github
/
ChatGLM2-6B
mirror of https://github.com/THUDM/ChatGLM2-6B
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

Update openai_api.py 

使用chunk.model_dump_json方法，替换chunk.json方法，以修复报错dumps_kwargs keyword arguments are no longer supported
pull/484/head
Yif-Liu-Github 2023-08-21 23:59:34 +08:00 committed by GitHub
parent 80602dcae1
commit a51bdd761a
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 5 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
openai_api.py
View File

@ -135,7 +135,7 @@ async def predict(query: str, history: List[List[str]], model_id: str):
finish_reason=None finish_reason=None
) )
chunk = ChatCompletionResponse(model=model_id, choices=[choice_data], object="chat.completion.chunk") chunk = ChatCompletionResponse(model=model_id, choices=[choice_data], object="chat.completion.chunk")
yield "{}".format(chunk.json(exclude_unset=True, ensure_ascii=False)) yield "{}".format(chunk.model_dump_json(exclude_unset=True,exclude_none=True))
current_length = 0 current_length = 0
@ -152,7 +152,7 @@ async def predict(query: str, history: List[List[str]], model_id: str):
finish_reason=None finish_reason=None
) )
chunk = ChatCompletionResponse(model=model_id, choices=[choice_data], object="chat.completion.chunk") chunk = ChatCompletionResponse(model=model_id, choices=[choice_data], object="chat.completion.chunk")
yield "{}".format(chunk.json(exclude_unset=True, ensure_ascii=False)) yield "{}".format(chunk.model_dump_json(exclude_unset=True,exclude_none=True))
choice_data = ChatCompletionResponseStreamChoice( choice_data = ChatCompletionResponseStreamChoice(
@ -161,14 +161,14 @@ async def predict(query: str, history: List[List[str]], model_id: str):
finish_reason="stop" finish_reason="stop"
) )
chunk = ChatCompletionResponse(model=model_id, choices=[choice_data], object="chat.completion.chunk") chunk = ChatCompletionResponse(model=model_id, choices=[choice_data], object="chat.completion.chunk")
yield "{}".format(chunk.json(exclude_unset=True, ensure_ascii=False)) yield "{}".format(chunk.model_dump_json(exclude_unset=True,exclude_none=True))
yield '[DONE]' yield '[DONE]'
if __name__ == "__main__": if __name__ == "__main__":
tokenizer = AutoTokenizer.from_pretrained("THUDM/chatglm2-6b", trust_remote_code=True) tokenizer = AutoTokenizer.from_pretrained("chatglm2-6b", trust_remote_code=True)
model = AutoModel.from_pretrained("THUDM/chatglm2-6b", trust_remote_code=True).cuda() model = AutoModel.from_pretrained("chatglm2-6b", trust_remote_code=True).cuda()
# 多显卡支持，使用下面两行代替上面一行，将num_gpus改为你实际的显卡数量 # 多显卡支持，使用下面两行代替上面一行，将num_gpus改为你实际的显卡数量
# from utils import load_model_on_gpus # from utils import load_model_on_gpus
# model = load_model_on_gpus("THUDM/chatglm2-6b", num_gpus=2) # model = load_model_on_gpus("THUDM/chatglm2-6b", num_gpus=2)