Board Game Assistant using RAG

This repository is for learning how to use Retrieval-Augmented Generation (RAG) with Ollama to answer questions about 6 boardgames: Uno, Exploading Kittens, Bang!, Werewolves, Monopoly and Ticket to Ride. Follow the steps below to set up and run the project.

Prerequisites

Ensure you have the following installed:

• Python 3.8 or higher
• pip (Python package installer)

Installation

1. Clone the repository:

   git clone <repository-url>
   cd boardgame-rag-tutorial

2. Install the required Python packages:

   pip install -r requirements.txt

Setting Up the Database

1. Place your PDF documents in the data directory. You should name the files to the name of boardgames.

2. Populate the database with the documents:

   cd src
   python populate_database.py --reset

Querying the Database

To query the database, run the query.py script with your query text:

python query.py "Your query text here"

Project Structure