[pymvpa] Support vector regression analysis error

[jian hou]

Dear all,
I am trying to perform Epsilon-insensitive SVR with a linear kernel using a searchlight procedure with a 3-voxel radius. 
Here is my code:
clf = SVM(svm_impl='EPSILON_SVR', C = 0.1, tube_epsilon = 0.01, kernel='Linear')
cv = CrossValidation(clf, NFoldPartitioner(), postproc=mean_sample(),
            errorfx=corr_error, enable_ca=['training_stats', 'stats'])
sl = sphere_searchlight(cv, radius=3)
res = sl(my_dataset)
However, I got an error which read:
'str' object has no attribute 'params'
Can anyone help me with figuring this out?
Best,
Jian Hou
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/pkg-exppsy-pymvpa/attachments/20160220/c22aa733/attachment.html>