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

Add model card creation snippet to example scripts (#13730)

Browse Source 
* Update run_glue.py

* Update run_glue.py

* Add model creation snippet to other scripts

* Fix style
This commit is contained in:
Gunjan Chhablani
2021-09-24 19:21:46 +05:30
committed by GitHub
parent 66b01ce864
commit 38580455de
10 changed files with 113 additions and 91 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
examples/pytorch/language-modeling/run_clm.py
View File

@@ -500,17 +500,19 @@ def main():
trainer.log_metrics("eval", metrics)
trainer.save_metrics("eval", metrics)
if training_args.push_to_hub:
kwargs = {"finetuned_from": model_args.model_name_or_path, "tasks": "text-generation"}
if data_args.dataset_name is not None:
kwargs["dataset_tags"] = data_args.dataset_name
if data_args.dataset_config_name is not None:
kwargs["dataset_args"] = data_args.dataset_config_name
kwargs["dataset"] = f"{data_args.dataset_name} {data_args.dataset_config_name}"
else:
kwargs["dataset"] = data_args.dataset_name
kwargs = {"finetuned_from": model_args.model_name_or_path, "tasks": "text-generation"}
if data_args.dataset_name is not None:
kwargs["dataset_tags"] = data_args.dataset_name
if data_args.dataset_config_name is not None:
kwargs["dataset_args"] = data_args.dataset_config_name
kwargs["dataset"] = f"{data_args.dataset_name} {data_args.dataset_config_name}"
else:
kwargs["dataset"] = data_args.dataset_name
if training_args.push_to_hub:
trainer.push_to_hub(**kwargs)
else:
trainer.create_model_card(**kwargs)
def _mp_fn(index):

20
examples/pytorch/language-modeling/run_mlm.py
View File

@@ -528,17 +528,19 @@ def main():
trainer.log_metrics("eval", metrics)
trainer.save_metrics("eval", metrics)
if training_args.push_to_hub:
kwargs = {"finetuned_from": model_args.model_name_or_path, "tasks": "fill-mask"}
if data_args.dataset_name is not None:
kwargs["dataset_tags"] = data_args.dataset_name
if data_args.dataset_config_name is not None:
kwargs["dataset_args"] = data_args.dataset_config_name
kwargs["dataset"] = f"{data_args.dataset_name} {data_args.dataset_config_name}"
else:
kwargs["dataset"] = data_args.dataset_name
kwargs = {"finetuned_from": model_args.model_name_or_path, "tasks": "fill-mask"}
if data_args.dataset_name is not None:
kwargs["dataset_tags"] = data_args.dataset_name
if data_args.dataset_config_name is not None:
kwargs["dataset_args"] = data_args.dataset_config_name
kwargs["dataset"] = f"{data_args.dataset_name} {data_args.dataset_config_name}"
else:
kwargs["dataset"] = data_args.dataset_name
if training_args.push_to_hub:
trainer.push_to_hub(**kwargs)
else:
trainer.create_model_card(**kwargs)
def _mp_fn(index):

20
examples/pytorch/language-modeling/run_plm.py
View File

@@ -499,17 +499,19 @@ def main():
trainer.log_metrics("eval", metrics)
trainer.save_metrics("eval", metrics)
if training_args.push_to_hub:
kwargs = {"finetuned_from": model_args.model_name_or_path, "tasks": "language-modeling"}
if data_args.dataset_name is not None:
kwargs["dataset_tags"] = data_args.dataset_name
if data_args.dataset_config_name is not None:
kwargs["dataset_args"] = data_args.dataset_config_name
kwargs["dataset"] = f"{data_args.dataset_name} {data_args.dataset_config_name}"
else:
kwargs["dataset"] = data_args.dataset_name
kwargs = {"finetuned_from": model_args.model_name_or_path, "tasks": "language-modeling"}
if data_args.dataset_name is not None:
kwargs["dataset_tags"] = data_args.dataset_name
if data_args.dataset_config_name is not None:
kwargs["dataset_args"] = data_args.dataset_config_name
kwargs["dataset"] = f"{data_args.dataset_name} {data_args.dataset_config_name}"
else:
kwargs["dataset"] = data_args.dataset_name
if training_args.push_to_hub:
trainer.push_to_hub(**kwargs)
else:
trainer.create_model_card(**kwargs)
def _mp_fn(index):

20
examples/pytorch/multiple-choice/run_swag.py
View File

@@ -430,15 +430,19 @@ def main():
trainer.log_metrics("eval", metrics)
trainer.save_metrics("eval", metrics)
kwargs = dict(
finetuned_from=model_args.model_name_or_path,
tasks="multiple-choice",
dataset_tags="swag",
dataset_args="regular",
dataset="SWAG",
language="en",
)
if training_args.push_to_hub:
trainer.push_to_hub(
finetuned_from=model_args.model_name_or_path,
tasks="multiple-choice",
dataset_tags="swag",
dataset_args="regular",
dataset="SWAG",
language="en",
)
trainer.push_to_hub(**kwargs)
else:
trainer.create_model_card(**kwargs)
def _mp_fn(index):

20
examples/pytorch/question-answering/run_qa.py
View File

@@ -623,17 +623,19 @@ def main():
trainer.log_metrics("predict", metrics)
trainer.save_metrics("predict", metrics)
if training_args.push_to_hub:
kwargs = {"finetuned_from": model_args.model_name_or_path, "tasks": "question-answering"}
if data_args.dataset_name is not None:
kwargs["dataset_tags"] = data_args.dataset_name
if data_args.dataset_config_name is not None:
kwargs["dataset_args"] = data_args.dataset_config_name
kwargs["dataset"] = f"{data_args.dataset_name} {data_args.dataset_config_name}"
else:
kwargs["dataset"] = data_args.dataset_name
kwargs = {"finetuned_from": model_args.model_name_or_path, "tasks": "question-answering"}
if data_args.dataset_name is not None:
kwargs["dataset_tags"] = data_args.dataset_name
if data_args.dataset_config_name is not None:
kwargs["dataset_args"] = data_args.dataset_config_name
kwargs["dataset"] = f"{data_args.dataset_name} {data_args.dataset_config_name}"
else:
kwargs["dataset"] = data_args.dataset_name
if training_args.push_to_hub:
trainer.push_to_hub(**kwargs)
else:
trainer.create_model_card(**kwargs)
def _mp_fn(index):

20
examples/pytorch/question-answering/run_qa_beam_search.py
View File

@@ -656,17 +656,19 @@ def main():
trainer.log_metrics("predict", metrics)
trainer.save_metrics("predict", metrics)
if training_args.push_to_hub:
kwargs = {"finetuned_from": model_args.model_name_or_path, "tasks": "question-answering"}
if data_args.dataset_name is not None:
kwargs["dataset_tags"] = data_args.dataset_name
if data_args.dataset_config_name is not None:
kwargs["dataset_args"] = data_args.dataset_config_name
kwargs["dataset"] = f"{data_args.dataset_name} {data_args.dataset_config_name}"
else:
kwargs["dataset"] = data_args.dataset_name
kwargs = {"finetuned_from": model_args.model_name_or_path, "tasks": "question-answering"}
if data_args.dataset_name is not None:
kwargs["dataset_tags"] = data_args.dataset_name
if data_args.dataset_config_name is not None:
kwargs["dataset_args"] = data_args.dataset_config_name
kwargs["dataset"] = f"{data_args.dataset_name} {data_args.dataset_config_name}"
else:
kwargs["dataset"] = data_args.dataset_name
if training_args.push_to_hub:
trainer.push_to_hub(**kwargs)
else:
trainer.create_model_card(**kwargs)
def _mp_fn(index):

20
examples/pytorch/summarization/run_summarization.py
View File

@@ -622,17 +622,19 @@ def main():
with open(output_prediction_file, "w") as writer:
writer.write("\n".join(predictions))
if training_args.push_to_hub:
kwargs = {"finetuned_from": model_args.model_name_or_path, "tasks": "summarization"}
if data_args.dataset_name is not None:
kwargs["dataset_tags"] = data_args.dataset_name
if data_args.dataset_config_name is not None:
kwargs["dataset_args"] = data_args.dataset_config_name
kwargs["dataset"] = f"{data_args.dataset_name} {data_args.dataset_config_name}"
else:
kwargs["dataset"] = data_args.dataset_name
kwargs = {"finetuned_from": model_args.model_name_or_path, "tasks": "summarization"}
if data_args.dataset_name is not None:
kwargs["dataset_tags"] = data_args.dataset_name
if data_args.dataset_config_name is not None:
kwargs["dataset_args"] = data_args.dataset_config_name
kwargs["dataset"] = f"{data_args.dataset_name} {data_args.dataset_config_name}"
else:
kwargs["dataset"] = data_args.dataset_name
if training_args.push_to_hub:
trainer.push_to_hub(**kwargs)
else:
trainer.create_model_card(**kwargs)
return results

16
examples/pytorch/text-classification/run_glue.py
View File

@@ -546,15 +546,17 @@ def main():
item = label_list[item]
writer.write(f"{index}\t{item}\n")
if training_args.push_to_hub:
kwargs = {"finetuned_from": model_args.model_name_or_path, "tasks": "text-classification"}
if data_args.task_name is not None:
kwargs["language"] = "en"
kwargs["dataset_tags"] = "glue"
kwargs["dataset_args"] = data_args.task_name
kwargs["dataset"] = f"GLUE {data_args.task_name.upper()}"
kwargs = {"finetuned_from": model_args.model_name_or_path, "tasks": "text-classification"}
if data_args.task_name is not None:
kwargs["language"] = "en"
kwargs["dataset_tags"] = "glue"
kwargs["dataset_args"] = data_args.task_name
kwargs["dataset"] = f"GLUE {data_args.task_name.upper()}"
if training_args.push_to_hub:
trainer.push_to_hub(**kwargs)
else:
trainer.create_model_card(**kwargs)
def _mp_fn(index):

20
examples/pytorch/token-classification/run_ner.py
View File

@@ -542,17 +542,19 @@ def main():
for prediction in true_predictions:
writer.write(" ".join(prediction) + "\n")
if training_args.push_to_hub:
kwargs = {"finetuned_from": model_args.model_name_or_path, "tasks": "token-classification"}
if data_args.dataset_name is not None:
kwargs["dataset_tags"] = data_args.dataset_name
if data_args.dataset_config_name is not None:
kwargs["dataset_args"] = data_args.dataset_config_name
kwargs["dataset"] = f"{data_args.dataset_name} {data_args.dataset_config_name}"
else:
kwargs["dataset"] = data_args.dataset_name
kwargs = {"finetuned_from": model_args.model_name_or_path, "tasks": "token-classification"}
if data_args.dataset_name is not None:
kwargs["dataset_tags"] = data_args.dataset_name
if data_args.dataset_config_name is not None:
kwargs["dataset_args"] = data_args.dataset_config_name
kwargs["dataset"] = f"{data_args.dataset_name} {data_args.dataset_config_name}"
else:
kwargs["dataset"] = data_args.dataset_name
if training_args.push_to_hub:
trainer.push_to_hub(**kwargs)
else:
trainer.create_model_card(**kwargs)
def _mp_fn(index):

28
examples/pytorch/translation/run_translation.py
View File

@@ -590,21 +590,23 @@ def main():
with open(output_prediction_file, "w", encoding="utf-8") as writer:
writer.write("\n".join(predictions))
kwargs = {"finetuned_from": model_args.model_name_or_path, "tasks": "translation"}
if data_args.dataset_name is not None:
kwargs["dataset_tags"] = data_args.dataset_name
if data_args.dataset_config_name is not None:
kwargs["dataset_args"] = data_args.dataset_config_name
kwargs["dataset"] = f"{data_args.dataset_name} {data_args.dataset_config_name}"
else:
kwargs["dataset"] = data_args.dataset_name
languages = [l for l in [data_args.source_lang, data_args.target_lang] if l is not None]
if len(languages) > 0:
kwargs["language"] = languages
if training_args.push_to_hub:
kwargs = {"finetuned_from": model_args.model_name_or_path, "tasks": "translation"}
if data_args.dataset_name is not None:
kwargs["dataset_tags"] = data_args.dataset_name
if data_args.dataset_config_name is not None:
kwargs["dataset_args"] = data_args.dataset_config_name
kwargs["dataset"] = f"{data_args.dataset_name} {data_args.dataset_config_name}"
else:
kwargs["dataset"] = data_args.dataset_name
languages = [l for l in [data_args.source_lang, data_args.target_lang] if l is not None]
if len(languages) > 0:
kwargs["language"] = languages
trainer.push_to_hub(**kwargs)
else:
trainer.create_model_card(**kwargs)
return results