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

Style

Browse Source
This commit is contained in:
Sylvain Gugger
2021-04-06 19:54:13 -04:00
parent aef4cf8c52
commit fd338abdeb
2 changed files with 8 additions and 19 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
examples/question-answering/run_qa_beam_search_no_trainer.py
View File

@@ -76,9 +76,7 @@ def parse_args():
parser.add_argument( parser.add_argument(
"--preprocessing_num_workers", type=int, default=4, help="A csv or a json file containing the training data." "--preprocessing_num_workers", type=int, default=4, help="A csv or a json file containing the training data."
) )
parser.add_argument( parser.add_argument("--do_predict", action="store_true", help="Eval the question answering model")
"--do_predict", action="store_true", help="Eval the question answering model"
)
parser.add_argument( parser.add_argument(
"--validation_file", type=str, default=None, help="A csv or a json file containing the validation data." "--validation_file", type=str, default=None, help="A csv or a json file containing the validation data."
) )
@@ -396,7 +394,6 @@ def main():
return tokenized_examples return tokenized_examples
if "train" not in raw_datasets: if "train" not in raw_datasets:
raise ValueError("--do_train requires a train dataset") raise ValueError("--do_train requires a train dataset")
train_dataset = raw_datasets["train"] train_dataset = raw_datasets["train"]
@@ -481,7 +478,6 @@ def main():
return tokenized_examples return tokenized_examples
if "validation" not in raw_datasets: if "validation" not in raw_datasets:
raise ValueError("--do_eval requires a validation dataset") raise ValueError("--do_eval requires a validation dataset")
eval_examples = raw_datasets["validation"] eval_examples = raw_datasets["validation"]
@@ -539,11 +535,8 @@ def main():
train_dataset, shuffle=True, collate_fn=data_collator, batch_size=args.per_device_train_batch_size train_dataset, shuffle=True, collate_fn=data_collator, batch_size=args.per_device_train_batch_size
) )
eval_dataset.set_format(type="torch", columns=["attention_mask", "input_ids", "token_type_ids"]) eval_dataset.set_format(type="torch", columns=["attention_mask", "input_ids", "token_type_ids"])
eval_dataloader = DataLoader( eval_dataloader = DataLoader(eval_dataset, collate_fn=data_collator, batch_size=args.per_device_eval_batch_size)
eval_dataset, collate_fn=data_collator, batch_size=args.per_device_eval_batch_size
)
if args.do_predict: if args.do_predict:
test_dataset.set_format(type="torch", columns=["attention_mask", "input_ids", "token_type_ids"]) test_dataset.set_format(type="torch", columns=["attention_mask", "input_ids", "token_type_ids"])

8
examples/question-answering/run_qa_no_trainer.py
View File

@@ -81,9 +81,7 @@ def parse_args():
parser.add_argument( parser.add_argument(
"--preprocessing_num_workers", type=int, default=4, help="A csv or a json file containing the training data." "--preprocessing_num_workers", type=int, default=4, help="A csv or a json file containing the training data."
) )
parser.add_argument( parser.add_argument("--do_predict", action="store_true", help="Eval the question answering model")
"--do_predict", action="store_true", help="Eval the question answering model"
)
parser.add_argument( parser.add_argument(
"--validation_file", type=str, default=None, help="A csv or a json file containing the validation data." "--validation_file", type=str, default=None, help="A csv or a json file containing the validation data."
) )
@@ -543,9 +541,7 @@ def main():
) )
eval_dataset.set_format(type="torch", columns=["attention_mask", "input_ids", "token_type_ids"]) eval_dataset.set_format(type="torch", columns=["attention_mask", "input_ids", "token_type_ids"])
eval_dataloader = DataLoader( eval_dataloader = DataLoader(eval_dataset, collate_fn=data_collator, batch_size=args.per_device_eval_batch_size)
eval_dataset, collate_fn=data_collator, batch_size=args.per_device_eval_batch_size
)
if args.do_predict: if args.do_predict:
test_dataset.set_format(type="torch", columns=["attention_mask", "input_ids", "token_type_ids"]) test_dataset.set_format(type="torch", columns=["attention_mask", "input_ids", "token_type_ids"])