Bar mask transmission #190
Conversation
Incorporate Zernike interpolation from the non-uniform 2D grid data rather than simply grabbing nearest neighbor.
Currently, the bar mask has a transmission of 0 at x=0. Presumably, the intent was to set 0 transmission along the bar’s x-axis (at y=0). The current implementation creates a vertical line that is perpendicular to the wedge rather than through the wedge horizontally. This update should fix that issue.
|
aw, shoot. yes that was supposed to be y=0. Good catch. |
|
Still haven't gotten the hang of pull requests. It would be nice if I could send a pull request for only a single commit rather than others that have already been sorted out... |
|
The Travis tests failed for this. Upon investigation, this failure was because the fix does introduce significant enough changes in the coronagraphic PSFs to just slightly exceed the 1% threshold in the unit test for how much the total post-coronagraphic flux is allowed to differ from the expected value from prior calculations. But it's clear this is a real and necessary fix, and presumably the new value is more accurate than the old one. So I am going to go ahead and merge this, and will update the unit tests accordingly as part of some related work I'm doing for #191. |
Currently, the bar mask has a transmission of 0 at x=0. Presumably, the intent was to set 0 transmission along the bar’s x-axis (at y=0). The current implementation creates a vertical line that is perpendicular to the wedge rather than through the wedge horizontally. This update should fix that issue.