[ET-VK] Quantized Int8 Convolution #13817

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Merged

facebook-github-bot merged 10 commits into gh/SS-JIA/317/base from gh/SS-JIA/317/head

Sep 8, 2025

Contributor

SS-JIA commented Aug 29, 2025 •

edited

Loading

Stack from ghstack (oldest at bottom):

See the below diff; this diff implements int8 quantized conv2d using the quantized linear layer introduced below.

Note that the current implementation doesn't yet support depthwise convs; a specialized implementation will need to be added for that.

Differential Revision: D81330809


          [ET-VK] Quantized Int8 Convolution

a052433

See the below diff; this diff implements int8 quantized conv2d using the quantized linear layer introduced below.

Note that the current implementation doesn't yet support depthwise convs; a specialized implementation will need to be added for that.

Differential Revision: [D81330809](https://our.internmc.facebook.com/intern/diff/D81330809/)

[ghstack-poisoned]

SS-JIA requested review from kirklandsign and larryliu0820 as code owners

August 29, 2025 16:52

pytorch-bot bot commented Aug 29, 2025 •

edited

Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/13817

📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

❌ 2 New Failures

As of commit 1282461 with merge base 1a7441f ():

NEW FAILURES - The following jobs have failed:

Build documentation / build (buck2) / Build doc (gh)
At least one of the pre-conditions you specified did not hold
pull / android / run-emulator (gh)
The process '/usr/bin/sh' failed with exit code 255

This comment was automatically generated by Dr. CI and updates every 15 minutes.

This was referenced Aug 29, 2025

[ET-VK][ez] Add ability to check for dot product extension support + upgrade glslc #13814

Merged

[ET-VK] Introduce custom op correctness + speed testing suite & add vulkan operator testing to CI #13815

Merged

[ET-VK] Quantized Int8 Linear #13816

Merged

[ET-VK][AOT] Enable exporting Q8 Quantized Linear + Convolution #13818

Merged

github-actions bot commented Aug 29, 2025

This PR needs a `release notes:` label

If your change should be included in the release notes (i.e. would users of this library care about this change?), please use a label starting with release notes:. This helps us keep track and include your important work in the next release notes.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "release notes: none"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

meta-cla bot added the CLA Signed label

Contributor

facebook-github-bot commented Aug 29, 2025

This pull request was exported from Phabricator. Differential Revision: D81330809

facebook-github-bot added the fb-exported label


          Update on "[ET-VK] Quantized Int8 Convolution"

7482e5a

See the below diff; this diff implements int8 quantized conv2d using the quantized linear layer introduced below.

Note that the current implementation doesn't yet support depthwise convs; a specialized implementation will need to be added for that.

Differential Revision: [D81330809](https://our.internmc.facebook.com/intern/diff/D81330809/)

[ghstack-poisoned]

Contributor

facebook-github-bot commented Aug 29, 2025

This pull request was exported from Phabricator. Differential Revision: D81330809


          Update on "[ET-VK] Quantized Int8 Convolution"

ec9b15e

See the below diff; this diff implements int8 quantized conv2d using the quantized linear layer introduced below.

Note that the current implementation doesn't yet support depthwise convs; a specialized implementation will need to be added for that.

Differential Revision: [D81330809](https://our.internmc.facebook.com/intern/diff/D81330809/)

[ghstack-poisoned]

Contributor

facebook-github-bot commented Aug 29, 2025

This pull request was exported from Phabricator. Differential Revision: D81330809


          Update on "[ET-VK] Quantized Int8 Convolution"

97d037e

See the below diff; this diff implements int8 quantized conv2d using the quantized linear layer introduced below.

Note that the current implementation doesn't yet support depthwise convs; a specialized implementation will need to be added for that.

Differential Revision: [D81330809](https://our.internmc.facebook.com/intern/diff/D81330809/)

[ghstack-poisoned]

Contributor

facebook-github-bot commented Aug 30, 2025

This pull request was exported from Phabricator. Differential Revision: D81330809


          Update on "[ET-VK] Quantized Int8 Convolution"

62e4fc3

See the below diff; this diff implements int8 quantized conv2d using the quantized linear layer introduced below.

Note that the current implementation doesn't yet support depthwise convs; a specialized implementation will need to be added for that.

Differential Revision: [D81330809](https://our.internmc.facebook.com/intern/diff/D81330809/)

[ghstack-poisoned]

SS-JIA mentioned this pull request

[ET-VK][ez] Enable max_pool2d.default #13879

Merged

Contributor

facebook-github-bot commented Sep 2, 2025

This pull request was exported from Phabricator. Differential Revision: D81330809


          Update on "[ET-VK] Quantized Int8 Convolution"

86581a6

See the below diff; this diff implements int8 quantized conv2d using the quantized linear layer introduced below.

Note that the current implementation doesn't yet support depthwise convs; a specialized implementation will need to be added for that.

Differential Revision: [D81330809](https://our.internmc.facebook.com/intern/diff/D81330809/)

[ghstack-poisoned]

Contributor

facebook-github-bot commented Sep 4, 2025

This pull request was exported from Phabricator. Differential Revision: D81330809


          Update on "[ET-VK] Quantized Int8 Convolution"

1aa9a15

See the below diff; this diff implements int8 quantized conv2d using the quantized linear layer introduced below.

Note that the current implementation doesn't yet support depthwise convs; a specialized implementation will need to be added for that.

Differential Revision: [D81330809](https://our.internmc.facebook.com/intern/diff/D81330809/)

[ghstack-poisoned]

Contributor

facebook-github-bot commented Sep 4, 2025

This pull request was exported from Phabricator. Differential Revision: D81330809


          Update on "[ET-VK] Quantized Int8 Convolution"

8cae8d5

See the below diff; this diff implements int8 quantized conv2d using the quantized linear layer introduced below.

Note that the current implementation doesn't yet support depthwise convs; a specialized implementation will need to be added for that.

Differential Revision: [D81330809](https://our.internmc.facebook.com/intern/diff/D81330809/)

[ghstack-poisoned]

This was referenced Sep 5, 2025

[ET-VK] Fast path for choose_qparams #14019

Merged

[ET-VK] Implement linear_q4gsw #14020

Merged

Contributor

facebook-github-bot commented Sep 5, 2025

This pull request was exported from Phabricator. Differential Revision: D81330809

manuelcandales approved these changes

View reviewed changes

ssjia and others added 2 commits

September 7, 2025 10:54


          Update on "[ET-VK] Quantized Int8 Convolution"

1a48ceb

See the below diff; this diff implements int8 quantized conv2d using the quantized linear layer introduced below.

Note that the current implementation doesn't yet support depthwise convs; a specialized implementation will need to be added for that.

Differential Revision: [D81330809](https://our.internmc.facebook.com/intern/diff/D81330809/)

[ghstack-poisoned]


          Update on "[ET-VK] Quantized Int8 Convolution"

See the below diff; this diff implements int8 quantized conv2d using the quantized linear layer introduced below.

Note that the current implementation doesn't yet support depthwise convs; a specialized implementation will need to be added for that.

Differential Revision: [D81330809](https://our.internmc.facebook.com/intern/diff/D81330809/)

[ghstack-poisoned]

Contributor

facebook-github-bot commented Sep 7, 2025

This pull request was exported from Phabricator. Differential Revision: D81330809

facebook-github-bot merged commit 4eb279c into gh/SS-JIA/317/base

113 of 118 checks passed

facebook-github-bot deleted the gh/SS-JIA/317/head branch

September 8, 2025 00:05

facebook-github-bot temporarily deployed to cherry-pick-bot

September 8, 2025 00:05

— with

GitHub Actions Inactive

pytorchbot mentioned this pull request

[ET-VK] Quantized Int8 Convolution #14042

Merged

SS-JIA pushed a commit that referenced this pull request


          [ET-VK] Quantized Int8 Convolution

5dfcaa5

Pull Request resolved: #13817

See the below diff; this diff implements int8 quantized conv2d using the quantized linear layer introduced below.

Note that the current implementation doesn't yet support depthwise convs; a specialized implementation will need to be added for that.
ghstack-source-id: 308092875
@exported-using-ghexport

Differential Revision: [D81330809](https://our.internmc.facebook.com/intern/diff/D81330809/)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

CLA Signed fb-exported