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

Allow tokenization of sequences > 512 for caching

Browse Source 
For many applications requiring randomized data access, it's easier to cache the tokenized representations than the words. So why not turn this into a warning?
This commit is contained in:
Catalin Voss
2019-03-02 16:30:21 -08:00
committed by GitHub
parent 2152bfeae8
commit 9775b2eb27
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
pytorch_pretrained_bert/tokenization_openai.py
View File

@@ -232,7 +232,7 @@ class OpenAIGPTTokenizer(object):
else: else:
ids.append(self.encoder.get(token, 0)) ids.append(self.encoder.get(token, 0))
if len(ids) > self.max_len: if len(ids) > self.max_len:
raise ValueError( logger.warning(
"Token indices sequence length is longer than the specified maximum " "Token indices sequence length is longer than the specified maximum "
" sequence length for this OpenAI GPT model ({} > {}). Running this" " sequence length for this OpenAI GPT model ({} > {}). Running this"
" sequence through the model will result in indexing errors".format(len(ids), self.max_len) " sequence through the model will result in indexing errors".format(len(ids), self.max_len)