Question about Test function

[AlexTo]

Hi, I'd to have a question about the following block

rl_a3c_pytorch/test.py

Line 60 in eb5c9b9

if player.done and not player.info:

    if player.done and not player.info:
        state = player.env.reset()
        player.eps_len += 2
        player.state = torch.from_numpy(state).float()
        if gpu_id >= 0:
            with torch.cuda.device(gpu_id):
                player.state = player.state.cuda()
    elif player.info:

I don't quite understand when the info equals True or False, what is the meaning off having info=True and info=False ?

I can't seem to find a documentation about this info flag on Gym website :(

Thanks

[dgriff777]

Hi,

This is related to environment wrapper class EpisodicLifeEnv, found in environment.py, which makes the games episodic per life, for games with multiple lives, which can improve training speed. So if agent is in a game that has multiple lives the code you reference above is to help differentiate between when a game is truly over for test agent after loss of life and when a game is still playing for test agent after loss of life.