BALL
1.4.2
|
#include <BALL/SCORING/COMMON/fermiBaseFunction.h>
Public Member Functions | |
FermiBaseFunction () | |
FermiBaseFunction (float lower, float upper) | |
void | setLower (float lower) |
void | setUpper (float upper) |
float | calculate (float x) const |
float | calculate (float x, float lower, float upper) |
Public Member Functions inherited from BALL::ScoringBaseFunction | |
ScoringBaseFunction () | |
ScoringBaseFunction (float lower, float upper) | |
virtual | ~ScoringBaseFunction () |
Private Member Functions | |
void | computeFermiParameters_ () |
Private Attributes | |
float | a_ |
float | b_ |
Additional Inherited Members | |
Public Types inherited from BALL::ScoringBaseFunction | |
enum | BaseFunctionType { BASE_FUNCTION_TYPE__LINEAR, BASE_FUNCTION_TYPE__FERMI } |
Protected Attributes inherited from BALL::ScoringBaseFunction | |
float | lower_ |
float | upper_ |
Sigmoidal base function
Definition at line 13 of file fermiBaseFunction.h.
BALL::FermiBaseFunction::FermiBaseFunction | ( | ) |
The function is defined in such a way that the score approaches 1 for values < lower and 0 for values > upper. If you need contrary behaviour, you have to supply a "lower" which is greater than "upper". This will invert the function.
Reimplemented from BALL::ScoringBaseFunction.
Reimplemented from BALL::ScoringBaseFunction.
|
private |
|
virtual |
Reimplemented from BALL::ScoringBaseFunction.
|
virtual |
Reimplemented from BALL::ScoringBaseFunction.
|
private |
Definition at line 54 of file fermiBaseFunction.h.
|
private |
Definition at line 57 of file fermiBaseFunction.h.