Package org.neuroph.eval

Interface Summary

Interface	Description
Evaluator<T>	Generic interface for all types of evaluators

Class Summary

Class	Description
ClassifierEvaluator
ClassifierEvaluator.Binary	Binary evaluator used for computation of metrics in case when data has only one output result (one output neuron)
ClassifierEvaluator.MultiClass	Evaluator used for computation of metrics in case when data has multiple classes - one vs many classification
CrossValidationBak	This class implements multithreaded cross validation procedure.
ErrorEvaluator	Calculates scalar evaluation result using ErrorFunction
Evaluation	Evaluation service used to run different evaluators on trained neural network
EvaluationResult	Create class that will hold statistics for all evaluated datasets - avgs, mx, min, std, variation Result of the evaluation procedure
FoldResult	Result from single cross-validation fold, includes neural network, training and validation set, and fold evaluation results (at the moment only confsionMatrix) TODO: add different eveluation metrics, for regression too.
KFoldCrossValidation	This class implements multi-threaded cross validation procedure.