rahulkidambi / mobile-neurips2021 Goto Github PK

View Code? Open in Web Editor NEW

6.0 6.0 2.0 2.35 MB

This is the code repository for MobILE: Model-Based Imitation Learning from Observation Alone (NeurIPS 2021)

Shell 0.13% Python 99.77% Dockerfile 0.10%

mobile-neurips2021's People

Contributors

Stargazers

Watchers

Forkers

nathangavenski soroushmehr

mobile-neurips2021's Issues

Will there be a release for this work?

Hi, sorry for bothering you with that question.

But will there be a release in this repository for this work from NeurIPS?

Expert data

Hi guys, sorry for bothering you again. I'm really glad you guys released the code! But I cant seem to find an example of your expert data structure. Would you be able to upload just a short example?

Using MOBILE with discrete non MuJoCo envs

Hi guys, sorry for bothering you again, but I want to use MOBILE on my research and I'm trying to run him on classic controls tasks and other discrete environments.

I've tried adapting your current code to work with CartPole with no success. After the whole training procedure, the policy was outputting an average of 10 reward points. Is it possible to release the code that you used for the CartPole in your work? I would like to check what I'm doing wrong and have a base for adapting your algorithm for new environments.

Thanks!

Recommend Projects