-
-
Notifications
You must be signed in to change notification settings - Fork 63
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
test_reproject_healpix_to_image_footprint failures on Debian with ARM (and other) CPUs #414
Comments
@lpsinger - I also see something similar when I run the tests locally on M1 Mac (though not on Mac in the CI). It looks like in these cases, the following:
returns an array that has values in the invalid region that are not actually set to -1 but instead to e.g. 8 in the following example: ![]() Do you have any ideas why this would be platform-dependent? |
@olebole, we think that astropy/astropy-healpix#208 may have fixed this. We'll roll another release of astropy-healpix shortly. |
@olebole, we think that this should be fixed now if you updated to astropy-healpix 1.0.1. |
Yes, I can confirm this. After uploading healpix 1.0.1, the reproject tests on all platforms pass. |
On the Debian architectures with ARM (32+64 bit), RISCv64, all test_reproject_healpix_to_image_footprint(…, 'nearest-neighbor') test fail:
The ones with order='bilinear' pass.
Full build log for amd64.
In Intel (amd64, i386), PowerPC, s390x, all tests pass.
Environment:
The text was updated successfully, but these errors were encountered: