ikostrikov / pytorch-a3c

PyTorch implementation of Asynchronous Advantage Actor Critic (A3C) from "Asynchronous Methods for Deep Reinforcement Learning".
MIT License
1.23k stars 279 forks source link

NotImplementedError #66

Open ebasatemesgen opened 5 years ago

ebasatemesgen commented 5 years ago

hi, I wanted to test your code on my platform but there seems to be an error. Can you please help me fix it, I have attached the error log. Thank you.

error log.txt

jp18813100494 commented 5 years ago

I missed the same error "NotimplentmentedError". Can u tell me how to handle this?

verystrongjoe commented 5 years ago

instead of using create_atari_env() for creating gym enviornment, you need to use gym.make()

ebasatemesgen commented 5 years ago

Thanks a lot, it seems like its working now.

On Sat, 12 Oct 2019 at 10:58, UK, Jo notifications@github.com wrote:

instead of using create_atari_env() for creating gym enviornment, you need to use gym.make()

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/ikostrikov/pytorch-a3c/issues/66?email_source=notifications&email_token=ALRVVJP35IZCAQ74TMN3CTLQOF7UBA5CNFSM4ISUFMGKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEBBY63I#issuecomment-541298541, or unsubscribe https://github.com/notifications/unsubscribe-auth/ALRVVJPYVDDDE46DIIRYOKDQOF7UBANCNFSM4ISUFMGA .

gxywy commented 5 years ago

I missed the same error "NotimplentmentedError". How to solve it?

gxywy commented 5 years ago

instead of using create_atari_env() for creating gym enviornment, you need to use gym.make()

not work for me

gxywy commented 5 years ago

I figure out!

use gym.ObservationWrapper.__init__(self, env) instead of super(AtariRescale42x42, self).__init__(env) and super(NormalizedEnv, self).__init__(env)

then use def observation(self, observation): instead of def _observation(self, observation):

in envs.py