Aug 23, 2017 · 1 min read
Hi Arthur,
How do I identify which actionNumber in env.step(actionNumber) belongs to which action type out of left, right,up,down. Also since the first value returned by env.step() is the new state, so if for a period of time say 10 or 20 iterations I get different state value for the same actionNumber. I am confused.

Amazing post for the beginners. Cheers!!