Tracking integration for Image Question Answering

[osanseviero]

Tracking integration of task -Image Question Answering

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 huggingface/hub-docs#21

Integration guide: https://hf.co/docs/hub/adding-a-task

cc @mishig25 @NielsRogge

[NielsRogge]

We do have a model already in the library (ViLT) that can do VQA pretty well! https://huggingface.co/spaces/nielsr/vilt-vqa

Should be relatively easy to add a pipeline for it (the model isn't that complex compared to other ones, it just outputs logits) cc @Narsil

[osanseviero]

The pipeline is now here, so we can add this to the inference API + a widget for it (cc @mishig25). WDYT?

[osanseviero]

This is now fixed!