bc4rl

Bisimulation Critic for Reinforcement Learning

Usage:

python train.py <algo> <policy> <env> -d <device>

Example:

python train.py bsac BSACMlpPolicy lunarlander -d cuda:0