Tessellation neighbors #413
Conversation
samwaseda
requested review from
liamhuber and
pmrv
and removed request for
liamhuber
| @@ -174,6 +174,17 @@ def test_strain(self): | |||
| self.assertLess(np.absolute(eps_yz-strain[:, 1, 2]).max(), 0.01) | |||
| self.assertLess(np.absolute(eps_xz-strain[:, 0, 2]).max(), 0.01) | |||
|
|
|||
| def test_tessellations(self): | |||
There was a problem hiding this comment.
Please add msg kwarg to the tests to clarify the intent for skimmers.
| voro = Voronoi(positions) | ||
| x = positions[voro.ridge_points] | ||
| return indices[voro.ridge_points[ | ||
| np.isclose(self._structure.get_wrapped_coordinates(x), x).all(axis=-1).any(axis=-1) |
There was a problem hiding this comment.
This is beautful but kind of opaque...could we extract it to a little named method?
Pull Request Test Coverage Report for Build 1563680009
💛 - Coveralls |
|
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions. |
There was a problem hiding this comment.
A couple nits, but looks good.
Today is my first day of parental leave and then I'll be moving in January. I hope to be a but more involved again come March/April, but until then I'll probably only poke my nose into architecture stuff on base. Just a heads-up that future reviews from me might be tardy on the scale of months 😉
I added two algorithms to find neighbouring atoms:
To be honest, I don't really know whether the Delaunay tessellation finds its utility, but since the functionality was nearly the same I implemented it anyway.