Update repo to work with gymnasium #3
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
added 5 commits
February 11, 2023 14:17
…ustom defined game settings (from README)
…cv-python, tensorflow, gym (all which resulted in errors/were unable to install with newest version)
…structive compared to original file, updated custom env settings to match README
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Figured I would add this in case anyone else wants to use this repo with gymnasium. Gym is no longer being maintained and it is recommended that everyone switch to gymnasium instead, which has backward compatibility. This also enables use with other RL libraries Tianshou (already transitioned to gymnasium) and future compatibility with RLlib and stable_baselines3 (planning to transition afaik).
There are a few warnings due to new standards like flattening observation space, returning numpy arrays rather than tensors, defining a render_fps tag (found this made rendering a bit glitchy probably due to matplotlib), and returning extra "info" objects with the step function, but those are all minor and can be ignored.