Skip to content

This is a Python implementation of Alphazero (for chess) using a custom GUI

License

Notifications You must be signed in to change notification settings

DylanSnyder31/AlphaZero-Chess

Repository files navigation

Machine-Learning-Chess

This is a Python implemenatation of Google Deepmind's AlpahZero. This project uses the GUI and chess engine based on my last Repository.

Getting Started

Dependencies

pip install -r requirements.txt

Training

To continue training model run:

python Train_Agent.py

To train the model from scratch in deep_structure.py comment out these lines:

75| if True:
76|    data_NN = torch.load(file, map_location=lambda storage, loc: storage)
77|    self.Neural_Network_Architecture.load_state_dict(data_NN)

Playing

To play against the agent run:

python Play_Agent.py

Resources

Research Papers

Mastering The Game of Go Without Human knowledge

Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm

Articles

AlphaGo Zero - How and Why it Works

How to build your own AlphaZero AI using Python and Keras

A Simple Alpha(Go) Zero Tutorial

General Game-Playing With Monte Carlo Tree Search

Repositories

suragnair/alpha-zero-general

Zeta36/chess-alpha-zero

Licence

MIT © Dylan Snyder

Releases

No releases published

Packages

No packages published

Languages