Definition at line 47 of file ROCCurve.h.

Public Member Functions
	ROCCurve (const std::vector< Float_t > &mvaValues, const std::vector< Bool_t > &mvaTargets, const std::vector< Float_t > &mvaWeights)

	ROCCurve (const std::vector< Float_t > &mvaValues, const std::vector< Bool_t > &mvaTargets)

	ROCCurve (const std::vector< Float_t > &mvaSignal, const std::vector< Float_t > &mvaBackground, const std::vector< Float_t > &mvaSignalWeights, const std::vector< Float_t > &mvaBackgroundWeights)

	ROCCurve (const std::vector< Float_t > &mvaSignal, const std::vector< Float_t > &mvaBackground)

	~ROCCurve ()
	destructor More...

Double_t	GetEffSForEffB (Double_t effB, const UInt_t num_points=41)
	Calculate the signal efficiency (sensitivity) for a given background efficiency (sensitivity). More...

TGraph *	GetROCCurve (const UInt_t points=100)
	Returns a new TGraph containing the ROC curve. More...

Double_t	GetROCIntegral (const UInt_t points=41)
	Calculates the ROC integral (AUC) More...

Private Member Functions
std::vector< Double_t >	ComputeSensitivity (const UInt_t num_points)

std::vector< Double_t >	ComputeSpecificity (const UInt_t num_points)

MsgLogger &	Log () const
	message logger More...

Private Attributes
TGraph *	fGraph

MsgLogger *	fLogger

std::vector< Float_t >	fMvaBackground

std::vector< Float_t >	fMvaBackgroundWeights

std::vector< Float_t >	fMvaSignal

std::vector< Float_t >	fMvaSignalWeights

#include <TMVA/ROCCurve.h>

Constructor & Destructor Documentation

◆ ROCCurve() [1/4]

TMVA::ROCCurve::ROCCurve	(	const std::vector< Float_t > &	mvaValues,
		const std::vector< Bool_t > &	mvaTargets,
		const std::vector< Float_t > &	mvaWeights
	)

Definition at line 45 of file ROCCurve.cxx.

◆ ROCCurve() [2/4]

TMVA::ROCCurve::ROCCurve	(	const std::vector< Float_t > &	mvaValues,
		const std::vector< Bool_t > &	mvaTargets
	)

Definition at line 66 of file ROCCurve.cxx.

◆ ROCCurve() [3/4]

TMVA::ROCCurve::ROCCurve	(	const std::vector< Float_t > &	mvaSignal,
		const std::vector< Float_t > &	mvaBackground,
		const std::vector< Float_t > &	mvaSignalWeights,
		const std::vector< Float_t > &	mvaBackgroundWeights
	)

Definition at line 93 of file ROCCurve.cxx.

◆ ROCCurve() [4/4]

TMVA::ROCCurve::ROCCurve	(	const std::vector< Float_t > &	mvaSignal,
		const std::vector< Float_t > &	mvaBackground
	)

Definition at line 83 of file ROCCurve.cxx.

◆ ~ROCCurve()

TMVA::ROCCurve::~ROCCurve ( )

destructor

Definition at line 107 of file ROCCurve.cxx.

Member Function Documentation

◆ ComputeSensitivity()

std::vector< Double_t > TMVA::ROCCurve::ComputeSensitivity ( const UInt_t num_points )

private

Definition at line 154 of file ROCCurve.cxx.

◆ ComputeSpecificity()

std::vector< Double_t > TMVA::ROCCurve::ComputeSpecificity ( const UInt_t num_points )

private

Definition at line 115 of file ROCCurve.cxx.

◆ GetEffSForEffB()

Double_t TMVA::ROCCurve::GetEffSForEffB	(	Double_t	effB,
		const UInt_t	num_points = `41`
	)

Calculate the signal efficiency (sensitivity) for a given background efficiency (sensitivity).

Parameters

effB	Background efficiency for which to calculate signal efficiency.
num_points	Number of points used for the underlying histogram. The number of bins will be num_points - 1.

Definition at line 199 of file ROCCurve.cxx.

◆ GetROCCurve()

TGraph * TMVA::ROCCurve::GetROCCurve ( const UInt_t num_points = 100 )

Returns a new TGraph containing the ROC curve.

Specificity is on the x-axis, sensitivity on the y-axis.

Parameters

num_points Granularity of the resulting curve. The curve will be subdivided into num_points - 1 regions where the performance of the classifier is sampled. Larger number means more accurate, but more costly, evaluation.

Definition at line 256 of file ROCCurve.cxx.

◆ GetROCIntegral()

Double_t TMVA::ROCCurve::GetROCIntegral ( const UInt_t num_points = 41 )

Calculates the ROC integral (AUC)

Parameters

num_points Granularity of the resulting curve used for integration. The curve will be subdivided into num_points - 1 regions where the performance of the classifier is sampled. Larger number means more accurate, but more costly, evaluation.

Definition at line 230 of file ROCCurve.cxx.

◆ Log()

MsgLogger& TMVA::ROCCurve::Log ( ) const

inlineprivate

message logger

Definition at line 69 of file ROCCurve.h.

Member Data Documentation

◆ fGraph

TGraph* TMVA::ROCCurve::fGraph

private

Definition at line 71 of file ROCCurve.h.

◆ fLogger

MsgLogger* TMVA::ROCCurve::fLogger

mutableprivate

Definition at line 68 of file ROCCurve.h.

◆ fMvaBackground

std::vector<Float_t> TMVA::ROCCurve::fMvaBackground

private

Definition at line 74 of file ROCCurve.h.

◆ fMvaBackgroundWeights

std::vector<Float_t> TMVA::ROCCurve::fMvaBackgroundWeights

private

Definition at line 76 of file ROCCurve.h.

◆ fMvaSignal

std::vector<Float_t> TMVA::ROCCurve::fMvaSignal

private

Definition at line 73 of file ROCCurve.h.

◆ fMvaSignalWeights

std::vector<Float_t> TMVA::ROCCurve::fMvaSignalWeights

private

Definition at line 75 of file ROCCurve.h.

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

tmva/tmva/inc/TMVA/ROCCurve.h
tmva/tmva/src/ROCCurve.cxx

Public Member Functions

Private Member Functions

Private Attributes

Constructor & Destructor Documentation

◆ ROCCurve() [1/4]

◆ ROCCurve() [2/4]

◆ ROCCurve() [3/4]

◆ ROCCurve() [4/4]

◆ ~ROCCurve()

Member Function Documentation

◆ ComputeSensitivity()

◆ ComputeSpecificity()

◆ GetEffSForEffB()

◆ GetROCCurve()

◆ GetROCIntegral()

◆ Log()

Member Data Documentation

◆ fGraph

◆ fLogger

◆ fMvaBackground

◆ fMvaBackgroundWeights

◆ fMvaSignal

◆ fMvaSignalWeights