使用LangChain+Streamlit时遇openai.APIConnectionError连接错误求助

在基于LangChain和Streamlit开发项目过程中，调用OpenAI API时反复收到openai.APIConnectionError: Connection error错误，已参照OpenAI官方文档调试但未解决。

错误详细栈信息 #

Traceback (most recent call last):
  File "C:\Users\HP\anaconda3\envs\astro_bot\Lib\site-packages\streamlit\runtime\scriptrunner\script_runner.py", line 600, in _run_script
    exec(code, module.__dict__)
  File "C:\Users\HP\Desktop\vs_codes\astro_bot\bot.py", line 83, in <module>
    refined_query = query_refiner(conversation_string, query)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\HP\Desktop\vs_codes\astro_bot\utils.py", line 36, in query_refiner
    response = client.completions.create(
               ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\HP\anaconda3\envs\astro_bot\Lib\site-packages\openai\_utils\_utils.py", line 277, in wrapper
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\HP\anaconda3\envs\astro_bot\Lib\site-packages\openai\resources\completions.py", line 528, in create
    return self._post(
           ^^^^^^^^^^^
  File "C:\Users\HP\anaconda3\envs\astro_bot\Lib\site-packages\openai\_base_client.py", line 1240, in post
    return cast(ResponseT, self.request(cast_to, opts, stream=stream, stream_cls=stream_cls))
                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\HP\anaconda3\envs\astro_bot\Lib\site-packages\openai\_base_client.py", line 921, in request
    return self._request(
           ^^^^^^^^^^^^^^
  File "C:\Users\HP\anaconda3\envs\astro_bot\Lib\site-packages\openai\_base_client.py", line 976, in _request
    return self._retry_request(
           ^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\HP\anaconda3\envs\astro_bot\Lib\site-packages\openai\_base_client.py", line 1053, in _retry_request
    return self._request(
           ^^^^^^^^^^^^^^
  File "C:\Users\HP\anaconda3\envs\astro_bot\Lib\site-packages\openai\_base_client.py", line 976, in _request
    return self._retry_request(
           ^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\HP\anaconda3\envs\astro_bot\Lib\site-packages\openai\_base_client.py", line 1053, in _retry_request
    return self._request(
           ^^^^^^^^^^^^^^
  File "C:\Users\HP\anaconda3\envs\astro_bot\Lib\site-packages\openai\_base_client.py", line 986, in _request
    raise APIConnectionError(request=request) from err
openai.APIConnectionError: Connection error.

相关代码片段 #

with st.spinner("typing..."):
    conversation_string = get_conversation_string()
    # st.code(conversation_string)
    refined_query = query_refiner(conversation_string, query)

def query_refiner(conversation, query):
    response = client.completions.create(
    model="gpt-3.5-turbo",
    prompt=f"""Given the following user query and conversation log, formulate a question that would be the most relevant to provide the user with an answer from a knowledge base.\n\nCONVERSATION LOG: \n{conversation}\n\nQuery: {query}\n\nRefined Query:""",
    temperature=0.7,
    max_tokens=256,
    top_p=1,
    frequency_penalty=0,
    presence_penalty=0
    )
    return response['choices'][0]['text']

def get_conversation_string():
    conversation_string = ""
    for i in range(len(st.session_state['responses'])-1):
        conversation_string += "Human: "+st.session_state['requests'][i] + "\n"
        conversation_string += "Bot: "+ st.session_state['responses'][i+1] + "\n"
    return conversation_string

解决思路 #

• 检查网络连通性：
  在终端执行curl https://api.openai.com/v1/models测试能否直接访问OpenAI接口，确认本地网络无访问限制，可尝试切换网络（如手机热点）排查问题。

• 配置代理（若需）：
  若处于受限网络环境，给OpenAI客户端配置代理：

  from openai import OpenAI
  client = OpenAI(
      api_key="你的API密钥",
      proxy="http://代理地址:端口"  # 示例：http://127.0.0.1:7890
  )
  

  也可通过设置环境变量HTTP_PROXY和HTTPS_PROXY全局配置代理。

• 修正模型调用方式：
  gpt-3.5-turbo属于Chat模型，需使用client.chat.completions.create接口，而非completions.create，错误调用可能间接引发异常。修正后的代码：

  def query_refiner(conversation, query):
      response = client.chat.completions.create(
          model="gpt-3.5-turbo",
          messages=[
              {"role": "system", "content": "根据对话日志和用户查询，生成最适合从知识库获取答案的精炼问题"},
              {"role": "user", "content": f"对话日志:\n{conversation}\n\n用户查询:{query}\n\n精炼问题:"}
          ],
          temperature=0.7,
          max_tokens=256
      )
      return response.choices[0].message.content.strip()
  

• 验证客户端初始化：
  确保OpenAI客户端初始化时正确传入有效API密钥，可添加测试代码验证：

  models = client.models.list()
  print(models.data[0].id)  # 能打印模型ID说明初始化正常
  

• 添加重试机制：
  针对网络波动导致的连接失败，用tenacity库添加重试逻辑：

  from tenacity import retry, stop_after_attempt, wait_exponential, retry_if_exception_type
  from openai import APIConnectionError
  
  @retry(
      stop=stop_after_attempt(3),
      wait=wait_exponential(multiplier=1, min=2, max=10),
      retry=retry_if_exception_type(APIConnectionError)
  )
  def query_refiner(conversation, query):
      # 此处放API调用代码
  

内容的提问来源于stack exchange，提问作者Segun Odusina