gov.sandia.cognition.algorithm

Interface MeasurablePerformanceAlgorithm

All Known Subinterfaces:

ParameterCostMinimizer<ResultType>

All Known Implementing Classes:

AbstractBaumWelchAlgorithm, AbstractBracketedRootFinder, AbstractFactorizationMachineLearner, AbstractMinimizerBasedParameterCostMinimizer, AbstractParameterCostMinimizer, AbstractRootFinder, AffinityPropagation, BatchMultiPerceptron, BaumWelchAlgorithm, DirichletProcessClustering, DistributionParameterEstimator, FactorizationMachineAlternatingLeastSquares, FactorizationMachineStochasticGradient, FletcherXuHybridEstimation, GaussianContextRecognizer.Learner, GaussNewtonAlgorithm, GeneralizedHebbianAlgorithm, GeneticAlgorithm, KernelAdatron, KernelBasedIterativeRegression, KernelPerceptron, KMeansClusterer, KMeansClustererWithRemoval, LeastSquaresEstimator, LevenbergMarquardtEstimation, MetropolisHastingsAlgorithm, MiniBatchKMeansClusterer, MixtureOfGaussians.EMLearner, MixtureOfGaussians.Learner, OptimizedKMeansClusterer, ParallelBaumWelchAlgorithm, ParallelizedGeneticAlgorithm, ParallelizedKMeansClusterer, ParameterDerivativeFreeCostMinimizer, ParameterDifferentiableCostMinimizer, Perceptron, RootFinderBisectionMethod, RootFinderFalsePositionMethod, RootFinderNewtonsMethod, RootFinderRiddersMethod, RootFinderSecantMethod, ScalarMixtureDensityModel.EMLearner, SequentialMinimalOptimization, SimulatedAnnealer, SuccessiveOverrelaxation

@CodeReview(reviewer="Kevin R. Dixon",
            date="2008-12-02",
            changesNeeded=false,
            comments={"I do wonder if we should return a Collection of NamedValues here.","Otherwise, looks good."})
public interface MeasurablePerformanceAlgorithm

An interface for an algorithm that has a measurable quantity of performance that can be retrieved. Typically, this is used after each iteration, in conjunction with the IterativeAlgorithm interface, but it can also be used on its own to determine the performance of the algorithm after it has completed.

Since:

2.1

Author:

Justin Basilico

See Also:

IterativeAlgorithm

Method Summary

Modifier and Type	Method and Description
NamedValue<? extends java.lang.Number>	getPerformance() Gets the name-value pair that describes the current performance of the algorithm.

Method Detail

getPerformance

NamedValue<? extends java.lang.Number> getPerformance()

Gets the name-value pair that describes the current performance of the algorithm. For most algorithms, this is the value that they are attempting to optimize.

Returns:

The name-value pair that describes the current performance of the algorithm.