调用api文档时返回内容为空

[HhhBbWw]

问题描述 / Problem Description 我在试图调用api文档进行回答时，返回了空的text。

复现问题的步骤 / Steps to Reproduce

1. 启动了langchain api文档相关服务，如图所示。

2. 写了一个本地的调用代码，按照文档示例给定了输入。 下面图片为示例

3. print result，可以看到返回了空的answer。

访问方式或调用代码是否有问题？

[dosubot[bot]]

Based on the information you provided and the context from the Langchain-Chatchat repository, there are a few potential issues that could be causing your problem.

Firstly, there are two similar issues in the Langchain-Chatchat repository that might be relevant to your situation. The first one is [FEATURE] 能否支持本地的API调用, which suggests that you can call a local API by modifying the model_config.py file. You can add your local API address to the model configuration in model_config.py by referring to the suggestions in this issue. The second one is [BUG] 我无法调用该项目的api功能, but the context does not provide more details about this issue.

Secondly, there might be a problem with the way you're using the requests library to call the "Longchain-ChatChat API Server" version 1.2.7. The issue lies in how you process the response with iter_content and attempt to decode and load JSON data. A safer and more general approach would be to use response.json() to directly parse the JSON response if the API returns a standard JSON object. Here's a revised version of the critical part:

response = requests.post(url, headers=headers, json=data)
try:
    result = response.json()  # Directly parse the JSON response
except json.JSONDecodeError:
    print("Error decoding JSON from response")
    result = []

Thirdly, the expected format of the JSON payload for creating chat sessions, receiving feedback, and teaching the chatbot new responses in the "Longchain-ChatChat API Server" version 1.2.7 is as follows:

{
  "query": "What's the weather like today?",
  "history": [
    {
      "role": "user",
      "content": "What's the weather forecast for today?"
    },
    {
      "role": "assistant",
      "content": "The weather forecast for today is sunny."
    }
  ],
  "stream": false,
  "model_name": "gpt-3",
  "temperature": 0.7,
  "max_tokens": 150,
  "prompt_name": "default"
}

Lastly, the "Longchain-ChatChat API Server" version 1.2.7 handles the receiving of feedback through a specific API endpoint. The endpoint for receiving feedback and teaching new responses is /chat/feedback. However, the provided code does not explicitly detail how the feedback is used to teach new responses or improve the model's performance.

I hope this information helps you solve your problem. If you need further assistance, please provide more details about your issue.

Sources

• [FEATURE] 能否支持本地的API调用
• [BUG] 我无法调用该项目的api功能
• tests/api/test_stream_chat_api_thread.py
• server/chat/agent_chat.py
• server/api.py

About Dosu

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.

[github-actions[bot]]

这个问题已经被标记为 stale ，因为它已经超过 30 天没有任何活动。