Repository navigation
DM-55879: Correct the template variance for warping correlations - #477
Merged
Merged
Conversation
isullivan
force-pushed
the
tickets/DM-55879
branch
from
October 29, 2026 17:40
e1ca2e3 to
7582684
Compare
TallJimbo
approved these changes
Oct 29, 2026
| template variance if the plate scale of the coadd's constituent images | ||
| is different than the science image the template is being constructed | ||
| for. This should only be necessary if the coadd images were from a | ||
| different instrument than the science image. If the plate scale of the |
Member
There was a problem hiding this comment.
The "different instrument" sentence isn't technically correct or necessary; coadding to an instrument's native pixel scale has a lot of advantages, but there are valid reasons why one might not do that.
Contributor
Author
There was a problem hiding this comment.
I believe the "different instrument" statement is important. The correction factor it describes is due to the change in area of the pixels from the original coadded images to the science image, and does not depend on the pixelization of the coadd at all. This is separate from the warping correlation being added on this ticket, but I felt I should explain how the two correction factors interact.
Warping with a kernel normalized to unit sum moves noise power off the variance plane's diagonal and into covariance between neighboring pixels, which a variance plane cannot represent, leaving the variance too small for anything measured over more than one pixel.
The correction factors should now result in a better calculation of the variance, and doScaleVariance will actually make the result worse.
Warping all 16 patches took a couple seconds per test, and testRunCalibratesToNanojansky warped them twice more; it now relabels the patches' calibration instead. The getTemplate tests are about 6x faster with these changes.
isullivan
force-pushed
the
tickets/DM-55879
branch
from
October 29, 2026 22:35
7582684 to
57ad989
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This also turns doScaleVariance off by default for the template, since that sets the variance to an incorrect value.
The unit tests are slightly refactored, allowing them to run in 12s instead of 90s on my laptop.