Skip to content

clean and fix for new arch#1761

Merged
XuehaoSun merged 6 commits intomainfrom
hengguo/fix_refactor
Apr 30, 2026
Merged

clean and fix for new arch#1761
XuehaoSun merged 6 commits intomainfrom
hengguo/fix_refactor

Conversation

@n1ck-guo
Copy link
Copy Markdown
Contributor

Description

Please briefly describe your main changes, the motivation.

Type of Change

Bug fix

Related Issues

Fixes or relates to #

Checklist Before Submitting

  • My code has been tested locally.
  • Documentation has been updated as needed.
  • New or updated tests are included where applicable.
  • The CUDA CI has passed. You can trigger it by commenting /azp run Unit-Test-CUDA-AutoRound.

Signed-off-by: n1ck-guo <heng.guo@intel.com>
Copy link
Copy Markdown
Contributor

Copilot AI left a comment

Choose a reason for hiding this comment

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

Pull request overview

This PR aims to clean up and fix parts of the “new arch” compressor pipeline, mainly around model patch/rotation handling and log verbosity during model monkey-patching.

Changes:

  • Reduce verbosity of MiMo attention patch logging by switching several logger.info calls to logger.debug.
  • Refactor rotation handling in BaseCompressor by renaming transform_configsrotation_configs and moving rotation application into a dedicated _apply_rotations() phase called during post_init().
  • Adjust calibration dataset resolution logic (but the current change introduces a functional regression; see comments).

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.

File Description
auto_round/utils/common.py Lowers log level for MiMo attention forward patching messages.
auto_round/compressors_new/base.py Refactors rotation application into its own phase and renames related config storage; modifies calibration dataset selection logic.

Comment thread auto_round/compressors_new/base.py Outdated
Comment thread auto_round/compressors_new/base.py
n1ck-guo and others added 5 commits April 29, 2026 13:39
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Signed-off-by: n1ck-guo <heng.guo@intel.com>
Signed-off-by: n1ck-guo <heng.guo@intel.com>
This reverts commit b63dfee.
@xin3he
Copy link
Copy Markdown
Contributor

xin3he commented Apr 30, 2026

/azp run Unit-Test-CUDA-AutoRound

@azure-pipelines
Copy link
Copy Markdown

Azure Pipelines successfully started running 1 pipeline(s).

@XuehaoSun XuehaoSun merged commit 1c615f1 into main Apr 30, 2026
40 of 42 checks passed
@XuehaoSun XuehaoSun deleted the hengguo/fix_refactor branch April 30, 2026 13:24
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants