LWPR_Workspace is a structure that contains enough "working memory" for the LWPR computational routines. More...
#include <lwpr_aux.h>
Data Fields | |
int * | derivOk |
Used within lwpr_aux_update_distance_metric for storing which PLS directions can be trusted. | |
double * | storage |
Pointer to the allocated memory. | |
double * | dx |
Used to hold the difference between a normalised input vector and a RF's centre. | |
double * | dwdM |
Derivatives of the weight w with respect to LWPR_ReceptiveField.M. | |
double * | dJ2dM |
Derivatives of the cost J2 with respect to M. | |
double * | ddwdMdM |
2nd derivatives of w wrt. M | |
double * | ddJ2dMdM |
2nd derivatives of J2 wrt. M | |
double * | Ps |
Intermediate results used within lwpr_aux_update_distance_metric. | |
double * | Pse |
Intermediate results used within lwpr_aux_update_distance_metric. | |
double * | xu |
Used within PLS calculations (updated x). | |
double * | yres |
Intermediate results used within lwpr_aux_update_regression. | |
double * | ytarget |
Intermediate results used within lwpr_aux_update_regression. | |
double * | xres |
Intermediate results used within lwpr_aux_update_regression. | |
double * | xc |
Used to hold the difference between a normalised input vector and a RF's centre. | |
double * | xmz |
Updated mean of a RF. | |
double * | e_cv |
Intermediate results used within lwpr_aux_update_regression. | |
double * | s |
Intermediate results used within lwpr_aux_update_regression. | |
double * | dsdx |
Intermediate results used within lwpr_aux_predict_one_J. | |
double * | Dx |
Used to store RF.D * (x-RF.c). | |
double * | sum_dwdx |
Intermediate results used within lwpr_aux_predict_one_J. | |
double * | sum_ydwdx_wdydx |
Intermediate results used within lwpr_aux_predict_one_J. | |
double * | sum_ddwdxdx |
Intermediate results used within lwpr_aux_predict_one_gH. | |
double * | sum_ddRdxdx |
Intermediate results used within lwpr_aux_predict_one_gH. |
LWPR_Workspace is a structure that contains enough "working memory" for the LWPR computational routines.
It is automatically allocated within lwpr_init_model, with one LWPR_Workspace structure per thread. You should not have to handle any of its elements yourself.