search

Farama-Foundation / D4RL

A collection of reference environments for offline reinforcement learning
Apache License 2.0
1.32k stars 279 forks source link

CustomMDPPathCollector #9

Closed Jiankai-Sun closed 4 years ago

Jiankai-Sun commented 4 years ago

Firstly, thank you.

After python antmaze_bear.py, image

It seems that rlkit (already installed the latest version) doesn't contain CustomMDPPathCollector. Is this your custom class?

ruizhaogit commented 4 years ago

Thank you for the great work! @justinjfu
I also ran into the same error @Jiankai-Sun . It seems that "VAEPolicy" is also missing. Which "rlkit" fork are you using @justinjfu or is there an "rlkit" fork that has not been released yet?

aviralkumar2907 commented 4 years ago

Thanks for pointing out this issue.

We will release the rlkit fork we built off of soon, meanwhile, we have added support for D4RL in the BEAR repo, which can be found here: https://github.com/aviralkumar2907/BEAR, and that should run without any issues.

ruizhaogit commented 4 years ago

Couldn't reproduce the results reported in D4RL, see post here https://github.com/aviralkumar2907/BEAR/issues/5 @aviralkumar2907 @justinjfu