Reinforcement Learning library based on pytorch. Designed for research and experiments. Platform-agnostic (supports openai, marlo, more to be added)
6
stars
1
forks
source link
implement replay pre-fill for experience replay algorithms #26
Open
lucasosouza opened 5 years ago