ENH+FIX: Add 3dTproject AFNI interface, Fix OneDToolPy, Add -noFDR flag to 3dDeconvolve #2426
Conversation
mvdoc
changed the title
|
@effigies @mgxd I'm trying to fix 3dDeconvolve in that if I specify a number of jobs, MultiProc still believe it needs only one proc, see example below ( I tried to set Thanks |
|
You can set |
nipype/interfaces/afni/preprocess.py
Outdated
| * This is a file of 1s and 0s, indicating which | ||
| time points are to be included (1) and which are | ||
| to be excluded (0).""", | ||
| argstr="-censor %s") |
There was a problem hiding this comment.
This seems to be an input, so I think you probably need exists=True?
nipype/interfaces/afni/preprocess.py
Outdated
| bandpass = traits.Tuple( | ||
| traits.Float, traits.Float, | ||
| desc="""Remove all frequencies EXCEPT those in the range""", | ||
| argstr='-bandpass %f %f') |
There was a problem hiding this comment.
Does it make sense to use %g, which will generally give you nicer formatting? e.g. 2 vs 2.0000 or 1.395 vs 1.3949999999.
There are some other instances of %f in here. Your call whether to switch to %g.
|
Thanks for the comments @effigies . Added some changes |
nipype/interfaces/afni/model.py
Outdated
| @@ -153,7 +153,7 @@ class DeconvolveInputSpec(AFNICommandInputSpec): | |||
| 'instead of the bucket dataset, if possible.', | |||
| argstr='-cbucket %s') | |||
| out_file = File(desc='output statistics file', argstr='-bucket %s') | |||
| jobs = traits.Int( | |||
| num_threads = traits.Int( | |||
| desc='run the program with provided number of sub-processes', | |||
| argstr='-jobs %d') | |||
There was a problem hiding this comment.
Might want to set this to nohash=True, unless there's reason to think the number of threads affects the output.
|
@effigies - do you think the error we see is likely because of pybids changes, since we pull pybids from github master. i think i saw a reorganization of the dataset email somewhere. '/home/travis/pybids/bids/grabbids/tests/data/ds005' |
|
Yes, that changed in the latest PRs. I'll push a fix in the morning if I
don't find time tonight.
…On Feb 11, 2018 18:22, "Satrajit Ghosh" ***@***.***> wrote:
@effigies <https://github.com/effigies> - do you think the error we see
is likely because of pybids changes, since we pull pybids from github
master. i think i saw a reorganization of the dataset email somewhere.
'/home/travis/pybids/bids/grabbids/tests/data/ds005'
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#2426 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AAFF8pxLMyFqy6ZI0fPlFdDOl3claC0mks5tT3ZOgaJpZM4R4PjH>
.
|
|
@mvdoc - could you please merge with master and push? that should fix the travis error |
* master: (32 commits) DOC: Use pip-packaged numpydoc FIX: Update pybids data directory sty: spacing enh: improve travis build reliability enh: add test from nipy#2431 fix: set _id property of cloned node too fix: syntax sty: pep8 and better error messages Fix expected CL in doctest Change auto test accordingly enh+sty: autotest and styling enh: add c3d and c4d interface Fix issue nipy#2408 fix: updated custom search to current api fix out_file template bugs FIX: Do not cache grab_exts fix: remove deprecated output from _list_outputs ENH: Reorder traits for better error message ENH: Add info to ImageFile for better error messages ENH: Simplify listifying logic ...
|
@satra done |
Changes proposed in this pull request
interfaces.afni.preprocess.TProjectwrapping3dTprojectout_fileforOneDToolPyifcensor_motionis set-noFDRflag to3dDeconvolvejobstonum_threadsin3dDeconvolvefor consistency with other interfaces