[ENH] Getting pvalue computation ready

[adam2392]

Changes proposed in this pull request:

• finalizes API for estimating pvalue using Coleman, or permutation forest approach
• adds parallelization to the permutation forest sampling in the Coleman approach

Suggested # of trees is >=6_000. Suggested number of permutations in the Coleman approach is 10_000 or greater.

Before submitting

• I've read and followed all steps in the Making a pull request
  section of the CONTRIBUTING docs.
• I've updated or added any relevant docstrings following the syntax described in the
  Writing docstrings section of the CONTRIBUTING docs.
• If this PR fixes a bug, I've added a test that will fail without my fix.
• If this PR adds a new feature, I've added tests that sufficiently cover my new functionality.

After submitting

• All GitHub Actions jobs for my pull request have passed.

[sampan501]

This is basically just Coleman + new bootstrapping approach right?

[adam2392]

Yeah with some flexibility in the API

[sampan501]

Sweet, I'll merge when CIs are green

[codecov]

Codecov Report

Attention: 3 lines in your changes are missing coverage. Please review.

Comparison is base (d46e40b) 88.48% compared to head (5728550) 89.68%.

Files	Patch %	Lines
sktree/stats/forestht.py	92.30%	3 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #222      +/-   ##
==========================================
+ Coverage   88.48%   89.68%   +1.19%     
==========================================
  Files          54       54              
  Lines        4891     4963      +72     
==========================================
+ Hits         4328     4451     +123     
+ Misses        563      512      -51     

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

[sampan501]

@adam2392 I think the only thing missing is coverage. I don't believe there are any unit tests for this code

[adam2392]

@adam2392 I think the only thing missing is coverage. I don't believe there are any unit tests for this code

Will add

[adam2392]

Lmk wdyt @sampan501

Thru testing, I noticed that the permutation forest is different in how it computes the pvalue. It needs to check opposite directionality, since the null is slightly different I think. (so i.e. a bug), but lmk if you spot an issue

[sampan501]

Looks good except for some failed unit tests