HandyRL is a handy and simple framework based on Python and PyTorch for distributed reinforcement learning that is applicable to your own environments.
MIT License
282
stars
39
forks
source link
feature: return dict from evaluation function (same key) #338