Cross-validation part 2: using multiple classifiers

Using the cross-validation just implemented in ex_nfold_partitioner, run cross validation on our favourite dataset (s01 with t-stats per run). Run this with three classifiers: SVM, neareast neighbor, and naive bayesian, and print the classification accuracies.

As an additional exercise, implement cosmo_confusion_matrix and show the confusion between predicted and actual classes and show the confusion matrix for each classifier.

Hint: run_cross_validation_skl

Solution: run_cross_validation / run_cross_validation_pb