laszukdawid / ai-traineree

PyTorch agents and tools for (Deep) Reinforcement Learning

http://ai-traineree.readthedocs.io/

Apache License 2.0

25 stars 6 forks source link

bugfix examples/multi_agent/prison_iql.py #12

Closed stevens68 closed 3 years ago

stevens68 commented 3 years ago

Hi, great framework!

Seems to me that the following lines need should be changed in prison_iql.py

8: (version 2 is deprecated)

from pettingzoo.butterfly import prison_v3 as prison

14: to get 1 instead of (1,)

obs_size = ma_task.obs_size[0]

laszukdawid commented 3 years ago

Hi @stevens68,

Thanks for letting me know :) Got caught red-handed that I haven't run all the examples after update. Will update this shortly!

I'd super appreciate letting me know anything else you find. Just finishing a different part of a big project and moving back to AI Traineree shortly.

laszukdawid commented 3 years ago

Closing issue. Feel free to reopen or create a new one if there's more to be done.