Simple Local RAG Pipeline

This project implements a simple Local Retrieval-Augmented Generation (RAG) pipeline from scratch using Python and Hugging Face.

Features

• Inference: Utilizes an open-source LLM model from Hugging Face.
• Frameworks: Built with PyTorch and NumPy for calculations.
• Local Execution: Runs entirely on a local GPU.
• PDF Processing: Uses a PDF as the document source.

PDF Source

• Human Nutrition

Acknowledgements

• This project was built following a tutorial on YouTube