Robust subprocess sweep #308
Open
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
As described in CUNY-CL#302 this moves to a model where each sweep run is a subprocess, which seems likely to ensure the OS returns memory at the end of each.
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Outdated suggestions cannot be applied.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.
As described in #302, this moves the sweep to a model where each sweep run is a subprocess, which seems likely to ensure the OS returns memory at the end of each run. This model is one we piloted in UDTube and here the implementation is actually simpler because there's no need for temporary config files.
Testing confirms this works and that the data is being logged as expected to W&B, but I'd appreciate if a reviewer can try it out locally with an existing workflow.
One nice side effect of this is we no longer need to maintain the
train.trainmodule interface, which has no place in LightningCLI anyways.Closes #302.