[SyneRBI/SIRF] GitHub Actions fail on MR tests (Issue #1267)
Kris Thielemans
https://github.com/SyneRBI/SIRF/actions/runs/9556091923/job/26340671249#step:9:1053
E ValueError: cannot fill AcquisitionData of size (np.int32(512), np.int32(8), np.int32(256)) with data of size (1, 1, 8, 131072)
@evgueni-ovtchinnikov this is rather strange, as the recent commit didn't change anything in Gadgetron.
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.
Kris Thielemans
@evgueni-ovtchinnikov this prevents images being build, so needs urgent attention. sorry
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.
Evgueni Ovtchinnikov
the failure happens when testing test_imagedata_constructor.py, but SIRF master does not have this script!
apparently, GitHub Actions do some other branch - how can we find which one?
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.
Kris Thielemans
we have https://github.com/SyneRBI/SIRF/blob/master/.github/workflows/build-test.yml
and https://github.com/SyneRBI/SIRF-SuperBuild/blob/master/version_config.cmake of course.
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.
Evgueni Ovtchinnikov
the failure happens when testing test_imagedata_constructor.py, but SIRF master does not have this script!
found it (scrollbar does not show on VM unless you move the cursor to the window edge), sorry for the false alarm!
on my VM runs ok, so have to debug via GitHub Actions
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.
Evgueni Ovtchinnikov
looks fixed
likely reason was the change of numpy version, details tomorrow
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.
Kris Thielemans
ok. Indeed, we're getting numpy 2.0 now. It'd make sense that it's an external change causing this problem now.
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.
Evgueni Ovtchinnikov
fixed by 8e8d95b
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.
Evgueni Ovtchinnikov
Closed #1267 as completed.
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.