How do you go past the time limit but with still being able to reset when an agent dies? Or does it do that automatically
I am not quite sure what the question is. In the provided environments you can not go past the time limit (the episode terminates) and same happens when player dies. In both cases environment returns done=True and you have to call env.reset() to start a new game.
So there is no way to tell the difference between time up and death? Ok then thx
You could track the number of steps taken and if episode ends before the timelimit (18 000 for obtain-scenarios), then it ended because of timelimit.