Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/scar17off/chess-ai
Chess engine powered by neural networks featuring web and desktop interfaces, training capabilities, and grandmaster opening support.
https://github.com/scar17off/chess-ai
artificial-intelligence chess chess-engine chess-openings deep-learning game-ai gradio gui machine-learning neural-networks pygame python training-data web-interface
Last synced: 29 days ago
JSON representation
Chess engine powered by neural networks featuring web and desktop interfaces, training capabilities, and grandmaster opening support.
- Host: GitHub
- URL: https://github.com/scar17off/chess-ai
- Owner: scar17off
- License: mit
- Created: 2024-11-14T19:06:25.000Z (about 2 months ago)
- Default Branch: main
- Last Pushed: 2024-11-14T19:14:48.000Z (about 2 months ago)
- Last Synced: 2024-11-14T20:19:35.131Z (about 2 months ago)
- Topics: artificial-intelligence, chess, chess-engine, chess-openings, deep-learning, game-ai, gradio, gui, machine-learning, neural-networks, pygame, python, training-data, web-interface
- Language: Python
- Homepage:
- Size: 34.2 KB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE.md
Awesome Lists containing this project
README
# Chess AI with Neural Network
A sophisticated chess engine powered by a neural network, featuring both a graphical user interface and a web interface for gameplay and training.
## Features
- Neural network-based chess AI
- Interactive chess board visualization
- Multiple game modes (AI vs Human, Free Play)
- Training system with grandmaster-style game generation
- Model management and persistence
- Support for common chess openings
- Real-time move evaluation visualization
## Installation
1. Clone the repository
2. Install dependencies:
```bash
pip install -r requirements.txt
```
3. Generate training games:
```bash
python generate_training_games.py
```
## Usage
### Web Interface
Run the web interface with:
```bash
python webui.py
```
This launches a Gradio-based web interface with the following features:
- Interactive chess board
- Move input and validation
- AI move calculation
- Model selection and management
- Training game generation
- Model training interface
Features include:
- Pygame-based graphical interface
- Neural network visualization
- Real-time move evaluation
- Direct mouse interaction
## Training
The AI can be trained using:
1. Generated grandmaster-style games
2. Custom game recordings
To generate training games:
```bash
python generate_training_games.py
```
The system supports various grandmaster openings including:
- Sicilian Defense
- Ruy Lopez
- Italian Game
- French Defense
- Caro-Kann
- Queen's Gambit
- King's Indian
- And more...
## Configuration
Game parameters can be adjusted in `chess_config.json`:
```json
{
"DEFAULT_NUM_GAMES": 20,
"MIN_MOVES_PER_GAME": 30,
"MAX_MOVES_PER_GAME": 100,
"RANDOMIZATION_FACTOR": 0.2,
"CENTER_CONTROL_BONUS": 1.2,
"DEVELOPMENT_BONUS": 1.1,
"CHECK_BONUS": 0.5,
"CAPTURE_BONUS": 0.3
}
```
## Project Structure
- `webui.py`: Web interface implementation
- `neural_network.py`: Neural network model
- `game_recorder.py`: Game recording functionality
- `generate_training_games.py`: Training data generation
- `chess_config.json`: Configuration parameters
- `models/`: Saved model states
- `games/`: Recorded and generated games
- `pieces/`: Chess piece images
## Model Architecture
The neural network evaluates chess positions using:
- Piece positions and values
- Board control
- Development factors
- King safety
- Material balance
- Tactical opportunities
## Contributing
1. Fork the repository
2. Create a feature branch
3. Commit your changes
4. Push to the branch
5. Create a Pull Request
## License
This project is licensed under the MIT License - see the [LICENSE.md](LICENSE.md) file for details.