ENH Add SciPy tutorial dataset to fairlearn.datasets

[rensoostenbach]

This PR adds the SciPy tutorial dataset to fairlearn.datasets, as discussed in #1066

Completing this item requires:

• implement a function fairlearn.datasets.fetch_... (we need to come up with a name)
• adjust relevant unit tests in test.unit.datasets
• descriptive API reference (directly in the docstring)
• a user guide in docs.user_guide.datasets

Description

I have implemented the fetch function as fairlearn.datasets.fetch_diabetes_hospital, but ofcourse it would be easy to rename it to something else if prefered. I wasn't sure how detailed the User Guide should be, and also what parts should be included in the user guide. If there is anything that would be valuable to include that isn't in there yet, I would be happy to write that as well.

Tests

• existing tests adjusted

Documentation

• user guide added or updated
• API docs added or updated

Screenshots

[rensoostenbach]

I still need to write the User Guide for this PR, I will start next week with that.
I am not sure why flake8 is giving the following error at line 13:

D412 No blank lines allowed between a section header and its content

That line and the surrounding lines are done in the same way as in the other _fetch_dataset.py files.

[hildeweerts]

Just a few small comments!

[hildeweerts]

Tagging @LeJit and @MiroDudik who may have thoughts on what (else) should be in the user guide.

[codecov]

Codecov Report

❗ No coverage uploaded for pull request base (main@d5950c0). Click here to learn what that means.
The diff coverage is n/a.

❗ Current head 42d98f0 differs from pull request most recent head 971a349. Consider uploading reports for the commit 971a349 to get more accurate results

@@          Coverage Diff           @@
##             main   #1086   +/-   ##
======================================
  Coverage        ?   0.00%           
======================================
  Files           ?      51           
  Lines           ?    2180           
  Branches        ?       0           
======================================
  Hits            ?       0           
  Misses          ?    2180           
  Partials        ?       0           

Flag	Coverage Δ
unittests	0.00% <0.00%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update d5950c0...971a349. Read the comment docs.

[LeJit]

It may be worthing including a link to the full preprocessing.py script, in case any users are interested in all the changes we made to the original dataset.

[rensoostenbach]

It may be worthing including a link to the full preprocessing.py script, in case any users are interested in all the changes we made to the original dataset.

It is actually there already in line 35 of diabetes_hospital_data.rst, or would you (also) like to have it included somewhere else? @LeJit

[LeJit]

@rensoostenbach I was thinking more about referencing it in the docstring of fetch_diabetes_hospital in the paragraph where you talk about the SciPy tutorial.

[adrinjalali]

Merged with the latest main to have your docs rendered in our CI. You should do a git pull to have this on your local copy before you apply more changes.

[adrinjalali]

🤦🏼 OpenML's SSL certificate is expired, they're working on it.

[hildeweerts]

I think this PR is ready to get merged!