wombyz / custom-knowledge-chatbot Goto Github PK

I assume the last line should use "prompt" not "user_input" in the call to query()

 def generate_response(self, user_input):
        prompt = "\n".join([f"{message['role']}: {message['content']}" for message in self.chat_history[-5:]])
        prompt += f"\nUser: {user_input}"
        response = index.query(user_input)

Hello,

I encountered an issue while trying to clone the custom-knowledge-chatbot repository on a Windows machine. The issue is related to .rtf files with special characters (specifically question marks) in their file names. Windows does not allow the use of certain special characters in file names, including question marks, which leads to an error during the cloning process.

Here is an example of an invalid file path causing the issue:
custom-knowledge-chatbot/asos/Can you tell me more about your ASOS Premier service in Saudi Arabia? .rtf

I suggest updating the file names by removing or replacing the special characters (e.g., replacing the question mark with a hyphen) to make the repository compatible with Windows. This change would make it easier for Windows users to clone and work with your repository without encountering errors.

Thank you for your attention to this matter, and please let me know if you need any further information or assistance.

Very truly yours,

Reno Fernandez

PS: This email was generated using GPT-4. I figured you would appreciate that.

The section of code seems to have a issue stemming either from the mismatched names after the llama update or something else I'm not smart enough to figure out.

when I run the index of documents cell then returns an error
ValueError: The constructor now takes in a list of Node objects. Since you are passing in a list of Document objects, please use from_documents instead.

OpenAI is not defined error, we seem to be missing an import in the code?

Hi, the code does not work any more, but anyone can recover the right behaviour by installing old versions of the libraries:

Just use this:

!pip install llama_index==0.4.27
!pip install langchain==0.0.109