Framework for multivariate pattern analysis (MVPA)

Package Organization

The mvpa2 package contains the following subpackages and modules:

group Algorithms:
group Anatomical Atlases:
group Basic Data Structures:
group Classifiers (supervised learners):
group Feature Selections:
group Mappers (usually unsupervised learners):
group Measures:measures
group Miscellaneous:
 base misc support
group Unittests:
author:Michael Hanke, Yaroslav Halchenko, Nikolaas N. Oosterhof
requires:Python 2.6+
see:The PyMVPA webpage
see:GIT Repository Browser
license:The MIT License <>
copyright:© 2006-2016 Michael Hanke <>
copyright:© 2007-2016 Yaroslav O. Halchenko <>
copyright:© 2012-2016 Nikolaas N. Oosterhof <>


Doi(\*args, \*\*kwargs) Perform no good and no bad
get_random_seed() Generate a random int good for seeding RNG via seed function
pathjoin(a, \*p) Join two or more pathname components, inserting ‘/’ as needed.
seed([random_seed]) Uniform and combined seeding of all relevant random number generators.
test([limit, verbosity, exit_]) Runs the full or a subset of the PyMVPA unittest suite.
wtf([filename]) Report summary about PyMVPA and the system