Add example Zero-shot Image Classification with SigLIP / CLIP from scratch using KerasCV and KerasNLP

[awsaf49]

Issue Type

Feature Request

Source

source

Keras Version

Keras 3

Custom Code

Yes

OS Platform and Distribution

No response

Python version

No response

GPU model and memory

No response

Current Behavior?

Having an example of Zero-shot Image Classification with SigLIP or CLIP from scratch using both KerasCV and KerasNLP would be a great addition.

I have already published a notebook on Kaggle using Keras 3.0. I would love to add this example to keras.io.

This example is only having an issue with jax, but runs fine with tensorflow and torch, so it's very close to becoming backend-agnostic.

Reference:

• SigLIP: https://arxiv.org/abs/2303.15343
• CLIP: https://arxiv.org/abs/2103.00020

Standalone code to reproduce the issue or tutorial link

Relevant log output

No response

[awsaf49]

cc: @fchollet

[sachinprasadhs]

@awsaf49 , Feel free to create a PR with the example, team would be happy to review it.

[github-actions]

This issue is stale because it has been open for 14 days with no activity. It will be closed if no further activity occurs. Thank you.

[github-actions]

This issue was closed because it has been inactive for 28 days. Please reopen if you'd like to work on this further.