python-mvpa

multivariate pattern analysis with Python


PyMVPA eases pattern classification analyses of large datasets, with an accent on neuroimaging. It provides high-level abstraction of typical processing steps (e.g. data preparation, classification, feature selection, generalization testing), a number of implementations of some popular algorithms (e.g. kNN, GNB, Ridge Regressions, Sparse Multinomial Logistic Regression), and bindings to external machine learning libraries (libsvm, shogun).

While it is not limited to neuroimaging data (e.g. fMRI, or EEG) it is eminently suited for such datasets.

Related packages: python-mvpa-doc, python-mvpa-lib


Additional information
  • Michael Hanke, Yaroslav O. Halchenko, Per B. Sederberg, Stephen Jos{\'e} Hanson, James V. Haxby, Stefan Pollmann (2009). PyMVPA: A Python toolbox for multivariate pattern analysis of fMRI data. Neuroinformatics, 7, 37--53. URL DOI PubMed

Maintainer information

This software package is maintained for (Neuro)Debian by the follow individuals and/or groups:

Maintainer avatar
NeuroDebian Maintainers
Maintainer avatar
Michael Hanke
Maintainer avatar
Yaroslav Halchenko

In order to get support, or to get in touch with a maintainer, please click the ‘Help’ button at the top of the page.

Advanced user information

Version control system available: Browse sources

Package availability chart
Distribution Base version Our version Architectures
Debian GNU/Linux 6.0 (squeeze) 0.4.5~dev23-2 0.4.8-1~nd60+1 i386, amd64, sparc
Debian GNU/Linux 7.0 (wheezy) 0.4.8-1 0.4.8-1~nd70+1 i386, amd64, sparc
Debian GNU/Linux 8.0 (jessie)   0.4.8-1~nd70+1 i386, amd64, sparc
Debian unstable (sid) 0.4.8-3 0.4.8-1~nd+1 i386, amd64, sparc
Ubuntu 10.04 LTS “Lucid Lynx” (lucid) 0.4.3-2 0.4.8-1~nd10.04+1 i386, amd64, sparc
Ubuntu 12.04 LTS “Precise Pangolin” (precise) 0.4.7-2ubuntu1 0.4.8-1~nd11.10+1+nd12.04+1 i386, amd64, sparc
Ubuntu 14.04 “Trusty Tahr” (trusty) 0.4.8-3    
Ubuntu 14.10 “Utopic Unicorn” (utopic) 0.4.8-3    
Ubuntu 15.04 “Vivid Vervet” (vivid) 0.4.8-3    
The source code for this portal is licensed under the GPL-3 and is available on GitHub.