A Python-based Chess AI that uses a Convolutional Neural Network (CNN) to evaluate board positions and make strategic decisions. Trained on official tournament game data, this project explores how machine learning can simulate intelligent gameplay.
This project aims to build a basic Chess-playing AI using deep learning techniques. The model learns from board state evaluations and attempts to mimic human-like gameplay through pattern recognition and move prediction.
- Board state representation using UCI notation
- CNN-based evaluation of chess positions and move prediction
- Legal move generation and selection
- Support for playing against the AI
- Python 3
- PyTorch
- python-chess
- NumPy
- Jupyter Notebook