From a51bdd761a58a13f57e513d386a0b5ba50c105c5 Mon Sep 17 00:00:00 2001
From: Yif-Liu-Github <54025962+Yif-Liu-Github@users.noreply.github.com>
Date: Mon, 21 Aug 2023 23:59:34 +0800
Subject: [PATCH] Update openai_api.py
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

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
---
 openai_api.py | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/openai_api.py b/openai_api.py
index 7225562..a88e491 100644
--- a/openai_api.py
+++ b/openai_api.py
@@ -135,7 +135,7 @@ async def predict(query: str, history: List[List[str]], model_id: str):
         finish_reason=None
     )
     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
 
@@ -152,7 +152,7 @@ async def predict(query: str, history: List[List[str]], model_id: str):
             finish_reason=None
         )
         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(
@@ -161,14 +161,14 @@ async def predict(query: str, history: List[List[str]], model_id: str):
         finish_reason="stop"
     )
     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]'
 
 
 
 if __name__ == "__main__":
-    tokenizer = AutoTokenizer.from_pretrained("THUDM/chatglm2-6b", trust_remote_code=True)
-    model = AutoModel.from_pretrained("THUDM/chatglm2-6b", trust_remote_code=True).cuda()
+    tokenizer = AutoTokenizer.from_pretrained("chatglm2-6b", trust_remote_code=True)
+    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
     # from utils import load_model_on_gpus
     # model = load_model_on_gpus("THUDM/chatglm2-6b", num_gpus=2)