Allow MapieRegressor to use any split strategy

[thibaultcordier]

Description

Split-CP is only available with "split" keyword or ShuffleSplit class of scikit-learn.

This PR proposes to use more compatible split BaseCrossValidator/BaseShuffleSplit.

In practice, we need to use the predictions aggregation for cross-CPs and not for split-CPs. All you need to do is find out the number of splits using the get_n_splits method to know whether it's a split or a cross.

Fixes #385

Type of change

• Breaking change (fix or feature that would cause existing functionality to not work as expected)

How Has This Been Tested?

Please describe the tests that you ran to verify your changes. Provide instructions so we can reproduce. Please also list any relevant details for your test configuration

• Add test for the function check_no_agg_cv
• Add new split method with MapieRegressor
• Check that the other tests still work

Checklist

• I have read the contributing guidelines
• I have updated the HISTORY.rst and AUTHORS.rst files
• Linting passes successfully : make lint
• Typing passes successfully : make type-check
• Unit tests pass successfully : make tests
• Coverage is 100% : make coverage
• Documentation builds successfully : make doc

[thibaultcordier]

This PR also tidies things up a bit: the tests specific to CQR have been moved to the dedicated file (from test_utils.py to test_quantile_regression.py).

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (614293e) 100.00% compared to head (1fa8b7c) 100.00%.
Report is 34 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##            master      #386    +/-   ##
==========================================
  Coverage   100.00%   100.00%            
==========================================
  Files           39        39            
  Lines         4616      4633    +17     
  Branches       487       762   +275     
==========================================
+ Hits          4616      4633    +17     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[LacombeLouis]

Good PR! Thank you @thibaultcordier 😀
I am a little worried about adding another fit_attributes for custom algorithms. What do you think?

[vincentblot28]

Everything is fine, let's merge !