SG++-Doxygen-Documentation
sgpp::datadriven::AlgorithmAdaBoostBase Class Referenceabstract

#include <AlgorithmAdaBoostBase.hpp>

Inheritance diagram for sgpp::datadriven::AlgorithmAdaBoostBase:
sgpp::datadriven::AlgorithmAdaBoostIdentity

Public Member Functions

 AlgorithmAdaBoostBase (base::Grid &SparseGrid, size_t gridType, base::level_t gridLevel, base::DataMatrix &trainData, base::DataVector &trainDataClass, size_t NUM, double lambda, size_t IMAX, double eps, size_t IMAX_final, double eps_final, double firstLabel, double secondLabel, double threshold, double maxLambda, double minLambda, size_t searchNum, bool refine, size_t refineMode, size_t refineNum, size_t numberOfAda, double percentOfAda, size_t mode)
 Std-Constructor. More...
 
void classif (base::DataMatrix &testData, base::DataVector &algorithmClassTrain, base::DataVector &algorithmClassTest, base::DataMatrix &algorithmValueTrain, base::DataMatrix &algorithmValueTest)
 Performs a classify for the testing dataset according to the baselearners get from the algorithm. More...
 
void doAdaBoostR2 (base::DataMatrix &weights, base::DataMatrix &testData, base::DataMatrix &algorithmValueTrain, base::DataMatrix &algorithmValueTest, std::string lossFucType)
 Performs the Adaboost.R2(a regression algorithm) More...
 
void doAdaBoostRT (base::DataMatrix &weights, base::DataMatrix &testData, base::DataMatrix &algorithmValueTrain, base::DataMatrix &algorithmValueTest, double Tvalue, std::string powerType)
 Performs the Adaboost.RT(a regression algorithm) More...
 
void doDiscreteAdaBoost (base::DataVector &hypoWeight, base::DataVector &weightError, base::DataMatrix &weights, base::DataMatrix &decision, base::DataMatrix &testData, base::DataMatrix &algorithmValueTrain, base::DataMatrix &algorithmValueTest)
 Performs the Discrete Adaboost. More...
 
void doRealAdaBoost (base::DataMatrix &weights, base::DataMatrix &testData, base::DataMatrix &algorithmValueTrain, base::DataMatrix &algorithmValueTest)
 Performs the Real Adaboost. More...
 
void doRefinement (base::DataVector &alpha_ada, base::DataVector &weight_ada, size_t curBaseLearner)
 Performs refinement of grid to get an adaptive grid. More...
 
void eval (base::DataMatrix &testData, base::DataMatrix &algorithmValueTrain, base::DataMatrix &algorithmValueTest)
 Performs a real value calculate for the testing dataset. More...
 
void getAccuracy (base::DataMatrix &testData, base::DataVector &testDataClass, double *accuracy_train, double *accuracy_test)
 Performs an accuracy evaluation for the testing dataset. More...
 
void getAccuracyBL (base::DataMatrix &testData, base::DataVector &testDataClass, base::DataMatrix &algorithmValueTrain, base::DataMatrix &algorithmValueTest, double *accuracy_train, double *accuracy_test, size_t yourBaseLearner)
 Performs an accuracy evaluation for the testing dataset with a specified number of base learner. More...
 
size_t getActualBL ()
 Get the actual base learners after doing adaboosting. More...
 
size_t getMaxGridPoint (size_t baseLearner)
 Get the max GridPoint ever used in adaboosting. More...
 
size_t getMeanGridPoint (size_t baseLearner)
 Get the mean GridPoint ever used in adaboosting. More...
 
void getROC (base::DataMatrix &validationData, base::DataVector &validationDataClass, double *acc, double *sensitivity, double *specificity, double *precision, double *recall, double *fOneScore)
 Performs an evaluation to get ROC related parameter. More...
 
size_t getSumGridPoint (size_t baseLearner)
 Get the sum GridPoint ever used in adaboosting. More...
 
double hValue (double realValue)
 Performs a hypothesis classifier. More...
 
virtual ~AlgorithmAdaBoostBase ()
 Std-Deconstructor. More...
 

Protected Member Functions

virtual void alphaSolver (double &lambda, base::DataVector &weight, base::DataVector &alpha, bool final)=0
 Performs a solver to get alpha. More...
 

Protected Attributes

size_t actualBaseLearners
 Actual base learners number for Adaboosting. More...
 
size_t boostMode
 Set the boost mode (1: Discrete Adaboost, 2: Real Adaboost) More...
 
base::DataVectorclasses
 Pointer to the class(real value in regression) of the data vector. More...
 
base::DataMatrixdata
 Pointer to the data matrix. More...
 
size_t dim
 DataSet Dimension. More...
 
double epsilon
 Parameter for CG solver(during the refinement) More...
 
double epsilon_final
 Parameter for CG solver(for the last refinement) More...
 
base::Gridgrid
 the grid More...
 
size_t gridPoint
 Number of grid points. More...
 
size_t imax
 Parameter for CG solver(during the refinement) More...
 
size_t imax_final
 Parameter for CG solver(for the last refinement) More...
 
double labelOne
 One label of the DataSet. More...
 
double labelTwo
 Another label of the DataSet. More...
 
double lamb
 the lambda, the regularisation parameter More...
 
double lambLogMax
 Log of the Max lambda in searching for optimal lambda. More...
 
size_t lambSteps
 Number of iteration in searching for optimal lambda. More...
 
double lambStepsize
 Interval size with logrange used in searching optimal lambda. More...
 
base::level_t level
 grid level More...
 
base::DataVectormaxGridPoint
 Number of the maximum grid points used in the algorithm. More...
 
size_t numBaseLearners
 Number of base learner for Adaboosting. More...
 
size_t numData
 the size of the grid More...
 
size_t numOfAda
 Number of Grid points to refine. More...
 
double perOfAda
 Percentage of Grid points to refine(between 0 and 1) More...
 
bool refinement
 Judgement of grid refine. More...
 
size_t refineMode
 Select the refine mode(1:use grid number, 2: use grid number percentage) More...
 
size_t refineTimes
 Number of refinement with a certain percentage of Grid points. More...
 
base::DataVectorsumGridPoint
 Number of the sum grid points used in the algorithm. More...
 
double threshold
 Threshold to predict class. More...
 
size_t type
 type of grid possible value are 1, 2 or 3(1 = Linear Grid, 2 = LinearL0Boundary Grid, 3 = ModLinear Grid); More...
 

Constructor & Destructor Documentation

◆ AlgorithmAdaBoostBase()

sgpp::datadriven::AlgorithmAdaBoostBase::AlgorithmAdaBoostBase ( base::Grid SparseGrid,
size_t  gridType,
base::level_t  gridLevel,
base::DataMatrix trainData,
base::DataVector trainDataClass,
size_t  NUM,
double  lambda,
size_t  IMAX,
double  eps,
size_t  IMAX_final,
double  eps_final,
double  firstLabel,
double  secondLabel,
double  threshold,
double  maxLambda,
double  minLambda,
size_t  searchNum,
bool  refine,
size_t  refineMode,
size_t  refineNum,
size_t  numberOfAda,
double  percentOfAda,
size_t  mode 
)

Std-Constructor.

Parameters
SparseGridreference to the sparse grid
gridTypereference to the of grid type(1 = Linear Grid, 2 = LinearL0Boundary Grid, 3 = ModLinear Grid)
gridLevelreference to the level of grid
trainDatareference to the training dataset
trainDataClassreference to the class(real value in regression) of training dataset
NUMthe number of baselearner for Adaboosting
lambdathe regularisation parameter
IMAXthe parameter for ConjugateGradients
epsthe parameter for ConjugateGradients
IMAX_finalthe parameter for ConjugateGradients used for last refinement step
eps_finalthe parameter for ConjugateGradients used for last refinement step
firstLabelone label from training dataset
secondLabelanother label from training dataset
thresholdthe parameter for predicting a class
maxLambdathe max lambda used in searching optimal lambda
minLambdathe min lambda used in searching optimal lambda
searchNumthe searching times used in searching for optimal lambda
refinethe judgement of refine
refineModeSelect the refine mode
refineNumthe Number of refinement with a certain percentage of Grid points
numberOfAdathe number of Grid points to refine
percentOfAdathe percentage of Grid points to refine
modethe adaboost type to choose

References actualBaseLearners, boostMode, classes, data, dim, friedman::eps, epsilon, epsilon_final, sgpp::base::HashGridStorage::getDimension(), sgpp::base::DataMatrix::getNrows(), sgpp::base::HashGridStorage::getSize(), sgpp::base::Grid::getStorage(), grid, gridPoint, python.utils.sg_projections::gridStorage, imax, imax_final, labelOne, labelTwo, lamb, lambda, lambLogMax, lambSteps, lambStepsize, level, maxGridPoint, numBaseLearners, numData, numOfAda, perOfAda, python.uq.operations.discretizeProduct::refine(), refinement, refineMode, refineTimes, sumGridPoint, threshold, and type.

◆ ~AlgorithmAdaBoostBase()

sgpp::datadriven::AlgorithmAdaBoostBase::~AlgorithmAdaBoostBase ( )
virtual

Std-Deconstructor.

Member Function Documentation

◆ alphaSolver()

virtual void sgpp::datadriven::AlgorithmAdaBoostBase::alphaSolver ( double &  lambda,
base::DataVector weight,
base::DataVector alpha,
bool  final 
)
protectedpure virtual

Performs a solver to get alpha.

Parameters
lambdathe regularisation parameter
weightthe weights of examples
alphaoutput the coefficients of the sparse grid's basis functions
finaljudgement the final step of this base learner

Implemented in sgpp::datadriven::AlgorithmAdaBoostIdentity.

Referenced by doAdaBoostR2(), doAdaBoostRT(), doDiscreteAdaBoost(), doRealAdaBoost(), and doRefinement().

◆ classif()

void sgpp::datadriven::AlgorithmAdaBoostBase::classif ( base::DataMatrix testData,
base::DataVector algorithmClassTrain,
base::DataVector algorithmClassTest,
base::DataMatrix algorithmValueTrain,
base::DataMatrix algorithmValueTest 
)

Performs a classify for the testing dataset according to the baselearners get from the algorithm.

Parameters
testDatareference to the testing dataset
algorithmClassTrainreference to the class of training dataset got from the algorithm
algorithmClassTestreference to the class of testing dataset got from the algorithm
algorithmValueTrainthe matrix reference to the real value of training dataset got from the algorithm with diff base learners
algorithmValueTestthe matrix reference to the real value of testing dataset got from the algorithm with diff base learners

References eval(), sgpp::base::DataMatrix::get(), sgpp::base::DataMatrix::getNrows(), hValue(), python.statsfileInfo::i, numData, and sgpp::base::DataVector::set().

Referenced by getAccuracy(), and getROC().

◆ doAdaBoostR2()

void sgpp::datadriven::AlgorithmAdaBoostBase::doAdaBoostR2 ( base::DataMatrix weights,
base::DataMatrix testData,
base::DataMatrix algorithmValueTrain,
base::DataMatrix algorithmValueTest,
std::string  lossFucType 
)

Performs the Adaboost.R2(a regression algorithm)

Parameters
weightsthe matrix to store weights of every training date for every weak learner
testDatareference to the testing dataset
algorithmValueTrainthe matrix reference to the real value of training dataset got from the algorithm with diff base learners
algorithmValueTestthe matrix reference to the real value of testing dataset got from the algorithm with diff base learners
lossFucTypethe loss function type(linear, square or exponential)

References sgpp::base::DataVector::abs(), actualBaseLearners, alphaSolver(), sgpp::base::DataVector::axpy(), classes, sgpp::base::DataVector::componentwise_mult(), sgpp::base::DataVector::copyFrom(), python.leja::count, sgpp::base::Grid::createLinearBoundaryGrid(), sgpp::base::Grid::createLinearGrid(), sgpp::base::Grid::createModLinearGrid(), sgpp::op_factory::createOperationEval(), data, dim, doRefinement(), sgpp::base::DataVector::dotProduct(), sgpp::base::DataVector::get(), sgpp::base::DataMatrix::getColumn(), sgpp::base::Grid::getGenerator(), sgpp::base::DataMatrix::getNrows(), sgpp::base::DataMatrix::getRow(), sgpp::base::DataVector::getSize(), grid, gridPoint, python.statsfileInfo::i, lamb, level, sgpp::base::DataVector::max(), maxGridPoint, sgpp::base::DataVector::mult(), numBaseLearners, numData, sgpp::combigrid::pow(), refinement, sgpp::base::GridGenerator::regular(), sgpp::base::DataVector::resizeZero(), sgpp::base::DataVector::set(), sgpp::base::DataVector::setAll(), sgpp::base::DataMatrix::setColumn(), sgpp::base::DataVector::sqr(), sgpp::base::DataVector::sum(), sumGridPoint, type, and python.leja::weight.

◆ doAdaBoostRT()

void sgpp::datadriven::AlgorithmAdaBoostBase::doAdaBoostRT ( base::DataMatrix weights,
base::DataMatrix testData,
base::DataMatrix algorithmValueTrain,
base::DataMatrix algorithmValueTest,
double  Tvalue,
std::string  powerType 
)

Performs the Adaboost.RT(a regression algorithm)

Parameters
weightsthe matrix to store weights of every training date for every weak learner
testDatareference to the testing dataset
algorithmValueTrainthe matrix reference to the real value of training dataset got from the algorithm with diff base learners
algorithmValueTestthe matrix reference to the real value of testing dataset got from the algorithm with diff base learners
Tvaluethe threshold to demarcate the prediction correctness(only from 0 to 1)
powerTypethe error rate power coefficient(linear, square or cubic)

References actualBaseLearners, alphaSolver(), sgpp::base::DataVector::axpy(), classes, sgpp::base::DataVector::copyFrom(), python.leja::count, sgpp::base::Grid::createLinearBoundaryGrid(), sgpp::base::Grid::createLinearGrid(), sgpp::base::Grid::createModLinearGrid(), sgpp::op_factory::createOperationEval(), data, dim, doRefinement(), sgpp::base::DataVector::get(), sgpp::base::DataMatrix::getColumn(), sgpp::base::Grid::getGenerator(), sgpp::base::DataMatrix::getNrows(), sgpp::base::DataMatrix::getRow(), sgpp::base::DataVector::getSize(), grid, gridPoint, python.statsfileInfo::i, lamb, level, maxGridPoint, sgpp::base::DataVector::mult(), numBaseLearners, numData, refinement, sgpp::base::GridGenerator::regular(), sgpp::base::DataVector::resizeZero(), sgpp::base::DataVector::set(), sgpp::base::DataVector::setAll(), sgpp::base::DataMatrix::setColumn(), sgpp::base::DataVector::sum(), sumGridPoint, type, and python.leja::weight.

◆ doDiscreteAdaBoost()

void sgpp::datadriven::AlgorithmAdaBoostBase::doDiscreteAdaBoost ( base::DataVector hypoWeight,
base::DataVector weightError,
base::DataMatrix weights,
base::DataMatrix decision,
base::DataMatrix testData,
base::DataMatrix algorithmValueTrain,
base::DataMatrix algorithmValueTest 
)

Performs the Discrete Adaboost.

Parameters
hypoWeightthe vector to store hypothesis weights(Alpha-t)
weightErrorthe vector to store the weight error of each iteration
weightsthe matrix to store weights of every training date for every weak learner
decisionthe matrix to store the decision right or not according to the true class
testDatareference to the testing dataset
algorithmValueTrainthe matrix reference to the real value of training dataset got from the algorithm with diff base learners
algorithmValueTestthe matrix reference to the real value of testing dataset got from the algorithm with diff base learners

References actualBaseLearners, alphaSolver(), classes, sgpp::base::DataVector::copyFrom(), python.leja::count, sgpp::base::Grid::createLinearBoundaryGrid(), sgpp::base::Grid::createLinearGrid(), sgpp::base::Grid::createModLinearGrid(), sgpp::op_factory::createOperationEval(), data, dim, doRefinement(), sgpp::base::DataVector::dotProduct(), sgpp::base::DataVector::get(), sgpp::base::DataMatrix::get(), sgpp::base::Grid::getGenerator(), sgpp::base::DataMatrix::getNrows(), sgpp::base::DataMatrix::getRow(), sgpp::base::DataVector::getSize(), grid, gridPoint, hValue(), python.statsfileInfo::i, lamb, lambLogMax, lambSteps, lambStepsize, level, maxGridPoint, sgpp::base::DataVector::mult(), numBaseLearners, numData, refinement, sgpp::base::GridGenerator::regular(), sgpp::base::DataVector::resizeZero(), sgpp::base::DataVector::set(), sgpp::base::DataMatrix::set(), sgpp::base::DataVector::setAll(), sgpp::base::DataMatrix::setColumn(), sgpp::base::DataVector::sum(), sumGridPoint, type, and python.leja::weight.

Referenced by eval().

◆ doRealAdaBoost()

void sgpp::datadriven::AlgorithmAdaBoostBase::doRealAdaBoost ( base::DataMatrix weights,
base::DataMatrix testData,
base::DataMatrix algorithmValueTrain,
base::DataMatrix algorithmValueTest 
)

◆ doRefinement()

void sgpp::datadriven::AlgorithmAdaBoostBase::doRefinement ( base::DataVector alpha_ada,
base::DataVector weight_ada,
size_t  curBaseLearner 
)

Performs refinement of grid to get an adaptive grid.

Parameters
alpha_adathe coefficients of the sparse grid's basis functions and to be refined
weight_adathe weights of examples
curBaseLearnerthe current base learner

References alphaSolver(), sgpp::base::DataVector::get(), sgpp::base::Grid::getGenerator(), sgpp::base::HashGridStorage::getSize(), sgpp::base::Grid::getSize(), sgpp::base::Grid::getStorage(), grid, lamb, maxGridPoint, numOfAda, perOfAda, sgpp::base::GridGenerator::refine(), refineMode, refineTimes, sgpp::base::DataVector::resizeZero(), sgpp::base::DataVector::set(), and sumGridPoint.

Referenced by doAdaBoostR2(), doAdaBoostRT(), doDiscreteAdaBoost(), and doRealAdaBoost().

◆ eval()

void sgpp::datadriven::AlgorithmAdaBoostBase::eval ( base::DataMatrix testData,
base::DataMatrix algorithmValueTrain,
base::DataMatrix algorithmValueTest 
)

Performs a real value calculate for the testing dataset.

Parameters
testDatareference to the testing dataset
algorithmValueTrainthe matrix reference to the real value of training dataset got from the algorithm with diff base learners
algorithmValueTestthe matrix reference to the real value of testing dataset got from the algorithm with diff base learners

References boostMode, doDiscreteAdaBoost(), doRealAdaBoost(), numBaseLearners, numData, sgpp::base::DataVector::setAll(), and sgpp::base::DataMatrix::setAll().

Referenced by classif().

◆ getAccuracy()

void sgpp::datadriven::AlgorithmAdaBoostBase::getAccuracy ( base::DataMatrix testData,
base::DataVector testDataClass,
double *  accuracy_train,
double *  accuracy_test 
)

Performs an accuracy evaluation for the testing dataset.

Parameters
testDatareference to the testing dataset
testDataClassreference to the class of testing dataset
accuracy_trainreference to the accuracy for the training dataset
accuracy_testreference to the accuracy for the testing dataset

References classes, classif(), sgpp::base::DataVector::get(), sgpp::base::DataMatrix::getNrows(), sgpp::base::DataVector::getSize(), python.statsfileInfo::i, numBaseLearners, and numData.

◆ getAccuracyBL()

void sgpp::datadriven::AlgorithmAdaBoostBase::getAccuracyBL ( base::DataMatrix testData,
base::DataVector testDataClass,
base::DataMatrix algorithmValueTrain,
base::DataMatrix algorithmValueTest,
double *  accuracy_train,
double *  accuracy_test,
size_t  yourBaseLearner 
)

Performs an accuracy evaluation for the testing dataset with a specified number of base learner.

Parameters
testDatareference to the testing dataset
testDataClassreference to the class of testing dataset
algorithmValueTrainthe matrix reference to the real value of training dataset got from the algorithm with diff base learners
algorithmValueTestthe matrix reference to the real value of testing dataset got from the algorithm with diff base learners
accuracy_trainreference to the accuracy for the training dataset
accuracy_testreference to the accuracy for the testing dataset
yourBaseLearnerthe number of base learner specified

References classes, sgpp::base::DataVector::get(), sgpp::base::DataMatrix::get(), sgpp::base::DataMatrix::getNrows(), sgpp::base::DataVector::getSize(), hValue(), python.statsfileInfo::i, and numData.

◆ getActualBL()

size_t sgpp::datadriven::AlgorithmAdaBoostBase::getActualBL ( )

Get the actual base learners after doing adaboosting.

References actualBaseLearners.

◆ getMaxGridPoint()

size_t sgpp::datadriven::AlgorithmAdaBoostBase::getMaxGridPoint ( size_t  baseLearner)

Get the max GridPoint ever used in adaboosting.

Parameters
baseLearnernumber of baselearner

References sgpp::base::DataVector::get(), and maxGridPoint.

◆ getMeanGridPoint()

size_t sgpp::datadriven::AlgorithmAdaBoostBase::getMeanGridPoint ( size_t  baseLearner)

Get the mean GridPoint ever used in adaboosting.

Parameters
baseLearnernumber of baselearner

References sgpp::base::DataVector::get(), python.datasetAnalysis::mean, and sumGridPoint.

◆ getROC()

void sgpp::datadriven::AlgorithmAdaBoostBase::getROC ( base::DataMatrix validationData,
base::DataVector validationDataClass,
double *  acc,
double *  sensitivity,
double *  specificity,
double *  precision,
double *  recall,
double *  fOneScore 
)

Performs an evaluation to get ROC related parameter.

Parameters
validationDatareference to the validation dataset
validationDataClassreference to the class of validation dataset
accreference to the accuracy for the validation dataset
sensitivityreference to the sensitivity for the validation dataset
specificityreference to the specificity for the validation dataset
precisionreference to the precision for the validation dataset
recallreference to the recall for the validation dataset
fOneScorereference to the specificity for the validation dataset

References classif(), sgpp::base::DataVector::get(), sgpp::base::DataMatrix::getNrows(), sgpp::base::DataVector::getSize(), python.statsfileInfo::i, labelOne, labelTwo, numBaseLearners, and numData.

◆ getSumGridPoint()

size_t sgpp::datadriven::AlgorithmAdaBoostBase::getSumGridPoint ( size_t  baseLearner)

Get the sum GridPoint ever used in adaboosting.

Parameters
baseLearnernumber of baselearner

References sgpp::base::DataVector::get(), and sumGridPoint.

◆ hValue()

double sgpp::datadriven::AlgorithmAdaBoostBase::hValue ( double  realValue)

Performs a hypothesis classifier.

Parameters
realValuereal value of function

References labelOne, labelTwo, and threshold.

Referenced by classif(), doDiscreteAdaBoost(), and getAccuracyBL().

Member Data Documentation

◆ actualBaseLearners

size_t sgpp::datadriven::AlgorithmAdaBoostBase::actualBaseLearners
protected

Actual base learners number for Adaboosting.

Referenced by AlgorithmAdaBoostBase(), doAdaBoostR2(), doAdaBoostRT(), doDiscreteAdaBoost(), doRealAdaBoost(), and getActualBL().

◆ boostMode

size_t sgpp::datadriven::AlgorithmAdaBoostBase::boostMode
protected

Set the boost mode (1: Discrete Adaboost, 2: Real Adaboost)

Referenced by AlgorithmAdaBoostBase(), and eval().

◆ classes

base::DataVector* sgpp::datadriven::AlgorithmAdaBoostBase::classes
protected

◆ data

base::DataMatrix* sgpp::datadriven::AlgorithmAdaBoostBase::data
protected

◆ dim

◆ epsilon

double sgpp::datadriven::AlgorithmAdaBoostBase::epsilon
protected

Parameter for CG solver(during the refinement)

Referenced by AlgorithmAdaBoostBase(), and sgpp::datadriven::AlgorithmAdaBoostIdentity::alphaSolver().

◆ epsilon_final

double sgpp::datadriven::AlgorithmAdaBoostBase::epsilon_final
protected

Parameter for CG solver(for the last refinement)

Referenced by AlgorithmAdaBoostBase(), and sgpp::datadriven::AlgorithmAdaBoostIdentity::alphaSolver().

◆ grid

◆ gridPoint

size_t sgpp::datadriven::AlgorithmAdaBoostBase::gridPoint
protected

◆ imax

size_t sgpp::datadriven::AlgorithmAdaBoostBase::imax
protected

Parameter for CG solver(during the refinement)

Referenced by AlgorithmAdaBoostBase(), and sgpp::datadriven::AlgorithmAdaBoostIdentity::alphaSolver().

◆ imax_final

size_t sgpp::datadriven::AlgorithmAdaBoostBase::imax_final
protected

Parameter for CG solver(for the last refinement)

Referenced by AlgorithmAdaBoostBase(), and sgpp::datadriven::AlgorithmAdaBoostIdentity::alphaSolver().

◆ labelOne

double sgpp::datadriven::AlgorithmAdaBoostBase::labelOne
protected

One label of the DataSet.

Referenced by AlgorithmAdaBoostBase(), getROC(), and hValue().

◆ labelTwo

double sgpp::datadriven::AlgorithmAdaBoostBase::labelTwo
protected

Another label of the DataSet.

Referenced by AlgorithmAdaBoostBase(), getROC(), and hValue().

◆ lamb

double sgpp::datadriven::AlgorithmAdaBoostBase::lamb
protected

the lambda, the regularisation parameter

Referenced by AlgorithmAdaBoostBase(), doAdaBoostR2(), doAdaBoostRT(), doDiscreteAdaBoost(), doRealAdaBoost(), and doRefinement().

◆ lambLogMax

double sgpp::datadriven::AlgorithmAdaBoostBase::lambLogMax
protected

Log of the Max lambda in searching for optimal lambda.

Referenced by AlgorithmAdaBoostBase(), and doDiscreteAdaBoost().

◆ lambSteps

size_t sgpp::datadriven::AlgorithmAdaBoostBase::lambSteps
protected

Number of iteration in searching for optimal lambda.

Referenced by AlgorithmAdaBoostBase(), and doDiscreteAdaBoost().

◆ lambStepsize

double sgpp::datadriven::AlgorithmAdaBoostBase::lambStepsize
protected

Interval size with logrange used in searching optimal lambda.

Referenced by AlgorithmAdaBoostBase(), and doDiscreteAdaBoost().

◆ level

base::level_t sgpp::datadriven::AlgorithmAdaBoostBase::level
protected

◆ maxGridPoint

base::DataVector* sgpp::datadriven::AlgorithmAdaBoostBase::maxGridPoint
protected

Number of the maximum grid points used in the algorithm.

Referenced by AlgorithmAdaBoostBase(), doAdaBoostR2(), doAdaBoostRT(), doDiscreteAdaBoost(), doRealAdaBoost(), doRefinement(), and getMaxGridPoint().

◆ numBaseLearners

size_t sgpp::datadriven::AlgorithmAdaBoostBase::numBaseLearners
protected

◆ numData

size_t sgpp::datadriven::AlgorithmAdaBoostBase::numData
protected

◆ numOfAda

size_t sgpp::datadriven::AlgorithmAdaBoostBase::numOfAda
protected

Number of Grid points to refine.

Referenced by AlgorithmAdaBoostBase(), and doRefinement().

◆ perOfAda

double sgpp::datadriven::AlgorithmAdaBoostBase::perOfAda
protected

Percentage of Grid points to refine(between 0 and 1)

Referenced by AlgorithmAdaBoostBase(), and doRefinement().

◆ refinement

bool sgpp::datadriven::AlgorithmAdaBoostBase::refinement
protected

◆ refineMode

size_t sgpp::datadriven::AlgorithmAdaBoostBase::refineMode
protected

Select the refine mode(1:use grid number, 2: use grid number percentage)

Referenced by AlgorithmAdaBoostBase(), and doRefinement().

◆ refineTimes

size_t sgpp::datadriven::AlgorithmAdaBoostBase::refineTimes
protected

Number of refinement with a certain percentage of Grid points.

Referenced by AlgorithmAdaBoostBase(), and doRefinement().

◆ sumGridPoint

base::DataVector* sgpp::datadriven::AlgorithmAdaBoostBase::sumGridPoint
protected

◆ threshold

double sgpp::datadriven::AlgorithmAdaBoostBase::threshold
protected

Threshold to predict class.

Referenced by AlgorithmAdaBoostBase(), and hValue().

◆ type

size_t sgpp::datadriven::AlgorithmAdaBoostBase::type
protected

type of grid possible value are 1, 2 or 3(1 = Linear Grid, 2 = LinearL0Boundary Grid, 3 = ModLinear Grid);

Referenced by AlgorithmAdaBoostBase(), doAdaBoostR2(), doAdaBoostRT(), doDiscreteAdaBoost(), and doRealAdaBoost().


The documentation for this class was generated from the following files: