Loop should break when episode is done

[Driesssens]

Hi, I believe we should break out of the

for step in range(args.rnn_steps):

loop when done == True. Currently, when the environment indicates that the episode is done, the loop continues to go on for a couple of steps. Not sure how the Gym environment responds to that, but new values, rewards etc keep being added to the lists and that can't be good for training.

[mimoralea]

I agree... I had this issue and add it myself after line 146 here:

https://github.com/greydanus/baby-a3c/blob/master/baby-a3c.py#L144-L146