Apache Commons Math 3.6.1 · org.apache.commons.math3.optim.nonlinear.scalar.noderiv.CMAESOptimizer

Modifier and Type	Class and Description
private static class	CMAESOptimizer.DoubleIndex Used to sort fitness values.
private class	CMAESOptimizer.FitnessFunction Normalizes fitness values to the range [0,1].
public static class	CMAESOptimizer.PopulationSize Population size.
public static class	CMAESOptimizer.Sigma Input sigma values.
private static class	CMAESOptimizer.ValuePenaltyPair Stores the value and penalty (for repair of out of bounds point).

Field Summary

Modifier and Type	Field and Description
private RealMatrix	B Coordinate system.
private RealMatrix	BD B*D, stored for efficiency.
private RealMatrix	C Covariance matrix.
private double	cc Cumulation constant.
private double	ccov1 Learning rate for rank-one update.
private double	ccov1Sep Learning rate for rank-one update - diagonalOnly
private double	ccovmu Learning rate for rank-mu update'
private double	ccovmuSep Learning rate for rank-mu update - diagonalOnly
private final int	checkFeasableCount Determines how often a new random offspring is generated in case it is not feasible / beyond the defined limits, default is 0.
private double	chiN Expectation of \|\|N(0,I)\|\| == norm(randn(N,1)).
private double	cs Cumulation constant for step-size.
private RealMatrix	D Scaling.
private double	damps Damping for step-size.
private RealMatrix	diagC Diagonal of C, used for diagonalOnly.
private RealMatrix	diagD Diagonal of sqrt(D), stored for efficiency.
private int	diagonalOnly Defines the number of initial iterations, where the covariance matrix remains diagonal and the algorithm has internally linear time complexity.
private int	dimension Number of objective variables/problem dimension
private double[]	fitnessHistory History queue of best values.
private final boolean	generateStatistics Indicates whether statistic data is collected.
private int	historySize Size of history queue of best values.
private double[]	inputSigma
private final boolean	isActiveCMA Covariance update mechanism, default is active CMA.
private boolean	isMinimize Number of objective variables/problem dimension
private int	iterations Hides org.apache.commons.math3.optim.BaseOptimizer.iterations. Number of iterations already performed.
private int	lambda Population size, offspring number.
private double	logMu2 log(mu + 0.5), stored for efficiency.
private final int	maxIterations Maximal number of iterations allowed.
private int	mu Number of parents/points for recombination.
private double	mueff Variance-effectiveness of sum w_i x_i.
private double	normps Norm of ps, stored for efficiency.
private RealMatrix	pc Evolution path.
private RealMatrix	ps Evolution path for sigma.
private final RandomGenerator	random Random generator.
private double	sigma Overall standard deviation - search volume.
private final List<RealMatrix>	statisticsDHistory History of D matrix.
private final List<Double>	statisticsFitnessHistory History of fitness values.
private final List<RealMatrix>	statisticsMeanHistory History of mean matrix.
private final List<Double>	statisticsSigmaHistory History of sigma values.
private final double	stopFitness Limit for fitness value.
private double	stopTolFun Stop if fun-changes smaller stopTolFun.
private double	stopTolHistFun Stop if back fun-changes smaller stopTolHistFun.
private double	stopTolUpX Stop if x-changes larger stopTolUpX.
private double	stopTolX Stop if x-change smaller stopTolX.
private RealMatrix	weights Array for weighted recombination.
private RealMatrix	xmean Objective variables.

Access	Constructor and Description
public	CMAESOptimizer(int Maximal number of iterations. maxIterations, double Whether to stop if objective function value is smaller than `stopFitness`. stopFitness, boolean Chooses the covariance matrix update method. isActiveCMA, int Number of initial iterations, where the covariance matrix remains diagonal. diagonalOnly, int Determines how often new random objective variables are generated in case they are out of bounds. checkFeasableCount, RandomGenerator Random generator. random, boolean Whether statistic data is collected. generateStatistics, ConvergenceChecker<PointValuePair> Convergence checker. checker)

Method Summary

Modifier and Type	Method and Description
private void	checkParameters() Hides org.apache.commons.math3.optim.BaseMultivariateOptimizer.checkParameters. Checks dimensions and values of boundaries and inputSigma if defined.
private static void	copyColumn(final RealMatrix Source matrix. m1, int Source column. col1, RealMatrix Target matrix. m2, int Target column. col2) Copies a column from m1 to m2.
private static RealMatrix	Returns: the diagonal n-by-n matrix if m is a column matrix or the column matrix representing the diagonal if m is a n-by-n matrix. diag(final RealMatrix Input matrix. m)
private static RealMatrix	Returns: Matrix where the elements of m and n are element-wise divided. divide(final RealMatrix Input matrix 1. m, final RealMatrix Input matrix 2. n)
protected PointValuePair	doOptimize() Implements abstract org.apache.commons.math3.optim.BaseOptimizer.doOptimize. Performs the bulk of the optimization algorithm.
private static RealMatrix	Returns: n-by-m matrix of 0 values out of diagonal, and 1 values on the diagonal. eye(int Number of rows. n, int Number of columns. m)
public List<RealMatrix>	Returns: History of D matrix. getStatisticsDHistory()
public List<Double>	Returns: History of fitness values. getStatisticsFitnessHistory()
public List<RealMatrix>	Returns: History of mean matrix. getStatisticsMeanHistory()
public List<Double>	Returns: History of sigma values. getStatisticsSigmaHistory()
private void	initializeCMA(double[] Initial guess for the arguments of the fitness function. guess) Initialization of the dynamic search parameters
private static int[]	Returns: the inverse of the mapping defined by indices. inverse(final int[] Input index array. indices)
private static RealMatrix	Returns: Matrix representing the element-wise logarithm of m. log(final RealMatrix Input matrix m)
private static double	Returns: the maximum of the matrix element values. max(final RealMatrix Input matrix. m)
private static double	Returns: the maximum of the array values. max(final double[] Input array. m)
private static double	Returns: the minimum of the matrix element values. min(final RealMatrix Input matrix. m)
private static double	Returns: the minimum of the array values. min(final double[] Input array. m)
private static RealMatrix	Returns: n-by-m matrix filled with 1. ones(int Number of rows. n, int Number of columns. m)
public PointValuePair	Returns: a point/value pair that satisfies the convergence criteria. optimize(OptimizationData... Optimization data. In addition to those documented in `MultivariateOptimizer`, this method will register the following data: `Sigma` `PopulationSize` optData) Overrides org.apache.commons.math3.optim.nonlinear.scalar.MultivariateOptimizer.optimize. Stores data and performs the optimization.
protected void	parseOptimizationData(OptimizationData... Optimization data. The following data will be looked for: `Sigma` `PopulationSize` optData) Overrides org.apache.commons.math3.optim.nonlinear.scalar.MultivariateOptimizer.parseOptimizationData. Scans the list of (required and optional) optimization data that characterize the problem.
private static void	push(double[] History queue. vals, double Current best fitness value. val) Pushes the current best fitness value in a history queue.
private double[]	Returns: an array of Gaussian random numbers. randn(int Length of random array. size)
private RealMatrix	Returns: a 2-dimensional matrix of Gaussian random numbers. randn1(int Number of rows. size, int Population size. popSize)
private static RealMatrix	Returns: a matrix which replicates the input matrix in both directions. repmat(final RealMatrix Input matrix. mat, int Number of row replicates. n, int Number of column replicates. m)
private static int[]	Returns: the indices in inverse order (last is first). reverse(final int[] Input index array. indices)
private static RealMatrix	Returns: Matrix representing the selected columns. selectColumns(final RealMatrix Input matrix. m, final int[] Columns to select. cols)
private static RealMatrix	Returns: a sequence as column matrix. sequence(double Start value. start, double End value. end, double Step size. step)
private int[]	Returns: a sorted array of indices pointing into doubles. sortedIndices(final double[] Array of values to be sorted. doubles) Sorts fitness values.
private static RealMatrix	Returns: Matrix representing the element-wise square root of m. sqrt(final RealMatrix Input matrix. m)
private static RealMatrix	Returns: Matrix representing the element-wise square of m. square(final RealMatrix Input matrix. m)
private static RealMatrix	Returns: Row matrix representing the sums of the rows. sumRows(final RealMatrix Input matrix. m)
private static RealMatrix	Returns: the matrix where the elements of m and n are element-wise multiplied. times(final RealMatrix Input matrix 1. m, final RealMatrix Input matrix 2. n)
private static RealMatrix	Returns: Upper triangular part of matrix. triu(final RealMatrix Input matrix. m, int Diagonal position. k)
private void	updateBD(double Negative covariance factor. negccov) Update B and D from C.
private void	updateCovariance(boolean Flag indicating a small correction. hsig, final RealMatrix Fitness-sorted matrix of the argument vectors producing the current offspring. bestArx, final RealMatrix Unsorted matrix containing the gaussian random values of the current offspring. arz, final int[] Indices indicating the fitness-order of the current offspring. arindex, final RealMatrix xmean matrix of the previous generation. xold) Update of the covariance matrix C.
private void	updateCovarianceDiagonalOnly(boolean Flag indicating a small correction. hsig, final RealMatrix Fitness-sorted matrix of the gaussian random values of the current offspring. bestArz) Update of the covariance matrix C for diagonalOnly > 0
private boolean	Returns: hsig flag indicating a small correction. updateEvolutionPaths(RealMatrix Weighted row matrix of the gaussian random numbers generating the current offspring. zmean, RealMatrix xmean matrix of the previous generation. xold) Update of the evolution paths ps and pc.
private double	Returns: a double equal to maximum value minus minimum value. valueRange(final CMAESOptimizer.ValuePenaltyPair[] Array of valuePenaltyPairs to get range from. vpPairs) Get range of values.
private static RealMatrix	Returns: n-by-m matrix of zero values. zeros(int Number of rows. n, int Number of columns. m)

Inherited from org.apache.commons.math3.optim.nonlinear.scalar.MultivariateOptimizer:: computeObjectiveValue getGoalType

public Class CMAESOptimizer

Nested and Inner Type Summary

Field Summary

Constructor Summary

Method Summary