[hip-kernel-provider] Batch Norm forward inference functionality with variance variant & fused activation operations#5566
Open
saikubairkota wants to merge 3 commits intodevelopfrom
Conversation
saikubairkota
added
organization: streamhpc
…ded new plan with variance tensor as input to batchnorm fwd inference kernels
…orm forward inference with variance operation
…inference with activation fusion & added relevant tests
saikubairkota
force-pushed
the
streamhpc/saikubairkota/complete-bn-fwd-inference-functionality
branch
from
1f7c97e to
1586dd6
Compare
Contributor
Author
|
Question for the reviewer: Should we use a consistent namespace throughout because currently we use |
Codecov Report✅ All modified and coverable lines are covered by tests. ❌ Your project status has failed because the head coverage (77.21%) is below the target coverage (80.00%). You can increase the head coverage or adjust the target coverage. Additional details and impacted files@@ Coverage Diff @@
## develop #5566 +/- ##
========================================
Coverage 67.27% 67.27%
========================================
Files 1844 1844
Lines 284014 284014
Branches 39839 39839
========================================
+ Hits 191044 191045 +1
+ Misses 76512 76511 -1
Partials 16458 16458
*This pull request uses carry forward flags. Click here to find out more. 🚀 New features to boost your workflow:
|
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Motivation
This PR completes the Batch Norm Forward Inference implementation in the HIP kernel provider plugin by adding support for the remaining variance-based variant (it takes variance as input instead of the inverse variance). In addition, fused activation support is added in both Batch Norm Forward Inference variants.
Technical Details
BatchnormFwdInferenceWithVariancePlanto support the forward inference variant with variance input.BatchnormFwdInferencePlanto support fused activation operations.BatchnormPlanBuilderto enable graph construction for the new variance variant and the fused operations.Note: The activation functions and fused operations-related code in the plans have been added from our previous work in this PR.
Test Plan
Build the plugin and run all the unit and integration tests with
ninja check.Test Result
All unit and integration tests pass successfully on an MI210.
Submission Checklist