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

Tf model outputs (#6247)

Browse Source 
* TF outputs and test on BERT

* Albert to DistilBert

* All remaining TF models except T5

* Documentation

* One file forgotten

* TF outputs and test on BERT

* Albert to DistilBert

* All remaining TF models except T5

* Documentation

* One file forgotten

* Add new models and fix issues

* Quality improvements

* Add T5

* A bit of cleanup

* Fix for slow tests

* Style
This commit is contained in:
Sylvain Gugger
2020-08-05 11:34:39 -04:00
committed by GitHub
parent bd0eab351a
commit c67d1a0259
51 changed files with 3253 additions and 2430 deletions
Download Patch File Download Diff File Expand all files Collapse all files

55
tests/test_modeling_tf_bert.py
View File

@@ -118,6 +118,7 @@ class TFBertModelTester:
max_position_embeddings=self.max_position_embeddings,
type_vocab_size=self.type_vocab_size,
initializer_range=self.initializer_range,
return_dict=True,
)
return config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
@@ -130,18 +131,14 @@ class TFBertModelTester:
sequence_output, pooled_output = model(inputs)
inputs = [input_ids, input_mask]
sequence_output, pooled_output = model(inputs)
result = model(inputs)
sequence_output, pooled_output = model(input_ids)
result = 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]
list(result["last_hidden_state"].shape), [self.batch_size, self.seq_length, self.hidden_size]
)
self.parent.assertListEqual(list(result["pooled_output"].shape), [self.batch_size, self.hidden_size])
self.parent.assertListEqual(list(result["pooler_output"].shape), [self.batch_size, self.hidden_size])
def create_and_check_bert_lm_head(
self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
@@ -153,7 +150,7 @@ class TFBertModelTester:
"attention_mask": input_mask,
"token_type_ids": token_type_ids,
}
(prediction_scores,) = model(inputs)
prediction_scores = model(inputs)["logits"]
self.parent.assertListEqual(
list(prediction_scores.numpy().shape), [self.batch_size, self.seq_length, self.vocab_size]
)
@@ -167,39 +164,27 @@ class TFBertModelTester:
"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]
)
result = model(inputs)
self.parent.assertListEqual(list(result["logits"].shape), [self.batch_size, self.seq_length, self.vocab_size])
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 = TFBertForNextSentencePrediction(config=config)
inputs = {"input_ids": input_ids, "attention_mask": input_mask, "token_type_ids": token_type_ids}
(seq_relationship_score,) = model(inputs)
result = {
"seq_relationship_score": seq_relationship_score.numpy(),
}
self.parent.assertListEqual(list(result["seq_relationship_score"].shape), [self.batch_size, 2])
result = model(inputs)
self.parent.assertListEqual(list(result["logits"].shape), [self.batch_size, 2])
def create_and_check_bert_for_pretraining(
self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
):
model = TFBertForPreTraining(config=config)
inputs = {"input_ids": input_ids, "attention_mask": input_mask, "token_type_ids": token_type_ids}
prediction_scores, seq_relationship_score = model(inputs)
result = {
"prediction_scores": prediction_scores.numpy(),
"seq_relationship_score": seq_relationship_score.numpy(),
}
result = model(inputs)
self.parent.assertListEqual(
list(result["prediction_scores"].shape), [self.batch_size, self.seq_length, self.vocab_size]
list(result["prediction_logits"].shape), [self.batch_size, self.seq_length, self.vocab_size]
)
self.parent.assertListEqual(list(result["seq_relationship_score"].shape), [self.batch_size, 2])
self.parent.assertListEqual(list(result["seq_relationship_logits"].shape), [self.batch_size, 2])
def create_and_check_bert_for_sequence_classification(
self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
@@ -212,8 +197,7 @@ class TFBertModelTester:
"token_type_ids": token_type_ids,
}
(logits,) = model(inputs)
result = {"logits": logits.numpy()}
result = model(inputs)
self.parent.assertListEqual(list(result["logits"].shape), [self.batch_size, self.num_labels])
def create_and_check_bert_for_multiple_choice(
@@ -229,8 +213,7 @@ class TFBertModelTester:
"attention_mask": multiple_choice_input_mask,
"token_type_ids": multiple_choice_token_type_ids,
}
(logits,) = model(inputs)
result = {"logits": logits.numpy()}
result = model(inputs)
self.parent.assertListEqual(list(result["logits"].shape), [self.batch_size, self.num_choices])
def create_and_check_bert_for_token_classification(
@@ -243,10 +226,7 @@ class TFBertModelTester:
"attention_mask": input_mask,
"token_type_ids": token_type_ids,
}
(logits,) = model(inputs)
result = {
"logits": logits.numpy(),
}
result = model(inputs)
self.parent.assertListEqual(list(result["logits"].shape), [self.batch_size, self.seq_length, self.num_labels])
def create_and_check_bert_for_question_answering(
@@ -259,8 +239,7 @@ class TFBertModelTester:
"token_type_ids": token_type_ids,
}
start_logits, end_logits = model(inputs)
result = {"start_logits": start_logits.numpy(), "end_logits": end_logits.numpy()}
result = model(inputs)
self.parent.assertListEqual(list(result["start_logits"].shape), [self.batch_size, self.seq_length])
self.parent.assertListEqual(list(result["end_logits"].shape), [self.batch_size, self.seq_length])