Snake AI Training Project

Overview

This project implements a reinforcement learning approach to train an AI agent to play the classic Snake game using Deep Q-Learning (DQN).

Features

• Manual gameplay mode
• AI training with Deep Q-Network
• Dynamic speed control
• Automatic model weight saving and loading
• Pygame-based visualization

Requirements

• Python 3.8+
• Libraries:
  • pygame
  • torch
  • numpy

Project Structure

• normalGame.py: Manual gameplay mode
• train.py: AI training script
• Src/snake_trainer.py: Main training logic
• Src/snake_enviroment.py: Game environment
• Src/model/_3.py: DQN model definition

Training Process

• Uses epsilon-greedy exploration strategy
• Implements experience replay
• Periodically updates target network
• Saves model weights every 100 episodes

How to Run

Manual Gameplay

python normalGame.py

Train AI

python train.py

Training Controls

• Up/Down Arrow: Adjust training speed
• Spacebar: Pause/Resume training

Model Saving

• Weights automatically saved in model_weights/
• Latest weights loaded at training start

License

MIT License