Hello! Is there a way to enable online learning on a model deployed on a Huggingface space (or the Model Hub). Like the flag option in gradio apps to flag some examples, is there an option for the user to correct the output predicted? Thanks.
Thats a super cool idea. The machinery to build this exists, but you would need to actually hook the building blocks together.
We’ll work on a tutorial to streamline this at some point
Okay, I’d love to contribute with this project