Repository navigation
Qualcomm AI Engine Direct - Broadcast the constant base of quantized pow - #24431
Open
msluszniak wants to merge 1 commit into
Open
msluszniak wants to merge 1 commit into
msluszniak wants to merge 1 commit into
Conversation
LiftConstantScalarOperands lifts the base of pow.Scalar to a one-element constant. Quantized HTP ElementWisePower does not broadcast its first input, so only the first output element is computed and the rest are 0. Expand a constant base to the output shape.
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/24431
Note: Links to docs will display an error until the docs builds have been completed.
|
meta-cla
Bot
added
the
CLA Signed
This PR needs a
|
This branch has not been deployed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Quantized
pow.Scalar(c, x)on HTP returns 0 for every element but the first: the lifted one-element base is not broadcast by ElementWisePower. fp16 is unaffected. This expands a constant base to the output shape.TestQNNQuantizedOperator.test_qnn_backend_pow_scalarreproduces it on SM8850 (QAIRT 2.47): 99 of 100 outputs are 0 before, exact after. Found quantizing RF-DETR, whose sine position embedding computes10000 ** xin the graph.cc @nil-is-all