Skip to content

Use return_tensors="np" instead of "tf" #21266

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Rocketknight1 merged 1 commit into from
Jan 24, 2023
Merged

Use return_tensors="np" instead of "tf" #21266

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion examples/tensorflow/language-modeling/run_mlm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -499,7 +499,7 @@ def group_texts(examples):
# region TF Dataset preparation
num_replicas = training_args.strategy.num_replicas_in_sync
data_collator = DataCollatorForLanguageModeling(
tokenizer=tokenizer, mlm_probability=data_args.mlm_probability, return_tensors="tf"
tokenizer=tokenizer, mlm_probability=data_args.mlm_probability, return_tensors="np"
)
options = tf.data.Options()
options.experimental_distribute.auto_shard_policy = tf.data.experimental.AutoShardPolicy.OFF
Expand Down
4 changes: 2 additions & 2 deletions examples/tensorflow/multiple-choice/run_swag.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -105,7 +105,7 @@ def __call__(self, features):
padding=self.padding,
max_length=self.max_length,
pad_to_multiple_of=self.pad_to_multiple_of,
return_tensors="tf",
return_tensors="np",
)

# Un-flatten
Expand Down Expand Up @@ -410,7 +410,7 @@ def preprocess_function(examples):
)

if data_args.pad_to_max_length:
data_collator = DefaultDataCollator(return_tensors="tf")
data_collator = DefaultDataCollator(return_tensors="np")
else:
# custom class defined above, as HF has no data collator for multiple choice
data_collator = DataCollatorForMultipleChoice(tokenizer)
Expand Down
2 changes: 1 addition & 1 deletion examples/tensorflow/summarization/run_summarization.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -533,7 +533,7 @@ def postprocess_text(preds, labels):
model=model,
label_pad_token_id=label_pad_token_id,
pad_to_multiple_of=128, # Reduce the number of unique shapes for XLA, especially for generation
return_tensors="tf",
return_tensors="np",
)

dataset_options = tf.data.Options()
Expand Down
4 changes: 2 additions & 2 deletions examples/tensorflow/text-classification/run_glue.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -345,9 +345,9 @@ def preprocess_function(examples):
datasets = datasets.map(preprocess_function, batched=True, load_from_cache_file=not data_args.overwrite_cache)

if data_args.pad_to_max_length:
data_collator = DefaultDataCollator(return_tensors="tf")
data_collator = DefaultDataCollator(return_tensors="np")
else:
data_collator = DataCollatorWithPadding(tokenizer, return_tensors="tf")
data_collator = DataCollatorWithPadding(tokenizer, return_tensors="np")
# endregion

# region Metric function
Expand Down
2 changes: 1 addition & 1 deletion examples/tensorflow/token-classification/run_ner.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -396,7 +396,7 @@ def tokenize_and_align_labels(examples):

# We need the DataCollatorForTokenClassification here, as we need to correctly pad labels as
# well as inputs.
collate_fn = DataCollatorForTokenClassification(tokenizer=tokenizer, return_tensors="tf")
collate_fn = DataCollatorForTokenClassification(tokenizer=tokenizer, return_tensors="np")
num_replicas = training_args.strategy.num_replicas_in_sync
total_train_batch_size = training_args.per_device_train_batch_size * num_replicas

Expand Down
2 changes: 1 addition & 1 deletion examples/tensorflow/translation/run_translation.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -499,7 +499,7 @@ def preprocess_function(examples):
model=model,
label_pad_token_id=label_pad_token_id,
pad_to_multiple_of=64, # Reduce the number of unique shapes for XLA, especially for generation
return_tensors="tf",
return_tensors="np",
)
num_replicas = training_args.strategy.num_replicas_in_sync
total_train_batch_size = training_args.per_device_train_batch_size * num_replicas
Expand Down