[ET-VK] Add custom VkInt4WeightOnlyQuantizer for vulkan
#6234
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
## Context This diff adds the `VkInt4WeightOnlyQuantizer` class which enables 4-bit quantization of linear layers via source transformation. This quantizer class is copied from `torchao.quantization.GPTQ.WeightOnlyInt4Linear` with some minor changes as annotated in the implementation. Note that the pt2e quantization flow does not yet support groupwise quantization, so source transformation is the only way to perform groupwise quantization at the moment. Differential Revision: [D64406457](https://our.internmc.facebook.com/intern/diff/D64406457/) [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/6234
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit d2e8d29 with merge base 8673567 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
pytorch-bot
bot
added
ciflow/periodic
module: vulkan
facebook-github-bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D64406457 |
This was referenced Oct 15, 2024
## Context This diff adds the `VkInt4WeightOnlyQuantizer` class which enables 4-bit quantization of linear layers via source transformation. This quantizer class is copied from `torchao.quantization.GPTQ.WeightOnlyInt4Linear` with some minor changes as annotated in the implementation. Note that the pt2e quantization flow does not yet support groupwise quantization, so source transformation is the only way to perform groupwise quantization at the moment. Differential Revision: [D64406457](https://our.internmc.facebook.com/intern/diff/D64406457/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D64406457 |
## Context This diff adds the `VkInt4WeightOnlyQuantizer` class which enables 4-bit quantization of linear layers via source transformation. This quantizer class is copied from `torchao.quantization.GPTQ.WeightOnlyInt4Linear` with some minor changes as annotated in the implementation. Note that the pt2e quantization flow does not yet support groupwise quantization, so source transformation is the only way to perform groupwise quantization at the moment. Differential Revision: [D64406457](https://our.internmc.facebook.com/intern/diff/D64406457/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D64406457 |
## Context This diff adds the `VkInt4WeightOnlyQuantizer` class which enables 4-bit quantization of linear layers via source transformation. This quantizer class is copied from `torchao.quantization.GPTQ.WeightOnlyInt4Linear` with some minor changes as annotated in the implementation. Note that the pt2e quantization flow does not yet support groupwise quantization, so source transformation is the only way to perform groupwise quantization at the moment. Differential Revision: [D64406457](https://our.internmc.facebook.com/intern/diff/D64406457/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D64406457 |
junpi3
approved these changes
Oct 16, 2024
|
This pull request has been merged in 58ee33d. |
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.
Stack from ghstack (oldest at bottom):
export_llama#6235VkInt4WeightOnlyQuantizerfor vulkan #6234Context
This diff adds the
VkInt4WeightOnlyQuantizerclass which enables 4-bit quantization of linear layers via source transformation. This quantizer class is copied fromtorchao.quantization.GPTQ.WeightOnlyInt4Linearwith some minor changes as annotated in the implementation.Note that the pt2e quantization flow does not yet support groupwise quantization, so source transformation is the only way to perform groupwise quantization at the moment.
Differential Revision: D64406457