This content refers to an unreleased development version of PyMVPA

To provide the most recent news and documentation www.pymvpa.org reflects the development 0.6 series of PyMVPA. If you are interested in the documentation of the previous stable 0.4 series of PyMVPA, please visit v04.pymvpa.org.

Tiny Example of a Full Cross-Validation

Very, very simple example showing a complete cross-validation procedure with no fancy additions whatsoever.

# get PyMVPA running
from mvpa2.suite import *

# load PyMVPA example dataset
attr = SampleAttributes(os.path.join(pymvpa_dataroot,
                        'attributes_literal.txt'))
dataset = fmri_dataset(samples=os.path.join(pymvpa_dataroot, 'bold.nii.gz'),
                       targets=attr.targets, chunks=attr.chunks,
                       mask=os.path.join(pymvpa_dataroot, 'mask.nii.gz'))

# do chunkswise linear detrending on dataset
poly_detrend(dataset, polyord=1, chunks_attr='chunks')

# zscore dataset relative to baseline ('rest') mean
zscore(dataset, chunks_attr='chunks', param_est=('targets', ['rest']))

# select class face and house for this demo analysis
# would work with full datasets (just a little slower)
dataset = dataset[np.array([l in ['face', 'house'] for l in dataset.sa.targets],
                          dtype='bool')]

# setup cross validation procedure, using SMLR classifier
cv = CrossValidation(SMLR(), OddEvenPartitioner())

# and run it
error = np.mean(cv(dataset))

# UC: unique chunks, UT: unique targets
print "Error for %i-fold cross-validation on %i-class problem: %f" \
      % (len(dataset.UC), len(dataset.UT), error)

See also

The full source code of this example is included in the PyMVPA source distribution (doc/examples/start_easy.py).