Yoni Gozlan

Results 45 comments of Yoni Gozlan

Uniform kwargs for processors

Also added Udop to the list and to #32544 .

Uniform kwargs for processors

Hi @davidgxue and @MnCSSJ4x, I was wondering if you've made any recent progress on uniformizing LLaVa and Paligemma. They're some of the last image-text-to-text models left to uniformize, and it...

Uniform kwargs for processors

No problem @davidgxue! I will get started on LLaVa then

Uniformize kwargs for LLaVa processor and update docs

Now that https://github.com/huggingface/transformers/pull/33385 has been merged, this should be ready for review!

Image-Text-to-Text Support in Transformers Pipeline

Good timing ;) https://github.com/huggingface/transformers/pull/34170

[transformers snippet] Support pipeline VLMs

> The example is super nice, but not all image-text-to-text models support multiple images reliably. I'd go for a simpler single-image example for now. Agreed with this! another problem is...

Fix continue_final_message for image-text-to-text chat templates

Added one test for llava processor :). I could add one for every vlms processor that use chat template, but as they all use the same underlying `apply_chat_template`, I thought...

Add image text to text pipeline

Thanks for the feedback @knkski! Although it's not really an objective of this pipeline, I think we can try to add support and raise a warning at least, wdyt @Rocketknight1...

Add image text to text pipeline

@Rocketknight1 @knkski , text-only inference should be supported now :)

Add image text to text pipeline

There is still some issues with pipeline tests: - It seems that pipeline model tests are based on "tiny models" available on `hf-internal-testing`, but those tiny models don't seem to...