Qualcomm AI Engine Direct - Adding QNN backend support for reflection_pad3d core ATen op - #21502
Open
qti-horodnic wants to merge 1 commit into
Open
Qualcomm AI Engine Direct - Adding QNN backend support for reflection_pad3d core ATen op#21502qti-horodnic wants to merge 1 commit into
qti-horodnic wants to merge 1 commit into
Conversation
…_pad3d core ATen op
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/21502
Note: Links to docs will display an error until the docs builds have been completed.
|
Contributor
Author
|
@pytorchbot label "release notes: qualcomm" |
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Added support for
aten.reflection_pad3don the QNN HTP backend via a decomposition pass into theindex_select+catoperations.Note that QNN HTP's native Pad op with
MIRROR_REFLECTscheme only supports tensors up to rank 4. Sincereflection_pad3doperates on 5D tensors[N, C, D, H, W], it cannot be delegated directly. Using2dor1dvariants of this op in a decomposition is complicated, so this PR decomposesreflection_pad3dintoindex_selectandcat, which are both fully supported at 5D rank on HTP.This is done by:
index_selectto gather mirrored indices along the dimension.catto concatenate the reflected slices with the original tensor.Test plan