Add the workflow for co-working with the Gemini CLI. #2453
Conversation
Summary of ChangesHello @james77777778, 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 introduces comprehensive guidelines to the Highlights
Using Gemini Code AssistThe 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
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 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
|
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request adds valuable documentation to CONTRIBUTING.md on how to use the Gemini CLI for model development within Keras Hub. The instructions are clear and provide a concrete example with DINOV3. My review includes a couple of suggestions to better align the new workflow with the existing contribution guidelines in the STYLE_GUIDE.md, specifically regarding the required validation artifacts and script naming conventions.
| Next, let's instruct the CLI to construct a numerical validation test to ensure the implementation is correct. Before running this step, make sure you have installed the `transformers` library and have access to the `facebook/dinov3-*` presets. | ||
|
|
||
| ```shell | ||
| # In the CLI. | ||
| Create a numerical validation script `check_dinov3.py` for `DINOV3Backbone` at project root. Use the HF preset `facebook/dinov3-vits16-pretrain-lvd1689m` as a reference for the expected outputs. Remember to port the weights from HF to `DINOV3Backbone` within the script and refer to the existing implementation here: @keras_hub/src/utils/transformers/convert_dinov2.py | ||
| ``` | ||
|
|
||
| Now, instruct the CLI to run the script and correct any errors. If you are working within Conda environments, be sure to also instruct the CLI to use the appropriate environment for execution. | ||
|
|
||
| ```shell | ||
| # In the CLI. | ||
| Run @check_dinov3.py by `KERAS_BACKEND=jax conda run -n keras-hub-jax python check_dinov3.py`. Fix any errors encountered during execution. | ||
| ``` | ||
|
|
||
| During this phase, human intervention is often necessary. You will need to carefully review the CLI's modifications and provide guidance or even handcraft some details that the tool failed to implement correctly. |
There was a problem hiding this comment.
The workflow describes creating a Python script (check_dinov3.py) for numerical validation. While this is a useful step for development, the project's contribution guidelines require a Colab notebook demonstrating numerical equivalence for pull requests.1 Please consider adding a note to clarify that this script is a development tool and that a Colab notebook is the required final artifact for the PR. This will help contributors submit complete pull requests.
Style Guide References
Footnotes
-
The style guide mandates that every pull request must include Colab notebooks for numerical validation against the original model. ↩
james77777778
force-pushed
the
add-coworking-gemini-cli
branch
from
eb9e571 to
ac30d60
Compare
2 participants
Description of the change
This PR adds the detailed Gemini CLI instructions for implementing
DINOV3:#2444
I'm unsure about the location for this guideline. Let me know if you suggest placing it elsewhere.
cc @divyashreepathihalli
Reference
Colab Notebook
Checklist