Enable quantized add #8584
Merged
Enable quantized add #8584
Conversation
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/8584
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New FailureAs of commit 1a3bf95 with merge base e5dc18a ( NEW FAILURE - The following job has failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D69441041 |
Summary: Add a pattern for add ops, and call the existing quantized add kernel. Note that it's not capable of broadcasting yet, so we add an explicit check for sizes. Since that change is not guaranteed to improve performance but can reduce accuracy, we don't include it in the default quantizer and instead introduce a quantizer for WakeWord. In the future, we probably will want to maintain multiple custom quantizers for models in the same way we do here. If we want them to live somewhere else than quantizer.py (which would bloat), we can extract that in a later diff. Note that WW stateful tests are broken already, so they show as red in the test box. Reviewed By: zonglinpeng Differential Revision: D69441041
facebook-github-bot
force-pushed
the
export-D69441041
branch
from
0b0abf6 to
1a3bf95
Compare
|
This pull request was exported from Phabricator. Differential Revision: D69441041 |
zonglinpeng
approved these changes
Feb 20, 2025
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Outdated suggestions cannot be applied.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:
Add a pattern for add ops, and call the existing quantized add kernel. Note that it's not capable of broadcasting yet, so we add an explicit check for sizes.
Since that change is not guaranteed to improve performance but can reduce accuracy, we don't include it in the default quantizer and instead introduce a quantizer for WakeWord.
In the future, we probably will want to maintain multiple custom quantizers for models in the same way we do here. If we want them to live somewhere else than quantizer.py (which would bloat), we can extract that in a later diff.
Note that WW stateful tests are broken already, so they show as red in the test box.
Reviewed By: zonglinpeng
Differential Revision: D69441041