Final Project of Spring 2018 Artificial Intelligence at Tongji Univ.
A Chess AI based on Alphago and Alphago Zero
Currently finished supervised learning with alpha-beta search.
A pytorch version can be found here alpha-chess
ChessWarrior(white) vs. Stockfish(balck; elo 1600)
There are three modes: data train and play
- data: generate processed training data
- train: the main process to train neural network
- play: use the best model to play chess
python run.py -mode data
python run.py -mode train -lr 0.01
python run.py -mode play --no-cuda
pip install -r requirements.txt
- Reinforcement Learning
https://pan.baidu.com/s/1C2FVG9vQds-odrfgw5bUGQ passwd: imhk