feat(generativeai): Create genai_sdk_supervised_checkpoints_set_default_example.py

[yishan-pu]

Add an example to set default checkpoints for tuned models.

[gemini-code-assist]

Hello @yishan-pu, 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

This pull request adds a new example, genai_sdk_supervised_checkpoints_set_default_example.py, to demonstrate how to set default checkpoints for tuned models using the Gemini API. The example retrieves a tuned model, prints its default checkpoint and endpoint, updates the default checkpoint ID, and then prints the updated checkpoint and endpoint.

Highlights

• New Example: A new example script, genai_sdk_supervised_checkpoints_set_default_example.py, has been added to showcase setting default checkpoints for tuned models.
• Setting Default Checkpoints: The example demonstrates how to use the Gemini API to update the default checkpoint ID of a tuned model.
• Tuned Model Information: The example retrieves and prints the default checkpoint ID and endpoint of a tuned model before and after updating the checkpoint.

Changelog

• generative_ai/model_tuning/genai_sdk_supervised_checkpoints_set_default_example.py
  • Created a new example file to demonstrate setting default checkpoints for tuned models.
  • The example uses google.genai to interact with the Gemini API.
  • The example retrieves a tuned model and updates its default checkpoint ID.
  • The example prints the default checkpoint ID and endpoint of the tuned model before and after the update.

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.

---

A checkpoint's ID,
A model's guiding light,
Defaults now can shift.

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. ↩

[snippet-bot]

Here is the summary of possible violations 😱

There is a possible violation for not having product prefix.

• generative_ai/model_tuning/genai_sdk_supervised_checkpoints_set_default_example.py:23, tag genaisdk_gemini_tuning_checkpoints_set_default

The end of the violation section. All the stuff below is FYI purposes only.

---

Here is the summary of changes.

You are about to add 1 region tag.

• generative_ai/model_tuning/genai_sdk_supervised_checkpoints_set_default_example.py:23, tag genaisdk_gemini_tuning_checkpoints_set_default

---

This comment is generated by snippet-bot.
If you find problems with this result, please file an issue at:
https://github.com/googleapis/repo-automation-bots/issues.
To update this comment, add snippet-bot:force-run label or use the checkbox below:

• Refresh this comment

[gemini-code-assist]

Code Review

The code adds an example to set default checkpoints for tuned models using the genai SDK. The example seems straightforward and covers the basic functionality. However, there are a few areas where the code can be improved for clarity and robustness.

Summary of Findings

• Hardcoded values: The example uses hardcoded project and tuning job names. It would be better to use environment variables or configuration files to make the example more flexible and reusable.
• Error handling: The example lacks error handling. It should include try-except blocks to catch potential exceptions and provide informative error messages.
• Missing test case: There is no test case for this example. A test case should be added to ensure the example works as expected and to prevent regressions.

Merge Readiness

The code is functional but needs improvements in error handling, configuration, and testing. I recommend addressing these issues before merging. I am unable to directly approve the pull request, and other reviewers should review and approve this code before merging.

[gemini-code-assist]

It's good practice to include error handling. Wrap the API calls in try...except blocks to catch potential exceptions and provide informative error messages. This will make the example more robust and easier to debug.

  try:
    tuning_job = client.tunings.get(name=name)
    tuned_model = client.models.get(model=tuning_job.tuned_model.model)
  except Exception as e:
    print(f"Error getting tuning job or model: {e}")
    return None

[gemini-code-assist]

Add error handling here as well, similar to the previous API calls. This will ensure that the example handles potential errors when updating the model.

  try:
    tuned_model = client.models.update(
        model=tuned_model.name,
        config=types.UpdateModelConfig(default_checkpoint_id="1"),
    )
  except Exception as e:
    print(f"Error updating model: {e}")
    return None

[gemini-code-assist]

Consider using environment variables for the name to make the example more configurable and avoid hardcoding project-specific information. This would allow users to easily adapt the example to their own projects without modifying the code directly.

  name = os.getenv("TUNING_JOB_NAME", "projects/123456789012/locations/us-central1/tuningJobs/123456789012345")

[gemini-code-assist]

Consider adding a basic test case to supervised_tuning_examples_test.py to verify the functionality of this example. This will help ensure that the example works as expected and prevent regressions in the future.