JP-3551: Exclude overlapping background candidates #8744
Conversation
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## master #8744 +/- ##
==========================================
+ Coverage 60.75% 61.76% +1.01%
==========================================
Files 375 377 +2
Lines 38367 38743 +376
==========================================
+ Hits 23308 23929 +621
+ Misses 15059 14814 -245 ☔ View full report in Codecov by Sentry. |
There was a problem hiding this comment.
I left a couple comments about how sub-pixel dithered data might be handled. I don't think there is any good data examples to test it on for the S1600A1 5-point nodded data, but I see that some examples were added on the issue/ticket that give a general sense of the dither pattern behavior for sub-pixel dithers. Let me know if any other examples would help.
981312f to
e1d1244
Compare
|
@hayescr - I think I have the logic right for subpixel dithers now. I consolidated the S1600A1 overlap check with the existing check on primary dither position, so it's all in the same place now, and added some unit tests specifically for this condition. Let me know if you think this is sufficient now. |
e86737f to
49dfd22
Compare
|
Regression tests started here: |
|
@melanieclarke the changes look good to me! |
|
Report generation failed for the regression tests, but looking at the detailed output, I think this is the only real failure: This is the pool file I added to test these changes: the new association files for 5 point dithers with S1600A1 are subsets of the previous ones. |
49dfd22 to
69dd87e
Compare
25219e8 to
90be0a0
Compare
Uh oh!
There was an error while loading. Please reload this page.
90be0a0 to
003d423
Compare
|
Started regression tests on Jenkins here: |
Resolves JP-3551
Closes #8301
Update association files for S1600A1 with 5-point dithers to exclude the background candidate in the nearest dithers to the science.
Matching is done by primary dither position (computed from PATT_NUM and SUBPXPTS) since that is the information currently available in the pool. Background candidates with primary position <= 1 away from the science position are excluded from the pool. For example, if science is position 1, background at position 2 is excluded. If science is at position 3, backgrounds at position 2 and 4 are excluded. This depends on the 5-point dither retaining a fixed sequence of dither position order.
In this PR, I also moved the make_nod_asns function from the base class to the AsnMixin_Lv2Nod mix-in class, since it is only called in that case.
Checklist for PR authors (skip items if you don't have permissions or they are not applicable)
CHANGES.rstwithin the relevant release sectionupdated relevant documentationHow to run regression tests on a PR