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

fix distilbert in auto tokenizer

Browse Source
This commit is contained in:
thomwolf
2019-08-30 22:43:26 +02:00
parent 256086bc69
commit bfe93a5a21
1 changed files with 3 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
pytorch_transformers/tokenization_auto.py
View File

@@ -94,13 +94,13 @@ class AutoTokenizer(object):
Examples::
config = AutoTokenizer.from_pretrained('bert-base-uncased') # Download vocabulary from S3 and cache.
config = AutoTokenizer.from_pretrained('./test/bert_saved_model/') # E.g. tokenizer was saved using `save_pretrained('./test/saved_model/')`
tokenizer = AutoTokenizer.from_pretrained('bert-base-uncased') # Download vocabulary from S3 and cache.
tokenizer = AutoTokenizer.from_pretrained('./test/bert_saved_model/') # E.g. tokenizer was saved using `save_pretrained('./test/saved_model/')`
"""
if 'distilbert' in pretrained_model_name_or_path:
return DistilBertTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
if 'roberta' in pretrained_model_name_or_path:
elif 'roberta' in pretrained_model_name_or_path:
return RobertaTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
elif 'bert' in pretrained_model_name_or_path:
return BertTokenizer.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)