zuoxingdong / lagom

lagom: A PyTorch infrastructure for rapid prototyping of reinforcement learning algorithms.
MIT License
373 stars 30 forks source link

update envs: more efficient Atari wrapper, NormalizeAction #148

Closed zuoxingdong closed 5 years ago