Tracking integration for Video Classification

[osanseviero]

Tracking integration of task - Video Classification

Example discussion - https://huggingface.co/keras-io/video-transformers/discussions/1#62e9694f1ee086f609b4a500

Note that you're not expected to do all of the following steps. This PR helps track all the steps required to get a new task fully supported in the Hub 🔥

• [x] Integration with Inference API. Select at least one of the following:
  • [x] Added a transformers pipeline
  • [ ] Added to Community Inference API for 3rd party library
  • [ ] Added to Community Inference API for generic
• [x] Added basic UI elements (icon, order specification, etc)
• [ ] Added a widget

Integration guide: https://hf.co/docs/hub/models-tasks

[NielsRogge]

Would be great to add basic UI elements for it, as we now have video models as seen here: https://huggingface.co/models?other=video-classification

[osanseviero]

cc @NimaBoscarino or @nateraw would any of you be interested in working on this? This corresponds to the UI elements section at https://huggingface.co/docs/hub/models-tasks#ui-elements

[osanseviero]

Friendly ping

[NimaBoscarino]

I wish, but I don't have the bandwidth for it right now 😞

[osanseviero]

Now that we have the UI elements and the pipeline, we could add this to the inference API for transformers + add a widget. WDYT @Narsil @OlivierDehaene @mishig25 ?

[Narsil]

https://github.com/huggingface/api-inference/pull/1252

[osanseviero]

Very cool! The only missing step would be the widget

[osanseviero]

@nateraw would you be up for updating https://huggingface.co/tasks/video-classification to use the new pipeline?