tristandeleu / pytorch-maml-rl

Reinforcement Learning with Model-Agnostic Meta-Learning in Pytorch
MIT License
827 stars 158 forks source link

Normalized env bugfix #14

Closed vuoristo closed 6 years ago

vuoristo commented 6 years ago

Fixes running variance update bugs in NormalizedObservationWrapper and NormalizedRewardWrapper.

tristandeleu commented 6 years ago

Thank you for the fix! 👍