Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Smoothing kernel is not properly normalized in bkg generation #60

Open
samanthalwong opened this issue Oct 10, 2024 · 1 comment
Open
Assignees
Labels
bug Something isn't working invalid This doesn't seem right validation Issues holding up validation

Comments

@samanthalwong
Copy link
Collaborator

I can't replicate the sky map artifacts you're seeing (see below; I actually get the opposite where no smoothing has the edge effects but smoothing doesn't), but I believe that the smoothing impact on background rate is due to improper convolution with the Gaussian kernel.

I'll look into better ways to implement some sort of interpolation, whether it be convolving with a Gaussian with proper normalization or spline fitting, which is more similar to ED but also to how HESS smooths their 3D bkgs.

I'll open a different issue for this to keep better track of the actual issue and not my original issue on the wiki, which I haven't run into since we've resolved other issues in gammapy/gammapy-tools. We can reopen if this comes up again.

image
image

Originally posted by @samanthalwong in #58 (comment)

@samanthalwong samanthalwong self-assigned this Oct 10, 2024
@samanthalwong samanthalwong added bug Something isn't working invalid This doesn't seem right validation Issues holding up validation labels Oct 10, 2024
@samanthalwong
Copy link
Collaborator Author

samanthalwong commented Nov 1, 2024

Implemented in commit 3cd37f5

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working invalid This doesn't seem right validation Issues holding up validation
Projects
None yet
Development

No branches or pull requests

1 participant