README.md

RAG Chatbot

This is an LLM Chatbot powered by RAG. The tech stack includes Python, Langchain, OpenAI and Chroma DB vector store. Hosted on Streamlit.

LLM - Large Language Model
RAG - Retrieval Augumented Generation

For a video walkthrough, click this YouTube link to watch.

1. Create and activate virtual environment

python3 -m venv myvenv
source myvenv/bin/activate

2. Create and activate virtual environment on windows CMD

python -m venv myvenv
cd myvenv
.\Scripts\activate.bat 

3. Install libraries and dependencies

pip3 install -r requirements.txt

4. Get OpenAI API key

5. Run Streamlit app

streamlit run main.py

Split document and save to Supabase Vector database (Run once or only when you need to store a document)

python3 split_document.py

More Docs and Links

Streamlit Docs
Langchain Python Docs
Langchain Conversational RAG Docs

research a better chunck_size and chunck_overlap to use