Running NewSegment properly#258
Conversation
bthirion
left a comment
bthirion
left a comment
There was a problem hiding this comment.
Sounds good, but having a test would be much better imho
| # prepare template TPMs | ||
| tissue1 = ((os.path.join(SPM_DIR, tissue_path, 'TPM.nii'), 1), | ||
| 2, (True, True), (False, False)) | ||
| 2, (True, True), (True, True)) |
There was a problem hiding this comment.
A short comment on these parameters would be welcome
|
|
||
| normalize: bool, optional (default False) | ||
| flag indicating whether warped brain compartments (gm, wm, csf) are to | ||
| be generated (necessary if the caller wishes the brain later) |
There was a problem hiding this comment.
This is a flag related to spm.Segment. It doesn't hold for spm.NewSegment: to be removed.
| subject_data.hardlink_output_files(final=True) | ||
|
|
||
| finalize_report() | ||
| return subjects |
There was a problem hiding this comment.
Could we have a test that takes an anat image as input and checks that tpms are generated and / or normalized ?
|
This doesn't look good. |
%20anatomical_image.png)
|
Because of SPM8, I've tested under SPM12 and it's working. |
|
LGTM now. |
|
As long as the fsl_feeds example on circle-ci looks ugly (see thumbnail above), this looks like a regression to me. Could you please propagate your changes so that circle-ci looks good ? |
|
Looks like this PR is going to rot here really fast :/. Any updates on the raise issues ? |
|
No, please merge it. We badly need it. |
|
I think we need to drop spm8 support.
|
|
LGTM. |
3 participants
Trying to address #77
do_subject_newsegmentdo_subjects_darteldo_subject_normalize