Skip to content

Commit 0b083ee

Browse files
authored
fix: 修复openai模型在对接其他兼容openai接口平台时获取tokens错误 (#157)
1 parent d214e31 commit 0b083ee

File tree

2 files changed

+45
-3
lines changed

2 files changed

+45
-3
lines changed
Lines changed: 42 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,42 @@
1+
# coding=utf-8
2+
"""
3+
@project: maxkb
4+
@Author:虎
5+
@file: openai_chat_model.py
6+
@date:2024/4/18 15:28
7+
@desc:
8+
"""
9+
from typing import List
10+
11+
from langchain_core.messages import BaseMessage, get_buffer_string
12+
from langchain_openai import ChatOpenAI
13+
14+
15+
class TokenizerManage:
16+
tokenizer = None
17+
18+
@staticmethod
19+
def get_tokenizer():
20+
from transformers import GPT2TokenizerFast
21+
if TokenizerManage.tokenizer is None:
22+
TokenizerManage.tokenizer = GPT2TokenizerFast.from_pretrained('gpt2',
23+
cache_dir="/opt/maxkb/model/tokenizer",
24+
resume_download=False,
25+
force_download=False)
26+
return TokenizerManage.tokenizer
27+
28+
29+
class OpenAIChatModel(ChatOpenAI):
30+
def get_num_tokens_from_messages(self, messages: List[BaseMessage]) -> int:
31+
try:
32+
return super().get_num_tokens_from_messages(messages)
33+
except Exception as e:
34+
tokenizer = TokenizerManage.get_tokenizer()
35+
return sum([len(tokenizer.encode(get_buffer_string([m]))) for m in messages])
36+
37+
def get_num_tokens(self, text: str) -> int:
38+
try:
39+
return super().get_num_tokens(text)
40+
except Exception as e:
41+
tokenizer = TokenizerManage.get_tokenizer()
42+
return len(tokenizer.encode(text))

apps/setting/models_provider/impl/openai_model_provider/openai_model_provider.py

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -10,7 +10,6 @@
1010
from typing import Dict
1111

1212
from langchain.schema import HumanMessage
13-
from langchain_openai import ChatOpenAI
1413

1514
from common import forms
1615
from common.exception.app_exception import AppApiException
@@ -19,6 +18,7 @@
1918
from setting.models_provider.base_model_provider import IModelProvider, ModelProvideInfo, BaseModelCredential, \
2019
ModelInfo, \
2120
ModelTypeConst, ValidCode
21+
from setting.models_provider.impl.openai_model_provider.model.openai_chat_model import OpenAIChatModel
2222
from smartdoc.conf import PROJECT_DIR
2323

2424

@@ -71,8 +71,8 @@ class OpenAIModelProvider(IModelProvider):
7171
def get_dialogue_number(self):
7272
return 3
7373

74-
def get_model(self, model_type, model_name, model_credential: Dict[str, object], **model_kwargs) -> ChatOpenAI:
75-
azure_chat_open_ai = ChatOpenAI(
74+
def get_model(self, model_type, model_name, model_credential: Dict[str, object], **model_kwargs) -> OpenAIChatModel:
75+
azure_chat_open_ai = OpenAIChatModel(
7676
model=model_name,
7777
openai_api_base=model_credential.get('api_base'),
7878
openai_api_key=model_credential.get('api_key')

0 commit comments

Comments
 (0)