Congratulation to DeepMind! This is a reengineering implementation (on behalf of many other git repo in /support/) of DeepMind's Oct19th publication: [Mastering the Game of Go without Human Knowledge]. The supervised learning approach is more practical for individuals. (This repository has single purpose of education only)
Merge branch selfplay to master branch. The master branch contains the SL trained model (50% train acc, 40% test acc). The selfplay branch contains workable gtp.py and optimized mcts player mixin.
Merge branch selfplay to master branch. The master branch contains the SL trained model (50% train acc, 40% test acc). The selfplay branch contains workable gtp.py and optimized mcts player mixin.