search

keon / deep-q-learning

Minimal Deep Q Learning (DQN & DDQN) implementations in Keras
https://keon.io/deep-q-learning
MIT License
1.28k stars 454 forks source link

Fixed indentation #6

Closed kenfehling closed 7 years ago

keon commented 7 years ago

Thanks! but I was thinking of changing it as

target = (reward + self.gamma *
          np.amax(self.model.predict(next_state)[0]))

which I think is a cleaner code. It'd be cool if you could put it in this PR.

kenfehling commented 7 years ago

OK good idea, changed it

keon commented 7 years ago

Perfect. Thanks!