Skip to content

GitLab

Unverified Commit 41789c6c authored by Thomas Wolf's avatar Thomas Wolf Committed by GitHub
Browse files

Merge pull request #1059 from GuillemGSubies/master 

Better use of spacy tokenizer in open ai and xlm tokenizers
parents 260c8608 388e3251
Hide whitespace changes
Inline Side-by-side
Showing with 6 additions and 4 deletions
pytorch_transformers/tokenization_openai.py
View file @ 41789c6c
...@@ -89,8 +89,9 @@ class OpenAIGPTTokenizer(PreTrainedTokenizer): ...@@ -89,8 +89,9 @@ class OpenAIGPTTokenizer(PreTrainedTokenizer):
try: try:
import ftfy import ftfy
import spacy from spacy.lang.en import English
self.nlp = spacy.load('en', disable=['parser', 'tagger', 'ner', 'textcat']) _nlp = English()
self.nlp = _nlp.Defaults.create_tokenizer(_nlp)
self.fix_text = ftfy.fix_text self.fix_text = ftfy.fix_text
except ImportError: except ImportError:
logger.warning("ftfy or spacy is not installed using BERT BasicTokenizer instead of SpaCy & ftfy.") logger.warning("ftfy or spacy is not installed using BERT BasicTokenizer instead of SpaCy & ftfy.")
......
pytorch_transformers/tokenization_xlm.py
View file @ 41789c6c
...@@ -124,8 +124,9 @@ class XLMTokenizer(PreTrainedTokenizer): ...@@ -124,8 +124,9 @@ class XLMTokenizer(PreTrainedTokenizer):
**kwargs) **kwargs)
try: try:
import ftfy import ftfy
import spacy from spacy.lang.en import English
self.nlp = spacy.load('en', disable=['parser', 'tagger', 'ner', 'textcat']) _nlp = English()
self.nlp = _nlp.Defaults.create_tokenizer(_nlp)
self.fix_text = ftfy.fix_text self.fix_text = ftfy.fix_text
except ImportError: except ImportError:
logger.warning("ftfy or spacy is not installed using BERT BasicTokenizer instead of SpaCy & ftfy.") logger.warning("ftfy or spacy is not installed using BERT BasicTokenizer instead of SpaCy & ftfy.")
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment