Simple RAG example using LM Studio Server

Start LM Studio server running on port 1234.

This repo performs 3 functions:

Scrapes a website and follows links under the same path up to a maximum depth and outputs the scraped data to the data directory.
Runs an embedding model to embed the text into a Chroma vector database using disk storage (chroma_db directory)
Runs a Chat Bot that uses the embeddings to answer questions about the website main.py runs all 3 functions. Once the scraper and embeddings have been completed once, they do not need to be run again. You can simply run the chatbot.py file.

How to Run

Download the code:

git clone https://github.com/gregmeldrum/simple-rag-lmstudio.git
cd simple-rag-lmstudio

Optional - setup a virtual environment:

python -m venv venv
source venv/bin/activate

Download dependencies and run the functions

pip install -r requirements.txt
python main.py

After running main.py once, the embeddings are persisted, and you can comment out the scrape and embed lines when you re-run main.py.