theplu::yat::classifier Namespace Reference

Classifier related classes. More...

Classes
class	BootstrapSampler
	Class creating trainingset and validationset using bootstrapping. More...
class	ConsensusInputRanker
	Robust algorithm to rank rows in a data matrix versus a target vector. More...
class	CrossValidationSampler
	Class splitting a set into training set and validation set in a crossvalidation manner. More...
class	DataLookup1D
	Class for general vector view. More...
class	DataLookupWeighted1D
	Class for general weighted vector view. More...
class	EnsembleBuilder
	Class for ensembles of supervised classifiers. More...
class	FeatureSelector
	Interface class for FeatureSelection. More...
class	FeatureSelectorIR
	FeatureSelector using an InputRanker. More...
class	FeatureSelectorRandom
	Class for selection features by random. More...
class	GaussianKernelFunction
	Class for Gaussian kernel calculations. More...
class	IGP
	Class for In Group Proportions (IGP) See Kapp and Tibshirani, Biostatistics (2006). More...
class	InputRanker
	Class for ranking rows in a matrix, using a Score and a target vector. More...
class	IRRank
	Functor retrieving minus rank from a InputRanker to build a ConsensusInputRanker. More...
class	IRRetrieve
	Interface class for retrieving information from a InputRanker to build a ConsensusInputRanker. More...
class	Kernel
	Interface Class for Kernels. More...
class	Kernel_MEV
	Memory Efficient Kernel. More...
class	Kernel_SEV
	Speed Efficient Kernel. More...
class	KernelFunction
	Interface class calculating elements in Kernel. More...
class	KernelLookup
	Lookup into Kernel. More...
class	KNN
	Nearest Neighbor Classifier. More...
struct	KNN_ReciprocalDistance
	A model of the concept Neighbor Weighting Method to be used with KNN to weight the votes of the k nearest neighbors of a sample. More...
struct	KNN_ReciprocalRank
	A model of the concept Neighbor Weighting Method to be used with KNN to weight the votes of the k nearest neighbors of a sample. More...
struct	KNN_Uniform
	A model of the concept Neighbor Weighting Method to be used with KNN to weight the votes of the k nearest neighbors of a sample. More...
class	MatrixLookup
	General view into utility::Matrix. More...
class	MatrixLookupWeighted
	General view into utility::MatrixWeighted. More...
class	NBC
	Naive Bayesian Classifier. More...
class	NCC
	Nearest Centroid Classifier. More...
class	PolynomialKernelFunction
	Class for polynomial kernel calculations. More...
class	Sampler
	Interface class for dividing samples into training and validation. More...
class	SubsetGenerator
	Class splitting Data into training and validation set. More...
class	SupervisedClassifier
	Interface class for supervised classifiers that use data in a matrix format. More...
class	SVindex
class	SVM
	Support Vector Machine. More...
class	Target
	Class for containing sample labels. More...
Functions
std::ostream &	operator<< (std::ostream &s, const DataLookup1D &v)
	The output operator for DataLookup1D.
double	sum_weight (const DataLookupWeighted1D &)
std::ostream &	operator<< (std::ostream &s, const MatrixLookup &)
std::ostream &	operator<< (std::ostream &s, const MatrixLookupWeighted &)
std::ostream &	operator<< (std::ostream &, const Target &)
void	convert (const DataLookup1D &, utility::Vector &)
void	convert (const DataLookupWeighted1D &, utility::Vector &value, utility::Vector &weight)

---

Detailed Description

Classifier related classes.

All classes associated with usage of classifiers are defined within this namespace.

---

Function Documentation

void theplu::yat::classifier::convert	(	const DataLookupWeighted1D &	,
		utility::Vector &	value,
		utility::Vector &	weight
	)

Converts a DataLookupWeighted1D to two utility::vector

void theplu::yat::classifier::convert	(	const DataLookup1D &	,
		utility::Vector &
	)

Converts a DataLookup1D to a utility::vector

std::ostream& theplu::yat::classifier::operator<<	(	std::ostream &	,
		const Target &
	)

The output operator for the Target class.

std::ostream& theplu::yat::classifier::operator<<	(	std::ostream &	s,
		const MatrixLookupWeighted &
	)

The output operator MatrixLookupWeighted

For eacd data element data(i,j) is printed except those being associated with a zero weight for which nothing is printed.

std::ostream& theplu::yat::classifier::operator<<	(	std::ostream &	s,
		const MatrixLookup &
	)

The output operator MatrixLookup

std::ostream& theplu::yat::classifier::operator<<	(	std::ostream &	s,
		const DataLookup1D &	v
	)

The output operator for DataLookup1D.

Elements are separated by the character from the omanip fill. The following example will write the elements separated by tab.

	 char prev=s.fill('\t'); 
	 s << v;
	 s.fill(prev); 
	 

double theplu::yat::classifier::sum_weight

(

const DataLookupWeighted1D &

)

Returns:

sum of weights