Models a convex combination of inputs. More...

#include <shark/Models/ConvexCombination.h>

Inheritance diagram for shark::ConvexCombination:

Public Member Functions
	ConvexCombination ()
	CDefault Constructor; use setStructure later. More...

	ConvexCombination (std::size_t inputs, std::size_t outputs=1)
	Constructor creating a model with given dimnsionalities and optional offset term. More...

	ConvexCombination (RealMatrix const &matrix)
	Construction from matrix. More...

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

ConvexCombination &	operator= (ConvexCombination const &model)
	operator = More...

std::size_t	inputSize () const
	obtain the input dimension More...

std::size_t	outputSize () const
	obtain the output dimension More...

RealVector	parameterVector () const
	obtain the parameter vector More...

void	setParameterVector (RealVector const &newParameters)
	overwrite the parameter vector More...

std::size_t	numberOfParameters () const
	return the number of parameter More...

void	setStructure (std::size_t inputs, std::size_t outputs=1)
	overwrite structure and parameters More...

RealMatrix const &	weights () const

RealMatrix &	weights ()

boost::shared_ptr< State >	createState () const
	Creates an internal state of the model. More...

void	eval (BatchInputType const &inputs, BatchOutputType &outputs) const
	Evaluate the model: output = w * input. More...

void	eval (BatchInputType const &inputs, BatchOutputType &outputs, State &state) const
	Evaluate the model: output = w *input. More...

void	weightedParameterDerivative (BatchInputType const &patterns, RealMatrix const &coefficients, State const &state, RealVector &gradient) const
	Calculates the first derivative w.r.t the parameters and summing them up over all patterns of the last computed batch. More...

void	weightedInputDerivative (BatchInputType const &patterns, BatchOutputType const &coefficients, State const &state, BatchInputType &derivative) const
	Calculates the first derivative w.r.t the inputs and summs them up over all patterns of the last computed batch. More...

void	read (InArchive &archive)
	From ISerializable. More...

void	write (OutArchive &archive) const
	From ISerializable. More...

Public Member Functions inherited from shark::AbstractModel< RealVector, RealVector >
	AbstractModel ()

virtual	~AbstractModel ()

const Features &	features () const

virtual void	updateFeatures ()

bool	hasFirstParameterDerivative () const
	Returns true when the first parameter derivative is implemented. More...

bool	hasSecondParameterDerivative () const
	Returns true when the second parameter derivative is implemented. More...

bool	hasFirstInputDerivative () const
	Returns true when the first input derivative is implemented. More...

bool	hasSecondInputDerivative () const
	Returns true when the second parameter derivative is implemented. More...

bool	isSequential () const

virtual void	eval (InputType const &pattern, OutputType &output) const
	Standard interface for evaluating the response of the model to a single pattern. More...

Data< OutputType >	operator() (Data< InputType > const &patterns) const
	Model evaluation as an operator for a whole dataset. This is a convenience function. More...

OutputType	operator() (InputType const &pattern) const
	Model evaluation as an operator for a single pattern. This is a convenience function. More...

BatchOutputType	operator() (BatchInputType const &patterns) const
	Model evaluation as an operator for a single pattern. This is a convenience function. More...

virtual void	weightedParameterDerivative (BatchInputType const &pattern, BatchOutputType const &coefficients, State const &state, RealVector &derivative) const
	calculates the weighted sum of derivatives w.r.t the parameters. More...

virtual void	weightedParameterDerivative (BatchInputType const &pattern, BatchOutputType const &coefficients, Batch< RealMatrix >::type const &errorHessian, State const &state, RealVector &derivative, RealMatrix &hessian) const
	calculates the weighted sum of derivatives w.r.t the parameters More...

virtual void	weightedInputDerivative (BatchInputType const &pattern, BatchOutputType const &coefficients, typename Batch< RealMatrix >::type const &errorHessian, State const &state, RealMatrix &derivative, Batch< RealMatrix >::type &hessian) const
	calculates the weighted sum of derivatives w.r.t the inputs More...

virtual void	weightedDerivatives (BatchInputType const &patterns, BatchOutputType const &coefficients, State const &state, RealVector &parameterDerivative, BatchInputType &inputDerivative) const
	calculates weighted input and parameter derivative at the same time More...

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

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

Public Member Functions inherited from shark::ISerializable
virtual	~ISerializable ()
	Virtual d'tor. More...

void	load (InArchive &archive, unsigned int version)
	Versioned loading of components, calls read(...). More...

void	save (OutArchive &archive, unsigned int version) const
	Versioned storing of components, calls write(...). More...

	BOOST_SERIALIZATION_SPLIT_MEMBER ()

Friends
void	swap (ConvexCombination &model1, ConvexCombination &model2)
	swap More...

Additional Inherited Members
Public Types inherited from shark::AbstractModel< RealVector, RealVector >
enum	Feature

typedef RealVector	InputType
	Defines the input type of the model. More...

typedef RealVector	OutputType
	Defines the output type of the model. More...

typedef Batch< InputType >::type	BatchInputType
	defines the batch type of the input type. More...

typedef Batch< OutputType >::type	BatchOutputType
	defines the batch type of the output type More...

typedef TypedFlags< Feature >	Features

typedef TypedFeatureNotAvailableException< Feature >	FeatureNotAvailableException

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

Detailed Description

Models a convex combination of inputs.

For a given input vector x, the convex combination returns \( f_i(x) = sum_j w_{ij} x_j \), where \( w_i > 0 \) and \( sum_j w_{ij} = 1\), that is the outputs of the model are a convex combination of the inputs.

To ensure that the constraints are fulfilled, the model uses a different set of weights q_i and \( w_{ij} = exp(q_{ij})/sum_j exp(q_{ik}) \). As usual, this encoding is only used for the derivatives and the parameter vectors, not when the weights are explicitely set. In the latter case, the user must provide a set of suitable \( w_{ij} \).

Definition at line 51 of file ConvexCombination.h.

Constructor & Destructor Documentation

◆ ConvexCombination() [1/3]

shark::ConvexCombination::ConvexCombination ( )

inline

CDefault Constructor; use setStructure later.

Definition at line 58 of file ConvexCombination.h.

References shark::AbstractModel< RealVector, RealVector >::HAS_FIRST_INPUT_DERIVATIVE, shark::AbstractModel< RealVector, RealVector >::HAS_FIRST_PARAMETER_DERIVATIVE, and shark::AbstractModel< RealVector, RealVector >::m_features.

◆ ConvexCombination() [2/3]

shark::ConvexCombination::ConvexCombination	(	std::size_t	inputs,
		std::size_t	outputs = `1`
	)

inline

Constructor creating a model with given dimnsionalities and optional offset term.

Definition at line 64 of file ConvexCombination.h.

References shark::AbstractModel< RealVector, RealVector >::HAS_FIRST_INPUT_DERIVATIVE, shark::AbstractModel< RealVector, RealVector >::HAS_FIRST_PARAMETER_DERIVATIVE, and shark::AbstractModel< RealVector, RealVector >::m_features.

◆ ConvexCombination() [3/3]

shark::ConvexCombination::ConvexCombination ( RealMatrix const & matrix )

inline

Construction from matrix.

Definition at line 71 of file ConvexCombination.h.

References shark::AbstractModel< RealVector, RealVector >::HAS_FIRST_INPUT_DERIVATIVE, shark::AbstractModel< RealVector, RealVector >::HAS_FIRST_PARAMETER_DERIVATIVE, and shark::AbstractModel< RealVector, RealVector >::m_features.

Member Function Documentation

◆ createState()

boost::shared_ptr<State> shark::ConvexCombination::createState ( ) const

inlinevirtual

Creates an internal state of the model.

The state is needed when the derivatives are to be calculated. Eval can store a state which is then reused to speed up the calculations of the derivatives. This also allows eval to be evaluated in parallel!

Reimplemented from shark::AbstractModel< RealVector, RealVector >.

Definition at line 135 of file ConvexCombination.h.

◆ eval() [1/2]

void shark::ConvexCombination::eval	(	BatchInputType const &	inputs,
		BatchOutputType &	outputs
	)		const

inlinevirtual

Evaluate the model: output = w * input.

Reimplemented from shark::AbstractModel< RealVector, RealVector >.

Definition at line 140 of file ConvexCombination.h.

References remora::noalias(), remora::prod(), and remora::trans().

Referenced by eval().

◆ eval() [2/2]

void shark::ConvexCombination::eval	(	BatchInputType const &	inputs,
		BatchOutputType &	outputs,
		State &	state
	)		const

inlinevirtual

Evaluate the model: output = w *input.

Implements shark::AbstractModel< RealVector, RealVector >.

Definition at line 145 of file ConvexCombination.h.

References eval().

◆ inputSize()

std::size_t shark::ConvexCombination::inputSize ( ) const

inline

obtain the input dimension

Definition at line 93 of file ConvexCombination.h.

Referenced by weightedInputDerivative(), and weightedParameterDerivative().

◆ name()

std::string shark::ConvexCombination::name ( ) const

inlinevirtual

From INameable: return the class name.

Reimplemented from shark::INameable.

Definition at line 77 of file ConvexCombination.h.

◆ numberOfParameters()

std::size_t shark::ConvexCombination::numberOfParameters ( ) const

inlinevirtual

return the number of parameter

Reimplemented from shark::IParameterizable.

Definition at line 117 of file ConvexCombination.h.

Referenced by weightedParameterDerivative().

◆ operator=()

ConvexCombination& shark::ConvexCombination::operator= ( ConvexCombination const & model )

inline

operator =

Definition at line 86 of file ConvexCombination.h.

References swap.

◆ outputSize()

std::size_t shark::ConvexCombination::outputSize ( ) const

inline

obtain the output dimension

Definition at line 98 of file ConvexCombination.h.

Referenced by setParameterVector(), weightedInputDerivative(), and weightedParameterDerivative().

◆ parameterVector()

RealVector shark::ConvexCombination::parameterVector ( ) const

inlinevirtual

obtain the parameter vector

Reimplemented from shark::IParameterizable.

Definition at line 103 of file ConvexCombination.h.

References remora::to_vector().

◆ read()

void shark::ConvexCombination::read ( InArchive & archive )

inlinevirtual

From ISerializable.

Reimplemented from shark::AbstractModel< RealVector, RealVector >.

Definition at line 188 of file ConvexCombination.h.

◆ setParameterVector()

void shark::ConvexCombination::setParameterVector ( RealVector const & newParameters )

inlinevirtual

overwrite the parameter vector

Reimplemented from shark::IParameterizable.

Definition at line 108 of file ConvexCombination.h.

References remora::noalias(), outputSize(), remora::row(), remora::sum(), and remora::to_matrix().

◆ setStructure()

void shark::ConvexCombination::setStructure	(	std::size_t	inputs,
		std::size_t	outputs = `1`
	)

inline

overwrite structure and parameters

Definition at line 122 of file ConvexCombination.h.

References swap.

Referenced by main().

◆ weightedInputDerivative()

void shark::ConvexCombination::weightedInputDerivative	(	BatchInputType const &	patterns,
		BatchOutputType const &	coefficients,
		State const &	state,
		BatchInputType &	derivative
	)		const

inlinevirtual

Calculates the first derivative w.r.t the inputs and summs them up over all patterns of the last computed batch.

Reimplemented from shark::AbstractModel< RealVector, RealVector >.

Definition at line 174 of file ConvexCombination.h.

References inputSize(), remora::noalias(), outputSize(), remora::prod(), and SIZE_CHECK.

◆ weightedParameterDerivative()

void shark::ConvexCombination::weightedParameterDerivative	(	BatchInputType const &	patterns,
		RealMatrix const &	coefficients,
		State const &	state,
		RealVector &	gradient
	)		const

inline

Calculates the first derivative w.r.t the parameters and summing them up over all patterns of the last computed batch.

Definition at line 150 of file ConvexCombination.h.

References remora::inner_prod(), inputSize(), remora::noalias(), numberOfParameters(), outputSize(), remora::prod(), remora::row(), SIZE_CHECK, remora::to_matrix(), and remora::trans().

◆ weights() [1/2]

RealMatrix const& shark::ConvexCombination::weights ( ) const

inline

Definition at line 127 of file ConvexCombination.h.

◆ weights() [2/2]

RealMatrix& shark::ConvexCombination::weights ( )

inline

Definition at line 131 of file ConvexCombination.h.

◆ write()

void shark::ConvexCombination::write ( OutArchive & archive ) const

inlinevirtual

From ISerializable.

Reimplemented from shark::AbstractModel< RealVector, RealVector >.

Definition at line 192 of file ConvexCombination.h.

Friends And Related Function Documentation

◆ swap

void swap	(	ConvexCombination &	model1,
		ConvexCombination &	model2
	)

friend

swap

Definition at line 81 of file ConvexCombination.h.

Referenced by operator=(), and setStructure().

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

include/shark/Models/ConvexCombination.h

Public Member Functions

Friends

Additional Inherited Members

Detailed Description

Constructor & Destructor Documentation

◆ ConvexCombination() [1/3]

◆ ConvexCombination() [2/3]

◆ ConvexCombination() [3/3]

Member Function Documentation

◆ createState()

◆ eval() [1/2]

◆ eval() [2/2]

◆ inputSize()

◆ name()

◆ numberOfParameters()

◆ operator=()

◆ outputSize()

◆ parameterVector()

◆ read()

◆ setParameterVector()

◆ setStructure()

◆ weightedInputDerivative()

◆ weightedParameterDerivative()

◆ weights() [1/2]

◆ weights() [2/2]

◆ write()

Friends And Related Function Documentation

◆ swap