random notes

[amueller]

I'm not sure why "diamonds" is not included.
Also, you have "houses" and "boston" which are two housing datasets (though quite different in nature?). But neither is similar at all to the ames housing dataset or the king's county housing dataset (which have many high cardinality categorical variables).
So I'd probably include either of those?

Also, 2dplanes is a synthetic dataset, and so is mv.

[amueller]

I'm not sure if that's intended but it looks like in "bodyfat" the "density" basically seems to leak the target?

[amueller]

it says AutoPrice is there twice but I don't see it in https://new.openml.org/search?type=task&sort=runs&tags.tag=openml-reg19

[PhilippPro]

Dear Andreas, thanks for the feedback. I will have a closer look at these datasets.