Main Page   Modules   Data Structures   File List   Data Fields   Globals   Related Pages  

gnn_msre : Mean-Square-Relative-Error Criterion.
[gnn_criterion : Basic Criterion Function.]


Detailed Description

This datatype implements the MSRE criterion, given by

This function resembles the MSE criterion, but measures the relative error with respect to the target value. Minimizing this function is equivalent to minimize

Please note that zero target values do undefine the function. This criterion isn't appropiate for small target values.

In a batch training mode, where all patterns are used to determine the gradient , then the training optimizes the MAPE criterion.

Functions

double gnn_msre_e (gnn_criterion *crit, const gsl_vector *y, const gsl_vector *t)
 The evaluation function.

int gnn_msre_dy (gnn_criterion *crit, const gsl_vector *y, const gsl_vector *t, gsl_vector *dy)
 The gradient evaluation function.

gnn_criteriongnn_msre_new (size_t size)
 Creates a new gnn_msre criterion.


Function Documentation

int gnn_msre_dy gnn_criterion   crit,
const gsl_vector *    y,
const gsl_vector *    t,
gsl_vector *    dy
 

This function implements the gnn_msre criterion's gradient evaluation function given by

Parameters:
crit  A pointer to a gnn_msre criterion.
y  A pointer to an estimation vector .
t  A pointer to the desired target vector .
dy  A pointer to a buffer vector where the result should be placed.
Returns:
Returns 0 if succeeded.

Definition at line 139 of file gnn_msre.c.

double gnn_msre_e gnn_criterion   crit,
const gsl_vector *    y,
const gsl_vector *    t
 

This function corresponds to the evaluation of the MSRE criterion.

Parameters:
crit  A pointer to a gnn_msre criterion.
y  A pointer to an estimation vector .
t  A pointer to the desired target vector .
Returns:
A real number corresponding to the value of the criterion.

Definition at line 91 of file gnn_msre.c.

gnn_criterion* gnn_msre_new size_t    size
 

This function creates a new gnn_msre of the given size.

Parameters:
size  The size of the estimation and the target vector and .
Returns:
Returns a pointer to a new gnn_msre or NULL if failed.

Definition at line 190 of file gnn_msre.c.


Generated on Sun Jun 13 20:51:43 2004 for libgnn Gradient Retropropagation Machine Library by doxygen1.2.18