Package org.opencv.ml

Class ANN_MLP

java.lang.Object

org.opencv.core.Algorithm

org.opencv.ml.StatModel

org.opencv.ml.ANN_MLP

public class ANN_MLP
extends StatModel

Artificial Neural Networks - Multi-Layer Perceptrons. Unlike many other models in ML that are constructed and trained at once, in the MLP model these steps are separated. First, a network with the specified topology is created using the non-default constructor or the method ANN_MLP::create. All the weights are set to zeros. Then, the network is trained using a set of input and output vectors. The training procedure can be repeated more than once, that is, the weights can be adjusted based on the new training data. Additional flags for StatModel::train are available: ANN_MLP::TrainFlags. SEE: REF: ml_intro_ann

Field Summary

Fields

Modifier and Type	Field	Description
static int	ANNEAL
static int	BACKPROP
static int	GAUSSIAN
static int	IDENTITY
static int	LEAKYRELU
static int	NO_INPUT_SCALE
static int	NO_OUTPUT_SCALE
static int	RELU
static int	RPROP
static int	SIGMOID_SYM
static int	UPDATE_WEIGHTS

Fields inherited from class org.opencv.ml.StatModel

COMPRESSED_INPUT, PREPROCESSED_INPUT, RAW_OUTPUT, UPDATE_MODEL

Fields inherited from class org.opencv.core.Algorithm

nativeObj

Constructor Summary

Constructors

Modifier	Constructor	Description
protected	ANN_MLP(long addr)

Method Summary

Modifier and Type	Method	Description
static ANN_MLP	__fromPtr__(long addr)
static ANN_MLP	create()	Creates empty model Use StatModel::train to train the model, Algorithm::load<ANN_MLP>(filename) to load the pre-trained model.
protected void	finalize()
double	getAnnealCoolingRatio()	SEE: setAnnealCoolingRatio
double	getAnnealFinalT()	SEE: setAnnealFinalT
double	getAnnealInitialT()	SEE: setAnnealInitialT
int	getAnnealItePerStep()	SEE: setAnnealItePerStep
double	getBackpropMomentumScale()	SEE: setBackpropMomentumScale
double	getBackpropWeightScale()	SEE: setBackpropWeightScale
Mat	getLayerSizes()	Integer vector specifying the number of neurons in each layer including the input and output layers.
double	getRpropDW0()	SEE: setRpropDW0
double	getRpropDWMax()	SEE: setRpropDWMax
double	getRpropDWMin()	SEE: setRpropDWMin
double	getRpropDWMinus()	SEE: setRpropDWMinus
double	getRpropDWPlus()	SEE: setRpropDWPlus
TermCriteria	getTermCriteria()	SEE: setTermCriteria
int	getTrainMethod()	Returns current training method
Mat	getWeights(int layerIdx)
static ANN_MLP	load(String filepath)	Loads and creates a serialized ANN from a file Use ANN::save to serialize and store an ANN to disk.
void	setActivationFunction(int type)	Initialize the activation function for each neuron.
void	setActivationFunction(int type, double param1)	Initialize the activation function for each neuron.
void	setActivationFunction(int type, double param1, double param2)	Initialize the activation function for each neuron.
void	setAnnealCoolingRatio(double val)	getAnnealCoolingRatio SEE: getAnnealCoolingRatio
void	setAnnealFinalT(double val)	getAnnealFinalT SEE: getAnnealFinalT
void	setAnnealInitialT(double val)	getAnnealInitialT SEE: getAnnealInitialT
void	setAnnealItePerStep(int val)	getAnnealItePerStep SEE: getAnnealItePerStep
void	setBackpropMomentumScale(double val)	getBackpropMomentumScale SEE: getBackpropMomentumScale
void	setBackpropWeightScale(double val)	getBackpropWeightScale SEE: getBackpropWeightScale
void	setLayerSizes(Mat _layer_sizes)	Integer vector specifying the number of neurons in each layer including the input and output layers.
void	setRpropDW0(double val)	getRpropDW0 SEE: getRpropDW0
void	setRpropDWMax(double val)	getRpropDWMax SEE: getRpropDWMax
void	setRpropDWMin(double val)	getRpropDWMin SEE: getRpropDWMin
void	setRpropDWMinus(double val)	getRpropDWMinus SEE: getRpropDWMinus
void	setRpropDWPlus(double val)	getRpropDWPlus SEE: getRpropDWPlus
void	setTermCriteria(TermCriteria val)	getTermCriteria SEE: getTermCriteria
void	setTrainMethod(int method)	Sets training method and common parameters.
void	setTrainMethod(int method, double param1)	Sets training method and common parameters.
void	setTrainMethod(int method, double param1, double param2)	Sets training method and common parameters.

Methods inherited from class org.opencv.ml.StatModel

calcError, empty, getVarCount, isClassifier, isTrained, predict, predict, predict, train, train, train

Methods inherited from class org.opencv.core.Algorithm

clear, getDefaultName, getNativeObjAddr, save

Methods inherited from class java.lang.Object

clone, equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Details

IDENTITY

public static final int IDENTITY

See Also:

Constant Field Values

SIGMOID_SYM

public static final int SIGMOID_SYM

See Also:

Constant Field Values

GAUSSIAN

public static final int GAUSSIAN

See Also:

Constant Field Values

RELU

public static final int RELU

See Also:

Constant Field Values

LEAKYRELU

public static final int LEAKYRELU

See Also:

Constant Field Values

UPDATE_WEIGHTS

public static final int UPDATE_WEIGHTS

See Also:

Constant Field Values

NO_INPUT_SCALE

public static final int NO_INPUT_SCALE

See Also:

Constant Field Values

NO_OUTPUT_SCALE

public static final int NO_OUTPUT_SCALE

See Also:

Constant Field Values

BACKPROP

public static final int BACKPROP

See Also:

Constant Field Values

RPROP

public static final int RPROP

See Also:

Constant Field Values

ANNEAL

public static final int ANNEAL

See Also:

Constant Field Values

Constructor Details

ANN_MLP

protected ANN_MLP(long addr)

Method Details

__fromPtr__

public static ANN_MLP __fromPtr__(long addr)

setTrainMethod

public void setTrainMethod(int method, double param1, double param2)

Sets training method and common parameters.

Parameters:

method - Default value is ANN_MLP::RPROP. See ANN_MLP::TrainingMethods.

param1 - passed to setRpropDW0 for ANN_MLP::RPROP and to setBackpropWeightScale for ANN_MLP::BACKPROP and to initialT for ANN_MLP::ANNEAL.

param2 - passed to setRpropDWMin for ANN_MLP::RPROP and to setBackpropMomentumScale for ANN_MLP::BACKPROP and to finalT for ANN_MLP::ANNEAL.

setTrainMethod

public void setTrainMethod(int method, double param1)

Sets training method and common parameters.

Parameters:

method - Default value is ANN_MLP::RPROP. See ANN_MLP::TrainingMethods.

param1 - passed to setRpropDW0 for ANN_MLP::RPROP and to setBackpropWeightScale for ANN_MLP::BACKPROP and to initialT for ANN_MLP::ANNEAL.

setTrainMethod

public void setTrainMethod(int method)

Sets training method and common parameters.

Parameters:

method - Default value is ANN_MLP::RPROP. See ANN_MLP::TrainingMethods.

getTrainMethod

public int getTrainMethod()

Returns current training method

Returns:

automatically generated

setActivationFunction

public void setActivationFunction(int type, double param1, double param2)

Initialize the activation function for each neuron. Currently the default and the only fully supported activation function is ANN_MLP::SIGMOID_SYM.

Parameters:

type - The type of activation function. See ANN_MLP::ActivationFunctions.

param1 - The first parameter of the activation function, \(\alpha\). Default value is 0.

param2 - The second parameter of the activation function, \(\beta\). Default value is 0.

setActivationFunction

public void setActivationFunction(int type, double param1)

Initialize the activation function for each neuron. Currently the default and the only fully supported activation function is ANN_MLP::SIGMOID_SYM.

Parameters:

type - The type of activation function. See ANN_MLP::ActivationFunctions.

param1 - The first parameter of the activation function, \(\alpha\). Default value is 0.

setActivationFunction

public void setActivationFunction(int type)

Initialize the activation function for each neuron. Currently the default and the only fully supported activation function is ANN_MLP::SIGMOID_SYM.

Parameters:

type - The type of activation function. See ANN_MLP::ActivationFunctions.

setLayerSizes

public void setLayerSizes(Mat _layer_sizes)

Integer vector specifying the number of neurons in each layer including the input and output layers. The very first element specifies the number of elements in the input layer. The last element - number of elements in the output layer. Default value is empty Mat. SEE: getLayerSizes

Parameters:

_layer_sizes - automatically generated

getLayerSizes

public Mat getLayerSizes()

Integer vector specifying the number of neurons in each layer including the input and output layers. The very first element specifies the number of elements in the input layer. The last element - number of elements in the output layer. SEE: setLayerSizes

Returns:

automatically generated

getTermCriteria

public TermCriteria getTermCriteria()

SEE: setTermCriteria

Returns:

automatically generated

setTermCriteria

public void setTermCriteria(TermCriteria val)

getTermCriteria SEE: getTermCriteria

Parameters:

val - automatically generated

getBackpropWeightScale

public double getBackpropWeightScale()

SEE: setBackpropWeightScale

Returns:

automatically generated

setBackpropWeightScale

public void setBackpropWeightScale(double val)

getBackpropWeightScale SEE: getBackpropWeightScale

Parameters:

val - automatically generated

getBackpropMomentumScale

public double getBackpropMomentumScale()

SEE: setBackpropMomentumScale

Returns:

automatically generated

setBackpropMomentumScale

public void setBackpropMomentumScale(double val)

getBackpropMomentumScale SEE: getBackpropMomentumScale

Parameters:

val - automatically generated

getRpropDW0

public double getRpropDW0()

SEE: setRpropDW0

Returns:

automatically generated

setRpropDW0

public void setRpropDW0(double val)

getRpropDW0 SEE: getRpropDW0

Parameters:

val - automatically generated

getRpropDWPlus

public double getRpropDWPlus()

SEE: setRpropDWPlus

Returns:

automatically generated

setRpropDWPlus

public void setRpropDWPlus(double val)

getRpropDWPlus SEE: getRpropDWPlus

Parameters:

val - automatically generated

getRpropDWMinus

public double getRpropDWMinus()

SEE: setRpropDWMinus

Returns:

automatically generated

setRpropDWMinus

public void setRpropDWMinus(double val)

getRpropDWMinus SEE: getRpropDWMinus

Parameters:

val - automatically generated

getRpropDWMin

public double getRpropDWMin()

SEE: setRpropDWMin

Returns:

automatically generated

setRpropDWMin

public void setRpropDWMin(double val)

getRpropDWMin SEE: getRpropDWMin

Parameters:

val - automatically generated

getRpropDWMax

public double getRpropDWMax()

SEE: setRpropDWMax

Returns:

automatically generated

setRpropDWMax

public void setRpropDWMax(double val)

getRpropDWMax SEE: getRpropDWMax

Parameters:

val - automatically generated

getAnnealInitialT

public double getAnnealInitialT()

SEE: setAnnealInitialT

Returns:

automatically generated

setAnnealInitialT

public void setAnnealInitialT(double val)

getAnnealInitialT SEE: getAnnealInitialT

Parameters:

val - automatically generated

getAnnealFinalT

public double getAnnealFinalT()

SEE: setAnnealFinalT

Returns:

automatically generated

setAnnealFinalT

public void setAnnealFinalT(double val)

getAnnealFinalT SEE: getAnnealFinalT

Parameters:

val - automatically generated

getAnnealCoolingRatio

public double getAnnealCoolingRatio()

SEE: setAnnealCoolingRatio

Returns:

automatically generated

setAnnealCoolingRatio

public void setAnnealCoolingRatio(double val)

getAnnealCoolingRatio SEE: getAnnealCoolingRatio

Parameters:

val - automatically generated

getAnnealItePerStep

public int getAnnealItePerStep()

SEE: setAnnealItePerStep

Returns:

automatically generated

setAnnealItePerStep

public void setAnnealItePerStep(int val)

getAnnealItePerStep SEE: getAnnealItePerStep

Parameters:

val - automatically generated

getWeights

public Mat getWeights(int layerIdx)

create

public static ANN_MLP create()

Creates empty model Use StatModel::train to train the model, Algorithm::load<ANN_MLP>(filename) to load the pre-trained model. Note that the train method has optional flags: ANN_MLP::TrainFlags.

Returns:

automatically generated

load

public static ANN_MLP load(String filepath)

Loads and creates a serialized ANN from a file Use ANN::save to serialize and store an ANN to disk. Load the ANN from this file again, by calling this function with the path to the file.

Parameters:

filepath - path to serialized ANN

Returns:

automatically generated

finalize

protected void finalize() throws Throwable

Overrides:

finalize in class StatModel

Throws:

Throwable