AbstractMultiObjectiveOptimizer. More...
Namespaces | |
blas | |
statistics | |
tags | |
Tags are empty types which can be used as a function argument. | |
Classes | |
class | AbsoluteLoss |
absolute loss More... | |
class | AbstractBudgetMaintenanceStrategy |
This is the abstract interface for any budget maintenance strategy. More... | |
class | AbstractClustering |
Base class for clustering. More... | |
class | AbstractConstraintHandler |
Implements the base class for constraint handling. More... | |
class | AbstractCost |
Cost function interface. More... | |
class | AbstractDistribution |
Abstract class for distributions. More... | |
class | AbstractKernelFunction |
Base class of all Kernel functions. More... | |
class | AbstractLinearSvmTrainer |
Super class of all linear SVM trainers. More... | |
class | AbstractLineSearchOptimizer |
Basis class for line search methods. More... | |
class | AbstractLoss |
Loss function interface. More... | |
class | AbstractMetric |
class | AbstractModel |
Base class for all Models. More... | |
class | AbstractMultiObjectiveOptimizer |
base class for abstract multi-objective optimizers for arbitrary search spaces. More... | |
class | AbstractNearestNeighbors |
Interface for Nearest Neighbor queries. More... | |
class | AbstractObjectiveFunction |
Super class of all objective functions for optimization and learning. More... | |
class | AbstractOptimizer |
An optimizer that optimizes general objective functions. More... | |
class | AbstractSingleObjectiveOptimizer |
Base class for all single objective optimizer. More... | |
class | AbstractStoppingCriterion |
Base class for stopping criteria of optimization algorithms. More... | |
class | AbstractSvmTrainer |
Super class of all kernelized (non-linear) SVM trainers. More... | |
class | AbstractTrainer |
Superclass of supervised learning algorithms. More... | |
class | AbstractUnsupervisedTrainer |
Superclass of unsupervised learning algorithms. More... | |
class | AbstractWeightedTrainer |
Superclass of weighted supervised learning algorithms. More... | |
class | AbstractWeightedUnsupervisedTrainer |
Superclass of weighted unsupervised learning algorithms. More... | |
struct | Ackley |
Convex quadratic benchmark function with single dominant axis. More... | |
struct | AdditiveEpsilonIndicator |
Given a reference front R and an approximation F, calculates the additive approximation quality of F. More... | |
class | ARDKernelUnconstrained |
Automatic relevance detection kernel for unconstrained parameter optimization. More... | |
class | ArgMaxConverter |
Conversion of real-valued outputs to classes. More... | |
class | Autoencoder |
implements the autoencoder More... | |
class | BarsAndStripes |
Generates the Bars-And-Stripes problem. In this problem, a 4x4 image has either rows or columns of the same value. More... | |
struct | BaseFastNonDominatedSort |
Implements the well-known non-dominated sorting algorithm. More... | |
class | BaseRng |
Collection of different variate generators for different distributions. More... | |
struct | Batch |
class which helps using different batch types More... | |
class | Bernoulli |
This class simulates a "Bernoulli trial", which is like a coin toss. More... | |
class | BFGS |
Broyden, Fletcher, Goldfarb, Shannon algorithm for unconstraint optimization. More... | |
class | BiasSolver |
class | BiasSolverSimplex |
class | BinaryLayer |
Layer of binary units taking values in {0,1}. More... | |
class | BinaryTree |
Super class of binary space-partitioning trees. More... | |
class | Binomial |
Models a binomial distribution with parameters p and n. More... | |
class | BipolarLayer |
Layer of bipolar units taking values in {-1,1}. More... | |
struct | BitflipMutator |
Bitflip mutation operator. More... | |
class | BlockMatrix2x2 |
SVM regression matrix. More... | |
struct | BoundingBoxComputer |
Calculates bounding boxes. More... | |
class | BoxConstrainedProblem |
Quadratic program with box constraints. More... | |
struct | BoxConstrainedShrinkingProblem |
class | BoxConstraintHandler |
class | BoxedSVMProblem |
Boxed problem for alpha in [lower,upper]^n and equality constraints. More... | |
class | CachedMatrix |
Efficient quadratic matrix cache. More... | |
struct | CanBeCalled |
detects whether Functor(Argument) can be called. More... | |
class | CARTClassifier |
CART Classifier. More... | |
class | CARTTrainer |
Classification And Regression Trees CART. More... | |
class | Cauchy |
Cauchy distribution. More... | |
class | Centroids |
Clusters defined by centroids. More... | |
class | CG |
Conjugate-gradient method for unconstrained optimization. More... | |
class | Chessboard |
"chess board" problem for binary classification More... | |
struct | Cigar |
Convex quadratic benchmark function with single dominant axis. More... | |
class | CigarDiscus |
Convex quadratic benchmark function. More... | |
struct | CIGTAB1 |
Multi-objective optimization benchmark function CIGTAB 1. More... | |
struct | CIGTAB2 |
Multi-objective optimization benchmark function CIGTAB 2. More... | |
class | CircleInSquare |
class | ClusteringModel |
Abstract model with associated clustering object. More... | |
class | CMA |
Implements the CMA-ES. More... | |
struct | CMAChromosome |
Models a CMAChromosomeof the elitist (MO-)CMA-ES that encodes strategy parameters. More... | |
class | CMACMap |
The CMACMap class represents a linear combination of piecewise constant functions. More... | |
class | CMAIndividual |
class | CMSA |
Implements the CMSA. More... | |
class | CombinedObjectiveFunction |
Linear combination of objective functions. More... | |
class | ConcatenatedModel |
ConcatenatedModel concatenates two models such that the output of the first model is input to the second. More... | |
struct | ConstProxyReference |
sets the type of ProxxyReference More... | |
struct | ConstrainedSphere |
Constrained Sphere function. More... | |
class | ContrastiveDivergence |
Implements k-step Contrastive Divergence described by Hinton et al. (2006). More... | |
class | ConvexCombination |
Models a convex combination of inputs. More... | |
class | ConvolutionalRBM |
Implements a convolutional RBM with a single greyscale input imge and a set of squared image filters. More... | |
class | CrossEntropy |
Error measure for classication tasks that can be used as the objective function for training. More... | |
class | CrossEntropyIndependent |
Error measure for classification tasks of non exclusive attributes that can be used for model training. More... | |
class | CrossEntropyMethod |
Implements the Cross Entropy Method. More... | |
class | CrossValidationError |
Cross-validation error for selection of hyper-parameters. More... | |
class | CSvmDerivative |
This class provides two main member functions for computing the derivative of a C-SVM hypothesis w.r.t. its hyperparameters. The constructor takes a pointer to a KernelClassifier and an SvmTrainer, in the assumption that the former was trained by the latter. It heavily accesses their members to calculate the derivative of the alpha and offset values w.r.t. the SVM hyperparameters, that is, the regularization parameter C and the kernel parameters. This is done in the member function prepareCSvmParameterDerivative called by the constructor. After this initial, heavier computation step, modelCSvmParameterDerivative can be called on an input sample to the SVM model, and the method will yield the derivative of the hypothesis w.r.t. the SVM hyperparameters. More... | |
class | CSVMProblem |
Problem formulation for binary C-SVM problems. More... | |
class | CSvmTrainer |
Training of C-SVMs for binary classification. More... | |
class | CVFolds |
class | Data |
Data container. More... | |
class | DataDistribution |
A DataDistribution defines an unsupervised learning problem. More... | |
class | DataView |
Constant time Element-Lookup for Datasets. More... | |
class | DiagonalWithCircle |
class | DiffGeometric |
Random variable with diff geometric distribution. More... | |
class | DiffGeometric_distribution |
Implements a diff geometric distribution. More... | |
struct | DiffPowers |
class | Dirichlet |
Implements a Dirichlet distribution. More... | |
class | Dirichlet_distribution |
Dirichlet distribution. More... | |
class | DiscreteKernel |
Kernel on a finite, discrete space. More... | |
class | DiscreteLoss |
flexible loss for classification More... | |
class | DiscreteUniform |
Implements the discrete uniform distribution. More... | |
struct | Discus |
Convex quadratic benchmark function. More... | |
class | DistantModes |
Creates a set of pattern (each later representing a mode) which than are randomly perturbed to create the data set. The dataset was introduced in Desjardins et al. (2010) (Parallel Tempering for training restricted Boltzmann machines, AISTATS 2010) More... | |
class | DistTrainerContainer |
Container for known distribution trainers. More... | |
class | Divide |
Transformation function dividing the elements in a dataset by a scalar or component-wise by values stores in a vector. More... | |
class | DoublePole |
struct | DropoutNeuron |
Wraps a given neuron type and implements dropout for it. More... | |
struct | DTLZ1 |
Implements the benchmark function DTLZ1. More... | |
struct | DTLZ2 |
Implements the benchmark function DTLZ2. More... | |
struct | DTLZ3 |
Implements the benchmark function DTLZ3. More... | |
struct | DTLZ4 |
Implements the benchmark function DTLZ4. More... | |
struct | DTLZ5 |
Implements the benchmark function DTLZ5. More... | |
struct | DTLZ6 |
Implements the benchmark function DTLZ6. More... | |
struct | DTLZ7 |
Implements the benchmark function DTLZ7. More... | |
class | ElitistCMA |
Implements the elitist CMA-ES. More... | |
struct | ElitistSelection |
Survival selection to find the next parent set. More... | |
struct | ELLI1 |
Multi-objective optimization benchmark function ELLI1. More... | |
struct | ELLI2 |
Multi-objective optimization benchmark function ELLI2. More... | |
struct | Ellipsoid |
Convex quadratic benchmark function. More... | |
struct | EmptyState |
Default State of an Object which does not need a State. More... | |
struct | Energy |
The Energy function determining the Gibbs distribution of an RBM. More... | |
class | EnergyStoringTemperedMarkovChain |
Implements parallel tempering but also stores additional statistics on the energy differences. More... | |
class | EpsilonHingeLoss |
Hinge-loss for large margin regression. More... | |
class | EpsilonSvmTrainer |
Training of Epsilon-SVMs for regression. More... | |
struct | EPTournamentSelection |
Survival and mating selection to find the next parent set. More... | |
class | Erlang |
Erlang distributed random variable. More... | |
class | Erlang_distribution |
Implements an Erlang distribution. More... | |
class | ErrorFunction |
Objective function for supervised learning. More... | |
class | EvaluationArchive |
Objective function wrapper storing all function evaluations. More... | |
class | ExactGradient |
class | ExampleModifiedKernelMatrix |
class | Exception |
Top-level exception class of the shark library. More... | |
struct | FastSigmoidNeuron |
Fast sigmoidal function, which does not need to compute an exponential function. More... | |
class | FFNet |
Offers the functions to create and to work with a feed-forward network. More... | |
struct | FFNetStructures |
class | FisherLDA |
Fisher's Linear Discriminant Analysis for data compression. More... | |
struct | FitnessExtractor |
struct | Fonseca |
Bi-objective real-valued benchmark function proposed by Fonseca and Flemming. More... | |
class | Gamma |
Gamma distributed random variable. More... | |
class | Gamma_distribution |
Gamma distribution. More... | |
class | GaussianKernelMatrix |
Efficient special case if the kernel is Gaussian and the inputs are sparse vectors. More... | |
class | GaussianLayer |
A layer of Gaussian neurons. More... | |
class | GaussianNoiseModel |
Model which corrupts the data using gaussian noise. More... | |
class | GaussianRbfKernel |
Gaussian radial basis function kernel. More... | |
class | GeneralizationLoss |
The generalization loss calculates the relative increase of the validation error compared to the minimum training error. More... | |
class | GeneralizationQuotient |
SStopping criterion monitoring the quotient of generalization loss and training progress. More... | |
class | GeneralQuadraticProblem |
Most gneral problem formualtion, needs to be configured by hand. More... | |
class | GenericDistTrainer |
class | Geometric |
Implements the geometric distribution. More... | |
class | GibbsOperator |
Implements Block Gibbs Sampling related transition operators for various temperatures. More... | |
class | GridSearch |
Optimize by trying out a grid of configurations. More... | |
class | GruauPole |
Class for balancing two poles on a cart using a fitness function that punishes oscillating, i.e. quickly moving the cart back and forth to balance the poles. Based on code written by Verena Heidrich-Meisner for the paper. More... | |
struct | GSP |
Real-valued benchmark function with two objectives. More... | |
class | HardClusteringModel |
Model for "hard" clustering. More... | |
class | HierarchicalClustering |
Clusters defined by a binary space partitioning tree. More... | |
struct | Himmelblau |
Multi-modal two-dimensional continuous Himmelblau benchmark function. More... | |
class | HingeLoss |
Hinge-loss for large margin classification. More... | |
class | HMGSelectionCriterion |
class | HuberLoss |
Huber-loss for for robust regression. More... | |
class | HyperGeometric |
Random variable with a hypergeometric distribution. More... | |
class | HyperGeometric_distribution |
Hypergeometric distribution. More... | |
struct | HypervolumeApproximator |
Implements an FPRAS for approximating the volume of a set of high-dimensional objects. More... | |
struct | HypervolumeCalculator |
Implementation of the exact hypervolume calculation in m dimensions. More... | |
struct | HypervolumeIndicator |
Calculates the hypervolume covered by a front of non-dominated points. More... | |
struct | IdentityFitnessExtractor |
Functor that returns its argument without conversion. More... | |
struct | IHR1 |
Multi-objective optimization benchmark function IHR1. More... | |
struct | IHR2 |
Multi-objective optimization benchmark function IHR 2. More... | |
struct | IHR3 |
Multi-objective optimization benchmark function IHR3. More... | |
struct | IHR4 |
Multi-objective optimization benchmark function IHR 4. More... | |
struct | IHR6 |
Multi-objective optimization benchmark function IHR 6. More... | |
struct | ImageInformation |
Stores name and size of image externally. More... | |
class | ImagePatches |
Given a set of images, draws a set of image patches of a given size. More... | |
class | ImpulseNoiseModel |
Model which corrupts the data using Impulse noise. More... | |
class | INameable |
This class is an interface for all objects which can have a name. More... | |
class | IndexedIterator |
creates an Indexed Iterator, an Iterator which also carries index information using index() More... | |
class | IndicatorBasedMOCMA |
Implements the generational MO-CMA-ES. More... | |
class | IndicatorBasedRealCodedNSGAII |
Implements the NSGA-II. More... | |
struct | IndicatorBasedSelection |
Implements the well-known indicator-based selection strategy. More... | |
class | IndicatorBasedSteadyStateMOCMA |
Implements the \((\mu+1)\)-MO-CMA-ES. More... | |
class | Individual |
Individual is a simple templated class modelling an individual that acts as a candidate solution in an evolutionary algorithm. More... | |
struct | InvertedGenerationalDistance |
Inverted generational distance for comparing Pareto-front approximations. More... | |
class | IParameterizable |
Top level interface for everything that holds parameters. More... | |
class | IRpropMinus |
This class offers methods for the usage of the improved Resilient-Backpropagation-algorithm without weight-backtracking. More... | |
class | IRpropPlus |
This class offers methods for the usage of the improved Resilient-Backpropagation-algorithm with weight-backtracking. More... | |
class | IRpropPlusFull |
class | ISerializable |
Abstracts serializing functionality. More... | |
class | IterativeNNQuery |
Iterative nearest neighbors query. More... | |
class | JaakkolaHeuristic |
Jaakkola's heuristic and related quantities for Gaussian kernel selection. More... | |
class | KDTree |
KD-tree, a binary space-partitioning tree. More... | |
class | KernelBasisDistance |
Computes the squared distance between the optimal point in a basis to the point represented by a KernelExpansion. More... | |
class | KernelBudgetedSGDTrainer |
Budgeted stochastic gradient descent training for kernel-based models. More... | |
struct | KernelClassifier |
Linear classifier in a kernel feature space. More... | |
class | KernelExpansion |
Linear model in a kernel feature space. More... | |
class | KernelMatrix |
Kernel Gram matrix. More... | |
class | KernelMeanClassifier |
Kernelized mean-classifier. More... | |
class | KernelSGDTrainer |
Generic stochastic gradient descent training for kernel-based models. More... | |
class | KernelTargetAlignment |
Kernel Target Alignment - a measure of alignment of a kernel Gram matrix with labels. More... | |
struct | KeyValuePair |
Represents a Key-Value-Pair similar std::pair which is strictly ordered by it's key. More... | |
struct | KeyValueRange |
class | KHCTree |
KHC-tree, a binary space-partitioning tree. More... | |
class | LabeledData |
Data set for supervised learning. More... | |
class | LabeledDataDistribution |
A LabeledDataDistribution defines a supervised learning problem. More... | |
class | LabelOrder |
This will normalize the labels of a given dataset to 0..N-1. More... | |
class | LassoRegression |
LASSO Regression. More... | |
class | LBFGS |
Limited-Memory Broyden, Fletcher, Goldfarb, Shannon algorithm for unconstrained optimization. More... | |
class | LCTree |
LC-tree, a binary space-partitioning tree. More... | |
class | LDA |
Linear Discriminant Analysis (LDA) More... | |
struct | LeastContributorApproximator |
Approximately determines the point of a set contributing the least hypervolume. More... | |
struct | LibSVMSelectionCriterion |
class | LinearClassifier |
Basic linear classifier. More... | |
class | LinearCSvmTrainer |
class | LinearKernel |
Linear Kernel, parameter free. More... | |
class | LinearMcSvmADMTrainer |
class | LinearMcSvmATMTrainer |
class | LinearMcSvmATSTrainer |
class | LinearMcSvmCSTrainer |
class | LinearMcSvmLLWTrainer |
class | LinearMcSvmMMRTrainer |
class | LinearMcSvmOVATrainer |
class | LinearMcSvmReinforcedTrainer |
class | LinearMcSvmWWTrainer |
class | LinearModel |
Linear Prediction. More... | |
struct | LinearNeuron |
Linear activation Neuron. More... | |
class | LinearNorm |
Normalizes the (non-negative) input by dividing by the overall sum. More... | |
struct | LinearRankingSelection |
Implements a fitness-proportional selection scheme for mating selection that scales the fitness values linearly before carrying out the actual selection. More... | |
class | LinearRegression |
Linear Regression. More... | |
class | LineSearch |
Wrapper for the linesearch class of functions in the linear algebra library. More... | |
class | LMCMA |
Implements a Limited-Memory-CMA. More... | |
struct | LogisticNeuron |
Neuron which computes the Logistic (logistic) function with range [0,1]. More... | |
class | LogNormal |
Implements a log-normal distribution with parameters location m and Scale s. More... | |
class | LooError |
Leave-one-out error objective function. More... | |
class | LooErrorCSvm |
Leave-one-out error, specifically optimized for C-SVMs. More... | |
class | LRUCache |
Implements an LRU-Caching Strategy for arbitrary Cache-Lines. More... | |
struct | LZ1 |
Multi-objective optimization benchmark function LZ1. More... | |
struct | LZ2 |
Multi-objective optimization benchmark function LZ2. More... | |
struct | LZ3 |
Multi-objective optimization benchmark function LZ3. More... | |
struct | LZ4 |
Multi-objective optimization benchmark function LZ4. More... | |
struct | LZ5 |
Multi-objective optimization benchmark function LZ5. More... | |
struct | LZ6 |
Multi-objective optimization benchmark function LZ6. More... | |
struct | LZ7 |
Multi-objective optimization benchmark function LZ7. More... | |
struct | LZ8 |
Multi-objective optimization benchmark function LZ8. More... | |
struct | LZ9 |
class | MarkovChain |
A single Markov chain. More... | |
class | MarkovPole |
struct | MaximumGainCriterion |
Working set selection by maximization of the dual objective gain. More... | |
struct | MaximumGradientCriterion |
Working set selection by maximization of the projected gradient. More... | |
class | MaxIterations |
This stopping criterion stops after a fixed number of iterations. More... | |
class | McPegasos |
Pegasos solver for linear multi-class support vector machines. More... | |
class | McReinforcedSvmTrainer |
Training of reinforced-SVM for multi-category classification. More... | |
class | McSvmADMTrainer |
Training of ADM-SVMs for multi-category classification. More... | |
class | McSvmATMTrainer |
Training of ATM-SVMs for multi-category classification. More... | |
class | McSvmATSTrainer |
Training of ATS-SVMs for multi-category classification. More... | |
class | McSvmCSTrainer |
Training of the multi-category SVM by Crammer and Singer (CS). More... | |
class | McSvmLLWTrainer |
Training of the multi-category SVM by Lee, Lin and Wahba (LLW). More... | |
class | McSvmMMRTrainer |
Training of the maximum margin regression (MMR) multi-category SVM. More... | |
class | McSvmOVATrainer |
Training of a multi-category SVM by the one-versus-all (OVA) method. More... | |
class | McSvmWWTrainer |
Training of the multi-category SVM by Weston and Watkins (WW). More... | |
class | MeanModel |
Calculates the weighted mean of a set of models. More... | |
class | MergeBudgetMaintenanceStrategy |
Budget maintenance strategy that merges two vectors. More... | |
class | MergeBudgetMaintenanceStrategy< RealVector > |
Budget maintenance strategy merging vectors. More... | |
class | MissingFeaturesKernelExpansion |
Kernel expansion with missing features support. More... | |
class | MissingFeatureSvmTrainer |
Trainer for binary SVMs natively supporting missing features. More... | |
class | MklKernel |
Weighted sum of kernel functions. More... | |
class | MNIST |
Reads in the famous MNIST data in possibly binarized form. The MNIST database itself is not included in Shark, this class just helps loading it. More... | |
class | ModelKernel |
Kernel function that uses a Model as transformation function for another kernel. More... | |
class | ModifiedKernelMatrix |
Modified Kernel Gram matrix. More... | |
class | MonomialKernel |
Monomial kernel. Calculates \( \left\langle x_1, x_2 \right\rangle^m_exponent \). More... | |
class | MultiChainApproximator |
Approximates the gradient by taking samples from an ensemble of Markov chains running in parallel. More... | |
class | MultiNomialDistribution |
Implements a multinomial distribution. More... | |
struct | MultiplicativeEpsilonIndicator |
Given a reference front R and an approximation F, calculates the multiplicative approximation quality of F. More... | |
class | Multiply |
Transformation function multiplying the elements in a dataset by a scalar or component-wise by values stores in a vector. More... | |
class | MultiSequenceIterator |
Iterator which iterates of the elements of a nested sequence. More... | |
struct | MultiTaskSample |
Aggregation of input data and task index. More... | |
class | MultiVariateNormalDistribution |
Implements a multi-variate normal distribution with zero mean. More... | |
class | MultiVariateNormalDistributionCholesky |
Multivariate normal distribution with zero mean using a cholesky decomposition. More... | |
struct | MVPSelectionCriterion |
class | NBClassifier |
Naive Bayes classifier. More... | |
class | NBClassifierTrainer |
Trainer for naive Bayes classifier. More... | |
class | NearestNeighborClassifier |
Nearest Neighbor Classifier. More... | |
class | NearestNeighborRegression |
Nearest neighbor regression model. More... | |
class | NegativeAUC |
Negative area under the curve. More... | |
class | NegativeGaussianProcessEvidence |
Evidence for model selection of a regularization network/Gaussian process. More... | |
class | NegativeLogLikelihood |
Computes the negative log likelihood of a dataset under a model. More... | |
class | NegativeWilcoxonMannWhitneyStatistic |
Negative Wilcoxon-Mann-Whitney statistic. More... | |
class | NegExponential |
Implements the Negative exponential distribution. More... | |
class | NestedGridSearch |
Nested grid search. More... | |
class | NoisyErrorFunction |
Error Function which only uses a random fraction of data. More... | |
class | NonMarkovPole |
Objective function for single and double non-Markov poles. More... | |
class | Normal |
Implements a univariate normal (Gaussian) distribution. More... | |
class | NormalDistributedPoints |
Generates a set of normally distributed points. More... | |
class | NormalizeComponentsUnitInterval |
Train a model to normalize the components of a dataset to fit into the unit inverval. More... | |
class | NormalizeComponentsUnitVariance |
Train a linear model to normalize the components of a dataset to unit variance, and optionally to zero mean. More... | |
class | NormalizeComponentsWhitening |
Train a linear model to whiten the data. More... | |
class | NormalizeComponentsZCA |
Train a linear model to whiten the data. More... | |
class | NormalizedKernel |
Normalized version of a kernel function. More... | |
class | NormalizeKernelUnitVariance |
Determine the scaling factor of a ScaledKernel so that it has unit variance in feature space one on a given dataset. More... | |
class | Normalizer |
"Diagonal" linear model for data normalization. More... | |
class | NormalTrainer |
Trainer for normal distribution. More... | |
class | OneClassSvmTrainer |
Training of one-class SVMs. More... | |
class | OneNormRegularizer |
One-norm of the input as an objective function. More... | |
struct | OnePointCrossover |
Implements one-point crossover. More... | |
class | OneVersusOneClassifier |
One-versus-one Classifier. More... | |
class | OnlineRNNet |
A recurrent neural network regression model optimized for online learning. More... | |
class | OptimizationTrainer |
Wrapper for training schemes based on (iterative) optimization. More... | |
class | PairIterator |
A Pair-Iterator which gives a unified view of two ranges. More... | |
struct | PairRangeType |
struct | PairReference |
Given a type of pair and two iterators to zip together, returns the reference. More... | |
class | PamiToy |
struct | ParetoDominanceComparator |
Implementation of the Pareto-Dominance relation under the assumption of all objectives to be minimized. More... | |
class | PartlyPrecomputedMatrix |
Partly Precomputed version of a matrix for quadratic programming. More... | |
class | PCA |
Principal Component Analysis. More... | |
class | Pegasos |
Pegasos solver for linear (binary) support vector machines. More... | |
struct | PenalizingEvaluator |
Penalizing evaluator for scalar objective functions. More... | |
class | Perceptron |
Perceptron online learning algorithm. More... | |
class | PointSearch |
Optimize by trying out predefined configurations. More... | |
class | Poisson |
Implements a Poisson distribution with parameter mean. More... | |
class | PolynomialKernel |
Polynomial kernel. More... | |
struct | PolynomialMutator |
Polynomial mutation operator. More... | |
class | PopulationBasedStepSizeAdaptation |
Step size adaptation based on the success of the new population compared to the old. More... | |
class | PrecomputedMatrix |
Precomputed version of a matrix for quadratic programming. More... | |
class | ProductKernel |
Product of kernel functions. More... | |
class | ProjectBudgetMaintenanceStrategy |
Budget maintenance strategy that projects a vector. More... | |
class | ProjectBudgetMaintenanceStrategy< RealVector > |
Budget maintenance strategy that projects a vector. More... | |
class | ProxyIterator |
Creates an iterator which reinterpretes an object as a range. More... | |
class | QpBoxLinear |
Quadratic program solver for box-constrained problems with linear kernel. More... | |
class | QpBoxLinear< CompressedRealVector > |
class | QpConfig |
Super class of all support vector machine trainers. More... | |
class | QpMcBoxDecomp |
class | QpMcDecomp |
Quadratic program solver for multi class SVM problems. More... | |
class | QpMcLinear |
Generic solver skeleton for linear multi-class SVM problems. More... | |
class | QpMcLinearADM |
Solver for the multi-class SVM with absolute margin and discriminative maximum loss. More... | |
class | QpMcLinearATM |
Solver for the multi-class SVM with absolute margin and total maximum loss. More... | |
class | QpMcLinearATS |
Solver for the multi-class SVM with absolute margin and total sum loss. More... | |
class | QpMcLinearCS |
Solver for the multi-class SVM by Crammer & Singer. More... | |
class | QpMcLinearLLW |
Solver for the multi-class SVM by Lee, Lin & Wahba. More... | |
class | QpMcLinearMMR |
Solver for the multi-class maximum margin regression SVM. More... | |
class | QpMcLinearReinforced |
Solver for the "reinforced" multi-class SVM. More... | |
class | QpMcLinearWW |
Solver for the multi-class SVM by Weston & Watkins. More... | |
class | QpMcSimplexDecomp |
struct | QpSolutionProperties |
properties of the solution of a quadratic program More... | |
class | QpSolver |
Quadratic program solver. More... | |
class | QpSparseArray |
specialized container class for multi-class SVM problems More... | |
struct | QpStoppingCondition |
stopping conditions for quadratic programming More... | |
class | RadiusMarginQuotient |
radius margin quotions for binary SVMs More... | |
class | RBFLayer |
Implements a layer of radial basis functions in a neural network. More... | |
class | RBM |
stub for the RBM class. at the moment it is just a holder of the parameter set and the Energy. More... | |
struct | RealSpace |
The RealSpace can't be enumerated. Infinite values are just too much. More... | |
struct | RectifierNeuron |
Rectifier Neuron f(x) = max(0,x) More... | |
class | RecurrentStructure |
Offers a basic structure for recurrent networks. More... | |
class | RegularizationNetworkTrainer |
Training of a regularization network. More... | |
class | RegularizedKernelMatrix |
Kernel Gram matrix with modified diagonal. More... | |
class | RemoveBudgetMaintenanceStrategy |
Budget maintenance strategy that removes a vector. More... | |
struct | ResultSet |
class | RFClassifier |
Random Forest Classifier. More... | |
class | RFTrainer |
Random Forest. More... | |
class | RNNet |
A recurrent neural network regression model that learns with Back Propagation Through Time. More... | |
class | ROC |
ROC-Curve - false negatives over false positives. More... | |
struct | Rosenbrock |
Generalized Rosenbrock benchmark function. More... | |
struct | RotatedObjectiveFunction |
Rotates an objective function using a randomly initialized rotation. More... | |
struct | RouletteWheelSelection |
Fitness-proportional selection operator. More... | |
class | RpropMinus |
This class offers methods for the usage of the Resilient-Backpropagation-algorithm without weight-backtracking. More... | |
class | RpropPlus |
This class offers methods for the usage of the Resilient-Backpropagation-algorithm with weight-backtracking. More... | |
struct | Sampler |
Samples a random point. More... | |
class | ScaledKernel |
Scaled version of a kernel function. More... | |
struct | Schwefel |
Convex benchmark function. More... | |
class | ScopedHandle |
class | Shift |
Transformation function adding a vector or a scalar to the elements in a dataset. More... | |
class | Shifter |
Shifter problem. More... | |
class | SigmoidFitPlatt |
Optimizes the parameters of a sigmoid to fit a validation dataset via Platt's method. More... | |
class | SigmoidFitRpropNLL |
Optimizes the parameters of a sigmoid to fit a validation dataset via backpropagation on the negative log-likelihood. More... | |
class | SigmoidModel |
Standard sigmoid function. More... | |
class | SimpleNearestNeighbors |
Brute force optimized nearest neighbor implementation. More... | |
class | SimpleSigmoidModel |
Simple sigmoid function. More... | |
class | SimplexDownhill |
Simplex Downhill Method. More... | |
struct | SimulatedBinaryCrossover |
Simulated binary crossover operator. More... | |
class | SingleChainApproximator |
Approximates the gradient by taking samples from a single Markov chain. More... | |
struct | SingleObjectiveResultSet |
Result set for single objective algorithm. More... | |
class | SinglePole |
class | SMSEMOA |
Implements the SMS-EMOA. More... | |
class | SoftClusteringModel |
Model for "soft" clustering. More... | |
class | Softmax |
Softmax function. More... | |
class | SoftNearestNeighborClassifier |
SoftNearestNeighborClassifier returns a probabilistic classification by looking at the k nearest neighbors. More... | |
class | SparseAutoencoderError |
Error Function for Autoencoders and TiedAutoencoders which should be trained with sparse activation of the hidden neurons. More... | |
struct | Sphere |
Convex quadratic benchmark function. More... | |
class | SquaredEpsilonHingeLoss |
Hinge-loss for large margin regression using th squared two-norm. More... | |
class | SquaredHingeCSvmTrainer |
class | SquaredHingeLinearCSvmTrainer |
class | SquaredHingeLoss |
Squared Hinge-loss for large margin classification. More... | |
class | SquaredLoss |
squared loss for regression and classification More... | |
class | SquaredLoss< OutputType, unsigned int > |
class | SquaredLoss< Sequence, Sequence > |
struct | State |
Represents the State of an Object. More... | |
class | SteepestDescent |
Standard steepest descent. More... | |
class | SubrangeKernel |
Weighted sum of kernel functions. More... | |
class | SvmLogisticInterpretation |
Maximum-likelihood model selection score for binary support vector machines. More... | |
class | SvmProblem |
struct | SvmShrinkingProblem |
struct | TanhNeuron |
Neuron which computes the hyperbolic tangenst with range [-1,1]. More... | |
class | TanhSigmoidModel |
scaled Tanh sigmoid function More... | |
class | TemperedMarkovChain |
class | ThresholdConverter |
Convertion of real-valued outputs to classes 0 or 1. More... | |
class | ThresholdVectorConverter |
Convertion of real-vector outputs to vectors of class labels 0 or 1. More... | |
class | TiedAutoencoder |
implements the autoencoder with tied weights More... | |
class | Timer |
Timer abstraction with microsecond resolution. More... | |
struct | TournamentSelection |
Tournament selection operator. More... | |
class | TrainingError |
This stopping criterion tracks the improvement of the error function of the training error over an interval of iterations. More... | |
class | TrainingProgress |
This stopping criterion tracks the improvement of the training error over an interval of iterations. More... | |
struct | TransformedData |
class | TreeConstruction |
Stopping criteria for tree construction. More... | |
class | TreeNearestNeighbors |
Nearest Neighbors implementation using binary trees. More... | |
class | Truncate |
Transformation function truncating elements in a dataset. More... | |
class | TruncateAndRescale |
Transformation function first truncating and then rescaling elements in a dataset. More... | |
class | TruncatedExponential |
Implements a generator for the truncated exponential function. More... | |
class | TruncatedExponential_distribution |
boost random suitable distribution for an truncated exponential. See TruncatedExponential for more details. More... | |
class | TruncatedExponentialLayer |
A layer of truncated exponential neurons. More... | |
class | TrustRegionNewton |
Simple Trust-Region method based on the full Hessian matrix. More... | |
class | TukeyBiweightLoss |
Tukey's Biweight-loss for robust regression. More... | |
class | TwoNormRegularizer |
Two-norm of the input as an objective function. More... | |
class | TwoPointStepSizeAdaptation |
Step size adaptation based on the success of the new population compared to the old. More... | |
struct | TwoStateSpace |
The TwoStateSpace is a discrete Space with only two values, for example {0,1} or {-1,1}. More... | |
class | TypedFeatureNotAvailableException |
Exception indicating the attempt to use a feature which is not supported. More... | |
class | TypedFlags |
Flexible and extensible mechanisms for holding flags. More... | |
class | Uniform |
Implements a continuous uniform distribution. More... | |
class | UniformCrossover |
Uniform crossover of arbitrary individuals. More... | |
struct | UniformRankingSelection |
Selects individuals from the range of individual and offspring individuals. More... | |
class | UnlabeledData |
Data set for unsupervised learning. More... | |
struct | ValidatedSingleObjectiveResultSet |
Result set for validated points. More... | |
class | ValidatedStoppingCriterion |
Given the current Result set of the optimizer, calculates the validation error using a validation function and hands the results over to the underlying stopping criterion. More... | |
class | VDCMA |
struct | VectorMatrixTraits |
Template which finds for every Vector type the best fitting Matrix. More... | |
class | Wave |
Noisy sinc function: y = sin(x) / x + noise. More... | |
class | Weibull |
Weibull distributed random variable. More... | |
class | Weibull_distribution |
Weibull distribution. More... | |
class | WeightedLabeledData |
Weighted data set for supervised learning. More... | |
class | WeightedSumKernel |
Weighted sum of kernel functions. More... | |
class | WeightedUnlabeledData |
Weighted data set for unsupervised learning. More... | |
struct | WilcoxonRankSumTest |
Wilcoxon rank-sum test / Mann–Whitney U test. More... | |
struct | WS2MaximumGradientCriterion |
Working set selection by maximization of the projected gradient. More... | |
struct | ZDT1 |
Multi-objective optimization benchmark function ZDT1. More... | |
struct | ZDT2 |
Multi-objective optimization benchmark function ZDT2. More... | |
struct | ZDT3 |
Multi-objective optimization benchmark function ZDT3. More... | |
struct | ZDT4 |
Multi-objective optimization benchmark function ZDT4. More... | |
struct | ZDT6 |
Multi-objective optimization benchmark function ZDT6. More... | |
class | ZeroOneLoss |
0-1-loss for classification. More... | |
class | ZeroOneLoss< unsigned int, RealVector > |
0-1-loss for classification. More... | |
Enumerations | |
enum | AlphaStatus { AlphaFree = 0, AlphaLowerBound = 1, AlphaUpperBound = 2, AlphaDeactivated = 3 } |
enum | QpStopType { QpNone = 0, QpAccuracyReached = 1, QpMaxIterationsReached = 4, QpTimeout = 8 } |
enum | PartitionEstimationAlgorithm { AIS, AISMean, TwoSidedAISMean, AcceptanceRatio, AcceptanceRatioMean } |
Functions | |
SHARK_EXPORT_SYMBOL std::size_t | kMeans (Data< RealVector > const &data, std::size_t k, Centroids ¢roids, std::size_t maxIterations=0) |
The k-means clustering algorithm. More... | |
SHARK_EXPORT_SYMBOL std::size_t | kMeans (Data< RealVector > const &data, RBFLayer &model, std::size_t maxIterations=0) |
The k-means clustering algorithm for initializing an RBF Layer. More... | |
template<class InputType > | |
KernelExpansion< InputType > | kMeans (Data< InputType > const &dataset, std::size_t k, AbstractKernelFunction< InputType > &kernel, std::size_t maxIterations=0) |
template<class T > | |
T | maxExpInput () |
Maximum allowed input value for exp. More... | |
template<class T > | |
T | minExpInput () |
Minimum value for exp(x) allowed so that it is not 0. More... | |
template<class T > | |
boost::enable_if< boost::is_arithmetic< T >, T >::type | sqr (const T &x) |
Calculates x^2. More... | |
template<class T > | |
T | cube (const T &x) |
Calculates x^3. More... | |
template<class T > | |
boost::enable_if< boost::is_arithmetic< T >, T >::type | sigmoid (T x) |
Logistic function/logistic function. More... | |
template<class T > | |
T | safeExp (T x) |
Thresholded exp function, over- and underflow safe. More... | |
template<class T > | |
T | safeLog (T x) |
Thresholded log function, over- and underflow safe. More... | |
template<class T > | |
boost::enable_if< boost::is_arithmetic< T >, T >::type | softPlus (T x) |
Numerically stable version of the function log(1+exp(x)). More... | |
double | softPlus (double x) |
Numerically stable version of the function log(1+exp(x)). calculated with float precision to save some time. More... | |
template<class T > | |
T | copySign (T x, T y) |
double | trigamma (double x) |
Calculates the trigamma function -the derivative of the digamma function. More... | |
double | tetragamma (double x) |
Calculates the tetragamma function - the derivative of the trigamma function. More... | |
template<typename T , typename U > | |
ResultSet< T, U > | makeResultSet (T const &t, U const &u) |
Generates a typed solution given the search point and the corresponding objective function value. More... | |
template<class SearchPoint , class Result > | |
std::ostream & | operator<< (std::ostream &out, ResultSet< SearchPoint, Result > const &solution) |
template<class SearchPoint > | |
std::ostream & | operator<< (std::ostream &out, ValidatedSingleObjectiveResultSet< SearchPoint > const &solution) |
template<class RandomAccessIterator , class Rng > | |
void | partial_shuffle (RandomAccessIterator begin, RandomAccessIterator middle, RandomAccessIterator end, Rng &rng) |
random_shuffle algorithm which stops after acquiring the random subsequence for [begin,middle) More... | |
template<class Iterator , class Rng > | |
void | shuffle (Iterator begin, Iterator end, Rng &rng) |
random_shuffle algorithm which stops after acquiring the random subsequence for [begin,middle) More... | |
template<class RandomAccessIterator > | |
void | partial_shuffle (RandomAccessIterator begin, RandomAccessIterator middle, RandomAccessIterator end) |
random_shuffle algorithm which stops after acquiring the random subsequence for [begin,middle) More... | |
template<class Range > | |
boost::range_iterator< Range >::type | median_element (Range &range) |
Returns the iterator to the median element. after this call, the range is partially ordered. More... | |
template<class Range > | |
boost::range_iterator< Range >::type | median_element (Range const &rangeAdaptor) |
template<class Range > | |
boost::range_iterator< Range >::type | partitionEqually (Range &range) |
Partitions a range in two parts as equal in size as possible. More... | |
template<class Range > | |
boost::range_iterator< Range >::type | partitionEqually (Range const &rangeAdaptor) |
Partitions a range in two parts as equal in size as possible and returns it's result. More... | |
template<class K , class V > | |
void | swap (KeyValuePair< K, V > &pair1, KeyValuePair< K, V > &pair2) |
Swaps the contents of two instances of KeyValuePair. More... | |
template<class Key , class Value > | |
KeyValuePair< Key, Value > | makeKeyValuePair (Key const &key, Value const &value) |
Creates a KeyValuePair. More... | |
template<class Iterator1 , class Iterator2 > | |
KeyValueRange< Iterator1, Iterator2 > | zipKeyValuePairs (Iterator1 begin1, Iterator1 end1, Iterator2 begin2, Iterator2 end2) |
Zips two ranges together, interpreting the first range as Key which can be sorted. More... | |
template<class Range1 , class Range2 > | |
KeyValueRange< typename boost::range_iterator< Range1 >::type, typename boost::range_iterator< Range2 >::type > | zipKeyValuePairs (Range1 &range1, Range2 &range2) |
Zips two ranges together, interpreting the first range as Key which can be sorted. More... | |
template<class Range > | |
std::size_t | size (Range const &range) |
returns the size of a range More... | |
template<class Range > | |
boost::range_reference< Range >::type | get (Range &range, std::size_t i) |
returns the i-th element of a range More... | |
template<class Range > | |
boost::range_reference< Range const >::type | get (Range const &range, std::size_t i) |
template<class PairType , class Iterator1 , class Iterator2 > | |
boost::iterator_range< PairIterator< PairType, Iterator1, Iterator2 >> | zipPairRange (Iterator1 begin1, Iterator1 end1, Iterator2 begin2, Iterator2 end2) |
returns a paired zip range using pair type Pair This class must be specialized for every Pair to be used More... | |
template<class PairType , class Range1 , class Range2 > | |
PairRangeType< PairType, Range1, Range2 >::type | zipPairRange (Range1 &range1, Range2 &range2) |
returns a paired zip range using pair type Pair More... | |
template<class PairType , class Range1 , class Range2 > | |
PairRangeType< PairType, Range1 const, Range2 const >::type | zipPairRange (Range1 const &range1, Range2 const &range2) |
returns a paired zip range using pair type Pair More... | |
template<class T , class Range > | |
Batch< T >::type | createBatch (Range const &range) |
creates a batch from a range of inputs More... | |
template<class Range > | |
Batch< typename Range::value_type >::type | createBatch (Range const &range) |
template<class S > | |
S & | fusionize (detail::FusionFacade< S > &facade) |
template<class S > | |
S const & | fusionize (detail::FusionFacade< S > const &facade) |
template<class S > | |
boost::disable_if< detail::isFusionFacade< S >, S & >::type | fusionize (S &facade) |
template<class S > | |
boost::disable_if< detail::isFusionFacade< S >, S const &>::type | fusionize (S const &facade) |
template<class Range > | |
Data< typename boost::range_value< Range >::type > | createDataFromRange (Range const &inputs, std::size_t maximumBatchSize=0) |
creates a data object from a range of elements More... | |
template<class Range > | |
UnlabeledData< typename boost::range_value< Range >::type > | createUnlabeledDataFromRange (Range const &inputs, std::size_t maximumBatchSize=0) |
creates a data object from a range of elements More... | |
template<class Range1 , class Range2 > | |
LabeledData< typename boost::range_value< Range1 >::type, typename boost::range_value< Range2 >::type > | createLabeledDataFromRange (Range1 const &inputs, Range2 const &labels, std::size_t batchSize=0) |
creates a labeled data object from two ranges, representing inputs and labels More... | |
template<class T , class U > | |
std::ostream & | operator<< (std::ostream &stream, const LabeledData< T, U > &d) |
brief Outstream of elements for labeled data. More... | |
unsigned int | numberOfClasses (Data< unsigned int > const &labels) |
Return the number of classes of a set of class labels with unsigned int label encoding. More... | |
std::vector< std::size_t > | classSizes (Data< unsigned int > const &labels) |
Returns the number of members of each class in the dataset. More... | |
template<class InputType > | |
std::size_t | dataDimension (Data< InputType > const &dataset) |
Return the dimensionality of a dataset. More... | |
template<class InputType , class LabelType > | |
std::size_t | inputDimension (LabeledData< InputType, LabelType > const &dataset) |
Return the input dimensionality of a labeled dataset. More... | |
template<class InputType , class LabelType > | |
std::size_t | labelDimension (LabeledData< InputType, LabelType > const &dataset) |
Return the label/output dimensionality of a labeled dataset. More... | |
template<class InputType > | |
std::size_t | numberOfClasses (LabeledData< InputType, unsigned int > const &dataset) |
Return the number of classes (highest label value +1) of a classification dataset with unsigned int label encoding. More... | |
template<class InputType , class LabelType > | |
std::vector< std::size_t > | classSizes (LabeledData< InputType, LabelType > const &dataset) |
Returns the number of members of each class in the dataset. More... | |
template<class DatasetT > | |
DatasetT | indexedSubset (DatasetT const &dataset, typename DatasetT::IndexSet const &indices) |
template<class DatasetT > | |
DatasetT | rangeSubset (DatasetT const &dataset, std::size_t start, std::size_t end) |
Fill in the subset of batches [start,...,size+start[. More... | |
template<class DatasetT > | |
DatasetT | rangeSubset (DatasetT const &dataset, std::size_t size) |
Fill in the subset of batches [0,...,size[. More... | |
template<class T , class Functor > | |
boost::lazy_disable_if< CanBeCalled< Functor, typename Data< T >::batch_type >, TransformedData< Functor, T >>::type | transform (Data< T > const &data, Functor f) |
Transforms a dataset using a Functor f and returns the transformed result. More... | |
template<class T , class Functor > | |
boost::lazy_enable_if< CanBeCalled< Functor, typename Data< T >::batch_type >, TransformedData< Functor, T >>::type | transform (Data< T > const &data, Functor const &f) |
Transforms a dataset using a Functor f and returns the transformed result. More... | |
template<class I , class L , class Functor > | |
LabeledData< typename detail::TransformedDataElement< Functor, I >::type, L > | transformInputs (LabeledData< I, L > const &data, Functor const &f) |
Transforms the inputs of a dataset and return the transformed result. More... | |
template<class I , class L , class Functor > | |
LabeledData< I, typename detail::TransformedDataElement< Functor, L >::type > | transformLabels (LabeledData< I, L > const &data, Functor const &f) |
Transforms the labels of a dataset and returns the transformed result. More... | |
template<class FeatureSet > | |
Data< RealVector > | selectFeatures (Data< RealVector > const &data, FeatureSet const &features) |
Creates a copy o a dataset selecting only a certain set of features. More... | |
template<class T , class FeatureSet > | |
LabeledData< RealVector, T > | selectInputFeatures (LabeledData< RealVector, T > const &data, FeatureSet const &features) |
template<class DatasetT > | |
DatasetT | splitAtElement (DatasetT &data, std::size_t elementIndex) |
Removes the last part of a given dataset and returns a new split containing the removed elements. More... | |
template<class I > | |
void | repartitionByClass (LabeledData< I, unsigned int > &data, std::size_t batchSize=LabeledData< I, unsigned int >::DefaultBatchSize) |
reorders the dataset such, that points are grouped by labels More... | |
template<class I > | |
LabeledData< I, unsigned int > | binarySubProblem (LabeledData< I, unsigned int >const &data, unsigned int zeroClass, unsigned int oneClass) |
template<class I > | |
LabeledData< I, unsigned int > | oneVersusRestProblem (LabeledData< I, unsigned int >const &data, unsigned int oneClass) |
Construct a binary (two-class) one-versus-rest problem from a multi-class problem. More... | |
template<typename RowType > | |
RowType | getColumn (const Data< RowType > &data, std::size_t columnID) |
template<typename RowType > | |
void | setColumn (Data< RowType > &data, std::size_t columnID, RowType newColumn) |
template<class DatasetType , class IndexRange > | |
DataView< DatasetType > | subset (DataView< DatasetType > const &view, IndexRange const &indizes) |
creates a subset of a DataView with elements indexed by indices More... | |
template<class DatasetType > | |
DataView< DatasetType > | randomSubset (DataView< DatasetType > const &view, std::size_t size) |
creates a random subset of a DataView with given size More... | |
template<class DatasetType , class IndexRange > | |
DataView< DatasetType >::batch_type | subBatch (DataView< DatasetType > const &view, IndexRange const &indizes) |
Creates a batch given a set of indices. More... | |
template<class DatasetType > | |
DataView< DatasetType >::batch_type | randomSubBatch (DataView< DatasetType > const &view, std::size_t size) |
Creates a random batch of a given size. More... | |
template<class DatasetType > | |
DataView< DatasetType > | toView (DatasetType &set) |
Creates a View from a dataset. More... | |
template<class T > | |
DataView< T >::dataset_type | toDataset (DataView< T > const &view, std::size_t batchSize=DataView< T >::dataset_type::DefaultBatchSize) |
Creates a new dataset from a View. More... | |
template<class DatasetType > | |
std::size_t | numberOfClasses (DataView< DatasetType > const &view) |
template<class DatasetType > | |
std::size_t | inputDimension (DataView< DatasetType > const &view) |
Return the input dimensionality of the labeled dataset represented by the view. More... | |
template<class DatasetType > | |
std::size_t | labelDimension (DataView< DatasetType > const &view) |
Return the label dimensionality of the labeled dataset represented by the view. More... | |
template<class DatasetType > | |
std::size_t | dataDimension (DataView< DatasetType > const &view) |
Return the dimensionality of the dataset represented by the view. More... | |
template<typename VectorType > | |
void | importHDF5 (Data< VectorType > &data, const std::string &fileName, const std::string &datasetName) |
Import data from a HDF5 file. More... | |
template<typename VectorType , typename LabelType > | |
void | importHDF5 (LabeledData< VectorType, LabelType > &labeledData, const std::string &fileName, const std::string &data, const std::string &label) |
Import data to a LabeledData object from a HDF5 file. More... | |
template<typename VectorType > | |
void | importHDF5 (Data< VectorType > &data, const std::string &fileName, const std::vector< std::string > &cscDatasetName) |
Import data from HDF5 dataset of compressed sparse column format. More... | |
template<typename VectorType , typename LabelType > | |
void | importHDF5 (LabeledData< VectorType, LabelType > &labeledData, const std::string &fileName, const std::vector< std::string > &cscDatasetName, const std::string &label) |
Import data from HDF5 dataset of compressed sparse column format. More... | |
template<class T > | |
void | importPGM (std::string const &fileName, T &data, std::size_t &sx, std::size_t &sy) |
Import a PGM image from file. More... | |
template<class T > | |
void | exportPGM (std::string const &fileName, T const &data, std::size_t sx, std::size_t sy, bool normalize=false) |
Export a PGM image to file. More... | |
void | exportFiltersToPGMGrid (std::string const &basename, RealMatrix const &filters, std::size_t width, std::size_t height) |
Exports a set of filters as a grid image. More... | |
void | exportFiltersToPGMGrid (std::string const &basename, Data< RealVector > const &filters, std::size_t width, std::size_t height) |
Exports a set of filters as a grid image. More... | |
template<class T > | |
void | importPGMDir (const std::string &p, T &container, std::vector< ImageInformation > &info) |
Import PGM images scanning a directory recursively. More... | |
template<class T > | |
void | importPGMSet (const std::string &p, Data< T > &set, Data< ImageInformation > &setInfo) |
Import PGM images scanning a directory recursively. More... | |
template<class Vec1T , class Vec2T , class Vec3T > | |
void | meanvar (const Data< Vec1T > &data, blas::vector_container< Vec2T > &mean, blas::vector_container< Vec3T > &variance) |
Calculates the mean and variance values of the input data. More... | |
template<class Vec1T , class Vec2T , class MatT > | |
void | meanvar (const Data< Vec1T > &data, blas::vector_container< Vec2T > &mean, blas::matrix_container< MatT > &variance) |
Calculates the mean, variance and covariance values of the input data. More... | |
template<class VectorType > | |
VectorType | mean (Data< VectorType > const &data) |
Calculates the mean vector of the input vectors. More... | |
template<class VectorType > | |
VectorType | mean (UnlabeledData< VectorType > const &data) |
template<class VectorType > | |
VectorType | variance (const Data< VectorType > &data) |
Calculates the variance vector of the input vectors. More... | |
template<class VectorType > | |
VectorMatrixTraits< VectorType >::DenseMatrixType | covariance (const Data< VectorType > &data) |
Calculates the covariance matrix of the data vectors. More... | |
template<class T > | |
std::ostream & | operator<< (std::ostream &stream, const WeightedUnlabeledData< T > &d) |
brief Outstream of elements for weighted data. More... | |
template<class DataRange , class WeightRange > | |
boost::disable_if< boost::is_arithmetic< WeightRange >, WeightedUnlabeledData< typename boost::range_value< DataRange >::type > >::type | createUnlabeledDataFromRange (DataRange const &data, WeightRange const &weights, std::size_t batchSize=0) |
creates a weighted unweighted data object from two ranges, representing data and weights More... | |
template<class T , class U > | |
std::ostream & | operator<< (std::ostream &stream, const WeightedLabeledData< T, U > &d) |
brief Outstream of elements for weighted labeled data. More... | |
template<class InputType > | |
double | sumOfWeights (WeightedUnlabeledData< InputType > const &dataset) |
template<class InputType , class LabelType > | |
double | sumOfWeights (WeightedLabeledData< InputType, LabelType > const &dataset) |
std::size_t | numberOfClasses (WeightedUnlabeledData< unsigned int > const &labels) |
std::vector< std::size_t > | classSizes (WeightedUnlabeledData< unsigned int > const &labels) |
Returns the number of members of each class in the dataset. More... | |
template<class InputType > | |
std::size_t | dataDimension (WeightedUnlabeledData< InputType > const &dataset) |
Return the dimnsionality of points of a weighted dataset. More... | |
template<class InputType , class LabelType > | |
std::size_t | inputDimension (WeightedLabeledData< InputType, LabelType > const &dataset) |
Return the input dimensionality of a weighted labeled dataset. More... | |
template<class InputType , class LabelType > | |
std::size_t | labelDimension (WeightedLabeledData< InputType, LabelType > const &dataset) |
Return the label/output dimensionality of a labeled dataset. More... | |
template<class InputType > | |
std::size_t | numberOfClasses (WeightedLabeledData< InputType, unsigned int > const &dataset) |
Return the number of classes (highest label value +1) of a classification dataset with unsigned int label encoding. More... | |
template<class InputType , class LabelType > | |
std::vector< std::size_t > | classSizes (WeightedLabeledData< InputType, LabelType > const &dataset) |
Returns the number of members of each class in the dataset. More... | |
template<class InputRange , class LabelRange , class WeightRange > | |
boost::disable_if< boost::is_arithmetic< WeightRange >, WeightedLabeledData< typename boost::range_value< InputRange >::type, typename boost::range_value< LabelRange >::type >>::type | createLabeledDataFromRange (InputRange const &inputs, LabelRange const &labels, WeightRange const &weights, std::size_t batchSize=0) |
creates a weighted unweighted data object from two ranges, representing data and weights More... | |
template<class InputType , class LabelType > | |
WeightedLabeledData< InputType, LabelType > | bootstrap (LabeledData< InputType, LabelType > const &dataset, std::size_t bootStrapSize=0) |
Creates a bootstrap partition of a labeled dataset and returns it using weighting. More... | |
template<class InputType > | |
WeightedUnlabeledData< InputType > | bootstrap (UnlabeledData< InputType > const &dataset, std::size_t bootStrapSize=0) |
Creates a bootstrap partition of an unlabeled dataset and returns it using weighting. More... | |
SHARK_VECTOR_MATRIX_TYPEDEFS (long double, BigReal) | |
SHARK_VECTOR_MATRIX_TYPEDEFS (bool, Bool) | |
SHARK_VECTOR_MATRIX_ASSIGNMENT (BigReal) | |
template<class InputT , class IntermediateT , class OutputT > | |
detail::ConcatenatedModelWrapper< InputT, IntermediateT, OutputT > | operator>> (AbstractModel< InputT, IntermediateT > &firstModel, AbstractModel< IntermediateT, OutputT > &secondModel) |
Connects two AbstractModels so that the output of the first model is the input of the second. More... | |
template<class InputT , class IntermediateT , class OutputT > | |
detail::ConcatenatedModelList< InputT, IntermediateT, OutputT > | operator>> (const detail::ConcatenatedModelWrapperBase< InputT, IntermediateT > &firstModel, AbstractModel< IntermediateT, OutputT > &secondModel) |
Connects another AbstractModel two a previously created connection of models. More... | |
template<typename InputType , typename InputTypeT1 , typename InputTypeT2 > | |
double | evalSkipMissingFeatures (const AbstractKernelFunction< InputType > &kernelFunction, const InputTypeT1 &inputA, const InputTypeT2 &inputB) |
template<typename InputType , typename InputTypeT1 , typename InputTypeT2 , typename InputTypeT3 > | |
double | evalSkipMissingFeatures (const AbstractKernelFunction< InputType > &kernelFunction, const InputTypeT1 &inputA, const InputTypeT2 &inputB, InputTypeT3 const &missingness) |
template<class InputType , class M > | |
void | calculateRegularizedKernelMatrix (AbstractKernelFunction< InputType >const &kernel, Data< InputType > const &dataset, blas::matrix_expression< M > &matrix, double regularizer=0) |
Calculates the regularized kernel gram matrix of the points stored inside a dataset. More... | |
template<class InputType , class M > | |
void | calculateMixedKernelMatrix (AbstractKernelFunction< InputType >const &kernel, Data< InputType > const &dataset1, Data< InputType > const &dataset2, blas::matrix_expression< M > &matrix) |
Calculates the kernel gram matrix between two data sets. More... | |
template<class InputType > | |
RealMatrix | calculateRegularizedKernelMatrix (AbstractKernelFunction< InputType >const &kernel, Data< InputType > const &dataset, double regularizer=0) |
Calculates the regularized kernel gram matrix of the points stored inside a dataset. More... | |
template<class InputType > | |
RealMatrix | calculateMixedKernelMatrix (AbstractKernelFunction< InputType >const &kernel, Data< InputType > const &dataset1, Data< InputType > const &dataset2) |
Calculates the kernel gram matrix between two data sets. More... | |
template<class InputType , class WeightMatrix > | |
RealVector | calculateKernelMatrixParameterDerivative (AbstractKernelFunction< InputType > const &kernel, Data< InputType > const &dataset, WeightMatrix const &weights) |
Efficiently calculates the weighted derivative of a Kernel Gram Matrix w.r.t the Kernel Parameters. More... | |
template<class RngType > | |
bool | coinToss (RngType &rng, double pHeads) |
Flips a coin with probability of heads being pHeads by drawing random numbers from rng. More... | |
template<class RngType > | |
bool | coinToss (RngType &rng, unsigned int n, double pHeads) |
Returns a random number from Binomial(n,p) by drawing random numbers from rng. More... | |
template<typename RngType > | |
double | entropy (const Cauchy< RngType > &distribution) |
Returns the entropy of the Cauchy distribution. More... | |
template<class RngType > | |
std::size_t | discrete (RngType &rng, std::size_t low, std::size_t high) |
Draws a discrete number in {low,low+1,...,high} by drawing random numbers from rng. More... | |
template<typename Distribution > | |
double | entropy (Distribution &d, std::size_t trials=10000) |
Estimates the entropy of a distribution. The more trials the better is the estimate, but good estimates are slow. More... | |
ANNOUNCE_SHARK_RNG (shark::FastRngType, FastRng) | |
ANNOUNCE_SHARK_RNG (shark::DefaultRngType, Rng) | |
template<typename DistributionP , typename DistributionQ > | |
double | kullback_leiber_divergence (DistributionP &p, DistributionQ &q, std::size_t trials=10000) |
Estimates the kullback-leibler-divergence between two distributions. The more trials the better is the estimate, but good estimates are slow. More... | |
template<class RngType > | |
double | gauss (RngType &rng, double mean, double variance) |
Draws a number from the normal distribution with given mean and variance by drawing random numbers from rng. More... | |
template<class RngType > | |
double | uni (RngType &rng, double lower, double upper) |
Draws a number uniformly in [lower,upper] by drawing random numbers from rng. More... | |
template<typename Stream > | |
Stream & | operator<< (Stream &s, const WilcoxonRankSumTest::Element &element) |
template<class RBMType > | |
double | logPartitionFunction (RBMType const &rbm, double beta=1.0) |
Calculates the value of the partition function $Z$. More... | |
template<class RBMType > | |
double | negativeLogLikelihoodFromLogPartition (RBMType const &rbm, UnlabeledData< RealVector > const &inputs, double logPartition, double beta=1.0) |
Estimates the negative log-likelihood of a set of input vectors under the models distribution using the partition function. More... | |
template<class RBMType > | |
double | negativeLogLikelihood (RBMType const &rbm, UnlabeledData< RealVector > const &inputs, double beta=1.0) |
Estimates the negative log-likelihood of a set of input vectors under the models distribution. More... | |
double | estimateLogFreeEnergyFromEnergySamples (RealMatrix const &energyDiffUp, RealMatrix const &energyDiffDown, PartitionEstimationAlgorithm algorithm=AIS) |
template<class RBMType > | |
double | estimateLogFreeEnergy (RBMType &rbm, Data< RealVector > const &initDataset, RealVector const &beta, std::size_t samples, PartitionEstimationAlgorithm algorithm=AcceptanceRatioMean, float burnInPercentage=0.1) |
template<class RBMType > | |
double | annealedImportanceSampling (RBMType &rbm, RealVector const &beta, std::size_t samples) |
template<class RBMType > | |
double | estimateLogFreeEnergy (RBMType &rbm, Data< RealVector > const &initDataset, std::size_t chains, std::size_t samples, PartitionEstimationAlgorithm algorithm=AIS, float burnInPercentage=0.1) |
template<class RBMType > | |
double | annealedImportanceSampling (RBMType &rbm, std::size_t chains, std::size_t samples) |
template<typename Stream > | |
FrontType | read_front (Stream &in, std::size_t noObjectives, const std::string &separator=" ", std::size_t headerLines=0) |
bool | compare_fitness (const IndividualType &a, const IndividualType &b) |
template<class I , class L > | |
CVFolds< LabeledData< I, L > > | createCVIID (LabeledData< I, L > &set, std::size_t numberOfPartitions, std::size_t batchSize=Data< I >::DefaultBatchSize) |
Create a partition for cross validation. More... | |
template<class I , class L > | |
CVFolds< LabeledData< I, L > > | createCVSameSize (LabeledData< I, L > &set, std::size_t numberOfPartitions, std::size_t batchSize=LabeledData< I, L >::DefaultBatchSize) |
Create a partition for cross validation. More... | |
template<class I > | |
CVFolds< LabeledData< I, unsigned int > > | createCVSameSizeBalanced (LabeledData< I, unsigned int > &set, std::size_t numberOfPartitions, std::size_t batchSize=Data< I >::DefaultBatchSize, RecreationIndices *cv_indices=NULL) |
Create a partition for cross validation. More... | |
template<class I , class L > | |
CVFolds< LabeledData< I, L > > | createCVBatch (LabeledData< I, L > const &set, std::size_t numberOfPartitions) |
Create a partition for cross validation without changing the dataset. More... | |
template<class I , class L > | |
CVFolds< LabeledData< I, L > > | createCVIndexed (LabeledData< I, L > &set, std::size_t numberOfPartitions, std::vector< std::size_t > indices, std::size_t batchSize=Data< I >::DefaultBatchSize) |
Create a partition for cross validation from indices. More... | |
template<class I , class L > | |
CVFolds< LabeledData< I, L > > | createCVFullyIndexed (LabeledData< I, L > &set, std::size_t numberOfPartitions, RecreationIndices indices, std::size_t batchSize=Data< I >::DefaultBatchSize) |
Create a partition for cross validation from indices for both ordering and partitioning. More... | |
template<class T > | |
std::ostream & | operator<< (std::ostream &stream, const Data< T > &d) |
Outstream of elements. More... | |
void | import_libsvm (LabeledData< RealVector, unsigned int > &dataset, std::istream &stream, unsigned int highestIndex=0, std::size_t batchSize=LabeledData< RealVector, unsigned int >::DefaultBatchSize) |
Import data from a LIBSVM file. More... | |
void | import_libsvm (LabeledData< CompressedRealVector, unsigned int > &dataset, std::istream &stream, unsigned int highestIndex=0, std::size_t batchSize=LabeledData< RealVector, unsigned int >::DefaultBatchSize) |
Import data from a LIBSVM file. More... | |
void | import_libsvm (LabeledData< RealVector, unsigned int > &dataset, std::string fn, unsigned int highestIndex=0, std::size_t batchSize=LabeledData< RealVector, unsigned int >::DefaultBatchSize) |
Import data from a LIBSVM file. More... | |
void | import_libsvm (LabeledData< CompressedRealVector, unsigned int > &dataset, std::string fn, unsigned int highestIndex=0, std::size_t batchSize=LabeledData< RealVector, unsigned int >::DefaultBatchSize) |
Import data from a LIBSVM file. More... | |
template<typename InputType > | |
void | export_libsvm (LabeledData< InputType, unsigned int > &dataset, const std::string &fn, bool dense=false, bool oneMinusOne=true, bool sortLabels=false, bool append=false) |
Export data to LIBSVM format. More... | |
SHARK_EXPORT_SYMBOL void | importSparseData (LabeledData< RealVector, unsigned int > &dataset, std::istream &stream, unsigned int highestIndex=0, std::size_t batchSize=LabeledData< RealVector, unsigned int >::DefaultBatchSize) |
Import data from a sparse data (libSVM) file. More... | |
SHARK_EXPORT_SYMBOL void | importSparseData (LabeledData< CompressedRealVector, unsigned int > &dataset, std::istream &stream, unsigned int highestIndex=0, std::size_t batchSize=LabeledData< RealVector, unsigned int >::DefaultBatchSize) |
Import data from a sparse data (libSVM) file. More... | |
SHARK_EXPORT_SYMBOL void | importSparseData (LabeledData< RealVector, unsigned int > &dataset, std::string fn, unsigned int highestIndex=0, std::size_t batchSize=LabeledData< RealVector, unsigned int >::DefaultBatchSize) |
Import data from a sparse data (libSVM) file. More... | |
SHARK_EXPORT_SYMBOL void | importSparseData (LabeledData< CompressedRealVector, unsigned int > &dataset, std::string fn, unsigned int highestIndex=0, std::size_t batchSize=LabeledData< RealVector, unsigned int >::DefaultBatchSize) |
Import data from a sparse data (libSVM) file. More... | |
template<typename InputType > | |
void | exportSparseData (LabeledData< InputType, unsigned int > &dataset, const std::string &fn, bool dense=false, bool oneMinusOne=true, bool sortLabels=false, bool append=false) |
Export data to sparse data (libSVM) format. More... | |
template<class InputType , class OutputType > | |
void | initRandomNormal (AbstractModel< InputType, OutputType > &model, double s) |
Initialize model parameters normally distributed. More... | |
template<class InputType , class OutputType > | |
void | initRandomUniform (AbstractModel< InputType, OutputType > &model, double l, double h) |
Initialize model parameters uniformly at random. More... | |
Variables | |
static const double | SQRT_2_PI = boost::math::constants::root_two_pi<double>() |
Constant for sqrt( 2 * pi ). More... | |
enum | LabelPosition { FIRST_COLUMN, LAST_COLUMN } |
Position of the label in a CSV file. More... | |
SHARK_EXPORT_SYMBOL void | csvStringToData (Data< FloatVector > &data, std::string const &contents, char separator=',', char comment='#', std::size_t maximumBatchSize=Data< RealVector >::DefaultBatchSize) |
Import unlabeled vectors from a read-in character-separated value file. More... | |
SHARK_EXPORT_SYMBOL void | csvStringToData (Data< RealVector > &data, std::string const &contents, char separator=',', char comment='#', std::size_t maximumBatchSize=Data< RealVector >::DefaultBatchSize) |
Import unlabeled vectors from a read-in character-separated value file. More... | |
SHARK_EXPORT_SYMBOL void | csvStringToData (Data< unsigned int > &data, std::string const &contents, char separator=',', char comment='#', std::size_t maximumBatchSize=Data< unsigned int >::DefaultBatchSize) |
Import "csv" from string consisting only of a single unsigned int per row. More... | |
SHARK_EXPORT_SYMBOL void | csvStringToData (Data< int > &data, std::string const &contents, char separator=',', char comment='#', std::size_t maximumBatchSize=Data< int >::DefaultBatchSize) |
Import "csv" from string consisting only of a single int per row. More... | |
SHARK_EXPORT_SYMBOL void | csvStringToData (Data< float > &data, std::string const &contents, char separator=',', char comment='#', std::size_t maximumBatchSize=Data< double >::DefaultBatchSize) |
Import "csv" from string consisting only of a single double per row. More... | |
SHARK_EXPORT_SYMBOL void | csvStringToData (Data< double > &data, std::string const &contents, char separator=',', char comment='#', std::size_t maximumBatchSize=Data< double >::DefaultBatchSize) |
Import "csv" from string consisting only of a single double per row. More... | |
SHARK_EXPORT_SYMBOL void | csvStringToData (LabeledData< RealVector, unsigned int > &dataset, std::string const &contents, LabelPosition lp, char separator=',', char comment='#', std::size_t maximumBatchSize=LabeledData< RealVector, unsigned int >::DefaultBatchSize) |
Import labeled data from a character-separated value file. More... | |
SHARK_EXPORT_SYMBOL void | csvStringToData (LabeledData< FloatVector, unsigned int > &dataset, std::string const &contents, LabelPosition lp, char separator=',', char comment='#', std::size_t maximumBatchSize=LabeledData< RealVector, unsigned int >::DefaultBatchSize) |
Import labeled data from a character-separated value file. More... | |
SHARK_EXPORT_SYMBOL void | csvStringToData (LabeledData< RealVector, RealVector > &dataset, std::string const &contents, LabelPosition lp, std::size_t numberOfOutputs=1, char separator=',', char comment='#', std::size_t maximumBatchSize=LabeledData< RealVector, RealVector >::DefaultBatchSize) |
Import regression data from a read-in character-separated value file. More... | |
SHARK_EXPORT_SYMBOL void | csvStringToData (LabeledData< FloatVector, FloatVector > &dataset, std::string const &contents, LabelPosition lp, std::size_t numberOfOutputs=1, char separator=',', char comment='#', std::size_t maximumBatchSize=LabeledData< RealVector, RealVector >::DefaultBatchSize) |
Import regression data from a read-in character-separated value file. More... | |
template<class T > | |
void | importCSV (Data< T > &data, std::string fn, char separator=',', char comment='#', std::size_t maximumBatchSize=Data< T >::DefaultBatchSize, std::size_t titleLines=0) |
Import a Dataset from a csv file. More... | |
template<class T > | |
void | importCSV (LabeledData< blas::vector< T >, unsigned int > &data, std::string fn, LabelPosition lp, char separator=',', char comment='#', std::size_t maximumBatchSize=LabeledData< RealVector, unsigned int >::DefaultBatchSize) |
Import a labeled Dataset from a csv file. More... | |
template<class T > | |
void | importCSV (LabeledData< blas::vector< T >, blas::vector< T > > &data, std::string fn, LabelPosition lp, std::size_t numberOfOutputs=1, char separator=',', char comment='#', std::size_t maximumBatchSize=LabeledData< RealVector, RealVector >::DefaultBatchSize) |
Import a labeled Dataset from a csv file. More... | |
template<typename Type > | |
void | exportCSV (Data< Type > const &set, std::string fn, char separator=',', bool sci=true, unsigned int width=0) |
Format unlabeled data into a character-separated value file. More... | |
template<typename InputType , typename LabelType > | |
void | exportCSV (LabeledData< InputType, LabelType > const &dataset, std::string fn, LabelPosition lp, char separator=',', bool sci=true, unsigned int width=0) |
Format labeled data into a character-separated value file. More... | |
enum | KernelMatrixNormalizationType { NONE, MULTIPLICATIVE_TRACE_ONE, MULTIPLICATIVE_TRACE_N, MULTIPLICATIVE_VARIANCE_ONE, CENTER_ONLY, CENTER_AND_MULTIPLICATIVE_TRACE_ONE } |
template<typename InputType , typename LabelType > | |
void | exportKernelMatrix (LabeledData< InputType, LabelType > const &dataset, AbstractKernelFunction< InputType > &kernel, std::ostream &out, KernelMatrixNormalizationType normalizer=NONE, bool scientific=false, unsigned int fieldwidth=0) |
Write a kernel Gram matrix to stream. More... | |
template<typename InputType , typename LabelType > | |
void | exportKernelMatrix (LabeledData< InputType, LabelType > const &dataset, AbstractKernelFunction< InputType > &kernel, std::string fn, KernelMatrixNormalizationType normalizer=NONE, bool sci=false, unsigned int width=0) |
Write a kernel Gram matrix to file. More... | |
template<typename InputType , typename LabelType > | |
void | export_kernel_matrix (LabeledData< InputType, LabelType > const &dataset, AbstractKernelFunction< InputType > &kernel, std::ostream &out, KernelMatrixNormalizationType normalizer=NONE, bool scientific=false, unsigned int fieldwidth=0) |
template<typename InputType , typename LabelType > | |
void | export_kernel_matrix (LabeledData< InputType, LabelType > const &dataset, AbstractKernelFunction< InputType > &kernel, std::string fn, KernelMatrixNormalizationType normalizer=NONE, bool sci=false, unsigned int width=0) |
AbstractMultiObjectiveOptimizer.
Implements Block Gibbs Sampling.
Implements the Shifter benchmark problem.
Loads the MNIST benchmark problem.
Implements the DistantModes/ArtificialModes benchmark problem.
Implements the Bars & Stripes benchmark problem.
Implements the bipolar (-1,1) state neuron layer.
Typedefs for the Bipolar RBM.
Typedefs for the Binary-Binary RBM.
Wilcoxon Ranksum test implementation.
Calculate statistics given a range of values.
Implements a multi-variate normal distribution with zero mean.
Implements a multinomial distribution.
Implements a Weibull distribution.
Implements a uniform distribution.
Implements a truncated exponential.
Basic types and definitions of the Rng component.
Implements a poisson distribution.
Implements a univariate normal distribution.
Implements the negative exponential distribution.
Provides a log-normal distribution.
Provides a function for estimating the kullback-leibler-divergence.
Hypergeometric distribution.
This class subsumes several often used random number generators.
Implements a geometric distribution.
Implements a Gamma distribution.
Implements an Erlang distribution.
Provides a function for estimating the entropy of a distribution.
Discrete Uniform distribution.
Implements a dirichlet distribution.
Diff geometric distribution.
Standard Cauchy distribution.
Implements a binomial distribution.
Implements a Bernoulli distribution.
Abstract class for statistical distributions.
Maximum-likelihood model selection for binary support vector machines.
Specific error function for Feed-Forward-Networks which enforces it to have sparse hidden neuron activation.
ROC.
Regularizer.
Radius Margin Quotient for SVM model selection.
implements an error fucntion which only uses a random portion of the data for training
Negative Log Likelihood error function.
Evidence for model selection of a regularization network/Gaussian process.
Functions for measuring the area under the (ROC) curve.
Error measure for classication tasks, typically used for evaluation of results.
Implements Tukey's Biweight-loss function for robust regression.
Implements the Squared Error Loss function for regression.
Implements the Squared Hinge Loss function for maximum margin classification.
Implements the squard Hinge Loss function for maximum margin regression.
Implements the Huber loss function for robust regression.
Implements the Hinge Loss function for maximum margin classification.
Implements the Hinge Loss function for maximum margin regression.
Flexible error measure for classication tasks.
Error measure for classification tasks of non exclusive attributes that can be used for model training.
Error measure for classification tasks that can be used as the objective function for training.
super class of all loss functions
implements the absolute loss, which is the distance between labels and predictions
Leave-one-out error for C-SVMs.
Leave-one-out error.
Kernel Target Alignment - a measure of alignment of a kernel Gram matrix with labels.
Archive of evaluated points as an objective function wrapper.
error function for supervised learning
cross-validation error for selection of hyper-parameters
Multi-objective optimization benchmark function ZDT6.
Multi-objective optimization benchmark function ZDT4.
Multi-objective optimization benchmark function ZDT3.
Multi-objective optimization benchmark function ZDT2.
Multi-objective optimization benchmark function ZDT1.
Convex benchmark function.
Implements a wrapper over an m_objective function which just rotates its inputs.
Generalized Rosenbrock benchmark function.
Pole balancing simulation for double pole.
Pole balancing simulation for double poles.
Objective function for single and double poles with partial state information (non-Markovian task)
Objective function for single and double poles with full state information (Markovian task)
Multi-objective optimization benchmark function LZ9.
Multi-objective optimization benchmark function LZ8.
Multi-objective optimization benchmark function LZ7.
Multi-objective optimization benchmark function LZ6.
Multi-objective optimization benchmark function LZ5.
Multi-objective optimization benchmark function LZ4.
Multi-objective optimization benchmark function LZ3.
Multi-objective optimization benchmark function LZ2.
Multi-objective optimization benchmark function LZ1.
Multi-objective optimization benchmark function IHR 6.
Multi-objective optimization benchmark function IHR 4.
Multi-objective optimization benchmark function IHR 3.
Multi-objective optimization benchmark function IHR 2.
Multi-objective optimization benchmark function IHR 1.
Two-dimensional, real-valued Himmelblau function.
GSP benchmark function for multiobjective optimization.
Objective function for double non-Markov pole with special fitness functions.
Bi-objective real-valued benchmark function proposed by Fonseca and Flemming.
Multi-objective optimization benchmark function ELLI 2.
Multi-objective optimization benchmark function ELLI 1.
Objective function DTLZ7.
Objective function DTLZ6.
Objective function DTLZ5.
Objective function DTLZ4.
Objective function DTLZ3.
Objective function DTLZ2.
Objective function DTLZ1.
Multi-objective optimization benchmark function CIGTAB 2.
Multi-objective optimization benchmark function CIGTAB 1.
Convex quadratic benchmark function.
Convex quadratic benchmark function with single dominant axis.
cost function for quantitative judgement of deviations of predictions from target values
Base class for constraints.
Random Forest Classifier.
Tree for nearest neighbor search in data with low embedding dimension.
Tree for nearest neighbor search in kernel-induced feature spaces.
Tree for nearest neighbor search in low dimensions.
Cart Classifier.
Binary space-partitioning tree of data points.
Implements the autoencoder with tied weights.
Soft/probabilistic nearest neighbor classifier for vector-valued data.
Soft-max transformation.
Implements a simple sigmoidal model for sigmoidal fitting in a 2-d problem.
Offers a basic structure for recurrent networks.
Implements a radial basis function layer.
Offers the functions to create and to work with a recurrent neural network.
One-versus-one Classifier.
Model for scaling and translation of data vectors.
Nearest neighbor regression.
Nearest neighbor classification.
Implementation of Naive Bayes classifier.
Implements the Mean Model that can be used for ensemble classifiers.
Implements the linear norm class.
Implements the Linear Classifier class of the shark library.
Weighted sum of m_base kernels.
Variant of WeightedSumKernel which works on subranges of Vector inputs.
A kernel function that wraps a member kernel and multiplies it by a scalar.
Product of kernel functions.
Polynomial kernel.
Normalization of a kernel function.
Special kernel classes for multi-task and transfer learning.
monomial (polynomial) kernel
Weighted sum of base kernels, each acting on a subset of features only.
A kernel expansion with support of missing features.
linear kernel (standard inner product)
Collection of functions dealing with typical tasks of kernels.
Affine linear kernel function expansion.
Radial Gaussian kernel.
Do special kernel evaluation by skipping missing features.
Kernel on a finite, discrete space.
Derivative of a C-SVM hypothesis w.r.t. its hyperparameters.
Gaussian automatic relevance detection (ARD) kernel.
abstract super class of all metrics
abstract super class of all kernel functions
Implements a Model which corrupts the input by setting values of the input to a given value.
Implements a Feef-Forward multilayer perceptron.
Implements a Model using a linear function.
Format conversion models.
concatenation of two models, with type erasure
Model for "soft" clustering.
Hierarchical Clustering.
Model for "hard" clustering.
Super class for clustering models.
Clusters defined by centroids.
Super class for clustering definitions.
Implements the autoencoder.
base class for all models, as well as a specialized differentiable model
Helper functions for linear algebra component.
Algorithm for Singular Value Decomposition.
Some operations for matrices.
Algorithm for Triangular-Quadratic-Decomposition.
Some operations for creating rotation matrices.
Kernel Gram matrix with modified diagonal.
Partly Precomputed version of a matrix for quadratic programming.
Modified Kernel Gram matrix.
Helper functions to calculate several norms and distances.
Cache implementing an Least-Recently-Used Strategy.
Kernel Gram matrix.
Easy initialization of vectors.
Efficient special case if the kernel is gaussian and the inputs are sparse vectors.
Kernel matrix which supports kernel evaluations on data with missing features.
Algorithms for Eigenvalue decompositions.
Cholesky Decompositions for a Matrix A = LL^T.
Efficient quadratic matrix cache.
Precomputed version of a matrix for quadratic programming.
Sparse vector class.
Vector proxy classes.
expression templates for vector valued math
Implements the Dense vector class.
Implements a matrix with triangular storage layout.
Permutations of vectors and matrices.
Some special matrix-products.
Sparse matrix class.
Implements a set of matrices.
Matrix proxy classes.
Expression templates for expressions involving matrices.
Dense Matrix class.
Implements the lu decomposiion algorithm.
Assignment kernels for vector expressions.
Triangular solve kernel for vector expressions.
Triangular solve kernel for matrix epressions.
Triangular matrix-vector multiplication kernel.
Triangular Matrix-Matrix multiplication kernel.
Traits to obtain memory and storage sizes easily from expression templates.
Triangular packed matrix-vector multiplication.
Symmetric eigenvalue decomposition.
Dispatches the POTRF algorithm.
Kernels for matrix-expression assignments.
matrix-vector multiplication kernel
matrix-matrix multiplication kernel
dot-product of vectors
Implements the triangular packed matrix-vector multiplication.
Contains the lapack bindings for the symmetric eigenvalue problem syev.
Implements the default implementation of the POTRF algorithm.
Input and output of vectors and matrices.
forward declarations
Defines the basic types of CRTP base-classes.
Assignment and evaluation of vector expressions.
Entry Point for all Basic Linear Algebra(BLAS) in shark.
Weighted data sets for (un-)supervised learning.
some functions for vector valued statistics like mean, variance and covariance
Support for importing and exporting data from and to sparse data (libSVM) formatted data files.
Importing and exporting PGM images.
Deprecated import_libsvm and export_libsvm functions.
This will relabel a given dataset to have labels 0..N-1 (and vice versa)
Support for importing data from HDF5 file.
export precomputed kernel matrices (using libsvm format)
Fast lookup for elements in constant datasets.
Data for (un-)supervised learning.
Learning problems given by analytic distributions.
Tools for cross-validation.
Support for importing and exporting data from and to character separated value (CSV) files.
Defines an batch adptor for structures.
Defines the Batch Interface for a type, e.g., for every type a container with optimal structure.
Range which zips two ranges together while allowing a custom pair type.
A scoped_ptr like container for C type handles.
Helper Methods to use with boost range.
Provides a pair of Key and Value, as well as functions working with them.
Small Iterator collection.
Small General algorithm collection.
Template class checking whether for a functor F and Argument U, F(U) can be called.
Traits which allow to define ProxyReferences for types.
Timer abstraction with microsecond resolution.
Class which externalizes the state of an Object.
Result sets for algorithms.
Very basic math abstraction layer.
ISerializable interface.
IParameterizable interface.
INameable interface.
Flexible and extensible mechanisms for holding flags.
Optimization of the SigmoidModel according to Platt, 1999.
Random Forest Trainer.
Trainer for a Regularization Network or a Gaussian Process.
Principal Component Analysis.
Model training by means of a general purpose optimization procedure.
Trainer for One-Class Support Vector Machines.
Determine the scaling factor of a ScaledKernel so that it has unit variance in feature space one on a given dataset.
Data normalization to zero mean, unit variance and zero covariance while keping the original coordinate system.
Data normalization to zero mean, unit variance and zero covariance.
Data normalization to zero mean and unit variance.
Data normalization to the unit interval.
Trainer of Naive Bayes classifier.
Trainer for binary SVMs natively supporting missing features.
Trainer for the Multi-class Support Vector Machine by Weston and Watkins.
Trainer for One-versus-all (one-versus-rest) Multi-class Support Vector Machines.
Trainer for the Maximum Margin Regression Multi-class Support Vector Machine.
Trainer for the Multi-class Support Vector Machine by Lee, Lin, and Wahba.
Trainer for the Multi-class Support Vector Machine by Crammer and Singer.
Trainer for the ATS Multi-class Support Vector Machine.
Trainer for the ATM Multi-class Support Vector Machine.
Trainer for the ADM Multi-class Support Vector Machine.
Trainer for the Reinforced Multi-class Support Vector Machine.
Linear Regression.
LDA.
LASSO Regression.
Generic stochastic gradient descent training for kernel-based models.
Trainer for the Epsilon-Support Vector Machine for Regression.
Trainer for normal distribution.
Implementations of various distribution trainers.
Container for known distribution trainers.
Support Vector Machine Trainer for the standard C-SVM.
CART.
Remove budget maintenance strategy.
Project budget maintenance strategy.
Merge budget maintenance strategy.
Budgeted stochastic gradient descent training for kernel-based models.
Abstract Budget maintenance strategy.
Abstract Trainer Interface for trainers that support weighting.
Abstract Trainer Interface.
Abstract Support Vector Machine Trainer, general and linear case.
Stopping Criterion which evaluates the validation error and hands the result over to another stopping criterion.
Stopping Criterion which stops, when the training error seems to converge.
Stopping Criterion which stops, when the trainign error seems to converge.
Stopping Criterion which stops after a fixed number of iterations.
Stopping criterion monitoring the quotient of generalization loss and training progress.
Stopping Criterion which stops, when the generalization of the solution gets worse.
Defines a base class for stopping criteria of optimization algorithms.
Quadratic programming for Support Vector Machines.
Special container for certain coefficients describing multi-class SVMs.
General and specialized quadratic program classes and a generic solver.
Quadratic programming solvers for linear multi-class SVM training without bias.
Quadratic programming solver for multi-class SVMs.
Quadratic programming m_problem for multi-class SVMs.
Quadratic programming solver linear SVM training without bias.
Quadratic program definitions.
Pegasos solvers for linear SVMs.
Efficient Nearest neighbor queries.
Efficient brute force implementation of nearest neighbors.
Interface for nearest Neighbor queries.
The k-means clustering algorithm.
Jaakkola's heuristic and related quantities for Gaussian kernel selection.
Trust-Region Newton-Step Method.
implements different versions of Resilient Backpropagation of error.
CG.
BFGS.
Base class for Line Search Optimizer.
Implements the VD-CMA-ES Algorithm.
SteadyStateMOCMA.h.
Implements the SMS-EMOA.
Nelder-Mead Simplex Downhill Method.
IndicatorBasedRealCodedNSGAII.h.
Implementation of the Pareto-Dominance relation.
Implements a two point step size adaptation rule based on a line-search type of approach.
Roulette-Wheel-Selection using uniform selection probability assignment.
Implements tournament selection.
Implements fitness proportional selection.
Roulette-Wheel-Selection based on fitness-rank-based selection probability assignment.
Indicator-based selection strategy for multi-objective selection.
EP-Tournament selection operator.
Elitist Selection Operator suitable for (mu,lambda) and (mu+lambda) selection.
Uniform crossover of arbitrary individuals.
Simulated binary crossover operator.
Implements one-point crossover operator.
Implements the tep size adaptation based on the success of the new population compared to the old.
Polynomial mutation operator.
Bit flip mutation operator.
Implements the generational Multi-objective Covariance Matrix Adapation ES.
Approximately determines the individual contributing the least hypervolume.
Inverted generational distance for comparing Pareto-front approximations.
Calculates the hypervolume covered by a front of non-dominated points.
Calculates the additive approximation quality of a Pareto-front approximation.
Implementation of the exact hypervolume calculation in m dimensions.
Determine the volume of the union of objects by an FPRAS.
GridSearch.h.
Explicit traits for extracting fitness values from arbitrary types.
Implements the fast nondominated sort algorithm for pareto front calculation.
Implements the most recent version of the elitist CMA-ES.
Implements the Cross Entropy Algorithm.
Implements the CMSA.
Implements the most recent version of the non-elitist CMA-ES.
TypedIndividual.
CMAChromosomeof the CMA-ES.
AbstractSingleObjectiveOptimizer.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
Hansen, N. The CMA Evolution Startegy: A Tutorial, June 28, 2011 and the eqation numbers refer to this publication (retrieved April 2014).
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
The algorithm is described in
H. G. Beyer, B. Sendhoff (2008). Covariance Matrix Adaptation Revisited: The CMSA Evolution Strategy In Proceedings of the Tenth International Conference on Parallel Problem Solving from Nature (PPSN X), pp. 123-132, LNCS, Springer-Verlag
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
Christophe Thiery, Bruno Scherrer. Improvements on Learning Tetris with Cross Entropy. International Computer Games Association Journal, ICGA, 2009, 32. <inria-00418930>
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
The algorithm is based on
C. Igel, T. Suttorp, and N. Hansen. A Computational Efficient Covariance Matrix Update and a (1+1)-CMA for Evolution Strategies. In Proceedings of the Genetic and Evolutionary Computation Conference (GECCO 2006), pp. 453-460, ACM Press, 2006
D. V. Arnold and N. Hansen: Active covariance matrix adaptation for the (1+1)-CMA-ES. In Proceedings of the Genetic and Evolutionary Computation Conference (GECCO 2010): pp 385-392, ACM Press 2010
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
The algorithm is described in
Nicola Beume und Günter Rudolph. Faster S-Metric Calculation by Considering Dominated Hypervolume as Klee's Measure Problem. In: B. Kovalerchuk (ed.): Proceedings of the Second IASTED Conference on Computational Intelligence (CI 2006), pp. 231-236. ACTA Press: Anaheim, 2006.
A specialized algorithm is used for the three-objective case:
M. T. M. Emmerich and C. M. Fonseca. Computing hypervolume contributions in low dimensions: Asymptotically optimal algorithm and complexity results. In: Evolutionary Multi-Criterion Optimization (EMO) 2011. Vol. 6576 of Lecture Notes in Computer Science, pp. 121–135, Berlin: Springer, 2011.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
The algorithm is described in: James E. Baker. Adaptive Selection Methods for Genetic Algorithms. In John J. Grefenstette (ed.): Proceedings of the 1st International Conference on Genetic Algorithms (ICGA), pp. 101-111, Lawrence Erlbaum Associates, 1985
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
See http://en.wikipedia.org/wiki/Fitness_proportionate_selection
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
See http://en.wikipedia.org/wiki/Tournament_selection
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
The function is described in
Christian Igel, Nikolaus Hansen, and Stefan Roth. Covariance Matrix Adaptation for Multi-objective Optimization. Evolutionary Computation 15(1), pp. 1-28, 2007
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
See Nicola Beume, Boris Naujoks, and Michael Emmerich. SMS-EMOA: Multiobjective selection based on dominated hypervolume. European Journal of Operational Research, 181(3):1653-1669, 2007.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.Implements the VD-CMA-ES Algorithm
The VD-CMA-ES implements a restricted form of the CMA-ES where the covariance matrix is restriced to be (D+vv^T) where D is a diagonal matrix and v a single vector. Therefore this variant is capable of large-scale optimisation
For more reference, see the paper Akimoto, Y., A. Auger, and N. Hansen (2014). Comparison-Based Natural Gradient Optimization in High Dimension. To appear in Genetic and Evolutionary Computation Conference (GECCO 2014), Proceedings, ACM
The implementation differs from the paper to be closer to the reference implementation and to have better numerical accuracy.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
The Broyden, Fletcher, Goldfarb, Shannon (BFGS) algorithm is a quasi-Newton method for unconstrained real-valued optimization.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
Conjugate-gradient method for unconstraint optimization.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
The Limited-Memory Broyden, Fletcher, Goldfarb, Shannon (BFGS) algorithm is a quasi-Newton method for unconstrained real-valued optimization. See: http://en.wikipedia.org/wiki/LBFGS for details.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
A ProxyReference can be used in the context of abstract functions to bind several related types of arguments to a single proxy type. Main use are ublas expression templates so that vectors, matrix rows and subvectors can be treated as one argument type
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/. Shark linear algebra definitions
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.solves systems of triangular matrices
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
Often, you want to initialize a vector using already available data or you know a fixed initialization. In this case, this header helps, since it is possible to initialize a vector using init(vec) << a,b,c,...; also if you want to split a vector into several smaller parts, you can write init(vec) >> a,b,c,...; a,b,c are allowed to be vectors, vector expressions or single values. However, all vectors needs to be initialized to the correct size. It is checked in debug mode, that size(vec) equals size(a,b,c,...). The usage is not restricted to initialization, but can be used for any assignment where a vector is constructed from several parts. For example the construction of parameter vectors in AbstractModel::parameterVector.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
The function is described in
Christian Igel, Nikolaus Hansen, and Stefan Roth. Covariance Matrix Adaptation for Multi-objective Optimization. Evolutionary Computation 15(1), pp. 1-28, 2007
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
Class for balancing two poles on a cart using a fitness function that punishes oscillating, i.e. quickly moving the cart back and forth to balance the poles.
Based on code written by Verena Heidrich-Meisner for the paper
V. Heidrich-Meisner and C. Igel. Neuroevolution strategies for episodic reinforcement learning. Journal of Algorithms, 64(4):152–168, 2009.
Special fitness function from the paper
F. Gruau, D. Whitley, L. Pyeatt, A comparison between cellular encoding and direct encoding for genetic neural networks, in: J.R. Koza, D.E. Goldberg, D.B. Fogel, R.L. Riol (Eds.), Genetic Programming 1996: Proceedings of the First Annual Conference, MIT Press, 1996, pp. 81–89.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
Multi-modal benchmark function.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
The function is described in
H. Li and Q. Zhang. Multiobjective Optimization Problems with Complicated Pareto Sets, MOEA/D and NSGA-II, IEEE Trans on Evolutionary Computation, 2(12):284-302, April 2009.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
Class for balancing one or two poles on a cart using a fitness function that decreases the longer the pole(s) balance(s). Based on code written by Verena Heidrich-Meisner for the paper
V. Heidrich-Meisner and C. Igel. Neuroevolution strategies for episodic reinforcement learning. Journal of Algorithms, 64(4):152–168, 2009.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
Class for simulating two poles balancing on a cart. Based on code written by Verena Heidrich-Meisner for the paper
V. Heidrich-Meisner and C. Igel. Neuroevolution strategies for episodic reinforcement learning. Journal of Algorithms, 64(4):152–168, 2009.
which was in turn based on code available at http://webdocs.cs.ualberta.ca/~sutton/book/code/pole.c as of 2015/4/19, written by Rich Sutton and Chuck Anderson and later modified. Faustino Gomez wrote the physics code using the differential equations from Alexis Weiland's paper and added the Runge-Kutta solver.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
Class for simulating a single pole balancing on a cart. Based on code written by Verena Heidrich-Meisner for the paper
V. Heidrich-Meisner and C. Igel. Neuroevolution strategies for episodic reinforcement learning. Journal of Algorithms, 64(4):152–168, 2009.
which was in turn based on code available at http://webdocs.cs.ualberta.ca/~sutton/book/code/pole.c as of 2015/4/19, written by Rich Sutton and Chuck Anderson and later modified. Faustino Gomez wrote the physics code using the differential equations from Alexis Weiland's paper and added the Runge-Kutta solver.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This non-convex benchmark function for real-valued optimization is a generalization from two to multiple dimensions of a classic function first proposed in:
H. H. Rosenbrock. An automatic method for finding the greatest or least value of a function. The Computer Journal 3: 175–184, 1960
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
The function is described in
Eckart Zitzler, Kalyanmoy Deb, and Lothar Thiele. Comparison of Multiobjective Evolutionary Algorithms: Empirical Results. Evolutionary Computation 8(2):173-195, 2000
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This class offers convenience functions to generate numbers using a global random number generator from the following distributions:
Additionally this class offers a global random number generator. The default is the Mersenne Twister with a cycle length of $2^19937$. This generator can be used to construct additional distributions. The seed can be set via Rng::seed .
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
This file is part of Shark. http://image.diku.dk/shark/
Shark is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Shark is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with Shark. If not, see http://www.gnu.org/licenses/.
typedef IndicatorBasedMOCMA< LeastContributorApproximator< FastRng, HypervolumeCalculator > > shark::ApproximatedVolumeMOCMA |
typedef IndicatorBasedSteadyStateMOCMA< LeastContributorApproximator< FastRng, HypervolumeCalculator > > shark::ApproximatedVolumeSteadyStateMOCMA |
Definition at line 286 of file SteadyStateMOCMA.h.
Definition at line 53 of file BinaryRBM.h.
Definition at line 49 of file BinaryRBM.h.
Definition at line 48 of file BinaryRBM.h.
Definition at line 54 of file BinaryRBM.h.
Definition at line 52 of file BinaryRBM.h.
Definition at line 50 of file BinaryRBM.h.
typedef RBM<BinaryLayer,BinaryLayer, Rng::rng_type> shark::BinaryRBM |
Definition at line 47 of file BinaryRBM.h.
typedef TwoStateSpace<0,1> shark::BinarySpace |
Definition at line 84 of file TwoStateSpace.h.
Definition at line 53 of file BipolarRBM.h.
Definition at line 49 of file BipolarRBM.h.
Definition at line 48 of file BipolarRBM.h.
Definition at line 54 of file BipolarRBM.h.
Definition at line 52 of file BipolarRBM.h.
Definition at line 50 of file BipolarRBM.h.
typedef RBM<BipolarLayer,BipolarLayer, Rng::rng_type> shark::BipolarRBM |
Definition at line 47 of file BipolarRBM.h.
typedef LabeledData<RealVector, unsigned int> shark::ClassificationDataset |
typedef boost::property_map<Graph, boost::edge_color_t>::type shark::ColorMap |
typedef ARDKernelUnconstrained<CompressedRealVector> shark::CompressedARDKernel |
Definition at line 273 of file ArdKernel.h.
typedef LabeledData<CompressedRealVector, unsigned int> shark::CompressedClassificationDataset |
typedef LinearKernel<CompressedRealVector> shark::CompressedLinearKernel |
Definition at line 133 of file LinearKernel.h.
typedef ModelKernel<CompressedRealVector> shark::CompressedModelKernel |
Definition at line 275 of file ModelKernel.h.
typedef MonomialKernel<CompressedRealVector> shark::CompressedMonomialKernel |
Definition at line 193 of file MonomialKernel.h.
typedef NormalizedKernel<CompressedRealVector> shark::CompressedNormalizedKernel |
Definition at line 277 of file NormalizedKernel.h.
typedef PolynomialKernel<CompressedRealVector> shark::CompressedPolynomialKernel |
Definition at line 303 of file PolynomialKernel.h.
typedef GaussianRbfKernel<CompressedRealVector> shark::CompressedRbfKernel |
Definition at line 279 of file GaussianRbfKernel.h.
typedef ScaledKernel<CompressedRealVector> shark::CompressedScaledKernel |
Definition at line 160 of file ScaledKernel.h.
typedef WeightedSumKernel<CompressedRealVector> shark::CompressedWeightedSumKernel |
Definition at line 394 of file WeightedSumKernel.h.
typedef SubrangeKernel<CompressedRealVector> shark::CompressesSubrangeKernel |
Definition at line 209 of file SubrangeKernel.h.
Definition at line 52 of file ConvolutionalBinaryRBM.h.
Definition at line 48 of file ConvolutionalBinaryRBM.h.
Definition at line 47 of file ConvolutionalBinaryRBM.h.
typedef SingleChainApproximator<ConvolutionalBinaryPTChain> shark::ConvolutionalBinaryParallelTempering |
Definition at line 53 of file ConvolutionalBinaryRBM.h.
Definition at line 51 of file ConvolutionalBinaryRBM.h.
Definition at line 49 of file ConvolutionalBinaryRBM.h.
typedef ConvolutionalRBM<BinaryLayer,BinaryLayer, Rng::rng_type> shark::ConvolutionalBinaryRBM |
Definition at line 46 of file ConvolutionalBinaryRBM.h.
typedef boost::mt19937 shark::DefaultRngType |
Definition at line 272 of file ArdKernel.h.
typedef ARDKernelUnconstrained<ConstRealVectorRange> shark::DenseARDMklKernel |
Definition at line 274 of file ArdKernel.h.
typedef LinearKernel shark::DenseLinearKernel |
Definition at line 132 of file LinearKernel.h.
typedef ModelKernel<RealVector> shark::DenseModelKernel |
Definition at line 274 of file ModelKernel.h.
Definition at line 192 of file MonomialKernel.h.
Definition at line 276 of file NormalizedKernel.h.
Definition at line 302 of file PolynomialKernel.h.
Definition at line 278 of file GaussianRbfKernel.h.
typedef ScaledKernel shark::DenseScaledKernel |
Definition at line 159 of file ScaledKernel.h.
typedef ScaledKernel<ConstRealVectorRange> shark::DenseScaledMklKernel |
Definition at line 161 of file ScaledKernel.h.
typedef SubrangeKernel<RealVector> shark::DenseSubrangeKernel |
Definition at line 208 of file SubrangeKernel.h.
Definition at line 393 of file WeightedSumKernel.h.
typedef boost::graph_traits<Graph>::edge_descriptor shark::Edge |
Definition at line 287 of file SteadyStateMOCMA.h.
Definition at line 289 of file RealCodedNSGAII.h.
Default fast non-dominated sorting based on the Pareto-dominance relation.
Definition at line 126 of file FastNonDominatedSort.h.
typedef boost::rand48 shark::FastRngType |
typedef std::vector<SplitMatrixType> shark::ForestInfo |
Definition at line 44 of file RFClassifier.h.
typedef std::vector< shark::RealVector > shark::FrontType |
Definition at line 14 of file AdditiveEpsilonIndicatorMain.cpp.
Definition at line 54 of file GaussianBinaryRBM.h.
Definition at line 50 of file GaussianBinaryRBM.h.
Definition at line 49 of file GaussianBinaryRBM.h.
Definition at line 55 of file GaussianBinaryRBM.h.
Definition at line 53 of file GaussianBinaryRBM.h.
Definition at line 51 of file GaussianBinaryRBM.h.
typedef RBM<GaussianLayer,BinaryLayer, Rng::rng_type> shark::GaussianBinaryRBM |
Definition at line 48 of file GaussianBinaryRBM.h.
typedef boost::adjacency_matrix< boost::undirectedS, boost::property< boost::vertex_color_t, std::string >, boost::property< boost::edge_weight_t, double, boost::property< boost::edge_color_t, std::string > >, boost::no_property > shark::Graph |
typedef boost::archive::polymorphic_iarchive shark::InArchive |
Type of an archive to read from.
Definition at line 74 of file ISerializable.h.
typedef Individual< Tour,double > shark::IndividualType |
typedef IndicatorBasedMOCMA< HypervolumeIndicator > shark::MOCMA |
typedef AbstractObjectiveFunction< RealVector, RealVector > shark::MultiObjectiveFunction |
Definition at line 290 of file AbstractObjectiveFunction.h.
typedef boost::archive::polymorphic_oarchive shark::OutArchive |
Type of an archive to write to.
Definition at line 80 of file ISerializable.h.
typedef std::vector< IndividualType > shark::Population |
typedef blas::range shark::Range |
Definition at line 288 of file RealCodedNSGAII.h.
typedef std::pair< std::vector<std::size_t> , std::vector<std::size_t> > shark::RecreationIndices |
auxiliary typedef for createCVSameSizeBalanced and createCVFullyIndexed, stores location index in the first and partition index in the second
Definition at line 145 of file CVDatasetTools.h.
typedef LabeledData<RealVector, RealVector> shark::RegressionDataset |
typedef RpropPlus shark::Rprop93 |
typedef RpropMinus shark::Rprop94 |
typedef IRpropPlus shark::Rprop99 |
typedef IRpropMinus shark::Rprop99d |
typedef std::deque<RealVector> shark::Sequence |
typedef AbstractObjectiveFunction< RealVector, double > shark::SingleObjectiveFunction |
Definition at line 289 of file AbstractObjectiveFunction.h.
typedef CARTClassifier<RealVector>::SplitMatrixType shark::SplitMatrixType |
Definition at line 43 of file RFClassifier.h.
Definition at line 285 of file SteadyStateMOCMA.h.
typedef TwoStateSpace<-1,1> shark::SymmetricBinarySpace |
Definition at line 85 of file TwoStateSpace.h.
typedef boost::archive::polymorphic_text_iarchive shark::TextInArchive |
Definition at line 75 of file ISerializable.h.
typedef boost::archive::polymorphic_text_oarchive shark::TextOutArchive |
Definition at line 81 of file ISerializable.h.
Definition at line 53 of file TruncExpBinaryRBM.h.
Definition at line 49 of file TruncExpBinaryRBM.h.
Definition at line 48 of file TruncExpBinaryRBM.h.
Definition at line 54 of file TruncExpBinaryRBM.h.
Definition at line 52 of file TruncExpBinaryRBM.h.
Definition at line 50 of file TruncExpBinaryRBM.h.
typedef RBM<TruncExpBinaryEnergy, Rng::rng_type> shark::TruncExpBinaryRBM |
Definition at line 47 of file TruncExpBinaryRBM.h.
typedef boost::graph_traits<Graph>::vertex_descriptor shark::Vertex |
typedef boost::property_map<Graph, boost::edge_weight_t>::type shark::WeightMap |
enum shark::AlphaStatus |
Enumerator | |
---|---|
AlphaFree | |
AlphaLowerBound | |
AlphaUpperBound | |
AlphaDeactivated |
Definition at line 379 of file QpSolver.h.
Enumerator | |
---|---|
AIS | |
AISMean | |
TwoSidedAISMean | |
AcceptanceRatio | |
AcceptanceRatioMean |
Definition at line 111 of file analytics.h.
enum shark::QpStopType |
reason for the quadratic programming solver to stop the iterative optimization process
Enumerator | |
---|---|
QpNone | |
QpAccuracyReached | |
QpMaxIterationsReached | |
QpTimeout |
Definition at line 63 of file QuadraticProgram.h.
double shark::annealedImportanceSampling | ( | RBMType & | rbm, |
RealVector const & | beta, | ||
std::size_t | samples | ||
) |
Definition at line 171 of file analytics.h.
References shark::blas::soft_max(), and shark::blas::sum_rows().
Referenced by annealedImportanceSampling().
double shark::annealedImportanceSampling | ( | RBMType & | rbm, |
std::size_t | chains, | ||
std::size_t | samples | ||
) |
Definition at line 196 of file analytics.h.
References annealedImportanceSampling(), and beta.
shark::ANNOUNCE_SHARK_RNG | ( | shark::FastRngType | , |
FastRng | |||
) |
shark::ANNOUNCE_SHARK_RNG | ( | shark::DefaultRngType | , |
Rng | |||
) |
WeightedLabeledData< InputType, LabelType> shark::bootstrap | ( | LabeledData< InputType, LabelType > const & | dataset, |
std::size_t | bootStrapSize = 0 |
||
) |
Creates a bootstrap partition of a labeled dataset and returns it using weighting.
Bootstrapping resamples the dataset by drawing a set of points with replacement. Thus the sampled set will contain some points multiple times and some points not at all. Bootstrapping is usefull to obtain unbiased measurements of the mean and variance of an estimator.
Optionally the size of the bootstrap (that is, the number of sampled points) can be set. By default it is 0, which indicates that it is the same size as the original dataset.
Definition at line 649 of file WeightedDataset.h.
References discrete(), and shark::LabeledData< InputT, LabelT >::numberOfElements().
WeightedUnlabeledData<InputType> shark::bootstrap | ( | UnlabeledData< InputType > const & | dataset, |
std::size_t | bootStrapSize = 0 |
||
) |
Creates a bootstrap partition of an unlabeled dataset and returns it using weighting.
Bootstrapping resamples the dataset by drawing a set of points with replacement. Thus the sampled set will contain some points multiple times and some points not at all. Bootstrapping is usefull to obtain unbiased measurements of the mean and variance of an estimator.
Optionally the size of the bootstrap (that is, the number of sampled points) can be set. By default it is 0, which indicates that it is the same size as the original dataset.
Definition at line 675 of file WeightedDataset.h.
References discrete(), and shark::Data< Type >::numberOfElements().
RealVector shark::calculateKernelMatrixParameterDerivative | ( | AbstractKernelFunction< InputType > const & | kernel, |
Data< InputType > const & | dataset, | ||
WeightMatrix const & | weights | ||
) |
Efficiently calculates the weighted derivative of a Kernel Gram Matrix w.r.t the Kernel Parameters.
The formula is \( \sum_i \sum_j w_{ij} k(x_i,x_j)\) where w_ij are the weights of the gradient and x_i x_j are the datapoints defining the gram matrix and k is the kernel. For efficiency it is assumd that w_ij = w_ji. This method is only useful when the whole Kernel Gram Matrix neds to be computed to get the weights w_ij and only computing smaller blocks is not sufficient.
kernel | the kernel for which to calculate the kernel gram matrix |
dataset | the set of points used in the gram matrix |
weights | the weights of the derivative, they must be symmetric! |
Definition at line 181 of file KernelHelpers.h.
References shark::Data< Type >::batch(), shark::AbstractKernelFunction< InputTypeT >::createState(), shark::AbstractKernelFunction< InputTypeT >::eval(), shark::Data< Type >::numberOfBatches(), shark::IParameterizable::numberOfParameters(), size(), shark::blas::subrange(), and shark::AbstractKernelFunction< InputTypeT >::weightedParameterDerivative().
Referenced by shark::RadiusMarginQuotient< InputType, CacheType >::evalDerivative(), and shark::NegativeGaussianProcessEvidence< InputType, OutputType, LabelType >::evalDerivative().
void shark::calculateMixedKernelMatrix | ( | AbstractKernelFunction< InputType >const & | kernel, |
Data< InputType > const & | dataset1, | ||
Data< InputType > const & | dataset2, | ||
blas::matrix_expression< M > & | matrix | ||
) |
Calculates the kernel gram matrix between two data sets.
kernel | the kernel for which to calculate the kernel gram matrix |
dataset1 | the set of points corresponding to rows of the Gram matrix |
dataset2 | the set of points corresponding to columns of the Gram matrix |
matrix | the target kernel matrix |
Definition at line 95 of file KernelHelpers.h.
References shark::Data< Type >::batch(), shark::blas::noalias(), shark::Data< Type >::numberOfBatches(), shark::Data< Type >::numberOfElements(), SHARK_PARALLEL_FOR, size(), SIZE_CHECK, and shark::blas::subrange().
Referenced by calculateMixedKernelMatrix().
RealMatrix shark::calculateMixedKernelMatrix | ( | AbstractKernelFunction< InputType >const & | kernel, |
Data< InputType > const & | dataset1, | ||
Data< InputType > const & | dataset2 | ||
) |
Calculates the kernel gram matrix between two data sets.
kernel | the kernel for which to calculate the kernel gram matrix |
dataset1 | the set of points corresponding to rows of the Gram matrix |
dataset2 | the set of points corresponding to columns of the Gram matrix |
Definition at line 159 of file KernelHelpers.h.
References calculateMixedKernelMatrix().
void shark::calculateRegularizedKernelMatrix | ( | AbstractKernelFunction< InputType >const & | kernel, |
Data< InputType > const & | dataset, | ||
blas::matrix_expression< M > & | matrix, | ||
double | regularizer = 0 |
||
) |
Calculates the regularized kernel gram matrix of the points stored inside a dataset.
Regularization is applied by adding the regularizer on the diagonal
kernel | the kernel for which to calculate the kernel gram matrix |
dataset | the set of points used in the gram matrix |
matrix | the target kernel matrix |
regularizer | the regularizer of the matrix which is always >= 0. default is 0. |
Definition at line 52 of file KernelHelpers.h.
References shark::Data< Type >::batch(), shark::blas::noalias(), shark::Data< Type >::numberOfBatches(), shark::Data< Type >::numberOfElements(), SHARK_CHECK, SHARK_PARALLEL_FOR, size(), SIZE_CHECK, and shark::blas::subrange().
Referenced by calculateRegularizedKernelMatrix(), shark::NegativeGaussianProcessEvidence< InputType, OutputType, LabelType >::eval(), shark::NegativeGaussianProcessEvidence< InputType, OutputType, LabelType >::evalDerivative(), kMeans(), and shark::RegularizationNetworkTrainer< InputType >::train().
RealMatrix shark::calculateRegularizedKernelMatrix | ( | AbstractKernelFunction< InputType >const & | kernel, |
Data< InputType > const & | dataset, | ||
double | regularizer = 0 |
||
) |
Calculates the regularized kernel gram matrix of the points stored inside a dataset.
Regularization is applied by adding the regularizer on the diagonal
kernel | the kernel for which to calculate the kernel gram matrix |
dataset | the set of points used in the gram matrix |
regularizer | the regularizer of the matrix which is always >= 0. default is 0. |
Definition at line 141 of file KernelHelpers.h.
References calculateRegularizedKernelMatrix(), and SHARK_CHECK.
|
inline |
Returns the number of members of each class in the dataset.
Definition at line 579 of file WeightedDataset.h.
References classSizes().
|
inline |
Returns the number of members of each class in the dataset.
Definition at line 608 of file WeightedDataset.h.
References classSizes(), and shark::WeightedLabeledData< InputT, LabelT >::labels().
bool shark::coinToss | ( | RngType & | rng, |
unsigned int | n, | ||
double | pHeads | ||
) |
Returns a random number from Binomial(n,p) by drawing random numbers from rng.
Definition at line 123 of file Binomial.h.
bool shark::coinToss | ( | RngType & | rng, |
double | pHeads | ||
) |
Flips a coin with probability of heads being pHeads by drawing random numbers from rng.
Definition at line 138 of file Bernoulli.h.
Referenced by createProblem(), shark::CircleInSquare::draw(), shark::ImpulseNoiseModel::eval(), shark::DropoutNeuron< Neuron >::function(), shark::IndicatorBasedRealCodedNSGAII< Indicator >::generateOffspring(), main(), shark::UniformCrossover::operator()(), shark::PolynomialMutator::operator()(), shark::BitflipMutator::operator()(), shark::SimulatedBinaryCrossover< RealVector >::operator()(), shark::MultiNomialDistribution::operator()(), shark::BipolarLayer::sample(), and shark::BinaryLayer::sample().
bool shark::compare_fitness | ( | const IndividualType & | a, |
const IndividualType & | b | ||
) |
Definition at line 101 of file TSP.cpp.
References discrete(), shark::blas::max(), shark::Individual< PointType, FitnessTypeT, Chromosome >::searchPoint(), SIZE_CHECK, swap(), and shark::Individual< PointType, FitnessTypeT, Chromosome >::unpenalizedFitness().
Referenced by main().
VectorMatrixTraits<VectorType>::DenseMatrixType shark::covariance | ( | const Data< VectorType > & | data | ) |
Calculates the covariance matrix of the data vectors.
Referenced by createData(), mean(), shark::NormalDistributedPoints::NormalDistributedPoints(), shark::NormalizeComponentsWhitening::train(), and shark::NormalizeComponentsZCA::train().
creates a batch from a range of inputs
Definition at line 171 of file BatchInterface.h.
Referenced by createBatch(), shark::AbstractLoss< unsigned int, RealVector >::eval(), shark::AbstractKernelFunction< InputType >::eval(), shark::NormalizedKernel< InputType >::eval(), shark::CARTClassifier< RealVector >::eval(), shark::AbstractModel< InputT, unsigned int >::eval(), shark::AbstractLoss< unsigned int, RealVector >::evalDerivative(), shark::DataDistribution< RealVector >::generateDataset(), shark::LabeledDataDistribution< RealVector, unsigned int >::generateDataset(), shark::AbstractClustering< RealVector >::hardMembership(), shark::CSvmDerivative< InputType, CacheType >::modelCSvmParameterDerivative(), subBatch(), shark::KernelSGDTrainer< InputType, CacheType >::train(), shark::CSvmTrainer< InputType, CacheType >::train(), shark::KernelBudgetedSGDTrainer< InputType, CacheType >::train(), shark::LassoRegression< InputVectorType >::trainInternal(), shark::NormalizedKernel< InputType >::weightedInputDerivative(), shark::NormalizedKernel< InputType >::weightedParameterDerivative(), and shark::CSvmDerivative< InputType, CacheType >::write().
Batch<typename Range::value_type>::type shark::createBatch | ( | Range const & | range | ) |
Definition at line 176 of file BatchInterface.h.
References createBatch(), size(), and shark::blas::dense_matrix_adaptor< T, Orientation >::size1().
boost::disable_if< boost::is_arithmetic<WeightRange>, WeightedLabeledData< typename boost::range_value<InputRange>::type, typename boost::range_value<LabelRange>::type >>::type shark::createLabeledDataFromRange | ( | InputRange const & | inputs, |
LabelRange const & | labels, | ||
WeightRange const & | weights, | ||
std::size_t | batchSize = 0 |
||
) |
creates a weighted unweighted data object from two ranges, representing data and weights
Definition at line 622 of file WeightedDataset.h.
References createDataFromRange(), createLabeledDataFromRange(), SHARK_CHECK, and size().
boost::disable_if< boost::is_arithmetic<WeightRange>, WeightedUnlabeledData< typename boost::range_value<DataRange>::type > >::type shark::createUnlabeledDataFromRange | ( | DataRange const & | data, |
WeightRange const & | weights, | ||
std::size_t | batchSize = 0 |
||
) |
creates a weighted unweighted data object from two ranges, representing data and weights
Definition at line 423 of file WeightedDataset.h.
References createDataFromRange(), createUnlabeledDataFromRange(), SHARK_CHECK, and size().
std::size_t shark::dataDimension | ( | DataView< DatasetType > const & | view | ) |
Return the dimensionality of the dataset represented by the view.
Definition at line 327 of file DataView.h.
References dataDimension(), and shark::DataView< DatasetType >::dataset().
std::size_t shark::dataDimension | ( | WeightedUnlabeledData< InputType > const & | dataset | ) |
Return the dimnsionality of points of a weighted dataset.
Definition at line 585 of file WeightedDataset.h.
References dataDimension().
std::size_t shark::discrete | ( | RngType & | rng, |
std::size_t | low, | ||
std::size_t | high | ||
) |
Draws a discrete number in {low,low+1,...,high} by drawing random numbers from rng.
Definition at line 111 of file DiscreteUniform.h.
Referenced by shark::RemoveBudgetMaintenanceStrategy< InputType >::addToModel(), bootstrap(), compare_fitness(), createCVIID(), shark::IndicatorBasedSteadyStateMOCMA< Indicator >::doInit(), shark::SMSEMOA::doInit(), shark::IndicatorBasedMOCMA< Indicator >::doInit(), shark::IndicatorBasedRealCodedNSGAII< Indicator >::doInit(), shark::PamiToy::draw(), shark::ImagePatches::draw(), shark::IndicatorBasedSteadyStateMOCMA< Indicator >::generateOffspring(), getSamples(), main(), shark::OnePointCrossover::operator()(), shark::TournamentSelection< Predicate >::operator()(), shark::MultiNomialDistribution::operator()(), shark::Pegasos< VectorType >::solve(), shark::QpBoxLinear< InputT >::solve(), shark::QpMcLinear< InputT >::solve(), shark::McPegasos< VectorType >::solve(), shark::QpBoxLinear< CompressedRealVector >::solve(), shark::KernelSGDTrainer< InputType, CacheType >::train(), and shark::KernelBudgetedSGDTrainer< InputType, CacheType >::train().
double shark::entropy | ( | Distribution & | d, |
std::size_t | trials = 10000 |
||
) |
Estimates the entropy of a distribution. The more trials the better is the estimate, but good estimates are slow.
Distribution | The type of distribution, needs to provide operator()() for sampling purposes. |
[in,out] | d | Distribution instance to sample from. |
[in] | trials | The number of samples drawn from the distribution. Needs to be larger than 0. |
shark::Exception | if trials == 0. |
double shark::entropy | ( | const Cauchy< RngType > & | distribution | ) |
Returns the entropy of the Cauchy distribution.
Referenced by shark::Cauchy< RngType >::p().
double shark::estimateLogFreeEnergy | ( | RBMType & | rbm, |
Data< RealVector > const & | initDataset, | ||
RealVector const & | beta, | ||
std::size_t | samples, | ||
PartitionEstimationAlgorithm | algorithm = AcceptanceRatioMean , |
||
float | burnInPercentage = 0.1 |
||
) |
Definition at line 154 of file analytics.h.
References estimateLogFreeEnergyFromEnergySamples().
Referenced by estimateLogFreeEnergy().
double shark::estimateLogFreeEnergy | ( | RBMType & | rbm, |
Data< RealVector > const & | initDataset, | ||
std::size_t | chains, | ||
std::size_t | samples, | ||
PartitionEstimationAlgorithm | algorithm = AIS , |
||
float | burnInPercentage = 0.1 |
||
) |
Definition at line 182 of file analytics.h.
References beta, and estimateLogFreeEnergy().
|
inline |
Definition at line 119 of file analytics.h.
References AcceptanceRatio, AcceptanceRatioMean, AIS, AISMean, shark::blas::row(), shark::blas::soft_max(), shark::blas::sum_rows(), and TwoSidedAISMean.
Referenced by estimateLogFreeEnergy().
double shark::evalSkipMissingFeatures | ( | const AbstractKernelFunction< InputType > & | kernelFunction, |
const InputTypeT1 & | inputA, | ||
const InputTypeT2 & | inputB | ||
) |
Does a kernel function evaluation with Missing features in the inputs
kernelFunction | The kernel function used to do evaluation |
inputA | a input |
inputB | another input |
The kernel k(x,y) is evaluated taking missing features into account. For this it is checked whether a feature of x or y is nan and in this case the corresponding features in inputA and inputB won't be considered.
Definition at line 58 of file EvalSkipMissingFeatures.h.
References shark::AbstractKernelFunction< InputTypeT >::eval(), SHARKEXCEPTION, SIZE_CHECK, and shark::AbstractKernelFunction< InputTypeT >::supportsVariableInputSize().
Referenced by shark::MissingFeaturesKernelExpansion< InputType >::computeNorm(), shark::ExampleModifiedKernelMatrix< InputType, CacheType >::entry(), and shark::MissingFeaturesKernelExpansion< InputType >::eval().
double shark::evalSkipMissingFeatures | ( | const AbstractKernelFunction< InputType > & | kernelFunction, |
const InputTypeT1 & | inputA, | ||
const InputTypeT2 & | inputB, | ||
InputTypeT3 const & | missingness | ||
) |
Do kernel function evaluation while Missing features in the inputs
kernelFunction | The kernel function used to do evaluation |
inputA | a input |
inputB | another input |
missingness | used to decide which features in the inputs to take into consideration for the purpose of evaluation. If a feature is NaN, then the corresponding features in inputA and inputB won't be considered. |
Definition at line 105 of file EvalSkipMissingFeatures.h.
References shark::AbstractKernelFunction< InputTypeT >::eval(), SHARKEXCEPTION, SIZE_CHECK, and shark::AbstractKernelFunction< InputTypeT >::supportsVariableInputSize().
|
inline |
Exports a set of filters as a grid image.
It is assumed that the filters each form a row in the filter-matrix. Moreover, the sizes of the filter images has to be given and it must gold width*height=W.size2(). The filters a re printed on a single image as a grid. The grid will be close to square. And the image are separated by a black 1 pixel wide line. The output will be normalized so that all images are on the same scale.
basename | File to write to. ".pgm" is appended to the filename |
filters | Matrix storing the filters row by row |
width | Width of the filter image |
height | Height of th filter image |
Definition at line 190 of file Pgm.h.
References shark::blas::adapt_vector(), exportPGM(), shark::blas::min(), shark::blas::noalias(), shark::blas::row(), SIZE_CHECK, shark::blas::subrange(), and shark::blas::to_matrix().
Referenced by main().
|
inline |
Exports a set of filters as a grid image.
It is assumed that the filters each form a row in the filter-matrix. Moreover, the sizes of the filter images has to be given and it must gold width*height=W.size2(). The filters a re printed on a single image as a grid. The grid will be close to square. And the image are separated by a black 1 pixel wide line. The output will be normalized so that all images are on the same scale.
basename | File to write to. ".pgm" is appended to the filename |
filters | Matrix storing the filters row by row |
width | Width of the filter image |
height | Height of th filter image |
Definition at line 227 of file Pgm.h.
References shark::blas::adapt_vector(), shark::Data< Type >::batch(), dataDimension(), shark::Data< Type >::element(), exportPGM(), shark::blas::max(), shark::blas::min(), shark::blas::noalias(), shark::Data< Type >::numberOfBatches(), shark::Data< Type >::numberOfElements(), SIZE_CHECK, shark::blas::subrange(), and shark::blas::to_matrix().
void shark::exportPGM | ( | std::string const & | fileName, |
T const & | data, | ||
std::size_t | sx, | ||
std::size_t | sy, | ||
bool | normalize = false |
||
) |
Export a PGM image to file.
fileName | File to write to |
data | Linear object storing image |
sx | Width of image |
sy | Height of image |
normalize | Adjust values to [0,255], default false |
Definition at line 162 of file Pgm.h.
Referenced by exportFiltersToPGMGrid(), and main().
S& shark::fusionize | ( | detail::FusionFacade< S > & | facade | ) |
Definition at line 147 of file BatchInterfaceAdaptStruct.h.
S const& shark::fusionize | ( | detail::FusionFacade< S > const & | facade | ) |
Definition at line 151 of file BatchInterfaceAdaptStruct.h.
boost::disable_if<detail::isFusionFacade<S>,S&>::type shark::fusionize | ( | S & | facade | ) |
Definition at line 157 of file BatchInterfaceAdaptStruct.h.
boost::disable_if<detail::isFusionFacade<S>,S const& >::type shark::fusionize | ( | S const & | facade | ) |
Definition at line 162 of file BatchInterfaceAdaptStruct.h.
double shark::gauss | ( | RngType & | rng, |
double | mean, | ||
double | variance | ||
) |
Draws a number from the normal distribution with given mean and variance by drawing random numbers from rng.
Definition at line 112 of file Normal.h.
Referenced by shark::Wave::draw(), shark::PamiToy::draw(), shark::GaussianNoiseModel::eval(), shark::MultiVariateNormalDistributionCholesky::generate(), initRandomNormal(), shark::VDCMA::lambda(), shark::LMCMA::lambda(), main(), shark::MultiVariateNormalDistribution::operator()(), shark::Sphere::proposeStartingPoint(), shark::Schwefel::proposeStartingPoint(), and shark::ConstrainedSphere::proposeStartingPoint().
boost::range_reference<Range>::type shark::get | ( | Range & | range, |
std::size_t | i | ||
) |
returns the i-th element of a range
Definition at line 50 of file Range.h.
References size(), and SIZE_CHECK.
Referenced by shark::DataDistribution< RealVector >::generateDataset(), shark::LabeledDataDistribution< RealVector, unsigned int >::generateDataset(), main(), shark::QpBoxLinear< CompressedRealVector >::QpBoxLinear(), repartitionByClass(), and shark::McReinforcedSvmTrainer< InputType, CacheType >::train().
boost::range_reference<Range const>::type shark::get | ( | Range const & | range, |
std::size_t | i | ||
) |
Definition at line 58 of file Range.h.
References size(), and SIZE_CHECK.
void shark::importHDF5 | ( | Data< VectorType > & | data, |
const std::string & | fileName, | ||
const std::string & | datasetName | ||
) |
void shark::importHDF5 | ( | LabeledData< VectorType, LabelType > & | labeledData, |
const std::string & | fileName, | ||
const std::string & | data, | ||
const std::string & | label | ||
) |
Import data to a LabeledData object from a HDF5 file.
labeledData | Container storing the loaded data |
fileName | The name of HDF5 file to be read from |
data | the HDF5 dataset name for data |
label | the HDF5 dataset name for label |
VectorType | Type of object stored in Shark data container |
LableType | Type of label |
void shark::importHDF5 | ( | Data< VectorType > & | data, |
const std::string & | fileName, | ||
const std::vector< std::string > & | cscDatasetName | ||
) |
Import data from HDF5 dataset of compressed sparse column format.
data | Container storing the loaded data |
fileName | The name of HDF5 file to be read from |
cscDatasetName | the CSC dataset names used to construct a matrix |
VectorType | Type of object stored in Shark data container |
void shark::importHDF5 | ( | LabeledData< VectorType, LabelType > & | labeledData, |
const std::string & | fileName, | ||
const std::vector< std::string > & | cscDatasetName, | ||
const std::string & | label | ||
) |
Import data from HDF5 dataset of compressed sparse column format.
labeledData | Container storing the loaded data |
fileName | The name of HDF5 file to be read from |
cscDatasetName | the CSC dataset names used to construct a matrix |
label | the HDF5 dataset name for label |
VectorType | Type of object stored in Shark data container |
LabelType | Type of label |
void shark::importPGM | ( | std::string const & | fileName, |
T & | data, | ||
std::size_t & | sx, | ||
std::size_t & | sy | ||
) |
Import a PGM image from file.
fileName | The file to read from |
data | Linear object for storing image |
sx | Width of imported image |
sy | Height of imported image |
Definition at line 141 of file Pgm.h.
Referenced by importPGMDir().
void shark::importPGMDir | ( | const std::string & | p, |
T & | container, | ||
std::vector< ImageInformation > & | info | ||
) |
Import PGM images scanning a directory recursively.
p | Directory |
container | Container storing images |
info | Vector storing image informations |
Definition at line 281 of file Pgm.h.
References importPGM(), shark::ImageInformation::name, shark::ImageInformation::x, and shark::ImageInformation::y.
Referenced by importPGMSet().
void shark::importPGMSet | ( | const std::string & | p, |
Data< T > & | set, | ||
Data< ImageInformation > & | setInfo | ||
) |
Import PGM images scanning a directory recursively.
p | Directory |
set | Set storing images |
setInfo | Vector storing image informations |
Definition at line 311 of file Pgm.h.
References createDataFromRange(), and importPGMDir().
Referenced by main().
std::size_t shark::inputDimension | ( | DataView< DatasetType > const & | view | ) |
Return the input dimensionality of the labeled dataset represented by the view.
Definition at line 316 of file DataView.h.
References shark::DataView< DatasetType >::dataset(), and inputDimension().
std::size_t shark::inputDimension | ( | WeightedLabeledData< InputType, LabelType > const & | dataset | ) |
Return the input dimensionality of a weighted labeled dataset.
Definition at line 591 of file WeightedDataset.h.
References dataDimension(), and shark::WeightedLabeledData< InputT, LabelT >::inputs().
SHARK_EXPORT_SYMBOL std::size_t shark::kMeans | ( | Data< RealVector > const & | data, |
std::size_t | k, | ||
Centroids & | centroids, | ||
std::size_t | maxIterations = 0 |
||
) |
The k-means clustering algorithm.
data | vector-valued data to be clustered |
k | number of clusters |
centroids | centroids input/output |
maxIterations | maximum number of k-means iterations; 0: unlimited |
Referenced by main().
SHARK_EXPORT_SYMBOL std::size_t shark::kMeans | ( | Data< RealVector > const & | data, |
RBFLayer & | model, | ||
std::size_t | maxIterations = 0 |
||
) |
The k-means clustering algorithm for initializing an RBF Layer.
data | vector-valued data to be clustered |
model | RBFLayer input/output |
maxIterations | maximum number of k-means iterations; 0: unlimited |
KernelExpansion<InputType> shark::kMeans | ( | Data< InputType > const & | dataset, |
std::size_t | k, | ||
AbstractKernelFunction< InputType > & | kernel, | ||
std::size_t | maxIterations = 0 |
||
) |
Definition at line 108 of file KMeans.h.
References shark::KernelExpansion< InputType >::alpha(), shark::blas::arg_min(), calculateRegularizedKernelMatrix(), shark::blas::max(), shark::blas::noalias(), shark::Data< Type >::numberOfElements(), shark::KernelExpansion< InputType >::offset(), shark::blas::safe_div(), shark::KernelExpansion< InputType >::setStructure(), sqr(), and uni().
double shark::kullback_leiber_divergence | ( | DistributionP & | p, |
DistributionQ & | q, | ||
std::size_t | trials = 10000 |
||
) |
Estimates the kullback-leibler-divergence between two distributions. The more trials the better is the estimate, but good estimates are slow.
DistributionP | The type of distribution, needs to provide operator()() for sampling purposes. |
DistributionQ | The type of distribution, needs to provide operator()() for sampling purposes. |
[in,out] | p | Distribution instance to sample from. |
[in,out] | q | Distribution instance to sample from. |
[in] | trials | The number of samples drawn from the distribution. Needs to be larger than 0. |
shark::Exception | if trials == 0. |
Definition at line 49 of file KullbackLeiberDivergence.h.
std::size_t shark::labelDimension | ( | DataView< DatasetType > const & | view | ) |
Return the label dimensionality of the labeled dataset represented by the view.
Definition at line 321 of file DataView.h.
References shark::DataView< DatasetType >::dataset(), and labelDimension().
std::size_t shark::labelDimension | ( | WeightedLabeledData< InputType, LabelType > const & | dataset | ) |
Return the label/output dimensionality of a labeled dataset.
Definition at line 597 of file WeightedDataset.h.
References dataDimension(), and shark::WeightedLabeledData< InputT, LabelT >::labels().
double shark::logPartitionFunction | ( | RBMType const & | rbm, |
double | beta = 1.0 |
||
) |
Calculates the value of the partition function $Z$.
Only useful for small input and theoretical analysis
rbm | the RBM for which to calculate the function |
beta | the inverse temperature of the RBM. default is 1 |
Definition at line 48 of file analytics.h.
References beta.
Referenced by negativeLogLikelihood().
KeyValuePair<Key,Value> shark::makeKeyValuePair | ( | Key const & | key, |
Value const & | value | ||
) |
Creates a KeyValuePair.
Definition at line 80 of file KeyValuePair.h.
References shark::KeyValuePair< Key, Value >::key, shark::KeyValuePair< Key, Value >::operator<(), shark::KeyValuePair< Key, Value >::operator==(), swap(), and shark::KeyValuePair< Key, Value >::value.
ResultSet<T,U> shark::makeResultSet | ( | T const & | t, |
U const & | u | ||
) |
Generates a typed solution given the search point and the corresponding objective function value.
[in] | t | The search point. |
[in] | u | The objective function value. |
Definition at line 71 of file ResultSets.h.
VectorType shark::mean | ( | Data< VectorType > const & | data | ) |
Calculates the mean vector of the input vectors.
Referenced by createData(), shark::BaseRng< RNG >::diffGeom(), shark::ZDT6::eval(), shark::KernelTargetAlignment< InputType, LabelType >::evalDerivative(), shark::TruncatedExponentialLayer::expectedPhiValue(), exportKernelMatrix(), shark::BaseRng< RNG >::gauss(), getSamples(), shark::McPegasos< VectorType >::lossGradientADS(), shark::McPegasos< VectorType >::lossGradientATS(), mean(), shark::DiffGeometric< RngType >::p(), shark::BaseRng< RNG >::poisson(), shark::FisherLDA::setWhitening(), shark::statistics::Variance::statistics(), shark::NormalizeComponentsWhitening::train(), shark::NormalizeComponentsZCA::train(), shark::NormalTrainer::train(), shark::NormalizeComponentsUnitVariance< DataType >::train(), shark::QpMcLinearATS< InputT >::updateWeightVectors(), shark::QpMcLinearATM< InputT >::updateWeightVectors(), and shark::QpMcLinearReinforced< InputT >::updateWeightVectors().
VectorType shark::mean | ( | UnlabeledData< VectorType > const & | data | ) |
Definition at line 69 of file Statistics.h.
References covariance(), mean(), and variance().
void shark::meanvar | ( | const Data< Vec1T > & | data, |
blas::vector_container< Vec2T > & | mean, | ||
blas::vector_container< Vec3T > & | variance | ||
) |
Calculates the mean and variance values of the input data.
Referenced by main(), shark::NormalizeComponentsWhitening::train(), shark::NormalizeComponentsZCA::train(), and shark::NormalizeComponentsUnitVariance< DataType >::train().
void shark::meanvar | ( | const Data< Vec1T > & | data, |
blas::vector_container< Vec2T > & | mean, | ||
blas::matrix_container< MatT > & | variance | ||
) |
Calculates the mean, variance and covariance values of the input data.
Returns the iterator to the median element. after this call, the range is partially ordered.
After the call, all elements left of the median element are guaranteed to be <= median and all element on the right are >= median.
Definition at line 82 of file functional.h.
References size().
Referenced by median_element(), and partitionEqually().
boost::range_iterator<Range>::type shark::median_element | ( | Range const & | rangeAdaptor | ) |
Definition at line 95 of file functional.h.
References median_element().
double shark::negativeLogLikelihood | ( | RBMType const & | rbm, |
UnlabeledData< RealVector > const & | inputs, | ||
double | beta = 1.0 |
||
) |
Estimates the negative log-likelihood of a set of input vectors under the models distribution.
Only useful for small input and theoretical analysis
rbm | the Restricted Boltzmann machine for which the negative log likelihood of the data is to be calculated |
inputs | the input vectors |
beta | the inverse temperature of the RBM. default is 1 |
Definition at line 102 of file analytics.h.
References beta, logPartitionFunction(), and negativeLogLikelihoodFromLogPartition().
Referenced by shark::ExactGradient< RBMType >::eval(), and main().
double shark::negativeLogLikelihoodFromLogPartition | ( | RBMType const & | rbm, |
UnlabeledData< RealVector > const & | inputs, | ||
double | logPartition, | ||
double | beta = 1.0 |
||
) |
Estimates the negative log-likelihood of a set of input vectors under the models distribution using the partition function.
Only useful for small input and theoretical analysis
rbm | the Restricted Boltzmann machine for which the negative log likelihood of the data is to be calculated |
inputs | the input vectors |
logPartition | the logarithmic value of the partition function of the RBM. |
beta | the inverse temperature of the RBM. default is 1 |
Definition at line 79 of file analytics.h.
References shark::Data< Type >::batches(), beta, shark::blas::repeat(), and shark::blas::sum().
Referenced by negativeLogLikelihood().
std::size_t shark::numberOfClasses | ( | DataView< DatasetType > const & | view | ) |
Return the number of classes (size of the label vector) of a classification dataset with RealVector label encoding.
Definition at line 310 of file DataView.h.
References shark::DataView< DatasetType >::dataset(), and numberOfClasses().
|
inline |
Definition at line 574 of file WeightedDataset.h.
References numberOfClasses().
std::size_t shark::numberOfClasses | ( | WeightedLabeledData< InputType, unsigned int > const & | dataset | ) |
Return the number of classes (highest label value +1) of a classification dataset with unsigned int label encoding.
Definition at line 602 of file WeightedDataset.h.
References shark::WeightedLabeledData< InputT, LabelT >::labels(), and numberOfClasses().
std::ostream& shark::operator<< | ( | std::ostream & | out, |
ResultSet< SearchPoint, Result > const & | solution | ||
) |
Definition at line 75 of file ResultSets.h.
References shark::ResultSet< SearchPointT, ResultT >::value.
std::ostream& shark::operator<< | ( | std::ostream & | out, |
ValidatedSingleObjectiveResultSet< SearchPoint > const & | solution | ||
) |
Definition at line 131 of file ResultSets.h.
Stream& shark::operator<< | ( | Stream & | s, |
const WilcoxonRankSumTest::Element & | element | ||
) |
Definition at line 166 of file WilcoxonRankSumTest.h.
References shark::WilcoxonRankSumTest::Element::m_rank, shark::WilcoxonRankSumTest::Element::m_sample, and shark::WilcoxonRankSumTest::Element::m_value.
std::ostream& shark::operator<< | ( | std::ostream & | stream, |
const WeightedUnlabeledData< T > & | d | ||
) |
brief Outstream of elements for weighted data.
Definition at line 408 of file WeightedDataset.h.
std::ostream& shark::operator<< | ( | std::ostream & | stream, |
const WeightedLabeledData< T, U > & | d | ||
) |
brief Outstream of elements for weighted labeled data.
Definition at line 547 of file WeightedDataset.h.
detail::ConcatenatedModelWrapper<InputT,IntermediateT,OutputT> shark::operator>> | ( | AbstractModel< InputT, IntermediateT > & | firstModel, |
AbstractModel< IntermediateT, OutputT > & | secondModel | ||
) |
Connects two AbstractModels so that the output of the first model is the input of the second.
The type of the output of the first model must match the type of the input of the second model exactly.
Definition at line 334 of file ConcatenatedModel.h.
detail::ConcatenatedModelList<InputT,IntermediateT,OutputT> shark::operator>> | ( | const detail::ConcatenatedModelWrapperBase< InputT, IntermediateT > & | firstModel, |
AbstractModel< IntermediateT, OutputT > & | secondModel | ||
) |
Connects another AbstractModel two a previously created connection of models.
Definition at line 341 of file ConcatenatedModel.h.
void shark::partial_shuffle | ( | RandomAccessIterator | begin, |
RandomAccessIterator | middle, | ||
RandomAccessIterator | end, | ||
Rng & | rng | ||
) |
random_shuffle algorithm which stops after acquiring the random subsequence for [begin,middle)
Definition at line 46 of file functional.h.
Referenced by partial_shuffle(), randomSubBatch(), and randomSubset().
void shark::partial_shuffle | ( | RandomAccessIterator | begin, |
RandomAccessIterator | middle, | ||
RandomAccessIterator | end | ||
) |
random_shuffle algorithm which stops after acquiring the random subsequence for [begin,middle)
Definition at line 71 of file functional.h.
References partial_shuffle(), and uni().
Partitions a range in two parts as equal in size as possible.
The Algorithm partitions the range and returns the splitpoint. The elements in the range are ordered such that all elements in [begin,splitpoint) < [splitpoint,end). This partition is done such that the ranges are as equally sized as possible. It is guaranteed that the left range is not empty. However, if the range consists only of equal elements, the return value will be the end iterator indicating that there is no split possible. The whole algorithm runs in linear time by iterating 2 times over the sequence.
Definition at line 109 of file functional.h.
References median_element().
Referenced by partitionEqually(), and shark::BinaryTree< VectorType >::splitList().
boost::range_iterator<Range>::type shark::partitionEqually | ( | Range const & | rangeAdaptor | ) |
Partitions a range in two parts as equal in size as possible and returns it's result.
This the verison for adapted ranges.
Definition at line 139 of file functional.h.
References partitionEqually().
DataView<DatasetType>::batch_type shark::randomSubBatch | ( | DataView< DatasetType > const & | view, |
std::size_t | size | ||
) |
Creates a random batch of a given size.
view | the view from which the batch is to be created |
size | the size of the batch |
Definition at line 260 of file DataView.h.
References partial_shuffle(), shark::DataView< DatasetType >::size(), and subBatch().
DataView<DatasetType> shark::randomSubset | ( | DataView< DatasetType > const & | view, |
std::size_t | size | ||
) |
creates a random subset of a DataView with given size
view | the view for which the subset is to be created |
size | the size of the subset |
Definition at line 232 of file DataView.h.
References partial_shuffle(), shark::DataView< DatasetType >::size(), and subset().
Referenced by main().
FrontType shark::read_front | ( | Stream & | in, |
std::size_t | noObjectives, | ||
const std::string & | separator = " " , |
||
std::size_t | headerLines = 0 |
||
) |
Definition at line 17 of file AdditiveEpsilonIndicatorMain.cpp.
Referenced by main().
shark::SHARK_VECTOR_MATRIX_ASSIGNMENT | ( | BigReal | ) |
shark::SHARK_VECTOR_MATRIX_TYPEDEFS | ( | long | double, |
BigReal | |||
) |
shark::SHARK_VECTOR_MATRIX_TYPEDEFS | ( | bool | , |
Bool | |||
) |
void shark::shuffle | ( | Iterator | begin, |
Iterator | end, | ||
Rng & | rng | ||
) |
random_shuffle algorithm which stops after acquiring the random subsequence for [begin,middle)
Definition at line 60 of file functional.h.
References swap().
Referenced by createCVBatch(), shark::UnlabeledData< InputType >::shuffle(), and shark::LabeledData< InputType, LabelType >::shuffle().
std::size_t shark::size | ( | Range const & | range | ) |
returns the size of a range
This is just a fix to a bug of boost::size which produces a lot of useless warnings
Definition at line 44 of file Range.h.
Referenced by shark::blas::applyHouseholderOnTheLeft(), shark::blas::applyHouseholderOnTheRight(), shark::blas::approxsolveSymmPosDefSystem(), shark::blas::approxsolveSymmPosDefSystemInPlace(), shark::blas::kernels::assign(), shark::blas::assign(), shark::blas::kernels::assign_sparse(), shark::blas::axpy_prod(), calculateKernelMatrixParameterDerivative(), calculateMixedKernelMatrix(), shark::KHCTree< Container, CuttingAccuracy >::calculateNormal(), shark::LCTree< VectorType, CuttingAccuracy >::calculateNormal(), calculateRegularizedKernelMatrix(), shark::Chessboard::Chessboard(), classSizes(), createBatch(), createDataFromRange(), shark::blas::createHouseholderReflection(), createLabeledDataFromRange(), shark::GibbsOperator< RBMType >::createSample(), createUnlabeledDataFromRange(), shark::DataView< shark::Data< LabelType > const >::DataView(), shark::blas::diagonalMahalanobisDistance(), shark::blas::diagonalMahalanobisDistanceSqr(), shark::blas::diagonalMahalanobisNorm(), shark::blas::diagonalMahalanobisNormSqr(), shark::blas::distance(), shark::blas::distanceSqr(), shark::blas::divide_assign(), shark::blas::kernels::dot(), shark::blas::matrix_vector_range< M >::end(), shark::Energy< RBM >::energy(), shark::Energy< RBM >::energyFromHiddenInput(), shark::Energy< RBM >::energyFromVisibleInput(), shark::TruncatedExponentialLayer::energyTerm(), shark::ZeroOneLoss< LabelType, OutputType >::eval(), shark::HingeLoss::eval(), shark::SquaredHingeLoss::eval(), shark::MissingFeaturesKernelExpansion< InputType >::eval(), shark::ZeroOneLoss< unsigned int, RealVector >::eval(), shark::DiscreteKernel::eval(), shark::NormalizedKernel< InputType >::eval(), shark::NearestNeighborRegression< InputType >::eval(), shark::CARTClassifier< RealVector >::eval(), shark::NearestNeighborClassifier< InputType >::eval(), shark::ProductKernel< InputType >::eval(), shark::SoftNearestNeighborClassifier< InputType >::eval(), shark::NBClassifier< InputType, OutputType >::eval(), shark::OneVersusOneClassifier< InputType >::eval(), shark::SquaredLoss< Sequence, Sequence >::eval(), shark::NegativeWilcoxonMannWhitneyStatistic< LabelType, OutputType >::eval(), shark::ArgMaxConverter< KernelExpansion< InputType > >::eval(), shark::WeightedSumKernel< InputType >::eval(), shark::KernelExpansion< RealVector >::eval(), shark::HuberLoss::evalDerivative(), shark::TukeyBiweightLoss::evalDerivative(), shark::Ellipsoid::evalDerivative(), shark::HingeLoss::evalDerivative(), shark::SquaredHingeLoss::evalDerivative(), shark::Rosenbrock::evalDerivative(), shark::KernelTargetAlignment< InputType, LabelType >::evalDerivative(), shark::SquaredLoss< Sequence, Sequence >::evalDerivative(), shark::TruncatedExponentialLayer::expectedParameterDerivative(), shark::TruncatedExponentialLayer::expectedPhiValue(), exportKernelMatrix(), shark::AbstractKernelFunction< InputType >::featureDistanceSqr(), shark::blas::kernels::gemv(), shark::MultiVariateNormalDistributionCholesky::generate(), get(), shark::EnergyStoringTemperedMarkovChain< Operator >::getDownDifferences(), shark::NBClassifier< InputType, OutputType >::getFeatureDist(), shark::SimpleNearestNeighbors< InputType, LabelType >::getNeighbors(), shark::TreeNearestNeighbors< InputType, LabelType >::getNeighbors(), shark::EnergyStoringTemperedMarkovChain< Operator >::getUpDifferences(), shark::HierarchicalClustering< InputT >::hardMembership(), shark::AbstractClustering< RealVector >::hardMembership(), shark::BinaryTree< VectorType >::kernel(), shark::LabeledData< InputType, LabelType >::LabeledData(), shark::TruncatedExponentialLayer::logMarginalize(), shark::Energy< RBM >::logUnnormalizedProbabilityHidden(), shark::Energy< RBM >::logUnnormalizedProbabilityVisible(), shark::blas::lu_factorize(), main(), median_element(), shark::blas::minus_assign(), shark::blas::multiply_assign(), shark::IterativeNNQuery< DataContainer >::next(), shark::TruncatedExponentialLayer::numberOfParameters(), GaussianTaskKernel< InputTypeT >::numberOfTasks(), shark::UniformRankingSelection::operator()(), shark::ParetoDominanceComparator< Extractor >::operator()(), shark::LinearRankingSelection< Extractor >::operator()(), shark::ElitistSelection< Extractor >::operator()(), shark::IndicatorBasedSelection< shark::HypervolumeIndicator >::operator()(), shark::blas::operator*=(), shark::blas::operator+=(), shark::blas::operator-=(), shark::blas::operator/=(), shark::blas::operator<<(), shark::TruncatedExponentialLayer::parameterDerivative(), shark::TruncatedExponentialLayer::phi(), shark::blas::plus_assign(), shark::blas::randomRotationMatrix(), repartitionByClass(), shark::MultiVariateNormalDistribution::resize(), shark::blas::triangular_matrix< T, Orientation, TriangularType >::resize(), shark::MultiVariateNormalDistributionCholesky::resize(), shark::ROC::ROC(), shark::TruncatedExponentialLayer::sample(), shark::HypervolumeCalculator::serialize(), shark::OneClassSvmTrainer< InputType, CacheType >::setCacheSize(), shark::KernelSGDTrainer< InputType, CacheType >::setCacheSize(), shark::AbstractSvmTrainer< InputType, RealVector, KernelExpansion< InputType > >::setCacheSize(), shark::McPegasos< VectorType >::solve(), splitAtElement(), shark::statistics::FractionMissing::statistics(), shark::statistics::Mean::statistics(), shark::statistics::Variance::statistics(), shark::statistics::Quantile::statistics(), shark::GridSearch::step(), shark::TruncatedExponentialLayer::sufficientStatistics(), toDataset(), shark::blas::kernels::tpmv(), shark::McReinforcedSvmTrainer< InputType, CacheType >::train(), shark::CSvmTrainer< InputType, CacheType >::train(), shark::SquaredHingeCSvmTrainer< InputType, CacheType >::train(), shark::blas::kernels::trmv(), shark::blas::kernels::trsv(), shark::ScaledKernel< InputType >::weightedInputDerivative(), shark::NormalizedKernel< InputType >::weightedInputDerivative(), shark::WeightedSumKernel< InputType >::weightedInputDerivative(), and shark::NormalizedKernel< InputType >::weightedParameterDerivative().
DataView<DatasetType>::batch_type shark::subBatch | ( | DataView< DatasetType > const & | view, |
IndexRange const & | indizes | ||
) |
Creates a batch given a set of indices.
view | the view from which the batch is to be created |
indizes | the set of indizes defining the batch |
Definition at line 244 of file DataView.h.
References createBatch(), and subset().
Referenced by createCVFullyIndexed(), createCVIndexed(), and randomSubBatch().
DataView<DatasetType> shark::subset | ( | DataView< DatasetType > const & | view, |
IndexRange const & | indizes | ||
) |
creates a subset of a DataView with elements indexed by indices
view | the view for which the subset is to be created |
indizes | the index of the elements to be stored in the view |
Definition at line 222 of file DataView.h.
Referenced by shark::LooError< ModelTypeT, LabelType >::eval(), indexedSubset(), main(), randomSubset(), shark::KernelExpansion< RealVector >::sparsify(), and subBatch().
double shark::sumOfWeights | ( | WeightedUnlabeledData< InputType > const & | dataset | ) |
Definition at line 558 of file WeightedDataset.h.
References shark::blas::sum().
double shark::sumOfWeights | ( | WeightedLabeledData< InputType, LabelType > const & | dataset | ) |
Definition at line 566 of file WeightedDataset.h.
References shark::blas::sum().
void shark::swap | ( | KeyValuePair< K, V > & | pair1, |
KeyValuePair< K, V > & | pair2 | ||
) |
Swaps the contents of two instances of KeyValuePair.
Definition at line 72 of file KeyValuePair.h.
References shark::KeyValuePair< Key, Value >::key, and shark::KeyValuePair< Key, Value >::value.
Referenced by shark::SvmShrinkingProblem< Problem >::applyStep(), shark::blas::approxsolveSymmPosDefSystem(), shark::blas::approxsolveSymmPosDefSystemInPlace(), compare_fitness(), createCVFullyIndexed(), createCVIndexed(), shark::QpMcBoxDecomp< Matrix >::deactivateExample(), shark::QpMcSimplexDecomp< Matrix >::deactivateExample(), shark::QpMcDecomp< Matrix >::deactivateExample(), shark::QpMcBoxDecomp< Matrix >::deactivateVariable(), shark::QpMcSimplexDecomp< Matrix >::deactivateVariable(), shark::QpMcDecomp< Matrix >::deactivateVariable(), shark::MultiChainApproximator< MarkovChainType >::evalDerivative(), shark::ExampleModifiedKernelMatrix< InputType, CacheType >::flipColumnsAndRows(), shark::BlockMatrix2x2< Matrix >::flipColumnsAndRows(), shark::RegularizedKernelMatrix< InputType, CacheType >::flipColumnsAndRows(), shark::GaussianKernelMatrix< T, CacheType >::flipColumnsAndRows(), shark::KernelMatrix< InputType, CacheType >::flipColumnsAndRows(), shark::ModifiedKernelMatrix< InputType, CacheType >::flipColumnsAndRows(), shark::CachedMatrix< Matrix >::flipColumnsAndRows(), shark::GeneralQuadraticProblem< MatrixT >::flipCoordinates(), shark::BoxedSVMProblem< MatrixT >::flipCoordinates(), shark::BoxConstrainedProblem< Problem >::flipCoordinates(), shark::CSVMProblem< MatrixT >::flipCoordinates(), shark::SvmProblem< Problem >::flipCoordinates(), shark::BoxConstrainedShrinkingProblem< Problem >::flipCoordinates(), main(), makeKeyValuePair(), shark::WS2MaximumGradientCriterion::operator()(), shark::SimulatedBinaryCrossover< RealVector >::operator()(), shark::HMGSelectionCriterion::operator()(), shark::ConcatenatedModel< InputType, OutputType >::operator=(), shark::MultiChainApproximator< MarkovChainType >::setData(), shark::NoisyErrorFunction::setRegularizer(), shuffle(), shark::QpBoxLinear< InputT >::solve(), shark::QpMcLinear< InputT >::solve(), shark::QpBoxLinear< CompressedRealVector >::solve(), shark::KernelExpansion< RealVector >::sparsify(), shark::blas::triangular_matrix< T, Orientation, TriangularType >::swap(), shark::blas::matrix< double, blas::column_major >::swap(), shark::blas::compressed_vector< T, I >::swap(), shark::blas::matrix_set< RealMatrix >::swap(), shark::blas::compressed_matrix< T, I >::swap(), shark::blas::swap_rows(), shark::blas::swap_rows_inverted(), shark::LRUCache< QpFloatType >::swapLineIndices(), and shark::LassoRegression< InputVectorType >::trainInternal().
DataView<T>::dataset_type shark::toDataset | ( | DataView< T > const & | view, |
std::size_t | batchSize = DataView<T>::dataset_type::DefaultBatchSize |
||
) |
Creates a new dataset from a View.
When the elements of a View needs to be processed repeatedly it is often better to use the packed format of the Dataset again, since then the faster batch processing can be used
view | the view from which to create the new dataset |
batchSize | the size of the batches in the dataset |
Definition at line 289 of file DataView.h.
References shark::DataView< DatasetType >::dataset(), size(), and shark::DataView< DatasetType >::size().
Referenced by shark::LooError< ModelTypeT, LabelType >::eval(), main(), and shark::KernelExpansion< RealVector >::sparsify().
DataView<DatasetType> shark::toView | ( | DatasetType & | set | ) |
Creates a View from a dataset.
This is just a helper function to omit the actual type of the view
set | the dataset from which to create the view |
Definition at line 276 of file DataView.h.
Referenced by shark::KernelExpansion< RealVector >::sparsify().
double shark::uni | ( | RngType & | rng, |
double | lower, | ||
double | upper | ||
) |
Draws a number uniformly in [lower,upper] by drawing random numbers from rng.
Definition at line 123 of file Uniform.h.
Referenced by shark::PointSearch::configure(), createCVBatch(), shark::Chessboard::draw(), shark::Wave::draw(), shark::CircleInSquare::draw(), shark::DiagonalWithCircle::draw(), shark::ContrastiveDivergence< Operator >::evalDerivative(), shark::BoxConstraintHandler< SearchPointType >::generateRandomPoint(), shark::VDCMA::init(), shark::MarkovChain< Operator >::initializeChain(), shark::TemperedMarkovChain< Operator >::initializeChain(), initializeFFNet(), initializeWeights(), initRandomUniform(), kMeans(), main(), shark::RouletteWheelSelection::operator()(), shark::PolynomialMutator::operator()(), shark::SimulatedBinaryCrossover< RealVector >::operator()(), shark::Weibull_distribution< RealType >::operator()(), shark::Sampler< Rng >::operator()(), shark::HypervolumeApproximator< Rng >::operator()(), shark::Erlang_distribution< RealType >::operator()(), partial_shuffle(), shark::DiffPowers::proposeStartingPoint(), shark::Cigar::proposeStartingPoint(), shark::Himmelblau::proposeStartingPoint(), shark::Ackley::proposeStartingPoint(), shark::Discus::proposeStartingPoint(), shark::Ellipsoid::proposeStartingPoint(), shark::CigarDiscus::proposeStartingPoint(), shark::Rosenbrock::proposeStartingPoint(), shark::CIGTAB2::proposeStartingPoint(), shark::ELLI1::proposeStartingPoint(), shark::ELLI2::proposeStartingPoint(), shark::CIGTAB1::proposeStartingPoint(), shark::UnlabeledData< InputType >::shuffle(), shark::LabeledData< InputType, LabelType >::shuffle(), shark::QpBoxLinear< InputT >::solve(), shark::QpMcLinear< InputT >::solve(), shark::QpBoxLinear< CompressedRealVector >::solve(), and trainProblem().
VectorType shark::variance | ( | const Data< VectorType > & | data | ) |
Calculates the variance vector of the input vectors.
Referenced by getSamples(), mean(), shark::statistics::Variance::statistics(), shark::NormalTrainer::train(), and shark::NormalizeComponentsUnitVariance< DataType >::train().
KeyValueRange<Iterator1,Iterator2> shark::zipKeyValuePairs | ( | Iterator1 | begin1, |
Iterator1 | end1, | ||
Iterator2 | begin2, | ||
Iterator2 | end2 | ||
) |
Zips two ranges together, interpreting the first range as Key which can be sorted.
begin1 | beginning of first range |
end1 | end of first range |
begin2 | beginning of second range |
end2 | end of second range |
Definition at line 184 of file KeyValuePair.h.
Referenced by shark::BinaryTree< VectorType >::splitList().
KeyValueRange< typename boost::range_iterator<Range1>::type, typename boost::range_iterator<Range2>::type> shark::zipKeyValuePairs | ( | Range1 & | range1, |
Range2 & | range2 | ||
) |
Zips two ranges together, interpreting the first range as Key which can be sorted.
range1 | The Key range |
range2 | The value range |
Definition at line 198 of file KeyValuePair.h.
boost::iterator_range< PairIterator<PairType,Iterator1,Iterator2>> shark::zipPairRange | ( | Iterator1 | begin1, |
Iterator1 | end1, | ||
Iterator2 | begin2, | ||
Iterator2 | end2 | ||
) |
PairRangeType<PairType, Range1, Range2>::type shark::zipPairRange | ( | Range1 & | range1, |
Range2 & | range2 | ||
) |
PairRangeType<PairType, Range1 const, Range2 const>::type shark::zipPairRange | ( | Range1 const & | range1, |
Range2 const & | range2 | ||
) |