0-1-loss for classification. More...

#include <shark/ObjectiveFunctions/Loss/ZeroOneLoss.h>

Inheritance diagram for shark::ZeroOneLoss< unsigned int, RealVector >:

Public Types
typedef base_type::BatchLabelType	BatchLabelType

typedef base_type::BatchOutputType	BatchOutputType

Public Types inherited from shark::AbstractLoss< unsigned int, RealVector >
typedef RealVector	OutputType

typedef unsigned int	LabelType

typedef RealMatrix	MatrixType

typedef Batch< OutputType >::type	BatchOutputType

typedef Batch< LabelType >::type	BatchLabelType

typedef ConstProxyReference< LabelType const >::type	ConstLabelReference
	Const references to LabelType. More...

typedef ConstProxyReference< OutputType const >::type	ConstOutputReference
	Const references to OutputType. More...

Public Types inherited from shark::AbstractCost< unsigned int, RealVector >
enum	Feature
	list of features a cost function can have More...

typedef RealVector	OutputType

typedef unsigned int	LabelType

typedef Batch< OutputType >::type	BatchOutputType

typedef Batch< LabelType >::type	BatchLabelType

typedef TypedFlags< Feature >	Features

typedef TypedFeatureNotAvailableException< Feature >	FeatureNotAvailableException

Public Member Functions
	ZeroOneLoss (double threshold=0.0)

std::string	name () const
	From INameable: return the class name. More...

double	eval (BatchLabelType const &labels, BatchOutputType const &predictions) const

double	eval (Data< LabelType > const &targets, Data< OutputType > const &predictions, RealVector const &weights) const

Public Member Functions inherited from shark::AbstractLoss< unsigned int, RealVector >
	AbstractLoss ()

virtual double	eval (ConstLabelReference target, ConstOutputReference prediction) const
	evaluate the loss for a target and a prediction More...

double	eval (Data< LabelType > const &targets, Data< OutputType > const &predictions) const

virtual double	evalDerivative (ConstLabelReference target, ConstOutputReference prediction, OutputType &gradient) const
	evaluate the loss and its derivative for a target and a prediction More...

virtual double	evalDerivative (ConstLabelReference target, ConstOutputReference prediction, OutputType &gradient, MatrixType &hessian) const
	evaluate the loss and its first and second derivative for a target and a prediction More...

virtual double	evalDerivative (BatchLabelType const &target, BatchOutputType const &prediction, BatchOutputType &gradient) const
	evaluate the loss and the derivative w.r.t. the prediction More...

double	operator() (LabelType const &target, OutputType const &prediction) const
	evaluate the loss for a target and a prediction More...

double	operator() (BatchLabelType const &target, BatchOutputType const &prediction) const

Public Member Functions inherited from shark::AbstractCost< unsigned int, RealVector >
virtual	~AbstractCost ()

const Features &	features () const

virtual void	updateFeatures ()

bool	hasFirstDerivative () const
	returns true when the first parameter derivative is implemented More...

bool	isLossFunction () const
	returns true when the cost function is in fact a loss function More...

double	operator() (Data< LabelType > const &targets, Data< OutputType > const &predictions) const

Public Member Functions inherited from shark::INameable
virtual	~INameable ()

Additional Inherited Members
Protected Attributes inherited from shark::AbstractCost< unsigned int, RealVector >
Features	m_features

Detailed Description

template<>
class shark::ZeroOneLoss< unsigned int, RealVector >

0-1-loss for classification.

Definition at line 85 of file ZeroOneLoss.h.

Member Typedef Documentation

◆ BatchLabelType

typedef base_type::BatchLabelType shark::ZeroOneLoss< unsigned int, RealVector >::BatchLabelType

Definition at line 89 of file ZeroOneLoss.h.

◆ BatchOutputType

typedef base_type::BatchOutputType shark::ZeroOneLoss< unsigned int, RealVector >::BatchOutputType

Definition at line 90 of file ZeroOneLoss.h.

Constructor & Destructor Documentation

◆ ZeroOneLoss()

shark::ZeroOneLoss< unsigned int, RealVector >::ZeroOneLoss ( double threshold = 0.0 )

inline

constructor

Parameters

threshold in the case dim(predictions) == 1, predictions strictly larger than this parameter are regarded as belonging to the positive class

Definition at line 95 of file ZeroOneLoss.h.

Member Function Documentation

◆ eval() [1/2]

double shark::ZeroOneLoss< unsigned int, RealVector >::eval	(	BatchLabelType const &	labels,
		BatchOutputType const &	predictions
	)		const

inlinevirtual

Return zero if labels == arg max { predictions_i } and one otherwise, where the index i runs over the components of the predictions vector. A special version of dim(predictions) == 1 computes the predicted labels by thresholding at zero. Shark's label convention is used, saying that a positive value encodes class 0, a negative value encodes class 1.

Implements shark::AbstractLoss< unsigned int, RealVector >.

Definition at line 114 of file ZeroOneLoss.h.

References SIZE_CHECK.

Referenced by main().

◆ eval() [2/2]

double shark::ZeroOneLoss< unsigned int, RealVector >::eval	(	Data< LabelType > const &	targets,
		Data< OutputType > const &	predictions,
		RealVector const &	weights
	)		const

inline

Definition at line 125 of file ZeroOneLoss.h.

References shark::Data< Type >::batch(), shark::Data< Type >::numberOfBatches(), shark::Data< Type >::numberOfElements(), and SIZE_CHECK.

◆ name()

std::string shark::ZeroOneLoss< unsigned int, RealVector >::name ( ) const

inlinevirtual

From INameable: return the class name.

Reimplemented from shark::INameable.

Definition at line 101 of file ZeroOneLoss.h.

References shark::AbstractLoss< LabelType, LabelType >::eval().