![]() |
ROOT 6.14/05 Reference Guide |
An interface to calculate the "SeparationGain" for different separation criteria used in various training algorithms.
There are two things: the Separation Index, and the Separation Gain Separation Index: Measure of the "purity" of a sample. If all elements (events) in the sample belong to the same class (e.g. signal or background), than the separation index is 0 (meaning 100% purity (or 0% purity as it is symmetric. The index becomes maximal, for perfectly mixed samples eg. purity=50% , N_signal = N_bkg
Separation Gain: the measure of how the quality of separation of the sample increases by splitting the sample e.g. into a "left-node" and a "right-node" (N * Index_parent) - (N_left * Index_left) - (N_right * Index_right) this is then the quality criterion which is optimized for when trying to increase the information in the system (making the best selection
Definition at line 82 of file SeparationBase.h.
Public Member Functions | |
SeparationBase () | |
Constructor. More... | |
SeparationBase (const SeparationBase &s) | |
Copy constructor. More... | |
virtual | ~SeparationBase () |
const TString & | GetName () |
virtual Double_t | GetSeparationGain (const Double_t nSelS, const Double_t nSelB, const Double_t nTotS, const Double_t nTotB) |
Separation Gain: the measure of how the quality of separation of the sample increases by splitting the sample e.g. More... | |
virtual Double_t | GetSeparationIndex (const Double_t s, const Double_t b)=0 |
Protected Attributes | |
TString | fName |
Double_t | fPrecisionCut |
#include <TMVA/SeparationBase.h>
TMVA::SeparationBase::SeparationBase | ( | ) |
Constructor.
Definition at line 76 of file SeparationBase.cxx.
TMVA::SeparationBase::SeparationBase | ( | const SeparationBase & | s | ) |
Copy constructor.
Definition at line 86 of file SeparationBase.cxx.
|
inlinevirtual |
Definition at line 93 of file SeparationBase.h.
|
inline |
Definition at line 104 of file SeparationBase.h.
|
virtual |
Separation Gain: the measure of how the quality of separation of the sample increases by splitting the sample e.g.
into a "left-node" and a "right-node" (N * Index_parent) - (N_left * Index_left) - (N_right * Index_right) this is then the quality criterion which is optimized for when trying to increase the information in the system (making the best selection
Reimplemented in TMVA::SdivSqrtSplusB.
Definition at line 101 of file SeparationBase.cxx.
|
pure virtual |
Implemented in TMVA::GiniIndex, TMVA::GiniIndexWithLaplace, TMVA::SdivSqrtSplusB, TMVA::MisClassificationError, and TMVA::CrossEntropy.
|
protected |
Definition at line 108 of file SeparationBase.h.
|
protected |
Definition at line 110 of file SeparationBase.h.