Dear PyMVPA users, Sorry for posting this question again (as I haven't received any answer to my previous post). >From the permutation test example in the manual, I suppose the code for >performing permutation test for cross-validation is like: ====================clf = LinearNuSVMC()nd = MCNullDist(permutations=1000,tail='left',enable_states=['dist_samples'])terr = TransferError(clf,null_dist=nd)cv = CrossValidatedTransferError(terr,NFoldSplitter(),enable_states=['confusion'])error = cv(dataset)P_value = terr.null_probPerm_samples = nd.dist_samples==================== Is the above code correct? Is the permutation automatically performed for the training dataset in each fold? Many thanks,Meng
_______________________________________________ Pkg-ExpPsy-PyMVPA mailing list [email protected] http://lists.alioth.debian.org/mailman/listinfo/pkg-exppsy-pymvpa
