ENH: Conform confound regressor names to Derivatives RC2 #1343
Conversation
effigies
changed the title
|
For simplicity, I put a At this point, is there anything besides metadata left to update? |
| @@ -245,7 +255,7 @@ def _get_ica_confounds(ica_out_dir, skip_vols, newpath=None): | |||
| # add one to motion_ic_indices to match melodic report. | |||
| aroma_confounds = os.path.join(newpath, "AROMAAggrCompAROMAConfounds.tsv") | |||
| pd.DataFrame(aggr_confounds.T, | |||
| columns=['AROMAAggrComp%02d' % (x + 1) for x in motion_ic_indices]).to_csv( | |||
| columns=['aroma_motion_%02d' % (x + 1) for x in motion_ic_indices]).to_csv( | |||
There was a problem hiding this comment.
@jdkent I want your opinion on this change. We need to go to snake_case, and I didn't want to make it that much longer. I figured aroma_motion would indicate "components classified as motion by AROMA"...
There was a problem hiding this comment.
I think that sounds good 👍, that's how I read it too.
effigies
changed the title
effigies
changed the title
|
@oesteban I'm going to rebase this. I'm not sure what fixes you were going for, but these last merges look messed up. |
|
Yes, I did a bad merge. Please go ahead 👍 |
effigies
force-pushed
the
bids/derivatives_rc1
branch
from
24c41fd
to
b5d56aa
Compare
|
Assuming this passes, I think we should be good to go. |
|
@oesteban I'm seeing desc-aseg and label-aseg. I don't think we output label- anymore. Is this a thing where clearing the anatomical cache is required, or am I missing something? |
|
Looking into this right now |
| @@ -6,6 +6,8 @@ fmriprep/logs/CITATION.md | |||
| fmriprep/logs/CITATION.tex | |||
| fmriprep/sub-01 | |||
| fmriprep/sub-01/anat | |||
| fmriprep/sub-01/anat/sub-01_desc-aparcaseg_dseg.nii.gz | |||
| fmriprep/sub-01/anat/sub-01_desc-aseg_dseg.nii.gz | |||
| fmriprep/sub-01/anat/sub-01_desc-brain_mask.nii.gz | |||
| fmriprep/sub-01/anat/sub-01_desc-preproc_T1w.nii.gz | |||
| fmriprep/sub-01/anat/sub-01_dseg.nii.gz | |||
There was a problem hiding this comment.
Thought for another PR: Since we do sometimes have other dseg files, do we want to specify desc-fast or similar?
There was a problem hiding this comment.
That would be a good idea. I'd rather go with desc-3tissues and leave the provenance for the json file.
Changes proposed in this pull request
Updating confounds headers to match the RC2.
Known remaining tasks:
cosine_*headers in nipype or update in a separate function in nipype.Documentation that should be reviewed
Anywhere we discuss confounds. I believe we have example confounds that should be fixed, as well.