JunXiang
JunXiang
## Problem The **double quatations (")** which cannot use `parser.parse()` correctly. For example, completion from chain response as following: ```python text = """ Answer: { "product_name": "Control4® T3 Series 7\"...
I want to use `search_people` with parameter-`current_company` to find people. However, there are some problem in this method: 1. **Not all** the people will be return. (ex: company have 1000...
## Problem 1. The free model will use `HuggingFaceEmbeddings` which cause lots of **memory**. 2. The streamlit free server just have 1GB RAM. 3. When there have multiple users using...
## Problem The langchain retrievalQA will costs long time to get the answer. ## Solution 1. Use `streaming` to solve. ([Langchain Issue](https://github.com/langchain-ai/langchain/issues/3202)) 2. Use `streamlit_chat` to response streamling. But this...
## Why need supply openai api ? * `gpt4free` is not stable. * `gpt4free` not suitable for business project.