A class that implements standardization as a preprocessing step. More...

#include <Standardizer.h>

Inheritance diagram for iCub::learningmachine::Standardizer:

Public Member Functions
	Standardizer (double m=0., double s=1.)
	Constructor.

virtual void	update (double val)
	Feeds a single sample into the scaler, so that it can use this sample to update the offset and scale.

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

virtual bool	configure (yarp::os::Searchable &config)

Standardizer *	clone ()
	Asks the scaler to return a new object of its type.

virtual double	getDesiredMean ()
	Accessor for the desired mean.

virtual void	setDesiredMean (double m)
	Mutator for the desired mean.

virtual double	getDesiredStd ()
	Accessor for the desired standard deviation.

virtual void	setDesiredStd (double s)
	Mutator for the desired standard deviation.

Public Member Functions inherited from iCub::learningmachine::IScaler
	IScaler (double s=1., double o=0.)
	Constructor.

virtual	~IScaler ()
	Destructor (empty).

virtual double	transform (double val)
	Transforms a single sample value according to the state of the scaler.

virtual double	unTransform (double val)
	Untransforms a single sample value according to the state of the scaler.

std::string	getName () const
	Retrieve the name of this scaler.

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

virtual void	setUpdateEnabled (bool u)
	Mutator for the update state.

virtual bool	getUpdateEnabled ()
	Accessor for the update state.

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

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

Protected Member Functions
virtual void	writeBottle (yarp::os::Bottle &bot)
	Writes a serialization of the scaler into a bottle.

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

Protected Attributes
int	noSamples
	The number of samples that have been received so far.

double	mean
	Desired mean for the output distribution.

double	std
	Desired standard deviation for the output distribution.

double	runningMean
	Running mean based on the samples seen so far.

double	runningStd
	Running standard deviation based on the samples seen so far.

double	squaredErrors
	Temporary variable that counts the sum of the squared errors.

Protected Attributes inherited from iCub::learningmachine::IScaler
double	offset
	The offset for the linear transformation.

double	scale
	The scale for the linear transformation.

std::string	name
	The name of this type of scaler.

bool	updateEnabled
	Boolean indicating whether the scaler has to update each sample.

Detailed Description

A class that implements standardization as a preprocessing step.

Standardization is the process of converting all samples such that the set has a zero mean and unit standard deviation. In this particular implementation both the mean and standard deviation are calculated using a 'running' computation.

See also: iCub::learningmachine::IScaler

Author: Arjan Gijsberts

Definition at line 43 of file Standardizer.h.

Constructor & Destructor Documentation

◆ Standardizer()

iCub::learningmachine::Standardizer::Standardizer	(	double	m = `0.`,
		double	s = `1.`
	)

Constructor.

Parameters

m	desired output mean
s	desired output standard deviation

Definition at line 32 of file Standardizer.cpp.

Member Function Documentation

◆ clone()

Standardizer * iCub::learningmachine::Standardizer::clone ( )

inlinevirtual

Asks the scaler to return a new object of its type.

Returns: a fresh instance of the specified type

Implements iCub::learningmachine::IScaler.

Definition at line 112 of file Standardizer.h.

◆ configure()

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

virtual

Reimplemented from iCub::learningmachine::IScaler.

Definition at line 80 of file Standardizer.cpp.

◆ getDesiredMean()

virtual double iCub::learningmachine::Standardizer::getDesiredMean ( )

inlinevirtual

Accessor for the desired mean.

Definition at line 119 of file Standardizer.h.

◆ getDesiredStd()

virtual double iCub::learningmachine::Standardizer::getDesiredStd ( )

inlinevirtual

Accessor for the desired standard deviation.

Definition at line 131 of file Standardizer.h.

◆ getInfo()

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

virtual

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

Returns: the statistics of the machine

Reimplemented from iCub::learningmachine::IScaler.

Definition at line 56 of file Standardizer.cpp.

◆ readBottle()

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

protectedvirtual

Unserializes a scaler from a bottle.

Parameters

bot	the bottle

Reimplemented from iCub::learningmachine::IScaler.

Definition at line 72 of file Standardizer.cpp.

◆ setDesiredMean()

virtual void iCub::learningmachine::Standardizer::setDesiredMean ( double m )

inlinevirtual

Mutator for the desired mean.

Parameters

m	the new mean

Definition at line 126 of file Standardizer.h.

◆ setDesiredStd()

virtual void iCub::learningmachine::Standardizer::setDesiredStd ( double s )

inlinevirtual

Mutator for the desired standard deviation.

Parameters

s	the new standard deviation

Definition at line 138 of file Standardizer.h.

◆ update()

void iCub::learningmachine::Standardizer::update ( double val )

virtual

Feeds a single sample into the scaler, so that it can use this sample to update the offset and scale.

Parameters

value the sample value

Reimplemented from iCub::learningmachine::IScaler.

Definition at line 39 of file Standardizer.cpp.

◆ writeBottle()

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

protectedvirtual

Writes a serialization of the scaler into a bottle.

Parameters

bot	the bottle

Reimplemented from iCub::learningmachine::IScaler.

Definition at line 64 of file Standardizer.cpp.

Member Data Documentation

◆ mean

double iCub::learningmachine::Standardizer::mean

protected

Desired mean for the output distribution.

Definition at line 53 of file Standardizer.h.

◆ noSamples

int iCub::learningmachine::Standardizer::noSamples

protected

The number of samples that have been received so far.

Definition at line 48 of file Standardizer.h.

◆ runningMean

double iCub::learningmachine::Standardizer::runningMean

protected

Running mean based on the samples seen so far.

Definition at line 63 of file Standardizer.h.

◆ runningStd

double iCub::learningmachine::Standardizer::runningStd

protected

Running standard deviation based on the samples seen so far.

Definition at line 68 of file Standardizer.h.

◆ squaredErrors

double iCub::learningmachine::Standardizer::squaredErrors

protected

Temporary variable that counts the sum of the squared errors.

Definition at line 73 of file Standardizer.h.

◆ std

double iCub::learningmachine::Standardizer::std

protected

Desired standard deviation for the output distribution.

Definition at line 58 of file Standardizer.h.

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

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

Public Member Functions

Protected Member Functions

Protected Attributes

Detailed Description

Constructor & Destructor Documentation

◆ Standardizer()

Member Function Documentation

◆ clone()

◆ configure()

◆ getDesiredMean()

◆ getDesiredStd()

◆ getInfo()

◆ readBottle()

◆ setDesiredMean()

◆ setDesiredStd()

◆ update()

◆ writeBottle()

Member Data Documentation

◆ mean

◆ noSamples

◆ runningMean

◆ runningStd

◆ squaredErrors

◆ std