The sign in the equation of mean curvature was wrong, indeed. Thanks, @lilyminium for catching this.
The overall term used to calculate mean curvature was quite messy, and it didn't match the one I added to the docstrings.
I added two main changes in this PR:
[x] Reorganized equation of mean curvature (now matches docstrings).
[x] Updated tests.
To clarify, with the change of the sign 3 tests failed: test_analysis_mean, test_mean_curvature_all, test_mean_curvature_small. As you can verify, the change in the expected value of H is the sign, exclusively. These three tests were updated accordingly.
Just in case, I double-checked the term of Gaussian curvature, and it looks OK.
Since this change has an impact on the first tutorial included in docs, I opened #74 to update the tutorial once this PR is approved.
I'm very sorry for this terrible mistake, @lilyminium @orbeckst. I hope these changes fix #70.
Please let me know if I missing anything else here.
The sign in the equation of mean curvature was wrong, indeed. Thanks, @lilyminium for catching this. The overall term used to calculate mean curvature was quite messy, and it didn't match the one I added to the docstrings.
I added two main changes in this PR:
To clarify, with the change of the sign 3 tests failed:
test_analysis_mean
,test_mean_curvature_all
,test_mean_curvature_small
. As you can verify, the change in the expected value of H is the sign, exclusively. These three tests were updated accordingly.Just in case, I double-checked the term of Gaussian curvature, and it looks OK.
Since this change has an impact on the first tutorial included in docs, I opened #74 to update the tutorial once this PR is approved. I'm very sorry for this terrible mistake, @lilyminium @orbeckst. I hope these changes fix #70.
Please let me know if I missing anything else here.