M. Ernestus

After reading through the paper, I am using the following hyperparameter search space: | parameter | search space | |---------------------------------|-------------------------------------------------| | active_selection | True/False | | active_selection_oversampling | 2 to...

Hi @mschweizer, thanks a lot for bringing this up. @timbauman is actively working on this! Check out #712 and #711 if you want to stay updated about the progress.

Great! Let me know when the PR is ready for review!

Thanks for the feedback @qxcv! > One last piece of low-level feedback: `imitation_cli.utils` is a non-intuitive place to put configs (shouldn't it be `imitation_cli.configs` or something?). You are right, `utils`...

Hi @Bpoole908. Thanks for reporting this issue. Would you mind providing me with the sampled date in the two formats: - Once in the old `.npz` format - Once in...

@Bpoole908 @nil123532 If you could provide your problematic datasets this would be awesome. Then we can replicate and resolve the issue. Btw.: The v1.0 has some performance improvements in the...

It is not supported right now: https://github.com/HumanCompatibleAI/imitation/issues/651#issuecomment-1420219620 We are currently reworking the way data is handled in general. @AdamGleave do you think there is a quick fix? Should we aim...

As long as you don't plan on treating different parts of the observation space (e.g. applying a CNN to image data but taking in accelerator data directly) you should be...

Thanks a lot @taufeeque9 for adding this change. We will need it for #675 ! For my understanding: does the table show comparisons to the previous version of the implementation?

Hi @timokau. Thanks a lot for this hint! We will review the PC implementation in the coming weeks and then this information will be really valuable!