Add filter function to XNNPack Quantizer #10626
Merged
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/10626
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 251698a with merge base e912c65 ( 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: D73677442 |
mcr229
approved these changes
May 1, 2025
pssrawat
added a commit
to pssrawat/executorch
that referenced
this pull request
May 1, 2025
Summary: Like HTP quantizer, add support so that user can specify a filter function to xnnpack quantizer. If specified, we only quantize nodes that return True for the filter function as well. This allows a much finer control on how we quantize a graph. For multichannel ASR, we don't want to quantize certain nodes in certain layers of the encoder. These nodes don't have a proper module_name, so having a proper controlled suppression of quantization for such nodes is not feasible without a filter function. Reviewed By: mcr229 Differential Revision: D73677442
Summary: Pull Request resolved: pytorch#10626 Like HTP quantizer, add support so that user can specify a filter function to xnnpack quantizer. If specified, we only quantize nodes that return True for the filter function as well. This allows a much finer control on how we quantize a graph. For multichannel ASR, we don't want to quantize certain nodes in certain layers of the encoder. These nodes don't have a proper module_name, so having a proper controlled suppression of quantization for such nodes is not feasible without a filter function. Reviewed By: mcr229 Differential Revision: D73677442
|
This pull request was exported from Phabricator. Differential Revision: D73677442 |
jhelsby
pushed a commit
to jhelsby/executorch
that referenced
this pull request
May 9, 2025
Differential Revision: D73677442 Pull Request resolved: pytorch#10626
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:
Like HTP quantizer, add support so that user can specify a filter function to xnnpack quantizer. If specified, we only quantize nodes that return True for the filter function as well. This allows a much finer control on how we quantize a graph.
For multichannel ASR, we don't want to quantize certain nodes in certain layers of the encoder. These nodes don't have a proper module_name, so having a proper controlled suppression of quantization for such nodes is not feasible without a filter function.
Differential Revision: D73677442