Skip to content
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.

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

[GPU] Apply is_non_decompression_multiply() callback only for compressed models #21719

Conversation

sshlyapn
Copy link
Contributor

Details:

  • Fix quantized models performance issue caused by unintentional applying is_non_decompression_multiply() callback to quantized constants

Tickets:

  • 127833

@sshlyapn sshlyapn added category: GPU OpenVINO GPU plugin Code Freeze labels Dec 18, 2023
@sshlyapn sshlyapn added this to the 2023.3 milestone Dec 18, 2023
@sshlyapn sshlyapn requested review from a team as code owners December 18, 2023 07:19
Copy link
Contributor

@vladimir-paramuzov vladimir-paramuzov left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

lgtm if perf is ok

Copy link
Contributor

@e-ddykim e-ddykim left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think initial_transformations_manager.run_passes(func) needs to be called somewhere, doesn't it?

@sshlyapn sshlyapn force-pushed the fix_quantized_models_transformation_pass branch from 51389ea to f48f59c Compare December 18, 2023 07:34
@sshlyapn
Copy link
Contributor Author

I think initial_transformations_manager.run_passes(func) needs to be called somewhere, doesn't it?

Absolutely, thank you!

@p-durandin p-durandin enabled auto-merge (squash) December 18, 2023 07:47
@isanghao
Copy link
Contributor

Target network execution graph was recovered. Did not check the entire test scope.

@sshlyapn sshlyapn force-pushed the fix_quantized_models_transformation_pass branch from 98da00b to 2ef20c9 Compare December 18, 2023 08:10
@p-durandin p-durandin merged commit 032ac89 into openvinotoolkit:master Dec 18, 2023
83 checks passed
Lyamin-Roman pushed a commit to sshlyapn/openvino that referenced this pull request Dec 19, 2023
p-durandin pushed a commit that referenced this pull request Dec 20, 2023
* Revert "[GPU] Apply is_non_decompression_multiply() callback only for compressed models (#21719)"

This reverts commit 032ac89.

* [GPU] Apply MarkDequantizationSubgraph callback function only to compressed models
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
category: GPU OpenVINO GPU plugin Code Freeze
Projects
None yet
Development

Successfully merging this pull request may close these issues.

5 participants