EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

Fix TransformerTokenizer init for chatglm2 & 3 (#761)

Browse Source
This commit is contained in:
Ke Bao
2024-07-27 17:44:46 +08:00
committed by GitHub
parent ba29504b21
commit 3fdab91912
1 changed files with 21 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
python/sglang/srt/constrained/fsm_cache.py
View File

@@ -21,7 +21,27 @@ class FSMCache(BaseCache):
tokenizer = AutoTokenizer.from_pretrained(
tokenizer_path, **tokenizer_args_dict
)
self.outlines_tokenizer = TransformerTokenizer(tokenizer)
try:
self.outlines_tokenizer = TransformerTokenizer(tokenizer)
except AttributeError:
# FIXME: tmp fix for chatglm2 & chatglm3 (pad_token_id=0)
origin_pad_token_id = tokenizer.pad_token_id
def fset(self, value):
self._value = value
type(tokenizer).pad_token_id = property(
fget=type(tokenizer).pad_token_id.fget, fset=fset
)
self.outlines_tokenizer = TransformerTokenizer(tokenizer)
self.outlines_tokenizer.tokenizer.pad_token_id = origin_pad_token_id
self.outlines_tokenizer.pad_token_id = origin_pad_token_id
self.outlines_tokenizer.pad_token = (
self.outlines_tokenizer.tokenizer.pad_token
)
self.outlines_tokenizer.vocabulary = (
self.outlines_tokenizer.tokenizer.get_vocab()
)
else:
self.outlines_tokenizer = TransformerTokenizer(
tokenizer_path, **tokenizer_args_dict