Issues
- 0
Gobang version
#80 opened by YoujiaZhang - 24
Is it multiple searching at the same time?
#23 opened by apollo-time - 1
tensorflow.python.framework.errors_impl.InvalidArgumentError: Tensor input_1:0, specified in either feed_devices or fetch_devices was not found in the Graph
#58 opened by zky362550824 - 3
Cannot use multiple GPUs in self-play
#57 opened by AranKomat - 6
- 1
maybe a bug here
#51 opened by gooooloo - 15
- 7
- 2
AlphaZero Approach
#49 opened by JackThomson2 - 3
- 23
Performance Reports
#40 opened by mokemokechicken - 2
a question about reloading model
#48 opened by gooooloo - 3
Child seeds being identical to the parent seed may nullify the effect of multi-processing/threading
#47 opened by AranKomat - 21
It may forget pertinent information about positions that it no longer visits.
#38 opened by apollo-time - 39
Baseline Comparison?
#26 opened by mrlooi - 2
automatically ntest
#39 opened by gooooloo - 3
About the time of self-play
#24 opened by rtz19970824 - 1
The history dates of Challenge 3/4 are wrong.
#37 opened by sbbdms - 0
The sign of virtual loss is reversed
#35 opened by mokemokechicken - 2
What's different between Challenge 2 & 3?
#32 opened by gooooloo - 1
GPU ResourceExhaustedError after many times of Keras model.load() during self-play
#30 opened by HaoshengZou - 2
invalid correct moves
#29 opened by apollo-time - 5
About the optimizer?
#27 opened by wjllance - 1
404 in the new download script
#25 opened by evalon32 - 4
Another resign condition?
#11 opened by gooooloo - 4
Policy out softmax with illegal moves
#22 opened by apollo-time - 8
About MCTS
#17 opened by apollo-time - 5
Failed running GUI
#19 opened by mationai - 31
Great job!!
#1 opened by Zeta36 - 1
Drop wxPython?
#18 opened by breezewish - 1
Random flip and rotation when evaluate
#16 opened by apollo-time - 7
Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm
#13 opened by mokemokechicken - 1
What is action_by_value?
#15 opened by apollo-time - 1
Create .env file
#12 opened by wonderkid27 - 6
ChessAlpha Zero development
#10 opened by Zeta36 - 0
winner type is 'Winner' not 'Player'
#3 opened by mokemokechicken - 0
Implement resign
#8 opened by mokemokechicken - 3
- 0
- 1
L2 weight regularisation in loss?
#2 opened by gooooloo