Installation¶

Instructions are given for Linux and Python 2.7. The package is mainly developed and tested under Ubuntu 14.04, we then recommend this distribution. We officially support Ubuntu 12.04, Ubuntu 14.04, Debian stable (jessie), Fedora 21 and 22, and Arch Linux. See below for instructions for each distribution.

For windows users, a linux distribution can be installed easily inside a virtual machine such as Virtual Box. This wiki page explains how to install ubuntu from scratch in Virtual Box or you can get some free VirtualBox images.

Requirements / Dependencies¶

Dependencies are:

python 2.7 with development headers
numpy >= 1.0
scipy >= 0.7
matplotlib
nibabel
sympy (required by nipy)
nipy
a C compiler

Optional dependencies:

PyQt4 (viewer and XML editor)
joblib (local distributed computing)
paramiko (local network distributed computing)
soma-workflow (remote distributed computing)
scikit-learn (clustering)
sphinx (to generate documentation)
pygraphviz (for optimized graph operations and outputs)
munkres (parcellation operation)

Linux Installation¶

Please refer to the page corresponding to your distribution.

If you have another distribution or you are an advanced user, consider installing pyhrf in a virtual environment.