[pymvpa] Searchlight-MVPA-ValueError

[J]

Dear all,
I am trying to run a very simple searchlight on fMRI data via PyLab
(on Mac Leopard).

My code is as follows:

from mvpa.suite import *
import os
from matplotlib.pyplot import figure, show
from mvpa.misc.io.base import SampleAttributes
from mvpa.datasets.nifti import NiftiDataset

if __debug__:
    debug.active += ["SLC"]

attr = SampleAttributes(os.path.join(pymvpa_dataroot, 'attributes_test.txt'))
dataset = NiftiDataset(samples=os.path.join(pymvpa_dataroot,
'time_series_original_run_all.nii.gz'),
                       labels=attr.labels,
                       chunks=attr.chunks,
                       mask=os.path.join(pymvpa_dataroot,
'anatomy_mask.nii.gz'))
detrend(dataset, perchunk=True, model='linear')
zscore(dataset, perchunk=True, baselinelabels=[1], targetdtype='float32')

# choose classifier
clf = LinearCSVMC()

# setup measure to be computed by Searchlight
# cross-validated mean transfer using an Odd-Even dataset splitter
cv = CrossValidatedTransferError(TransferError(clf),
                                 OddEvenSplitter())

cv = CrossValidatedTransferError(
        transfer_error=TransferError(LinearCSVMC(),
        splitter=OddEvenSplitter())
s1 = Searchlight(cv, radius=5)
s1_map = s1(dataset)
dataset.map2Nifti(s1_map).save('searchlight_5mm.nii.gz')

---

this runs fine for a while and then it crashes and gives me the
following errors which I am not sure what they mean.

optimization finished, #iter = 59
nu = 0.775000
obj = -0.000003, rho = -0.999986
nSV = 67, nBSV = 57
Total nSV = 414
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)

/opt/local/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/site-packages/<ipython
console> in <module>()

/opt/local/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/site-packages/mvpa/measures/base.pyc
in __call__(self, dataset)
    103         container applying transformer if such is defined
    104         """
--> 105         result = self._call(dataset)
    106         result = self._postcall(dataset, result)
    107         return result

/opt/local/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/site-packages/mvpa/measures/searchlight.pyc
in _call(self, dataset)
    106
    107             # compute the datameasure and store in results

--> 108             measure = self.__datameasure(sphere)
    109             results.append(measure)
    110

/opt/local/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/site-packages/mvpa/measures/base.pyc
in __call__(self, dataset)
    103         container applying transformer if such is defined
    104         """
--> 105         result = self._call(dataset)
    106         result = self._postcall(dataset, result)
    107         return result

/opt/local/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/site-packages/mvpa/algorithms/cvtranserror.pyc
in _call(self, dataset)
    171
    172             # run the beast

--> 173             result = transerror(split[1], split[0])
    174
    175             # unbind the testdataset from the classifier


/opt/local/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/site-packages/mvpa/clfs/transerror.pyc
in __call__(self, testdataset, trainingdataset)
   1300         Returns a scalar value of the transfer error.
   1301         """
-> 1302         self._precall(testdataset, trainingdataset)
   1303         error = self._call(testdataset, trainingdataset)
   1304         self._postcall(testdataset, trainingdataset, error)

/opt/local/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/site-packages/mvpa/clfs/transerror.pyc
in _precall(self, testdataset, trainingdataset)
   1256                     self.__clf.states._changeTemporarily(
   1257                         enable_states=['training_confusion'])
-> 1258                 self.__clf.train(trainingdataset)
   1259                 if self.states.isEnabled('training_confusion'):
   1260                     self.training_confusion =
self.__clf.training_confusion

/opt/local/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/site-packages/mvpa/clfs/base.pyc
in train(self, dataset)
    366
    367         if dataset.nfeatures > 0:
--> 368             result = self._train(dataset)
    369         else:
    370             warning("Trying to train on dataset with no
features present")

/opt/local/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/site-packages/mvpa/clfs/libsvmc/svm.pyc
in _train(self, dataset)
    185                 libsvm_param._setParameter('weight', weight)
    186
--> 187         self.__model = svm.SVMModel(svmprob, libsvm_param)
    188
    189

/opt/local/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/site-packages/mvpa/clfs/libsvmc/_svm.pyc
in __init__(self, arg1, arg2)
    267             msg = svmc.svm_check_parameter(prob.prob, param.param)
    268             if msg:
--> 269                 raise ValueError, msg
    270             self.model = svmc.svm_train(prob.prob, param.param)
    271

ValueError: C <= 0


-------

Your input would be greatly appreciated.

Thanks a lot,
J