Project Module 1: RAG in asking questions and answering content extracted from PDF documents

Introduction

Large Language Models (LLMs): These models allow users to input any text and receive a relevant response. Popular applications include ChatGPT and Gemini.

Retrieval Augmented Generation (RAG): This technique enhances LLM responses by integrating content from a document source to answer an input query. This project demonstrates building a basic RAG program, applied to answering questions from course documents.

• Input: Document file and a related query.
• Output: The answer.

Code

Google Colab

1. Clone the repository. Then open and run RAG_pdf.file

git clone https://github.com/TungTSon/RAG_PDF.git

2. Upload a different PDF file and replace FILE_PATH with your path.
3. Run the provided cells.

Local Setup

If using Anaconda, create and activate an environment:

conda create -n my_env
conda activate my_env