Snake AI Q-Learning 🐍

A sophisticated implementation of a Snake game AI that utilizes Deep Q-Learning to master gameplay autonomously. This project demonstrates the practical application of reinforcement learning in game environments, showcasing how an AI agent can learn complex strategies through experience.

🌟 Features

• 🎮 Classic Snake game built with Pygame
• 🧠 Deep Q-Learning implementation with PyTorch
• 📈 Real-time visualization of training progress
• 💾 Automatic saving of best models
• 🎯 High score tracking

🔧 Technical Requirements

Core Dependencies

• Python 3.13.1
• PyTorch 2.9.0
  • torchvision 0.24.0
  • torchaudio 2.9.0
• Pygame 2.6.1 (SDL 2.28.4)
• Matplotlib 3.10.7
• NumPy 2.3.4
• IPython 9.6.0

Additional Dependencies

• filelock 3.20.0
• fsspec 2025.10.0
• jinja2 3.1.6
• networkx 3.5
• pillow 12.0.0
• typing-extensions 4.15.0
• MarkupSafe 3.0.3
• sympy 1.14.0
• contourpy 1.3.3
• cycler 0.12.1
• fonttools 4.60.1
• kiwisolver 1.4.9
• packaging 25.0
• pyparsing 3.2.5
• python-dateutil 2.9.0
• six 1.17.0

🚀 Quick Start

1. Clone the repository:

git clone https://github.com/Antwa-sensei253/Snake_AI_Q-learning.git
cd Snake_AI_Q-learning

2. Create and activate a virtual environment:

python -m venv .venv
# On Windows:
.\.venv\Scripts\activate

3. Install dependencies:

pip install torch torchvision torchaudio
pip install pygame numpy matplotlib

4. Run the training:

python agent.py

🎮 Game Environment

The game environment (game.py) is built with Pygame and includes:

• Classic snake movement mechanics
• Food spawning system
• Collision detection
• Score tracking
• FPS control

🤖 AI Agent

The AI agent (agent.py) uses Deep Q-Learning with:

• 11 input state features:
  • Danger detection in 3 directions
  • Current movement direction (4 values)
  • Food location relative to snake head (4 values)
• 3 possible actions:
  • [1,0,0] → Go straight
  • [0,1,0] → Turn right
  • [0,0,1] → Turn left

🧮 Neural Network

The model (model.py) architecture:

Input Layer (11) → Hidden Layer (256) → Output Layer (3)

💡 Training Process

• Experience replay with memory size of 100,000
• Epsilon-greedy exploration strategy
• Reward system:
  • Eating food: +10 points
  • Game over: -10 points
  • Survival: 0 points

📊 Performance Tracking

• Terminal output shows:
  • Current game number
  • Score
  • Record score
• Real-time plot displays:
  • Individual game scores
  • Moving average score

📁 Project Structure

Snake_AI_Q-learning/
├── agent.py     # AI agent and training loop
├── game.py      # Snake game environment
├── model.py     # Neural network architecture
├── plot.py      # Training visualization
└── model/       # Saved model weights

⚙️ Configuration

Key parameters in agent.py:

MAX_MEMORY = 100_000
BATCH_SIZE = 1000
LR = 0.001

🤝 Contributing

Contributions are welcome! Feel free to:

1. Fork the repository
2. Create a feature branch
3. Submit a pull request

📝 License

This project is open source and available under the MIT License.

🙏 Acknowledgments

Special thanks to:

• PyTorch team for the deep learning framework
• Pygame community for the game development library