Niels Cautaerts

Looks interesting but very computationally intensive I would imagine. Not sure about the RAM issue you describe though. For 4., would it make sense to identify the coordinates that each...

Do you mean with 2) to simulate patterns sampling all of SO(3) or will you apply the same logic as in the current template matching approach, i.e. optimizing the in-plane...

Also note that a cKDtree, point-distance based error metric approach will not be "fair" in polar coordinate space because the relative scale between r and theta is arbitrary. For a...

> the current template matching approach, just comparing vector positions. Not sure which one that would be ... > I think that there is some potential for simulating using real...

> multiplies the r position of the diffraction spots by the strip in theta at that r to get the cross correlation. Sort of but not entirely. In the CPU...

Here is an idea on using KDTrees which also kind of works in a similar way: ```python import numpy as np from scipy.spatial import KDTree from numba import njit from...

> Basically this changes the problem from a distance based comparison to more of a search problem using a presorted index. I'm not entirely sure how the KDTree works under...

ran into the same issue and indeed using master as suggested by @K41eb fixed the issue. It's a bit strange then that the documentation specifically suggests to check out a...

Thanks for pinging me for the discussion @CSSFrancis For me very simply the scope of diffsims is to easily and flexibly calculate kinematical diffraction intensities. An expanded scope may be...

> For me this is probably what I am most interested in adding. I think that its nice to have the ability to build simple structures simulate their kinematic diffraction...