This repository is based on the work and demonstration by TechLead on ChatGPT Retrieval. You can find the original YouTube Video here.

I also added a RAG implementation to the repo.

In addition, pipenv is utilized for environment management. Pls see details below.

Install pipenv if you don't have it already

pip install pipenv

Activate the virtual environment

pipenv shell

Install the dependencies specified in the Pipfile.lock file

pipenv install

Place your own data into the data/ folder.

Test reading data/data.txt file.

> python chatgpt.py "what is my dog's name"
Your dog's name is Sunny.

Test reading data/cat.pdf file.

> python chatgpt.py "what is my cat's name"
Your cat's name is Muffy.

In addition to what's in the original repo, I also experimented a bit more with RAG 🤩, following the LangChain doc, I added a RAG implementation that utilizes a webpage as the knowledge base. Pls run rag_web.py for fun, or check out rag_web.ipynb for sample results.