radius margin quotions for binary SVMs More...

#include <shark/ObjectiveFunctions/RadiusMarginQuotient.h>

Inheritance diagram for shark::RadiusMarginQuotient< InputType, CacheType >:

Classes
struct	Result

Public Types
typedef CacheType	QpFloatType

typedef KernelMatrix< InputType, QpFloatType >	KernelMatrixType

typedef CachedMatrix< KernelMatrixType >	CachedMatrixType

typedef LabeledData< InputType, unsigned int >	DatasetType

typedef AbstractKernelFunction< InputType >	KernelType

Public Types inherited from shark::AbstractObjectiveFunction< RealVector, double >
enum	Feature
	List of features that are supported by an implementation. More...

typedef RealVector	SearchPointType

typedef double	ResultType

typedef boost::mpl::if_< std::is_arithmetic< double >, SearchPointType, RealMatrix >::type	FirstOrderDerivative

typedef TypedFlags< Feature >	Features
	This statement declares the member m_features. See Core/Flags.h for details.

typedef TypedFeatureNotAvailableException< Feature >	FeatureNotAvailableException

Public Member Functions
	RadiusMarginQuotient (DatasetType const &dataset, KernelType *kernel)
	Constructor.

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

std::size_t	numberOfVariables () const
	Accesses the number of variables.

double	eval (SearchPointType const &parameters) const
	Evaluate the radius margin quotient.

double	evalDerivative (SearchPointType const &parameters, FirstOrderDerivative &derivative) const
	Evaluate the radius margin quotient and its first derivative.

Public Member Functions inherited from shark::AbstractObjectiveFunction< RealVector, double >
const Features &	features () const

virtual void	updateFeatures ()

bool	hasValue () const
	returns whether this function can calculate it's function value

bool	hasFirstDerivative () const
	returns whether this function can calculate the first derivative

bool	hasSecondDerivative () const
	returns whether this function can calculate the second derivative

bool	canProposeStartingPoint () const
	returns whether this function can propose a starting point.

bool	isConstrained () const
	returns whether this function can return

bool	hasConstraintHandler () const
	returns whether this function can return

bool	canProvideClosestFeasible () const
	Returns whether this function can calculate thee closest feasible to an infeasible point.

bool	isThreadSafe () const
	Returns true, when the function can be usd in parallel threads.

bool	isNoisy () const
	Returns true, when the function can be usd in parallel threads.

	AbstractObjectiveFunction ()
	Default ctor.

virtual	~AbstractObjectiveFunction ()
	Virtual destructor.

virtual void	init ()

void	setRng (random::rng_type *rng)
	Sets the Rng used by the objective function.

virtual bool	hasScalableDimensionality () const

virtual void	setNumberOfVariables (std::size_t numberOfVariables)
	Adjusts the number of variables if the function is scalable.

virtual std::size_t	numberOfObjectives () const

virtual bool	hasScalableObjectives () const

virtual void	setNumberOfObjectives (std::size_t numberOfObjectives)
	Adjusts the number of objectives if the function is scalable.

std::size_t	evaluationCounter () const
	Accesses the evaluation counter of the function.

AbstractConstraintHandler< SearchPointType > const &	getConstraintHandler () const
	Returns the constraint handler of the function if it has one.

virtual bool	isFeasible (const SearchPointType &input) const
	Tests whether a point in SearchSpace is feasible, e.g., whether the constraints are fulfilled.

virtual void	closestFeasible (SearchPointType &input) const
	If supported, the supplied point is repaired such that it satisfies all of the function's constraints.

virtual SearchPointType	proposeStartingPoint () const
	Proposes a starting point in the feasible search space of the function.

ResultType	operator() (SearchPointType const &input) const
	Evaluates the function. Useful together with STL-Algorithms like std::transform.

virtual ResultType	evalDerivative (SearchPointType const &input, SecondOrderDerivative &derivative) const
	Evaluates the objective function and calculates its gradient.

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

Protected Member Functions
Result	computeRadiusMargin () const

Protected Member Functions inherited from shark::AbstractObjectiveFunction< RealVector, double >
void	announceConstraintHandler (AbstractConstraintHandler< SearchPointType > const *handler)
	helper function which is called to announce the presence of an constraint handler.

Protected Attributes
KernelType *	mep_kernel
	underlying parameterized kernel object

DatasetType	m_dataset
	labeled data for radius and (hard) margin computation

Protected Attributes inherited from shark::AbstractObjectiveFunction< RealVector, double >
Features	m_features

std::size_t	m_evaluationCounter
	Evaluation counter, default value: 0.

AbstractConstraintHandler< SearchPointType > const *	m_constraintHandler

random::rng_type *	mep_rng

Detailed Description

template<class InputType, class CacheType = float>
class shark::RadiusMarginQuotient< InputType, CacheType >

radius margin quotions for binary SVMs

: The RadiusMarginQuotient is the quotient \( R^2 / \rho^2 \) of the radius R of the smallest sphere containing the training data and the margin \(\rho\) of a binary hard margin support vector machine. Both distances depend on the kernel function, and thus on its parameters. The radius margin quotient is a common objective function for the adaptation of the kernel parameters of a binary hard-margin SVM.

Definition at line 59 of file RadiusMarginQuotient.h.

Member Typedef Documentation

◆ CachedMatrixType

template<class InputType , class CacheType = float>

typedef CachedMatrix< KernelMatrixType > shark::RadiusMarginQuotient< InputType, CacheType >::CachedMatrixType

Definition at line 65 of file RadiusMarginQuotient.h.

◆ DatasetType

template<class InputType , class CacheType = float>

typedef LabeledData<InputType, unsigned int> shark::RadiusMarginQuotient< InputType, CacheType >::DatasetType

Definition at line 67 of file RadiusMarginQuotient.h.

◆ KernelMatrixType

template<class InputType , class CacheType = float>

typedef KernelMatrix<InputType, QpFloatType> shark::RadiusMarginQuotient< InputType, CacheType >::KernelMatrixType

Definition at line 64 of file RadiusMarginQuotient.h.

◆ KernelType

template<class InputType , class CacheType = float>

typedef AbstractKernelFunction<InputType> shark::RadiusMarginQuotient< InputType, CacheType >::KernelType

Definition at line 68 of file RadiusMarginQuotient.h.

◆ QpFloatType

template<class InputType , class CacheType = float>

typedef CacheType shark::RadiusMarginQuotient< InputType, CacheType >::QpFloatType

Definition at line 62 of file RadiusMarginQuotient.h.

Constructor & Destructor Documentation

◆ RadiusMarginQuotient()

template<class InputType , class CacheType = float>

shark::RadiusMarginQuotient< InputType, CacheType >::RadiusMarginQuotient	(	DatasetType const &	dataset,
		KernelType *	kernel
	)

inline

Constructor.

Definition at line 71 of file RadiusMarginQuotient.h.

References shark::AbstractObjectiveFunction< RealVector, double >::HAS_FIRST_DERIVATIVE, shark::AbstractObjectiveFunction< RealVector, double >::HAS_VALUE, shark::AbstractKernelFunction< InputTypeT >::hasFirstParameterDerivative(), shark::AbstractObjectiveFunction< RealVector, double >::m_features, and shark::RadiusMarginQuotient< InputType, CacheType >::mep_kernel.

Member Function Documentation

◆ computeRadiusMargin()

template<class InputType , class CacheType = float>

Result shark::RadiusMarginQuotient< InputType, CacheType >::computeRadiusMargin ( ) const

inlineprotected

Definition at line 140 of file RadiusMarginQuotient.h.

References shark::RadiusMarginQuotient< InputType, CacheType >::Result::alpha, shark::RadiusMarginQuotient< InputType, CacheType >::Result::beta, shark::LabeledData< InputT, LabelT >::inputs(), shark::LabeledData< InputT, LabelT >::labels(), shark::RadiusMarginQuotient< InputType, CacheType >::m_dataset, shark::RadiusMarginQuotient< InputType, CacheType >::mep_kernel, shark::LabeledData< InputT, LabelT >::numberOfElements(), shark::RadiusMarginQuotient< InputType, CacheType >::Result::R2, shark::QpSolver< Problem, SelectionStrategy >::solve(), shark::QpSolutionProperties::value, and shark::RadiusMarginQuotient< InputType, CacheType >::Result::w2.

Referenced by shark::RadiusMarginQuotient< InputType, CacheType >::eval(), and shark::RadiusMarginQuotient< InputType, CacheType >::evalDerivative().

◆ eval()

template<class InputType , class CacheType = float>

double shark::RadiusMarginQuotient< InputType, CacheType >::eval ( SearchPointType const & parameters ) const

inlinevirtual

Evaluate the radius margin quotient.

: The parameters are passed into the kernel, and the radius-margin quotient is computed w.r.t. the kernel-induced metric.

Reimplemented from shark::AbstractObjectiveFunction< RealVector, double >.

Definition at line 94 of file RadiusMarginQuotient.h.

References shark::RadiusMarginQuotient< InputType, CacheType >::computeRadiusMargin(), shark::LabeledData< InputT, LabelT >::empty(), shark::RadiusMarginQuotient< InputType, CacheType >::m_dataset, shark::AbstractObjectiveFunction< RealVector, double >::m_evaluationCounter, shark::RadiusMarginQuotient< InputType, CacheType >::mep_kernel, shark::IParameterizable< VectorType >::numberOfParameters(), shark::RadiusMarginQuotient< InputType, CacheType >::Result::R2, shark::IParameterizable< VectorType >::setParameterVector(), SHARK_RUNTIME_CHECK, SIZE_CHECK, and shark::RadiusMarginQuotient< InputType, CacheType >::Result::w2.

Referenced by main().

◆ evalDerivative()

template<class InputType , class CacheType = float>

double shark::RadiusMarginQuotient< InputType, CacheType >::evalDerivative	(	SearchPointType const &	parameters,
		FirstOrderDerivative &	derivative
	)		const

inlinevirtual

Evaluate the radius margin quotient and its first derivative.

: The parameters are passed into the kernel, and the radius-margin quotient and its derivative are computed w.r.t. the kernel-induced metric.

Reimplemented from shark::AbstractObjectiveFunction< RealVector, double >.

Definition at line 113 of file RadiusMarginQuotient.h.

References shark::RadiusMarginQuotient< InputType, CacheType >::Result::alpha, shark::RadiusMarginQuotient< InputType, CacheType >::Result::beta, shark::calculateKernelMatrixParameterDerivative(), shark::RadiusMarginQuotient< InputType, CacheType >::computeRadiusMargin(), shark::LabeledData< InputT, LabelT >::empty(), shark::LabeledData< InputT, LabelT >::inputs(), shark::RadiusMarginQuotient< InputType, CacheType >::m_dataset, shark::AbstractObjectiveFunction< RealVector, double >::m_evaluationCounter, shark::RadiusMarginQuotient< InputType, CacheType >::mep_kernel, shark::IParameterizable< VectorType >::numberOfParameters(), shark::RadiusMarginQuotient< InputType, CacheType >::Result::R2, shark::IParameterizable< VectorType >::setParameterVector(), SHARK_RUNTIME_CHECK, SIZE_CHECK, and shark::RadiusMarginQuotient< InputType, CacheType >::Result::w2.

◆ name()

template<class InputType , class CacheType = float>

std::string shark::RadiusMarginQuotient< InputType, CacheType >::name ( ) const

inlinevirtual

From INameable: return the class name.

Reimplemented from shark::INameable.

Definition at line 81 of file RadiusMarginQuotient.h.

◆ numberOfVariables()

template<class InputType , class CacheType = float>

std::size_t shark::RadiusMarginQuotient< InputType, CacheType >::numberOfVariables ( ) const

inlinevirtual

Accesses the number of variables.

Implements shark::AbstractObjectiveFunction< RealVector, double >.

Definition at line 84 of file RadiusMarginQuotient.h.

References shark::RadiusMarginQuotient< InputType, CacheType >::mep_kernel, and shark::IParameterizable< VectorType >::numberOfParameters().

Member Data Documentation

◆ m_dataset

template<class InputType , class CacheType = float>

DatasetType shark::RadiusMarginQuotient< InputType, CacheType >::m_dataset

protected

labeled data for radius and (hard) margin computation

Definition at line 186 of file RadiusMarginQuotient.h.

Referenced by shark::RadiusMarginQuotient< InputType, CacheType >::computeRadiusMargin(), shark::RadiusMarginQuotient< InputType, CacheType >::eval(), and shark::RadiusMarginQuotient< InputType, CacheType >::evalDerivative().

◆ mep_kernel

template<class InputType , class CacheType = float>

KernelType* shark::RadiusMarginQuotient< InputType, CacheType >::mep_kernel

protected

underlying parameterized kernel object

Definition at line 185 of file RadiusMarginQuotient.h.

Referenced by shark::RadiusMarginQuotient< InputType, CacheType >::computeRadiusMargin(), shark::RadiusMarginQuotient< InputType, CacheType >::eval(), shark::RadiusMarginQuotient< InputType, CacheType >::evalDerivative(), shark::RadiusMarginQuotient< InputType, CacheType >::numberOfVariables(), and shark::RadiusMarginQuotient< InputType, CacheType >::RadiusMarginQuotient().

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

include/shark/ObjectiveFunctions/RadiusMarginQuotient.h

Classes

Public Types

Public Member Functions

Protected Member Functions

Protected Attributes

Detailed Description

Member Typedef Documentation

◆ CachedMatrixType

◆ DatasetType

◆ KernelMatrixType

◆ KernelType

◆ QpFloatType

Constructor & Destructor Documentation

◆ RadiusMarginQuotient()

Member Function Documentation

◆ computeRadiusMargin()

◆ eval()

◆ evalDerivative()

◆ name()

◆ numberOfVariables()

Member Data Documentation

◆ m_dataset

◆ mep_kernel