[Relax] Fix batch normalization computation logic

[Asuka0630]

Dear reviewers,

Why
The previous implementation of batch_norm incorrectly conflated the computation of mean and variance between training and evaluation modes. Additionally, for '_native_batch_norm_legit.no_stats‘, using instance_norm to handle normalization ignored the batch dimension, leading to incorrect behavior.

How
This PR includes the following fixes:

1. Corrects the computation logic to properly distinguish between training and evaluation modes.
2. Ensures the batch dimension is properly accounted for in _batch_norm_legit_no_stats.

Environment
GPU: NVIDIA A100-SXM4-80GB

[gemini-code-assist]

Summary of Changes

Hello @Asuka0630, 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!

This pull request resolves critical issues in the batch normalization implementation within the Relax frontend for PyTorch. It ensures that the mean and variance are computed correctly based on whether the model is in training or evaluation mode, and it rectifies the handling of the batch dimension in instance normalization when no statistics are provided, leading to more accurate and robust model behavior.

Highlights

• Batch Normalization Logic Correction: The computation logic for batch normalization has been corrected to properly distinguish between training and evaluation modes, preventing incorrect conflation of mean and variance calculations.
• Instance Normalization Batch Dimension Fix: The _batch_norm_legit_no_stats function now correctly accounts for the batch dimension when using instance normalization, resolving an issue where it was previously ignored.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

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 by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

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 pull request 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 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. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

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 correctly fixes the batch normalization logic to distinguish between training and evaluation modes, and also adjusts the axes for _batch_norm_legit_no_stats to correctly account for the batch dimension. However, the change in python/tvm/topi/nn/batch_norm.py introduces a critical scoping bug that will cause a NameError. I've provided a fix for this bug. Additionally, I've suggested a comment update for improved code clarity in python/tvm/relax/frontend/torch/exported_program_translator.py.

[Asuka0630]

/gemini review

[gemini-code-assist]

Code Review

This pull request correctly addresses a critical bug in the batch normalization implementation where the logic for training and evaluation modes was swapped. The changes also properly handle the _native_batch_norm_legit.no_stats case from PyTorch by including the batch dimension during normalization. The test files have been updated accordingly to reflect these fixes. Overall, the changes are solid and significantly improve the correctness of the batch norm operator. I have one minor suggestion to improve code clarity and potentially performance in the batch_norm.py file.