Multivariate Pattern Analysis in Python |
The term Searchlight refers to an algorithm that runs a scalar DatasetMeasure on all possible spheres of a certain size within a dataset (that provides information about distances between feature locations). The measure typically computed is a cross-validated transfer error (see CrossValidatedTransferError). The idea to use a searchlight as a sensitivity analyzer on fMRI datasets stems from Kriegeskorte et al. (2006).
A searchlight analysis is can be easily performed. This examples shows a minimal draft of a complete analysis.
First import a necessary pieces of PyMVPA – this time each bit individually.
For the sake of simplicity, let’s use a small artificial dataset.
# overcomplicated way to generate an example dataset
ds = normalFeatureDataset(perlabel=10, nlabels=2, nchunks=2,
nfeatures=10, nonbogus_features=[3, 7],
snr=5.0)
dataset = MaskedDataset(samples=ds.samples, labels=ds.labels,
chunks=ds.chunks)
Now it only takes three lines for a searchlight analysis.
# setup measure to be computed in each sphere (cross-validated
# generalization error on odd/even splits)
cv = CrossValidatedTransferError(
TransferError(LinearCSVMC()),
OddEvenSplitter())
# setup searchlight with 5 mm radius and measure configured above
sl = Searchlight(cv, radius=5)
# run searchlight on dataset
sl_map = sl(dataset)
print 'Best performing sphere error:', min(sl_map)
If this analysis is done on a fMRI dataset using NiftiDataset the resulting searchlight map (sl_map) can be mapped back into the original dataspace and viewed as a brain overlay. Another example shows a typical application of this algorithm.
.. seealso::
The full source code of this example is included in the PyMVPA source distribution (doc/examples/searchlight_minimal.py).