mila-iqia / atari-representation-learning

Code for "Unsupervised State Representation Learning in Atari"
https://arxiv.org/abs/1906.08226
MIT License
240 stars 50 forks source link

Which version of gym[atari] was used? #74

Open Sopralapanca opened 8 months ago

Sopralapanca commented 8 months ago

I'm trying to test this implementation and I installed the packages according to the readme. Also, I have installed gym[accept-rom-licence] correctly.

I get this error

File "path-to-package/atariari/benchmark/wrapper.py", line 7, in step
    observation, reward, done, info = self.env.step(action)

This is probably because gym was updated and now step returns 5 values instead of 4.

So which version of gym was used for this project? I need it to install the correct version