Backport 2.2.0 PRs into 2.1.x branch #316
Conversation
The ``sloppy`` option was misused for debugging (also because the name was misused). This PR makes a better use of debug/sloppy boolean arguments. Backport niprepsgh-287 Backport niprepsgh-288
This is to bring SDCFlow's file filtering more in-line with approaches taken in other nipreps. To avoid potential collisions during `get()` parameter assigment, base / additional values are first merged into a dictionary, and then passed in as keyword arguments Backport niprepsgh-292
Codecov ReportBase: 83.03% // Head: 88.47% // Increases project coverage by
Additional details and impacted files@@ Coverage Diff @@
## maint/2.1.x #316 +/- ##
===============================================
+ Coverage 83.03% 88.47% +5.44%
===============================================
Files 25 25
Lines 1851 1927 +76
Branches 278 289 +11
===============================================
+ Hits 1537 1705 +168
+ Misses 282 188 -94
- Partials 32 34 +2
Flags with carried forward coverage won't be shown. Click here to find out more.
Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here. ☔ View full report at Codecov. |
|
Did we resolve a L/R flip in 2.2.0? Here's tmp/tests/test-fieldmap-masked.svg in this 2.1.x PR (left) and master (right).
|
|
I believe we have tests to check we are interpreting the output of TOPUP correctly. This is probably a flip in plotting rather than correction. Unfortunately, I would need time to check this more thoroughly - but your guess about #278 is very likely to be the source of the flip. |
|
No 2.1.2 release is planned. |
Backports #257, #284, #287, #288, #292, #293, #299, #310.
All of these should be uncontroversial to have in a bug-fix release. The main goal is to have a recent CircleCI run before a lot of changes to ensure that we're not regressing in artifacts.