multi-label classification / paperswithcode dataset

[ghost]

Hi guys,

Hope you are all well !

I was wondering if X-Transformer can handle multi-label classification with 1560 labels.

More precisely, I would like to apply it to paperswithcode dataset where labels are called tasks.

Refs:

• Repository
• Dataset

Thanks for any insights or inputs on that.

Cheers, X

[simonlevine]

@OctoberChang please correct me if I'm wrong but I believe so, but you'll need to write preprocessing logic. I don't think your dataset would benefit from PIFA either if the labels are single words. This is a magical repo for the exact XMLC task but also may be overkill for only 1500 labels, especially if each instance is only assigned to one label. I would try writing a pipeline with a transformer with a linear head (rather than the matcher/ranker) first and see if that's feasible.