Using pre-trained CLIP (Contrastive Language-Image Pre-Training) in Keras

[Marcelino Borges]

Hi. I would like to use this pre-trained model, available for pytorch in Keras.

https://github.com/openai/CLIP

Is there any CLIP pre-trained model available for Keras?

If there isn't, I was thinking about using the pytorch pre-trained model for extracting the features and, after that, using those features for feeding my keras model. Is it possible to convert pytorch tensors to keras tensors for doind that?

[Sayak Paul]

It's available via Hugging Face Transformers:

https://huggingface.co/docs/transformers/model_doc/clip (TFCLIPModel). 

Thanks,
Sayak | sayak.dev

--
You received this message because you are subscribed to the Google Groups "Keras-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to keras-users...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/keras-users/CANpcTFoOCohBmGQ3AJi3Vv7cB1fSW4ZnpHrVNm9dd9m_uVKTxw%40mail.gmail.com.