Yuan Chai

Results 3 issues of Yuan Chai

**功能描述 / Feature Description** 对于不同文件的存储的考虑: 1. 段落型知识文件-全部向量化 2. 问答对文件(jsonl个数)-只对query进行向量化,将answer存到metadata中,检索时,基于用户query和vs中的query进行匹配 3. 段落型和问答对两种类型可以分别设置不同的阈值,一般情况问答对的阈值会低一些 **解决的问题 / Problem Solved** 应对段落行知识文件和问答对两种情景 提高检索能力 现在我想用问答对jsonl文件存储到向量库,但只想向量化query部分,这个目前有实现吗?

enhancement

openai方式请求报错 ![image](https://github.com/modelscope/swift/assets/46665034/464060ea-30e3-4417-86bd-002fe0ffb110) **Describe the bug** INFO: 127.0.0.1:35572 - "POST /v1/chat/completions HTTP/1.1" 500 Internal Server Error ERROR: Exception in ASGI application Traceback (most recent call last): File "/home/chaiy/software/miniconda3/envs/swift/lib/python3.11/site-packages/uvicorn/protocols/http/httptools_impl.py", line 411, in...

我发现qwen的原生agent的react是放在了user上,而swift官方提供的ms_agent的react放在了system中,这两者有何区别,或者哪种更好? Qwen: ```bash system You are a helpful assistant. user Answer the following questions as best you can. You have access to the following APIs ``` ms_agent ```bash system Answer...