You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I've encountered a regression with coregistration between 0.0.13 and 0.0.16, affecting (at least) Tilt and VerticalShift when used in combination with NuthKaab.
Attached the relevant dh plots: coregistration result is wrong in 0.0.16 with any combination of NuthKaab + one of the two methods mentioned above; when used individually, NuthKaab and VerticalShift behave as expected also in 0.0.16. Also attached the input files to reproduce.
Thanks a lot for the reproducible example @MatteaE!
Probably a bug introduced in #436. Our tests for the coregistration are a bit old (2 years) and didn't catch this... It's one of our top priorities to update them.
I'll look at this in details later this week 🙂
rhugonnet
changed the title
Coregistration: Tilt and VerticalShift regression in 0.0.16
Coregistration: Tilt and VerticalShift pipelines in 0.0.16
Nov 27, 2023
Found the bug, 4 letters were missing in a variable of CoregPipeline.fit(). Added tests in #450 so that won't happen again 😄.
I'll release a 0.0.17 with the fix soon.
I've encountered a regression with coregistration between 0.0.13 and 0.0.16, affecting (at least) Tilt and VerticalShift when used in combination with NuthKaab.
Attached the relevant dh plots: coregistration result is wrong in 0.0.16 with any combination of NuthKaab + one of the two methods mentioned above; when used individually, NuthKaab and VerticalShift behave as expected also in 0.0.16. Also attached the input files to reproduce.
xdem_coreg_issue_data.zip
The text was updated successfully, but these errors were encountered: