search

SforAiDl / genrl

A PyTorch reinforcement learning library for generalizable and reproducible algorithm implementations with an aim to improve accessibility in RL
https://genrl.readthedocs.io
MIT License
403 stars 59 forks source link

Add support for depricated gym 0.10.5 to support multiagent-particle-env #325

Closed AdityaKapoor74 closed 3 years ago

AdityaKapoor74 commented 4 years ago

Will need to have support for gym version 0.10.5 as it is a need for multiagent-particle-env repo to work fine. Some issues are to fix imports in various files. For example gym.spaces.Space is gym.Space in 0.10.5

Sharad24 commented 4 years ago

Can you check if the multi agent algorithms work on something that has support for recent versions of gym? E.g. https://github.com/koulanurag/ma-gym

If they do work, I dont see a reason for adding support for deprecated softwares.