search

junxiaosong / AlphaZero_Gomoku

An implementation of the AlphaZero algorithm for Gomoku (also called Gobang or Five in a Row)
MIT License
3.34k stars 972 forks source link

Support tensorflow #9

Closed Kelvin-Zhong closed 6 years ago

Kelvin-Zhong commented 6 years ago

Support Tensorflow, already comment out the changes specific for Tensorflow.

Though I am still a N00b, I really like this repro, which is a super clear and easy learning materials :)

junxiaosong commented 6 years ago

My pleasure if it helps in any way. Please allow me some time to review the changes. Happy Chinese New Year!

junxiaosong commented 6 years ago

I have just merged your commits to support tensorflow. There seems to be some problems, say the L2 penalty part and the train_step() method in the file policy_value_net_tensorflow.py. I have tried to fix all the problems, but not sure doing it right or not. So if you have time, maybe you can review my changes to your commits to double check. Thanks.