DistanceMetric hotfix

[KulikDM]

All Submissions Basics:

• Have you followed the guidelines in our Contributing document?
• Have you checked to ensure there aren't other open Pull Requests for the same update/change?
• Have you checked all Issues to tie the PR to a specific one?

All Submissions Cores:

• Have you added an explanation of what your changes do and why you'd like us to include them?
• Have you written new tests for your core changes, as applicable?
• Have you successfully ran tests with your changes locally?
• Does your submission pass tests, including CircleCI, Travis CI, and AppVeyor?
• Does your submission have appropriate code coverage? The cutoff threshold is 95% by Coversall.

This is a hot fix for sklearn's DistanceMetric, an alternative to this PR and maybe perhaps better is to up the minimum sklearn version to 1.0.0 and change the DistanceMetric to the new API.

[coveralls]

Pull Request Test Coverage Report for Build 5467755404

• 4 of 4 (100.0%) changed or added relevant lines in 1 file are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage increased (+0.003%) to 93.604%

---

Totals
Change from base Build 5368111030:	0.003%
Covered Lines:	8707
Relevant Lines:	9302

---

💛 - Coveralls

[yzhao062]

Hi @KulikDM :) the newer version of scikit-learn -> 1.3.0 appears to cause many problems.
I have fixed others and the only failed tests now is test_threshold.py (see test results here). Would you mind taking a look to see what is might be the problem?

Cheers,
Yue

[KulikDM]

Hi @yzhao062, you are very right, scikit-learn=1.3.0 has caused a real change up to a lot of packages :(
The issue in test_threshold.py is being caused be a single test due to an external library scikit-lego. I've helped address the issue in PR 562. However, sckit-lego is currently having other issues with the new scikit-learn version and can't create a new release Issue 564. If you temporarily replace scikit-lego with https://github.com/koaning/scikit-lego/archive/main.zip in the requirement-ci.txt file test_threshold.py should pass. Otherwise, META could also be removed from test_threshold.py but that is a bit more of an extreme step. Let me know what you think should be done and I'll try to keep you posted on any updates from my side.

Best,
Daniel

[yzhao062]

Thanks for the thought! In that sense let us wait for the refresh of scikit-lego :) I mean waiting for the newer version of it on PyPI to automatically address the issue.

[KulikDM]

Hi @yzhao062 just finished helping to get scikit-lego working and I see that test_threshold.py is now passing all the tests :)

[yzhao062]

that is great!