This repository has been archived by the owner on Sep 18, 2024. It is now read-only.
[Retiarii] Policy-based RL Strategy #3650
Merged
Conversation
This file contains 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
liuzhe-lz
approved these changes
May 21, 2021
QuanluZhang
reviewed
May 25, 2021
| Takes ``ModelEvaluationEnv`` as input and return a policy. See ``_default_policy_fn`` for an example. | ||
| asynchronous : bool | ||
| If true, in each step, collector won't wait for all the envs to complete. | ||
| This should generally not affect the result, but might affect the efficiency. Note that a slightly more trials |
There was a problem hiding this comment.
don't understand, why asynchronous does not affect the result?
There was a problem hiding this comment.
Synchronous doesn't mean single-process sampling. Both synchronous and asynchronous has parallelism. "Asynchronous" induces a mechanism to give up on some environment when it's not finished.
Refer to https://tianshou.readthedocs.io/en/master/tutorials/cheatsheet.html#parallel-sampling if you feel interested. It's a bit complicated and I don't think I can make it clear here in a few words.
QuanluZhang
approved these changes
May 25, 2021
49 tasks
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.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.
This PR supports a family of RL strategy based on tianshou. The default built-in algorithm is PPO.
TODOs: