Nilearn is a Python module for fast and easy statistical learning on NeuroImaging data.
It leverages the scikit-learn Python toolbox for multivariate statistics with applications such as predictive modelling, classification, decoding, or connectivity analysis.
This work is made available by a community of people, amongst which the INRIA Parietal Project Team and the scikit-learn folks, in particular P. Gervais, A. Abraham, V. Michel, A. Gramfort, G. Varoquaux, F. Pedregosa, B. Thirion, M. Eickenberg, C. F. Gorgolewski, D. Bzdok, L. Esteve and B. Cipollini.
- Official source code repo: https://github.com/nilearn/nilearn/
- HTML documentation (stable release): http://nilearn.github.io/
The required dependencies to use the software are:
- Python >= 2.7,
- setuptools
- Numpy >= 1.6.1
- SciPy >= 0.9
- Scikit-learn >= 0.14.1
- Nibabel >= 1.2.0
If you are using nilearn plotting functionalities or running the examples, matplotlib >= 1.1.1 is required.
If you want to run the tests, you need nose >= 1.2.1 and coverage >= 3.6.
First make sure you have installed all the dependencies listed above. Then you can install nilearn by running the following command in a command prompt:
pip install -U --user nilearn
More detailed instructions are available at http://nilearn.github.io/introduction.html#installation.
Detailed instructions on how to contribute are available at http://nilearn.github.io/contributing.html