SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

Black 20 release

Browse Source
This commit is contained in:
Lysandre
2020-08-26 17:20:22 +02:00
parent e78c110338
commit a75c64d80c
191 changed files with 4807 additions and 3503 deletions
Download Patch File Download Diff File Expand all files Collapse all files

112
examples/bert-loses-patience/pabee/modeling_pabee_albert.py
View File

@@ -101,30 +101,30 @@ class AlbertModelWithPabee(AlbertModel):
regression=False,
):
r"""
Return:
:obj:`tuple(torch.FloatTensor)` comprising various elements depending on the configuration (:class:`~transformers.AlbertConfig`) and inputs:
last_hidden_state (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length, hidden_size)`):
Sequence of hidden-states at the output of the last layer of the model.
pooler_output (:obj:`torch.FloatTensor`: of shape :obj:`(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 pre-training.
Return:
:obj:`tuple(torch.FloatTensor)` comprising various elements depending on the configuration (:class:`~transformers.AlbertConfig`) and inputs:
last_hidden_state (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length, hidden_size)`):
Sequence of hidden-states at the output of the last layer of the model.
pooler_output (:obj:`torch.FloatTensor`: of shape :obj:`(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 pre-training.
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 (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_hidden_states=True``):
Tuple of :obj:`torch.FloatTensor` (one for the output of the embeddings + one for the output of each layer)
of shape :obj:`(batch_size, sequence_length, hidden_size)`.
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 (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_hidden_states=True``):
Tuple of :obj:`torch.FloatTensor` (one for the output of the embeddings + one for the output of each layer)
of shape :obj:`(batch_size, sequence_length, hidden_size)`.
Hidden-states of the model at the output of each layer plus the initial embedding outputs.
attentions (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_attentions=True``):
Tuple of :obj:`torch.FloatTensor` (one for each layer) of shape
:obj:`(batch_size, num_heads, sequence_length, sequence_length)`.
Hidden-states of the model at the output of each layer plus the initial embedding outputs.
attentions (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_attentions=True``):
Tuple of :obj:`torch.FloatTensor` (one for each layer) of shape
:obj:`(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.
Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
heads.
"""
if input_ids is not None and inputs_embeds is not None:
@@ -157,7 +157,10 @@ class AlbertModelWithPabee(AlbertModel):
res = []
for i in range(self.config.num_hidden_layers):
encoder_outputs = self.encoder.adaptive_forward(
encoder_outputs, current_layer=i, attention_mask=extended_attention_mask, head_mask=head_mask,
encoder_outputs,
current_layer=i,
attention_mask=extended_attention_mask,
head_mask=head_mask,
)
pooled_output = self.pooler_activation(self.pooler(encoder_outputs[0][:, 0]))
@@ -174,7 +177,10 @@ class AlbertModelWithPabee(AlbertModel):
for i in range(self.config.num_hidden_layers):
calculated_layer_num += 1
encoder_outputs = self.encoder.adaptive_forward(
encoder_outputs, current_layer=i, attention_mask=extended_attention_mask, head_mask=head_mask,
encoder_outputs,
current_layer=i,
attention_mask=extended_attention_mask,
head_mask=head_mask,
)
pooled_output = self.pooler_activation(self.pooler(encoder_outputs[0][:, 0]))
@@ -236,42 +242,42 @@ class AlbertForSequenceClassificationWithPabee(AlbertPreTrainedModel):
labels=None,
):
r"""
labels (:obj:`torch.LongTensor` of shape :obj:`(batch_size,)`, `optional`, defaults to :obj:`None`):
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).
labels (:obj:`torch.LongTensor` of shape :obj:`(batch_size,)`, `optional`, defaults to :obj:`None`):
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).
Returns:
:obj:`tuple(torch.FloatTensor)` comprising various elements depending on the configuration (:class:`~transformers.AlbertConfig`) 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 (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_hidden_states=True``):
Tuple of :obj:`torch.FloatTensor` (one for the output of the embeddings + one for the output of each layer)
of shape :obj:`(batch_size, sequence_length, hidden_size)`.
Returns:
:obj:`tuple(torch.FloatTensor)` comprising various elements depending on the configuration (:class:`~transformers.AlbertConfig`) 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 (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_hidden_states=True``):
Tuple of :obj:`torch.FloatTensor` (one for the output of the embeddings + one for the output of each layer)
of shape :obj:`(batch_size, sequence_length, hidden_size)`.
Hidden-states of the model at the output of each layer plus the initial embedding outputs.
attentions (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_attentions=True``):
Tuple of :obj:`torch.FloatTensor` (one for each layer) of shape
:obj:`(batch_size, num_heads, sequence_length, sequence_length)`.
Hidden-states of the model at the output of each layer plus the initial embedding outputs.
attentions (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_attentions=True``):
Tuple of :obj:`torch.FloatTensor` (one for each layer) of shape
:obj:`(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.
Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
heads.
Examples::
Examples::
from transformers import AlbertTokenizer
from pabee import AlbertForSequenceClassificationWithPabee
import torch
from transformers import AlbertTokenizer
from pabee import AlbertForSequenceClassificationWithPabee
import torch
tokenizer = AlbertTokenizer.from_pretrained('albert-base-v2')
model = AlbertForSequenceClassificationWithPabee.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)
loss, logits = outputs[:2]
tokenizer = AlbertTokenizer.from_pretrained('albert-base-v2')
model = AlbertForSequenceClassificationWithPabee.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)
loss, logits = outputs[:2]
"""

102
examples/bert-loses-patience/pabee/modeling_pabee_bert.py
View File

@@ -108,30 +108,30 @@ class BertModelWithPabee(BertModel):
regression=False,
):
r"""
Return:
:obj:`tuple(torch.FloatTensor)` comprising various elements depending on the configuration (:class:`~transformers.BertConfig`) and inputs:
last_hidden_state (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length, hidden_size)`):
Sequence of hidden-states at the output of the last layer of the model.
pooler_output (:obj:`torch.FloatTensor`: of shape :obj:`(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 pre-training.
Return:
:obj:`tuple(torch.FloatTensor)` comprising various elements depending on the configuration (:class:`~transformers.BertConfig`) and inputs:
last_hidden_state (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length, hidden_size)`):
Sequence of hidden-states at the output of the last layer of the model.
pooler_output (:obj:`torch.FloatTensor`: of shape :obj:`(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 pre-training.
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 (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_hidden_states=True``):
Tuple of :obj:`torch.FloatTensor` (one for the output of the embeddings + one for the output of each layer)
of shape :obj:`(batch_size, sequence_length, hidden_size)`.
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 (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_hidden_states=True``):
Tuple of :obj:`torch.FloatTensor` (one for the output of the embeddings + one for the output of each layer)
of shape :obj:`(batch_size, sequence_length, hidden_size)`.
Hidden-states of the model at the output of each layer plus the initial embedding outputs.
attentions (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_attentions=True``):
Tuple of :obj:`torch.FloatTensor` (one for each layer) of shape
:obj:`(batch_size, num_heads, sequence_length, sequence_length)`.
Hidden-states of the model at the output of each layer plus the initial embedding outputs.
attentions (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_attentions=True``):
Tuple of :obj:`torch.FloatTensor` (one for each layer) of shape
:obj:`(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.
Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
heads.
"""
if input_ids is not None and inputs_embeds is not None:
@@ -266,44 +266,44 @@ class BertForSequenceClassificationWithPabee(BertPreTrainedModel):
labels=None,
):
r"""
labels (:obj:`torch.LongTensor` of shape :obj:`(batch_size,)`, `optional`, defaults to :obj:`None`):
Labels for computing the sequence classification/regression loss.
Indices should be in :obj:`[0, ..., config.num_labels - 1]`.
If :obj:`config.num_labels == 1` a regression loss is computed (Mean-Square loss),
If :obj:`config.num_labels > 1` a classification loss is computed (Cross-Entropy).
labels (:obj:`torch.LongTensor` of shape :obj:`(batch_size,)`, `optional`, defaults to :obj:`None`):
Labels for computing the sequence classification/regression loss.
Indices should be in :obj:`[0, ..., config.num_labels - 1]`.
If :obj:`config.num_labels == 1` a regression loss is computed (Mean-Square loss),
If :obj:`config.num_labels > 1` a classification loss is computed (Cross-Entropy).
Returns:
:obj:`tuple(torch.FloatTensor)` comprising various elements depending on the configuration (:class:`~transformers.BertConfig`) and inputs:
loss (:obj:`torch.FloatTensor` of shape :obj:`(1,)`, `optional`, returned when :obj:`label` is provided):
Classification (or regression if config.num_labels==1) loss.
logits (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, config.num_labels)`):
Classification (or regression if config.num_labels==1) scores (before SoftMax).
hidden_states (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_hidden_states=True``):
Tuple of :obj:`torch.FloatTensor` (one for the output of the embeddings + one for the output of each layer)
of shape :obj:`(batch_size, sequence_length, hidden_size)`.
Returns:
:obj:`tuple(torch.FloatTensor)` comprising various elements depending on the configuration (:class:`~transformers.BertConfig`) and inputs:
loss (:obj:`torch.FloatTensor` of shape :obj:`(1,)`, `optional`, returned when :obj:`label` is provided):
Classification (or regression if config.num_labels==1) loss.
logits (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, config.num_labels)`):
Classification (or regression if config.num_labels==1) scores (before SoftMax).
hidden_states (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_hidden_states=True``):
Tuple of :obj:`torch.FloatTensor` (one for the output of the embeddings + one for the output of each layer)
of shape :obj:`(batch_size, sequence_length, hidden_size)`.
Hidden-states of the model at the output of each layer plus the initial embedding outputs.
attentions (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_attentions=True``):
Tuple of :obj:`torch.FloatTensor` (one for each layer) of shape
:obj:`(batch_size, num_heads, sequence_length, sequence_length)`.
Hidden-states of the model at the output of each layer plus the initial embedding outputs.
attentions (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_attentions=True``):
Tuple of :obj:`torch.FloatTensor` (one for each layer) of shape
:obj:`(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.
Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
heads.
Examples::
Examples::
from transformers import BertTokenizer, BertForSequenceClassification
from pabee import BertForSequenceClassificationWithPabee
import torch
from transformers import BertTokenizer, BertForSequenceClassification
from pabee import BertForSequenceClassificationWithPabee
import torch
tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
model = BertForSequenceClassificationWithPabee.from_pretrained('bert-base-uncased')
tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
model = BertForSequenceClassificationWithPabee.from_pretrained('bert-base-uncased')
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)
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]
loss, logits = outputs[:2]
"""

83
examples/bert-loses-patience/run_glue_with_pabee.py
View File

@@ -120,7 +120,10 @@ def train(args, train_dataset, model, tokenizer):
# 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,
model,
device_ids=[args.local_rank],
output_device=args.local_rank,
find_unused_parameters=True,
)
# Train!
@@ -151,13 +154,17 @@ def train(args, train_dataset, model, tokenizer):
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,
" Will skip the first %d steps in the first epoch",
steps_trained_in_current_epoch,
)
tr_loss, logging_loss = 0.0, 0.0
model.zero_grad()
train_iterator = trange(
epochs_trained, int(args.num_train_epochs), desc="Epoch", disable=args.local_rank not in [-1, 0],
epochs_trained,
int(args.num_train_epochs),
desc="Epoch",
disable=args.local_rank not in [-1, 0],
)
set_seed(args) # Added here for reproductibility
for _ in train_iterator:
@@ -372,7 +379,11 @@ def load_and_cache_examples(args, task, tokenizer, evaluate=False):
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,
examples,
tokenizer,
label_list=label_list,
max_length=args.max_seq_length,
output_mode=output_mode,
)
if args.local_rank in [-1, 0]:
logger.info("Saving features into cached file %s", cached_features_file)
@@ -434,15 +445,24 @@ def main():
help="The output directory where the model predictions and checkpoints will be written.",
)
parser.add_argument(
"--patience", default="0", type=str, required=False,
"--patience",
default="0",
type=str,
required=False,
)
parser.add_argument(
"--regression_threshold", default=0, type=float, required=False,
"--regression_threshold",
default=0,
type=float,
required=False,
)
# Other parameters
parser.add_argument(
"--config_name", default="", type=str, help="Pretrained config name or path if not the same as model_name",
"--config_name",
default="",
type=str,
help="Pretrained config name or path if not the same as model_name",
)
parser.add_argument(
"--tokenizer_name",
@@ -466,17 +486,27 @@ 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.")
parser.add_argument(
"--evaluate_during_training", action="store_true", help="Run evaluation during training at each logging step.",
"--evaluate_during_training",
action="store_true",
help="Run 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.",
"--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.",
"--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=1, type=int, help="Batch size per GPU/CPU for evaluation.",
"--per_gpu_eval_batch_size",
default=1,
type=int,
help="Batch size per GPU/CPU for evaluation.",
)
parser.add_argument(
"--gradient_accumulation_steps",
@@ -485,13 +515,19 @@ def main():
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.",
"--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 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, help="Max gradient norm.")
parser.add_argument(
"--num_train_epochs", default=3.0, type=float, help="Total number of training epochs to perform.",
"--num_train_epochs",
default=3.0,
type=float,
help="Total number of training epochs to perform.",
)
parser.add_argument(
"--max_steps",
@@ -503,7 +539,10 @@ def main():
parser.add_argument("--logging_steps", type=int, default=500, help="Log every X updates steps.")
parser.add_argument(
"--save_steps", type=int, default=500, help="Save checkpoint every X updates steps.",
"--save_steps",
type=int,
default=500,
help="Save checkpoint every X updates steps.",
)
parser.add_argument(
"--eval_all_checkpoints",
@@ -512,10 +551,14 @@ def main():
)
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",
"--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",
"--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")
@@ -532,7 +575,10 @@ def main():
"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",
"--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.")
@@ -634,7 +680,8 @@ def main():
print("Output Layers Parameters:", output_layers_param_num)
single_output_layer_param_num = sum(param.numel() for param in model.classifiers[0].parameters())
print(
"Added Output Layers Parameters:", output_layers_param_num - single_output_layer_param_num,
"Added Output Layers Parameters:",
output_layers_param_num - single_output_layer_param_num,
)
logger.info("Training/evaluation parameters %s", args)