Remove GPTQ Group/Dynamic Activation Ordering - #3038
Conversation
Signed-off-by: Roderick Wu <rowu@redhat.com>
|
Important Review skippedAuto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the ⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro Plus Run ID: You can disable this status message by setting the Use the checkbox below for a quick retry:
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
|
👋 Hi! Thank you for contributing to llm-compressor. Please add the ready label when the PR is ready for review. Note: This is required to complete the testing suite, please only add the label once the PR is code complete and local testing has been performed. |
|
The quality checks have failed. Please run |
Merge Protections🟢 Merge protection satisfied — ready to merge. Show 1 satisfied protection🟢 Require one maintainer reviewAll PRs must have at least one approving review from a maintainer before merging.
|
There was a problem hiding this comment.
Code Review
This pull request removes support for the deprecated ActivationOrdering.GROUP ("group") activation ordering strategy and its associated weight_g_idx parameter across the codebase, including documentation, modifiers, quantization logic, and tests. Feedback on the changes highlights three main areas of improvement: a critical correctness bug in quantize_weight where g_idx is not permuted when actorder is set to "static", an outdated return type annotation in quantize_weight that still expects a 5-tuple instead of the updated 2-tuple, and the need to explicitly raise a ValueError when actorder="group" is supplied to prevent silent failures.
|
The quality checks have failed. Please run |
kylesayrs
left a comment
There was a problem hiding this comment.
Looks great, awesome work
kylesayrs
left a comment
There was a problem hiding this comment.
Looks great, awesome work
|
The quality checks have failed. Please run |
Signed-off-by: Roderick Wu <rowu@redhat.com>
|
The quality checks have failed. Please run |
|
The quality checks have failed. Please run |
Signed-off-by: Roderick Wu <rowu@redhat.com>
Removing GPTQ activation order dynamic/group (already deprecated). Goes along with compressed-tensors change: vllm-project/compressed-tensors#840 --------- Signed-off-by: Roderick Wu <rowu@redhat.com> Co-authored-by: Roderick Wu <rowu@redhat.com> Co-authored-by: Dipika Sikka <dipikasikka1@gmail.com>
Removing GPTQ activation order dynamic/group (already deprecated). Goes along with compressed-tensors change:
vllm-project/compressed-tensors#840