Implementation of different sets (nim, hex and pentago) based on the baseline you can find here. The provided baseline has been simplified (games such as tictactoe othello have been removed) in order to focus on the games that are required.
To use a game of your choice, subclass the classes in Game.py and NeuralNet.py and implement their functions. Example implementations for Othello can be found in hex/hexGame.py and hex/pytorch/NNet.py.
Coach.py contains the core training loop and MCTS.py performs the Monte Carlo Tree Search. The parameters for the self-play can be specified in main.py. Additional neural network parameters are in hex/pytorch/NNet.py (cuda flag, batch size, epochs, learning rate etc.).
To start training a model for hex:
python main.pyChoose your framework and game in main.py.
To play against a pretrained model:
python pit.pyChoose your framework and game in pit.py.