choderalab / pymbar

Python implementation of the multistate Bennett acceptance ratio (MBAR)
http://pymbar.readthedocs.io
MIT License
240 stars 93 forks source link

Fixed example bug and updated output. #513

Closed mrshirts closed 1 year ago

mrshirts commented 1 year ago

Fixed a bug in the harmonic_oscillators.py example that wasn't actually retrieving the data from the results object for reverse exponential averaging.

Added logging to the two harmonic oscillator files.

Update the output of the example scripts since they were quite old.

codecov[bot] commented 1 year ago

Codecov Report

Merging #513 (336a0ee) into master (7b4f89f) will decrease coverage by 0.16%. The diff coverage is n/a.

:exclamation: Current head 336a0ee differs from pull request most recent head 41f8d8d. Consider uploading reports for the commit 41f8d8d to get more accurate results

mrshirts commented 1 year ago

Anyone have any thoughts why the doc build failed and the windows test timed out? This change would not really affect that much - it's adding logging instructions to 2 files, making one trivial change in processing outputs (doesn't change the algorithms run), and adds some text files.

mikemhenry commented 1 year ago

@mrshirts I can look at RTD, I think the windows test is more just being flaky, I will re-run that one

Lnaden commented 1 year ago

Windows is a slow to start test, and the trust-krylov solver in SciPy sometimes can take many thousands of iterations to converge instead of the ~30 based on the random data we generate. Together that can cause the windows' tests to time out. Re-running as @mikemhenry did works.