error msg for using smoothquant without calibration dataset

[ved1beta]

SUMMARY:
check for calibration dataset not present
#1510

[gemini-code-assist]

Hello @ved1beta, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

Summary of Changes

Hello team, Gemini here with a summary of this pull request. This PR, authored by ved1beta, addresses a potential issue when using the SmoothQuant modifier without providing the necessary calibration dataset. The core change introduces a check during the initialization phase of the SmoothQuant modifier to explicitly verify the presence of a calibration_dataloader in the provided state. If this required dataset is missing or None, the code now raises a clear ValueError. This enhancement provides immediate feedback to the user, preventing potential downstream failures or unexpected behavior that could occur if the calibration data was silently absent.

Highlights

• Error Handling: Implemented a specific check and raised a ValueError if the SmoothQuant modifier is initialized without a required calibration dataset.
• Input Validation: Ensures that the calibration_dataloader is available in the state object before proceeding with SmoothQuant initialization.

Changelog

• src/llmcompressor/modifiers/smoothquant/base.py
  • Added a check in the on_initialize method to see if state.calibration_dataloader exists and is not None (lines 130-137).
  • Raises a ValueError with a descriptive message if the calibration dataset is missing (lines 134-136).

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point in your pull request via creating an issue comment (i.e. comment on the pull request page) using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in issue comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist is currently in preview and may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments to provide feedback.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request introduces a validation check in the SmoothQuantModifier to ensure that a calibration dataset is available. This is a good addition.

Summary of Findings

• Calibration Data Access: The PR checks for state.calibration_dataloader to ensure a calibration dataset is present. However, the standard location for calibration data within the State object (as defined in llmcompressor.core.state.py) appears to be state.data.calib. Using a non-standard attribute state.calibration_dataloader might lead to incorrect validation, potentially raising errors when valid calibration data exists via state.data.calib, or vice-versa. This should be clarified or corrected to use the standard state.data.calib.

Merge Readiness

The pull request addresses an important validation step for SmoothQuantModifier. However, there's a significant concern regarding how the calibration data's presence is checked within the State object. This could lead to incorrect behavior. I recommend addressing the high-severity issue identified in the review comments before merging. As an AI, I am not authorized to approve pull requests; please ensure further review and approval from the maintainers after addressing the feedback.

[github-actions]

👋 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.

[gemini-code-assist]

Warning

You have reached your daily quota limit. Please wait up to 24 hours and I will start processing your requests again!

[kylesayrs]

Thank you for the contribution @ved1beta !

In the future, this may be incorporated into an earlier recipe validation step, but for now this will go a long way to helping users use smoothquant!

[kylesayrs]

Waiting for post release before merging