python-mvpa binary package in Ubuntu Precise i386

 Python module to ease pattern classification analyses of large
 datasets. 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.

Source package