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

Closing bracket in DistilBERT's token count.

Browse Source
This commit is contained in:
LysandreJik
2019-08-29 15:30:10 -04:00
parent fe8fb10b44
commit caf1d116a6
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/distillation/scripts/token_counts.py
View File

@@ -24,7 +24,7 @@ from utils import logger
if __name__ == '__main__': if __name__ == '__main__':
parser = argparse.ArgumentParser(description="Token Counts for smoothing the masking probabilities in MLM (cf XLM/word2vec)") parser = argparse.ArgumentParser(description="Token Counts for smoothing the masking probabilities in MLM (cf XLM/word2vec)")
parser.add_argument("--data_file", type=str, default="data/dump.bert-base-uncased.pickle", parser.add_argument("--data_file", type=str, default="data/dump.bert-base-uncased.pickle",
help="The binarized dataset." help="The binarized dataset.")
parser.add_argument("--token_counts_dump", type=str, default="data/token_counts.bert-base-uncased.pickle", parser.add_argument("--token_counts_dump", type=str, default="data/token_counts.bert-base-uncased.pickle",
help="The dump file.") help="The dump file.")
parser.add_argument("--vocab_size", default=30522, type=int) parser.add_argument("--vocab_size", default=30522, type=int)