Feed-Forward 2 layers Neural Network. More...

#include <neuralNetworks.h>

Inheritance diagram for iCub::ctrl::ff2LayNN:

Classes
struct	minmax

Public Member Functions
	ff2LayNN ()
	Create an empty network. More...

	ff2LayNN (const yarp::os::Property &options)
	Create and configure the network. More...

virtual bool	configure (const yarp::os::Property &options)
	Configure/reconfigure the network. More...

virtual bool	isValid () const
	Return the internal status after a configuration. More...

virtual yarp::sig::Vector	predict (const yarp::sig::Vector &x) const
	Predict the output given a certain input to the network. More...

virtual bool	getStructure (yarp::os::Property &options) const
	Retrieve the network structure as a Property object. More...

virtual bool	printStructure (std::ostream &stream=std::cout) const
	Dump tadily the network structure on the stream. More...

std::deque< yarp::sig::Vector > &	get_IW ()
	Retrieve first layer weights. More...

std::deque< yarp::sig::Vector > &	get_LW ()
	Retrieve second layer weights. More...

yarp::sig::Vector &	get_b1 ()
	Retrieve first layer bias. More...

yarp::sig::Vector &	get_b2 ()
	Retrieve second layer bias. More...

virtual yarp::sig::Vector	scaleInputToNetFormat (const yarp::sig::Vector &x) const
	Scale input to be used with the network. More...

virtual yarp::sig::Vector	scaleInputFromNetFormat (const yarp::sig::Vector &x) const
	Scale back input from the network's format. More...

virtual yarp::sig::Vector	scaleOutputToNetFormat (const yarp::sig::Vector &x) const
	Scale output to be used with the network. More...

virtual yarp::sig::Vector	scaleOutputFromNetFormat (const yarp::sig::Vector &x) const
	Scale back output from the network's format. More...

virtual yarp::sig::Vector	hiddenLayerFcn (const yarp::sig::Vector &x) const =0
	Hidden Layer Function. More...

virtual yarp::sig::Vector	outputLayerFcn (const yarp::sig::Vector &x) const =0
	Output Layer Function. More...

virtual yarp::sig::Vector	hiddenLayerGrad (const yarp::sig::Vector &x) const =0
	Gradient of the Hidden Layer Function. More...

virtual yarp::sig::Vector	outputLayerGrad (const yarp::sig::Vector &x) const =0
	Gradient of the Output Layer Function. More...

Protected Member Functions
void	prepare ()

void	setItem (yarp::os::Property &options, const std::string &tag, const yarp::sig::Vector &item) const

bool	getItem (const yarp::os::Property &options, const std::string &tag, yarp::sig::Vector &item) const

Protected Attributes
std::deque< yarp::sig::Vector >	IW

yarp::sig::Vector	b1

std::deque< yarp::sig::Vector >	LW

yarp::sig::Vector	b2

std::deque< minmax >	inMinMaxX

std::deque< minmax >	inMinMaxY

std::deque< minmax >	outMinMaxX

std::deque< minmax >	outMinMaxY

yarp::sig::Vector	inMinX

yarp::sig::Vector	inMinY

yarp::sig::Vector	inRatio

yarp::sig::Vector	outMinX

yarp::sig::Vector	outMinY

yarp::sig::Vector	outRatio

bool	configured

Detailed Description

Feed-Forward 2 layers Neural Network.

Useful to implement the networks trained via MATLAB (e.g. through nftool).

Definition at line 47 of file neuralNetworks.h.

Constructor & Destructor Documentation

◆ ff2LayNN() [1/2]

ff2LayNN::ff2LayNN ( )

Create an empty network.

Definition at line 26 of file neuralNetworks.cpp.

◆ ff2LayNN() [2/2]

iCub::ctrl::ff2LayNN::ff2LayNN ( const yarp::os::Property & options )

Create and configure the network.

Parameters

options contains the parameters to configure the network.

See also: configure

Member Function Documentation

◆ configure()

bool ff2LayNN::configure ( const yarp::os::Property & options )

virtual

Configure/reconfigure the network.

Parameters

options contains the parameters to configure the network according to the convention derived from MATLAB.

Returns: true/false on success/fail.

Note: Available options are:

numInputNodes <int> : input dimension.

numHiddenNodes <int> : number of nodes of the first layer (hidden).

numOutputNodes <int> : number of nodes of the second layer (output).

IW_<j> (<double> <double> ...) : jth weights vector of the hidden layer.

LW_<j> (<double> <double> ...) : jth weights vector of the output layer.

b1 (<double> <double> ...) : bias vector of the hidden layer.

b2 (<double> <double> ...) : bias vector of the output layer.

inMinMaxX_<j> (<min> <max>) : range of the jth input element.

inMinMaxY_<j> (<min> <max>) : range where to map the jth input element.

outMinMaxY_<j> (<min> <max>) : range of the jth output element.

outMinMaxX_<j> (<min> <max>) : range where to remap back the jth output element.

Note: all indexes are 0-based.

Definition at line 101 of file neuralNetworks.cpp.

◆ get_b1()

yarp::sig::Vector& iCub::ctrl::ff2LayNN::get_b1 ( )

inline

Retrieve first layer bias.

Returns: the list of first layer bias.

Definition at line 189 of file neuralNetworks.h.

◆ get_b2()

yarp::sig::Vector& iCub::ctrl::ff2LayNN::get_b2 ( )

inline

Retrieve second layer bias.

Returns: the list of second layer bias.

Definition at line 195 of file neuralNetworks.h.

◆ get_IW()

std::deque<yarp::sig::Vector>& iCub::ctrl::ff2LayNN::get_IW ( )

inline

Retrieve first layer weights.

Returns: the list of first layer weights.

Definition at line 177 of file neuralNetworks.h.

◆ get_LW()

std::deque<yarp::sig::Vector>& iCub::ctrl::ff2LayNN::get_LW ( )

inline

Retrieve second layer weights.

Returns: the list of second layer weights.

Definition at line 183 of file neuralNetworks.h.

◆ getItem()

bool ff2LayNN::getItem	(	const yarp::os::Property &	options,
		const std::string &	tag,
		yarp::sig::Vector &	item
	)		const

protected

Definition at line 58 of file neuralNetworks.cpp.

◆ getStructure()

bool ff2LayNN::getStructure ( yarp::os::Property & options ) const

virtual

Retrieve the network structure as a Property object.

Parameters

options is the output stream.

Returns: true/false on success/fail.

See also: configure

Definition at line 260 of file neuralNetworks.cpp.

◆ hiddenLayerFcn()

virtual yarp::sig::Vector iCub::ctrl::ff2LayNN::hiddenLayerFcn ( const yarp::sig::Vector & x ) const

pure virtual

Hidden Layer Function.

Parameters

x	is the input vector.

Returns: the output vector.

Implemented in iCub::ctrl::ff2LayNN_tansig_purelin.

◆ hiddenLayerGrad()

virtual yarp::sig::Vector iCub::ctrl::ff2LayNN::hiddenLayerGrad ( const yarp::sig::Vector & x ) const

pure virtual

Gradient of the Hidden Layer Function.

Parameters

x	is the input vector.

Returns: the output vector.

Implemented in iCub::ctrl::ff2LayNN_tansig_purelin.

◆ isValid()

bool ff2LayNN::isValid ( ) const

virtual

Return the internal status after a configuration.

Returns: true/false on success/fail.

Definition at line 40 of file neuralNetworks.cpp.

◆ outputLayerFcn()

virtual yarp::sig::Vector iCub::ctrl::ff2LayNN::outputLayerFcn ( const yarp::sig::Vector & x ) const

pure virtual

Output Layer Function.

Parameters

x	is the input vector.

Returns: the output vector.

Implemented in iCub::ctrl::ff2LayNN_tansig_purelin.

◆ outputLayerGrad()

virtual yarp::sig::Vector iCub::ctrl::ff2LayNN::outputLayerGrad ( const yarp::sig::Vector & x ) const

pure virtual

Gradient of the Output Layer Function.

Parameters

x	is the input vector.

Returns: the output vector.

Implemented in iCub::ctrl::ff2LayNN_tansig_purelin.

◆ predict()

Vector ff2LayNN::predict ( const yarp::sig::Vector & x ) const

virtual

Predict the output given a certain input to the network.

Parameters

x	is the actual input to the network.

Returns: the predicted output.

Definition at line 232 of file neuralNetworks.cpp.

◆ prepare()

void ff2LayNN::prepare ( )

protected

Definition at line 74 of file neuralNetworks.cpp.

◆ printStructure()

bool ff2LayNN::printStructure ( std::ostream & stream = std::cout ) const

virtual

Dump tadily the network structure on the stream.

Parameters

stream is the output stream.

Returns: true/false on success/fail.

Definition at line 329 of file neuralNetworks.cpp.

◆ scaleInputFromNetFormat()

Vector ff2LayNN::scaleInputFromNetFormat ( const yarp::sig::Vector & x ) const

virtual

Scale back input from the network's format.

Parameters

x	is the input vector.

Returns: the output vector.

Definition at line 211 of file neuralNetworks.cpp.

◆ scaleInputToNetFormat()

Vector ff2LayNN::scaleInputToNetFormat ( const yarp::sig::Vector & x ) const

virtual

Scale input to be used with the network.

Parameters

x	is the input vector.

Returns: the output vector.

Definition at line 204 of file neuralNetworks.cpp.

◆ scaleOutputFromNetFormat()

Vector ff2LayNN::scaleOutputFromNetFormat ( const yarp::sig::Vector & x ) const

virtual

Scale back output from the network's format.

Parameters

x	is the input vector.

Returns: the output vector.

Definition at line 225 of file neuralNetworks.cpp.

◆ scaleOutputToNetFormat()

Vector ff2LayNN::scaleOutputToNetFormat ( const yarp::sig::Vector & x ) const

virtual

Scale output to be used with the network.

Parameters

x	is the input vector.

Returns: the output vector.

Definition at line 218 of file neuralNetworks.cpp.

◆ setItem()

void ff2LayNN::setItem	(	yarp::os::Property &	options,
		const std::string &	tag,
		const yarp::sig::Vector &	item
	)		const

protected

Definition at line 47 of file neuralNetworks.cpp.

Member Data Documentation

◆ b1

yarp::sig::Vector iCub::ctrl::ff2LayNN::b1

protected

Definition at line 58 of file neuralNetworks.h.

◆ b2

yarp::sig::Vector iCub::ctrl::ff2LayNN::b2

protected

Definition at line 62 of file neuralNetworks.h.

◆ configured

bool iCub::ctrl::ff2LayNN::configured

protected

Definition at line 80 of file neuralNetworks.h.

◆ inMinMaxX

std::deque<minmax> iCub::ctrl::ff2LayNN::inMinMaxX

protected

Definition at line 65 of file neuralNetworks.h.

◆ inMinMaxY

std::deque<minmax> iCub::ctrl::ff2LayNN::inMinMaxY

protected

Definition at line 66 of file neuralNetworks.h.

◆ inMinX

yarp::sig::Vector iCub::ctrl::ff2LayNN::inMinX

protected

Definition at line 72 of file neuralNetworks.h.

◆ inMinY

yarp::sig::Vector iCub::ctrl::ff2LayNN::inMinY

protected

Definition at line 73 of file neuralNetworks.h.

◆ inRatio

yarp::sig::Vector iCub::ctrl::ff2LayNN::inRatio

protected

Definition at line 74 of file neuralNetworks.h.

◆ IW

std::deque<yarp::sig::Vector> iCub::ctrl::ff2LayNN::IW

protected

Definition at line 57 of file neuralNetworks.h.

◆ LW

std::deque<yarp::sig::Vector> iCub::ctrl::ff2LayNN::LW

protected

Definition at line 61 of file neuralNetworks.h.

◆ outMinMaxX

std::deque<minmax> iCub::ctrl::ff2LayNN::outMinMaxX

protected

Definition at line 69 of file neuralNetworks.h.

◆ outMinMaxY

std::deque<minmax> iCub::ctrl::ff2LayNN::outMinMaxY

protected

Definition at line 70 of file neuralNetworks.h.

◆ outMinX

yarp::sig::Vector iCub::ctrl::ff2LayNN::outMinX

protected

Definition at line 76 of file neuralNetworks.h.

◆ outMinY

yarp::sig::Vector iCub::ctrl::ff2LayNN::outMinY

protected

Definition at line 77 of file neuralNetworks.h.

◆ outRatio

yarp::sig::Vector iCub::ctrl::ff2LayNN::outRatio

protected

Definition at line 78 of file neuralNetworks.h.

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

icub-main/src/libraries/ctrlLib/include/iCub/ctrl/neuralNetworks.h
icub-main/src/libraries/ctrlLib/src/neuralNetworks.cpp

Classes

Public Member Functions

Protected Member Functions

Protected Attributes

Detailed Description

Constructor & Destructor Documentation

◆ ff2LayNN() [1/2]

◆ ff2LayNN() [2/2]

Member Function Documentation

◆ configure()

◆ get_b1()

◆ get_b2()

◆ get_IW()

◆ get_LW()

◆ getItem()

◆ getStructure()

◆ hiddenLayerFcn()

◆ hiddenLayerGrad()

◆ isValid()

◆ outputLayerFcn()

◆ outputLayerGrad()

◆ predict()

◆ prepare()

◆ printStructure()

◆ scaleInputFromNetFormat()

◆ scaleInputToNetFormat()

◆ scaleOutputFromNetFormat()

◆ scaleOutputToNetFormat()

◆ setItem()

Member Data Documentation

◆ b1

◆ b2

◆ configured

◆ inMinMaxX

◆ inMinMaxY

◆ inMinX

◆ inMinY

◆ inRatio

◆ IW

◆ LW

◆ outMinMaxX

◆ outMinMaxY

◆ outMinX

◆ outMinY

◆ outRatio