Gradient descent learning algorithm - HTML documentaiton under construction.

Inheritance:

Public Methods

	GradientDescent (double x = DEFAULT_ETA)
virtual	~GradientDescent ()
void	Update_delta_weights (unsigned n_layers, LayerNetTrain layer[] )
void	Update_weigths (unsigned n_layers, LayerNetTrain layer[])
void	Update_param (double err, unsigned iter)

Protected Methods

void	Zero_Delta_W (unsigned n_layers, LayerNetTrain layer[])

Inherited from Learning:

Public Methods

void Init_rate(double val=DEFAULT_ETA)

double Read_rate(void)

Protected Fields

static const double DEFAULT_ETA

static const double DEFAULT_DECR_FACTOR

static const double MIN_ETA

static const double DEFAULT_DECR_LIN_FACTOR

static const double DEFAULT_ALPHA

static const double DEFAULT_ETA_INCR

static const double DEFAULT_ETA_DECR

double eta

Documentation

Gradient descent learning class. It updates weights through simple gradient descent with fixed learning rate.

void Zero_Delta_W(unsigned n_layers, LayerNetTrain layer[])

GradientDescent(double x = DEFAULT_ETA)

virtual ~GradientDescent()

void Update_delta_weights(unsigned n_layers, LayerNetTrain layer[] )

void Update_weigths(unsigned n_layers, LayerNetTrain layer[])

void Update_param(double err, unsigned iter)

Direct child classes:

MomentumDescent
GradDescDecrEtaLin
GradDescDecrEtaExp
BoldDriver

Alphabetic index HTML hierarchy of classes or Java