One more question? Can anyone tell how to use the argument flags for example in Brain extraction (BET) the flags -R for robust. I twould be great if you could send an example of using flags..
The interface example includes one flag usage: https://nipype.readthedocs.io/en/latest/interfaces/generated/interfaces.fsl/preprocess.html#bet
but BET is also from FSL, so their mailing list should be a good place for that question as well !
Or at least the BET docs :) https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/BET/UserGuide Hope that helps !!
I have seen this. I have been using bash scripts previously using FSl command line. But nipype seems to be a lot more different when it comes to using flags. if anyone has used the flags somewhere in there script i would appreciate if they could share the usage of the flags. The BET example however talks about he flags but the usage is restricted to the Fractional intesity threshold. I need an example how to use -R (as in bash)
so in the interface example it would be
btr.inputs.robust = True
glad that helped :+1:
Thatworked
@mohitindelhi_gitlab, as already pointed out by @effigies such questions are great for neurostars.org (as more people will be able to help and benefit from it) and FSLeyes is not part of nipype. I would suggest you look into nibabel and nilearn. Especially the latter provides an amazing and divers functionality to plot all sorts of brain imaging data
Chris Gorgolewski This looks like an interesting model for nipype: https://nf-co.re/ Satrajit Ghosh (Gitter)
I would like to use the spm.FieldMap interface in nipype to correct some fMRI epi images for inhomogeneities in the magnetic field
I guess I can run FieldMap, first setting jobtype calculatevdm
then applyvdm
however from the documentation it is not so clear to me this would work as as output only a vdm file is given
Hi everyone, thanks a lot for the hard work! One question about the SPM interface's slice timing parameter (
nipype.interfaces.spm.SliceTiming): documentation says the parameter for the reference slice ref_slice needs to be of type int, but then it says it could be a 1-based index or a reference time point in msec (which would be a float). Right now, it throws an error when I try to use a time point for ref_slice, which would be a float (because I'm using TR/2). Maybe the input type for ref_slice should be changed? SPM12 takes either an integer or float, so ideally nipype should take either an integer or float as well. Thanks!!
Hi everyone, I would like to wrap a python function with an input that can be Either a list or a Str, and i would like to set a default value. But when i call method.default_values, the console returns undefined
There is my code sample:
My_input = traits.Either(traits.ListStr(('a', 'b'),usedefault=True), traits.Str(), desc='''description''', mandatory=False, usedefault=True)
In tthis exemple, i would like to have the list ("a","b) as default value
Thanks for you help and for all the work you are doing on nipype !
There is my code sample:
My_input = traits.Either(traits.ListStr(('a', 'b'),usedefault=True), traits.Str(), desc='''description''', mandatory=False, usedefault=True)
In tthis exemple, i would like to have the list ("a","b) as default value
Thanks for you help and for all the work you are doing on nipype !
Hi there, forgive me if this is the wrong place to ask this - I am having trouble using a Slicer interface (from nipype.interfaces.slicer.registration.brainsresample.BRAINSResample) because the workflow is not detecting the command. I do realize that BRAINSResample is a Slicer module, and can call it on the command line by going to the Slicer install directory and typing "./Slicer --launch BRAINSResample". I suspect that there is a problem with the visibility of the BRAINSresample module. Is it fair to say that if I can type "BRAINSResample" in a terminal window in any directory, then nipype should be able to call it? Note that I do have the Slicer path defined in the PATH environment variable in my ~.bashrc file - but I'm not sure if my python installation sees this environment variable?
hello, i have a project which depends on numpy<=1.14.3, as well as fitz, which depends on nipype, pip attempts to install nipype 1.2.1, which then errors out with
error: numpy 1.14.3 is installed but numpy>=1.15.3 is required by {'nipype'} which is strange because i see in the nipype requirements.txt depends on numpy>=1.9.0, so i wonder if it is something about your travis environment causing the egg to require a later version? if i examine the EGG_INFO in my site-packages, it does indeed depend on 1.15.3:
/opt/docker/ocrolus-core # grep numpy /usr/lib/python3.7/site-packages/nipype-1.2.1-py3.7.egg/EGG-INFO/requires.txt
numpy>=1.15.3
numpy>=1.15.3
Numpy had an issue that only exhibited in Python 3.7, which was fixed in 1.15.3: https://www.opensourceanswers.com/blog/you-shouldnt-use-python-37-for-data-science-right-now.html
If you need an earlier numpy, you should probably revert to Python 3.6.
It looks like that constraint has been in place since 1.1.8 (nipy/nipype@2bd4fc2), so I don't know what would be new.
Is it good to use Nipype for non-neuroimaging applications? I think Nipype amazing and so simple for creating pipelines and workflows, is there a similar library for this that is for general use and provide functionalities like creating workflows and nodes? Or can I use Nipype to create workflows with my own Python's functions?
Thank you
Yup, the core of nipype isn't really neuroimaging specific. Most of
nipype.algorithms and nipype.interfaces will be neuroimaging-related, but even there you have things like nipype.interfaces.base, nipype.interfaces.io, and nipype.interfaces.utility that are pretty generic.