Fixing Regression Tests for RM3 #1473
Conversation
kmruehl
left a comment
There was a problem hiding this comment.
@jleonqu Thank you for catching and resolving this error! My only recommendation is that the _org.mat files are very large. We only need output.bodies, so to save space I recommend resaving these files so that they only include the body class output. I'm also going to make some minor formatting revisions to these scripts.
699fac9 to
d67ec79
Compare
|
While reviewing this PR we noticed that the scripts were saving empty figures and that the comparison plots were not working properly. I pushed some revisions that resolve the issues with plotting the figures, but they may cause issues with the runner. I'll wait to see if the tests pass... |
|
The default appears to be that the figures are generating, which means that these tests should fail. This issue needs to be resolved. |
|
The test are passing, and this works for me. @jleonqu can you confirm? If it works for you, we can merge the PR |
|
All tests passed locally and the figures were generated correctly. The PR is ready to merge. |
* Match PR 1478 on main (#1479) * match PR 1478 changes on main --------- Co-authored-by: jtgrasb <87095491+jtgrasb@users.noreply.github.com> Co-authored-by: akeeste <akeeste@sandia.gov> * Fixing Regression Tests for RM3 (#1473) * Fixing Regression Tests for RM3 * Updating the correct _org.mat files * resolving figure issues --------- Co-authored-by: Ruehl <kmruehl@sandia.gov> --------- Co-authored-by: dforbush2 <dforbus@sandia.gov> Co-authored-by: jtgrasb <87095491+jtgrasb@users.noreply.github.com> Co-authored-by: akeeste <akeeste@sandia.gov> Co-authored-by: Jorge Leon <72461917+jleonqu@users.noreply.github.com>
This PR is solving Issue #1471
The baseline files for the RM3 regression tests have been updated and verified.
The tolerance value of 1e-10 was too tight, which was causing the tests to fail. The tolerance value has been updated to 1e-5.
The printPlotRegular.m and printPlotIrregular.m scripts have been updated to load the WEC-Sim run data that contains the most recent run and the results from the baseline file for comparison.