Replace _subsample with biom's optimized subsample

[wasade]

This PR uses the subsample method from biom. Subsample was recently extensively optimized by @sfiligoi. And, biom as a package was made Windows compatible by @mataton. However, the new version is not yet released. This PR sets the stage for using these improvements. I'll shift to a biom release at the next opportunity.

A few important items:

• some existing tests made I think slightly conservative assumptions
• this PR is a stepping stone to providing a 2D matrix or Table to subsample, but in my opinion we should defer those changes as a separate PR

Please complete the following checklist:

• I have read the contribution guidelines.

• I have documented all public-facing changes in the changelog.

• This pull request includes code, documentation, or other content derived from external source(s). If this is the case, ensure the external source's license is compatible with scikit-bio's license. Include the license in the licenses directory and add a comment in the code giving proper attribution. Ensure any other requirements set forth by the license and/or author are satisfied.

  • It is your responsibility to disclose code, documentation, or other content derived from external source(s). If you have questions about whether something can be included in the project or how to give proper attribution, include those questions in your pull request and a reviewer will assist you.

• This pull request does not include code, documentation, or other content derived from external source(s).

Note: This document may also be helpful to see some of the things code reviewers will be verifying when reviewing your pull request.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 98.47%. Comparing base (c1d6c18) to head (0b12fd0).

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #2016      +/-   ##
==========================================
- Coverage   98.47%   98.47%   -0.01%     
==========================================
  Files         184      184              
  Lines       31502    31498       -4     
  Branches     7673     7667       -6     
==========================================
- Hits        31021    31017       -4     
  Misses        464      464              
  Partials       17       17              

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

[qiyunzhu]

@wasade Since biom-format 2.1.16 is online, would it be the time to finish and merge this PR? I was able to setup a numpy 2.0 dev environment for scikit-bio, and found that the current bottleneck is the old _subsample code. Thank you!

  File "skbio/stats/__subsample.pyx", line 1, in init skbio.stats.__subsample
    # ----------------------------------------------------------------------------
ImportError: numpy.core.multiarray failed to import (auto-generated because you didn't call 'numpy.import_array()' after cimporting numpy; use '<void>numpy._import_array' to disable if you are certain you don't need it).

[wasade]

@qiyunzhu, if green then I think we're good for final review

[wasade]

I don't observe the same failure on x86_64, updating CI. There may be a wider variance on the failing test than what was originally tested for.

[wasade]

Okay, we're green

[qiyunzhu]

Looks great to me. Thanks! Just curious: What's the rationale of specifying "macos-13", "macos-14" instead of "macos-latest"?

[wasade]

Great!One is x86_64, other is M1On May 30, 2024, at 18:45, Qiyun Zhu ***@***.***> wrote: @qiyunzhu commented on this pull request. Looks great to me. Thanks! Just curious: What's the rationale of specifying "macos-13", "macos-14" instead of "macos-latest"? —Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you were mentioned.Message ID: ***@***.***>

[qiyunzhu]

@wasade Interesting! I did a quick search and found this comment. If there is an official source, it will be useful to comment in the yml file. But for now, let's just merge.

[wasade]

@qiyunzhu, here