dnrops
/
transformers
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

docs: add docs for map, and add num procs to load_dataset (#27520)

This commit is contained in:
Phuc Van Phan 2023-11-16 20:16:19 +07:00 committed by GitHub
parent 85fde09c97
commit 69c9b89fcb
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 3 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
examples/flax/speech-recognition/run_flax_speech_recognition_seq2seq.py
View File

@ -439,6 +439,7 @@ def main():
data_args.dataset_config_name,
split=data_args.train_split_name,
cache_dir=data_args.dataset_cache_dir,
num_proc=data_args.preprocessing_num_workers,
token=True if model_args.use_auth_token else None,
)
@ -448,6 +449,7 @@ def main():
data_args.dataset_config_name,
split=data_args.eval_split_name,
cache_dir=data_args.dataset_cache_dir,
num_proc=data_args.preprocessing_num_workers,
token=True if model_args.use_auth_token else None,
)
@ -551,7 +553,7 @@ def main():
prepare_dataset,
remove_columns=next(iter(raw_datasets.values())).column_names,
num_proc=num_workers,
desc="preprocess train dataset",
desc="preprocess train and eval dataset",
)
# filter training data with inputs longer than max_input_length