[Neuroimaging] Installing dipy without compiling

Ariel Rokem arokem at gmail.com
Thu Oct 8 18:09:51 CEST 2015


Hi Julien,

It should be possible for us to provide pre-compiled binaries for different
platforms. Ideally, that's what we should have on pypi, so that a pip
install doesn't require compilation. However, we are rather stretched as it
is, so we don't usually do that. It would be good to hear what exactly are
your needs: OSX would be easier to support (at least for machines that are
fairly up-to-date, no Power PCs, please...); Windows might be a bit more
challenging (though we do now have a few developers on the team who work on
that platform).

At the very least, I think that you should be able to compile dipy on your
target machine, and distribute that together with your toolbox. I don't see
any reason that wouldn't work.

Cheers,

Ariel


On Sun, Oct 4, 2015 at 8:38 AM, Julien Cohen-Adad <jcohen at polymtl.ca> wrote:

> Hi,
>
> For the software we are distributing (
> https://github.com/neuropoly/spinalcordtoolbox), we would like to include
> dipy.
>
> It's fairly easy to install it automatically via pip, however the OSX
> platform requires XCode (or other compilers). Since this is a problem for
> some of our users, I was wondering if there is a good way to precompile
> dipy and include it in our package. Currently we do include some binaries
> (e.g. ANTs) that work for OSX (>10.8) and most linux distributions
> therefore this solution is very easy to support.
>
> Thank you for your help,
>
> Julien
>
> --
> Julien Cohen-Adad, PhD
> Assistant Professor, Polytechnique Montreal
> Associate Director, Functional Neuroimaging Unit, University of Montreal
> Phone: 514 340 5121 (office: 2264);  Skype: jcohenadad
> Web: www.neuro.polymtl.ca
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20151008/043e09c6/attachment.html>


More information about the Neuroimaging mailing list