Thomas Li

This should be ready for another pass now. Primary mechanism for skips would be the ``_array_api_skips`` dictionary in ``sklearn/utils/_array_api.py`` I also added a ``skip_dask`` fixture/decorator for individual tests that test...

> Here is another pass of feedback, possibly final on my end. > > Would it be possible to open issues upstream, either in the dask, array-api-compat repos or the...

> Can we use array-api-compat to add `slogdet` to dask arrays? Then we'd avoid the problem that an estimator semi works. This is not possible yet. Dask doesn't expose a...

Thanks for updating. I'm back to working on this again now that my summer is over. Last time I remember, I think had ridge regression working locally (after wrangling some...

Thanks, I'll rebase this this week and let you know how it goes.

Made a bit of progress. Some notes I made along the way are: TODOs - [ ] Upstream fill_diagonal to array-api-extra (5 uses in sklearn) - [ ] Upstream isin/in1d...

I haven't touched this PR in a while (besides rebasing). I've opened an issue at array-api-extra, https://github.com/data-apis/array-api-extra/issues/268, for the missing functions and I think they're in favor of adding them....

Hm, this works on main for me. Can you try a newer version of pandas (e.g. 2.2.1)?

My bad, forgot to run the second command. Nice catch. Looks like ``parsing.concat_date_cols`` (which concatenates date_cols) is getting passed an ArrowExtensionArray. That behavior is deprecated (to be removed in 3.0),...

@WillAyd Mind if I push some fixes for the non string dtype builds to your branch? I finally got to taking a look at this PR and I have a...