Functions | |
param * | csom_lect (csom_t *csom, param *ptp, char *debq) |
int | csom_init (csom_t *csom) |
int | csom_calc (csom_t *csom) |
int | csom_destruct (csom_t *csom) |
int | imau1tocsomvect (float **data, int nb_attr,...) |
int | csomvecttoimau1 (csom_t *csom, int dimX, int dimY, imau1 *imres) |
int csom_calc | ( | csom_t * | csom | ) |
calculus using the Circular Self-Organizing Map nn,
the output is found in csom->csom_out,
csom | pointer to the structure of the neural network |
contors
euclidian distances
the winner
the distance from winner
STOP condition flag
a vector is a particular realization of the attribute space
distance left and right from the winner
rescaling variables
int csom_destruct | ( | csom_t * | csom | ) |
int csom_init | ( | csom_t * | csom | ) |
param* csom_lect | ( | csom_t * | csom, | |
param * | ptp, | |||
char * | debq | |||
) |
lecture des parametres de l'operateur csom->
des | pointeur descripteur | |
ptp | pointeur sur parametre courant deja alloue (a lire), | |
debq | pointeur de chaine de caracteres placee en debut de question |
int csomvecttoimau1 | ( | csom_t * | csom, | |
int | dimX, | |||
int | dimY, | |||
imau1 * | imres | |||
) |
transforms a computed csom structure into an imau1 image
csom | - pointer to a csom structure that has already been through csom_calc(..) | |
dimX | - the number of columns of the resulting image | |
dimY | - the number of rows of the resulting image | |
imres | - pointer to the resulting image |
int imau1tocsomvect | ( | float ** | data, | |
int | nb_attr, | |||
... | ||||
) |