hongshaorou/Tianshou

Fork 0

History

Dong Yan 029ab199f4 add softmax for mcts root node

2017-12-26 16:47:24 +08:00

code-verify

connect policy-value network with leela for code verification

2017-11-16 20:19:18 +08:00

__init__.py

compute score

2017-12-01 01:37:55 +08:00

.gitignore

replace try catch by isinstance collections.Hashable

2017-12-24 20:57:53 +08:00

data_statistic.py

count the winning rate for each player

2017-12-23 22:05:34 +08:00

engine.py

fix bug of game config and add profing functions to mcts

2017-12-24 17:43:45 +08:00

game.py

add role for mcts debug

2017-12-26 15:07:15 +08:00

go.py

change all copy to deepcopy

2017-12-24 14:41:40 +08:00

model.py

add softmax for mcts root node

2017-12-26 16:47:24 +08:00

multi_gpu.py

play

2017-12-09 21:41:11 +08:00

play.py

minor fixed

2017-12-25 16:51:44 +08:00

player.py

fix bug of game config and add profing functions to mcts

2017-12-24 17:43:45 +08:00

README.md

merge class strategy with class game. Next, merge Go with GoEnv

2017-12-15 22:19:44 +08:00

reversi.py

debug for go and reversi

2017-12-24 14:40:50 +08:00

utils.py

merge gtp

2017-11-28 17:00:10 +08:00

README.md

Reproduce AlphaGo Zero

Network.py

A reimplemented version of pv network which is presented in the nature paper.

code-verify:

Connecting our own policy-value neural network with leela-zero.

checkpoints:

Weights of the policy-value neural network

README.md

Reproduce AlphaGo Zero

Network.py

code-verify:

checkpoints:

File Specification