A generalized interface for a learning machine for offline and online learning machines (e.g. More...

#include <IMachineLearner.h>

Inheritance diagram for iCub::learningmachine::IMachineLearner:

Public Member Functions
	IMachineLearner ()
	Constructor. More...

virtual	~IMachineLearner ()
	Destructor (empty). More...

virtual void	feedSample (const yarp::sig::Vector &input, const yarp::sig::Vector &output)=0
	Provide the learning machine with an example of the desired mapping. More...

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

virtual Prediction	predict (const yarp::sig::Vector &input)=0
	Ask the learning machine to predict the output for a given input. More...

virtual IMachineLearner *	clone ()=0
	Asks the learning machine to return a clone of its type. More...

virtual void	reset ()=0
	Forget everything and start over. More...

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

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

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

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

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

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

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

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

Protected Member Functions
virtual void	writeBottle (yarp::os::Bottle &bot) const =0
	Writes a serialization of the machine into a bottle. More...

virtual void	readBottle (yarp::os::Bottle &bot)=0
	Unserializes a machine from a bottle. More...

Protected Attributes
std::string	name
	The name of this type of machine learner. More...

Detailed Description

A generalized interface for a learning machine for offline and online learning machines (e.g.

SVM, LSSVM, ANN). This interface extends the Learner interface, which imposes certain limitations. Learning machines that meet this interface can, if desired, be made into two executable modules by iCub::contrib::LearnModule and iCub::contrib::PredictModule.

The learning machine can be used for regression and classification from R^* to R^*.

Be aware that objects of this class type cannot be written directly on a port, although this interface implements Portable. The reason is that this is an abstract base class, which conflicts with the template port readers and writers. See MachinePortable for how to send Learning Machines over ports.

See also: iCub::contrib::Learner; iCub::contrib::IFixedSizeLearner; iCub::contrib::MachinePortable; iCub::contrib::MachineFactory

Author: Arjan Gijsberts

Definition at line 89 of file IMachineLearner.h.

Constructor & Destructor Documentation

◆ IMachineLearner()

iCub::learningmachine::IMachineLearner::IMachineLearner ( )

inline

Constructor.

Definition at line 118 of file IMachineLearner.h.

◆ ~IMachineLearner()

virtual iCub::learningmachine::IMachineLearner::~IMachineLearner ( )

inlinevirtual

Destructor (empty).

Definition at line 123 of file IMachineLearner.h.

Member Function Documentation

◆ clone()

virtual IMachineLearner* iCub::learningmachine::IMachineLearner::clone ( )

pure virtual

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

Returns: a clone of the current learner

Implemented in iCub::learningmachine::RLSLearner, iCub::learningmachine::LSSVMLearner, iCub::learningmachine::LinearGPRLearner, iCub::learningmachine::DummyLearner, and iCub::learningmachine::DatasetRecorder.

◆ feedSample()

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

pure virtual

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

Parameters

input	a sample input
output	the corresponding output

Implemented in iCub::learningmachine::RLSLearner, iCub::learningmachine::LSSVMLearner, iCub::learningmachine::LinearGPRLearner, iCub::learningmachine::IFixedSizeLearner, iCub::learningmachine::DummyLearner, and iCub::learningmachine::DatasetRecorder.

◆ fromString()

virtual bool iCub::learningmachine::IMachineLearner::fromString ( const std::string & str )

inlinevirtual

Asks the learning machine to initialize from a string serialization.

Returns: true on succes

Definition at line 217 of file IMachineLearner.h.

◆ getConfigHelp()

virtual std::string iCub::learningmachine::IMachineLearner::getConfigHelp ( )

inlinevirtual

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 in iCub::learningmachine::RLSLearner, iCub::learningmachine::LSSVMLearner, iCub::learningmachine::LinearGPRLearner, iCub::learningmachine::IFixedSizeLearner, and iCub::learningmachine::DatasetRecorder.

Definition at line 196 of file IMachineLearner.h.

◆ getInfo()

virtual std::string iCub::learningmachine::IMachineLearner::getInfo ( )

inlinevirtual

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

Returns: the information on the machine

Reimplemented in iCub::learningmachine::RLSLearner, iCub::learningmachine::LSSVMLearner, iCub::learningmachine::LinearGPRLearner, iCub::learningmachine::IFixedSizeLearner, iCub::learningmachine::DummyLearner, and iCub::learningmachine::DatasetRecorder.

Definition at line 186 of file IMachineLearner.h.

◆ getName()

std::string iCub::learningmachine::IMachineLearner::getName ( ) const

inline

Retrieve the name of this machine learning technique.

Returns: the name of this machine learner

Definition at line 228 of file IMachineLearner.h.

◆ predict()

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

pure virtual

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

Parameters

input the input

Returns: the expected output

Implemented in iCub::learningmachine::RLSLearner, iCub::learningmachine::LSSVMLearner, iCub::learningmachine::LinearGPRLearner, iCub::learningmachine::DummyLearner, and iCub::learningmachine::DatasetRecorder.

◆ read()

bool iCub::learningmachine::IMachineLearner::read ( yarp::os::ConnectionReader & connection )

inline

Definition at line 173 of file IMachineLearner.h.

◆ readBottle()

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

protectedpure 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

Implemented in iCub::learningmachine::RLSLearner, iCub::learningmachine::LSSVMLearner, iCub::learningmachine::LinearGPRLearner, iCub::learningmachine::IFixedSizeLearner, iCub::learningmachine::DummyLearner, and iCub::learningmachine::DatasetRecorder.

◆ reset()

virtual void iCub::learningmachine::IMachineLearner::reset ( )

pure virtual

Forget everything and start over.

Implemented in iCub::learningmachine::RLSLearner, iCub::learningmachine::LSSVMLearner, iCub::learningmachine::LinearGPRLearner, iCub::learningmachine::DummyLearner, and iCub::learningmachine::DatasetRecorder.

◆ setName()

void iCub::learningmachine::IMachineLearner::setName ( const std::string & name )

inline

Set the name of this machine learning technique.

Parameters

name	the new name

Definition at line 237 of file IMachineLearner.h.

◆ toString()

virtual std::string iCub::learningmachine::IMachineLearner::toString ( )

inlinevirtual

Asks the learning machine to return a string serialization.

Returns: a string serialization of the machine

Definition at line 206 of file IMachineLearner.h.

◆ train()

virtual void iCub::learningmachine::IMachineLearner::train ( )

inlinevirtual

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 in iCub::learningmachine::RLSLearner, iCub::learningmachine::LSSVMLearner, iCub::learningmachine::LinearGPRLearner, iCub::learningmachine::IFixedSizeLearner, iCub::learningmachine::DummyLearner, and iCub::learningmachine::DatasetRecorder.

Definition at line 139 of file IMachineLearner.h.

◆ write()

bool iCub::learningmachine::IMachineLearner::write ( yarp::os::ConnectionWriter & connection ) const

inline

Definition at line 164 of file IMachineLearner.h.

◆ writeBottle()

virtual void iCub::learningmachine::IMachineLearner::writeBottle ( yarp::os::Bottle & bot ) const

protectedpure virtual

Writes a serialization of the machine into a bottle.

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

Parameters

bot	the bottle containing the machine serialization

Implemented in iCub::learningmachine::IFixedSizeLearner, iCub::learningmachine::DummyLearner, and iCub::learningmachine::DatasetRecorder.

Member Data Documentation

◆ name

std::string iCub::learningmachine::IMachineLearner::name

protected

The name of this type of machine learner.

Definition at line 94 of file IMachineLearner.h.

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

icub-main/src/libraries/learningMachine/include/iCub/learningMachine/IMachineLearner.h

Public Member Functions

Protected Member Functions

Protected Attributes

Detailed Description

Constructor & Destructor Documentation

◆ IMachineLearner()

◆ ~IMachineLearner()

Member Function Documentation

◆ clone()

◆ feedSample()

◆ fromString()

◆ getConfigHelp()

◆ getInfo()

◆ getName()

◆ predict()

◆ read()

◆ readBottle()

◆ reset()

◆ setName()

◆ toString()

◆ train()

◆ write()

◆ writeBottle()

Member Data Documentation

◆ name