Snake Game AI

Overview

This is a Snake game AI implemented in Python using Pygame. The AI agent learns to play the Snake game by training a neural network model. A game like this is useful to know how reinforcement learning works!

This project was possible with the help of Patrick Lober's youtube video. He explained everything beautifully even to a beginner like me.

How to Run

1. Install the required dependencies: Pygame, NumPy, and Matplotlib.
2. Run the agent.py file to train the AI agent:

   python agent.py

Building blocks

• The AI agent uses a Deep Q-Learning algorithm to learn and improve its gameplay.
• The training progress is visualized using Matplotlib to show game scores and mean scores over time.
• Made with Pygame & Pytorch