Zeta36 / chess-alpha-zero

Chess reinforcement learning by AlphaGo Zero methods.
MIT License
2.14k stars 481 forks source link

Update requirements.txt. Add version keras #100

Closed maycuatroi closed 4 years ago

maycuatroi commented 4 years ago

Keras currently upgrated to version 2.3.1 and it made new issues :

tensorflow.python.framework.errors_impl.InvalidArgumentError: Tensor input_1:0, specified in either feed_devices or fetch_devices was not found in the Graph

Solution: Downgrade Keras version