---

gnn_rprop.h File Reference

#include "gnn_trainer.h"

Include dependency graph for gnn_rprop.h:

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Functions
gnn_trainer *	gnn_rprop_new (gnn_node *node, gnn_criterion *crit, gnn_dataset *data, double delta0, double deltamax)
	Creates a new RPROP trainer.
gnn_trainer *	gnn_rprop_standard_new (gnn_node *node, gnn_criterion *crit, gnn_dataset *data)
	Creates a standard RPROP trainer.
int	gnn_rprop_set_delta0 (gnn_trainer *trainer, double delta0)
	Sets .
double	gnn_rprop_get_delta0 (gnn_trainer *trainer)
	Gets .
int	gnn_rprop_set_deltamin (gnn_trainer *trainer, double deltamin)
	Sets .
double	gnn_rprop_get_deltamin (gnn_trainer *trainer)
	Gets .
int	gnn_rprop_set_deltamax (gnn_trainer *trainer, double deltamax)
	Sets .
double	gnn_rprop_get_deltamax (gnn_trainer *trainer)
	Gets .
int	gnn_rprop_set_nup (gnn_trainer *trainer, double nup)
	Sets .
double	gnn_rprop_get_nup (gnn_trainer *trainer)
int	gnn_rprop_set_num (gnn_trainer *trainer, double num)
	Sets .
double	gnn_rprop_get_num (gnn_trainer *trainer)

---

Function Documentation

double gnn_rprop_get_delta0 (  gnn_trainer *    trainer )

This function returns , the initial learning rate. Parameters: trainer  A pointer to a gnn_rprop. Returns: Returns . Definition at line 385 of file gnn_rprop.c.

double gnn_rprop_get_deltamax (  gnn_trainer *    trainer )

This function returns , the maximum learning rate. Parameters: trainer  A pointer to a gnn_rprop. Returns: Returns . Definition at line 491 of file gnn_rprop.c.

double gnn_rprop_get_deltamin (  gnn_trainer *    trainer )

This function returns , the minimum learning rate. Parameters: trainer  A pointer to a gnn_rprop. Returns: Returns . Definition at line 438 of file gnn_rprop.c.

double gnn_rprop_get_num (  gnn_trainer *    trainer )

Definition at line 595 of file gnn_rprop.c.

double gnn_rprop_get_nup (  gnn_trainer *    trainer )

Definition at line 543 of file gnn_rprop.c.

gnn_trainer* gnn_rprop_new (  gnn_node *    node, gnn_criterion *    crit, gnn_dataset *    data, double    delta0, double    deltamax )

This function creates a RPROP trainer (gnn_rprop). is the initial learning rate, and is the maximum learning rate. Parameters: node  A pointer to a gnn_node. crit  A pointer to a gnn_criterion : Basic Criterion Function.. data  A pointer to a gnn_dataset : Datasets for Training.. delta0  Initial learning rate . deltamax  Maximum learnig rate . Returns: Returns a pointer to a new gnn_rprop trainer. Definition at line 243 of file gnn_rprop.c.

int gnn_rprop_set_delta0 (  gnn_trainer *    trainer, double    delta0 )

This function sets a new value for , the initial learning rate. Its value should be stricly positive. Parameters: trainer  A pointer to a gnn_rprop. delta0  The new value for . Returns: Returns 0 if suceeded. Definition at line 356 of file gnn_rprop.c.

int gnn_rprop_set_deltamax (  gnn_trainer *    trainer, double    deltamax )

This function sets a new value for , the maximum learning rate. Its value should be stricly positive. Parameters: trainer  A pointer to a gnn_rprop. deltamax  The new value for . Returns: Returns 0 if suceeded. Definition at line 462 of file gnn_rprop.c.

int gnn_rprop_set_deltamin (  gnn_trainer *    trainer, double    deltamin )

This function sets a new value for , the minimum learning rate. Its value should be stricly positive. Parameters: trainer  A pointer to a gnn_rprop. deltamin  The new value for . Returns: Returns 0 if suceeded. Definition at line 409 of file gnn_rprop.c.

int gnn_rprop_set_num (  gnn_trainer *    trainer, double    num )

This function sets a new value for , the decrement rate. Its value should be within . Parameters: trainer  A pointer to a gnn_rprop. num  The new value for . Returns: Returns 0 if suceeded. Definition at line 567 of file gnn_rprop.c.

int gnn_rprop_set_nup (  gnn_trainer *    trainer, double    nup )

This function sets a new value for , the increment rate. Its value should be greater than 1. Parameters: trainer  A pointer to a gnn_rprop. nup  The new value for . Returns: Returns 0 if suceeded. Definition at line 515 of file gnn_rprop.c.

gnn_trainer* gnn_rprop_standard_new (  gnn_node *    node, gnn_criterion *    crit, gnn_dataset *    data )

This function creates a RPROP trainer (gnn_rprop). The initial learning rate is 0.1, and the maximum learning rate is 50. Parameters: node  A pointer to a gnn_node. crit  A pointer to a gnn_criterion : Basic Criterion Function.. data  A pointer to a gnn_dataset : Datasets for Training.. Returns: Returns a pointer to a new gnn_rprop trainer. Definition at line 334 of file gnn_rprop.c.

---