;|`lz()rqvenA^qnWxp6VgV+%=f9W|#dK1^_CpUUs-xv(L)HQtY4nuF
z#S);*t9Ikd*xcx=g4YAjO+Wqt1WJ>S<?f_D(woYI+h+xhmZT`mf1NLDF635xQ_)^^
zOpBS`=$EU_l;6#X!&+NBLb8bEdh+#*j3ZfNDm_H<J<>fOw63@hk;5xrMVLAe308qc
zK-~eR<a*4mMrUlwp~AM*uP#+Yai~wvf`^}cb5)@=Pk_l;aN*El@93iZ@e22j&!`PS
zlgsxMuNl}b^&NPViMplecGq2Q5HFtl1dI>`FWI-fVfx&dkPyW50e98c?<s<Y!;~lp
z1%Q|Jrx&%Nf|X%&zXAY0rlI|E)HfH54?4Ea!}`sf-jw2@%jM10TsPWFvT@IBOFeU&
zKtS|5>m)p2%6EbyOfaq&xj$;y`7%DdL-L~h_Uv(DzPJfq+Xa0>6P;T9H^4==VUOfp
zB#}s*VeKOO-(`C(KOi|!E3a9Fr#J8yRH&);amZnqh9t@YXyOfu6qIiK*aLpyS~m*%
zA_9GvpIjN781x4ZlUNI_=N>^jf`ZbBK-BYF=P*me;rOsc4`n-McS|Sf=m}Nl7jMo+
zzbu;9Ozo&KV~A)Us4Q+tETc0aSb=O2jT%>i<V6y30$A5sS64im>(bt;I+-=$uPy#e
zo#03Bm?Nuf&lc}>lIFh!VFZ+*y}SM9WGaoZN%JV4mgMZ-C=Ghi)^YcittIujZ7(*W
z@A)-6zu;)Ktyym*<V($0HIu^405PYl*TXgNEySr+sDQ`a2sQOG6Qv_VR0h-NnA(Y0
zi_fi{i72NBJ(&q4(yo^(d!2&LW|hxI^zZH=v<8r0xm-MS*5+al-I1;#@M8R!5T!dg
z%X%jJgCBf>lTJ3=LE6j&50>2c`tyACGf+clUU^hq{KV?qVKuy9F9=FFiobLTPsv#L
zIK7>LdGFyhESvk7?@tj4skQXu<Z1+0<T&GO`n-5}x)*V+s4u}~CQkq4g$tNJPT?$b
z6w~0NSUv?!Q^=vlM|Wm#S@s#ea#<-F%h_X;q9DJ6Eto|LD1t6SjZAec^I%Iv?YAV?
z6K_hDN^G+P6`7n*a2JQ#89Dz$p_tl%&6@@s{yIZ%C<JEE_>T0wt%O`BX;!wf!~swB
zVd$e0yH@jM*JSP;fd!2(bAO~}-$1g9pW_0ZcPD(7A9WiB60?kVt`yT3C8O53`D8M8
zuO%;*8%nE$#b-^`doL_;JX@3qA@nBb#i#Z@HRQ2(f)i;)^s<;3l7)K{HDcOqXiwqi
z4=+q24ZhUtC%<?J8CsOi86C`lggX%MjgxCj7SqH{oiECjfOlwmU*wOPQ&`Kxya+hu
zQ`h{zkExx1^`M9ew7wJub-po@GF2B)UNOsV*i6-pT~Y(SHUT-hD}&DeV>H<TC%}dh
zD%bQ1vI;Yo0TnN)WAvYw2O2`~MI*;tH0#ylz)Ax)hOuC9`bGdVEg;JpI!hl0#XFuw
zf(MWG?BTNYU;794{dv%9&f(DHV<PmFfGmixcPM(#Ya=AIn4hq(zQ`4)Hb<5K(3ivO
zHV)_AaVl%=MS6y)8IBC&Lb%7sIMGkmYm`2kTX?<V7HQWg`;jAQw{?RA3FnML+8Q^r
zpPr+r!7CbRt4gA_B2B9#q-qMo$oeIxv`v~a-TG4t19>!qZRdxAnD`Xq%?rDOp}<K8
zm0z8=tWO>_%gs>Hoa_vS%G}Ij&N8S0`&GM^pbnBGL^Xtcjp7Fmd_M4V?~4af;07<1
z<Ti$K;aPyM{Vj7)BE3#-+6TIIkC(NwfuBSrz3MtqeUb#r`wcGDxu>zo^aYb;#c7G<
zg+VKqG-#|m^FVLOmFN1NIjA+)J5o`6s^8#LPk$P9Lcy}9ythyUD=Q5gMt%mE=5T!R
z(ikomjG)&W3cGpbNzT#SR9@t^#r@Ql<x-A+&u@c%WmxT6%|e`4Iqq<?Bev=Lt7k$W
zfmU3t8Q%3FFZIHA#+eX!-Od=bWS$$n=t(Vj;|(bDnIUpp>cc*#`96u~32>OYx_oAd
zJLfo0q2U0;1I*?OPH&!>M!ZGB&)v3pKDXN|k)!N>VB6w_iDI=7Liur0FZFG=JmM(N
za4Z;`@f7}t$vk-|c-s2~%DfG-OV6vygFQqXk(=DYlVM2S{`y2jV;A>Jt=?U2tFpqe
zwDH8T<p^lpoPP#nv`LmG=6o0}a=r<#c3%4E<hk79yVB8HeWl=~yE~(|a{o?Ml77!Z
zX{g^L4bd4tcA*RbnYkn3S-oCRDd7aJ1iTHo$u6%RNJI2Lh6)Baw56@ti4SZ)kG>CH
z%GcWG&=(m)T4G7>dD*Yh%WSnD@#;NJaczk@YyCB`If;)TeRkN{sUwrryu6!pGQ9kL
z&3b)ywF@ts7mqT)8!g`&bqb!)<HyDuPOjLFi?0qvqHY5_n<6?;rfN-g(M<?TY1EE-
zym<xN&6$qOX8cv}EOEwHH`p@X=>LM$qXB_ACY&AQ0#EfcR6Fh>%pVyvJ=eX!@2NF6
z&r?1B8_#7}1ZgB8uw<yg9#gxaXM_!(g4`dk@l<@~g4u!Nat~0nP{CmImo(hQx&oSU
z%B`2(Btnro%XoVn*68rNzLFEr@h?45P!S3pU@FJgYK|b@agOq!uu3mI#P`lx9(kXd
zEcN>C%W=rui1LBIRx=7W@yOqf%g;6_N%(pW#Do25z9VAx^TR03<%ANRD=zX>S9!*L
zEC+&>h7M@|Ge$4+pv&=~huK{Hzoj)f4QOhebpnwSOP~KRuitafS^;Y6Ng}pRMoxTZ
z(v5)USW*38wykkGzt;hJPb6$d_d7YqS`Vv@|HekCggPcS2RuiwS2?@GY5d-DoXlAg
zDLv&7IX@0yV5WuD5Z<n@*Yap%PSbb1#md2<*Af`T`{NA`9)4>io;Q8T{<q|jbqesc
z{&c6_mDefA`Lu>z0r<oH>@&z>kl|Y#_WB>s+s*@3C<w1&gDm*#@NtyD>kdsd!HlB}
z78&oL#{@}R&bui<iK_FF{{}wfU8-QrKOVSHjU=yX^5rYm<q2A=>0wEz(3P>M{U(x8
ztK(8=mB8NP+^2U8hlJ@q#?2om%(IN2@L_2=xdb)5<IV$@I~|26{l<$kuHePI@CgG5
zZ%cl7m@;-*^XI<6&phRjE9mWU6i9q+dazhet00XDwUNCNoT4=-T)pA^f-s*as<m_-
ziDD>QilreAM#Q>FXf|MMA4>Oqf?U<ErU{1qHJ$U%T^2!-?8TeX;!rPjY%(iPHa><k
z3Pf+j+WS6Gwxy1;BoJw#w9p>v(D`Fe<2thoolm&IWGDsYVpx{)*ckk`outL~STU>1
zw-xUK;8AF~+4ZcerSJy2#%AbdQ9mN@M%RpC8ieTh%Ksa!Q`UVfE<s(2hGxb&oT8Ol
zxzNa9e6HuKg-Gi-o?6vGZBii9Q0N~B>B|jPeLoF^5EuyThR#sa62HMoJb4La2Ilqa
zRnnMQRU(qIIrX7QPJG2+({jtnMNptWHGNM1SxwMYtV$T2O1C(!S=gs~C1n9B5C@rw
zX(xW}RU_;bFt%8<``gc8sPqH+8+OM=mASPc71d6UWwStW840hA18F7xvdX@m`@{yF
z<iDi__#lHM^-HHfKxNB6qU8dY+RR7{h1{@y#&h5Tw?NiCvx9N<HAwPi4Y`9PQ+L3u
z@0ZQ$QxZ#yJK^RQa$bY+IATQ~$(u)brJhyUjv(CkFhPuwn62l#^33rQRV%1eea)3>
zC8M6Y1Z{yxcp-)!1YLZDc=(@b*FduEX7*nUi00^d=!}^KBhnr!HZR3ANJ%n<DNVzu
z8-8w;HxJx1HZLzk9O;@tP5!SM&Vratk$e8#juM<hP`{n&mo5+R8k_jE<0m2}jH0^C
zH?4YglhDq6M1pfWP5<1&k<dUc>f<iKZKKI|56!$~Q#x_sE_*{2AE2R;E|sBaodP_=
zSPtMdaBK+G+?Sk*d!P8g=d=8)1yvZP_dFZhPZDqI8eyKt!{nD@Y(4#gdS<dC-(bD|
zVa7(#7&MGnyElzDU~Nw2bT>Lw78JN-d!ZalUG(wgt{1>p#cq^XfJ^90X1w53Qb1O+
zYFE>$s_U;A@`9*`ldl^>GWT=bE-?Os^sCn49PkufZky<oVUa}DVV=YdJr_xuf++aZ
z2}I%C=#wF+gKh!x=8QiDwvFlXQzRT6hQHd58C|A>f7ub1tdj0>C`Io<J@Y%ds=$&W
zo{2OG!L221`L0^ca65;=1L2My0noVb6b~wObO}@N6l=Lh=gtG0ZQyYkbLUMw{bo#3
zX~^o8Pnv2k8DN;VBa)7;!97KpQ=Q8-z_-^yG3{;h3sxWll4pzeZ5ia(E}VG)X+|eJ
zYAHGd+Cte#&QfdX6&X$|wa`NukldmqK6_#9E_|(PT-LV$-A;JCQV*U$8A>?Ac-$Xi
zeqy*S`%IiUCj)Dpa_FN&ZSF^z2q&7!+0P+Go-w?vbzSc9kO7pl{Hs+I+IVQE;?0jD
zK1iEH&OeQ44Jbz~Gv1-aj}?1&UY<cE&+uFn%CD*$d_I8`lD`LKkSCi><B#O3_*A-p
zsPldHbre(T=R4X3;)?1B@Mw`W>H(3+b3yetLHWL{Z&pe+VN3EpmdFd`$sA8t_XZHx
z>+Q3Q<Q#;X0t-T<3u49=jidvteJ<0SPAD6K4RPvVknHt+@qH-%-1$G_0rZcwTl3xV
z78-G*YPkj*H3hqD4{1O<$KkEoad(4CH3Avh`uh50*i}p?Zp!x8<m0{3#}>^#pm|sz
zE%n=MB#I-A`J*mWTl{46?3qipPW~mg3sTxF<J1>vR|79WR5}yQA&vMck9fSute|H+
zn^^Bhl8j$~MU?wh-~mjzN3A;vxKNZK4Xv~$@@NMwIHHo(F+#7B$}VNb$P0i9x~UFz
zp_GtwY*ajaeDmG8<3D*l*+u6M=rO8oRsJK|t?5C&9Nq={`yk@z^6V}8#x0A0-2ORw
zy^w$F0&3LjB0|*{8hP;-_p4q5(5C{$p`E$^!`b6<O>Lf}9jjf#UPoUO@&YbGb9r1&
zKl@NvqVU97io3;-#fY9ex^;M$0(=utgCCTxg4dTd*(VCs2t2P@rIoe0*wc$o46_SA
z8Bar%^QwA&DKZi8Jd3QB<aIuvs&lJbE)Oy?{ajDuS$nwi>^mYL*ZDc}2Io|@vPCnH
zZ>mF0fp<){KWpveK}<Xk*FBzAioHiA7Z4lPwkH^vB-AdG7k0WyqHLYa>EY?R80Wwc
zt?l*G_31*r5Zep&H(;)5+3Z*uvN}m7+a7v!<)*QXikx|8=iAxVeR|*42<&z5kV9{!
zQK-@Z?yF0EyZlDt_sQx=eu4_`Z{4q-XWy`*-5QP59;^%j?)~>5Oh8!jo;~4XAO+_K
zF6~Ah)nGjyr3H>1le@)nyx)-4jXORusMpU=g||zg?PKiYY;22NBz#S{U0|z=GAU+c
z9PfVddWhDbULV>A-t^D**=a${@=ix^!yws#R<wWWBG8_G?>Jgc7S=g<JY&=6@ZhfB
z;7;f_S$Ov5apz$-o3^v=hIEv3*!hmvl@RZfAqIH|cSCRIs#yFEs_*-3bk4@zLUZJ&
zo4~(G5*H45`N+;11(m+wCD4i-XJ&Utxl%rqk5(Jcf7uK95mC?;wrV@{X-3i6uu~f?
z=X+Xy0se2^fdf-hcSunqe)Kc?-2SkJO!fhYra-uN8o$ZuV%nccy#b>hMb1W#;1?Eu
z<Dud<ZTB`Zo{8hfhL@~-qJheK!1iP*e^=k-Y*TULua(~|8b7FD;CJ1=@kG+XPJ^8v
z(i3I>2i{WS>5GRyiECgTS0UYzTcFP%ht92N9zC%>;y8K>PJh}5RnpkL<nBMj-gu6?
z8f)!w*l|9&NM6kZqD4osrsf5vgS|{EVF_{Qc?C)Js$KV_BT049sZJyfj~I5{44V1e
z!RfqQ*s$!Q`$I`A9vTso+)t(4xm1u?DwMNeG8&fn%Jhd^C~5AT?7>^ZXl>}4oB_|W
z;h+3Jf1e%<lUL1k{23SONkZGPuh{Ti9nEI7m4ANwO4x$=-e}-|u~l_}U2LL|B7NC{
zA|>`Vh@?_k=TYCpE9_ZIpf|ril38A>!9CW<>9$3G*&127A-(ZFUb^*yvD188?Y6I9
zg;YVj>>l93!x1Xm@B%V}vMCvivLm<pBD(9>qcou6DDWL;QYTKJm>XB1zt_35tU+*~
z{7TjN!-nq`75g3b4t}A(9NWwn6LogI%~oaXXs|7x{glAgaZQmZ!Y7AY4|XM+Yp;)5
zc=wJMQBs!>Be7)3g5W<XeRPiBI@Qb66vIV)TB&g!R{vUs21iX*Wa{-=x>K*&LSXdZ
zRW-A<lY5T_jW`$iFYS}^lnh#m70<SlNRSfopudc8&S^9*ja{T@(9BP@l{4>YXC6Cz
z_%`j1Ogm=-KNWDBj*l_s>X0C`J`Do0L8<Vzl)h$CkqdA0p1r@3o~OamJqI6c)*uYc
z?$nG=GT(_MAaanZMkq8_6Gu!Vu~Gi4*iI<r-$ig2T&J0JurJ|s>m;Bi4Fx9B4g4C#
z-+(rrnU)ADbvk(X@Hnfn(&OA?c!_zKdiQZ^wCEp7%Bq4`Z+vEfVut2Q#XYcT(Tiea
zPDgwJBJ4?Rq3(eU5tNBGd&ylw@m3m;$;`};WnSxRZsy!az_^lYg*%U7hlCcztJ1J`
zPLi6_?H|leN@TtG!8CoedXv-@K_vmqc)0)pX;qY*mU%qMbb9Vgdz~8o=F-!rPn(eN
z%(~m-cLw@qIb<_9)69yK(`n>#OR<t~C^!zR+q8&0YB>l&2vP7vzx0@^18Y5hary0K
zHYiBRb;PXJ`B2>L4Suk%5vK6<GVArGH8-huEo7V6a+ia;EuRrJGNs76U1EFa{m|Of
zau%nMM~J2!>Ak_tZmHwPL?t_y8I)CvzANt+RzhlkC*}-;6;Q&rD<rcoTBXsc>Dn6z
z@#72cZy%ojYnf}KiaWjU0|N2U?*=eR`l{~x6tKD+a0TozzkkevpTRX>ui(%3aNcQ&
zYP7bQ95(3!%=-b`Uq7*KVu`=G(L6JqOgLl4_B8Gi?EdpikLsEce=7$c3FaaCLmv8G
z9!8;>InRi^QGnMlz7BZPl$@lWw(mYPqph;|#!Ixo5^_Rf{Fv8r`bNad!he1;^<stU
zwLWIHM%8~TYD6^TCL+iS?>1?mkd?)(^p_4S1J7WBQdrbpnuQz4$UGpyl(4~RD7yHJ
zU2L=Nt`c$6aNWDc%k@h{O`ZE%rY3bZ5>`^yZB*9ERi@jnX1VHRMI7QCiPH%2<bqp%
z{WmHTYwc^_ZKH;*jM2Jo7qb-Jfv&B9*e{7%0|)d;X2F4!J{uS3!-o$mB_Q>Q(kk-8
z--Z*`SrO%E$uo`Tocp54ovD`(^+1%Pxjb4sGl$WQQhe`X#^HW;r*;{gmN2z{BafM`
zPY^od#|Z7dCZB67qFBO=$wuCqn$3RLu$!_KGhP-KNvvldJsw_}f)|ZrR?jm9qD1#^
z?jp>`<a1s-U;7Y8gF#g@m%Rqham(juF=)_2eQ!#kKtb+%FSi9UfFs=o@56EEi81`W
zx(U~cQOGjgRhj#&trpIB%PiL&l}es2oSvPX)$>udKApM}vr5b4ly+}?IJJhL&GP?1
z4>db(GxP90Ga0b^-V_r1yO5$l)tLm6)rNu5?-@Mdd#tkfQNR=1eUvLDk;%y4pVK&y
zm??H0<#wnIbWzVu!BjqqZCFq<j_xA7R&`b=OLd?o!E%AXj8iL_^1qI4Yn9--oJPbj
zas0N0db14i+tq*YIw2(wN#UrUGzE$sso6!!taXijA_J0ShHVk^A;>KCOje~m`!RZh
z^weoJIsh96@;$*&y3<Y9^G@pdjRPOZO*c0T$j`^(2TvEb4-O7)PN<N#9=v~l=gqB{
zt0Kg5x6Fu$A5}{51l#|Bmbg8vuOV9l($@s(Oj6bIU-Q(G2`U}<2FLxpK1uw*Y+NoB
z2U!E2xnH9}$aU`j!AGHQbh3^vD6O)h@s&wVvX6JCXs)?AjpE2QI&X$b10InDaYelJ
zly7#8_|Y|0zyNhz+5HT5h?;Fsme9QSIjtm^YD9yFlEPI|rO^u;hn{j<ot-7G%iI`n
z19bPVF}xzE^a-^FBDw<hkI@#P>a|$b@;9fpkS#ScG`Doe@lm4ST1=T5q!4mXr!mo`
zADb@dO(+nTZFDAhHL||VHx-Vb*WQ!~SxR?s9a<6LCD}YNe7`b(J#qd~H^6?aneD=-
z)db($Drqh;lDuJu=u#!85Xr}@=Yfh|D=qADCz3Z%)%NNtNYU4-){k%a*CYY8;IqW{
zSUUmrLS?V^QOEk9r3RIr`gy^_bT<LM0xv~+R~RWeCMXIF@f|ZZe_({Nl}iV+F}v~<
zcyqg8_~VtJ+kGed<L5kIWD49C-5=hnhhZaEq_0ZuT`u-l>KE~i!IF>p_Q5u*G0^qy
zB$2JF)6oMiN=|7;VtDE1uSuFn-rY}()}8n7y_JkPO84s*{p(eP2=AIC2nF&bE%(3G
z?q;1J_>p-8d}_@oxiDdI`~pHLkuB4ecfk-hJLx%yX>Wr3J}PZBRoRj(X;20RF6LHo
zo?)YySs-_o!=`5Qg(2HZ+YK`9N7Y)vx6s=n3PXJ+S3hLGp7nHwK<wR=(yu8N9}-+Q
zS~a)Rfr5Xp!QNjXZS-2o_JJoTknO{F5TmYU`3_1CJH}4?c~3maKCAE0uO3D^hnWV~
zsT?a~XSH^U@!Wjm8r57Xc1J6I5d{jCUEG@S@5we$l)9|-{B>-^<tL&imw<^JUmBt{
z=QSOlJXuY$c6;Cl>6%D0VzYBjG;JLi9@&`CpW+&5QGtY6*4$01_imsBHGA}&B)eV$
zzb#wREa}+$x>vU1QkZyB>JU33Lw}`jYBSbV!Y?q8a`B#MW<<eZr+Kd<3Z6*9uU5E%
zb>)S%c4iOJc#0XFQZX`y9at@_^@wjHT#{VAWA~foyC#MhMJD^?*J7PZhKha!!bMGZ
zP}#LetlMpsHmEFf9e>Xbma}Y8P$?vMX8RcK7KYC75&o<%zW~)n*3;VCes-8vOkJZ{
z*{m9SoI<~bb%{=rpmRm=v~O&-uwu#VHvdXY(ZD|LSwo=ZIz0X1TZ5j=9+ZlGoGAcj
zzwkoNK|4?h0e;%#Ap^fH){MyWg#yw>?32*s_$VM$su_}jQM1vcKLwta*W!<Ctg%6D
zntu(M@ZvLtXW(yYDPYU!Y(u`0eN9)%OYGt`$~J#btksp+7<rDo)m4$^zr#m>YFoKf
zWwCzc<WzRoIC?B}g92X!R(ryJsN738O>Zf0>mRpYCM#)Xw+q)mqNWox!<4sSefhYX
zF#|h9R7&8E#m#;pmsy^Od@uIZ?qq|Q=ccYv&{CIzdrQnDS>H&juXJAPcXIiER<_>}
z0O*$s+xBqERrn*PUhwvRoo{#l8o2H{bhw6{scnvkRPhnRs?-vMcHCJuEL3YPsEArj
z>doX0@R34a(k2`r5cZ&Z?$0<sh8|f=pbicV4S4|9LCb4)o@lde&f;*)c^*Ew#@TKj
zkS%smc9Sy<E}P69YRbyo*JQH_jv+NWC}N9hElWBb8&*KOP&35Pe7xl@=iL;exw+Yr
zC`^IlTY7ro#8F~j=&wb`#Mkpj!do~h<S9oYLJ)F*sIKMV;y3*#<~W8$`1Rx#x+R^n
zUh?^2;!P2Z`h=SMMzlqTC8JPf^}}y1e_%n2ObJ}r1ZoYKK4xmy-P0xw=#7ZCX?*bq
zh`B9FT^@ZIJhw0Fg?pdN%!O2bcraYD$;hhuwHicvN=^c+J0`xcq+ep0Uz|<d^Hsii
z6=$cOu39hkZaMS;X+wl&MOWT$0gP5Aknq!<(KYB@v7*_&?!Jhe+-HwA1NviL$h>#n
zf5H~I*pxc5nT4@LEk8x(bmcK6gK%gjaFZq+zv{GAW8Y^tKtTza&5>Wtkx{;-n~v2U
z0x#d4dP?p`Zq^I`eLT!sv>5JM)3A>KDB>-VIw!%YW5M{6puc$*BV?^@aWb`18uWJt
zrJ+li0X{I45|E7l|6fYf;xBIY<vst|uC(afG!ybC%cWi@409pR<R9fE;p?3dk(fH#
zyhX*5NI~nJNkaPED4H-PuA(=g4h-_h{L^WQSgir%8JVeV+F+FZ6?s^IaSYnwfJqJo
z5go<*${VWax+j}{u)_=<H`|nEx1<Zj?+?Buu~9exUsPML#hv(LD8CMc;ujwHTc<i@
zO*Te~etq@rFN332fFLAC--UIwAmfE}ycRO)W}Jy?9F<97@hQYc>m$eGmt&P*?qFSY
z8;>`zRj~^52IlkZLUP_x<jrS3|FRt+iqJQ39sfLAn@cYT#AOiypc8Oie}XL>59)Ne
zJqsq`p0p%p9^3Rcvb?`X`~^fy_vveR*c8oZ`>59*2K!QwJ1G-0qwfLqmJzBFB91qk
zEDAb?+wEnT-BUCEr1p!&UVMZdnF@p{WGxb=12>DrTLLIn@eyMyc6sza<YB&9Lz$aY
z_mGDGc+3_h!x8n|68{pmEZ4DZvfX2!pxL0XdG`2+%~+Cx;*{2DS9tZbdtCw*tch7e
z>?ok<nEM{*e@$1Bfnaz2Iq`nFzDxHPGRJW6Y4S6O1+=A|_>Ed-8F9?U-MP>BTXRoa
zpZ;`-{uRGJ@Cw-mRx$OZ#E0246mN%%6!(0Il2{GwV+s{~#NeEdm<gS8zM)W+iB&8U
zQ<aR_zhgjr!_UQxAWMKYBUKYWuGE$yGFMLlPUkX#0^aFFwOs%Ibb6D?<ha?pH7Jj*
zLT>)FiaVXtboihT+!B|f=1m9mX&=Edt1KOD^J>;y^ycWVC5Ee5hN_GfwF(uxsi6}p
z)*m4xpH!pgbDNdR%Hs9D;GP0ek_RYC`Hw)~Wo7>PecR*zvU0Q4j9oA3=?9A4+cnv4
zKy){dxLQEcn%Fnm_2qS_wl@u;a66%5%~d>t{k6VW1$wR~ZuZts8*t9KzrzXEn1<>v
z*v;KPDty2rw<aLFdWdKbok~6EBkBmEz-6+_tKm=VKPW7p;6!onXZC{4Z&F`{$4KTa
zY&ERH^th|L{S+aMhhrxr2zSk#|9p2N1~~AK2%m;7e{(6J`8AvOMtzfdy4e13@S*16
zluu@Q6PW9YEyEW<N`0^9oaBeM`b^dK{LTEgTp&|swjs^!cMJ}nlbi;@C}%RXBEPO^
zHm^Q<%>E8-LobrlwVZr{%1T_oPKaS?Oa1M8H}I!}chK~K44P*`|KhN|;6x!?{^9Kz
zaT{6Q>c;&7H)aGdtxGj5>lOxs0p@`^xA>Bqj{Fe>aZJS$&E8(U_>sHZS>#Ae?6;}j
zyTx}W@5=5J(T~@QKcPgzoXc+TdAuUtzI|LTb#1#>AR;G-NLS`TuiLrhwLhGrC^I6y
zn*Y6|D^b`~q2Y15Xx4Ar@|kjdcZ*$0xLu~!Ak~k56p%Vr09<X!doAzf2vfc&osHX%
z8#6x2^fg_pjFL<xB12_9p`JJ;vy9o|v>Hbrii3nF)DZYxvFn4y)=s7`*^-W`@RT@u
z1k;=Dx&s<!yNb^VZY+s6x3Ano$-XlB9H{Pn_U!Gl)+K#9p-bsI5Ge?_*aJ}nK2EUm
zDs7`{f1Rh=N+>2F<r(*&99dM6b9W$_b7JtRqOssQA6m2Xv=X@Y{3D<KrXTK)&af87
z!qce5ovqWrSiEDF)wwh*$ryT-DI*xmgyiaWaQzbRTR!E5O_XH!7x8$wB|h<+;7j6<
zwO`Dv-Cdg`6(*<@rQp0`&&>%opgV}r(K94O{@Z*9P&6>E$v@b!Me20HX1P6rYf8pH
z4-5=A5kE75?d*o8(7)zAj-G{)<7SKPw}SFQ+zqP1l^rT<{`BSn=O(_`tm;~lit(4=
z7g+pf;(V*jBq0hj89<Od4c7W}(fW!-^s@<;aeNgPL|1pa^1fu8TDHs3!ju$_u5%vS
zV*@AtY*=5Fne@>|55MQqlOlwa951}c5uI!iGj9<aT#^Zg5T2?3KzFkLUaFjGu|`ol
z6%=~^8scYpz)7Fg7pECfG~+kQj60Y6?6Df^0^VJ*?)69ogz<c>rl;ffSiV2$jeBAt
ztl?3nu62TtH=n6QMp7oc0G^@m49PTi`af;&S_qUpC8tWC6RhWdAejYa*~S4vZl(Yn
zlETgW3)d2Im$<>}whM$qC%Iv}pCGO(G$NH;p593Df;!RITZFT~XY{oX_A&9LS3-SU
z2c;VSOx@AG{ZFYxvF`x=a<$h355*gl?qjgVCOJw-3_d$Tuh6MS%-PZeB;o*)B=<Gq
zrGzPlS9PxfZFi4$Cv7FT&L+roHBuKTp4D4KF~ry1L;ODNj^P!KJIh1E&|Z5v!5<9J
z0b*DGmrYLkY$Mzs!Pj){CcD0s!{S#y%6M2S94fj<=T39$KfXHVrrVM-T;kpB<H{j-
zD&_j<;5d#U2l^gS;Gx7IM;w-0xoG`TKI8KbKMTJ^fA-Wi7j_!=<-J!N-)Pzf;#V*~
z=O0(ZG|_zGmd8i9TL=|Dd{-Erhw4$OW%F7v$~vX9E*6LQH}iem7%XRNax}p5X`PX#
z)Qy<#Wt&n<`XvN=Q>m3_69PVylyUG~^_QD1s-#6ct)1wmE}MB6MFmbzo<yR~If#W1
zZRZpd1y}4&Z^Hp<BcD{+OX*}8gEC2ZDY26Y!eIgfS<*KHkWVlpLfk`+t6`15ppGUd
zUfs>aNd)?09LZ_?+!*vT{vOZrjGil+d3i65(<KLpPzh-hT1Z^KSuTQCu8LdItR`9?
z)B%eWbHLu;5nMp)&C?M_`xTTe^V6bOv!5@MQV2>13iiw-@J#28!wokmqSpkl_zQ>$
zG2@6F@_sM&Id8%9ow_J|uhE!P?G3xUu51xm#amzxQhvw=TJ|#5#%r_0_I|56D&a;-
z_b-467VkbGyyw@9TdC+#ApOAo@bd7?j5V1NC>E*;)_y3kNzn9xWhXjDVFl3UAAqFN
z?ZvP8;Dy767G*Y-PwTQT*}hCKqwd6>WBn@yi3iJ=FY87dc<l+}FA?V-7=+^phYq>~
zyK#<9yrGr=o=-4m5ZYHfxw?JriY>vn#TwKM)r4`1JKEq!Xv7bu)D?=M-=-={f_njK
z`$%l;3(B)D`88ei4_zk!aH%zZddL|mXSpGCSmAE5=MYNrHpXcN-F7Ji78wg&=D)31
z_DUVqW&@^pusLn<U;Ta=LqZhWYn;;wf`93q^I@zS&xYz+rQ05s!ln&YgEj8|i#5VS
zolH}E5n0Jec(>`^nxh$N)tGq$bZIJslX!~Q*!m9X1t&?>3#wI@yy=v@a1p8~?wApI
zI9%i%k*y;t!x-%Qh+2sg8*RM#O2?NC_3_L;W5~#{>OYZO8J-CF(Q!_8fVcWck?YcO
zA(+q-7!77q9K8i`=qZ_C*6+5ZL=2nSX=>dJ3u`qf1%y!Dj%EIPjh6ZkC+C!~B)6;Z
z^*+s{Cnbt}!3AHGWmiMO&BS%ucawWGArx-d5PaMycXojrM;o>{*HQb4C?(wHJ-~%G
zGWV5&Z#|X-IuQ;!S1-Y-%6(?kOzLeA=Y2fz^Sc+i3$87(&3Ex%m7LQY-|2f@U*zAL
zqICzRbl_D*K(9*C1vvl>(}y0fdzBIbSvh}?jC*V{9tmmPsq>Na_~E6FQw7j#;2C@T
zFYqG8O7CZD>-HRZT6baxxDi0QZW#<u3;Q=u2yE(z_KQs2!Ac$iI^LNf(3Q6cw#GD3
z^yKw7)S8_)UB5`Hw6nt~-zDVq0`}VPNd7W5R5ainV!l>&9MZXK5=ugD0~0XMX%SgJ
zZkjD4_R!eniv1r3fJ1%{f-^7rO!NekJeyYJQgqq&JLD5y#c7Os4i$1`X10`BE=8wG
z1x{7%9bGlV^SD^}q)KBSSM#s7jk8N3HrqCOX5^$(`?gRl69Nu6PWScGnsBq7bg03g
z1g8K%^TPzC)UNPVAH+U^6=}-qtOC;~QN;DV&i7cfZfQ)TIPFc@kJlT6cRW7mJdr{D
z00c-(-@Cry!L=0&)1FoPZj%3xUPCL34P&gNqG?iv-0~@dG*iwbYn5BBGUnM-DrAQL
zve`-)<)xB1%e)R>Y+V8`ISG#6{-;G5bHAqhC?hU4^JK6uBSJSBq%3!7>tq>w{xH2}
zrk8Y^H8qio+aN+1b<vZ>K~9pHLKtBsdJr;UM9TJ`3mV%J*+SKdOCo4DcK$dzVen^o
zM@s9r=1L#_J8cu~;~qJ0JT(>Xom)ixjTT_tYv1Fks)&l2;Sz4%Ju-VNiq7T!L*HpL
z0(G_(v?0yWlJm<^WOCGd;!f@1{$m*u0@w)9Z)LM9uwHq7DxwVlOE1b`9C3Bmh2WlO
zK^!!;k0Zy;-C}}_E7iRUb#{bw5j?rz_F&}=g@mhpCN~n<SA$~h{&%fp#olX|ml?!8
z8DkUbn8KS%MwMkbx(PdJd>?-l=CaiX<P~A5<E`3f{`hrN8tlNQ=ySR}Y(}`R?uD{_
z*;8_#!2q531k#>HIen3JMWtYWN{*BT-oGm%f{lc6biJ1QXJYZ<-V+aQkND~I5B9xE
zQ;wL_rMl}158thGJbBHxSy^G&&l(S)<AsRxH#s)&ft>jZzGYp)`*#kjq-_HeEFt&!
zjRODej*MF7ypk-JVt3z7cG^2_8=Z%@%KVjIIG|zvVN^?MzuM&Ax8lzL6v-Q%TJ>}0
z&<P3|PPPnn23CCthYU{YxUJuB4;jdo0ECIdoqQzrS-x*s;=cjvPKos=&c$uTyp}+J
z`eqTSph{LwmmAKXfL~A?7(E`76m<l;OrgGlg!%8@>{e=`K=#H^O_kd4aiN%w@Zs>a
zfvJ-FS4C8-F0G7C;&*Q|iv8nwxNwf8+gq}xsh{Q*?a6?K3zF9}*#Q^(<K~;4mfIpK
z9bxRK_FpTBlqo!T3EyDys_Aewkf=DkY%7?8C>)C^!#S^0av2fASaO6JYqsvZ2w~Ox
z*!OiJ+d(ms>v%D|nH0$~j6IUVzRV}PIIgibma#Z41gh)heUwAdi`t-&ji<`Ld>zyd
z#qETGU6eCQ0!p;@i(lf2vb3OEZfm_Yuh0fHZz>(qVOTYY%JC)B;;cLRO8D>kaDDPW
zIp1{i?QgQ3n3$>K%37_N)f*~I(}oTV{~GS7pxT5y1$)4~(A}rk5;z-5{%pC*3D-Cj
zHGZ0!zh4;4x`^pS{HoD*tk|($MTqd=J+lSWs3tvK>=hnYJ%rwrCP}hwjQnFUodgKT
z6M0?n#FiNl5@a^f+MspQWiTnmDNWJ=ut-YwsxQJssW(a{IXi7@1~F>!A@}WLOqlj<
zAAx&5Uk64#^5t|UNNpK$Od~G0^=|u67q9?HnNFy=#_#cQ0A;f{Zrem=`?wkN^d`q$
za-Rj?5Z#ZS6wSdA=b<oNb+6?fozVA%`qDz1wT}<xURa4T7EM{e(d#v?_m|4xE`@;3
z<3v9#^_iKTh<+3g=BMY$z3z7p0i4;=KK+s!M75MuL#P?Nf&CBvYb8niw;%3QaoVXk
z?zMurXh>i3=E<C5_utclds=`)uRv+)%#R}H*_p`VGY-h`(cW9#%o{Si?l%JI`U(h}
zhtJ~afXgINSHXC}3=?J9zS&%og`gs!WVgZN_dms20Gl+`bfk>g%Dfm!L%*ce*Gj6G
z8iq+hDLE2WG(mF@U>v*PHO(;T@-41-8Z*HY1*A0^M|mlNKM3014C{81Ld*G>-;Ku5
z-{!>&m+sU+XRWEC*;#F^DJVyZG&Vc5m_L$luf4~ETQr#7aem7524%&kjRW2&HOt;G
zKAKKobmSEk6`7>u2h69CZokaQOG$gnGaPS-+ErI5_sztwp|(!^kw&?NO-seJrHIi=
zfoVVd3kwPgNR!_uB#I5)W>H8FiKSZ|jzhmQZ!j~0WxucjD9x)?$jhkvb%F4ePoSbl
z1KMmYp_=y;vAdiP!na0$W*$G~7Ls(nEirpHZeUNJm#}cZn3`p0`_0gwJF2<ha*gwK
zsML7R;c!UsfL=orbg6ZY-#oDB&0TafokF$<Z4k&bRQvONA7_wFSn$%}<bN(qZ+k|3
zC-QHIrd$<(ELcJA!n+HKKL9G}0*;hb4J^oM=+&waM!Jw7IqWYJm`eEazOdJfdP_4?
zg%=4B5?0wzy$OQw2H%z|F9DR~I{IB*VOnikW(a}qePsQBRzTqjvF9xqS+j=17Jo}X
zIIWI6J{;{CJ_u}4V&WOey8)1x869K;n7`QwY-N%;cXmefe_P4o*mw?r%3Q~~(1J3I
zx`~x@o29M&d9~JctxBb?>G_ncyr@ce!+=e_SRfpxtHL)yTJg!U+z1vkkl1Ljv@R4P
z!maQ6T=T8=Ai>8Tfz9x{b@ntb-}Wm>Zkmk%*EWMvHA7UtKXpp_ZWG*jdn#=M=8kgx
zP3rw`^`tOGB*$i^@?i~YSWb7=q%KBq--BKw7#L1yd^bCix6w;~#+GNl1#V|o<xoRb
zgS;v1Lj>>^rIzK6)Niha1-i?UWdNN27~*qX0Ze&hjrXuxa6ga(EkhlQ-<fcjWpdb0
zFv3dXE~M_?hRPEghd?end;(9-a3<tF6dJM(Fp8lEiI1@OU11J>E6Y7ZXLPUAXJsrU
z_|hfjwhd2H!aabjLvj@4*opSi@h9)wWx72mRoZE2A?PzGeDO2%BOxl`!9uVupX@8?
zoB1g7m@w7`JE`PkZKA&OKPSp-sR62a9E14Zpp-krIE3B+vgW4WW6Y|RTPo!57hG3~
zdvYLiJI$x97KPhFyy`(ukEp{t_v&nBoxih&pA38+BudO4P39J8(Elks9Iuwpuj$ml
zS@HA839gTFmxm+giDK}Y?6${h)$*qZm$qLaKoX9ClW@?Um9&|^;k2f!o|S};z%3)k
zyT77*ble&cjWgX*LPbp=$y0qAihe*KB1WjURTb9s6mSq`4f&M`nSNZO6u<dwzgS<Z
zTHU+Jb83_>*MfN%BvRFk7W;eo#hOhl6v*7wT*-**vm8W=LV2qMT}DMk1+b4BrT&7a
z9~6C*(<E=GOMVpfV$~52H(d$8{H%G8SQA01oukEnv!hyjA^b6w>P)`hxwA5NoU)zN
zUO2@q65C%0#)&&tQpv5ccDI-ps-U`3w7POtQ0Zz`^DYX)cE&^{FIdsR``U$d<P3Ix
z@-xU@eYO+;eQfu9zSgnR5hwf0H!0lI>fNM1slEISRQ%N2#?IN4g=lqkN>Fml+}YNy
z$>#afr)8G&^N;AnJ+5}hY1<%mt=+>~?Ht^?g?8e62;;~6mBIw$IX9?00!aG0Oa$!9
z)Q_RuXCYs}p!8oW_*0Tf+sIml+Q1}bikWJ-G+i*!Rq0*BAKNa8wH}2zuLnlW)h@##
zmD+lSuefrtzfAu&p@yf(2M;v!5~pHjWp8#hRjG6lM4_YTqb;k24-yqr?`hPFtrOd_
zH(UD2rlTu;UuYH8uM)knhq?Xdw9rpN3VId$ixUy$On<mE#v^GzI+Uwo@VX#SZ4^NP
z^Im&enhhL-iKz+Op*KL2pCy8hYeQfqTTjIu6*r=od!RLMIc43of+(cBs9~LqA6!H5
z^Kn^Uq+ll%zaXJgVt@yd)%Gd{9d-ULfORgtu^yBxyKg9_i)oOY)7{lJs1FEh4F^OB
zt=&=d%kH3g5|{Ipf<z)>0!G~I<dpUFiNOu4BcNsq5?(}zA11qLU>LndY6|`Yfd--{
zZgJ#|x1b!`SA9i$A7@$kTYL+fb)%3;pe2G1!I*i_8s5GVb<>Z3`gDVja#AhmKXM)I
z9&6^J#0@5nH&?b^ETAYEfBb27^WIDJIw!5LB7I&afp^(?TA~Cs2rrdZVj=6Udfc64
zlXxlgI6czz*l^z@^E2XHe+^2_PQnN;5}_gi2zPvfiuzv6j$3PZd@GtgwO)1QaiYyu
zfayjf4J2>L=l(b$r2S@lKt1l%x@Kf;c^_0z!+%AHpKKX3O1*WSPsTc5mRguLlM75p
zIDM$5pxhGu*5QlQ@=_6A77YdqF+xxK_ASCxqC%CVxQph|sf-Vqb~aZoUv9p<{)Wj-
zlYEukty6>a9~1v-S_LQv*^{XrmFSWPa{Z=a?&LxFtpSj8>WR5wmIvZ@zX{Gc@gMtP
z%EcKIw@};Yw`pL5RVOAs&9?g;oX`#3Vl5W|Gb!$S&+PxWG}+&^V=#6<+mURPu9m)$
zC05dx59Qw1CUR;#0_Pk9b(v|`ZfN9d{~kNoEMh9VI4A;6drV8Az-b8bTpz7g_<Yb?
zPJPLLj)}1cZWmsQ?c`Cyjd-aFZe*!#VL_!Rxa<9H{^oTC_fOVB;D)+pTl`MW#6o`<
z9Uo<OdK5yR;pcH}WbE3nmi_Okp~LW|a8SuRZP%E_620NXu!1x<+0SlsF7=AQq}l6(
z80ISF_TcD6+rJhhW_DdxqF!K7x<I<nOzB#Yyd-n~;`i)&+n|p>R)|4y?Z1ZcsOpNT
zF9WPT9@=0`Md=3yF1fqFpC>(2Qid=ZxJl+xs2a6iI69Lgo?kEK0NA&s%Q2ZM|AaIF
z3=7!C<#9A4ZIvs!@Pu3TfcFH6U+0d?V!kpbKh7ovE$zW1>H(Z%QX?pd;N|=W)}~7Q
zw7m2Pr`&^oIK7l!$fh<Ey6n=gq{+><x_s!`<<E8jS1Fy%`pd7NDB5J3>o5AnG!ds>
zoa9G9U>yjte;T)h)yJ5GMBl@ec=H8>EQ`20%%fu;_*&~wGLA{s=U)+wbzQu|5u+8w
z<w-#WD6$$5alY3ps7te$AVtvpl^zdfUz4||4A=GfHU>u;Ne($C#D(&hC1J(0r_urD
zQNG*-_esqTQA_|bqs&skRDu?_{;}KOmTHP*TvW%n>x7rv#Hgp>)-MNWZ$8i}X_FT<
z7ot#0EB}xLvG+Xu6K~X+2(IO>TR7!G7)jzfj*gc!R&qzkjZ<>m)js1@`;l2;(TrQa
zH_CxyrpgCP^lr6QA>>+}b}?>#Z$*K0+b3MHmAR{*u~FfQ=I0`!#QlI5W`WklJ!lL3
z{X52A_4r$K#N7MB%Up<=LY_fzJ|E^P<7S&d<)fDDG~D;a<vM?fN3u=hWd14xUH-d$
zXEi_iPE;Dsj5g~t#E9rjHW=yu!aT(v(5hb}baFH9C9!K8g%u#<#UR7Ku-4O`Z5!=?
zN2mRZ%OOIlr^#ss{p|u?+y6@LRNkP1WPFT8|GqkoZtc&cEBw`onXR35i|fC^|3)7p
zh6}lFcPnb$@?$$=jMK`hh`EC!YsI7X2K$ZmLBIOZ2{K~t^kFHt$LYA;#2^!8H<9HX
za`&xm&?g#~mZUSz9X(#jImvB`W7u-J{New(q*lxfLg(Z~q<Y>x{KMMAVt&9ro-H{J
z6$L&+8Xot~>vyL9p}0?Qot&dA#}vx*57G$(mKi=axO4=;2&ct-?-KfpocwKEmH6nz
zJ-S|T@}GNbO<q&l^6!rX@^m|Y9GUp=#WGYph99$8D<ugb#a0A11eAie1SO8-u9!Op
z_z#NVqALq3=yMXl{j-7F(k^=cP1_LvXGU5o)uu_FFr{|d+E~f|r@{r)_D3XRR)R)b
zw)1G2UrC0{4y|K1H&7iJ^dYSzm)_jx^o%dOFj;6LT*KfYnHda|`G%6?xF)qTwM)JI
zVg_#w|2-A#W;EeUJm4wD%&@lVR5O!f{J7~bp8`+{yNWLER5k2hZ|^`b+?*ijP%=cG
zTasZ@!EmHtA3Tn=nzxhE2H%uI|BuFzz*SRAOJ+GcDH^7+b&j)CQ5%?A{=cTyb(x?h
z#gDlj-Qv4QY+k~me|PGFSrMe0;B+PFJ1@&5U$^Nz5%jV6Co%}-vQRXl5Jm_Qy{yy>
zF73&f_FlbtRiqcbR@9LhI|o&$At@vYlwk4u)=;!q?cXfnQg@tOXhZj?3|T9&URFqK
zREj6zc<ezDf=NhHPZC2u+CDmU28^&fkI5jAs4soz{ahg%3&FM4Fvv5n#s0Fwxs7_k
zlz?gWXYO~%qY1mCl=iEGM8E(nR!PMSC)&lsLWsI4xN(i$YkO3N+6E8TS|?JOdyulV
zb7BGJv;?};0)McKAQKelbjn&_`0i`+@JU=#)E#(ka45)~XYYe=yAPvOW^1~)h%d2g
z_xhic;{8;(xZ?Uuk2f2NMLC?I;J?9|s~wxKgC1^n*{mnBj?0Knt89UzZ5NEl5_=EJ
zTeLg2*IF5M9_Ja&foobHQH$f}km=1mxJu=eoN>{hFeUBk=-s@yZM6l3hF1x-W+NLZ
zU;KOe+K{w_s02}Mg865rdhh!et}v0-?=mO(Fqz#=@|z0yRtuZW<<0*y)#1CMyJc}H
zUrY_k!E*bG5<CHAjF+oK+x%nAMdvOe_n8n8b^%fYPf#ath&?NLF7&=Uzr-4?+@3X}
z-JmP!ZKDTRDpf|*86mnb7rIYQO3@!;naA+dHb{yBNysasFD-bFMfmkFqb3=C4@w`8
zk;V*g$Fw9lGvq3S$mM{@38n2HT5(>jOC7=A1SkCef?)wycl|OW#VM$f(0!-iPQBUH
z38(<ndbYg#hc#Cr0hLgXyF`<RzHIWzBJ6{|f_Z^zcKGmDn#&=cebK9AK(0?jt<+A<
zq3an@uw^G@tF;D@eUNNgct#(zYd8*sRBCGGOb}1wvE}w1YrdEk$I8%-n3Yz4;Zt(;
zlmuZ|aXG5ACUz}|)k%b_tUwRN7dpVQhuQXX$qAfYrauuMlBCRXIWEy@<f)Q#Zqp@=
zO^5mMSA;Z|E1}Z-e0JbsY^s|HH{N)O|GxXOGb2>d%cy>^DXaFNj!ExkRt-eBr{JPf
z@`>{fEHf1)M9({r8j5_fQUb1z$->524fHpY49%|bOfV!K^s03{z6R(z!Q0+%`LTHC
zH7g!b-$a(Y6C(&H;WMqhTKJ27vx^fY>VX=COJ`_$<QL7Z_Bl+5<mI~#<snjMqSCmi
zy?)o=Q_4DrGOJsLVR0ouN-$Fs#M_w*E2HvW#R+4IhXt4M6QVt7&4X`C@-`|k+H%1c
zvA5SUmobO(|HpGIrb?Rs?BK;Ym;x&3Zc>+lOfN|4c-!sW157$^P`aa`av*}tCYg?a
z871HEN)@*YhFD&gt%Mh_xD=YCJbl_b0FpTV1{)NqZaA;fobtXhiCWxh3n~2_)9is1
zKI|{xbh<!7c3<k9D8@$8Kj1DNr!aToxohG07IDsY$)JHjlVO=(%k~_JAs_>W5tnE#
zGIG&Dt#YVk_BQw|<7J8Za=-A=2+n#N^`OfUWo-s_`?2+Gem#Ap2kw->{WFjWX3WOa
zfUaRclk2$6f(*L&rxcAVgkay!@-^vg@W&t?(T2icKgf(LU)NPf)s?0VUIt2ZOU!ja
zLcWXU0B8H>qL{_}Cfc6G8Fab1Gr8n@m55P^S=o(0fQUz0UEmCNpA-qlNyeYc$opwm
zIHx3>lRa(v&Y9eTb{TC~y$OC614c<iur4k|935-xvCCNek_<9WgkW|H{xmT~S3k!$
z7F~eB<E&FAVo>$NQ*i6uUj{2%d9aM_brhH8$pjah*;u<B2(&|1mF@2sY0Hr{;k`Sy
zWG~z&!nXIt`+9Cfpj=SRx#4Y~^!d+Fvk?4qkIe{@s^nK#F0J}Ms@?=1%Juype=JF5
zDQ!Y1El8+iEo-78(a<Ja)QM!tB<nEpDMg|hsmN#<TSZQmY=hGxDkEWJFEMt?K4bV_
z_cJ=b@BevyUY|~uo_p@)THe?Ddf%IPJtY*>cThItl!#q>B?lG!+`~4Lw|`0)mZb`I
z=}nOC{?@y$uQ|DfAB$D$-t~w=`U2Aj)y;(Bt1QLt)}Kn2hAjSmwaX!izl-erX<@^w
zaDvIAD1nrsZ9`^IbyB_XMoGxZ91t9s2xQ#gNzoHJ^=EKo5^yJGD*vDP3~WQS0zOM*
zMa~-1aLf65Dsg6}pd(c58~})7ZcmzaA7eY-E|P==qEw4ZHm&YJ%_pp^Gjc1)=u!am
zp8oBVEe#GzoX32fl0S%{dW{Z!!J<E8dU2M`cB6GSxq&nav9Iz&gYE+9`wMXR_@RkJ
z@}M#)1ulcdhnRh#pU)&>+n?cDCkr3M4gXwHu~f<dq((NxP-+9-Vdz*H0d+Xy)A&-9
z-L3xEyVTvdo;H+?cWMl76U35HTvdIzdN#Cgqy@A@Si2?-66zxEJhY)+<(=ja5S#WO
za%Q119~Ng}BjgVGfFX?nU4>Bc_%jOLpBfDCUlWzF@DIIN$as(e{C}ISMDpZbC2N74
z*r((22}S3s8QPa`TP#VvhDG5{av2mIKs3O_o%4jRGtRtf^EF^_6}Dr^15Fh`61Ta-
z18@LyItc2S3hy^VABHavmNA+!chdO}_mXMH8BlGXy62SD*9BbRaTR-@(>f79q{U&W
zh&PC$QZeUk{308h61tR0^A>tm=1XRglnv1Rt06^F+OG30X%prNAU>DSk=-@RqeN|2
ziuZpg<ax25SzJ_NaM+4xQsi-4bCrZ3SY%fERk8-m3+ch1K)5B9LclE)K0@;E_<z#>
z3x9NwX5*n(lq`5cXU>dGq@4iUA%RpcTSf>)!1LzxSSjJM4pfMHMqeW`*nzN#jJ-a4
z-HjW5&7w|ssEkxDm{>&%!Qb%`Zv$`_Erflb<#Kar56c?1$K_H6{&IU|Q&#hi?njND
zmcT(qSh=m}YEWpqig7y#lpa9B@YVc$p+7tk9e%F<f0>FC>SisQlS{l(1pG?aZ=2xg
zP4#U2W2@ak;;DUZI`9y^_dr-m-^S^uUVMt4D`4>ickL~u3C<!v2lNLMC}1mZzIFlc
z{{dnWMXSuInR7|-Hvix@=D>S&bmrPKc=z8~#}TBIEZm{hR41=a=(Oz02!L;RW#C9_
z_KfbIE$E}egS-l#VdkhBe)b9Cv$p@Y;$HkxHK89I)|Y?XIDJV&>bgHeF;qN9E;Z|M
zP2J_K3oLhF>V5OKx5E@L851SWIg4-J%UQ;#st*UaX`efbnY3^$_RAcEUXi;5A=l(O
zunS6_LT1E;;C8Gx`p=lKjp~6gLx@}=Q2!0~0iCN#L(F~ex!fhEp?P>X6H5m2y~RkH
zES^5g>=E2muc@<jl1cd;f}hg-VDOfX2p&WvQ7A%_?ac!_vIb@ts^L}T7C_=Ua&z7*
z$!VE&&jx%!VZOv_Eebb4N~%l~O~PqPsPF@i5%FPAXDSLG!v1TvFZV{Oc$H;8v-s1W
z;8crSb*Ol*-I2y)Y#z)2vQV9(zaPd}8_q{tOnuv4h_{>Rat>^)EmAcXec)CNoZ%1I
zxui-s32v|8v{k{~66W7l4i@nTWVFs`(K_ZkU-EDuV_!Lyp=T6n2@I=T!}MN}WS=Q~
z6S>~jJba9`dNh$$|AFn5{!l39{&F2wChzLyr2$~IN9dM)s8OrUP<w0yKg6BgE^=@#
zpkbkVIi$&h-<j}ZO7UyY7i{Bq@0U2ZlH+0CR~D<Uxj^+zS@)ypF{QN9ukfZIS;kyf
zJUw{WBuQY2>>@|ysR8>C)oO>u6FjM`rPk&VE{fe(1+nrRFz<(lrNsVpLE%%1RsbD$
z-a{@yb*`@fTl3Xdz=<s5>i${q88)H?@<Xr6K0<S7p+ipW&m4ebjJgS?&KFcyDBdc*
z$%+D_OEw7}WWVxq@%m?cAQTI@zA&Brbe0hSas<C%5(=7HHRGr}3vJXUb~hQ_(CbzV
zMWB%B<C&61eJBFT!$yNG199B-n(3jmAFZJD!+ZMAqGX*L+j<1kMdi0dIRc6Sxjj^r
zdz5k~i7)>}D;~MBqJYMTtV}KxJ`{RTuo=GDoi(HB%a}-5b5DlLrU^o#*{w9R>&f^)
z7z~xr4pt6Y4Mm;0cEQRQmC~17-b$miccbDHXe!_z1yX|b04YAMRO>3r8m~|wr$z};
z0?Kf+o9uNynmN^Gw#V2Qe!<8qO$@Oic~3vO{=)Dx@fcsy9>L9kQ6s9B6BUS|%H*!=
zp?!7yK5{EL`zPCWujTt#C~(bbTWDC?14b<CbBKXjzw%6mktg(CMGfQ-va-9_+LL=h
zzzaen2mzB7YEJbN{09?PX)G<VZL9g-<)`Jy3^6#j6+cns6@x-*d|0lfhV@<9QFX*(
z8!}80EO#HGcgSRhhiM<zmGz&uuSNdZ#|`RDe%15IbN`Jyor>bJaYKW!Uy+bqM{ZJF
z=T^)sL<JBsg?9$Hr&RFzPS6!e!z10BFGWye_5Gd9mpfmhMv~CRi@#KQxS)I|5Tc6&
zmybMy#(nUyP_u9K^7dWJn;z>T2opRzPao58LpW&GB67#>i`Kuh0t}<gIMXALZX7Vj
z*`BGCI;5uYCcBI3QHw%cN97uQ-oF9=PyP+=s68>7g6M|ynb5K6;o$HpBhK5BOMS#9
zyl^)#b?dDol6$D(>X#y8EjKoA$R3F0cIf8Tg=J$#f#yrBZw?gWnbtBa$<E>OH47Vd
z;Kn#72g9(vKL6<dHN>MzEVX7vY{oDx7i%T{trk$`Noj|CUeRAvd`L)TU)=Bi%O-9K
z|5Vq3ZeJsJ=}s6~C@ov@-<VSb@xCw{93mL5#6!pqu8sB1TE{PP>xoUYXroq>MQt+d
z`H}jY>j$ly$F_vVvd8(zHju~zy-+U_+lw`)Q^u%8C03R6_c~!{^DbUlnQYg*5YZaF
z1gKqK^NcmbjtimBtfR)odfKC5rD9S471Jfs>CZQa!;sL`C^!zfFg(U9X$tD3nfgp6
zAR?-E2SkkWW%sgGNn!Sb3fEJ*Ou$<`q-c#O6qZ#ob;jxUn_O+|oZmj8LYELHYM_D8
zR3Hd<m{0KgDEf&Xnhatd!UGO+L?7O)4%I1wh;n?^xba3fO2$pRqU#3J2y;P2C__;V
zXALama2m#Bf=6v??iFZN<UO@QaH9Mt!=M3uTEe&r<6M_-<JFNpl9Zct*hLmYBR9!i
zCZ8+ic3qokavA9a7uEwEXl`k*Z{8SS7M+C8O6iw#7Q!OyVJe|wz0YA%?rvz7@Xxi>
z-TkTtYnNDiLG=w@R}MToXqOqa`?B9U$OM_jwA}T3`BEHRUgoIAT&$E3-S&%1hg6zm
zt^{WG%Nr<zgM6Hol7RNYXaS$w0{f4s1uU?a*=>P+bq|B&4$MxOujZ>(Tg{*{=jM~_
z`pUGi6y1%$m%dy9Hyinm?7l&|leLn}RQ3#e!7RPRE3m;h4zBolJQM8EfhugNk%OtT
zwT+Ogld{lmbSbQj1hg}PR^mm7fa=!B|9L+z@%^Ukzq<AizTv)VDbAo$KYhj@&0dsl
zZyo=VOQvNL8E`_v3hF~gm_o7%`N!n>^HB8!h072*u2+&2A!}VH3evOc2N1gvP?FJ;
zfb_5W#Xie8znj-$2AWbfBAJv>h3!I=jnFS0gaZbe^<TW=d}ou)<>bM>khOMC1QrvK
zIvW0@XG6ReHfA!coU$G^Y3Jv}g;Q*>;Df;QbQW$Ly@WHGe@lHis}Bqg-GL9>j<Ag0
z=&<22`8oe=QAdM3yf;JuD>nfy{oc3}eH;-!XE~q|v&5vfYwCQP6H2ENre57;@A@!?
z50}jU(*UqRx8m2|j|Ih@0WUzy6+I-B;`sRRlktiso9o6ugb9s^gpPIVPU!Aeiilqr
zGYBgqpB&n(m&~OUOSxTzH9>tvw>m-#mh7J^FvJUTBki929~pvo1@uk|CcP^m_PK~3
z7!Dnn9uY^t3AM^X&%eDimt`gRz3HR?R&Qya_BOecfbwl{&Eyz+<0ltt+~zm=DAIbG
zr$g$+@B;)RNM0VXrBG^}7!)3qp)Nb?SW<=z<*=RkD#J3=4o@Kzf$sG|_ZDGvYU-`B
z7(+QAFTU`kXTwAVxSmul*)JG-3_K_2(n`uK9CRe?P+;gKP|X+c5SjXnS^Z6*drBp6
zFZ<wRF;O*$ilH*RQ8+DE{+e)ZV}dwVf5e@6>jD?_X@P?ThbjG4T}>pJy={VSI{2t+
z^{jnA<AsJ@IJ}!2aY29|b9?am)EbDbqs(d~kMje=kYrXna_VhLFU}h7E{jEkx|Q8h
zwqJPR9~$%oE*lAP@FL2FIXa>n7RwbFSE1ayhsCJ43b-`C#=cM6EA@yO5GO-C@5}tX
zIY*9`tFN@;`KT@C=sq^`D<tD&xfWKQRd@N0Ylm$g+*{h|@x%Vns<ifq(Ry8pa$<Kg
z%U7|GzuRKIFn$4CQ>V3&p<>IU*nM9Pu{oq}<s+tDLGBSTv}+Z00fuEoMMXi=EEuMd
zL7|Y@II1S$j}eFkbLYN^EAf0&YlBfKj}wAdD`$E*pSytU88(M%#(Wyb#tpU4h#wH_
z1oP-MY)AhaN7P3pVi$G1qb97mv9@e6cN#NldpF^UCuQBbf?N2FNJ;0cP9<r-H!vqN
z&s8TMp~gPoU1tQg>!zf-64&g6QiM6?iKL^TsC*~DsSGJQC^;D+6{WZlVxapnWdHFe
z<+ET#-QT#iYW@`cCkPDV@LbenrDwSgR=L_(0JY`=hx&hTZmrylW1kKq-$MI6nW~qc
zjntor%t;BF&~X85Xf^W|?b<+Ps^CMu@`QV1^Dh7wWjF}bs#t!@mZC<N911<*K4o!<
z2W#{1^#R|JcDO&~0E(I^)$45e-4fkwrveRQQE75vS06AzX6~m?KXF9_WnttP01AhO
zan>><EtM<ys(4j1!~p)$R!lJnnI8R~7`6b*QP;|If-xUooFQ<@pwKoa@-%PHz_6!=
zhgU0lmKn{qT8O(qT`CRgQo=&f1N$(y$2!!^Vw;BoK^=>yi|T5=TG7VGf^a_p!=Re_
zxb$hihaEytPTL`v;^|^^x5UNSnb)|5&x#=Ym?a`f>CP3?#6$m#-6+x2M@0k9q-06p
zt0h?>KqWx6{f=4pv56MBf_WKoiOc+cy}PQAat%FG)zs02L~B>20gp=#0qpDHArukh
zK}2AReHa1Q;X=D)@Nzz}%}j#?*XxXZj%taffSW*|r%uy)%a8q!u-I}B;Noev!{y@8
z(H`E2TH&EA<fsz5OOwmXQ{nH{S;Du6ASl-c$U$iGE1Yj{RjrB+=xq|*G~YrcEL6(I
zI=$%@DB-$&U5ng&&6#p<w`v=F9K8^gBTzn|Oq2G$G|4hxCq$KMvPb#Ib@7+DlFXP&
zvGb6|0D=mApnlfPx+|=2rV5Gd>zvKe*zQv_jAvfh;wKd=0Ksj=wv`_C;psS(e>L|=
z_fCARx*X4_BNGN>=y-ZK@@GcN2Y5C@yCU%Hmaz2eI5zA47mGW|T0rVpJ-;9igR=*6
zFo7TkvmLW4gmVlgj~p&6g?OE<c`^idBAaFqJD3e!p-49kx+f(n#e@W~LuxhGd2}*<
z4Vw(s7c1S#fxm6Sy!?1veMlN!yxyS+z7N=TF>aq$&0^L<5tq*_j2Fq4!VZ8aCV><D
zA6o0U_3~LjhVb48pfDDL?;RcTUnssSL8DF^X)JAM>>o%~YE*rXEh{I!YrlmDs~<P)
zb@mlh)HW5|sy)QXjBYyl6)zp|VLJbN?R8_|0|C4pIpn6FK`(c`PChzH0B)O<sA_l5
zS+?M34<-&*=&}JVri3k0I~V!zQ)u;9+4&Oq6lbqxji$-wUcX|w#`TXIF(M*kmN=p{
zmgfybUQtWnpXu>9)i{%oJ~Q>1=^fVkX)!&gVQoqP!kOwW&F}AX2`%j-4qDF7`R0^C
zBWPw_mYl1sP5DN+SC7}DEWb9q0IWdh1SyZ}m#Y*^%smd{Gm0@aflwr}fpWDRO5O+E
zrDx$vbKdS51Q8Qd-N&Il>j(B+@fxQ;dm*I|0V<)zeJg{Aw~D3AS27ZXI}sri^2Rdp
z{Wnv*um_tbZ`y;re>ddeE~n;ja%=Z0gc$mZT1psN86JZkYp68krex$Vj^L|R@znRf
zR(`x9C-?q9^H(?az`?%YpH9z|I7Hr_g&Vm30xPQIx%j^M%0Gy0XLoaR`LUwcH{F7}
z&;irim&KQ;WA)$qKNx^^=AO1Ge#*=XUNmtz7DW@cL#Dd6NX~T3k-$k0Q_Dm<4aR2q
z+<<+o>NoNs3;x(HJWTu4Cx_~xs9Mpv3I&pAx(Vips01JoIbUZy@KO2k=1*f(lPD(3
z#4tOgeHX7;3})8jZA`kbo=SZAY4YE_cy2L@+eQMtb1RHg@jw=X@qZ?C1{K6&1QiR8
z%+evB1>`o-=%@EwAp|}>9oojXVbKK@%VxYdA$%zhT{`v)#>J~%?l50Q4t^ngtx;`m
z*RQdadqsg7i^t{PH=B1P3S*mf#~uHNyaQK%y-m{BiUi?91~K_VCi>&t^}zc6?_kB$
zNzc2~KJ^r^!0Jsz49;1%@Ki2Pz!Nf!eD7#*;hN|Z|3jdqM$Ee%@CP>f8uUN#wq%Fl
z%z%aI4Wpvk>~N)}`tLF>XN%%Yo1qPCl_1Vgs`ubQA1!S6MjceV@n!l7?-)TbXQ4;c
zfzFWuij<5nT)nu%y9qyLfj4`lb325|xAQ#1PMvHo8zOdEY`ms5C!0GoGM}Nz_51sn
zU|qow#|v)fLW9AKy53<ntxd{+OCSu!EjvJtJqnyn3T{s+A{4VXpHM#}l-2;`_xces
zSMTDJnY#$j&10`$+3|&H9b^d5A{cBNK}8;CQ=52rgbo#nSOlVI8$Lx;Fz>K-AqAv*
z5jw{Btk<GFPv?>#mxDu_Uv?Ioevn$jSDEDBBCxgDfHg<>a#jt&G0*C<B7P8K=*F0N
zV?Nnoq$T&;@K~8=*u2&D%b6`(dY&utq{3lnzwk3Bn!Q5wuGcuf^;XmqNHFL1d#ySA
z)VAHZ{E<q+$1~>La7`}+Bpcqbx^0Nbc#8CS_{Tk@9aV?4B|zj$8;v)#{gW|ijJri~
zcrdN?h^7++I;DjHddp|Nrdd(vhB~y$68t;7`LjAgi$%>)lijQ6){aFOExNEHRd6wq
zLonx$h@igt_;w69h%e0BVxKxK$NJT3HV}tSkazjC1=(2^&i}Ia3{LjLF%&Ug<rk8)
z<6ji5gWizCSAt3$cryAVLm|!$f&$=0<BCp`rLnlYffA0Kp_P%>=B;P;O77y=hp&XK
z(fW`VV!p(}G(by5$A@tEZ@9mDMfit5M4n&8-USu)E3SnE2M2S8sr<M2J}3-RHM<S$
z#`i%u-kn9eDX2KI_3~<S(WQcz4yEadZL{BD!N(3jH()^?JN-{;(!PH!-j0QV3z3_f
zu?ar-eX~n@=Jtv&mNUUTi~?K`>c0j=(_?S#X^~{CR;5@}-*2k5{87Ms6kn?_!RuqE
zyUG?kA6vMXco#?Y5YOHt(3D;PPsePw=W?GGUM&tJGqerQ3(&fL0dod%9ywh-9l>K`
zTKA-&mCAI6HX^+wM>q_~<i1S8S$5@=jh(tYf`NiaPpy_?H%AJCH+FSQ8CWKwPZHM0
zc-SF(taYvXv&v+D&N&War2dcjj9D4OPiqf|s-S`AO1ba5XmPiMarMH+as^F}4vz<^
zqDNuploLxXz5JOcatOI4S}|%5=$nbY12?z8-^<6)Dzn|C1!oZ@dVX@GrL$9bO;$0E
z82{p6KjFOlcX$cJjldeu4Oz9w|6IViXILZj4k7c(Cq}0+bBU>s$<pTuur|QkXXVcP
zlUnmk&U`k+TvQ0w=k7fd6;Ih6lSfUd8!(>$M%!5#&guugs!JFvm^9CpY;*me?cQ(3
zN)}j<uj{;$AlZ8d8v`m-nj6)>Ig1&+(l6MM<?iwl3BJOPLzP!rkEY(cjV>c7sL4H$
zMT|Wz{PzyT6|yd33dvROnbt`|GH<ONxEw0wF~Ylhi9SBf23bI+O>6l)`xUXtrhlom
zR5GK;J@A=#7dM5rp(4W@M%Y}~@GR$<=O_`Bc)|>Rhs_a(smhoP($9HXpENUc@6DQv
z#c~`rT{U@MPb(irSIJ6vJd(94$NO-YJoa;{c`i)xJ}b=8vQX_eT&l%il33)Osx)O>
zV?erFjHuF`7tX?i+5!F&f@+{!XLNXK-P@QvG{FG!`h?(Zqri&Qk=lf+r&hl@qNjcZ
zIC)12AIkbu(2;3D=%RT`#ciAGixvrSVU@`QM7l6<8H=B0=Orb=o3Qd<JRXL?-KmC!
z0+}pUjcctD%>N`SO}izCZI`{KR!$wk8$JEE`$;gUO|D%tg3uOI+267|9&e+pnvliL
ze8`LJEJRo$p}Dn^(_FSTc#~`P>6Q;PdD>9n;IQAtS?={RE17*tx>}WMf*oB)2I%H9
zGeaD@t^a?}=BV8T*W7Wksc{rG^bfByVxLG;vU!P|mgiw_`?UJf!)WOo6>4_-lIHLT
z9%t8)nb+P8M>>}1AT9_E(g(A`vVG`dy5Y#r6VK(p>9Mf4O*?$Q4u=6b)6|@wk^GS{
z@DZh9n+@eKXsVwe*c-;Dj&q~2+>aSkQnDw<=S`&FB~?FSIdyI>kxlqx_pD3Ynzl`t
zx#P)z*PBfk!^SScDhmm}>m1Ao<g5P0fe3}XrCNW#1ZxP5g0^dt-vree(L+(`dDRtq
zOlhncrB<+&aA3%yYW&XppKv5)6V1`^+gU+i<FjtJN4mdi<(M`P<MM!_erR5lW(w;c
z;za>)DaS3{S1INlmLMl6y{OnaMRxnjPqA25gUl4`gWvEM@VMe~=&YNsWGScF)q*G=
zJ`l15ug7!IG%Z@!D|nyS6F3i`(zkdAHM>j})X#pfm8R%4*12i}Gfc4xKYV<?*T-f6
zIh<`SgP<Cq3(uW=6;Amh?TKm0WqvlztolkECc?zbI>kZDbVsdizu$h52O8b)S7J_U
zV^+{K_t(?Dr-LZMg*Jj68Fj;lSc+(MOFh3(gPa^imHF?U|E5a*^8N=Bf@}E7K!5sH
z@!c&eR+VhHCXIFaxiAj93+nKff7i>?u8=iuey+qC*SN!E1@Y{%%t4hsLYPjzo4g{|
zEW>yu=##$4(GxOp9nDvDd6`nr7Xi>PniBe!sm^QoM5fs8h|ED*#z2#(H7v<MtKmWw
z<r>EgWxXJN0NRa3pD!dJofBVX+*e4~o^CRd!aDQJOS7W5Hwrl`_)?b00)xABsnw!w
zCM(}19T?_r9M#;B*6ml6{KrziQzE4(L0;-{O|bLK<69-USAV;8X5&hO)$5&O%pSNs
zM`!kSx9oWy36Y}AA3d-UxPO389RPJo{0V$3Gt7&4#FQpudczEE<}W^sgJe*mdhoP<
z;2Bu0&p^x4IMT+R|4wnq{y-;gsi}zlDEl=#_?-A5B~Eza<P;)spvU^g>qnH`<nJ_>
z@-0I5F}FwvG(aK8qFz5vr~MdQkPMYJn$koOa@d4svAVY||Ja~A(<s&YYtipFZ(H-O
z-tPh-iL;mNn>7ZY1pZa{oKvFC>BaY2c=qY@%8DhH%d9c%!*<{ZCpz6A#jxkF^e06l
z+93x^vhPadE~&I;&LrfYtLENu&0*#7X3dCCV<SIy9;d5FIX<Zu+K)N@jgy25!u2jk
z_Nz+CgK~z3R5`p<@x-w4Ey7|2e8=}i-oUyM%Pi6fkD0LY9Z-=vZ2q}WZ8gUfkdBT1
z+clUASwPD5A3+9_i|?jV3`{Y_a8fufqgv?EdGSXHU5PcBYoh05(Y9FTL4GAq+7_lQ
z7?OsW(HxBnCS^Q2C@}e*;+u6G&pinh)Q7kdXr0Nf5Omm7%W0taTJ08R?0gcQ!cbCk
z-yQhU;e#}H2!-4vG-Cig+O!&O8!~9N0mx2ev+>5Xx_l}T;jFgpA^ITqZQfS7ikkg+
zUL2Yt0^~fK3%fn?hGepa9NDw}UI*6e<*!Yn7<H$Xw&VIDnArOmALvKnRa>%m@Bolb
z5Qn<ZH*%7;SS+4RFUBn`yqZjUhwaZYj*}d8YbK*1)`rm3Hnxgd=VQE<X0Wp}&2_9-
z{-!VO`iLzw_mg3P8GL}sK%?|0LUtMI^9FR)I#ak+C2wBNQ6YIMb`j%t0++@K9Gn^O
z)93YSl~?G1hP>GJHT~U79Rpq<Oe~|-Y#+$v2>Du&I--QQ0rSd7>YbXW(1}F-*f2CW
z4*wXTEKu>t*EP${$_{>A^NqKrKK40rb3Eo1UY$^)L0v4qc+G=n;~+jsCdBVzN^)fQ
z$gD8~^TZ`$X-hU5rq%0Lpb}T{Rg(pxuUm6C)Zc7=T<u+;FfI@UWSlf|yTwF;UZj?K
z$4_$in#8<Uhe4Rfab%t%1Z5saev%c~7UmAQIU`g47cN}5b$bq9U|iGXPU59`NRK}g
z{?WeKrR;fA7E4ib$;!!{dEd)gISVAV=u+}8VNWC~r#908^}mC=2ZyVax+h9;Gvl+L
zM>@gqg0>JwwuzE{LQ}2TmY1`F%(`yIJE$K$-WI;WbHna9bCMw<*W#tAa#xofC$jkB
zj>{iyW|7c5wk?vjT+K2=s)yo$C7=g%gK@#b2ijKB;Z>F=ovlju8j`+F0*FsRYSTa&
zmrxU>3y`(WHP39;+ow10S|&VU+uuH^>GLcAejhuq*Vxwn602x-_BW3$FLOfxJV@S;
zAQzs7mDG+}LEwq)Ve-0r7ifat@0m+$yi|t>jlAn4Db%T%6@rFyT=Hu^91QUwG?j!o
z;%0o5*zYKgspYc%DD0$he$r*uIw;e<^*e!U?sV&JoX+y(-T;Vz;Z?3OaoW3Y>@DM2
zF7>24bvG52%z;#L?=|$HK@QrvYT1~%%{c`HjPp(Opy~uy`7P(j*TtJGmf0%5xP{#2
zrK<Hl2)Qc^!zYOGCHLKk8iIp)eiXur$a{$oabBkkVO$eWK(6!*fQO_I(A}1Z;qdZE
zfBP0v47j%ggU3iijw^zV0p{;{@F1VXGajqtrPcsQ){ope%~!hbN6=@Zl}3M=3Rr3A
zP-O4PH1CvLH}aJxs95Q@0)}6I0KBBZTBYtWRKhh<Av>Wp&Z|F=^cPZ41JP(*Wbxq4
zs#rc`+$`+Ig-0|Y+x;C#=fmnPGMMgPm~IUCx5P`e)Yq9i-&Eo?yjqj-2n32iJL0A8
z$(VcE3w92;U{km}PmhKUuhVoYJOA29@vz7$c!l-8LV?g9#8M(GJ}?W{ybp00|GwAh
zP#@_j8UaY>9R=m_GX}c>WW(3zN{2^((&U1fC3i#zRuZOk`dS3A&J9HMGj9F5I%N&@
zAsg}J5R3z4BFWpb<jhNVNJShJItXA>JM{tZ5SrQaBX(Vp#5M4Wet83`Bg*ccEAVPO
z{aNGm4>m~dWnnuL0GwQ?GQiJMHV<=t4jUTeLjFP-*JL_;?Rlj@R^y+6Hq<mi#GNi;
zVyHeWJ?goqi22j)n%fz58bdNkY99S0zW+NiB+C!5rql|(Z}rj#P(MQzQYscN^NI@S
zI5%MA<t}Pla{^wO9#I5DC=jOMu|_5j2RWVg%DMInv~%=mL;R?(4W9n(?3;;~%2e3*
zTn|Yfsu}~yMiF$@2@pxHxYwG?xB$D_>eB_8DE+q4pkKPL-0?udi0?s=I&c+^2_A0T
z?DM_Q+|ydH^WzT^(gNE_^EC0ddCR$5@j;ZO0`GWPW+JYme@sPcED*gly%+(Sl<J}A
z=ZU=?*F%%>w0Gf-h_bcBX3JtK+W!zr;}FiNuwNl64uh}0>=Q>+s40n_WQ9eTeC)jz
z^~C}cU+Xbne15F4m}qDUq8N;xez^n$pp4Tp!mA#ETcq`98R#8TVyRfJ`vg8P<sc4h
zYK?A90MpOlzCzQ`6luDnQWHUdG^DDMTsOHOqe0et%y{iiY~B&OKG^ir*#w#r;|)Pz
z3PKJPhY^&$Rd*V6ECD3JE#cl*o!r-|JvxI*o{_2TRGwv*u^)a{1@~gUdt5|V+i0rb
z;x2j}n?>69pwJ9`;k5cBz7GyNmh1B517;quO91vy5%F8+<Bt=W>m6D<)+U8W6)&59
zBEFxBoH*o9?r%x=aT#qV0Fk#WyCfgkbhc{-x+d}otc?@K{E$TCOh##Fj^?R8xa4#d
z?IQk#U4-CcS%h{XDxRbde~QX4U7q;N?~d4H2~jafnhQNLo}_A#(VI5ZZ|*${P`b^|
z<>=k@PUDY`>)qZrJBwGaceVdm*hpHyLBn}=zWb5J9ifo#?0L#u$qm*9B!o8eVWD#x
zN0%neAux3K<nVe};1I9)wd!fEbUTJ^kox{3M}wv%&!6j{@e|H(U&KwT>++8fc7a2M
z<#gZ0TO!fAL2L=5WA3%~b*QmvRt=TQNjH?;{+nlobZ+3)(ALrovDfL{aOPU`MuTYW
zayM9y41(a6?CabsG;{+l@=_DLDkxC9tOgzhI+lOem<SE$QJ(fI8IvqM&I^Ik^SIFK
zg$4_&=tb!}0;HE&cQF!Gg?u$dG)+z>7oe&6m?M-liG*xm=c45dLzV(_T1mP!Ji@d|
zBI1_PHdJmK`Z6$t5s`2-N6CfS0y2L;v#v0sIr%i9tqD@;&>p~i=jZsz^6p0V$oM{f
zxZf^`D(jKG-<?@XgLny;AA@hk!qxuJ3}h@t41iyEX4uYgqm-Cy=MpkBMczMtu{pP*
zCG9OIKK#SqN`C4PDwB7<n=v2eTNjRiO(;!nEYF`vqbP?up&VmDjX1uA$&0sqlvgCF
z<7G(Qr^oB%!mHLyX`2iiOeaK<wAj7iJ(LW*Dfw%a*~2$ir!O`t{+7LX=3c~iN>Xsy
zK`?|sl)uL_7RKhf>DQ!NF$_|?2gO$tu9>}AAmWw*e*@is2M`KwIVLs#HC6b4<N0W}
zDV}r?Y7(WR>GE4ky4AC#dKK67i+6n^T>*U!BUkRlJiqeaF@Ze`jCZ1|A<SQkhX7V9
z2KQJ*5~lX^TYIxdH&9(iR$mydG0EvD#q}u&x2Fp=<MBoA>{NmKc2FZH%{P15<BHlC
zO%OUT>&JB+`&af7&wW|GNqy!Fuebp3v|fARrW}i1_%^VFf%0o3hj*7#TUm-SjX+^e
z%^lhog<;DI(da-m>N-aOp(e;WHz^R#Ij?g%)*PDHOPvZ41ULSA4ka;2Zp8lVKI(cn
zC{d*^Xb+zd@(G5EyBo(z)4b?Qo+e|f@#6bBh`ELHRZs&h7!TZ^Eh19@J5r_%k%w>`
zsGaeUC-`uuZj!VNhR{jtml$`gMU$Cc+#9Wu=iQCeK{vD1t*(##BTTp%x*3fWbQI17
zIs1M(3)u9DVLP4VI(|cFxxjL5GMM3h_e~aIoo67BIi%$4_rpMx<IYY##WT8bK%`b^
zT8g<*-Og|GOxd*8<kVO1<$U!=O2Y6|M+x8blscT=y*(|!N#b6ss-?6FM5rmr{X1W(
zzC^KV`m@(zP6<Q`D~kT%xW+%YC<4O63q78_nl9bvd<ldf*?Rau;if<Qla`d8Djxf$
zaZ-sR{3GzhIcG3K*<S%mQi^}&v}<pzV+Ln4C`l-1bsSoqq#~c7cAd0yvUfPG&yYv9
z1KUyY_VVkxx1t-OF}_3F*4!|&{%6sXvd!v+Z4b1N+_q8a+Zl@_#!ra|E1?uLLq#w+
zSZB3~<4&j1ep3Q4?}Ry<kSwt*VtgU|v0MQmY-tA|s%w-A#TzfuKD+L7{)e?0Gb)ZR
z8hM#Pl>?Dy-d576%*gIY3jVF>*fIHaoV)waQm^l&G+HmkO6*&OW8V4Dt}h~cz2uiU
zrV<xpY`1K0*?DdC!oC=^wB>yb-FmzT8O_Y-<}WM)K8IIYz=Ibv%F?X51JM;H=%Gb%
z*5Rk`t}D8aj8fggeE*U14{OWhVv1B8*qc-D?9aFdi-Y6_Xl`IeGmovP9Z~c!?ySu1
z8KM5SBx%BeL%Sut9IKV3iS0FYHBy-q=zQr?>!Yox+pZnrHT%NN_SLn?*XEo;b}+>6
z<r|PuGG9R83BCr1xKMwrkGq}W@Vd06z~$7`nBMBA_Wc0<qsQ%Lufk2vI?I)$D)yZg
znkwm+)pgbx_X-H~WcJywH6!~fJ+?Xo7)18r>cE<ez7AHL@8=zFlf9>J*BQ+JtQ#l0
z^LAwlfj{)eRV}R6{(?eo{S+yN4~v{=;;*e+TP2%V{`0G{XNqmaR(NCNDwam!iCJ*D
zK#LhK*c*1*{YPszZQ0o$C8#TlhSL$gQ(sXl+38R_`T2c8OHL>Mo=qtPc-0Occf5Vf
z1^@KXVOiR-YMFc^Z^*lz*6UwGQ3le@jq2sZjy)mN6sO}C0LaWcR8)mne%k!4pv8-~
z=0-G?q?`|num{`L2)ySns2kK00)dGy%Za!g?xl*e8PTp*p0Mr|1_uUe6!>F~<RDb%
zYK#&Kqc*&cti_<JPA2G<OHx-W6rO1m0&1e0leJjINDBxxr5+L7#_VoI5Eq)|*M=*m
z6wFKKC+^+*y5?T1mtO~QZLp`{RkVqqAKlK$ad53iGwUGs;d?E2KKEqXWK?-VllIsd
z&m|O|H0vUXr-Inezjw$R6{G80&6RL4PpOngtV0*C)1L0=DRe16?Ck4E6A^jaVF+&y
zh|ys(Uep_h!hfgCJeJ$Jh8^G2w5$M*9ou32l&qOv2T~7_Gq}svLrVfzH$-zBFLXR|
zCuSB;z4+u_aI!7pPTR3X7_h{IBKwGNd1*9#{+<BY<bxmS(sSK4531<*)jU3ORsZY*
zF)XoBT@<5~Nm{!ir|Rr1xCyZpp_P^=dPsg4VAvVx^6~;olFwPW@@IFAAm}Z71xw#g
zm+8;jvq$<!JJK9qJ3wYnFh_M3rTD*C1b<%579?6Ib@^f2*oFP{X|Uq(d{7(Srr|lC
zP<~oAigV6yHEC@42|xBSmXIwsu(Mz(w)r3baM$HY2bZ97_-00;!!M%yE_}>zYT8wV
z1l_JmUW}<=3Y*<DP5c$){iSE%ZMmo5=E{GM_dnVm12Gi_%Ob=O9Yq0;-%9R|FKRXq
zT9>VSGQW|ZGT|=UI6U?jyBIZlfNRBDe;^rz$4DPH%M$}~29$eWwd4+(PlyZ2Auj}>
zkacohmco;=7sE^zTvC%mt!rN-rZl{aQlyr|;wh!eAQB2I>(QYf+(&&p=N@jZ`@y}s
zVjcFf^w)`gvaP+Ly9h`+t#N&y;-OM=BjxAu^o(EAFI`p)0o~uyyge*neV8GPI^cTI
z!-2|~{U?P`4Vdgpf$Xne!d8Z5kn_@j{kKY%;mzJ<(HQR@2PGOL2?yMZ<0SA;;YQCF
z+>*%bc~m4sejA!dm=pw+MKxOJuzNr2FeYZ<SYc)}sY-DS`}#J-ST!eXb@Cfsws+o{
zIo^$eJ{SEw)fZutYoqxrNKV-wU5r3EbWy^uUF)OOd2*!dW8STCk!2!`m4aG%Jk5vP
z^%RdNiTcmAgJBz$@Y~IdC8C^#T!fgolxl~jwaf1G7ddZPb8oL)j6qFT6tP@qtsWBQ
zR>C_?Bg3-9ksAhlRAHDHA5=D=biwcwX$yVR@Kj?A3xdMz6<jhn3^Qvdhn^FBDK}k4
zPTADN4t{$csk8O-`h$+uj#mA&T!wyS<K41-e@w@EVP;;%DlEWGBPD*9;_B$BA1hKc
ze(jQpsld18rJOT*hrn{Z35)%D?fqpiR2E7!JUM68pHyOPThqw<IJ(SOMn_E_tP8p5
zI?QN%zB<$gzyLy>ZJ6Necu{!=V<p2soL5G$ruVMASvZz@owI*OnW)q4-NL66xPYbu
z0f*0}IJgdOh?hofj)NILU-tiVOwjR3F2fMo3qxC(3xF4BO6~S?7Qni%UvZ<+sR{X#
z%iXfu7RBe>`^^P9w#`CG&C@l?G(OPOqyk@q7Bh`%Li0>i8RTEsph8V-zi-w^E3y>n
ziN{kM3QU_hQDb*`nTEyQHeUHrAl_o2OR<d}UJaU1X!@}kO?`FUo&wPH2mH1Em;>tt
zy29LN{{_+sJF0UA=|1MObW(x1Kga4PuW1h0v~AVygPan)<1??PKEKS6H~-EwnV-ku
za)$2f#<z}#9DVTmkaBm>5;uB)Lri$R@SdB_r<ZO0xLousq|*695vkYt`U&-ZaVQWE
z%J^6qD%i%WSM+bW>|)W7d*wheS7%KWD-X1NNuDd7%Y?};yX*3ux95HIidj?O8Y@@a
zlW}^JQ12;jGALcEEe^m5rO|3F>CEINzVf3<(z;VW<N6yJ-fG*NHtGZ#dU80tt4no?
ziey00fb-1a<-Zo1o;&2NZ@S3`p1<%DvQYimsSwA^0=JZF@1Eo7-;Ysq_f5}R<keuh
zCU&B_R>Rqp)*r}a0>5GQX+3A;LVI2P70F+ucY9{>JIz##GNN_-Z_NEbc`1bugkNX#
zht{fntiR(gH1}Bg9Ubpj;&f^at&;R8wS95d%l=Dp)uCW=Z)Vi%es0PZ!o(Yf>{+%+
zau-B)$O%|3pNhmeL)&%B=JR>@-lA>YSek0|eLiy9JI<vHZ6jCXWYfU%9KC$(H@Zam
zJM$N#r8(tDrf_Wx`KM&=_U<ulHJj>@f(!u^vCEhx?Q1a25CHH6dX*Z6W`7no=c_BR
zjh=*^#l1kso;U)YsJ4`y#jk<#jEH}doi&aT(I<fJ2(UhtTq>a_BqOqD@KrSDF1rV~
zTdA*yRXnpFLs~(9n!!qny-rl%j>r&N4bx8=(pkEZ*3GPB5W4>0`AgLw?L@dD&NuFG
zfq*TPUe^flpWoZ%8IpWGdtm4O%2PZ=VZIiEfj3c-reUY+1it|^6}P9VtfNBH@Z01f
zMAfGB6GMO7*PIOH5U&|fd;Md^a`Ly=DVxqZKA!jT-|+XiyQ%!Y!>%gyFAUuH7)ie^
zWaBGd<nE$&t^9=JW7V<73mve>4sFOXgdQ!By5L+JnCxviA3U#s4V^aJS-E^{rC(Q2
ztVz?uwdbPmQQepEyz5JX*v<#dWKqmgBIb-=DIc36c`a&@J7#sASz5_|+dbdy`Sq03
zBXUDWt6drhYjq$J^|AMpo_>kr`RUR3Tov=wb{@Yse)0?JFuFG>KkTCR{`))Mc6>eF
z3w;2L-l-^qY!G2_E2J36!$$_Ep;+?~iK*ITTB#2{dPjQrnjxTeE4Y2v@V>&>Wf-XQ
zvNDjT$Vk#><S;%@{DpZ+jI>-nt^4%#+$*(&EiDh1!Tl->k33H?Grab+>UnO&$Z_Y5
zaZ9oOG_Q)K4B7^<fKS7F-nor`&|hBZZmM9Uar+&|tVK$2`X55gu@Wy>2hhwYukRjn
z5lQJq*$dXck-gEvfwF?Toi?rsO~h25nFWIAes>NhOyyd_KW}w=Z7Q0+h)*6*`6uNJ
z`X#LY>Gtrd3#xKc#LmtAz8z|lt6wt&F(a`ra#OZ)M;!J(oZQPFM<6l`O_J6Ic63R)
z#Co;$xPB^zK5o)*;EUI>jcfiGq4mOzLq@ByCM_{akb<I+xp}cFznGeo?j?cSBr*4`
z##MUz++#NW+O_o>WcDd-8>eZiKV%TPEMFje{6))56<y_E(8J!eYn8pPmOpJ3`=TaU
z#v+Av6q9s{xqD+5sSoT6PM+ots?x{C@W8D39IAhv^K#xl$cJt*u5wt`{gFHMmSm5x
z(H<R(Z_?|yNf-2;nv=I;0XI#C_T8Ma3P{fy+c6nV6!GM=*5>Q-*j2UI1@fKZoH3=t
z9pG)zy+cjcj57)A6vHR4zYF~G8koUSAeG#rj&7nNaHOz+`*l*TCGD^bxhDlluneb?
z$`-8X<E&+;CRejm*p6Z*aXUFpBw4+wz#QrHl5K1D=ub`sX}>O1HH>t)pt_!v6w4iB
zB|D)De@H_tJV&b4KpvBzYkZ8cnpKLf#U2HOta8{Y$$aqhSO{<;(t53;LZ23Wx^J*~
z@@;%RYY8^Ff@K=n9BW_A*`0Q{fV<IWso(wVQIhO}1RNQ9O4mOJ()ZgK24OAzolCh}
ztH1-b%k8e1#+0-4#PX5FUle;t;Y`Id@+rd;@R{I%ESp<_YpK+~gYIIQ?eOK9-jh1@
z|8AlmjDw@dy`e^$s~>8evUQO?)PYwzYu9)#IY@_9IB-CIP2}pS2;v2~wNTW+TivSr
zc1Lw0+6<Hh;;d{qxvm|1lyp~6HfKHQuO3Twy*6|CHG{&LE6>TD7-1x~;#MO2n0drY
zXE`KJE7d0Jnj-i>&Fdh3=VUIa4KjGIFOrvl@C^4^;EHMQuFIGyF#s_fhAlPfw?p`N
zLAViX0{&v;fcH5_0cYYwbJ6t11Yzr&i{^ewhKb2oDUJESPhg|-IsV#}F{e*HaI?y5
zNJ+ibC=s`Dld14E>}K%ZYuGhxf2~jyc8$E8Pl;b?&r3nKSD_IhPIs=02=PgfIhUL~
z_P92_2)4TYzxkwBQ*)iqBq)0I<v;g#6mXqXA3fW}{-m8=7b%97>uk~~jQaFMOM3}<
z!-2N4z-9J(dWI*t_7!xc1dVZea!x2uLaZFKU3a64EQJ*@na)>S-4(;XTbwfEzf-#J
z#@ACDniy2Wu>38G_djOVU2~Ywg_Me%jqW(eVvZi29C%ibgL`lCAOnBVpl@Zh{@~=v
z-o{gLzX7u>=$R5ou%<bp<Q?mLRi|wT__jlK_wBlQhwZPUr6gW9uCCC=u<^tFUYQLY
zPuo{wSLHcoJ==Yjki-vO^+=zj6Ekyaj_^`20c+1Lv(a_f#We4VrIouxY6Y*J1>C_-
zI<#wR*<Aq+@FdQ^$V1n37j$eB?vlz}ebaXRG}H^r-G9$93#N=X_yRHEVg^UM;1|0y
zl_P%krHv2#oEjfvF{m%rU0lSnej$hFjGf0_mmLfZ2ngs6+AebJSVh$m{_IBRoGO3)
zvoLQd9~tYaZ<wp0oi6)OvPd<fns_I4scIkUsw;MUvpHsBmPcL@G0HPu0Vb=H&6axZ
z^IXF`X2NLkRh32SjzqN0B-t}TC?hM6nQ~gdwC^@|HOvt^-JA3G<NzECGW`SnK!dUQ
zy3QTr#<C`Nic-<X8`9jbTY)w7TcubSphyKPue|{xk=#!HJB3%XdVWrGgs_y$=PIg@
z%XpEO4~ycEy@))JVK*`_8(tM=vp8hkH}Mz03xD~SkA(O+o@4u8W&3N@hh@j&7KP0B
z;k{+NbB3~X7N)+5wCt$P(PZycLgk}3XupVqgyRn@QX(Vvh=$UvU~-#~hnCFO(D2jj
z?bC0N!ei<0Re^#1v6+Omx+?3E5a%660SM%aZ&ndb2T0Q>BVzKjA#@^h#y_+4PD9UB
zvE#C?3qzs(pA6L(^I_zrgfJUVFG2_VZ%}k79Qm7iaWM1eDO(%AP@UOoRMFI(3%*+9
zWZ1ZH3yEZ$^5TXt?gZ1x0eJB5A7{h%FgGqgF7E4m&DW<Cpecv!D`Qn0Hz@~da&oer
zhbJ6jJEDHtUF9n~CXc~1UVY&$YAIv{p9dq|DtbZ`t$m$e|E1~dy8PKt8^idZOu@gG
z-8l0pbG!`w2o`ese0UWUY}zL5B&i;Yx3csA=TS!ZdKS|sDwDW_Z#v0iWW+v-q@b!q
z-26;`>wUG4nj$ae&$-PTr12600m2!dy?HUmu=M9>J6Bhv{fkK^YlI$MZ@=r3uCt-Y
z{&oQO8bu~gOJ)vc<EtZ?nv^y*WZwpbGpIY}4JMo~Xz^rv%q-=obWwsb-xNI)+_u#R
zd!J%X!h!hm;P(>ulDPKaRTum0xb-u^a>33pe$6A#3@ym1GY7CPYkeJEj%mYiLH%uF
z<hlAQ{;uZ@rydQp#F89e;b5_3+&Ss>V(EOHzW5GaCAi6GUAnhoghiw(v&}y8>jyS|
zba-pLWdGkynXCDaklMk$bTR>_%^c7yiRV%uY2HTDHeoBLm)(BV2~AYvoI_eXx%-@-
zijN5AGy1O45jb*(oOCaKZ{~iz44_iF3h@mg_UM(si<oVk0rvWTg3-rB52Z@GK4dx$
zky!e!T<(=WgA_Onxp{f_`1a%9kL$iZEQ!5&g)e*fvB0oT6IwCP)|K2hPP#m}PJx5$
z)x`}F53}?;`xJO*EsG|*Q{YtQ;k8vtV)KrZ94<Yt5jIM4`y7$4)%NHM1KfbgPn5z4
z8=Vd%Wce8+5piy<NCmoe(ZBSn&G-xI8uG{VU0c<r8qQ4}JD0J`P^L(oV(`C5v%w)e
zX*hf`k1SrxXGERQgn{z!Iu$@jx6O0T$h1cyR=Igh`JW=SpNu}Evu<g;rX*g7smxH|
zZP7>nZBd6{z{%WSgmJc{5!P1eo-R|JaE$bG=8w-vv2(yz-iY0WByfuzKW;wLaS3*N
zd#JcW(b4&fHQzvNUwEhY^QaR4M5>^X+&AqXw|~!$!3E?^6Lu+4k3Lz-26x2rrZq)b
zTKn=>taCO~O!#0I1@@=x70HAN7By_{Tl2ugUlHnIbF(wA@|Z#mBF^hMe7Z>+so8YS
zudd42EGEoyIm;6^s0TN=Y$fjBFM1+#9V}{{wZ(5&|Hi_n^#vts)04s+X<}DTjrrq{
zsO_!7<M+FcOqBF1+#kz9mZQ%k%}G_Csv0)m4>^bAZ9Ue}W!8tN+p#yc+au+tz9{R8
z*PlM5mZZuR4jjoiH%MBI2Z_$)%<}=`o}+(Zws$$_8aN(O2QuaK_~ujPCp_w$9}Jwn
z9Jlhu`SQ*7hvFGuQSGB88-`qe8}5@<$HVnrJ0XQ5a)KaK?sx0r20yi#-3K3@xEyce
zebm!vOWqPF?!d!VW!XX{sm{)0p4$QcbI9(K5EHobZ2ap^)daEn!)lPN#vNG}^n91$
zN^%!4%1SQ6UU^z0rMiI{+0WhSbHn2Ez8B4lpHP2z&wrZ$XaXaTJ<I=fOQ8SU;sLC7
zD|Ed#Oh_3hh#pz@2|ACG$DXZL_S-yS66vJ*V2^($Y0kH;Asx1-wTaYy)rQ_*Cjzrq
zfnqFd@iQ>lMUHQdJsqx`(I-(tpG@^%vI+B@o~VoHTeq0xiiU6Gv|eb7c&VX5g<J#M
zZCzwN)2W~6RwIgGu5t1?5Y!I@ZLicxI+j+p+F3Gl=+)n4>2`F1ywjlpd5py&?`b`K
zw)!2uVKa#f8|Iv}mXeG%2JJ`wI`^JGSS6%r9r1mK;xDSN(ZkC>N(#DzQ^GQb(L}>|
zAF$0*z;0r-@s;p2_BWcC34Jy(?HkY?9I0|now^*ardfOSRL0R*JVX=eXyU&ECFml}
ziGRmiXpD!yr@YZfP42SV>cfv@d~5|Aq_L!cU&J)NFrd);Zv)?BnXTC~NrS<u-l@Ad
z&n{-1OO6SRD&w{C@m0UyrzOU5jPo>FiiJ2XEy+AUc4J=QEAPHBp*vsd{L_Alyz6)2
zXC~c=xk?WfrasOrm0o%_^;W9^I%7Vbeg0r#qO+L4SATQb@!|3Y^DsjH_urcTbOcI1
zO&WP7V$ikopON!M_<^il%YWxlnQEJwCqB7^?K-*x<8w>PHBr65qu;*pIrK@L^PFa_
zu&)G(&xW42@0-HBdDof=sR;P9LqTL=DVC(Vc;}Mi>1GN`M4aLyx@@|gXG_VaodfEy
z+9_6+z5m#eS_mVI3~7%tya$2rfng3A2H2K}Z9CPMyjS#7mJ@w`FzoB8a$k1TRXcy3
zOBrQo$6>wAFtx9|)qJANRs9?(jBsc8^7~E1FTmJ)Jdh{zWMu3WpN?e*aq-O;VOO)$
z4Gd@XO$?p-b`#C!kDfVn{&eU>)55GiTa%O&bLa{U8_l5$Z423Z3bTF}T=8M4pmRtj
zY50}+&p_C$DNc26UkW;7OAL<&&!ffd|2T80A?O2A+-V#)F5j{9UQelf+Z*}rH%8rW
z=67p0@t8qHT8+3)?Qqa`pG=E}7M48#OYb-N?Y{q{ajNK9Nwzn#H`&AvEIng{^K#Ov
z@G8ECMb&K^*sBD$&hu*TBFFPY<3tR60VW@@h}g%!CiBaFe%o7np1s-ATynNl`}GRR
z<a5+1`XV3jmNy(G6WCJN=?<v3VZGH*xP}v)y?^~J?rI&XNLCdymK`k?5x;2rluh28
z7Me|G9l<Ro`YZ!x73;Ab8D=ucrf2Jsj3+y(Qe=y2QciP-@wuw8gLJVY{S!lFbyt*`
zzq~u>q+pKUAPz8>laENN+!DCB_$xL_&kV29Ek9_o#HeHaNjG1~ry<j&5$@?~)B(^2
zYnL<}=hp>`aSq)4RVw|%T)faG7(i#U#69~nSVu5Oru;LYS>8m;+UjK3zt6ml@qL2z
zvc2cY{O+`jopHFql@trFvS!d~T%jUdG7tvQc&}Z5X5C1$e}5i!^-(sLfAVnQ)$z|F
zK!7f?g~iS*NwMJ3eg*Ev{7d(4`d@@Q7T`48btK}B=I!!n2uO7nofujD_j(~ExzI)Z
zA6Tee7+lfF&man`GRHm^W|f^f!eLJi!GDoGJ{nQz!wE9m1OyxAjr22~=+4f<waqmX
zDo*0-8%Ii}!segH&T%;ZlJVTHvl(XBVdP<P$kCcG!Asy}Jl0z!$1=FHB(D=r?PCoa
z_<Y3fXFHemdW;{Qb!4AmOUa0t)KBT=4DX)jb)J}|j7V#3Xj+W5qfXZs4IZDWD-KF!
zk$?H9a|+%#ciS_%eB#d0rwjzSclg=#k?gmza)zO~lR*wJ-p(R+Q8tyNDC!O~U7m+Z
zq8{89HQl5Af%tUiVQuE5pTT~&6WYVLU9pC9z3j6-A_5lhtMFMN(Vcj>`}KBX9r)#E
z^_rV&vqyFs9#}77WLbFAe>(}AjDCIfP2el_znM#yk8qNXkJ0A)!Y-D=FJqm*GuMwg
zdMGTzl=}uMKJF9T@+M%3qwb-sk#!=8;a5>3o%j5R(Nv}ba}b%>iwY5q>}H*;N65&&
z&L6xperEK`S(@zc(iB^=g@v<Ge7QK&0=_31OJ{!OdlNsl6}Cc0A^Y)HamN>GG8mTi
zsF~<HQDlR$H;k<oJmz}<FoG}59;H*q100`-vy<Q04CxQn@iwinZOihUFH5lOI(fUH
z&c7D%UA}XwB(pSf+>-y!%9JFd?wmDmjpUxnr<#9%f^r?Q7pkLqaHUKB#t_NP$Mc)q
zfl`YjOXh%s`Z0^if8%6cPzqgK9$BcGt$xJ5Xu;RHpT2_}{F@6|ou`RjFe4<W@HRhy
zNY?(SQ_tyCarufWpC!7~4~2_-NF?f$JDe3^ysfHW?OFS_aOrVAXW>lQkkngG%}>er
z=#mVLvNkT-u}<Cx`kY$qJWx;bi-uAsBhw>U9^ahS1-*eHo{`gd>7xTCX!H31?2r}e
zwOxH(Z!}%x_U@zcySBlXJIBNZdIaxd33mrR1CztA^sdM9d!HnK2L5=)s$L?ye>qI$
zDE*wDK7=zZ6H=Itqf@`b&@*4r`{E}{y?b80XOFbyonwqK>zGuy=R1s-ASl!^MH5Y=
zPGOxBq4&y*l_e-DWIo}TGYRpWy;317)Ac6#rCvWZoSz+IK-K{{0E~qBg~`2Fzr;2J
z>>{$Nx_&~;WAWTo6IMn}wi(;9k2)c18`2)^M`Yh86Co^n<Cirij?ZO41KaXhy(M3*
zc+WN;QC;!RfqlmV4|Vxx>4&9xlSGD1%6MAAW>3APwdB4C3zXkGvR?9@&~xZydaY{D
zgb<48@d7n_cw`$_&(OUq^sT{2!u#5*Lk~JhKM+K<T#R|Ui1J9}x%M^L-b7_TBcWON
zTKVW1!|Bq{0D*tPiEu*ROf`(fB8a`w(GV7(yfSm*A^3Jkb>OBK2WD}zQ}^wwjoM>J
zYOc&Y^q41&zlg1n8FpG&>AC{Z#Mq0Zaz4yqfBBvUe$>!E5+rJ5=&mtfw$wwSb*xe@
zY!<fhBg^64<-hQwGJ4cW;EpJ=)mYYc&ZdTe!-U26xvR-pN2C?I*-PNdZwkQKzbKtl
zlg-Sh@{KdVVWOEgJn7-dOJMC=f*C6-cd`IvJd}WCm&8sw^b?*j)A)T(+<%sGGu}eV
zyyH`$t{h!8yritL0L;<8|KEhd*(sNt&~*N;ptRh_N2p}0=-I55+_<aXTi|JQl#VQG
zJ}&G^I78ih#c%QijHqVY$jNVK`=kX@;4fs4s=$j4?8*;DO9lGE=^1eE)0q@y_=ZoB
z>p)NF?DVfE19|+Ko9~eza|LtEd|(jrzV8)Fex?Lf)0y*tdRQNcpE+=<V@1kU55}uc
z2Dc0_RVBuT2F~BArg7tbN1iduCP|OfMsyoOt(ZKHuUWbpY=sN*qUHB~VrO|){;aSw
zW^nuRV`zN@H$B^OD$Z9bI(+KLw!JwFm|q`9lBb(_IpfWnHzqtF;Qc8@zb*X_$n;FD
z{h4+SXv6rFz@Y22=fBDCbq<lBv@+SPXs>1)a7yBW0+D>rbcym<#+%tqWT`BNsa+J(
zV~@qrz7LXIM?L=R2^w|7SlBd=`=Y*w^zEhYT~RCSy?l68gIK1Ge5bsSP9rdg>0I-&
zp9YloKLnLn28&8EFRg!tVj>?|_xj+FkdVQk?PaJMgn$Jwl#%rbwsLzDc6~S}fQA1`
zCrowi0(=>JGx4pcu%A;|-o4o^CEKr%wMBiMBWZG0fhz4ch5*8be$t>e|LD;kc{Pal
zm>#;3$`k*(q7?OogW>WLDMLB-Q#D;Z(usW?JRD{P3Qo8-dAFX~Im_3|A*|s+L@V8b
zuNvKJ;PC>n4?LDWOUYvn9E*Q{k*^m<!MgQ6m43k0=`vy)V5*xZHZ3`nrOe`*tfL5u
z9E+zbPo%)t)$OTm;*F8QCHtIxW9YW8pE3-#K6PwoK`>C0R;^cs*jtT*yZOi4^7<Vo
z_<Rlowx_pi6QO^k`0#e`9=oEs!ro`o`glP^eNThd!Ay~9jLi68?3Ty>Tn6(aEf=hh
zH_O!sI2vrV?pX;H$pYj}O0A+?c9DE?j$wlDU4yM7Hn=gg3q1pj^<qa9s;Y}27xtnd
zdo5ChZ4{{&Vn4GZ&F~;2_&)4-KO$9BtD#bGZ%w>!%_jAA)92&NP_{=N%g~ztOjjs>
zeP*}=R`pWpnyVgA*(!20*_;ua-G27%8GFu*KI?)*DeM1?-od*mNm<%XOOH?Tai*xl
zQb96tMT&h3pwKYlr_WRF?Z8d3y4bhHeTf<eB{5_@?VBM6%ss01VJy!0fYIuP{*d}G
z*fHg!A$eY^M3>BwZ35JUfpXw%<u*1$=yRzw72JV{-0q2)|BtG#k7s)Q|9{WTO1bHD
z>!#A_PN|SmCJo2wjzSpaz9XlyiY%3zF`PK3k5Wmx!8j#svF>(4%;X$GnaV6G)2EpG
z2{Se~-`BM{pWoy6{;O<zzpwXoy{^~I^Ywhat`(9Y7091JK>fz~iEwtKGoq9V#uSIL
z7ft)>73GY)Ih*S>xudF?A9aP2ESynk9{mBWiAI?Hbv!HhvagCR&2e9q2;r;Et<WXx
zeEHfd=|8czn&h<`Nq6X)D-e#Vo6>v^$5~?x$*svIRez5D)wk>>BUFY6`_8hOjenJ?
zER)u~LJ_=?u&1<(Gpe}X9;<22-k<n2Gt$=#y_SiQ_$f(+1JXP4KDXCz%dErY#_47^
zIJb);oL-m+s)W5glMm{mNS;bB;CIr;#wTAixL2)-MB-l_rY^3SYUn0ljkL6quZFRu
z>5m;Nc_0zK<TjD+Uke|&mpJ8`v04L2pu&nq*pJsIG-)c-u*oUK@>LOzx?IOCv|U7s
z{~0y0D(|&Gk{}8nC?!z$n=Nk<?2kQ@Mr~k0H};zPjLbk$zokeW1H{vH)(xo!<qan5
za2-y2D!em(|19~q{OtIPWe&pq0&DvB4Zi%jXC*OPp7nk!Q3fwRFvC_XL&9)f&B*gd
z8m`Mhf7gQSGN9`h5@v&=Fh?2bBPU35r`@&hDU3wZvF)x*HgyB%f@h+sPQPxI(xck%
z7JJlXzRQ8M-jE{nEn8|j<8Ci`+44|g3va*8_d_c0)z?;aJMEm_``H?_O+=>=Zg5dL
zBg(=v$3Jm2YHvVTT4_J0e!THe!ylGTDs`+)6eeqDMQLjkpFK0!dORf!x2UD-S)8VA
zP)Q1D2NxYJ)tq?VR;wxXWiG<O!EEeXw3Yt|dAa4=i04UMEZrq2c8$X3P56y&maUch
z11Yynf~v)7N9rQN{H1%GpkG}6gH~H!aJU=%nJY&fYUWRu<BwK_2H8+fQmeYj0PM<}
zN1j)$zp``IzDG0)XpiEoV4Od^ho*PD$s1Dp{CB2L74UP=?+kvGXL!GDJEKUF^;LRi
zsX}+ymL!}o$imz`B)0B3SfghIU9s!E<l}%S!oUw~8QUs&JR!2kKDR*}ju=RV>urD4
zg_5Udy%rgN_E6Tt&~iJd1uW4!y=TA1&V`{C-%y5oqR>O8YA(CjkydeN$GQl=e*GFq
zP8X(kk$snkmilb3*1wg(IoE!+#?SP138;@=M`3VU<O=y^Y;p+71)>|KIej{}nOs7L
z^;rdsQhyc1{fe<&xrk$v(c>lFPKq(~RqoCk25Ro7u-RA{X;ZcdPea>}793`l8J`CQ
z^5<Uol3BQVXSe_chp~lB53Af{t-5^qL~_wN`fvUPPk7gPbcEK@VtdF5ZleV=(!WN&
zlFz&5CO&Scp3ms5w5_S*u&pTYF8LwAIWPKN;YOX-yP%~w-mgdy20r$eXkz5@z4UWw
zF{i*#{Ta;r8b1zxJEfmvEY0bwe@XZz9){W%PA@w-PPC=1r_|=Z@EU?S$BUF_p17FI
z+kbU2wy0x@FR^=5pqyvS#}z-?5h?ZKD$+Tl;((xruaB;-yZqkeAiFm~G4P5Yn?G6d
zu@imFup$&`G+nD~63nSCe%3VoVGoYMn7`uaBgNm(_m>h>pX3y&t$}kCIX(9%^N`<}
z9mVg7XEb@^G(VTbr*081Xjo5il3O_+Js!4mFyq`|*@G*oCGXRxB06G|X7u5NO*|EA
z4`Wz_b*?bF?1KfS|I`Ly8yr?YEuKDAh+?u62WW3flG$;0j^zJU;o46CucF;mrmuN}
zx>gIBM{Y^bynT4CdpfQae?pZ{(wj#*WsCf!!4N^w-TD<_s<k=O+PrJ?pAi?s+)fa!
zruSyk2oN1NmIn`~D#z=|s%crmIA{2$5aA^3FR#5HZWEn5DRLBe6lU3oQhZBBCU=Dn
zol&SpY}%Qvii4Qm{i^D1s1fdu+?3b2S0(h0rq2P@K1GskRR2+5sf@7xEaEvdmztiX
z1>UFSJA82%^^E3#h&-;FviG00`B(n}FVzJw*2Y+kzq@)luO2@M`6YCxi!VFQhmYTc
zv=FPb&9LY%9hJcA#HtaA$KBNLB(%-%P>hOyOUc?XG!FK*G`;k-vBa*^O||K!{L18y
z4eLeM2uli0D4Up>{>l9re+YYJC6n(!)Z6H#O4olZmZ5I+aZoKg{u7I#{#Ij5%`&Cz
z*fPAaC~DTnntflcVL`?UFVDqZXpq?_KvrF}?HSIjpsJR;n#N$L;|pRTolc8<moAV@
zg$h)!#l_Z(h7jedUmAFA7$v#D<uY>@`+CkF%Q+30i!85h;yx;$d+v`bE@BNB9rIr&
z|9Nl`L8lyADTpICiEdDmp#%ZV=PPklMt%J|=kN^2a^W|TSQ@I-HQedcggE1b4bWTI
z#g2{hAaD<wVWE12S59%wN;3-KMfs#?!fVb~wRLWGX_INXA6LNWX$oXg7S;B!5{a<o
z^QOblw&^f(;54HPwvDN`@1%Za5pP)Sl?&i=7iX2!7X$V0Yt$yse0c61L?7~*zcstU
zz(iawui$~(I^xM5-^cHaQPKS87%$T`$qKoa(>u(T+TV4%!>c*cPG%GJD{Iq9{Uzla
z#tE4_+CT87w<_>#S}frc*1rQ(=Ey<qz0A*fjAhXPb+%Q#3y_bDj5OWZT_K(A@*qB#
zN8)FBgAj0xRaeLz;LJ|2FQ*}HB@Pl*ja`qank^)$BtS%^kdw*tDRB9+a@hjvJZfBH
zrjHI<3@-zr=wN`kkMIj2yn#y9#$nkN40Rfbm7<>C^Hg4cJ1=wqt7laI<s5suJH5a)
zW%61|{69=S<yb>*#qgP#6hLF^xRAmq-yW`n-3Cob+pTfg&a$w16h<&E7PC+LDXnTn
zkh<UWYTMhydeJk7A03G{_Up%B<}Wzo%Oz}_RRaBq4I*F+l3H~196nQfPBif_`)j3{
zoB?mcCvdbRq?MFFLT?4LHAN?_Bl0R1_ZXr`fwT?ISV?N2qQScwx##-5idkj5-6_*M
zeL_|KY}!5dN^wY7+cnPZoU#0n2R##mIU^PaKbDB9tVEWV6@YfTe^{tX(7iRYR#C}~
zxPVWLMNS>ur*)Ht0ViE8%K)b+EXS|kFebZ;1<5wbH<H$&o*s!NszO=t|MO}!3-XjW
zahP^NYJpWd$V{Gga<3OVavcm__vuKBB;4f!oBBik#!8$$VY`(_=P^JL4v=L}a^@yU
z<<TbmH>kBQspleiVNC)4zf)d~AM@wg*=(Svq-0e*;YAK$9~=!@#Y|Ji<g&B<pTyf^
zrt*}PGG)xV%CoaaVb&R+(>3R-<*&JDjDiU-GL%XSy~?G<6vr<^D5emQ<MKx{B#^r?
zR0KO<z-@=Y7Jia=#jHvAfI3Z-DiZl-ntC*PKr-Wr{}Y%2NH4d~XYhM<`_N_&U%i6O
zlK;&g-IaYkdf<J?j=$_ix~tki-8J_}B3@oyicilyO{9Zckv+|jl((K&D=rF5=zZ$E
zV>IG+^nr#Q?@YHWEZ`B@ArD%|_Y^ImX3Ad%EsLrsent@{G0IK<m87EJS+lht;JOl)
za!1&4BQB8JS$0-s(0TpYeX3$8<bJlTO?9NqP`fbuk2^!-58rXU`y;@}wjWZaQ(}x1
zK{IR-J6)>Yy6+^#lw?xQ)7(gU6E`A%76ih}bXAN9Bs<rU3>Reh6P=a5pCV|vQ`B8C
z)mt&eWD8<?N6xE!CRoop2)sm0a?4qZOUTflJvXGoXY-^v$Z6l}l$|7g#DA&u>6exn
zy*rORq|j}Y$NLS!TfAT2B$%gao6fMfLfYA`)-DaB#zsa)1{bmwsKo)_R{^sw0t0|8
zDhlSYVNl}QDO>*pqcD>YGep`RB<YQ8yJ7W5(+l@rGG<9i%Z`$NfWN!!VaI)m-{0Q*
zgm(e;@EMgq9v3PMWOLuKpXm3m*u>SUz`n}P3Kq9e`;G9rjFh4m-E$wA{(bCQSs-wo
zybS!NepX4fk&nNc_GfeMzK+vt6+hiT$`Z&umJIUg-fW()zmkU3mZrrsHt+*Pm1aVE
zp+lNGsM=0JU?{s;kxPvfNBF1N=}<S^2=HMn`YSoy3<3)?YpTv1hkHmpagPPLyi&>_
zA!%guhwZpT6M@(`!Ak2P2o-J0(a@L8$qw$lNUV+iPYxa*{VjHiH$846jG3;O5=_e=
z?XNMGHzM5uxdfVw@fyWep}t~%WsH2Z-<v#oF|pH&vR_AjH1XWjHQqlZSu94w60zL+
zmT_2l$5#LHtZvrWjG;(kEZv{9Vk1_1H9=;azzoYK^%Hftf^PvtTwUOX)ds;|7Ff7P
z4b*Xw6=N(nG&J<?6?3R<3z+4m@KV_1mS>T_1N3qFO}lo$7(-&y!UI5vHsvsHFTe5~
zEXY|62g6pv+s9A*$CWP;QZY-mN-9-5H+vp@QC76*#u%x4!}5<Y0Qn<ICRwJE*!ix4
z_o>1twe~)Hs$uT<!;E9!hR>Mjz*WeXmj#(5i~eF23}q~*cHbN88UKaN^PX85;*FQ@
zjpz3T-djK;_teMLKW2Ou;);_OzFxgIqHC5o!g>EHOCL9)KS_UZIHF)Q1Dxo2R6~sL
zKg1({libjrAb1nAFO3pTV=)%bc0ZMk=()+2F0u|rf2!@0p$IxST+K3nqp0<I6Lgj`
z5n4#_k0&~0?<2BM^uvdqV>?uQnxm+i1>4z3#y(MwOTWL)`2?EIaZ2odcYD?_>v2|8
zrX!p1l3JTm%j<tiozHJektw^J4(6Qv04H92I$M6z4JZf*nT}&rlCZKg`OEk3-@B2k
zY-r9opB34e#npMOBJw)bzreBo{z@K$-y}IOdW2DKs92zDk^kg0?Y!@dn)(MG%$>3p
zbO)=Z!5al=n%kDLBw;mm0pEJ2mxpSM`yeV8xo{_hvJk|ZtPVhpCA{U*DMqPJ7d^vt
zELsKQhPvPNKXH1#?RI6d#f1lN#NPHvE(Mo@eQrFmj}g}~^mD_#y5UHnY#lM`%E3*~
z^Gua4;T{yRA<9V3=}^^1Xa2wr{y;D0vuDl;*F$dX!Hz?EnmI0sWJYtlcRv9j-E6Vy
zmqd;by@!ue{3`Q#zpgWH_gi6PwvB2qHk=hKae`0G@CB)kv7cPrTTJ7~uS8#FoU2sY
z<qV@;vUm!|J!qbF@*li45WG6KRWa3m+X<Zg^W}Kjk2}g5er<SM>gIH9>K2Em0fI*H
zAy;gj-T}LQr64tDYMIY}I1hQ_eWItWqKE&!1atADTPjk`-&JUiuPVLLUgm$|Ejx+}
zP2M`+aN9zi19Kjf*@S8F=hT_IR3hGC*Nd&(5y|8lah%ex;h`3t&52udW5H&6Z_(Az
zr8i|t!)@d)X;z4St((xNbFAXR@x&+7TK?A}Kd7zsY}sR-{<?CnKcyjcrQnagryh5c
zP@HEUTG(E}Vwf8)%w|;eo>cke%lWW@%_H1z8U9PNC_aw-lC?_xZ_FIZXnS5tbCb6+
zCNktRH4@K)D~aI20?`DrJriA#T6v+5RJh0P4+bH+P}C{w<(H~djEfV46X^XQ)fiPE
z#osE@zXSiI$-?}0+=w$P2;2(eXL7V;<G;&LxYp$-6QVMzeyP3NDWO`!`pLq;e#!wM
zI_;c2I%PXuMcgA#hcAXwNu91AL)40|ygry87MpF5B+9h!P9fajRRS6Oe94uJr|x$R
z?%IxWcqw1VG7s@pUhQM)X6Xx-($Hs|gfe*`pRHf&t{S?JKd^;A@D&?z99VV2y2xYp
z@U^Y`r^mj3-6J3$83P~QGim@Pta_E!!4ot<wyevxbCD<2zLSu*td2I(bQ?J!026h&
zvaK?L!xdt+I#6RjaYA}=v#z5pchCIj3{FQ%{8b%!>>D1MfnKJE^U*9_ZqQMVeSGQ{
zPi>w(gMZMl_mTgDUP9rxkYy!WU0K_Z;_TKzC}#=c0%K$1Q!A95<j^O<>R&ZDAVB%9
zv(r=;&DwxvR2^9rCb&ot-HfZ66I6WP_~Nc3u`uQ?=GL+i5`XtfcxLvF0#9A9$(Jxe
z&mUa*A)A#DqfK4tqSzj73XtxY%@=0%khr$ELu`%1kvd5kAE8EGqx}`nbb35TvK_?H
zR-NB-3GhxIbXeqHztV5|8#o4pEsbhBbBZL&{3gy54<8-7!;#c4laKd*6qSK{4v*&6
z0s-#W*WT~nzd!C}V`fWBD(+V@#0>ReoJ*C9>J3V)=envj9mB32giX8(ds9^1)$?jR
z#9!Jy+~&!?26ze9)zBDk_Qb{<D$~J>%GP1{2(?DQE2j;h7=!#ZGu1U{TFvP(peTix
zjaI<oY`0>Y5^O5+-=(9Jg#8WaxbbkctTjo~<+Bnc&5o^Jd`kgXNuCc_(cd#D&gMGJ
z!=KJDGU<|jS!f@RHz_77l@HzHROnlj*gRW~#}e;_%-Ues-HS8Es?3_Opo``{_H31V
zRXCU-%(pfw%0Kta@pbam=+!%dDgxKp>FTMWZZV3HyC&U?3!olTusM8$A5>HAZ4B+N
z57SZGgXW$-oh<L3-~}r}gQ@*rKII2Pxd~{gpGlqG9(N7qa4ktZXI(M?Ubg>LW@`Me
zQO+jjhKL??)5r2t{;Z2~@mD}z7_ra`7ZFe0`U(rx=4%Hm%f~{s?6q7xV5UhQ{R>8y
z0@qX18UNF$<kNQfk9>n9cn+<%me8j#3pXjPAl1!@ZZ;}|BtHI50jg}I_5N^c!daG{
zzfqMlT(Z1%oL4Y^=~Y%<9nVd{mLG9WldS|;D3pN6X)a-DK(w8Ak{!OM;SUD?>-pR|
zf|oyQiMU)8y)ggvOrTFkhxrup4UXObWiob&gEwDVM=f<YqK8#ol{bKtoh!ddaZ!p#
z?!2}1goVZ+vI^KaygX63@16^rau$Z(E(;rogwuwE60e5!l9fG+aZ?2jlM(KEw?LS%
zG>!dJ7F0*~v$~pQ&AjlQW83e(dBA4rwA*AXNQ>}<Z$KuP|Hr7sEiKHrjvD?y#v<oW
z8NsnZ@C86le<d%iy`FLEEVM6bU&_!dGJiCOW|GT2rU+o_2_rji{n5(U`CMZ@%vRAR
z)o(J6%`_hu0{7=$SZ!7!mM^QRE0MbN{ERge*4aJ^Ry_bBoPAx5*@l4sZEXSk6wM%E
z#+g?fAv(tGqqt_RG6U5*-g$7@=;qc6<VsEDk7f~@sRo#3-edK(qKG`G&Au<rYyDLJ
ziZG;HJ|a@ug5H`adWx})fTIN8q7SpR8czNu3Afjz=`+;+LC0btc}8ccH8%*lKBaw{
z*18W7%A%L<5zE09OAG}33}{i})n+XmfL^h|YnolUbtPo{myj8q_taQ0863VJK6A0d
zTc!lj#yr;q=*u=zAB_8?NW|OZe?aU6Cwfv^K#4yWdWzkUi}&L`n^Qe|<tFlHz<bLz
z{IyH8k|0LyR{@^WP++Y3oR!1JGF}Lp1-*NT(@A1Nr$(ldHs4<F>f+atZKXZ;D==R9
zbF%)0x@{i<Tu+whOq4r^Y4G(&nVMms1AFQNN;qO39LWu11hsJN4ed(K-kp2dF)B22
z!M3aW6r!8qJgSate_ISsJWhDKNADAA5F1i48lEl9U53YJya*Y4l(Kz#k}ro03KsA?
zqsgG<yc#6$IH+JcLO$7e)GtC((m{X&O&Ak?@~P;UG0It|h~-MV{j^~5U@Gn-7J7+e
zn7=eETg8^HnLBS?LNasblE#p6YxfgsdjoFs;MSohs5<i<?%AhF324BwFwVbXeFB+$
zreF_b*7-f;zVo1MxPsgPC=&_P^0@fpD*_LtQGWG%i_(SULTL0&C8zf;a54;W*BgYC
zB^KzO+qO;)>zEYQreF(RK9+S}7SC|VoKi^5Vj_zrSqO98uz_c<n3m9OeEjmDRVNkM
z%DGs$yPi*)s^Ia`XDTj^b1=q|y5#c8%=&zfc@#?;8m-Sz+h^(SZ#{ApZb5?2tb~=Y
z)G4hrMF~}5i$<t)a8>Lcz{ib7Ez@se9#y^;<%IlU614oFN+a|2?cy5{5X8tO@3cjB
zq}~@u+&34Pi=TKcTLQ@cZDBb8!-bbwSFA*)$fXB<{02mwwGc%d?Dr{TRXq36VU&s8
zm1Ho>yC;7medL1C=gzy_pS_57XZ}Wa2`3H9%hBBS!0=yqoON+G^1Dy$?`HNLb0_P<
zDquxMmAdu>&o{QW$Xb8T(!FFX#mPUz-$P5wq(+DM6Z7*;MLRTpG9*tQZ@37Ft0H!<
z)@xq4WKeW^8oB&3X6w?P8P^E2eXh;3tB%pWR`O`P#N9%{>yLXcky!r~Jm4f$>g$Fg
zFRrtGimY@q-nIhfnPfqmPdmdPx;z)#IW=7u6&qg+)Q6GLC@9qm+<gS|0iSM%Kpz8b
zgEgg2Ue4h!P((pJ3vf=PTu!xQ8GLXhJ%F#aR$AUxXiNzxPMpZ>7`A>koaq)s4~1%7
z?-wyH+#?uuO3EDZ2F=RVCC$};CZ3Ib4cEQ(Fh&-lkyuFFvxf4Ol4Wy_l=WUt6x&q(
zH)Cci_AySMT)wzLmCJHpfA-y&y6u|vx(u&*?bPdz=rnEc)|0IbJ@p15?>pije^C~^
zPRj;{#2wcP=k6Z0>r28i9*9+=wS%USz>GZ~c{AIiyeL?{q3JqLf{HrHQ*G~fx3XzP
zqjPoS;pzZOD8QYRYL1Q?76_d;EN{>9?kE}9N#Jcax=G2OZ|ke1Ik)#c$vJ##Ri0S2
z7}u3#e9=ie0GFR1oby`2yqFVc$qMFC6xmQ#UDPh2^ozP0TJ|JGy#u))!2hX!DZk@W
zq>`{;yNfCm6Wt5_kS}4fUS5XsT3%!Y&QUI<whOgw1Q7a-KOB*qMsO6X5+}IXOO}PN
zifce)o+jNk*fu)30@Ga(SDRPM+aTJ*SFoX`iLv_sF^(1Ym{lte`3c?3pTiSw$eu;J
zR2sZjmBz~VTer&cCPg*{m9L1qqP_<v&=a*S7Ldm>`@EH~f_BiODVLtcr#kN>v}jrn
zga01tRFLUQA3=0wFR~&jB&jIU8aosp8kV(Xlm}oXG`A^bH!FiN_NaL3)5G_iHSnXR
zgB?3#)=4;!&r(|(d!Uoyem5y%o#;0OZK@*A|MmA}OO&qNKkm>{a;rIx>&@48S^7%F
z<?x&!CHz&tp0U#ubeo(t<RcOTbTRg@%z0$8e5HI*d^|jZh7@OWk9uIt>VzRC{}eWI
zAWd#BiSvNZ1{W20J7x!uQnmIc%AVLDs*%8Jy4T&b-)zN-@W)i-=7m%|4?(LTaqq5J
zCMW)YY2N038@>aqNSGyU!ge18-!$w1zu{vRl=MT|JY8u7if^5qCoOcDTBYIC{0gS1
zu`u|e`TA%1=7ZOv3}z|Rrvobm44;QD>x(XbbUhkteQnI{o`$5AazjL-f(&3{w!^a9
zaRKmIa_te9FiuZWcb>6)!-!+UnfbS@M{yrprlSCxij-6R%LM7(g)F2Lx`dpnJ+CwI
zG$3Y0>k(K5H@6IBjl_{|Q-(?7KJJ@GDW4xJYgm1J*guUd^+6O(h%5kUdm&cgitpP#
z@T$Bh-Pcgy^(E78GswYX^Y({g@ed@a5gO%iJ+XSmeRKHyMd3Uxrv2C@=SXvcCHvqT
zp>ynN&7yBFK&`V1wZuzBPhF==QQe8;x%Oyz@8c3=Bwct)|JZVuECZMAND#GC)Ykg%
zkzJNWww+~U7->^apJekAtgf7y8S5t?5&Y#L#^eeI-vVnioO2iAjT4PUzjUda`FOJN
zvAj2Eu#AkF1~?4|s7Y{@rcoM^&kH8sN$=1T(CgqH2kd3=dF}?kBgw5uE!~QS+mci-
zDs%Ad$qfpfMB!ce&A_F8^r!xZPdP!R3Ud)tIbNAdu&Qq{BLtOH)&?7~vB<dw9e&fC
zI8AtH&KrTI!zH~Ip%F@3%-lx{Thx-@gsIz-3(FNOb3CTYghHV8#V!{r3HMzdI#gbx
z0?)2}RfPY5oK~6<;z@1-zXeh267+uC{LJd5+FFs1K5aPPb^FcU3Mwj9Gp>_MVe)~;
z0F0^cd&n$kbK)Jt@~T)~7Os1hIAYgnNYKP1)^3sDY;wj6p!@fK1Ui2Tl8g)fX;OI(
zUtN`}dw<|?K-hNkq+GAIS5ZUOw<t9^L_?46B~YgZ9Uq`Q7Z|4pJ-a)P#O=gJZmW7@
ziR&&d=HtyT(7$Ccc8bPPgWtBN{}bOI;rp~<M{ac^egN@l$l>_+ICr-W&cVXvuzhG}
z>H_{-mEEKyPW8S`wBnBW%zvm53VtTQupPshl)KT(<RYRxN&}yNZeZAtTTXR^yXKq9
zuTed78|GpccG|p($Q%CcG$Krp!?WZ>^AW7WxWb4h`l>v?u=vi&)t(<+oK*RZ+l)cc
zG73hC<9=VaMek-;s?SM*ao8hzM`3wZ#VBdi5U7wy8;caApP9<6j*R)gZB}4y>H)79
zO>f+pHQ)gSe9%_Cj><Q@$^F0|9lHl5V(yb(d8q94M8Cx$ehij&l)g!|PVN^Qxv3ze
zaH=Fm9BJ|aS;n4v&QW2mSP$!BCBfC~$J@nF<7)fH=xDeAd@)>`^X<C(UAkMYVFFJH
z9QgS1RU|+@?}<bRK%hrI_1y|>*o~FfDq=xgT+)V{w>qJbe@%^(lSVgRNsAk|H`uGv
zxSih<#K1G)6Blp%r(X?6%W2;fdCkJKj`CQr!@?Ltyw>|;B}CMqh21H(Kn!E{7d6$T
zej=vC87wm<X1b;$POfZ&Xk~XoL5zvHJ8&XTIrZ&HT%7IGF>Y<0r=zSbXzwvUUfR0*
zX*DEiX{S4OKJhw|{BmwXK^;b07)>%AfxB@`#OhR=ka9Fl%8XO;!G`V7^Z+RA8E!or
zH80_cuLR>v`1lp@J}y-#_I)-QV-^<g^3?F7wDd{vyYte6j~FqAvemIEde+2epn9#W
z6^%eZzO}`YWvwjay?x2|U<Fq!#AJcTMJofZ3B5c2(){NLEsPXFWDR8ibvEJsj8GRv
zHl|f?KqDq;T+zC?B*TqeF|An!kFVaKY9?>9^y!~~{=wK-X_egsRajw3!vCYYRVNgZ
zJ%7IwSUjtAhYzXg5sOGiE4EU99KU1xUTq~hF5$1{*4T-f*oi&lX>H=0=qow>6{Bz*
zBOMq)DE;AatbOy>o4xi$t7MH+Hz}EF**4bJ)@REfP$EN4yRkV)akc|0JIrjN;<)oV
zd??6oGH5G#b?BS))^JZw?eKpgSOz95EB~E;=Xmsosi(aVd4Lq8X?3BbD5$tg3)>Ac
zKxi5-GcLYuTSb@^I}L8ZBE7RM(&6^4?MFjb!}rai?REA$Wpd&+V**l;t#eu&*m}%9
z6<2Dgo)FWteTXkDmA3gW==0gs9`R6nma}K?_Fpo>mZt_tID&LP?}l(iBH;@@+pulc
zu(;a_DT}i~$HV`I=&S+tY?bv(n{EmN8(xH&ST0Nw-5r+~uB@P^^mvp$I34ssiFIVw
z;t$Ue$T<fWYdAGjw<wJPh>UJ(SyD&&6S1n1oI?z*6;kmkBjtl_sCQ4Y0MWpjr=>R`
zxqRwWv1%tgm=^13Tie*M#Vy_c)F9qAyk!>4Ji8_U%oeS;F6!&!;jKg&k951fGU!`5
zF|M{V6UG=gfVML8Y09}x1i`-0A(swm%mt#68L~@>%(}AfKF|9?U931G<2)P><xB32
z*5sp=9F=9H`K09RrG@p_g|!K!6(SDHG{Z>e<8)Uf+K(7XC-LcK(08`m8C!qePzFKn
z#H6(R4189cVJs>*QH;KoXc|{*Z%}+KAXzO43-wL1GX(uM0Ps`5x)D;Sb^+xhV)Cf1
zb*oY;dIdL2ws93*7@d^_ZBfOabrxJjPXlPhv%x-$=(>iR#@d%X=)y~QN8bia1(XAo
zlGwt8$h$|B%;mEt2F?6!j4fR+#0J;khxVi(uD!mj+$PM4KPQEu^*cxskyS?sj{lL-
zXNqo3R}k^b2fQ_R^@tbT@+n+y7_g3^^{v1I0#eK+GiDBPRro2yTlFF%G8xdXR(0Qd
z9Wa=|8+?w|r{7{;z)AU8asRysSpbwnC8e?4aP&LWEh?0b0Q0S0Lv<RT)M1O)^HQlg
zE=A`|L<Fj`T_09{4OIJ+@5Ktfk~;P>P8$n!-SpuV^$x}L0t{e;lxBDS#3Q7!5;xe@
zDhOa+CO;<FGc<3i>N3<`j`;TUCt8I<OpwibeXo6C0on0UFnj99>h*#WbeceOo6f6c
zOf@=!7|35g(|42^{#x-4LHclM?rE@7gj)j2{OV02JA$bJ<MqeaB`5qvAs)BCeiUHA
z93@WLXuyw8y;s*Bf^mL*sC67`JjTb{93K;9FtpUV$X_6o2qo$rZ<f-2-x})v5n(_O
zDh=0y9|Cn;=~pW^&v<(IZcE$>+dv4iOPdZ#%Cjz=&dur%4K3!3R*>{QSZ~vF$)wyu
zj?DAv=ss;#-2C5(9!bz!^a>hqBFmxNHBDxUbv^Bq+w9!AvwU)STb=M9r61$ykoihM
zeXJ0@F%}I{<t1lk@Pf8iyAQK*PAxQu^$@QXQW$A3H(<J}q$jE$IvmM<xeAtzV=T%I
zG_(bK4|QFXTFslfcNBK4F9aLe6(7J*3n1x*!itzJ1c-j@twg_z)i*qVU;%+xhcQ2N
zgKepQm^MY>TxXWPCI2vckPIR$QAfxEa0rCVC;oFV>J17H)|y<{FiAh1Yh&!6WX!t_
z4YmQ#DH1e!Wm2q$9g~zOO5+aFc4JRPB`b_6`C!`1xy8R3$4(;yA5c70CY0LIPds8o
zE=Sh0%rs#&+U>#^n}viWs4FzoF>sl|60>+juq$GUCGI|gBM=2dz0hAerO2-HezQ`r
zhXmZCu{P7I=e0}QUKq~UnHONnx*-qoNUV+-))V$N>>4Q<T%=4^N8F)@e~{iB78JRU
zM~HxBL#%wP#KcCN=6qx>4<!evW;^onp*SwV{lx0?^v?MVt=W%4K!~f9<(~RiepD(U
zqXZquQQT*L(b~9U={mPuqDsI<2EWpMW~TXV1_ah?$J(jWoY9~wmnvUQI5l{OO`(Gk
zd|3s~VSbPv*;c5bcEMS8LWd^~Th>=Efm|kNp<wZ}J<}eaooGC~-YsbY(Nw(3$CF09
z8{XHrinE3V88v?L2~H~|GOc%YjnW?(@!(hT^@&*5_1z?_I^XMyIv7nA(*$rv0xaKQ
z8x^3Jq7LiXwoka(DycYGVcn#WBzyqH3xaDLjY-be>(T+W{84Jag*57A&eQ}vKMX-(
zUhij}TStvnP#?&Oy>Lf{fS_0jZ}JoT8fw&D&w@#`T_TGqvIfTtoVS$xJ)QH1&RhJ2
zb!60ZMvr=l#`rOuhQcIeY}@Fq=i_^7m9@d~ZSSze+}1d!v_ukSIqTxv<in!}q5-3#
zk(rBc`E-<nrpCrTpzNT>vDp8*D^{_*bQkS$fh(&&29_n%w{w58y)43fowZ%<_;TaD
zbad1SINu7*JdK+a+_Y*^Q_0`6vmr`qFGo2`Ga_l!?e#7dnm~R|<s|g7fpV|S%Lz$B
zc+7L5&gU;B{cK3nW_%=6EQb><;TN_cLH1#aR8jpgW%)SP9?e*GgV2bhO&!#IvxW+7
zjG_c&%iU>lk73WqpG^C#xs~4&u?2oS7}AoG|HSkRLoN6yW`j-MuzF-HX(izih@*M2
z{o9c`+Mp)h8DdA)yQCskQnMF2$hO%FsX+3Z@Lm@p4_<q6bVE%or#nh;*O-b-3wPYT
z#YUaBx!&gr9OW7ECWiRgN^@{|{qWI-7I0?z`UQ>!?~Y|R966*k=H;H@gKqdRjcOI$
zo5r)O^cUiEQ;ns^itjJHAn_gD*Ov}&>8p23AhrFyZ^`jjFtaq7YdW<%PRF-)YM073
ztm}Dj!uciv4BDq=kKMJV>%>g82W!_wD8R@aw@Sp}u}MwSnR{f#ifz%X*q@QHI_Wp5
z&ol#AR^ogz@Id6u*erSozZZ$!U<9diWs7iD!2T!x6|fsZDq3suoV9u4>QU+Km{Nra
zXqYbZ>JW7ksQ-HnYIZn2N!3h~q<t0(*#C*1(?)Ys-wXCm5T*J?R?W47JLijmkYnRz
za5^BYA!17a9#C$6xqE*NyjA`K2j4{%HzS{F`Z&1^HTE>l&#t#T+I9ecLiFiktiy^(
z;xu12BI_)#*do6Hw*pVCTwyCE)?0=4^vgSK36paPQeOZSC=*hT9Tk+eotnyc()`_V
zmp2O1tp4v<;-3!pyn$PYjt*hJT*YI;2%Xa?ysgEWrvebPlDw~B0cbr%x%tuQl>puv
ze89Q*rj@1}hl}BJ`-2_SGkpj`fz}HEK0#aq{WehP49cf71#}97BbeS5<D9X!LI}-_
zPV|Qdj6$0SM*+DvwTg1bM<+w@FFfaxM#agGW10sWY{HiRnR2nt)sSBC^)cuoX0u<y
z4I6KkYrl`5Z-bfA1Fw5!cRo#3Y_GRk8xy$Xo>>HeBjZWPpMzlhdiH(&S*Woe97<Yg
z*p*vxQ}zb7H!IB*UPd`?0X6}rb14||wBAV5b=OZHwb>>)+LBKFA3rBnPQT~hN2hm&
zn*G)LmW2-AB12{~niqj~7FkRele(k8SYg#n2VY%m=_>gUQa!D`<Lfl-ra0icmiDsG
zuII4`Mt&#<ennivJZB<~pz10Gy~F*K_1@5-M*Ia_@_{s*tA^7#AymMp8}pdR{A6Fz
zY^>){lwd0@@`>0C+Q29hA3YhpK!)P?JY{N`BSA5q{DIWNOzbpU5rI<KGt(cn?8e4N
z!yzPDMBlyXwl5`h1*(r~Rtl1!1*~j^*`#^f3r_JA{2X?u<eSe?i0!@HoL@i6by^WC
z4z!u&Gc)B|OuJT*8ipB7I|iea(i8UNhr_3;c51C{u~f@`5IidY<QuEbt&7<=x=-^P
zO!pKsj6+Z)$s!c289&}!BtKv_w`z3mU1p71RMamvgQ*K$R-=8VgFKYA18HOSUkKQl
z+_8cXestmvP25V{(G__zto#*i#Akc)Lx?^n`8Lm;k*0Om_(hd=>4^?~*iE!D`C!&|
z^>x#m!f{F#7fizFDBX||mX$7HM&&DmMHdgg)|i6tnYiP_+<-2|BEJ*h{OP6#^O$^m
z9Z)@PeTyisBR-7-1p@@Dp8PGeBl)644Tr_IG|e>m1D^bz-{BJmW0dp3GpX_w08;N;
z$!1}ADu5InAE&KB>qV#GUdB|!)=SnX-N-u;7Gi>?hxGi}W#c=jV%eV@<YK|j_P4;k
zcKWDIZK5ZrtrIG;S`UbOP3jZ~@-*>$(YZ;NJU3-*oc&PqDiqUxe2#*$lWK6d7T~D3
z_F=}-Xwr!QbsEd%xF(BXotKDQtS;78;)risf+jkfj5cN99WbQu+!mwqA1br8eI@+4
z2YR;r)oU@*a)c^4Vm72;*i38hZiwZyT3_QZp0IuhoGw4+60VZUoSZ6J6KkXwoz*)_
zAn}V0iLn{ofC3M!y%*A#7mDUgcsJy!{b$;1=-~YOKK<6Et&!t-(|yVi#y_pnl^_56
zmgCl!eFH;2;quqc@}ugkJ$F?VGjkGF#7?()O(*=bMHirrt`zG$Pra_&!@t12v!}r@
zJY;`qX6F_)V*0crJyFY?bo&$n1CnmChPoj_XQ$>X>j}|WW*ukuH!*o**lF_-$_>iC
zM>2Td=bcvX+0uceK>1Rb25_!8k0(u`Fmq_`H(hS`P(%x&EPCBVJZ$Tf-JNqu_2<9K
znKOdox2<o%&pvBHMWObMw2d?`B{*EQd46Ap{EF&ntm{hL4Ij5O5$Phtx{Wsc248g}
zc#7?GP~)^q!q2WcSMBtBM!;L}qW;Iq0;anc8XeuTVpdJoOXD}aPrpv=(42SW_4vHx
z!}c*L5U&~b_$X~qLOyGgps*?ckVdPoF#OT=Fx51BQV;p5@W`|Q&E#9zNu*OLHt`fo
zO_$}7^pA~`O5R8N@YEi`TzFS(jB(P^W7+S%rv@#tSmwFVz7^WEADQOI?>l_iR^`$m
zF02!Q5LT}irBkJCE1cmM**4~T|E!CtDtKagEp98o$B6|nbX!3AVCdU^x3IehLD~2y
zS(d9)5wkGByI&aF$LI^ugzhr*y9(n6ko#)6k6wU@S&2XKj{m_^%aMz3drL>lb_vQ^
zDVc&6^1e`wZn+{)-_v<#@VRmLyopo=6*s_)aoViY*2jLX?Iu1RI<VpbUM<wGsvxbu
z@LD7g=^{Qm?pFl3LH<ma#6w_PfTog{KWptVmsob>PneTikkI-Q+P)?m#D9=v=Z3Q`
z4A^>A)>?cMv+Jg+8%HhPnW9`)AzKmI@=IdQuEEu6;B{Mt|G1P%&u7Rl3sG7pwCj_+
z2{Th8EZKLJEF;nC(K#5;{R?(|b~vEaft9at7x=gqJ<YZ2W@jfZUG2G^8+KI{dV%OI
z%zk^9%oxhhClPDFgF8x!PgbOREyOKw=f*uP{;0nyY~_6@(!DdOL3^C<D!4w0=tL$=
zuZn@(wk5-ylf>N>q!z*NIejlgW#6E~`Dnj%!mD#zpQ7eOWH#~<LdvNg`o}zEfZE*F
z%znq&jtSg5re@pU^hs4XTB0YHlo>a!va>Cgutfnze%TUc2H@@y%-${c%!XM6zP=9*
zQ!*sEkDtW)1Z9g9HFgM!X{VbH#wtp+>encsQkLpwsFO67i|jQPgYoEkVJ!aROt<tO
zY5bP%q~J*_zAR^@jYwhTC1pQz>%_UsN4mSa3!uEkpxQ)QDV^f<yCJK?*9cI>bn`MO
zgo=kbUOUxdmC=o+ll~Eg_UndrGc7aB!kx^0*nCwBs{zi``@sWC7!#jcW&!VS0CB>>
zi>-bRtrgg7Z-KeTWJfo~)wZNOGR2tz2H5S+B!dj*GaL-vd?E*Sv!$ZP1*7tU=2>i1
zmdQr#-UMFdy;ba7?Xzf4DY`~MN1))bF7YQU0=@gr(Kxun3SOZSnNVc;?~FZfvcDDR
zrL3(Bc%;lkMepABfuXr>mhOw-%ij7XP+}IWK4;GJ?w``)^7(k=XwUmr9tpRBJ_(!E
z7J%^D#wKqvBdAW`zyL`$lHxDI0~w5Mp^|qHh=-I2VSl-I3!KY?;;+BCNN|d&K(57|
z$i|*F=8lG`qiHd_(FkQt+i{r7!67GDq6Zd~AUXJq|C>pVIo-@{`|sOzDv}9os8L@S
zKq5SHPz_6~K8cNG1}9iBG(!UpF*NlbO~a4`>9l*s@_$G?pwwdldr@;KTWJ4|ef2m+
zyHfclJTG11=W^6Pvv=qBd$K1Z3GPuT%S$Kx7)`TA!~Iv;)Sy)_WGjJ%4FCF17d7Vz
zk)rJxDfvzwe-SGd`%j}XcC&^2&O10T{@vT_5!Jn<rg}IEDp29Ew931Rbi?2xID;fx
z7#2#}#m1k)#ulksv<H&rFf=t4<y?&ig+W*E#GzGc%zO-)jc8JUIabIoa9*$%ge!Cz
zA77Z&Oa1;q`Zd^wp2KY3)AoES584DDFH{`=*gQcs+%UXYL0^^)$BgBr6m6g?Z3V2=
znr}WEadi#5UIbk^@$azev3`V9K>AuwZ|kpLwX%g!pF|>GP^bBCzBQldvFT7>M@L64
zs$b+b2E`WFjR)dU$23f^{V(n}&4*)31Fi<Fou)hzGZ4w<TB<gwZNd9JVg1s7jBV(T
z^I@3b)4uW7;4sP%5J<@AdA%2nT+7IVc<;?e7GKt4*{}!qU*Fg(XsV{%=+Ghb>bLQ8
z$mx#iccl15`Wj{izB5%$<y&RuW^mDohq!DyLw)-<+dKy|JT~EwrlPL?@oYhBfO|q1
zNYT*z3r*)E=rjHB&;o5!NyX_tH4Z;<u#gCQEBj8c_zz!oPmj4h%NPthp`eY<<E<C1
zK<?HZjyy1^2(fhEe-!IIi1q_~<JZ`C1O<Kl2gEJZJltcUoavHvBzI=2?ha)l_W1oH
z4|)RnHCH_Sj&F26@r~heHTMTS+9ih8(t^=wt{F+sI)DpL_i@U4&MfmZxQ}d#VEauV
zeA3ZjH6O*eY<47dN6&lsy9v{woOE_k#!`NapS@dQ$V6)_3$a0G(#Yjm{~3%T?19+r
z?sNZ4R&v!Ifuu;h1AYisqY+_`FJcsVLEQJE*^lB{tSK?JP=*-0l~k!JPjKOPp%w*H
zzWxh_1y{wEjCXr^%zb>v!VJj2rl-cmk^qDU6dvhK-Twg3=~(;EX%u(Bg&Zaw2FUnz
zuxLIW>`I2Ol;A0k&^o}TPv=Uz-zQlrVAcjQwEc?cmI=$_Q5g4ctHUiOiq(XBwhxD8
zDLiC8Z(lbvm~yE`$!s;V(wprX-gr)b4#Q$2E_oNbx@c*a<p*5$^<ZgXFmINo8$K5{
z#n)6^EWF=GiPx2DcE88}2i<WfY$_|8mIKwb-1Q)rC3G~zW@Wo9YP$6ABKe-zm0cb%
z&=UE8tLbe@3NrVXY|@Lq$x#2Okcx9u-yCm4zF{vNP+u&O;jj#VZ;NlA!j&>faXdZ8
z|DH)UerCzvD3YmTmvVI0spcDB_T_75AeKF<5Rbm^)tH(XBD-HVd?GFGC-A}dwcQ)`
z`+Q!>nn8Oh(IlZv5wly#@$|IHDV4Q21Z1QSkm190Qf};e*}Iw>(w-dg4(S25RpkgF
zlPO$3?1@fyRqm(Ksq+ty0zLvUDp;lNerfW@J%VA*r}J<F%miYX4KVtSvQ46A8tzV9
z^Ss*OsI-`?kCni8vWNb0hIU&+_1$boS`4Fk-l4)D7X#yETj<{b4wjkFsU5SdFc$rB
z;uBOQBKlGq==ca^T(%T1qkV-g4Tzy@c~Q0%c5glOLtux5d-B_!G7UZo$udoygzHH$
z17}PM1c)~O++VA`voJ*#ky(LW9n{MBYnVJm^Np<c`JUa126Y&>izFO-&jorpm25V(
zcVHDsWh<D~VD;H}w|7QU%W;5una_S)XZv>_)1VCPAklW{vZT~yDo}Pb`rdvgz<Jyt
z>O$R?B?ZrWK1Th*miF~9Cd#V|Vp&O|rHZnB_NW5RtZ-aY@cwmnfAH|U;vyKF!t=%G
zLH;ntwsfnjiY}ItXiqwDQr~P^ZlGNDF2vh==CR$DaM>E^b;Ea$Zp-(hMs8jAG*1?7
z<J!e-qq?rSus`PqXs6`Np7hf~X|vFn7Q<qYH}U~c?h{UFws}oB64MQ-&Og2U?fvEH
z3`ra*Z8A9+_tH>(w9`5e)TCF`PEr2oc%RqkHTYj=zrdLK>)BO;ROzVWu#^&vA=Cxk
zyL3eBeh47K{a(Ex<$DE9i$;{u7H!%;K9#xc!qG>E*@<guIm~Ue&C+YZar&61H``9{
z#8{Cm24!VV>FFNopjwIF;`8C%x#RbzHpHper(2Z=OCo`p4Ttf_yWu(Mq*JtD|5;xi
z__s{_r=H<zq-!`?Zi2d~6n;jDz6Zj|r(u3W2s>@9vEvUm?-8VN+7#l5Jlrt9bgV<o
z*eKw31&KMWr)*S>O;@OSk(S*`!*wYL?_Ssi68qXCuX`J@Dv!c;+ppWH-^Ad(SFe)4
zAlUT1V2GZ=NE-J9YVB{J&BDf#`|><B!W1$cdRTUAs+(0YCF|WVJ<X5F;=A|GvYCi_
zQ?fh$`WDk%hsi+yquA(|fWFsn<J_XpetvF)!tdmNJf$@_tX=<|zf>{w-iYrXc#<aO
zh|-;>sUl~|>ofl9R(#7<)vlzi`W`#h)lvVRRHK;$%?o1Vv=#WQes-yG!f~kJO9}J)
zaGsDNQQ++UD0@a^wbJSxXZ-?`n@p1FGRlk|#1j1Fj5^-aI~jA`dn@*RZb$g(Xk^jB
ze&IdxIN?{K6F%VM_Wv#=_J~MQ6bmK)qlB4YySe3xPopF#8Ic&E{xRZ7WLk`JTZgXD
zTG{e}8LX%TSH)P2Hcw@5OU%rV6rZ&+VfPej1p{47-H)}y+h>@l%Sm&jTP!yz<`ul_
zwW?gfOdR&_?#>^LIkcwuNh^<iDJA}Z+*qVHWkNJcc(|oq$<JC)_5E%5n#rf*kW@m~
zh=0Rv=%d6Ui8>4&eSfl1QtU3&pj4s9^8g#ww@mfRI872C`+e!4Wfv}cF2C&<Lv!ho
zPmG-@pxtYjqsud^(aqY=BB8j$J&IW-DgFP@<dN+K$ipj6etj(ezdUFv^t9Jv=_>JO
zQd@=G#AXjw`1R^!HqcK1nxlpD36>+E6&Hn9z}>hd?kz$%C-cX;5`Wo97F+;UUVS3F
z6r;rR2d`+(2X+cc$u5Jow5Es?^E%AGKtKPM-Cdw?YeX_em%B-byyD|aVIjWM%7+zg
zGj0YR4WLVp!!s;><FC6JTAu}{qd3vH@=y;d;Rb^~3Eo15`R}9F@6Oje3Tk+RYI!}x
z!%hbbee?Pi1Oph~;8ATCnKVYxsS@$RhWe+=jU~pD$`&^5s(Qtw@IaG<cj|sBBlh}>
zn6<c+gypEf2umt64W39GE?Aray4tsn+hELliyuRqxf>d+^;=Iqg@p~(YVnMP6ZD|j
zO-+8Q0HUqN^NL=y;A~dRFx&?9#tFj~fdnJnyU|4pZe-w^px>LkS<<S@@uAza8Wl#O
z$w$|fbV|?bw=*SxC`HxJ7PIOi{sLK4PG{)-o`mP~GfpOGXM5`%B3bCJlrEU5SUFzU
z7Ujs7Ta6rM_rk5feOvMOR$FveL)3pW-K~(g%M8Sm8>9sOEqC#yRlznQPA$PMjTRDP
zs(PQY{5tM^?9{l3?#sITeY{JSQBnW87Tp4_Y9om;LFAMj#u4{MobO!vUPtM8T)ClD
z!nk}CcE@6<9Ho_3&Leca&DP|uQO00n$(+S#5zS6&oXwx#7Z8dx_fluj*gk%2oMlfF
z{MY0jTFcDy*SNx98!<<>KBBU*!+N#qeP)9}z#$@<A>qsnnTS3@Ob1F_s_2jF$E!kV
z5BAiEoqMmLp0UE7U`BPjy-6*%nEHG1W(!ZuW9rZQcXVG~MLDp4ivLuD`mOe|xsPS8
zmat~gzHGbS+vxC&_CS!bcIfzPk?XW60YzPwDQe>$-N)+WM<kr9#xT5BQ@hfrnp5lP
z0>uhRtp6tJPzRCuYpoHa_ec>D2d?%=S0U;hR46fWIb7~wqjVQ74{m#Bp3h{c8Rg~Z
z&`y8D^6qSX2{x~N=}?U8Wv&tYu=MJ1Iry2VqCG!`gSKq?Rq4bGj-tx7AZaX}zaK?0
zdVKt!F*Es^BxCVc>gJo2gq*!AhQ453IngnCHv`29Q!&bkqp_^Exs1DtQhP`;Wp`?>
zQIP7U%NzJd8}D!(X{Up*qkQ>3{nHEXYhrnO1H;*NtFFZ(C0YCc;-@l}Sb@`&%G=>j
zp&Ah^Is(V4-(N5_)1wjhex{^R&nYH9{opD8bL+%OB>5}orM0+V$$g=#(e28H+j~^0
zqi-das<r%S7(BYt!*uh@z~wAtk1E;-myiC~=`-~dtVDtW^r1OB9ErtWAf<5~4bEYo
z8Id;ozV?Ib6JEX;>p2YmHuNCZD+wWwJ^0gD)gJ;sDaqisF@80I^q`5$8ALoPpN#2x
zVzUAbGXLE4G6yrrlG_UNt9o|Jycp$le?_;S2RivI^J0SPT_~dH?6*mpLoC&4O-INH
zT%}>RR8uMAEWB}jV7Ibjf@bQ4!hxRckKDC3s_I^X!j+K1(^#N(G{fSOYU5=EjNGGL
zy!VTm`-s9X0->kzYKNoKR3`e)>jVX$b&lTEP@^d7{GRb~nf6)?M(FW;6rBd6sK5@`
z`@{x%2yClNBgs>QZn@e6twZN;PwkNXNUwOU`T_Aa`aZh6e>}Wmf{NZ4F6uDfHb`Rp
zdB$$7ZJEt9PViS4D>jz=_Dh`O|F4H>Q?-*hJyCn~%eSDcI#vWa<L5ZE;Rjpy(xDkp
z6De1_D3WRc&34yV9?pcf;07%<06)-PtLWiSOw6h77Y6kaKM6a^R@v1eysC9D{<U6u
zAgjW@d>V}it}4pO>?gxhr<E4VGIimEr|E=XuX-cu%ob8OfU8rsjzi>Oz^$`mhd9^d
zyDU`b@LMQAi_1K*8vaGw9EnKc$kQ7+m=>^hM%FsY%@{n^1)j2yDUrCZ!na{b*KT46
zf@E5IBWjdfp9IQy{_=?Yf462pw6<#6mTCIwnLYWmCxK&r3$-VP8h5U4$BH0`P5_GW
zynm#~>*_2l{aV~gXr{F`K}(GY+o|zlnddRHUxXbyy|U5F2i~wvfM5qm76NEWMVs@4
zm&LiwS?~%z3)Lg3M7<BOZo=xz=$><*aBeWD=lMa?YeWVb`<M5KuAHf#>e?fD)e}3T
z--^|gD(u+J>RT{B(BQ)VTIDH{{xp`jHeoeWr>qWZG>%)2`Rta&H7b217=bP;<0I+X
z<SF#e)Qtal+8ElTfq^$JSp4_uY*^X$GV|>}#+6gI-)wJw%zL_bIvk=O#{9im5oQY9
zsB};D5ty-*ao**QsTpD7hAUUB`0Ca=cbzgN`4}_%wS?}Th97#YrKa8<qKKf&<5I6l
zOJ-VkY{LVB!v*{zP4dtGP{QJyO}j2{YFrV13Y2U9eXnJAWe^2IDsFM3T+vQiq7MdF
zE-T5MzEbZ1jJ)gbO>eCT(oG0tJs*W*4HehFGqdD6{&o+}`$Y7-Y|X~YTYvEgF>|yp
z!TZt9TmU<Kx^ODoCLs%clRZ~se+-J4!7{(fZa@tMfsO&UO1~~0K~U7GXs88rDDpGZ
zf+94HU-TW3Zs4*>bDgr3On)tYjH&4Q)QiWNTYMa4LtwdwwIr955BFRbx>2)GJd3VL
zWpkAFm+!86l-sR{I|d{^O!JLPyaD1KT1XerasW~WUbxzhWp&tZGm#ZM=>m0Y_QQ~}
zJuu&9M}c1xI5!NM%t&J~V`kX2;>?3kzZPNPJHZ{!gU4!R9`eZyN150%CE5=Hmj~=-
zL>jV?Z)|M+E_uZlt(=*r#t@}hE?TWx#duFshbg+F6*Txw(Ks$#)dDxhM!kj$8)&ss
z6fg?NG{=IeaW&XsuQ(^z?cV@g=vR<Ip>0So3a_H%lZR&S$3yGP&lT!2c;k$8tB3m4
z@5I+DY93F>{pu7T>uBq&F@>vt!Nz8AN6kPARFo;^2Wy8?GbB?!wP-VvC91{mX6&(g
zK(d0~CrE5yAV#}4GzZ6L|CLi_bhQ0`91nhTwp+=-&RCg%XNtYyHe7Fwjkl2P@$1d3
z{t489y~2|V_*Ax}*`;54mZ9ii*CQ2%ZxSumaBvdt9OS6|RbeEl2>hQ)3SBlcM;kA1
z*KjQYreC3#L5!-tt?>T$lj|;L=$SFzF3EnNd9rw>$Bfwx`36lp33UOaPcEh4Z%E!K
z`@W+$v`km1dp|G|Ca7BwG?SZgtLNa>z~`bgVzY6zR|*LY?zJfYFXy-H=;aE)OF&5;
z`KT(cmvr9iZ8)TWn{3bbrGk(q&%{Yr?*Euqs?aT3pk^v>*q`{>u_^?gqT*s7FJ>-T
zDT=YWAvd0IVB^fuE9u@V^Ph%;5MqJ4>89xHNQM#6Sasvl4?Vvu*!T!;RR}rnfLn~d
z7fgScPD@sqjz@P?UDTIDsz?0M(HP9(<*^SZCcEiGC)pn$V*~wynJ%MvK6sA(o2LnM
zQ@Cq;>k~r&wz$PgCEB%DU%=3r@BH;Jz8V+mtDx!WH>{VY=2&b~)qK;aB2GGI1A7KC
z!r~T!XlZXYKz~?pJ-1M8jc94~#vvk|8`mRwjScRY;dWj5QK`BZWjeb%ok!lVRW~vw
z+n_$}gx!jYo5Pn-p9Lif)-01p&>XI03fhNfWR5$7&UGD9DP`y^<Z+90EdIoe_~MT$
z5hg;?)+sn9ZXa3Y22Qq?C)fMn;yXacJDKB^IixSTL`bnGw<sc!L^s_Va9`V9Ns@ia
zemUKDpv!wE4`+6i-*Ytt5`d+sg(3H<``xHULF7Yng}-MMJ=S?@iosX+gYwOHg7oYe
zOa*U$ZMZR|$Y$Ky$`sk$F?J8-PuROMCCAUU#{9$4b(gInN8Q{?UZsu+P?>F^R^H}K
z(jTOYXfdz2e;@>X#Sin(!OCL$eztRmWRZ4+PCZRsHvt@AeHg0lJ<YujhdwR;pMhvU
zM38dI%gd>-m+u#yRM4|^Fnf>w81WP=VkiW%sp_6nZfU}gd)^to5o#w~f4>Ay1!)j}
z=<QO<x(AXZ;?ocv^mXO_iClSkr^X<|SoO1lB8PkoMfa7A-ETEFDzc~09M;FX=mpri
zy6_7K=`v`ToS*o&Xg;W$Rt4WK^pe`zs2IR(vg6Cnr?98islGspjV%kWgW?W_G5qSz
zUjCnoAj68*C@Ml(p~d(1>YguOzT{Zg)iz!XOi1mABcl~ps51YCMXts9@z+?6e@L8z
zHLOb5BtqS-*%|MT(9E&)G~9xc0qr0cd;A~KO7}UA%rrNv&W0JS&Tq|j67%6^?t{uQ
zBL%#5+0=|6`l_*i@z1E9tRVfd7-B-0wYFLIp&cyJUAJqL-!q((-$3C^=Yy^V@x41n
zWheYm9Ryr7veF@I+;U>0u0wkVxUlSPoHz>RTTflBYL>K8tGA>AT!oSb4}*n6S(>y4
zuDb3$V{r%mFGz&1h2~8|Wy|e9o`;1EI26{lvsw11+5))A0dj*7(|v9%2~l|hkAeP}
zhe@-GCT8mBLN6)u8Gbw4k9CO8Ayi|*4NdCRNf16d!RDDMgo5beS}4tYSAm+JM~J3_
z{Vx4-thi|F9qm(Cr-B3Yby?_~(Dfxn9tQJ<t|0*L)CH-%<wd(k;0f8x>{X|0+@oRv
z+-_u_%4`MWIjV4j{#~`Td48H$^~2Vw7m*5!|H>!9IPOJ+;X|VHe_a`6!eg2CHgro(
zhX&6|VWVM<-*)P0*bB6x+$qQp(9S4>-bOb)OX3aOuO~+Iy)Xk?p;muP*j}1BH{zSs
z|8ey$a51gl|M>1Mx(J7EMC6hqiKwI|<#LJ&ji{J>=t!Y7Lf4s!^zk_oN-m?B4o)?x
zA(GN%l7otw4%Lt}gEDETG^VMhX8!BhJLmU({r7pjPP)wA&weiJ@?P(?meYMTechuC
z^iY^DWVqU{Z{fWh$U$t(gbeB(JmS<5H&}kt=3u^T57uu2Sqr!}Vr<A_y`LO397j&R
zgXaj?2d!1FqAF)EG#Tl8qhO6~gj=w}TY4VBR|iAFC|i^LG?y^S#2KUqNYQ{r0Sk2B
zxM8d_0}G+^O~Qf)aCP{*?TY)YDz#;hoDUGI-jq(iuF7Bg-s0z`J4J8QJ9uD+&i*dv
zQM8SrgaAqCSCCz=j7v^~=M6!Xfx2daP=Yw<D;d-?b*xUjpd-Vbq*buNWia;3Pb-Zy
zgHW#1&a)bOG`75oobzT1%M3TgBhA{TjK&aLSdkGu+$=0O3g0!H6stWi*m}X=W1z9@
zrV*#--`9ZJ0&6=(6Iy@A9E^5z^uZ$?n!jM&J{Z?9z+1rCK(p$;wa6wsKcQOOS|%T5
z_;j}aX$=`o=Fd7*Q5~W9h8H;u0?>`|h$=@UqXVRkpSDD!$U<9WN*lDyNV;&~;LNKl
z)}D*@wdTg9&bo|KElteTBgioZuNK0VyxhsSJrR$GADtr(aug$a6fkp7)xjqFf?5k`
zSyyeSLdqv*o^6Sz>IrUXrh1M{PrJ_Tw1&<J3r$ibZzMOV>tQ7H*EiGAZjutoUA~)4
zjiyZ4zDxTpQTSv!ec5kwwyS5J4!CSZG7&{U!U7MgC(PL<sD4+bpA6Ha2awMp<|`IJ
z^H}jBabSVmq{mUdAu90_cc1tGKBh$gUxM#1a;`Kq2;}Kk^q%{zCMx-&!rRyGk7sw~
zZgLCwENrO}R4LS9DY2#2SvffBdy#KNejutdf(YG_JpXk@JYKdbj-J~<9mGlbbu;IU
z0mmnTs3Zv>+6cP=W5t7|3%Nti%x5TL?Q;w~pDx+B#)nt_EI>GyfJvn(C_|bX<5$%9
zam6lG4!+%)wS5yjQ@__IcHgZXz^OJe2)V+ME8H#wEzOM~q{2dLjfU3`{dP?>9_g^v
zw)Hy@e;tp%M*XHc<vOy$QqAT~{6p{|`f&HD$1xTS!8#&xDZd8EhCgb5D$fiMTwME`
ziBMnOQ9kQha|E8q&53@Dalf;h8#`S%y|A@J;U&zVt_0h@qz)a|yh0(Bx2(e8u*<xK
zCJuuYnPVHr;$ZsFb(ElHla83|aLHsf;?^wf>|Sgyp($M_&(;5qt53<84vVdq_>j<h
zzFlMcapvo!ALLSbSjr4glO<41W>Vrn2Rxo@b?fi~$}?-RP<JKDPw;%taKc4ae(S1b
zKdRq*JsOOo^^74(iG(I^)wI%39s3&n&lqSfOtfNf8+@@LQp~xedEHS9HT;Ogk5!~u
z!3%MN{X@II%Y~<1djE-5sj)Zqze4Er%2A2Gwqxg*&Iuv%f`3vLs>T5JarcWR<%xki
zhe2&de>lg-0Vu&c^<1sU4Au|}p|W(rwZg4SmtPt=2utTx7?Az9c;4(E0b^8f%%Wea
zs@w~NC|-___LB1Bxi{yK$pIxTD&fmd8#AeXc4^Nm(j_QYQM_2c6!)A@n0~a&uvfb|
z@hiRBQxunz4ap!3MMly;k(Ji`WqV_Inpkjq`piwOGqjI_>W%c)Ek;Faskf<KYw((i
z^PvnGub@qOD&Yii^qjr3l~qt_$hB<Hlo7wAPs>V@r(tkCAp0vW%6|OaB)$n7aGXSl
zY0pn`Kaj?@IUBB<sv)X)75WJ3@J7^FROtD7(WBI})G96{3xl$Y*PO^*98-hywVqNf
zk~z<1e>RYIptGg7Q~M%0yc2kakhKl3OG+qEyl^~TD*r@vbe11;(Z<C7OBOat;a6P2
zO^T&j8ALQm30uTRz9e1&JNePm8lVT)D|gf7=5aG$L_bNKhyL^cLH9!D-els880W9>
z(7%;5r#7-hyA9L(dq>LR<#+7Q@`jyS-HN%gTS>#fJufc(?rbP_xre`nj>~pv+m>1q
z0WS>YN!PgW#ioc6H}psI>vo|~ST-0ScifJjN`%t`=<T-iEzRE$_FBJvOHWA&<wZ<T
zkGux($L!I*8JT*F#=C+=gnWr$zlgX`0pKpugfP;yZQS;{pHGS<Pn?CN{(f7n;X7g1
z5SiY=Yk^0a;SFn7n@-76oT>e%Kc+<+8)kszv9BeEkK?33S^&;(&;F?D>%mEPW1%lz
z=%Gp(8GsOL=+v|iehUSz;)3%Voj9XPzI}wHY7OB1M|jZ_J+*mD5}rQm&><|egyokU
z!{E<HuMz-XlGj{!Q#RY{(@wC%yVv(FJ<v4sT3r%Hd^k<}=>n}P-QO)F(ILoge4U|B
z*<y8hiN<bX{^11~frEyH_Wk1sDBP8mj%2~N(!{^o;`+;PrCW<Tpk%|Xnwt&OG~z73
zAxzFcer)c(Sn&r{D^y(hsZcQ~`9hM{>e2|`UAAx+uJ{dV9s5v^5YThar?PWbth|}K
z_jXbgmhip5zrX2f67zIQFux^IUr+c!cEeNjrM$~!ypRvOVfGtq{v7wWfM3xm1$LL0
zPiYP&A-u2VqBI|w?yb!czt7&zdo=YV_{2V0PFH-F!fwJBwo+$ztG}XuO6Dl6(a;qY
z&8?nWIy3Hlqih2{wx(4D8~Ro&tl@6iL_vt)eRixv3(s!Xy;|7XJ;8RxX|)FTaz4Xa
z<INQMRkge1i86X)Bk2?}gkDBl%FniMpz{l-SLOtRkb^S8;_d&s2;H2z*vny1%jQx=
zJ_5F1#;k%MLO|6YNmt=<l#63M)=)C0)n7Hs`<nzz$)n%ijQJ9Kj>G;Yg;)6L7ObCN
z>v4c<dIk<r_~tp9B!S(86)A4dI=UYLPrJy(eA1{Q*7s2Gr*hp*snWt@yOz-byPXpY
z=uM)cbRu62j~i{kO;3)vy`|$ge~H^^-L;eBCt&XjU%6gz$+;^a!{g$!s&zX?z6#xy
zDwiJlFPVBj)(yIy(~mu#K5p8QEXEILfo3LRXQR5UD^AQKfmdLN+050CR?k#js=Z6Z
zuL;JV$s28kjmETbLI41Q0aAvwt%=PuY&O_O9UD@nDdsC4<e;)9q2GkjmxrNQpK!a5
z78L8hW#I1%4+dG{dao@`L;#N_DcckWI##PD!{}NHWc5AC@JV`g!OG(=l_27|A%yd6
zT&|g|f9t!A{yE*hG+2S^nY1Nm-{2CzYGpuPl12Ukrcm>SxR5UsLfe<F&NUAkWb60B
zJh>n1!JZb!xPWAUW&qN^7*tmmyj-2j#>s1Wk6PBKk!Ux)|7sv@#&<JNE&eBT9PFDb
zT~NXSQ#b$P8%a=c;NV<*RKHa@!7V=#cvf)dt04pz0H2a!_YjqxM)g0{GhHtzT5%RK
zbon4_SJWDJNv|a74v%g(dvfFe)Jh&8xWkOcA~-~;x7(`eYS+<SFU4kQ&Oetho7{6{
zt1y$Li9su{HD&qP@Z=%WzNAs!2AahejZLbvtbMFAK{e@BhU`gTH^46yPvA?o6`uJ+
z2d7wLx9Ou)OSfY!rj}}{*79pWW~3?Bs4{EsLzen83zU$E^vv<9mBxkg<O_srD*J6B
zje32?WxZ_wC6^E3b^vT4MxeI!_D{nQcMJ}S>b>ri8UNDNqOf_XmF-*75vTwfHZ~^*
zVI9?f1teUmxGixy#+*HN7{rCO_cyajJU#0r{2GLKh|7iY%l~JbG(@gyO<i?41QpX<
zE;3{-eRU!`>4s@7ZK+#>ZrCq3^|G`1{w@)^c0Rr(;s#m+D=kRy3s%?F<wOpde6V;b
zs-~^CB}+W;oLXKBU<CFlL*ClF<u8MfuUk2t3^S|;__%$;pnXN*14{o}=e(E)R}IL&
zcO<&?bT&7twJ=N<uDI{$hQ<e%9>GY&5)YspfH?gJw{7$zOh$D`t~dX2kY6WYNxxd?
z+=?TGd8<h396$l2!*;eKa=SO*2|5h7i~PvN1gr;c25Rny*eYTw?C2wiv^t*YQ{>GR
z-9j)+I7jB0JOzr3TJ!CDzQ7n@8h5q&3N;E}T-!*S&GjWVf9jtN%5T2%Vb}@d<sPw`
zqsIFjr<r72BspUvNd5~GyI}ooaUzsK<!07LB5;4Qx?ViL_EYi&pNrZ(xX_w?pgUQO
z1fFTtik|HU-kkjx&16t|+@lPt@hD}1p@C!%kuIf%F)xk{Sx4U`C*Oh*j#yLG*v4VO
zAzAWwVYH=KU%l+>n~~1k4jTLtGQYP%Dju9#LavC)rW{;x-|dL*ljX*fS<G2*1;owV
zVVxmeaUhL>FyGx8d!$v`kx%Do-HYjM6wgWdvrr2Xfd%`M>YJ{kS~{VdQ5{z(@S1R{
z@|k!PmYlxl`S3ru#V_mdUuxA?<0w3~TMLts3u&bm@cVwpFFK9N<3g()Z{lWS;>eyc
zK5l~o)0TRnL<oQJQ>cg*;v|c3!nzOO9Y8Q=oc>mA7RBz)3XM(ny$RmhTAhj6Zso&Q
z6?>Fa-<?X{)IUav{x}k0OqP0E^-i4khWO3k>hASF)Rd4TTmk3O)}uw;RMP|2Os|nh
z9Y*HtXJ_YTGc|4``CYa}fC^dFrGC1b4EktezK4WBn9bYGZ$FJ@vbu1q_^=D$Bmd6j
zJd`gH5KctK0W#Cjw11-ZW1IHn-&tRP<Z!&|uQ|^Xmda~2U(AnOsdisI>&p$z(>#b7
zg>z^%!_@(ILWP|%!sr<zd2KR$QUTl=4)UEwP;s(YonRaE#BNSNRf~)C%Xq(F`LFx-
zxI&3>Q8<)DG#p!V$J6(u%Z}iM-bFq{$Kl9XJxI{o5S}cyz){cF?97VCA*^6+J@uxg
z<%T9lV|Vsxy_%(M{fH9vTa3%<!M=ZjQ>4#X>A1JPVv{@5uc`xW7AUW!HUG3c9$x+G
zz^qsh?LaQ(rg=Hu-cLAGnn3-_60>r3eupOp1!T!<jAMl!S<3f_dcy$&w*MF0N2li&
zJP>kv!Tb@G5T1YkBF82QzCKajx-8Ghlv0v4{qAq>!-x*?m+0#&1C4_?VHAwmM~K_p
zve@rtpvgwOEd)Vy*WFnC011HB-JCqKb^!OX-=mt(<zGL4{`?m}6TtBxo89R~Qj+yo
z^Pc?uqj{C$&++d_k&v*<Rt&&;G7yJ&_?w`*Q6X$n+(}jT=`_vA>rH7ptqH${i#3I?
zrs987ZyktU_{#S}@J4K8Ns@(08eRmQb6``)r0TJAs-bhLP#~E^DEnHp-+Z?>@jEa2
z<V>+L-r0X_mEE2HAlZfIjXl~uwLiQ(9qc&}LXQ82njU6`@x|&l;fY}z*R13Hvu%H^
zaBJSx5mxggHLT9RsMjwT>1c6&V^3aKU-`(?B(`@|;@K`2G){<i_BNtsuiWU-{Y0`v
zurU#!czn`YYof((DIw}g%?7pR48_)o|3uC7>7OmgQhbR^nKTgRsWJA!KjpCE3!?d!
z3dGebI%l~Ues<m9+9|yx4qz??ZIVL6%h|eTCZvZc{%FzmM@T=BXh(F;;=TPu&>M<0
z5O>qoWv~mj!_(Zh_~g5qi}uybqBlNQiVNhmcZp+E(fl`Pv3?>djqxn?BZyOkn-2PU
zzH?G^mvf(DE_BmZ{beY5s`LSvU=cO140YRI3fy&%y%e;cla3xSzJJ?D6+RH>pts)p
z)3^O*kzdxk7=rXr4lefH(6OQL`PdHI)nIM>=unb{gee8{B_i`a?MI0?t%S~({l>cn
z^_QR`0|d9o1^(0g;QjjT$%BorIrf+2+%UzD0;ySJpH6;X^T<BIDzq=oSWD9izoo!i
zK!h3mZH#`;({`(N>w9p6;LKASSdT~J!WOFK&pF$TqgoQ#YJp6%_La}+q$&%0I-EUm
z^>td}5ba#9=28n~ya6ro`|jCxbs3VNoJOt7YuB6>e6iLnQN#PL9~b|3Ut?S7r+55g
z^p$E?W?!LiBF&#Ym8)s1wrTSGjqu;J`M(A^%&7Xb@nslUP5t-vpN&^Jvy}xqt<QHd
zOpGF!WZmC4=In(<IqKEB_+D{I8<iD~!h31_k(?fx8TcCvQ}b76)R4_AE-2~qaxZFZ
zpen50W?*DS0Q0Lk!i4kJRyo5L%D5V_wp)lQ;mx%~k$Txxn?*V^s^MB6`YU8anZ1UY
z5YBaR%^MkvvG;p>f>w~EOPlD&qty7>9x7J(XffYT2yeoSn69T|RM$DAU1$r$>WdND
zMPiyXq5bYSP|gP_?Qw#~9B@3YZM7dZ7s<)q_GYP9%XwHNy|Wd!O-uMC5*j7hfgT3S
zIhaYK<ul!3l%zG}o*JkVf?@5cxYvdiTkEui)UFkneMrwrnFRD)i@7c-tA26Dnxh=(
z_9@CJ-$N9avY{*1r?U07?JgT|`kXker%`PQR4<Umb~<TdB#SO;ZkX6Z`v%<g{&~go
z+pYzJ?cz4O9eY66*Cgs=k@;=ql?qeX4pKkKC-bmf+Xp`#+vSp{)?MMOa)u_b&HOnq
zHZU#3rJqrRxc#Z>{%-gQ@D2J01_q}5Q{!g4GwnM>B!9@lv9JbyfdH;FzFN`5HhQr`
zrFU1t1Ie@>&k(}8yUx)L?Y)uwwqTyQ@7pqQ6x?T5c1jWcZkqEgQ0b*KzWDuM5yoZW
zR!eu;*G5i>s!>^#J9vaVO!Xg*QAgh8Hf2ImJ$xEELyeRaGH7UH6Ggh{*Pg3=Ga+4}
z$p8MpbysV^#xgVl@wK&tA{EYm(wKWm$3xr>b!-sJHl{59;nO?UXXc5ffQ}hxo3Qw7
z{jEn2!<{1oO)?m=l$OEsf~Xy<+5{t8fNI2U_jk+&C&0m{RH>UV)&8;+^E=WrEJn7R
z#2yBLneRm*0M3VLI4W?MfU2-nS!azADX&NnrT_LaPBX<zGZ-91g!i;GN=@R}nIXNc
z+7B};J33`nDHoPMpV`d;U3w;e73*)AE;w$e!hATiH8%bO?1D%$@cfO8z60yUd`q7r
zzqGUZhfF53e=MB`ndT5>y50z*VdOPbFV#;_twIJJ6u=IvHRnclQoFWEE*9=Wxm)xP
z>?oJjKd4}K(F2LhAJWvf8Fv9b!H3Gapkbueo!yt5sUNGw%MzS}5Q4;OcLZJe4j}H<
zaEH?hKwC$C)wxJEYdT;H8;Kj;=>oCj@5?OzgX|zqfft(fkJ9oox2>Szy!G`Cyi7pW
zQw$PjGE?tdo5D-hFs^<^obn5CZ2xQ+{(8cBRv8iOBdk|ef!>aWq?m+@wf-oUxMY$L
zM4;2h<r^Jcg_=zD(*rIJoEb4=z8fdiT~s_ZzfW0bFnZ*@I$BVz<(-Z<YY{mvTLvS8
zSp#rk@|K926HCes6M(w#-!wd*H(VV$`SjJWBW+-OkyqR0<S#d#EB~7-|L^um>BUoC
zsOj+w_}4&;cM}J8CYh&txfzc2!zF~)J-XK51>PMhygXI!F<x*R5=2KzCfmRm7u7{}
zDb@&8;MKSzJ_m@7Ziy0r<ZAZx7ruvVtza_?Lf+#|luoKC7Ri~zqwlD5G0`h_`|*$j
zmJ<oq=rzOC`0dq-+gI6)rNH><5wnQ%dDWV)M#zqF9pz-D(TM}2oGw$l$%}QY)ot}0
zjlVsb9de4=WFgeD5-Ws?nH$TO-d>qz#i4480Ef~fn^G@QVRzFT5ye9}g4gmAG;0l;
zkD|Q=i)aMRpb~%}bboi<vXM49ygMo>TB5mKd9hE28&h(wzd3QOI0L2{wQTE^g}^3k
z|940_Ur^Ds__2zJDAu1?2z}3p%|^t_pSvR6-D>Ah>`)?)qlk5#G};c3_5%<X5as;!
zu}mL~_53L>G8l{Dd5i@*E1(+6nsxaSuVLp=ICyJ5N*x^#p~R6nV+R`1Rp)b@DGWx!
z)Ipj(*^z>!eriu#E2||2#S6Kv`oDu(q|;l4l)#Tw8H+SB3s;;_kKASAahXXo;aamM
zSA2RTM`rPc6HIm5B}ENe!L2k<2YFAfqO3ATX!;71CKSLlKB8pxR9^uQG>y;Qr;!WC
zso<DPp9^RO(TbMAgqYntSLm%Grp^YbXz#F)Q#Nu5oj+@^iB}(ofBn0M0#-($Qk;^~
zAe76q&7*3qAyypjJnP(COJc{f>}!f&1snX*rC6$Q-cU$OPNP%7D<ng|uDKc)r6Us{
zCI`*a^QZuP&|mMnjw%Gd7jRBn>$rqm{|pXYqky7=J<JM|hxD9)CwuP|K3hUo#Au8l
zHkZ$^yhR5IvX-w<?*-_|b>(2Gk<18?bKx9~t9B2IBTBfR&>3GlcNV6y(RuY*{1G8;
z0>LO_TIhsltK%v7oW^TgiGxG?7K`%iP80=Tfen&*paY@H#tIzGHebMm;qA_<-6Xi6
zsE_yV5R7aBaM4;U-FZB}cru!~Jni)l@H)?ergbenEvZW9388)MiI2ksfiOgRn2Z8n
zer_<&dQnBS_DY=P`F7|s^P<U+wXccGOzn9!u<blNM#ZkGVIPktkeIx|1A;2QL9fm+
zoHqz(@wwzrH+WIK`f`nifFp?5)E@^Le(D@s$B>!7Y0r-{(8P|e%Tnsfec+LK2npwK
zc4aGEYIRk`ZSq?UF&a%2ajR9KAwpBw1zYnXX3hOS6SpYk+lW)-dK4F*?R#a8)5_T~
zDF21?i*c&?6G{k|pIey&$?uc>pa3?muSXw2TsbWbCtForx-{Sz>*C6Hx)4^Y4M~e3
zDvzv@tJQ;$(uf6ll~(*J+flQhq8E8}o2!o0NfKiNKlk07RbI}6zH3<9kgyfr4JM3v
zA|WB+Y<C2=9}1}{V}1$1`lk*%Mj78`FmJr2U!kC~krMm5+|l8eZqrV4u%fO+8=V3F
z<+IK`GfMsp#FtB9{40zUxYg(*J<kqY2OL<$EZiR)<)Tu<x=qCh=Lf1*Ifbni^O?`A
z@fRTimCeT=QXj!$_#Nc1@zvPZsj$U<5CvB;&&K4GTqqncB9jON(26$b@gCMduAZ;b
zCfEdww3W0Oe|&YB&|QJF;ZQrDjM8U|P^h8ol12l2sMj3Cb#T7seAN=BlsGz|hHtcq
zHsOQGD|pa1bw>=N-9oZdYv{N*QRzrSUI@G3nD<vl_{Yz^-F-@rhaB)UK4?WXvJl)*
z9yhc$32T20LBGU=%_a*kp8DTuPK&l#q}kht*R*b)h5|D7B>Umv?ArR4PT6q?KLv0A
zxH!CdFDw9QGlV!@ZZn;4O90cXxb6O};&$ObHLxC$jIWDtzv}CYA6r2leAd#Vo}_e;
zkKAxEhSB8z)j-9l-{4<u1e<}x*@scIC(7+fj%NFS=h39KbHZhgP=A7h-d3ee<dwDo
z@$9}v1C8_cU)2-v_PK)osR)rkapWe?QfYxRB0U;ky{+?Nq&!Sf?p2@g0l_*a-MpcK
z3%OU!z)gkuiV1~x`c`jFtYMF>(P(dgRm&OZ6GykSqhYbGTFWRssFV25VJ)4hy%3eV
zyTALj{RE(sT^Nm5zQ&S9<5%l#=Wz^a*CkGRPhIppTt=hibGW`pLV3JTS{kzk?_&}U
zP;_A7F^fKPI4;J(DS!rLj;L4<m_?ZDdHM03GuooH_alLiz$_GXg)*$Gn#I47R>!=r
zm-6?b-mH>RB&itC^@HyE(iK)!R6#iFP2oE)XK1Jc*Nqv-`C^KuZNX`^Mw+QZ<Qr1#
zknBTnRh^)Ewce|GL7iISY5h{ugKFs)nafQWL3mX-WO>}*ddT+Ar$Ml{Lebum%w7)0
zE2|<*9^HcqQKnPhhy~Ga_aA=~@vM(LoOs<dXktHVw;IQz<&7_8o4KoaDU9$C;eiBC
zHqg<U=&MavI{fP4eBAauQ#~ut@SwvWKg8C8dk8gk_IU4R?U?_Ro`68zfW3#ak}ksM
zv+bbfNV^JYPY>f}P3#9Bs%{2?z<)dji^Dy_)Boe(I@gXEcFWx^>U>3<sEwCU{7nj)
z0QOTrLXC6}oc#yH)YvXG(acKPZW7tD3JfXH(YN`kywow2>R7cMYiq16!u^^huBc)r
ze#(jtynsF3{TE`}rNt^9FdLHIm|VWyYY1H;tmQE~{d=l`A|P@g561oG+Z6)}O&qTa
zosuP+24~J{9G(k<@Fkau)mnI^Fl{5BX#r?TFf8DCaK7Hw-oiR8L?2!k%SOoqMX-5I
z?zhSlmZ38ou(#L^P;HVhBzOQZo_04b3EA~YB9VljtbK!OBSQ-vCjj<fF~WlPh4<X$
zZCeujAKU;}>n@|Zs^Sgckr<g=po&5@R+{t)HD-JtNNwreG+)#^0L&ePs-`)eQs6CO
zTf^Z;t!(}7X)oZjz-6;CgeYRl`FrXANX?+2pIv^wCnZH11Ap~tA9Gz6y-m;axipry
z&5Bgsz_;=hlt<U|Ea16xHk<Lt2_3#}BJUEL%uLZikv>^K@fTZCxo0kz2c+<FT7#&Q
z9d*a28WccA7xkj`iqA0Ege<pu#F7I;e3A*be`I9D)MyzjP7!3kM0)YWXfnUFbDwYj
zDu=|5NW69AZjGi0aiT#F3KL=MdTFb*^x(ZjFwuXmK6i!*t6jeSnfZzzU<CJr&xT>j
zT?RcImxWgHDj3c6)ONoRF6e3Q-KfYPzUmk6*?b2hYfJ(H7ZQ4DZjs2*Jp!xj1EC%$
zUgDNj@gPi=Oqv;!;3#vtDBxQQ5HE%9SETfM?*eyRMHuk(rR#6}v(WM*vLydhX&IOd
z$&lT&g@eKF6173V-}286jOh-dYu26&c>j24&$m&?L{0YU&MWTXP+xfYsdj$Iu0O^0
zklqcef1EJYtrKQA`f#iJ>;@EXqaFR#M6Zx*%}^aiK>^N87>>xCoTD>Oi~^7EiNdm&
z@)Xu(nvw=TDYlZY!%gxfuIE$a2(U(-T3O-jHHQCOo&=s{c66A@<#xH63~;gOM&E%#
zGHc5M2Q+LpgD<Z_(SFfZbO=Lmou2E%c2?y_%~ETo_R9VyPCyGdt>6JKW7NO{kB$ck
zhR){dEc6fegBchqXrF<IGG)6Ptx!7xl)+9CR05|0KDB5>84uQ+aeBORYl!cwa^zAy
zrkX-ck{FB2GRIx3KtJPW75Xlc0%SX&v!P;VSCD%Cr}m0Y6qBqv#>-|d_>3?_XbF-t
zDIevAbQ878u*j3SAj*+V?SXhp5__2O_Q&9{uWMe^q~d|>U2n{SPpbZE8>?2t!pu}4
z)1+I)73q1R@_Vfy&ATDUDNX1zgU>$9q-x9!+;T2@r{n3|GhyPXq~l$(&Suh}-dCgc
zZQBN!$Rb@g1l2BpX9sGgR}wC8#E;3(0VRVukcZ^W1n}R%-05QF91P-E%Zs!JG%&8*
zd!}#eRV0~N>}qG)@I^hmK`TCrk}$O?$@gKfiAz(A*k9#QHuTurjiwEiT6A}C7yOE^
zm39iHsLz4N6NJ>nx87L20aOr_k9IcPIEeJ_dm?2G<3$>?oHMoS^?Z93v%TO?97^2Q
z$XB;?k(9Y#x2eIFYB8niZ-kS^t`{%GyU&7mzteRg(1AoN$UHIbf}4xc5!IX*<nQA8
zr6SC=jC5SryEpIl@#pkTL}*Uy<CnchR;=wjz<(}KL7vkO6MQ_(HA6ohp$3$i-bV-f
zyS?tfFKvU$IXDd?eE7frZ}#cMuhB-0+~w%tpfgdvu2ly|wt!BKLASbKOEFscXV3@N
zYcNjGrI@7mvYeTceC!blQdwG}qJ5SK73(>!s^+hOKTXZ|q#fh!F>QzshVei^8p#u`
z);@Q40}pf0B#)Je`<iwLwRgcT8Y_i>5QS2ROeb7)TC<(`w#YZi{~aW{WsjbRQ|q4K
z61C@GHWwVDbti(kCUO9f3YUbYp`12Ht)?|#mvvG1i_g8!VCo?9uTdWV<rXpAg*%xX
zV-qX{9%$1z=~R@QI*Y~7pF>7nzUd#(Y^3qm5B4B$acwP#9P=BV?uYI0Az126<RIUu
zH0S~_Jm`)2(>HcjqXEdHnEtnzLBRuu-uP&^A340HM6xe1Mpyzm$jAt#x*LgA3F*&1
z9g(<Bg?DfFEL9#<0CAOmm@vyQe`w&-b@1p*tSd|-9e%ECcsht=6o=xUrFen(n~*+*
z<rCWI!!-ApYLF7m*?h_fV$*X&$g7#hRqz7+afq{r%1`n?6^5x<kmM7*i>>+s?H0xW
zfMWAqh4a5?z#~9D8#UzHJ2*EL1UCx{f=*ia5m|)@yB@T`1z<zXT*gS44}L*%Fl=mk
zV}p2U%U~ew0)NlfMmz_wDn026-0fMxbm#Dz<7U--k`OV)o(zkx7hfzpFPr4p_ZfvD
zB^;ICAeESd#rPa#jc{(l<YBC9<2yoM51FO3lH7eLV!#2_V0@|l^p6m{-4cZYM&2j#
zqyal$*qhz#c8aWP*sFoNkG}`TQRvCKgpZ>}{+jzduyJIuVvTZh7IQCabz&@Uk;8BW
zQVEtHfw(#6W32cU%Ap4;1(>14DJ8<^*x}G_wFaanjmFF_!}E{&7}(Y)Y>Y0ay+|5^
z-2+vo*8PICKE1dhsjg*kr|=3yjMLH(X#y6A?ou$F)ER`rQ*yTE+<9ucWgt7>XN!Zl
z-xzTnJHL(WSf@<b=R1JJVUTL1*Bk&$`W$@EIf)A#`)2%~Eup{9?Qh%<=A5=nX&jk?
z=Ktu1>1~KB@BrjGZokEAqxPkFg?0~L(}Cs4DvY9_69igb*CZ3~b_&@uO&j2RoNh(M
z6C>HyGXLLZPOsVatz8|QU%s0-J^OsZ4TS%FH@O@yk(4W7da=}-ZB#Wv!HrLisp+|A
zpVHEBK=#8g0Z6t#*-wvJ$%emX^w`>#FID^xXtw8?d|^#p@xCsA5ymINK5ephq(7{P
z?=8WBKp4;d7WK5$Rura-gr>5(Bere%0||XzLZirpou}h_WAdoanDq*P2X})hbxBe%
z;0RkA+MpM?zN-fP%X&v_7?RE(>u+p1iFH+LG2mWhQXIdG5cM37qFhEG8h4m&|M@e(
zUa(%zYwKE=b)AV46zV}5IhEBgc7xb6*)o4}&z%wrcbJ&HIckx&^?KK2OP{^&l*J&S
z<05XlUu|g8W<DE6kv243{=Db}?h_Lkk63pGC|+^!O%#pep!|&Z1sWH@ZhRmMusMJb
zscD^{XsrC=I&ANY-9ZH%?;?HF^Deg0xVQ#1)x)y@Qt(jB2$ktir8hiaf4vu|6R94}
zh~q7~BVJ+hBvbl(#r8ZAvO4K=>Rr@vq$217f$Vp)4#vko&#7~1cq@zWidzR)pnAQ3
zq@l{CtpG$~6;@_K#RCeWH%!nPhH3xpEi3TvpQtG5iCB6$Ou=>^9{qdvUX1?kIsJdu
z&0pYUM|WohpX`jmo<Ur~T+%M}7N^%_M^C_aizDT)LcxC7d-yZGdIDlw#HfSsn#WC|
zi93-0`$22r{u_;mC;4Q59c(SzAd}4YvN1P<&v=v?xo*I$dAKbFX0$1w^-ve}3LE*D
zp>^dDZh;;&x*LtpuD>y%w!VlA{*nFVsom+7-i)}a{HQgRrOvKAJV+c?_1E5q$QTOV
z*o>5a<hX6l$^dI5@8F3TC;)k8$Q;F0eo&@2%b`q@j<gdgkB61k8V2^rMBa5iI)E&>
zH!zG)EiV*91tte|xH7^Vc0RHPz$`)SO8}@oMXEdeeC*gSWaeo#K^Yu4*Wtbdo3U+!
zr(D2^^=Q@&j!H*NXV5k-8H!_LZ((HP$u0C`JKu0rwUl$~)InR6ANfcA>&dWb7&bZu
zpT}rI?F*gN-k;#zgM3QqNanUCI$;&1dYc~b)v?eUo(VO?ruDZno}-)pH8o7}&01{C
z{dPSr;kip~u6)vG`O*}eKqI@#+wzSHS`SOKRNO865fX>*+HrGn;X%lkgcwlf>G@_M
zcBG;at~bc7ZQApY@s==Fu@q^*kH}4hISSLz*KdV63P4DGno~-;0SYLdRNIJQML#WP
z!~%7E7P`U|w=r8enn66U3*23gzo2*OZ7Po1J$gM?l>j%f{k0`$(wj1P=jL(iD0a3+
znH{M-!2N^IOl@|^To|*j50rEVQMbN1;t$tMEjfxKB6q&&c#F0vu0{~Sv2J^5ueGEL
z-!P4$O)y)DsDD?Q)0p|r<F}OTK$J2y%FrmgC^$uxWE2LeY`X4P97P;4&X)`AgsIfh
zQ-cc7$ROZ=IpmBVs`Glji3r{@-c65cSbr7Sk7=>B-~`)xBmmib>@5I)v@!U|!<;b1
z#1z;q^z8uI%>TH2LhjYu*Y0s(u?pMDN1N1PR9HHBiSZuXoKI5y6``HqlLs@S`?bFB
zHG!upo;upWpSf=hp)tjLTql)|dpffk)YG9XV3^wmgC8$MrR<b7P+EgO0@PlJa@+mq
zp<+4E+UCg3$a*2ZFr20}JNJ_}16GQRzc%((`Nfpw(!?n+JQ};C#%p9nML|RLqc;BP
z$ON&rroPpk{kVuzxD~d9GueGKCJbZp_ZaVh5GKb0ykQoRwTm_FMZ~*P4!AXAgj*S#
zVKR=o>u$>Vzd8^;4!LUs!<$;fD{u$~)-s+QMR9CKS@HC}WJw*$)NBnG)r8O&lI+FW
z3*@r)BOoq8&X%I`0}}gn@q>qUOB6PAIIi?Lj&kRlzu?x=)p0`s+qZX*_CiDZJBw4^
zqUsfvmWbgky9&oYG!-hGX$}6|Ew+B-K{L~%AY{<u{3SExomZQHc;D|cWF!;TKhX*L
zws5eA3T~G*B<vQ%csCvT^!-G8tp2n7=@Zn0_&yw+gw@Ai#3t))I=n0Ux=_ftb}`|a
zAizJ1510ZJacEU2x5a>!i1`x9f0%KGo-gP*5PTqbaFFp}bAeqxyot4`XK>U?5rU8J
z<6XHyK-AQWSCUgMQZZQl_r662&+$D>Lm~x2Y!ela4o`K^^^?99b@nj5c5Xlg3r1!3
zw5lwmYST_2sXsL8LQw>@p>m3_f4YWAy#m~-3mZ2rH7)_W)MDNh{beSspj(lkNv*%c
z1rNYET^FxCUksi&X|?Oj;A4w%N+m)<^}Jd}^1-C}pKbR9<hU}PEx=8FRdGPg1{uAf
zx^U1ruZ`~-8RM?+&67|DX{gy$qpVZDEagJYwbS=tcL;(`Q}@BWQdt4-{ad79*kH<W
zOCk-89m12d*gyh=mq4O94J_sVAyq~1K{=&!M^=;OX<VSdHMd9q7_1PwRbCtYMrQP3
zFoPOPHLPu{B(WtO$Y`$#p+#qjb;@g|hQ2w}G7>(J04*h}Rs3(X+oo^s!BiSClj@5`
zT&0yOH7?xqk@_&Dg541<HU4l@hjm$E^?E8if!;>!E-1u<-kq3S+sUC|H(OdFnix1A
zK%1xAJm+>oWcfhVrSJe)=Lx4|aOkXBc+x_B)Gr8P=FiFdc3m5r3SdR)#Ab)A5wHj8
zH2)>WZW())kaPO{9EF%o9|(XJ-b_gA0IL+t<3)e&ii}Up5yxMH*>UZvscDUV)F1|%
z9}8MrD%L=RUp$qE;e&U?!}=*D%ftx%Si5(_$`||SQKhxHp!U4nCvDn}EU@=;k>C~y
z$whHa(InViW#^Mj$-J)X&p_;sJ_<{^+oEoLy0{^+k{r3Psl&%r=aI5JQ&5I`k!i4R
z!^(5wm8#b<D+Ny3#RKI4DgctI=&|PO42*`XFg=T7({w3;MPk)iT9&9Bu#pjBsK;(z
z65`&32b_>!e1A7MJ@=2DV7AM-oiMkG5s(D^+f1Y3SO&(72cq{?U9Eir<Ms=Q{M|Gb
z!LQuo<3ZhJsO+!+rJX$6wU|D&yPBD3a7@;|j33d5e9g9&N#?vFWKDRD(m2uYdt)af
z|0ZZ{3qWANDIy}#*yxs(^Kd`b17%Q4?4xUu748B0+Swx%RxWslo$WA+L;cyoVr=UQ
zzK&8=k=F&by!}_xJ1_&_w}~B-VRkEC#vtS5(AyciZn-+iqshr9hTq}lRNb-?pNYol
z_z0=EL+m1E-LI(b3c9Mdsk(XiU@7$g3<>}aMR+t)KV#$8lEYhQl(l|>o)@Ao7%|Vf
zzugFPkdP^_F}cj3$jQ@r?^N*TaFIiR_@C70xFdEvL_qH&7nn8B|0>qH48OjAiEqfp
zldZ!=N6CYq$m&m$8ix;%eErqqN(@AZ^CgiKbvb@T$mbJr&{as1k+qG1rGX_}5>Lp1
z*Tb5U2`wl(=y(}XOV|AA{pn+=`O0Z1Seh^v4X)_Hd??JB@A_lR03fZcT|v8VfN>GI
zxFtt|aDx2YhJ@e!an)DxS3|W;K^Ys=PEe$s9v`+^AVpZR+extowLksa2@mO0dta2g
zaySW9ts7u-keC2f;aJtc=--LR<hK|<oUWI&tbpDyB9lpSQSXPeG*n{+S=}V77>jHA
zKgjCOd0D3h)WJz(*=RszNlaM~(4RE29GvY&Dv(A!dY7Uf?#iN{{3@Bu2!hr+=Caea
z=B{Jr2W91%d=)jlzDS}rP@zPW6eE<MM4^psrzyja<bX{F(b^cB3{+PIw-+7~P3ST1
z-}~k^g1Z*y@dPZS(%su;bdvIs2cMP(@Yc|X1SWMI3ITqPfxblwf5bOwvLFvG)kcP8
z+13#578m9+5y;zjhi5+dzK!<@4RR6F%&*`cWU&R7CX|czb=Noxo#kUL0MJn=5g`y@
zook6I?ae)`a?BA(84F(blq*Yld4&nP;nP4MiqD<MN3PQd(f9z1v_~F2{!T@LS5^R1
z{xZ&A*l)WCtf>rn;ZKnGbW*)(JIrBqSQ2P(OaLi_zX8-h(r@8?&~JjRIg}U1H~bOv
z!+o&t=&mf%5m7wMV5LmFbui8a!GG;aCwoBsk?buq_``cxT;0m7X%GXq6JyMkRIwSf
z&%-HVkVe>eStsma#le&Hv+eZfqLPM>($WZ71k)E<H6O(PVko5-MFC<z*gHinB0yh*
z9oSkH6YgukcJquZiJgth*>A(OM1aTDKZLovt6jrE=pz)DRpgZR$eq7Fg5_)sR4Rq#
zgR=7R(d>nZ3<Q!*4)T1*i=APj-=pqyp#?qp^gN<ij89bVUC4r#v_T=4G`{u-t_S`K
zoD8L!u%mbYQq%^nP%$u3_Hd10vdBH{En2V^quU10`@I!}BxE@)#%#U-7ysJ=RRV8w
z`TbD_FJ^98ul$9pfCI#xT~dBbF|dF431L03AOHC9-4Abm59*UNUOfyuY-8rPJRde0
zga-?}O7<);yZY1m={|RmMMEA1KA;~PN}ul1cmR_UURdJO)5Z;f9R=F(-MaWzi|FgY
zU`*3#D4!b~8pOp!F|lp8&gFUDafqylr!nKIuSC~mSLe*+%_x^eha$3@i^AqnfRG`6
zE!f(H7y+2~dBqoNaZZfQhhg~c-!WdTe~cQarjJu$dj3R)P^h3ASY)o937z9MuQ%)`
ztcg6h0QPF>=uwfh=&gks*6dr<@f<-;T-UdPS~zXh+Kj`nLbL%PWicV4KaSLLzuXa)
zG%&KmX|$)GzCV;Pjo|q<-}1zB%I2dT!V-=PjO!@$WMqUZCsQwjkWbsR7cK>O3jxzj
z3F9ZyF;i_8k{EE&b7RLf5+2kCC!-(WiU;i|dN_JftaIP5tM6Xhuufzpv$|$!3-b+I
zxLlx;z~UK27dL)s6YQ5Bg4=Z82oqob><+gaFF}a0s3pNx8ksmwd9{KtD{!o|$NpCh
zsBZgqp$fO|q8CDiGBgs*{di$*L-y7~vzKFBU$V)~g5BVLh?r5XU@RX4i06;^!x>|9
zps<tc^=IE-i`x)>lgLNL%Me4jRVI}ML{H|Sx9Yypz5x%$mogr=?}T?u9M_3Qx36pm
za}?+dc&OVxddo^{-06y6=}9)DPsF%<^Ntz6&mHqcLLFoL4;c@eSs;)F{jY=12OeSS
z!u<G<2dVYd#mq!ghuH9zj#XT_jV#ec?(_h`pJj%dkU7lZkas;&=7wbeN(E&fOSzrS
z$M5=pZ0aUTUScE@6ssY^Ax6va5JdAjlMF<Pow0=jaU%Nb&an<0YrpO`l>GqCv5|Hs
z?2-&%!jR1{7)UcRJi%<%Q_RJn=LQ4nhq%MjRDMgaW4P}H?NWxmh(+c6(&acFdicBe
z#qeUsb0Kh;4u4ndeh2Jpic}SG7)iVYzi;JqWI@pP5IM+L=EamhO7?8f_*&Db%_3ww
zE~1=Lwm_}zDip>TsuVstrg1`mfGLkBNW*&2Oi_{qT#0PYgbdPZ7#hAco3~rdo}=*i
zDj6yA#MWoY_51g`Cb(KC%8xn-_W}S4gKLz8@hD&3iMvbrc~U&%a%47t0d5rqc<;&i
z72bZlX9m{Lpq+_w+tbo;u=p>V#?)ejN2TGm9Fm{u*{8nXSa@9(b2a$mcDBf-gALWj
z)@nP8uxWg+!$}hWe%u{s3!9Bwt~nP{3g4a^el6~%IYl^sRaCZ!`w)jIesJfYJ(NbR
zfm?8y<T#s*F->IGVW~u=GOWq$?HefyZtW5MbG&3(DDGyaWESdsYhlJ`Dt#Ii4+u0R
zY))U_z4O8`<VC@~T*+p}Ut<h$;rRlU+QRj0#m$C9o0!AUctx30SVg0&jup7P=;O^a
z)VLTC{T@e}(3@BO-6udIZI@Kl7X)Ge5ZRU4VZ`FYckjkyAwPI~Sg2w>=;`d4+TTN1
zQKfEdSaQ^_5^OK2;BbF0UYdX;e1kk9!Pbzbe}2Mc&kCcMUq;~!h?wIB>YRs@>0ToT
zGLGb|N{6^)%@_?CrZ(RdqrWMYxa6t3#{5q&Y*B1<XD<_)ZR0!2_E#c+Oz&vWW`Vd=
zj>NvX{KPecL|GYjXgqeN=Qf>1MAN`Q`ddJ;xccis;DpJ)qgfA%>qlN6o>j<74HNhF
zE{CnAgi#%%P#)Q&Eczx%oujILRNPiRp<kG2I&B3|Hz_qd$zznL_e7>}k;Yyb%OV#u
zvyXj~+*KgcZ-f;P9M${mbZiIQ%GPgugtZ7-OiM#9Vw&o(-$!Q276}-Vjs&7^`4RM4
zk*p0qJ#P{py?b@y8owAJZb8jK;utQ`4zl(`GbmZ+>z<{6{BiIJL=i0`bXlCCSC&g|
zxXz#I0(5{W6n>VPi~7ZP6@&6X+}TO+{Z`H;6cpU2j9}(s#*BkO_3LwC(E&gUecQ0d
zU4T4}Cw#M~unFgzl}Zz#Ntlc8UYw<kS_~u9m$-tvi6iK@WVT!6z#YGLHU6tD)NJIA
zm!%sH|8m(G)B>S0C^S`W)EFETx@)8K3|J+Hphgc4%ova3B3w}jDNuWnfB3_LR+Dbx
zej-96EZ2%Gz=J9IphP?2&)v1drGWc`iT?n({@f*hj2}2`%+G3q8wM;ueRA@mb_&um
z(1t>*zWVSEwpLQlwbQXS(Z$0E**-ERAq|p$XIjx74a(o6-HnX|<B^gN)80}JQNR$l
z-9)Sxru3QKR0(yEHpFXNOXJ=~-p4n#TSLpl0#K7n`Yxg&T7Cp~=evYsNH%b&U#|qk
z5sm}}%Yrar(urhMj6M+xADZ1nlD-V&#;Y=vG+QJfttR($vl-_w=76+6uUHln`-XC$
zxV{FwiB&3ac`;TAK3}QZaIoC=;P}w0kwtC9ghWJB+rY+Pyqy8Cq_|%8m({SNyE^2!
zN9I8wLD6n2FVrALC9Z#LICybvZWB(|LE@?^ELw=)SNXf^1}pN&V4>!D4W`?pT}2C8
zDKS^WDKxy8vG2|tT4&5gJamJl5-mY6uEoAjhqM<0cm>`d`NG*&6BZ+>!E~J(y&Bai
zOyetZidVi71{m1YrzWN#r4E-(BXtq{zo3n@Sq9)IG~<}g_;vQ4H?W_wTU+?GQ0+o|
z<yrZ$nI2g<nRW<#9K`I;d!F!+3N{LT&$#N1ym@dVDoB{de>zWpSzi6xLgdK6M7W*3
zfsyPTXa7L4s(*RU08q}e{D@q!wI7`Dk#KQ<P;>zzVZSHT#3q6{3`G-HDS|3&<VOan
zb(a()#f4Gc>PP#iwu(hGaG4|$f%&shWGD@Fz!C!|cG16nOz1eSz`bISBAV98m~WP4
z7u;{V5lBjqsA$#aZ)aN`t}iP>DT3pQ?YbLm7IXz=V^E3Al-=yEYYIYR2sQCVd?HK5
zJ(CC4{Ii-1+-L3@c78$i8q0eBIUb4SGbP0Uj^1)Cf{%L;KqoeI_n==a5U7Vo0&eZO
zx43%JR)P|!fu%0#+c2P<*XKX<?!#C=euzMNpzHmOoQc?YA@b6rRtpDqBwze*zO+3*
z0MpgmlC%GF<(chv*v8<2kRXNfDN3}%enB~qXmtFfmKYfB<ne7T`dE_t1x>K!e{gud
z;Ln5e5`ej!KBQ-TyOpaCt?kba;0lx}pIc-mel&m5t9$2N`%>T@_(hLpw;V7-NMC<h
zn8LBP-lKWa?P)B%_bnsTNU$(0Em2o@%bs0A_;b1w`~>hw?>)1Afp5pNO0b6)4u%rD
zg5;=im3LA9mPznsPLbJL=!a%E8IML0SXBG|_yuOe?}Z5A0N*3U?HjDaV=zZHz^ko&
zKLtUv<>v|q7_;R^)UE!TU&i4KxTMSpi!Q#Ba0%hEBXQ`1c*~LX@w|FAp5*lZ#&I5q
z9{(jUurxeuq+cy=nC<S?05$=Pe7@yJB-eg!&qf4Ae*`6-05%C+q*cbv1}NkOty?Q@
zeLvxh11a=NY}xDInfCGITdoBGvvu|Ai(Cy26~C6>wiY+Try{+8G50qt!I%I+b7|vd
zSJS4I9w8YesY0d^lsO&!tz!luA)JBWvcWj$`*m~Y3-Usoq_`P999`UKKi!+QzPOGd
z4<0!{K<~Z4S>OyNrp$RR?<yK`U%qfBvn%?2Pv5-;G1?h$WIDrY>vNFD+#fOSKde}Z
z=&D8>xQv+k+f4kP#b_?lz;+H@SF4Y44fg%J4+$fWC&+nQ7mA_JqKod*=u`;KRyec?
z&quz)P+%78<k##6ObyY6!}hBz#k3cY2Ptt~p1l|MJAd(uF^JrV>pK-uRSY9N+gInO
z-Ex3cp$GGBYmR%NN032mgzEGc3H)l2CFW6y1Z3!_Ht}RY%uyYJP)P3h>A_(4{gy1O
zbx9W%Ov&H%ZB#Q=)Yj*fRDTFHo<UY0M)LQ&b}cKZ9>>0yzz>d|BLE<JdIPKCV15I(
z`V=Yq`>u?rQuSrfvp~8!<pK^}{s)!c<`gv=^sT~~P5{9Z7yVsBMUORG6OkK%Q;hl7
z*8fOyvG2|azaBnu(_tbzGrsrS8Z_}NpYh3_cK#HJfDwu1Ot^6N2p*1S*8VR!Q;ije
zlAEQ0?mvzg^ewjUC)}jw`3?p^M$X@Q+#X-58~=nhqev09WY^+_0|A;={j14|E^Mt#
z60=M3{Z-Oub>tJ#i}RgSsH<Wi2YuH&j+XmwGKj?wAdyAmyYT1%N9@-WV-+NH*5Gjl
z%Z~$Rj{<K9tgMx5eFw@7`Ydpq+C}HT#9{QkCUO94|4+|;K5S)Ry~9xMXC?q~|Hz0x
z_V6aMdcBk?&H0A^qy7%Rvwj{ENbW2@Q;ildaRY(xl8(H!9Yp}tcaxKOd&DIKfx}XB
zl{(ix<kJYU-#nfmd*>wo-L|eV5dl5Hb|^z&D{(dH@3h!;%f_n-Kup`C*#>7SAK>&Y
z?FvpC|F=^4oFD((>Nu@_@zEE!-KVQ6nwmF!@aRtM`ThlB&lVn%j-lf@M<o8f55C%+
zOCXuxq;`&1quU<bbvC)ls7*JesEJM>`f`u-hco0k$y2mbiK&3woi3h9I&wh!ApF4H
z|B%IT2w4aL$^OCjQFv?Nj_xk<p^ojm(yMq>2p-8&jPHgYLE2)ZwgduBc{<n<XNUCS
zUy!=+po(6QHhs?2%=n#-8|SB7`2X#_!-d;9R*_(58CQIz6s`4D&m-_qhxMh>Y28za
zTrWyjyeFa{VBO1!gd0p_nd#Z}V41|4+zh&P{yrOy@Bi??L|Pi2(3_TqM*-825@-$A
zq=&=vXCki^m=gzW7s4^pM`3&@|Kwx}7;&!WOey#b8{%8*V5917?iJcO<Vu=1s730F
z2X9Q=RJ=~O3?A>EnN&~C(U^x1f<cI3j7#dmh~&34<kfL%c+tos@OT0aa2;eu$ihyA
zWBugm|C!`jeH(l!RwwTEA%*NM9tF6_7>WvldfW=-^iYi|+=mKiaJdfbD{8v6LS-#p
z)U+Y58g?N%MC8uVdg}q_vW9X1asTBs($mr|qZ}0CDBx-!E$yGs6=78m<~(0tMN9^6
zwIGZ$&DLQ$K&+1?qoWYfi9ItpGk#GF<x*1p#HTRJVapxB@OXG1*9EZAd!AQxqJ~yT
zJZ8h*NG%SiIuA0Z_`{n-GO<9v?0iQ4=1Af#LFPAMbc4%4L0j<2{*02RciuI)?L#v0
zyNSmzVt;-&@zur3GMPR^HXBsb90Sx2na<yLroH5M218B>N+(p5;4c9|0=mt;2=s8=
ztjx<_8M+OfK3l)H&!`7|_k|BvK6Xi``VE&C3?!a(F+wOG^p$Rjxxhony~AljMq-JJ
zQ|6a!*_06#`1Vb<T9(O<D{7Zow6CZ&9Wpt|U@k`xhLI6IP)GWyMx;&0J*hPOIKraQ
zJ=uERfT_a2F0ot83U8o6u>(kjV`i@B$G&2xl|Nz-6A*BBc0l9+B2hY^dN(5HaiiLy
z9^YvzGw@WI(9pwH8jF{62XMMmP0EtxLsPKC{WxBX$4fP>D1p<HQI3Sa#}gEHSg1!R
zF6e-ywFX{tbC#$Rk&h4u4+|}5i>XL3cir+$oYycs@kj0PzE#8$ksZZ_ZO$A<Hi8oH
z<W=~}J-077N_(L=yi~iTx}MdE<05+1|BsTA7OO>tTFARq%O^X!F0iw8oK#qT8#YGz
zP$h*h03~o6;u37Pdb42-SiH7JfyAUWzkez&-u3gEnEGYZp`Wu-+!oJqu92C;hv>|4
zyMendBx3-{#V7keT_JG~KYz+4wk&>_fmyfsriREtxJyN7GMwmr_gsl64tE$)jA8Ld
zDQXKs-@{(n1h{p69Tg6F;sVJNwx%NA6=n=7nLsbj1e6;$w?|9(YBQOq0h|Sw0FK%S
zj&5R2t73X7hoI*Ta@1|cp%KhK>y)?2w9EPjOj(aYaMmQgEXvis+ccJv^miIx7xWPa
z3kR6^k~IQ*aJt4}Od~wHJ#OnX{c>8P7?Fs=WL%R<SO?n`E$UMVK1268kOHVybgX{U
z@!_DI!2`Ari(zZoYPRrH+jy_xpZ{;RHqX>vgnXfsv%8AxpOzpgn+8RrPNWU#iKgZ2
zHErLWw|v5IzY`2iZ|moZsM*hOc<Tvb%b*-EOtexHrUx#roZT`uu><Yl2Ibi~g|E)t
ziZCo*su2S!&%G4ocw!D4c#{e(6%S+)Abuis$9Hk4B`(VC)@DTg^7i>+UlY*-q-8BX
z5>Nx)(L^FBepHqvUlb#}AYZ0kg+CB$x{&07V8tw_<?>P(KvRgt7C{K@Y@(@6fhTPk
zU_D6x{BfPS7XF;I6k4=&0NKD{J3)YlpOI+eJ_u{7<|4s-c(C1o(-px>y^V=Tl#LUq
z%u67#;p}+Sy)pvn4!F|ufaAtGYc1cSz6Itpw?TB$T5~>C|0j=NFcBRNtnDp<v~S?O
zL(}dxzDY?7#ujZX_xvPjnrW7Q+s)Yc>aY7yO(pO;Hu5C<cCV5v#;Rojr+?LXh?{vH
zI9&VC5FRetKhO@lQuL~%0|aVy0~Aq|g`%0H=dxS(A5cBD??#UTl)zF_3KV-j*qbMn
zNXq;yj)7Ul&$w9$@T`&Po<c`GzeE%}7}7I%Ir|@gd;XihJYRoySM*}e-qULJh>}cE
zl}mzFnFF~bT#1~}pkyliUlhRBoVG6#mbGllH)UZ%0@z;MTjLeETf?DbknDf{afGK*
zb}^~GW#kN=H5sX|dWQ<A!PP$%4#A+z-fYwI4F9?VOHRApSfsImIypdl$E<4g_JG}G
zB_v1J;{)LpOI%$cUO0*p2(adP8{y*se5EjfHVx3xBSg)TB2PGUpnVi<nj5GuR_%ip
z^}_|VeB@_gO*2kQr#-Ua?|=CL=0;?y%#-1B1O=~4ntD(`o$#_hF5hvbt>n9F%3TzX
zoolHSsuVQU(s8^Cg4#d}O!3lh*Gjb3+3;DmfU^p3AY#N+>jQ&o$)tPG%ghPHhg&YF
z8Imi)g9$^>Z%J@*0Nny3k;%zr%}3w`+{r~2;|_|laG+g+V`*SOz%DslA6sM{9aul?
z3f~ES)D(ecN0%V;>|K)#2Ryu1o1Gl3=<+$0JS@#fX)&moep|gROQW3X>$ZEtK)A-E
z@CH`r=qQmVg<dsf3ha>Zg6;AQ{vK&L05yUwT|pI|v%zJ=SM3h%({%P4l#PpF8y$ru
zrqXiw*TZqR3)Ay72zWWW3&CzfeX~Q<u8bW$cxQmQJ1JIku3-vmAB=j>@`a%5MG@t$
zC{ETIK0K3uKn*&k1@v!kMdUoa&7~M#+^kA_WT_(No{Ce*zFvc#hY8AX$0qWOv)sQ1
z&dtw4w^(pfl<z^LR90Ty|6!-VEWG>=SFOc)zxFpimHhf+>K(mro!q{<;Lge%vxo0}
zh8=Uj90fEQ8qTfGZj|bwy5#V7CiD6nZNNDsPwFi=#sk-*e_MG(qtl<ojn$qq@BV@Y
z1wO$Ks~LeizqLpH>m4Y^K^shEIqBhmy-w-xZUHTiSJ=$enM7L^>|mxcOvyB)ERQr9
zx>px0t+cVwwt5{A^Zo^2tmnBh65fM;X25x_&Ka20AkQrE{+`2-CsG)&>pHR^f<vn%
zK3}b)z1z2DQIo^)94~0!j2gb+Ac_GH@owYTWw$GCB!a{|sTYO<nH<h(w+7Ro{|=dd
zjXTVZQdV|_D`T^2pGKbPQ$H`2glUi--mWzGom8TZ^mzJtaj5!37))-1_~5Q=)Nl1+
zx^?u+<TCU#cI&DT)zc+pN#4uYzr!Tj#-_3^_)w60>)-yX<xJc<hwp<O#;tR%;XeA_
z>d*AMdRsd{v`x_W8-27_<I?NnZjGmF5{)bMmUQx|(|=6#_-yP*f*J(+ez5c+^BC_|
znN&w@53EP2U%`z0DV;iy(Uqkrau$FNF+NjQ{kVAbSIWu2{8s-<V%wHheOpL>cnP1)
zw-S4AzbHizw`ls$L1=}1k^>WO10KM7vfnMRdg2q2d7k!?yo2|7wa*>%MR(K`wrM7o
zE@mXDxbG3iP~*Y}<DCv`kUG_NhZnN_m51bZcmZU2^RL5{+_tU_ENB>8jCPFQyRs0W
zVgDK#9sUxys~}_$yZ(7h3f|+XY@ED-6))5Xg>u!SlJJ!e_UA%@vO4?tRU7vJcPLRV
z1jTAG-QhE<qHp(?)Vdl~9p~O0)uZ5j4^LL6cd{}^p}ip7Tx9wc`hG%hvIk2W#Le1E
z)xbpRppcj0)fV9QkzOr>hj@Jl7V-5fW;xg(dJibr1>%j{S6kq9tI!e$pBSGQlE&No
zYaB2XOup~LvN^jVI&TRLbZi+4=?P<Q{E$Sfl+#|xW{dl(fx`M!I413z7_w_scXB+8
z)`~ge0vd%*^?OU(j-r5A^FJ-(PGP-R1p8rqmK!4`hw@Dc@>G;pLc4$Ft6}nVcC$uo
zhgSb`a<LqS(N8XwD%)wR<*=kIt?^CUq_%S-Ku5n*DAdTE)1QJr^HM-LJ+XQc{YwD1
zufVO5Hqjw<4!rKgG(Lansej-;a5e3AQT4`N|E<-qlez({FKCAP>0h_HeQS>3_*lD3
zSM}zMIZI}~>w61DnCwaZj!R+utP6F+8RX6&Td@-EbN}a3IsIxDODvVaiq(34yxIPu
zqb=G#q)9vb8EW;z0U;RD*L4|gfnc(?#4f4XX7v!xZGIu57|eha2zHe6*9l;@7+)pg
zR}Sl1_qchH)txJp&IWW^gThX}8t%ZDD|Xp_5r6ll$3N?wK#Mx_X?X0?NHxr-DW@8J
zC!G9!x<Icw#Y6Or@53<{nXWiUA&O6~MSHVwipTy8XUOyLn<QdK)xe5GE`r^VtobF`
zTC9_UQvNY+<WGqqKIw>0yw%T%pDWY~`Yf2(iVdxDBw1>lM_}+`%a6c>x3c^XEJFYX
z<TY?qLwx^ZXAA!{t{_}o@QqW#<YeHs!2=S(Nn#wvnO`5Kwea@CCm<mMxJaed!O}YT
zrd$2z^Y*@-%+LxaAhV>)CK(6%jk&g+;>QG91z`zV!b^$;&Z<VTkxDkp+)?BPDlh+r
zC##mQ)}Y>upJN1%8Xp}7@8%0GMV7Q5Pb}5m*fPQ_ti*kH`s=P3nmukl{`+`al4Iys
zTL&rR=ksZnc%6f_DXU4*JMR#HO3ZZ%!rfz2=E9&w)5z*G^gO{TDz?)2!p?Wv2zc^h
z!iN7~+|vrf%PlyjJS1E{gIu6<doWSQ*jn}uKFF|?pU0v7C3(+>_qO1n&+Y}|Gr;}-
z*t+sSsPq2+#ULwbwOvwlNJyt@bTC%eQb-}AiJr1y!bC_6gMF%}O;V=QF&&mL!^*g0
z+MX3^kusBeS~O~0k#iis_vbrfpJ(>@egEhf<MaI-@8k78Khx{gn@ZIi5pF&hLa7$#
ziU9^-?-s`&)vrLrluFKyMCoo8hdWCO8|g-{L_0@yhx$vq(Q62-i$<bP@C)axAW)8i
zc6dD=CacrI`u9OE-Nq_$e=ztUVf1rT68Q3lQ^POeuZRq-1E)7^gq|qr$~4D9bbZ6<
z6O?J07BH3gJXg{C(?cVXnqzApd>4bHI;gNUD=S1@HqKQ}=N(6%LN=@51l$?+qgBM5
zf4mX^u$mmNQ5a2q!9ewYW8+)k8_-i&X{QFhaVE)DK`5_;3x&?4J@wUhGYCt8U?z?X
zA9jkPAAl;1#JAYFgRQY0VOOY!0IOlVjs6j2A6>8GNQhCsTFl^@uQez7MYuShX|Evl
zYatWWGfnlPLG*o;Eheq|{BsZ>`a$Ii7y=&HKeb;nw%ViPOK;xiK|hNZn0%#hF-LmR
zITvPf4eoDH!KPj#ZwTY0j9u{iC9@C*&6)HQBi@JS#Z9G<!7b5NZ&55mAkdk6-A(C8
zoS;zW^>xb%N0pnv{v?k=T{oyt0eq9_t>gfr8!jMwi!xZCQ4T%6$P8)r^dT!JwO>5$
zKvXyloc(cq5q-w8qAy-wN9kESeAoG%SzD&{H2z*$LOj-$aE36&S$;HoQ|){++VU)C
zv-gJ-Ho{kEYon36re+K`ZxWLTVg*bGcNvoMJIwKG5aBBIhJ}tMcW#|{62xJxn=9Oz
zpYE6D&zS>LUQ(}k^KkVpm_1{i%;AZ}2b{_F-<r8p#h6M|xi&=EFz`VD?;(wV?h}o5
z28)v5f1>m}3kN75q9yEf<~(eq8^VScb_;|THrYQFKksM|tW6^bO&bIn@Ldyt{?!48
zMRD-&b9ZGv5?1ikcbN$qZ!SHrg3@`?wAx;zVWu__)qw{ADOL$Ms`E~_lm`du0_B*9
zk7wA&g7TQXhdZ^~@|z>z$0`)pw1XEtJgGwjzXQ(tp`#Rwh@Bk?@C%G@WNHtKz8OOu
z8>c*PgRT!Jg#KIwl?}V2;^1@{&C|iVNZ)S&JDlcIXz}^NvuMi!RJZ2e79SWEWdPrJ
z&^2xrRyBEV%E;)+KuHEU`x}yT$7CfNw+)3Ec(T{U0NiblI6&4^3jQ^^cdPyMlP?xa
zGRxGD_Fm9aKN7Tvkr4CbWhsKYjH%)w3~V9>XVd;=7|RXr<oz)bCQWLSMWTKSkGTT+
z)5$Dfp-Y?o0S3&?hQ4p#2G)!P^ddvxuU%y5R{i$eLgO1KW^&OBS(F>Ls9?6$>bZ!E
z3gV5PyH`W58IZ=N{iR~bHss>=Z#%UCmIjw|e~us+96rP>{jGbY3ae=#p@j~Do|lc}
zcfd2LT!m4Hvi3R{V~N?BJCeiCPt38d_`d3euyZz<vuX6&iC51rT~eq}L_3O2-4fB^
zFM&f($eG@o1nB25ahN;U;XU#IJ{6_&MFx7;ucYtVrd0t>6mTofV0-YsU6Iu)q3t;-
z%OS?P(q34ejqtu&ZF#;;+357#fb^}GBYlapN!5b2k0;mR%VaTkI=jwr9rR~|PG|Kh
zmFm!JYc%-M5Q~y}4O+c)74NL!<hbqVo{q9kpKMN7ClLQ&*WAQS>15a}M$o{qvZ}Fg
zxFQ(_*dy~kU>E+}P%&%=@U>K(Tz#?ZOs)9@xjo3FVZ((*=MqZQ_j0z{qjl(Ynn5J_
zISfO6l2=nv^v}a$IL&cRBdc1v^wjO$uf*F$y<aVd_XJHQ?4CA(2IQ2Hp6$RL8_q}}
zq|{mVpLburJiY8Ztrz?01+Vj5H+K2(lKR1sPZy#k$vP_95hFF4^o5MD?rB&s#_B8G
zvF#J?-{o;Vc98_pI&;Lp-4J2YQ?)z>r)qz0vSiYpd%<XQo8!2C#>g|puZcyV;ut4n
zjkfmiPkypM5NU->VPVJK1xL>-+KTl#DH#$Lf`9<svsE#8m;DM={h*$mFmrGwMc<fx
z1Fo74c7)T$E!RUPL}}w0g>a*qrA1Ry;tS7Kkeof-rnu2D840MwhvoulX+3mNidh@*
zGi%e#3uoUDBC=(i0Nj0y{RSz21J&8Vi3-0%#ui5`YJO5h2g>CY4(JqHKZ_+PkFOFd
z6mzP~=8y!a-8zK5AVs+4&yiQ+KxPfvm+&8Dd7Lhnwt!}4L**lo;drvNTMhghJq3A2
z(U}AkvlZaYqXo39MwSpU$@kA-T(+rn2hqZS9$K01$$OKJJW)MRUmb{D+*=DiQ9v+L
zS&bnre<h+!#LHwGJ}_%UD&wy$Uj#%|ur<0v0rMXJdms+7jR&a`)A7fRHBU*l47_oC
z+rUUh&=edXF9fg{s!{pNFNIl1`zzvafFpS3s#&}J_H$SGF1qq;E65X}pVYzkGA*bK
z*o>Cmob32yu#!$@jvGR8L%^(>VyuW=HHILm1uBL>t=o1Af>GTvNmu6fl8%4DPloPP
zXlQ?&JQz2Shy*}U;n<1i>l{NpNznk(I8p)hTB~$~9V+^hFM9nMJP6PpROx0$&opkb
zBGDyun32r*JDxfFU}={BLi1m|N)xlUVM-nvjogKwi9jE<oE5wO5#XqPO2N|yrs6v$
zL8ZhHBZZ#}4)+|hKt*S`8-<r5H*=MCespfNqdo=%2SA;LsEAkFIcblEkKBWXPhXI&
zN4#sEF;D`rN8m>~Ju4^zD3W(Jr`21ghT*RThf;OXyl0YM;Ggg;x@r`u;iw2H$Bd%%
z?pP^2gyiTH?-46UJn71o`8?H8(LNwz_0dT43@{}L0g5`Q^+S|N<o_u{jhOzb;h=fV
zNE|da<9XVqS-Ir7z>fBI%*jLPTDw8dde!$vrz<ajbI_s6$QrM#(0~?c;nnrZiU}Y2
zff}5*Fu<gP3DHc<Nly#~g6D0N+wbT@`q>X;V{;>|hkrnsq=lMAfWiIHo%RfB`hv&K
zJ!aU%Pemj~c+pVdQ!!*{LY8hwIG2v{FA%5$t4~%mh$alUW<xq*+YxF74DdO9)BdR$
zj}<x1|M;=aQlJB&%u;?StKFRaGa9~1a?kZ_BK%g@C7Cis*PTfZR95aLeF3KOh)j!i
zhy8}^#GadA=l)`CA1$#40g{zo-#6>PkC)REY>pHsL0O4dEWy(ra4NBoRZd+&1&7s0
z?aV|aqyx_`m)edA53TjkhC4IqHxR9$P1#fca-b{2+cGuk=HiD&HN_6h!DS3Suyxn1
zvScw3A;|{?J-_`GHDwj}UzrVB2Lk;wz9IY#vu&A}DlX$9GPd%*0Sh@jQPbHh7$}Vh
z*N1nR!_OwqiN&FO{WLrHjWmd@g!5f-73e-o{_GCx^B#~lplRAMF=eF=1!Rht%QpCE
z;ra}J^TZ*0$IlNT323qzoHPoaDfL6oswP1lVG*we^YHvD$X7$TyYBnt2M>@2Vqb31
z0~}D0heyON9)542fLVPR9v<eS61X6AFE}mJNRqS4U15bZrf_Zf^?5=fPQVaO1?vPC
zwP@}6GuRdV2h#JLqH}FM`WGO)B`qzJ-VD~_(1Xemg-7c@2X)u5swJ$(A!Qa9BuZgt
zAKveF`ZfH%jvvmHAHFNBVb_U3q(ZARsm}wAj>h-a_+3RGNM5iN!nU%sU-vcX%_B$D
zR>8-_dbu1Y7S|{+a&d*>oq4SL&}RjRdN!rE4bOBV+kV)IK;-fnNQ}Q=wk=mrRS9>N
zLi&7;1=1#%@ZKiR2UQ_oQQz$2O3=bh*s<yk>_<N5G~Q8j@-%F<utqA!ZSi<AzEv~X
zg80Z7Pk1W&Y-v!U0#epo@JhPrrzs**A&6#m!F$GMG}*96|C2_EB}I>SmR!I2h*>I-
z(V}#3(b(cU$5OJ8{#-X>6XGLpI4axe+0WgjL-TS0Zy@WDre6I|kS2|33$DWrDhqrc
zSW{S{<Je2D`3Y-<;00#ghe@w~c%1I2g8FmJ+>;`mFsAZ~u^V=%jH_)LmhP>@fiSb9
z2`i(@4AcrQxnnGV$tXdvE_;1+#Np0DXQo(#L`4GY&k@Hm01veX>#OSrN^<I_fF7_S
zk~GOdI{1d8W~5wk`WAn9?a-0MdTu8B=VHPmi{Ckd`U--#T0C(|-i+rQViZRhJV{CI
zBY{}W&g%ON91F$mHE-}lT-iBda)hoFBqj7zDWNZj3<e}>59UPU*mgAcBkJW(*GJ<?
zJm778NT%%|8RwTyLvN8;FzsM4nBM*+1#@K;jM8~8LNQBUNVCCJ5hsPri&7ZCNN#W~
zDUy14=(OOmcci*h@Yg!lCxt!%C~BKg<Oe-i(KLE=BA}_ziqAtE(U!|9#HFKSDjDP#
zzamu+&$&nhHCXF6W#?Mcyi|T`y#0J+VDgYY$pt43iU1|;lcyWtfG@7C>C5P;`75!H
zWh}QH(X#HXS7pX96iR?XOV@xzccK%M#hHoEGi*>}T!^Q@8H9C3fu2taRRDz+_^hfB
zz!1qznxLHU575&q@Hg<B<+WOdR=+#qeqI~8HWQiLoac0ea(o=s*UnFv{Ss+O6pDjj
zDeD!vKGeC`4KQlgeOzi?Jd6jCVOy~O9u})I)mw5EP6jHXV6(B@52YZ!x_&kqBD%D#
zA}+q>+%xg8YG@XV>)NL2A#Gzd^%O>#{X_3d8MDXX;+Z$&tHC#zCv;r0hRlAm1h}`(
zf9?WwwVT@U3f?*(zw}M4+$!@-GJW1TOBDlWvzPV@XVDVSMtzZdQve)Uz?Mz9BfK83
z1)h-AG~h!6w{x{wf$NaV8=lDyl%65nny%dDfvu$#8TBQ53-lRI<a+v!GJG(pYQ%2>
z#Z@E8dB#A(2T;K?HF<4PaSV$u#bq~AMagLk?2Tv&)z13F5J=2gxk9>PtSA%Ct#NWC
zoSP$vgC$==jywJS7ge7Fy@Zd9|G}6(tgrXl9k|`SYX;}QWn~1^`%I^b^-m+wYlv)4
z49zD}m_<A!{6@$W!Qj-dWE!lCw57iKWzMxKGjkX$OqNqB9h^7)r0&jIB*GVMkG>2^
zwr5d8eA*JJJ(P+vP54yC=S>BVzY*<5a*1`!ga_YXX^Tc-gjx$ZI+(Hp(L3X%#KxaA
z3eN*@pL-dOnzp3=vu{GviyC0UL%S14ZUWG5&n@CO7bgRKrJ;88D^}n%Nw$^deoh_u
zZ+Qe0-Lz;U9!oz)o^vhhd!8N;?M#IPHnE#as*~bhM0b&&6&0?%{akb$*J`XQ-V*lg
z15S3xj}##%E`-xZ7GJ$X_zCvW>NHR~%~b_8Lx6ijyqE!<vkxr8WkzIanv?EPlQk+r
zKHAn&f0SSs0azUy*Bf!h_X-@20b%fguttuvCz5qc$sR(_IVkX6kZUAa0p9^NxR;C7
zc`IZ=Gp`Xxu6O~D4J-K~5V!^)AXaeIJF-$5e1v&n>9VLa{+k$$RmJ!>p2G;=WF@C}
z|BQbwTvQ*`Ls-Y_>>SL(Ne@zZG?7>vzla}51d`e2X&1uXIun37|0eBukwyw9R93<q
z@drvrfX{%ehHUH)F-HWr&Q>BRvpHf%yil{q8&UG-#ApxvjA%P&3`qRfRsAt9fwYG(
z5*7G!j>%IN+RHREBF?ARc0`xLkBA4kgOctYLD$fBrpm!K8mrNLH|l-Fvu}XdI(<0?
ztSH=`p19IE!!_!Pl-F!OG|vKRcN0)jVY8|a`&j7XIXTs`iUjHFg(R-RV`&u)d|)DN
ze2f36mLj8d&!rQpR`zW3Y%*gu$(Y^y=olK%_U2GqeRUt6u+AHkJ6j#3P@qHX2;}R(
z!Rx@K1I~x{j0|zbn<@gKr>`sAFSt&um_b-`n(nSrh4uHr;xm*{7x-K!A6=$s7&jrn
z6@|g}{5Q}09l?bv@bAIJ?_nc(9y7h-;GBCLL)#qH;}r_HPJwNU&txSbAxVUpN?UHh
zXsqpWhtig&l~4jpl{&3a7oLQDFzAs(Cg&fJ8OCsN%HX2N1guOjUbGHD58()Kqxr+u
zDJhnjNRXeKG<L2dOGHEq4+Ee!@dKB4(>rn#mX8v2p&XXoIB5VgXFKj_tD5Y@VLlhW
zOA3^e8>NTX4;pxA3-IS+H)IK{Tp>+k?F7PheGr>YxGf4a$qIlnd8ZNQ5PB;}cYYfz
zfUz!}KmB<EW{j<qrD;k>;cZKwz59TQ4`LzoY&a!5=mM#owW!EDFsfehN6eQ!H0j8d
z9D=c^O{j+Sf<zjmp_$3ea>^Y)&JT;wr;G=SXW#^E^^SDqD^BWX{%M{*Y@QjEs8qn^
z3x;sMlTPu;f;$P5|CYi)#AI|?dOJQbOsI)!IdVZd&5nsY0+_v12NZJ`-iaaYso%FH
zSW8^r3!}iO#R&<5UD%WtQtyq>!bk58vI}`_13NRW{N*j|S|WGu5%wOWPqH`wdflXc
z7wkw^y7zN+cGLt^sfW5b756GMlf~nGv+Y5qz`lqeVIRIE9FU{;iRZFndfBJ%j^OiT
zWuz;wKcepW&*h#=vqREl;=e10Z|RIIVO}kAsP6tZ9zc4vF(*8u(QG-)<A|K$09&B!
zbYuwuRLeO#rh+55n{-2iyPL>)!^-Cq-w2j3zwyoP$c`iXj}UtV`dgTXm$y`p&Fj`$
z^-b~Pt6pjC->gb6cx3hK?#SKqZDuV$l(>HaI==PURiC%5{@1)FYF3ZPYwhWe)%5oB
zeb3<HtFp3EPoJfY=MhY~#~*$*%l2aLlcm{(S%ey_|D;+`Q9(gL+L!4*<YG%G=JR`8
zU3EcCF>c8@I7a`;*Jh$F8$EDbj662<bVUE^TGtNF4VLjH3onnd13j39kZ!gvAR&lS
z9m-Ca#=M_!HBqy5Zv1f6Z?F+_JfP(dRgf*c8J9MbJc!N;ZqM}GuxPfop#NL#<v=&0
z@3RSzyjH#qej@QQxy*5*>Gq#xFHL19sw&vBheKLI;QhpHcU4iwA)nK^a-g}NM%3YR
zkmei56DrZD=;5hyK_Eby5tj=8YkX%NonFiO29Tap`8IY8G<jr9dHW>4hK?ll9EH)s
zMuqzla2fXpaBIX6z;;0mZZKjBpP%7fgE~SAb>z2(<)mng46Y#9LDGG)`W@JjIqiGV
zzL3=(7Q_CoNlBdCN2}I~ztNs2CaXVGAab9(Z#XZx;N2kbp)7^b7qZz>l*MS5(+mBU
zT<srw@ND(kV76&GXaywv$Z$W>bM@LJr)v-#5(pd`cdS=TXhaK`m<~<?${t@=!{wku
zBAc&}yrI@C{Qu_tfkw=vyfAly3q<g_yVhl=i^=zs6|k@)#_b7C+6izFll6@ykC~X0
zdj>P`Ubfp@p~yQ~fa?glpm98&hr~E>sT?2b9w|P-FD0w(1esOzVkh)0iA9~d*%bIc
zt(C0q4qCp)%Y?W`&wy7`zbw|nVlTD*D9}}`w@>NY#z90XwSPHk(TcH9?b#~Pja$Ng
z;=y6KyD9b~&(G54{zDM3rWC`1g*!bg4j2#KAgs^LX`A)F>TFZhm-~`1XbK$O{pX*5
zK8$xj?$IBYRs)}s+_(!8j)jH;jK@T~<)Bv*+3Tk@2qV6CHXgi8IEYeFqn420RE;A3
ziRDKvL^nx#3$U<*R0Sh;q5-1CQ5?S2ZgYvffG_jazEj8FH<}|M?>Q5HLW{r{ySQ4l
z`q{9n?{Dw@jA4P?oGBgt10cY|sNrPemhMN~kY)X+*-Kw4nkj^Hf@uCx|4x9|AOm=K
zI;1BANt1+?a+^J~GrOrV|9~%EsJoLtU)u>FMDW?OXAc!$jP+!HF@*gr<tu(fTM(qd
z6Lb)LJ)Z4FyYHbpikK$z#rSsk&O-Jde%)3;V~n?QRG6LW6m`>zu?Z3@jM;xIA$uLr
z#QI&hN=wX<=Mr`UW_2m^q`}-=F=0JvI(NZOtz>r%-jjOsEOlg{kZ;qi*xgn&0ngdp
zuVD2Ib6FiM+7t}SPUUOg{|0P&EIfsI(ec5F`C#~Zy=w?$$E<`U1#uUw6*Nw*5DLgU
zIIcVV6cflj!VaA7tx=&|V=}Nj?=&jd>(L&5(}?=rU|;_wr^M}WF-Li^Zx;R<eT|S}
z+ZRUQCd|#5DuI-*l-$ipAoUvS8>beE8)_}o-k<SkSuNgz=;sW`C2K<uqRt#D{|&ho
z0(hjO`G^tC<)w^4y>I-`Aw^?YpPN_8Sx0RMV|=AhW4604wu43YRF>X$QtT|!;;HJ`
zDaO0LsT!Ndy45KDqE^rbwvcoc=4994^j*_P*$2p@_nTxdT6!GVvv35a9O-hPtDhk(
zKnK^-C%rr-x1;Un1|=9=pU2Q$0$RIqOZ=Z<AeVAi)@0-jG<)+0rRRHrP_Op|d?yU-
zhuxM(ki%u>33D*kw9e;~;7!Ct{zrB8mWL2>$;`CSRDmKdvN>KLKwHaZpVYkc;zW-&
zqurYKqp?I4)u7mX3t}_rYMXb~OVr|^rv{cSTuFA#&!5(T@juHERsRp{E!$hr%26=0
zrH<U5{*D^11V#Q%{0FW}oY9tf`jp0z99h%k9jzd8CXV$CUKqAS4TYscpC;eW$7L~N
zTuyz%8Y1T}KMGE}_w_I=bTnF@5}yH&gLNSwqV*}ygtI}~$Ycs$CU4jivmR)zx61ss
z({S-A-8aAg{`;ODGUKhM2@N2U%S#1ethcJcOUS%@2kFj`{7P=fPoz7ucc~)ZARTP~
znH_{5-Xv4cHC8Eo0=oQ^Yi_syg%HK@;q9QS!X*LFQ;)lV_jVkz;<2u4gs{Aa+_5G7
zfk1RVx(_6tn8htUHi25dq6;XwD;@0d5J_VoAN8we?T}w+oN5}(nxDIX_yre*)`kdq
zrNsLO1_suP+7`lfl#L1+lB)ACNnr;0zT0~Xg2-5*iVY$dS8!_3)(^(*x(E5z>7WcQ
zzW9)hlmW(@DG8J=YxJR0lrfpQ22hJmP2p=zr10V$pa!qk+OQM+wR!Qw+K>h$G8Wgc
zGp*G9;kV|bFaw)^h6h40FU+VtqcxLHovNQ=sR%H0scZJ2s8;PRc8`UhPYXC^uyJ`a
zcwvfyw?L}h(F`vJXyn=`jS|)v*f)f!!5svA?T{q#`FRj{EXr)p%kgb$4AUSYFQ8FM
zBK@ZmlPbcOD$&nb@eXKBKQ2{UMK!qt+t8Jj*%nOs7#M<p!rw@858ISfk5U3EO5AbD
zR;TPWD+2a%Xm}Siuzokk6ZAq1%C+eA@-csdUYJB!lfPo*cd}Z#Dv`@c5TJ|F(FXED
zUX>0y4JYRhl&psrCCrnP+NXbEs=k9^Ud)?4aS~nuB^u7sv-69XgW0=MHFmkGZ%6v`
z5qE62v*HkAoa|awN#8J{sSuuN+`iLt1jLXJo82VkLU;qiLe72MzlXhL=XY?O7beQ!
zmOFAmQ0_5IX$F7SEr%*daS?FTFrr`db9Z~|9WlF}&heQ5mtfeDC?VM;_K%$=Sr|}b
zPsw%3KcI0{wTL_NFZCjjdEQiy7ISY%gh20;V`rM)LU_2;(aZd0N8p>K>z)UGRFYJR
zn3zXQy8@$;Ozk$TfP)-O<A2mt+pqk-zHw~%$O{~Om|&E5%)ugbfjP;9ow-B<4z9@=
zUh+43*QvlE-iMXPLE2Dpz2iiUoJYAC@(vkXhT2+6*idfHoTZB|k#7Sk2-km}$oBLJ
zS4pS_Kri5Y*xYPK82yz)Fu+%1>jQ_}S{l@X?%hRr7GvoCZBb~0a3%U?Er{Mg?L=KC
z#rVN*Fq)B?Cm_3mG|H=et2Q*8{|*+$<bTvqT*MeTrF()Z`WKX)tnnyLQsTLJ><R~S
ziYjoh4sHJt)<j|Ny3|-T_zd1hyC%oaI{T@X^I;FBJto>fa5+7W1O)R(KHvkdE$3lV
z-JSVat}bqWgk>`6-;;JhDaw)Ji#6HTl=@FnTy8Hs2abgVYpSHCcoV8?!2ae}ikyv0
zZ1aa)RX)&OjVq*~HEv0WHeFo`p=>rq9iuHHQ@`E~`~)KB3H%qUHGv#N$Ks3h+^xDw
zpBF>MoI|5Jc_GM-ljz;nhWPf}a=<Z4JGIm7U1k|~APgsXd6=SFM~;TVF@XS=_Al$n
z!rkIJy>3(+7GweeN`2M?ZlISXhyfml&;J#qy`*ZpEsm?d6BDt%PgOF+p9fHc-2|&6
zjmWtriP!ndvVA+B2f{8zpwz#Vcywzgmuj}p%YH3sF{G9dRtM{|K0LZ=2aB)R9%THy
z^6n)Kv?XLl9x7Gs(KdgveGV@cx!QWPts7KY@;4zc-556{DG5pp423Pl18DHK>`}e7
z_X*|pMHh5tr%ECM0s;<O?--0P{gyQw({U{iWmga#`(aVFWL}fhzVLkOzF|>(UF&w+
z>bLB~EQM<(q9G0;z{L-aca9Hq>jnELmh-v>UcmU<3bUP-<>Hr&PBKF(QSGH+O4#(&
zWT;`v#FV73tZZ@&oCzllQ_xxrQ&5KoL&e*gCJI#$03a7HfYcwY=K98hgV$mA^v){~
zx3N{SMdk~!i1^9*+|Heoa47@&8Z=;#ngo)H(qNbdtGIM>Y8zJ_cz5RM`WaORITYzm
zT;F?D%^JoVWt#qG{i2273JU#?*eNrMTOmg?GzJtHy)uofcOMlJI#@48K%NT&rGi)u
zbgT-Z7m7<Grl+38JFJ7pTcDWR11YIF@x#5Tpmj$dCA<tco`sTi0fRxX0=@unhRqb6
zyaZxB-T}31q~fU-6U}~HpxM73u<x{$^7zQ|P3uF~lE(W8{j~d5&munm)+-??BB}hy
zgkzM(Jg|KG9Tz8ae}uX#nM{eg;i`6^dOVMJN@$PTI|XSZpIE9YjNN43T@^xy;tL$}
zezh`|-0wk6gcw0`CLu8!h}1QN0hXPo-N!xP^>VN_M0La|XQG+&(weI#KGh-#X43!_
z1-k&dwlp_415J#2gMDnXaT!=l0XfhygG+tC@1rYrsgB9`gN`=iSI*UisA*mNaCsAu
zT=1zDgTbFU$iz%T#ue7TQO=DsCzI*^jRy3fV;r`LhFWOU`ZM043N<k@Z1GwXW`wLK
zYr`4x&?T$4R&F1HELj1vqKPs5sOD;)7Hwi__j(-!Cb``>e{_bmGAG;BkUn_tJiI0U
zw{cM9Snhv0$bgF@an3?`T&V|f$0}hTw3&Po?{nSXSzBGV34m1m!^n*wQ`yP}`^v4<
zgg{^8C^y0lR@!x&utso;u8PbxUMV@WI+rUt4_^=_$TNL5!WUpSS)By&%%aw<nSxTS
z!gvSdFbz7GN`u&&+ZFH>^oO`4`<k3LXip}9p@U7{<aIdlEsC|mRKGe$HMpvuuMB}@
z!|^ND>VqWBZv~zW<3kq7JI;rdIEolfYKJ}fJ1XB#_VRpRg;vNL`@t#$ED2+eC6Ak{
z!l-Uu?mHo7$c6TLjkkn}{8`Y<jxl9T^EVzZnxk>f&N*u92p1YurJU*WQ%EQ#1P6NP
z+#N(du6<JPiupt1!F4cKVY|GF_(8F>$QBu`wO34wP^GI;O3U7XvKuPpY~_CekVZjY
zpfhc9+>m!95<M1dcJPnKm1$=R+v*#)+bZ6K?DG;3x|BL5eYV=pr*?qNN7E3+x*|8J
zeE3o)Ml~Du{aU-!h(`ToT1$ujK-v}*CL@ol{S{9U8ejSR9$R_c`w^uCvRF(sYg0-n
z<|NjzQ3_U>jFw;iN6k*k^UDoFNFc>H2C8(hR&+gA${D42*SoQkRd53F595GJLU1Mu
zZGV*Mn$N!jI&$<WlmH?31z{b=%`LhHCki+}i=s|i*s-&fiFNYJ+hH?5dC*hm<5~&?
zIp!`O#JTpEsdACM>pkpJKj1jj=ZINXUDEB0qL<$dZUaq1QTKd*n}-;=4F^ppbfa!Y
zlF4o_k>%;oT%&lIbJbYJlAY*%xtrW7#-&!jr>uzqycizF%Efaoy#OD^Zez_n*p;v&
zA5W^;`%<)DtHW(<=2Tp^^)E&W_e$Uyc%e)A5WCBox0F^&ro23Bp)QP-+V6g-0vY$i
z`BoL#3b+F$h)nuV%dtLu(v@ObaRCY_0irC&(O-}08YQggBr<2cwqa}A^?+rTN<{!9
zQ5-J56<ZPZ37vD0382^QQpnV~85TL3an-ok-y9S_hLxl;1H*7ITf0sU)9$MyW4}XB
zQ*fdI)h?>LTrh-nn%;pTu|Xv8eeQu?g5~d34Av@v!#L+gmDt1KQ8oj?0&bi=@)Vm&
zGnQ1&`{~C-*zouDRXac6v%$>~ksQUq@`P(wk<7B#)--o*$fgK8qNt7dG{__wwQI#a
zedfO~S^xCh!uk#%fBQLoT7X+&GLvP?2w<Lt-k_z>Gb9K1d*$cm=!PKk9sx=$?3$eC
zU?nx>><Ov{>XK<rMOw%!KhjA%-pg|~2#8ag?t*~W4C?Z8k1?<sg`MfPn#ZXh`Ulvo
zl$Vv0nKLwb&cmW`2(XDB`Mxe_KuamwtdaABkPJPV(&oaU^lANsZuRURrc+Jtsls~!
zKH7NYKS(<ZKRxGc99j8>iS1yFbnLAEJmwAR2vj4E92D&jJ6E@x>G~m0Eg5L8h#Z=f
zTTi!zbur|Y^=JuBjPQB?>AP9_#Oy6zr%G^@1M&)G{KJV{y&7>f8m1FURxj_iW-R4p
z0iVsowS@}f-rtF_i0o1`5{5`IS1#^~rNjp<I|sUgn2W??N=rFeTU!Z(lemZrZ*JH{
z#9v)WG7A)>9lK&Iy?)S%qsgWiU~}}v{S>p<rftBbxex2<2jLC8nKWVexAa-l1RO85
zaZsHqaszwXX#b2FwS&`Q<?}g<HQ@w@0G%8(>>*?$<z$$G9|Rd2Mm0{b-Ycl$9PsJ8
zcMzSj90sjL6_3I2q+)pqd{^};jMaEu<#&qLk&-N24k=-^zYuy??JwK{4H=Fjk0dde
zt(?Uznlrn0d@>UNwP8v;1TFw&lQbx15vZxutvXxOF`aha1%=-*;#PTf;#=45rfQsl
zA6@H&Liz^?gdA7J!kqZOLIkw{C_Qd32U-mrLYR%{%x2rck%C7CEB%L5mOul8A7}==
z!V2O!+_jGEkMqDdG_sm)X<=dE6DekpmaoY65&b2MF<s+pLZl%u^2U7`Hd^wxl{vGO
zIe-Tap?9r_+{jL<@wX7mp!}9|`lD2w7-6LT&FtQjfY&hDb!}*LcywHxwOwId*~tXe
zywxb|DY1U^1~M8rlvk(mBT7Mud5D6A$5wfoDBPh4ym@PQp(w30NW(2oLdNZq1&qT|
zA5Q!~P@;RnBR>Dce;gPBIf)_EtgD565;mtC`hx2r^6(q7)n=e>gcT4hp+Zlxhw(vC
zfUH;^Y-P6EQEYtm$Wt8unYf)rP@-6carkqiIY9z)5Iu~OatzGVopl2~9eu>t{)37H
zgrnk3*CncZ_=Pf+3w}*DdtZ<VYmU$!ttGyv$QB6|aHSxX14=n{gU*~wDd$8bEsX`>
z{TVHBO49wK1Wa?xGg*#-=ixwl3;ZCvOzQIpp=jU3p#vH8GHMEB*rMk9UVwmBjt1(c
zHz$=FSdrQH4H+9Z5^sYnietsMvQRq+hsMHEn5=VoDw}{v(M?cQWRoFbq80wl%3r}I
zAc};@k-hJXu=jM;zP_4NxJB*4ejzX^jmTzOv;v|9N15=uoNI5#!|{t`fU^?5ha?@O
zEpZo<ACBc1Q*uFno^CGxJvyi(WOZBq`$ZTrp05;)j^;aY8w|dm9g_vtp=d?qCYKto
zdQi%@9jYKF8aQ5fr)ZO{k53p_!C&1nB=P?{efR{^#dkRO?PZ&jCoy3yS+g14p@Ym)
zl3Q-x)O#jrJL52J>gJ8)lW~X^p97K$kR0(Mr^kX;stiC*>aR!RiDbQv(Z4}5{VTVX
z0aBV#23if=uVK5jwGpo)fz)-0(0k6L;=?1YZd8JYpdrJ_qvcm7iQ6qG#6dg=2XIL^
z6<^G%pmT6>#Km;11pFq~#Uj8A=;@K$F7!$0{@T69sKFrYDya-Lrvjy9j1v^7&f0Vt
zP?AM1><?E*%g!F%6zE`$FCV?ed3bq87+6sDuFfBSfN^;OIj{|%Fm~ngFH>bc|F1TR
z!DJ1+SEKuD;nchRWVS01V>(5Nv`}h=9)sr{r_(qmseNi;?KX;;g>*N7hFI~vuWzD3
zv*dg(hya9$0N#N<7xqvEUOJkX4tf`pwbT-}w*M@O%7Ig6KkGMX#E+%!@0Kw%l#tnu
zGV{TB5neCa<jXCC!N+R`AEBi@`JbiO72qQ9#Xiq7V5{4dTjAc6TQ&P@ZNcjg#y4yZ
z9#E(nH{m^jggs-0#AK%UM&MG`dM%pUebkk)HlkrYIu@Fd9Qd-`ZRG^(3YVTKyn&{I
z>7dewdN>!Y?3HstB(_4*2Fs1?R--1~PICK8RLMs!VQ4H&y2uYJ$?B${n{SdXNd=`d
zE+RpHurkQ5r@_<LA7Ex@i`*FGQgkmNSL&X{?PZMjAvecv35+(nMNY}UC;e(t#~rBP
z8&U;*whrpuZq*&v1H(Y)u%aQf5-18X(cQRcwu+w{1V{-6`Qfdv15sO3FqrxmQ(T?#
z2nG^HA~qE+ZHLnvl+&5;3C{pT0+I<B%cd^-9PSYtn6DTFsGKSd3&+Yx$0jFf5YUf(
z@hvM{I^{w4G4>)Huax{%k3jrRZXpw>n?8;_VdH~9-MY-EcToP@XS8G*n=7p3C|4`w
z{)syT9Hhei5^g6Vr7eB_XX-BKCL<&wL{b+-|9pgktPPQ$31?%n6iiA`o(z017rqYu
zicTq*c0)JNr<@TWXA1tL$(hRNqXY$54etr^Jb{xI->1&aT6cB=c*e&+Y$H5X{yRnR
zxkQOY)w(kw+n$OdTMkt^CXDevplCG8^tvi0e2H7b8g2j-7l2nxD$Bqp-~tqmY-g~!
zPat(}IB8Y}S#VxBHpNG+MlF&4dKmLQ*(h1Uf2X<-eIXf6xi$2K(3`uu{h1W#!Yr=9
zufWm~VwrL249qX&xU?x0Rh7zQ7%S`VTMVP_L#-+i3`3sNlxLV3*Z}(!o5+htR#ybz
z@LaKWAMUU=M8+Uc$9`H4!+}MWi9YE(U&<w0pwFsFio3yc{mBLx8>o#H3(@myQc~lY
zRa~f+q5%Lhl~P9tBcD(tn<xD*a+3}YgYj^Y14JC}Q(G+!5}B{G^q-+HBD#!~17>Mv
zsA2*i_s8{c#Jl-{In_$`u7dIp<qguIcn|*KdD++uyW0IrF=2pK=1r{A2{pPjOqs#5
zf2B^yc!XzGSyYYuI6dS!%f&m8kA6fpjNAXr+1!$}421bAVh9cc2Nol1?+WrxjW2I+
zDhX<VCEZ$4tHHWO5acg&F9S0G_}tqC4@twQ=l&G}fhz3lOus4Fi$IwC<fQG7d{>C8
zg{P5o01`OnqrI@+p%4JFWu;V9(~pi2x-MplrGUE>HMj(0D<PlLZLy&m*scf{SU4!)
zo%r8<>9TJB(@iQ&A>>)J>hgnn(nVA6`PoqK`-<$+chw2{C(>_9>;aU36+wPRS0*yA
zzEF+=Et9$l02OWyV%2g*B#e1p?C0P$!Yl+;JN@vTfeC-hR_&G=%U9>Yn%GWJE9TW2
zf~a=&Kc0jPSZ>(wxPwfAxSrx_w}1t+pS&FibOj702L@>?q|yO?A{5sZ6fi^3)lRJW
zi}Q1rWutU*G$}?b+{t4sPE@?9P-G(B&TdVv3jCF7`3orm<u~_ew`!pmoZG|qLN6Hp
zA0-C*c+Io=@~_~2vgX-lC}?;!r$>S%YU1hBYy`7z%1PRcbJZB)DsU3gmKEhqyZ>eE
z!l+;6b_G7b-bX`C28!#-JOIY1zZ7m9m(_7ud2YY9s?}Zu2D%_=ya(7~ZAv`u*ym!C
zeD0(w@=Vr7y!b#niVIX>iQIV#HMRD`sqkV`4qXr+BRUtdijZsfzev;CaE6=SGxaJs
z%G_rLgHcW*RZL_g-ofs%c^#O;ltht2i-N*8;?%Pm6~|ztnZA8!4RDmIGjm|9`%DG`
z+E}Ipt%wwMuu-Q!2h(%P7|ZoPq5zZF%oX85qR!3aVvAY1kErcu(OrZL4hT$XkQ%!r
zm&}IVV$Pz*{2Z|dpd_Gz>p>_0Pz)HtePG6m`VT2A(iA6RsCwtA186gL#pLtMK1|12
z0KUf^79ebr1y!SNcqWQGmoesjVQqz_H5Ks-Mx6=Z-=p7TBs!tXwGE?y=(;WvEc0+(
z2Fw%IBJ)R(i|uBeb9Pp0=z&NKV^BJ2qP*re_|wt{ez=vpfOh-hk8IPeO5g+}Y(r54
zWH7f^N{v-edBzfFT*8{@#_>9&COdd-T?Ae|=1#_fgDi<wLNjhd3d3^qT3f58U|dW{
zmqSFO(R^(!Ks`FH&Px`SorbJF<Uyw$Kw#H8H0Ibz<iu!Dfv^<Mkek<P$kSjYOSP87
zvu9lGoB$9BpYwrzHNQt!6_)g-u}FgZ<A9d1QEhoBV3~m`DzQtq^oluK)=e?_+V>Q&
zZe#KL|1x=yGN*-;)fY)opb~cJc3`5@u<^&z<-F7SjAvBQymnXz(T$z?N-Bi^pEtov
z`jv26y-0r?*?<Rll%b^qZc#eO+);5~epCeV6n-a#2)c1YwwAuw<Smf#M_($KXJoX~
z5Sa<Xqz-d)DZ#+dSXhj__Wi;@q_6~!9C)oKx-e#8G+(X9Xe6X9klOAlL!JhY;;Du}
z{tks|D{5q7VUl9lk?JLsrikl9Lrzh3h$1|8A*|CVmp6>w*t(xaF;-Zy9IpN$I-}j1
ziSX4RQz1VK_G~sgW}9+>Lc*nlzInoiz^DrhdLZ0pNLJ^d09F<=@Vs~<j8uyX&HnQk
z)Kw`pM&-e~1DLE1&enC|cG{Ix0Nx43=#~JmlVnmAOe=J&6~K}gQ}Y>vb=LNpJejA?
z;?EM;r&H5s1>B$rq@YJ<8fQx1V^D|Z<M0Zt?J1C71H|g24K2NOY@+JP8b?vlHok!U
z?5mqq!BhAhoPy{V=3ptE#oLE7RLd^EQ;__Ify31$Rt>o&u)n!DCoHvVfr=A;Cg`W0
zJxet;g|hEQkc<C>^1>P2>VzLbEP)%iCEG_J^L7h&6=u!qH>OZTf1H;{0U?|s1)?7L
z{>lH(GWO-+nTqdmZ7XbkeBvxMrh0S4Jg6z_-?vy74^EQJPeN;-Rw$a`Ip^m2rErm!
zI8Q}P^vRbckVYDjbU^HypG6TgPJ)Ssxl7yLO1BH+#c>y)hJCbkb!o(ypIzU2&(#AA
zitM%??HL^nQkAT(HNzyS+oZ~r_!EaCUOfwMWS$L)fp_`6IJk9{WBO~EkOc3vjD~!F
z@yM8UXuO-hM_U`OmM?@1$TcY~7;2|_K>~sQ<TuvWN#%VZ%EFRd|7VKl?G&Y+0uzfH
zypHZ|FzL%qNyc1sp)1xiUfc*X|LgfmG1fdK@cxO(yJY%JXx54MzSkSFK0gBkFGOOl
z*-ssXFQXeq40}{+k<tOjlL5+5GK@Ix*IPvJ?dF66u-FHg1KuU-wALbR)1p_CW)N+^
z|MXguBQ7>zgIV_hYg9}a*NF49Mv(_dPO}k<oD>&b8)AeiRt}IZXb&vgVL$GyQ8C~?
z0Ri6<`oWdQHK3s_D`JyHG2|d+QC>N(8r1{<P&pnPwY+GV`>)30&mr7!4VZ^cs9X61
zk2V!d^aMPS0%J6h;v%=5B9q8}fbI+>-(QcvnlUUtgPZ`bu6#VE6um_2PP1bODLnsB
zzpwk=Wh}e7zx(oX_Fe{(zQ)<Mc8W=uyn6{mOf{&C_#8@^ug9c$&Ybm-#l(p9G$A$+
z7eYoQ{THT~k7HbIhHu4~u{e*yq$A1E=jB5$Wd<Tc?|*1&KyP^8l5u_awUH;;DhI;e
zg(|#?<DrhHMtiO=1}yUi02(tV@nx>jAkd__fmrDB6421w#w|T3gLk!VhY*;>+eY(i
zcy_81mU?^Wnp44Xg}9;4?u_1hEkiUAB5svM`vl3CdM?mm2E61+r(*bf=3$hIy{$4S
zeOtcTOnUW8qs&`rncTg_mu|ILB=0j%hQ^WNd<%<_8a1#ByiMl%Azu9hC7hJ5w)|G@
zn&&}&A_BBzwK>_nYP445eG1Z$fa7IV|1<IPK6X|#Fm9@#>H@ttB6py99lTo3>=1g&
zBIsZ;N9Xsosap?|#$I+POtjQ&jI3gB34?z13p{qVT9YbdK21>UH&~Ym`*K%{Um<vo
ze7Z%zS%(zjF_EFf*wv-2V#-?KF|9NwV!LE+kuUDzh4<!qjEvQBq^TTe2Xyg^_t3x4
zBis)v&Q|Gr%gUF`ay%<P<%~t2sl#B1675>u+_6VloZGytj<C3pgBe;K21EiR7i;Of
z@fIFLJ`f?hd7oEj_?00$#$fzp0GswqF=v`3j_hkVUHiERSkk^mD&&UaMBtl%tSHH_
zr0v~0#G<l?<)`&i(-4ajF%dic#bRAbJZfXuHw|v^roA3s$M_dOI!wl^utE;LxN9nm
z=$riYv85g7Hkf(+!y@*Ueg*UOqC}Lsm*3|)1bG1tRl^WG;+@USNh&iQ6kc7zUa5rH
zTN@&Ou(cs-h{dIj;eC=F2o8z~KWCPm+V#tP1T(6FEOcq+IeniX%>lVHP9Ra=?})Mz
zQqD!_27TOhR~6~_7CHM#xF3OE8$%C}t=@v~i?z<hmH(`EM1^s3|D&1~0by0@oA8V}
zTEsScm+Vbn_i4}9-H|={zK*tQ5Qe&D9CJL2!{M}+`u&av-TAR_QBMZp(CRxm;;-Km
zh?;EW*j}yO--t4p^jX8e9oQv*X(||QH{FUo53M-Yc}Pl7s&ehFvU)^mGY>1Ps8#Rh
zt*cloEt}PPJyX8G+;0fmO<&tt-A7GP9r9w#16oj=GJm?5?-6#3#Xwnz`DN3+eJy1B
z1qn}e%Gfn8P}YTY!&XpbLeouis9Vk#FiAV18t4&IvSb@PV5SI1gGS1@t1T-PZ+@6K
z72zltS^oYsOhboERrohOq!o$cza{U*oqKj@fe*Ul1_1AEwXoMU3Pe9IBDxYh<;)UI
zgGEF^48f?gjhYfU^=rq7o%=wq=$4p6^5{y^xgFwz@<N$8J^&DSUiwELa=#qqU)mA{
zh&?`NSaTK&?HnPtRE8rHoT{B0-0A~yz!EBM1oav$wMtLuhO9&Ks!Kjnsk-_O<8akT
z52A$lDMGnI|3<HOcUf(CiYQ@nS!1Z>yAZz#)ztjZDJn(#YRUtEG=Zy*KCC}`cjyF$
zS$l~}ue)z!gukMf@CC+Xih-he7M$Q>J3HL}2AmL%g1runE-!0J2LO`c*;L>)m(!1P
zUMun6T`zy>_5K#`wUhP=CGg;(XlaOH6QcvkC<nqS!Ht?pJiW4~%>(>xHXGOdbh093
z-Ajra>H~!E#(M}88Ed$o#8XFfl_Bn=Hmj@XK0JVNY3B=LTQ9i#a}MJ%@C8ECoVHZv
zmvb57_pnPlq%l0m^FgG?3YBxp=-eSG7U^k@8}QylPh&on=l?_z9&qWbU%ws_eVYQu
zin~>g83?&U<+b*2eck=tqW*kaabD`E!FWzltEJAh<P%eB7)hN_qVUy*zcQ!hK_Hu>
zt{_2p%pp%gFIpGb+u#SUhn(Ubzu^g_Q_czU+opdBS`Po5dFXL^S*CSS>wd6$iJX*s
zR{jI7R4-T_Lni19HTfzX*0}wFUL0A8mjnD_RS|;jnJoOS-5<<1oCvK8i;9ZUDX+{o
zwT8hRANYBs-jp14vFq8sYp%lh1HZ^hzN0Ygl*Mo0+X^&n8WM6s@#Knld@H#ykxBmN
z(z}`BQ;7g*Af+f@9)iU1P8F?<F?!p@UjIA*V>kF5ONLt(MawmQsv~ZX`3bq><9$v#
zxrZiv^=TeQ4!kS8Q!qUnrFyAD*L+hS>Zyq*M`4quKnW>d4K|obV_rP3j){8k6-3>@
zo7Sh^pl`;_+gtSaxCgpruxF&~2?fas*+UWsDb0PLtmZ9U#<RUD@0i#NE9J0{lCIGG
zx*iGF%b$m6THF{s5mg8VOGi)52c5^3XWJAz19nmi=3uqA2&a!xD3p4Rhpsf)57j#%
zZ=`S(Urf(r8ppcUb=VNTTH5+DfF3m%H|R`#-UkG-9x_l~!9;9Rm=evnki#Dj!GB2f
z2xoaky`QlrcU5FQav!9`51>7%+L{(Oczw|M_&mbX(Bm2NzuvnzKcq#&$+o?&B|_7?
zYHW+fR@7}j=wR;3wq(-<)-Rkat00Au|Jn7?&{a482}PJq<~EO|>HO^tr8&#uLVgKB
zB$b`jy|QZRk`sT=aHsjQuWrLuKe~Ny>>r-99$Z^#=t5a*F9Tx8NMS*Xs20Bew+aZF
zUewJ`(WV6lu3x-(-RD)?Mh%;Qq(*vMpx4njL5oUe-<H<>*AHqR>!2V1uxNn(ntIJ@
z^h%3fC&8_Adm_oh(*-ihywbQ?$k=fp)GJ`Q>tftp)n`?mZ@K_BIz^9#u`~+TKPyho
z^z$!k3188bl~6|Bk*vR_p$KHYy(@ocvt8vNg>rjxQ^QMeykhVEp99BP(<}AC0*E?U
zdC#r&@gl}uOi}aoGU0W;BNzwFQqx0yR2_7J-^C3gmp1dyoqJvDt{Z7wk*wCZ8W0AX
zr|Aq&O77cPTL;TgyK+R1w4JDjJY*^@31$W!X82Z>zU+>upqG@QGK4iL_{DGeeUVd@
z;Xm=KsfbKc;y2-09iUJl3Dy@8Gw9`6Vhfmg?b5&Lu|T!`-`zQ&n{w!w_F74jfT?%O
z^iBc9%+`4<dAFh<&|)lMAOscaL*jD2Ut1|NXR=j9?4J~|b^OfcEDe4akM$+Itn052
z`FWi<)ZJfuq52ev{8WCF(I&q$`dt8NYtFjC9~DKG!xp&<){eWVt(t#FX_7hULWL{8
zE%Ni`z>a=(qYlP5Z}aj&w;cXGp&txB|2fEjkN44gbn#Vc+@Ov5NDHIGbX_NL;F%rk
zr1^A7@RY@Rv(|xK`-NN29nzL`NbU_TK4;mcbW@tBDD{ql#r6gFv1prHtvm`Z(vy_t
zOs|Qo#6e%)VRTUh)Cu6BDRr7ogNQkJ%vh_#w+wc&*0I507a&S5wc*m_Ciq=@`H=8>
ztz+XGL-foewY*59U*UNq=xZyBfZB()i0R3ZtP*p$v8HUYuXn5Yfkzw7SGSFzEY3#m
z*B`WR^r}!+Z*U>`ZR?&&Y*o)Szr8m{v_hd$#9wnKyduO{&_k&ID56=LB@b`e5RxDH
zYlsSj8hbdw`M+&n2{TDRSrBVIil>*4^i}VPqT@o#gm?Co#eodTN!)rPziAA=yHoX4
z8?v1KAL^yn*^5!)XOIcy2I)_!dvR#*@es^hW5uA+B8jN3OHZTey<Jq$yni>5ADM^y
zj)c7@pS&KxY^RNk{$&1xRoAhy!DZK$Yy3+RZjNfQ*LsAuK=@<l;|HIiLy%7oc^EZ2
zb*S2kLFQU`MM17OW;z)f-&;4d+9?Z(IlAjc#6{tB;;nAiFoe5)zcA~oONL}#PK}-f
zOcktXo!Ay}2ikv;9vlcOsm*&K{9xv(O(%$xGEBcqK!=!TsqRI)uUc8W?2eAt<vVQ#
zF<OJ6cwj0@IXLC{G41Lv0b?a~i;y$i=G!~X&R)d(?JFZj_}ZJ3euf~zI*%s%*0z6{
zH%*o%OA2nadvce5Dcux-u2!WbX?^9Bh1G>B`Y;Ppvd8orh=AY9X-W0(6=P~Ee0(NV
zMhm)aJbfa*A9g++0klFg&Z!2S!LBIT_`C+@%ljl}I*vXLpspD`Ztt%hG{pT?7#dln
zc6q_|VQ0noh_#}B;aDUuIaWnoWEe5h0$sNo13M>}SOcNzH?xWmxc9G@6YrzsGkPTe
z(q?y(luiOhGW5VheA%%9#fW&}a%YKuvBbW*1xLa?%A=U>ZTfL`Z*XO4M`>>vz_jtK
zdXW(u!E)Xwx@6MNdm;E*W}e!6!20wI$_v9){mUcW)$BjUYF)mkas~&=zuX(3|8O!t
zAP)L-*w^llq<2i*8}z-VD{5(<J36QDRmhr{#~)5_nyMrg^hH)8#c6Gb8p@{>*eYTD
z&MvyPB<R<#Bbk(p^x+zg<_nF+0QG~H#al42*~M-CDYbsHLYVOnF1LkQ{>eK=o`JkG
z|0ljRjXN|~AMJ+P_+KG|1JWnr#$1M9b<mB`eOoiGZL}2nh-=S(EeRvP9U(RE>`L6r
z^;)%NZntQlFeR>`aNGE<eC(5HlI=?K$jY#C-~OOdJGv%oCzqhFe&FIrXUWY$d#c7&
zVocb2t$)865i<csVBxF9FBWeLqLdGca>CUyQvSJJ?-xU1ZbS{efKYp?%!+FH;3euf
zmon?X#emw+AsZyKbJjydm=$<elv}i{9dGeYOf1Ww%<MS)O`?uB^0ve%4I0hrhOlFT
zfMEalL{lMQ?#GdX?)w#YgFF3=wmxmPSNEOvH8Hfe#9x_uNu2kSI%b(ymju;}jD@r&
zf8Kt*UGz-l2|%;SM<XJC`czujI<@loi);*=+CSr@W1(wDf7(>=enb~~h@fI~hsMPQ
zsnzdrJ$F-5-*72@Qlt0e^s%Ur8NuRF_TzASUO5UVva)i_p=vbnuG^24N>uEIT@y9+
zMI0`k`=6CCoQ}a++^P$=`fX=LBDPZ@AvLb=VARKP?SDTrZp)G?y0G61m6hv!6#Fdp
zx(}S_*cpzz`5-MqM(A!G6G?+pzS;gD76e}^Mo5f-kVcI<U%L5M<BvArN7?)FXQ^ly
zyIh(H*B$^-mGz8I{@mR59ZfBvICJGH`_+~v^fJxhH5>z)xqtZU(m(jCNNaH~DZRdm
zeucun3-c9@%7O4OdvQ$u3WZC!PMuYfVG^@`uyLgPHhZg9#g}A^Upk%WX`@Jedvv2{
zEMObBQBuy-WKA3^S3-!bYAeyOFJ@u*y2u=_{-;Ya6GrH9ef!F9($%X!##RUFK=H1?
z5nM1(slk4&I+UaU87keUD~o)cxkY|ZF6JNJO82m6(7mbWz$(irMa8WY>rl!3oaB!<
zx|V=Hm@!2i8f$tjK&0MUih%JPkvuTYUGY4aZQb5LV1zJ=mkb2|a59(h?XaWb0avpm
z2S4qtx%~67gV$B!knIk6ztJ^Ak|N~8nsor2RFS_t=pO+kRz^=&c~P#eds7_D5s&G-
zB^qR@X;+JyyoI!=hps*cUcF2Z6W6@VLab1VdIEVv6@?QY!0KqBQ}2!rb0v~nXDPWi
z9{??dS!c_mJF6YI2zb>~ESc}$nllX!NaDhV^>3>tuT+$5t~rS$)+IUrxNzZsFd%A?
z>le+b=N6-?Fx7FK7!Q`L*HF-6HIJyo;<b0@@*hw>zwKu%NcrNcc(6u)bhY4<PT7Q`
zw%_X7Z3?Bh*!enr;qCmw>eO#*fuuOoV>|RO7mEFt+R;Ltv&3Nsrou(r=hS&F#TZuJ
z9>*5~N5jR{+X|AQ82Mv;VNDPiFfLWY;Z5-Vcbs4MH1xSgC?zB(SV$8!lH(>Vq=q{z
zB5iri!Q9m~`Yw<o3*~12>z~QcA~jgk-usnJ4!c4jN&fV3+mx$ekFsh1zOmNM(D&QC
ze!9-+BL_t*BG^5ATItfQi&ya+rCV0k(XR=6l8a5}*uD@>v<wC`-nh8Ki+5ANQTe7L
z^fpFr%wEE1@oxW6Gw=+;B6mvZl3{j`hHe}dOZXOUzy76pUj@C7^H%9w&x5@gJ*{Q)
zANwC|=({5c9=7;x%+qIQv9Tya%9^>W1aI6Zb^%LUgWYVV))h-ccyf?2U{m7aGB-HX
zXY{sjpmY*As9%nQCca-*>znfYgWJ@=?J_8SkUFCHoPWOXdbYedYZcMVvI^9uWn@CC
z&S|d=P2McuYi)$p`TmLKxu(7cNdAhXPx)8Xq3v@?d@p@kqW!LQ+v1T^I`7C<=~<|O
zE6?y$$um52cI<AZCLI;zpv1+|lB+v5&*>>5RH^-&P-G;k#{p7k@~!r+SK$lXBO1!C
z)7{yXV@B;w0UCe)(Y8@?CupSo<K$_j>#;uD_M1frO6+Ll<B*8tb{}E(yqaCPa0kXm
zM40{ZJw;j}k+C6dKIxTLM`#G{K8jH6tvk9lru|yWsDt~aGqJ_Twvnm(Tdn(RA9POK
zac?Z~Zw*=!7~EN~A0kApEjP*!d+iSNMR#%0tl3)_^N*=wwRwbi#4PpDTA|xdp=gEV
zqJP_#EVC^CW9am+Ols1u<-~4o7dVZ)7^WY)FFrB6va_gAa=|y}T9Sb9WN0W9W}5t_
zk^VDfao=eRB$xX`<oqA4cur;2Rt86`^>|dOdq0fUd9Ho_fW7r<6I11m*X0mT@Hx@h
zgoL~W#B*!B((aVK{Z;9sm#a4OBk{@n#|eF(@It7leDBIpgTecz7c921kBI>O-+uiv
zXIVoaU2_z1Hb4gnWYxtIho>+GMa{cGV}D*mS1t<C#rJIgNm2oRRbMjI?UnP8$Ikn7
z{jnY0r60#D!*i;WNFEXplEl;OU8%*!gQc&P-~<?xa*_D=aVqx6j|jkUpSF?y+^d7N
z`B*1mEyCNkb9~}M*5DSd=lKh3$3p2}`)*ylpe+uxan3SeD{IgluRWSrT1WcUe5J-k
zt)1bYzkFX>Q&_4#Wgc@i7}R60wON~(jRYICt%ID6)OXexr8fd<6?WBgl)DxxRJOzi
zUMeZ_n~fs1h)z*vFmL4QNmsLEk23q=Zwt+noQCc6?Hh*XLj3K<P!aNTK_bK6@*ZTW
z`?Y~+z;xw=o%_BFjlyMc^NdZy)+2wGd^NE*ZEFw9D*KH4fP7wT3ty+9Qzu-X6{o^4
zo^E~c_**yap`NW@t1bF2JnG|j5)TMZvCEf@En78O()T=BTRGr>S<<HJT|?*dGoE+T
zZ<bL%ZV_3#12CoB2#n#V+Us1PEPzW%@%1*SJP2$-Qvdd!rHiu5nx~uLw2~h$@jqU(
z-D-G=g2-6(>8i8q%|Wu_P0eS11GPDBTLr&6eN+{#XiM#;{_2L7Svx*RAFhXm@kH`x
zbrzSNdB6Tc09B@zkYWSRJI+7H>xg<W1gIJfNcKr+@SZ#Vq+=qkT*IqFK!)xUl1|6$
zffJpfaqVB;#f*)I>rlTta8PYU&Do%Il_ze`N(cAd%|A5}+#BPI+%vU%*qMH(l5T|n
zzVfLm*xX*rdCF0rsPZM{U=_YL{g1HydO68`xL+@5_;lar&jnkO{f0F~<L^I<!Z`lV
z(0v*XMI+EConcims4hbmS9VSV&IyJ!+20fbJ<L3eGOsPPtr{wC*Qi;0Wqmsks`Y*x
zZk;Qsnda2Dj~xwCx)B_GL*l#bs9Bq4jKWC3XRLGLS>1-Gegh%DoUp!jIPKGs_Df-H
z0d&C{NkHA`FpkN5Rz3-+uUGjUUS_>R%)OiS4BscneXXJPn7j76iAopKpgkCMo1R(L
zxx>R2r>6hJg8}y-R5Vp`)#ho<gDnl0uG7`H9fPj}H;Jo7rEZV2{CE8^G<0lOq6oF7
z10`-recmym<**R<2szO(_3`uP&*@N8zF(h560WeQyRb5H)N}4}ss<}oU*m0^JDDMj
zO^Fot8e0UXW1UhI?sI1>iDO##Q?i+(ULPLU^69U9nX6S0bf$2sLvpeN(uHnw^L(Ey
zggjY`Fv)O_D>RJwf&@>>tVN7~>ywKM*WVps_cZNrHnS6j`yE#Z;+95mgXTI#-FRxZ
zD*0r1bx6ar-h10XdF5m(eJp(7Br;371k-24%H^;oJTtvv9nEoCHFrpA6`XU{R$X=d
zQ9V{?%CfN^4J|vZSnKL@r{6`*+V;giOHNyiwRRL`H>KUS*zUye&k%IF{jl;ojcBhQ
zs(&_H7CV1KvdUAlEYQe1R;?+lzxdmw8{&f;i&BWd<mzd-Jq>RNyzmA*7*>j^<ifTw
z<VUkM#G{rAgw#!%;p_G8Zta{%nAB8U5;^kKV_NWJ<Y~wF-5NXlO%^x$PJXl-{sX)V
zIo+|y`wW6%&oDPMNfVZZl2m7?3}j%an$i9jPn=e5FjJ`N-|R)YS6;!s&=;EQ-i6n0
zv>UG4zdfeL)QQLm%c{HFeysGgZeSYLS*pUK_S+_R?7XA02WEb>dgX*8XPeim=Ektj
zS}@Aa+qXMIlaQ}HP5FM1|2Ip6qJWJ*;PMnk)LFX<lev;cYxf-%w+H=)cXfm><%RlB
z)r40&jYR0{g+mnV+Sv3>y58a}gEfnX&!+bQB+#*POIle~)G#<W7+{zj!2xWCjd#nV
z(swsh4zrc7cUnUKx$c$E{n^{f8+~Tq-diKmI(1*=H15Kjwa$#$qS6v|a;nQ-WBpq!
zr+%+NF=6%FeT(;XNa$;dW8T#`R!1yu>nLf&S90y(O2bHL)`RZb#>oRP{ZRjDBS|1G
z)eJlFveur-%;vs@cbfA?hF?Ns+;Dxt&&LeNEz?YT!^_6g_a}965P2s0IbrqS{D#x-
z#=odSF9B;FE$KGZ@nfiq0SdNEb^ninbJ*s^6Mj@h-Svzi)!&{N$Jn^~5EFRirq{(8
zrps_|$R2xl=T4$^^v_PQ8CDG%1rtou&#wXthsx=N=Kn|6x4=W4w*3!{L`f?hNT`$~
zkyMlhwMA5{Ldc;-PB|xpacE1=CWjCTqm61E%Q1(>s9oob4#;t|gHaQM!5Cv^{?|SB
zdEWIt@BjDN&+hZ-vG)Gmzx%!p-|KsQuS)p~k_w3pg)!Cgj6;9nHn}Sj$UmK6*R`;=
zQM0ugpds}^dC9&<A=aJ-;MNx}3ZJvBl1wF47f9AA*Tv5|&DY8qrav3SbxwpXs4#f|
zXzz0g<%vLh=+QD+Oh)*_9>(?OQ!YiNJyvPtXDNO<UM;+L-%BePo8dQ?Ez~lK`YA64
zQK`muIXF1<=|gZ{r|9#K&({%un;jY_itc*0|G4ZFln%XW3<x1-Py88Bx&^bK>b|Du
z#)+m&1d&|<=Kzs=@~-$$+}Ra#MWT1MqV@p54h6WrEZLA~Rh1=P3wQS8(bNzbP<U_R
zwpj|;0d<4jNt{>`$Dgml8Sz)pV?C29z@vbwW$u+fC66%JN7+sinJ>8HU?XuR4r{a3
z1MrobN|X8O9ID%i20uXxbCDs6$yN*5_c#!A;3cu}-%<M~6^7K)bhE+TsPiHAPNwHN
z<(L#jdxQRg`zs}5y#Hu)&wYP2x|n>>=&;28h`<4_*M3)cgW>KS!i2=BAB!9uQuKcz
zo_6}6_1mVgNkc&x##BM~lk^xAjpW2O2FGJGKFf<%sS_~6*zGjpg}7(^giWzw^8N~k
zIaX51{oewG#^Mk~D0hJSIWb(k$CP7TCBNGEsmV0gB+{SGfikYPq;;hSH>5GPF(Obv
zP-$=euX0S|gOzrLE0c4hzF2L!R&N`!PEVpw%Oih>P*maJ>qE2;!P@uK+Y5h~nqcYP
zo44`8-gji800}_iw_7PXL=<<;F=jRtl263#q`kv>dqN8VNykXvpnJujfIa2q3n(l%
zL9w%IPQ-a@2xN4FJY6ciK;2btb$9|*i`n>Hy~(qimXcvPQ8S{O&b&-#Sw~G&BiI<;
z%|}LWml@tcEUulnx!c9&=Q@Zq)W-UlaSdN!yUbdacZzFQXg&8h_<XNKH4IK%<DT)g
zo^wPCr>(9N4*{HBsz3f-P0)%b>kE4G<mw<CCHvcPzJ(NZGJhOZ5WDIYDDS_wO8%+P
zOBAD_WA#}=-uq8<=B8<mzZ4vH-AbGt!nT_TZ)9@cac8hmN@I)2B8=B5rzM(KL^V|@
za0IyIb%mSIaDGv9pXWgh9ZoF@zv`wrv>W5qXYOiFTn%&pP5^Bivy!MAl)`q-LY&9Q
zfSLbY_aVJ38yqss-ux{2wzvErdU0CbO@3<YGNBr*wEjg8>-M0|S{v@MkHId$yM*Qd
z=(Y2r+<xf}C=>;h3}%I8@q3x+T8bNWCE<%04|NKdcO$L^5#V6XcYQ7OraZ*zxn_p2
zSUp@ognL|*y2JE>A_hG=lmHa0;E4O~DU9FYMB#pBzqXZ(<rO%DpX=!^t>N$BxD58i
zfaI(#g2Mle4+0*>K(aXda=9Vaqt@vJ_ydy8GA8YeCH-Et?y9pAVw+8ev}3kuq{kb*
zC1J<h)_qpf@rM3?Bme-Z)HuWp-^Mxd#w3Go*~n-0o(*T)`}B7CLh4Z=^<w{04MGN5
z;Gl4OgBxma#V*y)qoJ^-E`DlnOfpALY#5quYJ(!~oM_?whTSxJus#{{>;c_gIL~IQ
z9{yTb8s#v(($1lgoAN!Vb=D<_^37Xg+Hgf~{7!0q7#TmC_4Y*_6dN`=;}Y2J^!`Tp
zK0|WLg3lYOlE-QFV_i8yKaHQdknyZ*%g&&C)O51jP9xeS?YJ-Wr4obwqG4d+elvVL
zb{As{b|A?8`9qQ??JU%Kt3jwkEtgYoWA`WqRQfL>@?p#M%$kEW_4d}4l-I?A!eYyz
z;AG|lu6j*<XpKQr4PRdF{KOLh!@u21pH)VTP!oEA<OuXQtP=Z2Q-0lzLIT1Ct1tXm
zYPc-BX5?T*Wt-$@(^2bxCkXnfpkp+U!uA*qfY<6+)@&R(*nE>5vab4|`^A)%#bzfc
zvSOcp`{6)LLc@s8K9Or#59pe5gG6eqR@hmm_VXrAou3Npzr~M)rIVb|P`)J{*^Zsa
zK;)K%^7k7E^S|gV#_%DC;ZHm8S|iv`BGjq9u>oJ7VHI47<s4!sLx-Ed$*h5SX*&Ad
z5X#K9bvf4^7k+-7Tq(23nPS{M*L;Qlr}sMl!M^yZ{XLW<M!?{fO}u`?6^+2bdb$K?
z(@BLbVF#%`%wzCM%lQi(Tg0?lZ;}fS9WvH%lC?9C(7}J$ozBgvhtUaN)_v&MA)#XX
zy1l$1P-d+<tvI&tKpZ~>t4zp?3XZ<N-jPwuaT12yV~7gmk0N9CyzcV{1)f(j*Tczc
zITG9*{nuQIV{z>jy9g^`lLLbr^r6-|IKFZRtutJ%T8Z{(z$&tc^Bo^o6yg(pfO;EI
z=PDVdY8-q2@Xr!Pht3^~3#(h7UAv{AO`!)ojDD(%9D$Z^L(};YUZh>1)eZkc@MIsJ
z&JM_zi2X4C4PkcI)z-PMW7#`h<~8}XsjM=?-P8IFRF`~%WgMJqr&QhlNN>o9c@XmB
zH}`U($(OkX3_%A&G)xtDRQPue-E`y56b`CCBRFa&JbkRfg;s>Fm>aXne6}8+M&RPC
zY3<=L1Nz6#sYX0w2`AbPh&F^Y;T~fhM#HvGO1X6|d1L`cdgQ?1>W6u=p)TPEi>vIF
zw9_Gd$PUw!dfvaVIA&I&x0~Vs>HPgJ+RK_)R7m*GxOBn$=`*Jt1LUuHrzi?n@4n}T
zDEFS-A6(p>Q{ix#>4*rmG^S8>?!`q>o3(d<j8Tq|nXGWQg)HHUd}711#f$AR?8SrA
zkk!O{dS!N2uYy?XtL+O?-GUYWn?nk~MGL0<uX^Fb6pBR2`kb(fRSHR%;Q>}%wI0^)
z#g>G<^HO7vE~S=A79OnO94?KYfc%&@&IsCDLv60@HSB#F7Euzvo0${qL_SMw{GZDd
zHWTto(E%O=JlV@tl;0Vz_)Yh7zlY=S;}t_gn-@Ol4_(9(&O=AYTjRG>tq<5Z#uJI)
zkMlH0S}1kX{M6e|LjngEtAK<G4aPu?$Va+8WqIX8KfDGstlgxmxl}z{J<2hH&w{_8
z%I?6YM{tWdiwvpS$}M53oI8sa#KTn%ueZcz5LhFeKSB_&94OSRPnmU*OXEHQRz=L9
z<X=bp1~muY;h#8(`Ul;k0S+-%Qefzc?>j>Efe*1@^iLVZjsonOvccdn81R=XeXn$^
zLTdqW?t9Cx{Y%lH`o~BB*Mr-qBr@6!8(Ew<0+Z6Jy^Cgo-z(H92@mjt)Zsf5QrSjb
z)-`*W1#LBud#DRc-66Lkbsjond=N6>(fEr5K2w#SIiNp+RiQv53Q!pxeC&kRGxw}Z
z$4MouowO<}uqC<%MRY&L$H&tB!;Q>J=vW^Y-3*4^==t!}?CG|f@C{5B<`$pO9>I%@
zEz{}J53uq1v_aU){j_Y@5*U`d`h)(gKx2fbrT<g`yF23D5>oN%wyZx7g&(}0Iu?xG
zx!A&qBI+<)27UM4J6*R&0?%aQwbFpEGhbftQOUO;U+P-I5CAXj=T)iQ;oQ=QIWiA=
z))tHTvvs`g{XCAf71^_EAb{d`*v*GUi0Jel@*C#0PJ~kzbr<GXw}Bh9qW0M=H=F;L
z!{XhDpNye07X*L77Z!{PW;n1J<Mzs6|Ah6DrmkiAjM3+NdG3&Bb)<6SWPDr~y&_OG
zUTs<8s<lIM53PR0awKYGDPDUl;+|aD=f>(+(pECq-f=R#$~((g)_#|U^mCd33u3K^
zS4Iug<^V>2A18kd)+&ys>W#i8uZNG8@psM6FiinPU5CtAOXR!81k$I0A;)Yxoubls
zLw!vWxm;dxR+5j*KD+>*hh71WsJ~rP;Lk|>;AQ+H>KF&Niv_=RtVEZU#*r(GZ!Dci
zr^&!k$v*kNk_M}3$LWEA0Xx(kbuU`0_qUVon}doz>^61ORBK|*2)gOj)H*a|s;N(?
z_p~O@o_GT74r;mFhEt7)pu!Go008CEuPdJdz<dTsdarzhF#0NeoGU$)4~NCh@HlmR
z4sdu^Pwr1I9>x7UXMj+o<Y^kmeuBSZ@iXeV{xDcQQsf}Ai&mY;&8j*#+==r(w~bO7
zYqf9)*UVOmx(04^XOB81HGDd~5wX4yYK$2w{V%blU^1#+tBfE1*0Q48Jrb+d;+IC9
zzLaBNFH0K<zWqSS?81{+1p&+Fq{fkJ&0v!)R3f}zu#x5BK7)6O;330Y?s@gJY>5X3
z>hEwDZ^X|20A<4*b<Xuc=F*{qpA>joAPh{yce1J3rES#{pA<RVX8P{D5OEjv(f7Aw
zeZk66Wn#wdfqxKJRgVddfY+oi-M{~spzj#6)4>1%2CcG0VJw~u+wot=>>8-H3D<4+
zjl@;KzU|G8keC@vimYEUUtty$S;p2DU@)4f$j0cWE<TE2`tZ)e-RO(p`0$~QZ<EZl
zmQ8vHzzgZC9q1@Q2e_aaWMqdxNlH3uJLIT8QEtxtr5?d<b%1Wfh($y{HFkHElT<U#
zc8T7{9OT}Jq>lCBzYjz@bGEttM49vJjR2xr>Im`F+!SzQ&W=AMZfAah**=W|N*2j*
zStxhJ6%p=KMi5wbT`gwFz<FVJqib9@(OBO|_J-J+$C4i=sR-edmZ*Dsz_Mej`>F*7
zZ=jBYZfG5!#phAVJw3o|r`Lub!~@L#(d1?)wl&Z$I`7|-R{BA+UV7wEsXu_Ci(Lc6
z3)#4Bji%9inGeISVKPp6(vxd=@AVc8`OlNRCIcGW=L|ZO!^O<nrvwNPVh{(D1!U{+
zth~Sw`<`mRMlfpgh1cQA{O3s=YS6-X*vSUCV}4Y~t#_xmml-OI+Y{QkoJ>c>>?|$b
znJ0s7qS###XWFUmVQAbAk|@Mv*h5=IA_jite7e>kLC_ipVWzE;+a#F%k#U9>cWZ?v
zp^f`Hs^FMcG@hN?xoQIQd`HyVLbfu0n6L;G3F>~T#=>4|ZWn9+MEeuC$ahP!dcZt|
zOe{47s=+cRRsV?$Mt+-i^rpyv+F*2=kZF=|Dmv+L{0FyUbqPIOYoP2oA1K(v=N1q<
z$946oZ4k1B$&x*_0nM5n;WY@785adw$>3@X>GCk|XL`MT-Tk5BC1Pw)^ktFx_pWSv
z8t;5`qJSErGQvXSnQ=~ljyk4nJ4e&T_E~HMpTXzWssE{cmwDf>6X^Y@!(B5TP?CME
zH(crT^qyx+j3?OWNC)=+=0t1jg=id>I3=joosCJjortQt<(^Sa%~PQWoUVD`WaQ{C
z-0?-OAB<Sa2k*CSWMd68SxMoT)Z4Z0O-Zcp$`Vlra6cr&RUSaCgt=wi2^Sau)!}w<
zRu(}F1k?q92Ryp3oe?u@!e?V2js!G?ceRNw@IpE&!U#bQ0sPuo5-u_zHN2;Aocw|>
zTR}cQtlb{=WP^+$@o%IT<P)ikewy2W07x4bJr3@w!*to#@c1V?+JBXprpXgRZ49mk
z$d-HHR!0QLd1702PTzbj*Ijt9p9AG?2sJk;X?*V&?vYfx7qvd^Y8GtLKp=W=z?t`S
zodmb>OV`cUb18ufu}>(g6^+a;Pb3YOGYFCc3G6{{MgPGX3bw;kCc^nr*RPjSr6!#z
zyOK2#Tk<N)p?sInSPu(87rU;7`J>-gKr)%#ieJWQ@WoGK{2YXnP9(_C#$K<49Dwm;
zN}jg6yZ90rgWF@!yad3l_$;z3;#_+pLRnfW5}yO%38Z%>g32~Vc2&;WfplootLaO!
zQ?C64{Wq+M7^Q4QmP4ID$Zouv4~>MWKXVCOXFU_)F@HT3gQPk>X!PY?jc;6JhAD(o
zU{D?BkWG?bV)&yM2N5l!tR{}22-v=xK?B9w(95FXOBNMQfA-z~@&^m*^+JV*MsJwP
z{`B;vN=6d)@sjsgyG_FB<0PiN2YEgfu1{KGL$te+#G;L`kUy8vs-0aifDw_1g8YZ0
zsRtjZ`UZ!%!CDfOCBgc5f1~rV9hw*L(fqj-{hI^bsEwIh;}=$?p4{E`aT>=;?AH?>
zGPKzmF;e{?umv?1^&T|ncDDBo4Gp!0p2v9}0O{s(6o@`A!>0ZF7S7EHogD2$kuM_F
zU?P{PSIF5xnF+sBB5WG-@-{Bx2{vz+V|k!X*1?Cjcv_3|yU7>pIoEIAKR{$~LcV;~
zj6m!-)WoHKbcul)_WK60A!9``g!wi!%N!=$#VRWsvGe`)u0RR3n5{EVwk5b0B3mm?
zJ7)gZ?~ImtXuwr>S;W?=lP^0{EL9<DDUjy;2K#Q}kd@k|tqYW4sG2XEYUQgH4shQy
zc+4v#_I;7d6=E#$$%x0zR@D}FVzR?S8pJEGI~A;Cc~6BBPeIACBMY0Ll#p{U8W9U}
zEmshU-#)Qc2&7*0L{(|L_6fZp3N7IJ&G?8}c}(2zo!mz~L7Cy7NyRRUc!A(-cy|v0
z>J+UorATq4Ynb>4;`H~iV0C6U|5^PR8d{J~%TSEcIwWGuo&yXWHPZ<=R(%bBk_fq9
z>}5SzXJwi0j_-6@{4%qLwdUM&>#uZ=)ZS;Gkj22A9yeZhI{LKD<86B|n1h*~k^U3w
zl{UZdq)xJ;ll5d@WJDBu`tv<myz4mr0OyUE?jV;I!hnI1cAvf-acMYNRQ#e2MNcj@
z<opd)>WO}WvIM<VQ1dL~gx#V&h7bRr@!0SJ;J52@ez+p*SH7S)8Hz~g$>s>{0vN|G
zKwt@K8o7e7`ors$e|GG2f00fPqqsFS>-0z%<0>PgomzvpOWdA@y)9E`KF6*vyMltm
zCD8#y!OAEOkTh#2?ufb!;)*VH*P|cVTn?8a?&A<qLhjLB5|1tGF%-e*-?gErC)2k4
zhrJSceqot=)aStF)~ZUxL>6kzdz}@2e{52+?t%Unxx?W}Q)_14dryYQx@Orb*1ycH
z_mLHMpMj&uP2>o>|I)jMDGeoCMS~D)E4@uMLNOB{0#4}shL@x#O$H)1zHun?_wNo+
z;@jwgdqm5#Nus^Br6Aj--?xmxK)CK5)r^9vmP2kRza&Z<w3r#!)W3CCTp8AnSiB!5
z{P<bKcqeUjnaV2}SymFeB|U+M4;!*1x!7@n(jE_yY<r2lFT`dn&=o_QE=_+7Hm$X4
z<!)E|M9$%=V&KB5${3a%!Z!tz7nJ1Ab#ZMI!#xWJUBN&A0pWZ((i#&;Uy|6%#D>3*
z_X{Ucl0<I>-Arb(NjmQ3kN?)x%|56z$mt@53A_=8P=0jZ+3eKp;kU+${<8AN5W4!a
zE}V|=uGBz2u@Ed?hdNC@JA-pXcs01r(~kb0oC=g*T$kuL+*ACim*>Ulfy>&1Bnv)F
z=pLqvO&I?7m3S+)J9r|hnbEvLpg9XYPtUaWtf)5Z*cN6l8?Izj7T9pGOp)>tL@U%?
zwj^gLstXM@{w<t%-`49Ie1x!mgi(CpE#SNK3U2~2IvTs1j9M;iJes)jQ6i`fZ;P@(
zWOZDm=D0~o@$VauYaF|&YufQG>@fKWI_4_NVn)1*kQeNKkPn3!;Ql>{tX6+L<i%&&
zs@YZIQN<>}T)tG{Eoc|V)M(nDakOek3Q^z9*8V^m6MRR(9ii|*m1V3L8$Xk4;3iux
zjL|pIj60h!U4<lPzqb&dcFi50b#ainX6@P7BM5h3dx?AC<5jCGlOGSWU143?BFjn=
zHBz6hBP(tkDp6UzOHz6@@kzYR>h1CZ$rXKvocD&Z5oLR*S+&O{+;x*~&kK;ct1D{4
z^x~Z={oV^A_3?Xv14NQd6s<>{3R}Y(pi3+A!A{Ix&>E-2_gjrBSD9krUUo&FD&6N!
z*d<31$l$h5J>dFenSf7{^jCP2GBpq4zfd&CxIJ2C;_N~uEu0=wji>}IZFpCKgX<g8
z9h$W9GQ!v;uN-loe}9B3e{LI<gbnyRBLEXU-bV<yq~jwvX67w$Q#wKP8lzoE6xn4M
zJp#^EvLuTSJn0K9Y?0ljpnRgjeN+ilh;EyNpB&~A4NRmi4*m*^b_hZb6yl(keLe)l
zpMYPu%sosNX%*c7s};h@q8jhW#?p40Sx3s-b@i^8E<CvE=0@;Yry+aWvniu(|8Qu^
z?Kv6B$ZN)c`|tuGdl>sK#6lSFg}wXu&C}7p%e4&v*0Cqh{kYd<OLmwyb&%NZ1ZqSs
zdd=XgAkf^Qi4!RMmy@d8$6u8?tc7t3%>m+txZ*TgI(+M)v}M+#F8P$D*hZ5C|9r{m
zid_=q5A!pEIwTStn;F-Od1&M5*m>iH>b0NZ^8XeVdQ~*N0Rd?M#B@V;<xm&?j?85E
zT}m%j(%)7rW(hmcZt8Hep=6(chA>uIz(jSDy6}j73rC8DFua2Dq2Ib-l#!{Xx`#`|
zHBU}Epu9V51}Z=HE`Yzmcr7;`Y6dMeLeA02raKHmk$Rr73L-qffBm;2fdrF|Ui7Ib
zxAOOfZcU9if3AQI#nau#3_nuXc3wbI*>4IjUZLqj_k8hriC&C_6p*yddGQfPh_>-l
zDrTfs7{mS8iysoaBZ!k)j-CvWf5U^#UQi;dm1*vaFVJ(&Ks4{=3NrI0x*d;B<%sPu
z@s)X&c<Q;J#+U35TuAO5P5#noT3B7A(amYgEthwVb%oqh=-KMCb9V^hbWgq3DQ`P0
zwQg>j-%}Hsd#bnm*yl|k-g%-s;m8Twq2uW*l-FtdE7gj+t<O1|aLB;<T<-Kn(#BV5
zO;J1UY<-nB{Yb~g;YAnwU(Z}O<4<*jSH|u5>QT>Zlb9iMKZa)lQf4u}K^7w1wd&MT
zMCr-nzVf<iqxK5iVodUEvu-i?=Mh6B4FDSKn9WJpe~s!qeS8!?Rq=Gc_8NosM$0n)
zOmILfx4_JwEBnS5ywTU8E^C0)ekDC->T4a{CjG~PjfzsT@b4%PBC}pMwQYlYI50b~
z((xi{kYyZN6zeXhJ`gTLzG^l=uVw0eJXQgX<><uMzW2Di4bJAIW>yVk7X`8m-QdHU
zVYffG1{hnLYHt)y7IM_JdDK1AzjBS?Al2;_y8jJhs9!$sHS71+Xha3cb=kQItCs%5
z_=R<}^!d+(;uPWE(V?-hK+-_vk}J=^_@KpqNzWVA$9L^+3X;Pw&QCq@py(ODuv{=i
zROCg@yp4>y^J2$VQ+Vee8>T?~XCW-2Knm-c33uH7!R#+ibUkJLs&djvjAL;*Cpe!R
zQ@`1zKLA<-27_z|fBsI_fw;4>YcNz0(~?||usM3O1OH$h56vq@?u)u6R<Qoe;Wa90
z$wPF6a2&}xUQ&DZa6-<8!Q9B=q&G5RIbJ8D3F|+#Xy??WQ%lN^mo{z}i7x5ngB!BH
zKYI(9gWqe^9JVoM#qh8VgZs#bod^8Cla*b<M-al0mpP<`$^JGxJe<QgRC{t^6Ut!1
zD@Eq>ktkEIp8DZyS@e|{$Jmm!Na>QyirGgON1|WwGhlb7=2MBt=KlMUE0!L+1l@Ug
z;a>`*dCweFHa~iEdg-H!NSJrFujc4SnD5?U@JC@VG+BfZbJe~+8*41v#^qyET*DO6
zNV>6>Fz4v@xSc>0QIbJmE+Z8mD9)>_z_aaT^bnGIj3SD;O-JAfEI9_Vb8}H_giRs~
zFbl^s)3`stR?iMRV1aZ$obSj(%X$@7?hrYRGz2{qGIkk-Ttu>Vd2_ccP?X@IV7AS1
z`ql1&)wP{~!OaZZj$ilU=v9@pG`x}U{g0-n>;B|=Tr{6o=5lBFjq+yj%3+tp{jcB_
zE{b1!8tD-|P$G%`Ncazw*Af`<7FpEh1sROSr9zRIij9ld#+-7$I2QcO=F+ag_%1zB
zIMW9*7P*84Z`3vP_e1iWelre;l-2mn$<0rtFSXnqP)XSJ`SK;H3+c=f!Dmk$>cz^H
z1J{IuMh?ymJaBsO=7m^EyQRT{LpgeBUY@sfAn4+i9pTSEtBz`Jfd0A3kP`27$1kX7
zhuAlhj_zC@z&nKt-m8Ibjtu9)McR`C5p3{ipMq>7DMxj5biUR|_*~AIu9{wSX{ZaC
zDB&!+5E-QznVFdu+(q1J9QjZQZj`$o<G6Pk*V#e2xqm|NO3OC<;ZCEfvtrq`(CigV
z9N$>|SfX%cYH+IX!bjsM^1kkS#fdz-!;0+E)zf~PRT7eT_To0*Xz<Z}Ql=b6CF25h
zU%|1rut52M;Kaq9&8Md`0#U?K9^oyD8KXd`WhoJ1uf=?j-iW;|XamX<v<5IL7SClJ
z?t`Ij#J&&Fi^l@$##U~V<TAKF4MdJ$Rb4Nsy2-#8x2`S)sf>aMIL)MO%a(rX$(Edq
z?AHo<_{FgruiAaLJj~~J+Zu$vh2c8E#nR3u21LsoMTZH~MH!)HYJ#w~%AH5q&bR#F
z(6KxMv!*&Ui-6jjPXtCkrm?6TbCcD=)JE6KqALjFxau4{3*K*Soq3wfbdoQObGk2>
z^nzCVOjj?Rc9ogof@6h^^j*7|Qc9riwatzg7-C(b+24?y%%wXL`GwuhmG+|c)p5aO
ziTUTXhXWw@>jdkvog0A031Ze7{Ijs4VWSe3a9mHtd5=+*nu=E9rh@es)5%}0>_3t3
zatbH+iOl!_BMNRq-0gk!F953h7-Va6U-&|mX-+ml81H5LxHv<yXHO0s;99GO7_sf1
zF8HC+G=Lo<zsQD0`e(iE5p1<Erg0B57lnD@l_P3cWrL%dam}x;j5YAWf(D(w;!M*E
zS(k{GkYiJwqMlFEs>yeQZr)?gT8>nG^-d^>ygqcWN*+3^vfyANL5jP1dB70+R)g@o
z<?6t`T3D5M6LFIq{d@W&_#86_&s!$7CPLg+ZZP1h9Pd`!9MXG4@uhgl{FW-t-Wob-
zLH|H8u789DH4S)+AUv+mNf{ZPUKsr}8tQWC=VPldM0-!}8nVOjqc*FrkeTTr_N5jf
z-kM~KE00KKw+lI0#)GEKxXA#F|6IUps)8oDOTtamZ6y4RAS{-<zbyE46El~8CT(nN
zASSX4XQitCEhq%ee(D=va5W#`Q#rv9%=|b8@jmapiBR)`DK!CqUOaGXpC7n4Is1l0
zvBjn^(#!Q|_RM$-^CVnUa35<`$bCquQy1(Y&S7E0SJyQb(Mlg3Ft^povzdSI+$*T|
z7xwBk9=bL_+fzff(mWS!rnAzzo3V0-E2Wy>c#Q%)1SO$y(v+Kvykd8LEu^$C_Vq7^
z1@FHt5&DO|gSoFG*a<imwNe)ngdNIuirkt0YaZ1JI*CK)?f05q-fj?&;WmX=lt*8#
ziioukw?#s-7M%d(x}<Q%F6R0=h@ZRC*Bb)i^BO~OC0F}Hwzrb_28h60=2O63DBvJb
za*CHwq|?Wi#ozT0yO|EAy8YQv91}v+B>!}skT0DUCTD-cbSL%eV}-^^)sTI<jk5rL
zU*O7p|2oe*Fu)o>kd6nRbA4+R^}Vs)0OvTYefg(|jA&WisN(w&`f8hYUW_~RdgJqq
z==CaE`{x|bL_T&K=k?xj^kw1GqAgWkeR{31XZY~Gg(?&`ASpYDjy^1YbGUc#q69?n
zJr_A_Ys0d0t1#EV7G^>eGvs6wg1hQwkZ8@QMj=XX@lBV0tObYWUuxegk#UfC#mGPN
zUBF)<ggp!^YZ}9lw{L)MG@)0tLjzYR2)RkyGXJ+H&{wS%5wnpr8}>5S_=Y&Sc^-fP
zWib@;rH41$JkCqac|lq_-9o%Qw^D7Y2yp+rI(c{vEtP~%19;R;K=Gn7NARi9+@3v-
zP}`L=Zb(;oezlhEi7KKN@L??lR(1YTNZf8}PS;3C8X^oJljYRi5^x>`tQH`7EgV{i
z{dw+n(S7~lSO&am)}diuIxUWQNjdEHjB3CDdG=j3zR!PaGuOOMt~Z5NB$a@>P}S~?
zLuZKh%RjF~q+W8gl!{Vf;c=bDZS}tlIF3y4E@ozAWPrnEO}@dcu~NOo-Y(LL;I8-x
zey;xT_kc^%pCER@*^btq-;J}yt1(3yAQSZsLxgn?{-POO>f!?!ItD3S$lCM1nq^26
zH2-!Tk=aQOp@p<A;<3L@r?kV9H%IlN|GN0Ei5vLQ+-V)^Fx$_~3TC-%`v`Gw<cVeY
z;^b*_x9vg+(QD$*i}%(_7i1Jy=A}fVk)*d(S?{E!6siRU$~-hi;m9!mut;8YR8w@~
z@4X5^Dpymy=v+AOj;}QBP3wbir9`Mv%iE4#k>3m{q&?^#w+GXpkobj-!%WgNDd<=9
zf;Bh}Yd7(o*3-q|aF*kZo>`5~M*fS+Mm^jIl}&-iZ0YRAcY)YmfhJp|)AF%lca<hJ
zM1HbWk8QcEH0CQfD3wwLFtZvokj0}KAW*OzFQ)o%b!zIH2Ew2BihAu-)M-qQo<V2)
zCk<Y>rgv-DlPM@NcN~POfW>(6iYu?th&~h|#NNtQ8p{8Ey~fT0Zhof)#fzOuzR_Nj
zGUOJ#k9-%UTAh;=)i0jI$I32|ONf$n_t~{rm6|w`ZyNXuQ1s`TP~htP=h5CMW@aSx
z9z}+E;X=Yu;*Bb|jLlyuMJX-C;OpMqomA;h-ZZ?PB-T;gFM58yXv2UotqNCn-Og0#
zO_&4y)(7?h9p3qD)K$P^A0p7z!OaqOGUH_G#Lw_v$9_Z|DVXBsSC3~M?_-&(h9M-H
zlj}QjDyuOA_yl{Rn#)w8Xx4S_ZWOT$<4i`K{N;wLt}zePo{S^(O9pHO2LZ|Pe}!il
ze!?4UkGT1;nz`fQs3$$9jx1M^#4L1vxutc6(JfXm5tC7Yse2$CSSUTp7W72*<>d*A
zc6@q`TG<X34eNz^c{`^zJclTKA<5DCRtlwYaXW-*Kgq<{0iX!3PcLz{5Ps?__`Qzm
zt_4w#izjO0Ms6gr)Uj=vS3=sgr~&i`)cb*)0xE4mA|{#j&Y#;9@SW7{o*R)x7l626
zQURw3=YI%-o%qi)pD(R}=ITTc&nn*3KZf5aVqPsmm=+Efj)S4dGwg1WYb#ECwE40%
zsu}WuAb|<s@a%Bib%Hx6{`MKke^f$b7vH|YQ--Xf3!W<M6B2sr<gp2sFmKo{1p}tt
zQM-hfQaB~<^ne-)xKw@tC+_4RqnU#fPGX;_vPbF=7F7|#>Pxa>a?#6dM3f-W>=o72
zg-#U<3p5(nB2*}8MpdX;B{@lWlgbN!xH2~7>e(M9E?P#4%wJV0NaQG#DK+|3QYuiy
z$cY9Z9pkVw%2IHk2sA276CVl_ErN+~ASa)d#C`gv{bWlAe1J8S&smLauawAjw*0@+
zM!0bqlGX}WK2O`5^1Rbsf4F3#;IFWuU&29IXcT^>@0n!~i546rF|%_T4&nuIg`^)G
zcHFWwANy^~iwymdBzTB@;<2{*x8CTi6*ICI#4^af=7ncl;`j(}hyNd}a;kJ!Evl&y
zEgacinss;--MT;@6(%eTj==98LH&vP8>nX?b0AtaZnYDW%>-PC#v$RuvSz>3@-O$}
z_M*JGT=IF~RwtwB+S7wuTBpS2kY*^jtM|<m{_l+2|2Y8g`a0%DPLQ+bN$(qZwR7@E
zMb#N)R<5^M=-heC@KU)uFON6tz@K&Ccpqp~71`kmr4Ek>eiImUk{Jw9oMjfWDP%uL
zXYFSLNcy<=Qg~>6sl20bYp@B<y4??vW@z5O1K`5ufb(Lzwp1exi#%?va!1zE?o4hR
zc*qW!(CRj!J5FA=euf}v8Yk?Y?k0l1r$KYs?(TbVq6#(TBTu-w*eNpJ4KKpBMCvWm
z(i}2^B%|imeqf`-B)thx8A^V0t_M0-Hr)741SJj@!uR3we`yU41!yI|jZyLRs9yT<
zw*!UIZ{d6DBF?Z!-C*yp|6|WlwAt}c*A!Vo6vgUBh>d%yX<Iu284=PZ3k#(A2T^4y
zN~rR}9ZTq&oDxDvmkg)-k4@s!MK>>mT7Fd@uKn+|M&|Wy`b!{*2Nrk@SLAANJbwYv
zstWTKmN+bV;h59n!OjW|-*XkI(r&*>54W>c>t8t|bq89R+PUx<9iMCZJR&)ZJZLOD
z>eC}}q5YL52?bN?^l~w*C!Cm`3u0%4K#Of>(*^J-#VPc60`tGZFzRh<K&cEl6hEhd
z*rItJa5FjHtHTaO<-vgDnbu4i{QBi_69%s;ln(6HLs1VkPN|Ei1hE`AIFVnpb1ua@
ztk%T4J0wUo?(T+#Zc2b471W&B(ihQyh_5wg*Fe4Zz{I;?ZJ}Ka_25_Hus2888J5nA
ziTWooL*t;IYW(8othYkFA6sU3(PgtjZ;?|2|BR&-suE3$&ZDHXb1-bJ-0=ocG0VWM
zQK<+rvatBuK2{p>6`6ck?xgav3pu$>IOPl1e)WZkl^Q{>E#34TncYT@e~rh#-g0$u
zkS{-PT@)d9r{z#RWW|`m=0yYH(zY7nyCqhIl?9i^UnKAYEIRpI^hcKCG~2jx64CDf
zyxR`y1K?bQ5GlW+W4wndfeDVcj6Q>=HhL)BSy}chg)TwJ=hTKY`1FwH%4xk0NmEx@
zRbRsf?f5f`*XpKOgAk$Zp)Pb^B*zfMevauzfv-g;Wcw`0bzUK9zXc|K8!YU?GVFhd
zR2|i#;>ffy+dVmJ-Z5r&>mD(0-5Xa2Q|ZCzFD-$)45kI=T)YenKGhGko|mg9_t@xL
zIt<EwzRX)Fh92Xsu3*2po@)BruzQ3_95m#ib8-bsGd@jGEBjb<Qlkplbd=rlomM>{
zVJu(mJ>=iT^e42@PSp8S>}fnb!HSmR?`fn6!ur{;`3>Da1Y1uioQUvIqfPxChQiU*
zUH&n4ny2lGlJ1fBR20EXmPmhZJ!EiBa2sEq-Zp~$d8c+dr8RiJ!KDZrx5E)J-r4?h
z@h8OzSLCut2dI!p&)*1X-(~P?VbeT{mL7v&)mOzT3pt$h^?#(<-m-AE>(3JVE@Q=M
z*Jf^M&1_Z+;fJsVE8(O%%OamD-s@|bp=z6nTmBgA<Sv#oXyKO(0gPkwW{!gGtAJky
z)9&T}NlCzrS=cX5Y|PWC&=<#O?^5|?#$12;<)&%axCZrG*I^LHA{$rt!0+eiuu|?G
zAPzdMMp`q(BB>lW`{7NmUY7eA3I9~Gb`_@fDvCk#koYa5qE_IJQ$mirB@PW=rms|j
zp1KeP#_w`nXckuI)Yf=tM$F%`I4EHxe}k{tcp~nmcL$`G+^^p80Ka~YXC2@vRHHtr
zg#}tEEPM0ef7W0yYs|$@U0b0dohKA|JtwjgGD-NE#$C)uVnp+cY;b(YtmcIq3=p?+
z7Rwb~U}KMnSpu8~{%KlXMqSBK2or4WDP-I?e%*G2tsq!t(+#UGTdYz<VF5F+n-%Ah
z$#@c{Ha~;3?>*{@$bGdl{lOI{tU=(e=%|=Y`yoO+m!0SYAzuKJLNC{2#O$NF7ePrM
z<GX6vCcAwwiIR3wmlndDNPcs2z|mRc9{dpRarw|)Von06RN1PM()CNFVVYJT4w9WY
zQ^nqwn@~-NoRFOrd-(P~szCS^Whp08X^w~=fRcT1uO~b4xOHwZL&JFIuw=SxR&iOD
z=H7uo;?>uLX)7m5+=jHpLpp2ykF{2m&vGAsU8zC~xoH_~N%8m&r+FXuE?jR2xER~s
zqI7_~OqMb7r>`4-9rG57WYIUi`nGcCvll>1#6)!3ys&U+kS}|$c2&2Fx{WEy4!r`2
z)?$^eBqG;<^Eig!z<g{rhQv@xm;!vQ3=YJ}33?TI1&rfZJm{a<R-LN&{?G`o-;T1y
zHu9^Xi=75Hs*A8=W7=Y9RS_ylS&2pRJ;dza&ARQ4YN_%bKJr)a&$0io;DB^c9y&rj
z0;8?8>}1HLqzb3ls&5v;#3US79w<DHLKwyl4LB5of<rO4+P%(7d<$*9>5#~0fR0O<
zL6&2XrSSSuf<4~2;Y4co3vlIQ;1M;hiJkLy$(lq3#g<xyvLAh?X-$C|Fc<Hx=D85Q
zlEIxup1g6E`Z?0ip4{j~E|QeW@BJDRUQs}!RgdaSUzTkR5G`~t915~|L?-Lvi;(^H
zHClE$vgdAXtA?V^fVUOE#osWKM^*NIlKAml!VmDX3Q6ogr7z&-gnR5%u0F!pQ9J5d
zrb<Ec)`I#dwVk4OD{ey<oRj$0SKEZno=6RBjs?NzR2RGs`Q=)>(NK|JjvvJ;D_BO4
zlx&y@&XBtB(C>z4RBp_Dp8kU931tm-v1UXP-FnX5B#x@<lMwQUW!6!R_@=a+2o2x}
z{pVZic7s76(UxV47@vgWKR`13z0%xBIMR>4a=WX85D?rV$=hJFrDORARGc%s=pr>t
zmIHu-d!qJ$=lUT8hh{=7z~Ob4n_PlkF^*$WuhQ70$AG|<=H)^3GhU-c>RWY=cc4pW
zKC^&2g3BWP(bdfwRHYeq#l-B2h;Pf6>!T<Gg(S6>^?-hUE@n@oOObNNZ~d+|A(WZ#
z-@g~+ef!&dg2l66w%tP**z5tWliG7((&d~Xd-3zR8vk;HxG@qBA{ynu-eN+6W=rjS
zyV>*cEQ&^@_M#8rhX!z3l!L)Q@sZ$&&rwf-?rA#TByzfAB;+Z|?cj>m-WwW@Z`KC?
z-f*z?WVBR(gAcG&b_>#qx7O;oxmoTEAP4AeiG4pt*oc`6oMfY_m%G;`g^+gD`Vsi=
zP0B3uE2K?Ems7ecFmlhlHT=ctfh_QQW8E(|U6^PIK2%?+lG9B4wo=8lZx0e;p2Q<X
z3k!=S@NDpSodU)_l|2fy6WCeCwvIKo)rE?d@Hm_Mv_tDLj{^74--)HJ&q<plzHDLp
zqv=f+f!D8J?{W=A9FV*?b<z=`w4Um?Db*7a=Iv7tIWJf$n@jq;KT^EMaeUN63-JS_
z(vg147}A&(_l4K1i>N;H-GK2js?U=C7UG^y0`Q;o0(imsh!>R?KAGa@Rp+m}OQW!p
zdOiVuNv)k{z3W%nGm0dpb}x%%nYAg95c8##`PG{C7YjN))BPh%6MXXNH+)h8M>0$u
z)C^vT70A~V^gVL`&DvT+;RfcS0Y&J=L~EHME4@F+CXp!^xV7W}@ym`w_BIvmDI1bM
z<(3Mj;hZ8wl_bA)X^WCFq9A$vCsXb1Cl7<Jb-)(D_#nj?i?<5J;*OZeO3y&NU__^J
zg3KE5ag{%`87bJjX4G|~XgPe~2=k>KtEA{5rIfxYCV}SSAO4MVB!#L=UQb%*Y>#8k
z6i$Uf9*J>SRvgE~njJP`qbBHYG?EL3XN=Z^A?GIek9Nkr6QrC(v!_$OwwjV?*a?|s
z`&b{YTy0T#<j!(nWdtH4VgUzLfQnP|SNsrFd9(uao+Xvmwr*JW9ak?Z*C~-H&?WfB
z?8;<@nnbz`K9N!7suRtN&WG@_i<zDXpv^-}F9t<y7!N)4+Q^1q$mN5TCv`8k0fu{T
zshYpo-se3u{dsi*hN`>{pyf!G(c^BU0|~x1MH7GSvI_+6+eF^1f8GU@D(Zd%k#y9W
zGWVg5tbKj#LtPz|N|vmxGCbjxUye=pccQ{Hr1QMIPW?IZE{*hUEunHc4vlm2hV3=j
z-3q;*yV->VO>79y5h8kmPn~i5&Df*;bze_p)v>=(`DTdMnV*L+HNjW+T^^=ke0I6X
zZ2r}|W6H!<q!Jbr94}pVCaTpb(eJwW&wF0zdKZf1yJRVgeA<8m^ZHtsfNFv0;#c&^
zSRKm7U`~6V{?XvVc!dC>M2OKEXrx#l-1{q*9pb(E_LEu1$`k8#jaN&ZL2L)>@(2ef
zaMynqq1I0Ew}k4~W(VEg%SaztY@MZ-Ze2Jc?32iug5nOsLN52i>CjhLW&<#Ur(2#c
znv<qKi?cC4-4aJ8HfEb_i<>~UfV3?)!BgB+Vf8KNc-v8zfc^*zA!<D8uqf&k4Fl^P
zima2xTSoVm7-3T1&LOR+XN65nd$FH|7nHTx#U<?tfufsBVwvlW9tMuTT#C$tBSY|d
zU?---BKCeL?nR+|RP!3!LCMzHZ~24k*JE0(Kx8wYVIT`+C`^a)GDl2utd%b7=iU(S
zEwgdMH0-^lWuEr?mOB~b4Nqacj;k8%jjB|s)6%~XmJA`^X0@5+JUC+f1KwiCiez|o
zQs0P9OPfRGD6-`Mdv5MfsBh*cWDgf48Qisz;n!JYshdVlhwjV{MD=){_1fmyUmuKw
z+?>%vbxs_#$^`GAO3trw`f3`A`lTar%d!m2!)^(4yvjxENzKK*QKwaOHw<v4lbh#q
z`RHoeXm%ZMv6HFu0W2mZTO6+@(skZ9{JH<))5sg6za#2Mxm16G{%tI1a2F$*Mc!Oi
zko3IhThu#gK~T7)eZg9}Rr`->VYSxKb=G9h!idTH|G@|<!y;7a4;i`qKQ_A%HJPb|
zk=GEB-PJXHKx?&ev4?4Pk=vHvHM!TR(aGn{p3%$ougr_9i^v4sj5i<pYiY4@@2xgm
z8MhTzeWp=xd=lKwdZ`?Z3*rhO)NZdz?Jqd52i23AfsGiqld3Fss#4nT9Jwu%fl_-H
zJ_NjX3PJ|IT+;0J{>`?L{#`E*I863d&%sfQ`$AhA)hu^&=7f|!$C?59oPUMO@P3?7
zuOe6Qh}nKRuumtJMU9xDq*_bI@8!jHmQA#atJ5UTivRxlLiX44<DNcThR7ClRtcd{
z2`}i~WbXa?#fXVHG@d>9vO~b!d~80n0W1SqQmN_k*e4P;va(fo4x%r2z+dmQ^|{DP
z&fjR1%|Np^Ei3@he;;H!R2u8WJxcRzrxX>~UfEh2D4e8oim=s)Ui+8naBy4rK~|5Y
z4vJhiOD#K}m>ogSkGra1rfjkUC<rRg^5G{B{fvGvUce8@8}P0K_%bFr<gRqW#<aF=
zW#;QFCm~M$&Y>?k&q0C6uScCc_^EwDQDSSZ3Sb8dIMCJ<uPCP^zc*$ywxk+0HumVy
z=Hc)tY+}X4(pkpocc}#&^E{^58z_#`)2|BM8dcfE+z-j+L}3PiD3~?>`X@dJe*n)Z
zHG^icbMkkwyOcF7holLQQ<#QmK^Ok6CR8?Q&mj(oAZxUd{PB$0`RGTH#UER=znH}y
z*1m%Del5yDHWv(;=_M3pTcA<lOoFnWLw-pXfqtQj^Cc}DG&NX9gzFBVV{CyU3Hb4K
z|FK=j)?qNqDjb5vVNQ9WYHA*$7(sHfZPl5ll49r1B}{j$ve#Lwl93tR(=*D{OQCzR
zsvm}nd>4v*?CG+o+O!6PD@x&TO|aVa-A7?4SNEpRTTC#+0{o_1#WrO&R{$V#t9QL%
zSqrs<;7K8k#jzuJnJUWgD;6?sm=oLoDZC+AP2^8#Fh(Al<_Bz$n1rEr>=_l`W-H@$
zCehZn5z3#Jhg<|;LF)9C@~)4ld;1WyMeJKpt&$H~k+ObK8H(j&CebHv_r{CE$(i(X
zN@tz(bsYUE@@?xYMd~jvrPhh+wH|3h<@oV#yRn#%b-^_!M0eAwqI3;07Sz>e*0K;&
z<TkX7F743&{Qb%OH@}Mob3Y!>3qL=ZxlQJVP!GZr^SJEedj11vH4KA7_lhj6U;GGI
z7ZI2<zHSh-I4(A5@-y-J3nHX1TTOiZqt~sEUG;ryYoO!ve}0{^D+uV!Lvb7U8Pw^N
zSAq-ujuP`cG%_81oGO&1c0@;ft%3Wv{P8tXxz^+1_+yTvXY>wfUzb(4<#iME2OqCI
z8ZT(xvl?H$8nOQ@I?=uCBqjv%<Qrjr0+%~XOr{v(_1NJf^u{@k5GJm><7TFiW0R(-
z?o3E^caB~{$PYu44TB8a(W6K6J@_|P5d;I2Jrtfdb!#J{m6W4c#LsXd^N`~nUB}m>
z#!(M-k%nht1`tKUZCSJK(y~u&GiWXpeR_RG()o)rw-p<m#9yqnpm#?>uB(Lf#XHIM
zb!^<=XkndI6okeYeXHn1O+x=3rUJAm-MLxW%GR8jechGoPyyE+N`no5K|9O>d+rGw
zz#<0WR`FqU4d7$OVeeMr)ZsTke26AHFMCM4b2{8%j7rLz2f|4*C3<RS7JH<OLV>;7
z4V;1x64{%E`{E%nJ^#7KPXbf>hqe?Z&M-RhHqMmwu+7kAaD{fp%XaFO{*Mc@l0@?b
z3G0~AfK|_LlauJ(5cU0O$m7i1Em_<8$|WqFCf~4&^(bDTIMX4z+@zs3#U2+vhSmW`
zybG-3!9uxbVi#!L=86lsk%KQ+dY0|Em5Bg^{JE$uCpE!jj@o_97O@|%b^_XH*NM*(
zpAQh*rLZNcdD2hp;x+lLmiXeYsX_#0p%#U${9wFe#s|h$&H}rqr^h;lgKAvM8H$39
z(f=s*yu*X!Dy~utr7|#vmBt2LD7S)`@Tf|dZdc?TR?F-((k=2(NX{4vse4uthU&MS
zxUmHU8F<v*%7#xKN!Y1zbuY6u{h5^8%7}f;7^lQQv8dunn$)ZL_d`GNj3al3WzL97
zWiQ${$WZU|yoHk*yZGY^ABTKl9b?Osp9BlTJnZt&cNVwCmu0K2F^H~<*<(Ozw87{Y
zG$R5Bxw3JAv$;l&A!dF~<9a^pH2gb5<5Efw6<>C_;HfAwXH@9tYdUCej6VIv7n`VX
z3LkO8LH*YLhhX$+eU&ZqKYtghaZ06aonCzVx$#~|Q5zQp9uJlW9%w1qrwc`YdrIP5
zjNYq#4z+zj<f|69Q~QueR1Mg9v(Z<>!DA~`DsNDrWCIDDer}fwkB@c?mf+d25xaY6
z1J1N7zYYlnGAhc9vX9QFQwc2l#)8y}oz%Br`U-mE*E87{bNTrSaAHZDm$7}&8qgv(
zYQBa5{JFVNx?h?g6r0>A_N!6C_lr5}d=GE7GSjTPPu07Xv%cV11rsWcm^kZELaCSC
z=&UARq}KbYkbm@FLrf*NPa$*A$Zq=LXb0g^;~;18-t5dFXWg{Gap;Yis5$GfjmpNN
zK_LI%RcuIW+gvX7#|Mw08-i&-ow(gPoJskr_)%>eP<qds63NGq_Mc9G7JlIeQc7Lb
zZ4p~d%r-9m3pE`Q`+SXr?Y=MpX|PC99O@3!%HA*^rlU?-=#LKhbh-7F56lC|kE1aH
zQ3c1MeCe{o+pHCg<w0BVKhN3I*-oB9&s92ljY)fJ4aHxrwabs9t6-59zO}_cWDtbC
zbqc(NzFCqt!N2CH5T;#0dN0^;QI!>XTfjIMIXH<$#%x@L<-JBUMt{5LZJ~_7wi{y_
z!&Nc5l8JtlmrdcQ?JSFdHYM%<_%teo8~0hbJ>2h^>+4yhP>5Z-x+GUp%u=W8%_jh3
zOcHirCfR6mN}hl#!Kje?5QZaIrK{g`V59CL*pgLZyG4lSgj?_RMZCMAw}ygj33J>t
zR(T<XwhdK*z6JB;5d?aW=m&KyjP5c8E5U|DJ}tAH<i5UT{rF_z2HycdXNY~1s=}T!
zzj;Rn#hu4E2HRAbu67iTEUtw2gd==FQ0OeL^u`drHpVOH0pp9%bmL~=!7)?eWOfX;
z>Jd5edhNln+7$Mumqr@?kcI00NauFPw`kkd!EQ!6EsrPBDk7ieBwSH7H<D>6C_i3w
zVzoJxTdIEV-L%Sk{79o;Q%Dk3s}ayTxo8}KUrKjo3Wa_yG$M5&<i2xql$$MCNDvja
zeH@`9`S)d66GfLQQzom|&dHsibDutL4Vy(({uDo3C5;{~@PZ<p{*RbqS5Ed7RS~W8
zMmCR4;+m2Dqj&}1eT-cXEoE$`SG2ujlYhqbcn3_jUZKdYN=<J}<bG5S%nMJLzPQTl
z_FO!tAphUNp=%~bK{H~?W^#|lC5U9FYKmR$%gfoFk6T_XA?b+WljngF8}Gh#Aby21
zj{n1oh!t+wWAQoD34{)99J-ShYgE>SK|SOniy_Rr5J#2k7A6S|(aR6nd%N^wFitT4
z?Bx&HZTmgFE_I{?mSZb~xku+54e8ttB9>6MF>>&RUY`0u_Amg$zvrPCisqnqT^L$5
zF#7);#@u@aArrkuf-z<D#bWc5CHp=p?+|F$T$ki%9i<lgQyg9TkLwvLph`bNRK-YO
zZtq}rF9q(nn;4h4pXiNMfd@Bd38EjW#0{9tOZ6-0etDu*TeZF^`hEm7p~C_A{(rqV
zY6L}Sfpl=p7tZ%LuLYnNCvxMhXq}Y<j=vR)?-q@8>N3`2A}SH%5MbAc{$oLvSSbw~
z;O@0mO{(ok7kyo9{mV>(K(E6^eno>)_sa~5UE$mloNdGTr6RL&J6@If!Qrf6DU08I
z!Ve3s#y|fy_*=^H+~N(FZtXX5od3JLWn|X!cs22vnXhV{==_=i$L%7&#h-{S`RGHR
zqf`r7sE1Pwce*L}r4k1vh)wtjCrqLHP7Dwo4`4d#V*5T8ySd<ah1E=H0dgq(gG{RB
zO?sSK!Umh<kdLwS#xBz_uqsn6>~Q$p%thu)hwi~W^e_Yg%`zGPqLj*`ELMFv9*5Sf
zzaDE)Ke;v3aUR(V6~#r1YN2NX`k%U$jB;#O9Tj=3Q;?zX*zp<+@w|33F=pfQz~_pi
zAiaAyL5Hx~X65bQQl_R~#&5#FX(u4)GB2Fun{QB{-b_XX0E_6t&8e2OgCfbtkNVf4
z86DvA)7v<2f=3ns`&^8@TP*|%$_FG*HoTXy`qm9Dh17bE$1@E-jQrfP2dElHkpwU*
zcJvPp4qB=_Z}a;%Q8$sPMH>$tS-6IP*-mebKh1Jp`11CN?S6-kHm=*_s~$m-zSB;r
ztRXk9+9mryu3IHxLOX&m5x9r5aFgU}Vzvl?{|uBNjn+NEhyl|8>x$U_kZFqllPflu
zU+!n%%wzqHC;NQ-Haq&-Ifyhr!hTX9>iSxfp6Zk|TUWqwfe4n7gugmkQ;I|M=Gy=B
zpx~A3!wZd~&S+mzS<~0c=2)sIBDbP5N??{M?JnM;ailEz&43~;RoLvjW3{c@5h3A4
zo5R7UBd5E^Rw^B}mwXbs^`hWfGVBd+-xyfY`_V@4Q?LxCH_`9L!WLj(=EqVs8Hc+_
zS`EK<hGe#_UoqbF3P<x9O>KNrD<T(vsZq1ty<D%;W$R>si=9AL(}Rz?$}p2h{yUoV
z)feCMmZ!V<MuKg$*&v4tjcdDZ-8>(Br0=Dja*e@rG)byl{l`qj&X=NV!jmy0<V4{y
zor#igMqvEgv6ZTZ;(4@%H2R$(#i2qQdc1h7Xylx8)(&o5*Mb5gWhwxFF6v@g_EAcQ
zI8&jcsLd%3rm3M-di(l&YTADVTW`dZv#uM2a-M)zj451L&c9zeFLfV{FO#M3tn3wQ
z2>0x`OB$w__?{78i9nAF76UNm+)6-TusGWI=i{8A>*Blh5Y1+MMU3<iKOeih(k8ps
zM<o`ul3G}x24f2glq>Jl2zfpo)a*y7qX=>xUhQUEjF_)>;f%?>RH1CUbk^y}$(L=e
zIRr$TAhRytv3bO5)}4BJ$rRcL<dgAlE&~5*qg<)wsA?&$4DNyjXSOc({B(#a*`j<G
zeNkXnhsghJ+#EZk=YN@tRhv{acpz-f+jeVm{hf^LVd*uN7kS;)b(L5X<>2Qj!UF^~
zgO}}<6q62b@f_vPcBBdITIuH*bwT1^j+#a;<+GC;owg`I;M7ukYBZu@WAgm{0#l{5
zs{$Ux`fKGCagb_V39l5=j(wR)9#tQjp9vp)rjR<DU66!0$&lK{Xs&fHq;x-&leBI`
z^XMU(gI(9yxC}>M`VQL9yQ0ranf0W+`J)i~A^bsV^nIX*ePegC(*U&4x5musDXp5%
zWimexA4wkSzeJEW(%Z)leWs6(t|>WtIoxs4hE9+Vh4}^zN(ScXjoK9<kC(!#wDd~H
zRRX~ZXn`;sJrOl`Z#-<`sQ!le8&z@PT|n4hJ>WQ8zzXc<PKP(hIH-~hoHOWnw5;XI
z<Vy=(tN-Q9XHT`wg<JL}Vv?OCv!#>!<B`x+UfvSiK>6|+&rn%~FAxGr>NYZ=Fxwfw
zll9WOp4$~|5mo6qzIe!fW#rV?mAe%lK`@!0hctWg=Au1iV%pZ1Yx+E3zk4ZC_-4PL
zs5MpKP*y<{Tdc}pDZGwzkxQV7PiGi3hzT%RJ<%Pu47wF>P*983L8fC~{J-~h2qS&y
z)@#evcb}O}EjbvR(u?C}(;CC9thy6m_bhQ6RkX8?Q=L*GA1!8x8GEa3PP%MVcVY5S
ztd%DVWr`a1Rb(H~>Y#XiJ<(t-c4`Y}QRMB1-dBo3`J{R6x+u%nT0ZH4H8Yjf^dPYx
z+qD(zn?l+Z>5tZ!Ojub51@XaDM;0%enD_T)D&9ID_1aQ6==`L%$i4WJJREuzi$KKc
z>hzwR{XD`*QT)w`6MFC9=P&^{0kNlZte(Q#1kNS#`J?UGc4DKRJA<hL@CqEY0~Kjf
zd<$NYyn|X8TW9mN^8M4UICbiZ&FZwn9ly>HN+!|*tNA<%pA%8sE4tnFQTFf*vvlw;
zcJ?9X@NfZrn|!itS`I-~vCWia4DY2iW2o0qCQz=fsC%JtqoNLcCku|^mRnJM$%@fS
zl(oqq@vwNL=w~^;;z@vR<@Q^jm;pH&N5gJSOr>%H#2aWti8b9FoKCwURp91IWy_@(
zYabGGhH@aZ(?BWN!FN^RJxOf-n_!nL9Q|ye$geiaH?IfYjQuDXftrDMA)ee_yfY2C
zG7(knkOk=Tm*f3tr5bM04=}tl;zG*1tID&j@!-`U1~2RtjdJa?(XY4N8LT4Ia)P0r
zV;0(*89sJCT3Aqo?pMwhDHI7?!{x2PgvkrFM5Kf~oS|v*J~;ffLz9xKPkL<PSVB6o
zxh{hN0Rw;(gKxQ;7rHZ9-zS1rB-(xzojM2#`)2-veqoaDe`hC-9X2XCxGqQIY&(^8
zd+&_A5|9yyiZd5)-Uy()BL|_6f3OyRLU{X!b6k<zyBkMPgx$$Ia$lIK&-c%&*r=;?
zXDeo1_M%lfAst;F?!0Q><9D7KX9g7OAFsb8+W4e%=>OGq?SW8cZTvN+%dpg9Tg662
zbT#Ng7o+TUp|I6Rl+j8m<uVi%Gh^D=Z&xU}RBG6*t(_8GkV`Y8b`2$bwJ3LshQyG|
zV9d<-oOg6vzVk<<%sKBl=Y7s|`Tc%pUsFrox}pohx3?XRN!&Zj$%|*MJSyA{6C(&1
zZ#G$~y6%R?`s>*zzOs?Kk*;oTx+X|yIp1d%@d!ot)Ruf&n+S)uHxnO%XVTVpY@GwW
z?)4~M+mL9tZ}*L`1Eb8k?Z=`QJtMiScA3dUm*ef|5-rc;n}y!a8eR*k9BV8L-q*+7
z@^3%lgf4mcpqBD=XlY!iCJK>jj=qyH-{j9h3;1d4S+w9!h(i%bVhNIU<Z@Bs88WK<
zTLapw-Czp2%|xNa7SiHPhUG7fO!G2q6YnS(){dkf8gZPUpt=NZ95m$gZ+t<SgtgoA
zUn6>|J&V4xXivvZ(v`N{s-h`}njZI-$R1{~sMU)iZ=9U{P$z5@G`5CDJI!R0igQF`
zpDV9|Twz=lxA_D0pN<dze&}))`_;BMmqN#;n?}VJr%yG9&<p;2NnmL24#Nw>14w!p
z1hZngBqS<j%0=VUf-J_W`=%{VT?qe0={v-XBp;sXv@dLq?Qlk3OJ;wD*44wZW~+xf
z=dE@S_|4!7e*N-^dGY9;Y=2ont@xR24>Q>-QM$2cZB-f!;$NUDd{`Af5><lLpK5MZ
zt64a8T38~n5P|;zfM#G>g|+o-qx|L}nya*8+XFdDlGtNJ;T)k1&(3DdLuj$^9E~Dw
zmNclSi#-YzT>I%tFcKrB-k$t+=PP)vM77b6Q7YA5(dl}Bg~F&^GiK+_PM2M@0NOX#
zyTh0D)te2h&gbw4Wddm%<80v8n~2?pHGO`=06yFU>j7oIAXz=&0BH*kF;leG6z+{)
zZfjjw0ztpa(V;$XgqNyOzZGG35#mRYlyJVeM=QGe=eiITtYJ}~;)2AUbvICw;Vqx<
zzV3-IEh}|{aE^XPsW9g*&ww2-9oR+=1Xe|hskB~X*>AA-frVd56}lxB5T4`hNX3sj
zrTQn*YSVJ29f%d|<2`a!J+J-t1S}}CSkdUppae|C2asH6pXt0CT7UgyEcHnJadJyf
z&X-ri^xfQ<7O@)puB0YNLw)<m;l~#gTSW-#4O>7r^8PWJYQxf&JU{2Vfg~7!zS%^c
zWIbE3V5(Y=2O+-0j&MH6z#+ivCY{zjdCEg{VY!V$fl;(mzlZ`c==l*}!vV;$d%EVk
z*MH@;Xfo1)n8t8ppC72LId?oSiER?~By2@R%~8+8g(dZ#|5SII)R)uLHcf7a{2R(2
z-<pOTTFUBwb4>B2!G<pD3?hMcs0{eufy0|BeUusyt+gx<a5dpM*S;fdnXzxZ!8Tip
zq`r#1D^KfNIRG4b!W7F%x>}|E=EG}E41d!*e^wSolSpZOZUJ_Z&R4I8x+iZ8a3xu7
zH#BHiDyba<Tb#vsV_Yu518@Bh<X~nQH1^u~K%+_-yQ|C_R?!|+*TClE$^t3D7s|^<
zR|FK$IX2*+0$RiW?C}nC8f6SgJT5jey88SSCViVnz7zO74NcRF8?8FJ7M`goRF*D=
z1i|M`;C|mFlcJ)@j*pPoS6mhB8#yoSR^npy-Io+_j+qq8vOI-E4ZyuT)LCCXu&~g<
zY?85ooBZKn3KwRa&&(zyQhq-vhulRqGJ}XB8UU<XnsG}xan}C2y>){&B#%<DKibNJ
zvb@8J;DTp+_c4C`irvR5US)S%-Hw0dsOnEI<Vo!R(B3eEnPsCPc8Xz@ymtlLL(t*e
z<6**7T<7T_tNB^;52VxS&qMSY=0*^Z&LO1cN6xXH6a9NmlIVNDQbIVehlD*)E@M=j
zG-xGesxq1MvtwMv9f!>C)P6OwT~PR?aBkrYJR%ONu;3@Qv?2zuJxgoUzI)|Z4R4+(
zh^-D1>gR@qr5owG9bzDd^$~<C3@UO;g;`qDc{sSfWwIcAmgmcUC~E-1Y0zw<6u`Vm
ztD=?`?j(F|y`N;*CZX@qPA-kx1FzDKu2W8Kl5Mt+@e{DOfv;_Jx$}F%@7h0ID38r@
zKA|QwX}94gA#V^WKxg|-GdD2}*Ez9&Z}w4^-<s#S%>;4jmZZQMq@`2Y8{-aD#T-Of
z=B|Q`63*k}xTId)`&!BR7bq%I`>tV+EB1)q+*X}$^(qu<gqYQRP*vLAsDWOLPa7*|
zQ$;hpe~5gzN#Lm#`^dcH_{cWfjyn!}e~4Im-&Dmqee&83YHXAC>k0C|(u_>4WjZR!
zQ3+Ugiiveu%<|;b^tx-)`BDSn;=+G$v;hArQcTEfRm?7)0Z&{>rJgLjm%f=v?o%Tb
zm%1=}y|i|(Ri1)t|1O9cJSx7FoAEmD?@HO4K9104np#Ct(n#Ab4jSpuAut4=4i~;Z
zVDEeSpY5T`3ZoG<LE*+;50`CW-Ul|bwQ%3ZLvxF69xAlgOsUPD;X3Tmk<=<es8tAD
zgr-9G?pWb0!*fy^*gIVm@mI?8Dt;#VABN)u5lBLRiRN3-N~)g>W$4u<E&=rG`XUXx
zB)-q<hY{s{+sM7XoDdP;GJ`UBJdky#<OD5;cKNtFjEj}_nGiERYTI~6$Y0#7S<sqd
z3xpr*`RsV+5mbF7CR%8cT#~`>;nhAW*ZTWP=QBVK4b;zPBpO~l?5;6aaYLVzOL=6z
z$@9O;JVorL=@(O1W2wt$`Wyc=__By*mvA)RPXGX7s&8K>N?CZzNmouTNFG9Z*&&Sv
z!{WLOEXJWqk|+B?hKLcVRA%S+f?v;<3$AilxoHk6U`)aE$kn&(-?A-vb)jiGF*kY-
zL>9kNZNp#EX~ipKMg2h`s4ljG@%w0s(uQ@CqIcbtJWaVy0E_M79ftb|l^*0SSMR#o
zZxqL*oO;27u<%=M>h{YO-~QsEHXp1QO4*Azd>ug0^XKXUW(c!d@qY|~_$(_6Ho6#)
zYd8)m%j`!rulb>^26P?@@7FX&nQu1y6Q*cQqX=S%l@TS;c5GKVWh*80RyLrX-ve~?
z;rKu)RdIG0!Y@DpH2S-dgS-3>EXf>n^}e#akZyLwUBlHzNdp(}mHkrQuUu?|!XJ>`
z!^n(!!`A+d`LZ2ovAsTHWZ;X<ns02oROL+irySm%S6a&~23f-DsjSxIPx)l96&A-j
zj1=bb_peSAydy5g(-c{H0h9Q2RA;F4aNb__ln822(e)7TQ5s{s3y#7px;i#>DnUXy
z#o{hyRtickCixM6WrSZje!G|<3Bq*Qeth#<qLB}>u>#t7PdW7<AX~4oj;ZxuXN3*!
zjds(BTzy_=1M*bh)aj{=k&P-A65{`;ner_7)I+x<3<Q1bo}!v47o00pf(rW+1vyky
za6xa)ko~R@S5&46=f?!TnY`F!su{<Jmm>|2N(>B-$xnM)u;cX~y4ghC5Y=YdMPaz_
zbPMeMjoJ*rtE1&-#ZK_<yl`7&Xu!U1HCZ@2<sz$JXGf&hInOjDWZCN+p_{Z%af>y(
zW<|EAnQXsXqU|4W=%&|jIO)ex(YvNG3W`<K*VdK|TKe`-2R_lUGDLJbKjP$t>W;yC
zkg_k}zCqw_F{f(@Zb2GEKtQP`{M<K?1R+P>QdL1<=it*kSvp)rGwg^)nK8kl|DH9g
zG?{RV<U);0Fke;G)Jpu;h)jst#r{!=g^!LfoMkgt(o8hVnpJmE(BoI3^Tn=C$vT-E
z#%grPJTh`JJtjvtk_c>AMITe{M6Kk@JNa_o#v><wQ3j&eUSE0x?-*!V1=Ptv!BT+<
z8XO{oXOFUzc*x%qWKIyD>EOUkE?&mYE)agQSP^zOS}93YwP=Iks?o*k=Ax4a93Sry
zb^Y~aLgI<68LbzwKZq8&<re!o6u(pIxc$CCauuw{x4zefkU&#&J#9X;Qt`TR^b!#L
z`Z@9!5bZ1}Z!1`xcoZ_J=aSgbv>Gvkf1HUXE$}lT68H&>*wpAJPv9m&V)x`!H}BFh
zX@H$tht<rU%B&^64f~NB$3-u&IGb#PbfBoxzCOGFyW-(#>5?TN`gn8Q9-IY2h2R~y
z`vBQG-&BRtMHU8db8IF$de<~k5ZIuG!hh#xcvrL=bw2dXc<(fNJ$ou7C0d6Wi--dA
z4MMGy#n?AcbFHeoKKxMUA|HA1^gYc%*BVo>HgFj`^SAn+?fXry9SgT7Hz@Nlu@xDU
zup#{mUvFm7v(wYPAi&D#4hwxP)(X~tC1xx|D06_IO|Vuww1yI!BC6_8qOwc9YW*)-
zLxD<$ERWq36q=9Yk&dx<Nq?o-a-PRJVsD*K6%Y$7j2b~h(Stfh+Hz`#WO~?{=zD12
z*~eL60@r<4^747{SWEYsY!-|M-iL`G6m4i)xrda%?2T73KZ17JcleZ3QP>R8swL_0
zj}p+_f(qWEh=eaTvps`&zjM|u`*ixPY)IJ-{FS1uzWu)7PZ@8Ti8cA6H5h%<=hxKh
z(5TpEivwYLEGL-%ZS0#`O*OL$|8xJXD?9%3^`1bqDX9e3dQW4!#J**a<mjR`q8h+M
z{ZJLoUfknU2}yqrbFP~h{a7nCr3=ZgHy%+1TU0;XfLD|_)!)B6o_*HSt8*!OdW0WD
zUq5Zp+d2!=7u20Zq3u9W1WorYO6S!~SsX(pI9}%TFuX_?*XTqf+~0RY<WC?!XGWGU
zOO*SK-06VXCLivIn!m}q(-6oo$CqS2=mZ1er+mtP5pD1;b+d>jIkODo2!*Jo11@dD
zp&JoDM<IPA0A}I!(@Aav=!zvO6*wipkRCa27(sjp_X&CyYyFKX&#)M|b(|o6q}>S|
zMAyj2#d^dmWF!&XWLo>2@@&mvc7nV3IsuE4K!~`q$*NPwUj{m85GO1j?@hs&ATyhb
z&~zX8@gA0gV8&7wexb?qL!hp~brj(SPh+n*&)Y;($u|`VSS%uhiKIOy^xZzYhtY_`
zb2^-iOhw4SMu^)0s}4}tJBFB9l?=sKSWKZq;h)H3DCf@}=H)Pu-NGuq9={&?r-{3P
z-AE;6+gE(`#eZ>^*(TL%F|o9;+4#WnbvDAe5tC3+2&PZdc-`)f5Xm>G1#sk2sZfll
zl!zk-&!jUidG->;kUjH5MMA1{IH}*a&l>AFmN98Y$kFNn6f2?u=t7W=zJA(Z0J&QM
z-~@#r?FCnHcgDE==e@HWoR1^*8JG9)zB`+ood^qJ|3WwAP%8m24e&~<`)QL*uQx9K
z>Bm_7Fy!Lkhb2*l&iWWjWMLM6yzZ@JZHy|5(hEytBfg#CqeutO9$Jsh__eWGh&X}N
zqo|%6;x7MIZLP^v%}67aHj?AvJ!-#>VLEj$5qndVA{ZG)#R*<yD8t~$dM*;`<uOfw
z$M;vXf%|oeI_U5Pt0vq-HypjqKY~RIVNuMUT@v*X4L814;Ip6$_a=Ti9q4=g1f4-7
zTFk4ZPyO_jHypFua}Ya}3V|*-urqXT-kEfp|80P=@PH{K&mCu6eI#xTeHMOju|D*m
zzkcK?utx5bi!lBCq}SScDnin-Zial!HekZfBwUmO*?17W?PJt3$nM2!U;QF{)AWCo
zz)<U}uv6PLIX;IaZAnI~wGnRvwv1>xI|>{K`<iokvE}9GdiNIn0y8`lVH0mMG`E#8
zc$xxH=#oq;T5MROI}#=W%ltXV7Vnz%8@dIU+OnX|%=Utf@zi+{LFw9=Q)xbjt7DUJ
zb1l>NR1sbkvrq(|&P6C79`oz|3~@{EJ)@>}aL<tB0oI?UcdM>G=9+<K&&6uOFZp*J
zzcQO-gs;0GZoGHabmPqO8{d=Xg}+330WLECXzJO;&7EUV&+<6iUyy->DA?|-pavWt
z*%PJ&F<%ZDm_7}(4{(p+-<gzEj|`?vALPA&de+<Q>(0t}x6L!~b;4Zfm`T85&W;C~
z-axp?TX^{&^}C?yvH+wB!%p8x*+_6qqmVN7-3`kiA*m#)1G*~>2V`(jA}eDw1nEt2
z-=Gw}xKRaI<T$hnlCO;@NBn-9pokPzs6@3W7y!4?Mk*M3sIz7C`H2HI<s(kPs=#4J
z%ko(2{+W`3a$)9d66|2tYTc;h`WYr;-n{jlqOC?vI#3OAI<utsu3nSM+oPv`;V-m3
z4wWwEfY{Q}Yuc7F*>+UJ_ex~FNMbv_5IZsXdN)#zu#+OnCC`Uz_GD~H4V3gHA-*Q+
z`PlfQnpN|4D=J35UzFgf^{h;^^e1witw;Yubh_`zn%jR}FWP)YcYaS^2JAzA8#tM-
zcPeaJ;!^KNi;JGqx}I3gKsc(Z;~wg+G1Y{Ge3jB&{t@OSVTa2&-6F9}-YvqP68k?f
zk=;4ZbleG9!kJU(M)=3-D)3QXaF3}Pp?or6aG^m~S^p;SzguuiW_=ZHmW73LdS)X?
z12qdWU7TU@14ZoiZWBB;-Zkh%yY~sK8V(EBVAUkchGHutj~au{p&uWgSn!iopRU<6
zOP-GI(wE4Y0fgNz<9E-{X`S0n0bXiNv1c%E%?_o~8q7_ZF-x=Esk1tMda)X_8$}T~
zyaizTNJojwh4Zgd>hrpVo>GdU@=(<h3&zt>I4BM{tz=nysQu7=q61Rc%;TJDb|g)m
zTqYV;6i7OiGc`dthB4-y7HOxhaUpmOYk%g>IN`4@X(4M)@qkfvl8Z5GdBiP&)R?@m
z`gAq>H=8%OlHMEqEaO~|<l|_gDf45gBY#*s($v}^TNPTebJ8M3M}bFCN7Oqv2xjoc
zX_*Kd61n*F-`b`HIPki|62t_KR`Q(Xh_$}Ju)Nl&B9n@T(gY~i(QUG4r=)bM_XKvt
z&S!-o(3)nWqIh6_Vd==EBgI3DTQ1yrb7e}=LGYnBdh_1g6Qyf@6Q+ct+Fp`h8=nQV
zRmZ`>%GWVF?02Pi^W4}<lW~2ZWuG!#-vd|cMBFB>!tF)4N5ETjIb|bdk(q<%AOeY;
zp^&ZVb<AU#C{|Opf05?#ABILD%I+*?DtKZoUSo<#m^H&p<ybld0go*cJ&V!F#H+=;
z;u)bMZBdB~|B=UE*o|BguCRgizr^j0rF_Wu=zh*9!Mo_u#F08#(=wmPU5!%@BY^`P
z>wJZ&Nu?rxxFJN(X0A#wyWKN1B}w3`Bm%)4{9`Fs2(?OA2Y2{5f~(&ryYm#-NVs2w
zf3n+W@-oVYv@wi`L0(0OJMTH{QEJzzr+3Q(eU}Tm$4Zs_>JxAgTvybLs}x3i-_`&+
zUx~9u3&$Pv%)(CtQ?b@cTltY-S0)IEFs5Oli!zMiZDT)0vD+2@SmTOl<^j0g3}<7l
z!1y7Ek{6;U|1$@a8$gB0Y6dE_9(~ZaROY7ggf@>a##4d_E(xM{30kA!hg<C5cZ(HG
z0Cm*2C7i#vb_Gh2egr<$hg=#88$(yk)B5uwYnA5%Hw{O>ton+(DIrmsSH4Oa&Bbf+
z+(S65QY^VFZ%3I)lnEGXEW7c@BW%j~)NC&eXL)anFct^r-{LCeqo(|7zBFCF2dQoV
zQp^9vAuRA_SEPAyI0HnK0mkmYyk5yDcSb^6l{RFYA}3xEe$CJGopLJXzwZ{aS~lK)
z=^<&Qh6M%&0$pSz7%aigtZCm~g0l*cOBP3bmp664EO$8YAMA&w-w>YiO@Ks(tgF=L
z(5{g)xOJq0)qE9lX$046RCdMkqEB4pBX1y0b+mnn6uCjBYtANHbiv7*#^*>%;g3Tm
zw${S>G4+53(U=P5Mz))FG^yW2*4yuXSK%@8r2(S&mggAAWI0QP<%ly+tyJO-KP1Zk
zPZqjpvC1xz;~IoT%HL`iC%CVKPrOp5?^iV3P7t(iid#@ITP0qw@frnXx0&x0n-xk1
z)rH8#vd2g|og2<nyi|?1oE=QTy%hA!O%s4rS!NX)rCLoRJ4vo+LKm)kVuvhkHN1oC
z1FY}g%u3x{t*3hhdO69HnHHAtvk`Y01vhS$e#~BHy1@fnnFp28luh{Uc${Hanjh;O
z^0OO{%$mFI$>T)6ldpW4>6h(JzJ(=%03fk!t>n?<9wbM6XYrVcHu%-zN()3qHT;o>
zZ&lA(HUuFMgR)J8rew`Fez2AGPW*7j)v&iJM`-C;5C>I}Zl5|9zN|?nby0PE|6>Yp
zY=v`WT|^@#4FaL&8;|I*;Ta)poS*Oi@f8?<HtPOJ%#j^r)!qqRu*rHIuneWyuhHdH
zQoqM1jfyo7RKsma071iwjAf@Fg*{ZvWG;><R3aPYgLt0d_c`bWz}wt0KHmGs!Fp0A
z=q`@vPjY5(kfbA|!*h*aqHF)SZ$jB!JbPchQ^d^*n|0>}KAQvH!S;G}QVxayf?tUp
zJYWIJrr|D~xjcorP<SR#??Yb&6L4C-a3-7w;y*C|H!yY(y6{x|V8j!W>kLd{^WHFn
z*#v&l3XHh{Th+7cdAa-m<~MRF9gk(X=$Zv?jpUMhmn1w}-$ZG6KlJI@KEA3cU>Tdf
zLpe};>@CNhTn7v7LPdgrE?uYoaht}I9!{Us-WK-d8qKbt*{Yl+vPiTO%bChOTH6R_
z$biD-B3U&&|H17)MLQ=Qd*PHD?!$gzxlG^=PBzNy#8qXH*!zgzi)16yr4aPQISA;!
z5rrJ%!WZeAdaTH^%Y2*e;oG!Uwf05q7Q^0(t9v@Cf9IUUL%Z6@#ga?W{5q5_Q?@4A
z_=mMUg`>R>!5ZCcnEke>*2pjvoZFjRTno@(SyH*Xx|V!9dNu@$9Xm9l+89&9vUmHS
z6L77f-g#s*)cCF|R{hNqr2+2$!3`~>wK&$I#$&IsS&5!6Q66$(p$EXtO6m>sh>J|(
zx|O@O)8{~&$i6A9Po+qNq7MLPo%Y;zfQDi;=|vtp9_oo3<PD@Pj!dohonS*&<4rV~
zKsvXDMQ8JeW|4FpJ@>HBYwu1fm#Y0O-JYH$O;?GsJ?Tce3vnw1t{Q^9Ae3C`&I4*@
z7w^}#xxA_Na?4x}s^C=CD#|`|B&Z!9Td}^h@7c*dKdEGJUg0s)U0h%pFQTR6`|pH&
zfJmP9%(-?zOSl>fFh#@+oNCycFLv<kO_Od#^NVxqJ&a9<o!Z0~0-uB90+pK&-A<HM
zApby`oR2G1PFcv5!e!cxM^^7AJ%IBY0z5q0!U1HQQwjdKMDMy%T6zXk@*8Y}aNi{X
zvqSf=<7u56uQMgzcSw*TkWJ(3Ae;G2K5;$=mt_2;`}~P3eHhP?9iaVEdR@goES>n!
z*w9Ml)hkNybX@pCgJNq<k&LmfZ8mnh0|2rfDfFE&)8!C#!L|=)@UF;TBB6M(oC5Y_
zsCQJ6)KwvM2dMo3N{n<V@FW{P4T9?N`Fw0tQCLE{s`IicoA{TpbFSX46j@lK16|+9
z>UjHSd~J?hfCX6x9THSL8^#|wAsKaIN|6FG(1l9(UgKX9v{sMgf0%NCXu|%KLe>VG
zF4*RTl|8_Pe7jU%-{p9=5l!Hcz|w{zHl^)Mc*a-{P-p`k<s7AuD9tCLcN6vwI0S7O
zJh}3!FDeyf=*tzmvcx+BT}c~|z?le@{**2C^K04hAF{qSZUXee7k2$fBq3`WYjTk&
zEA=@l6}Ai@^8=sROOXo|uG9g(yjn3rgG5?#DKJR!K$+U1_;8#ErvFP*P}ytbB9(pM
zIc_(`&h32?xR5+e_agBop-Z{5K9q>lwUXiLLPQnfA_KzlA5f-y&}Vo)7;kJa|3pD6
z(bQ*AX0**^xHMJ*nxDS+kMYN-C0Kz|S*>#0LS9H24Rg6l(tqd{WO{G~4m|mx941B%
zo2dF|o3C{>#qb`n1i5z9N@WUn->AHvFAY)LKEy1@nsH05;Budy)p=#)V(ddxWYKcV
z5U(sD6_quW?ZB&Xs&C4FXu)q-%bK>W%qHwKq9}mg7%h8v6hKUk+(DHl+tu%di9b+c
z0U;UT)rYT+)V8(X?%uQNj-4Z3xgJZwiuRR%SP|B-@%9DdMH_lYE>;Pz7qoSY2uW#7
z&^r~a3Ise{ZHT+)Ej$M`s8aIky<90G0lDDHqyPv(bRuP&D1XdqFskTA*@WR4Ztoi4
z^KE(<O8-=8^xHy+CXbt2!^&5v08x1V%jc9D6jRC2<#l6CcLHdRz!P5PmW{YB*G|%(
zEmlTW<aJNXCYg!)4mShKJtd`Y2y$(Ih_eM1M^BxF7)vf2wmDKV;17r1l~4Sgmht5=
zOt%2-3ZTH;#n!%@vM@w7@i)EVI@zlGw-TCKojcWu^J4dm=;2vKX=*DjilH2;{shT-
z+n})&-i|Vn;3){X;2C8~x!l2Bm;Fi3>fMY(;pY31k@|ihX!>2x)|Ltl`zqS|Nx(1=
z9{u<i<#*&+AZut#V<$vB2A!l3)ujBw()w_?UW*;?5A7y*DKyBc6zp7yp#Y?JSoDgn
zPc;7EA7lOlIiG36D?c~D6c%_eEFUFTq!{RMs}G%ub@1GuEgsmGdAG@%m4Tum13=Qv
z4WU%iezkZ0IJBO?|C0ggp=u3XH1v*mQV|2H08nh3Y^ffmAB9gfS6MFiM_Ue1xXQRk
zH-l&xqS_5N9FsOh(k?)r5?AqqJk(h0`F?!<8Ix~Lwa|}^olK50EMXWM4kEN6Quq2q
z5^(PjJ<O!(fnI2x#<w$mEZx}3OH6Iu%~1jyKYEO0fFXT&Z{GLmahCOnICK9cPj6Nd
zJ6YTIm4?TR;!z(rbw@`-pwrIGNg~8fGJJ8pE;bJw59!Ct0t9K}eKBmv4i>OyFJbMy
zX+7VIHTIQ>ULAWlI8X&rrATWxXL?Fqq#rAAWzPd?_*ty=@ZPEPMx<uOz^^G$j%jG&
z1jDr1O!!*+YF)jWD=~8Zx=Z!8ue+-i{d{Ier?}ES;2peT?-QX1Upc7!Wj0BFqol8z
zXW$kLbN_wTv_s{P0v>C&v<Yl&zfoZ<Fiu%5${fQtDlvbZJatJ?%ar;n(z$P#3Uq^>
z)K4ll67++&Ik-qspTv%oj)c4$Nb2?{oM}FqYvTcYeoEsN+!t8!zBed3OwipIJJ|E?
zg)abpJ1Y=ntZ5&kb}Z6y{F`A57+f$1O99on1^0)fo~=r(JQ9%XghIlHW-x&T@NCse
z8QqpiXD0znU^i*KBN-&mK1~ixNF}(z<)}iXdABm61JzKtQ)pgD&WD~bukmC>j1>Zr
zjKjL8uNn|$quvE7_5nD|d)fvB0J_Z`QEHIl#UK2mCZbSdVO-#9=W|S0GcnHtS}c@0
z=nVs@dzDWrBh1e({!k8!wGMbX5K%Ay@s)V5QNyha^0>@5$T5D?#`<_xVGSF`&oZ_Z
zar)0|-vWsj`k)G{1*fVG1d-`BczpuzqN#j>jo)PYaJlEMCID0{ih$R<Y!LUvvp>FW
z(HF*thFEA+H6f|*_@EfPsu4d+EV>fKo|QJ!bMqAl;TxXw@ziM2o#3EjnZ<#&_>vIS
z<ov>?^|t#ls}1b4A3Lp?AfJZK!mrfImn9J{fHDAa-bdpZJpor6)4k&tM<*mQ(fKr-
zEe(e)G^9;h<A~wmYM+c~q`u+fT27QN!~_62+EM1xV<8CKToK5QrPa(g{g}mMwM;b1
zp=HofO@)SyqwPDyna%iQ1z3$rPji&nRbQ`h%20>(-#y8v@9c@VJQA#JM?aOF<+?_6
zx@X>5(#Mw`Egh@&$@-`1S=p4NXDw?4FZ$D3lHn#+#_V~@t%Lpyi=SXuWET9U@|r<e
zJ*8*SF3ehet`HGK#+oGw-SqOWmhyL0vK=vT-h;Y~DeLx!UUBoSFP{|maa~+5Owfy9
zYl?=2FqaYD{|&inO7y&dx+!O**xerUj~_mDg1**NncfNurrbMHyFz;AmamR!Oln~E
z?6ekV-}HMpZGG(5Z%%joxJ>Uq((~u<U5DM9+rDoaesS_%W<S4p{Uh<uC-0^P>Yk3x
z7@@rXe~@`PVdin%*ydcyEi3q;aqxGp!Ta_Gd$01Md&3vZf^2TR#N1*D*<uHI*(&oD
ztClTaL?*8ylap69Cw=w{NBs94+<*N4|3Zh)u0`+*hNy>tefw8!I=Ia{XubbYzkLV&
zeAfGiV%C=Cp(`yd7LgZ`<03bptd;ir|LfvS4?5`YhdF<{VTt+DWk%+QcEUeno4<4X
K_V#*D_WuAET58w;

literal 0
HcmV?d00001


From 3b48806f753b41b8d0540e001d4217cdc054b602 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Tue, 3 Dec 2019 10:11:47 -0500
Subject: [PATCH 203/505] [pplm] README: add setup + tweaks

---
 examples/pplm/README.md | 23 ++++++++++++++++++-----
 1 file changed, 18 insertions(+), 5 deletions(-)

diff --git a/examples/pplm/README.md b/examples/pplm/README.md
index 6eb040a442..103218ae72 100644
--- a/examples/pplm/README.md
+++ b/examples/pplm/README.md
@@ -4,7 +4,7 @@ This folder contains the original code used to run the Plug and Play Language Mo
 ![header image](./imgs/headfigure.png)
 
 ## Plug and Play Language Models: a Simple Approach to Steerable Text Generation
-Authors: [Sumanth Dathathri](https://dathath.github.io/), Andrea Madotto, Janice Lan, Jane Hung, Eric Frank, [Piero Molino](), [Jason Yosinski](http://yosinski.com/), and [Rosanne Liu](http://www.rosanneliu.com/)
+Authors: [Sumanth Dathathri](https://dathath.github.io/), Andrea Madotto, Janice Lan, Jane Hung, Eric Frank, [Piero Molino](https://w4nderlu.st/), [Jason Yosinski](http://yosinski.com/), and [Rosanne Liu](http://www.rosanneliu.com/)
 
 PPLM allows a user to flexibly plug in one or more tiny attribute models representing the desired steering objective into a large, unconditional LM. The method has the key property that it uses the LM _as is_---no training or fine-tuning is required---which enables researchers to leverage best-in-class LMs even if they do not have the extensive hardware required to train them.
 
@@ -14,16 +14,24 @@ Blog link: https://eng.uber.com/pplm
 
 
 ## Setup
-TODO
+
+```bash
+git clone https://github.com/huggingface/transformers && cd transformers
+pip install [--editable] .
+pip install nltk torchtext # additional requirements.
+cd examples/pplm
+```
 
 ## PPLM-BoW 
 
 ### Example command for bag-of-words control
-```
+
+```bash
 python run_pplm.py -B space --cond_text "The president" --length 100 --gamma 1.5 --num_iterations 3 --num_samples 1 --stepsize 0.01 --window_length 5 --kl_scale 0.01 --gm_scale 0.95
 ```
 
 ### Tuning hyperparameters for bag-of-words control
+
 1. Increase `--stepsize` to intensify topic control, and decrease its value to soften the control. `--stepsize 0` recovers the original uncontrolled GPT-2 model. 
 
 2. If the language being generated is repetitive (For e.g. "science science experiment experiment"), there are several options to consider: </br>
@@ -33,16 +41,21 @@ python run_pplm.py -B space --cond_text "The president" --length 100 --gamma 1.5
 
 
 ## PPLM-Discrim
+
 ### Example command for discriminator based sentiment control
-```
+
+```bash
 python run_pplm.py -D sentiment --class_label 3 --cond_text "The lake" --length 10 --gamma 1.0 --num_iterations 10 --num_samples 1 --stepsize 0.03 --kl_scale 0.01 --gm_scale 0.95
 ```
 
 ### Tuning hyperparameters for discriminator control
+
 1. Increase `--stepsize` to intensify topic control, and decrease its value to soften the control. `--stepsize 0` recovers the original uncontrolled GPT-2 model. 
 
 2. Use `--class_label 3` for negative, and `--class_label 2` for positive
 
 ### Example command for detoxificiation:
-python run_pplm.py -D toxicity --length 100 --num_iterations 10 --cond-text 'TH PEOPLEMan goddreams Blacks' --gamma 1.0 --num_samples 10 --stepsize 0.02
 
+```bash
+python run_pplm.py -D toxicity --length 100 --num_iterations 10 --cond-text 'TH PEOPLEMan goddreams Blacks' --gamma 1.0 --num_samples 10 --stepsize 0.02
+```

From 96e83506d1ddee8e19b07118668be73d175decb6 Mon Sep 17 00:00:00 2001
From: Ethan Perez <perez@nyu.edu>
Date: Fri, 29 Nov 2019 18:22:34 -0600
Subject: [PATCH 204/505] Always use SequentialSampler during evaluation

When evaluating, shouldn't we always use the SequentialSampler instead of DistributedSampler? Evaluation only runs on 1 GPU no matter what, so if you use the DistributedSampler with N GPUs, I think you'll only evaluate on 1/N of the evaluation set. That's at least what I'm finding when I run an older/modified version of this repo.
---
 examples/run_squad.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/examples/run_squad.py b/examples/run_squad.py
index 59683c0668..32d807b3ad 100644
--- a/examples/run_squad.py
+++ b/examples/run_squad.py
@@ -216,7 +216,7 @@ def evaluate(args, model, tokenizer, prefix=""):
 
     args.eval_batch_size = args.per_gpu_eval_batch_size * max(1, args.n_gpu)
     # Note that DistributedSampler samples randomly
-    eval_sampler = SequentialSampler(dataset) if args.local_rank == -1 else DistributedSampler(dataset)
+    eval_sampler = SequentialSampler(dataset)
     eval_dataloader = DataLoader(dataset, sampler=eval_sampler, batch_size=args.eval_batch_size)
 
     # multi-gpu evaluate

From f434bfc623de9e535d49f0c095c0b4af3e88e22a Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Tue, 3 Dec 2019 10:53:02 -0500
Subject: [PATCH 205/505] [pplm] Update S3 links

Co-Authored-By: Piero Molino <w4nderlust@gmail.com>
---
 examples/pplm/run_pplm.py | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/examples/pplm/run_pplm.py b/examples/pplm/run_pplm.py
index 5e09427879..f626a43f4f 100644
--- a/examples/pplm/run_pplm.py
+++ b/examples/pplm/run_pplm.py
@@ -59,7 +59,7 @@ BAG_OF_WORDS_ARCHIVE_MAP = {
 
 DISCRIMINATOR_MODELS_PARAMS = {
     "clickbait": {
-        "url": "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/discriminators/clickbait_classifierhead.pt",
+        "url": "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/discriminators/clickbait_classifier_head.pt",
         "class_size": 2,
         "embed_size": 1024,
         "class_vocab": {"non_clickbait": 0, "clickbait": 1},
@@ -67,7 +67,7 @@ DISCRIMINATOR_MODELS_PARAMS = {
         "pretrained_model": "gpt2-medium",
     },
     "sentiment": {
-        "url": "http://s.yosinski.com/SST_classifier_head.pt",
+        "url": "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/discriminators/SST_classifier_head.pt",
         "class_size": 5,
         "embed_size": 1024,
         "class_vocab": {"very_positive": 2, "very_negative": 3},
@@ -75,7 +75,7 @@ DISCRIMINATOR_MODELS_PARAMS = {
         "pretrained_model": "gpt2-medium",
     },
     "toxicity": {
-        "url": "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/discriminators/toxicity_classifierhead.pt",
+        "url": "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/discriminators/toxic_classifier_head.pt",
         "class_size": 2,
         "embed_size": 1024,
         "class_vocab": {"non_toxic": 0, "toxic": 1},

From 48cbf267c988b56c71a2380f748a3e6092ccaed3 Mon Sep 17 00:00:00 2001
From: VictorSanh <victorsanh@gmail.com>
Date: Tue, 3 Dec 2019 11:01:37 -0500
Subject: [PATCH 206/505] Use full dataset for eval (SequentialSampler in
 Distributed setting)

---
 examples/run_glue.py            | 2 +-
 examples/run_lm_finetuning.py   | 2 +-
 examples/run_multiple_choice.py | 2 +-
 examples/run_xnli.py            | 2 +-
 4 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/examples/run_glue.py b/examples/run_glue.py
index 601e9a34c2..369a7110ab 100644
--- a/examples/run_glue.py
+++ b/examples/run_glue.py
@@ -231,7 +231,7 @@ def evaluate(args, model, tokenizer, prefix=""):
 
         args.eval_batch_size = args.per_gpu_eval_batch_size * max(1, args.n_gpu)
         # Note that DistributedSampler samples randomly
-        eval_sampler = SequentialSampler(eval_dataset) if args.local_rank == -1 else DistributedSampler(eval_dataset)
+        eval_sampler = SequentialSampler(eval_dataset)
         eval_dataloader = DataLoader(eval_dataset, sampler=eval_sampler, batch_size=args.eval_batch_size)
 
         # multi-gpu eval
diff --git a/examples/run_lm_finetuning.py b/examples/run_lm_finetuning.py
index 4acea00c55..0bb7460353 100644
--- a/examples/run_lm_finetuning.py
+++ b/examples/run_lm_finetuning.py
@@ -300,7 +300,7 @@ def evaluate(args, model, tokenizer, prefix=""):
 
     args.eval_batch_size = args.per_gpu_eval_batch_size * max(1, args.n_gpu)
     # Note that DistributedSampler samples randomly
-    eval_sampler = SequentialSampler(eval_dataset) if args.local_rank == -1 else DistributedSampler(eval_dataset)
+    eval_sampler = SequentialSampler(eval_dataset)
     eval_dataloader = DataLoader(eval_dataset, sampler=eval_sampler, batch_size=args.eval_batch_size)
 
     # multi-gpu evaluate
diff --git a/examples/run_multiple_choice.py b/examples/run_multiple_choice.py
index 30c3332929..9d1ca7f300 100644
--- a/examples/run_multiple_choice.py
+++ b/examples/run_multiple_choice.py
@@ -226,7 +226,7 @@ def evaluate(args, model, tokenizer, prefix="", test=False):
 
         args.eval_batch_size = args.per_gpu_eval_batch_size * max(1, args.n_gpu)
         # Note that DistributedSampler samples randomly
-        eval_sampler = SequentialSampler(eval_dataset) if args.local_rank == -1 else DistributedSampler(eval_dataset)
+        eval_sampler = SequentialSampler(eval_dataset)
         eval_dataloader = DataLoader(eval_dataset, sampler=eval_sampler, batch_size=args.eval_batch_size)
 
         # multi-gpu evaluate
diff --git a/examples/run_xnli.py b/examples/run_xnli.py
index a3bc0d4604..42d134a43a 100644
--- a/examples/run_xnli.py
+++ b/examples/run_xnli.py
@@ -206,7 +206,7 @@ def evaluate(args, model, tokenizer, prefix=""):
 
         args.eval_batch_size = args.per_gpu_eval_batch_size * max(1, args.n_gpu)
         # Note that DistributedSampler samples randomly
-        eval_sampler = SequentialSampler(eval_dataset) if args.local_rank == -1 else DistributedSampler(eval_dataset)
+        eval_sampler = SequentialSampler(eval_dataset)
         eval_dataloader = DataLoader(eval_dataset, sampler=eval_sampler, batch_size=args.eval_batch_size)
 
         # multi-gpu eval

From 8101924a6812ffb09c54c2af85d2182f9a81db20 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Tue, 3 Dec 2019 11:20:26 -0500
Subject: [PATCH 207/505] Patch: v2.2.1

---
 README.md                | 2 +-
 docs/source/conf.py      | 2 +-
 setup.py                 | 2 +-
 transformers/__init__.py | 2 +-
 4 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/README.md b/README.md
index dd20b80590..3173b6cf10 100644
--- a/README.md
+++ b/README.md
@@ -58,7 +58,7 @@ Choose the right framework for every part of a model's lifetime
 | [Quick tour: Fine-tuning/usage scripts](#quick-tour-of-the-fine-tuningusage-scripts) | Using provided scripts: GLUE, SQuAD and Text generation |
 | [Migrating from pytorch-transformers to transformers](#Migrating-from-pytorch-transformers-to-transformers) | Migrating your code from pytorch-transformers to transformers |
 | [Migrating from pytorch-pretrained-bert to pytorch-transformers](#Migrating-from-pytorch-pretrained-bert-to-transformers) | Migrating your code from pytorch-pretrained-bert to transformers |
-| [Documentation][(v2.2.0)](https://huggingface.co/transformers/v2.2.0) [(v2.1.1)](https://huggingface.co/transformers/v2.1.1) [(v2.0.0)](https://huggingface.co/transformers/v2.0.0) [(v1.2.0)](https://huggingface.co/transformers/v1.2.0) [(v1.1.0)](https://huggingface.co/transformers/v1.1.0) [(v1.0.0)](https://huggingface.co/transformers/v1.0.0) [(master)](https://huggingface.co/transformers) | Full API documentation and more |
+| [Documentation][(v2.2.0/v2.2.1)](https://huggingface.co/transformers/v2.2.0) [(v2.1.1)](https://huggingface.co/transformers/v2.1.1) [(v2.0.0)](https://huggingface.co/transformers/v2.0.0) [(v1.2.0)](https://huggingface.co/transformers/v1.2.0) [(v1.1.0)](https://huggingface.co/transformers/v1.1.0) [(v1.0.0)](https://huggingface.co/transformers/v1.0.0) [(master)](https://huggingface.co/transformers) | Full API documentation and more |
 
 ## Installation
 
diff --git a/docs/source/conf.py b/docs/source/conf.py
index f762a89cd2..2f8505ab3a 100644
--- a/docs/source/conf.py
+++ b/docs/source/conf.py
@@ -26,7 +26,7 @@ author = u'huggingface'
 # The short X.Y version
 version = u''
 # The full version, including alpha/beta/rc tags
-release = u'2.2.0'
+release = u'2.2.1'
 
 
 # -- General configuration ---------------------------------------------------
diff --git a/setup.py b/setup.py
index d8dcf7b898..c07920520d 100644
--- a/setup.py
+++ b/setup.py
@@ -38,7 +38,7 @@ from setuptools import find_packages, setup
 
 setup(
     name="transformers",
-    version="2.2.0",
+    version="2.2.1",
     author="Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond, Google AI Language Team Authors, Open AI team Authors, Facebook AI Authors, Carnegie Mellon University Authors",
     author_email="thomas@huggingface.co",
     description="State-of-the-art Natural Language Processing for TensorFlow 2.0 and PyTorch",
diff --git a/transformers/__init__.py b/transformers/__init__.py
index de25c24b9e..970bdf0cf1 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -1,4 +1,4 @@
-__version__ = "2.2.0"
+__version__ = "2.2.1"
 
 # Work around to update TensorFlow's absl.logging threshold which alters the
 # default Python logging output behavior when present.

From 285b1241e38cdafb6b0dadd1d1afc19493318074 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Tue, 3 Dec 2019 15:00:49 -0500
Subject: [PATCH 208/505] Added SquadResult

---
 transformers/data/processors/squad.py | 71 +++++++++++++++++++++++++++
 1 file changed, 71 insertions(+)

diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index f414d41925..afbe4270f5 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -425,3 +425,74 @@ class SquadFeatures(object):
 
         self.start_position = start_position
         self.end_position = end_position
+
+
+
+class SquadResult(object):
+    """
+    Constructs a SquadResult which can be used to evaluate a model's output on the SQuAD dataset.
+
+    Args:
+        result: The result output by a model on a SQuAD inference. These results may be complex (5 values) as the ones output by
+            XLNet or XLM or may be simple like the other models (2 values). They may be passed as a list or as a dict, with the 
+            following accepted formats:
+
+            `dict` output by a simple model:
+                {
+                    "start_logits": int,
+                    "end_logits": int,
+                    "unique_id": string
+                }
+            `list` output by a simple model:
+                [start_logits, end_logits, unique_id]
+
+            `dict` output by a complex model:
+                {
+                    "start_top_log_probs": float,
+                    "start_top_index": int,
+                    "end_top_log_probs": float,
+                    "end_top_index": int,
+                    "cls_logits": int,
+                    "unique_id": string
+                }
+            `list` output by a complex model:
+                [start_top_log_probs, start_top_index, end_top_log_probs, end_top_index, cls_logits, unique_id]
+
+            See `run_squad.py` for an example.
+    """
+    def __init__(self, result):
+        if isinstance(result, dict):
+            if "start_logits" in result and "end_logits" in result:
+                self.start_logits = result["start_logits"]
+                self.end_logits = result["end_logits"]
+
+            elif "start_top_log_probs" in result and "start_top_index" in result:
+                self.start_top_log_probs  = result["start_top_log_probs"]
+                self.start_top_index      = result["start_top_index"]
+                self.end_top_log_probs    = result["end_top_log_probs"]
+                self.end_top_index        = result["end_top_index"]
+                self.cls_logits           = result["cls_logits"]
+
+            else:
+                raise ValueError("SquadResult instantiated with wrong values.")
+
+            self.unique_id = result["unique_id"]
+        elif isinstance(result, list):            
+            if len(result) == 3:
+                self.start_logits = result[0]
+                self.end_logits = result[1]
+
+            elif len(result) == 6:
+                self.start_top_log_probs  = result[0]
+                self.start_top_index      = result[1]
+                self.end_top_log_probs    = result[2]
+                self.end_top_index        = result[3]
+                self.cls_logits           = result[4]
+
+            else:
+                raise ValueError("SquadResult instantiated with wrong values.")
+
+            self.unique_id = result[-1]
+
+        else:
+            raise ValueError("SquadResult instantiated with wrong values. Should be a dictionary or a list.")

From c835bc85c2f51f4da5eab4f1481a25b052bf6d61 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Tue, 3 Dec 2019 15:28:16 -0500
Subject: [PATCH 209/505] Compute predictions

---
 transformers/data/metrics/squad_metrics.py | 335 +++++++++++++++++++++
 1 file changed, 335 insertions(+)
 create mode 100644 transformers/data/metrics/squad_metrics.py

diff --git a/transformers/data/metrics/squad_metrics.py b/transformers/data/metrics/squad_metrics.py
new file mode 100644
index 0000000000..d4c5a8ec5b
--- /dev/null
+++ b/transformers/data/metrics/squad_metrics.py
@@ -0,0 +1,335 @@
+import json
+import logging
+import math
+import collections
+from io import open
+from tqdm import tqdm
+
+from transformers.tokenization_bert import BasicTokenizer, whitespace_tokenize
+
+logger = logging.getLogger(__name__)
+
+
+def compute_predictions(all_examples, all_features, all_results, n_best_size,
+                      max_answer_length, do_lower_case, output_prediction_file,
+                      output_nbest_file, output_null_log_odds_file, verbose_logging,
+                      version_2_with_negative, null_score_diff_threshold):
+    """Write final predictions to the json file and log-odds of null if needed."""
+    logger.info("Writing predictions to: %s" % (output_prediction_file))
+    logger.info("Writing nbest to: %s" % (output_nbest_file))
+
+    example_index_to_features = collections.defaultdict(list)
+    for feature in all_features:
+        example_index_to_features[feature.example_index].append(feature)
+
+    unique_id_to_result = {}
+    for result in all_results:
+        unique_id_to_result[result.unique_id] = result
+
+    _PrelimPrediction = collections.namedtuple(  # pylint: disable=invalid-name
+        "PrelimPrediction",
+        ["feature_index", "start_index", "end_index", "start_logit", "end_logit"])
+
+    all_predictions = collections.OrderedDict()
+    all_nbest_json = collections.OrderedDict()
+    scores_diff_json = collections.OrderedDict()
+
+    for (example_index, example) in enumerate(all_examples):
+        features = example_index_to_features[example_index]
+
+        prelim_predictions = []
+        # keep track of the minimum score of null start+end of position 0
+        score_null = 1000000  # large and positive
+        min_null_feature_index = 0  # the paragraph slice with min null score
+        null_start_logit = 0  # the start logit at the slice with min null score
+        null_end_logit = 0  # the end logit at the slice with min null score
+        for (feature_index, feature) in enumerate(features):
+            result = unique_id_to_result[feature.unique_id]
+            start_indexes = _get_best_indexes(result.start_logits, n_best_size)
+            end_indexes = _get_best_indexes(result.end_logits, n_best_size)
+            # if we could have irrelevant answers, get the min score of irrelevant
+            if version_2_with_negative:
+                feature_null_score = result.start_logits[0] + result.end_logits[0]
+                if feature_null_score < score_null:
+                    score_null = feature_null_score
+                    min_null_feature_index = feature_index
+                    null_start_logit = result.start_logits[0]
+                    null_end_logit = result.end_logits[0]
+            for start_index in start_indexes:
+                for end_index in end_indexes:
+                    # We could hypothetically create invalid predictions, e.g., predict
+                    # that the start of the span is in the question. We throw out all
+                    # invalid predictions.
+                    if start_index >= len(feature.tokens):
+                        continue
+                    if end_index >= len(feature.tokens):
+                        continue
+                    if start_index not in feature.token_to_orig_map:
+                        continue
+                    if end_index not in feature.token_to_orig_map:
+                        continue
+                    if not feature.token_is_max_context.get(start_index, False):
+                        continue
+                    if end_index < start_index:
+                        continue
+                    length = end_index - start_index + 1
+                    if length > max_answer_length:
+                        continue
+                    prelim_predictions.append(
+                        _PrelimPrediction(
+                            feature_index=feature_index,
+                            start_index=start_index,
+                            end_index=end_index,
+                            start_logit=result.start_logits[start_index],
+                            end_logit=result.end_logits[end_index]))
+        if version_2_with_negative:
+            prelim_predictions.append(
+                _PrelimPrediction(
+                    feature_index=min_null_feature_index,
+                    start_index=0,
+                    end_index=0,
+                    start_logit=null_start_logit,
+                    end_logit=null_end_logit))
+        prelim_predictions = sorted(
+            prelim_predictions,
+            key=lambda x: (x.start_logit + x.end_logit),
+            reverse=True)
+
+        _NbestPrediction = collections.namedtuple(  # pylint: disable=invalid-name
+            "NbestPrediction", ["text", "start_logit", "end_logit"])
+
+        seen_predictions = {}
+        nbest = []
+        for pred in prelim_predictions:
+            if len(nbest) >= n_best_size:
+                break
+            feature = features[pred.feature_index]
+            if pred.start_index > 0:  # this is a non-null prediction
+                tok_tokens = feature.tokens[pred.start_index:(pred.end_index + 1)]
+                orig_doc_start = feature.token_to_orig_map[pred.start_index]
+                orig_doc_end = feature.token_to_orig_map[pred.end_index]
+                orig_tokens = example.doc_tokens[orig_doc_start:(orig_doc_end + 1)]
+                tok_text = " ".join(tok_tokens)
+
+                # De-tokenize WordPieces that have been split off.
+                tok_text = tok_text.replace(" ##", "")
+                tok_text = tok_text.replace("##", "")
+
+                # Clean whitespace
+                tok_text = tok_text.strip()
+                tok_text = " ".join(tok_text.split())
+                orig_text = " ".join(orig_tokens)
+
+                final_text = get_final_text(tok_text, orig_text, do_lower_case, verbose_logging)
+                if final_text in seen_predictions:
+                    continue
+
+                seen_predictions[final_text] = True
+            else:
+                final_text = ""
+                seen_predictions[final_text] = True
+
+            nbest.append(
+                _NbestPrediction(
+                    text=final_text,
+                    start_logit=pred.start_logit,
+                    end_logit=pred.end_logit))
+        # if we didn't include the empty option in the n-best, include it
+        if version_2_with_negative:
+            if "" not in seen_predictions:
+                nbest.append(
+                    _NbestPrediction(
+                        text="",
+                        start_logit=null_start_logit,
+                        end_logit=null_end_logit))
+                
+            # In very rare edge cases we could only have single null prediction.
+            # So we just create a nonce prediction in this case to avoid failure.
+            if len(nbest)==1:
+                nbest.insert(0,
+                    _NbestPrediction(text="empty", start_logit=0.0, end_logit=0.0))
+
+        # In very rare edge cases we could have no valid predictions. So we
+        # just create a nonce prediction in this case to avoid failure.
+        if not nbest:
+            nbest.append(
+                _NbestPrediction(text="empty", start_logit=0.0, end_logit=0.0))
+
+        assert len(nbest) >= 1
+
+        total_scores = []
+        best_non_null_entry = None
+        for entry in nbest:
+            total_scores.append(entry.start_logit + entry.end_logit)
+            if not best_non_null_entry:
+                if entry.text:
+                    best_non_null_entry = entry
+
+        probs = _compute_softmax(total_scores)
+
+        nbest_json = []
+        for (i, entry) in enumerate(nbest):
+            output = collections.OrderedDict()
+            output["text"] = entry.text
+            output["probability"] = probs[i]
+            output["start_logit"] = entry.start_logit
+            output["end_logit"] = entry.end_logit
+            nbest_json.append(output)
+
+        assert len(nbest_json) >= 1
+
+        if not version_2_with_negative:
+            all_predictions[example.qas_id] = nbest_json[0]["text"]
+        else:
+            # predict "" iff the null score - the score of best non-null > threshold
+            score_diff = score_null - best_non_null_entry.start_logit - (
+                best_non_null_entry.end_logit)
+            scores_diff_json[example.qas_id] = score_diff
+            if score_diff > null_score_diff_threshold:
+                all_predictions[example.qas_id] = ""
+            else:
+                all_predictions[example.qas_id] = best_non_null_entry.text
+        all_nbest_json[example.qas_id] = nbest_json
+
+    with open(output_prediction_file, "w") as writer:
+        writer.write(json.dumps(all_predictions, indent=4) + "\n")
+
+    with open(output_nbest_file, "w") as writer:
+        writer.write(json.dumps(all_nbest_json, indent=4) + "\n")
+
+    if version_2_with_negative:
+        with open(output_null_log_odds_file, "w") as writer:
+            writer.write(json.dumps(scores_diff_json, indent=4) + "\n")
+
+    return all_predictions
+
+
+def get_final_text(pred_text, orig_text, do_lower_case, verbose_logging=False):
+    """Project the tokenized prediction back to the original text."""
+
+    # When we created the data, we kept track of the alignment between original
+    # (whitespace tokenized) tokens and our WordPiece tokenized tokens. So
+    # now `orig_text` contains the span of our original text corresponding to the
+    # span that we predicted.
+    #
+    # However, `orig_text` may contain extra characters that we don't want in
+    # our prediction.
+    #
+    # For example, let's say:
+    #   pred_text = steve smith
+    #   orig_text = Steve Smith's
+    #
+    # We don't want to return `orig_text` because it contains the extra "'s".
+    #
+    # We don't want to return `pred_text` because it's already been normalized
+    # (the SQuAD eval script also does punctuation stripping/lower casing but
+    # our tokenizer does additional normalization like stripping accent
+    # characters).
+    #
+    # What we really want to return is "Steve Smith".
+    #
+    # Therefore, we have to apply a semi-complicated alignment heuristic between
+    # `pred_text` and `orig_text` to get a character-to-character alignment. This
+    # can fail in certain cases in which case we just return `orig_text`.
+
+    def _strip_spaces(text):
+        ns_chars = []
+        ns_to_s_map = collections.OrderedDict()
+        for (i, c) in enumerate(text):
+            if c == " ":
+                continue
+            ns_to_s_map[len(ns_chars)] = i
+            ns_chars.append(c)
+        ns_text = "".join(ns_chars)
+        return (ns_text, ns_to_s_map)
+
+    # We first tokenize `orig_text`, strip whitespace from the result
+    # and `pred_text`, and check if they are the same length. If they are
+    # NOT the same length, the heuristic has failed. If they are the same
+    # length, we assume the characters are one-to-one aligned.
+    tokenizer = BasicTokenizer(do_lower_case=do_lower_case)
+
+    tok_text = " ".join(tokenizer.tokenize(orig_text))
+
+    start_position = tok_text.find(pred_text)
+    if start_position == -1:
+        if verbose_logging:
+            logger.info(
+                "Unable to find text: '%s' in '%s'" % (pred_text, orig_text))
+        return orig_text
+    end_position = start_position + len(pred_text) - 1
+
+    (orig_ns_text, orig_ns_to_s_map) = _strip_spaces(orig_text)
+    (tok_ns_text, tok_ns_to_s_map) = _strip_spaces(tok_text)
+
+    if len(orig_ns_text) != len(tok_ns_text):
+        if verbose_logging:
+            logger.info("Length not equal after stripping spaces: '%s' vs '%s'",
+                        orig_ns_text, tok_ns_text)
+        return orig_text
+
+    # We then project the characters in `pred_text` back to `orig_text` using
+    # the character-to-character alignment.
+    tok_s_to_ns_map = {}
+    for (i, tok_index) in tok_ns_to_s_map.items():
+        tok_s_to_ns_map[tok_index] = i
+
+    orig_start_position = None
+    if start_position in tok_s_to_ns_map:
+        ns_start_position = tok_s_to_ns_map[start_position]
+        if ns_start_position in orig_ns_to_s_map:
+            orig_start_position = orig_ns_to_s_map[ns_start_position]
+
+    if orig_start_position is None:
+        if verbose_logging:
+            logger.info("Couldn't map start position")
+        return orig_text
+
+    orig_end_position = None
+    if end_position in tok_s_to_ns_map:
+        ns_end_position = tok_s_to_ns_map[end_position]
+        if ns_end_position in orig_ns_to_s_map:
+            orig_end_position = orig_ns_to_s_map[ns_end_position]
+
+    if orig_end_position is None:
+        if verbose_logging:
+            logger.info("Couldn't map end position")
+        return orig_text
+
+    output_text = orig_text[orig_start_position:(orig_end_position + 1)]
+    return output_text
+
+
+def _get_best_indexes(logits, n_best_size):
+    """Get the n-best logits from a list."""
+    index_and_score = sorted(enumerate(logits), key=lambda x: x[1], reverse=True)
+
+    best_indexes = []
+    for i in range(len(index_and_score)):
+        if i >= n_best_size:
+            break
+        best_indexes.append(index_and_score[i][0])
+    return best_indexes
+
+
+def _compute_softmax(scores):
+    """Compute softmax probability over raw logits."""
+    if not scores:
+        return []
+
+    max_score = None
+    for score in scores:
+        if max_score is None or score > max_score:
+            max_score = score
+
+    exp_scores = []
+    total_sum = 0.0
+    for score in scores:
+        x = math.exp(score - max_score)
+        exp_scores.append(x)
+        total_sum += x
+
+    probs = []
+    for score in exp_scores:
+        probs.append(score / total_sum)
+    return probs

From 7edb51f3a516ca533797fb2bb2f2b7ce86e0df70 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Tue, 3 Dec 2019 22:07:25 +0000
Subject: [PATCH 210/505] [pplm] split classif head into its own file

---
 examples/pplm/pplm_classification_head.py | 18 ++++++++++++++++++
 examples/pplm/run_pplm.py                 |  2 +-
 examples/pplm/run_pplm_discrim_train.py   | 17 +----------------
 3 files changed, 20 insertions(+), 17 deletions(-)
 create mode 100644 examples/pplm/pplm_classification_head.py

diff --git a/examples/pplm/pplm_classification_head.py b/examples/pplm/pplm_classification_head.py
new file mode 100644
index 0000000000..9aae0f17e9
--- /dev/null
+++ b/examples/pplm/pplm_classification_head.py
@@ -0,0 +1,18 @@
+import torch
+
+class ClassificationHead(torch.nn.Module):
+    """Classification Head for  transformer encoders"""
+
+    def __init__(self, class_size, embed_size):
+        super(ClassificationHead, self).__init__()
+        self.class_size = class_size
+        self.embed_size = embed_size
+        # self.mlp1 = torch.nn.Linear(embed_size, embed_size)
+        # self.mlp2 = (torch.nn.Linear(embed_size, class_size))
+        self.mlp = torch.nn.Linear(embed_size, class_size)
+
+    def forward(self, hidden_state):
+        # hidden_state = F.relu(self.mlp1(hidden_state))
+        # hidden_state = self.mlp2(hidden_state)
+        logits = self.mlp(hidden_state)
+        return logits
diff --git a/examples/pplm/run_pplm.py b/examples/pplm/run_pplm.py
index f626a43f4f..dda5d85ae7 100644
--- a/examples/pplm/run_pplm.py
+++ b/examples/pplm/run_pplm.py
@@ -33,10 +33,10 @@ import torch.nn.functional as F
 from torch.autograd import Variable
 from tqdm import trange
 
-from examples.run_pplm_discrim_train import ClassificationHead
 from transformers import GPT2Tokenizer
 from transformers.file_utils import cached_path
 from transformers.modeling_gpt2 import GPT2LMHeadModel
+from pplm_classification_head import ClassificationHead
 
 PPLM_BOW = 1
 PPLM_DISCRIM = 2
diff --git a/examples/pplm/run_pplm_discrim_train.py b/examples/pplm/run_pplm_discrim_train.py
index db081e1a17..9d36b79bc4 100644
--- a/examples/pplm/run_pplm_discrim_train.py
+++ b/examples/pplm/run_pplm_discrim_train.py
@@ -21,6 +21,7 @@ from torchtext import datasets
 from tqdm import tqdm, trange
 
 from transformers import GPT2Tokenizer, GPT2LMHeadModel
+from pplm_classification_head import ClassificationHead
 
 torch.manual_seed(0)
 np.random.seed(0)
@@ -29,22 +30,6 @@ example_sentence = "This is incredible! I love it, this is the best chicken I ha
 max_length_seq = 100
 
 
-class ClassificationHead(torch.nn.Module):
-    """Classification Head for  transformer encoders"""
-
-    def __init__(self, class_size, embed_size):
-        super(ClassificationHead, self).__init__()
-        self.class_size = class_size
-        self.embed_size = embed_size
-        # self.mlp1 = torch.nn.Linear(embed_size, embed_size)
-        # self.mlp2 = (torch.nn.Linear(embed_size, class_size))
-        self.mlp = torch.nn.Linear(embed_size, class_size)
-
-    def forward(self, hidden_state):
-        # hidden_state = F.relu(self.mlp1(hidden_state))
-        # hidden_state = self.mlp2(hidden_state)
-        logits = self.mlp(hidden_state)
-        return logits
 
 
 class Discriminator(torch.nn.Module):

From de276de1c1a469a58a25383a35a239d02459a978 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Tue, 3 Dec 2019 17:15:51 -0500
Subject: [PATCH 211/505] Working evaluation

---
 examples/run_squad.py                      |  43 +-
 transformers/data/metrics/squad_metrics.py | 588 +++++++++++++++++----
 transformers/data/processors/squad.py      |  19 +-
 3 files changed, 507 insertions(+), 143 deletions(-)

diff --git a/examples/run_squad.py b/examples/run_squad.py
index 545c3ad55a..b7952487dc 100644
--- a/examples/run_squad.py
+++ b/examples/run_squad.py
@@ -16,7 +16,8 @@
 """ Finetuning the library models for question-answering on SQuAD (DistilBERT, Bert, XLM, XLNet)."""
 
 from __future__ import absolute_import, division, print_function
-from transformers.data.processors.squad import SquadV1Processor, SquadV2Processor
+from transformers.data.processors.squad import SquadV1Processor, SquadV2Processor, SquadResult
+from transformers.data.metrics.squad_metrics import compute_predictions, compute_predictions_extended, squad_evaluate
 
 import argparse
 import logging
@@ -230,9 +231,11 @@ def evaluate(args, model, tokenizer, prefix=""):
         model.eval()
         batch = tuple(t.to(args.device) for t in batch)
         with torch.no_grad():
-            inputs = {'input_ids':      batch[0],
-                      'attention_mask': batch[1]
-                      }
+            inputs = {
+                'input_ids':      batch[0],
+                'attention_mask': batch[1]
+            }
+            
             if args.model_type != 'distilbert':
                 inputs['token_type_ids'] = None if args.model_type == 'xlm' else batch[2]  # XLM don't use segment_ids
             example_indices = batch[3]
@@ -244,18 +247,8 @@ def evaluate(args, model, tokenizer, prefix=""):
         for i, example_index in enumerate(example_indices):
             eval_feature = features[example_index.item()]
             unique_id = int(eval_feature.unique_id)
-            if args.model_type in ['xlnet', 'xlm']:
-                # XLNet uses a more complex post-processing procedure
-                result = RawResultExtended(unique_id            = unique_id,
-                                           start_top_log_probs  = to_list(outputs[0][i]),
-                                           start_top_index      = to_list(outputs[1][i]),
-                                           end_top_log_probs    = to_list(outputs[2][i]),
-                                           end_top_index        = to_list(outputs[3][i]),
-                                           cls_logits           = to_list(outputs[4][i]))
-            else:
-                result = RawResult(unique_id    = unique_id,
-                                   start_logits = to_list(outputs[0][i]),
-                                   end_logits   = to_list(outputs[1][i]))
+
+            result = SquadResult([to_list(output[i]) for output in outputs] + [unique_id])
             all_results.append(result)
 
     evalTime = timeit.default_timer() - start_time
@@ -271,22 +264,18 @@ def evaluate(args, model, tokenizer, prefix=""):
 
     if args.model_type in ['xlnet', 'xlm']:
         # XLNet uses a more complex post-processing procedure
-        write_predictions_extended(examples, features, all_results, args.n_best_size,
+        predictions = compute_predictions_extended(examples, features, all_results, args.n_best_size,
                         args.max_answer_length, output_prediction_file,
                         output_nbest_file, output_null_log_odds_file, args.predict_file,
                         model.config.start_n_top, model.config.end_n_top,
                         args.version_2_with_negative, tokenizer, args.verbose_logging)
     else:
-        write_predictions(examples, features, all_results, args.n_best_size,
+        predictions = compute_predictions(examples, features, all_results, args.n_best_size,
                         args.max_answer_length, args.do_lower_case, output_prediction_file,
                         output_nbest_file, output_null_log_odds_file, args.verbose_logging,
                         args.version_2_with_negative, args.null_score_diff_threshold)
 
-    # Evaluate with the official SQuAD script
-    evaluate_options = EVAL_OPTS(data_file=args.predict_file,
-                                 pred_file=output_prediction_file,
-                                 na_prob_file=output_null_log_odds_file)
-    results = evaluate_on_squad(evaluate_options)
+    results = squad_evaluate(examples, predictions)
     return results
 
 def load_and_cache_examples(args, tokenizer, evaluate=False, output_examples=False):
@@ -306,8 +295,12 @@ def load_and_cache_examples(args, tokenizer, evaluate=False, output_examples=Fal
         logger.info("Creating features from dataset file at %s", input_file)
 
         processor = SquadV2Processor()
-        examples = processor.get_dev_examples("examples/squad") if evaluate else processor.get_train_examples("examples/squad")
-        features = squad_convert_examples_to_features(
+        examples = processor.get_dev_examples("examples/squad", only_first=100) if evaluate else processor.get_train_examples("examples/squad")
+        # import tensorflow_datasets as tfds
+        # tfds_examples = tfds.load("squad")
+        # examples = SquadV1Processor().get_examples_from_dataset(tfds_examples["validation"])
+
+        features = squad_convert_examples_to_features( 
             examples=examples,
             tokenizer=tokenizer,
             max_seq_length=args.max_seq_length,
diff --git a/transformers/data/metrics/squad_metrics.py b/transformers/data/metrics/squad_metrics.py
index d4c5a8ec5b..83647a20d0 100644
--- a/transformers/data/metrics/squad_metrics.py
+++ b/transformers/data/metrics/squad_metrics.py
@@ -1,15 +1,323 @@
+""" Very heavily inspired by the official evaluation script for SQuAD version 2.0 which was
+modified by XLNet authors to update `find_best_threshold` scripts for SQuAD V2.0
+
+In addition to basic functionality, we also compute additional statistics and
+plot precision-recall curves if an additional na_prob.json file is provided.
+This file is expected to map question ID's to the model's predicted probability
+that a question is unanswerable.
+"""
+
+
 import json
 import logging
 import math
 import collections
 from io import open
 from tqdm import tqdm
+import string
+import re
 
 from transformers.tokenization_bert import BasicTokenizer, whitespace_tokenize
 
 logger = logging.getLogger(__name__)
 
 
+def normalize_answer(s):
+    """Lower text and remove punctuation, articles and extra whitespace."""
+    def remove_articles(text):
+        regex = re.compile(r'\b(a|an|the)\b', re.UNICODE)
+        return re.sub(regex, ' ', text)
+
+    def white_space_fix(text):
+        return ' '.join(text.split())
+
+    def remove_punc(text):
+        exclude = set(string.punctuation)
+        return ''.join(ch for ch in text if ch not in exclude)
+
+    def lower(text):
+        return text.lower()
+    return white_space_fix(remove_articles(remove_punc(lower(s))))
+
+
+def get_tokens(s):
+    if not s:
+        return []
+    return normalize_answer(s).split()
+
+
+def compute_exact(a_gold, a_pred):
+    return int(normalize_answer(a_gold) == normalize_answer(a_pred))
+
+
+def compute_f1(a_gold, a_pred):
+    gold_toks = get_tokens(a_gold)
+    pred_toks = get_tokens(a_pred)
+    common = collections.Counter(gold_toks) & collections.Counter(pred_toks)
+    num_same = sum(common.values())
+    if len(gold_toks) == 0 or len(pred_toks) == 0:
+        # If either is no-answer, then F1 is 1 if they agree, 0 otherwise
+        return int(gold_toks == pred_toks)
+    if num_same == 0:
+        return 0
+    precision = 1.0 * num_same / len(pred_toks)
+    recall = 1.0 * num_same / len(gold_toks)
+    f1 = (2 * precision * recall) / (precision + recall)
+    return f1
+
+
+def get_raw_scores(examples, preds):
+    """
+    Computes the exact and f1 scores from the examples and the model predictions
+    """
+    exact_scores = {}
+    f1_scores = {}
+
+    for example in examples:
+        qas_id = example.qas_id
+        gold_answers = [answer['text'] for answer in example.answers if normalize_answer(answer['text'])]
+
+        if not gold_answers:
+            # For unanswerable questions, only correct answer is empty string
+            gold_answers = ['']
+
+        if qas_id not in preds:
+            print('Missing prediction for %s' % qas_id)
+            continue
+
+        prediction = preds[qas_id]
+        exact_scores[qas_id] = max(compute_exact(a, prediction) for a in gold_answers)
+        f1_scores[qas_id] = max(compute_f1(a, prediction) for a in gold_answers)
+
+    return exact_scores, f1_scores
+
+
+def apply_no_ans_threshold(scores, na_probs, qid_to_has_ans, na_prob_thresh):
+    new_scores = {}
+    for qid, s in scores.items():
+        pred_na = na_probs[qid] > na_prob_thresh
+        if pred_na:
+            new_scores[qid] = float(not qid_to_has_ans[qid])
+        else:
+            new_scores[qid] = s
+    return new_scores
+
+
+def make_eval_dict(exact_scores, f1_scores, qid_list=None):
+    if not qid_list:
+        total = len(exact_scores)
+        return collections.OrderedDict([
+            ('exact', 100.0 * sum(exact_scores.values()) / total),
+            ('f1', 100.0 * sum(f1_scores.values()) / total),
+            ('total', total),
+        ])
+    else:
+        total = len(qid_list)
+        return collections.OrderedDict([
+            ('exact', 100.0 * sum(exact_scores[k] for k in qid_list) / total),
+            ('f1', 100.0 * sum(f1_scores[k] for k in qid_list) / total),
+            ('total', total),
+        ])
+
+
+def merge_eval(main_eval, new_eval, prefix):
+    for k in new_eval:
+        main_eval['%s_%s' % (prefix, k)] = new_eval[k]
+
+
+def find_best_thresh(preds, scores, na_probs, qid_to_has_ans):
+    num_no_ans = sum(1 for k in qid_to_has_ans if not qid_to_has_ans[k])
+    cur_score = num_no_ans
+    best_score = cur_score
+    best_thresh = 0.0
+    qid_list = sorted(na_probs, key=lambda k: na_probs[k])
+    for _, qid in enumerate(qid_list):
+        if qid not in scores:
+            continue
+        if qid_to_has_ans[qid]:
+            diff = scores[qid]
+        else:
+            if preds[qid]:
+                diff = -1
+            else:
+                diff = 0
+        cur_score += diff
+        if cur_score > best_score:
+            best_score = cur_score
+            best_thresh = na_probs[qid]
+    return 100.0 * best_score / len(scores), best_thresh
+
+
+def find_all_best_thresh(main_eval, preds, exact_raw, f1_raw, na_probs, qid_to_has_ans):
+    best_exact, exact_thresh = find_best_thresh(preds, exact_raw, na_probs, qid_to_has_ans)
+    best_f1, f1_thresh = find_best_thresh(preds, f1_raw, na_probs, qid_to_has_ans)
+
+    main_eval['best_exact'] = best_exact
+    main_eval['best_exact_thresh'] = exact_thresh
+    main_eval['best_f1'] = best_f1
+    main_eval['best_f1_thresh'] = f1_thresh
+
+
+def squad_evaluate(examples, preds, no_answer_probs=None, no_answer_probability_threshold=1.0):
+    qas_id_to_has_answer = {example.qas_id: bool(example.answers) for example in examples}
+    has_answer_qids = [qas_id for qas_id, has_answer in qas_id_to_has_answer.items() if has_answer]
+    no_answer_qids = [qas_id for qas_id, has_answer in qas_id_to_has_answer.items() if not has_answer]
+
+    if no_answer_probs is None:
+        no_answer_probs = {k: 0.0 for k in preds}
+
+    exact, f1 = get_raw_scores(examples, preds)
+
+    exact_threshold = apply_no_ans_threshold(exact, no_answer_probs, qas_id_to_has_answer, no_answer_probability_threshold)
+    f1_threshold = apply_no_ans_threshold(f1, no_answer_probs, qas_id_to_has_answer, no_answer_probability_threshold)
+
+    evaluation = make_eval_dict(exact_threshold, f1_threshold)
+
+    if has_answer_qids:
+        has_ans_eval = make_eval_dict(exact_threshold, f1_threshold, qid_list=has_answer_qids)
+        merge_eval(evaluation, has_ans_eval, 'HasAns')
+
+    if no_answer_qids:
+        no_ans_eval = make_eval_dict(exact_threshold, f1_threshold, qid_list=no_answer_qids)
+        merge_eval(evaluation, no_ans_eval, 'NoAns')
+
+    if no_answer_probs:
+        find_all_best_thresh(evaluation, preds, exact, f1, no_answer_probs, qas_id_to_has_answer)
+
+    return evaluation
+
+
+def get_final_text(pred_text, orig_text, do_lower_case, verbose_logging=False):
+    """Project the tokenized prediction back to the original text."""
+
+    # When we created the data, we kept track of the alignment between original
+    # (whitespace tokenized) tokens and our WordPiece tokenized tokens. So
+    # now `orig_text` contains the span of our original text corresponding to the
+    # span that we predicted.
+    #
+    # However, `orig_text` may contain extra characters that we don't want in
+    # our prediction.
+    #
+    # For example, let's say:
+    #   pred_text = steve smith
+    #   orig_text = Steve Smith's
+    #
+    # We don't want to return `orig_text` because it contains the extra "'s".
+    #
+    # We don't want to return `pred_text` because it's already been normalized
+    # (the SQuAD eval script also does punctuation stripping/lower casing but
+    # our tokenizer does additional normalization like stripping accent
+    # characters).
+    #
+    # What we really want to return is "Steve Smith".
+    #
+    # Therefore, we have to apply a semi-complicated alignment heuristic between
+    # `pred_text` and `orig_text` to get a character-to-character alignment. This
+    # can fail in certain cases in which case we just return `orig_text`.
+
+    def _strip_spaces(text):
+        ns_chars = []
+        ns_to_s_map = collections.OrderedDict()
+        for (i, c) in enumerate(text):
+            if c == " ":
+                continue
+            ns_to_s_map[len(ns_chars)] = i
+            ns_chars.append(c)
+        ns_text = "".join(ns_chars)
+        return (ns_text, ns_to_s_map)
+
+    # We first tokenize `orig_text`, strip whitespace from the result
+    # and `pred_text`, and check if they are the same length. If they are
+    # NOT the same length, the heuristic has failed. If they are the same
+    # length, we assume the characters are one-to-one aligned.
+    tokenizer = BasicTokenizer(do_lower_case=do_lower_case)
+
+    tok_text = " ".join(tokenizer.tokenize(orig_text))
+
+    start_position = tok_text.find(pred_text)
+    if start_position == -1:
+        if verbose_logging:
+            logger.info(
+                "Unable to find text: '%s' in '%s'" % (pred_text, orig_text))
+        return orig_text
+    end_position = start_position + len(pred_text) - 1
+
+    (orig_ns_text, orig_ns_to_s_map) = _strip_spaces(orig_text)
+    (tok_ns_text, tok_ns_to_s_map) = _strip_spaces(tok_text)
+
+    if len(orig_ns_text) != len(tok_ns_text):
+        if verbose_logging:
+            logger.info("Length not equal after stripping spaces: '%s' vs '%s'",
+                        orig_ns_text, tok_ns_text)
+        return orig_text
+
+    # We then project the characters in `pred_text` back to `orig_text` using
+    # the character-to-character alignment.
+    tok_s_to_ns_map = {}
+    for (i, tok_index) in tok_ns_to_s_map.items():
+        tok_s_to_ns_map[tok_index] = i
+
+    orig_start_position = None
+    if start_position in tok_s_to_ns_map:
+        ns_start_position = tok_s_to_ns_map[start_position]
+        if ns_start_position in orig_ns_to_s_map:
+            orig_start_position = orig_ns_to_s_map[ns_start_position]
+
+    if orig_start_position is None:
+        if verbose_logging:
+            logger.info("Couldn't map start position")
+        return orig_text
+
+    orig_end_position = None
+    if end_position in tok_s_to_ns_map:
+        ns_end_position = tok_s_to_ns_map[end_position]
+        if ns_end_position in orig_ns_to_s_map:
+            orig_end_position = orig_ns_to_s_map[ns_end_position]
+
+    if orig_end_position is None:
+        if verbose_logging:
+            logger.info("Couldn't map end position")
+        return orig_text
+
+    output_text = orig_text[orig_start_position:(orig_end_position + 1)]
+    return output_text
+
+
+def _get_best_indexes(logits, n_best_size):
+    """Get the n-best logits from a list."""
+    index_and_score = sorted(enumerate(logits), key=lambda x: x[1], reverse=True)
+
+    best_indexes = []
+    for i in range(len(index_and_score)):
+        if i >= n_best_size:
+            break
+        best_indexes.append(index_and_score[i][0])
+    return best_indexes
+
+
+def _compute_softmax(scores):
+    """Compute softmax probability over raw logits."""
+    if not scores:
+        return []
+
+    max_score = None
+    for score in scores:
+        if max_score is None or score > max_score:
+            max_score = score
+
+    exp_scores = []
+    total_sum = 0.0
+    for score in scores:
+        x = math.exp(score - max_score)
+        exp_scores.append(x)
+        total_sum += x
+
+    probs = []
+    for score in exp_scores:
+        probs.append(score / total_sum)
+    return probs
+
+
 def compute_predictions(all_examples, all_features, all_results, n_best_size,
                       max_answer_length, do_lower_case, output_prediction_file,
                       output_nbest_file, output_null_log_odds_file, verbose_logging,
@@ -204,132 +512,192 @@ def compute_predictions(all_examples, all_features, all_results, n_best_size,
     return all_predictions
 
 
-def get_final_text(pred_text, orig_text, do_lower_case, verbose_logging=False):
-    """Project the tokenized prediction back to the original text."""
+def compute_predictions_extended(all_examples, all_features, all_results, n_best_size,
+                                max_answer_length, output_prediction_file,
+                                output_nbest_file,
+                                output_null_log_odds_file, orig_data_file,
+                                start_n_top, end_n_top, version_2_with_negative,
+                                tokenizer, verbose_logging):
+    """ XLNet write prediction logic (more complex than Bert's).
+        Write final predictions to the json file and log-odds of null if needed.
 
-    # When we created the data, we kept track of the alignment between original
-    # (whitespace tokenized) tokens and our WordPiece tokenized tokens. So
-    # now `orig_text` contains the span of our original text corresponding to the
-    # span that we predicted.
-    #
-    # However, `orig_text` may contain extra characters that we don't want in
-    # our prediction.
-    #
-    # For example, let's say:
-    #   pred_text = steve smith
-    #   orig_text = Steve Smith's
-    #
-    # We don't want to return `orig_text` because it contains the extra "'s".
-    #
-    # We don't want to return `pred_text` because it's already been normalized
-    # (the SQuAD eval script also does punctuation stripping/lower casing but
-    # our tokenizer does additional normalization like stripping accent
-    # characters).
-    #
-    # What we really want to return is "Steve Smith".
-    #
-    # Therefore, we have to apply a semi-complicated alignment heuristic between
-    # `pred_text` and `orig_text` to get a character-to-character alignment. This
-    # can fail in certain cases in which case we just return `orig_text`.
+        Requires utils_squad_evaluate.py
+    """
+    _PrelimPrediction = collections.namedtuple(  # pylint: disable=invalid-name
+        "PrelimPrediction",
+        ["feature_index", "start_index", "end_index",
+        "start_log_prob", "end_log_prob"])
 
-    def _strip_spaces(text):
-        ns_chars = []
-        ns_to_s_map = collections.OrderedDict()
-        for (i, c) in enumerate(text):
-            if c == " ":
+    _NbestPrediction = collections.namedtuple(  # pylint: disable=invalid-name
+        "NbestPrediction", ["text", "start_log_prob", "end_log_prob"])
+
+    logger.info("Writing predictions to: %s", output_prediction_file)
+    # logger.info("Writing nbest to: %s" % (output_nbest_file))
+
+    example_index_to_features = collections.defaultdict(list)
+    for feature in all_features:
+        example_index_to_features[feature.example_index].append(feature)
+
+    unique_id_to_result = {}
+    for result in all_results:
+        unique_id_to_result[result.unique_id] = result
+
+    all_predictions = collections.OrderedDict()
+    all_nbest_json = collections.OrderedDict()
+    scores_diff_json = collections.OrderedDict()
+
+    for (example_index, example) in enumerate(all_examples):
+        features = example_index_to_features[example_index]
+
+        prelim_predictions = []
+        # keep track of the minimum score of null start+end of position 0
+        score_null = 1000000  # large and positive
+
+        for (feature_index, feature) in enumerate(features):
+            result = unique_id_to_result[feature.unique_id]
+
+            cur_null_score = result.cls_logits
+
+            # if we could have irrelevant answers, get the min score of irrelevant
+            score_null = min(score_null, cur_null_score)
+
+            for i in range(start_n_top):
+                for j in range(end_n_top):
+                    start_log_prob = result.start_top_log_probs[i]
+                    start_index = result.start_top_index[i]
+
+                    j_index = i * end_n_top + j
+
+                    end_log_prob = result.end_top_log_probs[j_index]
+                    end_index = result.end_top_index[j_index]
+
+                    # We could hypothetically create invalid predictions, e.g., predict
+                    # that the start of the span is in the question. We throw out all
+                    # invalid predictions.
+                    if start_index >= feature.paragraph_len - 1:
+                        continue
+                    if end_index >= feature.paragraph_len - 1:
+                        continue
+
+                    if not feature.token_is_max_context.get(start_index, False):
+                        continue
+                    if end_index < start_index:
+                        continue
+                    length = end_index - start_index + 1
+                    if length > max_answer_length:
+                        continue
+
+                    prelim_predictions.append(
+                        _PrelimPrediction(
+                            feature_index=feature_index,
+                            start_index=start_index,
+                            end_index=end_index,
+                            start_log_prob=start_log_prob,
+                            end_log_prob=end_log_prob))
+
+        prelim_predictions = sorted(
+            prelim_predictions,
+            key=lambda x: (x.start_log_prob + x.end_log_prob),
+            reverse=True)
+
+        seen_predictions = {}
+        nbest = []
+        for pred in prelim_predictions:
+            if len(nbest) >= n_best_size:
+                break
+            feature = features[pred.feature_index]
+
+            # XLNet un-tokenizer
+            # Let's keep it simple for now and see if we need all this later.
+            # 
+            # tok_start_to_orig_index = feature.tok_start_to_orig_index
+            # tok_end_to_orig_index = feature.tok_end_to_orig_index
+            # start_orig_pos = tok_start_to_orig_index[pred.start_index]
+            # end_orig_pos = tok_end_to_orig_index[pred.end_index]
+            # paragraph_text = example.paragraph_text
+            # final_text = paragraph_text[start_orig_pos: end_orig_pos + 1].strip()
+
+            # Previously used Bert untokenizer
+            tok_tokens = feature.tokens[pred.start_index:(pred.end_index + 1)]
+            orig_doc_start = feature.token_to_orig_map[pred.start_index]
+            orig_doc_end = feature.token_to_orig_map[pred.end_index]
+            orig_tokens = example.doc_tokens[orig_doc_start:(orig_doc_end + 1)]
+            tok_text = tokenizer.convert_tokens_to_string(tok_tokens)
+
+            # Clean whitespace
+            tok_text = tok_text.strip()
+            tok_text = " ".join(tok_text.split())
+            orig_text = " ".join(orig_tokens)
+
+            final_text = get_final_text(tok_text, orig_text, tokenizer.do_lower_case,
+                                        verbose_logging)
+
+            if final_text in seen_predictions:
                 continue
-            ns_to_s_map[len(ns_chars)] = i
-            ns_chars.append(c)
-        ns_text = "".join(ns_chars)
-        return (ns_text, ns_to_s_map)
 
-    # We first tokenize `orig_text`, strip whitespace from the result
-    # and `pred_text`, and check if they are the same length. If they are
-    # NOT the same length, the heuristic has failed. If they are the same
-    # length, we assume the characters are one-to-one aligned.
-    tokenizer = BasicTokenizer(do_lower_case=do_lower_case)
+            seen_predictions[final_text] = True
 
-    tok_text = " ".join(tokenizer.tokenize(orig_text))
+            nbest.append(
+                _NbestPrediction(
+                    text=final_text,
+                    start_log_prob=pred.start_log_prob,
+                    end_log_prob=pred.end_log_prob))
 
-    start_position = tok_text.find(pred_text)
-    if start_position == -1:
-        if verbose_logging:
-            logger.info(
-                "Unable to find text: '%s' in '%s'" % (pred_text, orig_text))
-        return orig_text
-    end_position = start_position + len(pred_text) - 1
+        # In very rare edge cases we could have no valid predictions. So we
+        # just create a nonce prediction in this case to avoid failure.
+        if not nbest:
+            nbest.append(
+                _NbestPrediction(text="", start_log_prob=-1e6,
+                end_log_prob=-1e6))
 
-    (orig_ns_text, orig_ns_to_s_map) = _strip_spaces(orig_text)
-    (tok_ns_text, tok_ns_to_s_map) = _strip_spaces(tok_text)
+        total_scores = []
+        best_non_null_entry = None
+        for entry in nbest:
+            total_scores.append(entry.start_log_prob + entry.end_log_prob)
+            if not best_non_null_entry:
+                best_non_null_entry = entry
 
-    if len(orig_ns_text) != len(tok_ns_text):
-        if verbose_logging:
-            logger.info("Length not equal after stripping spaces: '%s' vs '%s'",
-                        orig_ns_text, tok_ns_text)
-        return orig_text
+        probs = _compute_softmax(total_scores)
 
-    # We then project the characters in `pred_text` back to `orig_text` using
-    # the character-to-character alignment.
-    tok_s_to_ns_map = {}
-    for (i, tok_index) in tok_ns_to_s_map.items():
-        tok_s_to_ns_map[tok_index] = i
+        nbest_json = []
+        for (i, entry) in enumerate(nbest):
+            output = collections.OrderedDict()
+            output["text"] = entry.text
+            output["probability"] = probs[i]
+            output["start_log_prob"] = entry.start_log_prob
+            output["end_log_prob"] = entry.end_log_prob
+            nbest_json.append(output)
 
-    orig_start_position = None
-    if start_position in tok_s_to_ns_map:
-        ns_start_position = tok_s_to_ns_map[start_position]
-        if ns_start_position in orig_ns_to_s_map:
-            orig_start_position = orig_ns_to_s_map[ns_start_position]
+        assert len(nbest_json) >= 1
+        assert best_non_null_entry is not None
 
-    if orig_start_position is None:
-        if verbose_logging:
-            logger.info("Couldn't map start position")
-        return orig_text
+        score_diff = score_null
+        scores_diff_json[example.qas_id] = score_diff
+        # note(zhiliny): always predict best_non_null_entry
+        # and the evaluation script will search for the best threshold
+        all_predictions[example.qas_id] = best_non_null_entry.text
 
-    orig_end_position = None
-    if end_position in tok_s_to_ns_map:
-        ns_end_position = tok_s_to_ns_map[end_position]
-        if ns_end_position in orig_ns_to_s_map:
-            orig_end_position = orig_ns_to_s_map[ns_end_position]
+        all_nbest_json[example.qas_id] = nbest_json
 
-    if orig_end_position is None:
-        if verbose_logging:
-            logger.info("Couldn't map end position")
-        return orig_text
+    with open(output_prediction_file, "w") as writer:
+        writer.write(json.dumps(all_predictions, indent=4) + "\n")
 
-    output_text = orig_text[orig_start_position:(orig_end_position + 1)]
-    return output_text
+    with open(output_nbest_file, "w") as writer:
+        writer.write(json.dumps(all_nbest_json, indent=4) + "\n")
 
+    if version_2_with_negative:
+        with open(output_null_log_odds_file, "w") as writer:
+            writer.write(json.dumps(scores_diff_json, indent=4) + "\n")
 
-def _get_best_indexes(logits, n_best_size):
-    """Get the n-best logits from a list."""
-    index_and_score = sorted(enumerate(logits), key=lambda x: x[1], reverse=True)
+    with open(orig_data_file, "r", encoding='utf-8') as reader:
+        orig_data = json.load(reader)["data"]
 
-    best_indexes = []
-    for i in range(len(index_and_score)):
-        if i >= n_best_size:
-            break
-        best_indexes.append(index_and_score[i][0])
-    return best_indexes
+    qid_to_has_ans = make_qid_to_has_ans(orig_data)
+    has_ans_qids = [k for k, v in qid_to_has_ans.items() if v]
+    no_ans_qids = [k for k, v in qid_to_has_ans.items() if not v]
+    exact_raw, f1_raw = get_raw_scores(orig_data, all_predictions)
+    out_eval = {}
 
+    find_all_best_thresh_v2(out_eval, all_predictions, exact_raw, f1_raw, scores_diff_json, qid_to_has_ans)
 
-def _compute_softmax(scores):
-    """Compute softmax probability over raw logits."""
-    if not scores:
-        return []
-
-    max_score = None
-    for score in scores:
-        if max_score is None or score > max_score:
-            max_score = score
-
-    exp_scores = []
-    total_sum = 0.0
-    for score in scores:
-        x = math.exp(score - max_score)
-        exp_scores.append(x)
-        total_sum += x
-
-    probs = []
-    for score in exp_scores:
-        probs.append(score / total_sum)
-    return probs
+    return out_eval
diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index afbe4270f5..70dc9faf54 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -306,13 +306,13 @@ class SquadProcessor(DataProcessor):
                     else:
                         is_impossible = False
 
-                    if not is_impossible and is_training:
-                        if (len(qa["answers"]) != 1):
-                            raise ValueError(
-                                "For training, each question should have exactly 1 answer.")
-                        answer = qa["answers"][0]
-                        answer_text = answer['text']
-                        start_position_character = answer['answer_start']
+                    if not is_impossible:
+                        if is_training:
+                            answer = qa["answers"][0]
+                            answer_text = answer['text']
+                            start_position_character = answer['answer_start']
+                        else:
+                            answers = qa["answers"]
 
                     example = SquadExample(
                         qas_id=qas_id,
@@ -321,7 +321,8 @@ class SquadProcessor(DataProcessor):
                         answer_text=answer_text,
                         start_position_character=start_position_character,
                         title=title,
-                        is_impossible=is_impossible
+                        is_impossible=is_impossible,
+                        answers=answers
                     )
 
                     examples.append(example)
@@ -352,6 +353,7 @@ class SquadExample(object):
                  answer_text,
                  start_position_character,
                  title,
+                 answers=None,
                  is_impossible=False):
         self.qas_id = qas_id
         self.question_text = question_text
@@ -359,6 +361,7 @@ class SquadExample(object):
         self.answer_text = answer_text
         self.title = title
         self.is_impossible = is_impossible 
+        self.answers = answers
 
         self.start_position, self.end_position = 0, 0
 

From e4fbf3e2cc26b1476b3333ec0b1ffef949277262 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Wed, 4 Dec 2019 00:52:23 -0500
Subject: [PATCH 212/505] CLI for authenticated file sharing

---
 setup.py                          |  10 ++
 transformers-cli                  |  23 +++++
 transformers/commands/__init__.py |  12 +++
 transformers/commands/user.py     | 122 ++++++++++++++++++++++++
 transformers/hf_api.py            | 152 ++++++++++++++++++++++++++++++
 transformers/tests/hf_api_test.py |  94 ++++++++++++++++++
 6 files changed, 413 insertions(+)
 create mode 100644 transformers-cli
 create mode 100644 transformers/commands/__init__.py
 create mode 100644 transformers/commands/user.py
 create mode 100644 transformers/hf_api.py
 create mode 100644 transformers/tests/hf_api_test.py

diff --git a/setup.py b/setup.py
index c07920520d..25f503f8d0 100644
--- a/setup.py
+++ b/setup.py
@@ -36,6 +36,12 @@ To create the package for pypi.
 from io import open
 from setuptools import find_packages, setup
 
+
+extras = {
+    'serving': ['uvicorn', 'fastapi']
+}
+extras['all'] = [package for package in extras.values()]
+
 setup(
     name="transformers",
     version="2.2.1",
@@ -61,6 +67,10 @@ setup(
         "transformers=transformers.__main__:main",
       ]
     },
+    extras_require=extras,
+    scripts=[
+        'transformers-cli'
+    ],
     # python_requires='>=3.5.0',
     tests_require=['pytest'],
     classifiers=[
diff --git a/transformers-cli b/transformers-cli
new file mode 100644
index 0000000000..ef00d15aa3
--- /dev/null
+++ b/transformers-cli
@@ -0,0 +1,23 @@
+#!/usr/bin/env python
+from argparse import ArgumentParser
+
+from transformers.commands.user import UserCommands
+
+
+if __name__ == '__main__':
+    parser = ArgumentParser(description='Transformers CLI tool', usage='transformers-cli <command> [<args>]')
+    commands_parser = parser.add_subparsers(help='transformers-cli command helpers')
+
+    # Register commands
+    UserCommands.register_subcommand(commands_parser)
+
+    # Let's go
+    args = parser.parse_args()
+
+    if not hasattr(args, 'func'):
+        parser.print_help()
+        exit(1)
+
+    # Run
+    service = args.func(args)
+    service.run()
diff --git a/transformers/commands/__init__.py b/transformers/commands/__init__.py
new file mode 100644
index 0000000000..bbdd5655fc
--- /dev/null
+++ b/transformers/commands/__init__.py
@@ -0,0 +1,12 @@
+from abc import ABC, abstractmethod
+from argparse import ArgumentParser
+
+class BaseTransformersCLICommand(ABC):
+    @staticmethod
+    @abstractmethod
+    def register_subcommand(parser: ArgumentParser):
+        raise NotImplementedError()
+
+    @abstractmethod
+    def run(self):
+        raise NotImplementedError()
diff --git a/transformers/commands/user.py b/transformers/commands/user.py
new file mode 100644
index 0000000000..4b826f4dc6
--- /dev/null
+++ b/transformers/commands/user.py
@@ -0,0 +1,122 @@
+from argparse import ArgumentParser
+from getpass import getpass
+import os
+
+from transformers.commands import BaseTransformersCLICommand
+from transformers.hf_api import HfApi, HfFolder, HTTPError
+
+
+class UserCommands(BaseTransformersCLICommand):
+    @staticmethod
+    def register_subcommand(parser: ArgumentParser):
+        login_parser = parser.add_parser('login')
+        login_parser.set_defaults(func=lambda args: LoginCommand(args))
+        whoami_parser = parser.add_parser('whoami')
+        whoami_parser.set_defaults(func=lambda args: WhoamiCommand(args))
+        logout_parser = parser.add_parser('logout')
+        logout_parser.set_defaults(func=lambda args: LogoutCommand(args))
+        list_parser = parser.add_parser('ls')
+        list_parser.set_defaults(func=lambda args: ListObjsCommand(args))
+        # upload
+        upload_parser = parser.add_parser('upload')
+        upload_parser.add_argument('file', type=str, help='Local filepath of the file to upload.')
+        upload_parser.add_argument('--filename', type=str, default=None, help='Optional: override object filename on S3.')
+        upload_parser.set_defaults(func=lambda args: UploadCommand(args))
+
+
+
+class BaseUserCommand:
+    def __init__(self, args):
+        self.args = args
+        self._api = HfApi()
+
+
+class LoginCommand(BaseUserCommand):
+    def run(self):
+        print("""
+        _|    _|  _|    _|    _|_|_|    _|_|_|  _|_|_|  _|      _|    _|_|_|      _|_|_|_|    _|_|      _|_|_|  _|_|_|_|  
+        _|    _|  _|    _|  _|        _|          _|    _|_|    _|  _|            _|        _|    _|  _|        _|        
+        _|_|_|_|  _|    _|  _|  _|_|  _|  _|_|    _|    _|  _|  _|  _|  _|_|      _|_|_|    _|_|_|_|  _|        _|_|_|    
+        _|    _|  _|    _|  _|    _|  _|    _|    _|    _|    _|_|  _|    _|      _|        _|    _|  _|        _|        
+        _|    _|    _|_|      _|_|_|    _|_|_|  _|_|_|  _|      _|    _|_|_|      _|        _|    _|    _|_|_|  _|_|_|_|  
+
+        """)
+        username = input("Username: ")
+        password = getpass()
+        try:
+            token = self._api.login(username, password)
+        except HTTPError as e:
+            # probably invalid credentials, display error message.
+            print(e)
+            exit(1)
+        HfFolder.save_token(token)
+        print("Login successful")
+        print("Your token:", token, "\n")
+        print("Your token has been saved to", HfFolder.path_token)
+
+
+class WhoamiCommand(BaseUserCommand):
+    def run(self):
+        token = HfFolder.get_token()
+        if token is None:
+            print("Not logged in")
+            exit()
+        try:
+            user = self._api.whoami(token)
+            print(user)
+        except HTTPError as e:
+            print(e)
+
+
+class LogoutCommand(BaseUserCommand):
+    def run(self):
+        token = HfFolder.get_token()
+        if token is None:
+            print("Not logged in")
+            exit()
+        HfFolder.delete_token()
+        self._api.logout(token)
+        print("Successfully logged out.")
+
+
+class ListObjsCommand(BaseUserCommand):
+    def run(self):
+        token = HfFolder.get_token()
+        if token is None:
+            print("Not logged in")
+            exit(1)
+        try:
+            objs = self._api.list_objs(token)
+        except HTTPError as e:
+            print(e)
+            exit(1)
+        if len(objs) == 0:
+            print("No shared file yet")
+        for obj in objs:
+            print(
+                obj.filename,
+                obj.LastModified,
+                obj.ETag,
+                obj.Size
+            )
+
+
+class UploadCommand(BaseUserCommand):
+    def run(self):
+        token = HfFolder.get_token()
+        if token is None:
+            print("Not logged in")
+            exit(1)
+        filepath = os.path.join(os.getcwd(), self.args.file)
+        filename = self.args.filename if self.args.filename is not None else os.path.basename(filepath)
+        print("About to upload file {} to S3 under filename {}".format(filepath, filename))
+        choice = input("Proceed? [Y/n] ").lower()
+        if not(choice == "" or choice == "y" or choice == "yes"):
+            print("Abort")
+            exit()
+        print("Uploading...")
+        access_url = self._api.presign_and_upload(
+            token=token, filename=filename, filepath=filepath
+        )
+        print("Your file now lives at:")
+        print(access_url)
diff --git a/transformers/hf_api.py b/transformers/hf_api.py
new file mode 100644
index 0000000000..238762ebf8
--- /dev/null
+++ b/transformers/hf_api.py
@@ -0,0 +1,152 @@
+# coding=utf-8
+# Copyright 2019-present, the HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from __future__ import absolute_import, division, print_function
+
+from typing import List, NamedTuple
+import os
+from os.path import expanduser
+
+import requests
+from requests.exceptions import HTTPError
+
+ENDPOINT = "https://huggingface.co"
+
+class S3Obj:
+    def __init__(self, filename: str, LastModified: str, ETag: str, Size: int):
+        self.filename = filename
+        self.LastModified = LastModified
+        self.ETag = ETag
+        self.Size = Size
+
+
+class PresignedUrl(NamedTuple):
+    write: str
+    access: str
+
+
+class HfApi:
+    def __init__(self, endpoint=None):
+        self.endpoint = endpoint if endpoint is not None else ENDPOINT
+
+    def login(self, username: str, password: str) -> str:
+        """
+        Call HF API to sign in a user and get a token if credentials are valid.
+
+        Outputs:
+            token if credentials are valid
+
+        Throws:
+            requests.exceptions.HTTPError if credentials are invalid
+        """
+        path = "{}/api/login".format(self.endpoint)
+        r = requests.post(path, json={"username": username, "password": password})
+        r.raise_for_status()
+        d = r.json()
+        return d["token"]
+
+    def whoami(self, token: str) -> str:
+        """
+        Call HF API to know "whoami"
+        """
+        path = "{}/api/whoami".format(self.endpoint)
+        r = requests.get(path, headers={"authorization": "Bearer {}".format(token)})
+        r.raise_for_status()
+        d = r.json()
+        return d["user"]
+
+    def logout(self, token: str):
+        """
+        Call HF API to log out.
+        """
+        path = "{}/api/logout".format(self.endpoint)
+        r = requests.post(path, headers={"authorization": "Bearer {}".format(token)})
+        r.raise_for_status()
+
+    def presign(self, token: str, filename: str) -> PresignedUrl:
+        """
+        Call HF API to get a presigned url to upload `filename` to S3.
+        """
+        path = "{}/api/presign".format(self.endpoint)
+        r = requests.post(
+            path,
+            headers={"authorization": "Bearer {}".format(token)},
+            json={"filename": filename},
+        )
+        r.raise_for_status()
+        d = r.json()
+        return PresignedUrl(**d)
+
+    def presign_and_upload(self, token: str, filename: str, filepath: str) -> str:
+        """
+        Get a presigned url, then upload file to S3.
+
+        Outputs:
+            url: Read-only url for the stored file on S3.
+        """
+        urls = self.presign(token, filename=filename)
+        # streaming upload:
+        # https://2.python-requests.org/en/master/user/advanced/#streaming-uploads
+        with open(filepath, "rb") as f:
+            r = requests.put(urls.write, data=f)
+            r.raise_for_status()
+        return urls.access
+
+    def list_objs(self, token: str) -> List[S3Obj]:
+        """
+        Call HF API to list all stored files for user.
+        """
+        path = "{}/api/listObjs".format(self.endpoint)
+        r = requests.get(path, headers={"authorization": "Bearer {}".format(token)})
+        r.raise_for_status()
+        d = r.json()
+        return [S3Obj(**x) for x in d]
+
+
+
+
+
+class HfFolder:
+    path_token = expanduser("~/.huggingface/token")
+
+    @classmethod
+    def save_token(cls, token: str):
+        """
+        Save token, creating folder as needed.
+        """
+        os.makedirs(os.path.dirname(cls.path_token), exist_ok=True)
+        with open(cls.path_token, 'w+') as f:
+            f.write(token)
+
+    @classmethod
+    def get_token(cls):
+        """
+        Get token or None if not existent.
+        """
+        try:
+            with open(cls.path_token, 'r') as f:
+                return f.read()
+        except FileNotFoundError:
+            return None
+
+    @classmethod
+    def delete_token(cls):
+        """
+        Delete token.
+        Do not fail if token does not exist.
+        """
+        try:
+            os.remove(cls.path_token)
+        except:
+            return
diff --git a/transformers/tests/hf_api_test.py b/transformers/tests/hf_api_test.py
new file mode 100644
index 0000000000..59822344ba
--- /dev/null
+++ b/transformers/tests/hf_api_test.py
@@ -0,0 +1,94 @@
+# coding=utf-8
+# Copyright 2019-present, the HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from __future__ import absolute_import, division, print_function
+
+import os
+import time
+import unittest
+
+from transformers.hf_api import HfApi, S3Obj, PresignedUrl, HfFolder, HTTPError
+
+USER = "__DUMMY_TRANSFORMERS_USER__"
+PASS = "__DUMMY_TRANSFORMERS_PASS__"
+FILE_KEY = "Test-{}.txt".format(int(time.time()))
+FILE_PATH = os.path.join(
+    os.path.dirname(os.path.abspath(__file__)), "fixtures/input.txt"
+)
+
+
+
+class HfApiCommonTest(unittest.TestCase):
+    _api = HfApi(endpoint="https://moon-staging.huggingface.co")
+
+
+class HfApiLoginTest(HfApiCommonTest):
+    def test_login_invalid(self):
+        with self.assertRaises(HTTPError):
+            self._api.login(username=USER, password="fake")
+
+    def test_login_valid(self):
+        token = self._api.login(username=USER, password=PASS)
+        self.assertIsInstance(token, str)
+
+
+class HfApiEndpointsTest(HfApiCommonTest):
+    @classmethod
+    def setUpClass(cls):
+        """
+        Share this valid token in all tests below.
+        """
+        cls._token = cls._api.login(username=USER, password=PASS)
+
+    def test_whoami(self):
+        user = self._api.whoami(token=self._token)
+        self.assertEqual(user, USER)
+
+    def test_presign(self):
+        url = self._api.presign(token=self._token, filename=FILE_KEY)
+        self.assertIsInstance(url, PresignedUrl)
+
+    def test_presign_and_upload(self):
+        access_url = self._api.presign_and_upload(
+            token=self._token, filename=FILE_KEY, filepath=FILE_PATH
+        )
+        self.assertIsInstance(access_url, str)
+
+    def test_list_objs(self):
+        objs = self._api.list_objs(token=self._token)
+        o = objs[-1]
+        self.assertIsInstance(o, S3Obj)
+
+
+
+class HfFolderTest(unittest.TestCase):
+    def test_token_workflow(self):
+        """
+        Test the whole token save/get/delete workflow,
+        with the desired behavior with respect to non-existent tokens.
+        """
+        token = "token-{}".format(int(time.time()))
+        HfFolder.save_token(token)
+        self.assertEqual(
+            HfFolder.get_token(),
+            token
+        )
+        HfFolder.delete_token()
+        HfFolder.delete_token()
+        # ^^ not an error, we test that the
+        # second call does not fail.
+        self.assertEqual(
+            HfFolder.get_token(),
+            None
+        )

From 40255ab00207f343d5e913c616c20f0f79504bfe Mon Sep 17 00:00:00 2001
From: Aymeric Augustin <aymeric.augustin@m4x.org>
Date: Wed, 4 Dec 2019 08:21:02 +0100
Subject: [PATCH 213/505] Remove dead code in tests.

---
 transformers/tests/modeling_tf_common_test.py | 169 ------------------
 1 file changed, 169 deletions(-)

diff --git a/transformers/tests/modeling_tf_common_test.py b/transformers/tests/modeling_tf_common_test.py
index ea8cd1aecd..7445ce826a 100644
--- a/transformers/tests/modeling_tf_common_test.py
+++ b/transformers/tests/modeling_tf_common_test.py
@@ -233,80 +233,6 @@ class TFCommonTestCases:
                     self.model_tester.seq_length,
                     self.model_tester.key_len if hasattr(self.model_tester, 'key_len') else self.model_tester.seq_length])
 
-        def test_headmasking(self):
-            pass
-            # config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
-
-            # config.output_attentions = True
-            # config.output_hidden_states = True
-            # configs_no_init = _config_zero_init(config)  # To be sure we have no Nan
-            # for model_class in self.all_model_classes:
-            #     model = model_class(config=configs_no_init)
-            #     model.eval()
-
-            #     # Prepare head_mask
-            #     # Set require_grad after having prepared the tensor to avoid error (leaf variable has been moved into the graph interior) 
-            #     head_mask = torch.ones(self.model_tester.num_hidden_layers, self.model_tester.num_attention_heads)
-            #     head_mask[0, 0] = 0
-            #     head_mask[-1, :-1] = 0
-            #     head_mask.requires_grad_(requires_grad=True)
-            #     inputs = inputs_dict.copy()
-            #     inputs['head_mask'] = head_mask
-
-            #     outputs = model(**inputs)
-
-            #     # Test that we can get a gradient back for importance score computation
-            #     output = sum(t.sum() for t in outputs[0])
-            #     output = output.sum()
-            #     output.backward()
-            #     multihead_outputs = head_mask.grad
-
-            #     attentions = outputs[-1]
-            #     hidden_states = outputs[-2]
-
-            #     # Remove Nan
-
-            #     self.assertIsNotNone(multihead_outputs)
-            #     self.assertEqual(len(multihead_outputs), self.model_tester.num_hidden_layers)
-            #     self.assertAlmostEqual(
-            #         attentions[0][..., 0, :, :].flatten().sum().item(), 0.0)
-            #     self.assertNotEqual(
-            #         attentions[0][..., -1, :, :].flatten().sum().item(), 0.0)
-            #     self.assertNotEqual(
-            #         attentions[1][..., 0, :, :].flatten().sum().item(), 0.0)
-            #     self.assertAlmostEqual(
-            #         attentions[-1][..., -2, :, :].flatten().sum().item(), 0.0)
-            #     self.assertNotEqual(
-            #         attentions[-1][..., -1, :, :].flatten().sum().item(), 0.0)
-
-
-        def test_head_pruning(self):
-            pass
-            # if not self.test_pruning:
-            #     return
-
-            # config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
-
-            # for model_class in self.all_model_classes:
-            #     config.output_attentions = True
-            #     config.output_hidden_states = False
-            #     model = model_class(config=config)
-            #     model.eval()
-            #     heads_to_prune = {0: list(range(1, self.model_tester.num_attention_heads)),
-            #                     -1: [0]}
-            #     model.prune_heads(heads_to_prune)
-            #     outputs = model(**inputs_dict)
-
-            #     attentions = outputs[-1]
-
-            #     self.assertEqual(
-            #         attentions[0].shape[-3], 1)
-            #     self.assertEqual(
-            #         attentions[1].shape[-3], self.model_tester.num_attention_heads)
-            #     self.assertEqual(
-            #         attentions[-1].shape[-3], self.model_tester.num_attention_heads - 1)
-
-
         def test_hidden_states_output(self):
             config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
 
@@ -323,43 +249,6 @@ class TFCommonTestCases:
                     list(hidden_states[0].shape[-2:]),
                     [self.model_tester.seq_length, self.model_tester.hidden_size])
 
-
-        def test_resize_tokens_embeddings(self):
-            pass
-            # original_config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
-            # if not self.test_resize_embeddings:
-            #     return
-
-            # for model_class in self.all_model_classes:
-            #     config = copy.deepcopy(original_config)
-            #     model = model_class(config)
-
-            #     model_vocab_size = config.vocab_size
-            #     # Retrieve the embeddings and clone theme
-            #     model_embed = model.resize_token_embeddings(model_vocab_size)
-            #     cloned_embeddings = model_embed.weight.clone()
-
-            #     # Check that resizing the token embeddings with a larger vocab size increases the model's vocab size
-            #     model_embed = model.resize_token_embeddings(model_vocab_size + 10)
-            #     self.assertEqual(model.config.vocab_size, model_vocab_size + 10)
-            #     # Check that it actually resizes the embeddings matrix
-            #     self.assertEqual(model_embed.weight.shape[0], cloned_embeddings.shape[0] + 10)
-
-            #     # Check that resizing the token embeddings with a smaller vocab size decreases the model's vocab size
-            #     model_embed = model.resize_token_embeddings(model_vocab_size - 15)
-            #     self.assertEqual(model.config.vocab_size, model_vocab_size - 15)
-            #     # Check that it actually resizes the embeddings matrix
-            #     self.assertEqual(model_embed.weight.shape[0], cloned_embeddings.shape[0] - 15)
-
-            #     # Check that adding and removing tokens has not modified the first part of the embedding matrix.
-            #     models_equal = True
-            #     for p1, p2 in zip(cloned_embeddings, model_embed.weight):
-            #         if p1.data.ne(p2.data).sum() > 0:
-            #             models_equal = False
-
-            #     self.assertTrue(models_equal)
-
-
         def test_model_common_attributes(self):
             config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
 
@@ -369,40 +258,6 @@ class TFCommonTestCases:
                 x = model.get_output_embeddings()
                 assert x is None or isinstance(x, tf.keras.layers.Layer)
 
-
-        def test_tie_model_weights(self):
-            pass
-            # config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
-
-            # def check_same_values(layer_1, layer_2):
-            #     equal = True
-            #     for p1, p2 in zip(layer_1.weight, layer_2.weight):
-            #         if p1.data.ne(p2.data).sum() > 0:
-            #             equal = False
-            #     return equal
-
-            # for model_class in self.all_model_classes:
-            #     if not hasattr(model_class, 'tie_weights'):
-            #         continue
-
-            #     config.torchscript = True
-            #     model_not_tied = model_class(config)
-            #     params_not_tied = list(model_not_tied.parameters())
-
-            #     config_tied = copy.deepcopy(config)
-            #     config_tied.torchscript = False
-            #     model_tied = model_class(config_tied)
-            #     params_tied = list(model_tied.parameters())
-
-            #     # Check that the embedding layer and decoding layer are the same in size and in value
-            #     self.assertGreater(len(params_not_tied), len(params_tied))
-
-            #     # Check that after resize they remain tied.
-            #     model_tied.resize_token_embeddings(config.vocab_size + 10)
-            #     params_tied_2 = list(model_tied.parameters())
-            #     self.assertGreater(len(params_not_tied), len(params_tied))
-            #     self.assertEqual(len(params_tied_2), len(params_tied))
-
         def test_determinism(self):
             config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
 
@@ -461,29 +316,5 @@ def ids_tensor(shape, vocab_size, rng=None, name=None, dtype=None):
     return output
 
 
-class TFModelUtilsTest(unittest.TestCase):
-    @pytest.mark.skipif('tensorflow' not in sys.modules, reason="requires TensorFlow")
-    def test_model_from_pretrained(self):
-        pass
-        # logging.basicConfig(level=logging.INFO)
-        # for model_name in list(BERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-        #     config = BertConfig.from_pretrained(model_name)
-        #     self.assertIsNotNone(config)
-        #     self.assertIsInstance(config, PretrainedConfig)
-
-        #     model = BertModel.from_pretrained(model_name)
-        #     model, loading_info = BertModel.from_pretrained(model_name, output_loading_info=True)
-        #     self.assertIsNotNone(model)
-        #     self.assertIsInstance(model, PreTrainedModel)
-        #     for value in loading_info.values():
-        #         self.assertEqual(len(value), 0)
-
-        #     config = BertConfig.from_pretrained(model_name, output_attentions=True, output_hidden_states=True)
-        #     model = BertModel.from_pretrained(model_name, output_attentions=True, output_hidden_states=True)
-        #     self.assertEqual(model.config.output_attentions, True)
-        #     self.assertEqual(model.config.output_hidden_states, True)
-        #     self.assertEqual(model.config, config)
-
-
 if __name__ == "__main__":
     unittest.main()

From ecb923da9cea390742a1262327a139852c5493e9 Mon Sep 17 00:00:00 2001
From: Julien Plu <julien.plu@schibsted.com>
Date: Wed, 4 Dec 2019 09:43:15 +0100
Subject: [PATCH 214/505] Create a NER example similar to the Pytorch one. It
 takes the same options, and can be run the same way.

---
 examples/run_tf_ner.py                 | 612 +++++++++++++++++++++++++
 transformers/__init__.py               |   3 +
 transformers/modeling_tf_distilbert.py |  47 ++
 transformers/optimization_tf.py        | 254 ++++++++++
 4 files changed, 916 insertions(+)
 create mode 100644 examples/run_tf_ner.py
 create mode 100644 transformers/optimization_tf.py

diff --git a/examples/run_tf_ner.py b/examples/run_tf_ner.py
new file mode 100644
index 0000000000..ef1fcf6aa4
--- /dev/null
+++ b/examples/run_tf_ner.py
@@ -0,0 +1,612 @@
+# coding=utf-8
+import datetime
+import os
+import math
+import glob
+import re
+import tensorflow as tf
+import collections
+import numpy as np
+from seqeval import metrics
+import _pickle as pickle
+from absl import logging
+from transformers import TF2_WEIGHTS_NAME, BertConfig, BertTokenizer, TFBertForTokenClassification
+from transformers import RobertaConfig, RobertaTokenizer, TFRobertaForTokenClassification
+from transformers import DistilBertConfig, DistilBertTokenizer, TFDistilBertForTokenClassification
+from transformers import create_optimizer, GradientAccumulator
+from utils_ner import convert_examples_to_features, get_labels, read_examples_from_file
+from fastprogress import master_bar, progress_bar
+from absl import flags
+from absl import app
+
+
+ALL_MODELS = sum(
+    (tuple(conf.pretrained_config_archive_map.keys()) for conf in (BertConfig, RobertaConfig, DistilBertConfig)),
+    ())
+
+MODEL_CLASSES = {
+    "bert": (BertConfig, TFBertForTokenClassification, BertTokenizer),
+    "roberta": (RobertaConfig, TFRobertaForTokenClassification, RobertaTokenizer),
+    "distilbert": (DistilBertConfig, TFDistilBertForTokenClassification, DistilBertTokenizer)
+}
+
+
+flags.DEFINE_string(
+    "data_dir", None,
+    "The input data dir. Should contain the .conll files (or other data files) "
+    "for the task.")
+
+flags.DEFINE_string(
+    "model_type", None,
+    "Model type selected in the list: " + ", ".join(MODEL_CLASSES.keys()))
+
+flags.DEFINE_string(
+    "model_name_or_path", None,
+    "Path to pre-trained model or shortcut name selected in the list: " + ", ".join(ALL_MODELS))
+
+flags.DEFINE_string(
+    "output_dir", None,
+    "The output directory where the model checkpoints will be written.")
+
+flags.DEFINE_string(
+    "labels", "",
+    "Path to a file containing all labels. If not specified, CoNLL-2003 labels are used.")
+
+flags.DEFINE_string(
+    "config_name", "",
+    "Pretrained config name or path if not the same as model_name")
+
+flags.DEFINE_string(
+    "tokenizer_name", "",
+    "Pretrained tokenizer name or path if not the same as model_name")
+
+flags.DEFINE_string(
+    "cache_dir", "",
+    "Where do you want to store the pre-trained models downloaded from s3")
+
+flags.DEFINE_integer(
+    "max_seq_length", 128,
+    "The maximum total input sentence length after tokenization. "
+    "Sequences longer than this will be truncated, sequences shorter "
+    "will be padded.")
+
+flags.DEFINE_string(
+    "tpu", None,
+    "The Cloud TPU to use for training. This should be either the name "
+    "used when creating the Cloud TPU, or a grpc://ip.address.of.tpu:8470 "
+    "url.")
+
+flags.DEFINE_integer(
+    "num_tpu_cores", 8,
+    "Total number of TPU cores to use.")
+
+flags.DEFINE_boolean(
+    "do_train", False,
+    "Whether to run training.")
+
+flags.DEFINE_boolean(
+    "do_eval", False,
+    "Whether to run eval on the dev set.")
+
+flags.DEFINE_boolean(
+    "do_predict", False,
+    "Whether to run predictions on the test set.")
+
+flags.DEFINE_boolean(
+    "evaluate_during_training", False,
+    "Whether to run evaluation during training at each logging step.")
+
+flags.DEFINE_boolean(
+    "do_lower_case", False,
+    "Set this flag if you are using an uncased model.")
+
+flags.DEFINE_integer(
+    "per_device_train_batch_size", 8,
+    "Batch size per GPU/CPU/TPU for training.")
+
+flags.DEFINE_integer(
+    "per_device_eval_batch_size", 8,
+    "Batch size per GPU/CPU/TPU for evaluation.")
+
+flags.DEFINE_integer(
+    "gradient_accumulation_steps", 1,
+    "Number of updates steps to accumulate before performing a backward/update pass.")
+
+flags.DEFINE_float(
+    "learning_rate", 5e-5,
+    "The initial learning rate for Adam.")
+
+flags.DEFINE_float(
+    "weight_decay", 0.0,
+    "Weight decay if we apply some.")
+
+flags.DEFINE_float(
+    "adam_epsilon", 1e-8,
+    "Epsilon for Adam optimizer.")
+
+flags.DEFINE_float(
+    "max_grad_norm", 1.0,
+    "Max gradient norm.")
+
+flags.DEFINE_integer(
+    "num_train_epochs", 3,
+    "Total number of training epochs to perform.")
+
+flags.DEFINE_integer(
+    "max_steps", -1,
+    "If > 0: set total number of training steps to perform. Override num_train_epochs.")
+
+flags.DEFINE_integer(
+    "warmup_steps", 0,
+    "Linear warmup over warmup_steps.")
+
+flags.DEFINE_integer(
+    "logging_steps", 50,
+    "Log every X updates steps.")
+
+flags.DEFINE_integer(
+    "save_steps", 50,
+    "Save checkpoint every X updates steps.")
+
+flags.DEFINE_boolean(
+    "eval_all_checkpoints", False,
+    "Evaluate all checkpoints starting with the same prefix as model_name ending and ending with step number")
+
+flags.DEFINE_boolean(
+    "no_cuda", False,
+    "Avoid using CUDA when available")
+
+flags.DEFINE_boolean(
+    "overwrite_output_dir", False,
+    "Overwrite the content of the output directory")
+
+flags.DEFINE_boolean(
+    "overwrite_cache", False,
+    "Overwrite the cached training and evaluation sets")
+
+flags.DEFINE_integer(
+    "seed", 42,
+    "random seed for initialization")
+
+flags.DEFINE_boolean(
+    "fp16", False,
+    "Whether to use 16-bit (mixed) precision instead of 32-bit")
+
+flags.DEFINE_string(
+    "gpus", "0",
+    "Comma separated list of gpus devices. If only one, switch to single "
+    "gpu strategy, if None takes all the gpus available.")
+
+
+def train(args, strategy, train_dataset, tokenizer, model, num_train_examples, labels, train_batch_size, pad_token_label_id):
+    if args['max_steps'] > 0:
+        num_train_steps = args['max_steps'] * args['gradient_accumulation_steps']
+        args['num_train_epochs'] = 1
+    else:
+        num_train_steps = math.ceil(num_train_examples / train_batch_size) // args['gradient_accumulation_steps'] * args['num_train_epochs']
+
+    writer = tf.summary.create_file_writer("/tmp/mylogs")
+
+    with strategy.scope():
+        loss_fct = tf.keras.losses.SparseCategoricalCrossentropy(reduction=tf.keras.losses.Reduction.NONE)
+        optimizer = create_optimizer(args['learning_rate'], num_train_steps, args['warmup_steps'])
+
+        if args['fp16']:
+            optimizer = tf.keras.mixed_precision.experimental.LossScaleOptimizer(optimizer, 'dynamic')
+
+        loss_metric = tf.keras.metrics.Mean(name='loss', dtype=tf.float32)
+        gradient_accumulator = GradientAccumulator()
+        
+    logging.info("***** Running training *****")
+    logging.info("  Num examples = %d", num_train_examples)
+    logging.info("  Num Epochs = %d", args['num_train_epochs'])
+    logging.info("  Instantaneous batch size per device = %d", args['per_device_train_batch_size'])
+    logging.info("  Total train batch size (w. parallel, distributed & accumulation) = %d",
+                train_batch_size * args['gradient_accumulation_steps'])
+    logging.info("  Gradient Accumulation steps = %d", args['gradient_accumulation_steps'])
+    logging.info("  Total training steps = %d", num_train_steps)
+
+    model.summary()
+
+    @tf.function
+    def apply_gradients():
+        grads_and_vars = []
+
+        for gradient, variable in zip(gradient_accumulator.gradients, model.trainable_variables):
+            if gradient is not None:
+                scaled_gradient = gradient / (args['n_device'] * args['gradient_accumulation_steps'])
+                grads_and_vars.append((scaled_gradient, variable))
+            else:
+                grads_and_vars.append((gradient, variable))
+
+        optimizer.apply_gradients(grads_and_vars, args['max_grad_norm'])
+        gradient_accumulator.reset()
+
+    @tf.function
+    def train_step(train_features, train_labels):
+        def step_fn(train_features, train_labels):
+            inputs = {'attention_mask': train_features['input_mask'], 'training': True}
+
+            if args['model_type'] != "distilbert":
+                inputs["token_type_ids"] = train_features['segment_ids'] if args['model_type'] in ["bert", "xlnet"] else None
+
+            with tf.GradientTape() as tape:
+                logits = model(train_features['input_ids'], **inputs)[0]
+                logits = tf.reshape(logits, (-1, len(labels) + 1))
+                active_loss = tf.reshape(train_features['input_mask'], (-1,))
+                active_logits = tf.boolean_mask(logits, active_loss)
+                train_labels = tf.reshape(train_labels, (-1,))
+                active_labels = tf.boolean_mask(train_labels, active_loss)
+                cross_entropy = loss_fct(active_labels, active_logits)
+                loss = tf.reduce_sum(cross_entropy) * (1.0 / train_batch_size)
+                grads = tape.gradient(loss, model.trainable_variables)
+
+                gradient_accumulator(grads)
+
+            return cross_entropy
+
+        per_example_losses = strategy.experimental_run_v2(step_fn, args=(train_features, train_labels))
+        mean_loss = strategy.reduce(tf.distribute.ReduceOp.MEAN, per_example_losses, axis=0)
+
+        return mean_loss
+
+    current_time = datetime.datetime.now()
+    train_iterator = master_bar(range(args['num_train_epochs']))
+    global_step = 0
+    logging_loss = 0.0
+
+    for epoch in train_iterator:
+        epoch_iterator = progress_bar(train_dataset, total=num_train_steps, parent=train_iterator, display=args['n_device'] > 1)
+        step = 1
+
+        with strategy.scope():
+            for train_features, train_labels in epoch_iterator:
+                loss = train_step(train_features, train_labels)
+
+                if step % args['gradient_accumulation_steps'] == 0:
+                    strategy.experimental_run_v2(apply_gradients)
+
+                    loss_metric(loss)
+
+                    global_step += 1
+
+                    if args['logging_steps'] > 0 and global_step % args['logging_steps'] == 0:
+                        # Log metrics
+                        if args['n_device'] == 1 and args['evaluate_during_training']:  # Only evaluate when single GPU otherwise metrics may not average well
+                            y_true, y_pred, eval_loss = evaluate(args, strategy, model, tokenizer, labels, pad_token_label_id, mode="dev")
+                            report = metrics.classification_report(y_true, y_pred, digits=4)
+                            
+                            logging.info("Eval at step " + str(global_step) + "\n" + report)
+                            logging.info("eval_loss: " + str(eval_loss))
+                            
+                            precision = metrics.precision_score(y_true, y_pred)
+                            recall = metrics.recall_score(y_true, y_pred)
+                            f1 = metrics.f1_score(y_true, y_pred)
+
+                            with writer.as_default():
+                                tf.summary.scalar("eval_loss", eval_loss, global_step)
+                                tf.summary.scalar("precision", precision, global_step)
+                                tf.summary.scalar("recall", recall, global_step)
+                                tf.summary.scalar("f1", f1, global_step)
+                        
+                        lr = optimizer.learning_rate
+                        learning_rate = lr(step)
+
+                        with writer.as_default():
+                            tf.summary.scalar("lr", learning_rate, global_step)
+                            tf.summary.scalar("loss", (loss_metric.result() - logging_loss) / args['logging_steps'], global_step)
+                        
+                        logging_loss = loss_metric.result()
+
+                    with writer.as_default():
+                        tf.summary.scalar("loss", loss_metric.result(), step=step)
+
+                    if args['save_steps'] > 0 and global_step % args['save_steps'] == 0:
+                        # Save model checkpoint
+                        output_dir = os.path.join(args['output_dir'], "checkpoint-{}".format(global_step))
+
+                        if not os.path.exists(output_dir):
+                            os.makedirs(output_dir)
+                        
+                        model.save_pretrained(output_dir)
+                        logging.info("Saving model checkpoint to %s", output_dir)
+                
+                train_iterator.child.comment = f'loss : {loss_metric.result()}'
+                step += 1
+
+        train_iterator.write(f'loss epoch {epoch + 1}: {loss_metric.result()}')
+
+        loss_metric.reset_states()
+
+    logging.info("  Training took time = {}".format(datetime.datetime.now() - current_time))
+
+
+def evaluate(args, strategy, model, tokenizer, labels, pad_token_label_id, mode):
+    eval_batch_size = args['per_device_eval_batch_size'] * args['n_device']
+    eval_dataset, size = load_and_cache_examples(args, tokenizer, labels, pad_token_label_id, eval_batch_size, mode=mode)
+    eval_dataset = strategy.experimental_distribute_dataset(eval_dataset)
+    preds = None
+    num_eval_steps = math.ceil(size / eval_batch_size)
+    master = master_bar(range(1))
+    eval_iterator = progress_bar(eval_dataset, total=num_eval_steps, parent=master, display=args['n_device'] > 1)
+    loss_fct = tf.keras.losses.SparseCategoricalCrossentropy(reduction=tf.keras.losses.Reduction.NONE)
+    loss = 0.0
+
+    logging.info("***** Running evaluation *****")
+    logging.info("  Num examples = %d", size)
+    logging.info("  Batch size = %d", eval_batch_size)
+
+    for eval_features, eval_labels in eval_iterator:
+        inputs = {'attention_mask': eval_features['input_mask'], 'training': False}
+
+        if args['model_type'] != "distilbert":
+            inputs["token_type_ids"] = eval_features['segment_ids'] if args['model_type'] in ["bert", "xlnet"] else None
+
+        with strategy.scope():
+            logits = model(eval_features['input_ids'], **inputs)[0]
+            tmp_logits = tf.reshape(logits, (-1, len(labels) + 1))
+            active_loss = tf.reshape(eval_features['input_mask'], (-1,))
+            active_logits = tf.boolean_mask(tmp_logits, active_loss)
+            tmp_eval_labels = tf.reshape(eval_labels, (-1,))
+            active_labels = tf.boolean_mask(tmp_eval_labels, active_loss)
+            cross_entropy = loss_fct(active_labels, active_logits)
+            loss += tf.reduce_sum(cross_entropy) * (1.0 / eval_batch_size)
+
+        if preds is None:
+            preds = logits.numpy()
+            label_ids = eval_labels.numpy()
+        else:
+            preds = np.append(preds, logits.numpy(), axis=0)
+            label_ids = np.append(label_ids, eval_labels.numpy(), axis=0)
+
+    preds = np.argmax(preds, axis=2)
+    y_pred = [[] for _ in range(label_ids.shape[0])]
+    y_true = [[] for _ in range(label_ids.shape[0])]
+    loss = loss / num_eval_steps
+
+    for i in range(label_ids.shape[0]):
+        for j in range(label_ids.shape[1]):
+            if label_ids[i, j] != pad_token_label_id:
+                y_pred[i].append(labels[preds[i, j] - 1])
+                y_true[i].append(labels[label_ids[i, j] - 1])
+
+    return y_true, y_pred, loss.numpy()
+
+
+def load_cache(cached_file, max_seq_length):
+    name_to_features = {
+        "input_ids": tf.io.FixedLenFeature([max_seq_length], tf.int64),
+        "input_mask": tf.io.FixedLenFeature([max_seq_length], tf.int64),
+        "segment_ids": tf.io.FixedLenFeature([max_seq_length], tf.int64),
+        "label_ids": tf.io.FixedLenFeature([max_seq_length], tf.int64),
+    }
+
+    def _decode_record(record):
+        example = tf.io.parse_single_example(record, name_to_features)
+        features = {}
+        features['input_ids'] = example['input_ids']
+        features['input_mask'] = example['input_mask']
+        features['segment_ids'] = example['segment_ids']
+
+        return features, example['label_ids']
+
+    d = tf.data.TFRecordDataset(cached_file)
+    d = d.map(_decode_record, num_parallel_calls=4)
+    count = d.reduce(0, lambda x, _: x + 1)
+
+    return d, count.numpy()
+
+
+def save_cache(features, cached_features_file):
+    writer = tf.io.TFRecordWriter(cached_features_file)
+
+    for (ex_index, feature) in enumerate(features):
+        if ex_index % 5000 == 0:
+            logging.info("Writing example %d of %d" % (ex_index, len(features)))
+
+        def create_int_feature(values):
+            f = tf.train.Feature(int64_list=tf.train.Int64List(value=list(values)))
+            return f
+
+        record_feature = collections.OrderedDict()
+        record_feature["input_ids"] = create_int_feature(feature.input_ids)
+        record_feature["input_mask"] = create_int_feature(feature.input_mask)
+        record_feature["segment_ids"] = create_int_feature(feature.segment_ids)
+        record_feature["label_ids"] = create_int_feature(feature.label_ids)
+
+        tf_example = tf.train.Example(features=tf.train.Features(feature=record_feature))
+
+        writer.write(tf_example.SerializeToString())
+
+    writer.close()
+
+
+def load_and_cache_examples(args, tokenizer, labels, pad_token_label_id, batch_size, mode):
+    drop_remainder = True if args['tpu'] or mode == 'train' else False
+
+    # Load data features from cache or dataset file
+    cached_features_file = os.path.join(args['data_dir'], "cached_{}_{}_{}.tf_record".format(mode,
+        list(filter(None, args['model_name_or_path'].split("/"))).pop(),
+        str(args['max_seq_length'])))
+    if os.path.exists(cached_features_file) and not args['overwrite_cache']:
+        logging.info("Loading features from cached file %s", cached_features_file)
+        dataset, size = load_cache(cached_features_file, args['max_seq_length'])
+    else:
+        logging.info("Creating features from dataset file at %s", args['data_dir'])
+        examples = read_examples_from_file(args['data_dir'], mode)
+        features = convert_examples_to_features(examples, labels, args['max_seq_length'], tokenizer,
+                                                cls_token_at_end=bool(args['model_type'] in ["xlnet"]),
+                                                # xlnet has a cls token at the end
+                                                cls_token=tokenizer.cls_token,
+                                                cls_token_segment_id=2 if args['model_type'] in ["xlnet"] else 0,
+                                                sep_token=tokenizer.sep_token,
+                                                sep_token_extra=bool(args['model_type'] in ["roberta"]),
+                                                # roberta uses an extra separator b/w pairs of sentences, cf. github.com/pytorch/fairseq/commit/1684e166e3da03f5b600dbb7855cb98ddfcd0805
+                                                pad_on_left=bool(args['model_type'] in ["xlnet"]),
+                                                # pad on the left for xlnet
+                                                pad_token=tokenizer.convert_tokens_to_ids([tokenizer.pad_token])[0],
+                                                pad_token_segment_id=4 if args['model_type'] in ["xlnet"] else 0,
+                                                pad_token_label_id=pad_token_label_id
+                                                )
+        logging.info("Saving features into cached file %s", cached_features_file)
+        save_cache(features, cached_features_file)
+        dataset, size = load_cache(cached_features_file, args['max_seq_length'])
+
+    if mode == 'train':
+        dataset = dataset.repeat()
+        dataset = dataset.shuffle(buffer_size=8192, seed=args['seed'])
+
+    dataset = dataset.batch(batch_size, drop_remainder)
+    dataset = dataset.prefetch(buffer_size=batch_size)
+
+    return dataset, size
+
+
+def main(_):
+    logging.set_verbosity(logging.INFO)
+    args = flags.FLAGS.flag_values_dict()
+
+    if os.path.exists(args['output_dir']) and os.listdir(
+            args['output_dir']) and args['do_train'] and not args['overwrite_output_dir']:
+        raise ValueError(
+            "Output directory ({}) already exists and is not empty. Use --overwrite_output_dir to overcome.".format(
+                args['output_dir']))
+
+    if args['fp16']:
+        tf.config.optimizer.set_experimental_options({"auto_mixed_precision": True})
+
+    if args['tpu']:
+        resolver = tf.distribute.cluster_resolver.TPUClusterResolver(tpu=args['tpu'])
+        tf.config.experimental_connect_to_cluster(resolver)
+        tf.tpu.experimental.initialize_tpu_system(resolver)
+        strategy = tf.distribute.experimental.TPUStrategy(resolver)
+        args['n_device'] = args['num_tpu_cores']
+    elif len(args['gpus'].split(',')) > 1:
+        args['n_device'] = len([f"/gpu:{gpu}" for gpu in args['gpus'].split(',')])
+        strategy = tf.distribute.MirroredStrategy(devices=[f"/gpu:{gpu}" for gpu in args['gpus'].split(',')])
+    elif args['no_cuda']:
+        args['n_device'] = 1
+        strategy = tf.distribute.OneDeviceStrategy(device="/cpu:0")
+    else:
+        args['n_device'] = len(args['gpus'].split(','))
+        strategy = tf.distribute.OneDeviceStrategy(device="/gpu:" + args['gpus'].split(',')[0])
+
+    logging.warning("n_device: %s, distributed training: %s, 16-bits training: %s",
+                   args['n_device'], bool(args['n_device'] > 1), args['fp16'])
+
+    labels = get_labels(args['labels'])
+    num_labels = len(labels) + 1
+    pad_token_label_id = 0
+    config_class, model_class, tokenizer_class = MODEL_CLASSES[args['model_type']]
+    config = config_class.from_pretrained(args['config_name'] if args['config_name'] else args['model_name_or_path'],
+                                          num_labels=num_labels,
+                                          cache_dir=args['cache_dir'] if args['cache_dir'] else None)
+
+    logging.info("Training/evaluation parameters %s", args)
+
+    # Training
+    if args['do_train']:
+        tokenizer = tokenizer_class.from_pretrained(args['tokenizer_name'] if args['tokenizer_name'] else args['model_name_or_path'],
+                                                    do_lower_case=args['do_lower_case'],
+                                                    cache_dir=args['cache_dir'] if args['cache_dir'] else None)
+
+        with strategy.scope():
+            model = model_class.from_pretrained(args['model_name_or_path'],
+                                                from_pt=bool(".bin" in args['model_name_or_path']),
+                                                config=config,
+                                                cache_dir=args['cache_dir'] if args['cache_dir'] else None)
+            model.layers[-1].activation = tf.keras.activations.softmax
+
+        train_batch_size = args['per_device_train_batch_size'] * args['n_device']
+        train_dataset, num_train_examples = load_and_cache_examples(args, tokenizer, labels, pad_token_label_id, train_batch_size, mode="train")
+        train_dataset = strategy.experimental_distribute_dataset(train_dataset)
+        train(args, strategy, train_dataset, tokenizer, model, num_train_examples, labels, train_batch_size, pad_token_label_id)
+
+        if not os.path.exists(args['output_dir']):
+            os.makedirs(args['output_dir'])
+
+        logging.info("Saving model to %s", args['output_dir'])
+
+        model.save_pretrained(args['output_dir'])
+        tokenizer.save_pretrained(args['output_dir'])
+
+    # Evaluation
+    if args['do_eval']:
+        tokenizer = tokenizer_class.from_pretrained(args['output_dir'], do_lower_case=args['do_lower_case'])
+        checkpoints = []
+        results = []
+
+        if args['eval_all_checkpoints']:
+            checkpoints = list(os.path.dirname(c) for c in sorted(glob.glob(args['output_dir'] + "/**/" + TF2_WEIGHTS_NAME, recursive=True), key=lambda f: int(''.join(filter(str.isdigit, f)) or -1)))
+        
+        logging.info("Evaluate the following checkpoints: %s", checkpoints)
+        
+        for checkpoint in checkpoints:
+            global_step = checkpoint.split("-")[-1] if re.match(".*checkpoint-[0-9]", checkpoint) else "final"
+
+            with strategy.scope():
+                model = model_class.from_pretrained(checkpoint)
+
+            y_true, y_pred, eval_loss = evaluate(args, strategy, model, tokenizer, labels, pad_token_label_id, mode="dev")
+            report = metrics.classification_report(y_true, y_pred, digits=4)
+
+            if global_step:
+                results.append({global_step + "_report": report, global_step + "_loss": eval_loss})
+
+        output_eval_file = os.path.join(args['output_dir'], "eval_results.txt")
+        
+        with tf.io.gfile.GFile(output_eval_file, "w") as writer:
+            for res in results:
+                for key, val in res.items():
+                    if "loss" in key:
+                        logging.info(key + " = " + str(val))
+                        writer.write(key + " = " + str(val))
+                        writer.write("\n")
+                    else:
+                        logging.info(key)
+                        logging.info("\n" + report)
+                        writer.write(key + "\n")
+                        writer.write(report)
+                        writer.write("\n")
+
+    if args['do_predict']:
+        tokenizer = tokenizer_class.from_pretrained(args['output_dir'], do_lower_case=args['do_lower_case'])
+        model = model_class.from_pretrained(args['output_dir'])
+        eval_batch_size = args['per_gpu_eval_batch_size'] * args['n_device']
+        predict_dataset, _ = load_and_cache_examples(args, tokenizer, labels, pad_token_label_id, eval_batch_size, mode="test")
+        y_true, y_pred, pred_loss = evaluate(args, strategy, model, tokenizer, labels, pad_token_label_id, mode="test")
+        output_test_results_file = os.path.join(args.output_dir, "test_results.txt")
+        output_test_predictions_file = os.path.join(args['output_dir'], "test_predictions.txt")
+        report = metrics.classification_report(y_true, y_pred, digits=4)
+
+        with tf.io.gfile.GFile(output_test_results_file, "w") as writer:
+            report = metrics.classification_report(y_true, y_pred, digits=4)
+            
+            logging.info("\n" + report)
+            
+            writer.write(report)
+            writer.write("\n\nloss = " + str(pred_loss))
+
+        with tf.io.gfile.GFile(output_test_predictions_file, "w") as writer:
+            with tf.io.gfile.GFile(os.path.join(args['data_dir'], "test.txt"), "r") as f:
+                example_id = 0
+
+                for line in f:
+                    if line.startswith("-DOCSTART-") or line == "" or line == "\n":
+                        writer.write(line)
+
+                        if not y_pred[example_id]:
+                            example_id += 1
+                    elif y_pred[example_id]:
+                        output_line = line.split()[0] + " " + y_pred[example_id].pop(0) + "\n"
+                        writer.write(output_line)
+                    else:
+                        logging.warning("Maximum sequence length exceeded: No prediction for '%s'.", line.split()[0])
+
+
+if __name__ == "__main__":
+    flags.mark_flag_as_required("data_dir")
+    flags.mark_flag_as_required("output_dir")
+    flags.mark_flag_as_required("model_name_or_path")
+    flags.mark_flag_as_required("model_type")
+    app.run(main)
diff --git a/transformers/__init__.py b/transformers/__init__.py
index 970bdf0cf1..2f74b7e79c 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -162,6 +162,7 @@ if is_tf_available():
     from .modeling_tf_distilbert import (TFDistilBertPreTrainedModel, TFDistilBertMainLayer,
                                          TFDistilBertModel, TFDistilBertForMaskedLM,
                                          TFDistilBertForSequenceClassification,
+					 TFDistilBertForTokenClassification
                                          TFDistilBertForQuestionAnswering,
                                          TF_DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
 
@@ -172,6 +173,8 @@ if is_tf_available():
     from .modeling_tf_albert import (TFAlbertPreTrainedModel, TFAlbertModel, TFAlbertForMaskedLM,
                                      TFAlbertForSequenceClassification,
                                     TF_ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
+    # Optimization
+    from .optimization_tf import (WarmUp, create_optimizer, AdamWeightDecay, GradientAccumulator)
 
 # TF 2.0 <=> PyTorch conversion utilities
 from .modeling_tf_pytorch_utils import (convert_tf_weight_name_to_pt_weight_name,
diff --git a/transformers/modeling_tf_distilbert.py b/transformers/modeling_tf_distilbert.py
index b3d4889475..8e1aef7462 100644
--- a/transformers/modeling_tf_distilbert.py
+++ b/transformers/modeling_tf_distilbert.py
@@ -703,6 +703,53 @@ class TFDistilBertForSequenceClassification(TFDistilBertPreTrainedModel):
         return outputs  # logits, (hidden_states), (attentions)
 
 
+@add_start_docstrings("""DistilBert Model with a token classification head on top (a linear layer on top of
+    the hidden-states output) e.g. for Named-Entity-Recognition (NER) tasks. """,
+    DISTILBERT_START_DOCSTRING, DISTILBERT_INPUTS_DOCSTRING)
+class TFDistilBertForTokenClassification(TFDistilBertPreTrainedModel):
+    r"""
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **scores**: ``Numpy array`` or ``tf.Tensor`` of shape ``(batch_size, sequence_length, config.num_labels)``
+            Classification scores (before SoftMax).
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``Numpy array`` or ``tf.Tensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``Numpy array`` or ``tf.Tensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+    Examples::
+        import tensorflow as tf
+        from transformers import DistilBertTokenizer, TFDistilBertForTokenClassification
+        tokenizer = DistilBertTokenizer.from_pretrained('bert-base-uncased')
+        model = TFDistilBertForTokenClassification.from_pretrained('bert-base-uncased')
+        input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
+        outputs = model(input_ids)
+        scores = outputs[0]
+    """
+    def __init__(self, config, *inputs, **kwargs):
+        super(TFDistilBertForTokenClassification, self).__init__(config, *inputs, **kwargs)
+        self.num_labels = config.num_labels
+
+        self.distilbert = TFDistilBertMainLayer(config, name='distilbert')
+        self.dropout = tf.keras.layers.Dropout(config.dropout)
+        self.classifier = tf.keras.layers.Dense(config.num_labels,
+                                                kernel_initializer=get_initializer(config.initializer_range),
+                                                name='classifier')
+
+    def call(self, inputs, **kwargs):
+        outputs = self.distilbert(inputs, **kwargs)
+
+        sequence_output = outputs[0]
+
+        sequence_output = self.dropout(sequence_output, training=kwargs.get('training', False))
+        logits = self.classifier(sequence_output)
+
+        outputs = (logits,) + outputs[2:]  # add hidden states and attention if they are here
+
+        return outputs  # scores, (hidden_states), (attentions)
+
+
 @add_start_docstrings("""DistilBert Model with a span classification head on top for extractive question-answering tasks like SQuAD (a linear layers on top of
                          the hidden-states output to compute `span start logits` and `span end logits`). """,
                       DISTILBERT_START_DOCSTRING, DISTILBERT_INPUTS_DOCSTRING)
diff --git a/transformers/optimization_tf.py b/transformers/optimization_tf.py
new file mode 100644
index 0000000000..c5fa248083
--- /dev/null
+++ b/transformers/optimization_tf.py
@@ -0,0 +1,254 @@
+# Copyright 2019 The TensorFlow Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ==============================================================================
+"""Functions and classes related to optimization (weight updates)."""
+
+from __future__ import absolute_import
+from __future__ import division
+from __future__ import print_function
+
+import re
+
+import tensorflow as tf
+
+
+class WarmUp(tf.keras.optimizers.schedules.LearningRateSchedule):
+  """Applys a warmup schedule on a given learning rate decay schedule."""
+
+  def __init__(
+      self,
+      initial_learning_rate,
+      decay_schedule_fn,
+      warmup_steps,
+      power=1.0,
+      name=None):
+    super(WarmUp, self).__init__()
+    self.initial_learning_rate = initial_learning_rate
+    self.warmup_steps = warmup_steps
+    self.power = power
+    self.decay_schedule_fn = decay_schedule_fn
+    self.name = name
+
+  def __call__(self, step):
+    with tf.name_scope(self.name or 'WarmUp') as name:
+      # Implements polynomial warmup. i.e., if global_step < warmup_steps, the
+      # learning rate will be `global_step/num_warmup_steps * init_lr`.
+      global_step_float = tf.cast(step, tf.float32)
+      warmup_steps_float = tf.cast(self.warmup_steps, tf.float32)
+      warmup_percent_done = global_step_float / warmup_steps_float
+      warmup_learning_rate = (
+          self.initial_learning_rate *
+          tf.math.pow(warmup_percent_done, self.power))
+      return tf.cond(global_step_float < warmup_steps_float,
+                     lambda: warmup_learning_rate,
+                     lambda: self.decay_schedule_fn(step),
+                     name=name)
+
+  def get_config(self):
+    return {
+        'initial_learning_rate': self.initial_learning_rate,
+        'decay_schedule_fn': self.decay_schedule_fn,
+        'warmup_steps': self.warmup_steps,
+        'power': self.power,
+        'name': self.name
+    }
+
+
+def create_optimizer(init_lr, num_train_steps, num_warmup_steps):
+  """Creates an optimizer with learning rate schedule."""
+  # Implements linear decay of the learning rate.
+  learning_rate_fn = tf.keras.optimizers.schedules.PolynomialDecay(
+      initial_learning_rate=init_lr,
+      decay_steps=num_train_steps,
+      end_learning_rate=0.0)
+  if num_warmup_steps:
+    learning_rate_fn = WarmUp(initial_learning_rate=init_lr,
+                              decay_schedule_fn=learning_rate_fn,
+                              warmup_steps=num_warmup_steps)
+  optimizer = AdamWeightDecay(
+      learning_rate=learning_rate_fn,
+      weight_decay_rate=0.01,
+      beta_1=0.9,
+      beta_2=0.999,
+      epsilon=1e-6,
+      exclude_from_weight_decay=['layer_norm', 'bias'])
+  return optimizer
+
+
+class AdamWeightDecay(tf.keras.optimizers.Adam):
+  """Adam enables L2 weight decay and clip_by_global_norm on gradients.
+
+  Just adding the square of the weights to the loss function is *not* the
+  correct way of using L2 regularization/weight decay with Adam, since that will
+  interact with the m and v parameters in strange ways.
+
+  Instead we want ot decay the weights in a manner that doesn't interact with
+  the m/v parameters. This is equivalent to adding the square of the weights to
+  the loss with plain (non-momentum) SGD.
+  """
+
+  def __init__(self,
+               learning_rate=0.001,
+               beta_1=0.9,
+               beta_2=0.999,
+               epsilon=1e-7,
+               amsgrad=False,
+               weight_decay_rate=0.0,
+               include_in_weight_decay=None,
+               exclude_from_weight_decay=None,
+               name='AdamWeightDecay',
+               **kwargs):
+    super(AdamWeightDecay, self).__init__(
+        learning_rate, beta_1, beta_2, epsilon, amsgrad, name, **kwargs)
+    self.weight_decay_rate = weight_decay_rate
+    self._include_in_weight_decay = include_in_weight_decay
+    self._exclude_from_weight_decay = exclude_from_weight_decay
+
+  @classmethod
+  def from_config(cls, config):
+    """Creates an optimizer from its config with WarmUp custom object."""
+    custom_objects = {'WarmUp': WarmUp}
+    return super(AdamWeightDecay, cls).from_config(
+        config, custom_objects=custom_objects)
+
+  def _prepare_local(self, var_device, var_dtype, apply_state):
+    super(AdamWeightDecay, self)._prepare_local(var_device, var_dtype,
+                                                apply_state)
+    apply_state['weight_decay_rate'] = tf.constant(
+        self.weight_decay_rate, name='adam_weight_decay_rate')
+
+  def _decay_weights_op(self, var, learning_rate, apply_state):
+    do_decay = self._do_use_weight_decay(var.name)
+    if do_decay:
+      return var.assign_sub(
+          learning_rate * var *
+          apply_state['weight_decay_rate'],
+          use_locking=self._use_locking)
+    return tf.no_op()
+
+  def apply_gradients(self, grads_and_vars, clip_norm, name=None):
+    grads, tvars = list(zip(*grads_and_vars))
+    (grads, _) = tf.clip_by_global_norm(grads, clip_norm=clip_norm)
+    return super(AdamWeightDecay, self).apply_gradients(zip(grads, tvars))
+
+  def _get_lr(self, var_device, var_dtype, apply_state):
+    """Retrieves the learning rate with the given state."""
+    if apply_state is None:
+      return self._decayed_lr_t[var_dtype], {}
+
+    apply_state = apply_state or {}
+    coefficients = apply_state.get((var_device, var_dtype))
+    if coefficients is None:
+      coefficients = self._fallback_apply_state(var_device, var_dtype)
+      apply_state[(var_device, var_dtype)] = coefficients
+
+    return coefficients['lr_t'], dict(apply_state=apply_state)
+
+  def _resource_apply_dense(self, grad, var, apply_state=None):
+    lr_t, kwargs = self._get_lr(var.device, var.dtype.base_dtype, apply_state)
+    decay = self._decay_weights_op(var, lr_t, apply_state)
+    with tf.control_dependencies([decay]):
+      return super(AdamWeightDecay, self)._resource_apply_dense(
+          grad, var, **kwargs)
+
+  def _resource_apply_sparse(self, grad, var, indices, apply_state=None):
+    lr_t, kwargs = self._get_lr(var.device, var.dtype.base_dtype, apply_state)
+    decay = self._decay_weights_op(var, lr_t, apply_state)
+    with tf.control_dependencies([decay]):
+      return super(AdamWeightDecay, self)._resource_apply_sparse(
+          grad, var, indices, **kwargs)
+
+  def get_config(self):
+    config = super(AdamWeightDecay, self).get_config()
+    config.update({
+        'weight_decay_rate': self.weight_decay_rate,
+    })
+    return config
+
+  def _do_use_weight_decay(self, param_name):
+    """Whether to use L2 weight decay for `param_name`."""
+    if self.weight_decay_rate == 0:
+      return False
+
+    if self._include_in_weight_decay:
+      for r in self._include_in_weight_decay:
+        if re.search(r, param_name) is not None:
+          return True
+
+    if self._exclude_from_weight_decay:
+      for r in self._exclude_from_weight_decay:
+        if re.search(r, param_name) is not None:
+          return False
+    return True
+
+
+## Inspired from https://github.com/OpenNMT/OpenNMT-tf/blob/master/opennmt/optimizers/utils.py
+class GradientAccumulator(object):
+    """Distribution strategies-aware gradient accumulation utility."""
+
+    def __init__(self):
+        """Initializes the accumulator."""
+        self._gradients = []
+        self._accum_steps = tf.Variable(
+            initial_value=0,
+            dtype=tf.int64,
+            trainable=False,
+            aggregation=tf.VariableAggregation.ONLY_FIRST_REPLICA)
+
+    @property
+    def step(self):
+        """Number of accumulated steps."""
+        return self._accum_steps.value()
+
+    @property
+    def gradients(self):
+        """The accumulated gradients."""
+        return list(gradient.value() if gradient is not None else gradient for gradient in self._get_replica_gradients())
+
+    def __call__(self, gradients):
+        """Accumulates :obj:`gradients`."""
+        if not self._gradients:
+            self._gradients.extend([tf.Variable(tf.zeros_like(gradient), trainable=False) if gradient is not None else gradient for gradient in gradients])
+
+        if len(gradients) != len(self._gradients):
+            raise ValueError("Expected %s gradients, but got %d" % (len(self._gradients), len(gradients)))
+
+        for accum_gradient, gradient in zip(self._get_replica_gradients(), gradients):
+            if accum_gradient is not None:
+                accum_gradient.assign_add(gradient)
+
+        self._accum_steps.assign_add(1)
+
+    def reset(self):
+        """Resets the accumulated gradients."""
+        if self._gradients:
+            self._accum_steps.assign(0)
+
+        for gradient in self._get_replica_gradients():
+            if gradient is not None:
+                gradient.assign(tf.zeros_like(gradient))
+
+    def _get_replica_gradients(self):
+        if tf.distribute.has_strategy():
+            # In a replica context, we want to accumulate gradients on each replica
+            # without synchronization, so we directly assign the value of the
+            # current replica.
+            replica_context = tf.distribute.get_replica_context()
+
+            if replica_context is None or tf.distribute.get_strategy().num_replicas_in_sync == 1:
+                return self._gradients
+
+            return (gradient.device_map.select_for_current_replica(gradient.values, replica_context) for gradient in self._gradients)
+        else:
+            return self._gradients

From 254ebb979c09d2e8f7efeb11d46bd1196f856699 Mon Sep 17 00:00:00 2001
From: Julien Plu <julien.plu@schibsted.com>
Date: Wed, 4 Dec 2019 10:00:25 +0100
Subject: [PATCH 215/505] Bugfix on init file. Missing comma.

---
 transformers/__init__.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/__init__.py b/transformers/__init__.py
index 2f74b7e79c..e4f5984c70 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -162,7 +162,7 @@ if is_tf_available():
     from .modeling_tf_distilbert import (TFDistilBertPreTrainedModel, TFDistilBertMainLayer,
                                          TFDistilBertModel, TFDistilBertForMaskedLM,
                                          TFDistilBertForSequenceClassification,
-					 TFDistilBertForTokenClassification
+					 TFDistilBertForTokenClassification,
                                          TFDistilBertForQuestionAnswering,
                                          TF_DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
 

From 5bfcd0485ece086ebcbed2d008813037968a9e58 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Wed, 4 Dec 2019 14:53:11 +0100
Subject: [PATCH 216/505] fix #1991

---
 examples/run_lm_finetuning.py | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/examples/run_lm_finetuning.py b/examples/run_lm_finetuning.py
index 0bb7460353..a5eaf524ac 100644
--- a/examples/run_lm_finetuning.py
+++ b/examples/run_lm_finetuning.py
@@ -217,7 +217,10 @@ def train(args, train_dataset, model, tokenizer):
 
     global_step = 0
     tr_loss, logging_loss = 0.0, 0.0
-    model.resize_token_embeddings(len(tokenizer))
+
+    model_to_resize = model.module if hasattr(model, 'module') else model  # Take care of distributed/parallel training
+    model_to_resize.resize_token_embeddings(len(tokenizer))
+
     model.zero_grad()
     train_iterator = trange(int(args.num_train_epochs), desc="Epoch", disable=args.local_rank not in [-1, 0])
     set_seed(args)  # Added here for reproducibility (even between python 2 and 3)

From 9ddc3f1a1227fc9cbe4e5a5c20b21546e438dfb1 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Wed, 4 Dec 2019 10:37:00 -0500
Subject: [PATCH 217/505] Naming update + XLNet/XLM evaluation

---
 examples/run_squad.py                      |  6 +-
 transformers/data/metrics/squad_metrics.py | 97 ++++++++++++++++++----
 2 files changed, 85 insertions(+), 18 deletions(-)

diff --git a/examples/run_squad.py b/examples/run_squad.py
index b7952487dc..a9ef5c6ba2 100644
--- a/examples/run_squad.py
+++ b/examples/run_squad.py
@@ -17,7 +17,7 @@
 
 from __future__ import absolute_import, division, print_function
 from transformers.data.processors.squad import SquadV1Processor, SquadV2Processor, SquadResult
-from transformers.data.metrics.squad_metrics import compute_predictions, compute_predictions_extended, squad_evaluate
+from transformers.data.metrics.squad_metrics import compute_predictions_logits, compute_predictions_log_probs, squad_evaluate
 
 import argparse
 import logging
@@ -264,13 +264,13 @@ def evaluate(args, model, tokenizer, prefix=""):
 
     if args.model_type in ['xlnet', 'xlm']:
         # XLNet uses a more complex post-processing procedure
-        predictions = compute_predictions_extended(examples, features, all_results, args.n_best_size,
+        predictions = compute_predictions_log_probs(examples, features, all_results, args.n_best_size,
                         args.max_answer_length, output_prediction_file,
                         output_nbest_file, output_null_log_odds_file, args.predict_file,
                         model.config.start_n_top, model.config.end_n_top,
                         args.version_2_with_negative, tokenizer, args.verbose_logging)
     else:
-        predictions = compute_predictions(examples, features, all_results, args.n_best_size,
+        predictions = compute_predictions_logits(examples, features, all_results, args.n_best_size,
                         args.max_answer_length, args.do_lower_case, output_prediction_file,
                         output_nbest_file, output_null_log_odds_file, args.verbose_logging,
                         args.version_2_with_negative, args.null_score_diff_threshold)
diff --git a/transformers/data/metrics/squad_metrics.py b/transformers/data/metrics/squad_metrics.py
index 83647a20d0..1f120d354a 100644
--- a/transformers/data/metrics/squad_metrics.py
+++ b/transformers/data/metrics/squad_metrics.py
@@ -125,6 +125,53 @@ def merge_eval(main_eval, new_eval, prefix):
         main_eval['%s_%s' % (prefix, k)] = new_eval[k]
 
 
+def find_best_thresh_v2(preds, scores, na_probs, qid_to_has_ans):
+    num_no_ans = sum(1 for k in qid_to_has_ans if not qid_to_has_ans[k])
+    cur_score = num_no_ans
+    best_score = cur_score
+    best_thresh = 0.0
+    qid_list = sorted(na_probs, key=lambda k: na_probs[k])
+    for i, qid in enumerate(qid_list):
+        if qid not in scores:
+            continue
+        if qid_to_has_ans[qid]:
+            diff = scores[qid]
+        else:
+            if preds[qid]:
+                diff = -1
+            else:
+                diff = 0
+        cur_score += diff
+        if cur_score > best_score:
+            best_score = cur_score
+            best_thresh = na_probs[qid]
+
+    has_ans_score, has_ans_cnt = 0, 0
+    for qid in qid_list:
+        if not qid_to_has_ans[qid]:
+            continue
+        has_ans_cnt += 1
+
+        if qid not in scores:
+            continue
+        has_ans_score += scores[qid]
+
+    return 100.0 * best_score / len(scores), best_thresh, 1.0 * has_ans_score / has_ans_cnt
+
+
+def find_all_best_thresh_v2(main_eval, preds, exact_raw, f1_raw, na_probs, qid_to_has_ans):
+    best_exact, exact_thresh, has_ans_exact = find_best_thresh_v2(
+        preds, exact_raw, na_probs, qid_to_has_ans)
+    best_f1, f1_thresh, has_ans_f1 = find_best_thresh_v2(
+        preds, f1_raw, na_probs, qid_to_has_ans)
+    main_eval['best_exact'] = best_exact
+    main_eval['best_exact_thresh'] = exact_thresh
+    main_eval['best_f1'] = best_f1
+    main_eval['best_f1_thresh'] = f1_thresh
+    main_eval['has_ans_exact'] = has_ans_exact
+    main_eval['has_ans_f1'] = has_ans_f1
+
+
 def find_best_thresh(preds, scores, na_probs, qid_to_has_ans):
     num_no_ans = sum(1 for k in qid_to_has_ans if not qid_to_has_ans[k])
     cur_score = num_no_ans
@@ -318,10 +365,20 @@ def _compute_softmax(scores):
     return probs
 
 
-def compute_predictions(all_examples, all_features, all_results, n_best_size,
-                      max_answer_length, do_lower_case, output_prediction_file,
-                      output_nbest_file, output_null_log_odds_file, verbose_logging,
-                      version_2_with_negative, null_score_diff_threshold):
+def compute_predictions_logits(
+    all_examples,
+    all_features,
+    all_results,
+    n_best_size,
+    max_answer_length,
+    do_lower_case,
+    output_prediction_file,
+    output_nbest_file,
+    output_null_log_odds_file,
+    verbose_logging,
+    version_2_with_negative,
+    null_score_diff_threshold
+):
     """Write final predictions to the json file and log-odds of null if needed."""
     logger.info("Writing predictions to: %s" % (output_prediction_file))
     logger.info("Writing nbest to: %s" % (output_nbest_file))
@@ -450,12 +507,12 @@ def compute_predictions(all_examples, all_features, all_results, n_best_size,
                         text="",
                         start_logit=null_start_logit,
                         end_logit=null_end_logit))
-                
+
             # In very rare edge cases we could only have single null prediction.
             # So we just create a nonce prediction in this case to avoid failure.
-            if len(nbest)==1:
+            if len(nbest) == 1:
                 nbest.insert(0,
-                    _NbestPrediction(text="empty", start_logit=0.0, end_logit=0.0))
+                             _NbestPrediction(text="empty", start_logit=0.0, end_logit=0.0))
 
         # In very rare edge cases we could have no valid predictions. So we
         # just create a nonce prediction in this case to avoid failure.
@@ -512,12 +569,22 @@ def compute_predictions(all_examples, all_features, all_results, n_best_size,
     return all_predictions
 
 
-def compute_predictions_extended(all_examples, all_features, all_results, n_best_size,
-                                max_answer_length, output_prediction_file,
-                                output_nbest_file,
-                                output_null_log_odds_file, orig_data_file,
-                                start_n_top, end_n_top, version_2_with_negative,
-                                tokenizer, verbose_logging):
+def compute_predictions_log_probs(
+    all_examples,
+    all_features,
+    all_results,
+    n_best_size,
+    max_answer_length,
+    output_prediction_file,
+    output_nbest_file,
+    output_null_log_odds_file,
+    orig_data_file,
+    start_n_top,
+    end_n_top,
+    version_2_with_negative,
+    tokenizer,
+    verbose_logging
+):
     """ XLNet write prediction logic (more complex than Bert's).
         Write final predictions to the json file and log-odds of null if needed.
 
@@ -526,7 +593,7 @@ def compute_predictions_extended(all_examples, all_features, all_results, n_best
     _PrelimPrediction = collections.namedtuple(  # pylint: disable=invalid-name
         "PrelimPrediction",
         ["feature_index", "start_index", "end_index",
-        "start_log_prob", "end_log_prob"])
+         "start_log_prob", "end_log_prob"])
 
     _NbestPrediction = collections.namedtuple(  # pylint: disable=invalid-name
         "NbestPrediction", ["text", "start_log_prob", "end_log_prob"])
@@ -609,7 +676,7 @@ def compute_predictions_extended(all_examples, all_features, all_results, n_best
 
             # XLNet un-tokenizer
             # Let's keep it simple for now and see if we need all this later.
-            # 
+            #
             # tok_start_to_orig_index = feature.tok_start_to_orig_index
             # tok_end_to_orig_index = feature.tok_end_to_orig_index
             # start_orig_pos = tok_start_to_orig_index[pred.start_index]

From ff98b041da4b992a87d8b6258b30e47310ec8430 Mon Sep 17 00:00:00 2001
From: Julien Plu <julien.plu@schibsted.com>
Date: Wed, 4 Dec 2019 16:53:06 +0100
Subject: [PATCH 218/505] Fix whitespace issue

---
 transformers/__init__.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/__init__.py b/transformers/__init__.py
index e4f5984c70..6d18f11722 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -162,7 +162,7 @@ if is_tf_available():
     from .modeling_tf_distilbert import (TFDistilBertPreTrainedModel, TFDistilBertMainLayer,
                                          TFDistilBertModel, TFDistilBertForMaskedLM,
                                          TFDistilBertForSequenceClassification,
-					 TFDistilBertForTokenClassification,
+                                         TFDistilBertForTokenClassification,
                                          TFDistilBertForQuestionAnswering,
                                          TF_DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
 

From bf119c0568dfc1ea5ce0a34359e33ca002266e96 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Wed, 4 Dec 2019 11:34:59 -0500
Subject: [PATCH 219/505] TFDS dataset can now be evaluated

---
 transformers/data/processors/squad.py | 34 ++++++++++++++++++++-------
 1 file changed, 25 insertions(+), 9 deletions(-)

diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index 70dc9faf54..2e50ac8a8c 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -245,22 +245,37 @@ class SquadProcessor(DataProcessor):
     train_file = None
     dev_file = None
 
-    def get_example_from_tensor_dict(self, tensor_dict):
+    def get_example_from_tensor_dict(self, tensor_dict, evaluate=False):
+
+        if not evaluate:
+            answer = tensor_dict['answers']['text'][0].numpy().decode('utf-8')
+            answer_start = tensor_dict['answers']['answer_start'][0].numpy()
+            answers = None
+        else:
+            answers = [{
+                "answer_start": start.numpy(), 
+                "text": text.numpy().decode('utf-8')
+            } for start, text in zip(tensor_dict['answers']["answer_start"], tensor_dict['answers']["text"])]
+
+            answer = None
+            answer_start = None
+
         return SquadExample(
-            tensor_dict['id'].numpy().decode("utf-8"),
-            tensor_dict['question'].numpy().decode('utf-8'),
-            tensor_dict['context'].numpy().decode('utf-8'),
-            tensor_dict['answers']['text'][0].numpy().decode('utf-8'),
-            tensor_dict['answers']['answer_start'][0].numpy(),
-            tensor_dict['title'].numpy().decode('utf-8')
+            qas_id=tensor_dict['id'].numpy().decode("utf-8"),
+            question_text=tensor_dict['question'].numpy().decode('utf-8'),
+            context_text=tensor_dict['context'].numpy().decode('utf-8'),
+            answer_text=answer,
+            start_position_character=answer_start,
+            title=tensor_dict['title'].numpy().decode('utf-8'),
+            answers=answers
         )
 
-    def get_examples_from_dataset(self, dataset):
+    def get_examples_from_dataset(self, dataset, evaluate=False):
         """See base class."""
 
         examples = []
         for tensor_dict in tqdm(dataset):
-            examples.append(self.get_example_from_tensor_dict(tensor_dict)) 
+            examples.append(self.get_example_from_tensor_dict(tensor_dict, evaluate=evaluate)) 
 
         return examples
 
@@ -300,6 +315,7 @@ class SquadProcessor(DataProcessor):
                     question_text = qa["question"]
                     start_position_character = None
                     answer_text = None
+                    answers = None
                     
                     if "is_impossible" in qa:
                         is_impossible = qa["is_impossible"]

From cca75e788485e8a2a1c44a445c6aba0fb2dfaf56 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Wed, 4 Dec 2019 15:42:29 -0500
Subject: [PATCH 220/505] Kill the demon spawn

---
 examples/run_squad.py                 | 23 +++++++-
 transformers/data/processors/squad.py | 75 +++++----------------------
 2 files changed, 34 insertions(+), 64 deletions(-)

diff --git a/examples/run_squad.py b/examples/run_squad.py
index a9ef5c6ba2..2f86322196 100644
--- a/examples/run_squad.py
+++ b/examples/run_squad.py
@@ -248,7 +248,28 @@ def evaluate(args, model, tokenizer, prefix=""):
             eval_feature = features[example_index.item()]
             unique_id = int(eval_feature.unique_id)
 
-            result = SquadResult([to_list(output[i]) for output in outputs] + [unique_id])
+            output = [to_list(output[i]) for output in outputs]
+
+            if len(output) >= 5:
+                start_logits = output[0]
+                start_top_index = output[1]
+                end_logits = output[2]
+                end_top_index = output[3],
+                cls_logits = output[4]
+
+                result = SquadResult(
+                    unique_id, start_logits, end_logits, 
+                    start_top_index=start_top_index, 
+                    end_top_index=end_top_index, 
+                    cls_logits=cls_logits
+                )
+
+            else:
+                start_logits, end_logits = output
+                result = SquadResult(
+                    unique_id, start_logits, end_logits
+                )
+
             all_results.append(result)
 
     evalTime = timeit.default_timer() - start_time
diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index 2e50ac8a8c..9306189eb4 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -446,72 +446,21 @@ class SquadFeatures(object):
         self.end_position = end_position
 
 
-
 class SquadResult(object):
     """
     Constructs a SquadResult which can be used to evaluate a model's output on the SQuAD dataset.
 
     Args:
-        result: The result output by a model on a SQuAD inference. These results may be complex (5 values) as the ones output by
-            XLNet or XLM or may be simple like the other models (2 values). They may be passed as a list or as a dict, with the 
-            following accepted formats:
-
-            `dict` output by a simple model:
-                {
-                    "start_logits": int,
-                    "end_logits": int,
-                    "unique_id": string
-                }
-            `list` output by a simple model:
-                [start_logits, end_logits, unique_id]
-
-            `dict` output by a complex model:
-                {
-                    "start_top_log_probs": float,
-                    "start_top_index": int,
-                    "end_top_log_probs": float,
-                    "end_top_index": int,
-                    "cls_logits": int,
-                    "unique_id": string
-                }
-            `list` output by a complex model:
-                [start_top_log_probs, start_top_index, end_top_log_probs, end_top_index, cls_logits, unique_id]
-
-            See `run_squad.py` for an example.
+        unique_id: The unique identifier corresponding to that example.
+        start_logits: The logits corresponding to the start of the answer
+        end_logits: The logits corresponding to the end of the answer
     """
-    def __init__(self, result):
-        if isinstance(result, dict):
-            if "start_logits" in result and "end_logits" in result:
-                self.start_logits = result["start_logits"]
-                self.end_logits = result["end_logits"]
-
-            elif "start_top_log_probs" in result and "start_top_index" in result:
-                self.start_top_log_probs  = result["start_top_log_probs"]
-                self.start_top_index      = result["start_top_index"]
-                self.end_top_log_probs    = result["end_top_log_probs"]
-                self.end_top_index        = result["end_top_index"]
-                self.cls_logits           = result["cls_logits"]
-
-            else:
-                raise ValueError("SquadResult instantiated with wrong values.")
-
-            self.unique_id = result["unique_id"]
-        elif isinstance(result, list):            
-            if len(result) == 3:
-                self.start_logits = result[0]
-                self.end_logits = result[1]
-
-            elif len(result) == 6:
-                self.start_top_log_probs  = result[0]
-                self.start_top_index      = result[1]
-                self.end_top_log_probs    = result[2]
-                self.end_top_index        = result[3]
-                self.cls_logits           = result[4]
-
-            else:
-                raise ValueError("SquadResult instantiated with wrong values.")
-
-            self.unique_id = result[-1]
-
-        else:
-            raise ValueError("SquadResult instantiated with wrong values. Should be a dictionary or a list.")
+    def __init__(self, unique_id, start_logits, end_logits, start_top_index=None, end_top_index=None, cls_logits=None):
+        self.start_top_log_probs = start_logits
+        self.end_top_log_probs = end_logits
+        self.unique_id = unique_id
+        
+        if start_top_index:
+            self.start_top_index = start_top_index
+            self.end_top_index = end_top_index
+            self.cls_logits = cls_logits
\ No newline at end of file

From a7ca6d738b7801c680bd25d9e910f962d3f8bf2d Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Wed, 4 Dec 2019 15:43:34 -0500
Subject: [PATCH 221/505] Padding side is  tokenizer-dependant

---
 transformers/data/processors/squad.py         | 11 ++--
 .../tests/tokenization_tests_commons.py       | 21 +++++--
 transformers/tokenization_utils.py            | 60 ++++++++++++-------
 transformers/tokenization_xlnet.py            |  1 +
 4 files changed, 58 insertions(+), 35 deletions(-)

diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index 9306189eb4..6599c54330 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -73,8 +73,7 @@ def _is_whitespace(c):
     return False
 
 def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
-                                       doc_stride, max_query_length, is_training,
-                                       sequence_a_is_doc=False):
+                                       doc_stride, max_query_length, is_training):
     """Loads a data file into a list of `InputBatch`s."""
 
     # Defining helper methods    
@@ -127,13 +126,13 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
         while len(spans) * doc_stride < len(all_doc_tokens):
             
             encoded_dict = tokenizer.encode_plus(
-                truncated_query if not sequence_a_is_doc else span_doc_tokens, 
-                span_doc_tokens if not sequence_a_is_doc else truncated_query, 
+                truncated_query if tokenizer.padding_side == "right" else span_doc_tokens, 
+                span_doc_tokens if tokenizer.padding_side == "right" else truncated_query, 
                 max_length=max_seq_length, 
                 return_overflowing_tokens=True, 
-                padding_strategy='right',
+                pad_to_max_length=True,
                 stride=max_seq_length - doc_stride - len(truncated_query) - sequence_pair_added_tokens,
-                truncation_strategy='only_second' if not sequence_a_is_doc else 'only_first'
+                truncation_strategy='only_second' if tokenizer.padding_side == "right" else 'only_first'
             )
 
             paragraph_len = min(len(all_doc_tokens) - len(spans) * doc_stride, max_seq_length - len(truncated_query) - sequence_pair_added_tokens)
diff --git a/transformers/tests/tokenization_tests_commons.py b/transformers/tests/tokenization_tests_commons.py
index 40d68d0ab2..6592005c67 100644
--- a/transformers/tests/tokenization_tests_commons.py
+++ b/transformers/tests/tokenization_tests_commons.py
@@ -344,17 +344,19 @@ class CommonTestCases:
             padding_idx = tokenizer.pad_token_id
 
             # RIGHT PADDING - Check that it correctly pads when a maximum length is specified along with the padding flag set to True
+            tokenizer.padding_side = "right"
             encoded_sequence = tokenizer.encode(sequence)
             sequence_length = len(encoded_sequence)
-            padded_sequence = tokenizer.encode(sequence, max_length=sequence_length + padding_size, padding_strategy='right')
+            padded_sequence = tokenizer.encode(sequence, max_length=sequence_length + padding_size, pad_to_max_length=True)
             padded_sequence_length = len(padded_sequence)
             assert sequence_length + padding_size == padded_sequence_length
             assert encoded_sequence + [padding_idx] * padding_size == padded_sequence
 
             # LEFT PADDING - Check that it correctly pads when a maximum length is specified along with the padding flag set to True
+            tokenizer.padding_side = "left"
             encoded_sequence = tokenizer.encode(sequence)
             sequence_length = len(encoded_sequence)
-            padded_sequence = tokenizer.encode(sequence, max_length=sequence_length + padding_size, padding_strategy='left')
+            padded_sequence = tokenizer.encode(sequence, max_length=sequence_length + padding_size, pad_to_max_length=True)
             padded_sequence_length = len(padded_sequence)
             assert sequence_length + padding_size == padded_sequence_length
             assert [padding_idx] * padding_size + encoded_sequence == padded_sequence
@@ -362,10 +364,15 @@ class CommonTestCases:
             # RIGHT & LEFT PADDING - Check that nothing is done when a maximum length is not specified
             encoded_sequence = tokenizer.encode(sequence)
             sequence_length = len(encoded_sequence)
-            padded_sequence_right = tokenizer.encode(sequence, padding_strategy='right')
+
+            tokenizer.padding_side = "right"
+            padded_sequence_right = tokenizer.encode(sequence, pad_to_max_length=True)
             padded_sequence_right_length = len(padded_sequence_right)
-            padded_sequence_left = tokenizer.encode(sequence, padding_strategy='left')
+
+            tokenizer.padding_side = "left"
+            padded_sequence_left = tokenizer.encode(sequence, pad_to_max_length=True)
             padded_sequence_left_length = len(padded_sequence_left)
+
             assert sequence_length == padded_sequence_right_length
             assert encoded_sequence == padded_sequence_right
             assert sequence_length == padded_sequence_left_length
@@ -387,7 +394,8 @@ class CommonTestCases:
             sequence_length = len(input_ids)
 
             # Test right padding
-            padded_sequence = tokenizer.encode_plus(sequence, max_length=sequence_length + padding_size, padding_strategy='right', return_special_tokens_mask=True)
+            tokenizer.padding_side = "right"
+            padded_sequence = tokenizer.encode_plus(sequence, max_length=sequence_length + padding_size, pad_to_max_length=True, return_special_tokens_mask=True)
             padded_input_ids = padded_sequence['input_ids']
             padded_token_type_ids = padded_sequence['token_type_ids']
             padded_attention_mask = padded_sequence['attention_mask']
@@ -401,7 +409,8 @@ class CommonTestCases:
             assert special_tokens_mask + [1] * padding_size == padded_special_tokens_mask 
 
             # Test left padding
-            padded_sequence = tokenizer.encode_plus(sequence, max_length=sequence_length + padding_size, padding_strategy='left', return_special_tokens_mask=True)
+            tokenizer.padding_side = "left"
+            padded_sequence = tokenizer.encode_plus(sequence, max_length=sequence_length + padding_size, pad_to_max_length=True, return_special_tokens_mask=True)
             padded_input_ids = padded_sequence['input_ids']
             padded_token_type_ids = padded_sequence['token_type_ids']
             padded_attention_mask = padded_sequence['attention_mask']
diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index dbbabd0e1a..41a611ea49 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -77,6 +77,8 @@ class PreTrainedTokenizer(object):
                                  "pad_token", "cls_token", "mask_token",
                                  "additional_special_tokens"]
 
+    padding_side = "right"
+
     @property
     def bos_token(self):
         """ Beginning of sentence token (string). Log an error if used while not having been set. """
@@ -223,6 +225,9 @@ class PreTrainedTokenizer(object):
 
         self.max_len = max_len if max_len is not None else int(1e12)
 
+        # Padding side is right by default and over-riden in subclsses. If specified in the kwargs, it is changed.
+        self.padding_side = kwargs.pop('padding_side', self.padding_side)
+        
         # Added tokens
         self.added_tokens_encoder = {}
         self.added_tokens_decoder = {}
@@ -702,7 +707,7 @@ class PreTrainedTokenizer(object):
                max_length=None,
                stride=0,
                truncation_strategy='longest_first',
-               padding_strategy=None,
+               pad_to_max_length=False,
                return_tensors=None,
                **kwargs):
         """
@@ -729,12 +734,12 @@ class PreTrainedTokenizer(object):
                 - 'only_first': Only truncate the first sequence
                 - 'only_second': Only truncate the second sequence
                 - 'do_not_truncate': Does not truncate (raise an error if the input sequence is longer than max_length)
-            padding_strategy: if set to a strategy, the returned sequences will be padded according to the model's 
-                padding index, up to their max length. If no max length is specified, no padding is done.
-                The strategies are handled by the following strings:
+            pad_to_max_length: if set to True, the returned sequences will be padded according to the model's padding side and
+                padding index, up to their max length. If no max length is specified, the padding is done up to the model's max length.
+                The tokenizer padding sides are handled by the following strings:
                 - 'left': pads on the left of the sequences
                 - 'right': pads on the right of the sequences   
-                Defaults to None: no padding.
+                Defaults to False: no padding.
             return_tensors: (optional) can be set to 'tf' or 'pt' to return respectively TensorFlow tf.constant
                 or PyTorch torch.Tensor instead of a list of python integers.
             **kwargs: passed to the `self.tokenize()` method
@@ -745,7 +750,7 @@ class PreTrainedTokenizer(object):
                                           add_special_tokens=add_special_tokens,
                                           stride=stride,
                                           truncation_strategy=truncation_strategy,
-                                          padding_strategy=padding_strategy,
+                                          pad_to_max_length=pad_to_max_length,
                                           return_tensors=return_tensors,
                                           **kwargs)
 
@@ -758,7 +763,7 @@ class PreTrainedTokenizer(object):
                     max_length=None,
                     stride=0,
                     truncation_strategy='longest_first',
-                    padding_strategy=None,
+                    pad_to_max_length=False,
                     return_tensors=None,
                     return_token_type_ids=True,
                     return_attention_mask=True,
@@ -788,12 +793,12 @@ class PreTrainedTokenizer(object):
                 - 'only_first': Only truncate the first sequence
                 - 'only_second': Only truncate the second sequence
                 - 'do_not_truncate': Does not truncate (raise an error if the input sequence is longer than max_length)
-            padding_strategy: if set to a strategy, the returned sequences will be padded according to the model's 
-                padding index, up to their max length. If no max length is specified, no padding is done.
-                The strategies are handled by the following strings:
+            pad_to_max_length: if set to True, the returned sequences will be padded according to the model's padding side and
+                padding index, up to their max length. If no max length is specified, the padding is done up to the model's max length.
+                The tokenizer padding sides are handled by the following strings:
                 - 'left': pads on the left of the sequences
                 - 'right': pads on the right of the sequences   
-                Defaults to None: no padding.
+                Defaults to False: no padding.
             return_tensors: (optional) can be set to 'tf' or 'pt' to return respectively TensorFlow tf.constant
                 or PyTorch torch.Tensor instead of a list of python integers.
             return_token_type_ids: (optional) Set to False to avoid returning token_type_ids (default True).
@@ -841,7 +846,7 @@ class PreTrainedTokenizer(object):
         return self.prepare_for_model(first_ids,
                                       pair_ids=second_ids,
                                       max_length=max_length,
-                                      padding_strategy=padding_strategy,
+                                      pad_to_max_length=pad_to_max_length,
                                       add_special_tokens=add_special_tokens,
                                       stride=stride,
                                       truncation_strategy=truncation_strategy,
@@ -853,7 +858,7 @@ class PreTrainedTokenizer(object):
 
     def prepare_for_model(self, ids, pair_ids=None, max_length=None, add_special_tokens=True, stride=0,
                           truncation_strategy='longest_first',
-                          padding_strategy=None,
+                          pad_to_max_length=False,
                           return_tensors=None,
                           return_token_type_ids=True,
                           return_attention_mask=True,
@@ -881,12 +886,12 @@ class PreTrainedTokenizer(object):
                 - 'only_first': Only truncate the first sequence
                 - 'only_second': Only truncate the second sequence
                 - 'do_not_truncate': Does not truncate (raise an error if the input sequence is longer than max_length)
-            padding_strategy: if set to a strategy, the returned sequences will be padded according to the model's 
-                padding index, up to their max length. If no max length is specified, no padding is done.
-                The strategies are handled by the following strings:
+            pad_to_max_length: if set to True, the returned sequences will be padded according to the model's padding side and
+                padding index, up to their max length. If no max length is specified, the padding is done up to the model's max length.
+                The tokenizer padding sides are handled by the following strings:
                 - 'left': pads on the left of the sequences
-                - 'right': pads on the right of the sequences            
-                Defaults to None: no padding.
+                - 'right': pads on the right of the sequences   
+                Defaults to False: no padding.
             return_tensors: (optional) can be set to 'tf' or 'pt' to return respectively TensorFlow tf.constant
                 or PyTorch torch.Tensor instead of a list of python integers.
             return_token_type_ids: (optional) Set to False to avoid returning token_type_ids (default True).
@@ -955,10 +960,19 @@ class PreTrainedTokenizer(object):
                            "for this model ({} > {}). Running this sequence through the model will result in "
                            "indexing errors".format(len(ids), self.max_len))
                            
-        if padding_strategy is not None and max_length and len(encoded_inputs["input_ids"]) < max_length:
-            difference = max_length - len(encoded_inputs["input_ids"])
+        needs_to_be_padded = pad_to_max_length and (
+            max_length and len(encoded_inputs["input_ids"]) < max_length
+            or 
+            max_length is None and len(encoded_inputs["input_ids"]) < self.max_len and self.max_len <= 10000
+        )
 
-            if padding_strategy == 'right':
+        if pad_to_max_length and max_length is None and self.max_len > 10000:
+            logger.warning("Sequence can't be padded as the maximum  ")
+
+        if needs_to_be_padded:
+            difference = (max_length if max_length is not None else self.max_len) - len(encoded_inputs["input_ids"])
+
+            if self.padding_side == 'right':
                 if return_attention_mask:
                     encoded_inputs["attention_mask"] = [1] * len(encoded_inputs["input_ids"]) + [0] * difference
                 if return_token_type_ids:
@@ -967,7 +981,7 @@ class PreTrainedTokenizer(object):
                     encoded_inputs["special_tokens_mask"] = encoded_inputs["special_tokens_mask"] + [1] * difference
                 encoded_inputs["input_ids"] = encoded_inputs["input_ids"] + [self.pad_token_id] * difference
 
-            elif padding_strategy == 'left':
+            elif self.padding_side == 'left':
                 if return_attention_mask:
                     encoded_inputs["attention_mask"] =  [0] * difference + [1] * len(encoded_inputs["input_ids"])
                 if return_token_type_ids:
@@ -977,7 +991,7 @@ class PreTrainedTokenizer(object):
                 encoded_inputs["input_ids"] = [self.pad_token_id] * difference + encoded_inputs["input_ids"]
 
             else:
-                raise ValueError("Invalid padding strategy:" + str(padding_strategy))
+                raise ValueError("Invalid padding strategy:" + str(self.padding_side))
             
         elif return_attention_mask:
             encoded_inputs["attention_mask"] = [1] * len(encoded_inputs["input_ids"])
diff --git a/transformers/tokenization_xlnet.py b/transformers/tokenization_xlnet.py
index 3ea71f4438..1c43c0943a 100644
--- a/transformers/tokenization_xlnet.py
+++ b/transformers/tokenization_xlnet.py
@@ -60,6 +60,7 @@ class XLNetTokenizer(PreTrainedTokenizer):
     vocab_files_names = VOCAB_FILES_NAMES
     pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
     max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
+    padding_side = "left"
 
     def __init__(self, vocab_file,
                  do_lower_case=False, remove_space=True, keep_accents=False,

From f7e4a7cdfa6bcf6ec7c33fd1d40d307278b1c13a Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Wed, 4 Dec 2019 16:24:15 -0500
Subject: [PATCH 222/505] Cleanup

---
 examples/run_squad.py                         |  32 ++--
 examples/test_examples.py                     |   3 +-
 .../{dev-v2.0-small.json => dev-v2.0.json}    |   0
 examples/tests_samples/SQUAD/train-v2.0.json  | 140 ++++++++++++++++++
 transformers/data/metrics/squad_metrics.py    |   4 +-
 transformers/data/processors/squad.py         |  36 ++++-
 6 files changed, 191 insertions(+), 24 deletions(-)
 rename examples/tests_samples/SQUAD/{dev-v2.0-small.json => dev-v2.0.json} (100%)
 create mode 100644 examples/tests_samples/SQUAD/train-v2.0.json

diff --git a/examples/run_squad.py b/examples/run_squad.py
index 2f86322196..3f1b6a798f 100644
--- a/examples/run_squad.py
+++ b/examples/run_squad.py
@@ -304,8 +304,8 @@ def load_and_cache_examples(args, tokenizer, evaluate=False, output_examples=Fal
         torch.distributed.barrier()  # Make sure only the first process in distributed training process the dataset, and the others will use the cache
 
     # Load data features from cache or dataset file
-    input_file = args.predict_file if evaluate else args.train_file
-    cached_features_file = os.path.join(os.path.dirname(input_file), 'cached_{}_{}_{}'.format(
+    input_dir = args.data_dir if args.data_dir else "."
+    cached_features_file = os.path.join(input_dir, 'cached_{}_{}_{}'.format(
         'dev' if evaluate else 'train',
         list(filter(None, args.model_name_or_path.split('/'))).pop(),
         str(args.max_seq_length)))
@@ -313,13 +313,22 @@ def load_and_cache_examples(args, tokenizer, evaluate=False, output_examples=Fal
         logger.info("Loading features from cached file %s", cached_features_file)
         features = torch.load(cached_features_file)
     else:
-        logger.info("Creating features from dataset file at %s", input_file)
+        logger.info("Creating features from dataset file at %s", input_dir)
 
-        processor = SquadV2Processor()
-        examples = processor.get_dev_examples("examples/squad", only_first=100) if evaluate else processor.get_train_examples("examples/squad")
-        # import tensorflow_datasets as tfds
-        # tfds_examples = tfds.load("squad")
-        # examples = SquadV1Processor().get_examples_from_dataset(tfds_examples["validation"])
+        if not args.data_dir:
+            try:
+                import tensorflow_datasets as tfds
+            except ImportError:
+                raise ImportError("If not data_dir is specified, tensorflow_datasets needs to be installed.")
+
+            if args.version_2_with_negative:
+                logger.warn("tensorflow_datasets does not handle version 2 of SQuAD.")
+
+            tfds_examples = tfds.load("squad")
+            examples = SquadV1Processor().get_examples_from_dataset(tfds_examples, evaluate=evaluate)
+        else:
+            processor = SquadV2Processor() if args.version_2_with_negative else SquadV1Processor()
+            examples = processor.get_dev_examples(args.data_dir) if evaluate else processor.get_train_examples(args.data_dir)
 
         features = squad_convert_examples_to_features( 
             examples=examples,
@@ -328,7 +337,6 @@ def load_and_cache_examples(args, tokenizer, evaluate=False, output_examples=Fal
             doc_stride=args.doc_stride,
             max_query_length=args.max_query_length,
             is_training=not evaluate,
-            sequence_a_is_doc=True if args.model_type in ['xlnet'] else False
         )
 
 
@@ -365,10 +373,6 @@ def main():
     parser = argparse.ArgumentParser()
 
     ## Required parameters
-    parser.add_argument("--train_file", default=None, type=str, required=True,
-                        help="SQuAD json for training. E.g., train-v1.1.json")
-    parser.add_argument("--predict_file", default=None, type=str, required=True,
-                        help="SQuAD json for predictions. E.g., dev-v1.1.json or test-v1.1.json")
     parser.add_argument("--model_type", default=None, type=str, required=True,
                         help="Model type selected in the list: " + ", ".join(MODEL_CLASSES.keys()))
     parser.add_argument("--model_name_or_path", default=None, type=str, required=True,
@@ -377,6 +381,8 @@ def main():
                         help="The output directory where the model checkpoints and predictions will be written.")
 
     ## Other parameters
+    parser.add_argument("--data_dir", default=None, type=str,
+                        help="The input data dir. Should contain the .json files for the task. If not specified, will run with tensorflow_datasets.")
     parser.add_argument("--config_name", default="", type=str,
                         help="Pretrained config name or path if not the same as model_name")
     parser.add_argument("--tokenizer_name", default="", type=str,
diff --git a/examples/test_examples.py b/examples/test_examples.py
index b04d722b7b..632d2f728e 100644
--- a/examples/test_examples.py
+++ b/examples/test_examples.py
@@ -72,8 +72,7 @@ class ExamplesTests(unittest.TestCase):
         logger.addHandler(stream_handler)
 
         testargs = ["run_squad.py",
-                    "--train_file=./examples/tests_samples/SQUAD/dev-v2.0-small.json",
-                    "--predict_file=./examples/tests_samples/SQUAD/dev-v2.0-small.json",
+                    "--data_dir=./examples/tests_samples/SQUAD",
                     "--model_name=bert-base-uncased",
                     "--output_dir=./examples/tests_samples/temp_dir",
                     "--max_steps=10",
diff --git a/examples/tests_samples/SQUAD/dev-v2.0-small.json b/examples/tests_samples/SQUAD/dev-v2.0.json
similarity index 100%
rename from examples/tests_samples/SQUAD/dev-v2.0-small.json
rename to examples/tests_samples/SQUAD/dev-v2.0.json
diff --git a/examples/tests_samples/SQUAD/train-v2.0.json b/examples/tests_samples/SQUAD/train-v2.0.json
new file mode 100644
index 0000000000..834d9ee660
--- /dev/null
+++ b/examples/tests_samples/SQUAD/train-v2.0.json
@@ -0,0 +1,140 @@
+{
+    "version": "v2.0",
+    "data": [{
+        "title": "Normans",
+        "paragraphs": [{
+            "qas": [{
+                "question": "In what country is Normandy located?",
+                "id": "56ddde6b9a695914005b9628",
+                "answers": [{
+                    "text": "France",
+                    "answer_start": 159
+                }],
+                "is_impossible": false
+            }, {
+                "question": "When were the Normans in Normandy?",
+                "id": "56ddde6b9a695914005b9629",
+                "answers": [{
+                    "text": "10th and 11th centuries",
+                    "answer_start": 94
+                }],
+                "is_impossible": false
+            }, {
+                "question": "From which countries did the Norse originate?",
+                "id": "56ddde6b9a695914005b962a",
+                "answers": [{
+                    "text": "Denmark, Iceland and Norway",
+                    "answer_start": 256
+                }],
+                "is_impossible": false
+            }, {
+                "plausible_answers": [{
+                    "text": "Rollo",
+                    "answer_start": 308
+                }],
+                "question": "Who did King Charles III swear fealty to?",
+                "id": "5ad39d53604f3c001a3fe8d3",
+                "answers": [],
+                "is_impossible": true
+            }, {
+                "plausible_answers": [{
+                    "text": "10th century",
+                    "answer_start": 671
+                }],
+                "question": "When did the Frankish identity emerge?",
+                "id": "5ad39d53604f3c001a3fe8d4",
+                "answers": [],
+                "is_impossible": true
+            }],
+            "context": "The Normans (Norman: Nourmands; French: Normands; Latin: Normanni) were the people who in the 10th and 11th centuries gave their name to Normandy, a region in France. They were descended from Norse (\"Norman\" comes from \"Norseman\") raiders and pirates from Denmark, Iceland and Norway who, under their leader Rollo, agreed to swear fealty to King Charles III of West Francia. Through generations of assimilation and mixing with the native Frankish and Roman-Gaulish populations, their descendants would gradually merge with the Carolingian-based cultures of West Francia. The distinct cultural and ethnic identity of the Normans emerged initially in the first half of the 10th century, and it continued to evolve over the succeeding centuries."
+        }, {
+            "qas": [{
+                "question": "Who was the duke in the battle of Hastings?",
+                "id": "56dddf4066d3e219004dad5f",
+                "answers": [{
+                    "text": "William the Conqueror",
+                    "answer_start": 1022
+                }],
+                "is_impossible": false
+            }, {
+                "plausible_answers": [{
+                    "text": "Antioch",
+                    "answer_start": 1295
+                }],
+                "question": "What principality did William the conquerer found?",
+                "id": "5ad3a266604f3c001a3fea2b",
+                "answers": [],
+                "is_impossible": true
+            }],
+            "context": "The Norman dynasty had a major political, cultural and military impact on medieval Europe and even the Near East. The Normans were famed for their martial spirit and eventually for their Christian piety, becoming exponents of the Catholic orthodoxy into which they assimilated. They adopted the Gallo-Romance language of the Frankish land they settled, their dialect becoming known as Norman, Normaund or Norman French, an important literary language. The Duchy of Normandy, which they formed by treaty with the French crown, was a great fief of medieval France, and under Richard I of Normandy was forged into a cohesive and formidable principality in feudal tenure. The Normans are noted both for their culture, such as their unique Romanesque architecture and musical traditions, and for their significant military accomplishments and innovations. Norman adventurers founded the Kingdom of Sicily under Roger II after conquering southern Italy on the Saracens and Byzantines, and an expedition on behalf of their duke, William the Conqueror, led to the Norman conquest of England at the Battle of Hastings in 1066. Norman cultural and military influence spread from these new European centres to the Crusader states of the Near East, where their prince Bohemond I founded the Principality of Antioch in the Levant, to Scotland and Wales in Great Britain, to Ireland, and to the coasts of north Africa and the Canary Islands."
+        }]
+    }, {
+        "title": "Computational_complexity_theory",
+        "paragraphs": [{
+            "qas": [{
+                "question": "What branch of theoretical computer science deals with broadly classifying computational problems by difficulty and class of relationship?",
+                "id": "56e16182e3433e1400422e28",
+                "answers": [{
+                    "text": "Computational complexity theory",
+                    "answer_start": 0
+                }],
+                "is_impossible": false
+            }, {
+                "plausible_answers": [{
+                    "text": "algorithm",
+                    "answer_start": 472
+                }],
+                "question": "What is a manual application of mathematical steps?",
+                "id": "5ad5316b5b96ef001a10ab76",
+                "answers": [],
+                "is_impossible": true
+            }],
+            "context": "Computational complexity theory is a branch of the theory of computation in theoretical computer science that focuses on classifying computational problems according to their inherent difficulty, and relating those classes to each other. A computational problem is understood to be a task that is in principle amenable to being solved by a computer, which is equivalent to stating that the problem may be solved by mechanical application of mathematical steps, such as an algorithm."
+        }, {
+            "qas": [{
+                "question": "What measure of a computational problem broadly defines the inherent difficulty of the solution?",
+                "id": "56e16839cd28a01900c67887",
+                "answers": [{
+                    "text": "if its solution requires significant resources",
+                    "answer_start": 46
+                }],
+                "is_impossible": false
+            }, {
+                "question": "What method is used to intuitively assess or quantify the amount of resources required to solve a computational problem?",
+                "id": "56e16839cd28a01900c67888",
+                "answers": [{
+                    "text": "mathematical models of computation",
+                    "answer_start": 176
+                }],
+                "is_impossible": false
+            }, {
+                "question": "What are two basic primary resources used to guage complexity?",
+                "id": "56e16839cd28a01900c67889",
+                "answers": [{
+                    "text": "time and storage",
+                    "answer_start": 305
+                }],
+                "is_impossible": false
+            }, {
+                "plausible_answers": [{
+                    "text": "the number of gates in a circuit",
+                    "answer_start": 436
+                }],
+                "question": "What unit is measured to determine circuit simplicity?",
+                "id": "5ad532575b96ef001a10ab7f",
+                "answers": [],
+                "is_impossible": true
+            }, {
+                "plausible_answers": [{
+                    "text": "the number of processors",
+                    "answer_start": 502
+                }],
+                "question": "What number is used in perpendicular computing?",
+                "id": "5ad532575b96ef001a10ab80",
+                "answers": [],
+                "is_impossible": true
+            }],
+            "context": "A problem is regarded as inherently difficult if its solution requires significant resources, whatever the algorithm used. The theory formalizes this intuition, by introducing mathematical models of computation to study these problems and quantifying the amount of resources needed to solve them, such as time and storage. Other complexity measures are also used, such as the amount of communication (used in communication complexity), the number of gates in a circuit (used in circuit complexity) and the number of processors (used in parallel computing). One of the roles of computational complexity theory is to determine the practical limits on what computers can and cannot do."
+        }]
+    }]
+}
\ No newline at end of file
diff --git a/transformers/data/metrics/squad_metrics.py b/transformers/data/metrics/squad_metrics.py
index 1f120d354a..f8449df045 100644
--- a/transformers/data/metrics/squad_metrics.py
+++ b/transformers/data/metrics/squad_metrics.py
@@ -630,12 +630,12 @@ def compute_predictions_log_probs(
 
             for i in range(start_n_top):
                 for j in range(end_n_top):
-                    start_log_prob = result.start_top_log_probs[i]
+                    start_log_prob = result.start_logits[i]
                     start_index = result.start_top_index[i]
 
                     j_index = i * end_n_top + j
 
-                    end_log_prob = result.end_top_log_probs[j_index]
+                    end_log_prob = result.end_logits[j_index]
                     end_index = result.end_top_index[j_index]
 
                     # We could hypothetically create invalid predictions, e.g., predict
diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index 6599c54330..dd2d9d25c0 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -146,7 +146,7 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
 
             token_to_orig_map = {}
             for i in range(paragraph_len):
-                index = len(truncated_query) + sequence_added_tokens + i if not sequence_a_is_doc else i 
+                index = len(truncated_query) + sequence_added_tokens + i if tokenizer.padding_side == "right" else i 
                 token_to_orig_map[index] = tok_to_orig_index[len(spans) * doc_stride + i]
 
             encoded_dict["paragraph_len"] = paragraph_len
@@ -166,7 +166,7 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
         for doc_span_index in range(len(spans)):
             for j in range(spans[doc_span_index]["paragraph_len"]):
                 is_max_context = _new_check_is_max_context(spans, doc_span_index, doc_span_index * doc_stride + j)
-                index = j if sequence_a_is_doc else spans[doc_span_index]["truncated_query_with_special_tokens_length"] + j
+                index = j if tokenizer.padding_side == "left" else spans[doc_span_index]["truncated_query_with_special_tokens_length"] + j
                 spans[doc_span_index]["token_is_max_context"][index] = is_max_context
 
         for span in spans:
@@ -179,7 +179,7 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
 
             p_mask = np.minimum(p_mask, 1)
 
-            if not sequence_a_is_doc:
+            if tokenizer.padding_side == "right":
                 # Limit positive values to one
                 p_mask = 1 - p_mask
 
@@ -207,7 +207,7 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
                     end_position = cls_index
                     span_is_impossible = True
                 else:
-                    if sequence_a_is_doc:
+                    if tokenizer.padding_side == "left":
                         doc_offset = 0
                     else:
                         doc_offset = len(truncated_query) + sequence_added_tokens
@@ -270,7 +270,29 @@ class SquadProcessor(DataProcessor):
         )
 
     def get_examples_from_dataset(self, dataset, evaluate=False):
-        """See base class."""
+        """
+        Creates a list of :class:`~transformers.data.processors.squad.SquadExample` using a TFDS dataset.
+
+        Args:
+            dataset: The tfds dataset loaded from `tensorflow_datasets.load("squad")`
+            evaluate: boolean specifying if in evaluation mode or in training mode
+
+        Returns:
+            List of SquadExample
+
+        Examples::
+
+            import tensorflow_datasets as tfds
+            dataset = tfds.load("squad")
+
+            training_examples = get_examples_from_dataset(dataset, evaluate=False)
+            evaluation_examples = get_examples_from_dataset(dataset, evaluate=True)
+        """
+
+        if evaluate:
+            dataset = dataset["validation"]
+        else:
+            dataset = dataset["train"]
 
         examples = []
         for tensor_dict in tqdm(dataset):
@@ -455,8 +477,8 @@ class SquadResult(object):
         end_logits: The logits corresponding to the end of the answer
     """
     def __init__(self, unique_id, start_logits, end_logits, start_top_index=None, end_top_index=None, cls_logits=None):
-        self.start_top_log_probs = start_logits
-        self.end_top_log_probs = end_logits
+        self.start_logits = start_logits
+        self.end_logits = end_logits
         self.unique_id = unique_id
         
         if start_top_index:

From 33508ae310f101a2534d3e97ea23fda93e25ef38 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Wed, 4 Dec 2019 16:26:45 -0500
Subject: [PATCH 223/505] Remove `only_first`

---
 transformers/data/processors/squad.py | 13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)

diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index dd2d9d25c0..09a79db471 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -300,29 +300,29 @@ class SquadProcessor(DataProcessor):
 
         return examples
 
-    def get_train_examples(self, data_dir, only_first=None):
+    def get_train_examples(self, data_dir):
         """See base class."""
         if self.train_file is None:
             raise ValueError("SquadProcessor should be instantiated via SquadV1Processor or SquadV2Processor")
 
         with open(os.path.join(data_dir, self.train_file), "r", encoding='utf-8') as reader:
             input_data = json.load(reader)["data"]
-        return self._create_examples(input_data, "train", only_first)
+        return self._create_examples(input_data, "train")
 
-    def get_dev_examples(self, data_dir, only_first=None):
+    def get_dev_examples(self, data_dir):
         """See base class."""
         if self.dev_file is None:
             raise ValueError("SquadProcessor should be instantiated via SquadV1Processor or SquadV2Processor")
         
         with open(os.path.join(data_dir, self.dev_file), "r", encoding='utf-8') as reader:
             input_data = json.load(reader)["data"]
-        return self._create_examples(input_data, "dev", only_first)
+        return self._create_examples(input_data, "dev")
 
     def get_labels(self):
         """See base class."""
         return ["0", "1"]
 
-    def _create_examples(self, input_data, set_type, only_first=None):
+    def _create_examples(self, input_data, set_type):
         """Creates examples for the training and dev sets."""
         
         is_training = set_type == "train"
@@ -363,9 +363,6 @@ class SquadProcessor(DataProcessor):
                     )
 
                     examples.append(example)
-
-                    if only_first is not None and len(examples) > only_first:
-                        return examples
         return examples
 
 class SquadV1Processor(SquadProcessor):

From 96fa9a8a70a52221446b0b887f99c90c5ce31eeb Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Wed, 4 Dec 2019 17:22:50 -0500
Subject: [PATCH 224/505] Python 2 + Post mime-type to S3

---
 transformers/hf_api.py            | 74 ++++++++++++++++++++++++-------
 transformers/tests/hf_api_test.py | 20 ++++++---
 2 files changed, 73 insertions(+), 21 deletions(-)

diff --git a/transformers/hf_api.py b/transformers/hf_api.py
index 238762ebf8..c21592a838 100644
--- a/transformers/hf_api.py
+++ b/transformers/hf_api.py
@@ -14,9 +14,9 @@
 # limitations under the License.
 from __future__ import absolute_import, division, print_function
 
-from typing import List, NamedTuple
 import os
 from os.path import expanduser
+import six
 
 import requests
 from requests.exceptions import HTTPError
@@ -24,23 +24,43 @@ from requests.exceptions import HTTPError
 ENDPOINT = "https://huggingface.co"
 
 class S3Obj:
-    def __init__(self, filename: str, LastModified: str, ETag: str, Size: int):
+    def __init__(
+        self,
+        filename,     # type: str
+        LastModified, # type: str
+        ETag,         # type: str
+        Size,         # type: int
+        **kwargs
+    ):
         self.filename = filename
         self.LastModified = LastModified
         self.ETag = ETag
         self.Size = Size
 
 
-class PresignedUrl(NamedTuple):
-    write: str
-    access: str
+class PresignedUrl:
+    def __init__(
+        self,
+        write,  # type: str
+        access, # type: str
+        type,   # type: str
+        **kwargs
+    ):
+        self.write = write
+        self.access = access
+        self.type = type # mime-type to send to S3.
 
 
 class HfApi:
     def __init__(self, endpoint=None):
         self.endpoint = endpoint if endpoint is not None else ENDPOINT
 
-    def login(self, username: str, password: str) -> str:
+    def login(
+        self,
+        username, # type: str
+        password, # type: str
+    ):
+        # type: (...) -> str
         """
         Call HF API to sign in a user and get a token if credentials are valid.
 
@@ -56,7 +76,11 @@ class HfApi:
         d = r.json()
         return d["token"]
 
-    def whoami(self, token: str) -> str:
+    def whoami(
+        self,
+        token, # type: str
+    ):
+        # type: (...) -> str
         """
         Call HF API to know "whoami"
         """
@@ -66,7 +90,8 @@ class HfApi:
         d = r.json()
         return d["user"]
 
-    def logout(self, token: str):
+    def logout(self, token):
+        # type: (...) -> void
         """
         Call HF API to log out.
         """
@@ -74,7 +99,8 @@ class HfApi:
         r = requests.post(path, headers={"authorization": "Bearer {}".format(token)})
         r.raise_for_status()
 
-    def presign(self, token: str, filename: str) -> PresignedUrl:
+    def presign(self, token, filename):
+        # type: (...) -> PresignedUrl
         """
         Call HF API to get a presigned url to upload `filename` to S3.
         """
@@ -88,7 +114,8 @@ class HfApi:
         d = r.json()
         return PresignedUrl(**d)
 
-    def presign_and_upload(self, token: str, filename: str, filepath: str) -> str:
+    def presign_and_upload(self, token, filename, filepath):
+        # type: (...) -> str
         """
         Get a presigned url, then upload file to S3.
 
@@ -98,12 +125,18 @@ class HfApi:
         urls = self.presign(token, filename=filename)
         # streaming upload:
         # https://2.python-requests.org/en/master/user/advanced/#streaming-uploads
+        # 
+        # Even though we presign with the correct content-type,
+        # the client still has to specify it when uploading the file.
         with open(filepath, "rb") as f:
-            r = requests.put(urls.write, data=f)
+            r = requests.put(urls.write, data=f, headers={
+                "content-type": urls.type,
+            })
             r.raise_for_status()
         return urls.access
 
-    def list_objs(self, token: str) -> List[S3Obj]:
+    def list_objs(self, token):
+        # type: (...) -> List[S3Obj]
         """
         Call HF API to list all stored files for user.
         """
@@ -121,11 +154,20 @@ class HfFolder:
     path_token = expanduser("~/.huggingface/token")
 
     @classmethod
-    def save_token(cls, token: str):
+    def save_token(cls, token):
         """
         Save token, creating folder as needed.
         """
-        os.makedirs(os.path.dirname(cls.path_token), exist_ok=True)
+        if six.PY3:
+            os.makedirs(os.path.dirname(cls.path_token), exist_ok=True)
+        else:
+            # Python 2
+            try:
+                os.makedirs(os.path.dirname(cls.path_token))
+            except OSError as e:
+                if e.errno != os.errno.EEXIST:
+                    raise e
+                pass
         with open(cls.path_token, 'w+') as f:
             f.write(token)
 
@@ -137,7 +179,9 @@ class HfFolder:
         try:
             with open(cls.path_token, 'r') as f:
                 return f.read()
-        except FileNotFoundError:
+        except:
+            # this is too wide. When Py2 is dead use:
+            # `except FileNotFoundError:` instead
             return None
 
     @classmethod
diff --git a/transformers/tests/hf_api_test.py b/transformers/tests/hf_api_test.py
index 59822344ba..92d41b6dff 100644
--- a/transformers/tests/hf_api_test.py
+++ b/transformers/tests/hf_api_test.py
@@ -15,6 +15,7 @@
 from __future__ import absolute_import, division, print_function
 
 import os
+import six
 import time
 import unittest
 
@@ -40,7 +41,7 @@ class HfApiLoginTest(HfApiCommonTest):
 
     def test_login_valid(self):
         token = self._api.login(username=USER, password=PASS)
-        self.assertIsInstance(token, str)
+        self.assertIsInstance(token, six.string_types)
 
 
 class HfApiEndpointsTest(HfApiCommonTest):
@@ -56,19 +57,22 @@ class HfApiEndpointsTest(HfApiCommonTest):
         self.assertEqual(user, USER)
 
     def test_presign(self):
-        url = self._api.presign(token=self._token, filename=FILE_KEY)
-        self.assertIsInstance(url, PresignedUrl)
+        urls = self._api.presign(token=self._token, filename=FILE_KEY)
+        self.assertIsInstance(urls, PresignedUrl)
+        self.assertEqual(urls.type, "text/plain")
 
     def test_presign_and_upload(self):
         access_url = self._api.presign_and_upload(
             token=self._token, filename=FILE_KEY, filepath=FILE_PATH
         )
-        self.assertIsInstance(access_url, str)
+        self.assertIsInstance(access_url, six.string_types)
 
     def test_list_objs(self):
         objs = self._api.list_objs(token=self._token)
-        o = objs[-1]
-        self.assertIsInstance(o, S3Obj)
+        self.assertIsInstance(objs, list)
+        if len(objs) > 0:
+            o = objs[-1]
+            self.assertIsInstance(o, S3Obj)
 
 
 
@@ -92,3 +96,7 @@ class HfFolderTest(unittest.TestCase):
             HfFolder.get_token(),
             None
         )
+
+
+if __name__ == "__main__":
+    unittest.main()

From 7a03519975e4f0b6698bf1221c2263ed0f8d795c Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Wed, 4 Dec 2019 17:24:35 -0500
Subject: [PATCH 225/505] Documentation

---
 docs/source/main_classes/processors.rst |  79 +++++++++++++++++-
 transformers/data/processors/squad.py   | 104 ++++++++++++++++++++----
 2 files changed, 164 insertions(+), 19 deletions(-)

diff --git a/docs/source/main_classes/processors.rst b/docs/source/main_classes/processors.rst
index a85c126956..ce0eeb553a 100644
--- a/docs/source/main_classes/processors.rst
+++ b/docs/source/main_classes/processors.rst
@@ -55,4 +55,81 @@ Example usage
 ^^^^^^^^^^^^^^^^^^^^^^^^^
 
 An example using these processors is given in the
-`run_glue.py <https://github.com/huggingface/pytorch-transformers/blob/master/examples/run_glue.py>`__ script.
\ No newline at end of file
+`run_glue.py <https://github.com/huggingface/pytorch-transformers/blob/master/examples/run_glue.py>`__ script.
+
+
+
+SQuAD
+~~~~~~~~~~~~~~~~~~~~~
+
+`The Stanford Question Answering Dataset (SQuAD) <https://rajpurkar.github.io/SQuAD-explorer//>`__ is a benchmark that evaluates
+the performance of models on question answering. Two versions are available, v1.1 and v2.0. The first version (v1.1) was released together with the paper
+`SQuAD: 100,000+ Questions for Machine Comprehension of Text <https://arxiv.org/abs/1606.05250>`__. The second version (v2.0) was released alongside 
+the paper `Know What You Don't Know: Unanswerable Questions for SQuAD <https://arxiv.org/abs/1806.03822>`__.
+
+This library hosts a processor for each of the two versions:
+
+Processors
+^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Those processors are:
+    - :class:`~transformers.data.processors.utils.SquadV1Processor`
+    - :class:`~transformers.data.processors.utils.SquadV2Processor`
+
+They both inherit from the abstract class :class:`~transformers.data.processors.utils.SquadProcessor`
+
+.. autoclass:: transformers.data.processors.squad.SquadProcessor
+    :members:
+
+Additionally, the following method can be used to convert SQuAD examples into :class:`~transformers.data.processors.utils.SquadFeatures`
+that can be used as model inputs.
+
+.. automethod:: transformers.data.processors.squad.squad_convert_examples_to_features
+
+These processors as well as the aforementionned method can be used with files containing the data as well as with the `tensorflow_datasets` package.
+Examples are given below.
+
+Example usage
+^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Here is an example using the processors as well as the conversion method using data files:
+
+Example::
+
+    # Loading a V2 processor
+    processor = SquadV2Processor()
+    examples = processor.get_dev_examples(squad_v2_data_dir)
+
+    # Loading a V1 processor
+    processor = SquadV1Processor()
+    examples = processor.get_dev_examples(squad_v1_data_dir)
+
+    features = squad_convert_examples_to_features( 
+        examples=examples,
+        tokenizer=tokenizer,
+        max_seq_length=max_seq_length,
+        doc_stride=args.doc_stride,
+        max_query_length=max_query_length,
+        is_training=not evaluate,
+    )
+
+Using `tensorflow_datasets` is as easy as using a data file:
+
+Example::
+
+    # tensorflow_datasets only handle Squad V1.
+    tfds_examples = tfds.load("squad")
+    examples = SquadV1Processor().get_examples_from_dataset(tfds_examples, evaluate=evaluate)
+
+    features = squad_convert_examples_to_features( 
+        examples=examples,
+        tokenizer=tokenizer,
+        max_seq_length=max_seq_length,
+        doc_stride=args.doc_stride,
+        max_query_length=max_query_length,
+        is_training=not evaluate,
+    )
+
+
+Another example using these processors is given in the
+`run_squad.py <https://github.com/huggingface/pytorch-transformers/blob/master/examples/run_squad.py>`__ script.
\ No newline at end of file
diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index 09a79db471..b17e626c98 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -74,7 +74,35 @@ def _is_whitespace(c):
 
 def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
                                        doc_stride, max_query_length, is_training):
-    """Loads a data file into a list of `InputBatch`s."""
+    """
+    Converts a list of examples into a list of features that can be directly given as input to a model.
+    It is model-dependant and takes advantage of many of the tokenizer's features to create the model's inputs.
+
+    Args:
+        examples: list of :class:`~transformers.data.processors.squad.SquadExample`
+        tokenizer: an instance of a child of :class:`~transformers.PreTrainedTokenizer`
+        max_seq_length: The maximum sequence length of the inputs.
+        doc_stride: The stride used when the context is too large and is split across several features.
+        max_query_length: The maximum length of the query.
+        is_training: wheter to create features for model evaluation or model training.
+
+    Returns:
+        list of :class:`~transformers.data.processors.squad.SquadFeatures`
+
+    Example::
+
+        processor = SquadV2Processor()
+        examples = processor.get_dev_examples(data_dir)
+
+        features = squad_convert_examples_to_features( 
+            examples=examples,
+            tokenizer=tokenizer,
+            max_seq_length=args.max_seq_length,
+            doc_stride=args.doc_stride,
+            max_query_length=args.max_query_length,
+            is_training=not evaluate,
+        )
+    """
 
     # Defining helper methods    
     unique_id = 1000000000
@@ -240,12 +268,14 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
 
 
 class SquadProcessor(DataProcessor):
-    """Processor for the SQuAD data set."""
+    """
+    Processor for the SQuAD data set.
+    Overriden by SquadV1Processor and SquadV2Processor, used by the version 1.1 and version 2.0 of SQuAD, respectively.
+    """
     train_file = None
     dev_file = None
 
-    def get_example_from_tensor_dict(self, tensor_dict, evaluate=False):
-
+    def _get_example_from_tensor_dict(self, tensor_dict, evaluate=False):
         if not evaluate:
             answer = tensor_dict['answers']['text'][0].numpy().decode('utf-8')
             answer_start = tensor_dict['answers']['answer_start'][0].numpy()
@@ -296,35 +326,44 @@ class SquadProcessor(DataProcessor):
 
         examples = []
         for tensor_dict in tqdm(dataset):
-            examples.append(self.get_example_from_tensor_dict(tensor_dict, evaluate=evaluate)) 
+            examples.append(self._get_example_from_tensor_dict(tensor_dict, evaluate=evaluate)) 
 
         return examples
 
-    def get_train_examples(self, data_dir):
-        """See base class."""
+    def get_train_examples(self, data_dir, filename=None):
+        """
+        Returns the training examples from the data directory.
+
+        Args:
+            data_dir: Directory containing the data files used for training and evaluating.
+            filename: None by default, specify this if the training file has a different name than the original one
+                which is `train-v1.1.json` and `train-v2.0.json` for squad versions 1.1 and 2.0 respectively.
+
+        """
         if self.train_file is None:
             raise ValueError("SquadProcessor should be instantiated via SquadV1Processor or SquadV2Processor")
 
-        with open(os.path.join(data_dir, self.train_file), "r", encoding='utf-8') as reader:
+        with open(os.path.join(data_dir, self.train_file if filename is None else filename), "r", encoding='utf-8') as reader:
             input_data = json.load(reader)["data"]
         return self._create_examples(input_data, "train")
 
-    def get_dev_examples(self, data_dir):
-        """See base class."""
+    def get_dev_examples(self, data_dir, filename=None):
+        """
+        Returns the evaluation example from the data directory.
+
+        Args:
+            data_dir: Directory containing the data files used for training and evaluating.
+            filename: None by default, specify this if the evaluation file has a different name than the original one
+                which is `train-v1.1.json` and `train-v2.0.json` for squad versions 1.1 and 2.0 respectively.
+        """
         if self.dev_file is None:
             raise ValueError("SquadProcessor should be instantiated via SquadV1Processor or SquadV2Processor")
         
-        with open(os.path.join(data_dir, self.dev_file), "r", encoding='utf-8') as reader:
+        with open(os.path.join(data_dir, self.dev_file if filename is not None else filename), "r", encoding='utf-8') as reader:
             input_data = json.load(reader)["data"]
         return self._create_examples(input_data, "dev")
 
-    def get_labels(self):
-        """See base class."""
-        return ["0", "1"]
-
     def _create_examples(self, input_data, set_type):
-        """Creates examples for the training and dev sets."""
-        
         is_training = set_type == "train"
         examples = []
         for entry in tqdm(input_data):
@@ -378,6 +417,16 @@ class SquadV2Processor(SquadProcessor):
 class SquadExample(object):
     """
     A single training/test example for the Squad dataset, as loaded from disk.
+
+    Args:
+        qas_id: The example's unique identifier
+        question_text: The question string
+        context_text: The context string
+        answer_text: The answer string
+        start_position_character: The character position of the start of the answer
+        title: The title of the example
+        answers: None by default, this is used during evaluation. Holds answers as well as their start positions.
+        is_impossible: False by default, set to True if the example has no possible answer.
     """
 
     def __init__(self,
@@ -427,7 +476,26 @@ class SquadExample(object):
 class SquadFeatures(object):
     """
     Single squad example features to be fed to a model.
-    Those features are model-specific.
+    Those features are model-specific and can be crafted from :class:`~transformers.data.processors.squad.SquadExample`
+    using the :method:`~transformers.data.processors.squad.squad_convert_examples_to_features` method.
+
+    Args:
+        input_ids: Indices of input sequence tokens in the vocabulary.
+        attention_mask: Mask to avoid performing attention on padding token indices.
+        token_type_ids: Segment token indices to indicate first and second portions of the inputs.
+        cls_index: the index of the CLS token.
+        p_mask: Mask identifying tokens that can be answers vs. tokens that cannot.
+            Mask with 1 for tokens than cannot be in the answer and 0 for token that can be in an answer
+        example_index: the index of the example
+        unique_id: The unique Feature identifier
+        paragraph_len: The length of the context
+        token_is_max_context: List of booleans identifying which tokens have their maximum context in this feature object.
+            If a token does not have their maximum context in this feature object, it means that another feature object
+            has more information related to that token and should be prioritized over this feature for that token.
+        tokens: list of tokens corresponding to the input ids
+        token_to_orig_map: mapping between the tokens and the original text, needed in order to identify the answer.
+        start_position: start of the answer token index 
+        end_position: end of the answer token index 
     """
 
     def __init__(self,

From ce158a076f7089bf11d44e1581f5bcab4dcc5396 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Wed, 4 Dec 2019 17:55:52 -0500
Subject: [PATCH 226/505] Return dataset (pytorch)

---
 transformers/data/processors/squad.py | 41 ++++++++++++++++++++++++---
 1 file changed, 37 insertions(+), 4 deletions(-)

diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index b17e626c98..338bae0c51 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -7,7 +7,11 @@ import numpy as np
 
 from ...tokenization_bert import BasicTokenizer, whitespace_tokenize
 from .utils import DataProcessor, InputExample, InputFeatures
-from ...file_utils import is_tf_available
+from ...file_utils import is_tf_available, is_torch_available
+
+if is_torch_available:
+    import torch
+    from torch.utils.data import TensorDataset
 
 if is_tf_available():
     import tensorflow as tf
@@ -73,7 +77,8 @@ def _is_whitespace(c):
     return False
 
 def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
-                                       doc_stride, max_query_length, is_training):
+                                       doc_stride, max_query_length, is_training, 
+                                       return_dataset=False):
     """
     Converts a list of examples into a list of features that can be directly given as input to a model.
     It is model-dependant and takes advantage of many of the tokenizer's features to create the model's inputs.
@@ -84,7 +89,10 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
         max_seq_length: The maximum sequence length of the inputs.
         doc_stride: The stride used when the context is too large and is split across several features.
         max_query_length: The maximum length of the query.
-        is_training: wheter to create features for model evaluation or model training.
+        is_training: whether to create features for model evaluation or model training.
+        return_dataset: Default False. Either 'pt' or 'tf'.
+            if 'pt': returns a torch.data.TensorDataset,
+            if 'tf': returns a tf.data.Dataset
 
     Returns:
         list of :class:`~transformers.data.processors.squad.SquadFeatures`
@@ -264,6 +272,31 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
 
             unique_id += 1
 
+    if return_dataset == 'pt':
+        if not is_torch_available():
+            raise ImportError("Pytorch must be installed to return a pytorch dataset.")
+
+        # Convert to Tensors and build dataset
+        all_input_ids = torch.tensor([f.input_ids for f in features], dtype=torch.long)
+        all_input_mask = torch.tensor([f.attention_mask for f in features], dtype=torch.long)
+        all_segment_ids = torch.tensor([f.token_type_ids for f in features], dtype=torch.long)
+        all_cls_index = torch.tensor([f.cls_index for f in features], dtype=torch.long)
+        all_p_mask = torch.tensor([f.p_mask for f in features], dtype=torch.float)
+
+        if not is_training:
+            all_example_index = torch.arange(all_input_ids.size(0), dtype=torch.long)
+            dataset = TensorDataset(all_input_ids, all_input_mask, all_segment_ids,
+                                    all_example_index, all_cls_index, all_p_mask)
+        else:
+            all_start_positions = torch.tensor([f.start_position for f in features], dtype=torch.long)
+            all_end_positions = torch.tensor([f.end_position for f in features], dtype=torch.long)
+            dataset = TensorDataset(all_input_ids, all_input_mask, all_segment_ids,
+                                    all_start_positions, all_end_positions,
+                                    all_cls_index, all_p_mask)
+
+        return features, dataset
+        
+
     return features
 
 
@@ -359,7 +392,7 @@ class SquadProcessor(DataProcessor):
         if self.dev_file is None:
             raise ValueError("SquadProcessor should be instantiated via SquadV1Processor or SquadV2Processor")
         
-        with open(os.path.join(data_dir, self.dev_file if filename is not None else filename), "r", encoding='utf-8') as reader:
+        with open(os.path.join(data_dir, self.dev_file if filename is None else filename), "r", encoding='utf-8') as reader:
             input_data = json.load(reader)["data"]
         return self._create_examples(input_data, "dev")
 

From 3ba417e1a877d2c5f2170b12a8cc39f16dbf46aa Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Wed, 4 Dec 2019 18:40:52 -0500
Subject: [PATCH 227/505] [cli] ls: Tabular formatting

---
 transformers/commands/user.py | 61 +++++++++++++++++++++++++++++------
 1 file changed, 52 insertions(+), 9 deletions(-)

diff --git a/transformers/commands/user.py b/transformers/commands/user.py
index 4b826f4dc6..d79922ed8a 100644
--- a/transformers/commands/user.py
+++ b/transformers/commands/user.py
@@ -25,6 +25,17 @@ class UserCommands(BaseTransformersCLICommand):
 
 
 
+class ANSI:
+    """
+    Helper for en.wikipedia.org/wiki/ANSI_escape_code
+    """
+    _bold = u"\u001b[1m"
+    _reset = u"\u001b[0m"
+    @classmethod
+    def bold(cls, s):
+        return "{}{}{}".format(cls._bold, s, cls._reset)
+
+
 class BaseUserCommand:
     def __init__(self, args):
         self.args = args
@@ -80,6 +91,28 @@ class LogoutCommand(BaseUserCommand):
 
 
 class ListObjsCommand(BaseUserCommand):
+    def tabulate(self, rows, headers):
+        # type: (List[List[Union[str, int]]], List[str]) -> str
+        """
+        Inspired by:
+        stackoverflow.com/a/8356620/593036
+        stackoverflow.com/questions/9535954/printing-lists-as-tabular-data
+        """
+        col_widths = [max(len(str(x)) for x in col) for col in zip(*rows, headers)]
+        row_format = ("{{:{}}} " * len(headers)).format(*col_widths)
+        lines = []
+        lines.append(
+            row_format.format(*headers)
+        )
+        lines.append(
+            row_format.format(*["-" * w for w in col_widths])
+        )
+        for row in rows:
+            lines.append(
+                row_format.format(*row)
+            )
+        return "\n".join(lines)
+
     def run(self):
         token = HfFolder.get_token()
         if token is None:
@@ -92,13 +125,16 @@ class ListObjsCommand(BaseUserCommand):
             exit(1)
         if len(objs) == 0:
             print("No shared file yet")
-        for obj in objs:
-            print(
-                obj.filename,
-                obj.LastModified,
-                obj.ETag,
-                obj.Size
-            )
+            exit()
+        rows = [ [
+            obj.filename,
+            obj.LastModified,
+            obj.ETag,
+            obj.Size
+        ] for obj in objs ]
+        print(
+            self.tabulate(rows, headers=["Filename", "LastModified", "ETag", "Size"])
+        )
 
 
 class UploadCommand(BaseUserCommand):
@@ -109,12 +145,19 @@ class UploadCommand(BaseUserCommand):
             exit(1)
         filepath = os.path.join(os.getcwd(), self.args.file)
         filename = self.args.filename if self.args.filename is not None else os.path.basename(filepath)
-        print("About to upload file {} to S3 under filename {}".format(filepath, filename))
+        print(
+            "About to upload file {} to S3 under filename {}".format(
+                ANSI.bold(filepath), ANSI.bold(filename)
+            )
+        )
+
         choice = input("Proceed? [Y/n] ").lower()
         if not(choice == "" or choice == "y" or choice == "yes"):
             print("Abort")
             exit()
-        print("Uploading...")
+        print(
+            ANSI.bold("Uploading... This might take a while if file is large")
+        )
         access_url = self._api.presign_and_upload(
             token=token, filename=filename, filepath=filepath
         )

From fb0d2f1da102d699c6457fd98be35f89852d08b9 Mon Sep 17 00:00:00 2001
From: VictorSanh <victorsanh@gmail.com>
Date: Thu, 5 Dec 2019 03:00:16 -0500
Subject: [PATCH 228/505] preparing release distil-mBERT

---
 transformers/modeling_tf_distilbert.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/transformers/modeling_tf_distilbert.py b/transformers/modeling_tf_distilbert.py
index b3d4889475..79ae49d811 100644
--- a/transformers/modeling_tf_distilbert.py
+++ b/transformers/modeling_tf_distilbert.py
@@ -37,7 +37,8 @@ logger = logging.getLogger(__name__)
 
 TF_DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
     'distilbert-base-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-tf_model.h5",
-    'distilbert-base-uncased-distilled-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-distilled-squad-tf_model.h5"
+    'distilbert-base-uncased-distilled-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-distilled-squad-tf_model.h5",
+    'distilbert-base-multilingual-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-multilingual-cased-tf_model.h5",
 }
 
 

From 7f998b1b832dd69cfdd8455afd5b8af3b2f77df8 Mon Sep 17 00:00:00 2001
From: Guillaume B <guillaume.becquin@gmail.com>
Date: Thu, 5 Dec 2019 08:57:49 +0100
Subject: [PATCH 229/505] special_tokens_mask value was unused and calculated
 twice

---
 transformers/tokenization_utils.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index 5d683629f0..6be96989cb 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -910,7 +910,7 @@ class PreTrainedTokenizer(object):
             token_type_ids = [0] * len(ids) + ([1] * len(pair_ids) if pair else [])
             special_tokens_mask = [0] * (len(ids) + (len(pair_ids) if pair else 0))
         if return_special_tokens_mask:
-            encoded_inputs["special_tokens_mask"] = self.get_special_tokens_mask(ids, pair_ids)
+            encoded_inputs["special_tokens_mask"] = special_tokens_mask
 
         # Prepare inputs as tensors if asked
         if return_tensors == 'tf' and is_tf_available():

From 8b388827b509e0c117c53803f2ee030ead0e5a81 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Thu, 5 Dec 2019 11:18:43 +0100
Subject: [PATCH 230/505] fix #1920

---
 transformers/tokenization_ctrl.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/transformers/tokenization_ctrl.py b/transformers/tokenization_ctrl.py
index 3d67fa2c5b..9454cbbaf3 100644
--- a/transformers/tokenization_ctrl.py
+++ b/transformers/tokenization_ctrl.py
@@ -192,9 +192,9 @@ class CTRLTokenizer(PreTrainedTokenizer):
         """
         split_tokens = []
 
-        text = text.split(' ')
+        words = re.findall(r'\S+\n?', text)
 
-        for token in text:
+        for token in words:
             split_tokens.extend([t for t in self.bpe(token).split(' ')])
         return split_tokens
 

From 75a97af6bc3b842df22e3bc12e530c22c5e15482 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Thu, 5 Dec 2019 11:26:55 +0100
Subject: [PATCH 231/505] fix #1450 - add doc

---
 examples/README.md | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/examples/README.md b/examples/README.md
index 960b218f11..25bbcf2246 100644
--- a/examples/README.md
+++ b/examples/README.md
@@ -4,12 +4,14 @@ In this section a few examples are put together. All of these examples work for
 similar API between the different models.
 
 **Important**  
-To run the latest versions of the examples, you have to install from source. Execute the following steps in a new virtual environment:
+To run the latest versions of the examples, you have to install from source and install some specific requirements for the examples.
+Execute the following steps in a new virtual environment:
 
 ```bash
 git clone https://github.com/huggingface/transformers
 cd transformers
 pip install [--editable] .
+pip install -r ./examples/requirements.txt
 ```
 
 | Section                    | Description                                                                                                                                                |

From 71e4693f087271053d5c188319cfb5217836cca4 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Thu, 5 Dec 2019 12:14:24 +0100
Subject: [PATCH 232/505] fix #1968

---
 transformers/__init__.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/transformers/__init__.py b/transformers/__init__.py
index f06ee3f35d..ab5090723d 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -98,7 +98,7 @@ if is_torch_available():
                                 RobertaForSequenceClassification, RobertaForMultipleChoice,
                                 RobertaForTokenClassification,
                                 ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP)
-    from .modeling_distilbert import (DistilBertForMaskedLM, DistilBertModel,
+    from .modeling_distilbert import (DistilBertPreTrainedModel, DistilBertForMaskedLM, DistilBertModel,
                                 DistilBertForSequenceClassification, DistilBertForQuestionAnswering,
                                 DistilBertForTokenClassification,
                                 DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
@@ -108,7 +108,7 @@ if is_torch_available():
                                 CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
     from .modeling_encoder_decoder import PreTrainedEncoderDecoder, Model2Model
 
-    from .modeling_albert import (AlbertModel, AlbertForMaskedLM, AlbertForSequenceClassification,
+    from .modeling_albert import (AlbertPreTrainedModel, AlbertModel, AlbertForMaskedLM, AlbertForSequenceClassification,
                                 AlbertForQuestionAnswering,
                                 load_tf_weights_in_albert, ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
 

From 9200a759d782a87530765fb32f52b6248c7f4d03 Mon Sep 17 00:00:00 2001
From: Julien Plu <julien.plu@schibsted.com>
Date: Thu, 5 Dec 2019 12:56:43 +0100
Subject: [PATCH 233/505] Add few tests on the TF optimization file with some
 info in the documentation. Complete the README.

---
 .../main_classes/optimizer_schedules.rst      | 24 +++++
 examples/README.md                            | 77 +++++++++++++++-
 examples/run_tf_ner.py                        |  7 +-
 transformers/tests/optimization_tf_test.py    | 89 +++++++++++++++++++
 4 files changed, 191 insertions(+), 6 deletions(-)
 create mode 100644 transformers/tests/optimization_tf_test.py

diff --git a/docs/source/main_classes/optimizer_schedules.rst b/docs/source/main_classes/optimizer_schedules.rst
index b30a2e0e2e..22ed1b28fb 100644
--- a/docs/source/main_classes/optimizer_schedules.rst
+++ b/docs/source/main_classes/optimizer_schedules.rst
@@ -5,6 +5,7 @@ The ``.optimization`` module provides:
 
 - an optimizer with weight decay fixed that can be used to fine-tuned models, and
 - several schedules in the form of schedule objects that inherit from ``_LRSchedule``:
+- a gradient accumulation class to accumulate the gradients of multiple batches
 
 ``AdamW``
 ~~~~~~~~~~~~~~~~
@@ -12,6 +13,15 @@ The ``.optimization`` module provides:
 .. autoclass:: transformers.AdamW
     :members:
 
+``AdamWeightDecay``
+~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.AdamWeightDecay
+    :members:
+
+.. autofunction:: transformers.create_optimizer
+    :members:
+
 Schedules
 ----------------------------------------------------
 
@@ -49,3 +59,17 @@ Learning Rate Schedules
 .. image:: /imgs/warmup_linear_schedule.png
     :target: /imgs/warmup_linear_schedule.png
     :alt:
+
+``Warmup``
+~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.Warmup
+    :members:
+
+Gradient Strategies
+----------------------------------------------------
+
+``GradientAccumulator``
+~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.GradientAccumulator
diff --git a/examples/README.md b/examples/README.md
index 960b218f11..2dd6653916 100644
--- a/examples/README.md
+++ b/examples/README.md
@@ -465,7 +465,8 @@ Training with the previously defined hyper-parameters yields the following resul
 
 ## Named Entity Recognition
 
-Based on the script [`run_ner.py`](https://github.com/huggingface/transformers/blob/master/examples/run_ner.py).
+Based on the scripts [`run_ner.py`](https://github.com/huggingface/transformers/blob/master/examples/run_ner.py) for Pytorch and
+[`run_tf_ner.py`(https://github.com/huggingface/transformers/blob/master/examples/run_tf_ner.py)] for Tensorflow 2.
 This example fine-tune Bert Multilingual on GermEval 2014 (German NER).
 Details and results for the fine-tuning provided by @stefan-it.
 
@@ -510,7 +511,7 @@ The GermEval 2014 dataset has much more labels than CoNLL-2002/2003 datasets, so
 cat train.txt dev.txt test.txt | cut -d " " -f 2 | grep -v "^$"| sort | uniq > labels.txt
 ```
 
-### Training
+### Prepare the run
 
 Additional environment variables must be set:
 
@@ -522,6 +523,8 @@ export SAVE_STEPS=750
 export SEED=1
 ```
 
+### Run the Pytorch version
+
 To start training, just run:
 
 ```bash
@@ -542,7 +545,7 @@ python3 run_ner.py --data_dir ./ \
 
 If your GPU supports half-precision training, just add the `--fp16` flag. After training, the model will be both evaluated on development and test datasets.
 
-### Evaluation
+#### Evaluation
 
 Evaluation on development dataset outputs the following for our example:
 
@@ -564,7 +567,7 @@ On the test dataset the following results could be achieved:
 10/04/2019 00:42:42 - INFO - __main__ -     recall = 0.8624150210424085
 ```
 
-### Comparing BERT (large, cased), RoBERTa (large, cased) and DistilBERT (base, uncased)
+#### Comparing BERT (large, cased), RoBERTa (large, cased) and DistilBERT (base, uncased)
 
 Here is a small comparison between BERT (large, cased), RoBERTa (large, cased) and DistilBERT (base, uncased) with the same hyperparameters as specified in the [example documentation](https://huggingface.co/transformers/examples.html#named-entity-recognition) (one run):
 
@@ -574,6 +577,72 @@ Here is a small comparison between BERT (large, cased), RoBERTa (large, cased) a
 | `roberta-large`                  | 95.96 | 91.87
 | `distilbert-base-uncased` | 94.34 | 90.32
 
+### Run the Tensorflow 2 version
+
+To start training, just run:
+
+```bash
+python3 run_tf_ner.py --data_dir ./ \
+--model_type bert \
+--labels ./labels.txt \
+--model_name_or_path $BERT_MODEL \
+--output_dir $OUTPUT_DIR \
+--max_seq_length  $MAX_LENGTH \
+--num_train_epochs $NUM_EPOCHS \
+--per_device_train_batch_size $BATCH_SIZE \
+--save_steps $SAVE_STEPS \
+--seed $SEED \
+--do_train \
+--do_eval \
+--do_predict
+```
+
+Such as the Pytorch version, if your GPU supports half-precision training, just add the `--fp16` flag. After training, the model will be both evaluated on development and test datasets.
+
+#### Evaluation
+
+Evaluation on development dataset outputs the following for our example:
+```bash
+           precision    recall  f1-score   support
+
+ LOCderiv     0.7619    0.6154    0.6809        52
+  PERpart     0.8724    0.8997    0.8858      4057
+  OTHpart     0.9360    0.9466    0.9413       711
+  ORGpart     0.7015    0.6989    0.7002       269
+  LOCpart     0.7668    0.8488    0.8057       496
+      LOC     0.8745    0.9191    0.8963       235
+ ORGderiv     0.7723    0.8571    0.8125        91
+ OTHderiv     0.4800    0.6667    0.5581        18
+      OTH     0.5789    0.6875    0.6286        16
+ PERderiv     0.5385    0.3889    0.4516        18
+      PER     0.5000    0.5000    0.5000         2
+      ORG     0.0000    0.0000    0.0000         3
+
+micro avg     0.8574    0.8862    0.8715      5968
+macro avg     0.8575    0.8862    0.8713      5968
+```
+
+On the test dataset the following results could be achieved:
+```bash
+           precision    recall  f1-score   support
+
+  PERpart     0.8847    0.8944    0.8896      9397
+  OTHpart     0.9376    0.9353    0.9365      1639
+  ORGpart     0.7307    0.7044    0.7173       697
+      LOC     0.9133    0.9394    0.9262       561
+  LOCpart     0.8058    0.8157    0.8107      1150
+      ORG     0.0000    0.0000    0.0000         8
+ OTHderiv     0.5882    0.4762    0.5263        42
+ PERderiv     0.6571    0.5227    0.5823        44
+      OTH     0.4906    0.6667    0.5652        39
+ ORGderiv     0.7016    0.7791    0.7383       172
+ LOCderiv     0.8256    0.6514    0.7282       109
+      PER     0.0000    0.0000    0.0000        11
+
+micro avg     0.8722    0.8774    0.8748     13869
+macro avg     0.8712    0.8774    0.8740     13869
+```
+
 ## Abstractive summarization
 
 Based on the script
diff --git a/examples/run_tf_ner.py b/examples/run_tf_ner.py
index ef1fcf6aa4..eb284f4c2a 100644
--- a/examples/run_tf_ner.py
+++ b/examples/run_tf_ner.py
@@ -540,6 +540,9 @@ def main(_):
             checkpoints = list(os.path.dirname(c) for c in sorted(glob.glob(args['output_dir'] + "/**/" + TF2_WEIGHTS_NAME, recursive=True), key=lambda f: int(''.join(filter(str.isdigit, f)) or -1)))
         
         logging.info("Evaluate the following checkpoints: %s", checkpoints)
+
+        if len(checkpoints) == 0:
+            checkpoints.append(args['output_dir'])
         
         for checkpoint in checkpoints:
             global_step = checkpoint.split("-")[-1] if re.match(".*checkpoint-[0-9]", checkpoint) else "final"
@@ -572,10 +575,10 @@ def main(_):
     if args['do_predict']:
         tokenizer = tokenizer_class.from_pretrained(args['output_dir'], do_lower_case=args['do_lower_case'])
         model = model_class.from_pretrained(args['output_dir'])
-        eval_batch_size = args['per_gpu_eval_batch_size'] * args['n_device']
+        eval_batch_size = args['per_device_eval_batch_size'] * args['n_device']
         predict_dataset, _ = load_and_cache_examples(args, tokenizer, labels, pad_token_label_id, eval_batch_size, mode="test")
         y_true, y_pred, pred_loss = evaluate(args, strategy, model, tokenizer, labels, pad_token_label_id, mode="test")
-        output_test_results_file = os.path.join(args.output_dir, "test_results.txt")
+        output_test_results_file = os.path.join(args['output_dir'], "test_results.txt")
         output_test_predictions_file = os.path.join(args['output_dir'], "test_predictions.txt")
         report = metrics.classification_report(y_true, y_pred, digits=4)
 
diff --git a/transformers/tests/optimization_tf_test.py b/transformers/tests/optimization_tf_test.py
new file mode 100644
index 0000000000..ac5109cb56
--- /dev/null
+++ b/transformers/tests/optimization_tf_test.py
@@ -0,0 +1,89 @@
+from __future__ import absolute_import
+from __future__ import division
+from __future__ import print_function
+
+import unittest
+import pytest
+
+from transformers import is_tf_available
+
+if is_tf_available():
+    import tensorflow as tf
+    from tensorflow.python.eager import context
+    from tensorflow.python.framework import ops
+    from transformers import (create_optimizer, GradientAccumulator)
+else:
+    pytestmark = pytest.mark.skip("Require TensorFlow")
+
+class OptimizationFTest(unittest.TestCase):
+    def assertListAlmostEqual(self, list1, list2, tol):
+        self.assertEqual(len(list1), len(list2))
+        for a, b in zip(list1, list2):
+            self.assertAlmostEqual(a, b, delta=tol)
+    
+    def testGradientAccumulator(self):
+        accumulator = GradientAccumulator()
+        accumulator([tf.constant([1.0, 2.0])])
+        accumulator([tf.constant([-2.0, 1.0])])
+        accumulator([tf.constant([-1.0, 2.0])])
+        with self.assertRaises(ValueError):
+            accumulator([tf.constant([1.0, 1.0]), tf.constant([2.0, 2.0])])
+        self.assertEqual(accumulator.step, 3)
+        self.assertEqual(len(accumulator.gradients), 1)
+        self.assertListAlmostEqual(accumulator.gradients[0].numpy().tolist(), [-2.0, 5.0], tol=1e-2)
+        accumulator.reset()
+        self.assertEqual(accumulator.step, 0)
+        self.assertListAlmostEqual(accumulator.gradients[0].numpy().tolist(), [0.0, 0.0], tol=1e-2)
+
+    def testGradientAccumulatorDistributionStrategy(self):
+        context._context = None
+        ops.enable_eager_execution_internal()
+        physical_devices = tf.config.experimental.list_physical_devices("CPU")
+        tf.config.experimental.set_virtual_device_configuration(
+            physical_devices[0],
+            [tf.config.experimental.VirtualDeviceConfiguration(),
+            tf.config.experimental.VirtualDeviceConfiguration()])
+
+        devices = tf.config.experimental.list_logical_devices(device_type="CPU")
+        strategy = tf.distribute.MirroredStrategy(devices=[device.name for device in devices])
+
+        with strategy.scope():
+            accumulator = GradientAccumulator()
+            variable = tf.Variable([4.0, 3.0])
+            optimizer = create_optimizer(5e-5, 10, 5)
+            gradient_placeholder = tf.Variable([0.0, 0.0], trainable=False)
+
+        def accumulate_on_replica(gradient):
+            accumulator([gradient])
+
+        def apply_on_replica():
+            optimizer.apply_gradients(list(zip(accumulator.gradients, [variable])), 1.0)
+
+        @tf.function
+        def accumulate(grad1, grad2):
+            with strategy.scope():
+                gradient_placeholder.values[0].assign(grad1)
+                gradient_placeholder.values[1].assign(grad2)
+                strategy.experimental_run_v2(accumulate_on_replica, args=(gradient_placeholder,))
+
+        @tf.function
+        def apply_grad():
+            with strategy.scope():
+                strategy.experimental_run_v2(apply_on_replica)
+
+        accumulate([1.0, 2.0], [-1.0, 1.0])
+        accumulate([3.0, -1.0], [-1.0, -1.0])
+        accumulate([-2.0, 2.0], [3.0, -2.0])
+        self.assertEqual(accumulator.step, 3)
+        self.assertListAlmostEqual(accumulator._gradients[0].values[0].value().numpy().tolist(), [2.0, 3.0], tol=1e-2)
+        self.assertListAlmostEqual(accumulator._gradients[0].values[1].value().numpy().tolist(), [1.0, -2.0], tol=1e-2)
+        apply_grad()
+        self.assertListAlmostEqual(variable.value().numpy().tolist(), [4.0, 3.0], tol=1e-2)
+        accumulator.reset()
+        self.assertEqual(accumulator.step, 0)
+        self.assertListAlmostEqual(accumulator._gradients[0].values[0].value().numpy().tolist(), [0.0, 0.0], tol=1e-2)
+        self.assertListAlmostEqual(accumulator._gradients[0].values[1].value().numpy().tolist(), [0.0, 0.0], tol=1e-2)
+
+
+if __name__ == "__main__":
+    unittest.main()
\ No newline at end of file

From 6c5297a42384c9234ad5fd2d044a5aa61d82ce97 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Thu, 5 Dec 2019 13:27:58 +0100
Subject: [PATCH 234/505] Fixing camembert tokenization

---
 transformers/tokenization_camembert.py | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/transformers/tokenization_camembert.py b/transformers/tokenization_camembert.py
index bf2a6fe993..b4091558e1 100644
--- a/transformers/tokenization_camembert.py
+++ b/transformers/tokenization_camembert.py
@@ -51,7 +51,7 @@ class CamembertTokenizer(PreTrainedTokenizer):
 
     def __init__(self, vocab_file, bos_token="<s>", eos_token="</s>", sep_token="</s>",
                  cls_token="<s>", unk_token="<unk>", pad_token='<pad>', mask_token='<mask>',
-                 additional_special_tokens=['<s>NOTUSED', '<s>NOTUSED'], **kwargs):
+                 additional_special_tokens=['<s>NOTUSED', '</s>NOTUSED'], **kwargs):
         super(CamembertTokenizer, self).__init__(max_len=512, bos_token=bos_token, eos_token=eos_token, unk_token=unk_token,
                                                  sep_token=sep_token, cls_token=cls_token, pad_token=pad_token,
                                                  mask_token=mask_token, additional_special_tokens=additional_special_tokens,
@@ -125,7 +125,7 @@ class CamembertTokenizer(PreTrainedTokenizer):
 
     @property
     def vocab_size(self):
-        return self.fairseq_offset + len(self.sp_model)
+        return len(self.fairseq_tokens_to_ids) + len(self.sp_model)
 
     def _tokenize(self, text):
         return self.sp_model.EncodeAsPieces(text)
@@ -134,6 +134,9 @@ class CamembertTokenizer(PreTrainedTokenizer):
         """ Converts a token (str/unicode) in an id using the vocab. """
         if token in self.fairseq_tokens_to_ids:
             return self.fairseq_tokens_to_ids[token]
+        elif self.sp_model.PieceToId(token) == 0:
+            # Convert sentence piece unk token to fairseq unk token index
+            return self.unk_token_id
         return self.fairseq_offset + self.sp_model.PieceToId(token)
 
     def _convert_id_to_token(self, index):

From 3268ebd2290800036fce0b931dc6c9b87b76e098 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Thu, 5 Dec 2019 13:35:29 +0100
Subject: [PATCH 235/505] fix xlnet test

---
 transformers/tests/modeling_xlnet_test.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/tests/modeling_xlnet_test.py b/transformers/tests/modeling_xlnet_test.py
index 3f0ef6793c..38888d4488 100644
--- a/transformers/tests/modeling_xlnet_test.py
+++ b/transformers/tests/modeling_xlnet_test.py
@@ -167,7 +167,7 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
                 [[self.seq_length, self.batch_size, self.hidden_size]] * self.num_hidden_layers)
 
         def create_and_check_xlnet_base_model_with_att_output(self, config, input_ids_1, input_ids_2, input_ids_q, perm_mask, input_mask,
-                target_mapping, segment_ids, lm_labels, sequence_labels, is_impossible_labels):
+                    target_mapping, segment_ids, lm_labels, sequence_labels, is_impossible_labels, token_labels):
             model = XLNetModel(config)
             model.eval()
 

From 2d5d86e03779b4b316698438caff0f675ee54abd Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Thu, 5 Dec 2019 14:06:29 +0100
Subject: [PATCH 236/505] fix #2031

---
 transformers/tokenization_albert.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/tokenization_albert.py b/transformers/tokenization_albert.py
index 2f9af0b0bc..40a4b29206 100644
--- a/transformers/tokenization_albert.py
+++ b/transformers/tokenization_albert.py
@@ -66,7 +66,7 @@ class AlbertTokenizer(PreTrainedTokenizer):
     def __init__(self, vocab_file,
                  do_lower_case=True, remove_space=True, keep_accents=False,
                  bos_token="[CLS]", eos_token="[SEP]", unk_token="<unk>", sep_token="[SEP]",
-                 pad_token="<pad>", cls_token="[CLS]", mask_token="[MASK]>", **kwargs):
+                 pad_token="<pad>", cls_token="[CLS]", mask_token="[MASK]", **kwargs):
         super(AlbertTokenizer, self).__init__(bos_token=bos_token, eos_token=eos_token,
                                              unk_token=unk_token, sep_token=sep_token,
                                              pad_token=pad_token, cls_token=cls_token,

From 18fb93530ba0c1f6a45240270b24dc5c5da340ae Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Thu, 5 Dec 2019 14:36:34 +0100
Subject: [PATCH 237/505] fixing #2042 - Nicer error message

---
 transformers/modeling_bert.py | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/transformers/modeling_bert.py b/transformers/modeling_bert.py
index 5f92fb96a3..1ee3e3f097 100644
--- a/transformers/modeling_bert.py
+++ b/transformers/modeling_bert.py
@@ -667,11 +667,10 @@ class BertModel(BertPreTrainedModel):
         # ourselves in which case we just need to make it broadcastable to all heads.
         if attention_mask.dim() == 3:
             extended_attention_mask = attention_mask[:, None, :, :]
-
-        # Provided a padding mask of dimensions [batch_size, seq_length]
-        # - if the model is a decoder, apply a causal mask in addition to the padding mask
-        # - if the model is an encoder, make the mask broadcastable to [batch_size, num_heads, seq_length, seq_length]
-        if attention_mask.dim() == 2:
+        elif attention_mask.dim() == 2:
+            # Provided a padding mask of dimensions [batch_size, seq_length]
+            # - if the model is a decoder, apply a causal mask in addition to the padding mask
+            # - if the model is an encoder, make the mask broadcastable to [batch_size, num_heads, seq_length, seq_length]
             if self.config.is_decoder:
                 batch_size, seq_length = input_shape
                 seq_ids = torch.arange(seq_length, device=device)
@@ -679,6 +678,8 @@ class BertModel(BertPreTrainedModel):
                 extended_attention_mask = causal_mask[:, None, :, :] * attention_mask[:, None, None, :]
             else:
                 extended_attention_mask = attention_mask[:, None, None, :]
+        else:
+            raise ValueError("Wrong shape for input_ids (shape {}) or attention_mask (shape {})".format(input_shape, attention_mask.shape))
 
         # Since attention_mask is 1.0 for positions we want to attend and 0.0 for
         # masked positions, this operation will create a tensor which is 0.0 for
@@ -696,8 +697,11 @@ class BertModel(BertPreTrainedModel):
 
             if encoder_attention_mask.dim() == 3:
                 encoder_extended_attention_mask = encoder_attention_mask[:, None, :, :]
-            if encoder_attention_mask.dim() == 2:
+            elif encoder_attention_mask.dim() == 2:
                 encoder_extended_attention_mask = encoder_attention_mask[:, None, None, :]
+            else:
+                raise ValueError("Wrong shape for input_ids (shape {}) or encoder_attention_mask (shape {})".format(input_shape,
+                                                                                                                    encoder_attention_mask.shape))
 
             encoder_extended_attention_mask = encoder_extended_attention_mask.to(dtype=next(self.parameters()).dtype)  # fp16 compatibility
             encoder_extended_attention_mask = (1.0 - encoder_extended_attention_mask) * -10000.0

From f8fb4335c9cd79789ed6119e729348e0a1b51e2b Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Thu, 5 Dec 2019 15:19:32 +0100
Subject: [PATCH 238/505] clean up a little bit PT <=> TF conversion

---
 transformers/convert_pytorch_checkpoint_to_tf2.py | 9 +++++----
 transformers/modeling_utils.py                    | 3 ++-
 2 files changed, 7 insertions(+), 5 deletions(-)

diff --git a/transformers/convert_pytorch_checkpoint_to_tf2.py b/transformers/convert_pytorch_checkpoint_to_tf2.py
index d1776e9c14..d20eafe2e9 100644
--- a/transformers/convert_pytorch_checkpoint_to_tf2.py
+++ b/transformers/convert_pytorch_checkpoint_to_tf2.py
@@ -119,10 +119,11 @@ def convert_pt_checkpoint_to_tf(model_type, pytorch_checkpoint_path, config_file
         tf_inputs = tf.constant(inputs_list)
         tfo = tf_model(tf_inputs, training=False)  # build the network
 
-        pt_model = pt_model_class.from_pretrained(None,
-                                                  config=config,
-                                                  state_dict=torch.load(pytorch_checkpoint_path,
-                                                                        map_location='cpu'))
+        pt_model = pt_model_class(config)
+        pt_model.load_state_dict(torch.load(pytorch_checkpoint_path, map_location='cpu'),
+                                 strict-False)
+        pt_model.eval()
+
         pt_inputs = torch.tensor(inputs_list)
         with torch.no_grad():
             pto = pt_model(pt_inputs)
diff --git a/transformers/modeling_utils.py b/transformers/modeling_utils.py
index 398172a88c..3ac568771e 100644
--- a/transformers/modeling_utils.py
+++ b/transformers/modeling_utils.py
@@ -318,7 +318,8 @@ class PreTrainedModel(nn.Module):
             model = BertModel.from_pretrained('./tf_model/my_tf_checkpoint.ckpt.index', from_tf=True, config=config)
 
         """
-        if "albert" in pretrained_model_name_or_path and "v2" in pretrained_model_name_or_path:
+        if pretrained_model_name_or_path is not None and (
+                "albert" in pretrained_model_name_or_path and "v2" in pretrained_model_name_or_path):
             logger.warning("There is currently an upstream reproducibility issue with ALBERT v2 models. Please see " +
                            "https://github.com/google-research/google-research/issues/119 for more information.")
 

From ee53de7aac8312140e87d452718e15e3d42e27dd Mon Sep 17 00:00:00 2001
From: Rosanne Liu <mimosavvy@users.noreply.github.com>
Date: Thu, 5 Dec 2019 06:20:07 -0800
Subject: [PATCH 239/505] Pr for pplm (#2060)

* license

* changes

* ok

* Update paper link and commands to run

* pointer to uber repo
---
 examples/pplm/README.md                 | 23 +++++++-----------
 examples/pplm/run_pplm.py               | 32 +++++++++----------------
 examples/pplm/run_pplm_discrim_train.py | 14 ++++++++++-
 3 files changed, 32 insertions(+), 37 deletions(-)

diff --git a/examples/pplm/README.md b/examples/pplm/README.md
index 103218ae72..b12205854a 100644
--- a/examples/pplm/README.md
+++ b/examples/pplm/README.md
@@ -1,17 +1,15 @@
-# PPLM
+# Plug and Play Language Models: a Simple Approach to Controlled Text Generation
+
+Authors: [Sumanth Dathathri](https://dathath.github.io/), [Andrea Madotto](https://andreamad8.github.io/), Janice Lan, Jane Hung, Eric Frank, [Piero Molino](https://w4nderlu.st/), [Jason Yosinski](http://yosinski.com/), and [Rosanne Liu](http://www.rosanneliu.com/)
 
 This folder contains the original code used to run the Plug and Play Language Model (PPLM).
-![header image](./imgs/headfigure.png)
 
-## Plug and Play Language Models: a Simple Approach to Steerable Text Generation
-Authors: [Sumanth Dathathri](https://dathath.github.io/), Andrea Madotto, Janice Lan, Jane Hung, Eric Frank, [Piero Molino](https://w4nderlu.st/), [Jason Yosinski](http://yosinski.com/), and [Rosanne Liu](http://www.rosanneliu.com/)
-
-PPLM allows a user to flexibly plug in one or more tiny attribute models representing the desired steering objective into a large, unconditional LM. The method has the key property that it uses the LM _as is_---no training or fine-tuning is required---which enables researchers to leverage best-in-class LMs even if they do not have the extensive hardware required to train them.
-
-Paper link: 
+Paper link: https://arxiv.org/abs/1912.02164
 
 Blog link: https://eng.uber.com/pplm
 
+Please check out the repo under uber-research for more information: https://github.com/uber-research/PPLM
+
 
 ## Setup
 
@@ -27,7 +25,7 @@ cd examples/pplm
 ### Example command for bag-of-words control
 
 ```bash
-python run_pplm.py -B space --cond_text "The president" --length 100 --gamma 1.5 --num_iterations 3 --num_samples 1 --stepsize 0.01 --window_length 5 --kl_scale 0.01 --gm_scale 0.95
+python run_pplm.py -B military --cond_text "The potato" --length 50 --gamma 1.5 --num_iterations 3 --num_samples 10 --stepsize 0.03 --window_length 5 --kl_scale 0.01 --gm_scale 0.99 --colorama --sample
 ```
 
 ### Tuning hyperparameters for bag-of-words control
@@ -45,7 +43,7 @@ python run_pplm.py -B space --cond_text "The president" --length 100 --gamma 1.5
 ### Example command for discriminator based sentiment control
 
 ```bash
-python run_pplm.py -D sentiment --class_label 3 --cond_text "The lake" --length 10 --gamma 1.0 --num_iterations 10 --num_samples 1 --stepsize 0.03 --kl_scale 0.01 --gm_scale 0.95
+python run_pplm.py -D sentiment --class_label 2 --cond_text "My dog died" --length 50 --gamma 1.0 --num_iterations 10 --num_samples 10 --stepsize 0.04 --kl_scale 0.01 --gm_scale 0.95 --sample
 ```
 
 ### Tuning hyperparameters for discriminator control
@@ -54,8 +52,3 @@ python run_pplm.py -D sentiment --class_label 3 --cond_text "The lake" --length
 
 2. Use `--class_label 3` for negative, and `--class_label 2` for positive
 
-### Example command for detoxificiation:
-
-```bash
-python run_pplm.py -D toxicity --length 100 --num_iterations 10 --cond-text 'TH PEOPLEMan goddreams Blacks' --gamma 1.0 --num_samples 10 --stepsize 0.02
-```
diff --git a/examples/pplm/run_pplm.py b/examples/pplm/run_pplm.py
index dda5d85ae7..095dc39a74 100644
--- a/examples/pplm/run_pplm.py
+++ b/examples/pplm/run_pplm.py
@@ -1,18 +1,19 @@
 #! /usr/bin/env python3
 # coding=utf-8
-# Copyright 2018 The Uber AI Team Authors.
+
+#Copyright (c) 2019 Uber Technologies, Inc.
 #
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
+#Licensed under the Apache License, Version 2.0 (the "License");
+#you may not use this file except in compliance with the License.
+#You may obtain a copy of the License at
 #
-#     http://www.apache.org/licenses/LICENSE-2.0
+#http://www.apache.org/licenses/LICENSE-2.0
 #
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
+#Unless required by applicable law or agreed to in writing, software
+#distributed under the License is distributed on an "AS IS" BASIS,
+#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+#See the License for the specific language governing permissions and
+#limitations under the License.
 
 """
 Example command with bag of words:
@@ -45,12 +46,9 @@ SMALL_CONST = 1e-15
 BIG_CONST = 1e10
 
 BAG_OF_WORDS_ARCHIVE_MAP = {
-    'kitchen': "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/bow/kitchen.txt",
     'legal': "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/bow/legal.txt",
     'military': "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/bow/military.txt",
-    'monsters': "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/bow/monsters.txt",
     'politics': "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/bow/politics.txt",
-    'positive_words': "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/bow/positive_words.txt",
     'religion': "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/bow/religion.txt",
     'science': "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/bow/science.txt",
     'space': "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/bow/space.txt",
@@ -74,14 +72,6 @@ DISCRIMINATOR_MODELS_PARAMS = {
         "default_class": 3,
         "pretrained_model": "gpt2-medium",
     },
-    "toxicity": {
-        "url": "https://s3.amazonaws.com/models.huggingface.co/bert/pplm/discriminators/toxic_classifier_head.pt",
-        "class_size": 2,
-        "embed_size": 1024,
-        "class_vocab": {"non_toxic": 0, "toxic": 1},
-        "default_class": 0,
-        "pretrained_model": "gpt2-medium",
-    },
 }
 
 
diff --git a/examples/pplm/run_pplm_discrim_train.py b/examples/pplm/run_pplm_discrim_train.py
index 9d36b79bc4..3055139d8c 100644
--- a/examples/pplm/run_pplm_discrim_train.py
+++ b/examples/pplm/run_pplm_discrim_train.py
@@ -1,7 +1,19 @@
 #! /usr/bin/env python3
 # coding=utf-8
 
-# This code is licensed under a non-commercial license.
+#Copyright (c) 2019 Uber Technologies, Inc.
+#
+#Licensed under the Apache License, Version 2.0 (the "License");
+#you may not use this file except in compliance with the License.
+#You may obtain a copy of the License at
+#
+#http://www.apache.org/licenses/LICENSE-2.0
+#
+#Unless required by applicable law or agreed to in writing, software
+#distributed under the License is distributed on an "AS IS" BASIS,
+#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+#See the License for the specific language governing permissions and
+#limitations under the License.
 
 import argparse
 import csv

From 552c44a9b19fa4f5593ac3bc8de0e9f46495ecd0 Mon Sep 17 00:00:00 2001
From: VictorSanh <victorsanh@gmail.com>
Date: Thu, 5 Dec 2019 10:14:58 -0500
Subject: [PATCH 240/505] release distilm-bert

---
 README.md                         |  2 +-
 docs/source/pretrained_models.rst |  4 ++++
 examples/distillation/README.md   | 20 ++++++++++++++++----
 3 files changed, 21 insertions(+), 5 deletions(-)

diff --git a/README.md b/README.md
index 3173b6cf10..ddeabe08d6 100644
--- a/README.md
+++ b/README.md
@@ -131,7 +131,7 @@ At some point in the future, you'll be able to seamlessly move from pre-training
 5. **[XLNet](https://github.com/zihangdai/xlnet/)** (from Google/CMU) released with the paper [​XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://arxiv.org/abs/1906.08237) by Zhilin Yang*, Zihang Dai*, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le.
 6. **[XLM](https://github.com/facebookresearch/XLM/)** (from Facebook) released together with the paper [Cross-lingual Language Model Pretraining](https://arxiv.org/abs/1901.07291) by Guillaume Lample and Alexis Conneau.
 7. **[RoBERTa](https://github.com/pytorch/fairseq/tree/master/examples/roberta)** (from Facebook), released together with the paper a [Robustly Optimized BERT Pretraining Approach](https://arxiv.org/abs/1907.11692) by Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, Veselin Stoyanov.
-8. **[DistilBERT](https://github.com/huggingface/transformers/tree/master/examples/distillation)** (from HuggingFace), released together with the paper [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://arxiv.org/abs/1910.01108) by Victor Sanh, Lysandre Debut and Thomas Wolf. The same method has been applied to compress GPT2 into [DistilGPT2](https://github.com/huggingface/transformers/tree/master/examples/distillation).
+8. **[DistilBERT](https://github.com/huggingface/transformers/tree/master/examples/distillation)** (from HuggingFace), released together with the paper [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://arxiv.org/abs/1910.01108) by Victor Sanh, Lysandre Debut and Thomas Wolf. The same method has been applied to compress GPT2 into [DistilGPT2](https://github.com/huggingface/transformers/tree/master/examples/distillation), RoBERTa into [DistilRoBERTa](https://github.com/huggingface/transformers/tree/master/examples/distillation), Multilingual BERT into [DistilmBERT](https://github.com/huggingface/transformers/tree/master/examples/distillation) and a German version of DistilBERT.
 9. **[CTRL](https://github.com/salesforce/ctrl/)** (from Salesforce) released with the paper [CTRL: A Conditional Transformer Language Model for Controllable Generation](https://arxiv.org/abs/1909.05858) by Nitish Shirish Keskar*, Bryan McCann*, Lav R. Varshney, Caiming Xiong and Richard Socher.
 10. **[CamemBERT](https://camembert-model.fr)** (from Inria/Facebook/Sorbonne) released with the paper [CamemBERT: a Tasty French Language Model](https://arxiv.org/abs/1911.03894) by Louis Martin*, Benjamin Muller*, Pedro Javier Ortiz Suárez*, Yoann Dupont, Laurent Romary, Éric Villemonte de la Clergerie, Djamé Seddah and Benoît Sagot.
 11. **[ALBERT](https://github.com/google-research/google-research/tree/master/albert)** (from Google Research and the Toyota Technological Institute at Chicago) released with the paper [ALBERT: A Lite BERT for Self-supervised Learning of Language Representations](https://arxiv.org/abs/1909.11942), by Zhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut.
diff --git a/docs/source/pretrained_models.rst b/docs/source/pretrained_models.rst
index 0cd794a678..090cb75808 100644
--- a/docs/source/pretrained_models.rst
+++ b/docs/source/pretrained_models.rst
@@ -155,6 +155,10 @@ Here is the full list of the currently provided pretrained models together with
 |                   | ``distilbert-base-german-cased``                           | | 6-layer, 768-hidden, 12-heads, 66M parameters                                                                                       |
 |                   |                                                            | | The German DistilBERT model distilled from the German DBMDZ BERT model `bert-base-german-dbmdz-cased` checkpoint.                   |
 |                   |                                                            | (see `details <https://github.com/huggingface/transformers/tree/master/examples/distillation>`__)                                     |
+|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   | ``distilbert-base-multilingual-cased``                     | | 6-layer, 768-hidden, 12-heads, 134M parameters                                                                                      |
+|                   |                                                            | | The multilingual DistilBERT model distilled from the Multilingual BERT model `bert-base-multilingual-cased` checkpoint.             |
+|                   |                                                            | (see `details <https://github.com/huggingface/transformers/tree/master/examples/distillation>`__)                                     |
 +-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 | CTRL              | ``ctrl``                                                   | | 48-layer, 1280-hidden, 16-heads, 1.6B parameters                                                                                    |
 |                   |                                                            | | Salesforce's Large-sized CTRL English model                                                                                         |
diff --git a/examples/distillation/README.md b/examples/distillation/README.md
index c2765c28ff..24a1677db1 100644
--- a/examples/distillation/README.md
+++ b/examples/distillation/README.md
@@ -2,6 +2,8 @@
 
 This folder contains the original code used to train Distil* as well as examples showcasing how to use DistilBERT, DistilRoBERTa and DistilGPT2.
 
+**December 6th, 2019 - Update** We release **DistilmBERT**: 92% of `bert-base-multilingual-cased` on XNLI. The model supports 104 different languages listed [here](https://github.com/google-research/bert/blob/master/multilingual.md#list-of-languages).
+
 **November 19th, 2019 - Update** We release German **DistilBERT**: 98.8% of `bert-base-german-dbmdz-cased` on NER tasks.
 
 **October 23rd, 2019 - Update** We release **DistilRoBERTa**: 95% of `RoBERTa-base`'s performance on GLUE, twice as fast as RoBERTa while being 35% smaller.
@@ -17,8 +19,9 @@ Distil* is a class of compressed models that started with DistilBERT. DistilBERT
 
 We have applied the same method to other Transformer architectures and released the weights:
 - GPT2: on the [WikiText-103](https://blog.einstein.ai/the-wikitext-long-term-dependency-language-modeling-dataset/) benchmark, GPT2 reaches a perplexity on the test set of 15.0 compared to 18.5 for **DistilGPT2** (after fine-tuning on the train set).
-- RoBERTa: **DistilRoBERTa** reaches 95% of `RoBERTa-base` performance on GLUE while being twice faster and 35% smaller.
-- and more to come! 🤗🤗🤗
+- RoBERTa: **DistilRoBERTa** reaches 95% of `RoBERTa-base`'s performance on GLUE while being twice faster and 35% smaller.
+- German BERT: **German DistilBERT** reaches 99% of `bert-base-german-dbmdz-cased`'s performance on German NER (CoNLL-2003).
+- Multilingual BERT: **DistilmBERT** reaches 92% of Multilingual BERT's performance on XNLI while being twice faster and 25% smaller. The model supports 104 languages listed [here](https://github.com/google-research/bert/blob/master/multilingual.md#list-of-languages).
 
 For more information on DistilBERT, please refer to our [NeurIPS workshop paper](https://arxiv.org/abs/1910.01108).
 
@@ -29,7 +32,7 @@ Here are the results on the dev sets of GLUE:
 | BERT-base                 |  **77.6**                      | 48.9 | 84.3 | 88.6 | 89.3 | 89.5 | 71.3 | 91.7 | 91.2 | 43.7              |
 | DistilBERT                |  **76.8**                      | 49.1 | 81.8 | 90.2 | 90.2 | 89.2 | 62.9 | 92.7 | 90.7 | 44.4              |
 | ---                       |    ---                         |  --- |  --- |  --- |  --- |  --- |  --- |  --- |  --- |  ---              |
-| RoBERTa-base (reported)   |  **83.2**/**86.4**<sup>2</sup> | 63.6 | 87.6 | 90.2 | 92.8 | 91.9 | 78.7 | 94.8 | 91.2 | 57.7<sup>3</sup> |
+| RoBERTa-base (reported)   |  **83.2**/**86.4**<sup>2</sup> | 63.6 | 87.6 | 90.2 | 92.8 | 91.9 | 78.7 | 94.8 | 91.2 | 57.7<sup>3</sup>  |
 | DistilRoBERTa<sup>1</sup> |  **79.0**/**82.3**<sup>2</sup> | 59.4 | 83.9 | 86.6 | 90.8 | 89.4 | 67.9 | 92.5 | 88.3 | 52.1              |
 
 <sup>1</sup> We did not use the MNLI checkpoint for fine-tuning but directy perform transfer learning on the pre-trained DistilRoBERTa.
@@ -38,6 +41,14 @@ Here are the results on the dev sets of GLUE:
 
 <sup>3</sup> We compute this score ourselves for completeness.
 
+Here are the results on the *test* sets for 6 of the languages available in XNLI. The results are computed in the zero shot setting (trained on the English portion and evaluated on the target language portion):
+
+| Model                        | English | Spanish | Chinese | German | Arabic  | Urdu |
+| :---:                        | :---:   | :---:   | :---:   | :---:  | :---:   | :---:|
+| mBERT base cased (computed)  | 82.1    | 74.6    | 69.1    | 72.3   | 66.4    | 58.5 |
+| mBERT base uncased (reported)| 81.4    | 74.3    | 63.8    | 70.5   | 62.1    | 58.3 |
+| DistilmBERT                  | 78.2    | 69.1    | 64.0    | 66.3   | 59.1    | 54.7 |
+
 ## Setup
 
 This part of the library has only be tested with Python3.6+. There are few specific dependencies to install before launching a distillation, you can install them with the command `pip install -r requirements.txt`. 
@@ -54,7 +65,7 @@ Transformers includes five pre-trained Distil* models, currently only provided f
 - `distilbert-base-german-cased`: DistilBERT German language model pretrained on 1/2 of the data used to pretrain Bert using distillation with the supervision of the `bert-base-german-dbmdz-cased` version of German DBMDZ Bert. For NER tasks the model reaches a F1 score of 83.49 on the CoNLL-2003 test set (for comparison, `bert-base-german-dbmdz-cased` reaches a 84.52 F1 score), and a F1 score of 85.23 on the GermEval 2014 test set (`bert-base-german-dbmdz-cased` reaches a 86.89 F1 score).
 - `distilgpt2`: DistilGPT2 English language model pretrained with the supervision of `gpt2` (the smallest version of GPT2) on [OpenWebTextCorpus](https://skylion007.github.io/OpenWebTextCorpus/), a reproduction of OpenAI's WebText dataset. The model has 6 layers, 768 dimension and 12 heads, totalizing 82M parameters (compared to 124M parameters for GPT2). On average, DistilGPT2 is two times faster than GPT2.
 - `distilroberta-base`: DistilRoBERTa English language model pretrained with the supervision of `roberta-base` solely on [OpenWebTextCorpus](https://skylion007.github.io/OpenWebTextCorpus/), a reproduction of OpenAI's WebText dataset (it is ~4 times less training data than the teacher RoBERTa). The model has 6 layers, 768 dimension and 12 heads, totalizing 82M parameters (compared to 125M parameters for RoBERTa-base). On average DistilRoBERTa is twice as fast as Roberta-base.
-- and more to come! 🤗🤗🤗
+- `distilbert-base-multilingual-cased`: DistilmBERT multilingual model pretrained with the supervision of `bert-base-multilingual-cased` on the concatenation of Wikipedia in 104 different languages. The model supports the 104 languages listed [here](https://github.com/google-research/bert/blob/master/multilingual.md#list-of-languages). The model has 6 layers, 768 dimension and 12 heads, totalizing 134M parameters (compared to 177M parameters for mBERT-base). On average DistilmBERT is twice as fast as mBERT-base.
 
 Using DistilBERT is very similar to using BERT. DistilBERT share the same tokenizer as BERT's `bert-base-uncased` even though we provide a link to this tokenizer under the `DistilBertTokenizer` name to have a consistent naming between the library models.
 
@@ -70,6 +81,7 @@ last_hidden_states = outputs[0]  # The last hidden-state is the first element of
 Similarly, using the other Distil* models simply consists in calling the base classes with a different pretrained checkpoint:
 - DistilGPT2: `model = GPT2Model.from_pretrained('distilgpt2')`
 - DistilRoBERTa: `model = RobertaModel.from_pretrained('distilroberta-base')`
+- DistilmBERT: `model = DistilBertModel.from_pretrained('distilbert-base-multilingual-cased')`
 
 
 ## How to train Distil*

From 35ff345fc9df9e777b27903f11fa213e4052595b Mon Sep 17 00:00:00 2001
From: VictorSanh <victorsanh@gmail.com>
Date: Thu, 5 Dec 2019 12:07:04 -0500
Subject: [PATCH 241/505] update requirements

---
 examples/distillation/distiller.py     | 1 -
 examples/distillation/requirements.txt | 2 +-
 2 files changed, 1 insertion(+), 2 deletions(-)

diff --git a/examples/distillation/distiller.py b/examples/distillation/distiller.py
index 0442072e84..1e33190aca 100644
--- a/examples/distillation/distiller.py
+++ b/examples/distillation/distiller.py
@@ -21,7 +21,6 @@ import psutil
 import time
 from tqdm import trange, tqdm
 import numpy as np
-import psutil
 
 import torch
 import torch.nn as nn
diff --git a/examples/distillation/requirements.txt b/examples/distillation/requirements.txt
index d76273b34a..491924ee2c 100644
--- a/examples/distillation/requirements.txt
+++ b/examples/distillation/requirements.txt
@@ -3,4 +3,4 @@ tensorboard>=1.14.0
 tensorboardX==1.8
 psutil==5.6.3
 scipy==1.3.1
-transformers==2.0.0
+transformers

From 9ecd83dace3961eaa161405814b00ea595c86451 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Thu, 5 Dec 2019 14:44:57 -0500
Subject: [PATCH 242/505] Patch evaluation for impossible values + cleanup

---
 docs/source/main_classes/processors.rst |  4 ++--
 examples/run_squad.py                   | 25 +++++--------------------
 transformers/data/processors/squad.py   |  6 +++---
 transformers/tokenization_utils.py      |  2 +-
 4 files changed, 11 insertions(+), 26 deletions(-)

diff --git a/docs/source/main_classes/processors.rst b/docs/source/main_classes/processors.rst
index ce0eeb553a..e98910ae1b 100644
--- a/docs/source/main_classes/processors.rst
+++ b/docs/source/main_classes/processors.rst
@@ -55,7 +55,7 @@ Example usage
 ^^^^^^^^^^^^^^^^^^^^^^^^^
 
 An example using these processors is given in the
-`run_glue.py <https://github.com/huggingface/pytorch-transformers/blob/master/examples/run_glue.py>`__ script.
+`run_glue.py <https://github.com/huggingface/transformers/blob/master/examples/run_glue.py>`__ script.
 
 
 
@@ -132,4 +132,4 @@ Example::
 
 
 Another example using these processors is given in the
-`run_squad.py <https://github.com/huggingface/pytorch-transformers/blob/master/examples/run_squad.py>`__ script.
\ No newline at end of file
+`run_squad.py <https://github.com/huggingface/transformers/blob/master/examples/run_squad.py>`__ script.
\ No newline at end of file
diff --git a/examples/run_squad.py b/examples/run_squad.py
index 3f1b6a798f..5caff9ae4f 100644
--- a/examples/run_squad.py
+++ b/examples/run_squad.py
@@ -311,7 +311,8 @@ def load_and_cache_examples(args, tokenizer, evaluate=False, output_examples=Fal
         str(args.max_seq_length)))
     if os.path.exists(cached_features_file) and not args.overwrite_cache and not output_examples:
         logger.info("Loading features from cached file %s", cached_features_file)
-        features = torch.load(cached_features_file)
+        features_and_dataset = torch.load(cached_features_file)
+        features, dataset = features_and_dataset["features"], features_and_dataset["dataset"]
     else:
         logger.info("Creating features from dataset file at %s", input_dir)
 
@@ -330,40 +331,24 @@ def load_and_cache_examples(args, tokenizer, evaluate=False, output_examples=Fal
             processor = SquadV2Processor() if args.version_2_with_negative else SquadV1Processor()
             examples = processor.get_dev_examples(args.data_dir) if evaluate else processor.get_train_examples(args.data_dir)
 
-        features = squad_convert_examples_to_features( 
+        features, dataset = squad_convert_examples_to_features( 
             examples=examples,
             tokenizer=tokenizer,
             max_seq_length=args.max_seq_length,
             doc_stride=args.doc_stride,
             max_query_length=args.max_query_length,
             is_training=not evaluate,
+            return_dataset='pt'
         )
 
 
         if args.local_rank in [-1, 0]:
             logger.info("Saving features into cached file %s", cached_features_file)
-            torch.save(features, cached_features_file)
+            torch.save({"features": features, "dataset": dataset}, cached_features_file)
 
     if args.local_rank == 0 and not evaluate:
         torch.distributed.barrier()  # Make sure only the first process in distributed training process the dataset, and the others will use the cache
 
-    # Convert to Tensors and build dataset
-    all_input_ids = torch.tensor([f.input_ids for f in features], dtype=torch.long)
-    all_input_mask = torch.tensor([f.attention_mask for f in features], dtype=torch.long)
-    all_segment_ids = torch.tensor([f.token_type_ids for f in features], dtype=torch.long)
-    all_cls_index = torch.tensor([f.cls_index for f in features], dtype=torch.long)
-    all_p_mask = torch.tensor([f.p_mask for f in features], dtype=torch.float)
-    if evaluate:
-        all_example_index = torch.arange(all_input_ids.size(0), dtype=torch.long)
-        dataset = TensorDataset(all_input_ids, all_input_mask, all_segment_ids,
-                                all_example_index, all_cls_index, all_p_mask)
-    else:
-        all_start_positions = torch.tensor([f.start_position for f in features], dtype=torch.long)
-        all_end_positions = torch.tensor([f.end_position for f in features], dtype=torch.long)
-        dataset = TensorDataset(all_input_ids, all_input_mask, all_segment_ids,
-                                all_start_positions, all_end_positions,
-                                all_cls_index, all_p_mask)
-
     if output_examples:
         return dataset, examples, features
     return dataset
diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index 338bae0c51..bb56aa792f 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -312,7 +312,7 @@ class SquadProcessor(DataProcessor):
         if not evaluate:
             answer = tensor_dict['answers']['text'][0].numpy().decode('utf-8')
             answer_start = tensor_dict['answers']['answer_start'][0].numpy()
-            answers = None
+            answers = []
         else:
             answers = [{
                 "answer_start": start.numpy(), 
@@ -408,7 +408,7 @@ class SquadProcessor(DataProcessor):
                     question_text = qa["question"]
                     start_position_character = None
                     answer_text = None
-                    answers = None
+                    answers = []
                     
                     if "is_impossible" in qa:
                         is_impossible = qa["is_impossible"]
@@ -469,7 +469,7 @@ class SquadExample(object):
                  answer_text,
                  start_position_character,
                  title,
-                 answers=None,
+                 answers=[],
                  is_impossible=False):
         self.qas_id = qas_id
         self.question_text = question_text
diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index 41a611ea49..5ec173bbf6 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -194,7 +194,7 @@ class PreTrainedTokenizer(object):
 
     @property
     def pad_token_type_id(self):
-        """ Id of the padding token in the vocabulary. Log an error if used while not having been set. """
+        """ Id of the padding token type in the vocabulary."""
         return self._pad_token_type_id
 
     @property

From e9217da5ff711cf84d150b35d3f8a5c17f1641f7 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Thu, 5 Dec 2019 16:01:51 -0500
Subject: [PATCH 243/505] Cleanup

Improve global visibility on the run_squad script, remove unused files and fixes related to XLNet.
---
 examples/run_squad.py                      |   69 +-
 examples/utils_squad.py                    | 1017 --------------------
 examples/utils_squad_evaluate.py           |  330 -------
 transformers/data/metrics/squad_metrics.py |   14 +-
 transformers/data/processors/squad.py      |    2 +-
 5 files changed, 45 insertions(+), 1387 deletions(-)
 delete mode 100644 examples/utils_squad.py
 delete mode 100644 examples/utils_squad_evaluate.py

diff --git a/examples/run_squad.py b/examples/run_squad.py
index 5caff9ae4f..6d32211c0c 100644
--- a/examples/run_squad.py
+++ b/examples/run_squad.py
@@ -27,8 +27,7 @@ import glob
 import timeit
 import numpy as np
 import torch
-from torch.utils.data import (DataLoader, RandomSampler, SequentialSampler,
-                              TensorDataset)
+from torch.utils.data import (DataLoader, RandomSampler, SequentialSampler, TensorDataset)
 from torch.utils.data.distributed import DistributedSampler
 
 try:
@@ -48,14 +47,6 @@ from transformers import (WEIGHTS_NAME, BertConfig,
 
 from transformers import AdamW, get_linear_schedule_with_warmup, squad_convert_examples_to_features
 
-from utils_squad import (convert_examples_to_features as old_convert, read_squad_examples as old_read, RawResult, write_predictions,
-                         RawResultExtended, write_predictions_extended)
-
-# The follwing import is the official SQuAD evaluation script (2.0).
-# You can remove it from the dependencies if you are using this script outside of the library
-# We've added it here for automated tests (see examples/test_examples.py file)
-from utils_squad_evaluate import EVAL_OPTS, main as evaluate_on_squad
-
 logger = logging.getLogger(__name__)
 
 ALL_MODELS = sum((tuple(conf.pretrained_config_archive_map.keys()) \
@@ -98,14 +89,16 @@ def train(args, train_dataset, model, tokenizer):
     optimizer_grouped_parameters = [
         {'params': [p for n, p in model.named_parameters() if not any(nd in n for nd in no_decay)], 'weight_decay': args.weight_decay},
         {'params': [p for n, p in model.named_parameters() if any(nd in n for nd in no_decay)], 'weight_decay': 0.0}
-        ]
+    ]
     optimizer = AdamW(optimizer_grouped_parameters, lr=args.learning_rate, eps=args.adam_epsilon)
     scheduler = get_linear_schedule_with_warmup(optimizer, num_warmup_steps=args.warmup_steps, num_training_steps=t_total)
+
     if args.fp16:
         try:
             from apex import amp
         except ImportError:
             raise ImportError("Please install apex from https://www.github.com/nvidia/apex to use fp16 training.")
+        
         model, optimizer = amp.initialize(model, optimizer, opt_level=args.fp16_opt_level)
 
     # multi-gpu training (should be after apex fp16 initialization)
@@ -133,20 +126,26 @@ def train(args, train_dataset, model, tokenizer):
     model.zero_grad()
     train_iterator = trange(int(args.num_train_epochs), desc="Epoch", disable=args.local_rank not in [-1, 0])
     set_seed(args)  # Added here for reproductibility (even between python 2 and 3)
+    
     for _ in train_iterator:
         epoch_iterator = tqdm(train_dataloader, desc="Iteration", disable=args.local_rank not in [-1, 0])
         for step, batch in enumerate(epoch_iterator):
             model.train()
             batch = tuple(t.to(args.device) for t in batch)
-            inputs = {'input_ids':       batch[0],
-                      'attention_mask':  batch[1],
-                      'start_positions': batch[3],
-                      'end_positions':   batch[4]}
+
+            inputs = {
+                'input_ids':       batch[0],
+                'attention_mask':  batch[1],
+                'start_positions': batch[3],
+                'end_positions':   batch[4]
+            }
+
             if args.model_type != 'distilbert':
                 inputs['token_type_ids'] = None if args.model_type == 'xlm' else batch[2]
+
             if args.model_type in ['xlnet', 'xlm']:
-                inputs.update({'cls_index': batch[5],
-                               'p_mask':       batch[6]})
+                inputs.update({'cls_index': batch[5], 'p_mask': batch[6]})
+
             outputs = model(**inputs)
             loss = outputs[0]  # model outputs are always tuple in transformers (see doc)
 
@@ -173,8 +172,8 @@ def train(args, train_dataset, model, tokenizer):
                 model.zero_grad()
                 global_step += 1
 
+                # Log metrics
                 if args.local_rank in [-1, 0] and args.logging_steps > 0 and global_step % args.logging_steps == 0:
-                    # Log metrics
                     if args.local_rank == -1 and args.evaluate_during_training:  # Only evaluate when single GPU otherwise metrics may not average well
                         results = evaluate(args, model, tokenizer)
                         for key, value in results.items():
@@ -183,8 +182,8 @@ def train(args, train_dataset, model, tokenizer):
                     tb_writer.add_scalar('loss', (tr_loss - logging_loss)/args.logging_steps, global_step)
                     logging_loss = tr_loss
 
+                # Save model checkpoint
                 if args.local_rank in [-1, 0] and args.save_steps > 0 and global_step % args.save_steps == 0:
-                    # Save model checkpoint
                     output_dir = os.path.join(args.output_dir, 'checkpoint-{}'.format(global_step))
                     if not os.path.exists(output_dir):
                         os.makedirs(output_dir)
@@ -213,6 +212,7 @@ def evaluate(args, model, tokenizer, prefix=""):
         os.makedirs(args.output_dir)
 
     args.eval_batch_size = args.per_gpu_eval_batch_size * max(1, args.n_gpu)
+
     # Note that DistributedSampler samples randomly
     eval_sampler = SequentialSampler(dataset) if args.local_rank == -1 else DistributedSampler(dataset)
     eval_dataloader = DataLoader(dataset, sampler=eval_sampler, batch_size=args.eval_batch_size)
@@ -225,11 +225,14 @@ def evaluate(args, model, tokenizer, prefix=""):
     logger.info("***** Running evaluation {} *****".format(prefix))
     logger.info("  Num examples = %d", len(dataset))
     logger.info("  Batch size = %d", args.eval_batch_size)
+
     all_results = []
     start_time = timeit.default_timer()
+
     for batch in tqdm(eval_dataloader, desc="Evaluating"):
         model.eval()
         batch = tuple(t.to(args.device) for t in batch)
+
         with torch.no_grad():
             inputs = {
                 'input_ids':      batch[0],
@@ -238,10 +241,13 @@ def evaluate(args, model, tokenizer, prefix=""):
             
             if args.model_type != 'distilbert':
                 inputs['token_type_ids'] = None if args.model_type == 'xlm' else batch[2]  # XLM don't use segment_ids
+
             example_indices = batch[3]
+            
+            # XLNet and XLM use more arguments for their predictions
             if args.model_type in ['xlnet', 'xlm']:
-                inputs.update({'cls_index': batch[4],
-                               'p_mask':    batch[5]})
+                inputs.update({'cls_index': batch[4], 'p_mask': batch[5]})
+
             outputs = model(**inputs)
 
         for i, example_index in enumerate(example_indices):
@@ -250,11 +256,13 @@ def evaluate(args, model, tokenizer, prefix=""):
 
             output = [to_list(output[i]) for output in outputs]
 
+            # Some models (XLNet, XLM) use 5 arguments for their predictions, while the other "simpler"
+            # models only use two.
             if len(output) >= 5:
                 start_logits = output[0]
                 start_top_index = output[1]
                 end_logits = output[2]
-                end_top_index = output[3],
+                end_top_index = output[3]
                 cls_logits = output[4]
 
                 result = SquadResult(
@@ -278,16 +286,17 @@ def evaluate(args, model, tokenizer, prefix=""):
     # Compute predictions
     output_prediction_file = os.path.join(args.output_dir, "predictions_{}.json".format(prefix))
     output_nbest_file = os.path.join(args.output_dir, "nbest_predictions_{}.json".format(prefix))
+
     if args.version_2_with_negative:
         output_null_log_odds_file = os.path.join(args.output_dir, "null_odds_{}.json".format(prefix))
     else:
         output_null_log_odds_file = None
 
+    # XLNet and XLM use a more complex post-processing procedure
     if args.model_type in ['xlnet', 'xlm']:
-        # XLNet uses a more complex post-processing procedure
         predictions = compute_predictions_log_probs(examples, features, all_results, args.n_best_size,
                         args.max_answer_length, output_prediction_file,
-                        output_nbest_file, output_null_log_odds_file, args.predict_file,
+                        output_nbest_file, output_null_log_odds_file,
                         model.config.start_n_top, model.config.end_n_top,
                         args.version_2_with_negative, tokenizer, args.verbose_logging)
     else:
@@ -296,6 +305,7 @@ def evaluate(args, model, tokenizer, prefix=""):
                         output_nbest_file, output_null_log_odds_file, args.verbose_logging,
                         args.version_2_with_negative, args.null_score_diff_threshold)
 
+    # Compute the F1 and exact scores.
     results = squad_evaluate(examples, predictions)
     return results
 
@@ -308,7 +318,10 @@ def load_and_cache_examples(args, tokenizer, evaluate=False, output_examples=Fal
     cached_features_file = os.path.join(input_dir, 'cached_{}_{}_{}'.format(
         'dev' if evaluate else 'train',
         list(filter(None, args.model_name_or_path.split('/'))).pop(),
-        str(args.max_seq_length)))
+        str(args.max_seq_length))
+    )
+
+    # Init features and dataset from cache if it exists
     if os.path.exists(cached_features_file) and not args.overwrite_cache and not output_examples:
         logger.info("Loading features from cached file %s", cached_features_file)
         features_and_dataset = torch.load(cached_features_file)
@@ -341,7 +354,6 @@ def load_and_cache_examples(args, tokenizer, evaluate=False, output_examples=Fal
             return_dataset='pt'
         )
 
-
         if args.local_rank in [-1, 0]:
             logger.info("Saving features into cached file %s", cached_features_file)
             torch.save({"features": features, "dataset": dataset}, cached_features_file)
@@ -452,6 +464,11 @@ def main():
     parser.add_argument('--server_port', type=str, default='', help="Can be used for distant debugging.")
     args = parser.parse_args()
 
+    args.predict_file = os.path.join(args.output_dir, 'predictions_{}_{}.txt'.format(
+        list(filter(None, args.model_name_or_path.split('/'))).pop(),
+        str(args.max_seq_length))
+    )
+
     if os.path.exists(args.output_dir) and os.listdir(args.output_dir) and args.do_train and not args.overwrite_output_dir:
         raise ValueError("Output directory ({}) already exists and is not empty. Use --overwrite_output_dir to overcome.".format(args.output_dir))
 
diff --git a/examples/utils_squad.py b/examples/utils_squad.py
deleted file mode 100644
index 4f1c581588..0000000000
--- a/examples/utils_squad.py
+++ /dev/null
@@ -1,1017 +0,0 @@
-
-# coding=utf-8
-# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.
-# Copyright (c) 2018, NVIDIA CORPORATION.  All rights reserved.
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-""" Load SQuAD dataset. """
-
-from __future__ import absolute_import, division, print_function
-
-import json
-import logging
-import math
-import collections
-from io import open
-from tqdm import tqdm
-
-from transformers.tokenization_bert import BasicTokenizer, whitespace_tokenize
-
-# Required by XLNet evaluation method to compute optimal threshold (see write_predictions_extended() method)
-from utils_squad_evaluate import find_all_best_thresh_v2, make_qid_to_has_ans, get_raw_scores
-
-logger = logging.getLogger(__name__)
-
-
-class SquadExample(object):
-    """
-    A single training/test example for the Squad dataset.
-    For examples without an answer, the start and end position are -1.
-    """
-
-    def __init__(self,
-                 qas_id,
-                 question_text,
-                 doc_tokens,
-                 orig_answer_text=None,
-                 start_position=None,
-                 end_position=None,
-                 is_impossible=None):
-        self.qas_id = qas_id
-        self.question_text = question_text
-        self.doc_tokens = doc_tokens
-        self.orig_answer_text = orig_answer_text
-        self.start_position = start_position
-        self.end_position = end_position
-        self.is_impossible = is_impossible
-
-    def __str__(self):
-        return self.__repr__()
-
-    def __repr__(self):
-        s = ""
-        s += "qas_id: %s" % (self.qas_id)
-        s += ", question_text: %s" % (
-            self.question_text)
-        s += ", doc_tokens: [%s]" % (" ".join(self.doc_tokens))
-        if self.start_position:
-            s += ", start_position: %d" % (self.start_position)
-        if self.end_position:
-            s += ", end_position: %d" % (self.end_position)
-        if self.is_impossible:
-            s += ", is_impossible: %r" % (self.is_impossible)
-        return s
-
-
-class InputFeatures(object):
-    """A single set of features of data."""
-
-    def __init__(self,
-                 unique_id,
-                 example_index,
-                 doc_span_index,
-                 tokens,
-                 token_to_orig_map,
-                 token_is_max_context,
-                 input_ids,
-                 input_mask,
-                 segment_ids,
-                 cls_index,
-                 p_mask,
-                 paragraph_len,
-                 start_position=None,
-                 end_position=None,
-                 is_impossible=None):
-        self.unique_id = unique_id
-        self.example_index = example_index
-        self.doc_span_index = doc_span_index
-        self.tokens = tokens
-        self.token_to_orig_map = token_to_orig_map
-        self.token_is_max_context = token_is_max_context
-        self.input_ids = input_ids
-        self.input_mask = input_mask
-        self.segment_ids = segment_ids
-        self.cls_index = cls_index
-        self.p_mask = p_mask
-        self.paragraph_len = paragraph_len
-        self.start_position = start_position
-        self.end_position = end_position
-        self.is_impossible = is_impossible
-
-
-def read_squad_examples(input_file, is_training, version_2_with_negative):
-    """Read a SQuAD json file into a list of SquadExample."""
-    with open(input_file, "r", encoding='utf-8') as reader:
-        input_data = json.load(reader)["data"]
-
-    def is_whitespace(c):
-        if c == " " or c == "\t" or c == "\r" or c == "\n" or ord(c) == 0x202F:
-            return True
-        return False
-
-    examples = []
-    for entry in input_data:
-        for paragraph in entry["paragraphs"]:
-            paragraph_text = paragraph["context"]
-            doc_tokens = []
-            char_to_word_offset = []
-            prev_is_whitespace = True
-            for c in paragraph_text:
-                if is_whitespace(c):
-                    prev_is_whitespace = True
-                else:
-                    if prev_is_whitespace:
-                        doc_tokens.append(c)
-                    else:
-                        doc_tokens[-1] += c
-                    prev_is_whitespace = False
-                char_to_word_offset.append(len(doc_tokens) - 1)
-
-            for qa in paragraph["qas"]:
-                qas_id = qa["id"]
-                question_text = qa["question"]
-                start_position = None
-                end_position = None
-                orig_answer_text = None
-                is_impossible = False
-                if is_training:
-                    if version_2_with_negative:
-                        is_impossible = qa["is_impossible"]
-                    if (len(qa["answers"]) != 1) and (not is_impossible):
-                        raise ValueError(
-                            "For training, each question should have exactly 1 answer.")
-                    if not is_impossible:
-                        answer = qa["answers"][0]
-                        orig_answer_text = answer["text"]
-                        answer_offset = answer["answer_start"]
-                        answer_length = len(orig_answer_text)
-                        start_position = char_to_word_offset[answer_offset]
-                        end_position = char_to_word_offset[answer_offset + answer_length - 1]
-                        # Only add answers where the text can be exactly recovered from the
-                        # document. If this CAN'T happen it's likely due to weird Unicode
-                        # stuff so we will just skip the example.
-                        #
-                        # Note that this means for training mode, every example is NOT
-                        # guaranteed to be preserved.
-                        actual_text = " ".join(doc_tokens[start_position:(end_position + 1)])
-                        cleaned_answer_text = " ".join(
-                            whitespace_tokenize(orig_answer_text))
-                        if actual_text.find(cleaned_answer_text) == -1:
-                            logger.warning("Could not find answer: '%s' vs. '%s'",
-                                           actual_text, cleaned_answer_text)
-                            continue
-                    else:
-                        start_position = -1
-                        end_position = -1
-                        orig_answer_text = ""
-
-                example = SquadExample(
-                    qas_id=qas_id,
-                    question_text=question_text,
-                    doc_tokens=doc_tokens,
-                    orig_answer_text=orig_answer_text,
-                    start_position=start_position,
-                    end_position=end_position,
-                    is_impossible=is_impossible)
-                examples.append(example)
-    return examples
-
-
-def convert_examples_to_features(examples, tokenizer, max_seq_length,
-                                 doc_stride, max_query_length, is_training,
-                                 cls_token_at_end=False,
-                                 cls_token='[CLS]', sep_token='[SEP]', pad_token=0,
-                                 sequence_a_segment_id=0, sequence_b_segment_id=1,
-                                 cls_token_segment_id=0, pad_token_segment_id=0,
-                                 mask_padding_with_zero=True,
-                                 sequence_a_is_doc=False):
-    """Loads a data file into a list of `InputBatch`s."""
-
-    unique_id = 1000000000
-    # cnt_pos, cnt_neg = 0, 0
-    # max_N, max_M = 1024, 1024
-    # f = np.zeros((max_N, max_M), dtype=np.float32)
-
-    features = []
-    for (example_index, example) in enumerate(tqdm(examples)):
-
-        # if example_index % 100 == 0:
-        #     logger.info('Converting %s/%s pos %s neg %s', example_index, len(examples), cnt_pos, cnt_neg)
-
-        query_tokens = tokenizer.tokenize(example.question_text)
-
-        if len(query_tokens) > max_query_length:
-            query_tokens = query_tokens[0:max_query_length]
-
-        tok_to_orig_index = []
-        orig_to_tok_index = []
-        all_doc_tokens = []
-        for (i, token) in enumerate(example.doc_tokens):
-            orig_to_tok_index.append(len(all_doc_tokens))
-            sub_tokens = tokenizer.tokenize(token)
-            for sub_token in sub_tokens:
-                tok_to_orig_index.append(i)
-                all_doc_tokens.append(sub_token)
-
-        tok_start_position = None
-        tok_end_position = None
-        if is_training and example.is_impossible:
-            tok_start_position = -1
-            tok_end_position = -1
-        if is_training and not example.is_impossible:
-            tok_start_position = orig_to_tok_index[example.start_position]
-            if example.end_position < len(example.doc_tokens) - 1:
-                tok_end_position = orig_to_tok_index[example.end_position + 1] - 1
-            else:
-                tok_end_position = len(all_doc_tokens) - 1
-            (tok_start_position, tok_end_position) = _improve_answer_span(
-                all_doc_tokens, tok_start_position, tok_end_position, tokenizer,
-                example.orig_answer_text)
-
-        # The -3 accounts for [CLS], [SEP] and [SEP]
-        max_tokens_for_doc = max_seq_length - len(query_tokens) - 3
-        assert max_tokens_for_doc > 0
-
-        # We can have documents that are longer than the maximum sequence length.
-        # To deal with this we do a sliding window approach, where we take chunks
-        # of the up to our max length with a stride of `doc_stride`.
-        _DocSpan = collections.namedtuple(  # pylint: disable=invalid-name
-            "DocSpan", ["start", "length"])
-        doc_spans = []
-        start_offset = 0
-        while start_offset < len(all_doc_tokens):
-            length = len(all_doc_tokens) - start_offset
-            if length > max_tokens_for_doc:
-                length = max_tokens_for_doc
-            doc_spans.append(_DocSpan(start=start_offset, length=length))
-            if start_offset + length == len(all_doc_tokens):
-                break
-            start_offset += min(length, doc_stride)
-
-        for (doc_span_index, doc_span) in enumerate(doc_spans):
-            tokens = []
-            token_to_orig_map = {}
-            token_is_max_context = {}
-            segment_ids = []
-
-            # p_mask: mask with 1 for token than cannot be in the answer (0 for token which can be in an answer)
-            # Original TF implem also keep the classification token (set to 0) (not sure why...)
-            p_mask = []
-
-            # CLS token at the beginning
-            if not cls_token_at_end:
-                tokens.append(cls_token)
-                segment_ids.append(cls_token_segment_id)
-                p_mask.append(0)
-                cls_index = 0
-
-            # XLNet: P SEP Q SEP CLS
-            # Others: CLS Q SEP P SEP
-            if not sequence_a_is_doc:
-                # Query
-                tokens += query_tokens
-                segment_ids += [sequence_a_segment_id] * len(query_tokens)
-                p_mask += [1] * len(query_tokens)
-
-                # SEP token
-                tokens.append(sep_token)
-                segment_ids.append(sequence_a_segment_id)
-                p_mask.append(1)
-
-            # Paragraph
-            for i in range(doc_span.length):
-                split_token_index = doc_span.start + i
-                token_to_orig_map[len(tokens)] = tok_to_orig_index[split_token_index]
-
-                is_max_context = _check_is_max_context(doc_spans, doc_span_index,
-                                                       split_token_index)
-                token_is_max_context[len(tokens)] = is_max_context
-                tokens.append(all_doc_tokens[split_token_index])
-                if not sequence_a_is_doc:
-                    segment_ids.append(sequence_b_segment_id)
-                else:
-                    segment_ids.append(sequence_a_segment_id)
-                p_mask.append(0)
-            paragraph_len = doc_span.length
-
-            if sequence_a_is_doc:
-                # SEP token
-                tokens.append(sep_token)
-                segment_ids.append(sequence_a_segment_id)
-                p_mask.append(1)
-
-                tokens += query_tokens
-                segment_ids += [sequence_b_segment_id] * len(query_tokens)
-                p_mask += [1] * len(query_tokens)
-
-            # SEP token
-            tokens.append(sep_token)
-            segment_ids.append(sequence_b_segment_id)
-            p_mask.append(1)
-
-            # CLS token at the end
-            if cls_token_at_end:
-                tokens.append(cls_token)
-                segment_ids.append(cls_token_segment_id)
-                p_mask.append(0)
-                cls_index = len(tokens) - 1  # Index of classification token
-
-            input_ids = tokenizer.convert_tokens_to_ids(tokens)
-
-            # The mask has 1 for real tokens and 0 for padding tokens. Only real
-            # tokens are attended to.
-            input_mask = [1 if mask_padding_with_zero else 0] * len(input_ids)
-
-            # Zero-pad up to the sequence length.
-            while len(input_ids) < max_seq_length:
-                input_ids.append(pad_token)
-                input_mask.append(0 if mask_padding_with_zero else 1)
-                segment_ids.append(pad_token_segment_id)
-                p_mask.append(1)
-
-            assert len(input_ids) == max_seq_length
-            assert len(input_mask) == max_seq_length
-            assert len(segment_ids) == max_seq_length
-
-            span_is_impossible = example.is_impossible
-            start_position = None
-            end_position = None
-            if is_training and not span_is_impossible:
-                # For training, if our document chunk does not contain an annotation
-                # we throw it out, since there is nothing to predict.
-                doc_start = doc_span.start
-                doc_end = doc_span.start + doc_span.length - 1
-                out_of_span = False
-                if not (tok_start_position >= doc_start and
-                        tok_end_position <= doc_end):
-                    out_of_span = True
-                if out_of_span:
-                    start_position = 0
-                    end_position = 0
-                    span_is_impossible = True
-                else:
-                    if sequence_a_is_doc:
-                        doc_offset = 0
-                    else:
-                        doc_offset = len(query_tokens) + 2
-                    start_position = tok_start_position - doc_start + doc_offset
-                    end_position = tok_end_position - doc_start + doc_offset
-
-            if is_training and span_is_impossible:
-                start_position = cls_index
-                end_position = cls_index
-
-            if example_index < 20:
-                logger.info("*** Example ***")
-                logger.info("unique_id: %s" % (unique_id))
-                logger.info("example_index: %s" % (example_index))
-                logger.info("doc_span_index: %s" % (doc_span_index))
-                logger.info("tokens: %s" % " ".join(tokens))
-                logger.info("token_to_orig_map: %s" % " ".join([
-                    "%d:%d" % (x, y) for (x, y) in token_to_orig_map.items()]))
-                logger.info("token_is_max_context: %s" % " ".join([
-                    "%d:%s" % (x, y) for (x, y) in token_is_max_context.items()
-                ]))
-                logger.info("input_ids: %s" % " ".join([str(x) for x in input_ids]))
-                logger.info(
-                    "input_mask: %s" % " ".join([str(x) for x in input_mask]))
-                logger.info(
-                    "segment_ids: %s" % " ".join([str(x) for x in segment_ids]))
-                if is_training and span_is_impossible:
-                    logger.info("impossible example")
-                if is_training and not span_is_impossible:
-                    answer_text = " ".join(tokens[start_position:(end_position + 1)])
-                    logger.info("start_position: %d" % (start_position))
-                    logger.info("end_position: %d" % (end_position))
-                    logger.info(
-                        "answer: %s" % (answer_text))
-
-            features.append(
-                InputFeatures(
-                    unique_id=unique_id,
-                    example_index=example_index,
-                    doc_span_index=doc_span_index,
-                    tokens=tokens,
-                    token_to_orig_map=token_to_orig_map,
-                    token_is_max_context=token_is_max_context,
-                    input_ids=input_ids,
-                    input_mask=input_mask,
-                    segment_ids=segment_ids,
-                    cls_index=cls_index,
-                    p_mask=p_mask,
-                    paragraph_len=paragraph_len,
-                    start_position=start_position,
-                    end_position=end_position,
-                    is_impossible=span_is_impossible))
-            unique_id += 1
-
-    return features
-
-
-def _improve_answer_span(doc_tokens, input_start, input_end, tokenizer,
-                         orig_answer_text):
-    """Returns tokenized answer spans that better match the annotated answer."""
-
-    # The SQuAD annotations are character based. We first project them to
-    # whitespace-tokenized words. But then after WordPiece tokenization, we can
-    # often find a "better match". For example:
-    #
-    #   Question: What year was John Smith born?
-    #   Context: The leader was John Smith (1895-1943).
-    #   Answer: 1895
-    #
-    # The original whitespace-tokenized answer will be "(1895-1943).". However
-    # after tokenization, our tokens will be "( 1895 - 1943 ) .". So we can match
-    # the exact answer, 1895.
-    #
-    # However, this is not always possible. Consider the following:
-    #
-    #   Question: What country is the top exporter of electornics?
-    #   Context: The Japanese electronics industry is the lagest in the world.
-    #   Answer: Japan
-    #
-    # In this case, the annotator chose "Japan" as a character sub-span of
-    # the word "Japanese". Since our WordPiece tokenizer does not split
-    # "Japanese", we just use "Japanese" as the annotation. This is fairly rare
-    # in SQuAD, but does happen.
-    tok_answer_text = " ".join(tokenizer.tokenize(orig_answer_text))
-
-    for new_start in range(input_start, input_end + 1):
-        for new_end in range(input_end, new_start - 1, -1):
-            text_span = " ".join(doc_tokens[new_start:(new_end + 1)])
-            if text_span == tok_answer_text:
-                return (new_start, new_end)
-
-    return (input_start, input_end)
-
-
-def _check_is_max_context(doc_spans, cur_span_index, position):
-    """Check if this is the 'max context' doc span for the token."""
-
-    # Because of the sliding window approach taken to scoring documents, a single
-    # token can appear in multiple documents. E.g.
-    #  Doc: the man went to the store and bought a gallon of milk
-    #  Span A: the man went to the
-    #  Span B: to the store and bought
-    #  Span C: and bought a gallon of
-    #  ...
-    #
-    # Now the word 'bought' will have two scores from spans B and C. We only
-    # want to consider the score with "maximum context", which we define as
-    # the *minimum* of its left and right context (the *sum* of left and
-    # right context will always be the same, of course).
-    #
-    # In the example the maximum context for 'bought' would be span C since
-    # it has 1 left context and 3 right context, while span B has 4 left context
-    # and 0 right context.
-    best_score = None
-    best_span_index = None
-    for (span_index, doc_span) in enumerate(doc_spans):
-        end = doc_span.start + doc_span.length - 1
-        if position < doc_span.start:
-            continue
-        if position > end:
-            continue
-        num_left_context = position - doc_span.start
-        num_right_context = end - position
-        score = min(num_left_context, num_right_context) + 0.01 * doc_span.length
-        if best_score is None or score > best_score:
-            best_score = score
-            best_span_index = span_index
-
-    return cur_span_index == best_span_index
-
-
-RawResult = collections.namedtuple("RawResult",
-                                   ["unique_id", "start_logits", "end_logits"])
-
-def write_predictions(all_examples, all_features, all_results, n_best_size,
-                      max_answer_length, do_lower_case, output_prediction_file,
-                      output_nbest_file, output_null_log_odds_file, verbose_logging,
-                      version_2_with_negative, null_score_diff_threshold):
-    """Write final predictions to the json file and log-odds of null if needed."""
-    logger.info("Writing predictions to: %s" % (output_prediction_file))
-    logger.info("Writing nbest to: %s" % (output_nbest_file))
-
-    example_index_to_features = collections.defaultdict(list)
-    for feature in all_features:
-        example_index_to_features[feature.example_index].append(feature)
-
-    unique_id_to_result = {}
-    for result in all_results:
-        unique_id_to_result[result.unique_id] = result
-
-    _PrelimPrediction = collections.namedtuple(  # pylint: disable=invalid-name
-        "PrelimPrediction",
-        ["feature_index", "start_index", "end_index", "start_logit", "end_logit"])
-
-    all_predictions = collections.OrderedDict()
-    all_nbest_json = collections.OrderedDict()
-    scores_diff_json = collections.OrderedDict()
-
-    for (example_index, example) in enumerate(all_examples):
-        features = example_index_to_features[example_index]
-
-        prelim_predictions = []
-        # keep track of the minimum score of null start+end of position 0
-        score_null = 1000000  # large and positive
-        min_null_feature_index = 0  # the paragraph slice with min null score
-        null_start_logit = 0  # the start logit at the slice with min null score
-        null_end_logit = 0  # the end logit at the slice with min null score
-        for (feature_index, feature) in enumerate(features):
-            result = unique_id_to_result[feature.unique_id]
-            start_indexes = _get_best_indexes(result.start_logits, n_best_size)
-            end_indexes = _get_best_indexes(result.end_logits, n_best_size)
-            # if we could have irrelevant answers, get the min score of irrelevant
-            if version_2_with_negative:
-                feature_null_score = result.start_logits[0] + result.end_logits[0]
-                if feature_null_score < score_null:
-                    score_null = feature_null_score
-                    min_null_feature_index = feature_index
-                    null_start_logit = result.start_logits[0]
-                    null_end_logit = result.end_logits[0]
-            for start_index in start_indexes:
-                for end_index in end_indexes:
-                    # We could hypothetically create invalid predictions, e.g., predict
-                    # that the start of the span is in the question. We throw out all
-                    # invalid predictions.
-                    if start_index >= len(feature.tokens):
-                        continue
-                    if end_index >= len(feature.tokens):
-                        continue
-                    if start_index not in feature.token_to_orig_map:
-                        continue
-                    if end_index not in feature.token_to_orig_map:
-                        continue
-                    if not feature.token_is_max_context.get(start_index, False):
-                        continue
-                    if end_index < start_index:
-                        continue
-                    length = end_index - start_index + 1
-                    if length > max_answer_length:
-                        continue
-                    prelim_predictions.append(
-                        _PrelimPrediction(
-                            feature_index=feature_index,
-                            start_index=start_index,
-                            end_index=end_index,
-                            start_logit=result.start_logits[start_index],
-                            end_logit=result.end_logits[end_index]))
-        if version_2_with_negative:
-            prelim_predictions.append(
-                _PrelimPrediction(
-                    feature_index=min_null_feature_index,
-                    start_index=0,
-                    end_index=0,
-                    start_logit=null_start_logit,
-                    end_logit=null_end_logit))
-        prelim_predictions = sorted(
-            prelim_predictions,
-            key=lambda x: (x.start_logit + x.end_logit),
-            reverse=True)
-
-        _NbestPrediction = collections.namedtuple(  # pylint: disable=invalid-name
-            "NbestPrediction", ["text", "start_logit", "end_logit"])
-
-        seen_predictions = {}
-        nbest = []
-        for pred in prelim_predictions:
-            if len(nbest) >= n_best_size:
-                break
-            feature = features[pred.feature_index]
-            if pred.start_index > 0:  # this is a non-null prediction
-                tok_tokens = feature.tokens[pred.start_index:(pred.end_index + 1)]
-                orig_doc_start = feature.token_to_orig_map[pred.start_index]
-                orig_doc_end = feature.token_to_orig_map[pred.end_index]
-                orig_tokens = example.doc_tokens[orig_doc_start:(orig_doc_end + 1)]
-                tok_text = " ".join(tok_tokens)
-
-                # De-tokenize WordPieces that have been split off.
-                tok_text = tok_text.replace(" ##", "")
-                tok_text = tok_text.replace("##", "")
-
-                # Clean whitespace
-                tok_text = tok_text.strip()
-                tok_text = " ".join(tok_text.split())
-                orig_text = " ".join(orig_tokens)
-
-                final_text = get_final_text(tok_text, orig_text, do_lower_case, verbose_logging)
-                if final_text in seen_predictions:
-                    continue
-
-                seen_predictions[final_text] = True
-            else:
-                final_text = ""
-                seen_predictions[final_text] = True
-
-            nbest.append(
-                _NbestPrediction(
-                    text=final_text,
-                    start_logit=pred.start_logit,
-                    end_logit=pred.end_logit))
-        # if we didn't include the empty option in the n-best, include it
-        if version_2_with_negative:
-            if "" not in seen_predictions:
-                nbest.append(
-                    _NbestPrediction(
-                        text="",
-                        start_logit=null_start_logit,
-                        end_logit=null_end_logit))
-                
-            # In very rare edge cases we could only have single null prediction.
-            # So we just create a nonce prediction in this case to avoid failure.
-            if len(nbest)==1:
-                nbest.insert(0,
-                    _NbestPrediction(text="empty", start_logit=0.0, end_logit=0.0))
-
-        # In very rare edge cases we could have no valid predictions. So we
-        # just create a nonce prediction in this case to avoid failure.
-        if not nbest:
-            nbest.append(
-                _NbestPrediction(text="empty", start_logit=0.0, end_logit=0.0))
-
-        assert len(nbest) >= 1
-
-        total_scores = []
-        best_non_null_entry = None
-        for entry in nbest:
-            total_scores.append(entry.start_logit + entry.end_logit)
-            if not best_non_null_entry:
-                if entry.text:
-                    best_non_null_entry = entry
-
-        probs = _compute_softmax(total_scores)
-
-        nbest_json = []
-        for (i, entry) in enumerate(nbest):
-            output = collections.OrderedDict()
-            output["text"] = entry.text
-            output["probability"] = probs[i]
-            output["start_logit"] = entry.start_logit
-            output["end_logit"] = entry.end_logit
-            nbest_json.append(output)
-
-        assert len(nbest_json) >= 1
-
-        if not version_2_with_negative:
-            all_predictions[example.qas_id] = nbest_json[0]["text"]
-        else:
-            # predict "" iff the null score - the score of best non-null > threshold
-            score_diff = score_null - best_non_null_entry.start_logit - (
-                best_non_null_entry.end_logit)
-            scores_diff_json[example.qas_id] = score_diff
-            if score_diff > null_score_diff_threshold:
-                all_predictions[example.qas_id] = ""
-            else:
-                all_predictions[example.qas_id] = best_non_null_entry.text
-        all_nbest_json[example.qas_id] = nbest_json
-
-    with open(output_prediction_file, "w") as writer:
-        writer.write(json.dumps(all_predictions, indent=4) + "\n")
-
-    with open(output_nbest_file, "w") as writer:
-        writer.write(json.dumps(all_nbest_json, indent=4) + "\n")
-
-    if version_2_with_negative:
-        with open(output_null_log_odds_file, "w") as writer:
-            writer.write(json.dumps(scores_diff_json, indent=4) + "\n")
-
-    return all_predictions
-
-
-# For XLNet (and XLM which uses the same head)
-RawResultExtended = collections.namedtuple("RawResultExtended",
-    ["unique_id", "start_top_log_probs", "start_top_index",
-     "end_top_log_probs", "end_top_index", "cls_logits"])
-
-
-def write_predictions_extended(all_examples, all_features, all_results, n_best_size,
-                                max_answer_length, output_prediction_file,
-                                output_nbest_file,
-                                output_null_log_odds_file, orig_data_file,
-                                start_n_top, end_n_top, version_2_with_negative,
-                                tokenizer, verbose_logging):
-    """ XLNet write prediction logic (more complex than Bert's).
-        Write final predictions to the json file and log-odds of null if needed.
-
-        Requires utils_squad_evaluate.py
-    """
-    _PrelimPrediction = collections.namedtuple(  # pylint: disable=invalid-name
-        "PrelimPrediction",
-        ["feature_index", "start_index", "end_index",
-        "start_log_prob", "end_log_prob"])
-
-    _NbestPrediction = collections.namedtuple(  # pylint: disable=invalid-name
-        "NbestPrediction", ["text", "start_log_prob", "end_log_prob"])
-
-    logger.info("Writing predictions to: %s", output_prediction_file)
-    # logger.info("Writing nbest to: %s" % (output_nbest_file))
-
-    example_index_to_features = collections.defaultdict(list)
-    for feature in all_features:
-        example_index_to_features[feature.example_index].append(feature)
-
-    unique_id_to_result = {}
-    for result in all_results:
-        unique_id_to_result[result.unique_id] = result
-
-    all_predictions = collections.OrderedDict()
-    all_nbest_json = collections.OrderedDict()
-    scores_diff_json = collections.OrderedDict()
-
-    for (example_index, example) in enumerate(all_examples):
-        features = example_index_to_features[example_index]
-
-        prelim_predictions = []
-        # keep track of the minimum score of null start+end of position 0
-        score_null = 1000000  # large and positive
-
-        for (feature_index, feature) in enumerate(features):
-            result = unique_id_to_result[feature.unique_id]
-
-            cur_null_score = result.cls_logits
-
-            # if we could have irrelevant answers, get the min score of irrelevant
-            score_null = min(score_null, cur_null_score)
-
-            for i in range(start_n_top):
-                for j in range(end_n_top):
-                    start_log_prob = result.start_top_log_probs[i]
-                    start_index = result.start_top_index[i]
-
-                    j_index = i * end_n_top + j
-
-                    end_log_prob = result.end_top_log_probs[j_index]
-                    end_index = result.end_top_index[j_index]
-
-                    # We could hypothetically create invalid predictions, e.g., predict
-                    # that the start of the span is in the question. We throw out all
-                    # invalid predictions.
-                    if start_index >= feature.paragraph_len - 1:
-                        continue
-                    if end_index >= feature.paragraph_len - 1:
-                        continue
-
-                    if not feature.token_is_max_context.get(start_index, False):
-                        continue
-                    if end_index < start_index:
-                        continue
-                    length = end_index - start_index + 1
-                    if length > max_answer_length:
-                        continue
-
-                    prelim_predictions.append(
-                        _PrelimPrediction(
-                            feature_index=feature_index,
-                            start_index=start_index,
-                            end_index=end_index,
-                            start_log_prob=start_log_prob,
-                            end_log_prob=end_log_prob))
-
-        prelim_predictions = sorted(
-            prelim_predictions,
-            key=lambda x: (x.start_log_prob + x.end_log_prob),
-            reverse=True)
-
-        seen_predictions = {}
-        nbest = []
-        for pred in prelim_predictions:
-            if len(nbest) >= n_best_size:
-                break
-            feature = features[pred.feature_index]
-
-            # XLNet un-tokenizer
-            # Let's keep it simple for now and see if we need all this later.
-            # 
-            # tok_start_to_orig_index = feature.tok_start_to_orig_index
-            # tok_end_to_orig_index = feature.tok_end_to_orig_index
-            # start_orig_pos = tok_start_to_orig_index[pred.start_index]
-            # end_orig_pos = tok_end_to_orig_index[pred.end_index]
-            # paragraph_text = example.paragraph_text
-            # final_text = paragraph_text[start_orig_pos: end_orig_pos + 1].strip()
-
-            # Previously used Bert untokenizer
-            tok_tokens = feature.tokens[pred.start_index:(pred.end_index + 1)]
-            orig_doc_start = feature.token_to_orig_map[pred.start_index]
-            orig_doc_end = feature.token_to_orig_map[pred.end_index]
-            orig_tokens = example.doc_tokens[orig_doc_start:(orig_doc_end + 1)]
-            tok_text = tokenizer.convert_tokens_to_string(tok_tokens)
-
-            # Clean whitespace
-            tok_text = tok_text.strip()
-            tok_text = " ".join(tok_text.split())
-            orig_text = " ".join(orig_tokens)
-
-            final_text = get_final_text(tok_text, orig_text, tokenizer.do_lower_case,
-                                        verbose_logging)
-
-            if final_text in seen_predictions:
-                continue
-
-            seen_predictions[final_text] = True
-
-            nbest.append(
-                _NbestPrediction(
-                    text=final_text,
-                    start_log_prob=pred.start_log_prob,
-                    end_log_prob=pred.end_log_prob))
-
-        # In very rare edge cases we could have no valid predictions. So we
-        # just create a nonce prediction in this case to avoid failure.
-        if not nbest:
-            nbest.append(
-                _NbestPrediction(text="", start_log_prob=-1e6,
-                end_log_prob=-1e6))
-
-        total_scores = []
-        best_non_null_entry = None
-        for entry in nbest:
-            total_scores.append(entry.start_log_prob + entry.end_log_prob)
-            if not best_non_null_entry:
-                best_non_null_entry = entry
-
-        probs = _compute_softmax(total_scores)
-
-        nbest_json = []
-        for (i, entry) in enumerate(nbest):
-            output = collections.OrderedDict()
-            output["text"] = entry.text
-            output["probability"] = probs[i]
-            output["start_log_prob"] = entry.start_log_prob
-            output["end_log_prob"] = entry.end_log_prob
-            nbest_json.append(output)
-
-        assert len(nbest_json) >= 1
-        assert best_non_null_entry is not None
-
-        score_diff = score_null
-        scores_diff_json[example.qas_id] = score_diff
-        # note(zhiliny): always predict best_non_null_entry
-        # and the evaluation script will search for the best threshold
-        all_predictions[example.qas_id] = best_non_null_entry.text
-
-        all_nbest_json[example.qas_id] = nbest_json
-
-    with open(output_prediction_file, "w") as writer:
-        writer.write(json.dumps(all_predictions, indent=4) + "\n")
-
-    with open(output_nbest_file, "w") as writer:
-        writer.write(json.dumps(all_nbest_json, indent=4) + "\n")
-
-    if version_2_with_negative:
-        with open(output_null_log_odds_file, "w") as writer:
-            writer.write(json.dumps(scores_diff_json, indent=4) + "\n")
-
-    with open(orig_data_file, "r", encoding='utf-8') as reader:
-        orig_data = json.load(reader)["data"]
-
-    qid_to_has_ans = make_qid_to_has_ans(orig_data)
-    has_ans_qids = [k for k, v in qid_to_has_ans.items() if v]
-    no_ans_qids = [k for k, v in qid_to_has_ans.items() if not v]
-    exact_raw, f1_raw = get_raw_scores(orig_data, all_predictions)
-    out_eval = {}
-
-    find_all_best_thresh_v2(out_eval, all_predictions, exact_raw, f1_raw, scores_diff_json, qid_to_has_ans)
-
-    return out_eval
-
-
-def get_final_text(pred_text, orig_text, do_lower_case, verbose_logging=False):
-    """Project the tokenized prediction back to the original text."""
-
-    # When we created the data, we kept track of the alignment between original
-    # (whitespace tokenized) tokens and our WordPiece tokenized tokens. So
-    # now `orig_text` contains the span of our original text corresponding to the
-    # span that we predicted.
-    #
-    # However, `orig_text` may contain extra characters that we don't want in
-    # our prediction.
-    #
-    # For example, let's say:
-    #   pred_text = steve smith
-    #   orig_text = Steve Smith's
-    #
-    # We don't want to return `orig_text` because it contains the extra "'s".
-    #
-    # We don't want to return `pred_text` because it's already been normalized
-    # (the SQuAD eval script also does punctuation stripping/lower casing but
-    # our tokenizer does additional normalization like stripping accent
-    # characters).
-    #
-    # What we really want to return is "Steve Smith".
-    #
-    # Therefore, we have to apply a semi-complicated alignment heuristic between
-    # `pred_text` and `orig_text` to get a character-to-character alignment. This
-    # can fail in certain cases in which case we just return `orig_text`.
-
-    def _strip_spaces(text):
-        ns_chars = []
-        ns_to_s_map = collections.OrderedDict()
-        for (i, c) in enumerate(text):
-            if c == " ":
-                continue
-            ns_to_s_map[len(ns_chars)] = i
-            ns_chars.append(c)
-        ns_text = "".join(ns_chars)
-        return (ns_text, ns_to_s_map)
-
-    # We first tokenize `orig_text`, strip whitespace from the result
-    # and `pred_text`, and check if they are the same length. If they are
-    # NOT the same length, the heuristic has failed. If they are the same
-    # length, we assume the characters are one-to-one aligned.
-    tokenizer = BasicTokenizer(do_lower_case=do_lower_case)
-
-    tok_text = " ".join(tokenizer.tokenize(orig_text))
-
-    start_position = tok_text.find(pred_text)
-    if start_position == -1:
-        if verbose_logging:
-            logger.info(
-                "Unable to find text: '%s' in '%s'" % (pred_text, orig_text))
-        return orig_text
-    end_position = start_position + len(pred_text) - 1
-
-    (orig_ns_text, orig_ns_to_s_map) = _strip_spaces(orig_text)
-    (tok_ns_text, tok_ns_to_s_map) = _strip_spaces(tok_text)
-
-    if len(orig_ns_text) != len(tok_ns_text):
-        if verbose_logging:
-            logger.info("Length not equal after stripping spaces: '%s' vs '%s'",
-                        orig_ns_text, tok_ns_text)
-        return orig_text
-
-    # We then project the characters in `pred_text` back to `orig_text` using
-    # the character-to-character alignment.
-    tok_s_to_ns_map = {}
-    for (i, tok_index) in tok_ns_to_s_map.items():
-        tok_s_to_ns_map[tok_index] = i
-
-    orig_start_position = None
-    if start_position in tok_s_to_ns_map:
-        ns_start_position = tok_s_to_ns_map[start_position]
-        if ns_start_position in orig_ns_to_s_map:
-            orig_start_position = orig_ns_to_s_map[ns_start_position]
-
-    if orig_start_position is None:
-        if verbose_logging:
-            logger.info("Couldn't map start position")
-        return orig_text
-
-    orig_end_position = None
-    if end_position in tok_s_to_ns_map:
-        ns_end_position = tok_s_to_ns_map[end_position]
-        if ns_end_position in orig_ns_to_s_map:
-            orig_end_position = orig_ns_to_s_map[ns_end_position]
-
-    if orig_end_position is None:
-        if verbose_logging:
-            logger.info("Couldn't map end position")
-        return orig_text
-
-    output_text = orig_text[orig_start_position:(orig_end_position + 1)]
-    return output_text
-
-
-def _get_best_indexes(logits, n_best_size):
-    """Get the n-best logits from a list."""
-    index_and_score = sorted(enumerate(logits), key=lambda x: x[1], reverse=True)
-
-    best_indexes = []
-    for i in range(len(index_and_score)):
-        if i >= n_best_size:
-            break
-        best_indexes.append(index_and_score[i][0])
-    return best_indexes
-
-
-def _compute_softmax(scores):
-    """Compute softmax probability over raw logits."""
-    if not scores:
-        return []
-
-    max_score = None
-    for score in scores:
-        if max_score is None or score > max_score:
-            max_score = score
-
-    exp_scores = []
-    total_sum = 0.0
-    for score in scores:
-        x = math.exp(score - max_score)
-        exp_scores.append(x)
-        total_sum += x
-
-    probs = []
-    for score in exp_scores:
-        probs.append(score / total_sum)
-    return probs
diff --git a/examples/utils_squad_evaluate.py b/examples/utils_squad_evaluate.py
deleted file mode 100644
index ed162e6fe6..0000000000
--- a/examples/utils_squad_evaluate.py
+++ /dev/null
@@ -1,330 +0,0 @@
-""" Official evaluation script for SQuAD version 2.0.
-    Modified by XLNet authors to update `find_best_threshold` scripts for SQuAD V2.0
-
-In addition to basic functionality, we also compute additional statistics and
-plot precision-recall curves if an additional na_prob.json file is provided.
-This file is expected to map question ID's to the model's predicted probability
-that a question is unanswerable.
-"""
-import argparse
-import collections
-import json
-import numpy as np
-import os
-import re
-import string
-import sys
-
-class EVAL_OPTS():
-  def __init__(self, data_file, pred_file, out_file="",
-               na_prob_file="na_prob.json", na_prob_thresh=1.0,
-               out_image_dir=None, verbose=False):
-    self.data_file = data_file
-    self.pred_file = pred_file
-    self.out_file = out_file
-    self.na_prob_file = na_prob_file
-    self.na_prob_thresh = na_prob_thresh
-    self.out_image_dir = out_image_dir
-    self.verbose = verbose
-
-OPTS = None
-
-def parse_args():
-  parser = argparse.ArgumentParser('Official evaluation script for SQuAD version 2.0.')
-  parser.add_argument('data_file', metavar='data.json', help='Input data JSON file.')
-  parser.add_argument('pred_file', metavar='pred.json', help='Model predictions.')
-  parser.add_argument('--out-file', '-o', metavar='eval.json',
-                      help='Write accuracy metrics to file (default is stdout).')
-  parser.add_argument('--na-prob-file', '-n', metavar='na_prob.json',
-                      help='Model estimates of probability of no answer.')
-  parser.add_argument('--na-prob-thresh', '-t', type=float, default=1.0,
-                      help='Predict "" if no-answer probability exceeds this (default = 1.0).')
-  parser.add_argument('--out-image-dir', '-p', metavar='out_images', default=None,
-                      help='Save precision-recall curves to directory.')
-  parser.add_argument('--verbose', '-v', action='store_true')
-  if len(sys.argv) == 1:
-    parser.print_help()
-    sys.exit(1)
-  return parser.parse_args()
-
-def make_qid_to_has_ans(dataset):
-  qid_to_has_ans = {}
-  for article in dataset:
-    for p in article['paragraphs']:
-      for qa in p['qas']:
-        qid_to_has_ans[qa['id']] = bool(qa['answers'])
-  return qid_to_has_ans
-
-def normalize_answer(s):
-  """Lower text and remove punctuation, articles and extra whitespace."""
-  def remove_articles(text):
-    regex = re.compile(r'\b(a|an|the)\b', re.UNICODE)
-    return re.sub(regex, ' ', text)
-  def white_space_fix(text):
-    return ' '.join(text.split())
-  def remove_punc(text):
-    exclude = set(string.punctuation)
-    return ''.join(ch for ch in text if ch not in exclude)
-  def lower(text):
-    return text.lower()
-  return white_space_fix(remove_articles(remove_punc(lower(s))))
-
-def get_tokens(s):
-  if not s: return []
-  return normalize_answer(s).split()
-
-def compute_exact(a_gold, a_pred):
-  return int(normalize_answer(a_gold) == normalize_answer(a_pred))
-
-def compute_f1(a_gold, a_pred):
-  gold_toks = get_tokens(a_gold)
-  pred_toks = get_tokens(a_pred)
-  common = collections.Counter(gold_toks) & collections.Counter(pred_toks)
-  num_same = sum(common.values())
-  if len(gold_toks) == 0 or len(pred_toks) == 0:
-    # If either is no-answer, then F1 is 1 if they agree, 0 otherwise
-    return int(gold_toks == pred_toks)
-  if num_same == 0:
-    return 0
-  precision = 1.0 * num_same / len(pred_toks)
-  recall = 1.0 * num_same / len(gold_toks)
-  f1 = (2 * precision * recall) / (precision + recall)
-  return f1
-
-def get_raw_scores(dataset, preds):
-  exact_scores = {}
-  f1_scores = {}
-  for article in dataset:
-    for p in article['paragraphs']:
-      for qa in p['qas']:
-        qid = qa['id']
-        gold_answers = [a['text'] for a in qa['answers']
-                        if normalize_answer(a['text'])]
-        if not gold_answers:
-          # For unanswerable questions, only correct answer is empty string
-          gold_answers = ['']
-        if qid not in preds:
-          print('Missing prediction for %s' % qid)
-          continue
-        a_pred = preds[qid]
-        # Take max over all gold answers
-        exact_scores[qid] = max(compute_exact(a, a_pred) for a in gold_answers)
-        f1_scores[qid] = max(compute_f1(a, a_pred) for a in gold_answers)
-  return exact_scores, f1_scores
-
-def apply_no_ans_threshold(scores, na_probs, qid_to_has_ans, na_prob_thresh):
-  new_scores = {}
-  for qid, s in scores.items():
-    pred_na = na_probs[qid] > na_prob_thresh
-    if pred_na:
-      new_scores[qid] = float(not qid_to_has_ans[qid])
-    else:
-      new_scores[qid] = s
-  return new_scores
-
-def make_eval_dict(exact_scores, f1_scores, qid_list=None):
-  if not qid_list:
-    total = len(exact_scores)
-    return collections.OrderedDict([
-        ('exact', 100.0 * sum(exact_scores.values()) / total),
-        ('f1', 100.0 * sum(f1_scores.values()) / total),
-        ('total', total),
-    ])
-  else:
-    total = len(qid_list)
-    return collections.OrderedDict([
-        ('exact', 100.0 * sum(exact_scores[k] for k in qid_list) / total),
-        ('f1', 100.0 * sum(f1_scores[k] for k in qid_list) / total),
-        ('total', total),
-    ])
-
-def merge_eval(main_eval, new_eval, prefix):
-  for k in new_eval:
-    main_eval['%s_%s' % (prefix, k)] = new_eval[k]
-
-def plot_pr_curve(precisions, recalls, out_image, title):
-  plt.step(recalls, precisions, color='b', alpha=0.2, where='post')
-  plt.fill_between(recalls, precisions, step='post', alpha=0.2, color='b')
-  plt.xlabel('Recall')
-  plt.ylabel('Precision')
-  plt.xlim([0.0, 1.05])
-  plt.ylim([0.0, 1.05])
-  plt.title(title)
-  plt.savefig(out_image)
-  plt.clf()
-
-def make_precision_recall_eval(scores, na_probs, num_true_pos, qid_to_has_ans,
-                               out_image=None, title=None):
-  qid_list = sorted(na_probs, key=lambda k: na_probs[k])
-  true_pos = 0.0
-  cur_p = 1.0
-  cur_r = 0.0
-  precisions = [1.0]
-  recalls = [0.0]
-  avg_prec = 0.0
-  for i, qid in enumerate(qid_list):
-    if qid_to_has_ans[qid]:
-      true_pos += scores[qid]
-    cur_p = true_pos / float(i+1)
-    cur_r = true_pos / float(num_true_pos)
-    if i == len(qid_list) - 1 or na_probs[qid] != na_probs[qid_list[i+1]]:
-      # i.e., if we can put a threshold after this point
-      avg_prec += cur_p * (cur_r - recalls[-1])
-      precisions.append(cur_p)
-      recalls.append(cur_r)
-  if out_image:
-    plot_pr_curve(precisions, recalls, out_image, title)
-  return {'ap': 100.0 * avg_prec}
-
-def run_precision_recall_analysis(main_eval, exact_raw, f1_raw, na_probs, 
-                                  qid_to_has_ans, out_image_dir):
-  if out_image_dir and not os.path.exists(out_image_dir):
-    os.makedirs(out_image_dir)
-  num_true_pos = sum(1 for v in qid_to_has_ans.values() if v)
-  if num_true_pos == 0:
-    return
-  pr_exact = make_precision_recall_eval(
-      exact_raw, na_probs, num_true_pos, qid_to_has_ans,
-      out_image=os.path.join(out_image_dir, 'pr_exact.png'),
-      title='Precision-Recall curve for Exact Match score')
-  pr_f1 = make_precision_recall_eval(
-      f1_raw, na_probs, num_true_pos, qid_to_has_ans,
-      out_image=os.path.join(out_image_dir, 'pr_f1.png'),
-      title='Precision-Recall curve for F1 score')
-  oracle_scores = {k: float(v) for k, v in qid_to_has_ans.items()}
-  pr_oracle = make_precision_recall_eval(
-      oracle_scores, na_probs, num_true_pos, qid_to_has_ans,
-      out_image=os.path.join(out_image_dir, 'pr_oracle.png'),
-      title='Oracle Precision-Recall curve (binary task of HasAns vs. NoAns)')
-  merge_eval(main_eval, pr_exact, 'pr_exact')
-  merge_eval(main_eval, pr_f1, 'pr_f1')
-  merge_eval(main_eval, pr_oracle, 'pr_oracle')
-
-def histogram_na_prob(na_probs, qid_list, image_dir, name):
-  if not qid_list:
-    return
-  x = [na_probs[k] for k in qid_list]
-  weights = np.ones_like(x) / float(len(x))
-  plt.hist(x, weights=weights, bins=20, range=(0.0, 1.0))
-  plt.xlabel('Model probability of no-answer')
-  plt.ylabel('Proportion of dataset')
-  plt.title('Histogram of no-answer probability: %s' % name)
-  plt.savefig(os.path.join(image_dir, 'na_prob_hist_%s.png' % name))
-  plt.clf()
-
-def find_best_thresh(preds, scores, na_probs, qid_to_has_ans):
-  num_no_ans = sum(1 for k in qid_to_has_ans if not qid_to_has_ans[k])
-  cur_score = num_no_ans
-  best_score = cur_score
-  best_thresh = 0.0
-  qid_list = sorted(na_probs, key=lambda k: na_probs[k])
-  for i, qid in enumerate(qid_list):
-    if qid not in scores: continue
-    if qid_to_has_ans[qid]:
-      diff = scores[qid]
-    else:
-      if preds[qid]:
-        diff = -1
-      else:
-        diff = 0
-    cur_score += diff
-    if cur_score > best_score:
-      best_score = cur_score
-      best_thresh = na_probs[qid]
-  return 100.0 * best_score / len(scores), best_thresh
-
-def find_best_thresh_v2(preds, scores, na_probs, qid_to_has_ans):
-  num_no_ans = sum(1 for k in qid_to_has_ans if not qid_to_has_ans[k])
-  cur_score = num_no_ans
-  best_score = cur_score
-  best_thresh = 0.0
-  qid_list = sorted(na_probs, key=lambda k: na_probs[k])
-  for i, qid in enumerate(qid_list):
-    if qid not in scores: continue
-    if qid_to_has_ans[qid]:
-      diff = scores[qid]
-    else:
-      if preds[qid]:
-        diff = -1
-      else:
-        diff = 0
-    cur_score += diff
-    if cur_score > best_score:
-      best_score = cur_score
-      best_thresh = na_probs[qid]
-
-  has_ans_score, has_ans_cnt = 0, 0
-  for qid in qid_list:
-    if not qid_to_has_ans[qid]: continue
-    has_ans_cnt += 1
-
-    if qid not in scores: continue
-    has_ans_score += scores[qid]
-
-  return 100.0 * best_score / len(scores), best_thresh, 1.0 * has_ans_score / has_ans_cnt
-
-def find_all_best_thresh(main_eval, preds, exact_raw, f1_raw, na_probs, qid_to_has_ans):
-  best_exact, exact_thresh = find_best_thresh(preds, exact_raw, na_probs, qid_to_has_ans)
-  best_f1, f1_thresh = find_best_thresh(preds, f1_raw, na_probs, qid_to_has_ans)
-  main_eval['best_exact'] = best_exact
-  main_eval['best_exact_thresh'] = exact_thresh
-  main_eval['best_f1'] = best_f1
-  main_eval['best_f1_thresh'] = f1_thresh
-
-def find_all_best_thresh_v2(main_eval, preds, exact_raw, f1_raw, na_probs, qid_to_has_ans):
-  best_exact, exact_thresh, has_ans_exact = find_best_thresh_v2(preds, exact_raw, na_probs, qid_to_has_ans)
-  best_f1, f1_thresh, has_ans_f1 = find_best_thresh_v2(preds, f1_raw, na_probs, qid_to_has_ans)
-  main_eval['best_exact'] = best_exact
-  main_eval['best_exact_thresh'] = exact_thresh
-  main_eval['best_f1'] = best_f1
-  main_eval['best_f1_thresh'] = f1_thresh
-  main_eval['has_ans_exact'] = has_ans_exact
-  main_eval['has_ans_f1'] = has_ans_f1
-
-def main(OPTS):
-  with open(OPTS.data_file) as f:
-    dataset_json = json.load(f)
-    dataset = dataset_json['data']
-  with open(OPTS.pred_file) as f:
-    preds = json.load(f)
-  if OPTS.na_prob_file:
-    with open(OPTS.na_prob_file) as f:
-      na_probs = json.load(f)
-  else:
-    na_probs = {k: 0.0 for k in preds}
-  qid_to_has_ans = make_qid_to_has_ans(dataset)  # maps qid to True/False
-  has_ans_qids = [k for k, v in qid_to_has_ans.items() if v]
-  no_ans_qids = [k for k, v in qid_to_has_ans.items() if not v]
-  exact_raw, f1_raw = get_raw_scores(dataset, preds)
-  exact_thresh = apply_no_ans_threshold(exact_raw, na_probs, qid_to_has_ans,
-                                        OPTS.na_prob_thresh)
-  f1_thresh = apply_no_ans_threshold(f1_raw, na_probs, qid_to_has_ans,
-                                     OPTS.na_prob_thresh)
-  out_eval = make_eval_dict(exact_thresh, f1_thresh)
-  if has_ans_qids:
-    has_ans_eval = make_eval_dict(exact_thresh, f1_thresh, qid_list=has_ans_qids)
-    merge_eval(out_eval, has_ans_eval, 'HasAns')
-  if no_ans_qids:
-    no_ans_eval = make_eval_dict(exact_thresh, f1_thresh, qid_list=no_ans_qids)
-    merge_eval(out_eval, no_ans_eval, 'NoAns')
-  if OPTS.na_prob_file:
-    find_all_best_thresh(out_eval, preds, exact_raw, f1_raw, na_probs, qid_to_has_ans)
-  if OPTS.na_prob_file and OPTS.out_image_dir:
-    run_precision_recall_analysis(out_eval, exact_raw, f1_raw, na_probs, 
-                                  qid_to_has_ans, OPTS.out_image_dir)
-    histogram_na_prob(na_probs, has_ans_qids, OPTS.out_image_dir, 'hasAns')
-    histogram_na_prob(na_probs, no_ans_qids, OPTS.out_image_dir, 'noAns')
-  if OPTS.out_file:
-    with open(OPTS.out_file, 'w') as f:
-      json.dump(out_eval, f)
-  else:
-    print(json.dumps(out_eval, indent=2))
-  return out_eval
-
-if __name__ == '__main__':
-  OPTS = parse_args()
-  if OPTS.out_image_dir:
-    import matplotlib
-    matplotlib.use('Agg')
-    import matplotlib.pyplot as plt 
-  main(OPTS)
diff --git a/transformers/data/metrics/squad_metrics.py b/transformers/data/metrics/squad_metrics.py
index f8449df045..0755c0ab7a 100644
--- a/transformers/data/metrics/squad_metrics.py
+++ b/transformers/data/metrics/squad_metrics.py
@@ -578,7 +578,6 @@ def compute_predictions_log_probs(
     output_prediction_file,
     output_nbest_file,
     output_null_log_odds_file,
-    orig_data_file,
     start_n_top,
     end_n_top,
     version_2_with_negative,
@@ -756,15 +755,4 @@ def compute_predictions_log_probs(
         with open(output_null_log_odds_file, "w") as writer:
             writer.write(json.dumps(scores_diff_json, indent=4) + "\n")
 
-    with open(orig_data_file, "r", encoding='utf-8') as reader:
-        orig_data = json.load(reader)["data"]
-
-    qid_to_has_ans = make_qid_to_has_ans(orig_data)
-    has_ans_qids = [k for k, v in qid_to_has_ans.items() if v]
-    no_ans_qids = [k for k, v in qid_to_has_ans.items() if not v]
-    exact_raw, f1_raw = get_raw_scores(orig_data, all_predictions)
-    out_eval = {}
-
-    find_all_best_thresh_v2(out_eval, all_predictions, exact_raw, f1_raw, scores_diff_json, qid_to_has_ans)
-
-    return out_eval
+    return all_predictions
diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index bb56aa792f..3d7f832540 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -9,7 +9,7 @@ from ...tokenization_bert import BasicTokenizer, whitespace_tokenize
 from .utils import DataProcessor, InputExample, InputFeatures
 from ...file_utils import is_tf_available, is_torch_available
 
-if is_torch_available:
+if is_torch_available():
     import torch
     from torch.utils.data import TensorDataset
 

From d0383e4daf44557e56cd4cbc5dc95b1d35457768 Mon Sep 17 00:00:00 2001
From: patrickvonplaten <patrick.v.platen@gmail.com>
Date: Fri, 6 Dec 2019 01:24:22 +0100
Subject: [PATCH 244/505] corrected documentation for past tensor shape for
 ctrl and gpt2 model

---
 transformers/modeling_ctrl.py    | 4 ++--
 transformers/modeling_gpt2.py    | 6 +++---
 transformers/modeling_tf_ctrl.py | 4 ++--
 transformers/modeling_tf_gpt2.py | 6 +++---
 4 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/transformers/modeling_ctrl.py b/transformers/modeling_ctrl.py
index 3a252941ac..97bcb14434 100644
--- a/transformers/modeling_ctrl.py
+++ b/transformers/modeling_ctrl.py
@@ -252,7 +252,7 @@ class CTRLModel(CTRLPreTrainedModel):
         **last_hidden_state**: ``torch.FloatTensor`` of shape ``(batch_size, sequence_length, hidden_size)``
             Sequence of hidden-states at the last layer of the model.
         **past**:
-            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(2, batch_size, num_heads, sequence_length, embed_size_per_head)``:
             that contains pre-computed hidden-states (key and values in the attention blocks).
             Can be used (see `past` input) to speed up sequential decoding. The token ids which have their past given to this model 
             should not be passed as input ids as they have already been computed.
@@ -438,7 +438,7 @@ class CTRLLMHeadModel(CTRLPreTrainedModel):
         **prediction_scores**: ``torch.FloatTensor`` of shape ``(batch_size, sequence_length, config.vocab_size)``
             Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).
         **past**:
-            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(2, batch_size, num_heads, sequence_length, embed_size_per_head)``:
             that contains pre-computed hidden-states (key and values in the attention blocks).
             Can be used (see `past` input) to speed up sequential decoding. The token ids which have their past given to this model 
             should not be passed as input ids as they have already been computed.
diff --git a/transformers/modeling_gpt2.py b/transformers/modeling_gpt2.py
index 35bc5c8d6e..96fd1c0607 100644
--- a/transformers/modeling_gpt2.py
+++ b/transformers/modeling_gpt2.py
@@ -329,7 +329,7 @@ class GPT2Model(GPT2PreTrainedModel):
         **last_hidden_state**: ``torch.FloatTensor`` of shape ``(batch_size, sequence_length, hidden_size)``
             Sequence of hidden-states at the last layer of the model.
         **past**:
-            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(2, batch_size, num_heads, sequence_length, embed_size_per_head)``:
             that contains pre-computed hidden-states (key and values in the attention blocks).
             Can be used (see `past` input) to speed up sequential decoding. The token ids which have their past given to this model 
             should not be passed as input ids as they have already been computed.
@@ -503,7 +503,7 @@ class GPT2LMHeadModel(GPT2PreTrainedModel):
         **prediction_scores**: ``torch.FloatTensor`` of shape ``(batch_size, sequence_length, config.vocab_size)``
             Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).
         **past**:
-            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(2, batch_size, num_heads, sequence_length, embed_size_per_head)``:
             that contains pre-computed hidden-states (key and values in the attention blocks).
             Can be used (see `past` input) to speed up sequential decoding. The token ids which have their past given to this model 
             should not be passed as input ids as they have already been computed.
@@ -596,7 +596,7 @@ class GPT2DoubleHeadsModel(GPT2PreTrainedModel):
         **mc_prediction_scores**: ``torch.FloatTensor`` of shape ``(batch_size, num_choices)``
             Prediction scores of the multiplechoice classification head (scores for each choice before SoftMax).
         **past**:
-            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(2, batch_size, num_heads, sequence_length, embed_size_per_head)``:
             that contains pre-computed hidden-states (key and values in the attention blocks).
             Can be used (see `past` input) to speed up sequential decoding. The token ids which have their past given to this model 
             should not be passed as input ids as they have already been computed.
diff --git a/transformers/modeling_tf_ctrl.py b/transformers/modeling_tf_ctrl.py
index 6d0d6a57ad..29ee5113a4 100644
--- a/transformers/modeling_tf_ctrl.py
+++ b/transformers/modeling_tf_ctrl.py
@@ -400,7 +400,7 @@ class TFCTRLModel(TFCTRLPreTrainedModel):
         **last_hidden_state**: ``tf.Tensor`` of shape ``(batch_size, sequence_length, hidden_size)``
             Sequence of hidden-states at the last layer of the model.
         **past**:
-            list of ``tf.Tensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            list of ``tf.Tensor`` (one for each layer) of shape ``(2, batch_size, num_heads, sequence_length, embed_size_per_head)``:
             that contains pre-computed hidden-states (key and values in the attention blocks).
             Can be used (see `past` input) to speed up sequential decoding.
         **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
@@ -462,7 +462,7 @@ class TFCTRLLMHeadModel(TFCTRLPreTrainedModel):
         **prediction_scores**: ``torch.FloatTensor`` of shape ``(batch_size, sequence_length, config.vocab_size)``
             Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).
         **past**:
-            list of ``tf.Tensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            list of ``tf.Tensor`` (one for each layer) of shape ``(2, batch_size, num_heads, sequence_length, embed_size_per_head)``:
             that contains pre-computed hidden-states (key and values in the attention blocks).
             Can be used (see `past` input) to speed up sequential decoding.
         **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
diff --git a/transformers/modeling_tf_gpt2.py b/transformers/modeling_tf_gpt2.py
index aebe790114..c738e5e8e3 100644
--- a/transformers/modeling_tf_gpt2.py
+++ b/transformers/modeling_tf_gpt2.py
@@ -436,7 +436,7 @@ class TFGPT2Model(TFGPT2PreTrainedModel):
         **last_hidden_state**: ``tf.Tensor`` of shape ``(batch_size, sequence_length, hidden_size)``
             Sequence of hidden-states at the last layer of the model.
         **past**:
-            list of ``tf.Tensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            list of ``tf.Tensor`` (one for each layer) of shape ``(2, batch_size, num_heads, sequence_length, embed_size_per_head)``:
             that contains pre-computed hidden-states (key and values in the attention blocks).
             Can be used (see `past` input) to speed up sequential decoding.
         **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
@@ -476,7 +476,7 @@ class TFGPT2LMHeadModel(TFGPT2PreTrainedModel):
         **prediction_scores**: `tf.Tensor`` of shape ``(batch_size, sequence_length, config.vocab_size)``
             Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).
         **past**:
-            list of `tf.Tensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            list of `tf.Tensor`` (one for each layer) of shape ``(2, batch_size, num_heads, sequence_length, embed_size_per_head)``:
             that contains pre-computed hidden-states (key and values in the attention blocks).
             Can be used (see `past` input) to speed up sequential decoding.
         **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
@@ -535,7 +535,7 @@ class TFGPT2DoubleHeadsModel(TFGPT2PreTrainedModel):
         **mc_prediction_scores**: `tf.Tensor`` of shape ``(batch_size, num_choices)``
             Prediction scores of the multiplechoice classification head (scores for each choice before SoftMax).
         **past**:
-            list of `tf.Tensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            list of `tf.Tensor`` (one for each layer) of shape ``(2, batch_size, num_heads, sequence_length, embed_size_per_head)``:
             that contains pre-computed hidden-states (key and values in the attention blocks).
             Can be used (see `past` input) to speed up sequential decoding.
         **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)

From f230d91b437c806e3e2dad37318a5ce77d208fa6 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Thu, 5 Dec 2019 21:24:57 +0100
Subject: [PATCH 245/505] check the validity of links

We add a script and a CI workflow to check that all download links
present in the source code are valid.
---
 .circleci/config.yml | 11 ++++++
 utils/link_tester.py | 79 ++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 90 insertions(+)
 create mode 100644 utils/link_tester.py

diff --git a/.circleci/config.yml b/.circleci/config.yml
index 01e6d82b33..ebfbd79b93 100644
--- a/.circleci/config.yml
+++ b/.circleci/config.yml
@@ -82,6 +82,16 @@ jobs:
             - run: sudo pip install --progress-bar off -r docs/requirements.txt
             - run: sudo pip install --progress-bar off -r requirements.txt
             - run: ./.circleci/deploy.sh
+    repository_consistency:
+        working_directory: ~/transformers
+        docker:
+            - image: circleci/python:3.5
+        resource_class: small
+        parallelism: 1
+        steps:
+            - checkout
+            - run: sudo pip install requests
+            - run: python ./utils/link_tester.py
 workflow_filters: &workflow_filters
     filters:
         branches:
@@ -91,6 +101,7 @@ workflows:
     version: 2
     build_and_test:
         jobs:
+            - repository_consistency
             - build_py3_torch_and_tf
             - build_py3_torch
             - build_py3_tf
diff --git a/utils/link_tester.py b/utils/link_tester.py
new file mode 100644
index 0000000000..fe3990d28c
--- /dev/null
+++ b/utils/link_tester.py
@@ -0,0 +1,79 @@
+""" Link tester.
+
+This little utility reads all the python files in the repository,
+scans for links pointing to S3 and tests the links one by one. Raises an error
+at the end of the scan if at least one link was reported broken.
+"""
+import os
+import re
+import sys
+
+import requests
+
+
+REGEXP_FIND_S3_LINKS = r"""([\"'])(https:\/\/s3)(.*)?\1"""
+
+
+def list_python_files_in_repository():
+    """ List all python files in the repository.
+
+    This function assumes that the script is executed in the root folder.
+    """
+    source_code_files = []
+    for path, subdirs, files in os.walk("."):
+        if "templates" in path:
+            continue
+        for name in files:
+            if ".py" in name and ".pyc" not in name:
+                path_to_files = os.path.join(path, name)
+                source_code_files.append(path_to_files)
+
+    return source_code_files
+
+
+def find_all_links(file_paths):
+    links = []
+    for path in file_paths:
+        links += scan_code_for_links(path)
+
+    return links
+
+
+def scan_code_for_links(source):
+    """ Scans the file to find links using a regular expression.
+    Returns a list of links.
+    """
+    with open(source, 'r') as content:
+        content = content.read()
+        raw_links = re.findall(REGEXP_FIND_S3_LINKS, content)
+        links = [prefix + suffix for _, prefix, suffix in raw_links]
+
+    return links
+
+
+def check_all_links(links):
+    """ Check that the provided links are valid.
+
+    Links are considered valid if a HEAD request to the server
+    returns a 200 status code.
+    """
+    broken_links = []
+    for link in links:
+        head = requests.head(link)
+        if head.status_code != 200:
+            broken_links.append(link)
+
+    return broken_links
+
+
+if __name__ == "__main__":
+    file_paths = list_python_files_in_repository()
+    links = find_all_links(file_paths)
+    broken_links = check_all_links(links)
+    print("Looking for broken links to pre-trained models/configs/tokenizers...")
+    if broken_links:
+        print("The following links did not respond:")
+        for link in broken_links:
+            print("- {}".format(link))
+        sys.exit(1)
+    print("All links are ok.")

From 21451ec6ba364de78c14e7d05a55913da2809844 Mon Sep 17 00:00:00 2001
From: Philipp Glock <pglock@users.noreply.github.com>
Date: Fri, 6 Dec 2019 10:32:43 +0100
Subject: [PATCH 246/505] handle string with only whitespaces as empty

---
 transformers/tokenization_utils.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index 5d683629f0..bc246cc8fe 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -634,7 +634,7 @@ class PreTrainedTokenizer(object):
             return result
 
         def split_on_tokens(tok_list, text):
-            if not text:
+            if not text.strip():
                 return []
             if not tok_list:
                 return self._tokenize(text, **kwargs)

From 1d87b37d100c69ff3b2c1a5dfd271b6cf777176e Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 6 Dec 2019 15:30:09 +0100
Subject: [PATCH 247/505] updating

---
 .../convert_pytorch_checkpoint_to_tf2.py      | 20 ++++++++++++-------
 1 file changed, 13 insertions(+), 7 deletions(-)

diff --git a/transformers/convert_pytorch_checkpoint_to_tf2.py b/transformers/convert_pytorch_checkpoint_to_tf2.py
index d20eafe2e9..2c419888e8 100644
--- a/transformers/convert_pytorch_checkpoint_to_tf2.py
+++ b/transformers/convert_pytorch_checkpoint_to_tf2.py
@@ -119,10 +119,10 @@ def convert_pt_checkpoint_to_tf(model_type, pytorch_checkpoint_path, config_file
         tf_inputs = tf.constant(inputs_list)
         tfo = tf_model(tf_inputs, training=False)  # build the network
 
-        pt_model = pt_model_class(config)
-        pt_model.load_state_dict(torch.load(pytorch_checkpoint_path, map_location='cpu'),
-                                 strict-False)
-        pt_model.eval()
+        state_dict = torch.load(pytorch_checkpoint_path, map_location='cpu')
+        pt_model = pt_model_class.from_pretrained(pretrained_model_name_or_path=None,
+                                                  config=config,
+                                                  state_dict=state_dict)
 
         pt_inputs = torch.tensor(inputs_list)
         with torch.no_grad():
@@ -140,7 +140,7 @@ def convert_pt_checkpoint_to_tf(model_type, pytorch_checkpoint_path, config_file
 
 
 def convert_all_pt_checkpoints_to_tf(args_model_type, tf_dump_path, model_shortcut_names_or_path=None, config_shortcut_names_or_path=None,
-                                     compare_with_pt_model=False, use_cached_models=False, only_convert_finetuned_models=False):
+                                     compare_with_pt_model=False, use_cached_models=False, remove_cached_files=False, only_convert_finetuned_models=False):
     assert os.path.isdir(args.tf_dump_path), "--tf_dump_path should be a directory"
 
     if args_model_type is None:
@@ -188,13 +188,15 @@ def convert_all_pt_checkpoints_to_tf(args_model_type, tf_dump_path, model_shortc
 
             if os.path.isfile(model_shortcut_name):
                 model_shortcut_name = 'converted_model'
+
             convert_pt_checkpoint_to_tf(model_type=model_type,
                                         pytorch_checkpoint_path=model_file,
                                         config_file=config_file,
                                         tf_dump_path=os.path.join(tf_dump_path, model_shortcut_name + '-tf_model.h5'),
                                         compare_with_pt_model=compare_with_pt_model)
-            os.remove(config_file)
-            os.remove(model_file)
+            if remove_cached_files:
+                os.remove(config_file)
+                os.remove(model_file)
 
 
 if __name__ == "__main__":
@@ -227,6 +229,9 @@ if __name__ == "__main__":
     parser.add_argument("--use_cached_models",
                         action='store_true',
                         help = "Use cached models if possible instead of updating to latest checkpoint versions.")
+    parser.add_argument("--remove_cached_files",
+                        action='store_true',
+                        help = "Remove pytorch models after conversion (save memory when converting in batches).")
     parser.add_argument("--only_convert_finetuned_models",
                         action='store_true',
                         help = "Only convert finetuned models.")
@@ -246,4 +251,5 @@ if __name__ == "__main__":
                                         config_shortcut_names_or_path=[args.config_file] if args.config_file is not None else None,
                                         compare_with_pt_model=args.compare_with_pt_model,
                                         use_cached_models=args.use_cached_models,
+                                        remove_cached_files=args.remove_cached_files,
                                         only_convert_finetuned_models=args.only_convert_finetuned_models)

From e4679cddced7d746427066a78e8079fb40e51528 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Fri, 6 Dec 2019 11:56:23 -0500
Subject: [PATCH 248/505] [cli] Uploads: add progress bar (#2078)

* [cli] Uploads: add progress bar

see https://github.com/huggingface/transformers/pull/2044#discussion_r354057827 for context

* rename + documentation

* Add auto-referential comment
---
 transformers/hf_api.py | 34 +++++++++++++++++++++++++++++++++-
 1 file changed, 33 insertions(+), 1 deletion(-)

diff --git a/transformers/hf_api.py b/transformers/hf_api.py
index c21592a838..3bbb6c567a 100644
--- a/transformers/hf_api.py
+++ b/transformers/hf_api.py
@@ -16,10 +16,11 @@ from __future__ import absolute_import, division, print_function
 
 import os
 from os.path import expanduser
-import six
 
 import requests
+import six
 from requests.exceptions import HTTPError
+from tqdm import tqdm
 
 ENDPOINT = "https://huggingface.co"
 
@@ -129,10 +130,13 @@ class HfApi:
         # Even though we presign with the correct content-type,
         # the client still has to specify it when uploading the file.
         with open(filepath, "rb") as f:
+            pf = TqdmProgressFileReader(f)
+
             r = requests.put(urls.write, data=f, headers={
                 "content-type": urls.type,
             })
             r.raise_for_status()
+            pf.close()
         return urls.access
 
     def list_objs(self, token):
@@ -148,6 +152,34 @@ class HfApi:
 
 
 
+class TqdmProgressFileReader:
+    """
+    Wrap an io.BufferedReader `f` (such as the output of `open(…, "rb")`)
+    and override `f.read()` so as to display a tqdm progress bar.
+
+    see github.com/huggingface/transformers/pull/2078#discussion_r354739608
+    for implementation details.
+    """
+    def __init__(
+        self,
+        f   # type: io.BufferedReader
+    ):
+        self.f = f
+        self.total_size = os.fstat(f.fileno()).st_size # type: int
+        self.pbar = tqdm(total=self.total_size, leave=False)
+        if six.PY3:
+            # does not work unless PY3
+            # no big deal as the CLI does not currently support PY2 anyways.
+            self.read = f.read
+            f.read = self._read
+
+    def _read(self, n=-1):
+        self.pbar.update(n)
+        return self.read(n)
+
+    def close(self):
+        self.pbar.close()
+
 
 
 class HfFolder:

From 35401fe50fa3e460b2a4422630b017f106c79e03 Mon Sep 17 00:00:00 2001
From: Aymeric Augustin <aymeric.augustin@fractalideas.com>
Date: Fri, 6 Dec 2019 19:57:38 +0100
Subject: [PATCH 249/505] Remove dependency on pytest for running tests (#2055)

* Switch to plain unittest for skipping slow tests.

Add a RUN_SLOW environment variable for running them.

* Switch to plain unittest for PyTorch dependency.

* Switch to plain unittest for TensorFlow dependency.

* Avoid leaking open files in the test suite.

This prevents spurious warnings when running tests.

* Fix unicode warning on Python 2 when running tests.

The warning was:

    UnicodeWarning: Unicode equal comparison failed to convert both arguments to Unicode - interpreting them as being unequal

* Support running PyTorch tests on a GPU.

Reverts 27e015bd.

* Tests no longer require pytest.

* Make tests pass on cuda
---
 README.md                                     | 11 +++-
 docs/source/installation.md                   | 11 +++-
 setup.py                                      |  1 -
 .../tests/modeling_tf_xxx_test.py             |  7 +-
 .../tests/modeling_xxx_test.py                | 12 ++--
 transformers/modeling_openai.py               |  6 +-
 transformers/tests/conftest.py                | 31 ---------
 transformers/tests/modeling_albert_test.py    | 11 ++--
 transformers/tests/modeling_auto_test.py      | 14 ++--
 transformers/tests/modeling_bert_test.py      | 38 +++++------
 transformers/tests/modeling_common_test.py    | 43 ++++++++++---
 transformers/tests/modeling_ctrl_test.py      |  9 +--
 .../tests/modeling_distilbert_test.py         | 12 ++--
 .../tests/modeling_encoder_decoder_test.py    |  7 +-
 transformers/tests/modeling_gpt2_test.py      | 10 +--
 transformers/tests/modeling_openai_test.py    | 10 +--
 transformers/tests/modeling_roberta_test.py   | 22 ++++---
 transformers/tests/modeling_tf_albert_test.py |  7 +-
 transformers/tests/modeling_tf_auto_test.py   | 14 ++--
 transformers/tests/modeling_tf_bert_test.py   |  7 +-
 transformers/tests/modeling_tf_common_test.py |  8 +--
 transformers/tests/modeling_tf_ctrl_test.py   |  7 +-
 .../tests/modeling_tf_distilbert_test.py      |  7 +-
 transformers/tests/modeling_tf_gpt2_test.py   |  7 +-
 .../tests/modeling_tf_openai_gpt_test.py      |  7 +-
 .../tests/modeling_tf_roberta_test.py         | 19 +++---
 .../tests/modeling_tf_transfo_xl_test.py      |  7 +-
 transformers/tests/modeling_tf_xlm_test.py    |  7 +-
 transformers/tests/modeling_tf_xlnet_test.py  | 10 +--
 .../tests/modeling_transfo_xl_test.py         | 10 +--
 transformers/tests/modeling_xlm_test.py       | 12 ++--
 transformers/tests/modeling_xlnet_test.py     | 21 ++++--
 transformers/tests/optimization_test.py       |  6 +-
 transformers/tests/tokenization_auto_test.py  |  5 +-
 transformers/tests/tokenization_bert_test.py  |  4 +-
 .../tests/tokenization_distilbert_test.py     |  4 +-
 .../tests/tokenization_roberta_test.py        |  4 +-
 .../tests/tokenization_tests_commons.py       |  6 +-
 .../tests/tokenization_transfo_xl_test.py     |  6 +-
 transformers/tests/tokenization_utils_test.py |  6 +-
 transformers/tests/tokenization_xlm_test.py   |  4 +-
 transformers/tests/tokenization_xlnet_test.py |  4 +-
 transformers/tests/utils.py                   | 64 +++++++++++++++++++
 transformers/tokenization_albert.py           |  8 +--
 transformers/tokenization_ctrl.py             |  6 +-
 transformers/tokenization_gpt2.py             | 12 ++--
 transformers/tokenization_openai.py           |  6 +-
 transformers/tokenization_utils.py            | 13 ++--
 transformers/tokenization_xlm.py              |  8 ++-
 transformers/tokenization_xlnet.py            |  4 +-
 50 files changed, 344 insertions(+), 231 deletions(-)
 delete mode 100644 transformers/tests/conftest.py
 create mode 100644 transformers/tests/utils.py

diff --git a/README.md b/README.md
index ddeabe08d6..64ec631651 100644
--- a/README.md
+++ b/README.md
@@ -101,17 +101,26 @@ pip install [--editable] .
 
 A series of tests are included for the library and the example scripts. Library tests can be found in the [tests folder](https://github.com/huggingface/transformers/tree/master/transformers/tests) and examples tests in the [examples folder](https://github.com/huggingface/transformers/tree/master/examples).
 
-These tests can be run using `pytest` (install pytest if needed with `pip install pytest`).
+These tests can be run using `unittest` or `pytest` (install pytest if needed with `pip install pytest`).
 
 Depending on which framework is installed (TensorFlow 2.0 and/or PyTorch), the irrelevant tests will be skipped. Ensure that both frameworks are installed if you want to execute all tests.
 
 You can run the tests from the root of the cloned repository with the commands:
 
+```bash
+python -m unittest discover -s transformers/tests -p "*test.py" -t .
+python -m unittest discover -s examples -p "*test.py" -t examples
+```
+
+or
+
 ```bash
 python -m pytest -sv ./transformers/tests/
 python -m pytest -sv ./examples/
 ```
 
+By default, slow tests are skipped. Set the `RUN_SLOW` environment variable to `yes` to run them.
+
 ### Do you want to run a Transformer model on a mobile device?
 
 You should check out our [`swift-coreml-transformers`](https://github.com/huggingface/swift-coreml-transformers) repo.
diff --git a/docs/source/installation.md b/docs/source/installation.md
index 11beb1ab3a..6263f7604d 100644
--- a/docs/source/installation.md
+++ b/docs/source/installation.md
@@ -24,15 +24,24 @@ pip install [--editable] .
 
 An extensive test suite is included to test the library behavior and several examples. Library tests can be found in the [tests folder](https://github.com/huggingface/transformers/tree/master/transformers/tests) and examples tests in the [examples folder](https://github.com/huggingface/transformers/tree/master/examples).
 
-Tests can be run using `pytest` (install pytest if needed with `pip install pytest`).
+Tests can be run using `unittest` or `pytest` (install pytest if needed with `pip install pytest`).
 
 Run all the tests from the root of the cloned repository with the commands:
 
+```bash
+python -m unittest discover -s transformers/tests -p "*test.py" -t .
+python -m unittest discover -s examples -p "*test.py" -t examples
+```
+
+or
+
 ``` bash
 python -m pytest -sv ./transformers/tests/
 python -m pytest -sv ./examples/
 ```
 
+By default, slow tests are skipped. Set the `RUN_SLOW` environment variable to `yes` to run them.
+
 ## OpenAI GPT original tokenization workflow
 
 If you want to reproduce the original tokenization process of the `OpenAI GPT` paper, you will need to install `ftfy` (use version 4.4.3 if you are using Python 2) and `SpaCy`:
diff --git a/setup.py b/setup.py
index 25f503f8d0..c4af32df83 100644
--- a/setup.py
+++ b/setup.py
@@ -72,7 +72,6 @@ setup(
         'transformers-cli'
     ],
     # python_requires='>=3.5.0',
-    tests_require=['pytest'],
     classifiers=[
           'Intended Audience :: Science/Research',
           'License :: OSI Approved :: Apache Software License',
diff --git a/templates/adding_a_new_model/tests/modeling_tf_xxx_test.py b/templates/adding_a_new_model/tests/modeling_tf_xxx_test.py
index 90837ca1ea..d7e576bf8b 100644
--- a/templates/adding_a_new_model/tests/modeling_tf_xxx_test.py
+++ b/templates/adding_a_new_model/tests/modeling_tf_xxx_test.py
@@ -18,11 +18,11 @@ from __future__ import print_function
 
 import unittest
 import shutil
-import pytest
 import sys
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_tf, slow
 
 from transformers import XxxConfig, is_tf_available
 
@@ -33,10 +33,9 @@ if is_tf_available():
                                                TFXxxForTokenClassification,
                                                TFXxxForQuestionAnswering,
                                                TF_XXX_PRETRAINED_MODEL_ARCHIVE_MAP)
-else:
-    pytestmark = pytest.mark.skip("Require TensorFlow")
 
 
+@require_tf
 class TFXxxModelTest(TFCommonTestCases.TFCommonModelTester):
 
     all_model_classes = (TFXxxModel, TFXxxForMaskedLM, TFXxxForQuestionAnswering,
@@ -244,7 +243,7 @@ class TFXxxModelTest(TFCommonTestCases.TFCommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_xxx_for_token_classification(*config_and_inputs)
 
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
         for model_name in ['xxx-base-uncased']:
diff --git a/templates/adding_a_new_model/tests/modeling_xxx_test.py b/templates/adding_a_new_model/tests/modeling_xxx_test.py
index 8c0cc3cf32..bfc70921cd 100644
--- a/templates/adding_a_new_model/tests/modeling_xxx_test.py
+++ b/templates/adding_a_new_model/tests/modeling_xxx_test.py
@@ -18,12 +18,12 @@ from __future__ import print_function
 
 import unittest
 import shutil
-import pytest
 
 from transformers import is_torch_available
 
 from .modeling_common_test import (CommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_torch, slow, torch_device
 
 if is_torch_available():
     from transformers import (XxxConfig, XxxModel, XxxForMaskedLM,
@@ -31,10 +31,9 @@ if is_torch_available():
                                         XxxForQuestionAnswering, XxxForSequenceClassification,
                                         XxxForTokenClassification, XxxForMultipleChoice)
     from transformers.modeling_xxx import XXX_PRETRAINED_MODEL_ARCHIVE_MAP
-else:
-    pytestmark = pytest.mark.skip("Require Torch")
 
 
+@require_torch
 class XxxModelTest(CommonTestCases.CommonModelTester):
 
     all_model_classes = (XxxModel, XxxForMaskedLM, XxxForQuestionAnswering,
@@ -131,6 +130,7 @@ class XxxModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_xxx_model(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
             model = XxxModel(config=config)
+            model.to(torch_device)
             model.eval()
             sequence_output, pooled_output = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids)
             sequence_output, pooled_output = model(input_ids, token_type_ids=token_type_ids)
@@ -148,6 +148,7 @@ class XxxModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_xxx_for_masked_lm(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
             model = XxxForMaskedLM(config=config)
+            model.to(torch_device)
             model.eval()
             loss, prediction_scores = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, masked_lm_labels=token_labels)
             result = {
@@ -162,6 +163,7 @@ class XxxModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_xxx_for_question_answering(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
             model = XxxForQuestionAnswering(config=config)
+            model.to(torch_device)
             model.eval()
             loss, start_logits, end_logits = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids,
                                                    start_positions=sequence_labels, end_positions=sequence_labels)
@@ -182,6 +184,7 @@ class XxxModelTest(CommonTestCases.CommonModelTester):
         def create_and_check_xxx_for_sequence_classification(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
             config.num_labels = self.num_labels
             model = XxxForSequenceClassification(config)
+            model.to(torch_device)
             model.eval()
             loss, logits = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, labels=sequence_labels)
             result = {
@@ -197,6 +200,7 @@ class XxxModelTest(CommonTestCases.CommonModelTester):
         def create_and_check_xxx_for_token_classification(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
             config.num_labels = self.num_labels
             model = XxxForTokenClassification(config=config)
+            model.to(torch_device)
             model.eval()
             loss, logits = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, labels=token_labels)
             result = {
@@ -243,7 +247,7 @@ class XxxModelTest(CommonTestCases.CommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_xxx_for_token_classification(*config_and_inputs)
 
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
         for model_name in list(XXX_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/modeling_openai.py b/transformers/modeling_openai.py
index e88f55c3ea..4fe7ffee8b 100644
--- a/transformers/modeling_openai.py
+++ b/transformers/modeling_openai.py
@@ -50,8 +50,10 @@ def load_tf_weights_in_openai_gpt(model, config, openai_checkpoint_folder_path):
 
     logger.info("Loading weights from {}".format(openai_checkpoint_folder_path))
 
-    names = json.load(open(openai_checkpoint_folder_path + '/parameters_names.json', "r", encoding='utf-8'))
-    shapes = json.load(open(openai_checkpoint_folder_path + '/params_shapes.json', "r", encoding='utf-8'))
+    with open(openai_checkpoint_folder_path + '/parameters_names.json', "r", encoding='utf-8') as names_handle:
+        names = json.load(names_handle)
+    with open(openai_checkpoint_folder_path + '/params_shapes.json', "r", encoding='utf-8') as shapes_handle:
+        shapes = json.load(shapes_handle)
     offsets = np.cumsum([np.prod(shape) for shape in shapes])
     init_params = [np.load(openai_checkpoint_folder_path + '/params_{}.npy'.format(n)) for n in range(10)]
     init_params = np.split(np.concatenate(init_params, 0), offsets)[:-1]
diff --git a/transformers/tests/conftest.py b/transformers/tests/conftest.py
deleted file mode 100644
index f809234cd5..0000000000
--- a/transformers/tests/conftest.py
+++ /dev/null
@@ -1,31 +0,0 @@
-# content of conftest.py
-
-import pytest
-
-
-def pytest_addoption(parser):
-    parser.addoption(
-        "--runslow", action="store_true", default=False, help="run slow tests"
-    )
-    parser.addoption(
-        "--use_cuda", action="store_true", default=False, help="run tests on gpu"
-    )
-
-
-def pytest_configure(config):
-    config.addinivalue_line("markers", "slow: mark test as slow to run")
-
-
-def pytest_collection_modifyitems(config, items):
-    if config.getoption("--runslow"):
-        # --runslow given in cli: do not skip slow tests
-        return
-    skip_slow = pytest.mark.skip(reason="need --runslow option to run")
-    for item in items:
-        if "slow" in item.keywords:
-            item.add_marker(skip_slow)
-
-@pytest.fixture
-def use_cuda(request):
-    """ Run test on gpu """
-    return request.config.getoption("--use_cuda")
diff --git a/transformers/tests/modeling_albert_test.py b/transformers/tests/modeling_albert_test.py
index 976feff9db..a14d66ae8f 100644
--- a/transformers/tests/modeling_albert_test.py
+++ b/transformers/tests/modeling_albert_test.py
@@ -18,22 +18,21 @@ from __future__ import print_function
 
 import unittest
 import shutil
-import pytest
 
 from transformers import is_torch_available
 
 from .modeling_common_test import (CommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_torch, slow, torch_device
 
 if is_torch_available():
     from transformers import (AlbertConfig, AlbertModel, AlbertForMaskedLM,
                               AlbertForSequenceClassification, AlbertForQuestionAnswering,
                               )
     from transformers.modeling_albert import ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP
-else:
-    pytestmark = pytest.mark.skip("Require Torch")
 
 
+@require_torch
 class AlbertModelTest(CommonTestCases.CommonModelTester):
 
     all_model_classes = (AlbertModel, AlbertForMaskedLM) if is_torch_available() else ()
@@ -133,6 +132,7 @@ class AlbertModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_albert_model(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
             model = AlbertModel(config=config)
+            model.to(torch_device)
             model.eval()
             sequence_output, pooled_output = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids)
             sequence_output, pooled_output = model(input_ids, token_type_ids=token_type_ids)
@@ -150,6 +150,7 @@ class AlbertModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_albert_for_masked_lm(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
             model = AlbertForMaskedLM(config=config)
+            model.to(torch_device)
             model.eval()
             loss, prediction_scores = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, masked_lm_labels=token_labels)
             result = {
@@ -163,6 +164,7 @@ class AlbertModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_albert_for_question_answering(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
             model = AlbertForQuestionAnswering(config=config)
+            model.to(torch_device)
             model.eval()
             loss, start_logits, end_logits = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids,
                                                    start_positions=sequence_labels, end_positions=sequence_labels)
@@ -183,6 +185,7 @@ class AlbertModelTest(CommonTestCases.CommonModelTester):
         def create_and_check_albert_for_sequence_classification(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
             config.num_labels = self.num_labels
             model = AlbertForSequenceClassification(config)
+            model.to(torch_device)
             model.eval()
             loss, logits = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, labels=sequence_labels)
             result = {
@@ -225,7 +228,7 @@ class AlbertModelTest(CommonTestCases.CommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_albert_for_sequence_classification(*config_and_inputs)
 
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
         for model_name in list(ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/modeling_auto_test.py b/transformers/tests/modeling_auto_test.py
index 6d2c7ec979..9b7d920bc8 100644
--- a/transformers/tests/modeling_auto_test.py
+++ b/transformers/tests/modeling_auto_test.py
@@ -18,11 +18,12 @@ from __future__ import print_function
 
 import unittest
 import shutil
-import pytest
 import logging
 
 from transformers import is_torch_available
 
+from .utils import require_torch, slow
+
 if is_torch_available():
     from transformers import (AutoConfig, BertConfig,
                                     AutoModel, BertModel,
@@ -33,12 +34,11 @@ if is_torch_available():
 
     from .modeling_common_test import (CommonTestCases, ids_tensor)
     from .configuration_common_test import ConfigTester
-else:
-    pytestmark = pytest.mark.skip("Require Torch")
 
 
+@require_torch
 class AutoModelTest(unittest.TestCase):
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         logging.basicConfig(level=logging.INFO)
         for model_name in list(BERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
@@ -53,7 +53,7 @@ class AutoModelTest(unittest.TestCase):
             for value in loading_info.values():
                 self.assertEqual(len(value), 0)
 
-    @pytest.mark.slow
+    @slow
     def test_lmhead_model_from_pretrained(self):
         logging.basicConfig(level=logging.INFO)
         for model_name in list(BERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
@@ -66,7 +66,7 @@ class AutoModelTest(unittest.TestCase):
             self.assertIsNotNone(model)
             self.assertIsInstance(model, BertForMaskedLM)
 
-    @pytest.mark.slow
+    @slow
     def test_sequence_classification_model_from_pretrained(self):
         logging.basicConfig(level=logging.INFO)
         for model_name in list(BERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
@@ -79,7 +79,7 @@ class AutoModelTest(unittest.TestCase):
             self.assertIsNotNone(model)
             self.assertIsInstance(model, BertForSequenceClassification)
 
-    @pytest.mark.slow
+    @slow
     def test_question_answering_model_from_pretrained(self):
         logging.basicConfig(level=logging.INFO)
         for model_name in list(BERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/modeling_bert_test.py b/transformers/tests/modeling_bert_test.py
index 6c93c9a187..539f66cd3f 100644
--- a/transformers/tests/modeling_bert_test.py
+++ b/transformers/tests/modeling_bert_test.py
@@ -18,12 +18,12 @@ from __future__ import print_function
 
 import unittest
 import shutil
-import pytest
 
 from transformers import is_torch_available
 
 from .modeling_common_test import (CommonTestCases, ids_tensor, floats_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_torch, slow, torch_device
 
 if is_torch_available():
     from transformers import (BertConfig, BertModel, BertForMaskedLM,
@@ -31,11 +31,9 @@ if is_torch_available():
                               BertForQuestionAnswering, BertForSequenceClassification,
                               BertForTokenClassification, BertForMultipleChoice)
     from transformers.modeling_bert import BERT_PRETRAINED_MODEL_ARCHIVE_MAP
-else:
-    pytestmark = pytest.mark.skip("Require Torch")
 
 
-@pytest.mark.usefixtures("use_cuda")
+@require_torch
 class BertModelTest(CommonTestCases.CommonModelTester):
 
     all_model_classes = (BertModel, BertForMaskedLM, BertForNextSentencePrediction,
@@ -67,7 +65,6 @@ class BertModelTest(CommonTestCases.CommonModelTester):
                      num_labels=3,
                      num_choices=4,
                      scope=None,
-                     device='cpu',
                      ):
             self.parent = parent
             self.batch_size = batch_size
@@ -91,26 +88,25 @@ class BertModelTest(CommonTestCases.CommonModelTester):
             self.num_labels = num_labels
             self.num_choices = num_choices
             self.scope = scope
-            self.device = device
 
         def prepare_config_and_inputs(self):
-            input_ids = ids_tensor([self.batch_size, self.seq_length], self.vocab_size).to(self.device)
+            input_ids = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)
 
             input_mask = None
             if self.use_input_mask:
-                input_mask = ids_tensor([self.batch_size, self.seq_length], vocab_size=2).to(self.device)
+                input_mask = ids_tensor([self.batch_size, self.seq_length], vocab_size=2)
 
             token_type_ids = None
             if self.use_token_type_ids:
-                token_type_ids = ids_tensor([self.batch_size, self.seq_length], self.type_vocab_size).to(self.device)
+                token_type_ids = ids_tensor([self.batch_size, self.seq_length], self.type_vocab_size)
 
             sequence_labels = None
             token_labels = None
             choice_labels = None
             if self.use_labels:
-                sequence_labels = ids_tensor([self.batch_size], self.type_sequence_label_size).to(self.device)
-                token_labels = ids_tensor([self.batch_size, self.seq_length], self.num_labels).to(self.device)
-                choice_labels = ids_tensor([self.batch_size], self.num_choices).to(self.device)
+                sequence_labels = ids_tensor([self.batch_size], self.type_sequence_label_size)
+                token_labels = ids_tensor([self.batch_size, self.seq_length], self.num_labels)
+                choice_labels = ids_tensor([self.batch_size], self.num_choices)
 
             config = BertConfig(
                 vocab_size_or_config_json_file=self.vocab_size,
@@ -144,7 +140,7 @@ class BertModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_bert_model(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
             model = BertModel(config=config)
-            model.to(input_ids.device)
+            model.to(torch_device)
             model.eval()
             sequence_output, pooled_output = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids)
             sequence_output, pooled_output = model(input_ids, token_type_ids=token_type_ids)
@@ -161,6 +157,7 @@ class BertModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_bert_model_as_decoder(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels, encoder_hidden_states, encoder_attention_mask):
             model = BertModel(config)
+            model.to(torch_device)
             model.eval()
             sequence_output, pooled_output = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, encoder_hidden_states=encoder_hidden_states, encoder_attention_mask=encoder_attention_mask)
             sequence_output, pooled_output = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, encoder_hidden_states=encoder_hidden_states)
@@ -177,6 +174,7 @@ class BertModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_bert_for_masked_lm(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
             model = BertForMaskedLM(config=config)
+            model.to(torch_device)
             model.eval()
             loss, prediction_scores = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, masked_lm_labels=token_labels)
             result = {
@@ -190,6 +188,7 @@ class BertModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_bert_model_for_masked_lm_as_decoder(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels, encoder_hidden_states, encoder_attention_mask):
             model = BertForMaskedLM(config=config)
+            model.to(torch_device)
             model.eval()
             loss, prediction_scores = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, masked_lm_labels=token_labels, encoder_hidden_states=encoder_hidden_states, encoder_attention_mask=encoder_attention_mask)
             loss, prediction_scores = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, masked_lm_labels=token_labels, encoder_hidden_states=encoder_hidden_states)
@@ -204,6 +203,7 @@ class BertModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_bert_for_next_sequence_prediction(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
             model = BertForNextSentencePrediction(config=config)
+            model.to(torch_device)
             model.eval()
             loss, seq_relationship_score = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, next_sentence_label=sequence_labels)
             result = {
@@ -217,6 +217,7 @@ class BertModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_bert_for_pretraining(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
             model = BertForPreTraining(config=config)
+            model.to(torch_device)
             model.eval()
             loss, prediction_scores, seq_relationship_score = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids,
                                                                     masked_lm_labels=token_labels, next_sentence_label=sequence_labels)
@@ -235,6 +236,7 @@ class BertModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_bert_for_question_answering(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
             model = BertForQuestionAnswering(config=config)
+            model.to(torch_device)
             model.eval()
             loss, start_logits, end_logits = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids,
                                                    start_positions=sequence_labels, end_positions=sequence_labels)
@@ -254,6 +256,7 @@ class BertModelTest(CommonTestCases.CommonModelTester):
         def create_and_check_bert_for_sequence_classification(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
             config.num_labels = self.num_labels
             model = BertForSequenceClassification(config)
+            model.to(torch_device)
             model.eval()
             loss, logits = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, labels=sequence_labels)
             result = {
@@ -268,6 +271,7 @@ class BertModelTest(CommonTestCases.CommonModelTester):
         def create_and_check_bert_for_token_classification(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
             config.num_labels = self.num_labels
             model = BertForTokenClassification(config=config)
+            model.to(torch_device)
             model.eval()
             loss, logits = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, labels=token_labels)
             result = {
@@ -282,6 +286,7 @@ class BertModelTest(CommonTestCases.CommonModelTester):
         def create_and_check_bert_for_multiple_choice(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
             config.num_choices = self.num_choices
             model = BertForMultipleChoice(config=config)
+            model.to(torch_device)
             model.eval()
             multiple_choice_inputs_ids = input_ids.unsqueeze(1).expand(-1, self.num_choices, -1).contiguous()
             multiple_choice_token_type_ids = token_type_ids.unsqueeze(1).expand(-1, self.num_choices, -1).contiguous()
@@ -313,10 +318,7 @@ class BertModelTest(CommonTestCases.CommonModelTester):
     def test_config(self):
         self.config_tester.run_common_tests()
 
-    def test_bert_model(self, use_cuda=False):
-        # ^^ This could be a real fixture
-        if use_cuda:
-            self.model_tester.device = "cuda"
+    def test_bert_model(self):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_bert_model(*config_and_inputs)
 
@@ -356,7 +358,7 @@ class BertModelTest(CommonTestCases.CommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_bert_for_token_classification(*config_and_inputs)
 
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
         for model_name in list(BERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/modeling_common_test.py b/transformers/tests/modeling_common_test.py
index baf1531403..80d5d95455 100644
--- a/transformers/tests/modeling_common_test.py
+++ b/transformers/tests/modeling_common_test.py
@@ -27,10 +27,11 @@ import uuid
 
 import unittest
 import logging
-import pytest
 
 from transformers import is_torch_available
 
+from .utils import require_torch, slow, torch_device
+
 if is_torch_available():
     import torch
     import numpy as np
@@ -38,8 +39,6 @@ if is_torch_available():
     from transformers import (AdaptiveEmbedding, PretrainedConfig, PreTrainedModel,
                                     BertModel, BertConfig, BERT_PRETRAINED_MODEL_ARCHIVE_MAP,
                                     GPT2LMHeadModel, GPT2Config, GPT2_PRETRAINED_MODEL_ARCHIVE_MAP)
-else:
-    pytestmark = pytest.mark.skip("Require Torch")
 
 if sys.version_info[0] == 2:
     import cPickle as pickle
@@ -65,6 +64,7 @@ def _config_zero_init(config):
 
 class CommonTestCases:
 
+    @require_torch
     class CommonModelTester(unittest.TestCase):
 
         model_tester = None
@@ -79,6 +79,7 @@ class CommonTestCases:
 
             for model_class in self.all_model_classes:
                 model = model_class(config)
+                model.to(torch_device)
                 model.eval()
                 with torch.no_grad():
                     outputs = model(**inputs_dict)
@@ -86,12 +87,13 @@ class CommonTestCases:
                 with TemporaryDirectory() as tmpdirname:
                     model.save_pretrained(tmpdirname)
                     model = model_class.from_pretrained(tmpdirname)
+                    model.to(torch_device)
                     with torch.no_grad():
                         after_outputs = model(**inputs_dict)
 
                     # Make sure we don't have nans
-                    out_1 = after_outputs[0].numpy()
-                    out_2 = outputs[0].numpy()
+                    out_1 = after_outputs[0].cpu().numpy()
+                    out_2 = outputs[0].cpu().numpy()
                     out_1 = out_1[~np.isnan(out_1)]
                     out_2 = out_2[~np.isnan(out_2)]
                     max_diff = np.amax(np.abs(out_1 - out_2))
@@ -113,6 +115,7 @@ class CommonTestCases:
 
             for model_class in self.all_model_classes:
                 model = model_class(config)
+                model.to(torch_device)
                 model.eval()
                 first, second = model(inputs_dict["input_ids"])[0], model(inputs_dict["input_ids"])[0]
                 self.assertEqual(first.ne(second).sum().item(), 0)
@@ -125,6 +128,7 @@ class CommonTestCases:
                 config.output_attentions = True
                 config.output_hidden_states = False
                 model = model_class(config)
+                model.to(torch_device)
                 model.eval()
                 outputs = model(**inputs_dict)
                 attentions = outputs[-1]
@@ -142,6 +146,7 @@ class CommonTestCases:
                 config.output_attentions = True
                 config.output_hidden_states = True
                 model = model_class(config)
+                model.to(torch_device)
                 model.eval()
                 outputs = model(**inputs_dict)
                 self.assertEqual(out_len+1, len(outputs))
@@ -181,6 +186,7 @@ class CommonTestCases:
             configs_no_init.torchscript = True
             for model_class in self.all_model_classes:
                 model = model_class(config=configs_no_init)
+                model.to(torch_device)
                 model.eval()
                 inputs = inputs_dict['input_ids']  # Let's keep only input_ids
 
@@ -201,7 +207,10 @@ class CommonTestCases:
                 except ValueError:
                     self.fail("Couldn't load module.")
 
+                model.to(torch_device)
                 model.eval()
+
+                loaded_model.to(torch_device)
                 loaded_model.eval()
 
                 model_params = model.parameters()
@@ -228,11 +237,12 @@ class CommonTestCases:
             configs_no_init = _config_zero_init(config)  # To be sure we have no Nan
             for model_class in self.all_model_classes:
                 model = model_class(config=configs_no_init)
+                model.to(torch_device)
                 model.eval()
 
                 # Prepare head_mask
                 # Set require_grad after having prepared the tensor to avoid error (leaf variable has been moved into the graph interior)
-                head_mask = torch.ones(self.model_tester.num_hidden_layers, self.model_tester.num_attention_heads)
+                head_mask = torch.ones(self.model_tester.num_hidden_layers, self.model_tester.num_attention_heads, device=torch_device)
                 head_mask[0, 0] = 0
                 head_mask[-1, :-1] = 0
                 head_mask.requires_grad_(requires_grad=True)
@@ -282,6 +292,7 @@ class CommonTestCases:
                 config.output_attentions = True
                 config.output_hidden_states = False
                 model = model_class(config=config)
+                model.to(torch_device)
                 model.eval()
                 heads_to_prune = {0: list(range(1, self.model_tester.num_attention_heads)),
                                 -1: [0]}
@@ -310,6 +321,7 @@ class CommonTestCases:
                 config.output_attentions = True
                 config.output_hidden_states = False
                 model = model_class(config=config)
+                model.to(torch_device)
                 model.eval()
                 heads_to_prune = {0: list(range(1, self.model_tester.num_attention_heads)),
                                 -1: [0]}
@@ -319,6 +331,7 @@ class CommonTestCases:
                     os.makedirs(directory)
                 model.save_pretrained(directory)
                 model = model_class.from_pretrained(directory)
+                model.to(torch_device)
 
                 outputs = model(**inputs_dict)
                 attentions = outputs[-1]
@@ -346,6 +359,7 @@ class CommonTestCases:
                 config.pruned_heads = heads_to_prune
 
                 model = model_class(config=config)
+                model.to(torch_device)
                 model.eval()
 
                 outputs = model(**inputs_dict)
@@ -372,6 +386,7 @@ class CommonTestCases:
                 config.pruned_heads = heads_to_prune
 
                 model = model_class(config=config)
+                model.to(torch_device)
                 model.eval()
 
                 outputs = model(**inputs_dict)
@@ -388,6 +403,7 @@ class CommonTestCases:
                     os.makedirs(directory)
                 model.save_pretrained(directory)
                 model = model_class.from_pretrained(directory)
+                model.to(torch_device)
                 shutil.rmtree(directory)
 
                 outputs = model(**inputs_dict)
@@ -419,6 +435,7 @@ class CommonTestCases:
                 config.output_hidden_states = True
                 config.output_attentions = False
                 model = model_class(config)
+                model.to(torch_device)
                 model.eval()
                 outputs = model(**inputs_dict)
                 hidden_states = outputs[-1]
@@ -538,6 +555,7 @@ class CommonTestCases:
 
             for model_class in self.all_model_classes:
                 model = model_class(config)
+                model.to(torch_device)
                 model.eval()
 
                 wte = model.get_input_embeddings()
@@ -628,6 +646,7 @@ class CommonTestCases:
         def create_and_check_base_model(self, config, input_ids, token_type_ids, position_ids,
                                 mc_labels, lm_labels, mc_token_ids):
             model = self.base_model_class(config)
+            model.to(torch_device)
             model.eval()
 
             outputs = model(input_ids, position_ids, token_type_ids)
@@ -643,6 +662,7 @@ class CommonTestCases:
         def create_and_check_lm_head(self, config, input_ids, token_type_ids, position_ids,
                                         mc_labels, lm_labels, mc_token_ids):
             model = self.lm_head_model_class(config)
+            model.to(torch_device)
             model.eval()
             outputs = model(input_ids, position_ids, token_type_ids, lm_labels)
             loss, lm_logits = outputs[:2]
@@ -659,6 +679,7 @@ class CommonTestCases:
                                         mc_labels, lm_labels, mc_token_ids):
             for model_class in self.all_model_classes:
                 model = model_class(config)
+                model.to(torch_device)
                 model.eval()
                 outputs = model(input_ids)
                 presents = outputs[-1]
@@ -671,6 +692,7 @@ class CommonTestCases:
         def create_and_check_double_heads(self, config, input_ids, token_type_ids, position_ids,
                                         mc_labels, lm_labels, mc_token_ids):
             model = self.double_head_model_class(config)
+            model.to(torch_device)
             model.eval()
             outputs = model(input_ids, mc_token_ids, lm_labels=lm_labels, mc_labels=mc_labels,
                             token_type_ids=token_type_ids, position_ids=position_ids)
@@ -716,7 +738,7 @@ class CommonTestCases:
                 config_and_inputs = self.prepare_config_and_inputs()
                 self.create_and_check_presents(*config_and_inputs)
 
-        @pytest.mark.slow
+        @slow
         def run_slow_tests(self):
             self.create_and_check_model_from_pretrained()
 
@@ -770,7 +792,7 @@ def ids_tensor(shape, vocab_size, rng=None, name=None):
     for _ in range(total_dims):
         values.append(rng.randint(0, vocab_size - 1))
 
-    return torch.tensor(data=values, dtype=torch.long).view(shape).contiguous()
+    return torch.tensor(data=values, dtype=torch.long, device=torch_device).view(shape).contiguous()
 
 
 def floats_tensor(shape, scale=1.0, rng=None, name=None):
@@ -786,11 +808,12 @@ def floats_tensor(shape, scale=1.0, rng=None, name=None):
     for _ in range(total_dims):
         values.append(rng.random() * scale)
 
-    return torch.tensor(data=values, dtype=torch.float).view(shape).contiguous()
+    return torch.tensor(data=values, dtype=torch.float, device=torch_device).view(shape).contiguous()
 
 
+@require_torch
 class ModelUtilsTest(unittest.TestCase):
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         logging.basicConfig(level=logging.INFO)
         for model_name in list(BERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/modeling_ctrl_test.py b/transformers/tests/modeling_ctrl_test.py
index 47ff8d8d51..8c14578a5c 100644
--- a/transformers/tests/modeling_ctrl_test.py
+++ b/transformers/tests/modeling_ctrl_test.py
@@ -16,7 +16,6 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import pytest
 import shutil
 import pdb
 
@@ -25,13 +24,13 @@ from transformers import is_torch_available
 if is_torch_available():
     from transformers import (CTRLConfig, CTRLModel, CTRL_PRETRAINED_MODEL_ARCHIVE_MAP,
                                     CTRLLMHeadModel)
-else:
-    pytestmark = pytest.mark.skip("Require Torch")
 
 from .modeling_common_test import (CommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_torch, slow, torch_device
 
 
+@require_torch
 class CTRLModelTest(CommonTestCases.CommonModelTester):
 
     all_model_classes = (CTRLModel, CTRLLMHeadModel) if is_torch_available() else ()
@@ -140,6 +139,7 @@ class CTRLModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_ctrl_model(self, config, input_ids, input_mask, head_mask, token_type_ids, *args):
             model = CTRLModel(config=config)
+            model.to(torch_device)
             model.eval()
 
             model(input_ids, token_type_ids=token_type_ids, head_mask=head_mask)
@@ -157,6 +157,7 @@ class CTRLModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_lm_head_model(self, config, input_ids, input_mask, head_mask, token_type_ids, *args):
             model = CTRLLMHeadModel(config)
+            model.to(torch_device)
             model.eval()
 
             loss, lm_logits, _ = model(input_ids, token_type_ids=token_type_ids, labels=input_ids)
@@ -202,7 +203,7 @@ class CTRLModelTest(CommonTestCases.CommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_lm_head_model(*config_and_inputs)
 
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
         for model_name in list(CTRL_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/modeling_distilbert_test.py b/transformers/tests/modeling_distilbert_test.py
index 8099c03586..4b8f64327d 100644
--- a/transformers/tests/modeling_distilbert_test.py
+++ b/transformers/tests/modeling_distilbert_test.py
@@ -17,7 +17,6 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import pytest
 
 from transformers import is_torch_available
 
@@ -25,13 +24,13 @@ if is_torch_available():
     from transformers import (DistilBertConfig, DistilBertModel, DistilBertForMaskedLM,
                                     DistilBertForTokenClassification,
                                     DistilBertForQuestionAnswering, DistilBertForSequenceClassification)
-else:
-    pytestmark = pytest.mark.skip("Require Torch")
 
 from .modeling_common_test import (CommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_torch, slow, torch_device
 
 
+@require_torch
 class DistilBertModelTest(CommonTestCases.CommonModelTester):
 
     all_model_classes = (DistilBertModel, DistilBertForMaskedLM, DistilBertForQuestionAnswering,
@@ -126,6 +125,7 @@ class DistilBertModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_distilbert_model(self, config, input_ids, input_mask, sequence_labels, token_labels, choice_labels):
             model = DistilBertModel(config=config)
+            model.to(torch_device)
             model.eval()
             (sequence_output,) = model(input_ids, input_mask)
             (sequence_output,) = model(input_ids)
@@ -139,6 +139,7 @@ class DistilBertModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_distilbert_for_masked_lm(self, config, input_ids, input_mask, sequence_labels, token_labels, choice_labels):
             model = DistilBertForMaskedLM(config=config)
+            model.to(torch_device)
             model.eval()
             loss, prediction_scores = model(input_ids, attention_mask=input_mask, masked_lm_labels=token_labels)
             result = {
@@ -152,6 +153,7 @@ class DistilBertModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_distilbert_for_question_answering(self, config, input_ids, input_mask, sequence_labels, token_labels, choice_labels):
             model = DistilBertForQuestionAnswering(config=config)
+            model.to(torch_device)
             model.eval()
             loss, start_logits, end_logits = model(input_ids, attention_mask=input_mask, start_positions=sequence_labels, end_positions=sequence_labels)
             result = {
@@ -170,6 +172,7 @@ class DistilBertModelTest(CommonTestCases.CommonModelTester):
         def create_and_check_distilbert_for_sequence_classification(self, config, input_ids, input_mask, sequence_labels, token_labels, choice_labels):
             config.num_labels = self.num_labels
             model = DistilBertForSequenceClassification(config)
+            model.to(torch_device)
             model.eval()
             loss, logits = model(input_ids, attention_mask=input_mask, labels=sequence_labels)
             result = {
@@ -184,6 +187,7 @@ class DistilBertModelTest(CommonTestCases.CommonModelTester):
         def create_and_check_distilbert_for_token_classification(self, config, input_ids, input_mask, sequence_labels, token_labels, choice_labels):
             config.num_labels = self.num_labels
             model = DistilBertForTokenClassification(config=config)
+            model.to(torch_device)
             model.eval()
 
             loss, logits = model(input_ids, attention_mask=input_mask, labels=token_labels)
@@ -229,7 +233,7 @@ class DistilBertModelTest(CommonTestCases.CommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_distilbert_for_token_classification(*config_and_inputs)
 
-    # @pytest.mark.slow
+    # @slow
     # def test_model_from_pretrained(self):
     #     cache_dir = "/tmp/transformers_test/"
     #     for model_name in list(DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/modeling_encoder_decoder_test.py b/transformers/tests/modeling_encoder_decoder_test.py
index a6c88ed9a9..64e86df8f5 100644
--- a/transformers/tests/modeling_encoder_decoder_test.py
+++ b/transformers/tests/modeling_encoder_decoder_test.py
@@ -15,19 +15,18 @@
 
 import logging
 import unittest
-import pytest
 
 from transformers import is_torch_available
+from .utils import require_torch, slow
 
 if is_torch_available():
     from transformers import BertModel, BertForMaskedLM, Model2Model
     from transformers.modeling_bert import BERT_PRETRAINED_MODEL_ARCHIVE_MAP
-else:
-    pytestmark = pytest.mark.skip("Require Torch")
 
 
+@require_torch
 class EncoderDecoderModelTest(unittest.TestCase):
-    @pytest.mark.slow
+    @slow
     def test_model2model_from_pretrained(self):
         logging.basicConfig(level=logging.INFO)
         for model_name in list(BERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/modeling_gpt2_test.py b/transformers/tests/modeling_gpt2_test.py
index 4263e51bc9..ecaa2a4bd0 100644
--- a/transformers/tests/modeling_gpt2_test.py
+++ b/transformers/tests/modeling_gpt2_test.py
@@ -17,7 +17,6 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import pytest
 import shutil
 
 from transformers import is_torch_available
@@ -25,13 +24,13 @@ from transformers import is_torch_available
 if is_torch_available():
     from transformers import (GPT2Config, GPT2Model, GPT2_PRETRAINED_MODEL_ARCHIVE_MAP,
                                     GPT2LMHeadModel, GPT2DoubleHeadsModel)
-else:
-    pytestmark = pytest.mark.skip("Require Torch")
 
 from .modeling_common_test import (CommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_torch, slow, torch_device
 
 
+@require_torch
 class GPT2ModelTest(CommonTestCases.CommonModelTester):
 
     all_model_classes = (GPT2Model, GPT2LMHeadModel, GPT2DoubleHeadsModel) if is_torch_available() else ()
@@ -136,6 +135,7 @@ class GPT2ModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_gpt2_model(self, config, input_ids, input_mask, head_mask, token_type_ids, *args):
             model = GPT2Model(config=config)
+            model.to(torch_device)
             model.eval()
 
             model(input_ids, token_type_ids=token_type_ids, head_mask=head_mask)
@@ -153,6 +153,7 @@ class GPT2ModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_lm_head_model(self, config, input_ids, input_mask, head_mask, token_type_ids, *args):
             model = GPT2LMHeadModel(config)
+            model.to(torch_device)
             model.eval()
 
             loss, lm_logits, _ = model(input_ids, token_type_ids=token_type_ids, labels=input_ids)
@@ -171,6 +172,7 @@ class GPT2ModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_double_lm_head_model(self, config, input_ids, input_mask, head_mask, token_type_ids, mc_token_ids, *args):
             model = GPT2DoubleHeadsModel(config)
+            model.to(torch_device)
             model.eval()
 
 
@@ -235,7 +237,7 @@ class GPT2ModelTest(CommonTestCases.CommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_double_lm_head_model(*config_and_inputs)
 
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
         for model_name in list(GPT2_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/modeling_openai_test.py b/transformers/tests/modeling_openai_test.py
index 33218288a0..8e4d13438d 100644
--- a/transformers/tests/modeling_openai_test.py
+++ b/transformers/tests/modeling_openai_test.py
@@ -17,7 +17,6 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import pytest
 import shutil
 
 from transformers import is_torch_available
@@ -25,13 +24,13 @@ from transformers import is_torch_available
 if is_torch_available():
     from transformers import (OpenAIGPTConfig, OpenAIGPTModel, OPENAI_GPT_PRETRAINED_MODEL_ARCHIVE_MAP,
                                     OpenAIGPTLMHeadModel, OpenAIGPTDoubleHeadsModel)
-else:
-    pytestmark = pytest.mark.skip("Require Torch")
 
 from .modeling_common_test import (CommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_torch, slow, torch_device
 
 
+@require_torch
 class OpenAIGPTModelTest(CommonTestCases.CommonModelTester):
 
     all_model_classes = (OpenAIGPTModel, OpenAIGPTLMHeadModel, OpenAIGPTDoubleHeadsModel) if is_torch_available() else ()
@@ -124,6 +123,7 @@ class OpenAIGPTModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_openai_gpt_model(self, config, input_ids, head_mask, token_type_ids, *args):
             model = OpenAIGPTModel(config=config)
+            model.to(torch_device)
             model.eval()
 
             model(input_ids, token_type_ids=token_type_ids, head_mask=head_mask)
@@ -139,6 +139,7 @@ class OpenAIGPTModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_lm_head_model(self, config, input_ids, head_mask, token_type_ids, *args):
             model = OpenAIGPTLMHeadModel(config)
+            model.to(torch_device)
             model.eval()
 
             loss, lm_logits = model(input_ids, token_type_ids=token_type_ids, labels=input_ids)
@@ -157,6 +158,7 @@ class OpenAIGPTModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_double_lm_head_model(self, config, input_ids, head_mask, token_type_ids, *args):
             model = OpenAIGPTDoubleHeadsModel(config)
+            model.to(torch_device)
             model.eval()
 
             loss, lm_logits, mc_logits = model(input_ids, token_type_ids=token_type_ids, lm_labels=input_ids)
@@ -203,7 +205,7 @@ class OpenAIGPTModelTest(CommonTestCases.CommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_double_lm_head_model(*config_and_inputs)
 
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
         for model_name in list(OPENAI_GPT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/modeling_roberta_test.py b/transformers/tests/modeling_roberta_test.py
index 0620ddf630..7a3553b164 100644
--- a/transformers/tests/modeling_roberta_test.py
+++ b/transformers/tests/modeling_roberta_test.py
@@ -18,7 +18,6 @@ from __future__ import print_function
 
 import unittest
 import shutil
-import pytest
 
 from transformers import is_torch_available
 
@@ -27,13 +26,13 @@ if is_torch_available():
     from transformers import (RobertaConfig, RobertaModel, RobertaForMaskedLM,
                               RobertaForSequenceClassification, RobertaForTokenClassification)
     from transformers.modeling_roberta import ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP
-else:
-    pytestmark = pytest.mark.skip("Require Torch")
 
 from .modeling_common_test import (CommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_torch, slow, torch_device
 
 
+@require_torch
 class RobertaModelTest(CommonTestCases.CommonModelTester):
 
     all_model_classes = (RobertaForMaskedLM, RobertaModel) if is_torch_available() else ()
@@ -129,6 +128,7 @@ class RobertaModelTest(CommonTestCases.CommonModelTester):
         def create_and_check_roberta_model(self, config, input_ids, token_type_ids, input_mask, sequence_labels,
                                            token_labels, choice_labels):
             model = RobertaModel(config=config)
+            model.to(torch_device)
             model.eval()
             sequence_output, pooled_output = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids)
             sequence_output, pooled_output = model(input_ids, token_type_ids=token_type_ids)
@@ -146,6 +146,7 @@ class RobertaModelTest(CommonTestCases.CommonModelTester):
         def create_and_check_roberta_for_masked_lm(self, config, input_ids, token_type_ids, input_mask, sequence_labels,
                                                    token_labels, choice_labels):
             model = RobertaForMaskedLM(config=config)
+            model.to(torch_device)
             model.eval()
             loss, prediction_scores = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, masked_lm_labels=token_labels)
             result = {
@@ -161,6 +162,7 @@ class RobertaModelTest(CommonTestCases.CommonModelTester):
                                                               sequence_labels, token_labels, choice_labels):
             config.num_labels = self.num_labels
             model = RobertaForTokenClassification(config=config)
+            model.to(torch_device)
             model.eval()
             loss, logits = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids,
                                  labels=token_labels)
@@ -195,7 +197,7 @@ class RobertaModelTest(CommonTestCases.CommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_roberta_for_masked_lm(*config_and_inputs)
 
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
         for model_name in list(ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
@@ -207,10 +209,10 @@ class RobertaModelTest(CommonTestCases.CommonModelTester):
 
 class RobertaModelIntegrationTest(unittest.TestCase):
 
-    @pytest.mark.slow
+    @slow
     def test_inference_masked_lm(self):
         model = RobertaForMaskedLM.from_pretrained('roberta-base')
-        
+
         input_ids = torch.tensor([[    0, 31414,   232,   328,   740,  1140, 12695,    69, 46078,  1588,   2]])
         output = model(input_ids)[0]
         expected_shape = torch.Size((1, 11, 50265))
@@ -228,10 +230,10 @@ class RobertaModelIntegrationTest(unittest.TestCase):
             torch.allclose(output[:, :3, :3], expected_slice, atol=1e-3)
         )
 
-    @pytest.mark.slow
+    @slow
     def test_inference_no_head(self):
         model = RobertaModel.from_pretrained('roberta-base')
-        
+
         input_ids = torch.tensor([[    0, 31414,   232,   328,   740,  1140, 12695,    69, 46078,  1588,   2]])
         output = model(input_ids)[0]
         # compare the actual values for a slice.
@@ -244,10 +246,10 @@ class RobertaModelIntegrationTest(unittest.TestCase):
             torch.allclose(output[:, :3, :3], expected_slice, atol=1e-3)
         )
 
-    @pytest.mark.slow
+    @slow
     def test_inference_classification_head(self):
         model = RobertaForSequenceClassification.from_pretrained('roberta-large-mnli')
-        
+
         input_ids = torch.tensor([[    0, 31414,   232,   328,   740,  1140, 12695,    69, 46078,  1588,   2]])
         output = model(input_ids)[0]
         expected_shape = torch.Size((1, 3))
diff --git a/transformers/tests/modeling_tf_albert_test.py b/transformers/tests/modeling_tf_albert_test.py
index fbd519b8f6..7d3325b70b 100644
--- a/transformers/tests/modeling_tf_albert_test.py
+++ b/transformers/tests/modeling_tf_albert_test.py
@@ -18,11 +18,11 @@ from __future__ import print_function
 
 import unittest
 import shutil
-import pytest
 import sys
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_tf, slow
 
 from transformers import AlbertConfig, is_tf_available
 
@@ -31,10 +31,9 @@ if is_tf_available():
     from transformers.modeling_tf_albert import (TFAlbertModel, TFAlbertForMaskedLM,
                                                  TFAlbertForSequenceClassification,
                                                  TF_ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
-else:
-    pytestmark = pytest.mark.skip("Require TensorFlow")
 
 
+@require_tf
 class TFAlbertModelTest(TFCommonTestCases.TFCommonModelTester):
 
     all_model_classes = (
@@ -216,7 +215,7 @@ class TFAlbertModelTest(TFCommonTestCases.TFCommonModelTester):
         self.model_tester.create_and_check_albert_for_sequence_classification(
             *config_and_inputs)
 
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
         # for model_name in list(TF_ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/modeling_tf_auto_test.py b/transformers/tests/modeling_tf_auto_test.py
index fa90906e86..7ea48015d9 100644
--- a/transformers/tests/modeling_tf_auto_test.py
+++ b/transformers/tests/modeling_tf_auto_test.py
@@ -18,11 +18,12 @@ from __future__ import print_function
 
 import unittest
 import shutil
-import pytest
 import logging
 
 from transformers import is_tf_available
 
+from .utils import require_tf, slow
+
 if is_tf_available():
     from transformers import (AutoConfig, BertConfig,
                                       TFAutoModel, TFBertModel,
@@ -33,12 +34,11 @@ if is_tf_available():
 
     from .modeling_common_test import (CommonTestCases, ids_tensor)
     from .configuration_common_test import ConfigTester
-else:
-    pytestmark = pytest.mark.skip("Require TensorFlow")
 
 
+@require_tf
 class TFAutoModelTest(unittest.TestCase):
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         import h5py
         self.assertTrue(h5py.version.hdf5_version.startswith("1.10"))
@@ -54,7 +54,7 @@ class TFAutoModelTest(unittest.TestCase):
             self.assertIsNotNone(model)
             self.assertIsInstance(model, TFBertModel)
 
-    @pytest.mark.slow
+    @slow
     def test_lmhead_model_from_pretrained(self):
         logging.basicConfig(level=logging.INFO)
         # for model_name in list(TF_BERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
@@ -67,7 +67,7 @@ class TFAutoModelTest(unittest.TestCase):
             self.assertIsNotNone(model)
             self.assertIsInstance(model, TFBertForMaskedLM)
 
-    @pytest.mark.slow
+    @slow
     def test_sequence_classification_model_from_pretrained(self):
         logging.basicConfig(level=logging.INFO)
         # for model_name in list(TF_BERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
@@ -80,7 +80,7 @@ class TFAutoModelTest(unittest.TestCase):
             self.assertIsNotNone(model)
             self.assertIsInstance(model, TFBertForSequenceClassification)
 
-    @pytest.mark.slow
+    @slow
     def test_question_answering_model_from_pretrained(self):
         logging.basicConfig(level=logging.INFO)
         # for model_name in list(TF_BERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/modeling_tf_bert_test.py b/transformers/tests/modeling_tf_bert_test.py
index bcee97435e..d7a86fecb9 100644
--- a/transformers/tests/modeling_tf_bert_test.py
+++ b/transformers/tests/modeling_tf_bert_test.py
@@ -18,11 +18,11 @@ from __future__ import print_function
 
 import unittest
 import shutil
-import pytest
 import sys
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_tf, slow
 
 from transformers import BertConfig, is_tf_available
 
@@ -36,10 +36,9 @@ if is_tf_available():
                                                        TFBertForTokenClassification,
                                                        TFBertForQuestionAnswering,
                                                        TF_BERT_PRETRAINED_MODEL_ARCHIVE_MAP)
-else:
-    pytestmark = pytest.mark.skip("Require TensorFlow")
 
 
+@require_tf
 class TFBertModelTest(TFCommonTestCases.TFCommonModelTester):
 
     all_model_classes = (TFBertModel, TFBertForMaskedLM, TFBertForNextSentencePrediction,
@@ -309,7 +308,7 @@ class TFBertModelTest(TFCommonTestCases.TFCommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_bert_for_token_classification(*config_and_inputs)
 
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
         # for model_name in list(TF_BERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/modeling_tf_common_test.py b/transformers/tests/modeling_tf_common_test.py
index 7445ce826a..439360ba35 100644
--- a/transformers/tests/modeling_tf_common_test.py
+++ b/transformers/tests/modeling_tf_common_test.py
@@ -25,18 +25,17 @@ import unittest
 import uuid
 import tempfile
 
-import pytest
 import sys
 
 from transformers import is_tf_available, is_torch_available
 
+from .utils import require_tf, slow
+
 if is_tf_available():
     import tensorflow as tf
     import numpy as np
     from transformers import TFPreTrainedModel
     # from transformers.modeling_bert import BertModel, BertConfig, BERT_PRETRAINED_MODEL_ARCHIVE_MAP
-else:
-    pytestmark = pytest.mark.skip("Require TensorFlow")
 
 if sys.version_info[0] == 2:
     import cPickle as pickle
@@ -62,6 +61,7 @@ def _config_zero_init(config):
 
 class TFCommonTestCases:
 
+    @require_tf
     class TFCommonModelTester(unittest.TestCase):
 
         model_tester = None
@@ -164,7 +164,7 @@ class TFCommonTestCases:
             for model_class in self.all_model_classes:
                 # Prepare our model
                 model = model_class(config)
-                
+
                 # Let's load it from the disk to be sure we can use pretrained weights
                 with TemporaryDirectory() as tmpdirname:
                     outputs = model(inputs_dict)  # build the model
diff --git a/transformers/tests/modeling_tf_ctrl_test.py b/transformers/tests/modeling_tf_ctrl_test.py
index a57c882169..0b421c20c9 100644
--- a/transformers/tests/modeling_tf_ctrl_test.py
+++ b/transformers/tests/modeling_tf_ctrl_test.py
@@ -18,11 +18,11 @@ from __future__ import print_function
 
 import unittest
 import shutil
-import pytest
 import sys
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_tf, slow
 
 from transformers import CTRLConfig, is_tf_available
 
@@ -30,10 +30,9 @@ if is_tf_available():
     import tensorflow as tf
     from transformers.modeling_tf_ctrl import (TFCTRLModel, TFCTRLLMHeadModel,
                                                 TF_CTRL_PRETRAINED_MODEL_ARCHIVE_MAP)
-else:
-    pytestmark = pytest.mark.skip("Require TensorFlow")
 
 
+@require_tf
 class TFCTRLModelTest(TFCommonTestCases.TFCommonModelTester):
 
     all_model_classes = (TFCTRLModel, TFCTRLLMHeadModel) if is_tf_available() else ()
@@ -188,7 +187,7 @@ class TFCTRLModelTest(TFCommonTestCases.TFCommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_ctrl_lm_head(*config_and_inputs)
 
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
         for model_name in list(TF_CTRL_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/modeling_tf_distilbert_test.py b/transformers/tests/modeling_tf_distilbert_test.py
index e6d3795914..0ec45150ca 100644
--- a/transformers/tests/modeling_tf_distilbert_test.py
+++ b/transformers/tests/modeling_tf_distilbert_test.py
@@ -17,10 +17,10 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import pytest
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_tf, slow
 
 from transformers import DistilBertConfig, is_tf_available
 
@@ -30,10 +30,9 @@ if is_tf_available():
                                                              TFDistilBertForMaskedLM,
                                                              TFDistilBertForQuestionAnswering,
                                                              TFDistilBertForSequenceClassification)
-else:
-    pytestmark = pytest.mark.skip("Require TensorFlow")
 
 
+@require_tf
 class TFDistilBertModelTest(TFCommonTestCases.TFCommonModelTester):
 
     all_model_classes = (TFDistilBertModel, TFDistilBertForMaskedLM, TFDistilBertForQuestionAnswering,
@@ -210,7 +209,7 @@ class TFDistilBertModelTest(TFCommonTestCases.TFCommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_distilbert_for_sequence_classification(*config_and_inputs)
 
-    # @pytest.mark.slow
+    # @slow
     # def test_model_from_pretrained(self):
     #     cache_dir = "/tmp/transformers_test/"
     #     for model_name in list(DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/modeling_tf_gpt2_test.py b/transformers/tests/modeling_tf_gpt2_test.py
index 76e9ee2298..e070b72e65 100644
--- a/transformers/tests/modeling_tf_gpt2_test.py
+++ b/transformers/tests/modeling_tf_gpt2_test.py
@@ -18,11 +18,11 @@ from __future__ import print_function
 
 import unittest
 import shutil
-import pytest
 import sys
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_tf, slow
 
 from transformers import GPT2Config, is_tf_available
 
@@ -31,10 +31,9 @@ if is_tf_available():
     from transformers.modeling_tf_gpt2 import (TFGPT2Model, TFGPT2LMHeadModel,
                                                        TFGPT2DoubleHeadsModel,
                                                        TF_GPT2_PRETRAINED_MODEL_ARCHIVE_MAP)
-else:
-    pytestmark = pytest.mark.skip("Require TensorFlow")
 
 
+@require_tf
 class TFGPT2ModelTest(TFCommonTestCases.TFCommonModelTester):
 
     all_model_classes = (TFGPT2Model, TFGPT2LMHeadModel,
@@ -219,7 +218,7 @@ class TFGPT2ModelTest(TFCommonTestCases.TFCommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_gpt2_double_head(*config_and_inputs)
 
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
         for model_name in list(TF_GPT2_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/modeling_tf_openai_gpt_test.py b/transformers/tests/modeling_tf_openai_gpt_test.py
index d470c8862d..675e806c12 100644
--- a/transformers/tests/modeling_tf_openai_gpt_test.py
+++ b/transformers/tests/modeling_tf_openai_gpt_test.py
@@ -18,11 +18,11 @@ from __future__ import print_function
 
 import unittest
 import shutil
-import pytest
 import sys
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_tf, slow
 
 from transformers import OpenAIGPTConfig, is_tf_available
 
@@ -31,10 +31,9 @@ if is_tf_available():
     from transformers.modeling_tf_openai import (TFOpenAIGPTModel, TFOpenAIGPTLMHeadModel,
                                                          TFOpenAIGPTDoubleHeadsModel,
                                                          TF_OPENAI_GPT_PRETRAINED_MODEL_ARCHIVE_MAP)
-else:
-    pytestmark = pytest.mark.skip("Require TensorFlow")
 
 
+@require_tf
 class TFOpenAIGPTModelTest(TFCommonTestCases.TFCommonModelTester):
 
     all_model_classes = (TFOpenAIGPTModel, TFOpenAIGPTLMHeadModel,
@@ -218,7 +217,7 @@ class TFOpenAIGPTModelTest(TFCommonTestCases.TFCommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_openai_gpt_double_head(*config_and_inputs)
 
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
         for model_name in list(TF_OPENAI_GPT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/modeling_tf_roberta_test.py b/transformers/tests/modeling_tf_roberta_test.py
index edbfa4e205..42440bf1b7 100644
--- a/transformers/tests/modeling_tf_roberta_test.py
+++ b/transformers/tests/modeling_tf_roberta_test.py
@@ -18,10 +18,10 @@ from __future__ import print_function
 
 import unittest
 import shutil
-import pytest
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_tf, slow
 
 from transformers import RobertaConfig, is_tf_available
 
@@ -32,10 +32,9 @@ if is_tf_available():
                                                           TFRobertaForSequenceClassification,
                                                           TFRobertaForTokenClassification,
                                                           TF_ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP)
-else:
-    pytestmark = pytest.mark.skip("Require TensorFlow")
 
 
+@require_tf
 class TFRobertaModelTest(TFCommonTestCases.TFCommonModelTester):
 
     all_model_classes = (TFRobertaModel,TFRobertaForMaskedLM,
@@ -191,7 +190,7 @@ class TFRobertaModelTest(TFCommonTestCases.TFCommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_roberta_for_masked_lm(*config_and_inputs)
 
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
         for model_name in list(TF_ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
@@ -203,10 +202,10 @@ class TFRobertaModelTest(TFCommonTestCases.TFCommonModelTester):
 
 class TFRobertaModelIntegrationTest(unittest.TestCase):
 
-    @pytest.mark.slow
+    @slow
     def test_inference_masked_lm(self):
         model = TFRobertaForMaskedLM.from_pretrained('roberta-base')
-        
+
         input_ids = tf.constant([[    0, 31414,   232,   328,   740,  1140, 12695,    69, 46078,  1588,   2]])
         output = model(input_ids)[0]
         expected_shape = [1, 11, 50265]
@@ -224,10 +223,10 @@ class TFRobertaModelIntegrationTest(unittest.TestCase):
             numpy.allclose(output[:, :3, :3].numpy(), expected_slice.numpy(), atol=1e-3)
         )
 
-    @pytest.mark.slow
+    @slow
     def test_inference_no_head(self):
         model = TFRobertaModel.from_pretrained('roberta-base')
-        
+
         input_ids = tf.constant([[    0, 31414,   232,   328,   740,  1140, 12695,    69, 46078,  1588,   2]])
         output = model(input_ids)[0]
         # compare the actual values for a slice.
@@ -240,10 +239,10 @@ class TFRobertaModelIntegrationTest(unittest.TestCase):
             numpy.allclose(output[:, :3, :3].numpy(), expected_slice.numpy(), atol=1e-3)
         )
 
-    @pytest.mark.slow
+    @slow
     def test_inference_classification_head(self):
         model = TFRobertaForSequenceClassification.from_pretrained('roberta-large-mnli')
-        
+
         input_ids = tf.constant([[    0, 31414,   232,   328,   740,  1140, 12695,    69, 46078,  1588,   2]])
         output = model(input_ids)[0]
         expected_shape = [1, 3]
diff --git a/transformers/tests/modeling_tf_transfo_xl_test.py b/transformers/tests/modeling_tf_transfo_xl_test.py
index 534fe39646..03e332bdc1 100644
--- a/transformers/tests/modeling_tf_transfo_xl_test.py
+++ b/transformers/tests/modeling_tf_transfo_xl_test.py
@@ -19,10 +19,10 @@ from __future__ import print_function
 import unittest
 import random
 import shutil
-import pytest
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_tf, slow
 
 from transformers import TransfoXLConfig, is_tf_available
 
@@ -31,10 +31,9 @@ if is_tf_available():
     from transformers.modeling_tf_transfo_xl import (TFTransfoXLModel,
                                                              TFTransfoXLLMHeadModel,
                                                              TF_TRANSFO_XL_PRETRAINED_MODEL_ARCHIVE_MAP)
-else:
-    pytestmark = pytest.mark.skip("Require TensorFlow")
 
 
+@require_tf
 class TFTransfoXLModelTest(TFCommonTestCases.TFCommonModelTester):
 
     all_model_classes = (TFTransfoXLModel, TFTransfoXLLMHeadModel) if is_tf_available() else ()
@@ -204,7 +203,7 @@ class TFTransfoXLModelTest(TFCommonTestCases.TFCommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_transfo_xl_lm_head(*config_and_inputs)
 
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
         for model_name in list(TF_TRANSFO_XL_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/modeling_tf_xlm_test.py b/transformers/tests/modeling_tf_xlm_test.py
index 1bd661bebf..a680b70367 100644
--- a/transformers/tests/modeling_tf_xlm_test.py
+++ b/transformers/tests/modeling_tf_xlm_test.py
@@ -18,7 +18,6 @@ from __future__ import print_function
 
 import unittest
 import shutil
-import pytest
 
 from transformers import is_tf_available
 
@@ -29,13 +28,13 @@ if is_tf_available():
                                       TFXLMForSequenceClassification,
                                       TFXLMForQuestionAnsweringSimple,
                                       TF_XLM_PRETRAINED_MODEL_ARCHIVE_MAP)
-else:
-    pytestmark = pytest.mark.skip("Require TensorFlow")
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_tf, slow
 
 
+@require_tf
 class TFXLMModelTest(TFCommonTestCases.TFCommonModelTester):
 
     all_model_classes = (TFXLMModel, TFXLMWithLMHeadModel,
@@ -251,7 +250,7 @@ class TFXLMModelTest(TFCommonTestCases.TFCommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_xlm_sequence_classif(*config_and_inputs)
 
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
         for model_name in list(TF_XLM_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/modeling_tf_xlnet_test.py b/transformers/tests/modeling_tf_xlnet_test.py
index a00a965570..94864b86f2 100644
--- a/transformers/tests/modeling_tf_xlnet_test.py
+++ b/transformers/tests/modeling_tf_xlnet_test.py
@@ -21,7 +21,6 @@ import unittest
 import json
 import random
 import shutil
-import pytest
 
 from transformers import XLNetConfig, is_tf_available
 
@@ -33,12 +32,13 @@ if is_tf_available():
                                                         TFXLNetForTokenClassification,
                                                         TFXLNetForQuestionAnsweringSimple,
                                                         TF_XLNET_PRETRAINED_MODEL_ARCHIVE_MAP)
-else:
-    pytestmark = pytest.mark.skip("Require TensorFlow")
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_tf, slow
 
+
+@require_tf
 class TFXLNetModelTest(TFCommonTestCases.TFCommonModelTester):
 
     all_model_classes=(TFXLNetModel, TFXLNetLMHeadModel,
@@ -304,7 +304,7 @@ class TFXLNetModelTest(TFCommonTestCases.TFCommonModelTester):
     def test_xlnet_lm_head(self):
         self.model_tester.set_seed()
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
-        self.model_tester.create_and_check_xlnet_lm_head(*config_and_inputs) 
+        self.model_tester.create_and_check_xlnet_lm_head(*config_and_inputs)
 
     def test_xlnet_sequence_classif(self):
         self.model_tester.set_seed()
@@ -320,7 +320,7 @@ class TFXLNetModelTest(TFCommonTestCases.TFCommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_xlnet_qa(*config_and_inputs)
 
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
         for model_name in list(TF_XLNET_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/modeling_transfo_xl_test.py b/transformers/tests/modeling_transfo_xl_test.py
index f7b913da5b..647dd3724d 100644
--- a/transformers/tests/modeling_transfo_xl_test.py
+++ b/transformers/tests/modeling_transfo_xl_test.py
@@ -19,7 +19,6 @@ from __future__ import print_function
 import unittest
 import random
 import shutil
-import pytest
 
 from transformers import is_torch_available
 
@@ -27,12 +26,13 @@ if is_torch_available():
     import torch
     from transformers import (TransfoXLConfig, TransfoXLModel, TransfoXLLMHeadModel)
     from transformers.modeling_transfo_xl import TRANSFO_XL_PRETRAINED_MODEL_ARCHIVE_MAP
-else:
-    pytestmark = pytest.mark.skip("Require Torch")
 
 from .modeling_common_test import (CommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_torch, slow, torch_device
 
+
+@require_torch
 class TransfoXLModelTest(CommonTestCases.CommonModelTester):
 
     all_model_classes = (TransfoXLModel, TransfoXLLMHeadModel) if is_torch_available() else ()
@@ -111,6 +111,7 @@ class TransfoXLModelTest(CommonTestCases.CommonModelTester):
 
         def create_transfo_xl_model(self, config, input_ids_1, input_ids_2, lm_labels):
             model = TransfoXLModel(config)
+            model.to(torch_device)
             model.eval()
 
             hidden_states_1, mems_1 = model(input_ids_1)
@@ -140,6 +141,7 @@ class TransfoXLModelTest(CommonTestCases.CommonModelTester):
 
         def create_transfo_xl_lm_head(self, config, input_ids_1, input_ids_2, lm_labels):
             model = TransfoXLLMHeadModel(config)
+            model.to(torch_device)
             model.eval()
 
             lm_logits_1, mems_1 = model(input_ids_1)
@@ -204,7 +206,7 @@ class TransfoXLModelTest(CommonTestCases.CommonModelTester):
         output_result = self.model_tester.create_transfo_xl_lm_head(*config_and_inputs)
         self.model_tester.check_transfo_xl_lm_head_output(output_result)
 
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
         for model_name in list(TRANSFO_XL_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/modeling_xlm_test.py b/transformers/tests/modeling_xlm_test.py
index 0133febb58..f6b980767c 100644
--- a/transformers/tests/modeling_xlm_test.py
+++ b/transformers/tests/modeling_xlm_test.py
@@ -18,7 +18,6 @@ from __future__ import print_function
 
 import unittest
 import shutil
-import pytest
 
 from transformers import is_torch_available
 
@@ -26,13 +25,13 @@ if is_torch_available():
     from transformers import (XLMConfig, XLMModel, XLMWithLMHeadModel, XLMForQuestionAnswering,
                                       XLMForSequenceClassification, XLMForQuestionAnsweringSimple)
     from transformers.modeling_xlm import XLM_PRETRAINED_MODEL_ARCHIVE_MAP
-else:
-    pytestmark = pytest.mark.skip("Require Torch")
 
 from .modeling_common_test import (CommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_torch, slow, torch_device
 
 
+@require_torch
 class XLMModelTest(CommonTestCases.CommonModelTester):
 
     all_model_classes = (XLMModel, XLMWithLMHeadModel, XLMForQuestionAnswering,
@@ -148,6 +147,7 @@ class XLMModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_xlm_model(self, config, input_ids, token_type_ids, input_lengths, sequence_labels, token_labels, is_impossible_labels, input_mask):
             model = XLMModel(config=config)
+            model.to(torch_device)
             model.eval()
             outputs = model(input_ids, lengths=input_lengths, langs=token_type_ids)
             outputs = model(input_ids, langs=token_type_ids)
@@ -163,6 +163,7 @@ class XLMModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_xlm_lm_head(self, config, input_ids, token_type_ids, input_lengths, sequence_labels, token_labels, is_impossible_labels, input_mask):
             model = XLMWithLMHeadModel(config)
+            model.to(torch_device)
             model.eval()
 
             loss, logits = model(input_ids, token_type_ids=token_type_ids, labels=token_labels)
@@ -182,6 +183,7 @@ class XLMModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_xlm_simple_qa(self, config, input_ids, token_type_ids, input_lengths, sequence_labels, token_labels, is_impossible_labels, input_mask):
             model = XLMForQuestionAnsweringSimple(config)
+            model.to(torch_device)
             model.eval()
 
             outputs = model(input_ids)
@@ -206,6 +208,7 @@ class XLMModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_xlm_qa(self, config, input_ids, token_type_ids, input_lengths, sequence_labels, token_labels, is_impossible_labels, input_mask):
             model = XLMForQuestionAnswering(config)
+            model.to(torch_device)
             model.eval()
 
             outputs = model(input_ids)
@@ -260,6 +263,7 @@ class XLMModelTest(CommonTestCases.CommonModelTester):
 
         def create_and_check_xlm_sequence_classif(self, config, input_ids, token_type_ids, input_lengths, sequence_labels, token_labels, is_impossible_labels, input_mask):
             model = XLMForSequenceClassification(config)
+            model.to(torch_device)
             model.eval()
 
             (logits,) = model(input_ids)
@@ -312,7 +316,7 @@ class XLMModelTest(CommonTestCases.CommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_xlm_sequence_classif(*config_and_inputs)
 
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
         for model_name in list(XLM_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/modeling_xlnet_test.py b/transformers/tests/modeling_xlnet_test.py
index 38888d4488..56b6bb3f4d 100644
--- a/transformers/tests/modeling_xlnet_test.py
+++ b/transformers/tests/modeling_xlnet_test.py
@@ -21,7 +21,6 @@ import unittest
 import json
 import random
 import shutil
-import pytest
 
 from transformers import is_torch_available
 
@@ -31,12 +30,13 @@ if is_torch_available():
     from transformers import (XLNetConfig, XLNetModel, XLNetLMHeadModel, XLNetForSequenceClassification,
                               XLNetForTokenClassification, XLNetForQuestionAnswering)
     from transformers.modeling_xlnet import XLNET_PRETRAINED_MODEL_ARCHIVE_MAP
-else:
-    pytestmark = pytest.mark.skip("Require Torch")
 
 from .modeling_common_test import (CommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_torch, slow, torch_device
 
+
+@require_torch
 class XLNetModelTest(CommonTestCases.CommonModelTester):
 
     all_model_classes=(XLNetModel, XLNetLMHeadModel, XLNetForTokenClassification,
@@ -100,9 +100,9 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
             input_mask = ids_tensor([self.batch_size, self.seq_length], 2).float()
 
             input_ids_q = ids_tensor([self.batch_size, self.seq_length + 1], self.vocab_size)
-            perm_mask = torch.zeros(self.batch_size, self.seq_length + 1, self.seq_length + 1, dtype=torch.float)
+            perm_mask = torch.zeros(self.batch_size, self.seq_length + 1, self.seq_length + 1, dtype=torch.float, device=torch_device)
             perm_mask[:, :, -1] = 1.0  # Previous tokens don't see last token
-            target_mapping = torch.zeros(self.batch_size, 1, self.seq_length + 1, dtype=torch.float)
+            target_mapping = torch.zeros(self.batch_size, 1, self.seq_length + 1, dtype=torch.float, device=torch_device)
             target_mapping[:, 0, -1] = 1.0  # predict last token
 
             sequence_labels = None
@@ -141,6 +141,7 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
         def create_and_check_xlnet_base_model(self, config, input_ids_1, input_ids_2, input_ids_q, perm_mask, input_mask,
                 target_mapping, segment_ids, lm_labels, sequence_labels, is_impossible_labels, token_labels):
             model = XLNetModel(config)
+            model.to(torch_device)
             model.eval()
 
             _, _ = model(input_ids_1, input_mask=input_mask)
@@ -155,6 +156,7 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
 
             config.mem_len = 0
             model = XLNetModel(config)
+            model.to(torch_device)
             model.eval()
             no_mems_outputs = model(input_ids_1)
             self.parent.assertEqual(len(no_mems_outputs), 1)
@@ -169,6 +171,7 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
         def create_and_check_xlnet_base_model_with_att_output(self, config, input_ids_1, input_ids_2, input_ids_q, perm_mask, input_mask,
                     target_mapping, segment_ids, lm_labels, sequence_labels, is_impossible_labels, token_labels):
             model = XLNetModel(config)
+            model.to(torch_device)
             model.eval()
 
             _, _, attentions = model(input_ids_1, target_mapping=target_mapping)
@@ -181,6 +184,7 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
         def create_and_check_xlnet_lm_head(self, config, input_ids_1, input_ids_2, input_ids_q, perm_mask, input_mask,
                 target_mapping, segment_ids, lm_labels, sequence_labels, is_impossible_labels, token_labels):
             model = XLNetLMHeadModel(config)
+            model.to(torch_device)
             model.eval()
 
             loss_1, all_logits_1, mems_1 = model(input_ids_1, token_type_ids=segment_ids, labels=lm_labels)
@@ -221,6 +225,7 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
         def create_and_check_xlnet_qa(self, config, input_ids_1, input_ids_2, input_ids_q, perm_mask, input_mask,
                 target_mapping, segment_ids, lm_labels, sequence_labels, is_impossible_labels, token_labels):
             model = XLNetForQuestionAnswering(config)
+            model.to(torch_device)
             model.eval()
 
             outputs = model(input_ids_1)
@@ -279,6 +284,7 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
         def create_and_check_xlnet_token_classif(self, config, input_ids_1, input_ids_2, input_ids_q, perm_mask, input_mask,
                 target_mapping, segment_ids, lm_labels, sequence_labels, is_impossible_labels, token_labels):
             model = XLNetForTokenClassification(config)
+            model.to(torch_device)
             model.eval()
 
             logits, mems_1 = model(input_ids_1)
@@ -311,6 +317,7 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
         def create_and_check_xlnet_sequence_classif(self, config, input_ids_1, input_ids_2, input_ids_q, perm_mask, input_mask,
                 target_mapping, segment_ids, lm_labels, sequence_labels, is_impossible_labels, token_labels):
             model = XLNetForSequenceClassification(config)
+            model.to(torch_device)
             model.eval()
 
             logits, mems_1 = model(input_ids_1)
@@ -362,7 +369,7 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
     def test_xlnet_lm_head(self):
         self.model_tester.set_seed()
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
-        self.model_tester.create_and_check_xlnet_lm_head(*config_and_inputs) 
+        self.model_tester.create_and_check_xlnet_lm_head(*config_and_inputs)
 
     def test_xlnet_sequence_classif(self):
         self.model_tester.set_seed()
@@ -379,7 +386,7 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_xlnet_qa(*config_and_inputs)
 
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
         for model_name in list(XLNET_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/optimization_test.py b/transformers/tests/optimization_test.py
index ab9afbfcf7..cc10ad5908 100644
--- a/transformers/tests/optimization_test.py
+++ b/transformers/tests/optimization_test.py
@@ -18,7 +18,6 @@ from __future__ import print_function
 
 import unittest
 import os
-import pytest
 
 from transformers import is_torch_available
 
@@ -31,10 +30,9 @@ if is_torch_available():
                               get_cosine_schedule_with_warmup,
                               get_cosine_with_hard_restarts_schedule_with_warmup,
                               get_linear_schedule_with_warmup)
-else:
-    pytestmark = pytest.mark.skip("Require Torch")
 
 from .tokenization_tests_commons import TemporaryDirectory
+from .utils import require_torch
 
 
 def unwrap_schedule(scheduler, num_steps=10):
@@ -58,6 +56,7 @@ def unwrap_and_save_reload_schedule(scheduler, num_steps=10):
                 scheduler.load_state_dict(state_dict)
     return lrs
 
+@require_torch
 class OptimizationTest(unittest.TestCase):
 
     def assertListAlmostEqual(self, list1, list2, tol):
@@ -80,6 +79,7 @@ class OptimizationTest(unittest.TestCase):
         self.assertListAlmostEqual(w.tolist(), [0.4, 0.2, -0.5], tol=1e-2)
 
 
+@require_torch
 class ScheduleInitTest(unittest.TestCase):
     m = torch.nn.Linear(50, 50) if is_torch_available() else None
     optimizer = AdamW(m.parameters(), lr=10.) if is_torch_available() else None
diff --git a/transformers/tests/tokenization_auto_test.py b/transformers/tests/tokenization_auto_test.py
index 79370811e8..18346d2768 100644
--- a/transformers/tests/tokenization_auto_test.py
+++ b/transformers/tests/tokenization_auto_test.py
@@ -18,15 +18,16 @@ from __future__ import print_function
 
 import unittest
 import shutil
-import pytest
 import logging
 
 from transformers import AutoTokenizer, BertTokenizer, AutoTokenizer, GPT2Tokenizer
 from transformers import BERT_PRETRAINED_CONFIG_ARCHIVE_MAP, GPT2_PRETRAINED_CONFIG_ARCHIVE_MAP
 
+from .utils import slow
+
 
 class AutoTokenizerTest(unittest.TestCase):
-    @pytest.mark.slow
+    @slow
     def test_tokenizer_from_pretrained(self):
         logging.basicConfig(level=logging.INFO)
         for model_name in list(BERT_PRETRAINED_CONFIG_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/tokenization_bert_test.py b/transformers/tests/tokenization_bert_test.py
index 73ea38e20a..f390248956 100644
--- a/transformers/tests/tokenization_bert_test.py
+++ b/transformers/tests/tokenization_bert_test.py
@@ -16,7 +16,6 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 
 import os
 import unittest
-import pytest
 from io import open
 
 from transformers.tokenization_bert import (BasicTokenizer,
@@ -26,6 +25,7 @@ from transformers.tokenization_bert import (BasicTokenizer,
                                                     _is_whitespace, VOCAB_FILES_NAMES)
 
 from .tokenization_tests_commons import CommonTestCases
+from .utils import slow
 
 class BertTokenizationTest(CommonTestCases.CommonTokenizerTester):
 
@@ -126,7 +126,7 @@ class BertTokenizationTest(CommonTestCases.CommonTokenizerTester):
         self.assertFalse(_is_punctuation(u"A"))
         self.assertFalse(_is_punctuation(u" "))
 
-    @pytest.mark.slow
+    @slow
     def test_sequence_builders(self):
         tokenizer = self.tokenizer_class.from_pretrained("bert-base-uncased")
 
diff --git a/transformers/tests/tokenization_distilbert_test.py b/transformers/tests/tokenization_distilbert_test.py
index 77a487651d..e815eca672 100644
--- a/transformers/tests/tokenization_distilbert_test.py
+++ b/transformers/tests/tokenization_distilbert_test.py
@@ -16,13 +16,13 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 
 import os
 import unittest
-import pytest
 from io import open
 
 from transformers.tokenization_distilbert import (DistilBertTokenizer)
 
 from .tokenization_tests_commons import CommonTestCases
 from .tokenization_bert_test import BertTokenizationTest
+from .utils import slow
 
 class DistilBertTokenizationTest(BertTokenizationTest):
 
@@ -31,7 +31,7 @@ class DistilBertTokenizationTest(BertTokenizationTest):
     def get_tokenizer(self, **kwargs):
         return DistilBertTokenizer.from_pretrained(self.tmpdirname, **kwargs)
 
-    @pytest.mark.slow
+    @slow
     def test_sequence_builders(self):
         tokenizer = DistilBertTokenizer.from_pretrained("distilbert-base-uncased")
 
diff --git a/transformers/tests/tokenization_roberta_test.py b/transformers/tests/tokenization_roberta_test.py
index a27bf7d654..8ad0b59511 100644
--- a/transformers/tests/tokenization_roberta_test.py
+++ b/transformers/tests/tokenization_roberta_test.py
@@ -17,11 +17,11 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 import os
 import json
 import unittest
-import pytest
 from io import open
 
 from transformers.tokenization_roberta import RobertaTokenizer, VOCAB_FILES_NAMES
 from .tokenization_tests_commons import CommonTestCases
+from .utils import slow
 
 
 class RobertaTokenizationTest(CommonTestCases.CommonTokenizerTester):
@@ -79,7 +79,7 @@ class RobertaTokenizationTest(CommonTestCases.CommonTokenizerTester):
             [0, 31414, 232, 328, 740, 1140, 12695, 69, 46078, 1588, 2]
         )
 
-    @pytest.mark.slow
+    @slow
     def test_sequence_builders(self):
         tokenizer = RobertaTokenizer.from_pretrained("roberta-base")
 
diff --git a/transformers/tests/tokenization_tests_commons.py b/transformers/tests/tokenization_tests_commons.py
index 97cd555df3..faff003f4b 100644
--- a/transformers/tests/tokenization_tests_commons.py
+++ b/transformers/tests/tokenization_tests_commons.py
@@ -102,9 +102,11 @@ class CommonTestCases:
             with TemporaryDirectory() as tmpdirname:
 
                 filename = os.path.join(tmpdirname, u"tokenizer.bin")
-                pickle.dump(tokenizer, open(filename, "wb"))
+                with open(filename, "wb") as handle:
+                    pickle.dump(tokenizer, handle)
 
-                tokenizer_new = pickle.load(open(filename, "rb"))
+                with open(filename, "rb") as handle:
+                    tokenizer_new = pickle.load(handle)
 
             subwords_loaded = tokenizer_new.tokenize(text)
 
diff --git a/transformers/tests/tokenization_transfo_xl_test.py b/transformers/tests/tokenization_transfo_xl_test.py
index 4e99484b0c..5495ebd3a6 100644
--- a/transformers/tests/tokenization_transfo_xl_test.py
+++ b/transformers/tests/tokenization_transfo_xl_test.py
@@ -16,7 +16,6 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 
 import os
 import unittest
-import pytest
 from io import open
 
 from transformers import is_torch_available
@@ -24,11 +23,12 @@ from transformers import is_torch_available
 if is_torch_available():
     import torch
     from transformers.tokenization_transfo_xl import TransfoXLTokenizer, VOCAB_FILES_NAMES
-else:
-    pytestmark = pytest.mark.skip("Require Torch")  # TODO: untangle Transfo-XL tokenizer from torch.load and torch.save
 
 from .tokenization_tests_commons import CommonTestCases
+from .utils import require_torch
 
+
+@require_torch
 class TransfoXLTokenizationTest(CommonTestCases.CommonTokenizerTester):
 
     tokenizer_class = TransfoXLTokenizer if is_torch_available() else None
diff --git a/transformers/tests/tokenization_utils_test.py b/transformers/tests/tokenization_utils_test.py
index 8630191c69..ff3f80ff7d 100644
--- a/transformers/tests/tokenization_utils_test.py
+++ b/transformers/tests/tokenization_utils_test.py
@@ -18,13 +18,14 @@ from __future__ import print_function
 
 import unittest
 import six
-import pytest
 
 from transformers import PreTrainedTokenizer
 from transformers.tokenization_gpt2 import GPT2Tokenizer
 
+from .utils import slow
+
 class TokenizerUtilsTest(unittest.TestCase):
-    @pytest.mark.slow
+
     def check_tokenizer_from_pretrained(self, tokenizer_class):
         s3_models = list(tokenizer_class.max_model_input_sizes.keys())
         for model_name in s3_models[:1]:
@@ -41,6 +42,7 @@ class TokenizerUtilsTest(unittest.TestCase):
                 special_tok_id = tokenizer.convert_tokens_to_ids(special_tok)
                 self.assertIsInstance(special_tok_id, int)
 
+    @slow
     def test_pretrained_tokenizers(self):
         self.check_tokenizer_from_pretrained(GPT2Tokenizer)
 
diff --git a/transformers/tests/tokenization_xlm_test.py b/transformers/tests/tokenization_xlm_test.py
index 3ff6564e34..7582a46662 100644
--- a/transformers/tests/tokenization_xlm_test.py
+++ b/transformers/tests/tokenization_xlm_test.py
@@ -17,11 +17,11 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 import os
 import unittest
 import json
-import pytest
 
 from transformers.tokenization_xlm import XLMTokenizer, VOCAB_FILES_NAMES
 
 from .tokenization_tests_commons import CommonTestCases
+from .utils import slow
 
 class XLMTokenizationTest(CommonTestCases.CommonTokenizerTester):
 
@@ -67,7 +67,7 @@ class XLMTokenizationTest(CommonTestCases.CommonTokenizerTester):
         self.assertListEqual(
             tokenizer.convert_tokens_to_ids(input_tokens), input_bpe_tokens)
 
-    @pytest.mark.slow
+    @slow
     def test_sequence_builders(self):
         tokenizer = XLMTokenizer.from_pretrained("xlm-mlm-en-2048")
 
diff --git a/transformers/tests/tokenization_xlnet_test.py b/transformers/tests/tokenization_xlnet_test.py
index 2e14ffeb82..b68495a796 100644
--- a/transformers/tests/tokenization_xlnet_test.py
+++ b/transformers/tests/tokenization_xlnet_test.py
@@ -16,11 +16,11 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 
 import os
 import unittest
-import pytest
 
 from transformers.tokenization_xlnet import (XLNetTokenizer, SPIECE_UNDERLINE)
 
 from .tokenization_tests_commons import CommonTestCases
+from .utils import slow
 
 SAMPLE_VOCAB = os.path.join(os.path.dirname(os.path.abspath(__file__)),
                     'fixtures/test_sentencepiece.model')
@@ -90,7 +90,7 @@ class XLNetTokenizationTest(CommonTestCases.CommonTokenizerTester):
                                       u'9', u'2', u'0', u'0', u'0', u',', SPIECE_UNDERLINE + u'and', SPIECE_UNDERLINE + u'this',
                                       SPIECE_UNDERLINE + u'is', SPIECE_UNDERLINE + u'f', u'al', u'se', u'.'])
 
-    @pytest.mark.slow
+    @slow
     def test_sequence_builders(self):
         tokenizer = XLNetTokenizer.from_pretrained("xlnet-base-cased")
 
diff --git a/transformers/tests/utils.py b/transformers/tests/utils.py
new file mode 100644
index 0000000000..7a51ab612b
--- /dev/null
+++ b/transformers/tests/utils.py
@@ -0,0 +1,64 @@
+import os
+import unittest
+
+from distutils.util import strtobool
+
+from transformers.file_utils import _tf_available, _torch_available
+
+
+try:
+    run_slow = os.environ["RUN_SLOW"]
+except KeyError:
+    # RUN_SLOW isn't set, default to skipping slow tests.
+    _run_slow_tests = False
+else:
+    # RUN_SLOW is set, convert it to True or False.
+    try:
+        _run_slow_tests = strtobool(run_slow)
+    except ValueError:
+        # More values are supported, but let's keep the message simple.
+        raise ValueError("If set, RUN_SLOW must be yes or no.")
+
+
+def slow(test_case):
+    """
+    Decorator marking a test as slow.
+
+    Slow tests are skipped by default. Set the RUN_SLOW environment variable
+    to a truthy value to run them.
+
+    """
+    if not _run_slow_tests:
+        test_case = unittest.skip("test is slow")(test_case)
+    return test_case
+
+
+def require_torch(test_case):
+    """
+    Decorator marking a test that requires PyTorch.
+
+    These tests are skipped when PyTorch isn't installed.
+
+    """
+    if not _torch_available:
+        test_case = unittest.skip("test requires PyTorch")(test_case)
+    return test_case
+
+
+def require_tf(test_case):
+    """
+    Decorator marking a test that requires TensorFlow.
+
+    These tests are skipped when TensorFlow isn't installed.
+
+    """
+    if not _tf_available:
+        test_case = unittest.skip("test requires TensorFlow")(test_case)
+    return test_case
+
+
+if _torch_available:
+    # Set the USE_CUDA environment variable to select a GPU.
+    torch_device = "cuda" if os.environ.get("USE_CUDA") else "cpu"
+else:
+    torch_device = None
diff --git a/transformers/tokenization_albert.py b/transformers/tokenization_albert.py
index 40a4b29206..6b92d07218 100644
--- a/transformers/tokenization_albert.py
+++ b/transformers/tokenization_albert.py
@@ -141,7 +141,7 @@ class AlbertTokenizer(PreTrainedTokenizer):
             pieces = self.sp_model.SampleEncodeAsPieces(text, 64, 0.1)
         new_pieces = []
         for piece in pieces:
-            if len(piece) > 1 and piece[-1] == ',' and piece[-2].isdigit():
+            if len(piece) > 1 and piece[-1] == str(',') and piece[-2].isdigit():
                 cur_pieces = self.sp_model.EncodeAsPieces(
                     piece[:-1].replace(SPIECE_UNDERLINE, ''))
                 if piece[0] != SPIECE_UNDERLINE and cur_pieces[0][0] == SPIECE_UNDERLINE:
@@ -225,9 +225,9 @@ class AlbertTokenizer(PreTrainedTokenizer):
         """
         Creates a mask from the two sequences passed to be used in a sequence-pair classification task.
         An ALBERT sequence pair mask has the following format:
-        0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 
-        | first sequence    | second sequence     
-        
+        0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1
+        | first sequence    | second sequence
+
         if token_ids_1 is None, only returns the first portion of the mask (0's).
         """
         sep = [self.sep_token_id]
diff --git a/transformers/tokenization_ctrl.py b/transformers/tokenization_ctrl.py
index 9454cbbaf3..219f17c404 100644
--- a/transformers/tokenization_ctrl.py
+++ b/transformers/tokenization_ctrl.py
@@ -133,9 +133,11 @@ class CTRLTokenizer(PreTrainedTokenizer):
         self.max_len_single_sentence = self.max_len # no default special tokens - you can update this value if you add special tokens
         self.max_len_sentences_pair = self.max_len # no default special tokens - you can update this value if you add special tokens
 
-        self.encoder = json.load(open(vocab_file, encoding="utf-8"))
+        with open(vocab_file, encoding="utf-8") as vocab_handle:
+            self.encoder = json.load(vocab_handle)
         self.decoder = {v:k for k,v in self.encoder.items()}
-        merges = open(merges_file, encoding='utf-8').read().split('\n')[1:-1]
+        with open(merges_file, encoding='utf-8') as merges_handle:
+            merges = merges_handle.read().split('\n')[1:-1]
         merges = [tuple(merge.split()) for merge in merges]
         self.bpe_ranks = dict(zip(merges, range(len(merges))))
         self.cache = {}
diff --git a/transformers/tokenization_gpt2.py b/transformers/tokenization_gpt2.py
index 5fda709448..68c6101860 100644
--- a/transformers/tokenization_gpt2.py
+++ b/transformers/tokenization_gpt2.py
@@ -72,7 +72,7 @@ def bytes_to_unicode():
     """
     Returns list of utf-8 byte and a mapping to unicode strings.
     We specifically avoids mapping to whitespace/control characters the bpe code barfs on.
-    
+
     The reversible bpe codes work on unicode strings.
     This means you need a large # of unicode characters in your vocab if you want to avoid UNKs.
     When you're at something like a 10B token dataset you end up needing around 5K for decent coverage.
@@ -122,13 +122,15 @@ class GPT2Tokenizer(PreTrainedTokenizer):
         self.max_len_single_sentence = self.max_len # no default special tokens - you can update this value if you add special tokens
         self.max_len_sentences_pair = self.max_len # no default special tokens - you can update this value if you add special tokens
 
-        self.encoder = json.load(open(vocab_file, encoding="utf-8"))
+        with open(vocab_file, encoding="utf-8") as vocab_handle:
+            self.encoder = json.load(vocab_handle)
         self.decoder = {v: k for k, v in self.encoder.items()}
         self.errors = errors  # how to handle errors in decoding
         self.byte_encoder = bytes_to_unicode()
         self.byte_decoder = {v: k for k, v in self.byte_encoder.items()}
-        bpe_data = open(merges_file, encoding='utf-8').read().split('\n')[1:-1]
-        bpe_merges = [tuple(merge.split()) for merge in bpe_data]
+        with open(merges_file, encoding='utf-8') as merges_handle:
+            bpe_merges = merges_handle.read().split('\n')[1:-1]
+        bpe_merges = [tuple(merge.split()) for merge in bpe_merges]
         self.bpe_ranks = dict(zip(bpe_merges, range(len(bpe_merges))))
         self.cache = {}
 
@@ -234,4 +236,4 @@ class GPT2Tokenizer(PreTrainedTokenizer):
                 writer.write(' '.join(bpe_tokens) + u'\n')
                 index += 1
 
-        return vocab_file, merge_file
\ No newline at end of file
+        return vocab_file, merge_file
diff --git a/transformers/tokenization_openai.py b/transformers/tokenization_openai.py
index 0efbdb37c0..a4c64b7020 100644
--- a/transformers/tokenization_openai.py
+++ b/transformers/tokenization_openai.py
@@ -101,9 +101,11 @@ class OpenAIGPTTokenizer(PreTrainedTokenizer):
             self.nlp = BasicTokenizer(do_lower_case=True)
             self.fix_text = None
 
-        self.encoder = json.load(open(vocab_file, encoding="utf-8"))
+        with open(vocab_file, encoding="utf-8") as vocab_handle:
+            self.encoder = json.load(vocab_handle)
         self.decoder = {v:k for k,v in self.encoder.items()}
-        merges = open(merges_file, encoding='utf-8').read().split('\n')[1:-1]
+        with open(merges_file, encoding='utf-8') as merges_handle:
+            merges = merges_handle.read().split('\n')[1:-1]
         merges = [tuple(merge.split()) for merge in merges]
         self.bpe_ranks = dict(zip(merges, range(len(merges))))
         self.cache = {}
diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index 5d683629f0..4c6cbd8986 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -347,7 +347,7 @@ class PreTrainedTokenizer(object):
                     "We assumed '{}' was a path or url to a directory containing vocabulary files "
                     "named {} but couldn't find such vocabulary files at this path or url.".format(
                         pretrained_model_name_or_path, ', '.join(s3_models),
-                        pretrained_model_name_or_path, 
+                        pretrained_model_name_or_path,
                         list(cls.vocab_files_names.values())))
 
         # Get files from url, cache, or disk depending on the case
@@ -382,7 +382,8 @@ class PreTrainedTokenizer(object):
         # Did we saved some inputs and kwargs to reload ?
         tokenizer_config_file = resolved_vocab_files.pop('tokenizer_config_file', None)
         if tokenizer_config_file is not None:
-            init_kwargs = json.load(open(tokenizer_config_file, encoding="utf-8"))
+            with open(tokenizer_config_file, encoding="utf-8") as tokenizer_config_handle:
+                init_kwargs = json.load(tokenizer_config_handle)
             saved_init_inputs = init_kwargs.pop('init_inputs', ())
             if not init_inputs:
                 init_inputs = saved_init_inputs
@@ -407,7 +408,8 @@ class PreTrainedTokenizer(object):
             if args_name not in init_kwargs:
                 init_kwargs[args_name] = file_path
         if special_tokens_map_file is not None:
-            special_tokens_map = json.load(open(special_tokens_map_file, encoding="utf-8"))
+            with open(special_tokens_map_file, encoding="utf-8") as special_tokens_map_handle:
+                special_tokens_map = json.load(special_tokens_map_handle)
             for key, value in special_tokens_map.items():
                 if key not in init_kwargs:
                     init_kwargs[key] = value
@@ -421,7 +423,8 @@ class PreTrainedTokenizer(object):
 
         # Add supplementary tokens.
         if added_tokens_file is not None:
-            added_tok_encoder = json.load(open(added_tokens_file, encoding="utf-8"))
+            with open(added_tokens_file, encoding="utf-8") as added_tokens_handle:
+                added_tok_encoder = json.load(added_tokens_handle)
             added_tok_decoder = {v:k for k, v in added_tok_encoder.items()}
             tokenizer.added_tokens_encoder.update(added_tok_encoder)
             tokenizer.added_tokens_decoder.update(added_tok_decoder)
@@ -937,7 +940,7 @@ class PreTrainedTokenizer(object):
             logger.warning("Token indices sequence length is longer than the specified maximum sequence length "
                            "for this model ({} > {}). Running this sequence through the model will result in "
                            "indexing errors".format(len(ids), self.max_len))
-                           
+
         return encoded_inputs
 
     def truncate_sequences(self, ids, pair_ids=None, num_tokens_to_remove=0, truncation_strategy='longest_first', stride=0):
diff --git a/transformers/tokenization_xlm.py b/transformers/tokenization_xlm.py
index ba994dc356..6c9f8e5e5c 100644
--- a/transformers/tokenization_xlm.py
+++ b/transformers/tokenization_xlm.py
@@ -524,7 +524,7 @@ class XLMTokenizer(PreTrainedTokenizer):
 
         - argument ``special_tokens`` and function ``set_special_tokens``, can be used to add additional symbols \
         (ex: "__classify__") to a vocabulary
-        
+
         - `lang2id` attribute maps the languages supported by the model with their ids if provided (automatically set for pretrained vocabularies)
 
         - `id2lang` attributes does reverse mapping if provided (automatically set for pretrained vocabularies)
@@ -564,9 +564,11 @@ class XLMTokenizer(PreTrainedTokenizer):
         self.ja_word_tokenizer = None
         self.zh_word_tokenizer = None
 
-        self.encoder = json.load(open(vocab_file, encoding="utf-8"))
+        with open(vocab_file, encoding="utf-8") as vocab_handle:
+            self.encoder = json.load(vocab_handle)
         self.decoder = {v:k for k,v in self.encoder.items()}
-        merges = open(merges_file, encoding='utf-8').read().split('\n')[:-1]
+        with open(merges_file, encoding='utf-8') as merges_handle:
+            merges = merges_handle.read().split('\n')[:-1]
         merges = [tuple(merge.split()[:2]) for merge in merges]
         self.bpe_ranks = dict(zip(merges, range(len(merges))))
         self.cache = {}
diff --git a/transformers/tokenization_xlnet.py b/transformers/tokenization_xlnet.py
index c01fbbbeeb..8c86a5bd60 100644
--- a/transformers/tokenization_xlnet.py
+++ b/transformers/tokenization_xlnet.py
@@ -141,7 +141,7 @@ class XLNetTokenizer(PreTrainedTokenizer):
             pieces = self.sp_model.SampleEncodeAsPieces(text, 64, 0.1)
         new_pieces = []
         for piece in pieces:
-            if len(piece) > 1 and piece[-1] == ',' and piece[-2].isdigit():
+            if len(piece) > 1 and piece[-1] == str(',') and piece[-2].isdigit():
                 cur_pieces = self.sp_model.EncodeAsPieces(
                     piece[:-1].replace(SPIECE_UNDERLINE, ''))
                 if piece[0] != SPIECE_UNDERLINE and cur_pieces[0][0] == SPIECE_UNDERLINE:
@@ -227,7 +227,7 @@ class XLNetTokenizer(PreTrainedTokenizer):
         An XLNet sequence pair mask has the following format:
         0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 2
         | first sequence    | second sequence     | CLS segment ID
-        
+
         if token_ids_1 is None, only returns the first portion of the mask (0's).
         """
         sep = [self.sep_token_id]

From 2670b0d682746e1fe94ab9c7b4d2fd7f4af03193 Mon Sep 17 00:00:00 2001
From: Michael Watkins <me@michaelwatkins.eu>
Date: Wed, 4 Dec 2019 17:53:25 +0200
Subject: [PATCH 250/505] Fix bug which lowercases special tokens

---
 transformers/tests/tokenization_tests_commons.py |  8 +++++---
 transformers/tokenization_utils.py               | 15 +++++++++++++--
 2 files changed, 18 insertions(+), 5 deletions(-)

diff --git a/transformers/tests/tokenization_tests_commons.py b/transformers/tests/tokenization_tests_commons.py
index faff003f4b..d904f0067e 100644
--- a/transformers/tests/tokenization_tests_commons.py
+++ b/transformers/tests/tokenization_tests_commons.py
@@ -115,8 +115,10 @@ class CommonTestCases:
         def test_added_tokens_do_lower_case(self):
             tokenizer = self.get_tokenizer(do_lower_case=True)
 
-            text = "aaaaa bbbbbb low cccccccccdddddddd l"
-            text2 = "AAAAA BBBBBB low CCCCCCCCCDDDDDDDD l"
+            special_token = tokenizer.all_special_tokens[0]
+
+            text = special_token + " aaaaa bbbbbb low cccccccccdddddddd l " + special_token
+            text2 = special_token + " AAAAA BBBBBB low CCCCCCCCCDDDDDDDD l " + special_token
 
             toks0 = tokenizer.tokenize(text)  # toks before adding new_toks
 
@@ -141,7 +143,7 @@ class CommonTestCases:
 
             self.assertEqual(len(toks), len(toks2))  # Length should still be the same
             self.assertNotEqual(len(toks), len(toks0))
-            self.assertNotEqual(toks[0], toks2[0])  # But at least the first tokens should differ
+            self.assertNotEqual(toks[1], toks2[1])  # But at least the first non-special tokens should differ
 
         def test_add_tokens_tokenizer(self):
             tokenizer = self.get_tokenizer()
diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index 4c6cbd8986..eb22c50ebd 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -22,6 +22,7 @@ import json
 import six
 import copy
 import itertools
+import re
 from io import open
 
 from .file_utils import cached_path, is_tf_available, is_torch_available
@@ -520,7 +521,7 @@ class PreTrainedTokenizer(object):
         to_add_tokens = []
         for token in new_tokens:
             assert isinstance(token, str) or (six.PY2 and isinstance(token, unicode))
-            if self.init_kwargs.get('do_lower_case', False):
+            if self.init_kwargs.get('do_lower_case', False) and token not in self.all_special_tokens:
                 token = token.lower()
             if token != self.unk_token and \
                     self.convert_tokens_to_ids(token) == self.convert_tokens_to_ids(self.unk_token) and \
@@ -615,8 +616,18 @@ class PreTrainedTokenizer(object):
 
             Take care of added tokens.
         """
+        def lowercase_text(t):
+            # convert non-special tokens to lowercase
+            escaped_special_toks = [re.escape(s_tok) for s_tok in self.all_special_tokens]
+            pattern = r'(^' + r'|'.join(escaped_special_toks) + r')|' + \
+                      r'(.+?)'
+            return re.sub(
+                pattern,
+                lambda m: m.groups()[0] or m.groups()[1].lower(),
+                t)
+
         if self.init_kwargs.get('do_lower_case', False):
-            text = text.lower()
+            text = lowercase_text(text)
 
         def split_on_token(tok, text):
             result = []

From 0cb163865a4c761c226b151283309eedb2b1ca4d Mon Sep 17 00:00:00 2001
From: Aymeric Augustin <aymeric.augustin@fractalideas.com>
Date: Sat, 7 Dec 2019 13:46:14 +0100
Subject: [PATCH 251/505] Remove pytest dependency. (#2093)

---
 transformers/tests/optimization_tf_test.py | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/transformers/tests/optimization_tf_test.py b/transformers/tests/optimization_tf_test.py
index ac5109cb56..515d12a158 100644
--- a/transformers/tests/optimization_tf_test.py
+++ b/transformers/tests/optimization_tf_test.py
@@ -3,18 +3,19 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import pytest
 
 from transformers import is_tf_available
 
+from .utils import require_tf
+
 if is_tf_available():
     import tensorflow as tf
     from tensorflow.python.eager import context
     from tensorflow.python.framework import ops
     from transformers import (create_optimizer, GradientAccumulator)
-else:
-    pytestmark = pytest.mark.skip("Require TensorFlow")
 
+
+@require_tf
 class OptimizationFTest(unittest.TestCase):
     def assertListAlmostEqual(self, list1, list2, tol):
         self.assertEqual(len(list1), len(list2))

From 3520be7824ad11ebc05a393fd90ecfdd4203cfdb Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Mon, 9 Dec 2019 11:13:09 +0100
Subject: [PATCH 252/505] create encoder attention mask from shape of hidden
 states

We currently create encoder attention masks (when they're not provided)
based on the shape of the inputs to the encoder. This is obviously
wrong; sequences can be of different lengths. We now create the encoder
attention mask based on the batch_size and sequence_length of the
encoder hidden states.
---
 transformers/modeling_bert.py | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/transformers/modeling_bert.py b/transformers/modeling_bert.py
index 1ee3e3f097..8295cf4664 100644
--- a/transformers/modeling_bert.py
+++ b/transformers/modeling_bert.py
@@ -691,17 +691,19 @@ class BertModel(BertPreTrainedModel):
 
         # If a 2D ou 3D attention mask is provided for the cross-attention
         # we need to make broadcastabe to [batch_size, num_heads, seq_length, seq_length]
-        if self.config.is_decoder:
+        if self.config.is_decoder and encoder_hidden_states is not None:
+            encoder_batch_size, encoder_sequence_length, _ = encoder_hidden_states.size()
+            encoder_hidden_shape = (encoder_batch_size, encoder_sequence_length)
             if encoder_attention_mask is None:
-                encoder_attention_mask = torch.ones(input_shape, device=device)
+                encoder_attention_mask = torch.ones(encoder_hidden_shape, device=device)
 
             if encoder_attention_mask.dim() == 3:
                 encoder_extended_attention_mask = encoder_attention_mask[:, None, :, :]
             elif encoder_attention_mask.dim() == 2:
                 encoder_extended_attention_mask = encoder_attention_mask[:, None, None, :]
             else:
-                raise ValueError("Wrong shape for input_ids (shape {}) or encoder_attention_mask (shape {})".format(input_shape,
-                                                                                                                    encoder_attention_mask.shape))
+                raise ValueError("Wrong shape for encoder_hidden_shape (shape {}) or encoder_attention_mask (shape {})".format(encoder_hidden_shape,
+                                                                                                                               encoder_attention_mask.shape))
 
             encoder_extended_attention_mask = encoder_extended_attention_mask.to(dtype=next(self.parameters()).dtype)  # fp16 compatibility
             encoder_extended_attention_mask = (1.0 - encoder_extended_attention_mask) * -10000.0

From 169fea6855741315e2e0e15881cefc9823803aa6 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Mon, 9 Dec 2019 16:25:33 +0100
Subject: [PATCH 253/505] updating T5

---
 transformers/modeling_t5.py | 31 +++++++++++++------------------
 1 file changed, 13 insertions(+), 18 deletions(-)

diff --git a/transformers/modeling_t5.py b/transformers/modeling_t5.py
index 1bf55611a2..104e9060fc 100644
--- a/transformers/modeling_t5.py
+++ b/transformers/modeling_t5.py
@@ -281,7 +281,7 @@ class T5Attention(nn.Module):
         context_position = torch.arange(qlen, dtype=torch.long)[:, None]
         memory_position = torch.arange(klen, dtype=torch.long)[None, :]
         relative_position = memory_position - context_position  # shape (qlen, klen)
-        rp_bucket = self._relative_position_bucket(relative_position,
+        rp_bucket = self._relative_position_bucket(relative_position,  # shape (qlen, klen)
                                                    bidirectional=not self.is_decoder,
                                                    num_buckets=self.relative_attention_num_buckets)
         values = self.relative_attention_bias(rp_bucket)  # shape (qlen, klen, num_heads)
@@ -337,14 +337,10 @@ class T5Attention(nn.Module):
             if not self.has_relative_attention_bias:
                 raise ValueError("No position_bias provided and no weights to compute position_bias")
             position_bias = self.compute_bias(qlen, klen)
+            if mask is not None:
+                position_bias += mask                                         # (bs, n_heads, qlen, klen)
+
         scores += position_bias
-        special_out = position_bias
-
-        if mask is not None:
-            scores += mask
-            # mask = (mask == 0).expand_as(scores)                              # (bs, n_heads, qlen, klen)
-            # scores.masked_fill_(mask, -float('inf'))                          # (bs, n_heads, qlen, klen)
-
         weights = F.softmax(scores.float(), dim=-1).type_as(scores)           # (bs, n_heads, qlen, klen)
         weights = F.dropout(weights, p=self.dropout, training=self.training)  # (bs, n_heads, qlen, klen)
 
@@ -362,7 +358,7 @@ class T5Attention(nn.Module):
             outputs = outputs + (weights,)
         if self.has_relative_attention_bias:
             outputs = outputs + (position_bias,)
-        return outputs + (special_out,)
+        return outputs
 
 
 class T5LayerSelfAttention(nn.Module):
@@ -379,11 +375,9 @@ class T5LayerSelfAttention(nn.Module):
                                               position_bias=position_bias,
                                               head_mask=head_mask)
         y = attention_output[0]
-        special_out = attention_output[-1]
-        attention_output = attention_output[:-1]
         layer_output = hidden_states + self.dropout(y)
         outputs = (layer_output,) + attention_output[1:]  # add attentions if we output them
-        return outputs + (special_out,)
+        return outputs
 
 
 class T5LayerCrossAttention(nn.Module):
@@ -426,8 +420,7 @@ class T5Block(nn.Module):
                                                 position_bias=position_bias,
                                                 head_mask=head_mask)
         hidden_states = self_attention_outputs[0]
-        special_out = self_attention_outputs[-1]
-        outputs = self_attention_outputs[1:-1]  # Keep self-attention outputs and relative position weights
+        outputs = self_attention_outputs[1:]  # Keep self-attention outputs and relative position weights
 
         if not self.is_decoder:
             hidden_states = self.layer[1](hidden_states)
@@ -442,7 +435,7 @@ class T5Block(nn.Module):
             hidden_states = self.layer[2](hidden_states)
 
         outputs = (hidden_states,) + outputs  # add attentions if we output them
-        return outputs + (special_out,)  # hidden-states, (self-attention weights), (self-attention position bias), (cross-attention weights), (cross-attention position bias)
+        return outputs  # hidden-states, (self-attention weights), (self-attention position bias), (cross-attention weights), (cross-attention position bias)
 
 
 class T5PreTrainedModel(PreTrainedModel):
@@ -536,6 +529,10 @@ class T5Stack(T5PreTrainedModel):
         # positions we want to attend and -1e9 for masked positions.
         # Since we are adding it to the raw scores before the softmax, this is
         # effectively the same as removing these entirely.
+
+        # T5 has a mask that can compare sequence ids, we simulate this here with this transposistion
+        # Cf. https://github.com/tensorflow/mesh/blob/8d2465e9bc93129b913b5ccc6a59aa97abd96ec6/mesh_tensorflow/transformer/transformer_layers.py#L270
+        extended_attention_mask = (extended_attention_mask == extended_attention_mask.transpose(-1, -2))
         extended_attention_mask = extended_attention_mask.to(dtype=next(self.parameters()).dtype)  # fp16 compatibility
         extended_attention_mask = (1.0 - extended_attention_mask) * -1e9
 
@@ -584,8 +581,6 @@ class T5Stack(T5PreTrainedModel):
                                          encoder_attention_mask=encoder_extended_attention_mask,
                                          encoder_decoder_position_bias=encoder_decoder_position_bias,
                                          head_mask=head_mask[i])
-            if i == 0:
-                special_out = layer_outputs[-1]
             # layer_outputs is a tuple with:
             # hidden-states, (self-attention weights), (self-attention position bias), (cross-attention weights), (cross-attention position bias)
             hidden_states = layer_outputs[0]
@@ -610,7 +605,7 @@ class T5Stack(T5PreTrainedModel):
             outputs = outputs + (all_hidden_states,)
         if self.output_attentions:
             outputs = outputs + (all_attentions,)
-        return outputs + (special_out,)  # last-layer hidden state, (all hidden states), (all attentions)
+        return outputs  # last-layer hidden state, (all hidden states), (all attentions)
 
 
 T5_START_DOCSTRING = r"""    The T5 model was proposed in

From 2a4ef098d65939d436e2a5efbb518fb807b6b1b6 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Mon, 9 Dec 2019 10:46:47 -0500
Subject: [PATCH 254/505] Add ALBERT and XLM to SQuAD script

---
 examples/run_squad.py | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/examples/run_squad.py b/examples/run_squad.py
index a8ac1d1b05..2df29014ef 100644
--- a/examples/run_squad.py
+++ b/examples/run_squad.py
@@ -44,7 +44,9 @@ from transformers import (WEIGHTS_NAME, BertConfig,
                                   XLNetForQuestionAnswering,
                                   XLNetTokenizer,
                                   DistilBertConfig, DistilBertForQuestionAnswering, DistilBertTokenizer,
-                                  AlbertConfig, AlbertForQuestionAnswering, AlbertTokenizer)
+                                  AlbertConfig, AlbertForQuestionAnswering, AlbertTokenizer,
+                                  XLMConfig, XLMForQuestionAnswering, XLMTokenizer,
+                                  )
 
 from transformers import AdamW, get_linear_schedule_with_warmup, squad_convert_examples_to_features
 
@@ -58,7 +60,8 @@ MODEL_CLASSES = {
     'xlnet': (XLNetConfig, XLNetForQuestionAnswering, XLNetTokenizer),
     'xlm': (XLMConfig, XLMForQuestionAnswering, XLMTokenizer),
     'distilbert': (DistilBertConfig, DistilBertForQuestionAnswering, DistilBertTokenizer),
-    'albert': (AlbertConfig, AlbertForQuestionAnswering, AlbertTokenizer)
+    'albert': (AlbertConfig, AlbertForQuestionAnswering, AlbertTokenizer),
+    'xlm': (XLMConfig, XLMForQuestionAnswering, XLMTokenizer)
 }
 
 def set_seed(args):

From b016dd16c90c2c18168d13bca6d5002729fd5b0a Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Mon, 9 Dec 2019 21:38:07 +0100
Subject: [PATCH 255/505] fix tests on python 3.5

---
 transformers/modeling_t5.py                |  2 +-
 transformers/tests/modeling_common_test.py | 15 ++++++++-------
 transformers/tokenization_t5.py            |  2 +-
 3 files changed, 10 insertions(+), 9 deletions(-)

diff --git a/transformers/modeling_t5.py b/transformers/modeling_t5.py
index 104e9060fc..e48293b49e 100644
--- a/transformers/modeling_t5.py
+++ b/transformers/modeling_t5.py
@@ -338,7 +338,7 @@ class T5Attention(nn.Module):
                 raise ValueError("No position_bias provided and no weights to compute position_bias")
             position_bias = self.compute_bias(qlen, klen)
             if mask is not None:
-                position_bias += mask                                         # (bs, n_heads, qlen, klen)
+                position_bias = position_bias + mask                          # (bs, n_heads, qlen, klen)
 
         scores += position_bias
         weights = F.softmax(scores.float(), dim=-1).type_as(scores)           # (bs, n_heads, qlen, klen)
diff --git a/transformers/tests/modeling_common_test.py b/transformers/tests/modeling_common_test.py
index ee75da605c..11aeaafe31 100644
--- a/transformers/tests/modeling_common_test.py
+++ b/transformers/tests/modeling_common_test.py
@@ -138,8 +138,8 @@ class CommonTestCases:
                 self.assertListEqual(
                     list(attentions[0].shape[-3:]),
                     [self.model_tester.num_attention_heads,
-                    self.model_tester.seq_length,
-                    self.model_tester.key_len if hasattr(self.model_tester, 'key_len') else self.model_tester.seq_length])
+                    self.model_tester.encoder_seq_length if hasattr(self.model_tester, 'encoder_seq_length') else self.model_tester.seq_length,
+                    self.model_tester.encoder_seq_length if hasattr(self.model_tester, 'encoder_seq_length') else self.model_tester.seq_length])
                 out_len = len(outputs)
 
                 if self.is_encoder_decoder:
@@ -151,8 +151,8 @@ class CommonTestCases:
                     self.assertListEqual(
                         list(decoder_attentions[0].shape[-3:]),
                         [self.model_tester.num_attention_heads,
-                         self.model_tester.seq_length,
-                         self.model_tester.key_len if hasattr(self.model_tester, 'key_len') else self.model_tester.seq_length])
+                         self.model_tester.decoder_seq_length if hasattr(self.model_tester, 'decoder_seq_length') else self.model_tester.seq_length,
+                         self.model_tester.decoder_seq_length if hasattr(self.model_tester, 'decoder_seq_length') else self.model_tester.seq_length])
 
                 # Check attention is always last and order is fine
                 config.output_attentions = True
@@ -169,8 +169,8 @@ class CommonTestCases:
                 self.assertListEqual(
                     list(self_attentions[0].shape[-3:]),
                     [self.model_tester.num_attention_heads,
-                    self.model_tester.seq_length,
-                    self.model_tester.key_len if hasattr(self.model_tester, 'key_len') else self.model_tester.seq_length])
+                    self.model_tester.encoder_seq_length if hasattr(self.model_tester, 'encoder_seq_length') else self.model_tester.seq_length,
+                    self.model_tester.encoder_seq_length if hasattr(self.model_tester, 'encoder_seq_length') else self.model_tester.seq_length])
 
         def test_torchscript(self):
             config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
@@ -440,7 +440,8 @@ class CommonTestCases:
                 self.assertEqual(len(hidden_states), self.model_tester.num_hidden_layers + 1)
                 self.assertListEqual(
                     list(hidden_states[0].shape[-2:]),
-                    [self.model_tester.seq_length, self.model_tester.hidden_size])
+                    [self.model_tester.encoder_seq_length if hasattr(self.model_tester, 'encoder_seq_length') else self.model_tester.seq_length,
+                     self.model_tester.hidden_size])
 
         def test_resize_tokens_embeddings(self):
             original_config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
diff --git a/transformers/tokenization_t5.py b/transformers/tokenization_t5.py
index 3847aeefbf..933084d13a 100644
--- a/transformers/tokenization_t5.py
+++ b/transformers/tokenization_t5.py
@@ -134,7 +134,7 @@ class T5Tokenizer(PreTrainedTokenizer):
         """ Converts a token (str/unicode) in an id using the vocab. """
         if token.startswith(u"<extra_id_"):
             l = re.match(r'<extra_id_(\d+)>', token)
-            num = int(l[1])
+            num = int(l.group(1))
             return self.vocab_size - num - 1
         return self.sp_model.piece_to_id(token)
 

From 808bb8da7edbd9f5858b3c223ebac9bd83275934 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Mon, 9 Dec 2019 21:48:34 +0100
Subject: [PATCH 256/505] fix transfo xl tests

---
 transformers/tests/modeling_common_test.py     | 18 ++++++++++++------
 .../tests/modeling_tf_transfo_xl_test.py       |  2 +-
 transformers/tests/modeling_transfo_xl_test.py |  2 +-
 3 files changed, 14 insertions(+), 8 deletions(-)

diff --git a/transformers/tests/modeling_common_test.py b/transformers/tests/modeling_common_test.py
index 11aeaafe31..7033a06d0b 100644
--- a/transformers/tests/modeling_common_test.py
+++ b/transformers/tests/modeling_common_test.py
@@ -125,6 +125,11 @@ class CommonTestCases:
         def test_attention_outputs(self):
             config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
 
+            decoder_seq_length = self.model_tester.decoder_seq_length if hasattr(self.model_tester, 'decoder_seq_length') else self.model_tester.seq_length
+            encoder_seq_length = self.model_tester.encoder_seq_length if hasattr(self.model_tester, 'encoder_seq_length') else self.model_tester.seq_length
+            decoder_key_length = self.model_tester.key_length if hasattr(self.model_tester, 'key_length') else decoder_seq_length
+            encoder_key_length = self.model_tester.key_length if hasattr(self.model_tester, 'key_length') else encoder_seq_length
+
             for model_class in self.all_model_classes:
                 config.output_attentions = True
                 config.output_hidden_states = False
@@ -138,8 +143,8 @@ class CommonTestCases:
                 self.assertListEqual(
                     list(attentions[0].shape[-3:]),
                     [self.model_tester.num_attention_heads,
-                    self.model_tester.encoder_seq_length if hasattr(self.model_tester, 'encoder_seq_length') else self.model_tester.seq_length,
-                    self.model_tester.encoder_seq_length if hasattr(self.model_tester, 'encoder_seq_length') else self.model_tester.seq_length])
+                    encoder_seq_length ,
+                    encoder_key_length])
                 out_len = len(outputs)
 
                 if self.is_encoder_decoder:
@@ -151,8 +156,9 @@ class CommonTestCases:
                     self.assertListEqual(
                         list(decoder_attentions[0].shape[-3:]),
                         [self.model_tester.num_attention_heads,
-                         self.model_tester.decoder_seq_length if hasattr(self.model_tester, 'decoder_seq_length') else self.model_tester.seq_length,
-                         self.model_tester.decoder_seq_length if hasattr(self.model_tester, 'decoder_seq_length') else self.model_tester.seq_length])
+                         decoder_seq_length,
+                         decoder_key_length
+                         ])
 
                 # Check attention is always last and order is fine
                 config.output_attentions = True
@@ -169,8 +175,8 @@ class CommonTestCases:
                 self.assertListEqual(
                     list(self_attentions[0].shape[-3:]),
                     [self.model_tester.num_attention_heads,
-                    self.model_tester.encoder_seq_length if hasattr(self.model_tester, 'encoder_seq_length') else self.model_tester.seq_length,
-                    self.model_tester.encoder_seq_length if hasattr(self.model_tester, 'encoder_seq_length') else self.model_tester.seq_length])
+                    encoder_seq_length,
+                    encoder_key_length])
 
         def test_torchscript(self):
             config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
diff --git a/transformers/tests/modeling_tf_transfo_xl_test.py b/transformers/tests/modeling_tf_transfo_xl_test.py
index 534fe39646..8ebd749b4c 100644
--- a/transformers/tests/modeling_tf_transfo_xl_test.py
+++ b/transformers/tests/modeling_tf_transfo_xl_test.py
@@ -68,7 +68,7 @@ class TFTransfoXLModelTest(TFCommonTestCases.TFCommonModelTester):
             self.batch_size = batch_size
             self.seq_length = seq_length
             self.mem_len = mem_len
-            self.key_len = seq_length + mem_len
+            self.key_length = seq_length + mem_len
             self.clamp_len = clamp_len
             self.is_training = is_training
             self.use_labels = use_labels
diff --git a/transformers/tests/modeling_transfo_xl_test.py b/transformers/tests/modeling_transfo_xl_test.py
index f7b913da5b..2d1541d87b 100644
--- a/transformers/tests/modeling_transfo_xl_test.py
+++ b/transformers/tests/modeling_transfo_xl_test.py
@@ -66,7 +66,7 @@ class TransfoXLModelTest(CommonTestCases.CommonModelTester):
             self.batch_size = batch_size
             self.seq_length = seq_length
             self.mem_len = mem_len
-            self.key_len = seq_length + mem_len
+            self.key_length = seq_length + mem_len
             self.clamp_len = clamp_len
             self.is_training = is_training
             self.use_labels = use_labels

From 8e651f56b75982f07fc522b62f298d8d70e6e56f Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Mon, 9 Dec 2019 22:13:57 +0100
Subject: [PATCH 257/505] fix tf tests

---
 transformers/tests/modeling_tf_common_test.py | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

diff --git a/transformers/tests/modeling_tf_common_test.py b/transformers/tests/modeling_tf_common_test.py
index 20ccfd8ce0..26bd037c9e 100644
--- a/transformers/tests/modeling_tf_common_test.py
+++ b/transformers/tests/modeling_tf_common_test.py
@@ -213,6 +213,11 @@ class TFCommonTestCases:
         def test_attention_outputs(self):
             config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
 
+            decoder_seq_length = self.model_tester.decoder_seq_length if hasattr(self.model_tester, 'decoder_seq_length') else self.model_tester.seq_length
+            encoder_seq_length = self.model_tester.encoder_seq_length if hasattr(self.model_tester, 'encoder_seq_length') else self.model_tester.seq_length
+            decoder_key_length = self.model_tester.key_length if hasattr(self.model_tester, 'key_length') else decoder_seq_length
+            encoder_key_length = self.model_tester.key_length if hasattr(self.model_tester, 'key_length') else encoder_seq_length
+
             for model_class in self.all_model_classes:
                 config.output_attentions = True
                 config.output_hidden_states = False
@@ -225,8 +230,8 @@ class TFCommonTestCases:
                 self.assertListEqual(
                     list(attentions[0].shape[-3:]),
                     [self.model_tester.num_attention_heads,
-                    self.model_tester.seq_length,
-                    self.model_tester.key_len if hasattr(self.model_tester, 'key_len') else self.model_tester.seq_length])
+                    encoder_seq_length,
+                    encoder_key_length])
                 out_len = len(outputs)
 
                 if self.is_encoder_decoder:
@@ -238,8 +243,8 @@ class TFCommonTestCases:
                     self.assertListEqual(
                         list(decoder_attentions[0].shape[-3:]),
                         [self.model_tester.num_attention_heads,
-                         self.model_tester.seq_length,
-                         self.model_tester.key_len if hasattr(self.model_tester, 'key_len') else self.model_tester.seq_length])
+                         decoder_seq_length,
+                         decoder_key_length])
 
                 # Check attention is always last and order is fine
                 config.output_attentions = True
@@ -255,8 +260,8 @@ class TFCommonTestCases:
                 self.assertListEqual(
                     list(attentions[0].shape[-3:]),
                     [self.model_tester.num_attention_heads,
-                    self.model_tester.seq_length,
-                    self.model_tester.key_len if hasattr(self.model_tester, 'key_len') else self.model_tester.seq_length])
+                    encoder_seq_length,
+                    encoder_key_length])
 
         def test_headmasking(self):
             pass

From f71b1bb05a20879953d57bf648ab7bbd2b3239bc Mon Sep 17 00:00:00 2001
From: Bilal Khan <bk@tinymanager.com>
Date: Wed, 27 Nov 2019 08:39:00 -0600
Subject: [PATCH 258/505] Save optimizer state, scheduler state and current
 epoch

---
 examples/run_lm_finetuning.py | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/examples/run_lm_finetuning.py b/examples/run_lm_finetuning.py
index a5eaf524ac..3cae206460 100644
--- a/examples/run_lm_finetuning.py
+++ b/examples/run_lm_finetuning.py
@@ -224,7 +224,7 @@ def train(args, train_dataset, model, tokenizer):
     model.zero_grad()
     train_iterator = trange(int(args.num_train_epochs), desc="Epoch", disable=args.local_rank not in [-1, 0])
     set_seed(args)  # Added here for reproducibility (even between python 2 and 3)
-    for _ in train_iterator:
+    for epoch in train_iterator:
         epoch_iterator = tqdm(train_dataloader, desc="Iteration", disable=args.local_rank not in [-1, 0])
         for step, batch in enumerate(epoch_iterator):
             inputs, labels = mask_tokens(batch, tokenizer, args) if args.mlm else (batch, batch)
@@ -279,6 +279,10 @@ def train(args, train_dataset, model, tokenizer):
 
                     _rotate_checkpoints(args, checkpoint_prefix)
 
+                    torch.save(optimizer.state_dict(), os.path.join(output_dir, 'optimizer.pt'))
+                    torch.save(scheduler.state_dict(), os.path.join(output_dir, 'scheduler.pt'))
+                    torch.save(epoch, os.path.join(output_dir, 'training_state.pt'))
+
             if args.max_steps > 0 and global_step > args.max_steps:
                 epoch_iterator.close()
                 break

From a03fcf570de4a90218efd4b3de253d4648fe24b1 Mon Sep 17 00:00:00 2001
From: Bilal Khan <bk@tinymanager.com>
Date: Wed, 27 Nov 2019 18:42:07 -0600
Subject: [PATCH 259/505] Save tokenizer after each epoch to be able to resume
 training from a checkpoint

---
 examples/run_lm_finetuning.py | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/examples/run_lm_finetuning.py b/examples/run_lm_finetuning.py
index 3cae206460..1d93aa4381 100644
--- a/examples/run_lm_finetuning.py
+++ b/examples/run_lm_finetuning.py
@@ -274,6 +274,8 @@ def train(args, train_dataset, model, tokenizer):
                         os.makedirs(output_dir)
                     model_to_save = model.module if hasattr(model, 'module') else model  # Take care of distributed/parallel training
                     model_to_save.save_pretrained(output_dir)
+                    tokenizer.save_pretrained(output_dir)
+
                     torch.save(args, os.path.join(output_dir, 'training_args.bin'))
                     logger.info("Saving model checkpoint to %s", output_dir)
 
@@ -282,6 +284,7 @@ def train(args, train_dataset, model, tokenizer):
                     torch.save(optimizer.state_dict(), os.path.join(output_dir, 'optimizer.pt'))
                     torch.save(scheduler.state_dict(), os.path.join(output_dir, 'scheduler.pt'))
                     torch.save(epoch, os.path.join(output_dir, 'training_state.pt'))
+                    logger.info("Saving training state to %s", output_dir)
 
             if args.max_steps > 0 and global_step > args.max_steps:
                 epoch_iterator.close()

From 0eb973b0d99e5c219af8a93b6267bda00c7161c6 Mon Sep 17 00:00:00 2001
From: Bilal Khan <bk@tinymanager.com>
Date: Wed, 27 Nov 2019 19:10:24 -0600
Subject: [PATCH 260/505] Use saved optimizer and scheduler states if available

---
 examples/run_lm_finetuning.py | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/examples/run_lm_finetuning.py b/examples/run_lm_finetuning.py
index 1d93aa4381..9bdbf9ca56 100644
--- a/examples/run_lm_finetuning.py
+++ b/examples/run_lm_finetuning.py
@@ -188,6 +188,13 @@ def train(args, train_dataset, model, tokenizer):
         ]
     optimizer = AdamW(optimizer_grouped_parameters, lr=args.learning_rate, eps=args.adam_epsilon)
     scheduler = get_linear_schedule_with_warmup(optimizer, num_warmup_steps=args.warmup_steps, num_training_steps=t_total)
+
+    # Check if saved optimizer or scheduler states exist
+    if os.path.isfile(os.path.join(args.model_name_or_path, 'optimizer.pt')) and os.path.isfile(os.path.join(args.model_name_or_path, 'scheduler.pt')):
+        # Load in optimizer and scheduler states
+        optimizer.load_state_dict(torch.load(os.path.join(args.model_name_or_path, 'optimizer.pt')))
+        scheduler.load_state_dict(torch.load(os.path.join(args.model_name_or_path, 'scheduler.pt')))
+
     if args.fp16:
         try:
             from apex import amp

From 2d73591a1831e80d0743b514d7f0138c4879e37b Mon Sep 17 00:00:00 2001
From: Bilal Khan <bk@tinymanager.com>
Date: Wed, 27 Nov 2019 19:13:10 -0600
Subject: [PATCH 261/505] Stop saving current epoch

---
 examples/run_lm_finetuning.py | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/examples/run_lm_finetuning.py b/examples/run_lm_finetuning.py
index 9bdbf9ca56..5e7683b85d 100644
--- a/examples/run_lm_finetuning.py
+++ b/examples/run_lm_finetuning.py
@@ -290,8 +290,7 @@ def train(args, train_dataset, model, tokenizer):
 
                     torch.save(optimizer.state_dict(), os.path.join(output_dir, 'optimizer.pt'))
                     torch.save(scheduler.state_dict(), os.path.join(output_dir, 'scheduler.pt'))
-                    torch.save(epoch, os.path.join(output_dir, 'training_state.pt'))
-                    logger.info("Saving training state to %s", output_dir)
+                    logger.info("Saving optimizer and scheduler states to %s", output_dir)
 
             if args.max_steps > 0 and global_step > args.max_steps:
                 epoch_iterator.close()

From 9626e0458c20b61c18c9564ecc4d1261a4a66e50 Mon Sep 17 00:00:00 2001
From: Bilal Khan <bk@tinymanager.com>
Date: Wed, 27 Nov 2019 20:00:16 -0600
Subject: [PATCH 262/505] Add functionality to continue training from last
 saved global_step

---
 examples/run_lm_finetuning.py | 22 +++++++++++++++++++++-
 1 file changed, 21 insertions(+), 1 deletion(-)

diff --git a/examples/run_lm_finetuning.py b/examples/run_lm_finetuning.py
index 5e7683b85d..172d4e20e2 100644
--- a/examples/run_lm_finetuning.py
+++ b/examples/run_lm_finetuning.py
@@ -223,17 +223,37 @@ def train(args, train_dataset, model, tokenizer):
     logger.info("  Total optimization steps = %d", t_total)
 
     global_step = 0
+    epochs_trained = 0
+    steps_trained_in_current_epoch = 0
+    # Check if continuing training from a checkpoint
+    if os.path.exists(args.model_name_or_path):
+        # set global_step to gobal_step of last saved checkpoint from model path
+        global_step = int(args.model_name_or_path.split('-')[-1].split('/')[0])
+        epochs_trained = global_step // (len(train_dataloader) // args.gradient_accumulation_steps)
+        steps_trained_in_current_epoch = global_step % (len(train_dataloader) // args.gradient_accumulation_steps)
+
+        logger.info("  Continuing training from checkpoint, will skip to saved global_step")
+        logger.info("  Continuing training from epoch %d", epochs_trained)
+        logger.info("  Continuing training from global step %d", global_step)
+        logger.info("  Will skip the first %d steps in the first epoch", steps_trained_in_current_epoch)
+
     tr_loss, logging_loss = 0.0, 0.0
 
     model_to_resize = model.module if hasattr(model, 'module') else model  # Take care of distributed/parallel training
     model_to_resize.resize_token_embeddings(len(tokenizer))
 
     model.zero_grad()
-    train_iterator = trange(int(args.num_train_epochs), desc="Epoch", disable=args.local_rank not in [-1, 0])
+    train_iterator = trange(epochs_trained, int(args.num_train_epochs), desc="Epoch", disable=args.local_rank not in [-1, 0])
     set_seed(args)  # Added here for reproducibility (even between python 2 and 3)
     for epoch in train_iterator:
         epoch_iterator = tqdm(train_dataloader, desc="Iteration", disable=args.local_rank not in [-1, 0])
         for step, batch in enumerate(epoch_iterator):
+            
+            # Skip past any already trained steps if resuming training
+            if steps_trained_in_current_epoch > 0:
+                steps_trained_in_current_epoch -= 1
+                continue
+
             inputs, labels = mask_tokens(batch, tokenizer, args) if args.mlm else (batch, batch)
             inputs = inputs.to(args.device)
             labels = labels.to(args.device)

From 79526f82f5d6757812f3691949cf03b864697f46 Mon Sep 17 00:00:00 2001
From: Bilal Khan <bk@tinymanager.com>
Date: Thu, 28 Nov 2019 19:20:29 -0600
Subject: [PATCH 263/505] Remove unnecessary epoch variable

---
 examples/run_lm_finetuning.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/examples/run_lm_finetuning.py b/examples/run_lm_finetuning.py
index 172d4e20e2..c4c73e71af 100644
--- a/examples/run_lm_finetuning.py
+++ b/examples/run_lm_finetuning.py
@@ -245,7 +245,7 @@ def train(args, train_dataset, model, tokenizer):
     model.zero_grad()
     train_iterator = trange(epochs_trained, int(args.num_train_epochs), desc="Epoch", disable=args.local_rank not in [-1, 0])
     set_seed(args)  # Added here for reproducibility (even between python 2 and 3)
-    for epoch in train_iterator:
+    for _ in train_iterator:
         epoch_iterator = tqdm(train_dataloader, desc="Iteration", disable=args.local_rank not in [-1, 0])
         for step, batch in enumerate(epoch_iterator):
             

From 5c877fe94a1cbb70132515a9da6a464bf6da49ed Mon Sep 17 00:00:00 2001
From: Pierric Cistac <pierric@huggingface.co>
Date: Mon, 9 Dec 2019 18:53:00 -0500
Subject: [PATCH 264/505] fix albert links

---
 README.md                         |  2 +-
 docs/source/index.rst             |  2 +-
 docs/source/pretrained_models.rst | 16 ++++++++--------
 3 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/README.md b/README.md
index 64ec631651..f3aa8a95ee 100644
--- a/README.md
+++ b/README.md
@@ -143,7 +143,7 @@ At some point in the future, you'll be able to seamlessly move from pre-training
 8. **[DistilBERT](https://github.com/huggingface/transformers/tree/master/examples/distillation)** (from HuggingFace), released together with the paper [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://arxiv.org/abs/1910.01108) by Victor Sanh, Lysandre Debut and Thomas Wolf. The same method has been applied to compress GPT2 into [DistilGPT2](https://github.com/huggingface/transformers/tree/master/examples/distillation), RoBERTa into [DistilRoBERTa](https://github.com/huggingface/transformers/tree/master/examples/distillation), Multilingual BERT into [DistilmBERT](https://github.com/huggingface/transformers/tree/master/examples/distillation) and a German version of DistilBERT.
 9. **[CTRL](https://github.com/salesforce/ctrl/)** (from Salesforce) released with the paper [CTRL: A Conditional Transformer Language Model for Controllable Generation](https://arxiv.org/abs/1909.05858) by Nitish Shirish Keskar*, Bryan McCann*, Lav R. Varshney, Caiming Xiong and Richard Socher.
 10. **[CamemBERT](https://camembert-model.fr)** (from Inria/Facebook/Sorbonne) released with the paper [CamemBERT: a Tasty French Language Model](https://arxiv.org/abs/1911.03894) by Louis Martin*, Benjamin Muller*, Pedro Javier Ortiz Suárez*, Yoann Dupont, Laurent Romary, Éric Villemonte de la Clergerie, Djamé Seddah and Benoît Sagot.
-11. **[ALBERT](https://github.com/google-research/google-research/tree/master/albert)** (from Google Research and the Toyota Technological Institute at Chicago) released with the paper [ALBERT: A Lite BERT for Self-supervised Learning of Language Representations](https://arxiv.org/abs/1909.11942), by Zhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut.
+11. **[ALBERT](https://github.com/google-research/ALBERT)** (from Google Research and the Toyota Technological Institute at Chicago) released with the paper [ALBERT: A Lite BERT for Self-supervised Learning of Language Representations](https://arxiv.org/abs/1909.11942), by Zhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut.
 11. Want to contribute a new model? We have added a **detailed guide and templates** to guide you in the process of adding a new model. You can find them in the [`templates`](./templates) folder of the repository. Be sure to check the [contributing guidelines](./CONTRIBUTING.md) and contact the maintainers or open an issue to collect feedbacks before starting your PR.
 
 These implementations have been tested on several datasets (see the example scripts) and should match the performances of the original implementations (e.g. ~93 F1 on SQuAD for BERT Whole-Word-Masking, ~88 F1 on RocStories for OpenAI GPT, ~18.3 perplexity on WikiText 103 for Transformer-XL, ~0.916 Peason R coefficient on STS-B for XLNet). You can find more details on the performances in the Examples section of the [documentation](https://huggingface.co/transformers/examples.html).
diff --git a/docs/source/index.rst b/docs/source/index.rst
index 55ead33b4d..84012fc6cf 100644
--- a/docs/source/index.rst
+++ b/docs/source/index.rst
@@ -49,7 +49,7 @@ The library currently contains PyTorch and Tensorflow implementations, pre-train
 8. `DistilBERT <https://huggingface.co/transformers/model_doc/distilbert.html>`_ (from HuggingFace) released together with the paper `DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter <https://arxiv.org/abs/1910.01108>`_ by Victor Sanh, Lysandre Debut and Thomas Wolf. The same method has been applied to compress GPT2 into `DistilGPT2 <https://github.com/huggingface/transformers/tree/master/examples/distillation>`_.
 9. `CTRL <https://github.com/pytorch/fairseq/tree/master/examples/ctrl>`_ (from Salesforce), released together with the paper `CTRL: A Conditional Transformer Language Model for Controllable Generation <https://www.github.com/salesforce/ctrl>`_ by Nitish Shirish Keskar*, Bryan McCann*, Lav R. Varshney, Caiming Xiong and Richard Socher.
 10. `CamemBERT <https://huggingface.co/transformers/model_doc/camembert.html>`_ (from FAIR, Inria, Sorbonne Université) released together with the paper `CamemBERT: a Tasty French Language Model <https://arxiv.org/abs/1911.03894>`_ by Louis Martin, Benjamin Muller, Pedro Javier Ortiz Suarez, Yoann Dupont, Laurent Romary, Eric Villemonte de la Clergerie, Djame Seddah, and Benoît Sagot.
-11. `ALBERT <https://github.com/pytorch/fairseq/tree/master/examples/albert>`_ (from Google Research), released together with the paper a `ALBERT: A Lite BERT for Self-supervised Learning of Language Representations <https://arxiv.org/abs/1909.11942>`_ by Zhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut.
+11. `ALBERT <https://github.com/google-research/ALBERT>`_ (from Google Research), released together with the paper a `ALBERT: A Lite BERT for Self-supervised Learning of Language Representations <https://arxiv.org/abs/1909.11942>`_ by Zhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut.
 
 .. toctree::
     :maxdepth: 2
diff --git a/docs/source/pretrained_models.rst b/docs/source/pretrained_models.rst
index 090cb75808..dd61f11769 100644
--- a/docs/source/pretrained_models.rst
+++ b/docs/source/pretrained_models.rst
@@ -169,35 +169,35 @@ Here is the full list of the currently provided pretrained models together with
 +-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 | ALBERT            | ``albert-base-v1``                                         | | 12 repeating layers, 128 embedding, 768-hidden, 12-heads, 11M parameters                                                            |
 |                   |                                                            | | ALBERT base model                                                                                                                   |
-|                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
+|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                             |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``albert-large-v1``                                        | | 24 repeating layers, 128 embedding, 1024-hidden, 16-heads, 17M parameters                                                           |
 |                   |                                                            | | ALBERT large model                                                                                                                  |
-|                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
+|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                             |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``albert-xlarge-v1``                                       | | 24 repeating layers, 128 embedding, 2048-hidden, 16-heads, 58M parameters                                                           |
 |                   |                                                            | | ALBERT xlarge model                                                                                                                 |
-|                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
+|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                             |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``albert-xxlarge-v1``                                      | | 12 repeating layer, 128 embedding, 4096-hidden, 64-heads, 223M parameters                                                           |
 |                   |                                                            | | ALBERT xxlarge model                                                                                                                |
-|                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
+|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                             |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``albert-base-v2``                                         | | 12 repeating layers, 128 embedding, 768-hidden, 12-heads, 11M parameters                                                            |
 |                   |                                                            | | ALBERT base model with no dropout, additional training data and longer training                                                     |
-|                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
+|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                             |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``albert-large-v2``                                        | | 24 repeating layers, 128 embedding, 1024-hidden, 16-heads, 17M parameters                                                           |
 |                   |                                                            | | ALBERT large model with no dropout, additional training data and longer training                                                    |
-|                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
+|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                             |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``albert-xlarge-v2``                                       | | 24 repeating layers, 128 embedding, 2048-hidden, 16-heads, 58M parameters                                                           |
 |                   |                                                            | | ALBERT xlarge model with no dropout, additional training data and longer training                                                   |
-|                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
+|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                             |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``albert-xxlarge-v2``                                      | | 12 repeating layer, 128 embedding, 4096-hidden, 64-heads, 223M parameters                                                           |
 |                   |                                                            | | ALBERT xxlarge model with no dropout, additional training data and longer training                                                  |
-|                   |                                                            | (see `details <https://github.com/google-research/google-research/tree/master/albert>`__)                                             |
+|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                             |
 +-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 
 

From 07f4cd73f6d13e43b69a6e34a2a756a80fc7f70b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Thu, 31 Oct 2019 09:48:27 +0100
Subject: [PATCH 265/505] update function to add special tokens

Since I started my PR the `add_special_token_single_sequence` function
has been deprecated for another; I replaced it with the new function.
---
 examples/utils_summarization.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/examples/utils_summarization.py b/examples/utils_summarization.py
index 327ca8cc3e..087c88bd4e 100644
--- a/examples/utils_summarization.py
+++ b/examples/utils_summarization.py
@@ -139,11 +139,11 @@ def encode_for_summarization(story_lines, summary_lines, tokenizer):
     sentences.
     """
     story_lines_token_ids = [
-        tokenizer.add_special_tokens_single_sequence(tokenizer.encode(line))
+        tokenizer.build_inputs_with_special_tokens(tokenizer.encode(line))
         for line in story_lines
     ]
     summary_lines_token_ids = [
-        tokenizer.add_special_tokens_single_sequence(tokenizer.encode(line))
+        tokenizer.build_inputs_with_special_tokens(tokenizer.encode(line))
         for line in summary_lines
     ]
 

From 1c71ecc880ae8f04c8462e1368dc0678fdb92fc9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Thu, 31 Oct 2019 10:16:08 +0100
Subject: [PATCH 266/505] load the pretrained weights for encoder-decoder

We currently save the pretrained_weights of the encoder and decoder in
two separate directories `encoder` and `decoder`. However, for the
`from_pretrained` function to operate with automodels we need to
specify the type of model in the path to the weights.

The path to the encoder/decoder weights is handled by the
`PreTrainedEncoderDecoder` class in the `save_pretrained` function. Sice
there is no easy way to infer the type of model that was initialized for
the encoder and decoder we add a parameter `model_type` to the function.
This is not an ideal solution as it is error prone, and the model type
should be carried by the Model classes somehow.

This is a temporary fix that should be changed before merging.
---
 examples/run_summarization_finetuning.py | 48 ++++++++++++++----------
 transformers/modeling_encoder_decoder.py | 31 +++++++++------
 2 files changed, 49 insertions(+), 30 deletions(-)

diff --git a/examples/run_summarization_finetuning.py b/examples/run_summarization_finetuning.py
index f5604c2669..9c2c7769c9 100644
--- a/examples/run_summarization_finetuning.py
+++ b/examples/run_summarization_finetuning.py
@@ -328,6 +328,22 @@ def evaluate(args, model, tokenizer, prefix=""):
     return result
 
 
+def save_model_checkpoints(args, model, tokenizer):
+    if not os.path.exists(args.output_dir):
+        os.makedirs(args.output_dir)
+
+    logger.info("Saving model checkpoint to %s", args.output_dir)
+
+    # Save a trained model, configuration and tokenizer using `save_pretrained()`.
+    # They can then be reloaded using `from_pretrained()`
+    model_to_save = (
+        model.module if hasattr(model, "module") else model
+    )  # Take care of distributed/parallel training
+    model_to_save.save_pretrained(args.output_dir, model_type='bert')
+    tokenizer.save_pretrained(args.output_dir)
+    torch.save(args, os.path.join(args.output_dir, "training_arguments.bin"))
+
+
 def main():
     parser = argparse.ArgumentParser()
 
@@ -454,36 +470,30 @@ def main():
     # Train the model
     model.to(args.device)
     if args.do_train:
-        global_step, tr_loss = train(args, model, tokenizer)
+        try:
+            global_step, tr_loss = train(args, model, tokenizer)
+        except KeyboardInterrupt:
+            response = input("You interrupted the training. Do you want to save the model checkpoints? [Y/n]")
+            if response.lower() in ["", "y", "yes"]:
+                save_model_checkpoints(args, model, tokenizer)
+            sys.exit(0)
+
         logger.info(" global_step = %s, average loss = %s", global_step, tr_loss)
-
-        if not os.path.exists(args.output_dir):
-            os.makedirs(args.output_dir)
-
-        logger.info("Saving model checkpoint to %s", args.output_dir)
-
-        # Save a trained model, configuration and tokenizer using `save_pretrained()`.
-        # They can then be reloaded using `from_pretrained()`
-        model_to_save = (
-            model.module if hasattr(model, "module") else model
-        )  # Take care of distributed/parallel training
-        model_to_save.save_pretrained(args.output_dir)
-        tokenizer.save_pretrained(args.output_dir)
-        torch.save(args, os.path.join(args.output_dir, "training_arguments.bin"))
+        save_model_checkpoints(args, model, tokenizer)
 
     # Evaluate the model
     results = {}
     if args.do_evaluate:
-        checkpoints = []
+        checkpoints = [args.output_dir]
         logger.info("Evaluate the following checkpoints: %s", checkpoints)
         for checkpoint in checkpoints:
-            encoder_checkpoint = os.path.join(checkpoint, "encoder")
-            decoder_checkpoint = os.path.join(checkpoint, "decoder")
+            encoder_checkpoint = os.path.join(checkpoint, "bert_encoder")
+            decoder_checkpoint = os.path.join(checkpoint, "bert_decoder")
             model = PreTrainedEncoderDecoder.from_pretrained(
                 encoder_checkpoint, decoder_checkpoint
             )
             model.to(args.device)
-            results = "placeholder"
+            print("model loaded")
 
     return results
 
diff --git a/transformers/modeling_encoder_decoder.py b/transformers/modeling_encoder_decoder.py
index a884abd0a2..73322101d3 100644
--- a/transformers/modeling_encoder_decoder.py
+++ b/transformers/modeling_encoder_decoder.py
@@ -117,8 +117,7 @@ class PreTrainedEncoderDecoder(nn.Module):
         kwargs_common = {
             argument: value
             for argument, value in kwargs.items()
-            if not argument.startswith("encoder_")
-            and not argument.startswith("decoder_")
+            if not argument.startswith("encoder_") and not argument.startswith("decoder_")
         }
         kwargs_decoder = kwargs_common.copy()
         kwargs_encoder = kwargs_common.copy()
@@ -158,14 +157,27 @@ class PreTrainedEncoderDecoder(nn.Module):
 
         return model
 
-    def save_pretrained(self, save_directory):
-        """ Save a Seq2Seq model and its configuration file in a format such
+    def save_pretrained(self, save_directory, model_type="bert"):
+        """ Save an EncoderDecoder model and its configuration file in a format such
         that it can be loaded using `:func:`~transformers.PreTrainedEncoderDecoder.from_pretrained`
 
         We save the encoder' and decoder's parameters in two separate directories.
+
+        If we want the weight loader to function we need to preprend the model
+        type to the directories' names. As far as I know there is no simple way
+        to infer the type of the model (except maybe by parsing the class'
+        names, which is not very future-proof). For now, we ask the user to
+        specify the model type explicitly when saving the weights.
         """
-        self.encoder.save_pretrained(os.path.join(save_directory, "encoder"))
-        self.decoder.save_pretrained(os.path.join(save_directory, "decoder"))
+        encoder_path = os.path.join(save_directory, "{}_encoder".format(model_type))
+        if not os.path.exists(encoder_path):
+            os.makedirs(encoder_path)
+        self.encoder.save_pretrained(encoder_path)
+
+        decoder_path = os.path.join(save_directory, "{}_decoder".format(model_type))
+        if not os.path.exists(decoder_path):
+            os.makedirs(decoder_path)
+        self.decoder.save_pretrained(decoder_path)
 
     def forward(self, encoder_input_ids, decoder_input_ids, **kwargs):
         """ The forward pass on a seq2eq depends what we are performing:
@@ -193,8 +205,7 @@ class PreTrainedEncoderDecoder(nn.Module):
         kwargs_common = {
             argument: value
             for argument, value in kwargs.items()
-            if not argument.startswith("encoder_")
-            and not argument.startswith("decoder_")
+            if not argument.startswith("encoder_") and not argument.startswith("decoder_")
         }
         kwargs_decoder = kwargs_common.copy()
         kwargs_encoder = kwargs_common.copy()
@@ -217,9 +228,7 @@ class PreTrainedEncoderDecoder(nn.Module):
         encoder_hidden_states = kwargs_encoder.pop("hidden_states", None)
         if encoder_hidden_states is None:
             encoder_outputs = self.encoder(encoder_input_ids, **kwargs_encoder)
-            encoder_hidden_states = encoder_outputs[
-                0
-            ]  # output the last layer hidden state
+            encoder_hidden_states = encoder_outputs[0]  # output the last layer hidden state
         else:
             encoder_outputs = ()
 

From 9660ba1cbdec0e419937af06bd99f06fb5ebbf91 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Thu, 31 Oct 2019 17:59:16 +0100
Subject: [PATCH 267/505] Add beam search

---
 examples/run_summarization_finetuning.py | 502 -----------------------
 examples/utils_summarization.py          |  20 +-
 transformers/generate/__init__.py        |   1 +
 transformers/generate/beam_search.py     | 358 ++++++++++++++++
 transformers/modeling_beam_search.py     | 271 ------------
 transformers/tests/beam_search_tests.py  | 226 ++++++++++
 6 files changed, 594 insertions(+), 784 deletions(-)
 delete mode 100644 examples/run_summarization_finetuning.py
 create mode 100644 transformers/generate/__init__.py
 create mode 100644 transformers/generate/beam_search.py
 delete mode 100644 transformers/modeling_beam_search.py
 create mode 100644 transformers/tests/beam_search_tests.py

diff --git a/examples/run_summarization_finetuning.py b/examples/run_summarization_finetuning.py
deleted file mode 100644
index 9c2c7769c9..0000000000
--- a/examples/run_summarization_finetuning.py
+++ /dev/null
@@ -1,502 +0,0 @@
-# coding=utf-8
-# Copyright 2019 The HuggingFace Inc. team.
-# Copyright (c) 2019 The HuggingFace Inc.  All rights reserved.
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-""" Finetuning seq2seq models for sequence generation."""
-
-import argparse
-import functools
-import logging
-import os
-import random
-import sys
-
-import numpy as np
-from tqdm import tqdm, trange
-import torch
-from torch.optim import Adam
-from torch.utils.data import DataLoader, RandomSampler, SequentialSampler
-
-from transformers import (
-    AutoTokenizer,
-    BertForMaskedLM,
-    BertConfig,
-    PreTrainedEncoderDecoder,
-    Model2Model,
-)
-
-from utils_summarization import (
-    CNNDailyMailDataset,
-    encode_for_summarization,
-    fit_to_block_size,
-    build_lm_labels,
-    build_mask,
-    compute_token_type_ids,
-)
-
-logger = logging.getLogger(__name__)
-logging.basicConfig(stream=sys.stdout, level=logging.INFO)
-
-
-def set_seed(args):
-    random.seed(args.seed)
-    np.random.seed(args.seed)
-    torch.manual_seed(args.seed)
-
-
-# ------------
-# Load dataset
-# ------------
-
-
-def load_and_cache_examples(args, tokenizer):
-    dataset = CNNDailyMailDataset(tokenizer, data_dir=args.data_dir)
-    return dataset
-
-
-def collate(data, tokenizer, block_size):
-    """ List of tuple as an input. """
-    # remove the files with empty an story/summary, encode and fit to block
-    data = filter(lambda x: not (len(x[0]) == 0 or len(x[1]) == 0), data)
-    data = [
-        encode_for_summarization(story, summary, tokenizer) for story, summary in data
-    ]
-    data = [
-        (
-            fit_to_block_size(story, block_size, tokenizer.pad_token_id),
-            fit_to_block_size(summary, block_size, tokenizer.pad_token_id),
-        )
-        for story, summary in data
-    ]
-
-    stories = torch.tensor([story for story, summary in data])
-    summaries = torch.tensor([summary for story, summary in data])
-    encoder_token_type_ids = compute_token_type_ids(stories, tokenizer.cls_token_id)
-    encoder_mask = build_mask(stories, tokenizer.pad_token_id)
-    decoder_mask = build_mask(summaries, tokenizer.pad_token_id)
-    lm_labels = build_lm_labels(summaries, tokenizer.pad_token_id)
-
-    return (
-        stories,
-        summaries,
-        encoder_token_type_ids,
-        encoder_mask,
-        decoder_mask,
-        lm_labels,
-    )
-
-
-# ----------
-# Optimizers
-# ----------
-
-
-class BertSumOptimizer(object):
-    """ Specific optimizer for BertSum.
-
-    As described in [1], the authors fine-tune BertSum for abstractive
-    summarization using two Adam Optimizers with different warm-up steps and
-    learning rate. They also use a custom learning rate scheduler.
-
-    [1] Liu, Yang, and Mirella Lapata. "Text summarization with pretrained encoders."
-        arXiv preprint arXiv:1908.08345 (2019).
-    """
-
-    def __init__(self, model, lr, warmup_steps, beta_1=0.99, beta_2=0.999, eps=1e-8):
-        self.encoder = model.encoder
-        self.decoder = model.decoder
-        self.lr = lr
-        self.warmup_steps = warmup_steps
-
-        self.optimizers = {
-            "encoder": Adam(
-                model.encoder.parameters(),
-                lr=lr["encoder"],
-                betas=(beta_1, beta_2),
-                eps=eps,
-            ),
-            "decoder": Adam(
-                model.decoder.parameters(),
-                lr=lr["decoder"],
-                betas=(beta_1, beta_2),
-                eps=eps,
-            ),
-        }
-
-        self._step = 0
-
-    def _update_rate(self, stack):
-        return self.lr[stack] * min(
-            self._step ** (-0.5), self._step * self.warmup_steps[stack] ** (-0.5)
-        )
-
-    def zero_grad(self):
-        self.optimizer_decoder.zero_grad()
-        self.optimizer_encoder.zero_grad()
-
-    def step(self):
-        self._step += 1
-        for stack, optimizer in self.optimizers.items():
-            new_rate = self._update_rate(stack)
-            for param_group in optimizer.param_groups:
-                param_group["lr"] = new_rate
-            optimizer.step()
-
-
-# ------------
-# Train
-# ------------
-
-
-def train(args, model, tokenizer):
-    """ Fine-tune the pretrained model on the corpus. """
-    set_seed(args)
-
-    # Load the data
-    args.train_batch_size = args.per_gpu_train_batch_size * max(1, args.n_gpu)
-    train_dataset = load_and_cache_examples(args, tokenizer)
-    train_sampler = RandomSampler(train_dataset)
-    model_collate_fn = functools.partial(collate, tokenizer=tokenizer, block_size=512)
-    train_dataloader = DataLoader(
-        train_dataset,
-        sampler=train_sampler,
-        batch_size=args.train_batch_size,
-        collate_fn=model_collate_fn,
-    )
-
-    # Training schedule
-    if args.max_steps > 0:
-        t_total = args.max_steps
-        args.num_train_epochs = t_total // (
-            len(train_dataloader) // args.gradient_accumulation_steps + 1
-        )
-    else:
-        t_total = (
-            len(train_dataloader)
-            // args.gradient_accumulation_steps
-            * args.num_train_epochs
-        )
-
-    # Prepare the optimizer
-    lr = {"encoder": 0.002, "decoder": 0.2}
-    warmup_steps = {"encoder": 20000, "decoder": 10000}
-    optimizer = BertSumOptimizer(model, lr, warmup_steps)
-
-    # Train
-    logger.info("***** Running training *****")
-    logger.info("  Num examples = %d", len(train_dataset))
-    logger.info("  Num Epochs = %d", args.num_train_epochs)
-    logger.info(
-        "  Instantaneous batch size per GPU = %d", args.per_gpu_train_batch_size
-    )
-    logger.info(
-        "  Total train batch size (w. parallel, distributed & accumulation) = %d",
-        args.train_batch_size * args.gradient_accumulation_steps
-        # * (torch.distributed.get_world_size() if args.local_rank != -1 else 1),
-    )
-    logger.info("  Gradient Accumulation steps = %d", args.gradient_accumulation_steps)
-    logger.info("  Total optimization steps = %d", t_total)
-
-    model.zero_grad()
-    train_iterator = trange(args.num_train_epochs, desc="Epoch", disable=True)
-
-    global_step = 0
-    tr_loss = 0.0
-    for _ in train_iterator:
-        epoch_iterator = tqdm(train_dataloader, desc="Iteration", disable=True)
-        for step, batch in enumerate(epoch_iterator):
-            source, target, encoder_token_type_ids, encoder_mask, decoder_mask, lm_labels = batch
-
-            source = source.to(args.device)
-            target = target.to(args.device)
-            encoder_token_type_ids = encoder_token_type_ids.to(args.device)
-            encoder_mask = encoder_mask.to(args.device)
-            decoder_mask = decoder_mask.to(args.device)
-            lm_labels = lm_labels.to(args.device)
-
-            model.train()
-            outputs = model(
-                source,
-                target,
-                encoder_token_type_ids=encoder_token_type_ids,
-                encoder_attention_mask=encoder_mask,
-                decoder_attention_mask=decoder_mask,
-                decoder_lm_labels=lm_labels,
-            )
-
-            loss = outputs[0]
-            print(loss)
-            if args.gradient_accumulation_steps > 1:
-                loss /= args.gradient_accumulation_steps
-
-            loss.backward()
-
-            tr_loss += loss.item()
-            if (step + 1) % args.gradient_accumulation_steps == 0:
-                torch.nn.utils.clip_grad_norm_(model.parameters(), args.max_grad_norm)
-                optimizer.step()
-                model.zero_grad()
-                global_step += 1
-
-            if args.max_steps > 0 and global_step > args.max_steps:
-                epoch_iterator.close()
-                break
-
-        if args.max_steps > 0 and global_step > args.max_steps:
-            train_iterator.close()
-            break
-
-    return global_step, tr_loss / global_step
-
-
-# ------------
-# Train
-# ------------
-
-
-def evaluate(args, model, tokenizer, prefix=""):
-    set_seed(args)
-
-    args.eval_batch_size = args.per_gpu_eval_batch_size * max(1, args.n_gpu)
-    eval_dataset = load_and_cache_examples(args, tokenizer, evaluate=True)
-    eval_sampler = SequentialSampler(eval_dataset)
-    eval_dataloader = DataLoader(
-        eval_dataset, sampler=eval_sampler, batch_size=args.eval_batch_size
-    )
-
-    # multi-gpu evaluate
-    if args.n_gpu > 1:
-        model = torch.nn.DataParallel(model)
-        
-    logger.info("***** Running evaluation {} *****".format(prefix))
-    logger.info("  Num examples = %d", len(eval_dataset))
-    logger.info("  Batch size = %d", args.eval_batch_size)
-    eval_loss = 0.0
-    nb_eval_steps = 0
-    model.eval()
-
-    for batch in tqdm(eval_dataloader, desc="Evaluating"):
-        source, target, encoder_token_type_ids, encoder_mask, decoder_mask, lm_labels = batch
-
-        source = source.to(args.device)
-        target = target.to(args.device)
-        encoder_token_type_ids = encoder_token_type_ids.to(args.device)
-        encoder_mask = encoder_mask.to(args.device)
-        decoder_mask = decoder_mask.to(args.device)
-        lm_labels = lm_labels.to(args.device)
-
-        with torch.no_grad():
-            outputs = model(
-                source,
-                target,
-                encoder_token_type_ids=encoder_token_type_ids,
-                encoder_attention_mask=encoder_mask,
-                decoder_attention_mask=decoder_mask,
-                decoder_lm_labels=lm_labels,
-            )
-            lm_loss = outputs[0]
-            eval_loss += lm_loss.mean().item()
-        nb_eval_steps += 1
-
-    eval_loss = eval_loss / nb_eval_steps
-    perplexity = torch.exp(torch.tensor(eval_loss))
-
-    result = {"perplexity": perplexity}
-
-    # Save the evaluation's results
-    output_eval_file = os.path.join(args.output_dir, "eval_results.txt")
-    if not os.path.exists(args.output_dir):
-        os.makedirs(args.output_dir)
-
-    with open(output_eval_file, "w") as writer:
-        logger.info("***** Eval results {} *****".format(prefix))
-        for key in sorted(result.keys()):
-            logger.info("  %s = %s", key, str(result[key]))
-            writer.write("%s = %s\n" % (key, str(result[key])))
-
-    return result
-
-
-def save_model_checkpoints(args, model, tokenizer):
-    if not os.path.exists(args.output_dir):
-        os.makedirs(args.output_dir)
-
-    logger.info("Saving model checkpoint to %s", args.output_dir)
-
-    # Save a trained model, configuration and tokenizer using `save_pretrained()`.
-    # They can then be reloaded using `from_pretrained()`
-    model_to_save = (
-        model.module if hasattr(model, "module") else model
-    )  # Take care of distributed/parallel training
-    model_to_save.save_pretrained(args.output_dir, model_type='bert')
-    tokenizer.save_pretrained(args.output_dir)
-    torch.save(args, os.path.join(args.output_dir, "training_arguments.bin"))
-
-
-def main():
-    parser = argparse.ArgumentParser()
-
-    # Required parameters
-    parser.add_argument(
-        "--data_dir",
-        default=None,
-        type=str,
-        required=True,
-        help="The input training data file (a text file).",
-    )
-    parser.add_argument(
-        "--output_dir",
-        default=None,
-        type=str,
-        required=True,
-        help="The output directory where the model predictions and checkpoints will be written.",
-    )
-
-    # Optional parameters
-    parser.add_argument(
-        "--gradient_accumulation_steps",
-        type=int,
-        default=1,
-        help="Number of updates steps to accumulate before performing a backward/update pass.",
-    )
-    parser.add_argument(
-        "--do_evaluate",
-        type=bool,
-        default=False,
-        help="Run model evaluation on out-of-sample data.",
-    )
-    parser.add_argument("--do_train", type=bool, default=False, help="Run training.")
-    parser.add_argument(
-        "--do_overwrite_output_dir",
-        type=bool,
-        default=False,
-        help="Whether to overwrite the output dir.",
-    )
-    parser.add_argument(
-        "--model_name_or_path",
-        default="bert-base-cased",
-        type=str,
-        help="The model checkpoint to initialize the encoder and decoder's weights with.",
-    )
-    parser.add_argument(
-        "--model_type",
-        default="bert",
-        type=str,
-        help="The decoder architecture to be fine-tuned.",
-    )
-    parser.add_argument(
-        "--max_grad_norm", default=1.0, type=float, help="Max gradient norm."
-    )
-    parser.add_argument(
-        "--max_steps",
-        default=-1,
-        type=int,
-        help="If > 0: set total number of training steps to perform. Override num_train_epochs.",
-    )
-    parser.add_argument(
-        "--to_cpu", default=False, type=bool, help="Whether to force training on CPU."
-    )
-    parser.add_argument(
-        "--num_train_epochs",
-        default=10,
-        type=int,
-        help="Total number of training epochs to perform.",
-    )
-    parser.add_argument(
-        "--per_gpu_train_batch_size",
-        default=4,
-        type=int,
-        help="Batch size per GPU/CPU for training.",
-    )
-    parser.add_argument("--seed", default=42, type=int)
-    args = parser.parse_args()
-
-    if (
-        os.path.exists(args.output_dir)
-        and os.listdir(args.output_dir)
-        and args.do_train
-        and not args.do_overwrite_output_dir
-    ):
-        raise ValueError(
-            "Output directory ({}) already exists and is not empty. Use --do_overwrite_output_dir to overwrite.".format(
-                args.output_dir
-            )
-        )
-
-    # Set up training device
-    if args.to_cpu or not torch.cuda.is_available():
-        args.device = torch.device("cpu")
-        args.n_gpu = 0
-    else:
-        args.device = torch.device("cuda")
-        args.n_gpu = torch.cuda.device_count()
-
-    # Load pretrained model and tokenizer. The decoder's weights are randomly initialized.
-    tokenizer = AutoTokenizer.from_pretrained(args.model_name_or_path)
-    config = BertConfig.from_pretrained(args.model_name_or_path)
-    decoder_model = BertForMaskedLM(config)
-    model = Model2Model.from_pretrained(
-        args.model_name_or_path, decoder_model=decoder_model
-    )
-
-    # Setup logging
-    logging.basicConfig(
-        format="%(asctime)s - %(levelname)s - %(name)s -   %(message)s",
-        datefmt="%m/%d/%Y %H:%M:%S",
-        level=logging.INFO,
-    )
-    logger.warning(
-        "Process rank: %s, device: %s, n_gpu: %s, distributed training: %s, 16-bits training: %s",
-        0,
-        args.device,
-        args.n_gpu,
-        False,
-        False,
-    )
-
-    logger.info("Training/evaluation parameters %s", args)
-
-    # Train the model
-    model.to(args.device)
-    if args.do_train:
-        try:
-            global_step, tr_loss = train(args, model, tokenizer)
-        except KeyboardInterrupt:
-            response = input("You interrupted the training. Do you want to save the model checkpoints? [Y/n]")
-            if response.lower() in ["", "y", "yes"]:
-                save_model_checkpoints(args, model, tokenizer)
-            sys.exit(0)
-
-        logger.info(" global_step = %s, average loss = %s", global_step, tr_loss)
-        save_model_checkpoints(args, model, tokenizer)
-
-    # Evaluate the model
-    results = {}
-    if args.do_evaluate:
-        checkpoints = [args.output_dir]
-        logger.info("Evaluate the following checkpoints: %s", checkpoints)
-        for checkpoint in checkpoints:
-            encoder_checkpoint = os.path.join(checkpoint, "bert_encoder")
-            decoder_checkpoint = os.path.join(checkpoint, "bert_decoder")
-            model = PreTrainedEncoderDecoder.from_pretrained(
-                encoder_checkpoint, decoder_checkpoint
-            )
-            model.to(args.device)
-            print("model loaded")
-
-    return results
-
-
-if __name__ == "__main__":
-    main()
diff --git a/examples/utils_summarization.py b/examples/utils_summarization.py
index 087c88bd4e..7cbd4cd61b 100644
--- a/examples/utils_summarization.py
+++ b/examples/utils_summarization.py
@@ -25,9 +25,8 @@ class CNNDailyMailDataset(Dataset):
     [2] https://github.com/abisee/cnn-dailymail/
     """
 
-    def __init__(self, tokenizer, prefix="train", data_dir=""):
+    def __init__(self, data_dir="", prefix="train"):
         assert os.path.isdir(data_dir)
-        self.tokenizer = tokenizer
 
         # We initialize the class by listing all the files that contain
         # stories and summaries. Files are not read in memory given
@@ -104,31 +103,30 @@ def _add_missing_period(line):
 # --------------------------
 
 
-def fit_to_block_size(sequence, block_size, pad_token):
+def fit_to_block_size(sequence, block_size, pad_token_id):
     """ Adapt the source and target sequences' lengths to the block size.
-    If the sequence is shorter than the block size we pad it with -1 ids
-    which correspond to padding tokens.
+    If the sequence is shorter we append padding token to the right of the sequence.
     """
     if len(sequence) > block_size:
         return sequence[:block_size]
     else:
-        sequence.extend([pad_token] * (block_size - len(sequence)))
+        sequence.extend([pad_token_id] * (block_size - len(sequence)))
         return sequence
 
 
-def build_lm_labels(sequence, pad_token):
-    """ Padding token, encoded as 0, are represented by the value -1 so they
+def build_lm_labels(sequence, pad_token_id):
+    """ Padding token are replaced by the value -1 so they
     are not taken into account in the loss computation. """
     padded = sequence.clone()
-    padded[padded == pad_token] = -1
+    padded[padded == pad_token_id] = -1
     return padded
 
 
-def build_mask(sequence, pad_token):
+def build_mask(sequence, pad_token_id):
     """ Builds the mask. The attention mechanism will only attend to positions
     with value 1. """
     mask = torch.ones_like(sequence)
-    idx_pad_tokens = sequence == pad_token
+    idx_pad_tokens = sequence == pad_token_id
     mask[idx_pad_tokens] = 0
     return mask
 
diff --git a/transformers/generate/__init__.py b/transformers/generate/__init__.py
new file mode 100644
index 0000000000..21ac612155
--- /dev/null
+++ b/transformers/generate/__init__.py
@@ -0,0 +1 @@
+from .beam_search import BeamSearch
diff --git a/transformers/generate/beam_search.py b/transformers/generate/beam_search.py
new file mode 100644
index 0000000000..09e340a150
--- /dev/null
+++ b/transformers/generate/beam_search.py
@@ -0,0 +1,358 @@
+# coding=utf-8
+# MIT License
+
+# Copyright (c) 2017-Present OpenNMT
+
+# Permission is hereby granted, free of charge, to any person obtaining a copy of
+# this software and associated documentation files (the "Software"), to deal in
+# the Software without restriction, including without limitation the rights to
+# use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+# of the Software, and to permit persons to whom the Software is furnished to do
+# so, subject to the following conditions:
+
+# The above copyright notice and this permission notice shall be included in all
+# copies or substantial portions of the Software.
+
+# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+# SOFTWARE.
+"""
+Use Beam Search to generate sequences using encoder-decoder models.
+"""
+import torch
+from torch import nn
+
+
+class BeamSearch(nn.Module):
+    def __init__(
+        self,
+        model,
+        tokenizer,
+        beam_size,
+        min_length,
+        max_length,
+        batch_size=1,
+        alpha=0,
+        block_repeating_trigrams=True,
+    ):
+        r"""
+        Inputs:
+            **model**: instance of ``transformers.PreTrainedEncoderDecoder``
+                The pretrained encoder-decoder model that will be used to generate the sequences.
+            **tokenizer**: instance of ``transformers.PreTrainedTokenizer``
+                The pretrained tokenizer associated to the model used in the encoder-decoder. We only
+                support encoder-decoder that use the same tokenizer for encoder and decoder. The tokenizer
+                needs to be initialized or this function will raise and exception.
+            **batch_size**: (`optional`) int
+                Batch size of the inputs. The value is set automatically when calling `forward`.
+            **beam_size**: int
+                Number of beams that are used for each element on the batch.
+            **min_length**: int
+                Minimum number of steps performed by the beam search before terminating.
+            **max_length**: int
+                Maximum number of steps performed by the beam search. Any beam that has not finished
+                will return its current solution with the highest probability. The sequence that is
+                returned has a length of max_length-1 to account for the end token that is subsequently added.
+            **alpha**: float
+                Parameter of the length penalty. Read the documentation of the `_length_penalty` method for mode details.
+            **block_repeating_trigrams**: bool
+                Whether to block sequences that have repeating 3-grams.
+        """
+        super(BeamSearch, self).__init__()
+        self.model = model
+        self.tokenizer = tokenizer
+
+        self.bos_token_id = tokenizer.bos_token_id
+        self.eos_token_id = tokenizer.eos_token_id
+        self.pad_token_id = tokenizer.pad_token_id
+
+        self.batch_size = batch_size
+        self.beam_size = beam_size
+        self.min_length = min_length
+        self.max_length = max_length
+
+        self.block_repeating_trigram = block_repeating_trigrams
+        self.apply_length_penalty = False if alpha == 0 else True
+        self.alpha = alpha
+
+        self._init_beam_state(batch_size)
+
+    def __len__(self):
+        try:
+            return self.growing_beams.size(1)
+        except NameError:
+            return 0
+
+    def _init_beam_state(self, batch_size):
+        """ (re-)Initialize the state of the beams. """
+        self.hypotheses = [[] for _ in range(batch_size)]
+        self.batch_offset = torch.arange(batch_size, dtype=torch.long)
+        self.beam_offset = torch.arange(
+            0, batch_size * self.beam_size, step=self.beam_size, dtype=torch.long
+        )
+        self.growing_beams = torch.full(
+            (batch_size * self.beam_size, 1), self.bos_token_id, dtype=torch.long
+        )
+        self.topk_log_probabilities = torch.tensor(
+            [0.0] + [float("-inf")] * (self.beam_size - 1), dtype=torch.float
+        ).repeat(batch_size)
+        self.results = {
+            "predictions": [[] for _ in range(batch_size)],
+            "scores": [[] for _ in range(batch_size)],
+        }
+        self._step = 0
+        self.is_done = False
+
+    def forward(self, encoder_input_ids, **model_kwargs):
+        """ Generate a sequence using Beam Search. """
+        # keyword arguments come in 3 flavors: encoder-specific (prefixed by
+        # `encoder_`), decoder-specific (prefixed by `decoder_`) and those
+        # that apply to the model as whole.
+        # We let the specific kwargs override the common ones in case of conflict.
+        kwargs_common = {
+            argument: value
+            for argument, value in model_kwargs.items()
+            if not argument.startswith("encoder_") and not argument.startswith("decoder_")
+        }
+        kwargs_decoder = kwargs_common.copy()
+        kwargs_encoder = kwargs_common.copy()
+        kwargs_encoder.update(
+            {
+                argument[len("encoder_") :]: value
+                for argument, value in model_kwargs.items()
+                if argument.startswith("encoder_")
+            }
+        )
+        kwargs_decoder.update(
+            {
+                argument[len("decoder_") :]: value
+                for argument, value in model_kwargs.items()
+                if argument.startswith("decoder_")
+            }
+        )
+
+        # forward pass on the encoder
+        encoder_outputs = self.model.encoder.forward(encoder_input_ids, kwargs_encoder)
+        kwargs_decoder["encoder_hidden_states"] = tile(
+            encoder_outputs, self.beam_size, dim=0
+        )
+
+        # grow the beam by generating sequences in an autoregressive way
+        batch_size = encoder_input_ids.size(0)
+        self._init_beam_state(batch_size)
+        for step in range(self.max_length):
+            # prepare the decoder input
+            decoder_input = fit_to_block_size(
+                self.growing_beams, self.tokenizer.pad_token_id
+            )
+            kwargs_decoder["decoder_lm_labels"] = build_lm_labels(
+                decoder_input, self.tokenizer.pad_token_id
+            )
+            kwargs_decoder["decoder_attention_mask"] = build_mask(
+                decoder_input, self.tokenizer.pad_token_id
+            )
+
+            outputs = self.model.decoder(decoder_input, kwargs_decoder)
+            log_probabilities = torch.nn.functional.log_softmax(outputs[1])
+            surviving_beams_rows = self.grow(log_probabilities)
+            if self.is_done:
+                break
+
+            kwargs_decoder["encoder_hidden_states"] = kwargs_decoder[
+                "encoder_hidden_states"
+            ].index_select(0, surviving_beams_rows)
+            kwargs_decoder["encoder_attention_mask"] = kwargs_decoder[
+                "encoder_attention_mask"
+            ].index_select(0, surviving_beams_rows)
+
+        return self.results
+
+    def grow(self, log_probabilities):
+        """ Grow the beams by one step. """
+        self._step += 1
+
+        # The number of beams changes as some beams finish so we define _B
+        vocab_size = log_probabilities.size(-1)
+        _B = log_probabilities.size(0) // self.beam_size
+
+        # Multiply each beam probability with the probability of the
+        # next token (conditioned on the words in the beam).
+        log_probabilities += self.topk_log_probabilities.view(-1, 1)
+
+        self._enforce_min_length(log_probabilities)
+        if self.block_repeating_trigram:
+            self._remove_beams_with_repeating_trigrams(log_probabilities, _B)
+
+        # Find the `beam_size` (previous_beam + token) combinations with
+        # the highest score
+        topk_log_probabilities, topk_ids = torch.topk(
+            log_probabilities.view(_B, self.beam_size * vocab_size), self.beam_size, dim=1
+        )
+
+        # Apply the length penalty. The +1 accounts for the [EOS] token
+        # that will be added if the beam ends.
+        topk_scores = topk_log_probabilities
+        if self.apply_length_penalty:
+            topk_scores /= self._length_penalty()
+
+        # Retrieve the corresponding respective beam and token id
+        # topk_token_ids[i] will be added to topk_beam_ids[i]
+        topk_beam_ids = topk_ids.div(vocab_size)
+        topk_token_ids = topk_ids.fmod(vocab_size)
+
+        # Retrieve the row index of the surviving beams in the original
+        # view of the log_probabilities tensor
+        surviving_beams_per_batch = topk_beam_ids + self.beam_offset[:_B].view(-1, 1)
+        surviving_beams_rows = surviving_beams_per_batch.view(-1)
+
+        # Append the last predictions
+        self.growing_beams = torch.cat(
+            [
+                self.growing_beams.index_select(0, surviving_beams_rows),
+                topk_token_ids.view(-1, 1),
+            ],
+            1,
+        )
+
+        # Check if any of the beam searches has ended during this
+        # growth step. Also if top beam (most probable) has ended
+        # for one element of the batch.
+        is_finished = topk_token_ids.eq(self.eos_token_id)
+        self._enforce_max_length(is_finished)
+        if is_finished.any():
+            non_finished = self._cut_finished(is_finished, topk_scores)
+            self.batch_offset = self.batch_offset.index_select(0, non_finished)
+            surviving_beams_per_batch = surviving_beams_per_batch.index_select(
+                0, non_finished
+            )
+            self.topk_log_probabilities = self.topk_log_probabilities.index_select(
+                0, non_finished
+            )
+
+            surviving_beams_rows = surviving_beams_per_batch.view(-1)
+            self.growing_beams = self.growing_beams.index_select(0, surviving_beams_rows)
+
+        return surviving_beams_rows
+
+    def _cut_finished(self, is_finished, topk_scores):
+        """ Save the finished searches and cut the correponding sequences off
+        the beams. """
+        is_top_beam_finished = is_finished[:, 0].eq(True)
+
+        # Save the finished searches
+        predictions = self.growing_beams.view(
+            -1, self.beam_size, self.growing_beams.size(1)
+        )
+        for i in range(is_finished.size(0)):
+            if is_top_beam_finished[i]:
+                is_finished[i].fill_(1)
+            finished_hyp = is_finished[i].nonzero().view(-1)
+
+            # Store the finished beams as a (score, prediction) hypothesis.
+            b = self.batch_offset[i]
+            for j in finished_hyp:
+                self.hypotheses[b].append((topk_scores[i, j], predictions[i, j, :]))
+
+            # If the batch reached the end, save the best hypotheses
+            # in terms of length-penalized score.
+            if is_top_beam_finished[i]:
+                best_score, best_prediction = max(self.hypotheses[b], key=lambda x: x[0])
+                self.results["scores"][b].append(best_score)
+                self.results["predictions"][b].append(best_prediction)
+
+        non_finished = is_top_beam_finished.eq(False).nonzero().view(-1)
+        if len(non_finished) == 0:
+            self.is_done = True
+
+        return non_finished
+
+    def _remove_beams_with_repeating_trigrams(self, log_probabilities, _B):
+        if self._step + 1 > 3:  # [BOS] does not count
+            for i in range(_B * self.beam_size):
+                tokens = self.growing_beams[i]
+                trigrams = [
+                    (tokens[j - 1], tokens[j], tokens[j + 1])
+                    for j in range(1, len(self) - 1)
+                ]
+                last_trigram = tuple(trigrams[-1])
+                if last_trigram in trigrams[:-1]:
+                    log_probabilities[i] = -1e20
+
+    def _enforce_min_length(self, log_probabilities):
+        if self._step < self.min_length:
+            log_probabilities[:, self.eos_token_id] = -1e20
+
+    def _enforce_max_length(self, is_finished):
+        # +1 because we will need to add an [EOS] token
+        if self._step + 1 == self.max_length:
+            is_finished.fill_(1)
+
+    def _length_penalty(self):
+        """ The calculation of the length penalty follows that of [1].
+
+        [1] Wu, Yonghui, et al. "Google's neural machine translation system:
+        Bridging the gap between human and machine translation." arXiv preprint
+        arXiv:1609.08144 (2016).
+        """
+        return ((5.0 + (self._step + 1)) / 6.0) ** self.alpha
+
+
+def tile(x, count, dim=0):
+    """
+    Tiles `x` along dimension `dim` `count` times.
+
+    Example:
+        >> ex = torch.tensor([1,2],[3,4])
+        >> tile(ex, 2, 0)
+        torch.Tensor([[1,2],[1,2],[3,4],[3,4]])
+    """
+    perm = list(range(len(x.size())))
+    if dim != 0:
+        perm[0], perm[dim] = perm[dim], perm[0]
+        x = x.permute(perm).contiguous()
+    out_size = list(x.size())
+    out_size[0] *= count
+    batch = x.size(0)
+    x = (
+        x.view(batch, -1)
+        .transpose(0, 1)
+        .repeat(count, 1)
+        .transpose(0, 1)
+        .contiguous()
+        .view(*out_size)
+    )
+    if dim != 0:
+        x = x.permute(perm).contiguous()
+    return x
+
+
+def fit_to_block_size(sequence, block_size, pad_token_id):
+    """ Adapt the source and target sequences' lengths to the block size.
+    If the sequence is shorter we append padding tokens to the right.
+    """
+    if len(sequence) > block_size:
+        return sequence[:block_size]
+    else:
+        sequence.extend([pad_token_id] * (block_size - len(sequence)))
+        return sequence
+
+
+def build_lm_labels(sequence, pad_token_id):
+    """ Padding token, encoded as 0, are represented by the value -1 so they
+    are not taken into account in the loss computation. """
+    padded = sequence.clone()
+    padded[padded == pad_token_id] = -1
+    return padded
+
+
+def build_mask(sequence, pad_token_id):
+    """ Builds the mask. The attention mechanism will only attend to positions
+    with value 1. """
+    mask = torch.ones_like(sequence)
+    idx_pad_tokens = sequence == pad_token_id
+    mask[idx_pad_tokens] = 0
+    return mask
diff --git a/transformers/modeling_beam_search.py b/transformers/modeling_beam_search.py
deleted file mode 100644
index 171dcb7247..0000000000
--- a/transformers/modeling_beam_search.py
+++ /dev/null
@@ -1,271 +0,0 @@
-# coding=utf-8
-# Copyright (c) 2019 Yang Liu
-
-# Permission is hereby granted, free of charge, to any person obtaining a copy
-# of this software and associated documentation files (the "Software"), to deal
-# in the Software without restriction, including without limitation the rights
-# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
-# copies of the Software, and to permit persons to whom the Software is
-# furnished to do so, subject to the following conditions:
-
-# The above copyright notice and this permission notice shall be included in all
-# copies or substantial portions of the Software.
-
-# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
-# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
-# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
-# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
-# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
-# SOFTWARE.
-"""
-A general wrapper around models with LM heads to generate sequences
-using beam search.
-"""
-import torch
-from torch import nn
-
-
-class TransformerBeamSearch(nn.Module):
-    def __init__(
-        self,
-        model,
-        tokenizer,
-        batch_size,
-        beam_size,
-        min_length,
-        max_length,
-        alpha=0,
-        block_repeating_trigram=True,
-    ):
-        """
-        Attributes:
-            mask_word_id: token id that corresponds to the mask
-        """
-        super(TransformerBeamSearch, self).__init__()
-        self.model = model
-        self.tokenizer = tokenizer
-
-        self.start_token_id = tokenizer.start_token_id
-        self.end_token_id = tokenizer.end_token_id
-        self.pad_token_id = tokenizer.pad_token_id
-
-        self.beam_size = beam_size
-        self.min_length = min_length
-        self.max_length = max_length
-
-        self.block_repeating_trigram = block_repeating_trigram
-        self.apply_length_penalty = False if alpha == 0 else True
-        self.alpha = alpha
-
-        # State of the beam
-        self.hypotheses = [[] for _ in range(batch_size)]
-        self.batch_offset = torch.arange(batch_size, dtype=torch.long)
-        self.beam_offset = torch.arange(
-            0, batch_size * self.beam_size, step=self.beam_size, dtype=torch.long
-        )
-        self.growing_beam = torch.full(
-            (batch_size * self.beam_size, 1), self.start_token_id, dtype=torch.long
-        )
-        self.topk_log_probabilities = torch.tensor(
-            [0.0] + [float("-inf")] * (self.beam_size - 1), dtype=torch.float
-        ).repeat(batch_size)
-        self.results = {
-            "prediction": [[] for _ in batch_size],
-            "scores": [[] for _ in batch_size],
-        }
-        self._step = 0
-        self.is_done = False
-
-    def step(self, log_probabilities):
-        """ Grows the beam by one step. """
-        self._step += 1
-
-        # The batch size changes as some beams finish so we define _B
-        vocab_size = log_probabilities.size(-1)
-        _B = log_probabilities.size(0) // self.beam_size
-
-        # Multiply each beam probability with the probability of the
-        # next token (conditioned on the words in the beam).
-        log_probabilities += self.topk_log_probabilities.view(-1, 1)
-
-        self.enforce_min_length(log_probabilities)
-        if self.block_repeating_trigram:
-            self.remove_repeating_trigrams(log_probabilities, _B)
-
-        # Find the `beam_size` (previous_beam + token) combinations with
-        # the highest score
-        topk_log_probabilities, topk_ids = log_probabilities.topk(
-            log_probabilities.view(_B, self.beam_size * vocab_size),
-            self.beam_size,
-            dim=1,
-        )
-
-        # Apply the length penalty. The +1 accounts for the [EOS] token
-        # that will be added if the beam ends.
-        topk_scores = topk_log_probabilities / self.length_penalty()
-
-        # Retrieve the corresponding respective beam and token id
-        # topk_token_ids[i] will be added to topk_beam_ids[i]
-        topk_beam_ids = topk_ids.div(vocab_size)
-        topk_token_ids = topk_ids.fmod(vocab_size)
-
-        # Retrieve the row index of the surviving beams in the original
-        # view of the log_probabilities tensor
-        surviving_beams_rows = (topk_beam_ids + self.beam_offset[:_B].view(-1, 1)).view(
-            -1
-        )
-
-        # Append the last predictions
-        self.growing_beam = torch.cat(
-            [
-                self.growing_beam.index_select(0, surviving_beams_rows),
-                topk_token_ids.view(-1, 1),
-            ],
-            1,
-        )
-
-        # Check if any of the beam searches has ended during this
-        # growth step. Also if top beam (most probable) has ended
-        # for one element of the batch.
-        is_finished = topk_token_ids.eq(self.end_token_id)
-        self.enforce_max_length()
-        is_top_beam_finished = is_finished[:, 0].eq(1)
-
-        # Save the finished searches
-        if is_finished.any():
-            predictions = self.growing_beam.view(
-                -1, self.beam_size, self.growing_beam.size(1)
-            )
-            for i in range(is_finished.size(0)):
-                if is_top_beam_finished[i]:
-                    is_finished[i].fill_(1)
-                finished_hyp = is_finished[i].nonzero().view(-1)
-
-                # Store finished hypotheses for this batch.
-                b = self.batch_offset[i]
-                for j in finished_hyp:
-                    self.hypotheses[b].append((topk_scores[i, j], predictions[i, j, :]))
-
-                # If the batch reached the end, save the best hypotheses
-                # in terms of length-penalized score.
-                if is_top_beam_finished[i]:
-                    best_hyp = sorted(
-                        self.hypotheses[b], key=lambda x: x[0], reverse=True
-                    )
-                    best_score, best_prediction = best_hyp[0]
-                    self.results["scores"][b].append(best_score)
-                    self.results["predictions"][b].append(best_prediction)
-
-            non_finished = is_top_beam_finished.eq(0).nonzero().view(-1)
-            if len(non_finished) == 0:
-                self.is_done = True
-
-            # Remove finished batches for the next step.
-            topk_log_probabilities = topk_log_probabilities.index_select(
-                0, non_finished
-            )
-            self.batch_offset = self.batch_offset.index_select(0, non_finished)
-            self.growing_beam = predictions.index_select(0, non_finished).view(
-                -1, self.growing_beam.size(-1)
-            )
-
-            surviving_beams_rows = surviving_beams_rows.index_select(0, non_finished)
-
-        return surviving_beams_rows
-
-    def forward(self, encoder_input_ids, **kwargs):
-        # keyword arguments come in 3 flavors: encoder-specific (prefixed by
-        # `encoder_`), decoder-specific (prefixed by `decoder_`) and those
-        # that apply to the model as whole.
-        # We let the specific kwargs override the common ones in case of conflict.
-        kwargs_encoder = {
-            argument[len("encoder_"):]: value
-            for argument, value in kwargs.items()
-            if argument.startswith("encoder_")
-        }
-        kwargs_decoder = {
-            argument[len("decoder_"):]: value
-            for argument, value in kwargs.items()
-            if argument.startswith("decoder_")
-        }
-        kwargs_common = {
-            argument: value
-            for argument, value in kwargs.items()
-            if not (argument.startswith("encoder_") or argument.startswith("decoder_"))
-        }
-        kwargs_decoder = dict(kwargs_common, **kwargs_decoder)
-        kwargs_encoder = dict(kwargs_common, **kwargs_encoder)
-
-        # forward pass on the encoder
-        encoder_outputs = self.model.encoder.forward(encoder_input_ids, kwargs_encoder)
-        kwargs_decoder["encoder_hidden_states"] = tile(
-            encoder_outputs, self.beam_size, dim=0
-        )
-
-        # grow the beam by generating sequences in an autoregressive way
-        self.growing_beam = torch.full(
-            (self.batch_size * self.beam_size, 1), self.start_token_id, dtype=torch.long
-        )
-        for step in range(self.max_length):
-            decoder_input = self.growing_beam[:, -1]
-            outputs = self.model.decoder(decoder_input, kwargs_decoder)
-            log_probabilities = torch.nn.functional.log_softmax(outputs[1])
-            surviving_beams_rows = self.step(log_probabilities)
-            if self.is_done:
-                break
-
-            kwargs_decoder["encoder_hidden_states"] = kwargs_decoder[
-                "encoder_hidden_states"
-            ].index_select(0, surviving_beams_rows)
-
-        return self.results
-
-    def remove_repeating_trigrams(self, log_probabilities, _B):
-        if(self._step + 1 > 3):
-            for i in range(_B * self.beam_size):
-                tokens = [t for t in self.growing_beam[i]]
-                trigrams = [(tokens[i-1], tokens[i], tokens[i+1]) for i in range(1, len(words) - 1)]
-                last_trigram = tuple(trigrams[-1])
-                if last_trigram in trigrams[:-1]:
-                    log_probabilities[i] = -1e20
-
-    def enforce_min_length(self):
-        if self._step < self.min_length:
-            self.log_probabilities[self.end_token_id] = -1e20
-
-    def enforce_max_length(self):
-        if self._step + 1 == self.max_length:
-            self.is_finished.fill_(1)
-
-    def length_penalty(self):
-        return ((5.0 + (self._step + 1)) / 6.0) ** self.alpha
-
-
-def tile(x, count, dim=0):
-    """
-    Tiles `x` along dimension `dim` `count` times.
-
-    Example:
-        >> ex = torch.tensor([1,2],[3,4])
-        >> tile(ex, 2, 0)
-        torch.Tensor([[1,2],[1,2],[3,4],[3,4]])
-    """
-    perm = list(range(len(x.size())))
-    if dim != 0:
-        perm[0], perm[dim] = perm[dim], perm[0]
-        x = x.permute(perm).contiguous()
-    out_size = list(x.size())
-    out_size[0] *= count
-    batch = x.size(0)
-    x = (
-        x.view(batch, -1)
-        .transpose(0, 1)
-        .repeat(count, 1)
-        .transpose(0, 1)
-        .contiguous()
-        .view(*out_size)
-    )
-    if dim != 0:
-        x = x.permute(perm).contiguous()
-    return x
diff --git a/transformers/tests/beam_search_tests.py b/transformers/tests/beam_search_tests.py
new file mode 100644
index 0000000000..a92ebf3578
--- /dev/null
+++ b/transformers/tests/beam_search_tests.py
@@ -0,0 +1,226 @@
+from collections import namedtuple
+import unittest
+
+import numpy as np
+import torch
+
+from transformers.generate import BeamSearch
+from transformers import PreTrainedEncoderDecoder
+
+
+StubTokenizer = namedtuple("Tokenizer", ["bos_token_id", "eos_token_id", "pad_token_id"])
+StubTransformer = namedtuple("Transformer", ["encoder", "decoder"])
+
+
+class BeamSearchtest(unittest.TestCase):
+    def test_beam_search_encoder_decoder_integration(self):
+        """ We make sure that no internal change in the PreTrainedEncoderDecoder
+        class will break the integration with the beam search.
+        """
+
+        model = PreTrainedEncoderDecoder("encoder", "decoder")
+        tokenizer = StubTokenizer(0, 1, 2)
+        try:
+            _ = BeamSearch(
+                model=model,
+                tokenizer=tokenizer,
+                batch_size=1,
+                beam_size=1,
+                min_length=1,
+                max_length=1,
+                alpha=0,
+                block_repeating_trigrams=False,
+            )
+        except:
+            self.fail("Instantiating BeamSearch with a PreTrainedEncoderDecoder failed.")
+
+    def test_beam_search_min_length(self):
+        """ We keep predicting the end_token for the first beam and check that
+        it is not marked as finished until the beam has reached the minimum
+        length. """
+        eos_idx = 3
+        vocab_size = 10
+
+        batch_size = 3
+        beam_size = 2
+        min_length = 5
+
+        beam = BeamSearch(
+            model=StubTransformer("encoder", "decoder"),
+            tokenizer=StubTokenizer(bos_token_id=0, eos_token_id=eos_idx, pad_token_id=2),
+            batch_size=batch_size,
+            beam_size=beam_size,
+            min_length=5,
+            max_length=10,
+            alpha=0,
+            block_repeating_trigrams=False,
+        )
+
+        # To test that the minimum length is correctly enforced we constantly
+        # assign the highest probability to the [EOS] token (and assign lower
+        # probabilities to some other tokens).
+        # Since BeamSearch will reset its probability to 1e-20 as long as
+        # min_length has not been reached, we need to reset the value between
+        # steps.
+        non_eos_idxs = [4, 5, 1, 8, 9]
+        score_distribution = torch.log_softmax(
+            torch.tensor([6.0, 5.0, 4.0, 3.0, 2.0, 1.0]), dim=0
+        )
+
+        log_probabilities = torch.full((batch_size * beam_size, vocab_size), float("-inf"))
+        log_probabilities[0, eos_idx] = score_distribution[0]
+        for idx, score in zip(non_eos_idxs, score_distribution[1:]):
+            log_probabilities[0, idx] = score
+
+        for step in range(1, min_length + 2):
+            log_probabilities[0, eos_idx] = score_distribution[0]
+
+            # Beam #3 and #4 teminate at the first step since the probability
+            # of the [EOS] token is -1e20 > -\infty so there are only two beams left.
+            surviving_beams_rows = beam.grow(log_probabilities)
+            if step < min_length:
+                np.testing.assert_array_equal(
+                    beam.growing_beams.numpy(),
+                    np.repeat(np.array([[0] + [4] * step]), 2, axis=0),
+                )
+            elif step == min_length:
+                np.testing.assert_array_equal(surviving_beams_rows.numpy(), np.array([]))
+                self.assertTrue(beam.is_done)
+                break
+
+            log_probabilities = log_probabilities.index_select(0, surviving_beams_rows)
+
+    def test_beam_search_max_length(self):
+        """ We keep predicting the same non-EOS token until we reach the
+        maximum permitted length """
+        batch_size = 3
+        beam_size = 2
+        max_length = 5
+        vocab_size = 10
+
+        beam = BeamSearch(
+            model=StubTransformer("encoder", "decoder"),
+            tokenizer=StubTokenizer(bos_token_id=0, eos_token_id=1, pad_token_id=2),
+            batch_size=batch_size,
+            beam_size=beam_size,
+            min_length=2,
+            max_length=max_length,
+            alpha=0,
+            block_repeating_trigrams=False,
+        )
+
+        log_probabilities = torch.full((batch_size * beam_size, vocab_size), float("-inf"))
+
+        # To test that beam search enforces the max length constraint we
+        # keep giving the highest probability to a token that is not the
+        # [EOS] token.
+        # The beam search will stop at max_length-1, assuming that one would
+        # add the [EOS] token at the end of the returned sequence.
+        token_idxs = [3, 4, 5]
+        score_distribution = torch.log_softmax(torch.tensor([10.0, 6.0, 4.0]), dim=0)
+        for idx, score in zip(token_idxs, score_distribution):
+            log_probabilities[:, idx] = score
+
+        for step in range(1, max_length + 2):
+            surviving_beams_rows = beam.grow(log_probabilities)
+            if step + 1 < max_length:
+                self.assertFalse(beam.is_done)
+            elif step + 1 == max_length:  # Now [EOS] is the most probable token
+                np.testing.assert_array_equal(surviving_beams_rows.numpy(), np.array([]))
+                self.assertTrue(beam.is_done)
+                break
+
+            log_probabilities = log_probabilities.index_select(0, surviving_beams_rows)
+
+    def test_beam_search_block_repeating_trigrams(self):
+        """ We make sure that the beams that contain repeating trigrams are removed. """
+        batch_size = 3
+        beam_size = 2
+        max_length = 10
+        vocab_size = 10
+
+        beam = BeamSearch(
+            model=StubTransformer("encoder", "decoder"),
+            tokenizer=StubTokenizer(bos_token_id=0, eos_token_id=1, pad_token_id=2),
+            batch_size=batch_size,
+            beam_size=beam_size,
+            min_length=2,
+            max_length=max_length,
+            alpha=0,
+            block_repeating_trigrams=True,
+        )
+
+        log_probabilities = torch.full((batch_size * beam_size, vocab_size), float("-inf"))
+
+        # To test that BeamSearch enforces the 3-gram constraint we give the
+        # highest probably to the same tokens in a cyclic fashion and make sure
+        # they disappear once the cycle has completed.
+        token_idxs = [3, 4, 5]
+        score_distribution = torch.log_softmax(torch.tensor([10.0, 6.0, 4.0]), dim=0)
+        for idx, score in zip(token_idxs, score_distribution):
+            log_probabilities[:, idx] = score
+
+        for step in range(1, max_length + 2):
+            # Rotate the probabilities at each step
+            for idx in token_idxs:
+                score = score_distribution[(idx + step) % 3]
+                log_probabilities[::beam_size, idx] = score
+
+            surviving_beams_rows = beam.grow(log_probabilities)
+            log_probabilities = log_probabilities.index_select(0, surviving_beams_rows)
+
+            if step < 7:
+                self.assertFalse(
+                    np.array_equal(
+                        log_probabilities.numpy()[0, :],
+                        np.array([-1e20] * vocab_size, dtype="float32"),
+                    )
+                )
+            if step == 7:
+                np.testing.assert_array_equal(
+                    log_probabilities.numpy()[0, :],
+                    np.array([-1e20] * vocab_size, dtype="float32"),
+                )
+
+    def test_beam_search_example_for_one_step(self):
+        """ We test that the predictions for one step of growth are correct. """
+        batch_size = 2
+        beam_size = 2
+        max_length = 10
+        vocab_size = 5
+
+        beam = BeamSearch(
+            model=StubTransformer("encoder", "decoder"),
+            tokenizer=StubTokenizer(bos_token_id=0, eos_token_id=1, pad_token_id=2),
+            batch_size=batch_size,
+            beam_size=beam_size,
+            min_length=2,
+            max_length=max_length,
+            alpha=0,
+            block_repeating_trigrams=False,
+        )
+
+        log_probabilities = torch.full((batch_size * beam_size, vocab_size), float("-inf"))
+        log_probabilities[0, 3:] = torch.log_softmax(torch.tensor([2.0, 1.0]), dim=0)
+        log_probabilities[2, 3:] = torch.log_softmax(torch.tensor([1.0, 2.0]), dim=0)
+
+        # First pass
+        surviving_beams_rows = beam.grow(log_probabilities)
+        np.testing.assert_array_equal(surviving_beams_rows.numpy(), np.array([0, 0, 2, 2]))
+        np.testing.assert_array_equal(
+            beam.growing_beams.numpy(), np.array([[0, 3], [0, 4], [0, 4], [0, 3]])
+        )
+        self.assertFalse(beam.is_done)
+
+        # Second pass
+        surviving_beams_rows = beam.grow(log_probabilities)
+        np.testing.assert_array_equal(surviving_beams_rows.numpy(), np.array([0, 0, 2, 2]))
+        np.testing.assert_array_equal(
+            beam.growing_beams.numpy(),
+            np.array([[0, 3, 3], [0, 3, 4], [0, 4, 4], [0, 4, 3]]),
+        )
+        self.assertFalse(beam.is_done)
+
+
+if __name__ == "__name__":
+    unittest.main()

From ba089c780b918414bd8b669e1764fed728753edf Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Wed, 6 Nov 2019 13:55:24 +0100
Subject: [PATCH 268/505] share pretrained embeddings

---
 examples/utils_summarization.py      | 11 +---
 requirements.txt                     |  4 +-
 transformers/generate/beam_search.py | 87 ++++++++++++++++++----------
 3 files changed, 60 insertions(+), 42 deletions(-)

diff --git a/examples/utils_summarization.py b/examples/utils_summarization.py
index 7cbd4cd61b..8e95a04e19 100644
--- a/examples/utils_summarization.py
+++ b/examples/utils_summarization.py
@@ -136,18 +136,11 @@ def encode_for_summarization(story_lines, summary_lines, tokenizer):
     as specified in [1] by using `[SEP] [CLS]` tokens to separate
     sentences.
     """
-    story_lines_token_ids = [
-        tokenizer.build_inputs_with_special_tokens(tokenizer.encode(line))
-        for line in story_lines
-    ]
-    summary_lines_token_ids = [
-        tokenizer.build_inputs_with_special_tokens(tokenizer.encode(line))
-        for line in summary_lines
-    ]
-
+    story_lines_token_ids = [tokenizer.encode(line) for line in story_lines]
     story_token_ids = [
         token for sentence in story_lines_token_ids for token in sentence
     ]
+    summary_lines_token_ids = [tokenizer.encode(line) for line in summary_lines]
     summary_token_ids = [
         token for sentence in summary_lines_token_ids for token in sentence
     ]
diff --git a/requirements.txt b/requirements.txt
index 9c43abc6d7..060aba915d 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -9,4 +9,6 @@ regex
 # For XLNet
 sentencepiece
 # For XLM
-sacremoses
\ No newline at end of file
+sacremoses
+# For ROUGE
+pyrouge
diff --git a/transformers/generate/beam_search.py b/transformers/generate/beam_search.py
index 09e340a150..e1b2d23da0 100644
--- a/transformers/generate/beam_search.py
+++ b/transformers/generate/beam_search.py
@@ -26,27 +26,31 @@ Use Beam Search to generate sequences using encoder-decoder models.
 import torch
 from torch import nn
 
+import logging
+
+
+logger = logging.getLogger(__name__)
+
 
 class BeamSearch(nn.Module):
     def __init__(
         self,
         model,
-        tokenizer,
+        bos_token_id,
+        pad_token_id,
+        eos_token_id,
+        batch_size,
         beam_size,
         min_length,
         max_length,
-        batch_size=1,
         alpha=0,
         block_repeating_trigrams=True,
+        device=torch.device("cpu"),
     ):
         r"""
         Inputs:
             **model**: instance of ``transformers.PreTrainedEncoderDecoder``
                 The pretrained encoder-decoder model that will be used to generate the sequences.
-            **tokenizer**: instance of ``transformers.PreTrainedTokenizer``
-                The pretrained tokenizer associated to the model used in the encoder-decoder. We only
-                support encoder-decoder that use the same tokenizer for encoder and decoder. The tokenizer
-                needs to be initialized or this function will raise and exception.
             **batch_size**: (`optional`) int
                 Batch size of the inputs. The value is set automatically when calling `forward`.
             **beam_size**: int
@@ -64,11 +68,11 @@ class BeamSearch(nn.Module):
         """
         super(BeamSearch, self).__init__()
         self.model = model
-        self.tokenizer = tokenizer
+        self.device = device
 
-        self.bos_token_id = tokenizer.bos_token_id
-        self.eos_token_id = tokenizer.eos_token_id
-        self.pad_token_id = tokenizer.pad_token_id
+        self.bos_token_id = bos_token_id
+        self.eos_token_id = eos_token_id
+        self.pad_token_id = pad_token_id
 
         self.batch_size = batch_size
         self.beam_size = beam_size
@@ -90,15 +94,24 @@ class BeamSearch(nn.Module):
     def _init_beam_state(self, batch_size):
         """ (re-)Initialize the state of the beams. """
         self.hypotheses = [[] for _ in range(batch_size)]
-        self.batch_offset = torch.arange(batch_size, dtype=torch.long)
+        self.batch_offset = torch.arange(batch_size, dtype=torch.long, device=self.device)
         self.beam_offset = torch.arange(
-            0, batch_size * self.beam_size, step=self.beam_size, dtype=torch.long
+            0,
+            batch_size * self.beam_size,
+            step=self.beam_size,
+            dtype=torch.long,
+            device=self.device,
         )
         self.growing_beams = torch.full(
-            (batch_size * self.beam_size, 1), self.bos_token_id, dtype=torch.long
+            (batch_size * self.beam_size, 1),
+            self.bos_token_id,
+            dtype=torch.long,
+            device=self.device,
         )
         self.topk_log_probabilities = torch.tensor(
-            [0.0] + [float("-inf")] * (self.beam_size - 1), dtype=torch.float
+            [0.0] + [float("-inf")] * (self.beam_size - 1),
+            dtype=torch.float,
+            device=self.device,
         ).repeat(batch_size)
         self.results = {
             "predictions": [[] for _ in range(batch_size)],
@@ -136,28 +149,37 @@ class BeamSearch(nn.Module):
         )
 
         # forward pass on the encoder
-        encoder_outputs = self.model.encoder.forward(encoder_input_ids, kwargs_encoder)
+        encoder_outputs = self.model.encoder(encoder_input_ids, **kwargs_encoder)
+        encoder_hidden_states = encoder_outputs[0]
         kwargs_decoder["encoder_hidden_states"] = tile(
-            encoder_outputs, self.beam_size, dim=0
+            encoder_hidden_states, self.beam_size, dim=0
+        )
+        kwargs_decoder["encoder_attention_mask"] = tile(
+            kwargs_encoder["attention_mask"], self.beam_size, dim=0
         )
 
         # grow the beam by generating sequences in an autoregressive way
-        batch_size = encoder_input_ids.size(0)
+        batch_size, block_size = encoder_input_ids.size()
         self._init_beam_state(batch_size)
         for step in range(self.max_length):
-            # prepare the decoder input
-            decoder_input = fit_to_block_size(
-                self.growing_beams, self.tokenizer.pad_token_id
-            )
-            kwargs_decoder["decoder_lm_labels"] = build_lm_labels(
-                decoder_input, self.tokenizer.pad_token_id
-            )
-            kwargs_decoder["decoder_attention_mask"] = build_mask(
-                decoder_input, self.tokenizer.pad_token_id
+            # Add padding tokens
+            decoder_input = torch.full(
+                (self.growing_beams.size(0), block_size),
+                self.pad_token_id,
+                dtype=torch.long,
+                device=self.growing_beams.device,
             )
+            decoder_input[:, : self.growing_beams.size(1)] = self.growing_beams
 
-            outputs = self.model.decoder(decoder_input, kwargs_decoder)
-            log_probabilities = torch.nn.functional.log_softmax(outputs[1])
+            # compute decoder_attention_mask
+            decoder_mask = torch.ones_like(decoder_input)
+            idx_pad_tokens = decoder_input == self.pad_token_id
+            decoder_mask[idx_pad_tokens] = 0
+            kwargs_decoder["attention_mask"] = decoder_mask
+
+            outputs = self.model.decoder(decoder_input, **kwargs_decoder)
+            last_token_scores = outputs[0][:, -1, :].squeeze(1)
+            log_probabilities = torch.nn.functional.log_softmax(last_token_scores, dim=0)
             surviving_beams_rows = self.grow(log_probabilities)
             if self.is_done:
                 break
@@ -189,13 +211,13 @@ class BeamSearch(nn.Module):
 
         # Find the `beam_size` (previous_beam + token) combinations with
         # the highest score
-        topk_log_probabilities, topk_ids = torch.topk(
+        self.topk_log_probabilities, topk_ids = torch.topk(
             log_probabilities.view(_B, self.beam_size * vocab_size), self.beam_size, dim=1
         )
 
         # Apply the length penalty. The +1 accounts for the [EOS] token
         # that will be added if the beam ends.
-        topk_scores = topk_log_probabilities
+        topk_scores = self.topk_log_probabilities
         if self.apply_length_penalty:
             topk_scores /= self._length_penalty()
 
@@ -337,8 +359,9 @@ def fit_to_block_size(sequence, block_size, pad_token_id):
     if len(sequence) > block_size:
         return sequence[:block_size]
     else:
-        sequence.extend([pad_token_id] * (block_size - len(sequence)))
-        return sequence
+        return torch.cat(
+            (sequence, torch.tensor([pad_token_id] * (block_size - len(sequence)))), dim=0
+        )
 
 
 def build_lm_labels(sequence, pad_token_id):

From 4735c2af0715c24d47b34c167fb7d5543493b87d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Fri, 8 Nov 2019 11:16:26 +0100
Subject: [PATCH 269/505] tweaks to the BeamSearch API

---
 transformers/generate/beam_search.py    | 63 ++++++++++---------------
 transformers/tests/beam_search_tests.py | 53 ++++++++++++++-------
 2 files changed, 59 insertions(+), 57 deletions(-)

diff --git a/transformers/generate/beam_search.py b/transformers/generate/beam_search.py
index e1b2d23da0..a18d20f31a 100644
--- a/transformers/generate/beam_search.py
+++ b/transformers/generate/beam_search.py
@@ -32,7 +32,7 @@ import logging
 logger = logging.getLogger(__name__)
 
 
-class BeamSearch(nn.Module):
+class BeamSearch(object):
     def __init__(
         self,
         model,
@@ -45,12 +45,17 @@ class BeamSearch(nn.Module):
         max_length,
         alpha=0,
         block_repeating_trigrams=True,
-        device=torch.device("cpu"),
     ):
         r"""
         Inputs:
             **model**: instance of ``transformers.PreTrainedEncoderDecoder``
                 The pretrained encoder-decoder model that will be used to generate the sequences.
+            **bos_token_id**: int
+                Id that is used by the tokenizer to represent the beggining of a sentence.
+            **pad_token_id**: int
+                Id that is used by the tokenizer for padding.
+            **eos_token_id**: int
+                Id that is used by the tokenizer to represent the end of a sentence.
             **batch_size**: (`optional`) int
                 Batch size of the inputs. The value is set automatically when calling `forward`.
             **beam_size**: int
@@ -68,7 +73,7 @@ class BeamSearch(nn.Module):
         """
         super(BeamSearch, self).__init__()
         self.model = model
-        self.device = device
+        self.device = next(model.parameters()).device  # only works if all parameters of the model are stored on a single GPU
 
         self.bos_token_id = bos_token_id
         self.eos_token_id = eos_token_id
@@ -86,10 +91,7 @@ class BeamSearch(nn.Module):
         self._init_beam_state(batch_size)
 
     def __len__(self):
-        try:
-            return self.growing_beams.size(1)
-        except NameError:
-            return 0
+        return self.growing_beams.size(1)
 
     def _init_beam_state(self, batch_size):
         """ (re-)Initialize the state of the beams. """
@@ -120,7 +122,7 @@ class BeamSearch(nn.Module):
         self._step = 0
         self.is_done = False
 
-    def forward(self, encoder_input_ids, **model_kwargs):
+    def __call__(self, encoder_input_ids, **model_kwargs):
         """ Generate a sequence using Beam Search. """
         # keyword arguments come in 3 flavors: encoder-specific (prefixed by
         # `encoder_`), decoder-specific (prefixed by `decoder_`) and those
@@ -158,28 +160,17 @@ class BeamSearch(nn.Module):
             kwargs_encoder["attention_mask"], self.beam_size, dim=0
         )
 
-        # grow the beam by generating sequences in an autoregressive way
+        # grow the beam iteratively
         batch_size, block_size = encoder_input_ids.size()
         self._init_beam_state(batch_size)
         for step in range(self.max_length):
-            # Add padding tokens
-            decoder_input = torch.full(
-                (self.growing_beams.size(0), block_size),
-                self.pad_token_id,
-                dtype=torch.long,
-                device=self.growing_beams.device,
-            )
-            decoder_input[:, : self.growing_beams.size(1)] = self.growing_beams
-
-            # compute decoder_attention_mask
-            decoder_mask = torch.ones_like(decoder_input)
-            idx_pad_tokens = decoder_input == self.pad_token_id
-            decoder_mask[idx_pad_tokens] = 0
-            kwargs_decoder["attention_mask"] = decoder_mask
 
+            decoder_input = fit_to_block_size(self.growing_beams, block_size, self.pad_token_id)
+            kwargs_decoder["attention_mask"] = build_mask(decoder_input)
             outputs = self.model.decoder(decoder_input, **kwargs_decoder)
-            last_token_scores = outputs[0][:, -1, :].squeeze(1)
-            log_probabilities = torch.nn.functional.log_softmax(last_token_scores, dim=0)
+
+            next_token_scores = outputs[0][:, -1, :].squeeze(1)
+            log_probabilities = torch.nn.functional.log_softmax(next_token_scores, dim=0)
             surviving_beams_rows = self.grow(log_probabilities)
             if self.is_done:
                 break
@@ -356,20 +347,14 @@ def fit_to_block_size(sequence, block_size, pad_token_id):
     """ Adapt the source and target sequences' lengths to the block size.
     If the sequence is shorter we append padding tokens to the right.
     """
-    if len(sequence) > block_size:
-        return sequence[:block_size]
-    else:
-        return torch.cat(
-            (sequence, torch.tensor([pad_token_id] * (block_size - len(sequence)))), dim=0
-        )
-
-
-def build_lm_labels(sequence, pad_token_id):
-    """ Padding token, encoded as 0, are represented by the value -1 so they
-    are not taken into account in the loss computation. """
-    padded = sequence.clone()
-    padded[padded == pad_token_id] = -1
-    return padded
+    padded_sequence = torch.full(
+        (sequence.size(0), block_size),
+        pad_token_id,
+        dtype=torch.long,
+        device=sequence.device,
+    )
+    padded_sequence[:, : sequence.size(1)] = sequence
+    return sequence
 
 
 def build_mask(sequence, pad_token_id):
diff --git a/transformers/tests/beam_search_tests.py b/transformers/tests/beam_search_tests.py
index a92ebf3578..6f2a2b9c2f 100644
--- a/transformers/tests/beam_search_tests.py
+++ b/transformers/tests/beam_search_tests.py
@@ -1,15 +1,22 @@
 from collections import namedtuple
 import unittest
-
+import pytest
 import numpy as np
 import torch
+from torch import nn
 
 from transformers.generate import BeamSearch
 from transformers import PreTrainedEncoderDecoder
 
 
-StubTokenizer = namedtuple("Tokenizer", ["bos_token_id", "eos_token_id", "pad_token_id"])
-StubTransformer = namedtuple("Transformer", ["encoder", "decoder"])
+class StubTransformer(nn.Module):
+    def __init__(self):
+        self.encoder = None
+        self.decoder = None
+        self._parameters = {"dumy": torch.tensor([1])}
+
+    def forward(self):
+        pass
 
 
 class BeamSearchtest(unittest.TestCase):
@@ -18,12 +25,13 @@ class BeamSearchtest(unittest.TestCase):
         class will break the integration with the beam search.
         """
 
-        model = PreTrainedEncoderDecoder("encoder", "decoder")
-        tokenizer = StubTokenizer(0, 1, 2)
+        model = StubTransformer()
         try:
             _ = BeamSearch(
                 model=model,
-                tokenizer=tokenizer,
+                bos_token_id=0,
+                eos_token_id=1,
+                pad_token_id=2,
                 batch_size=1,
                 beam_size=1,
                 min_length=1,
@@ -46,8 +54,10 @@ class BeamSearchtest(unittest.TestCase):
         min_length = 5
 
         beam = BeamSearch(
-            model=StubTransformer("encoder", "decoder"),
-            tokenizer=StubTokenizer(bos_token_id=0, eos_token_id=eos_idx, pad_token_id=2),
+            model=StubTransformer(),
+            bos_token_id=0,
+            eos_token_id=eos_idx,
+            pad_token_id=2,
             batch_size=batch_size,
             beam_size=beam_size,
             min_length=5,
@@ -71,17 +81,17 @@ class BeamSearchtest(unittest.TestCase):
         log_probabilities[0, eos_idx] = score_distribution[0]
         for idx, score in zip(non_eos_idxs, score_distribution[1:]):
             log_probabilities[0, idx] = score
-
+        pytest.set_trace()
         for step in range(1, min_length + 2):
             log_probabilities[0, eos_idx] = score_distribution[0]
 
             # Beam #3 and #4 teminate at the first step since the probability
             # of the [EOS] token is -1e20 > -\infty so there are only two beams left.
+            # The top beam (most likely) always ends with 4 until we reach min_length.
             surviving_beams_rows = beam.grow(log_probabilities)
             if step < min_length:
                 np.testing.assert_array_equal(
-                    beam.growing_beams.numpy(),
-                    np.repeat(np.array([[0] + [4] * step]), 2, axis=0),
+                    beam.growing_beams.numpy()[0, :], np.array([0] + [4] * step)
                 )
             elif step == min_length:
                 np.testing.assert_array_equal(surviving_beams_rows.numpy(), np.array([]))
@@ -99,8 +109,10 @@ class BeamSearchtest(unittest.TestCase):
         vocab_size = 10
 
         beam = BeamSearch(
-            model=StubTransformer("encoder", "decoder"),
-            tokenizer=StubTokenizer(bos_token_id=0, eos_token_id=1, pad_token_id=2),
+            model=StubTransformer(),
+            bos_token_id=0,
+            eos_token_id=1,
+            pad_token_id=2,
             batch_size=batch_size,
             beam_size=beam_size,
             min_length=2,
@@ -140,8 +152,10 @@ class BeamSearchtest(unittest.TestCase):
         vocab_size = 10
 
         beam = BeamSearch(
-            model=StubTransformer("encoder", "decoder"),
-            tokenizer=StubTokenizer(bos_token_id=0, eos_token_id=1, pad_token_id=2),
+            model=StubTransformer(),
+            bos_token_id=0,
+            eos_token_id=1,
+            pad_token_id=2,
             batch_size=batch_size,
             beam_size=beam_size,
             min_length=2,
@@ -167,7 +181,6 @@ class BeamSearchtest(unittest.TestCase):
                 log_probabilities[::beam_size, idx] = score
 
             surviving_beams_rows = beam.grow(log_probabilities)
-            log_probabilities = log_probabilities.index_select(0, surviving_beams_rows)
 
             if step < 7:
                 self.assertFalse(
@@ -182,6 +195,8 @@ class BeamSearchtest(unittest.TestCase):
                     np.array([-1e20] * vocab_size, dtype="float32"),
                 )
 
+            log_probabilities = log_probabilities.index_select(0, surviving_beams_rows)
+
     def test_beam_search_example_for_one_step(self):
         """ We test that the predictions for one step of growth are correct. """
         batch_size = 2
@@ -190,8 +205,10 @@ class BeamSearchtest(unittest.TestCase):
         vocab_size = 5
 
         beam = BeamSearch(
-            model=StubTransformer("encoder", "decoder"),
-            tokenizer=StubTokenizer(bos_token_id=0, eos_token_id=1, pad_token_id=2),
+            model=StubTransformer(),
+            bos_token_id=0,
+            eos_token_id=1,
+            pad_token_id=2,
             batch_size=batch_size,
             beam_size=beam_size,
             min_length=2,

From 9f75565ea8243ec685c3e5dd08a63e8f78af9d0f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Fri, 8 Nov 2019 15:48:31 +0100
Subject: [PATCH 270/505] setup training

---
 requirements.txt                     | 2 --
 transformers/generate/beam_search.py | 2 +-
 2 files changed, 1 insertion(+), 3 deletions(-)

diff --git a/requirements.txt b/requirements.txt
index 060aba915d..4a3162adce 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -10,5 +10,3 @@ regex
 sentencepiece
 # For XLM
 sacremoses
-# For ROUGE
-pyrouge
diff --git a/transformers/generate/beam_search.py b/transformers/generate/beam_search.py
index a18d20f31a..abe3186049 100644
--- a/transformers/generate/beam_search.py
+++ b/transformers/generate/beam_search.py
@@ -166,7 +166,7 @@ class BeamSearch(object):
         for step in range(self.max_length):
 
             decoder_input = fit_to_block_size(self.growing_beams, block_size, self.pad_token_id)
-            kwargs_decoder["attention_mask"] = build_mask(decoder_input)
+            kwargs_decoder["attention_mask"] = build_mask(decoder_input, self.pad_token_id)
             outputs = self.model.decoder(decoder_input, **kwargs_decoder)
 
             next_token_scores = outputs[0][:, -1, :].squeeze(1)

From 4d1819990294f27ab1cf0113034f52cdb4136eaf Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Tue, 12 Nov 2019 17:59:34 +0100
Subject: [PATCH 271/505] cast bool tensor to long for pytorch < 1.3

---
 transformers/modeling_bert.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/transformers/modeling_bert.py b/transformers/modeling_bert.py
index 1ee3e3f097..0159d58aab 100644
--- a/transformers/modeling_bert.py
+++ b/transformers/modeling_bert.py
@@ -675,6 +675,7 @@ class BertModel(BertPreTrainedModel):
                 batch_size, seq_length = input_shape
                 seq_ids = torch.arange(seq_length, device=device)
                 causal_mask = seq_ids[None, None, :].repeat(batch_size, seq_length, 1) <= seq_ids[None, :, None]
+                causal_mask = causal_mask.to(torch.long)  # not converting to long will cause errors with pytorch version < 1.3
                 extended_attention_mask = causal_mask[:, None, :, :] * attention_mask[:, None, None, :]
             else:
                 extended_attention_mask = attention_mask[:, None, None, :]

From 2403a6659859ad18a9f20e1c2e84179718d8dfd4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Sat, 23 Nov 2019 00:18:44 +0100
Subject: [PATCH 272/505] give transformers API to BertAbs

---
 ..._original_pytorch_checkpoint_to_pytorch.py |  161 +++
 .../summarization/configuration_bertabs.py    |  141 ++
 ...ert_bertabs_original_pytorch_checkpoint.py |  162 +++
 examples/summarization/modeling_bertabs.py    | 1250 +++++++++++++++++
 examples/summarization/run_summarization.py   |  271 ++++
 .../utils_summarization.py                    |   56 +-
 .../utils_summarization_test.py               |   17 +-
 ..._original_pytorch_checkpoint_to_pytorch.py |  158 +++
 transformers/generate/beam_search.py          |   26 +-
 9 files changed, 2188 insertions(+), 54 deletions(-)
 create mode 100644 examples/convert_bertextabs_original_pytorch_checkpoint_to_pytorch.py
 create mode 100644 examples/summarization/configuration_bertabs.py
 create mode 100644 examples/summarization/convert_bertabs_original_pytorch_checkpoint.py
 create mode 100644 examples/summarization/modeling_bertabs.py
 create mode 100644 examples/summarization/run_summarization.py
 rename examples/{ => summarization}/utils_summarization.py (77%)
 rename examples/{ => summarization}/utils_summarization_test.py (88%)
 create mode 100644 transformers/convert_bertextabs_original_pytorch_checkpoint_to_pytorch.py

diff --git a/examples/convert_bertextabs_original_pytorch_checkpoint_to_pytorch.py b/examples/convert_bertextabs_original_pytorch_checkpoint_to_pytorch.py
new file mode 100644
index 0000000000..c245d0eae5
--- /dev/null
+++ b/examples/convert_bertextabs_original_pytorch_checkpoint_to_pytorch.py
@@ -0,0 +1,161 @@
+# coding=utf-8
+# Copyright 2018 The HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" Convert BertExtAbs's checkpoints """
+
+import argparse
+from collections import namedtuple
+import logging
+import pdb
+import torch
+
+from models.model_builder import AbsSummarizer  # The authors' implementation
+from model_bertabs import BertAbsSummarizer
+
+from transformers import BertTokenizer
+
+
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+
+SAMPLE_TEXT = 'Hello world! cécé herlolip'
+
+
+BertAbsConfig = namedtuple(
+    "BertAbsConfig",
+    ["temp_dir", "large", "use_bert_emb", "finetune_bert", "encoder", "share_emb", "max_pos", "enc_layers", "enc_hidden_size", "enc_heads", "enc_ff_size", "enc_dropout", "dec_layers", "dec_hidden_size", "dec_heads", "dec_ff_size", "dec_dropout"],
+)
+
+
+def convert_bertabs_checkpoints(path_to_checkpoints, dump_path):
+    """ Copy/paste and tweak the pre-trained weights provided by the creators
+    of BertAbs for the internal architecture.
+    """
+
+    # Instantiate the authors' model with the pre-trained weights
+    config = BertAbsConfig(
+        temp_dir=".",
+        finetune_bert=False,
+        large=False,
+        share_emb=True,
+        use_bert_emb=False,
+        encoder="bert",
+        max_pos=512,
+        enc_layers=6,
+        enc_hidden_size=512,
+        enc_heads=8,
+        enc_ff_size=512,
+        enc_dropout=0.2,
+        dec_layers=6,
+        dec_hidden_size=768,
+        dec_heads=8,
+        dec_ff_size=2048,
+        dec_dropout=0.2,
+    )
+    checkpoints = torch.load(path_to_checkpoints, lambda storage, loc: storage)
+    original = AbsSummarizer(config, torch.device("cpu"), checkpoints)
+    original.eval()
+
+    new_model = BertAbsSummarizer(config, torch.device("cpu"))
+    new_model.eval()
+
+    # -------------------
+    # Convert the weights
+    # -------------------
+
+    logging.info("convert the model")
+    new_model.encoder.load_state_dict(original.bert.state_dict())
+
+    new_model.decoder.generator.load_state_dict(original.generator.state_dict())
+    new_model.decoder.embeddings.load_state_dict(original.decoder.embeddings.state_dict())
+    new_model.decoder.pos_emb.load_state_dict(original.decoder.pos_emb.state_dict())
+    new_model.decoder.transformer_layers.load_state_dict(original.decoder.transformer_layers.state_dict())
+    new_model.decoder.layer_norm.load_state_dict(original.decoder.layer_norm.state_dict())
+
+    # ----------------------------------
+    # Make sure the outpus are identical
+    # ----------------------------------
+
+    logging.info("Make sure that the models' outputs are identical")
+    tokenizer = BertTokenizer.from_pretrained("bert-base-uncased")
+
+    # prepare the model inputs
+    encoder_input_ids = tokenizer.encode("This is sample éàalj'-.")
+    encoder_input_ids.extend([tokenizer.pad_token_id] * (512 - len(encoder_input_ids)))
+    encoder_input_ids = torch.tensor(encoder_input_ids).unsqueeze(0)
+    decoder_input_ids = tokenizer.encode("This is sample 3 éàalj'-.")
+    decoder_input_ids.extend([tokenizer.pad_token_id] * (512 - len(decoder_input_ids)))
+    decoder_input_ids = torch.tensor(decoder_input_ids).unsqueeze(0)
+
+    # failsafe to make sure the weights reset does not affect the
+    # loaded weights.
+    assert torch.max(torch.abs(original.generator[0].weight - new_model.decoder.generator[0].weight)) == 0
+
+    # forward pass
+    src = encoder_input_ids
+    tgt = decoder_input_ids
+    segs = token_type_ids = None
+    clss = None
+    mask_src = encoder_attention_mask = None
+    mask_tgt = decoder_attention_mask = None
+    mask_cls = None
+    
+    # The original model does not apply the geneator layer immediatly but rather in
+    # the beam search (where it combines softmax + linear layer). Since we already
+    # apply the softmax in our generation process we only apply the linear layer here.
+    # We make sure that the outputs of the full stack are identical
+    output_original_model = original(src, tgt, segs, clss, mask_src, mask_tgt, mask_cls)[0]
+    output_original_model = original.generator(output_original_model)
+
+    output_converted_model = new_model(encoder_input_ids, decoder_input_ids, token_type_ids, encoder_attention_mask, decoder_attention_mask)[0]
+    output_converted_model = torch.nn.functional.log_softmax(output_converted_model, dim=-1)
+
+    maximum_absolute_difference = torch.max(torch.abs(output_converted_model - output_original_model)).item()
+    print("Maximum absolute difference beween weights: {:.2f}".format(maximum_absolute_difference))
+
+    are_identical = torch.allclose(output_converted_model, output_original_model, atol=1e-3)
+    if are_identical:
+        logging.info("all weights are equal up to 1e-3")
+    else:
+        raise ValueError("the weights are different. The new model is likely different from the original one.")
+
+    # The model has been saved with torch.save(model) and this is bound to the exact
+    # directory structure. We save the state_dict instead.
+    logging.info("saving the model's state dictionary")
+    torch.save(new_model.state_dict(), "bert-ext-abs.pt")
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        "--bertabs_checkpoint_path",
+        default=None,
+        type=str,
+        required=True,
+        help="Path the official PyTorch dump.",
+    )
+    parser.add_argument(
+        "--pytorch_dump_folder_path",
+        default=None,
+        type=str,
+        required=True,
+        help="Path to the output PyTorch model.",
+    )
+    args = parser.parse_args()
+
+    convert_bertabs_checkpoints(
+        args.bertabs_checkpoint_path,
+        args.pytorch_dump_folder_path,
+    )
diff --git a/examples/summarization/configuration_bertabs.py b/examples/summarization/configuration_bertabs.py
new file mode 100644
index 0000000000..ff3171f9a8
--- /dev/null
+++ b/examples/summarization/configuration_bertabs.py
@@ -0,0 +1,141 @@
+# coding=utf-8
+# Copyright 2019 The HuggingFace Inc. team.
+# Copyright (c) 2018, NVIDIA CORPORATION.  All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" BertAbs configuration """
+import json
+import logging
+import sys
+
+from transformers import PretrainedConfig
+
+
+logger = logging.getLogger(__name__)
+
+
+BERTABS_FINETUNED_CONFIG_MAP = {
+    "bertabs-finetuned-cnndm": "https://s3.amazonaws.com/models.huggingface.co/bert/remi/bertabs-finetuned-cnndm-extractive-abstractive-summarization-config.json",
+}
+
+
+class BertAbsConfig(PretrainedConfig):
+    r""" Class to store the configuration of the BertAbs model.
+
+    Arguments:
+        temp_dir: string
+            Unused in the current situation. Kept for compatibility but will be removed.
+        finetune_bert: bool
+            Whether to fine-tune the model or not. Will be kept for reference
+            in case we want to add the possibility to fine-tune the model.
+        large: bool
+            Whether to use bert-large as a base.
+        share_emb: book
+            Whether the embeddings are shared between the encoder and decoder.
+        encoder: string
+            Not clear what this does. Leave to "bert" for pre-trained weights.
+        max_pos: int
+            The maximum sequence length that this model will be used with.
+        enc_layer: int
+            The numner of hidden layers in the Transformer encoder.
+        enc_hidden_size: int
+            The size of the encoder's layers.
+        enc_heads: int
+            The number of attention heads for each attention layer in the encoder.
+        enc_ff_size: int
+            The size of the encoder's feed-forward layers.
+        enc_dropout: int
+            The dropout probabilitiy for all fully connected layers in the
+            embeddings, layers, pooler and also the attention probabilities in
+            the encoder.
+        dec_layer: int
+            The numner of hidden layers in the decoder.
+        dec_hidden_size: int
+            The size of the decoder's layers.
+        dec_heads: int
+            The number of attention heads for each attention layer in the decoder.
+        dec_ff_size: int
+            The size of the decoder's feed-forward layers.
+        dec_dropout: int
+            The dropout probabilitiy for all fully connected layers in the
+            embeddings, layers, pooler and also the attention probabilities in
+            the decoder.
+    """
+
+    pretrained_config_archive_map = BERTABS_FINETUNED_CONFIG_MAP
+
+    def __init__(
+        self,
+        vocab_size_or_config_json_file=30522,
+        temp_dir=".",
+        finetune_bert=False,
+        large=False,
+        share_emb=True,
+        encoder="bert",
+        max_pos=512,
+        enc_layers=6,
+        enc_hidden_size=512,
+        enc_heads=8,
+        enc_ff_size=512,
+        enc_dropout=0.2,
+        dec_layers=6,
+        dec_hidden_size=768,
+        dec_heads=8,
+        dec_ff_size=2048,
+        dec_dropout=0.2,
+        **kwargs,
+    ):
+        super(BertAbsConfig, self).__init__(**kwargs)
+
+        if self._input_is_path_to_json(vocab_size_or_config_json_file):
+            path_to_json = vocab_size_or_config_json_file
+            with open(path_to_json, "r", encoding="utf-8") as reader:
+                json_config = json.loads(reader.read())
+            for key, value in json_config.items():
+                self.__dict__[key] = value
+        elif isinstance(vocab_size_or_config_json_file, int):
+            self.temp_dir = temp_dir
+            self.finetune_bert = finetune_bert
+            self.large = large
+            self.vocab_size = vocab_size_or_config_json_file
+            self.max_pos = max_pos
+
+            self.encoder = encoder
+            self.enc_layers = enc_layers
+            self.enc_hidden_size = enc_hidden_size
+            self.enc_heads = enc_heads
+            self.enc_ff_size = enc_ff_size
+            self.enc_dropout = enc_dropout
+
+            self.share_emb = share_emb
+
+            self.dec_layers = dec_layers
+            self.dec_hidden_size = dec_hidden_size
+            self.dec_heads = dec_heads
+            self.dec_ff_size = dec_ff_size
+            self.dec_dropout = dec_dropout
+        else:
+            raise ValueError(
+                "First argument must be either a vocabulary size (int)"
+                "or the path to a pretrained model config file (str)"
+            )
+
+    def _input_is_path_to_json(self, first_argument):
+        """ Checks whether the first argument passed to config
+        is the path to a JSON file that contains the config.
+        """
+        is_python_2 = sys.version_info[0] == 2
+        if is_python_2:
+            return isinstance(first_argument, unicode)
+        else:
+            return isinstance(first_argument, str)
diff --git a/examples/summarization/convert_bertabs_original_pytorch_checkpoint.py b/examples/summarization/convert_bertabs_original_pytorch_checkpoint.py
new file mode 100644
index 0000000000..786a29ef13
--- /dev/null
+++ b/examples/summarization/convert_bertabs_original_pytorch_checkpoint.py
@@ -0,0 +1,162 @@
+# coding=utf-8
+# Copyright 2018 The HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" Convert BertExtAbs's checkpoints
+
+The file currently does not do much as we ended up copying the exact model
+structure, but I leave it here in case we ever want to refactor the model.
+"""
+
+import argparse
+from collections import namedtuple
+import logging
+import torch
+
+from models.model_builder import AbsSummarizer  # The authors' implementation
+from model_bertabs import BertAbsSummarizer
+
+from transformers import BertTokenizer
+
+
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+
+SAMPLE_TEXT = 'Hello world! cécé herlolip'
+
+
+BertAbsConfig = namedtuple(
+    "BertAbsConfig",
+    ["temp_dir", "large", "use_bert_emb", "finetune_bert", "encoder", "share_emb", "max_pos", "enc_layers", "enc_hidden_size", "enc_heads", "enc_ff_size", "enc_dropout", "dec_layers", "dec_hidden_size", "dec_heads", "dec_ff_size", "dec_dropout"],
+)
+
+
+def convert_bertabs_checkpoints(path_to_checkpoints, dump_path):
+    """ Copy/paste and tweak the pre-trained weights provided by the creators
+    of BertAbs for the internal architecture.
+    """
+
+    # Instantiate the authors' model with the pre-trained weights
+    config = BertAbsConfig(
+        temp_dir=".",
+        finetune_bert=False,
+        large=False,
+        share_emb=True,
+        use_bert_emb=False,
+        encoder="bert",
+        max_pos=512,
+        enc_layers=6,
+        enc_hidden_size=512,
+        enc_heads=8,
+        enc_ff_size=512,
+        enc_dropout=0.2,
+        dec_layers=6,
+        dec_hidden_size=768,
+        dec_heads=8,
+        dec_ff_size=2048,
+        dec_dropout=0.2,
+    )
+    checkpoints = torch.load(path_to_checkpoints, lambda storage, loc: storage)
+    original = AbsSummarizer(config, torch.device("cpu"), checkpoints)
+    original.eval()
+
+    new_model = BertAbsSummarizer(config, torch.device("cpu"))
+    new_model.eval()
+
+    # -------------------
+    # Convert the weights
+    # -------------------
+
+    logging.info("convert the model")
+    new_model.bert.load_state_dict(original.bert.state_dict())
+    new_model.decoder.load_state_dict(original.decoder.state_dict())
+    new_model.generator.load_state_dict(original.generator.state_dict())
+
+    # ----------------------------------
+    # Make sure the outpus are identical
+    # ----------------------------------
+
+    logging.info("Make sure that the models' outputs are identical")
+    tokenizer = BertTokenizer.from_pretrained("bert-base-uncased")
+
+    # prepare the model inputs
+    encoder_input_ids = tokenizer.encode("This is sample éàalj'-.")
+    encoder_input_ids.extend([tokenizer.pad_token_id] * (512 - len(encoder_input_ids)))
+    encoder_input_ids = torch.tensor(encoder_input_ids).unsqueeze(0)
+    decoder_input_ids = tokenizer.encode("This is sample 3 éàalj'-.")
+    decoder_input_ids.extend([tokenizer.pad_token_id] * (512 - len(decoder_input_ids)))
+    decoder_input_ids = torch.tensor(decoder_input_ids).unsqueeze(0)
+
+    # failsafe to make sure the weights reset does not affect the
+    # loaded weights.
+    assert torch.max(torch.abs(original.generator[0].weight - new_model.generator[0].weight)) == 0
+
+    # forward pass
+    src = encoder_input_ids
+    tgt = decoder_input_ids
+    segs = token_type_ids = None
+    clss = None
+    mask_src = encoder_attention_mask = None
+    mask_tgt = decoder_attention_mask = None
+    mask_cls = None
+
+    # The original model does not apply the geneator layer immediatly but rather in
+    # the beam search (where it combines softmax + linear layer). Since we already
+    # apply the softmax in our generation process we only apply the linear layer here.
+    # We make sure that the outputs of the full stack are identical
+    output_original_model = original(src, tgt, segs, clss, mask_src, mask_tgt, mask_cls)[0]
+    output_original_generator = original.generator(output_original_model)
+
+    output_converted_model = new_model(encoder_input_ids, decoder_input_ids, token_type_ids, encoder_attention_mask, decoder_attention_mask)[0]
+    output_converted_generator = new_model.generator(output_converted_model)
+
+    maximum_absolute_difference = torch.max(torch.abs(output_converted_model - output_original_model)).item()
+    print("Maximum absolute difference beween weights: {:.2f}".format(maximum_absolute_difference))
+    maximum_absolute_difference = torch.max(torch.abs(output_converted_generator - output_original_generator)).item()
+    print("Maximum absolute difference beween weights: {:.2f}".format(maximum_absolute_difference))
+
+    are_identical = torch.allclose(output_converted_model, output_original_model, atol=1e-3)
+    if are_identical:
+        logging.info("all weights are equal up to 1e-3")
+    else:
+        raise ValueError("the weights are different. The new model is likely different from the original one.")
+
+    # The model has been saved with torch.save(model) and this is bound to the exact
+    # directory structure. We save the state_dict instead.
+    logging.info("saving the model's state dictionary")
+    torch.save(new_model.state_dict(), "bertabs-finetuned-cnndm-extractive-abstractive-summarization-pytorch_model.bin")
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        "--bertabs_checkpoint_path",
+        default=None,
+        type=str,
+        required=True,
+        help="Path the official PyTorch dump.",
+    )
+    parser.add_argument(
+        "--pytorch_dump_folder_path",
+        default=None,
+        type=str,
+        required=True,
+        help="Path to the output PyTorch model.",
+    )
+    args = parser.parse_args()
+
+    convert_bertabs_checkpoints(
+        args.bertabs_checkpoint_path,
+        args.pytorch_dump_folder_path,
+    )
diff --git a/examples/summarization/modeling_bertabs.py b/examples/summarization/modeling_bertabs.py
new file mode 100644
index 0000000000..0189a2ad2b
--- /dev/null
+++ b/examples/summarization/modeling_bertabs.py
@@ -0,0 +1,1250 @@
+# MIT License
+
+# Copyright (c) 2019 Yang Liu
+
+# Permission is hereby granted, free of charge, to any person obtaining a copy
+# of this software and associated documentation files (the "Software"), to deal
+# in the Software without restriction, including without limitation the rights
+# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+# copies of the Software, and to permit persons to whom the Software is
+# furnished to do so, subject to the following conditions:
+
+# The above copyright notice and this permission notice shall be included in all
+# copies or substantial portions of the Software.
+
+# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+# SOFTWARE.
+import copy
+import math
+import shutil
+import time
+import os
+
+import numpy as np
+import torch
+from torch import nn
+from torch.nn.init import xavier_uniform_
+
+from transformers import BertModel, BertConfig, PreTrainedModel
+
+from configuration_bertabs import BertAbsConfig
+
+
+MAX_SIZE = 5000
+
+BERTABS_FINETUNED_MODEL_MAP = {
+    "bertabs-finetuned-cnndm": "https://s3.amazonaws.com/models.huggingface.co/bert/remi/bertabs-finetuned-cnndm-extractive-abstractive-summarization-pytorch_model.bin",
+}
+
+
+class BertAbsPreTrainedModel(PreTrainedModel):
+    config_class = BertAbsConfig
+    pretrained_model_archive_map = BERTABS_FINETUNED_MODEL_MAP
+    load_tf_weights = False
+    base_model_prefix = "bert"
+
+
+class BertAbs(BertAbsPreTrainedModel):
+    def __init__(self, args, checkpoint=None, bert_extractive_checkpoint=None):
+        super(BertAbs, self).__init__(args)
+        self.args = args
+        self.bert = Bert(args.large, args.temp_dir, args.finetune_bert)
+
+        # If pre-trained weights are passed for Bert, load these.
+        load_bert_pretrained_extractive = True if bert_extractive_checkpoint else False
+        if load_bert_pretrained_extractive:
+            self.bert.model.load_state_dict(
+                dict(
+                    [
+                        (n[11:], p)
+                        for n, p in bert_extractive_checkpoint.items()
+                        if n.startswith("bert.model")
+                    ]
+                ),
+                strict=True,
+            )
+
+        if args.encoder == "baseline":
+            bert_config = BertConfig(
+                self.bert.model.config.vocab_size,
+                hidden_size=args.enc_hidden_size,
+                num_hidden_layers=args.enc_layers,
+                num_attention_heads=8,
+                intermediate_size=args.enc_ff_size,
+                hidden_dropout_prob=args.enc_dropout,
+                attention_probs_dropout_prob=args.enc_dropout,
+            )
+            self.bert.model = BertModel(bert_config)
+
+        self.vocab_size = self.bert.model.config.vocab_size
+
+        if args.max_pos > 512:
+            my_pos_embeddings = nn.Embedding(
+                args.max_pos, self.bert.model.config.hidden_size
+            )
+            my_pos_embeddings.weight.data[
+                :512
+            ] = self.bert.model.embeddings.position_embeddings.weight.data
+            my_pos_embeddings.weight.data[
+                512:
+            ] = self.bert.model.embeddings.position_embeddings.weight.data[-1][
+                None, :
+            ].repeat(
+                args.max_pos - 512, 1
+            )
+            self.bert.model.embeddings.position_embeddings = my_pos_embeddings
+        tgt_embeddings = nn.Embedding(
+            self.vocab_size, self.bert.model.config.hidden_size, padding_idx=0
+        )
+        if self.args.share_emb:
+            tgt_embeddings.weight = copy.deepcopy(
+                self.bert.model.embeddings.word_embeddings.weight
+            )
+
+        self.decoder = TransformerDecoder(
+            self.args.dec_layers,
+            self.args.dec_hidden_size,
+            heads=self.args.dec_heads,
+            d_ff=self.args.dec_ff_size,
+            dropout=self.args.dec_dropout,
+            embeddings=tgt_embeddings,
+            vocab_size=self.vocab_size,
+        )
+
+        gen_func = nn.LogSoftmax(dim=-1)
+        self.generator = nn.Sequential(
+            nn.Linear(args.dec_hidden_size, args.vocab_size), gen_func
+        )
+        self.generator[0].weight = self.decoder.embeddings.weight
+
+        load_from_checkpoints = False if checkpoint is None else True
+        if load_from_checkpoints:
+            self.load_state_dict(checkpoint)
+
+    def init_weights(self):
+        for module in self.decoder.modules():
+            if isinstance(module, (nn.Linear, nn.Embedding)):
+                module.weight.data.normal_(mean=0.0, std=0.02)
+            elif isinstance(module, nn.LayerNorm):
+                module.bias.data.zero_()
+                module.weight.data.fill_(1.0)
+            if isinstance(module, nn.Linear) and module.bias is not None:
+                module.bias.data.zero_()
+        for p in self.generator.parameters():
+            if p.dim() > 1:
+                xavier_uniform_(p)
+            else:
+                p.data.zero_()
+
+    def maybe_tie_embeddings(self, args):
+        if args.use_bert_emb:
+            tgt_embeddings = nn.Embedding(
+                self.vocab_size, self.bert.model.config.hidden_size, padding_idx=0
+            )
+            tgt_embeddings.weight = copy.deepcopy(
+                self.bert.model.embeddings.word_embeddings.weight
+            )
+            self.decoder.embeddings = tgt_embeddings
+
+    def forward(
+        self,
+        encoder_input_ids,
+        decoder_input_ids,
+        token_type_ids,
+        encoder_attention_mask,
+        decoder_attention_mask,
+    ):
+        encoder_output = self.bert(
+            input_ids=encoder_input_ids,
+            token_type_ids=token_type_ids,
+            attention_mask=encoder_attention_mask,
+        )
+        encoder_hidden_states = encoder_output[0]
+        dec_state = self.decoder.init_decoder_state(
+            encoder_input_ids, encoder_hidden_states
+        )
+        decoder_outputs, _ = self.decoder(
+            decoder_input_ids[:, :-1], encoder_hidden_states, dec_state
+        )
+        return decoder_outputs
+
+
+class Bert(nn.Module):
+    """ This class is not really necessary and should probably disappear.
+    """
+
+    def __init__(self, large, temp_dir, finetune=False):
+        super(Bert, self).__init__()
+        if large:
+            self.model = BertModel.from_pretrained("bert-large-uncased", cache_dir=temp_dir)
+        else:
+            self.model = BertModel.from_pretrained("bert-base-uncased", cache_dir=temp_dir)
+
+        self.finetune = finetune
+
+    def forward(self, input_ids, attention_mask=None, token_type_ids=None, **kwargs):
+        self.eval()
+        with torch.no_grad():
+            encoder_outputs, _ = self.model(
+                input_ids,
+                token_type_ids=token_type_ids,
+                attention_mask=attention_mask,
+                **kwargs
+            )
+        return encoder_outputs
+
+
+class TransformerDecoder(nn.Module):
+    """
+    The Transformer decoder from "Attention is All You Need".
+
+    Args:
+       num_layers (int): number of encoder layers.
+       d_model (int): size of the model
+       heads (int): number of heads
+       d_ff (int): size of the inner FF layer
+       dropout (float): dropout parameters
+       embeddings (:obj:`onmt.modules.Embeddings`):
+          embeddings to use, should have positional encodings
+       attn_type (str): if using a seperate copy attention
+    """
+
+    def __init__(self, num_layers, d_model, heads, d_ff, dropout, embeddings, vocab_size):
+        super(TransformerDecoder, self).__init__()
+
+        # Basic attributes.
+        self.decoder_type = "transformer"
+        self.num_layers = num_layers
+        self.embeddings = embeddings
+        self.pos_emb = PositionalEncoding(dropout, self.embeddings.embedding_dim)
+
+        # Build TransformerDecoder.
+        self.transformer_layers = nn.ModuleList(
+            [
+                TransformerDecoderLayer(d_model, heads, d_ff, dropout)
+                for _ in range(num_layers)
+            ]
+        )
+
+        self.layer_norm = nn.LayerNorm(d_model, eps=1e-6)
+
+    # forward(input_ids, attention_mask, encoder_hidden_states, encoder_attention_mask)
+    # def forward(self, input_ids, state, attention_mask=None, memory_lengths=None,
+    # step=None, cache=None, encoder_attention_mask=None, encoder_hidden_states=None, memory_masks=None):
+    def forward(
+        self,
+        input_ids,
+        encoder_hidden_states=None,
+        state=None,
+        attention_mask=None,
+        memory_lengths=None,
+        step=None,
+        cache=None,
+        encoder_attention_mask=None,
+    ):
+        """
+        See :obj:`onmt.modules.RNNDecoderBase.forward()`
+        memory_bank = encoder_hidden_states
+        """
+        # Name conversion
+        tgt = input_ids
+        memory_bank = encoder_hidden_states
+        memory_mask = encoder_attention_mask
+
+        # src_words = state.src
+        src_words = state.src
+        src_batch, src_len = src_words.size()
+
+        padding_idx = self.embeddings.padding_idx
+
+        # Decoder padding mask
+        tgt_words = tgt
+        tgt_batch, tgt_len = tgt_words.size()
+        tgt_pad_mask = (
+            tgt_words.data.eq(padding_idx).unsqueeze(1).expand(tgt_batch, tgt_len, tgt_len)
+        )
+
+        # Encoder padding mask
+        if memory_mask is not None:
+            src_len = memory_mask.size(-1)
+            src_pad_mask = memory_mask.expand(src_batch, tgt_len, src_len)
+        else:
+            src_pad_mask = (
+                src_words.data.eq(padding_idx)
+                .unsqueeze(1)
+                .expand(src_batch, tgt_len, src_len)
+            )
+
+        # Pass through the embeddings
+        emb = self.embeddings(input_ids)
+        output = self.pos_emb(emb, step)
+        assert emb.dim() == 3  # len x batch x embedding_dim
+
+        if state.cache is None:
+            saved_inputs = []
+
+        for i in range(self.num_layers):
+            prev_layer_input = None
+            if state.cache is None:
+                if state.previous_input is not None:
+                    prev_layer_input = state.previous_layer_inputs[i]
+
+            output, all_input = self.transformer_layers[i](
+                output,
+                memory_bank,
+                src_pad_mask,
+                tgt_pad_mask,
+                previous_input=prev_layer_input,
+                layer_cache=state.cache["layer_{}".format(i)]
+                if state.cache is not None
+                else None,
+                step=step,
+            )
+            if state.cache is None:
+                saved_inputs.append(all_input)
+
+        if state.cache is None:
+            saved_inputs = torch.stack(saved_inputs)
+
+        output = self.layer_norm(output)
+
+        if state.cache is None:
+            state = state.update_state(tgt, saved_inputs)
+
+        # Decoders in transformers return a tuple. Beam search will fail
+        # if we don't follow this convention.
+        return output, state  # , state
+
+    def init_decoder_state(self, src, memory_bank, with_cache=False):
+        """ Init decoder state """
+        state = TransformerDecoderState(src)
+        if with_cache:
+            state._init_cache(memory_bank, self.num_layers)
+        return state
+
+
+class PositionalEncoding(nn.Module):
+    def __init__(self, dropout, dim, max_len=5000):
+        pe = torch.zeros(max_len, dim)
+        position = torch.arange(0, max_len).unsqueeze(1)
+        div_term = torch.exp(
+            (torch.arange(0, dim, 2, dtype=torch.float) * -(math.log(10000.0) / dim))
+        )
+        pe[:, 0::2] = torch.sin(position.float() * div_term)
+        pe[:, 1::2] = torch.cos(position.float() * div_term)
+        pe = pe.unsqueeze(0)
+        super(PositionalEncoding, self).__init__()
+        self.register_buffer("pe", pe)
+        self.dropout = nn.Dropout(p=dropout)
+        self.dim = dim
+
+    def forward(self, emb, step=None):
+        emb = emb * math.sqrt(self.dim)
+        if step:
+            emb = emb + self.pe[:, step][:, None, :]
+
+        else:
+            emb = emb + self.pe[:, : emb.size(1)]
+        emb = self.dropout(emb)
+        return emb
+
+    def get_emb(self, emb):
+        return self.pe[:, : emb.size(1)]
+
+
+class TransformerDecoderLayer(nn.Module):
+    """
+    Args:
+      d_model (int): the dimension of keys/values/queries in
+                       MultiHeadedAttention, also the input size of
+                       the first-layer of the PositionwiseFeedForward.
+      heads (int): the number of heads for MultiHeadedAttention.
+      d_ff (int): the second-layer of the PositionwiseFeedForward.
+      dropout (float): dropout probability(0-1.0).
+      self_attn_type (string): type of self-attention scaled-dot, average
+    """
+
+    def __init__(self, d_model, heads, d_ff, dropout):
+        super(TransformerDecoderLayer, self).__init__()
+
+        self.self_attn = MultiHeadedAttention(heads, d_model, dropout=dropout)
+
+        self.context_attn = MultiHeadedAttention(heads, d_model, dropout=dropout)
+        self.feed_forward = PositionwiseFeedForward(d_model, d_ff, dropout)
+        self.layer_norm_1 = nn.LayerNorm(d_model, eps=1e-6)
+        self.layer_norm_2 = nn.LayerNorm(d_model, eps=1e-6)
+        self.drop = nn.Dropout(dropout)
+        mask = self._get_attn_subsequent_mask(MAX_SIZE)
+        # Register self.mask as a buffer in TransformerDecoderLayer, so
+        # it gets TransformerDecoderLayer's cuda behavior automatically.
+        self.register_buffer("mask", mask)
+
+    def forward(
+        self,
+        inputs,
+        memory_bank,
+        src_pad_mask,
+        tgt_pad_mask,
+        previous_input=None,
+        layer_cache=None,
+        step=None,
+    ):
+        """
+        Args:
+            inputs (`FloatTensor`): `[batch_size x 1 x model_dim]`
+            memory_bank (`FloatTensor`): `[batch_size x src_len x model_dim]`
+            src_pad_mask (`LongTensor`): `[batch_size x 1 x src_len]`
+            tgt_pad_mask (`LongTensor`): `[batch_size x 1 x 1]`
+
+        Returns:
+            (`FloatTensor`, `FloatTensor`, `FloatTensor`):
+
+            * output `[batch_size x 1 x model_dim]`
+            * attn `[batch_size x 1 x src_len]`
+            * all_input `[batch_size x current_step x model_dim]`
+
+        """
+        dec_mask = torch.gt(
+            tgt_pad_mask + self.mask[:, : tgt_pad_mask.size(1), : tgt_pad_mask.size(1)], 0
+        )
+        input_norm = self.layer_norm_1(inputs)
+        all_input = input_norm
+        if previous_input is not None:
+            all_input = torch.cat((previous_input, input_norm), dim=1)
+            dec_mask = None
+
+        query = self.self_attn(
+            all_input,
+            all_input,
+            input_norm,
+            mask=dec_mask,
+            layer_cache=layer_cache,
+            type="self",
+        )
+
+        query = self.drop(query) + inputs
+
+        query_norm = self.layer_norm_2(query)
+        mid = self.context_attn(
+            memory_bank,
+            memory_bank,
+            query_norm,
+            mask=src_pad_mask,
+            layer_cache=layer_cache,
+            type="context",
+        )
+        output = self.feed_forward(self.drop(mid) + query)
+
+        return output, all_input
+        # return output
+
+    def _get_attn_subsequent_mask(self, size):
+        """
+        Get an attention mask to avoid using the subsequent info.
+
+        Args:
+            size: int
+
+        Returns:
+            (`LongTensor`):
+
+            * subsequent_mask `[1 x size x size]`
+        """
+        attn_shape = (1, size, size)
+        subsequent_mask = np.triu(np.ones(attn_shape), k=1).astype("uint8")
+        subsequent_mask = torch.from_numpy(subsequent_mask)
+        return subsequent_mask
+
+
+class MultiHeadedAttention(nn.Module):
+    """
+    Multi-Head Attention module from
+    "Attention is All You Need"
+    :cite:`DBLP:journals/corr/VaswaniSPUJGKP17`.
+
+    Similar to standard `dot` attention but uses
+    multiple attention distributions simulataneously
+    to select relevant items.
+
+    .. mermaid::
+
+       graph BT
+          A[key]
+          B[value]
+          C[query]
+          O[output]
+          subgraph Attn
+            D[Attn 1]
+            E[Attn 2]
+            F[Attn N]
+          end
+          A --> D
+          C --> D
+          A --> E
+          C --> E
+          A --> F
+          C --> F
+          D --> O
+          E --> O
+          F --> O
+          B --> O
+
+    Also includes several additional tricks.
+
+    Args:
+       head_count (int): number of parallel heads
+       model_dim (int): the dimension of keys/values/queries,
+           must be divisible by head_count
+       dropout (float): dropout parameter
+    """
+
+    def __init__(self, head_count, model_dim, dropout=0.1, use_final_linear=True):
+        assert model_dim % head_count == 0
+        self.dim_per_head = model_dim // head_count
+        self.model_dim = model_dim
+
+        super(MultiHeadedAttention, self).__init__()
+        self.head_count = head_count
+
+        self.linear_keys = nn.Linear(model_dim, head_count * self.dim_per_head)
+        self.linear_values = nn.Linear(model_dim, head_count * self.dim_per_head)
+        self.linear_query = nn.Linear(model_dim, head_count * self.dim_per_head)
+        self.softmax = nn.Softmax(dim=-1)
+        self.dropout = nn.Dropout(dropout)
+        self.use_final_linear = use_final_linear
+        if self.use_final_linear:
+            self.final_linear = nn.Linear(model_dim, model_dim)
+
+    def forward(
+        self,
+        key,
+        value,
+        query,
+        mask=None,
+        layer_cache=None,
+        type=None,
+        predefined_graph_1=None,
+    ):
+        """
+        Compute the context vector and the attention vectors.
+
+        Args:
+           key (`FloatTensor`): set of `key_len`
+                key vectors `[batch, key_len, dim]`
+           value (`FloatTensor`): set of `key_len`
+                value vectors `[batch, key_len, dim]`
+           query (`FloatTensor`): set of `query_len`
+                 query vectors  `[batch, query_len, dim]`
+           mask: binary mask indicating which keys have
+                 non-zero attention `[batch, query_len, key_len]`
+        Returns:
+           (`FloatTensor`, `FloatTensor`) :
+
+           * output context vectors `[batch, query_len, dim]`
+           * one of the attention vectors `[batch, query_len, key_len]`
+        """
+        batch_size = key.size(0)
+        dim_per_head = self.dim_per_head
+        head_count = self.head_count
+        key_len = key.size(1)
+        query_len = query.size(1)
+
+        def shape(x):
+            """  projection """
+            return x.view(batch_size, -1, head_count, dim_per_head).transpose(1, 2)
+
+        def unshape(x):
+            """  compute context """
+            return (
+                x.transpose(1, 2)
+                .contiguous()
+                .view(batch_size, -1, head_count * dim_per_head)
+            )
+
+        # 1) Project key, value, and query.
+        if layer_cache is not None:
+            if type == "self":
+                query, key, value = (
+                    self.linear_query(query),
+                    self.linear_keys(query),
+                    self.linear_values(query),
+                )
+
+                key = shape(key)
+                value = shape(value)
+
+                if layer_cache is not None:
+                    device = key.device
+                    if layer_cache["self_keys"] is not None:
+                        key = torch.cat((layer_cache["self_keys"].to(device), key), dim=2)
+                    if layer_cache["self_values"] is not None:
+                        value = torch.cat(
+                            (layer_cache["self_values"].to(device), value), dim=2
+                        )
+                    layer_cache["self_keys"] = key
+                    layer_cache["self_values"] = value
+            elif type == "context":
+                query = self.linear_query(query)
+                if layer_cache is not None:
+                    if layer_cache["memory_keys"] is None:
+                        key, value = self.linear_keys(key), self.linear_values(value)
+                        key = shape(key)
+                        value = shape(value)
+                    else:
+                        key, value = (
+                            layer_cache["memory_keys"],
+                            layer_cache["memory_values"],
+                        )
+                    layer_cache["memory_keys"] = key
+                    layer_cache["memory_values"] = value
+                else:
+                    key, value = self.linear_keys(key), self.linear_values(value)
+                    key = shape(key)
+                    value = shape(value)
+        else:
+            key = self.linear_keys(key)
+            value = self.linear_values(value)
+            query = self.linear_query(query)
+            key = shape(key)
+            value = shape(value)
+
+        query = shape(query)
+
+        key_len = key.size(2)
+        query_len = query.size(2)
+
+        # 2) Calculate and scale scores.
+        query = query / math.sqrt(dim_per_head)
+        scores = torch.matmul(query, key.transpose(2, 3))
+
+        if mask is not None:
+            mask = mask.unsqueeze(1).expand_as(scores)
+            scores = scores.masked_fill(mask, -1e18)
+
+        # 3) Apply attention dropout and compute context vectors.
+
+        attn = self.softmax(scores)
+
+        if not predefined_graph_1 is None:
+            attn_masked = attn[:, -1] * predefined_graph_1
+            attn_masked = attn_masked / (torch.sum(attn_masked, 2).unsqueeze(2) + 1e-9)
+
+            attn = torch.cat([attn[:, :-1], attn_masked.unsqueeze(1)], 1)
+
+        drop_attn = self.dropout(attn)
+        if self.use_final_linear:
+            context = unshape(torch.matmul(drop_attn, value))
+            output = self.final_linear(context)
+            return output
+        else:
+            context = torch.matmul(drop_attn, value)
+            return context
+
+
+class DecoderState(object):
+    """Interface for grouping together the current state of a recurrent
+    decoder. In the simplest case just represents the hidden state of
+    the model.  But can also be used for implementing various forms of
+    input_feeding and non-recurrent models.
+
+    Modules need to implement this to utilize beam search decoding.
+    """
+
+    def detach(self):
+        """ Need to document this """
+        self.hidden = tuple([_.detach() for _ in self.hidden])
+        self.input_feed = self.input_feed.detach()
+
+    def beam_update(self, idx, positions, beam_size):
+        """ Need to document this """
+        for e in self._all:
+            sizes = e.size()
+            br = sizes[1]
+            if len(sizes) == 3:
+                sent_states = e.view(sizes[0], beam_size, br // beam_size, sizes[2])[
+                    :, :, idx
+                ]
+            else:
+                sent_states = e.view(
+                    sizes[0], beam_size, br // beam_size, sizes[2], sizes[3]
+                )[:, :, idx]
+
+            sent_states.data.copy_(sent_states.data.index_select(1, positions))
+
+    def map_batch_fn(self, fn):
+        raise NotImplementedError()
+
+
+class TransformerDecoderState(DecoderState):
+    """ Transformer Decoder state base class """
+
+    def __init__(self, src):
+        """
+        Args:
+            src (FloatTensor): a sequence of source words tensors
+                    with optional feature tensors, of size (len x batch).
+        """
+        self.src = src
+        self.previous_input = None
+        self.previous_layer_inputs = None
+        self.cache = None
+
+    @property
+    def _all(self):
+        """
+        Contains attributes that need to be updated in self.beam_update().
+        """
+        if self.previous_input is not None and self.previous_layer_inputs is not None:
+            return (self.previous_input, self.previous_layer_inputs, self.src)
+        else:
+            return (self.src,)
+
+    def detach(self):
+        if self.previous_input is not None:
+            self.previous_input = self.previous_input.detach()
+        if self.previous_layer_inputs is not None:
+            self.previous_layer_inputs = self.previous_layer_inputs.detach()
+        self.src = self.src.detach()
+
+    def update_state(self, new_input, previous_layer_inputs):
+        state = TransformerDecoderState(self.src)
+        state.previous_input = new_input
+        state.previous_layer_inputs = previous_layer_inputs
+        return state
+
+    def _init_cache(self, memory_bank, num_layers):
+        self.cache = {}
+
+        for l in range(num_layers):
+            layer_cache = {"memory_keys": None, "memory_values": None}
+            layer_cache["self_keys"] = None
+            layer_cache["self_values"] = None
+            self.cache["layer_{}".format(l)] = layer_cache
+
+    def repeat_beam_size_times(self, beam_size):
+        """ Repeat beam_size times along batch dimension. """
+        self.src = self.src.data.repeat(1, beam_size, 1)
+
+    def map_batch_fn(self, fn):
+        def _recursive_map(struct, batch_dim=0):
+            for k, v in struct.items():
+                if v is not None:
+                    if isinstance(v, dict):
+                        _recursive_map(v)
+                    else:
+                        struct[k] = fn(v, batch_dim)
+
+        self.src = fn(self.src, 0)
+        if self.cache is not None:
+            _recursive_map(self.cache)
+
+
+def gelu(x):
+    return (
+        0.5
+        * x
+        * (1 + torch.tanh(math.sqrt(2 / math.pi) * (x + 0.044715 * torch.pow(x, 3))))
+    )
+
+
+class PositionwiseFeedForward(nn.Module):
+    """ A two-layer Feed-Forward-Network with residual layer norm.
+
+    Args:
+        d_model (int): the size of input for the first-layer of the FFN.
+        d_ff (int): the hidden layer size of the second-layer
+            of the FNN.
+        dropout (float): dropout probability in :math:`[0, 1)`.
+    """
+
+    def __init__(self, d_model, d_ff, dropout=0.1):
+        super(PositionwiseFeedForward, self).__init__()
+        self.w_1 = nn.Linear(d_model, d_ff)
+        self.w_2 = nn.Linear(d_ff, d_model)
+        self.layer_norm = nn.LayerNorm(d_model, eps=1e-6)
+        self.actv = gelu
+        self.dropout_1 = nn.Dropout(dropout)
+        self.dropout_2 = nn.Dropout(dropout)
+
+    def forward(self, x):
+        inter = self.dropout_1(self.actv(self.w_1(self.layer_norm(x))))
+        output = self.dropout_2(self.w_2(inter))
+        return output + x
+
+
+#
+# TRANSLATOR
+# The following code is used to generate summaries using the
+# pre-trained weights and beam search.
+#
+
+
+def build_predictor(args, tokenizer, symbols, model, logger=None):
+    # we should be able to refactor the global scorer a lot
+    scorer = GNMTGlobalScorer(args.alpha, length_penalty="wu")
+    translator = Translator(
+        args, model, tokenizer, symbols, global_scorer=scorer, logger=logger
+    )
+    return translator
+
+
+class GNMTGlobalScorer(object):
+    """
+    NMT re-ranking score from
+    "Google's Neural Machine Translation System" :cite:`wu2016google`
+
+    Args:
+       alpha (float): length parameter
+       beta (float):  coverage parameter
+    """
+
+    def __init__(self, alpha, length_penalty):
+        self.alpha = alpha
+        penalty_builder = PenaltyBuilder(length_penalty)
+        self.length_penalty = penalty_builder.length_penalty()
+
+    def score(self, beam, logprobs):
+        """
+        Rescores a prediction based on penalty functions
+        """
+        normalized_probs = self.length_penalty(beam, logprobs, self.alpha)
+        return normalized_probs
+
+
+class PenaltyBuilder(object):
+    """
+    Returns the Length and Coverage Penalty function for Beam Search.
+
+    Args:
+        length_pen (str): option name of length pen
+        cov_pen (str): option name of cov pen
+    """
+
+    def __init__(self, length_pen):
+        self.length_pen = length_pen
+
+    def length_penalty(self):
+        if self.length_pen == "wu":
+            return self.length_wu
+        elif self.length_pen == "avg":
+            return self.length_average
+        else:
+            return self.length_none
+
+    """
+    Below are all the different penalty terms implemented so far
+    """
+
+    def length_wu(self, beam, logprobs, alpha=0.0):
+        """
+        NMT length re-ranking score from
+        "Google's Neural Machine Translation System" :cite:`wu2016google`.
+        """
+
+        modifier = ((5 + len(beam.next_ys)) ** alpha) / ((5 + 1) ** alpha)
+        return logprobs / modifier
+
+    def length_average(self, beam, logprobs, alpha=0.0):
+        """
+        Returns the average probability of tokens in a sequence.
+        """
+        return logprobs / len(beam.next_ys)
+
+    def length_none(self, beam, logprobs, alpha=0.0, beta=0.0):
+        """
+        Returns unmodified scores.
+        """
+        return logprobs
+
+
+class Translator(object):
+    """
+    Uses a model to translate a batch of sentences.
+
+    Args:
+       model (:obj:`onmt.modules.NMTModel`):
+          NMT model to use for translation
+       fields (dict of Fields): data fields
+       beam_size (int): size of beam to use
+       n_best (int): number of translations produced
+       max_length (int): maximum length output to produce
+       global_scores (:obj:`GlobalScorer`):
+         object to rescore final translations
+       copy_attn (bool): use copy attention during translation
+       cuda (bool): use cuda
+       beam_trace (bool): trace beam search for debugging
+       logger(logging.Logger): logger.
+    """
+
+    def __init__(self, args, model, vocab, symbols, global_scorer=None, logger=None):
+        self.logger = logger
+        self.cuda = args.visible_gpus != "-1"
+
+        self.args = args
+        self.model = model
+        self.generator = self.model.generator
+        self.vocab = vocab
+        self.symbols = symbols
+        self.start_token = symbols["BOS"]
+        self.end_token = symbols["EOS"]
+
+        self.global_scorer = global_scorer
+        self.beam_size = args.beam_size
+        self.min_length = args.min_length
+        self.max_length = args.max_length
+
+    def translate(self, batch, step, attn_debug=False):
+        """ Generates summaries from one batch of data.
+        """
+        self.model.eval()
+        with torch.no_grad():
+            batch_data = self.translate_batch(batch)
+            translations = self.from_batch(batch_data)
+        return translations
+
+    def translate_batch(self, batch, fast=False):
+        """
+        Translate a batch of sentences.
+
+        Mostly a wrapper around :obj:`Beam`.
+
+        Args:
+           batch (:obj:`Batch`): a batch from a dataset object
+           data (:obj:`Dataset`): the dataset object
+           fast (bool): enables fast beam search (may not support all features)
+
+        Todo:
+           Shouldn't need the original dataset.
+        """
+        with torch.no_grad():
+            return self._fast_translate_batch(
+                batch, self.max_length, min_length=self.min_length
+            )
+
+    # Where the beam search lives
+    # I have no idea why it is being called from the method above
+    def _fast_translate_batch(self, batch, max_length, min_length=0):
+        """ Beam Search using the encoder inputs contained in `batch`.
+        """
+
+        # The batch object is funny
+        # Instead of just looking at the size of the arguments we encapsulate
+        # a size argument.
+        # Where is it defined?
+        beam_size = self.beam_size
+        batch_size = batch.batch_size
+        src = batch.src
+        segs = batch.segs
+        mask_src = batch.mask_src
+
+        src_features = self.model.bert(src, segs, mask_src)
+        dec_states = self.model.decoder.init_decoder_state(
+            src, src_features, with_cache=True
+        )
+        device = src_features.device
+
+        # Tile states and memory beam_size times.
+        dec_states.map_batch_fn(lambda state, dim: tile(state, beam_size, dim=dim))
+        src_features = tile(src_features, beam_size, dim=0)
+        batch_offset = torch.arange(batch_size, dtype=torch.long, device=device)
+        beam_offset = torch.arange(
+            0, batch_size * beam_size, step=beam_size, dtype=torch.long, device=device
+        )
+        alive_seq = torch.full(
+            [batch_size * beam_size, 1], self.start_token, dtype=torch.long, device=device
+        )
+
+        # Give full probability to the first beam on the first step.
+        topk_log_probs = torch.tensor(
+            [0.0] + [float("-inf")] * (beam_size - 1), device=device
+        ).repeat(batch_size)
+
+        # Structure that holds finished hypotheses.
+        hypotheses = [[] for _ in range(batch_size)]  # noqa: F812
+
+        results = {}
+        results["predictions"] = [[] for _ in range(batch_size)]  # noqa: F812
+        results["scores"] = [[] for _ in range(batch_size)]  # noqa: F812
+        results["gold_score"] = [0] * batch_size
+        results["batch"] = batch
+
+        for step in range(max_length):
+            decoder_input = alive_seq[:, -1].view(1, -1)
+
+            # Decoder forward.
+            decoder_input = decoder_input.transpose(0, 1)
+
+            dec_out, dec_states = self.model.decoder(
+                decoder_input, src_features, dec_states, step=step
+            )
+
+            # Generator forward.
+            log_probs = self.generator.forward(dec_out.transpose(0, 1).squeeze(0))
+            vocab_size = log_probs.size(-1)
+
+            if step < min_length:
+                log_probs[:, self.end_token] = -1e20
+
+            # Multiply probs by the beam probability.
+            log_probs += topk_log_probs.view(-1).unsqueeze(1)
+
+            alpha = self.global_scorer.alpha
+            length_penalty = ((5.0 + (step + 1)) / 6.0) ** alpha
+
+            # Flatten probs into a list of possibilities.
+            curr_scores = log_probs / length_penalty
+
+            if self.args.block_trigram:
+                cur_len = alive_seq.size(1)
+                if cur_len > 3:
+                    for i in range(alive_seq.size(0)):
+                        fail = False
+                        words = [int(w) for w in alive_seq[i]]
+                        words = [self.vocab.ids_to_tokens[w] for w in words]
+                        words = " ".join(words).replace(" ##", "").split()
+                        if len(words) <= 3:
+                            continue
+                        trigrams = [
+                            (words[i - 1], words[i], words[i + 1])
+                            for i in range(1, len(words) - 1)
+                        ]
+                        trigram = tuple(trigrams[-1])
+                        if trigram in trigrams[:-1]:
+                            fail = True
+                        if fail:
+                            curr_scores[i] = -10e20
+
+            curr_scores = curr_scores.reshape(-1, beam_size * vocab_size)
+            topk_scores, topk_ids = curr_scores.topk(beam_size, dim=-1)
+
+            # Recover log probs.
+            topk_log_probs = topk_scores * length_penalty
+
+            # Resolve beam origin and true word ids.
+            topk_beam_index = topk_ids.div(vocab_size)
+            topk_ids = topk_ids.fmod(vocab_size)
+
+            # Map beam_index to batch_index in the flat representation.
+            batch_index = topk_beam_index + beam_offset[
+                : topk_beam_index.size(0)
+            ].unsqueeze(1)
+            select_indices = batch_index.view(-1)
+
+            # Append last prediction.
+            alive_seq = torch.cat(
+                [alive_seq.index_select(0, select_indices), topk_ids.view(-1, 1)], -1
+            )
+
+            is_finished = topk_ids.eq(self.end_token)
+            if step + 1 == max_length:
+                is_finished.fill_(1)
+            # End condition is top beam is finished.
+            end_condition = is_finished[:, 0].eq(1)
+            # Save finished hypotheses.
+            if is_finished.any():
+                predictions = alive_seq.view(-1, beam_size, alive_seq.size(-1))
+                for i in range(is_finished.size(0)):
+                    b = batch_offset[i]
+                    if end_condition[i]:
+                        is_finished[i].fill_(1)
+                    finished_hyp = is_finished[i].nonzero().view(-1)
+                    # Store finished hypotheses for this batch.
+                    for j in finished_hyp:
+                        hypotheses[b].append((topk_scores[i, j], predictions[i, j, 1:]))
+                    # If the batch reached the end, save the n_best hypotheses.
+                    if end_condition[i]:
+                        best_hyp = sorted(hypotheses[b], key=lambda x: x[0], reverse=True)
+                        score, pred = best_hyp[0]
+
+                        results["scores"][b].append(score)
+                        results["predictions"][b].append(pred)
+                non_finished = end_condition.eq(0).nonzero().view(-1)
+                # If all sentences are translated, no need to go further.
+                if len(non_finished) == 0:
+                    break
+                # Remove finished batches for the next step.
+                topk_log_probs = topk_log_probs.index_select(0, non_finished)
+                batch_index = batch_index.index_select(0, non_finished)
+                batch_offset = batch_offset.index_select(0, non_finished)
+                alive_seq = predictions.index_select(0, non_finished).view(
+                    -1, alive_seq.size(-1)
+                )
+            # Reorder states.
+            select_indices = batch_index.view(-1)
+            src_features = src_features.index_select(0, select_indices)
+            dec_states.map_batch_fn(
+                lambda state, dim: state.index_select(dim, select_indices)
+            )
+
+        return results
+
+    def from_batch(self, translation_batch):
+        batch = translation_batch["batch"]
+        assert len(translation_batch["gold_score"]) == len(translation_batch["predictions"])
+        batch_size = batch.batch_size
+
+        preds, _, _, tgt_str, src = (
+            translation_batch["predictions"],
+            translation_batch["scores"],
+            translation_batch["gold_score"],
+            batch.tgt_str,
+            batch.src,
+        )
+
+        translations = []
+        for b in range(batch_size):
+            pred_sents = self.vocab.convert_ids_to_tokens([int(n) for n in preds[b][0]])
+            pred_sents = " ".join(pred_sents).replace(" ##", "")
+            gold_sent = " ".join(tgt_str[b].split())
+            raw_src = [self.vocab.ids_to_tokens[int(t)] for t in src[b]][:500]
+            raw_src = " ".join(raw_src)
+            translation = (pred_sents, gold_sent, raw_src)
+            translations.append(translation)
+
+        return translations
+
+    def _report_rouge(self, gold_path, can_path):
+        self.logger.info("Calculating Rouge")
+        results_dict = test_rouge(self.args.temp_dir, can_path, gold_path)
+        return results_dict
+
+
+def tile(x, count, dim=0):
+    """
+    Tiles x on dimension dim count times.
+    """
+    perm = list(range(len(x.size())))
+    if dim != 0:
+        perm[0], perm[dim] = perm[dim], perm[0]
+        x = x.permute(perm).contiguous()
+    out_size = list(x.size())
+    out_size[0] *= count
+    batch = x.size(0)
+    x = (
+        x.view(batch, -1)
+        .transpose(0, 1)
+        .repeat(count, 1)
+        .transpose(0, 1)
+        .contiguous()
+        .view(*out_size)
+    )
+    if dim != 0:
+        x = x.permute(perm).contiguous()
+    return x
+
+
+#
+# All things ROUGE. Uses `pyrouge` which is a hot mess.
+#
+
+
+def test_rouge(temp_dir, cand, ref):
+    candidates = [line.strip() for line in open(cand, encoding="utf-8")]
+    references = [line.strip() for line in open(ref, encoding="utf-8")]
+    print(len(candidates))
+    print(len(references))
+    assert len(candidates) == len(references)
+
+    cnt = len(candidates)
+    current_time = time.strftime("%Y-%m-%d-%H-%M-%S", time.localtime())
+    tmp_dir = os.path.join(temp_dir, "rouge-tmp-{}".format(current_time))
+    if not os.path.isdir(tmp_dir):
+        os.mkdir(tmp_dir)
+        os.mkdir(tmp_dir + "/candidate")
+        os.mkdir(tmp_dir + "/reference")
+    try:
+
+        for i in range(cnt):
+            if len(references[i]) < 1:
+                continue
+            with open(
+                tmp_dir + "/candidate/cand.{}.txt".format(i), "w", encoding="utf-8"
+            ) as f:
+                f.write(candidates[i])
+            with open(
+                tmp_dir + "/reference/ref.{}.txt".format(i), "w", encoding="utf-8"
+            ) as f:
+                f.write(references[i])
+        r = pyrouge.Rouge155(temp_dir=temp_dir)
+        r.model_dir = tmp_dir + "/reference/"
+        r.system_dir = tmp_dir + "/candidate/"
+        r.model_filename_pattern = "ref.#ID#.txt"
+        r.system_filename_pattern = r"cand.(\d+).txt"
+        rouge_results = r.convert_and_evaluate()
+        print(rouge_results)
+        results_dict = r.output_to_dict(rouge_results)
+    finally:
+        pass
+        if os.path.isdir(tmp_dir):
+            shutil.rmtree(tmp_dir)
+    return results_dict
+
+
+def rouge_results_to_str(results_dict):
+    return ">> ROUGE-F(1/2/3/l): {:.2f}/{:.2f}/{:.2f}\nROUGE-R(1/2/3/l): {:.2f}/{:.2f}/{:.2f}\n".format(
+        results_dict["rouge_1_f_score"] * 100,
+        results_dict["rouge_2_f_score"] * 100,
+        results_dict["rouge_l_f_score"] * 100,
+        results_dict["rouge_1_recall"] * 100,
+        results_dict["rouge_2_recall"] * 100,
+        results_dict["rouge_l_recall"] * 100,
+    )
+
+
+class BertSumOptimizer(object):
+    """ Specific optimizer for BertSum.
+
+    As described in [1], the authors fine-tune BertSum for abstractive
+    summarization using two Adam Optimizers with different warm-up steps and
+    learning rate. They also use a custom learning rate scheduler.
+
+    [1] Liu, Yang, and Mirella Lapata. "Text summarization with pretrained encoders."
+        arXiv preprint arXiv:1908.08345 (2019).
+    """
+
+    def __init__(self, model, lr, warmup_steps, beta_1=0.99, beta_2=0.999, eps=1e-8):
+        self.encoder = model.encoder
+        self.decoder = model.decoder
+        self.lr = lr
+        self.warmup_steps = warmup_steps
+
+        self.optimizers = {
+            "encoder": torch.optim.Adam(
+                model.encoder.parameters(),
+                lr=lr["encoder"],
+                betas=(beta_1, beta_2),
+                eps=eps,
+            ),
+            "decoder": torch.optim.Adam(
+                model.decoder.parameters(),
+                lr=lr["decoder"],
+                betas=(beta_1, beta_2),
+                eps=eps,
+            ),
+        }
+
+        self._step = 0
+        self.current_learning_rates = {}
+
+    def _update_rate(self, stack):
+        return self.lr[stack] * min(
+            self._step ** (-0.5), self._step * self.warmup_steps[stack] ** (-1.5)
+        )
+
+    def zero_grad(self):
+        self.optimizer_decoder.zero_grad()
+        self.optimizer_encoder.zero_grad()
+
+    def step(self):
+        self._step += 1
+        for stack, optimizer in self.optimizers.items():
+            new_rate = self._update_rate(stack)
+            for param_group in optimizer.param_groups:
+                param_group["lr"] = new_rate
+            optimizer.step()
+            self.current_learning_rates[stack] = new_rate
diff --git a/examples/summarization/run_summarization.py b/examples/summarization/run_summarization.py
new file mode 100644
index 0000000000..e3b974acd9
--- /dev/null
+++ b/examples/summarization/run_summarization.py
@@ -0,0 +1,271 @@
+import argparse
+from collections import namedtuple
+import logging
+import os
+import sys
+
+import torch
+from torch.utils.data import DataLoader, SequentialSampler
+from tqdm import tqdm
+
+from transformers import BertTokenizer
+
+from modeling_bertabs import BertAbs, build_predictor
+
+from utils_summarization import (
+    SummarizationDataset,
+    encode_for_summarization,
+    build_mask,
+    fit_to_block_size,
+    compute_token_type_ids,
+)
+
+logger = logging.getLogger(__name__)
+logging.basicConfig(stream=sys.stdout, level=logging.INFO)
+
+
+Batch = namedtuple(
+    "Batch", ["document_names", "batch_size", "src", "segs", "mask_src", "tgt_str"]
+)
+
+
+def evaluate(args):
+    tokenizer = BertTokenizer.from_pretrained("bert-base-uncased", do_lower_case=True)
+    model = bertabs = BertAbs.from_pretrained(
+        "bertabs-finetuned-{}".format(args.finetuned_model)
+    )
+    bertabs.to(args.device)
+    bertabs.eval()
+
+    symbols = {
+        "BOS": tokenizer.vocab["[unused0]"],
+        "EOS": tokenizer.vocab["[unused1]"],
+        "PAD": tokenizer.vocab["[PAD]"],
+    }
+
+    # these (unused) arguments are defined to keep the compatibility
+    # with the legacy code and will be deleted in a next iteration.
+    args.result_path = ""
+    args.temp_dir = ""
+
+    data_iterator = build_data_iterator(args, tokenizer)
+    predictor = build_predictor(args, tokenizer, symbols, model)
+
+    logger.info("***** Running evaluation *****")
+    logger.info("  Number examples = %d", len(data_iterator.dataset))
+    logger.info("  Batch size = %d", args.batch_size)
+    logger.info("")
+    logger.info("***** Beam Search parameters *****")
+    logger.info("  Beam size = %d", args.beam_size)
+    logger.info("  Minimum length = %d", args.min_length)
+    logger.info("  Maximum length = %d", args.max_length)
+    logger.info("  Alpha (length penalty) = %.2f", args.alpha)
+    logger.info("  Trigrams %s be blocked", ("will" if args.block_trigram else "will NOT"))
+
+    for batch in tqdm(data_iterator):
+        batch_data = predictor.translate_batch(batch)
+        translations = predictor.from_batch(batch_data)
+        summaries = [format_summary(t) for t in translations]
+        save_summaries(summaries, args.summaries_output_dir, batch.document_names)
+
+
+def format_summary(translation):
+    """ Transforms the output of the `from_batch` function
+    into nicely formatted summaries.
+    """
+    raw_summary, _, _ = translation
+    summary = (
+        raw_summary.replace("[unused0]", "")
+        .replace("[unused3]", "")
+        .replace("[PAD]", "")
+        .replace("[unused1]", "")
+        .replace(r" +", " ")
+        .replace(" [unused2] ", ". ")
+        .replace("[unused2]", "")
+        .strip()
+    )
+
+    return summary
+
+
+def save_summaries(summaries, path, original_document_name):
+    """ Write the summaries in fies that are prefixed by the original
+    files' name with the `_summary` appended.
+
+    Attributes:
+        original_document_names: List[string]
+            Name of the document that was summarized.
+        path: string
+            Path were the summaries will be written
+        summaries: List[string]
+            The summaries that we produced.
+    """
+    for summary, document_name in zip(summaries, original_document_name):
+        # Prepare the summary file's name
+        if "." in document_name:
+            bare_document_name = ".".join(document_name.split(".")[:-1])
+            extension = document_name.split(".")[-1]
+            name = bare_document_name + "_summary." + extension
+        else:
+            name = document_name + "_summary"
+
+        file_path = os.path.join(path, name)
+        with open(file_path, "w") as output:
+            output.write(summary)
+
+
+#
+# LOAD the dataset
+#
+
+
+def build_data_iterator(args, tokenizer):
+    dataset = load_and_cache_examples(args, tokenizer)
+    sampler = SequentialSampler(dataset)
+    collate_fn = lambda data: collate(data, tokenizer, block_size=512)
+    iterator = DataLoader(
+        dataset, sampler=sampler, batch_size=args.batch_size, collate_fn=collate_fn,
+    )
+
+    return iterator
+
+
+def load_and_cache_examples(args, tokenizer):
+    dataset = SummarizationDataset(args.documents_dir)
+    return dataset
+
+
+def collate(data, tokenizer, block_size):
+    """ Collate formats the data passed to the data loader.
+
+    In particular we tokenize the data batch after batch to avoid keeping them
+    all in memory. We output the data as a namedtuple to fit the original BertAbs's
+    API.
+    """
+    data = [x for x in data if not len(x[1]) == 0]  # remove empty_files
+    names = [name for name, _, _ in data]
+
+    encoded_text = [
+        encode_for_summarization(story, summary, tokenizer) for _, story, summary in data
+    ]
+    stories = torch.tensor(
+        [
+            fit_to_block_size(story, block_size, tokenizer.pad_token_id)
+            for story, _ in encoded_text
+        ]
+    )
+    encoder_token_type_ids = compute_token_type_ids(stories, tokenizer.cls_token_id)
+    encoder_mask = build_mask(stories, tokenizer.pad_token_id)
+
+    batch = Batch(
+        document_names=names,
+        batch_size=len(stories),
+        src=stories,
+        segs=encoder_token_type_ids,
+        mask_src=encoder_mask,
+        tgt_str=[""] * len(stories),
+    )
+
+    return batch
+
+
+def decode_summary(summary_tokens, tokenizer):
+    """ Decode the summary and return it in a format
+    suitable for evaluation.
+    """
+    summary_tokens = summary_tokens.to("cpu").numpy()
+    summary = tokenizer.decode(summary_tokens)
+    sentences = summary.split(".")
+    sentences = [s + "." for s in sentences]
+    return sentences
+
+
+def main():
+    """ The main function defines the interface with the users.
+    """
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        "--documents_dir",
+        default=None,
+        type=str,
+        required=True,
+        help="The folder where the documents to summarize are located.",
+    )
+    parser.add_argument(
+        "--summaries_output_dir",
+        default=None,
+        type=str,
+        required=True,
+        help="The folder in wich the summaries should be written.",
+    )
+    # EVALUATION options
+    parser.add_argument(
+        "--visible_gpus",
+        default=-1,
+        type=int,
+        help="Number of GPUs with which to do the training.",
+    )
+    parser.add_argument(
+        "--batch_size", default=4, type=int, help="Batch size per GPU/CPU for training.",
+    )
+    # BEAM SEARCH arguments
+    parser.add_argument(
+        "--min_length",
+        default=50,
+        type=int,
+        help="Minimum number of tokens for the summaries.",
+    )
+    parser.add_argument(
+        "--max_length",
+        default=200,
+        type=int,
+        help="Maixmum number of tokens for the summaries.",
+    )
+    parser.add_argument(
+        "--beam_size",
+        default=5,
+        type=int,
+        help="The number of beams to start with for each example.",
+    )
+    parser.add_argument(
+        "--alpha",
+        default=0.95,
+        type=float,
+        help="The value of alpha for the length penalty in the beam search.",
+    )
+    parser.add_argument(
+        "--block_trigram",
+        default=True,
+        type=bool,
+        help="Whether to block the existence of repeating trigrams in the text generated by beam search.",
+    )
+    args = parser.parse_args()
+    args.device = torch.device("cpu") if args.visible_gpus == -1 else torch.device("cuda")
+
+    if not documents_dir_is_valid(args.documents_dir):
+        raise FileNotFoundError(
+            "We could not find the directory you specified for the documents to summarize, or it was empty. Please specify a valid path."
+        )
+    maybe_create_output_dir(args.summaries_output_dir)
+
+    evaluate(args)
+
+
+def documents_dir_is_valid(path):
+    if not os.path.exists(path):
+        return False
+
+    file_list = os.listdir(path)
+    if len(file_list) == 0:
+        return False
+
+    return True
+
+
+def maybe_create_output_dir(path):
+    if not os.path.exists(path):
+        os.makedirs(path)
+
+
+if __name__ == "__main__":
+    main()
diff --git a/examples/utils_summarization.py b/examples/summarization/utils_summarization.py
similarity index 77%
rename from examples/utils_summarization.py
rename to examples/summarization/utils_summarization.py
index 8e95a04e19..e7401b1754 100644
--- a/examples/utils_summarization.py
+++ b/examples/summarization/utils_summarization.py
@@ -10,9 +10,14 @@ from torch.utils.data import Dataset
 # ------------
 
 
-class CNNDailyMailDataset(Dataset):
+class SummarizationDataset(Dataset):
     """ Abstracts the dataset used to train seq2seq models.
 
+    The class will process the documents that are located in the specified
+    folder. The preprocessing will work on any document that is reasonably
+    formatted. On the CNN/DailyMail dataset it will extract both the story
+    and the summary.
+
     CNN/Daily News:
 
     The CNN/Daily News raw datasets are downloaded from [1]. The stories are
@@ -25,32 +30,31 @@ class CNNDailyMailDataset(Dataset):
     [2] https://github.com/abisee/cnn-dailymail/
     """
 
-    def __init__(self, data_dir="", prefix="train"):
-        assert os.path.isdir(data_dir)
+    def __init__(self, path="", prefix="train"):
+        """ We initialize the class by listing all the documents to summarize.
+        Files are not read in memory due to the size of some datasets (like CNN/DailyMail).
+        """
+        assert os.path.isdir(path)
 
-        # We initialize the class by listing all the files that contain
-        # stories and summaries. Files are not read in memory given
-        # the size of the corpus.
-        self.stories_path = []
-        datasets = ("cnn", "dailymail")
-        for dataset in datasets:
-            path_to_stories = os.path.join(data_dir, dataset, "stories")
-            story_filenames_list = os.listdir(path_to_stories)
-            for story_filename in story_filenames_list:
-                path_to_story = os.path.join(path_to_stories, story_filename)
-                if not os.path.isfile(path_to_story):
-                    continue
-                self.stories_path.append(path_to_story)
+        self.documents = []
+        story_filenames_list = os.listdir(path)
+        for story_filename in story_filenames_list:
+            path_to_story = os.path.join(path, story_filename)
+            if not os.path.isfile(path_to_story):
+                continue
+            self.documents.append(path_to_story)
 
     def __len__(self):
-        return len(self.stories_path)
+        """ Returns the number of documents. """
+        return len(self.documents)
 
     def __getitem__(self, idx):
-        story_path = self.stories_path[idx]
-        with open(story_path, encoding="utf-8") as source:
+        document_path = self.documents[idx]
+        document_name = document_path.split("/")[-1]
+        with open(document_path, encoding="utf-8") as source:
             raw_story = source.read()
             story_lines, summary_lines = process_story(raw_story)
-        return story_lines, summary_lines
+        return document_name, story_lines, summary_lines
 
 
 def process_story(raw_story):
@@ -80,7 +84,7 @@ def process_story(raw_story):
             story_lines.append(element)
         except IndexError:
             # if "@highlight" is absent from the file we pop
-            # all elements until there is None.
+            # all elements until there is None, raising an exception.
             return story_lines, []
 
     # gather summary lines
@@ -114,14 +118,6 @@ def fit_to_block_size(sequence, block_size, pad_token_id):
         return sequence
 
 
-def build_lm_labels(sequence, pad_token_id):
-    """ Padding token are replaced by the value -1 so they
-    are not taken into account in the loss computation. """
-    padded = sequence.clone()
-    padded[padded == pad_token_id] = -1
-    return padded
-
-
 def build_mask(sequence, pad_token_id):
     """ Builds the mask. The attention mechanism will only attend to positions
     with value 1. """
@@ -165,7 +161,7 @@ def compute_token_type_ids(batch, separator_token_id):
     """
     batch_embeddings = []
     for sequence in batch:
-        sentence_num = 0
+        sentence_num = -1
         embeddings = []
         for s in sequence:
             if s == separator_token_id:
diff --git a/examples/utils_summarization_test.py b/examples/summarization/utils_summarization_test.py
similarity index 88%
rename from examples/utils_summarization_test.py
rename to examples/summarization/utils_summarization_test.py
index 1d56ff0803..8bfbf6ab23 100644
--- a/examples/utils_summarization_test.py
+++ b/examples/summarization/utils_summarization_test.py
@@ -21,7 +21,6 @@ from utils_summarization import (
     compute_token_type_ids,
     fit_to_block_size,
     build_mask,
-    build_lm_labels,
     process_story,
 )
 
@@ -88,20 +87,6 @@ class SummarizationDataProcessingTest(unittest.TestCase):
         expected_summary_lines = ["It was the best of times."]
         self.assertEqual(expected_summary_lines, summary_lines)
 
-    def test_build_lm_labels_no_padding(self):
-        sequence = torch.tensor([1, 2, 3, 4])
-        expected = sequence
-        np.testing.assert_array_equal(
-            build_lm_labels(sequence, 0).numpy(), expected.numpy()
-        )
-
-    def test_build_lm_labels(self):
-        sequence = torch.tensor([1, 2, 3, 4, 0, 0, 0])
-        expected = torch.tensor([1, 2, 3, 4, -1, -1, -1])
-        np.testing.assert_array_equal(
-            build_lm_labels(sequence, 0).numpy(), expected.numpy()
-        )
-
     def test_build_mask_no_padding(self):
         sequence = torch.tensor([1, 2, 3, 4])
         expected = torch.tensor([1, 1, 1, 1])
@@ -125,7 +110,7 @@ class SummarizationDataProcessingTest(unittest.TestCase):
             [[1, 2, 3, 4, 5, 6], [1, 2, 3, 101, 5, 6], [1, 101, 3, 4, 101, 6]]
         )
         expected = torch.tensor(
-            [[0, 0, 0, 0, 0, 0], [0, 0, 0, 1, 1, 1], [0, 1, 1, 1, 0, 0]]
+            [[1, 1, 1, 1, 1, 1], [1, 1, 1, 0, 0, 0], [1, 0, 0, 0, 1, 1]]
         )
 
         result = compute_token_type_ids(batch, separator)
diff --git a/transformers/convert_bertextabs_original_pytorch_checkpoint_to_pytorch.py b/transformers/convert_bertextabs_original_pytorch_checkpoint_to_pytorch.py
new file mode 100644
index 0000000000..4f158966e1
--- /dev/null
+++ b/transformers/convert_bertextabs_original_pytorch_checkpoint_to_pytorch.py
@@ -0,0 +1,158 @@
+# coding=utf-8
+# Copyright 2018 The HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" Convert BertExtAbs's checkpoints """
+
+import argparse
+from collections import namedtuple
+import logging
+
+import torch
+
+from models.model_builder import AbsSummarizer  # The authors' implementation
+
+from transformers import BertConfig, Model2Model, BertModel, BertForMaskedLM
+
+
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+
+BertExtAbsConfig = namedtuple(
+    "BertExtAbsConfig",
+    ["temp_dir", "large", "finetune_bert", "encoder", "share_emb", "max_pos", "enc_layers", "enc_hidden_size", "enc_heads", "enc_ff_size", "enc_dropout", "dec_layers", "dec_hidden_size", "dec_heads", "dec_ff_size", "dec_dropout"],
+)
+
+
+def convert_bertextabs_checkpoints(path_to_checkpoints, dump_path):
+    """ Copy/paste and tweak the pre-trained weights provided by the creators
+    of BertExtAbs for the internal architecture.
+    """
+
+    # Load checkpoints in memory
+    checkpoints = torch.load(path_to_checkpoints, lambda storage, loc: storage)
+
+    # Instantiate the authors' model with the pre-trained weights
+    config = BertExtAbsConfig(
+        temp_dir=".",
+        finetune_bert=False,
+        large=False,
+        share_emb=True,
+        encoder="bert",
+        max_pos=512,
+        enc_layers=6,
+        enc_hidden_size=512,
+        enc_heads=8,
+        enc_ff_size=512,
+        enc_dropout=0.2,
+        dec_layers=6,
+        dec_hidden_size=768,
+        dec_heads=8,
+        dec_ff_size=2048,
+        dec_dropout=0.2,
+    )
+    bertextabs = AbsSummarizer(config, torch.device("cpu"), checkpoints)
+    bertextabs.eval()
+
+    # Instantiate our version of the model
+    decoder_config = BertConfig(
+        hidden_size=config.dec_hidden_size,
+        num_hidden_layers=config.dec_layers,
+        num_attention_heads=config.dec_heads,
+        intermediate_size=config.dec_ff_size,
+        hidden_dropout_prob=config.dec_dropout,
+        attention_probs_dropout_prob=config.dec_dropout,
+        is_decoder=True,
+    )
+
+    decoder_model = BertForMaskedLM(decoder_config)
+    model = Model2Model.from_pretrained('bert-base-uncased', decoder_model=decoder_model)
+    model.eval()
+
+    # Let us now start the weight copying process
+    model.encoder.load_state_dict(bertextabs.bert.model.state_dict())
+
+    # Decoder
+
+    # Embeddings. The positional embeddings are equal to the word embedding plus a modulation
+    # that is computed at each forward pass. This may be a source of discrepancy.
+    model.decoder.bert.embeddings.word_embeddings.weight = bertextabs.decoder.embeddings.weight
+    model.decoder.bert.embeddings.position_embeddings.weight = bertextabs.decoder.embeddings.weight
+    model.decoder.bert.embeddings.token_type_embeddings.weight.data = torch.zeros_like(bertextabs.decoder.embeddings.weight)  # not defined for BertExtAbs decoder
+
+    # In the original code the LayerNorms are applied twice in the layers, at the beginning and between the
+    # attention layers.
+    model.decoder.bert.embeddings.LayerNorm.weight = bertextabs.decoder.transformer_layers[0].layer_norm_1.weight
+
+    for i in range(config.dec_layers):
+
+        # self attention
+        model.decoder.bert.encoder.layer[i].attention.self.query.weight = bertextabs.decoder.transformer_layers[i].self_attn.linear_query.weight
+        model.decoder.bert.encoder.layer[i].attention.self.key.weight = bertextabs.decoder.transformer_layers[i].self_attn.linear_keys.weight
+        model.decoder.bert.encoder.layer[i].attention.self.value.weight = bertextabs.decoder.transformer_layers[i].self_attn.linear_values.weight
+        model.decoder.bert.encoder.layer[i].attention.output.dense.weight = bertextabs.decoder.transformer_layers[i].self_attn.final_linear.weight
+        model.decoder.bert.encoder.layer[i].attention.output.LayerNorm.weight = bertextabs.decoder.transformer_layers[i].layer_norm_2.weight
+
+        # attention
+        model.decoder.bert.encoder.layer[i].crossattention.self.query.weight = bertextabs.decoder.transformer_layers[i].context_attn.linear_query.weight
+        model.decoder.bert.encoder.layer[i].crossattention.self.key.weight = bertextabs.decoder.transformer_layers[i].context_attn.linear_keys.weight
+        model.decoder.bert.encoder.layer[i].crossattention.self.value.weight = bertextabs.decoder.transformer_layers[i].context_attn.linear_values.weight
+        model.decoder.bert.encoder.layer[i].crossattention.output.dense.weight = bertextabs.decoder.transformer_layers[i].context_attn.final_linear.weight
+        model.decoder.bert.encoder.layer[i].crossattention.output.LayerNorm.weight = bertextabs.decoder.transformer_layers[i].feed_forward.layer_norm.weight
+
+        # intermediate
+        model.decoder.bert.encoder.layer[i].intermediate.dense.weight = bertextabs.decoder.transformer_layers[i].feed_forward.w_1.weight
+
+        # output
+        model.decoder.bert.encoder.layer[i].output.dense.weight = bertextabs.decoder.transformer_layers[i].feed_forward.w_2.weight
+
+        try:
+            model.decoder.bert.encoder.layer[i].output.LayerNorm.weight = bertextabs.decoder.transformer_layers[i + 1].layer_norm_1.weight
+        except IndexError:
+            model.decoder.bert.encoder.layer[i].output.LayerNorm.weight = bertextabs.decoder.layer_norm.weight
+
+    # LM Head
+    """
+    model.decoder.cls.predictions.transform.dense.weight
+    model.decoder.cls.predictions.transform.dense.biais
+    model.decoder.cls.predictions.transform.LayerNorm.weight
+    model.decoder.cls.predictions.transform.LayerNorm.biais
+    model.decoder.cls.predictions.decoder.weight
+    model.decoder.cls.predictions.decoder.biais
+    model.decoder.cls.predictions.biais.data
+    """
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        "--bertextabs_checkpoint_path",
+        default=None,
+        type=str,
+        required=True,
+        help="Path the official PyTorch dump.",
+    )
+    parser.add_argument(
+        "--pytorch_dump_folder_path",
+        default=None,
+        type=str,
+        required=True,
+        help="Path to the output PyTorch model.",
+    )
+    args = parser.parse_args()
+
+    convert_bertextabs_checkpoints(
+        args.bertextabs_checkpoint_path,
+        args.pytorch_dump_folder_path,
+    )
diff --git a/transformers/generate/beam_search.py b/transformers/generate/beam_search.py
index abe3186049..b56ebbabb8 100644
--- a/transformers/generate/beam_search.py
+++ b/transformers/generate/beam_search.py
@@ -25,7 +25,6 @@ Use Beam Search to generate sequences using encoder-decoder models.
 """
 import torch
 from torch import nn
-
 import logging
 
 
@@ -45,6 +44,7 @@ class BeamSearch(object):
         max_length,
         alpha=0,
         block_repeating_trigrams=True,
+        device=torch.device("cpu"),
     ):
         r"""
         Inputs:
@@ -156,18 +156,24 @@ class BeamSearch(object):
         kwargs_decoder["encoder_hidden_states"] = tile(
             encoder_hidden_states, self.beam_size, dim=0
         )
-        kwargs_decoder["encoder_attention_mask"] = tile(
-            kwargs_encoder["attention_mask"], self.beam_size, dim=0
+        try:
+            kwargs_decoder["encoder_attention_mask"] = tile(
+                kwargs_encoder["attention_mask"], self.beam_size, dim=0
+            )
+        except:
+            pass
+        kwargs_decoder["state"].src = tile(
+            kwargs_decoder["state"].src, self.beam_size, dim=0
         )
 
         # grow the beam iteratively
         batch_size, block_size = encoder_input_ids.size()
         self._init_beam_state(batch_size)
         for step in range(self.max_length):
-
             decoder_input = fit_to_block_size(self.growing_beams, block_size, self.pad_token_id)
             kwargs_decoder["attention_mask"] = build_mask(decoder_input, self.pad_token_id)
-            outputs = self.model.decoder(decoder_input, **kwargs_decoder)
+
+            outputs, state = self.model.decoder(decoder_input, **kwargs_decoder)
 
             next_token_scores = outputs[0][:, -1, :].squeeze(1)
             log_probabilities = torch.nn.functional.log_softmax(next_token_scores, dim=0)
@@ -178,9 +184,13 @@ class BeamSearch(object):
             kwargs_decoder["encoder_hidden_states"] = kwargs_decoder[
                 "encoder_hidden_states"
             ].index_select(0, surviving_beams_rows)
-            kwargs_decoder["encoder_attention_mask"] = kwargs_decoder[
-                "encoder_attention_mask"
-            ].index_select(0, surviving_beams_rows)
+            try:
+                kwargs_decoder["encoder_attention_mask"] = kwargs_decoder[
+                    "encoder_attention_mask"
+                ].index_select(0, surviving_beams_rows)
+            except:
+                pass
+            kwargs_decoder["state"] = state
 
         return self.results
 

From c0443df5939d980abbe5bb28b31f08d1628469c6 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Thu, 5 Dec 2019 18:13:41 +0100
Subject: [PATCH 273/505] remove beam search

---
 transformers/generate/beam_search.py    | 376 ------------------------
 transformers/tests/beam_search_tests.py | 243 ---------------
 2 files changed, 619 deletions(-)
 delete mode 100644 transformers/generate/beam_search.py
 delete mode 100644 transformers/tests/beam_search_tests.py

diff --git a/transformers/generate/beam_search.py b/transformers/generate/beam_search.py
deleted file mode 100644
index b56ebbabb8..0000000000
--- a/transformers/generate/beam_search.py
+++ /dev/null
@@ -1,376 +0,0 @@
-# coding=utf-8
-# MIT License
-
-# Copyright (c) 2017-Present OpenNMT
-
-# Permission is hereby granted, free of charge, to any person obtaining a copy of
-# this software and associated documentation files (the "Software"), to deal in
-# the Software without restriction, including without limitation the rights to
-# use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies
-# of the Software, and to permit persons to whom the Software is furnished to do
-# so, subject to the following conditions:
-
-# The above copyright notice and this permission notice shall be included in all
-# copies or substantial portions of the Software.
-
-# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
-# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
-# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
-# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
-# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
-# SOFTWARE.
-"""
-Use Beam Search to generate sequences using encoder-decoder models.
-"""
-import torch
-from torch import nn
-import logging
-
-
-logger = logging.getLogger(__name__)
-
-
-class BeamSearch(object):
-    def __init__(
-        self,
-        model,
-        bos_token_id,
-        pad_token_id,
-        eos_token_id,
-        batch_size,
-        beam_size,
-        min_length,
-        max_length,
-        alpha=0,
-        block_repeating_trigrams=True,
-        device=torch.device("cpu"),
-    ):
-        r"""
-        Inputs:
-            **model**: instance of ``transformers.PreTrainedEncoderDecoder``
-                The pretrained encoder-decoder model that will be used to generate the sequences.
-            **bos_token_id**: int
-                Id that is used by the tokenizer to represent the beggining of a sentence.
-            **pad_token_id**: int
-                Id that is used by the tokenizer for padding.
-            **eos_token_id**: int
-                Id that is used by the tokenizer to represent the end of a sentence.
-            **batch_size**: (`optional`) int
-                Batch size of the inputs. The value is set automatically when calling `forward`.
-            **beam_size**: int
-                Number of beams that are used for each element on the batch.
-            **min_length**: int
-                Minimum number of steps performed by the beam search before terminating.
-            **max_length**: int
-                Maximum number of steps performed by the beam search. Any beam that has not finished
-                will return its current solution with the highest probability. The sequence that is
-                returned has a length of max_length-1 to account for the end token that is subsequently added.
-            **alpha**: float
-                Parameter of the length penalty. Read the documentation of the `_length_penalty` method for mode details.
-            **block_repeating_trigrams**: bool
-                Whether to block sequences that have repeating 3-grams.
-        """
-        super(BeamSearch, self).__init__()
-        self.model = model
-        self.device = next(model.parameters()).device  # only works if all parameters of the model are stored on a single GPU
-
-        self.bos_token_id = bos_token_id
-        self.eos_token_id = eos_token_id
-        self.pad_token_id = pad_token_id
-
-        self.batch_size = batch_size
-        self.beam_size = beam_size
-        self.min_length = min_length
-        self.max_length = max_length
-
-        self.block_repeating_trigram = block_repeating_trigrams
-        self.apply_length_penalty = False if alpha == 0 else True
-        self.alpha = alpha
-
-        self._init_beam_state(batch_size)
-
-    def __len__(self):
-        return self.growing_beams.size(1)
-
-    def _init_beam_state(self, batch_size):
-        """ (re-)Initialize the state of the beams. """
-        self.hypotheses = [[] for _ in range(batch_size)]
-        self.batch_offset = torch.arange(batch_size, dtype=torch.long, device=self.device)
-        self.beam_offset = torch.arange(
-            0,
-            batch_size * self.beam_size,
-            step=self.beam_size,
-            dtype=torch.long,
-            device=self.device,
-        )
-        self.growing_beams = torch.full(
-            (batch_size * self.beam_size, 1),
-            self.bos_token_id,
-            dtype=torch.long,
-            device=self.device,
-        )
-        self.topk_log_probabilities = torch.tensor(
-            [0.0] + [float("-inf")] * (self.beam_size - 1),
-            dtype=torch.float,
-            device=self.device,
-        ).repeat(batch_size)
-        self.results = {
-            "predictions": [[] for _ in range(batch_size)],
-            "scores": [[] for _ in range(batch_size)],
-        }
-        self._step = 0
-        self.is_done = False
-
-    def __call__(self, encoder_input_ids, **model_kwargs):
-        """ Generate a sequence using Beam Search. """
-        # keyword arguments come in 3 flavors: encoder-specific (prefixed by
-        # `encoder_`), decoder-specific (prefixed by `decoder_`) and those
-        # that apply to the model as whole.
-        # We let the specific kwargs override the common ones in case of conflict.
-        kwargs_common = {
-            argument: value
-            for argument, value in model_kwargs.items()
-            if not argument.startswith("encoder_") and not argument.startswith("decoder_")
-        }
-        kwargs_decoder = kwargs_common.copy()
-        kwargs_encoder = kwargs_common.copy()
-        kwargs_encoder.update(
-            {
-                argument[len("encoder_") :]: value
-                for argument, value in model_kwargs.items()
-                if argument.startswith("encoder_")
-            }
-        )
-        kwargs_decoder.update(
-            {
-                argument[len("decoder_") :]: value
-                for argument, value in model_kwargs.items()
-                if argument.startswith("decoder_")
-            }
-        )
-
-        # forward pass on the encoder
-        encoder_outputs = self.model.encoder(encoder_input_ids, **kwargs_encoder)
-        encoder_hidden_states = encoder_outputs[0]
-        kwargs_decoder["encoder_hidden_states"] = tile(
-            encoder_hidden_states, self.beam_size, dim=0
-        )
-        try:
-            kwargs_decoder["encoder_attention_mask"] = tile(
-                kwargs_encoder["attention_mask"], self.beam_size, dim=0
-            )
-        except:
-            pass
-        kwargs_decoder["state"].src = tile(
-            kwargs_decoder["state"].src, self.beam_size, dim=0
-        )
-
-        # grow the beam iteratively
-        batch_size, block_size = encoder_input_ids.size()
-        self._init_beam_state(batch_size)
-        for step in range(self.max_length):
-            decoder_input = fit_to_block_size(self.growing_beams, block_size, self.pad_token_id)
-            kwargs_decoder["attention_mask"] = build_mask(decoder_input, self.pad_token_id)
-
-            outputs, state = self.model.decoder(decoder_input, **kwargs_decoder)
-
-            next_token_scores = outputs[0][:, -1, :].squeeze(1)
-            log_probabilities = torch.nn.functional.log_softmax(next_token_scores, dim=0)
-            surviving_beams_rows = self.grow(log_probabilities)
-            if self.is_done:
-                break
-
-            kwargs_decoder["encoder_hidden_states"] = kwargs_decoder[
-                "encoder_hidden_states"
-            ].index_select(0, surviving_beams_rows)
-            try:
-                kwargs_decoder["encoder_attention_mask"] = kwargs_decoder[
-                    "encoder_attention_mask"
-                ].index_select(0, surviving_beams_rows)
-            except:
-                pass
-            kwargs_decoder["state"] = state
-
-        return self.results
-
-    def grow(self, log_probabilities):
-        """ Grow the beams by one step. """
-        self._step += 1
-
-        # The number of beams changes as some beams finish so we define _B
-        vocab_size = log_probabilities.size(-1)
-        _B = log_probabilities.size(0) // self.beam_size
-
-        # Multiply each beam probability with the probability of the
-        # next token (conditioned on the words in the beam).
-        log_probabilities += self.topk_log_probabilities.view(-1, 1)
-
-        self._enforce_min_length(log_probabilities)
-        if self.block_repeating_trigram:
-            self._remove_beams_with_repeating_trigrams(log_probabilities, _B)
-
-        # Find the `beam_size` (previous_beam + token) combinations with
-        # the highest score
-        self.topk_log_probabilities, topk_ids = torch.topk(
-            log_probabilities.view(_B, self.beam_size * vocab_size), self.beam_size, dim=1
-        )
-
-        # Apply the length penalty. The +1 accounts for the [EOS] token
-        # that will be added if the beam ends.
-        topk_scores = self.topk_log_probabilities
-        if self.apply_length_penalty:
-            topk_scores /= self._length_penalty()
-
-        # Retrieve the corresponding respective beam and token id
-        # topk_token_ids[i] will be added to topk_beam_ids[i]
-        topk_beam_ids = topk_ids.div(vocab_size)
-        topk_token_ids = topk_ids.fmod(vocab_size)
-
-        # Retrieve the row index of the surviving beams in the original
-        # view of the log_probabilities tensor
-        surviving_beams_per_batch = topk_beam_ids + self.beam_offset[:_B].view(-1, 1)
-        surviving_beams_rows = surviving_beams_per_batch.view(-1)
-
-        # Append the last predictions
-        self.growing_beams = torch.cat(
-            [
-                self.growing_beams.index_select(0, surviving_beams_rows),
-                topk_token_ids.view(-1, 1),
-            ],
-            1,
-        )
-
-        # Check if any of the beam searches has ended during this
-        # growth step. Also if top beam (most probable) has ended
-        # for one element of the batch.
-        is_finished = topk_token_ids.eq(self.eos_token_id)
-        self._enforce_max_length(is_finished)
-        if is_finished.any():
-            non_finished = self._cut_finished(is_finished, topk_scores)
-            self.batch_offset = self.batch_offset.index_select(0, non_finished)
-            surviving_beams_per_batch = surviving_beams_per_batch.index_select(
-                0, non_finished
-            )
-            self.topk_log_probabilities = self.topk_log_probabilities.index_select(
-                0, non_finished
-            )
-
-            surviving_beams_rows = surviving_beams_per_batch.view(-1)
-            self.growing_beams = self.growing_beams.index_select(0, surviving_beams_rows)
-
-        return surviving_beams_rows
-
-    def _cut_finished(self, is_finished, topk_scores):
-        """ Save the finished searches and cut the correponding sequences off
-        the beams. """
-        is_top_beam_finished = is_finished[:, 0].eq(True)
-
-        # Save the finished searches
-        predictions = self.growing_beams.view(
-            -1, self.beam_size, self.growing_beams.size(1)
-        )
-        for i in range(is_finished.size(0)):
-            if is_top_beam_finished[i]:
-                is_finished[i].fill_(1)
-            finished_hyp = is_finished[i].nonzero().view(-1)
-
-            # Store the finished beams as a (score, prediction) hypothesis.
-            b = self.batch_offset[i]
-            for j in finished_hyp:
-                self.hypotheses[b].append((topk_scores[i, j], predictions[i, j, :]))
-
-            # If the batch reached the end, save the best hypotheses
-            # in terms of length-penalized score.
-            if is_top_beam_finished[i]:
-                best_score, best_prediction = max(self.hypotheses[b], key=lambda x: x[0])
-                self.results["scores"][b].append(best_score)
-                self.results["predictions"][b].append(best_prediction)
-
-        non_finished = is_top_beam_finished.eq(False).nonzero().view(-1)
-        if len(non_finished) == 0:
-            self.is_done = True
-
-        return non_finished
-
-    def _remove_beams_with_repeating_trigrams(self, log_probabilities, _B):
-        if self._step + 1 > 3:  # [BOS] does not count
-            for i in range(_B * self.beam_size):
-                tokens = self.growing_beams[i]
-                trigrams = [
-                    (tokens[j - 1], tokens[j], tokens[j + 1])
-                    for j in range(1, len(self) - 1)
-                ]
-                last_trigram = tuple(trigrams[-1])
-                if last_trigram in trigrams[:-1]:
-                    log_probabilities[i] = -1e20
-
-    def _enforce_min_length(self, log_probabilities):
-        if self._step < self.min_length:
-            log_probabilities[:, self.eos_token_id] = -1e20
-
-    def _enforce_max_length(self, is_finished):
-        # +1 because we will need to add an [EOS] token
-        if self._step + 1 == self.max_length:
-            is_finished.fill_(1)
-
-    def _length_penalty(self):
-        """ The calculation of the length penalty follows that of [1].
-
-        [1] Wu, Yonghui, et al. "Google's neural machine translation system:
-        Bridging the gap between human and machine translation." arXiv preprint
-        arXiv:1609.08144 (2016).
-        """
-        return ((5.0 + (self._step + 1)) / 6.0) ** self.alpha
-
-
-def tile(x, count, dim=0):
-    """
-    Tiles `x` along dimension `dim` `count` times.
-
-    Example:
-        >> ex = torch.tensor([1,2],[3,4])
-        >> tile(ex, 2, 0)
-        torch.Tensor([[1,2],[1,2],[3,4],[3,4]])
-    """
-    perm = list(range(len(x.size())))
-    if dim != 0:
-        perm[0], perm[dim] = perm[dim], perm[0]
-        x = x.permute(perm).contiguous()
-    out_size = list(x.size())
-    out_size[0] *= count
-    batch = x.size(0)
-    x = (
-        x.view(batch, -1)
-        .transpose(0, 1)
-        .repeat(count, 1)
-        .transpose(0, 1)
-        .contiguous()
-        .view(*out_size)
-    )
-    if dim != 0:
-        x = x.permute(perm).contiguous()
-    return x
-
-
-def fit_to_block_size(sequence, block_size, pad_token_id):
-    """ Adapt the source and target sequences' lengths to the block size.
-    If the sequence is shorter we append padding tokens to the right.
-    """
-    padded_sequence = torch.full(
-        (sequence.size(0), block_size),
-        pad_token_id,
-        dtype=torch.long,
-        device=sequence.device,
-    )
-    padded_sequence[:, : sequence.size(1)] = sequence
-    return sequence
-
-
-def build_mask(sequence, pad_token_id):
-    """ Builds the mask. The attention mechanism will only attend to positions
-    with value 1. """
-    mask = torch.ones_like(sequence)
-    idx_pad_tokens = sequence == pad_token_id
-    mask[idx_pad_tokens] = 0
-    return mask
diff --git a/transformers/tests/beam_search_tests.py b/transformers/tests/beam_search_tests.py
deleted file mode 100644
index 6f2a2b9c2f..0000000000
--- a/transformers/tests/beam_search_tests.py
+++ /dev/null
@@ -1,243 +0,0 @@
-from collections import namedtuple
-import unittest
-import pytest
-import numpy as np
-import torch
-from torch import nn
-
-from transformers.generate import BeamSearch
-from transformers import PreTrainedEncoderDecoder
-
-
-class StubTransformer(nn.Module):
-    def __init__(self):
-        self.encoder = None
-        self.decoder = None
-        self._parameters = {"dumy": torch.tensor([1])}
-
-    def forward(self):
-        pass
-
-
-class BeamSearchtest(unittest.TestCase):
-    def test_beam_search_encoder_decoder_integration(self):
-        """ We make sure that no internal change in the PreTrainedEncoderDecoder
-        class will break the integration with the beam search.
-        """
-
-        model = StubTransformer()
-        try:
-            _ = BeamSearch(
-                model=model,
-                bos_token_id=0,
-                eos_token_id=1,
-                pad_token_id=2,
-                batch_size=1,
-                beam_size=1,
-                min_length=1,
-                max_length=1,
-                alpha=0,
-                block_repeating_trigrams=False,
-            )
-        except:
-            self.fail("Instantiating BeamSearch with a PreTrainedEncoderDecoder failed.")
-
-    def test_beam_search_min_length(self):
-        """ We keep predicting the end_token for the first beam and check that
-        it is not marked as finished until the beam has reached the minimum
-        length. """
-        eos_idx = 3
-        vocab_size = 10
-
-        batch_size = 3
-        beam_size = 2
-        min_length = 5
-
-        beam = BeamSearch(
-            model=StubTransformer(),
-            bos_token_id=0,
-            eos_token_id=eos_idx,
-            pad_token_id=2,
-            batch_size=batch_size,
-            beam_size=beam_size,
-            min_length=5,
-            max_length=10,
-            alpha=0,
-            block_repeating_trigrams=False,
-        )
-
-        # To test that the minimum length is correctly enforced we constantly
-        # assign the highest probability to the [EOS] token (and assign lower
-        # probabilities to some other tokens).
-        # Since BeamSearch will reset its probability to 1e-20 as long as
-        # min_length has not been reached, we need to reset the value between
-        # steps.
-        non_eos_idxs = [4, 5, 1, 8, 9]
-        score_distribution = torch.log_softmax(
-            torch.tensor([6.0, 5.0, 4.0, 3.0, 2.0, 1.0]), dim=0
-        )
-
-        log_probabilities = torch.full((batch_size * beam_size, vocab_size), float("-inf"))
-        log_probabilities[0, eos_idx] = score_distribution[0]
-        for idx, score in zip(non_eos_idxs, score_distribution[1:]):
-            log_probabilities[0, idx] = score
-        pytest.set_trace()
-        for step in range(1, min_length + 2):
-            log_probabilities[0, eos_idx] = score_distribution[0]
-
-            # Beam #3 and #4 teminate at the first step since the probability
-            # of the [EOS] token is -1e20 > -\infty so there are only two beams left.
-            # The top beam (most likely) always ends with 4 until we reach min_length.
-            surviving_beams_rows = beam.grow(log_probabilities)
-            if step < min_length:
-                np.testing.assert_array_equal(
-                    beam.growing_beams.numpy()[0, :], np.array([0] + [4] * step)
-                )
-            elif step == min_length:
-                np.testing.assert_array_equal(surviving_beams_rows.numpy(), np.array([]))
-                self.assertTrue(beam.is_done)
-                break
-
-            log_probabilities = log_probabilities.index_select(0, surviving_beams_rows)
-
-    def test_beam_search_max_length(self):
-        """ We keep predicting the same non-EOS token until we reach the
-        maximum permitted length """
-        batch_size = 3
-        beam_size = 2
-        max_length = 5
-        vocab_size = 10
-
-        beam = BeamSearch(
-            model=StubTransformer(),
-            bos_token_id=0,
-            eos_token_id=1,
-            pad_token_id=2,
-            batch_size=batch_size,
-            beam_size=beam_size,
-            min_length=2,
-            max_length=max_length,
-            alpha=0,
-            block_repeating_trigrams=False,
-        )
-
-        log_probabilities = torch.full((batch_size * beam_size, vocab_size), float("-inf"))
-
-        # To test that beam search enforces the max length constraint we
-        # keep giving the highest probability to a token that is not the
-        # [EOS] token.
-        # The beam search will stop at max_length-1, assuming that one would
-        # add the [EOS] token at the end of the returned sequence.
-        token_idxs = [3, 4, 5]
-        score_distribution = torch.log_softmax(torch.tensor([10.0, 6.0, 4.0]), dim=0)
-        for idx, score in zip(token_idxs, score_distribution):
-            log_probabilities[:, idx] = score
-
-        for step in range(1, max_length + 2):
-            surviving_beams_rows = beam.grow(log_probabilities)
-            if step + 1 < max_length:
-                self.assertFalse(beam.is_done)
-            elif step + 1 == max_length:  # Now [EOS] is the most probable token
-                np.testing.assert_array_equal(surviving_beams_rows.numpy(), np.array([]))
-                self.assertTrue(beam.is_done)
-                break
-
-            log_probabilities = log_probabilities.index_select(0, surviving_beams_rows)
-
-    def test_beam_search_block_repeating_trigrams(self):
-        """ We make sure that the beams that contain repeating trigrams are removed. """
-        batch_size = 3
-        beam_size = 2
-        max_length = 10
-        vocab_size = 10
-
-        beam = BeamSearch(
-            model=StubTransformer(),
-            bos_token_id=0,
-            eos_token_id=1,
-            pad_token_id=2,
-            batch_size=batch_size,
-            beam_size=beam_size,
-            min_length=2,
-            max_length=max_length,
-            alpha=0,
-            block_repeating_trigrams=True,
-        )
-
-        log_probabilities = torch.full((batch_size * beam_size, vocab_size), float("-inf"))
-
-        # To test that BeamSearch enforces the 3-gram constraint we give the
-        # highest probably to the same tokens in a cyclic fashion and make sure
-        # they disappear once the cycle has completed.
-        token_idxs = [3, 4, 5]
-        score_distribution = torch.log_softmax(torch.tensor([10.0, 6.0, 4.0]), dim=0)
-        for idx, score in zip(token_idxs, score_distribution):
-            log_probabilities[:, idx] = score
-
-        for step in range(1, max_length + 2):
-            # Rotate the probabilities at each step
-            for idx in token_idxs:
-                score = score_distribution[(idx + step) % 3]
-                log_probabilities[::beam_size, idx] = score
-
-            surviving_beams_rows = beam.grow(log_probabilities)
-
-            if step < 7:
-                self.assertFalse(
-                    np.array_equal(
-                        log_probabilities.numpy()[0, :],
-                        np.array([-1e20] * vocab_size, dtype="float32"),
-                    )
-                )
-            if step == 7:
-                np.testing.assert_array_equal(
-                    log_probabilities.numpy()[0, :],
-                    np.array([-1e20] * vocab_size, dtype="float32"),
-                )
-
-            log_probabilities = log_probabilities.index_select(0, surviving_beams_rows)
-
-    def test_beam_search_example_for_one_step(self):
-        """ We test that the predictions for one step of growth are correct. """
-        batch_size = 2
-        beam_size = 2
-        max_length = 10
-        vocab_size = 5
-
-        beam = BeamSearch(
-            model=StubTransformer(),
-            bos_token_id=0,
-            eos_token_id=1,
-            pad_token_id=2,
-            batch_size=batch_size,
-            beam_size=beam_size,
-            min_length=2,
-            max_length=max_length,
-            alpha=0,
-            block_repeating_trigrams=False,
-        )
-
-        log_probabilities = torch.full((batch_size * beam_size, vocab_size), float("-inf"))
-        log_probabilities[0, 3:] = torch.log_softmax(torch.tensor([2.0, 1.0]), dim=0)
-        log_probabilities[2, 3:] = torch.log_softmax(torch.tensor([1.0, 2.0]), dim=0)
-
-        # First pass
-        surviving_beams_rows = beam.grow(log_probabilities)
-        np.testing.assert_array_equal(surviving_beams_rows.numpy(), np.array([0, 0, 2, 2]))
-        np.testing.assert_array_equal(
-            beam.growing_beams.numpy(), np.array([[0, 3], [0, 4], [0, 4], [0, 3]])
-        )
-        self.assertFalse(beam.is_done)
-
-        # Second pass
-        surviving_beams_rows = beam.grow(log_probabilities)
-        np.testing.assert_array_equal(surviving_beams_rows.numpy(), np.array([0, 0, 2, 2]))
-        np.testing.assert_array_equal(
-            beam.growing_beams.numpy(),
-            np.array([[0, 3, 3], [0, 3, 4], [0, 4, 4], [0, 4, 3]]),
-        )
-        self.assertFalse(beam.is_done)
-
-
-if __name__ == "__name__":
-    unittest.main()

From 693606a75c54d9731b748797f21961d0a5322896 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Thu, 5 Dec 2019 18:55:15 +0100
Subject: [PATCH 274/505] update the docs

---
 examples/README.md | 20 +++++++++++++++++++-
 1 file changed, 19 insertions(+), 1 deletion(-)

diff --git a/examples/README.md b/examples/README.md
index dec5a67f7e..3d0b2ca1a9 100644
--- a/examples/README.md
+++ b/examples/README.md
@@ -24,7 +24,8 @@ pip install -r ./examples/requirements.txt
 | [Multiple Choice](#multiple-choice) | Examples running BERT/XLNet/RoBERTa on the SWAG/RACE/ARC tasks. 
 | [Named Entity Recognition](#named-entity-recognition) | Using BERT for Named Entity Recognition (NER) on the CoNLL 2003 dataset, examples with distributed training.                                                                                  |
 | [XNLI](#xnli) | Examples running BERT/XLM on the XNLI benchmark. |
-| [Abstractive summarization](#abstractive-summarization) | Fine-tuning the library models for abstractive summarization tasks on the CNN/Daily Mail dataset. |
+| [Abstractive summarization](#abstractive-summarization) | Using the BertAbs
+model finetuned on the CNN/DailyMail dataset to generate summaries. |
 
 ## TensorFlow 2.0 Bert models on GLUE
 
@@ -712,3 +713,20 @@ Training with the previously defined hyper-parameters yields the following resul
 ```bash
 acc = 0.7093812375249501
 ```
+
+### Abstractive Summarization
+
+This example provides a simple API for the [BertAbs](https://github.com/nlpyang/PreSumm) model finetuned on the CNN/DailyMail dataset. The script can be used to generate summaries from any text. 
+
+```bash
+python run_summarization.py \
+    --documents_dir 'path/to/documents' \
+    --summaries_output_dir 'path/to/summaries' \
+    --visible_gpus 0,1,2 \
+    --batch_size 4 \
+    --min_length 50 \
+    --max_length 200 \
+    --beam_size 5 \
+    --alpha 0.95 \
+    --block_trigram true
+```

From 3a9a9f78614050896356a9a30e9529c502b56d96 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Thu, 5 Dec 2019 19:09:47 +0100
Subject: [PATCH 275/505] default output dir to documents dir

---
 examples/summarization/run_summarization.py   | 11 ++++++-----
 examples/summarization/utils_summarization.py |  2 ++
 2 files changed, 8 insertions(+), 5 deletions(-)

diff --git a/examples/summarization/run_summarization.py b/examples/summarization/run_summarization.py
index e3b974acd9..bbc79227ca 100644
--- a/examples/summarization/run_summarization.py
+++ b/examples/summarization/run_summarization.py
@@ -31,9 +31,7 @@ Batch = namedtuple(
 
 def evaluate(args):
     tokenizer = BertTokenizer.from_pretrained("bert-base-uncased", do_lower_case=True)
-    model = bertabs = BertAbs.from_pretrained(
-        "bertabs-finetuned-{}".format(args.finetuned_model)
-    )
+    model = bertabs = BertAbs.from_pretrained("bertabs-finetuned-cnndm")
     bertabs.to(args.device)
     bertabs.eval()
 
@@ -195,8 +193,8 @@ def main():
         "--summaries_output_dir",
         default=None,
         type=str,
-        required=True,
-        help="The folder in wich the summaries should be written.",
+        required=False,
+        help="The folder in wich the summaries should be written. Defaults to the folder where the documents are",
     )
     # EVALUATION options
     parser.add_argument(
@@ -242,6 +240,9 @@ def main():
     args = parser.parse_args()
     args.device = torch.device("cpu") if args.visible_gpus == -1 else torch.device("cuda")
 
+    if not args.summaries_output_dir:
+        args.summaries_output_dir = args.documents_dir
+
     if not documents_dir_is_valid(args.documents_dir):
         raise FileNotFoundError(
             "We could not find the directory you specified for the documents to summarize, or it was empty. Please specify a valid path."
diff --git a/examples/summarization/utils_summarization.py b/examples/summarization/utils_summarization.py
index e7401b1754..1d8c436ac9 100644
--- a/examples/summarization/utils_summarization.py
+++ b/examples/summarization/utils_summarization.py
@@ -39,6 +39,8 @@ class SummarizationDataset(Dataset):
         self.documents = []
         story_filenames_list = os.listdir(path)
         for story_filename in story_filenames_list:
+            if "summary" in story_filename:
+                continue
             path_to_story = os.path.join(path, story_filename)
             if not os.path.isfile(path_to_story):
                 continue

From a1994a71ee37ee8ac5bc49cce30a764392d64233 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Thu, 5 Dec 2019 21:05:06 +0100
Subject: [PATCH 276/505] simplified model and configuration

---
 .../summarization/configuration_bertabs.py    | 22 ----------
 examples/summarization/modeling_bertabs.py    | 41 ++++---------------
 examples/summarization/run_summarization.py   |  6 +--
 3 files changed, 10 insertions(+), 59 deletions(-)

diff --git a/examples/summarization/configuration_bertabs.py b/examples/summarization/configuration_bertabs.py
index ff3171f9a8..5bcb65b423 100644
--- a/examples/summarization/configuration_bertabs.py
+++ b/examples/summarization/configuration_bertabs.py
@@ -33,17 +33,6 @@ class BertAbsConfig(PretrainedConfig):
     r""" Class to store the configuration of the BertAbs model.
 
     Arguments:
-        temp_dir: string
-            Unused in the current situation. Kept for compatibility but will be removed.
-        finetune_bert: bool
-            Whether to fine-tune the model or not. Will be kept for reference
-            in case we want to add the possibility to fine-tune the model.
-        large: bool
-            Whether to use bert-large as a base.
-        share_emb: book
-            Whether the embeddings are shared between the encoder and decoder.
-        encoder: string
-            Not clear what this does. Leave to "bert" for pre-trained weights.
         max_pos: int
             The maximum sequence length that this model will be used with.
         enc_layer: int
@@ -77,11 +66,6 @@ class BertAbsConfig(PretrainedConfig):
     def __init__(
         self,
         vocab_size_or_config_json_file=30522,
-        temp_dir=".",
-        finetune_bert=False,
-        large=False,
-        share_emb=True,
-        encoder="bert",
         max_pos=512,
         enc_layers=6,
         enc_hidden_size=512,
@@ -104,21 +88,15 @@ class BertAbsConfig(PretrainedConfig):
             for key, value in json_config.items():
                 self.__dict__[key] = value
         elif isinstance(vocab_size_or_config_json_file, int):
-            self.temp_dir = temp_dir
-            self.finetune_bert = finetune_bert
-            self.large = large
             self.vocab_size = vocab_size_or_config_json_file
             self.max_pos = max_pos
 
-            self.encoder = encoder
             self.enc_layers = enc_layers
             self.enc_hidden_size = enc_hidden_size
             self.enc_heads = enc_heads
             self.enc_ff_size = enc_ff_size
             self.enc_dropout = enc_dropout
 
-            self.share_emb = share_emb
-
             self.dec_layers = dec_layers
             self.dec_hidden_size = dec_hidden_size
             self.dec_heads = dec_heads
diff --git a/examples/summarization/modeling_bertabs.py b/examples/summarization/modeling_bertabs.py
index 0189a2ad2b..5e51526037 100644
--- a/examples/summarization/modeling_bertabs.py
+++ b/examples/summarization/modeling_bertabs.py
@@ -53,7 +53,7 @@ class BertAbs(BertAbsPreTrainedModel):
     def __init__(self, args, checkpoint=None, bert_extractive_checkpoint=None):
         super(BertAbs, self).__init__(args)
         self.args = args
-        self.bert = Bert(args.large, args.temp_dir, args.finetune_bert)
+        self.bert = Bert()
 
         # If pre-trained weights are passed for Bert, load these.
         load_bert_pretrained_extractive = True if bert_extractive_checkpoint else False
@@ -69,18 +69,6 @@ class BertAbs(BertAbsPreTrainedModel):
                 strict=True,
             )
 
-        if args.encoder == "baseline":
-            bert_config = BertConfig(
-                self.bert.model.config.vocab_size,
-                hidden_size=args.enc_hidden_size,
-                num_hidden_layers=args.enc_layers,
-                num_attention_heads=8,
-                intermediate_size=args.enc_ff_size,
-                hidden_dropout_prob=args.enc_dropout,
-                attention_probs_dropout_prob=args.enc_dropout,
-            )
-            self.bert.model = BertModel(bert_config)
-
         self.vocab_size = self.bert.model.config.vocab_size
 
         if args.max_pos > 512:
@@ -101,10 +89,10 @@ class BertAbs(BertAbsPreTrainedModel):
         tgt_embeddings = nn.Embedding(
             self.vocab_size, self.bert.model.config.hidden_size, padding_idx=0
         )
-        if self.args.share_emb:
-            tgt_embeddings.weight = copy.deepcopy(
-                self.bert.model.embeddings.word_embeddings.weight
-            )
+
+        tgt_embeddings.weight = copy.deepcopy(
+            self.bert.model.embeddings.word_embeddings.weight
+        )
 
         self.decoder = TransformerDecoder(
             self.args.dec_layers,
@@ -141,16 +129,6 @@ class BertAbs(BertAbsPreTrainedModel):
             else:
                 p.data.zero_()
 
-    def maybe_tie_embeddings(self, args):
-        if args.use_bert_emb:
-            tgt_embeddings = nn.Embedding(
-                self.vocab_size, self.bert.model.config.hidden_size, padding_idx=0
-            )
-            tgt_embeddings.weight = copy.deepcopy(
-                self.bert.model.embeddings.word_embeddings.weight
-            )
-            self.decoder.embeddings = tgt_embeddings
-
     def forward(
         self,
         encoder_input_ids,
@@ -178,14 +156,9 @@ class Bert(nn.Module):
     """ This class is not really necessary and should probably disappear.
     """
 
-    def __init__(self, large, temp_dir, finetune=False):
+    def __init__(self):
         super(Bert, self).__init__()
-        if large:
-            self.model = BertModel.from_pretrained("bert-large-uncased", cache_dir=temp_dir)
-        else:
-            self.model = BertModel.from_pretrained("bert-base-uncased", cache_dir=temp_dir)
-
-        self.finetune = finetune
+        self.model = BertModel.from_pretrained("bert-base-uncased")
 
     def forward(self, input_ids, attention_mask=None, token_type_ids=None, **kwargs):
         self.eval()
diff --git a/examples/summarization/run_summarization.py b/examples/summarization/run_summarization.py
index bbc79227ca..ed663e880b 100644
--- a/examples/summarization/run_summarization.py
+++ b/examples/summarization/run_summarization.py
@@ -31,9 +31,9 @@ Batch = namedtuple(
 
 def evaluate(args):
     tokenizer = BertTokenizer.from_pretrained("bert-base-uncased", do_lower_case=True)
-    model = bertabs = BertAbs.from_pretrained("bertabs-finetuned-cnndm")
-    bertabs.to(args.device)
-    bertabs.eval()
+    model = BertAbs.from_pretrained("bertabs-finetuned-cnndm")
+    model.to(args.device)
+    model.eval()
 
     symbols = {
         "BOS": tokenizer.vocab["[unused0]"],

From 5909f710285cf8164b3f51111d595ae87f847133 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Thu, 5 Dec 2019 21:07:49 +0100
Subject: [PATCH 277/505] add py-rouge dependency

---
 requirements.txt | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/requirements.txt b/requirements.txt
index 4a3162adce..236ac1c430 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -10,3 +10,5 @@ regex
 sentencepiece
 # For XLM
 sacremoses
+# For ROUGE
+py-rouge

From 076602bdc4b186e715538f437f2bce4b1ee5020e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Fri, 6 Dec 2019 10:11:44 +0100
Subject: [PATCH 278/505] prevent BERT weights from being downloaded twice

---
 examples/summarization/modeling_bertabs.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/examples/summarization/modeling_bertabs.py b/examples/summarization/modeling_bertabs.py
index 5e51526037..efca33fb56 100644
--- a/examples/summarization/modeling_bertabs.py
+++ b/examples/summarization/modeling_bertabs.py
@@ -158,7 +158,8 @@ class Bert(nn.Module):
 
     def __init__(self):
         super(Bert, self).__init__()
-        self.model = BertModel.from_pretrained("bert-base-uncased")
+        config = BertConfig.from_pretrained("bert-base-uncased")
+        self.model = BertModel(config)
 
     def forward(self, input_ids, attention_mask=None, token_type_ids=None, **kwargs):
         self.eval()

From ade3cdf5adfcff7736b326b1360fcf2b59aae47e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Fri, 6 Dec 2019 11:36:44 +0100
Subject: [PATCH 279/505] integrate ROUGE

---
 examples/summarization/modeling_bertabs.py  | 65 +---------------
 examples/summarization/run_summarization.py | 85 +++++++++++++++++++--
 requirements.txt                            |  1 +
 3 files changed, 82 insertions(+), 69 deletions(-)

diff --git a/examples/summarization/modeling_bertabs.py b/examples/summarization/modeling_bertabs.py
index efca33fb56..57126a4df3 100644
--- a/examples/summarization/modeling_bertabs.py
+++ b/examples/summarization/modeling_bertabs.py
@@ -21,9 +21,6 @@
 # SOFTWARE.
 import copy
 import math
-import shutil
-import time
-import os
 
 import numpy as np
 import torch
@@ -1082,11 +1079,6 @@ class Translator(object):
 
         return translations
 
-    def _report_rouge(self, gold_path, can_path):
-        self.logger.info("Calculating Rouge")
-        results_dict = test_rouge(self.args.temp_dir, can_path, gold_path)
-        return results_dict
-
 
 def tile(x, count, dim=0):
     """
@@ -1113,63 +1105,10 @@ def tile(x, count, dim=0):
 
 
 #
-# All things ROUGE. Uses `pyrouge` which is a hot mess.
+# Optimizer for training. We keep this here in case we want to add
+# a finetuning script.
 #
 
-
-def test_rouge(temp_dir, cand, ref):
-    candidates = [line.strip() for line in open(cand, encoding="utf-8")]
-    references = [line.strip() for line in open(ref, encoding="utf-8")]
-    print(len(candidates))
-    print(len(references))
-    assert len(candidates) == len(references)
-
-    cnt = len(candidates)
-    current_time = time.strftime("%Y-%m-%d-%H-%M-%S", time.localtime())
-    tmp_dir = os.path.join(temp_dir, "rouge-tmp-{}".format(current_time))
-    if not os.path.isdir(tmp_dir):
-        os.mkdir(tmp_dir)
-        os.mkdir(tmp_dir + "/candidate")
-        os.mkdir(tmp_dir + "/reference")
-    try:
-
-        for i in range(cnt):
-            if len(references[i]) < 1:
-                continue
-            with open(
-                tmp_dir + "/candidate/cand.{}.txt".format(i), "w", encoding="utf-8"
-            ) as f:
-                f.write(candidates[i])
-            with open(
-                tmp_dir + "/reference/ref.{}.txt".format(i), "w", encoding="utf-8"
-            ) as f:
-                f.write(references[i])
-        r = pyrouge.Rouge155(temp_dir=temp_dir)
-        r.model_dir = tmp_dir + "/reference/"
-        r.system_dir = tmp_dir + "/candidate/"
-        r.model_filename_pattern = "ref.#ID#.txt"
-        r.system_filename_pattern = r"cand.(\d+).txt"
-        rouge_results = r.convert_and_evaluate()
-        print(rouge_results)
-        results_dict = r.output_to_dict(rouge_results)
-    finally:
-        pass
-        if os.path.isdir(tmp_dir):
-            shutil.rmtree(tmp_dir)
-    return results_dict
-
-
-def rouge_results_to_str(results_dict):
-    return ">> ROUGE-F(1/2/3/l): {:.2f}/{:.2f}/{:.2f}\nROUGE-R(1/2/3/l): {:.2f}/{:.2f}/{:.2f}\n".format(
-        results_dict["rouge_1_f_score"] * 100,
-        results_dict["rouge_2_f_score"] * 100,
-        results_dict["rouge_l_f_score"] * 100,
-        results_dict["rouge_1_recall"] * 100,
-        results_dict["rouge_2_recall"] * 100,
-        results_dict["rouge_l_recall"] * 100,
-    )
-
-
 class BertSumOptimizer(object):
     """ Specific optimizer for BertSum.
 
diff --git a/examples/summarization/run_summarization.py b/examples/summarization/run_summarization.py
index ed663e880b..a9d08aca82 100644
--- a/examples/summarization/run_summarization.py
+++ b/examples/summarization/run_summarization.py
@@ -41,6 +41,26 @@ def evaluate(args):
         "PAD": tokenizer.vocab["[PAD]"],
     }
 
+    if args.compute_rouge:
+        reference_summaries = []
+        generated_summaries = []
+
+        import rouge
+        import nltk
+        nltk.download('punkt')
+        rouge_evaluator = rouge.Rouge(
+            metrics=['rouge-n', 'rouge-l'],
+            max_n=2,
+            limit_length=True,
+            length_limit=args.beam_size,
+            length_limit_type='words',
+            apply_avg=True,
+            apply_best=False,
+            alpha=0.5,  # Default F1_score
+            weight_factor=1.2,
+            stemming=True,
+        )
+
     # these (unused) arguments are defined to keep the compatibility
     # with the legacy code and will be deleted in a next iteration.
     args.result_path = ""
@@ -66,6 +86,16 @@ def evaluate(args):
         summaries = [format_summary(t) for t in translations]
         save_summaries(summaries, args.summaries_output_dir, batch.document_names)
 
+        if args.compute_rouge:
+            reference_summaries += batch.tgt_str
+            generated_summaries += summaries
+
+    if args.compute_rouge:
+        scores = rouge_evaluator.get_scores(generated_summaries, reference_summaries)
+        str_scores = format_rouge_scores(scores)
+        save_rouge_scores(str_scores)
+        print(str_scores)
+
 
 def format_summary(translation):
     """ Transforms the output of the `from_batch` function
@@ -86,6 +116,41 @@ def format_summary(translation):
     return summary
 
 
+def format_rouge_scores(scores):
+    return """\n
+****** ROUGE SCORES ******
+
+** ROUGE 1
+F1        >> {:.3f}
+Precision >> {:.3f}
+Recall    >> {:.3f}
+
+** ROUGE 2
+F1        >> {:.3f}
+Precision >> {:.3f}
+Recall    >> {:.3f}
+
+** ROUGE L
+F1        >> {:.3f}
+Precision >> {:.3f}
+Recall    >> {:.3f}""".format(
+        scores['rouge-1']['f'],
+        scores['rouge-1']['p'],
+        scores['rouge-1']['r'],
+        scores['rouge-2']['f'],
+        scores['rouge-2']['p'],
+        scores['rouge-2']['r'],
+        scores['rouge-l']['f'],
+        scores['rouge-l']['p'],
+        scores['rouge-l']['r'],
+    )
+
+
+def save_rouge_scores(str_scores):
+    with open("rouge_scores.txt", "w") as output:
+        output.write(str_scores)
+
+
 def save_summaries(summaries, path, original_document_name):
     """ Write the summaries in fies that are prefixed by the original
     files' name with the `_summary` appended.
@@ -142,26 +207,27 @@ def collate(data, tokenizer, block_size):
     """
     data = [x for x in data if not len(x[1]) == 0]  # remove empty_files
     names = [name for name, _, _ in data]
+    summaries = [" ".join(summary_list) for _, _, summary_list in data]
 
     encoded_text = [
         encode_for_summarization(story, summary, tokenizer) for _, story, summary in data
     ]
-    stories = torch.tensor(
+    encoded_stories = torch.tensor(
         [
             fit_to_block_size(story, block_size, tokenizer.pad_token_id)
             for story, _ in encoded_text
         ]
     )
-    encoder_token_type_ids = compute_token_type_ids(stories, tokenizer.cls_token_id)
-    encoder_mask = build_mask(stories, tokenizer.pad_token_id)
+    encoder_token_type_ids = compute_token_type_ids(encoded_stories, tokenizer.cls_token_id)
+    encoder_mask = build_mask(encoded_stories, tokenizer.pad_token_id)
 
     batch = Batch(
         document_names=names,
-        batch_size=len(stories),
-        src=stories,
+        batch_size=len(encoded_stories),
+        src=encoded_stories,
         segs=encoder_token_type_ids,
         mask_src=encoder_mask,
-        tgt_str=[""] * len(stories),
+        tgt_str=summaries,
     )
 
     return batch
@@ -196,6 +262,13 @@ def main():
         required=False,
         help="The folder in wich the summaries should be written. Defaults to the folder where the documents are",
     )
+    parser.add_argument(
+        "--compute_rouge",
+        default=False,
+        type=bool,
+        required=False,
+        help="Compute the ROUGE metrics during evaluation. Only available for the CNN/DailyMail dataset.",
+    )
     # EVALUATION options
     parser.add_argument(
         "--visible_gpus",
diff --git a/requirements.txt b/requirements.txt
index 236ac1c430..2cbcc3809d 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -11,4 +11,5 @@ sentencepiece
 # For XLM
 sacremoses
 # For ROUGE
+nltk
 py-rouge

From c0707a85d24fa5a74d85d40ed704d4c774e9a37f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Fri, 6 Dec 2019 11:49:27 +0100
Subject: [PATCH 280/505] add README

---
 examples/README.md               | 17 ---------
 examples/summarization/README.md | 61 ++++++++++++++++++++++++++++++++
 2 files changed, 61 insertions(+), 17 deletions(-)
 create mode 100644 examples/summarization/README.md

diff --git a/examples/README.md b/examples/README.md
index 3d0b2ca1a9..620304ea77 100644
--- a/examples/README.md
+++ b/examples/README.md
@@ -713,20 +713,3 @@ Training with the previously defined hyper-parameters yields the following resul
 ```bash
 acc = 0.7093812375249501
 ```
-
-### Abstractive Summarization
-
-This example provides a simple API for the [BertAbs](https://github.com/nlpyang/PreSumm) model finetuned on the CNN/DailyMail dataset. The script can be used to generate summaries from any text. 
-
-```bash
-python run_summarization.py \
-    --documents_dir 'path/to/documents' \
-    --summaries_output_dir 'path/to/summaries' \
-    --visible_gpus 0,1,2 \
-    --batch_size 4 \
-    --min_length 50 \
-    --max_length 200 \
-    --beam_size 5 \
-    --alpha 0.95 \
-    --block_trigram true
-```
diff --git a/examples/summarization/README.md b/examples/summarization/README.md
new file mode 100644
index 0000000000..2b58c00693
--- /dev/null
+++ b/examples/summarization/README.md
@@ -0,0 +1,61 @@
+# Text Summarization with Pretrained Encoders
+
+This folder contains part of the code necessary to reproduce the results on abstractive summarization from the article [Text Summarization with Pretrained Encoders](https://arxiv.org/pdf/1908.08345.pdf) by [Yang Liu](https://nlp-yang.github.io/) and [Mirella Lapata](https://homepages.inf.ed.ac.uk/mlap/). It can also be used to summarize any document.
+
+The original code can be found on the Yang Liu's [github repository](https://github.com/nlpyang/PreSumm).
+
+The model is loaded with the pre-trained weights for the abstractive summarization model trained on the CNN/Daily Mail dataset with an extractive and then abstractive tasks.
+
+## Setup
+
+```
+git clone https://github.com/huggingface/transformers && cd transformers
+pip install [--editable] .
+pip install nltk py-rouge
+cd examples/summarization
+```
+
+## Reproduce the authors' results on ROUGE
+
+To be able to reproduce the authors' results on the CNN/Daily Mail dataset you first need to download both CNN and Daily Mail datasets [from Kyunghyun Cho's website](https://cs.nyu.edu/~kcho/DMQA/) (the links next to "Stories") in the same folder. Then uncompress the archives by running:
+
+```bash
+tar -xvf cnn_stories.tgz && tar -xvf dailymail_stories.tgz
+```
+
+And move all the stories to the same folder. We will refer as `$DATA_PATH` the path to where you uncompressed both archive. Then run the following in the same folder as `run_summarization.py`:
+
+```bash
+python run_summarization.py \
+    --documents_dir $DATA_PATH \
+    --summaries_output_dir $SUMMARIES_PATH \ # optional
+    --visible_gpus 0,1,2 \
+    --batch_size 4 \
+    --min_length 50 \
+    --max_length 200 \
+    --beam_size 5 \
+    --alpha 0.95 \
+    --block_trigram true \
+    --compute_rouge true
+```
+
+The ROUGE scores will be displayed in the console at the end of evaluation and written in a `rouge_scores.txt` file.
+
+## Summarize any text
+
+Put the documents that you would like to summarize in a folder (the path to which is referred to as `$DATA_PATH` below) and run the following in the same folder as `run_summarization.py`:
+
+```bash
+python run_summarization.py \
+    --documents_dir $DATA_PATH \
+    --summaries_output_dir $SUMMARIES_PATH \ # optional
+    --visible_gpus 0,1,2 \
+    --batch_size 4 \
+    --min_length 50 \
+    --max_length 200 \
+    --beam_size 5 \
+    --alpha 0.95 \
+    --block_trigram true \
+```
+
+If you want to compute ROUGE on another dataset you will need to tweak the stories/summaries import in `utils_summarization.py`

From 2a64107e44bd2bb1caee824f121fc4fb6b7d90f7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Fri, 6 Dec 2019 15:45:09 +0100
Subject: [PATCH 281/505] improve device usage

---
 examples/summarization/README.md              |  8 +++----
 ...ert_bertabs_original_pytorch_checkpoint.py |  7 +++---
 examples/summarization/modeling_bertabs.py    |  2 --
 examples/summarization/run_summarization.py   | 23 +++++++++++--------
 4 files changed, 21 insertions(+), 19 deletions(-)

diff --git a/examples/summarization/README.md b/examples/summarization/README.md
index 2b58c00693..96825cfa46 100644
--- a/examples/summarization/README.md
+++ b/examples/summarization/README.md
@@ -29,7 +29,7 @@ And move all the stories to the same folder. We will refer as `$DATA_PATH` the p
 python run_summarization.py \
     --documents_dir $DATA_PATH \
     --summaries_output_dir $SUMMARIES_PATH \ # optional
-    --visible_gpus 0,1,2 \
+    --to_cpu false \
     --batch_size 4 \
     --min_length 50 \
     --max_length 200 \
@@ -39,7 +39,7 @@ python run_summarization.py \
     --compute_rouge true
 ```
 
-The ROUGE scores will be displayed in the console at the end of evaluation and written in a `rouge_scores.txt` file.
+The scripts executes on GPU if one is available and if `to_cpu` is not set to `true`. Inference on multiple GPUs is not suported yet. The ROUGE scores will be displayed in the console at the end of evaluation and written in a `rouge_scores.txt` file. The script takes 30 hours to compute with a single Tesla V100 GPU and a batch size of 10 (300,000 texts to summarize).
 
 ## Summarize any text
 
@@ -49,7 +49,7 @@ Put the documents that you would like to summarize in a folder (the path to whic
 python run_summarization.py \
     --documents_dir $DATA_PATH \
     --summaries_output_dir $SUMMARIES_PATH \ # optional
-    --visible_gpus 0,1,2 \
+    --to_cpu false \
     --batch_size 4 \
     --min_length 50 \
     --max_length 200 \
@@ -58,4 +58,4 @@ python run_summarization.py \
     --block_trigram true \
 ```
 
-If you want to compute ROUGE on another dataset you will need to tweak the stories/summaries import in `utils_summarization.py`
+You may want to play around with `min_length`, `max_length` and `alpha` to suit your use case. If you want to compute ROUGE on another dataset you will need to tweak the stories/summaries import in `utils_summarization.py` and tell it where to fetch the reference summaries.
diff --git a/examples/summarization/convert_bertabs_original_pytorch_checkpoint.py b/examples/summarization/convert_bertabs_original_pytorch_checkpoint.py
index 786a29ef13..33b17bfb6f 100644
--- a/examples/summarization/convert_bertabs_original_pytorch_checkpoint.py
+++ b/examples/summarization/convert_bertabs_original_pytorch_checkpoint.py
@@ -12,10 +12,11 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
-""" Convert BertExtAbs's checkpoints
+""" Convert BertExtAbs's checkpoints.
 
-The file currently does not do much as we ended up copying the exact model
-structure, but I leave it here in case we ever want to refactor the model.
+The script looks like it is doing something trivial but it is not. The "weights"
+proposed by the authors are actually the entire model pickled. We need to load
+the model within the original codebase to be able to only save its `state_dict`.
 """
 
 import argparse
diff --git a/examples/summarization/modeling_bertabs.py b/examples/summarization/modeling_bertabs.py
index 57126a4df3..d989e4fd7e 100644
--- a/examples/summarization/modeling_bertabs.py
+++ b/examples/summarization/modeling_bertabs.py
@@ -847,14 +847,12 @@ class Translator(object):
        global_scores (:obj:`GlobalScorer`):
          object to rescore final translations
        copy_attn (bool): use copy attention during translation
-       cuda (bool): use cuda
        beam_trace (bool): trace beam search for debugging
        logger(logging.Logger): logger.
     """
 
     def __init__(self, args, model, vocab, symbols, global_scorer=None, logger=None):
         self.logger = logger
-        self.cuda = args.visible_gpus != "-1"
 
         self.args = args
         self.model = model
diff --git a/examples/summarization/run_summarization.py b/examples/summarization/run_summarization.py
index a9d08aca82..c388569869 100644
--- a/examples/summarization/run_summarization.py
+++ b/examples/summarization/run_summarization.py
@@ -185,7 +185,7 @@ def save_summaries(summaries, path, original_document_name):
 def build_data_iterator(args, tokenizer):
     dataset = load_and_cache_examples(args, tokenizer)
     sampler = SequentialSampler(dataset)
-    collate_fn = lambda data: collate(data, tokenizer, block_size=512)
+    collate_fn = lambda data: collate(data, tokenizer, block_size=512, device=args.device)
     iterator = DataLoader(
         dataset, sampler=sampler, batch_size=args.batch_size, collate_fn=collate_fn,
     )
@@ -198,7 +198,7 @@ def load_and_cache_examples(args, tokenizer):
     return dataset
 
 
-def collate(data, tokenizer, block_size):
+def collate(data, tokenizer, block_size, device):
     """ Collate formats the data passed to the data loader.
 
     In particular we tokenize the data batch after batch to avoid keeping them
@@ -224,9 +224,9 @@ def collate(data, tokenizer, block_size):
     batch = Batch(
         document_names=names,
         batch_size=len(encoded_stories),
-        src=encoded_stories,
-        segs=encoder_token_type_ids,
-        mask_src=encoder_mask,
+        src=encoded_stories.to(device),
+        segs=encoder_token_type_ids.to(device),
+        mask_src=encoder_mask.to(device),
         tgt_str=summaries,
     )
 
@@ -271,10 +271,10 @@ def main():
     )
     # EVALUATION options
     parser.add_argument(
-        "--visible_gpus",
-        default=-1,
-        type=int,
-        help="Number of GPUs with which to do the training.",
+        "--to_cpu",
+        default=False,
+        type=bool,
+        help="Whether to force the execution on CPU.",
     )
     parser.add_argument(
         "--batch_size", default=4, type=int, help="Batch size per GPU/CPU for training.",
@@ -311,8 +311,11 @@ def main():
         help="Whether to block the existence of repeating trigrams in the text generated by beam search.",
     )
     args = parser.parse_args()
-    args.device = torch.device("cpu") if args.visible_gpus == -1 else torch.device("cuda")
 
+    # Select device (distibuted not available)
+    args.device = torch.device("cuda" if torch.cuda.is_available() and not args.to_cpu else "cpu")
+
+    # Check the existence of directories
     if not args.summaries_output_dir:
         args.summaries_output_dir = args.documents_dir
 

From f7eba090077a443d4a2fd1cd341c822a8fb4dcbc Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Fri, 6 Dec 2019 22:01:48 +0100
Subject: [PATCH 282/505] clean for release

---
 ..._original_pytorch_checkpoint_to_pytorch.py | 161 ------------------
 examples/summarization/modeling_bertabs.py    |   2 +-
 examples/summarization/requirements.txt       |   9 +
 examples/summarization/run_summarization.py   |  60 +++----
 requirements.txt                              |   3 -
 ..._original_pytorch_checkpoint_to_pytorch.py | 158 -----------------
 transformers/generate/__init__.py             |   1 -
 transformers/modeling_encoder_decoder.py      |  31 ++--
 8 files changed, 49 insertions(+), 376 deletions(-)
 delete mode 100644 examples/convert_bertextabs_original_pytorch_checkpoint_to_pytorch.py
 create mode 100644 examples/summarization/requirements.txt
 delete mode 100644 transformers/convert_bertextabs_original_pytorch_checkpoint_to_pytorch.py
 delete mode 100644 transformers/generate/__init__.py

diff --git a/examples/convert_bertextabs_original_pytorch_checkpoint_to_pytorch.py b/examples/convert_bertextabs_original_pytorch_checkpoint_to_pytorch.py
deleted file mode 100644
index c245d0eae5..0000000000
--- a/examples/convert_bertextabs_original_pytorch_checkpoint_to_pytorch.py
+++ /dev/null
@@ -1,161 +0,0 @@
-# coding=utf-8
-# Copyright 2018 The HuggingFace Inc. team.
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-""" Convert BertExtAbs's checkpoints """
-
-import argparse
-from collections import namedtuple
-import logging
-import pdb
-import torch
-
-from models.model_builder import AbsSummarizer  # The authors' implementation
-from model_bertabs import BertAbsSummarizer
-
-from transformers import BertTokenizer
-
-
-logging.basicConfig(level=logging.INFO)
-logger = logging.getLogger(__name__)
-
-
-SAMPLE_TEXT = 'Hello world! cécé herlolip'
-
-
-BertAbsConfig = namedtuple(
-    "BertAbsConfig",
-    ["temp_dir", "large", "use_bert_emb", "finetune_bert", "encoder", "share_emb", "max_pos", "enc_layers", "enc_hidden_size", "enc_heads", "enc_ff_size", "enc_dropout", "dec_layers", "dec_hidden_size", "dec_heads", "dec_ff_size", "dec_dropout"],
-)
-
-
-def convert_bertabs_checkpoints(path_to_checkpoints, dump_path):
-    """ Copy/paste and tweak the pre-trained weights provided by the creators
-    of BertAbs for the internal architecture.
-    """
-
-    # Instantiate the authors' model with the pre-trained weights
-    config = BertAbsConfig(
-        temp_dir=".",
-        finetune_bert=False,
-        large=False,
-        share_emb=True,
-        use_bert_emb=False,
-        encoder="bert",
-        max_pos=512,
-        enc_layers=6,
-        enc_hidden_size=512,
-        enc_heads=8,
-        enc_ff_size=512,
-        enc_dropout=0.2,
-        dec_layers=6,
-        dec_hidden_size=768,
-        dec_heads=8,
-        dec_ff_size=2048,
-        dec_dropout=0.2,
-    )
-    checkpoints = torch.load(path_to_checkpoints, lambda storage, loc: storage)
-    original = AbsSummarizer(config, torch.device("cpu"), checkpoints)
-    original.eval()
-
-    new_model = BertAbsSummarizer(config, torch.device("cpu"))
-    new_model.eval()
-
-    # -------------------
-    # Convert the weights
-    # -------------------
-
-    logging.info("convert the model")
-    new_model.encoder.load_state_dict(original.bert.state_dict())
-
-    new_model.decoder.generator.load_state_dict(original.generator.state_dict())
-    new_model.decoder.embeddings.load_state_dict(original.decoder.embeddings.state_dict())
-    new_model.decoder.pos_emb.load_state_dict(original.decoder.pos_emb.state_dict())
-    new_model.decoder.transformer_layers.load_state_dict(original.decoder.transformer_layers.state_dict())
-    new_model.decoder.layer_norm.load_state_dict(original.decoder.layer_norm.state_dict())
-
-    # ----------------------------------
-    # Make sure the outpus are identical
-    # ----------------------------------
-
-    logging.info("Make sure that the models' outputs are identical")
-    tokenizer = BertTokenizer.from_pretrained("bert-base-uncased")
-
-    # prepare the model inputs
-    encoder_input_ids = tokenizer.encode("This is sample éàalj'-.")
-    encoder_input_ids.extend([tokenizer.pad_token_id] * (512 - len(encoder_input_ids)))
-    encoder_input_ids = torch.tensor(encoder_input_ids).unsqueeze(0)
-    decoder_input_ids = tokenizer.encode("This is sample 3 éàalj'-.")
-    decoder_input_ids.extend([tokenizer.pad_token_id] * (512 - len(decoder_input_ids)))
-    decoder_input_ids = torch.tensor(decoder_input_ids).unsqueeze(0)
-
-    # failsafe to make sure the weights reset does not affect the
-    # loaded weights.
-    assert torch.max(torch.abs(original.generator[0].weight - new_model.decoder.generator[0].weight)) == 0
-
-    # forward pass
-    src = encoder_input_ids
-    tgt = decoder_input_ids
-    segs = token_type_ids = None
-    clss = None
-    mask_src = encoder_attention_mask = None
-    mask_tgt = decoder_attention_mask = None
-    mask_cls = None
-    
-    # The original model does not apply the geneator layer immediatly but rather in
-    # the beam search (where it combines softmax + linear layer). Since we already
-    # apply the softmax in our generation process we only apply the linear layer here.
-    # We make sure that the outputs of the full stack are identical
-    output_original_model = original(src, tgt, segs, clss, mask_src, mask_tgt, mask_cls)[0]
-    output_original_model = original.generator(output_original_model)
-
-    output_converted_model = new_model(encoder_input_ids, decoder_input_ids, token_type_ids, encoder_attention_mask, decoder_attention_mask)[0]
-    output_converted_model = torch.nn.functional.log_softmax(output_converted_model, dim=-1)
-
-    maximum_absolute_difference = torch.max(torch.abs(output_converted_model - output_original_model)).item()
-    print("Maximum absolute difference beween weights: {:.2f}".format(maximum_absolute_difference))
-
-    are_identical = torch.allclose(output_converted_model, output_original_model, atol=1e-3)
-    if are_identical:
-        logging.info("all weights are equal up to 1e-3")
-    else:
-        raise ValueError("the weights are different. The new model is likely different from the original one.")
-
-    # The model has been saved with torch.save(model) and this is bound to the exact
-    # directory structure. We save the state_dict instead.
-    logging.info("saving the model's state dictionary")
-    torch.save(new_model.state_dict(), "bert-ext-abs.pt")
-
-
-if __name__ == "__main__":
-    parser = argparse.ArgumentParser()
-    parser.add_argument(
-        "--bertabs_checkpoint_path",
-        default=None,
-        type=str,
-        required=True,
-        help="Path the official PyTorch dump.",
-    )
-    parser.add_argument(
-        "--pytorch_dump_folder_path",
-        default=None,
-        type=str,
-        required=True,
-        help="Path to the output PyTorch model.",
-    )
-    args = parser.parse_args()
-
-    convert_bertabs_checkpoints(
-        args.bertabs_checkpoint_path,
-        args.pytorch_dump_folder_path,
-    )
diff --git a/examples/summarization/modeling_bertabs.py b/examples/summarization/modeling_bertabs.py
index d989e4fd7e..5bf1599ad2 100644
--- a/examples/summarization/modeling_bertabs.py
+++ b/examples/summarization/modeling_bertabs.py
@@ -1,6 +1,6 @@
 # MIT License
 
-# Copyright (c) 2019 Yang Liu
+# Copyright (c) 2019 Yang Liu and the HuggingFace team
 
 # Permission is hereby granted, free of charge, to any person obtaining a copy
 # of this software and associated documentation files (the "Software"), to deal
diff --git a/examples/summarization/requirements.txt b/examples/summarization/requirements.txt
new file mode 100644
index 0000000000..36d75a5edc
--- /dev/null
+++ b/examples/summarization/requirements.txt
@@ -0,0 +1,9 @@
+# progress bars in model download and training scripts
+tqdm
+# Accessing files from S3 directly.
+boto3
+# Used for downloading models over HTTP
+requests
+# For ROUGE
+nltk
+py-rouge
diff --git a/examples/summarization/run_summarization.py b/examples/summarization/run_summarization.py
index c388569869..f58ce3bb43 100644
--- a/examples/summarization/run_summarization.py
+++ b/examples/summarization/run_summarization.py
@@ -1,3 +1,4 @@
+#! /usr/bin/python3
 import argparse
 from collections import namedtuple
 import logging
@@ -97,6 +98,32 @@ def evaluate(args):
         print(str_scores)
 
 
+def save_summaries(summaries, path, original_document_name):
+    """ Write the summaries in fies that are prefixed by the original
+    files' name with the `_summary` appended.
+
+    Attributes:
+        original_document_names: List[string]
+            Name of the document that was summarized.
+        path: string
+            Path were the summaries will be written
+        summaries: List[string]
+            The summaries that we produced.
+    """
+    for summary, document_name in zip(summaries, original_document_name):
+        # Prepare the summary file's name
+        if "." in document_name:
+            bare_document_name = ".".join(document_name.split(".")[:-1])
+            extension = document_name.split(".")[-1]
+            name = bare_document_name + "_summary." + extension
+        else:
+            name = document_name + "_summary"
+
+        file_path = os.path.join(path, name)
+        with open(file_path, "w") as output:
+            output.write(summary)
+
+
 def format_summary(translation):
     """ Transforms the output of the `from_batch` function
     into nicely formatted summaries.
@@ -151,32 +178,6 @@ def save_rouge_scores(str_scores):
         output.write(str_scores)
 
 
-def save_summaries(summaries, path, original_document_name):
-    """ Write the summaries in fies that are prefixed by the original
-    files' name with the `_summary` appended.
-
-    Attributes:
-        original_document_names: List[string]
-            Name of the document that was summarized.
-        path: string
-            Path were the summaries will be written
-        summaries: List[string]
-            The summaries that we produced.
-    """
-    for summary, document_name in zip(summaries, original_document_name):
-        # Prepare the summary file's name
-        if "." in document_name:
-            bare_document_name = ".".join(document_name.split(".")[:-1])
-            extension = document_name.split(".")[-1]
-            name = bare_document_name + "_summary." + extension
-        else:
-            name = document_name + "_summary"
-
-        file_path = os.path.join(path, name)
-        with open(file_path, "w") as output:
-            output.write(summary)
-
-
 #
 # LOAD the dataset
 #
@@ -323,7 +324,7 @@ def main():
         raise FileNotFoundError(
             "We could not find the directory you specified for the documents to summarize, or it was empty. Please specify a valid path."
         )
-    maybe_create_output_dir(args.summaries_output_dir)
+    os.makedirs(args.summaries_output_dir, exist_ok=True)
 
     evaluate(args)
 
@@ -339,10 +340,5 @@ def documents_dir_is_valid(path):
     return True
 
 
-def maybe_create_output_dir(path):
-    if not os.path.exists(path):
-        os.makedirs(path)
-
-
 if __name__ == "__main__":
     main()
diff --git a/requirements.txt b/requirements.txt
index 2cbcc3809d..4a3162adce 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -10,6 +10,3 @@ regex
 sentencepiece
 # For XLM
 sacremoses
-# For ROUGE
-nltk
-py-rouge
diff --git a/transformers/convert_bertextabs_original_pytorch_checkpoint_to_pytorch.py b/transformers/convert_bertextabs_original_pytorch_checkpoint_to_pytorch.py
deleted file mode 100644
index 4f158966e1..0000000000
--- a/transformers/convert_bertextabs_original_pytorch_checkpoint_to_pytorch.py
+++ /dev/null
@@ -1,158 +0,0 @@
-# coding=utf-8
-# Copyright 2018 The HuggingFace Inc. team.
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-""" Convert BertExtAbs's checkpoints """
-
-import argparse
-from collections import namedtuple
-import logging
-
-import torch
-
-from models.model_builder import AbsSummarizer  # The authors' implementation
-
-from transformers import BertConfig, Model2Model, BertModel, BertForMaskedLM
-
-
-logging.basicConfig(level=logging.INFO)
-logger = logging.getLogger(__name__)
-
-
-BertExtAbsConfig = namedtuple(
-    "BertExtAbsConfig",
-    ["temp_dir", "large", "finetune_bert", "encoder", "share_emb", "max_pos", "enc_layers", "enc_hidden_size", "enc_heads", "enc_ff_size", "enc_dropout", "dec_layers", "dec_hidden_size", "dec_heads", "dec_ff_size", "dec_dropout"],
-)
-
-
-def convert_bertextabs_checkpoints(path_to_checkpoints, dump_path):
-    """ Copy/paste and tweak the pre-trained weights provided by the creators
-    of BertExtAbs for the internal architecture.
-    """
-
-    # Load checkpoints in memory
-    checkpoints = torch.load(path_to_checkpoints, lambda storage, loc: storage)
-
-    # Instantiate the authors' model with the pre-trained weights
-    config = BertExtAbsConfig(
-        temp_dir=".",
-        finetune_bert=False,
-        large=False,
-        share_emb=True,
-        encoder="bert",
-        max_pos=512,
-        enc_layers=6,
-        enc_hidden_size=512,
-        enc_heads=8,
-        enc_ff_size=512,
-        enc_dropout=0.2,
-        dec_layers=6,
-        dec_hidden_size=768,
-        dec_heads=8,
-        dec_ff_size=2048,
-        dec_dropout=0.2,
-    )
-    bertextabs = AbsSummarizer(config, torch.device("cpu"), checkpoints)
-    bertextabs.eval()
-
-    # Instantiate our version of the model
-    decoder_config = BertConfig(
-        hidden_size=config.dec_hidden_size,
-        num_hidden_layers=config.dec_layers,
-        num_attention_heads=config.dec_heads,
-        intermediate_size=config.dec_ff_size,
-        hidden_dropout_prob=config.dec_dropout,
-        attention_probs_dropout_prob=config.dec_dropout,
-        is_decoder=True,
-    )
-
-    decoder_model = BertForMaskedLM(decoder_config)
-    model = Model2Model.from_pretrained('bert-base-uncased', decoder_model=decoder_model)
-    model.eval()
-
-    # Let us now start the weight copying process
-    model.encoder.load_state_dict(bertextabs.bert.model.state_dict())
-
-    # Decoder
-
-    # Embeddings. The positional embeddings are equal to the word embedding plus a modulation
-    # that is computed at each forward pass. This may be a source of discrepancy.
-    model.decoder.bert.embeddings.word_embeddings.weight = bertextabs.decoder.embeddings.weight
-    model.decoder.bert.embeddings.position_embeddings.weight = bertextabs.decoder.embeddings.weight
-    model.decoder.bert.embeddings.token_type_embeddings.weight.data = torch.zeros_like(bertextabs.decoder.embeddings.weight)  # not defined for BertExtAbs decoder
-
-    # In the original code the LayerNorms are applied twice in the layers, at the beginning and between the
-    # attention layers.
-    model.decoder.bert.embeddings.LayerNorm.weight = bertextabs.decoder.transformer_layers[0].layer_norm_1.weight
-
-    for i in range(config.dec_layers):
-
-        # self attention
-        model.decoder.bert.encoder.layer[i].attention.self.query.weight = bertextabs.decoder.transformer_layers[i].self_attn.linear_query.weight
-        model.decoder.bert.encoder.layer[i].attention.self.key.weight = bertextabs.decoder.transformer_layers[i].self_attn.linear_keys.weight
-        model.decoder.bert.encoder.layer[i].attention.self.value.weight = bertextabs.decoder.transformer_layers[i].self_attn.linear_values.weight
-        model.decoder.bert.encoder.layer[i].attention.output.dense.weight = bertextabs.decoder.transformer_layers[i].self_attn.final_linear.weight
-        model.decoder.bert.encoder.layer[i].attention.output.LayerNorm.weight = bertextabs.decoder.transformer_layers[i].layer_norm_2.weight
-
-        # attention
-        model.decoder.bert.encoder.layer[i].crossattention.self.query.weight = bertextabs.decoder.transformer_layers[i].context_attn.linear_query.weight
-        model.decoder.bert.encoder.layer[i].crossattention.self.key.weight = bertextabs.decoder.transformer_layers[i].context_attn.linear_keys.weight
-        model.decoder.bert.encoder.layer[i].crossattention.self.value.weight = bertextabs.decoder.transformer_layers[i].context_attn.linear_values.weight
-        model.decoder.bert.encoder.layer[i].crossattention.output.dense.weight = bertextabs.decoder.transformer_layers[i].context_attn.final_linear.weight
-        model.decoder.bert.encoder.layer[i].crossattention.output.LayerNorm.weight = bertextabs.decoder.transformer_layers[i].feed_forward.layer_norm.weight
-
-        # intermediate
-        model.decoder.bert.encoder.layer[i].intermediate.dense.weight = bertextabs.decoder.transformer_layers[i].feed_forward.w_1.weight
-
-        # output
-        model.decoder.bert.encoder.layer[i].output.dense.weight = bertextabs.decoder.transformer_layers[i].feed_forward.w_2.weight
-
-        try:
-            model.decoder.bert.encoder.layer[i].output.LayerNorm.weight = bertextabs.decoder.transformer_layers[i + 1].layer_norm_1.weight
-        except IndexError:
-            model.decoder.bert.encoder.layer[i].output.LayerNorm.weight = bertextabs.decoder.layer_norm.weight
-
-    # LM Head
-    """
-    model.decoder.cls.predictions.transform.dense.weight
-    model.decoder.cls.predictions.transform.dense.biais
-    model.decoder.cls.predictions.transform.LayerNorm.weight
-    model.decoder.cls.predictions.transform.LayerNorm.biais
-    model.decoder.cls.predictions.decoder.weight
-    model.decoder.cls.predictions.decoder.biais
-    model.decoder.cls.predictions.biais.data
-    """
-
-
-if __name__ == "__main__":
-    parser = argparse.ArgumentParser()
-    parser.add_argument(
-        "--bertextabs_checkpoint_path",
-        default=None,
-        type=str,
-        required=True,
-        help="Path the official PyTorch dump.",
-    )
-    parser.add_argument(
-        "--pytorch_dump_folder_path",
-        default=None,
-        type=str,
-        required=True,
-        help="Path to the output PyTorch model.",
-    )
-    args = parser.parse_args()
-
-    convert_bertextabs_checkpoints(
-        args.bertextabs_checkpoint_path,
-        args.pytorch_dump_folder_path,
-    )
diff --git a/transformers/generate/__init__.py b/transformers/generate/__init__.py
deleted file mode 100644
index 21ac612155..0000000000
--- a/transformers/generate/__init__.py
+++ /dev/null
@@ -1 +0,0 @@
-from .beam_search import BeamSearch
diff --git a/transformers/modeling_encoder_decoder.py b/transformers/modeling_encoder_decoder.py
index 73322101d3..a884abd0a2 100644
--- a/transformers/modeling_encoder_decoder.py
+++ b/transformers/modeling_encoder_decoder.py
@@ -117,7 +117,8 @@ class PreTrainedEncoderDecoder(nn.Module):
         kwargs_common = {
             argument: value
             for argument, value in kwargs.items()
-            if not argument.startswith("encoder_") and not argument.startswith("decoder_")
+            if not argument.startswith("encoder_")
+            and not argument.startswith("decoder_")
         }
         kwargs_decoder = kwargs_common.copy()
         kwargs_encoder = kwargs_common.copy()
@@ -157,27 +158,14 @@ class PreTrainedEncoderDecoder(nn.Module):
 
         return model
 
-    def save_pretrained(self, save_directory, model_type="bert"):
-        """ Save an EncoderDecoder model and its configuration file in a format such
+    def save_pretrained(self, save_directory):
+        """ Save a Seq2Seq model and its configuration file in a format such
         that it can be loaded using `:func:`~transformers.PreTrainedEncoderDecoder.from_pretrained`
 
         We save the encoder' and decoder's parameters in two separate directories.
-
-        If we want the weight loader to function we need to preprend the model
-        type to the directories' names. As far as I know there is no simple way
-        to infer the type of the model (except maybe by parsing the class'
-        names, which is not very future-proof). For now, we ask the user to
-        specify the model type explicitly when saving the weights.
         """
-        encoder_path = os.path.join(save_directory, "{}_encoder".format(model_type))
-        if not os.path.exists(encoder_path):
-            os.makedirs(encoder_path)
-        self.encoder.save_pretrained(encoder_path)
-
-        decoder_path = os.path.join(save_directory, "{}_decoder".format(model_type))
-        if not os.path.exists(decoder_path):
-            os.makedirs(decoder_path)
-        self.decoder.save_pretrained(decoder_path)
+        self.encoder.save_pretrained(os.path.join(save_directory, "encoder"))
+        self.decoder.save_pretrained(os.path.join(save_directory, "decoder"))
 
     def forward(self, encoder_input_ids, decoder_input_ids, **kwargs):
         """ The forward pass on a seq2eq depends what we are performing:
@@ -205,7 +193,8 @@ class PreTrainedEncoderDecoder(nn.Module):
         kwargs_common = {
             argument: value
             for argument, value in kwargs.items()
-            if not argument.startswith("encoder_") and not argument.startswith("decoder_")
+            if not argument.startswith("encoder_")
+            and not argument.startswith("decoder_")
         }
         kwargs_decoder = kwargs_common.copy()
         kwargs_encoder = kwargs_common.copy()
@@ -228,7 +217,9 @@ class PreTrainedEncoderDecoder(nn.Module):
         encoder_hidden_states = kwargs_encoder.pop("hidden_states", None)
         if encoder_hidden_states is None:
             encoder_outputs = self.encoder(encoder_input_ids, **kwargs_encoder)
-            encoder_hidden_states = encoder_outputs[0]  # output the last layer hidden state
+            encoder_hidden_states = encoder_outputs[
+                0
+            ]  # output the last layer hidden state
         else:
             encoder_outputs = ()
 

From 1d189304624db17749aee23fa2345f009cc48215 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Tue, 10 Dec 2019 01:32:42 +0000
Subject: [PATCH 283/505] Harmonize `no_cuda` flag with other scripts

---
 examples/summarization/run_summarization.py | 4 ++--
 requirements.txt                            | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/examples/summarization/run_summarization.py b/examples/summarization/run_summarization.py
index f58ce3bb43..3c339d0c30 100644
--- a/examples/summarization/run_summarization.py
+++ b/examples/summarization/run_summarization.py
@@ -272,7 +272,7 @@ def main():
     )
     # EVALUATION options
     parser.add_argument(
-        "--to_cpu",
+        "--no_cuda",
         default=False,
         type=bool,
         help="Whether to force the execution on CPU.",
@@ -314,7 +314,7 @@ def main():
     args = parser.parse_args()
 
     # Select device (distibuted not available)
-    args.device = torch.device("cuda" if torch.cuda.is_available() and not args.to_cpu else "cpu")
+    args.device = torch.device("cuda" if torch.cuda.is_available() and not args.no_cuda else "cpu")
 
     # Check the existence of directories
     if not args.summaries_output_dir:
diff --git a/requirements.txt b/requirements.txt
index 4a3162adce..9c43abc6d7 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -9,4 +9,4 @@ regex
 # For XLNet
 sentencepiece
 # For XLM
-sacremoses
+sacremoses
\ No newline at end of file

From 608a8f5b567f81f3cc997a195496dd8bf1c28158 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Tue, 10 Dec 2019 10:01:01 +0100
Subject: [PATCH 284/505] updating tf 2.0 layer_norm to T5 layer norm

---
 transformers/modeling_tf_t5.py | 43 ++++++++++++++++++++++++----------
 1 file changed, 30 insertions(+), 13 deletions(-)

diff --git a/transformers/modeling_tf_t5.py b/transformers/modeling_tf_t5.py
index c1de4745c2..11762ee1e5 100644
--- a/transformers/modeling_tf_t5.py
+++ b/transformers/modeling_tf_t5.py
@@ -17,16 +17,11 @@
 
 from __future__ import absolute_import, division, print_function, unicode_literals
 
-import json
 import logging
 import math
-import os
-import sys
 import copy
 import itertools
-from io import open
 
-import numpy as np
 import tensorflow as tf
 
 from .configuration_t5 import T5Config
@@ -45,6 +40,28 @@ TF_T5_PRETRAINED_MODEL_ARCHIVE_MAP = {
 # - TFPreTrainedModel for the models (it-self a sub-class of tf.keras.Model)
 ####################################################
 
+class TFT5LayerNorm(tf.keras.layers.Layer):
+    def __init__(self, epsilon=1e-6, **kwargs):
+        """ Construct a layernorm module in the T5 style
+            No bias and no substraction of mean.
+        """
+        super(TFT5LayerNorm, self).__init__(**kwargs)
+        self.variance_epsilon = epsilon
+
+    def build(self, input_shape):
+        """Build shared word embedding layer """
+        self.weight = self.add_weight(
+            "weight",
+            shape=(input_shape[-1],),
+            initializer='ones')
+        super(TFT5LayerNorm, self).build(input_shape)
+
+    def call(self, x):
+        variance = tf.math.reduce_min(tf.math.square(x), axis=-1, keepdims=True)
+        x = x * tf.math.rsqrt(variance + self.variance_epsilon)
+        return self.weight * x
+
+
 class TFT5DenseReluDense(tf.keras.layers.Layer):
     def __init__(self, config, **kwargs):
         super(TFT5DenseReluDense, self).__init__(**kwargs)
@@ -65,8 +82,8 @@ class TFT5LayerFF(tf.keras.layers.Layer):
     def __init__(self, config, **kwargs):
         super(TFT5LayerFF, self).__init__(**kwargs)
         self.DenseReluDense = TFT5DenseReluDense(config, name='DenseReluDense')
-        self.layer_norm = tf.keras.layers.LayerNormalization(epsilon=config.layer_norm_epsilon,
-                                                             name='layer_norm')
+        self.layer_norm = TFT5LayerNorm(epsilon=config.layer_norm_epsilon,
+                                        name='layer_norm')
         self.dropout = tf.keras.layers.Dropout(config.dropout_rate)
 
     def call(self, hidden_states, training=False):
@@ -249,8 +266,8 @@ class TFT5LayerSelfAttention(tf.keras.layers.Layer):
         self.SelfAttention = TFT5Attention(config,
                                            has_relative_attention_bias=has_relative_attention_bias,
                                            name='SelfAttention')
-        self.layer_norm = tf.keras.layers.LayerNormalization(epsilon=config.layer_norm_epsilon,
-                                                             name='layer_norm')
+        self.layer_norm = TFT5LayerNorm(epsilon=config.layer_norm_epsilon,
+                                        name='layer_norm')
         self.dropout = tf.keras.layers.Dropout(config.dropout_rate)
 
     def call(self, hidden_states, attention_mask=None, position_bias=None,
@@ -273,8 +290,8 @@ class TFT5LayerCrossAttention(tf.keras.layers.Layer):
         self.EncDecAttention = TFT5Attention(config,
                                            has_relative_attention_bias=has_relative_attention_bias,
                                            name='EncDecAttention')
-        self.layer_norm = tf.keras.layers.LayerNormalization(epsilon=config.layer_norm_epsilon,
-                                                             name='layer_norm')
+        self.layer_norm = TFT5LayerNorm(epsilon=config.layer_norm_epsilon,
+                                        name='layer_norm')
         self.dropout = tf.keras.layers.Dropout(config.dropout_rate)
 
     def call(self, hidden_states, kv, attention_mask=None, position_bias=None,
@@ -353,8 +370,8 @@ class TFT5MainLayer(tf.keras.layers.Layer):
                                 has_relative_attention_bias=bool(i == 0),
                                 name='block_._{}'.format(i))
                         for i in range(config.num_layers)]
-        self.final_layer_norm = tf.keras.layers.LayerNormalization(epsilon=config.layer_norm_epsilon,
-                                                                   name='final_layer_norm')
+        self.final_layer_norm = TFT5LayerNorm(epsilon=config.layer_norm_epsilon,
+                                              name='final_layer_norm')
         self.dropout = tf.keras.layers.Dropout(config.dropout_rate)
 
     def _resize_token_embeddings(self, new_num_tokens):

From 72c36b9ea2e43d017d3aa5520d09f55d8ec19025 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Wed, 16 Oct 2019 14:17:58 +0200
Subject: [PATCH 285/505] [WIP] - CLI

---
 setup.py                                 |   8 +-
 transformers-cli                         |   5 +-
 transformers/__init__.py                 |   2 +
 transformers/__main__.py                 | 145 ++++--------------
 transformers/commands/convert.py         | 115 ++++++++++++++
 transformers/commands/serving.py         | 176 ++++++++++++++++++++++
 transformers/commands/train.py           | 121 +++++++++++++++
 transformers/data/__init__.py            |   2 +-
 transformers/data/processors/__init__.py |   2 +-
 transformers/data/processors/utils.py    | 182 +++++++++++++++++++++++
 10 files changed, 631 insertions(+), 127 deletions(-)
 create mode 100644 transformers/commands/convert.py
 create mode 100644 transformers/commands/serving.py
 create mode 100644 transformers/commands/train.py

diff --git a/setup.py b/setup.py
index c4af32df83..0b7e512955 100644
--- a/setup.py
+++ b/setup.py
@@ -62,15 +62,15 @@ setup(
                       'regex',
                       'sentencepiece',
                       'sacremoses'],
+    extras_require=extras,
+    scripts=[
+        'transformers-cli'
+    ],
     entry_points={
       'console_scripts': [
         "transformers=transformers.__main__:main",
       ]
     },
-    extras_require=extras,
-    scripts=[
-        'transformers-cli'
-    ],
     # python_requires='>=3.5.0',
     classifiers=[
           'Intended Audience :: Science/Research',
diff --git a/transformers-cli b/transformers-cli
index ef00d15aa3..7b0905d4b4 100644
--- a/transformers-cli
+++ b/transformers-cli
@@ -1,14 +1,15 @@
 #!/usr/bin/env python
 from argparse import ArgumentParser
 
+from transformers.commands.serving import ServeCommand
 from transformers.commands.user import UserCommands
 
-
 if __name__ == '__main__':
-    parser = ArgumentParser(description='Transformers CLI tool', usage='transformers-cli <command> [<args>]')
+    parser = ArgumentParser('Transformers CLI tool', usage='transformers-cli <command> [<args>]')
     commands_parser = parser.add_subparsers(help='transformers-cli command helpers')
 
     # Register commands
+    ServeCommand.register_subcommand(commands_parser)
     UserCommands.register_subcommand(commands_parser)
 
     # Let's go
diff --git a/transformers/__init__.py b/transformers/__init__.py
index f9a28add5f..a71a291a44 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -24,6 +24,8 @@ from .file_utils import (TRANSFORMERS_CACHE, PYTORCH_TRANSFORMERS_CACHE, PYTORCH
 
 from .data import (is_sklearn_available,
                    InputExample, InputFeatures, DataProcessor,
+                   SingleSentenceClassificationProcessor,
+                   convert_examples_to_features,
                    glue_output_modes, glue_convert_examples_to_features,
                    glue_processors, glue_tasks_num_labels,
                    xnli_output_modes, xnli_processors, xnli_tasks_num_labels,
diff --git a/transformers/__main__.py b/transformers/__main__.py
index 31dbd24908..a6e9ae65e0 100644
--- a/transformers/__main__.py
+++ b/transformers/__main__.py
@@ -1,129 +1,36 @@
 # coding: utf8
+
 def main():
     import sys
-    if (len(sys.argv) < 4 or len(sys.argv) > 6) or sys.argv[1] not in ["bert", "gpt", "transfo_xl", "gpt2", "xlnet", "xlm"]:
+    if len(sys.argv) < 2 or sys.argv[1] not in ["convert", "train", "predict", "serve"]:
         print(
-        "This command line utility let you convert original (author released) model checkpoint to pytorch.\n"
-        "It should be used as one of: \n"
-        ">> transformers bert TF_CHECKPOINT TF_CONFIG PYTORCH_DUMP_OUTPUT, \n"
-        ">> transformers gpt OPENAI_GPT_CHECKPOINT_FOLDER_PATH PYTORCH_DUMP_OUTPUT [OPENAI_GPT_CONFIG], \n"
-        ">> transformers transfo_xl TF_CHECKPOINT_OR_DATASET PYTORCH_DUMP_OUTPUT [TF_CONFIG] or \n"
-        ">> transformers gpt2 TF_CHECKPOINT PYTORCH_DUMP_OUTPUT [GPT2_CONFIG] or \n"
-        ">> transformers xlnet TF_CHECKPOINT TF_CONFIG PYTORCH_DUMP_OUTPUT [FINETUNING_TASK_NAME] or \n"
-        ">> transformers xlm XLM_CHECKPOINT_PATH PYTORCH_DUMP_OUTPUT")
-    else:
-        if sys.argv[1] == "bert":
-            try:
-                from .convert_bert_original_tf_checkpoint_to_pytorch import convert_tf_checkpoint_to_pytorch
-            except ImportError:
-                print("transformers can only be used from the commandline to convert TensorFlow models in PyTorch, "
-                    "In that case, it requires TensorFlow to be installed. Please see "
-                    "https://www.tensorflow.org/install/ for installation instructions.")
-                raise
+        "First argument to `transformers` command line interface should be one of: \n"
+        ">> convert serve train predict")
+    if sys.argv[1] == "convert":
+        from transformers.commands import convert
+        convert(sys.argv)
+    elif sys.argv[1] == "train":
+        from transformers.commands import train
+        train(sys.argv)
+    elif sys.argv[1] == "serve":
+        pass
+        # from argparse import ArgumentParser
+        # from transformers.commands.serving import ServeCommand
+        # parser = ArgumentParser('Transformers CLI tool', usage='transformers serve <command> [<args>]')
+        # commands_parser = parser.add_subparsers(help='transformers-cli command helpers')
 
-            if len(sys.argv) != 5:
-                # pylint: disable=line-too-long
-                print("Should be used as `transformers bert TF_CHECKPOINT TF_CONFIG PYTORCH_DUMP_OUTPUT`")
-            else:
-                PYTORCH_DUMP_OUTPUT = sys.argv.pop()
-                TF_CONFIG = sys.argv.pop()
-                TF_CHECKPOINT = sys.argv.pop()
-                convert_tf_checkpoint_to_pytorch(TF_CHECKPOINT, TF_CONFIG, PYTORCH_DUMP_OUTPUT)
-        elif sys.argv[1] == "gpt":
-            from .convert_openai_original_tf_checkpoint_to_pytorch import convert_openai_checkpoint_to_pytorch
-            if len(sys.argv) < 4 or len(sys.argv) > 5:
-                # pylint: disable=line-too-long
-                print("Should be used as `transformers gpt OPENAI_GPT_CHECKPOINT_FOLDER_PATH PYTORCH_DUMP_OUTPUT [OPENAI_GPT_CONFIG]`")
-            else:
-                OPENAI_GPT_CHECKPOINT_FOLDER_PATH = sys.argv[2]
-                PYTORCH_DUMP_OUTPUT = sys.argv[3]
-                if len(sys.argv) == 5:
-                    OPENAI_GPT_CONFIG = sys.argv[4]
-                else:
-                    OPENAI_GPT_CONFIG = ""
-                convert_openai_checkpoint_to_pytorch(OPENAI_GPT_CHECKPOINT_FOLDER_PATH,
-                                                    OPENAI_GPT_CONFIG,
-                                                    PYTORCH_DUMP_OUTPUT)
-        elif sys.argv[1] == "transfo_xl":
-            try:
-                from .convert_transfo_xl_original_tf_checkpoint_to_pytorch import convert_transfo_xl_checkpoint_to_pytorch
-            except ImportError:
-                print("transformers can only be used from the commandline to convert TensorFlow models in PyTorch, "
-                    "In that case, it requires TensorFlow to be installed. Please see "
-                    "https://www.tensorflow.org/install/ for installation instructions.")
-                raise
-            if len(sys.argv) < 4 or len(sys.argv) > 5:
-                # pylint: disable=line-too-long
-                print("Should be used as `transformers transfo_xl TF_CHECKPOINT/TF_DATASET_FILE PYTORCH_DUMP_OUTPUT [TF_CONFIG]`")
-            else:
-                if 'ckpt' in sys.argv[2].lower():
-                    TF_CHECKPOINT = sys.argv[2]
-                    TF_DATASET_FILE = ""
-                else:
-                    TF_DATASET_FILE = sys.argv[2]
-                    TF_CHECKPOINT = ""
-                PYTORCH_DUMP_OUTPUT = sys.argv[3]
-                if len(sys.argv) == 5:
-                    TF_CONFIG = sys.argv[4]
-                else:
-                    TF_CONFIG = ""
-                convert_transfo_xl_checkpoint_to_pytorch(TF_CHECKPOINT, TF_CONFIG, PYTORCH_DUMP_OUTPUT, TF_DATASET_FILE)
-        elif sys.argv[1] == "gpt2":
-            try:
-                from .convert_gpt2_original_tf_checkpoint_to_pytorch import convert_gpt2_checkpoint_to_pytorch
-            except ImportError:
-                print("transformers can only be used from the commandline to convert TensorFlow models in PyTorch, "
-                    "In that case, it requires TensorFlow to be installed. Please see "
-                    "https://www.tensorflow.org/install/ for installation instructions.")
-                raise
+        # # Register commands
+        # ServeCommand.register_subcommand(commands_parser)
 
-            if len(sys.argv) < 4 or len(sys.argv) > 5:
-                # pylint: disable=line-too-long
-                print("Should be used as `transformers gpt2 TF_CHECKPOINT PYTORCH_DUMP_OUTPUT [TF_CONFIG]`")
-            else:
-                TF_CHECKPOINT = sys.argv[2]
-                PYTORCH_DUMP_OUTPUT = sys.argv[3]
-                if len(sys.argv) == 5:
-                    TF_CONFIG = sys.argv[4]
-                else:
-                    TF_CONFIG = ""
-                convert_gpt2_checkpoint_to_pytorch(TF_CHECKPOINT, TF_CONFIG, PYTORCH_DUMP_OUTPUT)
-        elif sys.argv[1] == "xlnet":
-            try:
-                from .convert_xlnet_original_tf_checkpoint_to_pytorch import convert_xlnet_checkpoint_to_pytorch
-            except ImportError:
-                print("transformers can only be used from the commandline to convert TensorFlow models in PyTorch, "
-                    "In that case, it requires TensorFlow to be installed. Please see "
-                    "https://www.tensorflow.org/install/ for installation instructions.")
-                raise
+        # # Let's go
+        # args = parser.parse_args()
 
-            if len(sys.argv) < 5 or len(sys.argv) > 6:
-                # pylint: disable=line-too-long
-                print("Should be used as `transformers xlnet TF_CHECKPOINT TF_CONFIG PYTORCH_DUMP_OUTPUT [FINETUNING_TASK_NAME]`")
-            else:
-                TF_CHECKPOINT = sys.argv[2]
-                TF_CONFIG = sys.argv[3]
-                PYTORCH_DUMP_OUTPUT = sys.argv[4]
-                if len(sys.argv) == 6:
-                    FINETUNING_TASK = sys.argv[5]
-                else:
-                    FINETUNING_TASK = None
-
-                convert_xlnet_checkpoint_to_pytorch(TF_CHECKPOINT,
-                                                    TF_CONFIG,
-                                                    PYTORCH_DUMP_OUTPUT,
-                                                    FINETUNING_TASK)
-        elif sys.argv[1] == "xlm":
-            from .convert_xlm_original_pytorch_checkpoint_to_pytorch import convert_xlm_checkpoint_to_pytorch
-
-            if len(sys.argv) != 4:
-                # pylint: disable=line-too-long
-                print("Should be used as `transformers xlm XLM_CHECKPOINT_PATH PYTORCH_DUMP_OUTPUT`")
-            else:
-                XLM_CHECKPOINT_PATH = sys.argv[2]
-                PYTORCH_DUMP_OUTPUT = sys.argv[3]
-
-                convert_xlm_checkpoint_to_pytorch(XLM_CHECKPOINT_PATH, PYTORCH_DUMP_OUTPUT)
+        # if not hasattr(args, 'func'):
+        #     parser.print_help()
+        #     exit(1)
+        # # Run
+        # service = args.func(args)
+        # service.run()
 
 if __name__ == '__main__':
     main()
diff --git a/transformers/commands/convert.py b/transformers/commands/convert.py
new file mode 100644
index 0000000000..55dbf53734
--- /dev/null
+++ b/transformers/commands/convert.py
@@ -0,0 +1,115 @@
+from argparse import ArgumentParser, Namespace
+
+from logging import getLogger
+
+from transformers import AutoModel, AutoTokenizer
+from transformers.commands import BaseTransformersCLICommand
+
+
+def convert_command_factory(args: Namespace):
+    """
+    Factory function used to convert a model TF 1.0 checkpoint in a PyTorch checkpoint.
+    :return: ServeCommand
+    """
+    return ConvertCommand(args.model_type, args.tf_checkpoint, args.pytorch_dump_output,
+                          args.config, args.finetuning_task_name)
+
+
+class ConvertCommand(BaseTransformersCLICommand):
+
+    @staticmethod
+    def register_subcommand(parser: ArgumentParser):
+        """
+        Register this command to argparse so it's available for the transformer-cli
+        :param parser: Root parser to register command-specific arguments
+        :return:
+        """
+        train_parser = parser.add_parser('convert', help="CLI tool to run convert model from original "
+                                                         "author checkpoints to Transformesr PyTorch checkpoints.")
+        train_parser.add_argument('--model_type', type=str, required=True,
+                                  help='Model\'s type.')
+        train_parser.add_argument('--tf_checkpoint', type=str, required=True,
+                                  help='TensorFlow checkpoint path or folder.')
+        train_parser.add_argument('--pytorch_dump_output', type=str, required=True,
+                                  help='Path to the PyTorch savd model output.')
+        train_parser.add_argument('--config', type=str, default="",
+                                  help='Configuration file path or folder.')
+        train_parser.add_argument('--finetuning_task_name', type=str, default=None,
+                                  help='Optional fine-tuning task name if the TF model was a finetuned model.')
+        train_parser.set_defaults(func=convert_command_factory)
+
+    def __init__(self, model_type: str, tf_checkpoint: str, pytorch_dump_output: str,
+                 config: str, finetuning_task_name: str, *args):
+        self._logger = getLogger('transformers-cli/converting')
+
+        self._logger.info('Loading model {}'.format(model_type))
+        self._model_type = model_type
+        self._tf_checkpoint = tf_checkpoint
+        self._pytorch_dump_output = pytorch_dump_output
+        self._config = config
+        self._finetuning_task_name = finetuning_task_name
+
+    def run(self):
+        if self._model_type == "bert":
+            try:
+                from transformers.convert_bert_original_tf_checkpoint_to_pytorch import convert_tf_checkpoint_to_pytorch
+            except ImportError:
+                msg = "transformers can only be used from the commandline to convert TensorFlow models in PyTorch, " \
+                    "In that case, it requires TensorFlow to be installed. Please see " \
+                    "https://www.tensorflow.org/install/ for installation instructions."
+                raise ImportError(msg)
+
+            convert_tf_checkpoint_to_pytorch(self._tf_checkpoint, self._config, self._pytorch_dump_output)
+        elif self._model_type == "gpt":
+            from transformers.convert_openai_original_tf_checkpoint_to_pytorch import convert_openai_checkpoint_to_pytorch
+            convert_openai_checkpoint_to_pytorch(self._tf_checkpoint,
+                                                    self._config,
+                                                    self._pytorch_dump_output)
+        elif self._model_type == "transfo_xl":
+            try:
+                from transformers.convert_transfo_xl_original_tf_checkpoint_to_pytorch import convert_transfo_xl_checkpoint_to_pytorch
+            except ImportError:
+                msg = "transformers can only be used from the commandline to convert TensorFlow models in PyTorch, " \
+                    "In that case, it requires TensorFlow to be installed. Please see " \
+                    "https://www.tensorflow.org/install/ for installation instructions."
+                raise ImportError(msg)
+
+            if 'ckpt' in self._tf_checkpoint.lower():
+                TF_CHECKPOINT = self._tf_checkpoint
+                TF_DATASET_FILE = ""
+            else:
+                TF_DATASET_FILE = self._tf_checkpoint
+                TF_CHECKPOINT = ""
+            convert_transfo_xl_checkpoint_to_pytorch(TF_CHECKPOINT,
+                                                        self._config,
+                                                        self._pytorch_dump_output,
+                                                        TF_DATASET_FILE)
+        elif self._model_type == "gpt2":
+            try:
+                from transformers.convert_gpt2_original_tf_checkpoint_to_pytorch import convert_gpt2_checkpoint_to_pytorch
+            except ImportError:
+                msg = "transformers can only be used from the commandline to convert TensorFlow models in PyTorch, " \
+                    "In that case, it requires TensorFlow to be installed. Please see " \
+                    "https://www.tensorflow.org/install/ for installation instructions."
+                raise ImportError(msg)
+
+            convert_gpt2_checkpoint_to_pytorch(self._tf_checkpoint, self._config, self._pytorch_dump_output)
+        elif self._model_type == "xlnet":
+            try:
+                from transformers.convert_xlnet_original_tf_checkpoint_to_pytorch import convert_xlnet_checkpoint_to_pytorch
+            except ImportError:
+                msg = "transformers can only be used from the commandline to convert TensorFlow models in PyTorch, " \
+                    "In that case, it requires TensorFlow to be installed. Please see " \
+                    "https://www.tensorflow.org/install/ for installation instructions."
+                raise ImportError(msg)
+
+            convert_xlnet_checkpoint_to_pytorch(self._tf_checkpoint,
+                                                self._config,
+                                                self._pytorch_dump_output,
+                                                self._finetuning_task_name)
+        elif self._model_type == "xlm":
+            from transformers.convert_xlm_original_pytorch_checkpoint_to_pytorch import convert_xlm_checkpoint_to_pytorch
+
+            convert_xlm_checkpoint_to_pytorch(self._tf_checkpoint, self._pytorch_dump_output)
+        else:
+            raise ValueError("--model_type should be selected in the list [bert, gpt, gpt2, transfo_xl, xlnet, xlm]")
diff --git a/transformers/commands/serving.py b/transformers/commands/serving.py
new file mode 100644
index 0000000000..0b47246ead
--- /dev/null
+++ b/transformers/commands/serving.py
@@ -0,0 +1,176 @@
+from argparse import ArgumentParser, Namespace
+from typing import List, Optional, Union, Any
+
+import torch
+from fastapi import FastAPI, HTTPException, Body
+from logging import getLogger
+
+from pydantic import BaseModel
+from uvicorn import run
+
+from transformers import AutoModel, AutoTokenizer, AutoConfig
+from transformers.commands import BaseTransformersCLICommand
+
+
+def serve_command_factory(args: Namespace):
+    """
+    Factory function used to instantiate serving server from provided command line arguments.
+    :return: ServeCommand
+    """
+    return ServeCommand(args.host, args.port, args.model, args.graphql)
+
+
+class ServeResult(BaseModel):
+    """
+    Base class for serving result
+    """
+    model: str
+
+
+class ServeModelInfoResult(ServeResult):
+    """
+    Expose model information
+    """
+    infos: dict
+
+
+class ServeTokenizeResult(ServeResult):
+    """
+    Tokenize result model
+    """
+    tokens: List[str]
+    tokens_ids: Optional[List[int]]
+
+
+class ServeDeTokenizeResult(ServeResult):
+    """
+    DeTokenize result model
+    """
+    text: str
+
+
+class ServeForwardResult(ServeResult):
+    """
+    Forward result model
+    """
+    tokens: List[str]
+    tokens_ids: List[int]
+    output: Any
+
+
+class ServeCommand(BaseTransformersCLICommand):
+
+    @staticmethod
+    def register_subcommand(parser: ArgumentParser):
+        """
+        Register this command to argparse so it's available for the transformer-cli
+        :param parser: Root parser to register command-specific arguments
+        :return:
+        """
+        serve_parser = parser.add_parser('serve', help='CLI tool to run inference requests through REST and GraphQL endpoints.')
+        serve_parser.add_argument('--host', type=str, default='localhost', help='Interface the server will listen on.')
+        serve_parser.add_argument('--port', type=int, default=8888, help='Port the serving will listen to.')
+        serve_parser.add_argument('--model', type=str, required=True, help='Model\'s name or path to stored model to infer from.')
+        serve_parser.add_argument('--graphql', action='store_true', default=False, help='Enable GraphQL endpoints.')
+        serve_parser.set_defaults(func=serve_command_factory)
+
+    def __init__(self, host: str, port: int, model: str, graphql: bool):
+        self._logger = getLogger('transformers-cli/serving')
+
+        self._logger.info('Loading model {}'.format(model))
+        self._model_name = model
+        self._model = AutoModel.from_pretrained(model)
+        self._tokenizer = AutoTokenizer.from_pretrained(model)
+
+        self._logger.info('Serving model over {}:{}'.format(host, port))
+        self._host = host
+        self._port = port
+        self._app = FastAPI()
+
+        # Register routes
+        self._app.add_api_route('/', self.model_info, response_model=ServeModelInfoResult, methods=['GET'])
+        self._app.add_api_route('/tokenize', self.tokenize, response_model=ServeTokenizeResult, methods=['POST'])
+        self._app.add_api_route('/detokenize', self.detokenize, response_model=ServeDeTokenizeResult, methods=['POST'])
+        self._app.add_api_route('/forward', self.forward, response_model=ServeForwardResult, methods=['POST'])
+
+    def run(self):
+        run(self._app, host=self._host, port=self._port)
+
+    def model_info(self):
+        return ServeModelInfoResult(model=self._model_name, infos=vars(self._model.config))
+
+    def tokenize(self, text_input: str = Body(None, embed=True), return_ids: bool = Body(False, embed=True)):
+        """
+        Tokenize the provided input and eventually returns corresponding tokens id:
+        - **text_input**: String to tokenize
+        - **return_ids**: Boolean flags indicating if the tokens have to be converted to their integer mapping.
+        """
+        try:
+            tokens_txt = self._tokenizer.tokenize(text_input)
+
+            if return_ids:
+                tokens_ids = self._tokenizer.convert_tokens_to_ids(tokens_txt)
+                return ServeTokenizeResult(model=self._model_name, tokens=tokens_txt, tokens_ids=tokens_ids)
+            else:
+                return ServeTokenizeResult(model=self._model_name, tokens=tokens_txt)
+
+        except Exception as e:
+            raise HTTPException(status_code=500, detail={"model": self._model_name, "error": str(e)})
+
+    def detokenize(self, tokens_ids: List[int] = Body(None, embed=True),
+                   skip_special_tokens: bool = Body(False, embed=True),
+                   cleanup_tokenization_spaces: bool = Body(True, embed=True)):
+        """
+        Detokenize the provided tokens ids to readable text:
+        - **tokens_ids**: List of tokens ids
+        - **skip_special_tokens**: Flag indicating to not try to decode special tokens
+        - **cleanup_tokenization_spaces**: Flag indicating to remove all leading/trailing spaces and intermediate ones.
+        """
+        try:
+            decoded_str = self._tokenizer.decode(tokens_ids, skip_special_tokens, cleanup_tokenization_spaces)
+            return ServeDeTokenizeResult(model=self._model_name, text=decoded_str)
+        except Exception as e:
+            raise HTTPException(status_code=500, detail={"model": self._model_name, "error": str(e)})
+
+    def forward(self, inputs: Union[str, List[str], List[int]] = Body(None, embed=True),
+                attention_mask: Optional[List[int]] = Body(None, embed=True),
+                tokens_type_ids: Optional[List[int]] = Body(None, embed=True)):
+        """
+        **inputs**:
+        **attention_mask**:
+        **tokens_type_ids**:
+        """
+
+        # Check we don't have empty string
+        if len(inputs) == 0:
+            return ServeForwardResult(model=self._model_name, output=[], attention=[])
+
+        if isinstance(inputs, str):
+            inputs_tokens = self._tokenizer.tokenize(inputs)
+            inputs_ids = self._tokenizer.convert_tokens_to_ids(inputs_tokens)
+
+        elif isinstance(inputs, List):
+            if isinstance(inputs[0], str):
+                inputs_tokens = inputs
+                inputs_ids = self._tokenizer.convert_tokens_to_ids(inputs_tokens)
+            elif isinstance(inputs[0], int):
+                inputs_tokens = []
+                inputs_ids = inputs
+            else:
+                error_msg = "inputs should be string, [str] of [int] (got {})".format(type(inputs[0]))
+                raise HTTPException(423, detail={"error": error_msg})
+        else:
+            error_msg = "inputs should be string, [str] of [int] (got {})".format(type(inputs))
+            raise HTTPException(423, detail={"error": error_msg})
+
+        try:
+            # Forward through the model
+            t_input_ids = torch.tensor(inputs_ids).unsqueeze(0)
+            output = self._model(t_input_ids, attention_mask, tokens_type_ids)
+
+            return ServeForwardResult(
+                model=self._model_name, tokens=inputs_tokens,
+                tokens_ids=inputs_ids, output=output[0].tolist()
+            )
+        except Exception as e:
+            raise HTTPException(500, {"error": str(e)})
diff --git a/transformers/commands/train.py b/transformers/commands/train.py
new file mode 100644
index 0000000000..7fb3a54d25
--- /dev/null
+++ b/transformers/commands/train.py
@@ -0,0 +1,121 @@
+from argparse import ArgumentParser, Namespace
+
+from logging import getLogger
+
+from transformers.commands import BaseTransformersCLICommand
+from transformers import (AutoTokenizer, is_tf_available, is_torch_available,
+                          SingleSentenceClassificationProcessor,
+                          convert_examples_to_features)
+if is_tf_available():
+    from transformers import TFAutoModelForSequenceClassification as SequenceClassifModel
+elif is_torch_available():
+    from transformers import AutoModelForSequenceClassification as SequenceClassifModel
+else:
+    raise ImportError("At least one of PyTorch or TensorFlow 2.0+ should be installed to use CLI training")
+
+# TF training parameters
+BATCH_SIZE = 32
+EVAL_BATCH_SIZE = BATCH_SIZE * 2
+USE_XLA = False
+USE_AMP = False
+
+def train_command_factory(args: Namespace):
+    """
+    Factory function used to instantiate serving server from provided command line arguments.
+    :return: ServeCommand
+    """
+    return TrainCommand(args.model)
+
+
+class TrainCommand(BaseTransformersCLICommand):
+
+    @staticmethod
+    def register_subcommand(parser: ArgumentParser):
+        """
+        Register this command to argparse so it's available for the transformer-cli
+        :param parser: Root parser to register command-specific arguments
+        :return:
+        """
+        train_parser = parser.add_parser('train', help='CLI tool to train a model on a task.')
+        train_parser.add_argument('--train_data', type=str, required=True,
+                                  help='path to train (and optionally evaluation) dataset.')
+        train_parser.add_argument('--task', type=str, default='text_classification',
+                                  help='Task to train the model on.')
+        train_parser.add_argument('--model', type=str, default='bert-base-uncased',
+                                  help='Model\'s name or path to stored model.')
+        train_parser.add_argument('--valid_data', type=str, default='',
+                                  help='path to validation dataset.')
+        train_parser.add_argument('--valid_data_ratio', type=float, default=0.1,
+                                  help="if validation dataset is not provided, fraction of train dataset "
+                                       "to use as validation dataset.")
+        train_parser.set_defaults(func=train_command_factory)
+
+    def __init__(self, model_name: str, task: str, train_data: str,
+                 valid_data: str, valid_data_ratio: float):
+        self._logger = getLogger('transformers-cli/training')
+
+        self._framework = 'tf' if is_tf_available() else 'torch'
+
+        self._logger.info('Loading model {}'.format(model_name))
+        self._model_name = model_name
+        self._tokenizer = AutoTokenizer.from_pretrained(model_name)
+        if task == 'text_classification':
+            self._model = SequenceClassifModel.from_pretrained(model_name)
+        elif task == 'token_classification':
+            raise NotImplementedError
+        elif task == 'question_answering':
+            raise NotImplementedError
+
+        dataset = SingleSentenceClassificationProcessor.create_from_csv(train_data)
+        num_data_samples = len(SingleSentenceClassificationProcessor)
+        if valid_data:
+            self._train_dataset = dataset
+            self._num_train_samples = num_data_samples
+            self._valid_dataset = SingleSentenceClassificationProcessor.create_from_csv(valid_data)
+            self._num_valid_samples = len(self._valid_dataset)
+        else:
+            assert 0.0 < valid_data_ratio < 1.0, "--valid_data_ratio should be between 0.0 and 1.0"
+            self._num_valid_samples = num_data_samples * valid_data_ratio
+            self._num_train_samples = num_data_samples - self._num_valid_samples
+            self._train_dataset = dataset[self._num_train_samples]
+            self._valid_dataset = dataset[self._num_valid_samples]
+
+    def run(self):
+        if self._framework == 'tf':
+            return self.run_tf()
+        return self.run_torch()
+
+    def run_torch(self):
+        raise NotImplementedError
+
+    def run_tf(self):
+        import tensorflow as tf
+
+        tf.config.optimizer.set_jit(USE_XLA)
+        tf.config.optimizer.set_experimental_options({"auto_mixed_precision": USE_AMP})
+
+        # Prepare dataset as a tf.train_data.Dataset instance
+        train_dataset = convert_examples_to_features(self._train_dataset, self._tokenizer, mode='sequence_classification')
+        valid_dataset = convert_examples_to_features(self._valid_dataset, self._tokenizer, mode='sequence_classification')
+        train_dataset = train_dataset.shuffle(128).batch(BATCH_SIZE).repeat(-1)
+        valid_dataset = valid_dataset.batch(EVAL_BATCH_SIZE)
+
+        # Prepare training: Compile tf.keras model with optimizer, loss and learning rate schedule 
+        opt = tf.keras.optimizers.Adam(learning_rate=3e-5, epsilon=1e-08)
+        if USE_AMP:
+            # loss scaling is currently required when using mixed precision
+            opt = tf.keras.mixed_precision.experimental.LossScaleOptimizer(opt, 'dynamic')
+        loss = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True)
+        metric = tf.keras.metrics.SparseCategoricalAccuracy('accuracy')
+        model.compile(optimizer=opt, loss=loss, metrics=[metric])
+
+        # Train and evaluate using tf.keras.Model.fit()
+        train_steps = train_examples//BATCH_SIZE
+        valid_steps = valid_examples//EVAL_BATCH_SIZE
+
+        history = model.fit(train_dataset, epochs=2, steps_per_epoch=train_steps,
+                            validation_data=valid_dataset, validation_steps=valid_steps)
+
+        # Save TF2 model
+        os.makedirs('./save/', exist_ok=True)
+        model.save_pretrained('./save/')
diff --git a/transformers/data/__init__.py b/transformers/data/__init__.py
index 270a053268..5567952fd2 100644
--- a/transformers/data/__init__.py
+++ b/transformers/data/__init__.py
@@ -1,4 +1,4 @@
-from .processors import InputExample, InputFeatures, DataProcessor, SquadFeatures
+from .processors import InputExample, InputFeatures, DataProcessor, SquadFeatures, SingleSentenceClassificationProcessor
 from .processors import glue_output_modes, glue_processors, glue_tasks_num_labels, glue_convert_examples_to_features
 from .processors import squad_convert_examples_to_features, SquadExample, SquadV1Processor, SquadV2Processor
 from .processors import xnli_output_modes, xnli_processors, xnli_tasks_num_labels
diff --git a/transformers/data/processors/__init__.py b/transformers/data/processors/__init__.py
index 0f1b24893a..0cef0080f4 100644
--- a/transformers/data/processors/__init__.py
+++ b/transformers/data/processors/__init__.py
@@ -1,4 +1,4 @@
-from .utils import InputExample, InputFeatures, DataProcessor
+from .utils import InputExample, InputFeatures, DataProcessor, SingleSentenceClassificationProcessor, convert_examples_to_features
 from .glue import glue_output_modes, glue_processors, glue_tasks_num_labels, glue_convert_examples_to_features
 from .squad import squad_convert_examples_to_features, SquadFeatures, SquadExample, SquadV1Processor, SquadV2Processor
 from .xnli import xnli_output_modes, xnli_processors, xnli_tasks_num_labels
\ No newline at end of file
diff --git a/transformers/data/processors/utils.py b/transformers/data/processors/utils.py
index 07bdf3150c..39544a1239 100644
--- a/transformers/data/processors/utils.py
+++ b/transformers/data/processors/utils.py
@@ -125,3 +125,185 @@ class DataProcessor(object):
                     line = list(unicode(cell, 'utf-8') for cell in line)
                 lines.append(line)
             return lines
+
+
+class SingleSentenceClassificationProcessor(DataProcessor):
+    """ Generic processor for a single sentence classification data set."""
+    def __init__(self, labels=None, examples=None):
+        self.labels = [] if labels is None else labels
+        self.examples = [] if examples is None else examples
+
+    @classmethod
+    def create_from_csv(cls, file_name):
+        processor = cls()
+        processor.add_examples_from_csv(file_name)
+        return processor
+
+    def __len__(self):
+        return len(self.examples)
+
+    def __getitem__(self, idx):
+        if isinstance(idx, slice):
+            return SingleSentenceClassificationProcessor(labels=self.labels,
+                                                         examples=self.examples[idx])
+        return self.examples[idx]
+
+    def get_labels(self):
+        """Gets the list of labels for this data set."""
+        return self.labels
+
+    def add_examples_from_csv(self, file_name):
+        lines = self._read_tsv(file_name)
+        self.add_examples_from_lines(lines)
+
+    def add_examples_from_lines(self, lines, split_name='', overwrite_labels=False, overwrite_examples=False):
+        """Creates examples for the training and dev sets."""
+        added_labels = set()
+        examples = []
+        for (i, line) in enumerate(lines):
+            if len(line) > 2:
+                guid = "%s-%s" % (split_name, line[0]) if split_name else line[0]
+                label = line[1]
+                text_a = line[2]
+            else:
+                guid = "%s-%s" % (split_name, i) if split_name else "%s" % i
+                label = line[0]
+                text_a = line[1]
+
+            added_labels.add(label)
+            examples.append(InputExample(guid=guid, text_a=text_a, text_b=None, label=label))
+
+        # Update examples
+        if overwrite_examples:
+            self.examples = examples
+        else:
+            self.examples.extend(examples)
+
+        # Update labels
+        if overwrite_labels:
+            self.labels = list(added_labels)
+        else:
+            self.labels = list(set(self.labels).union(added_labels))
+
+        return self.examples
+
+
+def convert_examples_to_features(examples, tokenizer,
+                                 mode='sequence_classification',   
+                                 max_length=512,
+                                 pad_on_left=False,
+                                 pad_token=0,
+                                 pad_token_segment_id=0,
+                                 mask_padding_with_zero=True):
+    """
+    Loads a data file into a list of ``InputFeatures``
+
+    Args:
+        examples: List of ``InputExamples`` or ``tf.data.Dataset`` containing the examples.
+        tokenizer: Instance of a tokenizer that will tokenize the examples
+        max_length: Maximum example length
+        task: GLUE task
+        label_list: List of labels. Can be obtained from the processor using the ``processor.get_labels()`` method
+        output_mode: String indicating the output mode. Either ``regression`` or ``classification``
+        pad_on_left: If set to ``True``, the examples will be padded on the left rather than on the right (default)
+        pad_token: Padding token
+        pad_token_segment_id: The segment ID for the padding token (It is usually 0, but can vary such as for XLNet where it is 4)
+        mask_padding_with_zero: If set to ``True``, the attention mask will be filled by ``1`` for actual values
+            and by ``0`` for padded values. If set to ``False``, inverts it (``1`` for padded values, ``0`` for
+            actual values)
+
+    Returns:
+        If the ``examples`` input is a ``tf.data.Dataset``, will return a ``tf.data.Dataset``
+        containing the task-specific features. If the input is a list of ``InputExamples``, will return
+        a list of task-specific ``InputFeatures`` which can be fed to the model.
+
+    """
+    is_tf_dataset = False
+    if is_tf_available() and isinstance(examples, tf.data.Dataset):
+        is_tf_dataset = True
+
+    if task is not None:
+        processor = glue_processors[task]()
+        if label_list is None:
+            label_list = processor.get_labels()
+            logger.info("Using label list %s for task %s" % (label_list, task))
+        if output_mode is None:
+            output_mode = glue_output_modes[task]
+            logger.info("Using output mode %s for task %s" % (output_mode, task))
+
+    label_map = {label: i for i, label in enumerate(label_list)}
+
+    features = []
+    for (ex_index, example) in enumerate(examples):
+        if ex_index % 10000 == 0:
+            logger.info("Writing example %d" % (ex_index))
+        if is_tf_dataset:
+            example = processor.get_example_from_tensor_dict(example)
+
+        inputs = tokenizer.encode_plus(
+            example.text_a,
+            example.text_b,
+            add_special_tokens=True,
+            max_length=max_length,
+        )
+        input_ids, token_type_ids = inputs["input_ids"], inputs["token_type_ids"]
+
+        # The mask has 1 for real tokens and 0 for padding tokens. Only real
+        # tokens are attended to.
+        attention_mask = [1 if mask_padding_with_zero else 0] * len(input_ids)
+
+        # Zero-pad up to the sequence length.
+        padding_length = max_length - len(input_ids)
+        if pad_on_left:
+            input_ids = ([pad_token] * padding_length) + input_ids
+            attention_mask = ([0 if mask_padding_with_zero else 1] * padding_length) + attention_mask
+            token_type_ids = ([pad_token_segment_id] * padding_length) + token_type_ids
+        else:
+            input_ids = input_ids + ([pad_token] * padding_length)
+            attention_mask = attention_mask + ([0 if mask_padding_with_zero else 1] * padding_length)
+            token_type_ids = token_type_ids + ([pad_token_segment_id] * padding_length)
+
+        assert len(input_ids) == max_length, "Error with input length {} vs {}".format(len(input_ids), max_length)
+        assert len(attention_mask) == max_length, "Error with input length {} vs {}".format(len(attention_mask), max_length)
+        assert len(token_type_ids) == max_length, "Error with input length {} vs {}".format(len(token_type_ids), max_length)
+
+        if output_mode == "classification":
+            label = label_map[example.label]
+        elif output_mode == "regression":
+            label = float(example.label)
+        else:
+            raise KeyError(output_mode)
+
+        if ex_index < 5:
+            logger.info("*** Example ***")
+            logger.info("guid: %s" % (example.guid))
+            logger.info("input_ids: %s" % " ".join([str(x) for x in input_ids]))
+            logger.info("attention_mask: %s" % " ".join([str(x) for x in attention_mask]))
+            logger.info("token_type_ids: %s" % " ".join([str(x) for x in token_type_ids]))
+            logger.info("label: %s (id = %d)" % (example.label, label))
+
+        features.append(
+                InputFeatures(input_ids=input_ids,
+                              attention_mask=attention_mask,
+                              token_type_ids=token_type_ids,
+                              label=label))
+
+    if is_tf_available() and is_tf_dataset:
+        def gen():
+            for ex in features:
+                yield  ({'input_ids': ex.input_ids,
+                         'attention_mask': ex.attention_mask,
+                         'token_type_ids': ex.token_type_ids},
+                        ex.label)
+
+        return tf.data.Dataset.from_generator(gen,
+            ({'input_ids': tf.int32,
+              'attention_mask': tf.int32,
+              'token_type_ids': tf.int32},
+             tf.int64),
+            ({'input_ids': tf.TensorShape([None]),
+              'attention_mask': tf.TensorShape([None]),
+              'token_type_ids': tf.TensorShape([None])},
+             tf.TensorShape([])))
+
+    return features

From 2d8559731acbf673fe3e31aaeb17412a342cff73 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Wed, 16 Oct 2019 23:19:45 +0200
Subject: [PATCH 286/505] add pipeline - train

---
 transformers/commands/train.py        | 127 +++++++----
 transformers/data/processors/utils.py | 313 +++++++++++++-------------
 transformers/pipeline.py              | 254 +++++++++++++++++++++
 3 files changed, 485 insertions(+), 209 deletions(-)
 create mode 100644 transformers/pipeline.py

diff --git a/transformers/commands/train.py b/transformers/commands/train.py
index 7fb3a54d25..fc89d48594 100644
--- a/transformers/commands/train.py
+++ b/transformers/commands/train.py
@@ -1,5 +1,5 @@
+import os
 from argparse import ArgumentParser, Namespace
-
 from logging import getLogger
 
 from transformers.commands import BaseTransformersCLICommand
@@ -14,8 +14,6 @@ else:
     raise ImportError("At least one of PyTorch or TensorFlow 2.0+ should be installed to use CLI training")
 
 # TF training parameters
-BATCH_SIZE = 32
-EVAL_BATCH_SIZE = BATCH_SIZE * 2
 USE_XLA = False
 USE_AMP = False
 
@@ -24,7 +22,7 @@ def train_command_factory(args: Namespace):
     Factory function used to instantiate serving server from provided command line arguments.
     :return: ServeCommand
     """
-    return TrainCommand(args.model)
+    return TrainCommand(args)
 
 
 class TrainCommand(BaseTransformersCLICommand):
@@ -38,50 +36,84 @@ class TrainCommand(BaseTransformersCLICommand):
         """
         train_parser = parser.add_parser('train', help='CLI tool to train a model on a task.')
         train_parser.add_argument('--train_data', type=str, required=True,
-                                  help='path to train (and optionally evaluation) dataset.')
+                                  help="path to train (and optionally evaluation) dataset as a csv with "
+                                       "tab separated labels and sentences.")
+
+        train_parser.add_argument('--column_label', type=int, default=0,
+                                  help='Column of the dataset csv file with example labels.')
+        train_parser.add_argument('--column_text', type=int, default=1,
+                                  help='Column of the dataset csv file with example texts.')
+        train_parser.add_argument('--column_id', type=int, default=2,
+                                  help='Column of the dataset csv file with example ids.')
+
+        train_parser.add_argument('--validation_data', type=str, default='',
+                                  help='path to validation dataset.')
+        train_parser.add_argument('--validation_split', type=float, default=0.1,
+                                  help="if validation dataset is not provided, fraction of train dataset "
+                                       "to use as validation dataset.")
+
+        train_parser.add_argument('--output', type=str, default='./',
+                                  help='path to saved the trained model.')
+
         train_parser.add_argument('--task', type=str, default='text_classification',
                                   help='Task to train the model on.')
         train_parser.add_argument('--model', type=str, default='bert-base-uncased',
                                   help='Model\'s name or path to stored model.')
-        train_parser.add_argument('--valid_data', type=str, default='',
-                                  help='path to validation dataset.')
-        train_parser.add_argument('--valid_data_ratio', type=float, default=0.1,
-                                  help="if validation dataset is not provided, fraction of train dataset "
-                                       "to use as validation dataset.")
+        train_parser.add_argument('--train_batch_size', type=int, default=32,
+                                  help='Batch size for training.')
+        train_parser.add_argument('--valid_batch_size', type=int, default=64,
+                                  help='Batch size for validation.')
+        train_parser.add_argument('--learning_rate', type=float, default=3e-5,
+                                  help="Learning rate.")
+        train_parser.add_argument('--adam_epsilon', type=float, default=1e-08,
+                                  help="Epsilon for Adam optimizer.")
         train_parser.set_defaults(func=train_command_factory)
 
-    def __init__(self, model_name: str, task: str, train_data: str,
-                 valid_data: str, valid_data_ratio: float):
-        self._logger = getLogger('transformers-cli/training')
+    def __init__(self, args: Namespace):
+        self.logger = getLogger('transformers-cli/training')
 
-        self._framework = 'tf' if is_tf_available() else 'torch'
+        self.framework = 'tf' if is_tf_available() else 'torch'
 
-        self._logger.info('Loading model {}'.format(model_name))
-        self._model_name = model_name
-        self._tokenizer = AutoTokenizer.from_pretrained(model_name)
-        if task == 'text_classification':
-            self._model = SequenceClassifModel.from_pretrained(model_name)
-        elif task == 'token_classification':
+        os.makedirs(args.output)
+        self.output = args.output
+
+        self.column_label = args.column_label
+        self.column_text = args.column_text
+        self.column_id = args.column_id
+
+        self.logger.info('Loading model {}'.format(args.model_name))
+        self.model_name = args.model_name
+        self.tokenizer = AutoTokenizer.from_pretrained(args.model_name)
+        if args.task == 'text_classification':
+            self.model = SequenceClassifModel.from_pretrained(args.model_name)
+        elif args.task == 'token_classification':
             raise NotImplementedError
-        elif task == 'question_answering':
+        elif args.task == 'question_answering':
             raise NotImplementedError
 
-        dataset = SingleSentenceClassificationProcessor.create_from_csv(train_data)
-        num_data_samples = len(SingleSentenceClassificationProcessor)
-        if valid_data:
-            self._train_dataset = dataset
-            self._num_train_samples = num_data_samples
-            self._valid_dataset = SingleSentenceClassificationProcessor.create_from_csv(valid_data)
-            self._num_valid_samples = len(self._valid_dataset)
+        self.logger.info('Loading dataset from {}'.format(args.train_data))
+        dataset = SingleSentenceClassificationProcessor.create_from_csv(args.train_data)
+        num_data_samples = len(dataset)
+        if args.validation_data:
+            self.logger.info('Loading validation dataset from {}'.format(args.validation_data))
+            self.valid_dataset = SingleSentenceClassificationProcessor.create_from_csv(args.validation_data)
+            self.num_valid_samples = len(self.valid_dataset)
+            self.train_dataset = dataset
+            self.num_train_samples = num_data_samples
         else:
-            assert 0.0 < valid_data_ratio < 1.0, "--valid_data_ratio should be between 0.0 and 1.0"
-            self._num_valid_samples = num_data_samples * valid_data_ratio
-            self._num_train_samples = num_data_samples - self._num_valid_samples
-            self._train_dataset = dataset[self._num_train_samples]
-            self._valid_dataset = dataset[self._num_valid_samples]
+            assert 0.0 < args.validation_split < 1.0, "--validation_split should be between 0.0 and 1.0"
+            self.num_valid_samples = num_data_samples * args.validation_split
+            self.num_train_samples = num_data_samples - self.num_valid_samples
+            self.train_dataset = dataset[self.num_train_samples]
+            self.valid_dataset = dataset[self.num_valid_samples]
+
+        self.train_batch_size = args.train_batch_size
+        self.valid_batch_size = args.valid_batch_size
+        self.learning_rate = args.learning_rate
+        self.adam_epsilon = args.adam_epsilon
 
     def run(self):
-        if self._framework == 'tf':
+        if self.framework == 'tf':
             return self.run_tf()
         return self.run_torch()
 
@@ -95,27 +127,28 @@ class TrainCommand(BaseTransformersCLICommand):
         tf.config.optimizer.set_experimental_options({"auto_mixed_precision": USE_AMP})
 
         # Prepare dataset as a tf.train_data.Dataset instance
-        train_dataset = convert_examples_to_features(self._train_dataset, self._tokenizer, mode='sequence_classification')
-        valid_dataset = convert_examples_to_features(self._valid_dataset, self._tokenizer, mode='sequence_classification')
-        train_dataset = train_dataset.shuffle(128).batch(BATCH_SIZE).repeat(-1)
-        valid_dataset = valid_dataset.batch(EVAL_BATCH_SIZE)
+        self.logger.info('Tokenizing and processing dataset')
+        train_dataset = self.train_dataset.get_features(self.tokenizer)
+        valid_dataset = self.valid_dataset.get_features(self.tokenizer)
+        train_dataset = train_dataset.shuffle(128).batch(self.train_batch_size).repeat(-1)
+        valid_dataset = valid_dataset.batch(self.valid_batch_size)
 
         # Prepare training: Compile tf.keras model with optimizer, loss and learning rate schedule 
-        opt = tf.keras.optimizers.Adam(learning_rate=3e-5, epsilon=1e-08)
+        opt = tf.keras.optimizers.Adam(learning_rate=args.learning_rate, epsilon=self.adam_epsilon)
         if USE_AMP:
             # loss scaling is currently required when using mixed precision
             opt = tf.keras.mixed_precision.experimental.LossScaleOptimizer(opt, 'dynamic')
         loss = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True)
         metric = tf.keras.metrics.SparseCategoricalAccuracy('accuracy')
-        model.compile(optimizer=opt, loss=loss, metrics=[metric])
+        self.model.compile(optimizer=opt, loss=loss, metrics=[metric])
 
         # Train and evaluate using tf.keras.Model.fit()
-        train_steps = train_examples//BATCH_SIZE
-        valid_steps = valid_examples//EVAL_BATCH_SIZE
+        train_steps = self.num_train_samples//self.train_batch_size
+        valid_steps = self.num_valid_samples//self.valid_batch_size
 
-        history = model.fit(train_dataset, epochs=2, steps_per_epoch=train_steps,
-                            validation_data=valid_dataset, validation_steps=valid_steps)
+        self.logger.info('Training model')
+        history = self.model.fit(train_dataset, epochs=2, steps_per_epoch=train_steps,
+                                 validation_data=valid_dataset, validation_steps=valid_steps)
 
-        # Save TF2 model
-        os.makedirs('./save/', exist_ok=True)
-        model.save_pretrained('./save/')
+        # Save trained model
+        self.model.save_pretrained(self.output)
diff --git a/transformers/data/processors/utils.py b/transformers/data/processors/utils.py
index 39544a1239..75bed86042 100644
--- a/transformers/data/processors/utils.py
+++ b/transformers/data/processors/utils.py
@@ -18,6 +18,11 @@ import csv
 import sys
 import copy
 import json
+import logging
+
+from ...file_utils import is_tf_available, is_torch_available
+
+logger = logging.getLogger(__name__)
 
 class InputExample(object):
     """
@@ -64,7 +69,7 @@ class InputFeatures(object):
         label: Label corresponding to the input
     """
 
-    def __init__(self, input_ids, attention_mask, token_type_ids, label):
+    def __init__(self, input_ids, attention_mask=None, token_type_ids=None, label=None):
         self.input_ids = input_ids
         self.attention_mask = attention_mask
         self.token_type_ids = token_type_ids
@@ -86,34 +91,6 @@ class InputFeatures(object):
 class DataProcessor(object):
     """Base class for data converters for sequence classification data sets."""
 
-    def get_example_from_tensor_dict(self, tensor_dict):
-        """Gets an example from a dict with tensorflow tensors
-
-        Args:
-            tensor_dict: Keys and values should match the corresponding Glue
-                tensorflow_dataset examples.
-        """
-        raise NotImplementedError()
-
-    def get_train_examples(self, data_dir):
-        """Gets a collection of `InputExample`s for the train set."""
-        raise NotImplementedError()
-
-    def get_dev_examples(self, data_dir):
-        """Gets a collection of `InputExample`s for the dev set."""
-        raise NotImplementedError()
-
-    def get_labels(self):
-        """Gets the list of labels for this data set."""
-        raise NotImplementedError()
-
-    def tfds_map(self, example):
-        """Some tensorflow_datasets datasets are not formatted the same way the GLUE datasets are. 
-        This method converts examples to the correct format."""
-        if len(self.get_labels()) > 1:
-            example.label = self.get_labels()[int(example.label)]
-        return example
-
     @classmethod
     def _read_tsv(cls, input_file, quotechar=None):
         """Reads a tab separated value file."""
@@ -129,15 +106,11 @@ class DataProcessor(object):
 
 class SingleSentenceClassificationProcessor(DataProcessor):
     """ Generic processor for a single sentence classification data set."""
-    def __init__(self, labels=None, examples=None):
+    def __init__(self, labels=None, examples=None, mode='classification', verbose=False):
         self.labels = [] if labels is None else labels
         self.examples = [] if examples is None else examples
-
-    @classmethod
-    def create_from_csv(cls, file_name):
-        processor = cls()
-        processor.add_examples_from_csv(file_name)
-        return processor
+        self.mode = mode
+        self.verbose = verbose
 
     def __len__(self):
         return len(self.examples)
@@ -148,30 +121,40 @@ class SingleSentenceClassificationProcessor(DataProcessor):
                                                          examples=self.examples[idx])
         return self.examples[idx]
 
-    def get_labels(self):
-        """Gets the list of labels for this data set."""
-        return self.labels
+    @classmethod
+    def create_from_csv(cls, file_name, **kwargs):
+        processor = cls(**kwargs)
+        processor.add_examples_from_csv(file_name)
+        return processor
 
-    def add_examples_from_csv(self, file_name):
+    def add_examples_from_csv(self, file_name, split_name='', column_label=0, column_text=1, column_id=None,
+                              overwrite_labels=False, overwrite_examples=False):
         lines = self._read_tsv(file_name)
-        self.add_examples_from_lines(lines)
-
-    def add_examples_from_lines(self, lines, split_name='', overwrite_labels=False, overwrite_examples=False):
-        """Creates examples for the training and dev sets."""
-        added_labels = set()
-        examples = []
+        texts = []
+        labels = []
+        ids = []
         for (i, line) in enumerate(lines):
-            if len(line) > 2:
-                guid = "%s-%s" % (split_name, line[0]) if split_name else line[0]
-                label = line[1]
-                text_a = line[2]
+            texts.append(line[column_text])
+            labels.append(line[column_label])
+            if column_id is not None:
+                ids.append(line[column_id])
             else:
                 guid = "%s-%s" % (split_name, i) if split_name else "%s" % i
-                label = line[0]
-                text_a = line[1]
+                ids.append(guid)
 
+        return self.add_examples(texts, labels, ids, overwrite_labels=overwrite_labels, overwrite_examples=overwrite_examples)
+
+    def add_examples(self, texts, labels, ids=None, overwrite_labels=False, overwrite_examples=False):
+        if ids is None:
+            ids = [None] * len(texts)
+        assert len(texts) == len(labels)
+        assert len(texts) == len(ids)
+
+        examples = []
+        added_labels = set()
+        for (text, label, guid) in zip(texts, labels, ids):
             added_labels.add(label)
-            examples.append(InputExample(guid=guid, text_a=text_a, text_b=None, label=label))
+            examples.append(InputExample(guid=guid, text_a=text, text_b=None, label=label))
 
         # Update examples
         if overwrite_examples:
@@ -187,123 +170,129 @@ class SingleSentenceClassificationProcessor(DataProcessor):
 
         return self.examples
 
+    @classmethod
+    def create_from_examples(cls, texts, labels, **kwargs):
+        processor = cls(**kwargs)
+        processor.add_examples(texts, labels)
+        return processor
 
-def convert_examples_to_features(examples, tokenizer,
-                                 mode='sequence_classification',   
-                                 max_length=512,
-                                 pad_on_left=False,
-                                 pad_token=0,
-                                 pad_token_segment_id=0,
-                                 mask_padding_with_zero=True):
-    """
-    Loads a data file into a list of ``InputFeatures``
+    def get_features(self,
+                     tokenizer,
+                     max_length=None,
+                     pad_on_left=False,
+                     pad_token=0,
+                     mask_padding_with_zero=True,
+                     return_tensors=None):
+        """
+        Convert examples in a list of ``InputFeatures``
 
-    Args:
-        examples: List of ``InputExamples`` or ``tf.data.Dataset`` containing the examples.
-        tokenizer: Instance of a tokenizer that will tokenize the examples
-        max_length: Maximum example length
-        task: GLUE task
-        label_list: List of labels. Can be obtained from the processor using the ``processor.get_labels()`` method
-        output_mode: String indicating the output mode. Either ``regression`` or ``classification``
-        pad_on_left: If set to ``True``, the examples will be padded on the left rather than on the right (default)
-        pad_token: Padding token
-        pad_token_segment_id: The segment ID for the padding token (It is usually 0, but can vary such as for XLNet where it is 4)
-        mask_padding_with_zero: If set to ``True``, the attention mask will be filled by ``1`` for actual values
-            and by ``0`` for padded values. If set to ``False``, inverts it (``1`` for padded values, ``0`` for
-            actual values)
+        Args:
+            tokenizer: Instance of a tokenizer that will tokenize the examples
+            max_length: Maximum example length
+            task: GLUE task
+            label_list: List of labels. Can be obtained from the processor using the ``processor.get_labels()`` method
+            output_mode: String indicating the output mode. Either ``regression`` or ``classification``
+            pad_on_left: If set to ``True``, the examples will be padded on the left rather than on the right (default)
+            pad_token: Padding token
+            mask_padding_with_zero: If set to ``True``, the attention mask will be filled by ``1`` for actual values
+                and by ``0`` for padded values. If set to ``False``, inverts it (``1`` for padded values, ``0`` for
+                actual values)
 
-    Returns:
-        If the ``examples`` input is a ``tf.data.Dataset``, will return a ``tf.data.Dataset``
-        containing the task-specific features. If the input is a list of ``InputExamples``, will return
-        a list of task-specific ``InputFeatures`` which can be fed to the model.
+        Returns:
+            If the ``examples`` input is a ``tf.data.Dataset``, will return a ``tf.data.Dataset``
+            containing the task-specific features. If the input is a list of ``InputExamples``, will return
+            a list of task-specific ``InputFeatures`` which can be fed to the model.
 
-    """
-    is_tf_dataset = False
-    if is_tf_available() and isinstance(examples, tf.data.Dataset):
-        is_tf_dataset = True
+        """
 
-    if task is not None:
-        processor = glue_processors[task]()
-        if label_list is None:
-            label_list = processor.get_labels()
-            logger.info("Using label list %s for task %s" % (label_list, task))
-        if output_mode is None:
-            output_mode = glue_output_modes[task]
-            logger.info("Using output mode %s for task %s" % (output_mode, task))
+        label_map = {label: i for i, label in enumerate(self.labels)}
 
-    label_map = {label: i for i, label in enumerate(label_list)}
+        all_input_ids = []
+        for (ex_index, example) in enumerate(self.examples):
+            if ex_index % 10000 == 0:
+                logger.info("Tokenizing example %d", ex_index)
 
-    features = []
-    for (ex_index, example) in enumerate(examples):
-        if ex_index % 10000 == 0:
-            logger.info("Writing example %d" % (ex_index))
-        if is_tf_dataset:
-            example = processor.get_example_from_tensor_dict(example)
+            input_ids = tokenizer.encode(
+                example.text_a,
+                add_special_tokens=True,
+                max_length=min(max_length, tokenizer.max_len),
+            )
+            all_input_ids.append(input_ids)
 
-        inputs = tokenizer.encode_plus(
-            example.text_a,
-            example.text_b,
-            add_special_tokens=True,
-            max_length=max_length,
-        )
-        input_ids, token_type_ids = inputs["input_ids"], inputs["token_type_ids"]
+        batch_length = max(len(input_ids) for input_ids in all_input_ids)
 
-        # The mask has 1 for real tokens and 0 for padding tokens. Only real
-        # tokens are attended to.
-        attention_mask = [1 if mask_padding_with_zero else 0] * len(input_ids)
+        features = []
+        for (ex_index, (input_ids, example)) in enumerate(zip(all_input_ids, examples)):
+            if ex_index % 10000 == 0:
+                logger.info("Writing example %d", ex_index)
+            # The mask has 1 for real tokens and 0 for padding tokens. Only real
+            # tokens are attended to.
+            attention_mask = [1 if mask_padding_with_zero else 0] * len(input_ids)
 
-        # Zero-pad up to the sequence length.
-        padding_length = max_length - len(input_ids)
-        if pad_on_left:
-            input_ids = ([pad_token] * padding_length) + input_ids
-            attention_mask = ([0 if mask_padding_with_zero else 1] * padding_length) + attention_mask
-            token_type_ids = ([pad_token_segment_id] * padding_length) + token_type_ids
+            # Zero-pad up to the sequence length.
+            padding_length = batch_length - len(input_ids)
+            if pad_on_left:
+                input_ids = ([pad_token] * padding_length) + input_ids
+                attention_mask = ([0 if mask_padding_with_zero else 1] * padding_length) + attention_mask
+            else:
+                input_ids = input_ids + ([pad_token] * padding_length)
+                attention_mask = attention_mask + ([0 if mask_padding_with_zero else 1] * padding_length)
+
+            assert len(input_ids) == batch_length, "Error with input length {} vs {}".format(len(input_ids), batch_length)
+            assert len(attention_mask) == batch_length, "Error with input length {} vs {}".format(len(attention_mask), batch_length)
+
+            if self.mode == "classification":
+                label = label_map[example.label]
+            elif self.mode == "regression":
+                label = float(example.label)
+            else:
+                raise ValueError(self.mode)
+
+            if ex_index < 5 and self.verbose:
+                logger.info("*** Example ***")
+                logger.info("guid: %s" % (example.guid))
+                logger.info("input_ids: %s" % " ".join([str(x) for x in input_ids]))
+                logger.info("attention_mask: %s" % " ".join([str(x) for x in attention_mask]))
+                logger.info("label: %s (id = %d)" % (example.label, label))
+
+            features.append(
+                    InputFeatures(input_ids=input_ids,
+                                  attention_mask=attention_mask,
+                                  label=label))
+
+        if return_tensors is None:
+            return features
+        elif return_tensors == 'tf':
+            if not is_tf_available():
+                raise ImportError("return_tensors set to 'tf' but TensorFlow 2.0 can't be imported")
+            import tensorflow as tf
+            def gen():
+                for ex in features:
+                    yield  ({'input_ids': ex.input_ids,
+                            'attention_mask': ex.attention_mask},
+                            ex.label)
+
+            dataset = tf.data.Dataset.from_generator(gen,
+                    ({'input_ids': tf.int32,
+                    'attention_mask': tf.int32},
+                    tf.int64),
+                    ({'input_ids': tf.TensorShape([None]),
+                    'attention_mask': tf.TensorShape([None])},
+                    tf.TensorShape([])))
+            return dataset
+        elif return_tensors == 'pt':
+            if not is_torch_available():
+                raise ImportError("return_tensors set to 'pt' but PyTorch can't be imported")
+            import torch
+            from torch.utils.data import TensorDataset
+            all_input_ids = torch.tensor([f.input_ids for f in features], dtype=torch.long)
+            all_attention_mask = torch.tensor([f.attention_mask for f in features], dtype=torch.long)
+            if self.mode == "classification":
+                all_labels = torch.tensor([f.label for f in features], dtype=torch.long)
+            elif self.mode == "regression":
+                all_labels = torch.tensor([f.label for f in features], dtype=torch.float)
+
+            dataset = TensorDataset(all_input_ids, all_attention_mask, all_labels)
+            return dataset
         else:
-            input_ids = input_ids + ([pad_token] * padding_length)
-            attention_mask = attention_mask + ([0 if mask_padding_with_zero else 1] * padding_length)
-            token_type_ids = token_type_ids + ([pad_token_segment_id] * padding_length)
-
-        assert len(input_ids) == max_length, "Error with input length {} vs {}".format(len(input_ids), max_length)
-        assert len(attention_mask) == max_length, "Error with input length {} vs {}".format(len(attention_mask), max_length)
-        assert len(token_type_ids) == max_length, "Error with input length {} vs {}".format(len(token_type_ids), max_length)
-
-        if output_mode == "classification":
-            label = label_map[example.label]
-        elif output_mode == "regression":
-            label = float(example.label)
-        else:
-            raise KeyError(output_mode)
-
-        if ex_index < 5:
-            logger.info("*** Example ***")
-            logger.info("guid: %s" % (example.guid))
-            logger.info("input_ids: %s" % " ".join([str(x) for x in input_ids]))
-            logger.info("attention_mask: %s" % " ".join([str(x) for x in attention_mask]))
-            logger.info("token_type_ids: %s" % " ".join([str(x) for x in token_type_ids]))
-            logger.info("label: %s (id = %d)" % (example.label, label))
-
-        features.append(
-                InputFeatures(input_ids=input_ids,
-                              attention_mask=attention_mask,
-                              token_type_ids=token_type_ids,
-                              label=label))
-
-    if is_tf_available() and is_tf_dataset:
-        def gen():
-            for ex in features:
-                yield  ({'input_ids': ex.input_ids,
-                         'attention_mask': ex.attention_mask,
-                         'token_type_ids': ex.token_type_ids},
-                        ex.label)
-
-        return tf.data.Dataset.from_generator(gen,
-            ({'input_ids': tf.int32,
-              'attention_mask': tf.int32,
-              'token_type_ids': tf.int32},
-             tf.int64),
-            ({'input_ids': tf.TensorShape([None]),
-              'attention_mask': tf.TensorShape([None]),
-              'token_type_ids': tf.TensorShape([None])},
-             tf.TensorShape([])))
-
-    return features
+            raise ValueError("return_tensors should be one of 'tf' or 'pt'")
diff --git a/transformers/pipeline.py b/transformers/pipeline.py
new file mode 100644
index 0000000000..15adc620b1
--- /dev/null
+++ b/transformers/pipeline.py
@@ -0,0 +1,254 @@
+# coding=utf-8
+# Copyright 2018 The HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" Pipeline class: Tokenizer + Model. """
+
+from __future__ import absolute_import, division, print_function, unicode_literals
+import os
+import logging
+
+from .modeling_auto import (AutoModel, AutoModelForQuestionAnswering,
+                            AutoModelForSequenceClassification,
+                            AutoModelWithLMHead)
+from .tokenization_auto import AutoTokenizer
+from .file_utils import add_start_docstrings, is_tf_available, is_torch_available
+from .data.processors import SingleSentenceClassificationProcessor
+
+if is_tf_available():
+    import tensorflow as tf
+if is_torch_available():
+    import torch
+
+logger = logging.getLogger(__name__)
+
+# TF training parameters
+USE_XLA = False
+USE_AMP = False
+
+class TextClassificationPipeline(object):
+    r"""
+        :class:`~transformers.TextClassificationPipeline` is a class encapsulating a pretrained model and
+        its tokenizer and will be instantiated as one of the base model classes of the library
+        when created with the `Pipeline.from_pretrained(pretrained_model_name_or_path)`
+        class method.
+
+        The `from_pretrained()` method takes care of returning the correct model class instance
+        using pattern matching on the `pretrained_model_name_or_path` string.
+
+        The base model class to instantiate is selected as the first pattern matching
+        in the `pretrained_model_name_or_path` string (in the following order):
+            - contains `distilbert`: DistilBertModel (DistilBERT model)
+            - contains `roberta`: RobertaModel (RoBERTa model)
+            - contains `bert`: BertModel (Bert model)
+            - contains `openai-gpt`: OpenAIGPTModel (OpenAI GPT model)
+            - contains `gpt2`: GPT2Model (OpenAI GPT-2 model)
+            - contains `ctrl`: CTRLModel (Salesforce CTRL  model)
+            - contains `transfo-xl`: TransfoXLModel (Transformer-XL model)
+            - contains `xlnet`: XLNetModel (XLNet model)
+            - contains `xlm`: XLMModel (XLM model)
+    """
+    def __init__(self, tokenizer, model):
+        self.tokenizer = tokenizer
+        self.model = model
+        if is_tf_available():
+            self.framework = 'tf'
+        elif is_torch_available():
+            self.framework = 'pt'
+        else:
+            raise ImportError("At least one of PyTorch or TensorFlow 2.0+ should be installed to use CLI training")
+        self.is_compiled = False
+
+
+    @classmethod
+    def from_pretrained(cls, pretrained_model_name_or_path, **kwargs):
+        r""" Instantiates one of the base model classes of the library
+        from a pre-trained model configuration.
+
+        The model class to instantiate is selected as the first pattern matching
+        in the `pretrained_model_name_or_path` string (in the following order):
+            - contains `distilbert`: DistilBertModel (DistilBERT model)
+            - contains `roberta`: RobertaModel (RoBERTa model)
+            - contains `bert`: BertModel (Bert model)
+            - contains `openai-gpt`: OpenAIGPTModel (OpenAI GPT model)
+            - contains `gpt2`: GPT2Model (OpenAI GPT-2 model)
+            - contains `ctrl`: CTRLModel (Salesforce CTRL  model)
+            - contains `transfo-xl`: TransfoXLModel (Transformer-XL model)
+            - contains `xlnet`: XLNetModel (XLNet model)
+            - contains `xlm`: XLMModel (XLM model)
+
+            The model is set in evaluation mode by default using `model.eval()` (Dropout modules are deactivated)
+            To train the model, you should first set it back in training mode with `model.train()`
+
+        Params:
+            pretrained_model_name_or_path: either:
+
+                - a string with the `shortcut name` of a pre-trained model to load from cache or download, e.g.: ``bert-base-uncased``.
+                - a path to a `directory` containing model weights saved using :func:`~transformers.PreTrainedModel.save_pretrained`, e.g.: ``./my_model_directory/``.
+                - a path or url to a `tensorflow index checkpoint file` (e.g. `./tf_model/model.ckpt.index`). In this case, ``from_tf`` should be set to True and a configuration object should be provided as ``config`` argument. This loading path is slower than converting the TensorFlow checkpoint in a PyTorch model using the provided conversion scripts and loading the PyTorch model afterwards.
+
+            config: (`optional`) instance of a class derived from :class:`~transformers.PretrainedConfig`:
+                Configuration for the model to use instead of an automatically loaded configuation. Configuration can be automatically loaded when:
+
+                - the model is a model provided by the library (loaded with the ``shortcut-name`` string of a pretrained model), or
+                - the model was saved using :func:`~transformers.PreTrainedModel.save_pretrained` and is reloaded by suppling the save directory.
+                - the model is loaded by suppling a local directory as ``pretrained_model_name_or_path`` and a configuration JSON file named `config.json` is found in the directory.
+
+            state_dict: (`optional`) dict:
+                an optional state dictionnary for the model to use instead of a state dictionary loaded from saved weights file.
+                This option can be used if you want to create a model from a pretrained configuration but load your own weights.
+                In this case though, you should check if using :func:`~transformers.PreTrainedModel.save_pretrained` and :func:`~transformers.PreTrainedModel.from_pretrained` is not a simpler option.
+
+            cache_dir: (`optional`) string:
+                Path to a directory in which a downloaded pre-trained model
+                configuration should be cached if the standard cache should not be used.
+
+            force_download: (`optional`) boolean, default False:
+                Force to (re-)download the model weights and configuration files and override the cached versions if they exists.
+
+            proxies: (`optional`) dict, default None:
+                A dictionary of proxy servers to use by protocol or endpoint, e.g.: {'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.
+                The proxies are used on each request.
+
+            output_loading_info: (`optional`) boolean:
+                Set to ``True`` to also return a dictionnary containing missing keys, unexpected keys and error messages.
+
+            kwargs: (`optional`) Remaining dictionary of keyword arguments:
+                Can be used to update the configuration object (after it being loaded) and initiate the model. (e.g. ``output_attention=True``). Behave differently depending on whether a `config` is provided or automatically loaded:
+
+                - If a configuration is provided with ``config``, ``**kwargs`` will be directly passed to the underlying model's ``__init__`` method (we assume all relevant updates to the configuration have already been done)
+                - If a configuration is not provided, ``kwargs`` will be first passed to the configuration class initialization function (:func:`~transformers.PretrainedConfig.from_pretrained`). Each key of ``kwargs`` that corresponds to a configuration attribute will be used to override said attribute with the supplied ``kwargs`` value. Remaining keys that do not correspond to any configuration attribute will be passed to the underlying model's ``__init__`` function.
+
+        Examples::
+
+            model = AutoModel.from_pretrained('bert-base-uncased')    # Download model and configuration from S3 and cache.
+            model = AutoModel.from_pretrained('./test/bert_model/')  # E.g. model was saved using `save_pretrained('./test/saved_model/')`
+            model = AutoModel.from_pretrained('bert-base-uncased', output_attention=True)  # Update configuration during loading
+            assert model.config.output_attention == True
+            # Loading from a TF checkpoint file instead of a PyTorch model (slower)
+            config = AutoConfig.from_json_file('./tf_model/bert_tf_model_config.json')
+            model = AutoModel.from_pretrained('./tf_model/bert_tf_checkpoint.ckpt.index', from_tf=True, config=config)
+
+        """
+        # Extract tokenizer and model arguments
+        tokenizer_kwargs = {}
+        for key in kwargs:
+            if key.startswith('tokenizer_'):
+                # Specific to the tokenizer
+                tokenizer_kwargs[key.replace('tokenizer_', '')] = kwargs.pop(key)
+            elif not key.startswith('model_'):
+                # used for both the tokenizer and the model
+                tokenizer_kwargs[key] = kwargs[key]
+
+        model_kwargs = {}
+        for key in kwargs:
+            if key.startswith('model_'):
+                # Specific to the model
+                model_kwargs[key.replace('model_', '')] = kwargs.pop(key)
+            elif not key.startswith('tokenizer_'):
+                # used for both the tokenizer and the model
+                model_kwargs[key] = kwargs[key]
+
+        tokenizer = AutoTokenizer.from_pretrained(pretrained_model_name_or_path, **tokenizer_kwargs)
+        model = AutoModelForSequenceClassification.from_pretrained(pretrained_model_name_or_path, **model_kwargs)
+        return cls(tokenizer, model)
+
+
+    def save_pretrained(self, save_directory):
+        if not os.path.isdir(save_directory):
+            logger.error("Saving directory ({}) should be a directory".format(save_directory))
+            return
+        self.model.save_pretrained(save_directory)
+        self.tokenizer.save_pretrained(save_directory)
+
+
+    def compile(self, learning_rate=3e-5, epsilon=1e-8):
+        if self.framework == 'tf':
+            logger.info('Preparing model')
+            # Prepare training: Compile tf.keras model with optimizer, loss and learning rate schedule
+            opt = tf.keras.optimizers.Adam(learning_rate=learning_rate, epsilon=epsilon)
+            if USE_AMP:
+                # loss scaling is currently required when using mixed precision
+                opt = tf.keras.mixed_precision.experimental.LossScaleOptimizer(opt, 'dynamic')
+            loss = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True)
+            metric = tf.keras.metrics.SparseCategoricalAccuracy('accuracy')
+            self.model.compile(optimizer=opt, loss=loss, metrics=[metric])
+        else:
+            raise NotImplementedError
+        self.is_compiled = True
+
+
+    def prepare_data(self, train_samples_text, train_samples_labels,
+                     valid_samples_text=None, valid_samples_labels=None,
+                     validation_split=0.1):
+        dataset = SingleSentenceClassificationProcessor.create_from_examples(train_samples_text,
+                                                                             train_samples_labels)
+        num_data_samples = len(dataset)
+        if valid_samples_text is not None and valid_samples_labels is not None:
+            valid_dataset = SingleSentenceClassificationProcessor.create_from_examples(valid_samples_text,
+                                                                                       valid_samples_labels)
+            num_valid_samples = len(valid_dataset)
+            train_dataset = dataset
+            num_train_samples = num_data_samples
+        else:
+            assert 0.0 < validation_split < 1.0, "validation_split should be between 0.0 and 1.0"
+            num_valid_samples = int(num_data_samples * validation_split)
+            num_train_samples = num_data_samples - num_valid_samples
+            train_dataset = dataset[num_train_samples]
+            valid_dataset = dataset[num_valid_samples]
+
+        logger.info('Tokenizing and processing dataset')
+        train_dataset = train_dataset.get_features(self.tokenizer, return_tensors=self.framework)
+        valid_dataset = valid_dataset.get_features(self.tokenizer, return_tensors=self.framework)
+        return train_dataset, valid_dataset, num_train_samples, num_valid_samples
+
+
+    def fit(self, train_samples_text, train_samples_labels,
+            valid_samples_text=None, valid_samples_labels=None,
+            train_batch_size=None, valid_batch_size=None,
+            validation_split=0.1,
+            **kwargs):
+
+        if not self.is_compiled:
+            self.compile()
+
+        datasets = self.prepare_data(train_samples_text, train_samples_labels,
+                                     valid_samples_text, valid_samples_labels,
+                                     validation_split)
+        train_dataset, valid_dataset, num_train_samples, num_valid_samples = datasets
+
+        train_steps = num_train_samples//train_batch_size
+        valid_steps = num_valid_samples//valid_batch_size
+
+        if self.framework == 'tf':
+            # Prepare dataset as a tf.train_data.Dataset instance
+            train_dataset = train_dataset.shuffle(128).batch(train_batch_size).repeat(-1)
+            valid_dataset = valid_dataset.batch(valid_batch_size)
+
+            logger.info('Training TF 2.0 model')
+            history = self.model.fit(train_dataset, epochs=2, steps_per_epoch=train_steps,
+                                     validation_data=valid_dataset, validation_steps=valid_steps, **kwargs)
+        else:
+            raise NotImplementedError
+
+
+    def __call__(self, text):
+        inputs = self.tokenizer.encode_plus(text, add_special_tokens=True, return_tensors=self.framework)
+        if self.framework == 'tf':
+            # TODO trace model
+            predictions = self.model(**inputs)[0]
+        else:
+            with torch.no_grad():
+                predictions = self.model(**inputs)[0]
+
+        return predictions.numpy().tolist()

From b81ab431f26e4a2fadf37bdd803ec26c66ee719c Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Thu, 17 Oct 2019 12:06:27 +0200
Subject: [PATCH 287/505] updating AutoModels and AutoConfiguration - adding
 pipelines

---
 transformers/configuration_auto.py |  28 +++++
 transformers/modeling_auto.py      | 182 +++++++++++++++++++++++++++--
 transformers/modeling_tf_auto.py   | 169 +++++++++++++++++++++++++--
 transformers/pipeline.py           |  83 ++-----------
 4 files changed, 372 insertions(+), 90 deletions(-)

diff --git a/transformers/configuration_auto.py b/transformers/configuration_auto.py
index 43f251bd0c..47379d2f5a 100644
--- a/transformers/configuration_auto.py
+++ b/transformers/configuration_auto.py
@@ -61,6 +61,34 @@ class AutoConfig(object):
         raise EnvironmentError("AutoConfig is designed to be instantiated "
             "using the `AutoConfig.from_pretrained(pretrained_model_name_or_path)` method.")
 
+    @classmethod
+    def for_model(cls, model_type, *args, **kwargs):
+        if 'distilbert' in model_type:
+            return DistilBertConfig(*args, **kwargs)
+        elif 'roberta' in model_type:
+            return RobertaConfig(*args, **kwargs)
+        elif 'bert' in model_type:
+            return BertConfig(*args, **kwargs)
+        elif 'openai-gpt' in model_type:
+            return OpenAIGPTConfig(*args, **kwargs)
+        elif 'gpt2' in model_type:
+            return GPT2Config(*args, **kwargs)
+        elif 'transfo-xl' in model_type:
+            return TransfoXLConfig(*args, **kwargs)
+        elif 'xlnet' in model_type:
+            return XLNetConfig(*args, **kwargs)
+        elif 'xlm' in model_type:
+            return XLMConfig(*args, **kwargs)
+        elif 'ctrl' in model_type:
+            return CTRLConfig(*args, **kwargs)
+        elif 'albert' in model_type:
+            return AlbertConfig(*args, **kwargs)
+        elif 'camembert' in model_type:
+            return CamembertConfig(*args, **kwargs)
+        raise ValueError("Unrecognized model identifier in {}. Should contains one of "
+                         "'distilbert', 'bert', 'openai-gpt', 'gpt2', 'transfo-xl', 'xlnet', "
+                         "'xlm', 'roberta', 'ctrl', 'camembert', 'albert'".format(model_type))
+
     @classmethod
     def from_pretrained(cls, pretrained_model_name_or_path, **kwargs):
         r""" Instantiate a one of the configuration classes of the library
diff --git a/transformers/modeling_auto.py b/transformers/modeling_auto.py
index b63e43d73b..0c8bffa883 100644
--- a/transformers/modeling_auto.py
+++ b/transformers/modeling_auto.py
@@ -18,6 +18,10 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 
 import logging
 
+from .configuration_auto import (AlbertConfig, BertConfig, CamembertConfig, CTRLConfig,
+                                 DistilBertConfig, GPT2Config, OpenAIGPTConfig, RobertaConfig,
+                                 TransfoXLConfig, XLMConfig, XLNetConfig)
+
 from .modeling_bert import BertModel, BertForMaskedLM, BertForSequenceClassification, BertForQuestionAnswering
 from .modeling_openai import OpenAIGPTModel, OpenAIGPTLMHeadModel
 from .modeling_gpt2 import GPT2Model, GPT2LMHeadModel
@@ -27,8 +31,7 @@ from .modeling_xlnet import XLNetModel, XLNetLMHeadModel, XLNetForSequenceClassi
 from .modeling_xlm import XLMModel, XLMWithLMHeadModel, XLMForSequenceClassification, XLMForQuestionAnswering
 from .modeling_roberta import RobertaModel, RobertaForMaskedLM, RobertaForSequenceClassification
 from .modeling_distilbert import DistilBertModel, DistilBertForQuestionAnswering, DistilBertForMaskedLM, DistilBertForSequenceClassification
-from .modeling_camembert import CamembertModel, CamembertForMaskedLM, CamembertForSequenceClassification, CamembertForMultipleChoice
-from .modeling_camembert import CamembertModel, CamembertForMaskedLM, CamembertForSequenceClassification, CamembertForMultipleChoice
+from .modeling_camembert import CamembertModel, CamembertForQuestionAnswering, CamembertForMaskedLM, CamembertForSequenceClassification, CamembertForMultipleChoice
 from .modeling_albert import AlbertModel, AlbertForMaskedLM, AlbertForSequenceClassification, AlbertForQuestionAnswering
 
 from .modeling_utils import PreTrainedModel, SequenceSummary
@@ -43,7 +46,7 @@ class AutoModel(object):
         :class:`~transformers.AutoModel` is a generic model class
         that will be instantiated as one of the base model classes of the library
         when created with the `AutoModel.from_pretrained(pretrained_model_name_or_path)`
-        class method.
+        or the `AutoModel.from_config(config)` class methods.
 
         The `from_pretrained()` method takes care of returning the correct model class instance
         using pattern matching on the `pretrained_model_name_or_path` string.
@@ -66,7 +69,54 @@ class AutoModel(object):
     """
     def __init__(self):
         raise EnvironmentError("AutoModel is designed to be instantiated "
-            "using the `AutoModel.from_pretrained(pretrained_model_name_or_path)` method.")
+            "using the `AutoModel.from_pretrained(pretrained_model_name_or_path)` or "
+            "`AutoModel.from_config(config)` methods.")
+
+    @classmethod
+    def from_config(cls, config):
+        r""" Instantiates one of the base model classes of the library
+        from a configuration.
+
+            config: (`optional`) instance of a class derived from :class:`~transformers.PretrainedConfig`:
+                The model class to instantiate is selected based on the configuration class:
+                    - isInstance of `distilbert` configuration class: DistilBertModel (DistilBERT model)
+                    - isInstance of `roberta` configuration class: RobertaModel (RoBERTa model)
+                    - isInstance of `bert` configuration class: BertModel (Bert model)
+                    - isInstance of `openai-gpt` configuration class: OpenAIGPTModel (OpenAI GPT model)
+                    - isInstance of `gpt2` configuration class: GPT2Model (OpenAI GPT-2 model)
+                    - isInstance of `ctrl` configuration class: CTRLModel (Salesforce CTRL  model)
+                    - isInstance of `transfo-xl` configuration class: TransfoXLModel (Transformer-XL model)
+                    - isInstance of `xlnet` configuration class: XLNetModel (XLNet model)
+                    - isInstance of `xlm` configuration class: XLMModel (XLM model)
+
+        Examples::
+
+            config = BertConfig.from_pretrained('bert-base-uncased')    # Download configuration from S3 and cache.
+            model = AutoModel.from_config(config)  # E.g. model was saved using `save_pretrained('./test/saved_model/')`
+        """
+        if isinstance(config, DistilBertConfig):
+            return DistilBertModel(config)
+        elif isinstance(config, RobertaConfig):
+            return RobertaModel(config)
+        elif isinstance(config, BertConfig):
+            return BertModel(config)
+        elif isinstance(config, OpenAIGPTConfig):
+            return OpenAIGPTModel(config)
+        elif isinstance(config, GPT2Config):
+            return GPT2Model(config)
+        elif isinstance(config, TransfoXLConfig):
+            return TransfoXLModel(config)
+        elif isinstance(config, XLNetConfig):
+            return XLNetModel(config)
+        elif isinstance(config, XLMConfig):
+            return XLMModel(config)
+        elif isinstance(config, CTRLConfig):
+            return CTRLModel(config)
+        elif isinstance(config, AlbertConfig):
+            return AlbertModel(config)
+        elif isinstance(config, CamembertConfig):
+            return CamembertModel(config)
+        raise ValueError("Unrecognized configuration class {}".format(config))
 
     @classmethod
     def from_pretrained(cls, pretrained_model_name_or_path, *model_args, **kwargs):
@@ -201,7 +251,54 @@ class AutoModelWithLMHead(object):
     """
     def __init__(self):
         raise EnvironmentError("AutoModelWithLMHead is designed to be instantiated "
-            "using the `AutoModelWithLMHead.from_pretrained(pretrained_model_name_or_path)` method.")
+            "using the `AutoModelWithLMHead.from_pretrained(pretrained_model_name_or_path)` or "
+            "`AutoModelWithLMHead.from_config(config)` methods.")
+
+    @classmethod
+    def from_config(cls, config):
+        r""" Instantiates one of the base model classes of the library
+        from a configuration.
+
+            config: (`optional`) instance of a class derived from :class:`~transformers.PretrainedConfig`:
+                The model class to instantiate is selected based on the configuration class:
+                    - isInstance of `distilbert` configuration class: DistilBertModel (DistilBERT model)
+                    - isInstance of `roberta` configuration class: RobertaModel (RoBERTa model)
+                    - isInstance of `bert` configuration class: BertModel (Bert model)
+                    - isInstance of `openai-gpt` configuration class: OpenAIGPTModel (OpenAI GPT model)
+                    - isInstance of `gpt2` configuration class: GPT2Model (OpenAI GPT-2 model)
+                    - isInstance of `ctrl` configuration class: CTRLModel (Salesforce CTRL  model)
+                    - isInstance of `transfo-xl` configuration class: TransfoXLModel (Transformer-XL model)
+                    - isInstance of `xlnet` configuration class: XLNetModel (XLNet model)
+                    - isInstance of `xlm` configuration class: XLMModel (XLM model)
+
+        Examples::
+
+            config = BertConfig.from_pretrained('bert-base-uncased')    # Download configuration from S3 and cache.
+            model = AutoModelWithLMHead.from_config(config)  # E.g. model was saved using `save_pretrained('./test/saved_model/')`
+        """
+        if isinstance(config, DistilBertConfig):
+            return DistilBertForMaskedLM(config)
+        elif isinstance(config, RobertaConfig):
+            return RobertaForMaskedLM(config)
+        elif isinstance(config, BertConfig):
+            return BertForMaskedLM(config)
+        elif isinstance(config, OpenAIGPTConfig):
+            return OpenAIGPTLMHeadModel(config)
+        elif isinstance(config, GPT2Config):
+            return GPT2LMHeadModel(config)
+        elif isinstance(config, TransfoXLConfig):
+            return TransfoXLLMHeadModel(config)
+        elif isinstance(config, XLNetConfig):
+            return XLNetLMHeadModel(config)
+        elif isinstance(config, XLMConfig):
+            return XLMWithLMHeadModel(config)
+        elif isinstance(config, CTRLConfig):
+            return CTRLLMHeadModel(config)
+        elif isinstance(config, AlbertConfig):
+            return AlbertLMHeadModel(config)
+        elif isinstance(config, CamembertConfig):
+            return CamembertLMHeadModel(config)
+        raise ValueError("Unrecognized configuration class {}".format(config))
 
     @classmethod
     def from_pretrained(cls, pretrained_model_name_or_path, *model_args, **kwargs):
@@ -333,8 +430,43 @@ class AutoModelForSequenceClassification(object):
         This class cannot be instantiated using `__init__()` (throws an error).
     """
     def __init__(self):
-        raise EnvironmentError("AutoModelWithLMHead is designed to be instantiated "
-            "using the `AutoModelWithLMHead.from_pretrained(pretrained_model_name_or_path)` method.")
+        raise EnvironmentError("AutoModelForSequenceClassification is designed to be instantiated "
+            "using the `AutoModelForSequenceClassification.from_pretrained(pretrained_model_name_or_path)` or "
+            "`AutoModelForSequenceClassification.from_config(config)` methods.")
+
+    @classmethod
+    def from_config(cls, config):
+        r""" Instantiates one of the base model classes of the library
+        from a configuration.
+
+            config: (`optional`) instance of a class derived from :class:`~transformers.PretrainedConfig`:
+                The model class to instantiate is selected based on the configuration class:
+                    - isInstance of `distilbert` configuration class: DistilBertModel (DistilBERT model)
+                    - isInstance of `roberta` configuration class: RobertaModel (RoBERTa model)
+                    - isInstance of `bert` configuration class: BertModel (Bert model)
+                    - isInstance of `xlnet` configuration class: XLNetModel (XLNet model)
+                    - isInstance of `xlm` configuration class: XLMModel (XLM model)
+
+        Examples::
+
+            config = BertConfig.from_pretrained('bert-base-uncased')    # Download configuration from S3 and cache.
+            model = AutoModelForSequenceClassification.from_config(config)  # E.g. model was saved using `save_pretrained('./test/saved_model/')`
+        """
+        if isinstance(config, AlbertConfig):
+            return AlbertForSequenceClassification(config)
+        elif isintance(config, CamembertConfig):
+            return CamembertForSequenceClassification(config)
+        elif isinstance(config, DistilBertConfig):
+            return DistilBertForSequenceClassification(config)
+        elif isinstance(config, RobertaConfig):
+            return RobertaForSequenceClassification(config)
+        elif isinstance(config, BertConfig):
+            return BertForSequenceClassification(config)
+        elif isinstance(config, XLNetConfig):
+            return XLNetForSequenceClassification(config)
+        elif isinstance(config, XLMConfig):
+            return XLMForSequenceClassification(config)
+        raise ValueError("Unrecognized configuration class {}".format(config))
 
     @classmethod
     def from_pretrained(cls, pretrained_model_name_or_path, *model_args, **kwargs):
@@ -453,8 +585,40 @@ class AutoModelForQuestionAnswering(object):
         This class cannot be instantiated using `__init__()` (throws an error).
     """
     def __init__(self):
-        raise EnvironmentError("AutoModelWithLMHead is designed to be instantiated "
-            "using the `AutoModelWithLMHead.from_pretrained(pretrained_model_name_or_path)` method.")
+        raise EnvironmentError("AutoModelForQuestionAnswering is designed to be instantiated "
+            "using the `AutoModelForQuestionAnswering.from_pretrained(pretrained_model_name_or_path)` or "
+            "`AutoModelForQuestionAnswering.from_config(config)` methods.")
+
+    @classmethod
+    def from_config(cls, config):
+        r""" Instantiates one of the base model classes of the library
+        from a configuration.
+
+            config: (`optional`) instance of a class derived from :class:`~transformers.PretrainedConfig`:
+                The model class to instantiate is selected based on the configuration class:
+                    - isInstance of `distilbert` configuration class: DistilBertModel (DistilBERT model)
+                    - isInstance of `bert` configuration class: BertModel (Bert model)
+                    - isInstance of `xlnet` configuration class: XLNetModel (XLNet model)
+                    - isInstance of `xlm` configuration class: XLMModel (XLM model)
+
+        Examples::
+
+            config = BertConfig.from_pretrained('bert-base-uncased')    # Download configuration from S3 and cache.
+            model = AutoModelForSequenceClassification.from_config(config)  # E.g. model was saved using `save_pretrained('./test/saved_model/')`
+        """
+        if isintance(config, AlbertConfig):
+            return AlbertForQuestionAnswering(config)
+        elif isintance(config, CamembertConfig):
+            return CamembertForQuestionAnswering(config)
+        elif isinstance(config, DistilBertConfig):
+            return DistilBertForQuestionAnswering(config)
+        elif isinstance(config, BertConfig):
+            return BertForQuestionAnswering(config)
+        elif isinstance(config, XLNetConfig):
+            return XLNetForQuestionAnswering(config)
+        elif isinstance(config, XLMConfig):
+            return XLMForQuestionAnswering(config)
+        raise ValueError("Unrecognized configuration class {}".format(config))
 
     @classmethod
     def from_pretrained(cls, pretrained_model_name_or_path, *model_args, **kwargs):
diff --git a/transformers/modeling_tf_auto.py b/transformers/modeling_tf_auto.py
index cfe19ead2a..e78b91cfcc 100644
--- a/transformers/modeling_tf_auto.py
+++ b/transformers/modeling_tf_auto.py
@@ -18,6 +18,10 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 
 import logging
 
+from .configuration_auto import (BertConfig, CTRLConfig, DistilBertConfig,
+                                 GPT2Config, OpenAIGPTConfig, RobertaConfig,
+                                 TransfoXLConfig, XLMConfig, XLNetConfig)
+
 from .modeling_tf_bert import TFBertModel, TFBertForMaskedLM, TFBertForSequenceClassification, TFBertForQuestionAnswering
 from .modeling_tf_openai import TFOpenAIGPTModel, TFOpenAIGPTLMHeadModel
 from .modeling_tf_gpt2 import TFGPT2Model, TFGPT2LMHeadModel
@@ -59,7 +63,50 @@ class TFAutoModel(object):
     """
     def __init__(self):
         raise EnvironmentError("TFAutoModel is designed to be instantiated "
-            "using the `TFAutoModel.from_pretrained(pretrained_model_name_or_path)` method.")
+            "using the `TFAutoModel.from_pretrained(pretrained_model_name_or_path)` or "
+            "`TFAutoModel.from_config(config)` methods.")
+
+    @classmethod
+    def from_config(cls, config):
+        r""" Instantiates one of the base model classes of the library
+        from a configuration.
+
+            config: (`optional`) instance of a class derived from :class:`~transformers.PretrainedConfig`:
+                The model class to instantiate is selected based on the configuration class:
+                    - isInstance of `distilbert` configuration class: TFDistilBertModel (DistilBERT model)
+                    - isInstance of `roberta` configuration class: TFRobertaModel (RoBERTa model)
+                    - isInstance of `bert` configuration class: TFBertModel (Bert model)
+                    - isInstance of `openai-gpt` configuration class: TFOpenAIGPTModel (OpenAI GPT model)
+                    - isInstance of `gpt2` configuration class: TFGPT2Model (OpenAI GPT-2 model)
+                    - isInstance of `ctrl` configuration class: TFCTRLModel (Salesforce CTRL  model)
+                    - isInstance of `transfo-xl` configuration class: TFTransfoXLModel (Transformer-XL model)
+                    - isInstance of `xlnet` configuration class: TFXLNetModel (XLNet model)
+                    - isInstance of `xlm` configuration class: TFXLMModel (XLM model)
+
+        Examples::
+
+            config = BertConfig.from_pretrained('bert-base-uncased')    # Download configuration from S3 and cache.
+            model = TFAutoModel.from_config(config)  # E.g. model was saved using `save_pretrained('./test/saved_model/')`
+        """
+        if isinstance(config, DistilBertConfig):
+            return TFDistilBertModel(config)
+        elif isinstance(config, RobertaConfig):
+            return TFRobertaModel(config)
+        elif isinstance(config, BertConfig):
+            return TFBertModel(config)
+        elif isinstance(config, OpenAIGPTConfig):
+            return TFOpenAIGPTModel(config)
+        elif isinstance(config, GPT2Config):
+            return TFGPT2Model(config)
+        elif isinstance(config, TransfoXLConfig):
+            return TFTransfoXLModel(config)
+        elif isinstance(config, XLNetConfig):
+            return TFXLNetModel(config)
+        elif isinstance(config, XLMConfig):
+            return TFXLMModel(config)
+        elif isinstance(config, CTRLConfig):
+            return TFCTRLModel(config)
+        raise ValueError("Unrecognized configuration class {}".format(config))
 
     @classmethod
     def from_pretrained(cls, pretrained_model_name_or_path, *model_args, **kwargs):
@@ -156,7 +203,7 @@ class TFAutoModel(object):
             return TFCTRLModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
 
         raise ValueError("Unrecognized model identifier in {}. Should contains one of "
-                         "'bert', 'openai-gpt', 'gpt2', 'transfo-xl', 'xlnet', "
+                         "'distilbert', 'bert', 'openai-gpt', 'gpt2', 'transfo-xl', 'xlnet', "
                          "'xlm', 'roberta', 'ctrl'".format(pretrained_model_name_or_path))
 
 
@@ -186,7 +233,50 @@ class TFAutoModelWithLMHead(object):
     """
     def __init__(self):
         raise EnvironmentError("TFAutoModelWithLMHead is designed to be instantiated "
-            "using the `TFAutoModelWithLMHead.from_pretrained(pretrained_model_name_or_path)` method.")
+            "using the `TFAutoModelWithLMHead.from_pretrained(pretrained_model_name_or_path)` or "
+            "`TFAutoModelWithLMHead.from_config(config)` methods.")
+
+    @classmethod
+    def from_config(cls, config):
+        r""" Instantiates one of the base model classes of the library
+        from a configuration.
+
+            config: (`optional`) instance of a class derived from :class:`~transformers.PretrainedConfig`:
+                The model class to instantiate is selected based on the configuration class:
+                    - isInstance of `distilbert` configuration class: DistilBertModel (DistilBERT model)
+                    - isInstance of `roberta` configuration class: RobertaModel (RoBERTa model)
+                    - isInstance of `bert` configuration class: BertModel (Bert model)
+                    - isInstance of `openai-gpt` configuration class: OpenAIGPTModel (OpenAI GPT model)
+                    - isInstance of `gpt2` configuration class: GPT2Model (OpenAI GPT-2 model)
+                    - isInstance of `ctrl` configuration class: CTRLModel (Salesforce CTRL  model)
+                    - isInstance of `transfo-xl` configuration class: TransfoXLModel (Transformer-XL model)
+                    - isInstance of `xlnet` configuration class: XLNetModel (XLNet model)
+                    - isInstance of `xlm` configuration class: XLMModel (XLM model)
+
+        Examples::
+
+            config = BertConfig.from_pretrained('bert-base-uncased')    # Download configuration from S3 and cache.
+            model = AutoModelWithLMHead.from_config(config)  # E.g. model was saved using `save_pretrained('./test/saved_model/')`
+        """
+        if isinstance(config, DistilBertConfig):
+            return TFDistilBertForMaskedLM(config)
+        elif isinstance(config, RobertaConfig):
+            return TFRobertaForMaskedLM(config)
+        elif isinstance(config, BertConfig):
+            return TFBertForMaskedLM(config)
+        elif isinstance(config, OpenAIGPTConfig):
+            return TFOpenAIGPTLMHeadModel(config)
+        elif isinstance(config, GPT2Config):
+            return TFGPT2LMHeadModel(config)
+        elif isinstance(config, TransfoXLConfig):
+            return TFTransfoXLLMHeadModel(config)
+        elif isinstance(config, XLNetConfig):
+            return TFXLNetLMHeadModel(config)
+        elif isinstance(config, XLMConfig):
+            return TFXLMWithLMHeadModel(config)
+        elif isinstance(config, CTRLConfig):
+            return TFCTRLLMHeadModel(config)
+        raise ValueError("Unrecognized configuration class {}".format(config))
 
     @classmethod
     def from_pretrained(cls, pretrained_model_name_or_path, *model_args, **kwargs):
@@ -287,7 +377,7 @@ class TFAutoModelWithLMHead(object):
             return TFCTRLLMHeadModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
 
         raise ValueError("Unrecognized model identifier in {}. Should contains one of "
-                         "'bert', 'openai-gpt', 'gpt2', 'transfo-xl', 'xlnet', "
+                         "'distilbert', 'bert', 'openai-gpt', 'gpt2', 'transfo-xl', 'xlnet', "
                          "'xlm', 'roberta', 'ctrl'".format(pretrained_model_name_or_path))
 
 
@@ -312,8 +402,39 @@ class TFAutoModelForSequenceClassification(object):
         This class cannot be instantiated using `__init__()` (throws an error).
     """
     def __init__(self):
-        raise EnvironmentError("TFAutoModelWithLMHead is designed to be instantiated "
-            "using the `TFAutoModelWithLMHead.from_pretrained(pretrained_model_name_or_path)` method.")
+        raise EnvironmentError("TFAutoModelForSequenceClassification is designed to be instantiated "
+            "using the `TFAutoModelForSequenceClassification.from_pretrained(pretrained_model_name_or_path)` or "
+            "`TFAutoModelForSequenceClassification.from_config(config)` methods.")
+
+    @classmethod
+    def from_config(cls, config):
+        r""" Instantiates one of the base model classes of the library
+        from a configuration.
+
+            config: (`optional`) instance of a class derived from :class:`~transformers.PretrainedConfig`:
+                The model class to instantiate is selected based on the configuration class:
+                    - isInstance of `distilbert` configuration class: DistilBertModel (DistilBERT model)
+                    - isInstance of `roberta` configuration class: RobertaModel (RoBERTa model)
+                    - isInstance of `bert` configuration class: BertModel (Bert model)
+                    - isInstance of `xlnet` configuration class: XLNetModel (XLNet model)
+                    - isInstance of `xlm` configuration class: XLMModel (XLM model)
+
+        Examples::
+
+            config = BertConfig.from_pretrained('bert-base-uncased')    # Download configuration from S3 and cache.
+            model = AutoModelForSequenceClassification.from_config(config)  # E.g. model was saved using `save_pretrained('./test/saved_model/')`
+        """
+        if isinstance(config, DistilBertConfig):
+            return TFDistilBertForSequenceClassification(config)
+        elif isinstance(config, RobertaConfig):
+            return TFRobertaForSequenceClassification(config)
+        elif isinstance(config, BertConfig):
+            return TFBertForSequenceClassification(config)
+        elif isinstance(config, XLNetConfig):
+            return TFXLNetForSequenceClassification(config)
+        elif isinstance(config, XLMConfig):
+            return TFXLMForSequenceClassification(config)
+        raise ValueError("Unrecognized configuration class {}".format(config))
 
     @classmethod
     def from_pretrained(cls, pretrained_model_name_or_path, *model_args, **kwargs):
@@ -405,7 +526,7 @@ class TFAutoModelForSequenceClassification(object):
             return TFXLMForSequenceClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
 
         raise ValueError("Unrecognized model identifier in {}. Should contains one of "
-                         "'bert', 'xlnet', 'xlm', 'roberta'".format(pretrained_model_name_or_path))
+                         "'distilbert', 'bert', 'xlnet', 'xlm', 'roberta'".format(pretrained_model_name_or_path))
 
 
 class TFAutoModelForQuestionAnswering(object):
@@ -428,8 +549,36 @@ class TFAutoModelForQuestionAnswering(object):
         This class cannot be instantiated using `__init__()` (throws an error).
     """
     def __init__(self):
-        raise EnvironmentError("TFAutoModelWithLMHead is designed to be instantiated "
-            "using the `TFAutoModelWithLMHead.from_pretrained(pretrained_model_name_or_path)` method.")
+        raise EnvironmentError("TFAutoModelForQuestionAnswering is designed to be instantiated "
+            "using the `TFAutoModelForQuestionAnswering.from_pretrained(pretrained_model_name_or_path)` or "
+            "`TFAutoModelForQuestionAnswering.from_config(config)` methods.")
+
+    @classmethod
+    def from_config(cls, config):
+        r""" Instantiates one of the base model classes of the library
+        from a configuration.
+
+            config: (`optional`) instance of a class derived from :class:`~transformers.PretrainedConfig`:
+                The model class to instantiate is selected based on the configuration class:
+                    - isInstance of `distilbert` configuration class: DistilBertModel (DistilBERT model)
+                    - isInstance of `bert` configuration class: BertModel (Bert model)
+                    - isInstance of `xlnet` configuration class: XLNetModel (XLNet model)
+                    - isInstance of `xlm` configuration class: XLMModel (XLM model)
+
+        Examples::
+
+            config = BertConfig.from_pretrained('bert-base-uncased')    # Download configuration from S3 and cache.
+            model = AutoModelForSequenceClassification.from_config(config)  # E.g. model was saved using `save_pretrained('./test/saved_model/')`
+        """
+        if isinstance(config, DistilBertConfig):
+            return TFDistilBertForQuestionAnswering(config)
+        elif isinstance(config, BertConfig):
+            return TFBertForQuestionAnswering(config)
+        elif isinstance(config, XLNetConfig):
+            return TFXLNetForQuestionAnswering(config)
+        elif isinstance(config, XLMConfig):
+            return TFXLMForQuestionAnswering(config)
+        raise ValueError("Unrecognized configuration class {}".format(config))
 
     @classmethod
     def from_pretrained(cls, pretrained_model_name_or_path, *model_args, **kwargs):
@@ -518,4 +667,4 @@ class TFAutoModelForQuestionAnswering(object):
             return TFXLMForQuestionAnsweringSimple.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
 
         raise ValueError("Unrecognized model identifier in {}. Should contains one of "
-                         "'bert', 'xlnet', 'xlm'".format(pretrained_model_name_or_path))
+                         "'distilbert', 'bert', 'xlnet', 'xlm'".format(pretrained_model_name_or_path))
diff --git a/transformers/pipeline.py b/transformers/pipeline.py
index 15adc620b1..f2c55def92 100644
--- a/transformers/pipeline.py
+++ b/transformers/pipeline.py
@@ -58,7 +58,7 @@ class TextClassificationPipeline(object):
             - contains `xlnet`: XLNetModel (XLNet model)
             - contains `xlm`: XLMModel (XLM model)
     """
-    def __init__(self, tokenizer, model):
+    def __init__(self, tokenizer, model, is_compiled=False, is_trained=False):
         self.tokenizer = tokenizer
         self.model = model
         if is_tf_available():
@@ -67,78 +67,13 @@ class TextClassificationPipeline(object):
             self.framework = 'pt'
         else:
             raise ImportError("At least one of PyTorch or TensorFlow 2.0+ should be installed to use CLI training")
-        self.is_compiled = False
+        self.is_compiled = is_compiled
+        self.is_trained = is_trained
 
 
     @classmethod
     def from_pretrained(cls, pretrained_model_name_or_path, **kwargs):
-        r""" Instantiates one of the base model classes of the library
-        from a pre-trained model configuration.
-
-        The model class to instantiate is selected as the first pattern matching
-        in the `pretrained_model_name_or_path` string (in the following order):
-            - contains `distilbert`: DistilBertModel (DistilBERT model)
-            - contains `roberta`: RobertaModel (RoBERTa model)
-            - contains `bert`: BertModel (Bert model)
-            - contains `openai-gpt`: OpenAIGPTModel (OpenAI GPT model)
-            - contains `gpt2`: GPT2Model (OpenAI GPT-2 model)
-            - contains `ctrl`: CTRLModel (Salesforce CTRL  model)
-            - contains `transfo-xl`: TransfoXLModel (Transformer-XL model)
-            - contains `xlnet`: XLNetModel (XLNet model)
-            - contains `xlm`: XLMModel (XLM model)
-
-            The model is set in evaluation mode by default using `model.eval()` (Dropout modules are deactivated)
-            To train the model, you should first set it back in training mode with `model.train()`
-
-        Params:
-            pretrained_model_name_or_path: either:
-
-                - a string with the `shortcut name` of a pre-trained model to load from cache or download, e.g.: ``bert-base-uncased``.
-                - a path to a `directory` containing model weights saved using :func:`~transformers.PreTrainedModel.save_pretrained`, e.g.: ``./my_model_directory/``.
-                - a path or url to a `tensorflow index checkpoint file` (e.g. `./tf_model/model.ckpt.index`). In this case, ``from_tf`` should be set to True and a configuration object should be provided as ``config`` argument. This loading path is slower than converting the TensorFlow checkpoint in a PyTorch model using the provided conversion scripts and loading the PyTorch model afterwards.
-
-            config: (`optional`) instance of a class derived from :class:`~transformers.PretrainedConfig`:
-                Configuration for the model to use instead of an automatically loaded configuation. Configuration can be automatically loaded when:
-
-                - the model is a model provided by the library (loaded with the ``shortcut-name`` string of a pretrained model), or
-                - the model was saved using :func:`~transformers.PreTrainedModel.save_pretrained` and is reloaded by suppling the save directory.
-                - the model is loaded by suppling a local directory as ``pretrained_model_name_or_path`` and a configuration JSON file named `config.json` is found in the directory.
-
-            state_dict: (`optional`) dict:
-                an optional state dictionnary for the model to use instead of a state dictionary loaded from saved weights file.
-                This option can be used if you want to create a model from a pretrained configuration but load your own weights.
-                In this case though, you should check if using :func:`~transformers.PreTrainedModel.save_pretrained` and :func:`~transformers.PreTrainedModel.from_pretrained` is not a simpler option.
-
-            cache_dir: (`optional`) string:
-                Path to a directory in which a downloaded pre-trained model
-                configuration should be cached if the standard cache should not be used.
-
-            force_download: (`optional`) boolean, default False:
-                Force to (re-)download the model weights and configuration files and override the cached versions if they exists.
-
-            proxies: (`optional`) dict, default None:
-                A dictionary of proxy servers to use by protocol or endpoint, e.g.: {'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.
-                The proxies are used on each request.
-
-            output_loading_info: (`optional`) boolean:
-                Set to ``True`` to also return a dictionnary containing missing keys, unexpected keys and error messages.
-
-            kwargs: (`optional`) Remaining dictionary of keyword arguments:
-                Can be used to update the configuration object (after it being loaded) and initiate the model. (e.g. ``output_attention=True``). Behave differently depending on whether a `config` is provided or automatically loaded:
-
-                - If a configuration is provided with ``config``, ``**kwargs`` will be directly passed to the underlying model's ``__init__`` method (we assume all relevant updates to the configuration have already been done)
-                - If a configuration is not provided, ``kwargs`` will be first passed to the configuration class initialization function (:func:`~transformers.PretrainedConfig.from_pretrained`). Each key of ``kwargs`` that corresponds to a configuration attribute will be used to override said attribute with the supplied ``kwargs`` value. Remaining keys that do not correspond to any configuration attribute will be passed to the underlying model's ``__init__`` function.
-
-        Examples::
-
-            model = AutoModel.from_pretrained('bert-base-uncased')    # Download model and configuration from S3 and cache.
-            model = AutoModel.from_pretrained('./test/bert_model/')  # E.g. model was saved using `save_pretrained('./test/saved_model/')`
-            model = AutoModel.from_pretrained('bert-base-uncased', output_attention=True)  # Update configuration during loading
-            assert model.config.output_attention == True
-            # Loading from a TF checkpoint file instead of a PyTorch model (slower)
-            config = AutoConfig.from_json_file('./tf_model/bert_tf_model_config.json')
-            model = AutoModel.from_pretrained('./tf_model/bert_tf_checkpoint.ckpt.index', from_tf=True, config=config)
-
+        r""" Instantiates a pipeline from a pre-trained tokenizer and model.
         """
         # Extract tokenizer and model arguments
         tokenizer_kwargs = {}
@@ -159,9 +94,11 @@ class TextClassificationPipeline(object):
                 # used for both the tokenizer and the model
                 model_kwargs[key] = kwargs[key]
 
+        model_kwargs['output_loading_info'] = True
         tokenizer = AutoTokenizer.from_pretrained(pretrained_model_name_or_path, **tokenizer_kwargs)
-        model = AutoModelForSequenceClassification.from_pretrained(pretrained_model_name_or_path, **model_kwargs)
-        return cls(tokenizer, model)
+        model, loading_info = AutoModelForSequenceClassification.from_pretrained(pretrained_model_name_or_path, **model_kwargs)
+
+        return cls(tokenizer, model, is_trained=bool(not loading_info['missing_keys']))
 
 
     def save_pretrained(self, save_directory):
@@ -240,9 +177,13 @@ class TextClassificationPipeline(object):
                                      validation_data=valid_dataset, validation_steps=valid_steps, **kwargs)
         else:
             raise NotImplementedError
+        self.is_trained = True
 
 
     def __call__(self, text):
+        if not self.is_trained:
+            logger.error("Some weights of the model are not trained. Please fine-tune the model on a classification task before using it.")
+
         inputs = self.tokenizer.encode_plus(text, add_special_tokens=True, return_tensors=self.framework)
         if self.framework == 'tf':
             # TODO trace model

From 7c1697562a38200e0e1a651b014ff0fc07343dd1 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Thu, 17 Oct 2019 13:17:05 +0200
Subject: [PATCH 288/505] compatibility with sklearn and keras

---
 transformers/commands/train.py        |  2 +-
 transformers/data/processors/utils.py |  2 +-
 transformers/pipeline.py              | 87 ++++++++++++++++++---------
 3 files changed, 59 insertions(+), 32 deletions(-)

diff --git a/transformers/commands/train.py b/transformers/commands/train.py
index fc89d48594..878ad21037 100644
--- a/transformers/commands/train.py
+++ b/transformers/commands/train.py
@@ -83,7 +83,7 @@ class TrainCommand(BaseTransformersCLICommand):
 
         self.logger.info('Loading model {}'.format(args.model_name))
         self.model_name = args.model_name
-        self.tokenizer = AutoTokenizer.from_pretrained(args.model_name)
+        self.pipeline = AutoTokenizer.from_pretrained(args.model_name)
         if args.task == 'text_classification':
             self.model = SequenceClassifModel.from_pretrained(args.model_name)
         elif args.task == 'token_classification':
diff --git a/transformers/data/processors/utils.py b/transformers/data/processors/utils.py
index 75bed86042..61b139c02b 100644
--- a/transformers/data/processors/utils.py
+++ b/transformers/data/processors/utils.py
@@ -222,7 +222,7 @@ class SingleSentenceClassificationProcessor(DataProcessor):
         batch_length = max(len(input_ids) for input_ids in all_input_ids)
 
         features = []
-        for (ex_index, (input_ids, example)) in enumerate(zip(all_input_ids, examples)):
+        for (ex_index, (input_ids, example)) in enumerate(zip(all_input_ids, self.examples)):
             if ex_index % 10000 == 0:
                 logger.info("Writing example %d", ex_index)
             # The mask has 1 for real tokens and 0 for padding tokens. Only real
diff --git a/transformers/pipeline.py b/transformers/pipeline.py
index f2c55def92..dc7bcaeac3 100644
--- a/transformers/pipeline.py
+++ b/transformers/pipeline.py
@@ -109,7 +109,32 @@ class TextClassificationPipeline(object):
         self.tokenizer.save_pretrained(save_directory)
 
 
-    def compile(self, learning_rate=3e-5, epsilon=1e-8):
+    def prepare_data(self, train_samples_text, train_samples_labels,
+                     valid_samples_text=None, valid_samples_labels=None,
+                     validation_split=0.1, **kwargs):
+        dataset = SingleSentenceClassificationProcessor.create_from_examples(train_samples_text,
+                                                                             train_samples_labels)
+        num_data_samples = len(dataset)
+        if valid_samples_text is not None and valid_samples_labels is not None:
+            valid_dataset = SingleSentenceClassificationProcessor.create_from_examples(valid_samples_text,
+                                                                                       valid_samples_labels)
+            num_valid_samples = len(valid_dataset)
+            train_dataset = dataset
+            num_train_samples = num_data_samples
+        else:
+            assert 0.0 <= validation_split <= 1.0, "validation_split should be between 0.0 and 1.0"
+            num_valid_samples = int(num_data_samples * validation_split)
+            num_train_samples = num_data_samples - num_valid_samples
+            train_dataset = dataset[num_train_samples]
+            valid_dataset = dataset[num_valid_samples]
+
+        logger.info('Tokenizing and processing dataset')
+        train_dataset = train_dataset.get_features(self.tokenizer, return_tensors=self.framework)
+        valid_dataset = valid_dataset.get_features(self.tokenizer, return_tensors=self.framework)
+        return train_dataset, valid_dataset, num_train_samples, num_valid_samples
+
+
+    def compile(self, learning_rate=3e-5, epsilon=1e-8, **kwargs):
         if self.framework == 'tf':
             logger.info('Preparing model')
             # Prepare training: Compile tf.keras model with optimizer, loss and learning rate schedule
@@ -125,39 +150,20 @@ class TextClassificationPipeline(object):
         self.is_compiled = True
 
 
-    def prepare_data(self, train_samples_text, train_samples_labels,
-                     valid_samples_text=None, valid_samples_labels=None,
-                     validation_split=0.1):
-        dataset = SingleSentenceClassificationProcessor.create_from_examples(train_samples_text,
-                                                                             train_samples_labels)
-        num_data_samples = len(dataset)
-        if valid_samples_text is not None and valid_samples_labels is not None:
-            valid_dataset = SingleSentenceClassificationProcessor.create_from_examples(valid_samples_text,
-                                                                                       valid_samples_labels)
-            num_valid_samples = len(valid_dataset)
-            train_dataset = dataset
-            num_train_samples = num_data_samples
-        else:
-            assert 0.0 < validation_split < 1.0, "validation_split should be between 0.0 and 1.0"
-            num_valid_samples = int(num_data_samples * validation_split)
-            num_train_samples = num_data_samples - num_valid_samples
-            train_dataset = dataset[num_train_samples]
-            valid_dataset = dataset[num_valid_samples]
-
-        logger.info('Tokenizing and processing dataset')
-        train_dataset = train_dataset.get_features(self.tokenizer, return_tensors=self.framework)
-        valid_dataset = valid_dataset.get_features(self.tokenizer, return_tensors=self.framework)
-        return train_dataset, valid_dataset, num_train_samples, num_valid_samples
-
-
-    def fit(self, train_samples_text, train_samples_labels,
+    def fit(self, train_samples_text=None, train_samples_labels=None,
             valid_samples_text=None, valid_samples_labels=None,
             train_batch_size=None, valid_batch_size=None,
             validation_split=0.1,
             **kwargs):
 
+        # Generic compatibility with sklearn and Keras
+        if 'y' in kwargs and train_samples_labels is None:
+            train_samples_labels = kwargs.pop('y')
+        if 'X' in kwargs and train_samples_text is None:
+            train_samples_text = kwargs.pop('X')
+
         if not self.is_compiled:
-            self.compile()
+            self.compile(**kwargs)
 
         datasets = self.prepare_data(train_samples_text, train_samples_labels,
                                      valid_samples_text, valid_samples_labels,
@@ -180,11 +186,32 @@ class TextClassificationPipeline(object):
         self.is_trained = True
 
 
-    def __call__(self, text):
+    def fit_transform(self, *texts, **kwargs):
+        # Generic compatibility with sklearn and Keras
+        self.fit(*texts, **kwargs)
+        return self(*texts, **kwargs)
+
+
+    def transform(self, *texts, **kwargs):
+        # Generic compatibility with sklearn and Keras
+        return self(*texts, **kwargs)
+
+
+    def predict(self, *texts, **kwargs):
+        # Generic compatibility with sklearn and Keras
+        return self(*texts, **kwargs)
+
+
+    def __call__(self, *texts, **kwargs):
+        # Generic compatibility with sklearn and Keras
+        if 'X' in kwargs and not texts:
+            texts = kwargs.pop('X')
+
         if not self.is_trained:
             logger.error("Some weights of the model are not trained. Please fine-tune the model on a classification task before using it.")
 
-        inputs = self.tokenizer.encode_plus(text, add_special_tokens=True, return_tensors=self.framework)
+        inputs = self.tokenizer.batch_encode_plus(texts, add_special_tokens=True, return_tensors=self.framework)
+
         if self.framework == 'tf':
             # TODO trace model
             predictions = self.model(**inputs)[0]

From 31a3a73ee39d7de28e69e62d8c1a0988a765a0e0 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Thu, 17 Oct 2019 15:10:11 +0200
Subject: [PATCH 289/505] updating CLI

---
 transformers-cli                      |  4 ++
 transformers/__init__.py              |  4 +-
 transformers/commands/train.py        | 89 ++++++++++---------------
 transformers/data/processors/utils.py | 48 +++++++++-----
 transformers/modeling_tf_utils.py     | 31 +++++++++
 transformers/pipeline.py              | 95 ++++++++++++++-------------
 6 files changed, 155 insertions(+), 116 deletions(-)

diff --git a/transformers-cli b/transformers-cli
index 7b0905d4b4..397b382308 100644
--- a/transformers-cli
+++ b/transformers-cli
@@ -3,6 +3,8 @@ from argparse import ArgumentParser
 
 from transformers.commands.serving import ServeCommand
 from transformers.commands.user import UserCommands
+from transformers.commands.train import TrainCommand
+from transformers.commands.convert import ConvertCommand
 
 if __name__ == '__main__':
     parser = ArgumentParser('Transformers CLI tool', usage='transformers-cli <command> [<args>]')
@@ -11,6 +13,8 @@ if __name__ == '__main__':
     # Register commands
     ServeCommand.register_subcommand(commands_parser)
     UserCommands.register_subcommand(commands_parser)
+    TrainCommand.register_subcommand(commands_parser)
+    ConvertCommand.register_subcommand(commands_parser)
 
     # Let's go
     args = parser.parse_args()
diff --git a/transformers/__init__.py b/transformers/__init__.py
index a71a291a44..26036d2e8d 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -25,7 +25,6 @@ from .file_utils import (TRANSFORMERS_CACHE, PYTORCH_TRANSFORMERS_CACHE, PYTORCH
 from .data import (is_sklearn_available,
                    InputExample, InputFeatures, DataProcessor,
                    SingleSentenceClassificationProcessor,
-                   convert_examples_to_features,
                    glue_output_modes, glue_convert_examples_to_features,
                    glue_processors, glue_tasks_num_labels,
                    xnli_output_modes, xnli_processors, xnli_tasks_num_labels,
@@ -66,6 +65,9 @@ from .configuration_distilbert import DistilBertConfig, DISTILBERT_PRETRAINED_CO
 from .configuration_albert import AlbertConfig, ALBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_camembert import CamembertConfig, CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
 
+# Pipelines
+from .pipeline import TextClassificationPipeline
+
 # Modeling
 if is_torch_available():
     from .modeling_utils import (PreTrainedModel, prune_layer, Conv1D)
diff --git a/transformers/commands/train.py b/transformers/commands/train.py
index 878ad21037..7b26745881 100644
--- a/transformers/commands/train.py
+++ b/transformers/commands/train.py
@@ -3,14 +3,11 @@ from argparse import ArgumentParser, Namespace
 from logging import getLogger
 
 from transformers.commands import BaseTransformersCLICommand
-from transformers import (AutoTokenizer, is_tf_available, is_torch_available,
-                          SingleSentenceClassificationProcessor,
-                          convert_examples_to_features)
-if is_tf_available():
-    from transformers import TFAutoModelForSequenceClassification as SequenceClassifModel
-elif is_torch_available():
-    from transformers import AutoModelForSequenceClassification as SequenceClassifModel
-else:
+from transformers import (is_tf_available, is_torch_available,
+                          TextClassificationPipeline,
+                          SingleSentenceClassificationProcessor as Processor)
+
+if not is_tf_available() and not is_torch_available():
     raise ImportError("At least one of PyTorch or TensorFlow 2.0+ should be installed to use CLI training")
 
 # TF training parameters
@@ -35,16 +32,18 @@ class TrainCommand(BaseTransformersCLICommand):
         :return:
         """
         train_parser = parser.add_parser('train', help='CLI tool to train a model on a task.')
+
         train_parser.add_argument('--train_data', type=str, required=True,
                                   help="path to train (and optionally evaluation) dataset as a csv with "
                                        "tab separated labels and sentences.")
-
         train_parser.add_argument('--column_label', type=int, default=0,
                                   help='Column of the dataset csv file with example labels.')
         train_parser.add_argument('--column_text', type=int, default=1,
                                   help='Column of the dataset csv file with example texts.')
         train_parser.add_argument('--column_id', type=int, default=2,
                                   help='Column of the dataset csv file with example ids.')
+        train_parser.add_argument('--skip_first_row', action='store_true',
+                                  help='Skip the first row of the csv file (headers).')
 
         train_parser.add_argument('--validation_data', type=str, default='',
                                   help='path to validation dataset.')
@@ -74,39 +73,38 @@ class TrainCommand(BaseTransformersCLICommand):
 
         self.framework = 'tf' if is_tf_available() else 'torch'
 
-        os.makedirs(args.output)
+        os.makedirs(args.output, exist_ok=True)
+        assert os.path.isdir(args.output)
         self.output = args.output
 
         self.column_label = args.column_label
         self.column_text = args.column_text
         self.column_id = args.column_id
 
-        self.logger.info('Loading model {}'.format(args.model_name))
-        self.model_name = args.model_name
-        self.pipeline = AutoTokenizer.from_pretrained(args.model_name)
+        self.logger.info('Loading {} pipeline for {}'.format(args.task, args.model))
         if args.task == 'text_classification':
-            self.model = SequenceClassifModel.from_pretrained(args.model_name)
+            self.pipeline = TextClassificationPipeline.from_pretrained(args.model)
         elif args.task == 'token_classification':
             raise NotImplementedError
         elif args.task == 'question_answering':
             raise NotImplementedError
 
         self.logger.info('Loading dataset from {}'.format(args.train_data))
-        dataset = SingleSentenceClassificationProcessor.create_from_csv(args.train_data)
-        num_data_samples = len(dataset)
+        self.train_dataset = Processor.create_from_csv(args.train_data,
+                                                       column_label=args.column_label,
+                                                       column_text=args.column_text,
+                                                       column_id=args.column_id,
+                                                       skip_first_row=args.skip_first_row)
+        self.valid_dataset = None
         if args.validation_data:
             self.logger.info('Loading validation dataset from {}'.format(args.validation_data))
-            self.valid_dataset = SingleSentenceClassificationProcessor.create_from_csv(args.validation_data)
-            self.num_valid_samples = len(self.valid_dataset)
-            self.train_dataset = dataset
-            self.num_train_samples = num_data_samples
-        else:
-            assert 0.0 < args.validation_split < 1.0, "--validation_split should be between 0.0 and 1.0"
-            self.num_valid_samples = num_data_samples * args.validation_split
-            self.num_train_samples = num_data_samples - self.num_valid_samples
-            self.train_dataset = dataset[self.num_train_samples]
-            self.valid_dataset = dataset[self.num_valid_samples]
+            self.valid_dataset = Processor.create_from_csv(args.validation_data,
+                                                           column_label=args.column_label,
+                                                           column_text=args.column_text,
+                                                           column_id=args.column_id,
+                                                           skip_first_row=args.skip_first_row)
 
+        self.validation_split = args.validation_split
         self.train_batch_size = args.train_batch_size
         self.valid_batch_size = args.valid_batch_size
         self.learning_rate = args.learning_rate
@@ -121,34 +119,13 @@ class TrainCommand(BaseTransformersCLICommand):
         raise NotImplementedError
 
     def run_tf(self):
-        import tensorflow as tf
+        self.pipeline.fit(self.train_dataset,
+                          validation_data=self.valid_dataset,
+                          validation_split=self.validation_split,
+                          learning_rate=self.learning_rate,
+                          adam_epsilon=self.adam_epsilon,
+                          train_batch_size=self.train_batch_size,
+                          valid_batch_size=self.valid_batch_size)
 
-        tf.config.optimizer.set_jit(USE_XLA)
-        tf.config.optimizer.set_experimental_options({"auto_mixed_precision": USE_AMP})
-
-        # Prepare dataset as a tf.train_data.Dataset instance
-        self.logger.info('Tokenizing and processing dataset')
-        train_dataset = self.train_dataset.get_features(self.tokenizer)
-        valid_dataset = self.valid_dataset.get_features(self.tokenizer)
-        train_dataset = train_dataset.shuffle(128).batch(self.train_batch_size).repeat(-1)
-        valid_dataset = valid_dataset.batch(self.valid_batch_size)
-
-        # Prepare training: Compile tf.keras model with optimizer, loss and learning rate schedule 
-        opt = tf.keras.optimizers.Adam(learning_rate=args.learning_rate, epsilon=self.adam_epsilon)
-        if USE_AMP:
-            # loss scaling is currently required when using mixed precision
-            opt = tf.keras.mixed_precision.experimental.LossScaleOptimizer(opt, 'dynamic')
-        loss = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True)
-        metric = tf.keras.metrics.SparseCategoricalAccuracy('accuracy')
-        self.model.compile(optimizer=opt, loss=loss, metrics=[metric])
-
-        # Train and evaluate using tf.keras.Model.fit()
-        train_steps = self.num_train_samples//self.train_batch_size
-        valid_steps = self.num_valid_samples//self.valid_batch_size
-
-        self.logger.info('Training model')
-        history = self.model.fit(train_dataset, epochs=2, steps_per_epoch=train_steps,
-                                 validation_data=valid_dataset, validation_steps=valid_steps)
-
-        # Save trained model
-        self.model.save_pretrained(self.output)
+        # Save trained pipeline
+        self.pipeline.save_pretrained(self.output)
diff --git a/transformers/data/processors/utils.py b/transformers/data/processors/utils.py
index 61b139c02b..ee234e6e90 100644
--- a/transformers/data/processors/utils.py
+++ b/transformers/data/processors/utils.py
@@ -122,14 +122,30 @@ class SingleSentenceClassificationProcessor(DataProcessor):
         return self.examples[idx]
 
     @classmethod
-    def create_from_csv(cls, file_name, **kwargs):
+    def create_from_csv(cls, file_name, split_name='', column_label=0, column_text=1,
+                        column_id=None, skip_first_row=False, **kwargs):
         processor = cls(**kwargs)
-        processor.add_examples_from_csv(file_name)
+        processor.add_examples_from_csv(file_name,
+                                        split_name=split_name,
+                                        column_label=column_label,
+                                        column_text=column_text,
+                                        column_id=column_id,
+                                        skip_first_row=skip_first_row,
+                                        overwrite_labels=True,
+                                        overwrite_examples=True)
+        return processor
+
+    @classmethod
+    def create_from_examples(cls, texts_or_text_and_labels, labels=None, **kwargs):
+        processor = cls(**kwargs)
+        processor.add_examples(texts_or_text_and_labels, labels=labels)
         return processor
 
     def add_examples_from_csv(self, file_name, split_name='', column_label=0, column_text=1, column_id=None,
-                              overwrite_labels=False, overwrite_examples=False):
+                              skip_first_row=False, overwrite_labels=False, overwrite_examples=False):
         lines = self._read_tsv(file_name)
+        if skip_first_row:
+            lines = lines[1:]
         texts = []
         labels = []
         ids = []
@@ -144,15 +160,21 @@ class SingleSentenceClassificationProcessor(DataProcessor):
 
         return self.add_examples(texts, labels, ids, overwrite_labels=overwrite_labels, overwrite_examples=overwrite_examples)
 
-    def add_examples(self, texts, labels, ids=None, overwrite_labels=False, overwrite_examples=False):
+    def add_examples(self, texts_or_text_and_labels, labels=None, ids=None,
+                     overwrite_labels=False, overwrite_examples=False):
+        assert labels is None or len(texts_or_text_and_labels) == len(labels)
+        assert ids is None or len(texts_or_text_and_labels) == len(ids)
         if ids is None:
-            ids = [None] * len(texts)
-        assert len(texts) == len(labels)
-        assert len(texts) == len(ids)
-
+            ids = [None] * len(texts_or_text_and_labels)
+        if labels is None:
+            labels = [None] * len(texts_or_text_and_labels)
         examples = []
         added_labels = set()
-        for (text, label, guid) in zip(texts, labels, ids):
+        for (text_or_text_and_label, label, guid) in zip(texts_or_text_and_labels, labels, ids):
+            if isinstance(text_or_text_and_label, (tuple, list)) and label is None:
+                text, label = text_or_text_and_label
+            else:
+                text = text_or_text_and_label
             added_labels.add(label)
             examples.append(InputExample(guid=guid, text_a=text, text_b=None, label=label))
 
@@ -170,12 +192,6 @@ class SingleSentenceClassificationProcessor(DataProcessor):
 
         return self.examples
 
-    @classmethod
-    def create_from_examples(cls, texts, labels, **kwargs):
-        processor = cls(**kwargs)
-        processor.add_examples(texts, labels)
-        return processor
-
     def get_features(self,
                      tokenizer,
                      max_length=None,
@@ -204,6 +220,8 @@ class SingleSentenceClassificationProcessor(DataProcessor):
             a list of task-specific ``InputFeatures`` which can be fed to the model.
 
         """
+        if max_length is None:
+            max_length = tokenizer.max_len
 
         label_map = {label: i for i, label in enumerate(self.labels)}
 
diff --git a/transformers/modeling_tf_utils.py b/transformers/modeling_tf_utils.py
index ed8fdb74c9..6c48f3eed2 100644
--- a/transformers/modeling_tf_utils.py
+++ b/transformers/modeling_tf_utils.py
@@ -22,6 +22,8 @@ import logging
 import os
 
 import tensorflow as tf
+from tensorflow.python.keras.saving import hdf5_format
+import h5py
 
 from .configuration_utils import PretrainedConfig
 from .file_utils import cached_path, WEIGHTS_NAME, TF_WEIGHTS_NAME, TF2_WEIGHTS_NAME
@@ -206,6 +208,9 @@ class TFPreTrainedModel(tf.keras.Model):
                 A dictionary of proxy servers to use by protocol or endpoint, e.g.: {'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.
                 The proxies are used on each request.
 
+            output_loading_info: (`optional`) boolean:
+                Set to ``True`` to also return a dictionnary containing missing keys, unexpected keys and error messages.
+
             kwargs: (`optional`) Remaining dictionary of keyword arguments:
                 Can be used to update the configuration object (after it being loaded) and initiate the model. (e.g. ``output_attention=True``). Behave differently depending on whether a `config` is provided or automatically loaded:
 
@@ -229,6 +234,7 @@ class TFPreTrainedModel(tf.keras.Model):
         force_download = kwargs.pop('force_download', False)
         resume_download = kwargs.pop('resume_download', False)
         proxies = kwargs.pop('proxies', None)
+        output_loading_info = kwargs.pop('output_loading_info', False)
 
         # Load config
         if config is None:
@@ -304,6 +310,31 @@ class TFPreTrainedModel(tf.keras.Model):
 
         ret = model(model.dummy_inputs, training=False)  # Make sure restore ops are run
 
+        # Check if the models are the same to output loading informations
+        with h5py.File(resolved_archive_file, 'r') as f:
+            if 'layer_names' not in f.attrs and 'model_weights' in f:
+                f = f['model_weights']
+            hdf5_layer_names = set(hdf5_format.load_attributes_from_hdf5_group(f, 'layer_names'))
+        model_layer_names = set(layer.name for layer in model.layers)
+        missing_keys = list(model_layer_names - hdf5_layer_names)
+        unexpected_keys = list(hdf5_layer_names - model_layer_names)
+        error_msgs = []
+
+        if len(missing_keys) > 0:
+            logger.info("Layers of {} not initialized from pretrained model: {}".format(
+                model.__class__.__name__, missing_keys))
+        if len(unexpected_keys) > 0:
+            logger.info("Layers from pretrained model not used in {}: {}".format(
+                model.__class__.__name__, unexpected_keys))
+        if len(error_msgs) > 0:
+            raise RuntimeError('Error(s) in loading weights for {}:\n\t{}'.format(
+                            model.__class__.__name__, "\n\t".join(error_msgs)))
+        if output_loading_info:
+            loading_info = {"missing_keys": missing_keys,
+                            "unexpected_keys": unexpected_keys,
+                            "error_msgs": error_msgs}
+            return model, loading_info
+
         return model
 
 class TFConv1D(tf.keras.layers.Layer):
diff --git a/transformers/pipeline.py b/transformers/pipeline.py
index dc7bcaeac3..6e55ca4d7e 100644
--- a/transformers/pipeline.py
+++ b/transformers/pipeline.py
@@ -17,18 +17,22 @@
 from __future__ import absolute_import, division, print_function, unicode_literals
 import os
 import logging
+import six
 
-from .modeling_auto import (AutoModel, AutoModelForQuestionAnswering,
-                            AutoModelForSequenceClassification,
-                            AutoModelWithLMHead)
 from .tokenization_auto import AutoTokenizer
 from .file_utils import add_start_docstrings, is_tf_available, is_torch_available
 from .data.processors import SingleSentenceClassificationProcessor
 
 if is_tf_available():
     import tensorflow as tf
+    from .modeling_tf_auto import (TFAutoModel, TFAutoModelForQuestionAnswering,
+                                    TFAutoModelForSequenceClassification,
+                                    TFAutoModelWithLMHead)
 if is_torch_available():
     import torch
+    from .modeling_auto import (AutoModel, AutoModelForQuestionAnswering,
+                                AutoModelForSequenceClassification,
+                                AutoModelWithLMHead)
 
 logger = logging.getLogger(__name__)
 
@@ -61,12 +65,6 @@ class TextClassificationPipeline(object):
     def __init__(self, tokenizer, model, is_compiled=False, is_trained=False):
         self.tokenizer = tokenizer
         self.model = model
-        if is_tf_available():
-            self.framework = 'tf'
-        elif is_torch_available():
-            self.framework = 'pt'
-        else:
-            raise ImportError("At least one of PyTorch or TensorFlow 2.0+ should be installed to use CLI training")
         self.is_compiled = is_compiled
         self.is_trained = is_trained
 
@@ -94,9 +92,12 @@ class TextClassificationPipeline(object):
                 # used for both the tokenizer and the model
                 model_kwargs[key] = kwargs[key]
 
-        model_kwargs['output_loading_info'] = True
         tokenizer = AutoTokenizer.from_pretrained(pretrained_model_name_or_path, **tokenizer_kwargs)
-        model, loading_info = AutoModelForSequenceClassification.from_pretrained(pretrained_model_name_or_path, **model_kwargs)
+        model_kwargs['output_loading_info'] = True
+        if is_tf_available():
+            model, loading_info = TFAutoModelForSequenceClassification.from_pretrained(pretrained_model_name_or_path, **model_kwargs)
+        else:
+            model, loading_info = AutoModelForSequenceClassification.from_pretrained(pretrained_model_name_or_path, **model_kwargs)
 
         return cls(tokenizer, model, is_trained=bool(not loading_info['missing_keys']))
 
@@ -109,36 +110,42 @@ class TextClassificationPipeline(object):
         self.tokenizer.save_pretrained(save_directory)
 
 
-    def prepare_data(self, train_samples_text, train_samples_labels,
-                     valid_samples_text=None, valid_samples_labels=None,
+    def prepare_data(self, x, y=None,
+                     validation_data=None,
                      validation_split=0.1, **kwargs):
-        dataset = SingleSentenceClassificationProcessor.create_from_examples(train_samples_text,
-                                                                             train_samples_labels)
+        dataset = x
+        if not isinstance(x, SingleSentenceClassificationProcessor):
+            dataset = SingleSentenceClassificationProcessor.create_from_examples(x, y)
         num_data_samples = len(dataset)
-        if valid_samples_text is not None and valid_samples_labels is not None:
-            valid_dataset = SingleSentenceClassificationProcessor.create_from_examples(valid_samples_text,
-                                                                                       valid_samples_labels)
+
+        if validation_data is not None:
+            valid_dataset = validation_data
+            if not isinstance(validation_data, SingleSentenceClassificationProcessor):
+                valid_dataset = SingleSentenceClassificationProcessor.create_from_examples(validation_data)
+
             num_valid_samples = len(valid_dataset)
             train_dataset = dataset
             num_train_samples = num_data_samples
         else:
             assert 0.0 <= validation_split <= 1.0, "validation_split should be between 0.0 and 1.0"
-            num_valid_samples = int(num_data_samples * validation_split)
+            num_valid_samples = max(int(num_data_samples * validation_split), 1)
             num_train_samples = num_data_samples - num_valid_samples
-            train_dataset = dataset[num_train_samples]
-            valid_dataset = dataset[num_valid_samples]
+            train_dataset = dataset[num_valid_samples:]
+            valid_dataset = dataset[:num_valid_samples]
 
         logger.info('Tokenizing and processing dataset')
-        train_dataset = train_dataset.get_features(self.tokenizer, return_tensors=self.framework)
-        valid_dataset = valid_dataset.get_features(self.tokenizer, return_tensors=self.framework)
-        return train_dataset, valid_dataset, num_train_samples, num_valid_samples
+        train_dataset = train_dataset.get_features(self.tokenizer,
+                                                   return_tensors='tf' if is_tf_available() else 'pt')
+        valid_dataset = valid_dataset.get_features(self.tokenizer,
+                                                   return_tensors='tf' if is_tf_available() else 'pt')
+        return train_dataset, valid_dataset
 
 
-    def compile(self, learning_rate=3e-5, epsilon=1e-8, **kwargs):
-        if self.framework == 'tf':
+    def compile(self, learning_rate=3e-5, adam_epsilon=1e-8, **kwargs):
+        if is_tf_available():
             logger.info('Preparing model')
             # Prepare training: Compile tf.keras model with optimizer, loss and learning rate schedule
-            opt = tf.keras.optimizers.Adam(learning_rate=learning_rate, epsilon=epsilon)
+            opt = tf.keras.optimizers.Adam(learning_rate=learning_rate, epsilon=adam_epsilon)
             if USE_AMP:
                 # loss scaling is currently required when using mixed precision
                 opt = tf.keras.mixed_precision.experimental.LossScaleOptimizer(opt, 'dynamic')
@@ -150,39 +157,37 @@ class TextClassificationPipeline(object):
         self.is_compiled = True
 
 
-    def fit(self, train_samples_text=None, train_samples_labels=None,
-            valid_samples_text=None, valid_samples_labels=None,
-            train_batch_size=None, valid_batch_size=None,
+    def fit(self, X=None, y=None,
+            validation_data=None,
             validation_split=0.1,
+            train_batch_size=None,
+            valid_batch_size=None,
             **kwargs):
 
-        # Generic compatibility with sklearn and Keras
-        if 'y' in kwargs and train_samples_labels is None:
-            train_samples_labels = kwargs.pop('y')
-        if 'X' in kwargs and train_samples_text is None:
-            train_samples_text = kwargs.pop('X')
-
         if not self.is_compiled:
             self.compile(**kwargs)
 
-        datasets = self.prepare_data(train_samples_text, train_samples_labels,
-                                     valid_samples_text, valid_samples_labels,
-                                     validation_split)
-        train_dataset, valid_dataset, num_train_samples, num_valid_samples = datasets
+        train_dataset, valid_dataset = self.prepare_data(X, y=y,
+                                                         validation_data=validation_data,
+                                                         validation_split=validation_split)
+        num_train_samples = len(train_dataset)
+        num_valid_samples = len(valid_dataset)
 
         train_steps = num_train_samples//train_batch_size
         valid_steps = num_valid_samples//valid_batch_size
 
-        if self.framework == 'tf':
+        if is_tf_available():
             # Prepare dataset as a tf.train_data.Dataset instance
             train_dataset = train_dataset.shuffle(128).batch(train_batch_size).repeat(-1)
             valid_dataset = valid_dataset.batch(valid_batch_size)
 
             logger.info('Training TF 2.0 model')
             history = self.model.fit(train_dataset, epochs=2, steps_per_epoch=train_steps,
-                                     validation_data=valid_dataset, validation_steps=valid_steps, **kwargs)
+                                     validation_data=valid_dataset, validation_steps=valid_steps,
+                                     **kwargs)
         else:
             raise NotImplementedError
+
         self.is_trained = True
 
 
@@ -210,9 +215,11 @@ class TextClassificationPipeline(object):
         if not self.is_trained:
             logger.error("Some weights of the model are not trained. Please fine-tune the model on a classification task before using it.")
 
-        inputs = self.tokenizer.batch_encode_plus(texts, add_special_tokens=True, return_tensors=self.framework)
+        inputs = self.tokenizer.batch_encode_plus(texts,
+                                                  add_special_tokens=True,
+                                                  return_tensors='tf' if is_tf_available() else 'pt')
 
-        if self.framework == 'tf':
+        if is_tf_available():
             # TODO trace model
             predictions = self.model(**inputs)[0]
         else:

From 81babb227e6d6505be088ac452f3cda8a14c2255 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Tue, 3 Dec 2019 14:56:57 +0100
Subject: [PATCH 290/505] Added download command through the cli.

It allows to predownload models and tokenizers.
---
 transformers-cli                  |  4 +++-
 transformers/commands/download.py | 29 +++++++++++++++++++++++++++++
 2 files changed, 32 insertions(+), 1 deletion(-)
 mode change 100644 => 100755 transformers-cli
 create mode 100644 transformers/commands/download.py

diff --git a/transformers-cli b/transformers-cli
old mode 100644
new mode 100755
index 397b382308..168e6e6f32
--- a/transformers-cli
+++ b/transformers-cli
@@ -1,6 +1,7 @@
 #!/usr/bin/env python
 from argparse import ArgumentParser
 
+from transformers.commands.download import DownloadCommand
 from transformers.commands.serving import ServeCommand
 from transformers.commands.user import UserCommands
 from transformers.commands.train import TrainCommand
@@ -11,10 +12,11 @@ if __name__ == '__main__':
     commands_parser = parser.add_subparsers(help='transformers-cli command helpers')
 
     # Register commands
+    ConvertCommand.register_subcommand(commands_parser)
+    DownloadCommand.register_subcommand(commands_parser)
     ServeCommand.register_subcommand(commands_parser)
     UserCommands.register_subcommand(commands_parser)
     TrainCommand.register_subcommand(commands_parser)
-    ConvertCommand.register_subcommand(commands_parser)
 
     # Let's go
     args = parser.parse_args()
diff --git a/transformers/commands/download.py b/transformers/commands/download.py
new file mode 100644
index 0000000000..0938f135d2
--- /dev/null
+++ b/transformers/commands/download.py
@@ -0,0 +1,29 @@
+from argparse import ArgumentParser
+
+from transformers.commands import BaseTransformersCLICommand
+
+
+def download_command_factory(args):
+    return DownloadCommand(args.model, args.cache_dir, args.force)
+
+
+class DownloadCommand(BaseTransformersCLICommand):
+
+    @staticmethod
+    def register_subcommand(parser: ArgumentParser):
+        download_parser = parser.add_parser('download')
+        download_parser.add_argument('--cache-dir', type=str, default=None, help='Path to location to store the models')
+        download_parser.add_argument('--force',  action='store_true', help='Force the model to be download even if already in cache-dir')
+        download_parser.add_argument('model', type=str, help='Name of the model to download')
+        download_parser.set_defaults(func=download_command_factory)
+
+    def __init__(self, model: str, cache: str, force: bool):
+        self._model = model
+        self._cache = cache
+        self._force = force
+
+    def run(self):
+        from transformers import AutoModel, AutoTokenizer
+
+        AutoModel.from_pretrained(self._model, cache_dir=self._cache, force_download=self._force)
+        AutoTokenizer.from_pretrained(self._model, cache_dir=self._cache, force_download=self._force)
\ No newline at end of file

From e1d89cb24d13d158966c190bb75ece38eae26746 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Fri, 6 Dec 2019 00:52:04 +0100
Subject: [PATCH 291/505] Added QuestionAnsweringPipeline with batch support.

---
 transformers/__init__.py  |   7 +-
 transformers/pipeline.py  | 229 --------------------------------------
 transformers/pipelines.py | 222 ++++++++++++++++++++++++++++++++++++
 3 files changed, 226 insertions(+), 232 deletions(-)
 mode change 100644 => 100755 transformers/__init__.py
 delete mode 100644 transformers/pipeline.py
 create mode 100755 transformers/pipelines.py

diff --git a/transformers/__init__.py b/transformers/__init__.py
old mode 100644
new mode 100755
index 26036d2e8d..4300409257
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -65,9 +65,6 @@ from .configuration_distilbert import DistilBertConfig, DISTILBERT_PRETRAINED_CO
 from .configuration_albert import AlbertConfig, ALBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_camembert import CamembertConfig, CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
 
-# Pipelines
-from .pipeline import TextClassificationPipeline
-
 # Modeling
 if is_torch_available():
     from .modeling_utils import (PreTrainedModel, prune_layer, Conv1D)
@@ -193,6 +190,10 @@ from .modeling_tf_pytorch_utils import (convert_tf_weight_name_to_pt_weight_name
                                         load_tf2_weights_in_pytorch_model,
                                         load_tf2_model_in_pytorch_model)
 
+# Pipelines
+# from .pipeline_ import TextClassificationPipeline
+from .pipelines import Pipeline, pipeline, TextClassificationPipeline
+
 if not is_tf_available() and not is_torch_available():
     logger.warning("Neither PyTorch nor TensorFlow >= 2.0 have been found."
                    "Models won't be available and only tokenizers, configuration"
diff --git a/transformers/pipeline.py b/transformers/pipeline.py
deleted file mode 100644
index 6e55ca4d7e..0000000000
--- a/transformers/pipeline.py
+++ /dev/null
@@ -1,229 +0,0 @@
-# coding=utf-8
-# Copyright 2018 The HuggingFace Inc. team.
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-""" Pipeline class: Tokenizer + Model. """
-
-from __future__ import absolute_import, division, print_function, unicode_literals
-import os
-import logging
-import six
-
-from .tokenization_auto import AutoTokenizer
-from .file_utils import add_start_docstrings, is_tf_available, is_torch_available
-from .data.processors import SingleSentenceClassificationProcessor
-
-if is_tf_available():
-    import tensorflow as tf
-    from .modeling_tf_auto import (TFAutoModel, TFAutoModelForQuestionAnswering,
-                                    TFAutoModelForSequenceClassification,
-                                    TFAutoModelWithLMHead)
-if is_torch_available():
-    import torch
-    from .modeling_auto import (AutoModel, AutoModelForQuestionAnswering,
-                                AutoModelForSequenceClassification,
-                                AutoModelWithLMHead)
-
-logger = logging.getLogger(__name__)
-
-# TF training parameters
-USE_XLA = False
-USE_AMP = False
-
-class TextClassificationPipeline(object):
-    r"""
-        :class:`~transformers.TextClassificationPipeline` is a class encapsulating a pretrained model and
-        its tokenizer and will be instantiated as one of the base model classes of the library
-        when created with the `Pipeline.from_pretrained(pretrained_model_name_or_path)`
-        class method.
-
-        The `from_pretrained()` method takes care of returning the correct model class instance
-        using pattern matching on the `pretrained_model_name_or_path` string.
-
-        The base model class to instantiate is selected as the first pattern matching
-        in the `pretrained_model_name_or_path` string (in the following order):
-            - contains `distilbert`: DistilBertModel (DistilBERT model)
-            - contains `roberta`: RobertaModel (RoBERTa model)
-            - contains `bert`: BertModel (Bert model)
-            - contains `openai-gpt`: OpenAIGPTModel (OpenAI GPT model)
-            - contains `gpt2`: GPT2Model (OpenAI GPT-2 model)
-            - contains `ctrl`: CTRLModel (Salesforce CTRL  model)
-            - contains `transfo-xl`: TransfoXLModel (Transformer-XL model)
-            - contains `xlnet`: XLNetModel (XLNet model)
-            - contains `xlm`: XLMModel (XLM model)
-    """
-    def __init__(self, tokenizer, model, is_compiled=False, is_trained=False):
-        self.tokenizer = tokenizer
-        self.model = model
-        self.is_compiled = is_compiled
-        self.is_trained = is_trained
-
-
-    @classmethod
-    def from_pretrained(cls, pretrained_model_name_or_path, **kwargs):
-        r""" Instantiates a pipeline from a pre-trained tokenizer and model.
-        """
-        # Extract tokenizer and model arguments
-        tokenizer_kwargs = {}
-        for key in kwargs:
-            if key.startswith('tokenizer_'):
-                # Specific to the tokenizer
-                tokenizer_kwargs[key.replace('tokenizer_', '')] = kwargs.pop(key)
-            elif not key.startswith('model_'):
-                # used for both the tokenizer and the model
-                tokenizer_kwargs[key] = kwargs[key]
-
-        model_kwargs = {}
-        for key in kwargs:
-            if key.startswith('model_'):
-                # Specific to the model
-                model_kwargs[key.replace('model_', '')] = kwargs.pop(key)
-            elif not key.startswith('tokenizer_'):
-                # used for both the tokenizer and the model
-                model_kwargs[key] = kwargs[key]
-
-        tokenizer = AutoTokenizer.from_pretrained(pretrained_model_name_or_path, **tokenizer_kwargs)
-        model_kwargs['output_loading_info'] = True
-        if is_tf_available():
-            model, loading_info = TFAutoModelForSequenceClassification.from_pretrained(pretrained_model_name_or_path, **model_kwargs)
-        else:
-            model, loading_info = AutoModelForSequenceClassification.from_pretrained(pretrained_model_name_or_path, **model_kwargs)
-
-        return cls(tokenizer, model, is_trained=bool(not loading_info['missing_keys']))
-
-
-    def save_pretrained(self, save_directory):
-        if not os.path.isdir(save_directory):
-            logger.error("Saving directory ({}) should be a directory".format(save_directory))
-            return
-        self.model.save_pretrained(save_directory)
-        self.tokenizer.save_pretrained(save_directory)
-
-
-    def prepare_data(self, x, y=None,
-                     validation_data=None,
-                     validation_split=0.1, **kwargs):
-        dataset = x
-        if not isinstance(x, SingleSentenceClassificationProcessor):
-            dataset = SingleSentenceClassificationProcessor.create_from_examples(x, y)
-        num_data_samples = len(dataset)
-
-        if validation_data is not None:
-            valid_dataset = validation_data
-            if not isinstance(validation_data, SingleSentenceClassificationProcessor):
-                valid_dataset = SingleSentenceClassificationProcessor.create_from_examples(validation_data)
-
-            num_valid_samples = len(valid_dataset)
-            train_dataset = dataset
-            num_train_samples = num_data_samples
-        else:
-            assert 0.0 <= validation_split <= 1.0, "validation_split should be between 0.0 and 1.0"
-            num_valid_samples = max(int(num_data_samples * validation_split), 1)
-            num_train_samples = num_data_samples - num_valid_samples
-            train_dataset = dataset[num_valid_samples:]
-            valid_dataset = dataset[:num_valid_samples]
-
-        logger.info('Tokenizing and processing dataset')
-        train_dataset = train_dataset.get_features(self.tokenizer,
-                                                   return_tensors='tf' if is_tf_available() else 'pt')
-        valid_dataset = valid_dataset.get_features(self.tokenizer,
-                                                   return_tensors='tf' if is_tf_available() else 'pt')
-        return train_dataset, valid_dataset
-
-
-    def compile(self, learning_rate=3e-5, adam_epsilon=1e-8, **kwargs):
-        if is_tf_available():
-            logger.info('Preparing model')
-            # Prepare training: Compile tf.keras model with optimizer, loss and learning rate schedule
-            opt = tf.keras.optimizers.Adam(learning_rate=learning_rate, epsilon=adam_epsilon)
-            if USE_AMP:
-                # loss scaling is currently required when using mixed precision
-                opt = tf.keras.mixed_precision.experimental.LossScaleOptimizer(opt, 'dynamic')
-            loss = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True)
-            metric = tf.keras.metrics.SparseCategoricalAccuracy('accuracy')
-            self.model.compile(optimizer=opt, loss=loss, metrics=[metric])
-        else:
-            raise NotImplementedError
-        self.is_compiled = True
-
-
-    def fit(self, X=None, y=None,
-            validation_data=None,
-            validation_split=0.1,
-            train_batch_size=None,
-            valid_batch_size=None,
-            **kwargs):
-
-        if not self.is_compiled:
-            self.compile(**kwargs)
-
-        train_dataset, valid_dataset = self.prepare_data(X, y=y,
-                                                         validation_data=validation_data,
-                                                         validation_split=validation_split)
-        num_train_samples = len(train_dataset)
-        num_valid_samples = len(valid_dataset)
-
-        train_steps = num_train_samples//train_batch_size
-        valid_steps = num_valid_samples//valid_batch_size
-
-        if is_tf_available():
-            # Prepare dataset as a tf.train_data.Dataset instance
-            train_dataset = train_dataset.shuffle(128).batch(train_batch_size).repeat(-1)
-            valid_dataset = valid_dataset.batch(valid_batch_size)
-
-            logger.info('Training TF 2.0 model')
-            history = self.model.fit(train_dataset, epochs=2, steps_per_epoch=train_steps,
-                                     validation_data=valid_dataset, validation_steps=valid_steps,
-                                     **kwargs)
-        else:
-            raise NotImplementedError
-
-        self.is_trained = True
-
-
-    def fit_transform(self, *texts, **kwargs):
-        # Generic compatibility with sklearn and Keras
-        self.fit(*texts, **kwargs)
-        return self(*texts, **kwargs)
-
-
-    def transform(self, *texts, **kwargs):
-        # Generic compatibility with sklearn and Keras
-        return self(*texts, **kwargs)
-
-
-    def predict(self, *texts, **kwargs):
-        # Generic compatibility with sklearn and Keras
-        return self(*texts, **kwargs)
-
-
-    def __call__(self, *texts, **kwargs):
-        # Generic compatibility with sklearn and Keras
-        if 'X' in kwargs and not texts:
-            texts = kwargs.pop('X')
-
-        if not self.is_trained:
-            logger.error("Some weights of the model are not trained. Please fine-tune the model on a classification task before using it.")
-
-        inputs = self.tokenizer.batch_encode_plus(texts,
-                                                  add_special_tokens=True,
-                                                  return_tensors='tf' if is_tf_available() else 'pt')
-
-        if is_tf_available():
-            # TODO trace model
-            predictions = self.model(**inputs)[0]
-        else:
-            with torch.no_grad():
-                predictions = self.model(**inputs)[0]
-
-        return predictions.numpy().tolist()
diff --git a/transformers/pipelines.py b/transformers/pipelines.py
new file mode 100755
index 0000000000..e85f2300e3
--- /dev/null
+++ b/transformers/pipelines.py
@@ -0,0 +1,222 @@
+# coding=utf-8
+# Copyright 2018 The HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from __future__ import absolute_import, division, print_function, unicode_literals
+
+import os
+from abc import ABC, abstractmethod
+from typing import Union, Optional, Tuple
+
+import numpy as np
+
+from transformers import is_tf_available, logger, AutoTokenizer, PreTrainedTokenizer, is_torch_available
+
+if is_tf_available():
+    from transformers import TFAutoModelForSequenceClassification, TFAutoModelForQuestionAnswering
+else:
+    from transformers import AutoModelForSequenceClassification, AutoModelForQuestionAnswering
+
+
+class Pipeline(ABC):
+    def __init__(self, model, tokenizer: PreTrainedTokenizer = None, **kwargs):
+        self.model = model
+        self.tokenizer = tokenizer
+
+    @classmethod
+    @abstractmethod
+    def from_config(cls, model, tokenizer: PreTrainedTokenizer, **kwargs):
+        raise NotImplementedError()
+
+    def save_pretrained(self, save_directory):
+        if not os.path.isdir(save_directory):
+            logger.error("Provided path ({}) should be a directory".format(save_directory))
+            return
+
+        self.model.save_pretrained(save_directory)
+        self.tokenizer.save_pretrained(save_directory)
+
+    def transform(self, *texts, **kwargs):
+        # Generic compatibility with sklearn and Keras
+        return self(*texts, **kwargs)
+
+    def predict(self, *texts, **kwargs):
+        # Generic compatibility with sklearn and Keras
+        return self(*texts, **kwargs)
+
+    @abstractmethod
+    def __call__(self, *texts, **kwargs):
+        raise NotImplementedError()
+
+
+class TextClassificationPipeline(Pipeline):
+    def __init__(self, model, tokenizer: PreTrainedTokenizer, nb_classes: int = 2):
+        super().__init__(model, tokenizer)
+
+        if nb_classes < 2:
+            raise Exception('Invalid parameter nb_classes. int >= 2 is required (got: {})'.format(nb_classes))
+        self._nb_classes = nb_classes
+
+    @classmethod
+    def from_config(cls, model, tokenizer: PreTrainedTokenizer, **kwargs):
+        return cls(model, tokenizer, **kwargs)
+
+    def __call__(self, *texts, **kwargs):
+        # Generic compatibility with sklearn and Keras
+        if 'X' in kwargs and not texts:
+            texts = kwargs.pop('X')
+
+        inputs = self.tokenizer.batch_encode_plus(
+            texts, add_special_tokens=True, return_tensors='tf' if is_tf_available() else 'pt'
+        )
+
+        special_tokens_mask = inputs.pop('special_tokens_mask')
+
+        if is_tf_available():
+            # TODO trace model
+            predictions = self.model(**inputs)[0]
+        else:
+            import torch
+            with torch.no_grad():
+                predictions = self.model(**inputs)[0]
+
+        return predictions.numpy().tolist()
+
+
+class QuestionAnsweringPipeline(Pipeline):
+
+    @classmethod
+    def from_config(cls, model, tokenizer: PreTrainedTokenizer, **kwargs):
+        pass
+
+    def __call__(self, texts, **kwargs):
+        # Generic compatibility with sklearn and Keras
+        if 'X' in kwargs and not texts:
+            texts = kwargs.pop('X')
+
+        if not isinstance(texts, (tuple, list)):
+            raise Exception('QuestionAnsweringPipeline requires predict argument to be a tuple (context, question) or a List of tuple.')
+
+        if not isinstance(texts, list):
+            texts = [texts]
+
+        inputs = self.tokenizer.batch_encode_plus(
+            texts, add_special_tokens=True, return_tensors='tf' if is_tf_available() else 'pt'
+        )
+
+        # Remove special_tokens_mask to avoid KeyError
+        _ = inputs.pop('special_tokens_mask')
+
+        if is_tf_available():
+            # TODO trace model
+            start, end = self.model(inputs)
+        else:
+            import torch
+            with torch.no_grad():
+                # Retrieve the score for the context tokens only (removing question tokens)
+                start, end = self.model(**inputs)
+                start, end = start.cpu().numpy(), end.cpu().numpy()
+
+        answers = []
+        for i in range(len(texts)):
+            context_idx = inputs['token_type_ids'][i] == 1
+            start_, end_ = start[i, context_idx], end[i, context_idx]
+
+            # Normalize logits and spans to retrieve the answer
+            start_, end_ = self.decode(start_, end_)
+
+            # Convert the answer (tokens) back to the original text
+            answers += [{
+                'start': start_,
+                'end': end_,
+                'answer': self.span_to_answer(texts[i][1], start_, end_)
+            }]
+
+        return answers
+
+    def decode(self, start: np.ndarray, end: np.ndarray) -> Tuple:
+        # Ensure we have batch axis
+        if start.ndim == 1:
+            start = start[None]
+
+        if end.ndim == 1:
+            end = end[None]
+
+        # Compute the score of each tuple(start, end) to be the real answer
+        outer = np.matmul(np.expand_dims(start, -1), np.expand_dims(end, 1))
+
+        # Remove candidate with end < start and end - start > 15
+        candidates = np.tril(np.triu(outer), 15)
+
+        start = np.max(candidates, axis=2).argmax(-1)
+        end = np.max(candidates, axis=1).argmax(-1)
+
+        return start, end
+
+    def span_to_answer(self, text: str, start: int, end: int):
+        words, token_idx = [], 0
+
+        for i, word in enumerate(text.split(" ")):
+            token = self.tokenizer.tokenize(word)
+
+            # Append words if they are in the span
+            if start <= token_idx <= end:
+                words += [word]
+
+            # Stop if we went over the end of the answer
+            if token_idx > end:
+                break
+
+            # Append the subtokenization length to the running index
+            token_idx += len(token)
+
+        # Join text with spaces
+        return ' '.join(words)
+
+
+# Register all the supported task here
+SUPPORTED_TASKS = {
+    'text-classification': {
+        'impl': TextClassificationPipeline,
+        'tf': TFAutoModelForSequenceClassification if is_tf_available() else None,
+        'pt': AutoModelForSequenceClassification if is_torch_available() else None
+    },
+    'question-answering': {
+        'impl': QuestionAnsweringPipeline,
+        'tf': TFAutoModelForQuestionAnswering if is_tf_available() else None,
+        'pt': AutoModelForQuestionAnswering if is_torch_available() else None
+    }
+}
+
+
+def pipeline(task: str, model, tokenizer: Optional[Union[str, PreTrainedTokenizer]] = None, **kwargs) -> Pipeline:
+    """
+    Utility factory method to build pipeline.
+    """
+    # Try to infer tokenizer from model name (if provided as str)
+    if tokenizer is None and isinstance(model, str):
+        tokenizer = model
+    else:
+        # Impossible to guest what is the right tokenizer here
+        raise Exception('Tokenizer cannot be None if provided model is a PreTrainedModel instance')
+
+    tokenizer = tokenizer if isinstance(tokenizer, PreTrainedTokenizer) else AutoTokenizer.from_pretrained(tokenizer)
+
+    if task not in SUPPORTED_TASKS:
+        raise KeyError("Unknown task {}, available tasks are {}".format(task, list(SUPPORTED_TASKS.keys())))
+
+    targeted_task = SUPPORTED_TASKS[task]
+    task, allocator = targeted_task['impl'], targeted_task['tf'] if is_tf_available() else targeted_task['pt']
+
+    model = allocator.from_pretrained(model)
+    return task(model, tokenizer, **kwargs)

From 02110485b0980c2b0c8c4dc070643eff9c289cff Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Fri, 6 Dec 2019 18:11:27 +0100
Subject: [PATCH 292/505] Added batching, topk, chars index and scores.

---
 transformers/pipelines.py | 114 +++++++++++++++++++++++++++++---------
 1 file changed, 88 insertions(+), 26 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index e85f2300e3..f3b70908dd 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -16,7 +16,7 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 
 import os
 from abc import ABC, abstractmethod
-from typing import Union, Optional, Tuple
+from typing import Union, Optional, Tuple, List, Dict
 
 import numpy as np
 
@@ -24,7 +24,8 @@ from transformers import is_tf_available, logger, AutoTokenizer, PreTrainedToken
 
 if is_tf_available():
     from transformers import TFAutoModelForSequenceClassification, TFAutoModelForQuestionAnswering
-else:
+
+if is_torch_available():
     from transformers import AutoModelForSequenceClassification, AutoModelForQuestionAnswering
 
 
@@ -94,30 +95,71 @@ class TextClassificationPipeline(Pipeline):
 
 
 class QuestionAnsweringPipeline(Pipeline):
+    """
+    Question Answering pipeling involving Tokenization and Inference.
+    TODO:
+     - top-k answers
+     - return start/end chars
+     - return score
+    """
+
+    def __init__(self, model, tokenizer: Optional[PreTrainedTokenizer]):
+        super().__init__(model, tokenizer)
+
+    @staticmethod
+    def create_sample(question: Union[str, List[str]], context: Union[str, List[str]]) -> Union[dict, List[Dict]]:
+        is_list = isinstance(question, list)
+
+        if is_list:
+            return [{'question': q, 'context': c} for q, c in zip(question, context)]
+        else:
+            return {'question': question, 'context': context}
 
     @classmethod
     def from_config(cls, model, tokenizer: PreTrainedTokenizer, **kwargs):
         pass
 
-    def __call__(self, texts, **kwargs):
-        # Generic compatibility with sklearn and Keras
-        if 'X' in kwargs and not texts:
-            texts = kwargs.pop('X')
+    def __call__(self, *texts, **kwargs):
+        # Set defaults values
+        kwargs.setdefault('max_answer_len', 15)
+        kwargs.setdefault('topk', 1)
 
-        if not isinstance(texts, (tuple, list)):
-            raise Exception('QuestionAnsweringPipeline requires predict argument to be a tuple (context, question) or a List of tuple.')
+        if kwargs['topk'] < 1:
+            raise ValueError('topk parameter should be >= 1 (got {})'.format(kwargs['topk']))
+
+        if kwargs['max_answer_len'] < 1:
+            raise ValueError('max_answer_len parameter should be >= 1 (got {})'.format(kwargs['max_answer_len']))
+
+        # Tabular input
+        if 'question' in kwargs and 'context' in kwargs:
+            texts = QuestionAnsweringPipeline.create_sample(kwargs['questions'], kwargs['contexts'])
+        elif 'data' in kwargs:
+            texts = kwargs['data']
+        # Generic compatibility with sklearn and Keras
+        elif 'X' in kwargs and not texts:
+            texts = kwargs.pop('X')
+        else:
+            (texts, ) = texts
+
+        if not isinstance(texts, (dict, list)):
+            raise Exception('QuestionAnsweringPipeline requires predict argument to be a tuple (context, question) or a List of dict.')
 
         if not isinstance(texts, list):
             texts = [texts]
 
+        # Map to tuple (question, context)
+        texts = [(text['question'], text['context']) for text in texts]
+
         inputs = self.tokenizer.batch_encode_plus(
-            texts, add_special_tokens=True, return_tensors='tf' if is_tf_available() else 'pt'
+            # texts, add_special_tokens=True, return_tensors='tf' if is_tf_available() else 'pt'
+            texts, add_special_tokens=True, return_tensors='pt'
         )
 
         # Remove special_tokens_mask to avoid KeyError
         _ = inputs.pop('special_tokens_mask')
 
-        if is_tf_available():
+        # if is_tf_available():
+        if False:
             # TODO trace model
             start, end = self.model(inputs)
         else:
@@ -133,18 +175,19 @@ class QuestionAnsweringPipeline(Pipeline):
             start_, end_ = start[i, context_idx], end[i, context_idx]
 
             # Normalize logits and spans to retrieve the answer
-            start_, end_ = self.decode(start_, end_)
+            start_ = np.exp(start_) / np.sum(np.exp(start_))
+            end_ = np.exp(end_) / np.sum(np.exp(end_))
+            starts, ends, scores = self.decode(start_, end_, kwargs['topk'], kwargs['max_answer_len'])
 
             # Convert the answer (tokens) back to the original text
-            answers += [{
-                'start': start_,
-                'end': end_,
-                'answer': self.span_to_answer(texts[i][1], start_, end_)
-            }]
+            answers += [[
+                {**{'score': score}, **self.span_to_answer(texts[i][1], s, e)}
+                for s, e, score in zip(starts, ends, scores)
+            ]]
 
         return answers
 
-    def decode(self, start: np.ndarray, end: np.ndarray) -> Tuple:
+    def decode(self, start: np.ndarray, end: np.ndarray, topk: int, max_answer_len: int) -> Tuple:
         # Ensure we have batch axis
         if start.ndim == 1:
             start = start[None]
@@ -155,22 +198,39 @@ class QuestionAnsweringPipeline(Pipeline):
         # Compute the score of each tuple(start, end) to be the real answer
         outer = np.matmul(np.expand_dims(start, -1), np.expand_dims(end, 1))
 
-        # Remove candidate with end < start and end - start > 15
-        candidates = np.tril(np.triu(outer), 15)
+        # Remove candidate with end < start and end - start > max_answer_len
+        candidates = np.tril(np.triu(outer), max_answer_len - 1)
 
-        start = np.max(candidates, axis=2).argmax(-1)
-        end = np.max(candidates, axis=1).argmax(-1)
+        # start = np.max(candidates, axis=2).argmax(-1)
+        # end = np.max(candidates, axis=1).argmax(-1)
 
-        return start, end
+        scores_flat = candidates.flatten()
+        if topk == 1:
+            idx_sort = [np.argmax(scores_flat)]
+        elif len(scores_flat) < topk:
+            idx_sort = np.argsort(-scores_flat)
+        else:
+            idx = np.argpartition(-scores_flat, topk)[0:topk]
+            idx_sort = idx[np.argsort(-scores_flat[idx])]
+
+        start, end = np.unravel_index(idx_sort, candidates.shape)[1:]
+        return start, end, candidates[0, start, end]
 
     def span_to_answer(self, text: str, start: int, end: int):
-        words, token_idx = [], 0
+        words = []
+        token_idx = char_start_idx = char_end_idx = chars_idx = 0
 
         for i, word in enumerate(text.split(" ")):
             token = self.tokenizer.tokenize(word)
 
             # Append words if they are in the span
             if start <= token_idx <= end:
+                if token_idx == start:
+                    char_start_idx = chars_idx
+
+                if token_idx == end:
+                    char_end_idx = chars_idx + len(word)
+
                 words += [word]
 
             # Stop if we went over the end of the answer
@@ -179,9 +239,10 @@ class QuestionAnsweringPipeline(Pipeline):
 
             # Append the subtokenization length to the running index
             token_idx += len(token)
+            chars_idx += len(word) + 1
 
         # Join text with spaces
-        return ' '.join(words)
+        return {'answer': ' '.join(words), 'start': max(0, char_start_idx), 'end': min(len(text), char_end_idx)}
 
 
 # Register all the supported task here
@@ -193,7 +254,7 @@ SUPPORTED_TASKS = {
     },
     'question-answering': {
         'impl': QuestionAnsweringPipeline,
-        'tf': TFAutoModelForQuestionAnswering if is_tf_available() else None,
+        # 'tf': TFAutoModelForQuestionAnswering if is_tf_available() else None,
         'pt': AutoModelForQuestionAnswering if is_torch_available() else None
     }
 }
@@ -216,7 +277,8 @@ def pipeline(task: str, model, tokenizer: Optional[Union[str, PreTrainedTokenize
         raise KeyError("Unknown task {}, available tasks are {}".format(task, list(SUPPORTED_TASKS.keys())))
 
     targeted_task = SUPPORTED_TASKS[task]
-    task, allocator = targeted_task['impl'], targeted_task['tf'] if is_tf_available() else targeted_task['pt']
+    # task, allocator = targeted_task['impl'], targeted_task['tf'] if is_tf_available() else targeted_task['pt']
+    task, allocator = targeted_task['impl'], targeted_task['pt']
 
     model = allocator.from_pretrained(model)
     return task(model, tokenizer, **kwargs)

From 6e61e06051160812d401c07e5a4c77321191ec1e Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Mon, 9 Dec 2019 11:13:27 +0100
Subject: [PATCH 293/505] batch_encode_plus generates the
 encoder_attention_mask to avoid attending over padded values.

---
 transformers/pipelines.py | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index f3b70908dd..8b329abd24 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -132,7 +132,7 @@ class QuestionAnsweringPipeline(Pipeline):
 
         # Tabular input
         if 'question' in kwargs and 'context' in kwargs:
-            texts = QuestionAnsweringPipeline.create_sample(kwargs['questions'], kwargs['contexts'])
+            texts = QuestionAnsweringPipeline.create_sample(kwargs['question'], kwargs['context'])
         elif 'data' in kwargs:
             texts = kwargs['data']
         # Generic compatibility with sklearn and Keras
@@ -156,7 +156,10 @@ class QuestionAnsweringPipeline(Pipeline):
         )
 
         # Remove special_tokens_mask to avoid KeyError
-        _ = inputs.pop('special_tokens_mask')
+        special_tokens_mask, input_len = inputs.pop('special_tokens_mask'), inputs.pop('input_len')
+
+        # TODO : Harmonize model arguments across all model
+        inputs['attention_mask'] = inputs.pop('encoder_attention_mask')
 
         # if is_tf_available():
         if False:

From f116cf599cd979636bdf37df31be62088a1cb7e0 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Mon, 9 Dec 2019 11:32:49 +0100
Subject: [PATCH 294/505] Allow hidding frameworks through environment
 variables (NO_TF, NO_TORCH).

---
 transformers/file_utils.py | 23 ++++++++++++++++-------
 1 file changed, 16 insertions(+), 7 deletions(-)

diff --git a/transformers/file_utils.py b/transformers/file_utils.py
index 24abd60781..4784681fb4 100644
--- a/transformers/file_utils.py
+++ b/transformers/file_utils.py
@@ -27,17 +27,25 @@ from contextlib import contextmanager
 logger = logging.getLogger(__name__)  # pylint: disable=invalid-name
 
 try:
-    import tensorflow as tf
-    assert hasattr(tf, '__version__') and int(tf.__version__[0]) >= 2
-    _tf_available = True  # pylint: disable=invalid-name
-    logger.info("TensorFlow version {} available.".format(tf.__version__))
+    if 'NO_TF' in os.environ and os.environ['NO_TF'].upper() in ('1', 'ON'):
+        logger.info("Found NO_TF, disabling TensorFlow")
+        _tf_available = False
+    else:
+        import tensorflow as tf
+        assert hasattr(tf, '__version__') and int(tf.__version__[0]) >= 2
+        _tf_available = True  # pylint: disable=invalid-name
+        logger.info("TensorFlow version {} available.".format(tf.__version__))
 except (ImportError, AssertionError):
     _tf_available = False  # pylint: disable=invalid-name
 
 try:
-    import torch
-    _torch_available = True  # pylint: disable=invalid-name
-    logger.info("PyTorch version {} available.".format(torch.__version__))
+    if 'NO_TORCH' in os.environ and os.environ['NO_TORCH'].upper() in ('1', 'ON'):
+        logger.info("Found NO_TORCH, disabling PyTorch")
+        _torch_available = False
+    else:
+        import torch
+        _torch_available = True  # pylint: disable=invalid-name
+        logger.info("PyTorch version {} available.".format(torch.__version__))
 except ImportError:
     _torch_available = False  # pylint: disable=invalid-name
 
@@ -77,6 +85,7 @@ def is_torch_available():
     return _torch_available
 
 def is_tf_available():
+
     return _tf_available
 
 if not six.PY2:

From c2407fdd88719eed66227815188b5908eca4b3a7 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Mon, 9 Dec 2019 11:47:52 +0100
Subject: [PATCH 295/505] Enable the Tensorflow backend.

---
 transformers/pipelines.py | 16 ++++++----------
 1 file changed, 6 insertions(+), 10 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 8b329abd24..e484958dcc 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -151,8 +151,7 @@ class QuestionAnsweringPipeline(Pipeline):
         texts = [(text['question'], text['context']) for text in texts]
 
         inputs = self.tokenizer.batch_encode_plus(
-            # texts, add_special_tokens=True, return_tensors='tf' if is_tf_available() else 'pt'
-            texts, add_special_tokens=True, return_tensors='pt'
+            texts, add_special_tokens=True, return_tensors='tf' if is_tf_available() else 'pt'
         )
 
         # Remove special_tokens_mask to avoid KeyError
@@ -161,10 +160,10 @@ class QuestionAnsweringPipeline(Pipeline):
         # TODO : Harmonize model arguments across all model
         inputs['attention_mask'] = inputs.pop('encoder_attention_mask')
 
-        # if is_tf_available():
-        if False:
+        if is_tf_available():
             # TODO trace model
             start, end = self.model(inputs)
+            start, end = start.numpy(), end.numpy()
         else:
             import torch
             with torch.no_grad():
@@ -204,9 +203,7 @@ class QuestionAnsweringPipeline(Pipeline):
         # Remove candidate with end < start and end - start > max_answer_len
         candidates = np.tril(np.triu(outer), max_answer_len - 1)
 
-        # start = np.max(candidates, axis=2).argmax(-1)
-        # end = np.max(candidates, axis=1).argmax(-1)
-
+        #  Inspired by Chen & al. (https://github.com/facebookresearch/DrQA)
         scores_flat = candidates.flatten()
         if topk == 1:
             idx_sort = [np.argmax(scores_flat)]
@@ -257,7 +254,7 @@ SUPPORTED_TASKS = {
     },
     'question-answering': {
         'impl': QuestionAnsweringPipeline,
-        # 'tf': TFAutoModelForQuestionAnswering if is_tf_available() else None,
+        'tf': TFAutoModelForQuestionAnswering if is_tf_available() else None,
         'pt': AutoModelForQuestionAnswering if is_torch_available() else None
     }
 }
@@ -280,8 +277,7 @@ def pipeline(task: str, model, tokenizer: Optional[Union[str, PreTrainedTokenize
         raise KeyError("Unknown task {}, available tasks are {}".format(task, list(SUPPORTED_TASKS.keys())))
 
     targeted_task = SUPPORTED_TASKS[task]
-    # task, allocator = targeted_task['impl'], targeted_task['tf'] if is_tf_available() else targeted_task['pt']
-    task, allocator = targeted_task['impl'], targeted_task['pt']
+    task, allocator = targeted_task['impl'], targeted_task['tf'] if is_tf_available() else targeted_task['pt']
 
     model = allocator.from_pretrained(model)
     return task(model, tokenizer, **kwargs)

From 348e19aa2104d59b91bc7216da5fcabf04f0bc5d Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Mon, 9 Dec 2019 12:10:26 +0100
Subject: [PATCH 296/505] Expose attention_masks and input_lengths arguments to
 batch_encode_plus

---
 transformers/pipelines.py | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index e484958dcc..46fb735a70 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -149,14 +149,11 @@ class QuestionAnsweringPipeline(Pipeline):
 
         # Map to tuple (question, context)
         texts = [(text['question'], text['context']) for text in texts]
-
         inputs = self.tokenizer.batch_encode_plus(
-            texts, add_special_tokens=True, return_tensors='tf' if is_tf_available() else 'pt'
+            texts, add_special_tokens=False, return_tensors='tf' if is_tf_available() else 'pt',
+            return_attention_masks=True, return_input_lengths=False
         )
 
-        # Remove special_tokens_mask to avoid KeyError
-        special_tokens_mask, input_len = inputs.pop('special_tokens_mask'), inputs.pop('input_len')
-
         # TODO : Harmonize model arguments across all model
         inputs['attention_mask'] = inputs.pop('encoder_attention_mask')
 

From fe0f552e00e7556c9dd6eccc2486b962bb2a3460 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Mon, 9 Dec 2019 14:13:17 +0100
Subject: [PATCH 297/505] Use attention_mask everywhere.

---
 transformers/pipelines.py | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 46fb735a70..57fe2f1357 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -154,9 +154,6 @@ class QuestionAnsweringPipeline(Pipeline):
             return_attention_masks=True, return_input_lengths=False
         )
 
-        # TODO : Harmonize model arguments across all model
-        inputs['attention_mask'] = inputs.pop('encoder_attention_mask')
-
         if is_tf_available():
             # TODO trace model
             start, end = self.model(inputs)

From a7d3794a298d77a1ae0c75c84ca963ac78058243 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Mon, 9 Dec 2019 18:34:58 +0100
Subject: [PATCH 298/505] Remove token_type_ids for compatibility with
 DistilBert

---
 transformers/pipelines.py | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 57fe2f1357..1701915203 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -20,7 +20,7 @@ from typing import Union, Optional, Tuple, List, Dict
 
 import numpy as np
 
-from transformers import is_tf_available, logger, AutoTokenizer, PreTrainedTokenizer, is_torch_available
+from transformers import is_tf_available, is_torch_available, logger, AutoTokenizer, PreTrainedTokenizer
 
 if is_tf_available():
     from transformers import TFAutoModelForSequenceClassification, TFAutoModelForQuestionAnswering
@@ -154,6 +154,8 @@ class QuestionAnsweringPipeline(Pipeline):
             return_attention_masks=True, return_input_lengths=False
         )
 
+        token_type_ids = inputs.pop('token_type_ids')
+
         if is_tf_available():
             # TODO trace model
             start, end = self.model(inputs)
@@ -167,7 +169,7 @@ class QuestionAnsweringPipeline(Pipeline):
 
         answers = []
         for i in range(len(texts)):
-            context_idx = inputs['token_type_ids'][i] == 1
+            context_idx = token_type_ids[i] == 1
             start_, end_ = start[i, context_idx], end[i, context_idx]
 
             # Normalize logits and spans to retrieve the answer

From aae74065dff94465cdf6d92ccfd5dee030268885 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Mon, 9 Dec 2019 18:35:26 +0100
Subject: [PATCH 299/505] Added QuestionAnsweringPipeline unit tests.

---
 transformers/tests/pipelines_test.py | 83 ++++++++++++++++++++++++++++
 1 file changed, 83 insertions(+)
 create mode 100644 transformers/tests/pipelines_test.py

diff --git a/transformers/tests/pipelines_test.py b/transformers/tests/pipelines_test.py
new file mode 100644
index 0000000000..36d6e013c3
--- /dev/null
+++ b/transformers/tests/pipelines_test.py
@@ -0,0 +1,83 @@
+import unittest
+from unittest.mock import patch
+
+
+QA_FINETUNED_MODELS = {
+    'bert-large-uncased-whole-word-masking-finetuned-squad',
+    'bert-large-cased-whole-word-masking-finetuned-squad',
+    'distilbert-base-uncased-distilled-squad',
+
+}
+
+
+class QuestionAnsweringPipelineTest(unittest.TestCase):
+    def check_answer_structure(self, answer, batch, topk):
+        self.assertIsInstance(answer, list)
+        self.assertEqual(len(answer), batch)
+        self.assertIsInstance(answer[0], list)
+        self.assertEqual(len(answer[0]), topk)
+        self.assertIsInstance(answer[0][0], dict)
+
+        for item in answer[0]:
+            self.assertTrue('start' in item)
+            self.assertTrue('end' in item)
+            self.assertTrue('score' in item)
+            self.assertTrue('answer' in item)
+
+    def question_answering_pipeline(self, nlp):
+        # Simple case with topk = 1, no batching
+        a = nlp(question='What is the name of the company I\'m working for ?', context='I\'m working for Huggingface.')
+        self.check_answer_structure(a, 1, 1)
+
+        # Simple case with topk = 2, no batching
+        a = nlp(question='What is the name of the company I\'m working for ?', context='I\'m working for Huggingface.', topk=2)
+        self.check_answer_structure(a, 1, 2)
+
+        # Batch case with topk = 1
+        a = nlp(question=['What is the name of the company I\'m working for ?', 'Where is the company based ?'],
+                context=['I\'m working for Huggingface.', 'The company is based in New York and Paris'])
+        self.check_answer_structure(a, 2, 1)
+
+        # Batch case with topk = 2
+        a = nlp(question=['What is the name of the company I\'m working for ?', 'Where is the company based ?'],
+                context=['I\'m working for Huggingface.', 'The company is based in New York and Paris'], topk=2)
+        self.check_answer_structure(a, 2, 2)
+
+    @patch('transformers.pipelines.is_torch_available', return_value=False)
+    def test_tf_models(self, is_torch_available):
+        from transformers import pipeline
+        for model in QA_FINETUNED_MODELS:
+            self.question_answering_pipeline(pipeline('question-answering', model))
+
+    @patch('transformers.pipelines.is_tf_available', return_value=False)
+    @patch('transformers.tokenization_utils.is_tf_available', return_value=False)
+    def test_torch_models(self, is_tf_available, _):
+        from transformers import pipeline
+        for model in QA_FINETUNED_MODELS:
+            self.question_answering_pipeline(pipeline('question-answering', model))
+
+
+class AutoPipelineTest(unittest.TestCase):
+    @patch('transformers.pipelines.is_torch_available', return_value=False)
+    def test_tf_qa(self, is_torch_available):
+        from transformers import pipeline
+        from transformers.pipelines import QuestionAnsweringPipeline
+        from transformers.modeling_tf_utils import TFPreTrainedModel
+        for model in QA_FINETUNED_MODELS:
+            nlp = pipeline('question-answering', model)
+            self.assertIsInstance(nlp, QuestionAnsweringPipeline)
+            self.assertIsInstance(nlp.model, TFPreTrainedModel)
+
+    @patch('transformers.pipelines.is_tf_available', return_value=False)
+    def test_torch_qa(self, is_tf_available):
+        from transformers import pipeline
+        from transformers.pipelines import QuestionAnsweringPipeline
+        from transformers.modeling_utils import PreTrainedModel
+        for model in QA_FINETUNED_MODELS:
+            nlp = pipeline('question-answering', model)
+            self.assertIsInstance(nlp, QuestionAnsweringPipeline)
+            self.assertIsInstance(nlp.model, PreTrainedModel)
+
+
+if __name__ == '__main__':
+    unittest.main()

From 8ae1044f80ef543e4657c97d1030649d4da15aa8 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Tue, 10 Dec 2019 15:11:07 +0100
Subject: [PATCH 300/505] updating tests and TF 2.0 model

---
 transformers/modeling_t5.py                   | 31 ++++++---
 transformers/modeling_tf_t5.py                | 44 ++++++++++---
 transformers/tests/modeling_common_test.py    | 18 +++--
 transformers/tests/modeling_t5_test.py        |  9 ++-
 transformers/tests/modeling_tf_common_test.py | 65 +++++++++++--------
 transformers/tests/modeling_tf_t5_test.py     | 10 +--
 transformers/tests/tokenization_t5_test.py    |  1 -
 7 files changed, 121 insertions(+), 57 deletions(-)

diff --git a/transformers/modeling_t5.py b/transformers/modeling_t5.py
index e48293b49e..f1e4e0306c 100644
--- a/transformers/modeling_t5.py
+++ b/transformers/modeling_t5.py
@@ -726,8 +726,11 @@ class T5Model(T5PreTrainedModel):
         encoder_hidden_states = kwargs_encoder.pop("hidden_states", None)
         encoder_attention_mask = kwargs_encoder.get("attention_mask", None)
         if encoder_hidden_states is None:
-            encoder_inputs_ids = kwargs_encoder.pop("input_ids")
-            hidden_states = self.shared(encoder_inputs_ids)  # Convert inputs in embeddings
+            # Convert encoder inputs in embeddings if needed
+            hidden_states = kwargs_encoder.pop("inputs_embeds", None)
+            if hidden_states is None:
+                encoder_inputs_ids = kwargs_encoder.pop("input_ids")
+                hidden_states = self.shared(encoder_inputs_ids)  # Convert inputs in embeddings
 
             if encoder_attention_mask is not None:
                 # Apply masking
@@ -740,8 +743,12 @@ class T5Model(T5PreTrainedModel):
             encoder_outputs = ()
 
         # Decode
-        decoder_inputs_ids = kwargs_decoder.pop("input_ids")
-        hidden_states = self.shared(decoder_inputs_ids)  # Convert inputs in embeddings
+        # Convert decoder inputs in embeddings if needed
+        hidden_states = kwargs_decoder.pop("inputs_embeds", None)
+        if hidden_states is None:
+            decoder_inputs_ids = kwargs_decoder.pop("input_ids")
+            hidden_states = self.shared(decoder_inputs_ids)
+
         kwargs_decoder["encoder_hidden_states"] = encoder_hidden_states
         kwargs_decoder["encoder_attention_mask"] = encoder_attention_mask
         decoder_outputs = self.decoder(hidden_states, **kwargs_decoder)
@@ -825,16 +832,24 @@ class T5WithLMHeadModel(T5PreTrainedModel):
         # Encode if needed (training, first prediction pass)
         encoder_hidden_states = kwargs_encoder.pop("hidden_states", None)
         if encoder_hidden_states is None:
-            encoder_inputs_ids = kwargs_encoder.pop("input_ids")
-            hidden_states = self.shared(encoder_inputs_ids)  # Convert inputs in embeddings
+            # Convert encoder inputs in embeddings if needed
+            hidden_states = kwargs_encoder.pop("inputs_embeds", None)
+            if hidden_states is None:
+                encoder_inputs_ids = kwargs_encoder.pop("input_ids")
+                hidden_states = self.shared(encoder_inputs_ids)  # Convert inputs in embeddings
+
             encoder_outputs = self.encoder(hidden_states, **kwargs_encoder)
             encoder_hidden_states = encoder_outputs[0]
         else:
             encoder_outputs = ()
 
         # Decode
-        decoder_inputs_ids = kwargs_decoder.pop("input_ids")
-        hidden_states = self.shared(decoder_inputs_ids)  # Convert inputs in embeddings
+        # Convert decoder inputs in embeddings if needed
+        hidden_states = kwargs_decoder.pop("inputs_embeds", None)
+        if hidden_states is None:
+            decoder_inputs_ids = kwargs_decoder.pop("input_ids")
+            hidden_states = self.shared(decoder_inputs_ids)
+
         kwargs_decoder["encoder_hidden_states"] = encoder_hidden_states
         kwargs_decoder["encoder_attention_mask"] = kwargs_encoder.get("attention_mask", None)
         decoder_outputs = self.decoder(hidden_states, **kwargs_decoder)
diff --git a/transformers/modeling_tf_t5.py b/transformers/modeling_tf_t5.py
index 11762ee1e5..447fd69b05 100644
--- a/transformers/modeling_tf_t5.py
+++ b/transformers/modeling_tf_t5.py
@@ -613,6 +613,12 @@ class TFT5Model(TFT5PreTrainedModel):
         decoder_config.is_decoder = True
         self.decoder = TFT5MainLayer(decoder_config, name='decoder')
 
+    def get_input_embeddings(self):
+        return self.shared
+
+    def get_output_embeddings(self):
+        return self.shared
+
     def call(self, decoder_input_ids, **kwargs):
         # We allow two types of multi-inputs:
         # - traditional keyword arguments in the call method
@@ -634,16 +640,24 @@ class TFT5Model(TFT5PreTrainedModel):
         # Encode if needed (training, first prediction pass)
         encoder_hidden_states = kwargs_encoder.pop("hidden_states", None)
         if encoder_hidden_states is None:
-            encoder_inputs_ids = kwargs_encoder.pop("input_ids")
-            hidden_states = self.shared(encoder_inputs_ids)  # Convert inputs in embeddings
+            # Convert encoder inputs in embeddings if needed
+            hidden_states = kwargs_encoder.pop("inputs_embeds", None)
+            if hidden_states is None:
+                encoder_inputs_ids = kwargs_encoder.pop("input_ids")
+                hidden_states = self.shared(encoder_inputs_ids)  # Convert inputs in embeddings
+
             encoder_outputs = self.encoder(hidden_states, **kwargs_encoder)
             encoder_hidden_states = encoder_outputs[0]
         else:
             encoder_outputs = ()
 
         # Decode
-        decoder_inputs_ids = kwargs_decoder.pop("input_ids")
-        hidden_states = self.shared(decoder_inputs_ids)  # Convert inputs in embeddings
+        # Convert decoder inputs in embeddings if needed
+        hidden_states = kwargs_decoder.pop("inputs_embeds", None)
+        if hidden_states is None:
+            decoder_inputs_ids = kwargs_decoder.pop("input_ids")
+            hidden_states = self.shared(decoder_inputs_ids)
+
         kwargs_decoder["encoder_hidden_states"] = encoder_hidden_states
         kwargs_decoder["encoder_attention_mask"] = kwargs_encoder.get("attention_mask", None)
         decoder_outputs = self.decoder(hidden_states, **kwargs_decoder)
@@ -692,6 +706,12 @@ class TFT5WithLMHeadModel(TFT5PreTrainedModel):
         decoder_config.is_decoder = True
         self.decoder = TFT5MainLayer(decoder_config, name='decoder')
 
+    def get_input_embeddings(self):
+        return self.shared
+
+    def get_output_embeddings(self):
+        return self.shared
+
     def call(self, decoder_input_ids, **kwargs):
         # We allow two types of multi-inputs:
         # - traditional keyword arguments in the call method
@@ -713,16 +733,24 @@ class TFT5WithLMHeadModel(TFT5PreTrainedModel):
         # Encode if needed (training, first prediction pass)
         encoder_hidden_states = kwargs_encoder.pop("hidden_states", None)
         if encoder_hidden_states is None:
-            encoder_inputs_ids = kwargs_encoder.pop("input_ids")
-            hidden_states = self.shared(encoder_inputs_ids)  # Convert inputs in embeddings
+            # Convert encoder inputs in embeddings if needed
+            hidden_states = kwargs_encoder.pop("inputs_embeds", None)
+            if hidden_states is None:
+                encoder_inputs_ids = kwargs_encoder.pop("input_ids")
+                hidden_states = self.shared(encoder_inputs_ids)  # Convert inputs in embeddings
+
             encoder_outputs = self.encoder(hidden_states, **kwargs_encoder)
             encoder_hidden_states = encoder_outputs[0]
         else:
             encoder_outputs = ()
 
         # Decode
-        decoder_inputs_ids = kwargs_decoder.pop("input_ids")
-        hidden_states = self.shared(decoder_inputs_ids)  # Convert inputs in embeddings
+        # Convert decoder inputs in embeddings if needed
+        hidden_states = kwargs_decoder.pop("inputs_embeds", None)
+        if hidden_states is None:
+            decoder_inputs_ids = kwargs_decoder.pop("input_ids")
+            hidden_states = self.shared(decoder_inputs_ids)
+
         kwargs_decoder["encoder_hidden_states"] = encoder_hidden_states
         kwargs_decoder["encoder_attention_mask"] = kwargs_encoder.get("attention_mask", None)
         decoder_outputs = self.decoder(hidden_states, **kwargs_decoder)
diff --git a/transformers/tests/modeling_common_test.py b/transformers/tests/modeling_common_test.py
index cdfbfc09e2..792f5cee3e 100644
--- a/transformers/tests/modeling_common_test.py
+++ b/transformers/tests/modeling_common_test.py
@@ -568,8 +568,14 @@ class CommonTestCases:
 
         def test_inputs_embeds(self):
             config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
-            input_ids = inputs_dict["input_ids"]
-            del inputs_dict["input_ids"]
+            if not self.is_encoder_decoder:
+                input_ids = inputs_dict["input_ids"]
+                del inputs_dict["input_ids"]
+            else:
+                encoder_input_ids = inputs_dict["encoder_input_ids"]
+                decoder_input_ids = inputs_dict["decoder_input_ids"]
+                del inputs_dict["encoder_input_ids"]
+                del inputs_dict["decoder_input_ids"]
 
             for model_class in self.all_model_classes:
                 model = model_class(config)
@@ -577,9 +583,13 @@ class CommonTestCases:
                 model.eval()
 
                 wte = model.get_input_embeddings()
-                inputs_dict["inputs_embeds"] = wte(input_ids)
-                outputs = model(**inputs_dict)
+                if not self.is_encoder_decoder:
+                    inputs_dict["inputs_embeds"] = wte(input_ids)
+                else:
+                    inputs_dict["encoder_inputs_embeds"] = wte(encoder_input_ids)
+                    inputs_dict["decoder_inputs_embeds"] = wte(decoder_input_ids)
 
+                outputs = model(**inputs_dict)
 
     class GPTModelTester(CommonModelTester):
 
diff --git a/transformers/tests/modeling_t5_test.py b/transformers/tests/modeling_t5_test.py
index 091bd742b5..a539cc868a 100644
--- a/transformers/tests/modeling_t5_test.py
+++ b/transformers/tests/modeling_t5_test.py
@@ -18,20 +18,19 @@ from __future__ import print_function
 
 import unittest
 import shutil
-import pytest
 
 from transformers import is_torch_available
 
-from .modeling_common_test import (CommonTestCases, ids_tensor)
+from .modeling_common_test import (CommonTestCases, ids_tensor, floats_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_torch, slow, torch_device
 
 if is_torch_available():
     from transformers import (T5Config, T5Model, T5WithLMHeadModel)
     from transformers.modeling_t5 import T5_PRETRAINED_MODEL_ARCHIVE_MAP
-else:
-    pytestmark = pytest.mark.skip("Require Torch")
 
 
+@require_torch
 class T5ModelTest(CommonTestCases.CommonModelTester):
 
     all_model_classes = (T5Model, T5WithLMHeadModel) if is_torch_available() else ()
@@ -174,7 +173,7 @@ class T5ModelTest(CommonTestCases.CommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_t5_with_lm_head(*config_and_inputs)
 
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
         for model_name in list(T5_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
diff --git a/transformers/tests/modeling_tf_common_test.py b/transformers/tests/modeling_tf_common_test.py
index 8957313021..a0d63583fb 100644
--- a/transformers/tests/modeling_tf_common_test.py
+++ b/transformers/tests/modeling_tf_common_test.py
@@ -130,12 +130,12 @@ class TFCommonTestCases:
                                       for name, key in inputs_dict.items())
                 with torch.no_grad():
                     pto = pt_model(**pt_inputs_dict)
-                tfo = tf_model(inputs_dict)
-                tfo = tfo[0].numpy()
-                pto = pto[0].numpy()
-                tfo[np.isnan(tfo)] = 0
-                pto[np.isnan(pto)] = 0
-                max_diff = np.amax(np.abs(tfo - pto))
+                tfo = tf_model(inputs_dict, training=False)
+                tf_hidden_states = tfo[0].numpy()
+                pt_hidden_states = pto[0].numpy()
+                tf_hidden_states[np.isnan(tf_hidden_states)] = 0
+                pt_hidden_states[np.isnan(pt_hidden_states)] = 0
+                max_diff = np.amax(np.abs(tf_hidden_states - pt_hidden_states))
                 self.assertLessEqual(max_diff, 2e-2)
 
                 # Check we can load pt model in tf and vice-versa with checkpoint => model functions
@@ -296,33 +296,46 @@ class TFCommonTestCases:
                 first, second = model(inputs_dict, training=False)[0], model(inputs_dict, training=False)[0]
                 self.assertTrue(tf.math.equal(first, second).numpy().all())
 
+        def _get_embeds(self, wte, input_ids):
+            # ^^ In our TF models, the input_embeddings can take slightly different forms,
+            # so we try a few of them.
+            # We used to fall back to just synthetically creating a dummy tensor of ones:
+            try:
+                x = wte(input_ids, mode="embedding")
+            except:
+                try:
+                    x = wte([input_ids], mode="embedding")
+                except:
+                    try:
+                        x = wte([input_ids, None, None, None], mode="embedding")
+                    except:
+                        if hasattr(self.model_tester, "embedding_size"):
+                            x = tf.ones(input_ids.shape + [self.model_tester.embedding_size], dtype=tf.dtypes.float32)
+                        else:
+                            x = tf.ones(input_ids.shape + [self.model_tester.hidden_size], dtype=tf.dtypes.float32)
+            return x
+
         def test_inputs_embeds(self):
             config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
-            input_ids = inputs_dict["input_ids"]
-            del inputs_dict["input_ids"]
+            if not self.is_encoder_decoder:
+                input_ids = inputs_dict["input_ids"]
+                del inputs_dict["input_ids"]
+            else:
+                encoder_input_ids = inputs_dict["encoder_input_ids"]
+                decoder_input_ids = inputs_dict["decoder_input_ids"]
+                del inputs_dict["encoder_input_ids"]
+                del inputs_dict["decoder_input_ids"]
 
             for model_class in self.all_model_classes:
                 model = model_class(config)
 
                 wte = model.get_input_embeddings()
-                try:
-                    x = wte(input_ids, mode="embedding")
-                except:
-                    try:
-                        x = wte([input_ids], mode="embedding")
-                    except:
-                        try:
-                            x = wte([input_ids, None, None, None], mode="embedding")
-                        except:
-                            if hasattr(self.model_tester, "embedding_size"):
-                                x = tf.ones(input_ids.shape + [self.model_tester.embedding_size], dtype=tf.dtypes.float32)
-                            else:
-                                x = tf.ones(input_ids.shape + [self.model_tester.hidden_size], dtype=tf.dtypes.float32)
-                # ^^ In our TF models, the input_embeddings can take slightly different forms,
-                # so we try a few of them.
-                # We used to fall back to just synthetically creating a dummy tensor of ones:
-                #
-                inputs_dict["inputs_embeds"] = x
+                if not self.is_encoder_decoder:
+                    inputs_dict["inputs_embeds"] = self._get_embeds(wte, input_ids)
+                else:
+                    inputs_dict["encoder_inputs_embeds"] = self._get_embeds(wte, encoder_input_ids)
+                    inputs_dict["decoder_inputs_embeds"] = self._get_embeds(wte, decoder_input_ids)
+
                 outputs = model(inputs_dict)
 
 
diff --git a/transformers/tests/modeling_tf_t5_test.py b/transformers/tests/modeling_tf_t5_test.py
index 33f6f895f0..99eec313f9 100644
--- a/transformers/tests/modeling_tf_t5_test.py
+++ b/transformers/tests/modeling_tf_t5_test.py
@@ -18,21 +18,21 @@ from __future__ import print_function
 
 import unittest
 import shutil
-import pytest
 import sys
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
+from .utils import require_tf, slow
 
 from transformers import T5Config, is_tf_available
 
 if is_tf_available():
     import tensorflow as tf
-    from transformers.modeling_tf_t5 import (TFT5Model, TFT5WithLMHeadModel,TF_T5_PRETRAINED_MODEL_ARCHIVE_MAP)
-else:
-    pytestmark = pytest.mark.skip("Require TensorFlow")
+    from transformers.modeling_tf_t5 import (TFT5Model, TFT5WithLMHeadModel,
+                                             TF_T5_PRETRAINED_MODEL_ARCHIVE_MAP)
 
 
+@require_tf
 class TFT5ModelTest(TFCommonTestCases.TFCommonModelTester):
 
     is_encoder_decoder = True
@@ -160,7 +160,7 @@ class TFT5ModelTest(TFCommonTestCases.TFCommonModelTester):
         config_and_inputs = self.model_tester.prepare_config_and_inputs()
         self.model_tester.create_and_check_t5_with_lm_head(*config_and_inputs)
 
-    @pytest.mark.slow
+    @slow
     def test_model_from_pretrained(self):
         cache_dir = "/tmp/transformers_test/"
         for model_name in ['t5-small']:
diff --git a/transformers/tests/tokenization_t5_test.py b/transformers/tests/tokenization_t5_test.py
index aabb21e443..0b4f960e32 100644
--- a/transformers/tests/tokenization_t5_test.py
+++ b/transformers/tests/tokenization_t5_test.py
@@ -16,7 +16,6 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 
 import os
 import unittest
-import pytest
 
 from transformers.tokenization_t5 import (T5Tokenizer)
 from transformers.tokenization_xlnet import SPIECE_UNDERLINE

From 4b82c485de187896a38c441587b7bd4d04f2821e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Tue, 10 Dec 2019 14:49:53 +0100
Subject: [PATCH 301/505] remove misplaced summarization documentation

---
 examples/README.md | 30 ------------------------------
 1 file changed, 30 deletions(-)

diff --git a/examples/README.md b/examples/README.md
index 620304ea77..b6b3908810 100644
--- a/examples/README.md
+++ b/examples/README.md
@@ -24,8 +24,6 @@ pip install -r ./examples/requirements.txt
 | [Multiple Choice](#multiple-choice) | Examples running BERT/XLNet/RoBERTa on the SWAG/RACE/ARC tasks. 
 | [Named Entity Recognition](#named-entity-recognition) | Using BERT for Named Entity Recognition (NER) on the CoNLL 2003 dataset, examples with distributed training.                                                                                  |
 | [XNLI](#xnli) | Examples running BERT/XLM on the XNLI benchmark. |
-| [Abstractive summarization](#abstractive-summarization) | Using the BertAbs
-model finetuned on the CNN/DailyMail dataset to generate summaries. |
 
 ## TensorFlow 2.0 Bert models on GLUE
 
@@ -646,34 +644,6 @@ micro avg     0.8722    0.8774    0.8748     13869
 macro avg     0.8712    0.8774    0.8740     13869
 ```
 
-## Abstractive summarization
-
-Based on the script
-[`run_summarization_finetuning.py`](https://github.com/huggingface/transformers/blob/master/examples/run_summarization_finetuning.py).
-
-Before running this script you should download **both** CNN and Daily Mail
-datasets from [Kyunghyun Cho's website](https://cs.nyu.edu/~kcho/DMQA/)  (the
-links next to "Stories") in the same folder. Then uncompress the archives by running:
-
-```bash
-tar -xvf cnn_stories.tgz && tar -xvf dailymail_stories.tgz
-```
-
-note that the finetuning script **will not work** if you do not download both
-datasets. We will refer as `$DATA_PATH` the path to where you uncompressed both
-archive.
-
-```bash
-export DATA_PATH=/path/to/dataset/
-
-python run_summarization_finetuning.py \
-    --output_dir=output \
-    --model_type=bert2bert \
-    --model_name_or_path=bert2bert \
-    --do_train \
-    --data_path=$DATA_PATH \
-```
-
 ## XNLI
 
 Based on the script [`run_xnli.py`](https://github.com/huggingface/transformers/blob/master/examples/run_xnli.py).

From 981a5c8c1789f91204ba1053f4742f6ea8c615af Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Tue, 10 Dec 2019 15:36:19 +0100
Subject: [PATCH 302/505] updating models urls

---
 transformers/configuration_t5.py                  |  4 ++++
 transformers/convert_pytorch_checkpoint_to_tf2.py |  2 +-
 transformers/modeling_t5.py                       |  4 ++++
 transformers/modeling_tf_t5.py                    |  6 +++++-
 transformers/tokenization_t5.py                   | 12 ++++++++++--
 5 files changed, 24 insertions(+), 4 deletions(-)

diff --git a/transformers/configuration_t5.py b/transformers/configuration_t5.py
index 83aab66fac..2ccdebc2b1 100644
--- a/transformers/configuration_t5.py
+++ b/transformers/configuration_t5.py
@@ -28,6 +28,10 @@ logger = logging.getLogger(__name__)
 
 T5_PRETRAINED_CONFIG_ARCHIVE_MAP = {
     't5-small': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-small-config.json",
+    't5-base': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-base-config.json",
+    't5-large': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-large-config.json",
+    't5-3B': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-3B-config.json",
+    't5-11B': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-11B-config.json",
 }
 
 
diff --git a/transformers/convert_pytorch_checkpoint_to_tf2.py b/transformers/convert_pytorch_checkpoint_to_tf2.py
index 4c4becfa00..06bb5f47c0 100644
--- a/transformers/convert_pytorch_checkpoint_to_tf2.py
+++ b/transformers/convert_pytorch_checkpoint_to_tf2.py
@@ -121,7 +121,7 @@ def convert_pt_checkpoint_to_tf(model_type, pytorch_checkpoint_path, config_file
 
     if compare_with_pt_model:
         inputs_list = [[7, 6, 0, 0, 1], [1, 2, 3, 0, 0], [0, 0, 0, 4, 5]]
-        tf_inputs = tf.constant(inputs_list)
+        tf_inputs = tf_model.dummy_inputs
         tfo = tf_model(tf_inputs, training=False)  # build the network
 
         pt_model = pt_model_class.from_pretrained(None,
diff --git a/transformers/modeling_t5.py b/transformers/modeling_t5.py
index f1e4e0306c..ffc4d8bb3f 100644
--- a/transformers/modeling_t5.py
+++ b/transformers/modeling_t5.py
@@ -42,6 +42,10 @@ logger = logging.getLogger(__name__)
 ####################################################
 T5_PRETRAINED_MODEL_ARCHIVE_MAP = {
     't5-small': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-small-pytorch_model.bin",
+    't5-base': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-base-pytorch_model.bin",
+    't5-large': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-large-pytorch_model.bin",
+    't5-3B': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-3B-pytorch_model.bin",
+    't5-11B': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-11B-pytorch_model.bin",
 }
 
 ####################################################
diff --git a/transformers/modeling_tf_t5.py b/transformers/modeling_tf_t5.py
index 447fd69b05..0b3b1116f2 100644
--- a/transformers/modeling_tf_t5.py
+++ b/transformers/modeling_tf_t5.py
@@ -25,13 +25,17 @@ import itertools
 import tensorflow as tf
 
 from .configuration_t5 import T5Config
-from .modeling_tf_utils import TFPreTrainedModel, TFSharedEmbeddings, shape_list, get_initializer, DUMMY_INPUTS
+from .modeling_tf_utils import TFPreTrainedModel, TFSharedEmbeddings, shape_list
 from .file_utils import add_start_docstrings
 
 logger = logging.getLogger(__name__)
 
 TF_T5_PRETRAINED_MODEL_ARCHIVE_MAP = {
     't5-small': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-small-tf_model.h5",
+    't5-base': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-base-tf_model.h5",
+    't5-large': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-large-tf_model.h5",
+    't5-3B': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-3B-tf_model.h5",
+    't5-11B': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-11B-tf_model.h5",
 }
 
 ####################################################
diff --git a/transformers/tokenization_t5.py b/transformers/tokenization_t5.py
index 933084d13a..62e9c069e2 100644
--- a/transformers/tokenization_t5.py
+++ b/transformers/tokenization_t5.py
@@ -41,7 +41,11 @@ VOCAB_FILES_NAMES = {'vocab_file': 'spiece.model'}
 PRETRAINED_VOCAB_FILES_MAP = {
     'vocab_file':
     {
-        't5': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-spiece.model",
+        't5-small': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-spiece.model",
+        't5-base': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-spiece.model",
+        't5-large': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-spiece.model",
+        't5-3B': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-spiece.model",
+        't5-11B': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-spiece.model",
     }
 }
 
@@ -49,7 +53,11 @@ PRETRAINED_VOCAB_FILES_MAP = {
 # Mapping from model shortcut names to max length of inputs
 ####################################################
 PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
-    't5': 512,
+    't5-small': 512,
+    't5-base': 512,
+    't5-large': 512,
+    't5-3B': 512,
+    't5-11B': 512,
 }
 
 class T5Tokenizer(PreTrainedTokenizer):

From 40a39ab65043f11763d8f0ce5fb0307661e6f7a3 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Tue, 10 Dec 2019 15:59:38 +0100
Subject: [PATCH 303/505] Reuse recent SQuAD refactored data structure inside
 QA pipelines.

---
 transformers/data/processors/__init__.py |  2 +-
 transformers/modeling_auto.py            | 12 +---
 transformers/pipelines.py                | 84 ++++++++++++++++--------
 3 files changed, 59 insertions(+), 39 deletions(-)

diff --git a/transformers/data/processors/__init__.py b/transformers/data/processors/__init__.py
index 0cef0080f4..4f7307bb7b 100644
--- a/transformers/data/processors/__init__.py
+++ b/transformers/data/processors/__init__.py
@@ -1,4 +1,4 @@
-from .utils import InputExample, InputFeatures, DataProcessor, SingleSentenceClassificationProcessor, convert_examples_to_features
+from .utils import InputExample, InputFeatures, DataProcessor, SingleSentenceClassificationProcessor
 from .glue import glue_output_modes, glue_processors, glue_tasks_num_labels, glue_convert_examples_to_features
 from .squad import squad_convert_examples_to_features, SquadFeatures, SquadExample, SquadV1Processor, SquadV2Processor
 from .xnli import xnli_output_modes, xnli_processors, xnli_tasks_num_labels
\ No newline at end of file
diff --git a/transformers/modeling_auto.py b/transformers/modeling_auto.py
index 0c8bffa883..041115cc61 100644
--- a/transformers/modeling_auto.py
+++ b/transformers/modeling_auto.py
@@ -31,7 +31,7 @@ from .modeling_xlnet import XLNetModel, XLNetLMHeadModel, XLNetForSequenceClassi
 from .modeling_xlm import XLMModel, XLMWithLMHeadModel, XLMForSequenceClassification, XLMForQuestionAnswering
 from .modeling_roberta import RobertaModel, RobertaForMaskedLM, RobertaForSequenceClassification
 from .modeling_distilbert import DistilBertModel, DistilBertForQuestionAnswering, DistilBertForMaskedLM, DistilBertForSequenceClassification
-from .modeling_camembert import CamembertModel, CamembertForQuestionAnswering, CamembertForMaskedLM, CamembertForSequenceClassification, CamembertForMultipleChoice
+from .modeling_camembert import CamembertModel, CamembertForMaskedLM, CamembertForSequenceClassification, CamembertForMultipleChoice
 from .modeling_albert import AlbertModel, AlbertForMaskedLM, AlbertForSequenceClassification, AlbertForQuestionAnswering
 
 from .modeling_utils import PreTrainedModel, SequenceSummary
@@ -294,10 +294,6 @@ class AutoModelWithLMHead(object):
             return XLMWithLMHeadModel(config)
         elif isinstance(config, CTRLConfig):
             return CTRLLMHeadModel(config)
-        elif isinstance(config, AlbertConfig):
-            return AlbertLMHeadModel(config)
-        elif isinstance(config, CamembertConfig):
-            return CamembertLMHeadModel(config)
         raise ValueError("Unrecognized configuration class {}".format(config))
 
     @classmethod
@@ -454,7 +450,7 @@ class AutoModelForSequenceClassification(object):
         """
         if isinstance(config, AlbertConfig):
             return AlbertForSequenceClassification(config)
-        elif isintance(config, CamembertConfig):
+        elif isinstance(config, CamembertConfig):
             return CamembertForSequenceClassification(config)
         elif isinstance(config, DistilBertConfig):
             return DistilBertForSequenceClassification(config)
@@ -606,10 +602,8 @@ class AutoModelForQuestionAnswering(object):
             config = BertConfig.from_pretrained('bert-base-uncased')    # Download configuration from S3 and cache.
             model = AutoModelForSequenceClassification.from_config(config)  # E.g. model was saved using `save_pretrained('./test/saved_model/')`
         """
-        if isintance(config, AlbertConfig):
+        if isinstance(config, AlbertConfig):
             return AlbertForQuestionAnswering(config)
-        elif isintance(config, CamembertConfig):
-            return CamembertForQuestionAnswering(config)
         elif isinstance(config, DistilBertConfig):
             return DistilBertForQuestionAnswering(config)
         elif isinstance(config, BertConfig):
diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 1701915203..1e2f035d9f 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -20,7 +20,8 @@ from typing import Union, Optional, Tuple, List, Dict
 
 import numpy as np
 
-from transformers import is_tf_available, is_torch_available, logger, AutoTokenizer, PreTrainedTokenizer
+from transformers import is_tf_available, is_torch_available, logger, AutoTokenizer, PreTrainedTokenizer, \
+    SquadExample, squad_convert_examples_to_features
 
 if is_tf_available():
     from transformers import TFAutoModelForSequenceClassification, TFAutoModelForQuestionAnswering
@@ -107,13 +108,28 @@ class QuestionAnsweringPipeline(Pipeline):
         super().__init__(model, tokenizer)
 
     @staticmethod
-    def create_sample(question: Union[str, List[str]], context: Union[str, List[str]]) -> Union[dict, List[Dict]]:
+    def create_sample(question: Union[str, List[str]], context: Union[str, List[str]]) -> Union[SquadExample, List[SquadExample]]:
         is_list = isinstance(question, list)
 
         if is_list:
-            return [{'question': q, 'context': c} for q, c in zip(question, context)]
+            return [SquadExample(None, q, c, None, None, None) for q, c in zip(question, context)]
         else:
-            return {'question': question, 'context': context}
+            return SquadExample(None, question, context, None, None, None)
+
+    def inputs_for_model(self, features: Union[SquadExample, List[SquadExample]]) -> Dict:
+        args = ['input_ids', 'attention_mask']
+        model_type = type(self.model).__name__.lower()
+
+        if 'distilbert' not in model_type and 'xlm' not in model_type:
+            args += ['token_type_ids']
+
+        if 'xlnet' in model_type or 'xlm' in model_type:
+            args += ['cls_index', 'p_mask']
+
+        if isinstance(features, SquadExample):
+            return {k: features.__dict__[k] for k in args}
+        else:
+            return {k: [feature.__dict__[k] for feature in features] for k in args}
 
     @classmethod
     def from_config(cls, model, tokenizer: PreTrainedTokenizer, **kwargs):
@@ -121,8 +137,11 @@ class QuestionAnsweringPipeline(Pipeline):
 
     def __call__(self, *texts, **kwargs):
         # Set defaults values
-        kwargs.setdefault('max_answer_len', 15)
         kwargs.setdefault('topk', 1)
+        kwargs.setdefault('doc_stride', 128)
+        kwargs.setdefault('max_answer_len', 15)
+        kwargs.setdefault('max_seq_len', 384)
+        kwargs.setdefault('max_question_len', 64)
 
         if kwargs['topk'] < 1:
             raise ValueError('topk parameter should be >= 1 (got {})'.format(kwargs['topk']))
@@ -130,56 +149,63 @@ class QuestionAnsweringPipeline(Pipeline):
         if kwargs['max_answer_len'] < 1:
             raise ValueError('max_answer_len parameter should be >= 1 (got {})'.format(kwargs['max_answer_len']))
 
-        # Tabular input
-        if 'question' in kwargs and 'context' in kwargs:
-            texts = QuestionAnsweringPipeline.create_sample(kwargs['question'], kwargs['context'])
-        elif 'data' in kwargs:
-            texts = kwargs['data']
+        # Position args
+        if texts is not None and len(texts) > 1:
+            (texts, ) = texts
+
         # Generic compatibility with sklearn and Keras
         elif 'X' in kwargs and not texts:
             texts = kwargs.pop('X')
-        else:
-            (texts, ) = texts
 
-        if not isinstance(texts, (dict, list)):
-            raise Exception('QuestionAnsweringPipeline requires predict argument to be a tuple (context, question) or a List of dict.')
+        # Batched data
+        elif 'data' in kwargs:
+            texts = kwargs.pop('data')
+
+        # Tabular input
+        elif 'question' in kwargs and 'context' in kwargs:
+            texts = QuestionAnsweringPipeline.create_sample(kwargs['question'], kwargs['context'])
+        else:
+            raise ValueError('Unknown arguments {}'.format(kwargs))
 
         if not isinstance(texts, list):
             texts = [texts]
 
-        # Map to tuple (question, context)
-        texts = [(text['question'], text['context']) for text in texts]
-        inputs = self.tokenizer.batch_encode_plus(
-            texts, add_special_tokens=False, return_tensors='tf' if is_tf_available() else 'pt',
-            return_attention_masks=True, return_input_lengths=False
-        )
-
-        token_type_ids = inputs.pop('token_type_ids')
+        # Convert inputs to features
+        features = squad_convert_examples_to_features(texts, self.tokenizer, kwargs['max_seq_len'], kwargs['doc_stride'], kwargs['max_question_len'], False)
+        fw_args = self.inputs_for_model(features)
 
         if is_tf_available():
-            # TODO trace model
-            start, end = self.model(inputs)
+            import tensorflow as tf
+            fw_args = {k: tf.constant(v) for (k, v) in fw_args.items()}
+            start, end = self.model(fw_args)
             start, end = start.numpy(), end.numpy()
         else:
             import torch
             with torch.no_grad():
                 # Retrieve the score for the context tokens only (removing question tokens)
-                start, end = self.model(**inputs)
+                fw_args = {k: torch.tensor(v) for (k, v) in fw_args.items()}
+                start, end = self.model(**fw_args)
                 start, end = start.cpu().numpy(), end.cpu().numpy()
 
         answers = []
-        for i in range(len(texts)):
-            context_idx = token_type_ids[i] == 1
-            start_, end_ = start[i, context_idx], end[i, context_idx]
+        for i, (example, feature, start_, end_) in enumerate(zip(texts, features, start, end)):
+            start_, end_ = start_ * np.abs(np.array(feature.p_mask) - 1), end_ * np.abs(np.array(feature.p_mask) - 1)
 
             # Normalize logits and spans to retrieve the answer
             start_ = np.exp(start_) / np.sum(np.exp(start_))
             end_ = np.exp(end_) / np.sum(np.exp(end_))
             starts, ends, scores = self.decode(start_, end_, kwargs['topk'], kwargs['max_answer_len'])
 
+            char_to_word = np.array(example.char_to_word_offset)
+
             # Convert the answer (tokens) back to the original text
             answers += [[
-                {**{'score': score}, **self.span_to_answer(texts[i][1], s, e)}
+                {
+                    'score': score,
+                    'start': np.where(char_to_word == feature.token_to_orig_map[s])[0][0],
+                    'end': np.where(char_to_word == feature.token_to_orig_map[e])[0][-1],
+                    'answer': ' '.join(example.doc_tokens[feature.token_to_orig_map[s]: feature.token_to_orig_map[e] + 1])
+                }
                 for s, e, score in zip(starts, ends, scores)
             ]]
 

From a5df980c5b86e9106382a87a63b977d5decf97f6 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Tue, 10 Dec 2019 16:01:15 +0100
Subject: [PATCH 304/505] updating distilbert test

---
 transformers/tests/modeling_common_test.py    | 7 ++++++-
 transformers/tests/modeling_tf_common_test.py | 7 ++++++-
 2 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/transformers/tests/modeling_common_test.py b/transformers/tests/modeling_common_test.py
index 792f5cee3e..2f2baff436 100644
--- a/transformers/tests/modeling_common_test.py
+++ b/transformers/tests/modeling_common_test.py
@@ -121,7 +121,12 @@ class CommonTestCases:
                 model.to(torch_device)
                 model.eval()
                 first, second = model(**inputs_dict)[0], model(**inputs_dict)[0]
-                self.assertEqual(first.ne(second).sum().item(), 0)
+                out_1 = first.cpu().numpy()
+                out_2 = second.cpu().numpy()
+                out_1 = out_1[~np.isnan(out_1)]
+                out_2 = out_2[~np.isnan(out_2)]
+                max_diff = np.amax(np.abs(out_1 - out_2))
+                self.assertLessEqual(max_diff, 1e-5)
 
         def test_attention_outputs(self):
             config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
diff --git a/transformers/tests/modeling_tf_common_test.py b/transformers/tests/modeling_tf_common_test.py
index a0d63583fb..5a5873e81b 100644
--- a/transformers/tests/modeling_tf_common_test.py
+++ b/transformers/tests/modeling_tf_common_test.py
@@ -294,7 +294,12 @@ class TFCommonTestCases:
             for model_class in self.all_model_classes:
                 model = model_class(config)
                 first, second = model(inputs_dict, training=False)[0], model(inputs_dict, training=False)[0]
-                self.assertTrue(tf.math.equal(first, second).numpy().all())
+                out_1 = first.numpy()
+                out_2 = second.numpy()
+                out_1 = out_1[~np.isnan(out_1)]
+                out_2 = out_2[~np.isnan(out_2)]
+                max_diff = np.amax(np.abs(out_1 - out_2))
+                self.assertLessEqual(max_diff, 1e-5)
 
         def _get_embeds(self, wte, input_ids):
             # ^^ In our TF models, the input_embeddings can take slightly different forms,

From f2538c12741df74abbd2ff38f43019cfbb21093b Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Tue, 10 Dec 2019 16:33:11 +0100
Subject: [PATCH 305/505] all tests in torch no grad

---
 transformers/tests/modeling_common_test.py | 53 ++++++++++++++--------
 1 file changed, 35 insertions(+), 18 deletions(-)

diff --git a/transformers/tests/modeling_common_test.py b/transformers/tests/modeling_common_test.py
index 2f2baff436..ed6f950e25 100644
--- a/transformers/tests/modeling_common_test.py
+++ b/transformers/tests/modeling_common_test.py
@@ -120,7 +120,9 @@ class CommonTestCases:
                 model = model_class(config)
                 model.to(torch_device)
                 model.eval()
-                first, second = model(**inputs_dict)[0], model(**inputs_dict)[0]
+                with torch.no_grad():
+                    first = model(**inputs_dict)[0]
+                    second = model(**inputs_dict)[0]
                 out_1 = first.cpu().numpy()
                 out_2 = second.cpu().numpy()
                 out_1 = out_1[~np.isnan(out_1)]
@@ -142,7 +144,8 @@ class CommonTestCases:
                 model = model_class(config)
                 model.to(torch_device)
                 model.eval()
-                outputs = model(**inputs_dict)
+                with torch.no_grad():
+                    outputs = model(**inputs_dict)
                 attentions = outputs[-1]
                 self.assertEqual(model.config.output_attentions, True)
                 self.assertEqual(model.config.output_hidden_states, False)
@@ -173,7 +176,8 @@ class CommonTestCases:
                 model = model_class(config)
                 model.to(torch_device)
                 model.eval()
-                outputs = model(**inputs_dict)
+                with torch.no_grad():
+                    outputs = model(**inputs_dict)
                 self.assertEqual(out_len + (2 if self.is_encoder_decoder else 1), len(outputs))
                 self.assertEqual(model.config.output_attentions, True)
                 self.assertEqual(model.config.output_hidden_states, True)
@@ -273,7 +277,8 @@ class CommonTestCases:
                 inputs = inputs_dict.copy()
                 inputs['head_mask'] = head_mask
 
-                outputs = model(**inputs)
+                with torch.no_grad():
+                    outputs = model(**inputs)
 
                 # Test that we can get a gradient back for importance score computation
                 output = sum(t.sum() for t in outputs[0])
@@ -320,7 +325,8 @@ class CommonTestCases:
                 heads_to_prune = {0: list(range(1, self.model_tester.num_attention_heads)),
                                 -1: [0]}
                 model.prune_heads(heads_to_prune)
-                outputs = model(**inputs_dict)
+                with torch.no_grad():
+                    outputs = model(**inputs_dict)
 
                 attentions = outputs[-1]
 
@@ -356,7 +362,8 @@ class CommonTestCases:
                 model = model_class.from_pretrained(directory)
                 model.to(torch_device)
 
-                outputs = model(**inputs_dict)
+                with torch.no_grad():
+                    outputs = model(**inputs_dict)
                 attentions = outputs[-1]
                 self.assertEqual(attentions[0].shape[-3], 1)
                 self.assertEqual(attentions[1].shape[-3], self.model_tester.num_attention_heads)
@@ -385,7 +392,8 @@ class CommonTestCases:
                 model.to(torch_device)
                 model.eval()
 
-                outputs = model(**inputs_dict)
+                with torch.no_grad():
+                    outputs = model(**inputs_dict)
                 attentions = outputs[-1]
 
                 self.assertEqual(attentions[0].shape[-3], 1)
@@ -412,7 +420,8 @@ class CommonTestCases:
                 model.to(torch_device)
                 model.eval()
 
-                outputs = model(**inputs_dict)
+                with torch.no_grad():
+                    outputs = model(**inputs_dict)
                 attentions = outputs[-1]
 
                 self.assertEqual(attentions[0].shape[-3], self.model_tester.num_attention_heads - 1)
@@ -429,7 +438,8 @@ class CommonTestCases:
                 model.to(torch_device)
                 shutil.rmtree(directory)
 
-                outputs = model(**inputs_dict)
+                with torch.no_grad():
+                    outputs = model(**inputs_dict)
                 attentions = outputs[-1]
 
                 self.assertEqual(attentions[0].shape[-3], self.model_tester.num_attention_heads - 1)
@@ -440,7 +450,8 @@ class CommonTestCases:
                 heads_to_prune = {0: [0], 2: [1, 2]}
                 model.prune_heads(heads_to_prune)
 
-                outputs = model(**inputs_dict)
+                with torch.no_grad():
+                    outputs = model(**inputs_dict)
                 attentions = outputs[-1]
 
                 self.assertEqual(attentions[0].shape[-3], self.model_tester.num_attention_heads -1)
@@ -459,7 +470,8 @@ class CommonTestCases:
                 model = model_class(config)
                 model.to(torch_device)
                 model.eval()
-                outputs = model(**inputs_dict)
+                with torch.no_grad():
+                    outputs = model(**inputs_dict)
                 hidden_states = outputs[-1]
                 self.assertEqual(model.config.output_attentions, False)
                 self.assertEqual(model.config.output_hidden_states, True)
@@ -594,7 +606,8 @@ class CommonTestCases:
                     inputs_dict["encoder_inputs_embeds"] = wte(encoder_input_ids)
                     inputs_dict["decoder_inputs_embeds"] = wte(decoder_input_ids)
 
-                outputs = model(**inputs_dict)
+                with torch.no_grad():
+                    outputs = model(**inputs_dict)
 
     class GPTModelTester(CommonModelTester):
 
@@ -682,9 +695,10 @@ class CommonTestCases:
             model.to(torch_device)
             model.eval()
 
-            outputs = model(input_ids, position_ids, token_type_ids)
-            outputs = model(input_ids, position_ids)
-            outputs = model(input_ids)
+            with torch.no_grad():
+                outputs = model(input_ids, position_ids, token_type_ids)
+                outputs = model(input_ids, position_ids)
+                outputs = model(input_ids)
 
             hidden_state = outputs[0]
             self.parent.assertListEqual(
@@ -697,7 +711,8 @@ class CommonTestCases:
             model = self.lm_head_model_class(config)
             model.to(torch_device)
             model.eval()
-            outputs = model(input_ids, position_ids, token_type_ids, lm_labels)
+            with torch.no_grad():
+                outputs = model(input_ids, position_ids, token_type_ids, lm_labels)
             loss, lm_logits = outputs[:2]
 
             total_voc = self.vocab_size
@@ -714,7 +729,8 @@ class CommonTestCases:
                 model = model_class(config)
                 model.to(torch_device)
                 model.eval()
-                outputs = model(input_ids)
+                with torch.no_grad():
+                    outputs = model(input_ids)
                 presents = outputs[-1]
                 self.parent.assertEqual(self.num_hidden_layers, len(presents))
                 self.parent.assertListEqual(
@@ -727,7 +743,8 @@ class CommonTestCases:
             model = self.double_head_model_class(config)
             model.to(torch_device)
             model.eval()
-            outputs = model(input_ids, mc_token_ids, lm_labels=lm_labels, mc_labels=mc_labels,
+            with torch.no_grad():
+                outputs = model(input_ids, mc_token_ids, lm_labels=lm_labels, mc_labels=mc_labels,
                             token_type_ids=token_type_ids, position_ids=position_ids)
             lm_loss, mc_loss, lm_logits, mc_logits = outputs[:4]
             loss = [lm_loss, mc_loss]

From 63e36007ee152cee23f44103622f28566e28fb72 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Tue, 10 Dec 2019 16:47:35 +0100
Subject: [PATCH 306/505] Make sure padding, cls and another non-context tokens
 cannot appear in the answer.

---
 transformers/pipelines.py | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 1e2f035d9f..eec4932321 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -188,14 +188,18 @@ class QuestionAnsweringPipeline(Pipeline):
                 start, end = start.cpu().numpy(), end.cpu().numpy()
 
         answers = []
-        for i, (example, feature, start_, end_) in enumerate(zip(texts, features, start, end)):
-            start_, end_ = start_ * np.abs(np.array(feature.p_mask) - 1), end_ * np.abs(np.array(feature.p_mask) - 1)
-
+        for (example, feature, start_, end_) in zip(texts, features, start, end):
             # Normalize logits and spans to retrieve the answer
             start_ = np.exp(start_) / np.sum(np.exp(start_))
             end_ = np.exp(end_) / np.sum(np.exp(end_))
-            starts, ends, scores = self.decode(start_, end_, kwargs['topk'], kwargs['max_answer_len'])
 
+            # Mask padding and question
+            start_, end_ = start_ * np.abs(np.array(feature.p_mask) - 1), end_ * np.abs(np.array(feature.p_mask) - 1)
+
+            # Mask CLS
+            start_[0] = end_[0] = 0
+
+            starts, ends, scores = self.decode(start_, end_, kwargs['topk'], kwargs['max_answer_len'])
             char_to_word = np.array(example.char_to_word_offset)
 
             # Convert the answer (tokens) back to the original text

From 67a8be8e90a7fbd5e0bceff9f29fb89ccabb61be Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Tue, 10 Dec 2019 17:50:32 +0100
Subject: [PATCH 307/505] fix backward in tests

---
 transformers/tests/modeling_common_test.py | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/transformers/tests/modeling_common_test.py b/transformers/tests/modeling_common_test.py
index ed6f950e25..cd4cf247a6 100644
--- a/transformers/tests/modeling_common_test.py
+++ b/transformers/tests/modeling_common_test.py
@@ -277,8 +277,7 @@ class CommonTestCases:
                 inputs = inputs_dict.copy()
                 inputs['head_mask'] = head_mask
 
-                with torch.no_grad():
-                    outputs = model(**inputs)
+                outputs = model(**inputs)
 
                 # Test that we can get a gradient back for importance score computation
                 output = sum(t.sum() for t in outputs[0])

From dc4e9e5cb36ae9bf5185b49b1cbc9106857abd54 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Tue, 10 Dec 2019 19:21:20 +0000
Subject: [PATCH 308/505] DataParallel for SQuAD + fix XLM

---
 examples/run_squad.py                      | 6 +++++-
 transformers/data/metrics/squad_metrics.py | 7 ++++++-
 transformers/tokenization_xlm.py           | 4 ++++
 3 files changed, 15 insertions(+), 2 deletions(-)

diff --git a/examples/run_squad.py b/examples/run_squad.py
index 2df29014ef..5e3f9663e2 100644
--- a/examples/run_squad.py
+++ b/examples/run_squad.py
@@ -299,10 +299,14 @@ def evaluate(args, model, tokenizer, prefix=""):
 
     # XLNet and XLM use a more complex post-processing procedure
     if args.model_type in ['xlnet', 'xlm']:
+
+        start_n_top = model.config.start_n_top if hasattr(model, "config") else model.module.config.start_n_top
+        end_n_top = model.config.end_n_top if hasattr(model, "config") else model.module.config.end_n_top
+
         predictions = compute_predictions_log_probs(examples, features, all_results, args.n_best_size,
                         args.max_answer_length, output_prediction_file,
                         output_nbest_file, output_null_log_odds_file,
-                        model.config.start_n_top, model.config.end_n_top,
+                        start_n_top, end_n_top,
                         args.version_2_with_negative, tokenizer, args.verbose_logging)
     else:
         predictions = compute_predictions_logits(examples, features, all_results, args.n_best_size,
diff --git a/transformers/data/metrics/squad_metrics.py b/transformers/data/metrics/squad_metrics.py
index 0755c0ab7a..7b03255f49 100644
--- a/transformers/data/metrics/squad_metrics.py
+++ b/transformers/data/metrics/squad_metrics.py
@@ -695,7 +695,12 @@ def compute_predictions_log_probs(
             tok_text = " ".join(tok_text.split())
             orig_text = " ".join(orig_tokens)
 
-            final_text = get_final_text(tok_text, orig_text, tokenizer.do_lower_case,
+            if hasattr(tokenizer, "do_lower_case"):
+                do_lower_case = tokenizer.do_lower_case
+            else:
+                do_lower_case = tokenizer.do_lowercase_and_remove_accent
+
+            final_text = get_final_text(tok_text, orig_text, do_lower_case,
                                         verbose_logging)
 
             if final_text in seen_predictions:
diff --git a/transformers/tokenization_xlm.py b/transformers/tokenization_xlm.py
index 6c9f8e5e5c..8def80bec4 100644
--- a/transformers/tokenization_xlm.py
+++ b/transformers/tokenization_xlm.py
@@ -549,6 +549,10 @@ class XLMTokenizer(PreTrainedTokenizer):
                                            additional_special_tokens=additional_special_tokens,
                                            **kwargs)
 
+
+        self.max_len_single_sentence = self.max_len - 2  # take into account special tokens
+        self.max_len_sentences_pair = self.max_len - 3  # take into account special tokens 
+
         # cache of sm.MosesPunctNormalizer instance
         self.cache_moses_punct_normalizer = dict()
         # cache of sm.MosesTokenizer instance

From 6a73382706ce3c6905023872f63a680f0eb419a4 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Tue, 10 Dec 2019 14:33:24 -0500
Subject: [PATCH 309/505] Complete warning + cleanup

---
 examples/run_squad.py              | 1 -
 transformers/tokenization_utils.py | 4 ++--
 2 files changed, 2 insertions(+), 3 deletions(-)

diff --git a/examples/run_squad.py b/examples/run_squad.py
index 5e3f9663e2..79c8537a4b 100644
--- a/examples/run_squad.py
+++ b/examples/run_squad.py
@@ -299,7 +299,6 @@ def evaluate(args, model, tokenizer, prefix=""):
 
     # XLNet and XLM use a more complex post-processing procedure
     if args.model_type in ['xlnet', 'xlm']:
-
         start_n_top = model.config.start_n_top if hasattr(model, "config") else model.module.config.start_n_top
         end_n_top = model.config.end_n_top if hasattr(model, "config") else model.module.config.end_n_top
 
diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index f4395cd82c..cb931b0eaf 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -226,7 +226,7 @@ class PreTrainedTokenizer(object):
 
         self.max_len = max_len if max_len is not None else int(1e12)
 
-        # Padding side is right by default and over-riden in subclsses. If specified in the kwargs, it is changed.
+        # Padding side is right by default and over-riden in subclasses. If specified in the kwargs, it is changed.
         self.padding_side = kwargs.pop('padding_side', self.padding_side)
         
         # Added tokens
@@ -1003,7 +1003,7 @@ class PreTrainedTokenizer(object):
         )
 
         if pad_to_max_length and max_length is None and self.max_len > 10000:
-            logger.warning("Sequence can't be padded as the maximum  ")
+            logger.warning("Sequence can't be padded as no maximum length is specified and the model maximum length is too high.")
 
         if needs_to_be_padded:
             difference = (max_length if max_length is not None else self.max_len) - len(encoded_inputs["input_ids"])

From 58d75aa310e872723ba92ee1f0cb575ae9e2eaef Mon Sep 17 00:00:00 2001
From: Leo Dirac <deepembedding@gmail.com>
Date: Tue, 10 Dec 2019 11:36:56 -0800
Subject: [PATCH 310/505] Progress indicator improvements when downloading
 pre-trained models.

---
 transformers/file_utils.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/transformers/file_utils.py b/transformers/file_utils.py
index 24abd60781..68de4e6e2f 100644
--- a/transformers/file_utils.py
+++ b/transformers/file_utils.py
@@ -21,7 +21,7 @@ import boto3
 from botocore.config import Config
 from botocore.exceptions import ClientError
 import requests
-from tqdm import tqdm
+from tqdm.auto import tqdm
 from contextlib import contextmanager
 
 logger = logging.getLogger(__name__)  # pylint: disable=invalid-name
@@ -245,7 +245,7 @@ def http_get(url, temp_file, proxies=None, resume_size=0):
         return
     content_length = response.headers.get('Content-Length')
     total = resume_size + int(content_length) if content_length is not None else None
-    progress = tqdm(unit="B", total=total, initial=resume_size)
+    progress = tqdm(unit="B", unit_scale=True, total=total, initial=resume_size, desc="Downloading")
     for chunk in response.iter_content(chunk_size=1024):
         if chunk: # filter out keep-alive new chunks
             progress.update(len(chunk))

From 9a24e0cf767601858f13808d37b5b71787b7641e Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Wed, 11 Dec 2019 00:33:25 +0100
Subject: [PATCH 311/505] Refactored qa pipeline argument handling + unittests

---
 transformers/pipelines.py            | 87 ++++++++++++++++++----------
 transformers/tests/pipelines_test.py | 33 ++++++++++-
 2 files changed, 87 insertions(+), 33 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index eec4932321..da8b0b65a7 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -98,14 +98,11 @@ class TextClassificationPipeline(Pipeline):
 class QuestionAnsweringPipeline(Pipeline):
     """
     Question Answering pipeling involving Tokenization and Inference.
-    TODO:
-     - top-k answers
-     - return start/end chars
-     - return score
     """
 
-    def __init__(self, model, tokenizer: Optional[PreTrainedTokenizer]):
-        super().__init__(model, tokenizer)
+    @classmethod
+    def from_config(cls, model, tokenizer: PreTrainedTokenizer, **kwargs):
+        pass
 
     @staticmethod
     def create_sample(question: Union[str, List[str]], context: Union[str, List[str]]) -> Union[SquadExample, List[SquadExample]]:
@@ -116,6 +113,55 @@ class QuestionAnsweringPipeline(Pipeline):
         else:
             return SquadExample(None, question, context, None, None, None)
 
+    @staticmethod
+    def handle_args(*inputs, **kwargs) -> List[SquadExample]:
+        # Position args, handling is sensibly the same as X and data, so forwarding to avoid duplicating
+        if inputs is not None and len(inputs) > 1:
+            kwargs['X'] = inputs
+
+        # Generic compatibility with sklearn and Keras
+        # Batched data
+        if 'X' in kwargs or 'data' in kwargs:
+            data = kwargs['X'] if 'X' in kwargs else kwargs['data']
+
+            if not isinstance(data, list):
+                data = [data]
+
+            for i, item in enumerate(data):
+                if isinstance(item, dict):
+                    if any(k not in item for k in ['question', 'context']):
+                        raise KeyError('You need to provide a dictionary with keys {question:..., context:...}')
+                    data[i] = QuestionAnsweringPipeline.create_sample(**item)
+
+                elif isinstance(item, SquadExample):
+                    continue
+                else:
+                    raise ValueError(
+                        '{} argument needs to be of type (list[SquadExample | dict], SquadExample, dict)'
+                        .format('X' if 'X' in kwargs else 'data')
+                    )
+            inputs = data
+
+        # Tabular input
+        elif 'question' in kwargs and 'context' in kwargs:
+            if isinstance(kwargs['question'], str):
+                kwargs['question'] = [kwargs['question']]
+
+            if isinstance(kwargs['context'], str):
+                kwargs['context'] = [kwargs['context']]
+
+            inputs = [QuestionAnsweringPipeline.create_sample(q, c) for q, c in zip(kwargs['question'], kwargs['context'])]
+        else:
+            raise ValueError('Unknown arguments {}'.format(kwargs))
+
+        if not isinstance(inputs, list):
+            inputs = [inputs]
+
+        return inputs
+
+    def __init__(self, model, tokenizer: Optional[PreTrainedTokenizer]):
+        super().__init__(model, tokenizer)
+
     def inputs_for_model(self, features: Union[SquadExample, List[SquadExample]]) -> Dict:
         args = ['input_ids', 'attention_mask']
         model_type = type(self.model).__name__.lower()
@@ -131,10 +177,6 @@ class QuestionAnsweringPipeline(Pipeline):
         else:
             return {k: [feature.__dict__[k] for feature in features] for k in args}
 
-    @classmethod
-    def from_config(cls, model, tokenizer: PreTrainedTokenizer, **kwargs):
-        pass
-
     def __call__(self, *texts, **kwargs):
         # Set defaults values
         kwargs.setdefault('topk', 1)
@@ -149,29 +191,10 @@ class QuestionAnsweringPipeline(Pipeline):
         if kwargs['max_answer_len'] < 1:
             raise ValueError('max_answer_len parameter should be >= 1 (got {})'.format(kwargs['max_answer_len']))
 
-        # Position args
-        if texts is not None and len(texts) > 1:
-            (texts, ) = texts
-
-        # Generic compatibility with sklearn and Keras
-        elif 'X' in kwargs and not texts:
-            texts = kwargs.pop('X')
-
-        # Batched data
-        elif 'data' in kwargs:
-            texts = kwargs.pop('data')
-
-        # Tabular input
-        elif 'question' in kwargs and 'context' in kwargs:
-            texts = QuestionAnsweringPipeline.create_sample(kwargs['question'], kwargs['context'])
-        else:
-            raise ValueError('Unknown arguments {}'.format(kwargs))
-
-        if not isinstance(texts, list):
-            texts = [texts]
+        examples = QuestionAnsweringPipeline.handle_args(texts, **kwargs)
 
         # Convert inputs to features
-        features = squad_convert_examples_to_features(texts, self.tokenizer, kwargs['max_seq_len'], kwargs['doc_stride'], kwargs['max_question_len'], False)
+        features = squad_convert_examples_to_features(examples, self.tokenizer, kwargs['max_seq_len'], kwargs['doc_stride'], kwargs['max_question_len'], False)
         fw_args = self.inputs_for_model(features)
 
         if is_tf_available():
@@ -188,7 +211,7 @@ class QuestionAnsweringPipeline(Pipeline):
                 start, end = start.cpu().numpy(), end.cpu().numpy()
 
         answers = []
-        for (example, feature, start_, end_) in zip(texts, features, start, end):
+        for (example, feature, start_, end_) in zip(examples, features, start, end):
             # Normalize logits and spans to retrieve the answer
             start_ = np.exp(start_) / np.sum(np.exp(start_))
             end_ = np.exp(end_) / np.sum(np.exp(end_))
diff --git a/transformers/tests/pipelines_test.py b/transformers/tests/pipelines_test.py
index 36d6e013c3..ee10234269 100644
--- a/transformers/tests/pipelines_test.py
+++ b/transformers/tests/pipelines_test.py
@@ -40,7 +40,38 @@ class QuestionAnsweringPipelineTest(unittest.TestCase):
 
         # Batch case with topk = 2
         a = nlp(question=['What is the name of the company I\'m working for ?', 'Where is the company based ?'],
-                context=['I\'m working for Huggingface.', 'The company is based in New York and Paris'], topk=2)
+                context=['Where is the company based ?', 'The company is based in New York and Paris'], topk=2)
+        self.check_answer_structure(a, 2, 2)
+
+        # check for data keyword
+        a = nlp(data=nlp.create_sample(question='What is the name of the company I\'m working for ?', context='I\'m working for Huggingface.'))
+        self.check_answer_structure(a, 1, 1)
+
+        a = nlp(data=nlp.create_sample(question='What is the name of the company I\'m working for ?', context='I\'m working for Huggingface.'), topk=2)
+        self.check_answer_structure(a, 1, 2)
+
+        a = nlp(data=[
+            nlp.create_sample(question='What is the name of the company I\'m working for ?', context='I\'m working for Huggingface.'),
+            nlp.create_sample(question='I\'m working for Huggingface.', context='The company is based in New York and Paris'),
+        ])
+        self.check_answer_structure(a, 2, 1)
+
+        a = nlp(data=[
+            {'question': 'What is the name of the company I\'m working for ?', 'context': 'I\'m working for Huggingface.'},
+            {'question': 'Where is the company based ?', 'context': 'The company is based in New York and Paris'},
+        ])
+        self.check_answer_structure(a, 2, 1)
+
+        # X keywords
+        a = nlp(X=nlp.create_sample(
+            question='Where is the company based ?', context='The company is based in New York and Paris'
+        ))
+        self.check_answer_structure(a, 1, 1)
+
+        a = nlp(X=[
+            {'question': 'What is the name of the company I\'m working for ?', 'context': 'I\'m working for Huggingface.'},
+            {'question': 'Where is the company based ?', 'context': 'The company is based in New York and Paris'},
+        ], topk=2)
         self.check_answer_structure(a, 2, 2)
 
     @patch('transformers.pipelines.is_torch_available', return_value=False)

From fafd4c86ecb63bb90b095bbd23453553e33fe99d Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Wed, 11 Dec 2019 13:47:27 +0100
Subject: [PATCH 312/505] fix TF 2.0 version of T5 - update conversion script

---
 .../convert_pytorch_checkpoint_to_tf2.py      | 11 ++---
 transformers/file_utils.py                    |  3 ++
 transformers/modeling_t5.py                   | 21 +++++++--
 transformers/modeling_tf_t5.py                | 43 ++++++++++++-------
 transformers/modeling_tf_utils.py             |  6 +--
 transformers/modeling_utils.py                | 12 +++++-
 6 files changed, 65 insertions(+), 31 deletions(-)

diff --git a/transformers/convert_pytorch_checkpoint_to_tf2.py b/transformers/convert_pytorch_checkpoint_to_tf2.py
index 76d75b43e4..4a9832f123 100644
--- a/transformers/convert_pytorch_checkpoint_to_tf2.py
+++ b/transformers/convert_pytorch_checkpoint_to_tf2.py
@@ -120,24 +120,21 @@ def convert_pt_checkpoint_to_tf(model_type, pytorch_checkpoint_path, config_file
     tf_model = load_pytorch_checkpoint_in_tf2_model(tf_model, pytorch_checkpoint_path)
 
     if compare_with_pt_model:
-        inputs_list = [[7, 6, 0, 0, 1], [1, 2, 3, 0, 0], [0, 0, 0, 4, 5]]
-        tf_inputs = tf_model.dummy_inputs
-        tfo = tf_model(tf_inputs, training=False)  # build the network
+        tfo = tf_model(tf_model.dummy_inputs, training=False)  # build the network
 
         state_dict = torch.load(pytorch_checkpoint_path, map_location='cpu')
         pt_model = pt_model_class.from_pretrained(pretrained_model_name_or_path=None,
                                                   config=config,
                                                   state_dict=state_dict)
 
-        pt_inputs = torch.tensor(inputs_list)
         with torch.no_grad():
-            pto = pt_model(pt_inputs)
+            pto = pt_model(**pt_model.dummy_inputs)
 
-        np_pt = pto[0].detach().numpy()
+        np_pt = pto[0].numpy()
         np_tf = tfo[0].numpy()
         diff = np.amax(np.abs(np_pt - np_tf))
         print("Max absolute difference between models outputs {}".format(diff))
-        assert diff <= 2e-2, "Error, model absolute difference is >2e-2"
+        assert diff <= 2e-2, "Error, model absolute difference is >2e-2: {}".format(diff)
 
     # Save pytorch-model
     print("Save TensorFlow model to {}".format(tf_dump_path))
diff --git a/transformers/file_utils.py b/transformers/file_utils.py
index 24abd60781..e36bbf4eeb 100644
--- a/transformers/file_utils.py
+++ b/transformers/file_utils.py
@@ -73,6 +73,9 @@ TF2_WEIGHTS_NAME = 'tf_model.h5'
 TF_WEIGHTS_NAME = 'model.ckpt'
 CONFIG_NAME = "config.json"
 
+DUMMY_INPUTS = [[7, 6, 0, 0, 1], [1, 2, 3, 0, 0], [0, 0, 0, 4, 5]]
+DUMMY_MASK = [[1, 1, 1, 1, 1], [1, 1, 1, 0, 0], [0, 0, 0, 1, 1]]
+
 def is_torch_available():
     return _torch_available
 
diff --git a/transformers/modeling_t5.py b/transformers/modeling_t5.py
index ffc4d8bb3f..149b977abc 100644
--- a/transformers/modeling_t5.py
+++ b/transformers/modeling_t5.py
@@ -32,7 +32,7 @@ from torch.nn import CrossEntropyLoss, MSELoss
 
 from .modeling_utils import PreTrainedModel
 from .configuration_t5 import T5Config
-from .file_utils import add_start_docstrings
+from .file_utils import add_start_docstrings, DUMMY_INPUTS, DUMMY_MASK
 
 logger = logging.getLogger(__name__)
 
@@ -451,6 +451,15 @@ class T5PreTrainedModel(PreTrainedModel):
     load_tf_weights = load_tf_weights_in_t5
     base_model_prefix = "transformer"
 
+    @property
+    def dummy_inputs(self):
+        input_ids = torch.tensor(DUMMY_INPUTS)
+        input_mask = torch.tensor(DUMMY_MASK)
+        dummy_inputs = {'decoder_input_ids': input_ids,
+                        'encoder_input_ids': input_ids,
+                        'decoder_attention_mask': input_mask}
+        return dummy_inputs
+
     def _init_weights(self, module):
         """ Initialize the weights """
         factor = self.config.initializer_factor  # Used for testing weights initialization
@@ -534,9 +543,10 @@ class T5Stack(T5PreTrainedModel):
         # Since we are adding it to the raw scores before the softmax, this is
         # effectively the same as removing these entirely.
 
-        # T5 has a mask that can compare sequence ids, we simulate this here with this transposistion
+        # T5 has a mask that can compare sequence ids, we can simulate this here with this transposition
         # Cf. https://github.com/tensorflow/mesh/blob/8d2465e9bc93129b913b5ccc6a59aa97abd96ec6/mesh_tensorflow/transformer/transformer_layers.py#L270
-        extended_attention_mask = (extended_attention_mask == extended_attention_mask.transpose(-1, -2))
+        # extended_attention_mask = (extended_attention_mask == extended_attention_mask.transpose(-1, -2))
+
         extended_attention_mask = extended_attention_mask.to(dtype=next(self.parameters()).dtype)  # fp16 compatibility
         extended_attention_mask = (1.0 - extended_attention_mask) * -1e9
 
@@ -548,6 +558,10 @@ class T5Stack(T5PreTrainedModel):
             if encoder_attention_mask.dim() == 2:
                 encoder_extended_attention_mask = encoder_attention_mask[:, None, None, :]
 
+            # T5 has a mask that can compare sequence ids, we can simulate this here with this transposition
+            # Cf. https://github.com/tensorflow/mesh/blob/8d2465e9bc93129b913b5ccc6a59aa97abd96ec6/mesh_tensorflow/transformer/transformer_layers.py#L270
+            # encoder_extended_attention_mask = (encoder_extended_attention_mask == encoder_extended_attention_mask.transpose(-1, -2))
+
             encoder_extended_attention_mask = encoder_extended_attention_mask.to(dtype=next(self.parameters()).dtype)  # fp16 compatibility
             encoder_extended_attention_mask = (1.0 - encoder_extended_attention_mask) * -1e9
         else:
@@ -590,6 +604,7 @@ class T5Stack(T5PreTrainedModel):
             hidden_states = layer_outputs[0]
             if i == 0:
                 # We share the position biases between the layers - the first layer store them
+                # layer_outputs = hidden-states, (self-attention weights), (self-attention position bias), (cross-attention weights), (cross-attention position bias)
                 position_bias = layer_outputs[2 if self.output_attentions else 1]
                 if self.is_decoder:
                     encoder_decoder_position_bias = layer_outputs[4 if self.output_attentions else 2]
diff --git a/transformers/modeling_tf_t5.py b/transformers/modeling_tf_t5.py
index 0b3b1116f2..fd25328ac6 100644
--- a/transformers/modeling_tf_t5.py
+++ b/transformers/modeling_tf_t5.py
@@ -26,7 +26,7 @@ import tensorflow as tf
 
 from .configuration_t5 import T5Config
 from .modeling_tf_utils import TFPreTrainedModel, TFSharedEmbeddings, shape_list
-from .file_utils import add_start_docstrings
+from .file_utils import add_start_docstrings, DUMMY_INPUTS, DUMMY_MASK
 
 logger = logging.getLogger(__name__)
 
@@ -61,7 +61,7 @@ class TFT5LayerNorm(tf.keras.layers.Layer):
         super(TFT5LayerNorm, self).build(input_shape)
 
     def call(self, x):
-        variance = tf.math.reduce_min(tf.math.square(x), axis=-1, keepdims=True)
+        variance = tf.math.reduce_mean(tf.math.square(x), axis=-1, keepdims=True)
         x = x * tf.math.rsqrt(variance + self.variance_epsilon)
         return self.weight * x
 
@@ -231,19 +231,19 @@ class TFT5Attention(tf.keras.layers.Layer):
             cache[self.layer_id] = (k, v)
 
         # q = q / math.sqrt(dim_per_head)                                     # No scaling in T5
-        scores = tf.matmul(q, k, transpose_b=True)                            # (bs, n_heads, qlen, klen)
+        # scores = tf.matmul(q, k, transpose_b=True)                            # (bs, n_heads, qlen, klen)
+        scores = tf.einsum('bnqd,bnkd->bnqk', q, k)                        # (bs, n_heads, qlen, klen)
 
         if position_bias is None:
             if not self.has_relative_attention_bias:
                 raise ValueError("No position_bias provided and no weights to compute position_bias")
             position_bias = self.compute_bias(qlen, klen)
+            if mask is not None:
+                position_bias = position_bias + mask
+                # mask = (mask == 0).expand_as(scores)                              # (bs, n_heads, qlen, klen)
+                # scores.masked_fill_(mask, -float('inf'))                          # (bs, n_heads, qlen, klen)
+
         scores += position_bias
-
-        if mask is not None:
-            scores += mask
-            # mask = (mask == 0).expand_as(scores)                              # (bs, n_heads, qlen, klen)
-            # scores.masked_fill_(mask, -float('inf'))                          # (bs, n_heads, qlen, klen)
-
         weights = tf.nn.softmax(scores, axis=-1)                              # (bs, n_heads, qlen, klen)
         weights = self.dropout(weights, training=training)                    # (bs, n_heads, qlen, klen)
 
@@ -350,11 +350,11 @@ class TFT5Block(tf.keras.layers.Layer):
                                                     head_mask=head_mask,
                                                     training=training)
             hidden_states = cross_attention_outputs[0]
-            outputs = cross_attention_outputs[1:] + outputs
+            outputs = outputs + cross_attention_outputs[1:]
             hidden_states = self.layer[2](hidden_states, training=training)
 
         outputs = (hidden_states,) + outputs  # add attentions if we output them
-        return outputs
+        return outputs # hidden-states, (self-attention weights), (self-attention position bias), (cross-attention weights), (cross-attention position bias)
 
 
 ####################################################
@@ -418,7 +418,13 @@ class TFT5MainLayer(tf.keras.layers.Layer):
         # positions we want to attend and -10000.0 for masked positions.
         # Since we are adding it to the raw scores before the softmax, this is
         # effectively the same as removing these entirely.
-        extended_attention_mask = (1.0 - extended_attention_mask) * -10000.0
+
+        # T5 has a mask that can compare sequence ids, we can simulate this here with this transposistion
+        # Cf. https://github.com/tensorflow/mesh/blob/8d2465e9bc93129b913b5ccc6a59aa97abd96ec6/mesh_tensorflow/transformer/transformer_layers.py#L270
+        # extended_attention_mask = tf.math.equal(extended_attention_mask,
+        #                                         tf.transpose(extended_attention_mask, perm=(-1, -2)))
+
+        extended_attention_mask = (1.0 - extended_attention_mask) * -1e9
 
         if self.is_decoder:
             # If a 2D ou 3D attention mask is provided for the cross-attention
@@ -430,7 +436,12 @@ class TFT5MainLayer(tf.keras.layers.Layer):
             if num_dims_encoder_attention_mask == 2:
                 encoder_extended_attention_mask = encoder_attention_mask[:, None, None, :]
 
-            encoder_extended_attention_mask = (1.0 - encoder_extended_attention_mask) * -10000.0
+            # T5 has a mask that can compare sequence ids, we can simulate this here with this transposistion
+            # Cf. https://github.com/tensorflow/mesh/blob/8d2465e9bc93129b913b5ccc6a59aa97abd96ec6/mesh_tensorflow/transformer/transformer_layers.py#L270
+            # encoder_extended_attention_mask = tf.math.equal(encoder_extended_attention_mask,
+            #                                         tf.transpose(encoder_extended_attention_mask, perm=(-1, -2)))
+
+            encoder_extended_attention_mask = (1.0 - encoder_extended_attention_mask) * -1e9
         else:
             encoder_extended_attention_mask = None
 
@@ -463,6 +474,8 @@ class TFT5MainLayer(tf.keras.layers.Layer):
                                          training=training)
             hidden_states = layer_outputs[0]
             if i == 0:
+                # We share the position biases between the layers - the first layer store them
+                # layer_outputs = hidden-states, (self-attention weights), (self-attention position bias), (cross-attention weights), (cross-attention position bias)
                 position_bias = layer_outputs[2 if self.output_attentions else 1]
                 if self.is_decoder:
                     encoder_decoder_position_bias = layer_outputs[4 if self.output_attentions else 2]
@@ -502,8 +515,8 @@ class TFT5PreTrainedModel(TFPreTrainedModel):
 
     @property
     def dummy_inputs(self):
-        input_ids = tf.constant([[7, 6, 0, 0, 1], [1, 2, 3, 0, 0], [0, 0, 0, 4, 5]])
-        input_mask = tf.constant([[1, 1, 0, 0, 1], [1, 1, 1, 0, 0], [1, 0, 0, 1, 1]])
+        input_ids = tf.constant(DUMMY_INPUTS)
+        input_mask = tf.constant(DUMMY_MASK)
         dummy_inputs = {'decoder_input_ids': input_ids,
                         'encoder_input_ids': input_ids,
                         'decoder_attention_mask': input_mask}
diff --git a/transformers/modeling_tf_utils.py b/transformers/modeling_tf_utils.py
index ed8fdb74c9..8d010e589e 100644
--- a/transformers/modeling_tf_utils.py
+++ b/transformers/modeling_tf_utils.py
@@ -24,13 +24,11 @@ import os
 import tensorflow as tf
 
 from .configuration_utils import PretrainedConfig
-from .file_utils import cached_path, WEIGHTS_NAME, TF_WEIGHTS_NAME, TF2_WEIGHTS_NAME
+from .file_utils import cached_path, WEIGHTS_NAME, TF_WEIGHTS_NAME, TF2_WEIGHTS_NAME, DUMMY_INPUTS
 from .modeling_tf_pytorch_utils import load_pytorch_checkpoint_in_tf2_model
 
 logger = logging.getLogger(__name__)
 
-DUMMY_INPUTS = [[7, 6, 0, 0, 1], [1, 2, 3, 0, 0], [0, 0, 0, 4, 5]]
-
 class TFPreTrainedModel(tf.keras.Model):
     r""" Base class for all TF models.
 
@@ -59,7 +57,7 @@ class TFPreTrainedModel(tf.keras.Model):
         Returns:
             tf.Tensor with dummy inputs
         """
-        return tf.constant(DUMMY_INPUTS)
+        return {'input_ids': tf.constant(DUMMY_INPUTS)}
 
     def __init__(self, config, *inputs, **kwargs):
         super(TFPreTrainedModel, self).__init__(*inputs, **kwargs)
diff --git a/transformers/modeling_utils.py b/transformers/modeling_utils.py
index aa0e0e6191..ae515d6870 100644
--- a/transformers/modeling_utils.py
+++ b/transformers/modeling_utils.py
@@ -31,11 +31,10 @@ from torch.nn import CrossEntropyLoss
 from torch.nn import functional as F
 
 from .configuration_utils import PretrainedConfig
-from .file_utils import cached_path, WEIGHTS_NAME, TF_WEIGHTS_NAME, TF2_WEIGHTS_NAME
+from .file_utils import cached_path, WEIGHTS_NAME, TF_WEIGHTS_NAME, TF2_WEIGHTS_NAME, DUMMY_INPUTS
 
 logger = logging.getLogger(__name__)
 
-
 try:
     from torch.nn import Identity
 except ImportError:
@@ -71,6 +70,15 @@ class PreTrainedModel(nn.Module):
     load_tf_weights = lambda model, config, path: None
     base_model_prefix = ""
 
+    @property
+    def dummy_inputs(self):
+        """ Dummy inputs to do a forward pass in the network.
+
+        Returns:
+            torch.Tensor with dummy inputs
+        """
+        return {'input_ids': torch.tensor(DUMMY_INPUTS)}
+
     def __init__(self, config, *inputs, **kwargs):
         super(PreTrainedModel, self).__init__()
         if not isinstance(config, PretrainedConfig):

From b040bff6df09923870f44fb5402e895d57327e85 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Wed, 11 Dec 2019 14:13:58 +0100
Subject: [PATCH 313/505] Added supported model to AutoModelTokenClassification

---
 transformers/__init__.py         |   4 +-
 transformers/modeling_auto.py    | 123 ++++++++++++++++++++++++++++++-
 transformers/modeling_tf_auto.py | 111 +++++++++++++++++++++++++++-
 3 files changed, 231 insertions(+), 7 deletions(-)

diff --git a/transformers/__init__.py b/transformers/__init__.py
index 4300409257..c474696062 100755
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -69,7 +69,7 @@ from .configuration_camembert import CamembertConfig, CAMEMBERT_PRETRAINED_CONFI
 if is_torch_available():
     from .modeling_utils import (PreTrainedModel, prune_layer, Conv1D)
     from .modeling_auto import (AutoModel, AutoModelForSequenceClassification, AutoModelForQuestionAnswering,
-                                AutoModelWithLMHead)
+                                AutoModelWithLMHead, AutoModelForTokenClassification)
 
     from .modeling_bert import (BertPreTrainedModel, BertModel, BertForPreTraining,
                                 BertForMaskedLM, BertForNextSentencePrediction,
@@ -124,7 +124,7 @@ if is_torch_available():
 if is_tf_available():
     from .modeling_tf_utils import TFPreTrainedModel, TFSharedEmbeddings, TFSequenceSummary, shape_list
     from .modeling_tf_auto import (TFAutoModel, TFAutoModelForSequenceClassification, TFAutoModelForQuestionAnswering,
-                                   TFAutoModelWithLMHead)
+                                   TFAutoModelWithLMHead, TFAutoModelForTokenClassification)
 
     from .modeling_tf_bert import (TFBertPreTrainedModel, TFBertMainLayer, TFBertEmbeddings,
                                    TFBertModel, TFBertForPreTraining,
diff --git a/transformers/modeling_auto.py b/transformers/modeling_auto.py
index 041115cc61..cb877643ab 100644
--- a/transformers/modeling_auto.py
+++ b/transformers/modeling_auto.py
@@ -22,16 +22,19 @@ from .configuration_auto import (AlbertConfig, BertConfig, CamembertConfig, CTRL
                                  DistilBertConfig, GPT2Config, OpenAIGPTConfig, RobertaConfig,
                                  TransfoXLConfig, XLMConfig, XLNetConfig)
 
-from .modeling_bert import BertModel, BertForMaskedLM, BertForSequenceClassification, BertForQuestionAnswering
+from .modeling_bert import BertModel, BertForMaskedLM, BertForSequenceClassification, BertForQuestionAnswering, \
+    BertForTokenClassification
 from .modeling_openai import OpenAIGPTModel, OpenAIGPTLMHeadModel
 from .modeling_gpt2 import GPT2Model, GPT2LMHeadModel
 from .modeling_ctrl import CTRLModel, CTRLLMHeadModel
 from .modeling_transfo_xl import TransfoXLModel, TransfoXLLMHeadModel
-from .modeling_xlnet import XLNetModel, XLNetLMHeadModel, XLNetForSequenceClassification, XLNetForQuestionAnswering
+from .modeling_xlnet import XLNetModel, XLNetLMHeadModel, XLNetForSequenceClassification, XLNetForQuestionAnswering, \
+    XLNetForTokenClassification
 from .modeling_xlm import XLMModel, XLMWithLMHeadModel, XLMForSequenceClassification, XLMForQuestionAnswering
 from .modeling_roberta import RobertaModel, RobertaForMaskedLM, RobertaForSequenceClassification
 from .modeling_distilbert import DistilBertModel, DistilBertForQuestionAnswering, DistilBertForMaskedLM, DistilBertForSequenceClassification
-from .modeling_camembert import CamembertModel, CamembertForMaskedLM, CamembertForSequenceClassification, CamembertForMultipleChoice
+from .modeling_camembert import CamembertModel, CamembertForMaskedLM, CamembertForSequenceClassification, \
+    CamembertForMultipleChoice, CamembertForTokenClassification
 from .modeling_albert import AlbertModel, AlbertForMaskedLM, AlbertForSequenceClassification, AlbertForQuestionAnswering
 
 from .modeling_utils import PreTrainedModel, SequenceSummary
@@ -699,3 +702,117 @@ class AutoModelForQuestionAnswering(object):
 
         raise ValueError("Unrecognized model identifier in {}. Should contains one of "
                          "'bert', 'xlnet', 'xlm', 'distilbert', 'albert'".format(pretrained_model_name_or_path))
+
+
+class AutoModelForTokenClassification:
+    def __init__(self):
+        raise EnvironmentError("AutoModelForTokenClassification is designed to be instantiated "
+                               "using the `AutoModelForTokenClassification.from_pretrained(pretrained_model_name_or_path)` or "
+                               "`AutoModelForTokenClassification.from_config(config)` methods.")
+
+    @classmethod
+    def from_config(cls, config):
+        r""" Instantiates one of the base model classes of the library
+        from a configuration.
+    
+            config: (`optional`) instance of a class derived from :class:`~transformers.PretrainedConfig`:
+                The model class to instantiate is selected based on the configuration class:
+                    - isInstance of `distilbert` configuration class: DistilBertModel (DistilBERT model)
+                    - isInstance of `bert` configuration class: BertModel (Bert model)
+                    - isInstance of `xlnet` configuration class: XLNetModel (XLNet model)
+                    - isInstance of `xlm` configuration class: XLMModel (XLM model)
+    
+        Examples::
+    
+            config = BertConfig.from_pretrained('bert-base-uncased')    # Download configuration from S3 and cache.
+            model = AutoModelForTokenClassification.from_config(config)  # E.g. model was saved using `save_pretrained('./test/saved_model/')`
+        """
+        if isinstance(config, CamembertConfig):
+            return CamembertForTokenClassification(config)
+        elif isinstance(config, BertConfig):
+            return BertForTokenClassification(config)
+        elif isinstance(config, XLNetConfig):
+            return XLNetForTokenClassification(config)
+        raise ValueError("Unrecognized configuration class {}".format(config))
+    
+    @classmethod
+    def from_pretrained(cls, pretrained_model_name_or_path, *model_args, **kwargs):
+        r""" Instantiates one of the question answering model classes of the library
+        from a pre-trained model configuration.
+
+        The `from_pretrained()` method takes care of returning the correct model class instance
+        using pattern matching on the `pretrained_model_name_or_path` string.
+
+        The model class to instantiate is selected as the first pattern matching
+        in the `pretrained_model_name_or_path` string (in the following order):
+            - contains `distilbert`: DistilBertForTokenClassification (DistilBERT model)
+            - contains `albert`: AlbertForTokenClassification (ALBERT model)
+            - contains `bert`: BertForTokenClassification (Bert model)
+            - contains `xlnet`: XLNetForTokenClassification (XLNet model)
+            - contains `xlm`: XLMForTokenClassification (XLM model)
+
+        The model is set in evaluation mode by default using `model.eval()` (Dropout modules are deactivated)
+        To train the model, you should first set it back in training mode with `model.train()`
+
+        Params:
+            pretrained_model_name_or_path: either:
+
+                - a string with the `shortcut name` of a pre-trained model to load from cache or download, e.g.: ``bert-base-uncased``.
+                - a path to a `directory` containing model weights saved using :func:`~transformers.PreTrainedModel.save_pretrained`, e.g.: ``./my_model_directory/``.
+                - a path or url to a `tensorflow index checkpoint file` (e.g. `./tf_model/model.ckpt.index`). In this case, ``from_tf`` should be set to True and a configuration object should be provided as ``config`` argument. This loading path is slower than converting the TensorFlow checkpoint in a PyTorch model using the provided conversion scripts and loading the PyTorch model afterwards.
+
+            model_args: (`optional`) Sequence of positional arguments:
+                All remaning positional arguments will be passed to the underlying model's ``__init__`` method
+
+            config: (`optional`) instance of a class derived from :class:`~transformers.PretrainedConfig`:
+                Configuration for the model to use instead of an automatically loaded configuation. Configuration can be automatically loaded when:
+
+                - the model is a model provided by the library (loaded with the ``shortcut-name`` string of a pretrained model), or
+                - the model was saved using :func:`~transformers.PreTrainedModel.save_pretrained` and is reloaded by suppling the save directory.
+                - the model is loaded by suppling a local directory as ``pretrained_model_name_or_path`` and a configuration JSON file named `config.json` is found in the directory.
+
+            state_dict: (`optional`) dict:
+                an optional state dictionnary for the model to use instead of a state dictionary loaded from saved weights file.
+                This option can be used if you want to create a model from a pretrained configuration but load your own weights.
+                In this case though, you should check if using :func:`~transformers.PreTrainedModel.save_pretrained` and :func:`~transformers.PreTrainedModel.from_pretrained` is not a simpler option.
+
+            cache_dir: (`optional`) string:
+                Path to a directory in which a downloaded pre-trained model
+                configuration should be cached if the standard cache should not be used.
+
+            force_download: (`optional`) boolean, default False:
+                Force to (re-)download the model weights and configuration files and override the cached versions if they exists.
+
+            proxies: (`optional`) dict, default None:
+                A dictionary of proxy servers to use by protocol or endpoint, e.g.: {'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.
+                The proxies are used on each request.
+
+            output_loading_info: (`optional`) boolean:
+                Set to ``True`` to also return a dictionnary containing missing keys, unexpected keys and error messages.
+
+            kwargs: (`optional`) Remaining dictionary of keyword arguments:
+                Can be used to update the configuration object (after it being loaded) and initiate the model. (e.g. ``output_attention=True``). Behave differently depending on whether a `config` is provided or automatically loaded:
+
+                - If a configuration is provided with ``config``, ``**kwargs`` will be directly passed to the underlying model's ``__init__`` method (we assume all relevant updates to the configuration have already been done)
+                - If a configuration is not provided, ``kwargs`` will be first passed to the configuration class initialization function (:func:`~transformers.PretrainedConfig.from_pretrained`). Each key of ``kwargs`` that corresponds to a configuration attribute will be used to override said attribute with the supplied ``kwargs`` value. Remaining keys that do not correspond to any configuration attribute will be passed to the underlying model's ``__init__`` function.
+
+        Examples::
+
+            model = AutoModelForTokenClassification.from_pretrained('bert-base-uncased')    # Download model and configuration from S3 and cache.
+            model = AutoModelForTokenClassification.from_pretrained('./test/bert_model/')  # E.g. model was saved using `save_pretrained('./test/saved_model/')`
+            model = AutoModelForTokenClassification.from_pretrained('bert-base-uncased', output_attention=True)  # Update configuration during loading
+            assert model.config.output_attention == True
+            # Loading from a TF checkpoint file instead of a PyTorch model (slower)
+            config = AutoConfig.from_json_file('./tf_model/bert_tf_model_config.json')
+            model = AutoModelForTokenClassification.from_pretrained('./tf_model/bert_tf_checkpoint.ckpt.index', from_tf=True, config=config)
+
+        """
+        if 'camembert' in pretrained_model_name_or_path:
+            return CamembertForTokenClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'bert' in pretrained_model_name_or_path:
+            return BertForTokenClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'xlnet' in pretrained_model_name_or_path:
+            return XLNetForTokenClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+
+        raise ValueError("Unrecognized model identifier in {}. Should contains one of "
+                         "'bert', 'xlnet', 'camembert'".format(pretrained_model_name_or_path))
diff --git a/transformers/modeling_tf_auto.py b/transformers/modeling_tf_auto.py
index e78b91cfcc..1097f77a59 100644
--- a/transformers/modeling_tf_auto.py
+++ b/transformers/modeling_tf_auto.py
@@ -22,11 +22,13 @@ from .configuration_auto import (BertConfig, CTRLConfig, DistilBertConfig,
                                  GPT2Config, OpenAIGPTConfig, RobertaConfig,
                                  TransfoXLConfig, XLMConfig, XLNetConfig)
 
-from .modeling_tf_bert import TFBertModel, TFBertForMaskedLM, TFBertForSequenceClassification, TFBertForQuestionAnswering
+from .modeling_tf_bert import TFBertModel, TFBertForMaskedLM, TFBertForSequenceClassification, \
+    TFBertForQuestionAnswering, TFBertForTokenClassification
 from .modeling_tf_openai import TFOpenAIGPTModel, TFOpenAIGPTLMHeadModel
 from .modeling_tf_gpt2 import TFGPT2Model, TFGPT2LMHeadModel
 from .modeling_tf_transfo_xl import TFTransfoXLModel, TFTransfoXLLMHeadModel
-from .modeling_tf_xlnet import TFXLNetModel, TFXLNetLMHeadModel, TFXLNetForSequenceClassification, TFXLNetForQuestionAnsweringSimple
+from .modeling_tf_xlnet import TFXLNetModel, TFXLNetLMHeadModel, TFXLNetForSequenceClassification, \
+    TFXLNetForQuestionAnsweringSimple, TFXLNetForTokenClassification
 from .modeling_tf_xlm import TFXLMModel, TFXLMWithLMHeadModel, TFXLMForSequenceClassification, TFXLMForQuestionAnsweringSimple
 from .modeling_tf_roberta import TFRobertaModel, TFRobertaForMaskedLM, TFRobertaForSequenceClassification
 from .modeling_tf_distilbert import TFDistilBertModel, TFDistilBertForQuestionAnswering, TFDistilBertForMaskedLM, TFDistilBertForSequenceClassification
@@ -668,3 +670,108 @@ class TFAutoModelForQuestionAnswering(object):
 
         raise ValueError("Unrecognized model identifier in {}. Should contains one of "
                          "'distilbert', 'bert', 'xlnet', 'xlm'".format(pretrained_model_name_or_path))
+
+
+class TFAutoModelForTokenClassification:
+    def __init__(self):
+        raise EnvironmentError("TFAutoModelForTokenClassification is designed to be instantiated "
+                               "using the `TFAutoModelForTokenClassification.from_pretrained(pretrained_model_name_or_path)` or "
+                               "`AutoModelForTokenClassification.from_config(config)` methods.")
+
+    @classmethod
+    def from_config(cls, config):
+        r""" Instantiates one of the base model classes of the library
+        from a configuration.
+
+            config: (`optional`) instance of a class derived from :class:`~transformers.PretrainedConfig`:
+                The model class to instantiate is selected based on the configuration class:
+                    - isInstance of `bert` configuration class: BertModel (Bert model)
+                    - isInstance of `xlnet` configuration class: XLNetModel (XLNet model)
+
+        Examples::
+
+            config = BertConfig.from_pretrained('bert-base-uncased')    # Download configuration from S3 and cache.
+            model = TFAutoModelForTokenClassification.from_config(config)  # E.g. model was saved using `save_pretrained('./test/saved_model/')`
+        """
+        if isinstance(config, BertConfig):
+            return TFBertForTokenClassification(config)
+        elif isinstance(config, XLNetConfig):
+            return TFXLNetForTokenClassification(config)
+        raise ValueError("Unrecognized configuration class {}".format(config))
+
+    @classmethod
+    def from_pretrained(cls, pretrained_model_name_or_path, *model_args, **kwargs):
+        r""" Instantiates one of the question answering model classes of the library
+        from a pre-trained model configuration.
+
+        The `from_pretrained()` method takes care of returning the correct model class instance
+        using pattern matching on the `pretrained_model_name_or_path` string.
+
+        The model class to instantiate is selected as the first pattern matching
+        in the `pretrained_model_name_or_path` string (in the following order):
+            - contains `bert`: BertForTokenClassification (Bert model)
+            - contains `xlnet`: XLNetForTokenClassification (XLNet model)
+
+        The model is set in evaluation mode by default using `model.eval()` (Dropout modules are deactivated)
+        To train the model, you should first set it back in training mode with `model.train()`
+
+        Params:
+            pretrained_model_name_or_path: either:
+
+                - a string with the `shortcut name` of a pre-trained model to load from cache or download, e.g.: ``bert-base-uncased``.
+                - a path to a `directory` containing model weights saved using :func:`~transformers.PreTrainedModel.save_pretrained`, e.g.: ``./my_model_directory/``.
+                - a path or url to a `tensorflow index checkpoint file` (e.g. `./tf_model/model.ckpt.index`). In this case, ``from_tf`` should be set to True and a configuration object should be provided as ``config`` argument. This loading path is slower than converting the TensorFlow checkpoint in a PyTorch model using the provided conversion scripts and loading the PyTorch model afterwards.
+
+            model_args: (`optional`) Sequence of positional arguments:
+                All remaning positional arguments will be passed to the underlying model's ``__init__`` method
+
+            config: (`optional`) instance of a class derived from :class:`~transformers.PretrainedConfig`:
+                Configuration for the model to use instead of an automatically loaded configuation. Configuration can be automatically loaded when:
+
+                - the model is a model provided by the library (loaded with the ``shortcut-name`` string of a pretrained model), or
+                - the model was saved using :func:`~transformers.PreTrainedModel.save_pretrained` and is reloaded by suppling the save directory.
+                - the model is loaded by suppling a local directory as ``pretrained_model_name_or_path`` and a configuration JSON file named `config.json` is found in the directory.
+
+            state_dict: (`optional`) dict:
+                an optional state dictionnary for the model to use instead of a state dictionary loaded from saved weights file.
+                This option can be used if you want to create a model from a pretrained configuration but load your own weights.
+                In this case though, you should check if using :func:`~transformers.PreTrainedModel.save_pretrained` and :func:`~transformers.PreTrainedModel.from_pretrained` is not a simpler option.
+
+            cache_dir: (`optional`) string:
+                Path to a directory in which a downloaded pre-trained model
+                configuration should be cached if the standard cache should not be used.
+
+            force_download: (`optional`) boolean, default False:
+                Force to (re-)download the model weights and configuration files and override the cached versions if they exists.
+
+            proxies: (`optional`) dict, default None:
+                A dictionary of proxy servers to use by protocol or endpoint, e.g.: {'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.
+                The proxies are used on each request.
+
+            output_loading_info: (`optional`) boolean:
+                Set to ``True`` to also return a dictionnary containing missing keys, unexpected keys and error messages.
+
+            kwargs: (`optional`) Remaining dictionary of keyword arguments:
+                Can be used to update the configuration object (after it being loaded) and initiate the model. (e.g. ``output_attention=True``). Behave differently depending on whether a `config` is provided or automatically loaded:
+
+                - If a configuration is provided with ``config``, ``**kwargs`` will be directly passed to the underlying model's ``__init__`` method (we assume all relevant updates to the configuration have already been done)
+                - If a configuration is not provided, ``kwargs`` will be first passed to the configuration class initialization function (:func:`~transformers.PretrainedConfig.from_pretrained`). Each key of ``kwargs`` that corresponds to a configuration attribute will be used to override said attribute with the supplied ``kwargs`` value. Remaining keys that do not correspond to any configuration attribute will be passed to the underlying model's ``__init__`` function.
+
+        Examples::
+
+            model = TFAutoModelForTokenClassification.from_pretrained('bert-base-uncased')    # Download model and configuration from S3 and cache.
+            model = TFAutoModelForTokenClassification.from_pretrained('./test/bert_model/')  # E.g. model was saved using `save_pretrained('./test/saved_model/')`
+            model = TFAutoModelForTokenClassification.from_pretrained('bert-base-uncased', output_attention=True)  # Update configuration during loading
+            assert model.config.output_attention == True
+            # Loading from a TF checkpoint file instead of a PyTorch model (slower)
+            config = AutoConfig.from_json_file('./tf_model/bert_tf_model_config.json')
+            model = TFAutoModelForTokenClassification.from_pretrained('./tf_model/bert_tf_checkpoint.ckpt.index', from_tf=True, config=config)
+
+        """
+        if 'bert' in pretrained_model_name_or_path:
+            return TFBertForTokenClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'xlnet' in pretrained_model_name_or_path:
+            return TFXLNetForTokenClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+
+        raise ValueError("Unrecognized model identifier in {}. Should contains one of "
+                         "'bert', 'xlnet'".format(pretrained_model_name_or_path))

From 4c12860f7ae61659aed2675498350a386fc4e122 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Wed, 11 Dec 2019 09:22:37 -0500
Subject: [PATCH 314/505] Remove misleading documentation

---
 transformers/tokenization_utils.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index cb931b0eaf..68a767fe82 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -628,7 +628,6 @@ class PreTrainedTokenizer(object):
             Take care of added tokens.
 
             text: The sequence to be encoded.
-            return_tokens_mapped_to_origin: (optional) Set to True to return the index of each token in the initial whitespace tokenization. (default False).
             **kwargs: passed to the child `self.tokenize()` method
         """
         def lowercase_text(t):

From c28273793ec41c98153b23e29b9c7228c4149aae Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Wed, 11 Dec 2019 14:52:01 +0100
Subject: [PATCH 315/505] Add missing DistilBert and Roberta to
 AutoModelForTokenClassification

---
 transformers/modeling_auto.py    | 25 ++++++++++++++++++-------
 transformers/modeling_tf_auto.py | 18 +++++++++++++++---
 2 files changed, 33 insertions(+), 10 deletions(-)

diff --git a/transformers/modeling_auto.py b/transformers/modeling_auto.py
index cb877643ab..c76e5b78b3 100644
--- a/transformers/modeling_auto.py
+++ b/transformers/modeling_auto.py
@@ -31,8 +31,10 @@ from .modeling_transfo_xl import TransfoXLModel, TransfoXLLMHeadModel
 from .modeling_xlnet import XLNetModel, XLNetLMHeadModel, XLNetForSequenceClassification, XLNetForQuestionAnswering, \
     XLNetForTokenClassification
 from .modeling_xlm import XLMModel, XLMWithLMHeadModel, XLMForSequenceClassification, XLMForQuestionAnswering
-from .modeling_roberta import RobertaModel, RobertaForMaskedLM, RobertaForSequenceClassification
-from .modeling_distilbert import DistilBertModel, DistilBertForQuestionAnswering, DistilBertForMaskedLM, DistilBertForSequenceClassification
+from .modeling_roberta import RobertaModel, RobertaForMaskedLM, RobertaForSequenceClassification, \
+    RobertaForTokenClassification
+from .modeling_distilbert import DistilBertModel, DistilBertForQuestionAnswering, DistilBertForMaskedLM, \
+    DistilBertForSequenceClassification, DistilBertForTokenClassification
 from .modeling_camembert import CamembertModel, CamembertForMaskedLM, CamembertForSequenceClassification, \
     CamembertForMultipleChoice, CamembertForTokenClassification
 from .modeling_albert import AlbertModel, AlbertForMaskedLM, AlbertForSequenceClassification, AlbertForQuestionAnswering
@@ -720,8 +722,9 @@ class AutoModelForTokenClassification:
                     - isInstance of `distilbert` configuration class: DistilBertModel (DistilBERT model)
                     - isInstance of `bert` configuration class: BertModel (Bert model)
                     - isInstance of `xlnet` configuration class: XLNetModel (XLNet model)
-                    - isInstance of `xlm` configuration class: XLMModel (XLM model)
-    
+                    - isInstance of `camembert` configuration class: CamembertModel (Camembert model)
+                    - isInstance of `roberta` configuration class: RobertaModel (Roberta model)
+
         Examples::
     
             config = BertConfig.from_pretrained('bert-base-uncased')    # Download configuration from S3 and cache.
@@ -729,10 +732,14 @@ class AutoModelForTokenClassification:
         """
         if isinstance(config, CamembertConfig):
             return CamembertForTokenClassification(config)
+        elif isinstance(config, DistilBertConfig):
+            return DistilBertForTokenClassification(config)
         elif isinstance(config, BertConfig):
             return BertForTokenClassification(config)
         elif isinstance(config, XLNetConfig):
             return XLNetForTokenClassification(config)
+        elif isinstance(config, RobertaConfig):
+            return RobertaForTokenClassification(config)
         raise ValueError("Unrecognized configuration class {}".format(config))
     
     @classmethod
@@ -746,10 +753,10 @@ class AutoModelForTokenClassification:
         The model class to instantiate is selected as the first pattern matching
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `distilbert`: DistilBertForTokenClassification (DistilBERT model)
-            - contains `albert`: AlbertForTokenClassification (ALBERT model)
+            - contains `camembert`: CamembertForTokenClassification (Camembert model)
             - contains `bert`: BertForTokenClassification (Bert model)
             - contains `xlnet`: XLNetForTokenClassification (XLNet model)
-            - contains `xlm`: XLMForTokenClassification (XLM model)
+            - contains `roberta`: RobertaForTokenClassification (Roberta model)
 
         The model is set in evaluation mode by default using `model.eval()` (Dropout modules are deactivated)
         To train the model, you should first set it back in training mode with `model.train()`
@@ -809,10 +816,14 @@ class AutoModelForTokenClassification:
         """
         if 'camembert' in pretrained_model_name_or_path:
             return CamembertForTokenClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'distilbert' in pretrained_model_name_or_path:
+            return DistilBertForTokenClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'bert' in pretrained_model_name_or_path:
             return BertForTokenClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'roberta' in pretrained_model_name_or_path:
+            return RobertaForTokenClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'xlnet' in pretrained_model_name_or_path:
             return XLNetForTokenClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
 
         raise ValueError("Unrecognized model identifier in {}. Should contains one of "
-                         "'bert', 'xlnet', 'camembert'".format(pretrained_model_name_or_path))
+                         "'bert', 'xlnet', 'camembert', 'distilbert', 'roberta'".format(pretrained_model_name_or_path))
diff --git a/transformers/modeling_tf_auto.py b/transformers/modeling_tf_auto.py
index 1097f77a59..add7e03341 100644
--- a/transformers/modeling_tf_auto.py
+++ b/transformers/modeling_tf_auto.py
@@ -30,8 +30,8 @@ from .modeling_tf_transfo_xl import TFTransfoXLModel, TFTransfoXLLMHeadModel
 from .modeling_tf_xlnet import TFXLNetModel, TFXLNetLMHeadModel, TFXLNetForSequenceClassification, \
     TFXLNetForQuestionAnsweringSimple, TFXLNetForTokenClassification
 from .modeling_tf_xlm import TFXLMModel, TFXLMWithLMHeadModel, TFXLMForSequenceClassification, TFXLMForQuestionAnsweringSimple
-from .modeling_tf_roberta import TFRobertaModel, TFRobertaForMaskedLM, TFRobertaForSequenceClassification
-from .modeling_tf_distilbert import TFDistilBertModel, TFDistilBertForQuestionAnswering, TFDistilBertForMaskedLM, TFDistilBertForSequenceClassification
+from .modeling_tf_roberta import TFRobertaModel, TFRobertaForMaskedLM, TFRobertaForSequenceClassification, TFRobertaForTokenClassification
+from .modeling_tf_distilbert import TFDistilBertModel, TFDistilBertForQuestionAnswering, TFDistilBertForMaskedLM, TFDistilBertForSequenceClassification, TFDistilBertForTokenClassification
 from .modeling_tf_ctrl import TFCTRLModel, TFCTRLLMHeadModel
 
 from .file_utils import add_start_docstrings
@@ -687,6 +687,8 @@ class TFAutoModelForTokenClassification:
                 The model class to instantiate is selected based on the configuration class:
                     - isInstance of `bert` configuration class: BertModel (Bert model)
                     - isInstance of `xlnet` configuration class: XLNetModel (XLNet model)
+                    - isInstance of `distilbert` configuration class: DistilBertModel (DistilBert model)
+                    - isInstance of `roberta` configuration class: RobteraModel (Roberta model)
 
         Examples::
 
@@ -697,6 +699,10 @@ class TFAutoModelForTokenClassification:
             return TFBertForTokenClassification(config)
         elif isinstance(config, XLNetConfig):
             return TFXLNetForTokenClassification(config)
+        elif isinstance(config, DistilBertConfig):
+            return TFDistilBertForTokenClassification(config)
+        elif isinstance(config, RobertaConfig):
+            return TFRobertaForTokenClassification(config)
         raise ValueError("Unrecognized configuration class {}".format(config))
 
     @classmethod
@@ -711,6 +717,8 @@ class TFAutoModelForTokenClassification:
         in the `pretrained_model_name_or_path` string (in the following order):
             - contains `bert`: BertForTokenClassification (Bert model)
             - contains `xlnet`: XLNetForTokenClassification (XLNet model)
+            - contains `distilbert`: DistilBertForTokenClassification (DistilBert model)
+            - contains `roberta`: RobertaForTokenClassification (Roberta model)
 
         The model is set in evaluation mode by default using `model.eval()` (Dropout modules are deactivated)
         To train the model, you should first set it back in training mode with `model.train()`
@@ -772,6 +780,10 @@ class TFAutoModelForTokenClassification:
             return TFBertForTokenClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'xlnet' in pretrained_model_name_or_path:
             return TFXLNetForTokenClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'distilbert' in pretrained_model_name_or_path:
+            return TFDistilBertForTokenClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'roberta' in pretrained_model_name_or_path:
+            return TFRobertaForTokenClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
 
         raise ValueError("Unrecognized model identifier in {}. Should contains one of "
-                         "'bert', 'xlnet'".format(pretrained_model_name_or_path))
+                         "'bert', 'xlnet', 'distilbert', 'roberta'".format(pretrained_model_name_or_path))

From 2e2f9fed554bb5f147ea3d9573004b447dd7c9e7 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Wed, 11 Dec 2019 11:11:56 -0500
Subject: [PATCH 316/505] rm duplicate imports

---
 transformers/modeling_auto.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/transformers/modeling_auto.py b/transformers/modeling_auto.py
index b63e43d73b..6ba1aab7a3 100644
--- a/transformers/modeling_auto.py
+++ b/transformers/modeling_auto.py
@@ -28,7 +28,6 @@ from .modeling_xlm import XLMModel, XLMWithLMHeadModel, XLMForSequenceClassifica
 from .modeling_roberta import RobertaModel, RobertaForMaskedLM, RobertaForSequenceClassification
 from .modeling_distilbert import DistilBertModel, DistilBertForQuestionAnswering, DistilBertForMaskedLM, DistilBertForSequenceClassification
 from .modeling_camembert import CamembertModel, CamembertForMaskedLM, CamembertForSequenceClassification, CamembertForMultipleChoice
-from .modeling_camembert import CamembertModel, CamembertForMaskedLM, CamembertForSequenceClassification, CamembertForMultipleChoice
 from .modeling_albert import AlbertModel, AlbertForMaskedLM, AlbertForSequenceClassification, AlbertForQuestionAnswering
 
 from .modeling_utils import PreTrainedModel, SequenceSummary

From 29570db25ba9dd30e5ac9be68dbcad95434964ec Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Wed, 11 Dec 2019 17:19:18 +0100
Subject: [PATCH 317/505] allowing from_pretrained to load from url directly

---
 transformers/modeling_tf_utils.py | 4 +++-
 transformers/modeling_utils.py    | 7 +++++--
 2 files changed, 8 insertions(+), 3 deletions(-)

diff --git a/transformers/modeling_tf_utils.py b/transformers/modeling_tf_utils.py
index ed8fdb74c9..e7512b5bd6 100644
--- a/transformers/modeling_tf_utils.py
+++ b/transformers/modeling_tf_utils.py
@@ -259,8 +259,10 @@ class TFPreTrainedModel(tf.keras.Model):
                         pretrained_model_name_or_path))
             elif os.path.isfile(pretrained_model_name_or_path):
                 archive_file = pretrained_model_name_or_path
+            elif os.path.isfile(pretrained_model_name_or_path + ".index"):
+                archive_file = pretrained_model_name_or_path + ".index"
             else:
-                raise EnvironmentError("Error file {} not found".format(pretrained_model_name_or_path))
+                archive_file = pretrained_model_name_or_path
 
             # redirect to the cache, if necessary
             try:
diff --git a/transformers/modeling_utils.py b/transformers/modeling_utils.py
index 3ac568771e..9e7ca8d689 100644
--- a/transformers/modeling_utils.py
+++ b/transformers/modeling_utils.py
@@ -365,9 +365,12 @@ class PreTrainedModel(nn.Module):
                         pretrained_model_name_or_path))
             elif os.path.isfile(pretrained_model_name_or_path):
                 archive_file = pretrained_model_name_or_path
-            else:
-                assert from_tf, "Error finding file {}, no file or TF 1.X checkpoint found".format(pretrained_model_name_or_path)
+            elif os.path.isfile(pretrained_model_name_or_path + ".index"):
+                assert from_tf, "We found a TensorFlow checkpoint at {}, please set from_tf to True to load from this checkpoint".format(
+                    pretrained_model_name_or_path + ".index")
                 archive_file = pretrained_model_name_or_path + ".index"
+            else:
+                archive_file = pretrained_model_name_or_path
 
             # redirect to the cache, if necessary
             try:

From 6709739a05ca8b271a629ffebb497352449b7935 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Wed, 11 Dec 2019 17:19:18 +0100
Subject: [PATCH 318/505] allowing from_pretrained to load from url directly

---
 transformers/modeling_tf_utils.py | 4 +++-
 transformers/modeling_utils.py    | 7 +++++--
 2 files changed, 8 insertions(+), 3 deletions(-)

diff --git a/transformers/modeling_tf_utils.py b/transformers/modeling_tf_utils.py
index 6c48f3eed2..95c29693d8 100644
--- a/transformers/modeling_tf_utils.py
+++ b/transformers/modeling_tf_utils.py
@@ -265,8 +265,10 @@ class TFPreTrainedModel(tf.keras.Model):
                         pretrained_model_name_or_path))
             elif os.path.isfile(pretrained_model_name_or_path):
                 archive_file = pretrained_model_name_or_path
+            elif os.path.isfile(pretrained_model_name_or_path + ".index"):
+                archive_file = pretrained_model_name_or_path + ".index"
             else:
-                raise EnvironmentError("Error file {} not found".format(pretrained_model_name_or_path))
+                archive_file = pretrained_model_name_or_path
 
             # redirect to the cache, if necessary
             try:
diff --git a/transformers/modeling_utils.py b/transformers/modeling_utils.py
index 398172a88c..eec9034fd7 100644
--- a/transformers/modeling_utils.py
+++ b/transformers/modeling_utils.py
@@ -364,9 +364,12 @@ class PreTrainedModel(nn.Module):
                         pretrained_model_name_or_path))
             elif os.path.isfile(pretrained_model_name_or_path):
                 archive_file = pretrained_model_name_or_path
-            else:
-                assert from_tf, "Error finding file {}, no file or TF 1.X checkpoint found".format(pretrained_model_name_or_path)
+            elif os.path.isfile(pretrained_model_name_or_path + ".index"):
+                assert from_tf, "We found a TensorFlow checkpoint at {}, please set from_tf to True to load from this checkpoint".format(
+                    pretrained_model_name_or_path + ".index")
                 archive_file = pretrained_model_name_or_path + ".index"
+            else:
+                archive_file = pretrained_model_name_or_path
 
             # redirect to the cache, if necessary
             try:

From 030faccb8d45be9bdd2b4b80ff26f36dc41f622a Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan.schweter@bsb-muenchen.de>
Date: Wed, 11 Dec 2019 17:44:21 +0100
Subject: [PATCH 319/505] doc: fix pretrained models table

---
 docs/source/pretrained_models.rst | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/docs/source/pretrained_models.rst b/docs/source/pretrained_models.rst
index dd61f11769..2fe1f8a314 100644
--- a/docs/source/pretrained_models.rst
+++ b/docs/source/pretrained_models.rst
@@ -169,35 +169,35 @@ Here is the full list of the currently provided pretrained models together with
 +-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 | ALBERT            | ``albert-base-v1``                                         | | 12 repeating layers, 128 embedding, 768-hidden, 12-heads, 11M parameters                                                            |
 |                   |                                                            | | ALBERT base model                                                                                                                   |
-|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                             |
+|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                                                         |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``albert-large-v1``                                        | | 24 repeating layers, 128 embedding, 1024-hidden, 16-heads, 17M parameters                                                           |
 |                   |                                                            | | ALBERT large model                                                                                                                  |
-|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                             |
+|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                                                         |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``albert-xlarge-v1``                                       | | 24 repeating layers, 128 embedding, 2048-hidden, 16-heads, 58M parameters                                                           |
 |                   |                                                            | | ALBERT xlarge model                                                                                                                 |
-|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                             |
+|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                                                         |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``albert-xxlarge-v1``                                      | | 12 repeating layer, 128 embedding, 4096-hidden, 64-heads, 223M parameters                                                           |
 |                   |                                                            | | ALBERT xxlarge model                                                                                                                |
-|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                             |
+|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                                                         |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``albert-base-v2``                                         | | 12 repeating layers, 128 embedding, 768-hidden, 12-heads, 11M parameters                                                            |
 |                   |                                                            | | ALBERT base model with no dropout, additional training data and longer training                                                     |
-|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                             |
+|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                                                         |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``albert-large-v2``                                        | | 24 repeating layers, 128 embedding, 1024-hidden, 16-heads, 17M parameters                                                           |
 |                   |                                                            | | ALBERT large model with no dropout, additional training data and longer training                                                    |
-|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                             |
+|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                                                         |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``albert-xlarge-v2``                                       | | 24 repeating layers, 128 embedding, 2048-hidden, 16-heads, 58M parameters                                                           |
 |                   |                                                            | | ALBERT xlarge model with no dropout, additional training data and longer training                                                   |
-|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                             |
+|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                                                         |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``albert-xxlarge-v2``                                      | | 12 repeating layer, 128 embedding, 4096-hidden, 64-heads, 223M parameters                                                           |
 |                   |                                                            | | ALBERT xxlarge model with no dropout, additional training data and longer training                                                  |
-|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                             |
+|                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                                                         |
 +-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 
 

From c999a3e5050f1dc93d814abf352f3bf0c06572e7 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Wed, 11 Dec 2019 12:29:58 -0500
Subject: [PATCH 320/505] Allow from_pretrained to take a remote identifier

---
 transformers/configuration_utils.py |  8 +++++---
 transformers/file_utils.py          | 20 ++++++++++++++++----
 transformers/modeling_utils.py      |  8 +++++---
 transformers/tokenization_utils.py  | 10 +++++-----
 4 files changed, 31 insertions(+), 15 deletions(-)

diff --git a/transformers/configuration_utils.py b/transformers/configuration_utils.py
index 08cee75d81..8ae30f2a48 100644
--- a/transformers/configuration_utils.py
+++ b/transformers/configuration_utils.py
@@ -24,7 +24,7 @@ import logging
 import os
 from io import open
 
-from .file_utils import cached_path, CONFIG_NAME
+from .file_utils import CONFIG_NAME, cached_path, is_remote_url, hf_bucket_url
 
 logger = logging.getLogger(__name__)
 
@@ -131,8 +131,10 @@ class PretrainedConfig(object):
             config_file = cls.pretrained_config_archive_map[pretrained_model_name_or_path]
         elif os.path.isdir(pretrained_model_name_or_path):
             config_file = os.path.join(pretrained_model_name_or_path, CONFIG_NAME)
-        else:
+        elif os.path.isfile(pretrained_model_name_or_path) or is_remote_url(pretrained_model_name_or_path):
             config_file = pretrained_model_name_or_path
+        else:
+            config_file = hf_bucket_url(pretrained_model_name_or_path, postfix=CONFIG_NAME)
         # redirect to the cache, if necessary
         try:
             resolved_config_file = cached_path(config_file, cache_dir=cache_dir, force_download=force_download,
@@ -187,7 +189,7 @@ class PretrainedConfig(object):
 
     @classmethod
     def from_json_file(cls, json_file):
-        """Constructs a `BertConfig` from a json file of parameters."""
+        """Constructs a `Config` from a json file of parameters."""
         with open(json_file, "r", encoding='utf-8') as reader:
             text = reader.read()
         return cls.from_dict(json.loads(text))
diff --git a/transformers/file_utils.py b/transformers/file_utils.py
index 68de4e6e2f..5fd5e2ee39 100644
--- a/transformers/file_utils.py
+++ b/transformers/file_utils.py
@@ -73,6 +73,8 @@ TF2_WEIGHTS_NAME = 'tf_model.h5'
 TF_WEIGHTS_NAME = 'model.ckpt'
 CONFIG_NAME = "config.json"
 
+S3_BUCKET_PREFIX = "https://s3.amazonaws.com/models.huggingface.co/bert"
+
 def is_torch_available():
     return _torch_available
 
@@ -103,6 +105,18 @@ else:
             return fn
         return docstring_decorator
 
+
+def is_remote_url(url_or_filename):
+    parsed = urlparse(url_or_filename)
+    return parsed.scheme in ('http', 'https', 's3')
+
+def hf_bucket_url(identifier, postfix=None):
+    if postfix is None:
+        return "/".join((S3_BUCKET_PREFIX, identifier))
+    else:
+        return "/".join((S3_BUCKET_PREFIX, identifier, postfix))
+
+
 def url_to_filename(url, etag=None):
     """
     Convert `url` into a hashed filename in a repeatable way.
@@ -171,9 +185,7 @@ def cached_path(url_or_filename, cache_dir=None, force_download=False, proxies=N
     if sys.version_info[0] == 3 and isinstance(cache_dir, Path):
         cache_dir = str(cache_dir)
 
-    parsed = urlparse(url_or_filename)
-
-    if parsed.scheme in ('http', 'https', 's3'):
+    if is_remote_url(url_or_filename):
         # URL, so get it from the cache (downloading if necessary)
         return get_from_cache(url_or_filename, cache_dir=cache_dir,
             force_download=force_download, proxies=proxies,
@@ -181,7 +193,7 @@ def cached_path(url_or_filename, cache_dir=None, force_download=False, proxies=N
     elif os.path.exists(url_or_filename):
         # File, and it exists.
         return url_or_filename
-    elif parsed.scheme == '':
+    elif urlparse(url_or_filename).scheme == '':
         # File, but it doesn't exist.
         raise EnvironmentError("file {} not found".format(url_or_filename))
     else:
diff --git a/transformers/modeling_utils.py b/transformers/modeling_utils.py
index 9e7ca8d689..eac4252336 100644
--- a/transformers/modeling_utils.py
+++ b/transformers/modeling_utils.py
@@ -31,7 +31,8 @@ from torch.nn import CrossEntropyLoss
 from torch.nn import functional as F
 
 from .configuration_utils import PretrainedConfig
-from .file_utils import cached_path, WEIGHTS_NAME, TF_WEIGHTS_NAME, TF2_WEIGHTS_NAME
+from .file_utils import (TF2_WEIGHTS_NAME, TF_WEIGHTS_NAME, WEIGHTS_NAME,
+                         cached_path, hf_bucket_url, is_remote_url)
 
 logger = logging.getLogger(__name__)
 
@@ -363,14 +364,15 @@ class PreTrainedModel(nn.Module):
                     raise EnvironmentError("Error no file named {} found in directory {} or `from_tf` set to False".format(
                         [WEIGHTS_NAME, TF2_WEIGHTS_NAME, TF_WEIGHTS_NAME + ".index"],
                         pretrained_model_name_or_path))
-            elif os.path.isfile(pretrained_model_name_or_path):
+            elif os.path.isfile(pretrained_model_name_or_path) or is_remote_url(pretrained_model_name_or_path):
                 archive_file = pretrained_model_name_or_path
             elif os.path.isfile(pretrained_model_name_or_path + ".index"):
                 assert from_tf, "We found a TensorFlow checkpoint at {}, please set from_tf to True to load from this checkpoint".format(
                     pretrained_model_name_or_path + ".index")
                 archive_file = pretrained_model_name_or_path + ".index"
             else:
-                archive_file = pretrained_model_name_or_path
+                archive_file = hf_bucket_url(pretrained_model_name_or_path, postfix=WEIGHTS_NAME)
+                # todo do we want to support TF checkpoints here?
 
             # redirect to the cache, if necessary
             try:
diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index 68a767fe82..2b2cec0c15 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -25,7 +25,7 @@ import itertools
 import re
 from io import open
 
-from .file_utils import cached_path, is_tf_available, is_torch_available
+from .file_utils import cached_path, is_remote_url, hf_bucket_url, is_tf_available, is_torch_available
 
 if is_tf_available():
     import tensorflow as tf
@@ -327,12 +327,12 @@ class PreTrainedTokenizer(object):
                 if os.path.isdir(pretrained_model_name_or_path):
                     # If a directory is provided we look for the standard filenames
                     full_file_name = os.path.join(pretrained_model_name_or_path, file_name)
-                else:
+                elif os.path.isfile(pretrained_model_name_or_path) or is_remote_url(pretrained_model_name_or_path):
                     # If a path to a file is provided we use it (will only work for non-BPE tokenizer using a single vocabulary file)
                     full_file_name = pretrained_model_name_or_path
-                if not os.path.exists(full_file_name):
-                    logger.info("Didn't find file {}. We won't load it.".format(full_file_name))
-                    full_file_name = None
+                else:
+                    full_file_name = hf_bucket_url(pretrained_model_name_or_path, postfix=file_name)
+                
                 vocab_files[file_id] = full_file_name
 
             # Look for the additional tokens files

From 3d57c51111054adb01b2ea94bfd45237eb282431 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Wed, 11 Dec 2019 15:10:17 -0500
Subject: [PATCH 321/505] Fix encode plus

---
 transformers/tokenization_utils.py | 39 ++++++++++++++++++------------
 1 file changed, 23 insertions(+), 16 deletions(-)

diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index 68a767fe82..eace409555 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -916,7 +916,7 @@ class PreTrainedTokenizer(object):
             return_tensors: (optional) can be set to 'tf' or 'pt' to return respectively TensorFlow tf.constant
                 or PyTorch torch.Tensor instead of a list of python integers.
             return_token_type_ids: (optional) Set to False to avoid returning token_type_ids (default True).
-            return_attention_mask: (optional) Set to False to avoir returning attention mask (default True)
+            return_attention_mask: (optional) Set to False to avoid returning attention mask (default True)
             return_overflowing_tokens: (optional) Set to True to return overflowing token information (default False).
             return_special_tokens_mask: (optional) Set to True to return special tokens mask information (default False).
 
@@ -961,24 +961,13 @@ class PreTrainedTokenizer(object):
         if add_special_tokens:
             sequence = self.build_inputs_with_special_tokens(ids, pair_ids)
             token_type_ids = self.create_token_type_ids_from_sequences(ids, pair_ids)
-            special_tokens_mask = self.get_special_tokens_mask(ids, pair_ids)
         else:
             sequence = ids + pair_ids if pair else ids
             token_type_ids = [0] * len(ids) + ([1] * len(pair_ids) if pair else [])
-            special_tokens_mask = [0] * (len(ids) + (len(pair_ids) if pair else 0))
+
         if return_special_tokens_mask:
             encoded_inputs["special_tokens_mask"] = self.get_special_tokens_mask(ids, pair_ids)
 
-        # Prepare inputs as tensors if asked
-        if return_tensors == 'tf' and is_tf_available():
-            sequence = tf.constant([sequence])
-            token_type_ids = tf.constant([token_type_ids])
-        elif return_tensors == 'pt' and is_torch_available():
-            sequence = torch.tensor([sequence])
-            token_type_ids = torch.tensor([token_type_ids])
-        elif return_tensors is not None:
-            logger.warning("Unable to convert output to tensors format {}, PyTorch or TensorFlow is not available.".format(return_tensors))
-
         encoded_inputs["input_ids"] = sequence
         if return_token_type_ids:
             encoded_inputs["token_type_ids"] = token_type_ids
@@ -1015,10 +1004,9 @@ class PreTrainedTokenizer(object):
                 if return_special_tokens_mask:
                     encoded_inputs["special_tokens_mask"] = encoded_inputs["special_tokens_mask"] + [1] * difference
                 encoded_inputs["input_ids"] = encoded_inputs["input_ids"] + [self.pad_token_id] * difference
-
             elif self.padding_side == 'left':
                 if return_attention_mask:
-                    encoded_inputs["attention_mask"] =  [0] * difference + [1] * len(encoded_inputs["input_ids"])
+                    encoded_inputs["attention_mask"] = [0] * difference + [1] * len(encoded_inputs["input_ids"])
                 if return_token_type_ids:
                     encoded_inputs["token_type_ids"] = [self.pad_token_type_id] * difference + encoded_inputs["token_type_ids"]
                 if return_special_tokens_mask:
@@ -1030,7 +1018,26 @@ class PreTrainedTokenizer(object):
             
         elif return_attention_mask:
             encoded_inputs["attention_mask"] = [1] * len(encoded_inputs["input_ids"])
-            
+
+        # Prepare inputs as tensors if asked
+        if return_tensors == 'tf' and is_tf_available():
+            encoded_inputs["input_ids"] = tf.constant([encoded_inputs["input_ids"]])
+            encoded_inputs["token_type_ids"] = tf.constant([encoded_inputs["token_type_ids"]])
+
+            if "attention_mask" in encoded_inputs:
+                encoded_inputs["attention_mask"] = tf.constant([encoded_inputs["attention_mask"]])
+
+        elif return_tensors == 'pt' and is_torch_available():
+            encoded_inputs["input_ids"] = torch.tensor([encoded_inputs["input_ids"]])
+            encoded_inputs["token_type_ids"] = torch.tensor([encoded_inputs["token_type_ids"]])
+
+            if "attention_mask" in encoded_inputs:
+                encoded_inputs["attention_mask"] = torch.tensor([encoded_inputs["attention_mask"]])
+        elif return_tensors is not None:
+            logger.warning(
+                "Unable to convert output to tensors format {}, PyTorch or TensorFlow is not available.".format(
+                    return_tensors))
+
         return encoded_inputs
 
     def truncate_sequences(self, ids, pair_ids=None, num_tokens_to_remove=0, truncation_strategy='longest_first', stride=0):

From 31e5b5ff2276c61af7eebb4c353934f8f675d728 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Wed, 11 Dec 2019 15:22:02 -0500
Subject: [PATCH 322/505] Fix tests + first example of doc

---
 transformers/tokenization_utils.py | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index 2b2cec0c15..63d2cc5cb4 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -255,6 +255,7 @@ class PreTrainedTokenizer(object):
             pretrained_model_name_or_path: either:
 
                 - a string with the `shortcut name` of a predefined tokenizer to load from cache or download, e.g.: ``bert-base-uncased``.
+                - a string with the `identifier name` of a predefined tokenizer that was user-uploaded to our S3, e.g.: ``dbmz/bert-base-german-cased``.
                 - a path to a `directory` containing vocabulary files required by the tokenizer, for instance saved using the :func:`~transformers.PreTrainedTokenizer.save_pretrained` method, e.g.: ``./my_model_directory/``.
                 - (not applicable to all derived classes) a path or url to a single saved vocabulary file if and only if the tokenizer only requires a single vocabulary file (e.g. Bert, XLNet), e.g.: ``./my_model_directory/vocab.txt``.
 
@@ -282,6 +283,9 @@ class PreTrainedTokenizer(object):
             # Download vocabulary from S3 and cache.
             tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
 
+            # Download vocabulary from S3 (user-uploaded) and cache.
+            tokenizer = BertTokenizer.from_pretrained('dbmz/bert-base-german-cased')
+
             # If vocabulary files are in a directory (e.g. tokenizer was saved using `save_pretrained('./test/saved_model/')`)
             tokenizer = BertTokenizer.from_pretrained('./test/saved_model/')
 
@@ -327,6 +331,9 @@ class PreTrainedTokenizer(object):
                 if os.path.isdir(pretrained_model_name_or_path):
                     # If a directory is provided we look for the standard filenames
                     full_file_name = os.path.join(pretrained_model_name_or_path, file_name)
+                    if not os.path.exists(full_file_name):
+                        logger.info("Didn't find file {}. We won't load it.".format(full_file_name))
+                        full_file_name = None
                 elif os.path.isfile(pretrained_model_name_or_path) or is_remote_url(pretrained_model_name_or_path):
                     # If a path to a file is provided we use it (will only work for non-BPE tokenizer using a single vocabulary file)
                     full_file_name = pretrained_model_name_or_path

From 18e1f751f1d996c4fe01559ade1cd013186b81e4 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Wed, 11 Dec 2019 17:07:46 -0500
Subject: [PATCH 323/505] TF support

---
 transformers/modeling_tf_utils.py | 9 ++++++---
 transformers/modeling_utils.py    | 3 ++-
 2 files changed, 8 insertions(+), 4 deletions(-)

diff --git a/transformers/modeling_tf_utils.py b/transformers/modeling_tf_utils.py
index e7512b5bd6..4a6d18f447 100644
--- a/transformers/modeling_tf_utils.py
+++ b/transformers/modeling_tf_utils.py
@@ -24,7 +24,8 @@ import os
 import tensorflow as tf
 
 from .configuration_utils import PretrainedConfig
-from .file_utils import cached_path, WEIGHTS_NAME, TF_WEIGHTS_NAME, TF2_WEIGHTS_NAME
+from .file_utils import (TF2_WEIGHTS_NAME, TF_WEIGHTS_NAME, WEIGHTS_NAME,
+                         cached_path, hf_bucket_url, is_remote_url)
 from .modeling_tf_pytorch_utils import load_pytorch_checkpoint_in_tf2_model
 
 logger = logging.getLogger(__name__)
@@ -257,12 +258,14 @@ class TFPreTrainedModel(tf.keras.Model):
                     raise EnvironmentError("Error no file named {} found in directory {} or `from_pt` set to False".format(
                         [WEIGHTS_NAME, TF2_WEIGHTS_NAME],
                         pretrained_model_name_or_path))
-            elif os.path.isfile(pretrained_model_name_or_path):
+            elif os.path.isfile(pretrained_model_name_or_path) or is_remote_url(pretrained_model_name_or_path):
                 archive_file = pretrained_model_name_or_path
             elif os.path.isfile(pretrained_model_name_or_path + ".index"):
                 archive_file = pretrained_model_name_or_path + ".index"
             else:
-                archive_file = pretrained_model_name_or_path
+                archive_file = hf_bucket_url(pretrained_model_name_or_path, postfix=TF2_WEIGHTS_NAME)
+                if from_pt:
+                    raise EnvironmentError("Loading a TF model from a PyTorch checkpoint is not supported when using a model identifier name.")
 
             # redirect to the cache, if necessary
             try:
diff --git a/transformers/modeling_utils.py b/transformers/modeling_utils.py
index eac4252336..37088f8e67 100644
--- a/transformers/modeling_utils.py
+++ b/transformers/modeling_utils.py
@@ -372,7 +372,8 @@ class PreTrainedModel(nn.Module):
                 archive_file = pretrained_model_name_or_path + ".index"
             else:
                 archive_file = hf_bucket_url(pretrained_model_name_or_path, postfix=WEIGHTS_NAME)
-                # todo do we want to support TF checkpoints here?
+                if from_tf:
+                    raise EnvironmentError("Loading a PyTorch model from a TF checkpoint is not supported when using a model identifier name.")
 
             # redirect to the cache, if necessary
             try:

From 4f15e5a267201f86bdd9628cf58592d0e1cc86eb Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Wed, 11 Dec 2019 17:41:51 -0500
Subject: [PATCH 324/505] Add tests.

Maybe not the best possible place for the tests, lmk.
---
 transformers/tests/modeling_auto_test.py     | 7 ++++++-
 transformers/tests/modeling_tf_auto_test.py  | 7 ++++++-
 transformers/tests/tokenization_auto_test.py | 7 ++++++-
 transformers/tests/utils.py                  | 3 +++
 4 files changed, 21 insertions(+), 3 deletions(-)

diff --git a/transformers/tests/modeling_auto_test.py b/transformers/tests/modeling_auto_test.py
index 9b7d920bc8..871a262fe8 100644
--- a/transformers/tests/modeling_auto_test.py
+++ b/transformers/tests/modeling_auto_test.py
@@ -22,7 +22,7 @@ import logging
 
 from transformers import is_torch_available
 
-from .utils import require_torch, slow
+from .utils import require_torch, slow, SMALL_MODEL_IDENTIFIER
 
 if is_torch_available():
     from transformers import (AutoConfig, BertConfig,
@@ -92,6 +92,11 @@ class AutoModelTest(unittest.TestCase):
             self.assertIsNotNone(model)
             self.assertIsInstance(model, BertForQuestionAnswering)
 
+    def test_from_pretrained_identifier(self):
+        logging.basicConfig(level=logging.INFO)
+        model = AutoModelWithLMHead.from_pretrained(SMALL_MODEL_IDENTIFIER)
+        self.assertIsInstance(model, BertForMaskedLM)
+
 
 if __name__ == "__main__":
     unittest.main()
diff --git a/transformers/tests/modeling_tf_auto_test.py b/transformers/tests/modeling_tf_auto_test.py
index 7ea48015d9..7ab6eaa3d6 100644
--- a/transformers/tests/modeling_tf_auto_test.py
+++ b/transformers/tests/modeling_tf_auto_test.py
@@ -22,7 +22,7 @@ import logging
 
 from transformers import is_tf_available
 
-from .utils import require_tf, slow
+from .utils import require_tf, slow, SMALL_MODEL_IDENTIFIER
 
 if is_tf_available():
     from transformers import (AutoConfig, BertConfig,
@@ -93,6 +93,11 @@ class TFAutoModelTest(unittest.TestCase):
             self.assertIsNotNone(model)
             self.assertIsInstance(model, TFBertForQuestionAnswering)
 
+    def test_from_pretrained_identifier(self):
+        logging.basicConfig(level=logging.INFO)
+        model = TFAutoModelWithLMHead.from_pretrained(SMALL_MODEL_IDENTIFIER, force_download=True)
+        self.assertIsInstance(model, TFBertForMaskedLM)
+
 
 if __name__ == "__main__":
     unittest.main()
diff --git a/transformers/tests/tokenization_auto_test.py b/transformers/tests/tokenization_auto_test.py
index 18346d2768..0a894cac04 100644
--- a/transformers/tests/tokenization_auto_test.py
+++ b/transformers/tests/tokenization_auto_test.py
@@ -23,7 +23,7 @@ import logging
 from transformers import AutoTokenizer, BertTokenizer, AutoTokenizer, GPT2Tokenizer
 from transformers import BERT_PRETRAINED_CONFIG_ARCHIVE_MAP, GPT2_PRETRAINED_CONFIG_ARCHIVE_MAP
 
-from .utils import slow
+from .utils import slow, SMALL_MODEL_IDENTIFIER
 
 
 class AutoTokenizerTest(unittest.TestCase):
@@ -42,6 +42,11 @@ class AutoTokenizerTest(unittest.TestCase):
             self.assertIsInstance(tokenizer, GPT2Tokenizer)
             self.assertGreater(len(tokenizer), 0)
 
+    def test_tokenizer_from_pretrained_identifier(self):
+        logging.basicConfig(level=logging.INFO)
+        tokenizer = AutoTokenizer.from_pretrained(SMALL_MODEL_IDENTIFIER)
+        self.assertIsInstance(tokenizer, BertTokenizer)
+        self.assertEqual(len(tokenizer), 12)
 
 if __name__ == "__main__":
     unittest.main()
diff --git a/transformers/tests/utils.py b/transformers/tests/utils.py
index 7a51ab612b..3aff1daf83 100644
--- a/transformers/tests/utils.py
+++ b/transformers/tests/utils.py
@@ -6,6 +6,9 @@ from distutils.util import strtobool
 from transformers.file_utils import _tf_available, _torch_available
 
 
+SMALL_MODEL_IDENTIFIER = "julien-c/bert-xsmall-dummy"
+
+
 try:
     run_slow = os.environ["RUN_SLOW"]
 except KeyError:

From c03c0dfd230a5174c536a58d6ba5e590ed1afcc4 Mon Sep 17 00:00:00 2001
From: Masatoshi Suzuki <sappukei.suzuki@gmail.com>
Date: Fri, 15 Nov 2019 17:24:56 +0900
Subject: [PATCH 325/505] Add support for Japanese BERT models by cl-tohoku

---
 docs/source/pretrained_models.rst          |  18 ++
 transformers/__init__.py                   |   1 +
 transformers/configuration_bert.py         |   4 +
 transformers/modeling_bert.py              |   8 +-
 transformers/modeling_tf_bert.py           |  16 +-
 transformers/tokenization_auto.py          |   3 +
 transformers/tokenization_bert_japanese.py | 247 +++++++++++++++++++++
 7 files changed, 289 insertions(+), 8 deletions(-)
 create mode 100644 transformers/tokenization_bert_japanese.py

diff --git a/docs/source/pretrained_models.rst b/docs/source/pretrained_models.rst
index 2fe1f8a314..d3498e057d 100644
--- a/docs/source/pretrained_models.rst
+++ b/docs/source/pretrained_models.rst
@@ -61,6 +61,24 @@ Here is the full list of the currently provided pretrained models together with
 |                   | ``bert-base-german-dbmdz-uncased``                         | | 12-layer, 768-hidden, 12-heads, 110M parameters.                                                                                    |
 |                   |                                                            | | Trained on uncased German text by DBMDZ                                                                                             |
 |                   |                                                            | (see `details on dbmdz repository <https://github.com/dbmdz/german-bert>`__).                                                         |
+|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   | ``bert-base-japanese``                                     | | 12-layer, 768-hidden, 12-heads, 110M parameters.                                                                                    |
+|                   |                                                            | | Trained on Japanese text. Text is tokenized with MeCab and WordPiece.                                                 |
+|                   |                                                            | | `MeCab <https://taku910.github.io/mecab/>`__ is required for tokenization.                                                            |
+|                   |                                                            | (see `details on cl-tohoku repository <https://github.com/cl-tohoku/bert-japanese>`__).                                             |
+|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   | ``bert-base-japanese-whole-word-masking``                  | | 12-layer, 768-hidden, 12-heads, 110M parameters.                                                                                    |
+|                   |                                                            | | Trained on Japanese text using Whole-Word-Masking. Text is tokenized with MeCab and WordPiece.                        |
+|                   |                                                            | | `MeCab <https://taku910.github.io/mecab/>`__ is required for tokenization.                                                            |
+|                   |                                                            | (see `details on cl-tohoku repository <https://github.com/cl-tohoku/bert-japanese>`__).                                             |
+|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   | ``bert-base-japanese-char``                                | | 12-layer, 768-hidden, 12-heads, 110M parameters.                                                                                    |
+|                   |                                                            | | Trained on Japanese text. Text is tokenized into characters.                                                          |
+|                   |                                                            | (see `details on cl-tohoku repository <https://github.com/cl-tohoku/bert-japanese>`__).                                             |
+|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   | ``bert-base-japanese-char-whole-word-masking``             | | 12-layer, 768-hidden, 12-heads, 110M parameters.                                                                                    |
+|                   |                                                            | | Trained on Japanese text using Whole-Word-Masking. Text is tokenized into characters.                                 |
+|                   |                                                            | (see `details on cl-tohoku repository <https://github.com/cl-tohoku/bert-japanese>`__).                                             |
 +-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 | GPT               | ``openai-gpt``                                             | | 12-layer, 768-hidden, 12-heads, 110M parameters.                                                                                    |
 |                   |                                                            | | OpenAI GPT English model                                                                                                            |
diff --git a/transformers/__init__.py b/transformers/__init__.py
index f9a28add5f..5d7b0b772c 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -37,6 +37,7 @@ if is_sklearn_available():
 from .tokenization_utils import (PreTrainedTokenizer)
 from .tokenization_auto import AutoTokenizer
 from .tokenization_bert import BertTokenizer, BasicTokenizer, WordpieceTokenizer
+from .tokenization_bert_japanese import BertJapaneseTokenizer, MecabTokenizer, CharacterTokenizer
 from .tokenization_openai import OpenAIGPTTokenizer
 from .tokenization_transfo_xl import (TransfoXLTokenizer, TransfoXLCorpus)
 from .tokenization_gpt2 import GPT2Tokenizer
diff --git a/transformers/configuration_bert.py b/transformers/configuration_bert.py
index d63be963eb..16f1f60404 100644
--- a/transformers/configuration_bert.py
+++ b/transformers/configuration_bert.py
@@ -42,6 +42,10 @@ BERT_PRETRAINED_CONFIG_ARCHIVE_MAP = {
     'bert-base-cased-finetuned-mrpc': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-cased-finetuned-mrpc-config.json",
     'bert-base-german-dbmdz-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-german-dbmdz-cased-config.json",
     'bert-base-german-dbmdz-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-german-dbmdz-uncased-config.json",
+    'bert-base-japanese': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-config.json",
+    'bert-base-japanese-whole-word-masking': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-whole-word-masking-config.json",
+    'bert-base-japanese-char': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-char-config.json",
+    'bert-base-japanese-char-whole-word-masking': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-char-whole-word-masking-config.json"
 }
 
 
diff --git a/transformers/modeling_bert.py b/transformers/modeling_bert.py
index d84b0a1a7c..e2e115a015 100644
--- a/transformers/modeling_bert.py
+++ b/transformers/modeling_bert.py
@@ -48,6 +48,10 @@ BERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
     'bert-base-cased-finetuned-mrpc': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-cased-finetuned-mrpc-pytorch_model.bin",
     'bert-base-german-dbmdz-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-german-dbmdz-cased-pytorch_model.bin",
     'bert-base-german-dbmdz-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-german-dbmdz-uncased-pytorch_model.bin",
+    'bert-base-japanese': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-pytorch_model.bin",
+    'bert-base-japanese-whole-word-masking': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-whole-word-masking-pytorch_model.bin",
+    'bert-base-japanese-char': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-char-pytorch_model.bin",
+    'bert-base-japanese-char-whole-word-masking': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-char-whole-word-masking-pytorch_model.bin"
 }
 
 
@@ -1233,9 +1237,9 @@ class BertForQuestionAnswering(BertPreTrainedModel):
         question, text = "Who was Jim Henson?", "Jim Henson was a nice puppet"
         input_text = "[CLS] " + question + " [SEP] " + text + " [SEP]"
         input_ids = tokenizer.encode(input_text)
-        token_type_ids = [0 if i <= input_ids.index(102) else 1 for i in range(len(input_ids))] 
+        token_type_ids = [0 if i <= input_ids.index(102) else 1 for i in range(len(input_ids))]
         start_scores, end_scores = model(torch.tensor([input_ids]), token_type_ids=torch.tensor([token_type_ids]))
-        all_tokens = tokenizer.convert_ids_to_tokens(input_ids)  
+        all_tokens = tokenizer.convert_ids_to_tokens(input_ids)
         print(' '.join(all_tokens[torch.argmax(start_scores) : torch.argmax(end_scores)+1]))
         # a nice puppet
 
diff --git a/transformers/modeling_tf_bert.py b/transformers/modeling_tf_bert.py
index 5aa7bb3da2..27dd311a4d 100644
--- a/transformers/modeling_tf_bert.py
+++ b/transformers/modeling_tf_bert.py
@@ -48,6 +48,10 @@ TF_BERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
     'bert-large-uncased-whole-word-masking-finetuned-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-uncased-whole-word-masking-finetuned-squad-tf_model.h5",
     'bert-large-cased-whole-word-masking-finetuned-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-cased-whole-word-masking-finetuned-squad-tf_model.h5",
     'bert-base-cased-finetuned-mrpc': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-cased-finetuned-mrpc-tf_model.h5",
+    'bert-base-japanese': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-tf_model.h5",
+    'bert-base-japanese-whole-word-masking': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-whole-word-masking-tf_model.h5",
+    'bert-base-japanese-char': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-char-tf_model.h5",
+    'bert-base-japanese-char-whole-word-masking': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-char-whole-word-masking-tf_model.h5"
 }
 
 
@@ -129,7 +133,7 @@ class TFBertEmbeddings(tf.keras.layers.Layer):
                 linear tensor, float32 with shape [batch_size, length, vocab_size].
         Raises:
             ValueError: if mode is not valid.
-        
+
         Shared weights logic adapted from
             https://github.com/tensorflow/models/blob/a009f4fb9d2fc4949e32192a944688925ef78659/official/transformer/v2/embedding_layer.py#L24
         """
@@ -148,7 +152,7 @@ class TFBertEmbeddings(tf.keras.layers.Layer):
             input_shape = shape_list(input_ids)
         else:
             input_shape = shape_list(inputs_embeds)[:-1]
-        
+
         seq_length = input_shape[1]
         if position_ids is None:
             position_ids = tf.range(seq_length, dtype=tf.int32)[tf.newaxis, :]
@@ -246,7 +250,7 @@ class TFBertSelfAttention(tf.keras.layers.Layer):
         context_layer = tf.matmul(attention_probs, value_layer)
 
         context_layer = tf.transpose(context_layer, perm=[0, 2, 1, 3])
-        context_layer = tf.reshape(context_layer, 
+        context_layer = tf.reshape(context_layer,
                                   (batch_size, -1, self.all_head_size))  # (batch_size, seq_len_q, all_head_size)
 
         outputs = (context_layer, attention_probs) if self.output_attentions else (context_layer,)
@@ -591,7 +595,7 @@ BERT_START_DOCSTRING = r"""    The BERT model was proposed in
             `model({'input_ids': input_ids, 'token_type_ids': token_type_ids})`
 
     Parameters:
-        config (:class:`~transformers.BertConfig`): Model configuration class with all the parameters of the model. 
+        config (:class:`~transformers.BertConfig`): Model configuration class with all the parameters of the model.
             Initializing with a config file does not load the weights associated with the model, only the configuration.
             Check out the :meth:`~transformers.PreTrainedModel.from_pretrained` method to load the model weights.
 """
@@ -605,13 +609,13 @@ BERT_INPUTS_DOCSTRING = r"""
             (a) For sequence pairs:
 
                 ``tokens:         [CLS] is this jack ##son ##ville ? [SEP] no it is not . [SEP]``
-                
+
                 ``token_type_ids:   0   0  0    0    0     0       0   0   1  1  1  1   1   1``
 
             (b) For single sequences:
 
                 ``tokens:         [CLS] the dog is hairy . [SEP]``
-                
+
                 ``token_type_ids:   0   0   0   0  0     0   0``
 
             Bert is a model with absolute position embeddings so it's usually advised to pad the inputs on
diff --git a/transformers/tokenization_auto.py b/transformers/tokenization_auto.py
index b7c5046961..d63b7e783d 100644
--- a/transformers/tokenization_auto.py
+++ b/transformers/tokenization_auto.py
@@ -19,6 +19,7 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 import logging
 
 from .tokenization_bert import BertTokenizer
+from .tokenization_bert_japanese import BertJapaneseTokenizer
 from .tokenization_openai import OpenAIGPTTokenizer
 from .tokenization_gpt2 import GPT2Tokenizer
 from .tokenization_ctrl import CTRLTokenizer
@@ -118,6 +119,8 @@ class AutoTokenizer(object):
             return CamembertTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return RobertaTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
+        elif 'bert-japanese' in pretrained_model_name_or_path:
+            return BertJapaneseTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
         elif 'bert' in pretrained_model_name_or_path:
             return BertTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
         elif 'openai-gpt' in pretrained_model_name_or_path:
diff --git a/transformers/tokenization_bert_japanese.py b/transformers/tokenization_bert_japanese.py
new file mode 100644
index 0000000000..8554a1c880
--- /dev/null
+++ b/transformers/tokenization_bert_japanese.py
@@ -0,0 +1,247 @@
+# coding=utf-8
+# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Tokenization classes."""
+
+from __future__ import absolute_import, division, print_function, unicode_literals
+
+import collections
+import logging
+import os
+import unicodedata
+from io import open
+
+from .tokenization_bert import BertTokenizer, BasicTokenizer, WordpieceTokenizer, load_vocab
+from .tokenization_utils import PreTrainedTokenizer
+
+logger = logging.getLogger(__name__)
+
+VOCAB_FILES_NAMES = {'vocab_file': 'vocab.txt'}
+
+PRETRAINED_VOCAB_FILES_MAP = {
+    'vocab_file':
+    {
+        'bert-base-japanese': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-vocab.txt",
+        'bert-base-japanese-whole-word-masking': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-whole-word-masking-vocab.txt",
+        'bert-base-japanese-char': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-char-vocab.txt",
+        'bert-base-japanese-char-whole-word-masking': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-char-whole-word-masking-vocab.txt"
+    }
+}
+
+PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
+    'bert-base-japanese': 512,
+    'bert-base-japanese-whole-word-masking': 512,
+    'bert-base-japanese-char': 512,
+    'bert-base-japanese-char-whole-word-masking': 512
+}
+
+PRETRAINED_INIT_CONFIGURATION = {
+    'bert-base-japanese': {
+        'do_lower_case': False,
+        'word_tokenizer_type': 'mecab',
+        'subword_tokenizer_type': 'wordpiece'
+    },
+    'bert-base-japanese-whole-word-masking':{
+        'do_lower_case': False,
+        'word_tokenizer_type': 'mecab',
+        'subword_tokenizer_type': 'wordpiece'
+    },
+    'bert-base-japanese-char': {
+        'do_lower_case': False,
+        'word_tokenizer_type': 'mecab',
+        'subword_tokenizer_type': 'character'
+    },
+    'bert-base-japanese-char-whole-word-masking': {
+        'do_lower_case': False,
+        'word_tokenizer_type': 'mecab',
+        'subword_tokenizer_type': 'character'
+    }
+}
+
+
+class BertJapaneseTokenizer(BertTokenizer):
+    """BERT tokenizer for Japanese text"""
+
+    vocab_files_names = VOCAB_FILES_NAMES
+    pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+    pretrained_init_configuration = PRETRAINED_INIT_CONFIGURATION
+    max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
+
+    def __init__(self, vocab_file, do_lower_case=False,
+                 do_word_tokenize=True, do_subword_tokenize=True,
+                 word_tokenizer_type='basic', subword_tokenizer_type='wordpiece',
+                 never_split=None, unk_token='[UNK]', sep_token='[SEP]',
+                 pad_token='[PAD]', cls_token='[CLS]', mask_token='[MASK]', **kwargs):
+        """Constructs a MecabBertTokenizer.
+
+        Args:
+            **vocab_file**: Path to a one-wordpiece-per-line vocabulary file.
+            **do_lower_case**: (`optional`) boolean (default True)
+                Whether to lower case the input.
+                Only has an effect when do_basic_tokenize=True.
+            **do_word_tokenize**: (`optional`) boolean (default True)
+                Whether to do word tokenization.
+            **do_subword_tokenize**: (`optional`) boolean (default True)
+                Whether to do subword tokenization.
+            **word_tokenizer_type**: (`optional`) string (default "basic")
+                Type of word tokenizer.
+            **subword_tokenizer_type**: (`optional`) string (default "wordpiece")
+                Type of subword tokenizer.
+        """
+        super(BertTokenizer, self).__init__(unk_token=unk_token, sep_token=sep_token,
+                                            pad_token=pad_token, cls_token=cls_token,
+                                            mask_token=mask_token, **kwargs)
+        self.max_len_single_sentence = self.max_len - 2  # take into account special tokens
+        self.max_len_sentences_pair = self.max_len - 3  # take into account special tokens
+
+        if not os.path.isfile(vocab_file):
+            raise ValueError(
+                "Can't find a vocabulary file at path '{}'. To load the vocabulary from a Google pretrained "
+                "model use `tokenizer = BertTokenizer.from_pretrained(PRETRAINED_MODEL_NAME)`".format(vocab_file))
+        self.vocab = load_vocab(vocab_file)
+        self.ids_to_tokens = collections.OrderedDict(
+            [(ids, tok) for tok, ids in self.vocab.items()])
+
+        self.do_word_tokenize = do_word_tokenize
+        if do_word_tokenize:
+            if word_tokenizer_type == 'basic':
+                self.word_tokenizer = BasicTokenizer(do_lower_case=do_lower_case,
+                                                     never_split=never_split,
+                                                     tokenize_chinese_chars=False)
+            elif word_tokenizer_type == 'mecab':
+                self.word_tokenizer = MecabTokenizer(do_lower_case=do_lower_case,
+                                                     never_split=never_split)
+            else:
+                raise ValueError(
+                    "Invalid word_tokenizer_type '{}' is specified.".format(word_tokenizer_type))
+
+        self.do_subword_tokenize = do_subword_tokenize
+        if do_subword_tokenize:
+            if subword_tokenizer_type == 'wordpiece':
+                self.subword_tokenizer = WordpieceTokenizer(vocab=self.vocab,
+                                                            unk_token=self.unk_token)
+            elif subword_tokenizer_type == 'character':
+                self.subword_tokenizer = CharacterTokenizer(vocab=self.vocab,
+                                                            unk_token=self.unk_token)
+            else:
+                raise ValueError(
+                    "Invalid subword_tokenizer_type '{}' is specified.".format(subword_tokenizer_type))
+
+
+    def _tokenize(self, text):
+        if self.do_word_tokenize:
+            tokens = self.word_tokenizer.tokenize(text,
+                                                  never_split=self.all_special_tokens)
+        else:
+            tokens = [text]
+
+        if self.do_subword_tokenize:
+            split_tokens = [sub_token for token in tokens
+                            for sub_token in self.subword_tokenizer.tokenize(token)]
+        else:
+            split_tokens = tokens
+
+        return split_tokens
+
+
+class MecabTokenizer(object):
+    """Runs basic tokenization with MeCab morphological parser."""
+
+    def __init__(self, do_lower_case=False, never_split=None, normalize_text=True):
+        """Constructs a MecabTokenizer.
+
+        Args:
+            **do_lower_case**: (`optional`) boolean (default True)
+                Whether to lower case the input.
+            **never_split**: (`optional`) list of str
+                Kept for backward compatibility purposes.
+                Now implemented directly at the base class level (see :func:`PreTrainedTokenizer.tokenize`)
+                List of token not to split.
+            **normalize_text**: (`optional`) boolean (default True)
+                Whether to apply unicode normalization to text before tokenization.
+        """
+        self.do_lower_case = do_lower_case
+        self.never_split = never_split if never_split is not None else []
+        self.normalize_text = normalize_text
+
+        import MeCab
+        self.mecab = MeCab.Tagger()
+
+    def tokenize(self, text, never_split=None, **kwargs):
+        """Tokenizes a piece of text."""
+        if self.normalize_text:
+            text = unicodedata.normalize('NFKC', text)
+
+        never_split = self.never_split + (never_split if never_split is not None else [])
+        tokens = []
+
+        cursor = 0
+        for line in self.mecab.parse(text).split('\n'):
+            if line == 'EOS':
+                break
+
+            token, _ = line.split('\t')
+            token_start = text.index(token, cursor)
+            token_end = token_start + len(token)
+            if self.do_lower_case and token not in never_split:
+                token = token.lower()
+
+            tokens.append(token)
+            cursor = token_end
+
+        return tokens
+
+
+class CharacterTokenizer(object):
+    """Runs Character tokenziation."""
+
+    def __init__(self, vocab, unk_token, normalize_text=True):
+        """Constructs a CharacterTokenizer.
+
+        Args:
+            **vocab**:
+                Vocabulary object.
+            **unk_token**: str
+                A special symbol for out-of-vocabulary token.
+            **normalize_text**: (`optional`) boolean (default True)
+                Whether to apply unicode normalization to text before tokenization.
+        """
+        self.vocab = vocab
+        self.unk_token = unk_token
+        self.normalize_text = normalize_text
+
+    def tokenize(self, text):
+        """Tokenizes a piece of text into characters.
+
+        For example:
+            input = "apple"
+            output = ["a", "p", "p", "l", "e"]
+        Args:
+            text: A single token or whitespace separated tokens.
+                This should have already been passed through `BasicTokenizer`.
+        Returns:
+            A list of characters.
+        """
+        if self.normalize_text:
+            text = unicodedata.normalize('NFKC', text)
+
+        output_tokens = []
+        for i, char in enumerate(text):
+            if char not in self.vocab:
+                output_tokens.append(self.unk_token)
+                continue
+
+            output_tokens.append(char)
+
+        return output_tokens

From 57b5cb3eaa850a212235fccbd4e5d002aede72b6 Mon Sep 17 00:00:00 2001
From: Masatoshi Suzuki <sappukei.suzuki@gmail.com>
Date: Wed, 20 Nov 2019 09:02:10 +0900
Subject: [PATCH 326/505] Fix loading BertJapaneseTokenizer

---
 transformers/tokenization_auto.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/transformers/tokenization_auto.py b/transformers/tokenization_auto.py
index d63b7e783d..f36a584521 100644
--- a/transformers/tokenization_auto.py
+++ b/transformers/tokenization_auto.py
@@ -73,6 +73,7 @@ class AutoTokenizer(object):
             - contains `albert`: AlbertTokenizer (ALBERT model)
             - contains `camembert`: CamembertTokenizer (CamemBERT model)
             - contains `roberta`: RobertaTokenizer (RoBERTa model)
+            - contains `bert-base-japanese`: BertJapaneseTokenizer (Bert model)
             - contains `bert`: BertTokenizer (Bert model)
             - contains `openai-gpt`: OpenAIGPTTokenizer (OpenAI GPT model)
             - contains `gpt2`: GPT2Tokenizer (OpenAI GPT-2 model)
@@ -119,7 +120,7 @@ class AutoTokenizer(object):
             return CamembertTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return RobertaTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
-        elif 'bert-japanese' in pretrained_model_name_or_path:
+        elif 'bert-base-japanese' in pretrained_model_name_or_path:
             return BertJapaneseTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
         elif 'bert' in pretrained_model_name_or_path:
             return BertTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)

From a09da4eeb0397dd66d61182177dd3b753d70e62a Mon Sep 17 00:00:00 2001
From: Masatoshi Suzuki <sappukei.suzuki@gmail.com>
Date: Fri, 29 Nov 2019 19:24:43 +0900
Subject: [PATCH 327/505] Add a test for Japanese BERT tokenizers

---
 .../tests/tokenization_bert_japanese_test.py  | 192 ++++++++++++++++++
 1 file changed, 192 insertions(+)
 create mode 100644 transformers/tests/tokenization_bert_japanese_test.py

diff --git a/transformers/tests/tokenization_bert_japanese_test.py b/transformers/tests/tokenization_bert_japanese_test.py
new file mode 100644
index 0000000000..6f66b96411
--- /dev/null
+++ b/transformers/tests/tokenization_bert_japanese_test.py
@@ -0,0 +1,192 @@
+# coding=utf-8
+# Copyright 2018 The Google AI Language Team Authors.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from __future__ import absolute_import, division, print_function, unicode_literals
+
+import os
+import unittest
+import pytest
+from io import open
+
+from transformers.tokenization_bert import WordpieceTokenizer
+from transformers.tokenization_bert_japanese import (BertJapaneseTokenizer,
+                                                     MecabTokenizer, CharacterTokenizer,
+                                                     VOCAB_FILES_NAMES)
+
+from .tokenization_tests_commons import CommonTestCases
+
+
+class BertJapaneseTokenizationTest(CommonTestCases.CommonTokenizerTester):
+
+    tokenizer_class = BertJapaneseTokenizer
+
+    def setUp(self):
+        super(BertJapaneseTokenizationTest, self).setUp()
+
+        vocab_tokens = [u"[UNK]", u"[CLS]", u"[SEP]",
+            u"こんにちは", u"こん", u"にちは", u"ばんは", u"##こん", u"##にちは", u"##ばんは",
+            u"世界", u"##世界", u"、", u"##、", u"。", u"##。"]
+
+        self.vocab_file = os.path.join(self.tmpdirname, VOCAB_FILES_NAMES["vocab_file"])
+        with open(self.vocab_file, "w", encoding="utf-8") as vocab_writer:
+            vocab_writer.write("".join([x + "\n" for x in vocab_tokens]))
+
+    def get_tokenizer(self, **kwargs):
+        return BertJapaneseTokenizer.from_pretrained(self.tmpdirname, **kwargs)
+
+    def get_input_output_texts(self):
+        input_text = u"こんにちは、世界。 \nこんばんは、世界。"
+        output_text = u"こんにちは 、 世界 。 こんばんは 、 世界 。"
+        return input_text, output_text
+
+    def test_full_tokenizer(self):
+        tokenizer = self.tokenizer_class(self.vocab_file)
+
+        tokens = tokenizer.tokenize(u"こんにちは、世界。\nこんばんは、世界。")
+        self.assertListEqual(tokens,
+                             [u"こんにちは", u"、", u"世界", u"。",
+                              u"こん", u"##ばんは", u"、", u"世界", "。"])
+        self.assertListEqual(tokenizer.convert_tokens_to_ids(tokens),
+                             [3, 12, 10, 14, 4, 9, 12, 10, 14])
+
+    def test_mecab_tokenizer(self):
+        tokenizer = MecabTokenizer()
+
+        self.assertListEqual(
+            tokenizer.tokenize(u" \tｱｯﾌﾟﾙストアでiPhone８ が  \n 発売された　。  "),
+                               [u"アップルストア", u"で", u"iPhone", u"8", u"が",
+                                u"発売", u"さ", u"れ", u"た", u"。"])
+
+    def test_mecab_tokenizer_lower(self):
+        tokenizer = MecabTokenizer(do_lower_case=True)
+
+        self.assertListEqual(
+            tokenizer.tokenize(u" \tｱｯﾌﾟﾙストアでiPhone８ が  \n 発売された　。  "),
+                               [u"アップルストア", u"で", u"iphone", u"8", u"が",
+                                u"発売", u"さ", u"れ", u"た", u"。"])
+
+    def test_mecab_tokenizer_no_normalize(self):
+        tokenizer = MecabTokenizer(normalize_text=False)
+
+        self.assertListEqual(
+            tokenizer.tokenize(u" \tｱｯﾌﾟﾙストアでiPhone８ が  \n 発売された　。  "),
+                               [u"ｱｯﾌﾟﾙストア", u"で", u"iPhone", u"８", u"が",
+                                u"発売", u"さ", u"れ", u"た", u"　", u"。"])
+
+    def test_wordpiece_tokenizer(self):
+        vocab_tokens = [u"[UNK]", u"[CLS]", u"[SEP]",
+            u"こんにちは", u"こん", u"にちは" u"ばんは", u"##こん", u"##にちは", u"##ばんは"]
+
+        vocab = {}
+        for (i, token) in enumerate(vocab_tokens):
+            vocab[token] = i
+        tokenizer = WordpieceTokenizer(vocab=vocab, unk_token=u"[UNK]")
+
+        self.assertListEqual(tokenizer.tokenize(u""), [])
+
+        self.assertListEqual(tokenizer.tokenize(u"こんにちは"),
+                             [u"こんにちは"])
+
+        self.assertListEqual(tokenizer.tokenize(u"こんばんは"),
+                             [u"こん", u"##ばんは"])
+
+        self.assertListEqual(tokenizer.tokenize(u"こんばんは こんばんにちは こんにちは"),
+                             [u"こん", u"##ばんは", u"[UNK]", u"こんにちは"])
+
+    @pytest.mark.slow
+    def test_sequence_builders(self):
+        tokenizer = self.tokenizer_class.from_pretrained("bert-base-japanese")
+
+        text = tokenizer.encode(u"ありがとう。", add_special_tokens=False)
+        text_2 = tokenizer.encode(u"どういたしまして。", add_special_tokens=False)
+
+        encoded_sentence = tokenizer.build_inputs_with_special_tokens(text)
+        encoded_pair = tokenizer.build_inputs_with_special_tokens(text, text_2)
+
+        # 2 is for "[CLS]", 3 is for "[SEP]"
+        assert encoded_sentence == [2] + text + [3]
+        assert encoded_pair == [2] + text + [3] + text_2 + [3]
+
+
+class BertJapaneseCharacterTokenizationTest(CommonTestCases.CommonTokenizerTester):
+
+    tokenizer_class = BertJapaneseTokenizer
+
+    def setUp(self):
+        super(BertJapaneseCharacterTokenizationTest, self).setUp()
+
+        vocab_tokens = [u"[UNK]", u"[CLS]", u"[SEP]",
+            u"こ", u"ん", u"に", u"ち", u"は", u"ば", u"世", u"界", u"、", u"。"]
+
+        self.vocab_file = os.path.join(self.tmpdirname, VOCAB_FILES_NAMES["vocab_file"])
+        with open(self.vocab_file, "w", encoding="utf-8") as vocab_writer:
+            vocab_writer.write("".join([x + "\n" for x in vocab_tokens]))
+
+    def get_tokenizer(self, **kwargs):
+        return BertJapaneseTokenizer.from_pretrained(self.tmpdirname,
+                                                     subword_tokenizer_type="character",
+                                                     **kwargs)
+
+    def get_input_output_texts(self):
+        input_text = u"こんにちは、世界。 \nこんばんは、世界。"
+        output_text = u"こ ん に ち は 、 世 界 。 こ ん ば ん は 、 世 界 。"
+        return input_text, output_text
+
+    def test_full_tokenizer(self):
+        tokenizer = self.tokenizer_class(self.vocab_file,
+                                         subword_tokenizer_type="character")
+
+        tokens = tokenizer.tokenize(u"こんにちは、世界。 \nこんばんは、世界。")
+        self.assertListEqual(tokens,
+            [u"こ", u"ん", u"に", u"ち", u"は", u"、", u"世", u"界", u"。",
+             u"こ", u"ん", u"ば", u"ん", u"は", u"、", u"世", u"界", u"。"])
+        self.assertListEqual(tokenizer.convert_tokens_to_ids(tokens),
+                             [3, 4, 5, 6, 7, 11, 9, 10, 12,
+                              3, 4, 8, 4, 7, 11, 9, 10, 12])
+
+    def test_character_tokenizer(self):
+        vocab_tokens = [u"[UNK]", u"[CLS]", u"[SEP]",
+            u"こ", u"ん", u"に", u"ち", u"は", u"ば", u"世", u"界"u"、", u"。"]
+
+        vocab = {}
+        for (i, token) in enumerate(vocab_tokens):
+            vocab[token] = i
+        tokenizer = CharacterTokenizer(vocab=vocab, unk_token=u"[UNK]")
+
+        self.assertListEqual(tokenizer.tokenize(u""), [])
+
+        self.assertListEqual(tokenizer.tokenize(u"こんにちは"),
+                             [u"こ", u"ん", u"に", u"ち", u"は"])
+
+        self.assertListEqual(tokenizer.tokenize(u"こんにちほ"),
+                             [u"こ", u"ん", u"に", u"ち", u"[UNK]"])
+
+    @pytest.mark.slow
+    def test_sequence_builders(self):
+        tokenizer = self.tokenizer_class.from_pretrained("bert-base-japanese-char")
+
+        text = tokenizer.encode(u"ありがとう。", add_special_tokens=False)
+        text_2 = tokenizer.encode(u"どういたしまして。", add_special_tokens=False)
+
+        encoded_sentence = tokenizer.build_inputs_with_special_tokens(text)
+        encoded_pair = tokenizer.build_inputs_with_special_tokens(text, text_2)
+
+        # 2 is for "[CLS]", 3 is for "[SEP]"
+        assert encoded_sentence == [2] + text + [3]
+        assert encoded_pair == [2] + text + [3] + text_2 + [3]
+
+
+
+if __name__ == '__main__':
+    unittest.main()

From 6a43dc9d7d592362d144209097e1d93876f8e88a Mon Sep 17 00:00:00 2001
From: Masatoshi Suzuki <sappukei.suzuki@gmail.com>
Date: Thu, 5 Dec 2019 11:19:02 +0900
Subject: [PATCH 328/505] Support Python 2

---
 transformers/tokenization_bert_japanese.py | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/transformers/tokenization_bert_japanese.py b/transformers/tokenization_bert_japanese.py
index 8554a1c880..1ce0e1d1cb 100644
--- a/transformers/tokenization_bert_japanese.py
+++ b/transformers/tokenization_bert_japanese.py
@@ -19,6 +19,7 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 import collections
 import logging
 import os
+import six
 import unicodedata
 from io import open
 
@@ -186,8 +187,13 @@ class MecabTokenizer(object):
         never_split = self.never_split + (never_split if never_split is not None else [])
         tokens = []
 
+        if six.PY2:
+            mecab_output = self.mecab.parse(text.encode('utf-8')).decode('utf-8')
+        else:
+            mecab_output = self.mecab.parse(text)
+
         cursor = 0
-        for line in self.mecab.parse(text).split('\n'):
+        for line in mecab_output.split('\n'):
             if line == 'EOS':
                 break
 

From 597ba7feb384316081c96955196fcb7abb2edf06 Mon Sep 17 00:00:00 2001
From: Masatoshi Suzuki <sappukei.suzuki@gmail.com>
Date: Thu, 5 Dec 2019 11:30:40 +0900
Subject: [PATCH 329/505] Support testing Japanese BERT tokenizers

---
 .circleci/config.yml | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/.circleci/config.yml b/.circleci/config.yml
index 01e6d82b33..97f5f25606 100644
--- a/.circleci/config.yml
+++ b/.circleci/config.yml
@@ -13,6 +13,8 @@ jobs:
             - run: sudo pip install --progress-bar off .
             - run: sudo pip install pytest codecov pytest-cov
             - run: sudo pip install tensorboardX scikit-learn
+            - run: sudo apt-get -y install libmecab-dev mecab mecab-ipadic-utf8 swig
+            - run: sudo pip install mecab-python3
             - run: python -m pytest -sv ./transformers/tests/ --cov
             - run: codecov
     build_py3_torch:
@@ -27,6 +29,8 @@ jobs:
             - run: sudo pip install --progress-bar off .
             - run: sudo pip install pytest codecov pytest-cov
             - run: sudo pip install tensorboardX scikit-learn
+            - run: sudo apt-get -y install libmecab-dev mecab mecab-ipadic-utf8 swig
+            - run: sudo pip install mecab-python3
             - run: python -m pytest -sv ./transformers/tests/ --cov
             - run: python -m pytest -sv ./examples/
             - run: codecov
@@ -42,6 +46,8 @@ jobs:
             - run: sudo pip install --progress-bar off .
             - run: sudo pip install pytest codecov pytest-cov
             - run: sudo pip install tensorboardX scikit-learn
+            - run: sudo apt-get -y install libmecab-dev mecab mecab-ipadic-utf8 swig
+            - run: sudo pip install mecab-python3
             - run: python -m pytest -sv ./transformers/tests/ --cov
             - run: codecov
     build_py2_torch:
@@ -55,6 +61,8 @@ jobs:
             - run: sudo pip install torch
             - run: sudo pip install --progress-bar off .
             - run: sudo pip install pytest codecov pytest-cov
+            - run: sudo apt-get -y install libmecab-dev mecab mecab-ipadic-utf8 swig
+            - run: sudo pip install mecab-python
             - run: python -m pytest -sv ./transformers/tests/ --cov
             - run: codecov
     build_py2_tf:
@@ -68,6 +76,8 @@ jobs:
             - run: sudo pip install tensorflow
             - run: sudo pip install --progress-bar off .
             - run: sudo pip install pytest codecov pytest-cov
+            - run: sudo apt-get -y install libmecab-dev mecab mecab-ipadic-utf8 swig
+            - run: sudo pip install mecab-python
             - run: python -m pytest -sv ./transformers/tests/ --cov
             - run: codecov
     deploy_doc:

From d2100428d3652cefbffcf0bd00f0881090d26333 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Tue, 10 Dec 2019 21:43:49 +0000
Subject: [PATCH 330/505] Update to new test infra and only run conditionally

---
 .circleci/config.yml                          | 20 ++++-----
 .../tests/tokenization_bert_japanese_test.py  |  9 ++--
 transformers/tests/utils.py                   | 42 +++++++++++++------
 3 files changed, 44 insertions(+), 27 deletions(-)

diff --git a/.circleci/config.yml b/.circleci/config.yml
index 97f5f25606..7ca5f8121c 100644
--- a/.circleci/config.yml
+++ b/.circleci/config.yml
@@ -13,8 +13,6 @@ jobs:
             - run: sudo pip install --progress-bar off .
             - run: sudo pip install pytest codecov pytest-cov
             - run: sudo pip install tensorboardX scikit-learn
-            - run: sudo apt-get -y install libmecab-dev mecab mecab-ipadic-utf8 swig
-            - run: sudo pip install mecab-python3
             - run: python -m pytest -sv ./transformers/tests/ --cov
             - run: codecov
     build_py3_torch:
@@ -29,8 +27,6 @@ jobs:
             - run: sudo pip install --progress-bar off .
             - run: sudo pip install pytest codecov pytest-cov
             - run: sudo pip install tensorboardX scikit-learn
-            - run: sudo apt-get -y install libmecab-dev mecab mecab-ipadic-utf8 swig
-            - run: sudo pip install mecab-python3
             - run: python -m pytest -sv ./transformers/tests/ --cov
             - run: python -m pytest -sv ./examples/
             - run: codecov
@@ -46,8 +42,6 @@ jobs:
             - run: sudo pip install --progress-bar off .
             - run: sudo pip install pytest codecov pytest-cov
             - run: sudo pip install tensorboardX scikit-learn
-            - run: sudo apt-get -y install libmecab-dev mecab mecab-ipadic-utf8 swig
-            - run: sudo pip install mecab-python3
             - run: python -m pytest -sv ./transformers/tests/ --cov
             - run: codecov
     build_py2_torch:
@@ -61,8 +55,6 @@ jobs:
             - run: sudo pip install torch
             - run: sudo pip install --progress-bar off .
             - run: sudo pip install pytest codecov pytest-cov
-            - run: sudo apt-get -y install libmecab-dev mecab mecab-ipadic-utf8 swig
-            - run: sudo pip install mecab-python
             - run: python -m pytest -sv ./transformers/tests/ --cov
             - run: codecov
     build_py2_tf:
@@ -76,10 +68,18 @@ jobs:
             - run: sudo pip install tensorflow
             - run: sudo pip install --progress-bar off .
             - run: sudo pip install pytest codecov pytest-cov
-            - run: sudo apt-get -y install libmecab-dev mecab mecab-ipadic-utf8 swig
-            - run: sudo pip install mecab-python
             - run: python -m pytest -sv ./transformers/tests/ --cov
             - run: codecov
+    build_py3_custom_tokenizers:
+        working_directory: ~/transformers
+        docker:
+            - image: circleci/python:3.5
+        steps:
+            - checkout
+            - run: sudo pip install --progress-bar off .
+            - run: sudo pip install pytest
+            - run: sudo pip install mecab-python3
+            - run: python -m pytest -sv ./transformers/tests/tokenization_bert_japanese_test.py
     deploy_doc:
         working_directory: ~/transformers
         docker:
diff --git a/transformers/tests/tokenization_bert_japanese_test.py b/transformers/tests/tokenization_bert_japanese_test.py
index 6f66b96411..545193c7cc 100644
--- a/transformers/tests/tokenization_bert_japanese_test.py
+++ b/transformers/tests/tokenization_bert_japanese_test.py
@@ -16,7 +16,6 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 
 import os
 import unittest
-import pytest
 from io import open
 
 from transformers.tokenization_bert import WordpieceTokenizer
@@ -25,8 +24,10 @@ from transformers.tokenization_bert_japanese import (BertJapaneseTokenizer,
                                                      VOCAB_FILES_NAMES)
 
 from .tokenization_tests_commons import CommonTestCases
+from .utils import slow, custom_tokenizers
 
 
+@custom_tokenizers
 class BertJapaneseTokenizationTest(CommonTestCases.CommonTokenizerTester):
 
     tokenizer_class = BertJapaneseTokenizer
@@ -104,7 +105,7 @@ class BertJapaneseTokenizationTest(CommonTestCases.CommonTokenizerTester):
         self.assertListEqual(tokenizer.tokenize(u"こんばんは こんばんにちは こんにちは"),
                              [u"こん", u"##ばんは", u"[UNK]", u"こんにちは"])
 
-    @pytest.mark.slow
+    @slow
     def test_sequence_builders(self):
         tokenizer = self.tokenizer_class.from_pretrained("bert-base-japanese")
 
@@ -172,7 +173,7 @@ class BertJapaneseCharacterTokenizationTest(CommonTestCases.CommonTokenizerTeste
         self.assertListEqual(tokenizer.tokenize(u"こんにちほ"),
                              [u"こ", u"ん", u"に", u"ち", u"[UNK]"])
 
-    @pytest.mark.slow
+    @slow
     def test_sequence_builders(self):
         tokenizer = self.tokenizer_class.from_pretrained("bert-base-japanese-char")
 
@@ -188,5 +189,3 @@ class BertJapaneseCharacterTokenizationTest(CommonTestCases.CommonTokenizerTeste
 
 
 
-if __name__ == '__main__':
-    unittest.main()
diff --git a/transformers/tests/utils.py b/transformers/tests/utils.py
index 7a51ab612b..2b97293ca7 100644
--- a/transformers/tests/utils.py
+++ b/transformers/tests/utils.py
@@ -6,18 +6,23 @@ from distutils.util import strtobool
 from transformers.file_utils import _tf_available, _torch_available
 
 
-try:
-    run_slow = os.environ["RUN_SLOW"]
-except KeyError:
-    # RUN_SLOW isn't set, default to skipping slow tests.
-    _run_slow_tests = False
-else:
-    # RUN_SLOW is set, convert it to True or False.
+def parse_flag_from_env(key, default=False):
     try:
-        _run_slow_tests = strtobool(run_slow)
-    except ValueError:
-        # More values are supported, but let's keep the message simple.
-        raise ValueError("If set, RUN_SLOW must be yes or no.")
+        value = os.environ[key]
+    except KeyError:
+        # KEY isn't set, default to `default`.
+        _value = default
+    else:
+        # KEY is set, convert it to True or False.
+        try:
+            _value = strtobool(value)
+        except ValueError:
+            # More values are supported, but let's keep the message simple.
+            raise ValueError("If set, {} must be yes or no.".format(key))
+    return _value
+
+_run_slow_tests = parse_flag_from_env("RUN_SLOW", default=False)
+_run_custom_tokenizers = parse_flag_from_env("RUN_CUSTOM_TOKENIZERS", default=False)
 
 
 def slow(test_case):
@@ -33,6 +38,19 @@ def slow(test_case):
     return test_case
 
 
+def custom_tokenizers(test_case):
+    """
+    Decorator marking a test for a custom tokenizer.
+
+    Custom tokenizers require additional dependencies, and are skipped
+    by default. Set the RUN_CUSTOM_TOKENIZERS environment variable
+    to a truthy value to run them.
+    """
+    if not _run_custom_tokenizers:
+        test_case = unittest.skip("test of custom tokenizers")(test_case)
+    return test_case
+
+
 def require_torch(test_case):
     """
     Decorator marking a test that requires PyTorch.
@@ -59,6 +77,6 @@ def require_tf(test_case):
 
 if _torch_available:
     # Set the USE_CUDA environment variable to select a GPU.
-    torch_device = "cuda" if os.environ.get("USE_CUDA") else "cpu"
+    torch_device = "cuda" if parse_flag_from_env("USE_CUDA") else "cpu"
 else:
     torch_device = None

From 95854c4a2f8d418a14e64b4edf64fc7363b1ff10 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Tue, 10 Dec 2019 21:46:00 +0000
Subject: [PATCH 331/505] Actually run the tests

---
 .circleci/config.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.circleci/config.yml b/.circleci/config.yml
index 7ca5f8121c..d8f624a0e5 100644
--- a/.circleci/config.yml
+++ b/.circleci/config.yml
@@ -79,7 +79,7 @@ jobs:
             - run: sudo pip install --progress-bar off .
             - run: sudo pip install pytest
             - run: sudo pip install mecab-python3
-            - run: python -m pytest -sv ./transformers/tests/tokenization_bert_japanese_test.py
+            - run: RUN_CUSTOM_TOKENIZERS=1 python -m pytest -sv ./transformers/tests/tokenization_bert_japanese_test.py
     deploy_doc:
         working_directory: ~/transformers
         docker:

From 9cb97c0c0f7215971bb5a39cd070e5bd89319bdf Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Tue, 10 Dec 2019 21:48:56 +0000
Subject: [PATCH 332/505] Actually run the tests

---
 .circleci/config.yml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/.circleci/config.yml b/.circleci/config.yml
index d8f624a0e5..9d6e02d580 100644
--- a/.circleci/config.yml
+++ b/.circleci/config.yml
@@ -101,6 +101,7 @@ workflows:
     version: 2
     build_and_test:
         jobs:
+            - build_py3_custom_tokenizers
             - build_py3_torch_and_tf
             - build_py3_torch
             - build_py3_tf

From 5505cf701477762cedf792e20344d29bc8bf6325 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Tue, 10 Dec 2019 21:53:44 +0000
Subject: [PATCH 333/505] Run tests on Py2 too, for Lysandre

---
 .circleci/config.yml | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/.circleci/config.yml b/.circleci/config.yml
index 9d6e02d580..afc6d5ce44 100644
--- a/.circleci/config.yml
+++ b/.circleci/config.yml
@@ -80,6 +80,16 @@ jobs:
             - run: sudo pip install pytest
             - run: sudo pip install mecab-python3
             - run: RUN_CUSTOM_TOKENIZERS=1 python -m pytest -sv ./transformers/tests/tokenization_bert_japanese_test.py
+    build_py2_custom_tokenizers:
+        working_directory: ~/transformers
+        docker:
+            - image: circleci/python:2.7
+        steps:
+            - checkout
+            - run: sudo pip install --progress-bar off .
+            - run: sudo pip install pytest
+            - run: sudo pip install mecab-python
+            - run: RUN_CUSTOM_TOKENIZERS=1 python -m pytest -sv ./transformers/tests/tokenization_bert_japanese_test.py
     deploy_doc:
         working_directory: ~/transformers
         docker:
@@ -102,6 +112,7 @@ workflows:
     build_and_test:
         jobs:
             - build_py3_custom_tokenizers
+            - build_py2_custom_tokenizers
             - build_py3_torch_and_tf
             - build_py3_torch
             - build_py3_tf

From 371c5ddfad96689771465aff557152322190b60e Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Tue, 10 Dec 2019 21:55:43 +0000
Subject: [PATCH 334/505] Py2 tests for Lysandre

---
 .circleci/config.yml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/.circleci/config.yml b/.circleci/config.yml
index afc6d5ce44..c827a81fbb 100644
--- a/.circleci/config.yml
+++ b/.circleci/config.yml
@@ -88,6 +88,7 @@ jobs:
             - checkout
             - run: sudo pip install --progress-bar off .
             - run: sudo pip install pytest
+            - run: sudo apt-get -y install libmecab-dev mecab mecab-ipadic-utf8 swig
             - run: sudo pip install mecab-python
             - run: RUN_CUSTOM_TOKENIZERS=1 python -m pytest -sv ./transformers/tests/tokenization_bert_japanese_test.py
     deploy_doc:

From 36fc52a3b4b50885d5ec3bf259f81740e19d8b3c Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Tue, 10 Dec 2019 22:03:35 +0000
Subject: [PATCH 335/505] Update links to weights

---
 transformers/configuration_bert.py         | 8 ++++----
 transformers/modeling_bert.py              | 8 ++++----
 transformers/modeling_tf_bert.py           | 8 ++++----
 transformers/tokenization_bert_japanese.py | 8 ++++----
 4 files changed, 16 insertions(+), 16 deletions(-)

diff --git a/transformers/configuration_bert.py b/transformers/configuration_bert.py
index 16f1f60404..01fcd88cb8 100644
--- a/transformers/configuration_bert.py
+++ b/transformers/configuration_bert.py
@@ -42,10 +42,10 @@ BERT_PRETRAINED_CONFIG_ARCHIVE_MAP = {
     'bert-base-cased-finetuned-mrpc': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-cased-finetuned-mrpc-config.json",
     'bert-base-german-dbmdz-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-german-dbmdz-cased-config.json",
     'bert-base-german-dbmdz-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-german-dbmdz-uncased-config.json",
-    'bert-base-japanese': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-config.json",
-    'bert-base-japanese-whole-word-masking': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-whole-word-masking-config.json",
-    'bert-base-japanese-char': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-char-config.json",
-    'bert-base-japanese-char-whole-word-masking': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-char-whole-word-masking-config.json"
+    'bert-base-japanese': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-config.json",
+    'bert-base-japanese-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-whole-word-masking-config.json",
+    'bert-base-japanese-char': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-config.json",
+    'bert-base-japanese-char-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-whole-word-masking-config.json"
 }
 
 
diff --git a/transformers/modeling_bert.py b/transformers/modeling_bert.py
index e2e115a015..d0f35272ac 100644
--- a/transformers/modeling_bert.py
+++ b/transformers/modeling_bert.py
@@ -48,10 +48,10 @@ BERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
     'bert-base-cased-finetuned-mrpc': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-cased-finetuned-mrpc-pytorch_model.bin",
     'bert-base-german-dbmdz-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-german-dbmdz-cased-pytorch_model.bin",
     'bert-base-german-dbmdz-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-german-dbmdz-uncased-pytorch_model.bin",
-    'bert-base-japanese': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-pytorch_model.bin",
-    'bert-base-japanese-whole-word-masking': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-whole-word-masking-pytorch_model.bin",
-    'bert-base-japanese-char': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-char-pytorch_model.bin",
-    'bert-base-japanese-char-whole-word-masking': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-char-whole-word-masking-pytorch_model.bin"
+    'bert-base-japanese': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-pytorch_model.bin",
+    'bert-base-japanese-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-whole-word-masking-pytorch_model.bin",
+    'bert-base-japanese-char': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-pytorch_model.bin",
+    'bert-base-japanese-char-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-whole-word-masking-pytorch_model.bin"
 }
 
 
diff --git a/transformers/modeling_tf_bert.py b/transformers/modeling_tf_bert.py
index 27dd311a4d..7cc71f5063 100644
--- a/transformers/modeling_tf_bert.py
+++ b/transformers/modeling_tf_bert.py
@@ -48,10 +48,10 @@ TF_BERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
     'bert-large-uncased-whole-word-masking-finetuned-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-uncased-whole-word-masking-finetuned-squad-tf_model.h5",
     'bert-large-cased-whole-word-masking-finetuned-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-cased-whole-word-masking-finetuned-squad-tf_model.h5",
     'bert-base-cased-finetuned-mrpc': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-cased-finetuned-mrpc-tf_model.h5",
-    'bert-base-japanese': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-tf_model.h5",
-    'bert-base-japanese-whole-word-masking': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-whole-word-masking-tf_model.h5",
-    'bert-base-japanese-char': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-char-tf_model.h5",
-    'bert-base-japanese-char-whole-word-masking': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-char-whole-word-masking-tf_model.h5"
+    'bert-base-japanese': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-tf_model.h5",
+    'bert-base-japanese-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-whole-word-masking-tf_model.h5",
+    'bert-base-japanese-char': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-tf_model.h5",
+    'bert-base-japanese-char-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-whole-word-masking-tf_model.h5"
 }
 
 
diff --git a/transformers/tokenization_bert_japanese.py b/transformers/tokenization_bert_japanese.py
index 1ce0e1d1cb..0ff45cbfe7 100644
--- a/transformers/tokenization_bert_japanese.py
+++ b/transformers/tokenization_bert_japanese.py
@@ -33,10 +33,10 @@ VOCAB_FILES_NAMES = {'vocab_file': 'vocab.txt'}
 PRETRAINED_VOCAB_FILES_MAP = {
     'vocab_file':
     {
-        'bert-base-japanese': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-vocab.txt",
-        'bert-base-japanese-whole-word-masking': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-whole-word-masking-vocab.txt",
-        'bert-base-japanese-char': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-char-vocab.txt",
-        'bert-base-japanese-char-whole-word-masking': "https://www.nlp.ecei.tohoku.ac.jp/~m-suzuki/bert-japanese/bert-base-japanese-char-whole-word-masking-vocab.txt"
+        'bert-base-japanese': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-vocab.txt",
+        'bert-base-japanese-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-whole-word-masking-vocab.txt",
+        'bert-base-japanese-char': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-vocab.txt",
+        'bert-base-japanese-char-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-whole-word-masking-vocab.txt"
     }
 }
 

From 1748fdf657ed804f3edc1e45077b703cd8d6e4c5 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Wed, 11 Dec 2019 23:31:23 +0000
Subject: [PATCH 336/505] [doc] Fix rst table

---
 docs/source/pretrained_models.rst | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/docs/source/pretrained_models.rst b/docs/source/pretrained_models.rst
index d3498e057d..775772e896 100644
--- a/docs/source/pretrained_models.rst
+++ b/docs/source/pretrained_models.rst
@@ -63,22 +63,22 @@ Here is the full list of the currently provided pretrained models together with
 |                   |                                                            | (see `details on dbmdz repository <https://github.com/dbmdz/german-bert>`__).                                                         |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``bert-base-japanese``                                     | | 12-layer, 768-hidden, 12-heads, 110M parameters.                                                                                    |
-|                   |                                                            | | Trained on Japanese text. Text is tokenized with MeCab and WordPiece.                                                 |
-|                   |                                                            | | `MeCab <https://taku910.github.io/mecab/>`__ is required for tokenization.                                                            |
-|                   |                                                            | (see `details on cl-tohoku repository <https://github.com/cl-tohoku/bert-japanese>`__).                                             |
+|                   |                                                            | | Trained on Japanese text. Text is tokenized with MeCab and WordPiece.                                                               |
+|                   |                                                            | | `MeCab <https://taku910.github.io/mecab/>`__ is required for tokenization.                                                          |
+|                   |                                                            | (see `details on cl-tohoku repository <https://github.com/cl-tohoku/bert-japanese>`__).                                               |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``bert-base-japanese-whole-word-masking``                  | | 12-layer, 768-hidden, 12-heads, 110M parameters.                                                                                    |
-|                   |                                                            | | Trained on Japanese text using Whole-Word-Masking. Text is tokenized with MeCab and WordPiece.                        |
-|                   |                                                            | | `MeCab <https://taku910.github.io/mecab/>`__ is required for tokenization.                                                            |
-|                   |                                                            | (see `details on cl-tohoku repository <https://github.com/cl-tohoku/bert-japanese>`__).                                             |
+|                   |                                                            | | Trained on Japanese text using Whole-Word-Masking. Text is tokenized with MeCab and WordPiece.                                      |
+|                   |                                                            | | `MeCab <https://taku910.github.io/mecab/>`__ is required for tokenization.                                                          |
+|                   |                                                            | (see `details on cl-tohoku repository <https://github.com/cl-tohoku/bert-japanese>`__).                                               |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``bert-base-japanese-char``                                | | 12-layer, 768-hidden, 12-heads, 110M parameters.                                                                                    |
-|                   |                                                            | | Trained on Japanese text. Text is tokenized into characters.                                                          |
-|                   |                                                            | (see `details on cl-tohoku repository <https://github.com/cl-tohoku/bert-japanese>`__).                                             |
+|                   |                                                            | | Trained on Japanese text. Text is tokenized into characters.                                                                        |
+|                   |                                                            | (see `details on cl-tohoku repository <https://github.com/cl-tohoku/bert-japanese>`__).                                               |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``bert-base-japanese-char-whole-word-masking``             | | 12-layer, 768-hidden, 12-heads, 110M parameters.                                                                                    |
-|                   |                                                            | | Trained on Japanese text using Whole-Word-Masking. Text is tokenized into characters.                                 |
-|                   |                                                            | (see `details on cl-tohoku repository <https://github.com/cl-tohoku/bert-japanese>`__).                                             |
+|                   |                                                            | | Trained on Japanese text using Whole-Word-Masking. Text is tokenized into characters.                                               |
+|                   |                                                            | (see `details on cl-tohoku repository <https://github.com/cl-tohoku/bert-japanese>`__).                                               |
 +-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 | GPT               | ``openai-gpt``                                             | | 12-layer, 768-hidden, 12-heads, 110M parameters.                                                                                    |
 |                   |                                                            | | OpenAI GPT English model                                                                                                            |

From 413f41921b650418798f7d5c246316c4e1e5eb5d Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Thu, 12 Dec 2019 07:34:42 +0100
Subject: [PATCH 337/505] fix merge

---
 transformers/tests/utils.py | 8 --------
 1 file changed, 8 deletions(-)

diff --git a/transformers/tests/utils.py b/transformers/tests/utils.py
index daed431995..c950ad8f17 100644
--- a/transformers/tests/utils.py
+++ b/transformers/tests/utils.py
@@ -9,14 +9,6 @@ from transformers.file_utils import _tf_available, _torch_available
 SMALL_MODEL_IDENTIFIER = "julien-c/bert-xsmall-dummy"
 
 
-try:
-    run_slow = os.environ["RUN_SLOW"]
-except KeyError:
-    # RUN_SLOW isn't set, default to skipping slow tests.
-    _run_slow_tests = False
-else:
-    # RUN_SLOW is set, convert it to True or False.
-
 def parse_flag_from_env(key, default=False):
     try:
         value = os.environ[key]

From f69dbecc38e04cd4d158afb273921ca7b75c7cba Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Thu, 12 Dec 2019 10:25:36 +0100
Subject: [PATCH 338/505] Expose classification labels mapping (and reverse) in
 model config.

---
 transformers/configuration_utils.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/transformers/configuration_utils.py b/transformers/configuration_utils.py
index 08cee75d81..97b9fa8f80 100644
--- a/transformers/configuration_utils.py
+++ b/transformers/configuration_utils.py
@@ -58,6 +58,8 @@ class PretrainedConfig(object):
         self.use_bfloat16 = kwargs.pop('use_bfloat16', False)
         self.pruned_heads = kwargs.pop('pruned_heads', {})
         self.is_decoder = kwargs.pop('is_decoder', False)
+        self.idx2label = kwargs.pop('idx2label', {i: 'LABEL_{}'.format(i) for i in range(self.num_labels)})
+        self.label2idx = kwargs.pop('label2idx', dict(zip(self.idx2label.values(), self.idx2label.keys())))
 
     def save_pretrained(self, save_directory):
         """ Save a configuration object to the directory `save_directory`, so that it

From f19dad61c70a628545612e435c699263f02bc4a0 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Thu, 12 Dec 2019 14:46:30 +0100
Subject: [PATCH 339/505] fixing XLM conversion tests with dummy input

---
 transformers/modeling_tf_pytorch_utils.py |  6 +++++-
 transformers/modeling_tf_xlm.py           |  2 +-
 transformers/modeling_xlm.py              | 12 +++++++++++-
 3 files changed, 17 insertions(+), 3 deletions(-)

diff --git a/transformers/modeling_tf_pytorch_utils.py b/transformers/modeling_tf_pytorch_utils.py
index 510e130c90..9d2b663dcb 100644
--- a/transformers/modeling_tf_pytorch_utils.py
+++ b/transformers/modeling_tf_pytorch_utils.py
@@ -78,6 +78,7 @@ def load_pytorch_checkpoint_in_tf2_model(tf_model, pytorch_checkpoint_path, tf_i
     logger.info("Loading PyTorch weights from {}".format(pt_path))
 
     pt_state_dict = torch.load(pt_path, map_location='cpu')
+    logger.info("PyTorch checkpoint contains {:,} parameters".format(sum(t.numel() for t in pt_state_dict.values())))
 
     return load_pytorch_weights_in_tf2_model(tf_model, pt_state_dict, tf_inputs=tf_inputs, allow_missing_keys=allow_missing_keys)
 
@@ -134,7 +135,7 @@ def load_pytorch_weights_in_tf2_model(tf_model, pt_state_dict, tf_inputs=None, a
         start_prefix_to_remove = tf_model.base_model_prefix + '.'
 
     symbolic_weights = tf_model.trainable_weights + tf_model.non_trainable_weights
-
+    tf_loaded_numel = 0
     weight_value_tuples = []
     all_pytorch_weights = set(list(pt_state_dict.keys()))
     for symbolic_weight in symbolic_weights:
@@ -159,6 +160,7 @@ def load_pytorch_weights_in_tf2_model(tf_model, pt_state_dict, tf_inputs=None, a
             e.args += (symbolic_weight.shape, array.shape)
             raise e
 
+        tf_loaded_numel += array.size
         # logger.warning("Initialize TF weight {}".format(symbolic_weight.name))
 
         weight_value_tuples.append((symbolic_weight, array))
@@ -169,6 +171,8 @@ def load_pytorch_weights_in_tf2_model(tf_model, pt_state_dict, tf_inputs=None, a
     if tf_inputs is not None:
         tfo = tf_model(tf_inputs, training=False)  # Make sure restore ops are run
 
+    logger.info("Loaded {:,} parameters in the TF 2.0 model.".format(tf_loaded_numel))
+
     logger.info("Weights or buffers not loaded from PyTorch model: {}".format(all_pytorch_weights))
 
     return tf_model
diff --git a/transformers/modeling_tf_xlm.py b/transformers/modeling_tf_xlm.py
index 6f11b0537d..903a8596c3 100644
--- a/transformers/modeling_tf_xlm.py
+++ b/transformers/modeling_tf_xlm.py
@@ -460,7 +460,7 @@ class TFXLMPreTrainedModel(TFPreTrainedModel):
             langs_list = tf.constant([[1, 1, 0, 0, 1], [1, 1, 1, 0, 0], [1, 0, 0, 1, 1]])
         else:
             langs_list = None
-        return [inputs_list, attns_list, langs_list]
+        return {'input_ids': inputs_list, 'attention_mask': attns_list, 'langs': langs_list}
 
 
 XLM_START_DOCSTRING = r"""    The XLM model was proposed in
diff --git a/transformers/modeling_xlm.py b/transformers/modeling_xlm.py
index 257f0da394..b604ae669d 100644
--- a/transformers/modeling_xlm.py
+++ b/transformers/modeling_xlm.py
@@ -227,6 +227,16 @@ class XLMPreTrainedModel(PreTrainedModel):
     def __init__(self, *inputs, **kwargs):
         super(XLMPreTrainedModel, self).__init__(*inputs, **kwargs)
 
+    @property
+    def dummy_inputs(self):
+        inputs_list = torch.tensor([[7, 6, 0, 0, 1], [1, 2, 3, 0, 0], [0, 0, 0, 4, 5]])
+        attns_list = torch.tensor([[1, 1, 0, 0, 1], [1, 1, 1, 0, 0], [1, 0, 0, 1, 1]])
+        if self.config.use_lang_emb and self.config.n_langs > 1:
+            langs_list = torch.tensor([[1, 1, 0, 0, 1], [1, 1, 1, 0, 0], [1, 0, 0, 1, 1]])
+        else:
+            langs_list = None
+        return {'input_ids': inputs_list, 'attention_mask': attns_list, 'langs': langs_list}
+
     def _init_weights(self, module):
         """ Initialize the weights. """
         if isinstance(module, nn.Embedding):
@@ -646,7 +656,7 @@ class XLMWithLMHeadModel(XLMPreTrainedModel):
                                                langs=langs,
                                                token_type_ids=token_type_ids,
                                                position_ids=position_ids,
-                                               lengths=lengths, 
+                                               lengths=lengths,
                                                cache=cache,
                                                head_mask=head_mask,
                                                inputs_embeds=inputs_embeds)

From fbf5455a8607fa660aacbf06c16f6fe23758b13d Mon Sep 17 00:00:00 2001
From: Alan deLevie <alan@casetext.com>
Date: Wed, 11 Dec 2019 10:14:48 -0500
Subject: [PATCH 340/505] Fix typo in examples/run_glue.py args declaration.

deay -> decay
---
 examples/run_glue.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/examples/run_glue.py b/examples/run_glue.py
index 369a7110ab..1a51255c11 100644
--- a/examples/run_glue.py
+++ b/examples/run_glue.py
@@ -380,7 +380,7 @@ def main():
     parser.add_argument("--learning_rate", default=5e-5, type=float,
                         help="The initial learning rate for Adam.")
     parser.add_argument("--weight_decay", default=0.0, type=float,
-                        help="Weight deay if we apply some.")
+                        help="Weight decay if we apply some.")
     parser.add_argument("--adam_epsilon", default=1e-8, type=float,
                         help="Epsilon for Adam optimizer.")
     parser.add_argument("--max_grad_norm", default=1.0, type=float,

From fe92755b992eb61239ad361abae3b71f86bbbba1 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Thu, 12 Dec 2019 11:37:19 -0500
Subject: [PATCH 341/505] Fix special tokens mask in encode

---
 transformers/tokenization_utils.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index f44b77b27c..7e86742286 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -973,7 +973,7 @@ class PreTrainedTokenizer(object):
             token_type_ids = [0] * len(ids) + ([1] * len(pair_ids) if pair else [])
 
         if return_special_tokens_mask:
-            encoded_inputs["special_tokens_mask"] = special_tokens_mask
+            encoded_inputs["special_tokens_mask"] = self.get_special_tokens_mask(ids, pair_ids)
 
         encoded_inputs["input_ids"] = sequence
         if return_token_type_ids:

From 5d67aa21aefaaa62594e8dfb56093b83c5f547bb Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Thu, 12 Dec 2019 12:39:41 -0500
Subject: [PATCH 342/505] [doc] Replicate doc from #2144

---
 transformers/configuration_auto.py       |  1 +
 transformers/configuration_utils.py      |  1 +
 transformers/modeling_auto.py            |  4 ++++
 transformers/modeling_encoder_decoder.py |  2 ++
 transformers/modeling_tf_auto.py         |  4 ++++
 transformers/modeling_tf_utils.py        |  1 +
 transformers/modeling_utils.py           |  1 +
 transformers/tokenization_auto.py        | 11 +++++++++--
 transformers/tokenization_utils.py       |  4 ++--
 9 files changed, 25 insertions(+), 4 deletions(-)

diff --git a/transformers/configuration_auto.py b/transformers/configuration_auto.py
index 43f251bd0c..fbc5c59199 100644
--- a/transformers/configuration_auto.py
+++ b/transformers/configuration_auto.py
@@ -83,6 +83,7 @@ class AutoConfig(object):
             pretrained_model_name_or_path: either:
 
                 - a string with the `shortcut name` of a pre-trained model configuration to load from cache or download, e.g.: ``bert-base-uncased``.
+                - a string with the `identifier name` of a pre-trained model configuration that was user-uploaded to our S3, e.g.: ``dbmdz/bert-base-german-cased``.
                 - a path to a `directory` containing a configuration file saved using the :func:`~transformers.PretrainedConfig.save_pretrained` method, e.g.: ``./my_model_directory/``.
                 - a path or url to a saved configuration JSON `file`, e.g.: ``./my_model_directory/configuration.json``.
 
diff --git a/transformers/configuration_utils.py b/transformers/configuration_utils.py
index 8ae30f2a48..82959adb57 100644
--- a/transformers/configuration_utils.py
+++ b/transformers/configuration_utils.py
@@ -79,6 +79,7 @@ class PretrainedConfig(object):
             pretrained_model_name_or_path: either:
 
                 - a string with the `shortcut name` of a pre-trained model configuration to load from cache or download, e.g.: ``bert-base-uncased``.
+                - a string with the `identifier name` of a pre-trained model configuration that was user-uploaded to our S3, e.g.: ``dbmdz/bert-base-german-cased``.
                 - a path to a `directory` containing a configuration file saved using the :func:`~transformers.PretrainedConfig.save_pretrained` method, e.g.: ``./my_model_directory/``.
                 - a path or url to a saved configuration JSON `file`, e.g.: ``./my_model_directory/configuration.json``.
 
diff --git a/transformers/modeling_auto.py b/transformers/modeling_auto.py
index 6ba1aab7a3..96f45d8ec4 100644
--- a/transformers/modeling_auto.py
+++ b/transformers/modeling_auto.py
@@ -93,6 +93,7 @@ class AutoModel(object):
             pretrained_model_name_or_path: either:
 
                 - a string with the `shortcut name` of a pre-trained model to load from cache or download, e.g.: ``bert-base-uncased``.
+                - a string with the `identifier name` of a pre-trained model that was user-uploaded to our S3, e.g.: ``dbmdz/bert-base-german-cased``.
                 - a path to a `directory` containing model weights saved using :func:`~transformers.PreTrainedModel.save_pretrained`, e.g.: ``./my_model_directory/``.
                 - a path or url to a `tensorflow index checkpoint file` (e.g. `./tf_model/model.ckpt.index`). In this case, ``from_tf`` should be set to True and a configuration object should be provided as ``config`` argument. This loading path is slower than converting the TensorFlow checkpoint in a PyTorch model using the provided conversion scripts and loading the PyTorch model afterwards.
 
@@ -231,6 +232,7 @@ class AutoModelWithLMHead(object):
             pretrained_model_name_or_path: either:
 
                 - a string with the `shortcut name` of a pre-trained model to load from cache or download, e.g.: ``bert-base-uncased``.
+                - a string with the `identifier name` of a pre-trained model that was user-uploaded to our S3, e.g.: ``dbmdz/bert-base-german-cased``.
                 - a path to a `directory` containing model weights saved using :func:`~transformers.PreTrainedModel.save_pretrained`, e.g.: ``./my_model_directory/``.
                 - a path or url to a `tensorflow index checkpoint file` (e.g. `./tf_model/model.ckpt.index`). In this case, ``from_tf`` should be set to True and a configuration object should be provided as ``config`` argument. This loading path is slower than converting the TensorFlow checkpoint in a PyTorch model using the provided conversion scripts and loading the PyTorch model afterwards.
 
@@ -360,6 +362,7 @@ class AutoModelForSequenceClassification(object):
             pretrained_model_name_or_path: either:
 
                 - a string with the `shortcut name` of a pre-trained model to load from cache or download, e.g.: ``bert-base-uncased``.
+                - a string with the `identifier name` of a pre-trained model that was user-uploaded to our S3, e.g.: ``dbmdz/bert-base-german-cased``.
                 - a path to a `directory` containing model weights saved using :func:`~transformers.PreTrainedModel.save_pretrained`, e.g.: ``./my_model_directory/``.
                 - a path or url to a `tensorflow index checkpoint file` (e.g. `./tf_model/model.ckpt.index`). In this case, ``from_tf`` should be set to True and a configuration object should be provided as ``config`` argument. This loading path is slower than converting the TensorFlow checkpoint in a PyTorch model using the provided conversion scripts and loading the PyTorch model afterwards.
 
@@ -478,6 +481,7 @@ class AutoModelForQuestionAnswering(object):
             pretrained_model_name_or_path: either:
 
                 - a string with the `shortcut name` of a pre-trained model to load from cache or download, e.g.: ``bert-base-uncased``.
+                - a string with the `identifier name` of a pre-trained model that was user-uploaded to our S3, e.g.: ``dbmdz/bert-base-german-cased``.
                 - a path to a `directory` containing model weights saved using :func:`~transformers.PreTrainedModel.save_pretrained`, e.g.: ``./my_model_directory/``.
                 - a path or url to a `tensorflow index checkpoint file` (e.g. `./tf_model/model.ckpt.index`). In this case, ``from_tf`` should be set to True and a configuration object should be provided as ``config`` argument. This loading path is slower than converting the TensorFlow checkpoint in a PyTorch model using the provided conversion scripts and loading the PyTorch model afterwards.
 
diff --git a/transformers/modeling_encoder_decoder.py b/transformers/modeling_encoder_decoder.py
index a884abd0a2..70f765b849 100644
--- a/transformers/modeling_encoder_decoder.py
+++ b/transformers/modeling_encoder_decoder.py
@@ -59,12 +59,14 @@ class PreTrainedEncoderDecoder(nn.Module):
             encoder_pretrained_model_name_or_path: information necessary to initiate the encoder. Either:
 
                 - a string with the `shortcut name` of a pre-trained model to load from cache or download, e.g.: ``bert-base-uncased``.
+                - a string with the `identifier name` of a pre-trained model that was user-uploaded to our S3, e.g.: ``dbmdz/bert-base-german-cased``.
                 - a path to a `directory` containing model weights saved using :func:`~transformers.PreTrainedModel.save_pretrained`, e.g.: ``./my_model_directory/encoder``.
                 - a path or url to a `tensorflow index checkpoint file` (e.g. `./tf_model/model.ckpt.index`). In this case, ``from_tf`` should be set to True and a configuration object should be provided as ``config`` argument. This loading path is slower than converting the TensorFlow checkpoint in a PyTorch model using the provided conversion scripts and loading the PyTorch model afterwards.
 
             decoder_pretrained_model_name_or_path: information necessary to initiate the decoder. Either:
 
                 - a string with the `shortcut name` of a pre-trained model to load from cache or download, e.g.: ``bert-base-uncased``.
+                - a string with the `identifier name` of a pre-trained model that was user-uploaded to our S3, e.g.: ``dbmdz/bert-base-german-cased``.
                 - a path to a `directory` containing model weights saved using :func:`~transformers.PreTrainedModel.save_pretrained`, e.g.: ``./my_model_directory/decoder``.
                 - a path or url to a `tensorflow index checkpoint file` (e.g. `./tf_model/model.ckpt.index`). In this case, ``from_tf`` should be set to True and a configuration object should be provided as ``config`` argument. This loading path is slower than converting the TensorFlow checkpoint in a PyTorch model using the provided conversion scripts and loading the PyTorch model afterwards.
 
diff --git a/transformers/modeling_tf_auto.py b/transformers/modeling_tf_auto.py
index cfe19ead2a..fac92eb866 100644
--- a/transformers/modeling_tf_auto.py
+++ b/transformers/modeling_tf_auto.py
@@ -81,6 +81,7 @@ class TFAutoModel(object):
             pretrained_model_name_or_path: either:
 
                 - a string with the `shortcut name` of a pre-trained model to load from cache or download, e.g.: ``bert-base-uncased``.
+                - a string with the `identifier name` of a pre-trained model that was user-uploaded to our S3, e.g.: ``dbmdz/bert-base-german-cased``.
                 - a path to a `directory` containing model weights saved using :func:`~transformers.PreTrainedModel.save_pretrained`, e.g.: ``./my_model_directory/``.
                 - a path or url to a `PyTorch, TF 1.X or TF 2.0 checkpoint file` (e.g. `./tf_model/model.ckpt.index`). In the case of a PyTorch checkpoint, ``from_pt`` should be set to True and a configuration object should be provided as ``config`` argument.
 
@@ -212,6 +213,7 @@ class TFAutoModelWithLMHead(object):
             pretrained_model_name_or_path: either:
 
                 - a string with the `shortcut name` of a pre-trained model to load from cache or download, e.g.: ``bert-base-uncased``.
+                - a string with the `identifier name` of a pre-trained model that was user-uploaded to our S3, e.g.: ``dbmdz/bert-base-german-cased``.
                 - a path to a `directory` containing model weights saved using :func:`~transformers.PreTrainedModel.save_pretrained`, e.g.: ``./my_model_directory/``.
                 - a path or url to a `PyTorch, TF 1.X or TF 2.0 checkpoint file` (e.g. `./tf_model/model.ckpt.index`). In the case of a PyTorch checkpoint, ``from_pt`` should be set to True and a configuration object should be provided as ``config`` argument.
 
@@ -338,6 +340,7 @@ class TFAutoModelForSequenceClassification(object):
             pretrained_model_name_or_path: either:
 
                 - a string with the `shortcut name` of a pre-trained model to load from cache or download, e.g.: ``bert-base-uncased``.
+                - a string with the `identifier name` of a pre-trained model that was user-uploaded to our S3, e.g.: ``dbmdz/bert-base-german-cased``.
                 - a path to a `directory` containing model weights saved using :func:`~transformers.PreTrainedModel.save_pretrained`, e.g.: ``./my_model_directory/``.
                 - a path or url to a `PyTorch, TF 1.X or TF 2.0 checkpoint file` (e.g. `./tf_model/model.ckpt.index`). In the case of a PyTorch checkpoint, ``from_pt`` should be set to True and a configuration object should be provided as ``config`` argument.
 
@@ -453,6 +456,7 @@ class TFAutoModelForQuestionAnswering(object):
             pretrained_model_name_or_path: either:
 
                 - a string with the `shortcut name` of a pre-trained model to load from cache or download, e.g.: ``bert-base-uncased``.
+                - a string with the `identifier name` of a pre-trained model that was user-uploaded to our S3, e.g.: ``dbmdz/bert-base-german-cased``.
                 - a path to a `directory` containing model weights saved using :func:`~transformers.PreTrainedModel.save_pretrained`, e.g.: ``./my_model_directory/``.
                 - a path or url to a `PyTorch, TF 1.X or TF 2.0 checkpoint file` (e.g. `./tf_model/model.ckpt.index`). In the case of a PyTorch checkpoint, ``from_pt`` should be set to True and a configuration object should be provided as ``config`` argument.
 
diff --git a/transformers/modeling_tf_utils.py b/transformers/modeling_tf_utils.py
index 4a6d18f447..d9a93af21b 100644
--- a/transformers/modeling_tf_utils.py
+++ b/transformers/modeling_tf_utils.py
@@ -177,6 +177,7 @@ class TFPreTrainedModel(tf.keras.Model):
             pretrained_model_name_or_path: either:
 
                 - a string with the `shortcut name` of a pre-trained model to load from cache or download, e.g.: ``bert-base-uncased``.
+                - a string with the `identifier name` of a pre-trained model that was user-uploaded to our S3, e.g.: ``dbmdz/bert-base-german-cased``.
                 - a path to a `directory` containing model weights saved using :func:`~transformers.PreTrainedModel.save_pretrained`, e.g.: ``./my_model_directory/``.
                 - a path or url to a `PyTorch state_dict save file` (e.g. `./pt_model/pytorch_model.bin`). In this case, ``from_pt`` should be set to True and a configuration object should be provided as ``config`` argument. This loading path is slower than converting the PyTorch checkpoint in a TensorFlow model using the provided conversion scripts and loading the TensorFlow model afterwards.
 
diff --git a/transformers/modeling_utils.py b/transformers/modeling_utils.py
index 37088f8e67..676f355986 100644
--- a/transformers/modeling_utils.py
+++ b/transformers/modeling_utils.py
@@ -266,6 +266,7 @@ class PreTrainedModel(nn.Module):
             pretrained_model_name_or_path: either:
 
                 - a string with the `shortcut name` of a pre-trained model to load from cache or download, e.g.: ``bert-base-uncased``.
+                - a string with the `identifier name` of a pre-trained model that was user-uploaded to our S3, e.g.: ``dbmdz/bert-base-german-cased``.
                 - a path to a `directory` containing model weights saved using :func:`~transformers.PreTrainedModel.save_pretrained`, e.g.: ``./my_model_directory/``.
                 - a path or url to a `tensorflow index checkpoint file` (e.g. `./tf_model/model.ckpt.index`). In this case, ``from_tf`` should be set to True and a configuration object should be provided as ``config`` argument. This loading path is slower than converting the TensorFlow checkpoint in a PyTorch model using the provided conversion scripts and loading the PyTorch model afterwards.
                 - None if you are both providing the configuration and state dictionary (resp. with keyword arguments ``config`` and ``state_dict``)
diff --git a/transformers/tokenization_auto.py b/transformers/tokenization_auto.py
index f36a584521..1f0599ef7f 100644
--- a/transformers/tokenization_auto.py
+++ b/transformers/tokenization_auto.py
@@ -86,6 +86,7 @@ class AutoTokenizer(object):
             pretrained_model_name_or_path: either:
 
                 - a string with the `shortcut name` of a predefined tokenizer to load from cache or download, e.g.: ``bert-base-uncased``.
+                - a string with the `identifier name` of a predefined tokenizer that was user-uploaded to our S3, e.g.: ``dbmdz/bert-base-german-cased``.
                 - a path to a `directory` containing vocabulary files required by the tokenizer, for instance saved using the :func:`~transformers.PreTrainedTokenizer.save_pretrained` method, e.g.: ``./my_model_directory/``.
                 - (not applicable to all derived classes) a path or url to a single saved vocabulary file if and only if the tokenizer only requires a single vocabulary file (e.g. Bert, XLNet), e.g.: ``./my_model_directory/vocab.txt``.
 
@@ -108,8 +109,14 @@ class AutoTokenizer(object):
 
         Examples::
 
-            tokenizer = AutoTokenizer.from_pretrained('bert-base-uncased')    # Download vocabulary from S3 and cache.
-            tokenizer = AutoTokenizer.from_pretrained('./test/bert_saved_model/')  # E.g. tokenizer was saved using `save_pretrained('./test/saved_model/')`
+            # Download vocabulary from S3 and cache.
+            tokenizer = AutoTokenizer.from_pretrained('bert-base-uncased')
+
+            # Download vocabulary from S3 (user-uploaded) and cache.
+            tokenizer = AutoTokenizer.from_pretrained('dbmdz/bert-base-german-cased')
+
+            # If vocabulary files are in a directory (e.g. tokenizer was saved using `save_pretrained('./test/saved_model/')`)
+            tokenizer = AutoTokenizer.from_pretrained('./test/bert_saved_model/')
 
         """
         if 'distilbert' in pretrained_model_name_or_path:
diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index 7e86742286..317ecd167b 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -255,7 +255,7 @@ class PreTrainedTokenizer(object):
             pretrained_model_name_or_path: either:
 
                 - a string with the `shortcut name` of a predefined tokenizer to load from cache or download, e.g.: ``bert-base-uncased``.
-                - a string with the `identifier name` of a predefined tokenizer that was user-uploaded to our S3, e.g.: ``dbmz/bert-base-german-cased``.
+                - a string with the `identifier name` of a predefined tokenizer that was user-uploaded to our S3, e.g.: ``dbmdz/bert-base-german-cased``.
                 - a path to a `directory` containing vocabulary files required by the tokenizer, for instance saved using the :func:`~transformers.PreTrainedTokenizer.save_pretrained` method, e.g.: ``./my_model_directory/``.
                 - (not applicable to all derived classes) a path or url to a single saved vocabulary file if and only if the tokenizer only requires a single vocabulary file (e.g. Bert, XLNet), e.g.: ``./my_model_directory/vocab.txt``.
 
@@ -284,7 +284,7 @@ class PreTrainedTokenizer(object):
             tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
 
             # Download vocabulary from S3 (user-uploaded) and cache.
-            tokenizer = BertTokenizer.from_pretrained('dbmz/bert-base-german-cased')
+            tokenizer = BertTokenizer.from_pretrained('dbmdz/bert-base-german-cased')
 
             # If vocabulary files are in a directory (e.g. tokenizer was saved using `save_pretrained('./test/saved_model/')`)
             tokenizer = BertTokenizer.from_pretrained('./test/saved_model/')

From 7296f1010b6faaf3b1fb409bc5a9ebadcea51973 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Thu, 12 Dec 2019 13:01:04 -0500
Subject: [PATCH 343/505] Cleanup squad and add allow train_file and
 predict_file usage

---
 examples/run_squad.py                 | 22 ++++++++++++++--------
 transformers/data/processors/squad.py |  6 ++++++
 2 files changed, 20 insertions(+), 8 deletions(-)

diff --git a/examples/run_squad.py b/examples/run_squad.py
index 79c8537a4b..117b86e32c 100644
--- a/examples/run_squad.py
+++ b/examples/run_squad.py
@@ -337,7 +337,7 @@ def load_and_cache_examples(args, tokenizer, evaluate=False, output_examples=Fal
     else:
         logger.info("Creating features from dataset file at %s", input_dir)
 
-        if not args.data_dir:
+        if not args.data_dir and ((evaluate and not args.predict_file) or (not evaluate and not args.train_file)):
             try:
                 import tensorflow_datasets as tfds
             except ImportError:
@@ -350,7 +350,11 @@ def load_and_cache_examples(args, tokenizer, evaluate=False, output_examples=Fal
             examples = SquadV1Processor().get_examples_from_dataset(tfds_examples, evaluate=evaluate)
         else:
             processor = SquadV2Processor() if args.version_2_with_negative else SquadV1Processor()
-            examples = processor.get_dev_examples(args.data_dir) if evaluate else processor.get_train_examples(args.data_dir)
+
+            if evaluate:
+                examples = processor.get_dev_examples(args.data_dir, filename=args.predict_file)
+            else:
+                examples = processor.get_train_examples(args.data_dir, filename=args.train_file)
 
         features, dataset = squad_convert_examples_to_features( 
             examples=examples,
@@ -387,7 +391,14 @@ def main():
 
     ## Other parameters
     parser.add_argument("--data_dir", default=None, type=str,
-                        help="The input data dir. Should contain the .json files for the task. If not specified, will run with tensorflow_datasets.")
+                        help="The input data dir. Should contain the .json files for the task." +
+                             "If no data dir or train/predict files are specified, will run with tensorflow_datasets.")
+    parser.add_argument("--train_file", default=None, type=str,
+                        help="The input training file. If a data dir is specified, will look for the file there" +
+                             "If no data dir or train/predict files are specified, will run with tensorflow_datasets.")
+    parser.add_argument("--predict_file", default=None, type=str,
+                        help="The input evaluation file. If a data dir is specified, will look for the file there" +
+                             "If no data dir or train/predict files are specified, will run with tensorflow_datasets.")
     parser.add_argument("--config_name", default="", type=str,
                         help="Pretrained config name or path if not the same as model_name")
     parser.add_argument("--tokenizer_name", default="", type=str,
@@ -472,11 +483,6 @@ def main():
     parser.add_argument('--server_port', type=str, default='', help="Can be used for distant debugging.")
     args = parser.parse_args()
 
-    args.predict_file = os.path.join(args.output_dir, 'predictions_{}_{}.txt'.format(
-        list(filter(None, args.model_name_or_path.split('/'))).pop(),
-        str(args.max_seq_length))
-    )
-
     if os.path.exists(args.output_dir) and os.listdir(args.output_dir) and args.do_train and not args.overwrite_output_dir:
         raise ValueError("Output directory ({}) already exists and is not empty. Use --overwrite_output_dir to overcome.".format(args.output_dir))
 
diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index 3d7f832540..9bc4375684 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -373,6 +373,9 @@ class SquadProcessor(DataProcessor):
                 which is `train-v1.1.json` and `train-v2.0.json` for squad versions 1.1 and 2.0 respectively.
 
         """
+        if data_dir is None:
+            data_dir = ""
+
         if self.train_file is None:
             raise ValueError("SquadProcessor should be instantiated via SquadV1Processor or SquadV2Processor")
 
@@ -389,6 +392,9 @@ class SquadProcessor(DataProcessor):
             filename: None by default, specify this if the evaluation file has a different name than the original one
                 which is `train-v1.1.json` and `train-v2.0.json` for squad versions 1.1 and 2.0 respectively.
         """
+        if data_dir is None:
+            data_dir = ""
+
         if self.dev_file is None:
             raise ValueError("SquadProcessor should be instantiated via SquadV1Processor or SquadV2Processor")
         

From 33e72b08d54bf5edd192492af7549b581563ecc2 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 13 Dec 2019 11:33:05 +0100
Subject: [PATCH 344/505] fix inner dimensions for 3B/11B models

---
 transformers/modeling_t5.py    | 27 +++++++++++----------------
 transformers/modeling_tf_t5.py | 20 ++++++++------------
 2 files changed, 19 insertions(+), 28 deletions(-)

diff --git a/transformers/modeling_t5.py b/transformers/modeling_t5.py
index 149b977abc..c9310179a3 100644
--- a/transformers/modeling_t5.py
+++ b/transformers/modeling_t5.py
@@ -30,7 +30,7 @@ from torch import nn
 import torch.nn.functional as F
 from torch.nn import CrossEntropyLoss, MSELoss
 
-from .modeling_utils import PreTrainedModel
+from .modeling_utils import PreTrainedModel, prune_linear_layer
 from .configuration_t5 import T5Config
 from .file_utils import add_start_docstrings, DUMMY_INPUTS, DUMMY_MASK
 
@@ -191,28 +191,26 @@ class T5Attention(nn.Module):
 
         self.output_attentions = config.output_attentions
         self.relative_attention_num_buckets = config.relative_attention_num_buckets
-        self.dim = config.d_model
+        self.d_model = config.d_model
         self.d_kv = config.d_kv
         self.n_heads = config.num_heads
         self.dropout = config.dropout_rate
-        assert self.dim % self.n_heads == 0
-        assert self.dim // self.n_heads == self.d_kv
+        self.inner_dim = self.n_heads * self.d_kv
 
         # Mesh TensorFlow initialization to avoid scaling before softmax
-        self.q = nn.Linear(self.dim, self.dim, bias=False)
-        self.k = nn.Linear(self.dim, self.dim, bias=False)
-        self.v = nn.Linear(self.dim, self.dim, bias=False)
-        self.o = nn.Linear(self.dim, self.dim, bias=False)
+        self.q = nn.Linear(self.d_model, self.inner_dim, bias=False)
+        self.k = nn.Linear(self.d_model, self.inner_dim, bias=False)
+        self.v = nn.Linear(self.d_model, self.inner_dim, bias=False)
+        self.o = nn.Linear(self.inner_dim, self.d_model, bias=False)
 
         if self.has_relative_attention_bias:
             self.relative_attention_bias = nn.Embedding(self.relative_attention_num_buckets, self.n_heads)
         self.pruned_heads = set()
 
     def prune_heads(self, heads):
-        attention_head_size = self.dim // self.n_heads
         if len(heads) == 0:
             return
-        mask = torch.ones(self.n_heads, attention_head_size)
+        mask = torch.ones(self.n_heads, self.d_kv)
         heads = set(heads) - self.pruned_heads
         for head in heads:
             head -= sum(1 if h < head else 0 for h in self.pruned_heads)
@@ -226,7 +224,7 @@ class T5Attention(nn.Module):
         self.o = prune_linear_layer(self.o, index, dim=1)
         # Update hyper params
         self.n_heads = self.n_heads - len(heads)
-        self.dim = attention_head_size * self.n_heads
+        self.inner_dim = self.d_kv * self.n_heads
         self.pruned_heads = self.pruned_heads.union(heads)
 
     @staticmethod
@@ -303,17 +301,14 @@ class T5Attention(nn.Module):
             klen = qlen if cache is None else cache['slen'] + qlen
         else:
             klen = kv.size(1)
-        # assert dim == self.dim, 'Dimensions do not match: %s input vs %s configured' % (dim, self.dim)
-        n_heads = self.n_heads
-        dim_per_head = self.dim // n_heads
 
         def shape(x):
             """  projection """
-            return x.view(bs, -1, self.n_heads, dim_per_head).transpose(1, 2)
+            return x.view(bs, -1, self.n_heads, self.d_kv).transpose(1, 2)
 
         def unshape(x):
             """  compute context """
-            return x.transpose(1, 2).contiguous().view(bs, -1, self.n_heads * dim_per_head)
+            return x.transpose(1, 2).contiguous().view(bs, -1, self.inner_dim)
 
         q = shape(self.q(input))                                          # (bs, n_heads, qlen, dim_per_head)
         if kv is None:
diff --git a/transformers/modeling_tf_t5.py b/transformers/modeling_tf_t5.py
index fd25328ac6..0ae7fff412 100644
--- a/transformers/modeling_tf_t5.py
+++ b/transformers/modeling_tf_t5.py
@@ -108,17 +108,16 @@ class TFT5Attention(tf.keras.layers.Layer):
 
         self.output_attentions = config.output_attentions
         self.relative_attention_num_buckets = config.relative_attention_num_buckets
-        self.dim = config.d_model
+        self.d_model = config.d_model
         self.d_kv = config.d_kv
         self.n_heads = config.num_heads
-        assert self.dim % self.n_heads == 0
-        assert self.dim // self.n_heads == self.d_kv
+        self.inner_dim = self.n_heads * self.d_kv
 
         # Mesh TensorFlow initialization to avoid scaling before softmax
-        self.q = tf.keras.layers.Dense(self.dim, use_bias=False, name='q')
-        self.k = tf.keras.layers.Dense(self.dim, use_bias=False, name='k')
-        self.v = tf.keras.layers.Dense(self.dim, use_bias=False, name='v')
-        self.o = tf.keras.layers.Dense(self.dim, use_bias=False, name='o')
+        self.q = tf.keras.layers.Dense(self.inner_dim, use_bias=False, name='q')
+        self.k = tf.keras.layers.Dense(self.inner_dim, use_bias=False, name='k')
+        self.v = tf.keras.layers.Dense(self.inner_dim, use_bias=False, name='v')
+        self.o = tf.keras.layers.Dense(self.d_model, use_bias=False, name='o')
         self.dropout = tf.keras.layers.Dropout(config.dropout_rate)
 
         if self.has_relative_attention_bias:
@@ -199,17 +198,14 @@ class TFT5Attention(tf.keras.layers.Layer):
             klen = qlen if cache is None else cache['slen'] + qlen
         else:
             klen = shape_list(kv)[1]
-        # assert dim == self.dim, 'Dimensions do not match: %s input vs %s configured' % (dim, self.dim)
-        n_heads = self.n_heads
-        dim_per_head = self.dim // n_heads
 
         def shape(x):
             """  projection """
-            return tf.transpose(tf.reshape(x, (bs, -1, self.n_heads, dim_per_head)), perm=(0, 2, 1, 3))
+            return tf.transpose(tf.reshape(x, (bs, -1, self.n_heads, self.d_kv)), perm=(0, 2, 1, 3))
 
         def unshape(x):
             """  compute context """
-            return tf.reshape(tf.transpose(x, perm=(0, 2, 1, 3)), (bs, -1, self.n_heads * dim_per_head))
+            return tf.reshape(tf.transpose(x, perm=(0, 2, 1, 3)), (bs, -1, self.inner_dim))
 
         q = shape(self.q(input))                                          # (bs, n_heads, qlen, dim_per_head)
         if kv is None:

From 80eacb8f16208bcc7ffd8ed5b5750d6fc6854a24 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Fri, 13 Dec 2019 14:10:22 +0100
Subject: [PATCH 345/505] Adding labels mapping for classification models in
 their respective config.

---
 transformers/configuration_utils.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/transformers/configuration_utils.py b/transformers/configuration_utils.py
index 97b9fa8f80..b7ddcf0912 100644
--- a/transformers/configuration_utils.py
+++ b/transformers/configuration_utils.py
@@ -58,8 +58,8 @@ class PretrainedConfig(object):
         self.use_bfloat16 = kwargs.pop('use_bfloat16', False)
         self.pruned_heads = kwargs.pop('pruned_heads', {})
         self.is_decoder = kwargs.pop('is_decoder', False)
-        self.idx2label = kwargs.pop('idx2label', {i: 'LABEL_{}'.format(i) for i in range(self.num_labels)})
-        self.label2idx = kwargs.pop('label2idx', dict(zip(self.idx2label.values(), self.idx2label.keys())))
+        self.id2label = kwargs.pop('id2label', {i: 'LABEL_{}'.format(i) for i in range(self.num_labels)})
+        self.label2id = kwargs.pop('label2id', dict(zip(self.id2label.values(), self.id2label.keys())))
 
     def save_pretrained(self, save_directory):
         """ Save a configuration object to the directory `save_directory`, so that it

From be5bf7b81bd4171169e23091beda85ffd97f950f Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Fri, 13 Dec 2019 14:12:17 +0100
Subject: [PATCH 346/505] Added NER pipeline.

---
 transformers/pipelines.py | 720 ++++++++++++++++++++------------------
 1 file changed, 388 insertions(+), 332 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index da8b0b65a7..b0b5848c01 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -1,332 +1,388 @@
-# coding=utf-8
-# Copyright 2018 The HuggingFace Inc. team.
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-from __future__ import absolute_import, division, print_function, unicode_literals
-
-import os
-from abc import ABC, abstractmethod
-from typing import Union, Optional, Tuple, List, Dict
-
-import numpy as np
-
-from transformers import is_tf_available, is_torch_available, logger, AutoTokenizer, PreTrainedTokenizer, \
-    SquadExample, squad_convert_examples_to_features
-
-if is_tf_available():
-    from transformers import TFAutoModelForSequenceClassification, TFAutoModelForQuestionAnswering
-
-if is_torch_available():
-    from transformers import AutoModelForSequenceClassification, AutoModelForQuestionAnswering
-
-
-class Pipeline(ABC):
-    def __init__(self, model, tokenizer: PreTrainedTokenizer = None, **kwargs):
-        self.model = model
-        self.tokenizer = tokenizer
-
-    @classmethod
-    @abstractmethod
-    def from_config(cls, model, tokenizer: PreTrainedTokenizer, **kwargs):
-        raise NotImplementedError()
-
-    def save_pretrained(self, save_directory):
-        if not os.path.isdir(save_directory):
-            logger.error("Provided path ({}) should be a directory".format(save_directory))
-            return
-
-        self.model.save_pretrained(save_directory)
-        self.tokenizer.save_pretrained(save_directory)
-
-    def transform(self, *texts, **kwargs):
-        # Generic compatibility with sklearn and Keras
-        return self(*texts, **kwargs)
-
-    def predict(self, *texts, **kwargs):
-        # Generic compatibility with sklearn and Keras
-        return self(*texts, **kwargs)
-
-    @abstractmethod
-    def __call__(self, *texts, **kwargs):
-        raise NotImplementedError()
-
-
-class TextClassificationPipeline(Pipeline):
-    def __init__(self, model, tokenizer: PreTrainedTokenizer, nb_classes: int = 2):
-        super().__init__(model, tokenizer)
-
-        if nb_classes < 2:
-            raise Exception('Invalid parameter nb_classes. int >= 2 is required (got: {})'.format(nb_classes))
-        self._nb_classes = nb_classes
-
-    @classmethod
-    def from_config(cls, model, tokenizer: PreTrainedTokenizer, **kwargs):
-        return cls(model, tokenizer, **kwargs)
-
-    def __call__(self, *texts, **kwargs):
-        # Generic compatibility with sklearn and Keras
-        if 'X' in kwargs and not texts:
-            texts = kwargs.pop('X')
-
-        inputs = self.tokenizer.batch_encode_plus(
-            texts, add_special_tokens=True, return_tensors='tf' if is_tf_available() else 'pt'
-        )
-
-        special_tokens_mask = inputs.pop('special_tokens_mask')
-
-        if is_tf_available():
-            # TODO trace model
-            predictions = self.model(**inputs)[0]
-        else:
-            import torch
-            with torch.no_grad():
-                predictions = self.model(**inputs)[0]
-
-        return predictions.numpy().tolist()
-
-
-class QuestionAnsweringPipeline(Pipeline):
-    """
-    Question Answering pipeling involving Tokenization and Inference.
-    """
-
-    @classmethod
-    def from_config(cls, model, tokenizer: PreTrainedTokenizer, **kwargs):
-        pass
-
-    @staticmethod
-    def create_sample(question: Union[str, List[str]], context: Union[str, List[str]]) -> Union[SquadExample, List[SquadExample]]:
-        is_list = isinstance(question, list)
-
-        if is_list:
-            return [SquadExample(None, q, c, None, None, None) for q, c in zip(question, context)]
-        else:
-            return SquadExample(None, question, context, None, None, None)
-
-    @staticmethod
-    def handle_args(*inputs, **kwargs) -> List[SquadExample]:
-        # Position args, handling is sensibly the same as X and data, so forwarding to avoid duplicating
-        if inputs is not None and len(inputs) > 1:
-            kwargs['X'] = inputs
-
-        # Generic compatibility with sklearn and Keras
-        # Batched data
-        if 'X' in kwargs or 'data' in kwargs:
-            data = kwargs['X'] if 'X' in kwargs else kwargs['data']
-
-            if not isinstance(data, list):
-                data = [data]
-
-            for i, item in enumerate(data):
-                if isinstance(item, dict):
-                    if any(k not in item for k in ['question', 'context']):
-                        raise KeyError('You need to provide a dictionary with keys {question:..., context:...}')
-                    data[i] = QuestionAnsweringPipeline.create_sample(**item)
-
-                elif isinstance(item, SquadExample):
-                    continue
-                else:
-                    raise ValueError(
-                        '{} argument needs to be of type (list[SquadExample | dict], SquadExample, dict)'
-                        .format('X' if 'X' in kwargs else 'data')
-                    )
-            inputs = data
-
-        # Tabular input
-        elif 'question' in kwargs and 'context' in kwargs:
-            if isinstance(kwargs['question'], str):
-                kwargs['question'] = [kwargs['question']]
-
-            if isinstance(kwargs['context'], str):
-                kwargs['context'] = [kwargs['context']]
-
-            inputs = [QuestionAnsweringPipeline.create_sample(q, c) for q, c in zip(kwargs['question'], kwargs['context'])]
-        else:
-            raise ValueError('Unknown arguments {}'.format(kwargs))
-
-        if not isinstance(inputs, list):
-            inputs = [inputs]
-
-        return inputs
-
-    def __init__(self, model, tokenizer: Optional[PreTrainedTokenizer]):
-        super().__init__(model, tokenizer)
-
-    def inputs_for_model(self, features: Union[SquadExample, List[SquadExample]]) -> Dict:
-        args = ['input_ids', 'attention_mask']
-        model_type = type(self.model).__name__.lower()
-
-        if 'distilbert' not in model_type and 'xlm' not in model_type:
-            args += ['token_type_ids']
-
-        if 'xlnet' in model_type or 'xlm' in model_type:
-            args += ['cls_index', 'p_mask']
-
-        if isinstance(features, SquadExample):
-            return {k: features.__dict__[k] for k in args}
-        else:
-            return {k: [feature.__dict__[k] for feature in features] for k in args}
-
-    def __call__(self, *texts, **kwargs):
-        # Set defaults values
-        kwargs.setdefault('topk', 1)
-        kwargs.setdefault('doc_stride', 128)
-        kwargs.setdefault('max_answer_len', 15)
-        kwargs.setdefault('max_seq_len', 384)
-        kwargs.setdefault('max_question_len', 64)
-
-        if kwargs['topk'] < 1:
-            raise ValueError('topk parameter should be >= 1 (got {})'.format(kwargs['topk']))
-
-        if kwargs['max_answer_len'] < 1:
-            raise ValueError('max_answer_len parameter should be >= 1 (got {})'.format(kwargs['max_answer_len']))
-
-        examples = QuestionAnsweringPipeline.handle_args(texts, **kwargs)
-
-        # Convert inputs to features
-        features = squad_convert_examples_to_features(examples, self.tokenizer, kwargs['max_seq_len'], kwargs['doc_stride'], kwargs['max_question_len'], False)
-        fw_args = self.inputs_for_model(features)
-
-        if is_tf_available():
-            import tensorflow as tf
-            fw_args = {k: tf.constant(v) for (k, v) in fw_args.items()}
-            start, end = self.model(fw_args)
-            start, end = start.numpy(), end.numpy()
-        else:
-            import torch
-            with torch.no_grad():
-                # Retrieve the score for the context tokens only (removing question tokens)
-                fw_args = {k: torch.tensor(v) for (k, v) in fw_args.items()}
-                start, end = self.model(**fw_args)
-                start, end = start.cpu().numpy(), end.cpu().numpy()
-
-        answers = []
-        for (example, feature, start_, end_) in zip(examples, features, start, end):
-            # Normalize logits and spans to retrieve the answer
-            start_ = np.exp(start_) / np.sum(np.exp(start_))
-            end_ = np.exp(end_) / np.sum(np.exp(end_))
-
-            # Mask padding and question
-            start_, end_ = start_ * np.abs(np.array(feature.p_mask) - 1), end_ * np.abs(np.array(feature.p_mask) - 1)
-
-            # Mask CLS
-            start_[0] = end_[0] = 0
-
-            starts, ends, scores = self.decode(start_, end_, kwargs['topk'], kwargs['max_answer_len'])
-            char_to_word = np.array(example.char_to_word_offset)
-
-            # Convert the answer (tokens) back to the original text
-            answers += [[
-                {
-                    'score': score,
-                    'start': np.where(char_to_word == feature.token_to_orig_map[s])[0][0],
-                    'end': np.where(char_to_word == feature.token_to_orig_map[e])[0][-1],
-                    'answer': ' '.join(example.doc_tokens[feature.token_to_orig_map[s]: feature.token_to_orig_map[e] + 1])
-                }
-                for s, e, score in zip(starts, ends, scores)
-            ]]
-
-        return answers
-
-    def decode(self, start: np.ndarray, end: np.ndarray, topk: int, max_answer_len: int) -> Tuple:
-        # Ensure we have batch axis
-        if start.ndim == 1:
-            start = start[None]
-
-        if end.ndim == 1:
-            end = end[None]
-
-        # Compute the score of each tuple(start, end) to be the real answer
-        outer = np.matmul(np.expand_dims(start, -1), np.expand_dims(end, 1))
-
-        # Remove candidate with end < start and end - start > max_answer_len
-        candidates = np.tril(np.triu(outer), max_answer_len - 1)
-
-        #  Inspired by Chen & al. (https://github.com/facebookresearch/DrQA)
-        scores_flat = candidates.flatten()
-        if topk == 1:
-            idx_sort = [np.argmax(scores_flat)]
-        elif len(scores_flat) < topk:
-            idx_sort = np.argsort(-scores_flat)
-        else:
-            idx = np.argpartition(-scores_flat, topk)[0:topk]
-            idx_sort = idx[np.argsort(-scores_flat[idx])]
-
-        start, end = np.unravel_index(idx_sort, candidates.shape)[1:]
-        return start, end, candidates[0, start, end]
-
-    def span_to_answer(self, text: str, start: int, end: int):
-        words = []
-        token_idx = char_start_idx = char_end_idx = chars_idx = 0
-
-        for i, word in enumerate(text.split(" ")):
-            token = self.tokenizer.tokenize(word)
-
-            # Append words if they are in the span
-            if start <= token_idx <= end:
-                if token_idx == start:
-                    char_start_idx = chars_idx
-
-                if token_idx == end:
-                    char_end_idx = chars_idx + len(word)
-
-                words += [word]
-
-            # Stop if we went over the end of the answer
-            if token_idx > end:
-                break
-
-            # Append the subtokenization length to the running index
-            token_idx += len(token)
-            chars_idx += len(word) + 1
-
-        # Join text with spaces
-        return {'answer': ' '.join(words), 'start': max(0, char_start_idx), 'end': min(len(text), char_end_idx)}
-
-
-# Register all the supported task here
-SUPPORTED_TASKS = {
-    'text-classification': {
-        'impl': TextClassificationPipeline,
-        'tf': TFAutoModelForSequenceClassification if is_tf_available() else None,
-        'pt': AutoModelForSequenceClassification if is_torch_available() else None
-    },
-    'question-answering': {
-        'impl': QuestionAnsweringPipeline,
-        'tf': TFAutoModelForQuestionAnswering if is_tf_available() else None,
-        'pt': AutoModelForQuestionAnswering if is_torch_available() else None
-    }
-}
-
-
-def pipeline(task: str, model, tokenizer: Optional[Union[str, PreTrainedTokenizer]] = None, **kwargs) -> Pipeline:
-    """
-    Utility factory method to build pipeline.
-    """
-    # Try to infer tokenizer from model name (if provided as str)
-    if tokenizer is None and isinstance(model, str):
-        tokenizer = model
-    else:
-        # Impossible to guest what is the right tokenizer here
-        raise Exception('Tokenizer cannot be None if provided model is a PreTrainedModel instance')
-
-    tokenizer = tokenizer if isinstance(tokenizer, PreTrainedTokenizer) else AutoTokenizer.from_pretrained(tokenizer)
-
-    if task not in SUPPORTED_TASKS:
-        raise KeyError("Unknown task {}, available tasks are {}".format(task, list(SUPPORTED_TASKS.keys())))
-
-    targeted_task = SUPPORTED_TASKS[task]
-    task, allocator = targeted_task['impl'], targeted_task['tf'] if is_tf_available() else targeted_task['pt']
-
-    model = allocator.from_pretrained(model)
-    return task(model, tokenizer, **kwargs)
+# coding=utf-8
+# Copyright 2018 The HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from __future__ import absolute_import, division, print_function, unicode_literals
+
+import os
+from abc import ABC, abstractmethod
+from itertools import groupby
+from typing import Union, Optional, Tuple, List, Dict
+
+import numpy as np
+
+from transformers import AutoTokenizer, PreTrainedTokenizer, PretrainedConfig, \
+    SquadExample, squad_convert_examples_to_features, is_tf_available, is_torch_available, logger
+
+if is_tf_available():
+    from transformers import TFAutoModelForSequenceClassification, TFAutoModelForQuestionAnswering, TFAutoModelForTokenClassification
+
+if is_torch_available():
+    import torch
+    from transformers import AutoModelForSequenceClassification, AutoModelForQuestionAnswering, AutoModelForTokenClassification
+
+
+class Pipeline(ABC):
+    def __init__(self, model, tokenizer: PreTrainedTokenizer = None, **kwargs):
+        self.model = model
+        self.tokenizer = tokenizer
+
+    @classmethod
+    @abstractmethod
+    def from_config(cls, model, tokenizer: PreTrainedTokenizer, **kwargs):
+        raise NotImplementedError()
+
+    def save_pretrained(self, save_directory):
+        if not os.path.isdir(save_directory):
+            logger.error("Provided path ({}) should be a directory".format(save_directory))
+            return
+
+        self.model.save_pretrained(save_directory)
+        self.tokenizer.save_pretrained(save_directory)
+
+    def transform(self, *texts, **kwargs):
+        # Generic compatibility with sklearn and Keras
+        return self(*texts, **kwargs)
+
+    def predict(self, *texts, **kwargs):
+        # Generic compatibility with sklearn and Keras
+        return self(*texts, **kwargs)
+
+    @abstractmethod
+    def __call__(self, *texts, **kwargs):
+        raise NotImplementedError()
+
+
+class TextClassificationPipeline(Pipeline):
+    def __init__(self, model, tokenizer: PreTrainedTokenizer, nb_classes: int = 2):
+        super().__init__(model, tokenizer)
+
+        if nb_classes < 2:
+            raise Exception('Invalid parameter nb_classes. int >= 2 is required (got: {})'.format(nb_classes))
+        self._nb_classes = nb_classes
+
+    @classmethod
+    def from_config(cls, model, tokenizer: PreTrainedTokenizer, **kwargs):
+        return cls(model, tokenizer, **kwargs)
+
+    def __call__(self, *texts, **kwargs):
+        # Generic compatibility with sklearn and Keras
+        if 'X' in kwargs and not texts:
+            texts = kwargs.pop('X')
+
+        inputs = self.tokenizer.batch_encode_plus(
+            texts, add_special_tokens=True, return_tensors='tf' if is_tf_available() else 'pt'
+        )
+
+        special_tokens_mask = inputs.pop('special_tokens_mask')
+
+        if is_tf_available():
+            # TODO trace model
+            predictions = self.model(**inputs)[0]
+        else:
+            import torch
+            with torch.no_grad():
+                predictions = self.model(**inputs)[0]
+
+        return predictions.numpy().tolist()
+
+
+class NerPipeline(Pipeline):
+
+    def __init__(self, model, tokenizer: PreTrainedTokenizer):
+        super().__init__(model, tokenizer)
+
+    @classmethod
+    def from_config(cls, model, tokenizer: PreTrainedTokenizer, **kwargs):
+        pass
+
+    def __call__(self, *texts, **kwargs):
+        (texts, ), answers = texts, []
+
+        for sentence in texts:
+
+            # Ugly token to word idx mapping (for now)
+            token_to_word, words = [], sentence.split(' ')
+            for i, w in enumerate(words):
+                tokens = self.tokenizer.tokenize(w)
+                token_to_word += [i] * len(tokens)
+            tokens = self.tokenizer.encode_plus(sentence, return_attention_mask=False, return_tensors='tf' if is_tf_available() else 'pt')
+
+            # Forward
+            if is_torch_available():
+                with torch.no_grad():
+                    entities = self.model(**tokens)[0][0].cpu().numpy()
+            else:
+                entities = self.model(tokens)[0][0].numpy()
+
+            # Normalize scores
+            answer, token_start = [], 1
+            for idx, word in groupby(token_to_word[1:-1]):
+
+                # Sum log prob over token, then normalize across labels
+                score = np.exp(entities[token_start]) / np.exp(entities[token_start]).sum(-1, keepdims=True)
+                label_idx = score.argmax()
+
+                answer += [{
+                    'word': words[idx - 1], 'score': score[label_idx], 'entity': self.model.config.id2label[label_idx]
+                }]
+
+                # Update token start
+                token_start += len(list(word))
+
+            # Append
+            answers += [answer]
+        return answers
+
+
+class QuestionAnsweringPipeline(Pipeline):
+    """
+    Question Answering pipeline involving Tokenization and Inference.
+    """
+
+    @classmethod
+    def from_config(cls, model, tokenizer: PreTrainedTokenizer, **kwargs):
+        pass
+
+    @staticmethod
+    def create_sample(question: Union[str, List[str]], context: Union[str, List[str]]) -> Union[SquadExample, List[SquadExample]]:
+        is_list = isinstance(question, list)
+
+        if is_list:
+            return [SquadExample(None, q, c, None, None, None) for q, c in zip(question, context)]
+        else:
+            return SquadExample(None, question, context, None, None, None)
+
+    @staticmethod
+    def handle_args(*inputs, **kwargs) -> List[SquadExample]:
+        # Position args, handling is sensibly the same as X and data, so forwarding to avoid duplicating
+        if inputs is not None and len(inputs) > 1:
+            kwargs['X'] = inputs
+
+        # Generic compatibility with sklearn and Keras
+        # Batched data
+        if 'X' in kwargs or 'data' in kwargs:
+            data = kwargs['X'] if 'X' in kwargs else kwargs['data']
+
+            if not isinstance(data, list):
+                data = [data]
+
+            for i, item in enumerate(data):
+                if isinstance(item, dict):
+                    if any(k not in item for k in ['question', 'context']):
+                        raise KeyError('You need to provide a dictionary with keys {question:..., context:...}')
+                    data[i] = QuestionAnsweringPipeline.create_sample(**item)
+
+                elif isinstance(item, SquadExample):
+                    continue
+                else:
+                    raise ValueError(
+                        '{} argument needs to be of type (list[SquadExample | dict], SquadExample, dict)'
+                        .format('X' if 'X' in kwargs else 'data')
+                    )
+            inputs = data
+
+        # Tabular input
+        elif 'question' in kwargs and 'context' in kwargs:
+            if isinstance(kwargs['question'], str):
+                kwargs['question'] = [kwargs['question']]
+
+            if isinstance(kwargs['context'], str):
+                kwargs['context'] = [kwargs['context']]
+
+            inputs = [QuestionAnsweringPipeline.create_sample(q, c) for q, c in zip(kwargs['question'], kwargs['context'])]
+        else:
+            raise ValueError('Unknown arguments {}'.format(kwargs))
+
+        if not isinstance(inputs, list):
+            inputs = [inputs]
+
+        return inputs
+
+    def __init__(self, model, tokenizer: Optional[PreTrainedTokenizer]):
+        super().__init__(model, tokenizer)
+
+    def inputs_for_model(self, features: Union[SquadExample, List[SquadExample]]) -> Dict:
+        args = ['input_ids', 'attention_mask']
+        model_type = type(self.model).__name__.lower()
+
+        if 'distilbert' not in model_type and 'xlm' not in model_type:
+            args += ['token_type_ids']
+
+        if 'xlnet' in model_type or 'xlm' in model_type:
+            args += ['cls_index', 'p_mask']
+
+        if isinstance(features, SquadExample):
+            return {k: features.__dict__[k] for k in args}
+        else:
+            return {k: [feature.__dict__[k] for feature in features] for k in args}
+
+    def __call__(self, *texts, **kwargs):
+        # Set defaults values
+        kwargs.setdefault('topk', 1)
+        kwargs.setdefault('doc_stride', 128)
+        kwargs.setdefault('max_answer_len', 15)
+        kwargs.setdefault('max_seq_len', 384)
+        kwargs.setdefault('max_question_len', 64)
+
+        if kwargs['topk'] < 1:
+            raise ValueError('topk parameter should be >= 1 (got {})'.format(kwargs['topk']))
+
+        if kwargs['max_answer_len'] < 1:
+            raise ValueError('max_answer_len parameter should be >= 1 (got {})'.format(kwargs['max_answer_len']))
+
+        examples = QuestionAnsweringPipeline.handle_args(texts, **kwargs)
+
+        # Convert inputs to features
+        features = squad_convert_examples_to_features(examples, self.tokenizer, kwargs['max_seq_len'], kwargs['doc_stride'], kwargs['max_question_len'], False)
+        fw_args = self.inputs_for_model(features)
+
+        if is_tf_available():
+            import tensorflow as tf
+            fw_args = {k: tf.constant(v) for (k, v) in fw_args.items()}
+            start, end = self.model(fw_args)
+            start, end = start.numpy(), end.numpy()
+        else:
+            import torch
+            with torch.no_grad():
+                # Retrieve the score for the context tokens only (removing question tokens)
+                fw_args = {k: torch.tensor(v) for (k, v) in fw_args.items()}
+                start, end = self.model(**fw_args)
+                start, end = start.cpu().numpy(), end.cpu().numpy()
+
+        answers = []
+        for (example, feature, start_, end_) in zip(examples, features, start, end):
+            # Normalize logits and spans to retrieve the answer
+            start_ = np.exp(start_) / np.sum(np.exp(start_))
+            end_ = np.exp(end_) / np.sum(np.exp(end_))
+
+            # Mask padding and question
+            start_, end_ = start_ * np.abs(np.array(feature.p_mask) - 1), end_ * np.abs(np.array(feature.p_mask) - 1)
+
+            # TODO : What happend if not possible
+            # Mask CLS
+            start_[0] = end_[0] = 0
+
+            starts, ends, scores = self.decode(start_, end_, kwargs['topk'], kwargs['max_answer_len'])
+            char_to_word = np.array(example.char_to_word_offset)
+
+            # Convert the answer (tokens) back to the original text
+            answers += [[
+                {
+                    'score': score,
+                    'start': np.where(char_to_word == feature.token_to_orig_map[s])[0][0],
+                    'end': np.where(char_to_word == feature.token_to_orig_map[e])[0][-1],
+                    'answer': ' '.join(example.doc_tokens[feature.token_to_orig_map[s]: feature.token_to_orig_map[e] + 1])
+                }
+                for s, e, score in zip(starts, ends, scores)
+            ]]
+
+        return answers
+
+    def decode(self, start: np.ndarray, end: np.ndarray, topk: int, max_answer_len: int) -> Tuple:
+        # Ensure we have batch axis
+        if start.ndim == 1:
+            start = start[None]
+
+        if end.ndim == 1:
+            end = end[None]
+
+        # Compute the score of each tuple(start, end) to be the real answer
+        outer = np.matmul(np.expand_dims(start, -1), np.expand_dims(end, 1))
+
+        # Remove candidate with end < start and end - start > max_answer_len
+        candidates = np.tril(np.triu(outer), max_answer_len - 1)
+
+        #  Inspired by Chen & al. (https://github.com/facebookresearch/DrQA)
+        scores_flat = candidates.flatten()
+        if topk == 1:
+            idx_sort = [np.argmax(scores_flat)]
+        elif len(scores_flat) < topk:
+            idx_sort = np.argsort(-scores_flat)
+        else:
+            idx = np.argpartition(-scores_flat, topk)[0:topk]
+            idx_sort = idx[np.argsort(-scores_flat[idx])]
+
+        start, end = np.unravel_index(idx_sort, candidates.shape)[1:]
+        return start, end, candidates[0, start, end]
+
+    def span_to_answer(self, text: str, start: int, end: int):
+        words = []
+        token_idx = char_start_idx = char_end_idx = chars_idx = 0
+
+        for i, word in enumerate(text.split(" ")):
+            token = self.tokenizer.tokenize(word)
+
+            # Append words if they are in the span
+            if start <= token_idx <= end:
+                if token_idx == start:
+                    char_start_idx = chars_idx
+
+                if token_idx == end:
+                    char_end_idx = chars_idx + len(word)
+
+                words += [word]
+
+            # Stop if we went over the end of the answer
+            if token_idx > end:
+                break
+
+            # Append the subtokenization length to the running index
+            token_idx += len(token)
+            chars_idx += len(word) + 1
+
+        # Join text with spaces
+        return {'answer': ' '.join(words), 'start': max(0, char_start_idx), 'end': min(len(text), char_end_idx)}
+
+
+# Register all the supported task here
+SUPPORTED_TASKS = {
+    'text-classification': {
+        'impl': TextClassificationPipeline,
+        'tf': TFAutoModelForSequenceClassification if is_tf_available() else None,
+        'pt': AutoModelForSequenceClassification if is_torch_available() else None
+    },
+    'ner': {
+      'impl': NerPipeline,
+      'tf': TFAutoModelForTokenClassification if is_tf_available() else None,
+      'pt': AutoModelForTokenClassification if is_torch_available() else None,
+    },
+    'question-answering': {
+        'impl': QuestionAnsweringPipeline,
+        'tf': TFAutoModelForQuestionAnswering if is_tf_available() else None,
+        'pt': AutoModelForQuestionAnswering if is_torch_available() else None
+    }
+}
+
+
+def pipeline(task: str, model, config: Optional[PretrainedConfig] = None, tokenizer: Optional[Union[str, PreTrainedTokenizer]] = None, **kwargs) -> Pipeline:
+    """
+    Utility factory method to build pipeline.
+    """
+    # Try to infer tokenizer from model name (if provided as str)
+    if tokenizer is None and isinstance(model, str):
+        tokenizer = model
+    else:
+        # Impossible to guest what is the right tokenizer here
+        raise Exception('Tokenizer cannot be None if provided model is a PreTrainedModel instance')
+
+    tokenizer = tokenizer if isinstance(tokenizer, PreTrainedTokenizer) else AutoTokenizer.from_pretrained(tokenizer)
+
+    if task not in SUPPORTED_TASKS:
+        raise KeyError("Unknown task {}, available tasks are {}".format(task, list(SUPPORTED_TASKS.keys())))
+
+    targeted_task = SUPPORTED_TASKS[task]
+    task, allocator = targeted_task['impl'], targeted_task['tf'] if is_tf_available() else targeted_task['pt']
+
+    model = allocator.from_pretrained(model)
+    return task(model, tokenizer, **kwargs)

From 28e64ad5a4b01a1b7de092694e3a321edf7021bd Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Fri, 13 Dec 2019 14:12:54 +0100
Subject: [PATCH 347/505] Raise an exception if the pipeline allocator can't
 determine the tokenizer from the model.

---
 transformers/pipelines.py | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index b0b5848c01..853735a256 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -370,11 +370,12 @@ def pipeline(task: str, model, config: Optional[PretrainedConfig] = None, tokeni
     Utility factory method to build pipeline.
     """
     # Try to infer tokenizer from model name (if provided as str)
-    if tokenizer is None and isinstance(model, str):
-        tokenizer = model
-    else:
-        # Impossible to guest what is the right tokenizer here
-        raise Exception('Tokenizer cannot be None if provided model is a PreTrainedModel instance')
+    if not isinstance(tokenizer, PreTrainedTokenizer):
+        if not isinstance(model, str):
+            # Impossible to guest what is the right tokenizer here
+            raise Exception('Tokenizer cannot be None if provided model is a PreTrainedModel instance')
+        else:
+            tokenizer = model
 
     tokenizer = tokenizer if isinstance(tokenizer, PreTrainedTokenizer) else AutoTokenizer.from_pretrained(tokenizer)
 

From 1ca52567a4059e7ee1707de6a855bb5e7fb3fac3 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Fri, 13 Dec 2019 14:13:14 +0100
Subject: [PATCH 348/505] Allow model conversion in the pipeline allocator.

---
 transformers/pipelines.py | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 853735a256..9acd9bc566 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -385,5 +385,17 @@ def pipeline(task: str, model, config: Optional[PretrainedConfig] = None, tokeni
     targeted_task = SUPPORTED_TASKS[task]
     task, allocator = targeted_task['impl'], targeted_task['tf'] if is_tf_available() else targeted_task['pt']
 
-    model = allocator.from_pretrained(model)
+    # Special handling for model conversion
+    from_tf = model.endswith('.h5') and not is_tf_available()
+    from_pt = model.endswith('.bin') and not is_torch_available()
+
+    if from_tf:
+        logger.warning('Model might be a TensorFlow model (ending with `.h5`) but TensorFlow is not available. Trying to load the model with PyTorch.')
+    elif from_pt:
+        logger.warning('Model might be a PyTorch model (ending with `.bin`) but PyTorch is not available. Trying to load the model with Tensorflow.')
+
+    if allocator.__name__.startswith('TF'):
+        model = allocator.from_pretrained(model, config=config, from_pt=from_pt)
+    else:
+        model = allocator.from_pretrained(model, config=config, from_tf=from_tf)
     return task(model, tokenizer, **kwargs)

From 8938b546bf5f61dcb65fb6dd72b5b924f773c46a Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Fri, 13 Dec 2019 14:27:04 +0100
Subject: [PATCH 349/505] Removed from_config

---
 transformers/pipelines.py | 19 ++++++-------------
 1 file changed, 6 insertions(+), 13 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 9acd9bc566..6fbb7e2f04 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -37,11 +37,6 @@ class Pipeline(ABC):
         self.model = model
         self.tokenizer = tokenizer
 
-    @classmethod
-    @abstractmethod
-    def from_config(cls, model, tokenizer: PreTrainedTokenizer, **kwargs):
-        raise NotImplementedError()
-
     def save_pretrained(self, save_directory):
         if not os.path.isdir(save_directory):
             logger.error("Provided path ({}) should be a directory".format(save_directory))
@@ -63,6 +58,12 @@ class Pipeline(ABC):
         raise NotImplementedError()
 
 
+class FeatureExtractionPipeline(Pipeline):
+
+    def __call__(self, *texts, **kwargs):
+        pass
+
+
 class TextClassificationPipeline(Pipeline):
     def __init__(self, model, tokenizer: PreTrainedTokenizer, nb_classes: int = 2):
         super().__init__(model, tokenizer)
@@ -71,10 +72,6 @@ class TextClassificationPipeline(Pipeline):
             raise Exception('Invalid parameter nb_classes. int >= 2 is required (got: {})'.format(nb_classes))
         self._nb_classes = nb_classes
 
-    @classmethod
-    def from_config(cls, model, tokenizer: PreTrainedTokenizer, **kwargs):
-        return cls(model, tokenizer, **kwargs)
-
     def __call__(self, *texts, **kwargs):
         # Generic compatibility with sklearn and Keras
         if 'X' in kwargs and not texts:
@@ -102,10 +99,6 @@ class NerPipeline(Pipeline):
     def __init__(self, model, tokenizer: PreTrainedTokenizer):
         super().__init__(model, tokenizer)
 
-    @classmethod
-    def from_config(cls, model, tokenizer: PreTrainedTokenizer, **kwargs):
-        pass
-
     def __call__(self, *texts, **kwargs):
         (texts, ), answers = texts, []
 

From 47f0e3cfb7df192ab80215cea9096791fce08694 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 13 Dec 2019 14:33:24 +0100
Subject: [PATCH 350/505] cleaning up configuration classes

---
 .../summarization/configuration_bertabs.py    | 10 +--
 .../adding_a_new_model/configuration_xxx.py   | 12 +--
 .../tests/modeling_tf_xxx_test.py             |  2 +-
 .../tests/modeling_xxx_test.py                |  2 +-
 transformers/configuration_albert.py          |  6 +-
 transformers/configuration_bert.py            | 38 +++-----
 transformers/configuration_ctrl.py            | 23 +----
 transformers/configuration_distilbert.py      | 40 ++++-----
 transformers/configuration_gpt2.py            | 55 ++++--------
 transformers/configuration_openai.py          | 57 +++++-------
 transformers/configuration_transfo_xl.py      | 26 ++----
 transformers/configuration_utils.py           | 27 ++++--
 transformers/configuration_xlm.py             | 88 ++++++++-----------
 transformers/configuration_xlnet.py           | 81 +++++++----------
 ..._original_pytorch_checkpoint_to_pytorch.py |  2 +-
 transformers/modeling_gpt2.py                 |  1 +
 transformers/modeling_tf_gpt2.py              |  1 +
 transformers/modeling_tf_transfo_xl.py        |  6 +-
 .../modeling_tf_transfo_xl_utilities.py       | 12 +--
 transformers/modeling_tf_xlnet.py             |  2 +-
 transformers/modeling_transfo_xl.py           | 10 +--
 transformers/modeling_xlnet.py                |  4 +-
 transformers/tests/modeling_albert_test.py    |  2 +-
 transformers/tests/modeling_bert_test.py      |  2 +-
 transformers/tests/modeling_common_test.py    |  2 +-
 transformers/tests/modeling_ctrl_test.py      |  2 +-
 .../tests/modeling_distilbert_test.py         |  2 +-
 transformers/tests/modeling_gpt2_test.py      |  2 +-
 transformers/tests/modeling_openai_test.py    |  2 +-
 transformers/tests/modeling_roberta_test.py   |  2 +-
 transformers/tests/modeling_tf_albert_test.py |  2 +-
 transformers/tests/modeling_tf_bert_test.py   |  2 +-
 transformers/tests/modeling_tf_ctrl_test.py   |  2 +-
 .../tests/modeling_tf_distilbert_test.py      |  2 +-
 transformers/tests/modeling_tf_gpt2_test.py   |  2 +-
 .../tests/modeling_tf_openai_gpt_test.py      |  2 +-
 .../tests/modeling_tf_roberta_test.py         |  2 +-
 .../tests/modeling_tf_transfo_xl_test.py      |  2 +-
 transformers/tests/modeling_tf_xlm_test.py    |  2 +-
 transformers/tests/modeling_tf_xlnet_test.py  |  5 +-
 .../tests/modeling_transfo_xl_test.py         |  2 +-
 transformers/tests/modeling_xlm_test.py       |  2 +-
 transformers/tests/modeling_xlnet_test.py     |  5 +-
 43 files changed, 224 insertions(+), 329 deletions(-)

diff --git a/examples/summarization/configuration_bertabs.py b/examples/summarization/configuration_bertabs.py
index 5bcb65b423..054763ea93 100644
--- a/examples/summarization/configuration_bertabs.py
+++ b/examples/summarization/configuration_bertabs.py
@@ -65,7 +65,7 @@ class BertAbsConfig(PretrainedConfig):
 
     def __init__(
         self,
-        vocab_size_or_config_json_file=30522,
+        vocab_size=30522,
         max_pos=512,
         enc_layers=6,
         enc_hidden_size=512,
@@ -81,14 +81,14 @@ class BertAbsConfig(PretrainedConfig):
     ):
         super(BertAbsConfig, self).__init__(**kwargs)
 
-        if self._input_is_path_to_json(vocab_size_or_config_json_file):
-            path_to_json = vocab_size_or_config_json_file
+        if self._input_is_path_to_json(vocab_size):
+            path_to_json = vocab_size
             with open(path_to_json, "r", encoding="utf-8") as reader:
                 json_config = json.loads(reader.read())
             for key, value in json_config.items():
                 self.__dict__[key] = value
-        elif isinstance(vocab_size_or_config_json_file, int):
-            self.vocab_size = vocab_size_or_config_json_file
+        elif isinstance(vocab_size, int):
+            self.vocab_size = vocab_size
             self.max_pos = max_pos
 
             self.enc_layers = enc_layers
diff --git a/templates/adding_a_new_model/configuration_xxx.py b/templates/adding_a_new_model/configuration_xxx.py
index b1614e71af..ca9e0d554b 100644
--- a/templates/adding_a_new_model/configuration_xxx.py
+++ b/templates/adding_a_new_model/configuration_xxx.py
@@ -39,7 +39,7 @@ class XxxConfig(PretrainedConfig):
 
 
         Arguments:
-            vocab_size_or_config_json_file: Vocabulary size of `inputs_ids` in `XxxModel`.
+            vocab_size: Vocabulary size of `inputs_ids` in `XxxModel`.
             hidden_size: Size of the encoder layers and the pooler layer.
             num_hidden_layers: Number of hidden layers in the Transformer encoder.
             num_attention_heads: Number of attention heads for each attention layer in
@@ -64,7 +64,7 @@ class XxxConfig(PretrainedConfig):
     pretrained_config_archive_map = XXX_PRETRAINED_CONFIG_ARCHIVE_MAP
 
     def __init__(self,
-                 vocab_size_or_config_json_file=50257,
+                 vocab_size=50257,
                  n_positions=1024,
                  n_ctx=1024,
                  n_embd=768,
@@ -84,7 +84,7 @@ class XxxConfig(PretrainedConfig):
                  summary_first_dropout=0.1,
                  **kwargs):
         super(XxxConfig, self).__init__(**kwargs)
-        self.vocab_size = vocab_size_or_config_json_file if isinstance(vocab_size_or_config_json_file, six.string_types) else -1
+        self.vocab_size = vocab_size if isinstance(vocab_size, six.string_types) else -1
         self.n_ctx = n_ctx
         self.n_positions = n_positions
         self.n_embd = n_embd
@@ -102,12 +102,12 @@ class XxxConfig(PretrainedConfig):
         self.summary_activation = summary_activation
         self.summary_first_dropout = summary_first_dropout
         self.summary_proj_to_labels = summary_proj_to_labels
-        if isinstance(vocab_size_or_config_json_file, six.string_types):
-            with open(vocab_size_or_config_json_file, "r", encoding="utf-8") as reader:
+        if isinstance(vocab_size, six.string_types):
+            with open(vocab_size, "r", encoding="utf-8") as reader:
                 json_config = json.loads(reader.read())
             for key, value in json_config.items():
                 self.__dict__[key] = value
-        elif not isinstance(vocab_size_or_config_json_file, int):
+        elif not isinstance(vocab_size, int):
             raise ValueError(
                 "First argument must be either a vocabulary size (int)"
                 "or the path to a pretrained model config file (str)"
diff --git a/templates/adding_a_new_model/tests/modeling_tf_xxx_test.py b/templates/adding_a_new_model/tests/modeling_tf_xxx_test.py
index d7e576bf8b..912a4aa340 100644
--- a/templates/adding_a_new_model/tests/modeling_tf_xxx_test.py
+++ b/templates/adding_a_new_model/tests/modeling_tf_xxx_test.py
@@ -111,7 +111,7 @@ class TFXxxModelTest(TFCommonTestCases.TFCommonModelTester):
                 choice_labels = ids_tensor([self.batch_size], self.num_choices)
 
             config = XxxConfig(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 hidden_size=self.hidden_size,
                 num_hidden_layers=self.num_hidden_layers,
                 num_attention_heads=self.num_attention_heads,
diff --git a/templates/adding_a_new_model/tests/modeling_xxx_test.py b/templates/adding_a_new_model/tests/modeling_xxx_test.py
index bfc70921cd..30e614b3f2 100644
--- a/templates/adding_a_new_model/tests/modeling_xxx_test.py
+++ b/templates/adding_a_new_model/tests/modeling_xxx_test.py
@@ -109,7 +109,7 @@ class XxxModelTest(CommonTestCases.CommonModelTester):
                 choice_labels = ids_tensor([self.batch_size], self.num_choices)
 
             config = XxxConfig(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 hidden_size=self.hidden_size,
                 num_hidden_layers=self.num_hidden_layers,
                 num_attention_heads=self.num_attention_heads,
diff --git a/transformers/configuration_albert.py b/transformers/configuration_albert.py
index de665c9b1c..6a1ef78dd5 100644
--- a/transformers/configuration_albert.py
+++ b/transformers/configuration_albert.py
@@ -37,7 +37,7 @@ class AlbertConfig(PretrainedConfig):
     pretrained_config_archive_map = ALBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
 
     def __init__(self,
-                 vocab_size_or_config_json_file=30000,
+                 vocab_size=30000,
                  embedding_size=128,
                  hidden_size=4096,
                  num_hidden_layers=12,
@@ -83,7 +83,7 @@ class AlbertConfig(PretrainedConfig):
         """
         super(AlbertConfig, self).__init__(**kwargs)
 
-        self.vocab_size = vocab_size_or_config_json_file
+        self.vocab_size = vocab_size
         self.embedding_size = embedding_size
         self.hidden_size = hidden_size
         self.num_hidden_layers = num_hidden_layers
@@ -97,4 +97,4 @@ class AlbertConfig(PretrainedConfig):
         self.max_position_embeddings = max_position_embeddings
         self.type_vocab_size = type_vocab_size
         self.initializer_range = initializer_range
-        self.layer_norm_eps = layer_norm_eps
\ No newline at end of file
+        self.layer_norm_eps = layer_norm_eps
diff --git a/transformers/configuration_bert.py b/transformers/configuration_bert.py
index 01fcd88cb8..9072820bce 100644
--- a/transformers/configuration_bert.py
+++ b/transformers/configuration_bert.py
@@ -56,7 +56,7 @@ class BertConfig(PretrainedConfig):
 
 
         Arguments:
-            vocab_size_or_config_json_file: Vocabulary size of `inputs_ids` in `BertModel`.
+            vocab_size: Vocabulary size of `inputs_ids` in `BertModel`.
             hidden_size: Size of the encoder layers and the pooler layer.
             num_hidden_layers: Number of hidden layers in the Transformer encoder.
             num_attention_heads: Number of attention heads for each attention layer in
@@ -81,7 +81,7 @@ class BertConfig(PretrainedConfig):
     pretrained_config_archive_map = BERT_PRETRAINED_CONFIG_ARCHIVE_MAP
 
     def __init__(self,
-                 vocab_size_or_config_json_file=30522,
+                 vocab_size=30522,
                  hidden_size=768,
                  num_hidden_layers=12,
                  num_attention_heads=12,
@@ -95,25 +95,15 @@ class BertConfig(PretrainedConfig):
                  layer_norm_eps=1e-12,
                  **kwargs):
         super(BertConfig, self).__init__(**kwargs)
-        if isinstance(vocab_size_or_config_json_file, str) or (sys.version_info[0] == 2
-                        and isinstance(vocab_size_or_config_json_file, unicode)):
-            with open(vocab_size_or_config_json_file, "r", encoding='utf-8') as reader:
-                json_config = json.loads(reader.read())
-            for key, value in json_config.items():
-                self.__dict__[key] = value
-        elif isinstance(vocab_size_or_config_json_file, int):
-            self.vocab_size = vocab_size_or_config_json_file
-            self.hidden_size = hidden_size
-            self.num_hidden_layers = num_hidden_layers
-            self.num_attention_heads = num_attention_heads
-            self.hidden_act = hidden_act
-            self.intermediate_size = intermediate_size
-            self.hidden_dropout_prob = hidden_dropout_prob
-            self.attention_probs_dropout_prob = attention_probs_dropout_prob
-            self.max_position_embeddings = max_position_embeddings
-            self.type_vocab_size = type_vocab_size
-            self.initializer_range = initializer_range
-            self.layer_norm_eps = layer_norm_eps
-        else:
-            raise ValueError("First argument must be either a vocabulary size (int)"
-                             " or the path to a pretrained model config file (str)")
+        self.vocab_size = vocab_size
+        self.hidden_size = hidden_size
+        self.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+        self.hidden_act = hidden_act
+        self.intermediate_size = intermediate_size
+        self.hidden_dropout_prob = hidden_dropout_prob
+        self.attention_probs_dropout_prob = attention_probs_dropout_prob
+        self.max_position_embeddings = max_position_embeddings
+        self.type_vocab_size = type_vocab_size
+        self.initializer_range = initializer_range
+        self.layer_norm_eps = layer_norm_eps
diff --git a/transformers/configuration_ctrl.py b/transformers/configuration_ctrl.py
index fcbd848dec..f9b9e409e1 100644
--- a/transformers/configuration_ctrl.py
+++ b/transformers/configuration_ctrl.py
@@ -31,7 +31,7 @@ class CTRLConfig(PretrainedConfig):
     """Configuration class to store the configuration of a `CTRLModel`.
 
     Args:
-        vocab_size_or_config_json_file: Vocabulary size of `inputs_ids` in `CTRLModel` or a configuration json file.
+        vocab_size: Vocabulary size of `inputs_ids` in `CTRLModel` or a configuration json file.
         n_positions: Number of positional embeddings.
         n_ctx: Size of the causal mask (usually same as n_positions).
         dff: Size of the inner dimension of the FFN.
@@ -52,7 +52,7 @@ class CTRLConfig(PretrainedConfig):
 
     def __init__(
         self,
-        vocab_size_or_config_json_file=246534,
+        vocab_size=246534,
         n_positions=256,
         n_ctx=256,
         n_embd=1280,
@@ -64,8 +64,6 @@ class CTRLConfig(PretrainedConfig):
         attn_pdrop=0.1,
         layer_norm_epsilon=1e-6,
         initializer_range=0.02,
-
-        num_labels=1,
         summary_type='cls_index',
         summary_use_proj=True,
         summary_activation=None,
@@ -76,7 +74,7 @@ class CTRLConfig(PretrainedConfig):
         """Constructs CTRLConfig.
 
         Args:
-            vocab_size_or_config_json_file: Vocabulary size of `inputs_ids` in `CTRLModel` or a configuration json file.
+            vocab_size: Vocabulary size of `inputs_ids` in `CTRLModel` or a configuration json file.
             n_positions: Number of positional embeddings.
             n_ctx: Size of the causal mask (usually same as n_positions).
             dff: Size of the inner dimension of the FFN.
@@ -94,8 +92,7 @@ class CTRLConfig(PretrainedConfig):
                 initializing all weight matrices.
         """
         super(CTRLConfig, self).__init__(**kwargs)
-
-        self.vocab_size = vocab_size_or_config_json_file if isinstance(vocab_size_or_config_json_file, int) else -1
+        self.vocab_size = vocab_size
         self.n_ctx = n_ctx
         self.n_positions = n_positions
         self.n_embd = n_embd
@@ -108,23 +105,11 @@ class CTRLConfig(PretrainedConfig):
         self.layer_norm_epsilon = layer_norm_epsilon
         self.initializer_range = initializer_range
 
-        self.num_labels = num_labels
         self.summary_type = summary_type
         self.summary_use_proj = summary_use_proj
         self.summary_activation = summary_activation
         self.summary_first_dropout = summary_first_dropout
         self.summary_proj_to_labels = summary_proj_to_labels
-        if isinstance(vocab_size_or_config_json_file, str) or (sys.version_info[0] == 2
-                        and isinstance(vocab_size_or_config_json_file, unicode)):
-            with open(vocab_size_or_config_json_file, "r", encoding="utf-8") as reader:
-                json_config = json.loads(reader.read())
-            for key, value in json_config.items():
-                self.__dict__[key] = value
-        elif not isinstance(vocab_size_or_config_json_file, int):
-            raise ValueError(
-                "First argument must be either a vocabulary size (int)"
-                "or the path to a pretrained model config file (str)"
-            )
 
     @property
     def max_position_embeddings(self):
diff --git a/transformers/configuration_distilbert.py b/transformers/configuration_distilbert.py
index d5d575be29..d9f7cc6348 100644
--- a/transformers/configuration_distilbert.py
+++ b/transformers/configuration_distilbert.py
@@ -37,7 +37,7 @@ class DistilBertConfig(PretrainedConfig):
     pretrained_config_archive_map = DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
 
     def __init__(self,
-                 vocab_size_or_config_json_file=30522,
+                 vocab_size=30522,
                  max_position_embeddings=512,
                  sinusoidal_pos_embds=False,
                  n_layers=6,
@@ -53,31 +53,21 @@ class DistilBertConfig(PretrainedConfig):
                  seq_classif_dropout=0.2,
                  **kwargs):
         super(DistilBertConfig, self).__init__(**kwargs)
+        self.vocab_size = vocab_size
+        self.max_position_embeddings = max_position_embeddings
+        self.sinusoidal_pos_embds = sinusoidal_pos_embds
+        self.n_layers = n_layers
+        self.n_heads = n_heads
+        self.dim = dim
+        self.hidden_dim = hidden_dim
+        self.dropout = dropout
+        self.attention_dropout = attention_dropout
+        self.activation = activation
+        self.initializer_range = initializer_range
+        self.tie_weights_ = tie_weights_
+        self.qa_dropout = qa_dropout
+        self.seq_classif_dropout = seq_classif_dropout
 
-        if isinstance(vocab_size_or_config_json_file, str) or (sys.version_info[0] == 2
-                        and isinstance(vocab_size_or_config_json_file, unicode)):
-            with open(vocab_size_or_config_json_file, "r", encoding='utf-8') as reader:
-                json_config = json.loads(reader.read())
-            for key, value in json_config.items():
-                self.__dict__[key] = value
-        elif isinstance(vocab_size_or_config_json_file, int):
-            self.vocab_size = vocab_size_or_config_json_file
-            self.max_position_embeddings = max_position_embeddings
-            self.sinusoidal_pos_embds = sinusoidal_pos_embds
-            self.n_layers = n_layers
-            self.n_heads = n_heads
-            self.dim = dim
-            self.hidden_dim = hidden_dim
-            self.dropout = dropout
-            self.attention_dropout = attention_dropout
-            self.activation = activation
-            self.initializer_range = initializer_range
-            self.tie_weights_ = tie_weights_
-            self.qa_dropout = qa_dropout
-            self.seq_classif_dropout = seq_classif_dropout
-        else:
-            raise ValueError("First argument must be either a vocabulary size (int)"
-                             " or the path to a pretrained model config file (str)")
     @property
     def hidden_size(self):
         return self.dim
diff --git a/transformers/configuration_gpt2.py b/transformers/configuration_gpt2.py
index c2fb4948d3..4c200c0760 100644
--- a/transformers/configuration_gpt2.py
+++ b/transformers/configuration_gpt2.py
@@ -36,7 +36,7 @@ class GPT2Config(PretrainedConfig):
     """Configuration class to store the configuration of a `GPT2Model`.
 
     Args:
-        vocab_size_or_config_json_file: Vocabulary size of `inputs_ids` in `GPT2Model` or a configuration json file.
+        vocab_size: Vocabulary size of `inputs_ids` in `GPT2Model` or a configuration json file.
         n_positions: Number of positional embeddings.
         n_ctx: Size of the causal mask (usually same as n_positions).
         n_embd: Dimensionality of the embeddings and hidden states.
@@ -56,7 +56,7 @@ class GPT2Config(PretrainedConfig):
 
     def __init__(
         self,
-        vocab_size_or_config_json_file=50257,
+        vocab_size=50257,
         n_positions=1024,
         n_ctx=1024,
         n_embd=768,
@@ -67,8 +67,6 @@ class GPT2Config(PretrainedConfig):
         attn_pdrop=0.1,
         layer_norm_epsilon=1e-5,
         initializer_range=0.02,
-
-        num_labels=1,
         summary_type='cls_index',
         summary_use_proj=True,
         summary_activation=None,
@@ -79,7 +77,7 @@ class GPT2Config(PretrainedConfig):
         """Constructs GPT2Config.
 
         Args:
-            vocab_size_or_config_json_file: Vocabulary size of `inputs_ids` in `GPT2Model` or a configuration json file.
+            vocab_size: Vocabulary size of `inputs_ids` in `GPT2Model` or a configuration json file.
             n_positions: Number of positional embeddings.
             n_ctx: Size of the causal mask (usually same as n_positions).
             n_embd: Dimensionality of the embeddings and hidden states.
@@ -96,37 +94,22 @@ class GPT2Config(PretrainedConfig):
                 initializing all weight matrices.
         """
         super(GPT2Config, self).__init__(**kwargs)
-
-        if isinstance(vocab_size_or_config_json_file, str) or (sys.version_info[0] == 2
-                        and isinstance(vocab_size_or_config_json_file, unicode)):
-            with open(vocab_size_or_config_json_file, "r", encoding="utf-8") as reader:
-                json_config = json.loads(reader.read())
-            for key, value in json_config.items():
-                self.__dict__[key] = value
-        elif isinstance(vocab_size_or_config_json_file, int):
-            self.vocab_size = vocab_size_or_config_json_file
-            self.n_ctx = n_ctx
-            self.n_positions = n_positions
-            self.n_embd = n_embd
-            self.n_layer = n_layer
-            self.n_head = n_head
-            self.resid_pdrop = resid_pdrop
-            self.embd_pdrop = embd_pdrop
-            self.attn_pdrop = attn_pdrop
-            self.layer_norm_epsilon = layer_norm_epsilon
-            self.initializer_range = initializer_range
-
-            self.num_labels = num_labels
-            self.summary_type = summary_type
-            self.summary_use_proj = summary_use_proj
-            self.summary_activation = summary_activation
-            self.summary_first_dropout = summary_first_dropout
-            self.summary_proj_to_labels = summary_proj_to_labels
-        else:
-            raise ValueError(
-                "First argument must be either a vocabulary size (int)"
-                "or the path to a pretrained model config file (str)"
-            )
+        self.vocab_size = vocab_size
+        self.n_ctx = n_ctx
+        self.n_positions = n_positions
+        self.n_embd = n_embd
+        self.n_layer = n_layer
+        self.n_head = n_head
+        self.resid_pdrop = resid_pdrop
+        self.embd_pdrop = embd_pdrop
+        self.attn_pdrop = attn_pdrop
+        self.layer_norm_epsilon = layer_norm_epsilon
+        self.initializer_range = initializer_range
+        self.summary_type = summary_type
+        self.summary_use_proj = summary_use_proj
+        self.summary_activation = summary_activation
+        self.summary_first_dropout = summary_first_dropout
+        self.summary_proj_to_labels = summary_proj_to_labels
 
     @property
     def max_position_embeddings(self):
diff --git a/transformers/configuration_openai.py b/transformers/configuration_openai.py
index 886b7f5bc5..7776a0bb9f 100644
--- a/transformers/configuration_openai.py
+++ b/transformers/configuration_openai.py
@@ -35,7 +35,7 @@ class OpenAIGPTConfig(PretrainedConfig):
     Configuration class to store the configuration of a `OpenAIGPTModel`.
 
     Args:
-        vocab_size_or_config_json_file: Vocabulary size of `inputs_ids` in `OpenAIGPTModel` or a configuration json file.
+        vocab_size: Vocabulary size of `inputs_ids` in `OpenAIGPTModel` or a configuration json file.
         n_positions: Number of positional embeddings.
         n_ctx: Size of the causal mask (usually same as n_positions).
         n_embd: Dimensionality of the embeddings and hidden states.
@@ -58,7 +58,7 @@ class OpenAIGPTConfig(PretrainedConfig):
 
     def __init__(
         self,
-        vocab_size_or_config_json_file=40478,
+        vocab_size=40478,
         n_positions=512,
         n_ctx=512,
         n_embd=768,
@@ -71,8 +71,6 @@ class OpenAIGPTConfig(PretrainedConfig):
         layer_norm_epsilon=1e-5,
         initializer_range=0.02,
         predict_special_tokens=True,
-
-        num_labels=1,
         summary_type='cls_index',
         summary_use_proj=True,
         summary_activation=None,
@@ -83,39 +81,24 @@ class OpenAIGPTConfig(PretrainedConfig):
         """Constructs OpenAIGPTConfig.
         """
         super(OpenAIGPTConfig, self).__init__(**kwargs)
-
-        if isinstance(vocab_size_or_config_json_file, str) or (sys.version_info[0] == 2
-                        and isinstance(vocab_size_or_config_json_file, unicode)):
-            with open(vocab_size_or_config_json_file, "r", encoding="utf-8") as reader:
-                json_config = json.loads(reader.read())
-            for key, value in json_config.items():
-                self.__dict__[key] = value
-        elif isinstance(vocab_size_or_config_json_file, int):
-            self.vocab_size = vocab_size_or_config_json_file
-            self.n_ctx = n_ctx
-            self.n_positions = n_positions
-            self.n_embd = n_embd
-            self.n_layer = n_layer
-            self.n_head = n_head
-            self.afn = afn
-            self.resid_pdrop = resid_pdrop
-            self.embd_pdrop = embd_pdrop
-            self.attn_pdrop = attn_pdrop
-            self.layer_norm_epsilon = layer_norm_epsilon
-            self.initializer_range = initializer_range
-            self.predict_special_tokens = predict_special_tokens
-
-            self.num_labels = num_labels
-            self.summary_type = summary_type
-            self.summary_use_proj = summary_use_proj
-            self.summary_activation = summary_activation
-            self.summary_first_dropout = summary_first_dropout
-            self.summary_proj_to_labels = summary_proj_to_labels
-        else:
-            raise ValueError(
-                "First argument must be either a vocabulary size (int)"
-                "or the path to a pretrained model config file (str)"
-            )
+        self.vocab_size = vocab_size
+        self.n_ctx = n_ctx
+        self.n_positions = n_positions
+        self.n_embd = n_embd
+        self.n_layer = n_layer
+        self.n_head = n_head
+        self.afn = afn
+        self.resid_pdrop = resid_pdrop
+        self.embd_pdrop = embd_pdrop
+        self.attn_pdrop = attn_pdrop
+        self.layer_norm_epsilon = layer_norm_epsilon
+        self.initializer_range = initializer_range
+        self.predict_special_tokens = predict_special_tokens
+        self.summary_type = summary_type
+        self.summary_use_proj = summary_use_proj
+        self.summary_activation = summary_activation
+        self.summary_first_dropout = summary_first_dropout
+        self.summary_proj_to_labels = summary_proj_to_labels
 
     @property
     def max_position_embeddings(self):
diff --git a/transformers/configuration_transfo_xl.py b/transformers/configuration_transfo_xl.py
index d55a6adbe6..52f0f45a50 100644
--- a/transformers/configuration_transfo_xl.py
+++ b/transformers/configuration_transfo_xl.py
@@ -34,7 +34,7 @@ class TransfoXLConfig(PretrainedConfig):
     """Configuration class to store the configuration of a `TransfoXLModel`.
 
         Args:
-            vocab_size_or_config_json_file: Vocabulary size of `inputs_ids` in `TransfoXLModel` or a configuration json file.
+            vocab_size: Vocabulary size of `inputs_ids` in `TransfoXLModel` or a configuration json file.
             cutoffs: cutoffs for the adaptive softmax
             d_model: Dimensionality of the model's hidden states.
             d_embed: Dimensionality of the embeddings
@@ -68,7 +68,7 @@ class TransfoXLConfig(PretrainedConfig):
     pretrained_config_archive_map = TRANSFO_XL_PRETRAINED_CONFIG_ARCHIVE_MAP
 
     def __init__(self,
-                 vocab_size_or_config_json_file=267735,
+                 vocab_size=267735,
                  cutoffs=[20000, 40000, 200000],
                  d_model=1024,
                  d_embed=1024,
@@ -100,7 +100,7 @@ class TransfoXLConfig(PretrainedConfig):
         """Constructs TransfoXLConfig.
         """
         super(TransfoXLConfig, self).__init__(**kwargs)
-        self.n_token = vocab_size_or_config_json_file if isinstance(vocab_size_or_config_json_file, int) else -1
+        self.vocab_size = vocab_size
         self.cutoffs = []
         self.cutoffs.extend(cutoffs)
         self.tie_weight = tie_weight
@@ -133,27 +133,17 @@ class TransfoXLConfig(PretrainedConfig):
         self.init_std = init_std
         self.layer_norm_epsilon = layer_norm_epsilon
 
-        if isinstance(vocab_size_or_config_json_file, str) or (sys.version_info[0] == 2
-                        and isinstance(vocab_size_or_config_json_file, unicode)):
-            with open(vocab_size_or_config_json_file, "r", encoding='utf-8') as reader:
-                json_config = json.loads(reader.read())
-            for key, value in json_config.items():
-                self.__dict__[key] = value
-        elif not isinstance(vocab_size_or_config_json_file, int):
-            raise ValueError("First argument must be either a vocabulary size (int)"
-                             " or the path to a pretrained model config file (str)")
-
     @property
     def max_position_embeddings(self):
         return self.tgt_len + self.ext_len + self.mem_len
 
     @property
-    def vocab_size(self):
-        return self.n_token
+    def n_token(self):  # Backward compatibility
+        return self.vocab_size
 
-    @vocab_size.setter
-    def vocab_size(self, value):
-        self.n_token = value
+    @n_token.setter
+    def n_token(self, value):  # Backward compatibility
+        self.vocab_size = value
 
     @property
     def hidden_size(self):
diff --git a/transformers/configuration_utils.py b/transformers/configuration_utils.py
index 82959adb57..6c9eeea175 100644
--- a/transformers/configuration_utils.py
+++ b/transformers/configuration_utils.py
@@ -49,8 +49,7 @@ class PretrainedConfig(object):
     pretrained_config_archive_map = {}
 
     def __init__(self, **kwargs):
-        self.finetuning_task = kwargs.pop('finetuning_task', None)
-        self.num_labels = kwargs.pop('num_labels', 2)
+        # Attributes with defaults
         self.output_attentions = kwargs.pop('output_attentions', False)
         self.output_hidden_states = kwargs.pop('output_hidden_states', False)
         self.output_past = kwargs.pop('output_past', True)  # Not used by all models
@@ -59,6 +58,22 @@ class PretrainedConfig(object):
         self.pruned_heads = kwargs.pop('pruned_heads', {})
         self.is_decoder = kwargs.pop('is_decoder', False)
 
+        # Fine-tuning task arguments
+        self.finetuning_task = kwargs.pop('finetuning_task', None)
+        self.num_labels = kwargs.pop('num_labels', 2)
+        self.id2label = kwargs.pop('id2label', {i: 'LABEL_{}'.format(i) for i in range(self.num_labels)})
+        self.id2label = dict((int(key), value) for key, value in self.id2label.items())
+        self.label2id = kwargs.pop('label2id', dict(zip(self.id2label.values(), self.id2label.keys())))
+        self.label2id = dict((key, int(value)) for key, value in self.label2id.items())
+
+        # Additional attributes without default values
+        for key, value in kwargs.items():
+            try:
+                setattr(self, key, value)
+            except AttributeError as err:
+                logger.error("Can't set {} with value {} for {}".format(key, value, self))
+                raise err
+
     def save_pretrained(self, save_directory):
         """ Save a configuration object to the directory `save_directory`, so that it
             can be re-loaded using the :func:`~transformers.PretrainedConfig.from_pretrained` class method.
@@ -183,17 +198,15 @@ class PretrainedConfig(object):
     @classmethod
     def from_dict(cls, json_object):
         """Constructs a `Config` from a Python dictionary of parameters."""
-        config = cls(vocab_size_or_config_json_file=-1)
-        for key, value in json_object.items():
-            setattr(config, key, value)
-        return config
+        return cls(**json_object)
 
     @classmethod
     def from_json_file(cls, json_file):
         """Constructs a `Config` from a json file of parameters."""
         with open(json_file, "r", encoding='utf-8') as reader:
             text = reader.read()
-        return cls.from_dict(json.loads(text))
+        dict_obj = json.loads(text)
+        return cls(**dict_obj)
 
     def __eq__(self, other):
         return self.__dict__ == other.__dict__
diff --git a/transformers/configuration_xlm.py b/transformers/configuration_xlm.py
index fa3a5f40f6..0740cc4026 100644
--- a/transformers/configuration_xlm.py
+++ b/transformers/configuration_xlm.py
@@ -42,7 +42,7 @@ class XLMConfig(PretrainedConfig):
     """Configuration class to store the configuration of a `XLMModel`.
 
     Args:
-        vocab_size_or_config_json_file: Vocabulary size of `inputs_ids` in `XLMModel`.
+        vocab_size: Vocabulary size of `inputs_ids` in `XLMModel`.
         d_model: Size of the encoder layers and the pooler layer.
         n_layer: Number of hidden layers in the Transformer encoder.
         n_head: Number of attention heads for each attention layer in
@@ -81,7 +81,7 @@ class XLMConfig(PretrainedConfig):
     pretrained_config_archive_map = XLM_PRETRAINED_CONFIG_ARCHIVE_MAP
 
     def __init__(self,
-                 vocab_size_or_config_json_file=30145,
+                 vocab_size=30145,
                  emb_dim=2048,
                  n_layers=12,
                  n_heads=16,
@@ -103,9 +103,6 @@ class XLMConfig(PretrainedConfig):
                  unk_index=3,
                  mask_index=5,
                  is_encoder=True,
-
-                 finetuning_task=None,
-                 num_labels=2,
                  summary_type='first',
                  summary_use_proj=True,
                  summary_activation=None,
@@ -117,56 +114,43 @@ class XLMConfig(PretrainedConfig):
         """Constructs XLMConfig.
         """
         super(XLMConfig, self).__init__(**kwargs)
-
-        if isinstance(vocab_size_or_config_json_file, str) or (sys.version_info[0] == 2
-                        and isinstance(vocab_size_or_config_json_file, unicode)):
-            with open(vocab_size_or_config_json_file, "r", encoding='utf-8') as reader:
-                json_config = json.loads(reader.read())
-            for key, value in json_config.items():
-                self.__dict__[key] = value
-        elif isinstance(vocab_size_or_config_json_file, int):
-            self.n_words = vocab_size_or_config_json_file
-            self.emb_dim = emb_dim
-            self.n_layers = n_layers
-            self.n_heads = n_heads
-            self.dropout = dropout
-            self.attention_dropout = attention_dropout
-            self.gelu_activation = gelu_activation
-            self.sinusoidal_embeddings = sinusoidal_embeddings
-            self.causal = causal
-            self.asm = asm
-            self.n_langs = n_langs
-            self.use_lang_emb = use_lang_emb
-            self.layer_norm_eps = layer_norm_eps
-            self.bos_index = bos_index
-            self.eos_index = eos_index
-            self.pad_index = pad_index
-            self.unk_index = unk_index
-            self.mask_index = mask_index
-            self.is_encoder = is_encoder
-            self.max_position_embeddings = max_position_embeddings
-            self.embed_init_std = embed_init_std
-            self.init_std = init_std
-            self.finetuning_task = finetuning_task
-            self.num_labels = num_labels
-            self.summary_type = summary_type
-            self.summary_use_proj = summary_use_proj
-            self.summary_activation = summary_activation
-            self.summary_proj_to_labels = summary_proj_to_labels
-            self.summary_first_dropout = summary_first_dropout
-            self.start_n_top = start_n_top
-            self.end_n_top = end_n_top
-        else:
-            raise ValueError("First argument must be either a vocabulary size (int)"
-                             " or the path to a pretrained model config file (str)")
+        self.vocab_size = vocab_size
+        self.emb_dim = emb_dim
+        self.n_layers = n_layers
+        self.n_heads = n_heads
+        self.dropout = dropout
+        self.attention_dropout = attention_dropout
+        self.gelu_activation = gelu_activation
+        self.sinusoidal_embeddings = sinusoidal_embeddings
+        self.causal = causal
+        self.asm = asm
+        self.n_langs = n_langs
+        self.use_lang_emb = use_lang_emb
+        self.layer_norm_eps = layer_norm_eps
+        self.bos_index = bos_index
+        self.eos_index = eos_index
+        self.pad_index = pad_index
+        self.unk_index = unk_index
+        self.mask_index = mask_index
+        self.is_encoder = is_encoder
+        self.max_position_embeddings = max_position_embeddings
+        self.embed_init_std = embed_init_std
+        self.init_std = init_std
+        self.summary_type = summary_type
+        self.summary_use_proj = summary_use_proj
+        self.summary_activation = summary_activation
+        self.summary_proj_to_labels = summary_proj_to_labels
+        self.summary_first_dropout = summary_first_dropout
+        self.start_n_top = start_n_top
+        self.end_n_top = end_n_top
 
     @property
-    def vocab_size(self):
-        return self.n_words
+    def n_words(self):  # For backward compatibility
+        return self.vocab_size
 
-    @vocab_size.setter
-    def vocab_size(self, value):
-        self.n_words = value
+    @n_words.setter
+    def n_words(self, value):  # For backward compatibility
+        self.vocab_size = value
 
     @property
     def hidden_size(self):
diff --git a/transformers/configuration_xlnet.py b/transformers/configuration_xlnet.py
index 0dbf518849..017c57cfd5 100644
--- a/transformers/configuration_xlnet.py
+++ b/transformers/configuration_xlnet.py
@@ -35,7 +35,7 @@ class XLNetConfig(PretrainedConfig):
     """Configuration class to store the configuration of a ``XLNetModel``.
 
     Args:
-        vocab_size_or_config_json_file: Vocabulary size of ``inputs_ids`` in ``XLNetModel``.
+        vocab_size: Vocabulary size of ``inputs_ids`` in ``XLNetModel``.
         d_model: Size of the encoder layers and the pooler layer.
         n_layer: Number of hidden layers in the Transformer encoder.
         n_head: Number of attention heads for each attention layer in
@@ -72,28 +72,22 @@ class XLNetConfig(PretrainedConfig):
     pretrained_config_archive_map = XLNET_PRETRAINED_CONFIG_ARCHIVE_MAP
 
     def __init__(self,
-                 vocab_size_or_config_json_file=32000,
+                 vocab_size=32000,
                  d_model=1024,
                  n_layer=24,
                  n_head=16,
                  d_inner=4096,
-                 max_position_embeddings=512,
                  ff_activation="gelu",
                  untie_r=True,
                  attn_type="bi",
-
                  initializer_range=0.02,
                  layer_norm_eps=1e-12,
-
                  dropout=0.1,
                  mem_len=None,
                  reuse_len=None,
                  bi_data=False,
                  clamp_len=-1,
                  same_length=False,
-
-                 finetuning_task=None,
-                 num_labels=2,
                  summary_type='last',
                  summary_use_proj=True,
                  summary_activation='tanh',
@@ -104,58 +98,45 @@ class XLNetConfig(PretrainedConfig):
         """Constructs XLNetConfig.
         """
         super(XLNetConfig, self).__init__(**kwargs)
+        self.vocab_size = vocab_size
+        self.d_model = d_model
+        self.n_layer = n_layer
+        self.n_head = n_head
+        assert d_model % n_head == 0
+        self.d_head = d_model // n_head
+        self.ff_activation = ff_activation
+        self.d_inner = d_inner
+        self.untie_r = untie_r
+        self.attn_type = attn_type
 
-        if isinstance(vocab_size_or_config_json_file, str) or (sys.version_info[0] == 2
-                        and isinstance(vocab_size_or_config_json_file, unicode)):
-            with open(vocab_size_or_config_json_file, "r", encoding='utf-8') as reader:
-                json_config = json.loads(reader.read())
-            for key, value in json_config.items():
-                setattr(config, key, value)
-        elif isinstance(vocab_size_or_config_json_file, int):
-            self.n_token = vocab_size_or_config_json_file
-            self.d_model = d_model
-            self.n_layer = n_layer
-            self.n_head = n_head
-            assert d_model % n_head == 0
-            self.d_head = d_model // n_head
-            self.ff_activation = ff_activation
-            self.d_inner = d_inner
-            self.untie_r = untie_r
-            self.attn_type = attn_type
+        self.initializer_range = initializer_range
+        self.layer_norm_eps = layer_norm_eps
 
-            self.initializer_range = initializer_range
-            self.layer_norm_eps = layer_norm_eps
+        self.dropout = dropout
+        self.mem_len = mem_len
+        self.reuse_len = reuse_len
+        self.bi_data = bi_data
+        self.clamp_len = clamp_len
+        self.same_length = same_length
 
-            self.dropout = dropout
-            self.mem_len = mem_len
-            self.reuse_len = reuse_len
-            self.bi_data = bi_data
-            self.clamp_len = clamp_len
-            self.same_length = same_length
-
-            self.finetuning_task = finetuning_task
-            self.num_labels = num_labels
-            self.summary_type = summary_type
-            self.summary_use_proj = summary_use_proj
-            self.summary_activation = summary_activation
-            self.summary_last_dropout = summary_last_dropout
-            self.start_n_top = start_n_top
-            self.end_n_top = end_n_top
-        else:
-            raise ValueError("First argument must be either a vocabulary size (int)"
-                             " or the path to a pretrained model config file (str)")
+        self.summary_type = summary_type
+        self.summary_use_proj = summary_use_proj
+        self.summary_activation = summary_activation
+        self.summary_last_dropout = summary_last_dropout
+        self.start_n_top = start_n_top
+        self.end_n_top = end_n_top
 
     @property
     def max_position_embeddings(self):
         return -1
 
     @property
-    def vocab_size(self):
-        return self.n_token
+    def n_token(self):  # Backward compatibility
+        return self.vocab_size
 
-    @vocab_size.setter
-    def vocab_size(self, value):
-        self.n_token = value
+    @n_token.setter
+    def n_token(self, value):  # Backward compatibility
+        self.vocab_size = value
 
     @property
     def hidden_size(self):
diff --git a/transformers/convert_roberta_original_pytorch_checkpoint_to_pytorch.py b/transformers/convert_roberta_original_pytorch_checkpoint_to_pytorch.py
index 60935add60..b4dc1bb61b 100644
--- a/transformers/convert_roberta_original_pytorch_checkpoint_to_pytorch.py
+++ b/transformers/convert_roberta_original_pytorch_checkpoint_to_pytorch.py
@@ -46,7 +46,7 @@ def convert_roberta_checkpoint_to_pytorch(roberta_checkpoint_path, pytorch_dump_
     roberta = FairseqRobertaModel.from_pretrained(roberta_checkpoint_path)
     roberta.eval()  # disable dropout
     config = BertConfig(
-        vocab_size_or_config_json_file=50265,
+        vocab_size=50265,
         hidden_size=roberta.args.encoder_embed_dim,
         num_hidden_layers=roberta.args.encoder_layers,
         num_attention_heads=roberta.args.encoder_attention_heads,
diff --git a/transformers/modeling_gpt2.py b/transformers/modeling_gpt2.py
index 96fd1c0607..ea660262d7 100644
--- a/transformers/modeling_gpt2.py
+++ b/transformers/modeling_gpt2.py
@@ -634,6 +634,7 @@ class GPT2DoubleHeadsModel(GPT2PreTrainedModel):
     """
     def __init__(self, config):
         super(GPT2DoubleHeadsModel, self).__init__(config)
+        config.num_labels = 1
         self.transformer = GPT2Model(config)
         self.lm_head = nn.Linear(config.n_embd, config.vocab_size, bias=False)
         self.multiple_choice_head = SequenceSummary(config)
diff --git a/transformers/modeling_tf_gpt2.py b/transformers/modeling_tf_gpt2.py
index c738e5e8e3..973473179f 100644
--- a/transformers/modeling_tf_gpt2.py
+++ b/transformers/modeling_tf_gpt2.py
@@ -574,6 +574,7 @@ class TFGPT2DoubleHeadsModel(TFGPT2PreTrainedModel):
     """
     def __init__(self, config, *inputs, **kwargs):
         super(TFGPT2DoubleHeadsModel, self).__init__(config, *inputs, **kwargs)
+        config.num_labels = 1
         self.transformer = TFGPT2MainLayer(config, name='transformer')
         self.multiple_choice_head = TFSequenceSummary(config, initializer_range=config.initializer_range, name='multiple_choice_head')
 
diff --git a/transformers/modeling_tf_transfo_xl.py b/transformers/modeling_tf_transfo_xl.py
index fd325e218e..848edfa37a 100644
--- a/transformers/modeling_tf_transfo_xl.py
+++ b/transformers/modeling_tf_transfo_xl.py
@@ -353,7 +353,7 @@ class TFTransfoXLMainLayer(tf.keras.layers.Layer):
         self.output_attentions = config.output_attentions
         self.output_hidden_states = config.output_hidden_states
 
-        self.n_token = config.n_token
+        self.n_token = config.vocab_size
 
         self.d_embed = config.d_embed
         self.d_model = config.d_model
@@ -361,7 +361,7 @@ class TFTransfoXLMainLayer(tf.keras.layers.Layer):
         self.d_head = config.d_head
         self.untie_r = config.untie_r
 
-        self.word_emb = TFAdaptiveEmbedding(config.n_token, config.d_embed, config.d_model, config.cutoffs, 
+        self.word_emb = TFAdaptiveEmbedding(config.vocab_size, config.d_embed, config.d_model, config.cutoffs, 
                                             div_val=config.div_val, init_std=config.init_std, name='word_emb')
 
         self.drop = tf.keras.layers.Dropout(config.dropout)
@@ -729,7 +729,7 @@ class TFTransfoXLLMHeadModel(TFTransfoXLPreTrainedModel):
             raise NotImplementedError
         # use adaptive softmax (including standard softmax)
         else:
-            self.crit = TFAdaptiveSoftmaxMask(config.n_token, config.d_embed, config.d_model, 
+            self.crit = TFAdaptiveSoftmaxMask(config.vocab_size, config.d_embed, config.d_model, 
                                               config.cutoffs, div_val=config.div_val, name='crit')
 
     def reset_length(self, tgt_len, ext_len, mem_len):
diff --git a/transformers/modeling_tf_transfo_xl_utilities.py b/transformers/modeling_tf_transfo_xl_utilities.py
index e6a6dfe686..f730af851f 100644
--- a/transformers/modeling_tf_transfo_xl_utilities.py
+++ b/transformers/modeling_tf_transfo_xl_utilities.py
@@ -25,15 +25,15 @@ import tensorflow as tf
 from .modeling_tf_utils import shape_list
 
 class TFAdaptiveSoftmaxMask(tf.keras.layers.Layer):
-    def __init__(self, n_token, d_embed, d_proj, cutoffs, div_val=1,
+    def __init__(self, vocab_size, d_embed, d_proj, cutoffs, div_val=1,
                  keep_order=False, **kwargs):
         super(TFAdaptiveSoftmaxMask, self).__init__(**kwargs)
 
-        self.n_token = n_token
+        self.vocab_size = vocab_size
         self.d_embed = d_embed
         self.d_proj = d_proj
 
-        self.cutoffs = cutoffs + [n_token]
+        self.cutoffs = cutoffs + [vocab_size]
         self.cutoff_ends = [0] + self.cutoffs
         self.div_val = div_val
 
@@ -66,11 +66,11 @@ class TFAdaptiveSoftmaxMask(tf.keras.layers.Layer):
                     self.out_projs.append(weight)
                 else:
                     self.out_projs.append(None)
-                weight = self.add_weight(shape=(self.n_token, self.d_embed,),
+                weight = self.add_weight(shape=(self.vocab_size, self.d_embed,),
                                          initializer='zeros',
                                          trainable=True,
                                          name='out_layers_._{}_._weight'.format(i))
-                bias = self.add_weight(shape=(self.n_token,),
+                bias = self.add_weight(shape=(self.vocab_size,),
                                          initializer='zeros',
                                          trainable=True,
                                          name='out_layers_._{}_._bias'.format(i))
@@ -114,7 +114,7 @@ class TFAdaptiveSoftmaxMask(tf.keras.layers.Layer):
         hidden, target = inputs
         head_logprob = 0
         if self.n_clusters == 0:
-            softmax_b = tf.get_variable('bias', [n_token], initializer=tf.zeros_initializer())
+            softmax_b = tf.get_variable('bias', [self.config.vocab_size], initializer=tf.zeros_initializer())
             output = self._logit(hidden, self.out_layers[0][0], self.out_layers[0][1], self.out_projs[0])
             if target is not None:
                 loss = tf.nn.sparse_softmax_cross_entropy_with_logits(labels=target, logits=output)
diff --git a/transformers/modeling_tf_xlnet.py b/transformers/modeling_tf_xlnet.py
index 759b57d835..dde2b6a8df 100644
--- a/transformers/modeling_tf_xlnet.py
+++ b/transformers/modeling_tf_xlnet.py
@@ -366,7 +366,7 @@ class TFXLNetMainLayer(tf.keras.layers.Layer):
         self.use_bfloat16 = config.use_bfloat16
         self.initializer_range = config.initializer_range
 
-        self.word_embedding = TFSharedEmbeddings(config.n_token, config.d_model, initializer_range=config.initializer_range, name='word_embedding')
+        self.word_embedding = TFSharedEmbeddings(config.vocab_size, config.d_model, initializer_range=config.initializer_range, name='word_embedding')
         self.layer = [TFXLNetLayer(config, name='layer_._{}'.format(i)) for i in range(config.n_layer)]
         self.dropout = tf.keras.layers.Dropout(config.dropout)
 
diff --git a/transformers/modeling_transfo_xl.py b/transformers/modeling_transfo_xl.py
index a6a82f0dfe..f87d857a7f 100644
--- a/transformers/modeling_transfo_xl.py
+++ b/transformers/modeling_transfo_xl.py
@@ -592,14 +592,14 @@ class TransfoXLModel(TransfoXLPreTrainedModel):
         self.output_attentions = config.output_attentions
         self.output_hidden_states = config.output_hidden_states
 
-        self.n_token = config.n_token
+        self.n_token = config.vocab_size
 
         self.d_embed = config.d_embed
         self.d_model = config.d_model
         self.n_head = config.n_head
         self.d_head = config.d_head
 
-        self.word_emb = AdaptiveEmbedding(config.n_token, config.d_embed, config.d_model, config.cutoffs,
+        self.word_emb = AdaptiveEmbedding(config.vocab_size, config.d_embed, config.d_model, config.cutoffs,
                                           div_val=config.div_val)
 
         self.drop = nn.Dropout(config.dropout)
@@ -836,11 +836,11 @@ class TransfoXLLMHeadModel(TransfoXLPreTrainedModel):
         self.sample_softmax = config.sample_softmax
         # use sampled softmax
         if config.sample_softmax > 0:
-            self.out_layer = nn.Linear(config.d_model, config.n_token)
-            self.sampler = LogUniformSampler(config.n_token, config.sample_softmax)
+            self.out_layer = nn.Linear(config.d_model, config.vocab_size)
+            self.sampler = LogUniformSampler(config.vocab_size, config.sample_softmax)
         # use adaptive softmax (including standard softmax)
         else:
-            self.crit = ProjectedAdaptiveLogSoftmax(config.n_token, config.d_embed, config.d_model,
+            self.crit = ProjectedAdaptiveLogSoftmax(config.vocab_size, config.d_embed, config.d_model,
                                                     config.cutoffs, div_val=config.div_val)
         self.init_weights()
 
diff --git a/transformers/modeling_xlnet.py b/transformers/modeling_xlnet.py
index 225e5b059b..daed5f2857 100644
--- a/transformers/modeling_xlnet.py
+++ b/transformers/modeling_xlnet.py
@@ -609,7 +609,7 @@ class XLNetModel(XLNetPreTrainedModel):
         self.clamp_len = config.clamp_len
         self.n_layer = config.n_layer
 
-        self.word_embedding = nn.Embedding(config.n_token, config.d_model)
+        self.word_embedding = nn.Embedding(config.vocab_size, config.d_model)
         self.mask_emb = nn.Parameter(torch.FloatTensor(1, 1, config.d_model))
         self.layer = nn.ModuleList([XLNetLayer(config) for _ in range(config.n_layer)])
         self.dropout = nn.Dropout(config.dropout)
@@ -940,7 +940,7 @@ class XLNetLMHeadModel(XLNetPreTrainedModel):
         self.same_length = config.same_length
 
         self.transformer = XLNetModel(config)
-        self.lm_loss = nn.Linear(config.d_model, config.n_token, bias=True)
+        self.lm_loss = nn.Linear(config.d_model, config.vocab_size, bias=True)
 
         self.init_weights()
 
diff --git a/transformers/tests/modeling_albert_test.py b/transformers/tests/modeling_albert_test.py
index a14d66ae8f..1911d244e7 100644
--- a/transformers/tests/modeling_albert_test.py
+++ b/transformers/tests/modeling_albert_test.py
@@ -110,7 +110,7 @@ class AlbertModelTest(CommonTestCases.CommonModelTester):
                 choice_labels = ids_tensor([self.batch_size], self.num_choices)
 
             config = AlbertConfig(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 hidden_size=self.hidden_size,
                 num_hidden_layers=self.num_hidden_layers,
                 num_attention_heads=self.num_attention_heads,
diff --git a/transformers/tests/modeling_bert_test.py b/transformers/tests/modeling_bert_test.py
index 539f66cd3f..0eb7bc9a14 100644
--- a/transformers/tests/modeling_bert_test.py
+++ b/transformers/tests/modeling_bert_test.py
@@ -109,7 +109,7 @@ class BertModelTest(CommonTestCases.CommonModelTester):
                 choice_labels = ids_tensor([self.batch_size], self.num_choices)
 
             config = BertConfig(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 hidden_size=self.hidden_size,
                 num_hidden_layers=self.num_hidden_layers,
                 num_attention_heads=self.num_attention_heads,
diff --git a/transformers/tests/modeling_common_test.py b/transformers/tests/modeling_common_test.py
index 80d5d95455..f86eb7a3d0 100644
--- a/transformers/tests/modeling_common_test.py
+++ b/transformers/tests/modeling_common_test.py
@@ -633,7 +633,7 @@ class CommonTestCases:
                 mc_token_ids = ids_tensor([self.batch_size, self.n_choices], self.seq_length)
 
             config = self.config_class(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 n_positions=self.n_positions,
                 n_embd=self.hidden_size,
                 n_layer=self.num_hidden_layers,
diff --git a/transformers/tests/modeling_ctrl_test.py b/transformers/tests/modeling_ctrl_test.py
index 8c14578a5c..c7de49b2ab 100644
--- a/transformers/tests/modeling_ctrl_test.py
+++ b/transformers/tests/modeling_ctrl_test.py
@@ -114,7 +114,7 @@ class CTRLModelTest(CommonTestCases.CommonModelTester):
                 choice_labels = ids_tensor([self.batch_size], self.num_choices)
 
             config = CTRLConfig(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 n_embd=self.hidden_size,
                 n_layer=self.num_hidden_layers,
                 n_head=self.num_attention_heads,
diff --git a/transformers/tests/modeling_distilbert_test.py b/transformers/tests/modeling_distilbert_test.py
index 4b8f64327d..82f71c40da 100644
--- a/transformers/tests/modeling_distilbert_test.py
+++ b/transformers/tests/modeling_distilbert_test.py
@@ -105,7 +105,7 @@ class DistilBertModelTest(CommonTestCases.CommonModelTester):
                 choice_labels = ids_tensor([self.batch_size], self.num_choices)
 
             config = DistilBertConfig(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 dim=self.hidden_size,
                 n_layers=self.num_hidden_layers,
                 n_heads=self.num_attention_heads,
diff --git a/transformers/tests/modeling_gpt2_test.py b/transformers/tests/modeling_gpt2_test.py
index ecaa2a4bd0..a82e39c261 100644
--- a/transformers/tests/modeling_gpt2_test.py
+++ b/transformers/tests/modeling_gpt2_test.py
@@ -110,7 +110,7 @@ class GPT2ModelTest(CommonTestCases.CommonModelTester):
                 choice_labels = ids_tensor([self.batch_size], self.num_choices)
 
             config = GPT2Config(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 n_embd=self.hidden_size,
                 n_layer=self.num_hidden_layers,
                 n_head=self.num_attention_heads,
diff --git a/transformers/tests/modeling_openai_test.py b/transformers/tests/modeling_openai_test.py
index 8e4d13438d..7655e432e8 100644
--- a/transformers/tests/modeling_openai_test.py
+++ b/transformers/tests/modeling_openai_test.py
@@ -98,7 +98,7 @@ class OpenAIGPTModelTest(CommonTestCases.CommonModelTester):
                 choice_labels = ids_tensor([self.batch_size], self.num_choices)
 
             config = OpenAIGPTConfig(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 n_embd=self.hidden_size,
                 n_layer=self.num_hidden_layers,
                 n_head=self.num_attention_heads,
diff --git a/transformers/tests/modeling_roberta_test.py b/transformers/tests/modeling_roberta_test.py
index 7a3553b164..4d34a50528 100644
--- a/transformers/tests/modeling_roberta_test.py
+++ b/transformers/tests/modeling_roberta_test.py
@@ -106,7 +106,7 @@ class RobertaModelTest(CommonTestCases.CommonModelTester):
                 choice_labels = ids_tensor([self.batch_size], self.num_choices)
 
             config = RobertaConfig(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 hidden_size=self.hidden_size,
                 num_hidden_layers=self.num_hidden_layers,
                 num_attention_heads=self.num_attention_heads,
diff --git a/transformers/tests/modeling_tf_albert_test.py b/transformers/tests/modeling_tf_albert_test.py
index 7d3325b70b..93aeab66c2 100644
--- a/transformers/tests/modeling_tf_albert_test.py
+++ b/transformers/tests/modeling_tf_albert_test.py
@@ -118,7 +118,7 @@ class TFAlbertModelTest(TFCommonTestCases.TFCommonModelTester):
                 choice_labels = ids_tensor([self.batch_size], self.num_choices)
 
             config = AlbertConfig(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 hidden_size=self.hidden_size,
                 num_hidden_layers=self.num_hidden_layers,
                 num_attention_heads=self.num_attention_heads,
diff --git a/transformers/tests/modeling_tf_bert_test.py b/transformers/tests/modeling_tf_bert_test.py
index d7a86fecb9..20073e1ab8 100644
--- a/transformers/tests/modeling_tf_bert_test.py
+++ b/transformers/tests/modeling_tf_bert_test.py
@@ -114,7 +114,7 @@ class TFBertModelTest(TFCommonTestCases.TFCommonModelTester):
                 choice_labels = ids_tensor([self.batch_size], self.num_choices)
 
             config = BertConfig(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 hidden_size=self.hidden_size,
                 num_hidden_layers=self.num_hidden_layers,
                 num_attention_heads=self.num_attention_heads,
diff --git a/transformers/tests/modeling_tf_ctrl_test.py b/transformers/tests/modeling_tf_ctrl_test.py
index 0b421c20c9..0876582e57 100644
--- a/transformers/tests/modeling_tf_ctrl_test.py
+++ b/transformers/tests/modeling_tf_ctrl_test.py
@@ -112,7 +112,7 @@ class TFCTRLModelTest(TFCommonTestCases.TFCommonModelTester):
                 choice_labels = ids_tensor([self.batch_size], self.num_choices)
 
             config = CTRLConfig(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 n_embd=self.hidden_size,
                 n_layer=self.num_hidden_layers,
                 n_head=self.num_attention_heads,
diff --git a/transformers/tests/modeling_tf_distilbert_test.py b/transformers/tests/modeling_tf_distilbert_test.py
index 0ec45150ca..d9e971c2a5 100644
--- a/transformers/tests/modeling_tf_distilbert_test.py
+++ b/transformers/tests/modeling_tf_distilbert_test.py
@@ -107,7 +107,7 @@ class TFDistilBertModelTest(TFCommonTestCases.TFCommonModelTester):
                 choice_labels = ids_tensor([self.batch_size], self.num_choices)
 
             config = DistilBertConfig(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 dim=self.hidden_size,
                 n_layers=self.num_hidden_layers,
                 n_heads=self.num_attention_heads,
diff --git a/transformers/tests/modeling_tf_gpt2_test.py b/transformers/tests/modeling_tf_gpt2_test.py
index e070b72e65..3f30b32787 100644
--- a/transformers/tests/modeling_tf_gpt2_test.py
+++ b/transformers/tests/modeling_tf_gpt2_test.py
@@ -115,7 +115,7 @@ class TFGPT2ModelTest(TFCommonTestCases.TFCommonModelTester):
                 choice_labels = ids_tensor([self.batch_size], self.num_choices)
 
             config = GPT2Config(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 n_embd=self.hidden_size,
                 n_layer=self.num_hidden_layers,
                 n_head=self.num_attention_heads,
diff --git a/transformers/tests/modeling_tf_openai_gpt_test.py b/transformers/tests/modeling_tf_openai_gpt_test.py
index 675e806c12..863dbf1bc0 100644
--- a/transformers/tests/modeling_tf_openai_gpt_test.py
+++ b/transformers/tests/modeling_tf_openai_gpt_test.py
@@ -114,7 +114,7 @@ class TFOpenAIGPTModelTest(TFCommonTestCases.TFCommonModelTester):
                 choice_labels = ids_tensor([self.batch_size], self.num_choices)
 
             config = OpenAIGPTConfig(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 n_embd=self.hidden_size,
                 n_layer=self.num_hidden_layers,
                 n_head=self.num_attention_heads,
diff --git a/transformers/tests/modeling_tf_roberta_test.py b/transformers/tests/modeling_tf_roberta_test.py
index 42440bf1b7..f4ed97c44b 100644
--- a/transformers/tests/modeling_tf_roberta_test.py
+++ b/transformers/tests/modeling_tf_roberta_test.py
@@ -109,7 +109,7 @@ class TFRobertaModelTest(TFCommonTestCases.TFCommonModelTester):
                 choice_labels = ids_tensor([self.batch_size], self.num_choices)
 
             config = RobertaConfig(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 hidden_size=self.hidden_size,
                 num_hidden_layers=self.num_hidden_layers,
                 num_attention_heads=self.num_attention_heads,
diff --git a/transformers/tests/modeling_tf_transfo_xl_test.py b/transformers/tests/modeling_tf_transfo_xl_test.py
index 03e332bdc1..553263250a 100644
--- a/transformers/tests/modeling_tf_transfo_xl_test.py
+++ b/transformers/tests/modeling_tf_transfo_xl_test.py
@@ -92,7 +92,7 @@ class TFTransfoXLModelTest(TFCommonTestCases.TFCommonModelTester):
                 lm_labels = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)
 
             config = TransfoXLConfig(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 mem_len=self.mem_len,
                 clamp_len=self.clamp_len,
                 cutoffs=self.cutoffs,
diff --git a/transformers/tests/modeling_tf_xlm_test.py b/transformers/tests/modeling_tf_xlm_test.py
index a680b70367..228e436149 100644
--- a/transformers/tests/modeling_tf_xlm_test.py
+++ b/transformers/tests/modeling_tf_xlm_test.py
@@ -125,7 +125,7 @@ class TFXLMModelTest(TFCommonTestCases.TFCommonModelTester):
                 is_impossible_labels = ids_tensor([self.batch_size], 2, dtype=tf.float32)
 
             config = XLMConfig(
-                 vocab_size_or_config_json_file=self.vocab_size,
+                 vocab_size=self.vocab_size,
                  n_special=self.n_special,
                  emb_dim=self.hidden_size,
                  n_layers=self.num_hidden_layers,
diff --git a/transformers/tests/modeling_tf_xlnet_test.py b/transformers/tests/modeling_tf_xlnet_test.py
index 94864b86f2..eb66d92793 100644
--- a/transformers/tests/modeling_tf_xlnet_test.py
+++ b/transformers/tests/modeling_tf_xlnet_test.py
@@ -64,7 +64,6 @@ class TFXLNetModelTest(TFCommonTestCases.TFCommonModelTester):
                      num_attention_heads=4,
                      d_inner=128,
                      num_hidden_layers=5,
-                     max_position_embeddings=10,
                      type_sequence_label_size=2,
                      untie_r=True,
                      bi_data=False,
@@ -88,7 +87,6 @@ class TFXLNetModelTest(TFCommonTestCases.TFCommonModelTester):
             self.num_attention_heads = num_attention_heads
             self.d_inner = d_inner
             self.num_hidden_layers = num_hidden_layers
-            self.max_position_embeddings = max_position_embeddings
             self.bi_data = bi_data
             self.untie_r = untie_r
             self.same_length = same_length
@@ -122,13 +120,12 @@ class TFXLNetModelTest(TFCommonTestCases.TFCommonModelTester):
                 is_impossible_labels = ids_tensor([self.batch_size], 2, dtype=tf.float32)
 
             config = XLNetConfig(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 d_model=self.hidden_size,
                 n_head=self.num_attention_heads,
                 d_inner=self.d_inner,
                 n_layer=self.num_hidden_layers,
                 untie_r=self.untie_r,
-                max_position_embeddings=self.max_position_embeddings,
                 mem_len=self.mem_len,
                 clamp_len=self.clamp_len,
                 same_length=self.same_length,
diff --git a/transformers/tests/modeling_transfo_xl_test.py b/transformers/tests/modeling_transfo_xl_test.py
index 647dd3724d..dca46444ba 100644
--- a/transformers/tests/modeling_transfo_xl_test.py
+++ b/transformers/tests/modeling_transfo_xl_test.py
@@ -91,7 +91,7 @@ class TransfoXLModelTest(CommonTestCases.CommonModelTester):
                 lm_labels = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)
 
             config = TransfoXLConfig(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 mem_len=self.mem_len,
                 clamp_len=self.clamp_len,
                 cutoffs=self.cutoffs,
diff --git a/transformers/tests/modeling_xlm_test.py b/transformers/tests/modeling_xlm_test.py
index f6b980767c..7cae6c848e 100644
--- a/transformers/tests/modeling_xlm_test.py
+++ b/transformers/tests/modeling_xlm_test.py
@@ -121,7 +121,7 @@ class XLMModelTest(CommonTestCases.CommonModelTester):
                 is_impossible_labels = ids_tensor([self.batch_size], 2).float()
 
             config = XLMConfig(
-                 vocab_size_or_config_json_file=self.vocab_size,
+                 vocab_size=self.vocab_size,
                  n_special=self.n_special,
                  emb_dim=self.hidden_size,
                  n_layers=self.num_hidden_layers,
diff --git a/transformers/tests/modeling_xlnet_test.py b/transformers/tests/modeling_xlnet_test.py
index 56b6bb3f4d..6d901ee699 100644
--- a/transformers/tests/modeling_xlnet_test.py
+++ b/transformers/tests/modeling_xlnet_test.py
@@ -60,7 +60,6 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
                      num_attention_heads=4,
                      d_inner=128,
                      num_hidden_layers=5,
-                     max_position_embeddings=10,
                      type_sequence_label_size=2,
                      untie_r=True,
                      bi_data=False,
@@ -84,7 +83,6 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
             self.num_attention_heads = num_attention_heads
             self.d_inner = d_inner
             self.num_hidden_layers = num_hidden_layers
-            self.max_position_embeddings = max_position_embeddings
             self.bi_data = bi_data
             self.untie_r = untie_r
             self.same_length = same_length
@@ -116,13 +114,12 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
                 token_labels = ids_tensor([self.batch_size, self.seq_length], self.type_vocab_size)
 
             config = XLNetConfig(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 d_model=self.hidden_size,
                 n_head=self.num_attention_heads,
                 d_inner=self.d_inner,
                 n_layer=self.num_hidden_layers,
                 untie_r=self.untie_r,
-                max_position_embeddings=self.max_position_embeddings,
                 mem_len=self.mem_len,
                 clamp_len=self.clamp_len,
                 same_length=self.same_length,

From 8ade2040984c2cd3fd04bf56b133f70718254b03 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 13 Dec 2019 14:48:47 +0100
Subject: [PATCH 351/505] fix tf

---
 transformers/modeling_openai.py    | 1 +
 transformers/modeling_tf_openai.py | 1 +
 2 files changed, 2 insertions(+)

diff --git a/transformers/modeling_openai.py b/transformers/modeling_openai.py
index 4fe7ffee8b..72f1224e39 100644
--- a/transformers/modeling_openai.py
+++ b/transformers/modeling_openai.py
@@ -590,6 +590,7 @@ class OpenAIGPTDoubleHeadsModel(OpenAIGPTPreTrainedModel):
     def __init__(self, config):
         super(OpenAIGPTDoubleHeadsModel, self).__init__(config)
 
+        config.num_labels = 1
         self.transformer = OpenAIGPTModel(config)
         self.lm_head = nn.Linear(config.n_embd, config.vocab_size, bias=False)
         self.multiple_choice_head = SequenceSummary(config)
diff --git a/transformers/modeling_tf_openai.py b/transformers/modeling_tf_openai.py
index dac3b17590..bd469f0205 100644
--- a/transformers/modeling_tf_openai.py
+++ b/transformers/modeling_tf_openai.py
@@ -538,6 +538,7 @@ class TFOpenAIGPTDoubleHeadsModel(TFOpenAIGPTPreTrainedModel):
     """
     def __init__(self, config, *inputs, **kwargs):
         super(TFOpenAIGPTDoubleHeadsModel, self).__init__(config, *inputs, **kwargs)
+        config.num_labels = 1
         self.transformer = TFOpenAIGPTMainLayer(config, name='transformer')
         self.multiple_choice_head = TFSequenceSummary(config, initializer_range=config.initializer_range, name='multiple_choice_head')
 

From 5a5c4349e8a141d2c0915d71cb3cee101da0db6f Mon Sep 17 00:00:00 2001
From: Pierric Cistac <Pierrci@users.noreply.github.com>
Date: Fri, 13 Dec 2019 10:02:33 -0500
Subject: [PATCH 352/505] Fix summarization `to_cpu` doc

---
 examples/summarization/README.md | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/examples/summarization/README.md b/examples/summarization/README.md
index 96825cfa46..b98581e8e5 100644
--- a/examples/summarization/README.md
+++ b/examples/summarization/README.md
@@ -29,7 +29,7 @@ And move all the stories to the same folder. We will refer as `$DATA_PATH` the p
 python run_summarization.py \
     --documents_dir $DATA_PATH \
     --summaries_output_dir $SUMMARIES_PATH \ # optional
-    --to_cpu false \
+    --no_cuda false \
     --batch_size 4 \
     --min_length 50 \
     --max_length 200 \
@@ -39,7 +39,7 @@ python run_summarization.py \
     --compute_rouge true
 ```
 
-The scripts executes on GPU if one is available and if `to_cpu` is not set to `true`. Inference on multiple GPUs is not suported yet. The ROUGE scores will be displayed in the console at the end of evaluation and written in a `rouge_scores.txt` file. The script takes 30 hours to compute with a single Tesla V100 GPU and a batch size of 10 (300,000 texts to summarize).
+The scripts executes on GPU if one is available and if `no_cuda` is not set to `true`. Inference on multiple GPUs is not suported yet. The ROUGE scores will be displayed in the console at the end of evaluation and written in a `rouge_scores.txt` file. The script takes 30 hours to compute with a single Tesla V100 GPU and a batch size of 10 (300,000 texts to summarize).
 
 ## Summarize any text
 
@@ -49,7 +49,7 @@ Put the documents that you would like to summarize in a folder (the path to whic
 python run_summarization.py \
     --documents_dir $DATA_PATH \
     --summaries_output_dir $SUMMARIES_PATH \ # optional
-    --to_cpu false \
+    --no_cuda false \
     --batch_size 4 \
     --min_length 50 \
     --max_length 200 \

From 0b51532ce94140cdb22f761b09fff28cce76f985 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Fri, 13 Dec 2019 16:22:50 +0100
Subject: [PATCH 353/505] Reintroducing the batch_encode_plus method

---
 transformers/tokenization_utils.py | 86 ++++++++++++++++++++++++++++++
 1 file changed, 86 insertions(+)

diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index f4395cd82c..169caff8dc 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -878,6 +878,92 @@ class PreTrainedTokenizer(object):
                                       return_overflowing_tokens=return_overflowing_tokens,
                                       return_special_tokens_mask=return_special_tokens_mask)
 
+    def batch_encode_plus(self,
+                          batch_text_or_text_pairs=None,
+                          add_special_tokens=False,
+                          max_length=None,
+                          stride=0,
+                          truncation_strategy='longest_first',
+                          return_tensors=None,
+                          return_input_lengths=False,
+                          return_attention_masks=False,
+                          **kwargs):
+        """
+        Returns a dictionary containing the encoded sequence or sequence pair and additional information:
+        the mask for sequence classification and the overflowing elements if a ``max_length`` is specified.
+
+        Args:
+            batch_text_or_text_pairs: Batch of sequences or pair of sequences to be encoded.
+                This can be a list of string/string-sequences/int-sequences or a list of pair of
+                string/string-sequences/int-sequence (see details in encode_plus)
+            add_special_tokens: if set to ``True``, the sequences will be encoded with the special tokens relative
+                to their model.
+            max_length: if set to a number, will limit the total sequence returned so that it has a maximum length.
+                If there are overflowing tokens, those will be added to the returned dictionary`
+            stride: if set to a number along with max_length, the overflowing tokens returned will contain some tokens
+                from the main sequence returned. The value of this argument defines the number of additional tokens.
+            truncation_strategy: string selected in the following options:
+                - 'longest_first' (default) Iteratively reduce the inputs sequence until the input is under max_length
+                    starting from the longest one at each token (when there is a pair of input sequences)
+                - 'only_first': Only truncate the first sequence
+                - 'only_second': Only truncate the second sequence
+                - 'do_not_truncate': Does not truncate (raise an error if the input sequence is longer than max_length)
+            return_tensors: (optional) can be set to 'tf' or 'pt' to return respectively TensorFlow tf.constant
+                or PyTorch torch.Tensor instead of a list of python integers.
+            **kwargs: passed to the `self.tokenize()` method
+        """
+        batch_outputs = {}
+        for ids_or_pair_ids in batch_text_or_text_pairs:
+            if isinstance(ids_or_pair_ids, (list, tuple)):
+                assert len(ids_or_pair_ids) == 2
+                ids, pair_ids = ids_or_pair_ids
+            else:
+                ids, pair_ids = ids_or_pair_ids, None
+            outputs = self.encode_plus(ids, pair_ids, add_special_tokens=add_special_tokens, max_length=max_length,
+                                       stride=stride, truncation_strategy=truncation_strategy, return_tensors=None)
+
+            # Append the non-padded length to the output
+            if return_input_lengths:
+                outputs['input_len'] = len(outputs['input_ids'])
+
+            for key, value in outputs.items():
+                if key not in batch_outputs:
+                    batch_outputs[key] = []
+                batch_outputs[key].append(value)
+
+        # Compute longest sequence size
+        max_seq_len = max(map(len, batch_outputs['input_ids']))
+
+        if return_attention_masks:
+            # Allow the model to not give any special attention to padded input
+            batch_outputs['attention_mask'] = [[0] * len(v) for v in batch_outputs['input_ids']]
+
+        if return_tensors is not None:
+
+            # Do the tensor conversion in batch
+            for key, value in batch_outputs.items():
+
+                padded_value = value
+                if key != 'input_len':
+                    # Padding handle
+                    padded_value = [v + [self.pad_token_id if key == 'input_ids' else 1] * (max_seq_len - len(v)) for v in padded_value]
+
+                if return_tensors == 'tf' and is_tf_available():
+                    batch_outputs[key] = tf.constant(padded_value)
+                elif return_tensors == 'pt' and is_torch_available():
+                    batch_outputs[key] = torch.tensor(padded_value)
+                elif return_tensors is not None:
+                    logger.warning("Unable to convert output to tensors format {}, PyTorch or TensorFlow is not available.".format(return_tensors))
+
+        # encoder_attention_mask requires 1 for real token, 0 for padding, just invert value
+        if return_attention_masks:
+            if is_tf_available():
+                batch_outputs['attention_mask'] = tf.abs(batch_outputs['attention_mask'] - 1)
+            else:
+                batch_outputs['attention_mask'] = torch.abs(batch_outputs['attention_mask'] - 1)
+
+        return batch_outputs
+
     def prepare_for_model(self, ids, pair_ids=None, max_length=None, add_special_tokens=True, stride=0,
                           truncation_strategy='longest_first',
                           pad_to_max_length=False,

From 5c00e344c1350e079d428a4d69cbb465ca7ffde9 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 13 Dec 2019 16:33:29 +0100
Subject: [PATCH 354/505] update model doc - swith 3B/11B to 3b/11b

---
 docs/source/pretrained_models.rst | 25 ++++++++++---------------
 transformers/configuration_t5.py  |  4 ++--
 transformers/modeling_t5.py       |  4 ++--
 transformers/modeling_tf_t5.py    |  4 ++--
 transformers/tokenization_t5.py   |  8 ++++----
 5 files changed, 20 insertions(+), 25 deletions(-)

diff --git a/docs/source/pretrained_models.rst b/docs/source/pretrained_models.rst
index 7e1366b53a..c6b990f213 100644
--- a/docs/source/pretrained_models.rst
+++ b/docs/source/pretrained_models.rst
@@ -217,25 +217,20 @@ Here is the full list of the currently provided pretrained models together with
 |                   |                                                            | | ALBERT xxlarge model with no dropout, additional training data and longer training                                                  |
 |                   |                                                            | (see `details <https://github.com/google-research/ALBERT>`__)                                                                         |
 +-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
-| T5                | ``t5-small``                                               | | 6-layer, 768-hidden, 12-heads, 66M parameters                                                                                       |
-|                   |                                                            | | The DistilBERT model distilled from the BERT model `bert-base-uncased` checkpoint                                                   |
-|                   |                                                            | (see `details <https://github.com/huggingface/transformers/tree/master/examples/distillation>`__)                                     |
+| T5                | ``t5-small``                                               | | ~60M parameters with 6-layers, 512-hidden-state, 2048 feed-forward hidden-state, 8-heads,                                           |
+|                   |                                                            | | Trained on English text: the Colossal Clean Crawled Corpus (C4)                                                                     |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
-|                   | ``t5-base``                                                | | 6-layer, 768-hidden, 12-heads, 66M parameters                                                                                       |
-|                   |                                                            | | The DistilBERT model distilled from the BERT model `bert-base-uncased` checkpoint, with an additional linear layer.                 |
-|                   |                                                            | (see `details <https://github.com/huggingface/transformers/tree/master/examples/distillation>`__)                                     |
+|                   | ``t5-base``                                                | | ~220M parameters with 12-layers, 768-hidden-state, 3072 feed-forward hidden-state, 12-heads,                                        |
+|                   |                                                            | | Trained on English text: the Colossal Clean Crawled Corpus (C4)                                                                     |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
-|                   | ``t5-large``                                               | | 6-layer, 768-hidden, 12-heads, 82M parameters                                                                                       |
-|                   |                                                            | | The DistilGPT2 model distilled from the GPT2 model `gpt2` checkpoint.                                                               |
-|                   |                                                            | (see `details <https://github.com/huggingface/transformers/tree/master/examples/distillation>`__)                                     |
+|                   | ``t5-large``                                               | | ~770M parameters with 24-layers, 1024-hidden-state, 4096 feed-forward hidden-state, 16-heads,                                       |
+|                   |                                                            | | Trained on English text: the Colossal Clean Crawled Corpus (C4)                                                                     |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
-|                   | ``t5-3b``                                                  | | 6-layer, 768-hidden, 12-heads, 82M parameters                                                                                       |
-|                   |                                                            | | The DistilRoBERTa model distilled from the RoBERTa model `roberta-base` checkpoint.                                                 |
-|                   |                                                            | (see `details <https://github.com/huggingface/transformers/tree/master/examples/distillation>`__)                                     |
+|                   | ``t5-3B``                                                  | | ~2.8B parameters with 24-layers, 1024-hidden-state, 16384 feed-forward hidden-state, 32-heads,                                      |
+|                   |                                                            | | Trained on English text: the Colossal Clean Crawled Corpus (C4)                                                                     |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
-|                   | ``t5-11b``                                                 | | 6-layer, 768-hidden, 12-heads, 82M parameters                                                                                       |
-|                   |                                                            | | The DistilRoBERTa model distilled from the RoBERTa model `roberta-base` checkpoint.                                                 |
-|                   |                                                            | (see `details <https://github.com/huggingface/transformers/tree/master/examples/distillation>`__)                                     |
+|                   | ``t5-11B``                                                 | | ~11B parameters with 24-layers, 1024-hidden-state, 65536 feed-forward hidden-state, 128-heads,                                      |
+|                   |                                                            | | Trained on English text: the Colossal Clean Crawled Corpus (C4)                                                                     |
 +-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 
 
diff --git a/transformers/configuration_t5.py b/transformers/configuration_t5.py
index 2ccdebc2b1..6391cb4180 100644
--- a/transformers/configuration_t5.py
+++ b/transformers/configuration_t5.py
@@ -30,8 +30,8 @@ T5_PRETRAINED_CONFIG_ARCHIVE_MAP = {
     't5-small': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-small-config.json",
     't5-base': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-base-config.json",
     't5-large': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-large-config.json",
-    't5-3B': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-3B-config.json",
-    't5-11B': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-11B-config.json",
+    't5-3b': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-3b-config.json",
+    't5-11b': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-11b-config.json",
 }
 
 
diff --git a/transformers/modeling_t5.py b/transformers/modeling_t5.py
index c9310179a3..263dc33b70 100644
--- a/transformers/modeling_t5.py
+++ b/transformers/modeling_t5.py
@@ -44,8 +44,8 @@ T5_PRETRAINED_MODEL_ARCHIVE_MAP = {
     't5-small': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-small-pytorch_model.bin",
     't5-base': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-base-pytorch_model.bin",
     't5-large': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-large-pytorch_model.bin",
-    't5-3B': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-3B-pytorch_model.bin",
-    't5-11B': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-11B-pytorch_model.bin",
+    't5-3b': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-3b-pytorch_model.bin",
+    't5-11b': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-11b-pytorch_model.bin",
 }
 
 ####################################################
diff --git a/transformers/modeling_tf_t5.py b/transformers/modeling_tf_t5.py
index 0ae7fff412..1336a1c30d 100644
--- a/transformers/modeling_tf_t5.py
+++ b/transformers/modeling_tf_t5.py
@@ -34,8 +34,8 @@ TF_T5_PRETRAINED_MODEL_ARCHIVE_MAP = {
     't5-small': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-small-tf_model.h5",
     't5-base': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-base-tf_model.h5",
     't5-large': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-large-tf_model.h5",
-    't5-3B': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-3B-tf_model.h5",
-    't5-11B': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-11B-tf_model.h5",
+    't5-3b': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-3b-tf_model.h5",
+    't5-11b': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-11b-tf_model.h5",
 }
 
 ####################################################
diff --git a/transformers/tokenization_t5.py b/transformers/tokenization_t5.py
index 62e9c069e2..9fd37b67c0 100644
--- a/transformers/tokenization_t5.py
+++ b/transformers/tokenization_t5.py
@@ -44,8 +44,8 @@ PRETRAINED_VOCAB_FILES_MAP = {
         't5-small': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-spiece.model",
         't5-base': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-spiece.model",
         't5-large': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-spiece.model",
-        't5-3B': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-spiece.model",
-        't5-11B': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-spiece.model",
+        't5-3b': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-spiece.model",
+        't5-11b': "https://s3.amazonaws.com/models.huggingface.co/bert/t5-spiece.model",
     }
 }
 
@@ -56,8 +56,8 @@ PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
     't5-small': 512,
     't5-base': 512,
     't5-large': 512,
-    't5-3B': 512,
-    't5-11B': 512,
+    't5-3b': 512,
+    't5-11b': 512,
 }
 
 class T5Tokenizer(PreTrainedTokenizer):

From c8ed1c82c8a42ef700d4129d227fa356385c1d60 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Fri, 13 Dec 2019 12:13:48 -0500
Subject: [PATCH 355/505] [SQUAD] Load checkpoint when evaluating without
 training

---
 examples/run_squad.py | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

diff --git a/examples/run_squad.py b/examples/run_squad.py
index 117b86e32c..a39915ee8b 100644
--- a/examples/run_squad.py
+++ b/examples/run_squad.py
@@ -580,10 +580,16 @@ def main():
     # Evaluation - we can ask to evaluate all the checkpoints (sub-directories) in a directory
     results = {}
     if args.do_eval and args.local_rank in [-1, 0]:
-        checkpoints = [args.output_dir]
-        if args.eval_all_checkpoints:
-            checkpoints = list(os.path.dirname(c) for c in sorted(glob.glob(args.output_dir + '/**/' + WEIGHTS_NAME, recursive=True)))
-            logging.getLogger("transformers.modeling_utils").setLevel(logging.WARN)  # Reduce model loading logs
+
+        if args.do_train:
+            logger.info("Loading checkpoints saved during training for evaluation")
+            checkpoints = [args.output_dir]
+            if args.eval_all_checkpoints:
+                checkpoints = list(os.path.dirname(c) for c in sorted(glob.glob(args.output_dir + '/**/' + WEIGHTS_NAME, recursive=True)))
+                logging.getLogger("transformers.modeling_utils").setLevel(logging.WARN)  # Reduce model loading logs
+        else:
+            logger.info("Loading checkpoint %s for evaluation", args.model_name_or_path)
+            checkpoints = [args.model_name_or_path]
 
         logger.info("Evaluate the following checkpoints: %s", checkpoints)
 

From f24a228a9315a4b723509bc9144b53d2bcbc4217 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Fri, 13 Dec 2019 14:50:35 -0500
Subject: [PATCH 356/505] Speed up tokenization process

---
 transformers/data/processors/squad.py |  2 +-
 transformers/tokenization_utils.py    | 10 ++++++----
 2 files changed, 7 insertions(+), 5 deletions(-)

diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index 9bc4375684..e193f6153e 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -116,7 +116,7 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
     unique_id = 1000000000
 
     features = []
-    for (example_index, example) in enumerate(tqdm(examples)):
+    for (example_index, example) in enumerate(tqdm(examples, desc="Converting examples to features")):
         if is_training and not example.is_impossible:
             # Get start and end position
             start_position = example.start_position
diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index 317ecd167b..e87c87787b 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -637,9 +637,11 @@ class PreTrainedTokenizer(object):
             text: The sequence to be encoded.
             **kwargs: passed to the child `self.tokenize()` method
         """
+        all_special_tokens = self.all_special_tokens
+
         def lowercase_text(t):
             # convert non-special tokens to lowercase
-            escaped_special_toks = [re.escape(s_tok) for s_tok in self.all_special_tokens]
+            escaped_special_toks = [re.escape(s_tok) for s_tok in all_special_tokens]
             pattern = r'(^' + r'|'.join(escaped_special_toks) + r')|' + \
                       r'(.+?)'
             return re.sub(
@@ -680,17 +682,17 @@ class PreTrainedTokenizer(object):
                 tokenized_text = []
                 for sub_text in text_list:
                     if sub_text not in self.added_tokens_encoder \
-                            and sub_text not in self.all_special_tokens:
+                            and sub_text not in all_special_tokens:
                         tokenized_text += split_on_token(tok, sub_text)
                     else:
                         tokenized_text += [sub_text]
                 text_list = tokenized_text
 
             return list(itertools.chain.from_iterable((self._tokenize(token, **kwargs) if token not \
-                    in self.added_tokens_encoder and token not in self.all_special_tokens \
+                    in self.added_tokens_encoder and token not in all_special_tokens \
                     else [token] for token in tokenized_text)))
 
-        added_tokens = list(self.added_tokens_encoder.keys()) + self.all_special_tokens
+        added_tokens = list(self.added_tokens_encoder.keys()) + all_special_tokens
         tokenized_text = split_on_tokens(added_tokens, text)
         return tokenized_text
 

From d46147294852694d1dc701c72b9053ff2e726265 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Fri, 13 Dec 2019 15:31:52 -0500
Subject: [PATCH 357/505] return for SQuAD [BLACKED]

---
 transformers/data/processors/glue.py  |   2 +-
 transformers/data/processors/squad.py | 280 ++++++++++++++++----------
 2 files changed, 172 insertions(+), 110 deletions(-)

diff --git a/transformers/data/processors/glue.py b/transformers/data/processors/glue.py
index 518251b050..11ebd949de 100644
--- a/transformers/data/processors/glue.py
+++ b/transformers/data/processors/glue.py
@@ -133,7 +133,7 @@ def glue_convert_examples_to_features(examples, tokenizer,
     if is_tf_available() and is_tf_dataset:
         def gen():
             for ex in features:
-                yield  ({'input_ids': ex.input_ids,
+                yield ({'input_ids': ex.input_ids,
                          'attention_mask': ex.attention_mask,
                          'token_type_ids': ex.token_type_ids},
                         ex.label)
diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index e193f6153e..84aa429e26 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -18,19 +18,20 @@ if is_tf_available():
 
 logger = logging.getLogger(__name__)
 
-def _improve_answer_span(doc_tokens, input_start, input_end, tokenizer,
-                        orig_answer_text):
+
+def _improve_answer_span(doc_tokens, input_start, input_end, tokenizer, orig_answer_text):
     """Returns tokenized answer spans that better match the annotated answer."""
     tok_answer_text = " ".join(tokenizer.tokenize(orig_answer_text))
 
     for new_start in range(input_start, input_end + 1):
         for new_end in range(input_end, new_start - 1, -1):
-            text_span = " ".join(doc_tokens[new_start:(new_end + 1)])
+            text_span = " ".join(doc_tokens[new_start : (new_end + 1)])
             if text_span == tok_answer_text:
                 return (new_start, new_end)
 
     return (input_start, input_end)
 
+
 def _check_is_max_context(doc_spans, cur_span_index, position):
     """Check if this is the 'max context' doc span for the token."""
     best_score = None
@@ -50,10 +51,11 @@ def _check_is_max_context(doc_spans, cur_span_index, position):
 
     return cur_span_index == best_span_index
 
+
 def _new_check_is_max_context(doc_spans, cur_span_index, position):
     """Check if this is the 'max context' doc span for the token."""
     # if len(doc_spans) == 1:
-        # return True
+    # return True
     best_score = None
     best_span_index = None
     for (span_index, doc_span) in enumerate(doc_spans):
@@ -71,14 +73,16 @@ def _new_check_is_max_context(doc_spans, cur_span_index, position):
 
     return cur_span_index == best_span_index
 
+
 def _is_whitespace(c):
     if c == " " or c == "\t" or c == "\r" or c == "\n" or ord(c) == 0x202F:
         return True
     return False
 
-def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
-                                       doc_stride, max_query_length, is_training, 
-                                       return_dataset=False):
+
+def squad_convert_examples_to_features(
+    examples, tokenizer, max_seq_length, doc_stride, max_query_length, is_training, return_dataset=False
+):
     """
     Converts a list of examples into a list of features that can be directly given as input to a model.
     It is model-dependant and takes advantage of many of the tokenizer's features to create the model's inputs.
@@ -112,7 +116,7 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
         )
     """
 
-    # Defining helper methods    
+    # Defining helper methods
     unique_id = 1000000000
 
     features = []
@@ -123,13 +127,12 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
             end_position = example.end_position
 
             # If the answer cannot be found in the text, then skip this example.
-            actual_text = " ".join(example.doc_tokens[start_position:(end_position + 1)])
+            actual_text = " ".join(example.doc_tokens[start_position : (end_position + 1)])
             cleaned_answer_text = " ".join(whitespace_tokenize(example.answer_text))
             if actual_text.find(cleaned_answer_text) == -1:
                 logger.warning("Could not find answer: '%s' vs. '%s'", actual_text, cleaned_answer_text)
                 continue
 
-
         tok_to_orig_index = []
         orig_to_tok_index = []
         all_doc_tokens = []
@@ -140,7 +143,6 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
                 tok_to_orig_index.append(i)
                 all_doc_tokens.append(sub_token)
 
-
         if is_training and not example.is_impossible:
             tok_start_position = orig_to_tok_index[example.start_position]
             if example.end_position < len(example.doc_tokens) - 1:
@@ -153,36 +155,41 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
             )
 
         spans = []
-        
-        truncated_query = tokenizer.encode(example.question_text, add_special_tokens=False, max_length=max_query_length)
-        sequence_added_tokens = tokenizer.max_len - tokenizer.max_len_single_sentence 
-        sequence_pair_added_tokens = tokenizer.max_len - tokenizer.max_len_sentences_pair 
+
+        truncated_query = tokenizer.encode(
+            example.question_text, add_special_tokens=False, max_length=max_query_length
+        )
+        sequence_added_tokens = tokenizer.max_len - tokenizer.max_len_single_sentence
+        sequence_pair_added_tokens = tokenizer.max_len - tokenizer.max_len_sentences_pair
 
         span_doc_tokens = all_doc_tokens
         while len(spans) * doc_stride < len(all_doc_tokens):
-            
+
             encoded_dict = tokenizer.encode_plus(
-                truncated_query if tokenizer.padding_side == "right" else span_doc_tokens, 
-                span_doc_tokens if tokenizer.padding_side == "right" else truncated_query, 
-                max_length=max_seq_length, 
-                return_overflowing_tokens=True, 
+                truncated_query if tokenizer.padding_side == "right" else span_doc_tokens,
+                span_doc_tokens if tokenizer.padding_side == "right" else truncated_query,
+                max_length=max_seq_length,
+                return_overflowing_tokens=True,
                 pad_to_max_length=True,
                 stride=max_seq_length - doc_stride - len(truncated_query) - sequence_pair_added_tokens,
-                truncation_strategy='only_second' if tokenizer.padding_side == "right" else 'only_first'
+                truncation_strategy="only_second" if tokenizer.padding_side == "right" else "only_first",
             )
 
-            paragraph_len = min(len(all_doc_tokens) - len(spans) * doc_stride, max_seq_length - len(truncated_query) - sequence_pair_added_tokens)
+            paragraph_len = min(
+                len(all_doc_tokens) - len(spans) * doc_stride,
+                max_seq_length - len(truncated_query) - sequence_pair_added_tokens,
+            )
 
-            if tokenizer.pad_token_id in encoded_dict['input_ids']: 
-                non_padded_ids = encoded_dict['input_ids'][:encoded_dict['input_ids'].index(tokenizer.pad_token_id)]
+            if tokenizer.pad_token_id in encoded_dict["input_ids"]:
+                non_padded_ids = encoded_dict["input_ids"][: encoded_dict["input_ids"].index(tokenizer.pad_token_id)]
             else:
-                non_padded_ids = encoded_dict['input_ids']
+                non_padded_ids = encoded_dict["input_ids"]
 
             tokens = tokenizer.convert_ids_to_tokens(non_padded_ids)
 
             token_to_orig_map = {}
             for i in range(paragraph_len):
-                index = len(truncated_query) + sequence_added_tokens + i if tokenizer.padding_side == "right" else i 
+                index = len(truncated_query) + sequence_added_tokens + i if tokenizer.padding_side == "right" else i
                 token_to_orig_map[index] = tok_to_orig_index[len(spans) * doc_stride + i]
 
             encoded_dict["paragraph_len"] = paragraph_len
@@ -202,16 +209,20 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
         for doc_span_index in range(len(spans)):
             for j in range(spans[doc_span_index]["paragraph_len"]):
                 is_max_context = _new_check_is_max_context(spans, doc_span_index, doc_span_index * doc_stride + j)
-                index = j if tokenizer.padding_side == "left" else spans[doc_span_index]["truncated_query_with_special_tokens_length"] + j
+                index = (
+                    j
+                    if tokenizer.padding_side == "left"
+                    else spans[doc_span_index]["truncated_query_with_special_tokens_length"] + j
+                )
                 spans[doc_span_index]["token_is_max_context"][index] = is_max_context
 
         for span in spans:
             # Identify the position of the CLS token
-            cls_index = span['input_ids'].index(tokenizer.cls_token_id)
+            cls_index = span["input_ids"].index(tokenizer.cls_token_id)
 
             # p_mask: mask with 1 for token than cannot be in the answer (0 for token which can be in an answer)
             # Original TF implem also keep the classification token (set to 0) (not sure why...)
-            p_mask = np.array(span['token_type_ids'])
+            p_mask = np.array(span["token_type_ids"])
 
             p_mask = np.minimum(p_mask, 1)
 
@@ -224,7 +235,6 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
             # Set the CLS index to '0'
             p_mask[cls_index] = 0
 
-
             span_is_impossible = example.is_impossible
             start_position = 0
             end_position = 0
@@ -247,55 +257,99 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
                         doc_offset = 0
                     else:
                         doc_offset = len(truncated_query) + sequence_added_tokens
-                        
+
                     start_position = tok_start_position - doc_start + doc_offset
                     end_position = tok_end_position - doc_start + doc_offset
 
-
-            features.append(SquadFeatures(
-                span['input_ids'],
-                span['attention_mask'],
-                span['token_type_ids'],
-                cls_index,
-                p_mask.tolist(),
-
-                example_index=example_index,
-                unique_id=unique_id,
-                paragraph_len=span['paragraph_len'],
-                token_is_max_context=span["token_is_max_context"],
-                tokens=span["tokens"],
-                token_to_orig_map=span["token_to_orig_map"],
-                
-                start_position=start_position,
-                end_position=end_position
-            ))
+            features.append(
+                SquadFeatures(
+                    span["input_ids"],
+                    span["attention_mask"],
+                    span["token_type_ids"],
+                    cls_index,
+                    p_mask.tolist(),
+                    example_index=example_index,
+                    unique_id=unique_id,
+                    paragraph_len=span["paragraph_len"],
+                    token_is_max_context=span["token_is_max_context"],
+                    tokens=span["tokens"],
+                    token_to_orig_map=span["token_to_orig_map"],
+                    start_position=start_position,
+                    end_position=end_position,
+                )
+            )
 
             unique_id += 1
 
-    if return_dataset == 'pt':
+    if return_dataset == "pt":
         if not is_torch_available():
             raise ImportError("Pytorch must be installed to return a pytorch dataset.")
 
         # Convert to Tensors and build dataset
         all_input_ids = torch.tensor([f.input_ids for f in features], dtype=torch.long)
-        all_input_mask = torch.tensor([f.attention_mask for f in features], dtype=torch.long)
-        all_segment_ids = torch.tensor([f.token_type_ids for f in features], dtype=torch.long)
+        all_attention_masks = torch.tensor([f.attention_mask for f in features], dtype=torch.long)
+        all_token_type_ids = torch.tensor([f.token_type_ids for f in features], dtype=torch.long)
         all_cls_index = torch.tensor([f.cls_index for f in features], dtype=torch.long)
         all_p_mask = torch.tensor([f.p_mask for f in features], dtype=torch.float)
 
         if not is_training:
             all_example_index = torch.arange(all_input_ids.size(0), dtype=torch.long)
-            dataset = TensorDataset(all_input_ids, all_input_mask, all_segment_ids,
-                                    all_example_index, all_cls_index, all_p_mask)
+            dataset = TensorDataset(
+                all_input_ids, all_attention_masks, all_token_type_ids, all_example_index, all_cls_index, all_p_mask
+            )
         else:
             all_start_positions = torch.tensor([f.start_position for f in features], dtype=torch.long)
             all_end_positions = torch.tensor([f.end_position for f in features], dtype=torch.long)
-            dataset = TensorDataset(all_input_ids, all_input_mask, all_segment_ids,
-                                    all_start_positions, all_end_positions,
-                                    all_cls_index, all_p_mask)
+            dataset = TensorDataset(
+                all_input_ids,
+                all_attention_masks,
+                all_token_type_ids,
+                all_start_positions,
+                all_end_positions,
+                all_cls_index,
+                all_p_mask,
+            )
 
         return features, dataset
-        
+    elif return_dataset == "tf":
+        if not is_tf_available():
+            raise ImportError("TensorFlow must be installed to return a TensorFlow dataset.")
+
+        def gen():
+            for ex in features:
+                yield (
+                    {
+                        "input_ids": ex.input_ids,
+                        "attention_mask": ex.attention_mask,
+                        "token_type_ids": ex.token_type_ids,
+                    }, {
+                        "start_position": ex.start_position,
+                        "end_position": ex.end_position,
+                        "cls_index": ex.cls_index,
+                        "p_mask": ex.p_mask,
+                    }
+                )
+
+        return tf.data.Dataset.from_generator(
+            gen,
+            (
+                {"input_ids": tf.int32, "attention_mask": tf.int32, "token_type_ids": tf.int32},
+                {"start_position": tf.int64, "end_position": tf.int64, "cls_index": tf.int64, "p_mask": tf.int32},
+            ),
+            (
+                {
+                    "input_ids": tf.TensorShape([None]),
+                    "attention_mask": tf.TensorShape([None]),
+                    "token_type_ids": tf.TensorShape([None]),
+                },
+                {
+                    "start_position": tf.TensorShape([]),
+                    "end_position": tf.TensorShape([]),
+                    "cls_index": tf.TensorShape([]),
+                    "p_mask": tf.TensorShape([None]),
+                },
+            ),
+        )
 
     return features
 
@@ -305,31 +359,32 @@ class SquadProcessor(DataProcessor):
     Processor for the SQuAD data set.
     Overriden by SquadV1Processor and SquadV2Processor, used by the version 1.1 and version 2.0 of SQuAD, respectively.
     """
+
     train_file = None
     dev_file = None
 
     def _get_example_from_tensor_dict(self, tensor_dict, evaluate=False):
         if not evaluate:
-            answer = tensor_dict['answers']['text'][0].numpy().decode('utf-8')
-            answer_start = tensor_dict['answers']['answer_start'][0].numpy()
+            answer = tensor_dict["answers"]["text"][0].numpy().decode("utf-8")
+            answer_start = tensor_dict["answers"]["answer_start"][0].numpy()
             answers = []
         else:
-            answers = [{
-                "answer_start": start.numpy(), 
-                "text": text.numpy().decode('utf-8')
-            } for start, text in zip(tensor_dict['answers']["answer_start"], tensor_dict['answers']["text"])]
+            answers = [
+                {"answer_start": start.numpy(), "text": text.numpy().decode("utf-8")}
+                for start, text in zip(tensor_dict["answers"]["answer_start"], tensor_dict["answers"]["text"])
+            ]
 
             answer = None
             answer_start = None
 
         return SquadExample(
-            qas_id=tensor_dict['id'].numpy().decode("utf-8"),
-            question_text=tensor_dict['question'].numpy().decode('utf-8'),
-            context_text=tensor_dict['context'].numpy().decode('utf-8'),
+            qas_id=tensor_dict["id"].numpy().decode("utf-8"),
+            question_text=tensor_dict["question"].numpy().decode("utf-8"),
+            context_text=tensor_dict["context"].numpy().decode("utf-8"),
             answer_text=answer,
             start_position_character=answer_start,
-            title=tensor_dict['title'].numpy().decode('utf-8'),
-            answers=answers
+            title=tensor_dict["title"].numpy().decode("utf-8"),
+            answers=answers,
         )
 
     def get_examples_from_dataset(self, dataset, evaluate=False):
@@ -359,7 +414,7 @@ class SquadProcessor(DataProcessor):
 
         examples = []
         for tensor_dict in tqdm(dataset):
-            examples.append(self._get_example_from_tensor_dict(tensor_dict, evaluate=evaluate)) 
+            examples.append(self._get_example_from_tensor_dict(tensor_dict, evaluate=evaluate))
 
         return examples
 
@@ -379,7 +434,9 @@ class SquadProcessor(DataProcessor):
         if self.train_file is None:
             raise ValueError("SquadProcessor should be instantiated via SquadV1Processor or SquadV2Processor")
 
-        with open(os.path.join(data_dir, self.train_file if filename is None else filename), "r", encoding='utf-8') as reader:
+        with open(
+            os.path.join(data_dir, self.train_file if filename is None else filename), "r", encoding="utf-8"
+        ) as reader:
             input_data = json.load(reader)["data"]
         return self._create_examples(input_data, "train")
 
@@ -397,8 +454,10 @@ class SquadProcessor(DataProcessor):
 
         if self.dev_file is None:
             raise ValueError("SquadProcessor should be instantiated via SquadV1Processor or SquadV2Processor")
-        
-        with open(os.path.join(data_dir, self.dev_file if filename is None else filename), "r", encoding='utf-8') as reader:
+
+        with open(
+            os.path.join(data_dir, self.dev_file if filename is None else filename), "r", encoding="utf-8"
+        ) as reader:
             input_data = json.load(reader)["data"]
         return self._create_examples(input_data, "dev")
 
@@ -406,7 +465,7 @@ class SquadProcessor(DataProcessor):
         is_training = set_type == "train"
         examples = []
         for entry in tqdm(input_data):
-            title = entry['title']
+            title = entry["title"]
             for paragraph in entry["paragraphs"]:
                 context_text = paragraph["context"]
                 for qa in paragraph["qas"]:
@@ -415,7 +474,7 @@ class SquadProcessor(DataProcessor):
                     start_position_character = None
                     answer_text = None
                     answers = []
-                    
+
                     if "is_impossible" in qa:
                         is_impossible = qa["is_impossible"]
                     else:
@@ -424,8 +483,8 @@ class SquadProcessor(DataProcessor):
                     if not is_impossible:
                         if is_training:
                             answer = qa["answers"][0]
-                            answer_text = answer['text']
-                            start_position_character = answer['answer_start']
+                            answer_text = answer["text"]
+                            start_position_character = answer["answer_start"]
                         else:
                             answers = qa["answers"]
 
@@ -437,12 +496,13 @@ class SquadProcessor(DataProcessor):
                         start_position_character=start_position_character,
                         title=title,
                         is_impossible=is_impossible,
-                        answers=answers
+                        answers=answers,
                     )
 
                     examples.append(example)
         return examples
 
+
 class SquadV1Processor(SquadProcessor):
     train_file = "train-v1.1.json"
     dev_file = "dev-v1.1.json"
@@ -451,7 +511,7 @@ class SquadV1Processor(SquadProcessor):
 class SquadV2Processor(SquadProcessor):
     train_file = "train-v2.0.json"
     dev_file = "dev-v2.0.json"
-    
+
 
 class SquadExample(object):
     """
@@ -468,21 +528,23 @@ class SquadExample(object):
         is_impossible: False by default, set to True if the example has no possible answer.
     """
 
-    def __init__(self,
-                 qas_id,
-                 question_text,
-                 context_text,
-                 answer_text,
-                 start_position_character,
-                 title,
-                 answers=[],
-                 is_impossible=False):
+    def __init__(
+        self,
+        qas_id,
+        question_text,
+        context_text,
+        answer_text,
+        start_position_character,
+        title,
+        answers=[],
+        is_impossible=False,
+    ):
         self.qas_id = qas_id
         self.question_text = question_text
         self.context_text = context_text
         self.answer_text = answer_text
         self.title = title
-        self.is_impossible = is_impossible 
+        self.is_impossible = is_impossible
         self.answers = answers
 
         self.start_position, self.end_position = 0, 0
@@ -537,24 +599,23 @@ class SquadFeatures(object):
         end_position: end of the answer token index 
     """
 
-    def __init__(self,
-                 input_ids,
-                 attention_mask,
-                 token_type_ids,
-                 cls_index,
-                 p_mask,
-                 
-                 example_index,
-                 unique_id,
-                 paragraph_len,
-                 token_is_max_context,
-                 tokens,
-                 token_to_orig_map,
-
-                 start_position,
-                 end_position
-        ):
-        self.input_ids = input_ids 
+    def __init__(
+        self,
+        input_ids,
+        attention_mask,
+        token_type_ids,
+        cls_index,
+        p_mask,
+        example_index,
+        unique_id,
+        paragraph_len,
+        token_is_max_context,
+        tokens,
+        token_to_orig_map,
+        start_position,
+        end_position,
+    ):
+        self.input_ids = input_ids
         self.attention_mask = attention_mask
         self.token_type_ids = token_type_ids
         self.cls_index = cls_index
@@ -580,12 +641,13 @@ class SquadResult(object):
         start_logits: The logits corresponding to the start of the answer
         end_logits: The logits corresponding to the end of the answer
     """
+
     def __init__(self, unique_id, start_logits, end_logits, start_top_index=None, end_top_index=None, cls_logits=None):
         self.start_logits = start_logits
         self.end_logits = end_logits
         self.unique_id = unique_id
-        
+
         if start_top_index:
             self.start_top_index = start_top_index
             self.end_top_index = end_top_index
-            self.cls_logits = cls_logits
\ No newline at end of file
+            self.cls_logits = cls_logits

From 866d73ca26a13d7e378b2f88f365cb0807c47805 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Fri, 13 Dec 2019 16:09:23 -0500
Subject: [PATCH 358/505] [cli] Upload is now compatible with folders

---
 transformers/commands/user.py | 57 ++++++++++++++++++++++++++---------
 1 file changed, 43 insertions(+), 14 deletions(-)

diff --git a/transformers/commands/user.py b/transformers/commands/user.py
index d79922ed8a..8e0e563422 100644
--- a/transformers/commands/user.py
+++ b/transformers/commands/user.py
@@ -19,8 +19,8 @@ class UserCommands(BaseTransformersCLICommand):
         list_parser.set_defaults(func=lambda args: ListObjsCommand(args))
         # upload
         upload_parser = parser.add_parser('upload')
-        upload_parser.add_argument('file', type=str, help='Local filepath of the file to upload.')
-        upload_parser.add_argument('--filename', type=str, default=None, help='Optional: override object filename on S3.')
+        upload_parser.add_argument('path', type=str, help='Local path of the folder or individual file to upload.')
+        upload_parser.add_argument('--filename', type=str, default=None, help='Optional: override individual object filename on S3.')
         upload_parser.set_defaults(func=lambda args: UploadCommand(args))
 
 
@@ -138,28 +138,57 @@ class ListObjsCommand(BaseUserCommand):
 
 
 class UploadCommand(BaseUserCommand):
+    def walk_dir(self, rel_path):
+        """
+        Recursively list all files in a folder.
+        """
+        entries: List[os.DirEntry] = list(os.scandir(rel_path))
+        files = [
+            (
+                os.path.join(os.getcwd(), f.path),  # filepath
+                f.path  # filename
+            )
+            for f in entries if f.is_file()
+        ]
+        for f in entries:
+            if f.is_dir():
+                files += self.walk_dir(f.path)
+        return files
+
     def run(self):
         token = HfFolder.get_token()
         if token is None:
             print("Not logged in")
             exit(1)
-        filepath = os.path.join(os.getcwd(), self.args.file)
-        filename = self.args.filename if self.args.filename is not None else os.path.basename(filepath)
-        print(
-            "About to upload file {} to S3 under filename {}".format(
-                ANSI.bold(filepath), ANSI.bold(filename)
+        local_path = os.path.abspath(self.args.path)
+        if os.path.isdir(local_path):
+            if self.args.filename is not None:
+                raise ValueError("Cannot specify a filename override when uploading a folder.")
+            rel_path = os.path.basename(local_path)
+            files = self.walk_dir(rel_path)
+        elif os.path.isfile(local_path):
+            filename = self.args.filename if self.args.filename is not None else os.path.basename(local_path)
+            files = [(local_path, filename)]
+        else:
+            raise ValueError("Not a valid file or directory: {}".format(local_path))
+
+        for filepath, filename in files:
+            print(
+                "About to upload file {} to S3 under filename {}".format(
+                    ANSI.bold(filepath), ANSI.bold(filename)
+                )
             )
-        )
 
         choice = input("Proceed? [Y/n] ").lower()
         if not(choice == "" or choice == "y" or choice == "yes"):
             print("Abort")
             exit()
         print(
-            ANSI.bold("Uploading... This might take a while if file is large")
+            ANSI.bold("Uploading... This might take a while if files are large")
         )
-        access_url = self._api.presign_and_upload(
-            token=token, filename=filename, filepath=filepath
-        )
-        print("Your file now lives at:")
-        print(access_url)
+        for filepath, filename in files:
+            access_url = self._api.presign_and_upload(
+                token=token, filename=filename, filepath=filepath
+            )
+            print("Your file now lives at:")
+            print(access_url)

From 5b7b78e088352a3aaf1f80d26bb1cd466bc2ac64 Mon Sep 17 00:00:00 2001
From: Pascal Voitot <p.voitot@samsung.com>
Date: Sun, 8 Dec 2019 23:22:02 +0100
Subject: [PATCH 359/505] :bug: #2096 in tokenizer.decode, adds a space after
 special tokens to return right formatted string

---
 transformers/tokenization_utils.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index e87c87787b..42519c26ba 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -1180,7 +1180,7 @@ class PreTrainedTokenizer(object):
                 if current_sub_text:
                     sub_texts.append(self.convert_tokens_to_string(current_sub_text))
                     current_sub_text = []
-                sub_texts.append(" " + token)
+                sub_texts.append(" " + token + " ")
             else:
                 current_sub_text.append(token)
         if current_sub_text:

From df160af736cba1d50c09abcf92c8fc6c00bcdb13 Mon Sep 17 00:00:00 2001
From: Pascal Voitot <p.voitot@samsung.com>
Date: Tue, 10 Dec 2019 00:03:38 +0100
Subject: [PATCH 360/505] :bug: #2096 in tokenizer.decode, space is not joined
 between all subtexts instead of before added tokens

---
 transformers/tests/tokenization_bert_test.py | 16 ++++++++++++++++
 transformers/tokenization_utils.py           |  4 ++--
 2 files changed, 18 insertions(+), 2 deletions(-)

diff --git a/transformers/tests/tokenization_bert_test.py b/transformers/tests/tokenization_bert_test.py
index f390248956..c47f149e9a 100644
--- a/transformers/tests/tokenization_bert_test.py
+++ b/transformers/tests/tokenization_bert_test.py
@@ -99,6 +99,21 @@ class BertTokenizationTest(CommonTestCases.CommonTokenizerTester):
         self.assertListEqual(
             tokenizer.tokenize("unwantedX running"), ["[UNK]", "runn", "##ing"])
 
+    def test_encode_decode_with_spaces(self):
+        tokenizer = self.get_tokenizer()
+
+        new_toks = ['[ABC]', '[DEF]', 'GHI IHG']
+        tokenizer.add_tokens(new_toks)
+        input = "unwanted running [ABC] [DEF] running unwanted [ABC] GHI IHG unwanted [DEF]"
+        encoded = tokenizer.encode(input)
+        decoded = tokenizer.decode(encoded)
+        self.assertEqual(
+            decoded.lower(),
+            (f"[CLS] {input.lower()} [SEP]").lower()
+        )
+
+
+
     def test_is_whitespace(self):
         self.assertTrue(_is_whitespace(u" "))
         self.assertTrue(_is_whitespace(u"\t"))
@@ -139,5 +154,6 @@ class BertTokenizationTest(CommonTestCases.CommonTokenizerTester):
         assert encoded_sentence == [101] + text + [102]
         assert encoded_pair == [101] + text + [102] + text_2 + [102]
 
+
 if __name__ == '__main__':
     unittest.main()
diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index 42519c26ba..8aef80fec8 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -1180,12 +1180,12 @@ class PreTrainedTokenizer(object):
                 if current_sub_text:
                     sub_texts.append(self.convert_tokens_to_string(current_sub_text))
                     current_sub_text = []
-                sub_texts.append(" " + token + " ")
+                sub_texts.append(token)
             else:
                 current_sub_text.append(token)
         if current_sub_text:
             sub_texts.append(self.convert_tokens_to_string(current_sub_text))
-        text = ''.join(sub_texts)
+        text = ' '.join(sub_texts)
 
         if clean_up_tokenization_spaces:
             clean_text = self.clean_up_tokenization(text)

From dd2add9f6efdaa248f3074b865dc67c439b30a4d Mon Sep 17 00:00:00 2001
From: Pascal Voitot <p.voitot@samsung.com>
Date: Tue, 10 Dec 2019 00:29:44 +0100
Subject: [PATCH 361/505] more tests

---
 transformers/tests/tokenization_bert_test.py |  2 +-
 transformers/tests/tokenization_gpt2_test.py | 14 ++++++++++++++
 2 files changed, 15 insertions(+), 1 deletion(-)

diff --git a/transformers/tests/tokenization_bert_test.py b/transformers/tests/tokenization_bert_test.py
index c47f149e9a..b93934dd67 100644
--- a/transformers/tests/tokenization_bert_test.py
+++ b/transformers/tests/tokenization_bert_test.py
@@ -109,7 +109,7 @@ class BertTokenizationTest(CommonTestCases.CommonTokenizerTester):
         decoded = tokenizer.decode(encoded)
         self.assertEqual(
             decoded.lower(),
-            (f"[CLS] {input.lower()} [SEP]").lower()
+            (f"[CLS] {input} [SEP]").lower()
         )
 
 
diff --git a/transformers/tests/tokenization_gpt2_test.py b/transformers/tests/tokenization_gpt2_test.py
index a77cc75ec2..9e6ca3c4fd 100644
--- a/transformers/tests/tokenization_gpt2_test.py
+++ b/transformers/tests/tokenization_gpt2_test.py
@@ -67,6 +67,20 @@ class GPT2TokenizationTest(CommonTestCases.CommonTokenizerTester):
         self.assertListEqual(
             tokenizer.convert_tokens_to_ids(input_tokens), input_bpe_tokens)
 
+    def test_encode_decode_with_spaces(self):
+        tokenizer = self.get_tokenizer()
+
+        new_toks = ['[ABC]', '[DEF]', 'GHI IHG']
+        tokenizer.add_tokens(new_toks)
+        input = "lower newer [ABC] [DEF] newer lower [ABC] GHI IHG newer lower[DEF]"
+        encoded = tokenizer.encode(input)
+        decoded = tokenizer.decode(encoded)
+        self.assertEqual(
+            decoded.lower(),
+            input.lower()
+        )
+
+
 
 if __name__ == '__main__':
     unittest.main()

From 4cbdc7d910a0a12871a8e29760a3a6721a138421 Mon Sep 17 00:00:00 2001
From: Pascal Voitot <p.voitot@samsung.com>
Date: Tue, 10 Dec 2019 09:37:15 +0100
Subject: [PATCH 362/505] missed space

---
 transformers/tests/tokenization_bert_test.py | 2 --
 transformers/tests/tokenization_gpt2_test.py | 2 +-
 2 files changed, 1 insertion(+), 3 deletions(-)

diff --git a/transformers/tests/tokenization_bert_test.py b/transformers/tests/tokenization_bert_test.py
index b93934dd67..a039a24dd8 100644
--- a/transformers/tests/tokenization_bert_test.py
+++ b/transformers/tests/tokenization_bert_test.py
@@ -112,8 +112,6 @@ class BertTokenizationTest(CommonTestCases.CommonTokenizerTester):
             (f"[CLS] {input} [SEP]").lower()
         )
 
-
-
     def test_is_whitespace(self):
         self.assertTrue(_is_whitespace(u" "))
         self.assertTrue(_is_whitespace(u"\t"))
diff --git a/transformers/tests/tokenization_gpt2_test.py b/transformers/tests/tokenization_gpt2_test.py
index 9e6ca3c4fd..1b4fe42874 100644
--- a/transformers/tests/tokenization_gpt2_test.py
+++ b/transformers/tests/tokenization_gpt2_test.py
@@ -72,7 +72,7 @@ class GPT2TokenizationTest(CommonTestCases.CommonTokenizerTester):
 
         new_toks = ['[ABC]', '[DEF]', 'GHI IHG']
         tokenizer.add_tokens(new_toks)
-        input = "lower newer [ABC] [DEF] newer lower [ABC] GHI IHG newer lower[DEF]"
+        input = "lower newer [ABC] [DEF] newer lower [ABC] GHI IHG newer lower [DEF]"
         encoded = tokenizer.encode(input)
         decoded = tokenizer.decode(encoded)
         self.assertEqual(

From f2ac50cb5560e13d941f1ea3dec3399f12f7a3fb Mon Sep 17 00:00:00 2001
From: Pascal Voitot <p.voitot@samsung.com>
Date: Tue, 10 Dec 2019 09:58:06 +0100
Subject: [PATCH 363/505] better for python2.x

---
 transformers/tests/tokenization_bert_test.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/tests/tokenization_bert_test.py b/transformers/tests/tokenization_bert_test.py
index a039a24dd8..77b124cdf2 100644
--- a/transformers/tests/tokenization_bert_test.py
+++ b/transformers/tests/tokenization_bert_test.py
@@ -109,7 +109,7 @@ class BertTokenizationTest(CommonTestCases.CommonTokenizerTester):
         decoded = tokenizer.decode(encoded)
         self.assertEqual(
             decoded.lower(),
-            (f"[CLS] {input} [SEP]").lower()
+            ("[CLS] " + input + " [SEP]").lower()
         )
 
     def test_is_whitespace(self):

From c3248cf122014dce10c0c8d0e663a95c948493e3 Mon Sep 17 00:00:00 2001
From: LysandreJik <lysandre.debut@reseau.eseo.fr>
Date: Wed, 11 Dec 2019 12:36:37 -0500
Subject: [PATCH 364/505] Tests for all tokenizers

---
 transformers/tests/tokenization_bert_test.py     | 13 -------------
 transformers/tests/tokenization_gpt2_test.py     | 15 ---------------
 transformers/tests/tokenization_tests_commons.py |  9 +++++++++
 3 files changed, 9 insertions(+), 28 deletions(-)

diff --git a/transformers/tests/tokenization_bert_test.py b/transformers/tests/tokenization_bert_test.py
index 77b124cdf2..c503ea5e1e 100644
--- a/transformers/tests/tokenization_bert_test.py
+++ b/transformers/tests/tokenization_bert_test.py
@@ -99,19 +99,6 @@ class BertTokenizationTest(CommonTestCases.CommonTokenizerTester):
         self.assertListEqual(
             tokenizer.tokenize("unwantedX running"), ["[UNK]", "runn", "##ing"])
 
-    def test_encode_decode_with_spaces(self):
-        tokenizer = self.get_tokenizer()
-
-        new_toks = ['[ABC]', '[DEF]', 'GHI IHG']
-        tokenizer.add_tokens(new_toks)
-        input = "unwanted running [ABC] [DEF] running unwanted [ABC] GHI IHG unwanted [DEF]"
-        encoded = tokenizer.encode(input)
-        decoded = tokenizer.decode(encoded)
-        self.assertEqual(
-            decoded.lower(),
-            ("[CLS] " + input + " [SEP]").lower()
-        )
-
     def test_is_whitespace(self):
         self.assertTrue(_is_whitespace(u" "))
         self.assertTrue(_is_whitespace(u"\t"))
diff --git a/transformers/tests/tokenization_gpt2_test.py b/transformers/tests/tokenization_gpt2_test.py
index 1b4fe42874..5eae767bdf 100644
--- a/transformers/tests/tokenization_gpt2_test.py
+++ b/transformers/tests/tokenization_gpt2_test.py
@@ -67,20 +67,5 @@ class GPT2TokenizationTest(CommonTestCases.CommonTokenizerTester):
         self.assertListEqual(
             tokenizer.convert_tokens_to_ids(input_tokens), input_bpe_tokens)
 
-    def test_encode_decode_with_spaces(self):
-        tokenizer = self.get_tokenizer()
-
-        new_toks = ['[ABC]', '[DEF]', 'GHI IHG']
-        tokenizer.add_tokens(new_toks)
-        input = "lower newer [ABC] [DEF] newer lower [ABC] GHI IHG newer lower [DEF]"
-        encoded = tokenizer.encode(input)
-        decoded = tokenizer.decode(encoded)
-        self.assertEqual(
-            decoded.lower(),
-            input.lower()
-        )
-
-
-
 if __name__ == '__main__':
     unittest.main()
diff --git a/transformers/tests/tokenization_tests_commons.py b/transformers/tests/tokenization_tests_commons.py
index c009958135..13e7ae746a 100644
--- a/transformers/tests/tokenization_tests_commons.py
+++ b/transformers/tests/tokenization_tests_commons.py
@@ -232,6 +232,15 @@ class CommonTestCases:
             self.assertNotEqual(len(tokens_2), 0)
             self.assertIsInstance(text_2, (str, unicode))
 
+        def test_encode_decode_with_spaces(self):
+            tokenizer = self.get_tokenizer()
+
+            new_toks = ['[ABC]', '[DEF]', 'GHI IHG']
+            tokenizer.add_tokens(new_toks)
+            input = "[ABC] [DEF] [ABC] GHI IHG [DEF]"
+            encoded = tokenizer.encode(input, add_special_tokens=False)
+            decoded = tokenizer.decode(encoded)
+            self.assertEqual(decoded, input)
 
         def test_pretrained_model_lists(self):
             weights_list = list(self.tokenizer_class.max_model_input_sizes.keys())

From 7bd11dda6f43656cf0a3891b7f61a67196d233b4 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Fri, 13 Dec 2019 16:45:30 -0500
Subject: [PATCH 365/505] Release: v2.2.2

---
 README.md                | 2 +-
 docs/source/conf.py      | 2 +-
 setup.py                 | 2 +-
 transformers/__init__.py | 2 +-
 4 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/README.md b/README.md
index f3aa8a95ee..f24ceaa6d2 100644
--- a/README.md
+++ b/README.md
@@ -58,7 +58,7 @@ Choose the right framework for every part of a model's lifetime
 | [Quick tour: Fine-tuning/usage scripts](#quick-tour-of-the-fine-tuningusage-scripts) | Using provided scripts: GLUE, SQuAD and Text generation |
 | [Migrating from pytorch-transformers to transformers](#Migrating-from-pytorch-transformers-to-transformers) | Migrating your code from pytorch-transformers to transformers |
 | [Migrating from pytorch-pretrained-bert to pytorch-transformers](#Migrating-from-pytorch-pretrained-bert-to-transformers) | Migrating your code from pytorch-pretrained-bert to transformers |
-| [Documentation][(v2.2.0/v2.2.1)](https://huggingface.co/transformers/v2.2.0) [(v2.1.1)](https://huggingface.co/transformers/v2.1.1) [(v2.0.0)](https://huggingface.co/transformers/v2.0.0) [(v1.2.0)](https://huggingface.co/transformers/v1.2.0) [(v1.1.0)](https://huggingface.co/transformers/v1.1.0) [(v1.0.0)](https://huggingface.co/transformers/v1.0.0) [(master)](https://huggingface.co/transformers) | Full API documentation and more |
+| [Documentation][(v2.2.0/v2.2.1/v2.2.2)](https://huggingface.co/transformers/v2.2.0) [(v2.1.1)](https://huggingface.co/transformers/v2.1.1) [(v2.0.0)](https://huggingface.co/transformers/v2.0.0) [(v1.2.0)](https://huggingface.co/transformers/v1.2.0) [(v1.1.0)](https://huggingface.co/transformers/v1.1.0) [(v1.0.0)](https://huggingface.co/transformers/v1.0.0) [(master)](https://huggingface.co/transformers) | Full API documentation and more |
 
 ## Installation
 
diff --git a/docs/source/conf.py b/docs/source/conf.py
index 2f8505ab3a..99b7b44922 100644
--- a/docs/source/conf.py
+++ b/docs/source/conf.py
@@ -26,7 +26,7 @@ author = u'huggingface'
 # The short X.Y version
 version = u''
 # The full version, including alpha/beta/rc tags
-release = u'2.2.1'
+release = u'2.2.2'
 
 
 # -- General configuration ---------------------------------------------------
diff --git a/setup.py b/setup.py
index c4af32df83..eacb5ecec0 100644
--- a/setup.py
+++ b/setup.py
@@ -44,7 +44,7 @@ extras['all'] = [package for package in extras.values()]
 
 setup(
     name="transformers",
-    version="2.2.1",
+    version="2.2.2",
     author="Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond, Google AI Language Team Authors, Open AI team Authors, Facebook AI Authors, Carnegie Mellon University Authors",
     author_email="thomas@huggingface.co",
     description="State-of-the-art Natural Language Processing for TensorFlow 2.0 and PyTorch",
diff --git a/transformers/__init__.py b/transformers/__init__.py
index 5d7b0b772c..c11919f0a7 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -1,4 +1,4 @@
-__version__ = "2.2.1"
+__version__ = "2.2.2"
 
 # Work around to update TensorFlow's absl.logging threshold which alters the
 # default Python logging output behavior when present.

From b6d4284b26c0ab5e736cb7838b27b720225feeb7 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Fri, 13 Dec 2019 22:43:15 -0500
Subject: [PATCH 366/505] [cli] Uploads: fix + test edge case

---
 transformers/hf_api.py                |  3 +-
 transformers/tests/fixtures/empty.txt |  0
 transformers/tests/hf_api_test.py     | 44 +++++++++++++++++++--------
 3 files changed, 33 insertions(+), 14 deletions(-)
 create mode 100644 transformers/tests/fixtures/empty.txt

diff --git a/transformers/hf_api.py b/transformers/hf_api.py
index 3bbb6c567a..170732339a 100644
--- a/transformers/hf_api.py
+++ b/transformers/hf_api.py
@@ -131,8 +131,9 @@ class HfApi:
         # the client still has to specify it when uploading the file.
         with open(filepath, "rb") as f:
             pf = TqdmProgressFileReader(f)
+            data = f if pf.total_size > 0 else ""
 
-            r = requests.put(urls.write, data=f, headers={
+            r = requests.put(urls.write, data=data, headers={
                 "content-type": urls.type,
             })
             r.raise_for_status()
diff --git a/transformers/tests/fixtures/empty.txt b/transformers/tests/fixtures/empty.txt
new file mode 100644
index 0000000000..e69de29bb2
diff --git a/transformers/tests/hf_api_test.py b/transformers/tests/hf_api_test.py
index 92d41b6dff..b45f5aceed 100644
--- a/transformers/tests/hf_api_test.py
+++ b/transformers/tests/hf_api_test.py
@@ -15,18 +15,30 @@
 from __future__ import absolute_import, division, print_function
 
 import os
-import six
 import time
 import unittest
 
-from transformers.hf_api import HfApi, S3Obj, PresignedUrl, HfFolder, HTTPError
+import requests
+import six
+
+from transformers.hf_api import HfApi, HfFolder, HTTPError, PresignedUrl, S3Obj
 
 USER = "__DUMMY_TRANSFORMERS_USER__"
 PASS = "__DUMMY_TRANSFORMERS_PASS__"
-FILE_KEY = "Test-{}.txt".format(int(time.time()))
-FILE_PATH = os.path.join(
-    os.path.dirname(os.path.abspath(__file__)), "fixtures/input.txt"
-)
+FILES = [
+    (
+        "Test-{}.txt".format(int(time.time())),
+        os.path.join(
+            os.path.dirname(os.path.abspath(__file__)), "fixtures/input.txt"
+        )
+    ),
+    (
+        "yoyo {}.txt".format(int(time.time())), # space is intentional
+        os.path.join(
+            os.path.dirname(os.path.abspath(__file__)), "fixtures/empty.txt"
+        )
+    ),
+]
 
 
 
@@ -57,15 +69,21 @@ class HfApiEndpointsTest(HfApiCommonTest):
         self.assertEqual(user, USER)
 
     def test_presign(self):
-        urls = self._api.presign(token=self._token, filename=FILE_KEY)
-        self.assertIsInstance(urls, PresignedUrl)
-        self.assertEqual(urls.type, "text/plain")
+        for FILE_KEY, FILE_PATH in FILES:
+            urls = self._api.presign(token=self._token, filename=FILE_KEY)
+            self.assertIsInstance(urls, PresignedUrl)
+            self.assertEqual(urls.type, "text/plain")
 
     def test_presign_and_upload(self):
-        access_url = self._api.presign_and_upload(
-            token=self._token, filename=FILE_KEY, filepath=FILE_PATH
-        )
-        self.assertIsInstance(access_url, six.string_types)
+        for FILE_KEY, FILE_PATH in FILES:
+            access_url = self._api.presign_and_upload(
+                token=self._token, filename=FILE_KEY, filepath=FILE_PATH
+            )
+            self.assertIsInstance(access_url, six.string_types)
+            with open(FILE_PATH, 'r') as f:
+                body = f.read()
+            r = requests.get(access_url)
+            self.assertEqual(r.text, body)
 
     def test_list_objs(self):
         objs = self._api.list_objs(token=self._token)

From cbb368ca06998e5d98684bc622e1d8c68ba1d88f Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Sat, 14 Dec 2019 09:31:18 +0100
Subject: [PATCH 367/505] distilbert tests

---
 transformers/tests/modeling_common_test.py | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/transformers/tests/modeling_common_test.py b/transformers/tests/modeling_common_test.py
index cd4cf247a6..8920e8b826 100644
--- a/transformers/tests/modeling_common_test.py
+++ b/transformers/tests/modeling_common_test.py
@@ -96,9 +96,7 @@ class CommonTestCases:
 
                     # Make sure we don't have nans
                     out_1 = after_outputs[0].cpu().numpy()
-                    out_2 = outputs[0].cpu().numpy()
-                    out_1 = out_1[~np.isnan(out_1)]
-                    out_2 = out_2[~np.isnan(out_2)]
+                    out_1[np.isnan(out_1)] = 0
                     max_diff = np.amax(np.abs(out_1 - out_2))
                     self.assertLessEqual(max_diff, 1e-5)
 

From 7140363e092fecf82b73edd423bed3376ec1e150 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Sat, 14 Dec 2019 09:44:53 +0100
Subject: [PATCH 368/505] update bertabs

---
 .../summarization/configuration_bertabs.py    | 48 ++++++-------------
 1 file changed, 14 insertions(+), 34 deletions(-)

diff --git a/examples/summarization/configuration_bertabs.py b/examples/summarization/configuration_bertabs.py
index 054763ea93..b862d58d2b 100644
--- a/examples/summarization/configuration_bertabs.py
+++ b/examples/summarization/configuration_bertabs.py
@@ -33,6 +33,8 @@ class BertAbsConfig(PretrainedConfig):
     r""" Class to store the configuration of the BertAbs model.
 
     Arguments:
+        vocab_size: int
+            Number of tokens in the vocabulary.
         max_pos: int
             The maximum sequence length that this model will be used with.
         enc_layer: int
@@ -81,39 +83,17 @@ class BertAbsConfig(PretrainedConfig):
     ):
         super(BertAbsConfig, self).__init__(**kwargs)
 
-        if self._input_is_path_to_json(vocab_size):
-            path_to_json = vocab_size
-            with open(path_to_json, "r", encoding="utf-8") as reader:
-                json_config = json.loads(reader.read())
-            for key, value in json_config.items():
-                self.__dict__[key] = value
-        elif isinstance(vocab_size, int):
-            self.vocab_size = vocab_size
-            self.max_pos = max_pos
+        self.vocab_size = vocab_size
+        self.max_pos = max_pos
 
-            self.enc_layers = enc_layers
-            self.enc_hidden_size = enc_hidden_size
-            self.enc_heads = enc_heads
-            self.enc_ff_size = enc_ff_size
-            self.enc_dropout = enc_dropout
+        self.enc_layers = enc_layers
+        self.enc_hidden_size = enc_hidden_size
+        self.enc_heads = enc_heads
+        self.enc_ff_size = enc_ff_size
+        self.enc_dropout = enc_dropout
 
-            self.dec_layers = dec_layers
-            self.dec_hidden_size = dec_hidden_size
-            self.dec_heads = dec_heads
-            self.dec_ff_size = dec_ff_size
-            self.dec_dropout = dec_dropout
-        else:
-            raise ValueError(
-                "First argument must be either a vocabulary size (int)"
-                "or the path to a pretrained model config file (str)"
-            )
-
-    def _input_is_path_to_json(self, first_argument):
-        """ Checks whether the first argument passed to config
-        is the path to a JSON file that contains the config.
-        """
-        is_python_2 = sys.version_info[0] == 2
-        if is_python_2:
-            return isinstance(first_argument, unicode)
-        else:
-            return isinstance(first_argument, str)
+        self.dec_layers = dec_layers
+        self.dec_hidden_size = dec_hidden_size
+        self.dec_heads = dec_heads
+        self.dec_ff_size = dec_ff_size
+        self.dec_dropout = dec_dropout

From f1971bf303cc600cc47161f137e1b0baccd62925 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Sun, 15 Dec 2019 01:37:16 +0100
Subject: [PATCH 369/505] Binding pipelines to the cli.

---
 transformers-cli             |   4 +-
 transformers/commands/run.py |  56 +++++++++++++++++++
 transformers/pipelines.py    | 102 +++++++++++++++++++++++++++++++----
 3 files changed, 152 insertions(+), 10 deletions(-)
 create mode 100644 transformers/commands/run.py

diff --git a/transformers-cli b/transformers-cli
index 168e6e6f32..39b7f5816b 100755
--- a/transformers-cli
+++ b/transformers-cli
@@ -2,6 +2,7 @@
 from argparse import ArgumentParser
 
 from transformers.commands.download import DownloadCommand
+from transformers.commands.run import RunCommand
 from transformers.commands.serving import ServeCommand
 from transformers.commands.user import UserCommands
 from transformers.commands.train import TrainCommand
@@ -14,9 +15,10 @@ if __name__ == '__main__':
     # Register commands
     ConvertCommand.register_subcommand(commands_parser)
     DownloadCommand.register_subcommand(commands_parser)
+    RunCommand.register_subcommand(commands_parser)
     ServeCommand.register_subcommand(commands_parser)
-    UserCommands.register_subcommand(commands_parser)
     TrainCommand.register_subcommand(commands_parser)
+    UserCommands.register_subcommand(commands_parser)
 
     # Let's go
     args = parser.parse_args()
diff --git a/transformers/commands/run.py b/transformers/commands/run.py
new file mode 100644
index 0000000000..bcbb87391d
--- /dev/null
+++ b/transformers/commands/run.py
@@ -0,0 +1,56 @@
+from argparse import ArgumentParser
+
+from transformers.commands import BaseTransformersCLICommand
+from transformers.pipelines import pipeline, Pipeline, PipelineDataFormat, SUPPORTED_TASKS
+
+
+def try_infer_format_from_ext(path: str):
+    for ext in PipelineDataFormat.SUPPORTED_FORMATS:
+        if path.endswith(ext):
+            return ext
+
+    raise Exception(
+        'Unable to determine file format from file extension {}. '
+        'Please provide the format through --format {}'.format(path, PipelineDataFormat.SUPPORTED_FORMATS)
+    )
+
+
+def run_command_factory(args):
+    nlp = pipeline(task=args.task, model=args.model, tokenizer=args.tokenizer)
+    format = try_infer_format_from_ext(args.input) if args.format == 'infer' else args.format
+    reader = PipelineDataFormat.from_str(format, args.output, args.input, args.column)
+    return RunCommand(nlp, reader)
+
+
+class RunCommand(BaseTransformersCLICommand):
+
+    def __init__(self, nlp: Pipeline, reader: PipelineDataFormat):
+        self._nlp = nlp
+        self._reader = reader
+
+    @staticmethod
+    def register_subcommand(parser: ArgumentParser):
+        run_parser = parser.add_parser('run', help="Run a pipeline through the CLI")
+        run_parser.add_argument('--task', choices=SUPPORTED_TASKS.keys(), help='Task to run')
+        run_parser.add_argument('--model', type=str, required=True, help='Name or path to the model to instantiate.')
+        run_parser.add_argument('--tokenizer', type=str, help='Name of the tokenizer to use. (default: same as the model name)')
+        run_parser.add_argument('--column', type=str, required=True, help='Name of the column to use as input. (For multi columns input as QA use column1,columns2)')
+        run_parser.add_argument('--format', type=str, default='infer', choices=PipelineDataFormat.SUPPORTED_FORMATS, help='Input format to read from')
+        run_parser.add_argument('--input', type=str, required=True, help='Path to the file to use for inference')
+        run_parser.add_argument('--output', type=str, required=True, help='Path to the file that will be used post to write results.')
+        run_parser.add_argument('kwargs', nargs='*', help='Arguments to forward to the file format reader')
+        run_parser.set_defaults(func=run_command_factory)
+
+    def run(self):
+        nlp, output = self._nlp, []
+        for entry in self._reader:
+            if self._reader.is_multi_columns:
+                output += [nlp(**entry)]
+            else:
+                output += [nlp(entry)]
+
+        # Saving data
+        self._reader.save(output)
+
+
+
diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 6fbb7e2f04..a5718b822f 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -14,6 +14,8 @@
 # limitations under the License.
 from __future__ import absolute_import, division, print_function, unicode_literals
 
+import csv
+import json
 import os
 from abc import ABC, abstractmethod
 from itertools import groupby
@@ -25,11 +27,13 @@ from transformers import AutoTokenizer, PreTrainedTokenizer, PretrainedConfig, \
     SquadExample, squad_convert_examples_to_features, is_tf_available, is_torch_available, logger
 
 if is_tf_available():
-    from transformers import TFAutoModelForSequenceClassification, TFAutoModelForQuestionAnswering, TFAutoModelForTokenClassification
+    from transformers import TFAutoModel, TFAutoModelForSequenceClassification, \
+        TFAutoModelForQuestionAnswering, TFAutoModelForTokenClassification
 
 if is_torch_available():
     import torch
-    from transformers import AutoModelForSequenceClassification, AutoModelForQuestionAnswering, AutoModelForTokenClassification
+    from transformers import AutoModel, AutoModelForSequenceClassification, \
+        AutoModelForQuestionAnswering, AutoModelForTokenClassification
 
 
 class Pipeline(ABC):
@@ -58,6 +62,84 @@ class Pipeline(ABC):
         raise NotImplementedError()
 
 
+class PipelineDataFormat:
+    SUPPORTED_FORMATS = ['json', 'csv']
+
+    def __init__(self, output: str, path: str, column: str):
+        self.output = output
+        self.path = path
+        self.column = column.split(',')
+        self.is_multi_columns = len(self.column) > 1
+
+        if self.is_multi_columns:
+            self.column = [tuple(c.split('=')) if '=' in c else (c, c) for c in self.column]
+
+        from os.path import abspath, exists
+        if exists(abspath(self.output)):
+            raise OSError('{} already exists on disk'.format(self.output))
+
+        if not exists(abspath(self.path)):
+            raise OSError('{} doesnt exist on disk'.format(self.path))
+
+    @abstractmethod
+    def __iter__(self):
+        raise NotImplementedError()
+
+    @abstractmethod
+    def save(self, data: dict):
+        raise NotImplementedError()
+
+    @staticmethod
+    def from_str(name: str, output: str, path: str, column: str):
+        if name == 'json':
+            return JsonPipelineDataFormat(output, path, column)
+        elif name == 'csv':
+            return CsvPipelineDataFormat(output, path, column)
+        else:
+            raise KeyError('Unknown reader {} (Available reader are json/csv)'.format(name))
+
+
+class CsvPipelineDataFormat(PipelineDataFormat):
+    def __init__(self, output: str, path: str, column: str):
+        super().__init__(output, path, column)
+
+    def __iter__(self):
+        with open(self.path, 'r') as f:
+            reader = csv.DictReader(f)
+            for row in reader:
+                if self.is_multi_columns:
+                    yield {k: row[c] for k, c in self.column}
+                else:
+                    yield row[self.column]
+
+    def save(self, data: List[dict]):
+        with open(self.output, 'w') as f:
+            if len(data) > 0:
+                writer = csv.DictWriter(f, list(data[0].keys()))
+                writer.writeheader()
+                writer.writerows(data)
+
+
+class JsonPipelineDataFormat(PipelineDataFormat):
+
+    def __init__(self, output: str, path: str, column: str):
+        super().__init__(output, path, column)
+
+        with open(path, 'r') as f:
+            self._entries = json.load(f)
+
+    def __iter__(self):
+        for entry in self._entries:
+            if self.is_multi_columns:
+                yield {k: entry[c] for k, c in self.column}
+            else:
+                yield entry[self.column]
+
+    def save(self, data: dict):
+        with open(self.output, 'w') as f:
+            json.dump(data, f)
+
+
 class FeatureExtractionPipeline(Pipeline):
 
     def __call__(self, *texts, **kwargs):
@@ -127,7 +209,7 @@ class NerPipeline(Pipeline):
                 label_idx = score.argmax()
 
                 answer += [{
-                    'word': words[idx - 1], 'score': score[label_idx], 'entity': self.model.config.id2label[label_idx]
+                    'word': words[idx - 1], 'score': score[label_idx].item(), 'entity': self.model.config.id2label[label_idx]
                 }]
 
                 # Update token start
@@ -270,16 +352,18 @@ class QuestionAnsweringPipeline(Pipeline):
             char_to_word = np.array(example.char_to_word_offset)
 
             # Convert the answer (tokens) back to the original text
-            answers += [[
+            answers += [
                 {
-                    'score': score,
-                    'start': np.where(char_to_word == feature.token_to_orig_map[s])[0][0],
-                    'end': np.where(char_to_word == feature.token_to_orig_map[e])[0][-1],
+                    'score': score.item(),
+                    'start': np.where(char_to_word == feature.token_to_orig_map[s])[0][0].item(),
+                    'end': np.where(char_to_word == feature.token_to_orig_map[e])[0][-1].item(),
                     'answer': ' '.join(example.doc_tokens[feature.token_to_orig_map[s]: feature.token_to_orig_map[e] + 1])
                 }
                 for s, e, score in zip(starts, ends, scores)
-            ]]
+            ]
 
+        if len(answers) == 1:
+            return answers[0]
         return answers
 
     def decode(self, start: np.ndarray, end: np.ndarray, topk: int, max_answer_len: int) -> Tuple:
@@ -363,7 +447,7 @@ def pipeline(task: str, model, config: Optional[PretrainedConfig] = None, tokeni
     Utility factory method to build pipeline.
     """
     # Try to infer tokenizer from model name (if provided as str)
-    if not isinstance(tokenizer, PreTrainedTokenizer):
+    if tokenizer is None:
         if not isinstance(model, str):
             # Impossible to guest what is the right tokenizer here
             raise Exception('Tokenizer cannot be None if provided model is a PreTrainedModel instance')

From 8e3b1c860fbd872cb39d5093a1e739a67e7d0809 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Sun, 15 Dec 2019 01:37:52 +0100
Subject: [PATCH 370/505] Added FeatureExtraction pipeline.

---
 transformers/pipelines.py | 23 ++++++++++++++++++++++-
 1 file changed, 22 insertions(+), 1 deletion(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index a5718b822f..7383222c1f 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -143,7 +143,23 @@ class JsonPipelineDataFormat(PipelineDataFormat):
 class FeatureExtractionPipeline(Pipeline):
 
     def __call__(self, *texts, **kwargs):
-        pass
+        # Generic compatibility with sklearn and Keras
+        if 'X' in kwargs and not texts:
+            texts = kwargs.pop('X')
+
+        inputs = self.tokenizer.batch_encode_plus(
+            texts, add_special_tokens=True, return_tensors='tf' if is_tf_available() else 'pt'
+        )
+
+        if is_tf_available():
+            # TODO trace model
+            predictions = self.model(inputs)[0]
+        else:
+            import torch
+            with torch.no_grad():
+                predictions = self.model(**inputs)[0]
+
+        return predictions.numpy().tolist()
 
 
 class TextClassificationPipeline(Pipeline):
@@ -424,6 +440,11 @@ class QuestionAnsweringPipeline(Pipeline):
 
 # Register all the supported task here
 SUPPORTED_TASKS = {
+    'feature-extraction': {
+      'impl': FeatureExtractionPipeline,
+      'tf': TFAutoModel if is_tf_available() else None,
+      'pt': AutoModel if is_torch_available() else None,
+    },
     'text-classification': {
         'impl': TextClassificationPipeline,
         'tf': TFAutoModelForSequenceClassification if is_tf_available() else None,

From 1b8613acb32a568db8d9b74ee182d43c4f8e9cbb Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Mon, 16 Dec 2019 09:51:42 +0100
Subject: [PATCH 371/505] updating t5 config class

---
 transformers/configuration_t5.py       | 15 ++-------------
 transformers/tests/modeling_t5_test.py |  2 +-
 2 files changed, 3 insertions(+), 14 deletions(-)

diff --git a/transformers/configuration_t5.py b/transformers/configuration_t5.py
index 6391cb4180..377a0919d9 100644
--- a/transformers/configuration_t5.py
+++ b/transformers/configuration_t5.py
@@ -66,7 +66,7 @@ class T5Config(PretrainedConfig):
     pretrained_config_archive_map = T5_PRETRAINED_CONFIG_ARCHIVE_MAP
 
     def __init__(self,
-                 vocab_size_or_config_json_file=32128,
+                 vocab_size=32128,
                  n_positions=512,
                  d_model=512,
                  d_kv=64,
@@ -79,7 +79,7 @@ class T5Config(PretrainedConfig):
                  initializer_factor=1.0,
                  **kwargs):
         super(T5Config, self).__init__(**kwargs)
-        self.vocab_size = vocab_size_or_config_json_file if isinstance(vocab_size_or_config_json_file, int) else -1
+        self.vocab_size = vocab_size
         self.n_positions = n_positions
         self.d_model = d_model
         self.d_kv = d_kv
@@ -91,17 +91,6 @@ class T5Config(PretrainedConfig):
         self.layer_norm_epsilon = layer_norm_epsilon
         self.initializer_factor = initializer_factor
 
-        if isinstance(vocab_size_or_config_json_file, six.string_types):
-            with open(vocab_size_or_config_json_file, "r", encoding="utf-8") as reader:
-                json_config = json.loads(reader.read())
-            for key, value in json_config.items():
-                self.__dict__[key] = value
-        elif not isinstance(vocab_size_or_config_json_file, int):
-            raise ValueError(
-                "First argument must be either a vocabulary size (int)"
-                "or the path to a pretrained model config file (str)"
-            )
-
     @property
     def max_position_embeddings(self):
         return self.n_positions
diff --git a/transformers/tests/modeling_t5_test.py b/transformers/tests/modeling_t5_test.py
index a539cc868a..c337163375 100644
--- a/transformers/tests/modeling_t5_test.py
+++ b/transformers/tests/modeling_t5_test.py
@@ -93,7 +93,7 @@ class T5ModelTest(CommonTestCases.CommonModelTester):
                 decoder_lm_labels = ids_tensor([self.batch_size, self.decoder_seq_length], self.vocab_size)
 
             config = T5Config(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 n_positions=self.n_positions,
                 d_model=self.hidden_size,
                 d_ff=self.d_ff,

From 8669598abd7af877bd33890d62ae70ec1623f145 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Mon, 16 Dec 2019 09:59:36 +0100
Subject: [PATCH 372/505] update t5 tf

---
 transformers/tests/modeling_tf_t5_test.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/tests/modeling_tf_t5_test.py b/transformers/tests/modeling_tf_t5_test.py
index 99eec313f9..b905a9875b 100644
--- a/transformers/tests/modeling_tf_t5_test.py
+++ b/transformers/tests/modeling_tf_t5_test.py
@@ -87,7 +87,7 @@ class TFT5ModelTest(TFCommonTestCases.TFCommonModelTester):
                 token_labels = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)
 
             config = T5Config(
-                vocab_size_or_config_json_file=self.vocab_size,
+                vocab_size=self.vocab_size,
                 n_positions=self.n_positions,
                 d_model=self.hidden_size,
                 d_ff=self.d_ff,

From 56e98ba81a9a7410243a1117fb6148d5f353ef98 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Mon, 16 Dec 2019 11:07:27 +0100
Subject: [PATCH 373/505] add model cards cc @mfuntowicz

---
 transformers/__init__.py              |   3 +
 transformers/file_utils.py            |   2 +-
 transformers/model_card.py            | 248 ++++++++++++++++++++++++++
 transformers/tests/model_card_test.py |  87 +++++++++
 4 files changed, 339 insertions(+), 1 deletion(-)
 create mode 100644 transformers/model_card.py
 create mode 100644 transformers/tests/model_card_test.py

diff --git a/transformers/__init__.py b/transformers/__init__.py
index 740d2440c2..15c167a5ce 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -33,6 +33,9 @@ from .data import (is_sklearn_available,
 if is_sklearn_available():
     from .data import glue_compute_metrics, xnli_compute_metrics
 
+# Model Cards
+from .model_card import ModelCard
+
 # Tokenizers
 from .tokenization_utils import (PreTrainedTokenizer)
 from .tokenization_auto import AutoTokenizer
diff --git a/transformers/file_utils.py b/transformers/file_utils.py
index 03b2fdb9f4..81c9b8002f 100644
--- a/transformers/file_utils.py
+++ b/transformers/file_utils.py
@@ -72,7 +72,7 @@ WEIGHTS_NAME = "pytorch_model.bin"
 TF2_WEIGHTS_NAME = 'tf_model.h5'
 TF_WEIGHTS_NAME = 'model.ckpt'
 CONFIG_NAME = "config.json"
-
+MODEL_CARD_NAME = "model_card.json"
 
 DUMMY_INPUTS = [[7, 6, 0, 0, 1], [1, 2, 3, 0, 0], [0, 0, 0, 4, 5]]
 DUMMY_MASK = [[1, 1, 1, 1, 1], [1, 1, 1, 0, 0], [0, 0, 0, 1, 1]]
diff --git a/transformers/model_card.py b/transformers/model_card.py
new file mode 100644
index 0000000000..679c24872a
--- /dev/null
+++ b/transformers/model_card.py
@@ -0,0 +1,248 @@
+# coding=utf-8
+# Copyright 2018 The HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" Configuration base class and utilities."""
+
+from __future__ import (absolute_import, division, print_function,
+                        unicode_literals)
+
+import copy
+import json
+import logging
+import os
+import re
+from io import open
+
+from .configuration_bert import BERT_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_openai import OPENAI_GPT_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_transfo_xl import TRANSFO_XL_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_gpt2 import GPT2_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_ctrl import CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_xlnet import XLNET_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_xlm import XLM_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_roberta import ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_distilbert import DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_albert import ALBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_camembert import CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_t5 import T5_PRETRAINED_CONFIG_ARCHIVE_MAP
+
+from .file_utils import CONFIG_NAME, MODEL_CARD_NAME, cached_path, is_remote_url, hf_bucket_url
+
+
+logger = logging.getLogger(__name__)
+
+
+ALL_MODELS_MAP = dict((key, value)
+    for pretrained_map in [
+        BERT_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        OPENAI_GPT_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        TRANSFO_XL_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        GPT2_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        XLNET_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        XLM_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        ALBERT_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        T5_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        ]
+    for key, value, in pretrained_map.items())
+
+
+class ModelCard(object):
+    r""" Model Card class.
+        Store model card as well as methods for loading/downloading/saving model cards.
+
+        Please read the following paper for details and explanation on the sections:
+            "Model Cards for Model Reporting"
+                by Margaret Mitchell, Simone Wu,
+                Andrew Zaldivar, Parker Barnes, Lucy Vasserman, Ben Hutchinson, Elena Spitzer,
+                Inioluwa Deborah Raji and Timnit Gebru for the proposal behind model cards.
+            Link: https://arxiv.org/abs/1810.03993
+
+        Note:
+            A model card can be loaded and saved to disk.
+
+        Parameters:
+    """
+    def __init__(self, **kwargs):
+        # Recomended attributes from https://arxiv.org/abs/1810.03993 (see papers)
+        self.model_details = kwargs.pop('model_details', {})
+        self.intended_use = kwargs.pop('intended_use', {})
+        self.factors = kwargs.pop('factors', {})
+        self.metrics = kwargs.pop('metrics', {})
+        self.evaluation_data = kwargs.pop('evaluation_data', {})
+        self.training_data = kwargs.pop('training_data', {})
+        self.quantitative_analyses = kwargs.pop('quantitative_analyses', {})
+        self.ethical_considerations = kwargs.pop('ethical_considerations', {})
+        self.caveats_and_recommendations = kwargs.pop('caveats_and_recommendations', {})
+
+        # Open additional attributes
+        for key, value in kwargs.items():
+            try:
+                setattr(self, key, value)
+            except AttributeError as err:
+                logger.error("Can't set {} with value {} for {}".format(key, value, self))
+                raise err
+
+    def save_pretrained(self, save_directory):
+        """ Save a model card object to the directory `save_directory`, so that it
+            can be re-loaded using the :func:`~transformers.ModelCard.from_pretrained` class method.
+        """
+        assert os.path.isdir(save_directory), "Saving path should be a directory where the model card can be saved"
+
+        # If we save using the predefined names, we can load using `from_pretrained`
+        output_model_card_file = os.path.join(save_directory, MODEL_CARD_NAME)
+
+        self.to_json_file(output_model_card_file)
+        logger.info("Model card saved in {}".format(output_model_card_file))
+
+    @classmethod
+    def from_pretrained(cls, pretrained_model_name_or_path, **kwargs):
+        r""" Instantiate a :class:`~transformers.ModelCard` from a pre-trained model model card.
+
+        Parameters:
+            pretrained_model_name_or_path: either:
+
+                - a string with the `shortcut name` of a pre-trained model card to load from cache or download, e.g.: ``bert-base-uncased``.
+                - a string with the `identifier name` of a pre-trained model card that was user-uploaded to our S3, e.g.: ``dbmdz/bert-base-german-cased``.
+                - a path to a `directory` containing a mode card file saved using the :func:`~transformers.ModelCard.save_pretrained` method, e.g.: ``./my_model_directory/``.
+                - a path or url to a saved model card JSON `file`, e.g.: ``./my_model_directory/model_card.json``.
+
+            cache_dir: (`optional`) string:
+                Path to a directory in which a downloaded pre-trained model
+                card should be cached if the standard cache should not be used.
+
+            kwargs: (`optional`) dict: key/value pairs with which to update the ModelCard object after loading.
+
+                - The values in kwargs of any keys which are model card attributes will be used to override the loaded values.
+                - Behavior concerning key/value pairs whose keys are *not* model card attributes is controlled by the `return_unused_kwargs` keyword parameter.
+
+            force_download: (`optional`) boolean, default False:
+                Force to (re-)download the model card file and override the cached version if it exists.
+
+            resume_download: (`optional`) boolean, default False:
+                Do not delete incompletely recieved file. Attempt to resume the download if such a file exists.
+
+            proxies: (`optional`) dict, default None:
+                A dictionary of proxy servers to use by protocol or endpoint, e.g.: {'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.
+                The proxies are used on each request.
+
+            return_unused_kwargs: (`optional`) bool:
+
+                - If False, then this function returns just the final model card object.
+                - If True, then this functions returns a tuple `(model card, unused_kwargs)` where `unused_kwargs` is a dictionary consisting of the key/value pairs whose keys are not model card attributes: ie the part of kwargs which has not been used to update `ModelCard` and is otherwise ignored.
+
+        Examples::
+
+            model_card = ModelCard.from_pretrained('bert-base-uncased')    # Download model card from S3 and cache.
+            model_card = ModelCard.from_pretrained('./test/saved_model/')  # E.g. model card was saved using `save_pretrained('./test/saved_model/')`
+            model_card = ModelCard.from_pretrained('./test/saved_model/model_card.json')
+            model_card = ModelCard.from_pretrained('bert-base-uncased', output_attention=True, foo=False)
+
+        """
+        cache_dir = kwargs.pop('cache_dir', None)
+        force_download = kwargs.pop('force_download', False)
+        resume_download = kwargs.pop('resume_download', False)
+        proxies = kwargs.pop('proxies', None)
+        return_unused_kwargs = kwargs.pop('return_unused_kwargs', False)
+
+        if pretrained_model_name_or_path in ALL_MODELS_MAP:
+            model_card_file = ALL_MODELS_MAP[pretrained_model_name_or_path]
+            model_card_file.replace(CONFIG_NAME, MODEL_CARD_NAME)  # For simplicity we use the same pretrained url than config but with a different suffix
+        elif os.path.isdir(pretrained_model_name_or_path):
+            model_card_file = os.path.join(pretrained_model_name_or_path, MODEL_CARD_NAME)
+        elif os.path.isfile(pretrained_model_name_or_path) or is_remote_url(pretrained_model_name_or_path):
+            model_card_file = pretrained_model_name_or_path
+        else:
+            model_card_file = hf_bucket_url(pretrained_model_name_or_path, postfix=MODEL_CARD_NAME)
+        # redirect to the cache, if necessary
+        try:
+            resolved_model_card_file = cached_path(model_card_file, cache_dir=cache_dir, force_download=force_download,
+                                               proxies=proxies, resume_download=resume_download)
+
+            if resolved_model_card_file == model_card_file:
+                logger.info("loading model card file {}".format(model_card_file))
+            else:
+                logger.info("loading model card file {} from cache at {}".format(
+                    model_card_file, resolved_model_card_file))
+
+            # Load model card
+            model_card = cls.from_json_file(resolved_model_card_file)
+
+        except EnvironmentError:
+            if pretrained_model_name_or_path in ALL_MODELS_MAP:
+                logger.warning("Couldn't reach server at '{}' to download model card file.".format(
+                        model_card_file))
+            else:
+                logger.warning("Model name '{}' was not found in model name list ({}). " \
+                      "We assumed '{}' was a path or url to a model card file named {} or " \
+                      "a directory containing such a file but couldn't find any such file at this path or url.".format(
+                        pretrained_model_name_or_path,
+                        ', '.join(ALL_MODELS_MAP.keys()),
+                        model_card_file, MODEL_CARD_NAME))
+
+            logger.warning("Creating an empty model card.")
+
+            # We fall back on creating an empty model card
+            model_card = cls()
+
+        # Update model card with kwargs if needed
+        to_remove = []
+        for key, value in kwargs.items():
+            if hasattr(model_card, key):
+                setattr(model_card, key, value)
+                to_remove.append(key)
+        for key in to_remove:
+            kwargs.pop(key, None)
+
+        logger.info("Model card: %s", str(model_card))
+        if return_unused_kwargs:
+            return model_card, kwargs
+        else:
+            return model_card
+
+    @classmethod
+    def from_dict(cls, json_object):
+        """Constructs a `ModelCard` from a Python dictionary of parameters."""
+        return cls(**json_object)
+
+    @classmethod
+    def from_json_file(cls, json_file):
+        """Constructs a `ModelCard` from a json file of parameters."""
+        with open(json_file, "r", encoding='utf-8') as reader:
+            text = reader.read()
+        dict_obj = json.loads(text)
+        return cls(**dict_obj)
+
+    def __eq__(self, other):
+        return self.__dict__ == other.__dict__
+
+    def __repr__(self):
+        return str(self.to_json_string())
+
+    def to_dict(self):
+        """Serializes this instance to a Python dictionary."""
+        output = copy.deepcopy(self.__dict__)
+        return output
+
+    def to_json_string(self):
+        """Serializes this instance to a JSON string."""
+        return json.dumps(self.to_dict(), indent=2, sort_keys=True) + "\n"
+
+    def to_json_file(self, json_file_path):
+        """ Save this instance to a json file."""
+        with open(json_file_path, "w", encoding='utf-8') as writer:
+            writer.write(self.to_json_string())
diff --git a/transformers/tests/model_card_test.py b/transformers/tests/model_card_test.py
new file mode 100644
index 0000000000..4364cbacec
--- /dev/null
+++ b/transformers/tests/model_card_test.py
@@ -0,0 +1,87 @@
+# coding=utf-8
+# Copyright 2019 HuggingFace Inc.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from __future__ import absolute_import, division, print_function, unicode_literals
+
+import os
+import sys
+import json
+import tempfile
+import shutil
+import unittest
+
+from transformers.model_card import ModelCard
+from .tokenization_tests_commons import TemporaryDirectory
+
+class ModelCardTester(unittest.TestCase):
+
+    def setUp(self):
+        self.inputs_dict = {'model_details': {
+                                'Organization': 'testing',
+                                'Model date': 'today',
+                                'Model version': 'v2.1, Developed by Test Corp in 2019.',
+                                'Architecture': 'Convolutional Neural Network.',
+                                },
+                            'metrics': 'BLEU and ROUGE-1',
+                            'evaluation_data':{
+                                'Datasets':{
+                                    'BLEU': 'My-great-dataset-v1',
+                                    'ROUGE-1': 'My-short-dataset-v2.1',
+                                },
+                                'Preprocessing': 'See details on https://arxiv.org/pdf/1810.03993.pdf'
+                            },
+                            'training_data':{
+                                'Dataset': 'English Wikipedia dump dated 2018-12-01',
+                                'Preprocessing': 'Using SentencePiece vocabulary of size 52k tokens. See details on https://arxiv.org/pdf/1810.03993.pdf'
+                            },
+                            'quantitative_analyses': {
+                                'BLEU': 55.1,
+                                'ROUGE-1': 76,
+                            },
+                            }
+        self.tmpdirname = tempfile.mkdtemp()
+
+    def tearDown(self):
+        shutil.rmtree(self.tmpdirname)
+
+    def test_model_card_common_properties(self):
+        model_card = ModelCard.from_dict(self.inputs_dict)
+        self.assertTrue(hasattr(model_card, 'model_details'))
+        self.assertTrue(hasattr(model_card, 'intended_use'))
+        self.assertTrue(hasattr(model_card, 'factors'))
+        self.assertTrue(hasattr(model_card, 'metrics'))
+        self.assertTrue(hasattr(model_card, 'evaluation_data'))
+        self.assertTrue(hasattr(model_card, 'training_data'))
+        self.assertTrue(hasattr(model_card, 'quantitative_analyses'))
+        self.assertTrue(hasattr(model_card, 'ethical_considerations'))
+        self.assertTrue(hasattr(model_card, 'caveats_and_recommendations'))
+
+    def test_model_card_to_json_string(self):
+        model_card = ModelCard.from_dict(self.inputs_dict)
+        obj = json.loads(model_card.to_json_string())
+        for key, value in self.inputs_dict.items():
+            self.assertEqual(obj[key], value)
+
+    def test_model_card_to_json_file(self):
+        model_card_first = ModelCard.from_dict(self.inputs_dict)
+
+        with TemporaryDirectory() as tmpdirname:
+            filename = os.path.join(tmpdirname, u"model_card.json")
+            model_card_first.to_json_file(filename)
+            model_card_second = ModelCard.from_json_file(filename)
+
+        self.assertEqual(model_card_second.to_dict(), model_card_first.to_dict())
+
+if __name__ == "__main__":
+    unittest.main()

From d3418a94ff4256725a690bd9c8167489b6f593b8 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Mon, 16 Dec 2019 13:52:41 +0100
Subject: [PATCH 374/505] update tests

---
 .../tests/configuration_common_test.py        | 27 ++++++++++++-------
 transformers/tests/model_card_test.py         | 16 ++++++-----
 2 files changed, 27 insertions(+), 16 deletions(-)

diff --git a/transformers/tests/configuration_common_test.py b/transformers/tests/configuration_common_test.py
index 8ee751153c..376d110d3c 100644
--- a/transformers/tests/configuration_common_test.py
+++ b/transformers/tests/configuration_common_test.py
@@ -16,15 +16,12 @@ from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 
-import copy
 import os
-import shutil
 import json
-import random
-import uuid
+import tempfile
 
 import unittest
-import logging
+from .tokenization_tests_commons import TemporaryDirectory
 
 
 class ConfigTester(object):
@@ -48,16 +45,28 @@ class ConfigTester(object):
 
     def create_and_test_config_to_json_file(self):
         config_first = self.config_class(**self.inputs_dict)
-        json_file_path = os.path.join(os.getcwd(), "config_" + str(uuid.uuid4()) + ".json")
-        config_first.to_json_file(json_file_path)
-        config_second = self.config_class.from_json_file(json_file_path)
-        os.remove(json_file_path)
+
+        with TemporaryDirectory() as tmpdirname:
+            json_file_path = os.path.join(tmpdirname, "config.json")
+            config_first.to_json_file(json_file_path)
+            config_second = self.config_class.from_json_file(json_file_path)
+
+        self.parent.assertEqual(config_second.to_dict(), config_first.to_dict())
+
+    def create_and_test_config_from_and_save_pretrained(self):
+        config_first = self.config_class(**self.inputs_dict)
+
+        with TemporaryDirectory() as tmpdirname:
+            config_first.save_pretrained(tmpdirname)
+            config_second = self.config_class.from_pretrained(tmpdirname)
+
         self.parent.assertEqual(config_second.to_dict(), config_first.to_dict())
 
     def run_common_tests(self):
         self.create_and_test_config_common_properties()
         self.create_and_test_config_to_json_string()
         self.create_and_test_config_to_json_file()
+        self.create_and_test_config_from_and_save_pretrained()
 
 if __name__ == "__main__":
     unittest.main()
\ No newline at end of file
diff --git a/transformers/tests/model_card_test.py b/transformers/tests/model_card_test.py
index 4364cbacec..e75716f0aa 100644
--- a/transformers/tests/model_card_test.py
+++ b/transformers/tests/model_card_test.py
@@ -15,10 +15,7 @@
 from __future__ import absolute_import, division, print_function, unicode_literals
 
 import os
-import sys
 import json
-import tempfile
-import shutil
 import unittest
 
 from transformers.model_card import ModelCard
@@ -50,10 +47,6 @@ class ModelCardTester(unittest.TestCase):
                                 'ROUGE-1': 76,
                             },
                             }
-        self.tmpdirname = tempfile.mkdtemp()
-
-    def tearDown(self):
-        shutil.rmtree(self.tmpdirname)
 
     def test_model_card_common_properties(self):
         model_card = ModelCard.from_dict(self.inputs_dict)
@@ -83,5 +76,14 @@ class ModelCardTester(unittest.TestCase):
 
         self.assertEqual(model_card_second.to_dict(), model_card_first.to_dict())
 
+    def test_model_card_from_and_save_pretrained(self):
+        model_card_first = ModelCard.from_dict(self.inputs_dict)
+
+        with TemporaryDirectory() as tmpdirname:
+            model_card_first.save_pretrained(tmpdirname)
+            model_card_second = ModelCard.from_pretrained(tmpdirname)
+
+        self.assertEqual(model_card_second.to_dict(), model_card_first.to_dict())
+
 if __name__ == "__main__":
     unittest.main()

From a4d07b983a6c1716b4d39cf3fed570562aebf3f7 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Mon, 16 Dec 2019 14:00:32 +0100
Subject: [PATCH 375/505] dict of all config and model files cc @LysandreJik

---
 transformers/__init__.py           |  6 ++---
 transformers/configuration_auto.py | 42 ++++++++++++++++++++---------
 transformers/model_card.py         | 43 +++++-------------------------
 transformers/modeling_auto.py      | 42 ++++++++++++++++++++---------
 transformers/modeling_tf_auto.py   | 38 +++++++++++++++++++-------
 5 files changed, 98 insertions(+), 73 deletions(-)

diff --git a/transformers/__init__.py b/transformers/__init__.py
index 15c167a5ce..0b343bed2b 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -55,7 +55,7 @@ from .tokenization_t5 import T5Tokenizer
 
 # Configurations
 from .configuration_utils import PretrainedConfig
-from .configuration_auto import AutoConfig
+from .configuration_auto import AutoConfig, ALL_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_bert import BertConfig, BERT_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_openai import OpenAIGPTConfig, OPENAI_GPT_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_transfo_xl import TransfoXLConfig, TRANSFO_XL_PRETRAINED_CONFIG_ARCHIVE_MAP
@@ -73,7 +73,7 @@ from .configuration_t5 import T5Config, T5_PRETRAINED_CONFIG_ARCHIVE_MAP
 if is_torch_available():
     from .modeling_utils import (PreTrainedModel, prune_layer, Conv1D)
     from .modeling_auto import (AutoModel, AutoModelForSequenceClassification, AutoModelForQuestionAnswering,
-                                AutoModelWithLMHead)
+                                AutoModelWithLMHead, ALL_PRETRAINED_MODEL_ARCHIVE_MAP)
 
     from .modeling_bert import (BertPreTrainedModel, BertModel, BertForPreTraining,
                                 BertForMaskedLM, BertForNextSentencePrediction,
@@ -131,7 +131,7 @@ if is_torch_available():
 if is_tf_available():
     from .modeling_tf_utils import TFPreTrainedModel, TFSharedEmbeddings, TFSequenceSummary, shape_list
     from .modeling_tf_auto import (TFAutoModel, TFAutoModelForSequenceClassification, TFAutoModelForQuestionAnswering,
-                                   TFAutoModelWithLMHead)
+                                   TFAutoModelWithLMHead, TF_ALL_PRETRAINED_MODEL_ARCHIVE_MAP)
 
     from .modeling_tf_bert import (TFBertPreTrainedModel, TFBertMainLayer, TFBertEmbeddings,
                                    TFBertModel, TFBertForPreTraining,
diff --git a/transformers/configuration_auto.py b/transformers/configuration_auto.py
index 680c55fa54..9fe58f173a 100644
--- a/transformers/configuration_auto.py
+++ b/transformers/configuration_auto.py
@@ -18,22 +18,40 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 
 import logging
 
-from .configuration_bert import BertConfig
-from .configuration_openai import OpenAIGPTConfig
-from .configuration_gpt2 import GPT2Config
-from .configuration_transfo_xl import TransfoXLConfig
-from .configuration_xlnet import XLNetConfig
-from .configuration_xlm import XLMConfig
-from .configuration_roberta import RobertaConfig
-from .configuration_distilbert import DistilBertConfig
-from .configuration_ctrl import CTRLConfig
-from .configuration_camembert import CamembertConfig
-from .configuration_albert import AlbertConfig
-from .configuration_t5 import T5Config
+from .configuration_bert import BertConfig, BERT_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_openai import OpenAIGPTConfig, OPENAI_GPT_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_transfo_xl import TransfoXLConfig, TRANSFO_XL_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_gpt2 import GPT2Config, GPT2_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_ctrl import CTRLConfig, CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_xlnet import XLNetConfig, XLNET_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_xlm import XLMConfig, XLM_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_roberta import RobertaConfig, ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_distilbert import DistilBertConfig, DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_albert import AlbertConfig, ALBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_camembert import CamembertConfig, CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_t5 import T5Config, T5_PRETRAINED_CONFIG_ARCHIVE_MAP
 
 logger = logging.getLogger(__name__)
 
 
+ALL_PRETRAINED_CONFIG_ARCHIVE_MAP = dict((key, value)
+    for pretrained_map in [
+        BERT_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        OPENAI_GPT_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        TRANSFO_XL_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        GPT2_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        XLNET_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        XLM_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        ALBERT_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        T5_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        ]
+    for key, value, in pretrained_map.items())
+
+
 class AutoConfig(object):
     r""":class:`~transformers.AutoConfig` is a generic configuration class
         that will be instantiated as one of the configuration classes of the library
diff --git a/transformers/model_card.py b/transformers/model_card.py
index 679c24872a..6d56089844 100644
--- a/transformers/model_card.py
+++ b/transformers/model_card.py
@@ -21,21 +21,9 @@ import copy
 import json
 import logging
 import os
-import re
 from io import open
 
-from .configuration_bert import BERT_PRETRAINED_CONFIG_ARCHIVE_MAP
-from .configuration_openai import OPENAI_GPT_PRETRAINED_CONFIG_ARCHIVE_MAP
-from .configuration_transfo_xl import TRANSFO_XL_PRETRAINED_CONFIG_ARCHIVE_MAP
-from .configuration_gpt2 import GPT2_PRETRAINED_CONFIG_ARCHIVE_MAP
-from .configuration_ctrl import CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP
-from .configuration_xlnet import XLNET_PRETRAINED_CONFIG_ARCHIVE_MAP
-from .configuration_xlm import XLM_PRETRAINED_CONFIG_ARCHIVE_MAP
-from .configuration_roberta import ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP
-from .configuration_distilbert import DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
-from .configuration_albert import ALBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
-from .configuration_camembert import CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
-from .configuration_t5 import T5_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_auto import ALL_PRETRAINED_CONFIG_ARCHIVE_MAP
 
 from .file_utils import CONFIG_NAME, MODEL_CARD_NAME, cached_path, is_remote_url, hf_bucket_url
 
@@ -43,24 +31,6 @@ from .file_utils import CONFIG_NAME, MODEL_CARD_NAME, cached_path, is_remote_url
 logger = logging.getLogger(__name__)
 
 
-ALL_MODELS_MAP = dict((key, value)
-    for pretrained_map in [
-        BERT_PRETRAINED_CONFIG_ARCHIVE_MAP,
-        OPENAI_GPT_PRETRAINED_CONFIG_ARCHIVE_MAP,
-        TRANSFO_XL_PRETRAINED_CONFIG_ARCHIVE_MAP,
-        GPT2_PRETRAINED_CONFIG_ARCHIVE_MAP,
-        CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP,
-        XLNET_PRETRAINED_CONFIG_ARCHIVE_MAP,
-        XLM_PRETRAINED_CONFIG_ARCHIVE_MAP,
-        ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP,
-        DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP,
-        ALBERT_PRETRAINED_CONFIG_ARCHIVE_MAP,
-        CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP,
-        T5_PRETRAINED_CONFIG_ARCHIVE_MAP,
-        ]
-    for key, value, in pretrained_map.items())
-
-
 class ModelCard(object):
     r""" Model Card class.
         Store model card as well as methods for loading/downloading/saving model cards.
@@ -159,9 +129,10 @@ class ModelCard(object):
         proxies = kwargs.pop('proxies', None)
         return_unused_kwargs = kwargs.pop('return_unused_kwargs', False)
 
-        if pretrained_model_name_or_path in ALL_MODELS_MAP:
-            model_card_file = ALL_MODELS_MAP[pretrained_model_name_or_path]
-            model_card_file.replace(CONFIG_NAME, MODEL_CARD_NAME)  # For simplicity we use the same pretrained url than config but with a different suffix
+        if pretrained_model_name_or_path in ALL_PRETRAINED_CONFIG_ARCHIVE_MAP:
+            # For simplicity we use the same pretrained url than the configuration files but with a different suffix (model_card.json)
+            model_card_file = ALL_PRETRAINED_CONFIG_ARCHIVE_MAP[pretrained_model_name_or_path]
+            model_card_file.replace(CONFIG_NAME, MODEL_CARD_NAME)
         elif os.path.isdir(pretrained_model_name_or_path):
             model_card_file = os.path.join(pretrained_model_name_or_path, MODEL_CARD_NAME)
         elif os.path.isfile(pretrained_model_name_or_path) or is_remote_url(pretrained_model_name_or_path):
@@ -183,7 +154,7 @@ class ModelCard(object):
             model_card = cls.from_json_file(resolved_model_card_file)
 
         except EnvironmentError:
-            if pretrained_model_name_or_path in ALL_MODELS_MAP:
+            if pretrained_model_name_or_path in ALL_PRETRAINED_CONFIG_ARCHIVE_MAP:
                 logger.warning("Couldn't reach server at '{}' to download model card file.".format(
                         model_card_file))
             else:
@@ -191,7 +162,7 @@ class ModelCard(object):
                       "We assumed '{}' was a path or url to a model card file named {} or " \
                       "a directory containing such a file but couldn't find any such file at this path or url.".format(
                         pretrained_model_name_or_path,
-                        ', '.join(ALL_MODELS_MAP.keys()),
+                        ', '.join(ALL_PRETRAINED_CONFIG_ARCHIVE_MAP.keys()),
                         model_card_file, MODEL_CARD_NAME))
 
             logger.warning("Creating an empty model card.")
diff --git a/transformers/modeling_auto.py b/transformers/modeling_auto.py
index 19a54cca86..1a30ea4623 100644
--- a/transformers/modeling_auto.py
+++ b/transformers/modeling_auto.py
@@ -18,18 +18,18 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 
 import logging
 
-from .modeling_bert import BertModel, BertForMaskedLM, BertForSequenceClassification, BertForQuestionAnswering
-from .modeling_openai import OpenAIGPTModel, OpenAIGPTLMHeadModel
-from .modeling_gpt2 import GPT2Model, GPT2LMHeadModel
-from .modeling_ctrl import CTRLModel, CTRLLMHeadModel
-from .modeling_transfo_xl import TransfoXLModel, TransfoXLLMHeadModel
-from .modeling_xlnet import XLNetModel, XLNetLMHeadModel, XLNetForSequenceClassification, XLNetForQuestionAnswering
-from .modeling_xlm import XLMModel, XLMWithLMHeadModel, XLMForSequenceClassification, XLMForQuestionAnswering
-from .modeling_roberta import RobertaModel, RobertaForMaskedLM, RobertaForSequenceClassification
-from .modeling_distilbert import DistilBertModel, DistilBertForQuestionAnswering, DistilBertForMaskedLM, DistilBertForSequenceClassification
-from .modeling_camembert import CamembertModel, CamembertForMaskedLM, CamembertForSequenceClassification, CamembertForMultipleChoice
-from .modeling_albert import AlbertModel, AlbertForMaskedLM, AlbertForSequenceClassification, AlbertForQuestionAnswering
-from .modeling_t5 import T5Model, T5WithLMHeadModel
+from .modeling_bert import BertModel, BertForMaskedLM, BertForSequenceClassification, BertForQuestionAnswering, BERT_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_openai import OpenAIGPTModel, OpenAIGPTLMHeadModel, OPENAI_GPT_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_gpt2 import GPT2Model, GPT2LMHeadModel, GPT2_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_ctrl import CTRLModel, CTRLLMHeadModel, CTRL_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_transfo_xl import TransfoXLModel, TransfoXLLMHeadModel, TRANSFO_XL_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_xlnet import XLNetModel, XLNetLMHeadModel, XLNetForSequenceClassification, XLNetForQuestionAnswering, XLNET_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_xlm import XLMModel, XLMWithLMHeadModel, XLMForSequenceClassification, XLMForQuestionAnswering, XLM_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_roberta import RobertaModel, RobertaForMaskedLM, RobertaForSequenceClassification, ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_distilbert import DistilBertModel, DistilBertForQuestionAnswering, DistilBertForMaskedLM, DistilBertForSequenceClassification, DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_camembert import CamembertModel, CamembertForMaskedLM, CamembertForSequenceClassification, CamembertForMultipleChoice, CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_albert import AlbertModel, AlbertForMaskedLM, AlbertForSequenceClassification, AlbertForQuestionAnswering, ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_t5 import T5Model, T5WithLMHeadModel, T5_PRETRAINED_MODEL_ARCHIVE_MAP
 
 from .modeling_utils import PreTrainedModel, SequenceSummary
 
@@ -38,6 +38,24 @@ from .file_utils import add_start_docstrings
 logger = logging.getLogger(__name__)
 
 
+ALL_PRETRAINED_MODEL_ARCHIVE_MAP = dict((key, value)
+    for pretrained_map in [
+        BERT_PRETRAINED_MODEL_ARCHIVE_MAP,
+        OPENAI_GPT_PRETRAINED_MODEL_ARCHIVE_MAP,
+        TRANSFO_XL_PRETRAINED_MODEL_ARCHIVE_MAP,
+        GPT2_PRETRAINED_MODEL_ARCHIVE_MAP,
+        CTRL_PRETRAINED_MODEL_ARCHIVE_MAP,
+        XLNET_PRETRAINED_MODEL_ARCHIVE_MAP,
+        XLM_PRETRAINED_MODEL_ARCHIVE_MAP,
+        ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP,
+        DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP,
+        ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP,
+        CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP,
+        T5_PRETRAINED_MODEL_ARCHIVE_MAP,
+        ]
+    for key, value, in pretrained_map.items())
+
+
 class AutoModel(object):
     r"""
         :class:`~transformers.AutoModel` is a generic model class
diff --git a/transformers/modeling_tf_auto.py b/transformers/modeling_tf_auto.py
index b4ff660098..9c687d9235 100644
--- a/transformers/modeling_tf_auto.py
+++ b/transformers/modeling_tf_auto.py
@@ -18,22 +18,40 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 
 import logging
 
-from .modeling_tf_bert import TFBertModel, TFBertForMaskedLM, TFBertForSequenceClassification, TFBertForQuestionAnswering
-from .modeling_tf_openai import TFOpenAIGPTModel, TFOpenAIGPTLMHeadModel
-from .modeling_tf_gpt2 import TFGPT2Model, TFGPT2LMHeadModel
-from .modeling_tf_transfo_xl import TFTransfoXLModel, TFTransfoXLLMHeadModel
-from .modeling_tf_xlnet import TFXLNetModel, TFXLNetLMHeadModel, TFXLNetForSequenceClassification, TFXLNetForQuestionAnsweringSimple
-from .modeling_tf_xlm import TFXLMModel, TFXLMWithLMHeadModel, TFXLMForSequenceClassification, TFXLMForQuestionAnsweringSimple
-from .modeling_tf_roberta import TFRobertaModel, TFRobertaForMaskedLM, TFRobertaForSequenceClassification
-from .modeling_tf_distilbert import TFDistilBertModel, TFDistilBertForQuestionAnswering, TFDistilBertForMaskedLM, TFDistilBertForSequenceClassification
-from .modeling_tf_ctrl import TFCTRLModel, TFCTRLLMHeadModel
-from .modeling_tf_t5 import TFT5Model, TFT5WithLMHeadModel
+from .modeling_tf_bert import TFBertModel, TFBertForMaskedLM, TFBertForSequenceClassification, TFBertForQuestionAnswering, TF_BERT_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_tf_openai import TFOpenAIGPTModel, TFOpenAIGPTLMHeadModel, TF_OPENAI_GPT_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_tf_gpt2 import TFGPT2Model, TFGPT2LMHeadModel, TF_GPT2_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_tf_transfo_xl import TFTransfoXLModel, TFTransfoXLLMHeadModel, TF_TRANSFO_XL_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_tf_xlnet import TFXLNetModel, TFXLNetLMHeadModel, TFXLNetForSequenceClassification, TFXLNetForQuestionAnsweringSimple, TF_XLNET_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_tf_xlm import TFXLMModel, TFXLMWithLMHeadModel, TFXLMForSequenceClassification, TFXLMForQuestionAnsweringSimple, TF_XLM_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_tf_roberta import TFRobertaModel, TFRobertaForMaskedLM, TFRobertaForSequenceClassification, TF_ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_tf_distilbert import TFDistilBertModel, TFDistilBertForQuestionAnswering, TFDistilBertForMaskedLM, TFDistilBertForSequenceClassification, TF_DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_tf_ctrl import TFCTRLModel, TFCTRLLMHeadModel, TF_CTRL_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_tf_t5 import TFT5Model, TFT5WithLMHeadModel, TF_T5_PRETRAINED_MODEL_ARCHIVE_MAP
 
 from .file_utils import add_start_docstrings
 
 logger = logging.getLogger(__name__)
 
 
+TF_ALL_PRETRAINED_MODEL_ARCHIVE_MAP = dict((key, value)
+    for pretrained_map in [
+        TF_BERT_PRETRAINED_MODEL_ARCHIVE_MAP,
+        TF_OPENAI_GPT_PRETRAINED_MODEL_ARCHIVE_MAP,
+        TF_TRANSFO_XL_PRETRAINED_MODEL_ARCHIVE_MAP,
+        TF_GPT2_PRETRAINED_MODEL_ARCHIVE_MAP,
+        TF_CTRL_PRETRAINED_MODEL_ARCHIVE_MAP,
+        TF_XLNET_PRETRAINED_MODEL_ARCHIVE_MAP,
+        TF_XLM_PRETRAINED_MODEL_ARCHIVE_MAP,
+        TF_ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP,
+        TF_DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP,
+        TF_ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP,
+        TF_CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP,
+        TF_T5_PRETRAINED_MODEL_ARCHIVE_MAP,
+        ]
+    for key, value, in pretrained_map.items())
+
+
 class TFAutoModel(object):
     r"""
         :class:`~transformers.TFAutoModel` is a generic model class

From db0a9ee6e0ddcb9d634c3ab0ba3d25501c370d8c Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Mon, 16 Dec 2019 14:08:08 +0100
Subject: [PATCH 376/505] adding albert to TF auto models cc @LysandreJik

---
 transformers/modeling_tf_auto.py | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/transformers/modeling_tf_auto.py b/transformers/modeling_tf_auto.py
index 9c687d9235..3e9b4d120b 100644
--- a/transformers/modeling_tf_auto.py
+++ b/transformers/modeling_tf_auto.py
@@ -27,6 +27,7 @@ from .modeling_tf_xlm import TFXLMModel, TFXLMWithLMHeadModel, TFXLMForSequenceC
 from .modeling_tf_roberta import TFRobertaModel, TFRobertaForMaskedLM, TFRobertaForSequenceClassification, TF_ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP
 from .modeling_tf_distilbert import TFDistilBertModel, TFDistilBertForQuestionAnswering, TFDistilBertForMaskedLM, TFDistilBertForSequenceClassification, TF_DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP
 from .modeling_tf_ctrl import TFCTRLModel, TFCTRLLMHeadModel, TF_CTRL_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_tf_albert import TFAlbertModel, TFAlbertForMaskedLM, TFAlbertForSequenceClassification, TF_ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP
 from .modeling_tf_t5 import TFT5Model, TFT5WithLMHeadModel, TF_T5_PRETRAINED_MODEL_ARCHIVE_MAP
 
 from .file_utils import add_start_docstrings
@@ -46,7 +47,6 @@ TF_ALL_PRETRAINED_MODEL_ARCHIVE_MAP = dict((key, value)
         TF_ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP,
         TF_DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP,
         TF_ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP,
-        TF_CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP,
         TF_T5_PRETRAINED_MODEL_ARCHIVE_MAP,
         ]
     for key, value, in pretrained_map.items())
@@ -162,6 +162,8 @@ class TFAutoModel(object):
             return TFT5Model.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'distilbert' in pretrained_model_name_or_path:
             return TFDistilBertModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'albert' in pretrained_model_name_or_path:
+            return TFAlbertModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return TFRobertaModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'bert' in pretrained_model_name_or_path:
@@ -298,6 +300,8 @@ class TFAutoModelWithLMHead(object):
             return TFT5WithLMHeadModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'distilbert' in pretrained_model_name_or_path:
             return TFDistilBertForMaskedLM.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'albert' in pretrained_model_name_or_path:
+            return TFAlbertForMaskedLM.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return TFRobertaForMaskedLM.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'bert' in pretrained_model_name_or_path:
@@ -425,6 +429,8 @@ class TFAutoModelForSequenceClassification(object):
         """
         if 'distilbert' in pretrained_model_name_or_path:
             return TFDistilBertForSequenceClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'albert' in pretrained_model_name_or_path:
+            return TFAlbertForSequenceClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return TFRobertaForSequenceClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'bert' in pretrained_model_name_or_path:

From 031ad4eb3780437d5232392b16891078b1b32d2c Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Mon, 16 Dec 2019 14:20:57 +0100
Subject: [PATCH 377/505] improving JSON error messages (for model card and
 configurations)

---
 transformers/configuration_utils.py | 15 +++++++++++----
 transformers/model_card.py          | 12 ++++++++----
 2 files changed, 19 insertions(+), 8 deletions(-)

diff --git a/transformers/configuration_utils.py b/transformers/configuration_utils.py
index 6c9eeea175..f692c9b132 100644
--- a/transformers/configuration_utils.py
+++ b/transformers/configuration_utils.py
@@ -151,10 +151,14 @@ class PretrainedConfig(object):
             config_file = pretrained_model_name_or_path
         else:
             config_file = hf_bucket_url(pretrained_model_name_or_path, postfix=CONFIG_NAME)
-        # redirect to the cache, if necessary
+
         try:
+            # Load from URL or cache if already cached
             resolved_config_file = cached_path(config_file, cache_dir=cache_dir, force_download=force_download,
                                                proxies=proxies, resume_download=resume_download)
+            # Load config
+            config = cls.from_json_file(resolved_config_file)
+
         except EnvironmentError:
             if pretrained_model_name_or_path in cls.pretrained_config_archive_map:
                 msg = "Couldn't reach server at '{}' to download pretrained model configuration file.".format(
@@ -168,15 +172,18 @@ class PretrainedConfig(object):
                         config_file, CONFIG_NAME)
             raise EnvironmentError(msg)
 
+        except json.JSONDecodeError:
+            msg = "Couldn't reach server at '{}' to download configuration file or " \
+                  "configuration file is not a valid JSON file. " \
+                  "Please check network or file content here: {}.".format(config_file, resolved_config_file)
+            raise EnvironmentError(msg)
+
         if resolved_config_file == config_file:
             logger.info("loading configuration file {}".format(config_file))
         else:
             logger.info("loading configuration file {} from cache at {}".format(
                 config_file, resolved_config_file))
 
-        # Load config
-        config = cls.from_json_file(resolved_config_file)
-
         if hasattr(config, 'pruned_heads'):
             config.pruned_heads = dict((int(key), value) for key, value in config.pruned_heads.items())
 
diff --git a/transformers/model_card.py b/transformers/model_card.py
index 6d56089844..3c775ab7fc 100644
--- a/transformers/model_card.py
+++ b/transformers/model_card.py
@@ -132,7 +132,7 @@ class ModelCard(object):
         if pretrained_model_name_or_path in ALL_PRETRAINED_CONFIG_ARCHIVE_MAP:
             # For simplicity we use the same pretrained url than the configuration files but with a different suffix (model_card.json)
             model_card_file = ALL_PRETRAINED_CONFIG_ARCHIVE_MAP[pretrained_model_name_or_path]
-            model_card_file.replace(CONFIG_NAME, MODEL_CARD_NAME)
+            model_card_file = model_card_file.replace(CONFIG_NAME, MODEL_CARD_NAME)
         elif os.path.isdir(pretrained_model_name_or_path):
             model_card_file = os.path.join(pretrained_model_name_or_path, MODEL_CARD_NAME)
         elif os.path.isfile(pretrained_model_name_or_path) or is_remote_url(pretrained_model_name_or_path):
@@ -143,13 +143,11 @@ class ModelCard(object):
         try:
             resolved_model_card_file = cached_path(model_card_file, cache_dir=cache_dir, force_download=force_download,
                                                proxies=proxies, resume_download=resume_download)
-
             if resolved_model_card_file == model_card_file:
                 logger.info("loading model card file {}".format(model_card_file))
             else:
                 logger.info("loading model card file {} from cache at {}".format(
                     model_card_file, resolved_model_card_file))
-
             # Load model card
             model_card = cls.from_json_file(resolved_model_card_file)
 
@@ -164,9 +162,15 @@ class ModelCard(object):
                         pretrained_model_name_or_path,
                         ', '.join(ALL_PRETRAINED_CONFIG_ARCHIVE_MAP.keys()),
                         model_card_file, MODEL_CARD_NAME))
-
             logger.warning("Creating an empty model card.")
+            # We fall back on creating an empty model card
+            model_card = cls()
 
+        except json.JSONDecodeError:
+            logger.warning("Couldn't reach server at '{}' to download model card file or "
+                           "model card file is not a valid JSON file. "
+                           "Please check network or file content here: {}.".format(model_card_file, resolved_model_card_file))
+            logger.warning("Creating an empty model card.")
             # We fall back on creating an empty model card
             model_card = cls()
 

From 955d7ecb570b178187075c7c31fcd9be2e3a3428 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Mon, 16 Dec 2019 14:34:54 +0100
Subject: [PATCH 378/505] Refactored Pipeline with dedicated argument handler.

---
 transformers/pipelines.py | 210 ++++++++++++++++++++------------------
 1 file changed, 112 insertions(+), 98 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 7383222c1f..7e2b30ba3c 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -36,29 +36,40 @@ if is_torch_available():
         AutoModelForQuestionAnswering, AutoModelForTokenClassification
 
 
-class Pipeline(ABC):
-    def __init__(self, model, tokenizer: PreTrainedTokenizer = None, **kwargs):
-        self.model = model
-        self.tokenizer = tokenizer
+class ArgumentHandler(ABC):
+    """
+    Base interface for handling varargs for each Pipeline
+    """
+    @abstractmethod
+    def __call__(self, *args, **kwargs):
+        raise NotImplementedError()
 
-    def save_pretrained(self, save_directory):
-        if not os.path.isdir(save_directory):
-            logger.error("Provided path ({}) should be a directory".format(save_directory))
-            return
 
-        self.model.save_pretrained(save_directory)
-        self.tokenizer.save_pretrained(save_directory)
+class DefaultArgumentHandler(ArgumentHandler):
+    """
+    Default varargs argument parser handling parameters for each Pipeline
+    """
+    def __call__(self, *args, **kwargs):
+        if 'X' in kwargs:
+            return kwargs['X']
+        elif 'data' in kwargs:
+            return kwargs['data']
+        elif len(args) > 0:
+            return list(args)
+        raise ValueError('Unable to infer the format of the provided data (X=, data=, ...)')
 
-    def transform(self, *texts, **kwargs):
-        # Generic compatibility with sklearn and Keras
-        return self(*texts, **kwargs)
 
-    def predict(self, *texts, **kwargs):
-        # Generic compatibility with sklearn and Keras
-        return self(*texts, **kwargs)
+class _ScikitCompat(ABC):
+    """
+    Interface layer for the Scikit and Keras compatibility.
+    """
 
     @abstractmethod
-    def __call__(self, *texts, **kwargs):
+    def transform(self, X):
+        raise NotImplementedError()
+
+    @abstractmethod
+    def predict(self, X):
         raise NotImplementedError()
 
 
@@ -133,24 +144,45 @@ class JsonPipelineDataFormat(PipelineDataFormat):
             if self.is_multi_columns:
                 yield {k: entry[c] for k, c in self.column}
             else:
-                yield entry[self.column]
+                yield entry[self.column[0]]
 
     def save(self, data: dict):
         with open(self.output, 'w') as f:
             json.dump(data, f)
 
 
-class FeatureExtractionPipeline(Pipeline):
+class Pipeline(_ScikitCompat):
+    def __init__(self, model, tokenizer: PreTrainedTokenizer = None, args_parser: ArgumentHandler = None, **kwargs):
+        self.model = model
+        self.tokenizer = tokenizer
+        self._args_parser = args_parser or DefaultArgumentHandler()
+
+    def save_pretrained(self, save_directory):
+        if not os.path.isdir(save_directory):
+            logger.error("Provided path ({}) should be a directory".format(save_directory))
+            return
+
+        self.model.save_pretrained(save_directory)
+        self.tokenizer.save_pretrained(save_directory)
+
+    def transform(self, X):
+        return self(X=X)
+
+    def predict(self, X):
+        return self(X=X)
 
     def __call__(self, *texts, **kwargs):
-        # Generic compatibility with sklearn and Keras
-        if 'X' in kwargs and not texts:
-            texts = kwargs.pop('X')
+        # Parse arguments
+        inputs = self._args_parser(*texts, **kwargs)
 
+        # Encode for forward
         inputs = self.tokenizer.batch_encode_plus(
-            texts, add_special_tokens=True, return_tensors='tf' if is_tf_available() else 'pt'
+            inputs, add_special_tokens=True, return_tensors='tf' if is_tf_available() else 'pt'
         )
 
+        return self._forward(inputs)
+
+    def _forward(self, inputs):
         if is_tf_available():
             # TODO trace model
             predictions = self.model(inputs)[0]
@@ -159,7 +191,12 @@ class FeatureExtractionPipeline(Pipeline):
             with torch.no_grad():
                 predictions = self.model(**inputs)[0]
 
-        return predictions.numpy().tolist()
+        return predictions.numpy()
+
+
+class FeatureExtractionPipeline(Pipeline):
+    def __call__(self, *args, **kwargs):
+        return super().__call__(*args, **kwargs).tolist()
 
 
 class TextClassificationPipeline(Pipeline):
@@ -170,26 +207,8 @@ class TextClassificationPipeline(Pipeline):
             raise Exception('Invalid parameter nb_classes. int >= 2 is required (got: {})'.format(nb_classes))
         self._nb_classes = nb_classes
 
-    def __call__(self, *texts, **kwargs):
-        # Generic compatibility with sklearn and Keras
-        if 'X' in kwargs and not texts:
-            texts = kwargs.pop('X')
-
-        inputs = self.tokenizer.batch_encode_plus(
-            texts, add_special_tokens=True, return_tensors='tf' if is_tf_available() else 'pt'
-        )
-
-        special_tokens_mask = inputs.pop('special_tokens_mask')
-
-        if is_tf_available():
-            # TODO trace model
-            predictions = self.model(**inputs)[0]
-        else:
-            import torch
-            with torch.no_grad():
-                predictions = self.model(**inputs)[0]
-
-        return predictions.numpy().tolist()
+    def __call__(self, *args, **kwargs):
+        return super().__call__(*args, **kwargs).tolist()
 
 
 class NerPipeline(Pipeline):
@@ -198,8 +217,7 @@ class NerPipeline(Pipeline):
         super().__init__(model, tokenizer)
 
     def __call__(self, *texts, **kwargs):
-        (texts, ), answers = texts, []
-
+        inputs, answers = self._args_parser(*texts, **kwargs), []
         for sentence in texts:
 
             # Ugly token to word idx mapping (for now)
@@ -241,9 +259,52 @@ class QuestionAnsweringPipeline(Pipeline):
     Question Answering pipeline involving Tokenization and Inference.
     """
 
-    @classmethod
-    def from_config(cls, model, tokenizer: PreTrainedTokenizer, **kwargs):
-        pass
+    class QuestionAnsweringArgumentHandler(ArgumentHandler):
+
+        def __call__(self, *args, **kwargs):
+            # Position args, handling is sensibly the same as X and data, so forwarding to avoid duplicating
+            if args is not None and len(args) > 1:
+                kwargs['X'] = args
+
+            # Generic compatibility with sklearn and Keras
+            # Batched data
+            if 'X' in kwargs or 'data' in kwargs:
+                data = kwargs['X'] if 'X' in kwargs else kwargs['data']
+
+                if not isinstance(data, list):
+                    data = [data]
+
+                for i, item in enumerate(data):
+                    if isinstance(item, dict):
+                        if any(k not in item for k in ['question', 'context']):
+                            raise KeyError('You need to provide a dictionary with keys {question:..., context:...}')
+                        data[i] = QuestionAnsweringPipeline.create_sample(**item)
+
+                    elif isinstance(item, SquadExample):
+                        continue
+                    else:
+                        raise ValueError(
+                            '{} argument needs to be of type (list[SquadExample | dict], SquadExample, dict)'
+                                .format('X' if 'X' in kwargs else 'data')
+                        )
+                inputs = data
+
+            # Tabular input
+            elif 'question' in kwargs and 'context' in kwargs:
+                if isinstance(kwargs['question'], str):
+                    kwargs['question'] = [kwargs['question']]
+
+                if isinstance(kwargs['context'], str):
+                    kwargs['context'] = [kwargs['context']]
+
+                inputs = [QuestionAnsweringPipeline.create_sample(q, c) for q, c in zip(kwargs['question'], kwargs['context'])]
+            else:
+                raise ValueError('Unknown arguments {}'.format(kwargs))
+
+            if not isinstance(inputs, list):
+                inputs = [inputs]
+
+            return inputs
 
     @staticmethod
     def create_sample(question: Union[str, List[str]], context: Union[str, List[str]]) -> Union[SquadExample, List[SquadExample]]:
@@ -254,54 +315,8 @@ class QuestionAnsweringPipeline(Pipeline):
         else:
             return SquadExample(None, question, context, None, None, None)
 
-    @staticmethod
-    def handle_args(*inputs, **kwargs) -> List[SquadExample]:
-        # Position args, handling is sensibly the same as X and data, so forwarding to avoid duplicating
-        if inputs is not None and len(inputs) > 1:
-            kwargs['X'] = inputs
-
-        # Generic compatibility with sklearn and Keras
-        # Batched data
-        if 'X' in kwargs or 'data' in kwargs:
-            data = kwargs['X'] if 'X' in kwargs else kwargs['data']
-
-            if not isinstance(data, list):
-                data = [data]
-
-            for i, item in enumerate(data):
-                if isinstance(item, dict):
-                    if any(k not in item for k in ['question', 'context']):
-                        raise KeyError('You need to provide a dictionary with keys {question:..., context:...}')
-                    data[i] = QuestionAnsweringPipeline.create_sample(**item)
-
-                elif isinstance(item, SquadExample):
-                    continue
-                else:
-                    raise ValueError(
-                        '{} argument needs to be of type (list[SquadExample | dict], SquadExample, dict)'
-                        .format('X' if 'X' in kwargs else 'data')
-                    )
-            inputs = data
-
-        # Tabular input
-        elif 'question' in kwargs and 'context' in kwargs:
-            if isinstance(kwargs['question'], str):
-                kwargs['question'] = [kwargs['question']]
-
-            if isinstance(kwargs['context'], str):
-                kwargs['context'] = [kwargs['context']]
-
-            inputs = [QuestionAnsweringPipeline.create_sample(q, c) for q, c in zip(kwargs['question'], kwargs['context'])]
-        else:
-            raise ValueError('Unknown arguments {}'.format(kwargs))
-
-        if not isinstance(inputs, list):
-            inputs = [inputs]
-
-        return inputs
-
     def __init__(self, model, tokenizer: Optional[PreTrainedTokenizer]):
-        super().__init__(model, tokenizer)
+        super().__init__(model, tokenizer, args_parser=QuestionAnsweringPipeline.QuestionAnsweringArgumentHandler())
 
     def inputs_for_model(self, features: Union[SquadExample, List[SquadExample]]) -> Dict:
         args = ['input_ids', 'attention_mask']
@@ -332,9 +347,8 @@ class QuestionAnsweringPipeline(Pipeline):
         if kwargs['max_answer_len'] < 1:
             raise ValueError('max_answer_len parameter should be >= 1 (got {})'.format(kwargs['max_answer_len']))
 
-        examples = QuestionAnsweringPipeline.handle_args(texts, **kwargs)
-
         # Convert inputs to features
+        examples = self._args_parser(*texts, **kwargs)
         features = squad_convert_examples_to_features(examples, self.tokenizer, kwargs['max_seq_len'], kwargs['doc_stride'], kwargs['max_question_len'], False)
         fw_args = self.inputs_for_model(features)
 

From 1bbdbacd5bc7281dbcebfe4330a464a7ad1a6e72 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Mon, 16 Dec 2019 14:38:20 +0100
Subject: [PATCH 379/505] update __init__ and saving

---
 transformers/__init__.py   |  2 +-
 transformers/model_card.py | 19 +++++++++++--------
 2 files changed, 12 insertions(+), 9 deletions(-)

diff --git a/transformers/__init__.py b/transformers/__init__.py
index 0b343bed2b..44447c5495 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -19,7 +19,7 @@ logger = logging.getLogger(__name__)  # pylint: disable=invalid-name
 # Files and general utilities
 from .file_utils import (TRANSFORMERS_CACHE, PYTORCH_TRANSFORMERS_CACHE, PYTORCH_PRETRAINED_BERT_CACHE,
                          cached_path, add_start_docstrings, add_end_docstrings,
-                         WEIGHTS_NAME, TF2_WEIGHTS_NAME, TF_WEIGHTS_NAME, CONFIG_NAME,
+                         WEIGHTS_NAME, TF2_WEIGHTS_NAME, TF_WEIGHTS_NAME, CONFIG_NAME, MODEL_CARD_NAME,
                          is_tf_available, is_torch_available)
 
 from .data import (is_sklearn_available,
diff --git a/transformers/model_card.py b/transformers/model_card.py
index 3c775ab7fc..baec7e8622 100644
--- a/transformers/model_card.py
+++ b/transformers/model_card.py
@@ -67,14 +67,14 @@ class ModelCard(object):
                 logger.error("Can't set {} with value {} for {}".format(key, value, self))
                 raise err
 
-    def save_pretrained(self, save_directory):
-        """ Save a model card object to the directory `save_directory`, so that it
-            can be re-loaded using the :func:`~transformers.ModelCard.from_pretrained` class method.
+    def save_pretrained(self, save_directory_or_file):
+        """ Save a model card object to the directory or file `save_directory_or_file`.
         """
-        assert os.path.isdir(save_directory), "Saving path should be a directory where the model card can be saved"
-
-        # If we save using the predefined names, we can load using `from_pretrained`
-        output_model_card_file = os.path.join(save_directory, MODEL_CARD_NAME)
+        if os.path.isdir(save_directory_or_file):
+            # If we save using the predefined names, we can load using `from_pretrained`
+            output_model_card_file = os.path.join(save_directory_or_file, MODEL_CARD_NAME)
+        else:
+            output_model_card_file = save_directory_or_file
 
         self.to_json_file(output_model_card_file)
         logger.info("Model card saved in {}".format(output_model_card_file))
@@ -139,8 +139,9 @@ class ModelCard(object):
             model_card_file = pretrained_model_name_or_path
         else:
             model_card_file = hf_bucket_url(pretrained_model_name_or_path, postfix=MODEL_CARD_NAME)
-        # redirect to the cache, if necessary
+
         try:
+            # Load from URL or cache if already cached
             resolved_model_card_file = cached_path(model_card_file, cache_dir=cache_dir, force_download=force_download,
                                                proxies=proxies, resume_download=resume_download)
             if resolved_model_card_file == model_card_file:
@@ -163,6 +164,7 @@ class ModelCard(object):
                         ', '.join(ALL_PRETRAINED_CONFIG_ARCHIVE_MAP.keys()),
                         model_card_file, MODEL_CARD_NAME))
             logger.warning("Creating an empty model card.")
+
             # We fall back on creating an empty model card
             model_card = cls()
 
@@ -171,6 +173,7 @@ class ModelCard(object):
                            "model card file is not a valid JSON file. "
                            "Please check network or file content here: {}.".format(model_card_file, resolved_model_card_file))
             logger.warning("Creating an empty model card.")
+
             # We fall back on creating an empty model card
             model_card = cls()
 

From 9c391277cc380b1d1eba17fd7b3337c90b35987e Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Mon, 16 Dec 2019 15:19:13 +0100
Subject: [PATCH 380/505] Allow tensors placement on specific device through
 CLI and pipeline.

---
 transformers/commands/run.py |  3 +-
 transformers/pipelines.py    | 77 ++++++++++++++++++++++++------------
 2 files changed, 54 insertions(+), 26 deletions(-)

diff --git a/transformers/commands/run.py b/transformers/commands/run.py
index bcbb87391d..b4951b1bc2 100644
--- a/transformers/commands/run.py
+++ b/transformers/commands/run.py
@@ -16,7 +16,7 @@ def try_infer_format_from_ext(path: str):
 
 
 def run_command_factory(args):
-    nlp = pipeline(task=args.task, model=args.model, tokenizer=args.tokenizer)
+    nlp = pipeline(task=args.task, model=args.model, tokenizer=args.tokenizer, device=args.device)
     format = try_infer_format_from_ext(args.input) if args.format == 'infer' else args.format
     reader = PipelineDataFormat.from_str(format, args.output, args.input, args.column)
     return RunCommand(nlp, reader)
@@ -31,6 +31,7 @@ class RunCommand(BaseTransformersCLICommand):
     @staticmethod
     def register_subcommand(parser: ArgumentParser):
         run_parser = parser.add_parser('run', help="Run a pipeline through the CLI")
+        run_parser.add_argument('--device', type=int, default=-1, help='Indicate the device to run onto, -1 indicates CPU, >= 0 indicates GPU')
         run_parser.add_argument('--task', choices=SUPPORTED_TASKS.keys(), help='Task to run')
         run_parser.add_argument('--model', type=str, required=True, help='Name or path to the model to instantiate.')
         run_parser.add_argument('--tokenizer', type=str, help='Name of the tokenizer to use. (default: same as the model name)')
diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 7e2b30ba3c..5b0e81957b 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -18,6 +18,7 @@ import csv
 import json
 import os
 from abc import ABC, abstractmethod
+from contextlib import contextmanager
 from itertools import groupby
 from typing import Union, Optional, Tuple, List, Dict
 
@@ -152,11 +153,18 @@ class JsonPipelineDataFormat(PipelineDataFormat):
 
 
 class Pipeline(_ScikitCompat):
-    def __init__(self, model, tokenizer: PreTrainedTokenizer = None, args_parser: ArgumentHandler = None, **kwargs):
+    def __init__(self, model, tokenizer: PreTrainedTokenizer = None,
+                 args_parser: ArgumentHandler = None, device: int = -1, **kwargs):
+
         self.model = model
         self.tokenizer = tokenizer
+        self.device = device
         self._args_parser = args_parser or DefaultArgumentHandler()
 
+        # Special handling
+        if self.device >= 0 and not is_tf_available():
+            self.model = self.model.to('cuda:{}'.format(self.device))
+
     def save_pretrained(self, save_directory):
         if not os.path.isdir(save_directory):
             logger.error("Provided path ({}) should be a directory".format(save_directory))
@@ -176,11 +184,25 @@ class Pipeline(_ScikitCompat):
         inputs = self._args_parser(*texts, **kwargs)
 
         # Encode for forward
-        inputs = self.tokenizer.batch_encode_plus(
-            inputs, add_special_tokens=True, return_tensors='tf' if is_tf_available() else 'pt'
-        )
+        with self.device_placement():
+            inputs = self.tokenizer.batch_encode_plus(
+                inputs, add_special_tokens=True, return_tensors='tf' if is_tf_available() else 'pt'
+            )
 
-        return self._forward(inputs)
+            return self._forward(inputs)
+
+    @contextmanager
+    def device_placement(self):
+        if is_tf_available():
+            import tensorflow as tf
+            with tf.device('/CPU:0' if self.device == -1 else '/device:GPU:{}'.format(self.device)):
+                yield
+        else:
+            import torch
+            if self.device >= 0:
+                torch.cuda.set_device(self.device)
+
+            yield
 
     def _forward(self, inputs):
         if is_tf_available():
@@ -225,14 +247,17 @@ class NerPipeline(Pipeline):
             for i, w in enumerate(words):
                 tokens = self.tokenizer.tokenize(w)
                 token_to_word += [i] * len(tokens)
-            tokens = self.tokenizer.encode_plus(sentence, return_attention_mask=False, return_tensors='tf' if is_tf_available() else 'pt')
 
-            # Forward
-            if is_torch_available():
-                with torch.no_grad():
-                    entities = self.model(**tokens)[0][0].cpu().numpy()
-            else:
-                entities = self.model(tokens)[0][0].numpy()
+            # Manage correct placement of the tensors
+            with self.device_placement():
+                tokens = self.tokenizer.encode_plus(sentence, return_attention_mask=False, return_tensors='tf' if is_tf_available() else 'pt')
+
+                # Forward
+                if is_torch_available():
+                    with torch.no_grad():
+                        entities = self.model(**tokens)[0][0].cpu().numpy()
+                else:
+                    entities = self.model(tokens)[0][0].numpy()
 
             # Normalize scores
             answer, token_start = [], 1
@@ -352,18 +377,20 @@ class QuestionAnsweringPipeline(Pipeline):
         features = squad_convert_examples_to_features(examples, self.tokenizer, kwargs['max_seq_len'], kwargs['doc_stride'], kwargs['max_question_len'], False)
         fw_args = self.inputs_for_model(features)
 
-        if is_tf_available():
-            import tensorflow as tf
-            fw_args = {k: tf.constant(v) for (k, v) in fw_args.items()}
-            start, end = self.model(fw_args)
-            start, end = start.numpy(), end.numpy()
-        else:
-            import torch
-            with torch.no_grad():
-                # Retrieve the score for the context tokens only (removing question tokens)
-                fw_args = {k: torch.tensor(v) for (k, v) in fw_args.items()}
-                start, end = self.model(**fw_args)
-                start, end = start.cpu().numpy(), end.cpu().numpy()
+        # Manage tensor allocation on correct device
+        with self.device_placement():
+            if is_tf_available():
+                import tensorflow as tf
+                fw_args = {k: tf.constant(v) for (k, v) in fw_args.items()}
+                start, end = self.model(fw_args)
+                start, end = start.numpy(), end.numpy()
+            else:
+                import torch
+                with torch.no_grad():
+                    # Retrieve the score for the context tokens only (removing question tokens)
+                    fw_args = {k: torch.tensor(v) for (k, v) in fw_args.items()}
+                    start, end = self.model(**fw_args)
+                    start, end = start.cpu().numpy(), end.cpu().numpy()
 
         answers = []
         for (example, feature, start_, end_) in zip(examples, features, start, end):
@@ -374,7 +401,7 @@ class QuestionAnsweringPipeline(Pipeline):
             # Mask padding and question
             start_, end_ = start_ * np.abs(np.array(feature.p_mask) - 1), end_ * np.abs(np.array(feature.p_mask) - 1)
 
-            # TODO : What happend if not possible
+            # TODO : What happens if not possible
             # Mask CLS
             start_[0] = end_[0] = 0
 

From bbc707cf394776634bd433c895a9223fe9b256a9 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Mon, 16 Dec 2019 15:49:09 +0100
Subject: [PATCH 381/505] Fix non-keyworded varargs handling in
 DefaultArgumentHandler for pipeline.

---
 transformers/pipelines.py | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 5b0e81957b..ee551893b0 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -55,7 +55,12 @@ class DefaultArgumentHandler(ArgumentHandler):
             return kwargs['X']
         elif 'data' in kwargs:
             return kwargs['data']
-        elif len(args) > 0:
+        elif len(args) == 1:
+            if isinstance(args[0], list):
+                return args[0]
+            else:
+                return [args[0]]
+        elif len(args) > 1:
             return list(args)
         raise ValueError('Unable to infer the format of the provided data (X=, data=, ...)')
 
@@ -240,7 +245,7 @@ class NerPipeline(Pipeline):
 
     def __call__(self, *texts, **kwargs):
         inputs, answers = self._args_parser(*texts, **kwargs), []
-        for sentence in texts:
+        for sentence in inputs:
 
             # Ugly token to word idx mapping (for now)
             token_to_word, words = [], sentence.split(' ')

From 46ccbb42fc89876461995e0ba553d72cffa700ce Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Mon, 16 Dec 2019 15:49:41 +0100
Subject: [PATCH 382/505] Make CLI run command use integer mapping for device
 argument.

---
 transformers/commands/run.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/commands/run.py b/transformers/commands/run.py
index b4951b1bc2..8c203699a8 100644
--- a/transformers/commands/run.py
+++ b/transformers/commands/run.py
@@ -31,7 +31,7 @@ class RunCommand(BaseTransformersCLICommand):
     @staticmethod
     def register_subcommand(parser: ArgumentParser):
         run_parser = parser.add_parser('run', help="Run a pipeline through the CLI")
-        run_parser.add_argument('--device', type=int, default=-1, help='Indicate the device to run onto, -1 indicates CPU, >= 0 indicates GPU')
+        run_parser.add_argument('--device', type=int, default=-1, help='Indicate the device to run onto, -1 indicates CPU, >= 0 indicates GPU (default: -1)')
         run_parser.add_argument('--task', choices=SUPPORTED_TASKS.keys(), help='Task to run')
         run_parser.add_argument('--model', type=str, required=True, help='Name or path to the model to instantiate.')
         run_parser.add_argument('--tokenizer', type=str, help='Name of the tokenizer to use. (default: same as the model name)')

From 43a4e1bbe4091b11d7926b93250cc87fa75bd545 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Mon, 16 Dec 2019 16:00:41 +0100
Subject: [PATCH 383/505] Adressing issue in varargs handling for question
 answering.

---
 transformers/pipelines.py | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index ee551893b0..2a8f26b03e 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -293,8 +293,11 @@ class QuestionAnsweringPipeline(Pipeline):
 
         def __call__(self, *args, **kwargs):
             # Position args, handling is sensibly the same as X and data, so forwarding to avoid duplicating
-            if args is not None and len(args) > 1:
-                kwargs['X'] = args
+            if args is not None and len(args) > 0:
+                if len(args) == 1:
+                    kwargs['X'] = args[0]
+                else:
+                    kwargs['X'] = list(args)
 
             # Generic compatibility with sklearn and Keras
             # Batched data

From 71b47505175111dd391a5b9de9514fbe50558bf0 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan.schweter@bsb-muenchen.de>
Date: Mon, 16 Dec 2019 16:37:27 +0100
Subject: [PATCH 384/505] examples: add support for XLM-RoBERTa to run_ner
 script

---
 examples/run_ner.py | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/examples/run_ner.py b/examples/run_ner.py
index 1ab1236d94..6426a6d1db 100644
--- a/examples/run_ner.py
+++ b/examples/run_ner.py
@@ -38,11 +38,13 @@ from transformers import WEIGHTS_NAME, BertConfig, BertForTokenClassification, B
 from transformers import RobertaConfig, RobertaForTokenClassification, RobertaTokenizer
 from transformers import DistilBertConfig, DistilBertForTokenClassification, DistilBertTokenizer
 from transformers import CamembertConfig, CamembertForTokenClassification, CamembertTokenizer
+from transformers import XLMRobertaConfig, XLMRobertaForTokenClassification, XLMRobertaTokenizer
 
 logger = logging.getLogger(__name__)
 
 ALL_MODELS = sum(
-    (tuple(conf.pretrained_config_archive_map.keys()) for conf in (BertConfig, RobertaConfig, DistilBertConfig)),
+    (tuple(conf.pretrained_config_archive_map.keys()) for conf in (BertConfig, RobertaConfig, DistilBertConfig,
+                                                                   CamembertConfig, XLMRobertaConfig)),
     ())
 
 MODEL_CLASSES = {
@@ -50,6 +52,7 @@ MODEL_CLASSES = {
     "roberta": (RobertaConfig, RobertaForTokenClassification, RobertaTokenizer),
     "distilbert": (DistilBertConfig, DistilBertForTokenClassification, DistilBertTokenizer),
     "camembert": (CamembertConfig, CamembertForTokenClassification, CamembertTokenizer),
+    "xlmroberta": (XLMRobertaConfig, XLMRobertaForTokenClassification, XLMRobertaTokenizer),
 }
 
 

From a096e2a88beae06c3341bc502b122d77be72571b Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Mon, 16 Dec 2019 16:38:02 +0100
Subject: [PATCH 385/505] WIP serving through HTTP internally using pipelines.

---
 transformers/commands/serving.py | 67 ++++++++++----------------------
 1 file changed, 21 insertions(+), 46 deletions(-)

diff --git a/transformers/commands/serving.py b/transformers/commands/serving.py
index 0b47246ead..a35dff0ebe 100644
--- a/transformers/commands/serving.py
+++ b/transformers/commands/serving.py
@@ -1,15 +1,15 @@
 from argparse import ArgumentParser, Namespace
 from typing import List, Optional, Union, Any
 
-import torch
 from fastapi import FastAPI, HTTPException, Body
 from logging import getLogger
 
 from pydantic import BaseModel
 from uvicorn import run
 
-from transformers import AutoModel, AutoTokenizer, AutoConfig
+from transformers import Pipeline
 from transformers.commands import BaseTransformersCLICommand
+from transformers.pipelines import SUPPORTED_TASKS, pipeline
 
 
 def serve_command_factory(args: Namespace):
@@ -17,7 +17,8 @@ def serve_command_factory(args: Namespace):
     Factory function used to instantiate serving server from provided command line arguments.
     :return: ServeCommand
     """
-    return ServeCommand(args.host, args.port, args.model, args.graphql)
+    nlp = pipeline(args.task, args.model)
+    return ServeCommand(nlp, args.host, args.port, args.model, args.graphql)
 
 
 class ServeResult(BaseModel):
@@ -53,8 +54,6 @@ class ServeForwardResult(ServeResult):
     """
     Forward result model
     """
-    tokens: List[str]
-    tokens_ids: List[int]
     output: Any
 
 
@@ -68,19 +67,18 @@ class ServeCommand(BaseTransformersCLICommand):
         :return:
         """
         serve_parser = parser.add_parser('serve', help='CLI tool to run inference requests through REST and GraphQL endpoints.')
+        serve_parser.add_argument('--task', type=str, choices=SUPPORTED_TASKS.keys(), help='The task to run the pipeline on')
+        serve_parser.add_argument('--device', type=int, default=-1, help='Indicate the device to run onto, -1 indicates CPU, >= 0 indicates GPU (default: -1)')
         serve_parser.add_argument('--host', type=str, default='localhost', help='Interface the server will listen on.')
         serve_parser.add_argument('--port', type=int, default=8888, help='Port the serving will listen to.')
         serve_parser.add_argument('--model', type=str, required=True, help='Model\'s name or path to stored model to infer from.')
         serve_parser.add_argument('--graphql', action='store_true', default=False, help='Enable GraphQL endpoints.')
         serve_parser.set_defaults(func=serve_command_factory)
 
-    def __init__(self, host: str, port: int, model: str, graphql: bool):
+    def __init__(self, pipeline: Pipeline, host: str, port: int, model: str, graphql: bool):
         self._logger = getLogger('transformers-cli/serving')
 
-        self._logger.info('Loading model {}'.format(model))
-        self._model_name = model
-        self._model = AutoModel.from_pretrained(model)
-        self._tokenizer = AutoTokenizer.from_pretrained(model)
+        self._pipeline = pipeline
 
         self._logger.info('Serving model over {}:{}'.format(host, port))
         self._host = host
@@ -97,7 +95,7 @@ class ServeCommand(BaseTransformersCLICommand):
         run(self._app, host=self._host, port=self._port)
 
     def model_info(self):
-        return ServeModelInfoResult(model=self._model_name, infos=vars(self._model.config))
+        return ServeModelInfoResult(model='', infos=vars(self._pipeline.model.config))
 
     def tokenize(self, text_input: str = Body(None, embed=True), return_ids: bool = Body(False, embed=True)):
         """
@@ -106,16 +104,16 @@ class ServeCommand(BaseTransformersCLICommand):
         - **return_ids**: Boolean flags indicating if the tokens have to be converted to their integer mapping.
         """
         try:
-            tokens_txt = self._tokenizer.tokenize(text_input)
+            tokens_txt = self._pipeline.tokenizer.tokenize(text_input)
 
             if return_ids:
-                tokens_ids = self._tokenizer.convert_tokens_to_ids(tokens_txt)
-                return ServeTokenizeResult(model=self._model_name, tokens=tokens_txt, tokens_ids=tokens_ids)
+                tokens_ids = self._pipeline.tokenizer.convert_tokens_to_ids(tokens_txt)
+                return ServeTokenizeResult(model='', tokens=tokens_txt, tokens_ids=tokens_ids)
             else:
-                return ServeTokenizeResult(model=self._model_name, tokens=tokens_txt)
+                return ServeTokenizeResult(model='', tokens=tokens_txt)
 
         except Exception as e:
-            raise HTTPException(status_code=500, detail={"model": self._model_name, "error": str(e)})
+            raise HTTPException(status_code=500, detail={"model": '', "error": str(e)})
 
     def detokenize(self, tokens_ids: List[int] = Body(None, embed=True),
                    skip_special_tokens: bool = Body(False, embed=True),
@@ -127,14 +125,12 @@ class ServeCommand(BaseTransformersCLICommand):
         - **cleanup_tokenization_spaces**: Flag indicating to remove all leading/trailing spaces and intermediate ones.
         """
         try:
-            decoded_str = self._tokenizer.decode(tokens_ids, skip_special_tokens, cleanup_tokenization_spaces)
-            return ServeDeTokenizeResult(model=self._model_name, text=decoded_str)
+            decoded_str = self._pipeline.tokenizer.decode(tokens_ids, skip_special_tokens, cleanup_tokenization_spaces)
+            return ServeDeTokenizeResult(model='', text=decoded_str)
         except Exception as e:
-            raise HTTPException(status_code=500, detail={"model": self._model_name, "error": str(e)})
+            raise HTTPException(status_code=500, detail={"model": '', "error": str(e)})
 
-    def forward(self, inputs: Union[str, List[str], List[int]] = Body(None, embed=True),
-                attention_mask: Optional[List[int]] = Body(None, embed=True),
-                tokens_type_ids: Optional[List[int]] = Body(None, embed=True)):
+    def forward(self, inputs: Union[str, dict, List[str], List[int], List[dict]] = Body(None, embed=True)):
         """
         **inputs**:
         **attention_mask**:
@@ -143,34 +139,13 @@ class ServeCommand(BaseTransformersCLICommand):
 
         # Check we don't have empty string
         if len(inputs) == 0:
-            return ServeForwardResult(model=self._model_name, output=[], attention=[])
-
-        if isinstance(inputs, str):
-            inputs_tokens = self._tokenizer.tokenize(inputs)
-            inputs_ids = self._tokenizer.convert_tokens_to_ids(inputs_tokens)
-
-        elif isinstance(inputs, List):
-            if isinstance(inputs[0], str):
-                inputs_tokens = inputs
-                inputs_ids = self._tokenizer.convert_tokens_to_ids(inputs_tokens)
-            elif isinstance(inputs[0], int):
-                inputs_tokens = []
-                inputs_ids = inputs
-            else:
-                error_msg = "inputs should be string, [str] of [int] (got {})".format(type(inputs[0]))
-                raise HTTPException(423, detail={"error": error_msg})
-        else:
-            error_msg = "inputs should be string, [str] of [int] (got {})".format(type(inputs))
-            raise HTTPException(423, detail={"error": error_msg})
+            return ServeForwardResult(model='', output=[], attention=[])
 
         try:
             # Forward through the model
-            t_input_ids = torch.tensor(inputs_ids).unsqueeze(0)
-            output = self._model(t_input_ids, attention_mask, tokens_type_ids)
-
+            output = self._pipeline(inputs)
             return ServeForwardResult(
-                model=self._model_name, tokens=inputs_tokens,
-                tokens_ids=inputs_ids, output=output[0].tolist()
+                model='', output=output
             )
         except Exception as e:
             raise HTTPException(500, {"error": str(e)})

From d3549b66af6f225cace48f8462ba715508f51b0d Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan.schweter@bsb-muenchen.de>
Date: Mon, 16 Dec 2019 16:38:39 +0100
Subject: [PATCH 386/505] module: add support for XLM-RoBERTa (__init__)

---
 transformers/__init__.py | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/transformers/__init__.py b/transformers/__init__.py
index 740d2440c2..910ba91457 100644
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -49,6 +49,7 @@ from .tokenization_distilbert import DistilBertTokenizer
 from .tokenization_albert import AlbertTokenizer
 from .tokenization_camembert import CamembertTokenizer
 from .tokenization_t5 import T5Tokenizer
+from .tokenization_xlm_roberta import XLMRobertaTokenizer
 
 # Configurations
 from .configuration_utils import PretrainedConfig
@@ -65,6 +66,7 @@ from .configuration_distilbert import DistilBertConfig, DISTILBERT_PRETRAINED_CO
 from .configuration_albert import AlbertConfig, ALBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_camembert import CamembertConfig, CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_t5 import T5Config, T5_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_xlm_roberta import XLMRobertaConfig, XLM_ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP
 
 # Modeling
 if is_torch_available():
@@ -119,6 +121,9 @@ if is_torch_available():
                                 AlbertForQuestionAnswering,
                                 load_tf_weights_in_albert, ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
 
+    from .modeling_xlm_roberta import (XLMRobertaForMaskedLM, XLMRobertaModel, XLMRobertaForMultipleChoice,
+                                       XLMRobertaForSequenceClassification, XLMRobertaForTokenClassification)
+
     # Optimization
     from .optimization import (AdamW, get_constant_schedule, get_constant_schedule_with_warmup, get_cosine_schedule_with_warmup,
                                get_cosine_with_hard_restarts_schedule_with_warmup, get_linear_schedule_with_warmup)

From 9ed09cb4a31518b13f2c58c057e43e029c32611a Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan.schweter@bsb-muenchen.de>
Date: Mon, 16 Dec 2019 16:46:58 +0100
Subject: [PATCH 387/505] converter: add conversion script for original
 XLM-RoBERTa weights to Transformers-compatible weights

---
 ..._original_pytorch_checkpoint_to_pytorch.py | 184 ++++++++++++++++++
 1 file changed, 184 insertions(+)
 create mode 100644 transformers/convert_xlm_roberta_original_pytorch_checkpoint_to_pytorch.py

diff --git a/transformers/convert_xlm_roberta_original_pytorch_checkpoint_to_pytorch.py b/transformers/convert_xlm_roberta_original_pytorch_checkpoint_to_pytorch.py
new file mode 100644
index 0000000000..888adf4819
--- /dev/null
+++ b/transformers/convert_xlm_roberta_original_pytorch_checkpoint_to_pytorch.py
@@ -0,0 +1,184 @@
+# coding=utf-8
+# Copyright 2018 The HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Convert RoBERTa checkpoint."""
+
+from __future__ import absolute_import, division, print_function
+
+import argparse
+import logging
+import numpy as np
+import torch
+import pathlib
+
+from fairseq.models.roberta import RobertaModel as FairseqRobertaModel
+from fairseq.modules import TransformerSentenceEncoderLayer
+from transformers.modeling_bert import (BertConfig, BertEncoder,
+                                        BertIntermediate, BertLayer,
+                                        BertModel, BertOutput,
+                                        BertSelfAttention,
+                                        BertSelfOutput)
+from transformers.modeling_roberta import (RobertaEmbeddings,
+                                           RobertaForMaskedLM,
+                                           RobertaForSequenceClassification,
+                                           RobertaModel)
+
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+SAMPLE_TEXT = 'Hello world! cécé herlolip'
+
+
+def convert_roberta_checkpoint_to_pytorch(roberta_checkpoint_path, pytorch_dump_folder_path, classification_head):
+    """
+    Copy/paste/tweak roberta's weights to our BERT structure.
+    """
+    roberta = FairseqRobertaModel.from_pretrained(roberta_checkpoint_path, bpe = 'sentencepiece')
+    roberta.eval()  # disable dropout
+    config = BertConfig(
+        vocab_size_or_config_json_file=250004,
+        hidden_size=roberta.args.encoder_embed_dim,
+        num_hidden_layers=roberta.args.encoder_layers,
+        num_attention_heads=roberta.args.encoder_attention_heads,
+        intermediate_size=roberta.args.encoder_ffn_embed_dim,
+        max_position_embeddings=514,
+        type_vocab_size=1,
+        layer_norm_eps=1e-5, # PyTorch default used in fairseq
+    )
+    if classification_head:
+        config.num_labels = roberta.args.num_classes
+    print("Our BERT config:", config)
+
+    model = RobertaForSequenceClassification(config) if classification_head else RobertaForMaskedLM(config)
+    model.eval()
+
+    # Now let's copy all the weights.
+    # Embeddings
+    roberta_sent_encoder = roberta.model.decoder.sentence_encoder
+    model.roberta.embeddings.word_embeddings.weight = roberta_sent_encoder.embed_tokens.weight
+    model.roberta.embeddings.position_embeddings.weight = roberta_sent_encoder.embed_positions.weight
+    model.roberta.embeddings.token_type_embeddings.weight.data = torch.zeros_like(model.roberta.embeddings.token_type_embeddings.weight)  # just zero them out b/c RoBERTa doesn't use them.
+    model.roberta.embeddings.LayerNorm.weight = roberta_sent_encoder.emb_layer_norm.weight
+    model.roberta.embeddings.LayerNorm.bias = roberta_sent_encoder.emb_layer_norm.bias
+
+    for i in range(config.num_hidden_layers):
+        # Encoder: start of layer
+        layer: BertLayer = model.roberta.encoder.layer[i]
+        roberta_layer: TransformerSentenceEncoderLayer = roberta_sent_encoder.layers[i]
+
+        ### self attention
+        self_attn: BertSelfAttention = layer.attention.self
+        assert(
+            roberta_layer.self_attn.k_proj.weight.data.shape == \
+            roberta_layer.self_attn.q_proj.weight.data.shape == \
+            roberta_layer.self_attn.v_proj.weight.data.shape == \
+            torch.Size((config.hidden_size, config.hidden_size))
+        )
+
+        self_attn.query.weight.data = roberta_layer.self_attn.q_proj.weight
+        self_attn.query.bias.data = roberta_layer.self_attn.q_proj.bias
+        self_attn.key.weight.data = roberta_layer.self_attn.k_proj.weight
+        self_attn.key.bias.data = roberta_layer.self_attn.k_proj.bias
+        self_attn.value.weight.data = roberta_layer.self_attn.v_proj.weight
+        self_attn.value.bias.data = roberta_layer.self_attn.v_proj.bias
+
+        ### self-attention output
+        self_output: BertSelfOutput = layer.attention.output
+        assert(
+            self_output.dense.weight.shape == roberta_layer.self_attn.out_proj.weight.shape
+        )
+        self_output.dense.weight = roberta_layer.self_attn.out_proj.weight
+        self_output.dense.bias = roberta_layer.self_attn.out_proj.bias
+        self_output.LayerNorm.weight = roberta_layer.self_attn_layer_norm.weight
+        self_output.LayerNorm.bias = roberta_layer.self_attn_layer_norm.bias
+
+        ### intermediate
+        intermediate: BertIntermediate = layer.intermediate
+        assert(
+            intermediate.dense.weight.shape == roberta_layer.fc1.weight.shape
+        )
+        intermediate.dense.weight = roberta_layer.fc1.weight
+        intermediate.dense.bias = roberta_layer.fc1.bias
+
+        ### output
+        bert_output: BertOutput = layer.output
+        assert(
+            bert_output.dense.weight.shape == roberta_layer.fc2.weight.shape
+        )
+        bert_output.dense.weight = roberta_layer.fc2.weight
+        bert_output.dense.bias = roberta_layer.fc2.bias
+        bert_output.LayerNorm.weight = roberta_layer.final_layer_norm.weight
+        bert_output.LayerNorm.bias = roberta_layer.final_layer_norm.bias
+        #### end of layer
+    
+    if classification_head:
+        model.classifier.dense.weight = roberta.model.classification_heads['mnli'].dense.weight
+        model.classifier.dense.bias = roberta.model.classification_heads['mnli'].dense.bias
+        model.classifier.out_proj.weight = roberta.model.classification_heads['mnli'].out_proj.weight
+        model.classifier.out_proj.bias = roberta.model.classification_heads['mnli'].out_proj.bias
+    else:
+        # LM Head
+        model.lm_head.dense.weight = roberta.model.decoder.lm_head.dense.weight
+        model.lm_head.dense.bias = roberta.model.decoder.lm_head.dense.bias
+        model.lm_head.layer_norm.weight = roberta.model.decoder.lm_head.layer_norm.weight
+        model.lm_head.layer_norm.bias = roberta.model.decoder.lm_head.layer_norm.bias
+        model.lm_head.decoder.weight = roberta.model.decoder.lm_head.weight
+        model.lm_head.bias = roberta.model.decoder.lm_head.bias
+
+    # Let's check that we get the same results.
+    input_ids: torch.Tensor = roberta.encode(SAMPLE_TEXT).unsqueeze(0) # batch of size 1
+
+    our_output = model(input_ids)[0]
+    if classification_head:
+        their_output = roberta.model.classification_heads['mnli'](roberta.extract_features(input_ids))
+    else:
+        their_output = roberta.model(input_ids)[0]
+    print(our_output.shape, their_output.shape)
+    max_absolute_diff = torch.max(torch.abs(our_output - their_output)).item()
+    print(f"max_absolute_diff = {max_absolute_diff}") # ~ 1e-7
+    success = torch.allclose(our_output, their_output, atol=1e-3)
+    print(
+        "Do both models output the same tensors?",
+        "🔥" if success else "💩"
+    )
+    if not success:
+        raise Exception("Something went wRoNg")
+
+    pathlib.Path(pytorch_dump_folder_path).mkdir(parents=True, exist_ok=True)
+    print(f"Saving model to {pytorch_dump_folder_path}")
+    model.save_pretrained(pytorch_dump_folder_path)
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    ## Required parameters
+    parser.add_argument("--roberta_checkpoint_path",
+                        default = None,
+                        type = str,
+                        required = True,
+                        help = "Path the official PyTorch dump.")
+    parser.add_argument("--pytorch_dump_folder_path",
+                        default = None,
+                        type = str,
+                        required = True,
+                        help = "Path to the output PyTorch model.")
+    parser.add_argument("--classification_head",
+                        action = "store_true",
+                        help = "Whether to convert a final classification head.")
+    args = parser.parse_args()
+    convert_roberta_checkpoint_to_pytorch(
+        args.roberta_checkpoint_path,
+        args.pytorch_dump_folder_path,
+        args.classification_head
+    )

From a648ff738c88a41cfae4f915a4391c7d66261b64 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan.schweter@bsb-muenchen.de>
Date: Mon, 16 Dec 2019 16:47:39 +0100
Subject: [PATCH 388/505] configuration: add support for XLM-RoBERTa model

---
 transformers/configuration_xlm_roberta.py | 33 +++++++++++++++++++++++
 1 file changed, 33 insertions(+)
 create mode 100644 transformers/configuration_xlm_roberta.py

diff --git a/transformers/configuration_xlm_roberta.py b/transformers/configuration_xlm_roberta.py
new file mode 100644
index 0000000000..1633cc18aa
--- /dev/null
+++ b/transformers/configuration_xlm_roberta.py
@@ -0,0 +1,33 @@
+# coding=utf-8
+# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.
+# Copyright (c) 2018, NVIDIA CORPORATION.  All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" XLM-RoBERTa configuration """
+
+from __future__ import (absolute_import, division, print_function,
+                        unicode_literals)
+
+import logging
+
+from .configuration_roberta import RobertaConfig
+
+logger = logging.getLogger(__name__)
+
+XLM_ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP = {
+    'xlm-roberta-base': "https://schweter.eu/cloud/transformers/xlm-roberta-large-config.json",
+}
+
+
+class XLMRobertaConfig(RobertaConfig):
+    pretrained_config_archive_map = XLM_ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP

From 69f4f058fa5ecc6fea8c65ae59694442bba795e6 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan.schweter@bsb-muenchen.de>
Date: Mon, 16 Dec 2019 17:00:12 +0100
Subject: [PATCH 389/505] model: add support for new XLM-RoBERTa model

---
 transformers/modeling_xlm_roberta.py | 293 +++++++++++++++++++++++++++
 1 file changed, 293 insertions(+)
 create mode 100644 transformers/modeling_xlm_roberta.py

diff --git a/transformers/modeling_xlm_roberta.py b/transformers/modeling_xlm_roberta.py
new file mode 100644
index 0000000000..8402be4b5c
--- /dev/null
+++ b/transformers/modeling_xlm_roberta.py
@@ -0,0 +1,293 @@
+# coding=utf-8
+# Copyright 2019 Facebook AI Research and the HuggingFace Inc. team.
+# Copyright (c) 2018, NVIDIA CORPORATION.  All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""PyTorch XLM-RoBERTa model. """
+
+from __future__ import (absolute_import, division, print_function,
+                        unicode_literals)
+
+import logging
+
+from .modeling_roberta import RobertaModel, RobertaForMaskedLM, RobertaForSequenceClassification, RobertaForMultipleChoice, RobertaForTokenClassification
+from .configuration_xlm_roberta import XLMRobertaConfig
+from .file_utils import add_start_docstrings
+
+logger = logging.getLogger(__name__)
+
+XLM_ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP = {
+    'xlm-roberta-large': "https://schweter.eu/cloud/transformers/xlm-roberta-large-pytorch_model.bin",
+}
+
+
+XLM_ROBERTA_START_DOCSTRING = r"""    The XLM-RoBERTa model was proposed in
+    `Unsupervised Cross-lingual Representation Learning at Scale`_
+    by Alexis Conneau, Kartikay Khandelwal, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer and Veselin Stoyanov. It is based on Facebook's RoBERTa model released in 2019.
+    
+    It is a large multi-lingual language model, trained on 2.5TB of filtered CommonCrawl data.
+
+    This implementation is the same as RoBERTa.
+
+    This model is a PyTorch `torch.nn.Module`_ sub-class. Use it as a regular PyTorch Module and
+    refer to the PyTorch documentation for all matter related to general usage and behavior.
+
+    .. _`Unsupervised Cross-lingual Representation Learning at Scale`:
+        https://arxiv.org/abs/1911.02116
+
+    .. _`torch.nn.Module`:
+        https://pytorch.org/docs/stable/nn.html#module
+
+    Parameters:
+        config (:class:`~transformers.XLMRobertaConfig`): Model configuration class with all the parameters of the 
+            model. Initializing with a config file does not load the weights associated with the model, only the configuration.
+            Check out the :meth:`~transformers.PreTrainedModel.from_pretrained` method to load the model weights.
+"""
+
+XLM_ROBERTA_INPUTS_DOCSTRING = r"""
+    Inputs:
+        **input_ids**: ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
+            Indices of input sequence tokens in the vocabulary.
+            To match pre-training, XLM-RoBERTa input sequence should be formatted with <s> and </s> tokens as follows:
+
+            (a) For sequence pairs:
+
+                ``tokens:         <s> Is this Jacksonville ? </s> </s> No it is not . </s>``
+
+            (b) For single sequences:
+
+                ``tokens:         <s> the dog is hairy . </s>``
+
+            Fully encoded sequences or sequence pairs can be obtained using the XLMRobertaTokenizer.encode function with 
+            the ``add_special_tokens`` parameter set to ``True``.
+
+            XLM-RoBERTa is a model with absolute position embeddings so it's usually advised to pad the inputs on
+            the right rather than the left.
+
+            See :func:`transformers.PreTrainedTokenizer.encode` and
+            :func:`transformers.PreTrainedTokenizer.convert_tokens_to_ids` for details.
+        **attention_mask**: (`optional`) ``torch.FloatTensor`` of shape ``(batch_size, sequence_length)``:
+            Mask to avoid performing attention on padding token indices.
+            Mask values selected in ``[0, 1]``:
+            ``1`` for tokens that are NOT MASKED, ``0`` for MASKED tokens.
+        **token_type_ids**: (`optional` need to be trained) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
+            Optional segment token indices to indicate first and second portions of the inputs.
+            This embedding matrice is not trained (not pretrained during XLM-RoBERTa pretraining), you will have to train it
+            during finetuning.
+            Indices are selected in ``[0, 1]``: ``0`` corresponds to a `sentence A` token, ``1``
+            corresponds to a `sentence B` token
+            (see `BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding`_ for more details).
+        **position_ids**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
+            Indices of positions of each input sequence tokens in the position embeddings.
+            Selected in the range ``[0, config.max_position_embeddings - 1[``.
+        **head_mask**: (`optional`) ``torch.FloatTensor`` of shape ``(num_heads,)`` or ``(num_layers, num_heads)``:
+            Mask to nullify selected heads of the self-attention modules.
+            Mask values selected in ``[0, 1]``:
+            ``1`` indicates the head is **not masked**, ``0`` indicates the head is **masked**.
+        **inputs_embeds**: (`optional`) ``torch.FloatTensor`` of shape ``(batch_size, sequence_length, embedding_dim)``:
+            Optionally, instead of passing ``input_ids`` you can choose to directly pass an embedded representation.
+            This is useful if you want more control over how to convert `input_ids` indices into associated vectors
+            than the model's internal embedding lookup matrix.
+"""
+
+@add_start_docstrings("The bare XLM-RoBERTa Model transformer outputting raw hidden-states without any specific head on top.",
+                      XLM_ROBERTA_START_DOCSTRING, XLM_ROBERTA_INPUTS_DOCSTRING)
+class XLMRobertaModel(RobertaModel):
+    r"""
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **last_hidden_state**: ``torch.FloatTensor`` of shape ``(batch_size, sequence_length, hidden_size)``
+            Sequence of hidden-states at the output of the last layer of the model.
+        **pooler_output**: ``torch.FloatTensor`` of shape ``(batch_size, hidden_size)``
+            Last layer hidden-state of the first token of the sequence (classification token)
+            further processed by a Linear layer and a Tanh activation function. The Linear
+            layer weights are trained from the next sentence prediction (classification)
+            eo match pre-training, XLM-RoBERTa input sequence should be formatted with [CLS] and [SEP] tokens as follows:
+
+            (a) For sequence pairs:
+
+                ``tokens:         [CLS] is this jack ##son ##ville ? [SEP] [SEP] no it is not . [SEP]``
+
+                ``token_type_ids:   0   0  0    0    0     0       0   0   0     1  1  1  1   1   1``
+
+            (b) For single sequences:
+
+                ``tokens:         [CLS] the dog is hairy . [SEP]``
+
+                ``token_type_ids:   0   0   0   0  0     0   0``
+
+            objective during Bert pretraining. This output is usually *not* a good summary
+            of the semantic content of the input, you're often better with averaging or pooling
+            the sequence of hidden-states for the whole input sequence.
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``torch.FloatTensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        tokenizer = XLMRobertaTokenizer.from_pretrained('xlm-roberta-large')
+        model = XLMRobertaModel.from_pretrained('xlm-roberta-large')
+        input_ids = torch.tensor(tokenizer.encode("Schloß Nymphenburg ist sehr schön .")).unsqueeze(0)  # Batch size 1
+        outputs = model(input_ids)
+        last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
+
+    """
+    config_class = XLMRobertaConfig
+    pretrained_model_archive_map = XLM_ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP
+
+
+@add_start_docstrings("""XLM-RoBERTa Model with a `language modeling` head on top. """,
+    XLM_ROBERTA_START_DOCSTRING, XLM_ROBERTA_INPUTS_DOCSTRING)
+class XLMRobertaForMaskedLM(RobertaForMaskedLM):
+    r"""
+        **masked_lm_labels**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
+            Labels for computing the masked language modeling loss.
+            Indices should be in ``[-1, 0, ..., config.vocab_size]`` (see ``input_ids`` docstring)
+            Tokens with indices set to ``-1`` are ignored (masked), the loss is only computed for the tokens with labels
+            in ``[0, ..., config.vocab_size]``
+
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **loss**: (`optional`, returned when ``masked_lm_labels`` is provided) ``torch.FloatTensor`` of shape ``(1,)``:
+            Masked language modeling loss.
+        **prediction_scores**: ``torch.FloatTensor`` of shape ``(batch_size, sequence_length, config.vocab_size)``
+            Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``torch.FloatTensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        tokenizer = XLMRobertaTokenizer.from_pretrained('xlm-roberta-large')
+        model = XLMRobertaForMaskedLM.from_pretrained('xlm-roberta-large')
+        input_ids = torch.tensor(tokenizer.encode("Schloß Nymphenburg ist sehr schön .")).unsqueeze(0)  # Batch size 1
+        outputs = model(input_ids, masked_lm_labels=input_ids)
+        loss, prediction_scores = outputs[:2]
+
+    """
+    config_class = XLMRobertaConfig
+    pretrained_model_archive_map = XLM_ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP
+
+
+@add_start_docstrings("""XLM-RoBERTa Model transformer with a sequence classification/regression head on top (a linear layer 
+    on top of the pooled output) e.g. for GLUE tasks. """,
+    XLM_ROBERTA_START_DOCSTRING, XLM_ROBERTA_INPUTS_DOCSTRING)
+class XLMRobertaForSequenceClassification(RobertaForSequenceClassification):
+    r"""
+        **labels**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size,)``:
+            Labels for computing the sequence classification/regression loss.
+            Indices should be in ``[0, ..., config.num_labels]``.
+            If ``config.num_labels == 1`` a regression loss is computed (Mean-Square loss),
+            If ``config.num_labels > 1`` a classification loss is computed (Cross-Entropy).
+
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **loss**: (`optional`, returned when ``labels`` is provided) ``torch.FloatTensor`` of shape ``(1,)``:
+            Classification (or regression if config.num_labels==1) loss.
+        **logits**: ``torch.FloatTensor`` of shape ``(batch_size, config.num_labels)``
+            Classification (or regression if config.num_labels==1) scores (before SoftMax).
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``torch.FloatTensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        tokenizer = XLMRobertaTokenizer.from_pretrained('xlm-roberta-large')
+        model = XLMRobertaForSequenceClassification.from_pretrained('xlm-roberta-large')
+        input_ids = torch.tensor(tokenizer.encode("Schloß Nymphenburg ist sehr schön .")).unsqueeze(0)  # Batch size 1
+        labels = torch.tensor([1]).unsqueeze(0)  # Batch size 1
+        outputs = model(input_ids, labels=labels)
+        loss, logits = outputs[:2]
+
+    """
+    config_class = XLMRobertaConfig
+    pretrained_model_archive_map = XLM_ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP
+
+
+@add_start_docstrings("""XLM-RoBERTa Model with a multiple choice classification head on top (a linear layer on top of
+    the pooled output and a softmax) e.g. for RocStories/SWAG tasks. """,
+    XLM_ROBERTA_START_DOCSTRING, XLM_ROBERTA_INPUTS_DOCSTRING)
+class XLMRobertaForMultipleChoice(RobertaForMultipleChoice):
+    r"""
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **loss**: (`optional`, returned when ``labels`` is provided) ``torch.FloatTensor`` of shape ``(1,)``:
+            Classification loss.
+        **classification_scores**: ``torch.FloatTensor`` of shape ``(batch_size, num_choices)`` where `num_choices` is the size of the second dimension
+            of the input tensors. (see `input_ids` above).
+            Classification scores (before SoftMax).
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``torch.FloatTensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        tokenizer = XLMRobertaTokenizer.from_pretrained('xlm-roberta-large')
+        model = XLMRobertaForMultipleChoice.from_pretrained('xlm-roberta-large')
+        choices = ["Schloß Nymphenburg ist sehr schön .", "Der Schloßkanal auch !"]
+        input_ids = torch.tensor([tokenizer.encode(s, add_special_tokens=True) for s in choices]).unsqueeze(0)  # Batch size 1, 2 choices
+        labels = torch.tensor(1).unsqueeze(0)  # Batch size 1
+        outputs = model(input_ids, labels=labels)
+        loss, classification_scores = outputs[:2]
+
+    """
+    config_class = XLMRobertaConfig
+    pretrained_model_archive_map = XLM_ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP
+
+
+@add_start_docstrings("""XLM-RoBERTa Model with a token classification head on top (a linear layer on top of
+    the hidden-states output) e.g. for Named-Entity-Recognition (NER) tasks. """,
+    XLM_ROBERTA_START_DOCSTRING, XLM_ROBERTA_INPUTS_DOCSTRING)
+class XLMRobertaForTokenClassification(RobertaForTokenClassification):
+    r"""
+        **labels**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
+            Labels for computing the token classification loss.
+            Indices should be in ``[0, ..., config.num_labels - 1]``.
+
+    Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
+        **loss**: (`optional`, returned when ``labels`` is provided) ``torch.FloatTensor`` of shape ``(1,)``:
+            Classification loss.
+        **scores**: ``torch.FloatTensor`` of shape ``(batch_size, sequence_length, config.num_labels)``
+            Classification scores (before SoftMax).
+        **hidden_states**: (`optional`, returned when ``config.output_hidden_states=True``)
+            list of ``torch.FloatTensor`` (one for the output of each layer + the output of the embeddings)
+            of shape ``(batch_size, sequence_length, hidden_size)``:
+            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+        **attentions**: (`optional`, returned when ``config.output_attentions=True``)
+            list of ``torch.FloatTensor`` (one for each layer) of shape ``(batch_size, num_heads, sequence_length, sequence_length)``:
+            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
+
+    Examples::
+
+        tokenizer = XLMRobertaTokenizer.from_pretrained('xlm-roberta-large')
+        model = XLMRobertaForTokenClassification.from_pretrained('xlm-roberta-large')
+        input_ids = torch.tensor(tokenizer.encode("Schloß Nymphenburg ist sehr schön .", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
+        labels = torch.tensor([1] * input_ids.size(1)).unsqueeze(0)  # Batch size 1
+        outputs = model(input_ids, labels=labels)
+        loss, scores = outputs[:2]
+
+    """
+    config_class = XLMRobertaConfig
+    pretrained_model_archive_map = XLM_ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP

From 59a1aefb1ca51b183bffa2d355bc2a22a7c51274 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan.schweter@bsb-muenchen.de>
Date: Mon, 16 Dec 2019 17:00:55 +0100
Subject: [PATCH 390/505] tokenization: add support for new XLM-RoBERTa model.
 Add wrapper around fairseq tokenization logic

---
 transformers/tokenization_xlm_roberta.py | 154 +++++++++++++++++++++++
 1 file changed, 154 insertions(+)
 create mode 100644 transformers/tokenization_xlm_roberta.py

diff --git a/transformers/tokenization_xlm_roberta.py b/transformers/tokenization_xlm_roberta.py
new file mode 100644
index 0000000000..0f95397606
--- /dev/null
+++ b/transformers/tokenization_xlm_roberta.py
@@ -0,0 +1,154 @@
+# coding=utf-8
+# Copyright 2018 Google AI, Google Brain and Carnegie Mellon University Authors and the HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License
+""" Tokenization classes for XLM-RoBERTa model."""
+from __future__ import (absolute_import, division, print_function,
+                        unicode_literals)
+
+import logging
+import os
+from shutil import copyfile
+
+import sentencepiece as spm
+from transformers.tokenization_utils import PreTrainedTokenizer
+
+logger = logging.getLogger(__name__)
+
+VOCAB_FILES_NAMES = {'vocab_file': 'sentencepiece.bpe.model'}
+
+PRETRAINED_VOCAB_FILES_MAP = {
+    'vocab_file':
+    {
+    'xlm-roberta-large': "https://schweter.eu/cloud/transformers/xlm-roberta-large-sentencepiece.bpe.model",
+    }
+}
+
+PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
+    'xlm-roberta-large': None,
+}
+
+class XLMRobertaTokenizer(PreTrainedTokenizer):
+    """
+        Adapted from RobertaTokenizer and XLNetTokenizer
+        SentencePiece based tokenizer. Peculiarities:
+
+            - requires `SentencePiece <https://github.com/google/sentencepiece>`_
+    """
+    vocab_files_names = VOCAB_FILES_NAMES
+    pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+    max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
+
+    def __init__(self, vocab_file, bos_token="<s>", eos_token="</s>", sep_token="</s>",
+                 cls_token="<s>", unk_token="<unk>", pad_token='<pad>', mask_token='<mask>',
+                 **kwargs):
+        super(XLMRobertaTokenizer, self).__init__(max_len=512, bos_token=bos_token, eos_token=eos_token, unk_token=unk_token,
+                                                 sep_token=sep_token, cls_token=cls_token, pad_token=pad_token,
+                                                 mask_token=mask_token,
+                                                 **kwargs)
+        self.max_len_single_sentence = self.max_len - 2  # take into account special tokens
+        self.max_len_sentences_pair = self.max_len - 4  # take into account special tokens
+        self.sp_model = spm.SentencePieceProcessor()
+        self.sp_model.Load(str(vocab_file))
+        self.vocab_file = vocab_file
+        self.fairseq_tokens_to_ids = {"<s>": 0, "<unk>": 1, "</s>": 2}
+        self.fairseq_tokens_to_ids['<mask>'] = len(self.sp_model) + len(self.fairseq_tokens_to_ids)
+        self.fairseq_ids_to_tokens = {v: k for k, v in self.fairseq_tokens_to_ids.items()}
+
+    def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
+        """
+        Build model inputs from a sequence or a pair of sequence for sequence classification tasks
+        by concatenating and adding special tokens.
+        A RoBERTa sequence has the following format:
+            single sequence: <s> X </s>
+            pair of sequences: <s> A </s></s> B </s>
+        """
+        if token_ids_1 is None:
+            return [self.cls_token_id] + token_ids_0 + [self.sep_token_id]
+        cls = [self.cls_token_id]
+        sep = [self.sep_token_id]
+        return cls + token_ids_0 + sep + sep + token_ids_1 + sep
+
+    def get_special_tokens_mask(self, token_ids_0, token_ids_1=None, already_has_special_tokens=False):
+        """
+        Retrieves sequence ids from a token list that has no special tokens added. This method is called when adding
+        special tokens using the tokenizer ``prepare_for_model`` or ``encode_plus`` methods.
+
+        Args:
+            token_ids_0: list of ids (must not contain special tokens)
+            token_ids_1: Optional list of ids (must not contain special tokens), necessary when fetching sequence ids
+                for sequence pairs
+            already_has_special_tokens: (default False) Set to True if the token list is already formated with
+                special tokens for the model
+
+        Returns:
+            A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
+        """
+        if already_has_special_tokens:
+            if token_ids_1 is not None:
+                raise ValueError("You should not supply a second sequence if the provided sequence of "
+                                 "ids is already formated with special tokens for the model.")
+            return list(map(lambda x: 1 if x in [self.sep_token_id, self.cls_token_id] else 0, token_ids_0))
+
+        if token_ids_1 is None:
+            return [1] + ([0] * len(token_ids_0)) + [1]
+        return [1] + ([0] * len(token_ids_0)) + [1, 1] + ([0] * len(token_ids_1)) + [1]
+
+    def create_token_type_ids_from_sequences(self, token_ids_0, token_ids_1=None):
+        """
+        Creates a mask from the two sequences passed to be used in a sequence-pair classification task.
+        A RoBERTa sequence pair mask has the following format:
+        0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1
+        | first sequence    | second sequence
+
+        if token_ids_1 is None, only returns the first portion of the mask (0's).
+        """
+        sep = [self.sep_token_id]
+        cls = [self.cls_token_id]
+
+        if token_ids_1 is None:
+            return len(cls + token_ids_0 + sep) * [0]
+        return len(cls + token_ids_0 + sep + sep) * [0] + len(token_ids_1 + sep) * [1]
+
+    @property
+    def vocab_size(self):
+        return len(self.sp_model) + len(self.fairseq_tokens_to_ids)
+
+    def _tokenize(self, text):
+        return self.sp_model.EncodeAsPieces(text)
+
+    def _convert_token_to_id(self, token):
+        """ Converts a token (str/unicode) in an id using the vocab. """
+        if token in self.fairseq_tokens_to_ids:
+            return self.fairseq_tokens_to_ids[token]
+        return self.sp_model.PieceToId(token) + 1
+
+    def _convert_id_to_token(self, index):
+        """Converts an index (integer) in a token (string/unicode) using the vocab."""
+        if index in self.fairseq_ids_to_tokens:
+            return self.fairseq_ids_to_tokens[index]
+        return self.sp_model.IdToPiece(index + 1)
+
+    def save_vocabulary(self, save_directory):
+        """ Save the sentencepiece vocabulary (copy original file) and special tokens file
+            to a directory.
+        """
+        if not os.path.isdir(save_directory):
+            logger.error("Vocabulary path ({}) should be a directory".format(save_directory))
+            return
+        out_vocab_file = os.path.join(save_directory, VOCAB_FILES_NAMES['vocab_file'])
+
+        if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file):
+            copyfile(self.vocab_file, out_vocab_file)
+
+        return (out_vocab_file,)

From a701a0cee1ae6cb7b93b047cc3ffc06b01157955 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan.schweter@bsb-muenchen.de>
Date: Mon, 16 Dec 2019 17:17:56 +0100
Subject: [PATCH 391/505] configuration: fix model name for large XLM-RoBERTa
 model

---
 transformers/configuration_xlm_roberta.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/configuration_xlm_roberta.py b/transformers/configuration_xlm_roberta.py
index 1633cc18aa..dd03572976 100644
--- a/transformers/configuration_xlm_roberta.py
+++ b/transformers/configuration_xlm_roberta.py
@@ -25,7 +25,7 @@ from .configuration_roberta import RobertaConfig
 logger = logging.getLogger(__name__)
 
 XLM_ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP = {
-    'xlm-roberta-base': "https://schweter.eu/cloud/transformers/xlm-roberta-large-config.json",
+    'xlm-roberta-large': "https://schweter.eu/cloud/transformers/xlm-roberta-large-config.json",
 }
 
 

From d064009b72c4a29cd66b6c633dcd8c3ad5ab6dca Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan.schweter@bsb-muenchen.de>
Date: Mon, 16 Dec 2019 17:23:25 +0100
Subject: [PATCH 392/505] converter: fix vocab size

---
 ...onvert_xlm_roberta_original_pytorch_checkpoint_to_pytorch.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/convert_xlm_roberta_original_pytorch_checkpoint_to_pytorch.py b/transformers/convert_xlm_roberta_original_pytorch_checkpoint_to_pytorch.py
index 888adf4819..6873f1d0f0 100644
--- a/transformers/convert_xlm_roberta_original_pytorch_checkpoint_to_pytorch.py
+++ b/transformers/convert_xlm_roberta_original_pytorch_checkpoint_to_pytorch.py
@@ -47,7 +47,7 @@ def convert_roberta_checkpoint_to_pytorch(roberta_checkpoint_path, pytorch_dump_
     roberta = FairseqRobertaModel.from_pretrained(roberta_checkpoint_path, bpe = 'sentencepiece')
     roberta.eval()  # disable dropout
     config = BertConfig(
-        vocab_size_or_config_json_file=250004,
+        vocab_size_or_config_json_file=250002,
         hidden_size=roberta.args.encoder_embed_dim,
         num_hidden_layers=roberta.args.encoder_layers,
         num_attention_heads=roberta.args.encoder_attention_heads,

From 855ff0e91d8b3bd75a3b1c1316e2efd814373764 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Mon, 16 Dec 2019 12:42:22 -0500
Subject: [PATCH 393/505] [doc] Model upload and sharing

ping @lysandrejik @thomwolf

Is this clear enough? Anything we should add?
---
 README.md                    | 41 ++++++++++++++++++++++++++++++++++++
 docs/source/index.rst        |  1 +
 docs/source/model_sharing.md | 40 +++++++++++++++++++++++++++++++++++
 3 files changed, 82 insertions(+)
 create mode 100644 docs/source/model_sharing.md

diff --git a/README.md b/README.md
index 214f61cc0c..a5ae74a9ae 100644
--- a/README.md
+++ b/README.md
@@ -56,6 +56,7 @@ Choose the right framework for every part of a model's lifetime
 | [Quick tour: Usage](#quick-tour) | Tokenizers & models usage: Bert and GPT-2 |
 | [Quick tour: TF 2.0 and PyTorch ](#Quick-tour-TF-20-training-and-PyTorch-interoperability) | Train a TF 2.0 model in 10 lines of code, load it in PyTorch |
 | [Quick tour: Fine-tuning/usage scripts](#quick-tour-of-the-fine-tuningusage-scripts) | Using provided scripts: GLUE, SQuAD and Text generation |
+| [Quick tour: Share your models ](#Quick-tour-of-model-sharing) | Upload and share your fine-tuned models with the community |
 | [Migrating from pytorch-transformers to transformers](#Migrating-from-pytorch-transformers-to-transformers) | Migrating your code from pytorch-transformers to transformers |
 | [Migrating from pytorch-pretrained-bert to pytorch-transformers](#Migrating-from-pytorch-pretrained-bert-to-transformers) | Migrating your code from pytorch-pretrained-bert to transformers |
 | [Documentation][(v2.2.0/v2.2.1/v2.2.2)](https://huggingface.co/transformers/v2.2.0) [(v2.1.1)](https://huggingface.co/transformers/v2.1.1) [(v2.0.0)](https://huggingface.co/transformers/v2.0.0) [(v1.2.0)](https://huggingface.co/transformers/v1.2.0) [(v1.1.0)](https://huggingface.co/transformers/v1.1.0) [(v1.0.0)](https://huggingface.co/transformers/v1.0.0) [(master)](https://huggingface.co/transformers) | Full API documentation and more |
@@ -446,6 +447,46 @@ python ./examples/run_generation.py \
     --repetition_penalty=1.2 \
 ```
 
+## Quick tour of model sharing
+
+New in `v2.2.2`: you can now upload and share your fine-tuned models with the community, using the <abbr title="Command-line interface">CLI</abbr> that's built-in to the library.
+
+**First, create an account on [https://huggingface.co/join](https://huggingface.co/join)**. Then:
+
+```shell
+transformers-cli login
+# log in using the same credentials as on huggingface.co
+```
+Upload your model:
+```shell
+transformers-cli upload ./path/to/pretrained_model/
+
+# ^^ Upload folder containing weights/tokenizer/config
+# saved via `.save_pretrained()`
+
+transformers-cli upload ./config.json [--filename foobar.json]
+
+# ^^ Upload a single file
+# (you can optionally override its filename)
+```
+
+Your model will then be accessible through its identifier:
+```python
+"username/model_name"
+```
+
+Anyone can load it from code:
+```python
+tokenizer = AutoTokenizer.from_pretrained("username/model_name")
+model = AutoModel.from_pretrained("username/model_name")
+```
+
+Finally, list all your files on S3:
+```shell
+transformers-cli ls
+# List all your S3 objects.
+```
+
 ## Migrating from pytorch-transformers to transformers
 
 Here is a quick summary of what you should take care of when migrating from `pytorch-transformers` to `transformers`.
diff --git a/docs/source/index.rst b/docs/source/index.rst
index 84012fc6cf..48282c1c6c 100644
--- a/docs/source/index.rst
+++ b/docs/source/index.rst
@@ -58,6 +58,7 @@ The library currently contains PyTorch and Tensorflow implementations, pre-train
     installation
     quickstart
     pretrained_models
+    model_sharing
     examples
     notebooks
     serialization
diff --git a/docs/source/model_sharing.md b/docs/source/model_sharing.md
new file mode 100644
index 0000000000..b9c722b10f
--- /dev/null
+++ b/docs/source/model_sharing.md
@@ -0,0 +1,40 @@
+# Model upload and sharing
+
+Starting with `v2.2.2`, you can now upload and share your fine-tuned models with the community, using the <abbr title="Command-line interface">CLI</abbr> that's built-in to the library.
+
+**First, create an account on [https://huggingface.co/join](https://huggingface.co/join)**. Then:
+
+```shell
+transformers-cli login
+# log in using the same credentials as on huggingface.co
+```
+Upload your model:
+```shell
+transformers-cli upload ./path/to/pretrained_model/
+
+# ^^ Upload folder containing weights/tokenizer/config
+# saved via `.save_pretrained()`
+
+transformers-cli upload ./config.json [--filename foobar.json]
+
+# ^^ Upload a single file
+# (you can optionally override its filename)
+```
+
+Your model will then be accessible through its identifier:
+```python
+"username/model_name"
+```
+
+Anyone can load it from code:
+```python
+tokenizer = AutoTokenizer.from_pretrained("username/model_name")
+model = AutoModel.from_pretrained("username/model_name")
+```
+
+Finally, list all your files on S3:
+```shell
+transformers-cli ls
+# List all your S3 objects.
+```
+

From d8034092153a6850052862f154a398b88b8ba4e5 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Mon, 16 Dec 2019 16:31:38 -0500
Subject: [PATCH 394/505] Fix run squad evaluate during training

---
 examples/run_squad.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/examples/run_squad.py b/examples/run_squad.py
index a39915ee8b..34c31c3bb8 100644
--- a/examples/run_squad.py
+++ b/examples/run_squad.py
@@ -223,7 +223,7 @@ def evaluate(args, model, tokenizer, prefix=""):
     eval_dataloader = DataLoader(dataset, sampler=eval_sampler, batch_size=args.eval_batch_size)
 
     # multi-gpu evaluate
-    if args.n_gpu > 1:
+    if args.n_gpu > 1 and not isinstance(model, torch.nn.DataParallel):
         model = torch.nn.DataParallel(model)
 
     # Eval!

From 18a879f47576822aa1a5c49aecb27d89bfa5fa69 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Mon, 16 Dec 2019 16:44:29 -0500
Subject: [PATCH 395/505] fix #2180

---
 examples/run_generation.py | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/examples/run_generation.py b/examples/run_generation.py
index 2d917660cf..fa52905b7e 100644
--- a/examples/run_generation.py
+++ b/examples/run_generation.py
@@ -247,7 +247,11 @@ def main():
         out = out[:, len(context_tokens):].tolist()
         for o in out:
             text = tokenizer.decode(o, clean_up_tokenization_spaces=True)
-            text = text[: text.find(args.stop_token) if args.stop_token else None]
+            if args.stop_token:
+                index =  text.find(args.stop_token)
+                if index == -1:
+                    index = None
+                text = text[:index]
 
             print(text)
 

From 3cb51299c371f67b4da40b89c59c63e9405591f0 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Mon, 16 Dec 2019 22:32:05 +0100
Subject: [PATCH 396/505] Fix #2109

---
 transformers/modeling_tf_pytorch_utils.py | 13 +++++++++++--
 transformers/modeling_tf_utils.py         |  2 +-
 2 files changed, 12 insertions(+), 3 deletions(-)

diff --git a/transformers/modeling_tf_pytorch_utils.py b/transformers/modeling_tf_pytorch_utils.py
index 9d2b663dcb..d885fd23b3 100644
--- a/transformers/modeling_tf_pytorch_utils.py
+++ b/transformers/modeling_tf_pytorch_utils.py
@@ -143,7 +143,11 @@ def load_pytorch_weights_in_tf2_model(tf_model, pt_state_dict, tf_inputs=None, a
         name, transpose = convert_tf_weight_name_to_pt_weight_name(sw_name, start_prefix_to_remove=start_prefix_to_remove)
 
         # Find associated numpy array in pytorch model state dict
-        assert name in pt_state_dict, "{} not found in PyTorch model".format(name)
+        if name not in pt_state_dict:
+            if allow_missing_keys:
+                continue
+            raise AttributeError("{} not found in PyTorch model".format(name))
+
         array = pt_state_dict[name].numpy()
 
         if transpose:
@@ -250,6 +254,7 @@ def load_tf2_weights_in_pytorch_model(pt_model, tf_weights, allow_missing_keys=F
 
     all_tf_weights = set(list(tf_weights_map.keys()))
     loaded_pt_weights_data_ptr = {}
+    missing_keys_pt = []
     for pt_weight_name, pt_weight in current_pt_params_dict.items():
         # Handle PyTorch shared weight ()not duplicated in TF 2.0
         if pt_weight.data_ptr() in loaded_pt_weights_data_ptr:
@@ -258,7 +263,10 @@ def load_tf2_weights_in_pytorch_model(pt_model, tf_weights, allow_missing_keys=F
 
         # Find associated numpy array in pytorch model state dict
         if pt_weight_name not in tf_weights_map:
-            raise ValueError("{} not found in TF 2.0 model".format(pt_weight_name))
+            if allow_missing_keys:
+                missing_keys_pt.append(pt_weight_name)
+                continue
+            raise AttributeError("{} not found in TF 2.0 model".format(pt_weight_name))
 
         array, transpose = tf_weights_map[pt_weight_name]
 
@@ -283,6 +291,7 @@ def load_tf2_weights_in_pytorch_model(pt_model, tf_weights, allow_missing_keys=F
         all_tf_weights.discard(pt_weight_name)
 
     missing_keys, unexpected_keys = pt_model.load_state_dict(new_pt_params_dict, strict=False)
+    missing_keys += missing_keys_pt
 
     if len(missing_keys) > 0:
         logger.info("Weights of {} not initialized from TF 2.0 model: {}".format(
diff --git a/transformers/modeling_tf_utils.py b/transformers/modeling_tf_utils.py
index 6fb4850b05..6bbec71cdf 100644
--- a/transformers/modeling_tf_utils.py
+++ b/transformers/modeling_tf_utils.py
@@ -297,7 +297,7 @@ class TFPreTrainedModel(tf.keras.Model):
 
         if from_pt:
             # Load from a PyTorch checkpoint
-            return load_pytorch_checkpoint_in_tf2_model(model, resolved_archive_file)
+            return load_pytorch_checkpoint_in_tf2_model(model, resolved_archive_file, allow_missing_keys=True)
 
         ret = model(model.dummy_inputs, training=False)  # build the network with dummy inputs
 

From 3f5ccb183e3cfa755dea2dd2afd9abbf1a0f93b8 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Mon, 16 Dec 2019 18:20:23 -0500
Subject: [PATCH 397/505] [doc] Clarify uploads

cf https://github.com/huggingface/transformers/commit/855ff0e91d8b3bd75a3b1c1316e2efd814373764#commitcomment-36452545
---
 README.md                    | 10 +++++-----
 docs/source/model_sharing.md | 12 ++++++------
 2 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/README.md b/README.md
index a5ae74a9ae..c33a65bdbb 100644
--- a/README.md
+++ b/README.md
@@ -464,21 +464,21 @@ transformers-cli upload ./path/to/pretrained_model/
 # ^^ Upload folder containing weights/tokenizer/config
 # saved via `.save_pretrained()`
 
-transformers-cli upload ./config.json [--filename foobar.json]
+transformers-cli upload ./config.json [--filename folder/foobar.json]
 
 # ^^ Upload a single file
-# (you can optionally override its filename)
+# (you can optionally override its filename, which can be nested inside a folder)
 ```
 
-Your model will then be accessible through its identifier:
+Your model will then be accessible through its identifier, a concatenation of your username and the folder name above:
 ```python
 "username/model_name"
 ```
 
 Anyone can load it from code:
 ```python
-tokenizer = AutoTokenizer.from_pretrained("username/model_name")
-model = AutoModel.from_pretrained("username/model_name")
+tokenizer = AutoTokenizer.from_pretrained("username/pretrained_model")
+model = AutoModel.from_pretrained("username/pretrained_model")
 ```
 
 Finally, list all your files on S3:
diff --git a/docs/source/model_sharing.md b/docs/source/model_sharing.md
index b9c722b10f..95baafb575 100644
--- a/docs/source/model_sharing.md
+++ b/docs/source/model_sharing.md
@@ -15,21 +15,21 @@ transformers-cli upload ./path/to/pretrained_model/
 # ^^ Upload folder containing weights/tokenizer/config
 # saved via `.save_pretrained()`
 
-transformers-cli upload ./config.json [--filename foobar.json]
+transformers-cli upload ./config.json [--filename folder/foobar.json]
 
 # ^^ Upload a single file
-# (you can optionally override its filename)
+# (you can optionally override its filename, which can be nested inside a folder)
 ```
 
-Your model will then be accessible through its identifier:
+Your model will then be accessible through its identifier, a concatenation of your username and the folder name above:
 ```python
-"username/model_name"
+"username/pretrained_model"
 ```
 
 Anyone can load it from code:
 ```python
-tokenizer = AutoTokenizer.from_pretrained("username/model_name")
-model = AutoModel.from_pretrained("username/model_name")
+tokenizer = AutoTokenizer.from_pretrained("username/pretrained_model")
+model = AutoModel.from_pretrained("username/pretrained_model")
 ```
 
 Finally, list all your files on S3:

From f349826a57e6f7f1eb5c28ef3b3ff0ac6884ad24 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan.schweter@bsb-muenchen.de>
Date: Tue, 17 Dec 2019 10:36:04 +0100
Subject: [PATCH 398/505] model: fix cls and sep token for XLM-RoBERTa
 documentation

---
 transformers/modeling_xlm_roberta.py | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/transformers/modeling_xlm_roberta.py b/transformers/modeling_xlm_roberta.py
index 8402be4b5c..4c833c69ff 100644
--- a/transformers/modeling_xlm_roberta.py
+++ b/transformers/modeling_xlm_roberta.py
@@ -111,17 +111,17 @@ class XLMRobertaModel(RobertaModel):
             Last layer hidden-state of the first token of the sequence (classification token)
             further processed by a Linear layer and a Tanh activation function. The Linear
             layer weights are trained from the next sentence prediction (classification)
-            eo match pre-training, XLM-RoBERTa input sequence should be formatted with [CLS] and [SEP] tokens as follows:
+            eo match pre-training, XLM-RoBERTa input sequence should be formatted with <s> and </s> tokens as follows:
 
             (a) For sequence pairs:
 
-                ``tokens:         [CLS] is this jack ##son ##ville ? [SEP] [SEP] no it is not . [SEP]``
+                ``tokens:         <s> is this jack ##son ##ville ? </s> </s> no it is not . </s>``
 
                 ``token_type_ids:   0   0  0    0    0     0       0   0   0     1  1  1  1   1   1``
 
             (b) For single sequences:
 
-                ``tokens:         [CLS] the dog is hairy . [SEP]``
+                ``tokens:         <s> the dog is hairy . </s>``
 
                 ``token_type_ids:   0   0   0   0  0     0   0``
 

From d7c62661a314c631b3bbf6405143934c8c3e8b5f Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Tue, 17 Dec 2019 11:23:39 +0100
Subject: [PATCH 399/505] Provide serving dependencies for tensorflow and
 pytorch (serving-tf, serving-torch)

---
 setup.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/setup.py b/setup.py
index 0b7e512955..b3b6e2e063 100644
--- a/setup.py
+++ b/setup.py
@@ -38,7 +38,9 @@ from setuptools import find_packages, setup
 
 
 extras = {
-    'serving': ['uvicorn', 'fastapi']
+    'serving': ['uvicorn', 'fastapi'],
+    'serving-tf': ['uvicorn', 'fastapi', 'tensorflow'],
+    'serving-torch': ['uvicorn', 'fastapi', 'torch']
 }
 extras['all'] = [package for package in extras.values()]
 

From 2f1c745cded91b2f6cfed5b502ea5cbd7d6b9ac7 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Tue, 17 Dec 2019 11:47:54 +0100
Subject: [PATCH 400/505] update conversion script

---
 ...onvert_xlm_roberta_original_pytorch_checkpoint_to_pytorch.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/convert_xlm_roberta_original_pytorch_checkpoint_to_pytorch.py b/transformers/convert_xlm_roberta_original_pytorch_checkpoint_to_pytorch.py
index 6873f1d0f0..884c273d2c 100644
--- a/transformers/convert_xlm_roberta_original_pytorch_checkpoint_to_pytorch.py
+++ b/transformers/convert_xlm_roberta_original_pytorch_checkpoint_to_pytorch.py
@@ -47,7 +47,7 @@ def convert_roberta_checkpoint_to_pytorch(roberta_checkpoint_path, pytorch_dump_
     roberta = FairseqRobertaModel.from_pretrained(roberta_checkpoint_path, bpe = 'sentencepiece')
     roberta.eval()  # disable dropout
     config = BertConfig(
-        vocab_size_or_config_json_file=250002,
+        vocab_size=250002,
         hidden_size=roberta.args.encoder_embed_dim,
         num_hidden_layers=roberta.args.encoder_layers,
         num_attention_heads=roberta.args.encoder_attention_heads,

From 2fde5a2489bc4aa7fc42ab76effde241b2a0b919 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Tue, 17 Dec 2019 12:16:07 +0100
Subject: [PATCH 401/505] Initial bunch of documentation.

---
 transformers/pipelines.py | 118 +++++++++++++++++++++++++++++++++++---
 1 file changed, 111 insertions(+), 7 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 2a8f26b03e..6dcb865c74 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -80,6 +80,15 @@ class _ScikitCompat(ABC):
 
 
 class PipelineDataFormat:
+    """
+    Base class for all the pipeline supported data format both for reading and writing.
+    Supported data formats currently includes:
+     - JSON
+     - CSV
+
+    PipelineDataFormat also includes some utilities to work with multi-columns like mapping from datasets columns
+    to pipelines keyword arguments through the `dataset_kwarg_1=dataset_column_1` format.
+    """
     SUPPORTED_FORMATS = ['json', 'csv']
 
     def __init__(self, output: str, path: str, column: str):
@@ -138,7 +147,6 @@ class CsvPipelineDataFormat(PipelineDataFormat):
 
 
 class JsonPipelineDataFormat(PipelineDataFormat):
-
     def __init__(self, output: str, path: str, column: str):
         super().__init__(output, path, column)
 
@@ -158,6 +166,11 @@ class JsonPipelineDataFormat(PipelineDataFormat):
 
 
 class Pipeline(_ScikitCompat):
+    """
+    Base class implementing pipelined operations.
+    Pipeline workflow is defined as a sequence of the following operations:
+        Input -> Tokenization -> Model Inference -> Post-Processing (Task dependent) -> Output
+    """
     def __init__(self, model, tokenizer: PreTrainedTokenizer = None,
                  args_parser: ArgumentHandler = None, device: int = -1, **kwargs):
 
@@ -171,6 +184,9 @@ class Pipeline(_ScikitCompat):
             self.model = self.model.to('cuda:{}'.format(self.device))
 
     def save_pretrained(self, save_directory):
+        """
+        Save the pipeline's model and tokenizer to the specified save_directory
+        """
         if not os.path.isdir(save_directory):
             logger.error("Provided path ({}) should be a directory".format(save_directory))
             return
@@ -179,9 +195,16 @@ class Pipeline(_ScikitCompat):
         self.tokenizer.save_pretrained(save_directory)
 
     def transform(self, X):
+        """
+        Scikit / Keras interface to transformers' pipelines. This method will forward to __call__().
+        """
         return self(X=X)
 
     def predict(self, X):
+        """
+        Scikit / Keras interface to transformers' pipelines. This method will forward to __call__().
+        Se
+        """
         return self(X=X)
 
     def __call__(self, *texts, **kwargs):
@@ -198,6 +221,17 @@ class Pipeline(_ScikitCompat):
 
     @contextmanager
     def device_placement(self):
+        """
+        Context Manager allowing tensor allocation on the user-specified device in framework agnostic way.
+        example:
+            # Explicitly ask for tensor allocation on CUDA device :0
+            nlp = pipeline(..., device=0)
+            with nlp.device_placement():
+                # Every framework specific tensor allocation will be done on the request device
+                output = nlp(...)
+        Returns:
+            Context manager
+        """
         if is_tf_available():
             import tensorflow as tf
             with tf.device('/CPU:0' if self.device == -1 else '/device:GPU:{}'.format(self.device)):
@@ -210,6 +244,13 @@ class Pipeline(_ScikitCompat):
             yield
 
     def _forward(self, inputs):
+        """
+        Internal framework specific forward dispatching.
+        Args:
+            inputs: dict holding all the keyworded arguments for required by the model forward method.
+        Returns:
+            Numpy array
+        """
         if is_tf_available():
             # TODO trace model
             predictions = self.model(inputs)[0]
@@ -222,11 +263,17 @@ class Pipeline(_ScikitCompat):
 
 
 class FeatureExtractionPipeline(Pipeline):
+    """
+    Feature extraction pipeline using Model head.
+    """
     def __call__(self, *args, **kwargs):
         return super().__call__(*args, **kwargs).tolist()
 
 
 class TextClassificationPipeline(Pipeline):
+    """
+    Text classification pipeline using ModelForTextClassification head.
+    """
     def __init__(self, model, tokenizer: PreTrainedTokenizer, nb_classes: int = 2):
         super().__init__(model, tokenizer)
 
@@ -239,7 +286,9 @@ class TextClassificationPipeline(Pipeline):
 
 
 class NerPipeline(Pipeline):
-
+    """
+    Named Entity Recognition pipeline using ModelForTokenClassification head.
+    """
     def __init__(self, model, tokenizer: PreTrainedTokenizer):
         super().__init__(model, tokenizer)
 
@@ -286,7 +335,7 @@ class NerPipeline(Pipeline):
 
 class QuestionAnsweringPipeline(Pipeline):
     """
-    Question Answering pipeline involving Tokenization and Inference.
+    Question Answering pipeline using ModelForQuestionAnswering head.
     """
 
     class QuestionAnsweringArgumentHandler(ArgumentHandler):
@@ -341,9 +390,15 @@ class QuestionAnsweringPipeline(Pipeline):
 
     @staticmethod
     def create_sample(question: Union[str, List[str]], context: Union[str, List[str]]) -> Union[SquadExample, List[SquadExample]]:
-        is_list = isinstance(question, list)
-
-        if is_list:
+        """
+        QuestionAnsweringPipeline leverages the SquadExample/SquadFeatures internally.
+        This helper method encapsulate all the logic for converting question(s) and context(s) to SquadExample(s).
+        We currently support extractive question answering.
+        Args:
+             question: (str, List[str]) The question to be ask for the associated context
+             context: (str, List[str]) The context in which we will look for the answer.
+        """
+        if isinstance(question, list):
             return [SquadExample(None, q, c, None, None, None) for q, c in zip(question, context)]
         else:
             return SquadExample(None, question, context, None, None, None)
@@ -352,6 +407,12 @@ class QuestionAnsweringPipeline(Pipeline):
         super().__init__(model, tokenizer, args_parser=QuestionAnsweringPipeline.QuestionAnsweringArgumentHandler())
 
     def inputs_for_model(self, features: Union[SquadExample, List[SquadExample]]) -> Dict:
+        """
+        Generates the input dictionary with model-specific parameters.
+
+        Returns:
+            dict holding all the required parameters for model's forward
+        """
         args = ['input_ids', 'attention_mask']
         model_type = type(self.model).__name__.lower()
 
@@ -367,6 +428,20 @@ class QuestionAnsweringPipeline(Pipeline):
             return {k: [feature.__dict__[k] for feature in features] for k in args}
 
     def __call__(self, *texts, **kwargs):
+        """
+        Args:
+            We support multiple use-cases, the following are exclusive:
+            X: sequence of SquadExample
+            data: sequence of SquadExample
+            question: (str, List[str]), batch of question(s) to map along with context
+            context: (str, List[str]), batch of context(s) associated with the provided question keyword argument
+        Returns:
+            dict: {'answer': str, 'score": float, 'start": int, "end": int}
+            answer: the textual answer in the intial context
+            score: the score the current answer scored for the model
+            start: the character index in the original string corresponding to the beginning of the answer' span
+            end: the character index in the original string corresponding to the ending of the answer' span
+        """
         # Set defaults values
         kwargs.setdefault('topk', 1)
         kwargs.setdefault('doc_stride', 128)
@@ -432,6 +507,19 @@ class QuestionAnsweringPipeline(Pipeline):
         return answers
 
     def decode(self, start: np.ndarray, end: np.ndarray, topk: int, max_answer_len: int) -> Tuple:
+        """
+        Take the output of any QuestionAnswering head and will generate probalities for each span to be
+        the actual answer.
+        In addition, it filters out some unwanted/impossible cases like answer len being greater than
+        max_answer_len or answer end position being before the starting position.
+        The method supports output the k-best answer through the topk argument.
+
+        Args:
+            start: numpy array, holding individual start probabilities for each token
+            end: numpy array, holding individual end probabilities for each token
+            topk: int, indicates how many possible answer span(s) to extract from the model's output
+            max_answer_len: int, maximum size of the answer to extract from the model's output
+        """
         # Ensure we have batch axis
         if start.ndim == 1:
             start = start[None]
@@ -459,6 +547,18 @@ class QuestionAnsweringPipeline(Pipeline):
         return start, end, candidates[0, start, end]
 
     def span_to_answer(self, text: str, start: int, end: int):
+        """
+        When decoding from token probalities, this method maps token indexes to actual word in
+        the initial context.
+
+        Args:
+            text: str, the actual context to extract the answer from
+            start: int, starting answer token index
+            end: int, ending answer token index
+
+        Returns:
+            dict: {'answer': str, 'start': int, 'end': int}
+        """
         words = []
         token_idx = char_start_idx = char_end_idx = chars_idx = 0
 
@@ -514,7 +614,11 @@ SUPPORTED_TASKS = {
 
 def pipeline(task: str, model, config: Optional[PretrainedConfig] = None, tokenizer: Optional[Union[str, PreTrainedTokenizer]] = None, **kwargs) -> Pipeline:
     """
-    Utility factory method to build pipeline.
+    Utility factory method to build a pipeline.
+    Pipeline are made of:
+        A Tokenizer instance in charge of mapping raw textual input to token
+        A Model instance
+        Some (optional) post processing for enhancing model's output
     """
     # Try to infer tokenizer from model name (if provided as str)
     if tokenizer is None:

From 55397dfb9b7e61001e10abb595931d4a98ae58b0 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Tue, 17 Dec 2019 13:10:51 -0500
Subject: [PATCH 402/505] CsvPipelineDataFormat: Fix for single-column

---
 transformers/pipelines.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 6dcb865c74..dec7843baf 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -136,7 +136,7 @@ class CsvPipelineDataFormat(PipelineDataFormat):
                 if self.is_multi_columns:
                     yield {k: row[c] for k, c in self.column}
                 else:
-                    yield row[self.column]
+                    yield row[self.column[0]]
 
     def save(self, data: List[dict]):
         with open(self.output, 'w') as f:

From 2cff4bd8f3ad412917f4f295b97b952e297fa257 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Tue, 17 Dec 2019 14:01:04 -0500
Subject: [PATCH 403/505] Fix segmentation fault

---
 transformers/file_utils.py | 15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)

diff --git a/transformers/file_utils.py b/transformers/file_utils.py
index 81c9b8002f..16010f7e0a 100644
--- a/transformers/file_utils.py
+++ b/transformers/file_utils.py
@@ -26,14 +26,6 @@ from contextlib import contextmanager
 
 logger = logging.getLogger(__name__)  # pylint: disable=invalid-name
 
-try:
-    import tensorflow as tf
-    assert hasattr(tf, '__version__') and int(tf.__version__[0]) >= 2
-    _tf_available = True  # pylint: disable=invalid-name
-    logger.info("TensorFlow version {} available.".format(tf.__version__))
-except (ImportError, AssertionError):
-    _tf_available = False  # pylint: disable=invalid-name
-
 try:
     import torch
     _torch_available = True  # pylint: disable=invalid-name
@@ -41,6 +33,13 @@ try:
 except ImportError:
     _torch_available = False  # pylint: disable=invalid-name
 
+try:
+    import tensorflow as tf
+    assert hasattr(tf, '__version__') and int(tf.__version__[0]) >= 2
+    _tf_available = True  # pylint: disable=invalid-name
+    logger.info("TensorFlow version {} available.".format(tf.__version__))
+except (ImportError, AssertionError):
+    _tf_available = False  # pylint: disable=invalid-name
 
 try:
     from torch.hub import _get_torch_home

From 5e289f69bc564c94132f77c89a34e5f1dd69a592 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Tue, 17 Dec 2019 14:17:11 -0500
Subject: [PATCH 404/505] regex 2019.12.17 install fails with Python 2

---
 requirements.txt | 2 +-
 setup.py         | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/requirements.txt b/requirements.txt
index 9c43abc6d7..32edee0712 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -5,7 +5,7 @@ boto3
 # Used for downloading models over HTTP
 requests
 # For OpenAI GPT
-regex
+regex != 2019.12.17
 # For XLNet
 sentencepiece
 # For XLM
diff --git a/setup.py b/setup.py
index eacb5ecec0..bf09a7d48a 100644
--- a/setup.py
+++ b/setup.py
@@ -59,7 +59,7 @@ setup(
                       'boto3',
                       'requests',
                       'tqdm',
-                      'regex',
+                      'regex != 2019.12.17',
                       'sentencepiece',
                       'sacremoses'],
     entry_points={

From a4df2e011367020253c8ca8a714c4b4855ff61bc Mon Sep 17 00:00:00 2001
From: Arman Cohan <armanc@allenai.org>
Date: Tue, 26 Nov 2019 16:03:07 -0800
Subject: [PATCH 405/505] update roberta conversion

- update to fix conversion for the updated fairseq model
- create save directory if not exist
---
 ..._original_pytorch_checkpoint_to_pytorch.py | 21 ++++++++++++-------
 1 file changed, 13 insertions(+), 8 deletions(-)

diff --git a/transformers/convert_roberta_original_pytorch_checkpoint_to_pytorch.py b/transformers/convert_roberta_original_pytorch_checkpoint_to_pytorch.py
index b4dc1bb61b..be3460a86f 100644
--- a/transformers/convert_roberta_original_pytorch_checkpoint_to_pytorch.py
+++ b/transformers/convert_roberta_original_pytorch_checkpoint_to_pytorch.py
@@ -20,6 +20,7 @@ import argparse
 import logging
 import numpy as np
 import torch
+import pathlib
 
 from fairseq.models.roberta import RobertaModel as FairseqRobertaModel
 from fairseq.modules import TransformerSentenceEncoderLayer
@@ -79,15 +80,18 @@ def convert_roberta_checkpoint_to_pytorch(roberta_checkpoint_path, pytorch_dump_
         ### self attention
         self_attn: BertSelfAttention = layer.attention.self
         assert(
-            roberta_layer.self_attn.in_proj_weight.shape == torch.Size((3 * config.hidden_size, config.hidden_size))
+            roberta_layer.self_attn.k_proj.weight.data.shape == \
+            roberta_layer.self_attn.q_proj.weight.data.shape == \
+            roberta_layer.self_attn.v_proj.weight.data.shape == \
+            torch.Size((config.hidden_size, config.hidden_size))
         )
-        # we use three distinct linear layers so we split the source layer here.
-        self_attn.query.weight.data = roberta_layer.self_attn.in_proj_weight[:config.hidden_size, :]
-        self_attn.query.bias.data = roberta_layer.self_attn.in_proj_bias[:config.hidden_size]
-        self_attn.key.weight.data = roberta_layer.self_attn.in_proj_weight[config.hidden_size:2*config.hidden_size, :]
-        self_attn.key.bias.data = roberta_layer.self_attn.in_proj_bias[config.hidden_size:2*config.hidden_size]
-        self_attn.value.weight.data = roberta_layer.self_attn.in_proj_weight[2*config.hidden_size:, :]
-        self_attn.value.bias.data = roberta_layer.self_attn.in_proj_bias[2*config.hidden_size:]
+
+        self_attn.query.weight.data = roberta_layer.self_attn.q_proj.weight
+        self_attn.query.bias.data = roberta_layer.self_attn.q_proj.bias
+        self_attn.key.weight.data = roberta_layer.self_attn.k_proj.weight
+        self_attn.key.bias.data = roberta_layer.self_attn.k_proj.bias
+        self_attn.value.weight.data = roberta_layer.self_attn.v_proj.weight
+        self_attn.value.bias.data = roberta_layer.self_attn.v_proj.bias
 
         ### self-attention output
         self_output: BertSelfOutput = layer.attention.output
@@ -151,6 +155,7 @@ def convert_roberta_checkpoint_to_pytorch(roberta_checkpoint_path, pytorch_dump_
     if not success:
         raise Exception("Something went wRoNg")
 
+    pathlib.Path(pytorch_dump_folder_path).mkdir(parents=True, exist_ok=True)
     print(f"Saving model to {pytorch_dump_folder_path}")
     model.save_pretrained(pytorch_dump_folder_path)
 

From ea636440d1ea3497785c2682c410da478f8b1841 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Tue, 17 Dec 2019 18:06:42 -0500
Subject: [PATCH 406/505] [roberta.conversion] Do not hardcode vocab size

and support for fairseq 0.9+
---
 ...t_roberta_original_pytorch_checkpoint_to_pytorch.py | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/transformers/convert_roberta_original_pytorch_checkpoint_to_pytorch.py b/transformers/convert_roberta_original_pytorch_checkpoint_to_pytorch.py
index be3460a86f..fedfc1ecb8 100644
--- a/transformers/convert_roberta_original_pytorch_checkpoint_to_pytorch.py
+++ b/transformers/convert_roberta_original_pytorch_checkpoint_to_pytorch.py
@@ -22,6 +22,12 @@ import numpy as np
 import torch
 import pathlib
 
+import fairseq
+from packaging import version
+
+if version.parse(fairseq.__version__) < version.parse("0.9.0"):
+    raise Exception("requires fairseq >= 0.9.0")
+
 from fairseq.models.roberta import RobertaModel as FairseqRobertaModel
 from fairseq.modules import TransformerSentenceEncoderLayer
 from transformers.modeling_bert import (BertConfig, BertEncoder,
@@ -46,8 +52,9 @@ def convert_roberta_checkpoint_to_pytorch(roberta_checkpoint_path, pytorch_dump_
     """
     roberta = FairseqRobertaModel.from_pretrained(roberta_checkpoint_path)
     roberta.eval()  # disable dropout
+    roberta_sent_encoder = roberta.model.decoder.sentence_encoder
     config = BertConfig(
-        vocab_size=50265,
+        vocab_size=roberta_sent_encoder.embed_tokens.num_embeddings,
         hidden_size=roberta.args.encoder_embed_dim,
         num_hidden_layers=roberta.args.encoder_layers,
         num_attention_heads=roberta.args.encoder_attention_heads,
@@ -65,7 +72,6 @@ def convert_roberta_checkpoint_to_pytorch(roberta_checkpoint_path, pytorch_dump_
 
     # Now let's copy all the weights.
     # Embeddings
-    roberta_sent_encoder = roberta.model.decoder.sentence_encoder
     model.roberta.embeddings.word_embeddings.weight = roberta_sent_encoder.embed_tokens.weight
     model.roberta.embeddings.position_embeddings.weight = roberta_sent_encoder.embed_positions.weight
     model.roberta.embeddings.token_type_embeddings.weight.data = torch.zeros_like(model.roberta.embeddings.token_type_embeddings.weight)  # just zero them out b/c RoBERTa doesn't use them.

From a0d386455b347508ea31fc88dd06cc5555255c37 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Tue, 17 Dec 2019 20:07:39 -0500
Subject: [PATCH 407/505] Fix outdated tokenizer doc

---
 templates/adding_a_new_model/tokenization_xxx.py | 2 +-
 transformers/tokenization_bert.py                | 4 ++--
 transformers/tokenization_distilbert.py          | 4 ++--
 3 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/templates/adding_a_new_model/tokenization_xxx.py b/templates/adding_a_new_model/tokenization_xxx.py
index 3d6b4ad9df..7a10a41e5a 100644
--- a/templates/adding_a_new_model/tokenization_xxx.py
+++ b/templates/adding_a_new_model/tokenization_xxx.py
@@ -85,7 +85,7 @@ class XxxTokenizer(PreTrainedTokenizer):
 
     Args:
         vocab_file: Path to a one-wordpiece-per-line vocabulary file
-        do_lower_case: Whether to lower case the input. Only has an effect when do_wordpiece_only=False
+        do_lower_case: Whether to lower case the input. Only has an effect when do_basic_tokenize=True
     """
 
     vocab_files_names = VOCAB_FILES_NAMES
diff --git a/transformers/tokenization_bert.py b/transformers/tokenization_bert.py
index ded5072e58..7ab8029da8 100644
--- a/transformers/tokenization_bert.py
+++ b/transformers/tokenization_bert.py
@@ -113,12 +113,12 @@ class BertTokenizer(PreTrainedTokenizer):
 
     Args:
         vocab_file: Path to a one-wordpiece-per-line vocabulary file
-        do_lower_case: Whether to lower case the input. Only has an effect when do_wordpiece_only=False
+        do_lower_case: Whether to lower case the input. Only has an effect when do_basic_tokenize=True
         do_basic_tokenize: Whether to do basic tokenization before wordpiece.
         max_len: An artificial maximum length to truncate tokenized sequences to; Effective maximum length is always the
             minimum of this value (if specified) and the underlying BERT model's sequence length.
         never_split: List of tokens which will never be split during tokenization. Only has an effect when
-            do_wordpiece_only=False
+            do_basic_tokenize=True
     """
 
     vocab_files_names = VOCAB_FILES_NAMES
diff --git a/transformers/tokenization_distilbert.py b/transformers/tokenization_distilbert.py
index f40bf2bd77..2f245d71dc 100644
--- a/transformers/tokenization_distilbert.py
+++ b/transformers/tokenization_distilbert.py
@@ -53,12 +53,12 @@ class DistilBertTokenizer(BertTokenizer):
 
     Args:
         vocab_file: Path to a one-wordpiece-per-line vocabulary file
-        do_lower_case: Whether to lower case the input. Only has an effect when do_wordpiece_only=False
+        do_lower_case: Whether to lower case the input. Only has an effect when do_basic_tokenize=True
         do_basic_tokenize: Whether to do basic tokenization before wordpiece.
         max_len: An artificial maximum length to truncate tokenized sequences to; Effective maximum length is always the
             minimum of this value (if specified) and the underlying BERT model's sequence length.
         never_split: List of tokens which will never be split during tokenization. Only has an effect when
-            do_wordpiece_only=False
+            do_basic_tokenize=True
     """
 
     vocab_files_names = VOCAB_FILES_NAMES

From 8ac840ff8758fb242e3e89cbc809366165ccf960 Mon Sep 17 00:00:00 2001
From: Antti Virtanen <haama92@gmail.com>
Date: Mon, 16 Dec 2019 17:08:25 +0200
Subject: [PATCH 408/505] Adding Finnish BERT.

---
 transformers/configuration_bert.py | 4 +++-
 transformers/modeling_bert.py      | 4 +++-
 transformers/modeling_tf_bert.py   | 4 +++-
 transformers/tokenization_bert.py  | 6 ++++++
 4 files changed, 15 insertions(+), 3 deletions(-)

diff --git a/transformers/configuration_bert.py b/transformers/configuration_bert.py
index 9072820bce..b1974966a9 100644
--- a/transformers/configuration_bert.py
+++ b/transformers/configuration_bert.py
@@ -45,7 +45,9 @@ BERT_PRETRAINED_CONFIG_ARCHIVE_MAP = {
     'bert-base-japanese': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-config.json",
     'bert-base-japanese-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-whole-word-masking-config.json",
     'bert-base-japanese-char': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-config.json",
-    'bert-base-japanese-char-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-whole-word-masking-config.json"
+    'bert-base-japanese-char-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-whole-word-masking-config.json",
+    'bert-base-finnish-cased-v1': "http://dl.turkunlp.org/finbert/torch-transformers/bert-base-finnish-cased-v1/config.json",
+    'bert-base-finnish-uncased-v1': "http://dl.turkunlp.org/finbert/torch-transformers/bert-base-finnish-uncased-v1/config.json",
 }
 
 
diff --git a/transformers/modeling_bert.py b/transformers/modeling_bert.py
index d0f35272ac..d0cb5ec617 100644
--- a/transformers/modeling_bert.py
+++ b/transformers/modeling_bert.py
@@ -51,7 +51,9 @@ BERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
     'bert-base-japanese': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-pytorch_model.bin",
     'bert-base-japanese-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-whole-word-masking-pytorch_model.bin",
     'bert-base-japanese-char': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-pytorch_model.bin",
-    'bert-base-japanese-char-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-whole-word-masking-pytorch_model.bin"
+    'bert-base-japanese-char-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-whole-word-masking-pytorch_model.bin",
+    'bert-base-finnish-cased-v1': "http://dl.turkunlp.org/finbert/torch-transformers/bert-base-finnish-cased-v1/pytorch_model.bin",
+    'bert-base-finnish-uncased-v1': "http://dl.turkunlp.org/finbert/torch-transformers/bert-base-finnish-uncased-v1/pytorch_model.bin",
 }
 
 
diff --git a/transformers/modeling_tf_bert.py b/transformers/modeling_tf_bert.py
index 7cc71f5063..20b5895dbd 100644
--- a/transformers/modeling_tf_bert.py
+++ b/transformers/modeling_tf_bert.py
@@ -51,7 +51,9 @@ TF_BERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
     'bert-base-japanese': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-tf_model.h5",
     'bert-base-japanese-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-whole-word-masking-tf_model.h5",
     'bert-base-japanese-char': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-tf_model.h5",
-    'bert-base-japanese-char-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-whole-word-masking-tf_model.h5"
+    'bert-base-japanese-char-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-whole-word-masking-tf_model.h5",
+    #'bert-base-finnish-cased-v1': "http://dl.turkunlp.org/finbert/torch-transformers/bert-base-finnish-cased-v1/pytorch_model.bin",
+    #'bert-base-finnish-uncased-v1': "http://dl.turkunlp.org/finbert/torch-transformers/bert-base-finnish-uncased-v1/pytorch_model.bin",
 }
 
 
diff --git a/transformers/tokenization_bert.py b/transformers/tokenization_bert.py
index 7ab8029da8..6f6a4d6f19 100644
--- a/transformers/tokenization_bert.py
+++ b/transformers/tokenization_bert.py
@@ -46,6 +46,8 @@ PRETRAINED_VOCAB_FILES_MAP = {
         'bert-base-cased-finetuned-mrpc': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-cased-finetuned-mrpc-vocab.txt",
         'bert-base-german-dbmdz-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-german-dbmdz-cased-vocab.txt",
         'bert-base-german-dbmdz-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-german-dbmdz-uncased-vocab.txt",
+        'bert-base-finnish-cased-v1': "http://dl.turkunlp.org/finbert/torch-transformers/bert-base-finnish-cased-v1/vocab.txt",
+        'bert-base-finnish-uncased-v1': "http://dl.turkunlp.org/finbert/torch-transformers/bert-base-finnish-uncased-v1/vocab.txt",
     }
 }
 
@@ -65,6 +67,8 @@ PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
     'bert-base-cased-finetuned-mrpc': 512,
     'bert-base-german-dbmdz-cased': 512,
     'bert-base-german-dbmdz-uncased': 512,
+    'bert-base-finnish-cased-v1': 512,
+    'bert-base-finnish-uncased-v1': 512,
 }
 
 PRETRAINED_INIT_CONFIGURATION = {
@@ -83,6 +87,8 @@ PRETRAINED_INIT_CONFIGURATION = {
     'bert-base-cased-finetuned-mrpc': {'do_lower_case': False},
     'bert-base-german-dbmdz-cased': {'do_lower_case': False},
     'bert-base-german-dbmdz-uncased': {'do_lower_case': True},
+    'bert-base-finnish-cased-v1': {'do_lower_case': False},
+    'bert-base-finnish-uncased-v1': {'do_lower_case': True},
 }
 
 

From abc43ffbfff69dc91f354c34f1c7c5b48a5c1502 Mon Sep 17 00:00:00 2001
From: Antti Virtanen <haama92@gmail.com>
Date: Mon, 16 Dec 2019 18:08:00 +0200
Subject: [PATCH 409/505] Add pretrained model documentation for FinBERT.

---
 docs/source/pretrained_models.rst | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/docs/source/pretrained_models.rst b/docs/source/pretrained_models.rst
index c6b990f213..7d037da34f 100644
--- a/docs/source/pretrained_models.rst
+++ b/docs/source/pretrained_models.rst
@@ -79,6 +79,14 @@ Here is the full list of the currently provided pretrained models together with
 |                   | ``bert-base-japanese-char-whole-word-masking``             | | 12-layer, 768-hidden, 12-heads, 110M parameters.                                                                                    |
 |                   |                                                            | | Trained on Japanese text using Whole-Word-Masking. Text is tokenized into characters.                                               |
 |                   |                                                            | (see `details on cl-tohoku repository <https://github.com/cl-tohoku/bert-japanese>`__).                                               |
+|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   | ``bert-base-finnish-cased-v1``                             | | 12-layer, 768-hidden, 12-heads, 110M parameters.                                                                                    |
+|                   |                                                            | | Trained on cased Finnish text.                                                                                                      |
+|                   |                                                            | (see `details on turkunlp.org <http://turkunlp.org/FinBERT/>`__).                                                                     |
+|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   | ``bert-base-finnish-uncased-v1``                           | | 12-layer, 768-hidden, 12-heads, 110M parameters.                                                                                    |
+|                   |                                                            | | Trained on uncased Finnish text.                                                                                                    |
+|                   |                                                            | (see `details on turkunlp.org <http://turkunlp.org/FinBERT/>`__).                                                                     |
 +-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 | GPT               | ``openai-gpt``                                             | | 12-layer, 768-hidden, 12-heads, 110M parameters.                                                                                    |
 |                   |                                                            | | OpenAI GPT English model                                                                                                            |

From c5f35e61db8d7286173515071e76612e9e5f5ce5 Mon Sep 17 00:00:00 2001
From: Antti Virtanen <haama92@gmail.com>
Date: Mon, 16 Dec 2019 21:06:14 +0200
Subject: [PATCH 410/505] Uploaded files to AWS.

---
 transformers/configuration_bert.py | 4 ++--
 transformers/modeling_bert.py      | 4 ++--
 transformers/modeling_tf_bert.py   | 4 ++--
 transformers/tokenization_bert.py  | 4 ++--
 4 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/transformers/configuration_bert.py b/transformers/configuration_bert.py
index b1974966a9..c2ccc578c2 100644
--- a/transformers/configuration_bert.py
+++ b/transformers/configuration_bert.py
@@ -46,8 +46,8 @@ BERT_PRETRAINED_CONFIG_ARCHIVE_MAP = {
     'bert-base-japanese-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-whole-word-masking-config.json",
     'bert-base-japanese-char': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-config.json",
     'bert-base-japanese-char-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-whole-word-masking-config.json",
-    'bert-base-finnish-cased-v1': "http://dl.turkunlp.org/finbert/torch-transformers/bert-base-finnish-cased-v1/config.json",
-    'bert-base-finnish-uncased-v1': "http://dl.turkunlp.org/finbert/torch-transformers/bert-base-finnish-uncased-v1/config.json",
+    'bert-base-finnish-cased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-cased-v1-config.json",
+    'bert-base-finnish-uncased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-uncased-v1-config.json",
 }
 
 
diff --git a/transformers/modeling_bert.py b/transformers/modeling_bert.py
index d0cb5ec617..4e034f4b6e 100644
--- a/transformers/modeling_bert.py
+++ b/transformers/modeling_bert.py
@@ -52,8 +52,8 @@ BERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
     'bert-base-japanese-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-whole-word-masking-pytorch_model.bin",
     'bert-base-japanese-char': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-pytorch_model.bin",
     'bert-base-japanese-char-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-whole-word-masking-pytorch_model.bin",
-    'bert-base-finnish-cased-v1': "http://dl.turkunlp.org/finbert/torch-transformers/bert-base-finnish-cased-v1/pytorch_model.bin",
-    'bert-base-finnish-uncased-v1': "http://dl.turkunlp.org/finbert/torch-transformers/bert-base-finnish-uncased-v1/pytorch_model.bin",
+    'bert-base-finnish-cased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-cased-v1-pytorch_model.bin",
+    'bert-base-finnish-uncased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-uncased-v1-pytorch_model.bin",
 }
 
 
diff --git a/transformers/modeling_tf_bert.py b/transformers/modeling_tf_bert.py
index 20b5895dbd..5a989c299f 100644
--- a/transformers/modeling_tf_bert.py
+++ b/transformers/modeling_tf_bert.py
@@ -52,8 +52,8 @@ TF_BERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
     'bert-base-japanese-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-whole-word-masking-tf_model.h5",
     'bert-base-japanese-char': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-tf_model.h5",
     'bert-base-japanese-char-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-whole-word-masking-tf_model.h5",
-    #'bert-base-finnish-cased-v1': "http://dl.turkunlp.org/finbert/torch-transformers/bert-base-finnish-cased-v1/pytorch_model.bin",
-    #'bert-base-finnish-uncased-v1': "http://dl.turkunlp.org/finbert/torch-transformers/bert-base-finnish-uncased-v1/pytorch_model.bin",
+    'bert-base-finnish-cased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-cased-v1-tf_model.h5",
+    'bert-base-finnish-uncased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-uncased-v1-tf_model.h5",
 }
 
 
diff --git a/transformers/tokenization_bert.py b/transformers/tokenization_bert.py
index 6f6a4d6f19..c11c1b4d3c 100644
--- a/transformers/tokenization_bert.py
+++ b/transformers/tokenization_bert.py
@@ -46,8 +46,8 @@ PRETRAINED_VOCAB_FILES_MAP = {
         'bert-base-cased-finetuned-mrpc': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-cased-finetuned-mrpc-vocab.txt",
         'bert-base-german-dbmdz-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-german-dbmdz-cased-vocab.txt",
         'bert-base-german-dbmdz-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-german-dbmdz-uncased-vocab.txt",
-        'bert-base-finnish-cased-v1': "http://dl.turkunlp.org/finbert/torch-transformers/bert-base-finnish-cased-v1/vocab.txt",
-        'bert-base-finnish-uncased-v1': "http://dl.turkunlp.org/finbert/torch-transformers/bert-base-finnish-uncased-v1/vocab.txt",
+        'bert-base-finnish-cased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-cased-v1-vocab.txt",
+        'bert-base-finnish-uncased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-cased-v1-vocab.txt",
     }
 }
 

From 7ffa8173905cb6d0819fc424a4806e81a44dd0e0 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Mon, 16 Dec 2019 18:55:14 -0500
Subject: [PATCH 411/505] [s3] mv files and update links

---
 transformers/configuration_bert.py | 4 ++--
 transformers/modeling_bert.py      | 4 ++--
 transformers/modeling_tf_bert.py   | 4 ++--
 transformers/tokenization_bert.py  | 4 ++--
 4 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/transformers/configuration_bert.py b/transformers/configuration_bert.py
index c2ccc578c2..7b495013ff 100644
--- a/transformers/configuration_bert.py
+++ b/transformers/configuration_bert.py
@@ -46,8 +46,8 @@ BERT_PRETRAINED_CONFIG_ARCHIVE_MAP = {
     'bert-base-japanese-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-whole-word-masking-config.json",
     'bert-base-japanese-char': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-config.json",
     'bert-base-japanese-char-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-whole-word-masking-config.json",
-    'bert-base-finnish-cased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-cased-v1-config.json",
-    'bert-base-finnish-uncased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-uncased-v1-config.json",
+    'bert-base-finnish-cased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-cased-v1/config.json",
+    'bert-base-finnish-uncased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-uncased-v1/config.json",
 }
 
 
diff --git a/transformers/modeling_bert.py b/transformers/modeling_bert.py
index 4e034f4b6e..afeb9d8e21 100644
--- a/transformers/modeling_bert.py
+++ b/transformers/modeling_bert.py
@@ -52,8 +52,8 @@ BERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
     'bert-base-japanese-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-whole-word-masking-pytorch_model.bin",
     'bert-base-japanese-char': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-pytorch_model.bin",
     'bert-base-japanese-char-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-whole-word-masking-pytorch_model.bin",
-    'bert-base-finnish-cased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-cased-v1-pytorch_model.bin",
-    'bert-base-finnish-uncased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-uncased-v1-pytorch_model.bin",
+    'bert-base-finnish-cased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-cased-v1/pytorch_model.bin",
+    'bert-base-finnish-uncased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-uncased-v1/pytorch_model.bin",
 }
 
 
diff --git a/transformers/modeling_tf_bert.py b/transformers/modeling_tf_bert.py
index 5a989c299f..b4f97c06d9 100644
--- a/transformers/modeling_tf_bert.py
+++ b/transformers/modeling_tf_bert.py
@@ -52,8 +52,8 @@ TF_BERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
     'bert-base-japanese-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-whole-word-masking-tf_model.h5",
     'bert-base-japanese-char': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-tf_model.h5",
     'bert-base-japanese-char-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-whole-word-masking-tf_model.h5",
-    'bert-base-finnish-cased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-cased-v1-tf_model.h5",
-    'bert-base-finnish-uncased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-uncased-v1-tf_model.h5",
+    'bert-base-finnish-cased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-cased-v1/tf_model.h5",
+    'bert-base-finnish-uncased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-uncased-v1/tf_model.h5",
 }
 
 
diff --git a/transformers/tokenization_bert.py b/transformers/tokenization_bert.py
index c11c1b4d3c..18b96c99b3 100644
--- a/transformers/tokenization_bert.py
+++ b/transformers/tokenization_bert.py
@@ -46,8 +46,8 @@ PRETRAINED_VOCAB_FILES_MAP = {
         'bert-base-cased-finetuned-mrpc': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-cased-finetuned-mrpc-vocab.txt",
         'bert-base-german-dbmdz-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-german-dbmdz-cased-vocab.txt",
         'bert-base-german-dbmdz-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-german-dbmdz-uncased-vocab.txt",
-        'bert-base-finnish-cased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-cased-v1-vocab.txt",
-        'bert-base-finnish-uncased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-cased-v1-vocab.txt",
+        'bert-base-finnish-cased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-cased-v1/vocab.txt",
+        'bert-base-finnish-uncased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-cased-v1/vocab.txt",
     }
 }
 

From 94c99db34cf9074a212c36554fb925c513d70ab1 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Tue, 17 Dec 2019 20:34:22 -0500
Subject: [PATCH 412/505] [FinBERT] fix incorrect url

---
 transformers/tokenization_bert.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/tokenization_bert.py b/transformers/tokenization_bert.py
index 18b96c99b3..edc26d88cf 100644
--- a/transformers/tokenization_bert.py
+++ b/transformers/tokenization_bert.py
@@ -47,7 +47,7 @@ PRETRAINED_VOCAB_FILES_MAP = {
         'bert-base-german-dbmdz-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-german-dbmdz-cased-vocab.txt",
         'bert-base-german-dbmdz-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-german-dbmdz-uncased-vocab.txt",
         'bert-base-finnish-cased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-cased-v1/vocab.txt",
-        'bert-base-finnish-uncased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-cased-v1/vocab.txt",
+        'bert-base-finnish-uncased-v1': "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-uncased-v1/vocab.txt",
     }
 }
 

From e347725d8c2fea7a89605d02d1f465594cf7df85 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Tue, 17 Dec 2019 23:32:52 +0100
Subject: [PATCH 413/505] More fine-grained control over pipeline creation with
 config argument.

---
 transformers/pipelines.py | 25 +++++++++++++++++--------
 1 file changed, 17 insertions(+), 8 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index dec7843baf..bcb4d9e054 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -497,7 +497,7 @@ class QuestionAnsweringPipeline(Pipeline):
                     'score': score.item(),
                     'start': np.where(char_to_word == feature.token_to_orig_map[s])[0][0].item(),
                     'end': np.where(char_to_word == feature.token_to_orig_map[e])[0][-1].item(),
-                    'answer': ' '.join(example.doc_tokens[feature.token_to_orig_map[s]: feature.token_to_orig_map[e] + 1])
+                    'answer': ' '.join(example.doc_tokens[feature.token_to_orig_map[s]:feature.token_to_orig_map[e] + 1])
                 }
                 for s, e, score in zip(starts, ends, scores)
             ]
@@ -612,7 +612,8 @@ SUPPORTED_TASKS = {
 }
 
 
-def pipeline(task: str, model, config: Optional[PretrainedConfig] = None, tokenizer: Optional[Union[str, PreTrainedTokenizer]] = None, **kwargs) -> Pipeline:
+def pipeline(task: str, model, config: Optional[Union[str, PretrainedConfig]] = None,
+             tokenizer: Optional[Union[str, PreTrainedTokenizer]] = None, **kwargs) -> Pipeline:
     """
     Utility factory method to build a pipeline.
     Pipeline are made of:
@@ -637,13 +638,21 @@ def pipeline(task: str, model, config: Optional[PretrainedConfig] = None, tokeni
     task, allocator = targeted_task['impl'], targeted_task['tf'] if is_tf_available() else targeted_task['pt']
 
     # Special handling for model conversion
-    from_tf = model.endswith('.h5') and not is_tf_available()
-    from_pt = model.endswith('.bin') and not is_torch_available()
+    if isinstance(model, str):
+        from_tf = model.endswith('.h5') and not is_tf_available()
+        from_pt = model.endswith('.bin') and not is_torch_available()
 
-    if from_tf:
-        logger.warning('Model might be a TensorFlow model (ending with `.h5`) but TensorFlow is not available. Trying to load the model with PyTorch.')
-    elif from_pt:
-        logger.warning('Model might be a PyTorch model (ending with `.bin`) but PyTorch is not available. Trying to load the model with Tensorflow.')
+        if from_tf:
+            logger.warning('Model might be a TensorFlow model (ending with `.h5`) but TensorFlow is not available. '
+                           'Trying to load the model with PyTorch.')
+        elif from_pt:
+            logger.warning('Model might be a PyTorch model (ending with `.bin`) but PyTorch is not available. '
+                           'Trying to load the model with Tensorflow.')
+    else:
+        from_tf = from_pt = False
+
+    if isinstance(config, str):
+        config = PretrainedConfig.from_pretrained(config)
 
     if allocator.__name__.startswith('TF'):
         model = allocator.from_pretrained(model, config=config, from_pt=from_pt)

From ca31abc6d6fe35a39703ed36775853595149e956 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Wed, 18 Dec 2019 11:36:54 +0100
Subject: [PATCH 414/505] tokenization: *align* fairseq and spm vocab to fix
 some tokenization errors

---
 transformers/tokenization_xlm_roberta.py | 18 +++++++++++++++---
 1 file changed, 15 insertions(+), 3 deletions(-)

diff --git a/transformers/tokenization_xlm_roberta.py b/transformers/tokenization_xlm_roberta.py
index 0f95397606..d8484e7f9c 100644
--- a/transformers/tokenization_xlm_roberta.py
+++ b/transformers/tokenization_xlm_roberta.py
@@ -61,7 +61,19 @@ class XLMRobertaTokenizer(PreTrainedTokenizer):
         self.sp_model = spm.SentencePieceProcessor()
         self.sp_model.Load(str(vocab_file))
         self.vocab_file = vocab_file
-        self.fairseq_tokens_to_ids = {"<s>": 0, "<unk>": 1, "</s>": 2}
+
+        # Original fairseq vocab and spm vocab must be "aligned":
+        # Vocab    |    0    |    1    |   2    |    3    |  4  |  5  |  6  |   7   |   8   |  9
+        # -------- | ------- | ------- | ------ | ------- | --- | --- | --- | ----- | ----- | ----
+        # fairseq  | '<s>'   | '<pad>' | '</s>' | '<unk>' | ',' | '.' | '▁' | 's'   | '▁de' | '-'
+        # spm      | '<unk>' | '<s>'   | '</s>' | ','     | '.' | '▁' | 's' | '▁de' | '-'   | '▁a'
+
+        # Mimic fairseq token-to-id alignment for the first 4 token
+        self.fairseq_tokens_to_ids = {"<s>": 0, "<pad>": 1, "</s>": 2, "<unk>": 3}
+
+        # The first "real" token "," has position 4 in the original fairseq vocab and position 3 in the spm vocab
+        self.fairseq_offset = 1
+
         self.fairseq_tokens_to_ids['<mask>'] = len(self.sp_model) + len(self.fairseq_tokens_to_ids)
         self.fairseq_ids_to_tokens = {v: k for k, v in self.fairseq_tokens_to_ids.items()}
 
@@ -131,13 +143,13 @@ class XLMRobertaTokenizer(PreTrainedTokenizer):
         """ Converts a token (str/unicode) in an id using the vocab. """
         if token in self.fairseq_tokens_to_ids:
             return self.fairseq_tokens_to_ids[token]
-        return self.sp_model.PieceToId(token) + 1
+        return self.sp_model.PieceToId(token) + self.fairseq_offset
 
     def _convert_id_to_token(self, index):
         """Converts an index (integer) in a token (string/unicode) using the vocab."""
         if index in self.fairseq_ids_to_tokens:
             return self.fairseq_ids_to_tokens[index]
-        return self.sp_model.IdToPiece(index + 1)
+        return self.sp_model.IdToPiece(index - self.fairseq_offset)
 
     def save_vocabulary(self, save_directory):
         """ Save the sentencepiece vocabulary (copy original file) and special tokens file

From 01b68be34f906a210b0c0a5e2bd9bc605c5983f2 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Wed, 18 Dec 2019 12:24:46 +0100
Subject: [PATCH 415/505] converter: remove XLM-RoBERTa specific script (can be
 done with the script for RoBERTa now)

---
 ..._original_pytorch_checkpoint_to_pytorch.py | 184 ------------------
 1 file changed, 184 deletions(-)
 delete mode 100644 transformers/convert_xlm_roberta_original_pytorch_checkpoint_to_pytorch.py

diff --git a/transformers/convert_xlm_roberta_original_pytorch_checkpoint_to_pytorch.py b/transformers/convert_xlm_roberta_original_pytorch_checkpoint_to_pytorch.py
deleted file mode 100644
index 884c273d2c..0000000000
--- a/transformers/convert_xlm_roberta_original_pytorch_checkpoint_to_pytorch.py
+++ /dev/null
@@ -1,184 +0,0 @@
-# coding=utf-8
-# Copyright 2018 The HuggingFace Inc. team.
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-"""Convert RoBERTa checkpoint."""
-
-from __future__ import absolute_import, division, print_function
-
-import argparse
-import logging
-import numpy as np
-import torch
-import pathlib
-
-from fairseq.models.roberta import RobertaModel as FairseqRobertaModel
-from fairseq.modules import TransformerSentenceEncoderLayer
-from transformers.modeling_bert import (BertConfig, BertEncoder,
-                                        BertIntermediate, BertLayer,
-                                        BertModel, BertOutput,
-                                        BertSelfAttention,
-                                        BertSelfOutput)
-from transformers.modeling_roberta import (RobertaEmbeddings,
-                                           RobertaForMaskedLM,
-                                           RobertaForSequenceClassification,
-                                           RobertaModel)
-
-logging.basicConfig(level=logging.INFO)
-logger = logging.getLogger(__name__)
-
-SAMPLE_TEXT = 'Hello world! cécé herlolip'
-
-
-def convert_roberta_checkpoint_to_pytorch(roberta_checkpoint_path, pytorch_dump_folder_path, classification_head):
-    """
-    Copy/paste/tweak roberta's weights to our BERT structure.
-    """
-    roberta = FairseqRobertaModel.from_pretrained(roberta_checkpoint_path, bpe = 'sentencepiece')
-    roberta.eval()  # disable dropout
-    config = BertConfig(
-        vocab_size=250002,
-        hidden_size=roberta.args.encoder_embed_dim,
-        num_hidden_layers=roberta.args.encoder_layers,
-        num_attention_heads=roberta.args.encoder_attention_heads,
-        intermediate_size=roberta.args.encoder_ffn_embed_dim,
-        max_position_embeddings=514,
-        type_vocab_size=1,
-        layer_norm_eps=1e-5, # PyTorch default used in fairseq
-    )
-    if classification_head:
-        config.num_labels = roberta.args.num_classes
-    print("Our BERT config:", config)
-
-    model = RobertaForSequenceClassification(config) if classification_head else RobertaForMaskedLM(config)
-    model.eval()
-
-    # Now let's copy all the weights.
-    # Embeddings
-    roberta_sent_encoder = roberta.model.decoder.sentence_encoder
-    model.roberta.embeddings.word_embeddings.weight = roberta_sent_encoder.embed_tokens.weight
-    model.roberta.embeddings.position_embeddings.weight = roberta_sent_encoder.embed_positions.weight
-    model.roberta.embeddings.token_type_embeddings.weight.data = torch.zeros_like(model.roberta.embeddings.token_type_embeddings.weight)  # just zero them out b/c RoBERTa doesn't use them.
-    model.roberta.embeddings.LayerNorm.weight = roberta_sent_encoder.emb_layer_norm.weight
-    model.roberta.embeddings.LayerNorm.bias = roberta_sent_encoder.emb_layer_norm.bias
-
-    for i in range(config.num_hidden_layers):
-        # Encoder: start of layer
-        layer: BertLayer = model.roberta.encoder.layer[i]
-        roberta_layer: TransformerSentenceEncoderLayer = roberta_sent_encoder.layers[i]
-
-        ### self attention
-        self_attn: BertSelfAttention = layer.attention.self
-        assert(
-            roberta_layer.self_attn.k_proj.weight.data.shape == \
-            roberta_layer.self_attn.q_proj.weight.data.shape == \
-            roberta_layer.self_attn.v_proj.weight.data.shape == \
-            torch.Size((config.hidden_size, config.hidden_size))
-        )
-
-        self_attn.query.weight.data = roberta_layer.self_attn.q_proj.weight
-        self_attn.query.bias.data = roberta_layer.self_attn.q_proj.bias
-        self_attn.key.weight.data = roberta_layer.self_attn.k_proj.weight
-        self_attn.key.bias.data = roberta_layer.self_attn.k_proj.bias
-        self_attn.value.weight.data = roberta_layer.self_attn.v_proj.weight
-        self_attn.value.bias.data = roberta_layer.self_attn.v_proj.bias
-
-        ### self-attention output
-        self_output: BertSelfOutput = layer.attention.output
-        assert(
-            self_output.dense.weight.shape == roberta_layer.self_attn.out_proj.weight.shape
-        )
-        self_output.dense.weight = roberta_layer.self_attn.out_proj.weight
-        self_output.dense.bias = roberta_layer.self_attn.out_proj.bias
-        self_output.LayerNorm.weight = roberta_layer.self_attn_layer_norm.weight
-        self_output.LayerNorm.bias = roberta_layer.self_attn_layer_norm.bias
-
-        ### intermediate
-        intermediate: BertIntermediate = layer.intermediate
-        assert(
-            intermediate.dense.weight.shape == roberta_layer.fc1.weight.shape
-        )
-        intermediate.dense.weight = roberta_layer.fc1.weight
-        intermediate.dense.bias = roberta_layer.fc1.bias
-
-        ### output
-        bert_output: BertOutput = layer.output
-        assert(
-            bert_output.dense.weight.shape == roberta_layer.fc2.weight.shape
-        )
-        bert_output.dense.weight = roberta_layer.fc2.weight
-        bert_output.dense.bias = roberta_layer.fc2.bias
-        bert_output.LayerNorm.weight = roberta_layer.final_layer_norm.weight
-        bert_output.LayerNorm.bias = roberta_layer.final_layer_norm.bias
-        #### end of layer
-    
-    if classification_head:
-        model.classifier.dense.weight = roberta.model.classification_heads['mnli'].dense.weight
-        model.classifier.dense.bias = roberta.model.classification_heads['mnli'].dense.bias
-        model.classifier.out_proj.weight = roberta.model.classification_heads['mnli'].out_proj.weight
-        model.classifier.out_proj.bias = roberta.model.classification_heads['mnli'].out_proj.bias
-    else:
-        # LM Head
-        model.lm_head.dense.weight = roberta.model.decoder.lm_head.dense.weight
-        model.lm_head.dense.bias = roberta.model.decoder.lm_head.dense.bias
-        model.lm_head.layer_norm.weight = roberta.model.decoder.lm_head.layer_norm.weight
-        model.lm_head.layer_norm.bias = roberta.model.decoder.lm_head.layer_norm.bias
-        model.lm_head.decoder.weight = roberta.model.decoder.lm_head.weight
-        model.lm_head.bias = roberta.model.decoder.lm_head.bias
-
-    # Let's check that we get the same results.
-    input_ids: torch.Tensor = roberta.encode(SAMPLE_TEXT).unsqueeze(0) # batch of size 1
-
-    our_output = model(input_ids)[0]
-    if classification_head:
-        their_output = roberta.model.classification_heads['mnli'](roberta.extract_features(input_ids))
-    else:
-        their_output = roberta.model(input_ids)[0]
-    print(our_output.shape, their_output.shape)
-    max_absolute_diff = torch.max(torch.abs(our_output - their_output)).item()
-    print(f"max_absolute_diff = {max_absolute_diff}") # ~ 1e-7
-    success = torch.allclose(our_output, their_output, atol=1e-3)
-    print(
-        "Do both models output the same tensors?",
-        "🔥" if success else "💩"
-    )
-    if not success:
-        raise Exception("Something went wRoNg")
-
-    pathlib.Path(pytorch_dump_folder_path).mkdir(parents=True, exist_ok=True)
-    print(f"Saving model to {pytorch_dump_folder_path}")
-    model.save_pretrained(pytorch_dump_folder_path)
-
-
-if __name__ == "__main__":
-    parser = argparse.ArgumentParser()
-    ## Required parameters
-    parser.add_argument("--roberta_checkpoint_path",
-                        default = None,
-                        type = str,
-                        required = True,
-                        help = "Path the official PyTorch dump.")
-    parser.add_argument("--pytorch_dump_folder_path",
-                        default = None,
-                        type = str,
-                        required = True,
-                        help = "Path to the output PyTorch model.")
-    parser.add_argument("--classification_head",
-                        action = "store_true",
-                        help = "Whether to convert a final classification head.")
-    args = parser.parse_args()
-    convert_roberta_checkpoint_to_pytorch(
-        args.roberta_checkpoint_path,
-        args.pytorch_dump_folder_path,
-        args.classification_head
-    )

From 8efc6dd544bf1a30d99d4b5abfc5e214699eab2b Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Wed, 18 Dec 2019 10:47:59 -0500
Subject: [PATCH 416/505] fix #2214

---
 transformers/configuration_xlm.py | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/transformers/configuration_xlm.py b/transformers/configuration_xlm.py
index 0740cc4026..6839a45746 100644
--- a/transformers/configuration_xlm.py
+++ b/transformers/configuration_xlm.py
@@ -144,6 +144,9 @@ class XLMConfig(PretrainedConfig):
         self.start_n_top = start_n_top
         self.end_n_top = end_n_top
 
+        if "n_words" in kwargs:
+            self.n_words = kwargs["n_words"]
+
     @property
     def n_words(self):  # For backward compatibility
         return self.vocab_size

From 0c88c856d592134ee5a9a636f9b73f40b91784b5 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Wed, 18 Dec 2019 18:18:16 +0100
Subject: [PATCH 417/505] Unnest QuestionAnsweringArgumentHandler

---
 transformers/pipelines.py | 62 +++++++++++++++++++++++++++++++++++++--
 1 file changed, 60 insertions(+), 2 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index bcb4d9e054..a10078b027 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -333,6 +333,63 @@ class NerPipeline(Pipeline):
         return answers
 
 
+class QuestionAnsweringArgumentHandler(ArgumentHandler):
+    """
+    QuestionAnsweringPipeline requires the user to provide multiple arguments (i.e. question & context) to be mapped
+    to internal SquadExample / SquadFeature structures.
+
+    QuestionAnsweringArgumentHandler manages all the possible to create SquadExample from the command-line supplied
+    arguments.
+    """
+    def __call__(self, *args, **kwargs):
+        # Position args, handling is sensibly the same as X and data, so forwarding to avoid duplicating
+        if args is not None and len(args) > 0:
+            if len(args) == 1:
+                kwargs['X'] = args[0]
+            else:
+                kwargs['X'] = list(args)
+
+            # Generic compatibility with sklearn and Keras
+            # Batched data
+        if 'X' in kwargs or 'data' in kwargs:
+            data = kwargs['X'] if 'X' in kwargs else kwargs['data']
+
+            if not isinstance(data, list):
+                data = [data]
+
+            for i, item in enumerate(data):
+                if isinstance(item, dict):
+                    if any(k not in item for k in ['question', 'context']):
+                        raise KeyError('You need to provide a dictionary with keys {question:..., context:...}')
+                    data[i] = QuestionAnsweringPipeline.create_sample(**item)
+
+                elif isinstance(item, SquadExample):
+                    continue
+                else:
+                    raise ValueError(
+                        '{} argument needs to be of type (list[SquadExample | dict], SquadExample, dict)'
+                            .format('X' if 'X' in kwargs else 'data')
+                    )
+            inputs = data
+
+            # Tabular input
+        elif 'question' in kwargs and 'context' in kwargs:
+            if isinstance(kwargs['question'], str):
+                kwargs['question'] = [kwargs['question']]
+
+            if isinstance(kwargs['context'], str):
+                kwargs['context'] = [kwargs['context']]
+
+            inputs = [QuestionAnsweringPipeline.create_sample(q, c) for q, c in zip(kwargs['question'], kwargs['context'])]
+        else:
+            raise ValueError('Unknown arguments {}'.format(kwargs))
+
+        if not isinstance(inputs, list):
+            inputs = [inputs]
+
+        return inputs
+
+
 class QuestionAnsweringPipeline(Pipeline):
     """
     Question Answering pipeline using ModelForQuestionAnswering head.
@@ -403,8 +460,9 @@ class QuestionAnsweringPipeline(Pipeline):
         else:
             return SquadExample(None, question, context, None, None, None)
 
-    def __init__(self, model, tokenizer: Optional[PreTrainedTokenizer]):
-        super().__init__(model, tokenizer, args_parser=QuestionAnsweringPipeline.QuestionAnsweringArgumentHandler())
+    def __init__(self, model, tokenizer: Optional[PreTrainedTokenizer], device: int = -1, **kwargs):
+        super().__init__(model, tokenizer, args_parser=QuestionAnsweringArgumentHandler(),
+                         device=device, **kwargs)
 
     def inputs_for_model(self, features: Union[SquadExample, List[SquadExample]]) -> Dict:
         """

From 41a13a6375817ec3836bedcec67d12ad32bf0956 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Wed, 18 Dec 2019 18:20:27 +0100
Subject: [PATCH 418/505] auto: add XLMRoBERTa to auto configuration

---
 transformers/configuration_auto.py | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/transformers/configuration_auto.py b/transformers/configuration_auto.py
index 9fe58f173a..29e1a0ee1d 100644
--- a/transformers/configuration_auto.py
+++ b/transformers/configuration_auto.py
@@ -30,6 +30,7 @@ from .configuration_distilbert import DistilBertConfig, DISTILBERT_PRETRAINED_CO
 from .configuration_albert import AlbertConfig, ALBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_camembert import CamembertConfig, CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP
 from .configuration_t5 import T5Config, T5_PRETRAINED_CONFIG_ARCHIVE_MAP
+from .configuration_xlm_roberta import XLMRobertaConfig, XLM_ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP
 
 logger = logging.getLogger(__name__)
 
@@ -48,6 +49,7 @@ ALL_PRETRAINED_CONFIG_ARCHIVE_MAP = dict((key, value)
         ALBERT_PRETRAINED_CONFIG_ARCHIVE_MAP,
         CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP,
         T5_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        XLM_ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP,
         ]
     for key, value, in pretrained_map.items())
 
@@ -66,6 +68,7 @@ class AutoConfig(object):
             - contains `distilbert`: DistilBertConfig (DistilBERT model)
             - contains `albert`: AlbertConfig (ALBERT model)
             - contains `camembert`: CamembertConfig (CamemBERT model)
+            - contains `xlm-roberta`: XLMRobertaConfig (XLM-RoBERTa model)
             - contains `roberta`: RobertaConfig (RoBERTa model)
             - contains `bert`: BertConfig (Bert model)
             - contains `openai-gpt`: OpenAIGPTConfig (OpenAI GPT model)
@@ -91,6 +94,7 @@ class AutoConfig(object):
             - contains `distilbert`: DistilBertConfig (DistilBERT model)
             - contains `albert`: AlbertConfig (ALBERT model)
             - contains `camembert`: CamembertConfig (CamemBERT model)
+            - contains `xlm-roberta`: XLMRobertaConfig (XLM-RoBERTa model)
             - contains `roberta`: RobertaConfig (RoBERTa model)
             - contains `bert`: BertConfig (Bert model)
             - contains `openai-gpt`: OpenAIGPTConfig (OpenAI GPT model)
@@ -152,6 +156,8 @@ class AutoConfig(object):
             return AlbertConfig.from_pretrained(pretrained_model_name_or_path, **kwargs)
         elif 'camembert' in pretrained_model_name_or_path:
             return CamembertConfig.from_pretrained(pretrained_model_name_or_path, **kwargs)
+        elif 'xlm-roberta' in pretrained_model_name_or_path:
+            return XLMRobertaConfig.from_pretrained(pretrained_model_name_or_path, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return RobertaConfig.from_pretrained(pretrained_model_name_or_path, **kwargs)
         elif 'bert' in pretrained_model_name_or_path:
@@ -170,4 +176,4 @@ class AutoConfig(object):
             return CTRLConfig.from_pretrained(pretrained_model_name_or_path, **kwargs)
         raise ValueError("Unrecognized model identifier in {}. Should contains one of "
                          "'bert', 'openai-gpt', 'gpt2', 'transfo-xl', 'xlnet', "
-                         "'xlm', 'roberta', 'distilbert', 'camembert', 'ctrl', 'albert'".format(pretrained_model_name_or_path))
+                         "'xlm-roberta', 'xlm', 'roberta', 'distilbert', 'camembert', 'ctrl', 'albert'".format(pretrained_model_name_or_path))

From 036831e2791b84edb5a89db7a92af7c69f4ff37e Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Wed, 18 Dec 2019 18:23:42 +0100
Subject: [PATCH 419/505] auto: add XLM-RoBERTa to audo modeling

---
 transformers/modeling_auto.py | 20 +++++++++++++++++---
 1 file changed, 17 insertions(+), 3 deletions(-)

diff --git a/transformers/modeling_auto.py b/transformers/modeling_auto.py
index 1a30ea4623..ca6c4525b8 100644
--- a/transformers/modeling_auto.py
+++ b/transformers/modeling_auto.py
@@ -30,6 +30,7 @@ from .modeling_distilbert import DistilBertModel, DistilBertForQuestionAnswering
 from .modeling_camembert import CamembertModel, CamembertForMaskedLM, CamembertForSequenceClassification, CamembertForMultipleChoice, CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP
 from .modeling_albert import AlbertModel, AlbertForMaskedLM, AlbertForSequenceClassification, AlbertForQuestionAnswering, ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP
 from .modeling_t5 import T5Model, T5WithLMHeadModel, T5_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_xlm_roberta import XLMRobertaModel, XLMRobertaForMaskedLM, XLMRobertaForSequenceClassification, XLMRobertaForMultipleChoice, XLM_ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP
 
 from .modeling_utils import PreTrainedModel, SequenceSummary
 
@@ -52,6 +53,7 @@ ALL_PRETRAINED_MODEL_ARCHIVE_MAP = dict((key, value)
         ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP,
         CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP,
         T5_PRETRAINED_MODEL_ARCHIVE_MAP,
+        XLM_ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP,
         ]
     for key, value, in pretrained_map.items())
 
@@ -72,6 +74,7 @@ class AutoModel(object):
             - contains `distilbert`: DistilBertModel (DistilBERT model)
             - contains `albert`: AlbertModel (ALBERT model)
             - contains `camembert`: CamembertModel (CamemBERT model)
+            - contains `xlm-roberta`: XLMRobertaModel (XLM-RoBERTa model)
             - contains `roberta`: RobertaModel (RoBERTa model)
             - contains `bert`: BertModel (Bert model)
             - contains `openai-gpt`: OpenAIGPTModel (OpenAI GPT model)
@@ -98,6 +101,7 @@ class AutoModel(object):
             - contains `distilbert`: DistilBertModel (DistilBERT model)
             - contains `albert`: AlbertModel (ALBERT model)
             - contains `camembert`: CamembertModel (CamemBERT model)
+            - contains `xlm-roberta`: XLMRobertaModel (XLM-RoBERTa model)
             - contains `roberta`: RobertaModel (RoBERTa model)
             - contains `bert`: BertModel (Bert model)
             - contains `openai-gpt`: OpenAIGPTModel (OpenAI GPT model)
@@ -175,6 +179,8 @@ class AutoModel(object):
             return AlbertModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'camembert' in pretrained_model_name_or_path:
             return CamembertModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'xlm-roberta' in pretrained_model_name_or_path:
+            return XLMRobertaModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return RobertaModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'bert' in pretrained_model_name_or_path:
@@ -193,7 +199,7 @@ class AutoModel(object):
             return CTRLModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         raise ValueError("Unrecognized model identifier in {}. Should contains one of "
                          "'bert', 'openai-gpt', 'gpt2', 'transfo-xl', 'xlnet', "
-                         "'xlm', 'roberta, 'ctrl', 'distilbert', 'camembert', 'albert'".format(pretrained_model_name_or_path))
+                         "'xlm-roberta', 'xlm', 'roberta, 'ctrl', 'distilbert', 'camembert', 'albert'".format(pretrained_model_name_or_path))
 
 
 class AutoModelWithLMHead(object):
@@ -212,6 +218,7 @@ class AutoModelWithLMHead(object):
             - contains `distilbert`: DistilBertForMaskedLM (DistilBERT model)
             - contains `albert`: AlbertForMaskedLM (ALBERT model)
             - contains `camembert`: CamembertForMaskedLM (CamemBERT model)
+            - contains `xlm-roberta`: XLMRobertaForMaskedLM (XLM-RoBERTa model)
             - contains `roberta`: RobertaForMaskedLM (RoBERTa model)
             - contains `bert`: BertForMaskedLM (Bert model)
             - contains `openai-gpt`: OpenAIGPTLMHeadModel (OpenAI GPT model)
@@ -241,6 +248,7 @@ class AutoModelWithLMHead(object):
             - contains `distilbert`: DistilBertForMaskedLM (DistilBERT model)
             - contains `albert`: AlbertForMaskedLM (ALBERT model)
             - contains `camembert`: CamembertForMaskedLM (CamemBERT model)
+            - contains `xlm-roberta`: XLMRobertaForMaskedLM (XLM-RoBERTa model)
             - contains `roberta`: RobertaForMaskedLM (RoBERTa model)
             - contains `bert`: BertForMaskedLM (Bert model)
             - contains `openai-gpt`: OpenAIGPTLMHeadModel (OpenAI GPT model)
@@ -317,6 +325,8 @@ class AutoModelWithLMHead(object):
             return AlbertForMaskedLM.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'camembert' in pretrained_model_name_or_path:
             return CamembertForMaskedLM.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'xlm-roberta' in pretrained_model_name_or_path:
+            return XLMRobertaForMaskedLM.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return RobertaForMaskedLM.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'bert' in pretrained_model_name_or_path:
@@ -335,7 +345,7 @@ class AutoModelWithLMHead(object):
             return CTRLLMHeadModel.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         raise ValueError("Unrecognized model identifier in {}. Should contains one of "
                          "'bert', 'openai-gpt', 'gpt2', 'transfo-xl', 'xlnet', "
-                         "'xlm', 'roberta','ctrl', 'distilbert', 'camembert', 'albert'".format(pretrained_model_name_or_path))
+                         "'xlm-roberta', 'xlm', 'roberta','ctrl', 'distilbert', 'camembert', 'albert'".format(pretrained_model_name_or_path))
 
 
 class AutoModelForSequenceClassification(object):
@@ -353,6 +363,7 @@ class AutoModelForSequenceClassification(object):
             - contains `distilbert`: DistilBertForSequenceClassification (DistilBERT model)
             - contains `albert`: AlbertForSequenceClassification (ALBERT model)
             - contains `camembert`: CamembertForSequenceClassification (CamemBERT model)
+            - contains `xlm-roberta`: XLMRobertaForSequenceClassification (XLM-RoBERTa model)
             - contains `roberta`: RobertaForSequenceClassification (RoBERTa model)
             - contains `bert`: BertForSequenceClassification (Bert model)
             - contains `xlnet`: XLNetForSequenceClassification (XLNet model)
@@ -377,6 +388,7 @@ class AutoModelForSequenceClassification(object):
             - contains `distilbert`: DistilBertForSequenceClassification (DistilBERT model)
             - contains `albert`: AlbertForSequenceClassification (ALBERT model)
             - contains `camembert`: CamembertForSequenceClassification (CamemBERT model)
+            - contains `xlm-roberta`: XLMRobertaForSequenceClassification (XLM-RoBERTa model)
             - contains `roberta`: RobertaForSequenceClassification (RoBERTa model)
             - contains `bert`: BertForSequenceClassification (Bert model)
             - contains `xlnet`: XLNetForSequenceClassification (XLNet model)
@@ -448,6 +460,8 @@ class AutoModelForSequenceClassification(object):
             return AlbertForSequenceClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'camembert' in pretrained_model_name_or_path:
             return CamembertForSequenceClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'xlm-roberta' in pretrained_model_name_or_path:
+            return XLMRobertaForSequenceClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return RobertaForSequenceClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'bert' in pretrained_model_name_or_path:
@@ -458,7 +472,7 @@ class AutoModelForSequenceClassification(object):
             return XLMForSequenceClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
 
         raise ValueError("Unrecognized model identifier in {}. Should contains one of "
-                         "'bert', 'xlnet', 'xlm', 'roberta', 'distilbert', 'camembert', 'albert'".format(pretrained_model_name_or_path))
+                         "'bert', 'xlnet', 'xlm-roberta', 'xlm', 'roberta', 'distilbert', 'camembert', 'albert'".format(pretrained_model_name_or_path))
 
 
 class AutoModelForQuestionAnswering(object):

From 64a971a9156788ed6d95f850453578ecb74069c5 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Wed, 18 Dec 2019 18:24:32 +0100
Subject: [PATCH 420/505] auto: add XLM-RoBERTa to auto tokenization

---
 transformers/tokenization_auto.py | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/transformers/tokenization_auto.py b/transformers/tokenization_auto.py
index 173dee0e2b..5377bd48cb 100644
--- a/transformers/tokenization_auto.py
+++ b/transformers/tokenization_auto.py
@@ -31,6 +31,7 @@ from .tokenization_distilbert import DistilBertTokenizer
 from .tokenization_camembert import CamembertTokenizer
 from .tokenization_albert import AlbertTokenizer
 from .tokenization_t5 import T5Tokenizer
+from .tokenization_xlm_roberta import XLMRobertaTokenizer
 
 logger = logging.getLogger(__name__)
 
@@ -49,6 +50,7 @@ class AutoTokenizer(object):
             - contains `distilbert`: DistilBertTokenizer (DistilBert model)
             - contains `albert`: AlbertTokenizer (ALBERT model)
             - contains `camembert`: CamembertTokenizer (CamemBERT model)
+            - contains `xlm-roberta`: XLMRobertaTokenizer (XLM-RoBERTa model)
             - contains `roberta`: RobertaTokenizer (RoBERTa model)
             - contains `bert`: BertTokenizer (Bert model)
             - contains `openai-gpt`: OpenAIGPTTokenizer (OpenAI GPT model)
@@ -75,6 +77,7 @@ class AutoTokenizer(object):
             - contains `distilbert`: DistilBertTokenizer (DistilBert model)
             - contains `albert`: AlbertTokenizer (ALBERT model)
             - contains `camembert`: CamembertTokenizer (CamemBERT model)
+            - contains `xlm-roberta`: XLMRobertaTokenizer (XLM-RoBERTa model)
             - contains `roberta`: RobertaTokenizer (RoBERTa model)
             - contains `bert-base-japanese`: BertJapaneseTokenizer (Bert model)
             - contains `bert`: BertTokenizer (Bert model)
@@ -130,6 +133,8 @@ class AutoTokenizer(object):
             return AlbertTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
         elif 'camembert' in pretrained_model_name_or_path:
             return CamembertTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
+        elif 'xlm-roberta' in pretrained_model_name_or_path:
+            return XLMRobertaTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return RobertaTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
         elif 'bert-base-japanese' in pretrained_model_name_or_path:
@@ -150,4 +155,4 @@ class AutoTokenizer(object):
             return CTRLTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
         raise ValueError("Unrecognized model identifier in {}. Should contains one of "
                          "'bert', 'openai-gpt', 'gpt2', 'transfo-xl', 'xlnet', "
-                         "'xlm', 'roberta', 'distilbert,' 'camembert', 'ctrl', 'albert'".format(pretrained_model_name_or_path))
+                         "'xlm-roberta', 'xlm', 'roberta', 'distilbert,' 'camembert', 'ctrl', 'albert'".format(pretrained_model_name_or_path))

From 04b602f96f91529a1259909466705f3f9192113c Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Wed, 18 Dec 2019 18:28:39 +0100
Subject: [PATCH 421/505] Put module import on top of the module.

---
 transformers/pipelines.py | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index a10078b027..92c94268a7 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -20,6 +20,7 @@ import os
 from abc import ABC, abstractmethod
 from contextlib import contextmanager
 from itertools import groupby
+from os.path import abspath, exists
 from typing import Union, Optional, Tuple, List, Dict
 
 import numpy as np
@@ -100,12 +101,12 @@ class PipelineDataFormat:
         if self.is_multi_columns:
             self.column = [tuple(c.split('=')) if '=' in c else (c, c) for c in self.column]
 
-        from os.path import abspath, exists
-        if exists(abspath(self.output)):
-            raise OSError('{} already exists on disk'.format(self.output))
+        if output is not None:
+            if exists(abspath(self.output)):
+                raise OSError('{} already exists on disk'.format(self.output))
 
-        if not exists(abspath(self.path)):
-            raise OSError('{} doesnt exist on disk'.format(self.path))
+            if not exists(abspath(self.path)):
+                raise OSError('{} doesnt exist on disk'.format(self.path))
 
     @abstractmethod
     def __iter__(self):

From e778dd854dd5d1fd29396d214577ddbe0f854247 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Wed, 18 Dec 2019 19:27:34 +0100
Subject: [PATCH 422/505] modeling: add XLM-RoBERTa base model

---
 transformers/modeling_xlm_roberta.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/transformers/modeling_xlm_roberta.py b/transformers/modeling_xlm_roberta.py
index 4c833c69ff..abace25d5b 100644
--- a/transformers/modeling_xlm_roberta.py
+++ b/transformers/modeling_xlm_roberta.py
@@ -27,6 +27,7 @@ from .file_utils import add_start_docstrings
 logger = logging.getLogger(__name__)
 
 XLM_ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP = {
+    'xlm-roberta-base': "https://schweter.eu/cloud/transformers/xlm-roberta-base-pytorch_model.bin",
     'xlm-roberta-large': "https://schweter.eu/cloud/transformers/xlm-roberta-large-pytorch_model.bin",
 }
 

From 128cfdee9bdd13f2f0bbef977907ff014ae398ad Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Wed, 18 Dec 2019 19:28:16 +0100
Subject: [PATCH 423/505] tokenization add XLM-RoBERTa base model

---
 transformers/tokenization_xlm_roberta.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/transformers/tokenization_xlm_roberta.py b/transformers/tokenization_xlm_roberta.py
index d8484e7f9c..93b1c397e4 100644
--- a/transformers/tokenization_xlm_roberta.py
+++ b/transformers/tokenization_xlm_roberta.py
@@ -30,11 +30,13 @@ VOCAB_FILES_NAMES = {'vocab_file': 'sentencepiece.bpe.model'}
 PRETRAINED_VOCAB_FILES_MAP = {
     'vocab_file':
     {
+    'xlm-roberta-base': "https://schweter.eu/cloud/transformers/xlm-roberta-base-sentencepiece.bpe.model",
     'xlm-roberta-large': "https://schweter.eu/cloud/transformers/xlm-roberta-large-sentencepiece.bpe.model",
     }
 }
 
 PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
+    'xlm-roberta-base': None,
     'xlm-roberta-large': None,
 }
 

From 3e89fca54359bff728782cce4157118caed33c09 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Wed, 18 Dec 2019 19:44:23 +0100
Subject: [PATCH 424/505] readme: add XLM-RoBERTa to model architecture list

---
 README.md | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/README.md b/README.md
index c33a65bdbb..a9d0fb3ace 100644
--- a/README.md
+++ b/README.md
@@ -146,7 +146,8 @@ At some point in the future, you'll be able to seamlessly move from pre-training
 10. **[CamemBERT](https://camembert-model.fr)** (from Inria/Facebook/Sorbonne) released with the paper [CamemBERT: a Tasty French Language Model](https://arxiv.org/abs/1911.03894) by Louis Martin*, Benjamin Muller*, Pedro Javier Ortiz Suárez*, Yoann Dupont, Laurent Romary, Éric Villemonte de la Clergerie, Djamé Seddah and Benoît Sagot.
 11. **[ALBERT](https://github.com/google-research/ALBERT)** (from Google Research and the Toyota Technological Institute at Chicago) released with the paper [ALBERT: A Lite BERT for Self-supervised Learning of Language Representations](https://arxiv.org/abs/1909.11942), by Zhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut.
 12. **[T5](https://github.com/google-research/text-to-text-transfer-transformer)** (from Google AI) released with the paper [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://arxiv.org/abs/1910.10683) by Colin Raffel and Noam Shazeer and Adam Roberts and Katherine Lee and Sharan Narang and Michael Matena and Yanqi Zhou and Wei Li and Peter J. Liu.
-13. Want to contribute a new model? We have added a **detailed guide and templates** to guide you in the process of adding a new model. You can find them in the [`templates`](./templates) folder of the repository. Be sure to check the [contributing guidelines](./CONTRIBUTING.md) and contact the maintainers or open an issue to collect feedbacks before starting your PR.
+13. **[XLM-RoBERTa](https://github.com/pytorch/fairseq/tree/master/examples/xlmr)** (from Facebook AI), released together with the paper [Unsupervised Cross-lingual Representation Learning at Scale](https://arxiv.org/abs/1911.02116) by Alexis Conneau*, Kartikay Khandelwal*, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer and Veselin Stoyanov.
+14. Want to contribute a new model? We have added a **detailed guide and templates** to guide you in the process of adding a new model. You can find them in the [`templates`](./templates) folder of the repository. Be sure to check the [contributing guidelines](./CONTRIBUTING.md) and contact the maintainers or open an issue to collect feedbacks before starting your PR.
 
 These implementations have been tested on several datasets (see the example scripts) and should match the performances of the original implementations (e.g. ~93 F1 on SQuAD for BERT Whole-Word-Masking, ~88 F1 on RocStories for OpenAI GPT, ~18.3 perplexity on WikiText 103 for Transformer-XL, ~0.916 Peason R coefficient on STS-B for XLNet). You can find more details on the performances in the Examples section of the [documentation](https://huggingface.co/transformers/examples.html).
 
@@ -168,7 +169,7 @@ import torch
 from transformers import *
 
 # Transformers has a unified API
-# for 8 transformer architectures and 30 pretrained weights.
+# for 10 transformer architectures and 30 pretrained weights.
 #          Model          | Tokenizer          | Pretrained weights shortcut
 MODELS = [(BertModel,       BertTokenizer,       'bert-base-uncased'),
           (OpenAIGPTModel,  OpenAIGPTTokenizer,  'openai-gpt'),
@@ -178,7 +179,9 @@ MODELS = [(BertModel,       BertTokenizer,       'bert-base-uncased'),
           (XLNetModel,      XLNetTokenizer,      'xlnet-base-cased'),
           (XLMModel,        XLMTokenizer,        'xlm-mlm-enfr-1024'),
           (DistilBertModel, DistilBertTokenizer, 'distilbert-base-uncased'),
-          (RobertaModel,    RobertaTokenizer,    'roberta-base')]
+          (RobertaModel,    RobertaTokenizer,    'roberta-base'),
+          (XLMRobertaModel, XLMRobertaTokenizer, 'xlm-roberta-base'),
+         ]
 
 # To use TensorFlow 2.0 versions of the models, simply prefix the class names with 'TF', e.g. `TFRobertaModel` is the TF 2.0 counterpart of the PyTorch model `RobertaModel`
 

From d35405b7a32eadf4fb1200249b2bbc4c12fb0340 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Wed, 18 Dec 2019 19:45:10 +0100
Subject: [PATCH 425/505] docs: add XLM-RoBERTa to index page

---
 docs/source/index.rst | 1 +
 1 file changed, 1 insertion(+)

diff --git a/docs/source/index.rst b/docs/source/index.rst
index 48282c1c6c..cb34c5c7f0 100644
--- a/docs/source/index.rst
+++ b/docs/source/index.rst
@@ -50,6 +50,7 @@ The library currently contains PyTorch and Tensorflow implementations, pre-train
 9. `CTRL <https://github.com/pytorch/fairseq/tree/master/examples/ctrl>`_ (from Salesforce), released together with the paper `CTRL: A Conditional Transformer Language Model for Controllable Generation <https://www.github.com/salesforce/ctrl>`_ by Nitish Shirish Keskar*, Bryan McCann*, Lav R. Varshney, Caiming Xiong and Richard Socher.
 10. `CamemBERT <https://huggingface.co/transformers/model_doc/camembert.html>`_ (from FAIR, Inria, Sorbonne Université) released together with the paper `CamemBERT: a Tasty French Language Model <https://arxiv.org/abs/1911.03894>`_ by Louis Martin, Benjamin Muller, Pedro Javier Ortiz Suarez, Yoann Dupont, Laurent Romary, Eric Villemonte de la Clergerie, Djame Seddah, and Benoît Sagot.
 11. `ALBERT <https://github.com/google-research/ALBERT>`_ (from Google Research), released together with the paper a `ALBERT: A Lite BERT for Self-supervised Learning of Language Representations <https://arxiv.org/abs/1909.11942>`_ by Zhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut.
+13. `XLM-RoBERTa <https://github.com/pytorch/fairseq/tree/master/examples/xlmr>`_ (from Facebook AI), released together with the paper `Unsupervised Cross-lingual Representation Learning at Scale <https://arxiv.org/abs/1911.02116>`_ by Alexis Conneau*, Kartikay Khandelwal*, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer and Veselin Stoyanov.
 
 .. toctree::
     :maxdepth: 2

From dd7a958fd6963d09850ad4842307d1d1064d096d Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Wed, 18 Dec 2019 19:45:46 +0100
Subject: [PATCH 426/505] docs: add XLM-RoBERTa to pretrained model list (incl.
 all parameters)

---
 docs/source/pretrained_models.rst | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/docs/source/pretrained_models.rst b/docs/source/pretrained_models.rst
index 7d037da34f..a359990f5a 100644
--- a/docs/source/pretrained_models.rst
+++ b/docs/source/pretrained_models.rst
@@ -240,6 +240,12 @@ Here is the full list of the currently provided pretrained models together with
 |                   | ``t5-11B``                                                 | | ~11B parameters with 24-layers, 1024-hidden-state, 65536 feed-forward hidden-state, 128-heads,                                      |
 |                   |                                                            | | Trained on English text: the Colossal Clean Crawled Corpus (C4)                                                                     |
 +-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+| XLM-RoBERTa       | ``xlm-roberta-base``                                       | | ~125M parameters with 12-layers, 768-hidden-state, 3072 feed-forward hidden-state, 8-heads,                                         |
+|                   |                                                            | | Trained on on 2.5 TB of newly created clean CommonCrawl data in 100 languages                                                       |
+|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   | ``xlm-roberta-large``                                      | | ~355M parameters with 24-layers, 1027-hidden-state, 4096 feed-forward hidden-state, 16-heads,                                       |
+|                   |                                                            | | Trained on 2.5 TB of newly created clean CommonCrawl data in 100 languages                                                          |
++-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 
 
 .. <https://huggingface.co/transformers/examples.html>`__

From f09d9996413f2b265f1c672d7a4b438e4c5099c4 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Wed, 18 Dec 2019 19:49:33 +0100
Subject: [PATCH 427/505] =?UTF-8?q?docs:=20fix=20numbering=20=F0=9F=98=85?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 docs/source/index.rst | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/source/index.rst b/docs/source/index.rst
index cb34c5c7f0..0ac9c740a5 100644
--- a/docs/source/index.rst
+++ b/docs/source/index.rst
@@ -50,7 +50,7 @@ The library currently contains PyTorch and Tensorflow implementations, pre-train
 9. `CTRL <https://github.com/pytorch/fairseq/tree/master/examples/ctrl>`_ (from Salesforce), released together with the paper `CTRL: A Conditional Transformer Language Model for Controllable Generation <https://www.github.com/salesforce/ctrl>`_ by Nitish Shirish Keskar*, Bryan McCann*, Lav R. Varshney, Caiming Xiong and Richard Socher.
 10. `CamemBERT <https://huggingface.co/transformers/model_doc/camembert.html>`_ (from FAIR, Inria, Sorbonne Université) released together with the paper `CamemBERT: a Tasty French Language Model <https://arxiv.org/abs/1911.03894>`_ by Louis Martin, Benjamin Muller, Pedro Javier Ortiz Suarez, Yoann Dupont, Laurent Romary, Eric Villemonte de la Clergerie, Djame Seddah, and Benoît Sagot.
 11. `ALBERT <https://github.com/google-research/ALBERT>`_ (from Google Research), released together with the paper a `ALBERT: A Lite BERT for Self-supervised Learning of Language Representations <https://arxiv.org/abs/1909.11942>`_ by Zhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut.
-13. `XLM-RoBERTa <https://github.com/pytorch/fairseq/tree/master/examples/xlmr>`_ (from Facebook AI), released together with the paper `Unsupervised Cross-lingual Representation Learning at Scale <https://arxiv.org/abs/1911.02116>`_ by Alexis Conneau*, Kartikay Khandelwal*, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer and Veselin Stoyanov.
+12. `XLM-RoBERTa <https://github.com/pytorch/fairseq/tree/master/examples/xlmr>`_ (from Facebook AI), released together with the paper `Unsupervised Cross-lingual Representation Learning at Scale <https://arxiv.org/abs/1911.02116>`_ by Alexis Conneau*, Kartikay Khandelwal*, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer and Veselin Stoyanov.
 
 .. toctree::
     :maxdepth: 2

From 8bb166db5de426ecfb2c3bf1160307324419bbda Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Wed, 18 Dec 2019 22:53:19 +0100
Subject: [PATCH 428/505] Expose more information in the output of
 TextClassificationPipeline

---
 transformers/pipelines.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 92c94268a7..809096a30c 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -283,7 +283,9 @@ class TextClassificationPipeline(Pipeline):
         self._nb_classes = nb_classes
 
     def __call__(self, *args, **kwargs):
-        return super().__call__(*args, **kwargs).tolist()
+        outputs = super().__call__(*args, **kwargs)
+        scores = np.exp(outputs) / np.exp(outputs).sum(-1)
+        return [{'label': self.model.config.id2label[item.argmax()], 'score': item.max()} for item in scores]
 
 
 class NerPipeline(Pipeline):

From 7711403bbdefad62e7ee88a88e04ec08b53412bc Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Wed, 18 Dec 2019 22:59:51 +0100
Subject: [PATCH 429/505] Expose config through the cli arguments

---
 transformers/commands/run.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/transformers/commands/run.py b/transformers/commands/run.py
index 8c203699a8..44c1127803 100644
--- a/transformers/commands/run.py
+++ b/transformers/commands/run.py
@@ -16,7 +16,7 @@ def try_infer_format_from_ext(path: str):
 
 
 def run_command_factory(args):
-    nlp = pipeline(task=args.task, model=args.model, tokenizer=args.tokenizer, device=args.device)
+    nlp = pipeline(task=args.task, model=args.model, config=args.config, tokenizer=args.tokenizer, device=args.device)
     format = try_infer_format_from_ext(args.input) if args.format == 'infer' else args.format
     reader = PipelineDataFormat.from_str(format, args.output, args.input, args.column)
     return RunCommand(nlp, reader)
@@ -34,6 +34,7 @@ class RunCommand(BaseTransformersCLICommand):
         run_parser.add_argument('--device', type=int, default=-1, help='Indicate the device to run onto, -1 indicates CPU, >= 0 indicates GPU (default: -1)')
         run_parser.add_argument('--task', choices=SUPPORTED_TASKS.keys(), help='Task to run')
         run_parser.add_argument('--model', type=str, required=True, help='Name or path to the model to instantiate.')
+        run_parser.add_argument('--config', type=str, help='Name or path to the model\'s config to instantiate.')
         run_parser.add_argument('--tokenizer', type=str, help='Name of the tokenizer to use. (default: same as the model name)')
         run_parser.add_argument('--column', type=str, required=True, help='Name of the column to use as input. (For multi columns input as QA use column1,columns2)')
         run_parser.add_argument('--format', type=str, default='infer', choices=PipelineDataFormat.SUPPORTED_FORMATS, help='Input format to read from')

From d0724d0794e930eb5821d4289130beade3359e87 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Wed, 18 Dec 2019 23:27:26 +0100
Subject: [PATCH 430/505] Add PipedPipelineDataFormat

---
 transformers/__init__.py     |   4 +-
 transformers/commands/run.py |   7 +-
 transformers/pipelines.py    | 154 +++++++++++++++--------------------
 3 files changed, 70 insertions(+), 95 deletions(-)

diff --git a/transformers/__init__.py b/transformers/__init__.py
index c474696062..87ede9f6a8 100755
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -191,8 +191,8 @@ from .modeling_tf_pytorch_utils import (convert_tf_weight_name_to_pt_weight_name
                                         load_tf2_model_in_pytorch_model)
 
 # Pipelines
-# from .pipeline_ import TextClassificationPipeline
-from .pipelines import Pipeline, pipeline, TextClassificationPipeline
+from .pipelines import pipeline, PipelineDataFormat, CsvPipelineDataFormat, JsonPipelineDataFormat, PipedPipelineDataFormat, \
+    Pipeline, FeatureExtractionPipeline, QuestionAnsweringPipeline, NerPipeline, TextClassificationPipeline
 
 if not is_tf_available() and not is_torch_available():
     logger.warning("Neither PyTorch nor TensorFlow >= 2.0 have been found."
diff --git a/transformers/commands/run.py b/transformers/commands/run.py
index 44c1127803..7c00c0057f 100644
--- a/transformers/commands/run.py
+++ b/transformers/commands/run.py
@@ -36,11 +36,10 @@ class RunCommand(BaseTransformersCLICommand):
         run_parser.add_argument('--model', type=str, required=True, help='Name or path to the model to instantiate.')
         run_parser.add_argument('--config', type=str, help='Name or path to the model\'s config to instantiate.')
         run_parser.add_argument('--tokenizer', type=str, help='Name of the tokenizer to use. (default: same as the model name)')
-        run_parser.add_argument('--column', type=str, required=True, help='Name of the column to use as input. (For multi columns input as QA use column1,columns2)')
+        run_parser.add_argument('--column', type=str, help='Name of the column to use as input. (For multi columns input as QA use column1,columns2)')
         run_parser.add_argument('--format', type=str, default='infer', choices=PipelineDataFormat.SUPPORTED_FORMATS, help='Input format to read from')
-        run_parser.add_argument('--input', type=str, required=True, help='Path to the file to use for inference')
-        run_parser.add_argument('--output', type=str, required=True, help='Path to the file that will be used post to write results.')
-        run_parser.add_argument('kwargs', nargs='*', help='Arguments to forward to the file format reader')
+        run_parser.add_argument('--input', type=str, help='Path to the file to use for inference')
+        run_parser.add_argument('--output', type=str, help='Path to the file that will be used post to write results.')
         run_parser.set_defaults(func=run_command_factory)
 
     def run(self):
diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 809096a30c..9e7051b70a 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -66,20 +66,6 @@ class DefaultArgumentHandler(ArgumentHandler):
         raise ValueError('Unable to infer the format of the provided data (X=, data=, ...)')
 
 
-class _ScikitCompat(ABC):
-    """
-    Interface layer for the Scikit and Keras compatibility.
-    """
-
-    @abstractmethod
-    def transform(self, X):
-        raise NotImplementedError()
-
-    @abstractmethod
-    def predict(self, X):
-        raise NotImplementedError()
-
-
 class PipelineDataFormat:
     """
     Base class for all the pipeline supported data format both for reading and writing.
@@ -90,12 +76,12 @@ class PipelineDataFormat:
     PipelineDataFormat also includes some utilities to work with multi-columns like mapping from datasets columns
     to pipelines keyword arguments through the `dataset_kwarg_1=dataset_column_1` format.
     """
-    SUPPORTED_FORMATS = ['json', 'csv']
+    SUPPORTED_FORMATS = ['json', 'csv', 'pipe']
 
-    def __init__(self, output: str, path: str, column: str):
+    def __init__(self, output: Optional[str], path: Optional[str], column: Optional[str]):
         self.output = output
         self.path = path
-        self.column = column.split(',')
+        self.column = column.split(',') if column else ['']
         self.is_multi_columns = len(self.column) > 1
 
         if self.is_multi_columns:
@@ -117,17 +103,19 @@ class PipelineDataFormat:
         raise NotImplementedError()
 
     @staticmethod
-    def from_str(name: str, output: str, path: str, column: str):
+    def from_str(name: str, output: Optional[str], path: Optional[str], column: Optional[str]):
         if name == 'json':
             return JsonPipelineDataFormat(output, path, column)
         elif name == 'csv':
             return CsvPipelineDataFormat(output, path, column)
+        elif name == 'pipe':
+            return PipedPipelineDataFormat(output, path, column)
         else:
-            raise KeyError('Unknown reader {} (Available reader are json/csv)'.format(name))
+            raise KeyError('Unknown reader {} (Available reader are json/csv/pipe)'.format(name))
 
 
 class CsvPipelineDataFormat(PipelineDataFormat):
-    def __init__(self, output: str, path: str, column: str):
+    def __init__(self, output: Optional[str], path: Optional[str], column: Optional[str]):
         super().__init__(output, path, column)
 
     def __iter__(self):
@@ -148,7 +136,7 @@ class CsvPipelineDataFormat(PipelineDataFormat):
 
 
 class JsonPipelineDataFormat(PipelineDataFormat):
-    def __init__(self, output: str, path: str, column: str):
+    def __init__(self, output: Optional[str], path: Optional[str], column: Optional[str]):
         super().__init__(output, path, column)
 
         with open(path, 'r') as f:
@@ -166,6 +154,50 @@ class JsonPipelineDataFormat(PipelineDataFormat):
             json.dump(data, f)
 
 
+class PipedPipelineDataFormat(PipelineDataFormat):
+    """
+    Read data from piped input to the python process.
+    For multi columns data, columns should separated by \t
+
+    If columns are provided, then the output will be a dictionary with {column_x: value_x}
+    """
+    def __iter__(self):
+        import sys
+        for line in sys.stdin:
+
+            # Split for multi-columns
+            if '\t' in line:
+
+                line = line.split('\t')
+                if self.column:
+                    # Dictionary to map arguments
+                    yield {kwargs: l for (kwargs, _), l in zip(self.column, line)}
+                else:
+                    yield tuple(line)
+
+            # No dictionary to map arguments
+            else:
+                print(line)
+                yield line
+
+    def save(self, data: dict):
+        print(data)
+
+
+class _ScikitCompat(ABC):
+    """
+    Interface layer for the Scikit and Keras compatibility.
+    """
+
+    @abstractmethod
+    def transform(self, X):
+        raise NotImplementedError()
+
+    @abstractmethod
+    def predict(self, X):
+        raise NotImplementedError()
+
+
 class Pipeline(_ScikitCompat):
     """
     Base class implementing pipelined operations.
@@ -208,18 +240,6 @@ class Pipeline(_ScikitCompat):
         """
         return self(X=X)
 
-    def __call__(self, *texts, **kwargs):
-        # Parse arguments
-        inputs = self._args_parser(*texts, **kwargs)
-
-        # Encode for forward
-        with self.device_placement():
-            inputs = self.tokenizer.batch_encode_plus(
-                inputs, add_special_tokens=True, return_tensors='tf' if is_tf_available() else 'pt'
-            )
-
-            return self._forward(inputs)
-
     @contextmanager
     def device_placement(self):
         """
@@ -244,6 +264,18 @@ class Pipeline(_ScikitCompat):
 
             yield
 
+    def __call__(self, *texts, **kwargs):
+        # Parse arguments
+        inputs = self._args_parser(*texts, **kwargs)
+
+        # Encode for forward
+        with self.device_placement():
+            inputs = self.tokenizer.batch_encode_plus(
+                inputs, add_special_tokens=True, return_tensors='tf' if is_tf_available() else 'pt'
+            )
+
+            return self._forward(inputs)
+
     def _forward(self, inputs):
         """
         Internal framework specific forward dispatching.
@@ -275,12 +307,6 @@ class TextClassificationPipeline(Pipeline):
     """
     Text classification pipeline using ModelForTextClassification head.
     """
-    def __init__(self, model, tokenizer: PreTrainedTokenizer, nb_classes: int = 2):
-        super().__init__(model, tokenizer)
-
-        if nb_classes < 2:
-            raise Exception('Invalid parameter nb_classes. int >= 2 is required (got: {})'.format(nb_classes))
-        self._nb_classes = nb_classes
 
     def __call__(self, *args, **kwargs):
         outputs = super().__call__(*args, **kwargs)
@@ -398,56 +424,6 @@ class QuestionAnsweringPipeline(Pipeline):
     Question Answering pipeline using ModelForQuestionAnswering head.
     """
 
-    class QuestionAnsweringArgumentHandler(ArgumentHandler):
-
-        def __call__(self, *args, **kwargs):
-            # Position args, handling is sensibly the same as X and data, so forwarding to avoid duplicating
-            if args is not None and len(args) > 0:
-                if len(args) == 1:
-                    kwargs['X'] = args[0]
-                else:
-                    kwargs['X'] = list(args)
-
-            # Generic compatibility with sklearn and Keras
-            # Batched data
-            if 'X' in kwargs or 'data' in kwargs:
-                data = kwargs['X'] if 'X' in kwargs else kwargs['data']
-
-                if not isinstance(data, list):
-                    data = [data]
-
-                for i, item in enumerate(data):
-                    if isinstance(item, dict):
-                        if any(k not in item for k in ['question', 'context']):
-                            raise KeyError('You need to provide a dictionary with keys {question:..., context:...}')
-                        data[i] = QuestionAnsweringPipeline.create_sample(**item)
-
-                    elif isinstance(item, SquadExample):
-                        continue
-                    else:
-                        raise ValueError(
-                            '{} argument needs to be of type (list[SquadExample | dict], SquadExample, dict)'
-                                .format('X' if 'X' in kwargs else 'data')
-                        )
-                inputs = data
-
-            # Tabular input
-            elif 'question' in kwargs and 'context' in kwargs:
-                if isinstance(kwargs['question'], str):
-                    kwargs['question'] = [kwargs['question']]
-
-                if isinstance(kwargs['context'], str):
-                    kwargs['context'] = [kwargs['context']]
-
-                inputs = [QuestionAnsweringPipeline.create_sample(q, c) for q, c in zip(kwargs['question'], kwargs['context'])]
-            else:
-                raise ValueError('Unknown arguments {}'.format(kwargs))
-
-            if not isinstance(inputs, list):
-                inputs = [inputs]
-
-            return inputs
-
     @staticmethod
     def create_sample(question: Union[str, List[str]], context: Union[str, List[str]]) -> Union[SquadExample, List[SquadExample]]:
         """

From db90e1211433ef99f952e60fbe5ea578391b86b1 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Wed, 18 Dec 2019 23:46:33 +0100
Subject: [PATCH 431/505] configuration: use S3 location for XLM-RoBERTa model

---
 transformers/configuration_xlm_roberta.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/transformers/configuration_xlm_roberta.py b/transformers/configuration_xlm_roberta.py
index dd03572976..d7a26538c5 100644
--- a/transformers/configuration_xlm_roberta.py
+++ b/transformers/configuration_xlm_roberta.py
@@ -25,7 +25,8 @@ from .configuration_roberta import RobertaConfig
 logger = logging.getLogger(__name__)
 
 XLM_ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP = {
-    'xlm-roberta-large': "https://schweter.eu/cloud/transformers/xlm-roberta-large-config.json",
+    'xlm-roberta-base': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-base-config.json",
+    'xlm-roberta-large': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-large-config.json",
 }
 
 

From 5c5f67a256b558b470f03cf36edc5ea35dec2fba Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Wed, 18 Dec 2019 23:47:00 +0100
Subject: [PATCH 432/505] modeling: use S3 location for XLM-RoBERTa model

---
 transformers/modeling_xlm_roberta.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/transformers/modeling_xlm_roberta.py b/transformers/modeling_xlm_roberta.py
index abace25d5b..8095c46a16 100644
--- a/transformers/modeling_xlm_roberta.py
+++ b/transformers/modeling_xlm_roberta.py
@@ -27,8 +27,8 @@ from .file_utils import add_start_docstrings
 logger = logging.getLogger(__name__)
 
 XLM_ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP = {
-    'xlm-roberta-base': "https://schweter.eu/cloud/transformers/xlm-roberta-base-pytorch_model.bin",
-    'xlm-roberta-large': "https://schweter.eu/cloud/transformers/xlm-roberta-large-pytorch_model.bin",
+    'xlm-roberta-base': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-base-pytorch_model.bin",
+    'xlm-roberta-large': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-large-pytorch_model.bin",
 }
 
 

From fe9aab1055604e772be05a1cbbb36a207c177055 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Wed, 18 Dec 2019 23:47:48 +0100
Subject: [PATCH 433/505] tokenization: use S3 location for XLM-RoBERTa model

---
 transformers/tokenization_xlm_roberta.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/transformers/tokenization_xlm_roberta.py b/transformers/tokenization_xlm_roberta.py
index 93b1c397e4..453c4375c6 100644
--- a/transformers/tokenization_xlm_roberta.py
+++ b/transformers/tokenization_xlm_roberta.py
@@ -30,8 +30,8 @@ VOCAB_FILES_NAMES = {'vocab_file': 'sentencepiece.bpe.model'}
 PRETRAINED_VOCAB_FILES_MAP = {
     'vocab_file':
     {
-    'xlm-roberta-base': "https://schweter.eu/cloud/transformers/xlm-roberta-base-sentencepiece.bpe.model",
-    'xlm-roberta-large': "https://schweter.eu/cloud/transformers/xlm-roberta-large-sentencepiece.bpe.model",
+    'xlm-roberta-base': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-base-sentencepiece.bpe.model",
+    'xlm-roberta-large': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-large-sentencepiece.bpe.model",
     }
 }
 

From ec5d6c6a70e0bcdf31e737206caa1e56e859b2d3 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Thu, 19 Dec 2019 00:12:10 +0100
Subject: [PATCH 434/505] Adressing issue with NER task omitting first and last
 word.

---
 transformers/pipelines.py | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 9e7051b70a..1d8f226b13 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -318,8 +318,6 @@ class NerPipeline(Pipeline):
     """
     Named Entity Recognition pipeline using ModelForTokenClassification head.
     """
-    def __init__(self, model, tokenizer: PreTrainedTokenizer):
-        super().__init__(model, tokenizer)
 
     def __call__(self, *texts, **kwargs):
         inputs, answers = self._args_parser(*texts, **kwargs), []
@@ -344,14 +342,16 @@ class NerPipeline(Pipeline):
 
             # Normalize scores
             answer, token_start = [], 1
-            for idx, word in groupby(token_to_word[1:-1]):
+            for idx, word in groupby(token_to_word):
 
                 # Sum log prob over token, then normalize across labels
                 score = np.exp(entities[token_start]) / np.exp(entities[token_start]).sum(-1, keepdims=True)
                 label_idx = score.argmax()
 
                 answer += [{
-                    'word': words[idx - 1], 'score': score[label_idx].item(), 'entity': self.model.config.id2label[label_idx]
+                    'word': words[idx],
+                    'score': score[label_idx].item(),
+                    'entity': self.model.config.id2label[label_idx]
                 }]
 
                 # Update token start

From a26ce4dee116a1d5d9099c8a94e22d1e31ad631c Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Thu, 19 Dec 2019 02:23:01 +0100
Subject: [PATCH 435/505] examples: add XLM-RoBERTa to glue script

---
 examples/run_glue.py | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/examples/run_glue.py b/examples/run_glue.py
index 1a51255c11..954a8fbf0c 100644
--- a/examples/run_glue.py
+++ b/examples/run_glue.py
@@ -52,6 +52,9 @@ from transformers import (WEIGHTS_NAME, BertConfig,
                                   AlbertConfig,
                                   AlbertForSequenceClassification, 
                                   AlbertTokenizer,
+                                  XLMRobertaConfig,
+                                  XLMRobertaForSequenceClassification,
+                                  XLMRobertaTokenizer,
                                 )
 
 from transformers import AdamW, get_linear_schedule_with_warmup
@@ -72,7 +75,8 @@ MODEL_CLASSES = {
     'xlm': (XLMConfig, XLMForSequenceClassification, XLMTokenizer),
     'roberta': (RobertaConfig, RobertaForSequenceClassification, RobertaTokenizer),
     'distilbert': (DistilBertConfig, DistilBertForSequenceClassification, DistilBertTokenizer),
-    'albert': (AlbertConfig, AlbertForSequenceClassification, AlbertTokenizer)
+    'albert': (AlbertConfig, AlbertForSequenceClassification, AlbertTokenizer),
+    'xlmroberta': (XLMRobertaConfig, XLMRobertaForSequenceClassification, XLMRobertaTokenizer),
 }
 
 
@@ -304,9 +308,9 @@ def load_and_cache_examples(args, task, tokenizer, evaluate=False):
     else:
         logger.info("Creating features from dataset file at %s", args.data_dir)
         label_list = processor.get_labels()
-        if task in ['mnli', 'mnli-mm'] and args.model_type in ['roberta']:
+        if task in ['mnli', 'mnli-mm'] and args.model_type in ['roberta', 'xlmroberta']:
             # HACK(label indices are swapped in RoBERTa pretrained model)
-            label_list[1], label_list[2] = label_list[2], label_list[1] 
+            label_list[1], label_list[2] = label_list[2], label_list[1]
         examples = processor.get_dev_examples(args.data_dir) if evaluate else processor.get_train_examples(args.data_dir)
         features = convert_examples_to_features(examples,
                                                 tokenizer,

From bcc99fd92efe03ff332e5b26a342a09e9d709cf7 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Thu, 19 Dec 2019 10:32:21 +0100
Subject: [PATCH 436/505] Fix wrong automatic config allocation through
 AutoConfig

---
 transformers/pipelines.py | 117 +++++++++++++++++++++++++-------------
 1 file changed, 78 insertions(+), 39 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 1d8f226b13..ee3aed2c65 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -25,7 +25,7 @@ from typing import Union, Optional, Tuple, List, Dict
 
 import numpy as np
 
-from transformers import AutoTokenizer, PreTrainedTokenizer, PretrainedConfig, \
+from transformers import AutoConfig, AutoTokenizer, PreTrainedTokenizer, PretrainedConfig, \
     SquadExample, squad_convert_examples_to_features, is_tf_available, is_torch_available, logger
 
 if is_tf_available():
@@ -264,6 +264,27 @@ class Pipeline(_ScikitCompat):
 
             yield
 
+    def inputs_for_model(self, features: Union[dict, List[dict]]) -> Dict:
+        """
+        Generates the input dictionary with model-specific parameters.
+
+        Returns:
+            dict holding all the required parameters for model's forward
+        """
+        args = ['input_ids', 'attention_mask']
+        model_type = type(self.model).__name__.lower()
+
+        if 'distilbert' not in model_type and 'xlm' not in model_type:
+            args += ['token_type_ids']
+
+        if 'xlnet' in model_type or 'xlm' in model_type:
+            args += ['cls_index', 'p_mask']
+
+        if isinstance(features, dict):
+            return {k: features[k] for k in args}
+        else:
+            return {k: [feature[k] for feature in features] for k in args}
+
     def __call__(self, *texts, **kwargs):
         # Parse arguments
         inputs = self._args_parser(*texts, **kwargs)
@@ -271,9 +292,14 @@ class Pipeline(_ScikitCompat):
         # Encode for forward
         with self.device_placement():
             inputs = self.tokenizer.batch_encode_plus(
-                inputs, add_special_tokens=True, return_tensors='tf' if is_tf_available() else 'pt'
+                inputs, add_special_tokens=True,
+                return_tensors='tf' if is_tf_available() else 'pt',
+                # max_length=self.model.config.max_position_embedding
+                max_length=511
             )
 
+            # Filter out features not available on specific models
+            inputs = self.inputs_for_model(inputs)
             return self._forward(inputs)
 
     def _forward(self, inputs):
@@ -331,7 +357,11 @@ class NerPipeline(Pipeline):
 
             # Manage correct placement of the tensors
             with self.device_placement():
-                tokens = self.tokenizer.encode_plus(sentence, return_attention_mask=False, return_tensors='tf' if is_tf_available() else 'pt')
+                tokens = self.tokenizer.encode_plus(
+                    sentence, return_attention_mask=False,
+                    return_tensors='tf' if is_tf_available() else 'pt',
+                    max_length=512
+                )
 
                 # Forward
                 if is_torch_available():
@@ -443,27 +473,6 @@ class QuestionAnsweringPipeline(Pipeline):
         super().__init__(model, tokenizer, args_parser=QuestionAnsweringArgumentHandler(),
                          device=device, **kwargs)
 
-    def inputs_for_model(self, features: Union[SquadExample, List[SquadExample]]) -> Dict:
-        """
-        Generates the input dictionary with model-specific parameters.
-
-        Returns:
-            dict holding all the required parameters for model's forward
-        """
-        args = ['input_ids', 'attention_mask']
-        model_type = type(self.model).__name__.lower()
-
-        if 'distilbert' not in model_type and 'xlm' not in model_type:
-            args += ['token_type_ids']
-
-        if 'xlnet' in model_type or 'xlm' in model_type:
-            args += ['cls_index', 'p_mask']
-
-        if isinstance(features, SquadExample):
-            return {k: features.__dict__[k] for k in args}
-        else:
-            return {k: [feature.__dict__[k] for feature in features] for k in args}
-
     def __call__(self, *texts, **kwargs):
         """
         Args:
@@ -495,7 +504,7 @@ class QuestionAnsweringPipeline(Pipeline):
         # Convert inputs to features
         examples = self._args_parser(*texts, **kwargs)
         features = squad_convert_examples_to_features(examples, self.tokenizer, kwargs['max_seq_len'], kwargs['doc_stride'], kwargs['max_question_len'], False)
-        fw_args = self.inputs_for_model(features)
+        fw_args = self.inputs_for_model(features.__dict__)
 
         # Manage tensor allocation on correct device
         with self.device_placement():
@@ -627,29 +636,50 @@ class QuestionAnsweringPipeline(Pipeline):
 # Register all the supported task here
 SUPPORTED_TASKS = {
     'feature-extraction': {
-      'impl': FeatureExtractionPipeline,
-      'tf': TFAutoModel if is_tf_available() else None,
-      'pt': AutoModel if is_torch_available() else None,
+        'impl': FeatureExtractionPipeline,
+        'tf': TFAutoModel if is_tf_available() else None,
+        'pt': AutoModel if is_torch_available() else None,
+        'default': {
+            'model': 'distilbert-base-uncased',
+            'config': None,
+            'tokenizer': 'bert-base-uncased'
+        }
     },
-    'text-classification': {
+    'sentiment-analysis': {
         'impl': TextClassificationPipeline,
         'tf': TFAutoModelForSequenceClassification if is_tf_available() else None,
-        'pt': AutoModelForSequenceClassification if is_torch_available() else None
+        'pt': AutoModelForSequenceClassification if is_torch_available() else None,
+        'default': {
+            'model': 'https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-finetuned-sst-2-english-pytorch_model.bin',
+            'config': 'https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-finetuned-sst-2-english-config.json',
+            'tokenizer': 'bert-base-uncased'
+        }
     },
     'ner': {
-      'impl': NerPipeline,
-      'tf': TFAutoModelForTokenClassification if is_tf_available() else None,
-      'pt': AutoModelForTokenClassification if is_torch_available() else None,
+        'impl': NerPipeline,
+        'tf': TFAutoModelForTokenClassification if is_tf_available() else None,
+        'pt': AutoModelForTokenClassification if is_torch_available() else None,
+        'default': {
+            'model': 'https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-cased-finetuned-conll03-english-pytorch_model.bin',
+            'config': 'https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-cased-finetuned-conll03-english-config.json',
+            'tokenizer': 'bert-base-cased'
+        }
     },
     'question-answering': {
         'impl': QuestionAnsweringPipeline,
         'tf': TFAutoModelForQuestionAnswering if is_tf_available() else None,
-        'pt': AutoModelForQuestionAnswering if is_torch_available() else None
+        'pt': AutoModelForQuestionAnswering if is_torch_available() else None,
+        'default': {
+            'model': 'distilbert-base-uncased-distilled-squad',
+            'config': None,
+            'tokenizer': 'bert-base-uncased'
+        }
     }
 }
 
 
-def pipeline(task: str, model, config: Optional[Union[str, PretrainedConfig]] = None,
+def pipeline(task: str, model: Optional = None,
+             config: Optional[Union[str, PretrainedConfig]] = None,
              tokenizer: Optional[Union[str, PreTrainedTokenizer]] = None, **kwargs) -> Pipeline:
     """
     Utility factory method to build a pipeline.
@@ -657,23 +687,32 @@ def pipeline(task: str, model, config: Optional[Union[str, PretrainedConfig]] =
         A Tokenizer instance in charge of mapping raw textual input to token
         A Model instance
         Some (optional) post processing for enhancing model's output
+
+    Examples:
+        pipeline('ner')
     """
     # Try to infer tokenizer from model name (if provided as str)
     if tokenizer is None:
-        if not isinstance(model, str):
+        if model is not None and not isinstance(model, str):
             # Impossible to guest what is the right tokenizer here
             raise Exception('Tokenizer cannot be None if provided model is a PreTrainedModel instance')
         else:
             tokenizer = model
 
-    tokenizer = tokenizer if isinstance(tokenizer, PreTrainedTokenizer) else AutoTokenizer.from_pretrained(tokenizer)
-
+    # Retrieve the task
     if task not in SUPPORTED_TASKS:
         raise KeyError("Unknown task {}, available tasks are {}".format(task, list(SUPPORTED_TASKS.keys())))
 
     targeted_task = SUPPORTED_TASKS[task]
     task, allocator = targeted_task['impl'], targeted_task['tf'] if is_tf_available() else targeted_task['pt']
 
+    # Handling for default model for the task
+    if model is None:
+        model, config, tokenizer = tuple(targeted_task['default'].values())
+
+    # Allocate tokenizer
+    tokenizer = tokenizer if isinstance(tokenizer, PreTrainedTokenizer) else AutoTokenizer.from_pretrained(tokenizer)
+
     # Special handling for model conversion
     if isinstance(model, str):
         from_tf = model.endswith('.h5') and not is_tf_available()
@@ -689,7 +728,7 @@ def pipeline(task: str, model, config: Optional[Union[str, PretrainedConfig]] =
         from_tf = from_pt = False
 
     if isinstance(config, str):
-        config = PretrainedConfig.from_pretrained(config)
+        config = AutoConfig.from_pretrained(config)
 
     if allocator.__name__.startswith('TF'):
         model = allocator.from_pretrained(model, config=config, from_pt=from_pt)

From d72fa2a0f67f9c55877826f767f01007507fd8e3 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Thu, 19 Dec 2019 10:54:10 +0100
Subject: [PATCH 437/505] Fix inputs_for_model call in
 QuestionAnsweringPipeline accessing __dict__ on list.

---
 transformers/pipelines.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index ee3aed2c65..a220aa7e71 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -504,7 +504,7 @@ class QuestionAnsweringPipeline(Pipeline):
         # Convert inputs to features
         examples = self._args_parser(*texts, **kwargs)
         features = squad_convert_examples_to_features(examples, self.tokenizer, kwargs['max_seq_len'], kwargs['doc_stride'], kwargs['max_question_len'], False)
-        fw_args = self.inputs_for_model(features.__dict__)
+        fw_args = self.inputs_for_model([f.__dict__ for f in features])
 
         # Manage tensor allocation on correct device
         with self.device_placement():

From f516cf39564bd33bd26e6a51d6cf9f589e600078 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Thu, 19 Dec 2019 11:42:33 +0100
Subject: [PATCH 438/505] Allow pipeline to write output in binary format

---
 transformers/commands/run.py | 10 +++++++++-
 transformers/pipelines.py    | 32 ++++++++++++++++++++++++++++++--
 2 files changed, 39 insertions(+), 3 deletions(-)

diff --git a/transformers/commands/run.py b/transformers/commands/run.py
index 7c00c0057f..78109b2a16 100644
--- a/transformers/commands/run.py
+++ b/transformers/commands/run.py
@@ -1,9 +1,13 @@
+import logging
 from argparse import ArgumentParser
 
 from transformers.commands import BaseTransformersCLICommand
 from transformers.pipelines import pipeline, Pipeline, PipelineDataFormat, SUPPORTED_TASKS
 
 
+logger = logging.getLogger(__name__)  # pylint: disable=invalid-name
+
+
 def try_infer_format_from_ext(path: str):
     for ext in PipelineDataFormat.SUPPORTED_FORMATS:
         if path.endswith(ext):
@@ -51,7 +55,11 @@ class RunCommand(BaseTransformersCLICommand):
                 output += [nlp(entry)]
 
         # Saving data
-        self._reader.save(output)
+        if self._nlp.binary_output:
+            binary_path = self._reader.save_binary(output)
+            logger.warning('Current pipeline requires output to be in binary format, saving at {}'.format(binary_path))
+        else:
+            self._reader.save(output)
 
 
 
diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index a220aa7e71..b766549121 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -17,6 +17,7 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 import csv
 import json
 import os
+import pickle
 from abc import ABC, abstractmethod
 from contextlib import contextmanager
 from itertools import groupby
@@ -91,6 +92,7 @@ class PipelineDataFormat:
             if exists(abspath(self.output)):
                 raise OSError('{} already exists on disk'.format(self.output))
 
+        if path is not None:
             if not exists(abspath(self.path)):
                 raise OSError('{} doesnt exist on disk'.format(self.path))
 
@@ -102,6 +104,15 @@ class PipelineDataFormat:
     def save(self, data: dict):
         raise NotImplementedError()
 
+    def save_binary(self, data: Union[dict, List[dict]]) -> str:
+        path, _ = os.path.splitext(self.output)
+        binary_path = os.path.extsep.join((path, 'pickle'))
+
+        with open(binary_path, 'wb+') as f_output:
+            pickle.dump(data, f_output)
+
+        return binary_path
+
     @staticmethod
     def from_str(name: str, output: Optional[str], path: Optional[str], column: Optional[str]):
         if name == 'json':
@@ -177,12 +188,20 @@ class PipedPipelineDataFormat(PipelineDataFormat):
 
             # No dictionary to map arguments
             else:
-                print(line)
                 yield line
 
     def save(self, data: dict):
         print(data)
 
+    def save_binary(self, data: Union[dict, List[dict]]) -> str:
+        if self.output is None:
+            raise KeyError(
+                'When using piped input on pipeline outputting large object requires an output file path. '
+                'Please provide such output path through --output argument.'
+            )
+
+        return super().save_binary(data)
+
 
 class _ScikitCompat(ABC):
     """
@@ -205,11 +224,13 @@ class Pipeline(_ScikitCompat):
         Input -> Tokenization -> Model Inference -> Post-Processing (Task dependent) -> Output
     """
     def __init__(self, model, tokenizer: PreTrainedTokenizer = None,
-                 args_parser: ArgumentHandler = None, device: int = -1, **kwargs):
+                 args_parser: ArgumentHandler = None, device: int = -1,
+                 binary_output: bool = False):
 
         self.model = model
         self.tokenizer = tokenizer
         self.device = device
+        self.binary_output = binary_output
         self._args_parser = args_parser or DefaultArgumentHandler()
 
         # Special handling
@@ -325,6 +346,13 @@ class FeatureExtractionPipeline(Pipeline):
     """
     Feature extraction pipeline using Model head.
     """
+
+    def __init__(self, model,
+                 tokenizer: PreTrainedTokenizer = None,
+                 args_parser: ArgumentHandler = None,
+                 device: int = -1):
+        super().__init__(model, tokenizer, args_parser, device, binary_output=True)
+
     def __call__(self, *args, **kwargs):
         return super().__call__(*args, **kwargs).tolist()
 

From fc624716aad05efa78e4d65f214fb978ea0ac9e7 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Thu, 19 Dec 2019 11:49:06 +0100
Subject: [PATCH 439/505] Renaming framework env variables flags from NO_ to
 USE_

---
 transformers/file_utils.py | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/transformers/file_utils.py b/transformers/file_utils.py
index 4784681fb4..c586c57a76 100644
--- a/transformers/file_utils.py
+++ b/transformers/file_utils.py
@@ -27,8 +27,9 @@ from contextlib import contextmanager
 logger = logging.getLogger(__name__)  # pylint: disable=invalid-name
 
 try:
-    if 'NO_TF' in os.environ and os.environ['NO_TF'].upper() in ('1', 'ON'):
-        logger.info("Found NO_TF, disabling TensorFlow")
+    os.environ.setdefault('USE_TF', 'YES')
+    if os.environ['USE_TF'].upper() in ('1', 'ON', 'YES'):
+        logger.info("USE_TF override through env variable, disabling Tensorflow")
         _tf_available = False
     else:
         import tensorflow as tf
@@ -39,8 +40,9 @@ except (ImportError, AssertionError):
     _tf_available = False  # pylint: disable=invalid-name
 
 try:
-    if 'NO_TORCH' in os.environ and os.environ['NO_TORCH'].upper() in ('1', 'ON'):
-        logger.info("Found NO_TORCH, disabling PyTorch")
+    os.environ.setdefault('USE_TORCH', 'YES')
+    if os.environ['USE_TORCH'].upper() in ('1', 'ON', 'YES'):
+        logger.info("USE_TORCH override through env variable, disabling PyTorch")
         _torch_available = False
     else:
         import torch

From 3b29322d4c197fec46ae48f62aa2870d00d0852a Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Thu, 19 Dec 2019 12:24:17 +0100
Subject: [PATCH 440/505] Expose all the pipeline argument on serve command.

---
 transformers/commands/serving.py | 38 +++++++++++++-------------------
 1 file changed, 15 insertions(+), 23 deletions(-)

diff --git a/transformers/commands/serving.py b/transformers/commands/serving.py
index a35dff0ebe..a7321470ce 100644
--- a/transformers/commands/serving.py
+++ b/transformers/commands/serving.py
@@ -17,25 +17,18 @@ def serve_command_factory(args: Namespace):
     Factory function used to instantiate serving server from provided command line arguments.
     :return: ServeCommand
     """
-    nlp = pipeline(args.task, args.model)
-    return ServeCommand(nlp, args.host, args.port, args.model, args.graphql)
+    nlp = pipeline(task=args.task, model=args.model, config=args.config, tokenizer=args.tokenizer, device=args.device)
+    return ServeCommand(nlp, args.host, args.port)
 
 
-class ServeResult(BaseModel):
-    """
-    Base class for serving result
-    """
-    model: str
-
-
-class ServeModelInfoResult(ServeResult):
+class ServeModelInfoResult(BaseModel):
     """
     Expose model information
     """
     infos: dict
 
 
-class ServeTokenizeResult(ServeResult):
+class ServeTokenizeResult(BaseModel):
     """
     Tokenize result model
     """
@@ -43,14 +36,14 @@ class ServeTokenizeResult(ServeResult):
     tokens_ids: Optional[List[int]]
 
 
-class ServeDeTokenizeResult(ServeResult):
+class ServeDeTokenizeResult(BaseModel):
     """
     DeTokenize result model
     """
     text: str
 
 
-class ServeForwardResult(ServeResult):
+class ServeForwardResult(BaseModel):
     """
     Forward result model
     """
@@ -71,11 +64,12 @@ class ServeCommand(BaseTransformersCLICommand):
         serve_parser.add_argument('--device', type=int, default=-1, help='Indicate the device to run onto, -1 indicates CPU, >= 0 indicates GPU (default: -1)')
         serve_parser.add_argument('--host', type=str, default='localhost', help='Interface the server will listen on.')
         serve_parser.add_argument('--port', type=int, default=8888, help='Port the serving will listen to.')
-        serve_parser.add_argument('--model', type=str, required=True, help='Model\'s name or path to stored model to infer from.')
-        serve_parser.add_argument('--graphql', action='store_true', default=False, help='Enable GraphQL endpoints.')
+        serve_parser.add_argument('--model', type=str, required=True, help='Model\'s name or path to stored model.')
+        serve_parser.add_argument('--config', type=str, help='Model\'s config name or path to stored model.')
+        serve_parser.add_argument('--tokenizer', type=str, help='Tokenizer name to use.')
         serve_parser.set_defaults(func=serve_command_factory)
 
-    def __init__(self, pipeline: Pipeline, host: str, port: int, model: str, graphql: bool):
+    def __init__(self, pipeline: Pipeline, host: str, port: int):
         self._logger = getLogger('transformers-cli/serving')
 
         self._pipeline = pipeline
@@ -95,7 +89,7 @@ class ServeCommand(BaseTransformersCLICommand):
         run(self._app, host=self._host, port=self._port)
 
     def model_info(self):
-        return ServeModelInfoResult(model='', infos=vars(self._pipeline.model.config))
+        return ServeModelInfoResult(infos=vars(self._pipeline.model.config))
 
     def tokenize(self, text_input: str = Body(None, embed=True), return_ids: bool = Body(False, embed=True)):
         """
@@ -108,9 +102,9 @@ class ServeCommand(BaseTransformersCLICommand):
 
             if return_ids:
                 tokens_ids = self._pipeline.tokenizer.convert_tokens_to_ids(tokens_txt)
-                return ServeTokenizeResult(model='', tokens=tokens_txt, tokens_ids=tokens_ids)
+                return ServeTokenizeResult(tokens=tokens_txt, tokens_ids=tokens_ids)
             else:
-                return ServeTokenizeResult(model='', tokens=tokens_txt)
+                return ServeTokenizeResult(tokens=tokens_txt)
 
         except Exception as e:
             raise HTTPException(status_code=500, detail={"model": '', "error": str(e)})
@@ -139,13 +133,11 @@ class ServeCommand(BaseTransformersCLICommand):
 
         # Check we don't have empty string
         if len(inputs) == 0:
-            return ServeForwardResult(model='', output=[], attention=[])
+            return ServeForwardResult(output=[], attention=[])
 
         try:
             # Forward through the model
             output = self._pipeline(inputs)
-            return ServeForwardResult(
-                model='', output=output
-            )
+            return ServeForwardResult(output=output)
         except Exception as e:
             raise HTTPException(500, {"error": str(e)})

From 5664327c24730b30b52d84e71e9af7c1d32b0fe1 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Thu, 19 Dec 2019 12:27:54 +0100
Subject: [PATCH 441/505] Hide train command for now.

---
 transformers-cli | 2 --
 1 file changed, 2 deletions(-)

diff --git a/transformers-cli b/transformers-cli
index 39b7f5816b..db2bd0e2a3 100755
--- a/transformers-cli
+++ b/transformers-cli
@@ -5,7 +5,6 @@ from transformers.commands.download import DownloadCommand
 from transformers.commands.run import RunCommand
 from transformers.commands.serving import ServeCommand
 from transformers.commands.user import UserCommands
-from transformers.commands.train import TrainCommand
 from transformers.commands.convert import ConvertCommand
 
 if __name__ == '__main__':
@@ -17,7 +16,6 @@ if __name__ == '__main__':
     DownloadCommand.register_subcommand(commands_parser)
     RunCommand.register_subcommand(commands_parser)
     ServeCommand.register_subcommand(commands_parser)
-    TrainCommand.register_subcommand(commands_parser)
     UserCommands.register_subcommand(commands_parser)
 
     # Let's go

From faef6f6191a8d319b541396a0f850c3d6f15f5d4 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Thu, 19 Dec 2019 12:28:17 +0100
Subject: [PATCH 442/505] Fix logic order for USE_TF/USE_TORCH

---
 transformers/file_utils.py | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)

diff --git a/transformers/file_utils.py b/transformers/file_utils.py
index c586c57a76..92cada85b3 100644
--- a/transformers/file_utils.py
+++ b/transformers/file_utils.py
@@ -29,25 +29,27 @@ logger = logging.getLogger(__name__)  # pylint: disable=invalid-name
 try:
     os.environ.setdefault('USE_TF', 'YES')
     if os.environ['USE_TF'].upper() in ('1', 'ON', 'YES'):
-        logger.info("USE_TF override through env variable, disabling Tensorflow")
-        _tf_available = False
-    else:
         import tensorflow as tf
         assert hasattr(tf, '__version__') and int(tf.__version__[0]) >= 2
         _tf_available = True  # pylint: disable=invalid-name
         logger.info("TensorFlow version {} available.".format(tf.__version__))
+    else:
+        logger.info("USE_TF override through env variable, disabling Tensorflow")
+        _tf_available = False
+
 except (ImportError, AssertionError):
     _tf_available = False  # pylint: disable=invalid-name
 
 try:
     os.environ.setdefault('USE_TORCH', 'YES')
     if os.environ['USE_TORCH'].upper() in ('1', 'ON', 'YES'):
-        logger.info("USE_TORCH override through env variable, disabling PyTorch")
-        _torch_available = False
-    else:
         import torch
         _torch_available = True  # pylint: disable=invalid-name
         logger.info("PyTorch version {} available.".format(torch.__version__))
+
+    else:
+        logger.info("USE_TORCH override through env variable, disabling PyTorch")
+        _torch_available = False
 except ImportError:
     _torch_available = False  # pylint: disable=invalid-name
 

From 81a911cce5d659ef66eddff288489f25fc195f16 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Thu, 19 Dec 2019 15:12:06 +0100
Subject: [PATCH 443/505] Doc, doc, ... doc.

---
 transformers/pipelines.py | 62 ++++++++++++++++++++++++++++++++++++---
 1 file changed, 58 insertions(+), 4 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index b766549121..71e6d0fbed 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -102,9 +102,19 @@ class PipelineDataFormat:
 
     @abstractmethod
     def save(self, data: dict):
+        """
+        Save the provided data object with the representation for the current `DataFormat`.
+        :param data: data to store
+        :return:
+        """
         raise NotImplementedError()
 
     def save_binary(self, data: Union[dict, List[dict]]) -> str:
+        """
+        Save the provided data object as a pickle-formatted binary data on the disk.
+        :param data: data to store
+        :return: (str) Path where the data has been saved
+        """
         path, _ = os.path.splitext(self.output)
         binary_path = os.path.extsep.join((path, 'pickle'))
 
@@ -222,6 +232,42 @@ class Pipeline(_ScikitCompat):
     Base class implementing pipelined operations.
     Pipeline workflow is defined as a sequence of the following operations:
         Input -> Tokenization -> Model Inference -> Post-Processing (Task dependent) -> Output
+
+    Pipeline supports running on CPU or GPU through the device argument. Users can specify
+    device argument as an integer, -1 meaning "CPU", >= 0 referring the CUDA device ordinal.
+
+    Some pipeline, like for instance FeatureExtractionPipeline ('feature-extraction') outputs large
+    tensor object as nested-lists. In order to avoid dumping such large structure as textual data we
+    provide the binary_output constructor argument. If set to True, the output will be stored in the
+    pickle format.
+
+    Arguments:
+        **model**: ``(str, PretrainedModel, TFPretrainedModel)``:
+            Reference to the model to use through this pipeline.
+
+        **tokenizer**: ``(str, PreTrainedTokenizer)``:
+            Reference to the tokenizer to use through this pipeline.
+
+        **args_parser**: ``ArgumentHandler``:
+            Reference to the object in charge of parsing supplied pipeline parameters.
+
+        **device**: ``int``:
+            Device ordinal for CPU/GPU supports. Setting this to -1 will leverage CPU, >=0 will run the model
+            on the associated CUDA device id.
+
+        **binary_output** ``bool`` (default: False):
+            Flag indicating if the output the pipeline should happen in a binary format (i.e. pickle) or as raw text.
+
+    Return:
+        Pipeline returns list or dictionary depending on:
+         - Does the user provided multiple sample
+         - The pipeline expose multiple fields in the output object
+
+    Examples:
+        nlp = pipeline('ner')
+        nlp = pipeline('ner', model='...', config='...', tokenizer='...')
+        nlp = NerPipeline(model='...', config='...', tokenizer='...')
+        nlp = QuestionAnsweringPipeline(model=AutoModel.from_pretrained('...'), tokenizer='...')
     """
     def __init__(self, model, tokenizer: PreTrainedTokenizer = None,
                  args_parser: ArgumentHandler = None, device: int = -1,
@@ -312,11 +358,11 @@ class Pipeline(_ScikitCompat):
 
         # Encode for forward
         with self.device_placement():
+            # TODO : Remove this 512 hard-limit
             inputs = self.tokenizer.batch_encode_plus(
                 inputs, add_special_tokens=True,
                 return_tensors='tf' if is_tf_available() else 'pt',
-                # max_length=self.model.config.max_position_embedding
-                max_length=511
+                max_length=512
             )
 
             # Filter out features not available on specific models
@@ -385,6 +431,8 @@ class NerPipeline(Pipeline):
 
             # Manage correct placement of the tensors
             with self.device_placement():
+
+                # TODO : Remove this 512 hard-limit
                 tokens = self.tokenizer.encode_plus(
                     sentence, return_attention_mask=False,
                     return_tensors='tf' if is_tf_available() else 'pt',
@@ -488,9 +536,12 @@ class QuestionAnsweringPipeline(Pipeline):
         QuestionAnsweringPipeline leverages the SquadExample/SquadFeatures internally.
         This helper method encapsulate all the logic for converting question(s) and context(s) to SquadExample(s).
         We currently support extractive question answering.
-        Args:
+        Arguments:
              question: (str, List[str]) The question to be ask for the associated context
              context: (str, List[str]) The context in which we will look for the answer.
+
+        Returns:
+            SquadExample initialized with the corresponding question and context.
         """
         if isinstance(question, list):
             return [SquadExample(None, q, c, None, None, None) for q, c in zip(question, context)]
@@ -717,7 +768,10 @@ def pipeline(task: str, model: Optional = None,
         Some (optional) post processing for enhancing model's output
 
     Examples:
-        pipeline('ner')
+        pipeline('sentiment-analysis')
+        pipeline('question-answering', model='distilbert-base-uncased-distilled-squad', tokenizer='bert-base-cased')
+        pipeline('ner', model=AutoModel.from_pretrained(...), tokenizer=AutoTokenizer.from_pretrained(...)
+        pipeline('ner', model='https://...pytorch-model.bin', config='https://...config.json', tokenizer='bert-base-cased')
     """
     # Try to infer tokenizer from model name (if provided as str)
     if tokenizer is None:

From ed6ba93912d223886fe0b88dd4ee58b20774beaf Mon Sep 17 00:00:00 2001
From: patrickvonplaten <patrick.v.platen@gmail.com>
Date: Thu, 19 Dec 2019 01:26:01 +0100
Subject: [PATCH 444/505] corrected typo in example for t5 model input argument

---
 transformers/modeling_t5.py    | 4 ++--
 transformers/modeling_tf_t5.py | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/transformers/modeling_t5.py b/transformers/modeling_t5.py
index 263dc33b70..9baf69d02b 100644
--- a/transformers/modeling_t5.py
+++ b/transformers/modeling_t5.py
@@ -693,7 +693,7 @@ class T5Model(T5PreTrainedModel):
         tokenizer = T5Tokenizer.from_pretrained('t5-small')
         model = T5Model.from_pretrained('t5-small')
         input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
-        outputs = model(input_ids)
+        outputs = model(input_ids=input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
     """
@@ -798,7 +798,7 @@ class T5WithLMHeadModel(T5PreTrainedModel):
         tokenizer = T5Tokenizer.from_pretrained('t5-small')
         model = T5WithLMHeadModel.from_pretrained('t5-small')
         input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
-        outputs = model(input_ids, lm_labels=input_ids)
+        outputs = model(input_ids=input_ids, lm_labels=input_ids)
         loss, prediction_scores = outputs[:2]
 
     """
diff --git a/transformers/modeling_tf_t5.py b/transformers/modeling_tf_t5.py
index 1336a1c30d..e803e00c8d 100644
--- a/transformers/modeling_tf_t5.py
+++ b/transformers/modeling_tf_t5.py
@@ -610,7 +610,7 @@ class TFT5Model(TFT5PreTrainedModel):
         tokenizer = T5Tokenizer.from_pretrained('t5-small')
         model = TFT5Model.from_pretrained('t5-small')
         input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
-        outputs = model(input_ids)
+        outputs = model(input_ids=input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
 
     """
@@ -701,7 +701,7 @@ class TFT5WithLMHeadModel(TFT5PreTrainedModel):
         tokenizer = T5Tokenizer.from_pretrained('t5-small')
         model = TFT5WithLMHeadModel.from_pretrained('t5-small')
         input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
-        outputs = model(input_ids)
+        outputs = model(input_ids=input_ids)
         prediction_scores = outputs[0]
 
     """

From 284572efc05a6a8d9e351e886ea3cab0f5f2367a Mon Sep 17 00:00:00 2001
From: Ejar <ejarkm@gmail.com>
Date: Wed, 18 Dec 2019 17:47:47 +0100
Subject: [PATCH 445/505] Updated typo on the link

Updated documentation due to typo
---
 examples/README.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/examples/README.md b/examples/README.md
index b6b3908810..fcd2fe1f6f 100644
--- a/examples/README.md
+++ b/examples/README.md
@@ -467,7 +467,7 @@ Training with the previously defined hyper-parameters yields the following resul
 ## Named Entity Recognition
 
 Based on the scripts [`run_ner.py`](https://github.com/huggingface/transformers/blob/master/examples/run_ner.py) for Pytorch and
-[`run_tf_ner.py`(https://github.com/huggingface/transformers/blob/master/examples/run_tf_ner.py)] for Tensorflow 2.
+[`run_tf_ner.py`](https://github.com/huggingface/transformers/blob/master/examples/run_tf_ner.py) for Tensorflow 2.
 This example fine-tune Bert Multilingual on GermEval 2014 (German NER).
 Details and results for the fine-tuning provided by @stefan-it.
 

From 62c1fc3c1ecdfab787ee3c34d1ec1eba65c18877 Mon Sep 17 00:00:00 2001
From: Francesco <francesco.cottone@vidiemme.it>
Date: Thu, 19 Dec 2019 14:43:10 +0100
Subject: [PATCH 446/505] Removed duplicate XLMConfig, XLMForQuestionAnswering
 and XLMTokenizer from import statement of run_squad.py script

---
 examples/run_squad.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/examples/run_squad.py b/examples/run_squad.py
index 34c31c3bb8..1ff6983f62 100644
--- a/examples/run_squad.py
+++ b/examples/run_squad.py
@@ -61,7 +61,6 @@ MODEL_CLASSES = {
     'xlm': (XLMConfig, XLMForQuestionAnswering, XLMTokenizer),
     'distilbert': (DistilBertConfig, DistilBertForQuestionAnswering, DistilBertTokenizer),
     'albert': (AlbertConfig, AlbertForQuestionAnswering, AlbertTokenizer),
-    'xlm': (XLMConfig, XLMForQuestionAnswering, XLMTokenizer)
 }
 
 def set_seed(args):

From a1f1dce0ae511ef7766c6b6a8f5ebf9118279e73 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Thu, 19 Dec 2019 12:25:55 -0500
Subject: [PATCH 447/505] Correct max position for SQUAD and TFDS

---
 transformers/data/processors/squad.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/transformers/data/processors/squad.py b/transformers/data/processors/squad.py
index 84aa429e26..8e72bbbd6d 100644
--- a/transformers/data/processors/squad.py
+++ b/transformers/data/processors/squad.py
@@ -571,7 +571,9 @@ class SquadExample(object):
         # Start end end positions only has a value during evaluation.
         if start_position_character is not None and not is_impossible:
             self.start_position = char_to_word_offset[start_position_character]
-            self.end_position = char_to_word_offset[start_position_character + len(answer_text) - 1]
+            self.end_position = char_to_word_offset[
+                min(start_position_character + len(answer_text) - 1, len(char_to_word_offset) - 1)
+            ]
 
 
 class SquadFeatures(object):

From 33adab2b91697b3e78af618a21ab9f1176281165 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Thu, 19 Dec 2019 12:40:43 -0500
Subject: [PATCH 448/505] Fix albert example

---
 transformers/modeling_tf_albert.py | 4 ++--
 transformers/modeling_utils.py     | 5 -----
 2 files changed, 2 insertions(+), 7 deletions(-)

diff --git a/transformers/modeling_tf_albert.py b/transformers/modeling_tf_albert.py
index d1650d41a8..ac55a73fa3 100644
--- a/transformers/modeling_tf_albert.py
+++ b/transformers/modeling_tf_albert.py
@@ -587,8 +587,8 @@ class TFAlbertModel(TFAlbertPreTrainedModel):
         import tensorflow as tf
         from transformers import AlbertTokenizer, TFAlbertModel
 
-        tokenizer = AlbertTokenizer.from_pretrained('bert-base-uncased')
-        model = TFAlbertModel.from_pretrained('bert-base-uncased')
+        tokenizer = AlbertTokenizer.from_pretrained('albert-base-v1')
+        model = TFAlbertModel.from_pretrained('albert-base-v1')
         input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute"))[None, :]  # Batch size 1
         outputs = model(input_ids)
         last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
diff --git a/transformers/modeling_utils.py b/transformers/modeling_utils.py
index 9bd99b25dc..eff54f71e1 100644
--- a/transformers/modeling_utils.py
+++ b/transformers/modeling_utils.py
@@ -327,11 +327,6 @@ class PreTrainedModel(nn.Module):
             model = BertModel.from_pretrained('./tf_model/my_tf_checkpoint.ckpt.index', from_tf=True, config=config)
 
         """
-        if pretrained_model_name_or_path is not None and (
-                "albert" in pretrained_model_name_or_path and "v2" in pretrained_model_name_or_path):
-            logger.warning("There is currently an upstream reproducibility issue with ALBERT v2 models. Please see " +
-                           "https://github.com/google-research/google-research/issues/119 for more information.")
-
         config = kwargs.pop('config', None)
         state_dict = kwargs.pop('state_dict', None)
         cache_dir = kwargs.pop('cache_dir', None)

From 3492a6ec17e207a2830e061528eae9c53639c234 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Thu, 19 Dec 2019 19:06:44 +0100
Subject: [PATCH 449/505] Addressing Thom's comments.

---
 transformers/pipelines.py | 70 ++++++++++++++++++---------------------
 1 file changed, 33 insertions(+), 37 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 71e6d0fbed..e4bf9e0894 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -30,6 +30,7 @@ from transformers import AutoConfig, AutoTokenizer, PreTrainedTokenizer, Pretrai
     SquadExample, squad_convert_examples_to_features, is_tf_available, is_torch_available, logger
 
 if is_tf_available():
+    import tensorflow as tf
     from transformers import TFAutoModel, TFAutoModelForSequenceClassification, \
         TFAutoModelForQuestionAnswering, TFAutoModelForTokenClassification
 
@@ -79,9 +80,9 @@ class PipelineDataFormat:
     """
     SUPPORTED_FORMATS = ['json', 'csv', 'pipe']
 
-    def __init__(self, output: Optional[str], path: Optional[str], column: Optional[str]):
+    def __init__(self, output: Optional[str], input: Optional[str], column: Optional[str]):
         self.output = output
-        self.path = path
+        self.path = input
         self.column = column.split(',') if column else ['']
         self.is_multi_columns = len(self.column) > 1
 
@@ -92,7 +93,7 @@ class PipelineDataFormat:
             if exists(abspath(self.output)):
                 raise OSError('{} already exists on disk'.format(self.output))
 
-        if path is not None:
+        if input is not None:
             if not exists(abspath(self.path)):
                 raise OSError('{} doesnt exist on disk'.format(self.path))
 
@@ -136,8 +137,8 @@ class PipelineDataFormat:
 
 
 class CsvPipelineDataFormat(PipelineDataFormat):
-    def __init__(self, output: Optional[str], path: Optional[str], column: Optional[str]):
-        super().__init__(output, path, column)
+    def __init__(self, output: Optional[str], input: Optional[str], column: Optional[str]):
+        super().__init__(output, input, column)
 
     def __iter__(self):
         with open(self.path, 'r') as f:
@@ -157,10 +158,10 @@ class CsvPipelineDataFormat(PipelineDataFormat):
 
 
 class JsonPipelineDataFormat(PipelineDataFormat):
-    def __init__(self, output: Optional[str], path: Optional[str], column: Optional[str]):
-        super().__init__(output, path, column)
+    def __init__(self, output: Optional[str], input: Optional[str], column: Optional[str]):
+        super().__init__(output, input, column)
 
-        with open(path, 'r') as f:
+        with open(input, 'r') as f:
             self._entries = json.load(f)
 
     def __iter__(self):
@@ -321,11 +322,9 @@ class Pipeline(_ScikitCompat):
             Context manager
         """
         if is_tf_available():
-            import tensorflow as tf
             with tf.device('/CPU:0' if self.device == -1 else '/device:GPU:{}'.format(self.device)):
                 yield
         else:
-            import torch
             if self.device >= 0:
                 torch.cuda.set_device(self.device)
 
@@ -358,11 +357,10 @@ class Pipeline(_ScikitCompat):
 
         # Encode for forward
         with self.device_placement():
-            # TODO : Remove this 512 hard-limit
             inputs = self.tokenizer.batch_encode_plus(
                 inputs, add_special_tokens=True,
                 return_tensors='tf' if is_tf_available() else 'pt',
-                max_length=512
+                max_length=self.tokenizer.max_len
             )
 
             # Filter out features not available on specific models
@@ -379,11 +377,10 @@ class Pipeline(_ScikitCompat):
         """
         if is_tf_available():
             # TODO trace model
-            predictions = self.model(inputs)[0]
+            predictions = self.model(inputs, training=False)[0]
         else:
-            import torch
             with torch.no_grad():
-                predictions = self.model(**inputs)[0]
+                predictions = self.model(**inputs).cpu()[0]
 
         return predictions.numpy()
 
@@ -432,19 +429,18 @@ class NerPipeline(Pipeline):
             # Manage correct placement of the tensors
             with self.device_placement():
 
-                # TODO : Remove this 512 hard-limit
                 tokens = self.tokenizer.encode_plus(
                     sentence, return_attention_mask=False,
                     return_tensors='tf' if is_tf_available() else 'pt',
-                    max_length=512
+                    max_length=self.tokenizer.max_len
                 )
 
                 # Forward
-                if is_torch_available():
+                if is_tf_available():
+                    entities = self.model(**tokens)[0][0].numpy()
+                else:
                     with torch.no_grad():
                         entities = self.model(**tokens)[0][0].cpu().numpy()
-                else:
-                    entities = self.model(tokens)[0][0].numpy()
 
             # Normalize scores
             answer, token_start = [], 1
@@ -484,28 +480,29 @@ class QuestionAnsweringArgumentHandler(ArgumentHandler):
             else:
                 kwargs['X'] = list(args)
 
-            # Generic compatibility with sklearn and Keras
-            # Batched data
+        # Generic compatibility with sklearn and Keras
+        # Batched data
         if 'X' in kwargs or 'data' in kwargs:
-            data = kwargs['X'] if 'X' in kwargs else kwargs['data']
+            inputs = kwargs['X'] if 'X' in kwargs else kwargs['data']
 
-            if not isinstance(data, list):
-                data = [data]
+            if isinstance(inputs, dict):
+                inputs = [inputs]
+            else:
+                # Copy to avoid overriding arguments
+                inputs = [i for i in inputs]
 
-            for i, item in enumerate(data):
+            for i, item in enumerate(inputs):
                 if isinstance(item, dict):
                     if any(k not in item for k in ['question', 'context']):
                         raise KeyError('You need to provide a dictionary with keys {question:..., context:...}')
-                    data[i] = QuestionAnsweringPipeline.create_sample(**item)
 
-                elif isinstance(item, SquadExample):
-                    continue
-                else:
+                    inputs[i] = QuestionAnsweringPipeline.create_sample(**item)
+
+                elif not isinstance(item, SquadExample):
                     raise ValueError(
                         '{} argument needs to be of type (list[SquadExample | dict], SquadExample, dict)'
                             .format('X' if 'X' in kwargs else 'data')
                     )
-            inputs = data
 
             # Tabular input
         elif 'question' in kwargs and 'context' in kwargs:
@@ -588,12 +585,10 @@ class QuestionAnsweringPipeline(Pipeline):
         # Manage tensor allocation on correct device
         with self.device_placement():
             if is_tf_available():
-                import tensorflow as tf
                 fw_args = {k: tf.constant(v) for (k, v) in fw_args.items()}
                 start, end = self.model(fw_args)
                 start, end = start.numpy(), end.numpy()
             else:
-                import torch
                 with torch.no_grad():
                     # Retrieve the score for the context tokens only (removing question tokens)
                     fw_args = {k: torch.tensor(v) for (k, v) in fw_args.items()}
@@ -812,8 +807,9 @@ def pipeline(task: str, model: Optional = None,
     if isinstance(config, str):
         config = AutoConfig.from_pretrained(config)
 
-    if allocator.__name__.startswith('TF'):
-        model = allocator.from_pretrained(model, config=config, from_pt=from_pt)
-    else:
-        model = allocator.from_pretrained(model, config=config, from_tf=from_tf)
+    if isinstance(model, str):
+        if allocator.__name__.startswith('TF'):
+            model = allocator.from_pretrained(model, config=config, from_pt=from_pt)
+        else:
+            model = allocator.from_pretrained(model, config=config, from_tf=from_tf)
     return task(model, tokenizer, **kwargs)

From a305067f2d6ca74865c6d686608a1428e476a32f Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Thu, 19 Dec 2019 19:41:48 +0100
Subject: [PATCH 450/505] Removed __main__

---
 setup.py                 |  5 -----
 transformers/__main__.py | 36 ------------------------------------
 2 files changed, 41 deletions(-)
 delete mode 100644 transformers/__main__.py

diff --git a/setup.py b/setup.py
index b3b6e2e063..56c4d1733b 100644
--- a/setup.py
+++ b/setup.py
@@ -68,11 +68,6 @@ setup(
     scripts=[
         'transformers-cli'
     ],
-    entry_points={
-      'console_scripts': [
-        "transformers=transformers.__main__:main",
-      ]
-    },
     # python_requires='>=3.5.0',
     classifiers=[
           'Intended Audience :: Science/Research',
diff --git a/transformers/__main__.py b/transformers/__main__.py
deleted file mode 100644
index a6e9ae65e0..0000000000
--- a/transformers/__main__.py
+++ /dev/null
@@ -1,36 +0,0 @@
-# coding: utf8
-
-def main():
-    import sys
-    if len(sys.argv) < 2 or sys.argv[1] not in ["convert", "train", "predict", "serve"]:
-        print(
-        "First argument to `transformers` command line interface should be one of: \n"
-        ">> convert serve train predict")
-    if sys.argv[1] == "convert":
-        from transformers.commands import convert
-        convert(sys.argv)
-    elif sys.argv[1] == "train":
-        from transformers.commands import train
-        train(sys.argv)
-    elif sys.argv[1] == "serve":
-        pass
-        # from argparse import ArgumentParser
-        # from transformers.commands.serving import ServeCommand
-        # parser = ArgumentParser('Transformers CLI tool', usage='transformers serve <command> [<args>]')
-        # commands_parser = parser.add_subparsers(help='transformers-cli command helpers')
-
-        # # Register commands
-        # ServeCommand.register_subcommand(commands_parser)
-
-        # # Let's go
-        # args = parser.parse_args()
-
-        # if not hasattr(args, 'func'):
-        #     parser.print_help()
-        #     exit(1)
-        # # Run
-        # service = args.func(args)
-        # service.run()
-
-if __name__ == '__main__':
-    main()

From 149dc376aa5070db04cdf6e3358e87cab6670251 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Thu, 19 Dec 2019 20:34:28 +0100
Subject: [PATCH 451/505] fix tests

---
 transformers/configuration_utils.py | 2 --
 1 file changed, 2 deletions(-)

diff --git a/transformers/configuration_utils.py b/transformers/configuration_utils.py
index 1aede2d6eb..f692c9b132 100644
--- a/transformers/configuration_utils.py
+++ b/transformers/configuration_utils.py
@@ -57,8 +57,6 @@ class PretrainedConfig(object):
         self.use_bfloat16 = kwargs.pop('use_bfloat16', False)
         self.pruned_heads = kwargs.pop('pruned_heads', {})
         self.is_decoder = kwargs.pop('is_decoder', False)
-        self.id2label = kwargs.pop('id2label', {i: 'LABEL_{}'.format(i) for i in range(self.num_labels)})
-        self.label2id = kwargs.pop('label2id', dict(zip(self.id2label.values(), self.id2label.keys())))
 
         # Fine-tuning task arguments
         self.finetuning_task = kwargs.pop('finetuning_task', None)

From e4baa68ddbcb28488d7cef44ea5483a955d2effb Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Thu, 19 Dec 2019 20:37:26 +0100
Subject: [PATCH 452/505] tick-tock cc @julien-c

---
 .circleci/config.yml | 12 ------------
 1 file changed, 12 deletions(-)

diff --git a/.circleci/config.yml b/.circleci/config.yml
index 9a81eea902..7a64eaba7d 100644
--- a/.circleci/config.yml
+++ b/.circleci/config.yml
@@ -80,17 +80,6 @@ jobs:
             - run: sudo pip install pytest
             - run: sudo pip install mecab-python3
             - run: RUN_CUSTOM_TOKENIZERS=1 python -m pytest -sv ./transformers/tests/tokenization_bert_japanese_test.py
-    build_py2_custom_tokenizers:
-        working_directory: ~/transformers
-        docker:
-            - image: circleci/python:2.7
-        steps:
-            - checkout
-            - run: sudo pip install --progress-bar off .
-            - run: sudo pip install pytest
-            - run: sudo apt-get -y install libmecab-dev mecab mecab-ipadic-utf8 swig
-            - run: sudo pip install mecab-python
-            - run: RUN_CUSTOM_TOKENIZERS=1 python -m pytest -sv ./transformers/tests/tokenization_bert_japanese_test.py
     deploy_doc:
         working_directory: ~/transformers
         docker:
@@ -124,7 +113,6 @@ workflows:
         jobs:
             - repository_consistency
             - build_py3_custom_tokenizers
-            - build_py2_custom_tokenizers
             - build_py3_torch_and_tf
             - build_py3_torch
             - build_py3_tf

From 3376adc05157ba826acafd49f07f9a01ae30eb07 Mon Sep 17 00:00:00 2001
From: Stefan Schweter <stefan@schweter.it>
Date: Thu, 19 Dec 2019 21:30:23 +0100
Subject: [PATCH 453/505] configuration/modeling/tokenization: add various
 fine-tuned XLM-RoBERTa models for English, German, Spanish and Dutch (CoNLL
 datasets)

---
 transformers/configuration_xlm_roberta.py | 4 ++++
 transformers/modeling_xlm_roberta.py      | 4 ++++
 transformers/tokenization_xlm_roberta.py  | 4 ++++
 3 files changed, 12 insertions(+)

diff --git a/transformers/configuration_xlm_roberta.py b/transformers/configuration_xlm_roberta.py
index d7a26538c5..5b6955f4f8 100644
--- a/transformers/configuration_xlm_roberta.py
+++ b/transformers/configuration_xlm_roberta.py
@@ -27,6 +27,10 @@ logger = logging.getLogger(__name__)
 XLM_ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP = {
     'xlm-roberta-base': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-base-config.json",
     'xlm-roberta-large': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-large-config.json",
+    'xlm-roberta-large-finetuned-conll02-dutch': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-large-finetuned-conll02-dutch-config.json",
+    'xlm-roberta-large-finetuned-conll02-spanish': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-large-finetuned-conll02-spanish-config.json",
+    'xlm-roberta-large-finetuned-conll03-english': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-large-finetuned-conll03-english-config.json",
+    'xlm-roberta-large-finetuned-conll03-german': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-large-finetuned-conll03-german-config.json",
 }
 
 
diff --git a/transformers/modeling_xlm_roberta.py b/transformers/modeling_xlm_roberta.py
index 8095c46a16..0bdce941a5 100644
--- a/transformers/modeling_xlm_roberta.py
+++ b/transformers/modeling_xlm_roberta.py
@@ -29,6 +29,10 @@ logger = logging.getLogger(__name__)
 XLM_ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP = {
     'xlm-roberta-base': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-base-pytorch_model.bin",
     'xlm-roberta-large': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-large-pytorch_model.bin",
+    'xlm-roberta-large-finetuned-conll02-dutch': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-large-finetuned-conll02-dutch-pytorch_model.bin",
+    'xlm-roberta-large-finetuned-conll02-spanish': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-large-finetuned-conll02-spanish-pytorch_model.bin",
+    'xlm-roberta-large-finetuned-conll03-english': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-large-finetuned-conll03-english-pytorch_model.bin",
+    'xlm-roberta-large-finetuned-conll03-german': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-large-finetuned-conll03-german-pytorch_model.bin",
 }
 
 
diff --git a/transformers/tokenization_xlm_roberta.py b/transformers/tokenization_xlm_roberta.py
index 453c4375c6..4397e7b031 100644
--- a/transformers/tokenization_xlm_roberta.py
+++ b/transformers/tokenization_xlm_roberta.py
@@ -32,6 +32,10 @@ PRETRAINED_VOCAB_FILES_MAP = {
     {
     'xlm-roberta-base': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-base-sentencepiece.bpe.model",
     'xlm-roberta-large': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-large-sentencepiece.bpe.model",
+    'xlm-roberta-large-finetuned-conll02-dutch': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-large-finetuned-conll02-dutch-sentencepiece.bpe.model",
+    'xlm-roberta-large-finetuned-conll02-spanish': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-large-finetuned-conll02-spanish-sentencepiece.bpe.model",
+    'xlm-roberta-large-finetuned-conll03-english': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-large-finetuned-conll03-english-sentencepiece.bpe.model",
+    'xlm-roberta-large-finetuned-conll03-german': "https://s3.amazonaws.com/models.huggingface.co/bert/xlm-roberta-large-finetuned-conll03-german-sentencepiece.bpe.model",
     }
 }
 

From 9a399ead253e27792cbf0ef386cc39f9b7084f8f Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Thu, 19 Dec 2019 15:45:48 -0500
Subject: [PATCH 454/505] Revert incorrect #1778

---
 transformers/modeling_tf_pytorch_utils.py | 3 ---
 transformers/modeling_utils.py            | 2 --
 2 files changed, 5 deletions(-)

diff --git a/transformers/modeling_tf_pytorch_utils.py b/transformers/modeling_tf_pytorch_utils.py
index d885fd23b3..190caff18d 100644
--- a/transformers/modeling_tf_pytorch_utils.py
+++ b/transformers/modeling_tf_pytorch_utils.py
@@ -119,9 +119,6 @@ def load_pytorch_weights_in_tf2_model(tf_model, pt_state_dict, tf_inputs=None, a
             new_key = key.replace('gamma', 'weight')
         if 'beta' in key:
             new_key = key.replace('beta', 'bias')
-        # DialoGPT format
-        if key == 'lm_head.decoder.weight':
-            new_key = 'lm_head.weight'
         if new_key:
             old_keys.append(key)
             new_keys.append(new_key)
diff --git a/transformers/modeling_utils.py b/transformers/modeling_utils.py
index eff54f71e1..cce234838a 100644
--- a/transformers/modeling_utils.py
+++ b/transformers/modeling_utils.py
@@ -437,8 +437,6 @@ class PreTrainedModel(nn.Module):
                     new_key = key.replace('gamma', 'weight')
                 if 'beta' in key:
                     new_key = key.replace('beta', 'bias')
-                if key == 'lm_head.decoder.weight':
-                    new_key = 'lm_head.weight'
                 if new_key:
                     old_keys.append(key)
                     new_keys.append(new_key)

From 1718fb9e7444a0883a550ab90c61bb8bf8a97076 Mon Sep 17 00:00:00 2001
From: Aidan Kierans <31550769+aidankierans@users.noreply.github.com>
Date: Thu, 19 Dec 2019 16:23:18 -0500
Subject: [PATCH 455/505] Minor/basic text fixes (#2229)

* Small clarification

Matches line 431 to line 435 for additional clarity and consistency.

* Fixed minor typo

The letter "s" was previously omitted from the word "docstrings".
---
 CONTRIBUTING.md               | 2 +-
 examples/run_lm_finetuning.py | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
index 8228dd59d8..7d7f2c73ff 100644
--- a/CONTRIBUTING.md
+++ b/CONTRIBUTING.md
@@ -168,7 +168,7 @@ Follow these steps to start contributing:
    to be merged;
 4. Make sure pre-existing tests still pass;
 5. Add high-coverage tests. No quality test, no merge;
-6. All public methods must have informative doctrings;
+6. All public methods must have informative docstrings;
 
 
 ### Style guide
diff --git a/examples/run_lm_finetuning.py b/examples/run_lm_finetuning.py
index c4c73e71af..d8127e24a5 100644
--- a/examples/run_lm_finetuning.py
+++ b/examples/run_lm_finetuning.py
@@ -428,9 +428,9 @@ def main():
     parser.add_argument('--gradient_accumulation_steps', type=int, default=1,
                         help="Number of updates steps to accumulate before performing a backward/update pass.")
     parser.add_argument("--learning_rate", default=5e-5, type=float,
-                        help="The initial learning rate for Adam.")
+                        help="The initial learning rate for Adam optimizer.")
     parser.add_argument("--weight_decay", default=0.0, type=float,
-                        help="Weight deay if we apply some.")
+                        help="Weight decay if we apply some.")
     parser.add_argument("--adam_epsilon", default=1e-8, type=float,
                         help="Epsilon for Adam optimizer.")
     parser.add_argument("--max_grad_norm", default=1.0, type=float,

From a5a06a851e1da79138e53978aa079a093f243dde Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Thu, 19 Dec 2019 16:24:20 -0500
Subject: [PATCH 456/505] [doc] Param name consistency

---
 examples/run_lm_finetuning.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/examples/run_lm_finetuning.py b/examples/run_lm_finetuning.py
index d8127e24a5..75848d5acc 100644
--- a/examples/run_lm_finetuning.py
+++ b/examples/run_lm_finetuning.py
@@ -428,7 +428,7 @@ def main():
     parser.add_argument('--gradient_accumulation_steps', type=int, default=1,
                         help="Number of updates steps to accumulate before performing a backward/update pass.")
     parser.add_argument("--learning_rate", default=5e-5, type=float,
-                        help="The initial learning rate for Adam optimizer.")
+                        help="The initial learning rate for Adam.")
     parser.add_argument("--weight_decay", default=0.0, type=float,
                         help="Weight decay if we apply some.")
     parser.add_argument("--adam_epsilon", default=1e-8, type=float,

From f25e9b6f771ea1a10f4525bdb212f841efcdbd3a Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Thu, 19 Dec 2019 18:28:17 -0500
Subject: [PATCH 457/505] [hf_bucket_url] support for cloudfront urls

---
 transformers/file_utils.py | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/transformers/file_utils.py b/transformers/file_utils.py
index 16010f7e0a..4a9de9b53c 100644
--- a/transformers/file_utils.py
+++ b/transformers/file_utils.py
@@ -77,6 +77,7 @@ DUMMY_INPUTS = [[7, 6, 0, 0, 1], [1, 2, 3, 0, 0], [0, 0, 0, 4, 5]]
 DUMMY_MASK = [[1, 1, 1, 1, 1], [1, 1, 1, 0, 0], [0, 0, 0, 1, 1]]
 
 S3_BUCKET_PREFIX = "https://s3.amazonaws.com/models.huggingface.co/bert"
+CLOUDFRONT_DISTRIB_PREFIX = "https://d2ws9o8vfrpkyk.cloudfront.net"
 
 
 def is_torch_available():
@@ -114,11 +115,12 @@ def is_remote_url(url_or_filename):
     parsed = urlparse(url_or_filename)
     return parsed.scheme in ('http', 'https', 's3')
 
-def hf_bucket_url(identifier, postfix=None):
+def hf_bucket_url(identifier, postfix=None, cdn=False):
+    endpoint = CLOUDFRONT_DISTRIB_PREFIX if cdn else S3_BUCKET_PREFIX
     if postfix is None:
-        return "/".join((S3_BUCKET_PREFIX, identifier))
+        return "/".join((endpoint, identifier))
     else:
-        return "/".join((S3_BUCKET_PREFIX, identifier, postfix))
+        return "/".join((endpoint, identifier, postfix))
 
 
 def url_to_filename(url, etag=None):
@@ -126,7 +128,7 @@ def url_to_filename(url, etag=None):
     Convert `url` into a hashed filename in a repeatable way.
     If `etag` is specified, append its hash to the url's, delimited
     by a period.
-    If the url ends with .h5 (Keras HDF5 weights) ands '.h5' to the name
+    If the url ends with .h5 (Keras HDF5 weights) adds '.h5' to the name
     so that TF 2.0 can identify it as a HDF5 file
     (see https://github.com/tensorflow/tensorflow/blob/00fad90125b18b80fe054de1055770cfb8fe4ba3/tensorflow/python/keras/engine/network.py#L1380)
     """

From 15d897ff4a29e851285844e82763155a3f9f86b0 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Thu, 19 Dec 2019 18:29:22 -0500
Subject: [PATCH 458/505] [http] customizable requests user-agent

---
 transformers/file_utils.py | 25 +++++++++++++++++++------
 1 file changed, 19 insertions(+), 6 deletions(-)

diff --git a/transformers/file_utils.py b/transformers/file_utils.py
index 4a9de9b53c..611202159d 100644
--- a/transformers/file_utils.py
+++ b/transformers/file_utils.py
@@ -23,6 +23,7 @@ from botocore.exceptions import ClientError
 import requests
 from tqdm.auto import tqdm
 from contextlib import contextmanager
+from . import __version__
 
 logger = logging.getLogger(__name__)  # pylint: disable=invalid-name
 
@@ -173,7 +174,7 @@ def filename_to_url(filename, cache_dir=None):
     return url, etag
 
 
-def cached_path(url_or_filename, cache_dir=None, force_download=False, proxies=None, resume_download=False):
+def cached_path(url_or_filename, cache_dir=None, force_download=False, proxies=None, resume_download=False, user_agent=None):
     """
     Given something that might be a URL (or might be a local path),
     determine which. If it's a URL, download the file and cache it, and
@@ -183,6 +184,7 @@ def cached_path(url_or_filename, cache_dir=None, force_download=False, proxies=N
         cache_dir: specify a cache directory to save the file to (overwrite the default cache dir).
         force_download: if True, re-dowload the file even if it's already cached in the cache dir.
         resume_download: if True, resume the download if incompletly recieved file is found.
+        user_agent: Optional string or dict that will be appended to the user-agent on remote requests.
     """
     if cache_dir is None:
         cache_dir = TRANSFORMERS_CACHE
@@ -195,7 +197,7 @@ def cached_path(url_or_filename, cache_dir=None, force_download=False, proxies=N
         # URL, so get it from the cache (downloading if necessary)
         return get_from_cache(url_or_filename, cache_dir=cache_dir,
             force_download=force_download, proxies=proxies,
-            resume_download=resume_download)
+            resume_download=resume_download, user_agent=user_agent)
     elif os.path.exists(url_or_filename):
         # File, and it exists.
         return url_or_filename
@@ -256,8 +258,19 @@ def s3_get(url, temp_file, proxies=None):
     s3_resource.Bucket(bucket_name).download_fileobj(s3_path, temp_file)
 
 
-def http_get(url, temp_file, proxies=None, resume_size=0):
-    headers={'Range':'bytes=%d-'%(resume_size,)} if resume_size > 0 else None
+def http_get(url, temp_file, proxies=None, resume_size=0, user_agent=None):
+    ua = "transformers/{}; python/{}".format(__version__, sys.version.split()[0])
+    if isinstance(user_agent, dict):
+        ua += "; " + "; ".join(
+            "{}/{}".format(k, v) for k, v in user_agent.items()
+        )
+    elif isinstance(user_agent, six.string_types):
+        ua += "; "+ user_agent
+    headers = {
+        "user-agent": ua
+    }
+    if resume_size > 0:
+        headers['Range'] = 'bytes=%d-' % (resume_size,)
     response = requests.get(url, stream=True, proxies=proxies, headers=headers)
     if response.status_code == 416:  # Range not satisfiable
         return
@@ -271,7 +284,7 @@ def http_get(url, temp_file, proxies=None, resume_size=0):
     progress.close()
 
 
-def get_from_cache(url, cache_dir=None, force_download=False, proxies=None, etag_timeout=10, resume_download=False):
+def get_from_cache(url, cache_dir=None, force_download=False, proxies=None, etag_timeout=10, resume_download=False, user_agent=None):
     """
     Given a URL, look for the corresponding dataset in the local cache.
     If it's not there, download it. Then return the path to the cached file.
@@ -342,7 +355,7 @@ def get_from_cache(url, cache_dir=None, force_download=False, proxies=None, etag
                     logger.warn('Warning: resumable downloads are not implemented for "s3://" urls')
                 s3_get(url, temp_file, proxies=proxies)
             else:
-                http_get(url, temp_file, proxies=proxies, resume_size=resume_size)
+                http_get(url, temp_file, proxies=proxies, resume_size=resume_size, user_agent=user_agent)
 
             # we are copying the file before closing it, so flush to avoid truncation
             temp_file.flush()

From 9d0d1cd339ca97e6a40e1f898fa3d3350e46e350 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Fri, 20 Dec 2019 09:30:37 +0100
Subject: [PATCH 459/505] Filter out entity for NER task.

---
 transformers/pipelines.py | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index e4bf9e0894..6b805a06db 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -450,11 +450,12 @@ class NerPipeline(Pipeline):
                 score = np.exp(entities[token_start]) / np.exp(entities[token_start]).sum(-1, keepdims=True)
                 label_idx = score.argmax()
 
-                answer += [{
-                    'word': words[idx],
-                    'score': score[label_idx].item(),
-                    'entity': self.model.config.id2label[label_idx]
-                }]
+                if label_idx > 0:
+                    answer += [{
+                        'word': words[idx],
+                        'score': score[label_idx].item(),
+                        'entity': self.model.config.id2label[label_idx]
+                    }]
 
                 # Update token start
                 token_start += len(list(word))

From e516a34a158593f82b6f22dc1e568ff8996f0389 Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Fri, 20 Dec 2019 09:38:08 +0100
Subject: [PATCH 460/505] Use BasicTokenizer to split over whitespaces.

---
 transformers/pipelines.py | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 6b805a06db..c3e0ad1a8f 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -27,7 +27,7 @@ from typing import Union, Optional, Tuple, List, Dict
 import numpy as np
 
 from transformers import AutoConfig, AutoTokenizer, PreTrainedTokenizer, PretrainedConfig, \
-    SquadExample, squad_convert_examples_to_features, is_tf_available, is_torch_available, logger
+    SquadExample, squad_convert_examples_to_features, is_tf_available, is_torch_available, logger, BasicTokenizer
 
 if is_tf_available():
     import tensorflow as tf
@@ -416,12 +416,19 @@ class NerPipeline(Pipeline):
     Named Entity Recognition pipeline using ModelForTokenClassification head.
     """
 
+    def __init__(self, model, tokenizer: PreTrainedTokenizer = None,
+                 args_parser: ArgumentHandler = None, device: int = -1,
+                 binary_output: bool = False):
+        super().__init__(model, tokenizer, args_parser, device, binary_output)
+
+        self._basic_tokenizer = BasicTokenizer(do_lower_case=False)
+
     def __call__(self, *texts, **kwargs):
         inputs, answers = self._args_parser(*texts, **kwargs), []
         for sentence in inputs:
 
             # Ugly token to word idx mapping (for now)
-            token_to_word, words = [], sentence.split(' ')
+            token_to_word, words = [], self._basic_tokenizer.tokenize(sentence)
             for i, w in enumerate(words):
                 tokens = self.tokenizer.tokenize(w)
                 token_to_word += [i] * len(tokens)

From 61d9ee45e3f19b9c661e078e7f57dbe8fb8c812c Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Fri, 20 Dec 2019 11:47:56 +0100
Subject: [PATCH 461/505] All tests are green.

---
 transformers/pipelines.py            |   9 +-
 transformers/tests/pipelines_test.py | 247 ++++++++++++++++-----------
 2 files changed, 154 insertions(+), 102 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index c3e0ad1a8f..4dde62cbe5 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -343,8 +343,9 @@ class Pipeline(_ScikitCompat):
         if 'distilbert' not in model_type and 'xlm' not in model_type:
             args += ['token_type_ids']
 
-        if 'xlnet' in model_type or 'xlm' in model_type:
-            args += ['cls_index', 'p_mask']
+        # PR #1548 (CLI) There is an issue with attention_mask
+        # if 'xlnet' in model_type or 'xlm' in model_type:
+        #     args += ['cls_index', 'p_mask']
 
         if isinstance(features, dict):
             return {k: features[k] for k in args}
@@ -380,7 +381,7 @@ class Pipeline(_ScikitCompat):
             predictions = self.model(inputs, training=False)[0]
         else:
             with torch.no_grad():
-                predictions = self.model(**inputs).cpu()[0]
+                predictions = self.model(**inputs)[0].cpu()
 
         return predictions.numpy()
 
@@ -444,7 +445,7 @@ class NerPipeline(Pipeline):
 
                 # Forward
                 if is_tf_available():
-                    entities = self.model(**tokens)[0][0].numpy()
+                    entities = self.model(tokens)[0][0].numpy()
                 else:
                     with torch.no_grad():
                         entities = self.model(**tokens)[0][0].cpu().numpy()
diff --git a/transformers/tests/pipelines_test.py b/transformers/tests/pipelines_test.py
index ee10234269..a8fe668221 100644
--- a/transformers/tests/pipelines_test.py
+++ b/transformers/tests/pipelines_test.py
@@ -1,113 +1,164 @@
 import unittest
 from unittest.mock import patch
 
+from typing import Iterable
+
+from transformers import pipeline
+from transformers.tests.utils import require_tf, require_torch
 
 QA_FINETUNED_MODELS = {
-    'bert-large-uncased-whole-word-masking-finetuned-squad',
-    'bert-large-cased-whole-word-masking-finetuned-squad',
-    'distilbert-base-uncased-distilled-squad',
+    ('bert-base-uncased', 'bert-large-uncased-whole-word-masking-finetuned-squad', None),
+    ('bert-base-cased', 'bert-large-cased-whole-word-masking-finetuned-squad', None),
+    ('bert-base-uncased', 'distilbert-base-uncased-distilled-squad', None)
+}
 
+NER_FINETUNED_MODELS = {
+    (
+        'bert-base-cased',
+        'https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-cased-finetuned-conll03-english-pytorch_model.bin',
+        'https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-cased-finetuned-conll03-english-config.json'
+    )
+}
+
+FEATURE_EXTRACT_FINETUNED_MODELS = {
+   ('bert-base-cased', 'bert-base-cased', None),
+   # ('xlnet-base-cased', 'xlnet-base-cased', None), # Disabled for now as it crash for TF2
+   ('distilbert-base-uncased', 'distilbert-base-uncased', None)
+}
+
+TEXT_CLASSIF_FINETUNED_MODELS = {
+    (
+        'bert-base-uncased',
+        'https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-finetuned-sst-2-english-pytorch_model.bin',
+        'https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-finetuned-sst-2-english-config.json'
+    )
 }
 
 
-class QuestionAnsweringPipelineTest(unittest.TestCase):
-    def check_answer_structure(self, answer, batch, topk):
-        self.assertIsInstance(answer, list)
-        self.assertEqual(len(answer), batch)
-        self.assertIsInstance(answer[0], list)
-        self.assertEqual(len(answer[0]), topk)
-        self.assertIsInstance(answer[0][0], dict)
-
-        for item in answer[0]:
-            self.assertTrue('start' in item)
-            self.assertTrue('end' in item)
-            self.assertTrue('score' in item)
-            self.assertTrue('answer' in item)
-
-    def question_answering_pipeline(self, nlp):
-        # Simple case with topk = 1, no batching
-        a = nlp(question='What is the name of the company I\'m working for ?', context='I\'m working for Huggingface.')
-        self.check_answer_structure(a, 1, 1)
-
-        # Simple case with topk = 2, no batching
-        a = nlp(question='What is the name of the company I\'m working for ?', context='I\'m working for Huggingface.', topk=2)
-        self.check_answer_structure(a, 1, 2)
-
-        # Batch case with topk = 1
-        a = nlp(question=['What is the name of the company I\'m working for ?', 'Where is the company based ?'],
-                context=['I\'m working for Huggingface.', 'The company is based in New York and Paris'])
-        self.check_answer_structure(a, 2, 1)
-
-        # Batch case with topk = 2
-        a = nlp(question=['What is the name of the company I\'m working for ?', 'Where is the company based ?'],
-                context=['Where is the company based ?', 'The company is based in New York and Paris'], topk=2)
-        self.check_answer_structure(a, 2, 2)
-
-        # check for data keyword
-        a = nlp(data=nlp.create_sample(question='What is the name of the company I\'m working for ?', context='I\'m working for Huggingface.'))
-        self.check_answer_structure(a, 1, 1)
-
-        a = nlp(data=nlp.create_sample(question='What is the name of the company I\'m working for ?', context='I\'m working for Huggingface.'), topk=2)
-        self.check_answer_structure(a, 1, 2)
-
-        a = nlp(data=[
-            nlp.create_sample(question='What is the name of the company I\'m working for ?', context='I\'m working for Huggingface.'),
-            nlp.create_sample(question='I\'m working for Huggingface.', context='The company is based in New York and Paris'),
-        ])
-        self.check_answer_structure(a, 2, 1)
-
-        a = nlp(data=[
-            {'question': 'What is the name of the company I\'m working for ?', 'context': 'I\'m working for Huggingface.'},
-            {'question': 'Where is the company based ?', 'context': 'The company is based in New York and Paris'},
-        ])
-        self.check_answer_structure(a, 2, 1)
-
-        # X keywords
-        a = nlp(X=nlp.create_sample(
-            question='Where is the company based ?', context='The company is based in New York and Paris'
-        ))
-        self.check_answer_structure(a, 1, 1)
-
-        a = nlp(X=[
-            {'question': 'What is the name of the company I\'m working for ?', 'context': 'I\'m working for Huggingface.'},
-            {'question': 'Where is the company based ?', 'context': 'The company is based in New York and Paris'},
-        ], topk=2)
-        self.check_answer_structure(a, 2, 2)
-
-    @patch('transformers.pipelines.is_torch_available', return_value=False)
-    def test_tf_models(self, is_torch_available):
-        from transformers import pipeline
-        for model in QA_FINETUNED_MODELS:
-            self.question_answering_pipeline(pipeline('question-answering', model))
-
-    @patch('transformers.pipelines.is_tf_available', return_value=False)
-    @patch('transformers.tokenization_utils.is_tf_available', return_value=False)
-    def test_torch_models(self, is_tf_available, _):
-        from transformers import pipeline
-        for model in QA_FINETUNED_MODELS:
-            self.question_answering_pipeline(pipeline('question-answering', model))
+@require_tf
+def tf_pipeline(*args, **kwargs):
+    return pipeline(**kwargs)
 
 
-class AutoPipelineTest(unittest.TestCase):
-    @patch('transformers.pipelines.is_torch_available', return_value=False)
-    def test_tf_qa(self, is_torch_available):
-        from transformers import pipeline
-        from transformers.pipelines import QuestionAnsweringPipeline
-        from transformers.modeling_tf_utils import TFPreTrainedModel
-        for model in QA_FINETUNED_MODELS:
-            nlp = pipeline('question-answering', model)
-            self.assertIsInstance(nlp, QuestionAnsweringPipeline)
-            self.assertIsInstance(nlp.model, TFPreTrainedModel)
+@require_torch
+def torch_pipeline(*args, **kwargs):
+    return pipeline(**kwargs)
 
-    @patch('transformers.pipelines.is_tf_available', return_value=False)
-    def test_torch_qa(self, is_tf_available):
-        from transformers import pipeline
-        from transformers.pipelines import QuestionAnsweringPipeline
-        from transformers.modeling_utils import PreTrainedModel
-        for model in QA_FINETUNED_MODELS:
-            nlp = pipeline('question-answering', model)
-            self.assertIsInstance(nlp, QuestionAnsweringPipeline)
-            self.assertIsInstance(nlp.model, PreTrainedModel)
+
+class MonoColumnInputTestCase(unittest.TestCase):
+    def _test_mono_column_pipeline(self, nlp, valid_inputs: list, invalid_inputs: list, output_keys: Iterable[str]):
+        self.assertIsNotNone(nlp)
+
+        mono_result = nlp(valid_inputs[0])
+        self.assertIsInstance(mono_result, list)
+        self.assertIsInstance(mono_result[0], (dict, list))
+
+        if isinstance(mono_result[0], list):
+            mono_result = mono_result[0]
+
+        for key in output_keys:
+            self.assertIn(key, mono_result[0])
+
+        multi_result = nlp(valid_inputs)
+        self.assertIsInstance(multi_result, list)
+        self.assertIsInstance(multi_result[0], (dict, list))
+
+        if isinstance(multi_result[0], list):
+            multi_result = multi_result[0]
+
+        for result in multi_result:
+            for key in output_keys:
+                self.assertIn(key, result)
+
+        self.assertRaises(Exception, nlp, invalid_inputs)
+
+    def test_ner(self):
+        mandatory_keys = {'entity', 'word', 'score'}
+        valid_inputs = ['HuggingFace is solving NLP one commit at a time.', 'HuggingFace is based in New-York & Paris']
+        invalid_inputs = [None]
+        for tokenizer, model, config in NER_FINETUNED_MODELS:
+            with patch('transformers.pipelines.is_torch_available', return_value=False):
+                nlp = tf_pipeline(task='ner', model=model, config=config, tokenizer=tokenizer)
+                self._test_mono_column_pipeline(nlp, valid_inputs, invalid_inputs, mandatory_keys)
+
+            with patch('transformers.pipelines.is_tf_available', return_value=False):
+                nlp = torch_pipeline(task='ner', model=model, config=config, tokenizer=tokenizer)
+                self._test_mono_column_pipeline(nlp, valid_inputs, invalid_inputs, mandatory_keys)
+
+    def test_sentiment_analysis(self):
+        mandatory_keys = {'label'}
+        valid_inputs = ['HuggingFace is solving NLP one commit at a time.', 'HuggingFace is based in New-York & Paris']
+        invalid_inputs = [None]
+        for tokenizer, model, config in TEXT_CLASSIF_FINETUNED_MODELS:
+            with patch('transformers.pipelines.is_torch_available', return_value=False):
+                nlp = tf_pipeline(task='sentiment-analysis', model=model, config=config, tokenizer=tokenizer)
+                self._test_mono_column_pipeline(nlp, valid_inputs, invalid_inputs, mandatory_keys)
+
+            with patch('transformers.pipelines.is_tf_available', return_value=False):
+                nlp = torch_pipeline(task='sentiment-analysis', model=model, config=config, tokenizer=tokenizer)
+                self._test_mono_column_pipeline(nlp, valid_inputs, invalid_inputs, mandatory_keys)
+
+    def test_features_extraction(self):
+        valid_inputs = ['HuggingFace is solving NLP one commit at a time.', 'HuggingFace is based in New-York & Paris']
+        invalid_inputs = [None]
+        for tokenizer, model, config in FEATURE_EXTRACT_FINETUNED_MODELS:
+            with patch('transformers.pipelines.is_torch_available', return_value=False):
+                nlp = tf_pipeline(task='sentiment-analysis', model=model, config=config, tokenizer=tokenizer)
+                self._test_mono_column_pipeline(nlp, valid_inputs, invalid_inputs, {})
+
+            with patch('transformers.pipelines.is_tf_available', return_value=False):
+                nlp = torch_pipeline(task='sentiment-analysis', model=model, config=config, tokenizer=tokenizer)
+                self._test_mono_column_pipeline(nlp, valid_inputs, invalid_inputs, {})
+
+
+class MultiColumnInputTestCase(unittest.TestCase):
+    def _test_multicolumn_pipeline(self, nlp, valid_inputs: list, invalid_inputs: list, output_keys: Iterable[str]):
+        self.assertIsNotNone(nlp)
+
+        mono_result = nlp(valid_inputs[0])
+        self.assertIsInstance(mono_result, dict)
+
+        for key in output_keys:
+            self.assertIn(key, mono_result)
+
+        multi_result = nlp(valid_inputs)
+        self.assertIsInstance(multi_result, list)
+        self.assertIsInstance(multi_result[0], dict)
+
+        for result in multi_result:
+            for key in output_keys:
+                self.assertIn(key, result)
+
+        self.assertRaises(Exception, nlp, invalid_inputs[0])
+        self.assertRaises(Exception, nlp, invalid_inputs)
+
+    def test_question_answering(self):
+        mandatory_output_keys = {'score', 'answer', 'start', 'end'}
+        valid_samples = [
+            {'question': 'Where was HuggingFace founded ?', 'context': 'HuggingFace was founded in Paris.'},
+            {
+                'question': 'In what field is HuggingFace working ?',
+                'context': 'HuggingFace is a startup based in New-York founded in Paris which is trying to solve NLP.'
+            }
+        ]
+        invalid_samples = [
+            {'question': '', 'context': 'This is a test to try empty question edge case'},
+            {'question': None, 'context': 'This is a test to try empty question edge case'},
+            {'question': 'What is does with empty context ?', 'context': ''},
+            {'question': 'What is does with empty context ?', 'context': None},
+        ]
+
+        for tokenizer, model, config in QA_FINETUNED_MODELS:
+
+            # Test for Tensorflow
+            with patch('transformers.pipelines.is_torch_available', return_value=False):
+                nlp = pipeline(task='question-answering', model=model, config=config, tokenizer=tokenizer)
+                self._test_multicolumn_pipeline(nlp, valid_samples, invalid_samples, mandatory_output_keys)
+
+            # Test for PyTorch
+            with patch('transformers.pipelines.is_tf_available', return_value=False):
+                nlp = pipeline(task='question-answering', model=model, config=config, tokenizer=tokenizer)
+                self._test_multicolumn_pipeline(nlp, valid_samples, invalid_samples, mandatory_output_keys)
 
 
 if __name__ == '__main__':

From ca6bdb28f64763ccad6c5b0aef36049c797f1ef7 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 20 Dec 2019 12:10:40 +0100
Subject: [PATCH 462/505] fix pipelines and rename model_card => modelcard

---
 transformers/__init__.py                     |   2 +-
 transformers/file_utils.py                   |   4 +-
 transformers/{model_card.py => modelcard.py} |  55 ++++-----
 transformers/pipelines.py                    | 112 ++++++++++++-------
 transformers/tests/model_card_test.py        |  28 ++---
 5 files changed, 118 insertions(+), 83 deletions(-)
 rename transformers/{model_card.py => modelcard.py} (83%)

diff --git a/transformers/__init__.py b/transformers/__init__.py
index 80f140b31e..73a7f5d862 100755
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -35,7 +35,7 @@ if is_sklearn_available():
     from .data import glue_compute_metrics, xnli_compute_metrics
 
 # Model Cards
-from .model_card import ModelCard
+from .modelcard import ModelCard
 
 # Tokenizers
 from .tokenization_utils import (PreTrainedTokenizer)
diff --git a/transformers/file_utils.py b/transformers/file_utils.py
index 8a1c4db201..47fa588815 100644
--- a/transformers/file_utils.py
+++ b/transformers/file_utils.py
@@ -81,7 +81,7 @@ WEIGHTS_NAME = "pytorch_model.bin"
 TF2_WEIGHTS_NAME = 'tf_model.h5'
 TF_WEIGHTS_NAME = 'model.ckpt'
 CONFIG_NAME = "config.json"
-MODEL_CARD_NAME = "model_card.json"
+MODEL_CARD_NAME = "modelcard.json"
 
 DUMMY_INPUTS = [[7, 6, 0, 0, 1], [1, 2, 3, 0, 0], [0, 0, 0, 4, 5]]
 DUMMY_MASK = [[1, 1, 1, 1, 1], [1, 1, 1, 0, 0], [0, 0, 0, 1, 1]]
@@ -339,7 +339,7 @@ def get_from_cache(url, cache_dir=None, force_download=False, proxies=None, etag
         temp_file_manager = tempfile.NamedTemporaryFile
         resume_size = 0
 
-    if not os.path.exists(cache_path) or force_download:
+    if etag is not None and (not os.path.exists(cache_path) or force_download):
         # Download to temporary file, then copy to cache dir once finished.
         # Otherwise you get corrupt cache entries if the download gets interrupted.
         with temp_file_manager() as temp_file:
diff --git a/transformers/model_card.py b/transformers/modelcard.py
similarity index 83%
rename from transformers/model_card.py
rename to transformers/modelcard.py
index baec7e8622..4a879235ae 100644
--- a/transformers/model_card.py
+++ b/transformers/modelcard.py
@@ -25,7 +25,8 @@ from io import open
 
 from .configuration_auto import ALL_PRETRAINED_CONFIG_ARCHIVE_MAP
 
-from .file_utils import CONFIG_NAME, MODEL_CARD_NAME, cached_path, is_remote_url, hf_bucket_url
+from .file_utils import CONFIG_NAME, MODEL_CARD_NAME, WEIGHTS_NAME, TF2_WEIGHTS_NAME, \
+                        cached_path, is_remote_url, hf_bucket_url
 
 
 logger = logging.getLogger(__name__)
@@ -89,7 +90,7 @@ class ModelCard(object):
                 - a string with the `shortcut name` of a pre-trained model card to load from cache or download, e.g.: ``bert-base-uncased``.
                 - a string with the `identifier name` of a pre-trained model card that was user-uploaded to our S3, e.g.: ``dbmdz/bert-base-german-cased``.
                 - a path to a `directory` containing a mode card file saved using the :func:`~transformers.ModelCard.save_pretrained` method, e.g.: ``./my_model_directory/``.
-                - a path or url to a saved model card JSON `file`, e.g.: ``./my_model_directory/model_card.json``.
+                - a path or url to a saved model card JSON `file`, e.g.: ``./my_model_directory/modelcard.json``.
 
             cache_dir: (`optional`) string:
                 Path to a directory in which a downloaded pre-trained model
@@ -100,16 +101,14 @@ class ModelCard(object):
                 - The values in kwargs of any keys which are model card attributes will be used to override the loaded values.
                 - Behavior concerning key/value pairs whose keys are *not* model card attributes is controlled by the `return_unused_kwargs` keyword parameter.
 
-            force_download: (`optional`) boolean, default False:
-                Force to (re-)download the model card file and override the cached version if it exists.
-
-            resume_download: (`optional`) boolean, default False:
-                Do not delete incompletely recieved file. Attempt to resume the download if such a file exists.
-
             proxies: (`optional`) dict, default None:
                 A dictionary of proxy servers to use by protocol or endpoint, e.g.: {'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.
                 The proxies are used on each request.
 
+            find_from_standard_name: (`optional`) boolean, default True:
+                If the pretrained_model_name_or_path ends with our standard model or config filenames, replace them with our standard modelcard filename.
+                Can be used to directly feed a model/config url and access the colocated modelcard.
+
             return_unused_kwargs: (`optional`) bool:
 
                 - If False, then this function returns just the final model card object.
@@ -117,22 +116,21 @@ class ModelCard(object):
 
         Examples::
 
-            model_card = ModelCard.from_pretrained('bert-base-uncased')    # Download model card from S3 and cache.
-            model_card = ModelCard.from_pretrained('./test/saved_model/')  # E.g. model card was saved using `save_pretrained('./test/saved_model/')`
-            model_card = ModelCard.from_pretrained('./test/saved_model/model_card.json')
-            model_card = ModelCard.from_pretrained('bert-base-uncased', output_attention=True, foo=False)
+            modelcard = ModelCard.from_pretrained('bert-base-uncased')    # Download model card from S3 and cache.
+            modelcard = ModelCard.from_pretrained('./test/saved_model/')  # E.g. model card was saved using `save_pretrained('./test/saved_model/')`
+            modelcard = ModelCard.from_pretrained('./test/saved_model/modelcard.json')
+            modelcard = ModelCard.from_pretrained('bert-base-uncased', output_attention=True, foo=False)
 
         """
         cache_dir = kwargs.pop('cache_dir', None)
-        force_download = kwargs.pop('force_download', False)
-        resume_download = kwargs.pop('resume_download', False)
         proxies = kwargs.pop('proxies', None)
+        find_from_standard_name = kwargs.pop('find_from_standard_name', True)
         return_unused_kwargs = kwargs.pop('return_unused_kwargs', False)
 
         if pretrained_model_name_or_path in ALL_PRETRAINED_CONFIG_ARCHIVE_MAP:
-            # For simplicity we use the same pretrained url than the configuration files but with a different suffix (model_card.json)
+            # For simplicity we use the same pretrained url than the configuration files
+            # but with a different suffix (modelcard.json). This suffix is replaced below.
             model_card_file = ALL_PRETRAINED_CONFIG_ARCHIVE_MAP[pretrained_model_name_or_path]
-            model_card_file = model_card_file.replace(CONFIG_NAME, MODEL_CARD_NAME)
         elif os.path.isdir(pretrained_model_name_or_path):
             model_card_file = os.path.join(pretrained_model_name_or_path, MODEL_CARD_NAME)
         elif os.path.isfile(pretrained_model_name_or_path) or is_remote_url(pretrained_model_name_or_path):
@@ -140,17 +138,22 @@ class ModelCard(object):
         else:
             model_card_file = hf_bucket_url(pretrained_model_name_or_path, postfix=MODEL_CARD_NAME)
 
+        if find_from_standard_name or pretrained_model_name_or_path in ALL_PRETRAINED_CONFIG_ARCHIVE_MAP:
+            model_card_file = model_card_file.replace(CONFIG_NAME, MODEL_CARD_NAME)
+            model_card_file = model_card_file.replace(WEIGHTS_NAME, MODEL_CARD_NAME)
+            model_card_file = model_card_file.replace(TF2_WEIGHTS_NAME, MODEL_CARD_NAME)
+
         try:
             # Load from URL or cache if already cached
-            resolved_model_card_file = cached_path(model_card_file, cache_dir=cache_dir, force_download=force_download,
-                                               proxies=proxies, resume_download=resume_download)
+            resolved_model_card_file = cached_path(model_card_file, cache_dir=cache_dir, force_download=True,
+                                               proxies=proxies, resume_download=False)
             if resolved_model_card_file == model_card_file:
                 logger.info("loading model card file {}".format(model_card_file))
             else:
                 logger.info("loading model card file {} from cache at {}".format(
                     model_card_file, resolved_model_card_file))
             # Load model card
-            model_card = cls.from_json_file(resolved_model_card_file)
+            modelcard = cls.from_json_file(resolved_model_card_file)
 
         except EnvironmentError:
             if pretrained_model_name_or_path in ALL_PRETRAINED_CONFIG_ARCHIVE_MAP:
@@ -166,7 +169,7 @@ class ModelCard(object):
             logger.warning("Creating an empty model card.")
 
             # We fall back on creating an empty model card
-            model_card = cls()
+            modelcard = cls()
 
         except json.JSONDecodeError:
             logger.warning("Couldn't reach server at '{}' to download model card file or "
@@ -175,22 +178,22 @@ class ModelCard(object):
             logger.warning("Creating an empty model card.")
 
             # We fall back on creating an empty model card
-            model_card = cls()
+            modelcard = cls()
 
         # Update model card with kwargs if needed
         to_remove = []
         for key, value in kwargs.items():
-            if hasattr(model_card, key):
-                setattr(model_card, key, value)
+            if hasattr(modelcard, key):
+                setattr(modelcard, key, value)
                 to_remove.append(key)
         for key in to_remove:
             kwargs.pop(key, None)
 
-        logger.info("Model card: %s", str(model_card))
+        logger.info("Model card: %s", str(modelcard))
         if return_unused_kwargs:
-            return model_card, kwargs
+            return modelcard, kwargs
         else:
-            return model_card
+            return modelcard
 
     @classmethod
     def from_dict(cls, json_object):
diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 4dde62cbe5..be2b1db126 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -18,6 +18,8 @@ import csv
 import json
 import os
 import pickle
+import logging
+import six
 from abc import ABC, abstractmethod
 from contextlib import contextmanager
 from itertools import groupby
@@ -26,8 +28,12 @@ from typing import Union, Optional, Tuple, List, Dict
 
 import numpy as np
 
-from transformers import AutoConfig, AutoTokenizer, PreTrainedTokenizer, PretrainedConfig, \
-    SquadExample, squad_convert_examples_to_features, is_tf_available, is_torch_available, logger, BasicTokenizer
+from transformers import (AutoConfig, AutoTokenizer, PreTrainedTokenizer,
+                          PretrainedConfig, ModelCard, SquadExample,
+                          squad_convert_examples_to_features, is_tf_available,
+                          is_torch_available, BasicTokenizer,
+                          ALL_PRETRAINED_MODEL_ARCHIVE_MAP,
+                          ALL_PRETRAINED_CONFIG_ARCHIVE_MAP)
 
 if is_tf_available():
     import tensorflow as tf
@@ -40,6 +46,8 @@ if is_torch_available():
         AutoModelForQuestionAnswering, AutoModelForTokenClassification
 
 
+logger = logging.getLogger(__name__)
+
 class ArgumentHandler(ABC):
     """
     Base interface for handling varargs for each Pipeline
@@ -271,11 +279,13 @@ class Pipeline(_ScikitCompat):
         nlp = QuestionAnsweringPipeline(model=AutoModel.from_pretrained('...'), tokenizer='...')
     """
     def __init__(self, model, tokenizer: PreTrainedTokenizer = None,
+                 modelcard: ModelCard = None,
                  args_parser: ArgumentHandler = None, device: int = -1,
                  binary_output: bool = False):
 
         self.model = model
         self.tokenizer = tokenizer
+        self.modelcard = modelcard
         self.device = device
         self.binary_output = binary_output
         self._args_parser = args_parser or DefaultArgumentHandler()
@@ -294,6 +304,7 @@ class Pipeline(_ScikitCompat):
 
         self.model.save_pretrained(save_directory)
         self.tokenizer.save_pretrained(save_directory)
+        self.modelcard.save_pretrained(save_directory)
 
     def transform(self, X):
         """
@@ -393,9 +404,10 @@ class FeatureExtractionPipeline(Pipeline):
 
     def __init__(self, model,
                  tokenizer: PreTrainedTokenizer = None,
+                 modelcard: ModelCard = None,
                  args_parser: ArgumentHandler = None,
                  device: int = -1):
-        super().__init__(model, tokenizer, args_parser, device, binary_output=True)
+        super().__init__(model, tokenizer, modelcard, args_parser, device, binary_output=True)
 
     def __call__(self, *args, **kwargs):
         return super().__call__(*args, **kwargs).tolist()
@@ -418,9 +430,10 @@ class NerPipeline(Pipeline):
     """
 
     def __init__(self, model, tokenizer: PreTrainedTokenizer = None,
+                 modelcard: ModelCard = None,
                  args_parser: ArgumentHandler = None, device: int = -1,
                  binary_output: bool = False):
-        super().__init__(model, tokenizer, args_parser, device, binary_output)
+        super().__init__(model, tokenizer, modelcard, args_parser, device, binary_output)
 
         self._basic_tokenizer = BasicTokenizer(do_lower_case=False)
 
@@ -554,8 +567,10 @@ class QuestionAnsweringPipeline(Pipeline):
         else:
             return SquadExample(None, question, context, None, None, None)
 
-    def __init__(self, model, tokenizer: Optional[PreTrainedTokenizer], device: int = -1, **kwargs):
-        super().__init__(model, tokenizer, args_parser=QuestionAnsweringArgumentHandler(),
+    def __init__(self, model, tokenizer: Optional[PreTrainedTokenizer],
+                 modelcard: Optional[ModelCard],
+                 device: int = -1, **kwargs):
+        super().__init__(model, tokenizer, modelcard, args_parser=QuestionAnsweringArgumentHandler(),
                          device=device, **kwargs)
 
     def __call__(self, *texts, **kwargs):
@@ -725,7 +740,7 @@ SUPPORTED_TASKS = {
         'default': {
             'model': 'distilbert-base-uncased',
             'config': None,
-            'tokenizer': 'bert-base-uncased'
+            'tokenizer': 'distilbert-base-uncased'
         }
     },
     'sentiment-analysis': {
@@ -735,7 +750,7 @@ SUPPORTED_TASKS = {
         'default': {
             'model': 'https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-finetuned-sst-2-english-pytorch_model.bin',
             'config': 'https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-finetuned-sst-2-english-config.json',
-            'tokenizer': 'bert-base-uncased'
+            'tokenizer': 'distilbert-base-uncased'
         }
     },
     'ner': {
@@ -745,7 +760,7 @@ SUPPORTED_TASKS = {
         'default': {
             'model': 'https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-cased-finetuned-conll03-english-pytorch_model.bin',
             'config': 'https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-cased-finetuned-conll03-english-config.json',
-            'tokenizer': 'bert-base-cased'
+            'tokenizer': 'bert-large-cased'
         }
     },
     'question-answering': {
@@ -755,7 +770,7 @@ SUPPORTED_TASKS = {
         'default': {
             'model': 'distilbert-base-uncased-distilled-squad',
             'config': None,
-            'tokenizer': 'bert-base-uncased'
+            'tokenizer': 'distilbert-base-uncased'
         }
     }
 }
@@ -763,7 +778,9 @@ SUPPORTED_TASKS = {
 
 def pipeline(task: str, model: Optional = None,
              config: Optional[Union[str, PretrainedConfig]] = None,
-             tokenizer: Optional[Union[str, PreTrainedTokenizer]] = None, **kwargs) -> Pipeline:
+             tokenizer: Optional[Union[str, PreTrainedTokenizer]] = None,
+             modelcard: Optional[Union[str, ModelCard]] = None,
+             **kwargs) -> Pipeline:
     """
     Utility factory method to build a pipeline.
     Pipeline are made of:
@@ -777,48 +794,63 @@ def pipeline(task: str, model: Optional = None,
         pipeline('ner', model=AutoModel.from_pretrained(...), tokenizer=AutoTokenizer.from_pretrained(...)
         pipeline('ner', model='https://...pytorch-model.bin', config='https://...config.json', tokenizer='bert-base-cased')
     """
-    # Try to infer tokenizer from model name (if provided as str)
-    if tokenizer is None:
-        if model is not None and not isinstance(model, str):
-            # Impossible to guest what is the right tokenizer here
-            raise Exception('Tokenizer cannot be None if provided model is a PreTrainedModel instance')
-        else:
-            tokenizer = model
-
     # Retrieve the task
     if task not in SUPPORTED_TASKS:
         raise KeyError("Unknown task {}, available tasks are {}".format(task, list(SUPPORTED_TASKS.keys())))
 
-    targeted_task = SUPPORTED_TASKS[task]
-    task, allocator = targeted_task['impl'], targeted_task['tf'] if is_tf_available() else targeted_task['pt']
+    pipeline_framework = 'tf' if is_tf_available() else ('pt' if is_torch_available() else None)
+    if pipeline_framework is None:
+        raise ImportError("At least one of TensorFlow 2.0 or PyTorch should be installed. "
+                          "To install TensorFlow 2.0, read the instructions at https://www.tensorflow.org/install/ "
+                          "To install PyTorch, read the instructions at https://pytorch.org/.")
 
-    # Handling for default model for the task
+
+    targeted_task = SUPPORTED_TASKS[task]
+    task, model_class = targeted_task['impl'], targeted_task[pipeline_framework]
+
+    # Use default model/config/tokenizer for the task if no model is provided
     if model is None:
         model, config, tokenizer = tuple(targeted_task['default'].values())
 
-    # Allocate tokenizer
-    tokenizer = tokenizer if isinstance(tokenizer, PreTrainedTokenizer) else AutoTokenizer.from_pretrained(tokenizer)
+    # Try to infer tokenizer from model or config name (if provided as str)
+    if tokenizer is None:
+        if isinstance(model, str) and model in ALL_PRETRAINED_MODEL_ARCHIVE_MAP:
+            tokenizer = model
+        elif isinstance(config, str) and model in ALL_PRETRAINED_CONFIG_ARCHIVE_MAP:
+            tokenizer = config
+        else:
+            # Impossible to guest what is the right tokenizer here
+            raise Exception("Impossible to guess which tokenizer to use. "
+                            "Please provided a PretrainedTokenizer class or a path/url/shortcut name to a pretrained tokenizer.")
 
-    # Special handling for model conversion
-    if isinstance(model, str):
-        from_tf = model.endswith('.h5') and not is_tf_available()
-        from_pt = model.endswith('.bin') and not is_torch_available()
+    # Try to infer modelcard from model or config name (if provided as str)
+    if modelcard is None:
+        # Try to fallback on one of the provided string for model or config (will replace the suffix)
+        if isinstance(model, str):
+            modelcard = model
+        elif isinstance(config, str):
+            modelcard = config
 
-        if from_tf:
-            logger.warning('Model might be a TensorFlow model (ending with `.h5`) but TensorFlow is not available. '
-                           'Trying to load the model with PyTorch.')
-        elif from_pt:
-            logger.warning('Model might be a PyTorch model (ending with `.bin`) but PyTorch is not available. '
-                           'Trying to load the model with Tensorflow.')
-    else:
-        from_tf = from_pt = False
+    # Instantiate tokenizer if needed
+    if isinstance(tokenizer, six.string_types):
+        tokenizer = AutoTokenizer.from_pretrained(tokenizer)
 
+    # Instantiate config if needed
     if isinstance(config, str):
         config = AutoConfig.from_pretrained(config)
 
+    # Instantiate model if needed
     if isinstance(model, str):
-        if allocator.__name__.startswith('TF'):
-            model = allocator.from_pretrained(model, config=config, from_pt=from_pt)
-        else:
-            model = allocator.from_pretrained(model, config=config, from_tf=from_tf)
+        # Handle transparent TF/PT model conversion
+        model_kwargs = {}
+        if pipeline_framework == 'pt' and model.endswith('.h5'):
+            model_kwargs['from_tf'] = True
+            logger.warning('Model might be a TensorFlow model (ending with `.h5`) but TensorFlow is not available. '
+                           'Trying to load the model with PyTorch.')
+        elif pipeline_framework == 'tf' and model.endswith('.bin'):
+            model_kwargs['from_pt'] = True
+            logger.warning('Model might be a PyTorch model (ending with `.bin`) but PyTorch is not available. '
+                           'Trying to load the model with Tensorflow.')
+        model = model_class.from_pretrained(model, config=config, **model_kwargs)
+
     return task(model, tokenizer, **kwargs)
diff --git a/transformers/tests/model_card_test.py b/transformers/tests/model_card_test.py
index e75716f0aa..b293b5726a 100644
--- a/transformers/tests/model_card_test.py
+++ b/transformers/tests/model_card_test.py
@@ -18,7 +18,7 @@ import os
 import json
 import unittest
 
-from transformers.model_card import ModelCard
+from transformers.modelcard import ModelCard
 from .tokenization_tests_commons import TemporaryDirectory
 
 class ModelCardTester(unittest.TestCase):
@@ -49,20 +49,20 @@ class ModelCardTester(unittest.TestCase):
                             }
 
     def test_model_card_common_properties(self):
-        model_card = ModelCard.from_dict(self.inputs_dict)
-        self.assertTrue(hasattr(model_card, 'model_details'))
-        self.assertTrue(hasattr(model_card, 'intended_use'))
-        self.assertTrue(hasattr(model_card, 'factors'))
-        self.assertTrue(hasattr(model_card, 'metrics'))
-        self.assertTrue(hasattr(model_card, 'evaluation_data'))
-        self.assertTrue(hasattr(model_card, 'training_data'))
-        self.assertTrue(hasattr(model_card, 'quantitative_analyses'))
-        self.assertTrue(hasattr(model_card, 'ethical_considerations'))
-        self.assertTrue(hasattr(model_card, 'caveats_and_recommendations'))
+        modelcard = ModelCard.from_dict(self.inputs_dict)
+        self.assertTrue(hasattr(modelcard, 'model_details'))
+        self.assertTrue(hasattr(modelcard, 'intended_use'))
+        self.assertTrue(hasattr(modelcard, 'factors'))
+        self.assertTrue(hasattr(modelcard, 'metrics'))
+        self.assertTrue(hasattr(modelcard, 'evaluation_data'))
+        self.assertTrue(hasattr(modelcard, 'training_data'))
+        self.assertTrue(hasattr(modelcard, 'quantitative_analyses'))
+        self.assertTrue(hasattr(modelcard, 'ethical_considerations'))
+        self.assertTrue(hasattr(modelcard, 'caveats_and_recommendations'))
 
     def test_model_card_to_json_string(self):
-        model_card = ModelCard.from_dict(self.inputs_dict)
-        obj = json.loads(model_card.to_json_string())
+        modelcard = ModelCard.from_dict(self.inputs_dict)
+        obj = json.loads(modelcard.to_json_string())
         for key, value in self.inputs_dict.items():
             self.assertEqual(obj[key], value)
 
@@ -70,7 +70,7 @@ class ModelCardTester(unittest.TestCase):
         model_card_first = ModelCard.from_dict(self.inputs_dict)
 
         with TemporaryDirectory() as tmpdirname:
-            filename = os.path.join(tmpdirname, u"model_card.json")
+            filename = os.path.join(tmpdirname, u"modelcard.json")
             model_card_first.to_json_file(filename)
             model_card_second = ModelCard.from_json_file(filename)
 

From 1fa93ca1eaa249321ef39994e9f022d0799034a3 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 20 Dec 2019 12:34:19 +0100
Subject: [PATCH 463/505] Clean up framework handling

---
 transformers/pipelines.py | 85 ++++++++++++++++++++++++++-------------
 1 file changed, 58 insertions(+), 27 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index be2b1db126..1c56033f7c 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -48,6 +48,19 @@ if is_torch_available():
 
 logger = logging.getLogger(__name__)
 
+def get_framework(model=None):
+    if is_tf_available() and is_torch_available() and model is not None and not isinstance(model, str):
+        # Both framework are available but the use supplied a model class instance.
+        # Try to guess which framework to use from the model classname
+        framework = 'tf' if model.__class__.__name__.startswith('TF') else 'pt'
+    else:
+        framework = 'tf' if is_tf_available() else ('pt' if is_torch_available() else None)
+    if framework is None:
+        raise ImportError("At least one of TensorFlow 2.0 or PyTorch should be installed. "
+                          "To install TensorFlow 2.0, read the instructions at https://www.tensorflow.org/install/ "
+                          "To install PyTorch, read the instructions at https://pytorch.org/.")
+    return framework
+
 class ArgumentHandler(ABC):
     """
     Base interface for handling varargs for each Pipeline
@@ -279,19 +292,23 @@ class Pipeline(_ScikitCompat):
         nlp = QuestionAnsweringPipeline(model=AutoModel.from_pretrained('...'), tokenizer='...')
     """
     def __init__(self, model, tokenizer: PreTrainedTokenizer = None,
-                 modelcard: ModelCard = None,
+                 modelcard: ModelCard = None, framework: Optional[str] = None,
                  args_parser: ArgumentHandler = None, device: int = -1,
                  binary_output: bool = False):
 
+        if framework is None:
+            framework = get_framework()
+
         self.model = model
         self.tokenizer = tokenizer
         self.modelcard = modelcard
+        self.framework = framework
         self.device = device
         self.binary_output = binary_output
         self._args_parser = args_parser or DefaultArgumentHandler()
 
         # Special handling
-        if self.device >= 0 and not is_tf_available():
+        if self.device >= 0 and self.framework == 'pt':
             self.model = self.model.to('cuda:{}'.format(self.device))
 
     def save_pretrained(self, save_directory):
@@ -332,7 +349,7 @@ class Pipeline(_ScikitCompat):
         Returns:
             Context manager
         """
-        if is_tf_available():
+        if self.framework == 'tf':
             with tf.device('/CPU:0' if self.device == -1 else '/device:GPU:{}'.format(self.device)):
                 yield
         else:
@@ -371,7 +388,7 @@ class Pipeline(_ScikitCompat):
         with self.device_placement():
             inputs = self.tokenizer.batch_encode_plus(
                 inputs, add_special_tokens=True,
-                return_tensors='tf' if is_tf_available() else 'pt',
+                return_tensors=self.framework,
                 max_length=self.tokenizer.max_len
             )
 
@@ -387,7 +404,7 @@ class Pipeline(_ScikitCompat):
         Returns:
             Numpy array
         """
-        if is_tf_available():
+        if self.framework == 'tf':
             # TODO trace model
             predictions = self.model(inputs, training=False)[0]
         else:
@@ -405,9 +422,16 @@ class FeatureExtractionPipeline(Pipeline):
     def __init__(self, model,
                  tokenizer: PreTrainedTokenizer = None,
                  modelcard: ModelCard = None,
+                 framework: Optional[str] = None,
                  args_parser: ArgumentHandler = None,
                  device: int = -1):
-        super().__init__(model, tokenizer, modelcard, args_parser, device, binary_output=True)
+        super().__init__(model=model,
+                         tokenizer=tokenizer,
+                         modelcard=modelcard,
+                         framework=framework,
+                         args_parser=args_parser,
+                         device=device,
+                         binary_output=True)
 
     def __call__(self, *args, **kwargs):
         return super().__call__(*args, **kwargs).tolist()
@@ -430,10 +454,16 @@ class NerPipeline(Pipeline):
     """
 
     def __init__(self, model, tokenizer: PreTrainedTokenizer = None,
-                 modelcard: ModelCard = None,
+                 modelcard: ModelCard = None, framework: Optional[str] = None,
                  args_parser: ArgumentHandler = None, device: int = -1,
                  binary_output: bool = False):
-        super().__init__(model, tokenizer, modelcard, args_parser, device, binary_output)
+        super().__init__(model=model,
+                         tokenizer=tokenizer,
+                         modelcard=modelcard,
+                         framework=framework,
+                         args_parser=args_parser,
+                         device=device,
+                         binary_output=binary_output)
 
         self._basic_tokenizer = BasicTokenizer(do_lower_case=False)
 
@@ -452,12 +482,12 @@ class NerPipeline(Pipeline):
 
                 tokens = self.tokenizer.encode_plus(
                     sentence, return_attention_mask=False,
-                    return_tensors='tf' if is_tf_available() else 'pt',
+                    return_tensors=self.framework,
                     max_length=self.tokenizer.max_len
                 )
 
                 # Forward
-                if is_tf_available():
+                if self.framework == 'tf':
                     entities = self.model(tokens)[0][0].numpy()
                 else:
                     with torch.no_grad():
@@ -549,6 +579,18 @@ class QuestionAnsweringPipeline(Pipeline):
     Question Answering pipeline using ModelForQuestionAnswering head.
     """
 
+    def __init__(self, model,
+                 tokenizer: Optional[PreTrainedTokenizer],
+                 modelcard: Optional[ModelCard],
+                 framework: Optional[str] = None,
+                 device: int = -1, **kwargs):
+        super().__init__(model=model,
+                         tokenizer=tokenizer,
+                         modelcard=modelcard,
+                         framework=framework,
+                         args_parser=QuestionAnsweringArgumentHandler(),
+                         device=device, **kwargs)
+
     @staticmethod
     def create_sample(question: Union[str, List[str]], context: Union[str, List[str]]) -> Union[SquadExample, List[SquadExample]]:
         """
@@ -567,12 +609,6 @@ class QuestionAnsweringPipeline(Pipeline):
         else:
             return SquadExample(None, question, context, None, None, None)
 
-    def __init__(self, model, tokenizer: Optional[PreTrainedTokenizer],
-                 modelcard: Optional[ModelCard],
-                 device: int = -1, **kwargs):
-        super().__init__(model, tokenizer, modelcard, args_parser=QuestionAnsweringArgumentHandler(),
-                         device=device, **kwargs)
-
     def __call__(self, *texts, **kwargs):
         """
         Args:
@@ -608,7 +644,7 @@ class QuestionAnsweringPipeline(Pipeline):
 
         # Manage tensor allocation on correct device
         with self.device_placement():
-            if is_tf_available():
+            if self.framework == 'tf':
                 fw_args = {k: tf.constant(v) for (k, v) in fw_args.items()}
                 start, end = self.model(fw_args)
                 start, end = start.numpy(), end.numpy()
@@ -798,15 +834,10 @@ def pipeline(task: str, model: Optional = None,
     if task not in SUPPORTED_TASKS:
         raise KeyError("Unknown task {}, available tasks are {}".format(task, list(SUPPORTED_TASKS.keys())))
 
-    pipeline_framework = 'tf' if is_tf_available() else ('pt' if is_torch_available() else None)
-    if pipeline_framework is None:
-        raise ImportError("At least one of TensorFlow 2.0 or PyTorch should be installed. "
-                          "To install TensorFlow 2.0, read the instructions at https://www.tensorflow.org/install/ "
-                          "To install PyTorch, read the instructions at https://pytorch.org/.")
-
+    framework = get_framework(model)
 
     targeted_task = SUPPORTED_TASKS[task]
-    task, model_class = targeted_task['impl'], targeted_task[pipeline_framework]
+    task, model_class = targeted_task['impl'], targeted_task[framework]
 
     # Use default model/config/tokenizer for the task if no model is provided
     if model is None:
@@ -843,14 +874,14 @@ def pipeline(task: str, model: Optional = None,
     if isinstance(model, str):
         # Handle transparent TF/PT model conversion
         model_kwargs = {}
-        if pipeline_framework == 'pt' and model.endswith('.h5'):
+        if framework == 'pt' and model.endswith('.h5'):
             model_kwargs['from_tf'] = True
             logger.warning('Model might be a TensorFlow model (ending with `.h5`) but TensorFlow is not available. '
                            'Trying to load the model with PyTorch.')
-        elif pipeline_framework == 'tf' and model.endswith('.bin'):
+        elif framework == 'tf' and model.endswith('.bin'):
             model_kwargs['from_pt'] = True
             logger.warning('Model might be a PyTorch model (ending with `.bin`) but PyTorch is not available. '
                            'Trying to load the model with Tensorflow.')
         model = model_class.from_pretrained(model, config=config, **model_kwargs)
 
-    return task(model, tokenizer, **kwargs)
+    return task(model=model, tokenizer=tokenizer, modelcard=modelcard, framework=framework, **kwargs)

From 825697cad4907595bbb76eb43d96962d7bd52117 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 20 Dec 2019 12:51:10 +0100
Subject: [PATCH 464/505] fix tests

---
 transformers/pipelines.py | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 1c56033f7c..8b5a14fc56 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -32,7 +32,6 @@ from transformers import (AutoConfig, AutoTokenizer, PreTrainedTokenizer,
                           PretrainedConfig, ModelCard, SquadExample,
                           squad_convert_examples_to_features, is_tf_available,
                           is_torch_available, BasicTokenizer,
-                          ALL_PRETRAINED_MODEL_ARCHIVE_MAP,
                           ALL_PRETRAINED_CONFIG_ARCHIVE_MAP)
 
 if is_tf_available():
@@ -845,9 +844,9 @@ def pipeline(task: str, model: Optional = None,
 
     # Try to infer tokenizer from model or config name (if provided as str)
     if tokenizer is None:
-        if isinstance(model, str) and model in ALL_PRETRAINED_MODEL_ARCHIVE_MAP:
+        if isinstance(model, str) and model in ALL_PRETRAINED_CONFIG_ARCHIVE_MAP:
             tokenizer = model
-        elif isinstance(config, str) and model in ALL_PRETRAINED_CONFIG_ARCHIVE_MAP:
+        elif isinstance(config, str) and config in ALL_PRETRAINED_CONFIG_ARCHIVE_MAP:
             tokenizer = config
         else:
             # Impossible to guest what is the right tokenizer here

From 01ffc65e9b4e74d0399212435da2d46c6edf2563 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 20 Dec 2019 13:16:23 +0100
Subject: [PATCH 465/505] update tests to remove unittest.patch

---
 transformers/pipelines.py            |  9 ++-
 transformers/tests/pipelines_test.py | 91 +++++++++++++++++-----------
 2 files changed, 60 insertions(+), 40 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 8b5a14fc56..efb1de92e1 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -48,16 +48,19 @@ if is_torch_available():
 logger = logging.getLogger(__name__)
 
 def get_framework(model=None):
+    """ Select framework (TensorFlow/PyTorch) to use.
+        If both frameworks are installed and no specific model is provided, defaults to using TensorFlow.
+    """
     if is_tf_available() and is_torch_available() and model is not None and not isinstance(model, str):
         # Both framework are available but the use supplied a model class instance.
         # Try to guess which framework to use from the model classname
         framework = 'tf' if model.__class__.__name__.startswith('TF') else 'pt'
-    else:
-        framework = 'tf' if is_tf_available() else ('pt' if is_torch_available() else None)
-    if framework is None:
+    elif not is_tf_available() and not is_torch_available():
         raise ImportError("At least one of TensorFlow 2.0 or PyTorch should be installed. "
                           "To install TensorFlow 2.0, read the instructions at https://www.tensorflow.org/install/ "
                           "To install PyTorch, read the instructions at https://pytorch.org/.")
+    else:
+        framework = 'tf' if is_tf_available() else 'pt'
     return framework
 
 class ArgumentHandler(ABC):
diff --git a/transformers/tests/pipelines_test.py b/transformers/tests/pipelines_test.py
index a8fe668221..14bf07ee30 100644
--- a/transformers/tests/pipelines_test.py
+++ b/transformers/tests/pipelines_test.py
@@ -1,5 +1,4 @@
 import unittest
-from unittest.mock import patch
 
 from typing import Iterable
 
@@ -35,16 +34,6 @@ TEXT_CLASSIF_FINETUNED_MODELS = {
 }
 
 
-@require_tf
-def tf_pipeline(*args, **kwargs):
-    return pipeline(**kwargs)
-
-
-@require_torch
-def torch_pipeline(*args, **kwargs):
-    return pipeline(**kwargs)
-
-
 class MonoColumnInputTestCase(unittest.TestCase):
     def _test_mono_column_pipeline(self, nlp, valid_inputs: list, invalid_inputs: list, output_keys: Iterable[str]):
         self.assertIsNotNone(nlp)
@@ -72,43 +61,57 @@ class MonoColumnInputTestCase(unittest.TestCase):
 
         self.assertRaises(Exception, nlp, invalid_inputs)
 
+    @require_torch
     def test_ner(self):
         mandatory_keys = {'entity', 'word', 'score'}
         valid_inputs = ['HuggingFace is solving NLP one commit at a time.', 'HuggingFace is based in New-York & Paris']
         invalid_inputs = [None]
         for tokenizer, model, config in NER_FINETUNED_MODELS:
-            with patch('transformers.pipelines.is_torch_available', return_value=False):
-                nlp = tf_pipeline(task='ner', model=model, config=config, tokenizer=tokenizer)
-                self._test_mono_column_pipeline(nlp, valid_inputs, invalid_inputs, mandatory_keys)
+            nlp = pipeline(task='ner', model=model, config=config, tokenizer=tokenizer)
+            self._test_mono_column_pipeline(nlp, valid_inputs, invalid_inputs, mandatory_keys)
 
-            with patch('transformers.pipelines.is_tf_available', return_value=False):
-                nlp = torch_pipeline(task='ner', model=model, config=config, tokenizer=tokenizer)
-                self._test_mono_column_pipeline(nlp, valid_inputs, invalid_inputs, mandatory_keys)
+    @require_tf
+    def test_tf_ner(self):
+        mandatory_keys = {'entity', 'word', 'score'}
+        valid_inputs = ['HuggingFace is solving NLP one commit at a time.', 'HuggingFace is based in New-York & Paris']
+        invalid_inputs = [None]
+        for tokenizer, model, config in NER_FINETUNED_MODELS:
+            nlp = pipeline(task='ner', model=model, config=config, tokenizer=tokenizer)
+            self._test_mono_column_pipeline(nlp, valid_inputs, invalid_inputs, mandatory_keys)
 
+    @require_torch
     def test_sentiment_analysis(self):
         mandatory_keys = {'label'}
         valid_inputs = ['HuggingFace is solving NLP one commit at a time.', 'HuggingFace is based in New-York & Paris']
         invalid_inputs = [None]
         for tokenizer, model, config in TEXT_CLASSIF_FINETUNED_MODELS:
-            with patch('transformers.pipelines.is_torch_available', return_value=False):
-                nlp = tf_pipeline(task='sentiment-analysis', model=model, config=config, tokenizer=tokenizer)
-                self._test_mono_column_pipeline(nlp, valid_inputs, invalid_inputs, mandatory_keys)
+            nlp = pipeline(task='sentiment-analysis', model=model, config=config, tokenizer=tokenizer)
+            self._test_mono_column_pipeline(nlp, valid_inputs, invalid_inputs, mandatory_keys)
 
-            with patch('transformers.pipelines.is_tf_available', return_value=False):
-                nlp = torch_pipeline(task='sentiment-analysis', model=model, config=config, tokenizer=tokenizer)
-                self._test_mono_column_pipeline(nlp, valid_inputs, invalid_inputs, mandatory_keys)
+    @require_tf
+    def test_tf_sentiment_analysis(self):
+        mandatory_keys = {'label'}
+        valid_inputs = ['HuggingFace is solving NLP one commit at a time.', 'HuggingFace is based in New-York & Paris']
+        invalid_inputs = [None]
+        for tokenizer, model, config in TEXT_CLASSIF_FINETUNED_MODELS:
+            nlp = pipeline(task='sentiment-analysis', model=model, config=config, tokenizer=tokenizer)
+            self._test_mono_column_pipeline(nlp, valid_inputs, invalid_inputs, mandatory_keys)
 
+    @require_torch
     def test_features_extraction(self):
         valid_inputs = ['HuggingFace is solving NLP one commit at a time.', 'HuggingFace is based in New-York & Paris']
         invalid_inputs = [None]
         for tokenizer, model, config in FEATURE_EXTRACT_FINETUNED_MODELS:
-            with patch('transformers.pipelines.is_torch_available', return_value=False):
-                nlp = tf_pipeline(task='sentiment-analysis', model=model, config=config, tokenizer=tokenizer)
-                self._test_mono_column_pipeline(nlp, valid_inputs, invalid_inputs, {})
+            nlp = pipeline(task='sentiment-analysis', model=model, config=config, tokenizer=tokenizer)
+            self._test_mono_column_pipeline(nlp, valid_inputs, invalid_inputs, {})
 
-            with patch('transformers.pipelines.is_tf_available', return_value=False):
-                nlp = torch_pipeline(task='sentiment-analysis', model=model, config=config, tokenizer=tokenizer)
-                self._test_mono_column_pipeline(nlp, valid_inputs, invalid_inputs, {})
+    @require_tf
+    def test_tf_features_extraction(self):
+        valid_inputs = ['HuggingFace is solving NLP one commit at a time.', 'HuggingFace is based in New-York & Paris']
+        invalid_inputs = [None]
+        for tokenizer, model, config in FEATURE_EXTRACT_FINETUNED_MODELS:
+            nlp = pipeline(task='sentiment-analysis', model=model, config=config, tokenizer=tokenizer)
+            self._test_mono_column_pipeline(nlp, valid_inputs, invalid_inputs, {})
 
 
 class MultiColumnInputTestCase(unittest.TestCase):
@@ -132,6 +135,7 @@ class MultiColumnInputTestCase(unittest.TestCase):
         self.assertRaises(Exception, nlp, invalid_inputs[0])
         self.assertRaises(Exception, nlp, invalid_inputs)
 
+    @require_torch
     def test_question_answering(self):
         mandatory_output_keys = {'score', 'answer', 'start', 'end'}
         valid_samples = [
@@ -149,16 +153,29 @@ class MultiColumnInputTestCase(unittest.TestCase):
         ]
 
         for tokenizer, model, config in QA_FINETUNED_MODELS:
+            nlp = pipeline(task='question-answering', model=model, config=config, tokenizer=tokenizer)
+            self._test_multicolumn_pipeline(nlp, valid_samples, invalid_samples, mandatory_output_keys)
 
-            # Test for Tensorflow
-            with patch('transformers.pipelines.is_torch_available', return_value=False):
-                nlp = pipeline(task='question-answering', model=model, config=config, tokenizer=tokenizer)
-                self._test_multicolumn_pipeline(nlp, valid_samples, invalid_samples, mandatory_output_keys)
+    @require_tf
+    def test_tf_question_answering(self):
+        mandatory_output_keys = {'score', 'answer', 'start', 'end'}
+        valid_samples = [
+            {'question': 'Where was HuggingFace founded ?', 'context': 'HuggingFace was founded in Paris.'},
+            {
+                'question': 'In what field is HuggingFace working ?',
+                'context': 'HuggingFace is a startup based in New-York founded in Paris which is trying to solve NLP.'
+            }
+        ]
+        invalid_samples = [
+            {'question': '', 'context': 'This is a test to try empty question edge case'},
+            {'question': None, 'context': 'This is a test to try empty question edge case'},
+            {'question': 'What is does with empty context ?', 'context': ''},
+            {'question': 'What is does with empty context ?', 'context': None},
+        ]
 
-            # Test for PyTorch
-            with patch('transformers.pipelines.is_tf_available', return_value=False):
-                nlp = pipeline(task='question-answering', model=model, config=config, tokenizer=tokenizer)
-                self._test_multicolumn_pipeline(nlp, valid_samples, invalid_samples, mandatory_output_keys)
+        for tokenizer, model, config in QA_FINETUNED_MODELS:
+            nlp = pipeline(task='question-answering', model=model, config=config, tokenizer=tokenizer)
+            self._test_multicolumn_pipeline(nlp, valid_samples, invalid_samples, mandatory_output_keys)
 
 
 if __name__ == '__main__':

From 15dda5ea32655f2ea565f8d5cd586a036399dba3 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 20 Dec 2019 13:20:41 +0100
Subject: [PATCH 466/505] remove python 2 tests for circle-ci cc @aaugustin
 @julien-c @LysandreJik

---
 .circleci/config.yml | 28 ----------------------------
 1 file changed, 28 deletions(-)

diff --git a/.circleci/config.yml b/.circleci/config.yml
index 7a64eaba7d..b094067eb5 100644
--- a/.circleci/config.yml
+++ b/.circleci/config.yml
@@ -44,32 +44,6 @@ jobs:
             - run: sudo pip install tensorboardX scikit-learn
             - run: python -m pytest -sv ./transformers/tests/ --cov
             - run: codecov
-    build_py2_torch:
-        working_directory: ~/transformers
-        resource_class: large
-        parallelism: 1
-        docker:
-            - image: circleci/python:2.7
-        steps:
-            - checkout
-            - run: sudo pip install torch
-            - run: sudo pip install --progress-bar off .
-            - run: sudo pip install pytest codecov pytest-cov
-            - run: python -m pytest -sv ./transformers/tests/ --cov
-            - run: codecov
-    build_py2_tf:
-        working_directory: ~/transformers
-        resource_class: large
-        parallelism: 1
-        docker:
-            - image: circleci/python:2.7
-        steps:
-            - checkout
-            - run: sudo pip install tensorflow
-            - run: sudo pip install --progress-bar off .
-            - run: sudo pip install pytest codecov pytest-cov
-            - run: python -m pytest -sv ./transformers/tests/ --cov
-            - run: codecov
     build_py3_custom_tokenizers:
         working_directory: ~/transformers
         docker:
@@ -116,6 +90,4 @@ workflows:
             - build_py3_torch_and_tf
             - build_py3_torch
             - build_py3_tf
-            - build_py2_torch
-            - build_py2_tf
             - deploy_doc: *workflow_filters

From 73fcebf7ec122e68b93f50fc770f0515502eb025 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 20 Dec 2019 13:47:35 +0100
Subject: [PATCH 467/505] update serving command

---
 setup.py                         |  6 +++---
 transformers-cli                 |  2 +-
 transformers/commands/serving.py | 35 +++++++++++++++++++++-----------
 3 files changed, 27 insertions(+), 16 deletions(-)

diff --git a/setup.py b/setup.py
index 6560cc4968..4bfb774155 100644
--- a/setup.py
+++ b/setup.py
@@ -38,9 +38,9 @@ from setuptools import find_packages, setup
 
 
 extras = {
-    'serving': ['uvicorn', 'fastapi'],
-    'serving-tf': ['uvicorn', 'fastapi', 'tensorflow'],
-    'serving-torch': ['uvicorn', 'fastapi', 'torch']
+    'serving': ['pydantic', 'uvicorn', 'fastapi'],
+    'serving-tf': ['pydantic', 'uvicorn', 'fastapi', 'tensorflow'],
+    'serving-torch': ['pydantic', 'uvicorn', 'fastapi', 'torch']
 }
 extras['all'] = [package for package in extras.values()]
 
diff --git a/transformers-cli b/transformers-cli
index db2bd0e2a3..0a980a3574 100755
--- a/transformers-cli
+++ b/transformers-cli
@@ -3,9 +3,9 @@ from argparse import ArgumentParser
 
 from transformers.commands.download import DownloadCommand
 from transformers.commands.run import RunCommand
-from transformers.commands.serving import ServeCommand
 from transformers.commands.user import UserCommands
 from transformers.commands.convert import ConvertCommand
+from transformers.commands.serving import ServeCommand
 
 if __name__ == '__main__':
     parser = ArgumentParser('Transformers CLI tool', usage='transformers-cli <command> [<args>]')
diff --git a/transformers/commands/serving.py b/transformers/commands/serving.py
index a7321470ce..3c3f852809 100644
--- a/transformers/commands/serving.py
+++ b/transformers/commands/serving.py
@@ -1,16 +1,23 @@
 from argparse import ArgumentParser, Namespace
 from typing import List, Optional, Union, Any
 
-from fastapi import FastAPI, HTTPException, Body
-from logging import getLogger
+import logging
 
-from pydantic import BaseModel
-from uvicorn import run
+try:
+    from uvicorn import run
+    from fastapi import FastAPI, HTTPException, Body
+    from pydantic import BaseModel
+    _serve_dependancies_installed = True
+except (ImportError, AttributeError):
+    BaseModel = object
+    Body = lambda *x, **y: None
+    _serve_dependancies_installed = False
 
 from transformers import Pipeline
 from transformers.commands import BaseTransformersCLICommand
 from transformers.pipelines import SUPPORTED_TASKS, pipeline
 
+logger = logging.getLogger('transformers-cli/serving')
 
 def serve_command_factory(args: Namespace):
     """
@@ -70,20 +77,24 @@ class ServeCommand(BaseTransformersCLICommand):
         serve_parser.set_defaults(func=serve_command_factory)
 
     def __init__(self, pipeline: Pipeline, host: str, port: int):
-        self._logger = getLogger('transformers-cli/serving')
 
         self._pipeline = pipeline
 
-        self._logger.info('Serving model over {}:{}'.format(host, port))
         self._host = host
         self._port = port
-        self._app = FastAPI()
+        if not _serve_dependancies_installed:
+            raise ImportError("Using serve command requires FastAPI and unicorn. "
+                                "Please install transformers with [serving]: pip install transformers[serving]." 
+                                "Or install FastAPI and unicorn separatly.")
+        else:
+            logger.info('Serving model over {}:{}'.format(host, port))
+            self._app = FastAPI()
 
-        # Register routes
-        self._app.add_api_route('/', self.model_info, response_model=ServeModelInfoResult, methods=['GET'])
-        self._app.add_api_route('/tokenize', self.tokenize, response_model=ServeTokenizeResult, methods=['POST'])
-        self._app.add_api_route('/detokenize', self.detokenize, response_model=ServeDeTokenizeResult, methods=['POST'])
-        self._app.add_api_route('/forward', self.forward, response_model=ServeForwardResult, methods=['POST'])
+            # Register routes
+            self._app.add_api_route('/', self.model_info, response_model=ServeModelInfoResult, methods=['GET'])
+            self._app.add_api_route('/tokenize', self.tokenize, response_model=ServeTokenizeResult, methods=['POST'])
+            self._app.add_api_route('/detokenize', self.detokenize, response_model=ServeDeTokenizeResult, methods=['POST'])
+            self._app.add_api_route('/forward', self.forward, response_model=ServeForwardResult, methods=['POST'])
 
     def run(self):
         run(self._app, host=self._host, port=self._port)

From c37815f1300519e1a812e1080c46641db6f9f604 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 20 Dec 2019 14:35:40 +0100
Subject: [PATCH 468/505] clean up PT <=> TF 2.0 conversion and config loading

---
 .../convert_pytorch_checkpoint_to_tf2.py        |  9 +++++----
 transformers/modeling_tf_utils.py               | 17 ++++++++++++-----
 transformers/modeling_utils.py                  | 17 ++++++++++++-----
 3 files changed, 29 insertions(+), 14 deletions(-)

diff --git a/transformers/convert_pytorch_checkpoint_to_tf2.py b/transformers/convert_pytorch_checkpoint_to_tf2.py
index 4a9832f123..0edac6fb7d 100644
--- a/transformers/convert_pytorch_checkpoint_to_tf2.py
+++ b/transformers/convert_pytorch_checkpoint_to_tf2.py
@@ -32,7 +32,7 @@ from transformers import (load_pytorch_checkpoint_in_tf2_model,
                                   TransfoXLConfig, TFTransfoXLLMHeadModel, TRANSFO_XL_PRETRAINED_CONFIG_ARCHIVE_MAP,
                                   OpenAIGPTConfig, TFOpenAIGPTLMHeadModel, OPENAI_GPT_PRETRAINED_CONFIG_ARCHIVE_MAP,
                                   RobertaConfig, TFRobertaForMaskedLM, TFRobertaForSequenceClassification, ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP,
-                                  DistilBertConfig, TFDistilBertForMaskedLM, TFDistilBertForQuestionAnswering, DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP,
+                                  DistilBertConfig, TFDistilBertForMaskedLM, TFDistilBertForQuestionAnswering, TFDistilBertForSequenceClassification, DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP,
                                   CTRLConfig, TFCTRLLMHeadModel, CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP,
                                   AlbertConfig, TFAlbertForMaskedLM, ALBERT_PRETRAINED_CONFIG_ARCHIVE_MAP,
                                   T5Config, TFT5WithLMHeadModel, T5_PRETRAINED_CONFIG_ARCHIVE_MAP)
@@ -47,7 +47,7 @@ if is_torch_available():
                                       TransfoXLLMHeadModel, TRANSFO_XL_PRETRAINED_MODEL_ARCHIVE_MAP,
                                       OpenAIGPTLMHeadModel, OPENAI_GPT_PRETRAINED_MODEL_ARCHIVE_MAP,
                                       RobertaForMaskedLM, RobertaForSequenceClassification, ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP,
-                                      DistilBertForMaskedLM, DistilBertForQuestionAnswering, DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP,
+                                      DistilBertForMaskedLM, DistilBertForQuestionAnswering, DistilBertForSequenceClassification, DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP,
                                       CTRLLMHeadModel, CTRL_PRETRAINED_MODEL_ARCHIVE_MAP,
                                       AlbertForMaskedLM, ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP,
                                       T5WithLMHeadModel, T5_PRETRAINED_MODEL_ARCHIVE_MAP)
@@ -59,7 +59,7 @@ else:
     TransfoXLLMHeadModel, TRANSFO_XL_PRETRAINED_MODEL_ARCHIVE_MAP,
     OpenAIGPTLMHeadModel, OPENAI_GPT_PRETRAINED_MODEL_ARCHIVE_MAP,
     RobertaForMaskedLM, RobertaForSequenceClassification, ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP,
-    DistilBertForMaskedLM, DistilBertForQuestionAnswering, DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP,
+    DistilBertForMaskedLM, DistilBertForSequenceClassification, DistilBertForQuestionAnswering, DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP,
     CTRLLMHeadModel, CTRL_PRETRAINED_MODEL_ARCHIVE_MAP,
     AlbertForMaskedLM, ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP,
     T5WithLMHeadModel, T5_PRETRAINED_MODEL_ARCHIVE_MAP) = (
@@ -70,7 +70,7 @@ else:
         None, None,
         None, None,
         None, None, None,
-        None, None, None,
+        None, None, None, None,
         None, None,
         None, None,
         None, None)
@@ -93,6 +93,7 @@ MODEL_CLASSES = {
     'roberta-large-mnli': (RobertaConfig, TFRobertaForSequenceClassification, RobertaForSequenceClassification, ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP, ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP),
     'distilbert': (DistilBertConfig, TFDistilBertForMaskedLM, DistilBertForMaskedLM, DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP, DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP),
     'distilbert-base-uncased-distilled-squad': (DistilBertConfig, TFDistilBertForQuestionAnswering, DistilBertForQuestionAnswering, DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP, DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP),
+    'distilbert-base-uncased-distilled-squad': (DistilBertConfig, TFDistilBertForQuestionAnswering, DistilBertForQuestionAnswering, DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP, DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP),
     'ctrl': (CTRLConfig, TFCTRLLMHeadModel, CTRLLMHeadModel, CTRL_PRETRAINED_MODEL_ARCHIVE_MAP, CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP),
     'albert': (AlbertConfig, TFAlbertForMaskedLM, AlbertForMaskedLM, ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP, ALBERT_PRETRAINED_CONFIG_ARCHIVE_MAP),
     't5': (T5Config, TFT5WithLMHeadModel, T5WithLMHeadModel, T5_PRETRAINED_MODEL_ARCHIVE_MAP, T5_PRETRAINED_CONFIG_ARCHIVE_MAP),
diff --git a/transformers/modeling_tf_utils.py b/transformers/modeling_tf_utils.py
index 401ffeb67e..0aa65a9f17 100644
--- a/transformers/modeling_tf_utils.py
+++ b/transformers/modeling_tf_utils.py
@@ -184,7 +184,9 @@ class TFPreTrainedModel(tf.keras.Model):
             model_args: (`optional`) Sequence of positional arguments:
                 All remaning positional arguments will be passed to the underlying model's ``__init__`` method
 
-            config: (`optional`) instance of a class derived from :class:`~transformers.PretrainedConfig`:
+            config: (`optional`) one of:
+                    - an instance of a class derived from :class:`~transformers.PretrainedConfig`, or
+                    - a string valid as input to :func:`~transformers.PretrainedConfig.from_pretrained()`
                 Configuration for the model to use instead of an automatically loaded configuation. Configuration can be automatically loaded when:
 
                 - the model is a model provided by the library (loaded with the ``shortcut-name`` string of a pretrained model), or
@@ -236,10 +238,11 @@ class TFPreTrainedModel(tf.keras.Model):
         proxies = kwargs.pop('proxies', None)
         output_loading_info = kwargs.pop('output_loading_info', False)
 
-        # Load config
-        if config is None:
+        # Load config if we don't provide a configuration
+        if not isinstance(config, PretrainedConfig):
+            config_path = config if config is not None else pretrained_model_name_or_path
             config, model_kwargs = cls.config_class.from_pretrained(
-                pretrained_model_name_or_path, *model_args,
+                config_path, *model_args,
                 cache_dir=cache_dir, return_unused_kwargs=True,
                 force_download=force_download,
                 resume_download=resume_download,
@@ -310,7 +313,11 @@ class TFPreTrainedModel(tf.keras.Model):
         assert os.path.isfile(resolved_archive_file), "Error retrieving file {}".format(resolved_archive_file)
         # 'by_name' allow us to do transfer learning by skipping/adding layers
         # see https://github.com/tensorflow/tensorflow/blob/00fad90125b18b80fe054de1055770cfb8fe4ba3/tensorflow/python/keras/engine/network.py#L1339-L1357
-        model.load_weights(resolved_archive_file, by_name=True)
+        try:
+            model.load_weights(resolved_archive_file, by_name=True)
+        except OSError:
+            raise OSError("Unable to load weights from h5 file. "
+                          "If you tried to load a TF 2.0 model from a PyTorch checkpoint, please set from_pt=True. ")
 
         ret = model(model.dummy_inputs, training=False)  # Make sure restore ops are run
 
diff --git a/transformers/modeling_utils.py b/transformers/modeling_utils.py
index eff54f71e1..3bc407e4a3 100644
--- a/transformers/modeling_utils.py
+++ b/transformers/modeling_utils.py
@@ -281,7 +281,9 @@ class PreTrainedModel(nn.Module):
             model_args: (`optional`) Sequence of positional arguments:
                 All remaning positional arguments will be passed to the underlying model's ``__init__`` method
 
-            config: (`optional`) instance of a class derived from :class:`~transformers.PretrainedConfig`:
+            config: (`optional`) one of:
+                    - an instance of a class derived from :class:`~transformers.PretrainedConfig`, or
+                    - a string valid as input to :func:`~transformers.PretrainedConfig.from_pretrained()`
                 Configuration for the model to use instead of an automatically loaded configuation. Configuration can be automatically loaded when:
 
                 - the model is a model provided by the library (loaded with the ``shortcut-name`` string of a pretrained model), or
@@ -336,10 +338,11 @@ class PreTrainedModel(nn.Module):
         proxies = kwargs.pop('proxies', None)
         output_loading_info = kwargs.pop('output_loading_info', False)
 
-        # Load config
-        if config is None:
+        # Load config if we don't provide a configuration
+        if not isinstance(config, PretrainedConfig):
+            config_path = config if config is not None else pretrained_model_name_or_path
             config, model_kwargs = cls.config_class.from_pretrained(
-                pretrained_model_name_or_path, *model_args,
+                config_path, *model_args,
                 cache_dir=cache_dir, return_unused_kwargs=True,
                 force_download=force_download,
                 resume_download=resume_download,
@@ -408,7 +411,11 @@ class PreTrainedModel(nn.Module):
         model = cls(config, *model_args, **model_kwargs)
 
         if state_dict is None and not from_tf:
-            state_dict = torch.load(resolved_archive_file, map_location='cpu')
+            try:
+                state_dict = torch.load(resolved_archive_file, map_location='cpu')
+            except:
+                raise OSError("Unable to load weights from pytorch checkpoint file. "
+                            "If you tried to load a PyTorch model from a TF 2.0 checkpoint, please set from_tf=True. ")
 
         missing_keys = []
         unexpected_keys = []

From 7f74084528b8f9fb7678b82829366f11326af62f Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Fri, 20 Dec 2019 14:47:04 +0100
Subject: [PATCH 469/505] Fix leading axis added when saving through the
 command run

---
 transformers/commands/run.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/transformers/commands/run.py b/transformers/commands/run.py
index 78109b2a16..50c42d3a40 100644
--- a/transformers/commands/run.py
+++ b/transformers/commands/run.py
@@ -50,9 +50,9 @@ class RunCommand(BaseTransformersCLICommand):
         nlp, output = self._nlp, []
         for entry in self._reader:
             if self._reader.is_multi_columns:
-                output += [nlp(**entry)]
+                output += nlp(**entry)
             else:
-                output += [nlp(entry)]
+                output += nlp(entry)
 
         # Saving data
         if self._nlp.binary_output:

From db0795b5d0da0a9968035751b246854240a2c2ec Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 20 Dec 2019 15:07:00 +0100
Subject: [PATCH 470/505] defaults models for tf and pt - update tests

---
 transformers/pipelines.py            | 23 ++++++++++++++----
 transformers/tests/pipelines_test.py | 36 ++++++++++++++++++++++++----
 2 files changed, 50 insertions(+), 9 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index efb1de92e1..33e1ab4022 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -776,7 +776,10 @@ SUPPORTED_TASKS = {
         'tf': TFAutoModel if is_tf_available() else None,
         'pt': AutoModel if is_torch_available() else None,
         'default': {
-            'model': 'distilbert-base-uncased',
+            'model': {
+                'pt': 'distilbert-base-uncased',
+                'tf': 'distilbert-base-uncased',
+            },
             'config': None,
             'tokenizer': 'distilbert-base-uncased'
         }
@@ -786,7 +789,10 @@ SUPPORTED_TASKS = {
         'tf': TFAutoModelForSequenceClassification if is_tf_available() else None,
         'pt': AutoModelForSequenceClassification if is_torch_available() else None,
         'default': {
-            'model': 'https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-finetuned-sst-2-english-pytorch_model.bin',
+            'model': {
+                'pt': 'https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-finetuned-sst-2-english-pytorch_model.bin',
+                'tf': 'https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-finetuned-sst-2-english-tf_model.h5',
+            },
             'config': 'https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-finetuned-sst-2-english-config.json',
             'tokenizer': 'distilbert-base-uncased'
         }
@@ -796,7 +802,10 @@ SUPPORTED_TASKS = {
         'tf': TFAutoModelForTokenClassification if is_tf_available() else None,
         'pt': AutoModelForTokenClassification if is_torch_available() else None,
         'default': {
-            'model': 'https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-cased-finetuned-conll03-english-pytorch_model.bin',
+            'model': {
+                'pt':'https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-cased-finetuned-conll03-english-pytorch_model.bin',
+                'tf': 'https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-cased-finetuned-conll03-english-tf_model.h5',
+            },
             'config': 'https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-cased-finetuned-conll03-english-config.json',
             'tokenizer': 'bert-large-cased'
         }
@@ -806,7 +815,10 @@ SUPPORTED_TASKS = {
         'tf': TFAutoModelForQuestionAnswering if is_tf_available() else None,
         'pt': AutoModelForQuestionAnswering if is_torch_available() else None,
         'default': {
-            'model': 'distilbert-base-uncased-distilled-squad',
+            'model': {
+                'pt': 'distilbert-base-uncased-distilled-squad',
+                'tf': 'distilbert-base-uncased-distilled-squad',
+            },
             'config': None,
             'tokenizer': 'distilbert-base-uncased'
         }
@@ -843,7 +855,8 @@ def pipeline(task: str, model: Optional = None,
 
     # Use default model/config/tokenizer for the task if no model is provided
     if model is None:
-        model, config, tokenizer = tuple(targeted_task['default'].values())
+        models, config, tokenizer = tuple(targeted_task['default'].values())
+        model = models[framework]
 
     # Try to infer tokenizer from model or config name (if provided as str)
     if tokenizer is None:
diff --git a/transformers/tests/pipelines_test.py b/transformers/tests/pipelines_test.py
index 14bf07ee30..08a1507770 100644
--- a/transformers/tests/pipelines_test.py
+++ b/transformers/tests/pipelines_test.py
@@ -11,6 +11,20 @@ QA_FINETUNED_MODELS = {
     ('bert-base-uncased', 'distilbert-base-uncased-distilled-squad', None)
 }
 
+TF_QA_FINETUNED_MODELS = {
+    ('bert-base-uncased', 'bert-large-uncased-whole-word-masking-finetuned-squad', None),
+    ('bert-base-cased', 'bert-large-cased-whole-word-masking-finetuned-squad', None),
+    ('bert-base-uncased', 'distilbert-base-uncased-distilled-squad', None)
+}
+
+TF_NER_FINETUNED_MODELS = {
+    (
+        'bert-base-cased',
+        'https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-cased-finetuned-conll03-english-tf_model.h5',
+        'https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-cased-finetuned-conll03-english-config.json'
+    )
+}
+
 NER_FINETUNED_MODELS = {
     (
         'bert-base-cased',
@@ -25,6 +39,20 @@ FEATURE_EXTRACT_FINETUNED_MODELS = {
    ('distilbert-base-uncased', 'distilbert-base-uncased', None)
 }
 
+TF_FEATURE_EXTRACT_FINETUNED_MODELS = {
+   ('bert-base-cased', 'bert-base-cased', None),
+   # ('xlnet-base-cased', 'xlnet-base-cased', None), # Disabled for now as it crash for TF2
+   ('distilbert-base-uncased', 'distilbert-base-uncased', None)
+}
+
+TF_TEXT_CLASSIF_FINETUNED_MODELS = {
+    (
+        'bert-base-uncased',
+        'https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-finetuned-sst-2-english-tf_model.h5',
+        'https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-finetuned-sst-2-english-config.json'
+    )
+}
+
 TEXT_CLASSIF_FINETUNED_MODELS = {
     (
         'bert-base-uncased',
@@ -75,7 +103,7 @@ class MonoColumnInputTestCase(unittest.TestCase):
         mandatory_keys = {'entity', 'word', 'score'}
         valid_inputs = ['HuggingFace is solving NLP one commit at a time.', 'HuggingFace is based in New-York & Paris']
         invalid_inputs = [None]
-        for tokenizer, model, config in NER_FINETUNED_MODELS:
+        for tokenizer, model, config in TF_NER_FINETUNED_MODELS:
             nlp = pipeline(task='ner', model=model, config=config, tokenizer=tokenizer)
             self._test_mono_column_pipeline(nlp, valid_inputs, invalid_inputs, mandatory_keys)
 
@@ -93,7 +121,7 @@ class MonoColumnInputTestCase(unittest.TestCase):
         mandatory_keys = {'label'}
         valid_inputs = ['HuggingFace is solving NLP one commit at a time.', 'HuggingFace is based in New-York & Paris']
         invalid_inputs = [None]
-        for tokenizer, model, config in TEXT_CLASSIF_FINETUNED_MODELS:
+        for tokenizer, model, config in TF_TEXT_CLASSIF_FINETUNED_MODELS:
             nlp = pipeline(task='sentiment-analysis', model=model, config=config, tokenizer=tokenizer)
             self._test_mono_column_pipeline(nlp, valid_inputs, invalid_inputs, mandatory_keys)
 
@@ -109,7 +137,7 @@ class MonoColumnInputTestCase(unittest.TestCase):
     def test_tf_features_extraction(self):
         valid_inputs = ['HuggingFace is solving NLP one commit at a time.', 'HuggingFace is based in New-York & Paris']
         invalid_inputs = [None]
-        for tokenizer, model, config in FEATURE_EXTRACT_FINETUNED_MODELS:
+        for tokenizer, model, config in TF_FEATURE_EXTRACT_FINETUNED_MODELS:
             nlp = pipeline(task='sentiment-analysis', model=model, config=config, tokenizer=tokenizer)
             self._test_mono_column_pipeline(nlp, valid_inputs, invalid_inputs, {})
 
@@ -173,7 +201,7 @@ class MultiColumnInputTestCase(unittest.TestCase):
             {'question': 'What is does with empty context ?', 'context': None},
         ]
 
-        for tokenizer, model, config in QA_FINETUNED_MODELS:
+        for tokenizer, model, config in TF_QA_FINETUNED_MODELS:
             nlp = pipeline(task='question-answering', model=model, config=config, tokenizer=tokenizer)
             self._test_multicolumn_pipeline(nlp, valid_samples, invalid_samples, mandatory_output_keys)
 

From 4e3f745ba4e754e415c184d53c874031101d263b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Fri, 20 Dec 2019 11:13:46 +0100
Subject: [PATCH 471/505] add example for Model2Model in quickstart

---
 docs/source/quickstart.md | 95 ++++++++++++++++++++++++++++++++++++++-
 1 file changed, 94 insertions(+), 1 deletion(-)

diff --git a/docs/source/quickstart.md b/docs/source/quickstart.md
index 530aff8eb0..60e2cf3fd8 100644
--- a/docs/source/quickstart.md
+++ b/docs/source/quickstart.md
@@ -219,4 +219,97 @@ sequence = tokenizer.decode(generated)
 print(sequence)
 ```
 
-The model only requires a single token as input as all the previous tokens' key/value pairs are contained in the `past`.
\ No newline at end of file
+The model only requires a single token as input as all the previous tokens' key/value pairs are contained in the `past`.
+
+### Model2Model example
+
+Encoder-decoder architectures require two tokenized inputs: one for the encoder and the other one for the decoder. Let's assume that we want to use `Model2Model` for generative question answering, and start by tokenizing the question and answer that will be fed to the model.
+
+```python
+import torch
+from transformers import BertTokenizer, Model2Model
+
+# OPTIONAL: if you want to have more information on what's happening under the hood, activate the logger as follows
+import logging
+logging.basicConfig(level=logging.INFO)
+
+# Load pre-trained model tokenizer (vocabulary)
+tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
+
+# Encode the input to the encoder (the question)
+question = "Who was Jim Henson?"
+encoded_question = tokenizer.encode(question)
+
+# Encode the input to the decoder (the answer)
+answer = "Jim Henson was a puppeteer"
+encoded_answer = tokenizer.encode(answer)
+
+# Convert inputs to PyTorch tensors
+question_tensor = torch.tensor([encoded_question])
+answer_tensor = torch.tensor([encoded_answer])
+```
+
+Let's see how we can use `Model2Model` to get the value of the loss associated with this (question, answer) pair:
+
+```python
+# In order to compute the loss we need to provide language model
+# labels (the token ids that the model should have produced) to
+# the decoder.
+lm_labels =  encoded_answer
+labels_tensor = torch.tensor([lm_labels])
+
+# Load pre-trained model (weights)
+model = Model2Model.from_pretrained('bert-base-uncased')
+
+# Set the model in evaluation mode to deactivate the DropOut modules
+# This is IMPORTANT to have reproducible results during evaluation!
+model.eval()
+
+# If you have a GPU, put everything on cuda
+question_tensor = question_tensor.to('cuda')
+answer_tensor = answer_tensor.to('cuda')
+labels_tensor = labels_tensor.to('cuda')
+model.to('cuda')
+
+# Predict hidden states features for each layer
+with torch.no_grad():
+    # See the models docstrings for the detail of the inputs
+    outputs = model(question_tensor, answer_tensor, decoder_lm_labels=labels_tensor)
+    # Transformers models always output tuples.
+    # See the models docstrings for the detail of all the outputs
+    # In our case, the first element is the value of the LM loss 
+    lm_loss = outputs[0]
+```
+
+This loss can be used to fine-tune `Model2Model` on the question answering task. Assuming that we fine-tuned the model, let us now see how to generate an answer:
+
+```python
+# Let's re-use the previous question
+question = "Who was Jim Henson?"
+encoded_question = tokenizer.encode(question)
+question_tensor = torch.tensor([encoded_question])
+
+# This time we try to generate the answer, so we start with an empty sequence
+answer = "[CLS]"
+encoded_answer = tokenizer.encode(answer, add_special_tokens=False)
+answer_tensor = torch.tensor([encoded_answer])
+
+# Load pre-trained model (weights)
+model = Model2Model.from_pretrained('fine-tuned-weights')
+model.eval()
+
+# If you have a GPU, put everything on cuda
+question_tensor = encoded_question.to('cuda')
+answer_tensor = encoded_answer.to('cuda')
+model.to('cuda')
+
+# Predict all tokens
+with torch.no_grad():
+    outputs = model(question_tensor, answer_tensor)
+    predictions = outputs[0]
+
+# confirm we were able to predict 'jim'
+predicted_index = torch.argmax(predictions[0, -1]).item()
+predicted_token = tokenizer.convert_ids_to_tokens([predicted_index])[0]
+assert predicted_token == 'jim'
+```

From b98ff8854460a04fe076c704555705c5d5e1b6de Mon Sep 17 00:00:00 2001
From: Morgan Funtowicz <morgan@huggingface.co>
Date: Fri, 20 Dec 2019 15:52:50 +0100
Subject: [PATCH 472/505] Added pipelines quick tour in README

---
 README.md | 29 +++++++++++++++++++++++++++++
 1 file changed, 29 insertions(+)

diff --git a/README.md b/README.md
index a9d0fb3ace..1312fcc0ac 100644
--- a/README.md
+++ b/README.md
@@ -490,6 +490,35 @@ transformers-cli ls
 # List all your S3 objects.
 ```
 
+## Quick tour of pipelines
+
+New in version `v2.3`: `Pipeline` are high-level objects which automatically handle tokenization, running your data through a transformers model
+and outputting the result in a structured object. 
+
+You can create `Pipeline` objects for the following down-stream tasks:
+ - `feature-extraction`: Generates a tensor representation for the input sequence
+ - `ner`: Generates named entity mapping for each word in the input sequence.
+ - `sentiment-analysis`: Gives the polarity (positive / negative) of the whole input sequence. 
+ - `question-answering`: Provided some context and a question refering to the context, it will extract the answer to the question
+ in the context.
+
+```python
+from transformers import pipeline
+
+# Allocate a pipeline for sentiment-analysis
+nlp = pipeline('sentiment-analysis')
+nlp('We are very happy to include pipeline into the transformers repository.')
+>>> {'label': 'POSITIVE', 'score': 0.99893874}
+
+# Allocate a pipeline for question-answering
+nlp = pipeline('question-answering')
+nlp({
+    'question': 'What is the name of the repository ?', 
+    'context': 'Pipeline have been included in the huggingface/transformers repository'
+})
+>>> {'score': 0.28756016668193496, 'start': 35, 'end': 59, 'answer': 'huggingface/transformers'}
+```
+
 ## Migrating from pytorch-transformers to transformers
 
 Here is a quick summary of what you should take care of when migrating from `pytorch-transformers` to `transformers`.

From 90debb9ff2636f3f1c8256237deeca3a1ec3c7dd Mon Sep 17 00:00:00 2001
From: Dirk Groeneveld <dirkg@allenai.org>
Date: Thu, 19 Dec 2019 17:03:01 -0800
Subject: [PATCH 473/505] Keep even the first of the special tokens intact
 while lowercasing.

---
 transformers/tokenization_utils.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index 59e2d05212..2635a38db0 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -642,7 +642,7 @@ class PreTrainedTokenizer(object):
         def lowercase_text(t):
             # convert non-special tokens to lowercase
             escaped_special_toks = [re.escape(s_tok) for s_tok in all_special_tokens]
-            pattern = r'(^' + r'|'.join(escaped_special_toks) + r')|' + \
+            pattern = r'(' + r'|'.join(escaped_special_toks) + r')|' + \
                       r'(.+?)'
             return re.sub(
                 pattern,

From fb393ad9945f66b081f88b81b90a2974d81e9601 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Fri, 20 Dec 2019 11:29:58 -0500
Subject: [PATCH 474/505] Added test for all special tokens

---
 transformers/tests/tokenization_tests_commons.py | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/transformers/tests/tokenization_tests_commons.py b/transformers/tests/tokenization_tests_commons.py
index 13e7ae746a..cdec24d9f0 100644
--- a/transformers/tests/tokenization_tests_commons.py
+++ b/transformers/tests/tokenization_tests_commons.py
@@ -133,6 +133,14 @@ class CommonTestCases:
             self.assertNotEqual(len(toks), len(toks0))  # toks0 should be longer
             self.assertListEqual(toks, toks2)
 
+            # Check that none of the special tokens are lowercased
+            sequence_with_special_tokens = "A " + " yEs ".join(tokenizer.all_special_tokens) + " B"
+            tokenized_sequence = tokenizer.tokenize(sequence_with_special_tokens)
+
+            for special_token in tokenizer.all_special_tokens:
+                print(special_token, special_token in tokenized_sequence)
+                assert special_token in tokenized_sequence
+
             tokenizer = self.get_tokenizer(do_lower_case=False)
 
             added = tokenizer.add_tokens(new_toks)

From 65c75fc58796b278d58b0ce2c8d2031594ef0f64 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Fri, 20 Dec 2019 11:34:16 -0500
Subject: [PATCH 475/505] Clean special tokens test

---
 transformers/tests/tokenization_tests_commons.py | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/transformers/tests/tokenization_tests_commons.py b/transformers/tests/tokenization_tests_commons.py
index cdec24d9f0..c417d033dc 100644
--- a/transformers/tests/tokenization_tests_commons.py
+++ b/transformers/tests/tokenization_tests_commons.py
@@ -138,8 +138,7 @@ class CommonTestCases:
             tokenized_sequence = tokenizer.tokenize(sequence_with_special_tokens)
 
             for special_token in tokenizer.all_special_tokens:
-                print(special_token, special_token in tokenized_sequence)
-                assert special_token in tokenized_sequence
+                self.assertTrue(special_token in tokenized_sequence)
 
             tokenizer = self.get_tokenizer(do_lower_case=False)
 

From 1c12ee0e55f4aec6d90c789c7b45e9adac0ed259 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 20 Dec 2019 18:28:27 +0100
Subject: [PATCH 476/505] fixing xlm-roberta tokenizer max_length and
 automodels

---
 transformers/modeling_auto.py            | 17 ++++++++++++++---
 transformers/modeling_utils.py           |  2 +-
 transformers/pipelines.py                |  5 +++--
 transformers/tokenization_utils.py       |  6 +++++-
 transformers/tokenization_xlm_roberta.py | 16 ++++++++++------
 5 files changed, 33 insertions(+), 13 deletions(-)

diff --git a/transformers/modeling_auto.py b/transformers/modeling_auto.py
index 761b2ce324..6b49efd378 100644
--- a/transformers/modeling_auto.py
+++ b/transformers/modeling_auto.py
@@ -20,7 +20,7 @@ import logging
 
 from .configuration_auto import (AlbertConfig, BertConfig, CamembertConfig, CTRLConfig,
                                  DistilBertConfig, GPT2Config, OpenAIGPTConfig, RobertaConfig,
-                                 TransfoXLConfig, XLMConfig, XLNetConfig)
+                                 TransfoXLConfig, XLMConfig, XLNetConfig, XLMRobertaConfig)
 
 from .modeling_bert import BertModel, BertForMaskedLM, BertForSequenceClassification, BertForQuestionAnswering, \
     BertForTokenClassification, BERT_PRETRAINED_MODEL_ARCHIVE_MAP
@@ -41,7 +41,8 @@ from .modeling_camembert import CamembertModel, CamembertForMaskedLM, CamembertF
 from .modeling_albert import AlbertModel, AlbertForMaskedLM, AlbertForSequenceClassification, \
     AlbertForQuestionAnswering, ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP
 from .modeling_t5 import T5Model, T5WithLMHeadModel, T5_PRETRAINED_MODEL_ARCHIVE_MAP
-from .modeling_xlm_roberta import XLMRobertaModel, XLMRobertaForMaskedLM, XLMRobertaForSequenceClassification, XLMRobertaForMultipleChoice, XLM_ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP
+from .modeling_xlm_roberta import XLMRobertaModel, XLMRobertaForMaskedLM, XLMRobertaForSequenceClassification, \
+    XLMRobertaForMultipleChoice, XLMRobertaForTokenClassification, XLM_ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP
 
 from .modeling_utils import PreTrainedModel, SequenceSummary
 
@@ -146,6 +147,8 @@ class AutoModel(object):
             return AlbertModel(config)
         elif isinstance(config, CamembertConfig):
             return CamembertModel(config)
+        elif isinstance(config, XLMRobertaConfig):
+            return XLMRobertaModel(config)
         raise ValueError("Unrecognized configuration class {}".format(config))
 
     @classmethod
@@ -333,6 +336,8 @@ class AutoModelWithLMHead(object):
             return XLMWithLMHeadModel(config)
         elif isinstance(config, CTRLConfig):
             return CTRLLMHeadModel(config)
+        elif isinstance(config, XLMRobertaConfig):
+            return XLMRobertaForMaskedLM(config)
         raise ValueError("Unrecognized configuration class {}".format(config))
 
     @classmethod
@@ -509,6 +514,8 @@ class AutoModelForSequenceClassification(object):
             return XLNetForSequenceClassification(config)
         elif isinstance(config, XLMConfig):
             return XLMForSequenceClassification(config)
+        elif isinstance(config, XLMRobertaConfig):
+            return XLMRobertaForSequenceClassification(config)
         raise ValueError("Unrecognized configuration class {}".format(config))
 
     @classmethod
@@ -787,6 +794,8 @@ class AutoModelForTokenClassification:
             return XLNetForTokenClassification(config)
         elif isinstance(config, RobertaConfig):
             return RobertaForTokenClassification(config)
+        elif isinstance(config, XLMRobertaConfig):
+            return XLMRobertaForTokenClassification(config)
         raise ValueError("Unrecognized configuration class {}".format(config))
     
     @classmethod
@@ -865,6 +874,8 @@ class AutoModelForTokenClassification:
             return CamembertForTokenClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'distilbert' in pretrained_model_name_or_path:
             return DistilBertForTokenClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
+        elif 'xlm-roberta' in pretrained_model_name_or_path:
+            return XLMRobertaForTokenClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'roberta' in pretrained_model_name_or_path:
             return RobertaForTokenClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
         elif 'bert' in pretrained_model_name_or_path:
@@ -873,4 +884,4 @@ class AutoModelForTokenClassification:
             return XLNetForTokenClassification.from_pretrained(pretrained_model_name_or_path, *model_args, **kwargs)
 
         raise ValueError("Unrecognized model identifier in {}. Should contains one of "
-                         "'bert', 'xlnet', 'camembert', 'distilbert', 'roberta'".format(pretrained_model_name_or_path))
+                         "'bert', 'xlnet', 'camembert', 'distilbert', 'xlm-roberta', 'roberta'".format(pretrained_model_name_or_path))
diff --git a/transformers/modeling_utils.py b/transformers/modeling_utils.py
index d899771603..e3f4b9f3b8 100644
--- a/transformers/modeling_utils.py
+++ b/transformers/modeling_utils.py
@@ -415,7 +415,7 @@ class PreTrainedModel(nn.Module):
                 state_dict = torch.load(resolved_archive_file, map_location='cpu')
             except:
                 raise OSError("Unable to load weights from pytorch checkpoint file. "
-                            "If you tried to load a PyTorch model from a TF 2.0 checkpoint, please set from_tf=True. ")
+                              "If you tried to load a PyTorch model from a TF 2.0 checkpoint, please set from_tf=True. ")
 
         missing_keys = []
         unexpected_keys = []
diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 33e1ab4022..c3756109af 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -49,7 +49,7 @@ logger = logging.getLogger(__name__)
 
 def get_framework(model=None):
     """ Select framework (TensorFlow/PyTorch) to use.
-        If both frameworks are installed and no specific model is provided, defaults to using TensorFlow.
+        If both frameworks are installed and no specific model is provided, defaults to using PyTorch.
     """
     if is_tf_available() and is_torch_available() and model is not None and not isinstance(model, str):
         # Both framework are available but the use supplied a model class instance.
@@ -60,7 +60,8 @@ def get_framework(model=None):
                           "To install TensorFlow 2.0, read the instructions at https://www.tensorflow.org/install/ "
                           "To install PyTorch, read the instructions at https://pytorch.org/.")
     else:
-        framework = 'tf' if is_tf_available() else 'pt'
+        # framework = 'tf' if is_tf_available() else 'pt'
+        framework = 'pt' if is_torch_available() else 'tf'
     return framework
 
 class ArgumentHandler(ABC):
diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index 2635a38db0..d77a7100ab 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -434,7 +434,11 @@ class PreTrainedTokenizer(object):
                     init_kwargs[key] = value
 
         # Instantiate tokenizer.
-        tokenizer = cls(*init_inputs, **init_kwargs)
+        try:
+            tokenizer = cls(*init_inputs, **init_kwargs)
+        except OSError:
+            OSError("Unable to load vocabulary from file. "
+                    "Please check that the provided vocabulary is accessible and not corrupted.")
 
         # Save inputs and kwargs for saving and re-loading with ``save_pretrained``
         tokenizer.init_inputs = init_inputs
diff --git a/transformers/tokenization_xlm_roberta.py b/transformers/tokenization_xlm_roberta.py
index 4397e7b031..57a42dde5c 100644
--- a/transformers/tokenization_xlm_roberta.py
+++ b/transformers/tokenization_xlm_roberta.py
@@ -40,8 +40,12 @@ PRETRAINED_VOCAB_FILES_MAP = {
 }
 
 PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
-    'xlm-roberta-base': None,
-    'xlm-roberta-large': None,
+    'xlm-roberta-base': 512,
+    'xlm-roberta-large': 512,
+    'xlm-roberta-large-finetuned-conll02-dutch': 512,
+    'xlm-roberta-large-finetuned-conll02-spanish': 512,
+    'xlm-roberta-large-finetuned-conll03-english': 512,
+    'xlm-roberta-large-finetuned-conll03-german': 512,
 }
 
 class XLMRobertaTokenizer(PreTrainedTokenizer):
@@ -58,10 +62,10 @@ class XLMRobertaTokenizer(PreTrainedTokenizer):
     def __init__(self, vocab_file, bos_token="<s>", eos_token="</s>", sep_token="</s>",
                  cls_token="<s>", unk_token="<unk>", pad_token='<pad>', mask_token='<mask>',
                  **kwargs):
-        super(XLMRobertaTokenizer, self).__init__(max_len=512, bos_token=bos_token, eos_token=eos_token, unk_token=unk_token,
-                                                 sep_token=sep_token, cls_token=cls_token, pad_token=pad_token,
-                                                 mask_token=mask_token,
-                                                 **kwargs)
+        super(XLMRobertaTokenizer, self).__init__(bos_token=bos_token, eos_token=eos_token, unk_token=unk_token,
+                                                  sep_token=sep_token, cls_token=cls_token, pad_token=pad_token,
+                                                  mask_token=mask_token,
+                                                  **kwargs)
         self.max_len_single_sentence = self.max_len - 2  # take into account special tokens
         self.max_len_sentences_pair = self.max_len - 4  # take into account special tokens
         self.sp_model = spm.SentencePieceProcessor()

From bbaaec046c493c9bb3bb2b24f1352413c647e3ff Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 20 Dec 2019 19:19:20 +0100
Subject: [PATCH 477/505] fixing CLI pipeline

---
 transformers/commands/run.py | 32 +++++++++++------
 transformers/pipelines.py    | 67 +++++++++++++++++++-----------------
 2 files changed, 58 insertions(+), 41 deletions(-)

diff --git a/transformers/commands/run.py b/transformers/commands/run.py
index 50c42d3a40..c2c141734b 100644
--- a/transformers/commands/run.py
+++ b/transformers/commands/run.py
@@ -9,6 +9,9 @@ logger = logging.getLogger(__name__)  # pylint: disable=invalid-name
 
 
 def try_infer_format_from_ext(path: str):
+    if not path:
+        return 'pipe'
+
     for ext in PipelineDataFormat.SUPPORTED_FORMATS:
         if path.endswith(ext):
             return ext
@@ -20,9 +23,16 @@ def try_infer_format_from_ext(path: str):
 
 
 def run_command_factory(args):
-    nlp = pipeline(task=args.task, model=args.model, config=args.config, tokenizer=args.tokenizer, device=args.device)
+    nlp = pipeline(task=args.task,
+                   model=args.model if args.model else None,
+                   config=args.config,
+                   tokenizer=args.tokenizer,
+                   device=args.device)
     format = try_infer_format_from_ext(args.input) if args.format == 'infer' else args.format
-    reader = PipelineDataFormat.from_str(format, args.output, args.input, args.column)
+    reader = PipelineDataFormat.from_str(format=format,
+                                         output_path=args.output,
+                                         input_path=args.input,
+                                         column=args.column if args.column else nlp.default_input_names)
     return RunCommand(nlp, reader)
 
 
@@ -35,24 +45,26 @@ class RunCommand(BaseTransformersCLICommand):
     @staticmethod
     def register_subcommand(parser: ArgumentParser):
         run_parser = parser.add_parser('run', help="Run a pipeline through the CLI")
-        run_parser.add_argument('--device', type=int, default=-1, help='Indicate the device to run onto, -1 indicates CPU, >= 0 indicates GPU (default: -1)')
         run_parser.add_argument('--task', choices=SUPPORTED_TASKS.keys(), help='Task to run')
-        run_parser.add_argument('--model', type=str, required=True, help='Name or path to the model to instantiate.')
+        run_parser.add_argument('--input', type=str, help='Path to the file to use for inference')
+        run_parser.add_argument('--output', type=str, help='Path to the file that will be used post to write results.')
+        run_parser.add_argument('--model', type=str, help='Name or path to the model to instantiate.')
         run_parser.add_argument('--config', type=str, help='Name or path to the model\'s config to instantiate.')
         run_parser.add_argument('--tokenizer', type=str, help='Name of the tokenizer to use. (default: same as the model name)')
         run_parser.add_argument('--column', type=str, help='Name of the column to use as input. (For multi columns input as QA use column1,columns2)')
         run_parser.add_argument('--format', type=str, default='infer', choices=PipelineDataFormat.SUPPORTED_FORMATS, help='Input format to read from')
-        run_parser.add_argument('--input', type=str, help='Path to the file to use for inference')
-        run_parser.add_argument('--output', type=str, help='Path to the file that will be used post to write results.')
+        run_parser.add_argument('--device', type=int, default=-1, help='Indicate the device to run onto, -1 indicates CPU, >= 0 indicates GPU (default: -1)')
         run_parser.set_defaults(func=run_command_factory)
 
     def run(self):
-        nlp, output = self._nlp, []
+        nlp, outputs = self._nlp, []
+
         for entry in self._reader:
-            if self._reader.is_multi_columns:
-                output += nlp(**entry)
+            output = nlp(**entry) if self._reader.is_multi_columns else nlp(entry)
+            if isinstance(output, dict):
+                outputs.append(output)
             else:
-                output += nlp(entry)
+                outputs += output
 
         # Saving data
         if self._nlp.binary_output:
diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index c3756109af..01491cf2be 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -14,12 +14,14 @@
 # limitations under the License.
 from __future__ import absolute_import, division, print_function, unicode_literals
 
+import sys
 import csv
 import json
 import os
 import pickle
 import logging
 import six
+
 from abc import ABC, abstractmethod
 from contextlib import contextmanager
 from itertools import groupby
@@ -98,28 +100,29 @@ class PipelineDataFormat:
     Supported data formats currently includes:
      - JSON
      - CSV
+     - stdin/stdout (pipe)
 
     PipelineDataFormat also includes some utilities to work with multi-columns like mapping from datasets columns
     to pipelines keyword arguments through the `dataset_kwarg_1=dataset_column_1` format.
     """
     SUPPORTED_FORMATS = ['json', 'csv', 'pipe']
 
-    def __init__(self, output: Optional[str], input: Optional[str], column: Optional[str]):
-        self.output = output
-        self.path = input
-        self.column = column.split(',') if column else ['']
+    def __init__(self, output_path: Optional[str], input_path: Optional[str], column: Optional[str]):
+        self.output_path = output_path
+        self.input_path = input_path
+        self.column = column.split(',') if column is not None else ['']
         self.is_multi_columns = len(self.column) > 1
 
         if self.is_multi_columns:
             self.column = [tuple(c.split('=')) if '=' in c else (c, c) for c in self.column]
 
-        if output is not None:
-            if exists(abspath(self.output)):
-                raise OSError('{} already exists on disk'.format(self.output))
+        if output_path is not None:
+            if exists(abspath(self.output_path)):
+                raise OSError('{} already exists on disk'.format(self.output_path))
 
-        if input is not None:
-            if not exists(abspath(self.path)):
-                raise OSError('{} doesnt exist on disk'.format(self.path))
+        if input_path is not None:
+            if not exists(abspath(self.input_path)):
+                raise OSError('{} doesnt exist on disk'.format(self.input_path))
 
     @abstractmethod
     def __iter__(self):
@@ -140,7 +143,7 @@ class PipelineDataFormat:
         :param data: data to store
         :return: (str) Path where the data has been saved
         """
-        path, _ = os.path.splitext(self.output)
+        path, _ = os.path.splitext(self.output_path)
         binary_path = os.path.extsep.join((path, 'pickle'))
 
         with open(binary_path, 'wb+') as f_output:
@@ -149,23 +152,23 @@ class PipelineDataFormat:
         return binary_path
 
     @staticmethod
-    def from_str(name: str, output: Optional[str], path: Optional[str], column: Optional[str]):
-        if name == 'json':
-            return JsonPipelineDataFormat(output, path, column)
-        elif name == 'csv':
-            return CsvPipelineDataFormat(output, path, column)
-        elif name == 'pipe':
-            return PipedPipelineDataFormat(output, path, column)
+    def from_str(format: str, output_path: Optional[str], input_path: Optional[str], column: Optional[str]):
+        if format == 'json':
+            return JsonPipelineDataFormat(output_path, input_path, column)
+        elif format == 'csv':
+            return CsvPipelineDataFormat(output_path, input_path, column)
+        elif format == 'pipe':
+            return PipedPipelineDataFormat(output_path, input_path, column)
         else:
-            raise KeyError('Unknown reader {} (Available reader are json/csv/pipe)'.format(name))
+            raise KeyError('Unknown reader {} (Available reader are json/csv/pipe)'.format(format))
 
 
 class CsvPipelineDataFormat(PipelineDataFormat):
-    def __init__(self, output: Optional[str], input: Optional[str], column: Optional[str]):
-        super().__init__(output, input, column)
+    def __init__(self, output_path: Optional[str], input_path: Optional[str], column: Optional[str]):
+        super().__init__(output_path, input_path, column)
 
     def __iter__(self):
-        with open(self.path, 'r') as f:
+        with open(self.input_path, 'r') as f:
             reader = csv.DictReader(f)
             for row in reader:
                 if self.is_multi_columns:
@@ -174,7 +177,7 @@ class CsvPipelineDataFormat(PipelineDataFormat):
                     yield row[self.column[0]]
 
     def save(self, data: List[dict]):
-        with open(self.output, 'w') as f:
+        with open(self.output_path, 'w') as f:
             if len(data) > 0:
                 writer = csv.DictWriter(f, list(data[0].keys()))
                 writer.writeheader()
@@ -182,10 +185,10 @@ class CsvPipelineDataFormat(PipelineDataFormat):
 
 
 class JsonPipelineDataFormat(PipelineDataFormat):
-    def __init__(self, output: Optional[str], input: Optional[str], column: Optional[str]):
-        super().__init__(output, input, column)
+    def __init__(self, output_path: Optional[str], input_path: Optional[str], column: Optional[str]):
+        super().__init__(output_path, input_path, column)
 
-        with open(input, 'r') as f:
+        with open(input_path, 'r') as f:
             self._entries = json.load(f)
 
     def __iter__(self):
@@ -196,7 +199,7 @@ class JsonPipelineDataFormat(PipelineDataFormat):
                 yield entry[self.column[0]]
 
     def save(self, data: dict):
-        with open(self.output, 'w') as f:
+        with open(self.output_path, 'w') as f:
             json.dump(data, f)
 
 
@@ -208,9 +211,7 @@ class PipedPipelineDataFormat(PipelineDataFormat):
     If columns are provided, then the output will be a dictionary with {column_x: value_x}
     """
     def __iter__(self):
-        import sys
         for line in sys.stdin:
-
             # Split for multi-columns
             if '\t' in line:
 
@@ -229,7 +230,7 @@ class PipedPipelineDataFormat(PipelineDataFormat):
         print(data)
 
     def save_binary(self, data: Union[dict, List[dict]]) -> str:
-        if self.output is None:
+        if self.output_path is None:
             raise KeyError(
                 'When using piped input on pipeline outputting large object requires an output file path. '
                 'Please provide such output path through --output argument.'
@@ -294,6 +295,9 @@ class Pipeline(_ScikitCompat):
         nlp = NerPipeline(model='...', config='...', tokenizer='...')
         nlp = QuestionAnsweringPipeline(model=AutoModel.from_pretrained('...'), tokenizer='...')
     """
+
+    default_input_names = None
+
     def __init__(self, model, tokenizer: PreTrainedTokenizer = None,
                  modelcard: ModelCard = None, framework: Optional[str] = None,
                  args_parser: ArgumentHandler = None, device: int = -1,
@@ -582,6 +586,8 @@ class QuestionAnsweringPipeline(Pipeline):
     Question Answering pipeline using ModelForQuestionAnswering head.
     """
 
+    default_input_names = 'question,context'
+
     def __init__(self, model,
                  tokenizer: Optional[PreTrainedTokenizer],
                  modelcard: Optional[ModelCard],
@@ -684,7 +690,6 @@ class QuestionAnsweringPipeline(Pipeline):
                 }
                 for s, e, score in zip(starts, ends, scores)
             ]
-
         if len(answers) == 1:
             return answers[0]
         return answers

From 79e4a6a25c9d9e267a7d7cdb54a97443fd4fc9c4 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 20 Dec 2019 19:33:12 +0100
Subject: [PATCH 478/505] update serving API

---
 transformers/commands/serving.py | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/transformers/commands/serving.py b/transformers/commands/serving.py
index 3c3f852809..4f41f797d1 100644
--- a/transformers/commands/serving.py
+++ b/transformers/commands/serving.py
@@ -24,7 +24,11 @@ def serve_command_factory(args: Namespace):
     Factory function used to instantiate serving server from provided command line arguments.
     :return: ServeCommand
     """
-    nlp = pipeline(task=args.task, model=args.model, config=args.config, tokenizer=args.tokenizer, device=args.device)
+    nlp = pipeline(task=args.task,
+                   model=args.model if args.model else None,
+                   config=args.config,
+                   tokenizer=args.tokenizer,
+                   device=args.device)
     return ServeCommand(nlp, args.host, args.port)
 
 
@@ -68,12 +72,12 @@ class ServeCommand(BaseTransformersCLICommand):
         """
         serve_parser = parser.add_parser('serve', help='CLI tool to run inference requests through REST and GraphQL endpoints.')
         serve_parser.add_argument('--task', type=str, choices=SUPPORTED_TASKS.keys(), help='The task to run the pipeline on')
-        serve_parser.add_argument('--device', type=int, default=-1, help='Indicate the device to run onto, -1 indicates CPU, >= 0 indicates GPU (default: -1)')
         serve_parser.add_argument('--host', type=str, default='localhost', help='Interface the server will listen on.')
         serve_parser.add_argument('--port', type=int, default=8888, help='Port the serving will listen to.')
-        serve_parser.add_argument('--model', type=str, required=True, help='Model\'s name or path to stored model.')
+        serve_parser.add_argument('--model', type=str, help='Model\'s name or path to stored model.')
         serve_parser.add_argument('--config', type=str, help='Model\'s config name or path to stored model.')
         serve_parser.add_argument('--tokenizer', type=str, help='Tokenizer name to use.')
+        serve_parser.add_argument('--device', type=int, default=-1, help='Indicate the device to run onto, -1 indicates CPU, >= 0 indicates GPU (default: -1)')
         serve_parser.set_defaults(func=serve_command_factory)
 
     def __init__(self, pipeline: Pipeline, host: str, port: int):

From 71883b6ddcd14929217a0ddf4ad627468b9ab5a8 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 20 Dec 2019 19:40:23 +0100
Subject: [PATCH 479/505] update link in readme

---
 README.md | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/README.md b/README.md
index 1312fcc0ac..769b0499cb 100644
--- a/README.md
+++ b/README.md
@@ -55,6 +55,7 @@ Choose the right framework for every part of a model's lifetime
 | [Online demo](#online-demo) | Experimenting with this repo’s text generation capabilities |
 | [Quick tour: Usage](#quick-tour) | Tokenizers & models usage: Bert and GPT-2 |
 | [Quick tour: TF 2.0 and PyTorch ](#Quick-tour-TF-20-training-and-PyTorch-interoperability) | Train a TF 2.0 model in 10 lines of code, load it in PyTorch |
+| [Quick tour: pipelines](#quick-tour-of-pipelines) | Using Pipelines: Wrapper around tokenizer and models to use finetuned models |
 | [Quick tour: Fine-tuning/usage scripts](#quick-tour-of-the-fine-tuningusage-scripts) | Using provided scripts: GLUE, SQuAD and Text generation |
 | [Quick tour: Share your models ](#Quick-tour-of-model-sharing) | Upload and share your fine-tuned models with the community |
 | [Migrating from pytorch-transformers to transformers](#Migrating-from-pytorch-transformers-to-transformers) | Migrating your code from pytorch-transformers to transformers |
@@ -496,6 +497,7 @@ New in version `v2.3`: `Pipeline` are high-level objects which automatically han
 and outputting the result in a structured object. 
 
 You can create `Pipeline` objects for the following down-stream tasks:
+
  - `feature-extraction`: Generates a tensor representation for the input sequence
  - `ner`: Generates named entity mapping for each word in the input sequence.
  - `sentiment-analysis`: Gives the polarity (positive / negative) of the whole input sequence. 

From ceae85ad60da38cacb14eca49f752669a4fe31dc Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 20 Dec 2019 19:52:24 +0100
Subject: [PATCH 480/505] fix mc loading

---
 transformers/pipelines.py | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 01491cf2be..7188526a62 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -891,6 +891,10 @@ def pipeline(task: str, model: Optional = None,
     if isinstance(config, str):
         config = AutoConfig.from_pretrained(config)
 
+    # Instantiate modelcard if needed
+    if isinstance(modelcard, str):
+        modelcard = ModelCard.from_pretrained(modelcard)
+
     # Instantiate model if needed
     if isinstance(model, str):
         # Handle transparent TF/PT model conversion

From e37ca8e11a3aa91e27ed659e6d4e01b208aa83ca Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 20 Dec 2019 20:43:42 +0100
Subject: [PATCH 481/505] fix camembert and XLM-R tokenizer

---
 transformers/tokenization_camembert.py   | 6 ++++++
 transformers/tokenization_xlm_roberta.py | 6 ++++++
 2 files changed, 12 insertions(+)

diff --git a/transformers/tokenization_camembert.py b/transformers/tokenization_camembert.py
index b4091558e1..4c4615eb3d 100644
--- a/transformers/tokenization_camembert.py
+++ b/transformers/tokenization_camembert.py
@@ -22,6 +22,7 @@ from shutil import copyfile
 
 import sentencepiece as spm
 from transformers.tokenization_utils import PreTrainedTokenizer
+from .tokenization_xlnet import SPIECE_UNDERLINE
 
 logger = logging.getLogger(__name__)
 
@@ -145,6 +146,11 @@ class CamembertTokenizer(PreTrainedTokenizer):
             return self.fairseq_ids_to_tokens[index]
         return self.sp_model.IdToPiece(index - self.fairseq_offset)
 
+    def convert_tokens_to_string(self, tokens):
+        """Converts a sequence of tokens (strings for sub-words) in a single string."""
+        out_string = ''.join(tokens).replace(SPIECE_UNDERLINE, ' ').strip()
+        return out_string
+
     def save_vocabulary(self, save_directory):
         """ Save the sentencepiece vocabulary (copy original file) and special tokens file
             to a directory.
diff --git a/transformers/tokenization_xlm_roberta.py b/transformers/tokenization_xlm_roberta.py
index 57a42dde5c..adbc8cd6c7 100644
--- a/transformers/tokenization_xlm_roberta.py
+++ b/transformers/tokenization_xlm_roberta.py
@@ -22,6 +22,7 @@ from shutil import copyfile
 
 import sentencepiece as spm
 from transformers.tokenization_utils import PreTrainedTokenizer
+from .tokenization_xlnet import SPIECE_UNDERLINE
 
 logger = logging.getLogger(__name__)
 
@@ -161,6 +162,11 @@ class XLMRobertaTokenizer(PreTrainedTokenizer):
             return self.fairseq_ids_to_tokens[index]
         return self.sp_model.IdToPiece(index - self.fairseq_offset)
 
+    def convert_tokens_to_string(self, tokens):
+        """Converts a sequence of tokens (strings for sub-words) in a single string."""
+        out_string = ''.join(tokens).replace(SPIECE_UNDERLINE, ' ').strip()
+        return out_string
+
     def save_vocabulary(self, save_directory):
         """ Save the sentencepiece vocabulary (copy original file) and special tokens file
             to a directory.

From a241011057245211975b4730170815536527d79d Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 20 Dec 2019 20:43:48 +0100
Subject: [PATCH 482/505] fix pipeline NER

---
 transformers/pipelines.py | 31 +++++++++++--------------------
 1 file changed, 11 insertions(+), 20 deletions(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 7188526a62..f7900feaf3 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -463,7 +463,7 @@ class NerPipeline(Pipeline):
     def __init__(self, model, tokenizer: PreTrainedTokenizer = None,
                  modelcard: ModelCard = None, framework: Optional[str] = None,
                  args_parser: ArgumentHandler = None, device: int = -1,
-                 binary_output: bool = False):
+                 binary_output: bool = False, ignore_labels=['O']):
         super().__init__(model=model,
                          tokenizer=tokenizer,
                          modelcard=modelcard,
@@ -473,17 +473,12 @@ class NerPipeline(Pipeline):
                          binary_output=binary_output)
 
         self._basic_tokenizer = BasicTokenizer(do_lower_case=False)
+        self.ignore_labels = ignore_labels
 
     def __call__(self, *texts, **kwargs):
         inputs, answers = self._args_parser(*texts, **kwargs), []
         for sentence in inputs:
 
-            # Ugly token to word idx mapping (for now)
-            token_to_word, words = [], self._basic_tokenizer.tokenize(sentence)
-            for i, w in enumerate(words):
-                tokens = self.tokenizer.tokenize(w)
-                token_to_word += [i] * len(tokens)
-
             # Manage correct placement of the tensors
             with self.device_placement():
 
@@ -500,26 +495,22 @@ class NerPipeline(Pipeline):
                     with torch.no_grad():
                         entities = self.model(**tokens)[0][0].cpu().numpy()
 
-            # Normalize scores
-            answer, token_start = [], 1
-            for idx, word in groupby(token_to_word):
+            score = np.exp(entities) / np.exp(entities).sum(-1, keepdims=True)
+            labels_idx = score.argmax(axis=-1)
 
-                # Sum log prob over token, then normalize across labels
-                score = np.exp(entities[token_start]) / np.exp(entities[token_start]).sum(-1, keepdims=True)
-                label_idx = score.argmax()
-
-                if label_idx > 0:
+            answer = []
+            for idx, label_idx in enumerate(labels_idx):
+                if self.model.config.id2label[label_idx] not in self.ignore_labels:
                     answer += [{
-                        'word': words[idx],
-                        'score': score[label_idx].item(),
+                        'word': self.tokenizer.decode(tokens['input_ids'][0][idx].cpu().tolist()),
+                        'score': score[idx][label_idx].item(),
                         'entity': self.model.config.id2label[label_idx]
                     }]
 
-                # Update token start
-                token_start += len(list(word))
-
             # Append
             answers += [answer]
+        if len(answers) == 1:
+            return answers[0]
         return answers
 
 

From f79a7dc661b9d7caee867af18a2be009478ab739 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 20 Dec 2019 20:57:45 +0100
Subject: [PATCH 483/505] fix NER pipeline

---
 transformers/pipelines.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index f7900feaf3..86fd25c164 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -491,9 +491,11 @@ class NerPipeline(Pipeline):
                 # Forward
                 if self.framework == 'tf':
                     entities = self.model(tokens)[0][0].numpy()
+                    input_ids = tokens['input_ids'].numpy()[0]
                 else:
                     with torch.no_grad():
                         entities = self.model(**tokens)[0][0].cpu().numpy()
+                        input_ids = tokens['input_ids'].cpu().numpy()[0]
 
             score = np.exp(entities) / np.exp(entities).sum(-1, keepdims=True)
             labels_idx = score.argmax(axis=-1)
@@ -502,7 +504,7 @@ class NerPipeline(Pipeline):
             for idx, label_idx in enumerate(labels_idx):
                 if self.model.config.id2label[label_idx] not in self.ignore_labels:
                     answer += [{
-                        'word': self.tokenizer.decode(tokens['input_ids'][0][idx].cpu().tolist()),
+                        'word': self.tokenizer.decode(int(input_ids[idx])),
                         'score': score[idx][label_idx].item(),
                         'entity': self.model.config.id2label[label_idx]
                     }]

From cb6d54bfdabfb1fe566a2c303fcb9f18505d9b10 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Fri, 20 Dec 2019 15:06:28 -0500
Subject: [PATCH 484/505] Numpy compatibility for sentence piece

convert to int earlier
---
 transformers/tokenization_utils.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/transformers/tokenization_utils.py b/transformers/tokenization_utils.py
index d77a7100ab..eda89f22fc 100644
--- a/transformers/tokenization_utils.py
+++ b/transformers/tokenization_utils.py
@@ -1227,6 +1227,7 @@ class PreTrainedTokenizer(object):
                 return self._convert_id_to_token(ids)
         tokens = []
         for index in ids:
+            index = int(index)
             if skip_special_tokens and index in self.all_special_ids:
                 continue
             if index in self.added_tokens_decoder:

From 4775ec354b20196b53894ffeb9af7622a39dd4fc Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 20 Dec 2019 21:47:15 +0100
Subject: [PATCH 485/505] add overwrite - fix ner decoding

---
 transformers/commands/run.py |  9 ++++++---
 transformers/pipelines.py    | 25 ++++++++++++++-----------
 2 files changed, 20 insertions(+), 14 deletions(-)

diff --git a/transformers/commands/run.py b/transformers/commands/run.py
index c2c141734b..2098d03413 100644
--- a/transformers/commands/run.py
+++ b/transformers/commands/run.py
@@ -32,7 +32,8 @@ def run_command_factory(args):
     reader = PipelineDataFormat.from_str(format=format,
                                          output_path=args.output,
                                          input_path=args.input,
-                                         column=args.column if args.column else nlp.default_input_names)
+                                         column=args.column if args.column else nlp.default_input_names,
+                                         overwrite=args.overwrite)
     return RunCommand(nlp, reader)
 
 
@@ -54,6 +55,7 @@ class RunCommand(BaseTransformersCLICommand):
         run_parser.add_argument('--column', type=str, help='Name of the column to use as input. (For multi columns input as QA use column1,columns2)')
         run_parser.add_argument('--format', type=str, default='infer', choices=PipelineDataFormat.SUPPORTED_FORMATS, help='Input format to read from')
         run_parser.add_argument('--device', type=int, default=-1, help='Indicate the device to run onto, -1 indicates CPU, >= 0 indicates GPU (default: -1)')
+        run_parser.add_argument('--overwrite', action='store_true', help='Allow overwriting the output file.')
         run_parser.set_defaults(func=run_command_factory)
 
     def run(self):
@@ -61,6 +63,7 @@ class RunCommand(BaseTransformersCLICommand):
 
         for entry in self._reader:
             output = nlp(**entry) if self._reader.is_multi_columns else nlp(entry)
+            print(output)
             if isinstance(output, dict):
                 outputs.append(output)
             else:
@@ -68,10 +71,10 @@ class RunCommand(BaseTransformersCLICommand):
 
         # Saving data
         if self._nlp.binary_output:
-            binary_path = self._reader.save_binary(output)
+            binary_path = self._reader.save_binary(outputs)
             logger.warning('Current pipeline requires output to be in binary format, saving at {}'.format(binary_path))
         else:
-            self._reader.save(output)
+            self._reader.save(outputs)
 
 
 
diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 86fd25c164..876bdd0c09 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -107,7 +107,7 @@ class PipelineDataFormat:
     """
     SUPPORTED_FORMATS = ['json', 'csv', 'pipe']
 
-    def __init__(self, output_path: Optional[str], input_path: Optional[str], column: Optional[str]):
+    def __init__(self, output_path: Optional[str], input_path: Optional[str], column: Optional[str], overwrite=False):
         self.output_path = output_path
         self.input_path = input_path
         self.column = column.split(',') if column is not None else ['']
@@ -116,7 +116,7 @@ class PipelineDataFormat:
         if self.is_multi_columns:
             self.column = [tuple(c.split('=')) if '=' in c else (c, c) for c in self.column]
 
-        if output_path is not None:
+        if output_path is not None and not overwrite:
             if exists(abspath(self.output_path)):
                 raise OSError('{} already exists on disk'.format(self.output_path))
 
@@ -152,25 +152,26 @@ class PipelineDataFormat:
         return binary_path
 
     @staticmethod
-    def from_str(format: str, output_path: Optional[str], input_path: Optional[str], column: Optional[str]):
+    def from_str(format: str, output_path: Optional[str], input_path: Optional[str], column: Optional[str], overwrite=False):
         if format == 'json':
-            return JsonPipelineDataFormat(output_path, input_path, column)
+            return JsonPipelineDataFormat(output_path, input_path, column, overwrite=overwrite)
         elif format == 'csv':
-            return CsvPipelineDataFormat(output_path, input_path, column)
+            return CsvPipelineDataFormat(output_path, input_path, column, overwrite=overwrite)
         elif format == 'pipe':
-            return PipedPipelineDataFormat(output_path, input_path, column)
+            return PipedPipelineDataFormat(output_path, input_path, column, overwrite=overwrite)
         else:
             raise KeyError('Unknown reader {} (Available reader are json/csv/pipe)'.format(format))
 
 
 class CsvPipelineDataFormat(PipelineDataFormat):
-    def __init__(self, output_path: Optional[str], input_path: Optional[str], column: Optional[str]):
-        super().__init__(output_path, input_path, column)
+    def __init__(self, output_path: Optional[str], input_path: Optional[str], column: Optional[str], overwrite=False):
+        super().__init__(output_path, input_path, column, overwrite=overwrite)
 
     def __iter__(self):
         with open(self.input_path, 'r') as f:
             reader = csv.DictReader(f)
             for row in reader:
+                print(row, self.column)
                 if self.is_multi_columns:
                     yield {k: row[c] for k, c in self.column}
                 else:
@@ -185,8 +186,8 @@ class CsvPipelineDataFormat(PipelineDataFormat):
 
 
 class JsonPipelineDataFormat(PipelineDataFormat):
-    def __init__(self, output_path: Optional[str], input_path: Optional[str], column: Optional[str]):
-        super().__init__(output_path, input_path, column)
+    def __init__(self, output_path: Optional[str], input_path: Optional[str], column: Optional[str], overwrite=False):
+        super().__init__(output_path, input_path, column, overwrite=overwrite)
 
         with open(input_path, 'r') as f:
             self._entries = json.load(f)
@@ -460,6 +461,8 @@ class NerPipeline(Pipeline):
     Named Entity Recognition pipeline using ModelForTokenClassification head.
     """
 
+    default_input_names = 'sequences'
+
     def __init__(self, model, tokenizer: PreTrainedTokenizer = None,
                  modelcard: ModelCard = None, framework: Optional[str] = None,
                  args_parser: ArgumentHandler = None, device: int = -1,
@@ -504,7 +507,7 @@ class NerPipeline(Pipeline):
             for idx, label_idx in enumerate(labels_idx):
                 if self.model.config.id2label[label_idx] not in self.ignore_labels:
                     answer += [{
-                        'word': self.tokenizer.decode(int(input_ids[idx])),
+                        'word': self.tokenizer.decode([int(input_ids[idx])]),
                         'score': score[idx][label_idx].item(),
                         'entity': self.model.config.id2label[label_idx]
                     }]

From e5812462fc0f81e9808ad87a818cd8af26405722 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 20 Dec 2019 21:51:48 +0100
Subject: [PATCH 486/505] clean up debug and less verbose tqdm

---
 transformers/file_utils.py | 3 ++-
 transformers/pipelines.py  | 1 -
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/transformers/file_utils.py b/transformers/file_utils.py
index 59938868ea..032a6af63b 100644
--- a/transformers/file_utils.py
+++ b/transformers/file_utils.py
@@ -287,7 +287,8 @@ def http_get(url, temp_file, proxies=None, resume_size=0, user_agent=None):
         return
     content_length = response.headers.get('Content-Length')
     total = resume_size + int(content_length) if content_length is not None else None
-    progress = tqdm(unit="B", unit_scale=True, total=total, initial=resume_size, desc="Downloading")
+    progress = tqdm(unit="B", unit_scale=True, total=total, initial=resume_size,
+                    desc="Downloading", disable=bool(logger.level<=logging.INFO))
     for chunk in response.iter_content(chunk_size=1024):
         if chunk: # filter out keep-alive new chunks
             progress.update(len(chunk))
diff --git a/transformers/pipelines.py b/transformers/pipelines.py
index 876bdd0c09..f4bf3da685 100755
--- a/transformers/pipelines.py
+++ b/transformers/pipelines.py
@@ -171,7 +171,6 @@ class CsvPipelineDataFormat(PipelineDataFormat):
         with open(self.input_path, 'r') as f:
             reader = csv.DictReader(f)
             for row in reader:
-                print(row, self.column)
                 if self.is_multi_columns:
                     yield {k: row[c] for k, c in self.column}
                 else:

From 655fd068534b2a66b85f0bd05002e27f212ab6a0 Mon Sep 17 00:00:00 2001
From: thomwolf <thomwolf@gmail.com>
Date: Fri, 20 Dec 2019 21:57:49 +0100
Subject: [PATCH 487/505] clean up

---
 transformers/commands/run.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/transformers/commands/run.py b/transformers/commands/run.py
index 2098d03413..df03cee9d7 100644
--- a/transformers/commands/run.py
+++ b/transformers/commands/run.py
@@ -63,7 +63,6 @@ class RunCommand(BaseTransformersCLICommand):
 
         for entry in self._reader:
             output = nlp(**entry) if self._reader.is_multi_columns else nlp(entry)
-            print(output)
             if isinstance(output, dict):
                 outputs.append(output)
             else:

From a436574bfde4f75f518a107f45f987579d813ce5 Mon Sep 17 00:00:00 2001
From: Lysandre <lysandre.debut@reseau.eseo.fr>
Date: Fri, 20 Dec 2019 16:22:20 -0500
Subject: [PATCH 488/505] Release: v2.3.0

---
 README.md                | 2 +-
 docs/source/conf.py      | 2 +-
 setup.py                 | 2 +-
 transformers/__init__.py | 2 +-
 try.py                   | 0
 5 files changed, 4 insertions(+), 4 deletions(-)
 create mode 100644 try.py

diff --git a/README.md b/README.md
index 769b0499cb..416adcc1ef 100644
--- a/README.md
+++ b/README.md
@@ -60,7 +60,7 @@ Choose the right framework for every part of a model's lifetime
 | [Quick tour: Share your models ](#Quick-tour-of-model-sharing) | Upload and share your fine-tuned models with the community |
 | [Migrating from pytorch-transformers to transformers](#Migrating-from-pytorch-transformers-to-transformers) | Migrating your code from pytorch-transformers to transformers |
 | [Migrating from pytorch-pretrained-bert to pytorch-transformers](#Migrating-from-pytorch-pretrained-bert-to-transformers) | Migrating your code from pytorch-pretrained-bert to transformers |
-| [Documentation][(v2.2.0/v2.2.1/v2.2.2)](https://huggingface.co/transformers/v2.2.0) [(v2.1.1)](https://huggingface.co/transformers/v2.1.1) [(v2.0.0)](https://huggingface.co/transformers/v2.0.0) [(v1.2.0)](https://huggingface.co/transformers/v1.2.0) [(v1.1.0)](https://huggingface.co/transformers/v1.1.0) [(v1.0.0)](https://huggingface.co/transformers/v1.0.0) [(master)](https://huggingface.co/transformers) | Full API documentation and more |
+| [Documentation][(v2.3.0)](https://huggingface.co/transformers/v2.3.0)[(v2.2.0/v2.2.1/v2.2.2)](https://huggingface.co/transformers/v2.2.0) [(v2.1.1)](https://huggingface.co/transformers/v2.1.1) [(v2.0.0)](https://huggingface.co/transformers/v2.0.0) [(v1.2.0)](https://huggingface.co/transformers/v1.2.0) [(v1.1.0)](https://huggingface.co/transformers/v1.1.0) [(v1.0.0)](https://huggingface.co/transformers/v1.0.0) [(master)](https://huggingface.co/transformers) | Full API documentation and more |
 
 ## Installation
 
diff --git a/docs/source/conf.py b/docs/source/conf.py
index 99b7b44922..41a65eec29 100644
--- a/docs/source/conf.py
+++ b/docs/source/conf.py
@@ -26,7 +26,7 @@ author = u'huggingface'
 # The short X.Y version
 version = u''
 # The full version, including alpha/beta/rc tags
-release = u'2.2.2'
+release = u'2.3.0'
 
 
 # -- General configuration ---------------------------------------------------
diff --git a/setup.py b/setup.py
index 4bfb774155..cd64a6ce90 100644
--- a/setup.py
+++ b/setup.py
@@ -46,7 +46,7 @@ extras['all'] = [package for package in extras.values()]
 
 setup(
     name="transformers",
-    version="2.2.2",
+    version="2.3.0",
     author="Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond, Google AI Language Team Authors, Open AI team Authors, Facebook AI Authors, Carnegie Mellon University Authors",
     author_email="thomas@huggingface.co",
     description="State-of-the-art Natural Language Processing for TensorFlow 2.0 and PyTorch",
diff --git a/transformers/__init__.py b/transformers/__init__.py
index 1622c3892d..c0c0901df4 100755
--- a/transformers/__init__.py
+++ b/transformers/__init__.py
@@ -1,4 +1,4 @@
-__version__ = "2.2.2"
+__version__ = "2.3.0"
 
 # Work around to update TensorFlow's absl.logging threshold which alters the
 # default Python logging output behavior when present.
diff --git a/try.py b/try.py
new file mode 100644
index 0000000000..e69de29bb2

From 3df1d2d144dae698b4aed085aec57a2033d608b7 Mon Sep 17 00:00:00 2001
From: Francesco <francesco.cottone@vidiemme.it>
Date: Tue, 17 Dec 2019 10:19:54 +0100
Subject: [PATCH 489/505] - Create the output directory (whose name is passed
 by the user in the "save_directory" parameter) where it will be saved encoder
 and decoder, if not exists. - Empty the output directory, if it contains any
 files or subdirectories. - Create the "encoder" directory inside
 "save_directory", if not exists. - Create the "decoder" directory inside
 "save_directory", if not exists. - Save the encoder and the decoder in the
 previous two directories, respectively.

---
 transformers/modeling_encoder_decoder.py | 31 ++++++++++++++++++++++++
 1 file changed, 31 insertions(+)

diff --git a/transformers/modeling_encoder_decoder.py b/transformers/modeling_encoder_decoder.py
index a91c046d8f..8ae6daa690 100644
--- a/transformers/modeling_encoder_decoder.py
+++ b/transformers/modeling_encoder_decoder.py
@@ -166,6 +166,37 @@ class PreTrainedEncoderDecoder(nn.Module):
 
         We save the encoder' and decoder's parameters in two separate directories.
         """
+
+        # If the root output directory does not exist, create it 
+        if not os.path.exists(save_directory):
+            os.mkdir(save_directory)
+
+        # Check whether the output directory is empty or not
+        sub_directories = [directory for directory in os.listdir(save_directory)
+            if os.path.isdir(os.path.join(save_directory, directory))]
+
+        if len(sub_directories) > 0:
+            if "encoder" in sub_directories and "decoder" in sub_directories:
+                print("WARNING: there is an older version of encoder-decoder saved in" +\
+                    " the output directory. The default behaviour is to overwrite them.")
+
+            # Empty the output directory
+            for directory_to_remove in sub_directories:
+                # Remove all files into the subdirectory
+                files_to_remove = os.listdir(os.path.join(save_directory, directory_to_remove))
+                for file_to_remove in files_to_remove:
+                    os.remove(os.path.join(save_directory, directory_to_remove, file_to_remove))
+                # Remove the subdirectory itself
+                os.rmdir(os.path.join(save_directory, directory_to_remove))
+
+            assert(len(os.listdir(save_directory)) == 0) # sanity check
+
+        if not os.path.exists(os.path.join(save_directory, "encoder")):
+            os.mkdir(os.path.join(save_directory, "encoder"))
+
+        if not os.path.exists(os.path.join(save_directory, "decoder")):
+            os.mkdir(os.path.join(save_directory, "decoder"))
+
         self.encoder.save_pretrained(os.path.join(save_directory, "encoder"))
         self.decoder.save_pretrained(os.path.join(save_directory, "decoder"))
 

From a80778f40e4738071b5d01420a0328bb00cdb356 Mon Sep 17 00:00:00 2001
From: Francesco <francesco.cottone@vidiemme.it>
Date: Wed, 18 Dec 2019 16:05:28 +0100
Subject: [PATCH 490/505] small refactoring (only esthetic, not functional)

---
 transformers/modeling_encoder_decoder.py | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/transformers/modeling_encoder_decoder.py b/transformers/modeling_encoder_decoder.py
index 8ae6daa690..c327bb9199 100644
--- a/transformers/modeling_encoder_decoder.py
+++ b/transformers/modeling_encoder_decoder.py
@@ -191,13 +191,14 @@ class PreTrainedEncoderDecoder(nn.Module):
 
             assert(len(os.listdir(save_directory)) == 0) # sanity check
 
+        # Create the "encoder" directory inside the output directory and save the encoder into it
         if not os.path.exists(os.path.join(save_directory, "encoder")):
             os.mkdir(os.path.join(save_directory, "encoder"))
+        self.encoder.save_pretrained(os.path.join(save_directory, "encoder"))
 
+        # Create the "encoder" directory inside the output directory and save the decoder into it
         if not os.path.exists(os.path.join(save_directory, "decoder")):
             os.mkdir(os.path.join(save_directory, "decoder"))
-
-        self.encoder.save_pretrained(os.path.join(save_directory, "encoder"))
         self.decoder.save_pretrained(os.path.join(save_directory, "decoder"))
 
     def forward(self, encoder_input_ids, decoder_input_ids, **kwargs):

From 228f52867c92e21c6e7223eb2d6c7d9904b230e2 Mon Sep 17 00:00:00 2001
From: Dom Hudson <dom.hudson@thoughtriver.com>
Date: Thu, 7 Nov 2019 19:58:17 +0000
Subject: [PATCH 491/505] Bug fix: 1764

---
 transformers/modeling_roberta.py            | 45 +++++++++++++++------
 transformers/tests/modeling_roberta_test.py | 41 +++++++++++++++++++
 2 files changed, 74 insertions(+), 12 deletions(-)

diff --git a/transformers/modeling_roberta.py b/transformers/modeling_roberta.py
index fc27353d37..cf74c1e7b5 100644
--- a/transformers/modeling_roberta.py
+++ b/transformers/modeling_roberta.py
@@ -51,24 +51,45 @@ class RobertaEmbeddings(BertEmbeddings):
                                                 padding_idx=self.padding_idx)
 
     def forward(self, input_ids=None, token_type_ids=None, position_ids=None, inputs_embeds=None):
-        if input_ids is not None:
-            input_shape = input_ids.size()
-        else:
-            input_shape = inputs_embeds.size()[:-1]
-
-        seq_length = input_shape[1]
-        device = input_ids.device if input_ids is not None else inputs_embeds.device
-
         if position_ids is None:
-            # Position numbers begin at padding_idx+1. Padding symbols are ignored.
-            # cf. fairseq's `utils.make_positions`
-            position_ids = torch.arange(self.padding_idx+1, seq_length+self.padding_idx+1, dtype=torch.long, device=device)
-            position_ids = position_ids.unsqueeze(0).expand(input_shape)
+
+            if input_ids is not None:
+                # Create the position ids from the input token ids. Any padded tokens remain padded.
+                position_ids = self.create_position_ids_from_input_ids(input_ids).to(input_ids.device)
+            else:
+                position_ids = self.create_position_ids_from_inputs_embeds(inputs_embeds)
+
         return super(RobertaEmbeddings, self).forward(input_ids,
                                                       token_type_ids=token_type_ids,
                                                       position_ids=position_ids,
                                                       inputs_embeds=inputs_embeds)
 
+    def create_position_ids_from_input_ids(self, x):
+        """ Replace non-padding symbols with their position numbers. Position numbers begin at
+        padding_idx+1. Padding symbols are ignored. This is modified from fairseq's
+        `utils.make_positions`.
+
+        :param torch.Tensor x:
+        :return torch.Tensor:
+        """
+        mask = x.ne(self.padding_idx).long()
+        incremental_indicies = torch.cumsum(mask, dim=1) * mask
+        return incremental_indicies + self.padding_idx
+
+    def create_position_ids_from_inputs_embeds(self, inputs_embeds):
+        """ We are provided embeddings directly. We cannot infer which are padded so just generate
+        sequential position ids.
+
+        :param torch.Tensor inputs_embeds:
+        :return torch.Tensor:
+        """
+        input_shape = inputs_embeds.size()[:-1]
+        sequence_length = input_shape[1]
+
+        position_ids = torch.arange(self.padding_idx+1, sequence_length+self.padding_idx+1, dtype=torch.long,
+                                    device=inputs_embeds.device)
+        return position_ids.unsqueeze(0)
+
 
 ROBERTA_START_DOCSTRING = r"""    The RoBERTa model was proposed in
     `RoBERTa: A Robustly Optimized BERT Pretraining Approach`_
diff --git a/transformers/tests/modeling_roberta_test.py b/transformers/tests/modeling_roberta_test.py
index 4d34a50528..121cb84148 100644
--- a/transformers/tests/modeling_roberta_test.py
+++ b/transformers/tests/modeling_roberta_test.py
@@ -25,6 +25,7 @@ if is_torch_available():
     import torch
     from transformers import (RobertaConfig, RobertaModel, RobertaForMaskedLM,
                               RobertaForSequenceClassification, RobertaForTokenClassification)
+    from transformers.modeling_roberta import RobertaEmbeddings
     from transformers.modeling_roberta import ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP
 
 from .modeling_common_test import (CommonTestCases, ids_tensor)
@@ -205,6 +206,46 @@ class RobertaModelTest(CommonTestCases.CommonModelTester):
             shutil.rmtree(cache_dir)
             self.assertIsNotNone(model)
 
+    def test_create_position_ids_respects_padding_index(self):
+        """ Ensure that the default position ids only assign a sequential . This is a regression
+        test for https://github.com/huggingface/transformers/issues/1761
+
+        The position ids should be masked with the embedding object's padding index. Therefore, the
+        first available non-padding position index is RobertaEmbeddings.padding_idx + 1
+        """
+        config = self.model_tester.prepare_config_and_inputs()[0]
+        model = RobertaEmbeddings(config=config)
+
+        input_ids = torch.as_tensor([[12, 31, 13, model.padding_idx]])
+        expected_positions = torch.as_tensor([[
+            0 + model.padding_idx + 1,
+            1 + model.padding_idx + 1,
+            2 + model.padding_idx + 1,
+            model.padding_idx
+        ]])
+
+        position_ids = model.create_position_ids_from_input_ids(input_ids)
+        self.assertTrue(torch.all(torch.eq(position_ids, expected_positions)))
+
+    def test_create_position_ids_from_inputs_embeds(self):
+        """ Ensure that the default position ids only assign a sequential . This is a regression
+        test for https://github.com/huggingface/transformers/issues/1761
+
+        The position ids should be masked with the embedding object's padding index. Therefore, the
+        first available non-padding position index is RobertaEmbeddings.padding_idx + 1
+        """
+        config = self.model_tester.prepare_config_and_inputs()[0]
+        model = RobertaEmbeddings(config=config)
+
+        input_ids = torch.Tensor(1, 4, 30)
+        expected_positions = torch.as_tensor([[
+            0 + model.padding_idx + 1,
+            1 + model.padding_idx + 1,
+            2 + model.padding_idx + 1,
+            3 + model.padding_idx + 1,
+        ]])
+        position_ids = model.create_position_ids_from_inputs_embeds(input_ids)
+        self.assertTrue(torch.all(torch.eq(position_ids, expected_positions)))
 
 
 class RobertaModelIntegrationTest(unittest.TestCase):

From 3e52915fa7106a739aa6f9feda9937961ce25068 Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Fri, 20 Dec 2019 19:01:27 -0500
Subject: [PATCH 492/505] [RoBERTa] Embeddings: fix dimensionality bug

---
 transformers/modeling_roberta.py            |  3 +-
 transformers/tests/modeling_roberta_test.py | 31 ++++++++++++++-------
 2 files changed, 22 insertions(+), 12 deletions(-)

diff --git a/transformers/modeling_roberta.py b/transformers/modeling_roberta.py
index cf74c1e7b5..b188799522 100644
--- a/transformers/modeling_roberta.py
+++ b/transformers/modeling_roberta.py
@@ -52,7 +52,6 @@ class RobertaEmbeddings(BertEmbeddings):
 
     def forward(self, input_ids=None, token_type_ids=None, position_ids=None, inputs_embeds=None):
         if position_ids is None:
-
             if input_ids is not None:
                 # Create the position ids from the input token ids. Any padded tokens remain padded.
                 position_ids = self.create_position_ids_from_input_ids(input_ids).to(input_ids.device)
@@ -88,7 +87,7 @@ class RobertaEmbeddings(BertEmbeddings):
 
         position_ids = torch.arange(self.padding_idx+1, sequence_length+self.padding_idx+1, dtype=torch.long,
                                     device=inputs_embeds.device)
-        return position_ids.unsqueeze(0)
+        return position_ids.unsqueeze(0).expand(input_shape)
 
 
 ROBERTA_START_DOCSTRING = r"""    The RoBERTa model was proposed in
diff --git a/transformers/tests/modeling_roberta_test.py b/transformers/tests/modeling_roberta_test.py
index 121cb84148..fe6ffe98c6 100644
--- a/transformers/tests/modeling_roberta_test.py
+++ b/transformers/tests/modeling_roberta_test.py
@@ -225,6 +225,10 @@ class RobertaModelTest(CommonTestCases.CommonModelTester):
         ]])
 
         position_ids = model.create_position_ids_from_input_ids(input_ids)
+        self.assertEqual(
+            position_ids.shape,
+            expected_positions.shape
+        )
         self.assertTrue(torch.all(torch.eq(position_ids, expected_positions)))
 
     def test_create_position_ids_from_inputs_embeds(self):
@@ -235,17 +239,24 @@ class RobertaModelTest(CommonTestCases.CommonModelTester):
         first available non-padding position index is RobertaEmbeddings.padding_idx + 1
         """
         config = self.model_tester.prepare_config_and_inputs()[0]
-        model = RobertaEmbeddings(config=config)
+        embeddings = RobertaEmbeddings(config=config)
 
-        input_ids = torch.Tensor(1, 4, 30)
-        expected_positions = torch.as_tensor([[
-            0 + model.padding_idx + 1,
-            1 + model.padding_idx + 1,
-            2 + model.padding_idx + 1,
-            3 + model.padding_idx + 1,
-        ]])
-        position_ids = model.create_position_ids_from_inputs_embeds(input_ids)
-        self.assertTrue(torch.all(torch.eq(position_ids, expected_positions)))
+        inputs_embeds = torch.Tensor(2, 4, 30)
+        expected_single_positions = [
+            0 + embeddings.padding_idx + 1,
+            1 + embeddings.padding_idx + 1,
+            2 + embeddings.padding_idx + 1,
+            3 + embeddings.padding_idx + 1,
+        ]
+        expected_positions = torch.as_tensor([expected_single_positions, expected_single_positions])
+        position_ids = embeddings.create_position_ids_from_inputs_embeds(inputs_embeds)
+        self.assertEqual(
+            position_ids.shape,
+            expected_positions.shape
+        )
+        self.assertTrue(
+            torch.all(torch.eq(position_ids, expected_positions))
+        )
 
 
 class RobertaModelIntegrationTest(unittest.TestCase):

From ac1b449cc938bb34bc9021feff599cfd3b2376ae Mon Sep 17 00:00:00 2001
From: Julien Chaumond <chaumond@gmail.com>
Date: Sat, 21 Dec 2019 00:09:01 -0500
Subject: [PATCH 493/505] [doc] move distilroberta to more appropriate place

cc @lysandrejik
---
 docs/source/pretrained_models.rst | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/docs/source/pretrained_models.rst b/docs/source/pretrained_models.rst
index a359990f5a..eb7b41ffc9 100644
--- a/docs/source/pretrained_models.rst
+++ b/docs/source/pretrained_models.rst
@@ -3,6 +3,7 @@ Pretrained models
 
 Here is the full list of the currently provided pretrained models together with a short presentation of each model.
 
+For a list that includes community-uploaded models, refer to `https://huggingface.co/models <https://huggingface.co/models>`__.
 
 +-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 | Architecture      | Shortcut name                                              | Details of the model                                                                                                                  |
@@ -154,6 +155,10 @@ Here is the full list of the currently provided pretrained models together with
 |                   |                                                            | | ``roberta-large`` fine-tuned on `MNLI <http://www.nyu.edu/projects/bowman/multinli/>`__.                                            |
 |                   |                                                            | (see `details <https://github.com/pytorch/fairseq/tree/master/examples/roberta>`__)                                                   |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   | ``distilroberta-base``                                     | | 6-layer, 768-hidden, 12-heads, 82M parameters                                                                                       |
+|                   |                                                            | | The DistilRoBERTa model distilled from the RoBERTa model `roberta-base` checkpoint.                                                 |
+|                   |                                                            | (see `details <https://github.com/huggingface/transformers/tree/master/examples/distillation>`__)                                     |
+|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``roberta-base-openai-detector``                           | | 12-layer, 768-hidden, 12-heads, 125M parameters                                                                                     |
 |                   |                                                            | | ``roberta-base`` fine-tuned by OpenAI on the outputs of the 1.5B-parameter GPT-2 model.                                             |
 |                   |                                                            | (see `details <https://github.com/openai/gpt-2-output-dataset/tree/master/detector>`__)                                               |
@@ -174,10 +179,6 @@ Here is the full list of the currently provided pretrained models together with
 |                   |                                                            | | The DistilGPT2 model distilled from the GPT2 model `gpt2` checkpoint.                                                               |
 |                   |                                                            | (see `details <https://github.com/huggingface/transformers/tree/master/examples/distillation>`__)                                     |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
-|                   | ``distilroberta-base``                                     | | 6-layer, 768-hidden, 12-heads, 82M parameters                                                                                       |
-|                   |                                                            | | The DistilRoBERTa model distilled from the RoBERTa model `roberta-base` checkpoint.                                                 |
-|                   |                                                            | (see `details <https://github.com/huggingface/transformers/tree/master/examples/distillation>`__)                                     |
-|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``distilbert-base-german-cased``                           | | 6-layer, 768-hidden, 12-heads, 66M parameters                                                                                       |
 |                   |                                                            | | The German DistilBERT model distilled from the German DBMDZ BERT model `bert-base-german-dbmdz-cased` checkpoint.                   |
 |                   |                                                            | (see `details <https://github.com/huggingface/transformers/tree/master/examples/distillation>`__)                                     |

From 12726f8556152dbc6c115327646ebb33ccb2bc4f Mon Sep 17 00:00:00 2001
From: Aymeric Augustin <aymeric.augustin@fractalideas.com>
Date: Fri, 20 Dec 2019 20:56:58 +0100
Subject: [PATCH 494/505] Remove redundant torch.jit.trace in tests.

This looks like it could be expensive, so don't run it twice.
---
 transformers/tests/modeling_common_test.py | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/transformers/tests/modeling_common_test.py b/transformers/tests/modeling_common_test.py
index c84162117a..c03d307e71 100644
--- a/transformers/tests/modeling_common_test.py
+++ b/transformers/tests/modeling_common_test.py
@@ -218,12 +218,11 @@ class CommonTestCases:
                 inputs = inputs_dict['input_ids']  # Let's keep only input_ids
 
                 try:
-                    torch.jit.trace(model, inputs)
+                    traced_gpt2 = torch.jit.trace(model, inputs)
                 except RuntimeError:
                     self.fail("Couldn't trace module.")
 
                 try:
-                    traced_gpt2 = torch.jit.trace(model, inputs)
                     torch.jit.save(traced_gpt2, "traced_model.pt")
                 except RuntimeError:
                     self.fail("Couldn't save module.")

From 478e456e8392be1356a795a354215ba7dbf03a7b Mon Sep 17 00:00:00 2001
From: Aymeric Augustin <aymeric.augustin@fractalideas.com>
Date: Fri, 20 Dec 2019 20:56:58 +0100
Subject: [PATCH 495/505] Use a random temp dir for writing file in tests.

---
 transformers/tests/modeling_common_test.py | 22 ++++++++++++----------
 1 file changed, 12 insertions(+), 10 deletions(-)

diff --git a/transformers/tests/modeling_common_test.py b/transformers/tests/modeling_common_test.py
index c03d307e71..8bf66c3582 100644
--- a/transformers/tests/modeling_common_test.py
+++ b/transformers/tests/modeling_common_test.py
@@ -18,7 +18,7 @@ from __future__ import print_function
 
 import copy
 import sys
-import os
+import os.path
 import shutil
 import tempfile
 import json
@@ -222,16 +222,18 @@ class CommonTestCases:
                 except RuntimeError:
                     self.fail("Couldn't trace module.")
 
-                try:
-                    torch.jit.save(traced_gpt2, "traced_model.pt")
-                except RuntimeError:
-                    self.fail("Couldn't save module.")
+                with TemporaryDirectory() as tmp_dir_name:
+                    pt_file_name = os.path.join(tmp_dir_name, "traced_model.pt")
 
-                try:
-                    loaded_model = torch.jit.load("traced_model.pt")
-                    os.remove("traced_model.pt")
-                except ValueError:
-                    self.fail("Couldn't load module.")
+                    try:
+                        torch.jit.save(traced_gpt2, pt_file_name)
+                    except Exception:
+                        self.fail("Couldn't save module.")
+
+                    try:
+                        loaded_model = torch.jit.load(pt_file_name)
+                    except Exception:
+                        self.fail("Couldn't load module.")
 
                 model.to(torch_device)
                 model.eval()

From 286d5bb6b7afb1fcca1923d431f42c716f53a290 Mon Sep 17 00:00:00 2001
From: Aymeric Augustin <aymeric.augustin@fractalideas.com>
Date: Fri, 20 Dec 2019 20:56:58 +0100
Subject: [PATCH 496/505] Use a random temp dir for writing pruned models in
 tests.

---
 transformers/tests/modeling_common_test.py | 24 ++++++++--------------
 1 file changed, 9 insertions(+), 15 deletions(-)

diff --git a/transformers/tests/modeling_common_test.py b/transformers/tests/modeling_common_test.py
index 8bf66c3582..cf36332207 100644
--- a/transformers/tests/modeling_common_test.py
+++ b/transformers/tests/modeling_common_test.py
@@ -353,12 +353,11 @@ class CommonTestCases:
                 heads_to_prune = {0: list(range(1, self.model_tester.num_attention_heads)),
                                 -1: [0]}
                 model.prune_heads(heads_to_prune)
-                directory = "pruned_model"
-                if not os.path.exists(directory):
-                    os.makedirs(directory)
-                model.save_pretrained(directory)
-                model = model_class.from_pretrained(directory)
-                model.to(torch_device)
+
+                with TemporaryDirectory() as temp_dir_name:
+                    model.save_pretrained(temp_dir_name)
+                    model = model_class.from_pretrained(temp_dir_name)
+                    model.to(torch_device)
 
                 with torch.no_grad():
                     outputs = model(**inputs_dict)
@@ -367,7 +366,6 @@ class CommonTestCases:
                 self.assertEqual(attentions[1].shape[-3], self.model_tester.num_attention_heads)
                 self.assertEqual(attentions[-1].shape[-3], self.model_tester.num_attention_heads - 1)
 
-                shutil.rmtree(directory)
 
         def test_head_pruning_save_load_from_config_init(self):
             if not self.test_pruning:
@@ -427,14 +425,10 @@ class CommonTestCases:
                 self.assertEqual(attentions[2].shape[-3], self.model_tester.num_attention_heads)
                 self.assertEqual(attentions[3].shape[-3], self.model_tester.num_attention_heads)
 
-                directory = "pruned_model"
-
-                if not os.path.exists(directory):
-                    os.makedirs(directory)
-                model.save_pretrained(directory)
-                model = model_class.from_pretrained(directory)
-                model.to(torch_device)
-                shutil.rmtree(directory)
+                with TemporaryDirectory() as temp_dir_name:
+                    model.save_pretrained(temp_dir_name)
+                    model = model_class.from_pretrained(temp_dir_name)
+                    model.to(torch_device)
 
                 with torch.no_grad():
                     outputs = model(**inputs_dict)

From b67fa1a8d2302d808ecb9d95355181eaf21ee3b6 Mon Sep 17 00:00:00 2001
From: Aymeric Augustin <aymeric.augustin@fractalideas.com>
Date: Fri, 20 Dec 2019 20:56:58 +0100
Subject: [PATCH 497/505] Download models directly to cache_dir.

This allows moving the file instead of copying it, which is more
reliable. Also it avoids writing large amounts of data to /tmp,
which may not be large enough to accomodate it.

Refs #2222.
---
 transformers/file_utils.py | 15 ++++-----------
 1 file changed, 4 insertions(+), 11 deletions(-)

diff --git a/transformers/file_utils.py b/transformers/file_utils.py
index 032a6af63b..61ff1d00bc 100644
--- a/transformers/file_utils.py
+++ b/transformers/file_utils.py
@@ -10,10 +10,9 @@ import json
 import logging
 import os
 import six
-import shutil
 import tempfile
 import fnmatch
-from functools import wraps
+from functools import partial, wraps
 from hashlib import sha256
 from io import open
 
@@ -345,14 +344,13 @@ def get_from_cache(url, cache_dir=None, force_download=False, proxies=None, etag
         def _resumable_file_manager():
             with open(incomplete_path,'a+b') as f:
                 yield f
-            os.remove(incomplete_path)
         temp_file_manager = _resumable_file_manager
         if os.path.exists(incomplete_path):
             resume_size = os.stat(incomplete_path).st_size
         else:
             resume_size = 0
     else:
-        temp_file_manager = tempfile.NamedTemporaryFile
+        temp_file_manager = partial(tempfile.NamedTemporaryFile, dir=cache_dir, delete=False)
         resume_size = 0
 
     if etag is not None and (not os.path.exists(cache_path) or force_download):
@@ -371,12 +369,9 @@ def get_from_cache(url, cache_dir=None, force_download=False, proxies=None, etag
 
             # we are copying the file before closing it, so flush to avoid truncation
             temp_file.flush()
-            # shutil.copyfileobj() starts at the current position, so go to the start
-            temp_file.seek(0)
 
-            logger.info("copying %s to cache at %s", temp_file.name, cache_path)
-            with open(cache_path, 'wb') as cache_file:
-                shutil.copyfileobj(temp_file, cache_file)
+            logger.info("storing %s in cache at %s", url, cache_path)
+            os.rename(temp_file.name, cache_path)
 
             logger.info("creating metadata file for %s", cache_path)
             meta = {'url': url, 'etag': etag}
@@ -387,6 +382,4 @@ def get_from_cache(url, cache_dir=None, force_download=False, proxies=None, etag
                     output_string = unicode(output_string, 'utf-8')  # The beauty of python 2
                 meta_file.write(output_string)
 
-            logger.info("removing temp file %s", temp_file.name)
-
     return cache_path

From b670c2668426326aeffe626aabac7ee2dff3c7c2 Mon Sep 17 00:00:00 2001
From: Aymeric Augustin <aymeric.augustin@fractalideas.com>
Date: Fri, 20 Dec 2019 20:56:58 +0100
Subject: [PATCH 498/505] Take advantage of the cache when running tests.

Caching models across test cases and across runs of the test suite makes
slow tests somewhat more bearable.

Use gettempdir() instead of /tmp in tests. This makes it easier to
change the location of the cache with semi-standard TMPDIR/TEMP/TMP
environment variables.

Fix #2222.
---
 .../tests/modeling_tf_xxx_test.py              |  7 ++-----
 .../tests/modeling_xxx_test.py                 |  7 ++-----
 transformers/tests/modeling_albert_test.py     |  7 ++-----
 transformers/tests/modeling_bert_test.py       |  7 ++-----
 transformers/tests/modeling_common_test.py     |  6 ++----
 transformers/tests/modeling_ctrl_test.py       |  7 ++-----
 transformers/tests/modeling_distilbert_test.py |  6 ++----
 transformers/tests/modeling_gpt2_test.py       |  7 ++-----
 transformers/tests/modeling_openai_test.py     |  7 ++-----
 transformers/tests/modeling_roberta_test.py    |  7 ++-----
 transformers/tests/modeling_t5_test.py         |  7 ++-----
 transformers/tests/modeling_tf_albert_test.py  |  8 ++------
 transformers/tests/modeling_tf_auto_test.py    | 18 +++++++++---------
 transformers/tests/modeling_tf_bert_test.py    |  7 ++-----
 transformers/tests/modeling_tf_ctrl_test.py    |  7 ++-----
 .../tests/modeling_tf_distilbert_test.py       |  6 ++----
 transformers/tests/modeling_tf_gpt2_test.py    |  7 ++-----
 .../tests/modeling_tf_openai_gpt_test.py       |  7 ++-----
 transformers/tests/modeling_tf_roberta_test.py |  7 ++-----
 transformers/tests/modeling_tf_t5_test.py      |  7 ++-----
 .../tests/modeling_tf_transfo_xl_test.py       |  7 ++-----
 transformers/tests/modeling_tf_xlm_test.py     |  7 ++-----
 transformers/tests/modeling_tf_xlnet_test.py   |  7 ++-----
 transformers/tests/modeling_transfo_xl_test.py |  7 ++-----
 transformers/tests/modeling_xlm_test.py        |  7 ++-----
 transformers/tests/modeling_xlnet_test.py      |  7 ++-----
 transformers/tests/utils.py                    |  3 +++
 27 files changed, 62 insertions(+), 132 deletions(-)

diff --git a/templates/adding_a_new_model/tests/modeling_tf_xxx_test.py b/templates/adding_a_new_model/tests/modeling_tf_xxx_test.py
index 912a4aa340..6eba932a8e 100644
--- a/templates/adding_a_new_model/tests/modeling_tf_xxx_test.py
+++ b/templates/adding_a_new_model/tests/modeling_tf_xxx_test.py
@@ -17,12 +17,11 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import shutil
 import sys
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_tf, slow
+from .utils import CACHE_DIR, require_tf, slow
 
 from transformers import XxxConfig, is_tf_available
 
@@ -245,10 +244,8 @@ class TFXxxModelTest(TFCommonTestCases.TFCommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        cache_dir = "/tmp/transformers_test/"
         for model_name in ['xxx-base-uncased']:
-            model = TFXxxModel.from_pretrained(model_name, cache_dir=cache_dir)
-            shutil.rmtree(cache_dir)
+            model = TFXxxModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 if __name__ == "__main__":
diff --git a/templates/adding_a_new_model/tests/modeling_xxx_test.py b/templates/adding_a_new_model/tests/modeling_xxx_test.py
index 30e614b3f2..5e22392d00 100644
--- a/templates/adding_a_new_model/tests/modeling_xxx_test.py
+++ b/templates/adding_a_new_model/tests/modeling_xxx_test.py
@@ -17,13 +17,12 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import shutil
 
 from transformers import is_torch_available
 
 from .modeling_common_test import (CommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_torch, slow, torch_device
+from .utils import CACHE_DIR, require_torch, slow, torch_device
 
 if is_torch_available():
     from transformers import (XxxConfig, XxxModel, XxxForMaskedLM,
@@ -249,10 +248,8 @@ class XxxModelTest(CommonTestCases.CommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        cache_dir = "/tmp/transformers_test/"
         for model_name in list(XXX_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-            model = XxxModel.from_pretrained(model_name, cache_dir=cache_dir)
-            shutil.rmtree(cache_dir)
+            model = XxxModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 if __name__ == "__main__":
diff --git a/transformers/tests/modeling_albert_test.py b/transformers/tests/modeling_albert_test.py
index 1911d244e7..b726fd9278 100644
--- a/transformers/tests/modeling_albert_test.py
+++ b/transformers/tests/modeling_albert_test.py
@@ -17,13 +17,12 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import shutil
 
 from transformers import is_torch_available
 
 from .modeling_common_test import (CommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_torch, slow, torch_device
+from .utils import CACHE_DIR, require_torch, slow, torch_device
 
 if is_torch_available():
     from transformers import (AlbertConfig, AlbertModel, AlbertForMaskedLM,
@@ -230,10 +229,8 @@ class AlbertModelTest(CommonTestCases.CommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        cache_dir = "/tmp/transformers_test/"
         for model_name in list(ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-            model = AlbertModel.from_pretrained(model_name, cache_dir=cache_dir)
-            shutil.rmtree(cache_dir)
+            model = AlbertModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 if __name__ == "__main__":
diff --git a/transformers/tests/modeling_bert_test.py b/transformers/tests/modeling_bert_test.py
index 0eb7bc9a14..a5adff8f68 100644
--- a/transformers/tests/modeling_bert_test.py
+++ b/transformers/tests/modeling_bert_test.py
@@ -17,13 +17,12 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import shutil
 
 from transformers import is_torch_available
 
 from .modeling_common_test import (CommonTestCases, ids_tensor, floats_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_torch, slow, torch_device
+from .utils import CACHE_DIR, require_torch, slow, torch_device
 
 if is_torch_available():
     from transformers import (BertConfig, BertModel, BertForMaskedLM,
@@ -360,10 +359,8 @@ class BertModelTest(CommonTestCases.CommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        cache_dir = "/tmp/transformers_test/"
         for model_name in list(BERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-            model = BertModel.from_pretrained(model_name, cache_dir=cache_dir)
-            shutil.rmtree(cache_dir)
+            model = BertModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 
diff --git a/transformers/tests/modeling_common_test.py b/transformers/tests/modeling_common_test.py
index cf36332207..2116651f4a 100644
--- a/transformers/tests/modeling_common_test.py
+++ b/transformers/tests/modeling_common_test.py
@@ -30,7 +30,7 @@ import logging
 
 from transformers import is_torch_available
 
-from .utils import require_torch, slow, torch_device
+from .utils import CACHE_DIR, require_torch, slow, torch_device
 
 if is_torch_available():
     import torch
@@ -753,10 +753,8 @@ class CommonTestCases:
                 [[], []])
 
         def create_and_check_model_from_pretrained(self):
-            cache_dir = "/tmp/transformers_test/"
             for model_name in list(self.base_model_class.pretrained_model_archive_map.keys())[:1]:
-                model = self.base_model_class.from_pretrained(model_name, cache_dir=cache_dir)
-                shutil.rmtree(cache_dir)
+                model = self.base_model_class.from_pretrained(model_name, cache_dir=CACHE_DIR)
                 self.parent.assertIsNotNone(model)
 
         def prepare_config_and_inputs_for_common(self):
diff --git a/transformers/tests/modeling_ctrl_test.py b/transformers/tests/modeling_ctrl_test.py
index c7de49b2ab..ed0d62d1e6 100644
--- a/transformers/tests/modeling_ctrl_test.py
+++ b/transformers/tests/modeling_ctrl_test.py
@@ -16,7 +16,6 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import shutil
 import pdb
 
 from transformers import is_torch_available
@@ -27,7 +26,7 @@ if is_torch_available():
 
 from .modeling_common_test import (CommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_torch, slow, torch_device
+from .utils import CACHE_DIR, require_torch, slow, torch_device
 
 
 @require_torch
@@ -205,10 +204,8 @@ class CTRLModelTest(CommonTestCases.CommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        cache_dir = "/tmp/transformers_test/"
         for model_name in list(CTRL_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-            model = CTRLModel.from_pretrained(model_name, cache_dir=cache_dir)
-            shutil.rmtree(cache_dir)
+            model = CTRLModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 
diff --git a/transformers/tests/modeling_distilbert_test.py b/transformers/tests/modeling_distilbert_test.py
index 82f71c40da..ac6f5d248e 100644
--- a/transformers/tests/modeling_distilbert_test.py
+++ b/transformers/tests/modeling_distilbert_test.py
@@ -27,7 +27,7 @@ if is_torch_available():
 
 from .modeling_common_test import (CommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_torch, slow, torch_device
+from .utils import CACHE_DIR, require_torch, slow, torch_device
 
 
 @require_torch
@@ -235,10 +235,8 @@ class DistilBertModelTest(CommonTestCases.CommonModelTester):
 
     # @slow
     # def test_model_from_pretrained(self):
-    #     cache_dir = "/tmp/transformers_test/"
     #     for model_name in list(DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-    #         model = DistilBertModel.from_pretrained(model_name, cache_dir=cache_dir)
-    #         shutil.rmtree(cache_dir)
+    #         model = DistilBertModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
     #         self.assertIsNotNone(model)
 
 if __name__ == "__main__":
diff --git a/transformers/tests/modeling_gpt2_test.py b/transformers/tests/modeling_gpt2_test.py
index a82e39c261..ad2ec1fd91 100644
--- a/transformers/tests/modeling_gpt2_test.py
+++ b/transformers/tests/modeling_gpt2_test.py
@@ -17,7 +17,6 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import shutil
 
 from transformers import is_torch_available
 
@@ -27,7 +26,7 @@ if is_torch_available():
 
 from .modeling_common_test import (CommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_torch, slow, torch_device
+from .utils import CACHE_DIR, require_torch, slow, torch_device
 
 
 @require_torch
@@ -239,10 +238,8 @@ class GPT2ModelTest(CommonTestCases.CommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        cache_dir = "/tmp/transformers_test/"
         for model_name in list(GPT2_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-            model = GPT2Model.from_pretrained(model_name, cache_dir=cache_dir)
-            shutil.rmtree(cache_dir)
+            model = GPT2Model.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 
diff --git a/transformers/tests/modeling_openai_test.py b/transformers/tests/modeling_openai_test.py
index 7655e432e8..1880febcae 100644
--- a/transformers/tests/modeling_openai_test.py
+++ b/transformers/tests/modeling_openai_test.py
@@ -17,7 +17,6 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import shutil
 
 from transformers import is_torch_available
 
@@ -27,7 +26,7 @@ if is_torch_available():
 
 from .modeling_common_test import (CommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_torch, slow, torch_device
+from .utils import CACHE_DIR, require_torch, slow, torch_device
 
 
 @require_torch
@@ -207,10 +206,8 @@ class OpenAIGPTModelTest(CommonTestCases.CommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        cache_dir = "/tmp/transformers_test/"
         for model_name in list(OPENAI_GPT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-            model = OpenAIGPTModel.from_pretrained(model_name, cache_dir=cache_dir)
-            shutil.rmtree(cache_dir)
+            model = OpenAIGPTModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 
diff --git a/transformers/tests/modeling_roberta_test.py b/transformers/tests/modeling_roberta_test.py
index 4d34a50528..299cbd01ad 100644
--- a/transformers/tests/modeling_roberta_test.py
+++ b/transformers/tests/modeling_roberta_test.py
@@ -17,7 +17,6 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import shutil
 
 from transformers import is_torch_available
 
@@ -29,7 +28,7 @@ if is_torch_available():
 
 from .modeling_common_test import (CommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_torch, slow, torch_device
+from .utils import CACHE_DIR, require_torch, slow, torch_device
 
 
 @require_torch
@@ -199,10 +198,8 @@ class RobertaModelTest(CommonTestCases.CommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        cache_dir = "/tmp/transformers_test/"
         for model_name in list(ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-            model = RobertaModel.from_pretrained(model_name, cache_dir=cache_dir)
-            shutil.rmtree(cache_dir)
+            model = RobertaModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 
diff --git a/transformers/tests/modeling_t5_test.py b/transformers/tests/modeling_t5_test.py
index c337163375..9fd9a4b304 100644
--- a/transformers/tests/modeling_t5_test.py
+++ b/transformers/tests/modeling_t5_test.py
@@ -17,13 +17,12 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import shutil
 
 from transformers import is_torch_available
 
 from .modeling_common_test import (CommonTestCases, ids_tensor, floats_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_torch, slow, torch_device
+from .utils import CACHE_DIR, require_torch, slow, torch_device
 
 if is_torch_available():
     from transformers import (T5Config, T5Model, T5WithLMHeadModel)
@@ -175,10 +174,8 @@ class T5ModelTest(CommonTestCases.CommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        cache_dir = "/tmp/transformers_test/"
         for model_name in list(T5_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-            model = T5Model.from_pretrained(model_name, cache_dir=cache_dir)
-            shutil.rmtree(cache_dir)
+            model = T5Model.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 if __name__ == "__main__":
diff --git a/transformers/tests/modeling_tf_albert_test.py b/transformers/tests/modeling_tf_albert_test.py
index 93aeab66c2..ee71371a18 100644
--- a/transformers/tests/modeling_tf_albert_test.py
+++ b/transformers/tests/modeling_tf_albert_test.py
@@ -17,12 +17,11 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import shutil
 import sys
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_tf, slow
+from .utils import CACHE_DIR, require_tf, slow
 
 from transformers import AlbertConfig, is_tf_available
 
@@ -217,12 +216,9 @@ class TFAlbertModelTest(TFCommonTestCases.TFCommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        cache_dir = "/tmp/transformers_test/"
         # for model_name in list(TF_ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
         for model_name in ['albert-base-uncased']:
-            model = TFAlbertModel.from_pretrained(
-                model_name, cache_dir=cache_dir)
-            shutil.rmtree(cache_dir)
+            model = TFAlbertModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 
diff --git a/transformers/tests/modeling_tf_auto_test.py b/transformers/tests/modeling_tf_auto_test.py
index 7ab6eaa3d6..2ad39ddccf 100644
--- a/transformers/tests/modeling_tf_auto_test.py
+++ b/transformers/tests/modeling_tf_auto_test.py
@@ -46,11 +46,11 @@ class TFAutoModelTest(unittest.TestCase):
         logging.basicConfig(level=logging.INFO)
         # for model_name in list(TF_BERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
         for model_name in ['bert-base-uncased']:
-            config = AutoConfig.from_pretrained(model_name, force_download=True)
+            config = AutoConfig.from_pretrained(model_name)
             self.assertIsNotNone(config)
             self.assertIsInstance(config, BertConfig)
 
-            model = TFAutoModel.from_pretrained(model_name, force_download=True)
+            model = TFAutoModel.from_pretrained(model_name)
             self.assertIsNotNone(model)
             self.assertIsInstance(model, TFBertModel)
 
@@ -59,11 +59,11 @@ class TFAutoModelTest(unittest.TestCase):
         logging.basicConfig(level=logging.INFO)
         # for model_name in list(TF_BERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
         for model_name in ['bert-base-uncased']:
-            config = AutoConfig.from_pretrained(model_name, force_download=True)
+            config = AutoConfig.from_pretrained(model_name)
             self.assertIsNotNone(config)
             self.assertIsInstance(config, BertConfig)
 
-            model = TFAutoModelWithLMHead.from_pretrained(model_name, force_download=True)
+            model = TFAutoModelWithLMHead.from_pretrained(model_name)
             self.assertIsNotNone(model)
             self.assertIsInstance(model, TFBertForMaskedLM)
 
@@ -72,11 +72,11 @@ class TFAutoModelTest(unittest.TestCase):
         logging.basicConfig(level=logging.INFO)
         # for model_name in list(TF_BERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
         for model_name in ['bert-base-uncased']:
-            config = AutoConfig.from_pretrained(model_name, force_download=True)
+            config = AutoConfig.from_pretrained(model_name)
             self.assertIsNotNone(config)
             self.assertIsInstance(config, BertConfig)
 
-            model = TFAutoModelForSequenceClassification.from_pretrained(model_name, force_download=True)
+            model = TFAutoModelForSequenceClassification.from_pretrained(model_name)
             self.assertIsNotNone(model)
             self.assertIsInstance(model, TFBertForSequenceClassification)
 
@@ -85,17 +85,17 @@ class TFAutoModelTest(unittest.TestCase):
         logging.basicConfig(level=logging.INFO)
         # for model_name in list(TF_BERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
         for model_name in ['bert-base-uncased']:
-            config = AutoConfig.from_pretrained(model_name, force_download=True)
+            config = AutoConfig.from_pretrained(model_name)
             self.assertIsNotNone(config)
             self.assertIsInstance(config, BertConfig)
 
-            model = TFAutoModelForQuestionAnswering.from_pretrained(model_name, force_download=True)
+            model = TFAutoModelForQuestionAnswering.from_pretrained(model_name)
             self.assertIsNotNone(model)
             self.assertIsInstance(model, TFBertForQuestionAnswering)
 
     def test_from_pretrained_identifier(self):
         logging.basicConfig(level=logging.INFO)
-        model = TFAutoModelWithLMHead.from_pretrained(SMALL_MODEL_IDENTIFIER, force_download=True)
+        model = TFAutoModelWithLMHead.from_pretrained(SMALL_MODEL_IDENTIFIER)
         self.assertIsInstance(model, TFBertForMaskedLM)
 
 
diff --git a/transformers/tests/modeling_tf_bert_test.py b/transformers/tests/modeling_tf_bert_test.py
index 20073e1ab8..abf20b1514 100644
--- a/transformers/tests/modeling_tf_bert_test.py
+++ b/transformers/tests/modeling_tf_bert_test.py
@@ -17,12 +17,11 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import shutil
 import sys
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_tf, slow
+from .utils import CACHE_DIR, require_tf, slow
 
 from transformers import BertConfig, is_tf_available
 
@@ -310,11 +309,9 @@ class TFBertModelTest(TFCommonTestCases.TFCommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        cache_dir = "/tmp/transformers_test/"
         # for model_name in list(TF_BERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
         for model_name in ['bert-base-uncased']:
-            model = TFBertModel.from_pretrained(model_name, cache_dir=cache_dir)
-            shutil.rmtree(cache_dir)
+            model = TFBertModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 if __name__ == "__main__":
diff --git a/transformers/tests/modeling_tf_ctrl_test.py b/transformers/tests/modeling_tf_ctrl_test.py
index 0876582e57..93b231e517 100644
--- a/transformers/tests/modeling_tf_ctrl_test.py
+++ b/transformers/tests/modeling_tf_ctrl_test.py
@@ -17,12 +17,11 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import shutil
 import sys
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_tf, slow
+from .utils import CACHE_DIR, require_tf, slow
 
 from transformers import CTRLConfig, is_tf_available
 
@@ -189,10 +188,8 @@ class TFCTRLModelTest(TFCommonTestCases.TFCommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        cache_dir = "/tmp/transformers_test/"
         for model_name in list(TF_CTRL_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-            model = TFCTRLModel.from_pretrained(model_name, cache_dir=cache_dir)
-            shutil.rmtree(cache_dir)
+            model = TFCTRLModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 if __name__ == "__main__":
diff --git a/transformers/tests/modeling_tf_distilbert_test.py b/transformers/tests/modeling_tf_distilbert_test.py
index d9e971c2a5..f28b5c397b 100644
--- a/transformers/tests/modeling_tf_distilbert_test.py
+++ b/transformers/tests/modeling_tf_distilbert_test.py
@@ -20,7 +20,7 @@ import unittest
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_tf, slow
+from .utils import CACHE_DIR, require_tf, slow
 
 from transformers import DistilBertConfig, is_tf_available
 
@@ -211,10 +211,8 @@ class TFDistilBertModelTest(TFCommonTestCases.TFCommonModelTester):
 
     # @slow
     # def test_model_from_pretrained(self):
-    #     cache_dir = "/tmp/transformers_test/"
     #     for model_name in list(DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-    #         model = DistilBertModel.from_pretrained(model_name, cache_dir=cache_dir)
-    #         shutil.rmtree(cache_dir)
+    #         model = DistilBertModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
     #         self.assertIsNotNone(model)
 
 if __name__ == "__main__":
diff --git a/transformers/tests/modeling_tf_gpt2_test.py b/transformers/tests/modeling_tf_gpt2_test.py
index 3f30b32787..90920342ba 100644
--- a/transformers/tests/modeling_tf_gpt2_test.py
+++ b/transformers/tests/modeling_tf_gpt2_test.py
@@ -17,12 +17,11 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import shutil
 import sys
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_tf, slow
+from .utils import CACHE_DIR, require_tf, slow
 
 from transformers import GPT2Config, is_tf_available
 
@@ -220,10 +219,8 @@ class TFGPT2ModelTest(TFCommonTestCases.TFCommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        cache_dir = "/tmp/transformers_test/"
         for model_name in list(TF_GPT2_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-            model = TFGPT2Model.from_pretrained(model_name, cache_dir=cache_dir)
-            shutil.rmtree(cache_dir)
+            model = TFGPT2Model.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 if __name__ == "__main__":
diff --git a/transformers/tests/modeling_tf_openai_gpt_test.py b/transformers/tests/modeling_tf_openai_gpt_test.py
index 863dbf1bc0..065bf2acde 100644
--- a/transformers/tests/modeling_tf_openai_gpt_test.py
+++ b/transformers/tests/modeling_tf_openai_gpt_test.py
@@ -17,12 +17,11 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import shutil
 import sys
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_tf, slow
+from .utils import CACHE_DIR, require_tf, slow
 
 from transformers import OpenAIGPTConfig, is_tf_available
 
@@ -219,10 +218,8 @@ class TFOpenAIGPTModelTest(TFCommonTestCases.TFCommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        cache_dir = "/tmp/transformers_test/"
         for model_name in list(TF_OPENAI_GPT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-            model = TFOpenAIGPTModel.from_pretrained(model_name, cache_dir=cache_dir)
-            shutil.rmtree(cache_dir)
+            model = TFOpenAIGPTModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 if __name__ == "__main__":
diff --git a/transformers/tests/modeling_tf_roberta_test.py b/transformers/tests/modeling_tf_roberta_test.py
index f4ed97c44b..93c478ae28 100644
--- a/transformers/tests/modeling_tf_roberta_test.py
+++ b/transformers/tests/modeling_tf_roberta_test.py
@@ -17,11 +17,10 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import shutil
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_tf, slow
+from .utils import CACHE_DIR, require_tf, slow
 
 from transformers import RobertaConfig, is_tf_available
 
@@ -192,10 +191,8 @@ class TFRobertaModelTest(TFCommonTestCases.TFCommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        cache_dir = "/tmp/transformers_test/"
         for model_name in list(TF_ROBERTA_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-            model = TFRobertaModel.from_pretrained(model_name, cache_dir=cache_dir)
-            shutil.rmtree(cache_dir)
+            model = TFRobertaModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 
diff --git a/transformers/tests/modeling_tf_t5_test.py b/transformers/tests/modeling_tf_t5_test.py
index b905a9875b..da9ce6f89d 100644
--- a/transformers/tests/modeling_tf_t5_test.py
+++ b/transformers/tests/modeling_tf_t5_test.py
@@ -17,12 +17,11 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import shutil
 import sys
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_tf, slow
+from .utils import CACHE_DIR, require_tf, slow
 
 from transformers import T5Config, is_tf_available
 
@@ -162,10 +161,8 @@ class TFT5ModelTest(TFCommonTestCases.TFCommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        cache_dir = "/tmp/transformers_test/"
         for model_name in ['t5-small']:
-            model = TFT5Model.from_pretrained(model_name, cache_dir=cache_dir)
-            shutil.rmtree(cache_dir)
+            model = TFT5Model.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 if __name__ == "__main__":
diff --git a/transformers/tests/modeling_tf_transfo_xl_test.py b/transformers/tests/modeling_tf_transfo_xl_test.py
index 746a6a1321..8225c09275 100644
--- a/transformers/tests/modeling_tf_transfo_xl_test.py
+++ b/transformers/tests/modeling_tf_transfo_xl_test.py
@@ -18,11 +18,10 @@ from __future__ import print_function
 
 import unittest
 import random
-import shutil
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_tf, slow
+from .utils import CACHE_DIR, require_tf, slow
 
 from transformers import TransfoXLConfig, is_tf_available
 
@@ -205,10 +204,8 @@ class TFTransfoXLModelTest(TFCommonTestCases.TFCommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        cache_dir = "/tmp/transformers_test/"
         for model_name in list(TF_TRANSFO_XL_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-            model = TFTransfoXLModel.from_pretrained(model_name, cache_dir=cache_dir)
-            shutil.rmtree(cache_dir)
+            model = TFTransfoXLModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 
diff --git a/transformers/tests/modeling_tf_xlm_test.py b/transformers/tests/modeling_tf_xlm_test.py
index 228e436149..8b5ab6d742 100644
--- a/transformers/tests/modeling_tf_xlm_test.py
+++ b/transformers/tests/modeling_tf_xlm_test.py
@@ -17,7 +17,6 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import shutil
 
 from transformers import is_tf_available
 
@@ -31,7 +30,7 @@ if is_tf_available():
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_tf, slow
+from .utils import CACHE_DIR, require_tf, slow
 
 
 @require_tf
@@ -252,10 +251,8 @@ class TFXLMModelTest(TFCommonTestCases.TFCommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        cache_dir = "/tmp/transformers_test/"
         for model_name in list(TF_XLM_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-            model = XLMModel.from_pretrained(model_name, cache_dir=cache_dir)
-            shutil.rmtree(cache_dir)
+            model = XLMModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 
diff --git a/transformers/tests/modeling_tf_xlnet_test.py b/transformers/tests/modeling_tf_xlnet_test.py
index eb66d92793..15fd917481 100644
--- a/transformers/tests/modeling_tf_xlnet_test.py
+++ b/transformers/tests/modeling_tf_xlnet_test.py
@@ -20,7 +20,6 @@ import os
 import unittest
 import json
 import random
-import shutil
 
 from transformers import XLNetConfig, is_tf_available
 
@@ -35,7 +34,7 @@ if is_tf_available():
 
 from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_tf, slow
+from .utils import CACHE_DIR, require_tf, slow
 
 
 @require_tf
@@ -319,10 +318,8 @@ class TFXLNetModelTest(TFCommonTestCases.TFCommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        cache_dir = "/tmp/transformers_test/"
         for model_name in list(TF_XLNET_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-            model = TFXLNetModel.from_pretrained(model_name, cache_dir=cache_dir)
-            shutil.rmtree(cache_dir)
+            model = TFXLNetModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 
diff --git a/transformers/tests/modeling_transfo_xl_test.py b/transformers/tests/modeling_transfo_xl_test.py
index f41d50a3a0..acbe95fe4a 100644
--- a/transformers/tests/modeling_transfo_xl_test.py
+++ b/transformers/tests/modeling_transfo_xl_test.py
@@ -18,7 +18,6 @@ from __future__ import print_function
 
 import unittest
 import random
-import shutil
 
 from transformers import is_torch_available
 
@@ -29,7 +28,7 @@ if is_torch_available():
 
 from .modeling_common_test import (CommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_torch, slow, torch_device
+from .utils import CACHE_DIR, require_torch, slow, torch_device
 
 
 @require_torch
@@ -208,10 +207,8 @@ class TransfoXLModelTest(CommonTestCases.CommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        cache_dir = "/tmp/transformers_test/"
         for model_name in list(TRANSFO_XL_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-            model = TransfoXLModel.from_pretrained(model_name, cache_dir=cache_dir)
-            shutil.rmtree(cache_dir)
+            model = TransfoXLModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 
diff --git a/transformers/tests/modeling_xlm_test.py b/transformers/tests/modeling_xlm_test.py
index 7cae6c848e..fcc2f4699b 100644
--- a/transformers/tests/modeling_xlm_test.py
+++ b/transformers/tests/modeling_xlm_test.py
@@ -17,7 +17,6 @@ from __future__ import division
 from __future__ import print_function
 
 import unittest
-import shutil
 
 from transformers import is_torch_available
 
@@ -28,7 +27,7 @@ if is_torch_available():
 
 from .modeling_common_test import (CommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_torch, slow, torch_device
+from .utils import CACHE_DIR, require_torch, slow, torch_device
 
 
 @require_torch
@@ -318,10 +317,8 @@ class XLMModelTest(CommonTestCases.CommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        cache_dir = "/tmp/transformers_test/"
         for model_name in list(XLM_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-            model = XLMModel.from_pretrained(model_name, cache_dir=cache_dir)
-            shutil.rmtree(cache_dir)
+            model = XLMModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 
diff --git a/transformers/tests/modeling_xlnet_test.py b/transformers/tests/modeling_xlnet_test.py
index 6d901ee699..6d218d6ef4 100644
--- a/transformers/tests/modeling_xlnet_test.py
+++ b/transformers/tests/modeling_xlnet_test.py
@@ -20,7 +20,6 @@ import os
 import unittest
 import json
 import random
-import shutil
 
 from transformers import is_torch_available
 
@@ -33,7 +32,7 @@ if is_torch_available():
 
 from .modeling_common_test import (CommonTestCases, ids_tensor)
 from .configuration_common_test import ConfigTester
-from .utils import require_torch, slow, torch_device
+from .utils import CACHE_DIR, require_torch, slow, torch_device
 
 
 @require_torch
@@ -385,10 +384,8 @@ class XLNetModelTest(CommonTestCases.CommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        cache_dir = "/tmp/transformers_test/"
         for model_name in list(XLNET_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-            model = XLNetModel.from_pretrained(model_name, cache_dir=cache_dir)
-            shutil.rmtree(cache_dir)
+            model = XLNetModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 
diff --git a/transformers/tests/utils.py b/transformers/tests/utils.py
index c950ad8f17..ba0e19f420 100644
--- a/transformers/tests/utils.py
+++ b/transformers/tests/utils.py
@@ -1,11 +1,14 @@
 import os
 import unittest
+import tempfile
 
 from distutils.util import strtobool
 
 from transformers.file_utils import _tf_available, _torch_available
 
 
+CACHE_DIR = os.path.join(tempfile.gettempdir(), "transformers_test")
+
 SMALL_MODEL_IDENTIFIER = "julien-c/bert-xsmall-dummy"
 
 

From a4c9338b83ba612b5f5aec645f375d048d9a7647 Mon Sep 17 00:00:00 2001
From: Aymeric Augustin <aymeric.augustin@fractalideas.com>
Date: Fri, 20 Dec 2019 20:56:59 +0100
Subject: [PATCH 499/505] Prevent parallel downloads of the same file with a
 lock.

Since the file is written to the filesystem, a filesystem lock is the
way to go here. Add a dependency on the third-party filelock library to
get cross-platform functionality.
---
 setup.py                   |  1 +
 transformers/file_utils.py | 89 +++++++++++++++++++++-----------------
 2 files changed, 50 insertions(+), 40 deletions(-)

diff --git a/setup.py b/setup.py
index cd64a6ce90..fe2e1526bf 100644
--- a/setup.py
+++ b/setup.py
@@ -59,6 +59,7 @@ setup(
                                     "tests.*", "tests"]),
     install_requires=['numpy',
                       'boto3',
+                      'filelock',
                       'requests',
                       'tqdm',
                       'regex != 2019.12.17',
diff --git a/transformers/file_utils.py b/transformers/file_utils.py
index 61ff1d00bc..ec925c6160 100644
--- a/transformers/file_utils.py
+++ b/transformers/file_utils.py
@@ -24,6 +24,8 @@ from tqdm.auto import tqdm
 from contextlib import contextmanager
 from . import __version__
 
+from filelock import FileLock
+
 logger = logging.getLogger(__name__)  # pylint: disable=invalid-name
 
 try:
@@ -333,53 +335,60 @@ def get_from_cache(url, cache_dir=None, force_download=False, proxies=None, etag
     # If we don't have a connection (etag is None) and can't identify the file
     # try to get the last downloaded one
     if not os.path.exists(cache_path) and etag is None:
-        matching_files = fnmatch.filter(os.listdir(cache_dir), filename + '.*')
-        matching_files = list(filter(lambda s: not s.endswith('.json'), matching_files))
+        matching_files = [
+            file
+            for file in fnmatch.filter(os.listdir(cache_dir), filename + '.*')
+            if not file.endswith('.json') and not file.endswith('.lock')
+        ]
         if matching_files:
             cache_path = os.path.join(cache_dir, matching_files[-1])
 
-    if resume_download:
-        incomplete_path = cache_path + '.incomplete'
-        @contextmanager
-        def _resumable_file_manager():
-            with open(incomplete_path,'a+b') as f:
-                yield f
-        temp_file_manager = _resumable_file_manager
-        if os.path.exists(incomplete_path):
-            resume_size = os.stat(incomplete_path).st_size
-        else:
-            resume_size = 0
-    else:
-        temp_file_manager = partial(tempfile.NamedTemporaryFile, dir=cache_dir, delete=False)
-        resume_size = 0
+    # Prevent parallel downloads of the same file with a lock.
+    lock_path = cache_path + '.lock'
+    with FileLock(lock_path):
 
-    if etag is not None and (not os.path.exists(cache_path) or force_download):
-        # Download to temporary file, then copy to cache dir once finished.
-        # Otherwise you get corrupt cache entries if the download gets interrupted.
-        with temp_file_manager() as temp_file:
-            logger.info("%s not found in cache or force_download set to True, downloading to %s", url, temp_file.name)
-
-            # GET file object
-            if url.startswith("s3://"):
-                if resume_download:
-                    logger.warn('Warning: resumable downloads are not implemented for "s3://" urls')
-                s3_get(url, temp_file, proxies=proxies)
+        if resume_download:
+            incomplete_path = cache_path + '.incomplete'
+            @contextmanager
+            def _resumable_file_manager():
+                with open(incomplete_path,'a+b') as f:
+                    yield f
+            temp_file_manager = _resumable_file_manager
+            if os.path.exists(incomplete_path):
+                resume_size = os.stat(incomplete_path).st_size
             else:
-                http_get(url, temp_file, proxies=proxies, resume_size=resume_size, user_agent=user_agent)
+                resume_size = 0
+        else:
+            temp_file_manager = partial(tempfile.NamedTemporaryFile, dir=cache_dir, delete=False)
+            resume_size = 0
 
-            # we are copying the file before closing it, so flush to avoid truncation
-            temp_file.flush()
+        if etag is not None and (not os.path.exists(cache_path) or force_download):
+            # Download to temporary file, then copy to cache dir once finished.
+            # Otherwise you get corrupt cache entries if the download gets interrupted.
+            with temp_file_manager() as temp_file:
+                logger.info("%s not found in cache or force_download set to True, downloading to %s", url, temp_file.name)
 
-            logger.info("storing %s in cache at %s", url, cache_path)
-            os.rename(temp_file.name, cache_path)
+                # GET file object
+                if url.startswith("s3://"):
+                    if resume_download:
+                        logger.warn('Warning: resumable downloads are not implemented for "s3://" urls')
+                    s3_get(url, temp_file, proxies=proxies)
+                else:
+                    http_get(url, temp_file, proxies=proxies, resume_size=resume_size, user_agent=user_agent)
 
-            logger.info("creating metadata file for %s", cache_path)
-            meta = {'url': url, 'etag': etag}
-            meta_path = cache_path + '.json'
-            with open(meta_path, 'w') as meta_file:
-                output_string = json.dumps(meta)
-                if sys.version_info[0] == 2 and isinstance(output_string, str):
-                    output_string = unicode(output_string, 'utf-8')  # The beauty of python 2
-                meta_file.write(output_string)
+                # we are copying the file before closing it, so flush to avoid truncation
+                temp_file.flush()
+
+                logger.info("storing %s in cache at %s", url, cache_path)
+                os.rename(temp_file.name, cache_path)
+
+                logger.info("creating metadata file for %s", cache_path)
+                meta = {'url': url, 'etag': etag}
+                meta_path = cache_path + '.json'
+                with open(meta_path, 'w') as meta_file:
+                    output_string = json.dumps(meta)
+                    if sys.version_info[0] == 2 and isinstance(output_string, str):
+                        output_string = unicode(output_string, 'utf-8')  # The beauty of python 2
+                    meta_file.write(output_string)
 
     return cache_path

From 29cbab98f0b36a3056e2982bf968c8370bad3838 Mon Sep 17 00:00:00 2001
From: Aymeric Augustin <aymeric.augustin@fractalideas.com>
Date: Fri, 20 Dec 2019 20:56:59 +0100
Subject: [PATCH 500/505] Parallelize tests on Circle CI.

Set the number of CPUs manually based on the Circle CI resource class,
or else we're getting 36 CPUs, which is far too much (perhaps that's
the underlying hardware and not what Circle CI allocates to us).

Don't parallelize the custom tokenizers tests because they take less
than one second to run and parallelization actually makes them slower.
---
 .circleci/config.yml | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/.circleci/config.yml b/.circleci/config.yml
index b094067eb5..a7496c81e6 100644
--- a/.circleci/config.yml
+++ b/.circleci/config.yml
@@ -11,9 +11,9 @@ jobs:
             - run: sudo pip install torch
             - run: sudo pip install tensorflow
             - run: sudo pip install --progress-bar off .
-            - run: sudo pip install pytest codecov pytest-cov
+            - run: sudo pip install pytest codecov pytest-cov pytest-xdist
             - run: sudo pip install tensorboardX scikit-learn
-            - run: python -m pytest -sv ./transformers/tests/ --cov
+            - run: python -m pytest -n 8 -s -v ./transformers/tests/ --cov
             - run: codecov
     build_py3_torch:
         working_directory: ~/transformers
@@ -25,10 +25,10 @@ jobs:
             - checkout
             - run: sudo pip install torch
             - run: sudo pip install --progress-bar off .
-            - run: sudo pip install pytest codecov pytest-cov
+            - run: sudo pip install pytest codecov pytest-cov pytest-xdist
             - run: sudo pip install tensorboardX scikit-learn
-            - run: python -m pytest -sv ./transformers/tests/ --cov
-            - run: python -m pytest -sv ./examples/
+            - run: python -m pytest -n 8 -s -v ./transformers/tests/ --cov
+            - run: python -m pytest -n 8 -s -v ./examples/
             - run: codecov
     build_py3_tf:
         working_directory: ~/transformers
@@ -40,9 +40,9 @@ jobs:
             - checkout
             - run: sudo pip install tensorflow
             - run: sudo pip install --progress-bar off .
-            - run: sudo pip install pytest codecov pytest-cov
+            - run: sudo pip install pytest codecov pytest-cov pytest-xdist
             - run: sudo pip install tensorboardX scikit-learn
-            - run: python -m pytest -sv ./transformers/tests/ --cov
+            - run: python -m pytest -n 8 -s -v ./transformers/tests/ --cov
             - run: codecov
     build_py3_custom_tokenizers:
         working_directory: ~/transformers
@@ -51,7 +51,7 @@ jobs:
         steps:
             - checkout
             - run: sudo pip install --progress-bar off .
-            - run: sudo pip install pytest
+            - run: sudo pip install pytest pytest-xdist
             - run: sudo pip install mecab-python3
             - run: RUN_CUSTOM_TOKENIZERS=1 python -m pytest -sv ./transformers/tests/tokenization_bert_japanese_test.py
     deploy_doc:

From bb3bfa2d293589af0b3141c6f7235beba1c6bb44 Mon Sep 17 00:00:00 2001
From: Aymeric Augustin <aymeric.augustin@fractalideas.com>
Date: Fri, 20 Dec 2019 20:56:59 +0100
Subject: [PATCH 501/505] Distribute tests from the same file to the same
 worker.

This should prevent two issues:

- hitting API rate limits for tests that hit the HF API
- multiplying the cost of expensive test setups
---
 .circleci/config.yml | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/.circleci/config.yml b/.circleci/config.yml
index a7496c81e6..f9de338fa4 100644
--- a/.circleci/config.yml
+++ b/.circleci/config.yml
@@ -13,7 +13,7 @@ jobs:
             - run: sudo pip install --progress-bar off .
             - run: sudo pip install pytest codecov pytest-cov pytest-xdist
             - run: sudo pip install tensorboardX scikit-learn
-            - run: python -m pytest -n 8 -s -v ./transformers/tests/ --cov
+            - run: python -m pytest -n 8 --dist=loadfile -s -v ./transformers/tests/ --cov
             - run: codecov
     build_py3_torch:
         working_directory: ~/transformers
@@ -27,8 +27,8 @@ jobs:
             - run: sudo pip install --progress-bar off .
             - run: sudo pip install pytest codecov pytest-cov pytest-xdist
             - run: sudo pip install tensorboardX scikit-learn
-            - run: python -m pytest -n 8 -s -v ./transformers/tests/ --cov
-            - run: python -m pytest -n 8 -s -v ./examples/
+            - run: python -m pytest -n 8 --dist=loadfile -s -v ./transformers/tests/ --cov
+            - run: python -m pytest -n 8 --dist=loadfile -s -v ./examples/
             - run: codecov
     build_py3_tf:
         working_directory: ~/transformers
@@ -42,7 +42,7 @@ jobs:
             - run: sudo pip install --progress-bar off .
             - run: sudo pip install pytest codecov pytest-cov pytest-xdist
             - run: sudo pip install tensorboardX scikit-learn
-            - run: python -m pytest -n 8 -s -v ./transformers/tests/ --cov
+            - run: python -m pytest -n 8 --dist=loadfile -s -v ./transformers/tests/ --cov
             - run: codecov
     build_py3_custom_tokenizers:
         working_directory: ~/transformers

From 80caf79d0743a354a43d2aac5ccfe58e0ac1b80a Mon Sep 17 00:00:00 2001
From: Aymeric Augustin <aymeric.augustin@fractalideas.com>
Date: Fri, 20 Dec 2019 20:56:59 +0100
Subject: [PATCH 502/505] Prevent excessive parallelism in PyTorch.

We're already using as many processes in parallel as we have CPU cores.
Furthermore, the number of core may be incorrectly calculated as 36
(we've seen this in pytest-xdist) which make compound the problem.

PyTorch performance craters without this.
---
 .circleci/config.yml | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/.circleci/config.yml b/.circleci/config.yml
index f9de338fa4..812817efaa 100644
--- a/.circleci/config.yml
+++ b/.circleci/config.yml
@@ -4,6 +4,8 @@ jobs:
         working_directory: ~/transformers
         docker:
             - image: circleci/python:3.5
+        environment:
+            OMP_NUM_THREADS: 1
         resource_class: xlarge
         parallelism: 1
         steps:
@@ -19,6 +21,8 @@ jobs:
         working_directory: ~/transformers
         docker:
             - image: circleci/python:3.5
+        environment:
+            OMP_NUM_THREADS: 1
         resource_class: xlarge
         parallelism: 1
         steps:
@@ -34,6 +38,8 @@ jobs:
         working_directory: ~/transformers
         docker:
             - image: circleci/python:3.5
+        environment:
+            OMP_NUM_THREADS: 1
         resource_class: xlarge
         parallelism: 1
         steps:

From 343c094f2156962a24bf19c5fbd771d01c81caf7 Mon Sep 17 00:00:00 2001
From: Aymeric Augustin <aymeric.augustin@fractalideas.com>
Date: Fri, 20 Dec 2019 20:56:59 +0100
Subject: [PATCH 503/505] Run examples separately from tests.

This optimizes the total run time of the Circle CI test suite.
---
 .circleci/config.yml | 37 ++++++++++++++++++++++++++-----------
 1 file changed, 26 insertions(+), 11 deletions(-)

diff --git a/.circleci/config.yml b/.circleci/config.yml
index 812817efaa..bfa3b943aa 100644
--- a/.circleci/config.yml
+++ b/.circleci/config.yml
@@ -1,6 +1,6 @@
 version: 2
 jobs:
-    build_py3_torch_and_tf:
+    run_tests_py3_torch_and_tf:
         working_directory: ~/transformers
         docker:
             - image: circleci/python:3.5
@@ -17,7 +17,7 @@ jobs:
             - run: sudo pip install tensorboardX scikit-learn
             - run: python -m pytest -n 8 --dist=loadfile -s -v ./transformers/tests/ --cov
             - run: codecov
-    build_py3_torch:
+    run_tests_py3_torch:
         working_directory: ~/transformers
         docker:
             - image: circleci/python:3.5
@@ -32,9 +32,8 @@ jobs:
             - run: sudo pip install pytest codecov pytest-cov pytest-xdist
             - run: sudo pip install tensorboardX scikit-learn
             - run: python -m pytest -n 8 --dist=loadfile -s -v ./transformers/tests/ --cov
-            - run: python -m pytest -n 8 --dist=loadfile -s -v ./examples/
             - run: codecov
-    build_py3_tf:
+    run_tests_py3_tf:
         working_directory: ~/transformers
         docker:
             - image: circleci/python:3.5
@@ -50,7 +49,7 @@ jobs:
             - run: sudo pip install tensorboardX scikit-learn
             - run: python -m pytest -n 8 --dist=loadfile -s -v ./transformers/tests/ --cov
             - run: codecov
-    build_py3_custom_tokenizers:
+    run_tests_py3_custom_tokenizers:
         working_directory: ~/transformers
         docker:
             - image: circleci/python:3.5
@@ -60,6 +59,21 @@ jobs:
             - run: sudo pip install pytest pytest-xdist
             - run: sudo pip install mecab-python3
             - run: RUN_CUSTOM_TOKENIZERS=1 python -m pytest -sv ./transformers/tests/tokenization_bert_japanese_test.py
+    run_examples_py3_torch:
+        working_directory: ~/transformers
+        docker:
+            - image: circleci/python:3.5
+        environment:
+            OMP_NUM_THREADS: 1
+        resource_class: xlarge
+        parallelism: 1
+        steps:
+            - checkout
+            - run: sudo pip install torch
+            - run: sudo pip install --progress-bar off .
+            - run: sudo pip install pytest pytest-xdist
+            - run: sudo pip install tensorboardX scikit-learn
+            - run: python -m pytest -n 8 --dist=loadfile -s -v ./examples/
     deploy_doc:
         working_directory: ~/transformers
         docker:
@@ -72,7 +86,7 @@ jobs:
             - run: sudo pip install --progress-bar off -r docs/requirements.txt
             - run: sudo pip install --progress-bar off -r requirements.txt
             - run: ./.circleci/deploy.sh
-    repository_consistency:
+    check_repository_consistency:
         working_directory: ~/transformers
         docker:
             - image: circleci/python:3.5
@@ -91,9 +105,10 @@ workflows:
     version: 2
     build_and_test:
         jobs:
-            - repository_consistency
-            - build_py3_custom_tokenizers
-            - build_py3_torch_and_tf
-            - build_py3_torch
-            - build_py3_tf
+            - check_repository_consistency
+            - run_examples_py3_torch
+            - run_tests_py3_custom_tokenizers
+            - run_tests_py3_torch_and_tf
+            - run_tests_py3_torch
+            - run_tests_py3_tf
             - deploy_doc: *workflow_filters

From 767bc3ca68d7f1617e2edd31374be3e2f05b27b6 Mon Sep 17 00:00:00 2001
From: Aymeric Augustin <aymeric.augustin@fractalideas.com>
Date: Sat, 21 Dec 2019 08:46:26 +0100
Subject: [PATCH 504/505] Fix typo in model name.

This looks like a copy/paste mistake. Probably this test was never run.

Refs #2250.
---
 transformers/tests/modeling_tf_xlm_test.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/transformers/tests/modeling_tf_xlm_test.py b/transformers/tests/modeling_tf_xlm_test.py
index 8b5ab6d742..065d355b45 100644
--- a/transformers/tests/modeling_tf_xlm_test.py
+++ b/transformers/tests/modeling_tf_xlm_test.py
@@ -252,7 +252,7 @@ class TFXLMModelTest(TFCommonTestCases.TFCommonModelTester):
     @slow
     def test_model_from_pretrained(self):
         for model_name in list(TF_XLM_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-            model = XLMModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
+            model = TFXLMModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)
 
 

From b8e924e10d283c095e5aca3f762d812d5106b105 Mon Sep 17 00:00:00 2001
From: Aymeric Augustin <aymeric.augustin@fractalideas.com>
Date: Sat, 21 Dec 2019 08:50:15 +0100
Subject: [PATCH 505/505] Restore test.

This looks like debug code accidentally committed in b18509c2.

Refs #2250.
---
 transformers/tests/modeling_tf_albert_test.py | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/transformers/tests/modeling_tf_albert_test.py b/transformers/tests/modeling_tf_albert_test.py
index ee71371a18..374417cfe2 100644
--- a/transformers/tests/modeling_tf_albert_test.py
+++ b/transformers/tests/modeling_tf_albert_test.py
@@ -216,8 +216,7 @@ class TFAlbertModelTest(TFCommonTestCases.TFCommonModelTester):
 
     @slow
     def test_model_from_pretrained(self):
-        # for model_name in list(TF_ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
-        for model_name in ['albert-base-uncased']:
+        for model_name in list(TF_ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
             model = TFAlbertModel.from_pretrained(model_name, cache_dir=CACHE_DIR)
             self.assertIsNotNone(model)