Lucas Morin

@zimonitrome I was toying with your implementation to build a custom loss. But my model doesn't seem to learn (while it does with 'trivial' losses). Are you sure the gradient...

Thanks for the explanation. This is frightening as I might have used this before...

@cjnolet is cuml.experimental usually available on kaggle ?

@jrash I am curious about conformal prediction for binary classification. Do you happen to have a demo code for Mondrian Conformal for binary classification ?

Yes, this is what I was looking for. As the author says : "That does help as an interim solution, but I would still enjoy seeing the inclusion of the...

Thanks, I'll look into this. For path maybe the way to go is to allow for customizing task id, now I get random strings (cell id maybe ?). Now I...

I am trying to understand what would be needed. I am asking here and there and will propose something when I figure out what would word need.

I feel like: - Most ML algo would treat int as numerical. - Such a change would help the majority of people, while the edge case of encoding categorical as...

I am currently toying with multitargets approach ... I have a hard time defining a custom metric (haven't tried custom loss). Preds seems to be of size (len(y) x len(targets))...

Nice Idea to have a list like that :) Here are the two things that I would change: - Predict API. This has been mentioned. I would do something similar:...