Simple lflip #335
Simple lflip #335
Conversation
|
wetrun testing right now |
|
My wetrun test seems to have passed (I also modified the test to handle the new manualseg) Note that the templateseg neonate pipeline doesn't look great (the registration seems to sometimes place the hippocampus in the wrong place), and it may even have performed better using nnunet. I think this PR is ready for merge! |
|
Ok great - I'm at my computer now so I'll have a quick look now before merging |
hippunfold/config/snakebids.yml
Outdated
| @@ -356,37 +356,28 @@ xfm_identity_itk: resources/etc/identity_xfm_itk.txt | |||
|
|
|||
| #templates enabled for template-based segmentation are here | |||
| # TODO: should also perhaps include modalities avaialble, and any custom crop_native_res settings | |||
| template_based_segmentation: | |||
| template_available_hemis: | |||
There was a problem hiding this comment.
Why not put this inside the template_files dict to keep things together? I can do that in my edits..
There was a problem hiding this comment.
just made that edit -- generally trying to keep things hierarchical in the config (will make anotehr issue to do this en-masse, as the config right now is really messy with so many different top-level keys (which don't remain sorted in the nice way that we have them when deployed to an output folder)..
- adds a hemi_wildcards field instead
|
this is good to merge from my end -- I think you could do a Squash and merge instead, to keep the history clean (since all the commits can conceptually fit in a single commit) |
Here we remove all instances of flipping EXCEPT for templateseg and and shape_inject (where we now apply them to the templates instead of the images)