Use scikit-learn>=1.1.2

[raphaelvallat]

See #278 and scikit-learn/scikit-learn#23608

[musicinmybrain]

Works great as a downstream patch in Fedora Linux.

[codecov]

Codecov Report

Base: 98.58% // Head: 98.58% // No change to project coverage 👍

Coverage data is based on head (c198be8) compared to base (1f329ef).
Patch has no changes to coverable lines.

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #300   +/-   ##
=======================================
  Coverage   98.58%   98.58%           
=======================================
  Files          18       18           
  Lines        3320     3320           
  Branches      540      540           
=======================================
  Hits         3273     3273           
  Misses         25       25           
  Partials       22       22           

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[raphaelvallat]

Scikit-learn 1.1 and later requires Python 3.8 or newer (from sklearn documentation).

• To merge this PR, need to disable unit testing for Python 3.7

Update: to avoid pip conflicts, I think it is simpler to just remove any strict requirements for sklearn.