Tutorials examples for multiclass and multilabel labelling?

[JohnGiorgi]

Hi, thanks for making this public. I enjoyed the paper, and my team and I are excited to try this out in our work.

We have a multilabel problem and are struggling a little bit with how to apply FlyingSquid to this setup. C.2 of the appendix mentions that in the multiclass case, a one-versus-all scheme can be applied repeatedly. This makes sense in principle, but I was wondering if there was an example of this using LabelModel that could be provided? We went through the example notebooks, but each demonstrates a binary classification problem.

In the case of multilabel, I was thinking that you could follow a one-vs-all approach without a voting scheme as is typically done, such that each instance in the dataset can end up with multiple labels. Is this something that can be done with FlyingSquids LabelModel?

[JohnGiorgi]

This turned out to be straightforward.

For anyone who is interested, I just created one label_model per label and used each label_model to produce probabilistic_labels. This list of labels can be concatenated to yield a matrix of (number of examples, number of labels)!

[fredsala]

Hi John,

Thanks for your comments! Indeed, what you described is how we usually handle the multiclass case, but rather than concatenating, we combine all of the (binary) label model output parameters into one big parameter. For your case, with multilabels, concatenation should work.

Let us know if you have more questions!

[JieyuZ2]

Hi John,

Thanks for your comments! Indeed, what you described is how we usually handle the multiclass case, but rather than concatenating, we combine all of the (binary) label model output parameters into one big parameter. For your case, with multilabels, concatenation should work.

Let us know if you have more questions!

Hi Frederic,

Could you explain a little bit how to combine all the binary models into one big model?