feat: add empty_mps_cache handler for MPS device memory clearing#3765
Draft
Aanushka001 wants to merge 1 commit into
Draft
feat: add empty_mps_cache handler for MPS device memory clearing#3765Aanushka001 wants to merge 1 commit into
Aanushka001 wants to merge 1 commit into
Conversation
| def empty_mps_cache(_: Engine) -> None: | ||
| if hasattr(torch, "mps") and hasattr(torch.mps, "empty_cache"): | ||
| torch.mps.empty_cache() | ||
| import gc |
Collaborator
There was a problem hiding this comment.
imports should be on top
|
|
||
|
|
||
| def empty_mps_cache(_: Engine) -> None: | ||
| if hasattr(torch, "mps") and hasattr(torch.mps, "empty_cache"): |
Collaborator
There was a problem hiding this comment.
either we can check in the function or outside it, currently irt checks in both place which make it kinda redundant.
Collaborator
|
Hi, I also think we should add a separate parameter for the mps cache clearing instead of relying on pre_built cuda @vfdev-5 |
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
This PR adds MPS (Apple Metal Performance Shaders) cache clearing support, similar to the existing
empty_cuda_cache()for CUDA devices.Changes
empty_mps_cache()function that clears MPS cache viatorch.mps.empty_cache()_setup_common_training_handlers()when running on MPS devicesRelated
Based on original PR #3763