[pymvpa] Searchlight and permutation

Jan Derrfuss derrfuss at gmail.com
Thu Apr 9 18:02:01 UTC 2015


Hi all,

Now that Michael solved the timing mystery, I'm back to the question of 
how to get the p-values from a searchlight analysis with permutations. 
So, here's the current code:


repeater = Repeater(count=10)  # just 10 for testing

partitioner = NFoldPartitioner(cvtype=1)

permutator = AttributePermutator('targets', limit={'partitions': 1}, 
count=1)

null_cv = CrossValidation(clf,
                             ChainNode(
                                     [partitioner, permutator],
                                     space=partitioner.get_space()),
                             postproc=mean_sample())

distr_est = MCNullDist(repeater, tail='left',
                         measure=null_cv,
                         enable_ca=['dist_samples'])

cv_mc_corr = CrossValidation(clf,
                             partitioner,
                             postproc=mean_sample(),
                             null_dist=distr_est,
                             enable_ca=['stats'])

sl = sphere_searchlight(cv_mc_corr, radius = 3, postproc = 
mean_sample(), enable_ca=['stats'])


Based on sphere_searchlight? in iypthon I suspected that the answer 
might be pass_attr, but various variants I tried failed. Also, a number 
of mailing list threads I looked at turned out to be dead ends. Again, 
any input would be very much appreciated.

Jan




More information about the Pkg-ExpPsy-PyMVPA mailing list