Make it easier to change the surface normal in Bragg2DPtycho.from_dataset#63
Make it easier to change the surface normal in Bragg2DPtycho.from_dataset#63
Conversation
allevitan
left a comment
There was a problem hiding this comment.
This is really nice, seems like this is a super helpful feature!
Right now, if I'm reading the code right, it's written so that if there is an 'orientation' defined in dataset.sample_info, then that will override the provided surface_normal argument, and the same if a 'scattering_mode' argument is set.
I think it would make the most sense if surface_normal > scattering_mode > dataset.sample_info['orientation'], i.e. an explicitly defined surface_normal overrides everything, a scattering_mode overrides the info in dataset.sample_info, and dataset.sample_info serves as a fallback. Does that make sense?
|
Makes sense to me, and I think that sounds like a good idea. The latest push has this hierarchy implemented. Currently, it's set so that |
… an array with np.asarray, change the check for length-3 to cover all cases, and add a check to explicitly fail if the scattering_mode argument is improperly set, instead of silently falling back to default
|
I just added a few suggested tweaks to the branch, let me know if they look good:
If it looks good, let me know and I can clear my comments and merge it. |
|
Thanks for making these changes; they all look good to me. Would you do the honors of merging? |
Makes the surface normal vector a parameter in Bragg2DPtycho.from_dataset to make it easier for folks to define what the surface normal is.