Belle II Software  release-06-01-15
ECLChargedPidPDFs::VarTransfoSettings Class Reference

Class to hold parameters needed to perform pre-processing of input variables (e.g., gaussianisation, decorrelation) to build a multi-dimensional likelihood model. More...

#include <ECLChargedPidPDFs.h>

Collaboration diagram for ECLChargedPidPDFs::VarTransfoSettings:

Public Member Functions

 VarTransfoSettings ()
 Constructor.
 
 ~VarTransfoSettings ()
 Destructor.
 

Public Attributes

int nVars
 
std::string classPath
 Number of variables.
 
std::vector< int > nDivisions
 Path of the class used to get the variables transfo. More...
 
int nDivisionsMax
 Number of steps in which each variable range is sub-divided.
 
std::vector< double > cumulDist
 Maximal number of steps, across all variables.
 
std::vector< double > x
 Cumulative density function at each step.
 
std::vector< double > covMatrix
 Variable value at each step.
 
unsigned int ip
 Variables covariance matrix.
 
unsigned int jth
 p bin index
 
unsigned int gbin
 theta bin index
 

Detailed Description

Class to hold parameters needed to perform pre-processing of input variables (e.g., gaussianisation, decorrelation) to build a multi-dimensional likelihood model.

All matrices are stored in a linearised form, as vectors.

Definition at line 85 of file ECLChargedPidPDFs.h.

Member Data Documentation

◆ nDivisions

std::vector<int> nDivisions

Path of the class used to get the variables transfo.

Useful for debugging.

Definition at line 95 of file ECLChargedPidPDFs.h.


The documentation for this class was generated from the following file: