Add 16A8W quantization configuration utility for ARM backend #13898
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
This diff implements a 16A8W (16-bit activations, 8-bit weights) quantization configuration utility for the ExecutorTorch ARM backend, following the feedback from D79746479. ## Key Changes **1. New Quantization Configuration Function** - Add `get_16a8w_quantization_config()` in `fbcode/executorch/backends/arm/quantizer/arm_quantizer.py` - Provides 16-bit activations with HistogramObserver (better precision than 8A8W) - Maintains 8-bit weights with MinMaxObserver/PerChannelMinMaxObserver (memory efficient) - **Technically supported by TOSA through [EXT-INT16 extension/profile](https://www.mlplatform.org/tosa/tosa_spec.html#_conv2d)** ## Benefits - **Better Precision**: 16-bit activations provide higher precision than 8-bit. Useful for carrying precision for recurring neural nets. ghstack-source-id: 305991462 @exported-using-ghexport @bypass-github-export-checks @bypass-github-pytorch-ci-checks @bypass-github-executorch-ci-checks Differential Revision: [D81550512](https://our.internmc.facebook.com/intern/diff/D81550512/) [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/13898
Note: Links to docs will display an error until the docs builds have been completed. ❌ 2 New FailuresAs of commit 878f63f with merge base ae07cb6 ( NEW FAILURES - The following jobs have failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
meta-cla
bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D81550512 |
This diff implements a 16A8W (16-bit activations, 8-bit weights) quantization configuration utility for the ExecutorTorch ARM backend, following the feedback from D79746479. ## Key Changes **1. New Quantization Configuration Function** - Add `get_16a8w_quantization_config()` in `fbcode/executorch/backends/arm/quantizer/arm_quantizer.py` - Provides 16-bit activations with HistogramObserver (better precision than 8A8W) - Maintains 8-bit weights with MinMaxObserver/PerChannelMinMaxObserver (memory efficient) - **Technically supported by TOSA through [EXT-INT16 extension/profile](https://www.mlplatform.org/tosa/tosa_spec.html#_conv2d)** ## Benefits - **Better Precision**: 16-bit activations provide higher precision than 8-bit. Useful for carrying precision for recurring neural nets. exported-using-ghexport bypass-github-export-checks bypass-github-pytorch-ci-checks bypass-github-executorch-ci-checks Differential Revision: [D81550512](https://our.internmc.facebook.com/intern/diff/D81550512/) [ghstack-poisoned]
Ninja91
added a commit
that referenced
this pull request
Sep 3, 2025
Pull Request resolved: #13898 This diff implements a 16A8W (16-bit activations, 8-bit weights) quantization configuration utility for the ExecutorTorch ARM backend, following the feedback from D79746479. ## Key Changes **1. New Quantization Configuration Function** - Add `get_16a8w_quantization_config()` in `fbcode/executorch/backends/arm/quantizer/arm_quantizer.py` - Provides 16-bit activations with HistogramObserver (better precision than 8A8W) - Maintains 8-bit weights with MinMaxObserver/PerChannelMinMaxObserver (memory efficient) - **Technically supported by TOSA through [EXT-INT16 extension/profile](https://www.mlplatform.org/tosa/tosa_spec.html#_conv2d)** ## Benefits - **Better Precision**: 16-bit activations provide higher precision than 8-bit. Useful for carrying precision for recurring neural nets. ghstack-source-id: 307143911 @exported-using-ghexport ghstack-source-id: 307143911 Differential Revision: [D81550512](https://our.internmc.facebook.com/intern/diff/D81550512/)
|
This pull request was exported from Phabricator. Differential Revision: D81550512 |
shoumikhin
approved these changes
Sep 3, 2025
This diff implements a 16A8W (16-bit activations, 8-bit weights) quantization configuration utility for the ExecutorTorch ARM backend, following the feedback from D79746479. ## Key Changes **1. New Quantization Configuration Function** - Add `get_16a8w_quantization_config()` in `fbcode/executorch/backends/arm/quantizer/arm_quantizer.py` - Provides 16-bit activations with HistogramObserver (better precision than 8A8W) - Maintains 8-bit weights with MinMaxObserver/PerChannelMinMaxObserver (memory efficient) - **Technically supported by TOSA through [EXT-INT16 extension/profile](https://www.mlplatform.org/tosa/tosa_spec.html#_conv2d)** ## Benefits - **Better Precision**: 16-bit activations provide higher precision than 8-bit. Useful for carrying precision for recurring neural nets. exported-using-ghexport bypass-github-export-checks bypass-github-pytorch-ci-checks bypass-github-executorch-ci-checks Differential Revision: [D81550512](https://our.internmc.facebook.com/intern/diff/D81550512/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D81550512 |
This diff implements a 16A8W (16-bit activations, 8-bit weights) quantization configuration utility for the ExecutorTorch ARM backend, following the feedback from D79746479. ## Key Changes **1. New Quantization Configuration Function** - Add `get_16a8w_quantization_config()` in `fbcode/executorch/backends/arm/quantizer/arm_quantizer.py` - Provides 16-bit activations with HistogramObserver (better precision than 8A8W) - Maintains 8-bit weights with MinMaxObserver/PerChannelMinMaxObserver (memory efficient) - **Technically supported by TOSA through [EXT-INT16 extension/profile](https://www.mlplatform.org/tosa/tosa_spec.html#_conv2d)** ## Benefits - **Better Precision**: 16-bit activations provide higher precision than 8-bit. Useful for carrying precision for recurring neural nets. exported-using-ghexport bypass-github-export-checks bypass-github-pytorch-ci-checks bypass-github-executorch-ci-checks Differential Revision: [D81550512](https://our.internmc.facebook.com/intern/diff/D81550512/) [ghstack-poisoned]
This was referenced Sep 4, 2025
|
This pull request was exported from Phabricator. Differential Revision: D81550512 |
|
Hi @Ninja91 please add |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 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.
Stack from ghstack (oldest at bottom):
This diff implements a 16A8W (16-bit activations, 8-bit weights) quantization configuration utility for the ExecutorTorch ARM backend, following the feedback from D79746479.
Key Changes
1. New Quantization Configuration Function
get_16a8w_quantization_config()infbcode/executorch/backends/arm/quantizer/arm_quantizer.pyBenefits
@exported-using-ghexport
@bypass-github-export-checks
@bypass-github-pytorch-ci-checks
@bypass-github-executorch-ci-checks
Differential Revision: D81550512