BALL  1.4.2
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Properties Friends Macros Groups Pages
Public Member Functions | Private Member Functions | Private Attributes | List of all members
BALL::FermiBaseFunction Class Reference

#include <BALL/SCORING/COMMON/fermiBaseFunction.h>

Inheritance diagram for BALL::FermiBaseFunction:
BALL::ScoringBaseFunction

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_
 

Detailed Description

Sigmoidal base function

Definition at line 13 of file fermiBaseFunction.h.

Constructor & Destructor Documentation

BALL::FermiBaseFunction::FermiBaseFunction ( )
BALL::FermiBaseFunction::FermiBaseFunction ( float  lower,
float  upper 
)

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.

Member Function Documentation

float BALL::FermiBaseFunction::calculate ( float  x) const
virtual

Reimplemented from BALL::ScoringBaseFunction.

float BALL::FermiBaseFunction::calculate ( float  x,
float  lower,
float  upper 
)
virtual

Reimplemented from BALL::ScoringBaseFunction.

void BALL::FermiBaseFunction::computeFermiParameters_ ( )
private
void BALL::FermiBaseFunction::setLower ( float  lower)
virtual

Reimplemented from BALL::ScoringBaseFunction.

void BALL::FermiBaseFunction::setUpper ( float  upper)
virtual

Reimplemented from BALL::ScoringBaseFunction.

Member Data Documentation

float BALL::FermiBaseFunction::a_
private

Definition at line 54 of file fermiBaseFunction.h.

float BALL::FermiBaseFunction::b_
private

Definition at line 57 of file fermiBaseFunction.h.