update preset map for ViT model

[sonali-kumari1]

This PR updates PRESET_MAP for ViT model to align with the official naming convention used in Keras and Kaggle documentation.

[gemini-code-assist]

Summary of Changes

Hello @sonali-kumari1, 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 standardizes the naming conventions for Vision Transformer (ViT) model presets. By updating the keys in the PRESET_MAP dictionary, the changes ensure that the internal naming aligns with the official conventions used in Keras and Kaggle documentation, thereby improving clarity and maintainability for users interacting with these model checkpoints.

Highlights

• Preset Map Renaming: The keys within the PRESET_MAP dictionary for ViT models have been updated to align with official Keras and Kaggle naming conventions. This includes appending _imagenet for ImageNet-1k trained models and __imagenet21k for ImageNet-21k trained models.
• Example Usage Update: The example command in the script's docstring has been modified to reflect the new vit_base_patch16_224_imagenet preset name, ensuring consistency with the updated PRESET_MAP.

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 updates the preset names for the ViT model in the checkpoint conversion script to align with the official naming convention. While the changes for ImageNet-1k presets are correct, there is a critical inconsistency in the naming of ImageNet-21k presets. The script uses a double underscore (e.g., __imagenet21k) while the official preset definitions use a single underscore. This will lead to incorrectly named presets. I've provided a suggestion to fix this.

[sachinprasadhs]

Thanks for the PR, are all the models with the corrected name are loading properly both Keras and HF models?

[sonali-kumari1]

Thanks for the PR, are all the models with the corrected name are loading properly both Keras and HF models?

@sachinprasadhs, Few HF models are failing to load due to SafetensorError and FileNotFoundError.