[BUG] kmeanfold does xpass?

@kayhoogland @MBrouns did you folks also see this in your test results?

tests/test_meta/test_thresholder.py .................                                                                                                   [ 49%]
tests/test_metrics/test_correlation_score.py ..                                                                                                         [ 49%]
tests/test_metrics/test_neg_p_percent.py ...                                                                                                            [ 50%]
tests/test_model_selection/test_klusterfold.py ........X........X........X........X........X........X........X........X........X........X........X..... [ 57%]
...X........X........X........X........X........X........X........X........X........X........X........X........X........X........X........X             [ 67%]
tests/test_model_selection/test_timegapsplit.py ....                                                                                                    [ 67%]
tests/test_pandas_utils/test_pandas_utils.py ......                                                                                                     [ 68%]

[koaning]

I prefer with pytest.raises(ValueError): mainly because it is consistent with the rest of the API.

Also, I know that I am not the most senior pytest user. But I never heard of xpass before today. I can imagine introducing it can be very confusing for other users.

[MBrouns]

I didn’t look at the detailed test report, only checked whether the CI was green. I’m a bit surprised that an xpass actually results in the test suite as a whole passing. It might be preferable to use with pytest.raises(ValueError): instead in that case.