Package edu.cmu.minorthird.classify.experiments

Class Summary
ClassifiedDataset Pairs a dataset and a classifier, for easy inspection of the actions of a classifier.
ClassifiedDataset.ExplanationViewer  
CrossValidatedDataset View result of some sort of train/test experiment.
CrossValSplitter<T> Split into k separate disjoint folds, then return k train/test splits where each train set is the union of k-1 folds, and the test set is the k-th fold.
Evaluation Stores some detailed results of evaluating a classifier on data.
Evaluation.ConfusionMatrixViewer  
Evaluation.ElevenPointPrecisionViewer  
Evaluation.Matrix  
Evaluation.PropertyViewer  
Evaluation.ROCViewer  
EvaluationGroup A group of related evaluations.
Expt Simple experiment on a classifier.
FixedTestSetSplitter<T> Provides exactly one 'split', between the entire set given, and a fixed designated test set.
LeaveOneOutSplitter<T> Do N-fold cross-validation, where N is the number of different subpopulations.
RandomSplitter<T> Split into one train, one test partition.
SimpleRandomSplitter<T> Split into one train, one test partition.
StrataSorter Helper class for splitting up iterators over Examples, by class.
StratifiedCrossValSplitter Works with datasets of binary examples.
SubpopSorter<T> Helper class for splitting up iterators over Instances
SubsamplingCrossValSplitter<T> Variant of cross-validation in which not all training data is used.
Tester Test a classifier, in a number of ways.
WebmasterSplitter<T> A complicated splitter that stratifies samples according to an arbitrary "profile" property, and restricts train/test splits to not cross boundaries defined by "user" and "request" properties.