TMVACrossValidation.C File Reference

Detailed Description

This example explains how to use the cross-validation feature of TMVA.

It is validated the Fisher algorithm with a 5-fold cross-validation.

• Project : TMVA - a Root-integrated toolkit for multivariate data analysis
• Package : TMVA
• Exectuable: TMVACrossValidation

Processing /mnt/build/workspace/root-makedoc-v608/rootspi/rdoc/src/v6-08-00-patches/tutorials/tmva/TMVACrossValidation.C...
DataSetInfo              : [dataset] : Added class "Signal"
                         : Add Tree TreeS of type Signal with 6000 events
DataSetInfo              : [dataset] : Added class "Background"
                         : Add Tree TreeB of type Background with 6000 events
                         : Dataset[dataset] : Class index : 0  name : Signal
                         : Dataset[dataset] : Class index : 1  name : Background
                         : Evaluate method: Fisher
                         : Evaluation done.
<HEADER> CrossValidation          :  ==== Results ====
                         : Fold  0 ROC-Int : 0.9121
                         : Fold  1 ROC-Int : 0.8909
                         : Fold  2 ROC-Int : 0.8928
                         : Fold  3 ROC-Int : 0.8941
                         : Fold  4 ROC-Int : 0.8840
                         : ------------------------
                         : Average ROC-Int : 0.8948
                         : Std-Dev ROC-Int : 0.0104

#include "TFile.h"
#include "TTree.h"
#include "TString.h"
#include "TSystem.h"

#include "TMVA/DataLoader.h"
#include "TMVA/CrossValidation.h"
#include "TMVA/Tools.h"

void TMVACrossValidation()
{
   // This loads the library
   TMVA::Tools::Instance();

   // Load data
   TString fname = "./tmva_class_example.root";
   if (gSystem->AccessPathName(fname))
      gSystem->Exec("curl -O http://root.cern.ch/files/tmva_class_example.root");
   TFile *input = TFile::Open(fname);

   TTree* signalTree = (TTree*)input->Get("TreeS");
   TTree* background = (TTree*)input->Get("TreeB");

   // Setup dataloader
   TMVA::DataLoader* dataloader = new TMVA::DataLoader("dataset");

   dataloader->AddSignalTree(signalTree);
   dataloader->AddBackgroundTree(background);

   dataloader->AddVariable("var1");
   dataloader->AddVariable("var2");
   dataloader->AddVariable("var3");
   dataloader->AddVariable("var4");

   dataloader->PrepareTrainingAndTestTree("", "SplitMode=Random:NormMode=NumEvents:!V");

   // Setup cross-validation with Fisher method
   TMVA::CrossValidation cv(dataloader);
   cv.BookMethod(TMVA::Types::kFisher, "Fisher", "!H:!V:Fisher");

   // Run cross-validation and print results
   cv.Evaluate();
   TMVA::CrossValidationResult results = cv.GetResults();
   results.Print();
}

int main()
{
   TMVACrossValidation();
}

Author

Stefan Wunsch

Definition in file TMVACrossValidation.C.