1#ifndef ROOT_TMVA_ROCCalc 
    2#define ROOT_TMVA_ROCCalc 
1-D histogram with a double per channel (see TH1 documentation)
 
TH1 is the base class of all histogram classes in ROOT.
 
2-D histogram with a float per channel (see TH1 documentation)
 
Service class for 2-D histogram classes.
 
ostringstream derivative to redirect and format output
 
Double_t fSignalCut
MVA cut value for last demanded background rejection or signal efficiency.
 
Double_t Root(Double_t)
Root finding using Brents algorithm; taken from CERNLIB function RZERO.
 
void ApplySignalAndBackgroundStyle(TH1 *sig, TH1 *bkg, TH1 *any=nullptr)
 
Double_t GetROCIntegral()
code to compute the area under the ROC ( rej-vs-eff ) curve
 
Float_t fXmax
min and max of the mva distribution
 
TSpline * fSplmvaCumB
spline of cumulated mva distributions
 
Int_t fMaxIter
maximum number of iterations
 
Bool_t fStatus
false if is found some error in mvaS or mvaB
 
Double_t GetSignalReferenceCut()
 
Int_t fCutOrientation
+1 if larger mva value means more signal like, -1 otherwise
 
TH1 * GetPurity(Int_t nStot, Int_t nBtot)
 
Double_t GetEffSForEffBof(Double_t effBref, Double_t &effSerr)
get the signal efficiency for a particular background efficiency that will be the value of the effici...
 
ROCCalc(TH1 *mvaS, TH1 *mvaB)
 
TH1 * GetSignificance(Int_t nStot, Int_t nBtot)
 
MsgLogger & Log() const
message logger
 
TH1D * GetROC()
get the ROC curve
 
TH1 * fmvaB
the input mva distributions
 
Double_t fAbsTol
absolute tolerance deviation
 
TH1 * fmvaBpdf
the normalized (and rebinned) input mva distributions
 
Double_t GetEffForRoot(Double_t theCut)
returns efficiency as function of cut
 
Double_t fNevtS
number of signal events (used in error calculation)
 
Base class for spline implementation containing the Draw/Paint methods.
 
A TTree represents a columnar dataset.
 
create variable transformations