[Neuroimaging] Contributing

Matthew Brett matthew.brett at gmail.com
Wed Sep 14 12:24:46 EDT 2016


Hi,

On Tue, Sep 13, 2016 at 9:40 AM, Ruy Valle via Neuroimaging
<neuroimaging at python.org> wrote:
> Hi Ariel,
>
> Thank you for your response. I am doing fMRI analysis, mostly task-based up until now but I will be getting into resting-state analysis sooner or later. I will probably also start looking at EEG and TMS, but I am not sure how relevant that is for Nipy. I learned to use AFNI at their bootcamp and FSL through my supervisor at work. I recently (yesterday) started working on a project of mine through which I am hoping to clarify what the effect of including motion parameters (estimates of head motion calculated during volume registration/motion correction) in regression models used in fMRI software is. I would say that I have a fair amount of experience in Java but have not used it much as of late, am comfortable enough with Python, R, and MATLAB to really dive into them as much as needed, and have been learning C and Go recently.
>
> A colleague of mine introduced me to Nipy/Nipype around 5 months ago and I really liked the idea of it. I was still in college at the time and graduated this last semester from McGill University. I have been reading a book on modeling techniques and learning about statistics more generally, so contributing on that side could be nice. I also enjoy learning about algorithms (I implemented a heap sort algorithm in C a few weeks ago). To be honest, I find I still lack enough experience (both in software and in life in general) to pinpoint my preferences, and am open to trying new things, hopefully discovering what most appeals to me, increasing my skills, and making useful contributions along the way.
>

I find that the most productive way, is to start doing some analysis
I'm interested in, and then look to see how to do it in nipy (or dipy,
nibabel or nipype etc).  That gets me reading the code.   I start to
use the code, and that points out bugs, or documentation that could be
improved, or features I'd like.  That makes me look at the code and
maybe ask for help if it's hard to follow.  Then I make some edits and
put up a work-in-progress pull request.

Is that a practical path for you?

Cheers,

Matthew


More information about the Neuroimaging mailing list