Skip to content
New issue

Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? # to your account

[Proposal] Add MoveActionWrapper #456

Open
wants to merge 5 commits into
base: master
Choose a base branch
from

Conversation

younik
Copy link
Member

@younik younik commented Nov 19, 2024

Description

Whenever I work with Minigrid with full observability, I find the action space more complicated than should be. I believe the action space makes sense for agent view observation, but not with full observability, where using the directions (like FrozenLake) is easier.

As I already coded it for myself, I find it easier to open a PR than an issue but feel free to close it if you believe it is not a common use case.

Type of change

Please delete options that are not relevant.

  • New feature (non-breaking change which adds functionality)
  • This change requires a documentation update

Checklist:

  • I have run the pre-commit checks with pre-commit run --all-files (see CONTRIBUTING.md instructions to set it up)
  • I have commented my code, particularly in hard-to-understand areas
  • I have made corresponding changes to the documentation
  • My changes generate no new warnings
  • I have added tests that prove my fix is effective or that my feature works
  • New and existing unit tests pass locally with my changes

Copy link
Member

@pseudo-rnd-thoughts pseudo-rnd-thoughts left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@younik Apologies for only looking at this now.
I have a couple of questions

  • For debugging, this looks good but for implementation usage, taking extra steps to redirect an agent will affect TimeLimit. Could we cheat the environment by modifying the base agent direction to the intented direction then step forward? Only a single step will be taken in this case.
  • Why is the else case, action - 1?

# for free to join this conversation on GitHub. Already have an account? # to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants