Learning and Pruning Functions

krui_err krui_setLearnFunc( char *learning_function)
krui_err krui_setFFLearnFunc( char *FF_learning_function)
krui_err krui_setPrunnFunc( char *pruning_function)
selects the learning / subordinate learning / pruning function; returns an error code if the given function is unknown. char *krui_getLearnFunc( void )
char *krui_getFFLearnFunc( void )
char *krui_getPrunFunc( void )
returns the name of the current learning / subordinate learning / pruning function. The default learning / subordinate learning function is `Std_Backpropagation', the default pruning function is `MagPruning' (see also kr_def.h).

krui_err krui_learnAllPatterns( float *parameterInArray,
int NoOfInParams,
float * *parameterOutArray,
int *NoOfOutParams )
krui_err krui_learnAllPatternsFF( float *parameterInArray,
int NoOfInParams,
float * *parameterOutArray,
int *NoOfOutParams )
learns all patterns (each consisting of an input/output pair) using the current learning / subordinate learning function. parameterInArray contains the learning parameter(s). NoOfInParams stores the number of learning parameters. parameterOutArray returns the results of the learning function. This array is a static array defined in the learning function. *NoOfOutParams points to an integer value that contains the number of output parameters from the current learning function. The function returns an error code if memory allocation has failed or if the learning has failed. Patterns must be loaded before calling this function.

krui_err krui_learnSinglePattern( int pattern_no,
float *parameterInArray,
int NoOfInParams,
float * *parameterOutArray,
int *NoOfOutParams )
krui_err krui_learnSinglePatternFF( int pattern_no,
float *parameterInArray,
int NoOfInParams,
float * *parameterOutArray,
int *NoOfOutParams )
same as krui_learnAllPatterns( ... ) or krui_learnAllPatternsFF( ... ) respectively, but teaches only the current pattern.