SHOGUN  3.2.1
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Macros Groups Pages
Namespaces | Classes | Typedefs | Enumerations | Functions | Variables
shogun Namespace Reference

all of classes and functions are contained in the shogun namespace More...

Namespaces

namespace  linalg

Classes

class  DynArray
 Template Dynamic array class that creates an array that can be used like a list or an array. More...
class  Parallel
 Class Parallel provides helper functions for multithreading. More...
struct  TParameter
 parameter struct More...
class  Parameter
 Parameter class. More...
class  SGParamInfo
 Class that holds informations about a certain parameter of an CSGObject. Contains name, type, etc. This is used for mapping types that have changed in different versions of shogun. Instances of this class may be compared to each other. Ordering is based on name, equalness is based on all attributes. More...
class  ParameterMapElement
 Class to hold instances of a parameter map. Each element contains a key and a set of values, which each are of type SGParamInfo. May be compared to each other based on their keys. More...
class  ParameterMap
 Implements a map of ParameterMapElement instances Maps one key to a set of values. More...
class  SGDynamicRefObjectArray
 Dynamic array class for CRefObject pointers that creates an array that can be used like a list or an array. More...
class  CSGObject
 Class SGObject is the base class of all shogun objects. More...
class  SGRefObject
 Class SGRefObject is a reference count based memory management class. More...
class  Version
 Class Version provides version information. More...
class  CAveragedPerceptron
 Class Averaged Perceptron implements the standard linear (online) algorithm. Averaged perceptron is the simple extension of Perceptron. More...
class  CFeatureBlockLogisticRegression
 class FeatureBlockLogisticRegression, a linear binary logistic loss classifier for problems with complex feature relations. Currently two feature relations are supported - feature group (done via CIndexBlockGroup) and feature tree (done via CIndexTree). Handling of feature relations is done via L1/Lq (for groups) and L1/L2 (for trees) regularization. More...
class  CGaussianProcessClassification
 Class GaussianProcessClassification implements binary and multiclass classification based on Gaussian Processes. More...
class  CLDA
 Class LDA implements regularized Linear Discriminant Analysis. More...
class  CLPBoost
 Class LPBoost trains a linear classifier called Linear Programming Machine, i.e. a SVM using a \(\ell_1\) norm regularizer. More...
class  CLPM
 Class LPM trains a linear classifier called Linear Programming Machine, i.e. a SVM using a \(\ell_1\) norm regularizer. More...
class  CMKL
 Multiple Kernel Learning. More...
class  CMKLClassification
 Multiple Kernel Learning for two-class-classification. More...
class  CMKLMulticlass
 MKLMulticlass is a class for L1-norm Multiclass MKL. More...
class  MKLMulticlassGLPK
 MKLMulticlassGLPK is a helper class for MKLMulticlass. More...
class  MKLMulticlassGradient
 MKLMulticlassGradient is a helper class for MKLMulticlass. More...
class  MKLMulticlassOptimizationBase
 MKLMulticlassOptimizationBase is a helper class for MKLMulticlass. More...
class  CMKLOneClass
 Multiple Kernel Learning for one-class-classification. More...
class  CNearestCentroid
 Class NearestCentroid, an implementation of Nearest Shrunk Centroid classifier. More...
class  CPerceptron
 Class Perceptron implements the standard linear (online) perceptron. More...
class  CPluginEstimate
 class PluginEstimate More...
class  CCPLEXSVM
 CplexSVM a SVM solver implementation based on cplex (unfinished). More...
class  CGNPPLib
 class GNPPLib, a Library of solvers for Generalized Nearest Point Problem (GNPP). More...
class  CGNPPSVM
 class GNPPSVM More...
class  CGPBTSVM
 class GPBTSVM More...
class  CLibLinear
 This class provides an interface to the LibLinear library for large- scale linear learning focusing on SVM [1]. This is the classification interface. For regression, see CLibLinearRegression. There is also an online version, see COnlineLibLinear. More...
class  CLibSVM
 LibSVM. More...
class  CLibSVMOneClass
 class LibSVMOneClass More...
class  CMPDSVM
 class MPDSVM More...
class  CNewtonSVM
 NewtonSVM, In this Implementation linear SVM is trained in its primal form using Newton-like iterations. This Implementation is ported from the Olivier Chapelles fast newton based SVM solver, Which could be found here :http://mloss.org/software/view/30/ For further information on this implementation of SVM refer to this paper: http://www.kyb.mpg.de/publications/attachments/neco_%5B0%5D.pdf. More...
class  COnlineLibLinear
 Class implementing a purely online version of CLibLinear, using the L2R_L1LOSS_SVC_DUAL solver only. More...
class  COnlineSVMSGD
 class OnlineSVMSGD More...
class  CQPBSVMLib
 class QPBSVMLib More...
class  CSGDQN
 class SGDQN More...
class  CSVM
 A generic Support Vector Machine Interface. More...
class  CSVMLight
 class SVMlight More...
class  CSVMLightOneClass
 Trains a one class C SVM. More...
class  CSVMLin
 class SVMLin More...
class  CSVMOcas
 class SVMOcas More...
class  CSVMSGD
 class SVMSGD More...
class  CWDSVMOcas
 class WDSVMOcas More...
class  CVwCacheReader
 Base class from which all cache readers for VW should be derived. More...
class  CVwCacheWriter
 CVwCacheWriter is the base class for all VW cache creating classes. More...
class  CVwNativeCacheReader
 Class CVwNativeCacheReader reads from a cache exactly as that which has been produced by VW's default cache format. More...
class  CVwNativeCacheWriter
 Class CVwNativeCacheWriter writes a cache exactly as that which would be produced by VW's default cache format. More...
class  CVwAdaptiveLearner
 VwAdaptiveLearner uses an adaptive subgradient technique to update weights. More...
class  CVwNonAdaptiveLearner
 VwNonAdaptiveLearner uses a standard gradient descent weight update rule. More...
class  CVowpalWabbit
 Class CVowpalWabbit is the implementation of the online learning algorithm used in Vowpal Wabbit. More...
class  VwFeature
 One feature in VW. More...
class  VwExample
 Example class for VW. More...
class  VwLabel
 Class VwLabel holds a label object used by VW. More...
class  CVwEnvironment
 Class CVwEnvironment is the environment used by VW. More...
class  CVwLearner
 Base class for all VW learners. More...
class  CVwParser
 CVwParser is the object which provides the functions to parse examples from buffered input. More...
class  CVwRegressor
 Regressor used by VW. More...
class  CGMM
 Gaussian Mixture Model interface. More...
class  CHierarchical
 Agglomerative hierarchical single linkage clustering. More...
class  CKMeans
 KMeans clustering, partitions the data into k (a-priori specified) clusters. More...
class  CKMeansLloydImpl
class  CKMeansMiniBatchImpl
class  CConverter
 class Converter used to convert data More...
class  CDiffusionMaps
 class DiffusionMaps used to preprocess given data using Diffusion Maps dimensionality reduction technique as described in More...
class  CEmbeddingConverter
 class EmbeddingConverter (part of the Efficient Dimensionality Reduction Toolkit) used to construct embeddings of features, e.g. construct dense numeric embedding of string features More...
class  CFactorAnalysis
 class Factor Analysis used to embed data using Factor Analysis algorithm. More...
class  CHashedDocConverter
 This class can be used to convert a document collection contained in a CStringFeatures<char> object where each document is stored as a single vector into a hashed Bag-of-Words representation. Like in the standard Bag-of-Words representation, this class considers each document as a collection of tokens, which are then hashed into a new feature space of a specified dimension. This class is very flexible and allows the user to specify the tokenizer used to tokenize each document, specify whether the results should be normalized with regards to the sqrt of the document size, as well as to specify whether he wants to combine different tokens. The latter implements a k-skip n-grams approach, meaning that you can combine up to n tokens, while skipping up to k. Eg. for the tokens ["a", "b", "c", "d"], with n_grams = 2 and skips = 2, one would get the following combinations : ["a", "ab", "ac" (skipped 1), "ad" (skipped 2), "b", "bc", "bd" (skipped 1), "c", "cd", "d"]. More...
class  CHessianLocallyLinearEmbedding
 class HessianLocallyLinearEmbedding used to preprocess data using Hessian Locally Linear Embedding algorithm as described in More...
class  CFastICA
 class FastICA More...
class  CFFSep
 class FFSep More...
class  CICAConverter
 class ICAConverter Base class for ICA algorithms More...
class  CJade
 class Jade More...
class  CJediSep
 class JediSep More...
class  CSOBI
 class SOBI More...
class  CUWedgeSep
 class UWedgeSep More...
class  CIsomap
 class Isomap used to embed data using Isomap algorithm as described in More...
class  CKernelLocallyLinearEmbedding
 class KernelLocallyLinearEmbedding used to construct embeddings of data using kernel formulation of Locally Linear Embedding algorithm as described in More...
class  CLaplacianEigenmaps
 class LaplacianEigenmaps used to construct embeddings of data using Laplacian Eigenmaps algorithm as described in: More...
class  CLinearLocalTangentSpaceAlignment
 class LinearLocalTangentSpaceAlignment converter used to construct embeddings as described in: More...
class  CLocalityPreservingProjections
 class LocalityPreservingProjections used to compute embeddings of data using Locality Preserving Projections method as described in More...
class  CLocallyLinearEmbedding
 class LocallyLinearEmbedding used to embed data using Locally Linear Embedding algorithm described in More...
class  CLocalTangentSpaceAlignment
 class LocalTangentSpaceAlignment used to embed data using Local Tangent Space Alignment (LTSA) algorithm as described in: More...
class  CManifoldSculpting
 class CManifoldSculpting used to embed data using manifold sculpting embedding algorithm. More...
class  CMultidimensionalScaling
 class Multidimensionalscaling is used to perform multidimensional scaling (capable of landmark approximation if requested). More...
class  CNeighborhoodPreservingEmbedding
 NeighborhoodPreservingEmbedding converter used to construct embeddings as described in: More...
class  CStochasticProximityEmbedding
 class StochasticProximityEmbedding used to construct embeddings of data using the Stochastic Proximity algorithm. More...
class  CTDistributedStochasticNeighborEmbedding
 class CTDistributedStochasticNeighborEmbedding used to embed data using t-distributed stochastic neighbor embedding algorithm: http://jmlr.csail.mit.edu/papers/volume9/vandermaaten08a/vandermaaten08a.pdf. More...
class  CAttenuatedEuclideanDistance
 class AttenuatedEuclideanDistance More...
class  CBrayCurtisDistance
 class Bray-Curtis distance More...
class  CCanberraMetric
 class CanberraMetric More...
class  CCanberraWordDistance
 class CanberraWordDistance More...
class  CChebyshewMetric
 class ChebyshewMetric More...
class  CChiSquareDistance
 class ChiSquareDistance More...
class  CCosineDistance
 class CosineDistance More...
class  CCustomDistance
 The Custom Distance allows for custom user provided distance matrices. More...
class  CCustomMahalanobisDistance
 Class CustomMahalanobisDistance used to compute the distance between feature vectors \( \vec{x_i} \) and \( \vec{x_j} \) as \( (\vec{x_i} - \vec{x_j})^T \mathbf{M} (\vec{x_i} - \vec{x_j}) \), given the matrix \( \mathbf{M} \) which will be referred to as Mahalanobis matrix. More...
class  CDenseDistance
 template class DenseDistance More...
class  CDistance
 Class Distance, a base class for all the distances used in the Shogun toolbox. More...
class  CEuclideanDistance
 class EuclideanDistance More...
class  CGeodesicMetric
 class GeodesicMetric More...
class  CHammingWordDistance
 class HammingWordDistance More...
class  CJensenMetric
 class JensenMetric More...
class  CKernelDistance
 The Kernel distance takes a distance as input. More...
class  CMahalanobisDistance
 class MahalanobisDistance More...
class  CManhattanMetric
 class ManhattanMetric More...
class  CManhattanWordDistance
 class ManhattanWordDistance More...
class  CMinkowskiMetric
 class MinkowskiMetric More...
class  CRealDistance
 class RealDistance More...
class  CSparseDistance
 template class SparseDistance More...
class  CSparseEuclideanDistance
 class SparseEucldeanDistance More...
class  CStringDistance
 template class StringDistance More...
class  CTanimotoDistance
 class Tanimoto coefficient More...
class  CGaussianDistribution
 Dense version of the well-known Gaussian probability distribution, defined as

\[ \mathcal{N}_x(\mu,\Sigma)= \frac{1}{\sqrt{|2\pi\Sigma|}} \exp\left(-\frac{1}{2}(x-\mu)^T\Sigma^{-1}(x-\mu)\right) \]

. More...

class  CProbabilityDistribution
 A base class for representing n-dimensional probability distribution over the real numbers (64bit) for which various statistics can be computed and which can be sampled. More...
class  CDiscreteDistribution
 This is the base interface class for all discrete distributions. More...
class  CDistribution
 Base class Distribution from which all methods implementing a distribution are derived. More...
class  CEMBase
 This is the base class for Expectation Maximization (EM). EM for various purposes can be derived from this base class. This is a template class having a template member called data which can be used to store all parameters used and results calculated by the expectation and maximization steps of EM. More...
class  CEMMixtureModel
 This is the implementation of EM specialized for Mixture models. More...
class  CGaussian
 Gaussian distribution interface. More...
class  CGHMM
 class GHMM - this class is non-functional and was meant to implement a Generalize Hidden Markov Model (aka Semi Hidden Markov HMM). More...
class  CHistogram
 Class Histogram computes a histogram over all 16bit unsigned integers in the features. More...
class  Model
 class Model More...
class  CHMM
 Hidden Markov Model. More...
class  CKernelDensity
 This class implements the kernel density estimation technique. Kernel density estimation is a non-parametric way to estimate an unknown pdf. The pdf at a query point given finite training samples is calculated using the following formula : \ \(pdf(x')= \frac{1}{nh} \sum_{i=1}^n K(\frac{||x-x_i||}{h})\) \ K() in the above formula is called the kernel function and is controlled by the parameter h called kernel bandwidth. Presently, this class supports only Gaussian kernel which can be used with either Euclidean distance or Manhattan distance. This class makes use of 2 tree structures KD-tree and Ball tree for fast calculation. KD-trees are faster than ball trees at lower dimensions. In case of high dimensional data, ball tree tends to out-perform KD-tree. By default, the class used is Ball tree. More...
class  CLinearHMM
 The class LinearHMM is for learning Higher Order Markov chains. More...
struct  MixModelData
 This structure is used for storing data required for using the generic Expectation Maximization (EM) implemented by the template class CEMBase for mixture models like gaussian mixture model, multinomial mixture model etc. The EM specialized for mixture models is implemented by the class CEMMixtureModel which uses this MixModelData structure. More...
class  CMixtureModel
 This is the generic class for mixture models. The final distribution is a mixture of various simple distributions supplied by the user. More...
class  CPositionalPWM
 Positional PWM. More...
class  CCombinationRule
 CombinationRule abstract class The CombinationRule defines an interface to how to combine the classification or regression outputs of an ensemble of Machines. More...
class  CMajorityVote
 CMajorityVote is a CWeightedMajorityVote combiner, where each Machine's weight in the ensemble is 1.0. More...
class  CMeanRule
 CMeanRule simply averages the outputs of the Machines in the ensemble. More...
class  CWeightedMajorityVote
 Weighted Majority Vote implementation. More...
class  CBinaryClassEvaluation
 The class TwoClassEvaluation, a base class used to evaluate binary classification labels. More...
class  CClusteringAccuracy
 clustering accuracy More...
class  CClusteringEvaluation
 The base class used to evaluate clustering. More...
class  CClusteringMutualInformation
 clustering (normalized) mutual information More...
class  CContingencyTableEvaluation
 The class ContingencyTableEvaluation a base class used to evaluate 2-class classification with TP, FP, TN, FN rates. More...
class  CAccuracyMeasure
 class AccuracyMeasure used to measure accuracy of 2-class classifier. More...
class  CErrorRateMeasure
 class ErrorRateMeasure used to measure error rate of 2-class classifier. More...
class  CBALMeasure
 class BALMeasure used to measure balanced error of 2-class classifier. More...
class  CWRACCMeasure
 class WRACCMeasure used to measure weighted relative accuracy of 2-class classifier. More...
class  CF1Measure
 class F1Measure used to measure F1 score of 2-class classifier. More...
class  CCrossCorrelationMeasure
 class CrossCorrelationMeasure used to measure cross correlation coefficient of 2-class classifier. More...
class  CRecallMeasure
 class RecallMeasure used to measure recall of 2-class classifier. More...
class  CPrecisionMeasure
 class PrecisionMeasure used to measure precision of 2-class classifier. More...
class  CSpecificityMeasure
 class SpecificityMeasure used to measure specificity of 2-class classifier. More...
class  CCrossValidationResult
 type to encapsulate the results of an evaluation run. May contain confidence interval (if conf_int_alpha!=0). m_conf_int_alpha is the probability for an error, i.e. the value does not lie in the confidence interval. More...
class  CCrossValidation
 base class for cross-validation evaluation. Given a learning machine, a splitting strategy, an evaluation criterium, features and correspnding labels, this provides an interface for cross-validation. Results may be retrieved using the evaluate method. A number of repetitions may be specified for obtaining more accurate results. The arithmetic mean of different runs is returned along with confidence intervals, if a p-value is specified. Default number of runs is one, confidence interval combutation is disabled. More...
class  CCrossValidationMKLStorage
 Class for storing MKL weights in every fold of cross-validation. More...
class  CCrossValidationMulticlassStorage
 Class for storing multiclass evaluation information in every fold of cross-validation. More...
class  CCrossValidationOutput
 Class for managing individual folds in cross-validation. More...
class  CCrossValidationPrintOutput
 Class for outputting cross-validation intermediate results to the standard output. Simply prints all messages it gets. More...
class  CCrossValidationSplitting
 Implementation of normal cross-validation on the base of CSplittingStrategy. Produces subset index sets of equal size (at most one difference) More...
class  CDifferentiableFunction
 An abstract class that describes a differentiable function used for GradientEvaluation. More...
class  CEvaluation
 Class Evaluation, a base class for other classes used to evaluate labels, e.g. accuracy of classification or mean squared error of regression. More...
class  CEvaluationResult
 Abstract class that contains the result generated by the MachineEvaluation class. More...
class  CGradientCriterion
 Simple class which specifies the direction of gradient search. More...
class  CGradientEvaluation
 Class evaluates a machine using its associated differentiable function for the function value and its gradient with respect to parameters. More...
class  CGradientResult
 Container class that returns results from GradientEvaluation. It contains the function value as well as its gradient. More...
class  CLOOCrossValidationSplitting
 Implementation of Leave one out cross-validation on the base of CCrossValidationSplitting. Produces subset index sets consisting of one element,for each label. More...
class  CMachineEvaluation
 Machine Evaluation is an abstract class that evaluates a machine according to some criterion. More...
class  CMeanAbsoluteError
 Class MeanAbsoluteError used to compute an error of regression model. More...
class  CMeanSquaredError
 Class MeanSquaredError used to compute an error of regression model. More...
class  CMeanSquaredLogError
 Class CMeanSquaredLogError used to compute an error of regression model. More...
class  CMulticlassAccuracy
 The class MulticlassAccuracy used to compute accuracy of multiclass classification. More...
class  CMulticlassOVREvaluation
 The class MulticlassOVREvaluation used to compute evaluation parameters of multiclass classification via binary OvR decomposition and given binary evaluation technique. More...
class  CMultilabelAccuracy
 Class CMultilabelAccuracy used to compute accuracy of multilabel classification. More...
class  CPRCEvaluation
 Class PRCEvaluation used to evaluate PRC (Precision Recall Curve) and an area under PRC curve (auPRC). More...
class  CROCEvaluation
 Class ROCEvalution used to evaluate ROC (Receiver Operating Characteristic) and an area under ROC curve (auROC). More...
class  CSplittingStrategy
 Abstract base class for all splitting types. Takes a CLabels instance and generates a desired number of subsets which are being accessed by their indices via the method generate_subset_indices(...). More...
class  CStratifiedCrossValidationSplitting
 Implementation of stratified cross-validation on the base of CSplittingStrategy. Produces subset index sets of equal size (at most one difference) in which the label ratio is equal (at most one difference) to the label ratio of the specified labels. Do not use for regression since it may be impossible to distribute nice in that case. More...
class  CStructuredAccuracy
 class CStructuredAccuracy used to compute accuracy of structured classification More...
class  CAlphabet
 The class Alphabet implements an alphabet and alphabet utility functions. More...
class  CAttributeFeatures
 Implements attributed features, that is in the simplest case a number of (attribute, value) pairs. More...
class  CBinnedDotFeatures
 The class BinnedDotFeatures contains a 0-1 conversion of features into bins. More...
class  CCombinedDotFeatures
 Features that allow stacking of a number of DotFeatures. More...
class  CCombinedFeatures
 The class CombinedFeatures is used to combine a number of of feature objects into a single CombinedFeatures object. More...
class  CDataGenerator
 Class that is able to generate various data samples, which may be used for examples in SHOGUN. More...
class  CDenseFeatures
 The class DenseFeatures implements dense feature matrices. More...
class  CDenseSubsetFeatures
class  CDotFeatures
 Features that support dot products among other operations. More...
class  CDummyFeatures
 The class DummyFeatures implements features that only know the number of feature objects (but don't actually contain any). More...
class  CExplicitSpecFeatures
 Features that compute the Spectrum Kernel feature space explicitly. More...
class  CFactorGraphFeatures
 CFactorGraphFeatures maintains an array of factor graphs, each graph is a sample, i.e. an instance of structured input. More...
class  CFeatures
 The class Features is the base class of all feature objects. More...
class  CFKFeatures
 The class FKFeatures implements Fischer kernel features obtained from two Hidden Markov models. More...
class  CHashedDenseFeatures
 This class is identical to the CDenseFeatures class except that it hashes each dimension to a new feature space. More...
class  CHashedDocDotFeatures
 This class can be used to provide on-the-fly vectorization of a document collection. Like in the standard Bag-of-Words representation, this class considers each document as a collection of tokens, which are then hashed into a new feature space of a specified dimension. This class is very flexible and allows the user to specify the tokenizer used to tokenize each document, specify whether the results should be normalized with regards to the sqrt of the document size, as well as to specify whether he wants to combine different tokens. The latter implements a k-skip n-grams approach, meaning that you can combine up to n tokens, while skipping up to k. Eg. for the tokens ["a", "b", "c", "d"], with n_grams = 2 and skips = 2, one would get the following combinations : ["a", "ab", "ac" (skipped 1), "ad" (skipped 2), "b", "bc", "bd" (skipped 1), "c", "cd", "d"]. More...
class  CHashedSparseFeatures
 This class is identical to the CDenseFeatures class except that it hashes each dimension to a new feature space. More...
class  CHashedWDFeatures
 Features that compute the Weighted Degreee Kernel feature space explicitly. More...
class  CHashedWDFeaturesTransposed
 Features that compute the Weighted Degreee Kernel feature space explicitly. More...
class  CImplicitWeightedSpecFeatures
 Features that compute the Weighted Spectrum Kernel feature space explicitly. More...
class  CIndexFeatures
 The class IndexFeatures implements features that contain the index of the features. This features used in the CCustomKernel::init to make the subset of the kernel matrix. Initial CIndexFeature of row_idx and col_idx, pass them to the CCustomKernel::init(row_idx, col_idx), then use CCustomKernel::get_kernel_matrix() will get the sub kernel matrix specified by the row_idx and col_idx. More...
class  CLatentFeatures
 Latent Features class The class if for representing features for latent learning, e.g. LatentSVM. It's basically a very generic way of storing features of any (user-defined) form based on CData. More...
class  CLBPPyrDotFeatures
 Implements Local Binary Patterns with Scale Pyramids as dot features for a set of images. Expects the images to be loaded in a CDenseFeatures object. More...
class  CMatrixFeatures
 Class CMatrixFeatures used to represent data whose feature vectors are better represented with matrices rather than with unidimensional arrays or vectors. Optionally, it can be restricted that all the feature vectors have the same number of features. Set the attribute num_features different to zero to use this restriction. Allow feature vectors with different number of features by setting num_features equal to zero (default behaviour). More...
class  CPolyFeatures
 implement DotFeatures for the polynomial kernel More...
class  CRandomFourierDotFeatures
 This class implements the random fourier features for the DotFeatures framework. Basically upon the object creation it computes the random coefficients, namely w and b, that are needed for this method and then every time a vector is required it is computed based on the following formula z(x) = sqrt(2/D) * cos(w'*x + b), where D is the number of samples that are used. More...
class  CRandomKitchenSinksDotFeatures
 class that implements the Random Kitchen Sinks (RKS) for the DotFeatures as mentioned in http://books.nips.cc/papers/files/nips21/NIPS2008_0885.pdf. More...
class  CRealFileFeatures
 The class RealFileFeatures implements a dense double-precision floating point matrix from a file. More...
class  CSNPFeatures
 Features that compute the Weighted Degreee Kernel feature space explicitly. More...
class  CSparseFeatures
 Template class SparseFeatures implements sparse matrices. More...
class  CSparsePolyFeatures
 implement DotFeatures for the polynomial kernel More...
class  CGaussianBlobsDataGenerator
class  CMeanShiftDataGenerator
class  CStreamingDenseFeatures
 This class implements streaming features with dense feature vectors. More...
class  CStreamingDotFeatures
 Streaming features that support dot products among other operations. More...
class  CStreamingFeatures
 Streaming features are features which are used for online algorithms. More...
class  CStreamingHashedDenseFeatures
 This class acts as an alternative to the CStreamingDenseFeatures class and their difference is that the current example in this class is hashed into a smaller dimension dim. More...
class  CStreamingHashedDocDotFeatures
 This class implements streaming features for a document collection. Like in the standard Bag-of-Words representation, this class considers each document as a collection of tokens, which are then hashed into a new feature space of a specified dimension. This class is very flexible and allows the user to specify the tokenizer used to tokenize each document, specify whether the results should be normalized with regards to the sqrt of the document size, as well as to specify whether he wants to combine different tokens. The latter implements a k-skip n-grams approach, meaning that you can combine up to n tokens, while skipping up to k. Eg. for the tokens ["a", "b", "c", "d"], with n_grams = 2 and skips = 2, one would get the following combinations : ["a", "ab", "ac" (skipped 1), "ad" (skipped 2), "b", "bc", "bd" (skipped 1), "c", "cd", "d"]. More...
class  CStreamingHashedSparseFeatures
 This class acts as an alternative to the CStreamingSparseFeatures class and their difference is that the current example in this class is hashed into a smaller dimension dim. More...
class  CStreamingSparseFeatures
 This class implements streaming features with sparse feature vectors. The vector is represented as an SGSparseVector<T>. Each entry is of type SGSparseVectorEntry<T> with members `feat_index' and `entry'. More...
class  CStreamingStringFeatures
 This class implements streaming features as strings. More...
class  CStreamingVwFeatures
 This class implements streaming features for use with VW. More...
class  CStringFeatures
 Template class StringFeatures implements a list of strings. More...
class  CStringFileFeatures
 File based string features. More...
class  CSubset
 Wrapper class for an index subset which is used by SubsetStack. More...
class  CSubsetStack
 class to add subset support to another class. A CSubsetStackStack instance should be added and wrapper methods to all interfaces should be added. More...
class  CTOPFeatures
 The class TOPFeatures implements TOP kernel features obtained from two Hidden Markov models. More...
class  CWDFeatures
 Features that compute the Weighted Degreee Kernel feature space explicitly. More...
class  CBinaryFile
 A Binary file access class. More...
class  CBinaryStream
 memory mapped emulation via binary streams (files) More...
class  CCSVFile
 Class CSVFile used to read data from comma-separated values (CSV) files. See http://en.wikipedia.org/wiki/Comma-separated_values. More...
class  CFile
 A File access base class. More...
class  CIOBuffer
 An I/O buffer class. More...
class  CLibSVMFile
 read sparse real valued features in svm light format e.g. -1 1:10.0 2:100.2 1000:1.3 with -1 == (optional) label and dim 1 - value 10.0 dim 2 - value 100.2 dim 1000 - value 1.3 More...
class  CLineReader
 Class for buffered reading from a ascii file. More...
class  CMemoryMappedFile
 memory mapped file More...
class  CParser
 Class for reading from a string. More...
class  CProtobufFile
 Class for work with binary file in protobuf format. More...
class  CSerializableAsciiFile
 serializable ascii file More...
class  SerializableAsciiReader00
 Serializable ascii reader. More...
class  CSerializableFile
 serializable file More...
struct  substring
 struct Substring, specified by start position and end position. More...
class  SGIO
 Class SGIO, used to do input output operations throughout shogun. More...
class  CSimpleFile
 Template class SimpleFile to read and write from files. More...
class  CStreamingAsciiFile
 Class StreamingAsciiFile to read vector-by-vector from ASCII files. More...
class  CStreamingFile
 A Streaming File access class. More...
class  CStreamingFileFromDenseFeatures
 Class CStreamingFileFromDenseFeatures is a derived class of CStreamingFile which creates an input source for the online framework from a CDenseFeatures object. More...
class  CStreamingFileFromFeatures
 Class StreamingFileFromFeatures to read vector-by-vector from a CFeatures object. More...
class  CStreamingFileFromSparseFeatures
 Class CStreamingFileFromSparseFeatures is derived from CStreamingFile and provides an input source for the online framework. It uses an existing CSparseFeatures object to generate online examples. More...
class  CStreamingFileFromStringFeatures
 Class CStreamingFileFromStringFeatures is derived from CStreamingFile and provides an input source for the online framework from a CStringFeatures object. More...
class  CStreamingVwCacheFile
 Class StreamingVwCacheFile to read vector-by-vector from VW cache files. More...
class  CStreamingVwFile
 Class StreamingVwFile to read vector-by-vector from Vowpal Wabbit data files. It reads the example and label into one object of VwExample type. More...
class  CUAIFile
 Class UAIFILE used to read data from UAI files. See http://graphmod.ics.uci.edu/uai08/FileFormat for more details. More...
class  CANOVAKernel
 ANOVA (ANalysis Of VAriances) kernel. More...
class  CAUCKernel
 The AUC kernel can be used to maximize the area under the receiver operator characteristic curve (AUC) instead of margin in SVM training. More...
class  CBesselKernel
 the class Bessel kernel More...
class  CCauchyKernel
 Cauchy kernel. More...
class  CChi2Kernel
 The Chi2 kernel operating on realvalued vectors computes the chi-squared distance between sets of histograms. More...
class  CCircularKernel
 Circular kernel. More...
class  CCombinedKernel
 The Combined kernel is used to combine a number of kernels into a single CombinedKernel object by linear combination. More...
class  CConstKernel
 The Constant Kernel returns a constant for all elements. More...
class  CCustomKernel
 The Custom Kernel allows for custom user provided kernel matrices. More...
class  CDiagKernel
 The Diagonal Kernel returns a constant for the diagonal and zero otherwise. More...
class  CDistanceKernel
 The Distance kernel takes a distance as input. More...
class  CDotKernel
 Template class DotKernel is the base class for kernels working on DotFeatures. More...
class  CExponentialKernel
 The Exponential Kernel, closely related to the Gaussian Kernel computed on CDotFeatures. More...
class  CGaussianARDKernel
 Gaussian Kernel with Automatic Relevance Detection. More...
class  CGaussianKernel
 The well known Gaussian kernel (swiss army knife for SVMs) computed on CDotFeatures. More...
class  CGaussianShiftKernel
 An experimental kernel inspired by the WeightedDegreePositionStringKernel and the Gaussian kernel. More...
class  CGaussianShortRealKernel
 The well known Gaussian kernel (swiss army knife for SVMs) on dense short-real valued features. More...
class  CHistogramIntersectionKernel
 The HistogramIntersection kernel operating on realvalued vectors computes the histogram intersection distance between sets of histograms. Note: the current implementation assumes positive values for the histograms, and input vectors should sum to 1. More...
class  CInverseMultiQuadricKernel
 InverseMultiQuadricKernel. More...
class  CJensenShannonKernel
 The Jensen-Shannon kernel operating on real-valued vectors computes the Jensen-Shannon distance between the features. Often used in computer vision. More...
struct  K_THREAD_PARAM
class  CKernel
 The Kernel base class. More...
class  CLinearARDKernel
 Linear Kernel with Automatic Relevance Detection. More...
class  CLinearKernel
 Computes the standard linear kernel on CDotFeatures. More...
class  CLogKernel
 Log kernel. More...
class  CMultiquadricKernel
 MultiquadricKernel. More...
class  CAvgDiagKernelNormalizer
 Normalize the kernel by either a constant or the average value of the diagonal elements (depending on argument c of the constructor). More...
class  CDiceKernelNormalizer
 DiceKernelNormalizer performs kernel normalization inspired by the Dice coefficient (see http://en.wikipedia.org/wiki/Dice's_coefficient) More...
class  CFirstElementKernelNormalizer
 Normalize the kernel by a constant obtained from the first element of the kernel matrix, i.e. \( c=k({\bf x},{\bf x})\). More...
class  CIdentityKernelNormalizer
 Identity Kernel Normalization, i.e. no normalization is applied. More...
class  CKernelNormalizer
 The class Kernel Normalizer defines a function to post-process kernel values. More...
class  CRidgeKernelNormalizer
 Normalize the kernel by adding a constant term to its diagonal. This aids kernels to become positive definite (even though they are not - often caused by numerical problems). More...
class  CScatterKernelNormalizer
 the scatter kernel normalizer More...
class  CSqrtDiagKernelNormalizer
 SqrtDiagKernelNormalizer divides by the Square Root of the product of the diagonal elements. More...
class  CTanimotoKernelNormalizer
 TanimotoKernelNormalizer performs kernel normalization inspired by the Tanimoto coefficient (see http://en.wikipedia.org/wiki/Jaccard_index ) More...
class  CVarianceKernelNormalizer
 VarianceKernelNormalizer divides by the ``variance''. More...
class  CZeroMeanCenterKernelNormalizer
 ZeroMeanCenterKernelNormalizer centers the kernel in feature space. More...
class  CPolyKernel
 Computes the standard polynomial kernel on CDotFeatures. More...
class  CPowerKernel
 Power kernel. More...
class  CProductKernel
 The Product kernel is used to combine a number of kernels into a single ProductKernel object by element multiplication. More...
class  CPyramidChi2
 Pyramid Kernel over Chi2 matched histograms. More...
class  CRationalQuadraticKernel
 Rational Quadratic kernel. More...
class  CSigmoidKernel
 The standard Sigmoid kernel computed on dense real valued features. More...
class  CSparseKernel
 Template class SparseKernel, is the base class of kernels working on sparse features. More...
class  CSphericalKernel
 Spherical kernel. More...
class  CSplineKernel
 Computes the Spline Kernel function which is the cubic polynomial. More...
class  CCommUlongStringKernel
 The CommUlongString kernel may be used to compute the spectrum kernel from strings that have been mapped into unsigned 64bit integers. More...
class  CCommWordStringKernel
 The CommWordString kernel may be used to compute the spectrum kernel from strings that have been mapped into unsigned 16bit integers. More...
class  CDistantSegmentsKernel
 The distant segments kernel is a string kernel, which counts the number of substrings, so-called segments, at a certain distance from each other. More...
class  CFixedDegreeStringKernel
 The FixedDegree String kernel takes as input two strings of same size and counts the number of matches of length d. More...
class  CGaussianMatchStringKernel
 The class GaussianMatchStringKernel computes a variant of the Gaussian kernel on strings of same length. More...
class  CHistogramWordStringKernel
 The HistogramWordString computes the TOP kernel on inhomogeneous Markov Chains. More...
class  CLinearStringKernel
 Computes the standard linear kernel on dense char valued features. More...
class  CLocalAlignmentStringKernel
 The LocalAlignmentString kernel compares two sequences through all possible local alignments between the two sequences. More...
class  CLocalityImprovedStringKernel
 The LocalityImprovedString kernel is inspired by the polynomial kernel. Comparing neighboring characters it puts emphasize on local features. More...
class  CMatchWordStringKernel
 The class MatchWordStringKernel computes a variant of the polynomial kernel on strings of same length converted to a word alphabet. More...
class  COligoStringKernel
 This class offers access to the Oligo Kernel introduced by Meinicke et al. in 2004. More...
class  CPolyMatchStringKernel
 The class PolyMatchStringKernel computes a variant of the polynomial kernel on strings of same length. More...
class  CPolyMatchWordStringKernel
 The class PolyMatchWordStringKernel computes a variant of the polynomial kernel on word-features. More...
class  CRegulatoryModulesStringKernel
 The Regulaty Modules kernel, based on the WD kernel, as published in Schultheiss et al., Bioinformatics (2009) on regulatory sequences. More...
class  CSalzbergWordStringKernel
 The SalzbergWordString kernel implements the Salzberg kernel. More...
class  CSimpleLocalityImprovedStringKernel
 SimpleLocalityImprovedString kernel, is a ``simplified'' and better performing version of the Locality improved kernel. More...
class  CSNPStringKernel
 The class SNPStringKernel computes a variant of the polynomial kernel on strings of same length. More...
struct  SSKFeatures
 SSKFeatures. More...
class  CSparseSpatialSampleStringKernel
 Sparse Spatial Sample String Kernel by Pavel Kuksa pkuks.nosp@m.a@cs.nosp@m..rutg.nosp@m.ers..nosp@m.edu and Vladimir Pavlovic vladi.nosp@m.mir@.nosp@m.cs.ru.nosp@m.tger.nosp@m.s.edu More...
class  CSpectrumMismatchRBFKernel
 spectrum mismatch rbf kernel More...
class  CSpectrumRBFKernel
 spectrum rbf kernel More...
class  CStringKernel
 Template class StringKernel, is the base class of all String Kernels. More...
class  CSubsequenceStringKernel
 class SubsequenceStringKernel that implements String Subsequence Kernel (SSK) discussed by Lodhi et. al.[1]. A subsequence is any ordered sequence of \(n\) characters occurring in the text, though not necessarily contiguous. More formally, string \(u\) is a subsequence of string \(s\), iff there exists indices \(\mathbf{i}=(i_{1},\dots,i_{|u|})\), with \(1\le i_{1} \le \cdots \le i_{|u|} \le |s|\), such that \(u_{j}=s_{i_{j}}\) for \(j=1,\dots,|u|\), written as \(u=s[\mathbf{i}]\). The feature mapping \(\phi\) in this scenario is given by

\[ \phi_{u}(s)=\sum_{\mathbf{i}:u=s[\mathbf{i}]}\lambda^{l(\mathbf{i})} \]

for some \(lambda\le 1\), where \(l(\mathbf{i})\) is the length of the subsequence in \(s\), given by \(i_{|u|}-i_{1}+1\). The kernel here is an inner product in the feature space generated by all subsequences of length \(n\).

\[ K_{n}(s,t)=\sum_{u\in\Sigma^{n}}\langle \phi_{u}(s), \phi_{u}(t)\rangle = \sum_{u\in\Sigma^{n}}\sum_{\mathbf{i}:u=s[\mathbf{i}]} \sum_{\mathbf{j}:u=t[\mathbf{j}]}\lambda^{l(\mathbf{i})+l(\mathbf{j})} \]

Since the subsequences are weighted by the exponentially decaying factor \(\lambda\) of their full length in the text, more weight is given to those occurrences that are nearly contiguous. A direct computation is infeasible since the dimension of the feature space grows exponentially with \(n\). The paper describes an efficient computation approach using a dynamic programming technique. More...

class  CWeightedCommWordStringKernel
 The WeightedCommWordString kernel may be used to compute the weighted spectrum kernel (i.e. a spectrum kernel for 1 to K-mers, where each k-mer length is weighted by some coefficient \(\beta_k\)) from strings that have been mapped into unsigned 16bit integers. More...
class  CWeightedDegreePositionStringKernel
 The Weighted Degree Position String kernel (Weighted Degree kernel with shifts). More...
class  CWeightedDegreeStringKernel
 The Weighted Degree String kernel. More...
class  CTensorProductPairKernel
 Computes the Tensor Product Pair Kernel (TPPK). More...
class  CTStudentKernel
 Generalized T-Student kernel. More...
class  CWaveKernel
 Wave kernel. More...
class  CWaveletKernel
 the class WaveletKernel More...
class  CWeightedDegreeRBFKernel
 weighted degree RBF kernel More...
class  CBinaryLabels
 Binary Labels for binary classification. More...
class  CDenseLabels
 Dense integer or floating point labels. More...
class  CFactorGraphObservation
 Class CFactorGraphObservation is used as the structured output. More...
class  CFactorGraphLabels
 Class FactorGraphLabels used e.g. in the application of Structured Output (SO) learning with the FactorGraphModel. Each of the labels is represented by a graph. Each label is of type CFactorGraphObservation and all of them are stored in a CDynamicObjectArray. More...
class  CLabels
 The class Labels models labels, i.e. class assignments of objects. More...
class  CLabelsFactory
 The helper class to specialize base class instances of labels. More...
class  CLatentLabels
 abstract class for latent labels As latent labels always depends on the given application, this class only defines the API that the user has to implement for latent labels. More...
class  CMulticlassLabels
 Multiclass Labels for multi-class classification. More...
class  CMultilabelLabels
 Multilabel Labels for multi-label classification. More...
class  CRegressionLabels
 Real Labels are real-valued labels. More...
class  CStructuredLabels
 Base class of the labels used in Structured Output (SO) problems. More...
class  CLatentModel
 Abstract class CLatentModel It represents the application specific model and contains most of the application dependent logic to solve latent variable based problems. More...
class  CLatentSOSVM
 class Latent Structured Output SVM, an structured output based machine for classification problems with latent variables. More...
class  CLatentSVM
 LatentSVM class Latent SVM implementation based on [1]. For optimization this implementation uses SVMOcas. More...
class  CBitString
 a string class embedding a string in a compact bit representation More...
class  CCache
 Template class Cache implements a simple cache. More...
class  CCircularBuffer
 Implementation of circular buffer This buffer has logical structure such as queue (FIFO). But this queue is cyclic: tape, ends of which are connected, just instead tape there is block of physical memory. So, if you push big block of data it can be situated both at the end and the begin of buffer's memory. More...
class  CCompressor
 Compression library for compressing and decompressing buffers using one of the standard compression algorithms: More...
class  CJobResultAggregator
 Abstract base class that provides an interface for computing an aggeregation of the job results of independent computation jobs as they are submitted and also for finalizing the aggregation. More...
class  CStoreScalarAggregator
 Template class that aggregates scalar job results in each submit_result call, finalize then transforms current aggregation into a CScalarResult. More...
class  CStoreVectorAggregator
 Abstract template class that aggregates vector job results in each submit_result call, finalize is abstract. More...
class  CIndependentComputationEngine
 Abstract base class for solving multiple independent instances of CIndependentJob. It has one method, submit_job, which may add the job to an internal queue and might block if there is yet not space in the queue. After jobs are submitted, it might not yet be ready. wait_for_all waits until all jobs are completed, which must be called to guarantee that all jobs are finished. More...
class  CSerialComputationEngine
 Class that computes multiple independent instances of computation jobs sequentially. More...
class  CIndependentJob
 Abstract base for general computation jobs to be registered in CIndependentComputationEngine. compute method produces a job result and submits it to the internal JobResultAggregator. Each set of jobs that form a result will share the same job result aggregator. More...
class  CJobResult
 Base class that stores the result of an independent job. More...
class  CScalarResult
 Base class that stores the result of an independent job when the result is a scalar. More...
class  CVectorResult
 Base class that stores the result of an independent job when the result is a vector. More...
class  CData
 dummy data holder More...
struct  TSGDataType
 Datatypes that shogun supports. More...
class  CDelimiterTokenizer
 The class CDelimiterTokenizer is used to tokenize a SGVector<char> into tokens using custom chars as delimiters. One can set the delimiters to use by setting to 1 the appropriate index of the public field delimiters. Eg. to set as delimiter the character ':', one should do: tokenizer->delimiters[':'] = 1;. More...
class  CDynamicArray
 Template Dynamic array class that creates an array that can be used like a list or an array. More...
class  CDynamicObjectArray
 Dynamic array class for CSGObject pointers that creates an array that can be used like a list or an array. More...
class  CDynInt
 integer type of dynamic size More...
class  CGCArray
 Template class GCArray implements a garbage collecting static array. More...
class  CHash
 Collection of Hashing Functions. More...
class  CIndexBlock
 class IndexBlock used to represent contiguous indices of one group (e.g. block of related features) More...
class  CIndexBlockGroup
 class IndexBlockGroup used to represent group-based feature relation. More...
class  CIndexBlockRelation
 class IndexBlockRelation More...
class  CIndexBlockTree
 class IndexBlockTree used to represent tree guided feature relation. More...
class  CIndirectObject
 an array class that accesses elements indirectly via an index array. More...
class  v_array
 Class v_array taken directly from JL's implementation. More...
class  CJLCoverTreePoint
 Class Point to use with John Langford's CoverTree. This class must have some assoficated functions defined (distance, parse_points and print, see below) so it can be used with the CoverTree implementation. More...
class  CListElement
 Class ListElement, defines how an element of the the list looks like. More...
class  CList
 Class List implements a doubly connected list for low-level-objects. More...
class  CLock
 Class Lock used for synchronization in concurrent programs. More...
class  CMap
 the class CMap, a map based on the hash-table. w: http://en.wikipedia.org/wiki/Hash_table More...
class  CNGramTokenizer
 The class CNGramTokenizer is used to tokenize a SGVector<char> into n-grams. More...
class  RefCount
class  CSet
 the class CSet, a set based on the hash-table. w: http://en.wikipedia.org/wiki/Hash_table More...
class  SGMatrix
 shogun matrix More...
class  SGMatrixList
 shogun matrix list More...
class  SGNDArray
 shogun n-dimensional array More...
class  SGReferencedData
 shogun reference count managed data More...
class  SGSparseMatrix
 template class SGSparseMatrix More...
struct  SGSparseVectorEntry
 template class SGSparseVectorEntry More...
class  SGSparseVector
 template class SGSparseVector The assumtion is that the stored SGSparseVectorEntry<T>* vector is ordered by SGSparseVectorEntry.feat_index in non-decreasing order. This has to be assured by the user of the class. More...
class  SGString
 shogun string More...
class  SGStringList
 template class SGStringList More...
struct  IndexSorter
class  SGVector
 shogun vector More...
class  ShogunException
 Class ShogunException defines an exception which is thrown whenever an error inside of shogun occurs. More...
class  CSignal
 Class Signal implements signal handling to e.g. allow ctrl+c to cancel a long running process. More...
class  CStructuredData
 Base class of the components of StructuredLabels. More...
class  CTime
 Class Time that implements a stopwatch based on either cpu time or wall clock time. More...
class  CTokenizer
 The class CTokenizer acts as a base class in order to implement tokenizers. Sub-classes must implement the methods has_next(), next_token_idx() and get_copy(). More...
class  CTrie
 Template class Trie implements a suffix trie, i.e. a tree in which all suffixes up to a certain length are stored. More...
class  CAbsoluteDeviationLoss
 CAbsoluteDeviationLoss implements the absolute deviation loss function.
\(L(y_i,f(x_i)) = \mod{y_i-f(x_i)}\). More...
class  CExponentialLoss
 CExponentialLoss implements the exponential loss function.
\(L(y_i,f(x_i)) = \exp^{-y_if(x_i)}\). More...
class  CHingeLoss
 CHingeLoss implements the hinge loss function. More...
class  CHuberLoss
 CHuberLoss implements the Huber loss function. It behaves like SquaredLoss function at values below Huber delta and like absolute deviation at values greater than the delta. More...
class  CLogLoss
 CLogLoss implements the logarithmic loss function. More...
class  CLogLossMargin
 Class CLogLossMargin implements a margin-based log-likelihood loss function. More...
class  CLossFunction
 Class CLossFunction is the base class of all loss functions. More...
class  CSmoothHingeLoss
 CSmoothHingeLoss implements the smooth hinge loss function. More...
class  CSquaredHingeLoss
 Class CSquaredHingeLoss implements a squared hinge loss function. More...
class  CSquaredLoss
 CSquaredLoss implements the squared loss function. More...
class  CBaggingMachine
 : Bagging algorithm i.e. bootstrap aggregating More...
class  CBaseMulticlassMachine
class  CDistanceMachine
 A generic DistanceMachine interface. More...
class  CGaussianProcessMachine
 A base class for Gaussian Processes. More...
class  CConstMean
 The Const mean function class. More...
class  CDualVariationalGaussianLikelihood
 Class that models dual variational likelihood. More...
class  CEPInferenceMethod
 Class of the Expectation Propagation (EP) posterior approximation inference method. More...
class  CExactInferenceMethod
 The Gaussian exact form inference method class. More...
class  CFITCInferenceMethod
 The Fully Independent Conditional Training inference method class. More...
class  CGaussianLikelihood
 Class that models Gaussian likelihood. More...
class  CInferenceMethod
 The Inference Method base class. More...
class  CKLApproxDiagonalInferenceMethod
 The KL approximation inference method class. More...
class  CKLCholeskyInferenceMethod
 The KL approximation inference method class. More...
class  CKLCovarianceInferenceMethod
 The KL approximation inference method class. More...
class  CKLDualInferenceMethod
 The dual KL approximation inference method class. More...
class  CKLInferenceMethod
 The KL approximation inference method class. More...
class  CKLLowerTriangularInferenceMethod
 The KL approximation inference method class. More...
class  CLaplacianInferenceBase
 The Laplace approximation inference method base class. More...
class  CLikelihoodModel
 The Likelihood model base class. More...
class  CLogitDVGLikelihood
 Class that models dual variational logit likelihood. More...
class  CLogitLikelihood
 Class that models Logit likelihood. More...
class  CLogitVGLikelihood
 Class that models Logit likelihood and uses numerical integration to approximate the following variational expection of log likelihood

\[ \sum_{{i=1}^n}{E_{q(f_i|{\mu}_i,{\sigma}^2_i)}[logP(y_i|f_i)]} \]

. More...

class  CLogitVGPiecewiseBoundLikelihood
 Class that models Logit likelihood and uses variational piecewise bound to approximate the following variational expection of log likelihood

\[ \sum_{{i=1}^n}{E_{q(f_i|{\mu}_i,{\sigma}^2_i)}[logP(y_i|f_i)]} \]

where

\[ p(y_i|f_i) = \frac{exp(y_i*f_i)}{1+exp(f_i)}, y_i \in \{0,1\} \]

. More...

class  CMatrixOperations
 The helper class is used for Laplace and KL methods. More...
class  CMeanFunction
 An abstract class of the mean function. More...
class  CMultiLaplacianInferenceMethod
 The Laplace approximation inference method class for multi classification. More...
class  CNumericalVGLikelihood
 Class that models likelihood and uses numerical integration to approximate the following variational expection of log likelihood

\[ \sum_{{i=1}^n}{E_{q(f_i|{\mu}_i,{\sigma}^2_i)}[logP(y_i|f_i)]} \]

. More...

class  CProbitLikelihood
 Class that models Probit likelihood. More...
class  CProbitVGLikelihood
 Class that models Probit likelihood and uses numerical integration to approximate the following variational expection of log likelihood

\[ \sum_{{i=1}^n}{E_{q(f_i|{\mu}_i,{\sigma}^2_i)}[logP(y_i|f_i)]} \]

. More...

class  CSingleLaplacianInferenceMethod
 The SingleLaplace approximation inference method class for regression and binary Classification. More...
class  CSingleLaplacianInferenceMethodWithLBFGS
 The Laplace approximation inference method with LBFGS class for regression and binary classification. More...
class  CSoftMaxLikelihood
 Class that models Soft-Max likelihood. More...
class  CStudentsTLikelihood
 Class that models a Student's-t likelihood. More...
class  CStudentsTVGLikelihood
 Class that models Student's T likelihood and uses numerical integration to approximate the following variational expection of log likelihood

\[ \sum_{{i=1}^n}{E_{q(f_i|{\mu}_i,{\sigma}^2_i)}[logP(y_i|f_i)]} \]

. More...

class  CVariationalGaussianLikelihood
 The variational Gaussian Likelihood base class. The variational distribution is Gaussian. More...
class  CVariationalLikelihood
 The Variational Likelihood base class. More...
class  CZeroMean
 The zero mean function class. More...
class  CKernelMachine
 A generic KernelMachine interface. More...
class  CKernelMulticlassMachine
 generic kernel multiclass More...
class  CKernelStructuredOutputMachine
class  CLinearLatentMachine
 abstract implementaion of Linear Machine with latent variable This is the base implementation of all linear machines with latent variable. More...
class  CLinearMachine
 Class LinearMachine is a generic interface for all kinds of linear machines like classifiers. More...
class  CLinearMulticlassMachine
 generic linear multiclass machine More...
class  CLinearStructuredOutputMachine
class  CMachine
 A generic learning machine interface. More...
class  CMulticlassMachine
 experimental abstract generic multiclass machine class More...
class  CNativeMulticlassMachine
 experimental abstract native multiclass machine class More...
class  COnlineLinearMachine
 Class OnlineLinearMachine is a generic interface for linear machines like classifiers which work through online algorithms. More...
class  CRandomForest
 This class implements the Random Forests algorithm. In Random Forests algorithm, we train a number of randomized CART trees (see class CRandomCARTree) using the supplied training data. The number of trees to be trained is a parameter (called number of bags) controlled by the user. Test feature vectors are classified/regressed by combining the outputs of all these trained candidate trees using a combination rule (see class CCombinationRule). The feature for calculating out-of-box error is also provided to help determine the appropriate number of bags. The evaluatin criteria for calculating this out-of-box error is specified by the user (see class CEvaluation). More...
class  CStochasticGBMachine
 This class implements the stochastic gradient boosting algorithm for ensemble learning invented by Jerome H. Friedman. This class works with a variety of loss functions like squared loss, exponential loss, Huber loss etc which can be accessed through Shogun's CLossFunction interface (cf. http://www.shogun-toolbox.org/doc/en/latest/classshogun_1_1CLossFunction.html). Additionally, it can create an ensemble of any regressor class derived from the CMachine class (cf. http://www.shogun-toolbox.org/doc/en/latest/classshogun_1_1CMachine.html). For one dimensional optimization, this class uses the backtracking linesearch accessed via Shogun's L-BFGS class. A concise description of the algorithm implemented can be found in the following link : http://en.wikipedia.org/wiki/Gradient_boosting#Algorithm. More...
class  CStructuredOutputMachine
class  CApproxJointDiagonalizer
 Class ApproxJointDiagonalizer defines an Approximate Joint Diagonalizer (AJD) interface. More...
class  CFFDiag
 Class FFDiag. More...
class  CJADiag
 Class JADiag. More...
class  CJADiagOrth
 Class JADiagOrth. More...
class  CJediDiag
 Class Jedi. More...
class  CQDiag
 Class QDiag. More...
class  CUWedge
 Class UWedge. More...
class  CCplex
 Class CCplex to encapsulate access to the commercial cplex general purpose optimizer. More...
class  EigenSparseUtil
 This class contains some utilities for Eigen3 Sparse Matrix integration with shogun. Currently it provides a method for converting SGSparseMatrix to Eigen3 SparseMatrix. More...
class  CFunction
 Class of a function of one variable. More...
class  CIntegration
 Class that contains certain methods related to numerical integration. More...
class  CJacobiEllipticFunctions
 Class that contains methods for computing Jacobi elliptic functions related to complex analysis. These functions are inverse of the elliptic integral of first kind, i.e.

\[ u(k,m)=\int_{0}^{k}\frac{dt}{\sqrt{(1-t^{2})(1-m^{2}t^{2})}} =\int_{0}^{\varphi}\frac{d\theta}{\sqrt{(1-m^{2}sin^{2}\theta)}} \]

where \(k=sin\varphi\), \(t=sin\theta\) and parameter \(m, 0\le m \le 1\) is called modulus. Three main Jacobi elliptic functions are defined as \(sn(u,m)=k=sin\theta\), \(cn(u,m)=cos\theta=\sqrt{1-sn(u,m)^{2}}\) and \(dn(u,m)=\sqrt{1-m^{2}sn(u,m)^{2}}\). For \(k=1\), i.e. \(\varphi=\frac{\pi}{2}\), \(u(1,m)=K(m)\) is known as the complete elliptic integral of first kind. Similarly, \(u(1,m'))= K'(m')\), \(m'=\sqrt{1-m^{2}}\) is called the complementary complete elliptic integral of first kind. Jacobi functions are double periodic with quardratic periods \(K\) and \(K'\). More...

class  VectorDotProduct
 Abstract template base class for vector dot product computers. More...
class  CDirectEigenSolver
 Class that computes eigenvalues of a real valued, self-adjoint dense matrix linear operator using Eigen3. More...
class  CEigenSolver
 Abstract base class that provides an abstract compute method for computing eigenvalues of a real valued, self-adjoint linear operator. It also provides method for getting min and max eigenvalues. More...
class  CLanczosEigenSolver
 Class that computes eigenvalues of a real valued, self-adjoint linear operator using Lanczos algorithm. More...
class  CDenseMatrixOperator
 Class that represents a dense-matrix linear operator. It computes matrix-vector product \(Ax\) in its apply method, \(A\in\mathbb{C}^{m\times n},A:\mathbb{C}^{n}\rightarrow \mathbb{C}^{m}\) being the matrix operator and \(x\in\mathbb{C}^{n}\) being the vector. The result is a vector \(y\in\mathbb{C}^{m}\). More...
class  CLinearOperator
 Abstract template base class that represents a linear operator, e.g. a matrix. More...
class  CMatrixOperator
 Abstract base class that represents a matrix linear operator. It provides an interface to computes matrix-vector product \(Ax\) in its apply method, \(A\in\mathbb{C}^{m\times n},A:\mathbb{C}^{n} \rightarrow \mathbb{C}^{m}\) being the matrix operator and \(x\in \mathbb{C}^{n}\) being the vector. The result is a vector \(y\in \mathbb{C}^{m}\). More...
struct  SparsityStructure
 Struct that represents the sparsity structure of the Sparse Matrix in CRS. Implementation has been adapted from Krylstat (https://github.com/ Froskekongen/KRYLSTAT) library (c) Erlend Aune erlen.nosp@m.da@m.nosp@m.ath.n.nosp@m.tnu..nosp@m.no under GPL2+. More...
class  CSparseMatrixOperator
 Class that represents a sparse-matrix linear operator. It computes matrix-vector product \(Ax\) in its apply method, \(A\in\mathbb{C}^{m\times n},A:\mathbb{C}^{n}\rightarrow \mathbb{C}^{m}\) being the matrix operator and \(x\in\mathbb{C}^{n}\) being the vector. The result is a vector \(y\in\mathbb{C}^{m}\). More...
class  CCGMShiftedFamilySolver
 class that uses conjugate gradient method for solving a shifted linear system family where the linear opeator is real valued and symmetric positive definite, the vector is real valued, but the shifts are complex More...
class  CConjugateGradientSolver
 class that uses conjugate gradient method of solving a linear system involving a real valued linear operator and vector. Useful for large sparse systems involving sparse symmetric and positive-definite matrices. More...
class  CConjugateOrthogonalCGSolver
 class that uses conjugate orthogonal conjugate gradient method of solving a linear system involving a complex valued linear operator and vector. Useful for large sparse systems involving sparse symmetric matrices that are not Herimitian. More...
class  CDirectLinearSolverComplex
 Class that provides a solve method for complex dense-matrix linear systems. More...
class  CDirectSparseLinearSolver
 Class that provides a solve method for real sparse-matrix linear systems using LLT. More...
class  CIterativeLinearSolver
 abstract template base for all iterative linear solvers such as conjugate gradient (CG) solvers. provides interface for setting the iteration limit, relative/absolute tolerence. solve method is abstract. More...
class  CIterativeShiftedLinearFamilySolver
 abstract template base for CG based solvers to the solution of shifted linear systems of the form \((A+\sigma)x=b\) for several values of \(\sigma\) simultaneously, using only as many matrix-vector operations as the solution of a single system requires. This class adds another interface to the basic iterative linear solver that takes the shifts, \(\sigma\), and also weights, \(\alpha\), and returns the summation \(\sum_{i} \alpha_{i}x_{i}\), where \(x_{i}\) is the solution of the system \((A+\sigma_{i})x_{i}=b\). More...
struct  _IterInfo
 struct that contains current state of the iteration for iterative linear solvers More...
class  IterativeSolverIterator
 template class that is used as an iterator for an iterative linear solver. In the iteration of solving phase, each solver initializes the iteration with a maximum number of iteration limit, and relative/ absolute tolerence. They then call begin with the residual vector and continue until its end returns true, i.e. either it has converged or iteration count reached maximum limit. More...
class  CLinearSolver
 Abstract template base class that provides an abstract solve method for linear systems, that takes a linear operator \(A\), a vector \(b\), solves the system \(Ax=b\) and returns the vector \(x\). More...
class  CIndividualJobResultAggregator
 Class that aggregates vector job results in each submit_result call of jobs generated from rational approximation of linear operator function times a vector. finalize extracts the imaginary part of that aggregation, applies the linear operator to the aggregation, performs a dot product with the sample vector, multiplies with the constant multiplier (see CRationalApproximation) and stores the result as CScalarResult. More...
class  CDenseExactLogJob
 Class that represents the job of applying the log of a CDenseMatrixOperator on a real vector. More...
class  CRationalApproximationCGMJob
 Implementation of independent jobs that solves one whole family of shifted systems in rational approximation of linear operator function times a vector using CG-M linear solver. compute calls submit_results of the aggregator with CScalarResult (see CRationalApproximation) More...
class  CRationalApproximationIndividualJob
 Implementation of independent job that solves one of the family of shifted systems in rational approximation of linear operator function times a vector using a direct linear solver. The shift is moved inside the operator. compute calls submit_results of the aggregator with CVectorResult which is the solution vector for that shift multiplied by complex weight (See CRationalApproximation) More...
class  CLogDetEstimator
 Class to create unbiased estimators of \(log(\left|C\right|)= trace(log(C))\). For each estimate, it samples trace vectors (one by one) and calls submit_jobs of COperatorFunction, stores the resulting job result aggregator instances, calls wait_for_all of CIndependentComputationEngine to ensure that the job result aggregators are all up to date. Then simply computes running averages over the estimates. More...
class  CDenseMatrixExactLog
 Class that generates jobs for computing logarithm of a dense matrix linear operator. More...
class  CLogRationalApproximationCGM
 Implementaion of rational approximation of a operator-function times vector where the operator function is log of a linear operator. Each complex system generated from the shifts due to rational approximation of opertor- log times vector expression are solved at once with a shifted linear-family solver by the computation engine. generate_jobs generates one job per sample. More...
class  CLogRationalApproximationIndividual
 Implementaion of rational approximation of a operator-function times vector where the operator function is log of a dense-matrix. Each complex system generated from the shifts due to rational approximation of opertor- log times vector expression are solved individually with a complex linear solver by the computation engine. generate_jobs generates num_shifts number of jobs per trace sample. More...
class  COperatorFunction
 Abstract template base class for computing \(s^{T} f(C) s\) for a linear operator C and a vector s. submit_jobs method creates a bunch of jobs needed to solve for this particular \(s\) and attaches one unique job aggregator to each of them, then submits them all to the computation engine. More...
class  CRationalApproximation
 Abstract base class of the rational approximation of a function of a linear operator (A) times vector (v) using Cauchy's integral formula -

\[f(\text{A})\text{v}=\oint_{\Gamma}f(z)(z\text{I}-\text{A})^{-1} \text{v}dz\]

Computes eigenvalues of linear operator and uses Jacobi elliptic functions and conformal maps [2] for quadrature rule for discretizing the contour integral and computes complex shifts, weights and constant multiplier of the rational approximation of the above expression as

\[f(\text{A})\text{v}\approx \eta\text{A}\Im-\left(\sum_{l=1}^{N}\alpha_{l} (\text{A}-\sigma_{l}\text{I})^{-1}\text{v}\right)\]

where \(\alpha_{l},\sigma_{l}\in\mathbb{C}\) are respectively the shifts and weights of the linear systems generated from the rational approximation, and \(\eta\in\mathbb{R}\) is the constant multiplier, equals to \(\frac{-8K(\lambda_{m}\lambda_{M})^{\frac{1}{4}}}{k\pi N}\). More...

class  CNormalSampler
 Class that provides a sample method for Gaussian samples. More...
class  CTraceSampler
 Abstract template base class that provides an interface for sampling the trace of a linear operator using an abstract sample method. More...
class  CLoss
 Class which collects generic mathematical functions. More...
class  CMath
 Class which collects generic mathematical functions. More...
class  Munkres
 Munkres. More...
class  CRandom
 : Pseudo random number geneartor More...
class  CSparseInverseCovariance
 used to estimate inverse covariance matrix using graphical lasso More...
class  CStatistics
 Class that contains certain functions related to statistics, such as probability/cumulative distribution functions, different statistics, etc. More...
class  CLMNN
 Class LMNN that implements the distance metric learning technique Large Margin Nearest Neighbour (LMNN) described in. More...
class  CLMNNStatistics
 Class LMNNStatistics used to give access to intermediate results obtained training LMNN. More...
class  CGradientModelSelection
 Model selection class which searches for the best model by a gradient-search. More...
class  CGridSearchModelSelection
 Model selection class which searches for the best model by a grid- search. See CModelSelection for details. More...
class  CModelSelection
 Abstract base class for model selection. More...
class  CModelSelectionParameters
 Class to select parameters and their ranges for model selection. The structure is organized as a tree with different kinds of nodes, depending on the values of its member variables of name and CSGObject. More...
class  CParameterCombination
 Class that holds ONE combination of parameters for a learning machine. The structure is organized as a tree. Every node may hold a name or an instance of a Parameter class. Nodes may have children. The nodes are organized in such way, that every parameter of a model for model selection has one node and sub-parameters are stored in sub-nodes. Using a tree of this class, parameters of models may easily be set. There are these types of nodes: More...
class  CRandomSearchModelSelection
 Model selection class which searches for the best model by a random search. See CModelSelection for details. More...
class  CECOCAEDDecoder
class  CECOCDecoder
class  CECOCDiscriminantEncoder
class  CECOCEDDecoder
class  CECOCEncoder
 ECOCEncoder produce an ECOC codebook. More...
class  CECOCForestEncoder
class  CECOCHDDecoder
class  CECOCIHDDecoder
class  CECOCLLBDecoder
class  CECOCOVOEncoder
class  CECOCOVREncoder
class  CECOCRandomDenseEncoder
class  CECOCRandomSparseEncoder
class  CECOCSimpleDecoder
class  CECOCStrategy
class  CECOCUtil
class  CGaussianNaiveBayes
 Class GaussianNaiveBayes, a Gaussian Naive Bayes classifier. More...
class  CGMNPLib
 class GMNPLib Library of solvers for Generalized Minimal Norm Problem (GMNP). More...
class  CGMNPSVM
 Class GMNPSVM implements a one vs. rest MultiClass SVM. More...
class  CKNN
 Class KNN, an implementation of the standard k-nearest neigbor classifier. More...
class  CLaRank
 the LaRank multiclass SVM machine More...
class  CMCLDA
 Class MCLDA implements multiclass Linear Discriminant Analysis. More...
class  CMulticlassLibLinear
 multiclass LibLinear wrapper. Uses Crammer-Singer formulation and gradient descent optimization algorithm implemented in the LibLinear library. Regularized bias support is added using stacking bias 'feature' to hyperplanes normal vectors. More...
class  CMulticlassLibSVM
 class LibSVMMultiClass. Does one vs one classification. More...
class  CMulticlassLogisticRegression
 multiclass logistic regression More...
class  CMulticlassOCAS
 multiclass OCAS wrapper More...
class  CMulticlassOneVsOneStrategy
 multiclass one vs one strategy used to train generic multiclass machines for K-class problems with building voting-based ensemble of K*(K-1) binary classifiers multiclass probabilistic outputs can be obtained by using the heuristics described in [1] More...
class  CMulticlassOneVsRestStrategy
 multiclass one vs rest strategy used to train generic multiclass machines for K-class problems with building ensemble of K binary classifiers More...
class  CMulticlassStrategy
 class MulticlassStrategy used to construct generic multiclass classifiers with ensembles of binary classifiers More...
class  CMulticlassSVM
 class MultiClassSVM More...
class  CMulticlassTreeGuidedLogisticRegression
 multiclass tree guided logistic regression More...
class  CQDA
 Class QDA implements Quadratic Discriminant Analysis. More...
class  CRejectionStrategy
 base rejection strategy class More...
class  CThresholdRejectionStrategy
 threshold based rejection strategy More...
class  CDixonQTestRejectionStrategy
 simplified version of Dixon's Q test outlier based rejection strategy. Statistic values are taken from http://www.vias.org/tmdatanaleng/cc_outlier_tests_dixon.html More...
class  CScatterSVM
 ScatterSVM - Multiclass SVM. More...
class  CShareBoost
class  ShareBoostOptimizer
class  CBalancedConditionalProbabilityTree
class  CBallTree
 This class implements Ball tree. The ball tree is contructed using the top-down approach. cf. ftp://ftp.icsi.berkeley.edu/pub/techreports/1989/tr-89-063.pdf. More...
class  CBinaryTreeMachineNode
 The node of the tree structure forming a TreeMachine The node contains pointer to its parent and pointers to its 2 children: left child and right child. The node also contains data which can be of any type and has to be specified using template specifier. More...
class  CC45ClassifierTree
 Class C45ClassifierTree implements the C4.5 algorithm for decision tree learning. The algorithm steps are briefy explained below :
. More...
struct  C45TreeNodeData
 structure to store data of a node of C4.5 tree. This can be used as a template type in TreeMachineNode class. Ex: C4.5 algorithm uses nodes of type CTreeMachineNode<C45TreeNodeData> More...
class  CCARTree
 This class implements the Classification And Regression Trees algorithm by Breiman et al for decision tree learning. A CART tree is a binary decision tree that is constructed by splitting a node into two child nodes repeatedly, beginning with the root node that contains the whole dataset.

TREE GROWING PROCESS :
During the tree growing process, we recursively split a node into left child and right child so that the resulting nodes are "purest". We do this until any of the stopping criteria is met. To find the best split, we scan through all possible splits in all predictive attributes. The best split is one that maximises some splitting criterion. For classification tasks, ie. when the dependent attribute is categorical, the Gini index is used. For regression tasks, ie. when the dependent variable is continuous, least squares deviation is used. The algorithm uses two stopping criteria : if node becomes completely "pure", ie. all its members have identical dependent variable, or all of them have identical predictive attributes (independent variables).

. More...
struct  CARTreeNodeData
 structure to store data of a node of CART. This can be used as a template type in TreeMachineNode class. CART algorithm uses nodes of type CTreeMachineNode<CARTreeNodeData> More...
class  CCHAIDTree
 This class implements the CHAID algorithm proposed by Kass (1980) for decision tree learning. CHAID consists of three steps: merging, splitting and stopping. A tree is grown by repeatedly using these three steps on each node starting from the root node. CHAID accepts nominal or ordinal categorical predictors only. If predictors are continuous, they have to be transformed into ordinal predictors before tree growing.

CONVERTING CONTINUOUS PREDICTORS TO ORDINAL :
Continuous predictors are converted to ordinal by binning. The number of bins (K) has to be supplied by the user. Given K, a predictor is split in such a way that all the bins get the same number (more or less) of distinct predictor values. The maximum feature value in each bin is used as a breakpoint.

MERGING :
During the merging step, allowable pairs of categories of a predictor are evaluated for similarity. If the similarity of a pair is above a threshold, the categories constituting the pair are merged into a single category. The process is repeated until there is no pair left having high similarity between its categories. Similarity between categories is evaluated using the p_value

SPLITTING :
The splitting step selects which predictor to be used to best split the node. Selection is accomplished by comparing the adjusted p_value associated with each predictor. The predictor that has the smallest adjusted p_value is chosen for splitting the node.

STOPPING :
The tree growing process stops if any of the following conditions is satisfied :
. More...
struct  CHAIDTreeNodeData
 structure to store data of a node of CHAID. This can be used as a template type in TreeMachineNode class. CHAID algorithm uses nodes of type CTreeMachineNode<CHAIDTreeNodeData> More...
class  CConditionalProbabilityTree
struct  ConditionalProbabilityTreeNodeData
 struct to store data of node of conditional probability tree More...
class  CID3ClassifierTree
 class ID3ClassifierTree, implements classifier tree for discrete feature values using the ID3 algorithm. The training algorithm implemented is as follows : More...
struct  id3TreeNodeData
 structure to store data of a node of id3 tree. This can be used as a template type in TreeMachineNode class. Ex: id3 algorithm uses nodes of type CTreeMachineNode<id3TreeNodeData> More...
class  CKDTree
 This class implements KD-Tree. cf. http://www.autonlab.org/autonweb/14665/version/2/part/5/data/moore-tutorial.pdf. More...
class  CKNNHeap
 This class implements a specialized version of max heap structure. This heap specializes in storing the least k values seen so far along with the indices (or id) of the entities with which the values are associated. On calling the push method, it is automatically checked, if the new value supplied, is among the least k distances seen so far. Also, in case the heap is full already, the max among the stored values is automatically thrown out as the new value finds its proper place in the heap. More...
class  CNbodyTree
 This class implements genaralized tree for N-body problems like k-NN, kernel density estimation, 2 point correlation. More...
struct  NbodyTreeNodeData
 structure to store data of a node of N-Body tree. This can be used as a template type in TreeMachineNode class. N-Body tree building algorithm uses nodes of type CBinaryTreeMachineNode<NbodyTreeNodeData> More...
class  CRandomCARTree
 This class implements randomized CART algorithm used in the tree growing process of candidate trees in Random Forests algorithm. The tree growing process is different from the original CART algorithm because of the input attributes which are considered for each node split. In randomized CART, a few (fixed number) attributes are randomly chosen from all available attributes while deciding the best split. This is unlike the original CART where all available attributes are considered while deciding the best split. More...
class  CRandomConditionalProbabilityTree
class  CRelaxedTree
struct  RelaxedTreeNodeData
class  RelaxedTreeUtil
class  CTreeMachine
 class TreeMachine, a base class for tree based multiclass classifiers. This class is derived from CBaseMulticlassMachine and stores the root node (of class type CTreeMachineNode) to the tree structure More...
class  CTreeMachineNode
 The node of the tree structure forming a TreeMachine The node contains a pointer to its parent and a vector of pointers to its children. A node of this class can have only one parent but any number of children.The node also contains data which can be of any type and has to be specified using template specifier. More...
struct  VwConditionalProbabilityTreeNodeData
class  CVwConditionalProbabilityTree
class  CAutoencoder
 Represents a single layer neural autoencoder. More...
class  CConvolutionalFeatureMap
 Handles convolution and gradient calculation for a single feature map in a convolutional neural network. More...
class  CDeepAutoencoder
 Represents a muti-layer autoencoder. More...
class  CDeepBeliefNetwork
 A Deep Belief Network. More...
class  CNeuralConvolutionalLayer
 Main component in convolutional neural networks More...
class  CNeuralInputLayer
 Represents an input layer. The layer can be either connected to all the input features that a network receives (default) or connected to just a small part of those features. More...
class  CNeuralLayer
 Base class for neural network layers. More...
class  CNeuralLayers
 A class to construct neural layers. More...
class  CNeuralLinearLayer
 Neural layer with linear neurons, with an identity activation function. can be used as a hidden layer or an output layer. More...
class  CNeuralLogisticLayer
 Neural layer with linear neurons, with a logistic activation function. can be used as a hidden layer or an output layer. More...
class  CNeuralNetwork
 A generic multi-layer neural network. More...
class  CNeuralRectifiedLinearLayer
 Neural layer with rectified linear neurons. More...
class  CNeuralSoftmaxLayer
 Neural layer with linear neurons, with a softmax activation function. can be only be used as an output layer. Cross entropy error measure is used. More...
class  CRBM
 A Restricted Boltzmann Machine. More...
struct  tag_callback_data
struct  tag_iteration_data
struct  lbfgs_parameter_t
class  CBAHSIC
 Class CBAHSIC, that extends CKernelDependenceMaximization and uses HSIC [1] to compute dependence measures for feature selection using a backward elimination approach as described in [1]. This class serves as a convenience class that initializes the CDependenceMaximization::m_estimator with an instance of CHSIC and allows only BACKWARD_ELIMINATION algorithm to use which is set internally. Therefore, trying to use other algorithms by set_algorithm() will not work. Plese see the class documentation of CHSIC and [2] for more details on mathematical description of HSIC. More...
class  CDecompressString
 Preprocessor that decompresses compressed strings. More...
class  CDensePreprocessor
 Template class DensePreprocessor, base class for preprocessors (cf. CPreprocessor) that apply to CDenseFeatures (i.e. rectangular dense matrices) More...
class  CDependenceMaximization
 Class CDependenceMaximization, base class for all feature selection preprocessors which select a subset of features that shows maximum dependence between the features and the labels. This is done via an implementation of CIndependenceTest, m_estimator inside compute_measures() (see class documentation of CFeatureSelection), which performs a statistical test for a given feature \(\mathbf{X}_i\) from the set of features \(\mathbf{X}\), and the labels \(\mathbf{Y}\). The test checks

\[ \textbf{H}_0 : P\left(\mathbf{X}\setminus \mathbf{X}_i, \mathbf{Y}\right) =P\left(\mathbf{X}\setminus \mathbf{X}_i\right)P\left(\mathbf{Y}\right) \]

The test statistic is then used as a measure which signifies the independence between the rest of the features and the labels - higher the value of the test statistic, greater the dependency between the rest of the features and the class labels, and therefore lesser significant the current feature becomes. Therefore, highest scoring features are removed. The removal policy thus can only be N_LARGEST and PERCENTILE_LARGEST and it can be set via set_policy() call. remove_feats() method handles the removal of features based on the specified policy. More...

class  CDimensionReductionPreprocessor
 the class DimensionReductionPreprocessor, a base class for preprocessors used to lower the dimensionality of given simple features (dense matrices). More...
class  CFeatureSelection
 Template class CFeatureSelection, base class for all feature selection preprocessors which select a subset of features (dimensions in the feature matrix) to achieve a specified number of dimensions, m_target_dim from a given set of features. This class showcases all feature selection algorithms via a generic interface. Supported algorithms are specified by the enum EFeatureSelectionAlgorithm which can be set via set_algorithm() call. Supported wrapper algorithms are. More...
class  CFisherLDA
 Preprocessor FisherLDA attempts to model the difference between the classes of data by performing linear discriminant analysis on input feature vectors/matrices. When the init method in FisherLDA is called with proper feature matrix X(say N number of vectors and D feature dimensions) supplied via apply_to_feature_matrix or apply_to_feature_vector methods, this creates a transformation whose outputs are the reduced T-Dimensional & class-specific distribution (where T<= number of unique classes-1). The transformation matrix is essentially a DxT matrix, the columns of which correspond to the specified number of eigenvectors which maximizes the ratio of between class matrix to within class matrix. More...
class  CHomogeneousKernelMap
 Preprocessor HomogeneousKernelMap performs homogeneous kernel maps as described in. More...
class  CKernelDependenceMaximization
 Class CKernelDependenceMaximization, that uses an implementation of CKernelIndependenceTest to compute dependence measures for feature selection. Different kernels are used for labels and data. For the sake of computational convenience, the precompute() method is overridden to precompute the kernel for labels and save as an instance of CCustomKernel. More...
class  CKernelPCA
 Preprocessor KernelPCA performs kernel principal component analysis. More...
class  CLogPlusOne
 Preprocessor LogPlusOne does what the name says, it adds one to a dense real valued vector and takes the logarithm of each component of it. More...
class  CNormOne
 Preprocessor NormOne, normalizes vectors to have norm 1. More...
class  CPCA
 Preprocessor PCA performs principial component analysis on input feature vectors/matrices. When the init method in PCA is called with proper feature matrix X (with say N number of vectors and D feature dimension), a transformation matrix is computed and stored internally. This transformation matrix is then used to transform all D-dimensional feature vectors or feature matrices (with D feature dimensions) supplied via apply_to_feature_matrix or apply_to_feature_vector methods. This tranformation outputs the T-Dimensional approximation of all these input vectors and matrices (where T<=min(D,N)). The transformation matrix is essentially a DxT matrix, the columns of which correspond to the eigenvectors of the covariance matrix(XX') having top T eigenvalues. More...
class  CPNorm
 Preprocessor PNorm, normalizes vectors to have p-norm. More...
class  CPreprocessor
 Class Preprocessor defines a preprocessor interface. More...
class  CPruneVarSubMean
 Preprocessor PruneVarSubMean will substract the mean and remove features that have zero variance. More...
class  CRandomFourierGaussPreproc
 Preprocessor CRandomFourierGaussPreproc implements Random Fourier Features for the Gauss kernel a la Ali Rahimi and Ben Recht Nips2007 after preprocessing the features using them in a linear kernel approximates a gaussian kernel. More...
class  CRescaleFeatures
 Preprocessor RescaleFeautres is rescaling the range of features to make the features independent of each other and aims to scale the range in [0, 1] or [-1, 1]. More...
class  CSortUlongString
 Preprocessor SortUlongString, sorts the indivual strings in ascending order. More...
class  CSortWordString
 Preprocessor SortWordString, sorts the indivual strings in ascending order. More...
class  CSparsePreprocessor
 Template class SparsePreprocessor, base class for preprocessors (cf. CPreprocessor) that apply to CSparseFeatures. More...
class  CStringPreprocessor
 Template class StringPreprocessor, base class for preprocessors (cf. CPreprocessor) that apply to CStringFeatures (i.e. strings of variable length). More...
class  CSumOne
 Preprocessor SumOne, normalizes vectors to have sum 1. More...
class  CGaussianProcessRegression
 Class GaussianProcessRegression implements regression based on Gaussian Processes. More...
class  CKernelRidgeRegression
 Class KernelRidgeRegression implements Kernel Ridge Regression - a regularized least square method for classification and regression. More...
class  CLeastAngleRegression
 Class for Least Angle Regression, can be used to solve LASSO. More...
class  CLeastSquaresRegression
 class to perform Least Squares Regression More...
class  CLinearRidgeRegression
 Class LinearRidgeRegression implements Ridge Regression - a regularized least square method for classification and regression. More...
class  CLibLinearRegression
 This class provides an interface to the LibLinear library for large- scale linear learning focusing on SVM [1]. This is the regression interface. For classification, see CLibLinear. More...
class  CLibSVR
 Class LibSVR, performs support vector regression using LibSVM. More...
class  CMKLRegression
 Multiple Kernel Learning for regression. More...
class  CSVRLight
 Class SVRLight, performs support vector regression using SVMLight. More...
class  CHSIC
 This class implements the Hilbert Schmidtd Independence Criterion based independence test as described in [1]. More...
class  CHypothesisTest
 Hypothesis test base class. Provides an interface for statistical hypothesis testing via three methods: compute_statistic(), compute_p_value() and compute_threshold(). The second computes a p-value for the statistic computed by the first method. The p-value represents the position of the statistic in the null-distribution, i.e. the distribution of the statistic population given the null-hypothesis is true. (1-position = p-value). The third method, compute_threshold(), computes a threshold for a given test level which is needed to reject the null-hypothesis. More...
class  CIndependenceTest
 Provides an interface for performing the independence test. Given samples \(Z=\{(x_i,y_i)\}_{i=1}^m\) from the joint distribution \(\textbf{P}_{xy}\), does the joint distribution factorize as \(\textbf{P}_{xy}=\textbf{P}_x\textbf{P}_y\), i.e. product of the marginals? The null-hypothesis says yes, i.e. no dependence, the alternative hypothesis says no. More...
class  CKernelIndependenceTest
 Kernel independence test base class. Provides an interface for performing an independence test. Given samples \(Z=\{(x_i,y_i)\}_{i=1}^m\) from the joint distribution \(\textbf{P}_{xy}\), does the joint distribution factorize as \(\textbf{P}_{xy}=\textbf{P}_x\textbf{P}_y\), i.e. product of the marginals? More...
class  CKernelMeanMatching
 Kernel Mean Matching. More...
class  CKernelSelection
 Base class for kernel selection for kernel two-sample test statistic implementations (e.g. MMD). Provides abstract methods for selecting kernels and computing criteria or kernel weights for the implemented method. In order to implement new methods for kernel selection, simply write a new implementation of this class. More...
class  CKernelTwoSampleTest
 Kernel two sample test base class. Provides an interface for performing a two-sample test using a kernel, i.e. Given samples from two distributions \(p\) and \(q\), the null-hypothesis is: \(H_0: p=q\), the alternative hypothesis: \(H_1: p\neq q\). More...
class  CLinearTimeMMD
 This class implements the linear time Maximum Mean Statistic as described in [1] for streaming data (see CStreamingMMD for description). More...
class  CMMDKernelSelection
 Base class for kernel selection for MMD-based two-sample test statistic implementations. Provides abstract methods for selecting kernels and computing criteria or kernel weights for the implemented method. In order to implement new methods for kernel selection, simply write a new implementation of this class. More...
class  CMMDKernelSelectionComb
 Base class for kernel selection of combined kernels. Given an MMD instance whose underlying kernel is a combined one, this class provides an interface to select weights of this combined kernel. More...
class  CMMDKernelSelectionCombMaxL2
 Implementation of maximum MMD kernel selection for combined kernel. This class selects a combination of baseline kernels that maximises the the MMD for a combined kernel based on a L2-regularization approach. This boils down to solve the convex program

\[ \min_\beta \{\beta^T \beta \quad \text{s.t.}\quad \beta^T \eta=1, \beta\succeq 0\}, \]

where \(\eta\) is a vector whose elements are the MMDs of the baseline kernels. More...

class  CMMDKernelSelectionCombOpt
 Implementation of optimal kernel selection for combined kernel. This class selects a combination of baseline kernels that maximises the ratio of the MMD and its standard deviation for a combined kernel. This boils down to solve the convex program

\[ \min_\beta \{\beta^T (Q+\lambda_m) \beta \quad \text{s.t.}\quad \beta^T \eta=1, \beta\succeq 0\}, \]

where \(\eta\) is a vector whose elements are the MMDs of the baseline kernels and \(Q\) is a linear time estimate of the covariance of \(\eta\). More...

class  CMMDKernelSelectionMax
 Kernel selection class that selects the single kernel that maximises the MMD statistic. Works for CQuadraticTimeMMD and CLinearTimeMMD. This leads to a heuristic that is better than the standard median heuristic for Gaussian kernels. However, it comes with no guarantees. More...
class  CMMDKernelSelectionMedian
 Implements MMD kernel selection for a number of Gaussian baseline kernels via selecting the one with a bandwidth parameter that is closest to the median of all pairwise distances in the underlying data. Therefore, it only works for data to which a GaussianKernel can be applied, which are grouped under the class CDotFeatures in SHOGUN. More...
class  CMMDKernelSelectionOpt
 Implements optimal kernel selection for single kernels. Given a number of baseline kernels, this method selects the one that minimizes the type II error for a given type I error for a two-sample test. This only works for the CLinearTimeMMD statistic. More...
class  CNOCCO
 This class implements the NOrmalized Cross Covariance Operator (NOCCO) based independence test as described in [1]. More...
class  CQuadraticTimeMMD
 This class implements the quadratic time Maximum Mean Statistic as described in [1]. The MMD is the distance of two probability distributions \(p\) and \(q\) in a RKHS which we denote by

\[ \hat{\eta_k}=\text{MMD}[\mathcal{F},p,q]^2=\textbf{E}_{x,x'} \left[ k(x,x')\right]-2\textbf{E}_{x,y}\left[ k(x,y)\right] +\textbf{E}_{y,y'}\left[ k(y,y')\right]=||\mu_p - \mu_q||^2_\mathcal{F} \]

. More...

class  CStreamingMMD
 Abstract base class that provides an interface for performing kernel two-sample test on streaming data using Maximum Mean Discrepancy (MMD) as the test statistic. The MMD is the distance of two probability distributions \(p\) and \(q\) in a RKHS (see [1] for formal description). More...
class  CTwoSampleTest
 Provides an interface for performing the classical two-sample test i.e. Given samples from two distributions \(p\) and \(q\), the null-hypothesis is: \(H_0: p=q\), the alternative hypothesis: \(H_1: p\neq q\). More...
struct  BmrmStatistics
class  CCCSOSVM
 CCSOSVM. More...
class  CDisjointSet
 Class CDisjointSet data structure for linking graph nodes It's easy to identify connected graph, acyclic graph, roots of forest etc. please refer to http://en.wikipedia.org/wiki/Disjoint-set_data_structure. More...
class  CDualLibQPBMSOSVM
 Class DualLibQPBMSOSVM that uses Bundle Methods for Regularized Risk Minimization algorithms for structured output (SO) problems [1] presented in [2]. More...
class  CDynProg
 Dynamic Programming Class. More...
class  CFactorDataSource
 Class CFactorDataSource Source for factor data. In some cases, the same data can be shared by many factors. More...
class  CFactor
 Class CFactor A factor is defined on a clique in the factor graph. Each factor can have its own data, either dense, sparse or shared data. Note that currently this class is table factor oriented. More...
class  CFactorGraph
 Class CFactorGraph a factor graph is a structured input in general. More...
class  CFactorGraphModel
 CFactorGraphModel defines a model in terms of CFactorGraph and CMAPInference, where parameters are associated with factor types, in the model. There is a mapping vector records the locations of local factor parameters in the global parameter vector. More...
class  CFactorType
 Class CFactorType defines the way of factor parameterization. More...
class  CTableFactorType
 Class CTableFactorType the way that store assignments of variables and energies in a table or a multi-array. More...
class  CFWSOSVM
 Class CFWSOSVM solves SOSVM using Frank-Wolfe algorithm [1]. More...
struct  GCEdge
struct  GCNode
struct  GCNodePtr
class  CGraphCut
class  CHashedMultilabelModel
 Class CHashedMultilabelModel represents application specific model and contains application dependent logic for solving multilabel classification with feature hashing within a generic SO framework. We hash the feature of each class with a separate seed and put them in the same feature space (exploded feature space). More...
class  CHierarchicalMultilabelModel
 Class CHierarchicalMultilabelModel represents application specific model and contains application dependent logic for solving hierarchical multilabel classification[1] within a generic SO framework. More...
class  CHMSVMModel
 Class CHMSVMModel that represents the application specific model and contains the application dependent logic to solve Hidden Markov Support Vector Machines (HM-SVM) type of problems within a generic SO framework. More...
class  CIntronList
 class IntronList More...
struct  bmrm_ll
struct  ICP_stats
struct  line_search_res
class  CMAPInference
 Class CMAPInference performs MAP inference on a factor graph. Briefly, given a factor graph model, with features \(\bold{x}\), the prediction is obtained by \( {\arg\max} _{\bold{y}} P(\bold{Y} = \bold{y} | \bold{x}; \bold{w}) \). More...
class  CMAPInferImpl
 Class CMAPInferImpl abstract class of MAP inference implementation. More...
class  CMulticlassModel
 Class CMulticlassModel that represents the application specific model and contains the application dependent logic to solve multiclass classification within a generic SO framework. More...
struct  CRealNumber
 Class CRealNumber to be used in the application of Structured Output (SO) learning to multiclass classification. Even though it is likely that it does not make sense to consider real numbers as structured data, it has been made in this way because the basic type to use in structured labels needs to inherit from CStructuredData. More...
class  CMulticlassSOLabels
 Class CMulticlassSOLabels to be used in the application of Structured Output (SO) learning to multiclass classification. Each of the labels is represented by a real number and it is required that the values of the labels are in the set {0, 1, ..., num_classes-1}. Each label is of type CRealNumber and all of them are stored in a CDynamicObjectArray. More...
class  CMultilabelCLRModel
 Class MultilabelCLRModel represents application specific model and contains application dependent logic for solving multi-label classification using Calibrated Label Ranking (CLR) [1] method within a generic SO framework. More...
class  CMultilabelModel
 Class CMultilabelModel represents application specific model and contains application dependent logic for solving multilabel classification within a generic SO framework. More...
class  CSparseMultilabel
 Class CSparseMultilabel to be used in the application of Structured Output (SO) learning to Multilabel classification. More...
class  CMultilabelSOLabels
 Class CMultilabelSOLabels used in the application of Structured Output (SO) learning to Multilabel Classification. Labels are subsets of {0, 1, ..., num_classes-1}. Each of the label if of type CSparseMultilabel and all of them are stored in a CDynamicObjectArray. More...
class  CPlif
 class Plif More...
class  CPlifArray
 class PlifArray More...
class  CPlifBase
 class PlifBase More...
class  CPlifMatrix
 store plif arrays for all transitions in the model More...
class  CSegmentLoss
 class IntronList More...
class  CSequence
 Class CSequence to be used in the application of Structured Output (SO) learning to Hidden Markov Support Vector Machines (HM-SVM). More...
class  CSequenceLabels
 Class CSequenceLabels used e.g. in the application of Structured Output (SO) learning to Hidden Markov Support Vector Machines (HM-SVM). Each of the labels is represented by a sequence of integers. Each label is of type CSequence and all of them are stored in a CDynamicObjectArray. More...
class  CSOSVMHelper
 class CSOSVMHelper contains helper functions to compute primal objectives, dual objectives, average training losses, duality gaps etc. These values will be recorded to check convergence. This class is inspired by the matlab implementation of the block coordinate Frank-Wolfe SOSVM solver [1]. More...
class  CStateModel
 class CStateModel base, abstract class for the internal state representation used in the CHMSVMModel. More...
class  CStochasticSOSVM
 Class CStochasticSOSVM solves SOSVM using stochastic subgradient descent on the SVM primal problem [1], which is equivalent to SGD or Pegasos [2]. This class is inspired by the matlab SGD implementation in [3]. More...
struct  TMultipleCPinfo
struct  CResultSet
class  CStructuredModel
 Class CStructuredModel that represents the application specific model and contains most of the application dependent logic to solve structured output (SO) problems. The idea of this class is to be instantiated giving pointers to the functions that are dependent on the application, i.e. the combined feature representation \(\Psi(\bold{x},\bold{y})\) and the argmax function \( {\arg\max} _{\bold{y} \neq \bold{y}_i} \left \langle { \bold{w}, \Psi(\bold{x}_i,\bold{y}) } \right \rangle \). See: MulticlassModel.h and .cpp for an example of these functions implemented. More...
class  CTwoStateModel
 class CTwoStateModel class for the internal two-state representation used in the CHMSVMModel. More...
class  CDomainAdaptationMulticlassLibLinear
 domain adaptation multiclass LibLinear wrapper Source domain is assumed to b More...
class  CDomainAdaptationSVM
 class DomainAdaptationSVM More...
class  CDomainAdaptationSVMLinear
 class DomainAdaptationSVMLinear More...
class  MappedSparseMatrix
 mapped sparse matrix for representing graph relations of tasks More...
class  CLibLinearMTL
 class to implement LibLinear More...
class  CMultitaskClusteredLogisticRegression
 class MultitaskClusteredLogisticRegression, a classifier for multitask problems. Supports only task group relations. Based on solver ported from the MALSAR library. Assumes task in group are related with a clustered structure. More...
class  CMultitaskKernelMaskNormalizer
 The MultitaskKernel allows Multitask Learning via a modified kernel function. More...
class  CMultitaskKernelMaskPairNormalizer
 The MultitaskKernel allows Multitask Learning via a modified kernel function. More...
class  CMultitaskKernelMklNormalizer
 Base-class for parameterized Kernel Normalizers. More...
class  CMultitaskKernelNormalizer
 The MultitaskKernel allows Multitask Learning via a modified kernel function. More...
class  CMultitaskKernelPlifNormalizer
 The MultitaskKernel allows learning a piece-wise linear function (PLIF) via MKL. More...
class  CNode
 A CNode is an element of a CTaxonomy, which is used to describe hierarchical structure between tasks. More...
class  CTaxonomy
 CTaxonomy is used to describe hierarchical structure between tasks. More...
class  CMultitaskKernelTreeNormalizer
 The MultitaskKernel allows Multitask Learning via a modified kernel function based on taxonomy. More...
class  CMultitaskL12LogisticRegression
 class MultitaskL12LogisticRegression, a classifier for multitask problems. Supports only task group relations. Based on solver ported from the MALSAR library. More...
class  CMultitaskLeastSquaresRegression
 class Multitask Least Squares Regression, a machine to solve regression problems with a few tasks related via group or tree. Based on L1/Lq regression for groups and L1/L2 for trees. More...
class  CMultitaskLinearMachine
 class MultitaskLinearMachine, a base class for linear multitask classifiers More...
class  CMultitaskLogisticRegression
 class Multitask Logistic Regression used to solve classification problems with a few tasks related via group or tree. Based on L1/Lq regression for groups and L1/L2 for trees. More...
class  CMultitaskROCEvaluation
 Class MultitaskROCEvalution used to evaluate ROC (Receiver Operating Characteristic) and an area under ROC curve (auROC) of each task separately. More...
class  CMultitaskTraceLogisticRegression
 class MultitaskTraceLogisticRegression, a classifier for multitask problems. Supports only task group relations. Based on solver ported from the MALSAR library. More...
class  CTask
 class Task used to represent tasks in multitask learning. Essentially it represent a set of feature vector indices. More...
class  CTaskGroup
 class TaskGroup used to represent a group of tasks. Tasks in group do not overlap. More...
class  CTaskRelation
 used to represent tasks in multitask learning More...
class  CTaskTree
 class TaskTree used to represent a tree of tasks. Tree is constructed via task with subtasks (and subtasks of subtasks ..) passed to the TaskTree. More...
class  CGUIClassifier
 UI classifier. More...
class  CGUIConverter
 UI converter. More...
class  CGUIDistance
 UI distance. More...
class  CGUIFeatures
 UI features. More...
class  CGUIHMM
 UI HMM (Hidden Markov Model) More...
class  CGUIKernel
 UI kernel. More...
class  CGUILabels
 UI labels. More...
class  CGUIMath
 UI math. More...
class  CGUIPluginEstimate
 UI estimate. More...
class  CGUIPreprocessor
 UI preprocessor. More...
class  CGUIStructure
 UI structure. More...
class  CGUITime
 UI time. More...

Typedefs

typedef uint32_t(* hash_func_t )(substring, uint32_t)
 Hash function typedef, takes a substring and seed as parameters.
typedef uint32_t vw_size_t
 vw_size_t typedef to work across platforms
typedef int32_t(CVwParser::* parse_func )(CIOBuffer *, VwExample *&)
 Parse function typedef. Takes an IOBuffer and VwExample as arguments.
typedef float64_t KERNELCACHE_ELEM
typedef int64_t KERNELCACHE_IDX
typedef struct shogun::_IterInfo IterInfo
 struct that contains current state of the iteration for iterative linear solvers
typedef CBinaryTreeMachineNode
< VwConditionalProbabilityTreeNodeData
bnode_t
typedef struct tag_callback_data callback_data_t
typedef struct tag_iteration_data iteration_data_t
typedef int32_t(* line_search_proc )(int32_t n, float64_t *x, float64_t *f, float64_t *g, float64_t *s, float64_t *stp, const float64_t *xp, const float64_t *gp, float64_t *wa, callback_data_t *cd, const lbfgs_parameter_t *param)
typedef float64_t(* lbfgs_evaluate_t )(void *instance, const float64_t *x, float64_t *g, const int n, const float64_t step)
typedef int(* lbfgs_progress_t )(void *instance, const float64_t *x, const float64_t *g, const float64_t fx, const float64_t xnorm, const float64_t gnorm, const float64_t step, int n, int k, int ls)
typedef float64_t(* lbfgs_adjust_step_t )(void *instance, const float64_t *x, const float64_t *d, const int n, const float64_t step)
HMM specific types
typedef float64_t T_ALPHA_BETA_TABLE
 type for alpha/beta caching table
typedef uint8_t T_STATES
typedef T_STATESP_STATES
convenience typedefs
typedef CDynInt< uint64_t, 3 > uint192_t
 192 bit integer constructed out of 3 64bit uint64_t's
typedef CDynInt< uint64_t, 4 > uint256_t
 256 bit integer constructed out of 4 64bit uint64_t's
typedef CDynInt< uint64_t, 8 > uint512_t
 512 bit integer constructed out of 8 64bit uint64_t's
typedef CDynInt< uint64_t, 16 > uint1024_t
 1024 bit integer constructed out of 16 64bit uint64_t's

Enumerations

enum  EModelSelectionAvailability { MS_NOT_AVAILABLE = 0, MS_AVAILABLE = 1 }
enum  EGradientAvailability { GRADIENT_NOT_AVAILABLE = 0, GRADIENT_AVAILABLE = 1 }
enum  ELDAMethod { AUTO_LDA = 10, SVD_LDA = 20, FLD_LDA = 30 }
enum  LIBLINEAR_SOLVER_TYPE {
  L2R_LR, L2R_L2LOSS_SVC_DUAL, L2R_L2LOSS_SVC, L2R_L1LOSS_SVC_DUAL,
  L1R_L2LOSS_SVC, L1R_LR, L2R_LR_DUAL
}
enum  EVwCacheType { C_NATIVE = 0, C_PROTOBUF = 1 }
enum  E_VW_PARSER_TYPE { T_VW = 1, T_SVMLIGHT = 2, T_DENSE = 3 }
 The type of input to parse. More...
enum  EKMeansMethod { KMM_MINI_BATCH, KMM_LLOYD }
enum  ESPEStrategy { SPE_GLOBAL, SPE_LOCAL }
enum  EDistanceType {
  D_UNKNOWN = 0, D_MINKOWSKI = 10, D_MANHATTAN = 20, D_CANBERRA = 30,
  D_CHEBYSHEW = 40, D_GEODESIC = 50, D_JENSEN = 60, D_MANHATTANWORD = 70,
  D_HAMMINGWORD = 80, D_CANBERRAWORD = 90, D_SPARSEEUCLIDEAN = 100, D_EUCLIDEAN = 110,
  D_CHISQUARE = 120, D_TANIMOTO = 130, D_COSINE = 140, D_BRAYCURTIS = 150,
  D_CUSTOM = 160, D_ATTENUATEDEUCLIDEAN = 170, D_MAHALANOBIS = 180, D_DIRECTOR = 190,
  D_CUSTOMMAHALANOBIS = 200
}
enum  ECovType { FULL, DIAG, SPHERICAL }
enum  BaumWelchViterbiType {
  BW_NORMAL, BW_TRANS, BW_DEFINED, VIT_NORMAL,
  VIT_DEFINED
}
enum  EEvaluationMode { EM_KDTREE_SINGLE, EM_BALLTREE_SINGLE, EM_KDTREE_DUAL, EM_BALLTREE_DUAL }
enum  EContingencyTableMeasureType {
  ACCURACY = 0, ERROR_RATE = 10, BAL = 20, WRACC = 30,
  F1 = 40, CROSS_CORRELATION = 50, RECALL = 60, PRECISION = 70,
  SPECIFICITY = 80, CUSTOM = 999
}
enum  EEvaluationDirection { ED_MINIMIZE = 0, ED_MAXIMIZE = 1 }
enum  EEvaluationResultType { CROSSVALIDATION_RESULT = 0, GRADIENTEVALUATION_RESULT = 1 }
enum  EAlphabet {
  DNA = 0, RAWDNA = 1, RNA = 2, PROTEIN = 3,
  BINARY = 4, ALPHANUM = 5, CUBE = 6, RAWBYTE = 7,
  IUPAC_NUCLEIC_ACID = 8, IUPAC_AMINO_ACID = 9, NONE = 10, DIGIT = 11,
  DIGIT2 = 12, RAWDIGIT = 13, RAWDIGIT2 = 14, UNKNOWN = 15,
  SNP = 16, RAWSNP = 17
}
 Alphabet of charfeatures/observations. More...
enum  EFeatureType {
  F_UNKNOWN = 0, F_BOOL = 5, F_CHAR = 10, F_BYTE = 20,
  F_SHORT = 30, F_WORD = 40, F_INT = 50, F_UINT = 60,
  F_LONG = 70, F_ULONG = 80, F_SHORTREAL = 90, F_DREAL = 100,
  F_LONGREAL = 110, F_ANY = 1000
}
 shogun feature type More...
enum  EFeatureClass {
  C_UNKNOWN = 0, C_DENSE = 10, C_SPARSE = 20, C_STRING = 30,
  C_COMBINED = 40, C_COMBINED_DOT = 60, C_WD = 70, C_SPEC = 80,
  C_WEIGHTEDSPEC = 90, C_POLY = 100, C_STREAMING_DENSE = 110, C_STREAMING_SPARSE = 120,
  C_STREAMING_STRING = 130, C_STREAMING_VW = 140, C_BINNED_DOT = 150, C_DIRECTOR_DOT = 160,
  C_LATENT = 170, C_MATRIX = 180, C_FACTOR_GRAPH = 190, C_INDEX = 200,
  C_ANY = 1000
}
 shogun feature class More...
enum  EFeatureProperty { FP_NONE = 0, FP_DOT = 1, FP_STREAMING_DOT = 2 }
 shogun feature properties More...
enum  KernelName { GAUSSIAN, NOT_SPECIFIED }
enum  EMessageType {
  MSG_GCDEBUG = 0, MSG_DEBUG = 1, MSG_INFO = 2, MSG_NOTICE = 3,
  MSG_WARN = 4, MSG_ERROR = 5, MSG_CRITICAL = 6, MSG_ALERT = 7,
  MSG_EMERGENCY = 8, MSG_MESSAGEONLY = 9
}
enum  EMessageLocation { MSG_NONE = 0, MSG_FUNCTION = 1, MSG_LINE_AND_FILE = 2 }
enum  EOptimizationType { FASTBUTMEMHUNGRY, SLOWBUTMEMEFFICIENT }
enum  EKernelType {
  K_UNKNOWN = 0, K_LINEAR = 10, K_POLY = 20, K_GAUSSIAN = 30,
  K_GAUSSIANSHIFT = 32, K_GAUSSIANMATCH = 33, K_HISTOGRAM = 40, K_SALZBERG = 41,
  K_LOCALITYIMPROVED = 50, K_SIMPLELOCALITYIMPROVED = 60, K_FIXEDDEGREE = 70, K_WEIGHTEDDEGREE = 80,
  K_WEIGHTEDDEGREEPOS = 81, K_WEIGHTEDDEGREERBF = 82, K_WEIGHTEDCOMMWORDSTRING = 90, K_POLYMATCH = 100,
  K_ALIGNMENT = 110, K_COMMWORDSTRING = 120, K_COMMULONGSTRING = 121, K_SPECTRUMRBF = 122,
  K_SPECTRUMMISMATCHRBF = 123, K_COMBINED = 140, K_AUC = 150, K_CUSTOM = 160,
  K_SIGMOID = 170, K_CHI2 = 180, K_DIAG = 190, K_CONST = 200,
  K_DISTANCE = 220, K_LOCALALIGNMENT = 230, K_PYRAMIDCHI2 = 240, K_OLIGO = 250,
  K_MATCHWORD = 260, K_TPPK = 270, K_REGULATORYMODULES = 280, K_SPARSESPATIALSAMPLE = 290,
  K_HISTOGRAMINTERSECTION = 300, K_WAVELET = 310, K_WAVE = 320, K_CAUCHY = 330,
  K_TSTUDENT = 340, K_RATIONAL_QUADRATIC = 350, K_MULTIQUADRIC = 360, K_EXPONENTIAL = 370,
  K_SPHERICAL = 380, K_SPLINE = 390, K_ANOVA = 400, K_POWER = 410,
  K_LOG = 420, K_CIRCULAR = 430, K_INVERSEMULTIQUADRIC = 440, K_DISTANTSEGMENTS = 450,
  K_BESSEL = 460, K_JENSENSHANNON = 470, K_DIRECTOR = 480, K_PRODUCT = 490,
  K_LINEARARD = 500, K_GAUSSIANARD = 510, K_STREAMING = 520
}
enum  EKernelProperty { KP_NONE = 0, KP_LINADD = 1, KP_KERNCOMBINATION = 2, KP_BATCHEVALUATION = 4 }
enum  ENormalizerType { N_REGULAR = 0, N_MULTITASK = 1 }
enum  EWDKernType {
  E_WD = 0, E_EXTERNAL = 1, E_BLOCK_CONST = 2, E_BLOCK_LINEAR = 3,
  E_BLOCK_SQPOLY = 4, E_BLOCK_CUBICPOLY = 5, E_BLOCK_EXP = 6, E_BLOCK_LOG = 7
}
enum  E_COMPRESSION_TYPE {
  UNCOMPRESSED, LZO, GZIP, BZIP2,
  LZMA, SNAPPY
}
enum  EFeaturesContainer { FC_LHS = 0, FC_RHS = 1 }
enum  EStructuredDataType {
  SDT_UNKNOWN = 0, SDT_REAL = 1, SDT_SEQUENCE = 2, SDT_FACTOR_GRAPH = 3,
  SDT_SPARSE_MULTILABEL = 4
}
enum  ELossType {
  L_HINGELOSS = 0, L_SMOOTHHINGELOSS = 10, L_SQUAREDHINGELOSS = 20, L_SQUAREDLOSS = 30,
  L_EXPONENTIALLOSS = 40, L_ABSOLUTEDEVIATIONLOSS = 50, L_HUBERLOSS = 60, L_LOGLOSS = 100,
  L_LOGLOSSMARGIN = 110
}
 shogun loss type More...
enum  EInferenceType {
  INF_NONE = 0, INF_EXACT = 10, INF_FITC = 20, INF_LAPLACIAN = 30,
  INF_EP = 40, INF_KL = 50
}
enum  ELikelihoodModelType {
  LT_NONE = 0, LT_GAUSSIAN = 10, LT_STUDENTST = 20, LT_LOGIT = 30,
  LT_PROBIT = 40, LT_SOFTMAX = 50
}
enum  EMCSamplerType { MC_Probability, MC_Mean, MC_Variance }
enum  EMachineType {
  CT_NONE = 0, CT_LIGHT = 10, CT_LIGHTONECLASS = 11, CT_LIBSVM = 20,
  CT_LIBSVMONECLASS = 30, CT_LIBSVMMULTICLASS = 40, CT_MPD = 50, CT_GPBT = 60,
  CT_CPLEXSVM = 70, CT_PERCEPTRON = 80, CT_KERNELPERCEPTRON = 90, CT_LDA = 100,
  CT_LPM = 110, CT_LPBOOST = 120, CT_KNN = 130, CT_SVMLIN = 140,
  CT_KERNELRIDGEREGRESSION = 150, CT_GNPPSVM = 160, CT_GMNPSVM = 170, CT_SVMPERF = 200,
  CT_LIBSVR = 210, CT_SVRLIGHT = 220, CT_LIBLINEAR = 230, CT_KMEANS = 240,
  CT_HIERARCHICAL = 250, CT_SVMOCAS = 260, CT_WDSVMOCAS = 270, CT_SVMSGD = 280,
  CT_MKLMULTICLASS = 290, CT_MKLCLASSIFICATION = 300, CT_MKLONECLASS = 310, CT_MKLREGRESSION = 320,
  CT_SCATTERSVM = 330, CT_DASVM = 340, CT_LARANK = 350, CT_DASVMLINEAR = 360,
  CT_GAUSSIANNAIVEBAYES = 370, CT_AVERAGEDPERCEPTRON = 380, CT_SGDQN = 390, CT_CONJUGATEINDEX = 400,
  CT_LINEARRIDGEREGRESSION = 410, CT_LEASTSQUARESREGRESSION = 420, CT_QDA = 430, CT_NEWTONSVM = 440,
  CT_GAUSSIANPROCESSREGRESSION = 450, CT_LARS = 460, CT_MULTICLASS = 470, CT_DIRECTORLINEAR = 480,
  CT_DIRECTORKERNEL = 490, CT_LIBQPSOSVM = 500, CT_PRIMALMOSEKSOSVM = 510, CT_CCSOSVM = 520,
  CT_GAUSSIANPROCESSBINARY = 530, CT_GAUSSIANPROCESSMULTICLASS = 540, CT_STOCHASTICSOSVM = 550, CT_NEURALNETWORK = 560,
  CT_BAGGING = 570, CT_FWSOSVM = 580, CT_BCFWSOSVM = 590, CT_GAUSSIANPROCESSCLASS
}
enum  ESolverType {
  ST_AUTO = 0, ST_CPLEX = 1, ST_GLPK = 2, ST_NEWTON = 3,
  ST_DIRECT = 4, ST_ELASTICNET = 5, ST_BLOCK_NORM = 6
}
enum  EProblemType {
  PT_BINARY = 0, PT_REGRESSION = 1, PT_MULTICLASS = 2, PT_STRUCTURED = 3,
  PT_LATENT = 4, PT_CLASS = 5
}
enum  EStructRiskType {
  N_SLACK_MARGIN_RESCALING = 0, N_SLACK_SLACK_RESCALING = 1, ONE_SLACK_MARGIN_RESCALING = 2, ONE_SLACK_SLACK_RESCALING = 3,
  CUSTOMIZED_RISK = 4
}
enum  E_PROB_TYPE { E_LINEAR, E_QP }
enum  EDirectSolverType {
  DS_LLT = 0, DS_QR_NOPERM = 1, DS_QR_COLPERM = 2, DS_QR_FULLPERM = 3,
  DS_SVD = 4
}
enum  EOperatorFunction { OF_SQRT = 0, OF_LOG = 1, OF_POLY = 2, OF_UNDEFINED = 3 }
enum  ERangeType { R_LINEAR, R_EXP, R_LOG }
enum  EMSParamType {
  MSPT_NONE = 0, MSPT_FLOAT64, MSPT_INT32, MSPT_FLOAT64_VECTOR,
  MSPT_INT32_VECTOR, MSPT_FLOAT64_SGVECTOR, MSPT_INT32_SGVECTOR
}
enum  EProbHeuristicType {
  PROB_HEURIS_NONE = 0, OVA_NORM = 1, OVA_SOFTMAX = 2, OVO_PRICE = 3,
  OVO_HASTIE = 4, OVO_HAMAMURA = 5
}
enum  SCATTER_TYPE { NO_BIAS_LIBSVM, NO_BIAS_SVMLIGHT, TEST_RULE1, TEST_RULE2 }
enum  EAENoiseType { AENT_NONE = 0, AENT_DROPOUT = 1, AENT_GAUSSIAN = 2 }
 Determines the noise type for denoising autoencoders. More...
enum  EConvMapActivationFunction { CMAF_IDENTITY = 0, CMAF_LOGISTIC = 1, CMAF_RECTIFIED_LINEAR = 2 }
 Determines the activation function for neurons in a convolutional feature map. More...
enum  ENLAutoencoderPosition { NLAP_NONE = 0, NLAP_ENCODING = 1, NLAP_DECODING = 2 }
enum  ENNOptimizationMethod { NNOM_GRADIENT_DESCENT = 0, NNOM_LBFGS = 1 }
enum  ERBMMonitoringMethod { RBMMM_RECONSTRUCTION_ERROR = 0, RBMMM_PSEUDO_LIKELIHOOD = 1 }
enum  ERBMVisibleUnitType { RBMVUT_BINARY = 0, RBMVUT_GAUSSIAN = 1, RBMVUT_SOFTMAX = 2 }
enum  {
  LBFGS_SUCCESS = 0, LBFGS_CONVERGENCE = 0, LBFGS_STOP, LBFGS_ALREADY_MINIMIZED,
  LBFGSERR_UNKNOWNERROR = -1024, LBFGSERR_LOGICERROR, LBFGSERR_OUTOFMEMORY, LBFGSERR_CANCELED,
  LBFGSERR_INVALID_N, LBFGSERR_INVALID_N_SSE, LBFGSERR_INVALID_X_SSE, LBFGSERR_INVALID_EPSILON,
  LBFGSERR_INVALID_TESTPERIOD, LBFGSERR_INVALID_DELTA, LBFGSERR_INVALID_LINESEARCH, LBFGSERR_INVALID_MINSTEP,
  LBFGSERR_INVALID_MAXSTEP, LBFGSERR_INVALID_FTOL, LBFGSERR_INVALID_WOLFE, LBFGSERR_INVALID_GTOL,
  LBFGSERR_INVALID_XTOL, LBFGSERR_INVALID_MAXLINESEARCH, LBFGSERR_INVALID_ORTHANTWISE, LBFGSERR_INVALID_ORTHANTWISE_START,
  LBFGSERR_INVALID_ORTHANTWISE_END, LBFGSERR_OUTOFINTERVAL, LBFGSERR_INCORRECT_TMINMAX, LBFGSERR_ROUNDING_ERROR,
  LBFGSERR_MINIMUMSTEP, LBFGSERR_MAXIMUMSTEP, LBFGSERR_MAXIMUMLINESEARCH, LBFGSERR_MAXIMUMITERATION,
  LBFGSERR_WIDTHTOOSMALL, LBFGSERR_INVALIDPARAMETERS, LBFGSERR_INCREASEGRADIENT, LBFGSERR_INVALID_VALUE
}
enum  {
  LBFGS_LINESEARCH_DEFAULT = 0, LBFGS_LINESEARCH_MORETHUENTE = 0, LBFGS_LINESEARCH_BACKTRACKING_ARMIJO = 1, LBFGS_LINESEARCH_BACKTRACKING = 2,
  LBFGS_LINESEARCH_BACKTRACKING_WOLFE = 2, LBFGS_LINESEARCH_BACKTRACKING_STRONG_WOLFE = 3
}
enum  EFeatureSelectionAlgorithm { BACKWARD_ELIMINATION, FORWARD_SELECTION }
enum  EFeatureRemovalPolicy { N_SMALLEST, PERCENTILE_SMALLEST, N_LARGEST, PERCENTILE_LARGEST }
enum  EFLDAMethod { AUTO_FLDA = 10, CANVAR_FLDA = 20, CLASSIC_FLDA = 30 }
enum  HomogeneousKernelType { HomogeneousKernelIntersection = 0, HomogeneousKernelChi2, HomogeneousKernelJS }
 Type of kernel. More...
enum  HomogeneousKernelMapWindowType { HomogeneousKernelMapWindowUniform = 0, HomogeneousKernelMapWindowRectangular = 1 }
 Type of spectral windowing function. More...
enum  EPCAMethod { AUTO = 10, SVD = 20, EVD = 30 }
enum  EPCAMode { THRESHOLD, VARIANCE_EXPLAINED, FIXED_NUMBER }
enum  EPCAMemoryMode { MEM_REALLOCATE, MEM_IN_PLACE }
enum  EPreprocessorType {
  P_UNKNOWN = 0, P_NORMONE = 10, P_LOGPLUSONE = 20, P_SORTWORDSTRING = 30,
  P_SORTULONGSTRING = 40, P_SORTWORD = 50, P_PRUNEVARSUBMEAN = 60, P_DECOMPRESSSTRING = 70,
  P_DECOMPRESSCHARSTRING = 80, P_DECOMPRESSBYTESTRING = 90, P_DECOMPRESSWORDSTRING = 100, P_DECOMPRESSULONGSTRING = 110,
  P_RANDOMFOURIERGAUSS = 120, P_PCA = 130, P_KERNELPCA = 140, P_NORMDERIVATIVELEM3 = 150,
  P_DIMENSIONREDUCTIONPREPROCESSOR = 160, P_SUMONE = 170, P_HOMOGENEOUSKERNELMAP = 180, P_PNORM = 190,
  P_RESCALEFEATURES = 200, P_FISHERLDA = 210, P_BAHSIC = 220
}
enum  ETrainingType { PINV = 1, GS = 2 }
enum  ERegressionType { RT_NONE = 0, RT_LIGHT = 10, RT_LIBSVM = 20 }
 type of regressor More...
enum  LIBLINEAR_REGRESSION_TYPE { L2R_L2LOSS_SVR, L2R_L1LOSS_SVR_DUAL, L2R_L2LOSS_SVR_DUAL }
enum  EStatisticType { S_LINEAR_TIME_MMD, S_QUADRATIC_TIME_MMD, S_HSIC, S_NOCCO }
enum  ENullApproximationMethod {
  PERMUTATION, MMD2_SPECTRUM_DEPRECATED, MMD2_SPECTRUM, MMD2_GAMMA,
  MMD1_GAUSSIAN, HSIC_GAMMA
}
enum  EQuadraticMMDType {
  BIASED, BIASED_DEPRECATED, UNBIASED, UNBIASED_DEPRECATED,
  INCOMPLETE
}
enum  EQPType { MOSEK = 1, SVMLIGHT = 2 }
enum  ESolver { BMRM = 1, PPBMRM = 2, P3BMRM = 3, NCBM = 4 }
enum  ETerminalType { SOURCE = 0, SINK = 1 }
enum  EMAPInferType {
  TREE_MAX_PROD = 0, LOOPY_MAX_PROD = 1, LP_RELAXATION = 2, TRWS_MAX_PROD = 3,
  GRAPH_CUT = 4
}
enum  ETransformType {
  T_LINEAR, T_LOG, T_LOG_PLUS1, T_LOG_PLUS3,
  T_LINEAR_PLUS3
}
enum  EStateModelType { SMT_UNKNOWN = 0, SMT_TWO_STATE = 1 }

Functions

CSGObjectnew_sgserializable (const char *sgserializable_name, EPrimitiveType generic)
void init_shogun (void(*print_message)(FILE *target, const char *str), void(*print_warning)(FILE *target, const char *str), void(*print_error)(FILE *target, const char *str), void(*cancel_computations)(bool &delayed, bool &immediately))
void sg_global_print_default (FILE *target, const char *str)
void init_shogun_with_defaults ()
void exit_shogun ()
void set_global_io (SGIO *io)
SGIOget_global_io ()
void set_global_parallel (Parallel *parallel)
Parallelget_global_parallel ()
void set_global_version (Version *version)
Versionget_global_version ()
void set_global_math (CMath *math)
CMathget_global_math ()
void set_global_rand (CRandom *rand)
CRandomget_global_rand ()
void init_from_env ()
float32_t sd_offset_add (float32_t *weights, vw_size_t mask, VwFeature *begin, VwFeature *end, vw_size_t offset)
float32_t sd_offset_truncadd (float32_t *weights, vw_size_t mask, VwFeature *begin, VwFeature *end, vw_size_t offset, float32_t gravity)
float32_t one_pf_quad_predict (float32_t *weights, VwFeature &f, v_array< VwFeature > &cross_features, vw_size_t mask)
float32_t one_pf_quad_predict_trunc (float32_t *weights, VwFeature &f, v_array< VwFeature > &cross_features, vw_size_t mask, float32_t gravity)
float32_t real_weight (float32_t w, float32_t gravity)
template<class T >
void push (v_array< T > &v, const T &new_ele)
template<class T >
void alloc (v_array< T > &v, int length)
template<class T >
v_array< T > pop (v_array< v_array< T > > &stack)
float distance (CJLCoverTreePoint p1, CJLCoverTreePoint p2, float64_t upper_bound)
v_array< CJLCoverTreePointparse_points (CDistance *distance, EFeaturesContainer fc)
void print (CJLCoverTreePoint &p)
static const double * get_col (uint32_t j)
malsar_result_t malsar_clustered (CDotFeatures *features, double *y, double rho1, double rho2, const malsar_options &options)
malsar_result_t malsar_joint_feature_learning (CDotFeatures *features, double *y, double rho1, double rho2, const malsar_options &options)
malsar_result_t malsar_low_rank (CDotFeatures *features, double *y, double rho, const malsar_options &options)
void * sg_malloc (size_t size)
void * sg_calloc (size_t num, size_t size)
void sg_free (void *ptr)
void * sg_realloc (void *ptr, size_t size)
slep_result_t slep_mc_plain_lr (CDotFeatures *features, CMulticlassLabels *labels, float64_t z, const slep_options &options)
slep_result_t slep_mc_tree_lr (CDotFeatures *features, CMulticlassLabels *labels, float64_t z, const slep_options &options)
double compute_regularizer (double *w, double lambda, double lambda2, int n_vecs, int n_feats, int n_blocks, const slep_options &options)
double compute_lambda (double *ATx, double z, CDotFeatures *features, double *y, int n_vecs, int n_feats, int n_blocks, const slep_options &options)
void projection (double *w, double *v, int n_feats, int n_blocks, double lambda, double lambda2, double L, double *z, double *z0, const slep_options &options)
double search_point_gradient_and_objective (CDotFeatures *features, double *ATx, double *As, double *sc, double *y, int n_vecs, int n_feats, int n_tasks, double *g, double *gc, const slep_options &options)
slep_result_t slep_solver (CDotFeatures *features, double *y, double z, const slep_options &options)
void wrap_dsyev (char jobz, char uplo, int n, double *a, int lda, double *w, int *info)
void wrap_dgesvd (char jobu, char jobvt, int m, int n, double *a, int lda, double *sing, double *u, int ldu, double *vt, int ldvt, int *info)
void wrap_dgeqrf (int m, int n, double *a, int lda, double *tau, int *info)
void wrap_dorgqr (int m, int n, int k, double *a, int lda, double *tau, int *info)
void wrap_dsyevr (char jobz, char uplo, int n, double *a, int lda, int il, int iu, double *eigenvalues, double *eigenvectors, int *info)
void wrap_dsygvx (int itype, char jobz, char uplo, int n, double *a, int lda, double *b, int ldb, int il, int iu, double *eigenvalues, double *eigenvectors, int *info)
void wrap_dstemr (char jobz, char range, int n, double *diag, double *subdiag, double vl, double vu, int il, int iu, int *m, double *w, double *z__, int ldz, int nzc, int *isuppz, int tryrac, int *info)
virtual T compute (SGVector< T > vector1, SGVector< T > vector2) const
 Template class for Eigen3 dot product that performs dot product operation( \(\sum_{i=1}^d a_ib_i$ where $a,b$ are $d$-dimensional vectors) using Eigen3 Library */ template <class T> class DenseEigen3DotProduct : public VectorDotProduct<T, SGVector<T> > { public: /** Constructor */ DenseEigen3DotProduct() : VectorDotProduct<T, SGVector<T> >() { } /** Compute method which performs dot product operation (\){i=1}^d a_ib_i$ where $a,b$ are $d$-dimensional vectors)
virtual ~DenseEigen3DotProduct ()
template<class T >
SGVector< T > create_range_array (T min, T max, ERangeType type, T step, T type_base)
static larank_kcache_t * larank_kcache_create (CKernel *kernelfunc)
static void xtruncate (larank_kcache_t *self, int32_t k, int32_t nlen)
static void xpurge (larank_kcache_t *self)
static void larank_kcache_set_maximum_size (larank_kcache_t *self, int64_t entries)
static void larank_kcache_destroy (larank_kcache_t *self)
static void xminsize (larank_kcache_t *self, int32_t n)
static int32_t * larank_kcache_r2i (larank_kcache_t *self, int32_t n)
static void xextend (larank_kcache_t *self, int32_t k, int32_t nlen)
static void xswap (larank_kcache_t *self, int32_t i1, int32_t i2, int32_t r1, int32_t r2)
static void larank_kcache_swap_rr (larank_kcache_t *self, int32_t r1, int32_t r2)
static void larank_kcache_swap_ri (larank_kcache_t *self, int32_t r1, int32_t i2)
static float64_t xquery (larank_kcache_t *self, int32_t i, int32_t j)
static float64_t larank_kcache_query (larank_kcache_t *self, int32_t i, int32_t j)
static void larank_kcache_set_buddy (larank_kcache_t *self, larank_kcache_t *buddy)
static float32_tlarank_kcache_query_row (larank_kcache_t *self, int32_t i, int32_t len)
static int32_t line_search_backtracking (int32_t n, float64_t *x, float64_t *f, float64_t *g, float64_t *s, float64_t *stp, const float64_t *xp, const float64_t *gp, float64_t *wa, callback_data_t *cd, const lbfgs_parameter_t *param)
static int32_t line_search_backtracking_owlqn (int32_t n, float64_t *x, float64_t *f, float64_t *g, float64_t *s, float64_t *stp, const float64_t *xp, const float64_t *gp, float64_t *wp, callback_data_t *cd, const lbfgs_parameter_t *param)
static int32_t line_search_morethuente (int32_t n, float64_t *x, float64_t *f, float64_t *g, float64_t *s, float64_t *stp, const float64_t *xp, const float64_t *gp, float64_t *wa, callback_data_t *cd, const lbfgs_parameter_t *param)
static int32_t update_trial_interval (float64_t *x, float64_t *fx, float64_t *dx, float64_t *y, float64_t *fy, float64_t *dy, float64_t *t, float64_t *ft, float64_t *dt, const float64_t tmin, const float64_t tmax, int32_t *brackt)
static float64_t owlqn_x1norm (const float64_t *x, const int32_t start, const int32_t n)
static void owlqn_pseudo_gradient (float64_t *pg, const float64_t *x, const float64_t *g, const int32_t n, const float64_t c, const int32_t start, const int32_t end)
static void owlqn_project (float64_t *d, const float64_t *sign, const int32_t start, const int32_t end)
void lbfgs_parameter_init (lbfgs_parameter_t *param)
int32_t lbfgs (int32_t n, float64_t *x, float64_t *ptr_fx, lbfgs_evaluate_t proc_evaluate, lbfgs_progress_t proc_progress, void *instance, lbfgs_parameter_t *_param, lbfgs_adjust_step_t proc_adjust_step)
int lbfgs (int n, float64_t *x, float64_t *ptr_fx, lbfgs_evaluate_t proc_evaluate, lbfgs_progress_t proc_progress, void *instance, lbfgs_parameter_t *param, lbfgs_adjust_step_t proc_adjust_step=NULL)
void add_cutting_plane (bmrm_ll **tail, bool *map, float64_t *A, uint32_t free_idx, float64_t *cp_data, uint32_t dim)
void remove_cutting_plane (bmrm_ll **head, bmrm_ll **tail, bool *map, float64_t *icp)
void clean_icp (ICP_stats *icp_stats, BmrmStatistics &bmrm, bmrm_ll **head, bmrm_ll **tail, float64_t *&Hmat, float64_t *&diag_H, float64_t *&beta, bool *&map, uint32_t cleanAfter, float64_t *&b, uint32_t *&I, uint32_t cp_models)
static const float64_tget_col (uint32_t i)
BmrmStatistics svm_bmrm_solver (CDualLibQPBMSOSVM *machine, float64_t *W, float64_t TolRel, float64_t TolAbs, float64_t _lambda, uint32_t _BufSize, bool cleanICP, uint32_t cleanAfter, float64_t K, uint32_t Tmax, bool store_train_info)
float64_tget_cutting_plane (bmrm_ll *ptr)
uint32_t find_free_idx (bool *map, uint32_t size)
static const float64_tget_col (uint32_t i)
static line_search_res zoom (CDualLibQPBMSOSVM *machine, float64_t lambda, float64_t a_lo, float64_t a_hi, float64_t initial_fval, SGVector< float64_t > &initial_solution, SGVector< float64_t > &search_dir, float64_t wolfe_c1, float64_t wolfe_c2, float64_t init_lgrad, float64_t f_lo, float64_t g_lo, float64_t f_hi, float64_t g_hi)
std::vector< line_search_resline_search_with_strong_wolfe (CDualLibQPBMSOSVM *machine, float64_t lambda, float64_t initial_val, SGVector< float64_t > &initial_solution, SGVector< float64_t > &initial_grad, SGVector< float64_t > &search_dir, float64_t astart, float64_t amax=1.1, float64_t wolfe_c1=1E-4, float64_t wolfe_c2=0.9, float64_t max_iter=5)
void update_H (BmrmStatistics &ncbm, bmrm_ll *head, bmrm_ll *tail, SGMatrix< float64_t > &H, SGVector< float64_t > &diag_H, float64_t lambda, uint32_t maxCP, int32_t w_dim)
BmrmStatistics svm_ncbm_solver (CDualLibQPBMSOSVM *machine, float64_t *w, float64_t TolRel, float64_t TolAbs, float64_t _lambda, uint32_t _BufSize, bool cleanICP, uint32_t cleanAfter, bool is_convex, bool line_search, bool verbose)
static const float64_tget_col (uint32_t i)
BmrmStatistics svm_p3bm_solver (CDualLibQPBMSOSVM *machine, float64_t *W, float64_t TolRel, float64_t TolAbs, float64_t _lambda, uint32_t _BufSize, bool cleanICP, uint32_t cleanAfter, float64_t K, uint32_t Tmax, uint32_t cp_models, bool verbose)
static const float64_tget_col (uint32_t i)
BmrmStatistics svm_ppbm_solver (CDualLibQPBMSOSVM *machine, float64_t *W, float64_t TolRel, float64_t TolAbs, float64_t _lambda, uint32_t _BufSize, bool cleanICP, uint32_t cleanAfter, float64_t K, uint32_t Tmax, bool verbose)

Variables

Parallelsg_parallel = NULL
SGIOsg_io = NULL
Versionsg_version = NULL
CMathsg_math = NULL
CRandomsg_rand = NULL
void(* sg_print_message )(FILE *target, const char *str) = NULL
 function called to print normal messages
void(* sg_print_warning )(FILE *target, const char *str) = NULL
 function called to print warning messages
void(* sg_print_error )(FILE *target, const char *str) = NULL
 function called to print error messages
void(* sg_cancel_computations )(bool &delayed, bool &immediately) = NULL
 function called to cancel things
const int32_t quadratic_constant = 27942141
 Constant used while hashing/accessing quadratic features.
const int32_t constant_hash = 11650396
 Constant used to access the constant feature.
const uint32_t hash_base = 97562527
 Seed for hash.
const int32_t LOGSUM_TBL = 10000
static double * H_diag_matrix
static int H_diag_matrix_ld
static const float64_t Q_test_statistic_values [10][8]
static const lbfgs_parameter_t _defparam
static const uint32_t QPSolverMaxIter = 0xFFFFFFFF
static const float64_t epsilon = 0.0
static float64_tH
uint32_t BufSize
static float64_tHMatrix
static uint32_t maxCPs
static const float64_t epsilon = 0.0
static const uint32_t QPSolverMaxIter = 0xFFFFFFFF
static const float64_t epsilon = 0.0
static float64_tH
static float64_tH2
static const uint32_t QPSolverMaxIter = 0xFFFFFFFF
static const float64_t epsilon = 0.0
static float64_tH
static float64_tH2

Detailed Description

all of classes and functions are contained in the shogun namespace

Contains special purpose, algorithm specific functions. Uses the same backend as the Core module

Contains some utility functions. Uses the same backend as the Core module

Typedef Documentation

define shortcut for the node type

Definition at line 37 of file VwConditionalProbabilityTree.h.

Definition at line 89 of file lbfgs.cpp.

typedef uint32_t(* hash_func_t)(substring, uint32_t)

Hash function typedef, takes a substring and seed as parameters.

Definition at line 23 of file vw_constants.h.

Definition at line 97 of file lbfgs.cpp.

typedef struct shogun::_IterInfo IterInfo

struct that contains current state of the iteration for iterative linear solvers

kernel cache element

Definition at line 35 of file Kernel.h.

typedef int64_t KERNELCACHE_IDX

kernel cache index

Definition at line 46 of file Kernel.h.

typedef int32_t(* line_search_proc)(int32_t n, float64_t *x, float64_t *f, float64_t *g, float64_t *s, float64_t *stp, const float64_t *xp, const float64_t *gp, float64_t *wa, callback_data_t *cd, const lbfgs_parameter_t *param)

Definition at line 108 of file lbfgs.cpp.

typedef T_STATES * P_STATES

Definition at line 66 of file HMM.h.

typedef int32_t(CVwParser::* parse_func)(CIOBuffer *, VwExample *&)

Parse function typedef. Takes an IOBuffer and VwExample as arguments.

Definition at line 22 of file StreamingVwFile.h.

type for alpha/beta caching table

Definition at line 37 of file HMM.h.

typedef uint8_t T_STATES

type that is used for states. Probably uint8_t is enough if you have at most 256 states, however uint16_t/long/... is also possible although you might quickly run into memory problems

Definition at line 64 of file HMM.h.

typedef CDynInt<uint64_t,16> uint1024_t

1024 bit integer constructed out of 16 64bit uint64_t's

Definition at line 576 of file DynInt.h.

typedef CDynInt<uint64_t,3> uint192_t

192 bit integer constructed out of 3 64bit uint64_t's

Definition at line 567 of file DynInt.h.

typedef CDynInt<uint64_t,4> uint256_t

256 bit integer constructed out of 4 64bit uint64_t's

Definition at line 570 of file DynInt.h.

typedef CDynInt<uint64_t,8> uint512_t

512 bit integer constructed out of 8 64bit uint64_t's

Definition at line 573 of file DynInt.h.

typedef uint32_t vw_size_t

vw_size_t typedef to work across platforms

Definition at line 26 of file vw_constants.h.

Enumeration Type Documentation

Training type

Enumerator:
BW_NORMAL 

standard baum welch

BW_TRANS 

baum welch only for specified transitions

BW_DEFINED 

baum welch only for defined transitions/observations

VIT_NORMAL 

standard viterbi

VIT_DEFINED 

viterbi only for defined transitions/observations

Definition at line 71 of file HMM.h.

compression type

Enumerator:
UNCOMPRESSED 
LZO 
GZIP 
BZIP2 
LZMA 
SNAPPY 

Definition at line 21 of file Compressor.h.

Enumerator:
E_LINEAR 
E_QP 

Definition at line 29 of file Cplex.h.

The type of input to parse.

Enumerator:
T_VW 
T_SVMLIGHT 
T_DENSE 

Definition at line 30 of file VwParser.h.

Determines the noise type for denoising autoencoders.

Enumerator:
AENT_NONE 

No noise is applied

AENT_DROPOUT 

Each neuron in the input layer is randomly set to zero with some probability

AENT_GAUSSIAN 

Gaussian noise is added to neurons in the input layer

Definition at line 47 of file Autoencoder.h.

enum EAlphabet

Alphabet of charfeatures/observations.

Enumerator:
DNA 

DNA - letters A,C,G,T.

RAWDNA 

RAWDNA - letters 0,1,2,3.

RNA 

RNA - letters A,C,G,U.

PROTEIN 

PROTEIN - letters A-Z.

BINARY 
ALPHANUM 

ALPHANUM - [0-9A-Z].

CUBE 

CUBE - [1-6].

RAWBYTE 

RAW BYTE - [0-255].

IUPAC_NUCLEIC_ACID 

IUPAC_NUCLEIC_ACID.

IUPAC_AMINO_ACID 

IUPAC_AMINO_ACID.

NONE 

NONE - type has no alphabet.

DIGIT 

DIGIT - letters 0-9.

DIGIT2 

DIGIT2 - letters 0-2.

RAWDIGIT 

RAWDIGIT - 0-9.

RAWDIGIT2 

RAWDIGIT2 - 0-2.

UNKNOWN 

unknown alphabet

SNP 

SNP - letters A,C,G,T,0.

RAWSNP 

RAWSNP - letters 0,1,2,3,4.

Definition at line 23 of file Alphabet.h.

type of measure

Enumerator:
ACCURACY 
ERROR_RATE 
BAL 
WRACC 
F1 
CROSS_CORRELATION 
RECALL 
PRECISION 
SPECIFICITY 
CUSTOM 

Definition at line 27 of file ContingencyTableEvaluation.h.

Determines the activation function for neurons in a convolutional feature map.

Enumerator:
CMAF_IDENTITY 

Identity activation function: \( f(x) = x \)

CMAF_LOGISTIC 

Logistic activation function: \( f(x) = \frac{1}{1+exp(-x)} \)

CMAF_RECTIFIED_LINEAR 

Rectified linear activation function: \( f(x) = max(x,0) \)

Definition at line 46 of file ConvolutionalFeatureMap.h.

enum ECovType

Covariance type

Enumerator:
FULL 

full covariance

DIAG 

diagonal covariance

SPHERICAL 

spherical covariance

Definition at line 33 of file Gaussian.h.

solver type for direct solvers

Enumerator:
DS_LLT 
DS_QR_NOPERM 
DS_QR_COLPERM 
DS_QR_FULLPERM 
DS_SVD 

Definition at line 22 of file DirectLinearSolverComplex.h.

type of distance

Enumerator:
D_UNKNOWN 
D_MINKOWSKI 
D_MANHATTAN 
D_CANBERRA 
D_CHEBYSHEW 
D_GEODESIC 
D_JENSEN 
D_MANHATTANWORD 
D_HAMMINGWORD 
D_CANBERRAWORD 
D_SPARSEEUCLIDEAN 
D_EUCLIDEAN 
D_CHISQUARE 
D_TANIMOTO 
D_COSINE 
D_BRAYCURTIS 
D_CUSTOM 
D_ATTENUATEDEUCLIDEAN 
D_MAHALANOBIS 
D_DIRECTOR 
D_CUSTOMMAHALANOBIS 

Definition at line 34 of file Distance.h.

enum which used to define whether an evaluation measure has to be minimized or maximized

Enumerator:
ED_MINIMIZE 
ED_MAXIMIZE 

Definition at line 29 of file Evaluation.h.

Enumerator:
EM_KDTREE_SINGLE 
EM_BALLTREE_SINGLE 
EM_KDTREE_DUAL 
EM_BALLTREE_DUAL 

Definition at line 41 of file KernelDensity.h.

Type of evaluation result. Currently this includes Cross Validation and Gradient Evaluation

Enumerator:
CROSSVALIDATION_RESULT 
GRADIENTEVALUATION_RESULT 

Definition at line 23 of file EvaluationResult.h.

shogun feature class

Enumerator:
C_UNKNOWN 
C_DENSE 
C_SPARSE 
C_STRING 
C_COMBINED 
C_COMBINED_DOT 
C_WD 
C_SPEC 
C_WEIGHTEDSPEC 
C_POLY 
C_STREAMING_DENSE 
C_STREAMING_SPARSE 
C_STREAMING_STRING 
C_STREAMING_VW 
C_BINNED_DOT 
C_DIRECTOR_DOT 
C_LATENT 
C_MATRIX 
C_FACTOR_GRAPH 
C_INDEX 
C_ANY 

Definition at line 38 of file FeatureTypes.h.

shogun feature properties

Enumerator:
FP_NONE 
FP_DOT 
FP_STREAMING_DOT 

Definition at line 64 of file FeatureTypes.h.

Enum for feature removal policy in feature selection algorithms. See class documentation of CFeatureSelection for their descriptions.

Enumerator:
N_SMALLEST 
PERCENTILE_SMALLEST 
N_LARGEST 
PERCENTILE_LARGEST 

Definition at line 55 of file FeatureSelection.h.

Type used to indicate where to find (either lhs or rhs) the coordinate information of this point in the CDistance object associated

Enumerator:
FC_LHS 
FC_RHS 

Definition at line 107 of file JLCoverTreePoint.h.

Enum for feature selection algorithms. See class documentation of CFeatureSelection for their descriptions.

Enumerator:
BACKWARD_ELIMINATION 
FORWARD_SELECTION 

Definition at line 46 of file FeatureSelection.h.

shogun feature type

Enumerator:
F_UNKNOWN 
F_BOOL 
F_CHAR 
F_BYTE 
F_SHORT 
F_WORD 
F_INT 
F_UINT 
F_LONG 
F_ULONG 
F_SHORTREAL 
F_DREAL 
F_LONGREAL 
F_ANY 

Definition at line 19 of file FeatureTypes.h.

Matrix decomposition method for Fisher LDA

Enumerator:
AUTO_FLDA 

if N>D then CLASSIC_FLDA is chosen automatically else CANVAR_FLDA is chosen (D-dimensions N-number of vectors)

CANVAR_FLDA 

Cannonical Variable based FLDA.

CLASSIC_FLDA 

Classical Fisher Linear Discriminant Analysis.

Definition at line 49 of file FisherLDA.h.

gradient availability

Enumerator:
GRADIENT_NOT_AVAILABLE 
GRADIENT_AVAILABLE 

Definition at line 80 of file SGObject.h.

inference type

Enumerator:
INF_NONE 
INF_EXACT 
INF_FITC 
INF_LAPLACIAN 
INF_EP 
INF_KL 

Definition at line 32 of file InferenceMethod.h.

kernel property

Enumerator:
KP_NONE 
KP_LINADD 
KP_KERNCOMBINATION 
KP_BATCHEVALUATION 

Definition at line 121 of file Kernel.h.

kernel type

Enumerator:
K_UNKNOWN 
K_LINEAR 
K_POLY 
K_GAUSSIAN 
K_GAUSSIANSHIFT 
K_GAUSSIANMATCH 
K_HISTOGRAM 
K_SALZBERG 
K_LOCALITYIMPROVED 
K_SIMPLELOCALITYIMPROVED 
K_FIXEDDEGREE 
K_WEIGHTEDDEGREE 
K_WEIGHTEDDEGREEPOS 
K_WEIGHTEDDEGREERBF 
K_WEIGHTEDCOMMWORDSTRING 
K_POLYMATCH 
K_ALIGNMENT 
K_COMMWORDSTRING 
K_COMMULONGSTRING 
K_SPECTRUMRBF 
K_SPECTRUMMISMATCHRBF 
K_COMBINED 
K_AUC 
K_CUSTOM 
K_SIGMOID 
K_CHI2 
K_DIAG 
K_CONST 
K_DISTANCE 
K_LOCALALIGNMENT 
K_PYRAMIDCHI2 
K_OLIGO 
K_MATCHWORD 
K_TPPK 
K_REGULATORYMODULES 
K_SPARSESPATIALSAMPLE 
K_HISTOGRAMINTERSECTION 
K_WAVELET 
K_WAVE 
K_CAUCHY 
K_TSTUDENT 
K_RATIONAL_QUADRATIC 
K_MULTIQUADRIC 
K_EXPONENTIAL 
K_SPHERICAL 
K_SPLINE 
K_ANOVA 
K_POWER 
K_LOG 
K_CIRCULAR 
K_INVERSEMULTIQUADRIC 
K_DISTANTSEGMENTS 
K_BESSEL 
K_JENSENSHANNON 
K_DIRECTOR 
K_PRODUCT 
K_LINEARARD 
K_GAUSSIANARD 
K_STREAMING 

Definition at line 57 of file Kernel.h.

training method

Enumerator:
KMM_MINI_BATCH 
KMM_LLOYD 

Definition at line 29 of file KMeans.h.

enum ELDAMethod

Method for solving LDA

Enumerator:
AUTO_LDA 

if N>D then FLD_LDA is chosen automatically else SVD_LDA is chosen (D-dimensions N-number of vectors)

SVD_LDA 

Singular Value Decomposition based LDA.

FLD_LDA 

Fisher two class discrimiant based LDA.

Definition at line 27 of file LDA.h.

type of likelihood model

Enumerator:
LT_NONE 
LT_GAUSSIAN 
LT_STUDENTST 
LT_LOGIT 
LT_PROBIT 
LT_SOFTMAX 

Definition at line 26 of file LikelihoodModel.h.

enum ELossType

shogun loss type

Enumerator:
L_HINGELOSS 
L_SMOOTHHINGELOSS 
L_SQUAREDHINGELOSS 
L_SQUAREDLOSS 
L_EXPONENTIALLOSS 
L_ABSOLUTEDEVIATIONLOSS 
L_HUBERLOSS 
L_LOGLOSS 
L_LOGLOSSMARGIN 

Definition at line 29 of file LossFunction.h.

classifier type

Enumerator:
CT_NONE 
CT_LIGHT 
CT_LIGHTONECLASS 
CT_LIBSVM 
CT_LIBSVMONECLASS 
CT_LIBSVMMULTICLASS 
CT_MPD 
CT_GPBT 
CT_CPLEXSVM 
CT_PERCEPTRON 
CT_KERNELPERCEPTRON 
CT_LDA 
CT_LPM 
CT_LPBOOST 
CT_KNN 
CT_SVMLIN 
CT_KERNELRIDGEREGRESSION 
CT_GNPPSVM 
CT_GMNPSVM 
CT_SVMPERF 
CT_LIBSVR 
CT_SVRLIGHT 
CT_LIBLINEAR 
CT_KMEANS 
CT_HIERARCHICAL 
CT_SVMOCAS 
CT_WDSVMOCAS 
CT_SVMSGD 
CT_MKLMULTICLASS 
CT_MKLCLASSIFICATION 
CT_MKLONECLASS 
CT_MKLREGRESSION 
CT_SCATTERSVM 
CT_DASVM 
CT_LARANK 
CT_DASVMLINEAR 
CT_GAUSSIANNAIVEBAYES 
CT_AVERAGEDPERCEPTRON 
CT_SGDQN 
CT_CONJUGATEINDEX 
CT_LINEARRIDGEREGRESSION 
CT_LEASTSQUARESREGRESSION 
CT_QDA 
CT_NEWTONSVM 
CT_GAUSSIANPROCESSREGRESSION 
CT_LARS 
CT_MULTICLASS 
CT_DIRECTORLINEAR 
CT_DIRECTORKERNEL 
CT_LIBQPSOSVM 
CT_PRIMALMOSEKSOSVM 
CT_CCSOSVM 
CT_GAUSSIANPROCESSBINARY 
CT_GAUSSIANPROCESSMULTICLASS 
CT_STOCHASTICSOSVM 
CT_NEURALNETWORK 
CT_BAGGING 
CT_FWSOSVM 
CT_BCFWSOSVM 
CT_GAUSSIANPROCESSCLASS 

Definition at line 33 of file Machine.h.

the following inference methods are acceptable: Tree Max Product, Loopy Max Product, LP Relaxation, Sequential Tree Reweighted Max Product (TRW-S), Graph cuts

Enumerator:
TREE_MAX_PROD 
LOOPY_MAX_PROD 
LP_RELAXATION 
TRWS_MAX_PROD 
GRAPH_CUT 

Definition at line 29 of file MAPInference.h.

mc sampler type

Enumerator:
MC_Probability 
MC_Mean 
MC_Variance 

Definition at line 54 of file SoftMaxLikelihood.h.

The io functions can optionally prepend the function name or the line from where the print occurred.

Enumerator:
MSG_NONE 
MSG_FUNCTION 
MSG_LINE_AND_FILE 

Definition at line 62 of file SGIO.h.

The io libs output [DEBUG] etc in front of every message 'higher' messages filter output depending on the loglevel, i.e. CRITICAL messages will print all MSG_CRITICAL TO MSG_EMERGENCY messages.

Enumerator:
MSG_GCDEBUG 
MSG_DEBUG 
MSG_INFO 
MSG_NOTICE 
MSG_WARN 
MSG_ERROR 
MSG_CRITICAL 
MSG_ALERT 
MSG_EMERGENCY 
MSG_MESSAGEONLY 

Definition at line 45 of file SGIO.h.

model selection availability

Enumerator:
MS_NOT_AVAILABLE 
MS_AVAILABLE 

Definition at line 74 of file SGObject.h.

value type of a model selection parameter node

Enumerator:
MSPT_NONE 

no type

MSPT_FLOAT64 
MSPT_INT32 
MSPT_FLOAT64_VECTOR 
MSPT_INT32_VECTOR 
MSPT_FLOAT64_SGVECTOR 
MSPT_INT32_SGVECTOR 

Definition at line 32 of file ModelSelectionParameters.h.

For autoencoders, specifies the position of the layer in the autoencoder, i.e an encoding layer or a decoding layer

Enumerator:
NLAP_NONE 

The layer is not a part of an autoencoder

NLAP_ENCODING 

The layer is an encoding layer

NLAP_DECODING 

The layer is a decoding layer

Definition at line 49 of file NeuralLayer.h.

optimization method for neural networks

Enumerator:
NNOM_GRADIENT_DESCENT 
NNOM_LBFGS 

Definition at line 49 of file NeuralNetwork.h.

normalizer type

Enumerator:
N_REGULAR 
N_MULTITASK 

Definition at line 23 of file KernelNormalizer.h.

enum for different method to approximate null-distibution

Enumerator:
PERMUTATION 
MMD2_SPECTRUM_DEPRECATED 
MMD2_SPECTRUM 
MMD2_GAMMA 
MMD1_GAUSSIAN 
HSIC_GAMMA 

Definition at line 51 of file HypothesisTest.h.

linear operator function types

Enumerator:
OF_SQRT 
OF_LOG 
OF_POLY 
OF_UNDEFINED 

Definition at line 23 of file OperatorFunction.h.

optimization type

Enumerator:
FASTBUTMEMHUNGRY 
SLOWBUTMEMEFFICIENT 

Definition at line 50 of file Kernel.h.

memory usage by PCA : In-place or through reallocation

Enumerator:
MEM_REALLOCATE 

The feature matrix replaced by new matrix with target dims. This requires memory for old matrix as well as new matrix at once for a short amount of time initially.

MEM_IN_PLACE 

The feature matrix is modified in-place to generate the new matrix. Output matrix dimensions are changed to target dims, but actual matrix size remains same internally. Modifies initial data matrix

Definition at line 53 of file PCA.h.

enum EPCAMethod

Matrix decomposition method for PCA

Enumerator:
AUTO 

if N>D then EVD is chosen automatically else SVD is chosen (D-dimensions N-number of vectors)

SVD 

SVD based PCA. Time complexity ~14dn^2 (d-dimensions n-number of vectors)

EVD 

Eigenvalue decomposition of covariance matrix. Time complexity ~10d^3 (d-dimensions n-number of vectors)

Definition at line 27 of file PCA.h.

enum EPCAMode

mode of pca

Enumerator:
THRESHOLD 

cut by threshold

VARIANCE_EXPLAINED 

variance explained

FIXED_NUMBER 

keep fixed number of features

Definition at line 42 of file PCA.h.

enumeration of possible preprocessor types used by Shogun UI

Note to developer: any new preprocessor should be added here.

Enumerator:
P_UNKNOWN 
P_NORMONE 
P_LOGPLUSONE 
P_SORTWORDSTRING 
P_SORTULONGSTRING 
P_SORTWORD 
P_PRUNEVARSUBMEAN 
P_DECOMPRESSSTRING 
P_DECOMPRESSCHARSTRING 
P_DECOMPRESSBYTESTRING 
P_DECOMPRESSWORDSTRING 
P_DECOMPRESSULONGSTRING 
P_RANDOMFOURIERGAUSS 
P_PCA 
P_KERNELPCA 
P_NORMDERIVATIVELEM3 
P_DIMENSIONREDUCTIONPREPROCESSOR 
P_SUMONE 
P_HOMOGENEOUSKERNELMAP 
P_PNORM 
P_RESCALEFEATURES 
P_FISHERLDA 
P_BAHSIC 

Definition at line 32 of file Preprocessor.h.

multiclass prob output heuristics in [1] OVA_NORM: simple normalization of probabilites, eq.(6) OVA_SOFTMAX: normalizing using softmax function, eq.(7) OVO_PRICE: proposed by Price et al. see method 1 in [1] OVO_HASTIE: proposed by Hastie et al. see method 2 [9] in [1] OVO_HAMAMURA: proposed by Hamamura et al. see eq.(14) in [1]

[1] J. Milgram, M. Cheriet, R.Sabourin, "One Against One" or "One Against One": Which One is Better for Handwriting Recognition with SVMs?

Enumerator:
PROB_HEURIS_NONE 
OVA_NORM 
OVA_SOFTMAX 
OVO_PRICE 
OVO_HASTIE 
OVO_HAMAMURA 

Definition at line 36 of file MulticlassStrategy.h.

problem type

Enumerator:
PT_BINARY 
PT_REGRESSION 
PT_MULTICLASS 
PT_STRUCTURED 
PT_LATENT 
PT_CLASS 

Definition at line 110 of file Machine.h.

enum EQPType

Enum Training method selection

Enumerator:
MOSEK 

MOSEK.

SVMLIGHT 

SVM^Light

Definition at line 29 of file CCSOSVM.h.

Enum to select which statistic type of quadratic time MMD should be computed

Enumerator:
BIASED 
BIASED_DEPRECATED 
UNBIASED 
UNBIASED_DEPRECATED 
INCOMPLETE 

Definition at line 47 of file QuadraticTimeMMD.h.

enum ERangeType

type of range

Enumerator:
R_LINEAR 
R_EXP 
R_LOG 

Definition at line 26 of file ModelSelectionParameters.h.

Enumerator:
RBMMM_RECONSTRUCTION_ERROR 
RBMMM_PSEUDO_LIKELIHOOD 

Definition at line 49 of file RBM.h.

Enumerator:
RBMVUT_BINARY 
RBMVUT_GAUSSIAN 
RBMVUT_SOFTMAX 

Definition at line 55 of file RBM.h.

type of regressor

Enumerator:
RT_NONE 
RT_LIGHT 
RT_LIBSVM 

Definition at line 19 of file Regression.h.

enum ESolver

Enum Training method selection

Enumerator:
BMRM 

Standard BMRM algorithm.

PPBMRM 

Proximal Point BMRM (BMRM with prox-term)

P3BMRM 

Proximal Point P-BMRM (multiple cutting plane models)

NCBM 

Definition at line 27 of file DualLibQPBMSOSVM.h.

solver type

Enumerator:
ST_AUTO 
ST_CPLEX 
ST_GLPK 
ST_NEWTON 
ST_DIRECT 
ST_ELASTICNET 
ST_BLOCK_NORM 

Definition at line 98 of file Machine.h.

Stochastic Proximity Embedding (SPE) strategy

Enumerator:
SPE_GLOBAL 
SPE_LOCAL 

Definition at line 23 of file StochasticProximityEmbedding.h.

state model type

Enumerator:
SMT_UNKNOWN 
SMT_TWO_STATE 

Definition at line 20 of file StateModelTypes.h.

enum for different statistic types

Enumerator:
S_LINEAR_TIME_MMD 
S_QUADRATIC_TIME_MMD 
S_HSIC 
S_NOCCO 

Definition at line 42 of file HypothesisTest.h.

The structured empirical risk types, corresponding to different training objectives [1].

[1] T. Joachims, T. Finley, Chun-Nam Yu, Cutting-Plane Training of Structural SVMs, Machine Learning Journal, 2009.

Enumerator:
N_SLACK_MARGIN_RESCALING 
N_SLACK_SLACK_RESCALING 
ONE_SLACK_MARGIN_RESCALING 
ONE_SLACK_SLACK_RESCALING 
CUSTOMIZED_RISK 

Definition at line 35 of file StructuredOutputMachine.h.

structured data type

Enumerator:
SDT_UNKNOWN 
SDT_REAL 
SDT_SEQUENCE 
SDT_FACTOR_GRAPH 
SDT_SPARSE_MULTILABEL 

Definition at line 20 of file StructuredDataTypes.h.

Enumerator:
SOURCE 

source terminal

SINK 

sink terminal

Definition at line 31 of file GraphCut.h.

which training method to use for KRR

Enumerator:
PINV 

via pseudo inverse

GS 

or gauss-seidel iterative method

Definition at line 27 of file KernelRidgeRegression.h.

Ways to transform inputs

Enumerator:
T_LINEAR 

Linear.

T_LOG 

Logarithmic.

T_LOG_PLUS1 

Logarithmic (log(1+x))

T_LOG_PLUS3 

Logarithmic (log(3+x))

T_LINEAR_PLUS3 

Linear (3+x)

Definition at line 25 of file Plif.h.

Enum EVwCacheType specifies the type of cache used, either C_NATIVE or C_PROTOBUF.

Enumerator:
C_NATIVE 
C_PROTOBUF 

Definition at line 29 of file VwCacheReader.h.

WD kernel type

Enumerator:
E_WD 
E_EXTERNAL 
E_BLOCK_CONST 
E_BLOCK_LINEAR 
E_BLOCK_SQPOLY 
E_BLOCK_CUBICPOLY 
E_BLOCK_EXP 
E_BLOCK_LOG 

Definition at line 27 of file WeightedDegreeStringKernel.h.

Type of spectral windowing function.

Enumerator:
HomogeneousKernelMapWindowUniform 

uniform window

HomogeneousKernelMapWindowRectangular 

rectangular window

Definition at line 32 of file HomogeneousKernelMap.h.

Type of kernel.

Enumerator:
HomogeneousKernelIntersection 

intersection kernel

HomogeneousKernelChi2 

Chi2 kernel

HomogeneousKernelJS 

Jensen-Shannon kernel

Definition at line 25 of file HomogeneousKernelMap.h.

enum KernelName

names of kernels that can be approximated currently

Enumerator:
GAUSSIAN 

approximate gaussian kernel expects one parameter to be specified : kernel width

NOT_SPECIFIED 

not specified

Definition at line 26 of file RandomFourierDotFeatures.h.

liblinar regression solver type

Enumerator:
L2R_L2LOSS_SVR 

L2 regularized support vector regression with L2 epsilon tube loss.

L2R_L1LOSS_SVR_DUAL 

L2 regularized support vector regression with L1 epsilon tube loss.

L2R_L2LOSS_SVR_DUAL 

L2 regularized support vector regression with L2 epsilon tube loss (dual)

Definition at line 22 of file LibLinearRegression.h.

liblinar solver type

Enumerator:
L2R_LR 

L2 regularized linear logistic regression.

L2R_L2LOSS_SVC_DUAL 

L2 regularized SVM with L2-loss using dual coordinate descent.

L2R_L2LOSS_SVC 

L2 regularized SVM with L2-loss using newton in the primal.

L2R_L1LOSS_SVC_DUAL 

L2 regularized linear SVM with L1-loss using dual coordinate descent.

L1R_L2LOSS_SVC 

L1 regularized SVM with L2-loss using dual coordinate descent.

L1R_LR 

L1 regularized logistic regression.

L2R_LR_DUAL 

L2 regularized linear logistic regression via dual.

Definition at line 25 of file LibLinear.h.

scatter svm variant

Enumerator:
NO_BIAS_LIBSVM 

no bias w/ libsvm

NO_BIAS_SVMLIGHT 

no bias w/ svmlight

TEST_RULE1 

training with bias using test rule 1

TEST_RULE2 

training with bias using test rule 2

Definition at line 25 of file ScatterSVM.h.

Function Documentation

void add_cutting_plane ( bmrm_ll **  tail,
bool *  map,
float64_t A,
uint32_t  free_idx,
float64_t cp_data,
uint32_t  dim 
)

Add cutting plane

Parameters
tailPointer to the last CP entry
mapPointer to map storing info about CP physical memory
ACP physical memory
free_idxIndex to physical memory where the CP data will be stored
cp_dataCP data
dimDimension of CP data

Definition at line 29 of file libbmrm.cpp.

void shogun::alloc ( v_array< T > &  v,
int  length 
)

Used to modify the capacity of the vector

Parameters
vvector
lengththe new length of the vector

Definition at line 78 of file JLCoverTreePoint.h.

void clean_icp ( ICP_stats *  icp_stats,
BmrmStatistics &  bmrm,
bmrm_ll **  head,
bmrm_ll **  tail,
float64_t *&  H,
float64_t *&  diag_H,
float64_t *&  beta,
bool *&  map,
uint32_t  cleanAfter,
float64_t *&  b,
uint32_t *&  I,
uint32_t  cp_models = 0 
)

Clean-up in-active cutting planes

Definition at line 92 of file libbmrm.cpp.

virtual T shogun::compute ( SGVector< T >  vector1,
SGVector< T >  vector2 
) const
virtual

Template class for Eigen3 dot product that performs dot product operation( \(\sum_{i=1}^d a_ib_i$ where $a,b$ are $d$-dimensional vectors) using Eigen3 Library */ template <class T> class DenseEigen3DotProduct : public VectorDotProduct<T, SGVector<T> > { public: /** Constructor */ DenseEigen3DotProduct() : VectorDotProduct<T, SGVector<T> >() { } /** Compute method which performs dot product operation (\){i=1}^d a_ib_i$ where $a,b$ are $d$-dimensional vectors)

Parameters
vector1input vector
vector2input vector
Returns
the result
double shogun::compute_lambda ( double *  ATx,
double  z,
CDotFeatures *  features,
double *  y,
int  n_vecs,
int  n_feats,
int  n_blocks,
const slep_options &  options 
)

Definition at line 106 of file slep_solver.cpp.

double shogun::compute_regularizer ( double *  w,
double  lambda,
double  lambda2,
int  n_vecs,
int  n_feats,
int  n_blocks,
const slep_options &  options 
)

Definition at line 23 of file slep_solver.cpp.

SGVector<T> shogun::create_range_array ( min,
max,
ERangeType  type,
step,
type_base 
)

Creates an array of values specified by the parameters. A minimum and a maximum is specified, step interval, and an ERangeType (s. above) of the range, which is used to fill an array with concrete values. For some range types, a base is required. All values are given by void pointers to them (type conversion is done via m_value_type variable).

Parameters
minminimum of desired range. Requires min<max
maxmaximum of desired range. Requires min<max
typethe way the values are created, see ERangeType
stepincrement instaval for the values
type_basebase for EXP or LOG ranges

Definition at line 210 of file ModelSelectionParameters.h.

float shogun::distance ( CJLCoverTreePoint  p1,
CJLCoverTreePoint  p2,
float64_t  upper_bound 
)

Functions declared out of the class definition to respect JLCoverTree structure

Call m_distance->distance() with the proper index order depending on the feature containers in m_distance for each of the points

Definition at line 138 of file JLCoverTreePoint.h.

void exit_shogun ( )

This function must be called when one stops using libshogun. It will perform a number of cleanups

Definition at line 100 of file init.cpp.

uint32_t shogun::find_free_idx ( bool *  map,
uint32_t  size 
)

Get index of free slot for new cutting plane

Parameters
mapPointer to map storing info about CP physical memory
sizeSize of the CP buffer
Returns
Index of unoccupied memory field in CP physical memory

Definition at line 127 of file libbmrm.h.

static const float64_t* shogun::get_col ( uint32_t  i)
static

Definition at line 26 of file libncbm.cpp.

static const float64_t* shogun::get_col ( uint32_t  i)
static

Definition at line 28 of file libp3bm.cpp.

static const float64_t* shogun::get_col ( uint32_t  i)
static

Definition at line 28 of file libppbm.cpp.

static const double* shogun::get_col ( uint32_t  j)
static

Definition at line 28 of file malsar_clustered.cpp.

static const float64_t* shogun::get_col ( uint32_t  i)
static

Definition at line 179 of file libbmrm.cpp.

float64_t* shogun::get_cutting_plane ( bmrm_ll *  ptr)

Get cutting plane

Parameters
ptrPointer to some CP entry
Returns
Pointer to cutting plane at given entry

Definition at line 119 of file libbmrm.h.

SGIO * get_global_io ( )

get the global io object

Returns
io object

Definition at line 131 of file init.cpp.

CMath * get_global_math ( )

get the global math object

Returns
math object

Definition at line 170 of file init.cpp.

Parallel * get_global_parallel ( )

get the global parallel object

Returns
parallel object

Definition at line 144 of file init.cpp.

CRandom * get_global_rand ( )

get the global random object

Returns
random object

Definition at line 183 of file init.cpp.

Version * get_global_version ( )

get the global version object

Returns
version object

Definition at line 157 of file init.cpp.

void init_from_env ( )

Checks environment variables and modifies global objects

Definition at line 189 of file init.cpp.

void init_shogun ( void(*)(FILE *target, const char *str)  print_message = NULL,
void(*)(FILE *target, const char *str)  print_warning = NULL,
void(*)(FILE *target, const char *str)  print_error = NULL,
void(*)(bool &delayed, bool &immediately)  cancel_computations = NULL 
)

This function must be called before libshogun is used. Usually shogun does not provide any output messages (neither debugging nor error; apart from exceptions). This function allows one to specify customized output callback functions and a callback function to check for exceptions:

Parameters
print_messagefunction pointer to print a message
print_warningfunction pointer to print a warning message
print_errorfunction pointer to print an error message (this will be printed before shogun throws an exception)
cancel_computationsfunction pointer to check for exception

Definition at line 54 of file init.cpp.

void init_shogun_with_defaults ( )

init shogun with defaults

Definition at line 94 of file init.cpp.

static larank_kcache_t* shogun::larank_kcache_create ( CKernel kernelfunc)
static

Definition at line 65 of file LaRank.cpp.

static void shogun::larank_kcache_destroy ( larank_kcache_t *  self)
static

Definition at line 133 of file LaRank.cpp.

static float64_t shogun::larank_kcache_query ( larank_kcache_t *  self,
int32_t  i,
int32_t  j 
)
static

Definition at line 324 of file LaRank.cpp.

static float32_t* shogun::larank_kcache_query_row ( larank_kcache_t *  self,
int32_t  i,
int32_t  len 
)
static

Definition at line 356 of file LaRank.cpp.

static int32_t* shogun::larank_kcache_r2i ( larank_kcache_t *  self,
int32_t  n 
)
static

Definition at line 197 of file LaRank.cpp.

static void shogun::larank_kcache_set_buddy ( larank_kcache_t *  self,
larank_kcache_t *  buddy 
)
static

Definition at line 333 of file LaRank.cpp.

static void shogun::larank_kcache_set_maximum_size ( larank_kcache_t *  self,
int64_t  entries 
)
static

Definition at line 125 of file LaRank.cpp.

static void shogun::larank_kcache_swap_ri ( larank_kcache_t *  self,
int32_t  r1,
int32_t  i2 
)
static

Definition at line 290 of file LaRank.cpp.

static void shogun::larank_kcache_swap_rr ( larank_kcache_t *  self,
int32_t  r1,
int32_t  r2 
)
static

Definition at line 284 of file LaRank.cpp.

int32_t shogun::lbfgs ( int32_t  n,
float64_t x,
float64_t ptr_fx,
lbfgs_evaluate_t  proc_evaluate,
lbfgs_progress_t  proc_progress,
void *  instance,
lbfgs_parameter_t *  _param,
lbfgs_adjust_step_t  proc_adjust_step 
)

Definition at line 208 of file lbfgs.cpp.

static int32_t line_search_backtracking ( int32_t  n,
float64_t x,
float64_t f,
float64_t g,
float64_t s,
float64_t stp,
const float64_t xp,
const float64_t gp,
float64_t wa,
callback_data_t *  cd,
const lbfgs_parameter_t *  param 
)
static

Definition at line 606 of file lbfgs.cpp.

static int32_t line_search_backtracking_owlqn ( int32_t  n,
float64_t x,
float64_t f,
float64_t g,
float64_t s,
float64_t stp,
const float64_t xp,
const float64_t gp,
float64_t wp,
callback_data_t *  cd,
const lbfgs_parameter_t *  param 
)
static

Definition at line 722 of file lbfgs.cpp.

static int32_t line_search_morethuente ( int32_t  n,
float64_t x,
float64_t f,
float64_t g,
float64_t s,
float64_t stp,
const float64_t xp,
const float64_t gp,
float64_t wa,
callback_data_t *  cd,
const lbfgs_parameter_t *  param 
)
static

Definition at line 798 of file lbfgs.cpp.

std::vector<line_search_res> shogun::line_search_with_strong_wolfe ( CDualLibQPBMSOSVM *  machine,
float64_t  lambda,
float64_t  initial_val,
SGVector< float64_t > &  initial_solution,
SGVector< float64_t > &  initial_grad,
SGVector< float64_t > &  search_dir,
float64_t  astart,
float64_t  amax = 1.1,
float64_t  wolfe_c1 = 1E-4,
float64_t  wolfe_c2 = 0.9,
float64_t  max_iter = 5 
)

Definition at line 159 of file libncbm.cpp.

malsar_result_t malsar_clustered ( CDotFeatures *  features,
double *  y,
double  rho1,
double  rho2,
const malsar_options &  options 
)

Routine for learning a linear multitask logistic regression model using Clustered multitask learning algorithm.

Definition at line 33 of file malsar_clustered.cpp.

malsar_result_t malsar_joint_feature_learning ( CDotFeatures *  features,
double *  y,
double  rho1,
double  rho2,
const malsar_options &  options 
)

Routine for learning a linear multitask logistic regression model using Joint Feature algorithm.

Definition at line 24 of file malsar_joint_feature_learning.cpp.

malsar_result_t malsar_low_rank ( CDotFeatures *  features,
double *  y,
double  rho,
const malsar_options &  options 
)

Routine for learning a linear multitask logistic regression model using Low Rank multitask algorithm.

Definition at line 22 of file malsar_low_rank.cpp.

CSGObject * new_sgserializable ( const char *  sgserializable_name,
EPrimitiveType  generic 
)

new shogun serializable

Parameters
sgserializable_name
generic

Definition at line 2268 of file class_list.cpp.

float32_t one_pf_quad_predict ( float32_t weights,
VwFeature &  f,
v_array< VwFeature > &  cross_features,
vw_size_t  mask 
)

Get the prediction contribution from one feature.

Parameters
weightsweights
ffeature
cross_featurespaired features
maskmask
Returns
prediction from one feature

Definition at line 40 of file vw_math.cpp.

float32_t one_pf_quad_predict_trunc ( float32_t weights,
VwFeature &  f,
v_array< VwFeature > &  cross_features,
vw_size_t  mask,
float32_t  gravity 
)

Get the prediction contribution from one feature.

Weights are taken as truncated weights.

Parameters
weightsweights
ffeature
cross_featurespaired features
maskmask
gravityweight threshold value
Returns
prediction from one feature

Definition at line 48 of file vw_math.cpp.

static void owlqn_project ( float64_t d,
const float64_t sign,
const int32_t  start,
const int32_t  end 
)
static

Definition at line 1349 of file lbfgs.cpp.

static void owlqn_pseudo_gradient ( float64_t pg,
const float64_t x,
const float64_t g,
const int32_t  n,
const float64_t  c,
const int32_t  start,
const int32_t  end 
)
static

Definition at line 1306 of file lbfgs.cpp.

static float64_t owlqn_x1norm ( const float64_t x,
const int32_t  start,
const int32_t  n 
)
static

Definition at line 1290 of file lbfgs.cpp.

v_array< CJLCoverTreePoint > shogun::parse_points ( CDistance *  distance,
EFeaturesContainer  fc 
)

Fills up a v_array of CJLCoverTreePoint objects

Definition at line 183 of file JLCoverTreePoint.h.

v_array<T> shogun::pop ( v_array< v_array< T > > &  stack)

Returns the vector previous to the pointed one in the stack of vectors and decrements the index of the stack. No memory is freed here. If there are no vectors stored in the stack, create and return a new empty vector

Parameters
stackof vectors
Returns
the adequate vector according to the previous conditions

Definition at line 94 of file JLCoverTreePoint.h.

void shogun::print ( CJLCoverTreePoint &  p)

Print the information of the CoverTree point

Definition at line 207 of file JLCoverTreePoint.h.

void shogun::projection ( double *  w,
double *  v,
int  n_feats,
int  n_blocks,
double  lambda,
double  lambda2,
double  L,
double *  z,
double *  z0,
const slep_options &  options 
)

Definition at line 277 of file slep_solver.cpp.

void shogun::push ( v_array< T > &  v,
const T &  new_ele 
)

Insert a new element at the end of the vector

Parameters
vvector
new_eleelement to insert

Definition at line 61 of file JLCoverTreePoint.h.

float32_t shogun::real_weight ( float32_t  w,
float32_t  gravity 
)

Get the truncated weight value

Parameters
wweight
gravitythreshold for the weight
Returns
truncated weight

Definition at line 35 of file vw_math.h.

void remove_cutting_plane ( bmrm_ll **  head,
bmrm_ll **  tail,
bool *  map,
float64_t icp 
)

Remove cutting plane at given index

Parameters
headPointer to the first CP entry
tailPointer to the last CP entry
mapPointer to map storing info about CP physical memory
icpPointer to inactive CP that should be removed

Definition at line 59 of file libbmrm.cpp.

float32_t sd_offset_add ( float32_t weights,
vw_size_t  mask,
VwFeature *  begin,
VwFeature *  end,
vw_size_t  offset 
)

Dot product of feature vector with the weight vector with an offset added to the feature indices.

Parameters
weightsweight vector
maskmask
beginfirst feature of the vector
endlast feature of the vector
offsetindex offset
Returns
dot product

Definition at line 20 of file vw_math.cpp.

float32_t sd_offset_truncadd ( float32_t weights,
vw_size_t  mask,
VwFeature *  begin,
VwFeature *  end,
vw_size_t  offset,
float32_t  gravity 
)

Dot product of feature vector with the weight vector with an offset added to the feature indices.

Weights are taken as the truncated weights.

Parameters
weightsweights
maskmask
beginfirst feature of the vector
endlast feature of the vector
offsetindex offset
gravityweight threshold value
Returns
dot product

Definition at line 28 of file vw_math.cpp.

double shogun::search_point_gradient_and_objective ( CDotFeatures *  features,
double *  ATx,
double *  As,
double *  sc,
double *  y,
int  n_vecs,
int  n_feats,
int  n_tasks,
double *  g,
double *  gc,
const slep_options &  options 
)

Definition at line 313 of file slep_solver.cpp.

void set_global_io ( SGIO *  io)

set the global io object

Parameters
ioio object to use

Definition at line 124 of file init.cpp.

void set_global_math ( CMath *  math)

set the global math object

Parameters
mathmath object to use

Definition at line 163 of file init.cpp.

void set_global_parallel ( Parallel *  parallel)

set the global parallel object

Parameters
parallelparallel object to use

Definition at line 137 of file init.cpp.

void set_global_rand ( CRandom *  rand)

set the global random object

Parameters
randrandom object to use

Definition at line 176 of file init.cpp.

void set_global_version ( Version *  version)

set the global version object

Parameters
versionversion object to use

Definition at line 150 of file init.cpp.

void* shogun::sg_calloc ( size_t  num,
size_t  size 
)

Definition at line 227 of file memory.cpp.

void shogun::sg_free ( void *  ptr)

Definition at line 263 of file memory.cpp.

void shogun::sg_global_print_default ( FILE *  target,
const char *  str 
)

Definition at line 89 of file init.cpp.

void* shogun::sg_malloc ( size_t  size)

Definition at line 194 of file memory.cpp.

void* shogun::sg_realloc ( void *  ptr,
size_t  size 
)

Definition at line 279 of file memory.cpp.

slep_result_t slep_mc_plain_lr ( CDotFeatures features,
CMulticlassLabels labels,
float64_t  z,
const slep_options &  options 
)

Accelerated projected gradient solver for multiclass logistic regression problem with feature tree regularization.

Parameters
featuresfeatures to be used
labelslabels to be used
zregularization ratio
optionsoptions of solver

(n_vecs*n_classes);

Definition at line 27 of file slep_mc_plain_lr.cpp.

slep_result_t slep_mc_tree_lr ( CDotFeatures features,
CMulticlassLabels labels,
float64_t  z,
const slep_options &  options 
)

Accelerated projected gradient solver for multiclass logistic regression problem with feature tree regularization.

Parameters
featuresfeatures to be used
labelslabels to be used
zregularization ratio
optionsoptions of solver

(n_vecs*n_classes);

Definition at line 29 of file slep_mc_tree_lr.cpp.

slep_result_t slep_solver ( CDotFeatures *  features,
double *  y,
double  z,
const slep_options &  options 
)

Learning optimization task solver ported from the SLEP (Sparse LEarning Package) library.

Based on accelerated projected gradient method.

Supports two types of losses: logistic and least squares.

Supports multitask problems (task group [MULTITASK_GROUP] and task tree [MULTITASK_TREE] relations), problems with feature relations (feature group [FEATURE_GROUP] and feature tree [FEATURE_TREE]), basic regularized problems [PLAIN] and fused formulation.

Definition at line 402 of file slep_solver.cpp.

BmrmStatistics svm_bmrm_solver ( CDualLibQPBMSOSVM *  machine,
float64_t W,
float64_t  TolRel,
float64_t  TolAbs,
float64_t  _lambda,
uint32_t  _BufSize,
bool  cleanICP,
uint32_t  cleanAfter,
float64_t  K,
uint32_t  Tmax,
bool  store_train_info 
)

Standard BMRM Solver for Structured Output Learning

Parameters
machinePointer to the BMRM machine
WWeight vector
TolRelRelative tolerance
TolAbsAbsolute tolerance
_lambdaRegularization constant
_BufSizeSize of the CP buffer (i.e. maximal number of iterations)
cleanICPFlag that enables/disables inactive cutting plane removal feature
cleanAfterNumber of iterations that should be cutting plane inactive for to be removed
KParameter K
TmaxParameter Tmax
store_train_infoFlag that enable/disable store training infomation, e.g., primal, dual, training error
Returns
Structure with BMRM algorithm result

Definition at line 184 of file libbmrm.cpp.

BmrmStatistics svm_ncbm_solver ( CDualLibQPBMSOSVM *  machine,
float64_t w,
float64_t  TolRel,
float64_t  TolAbs,
float64_t  _lambda,
uint32_t  _BufSize,
bool  cleanICP,
uint32_t  cleanAfter,
bool  is_convex = false,
bool  line_search = true,
bool  verbose = false 
)

NCBM (non-convex bundle method) solver Solves any unconstrainedminimization problem in the form of: min lambda/2 ||w||^2 + R(w) where R(w) is a risk funciton of any kind.

Definition at line 308 of file libncbm.cpp.

BmrmStatistics svm_p3bm_solver ( CDualLibQPBMSOSVM *  machine,
float64_t W,
float64_t  TolRel,
float64_t  TolAbs,
float64_t  _lambda,
uint32_t  _BufSize,
bool  cleanICP,
uint32_t  cleanAfter,
float64_t  K,
uint32_t  Tmax,
uint32_t  cp_models,
bool  verbose 
)

Proximal Point P-BMRM (multiple cutting plane models) Solver for Structured Output Learning

Parameters
machinePointer to the BMRM machine
WWeight vector
TolRelRelative tolerance
TolAbsAbsolute tolerance
_lambdaRegularization constant
_BufSizeSize of the CP buffer (i.e. maximal number of iterations)
cleanICPFlag that enables/disables inactive cutting plane removal feature
cleanAfterNumber of iterations that should be cutting plane inactive for to be removed
KParameter K
TmaxParameter Tmax
cp_modelsCount of cutting plane models to be used
verboseFlag that enables/disables screen output
Returns
Structure with BMRM algorithm result

Definition at line 33 of file libp3bm.cpp.

BmrmStatistics svm_ppbm_solver ( CDualLibQPBMSOSVM *  machine,
float64_t W,
float64_t  TolRel,
float64_t  TolAbs,
float64_t  _lambda,
uint32_t  _BufSize,
bool  cleanICP,
uint32_t  cleanAfter,
float64_t  K,
uint32_t  Tmax,
bool  verbose 
)

Proximal Point BMRM Solver for Structured Output Learning

Parameters
machinePointer to the BMRM machine
WWeight vector
TolRelRelative tolerance
TolAbsAbsolute tolerance
_lambdaRegularization constant
_BufSizeSize of the CP buffer (i.e. maximal number of iterations)
cleanICPFlag that enables/disables inactive cutting plane removal feature
cleanAfterNumber of iterations that should be cutting plane inactive for to be removed
KParameter K
TmaxParameter Tmax
verboseFlag that enables/disables screen output
Returns
Structure with BMRM algorithm result

Definition at line 33 of file libppbm.cpp.

void shogun::update_H ( BmrmStatistics &  ncbm,
bmrm_ll *  head,
bmrm_ll *  tail,
SGMatrix< float64_t > &  H,
SGVector< float64_t > &  diag_H,
float64_t  lambda,
uint32_t  maxCP,
int32_t  w_dim 
)

Definition at line 274 of file libncbm.cpp.

static int32_t update_trial_interval ( float64_t x,
float64_t fx,
float64_t dx,
float64_t y,
float64_t fy,
float64_t dy,
float64_t t,
float64_t ft,
float64_t dt,
const float64_t  tmin,
const float64_t  tmax,
int32_t *  brackt 
)
static

Update a safeguarded trial value and interval for line search.

The parameter x represents the step with the least function value. The parameter t represents the current step. This function assumes that the derivative at the point of x in the direction of the step. If the bracket is set to true, the minimizer has been bracketed in an interval of uncertainty with endpoints between x and y.

Parameters
xThe pointer to the value of one endpoint.
fxThe pointer to the value of f(x).
dxThe pointer to the value of f'(x).
yThe pointer to the value of another endpoint.
fyThe pointer to the value of f(y).
dyThe pointer to the value of f'(y).
tThe pointer to the value of the trial value, t.
ftThe pointer to the value of f(t).
dtThe pointer to the value of f'(t).
tminThe minimum value for the trial value, t.
tmaxThe maximum value for the trial value, t.
bracktThe pointer to the predicate if the trial value is bracketed.
Return values
int32_tStatus value. Zero indicates a normal termination.
See Also
Jorge J. More and David J. Thuente. Line search algorithm with guaranteed sufficient decrease. ACM Transactions on Mathematical Software (TOMS), Vol 20, No 3, pp. 286-307, 1994.

Definition at line 1117 of file lbfgs.cpp.

void shogun::wrap_dgeqrf ( int  m,
int  n,
double *  a,
int  lda,
double *  tau,
int *  info 
)

Definition at line 271 of file lapack.cpp.

void shogun::wrap_dgesvd ( char  jobu,
char  jobvt,
int  m,
int  n,
double *  a,
int  lda,
double *  sing,
double *  u,
int  ldu,
double *  vt,
int  ldvt,
int *  info 
)

Definition at line 253 of file lapack.cpp.

void shogun::wrap_dorgqr ( int  m,
int  n,
int  k,
double *  a,
int  lda,
double *  tau,
int *  info 
)

Definition at line 289 of file lapack.cpp.

void shogun::wrap_dstemr ( char  jobz,
char  range,
int  n,
double *  diag,
double *  subdiag,
double  vl,
double  vu,
int  il,
int  iu,
int *  m,
double *  w,
double *  z__,
int  ldz,
int  nzc,
int *  isuppz,
int  tryrac,
int *  info 
)

Definition at line 373 of file lapack.cpp.

void shogun::wrap_dsyev ( char  jobz,
char  uplo,
int  n,
double *  a,
int  lda,
double *  w,
int *  info 
)

Definition at line 235 of file lapack.cpp.

void shogun::wrap_dsyevr ( char  jobz,
char  uplo,
int  n,
double *  a,
int  lda,
int  il,
int  iu,
double *  eigenvalues,
double *  eigenvectors,
int *  info 
)

Definition at line 307 of file lapack.cpp.

void shogun::wrap_dsygvx ( int  itype,
char  jobz,
char  uplo,
int  n,
double *  a,
int  lda,
double *  b,
int  ldb,
int  il,
int  iu,
double *  eigenvalues,
double *  eigenvectors,
int *  info 
)

Definition at line 342 of file lapack.cpp.

static void shogun::xextend ( larank_kcache_t *  self,
int32_t  k,
int32_t  nlen 
)
static

Definition at line 203 of file LaRank.cpp.

static void shogun::xminsize ( larank_kcache_t *  self,
int32_t  n 
)
static

Definition at line 166 of file LaRank.cpp.

static void shogun::xpurge ( larank_kcache_t *  self)
static

Definition at line 111 of file LaRank.cpp.

static float64_t shogun::xquery ( larank_kcache_t *  self,
int32_t  i,
int32_t  j 
)
static

Definition at line 296 of file LaRank.cpp.

static void shogun::xswap ( larank_kcache_t *  self,
int32_t  i1,
int32_t  i2,
int32_t  r1,
int32_t  r2 
)
static

Definition at line 222 of file LaRank.cpp.

static void shogun::xtruncate ( larank_kcache_t *  self,
int32_t  k,
int32_t  nlen 
)
static

Definition at line 85 of file LaRank.cpp.

static line_search_res shogun::zoom ( CDualLibQPBMSOSVM *  machine,
float64_t  lambda,
float64_t  a_lo,
float64_t  a_hi,
float64_t  initial_fval,
SGVector< float64_t > &  initial_solution,
SGVector< float64_t > &  search_dir,
float64_t  wolfe_c1,
float64_t  wolfe_c2,
float64_t  init_lgrad,
float64_t  f_lo,
float64_t  g_lo,
float64_t  f_hi,
float64_t  g_hi 
)
static

Definition at line 42 of file libncbm.cpp.

virtual shogun::~DenseEigen3DotProduct ( )
virtual

Destructor

Definition at line 71 of file DenseEigen3DotProduct.h.

Variable Documentation

const lbfgs_parameter_t _defparam
static
Initial value:
{
6, 1e-5, 0, 0.0,
1e-20, 1e20, 1e-4, 0.9, 0.9, 1.0e-16,
0.0, 0, 1,
}

Definition at line 99 of file lbfgs.cpp.

uint32_t BufSize

Definition at line 27 of file libbmrm.cpp.

const int32_t constant_hash = 11650396

Constant used to access the constant feature.

Definition at line 32 of file vw_constants.h.

const float64_t epsilon = 0.0
static

Definition at line 21 of file libp3bm.cpp.

const float64_t epsilon = 0.0
static

Definition at line 21 of file libppbm.cpp.

const float64_t epsilon = 0.0
static

Definition at line 24 of file libbmrm.cpp.

const float64_t epsilon = 0.0
static

Definition at line 24 of file libncbm.cpp.

float64_t* H
static

Definition at line 23 of file libp3bm.cpp.

float64_t* H
static

Definition at line 23 of file libppbm.cpp.

float64_t* H
static

Definition at line 26 of file libbmrm.cpp.

float64_t * H2
static

Definition at line 23 of file libp3bm.cpp.

float64_t * H2
static

Definition at line 23 of file libppbm.cpp.

double* H_diag_matrix
static

Definition at line 25 of file malsar_clustered.cpp.

int H_diag_matrix_ld
static

Definition at line 26 of file malsar_clustered.cpp.

const uint32_t hash_base = 97562527

Seed for hash.

Definition at line 35 of file vw_constants.h.

float64_t* HMatrix
static

Definition at line 22 of file libncbm.cpp.

const int32_t LOGSUM_TBL = 10000

span of the logsum table

Definition at line 23 of file LocalAlignmentStringKernel.h.

uint32_t maxCPs
static

Definition at line 23 of file libncbm.cpp.

const float64_t Q_test_statistic_values[10][8]
static
Initial value:
{
{0.713,0.683,0.637,0.597,0.551,0.477,0.409,0.325},
{0.627,0.604,0.568,0.538,0.503,0.450,0.401,0.339},
{0.539,0.517,0.484,0.456,0.425,0.376,0.332,0.278},
{0.490,0.469,0.438,0.412,0.382,0.337,0.295,0.246},
{0.460,0.439,0.410,0.384,0.355,0.312,0.272,0.226},
{0.437,0.417,0.388,0.363,0.336,0.294,0.256,0.211},
{0.422,0.403,0.374,0.349,0.321,0.280,0.244,0.201},
{0.408,0.389,0.360,0.337,0.310,0.270,0.234,0.192},
{0.397,0.377,0.350,0.326,0.300,0.261,0.226,0.185},
{0.387,0.368,0.341,0.317,0.292,0.253,0.219,0.179}
}

Definition at line 78 of file RejectionStrategy.h.

const uint32_t QPSolverMaxIter = 0xFFFFFFFF
static

Definition at line 20 of file libp3bm.cpp.

const uint32_t QPSolverMaxIter = 0xFFFFFFFF
static

Definition at line 20 of file libppbm.cpp.

const uint32_t QPSolverMaxIter = 0xFFFFFFFF
static

Definition at line 23 of file libbmrm.cpp.

const int32_t quadratic_constant = 27942141

Constant used while hashing/accessing quadratic features.

Definition at line 29 of file vw_constants.h.

void(* sg_cancel_computations)(bool &delayed, bool &immediately) = NULL

function called to cancel things

Definition at line 51 of file init.cpp.

SGIO * sg_io = NULL

shogun IO

Definition at line 36 of file init.cpp.

CMath* sg_math = NULL

Definition at line 38 of file init.cpp.

Parallel * sg_parallel = NULL

Definition at line 35 of file init.cpp.

void(* sg_print_error)(FILE *target, const char *str) = NULL

function called to print error messages

Definition at line 48 of file init.cpp.

void(* sg_print_message)(FILE *target, const char *str) = NULL

function called to print normal messages

Definition at line 42 of file init.cpp.

void(* sg_print_warning)(FILE *target, const char *str) = NULL

function called to print warning messages

Definition at line 45 of file init.cpp.

CRandom * sg_rand = NULL

random number generator

Definition at line 39 of file init.cpp.

Version * sg_version = NULL

Definition at line 37 of file init.cpp.


SHOGUN Machine Learning Toolbox - Documentation