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)
APV_MCTS hash table implementation: maximum speed for coroutine.
Learning rate exponential annealing for training large (2048) batch size.
Implement full identity residual network with global average pooling
Merge nextpy3 into mater:
Changes:
APV_MCTS hash table implementation: maximum speed for coroutine. Learning rate exponential annealing for training large (2048) batch size. Implement full identity residual network with global average pooling