You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The code for how a WCS solution is chosen isn't well documented in the readthedocs documentation. It would be good to add more information in a new section, or perhaps here.
For example, how the consistency checks work. How a radial_offset_check may result in a fit_quality of 2, and a better WCS solution than a fit solution with an RMS >10 mas.
Another example is how a solution with the minumum number of matches (-2) may be a better solution than a alignment solution with an RMS >10 mas.
Documentation of where the nmatches are set, and what the different variables refer to would also be good to include.
Also make it clear how the code picks an SVM reference image.
The text was updated successfully, but these errors were encountered:
The selection of the SVM reference image is done in tweakwcs. It looks like the code uses one of the two images that it determines to have the largest overlap in a list of images (unless order is specified by a user). This code is in tweakwcs.imalign.align_wcs().
Issue HLA-1425 was created on JIRA by Steve Goldman:
The code for how a WCS solution is chosen isn't well documented in the readthedocs documentation. It would be good to add more information in a new section, or perhaps here.
For example, how the consistency checks work. How a radial_offset_check may result in a fit_quality of 2, and a better WCS solution than a fit solution with an RMS >10 mas.
Another example is how a solution with the minumum number of matches (-2) may be a better solution than a alignment solution with an RMS >10 mas.
Documentation of where the nmatches are set, and what the different variables refer to would also be good to include.
Also make it clear how the code picks an SVM reference image.
The text was updated successfully, but these errors were encountered: