search

x35f / unstable_baselines

Re-implementations of SOTA RL algorithms.
127 stars 12 forks source link

UPDATE: support gym 0.26 #47

Closed x35f closed 1 year ago

x35f commented 1 year ago

Support the "truncated" return in env.step Fixed some bugs in PPO The AntTruncatedEnv and HumanoidTruncated env from MBPO are no longer supported by the new version of gym, please use the "old_gym" branch to run MBPO in these two tasks.