This repository was archived by the owner on Jun 3, 2025. It is now read-only.
Convert OPT MatMuls with quantized inputs to MatMulInteger #1585
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
dbogunowicz
previously approved these changes
May 29, 2023
|
Also, please update the PR description. |
Added the description |
…NX_export_OPT_matmuls
anmarques
approved these changes
Jun 7, 2023
src/sparseml/pytorch/sparsification/quantization/quantize_qat_export.py
Outdated
Show resolved
Hide resolved
bfineran
approved these changes
Jun 8, 2023
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.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.
This change enables converting the torch.bmm with two quantized inputs and non-quantized output to MatMulInteger. This procedure is mutually exclusive with the existing one for QATMatMul-based quantized matmuls.
The attached graph shows two MatMulInteger that are the results of this conversion on OPT-125m.
Note that the quantization of these MatMuls on OPT requires this PR.