Standard linear Bayesian regression or, equivalently, Gaussian Process Regression with a linear covariance function. More...

#include <LinearGPRLearner.h>

Inheritance diagram for iCub::learningmachine::LinearGPRLearner:

Public Member Functions
	LinearGPRLearner (unsigned int dom=1, unsigned int cod=1, double sigma=1.0)
	Constructor.

	LinearGPRLearner (const LinearGPRLearner &other)
	Copy constructor.

virtual	~LinearGPRLearner ()
	Destructor.

LinearGPRLearner &	operator= (const LinearGPRLearner &other)
	Assignment operator.

virtual void	feedSample (const yarp::sig::Vector &input, const yarp::sig::Vector &output)
	Provide the learning machine with an example of the desired mapping.

virtual void	train ()
	Train the learning machine on the examples that have been supplied so far.

virtual Prediction	predict (const yarp::sig::Vector &input)
	Ask the learning machine to predict the output for a given input.

void	reset ()
	Forget everything and start over.

LinearGPRLearner *	clone ()
	Asks the learning machine to return a clone of its type.

virtual std::string	getInfo ()
	Asks the learning machine to return a string containing information on its operation so far.

virtual std::string	getConfigHelp ()
	Asks the learning machine to return a string containing the list of configuration options that it supports.

virtual void	writeBottle (yarp::os::Bottle &bot)

virtual void	readBottle (yarp::os::Bottle &bot)
	Unserializes a machine from a bottle.

void	setDomainSize (unsigned int size)
	Mutator for the domain size.

void	setCoDomainSize (unsigned int size)
	Mutator for the codomain size.

void	setSigma (double s)
	Sets the signal noise \sigma to a specified value.

double	getSigma ()
	Accessor for the signal noise \sigma.

virtual bool	configure (yarp::os::Searchable &config)
	Change parameters.

Public Member Functions inherited from iCub::learningmachine::IFixedSizeLearner
	IFixedSizeLearner (unsigned int dom=1, unsigned int cod=1)
	Constructor.

unsigned int	getDomainSize () const
	Returns the size (dimensionality) of the input domain.

unsigned int	getCoDomainSize () const
	Returns the size (dimensionality) of the output domain (codomain).

Public Member Functions inherited from iCub::learningmachine::IMachineLearner
	IMachineLearner ()
	Constructor.

virtual	~IMachineLearner ()
	Destructor (empty).

virtual bool	open (yarp::os::Searchable &config)
	Initialize the object.

virtual bool	close ()
	Shut the object down.

bool	write (yarp::os::ConnectionWriter &connection) const

bool	read (yarp::os::ConnectionReader &connection)

virtual std::string	toString ()
	Asks the learning machine to return a string serialization.

virtual bool	fromString (const std::string &str)
	Asks the learning machine to initialize from a string serialization.

std::string	getName () const
	Retrieve the name of this machine learning technique.

void	setName (const std::string &name)
	Set the name of this machine learning technique.

Additional Inherited Members
Protected Member Functions inherited from iCub::learningmachine::IFixedSizeLearner
virtual bool	checkDomainSize (const yarp::sig::Vector &input)
	Checks whether the input is of the desired dimensionality.

virtual bool	checkCoDomainSize (const yarp::sig::Vector &output)
	Checks whether the output is of the desired dimensionality.

void	validateDomainSizes (const yarp::sig::Vector &input, const yarp::sig::Vector &output)
	Validates whether the input and output are of the desired dimensionality.

virtual void	writeBottle (yarp::os::Bottle &bot) const
	Writes a serialization of the machine into a bottle.

Protected Attributes inherited from iCub::learningmachine::IFixedSizeLearner
unsigned int	domainSize
	The dimensionality of the input domain.

unsigned int	coDomainSize
	The dimensionality of the output domain (codomain).

Protected Attributes inherited from iCub::learningmachine::IMachineLearner
std::string	name
	The name of this type of machine learner.

Detailed Description

Standard linear Bayesian regression or, equivalently, Gaussian Process Regression with a linear covariance function.

It uses a rank 1 update rule to incrementally update the Cholesky factor of the covariance matrix.

See: Gaussian Processes for Machine Learning. Carl Edward Rasmussen and Christopher K. I. Williams. The MIT Press, 2005.

Pattern Recognition and Machine Learning. Christopher M. Bishop. Springer-Verlag, 2006.

See also: iCub::learningmachine::IMachineLearner; iCub::learningmachine::IFixedSizeLearner; iCub::learningmachine::RLSLearner

Author: Arjan Gijsberts

Definition at line 59 of file LinearGPRLearner.h.

Constructor & Destructor Documentation

◆ LinearGPRLearner() [1/2]

iCub::learningmachine::LinearGPRLearner::LinearGPRLearner	(	unsigned int	dom = `1`,
		unsigned int	cod = `1`,
		double	sigma = `1.0`
	)

Constructor.

Parameters

dom	initial domain size
cod	initial codomain size
sigma	initial value for signal noise \sigma

Definition at line 41 of file LinearGPRLearner.cpp.

◆ LinearGPRLearner() [2/2]

iCub::learningmachine::LinearGPRLearner::LinearGPRLearner ( const LinearGPRLearner & other )

Copy constructor.

Definition at line 52 of file LinearGPRLearner.cpp.

◆ ~LinearGPRLearner()

iCub::learningmachine::LinearGPRLearner::~LinearGPRLearner ( )

virtual

Destructor.

Definition at line 57 of file LinearGPRLearner.cpp.

Member Function Documentation

◆ clone()

LinearGPRLearner * iCub::learningmachine::LinearGPRLearner::clone ( )

inlinevirtual

Asks the learning machine to return a clone of its type.

Returns: a clone of the current learner

Implements iCub::learningmachine::IMachineLearner.

Definition at line 134 of file LinearGPRLearner.h.

◆ configure()

bool iCub::learningmachine::LinearGPRLearner::configure ( yarp::os::Searchable & config )

virtual

Change parameters.

Reimplemented from iCub::learningmachine::IFixedSizeLearner.

Definition at line 170 of file LinearGPRLearner.cpp.

◆ feedSample()

void iCub::learningmachine::LinearGPRLearner::feedSample	(	const yarp::sig::Vector &	input,
		const yarp::sig::Vector &	output
	)

virtual

Provide the learning machine with an example of the desired mapping.

Parameters

input	a sample input
output	the corresponding output

Reimplemented from iCub::learningmachine::IFixedSizeLearner.

Definition at line 74 of file LinearGPRLearner.cpp.

◆ getConfigHelp()

std::string iCub::learningmachine::LinearGPRLearner::getConfigHelp ( )

virtual

Asks the learning machine to return a string containing the list of configuration options that it supports.

Returns: an informative description of the configuration options

Reimplemented from iCub::learningmachine::IFixedSizeLearner.

Definition at line 127 of file LinearGPRLearner.cpp.

◆ getInfo()

std::string iCub::learningmachine::LinearGPRLearner::getInfo ( )

virtual

Asks the learning machine to return a string containing information on its operation so far.

Returns: the information on the machine

Reimplemented from iCub::learningmachine::IFixedSizeLearner.

Definition at line 114 of file LinearGPRLearner.cpp.

◆ getSigma()

double iCub::learningmachine::LinearGPRLearner::getSigma ( )

Accessor for the signal noise \sigma.

Returns: the value of the parameter

Definition at line 165 of file LinearGPRLearner.cpp.

◆ operator=()

LinearGPRLearner & iCub::learningmachine::LinearGPRLearner::operator= ( const LinearGPRLearner & other )

Assignment operator.

Definition at line 60 of file LinearGPRLearner.cpp.

◆ predict()

Prediction iCub::learningmachine::LinearGPRLearner::predict ( const yarp::sig::Vector & input )

virtual

Ask the learning machine to predict the output for a given input.

Parameters

input the input

Returns: the expected output

Implements iCub::learningmachine::IMachineLearner.

Definition at line 93 of file LinearGPRLearner.cpp.

◆ readBottle()

void iCub::learningmachine::LinearGPRLearner::readBottle ( yarp::os::Bottle & bot )

virtual

Unserializes a machine from a bottle.

This method is internally referenced by the read method. Typically, subclasses should override this method instead of overriding the read method directly.

Parameters

bot	the bottle

Reimplemented from iCub::learningmachine::IFixedSizeLearner.

Definition at line 140 of file LinearGPRLearner.cpp.

◆ reset()

void iCub::learningmachine::LinearGPRLearner::reset ( )

virtual

Forget everything and start over.

Implements iCub::learningmachine::IMachineLearner.

Definition at line 107 of file LinearGPRLearner.cpp.

◆ setCoDomainSize()

void iCub::learningmachine::LinearGPRLearner::setCoDomainSize ( unsigned int size )

virtual

Mutator for the codomain size.

Parameters

size	the desired codomain size

Reimplemented from iCub::learningmachine::IFixedSizeLearner.

Definition at line 151 of file LinearGPRLearner.cpp.

◆ setDomainSize()

void iCub::learningmachine::LinearGPRLearner::setDomainSize ( unsigned int size )

virtual

Mutator for the domain size.

Parameters

size	the desired domain size

Reimplemented from iCub::learningmachine::IFixedSizeLearner.

Definition at line 146 of file LinearGPRLearner.cpp.

◆ setSigma()

void iCub::learningmachine::LinearGPRLearner::setSigma ( double s )

Sets the signal noise \sigma to a specified value.

This resets the machine.

Parameters

s	the desired value.

Definition at line 156 of file LinearGPRLearner.cpp.

◆ train()

void iCub::learningmachine::LinearGPRLearner::train ( )

virtual

Train the learning machine on the examples that have been supplied so far.

This method is primarily intended to be used for offline/batch learning machines. It explicitly initiates the training routine on those machines for the samples that have been collected so far.

Reimplemented from iCub::learningmachine::IFixedSizeLearner.

Definition at line 89 of file LinearGPRLearner.cpp.

◆ writeBottle()

void iCub::learningmachine::LinearGPRLearner::writeBottle ( yarp::os::Bottle & bot )

virtual

Definition at line 134 of file LinearGPRLearner.cpp.

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

icub-main/src/libraries/learningMachine/include/iCub/learningMachine/LinearGPRLearner.h
icub-main/src/libraries/learningMachine/src/LinearGPRLearner.cpp

Public Member Functions

Additional Inherited Members

Detailed Description

Constructor & Destructor Documentation

◆ LinearGPRLearner() [1/2]

◆ LinearGPRLearner() [2/2]

◆ ~LinearGPRLearner()

Member Function Documentation

◆ clone()

◆ configure()

◆ feedSample()

◆ getConfigHelp()

◆ getInfo()

◆ getSigma()

◆ operator=()

◆ predict()

◆ readBottle()

◆ reset()

◆ setCoDomainSize()

◆ setDomainSize()

◆ setSigma()

◆ train()

◆ writeBottle()