performance improvements

[shaivimalik]

Fixes #2063

Changes made in this Pull Request:

• improves performance as discussed in Numpy Performance Improvements #2063

PR Checklist

• Tests?
• Docs?
• CHANGELOG updated?
• Issue raised/referenced?

---

📚 Documentation preview 📚: https://readthedocs-preview--4089.org.readthedocs.build/en/4089/

[codecov]

Codecov Report

Patch coverage: 100.00% and project coverage change: -0.01 ⚠️

Comparison is base (628e0f7) 93.59% compared to head (174a3cc) 93.59%.

Additional details and impacted files

@@             Coverage Diff             @@
##           develop    #4089      +/-   ##
===========================================
- Coverage    93.59%   93.59%   -0.01%     
===========================================
  Files          192      192              
  Lines        25135    25133       -2     
  Branches      4056     4056              
===========================================
- Hits         23525    23523       -2     
  Misses        1092     1092              
  Partials       518      518              

Impacted Files	Coverage Δ
package/MDAnalysis/analysis/bat.py	94.57% <100.00%> (ø)
package/MDAnalysis/core/groups.py	97.51% <100.00%> (ø)
package/MDAnalysis/core/topologyattrs.py	96.12% <100.00%> (ø)
package/MDAnalysis/lib/transformations.py	78.49% <100.00%> (-0.07%)	⬇️

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 in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.

[shaivimalik]

@richardjgowers @orbeckst Hello! Can anybody review my PR?

[hmacdope]

Seems reasonable! Would you be able to see if this makes a change to the benchmarks?

See: https://github.com/MDAnalysis/mdanalysis/blob/develop/benchmarks/README.rst

[shaivimalik]

@hmacdope Yes. I will get back to you with the result.

[shaivimalik]

@hmacdope I ran the benchmarks 3 times and got the following results:

[shaivimalik]

Hi, @hmacdope I made changes to more files where using np.einsum in place of np.sum seemed logical on a different branch and got the following benchmark results:


If these results are up to the mark, I will make these changes to the current PR branch.

[hmacdope]

The decrease in performance in some cases is unexpected. Did you have stuff running in the background at the same time @g2707 ?

[shaivimalik]

@hmacdope No, I did not start any new applications during the benchmark.

[hmacdope]

@g2707 I can try on my computer tonight

[shaivimalik]

@hmacdope Should I add the new file changes before tonight?

[hmacdope]

Yep!

[github-actions]

Linter Bot Results:

Hi @g2707! Thanks for making this PR. We linted your code and found the following:

Some issues were found with the formatting of your code.

Code Location	Outcome
main package	⚠️ Possible failure
testsuite	✅ Passed

Please have a look at the darker-main-code and darker-test-code steps here for more details: https://github.com/MDAnalysis/mdanalysis/actions/runs/4561242103/jobs/8047069296

---

Please note: The black linter is purely informational, you can safely ignore these outcomes if there are no flake8 failures!

[shaivimalik]

Hi! @hmacdope Did you get a chance to run the benchmarks?

[hmacdope]

@g2707 yes I did thanks! Performance improvements look good! 2x runs shown below!

[hmacdope]

You will need a CHANGELOG entry (newest first), to add yourself to AUTHORS and as I think you are a new contributor we ask you to introduce yourself on the developer mailing list. ! Other than that looks good and all tests pass, just a few nitpicks.

[hmacdope]

Looks good to me! Introduce yourself on the developer mailing list and then I can merge. :)

https://groups.google.com/g/mdnalysis-devel

[shaivimalik]

@hmacdope Done!