[tune] Clean up temporary checkpoint directories for class trainables (ex: RLlib) #44366
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
Signed-off-by: Justin Yu <[email protected]>
matthewdeng
reviewed
Apr 19, 2024
| # NOTE: There's no need to supply the checkpoint directory inside | ||
| # the local trial dir, since it'll get persisted to the right location. | ||
| @contextmanager | ||
| def _create_checkpoint_dir(self, checkpoint_dir: Optional[str] = None): |
There was a problem hiding this comment.
Is this still used anywhere?
There was a problem hiding this comment.
Oh yeah I can delete this now.
…nup_cls_trainable_tmpdirs Signed-off-by: Justin Yu <[email protected]>
Signed-off-by: Justin Yu <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Why are these changes needed?
Class
Trainables currently create a temporary directory and provides it as the argument to the user-definedsave_checkpoint. The temporary checkpoint directory gets persisted to storage after the user code executes, but it never gets cleaned up after. This leads to a build-up of many checkpoint "staging" folders that need to be manually deleted. RLlib is affected due toAlgorithms being class trainables.This PR cleans up the checkpoint folders right after they've been copied to storage.
Related issue number
Closes #43092
Checks
git commit -s) in this PR.scripts/format.shto lint the changes in this PR.method in Tune, I've added it in
doc/source/tune/api/under thecorresponding
.rstfile.