| 1 |
bortigno |
1.1 |
// @(#)root/tmva $Id: TMVAClassification.cxx 37399 2010-12-08 15:22:07Z evt $
|
| 2 |
|
|
/**********************************************************************************
|
| 3 |
|
|
* Project : TMVA - a ROOT-integrated toolkit for multivariate data analysis *
|
| 4 |
|
|
* Package : TMVA *
|
| 5 |
|
|
* Exectuable: TMVAClassification *
|
| 6 |
|
|
* *
|
| 7 |
|
|
* This executable provides examples for the training and testing of the *
|
| 8 |
|
|
* TMVA classifiers. *
|
| 9 |
|
|
* *
|
| 10 |
|
|
* As input data is used a toy-MC sample consisting of four Gaussian-distributed *
|
| 11 |
|
|
* and linearly correlated input variables. *
|
| 12 |
|
|
* *
|
| 13 |
|
|
* The methods to be used can be switched on and off by means of booleans. *
|
| 14 |
|
|
* *
|
| 15 |
|
|
* Compile and run the example with the following commands *
|
| 16 |
|
|
* *
|
| 17 |
|
|
* make *
|
| 18 |
|
|
* ./TMVAClassification <Methods> *
|
| 19 |
|
|
* *
|
| 20 |
|
|
* where: <Methods> = "method1 method2" *
|
| 21 |
|
|
* are the TMVA classifier names *
|
| 22 |
|
|
* *
|
| 23 |
|
|
* example: *
|
| 24 |
|
|
* ./TMVAClassification Fisher LikelihoodPCA BDT *
|
| 25 |
|
|
* *
|
| 26 |
|
|
* If no method given, a default set is of classifiers is used *
|
| 27 |
|
|
* *
|
| 28 |
|
|
* The output file "TMVA.root" can be analysed with the use of dedicated *
|
| 29 |
|
|
* macros (simply say: root -l <../macros/macro.C>), which can be conveniently *
|
| 30 |
|
|
* invoked through a GUI launched by the command *
|
| 31 |
|
|
* *
|
| 32 |
|
|
* root -l ./TMVAGui.C *
|
| 33 |
|
|
**********************************************************************************/
|
| 34 |
|
|
|
| 35 |
|
|
#include <cstdlib>
|
| 36 |
|
|
#include <iostream>
|
| 37 |
|
|
#include <map>
|
| 38 |
|
|
#include <string>
|
| 39 |
|
|
|
| 40 |
|
|
#include "TChain.h"
|
| 41 |
|
|
#include "TFile.h"
|
| 42 |
|
|
#include "TTree.h"
|
| 43 |
|
|
#include "TString.h"
|
| 44 |
|
|
#include "TObjString.h"
|
| 45 |
|
|
#include "TSystem.h"
|
| 46 |
|
|
#include "TROOT.h"
|
| 47 |
|
|
|
| 48 |
|
|
#include "TMVA/Factory.h"
|
| 49 |
|
|
#include "TMVA/Tools.h"
|
| 50 |
|
|
#include "TMVA/Config.h"
|
| 51 |
|
|
|
| 52 |
|
|
#include "../macros/samples.h"
|
| 53 |
|
|
|
| 54 |
|
|
// read input data file with ascii format (otherwise ROOT) ?
|
| 55 |
|
|
Bool_t ReadDataFromAsciiIFormat = kFALSE;
|
| 56 |
|
|
|
| 57 |
|
|
int main( int argc, char** argv )
|
| 58 |
|
|
{
|
| 59 |
|
|
//---------------------------------------------------------------
|
| 60 |
|
|
// Default MVA methods to be trained + tested
|
| 61 |
|
|
std::map<std::string,int> Use;
|
| 62 |
|
|
|
| 63 |
|
|
// --- Cut optimisation
|
| 64 |
|
|
Use["Cuts"] = 1;
|
| 65 |
|
|
Use["CutsD"] = 0;
|
| 66 |
|
|
Use["CutsPCA"] = 0;
|
| 67 |
|
|
Use["CutsGA"] = 0;
|
| 68 |
|
|
Use["CutsSA"] = 0;
|
| 69 |
|
|
//
|
| 70 |
|
|
// --- 1-dimensional likelihood ("naive Bayes estimator")
|
| 71 |
|
|
Use["Likelihood"] = 1;
|
| 72 |
|
|
Use["LikelihoodD"] = 0; // the "D" extension indicates decorrelated input variables (see option strings)
|
| 73 |
|
|
Use["LikelihoodPCA"] = 0; // the "PCA" extension indicates PCA-transformed input variables (see option strings)
|
| 74 |
|
|
Use["LikelihoodKDE"] = 0;
|
| 75 |
|
|
Use["LikelihoodMIX"] = 0;
|
| 76 |
|
|
//
|
| 77 |
|
|
// --- Mutidimensional likelihood and Nearest-Neighbour methods
|
| 78 |
|
|
Use["PDERS"] = 0;
|
| 79 |
|
|
Use["PDERSD"] = 0;
|
| 80 |
|
|
Use["PDERSPCA"] = 0;
|
| 81 |
|
|
Use["PDEFoam"] = 0;
|
| 82 |
|
|
Use["PDEFoamBoost"] = 0; // uses generalised MVA method boosting
|
| 83 |
|
|
Use["KNN"] = 0; // k-nearest neighbour method
|
| 84 |
|
|
//
|
| 85 |
|
|
// --- Linear Discriminant Analysis
|
| 86 |
|
|
Use["LD"] = 0; // Linear Discriminant identical to Fisher
|
| 87 |
|
|
Use["Fisher"] = 0;
|
| 88 |
|
|
Use["FisherG"] = 0;
|
| 89 |
|
|
Use["BoostedFisher"] = 0; // uses generalised MVA method boosting
|
| 90 |
|
|
Use["HMatrix"] = 0;
|
| 91 |
|
|
//
|
| 92 |
|
|
// --- Function Discriminant analysis
|
| 93 |
|
|
Use["FDA_GA"] = 0; // minimisation of user-defined function using Genetics Algorithm
|
| 94 |
|
|
Use["FDA_SA"] = 0;
|
| 95 |
|
|
Use["FDA_MC"] = 0;
|
| 96 |
|
|
Use["FDA_MT"] = 0;
|
| 97 |
|
|
Use["FDA_GAMT"] = 0;
|
| 98 |
|
|
Use["FDA_MCMT"] = 0;
|
| 99 |
|
|
//
|
| 100 |
|
|
// --- Neural Networks (all are feed-forward Multilayer Perceptrons)
|
| 101 |
|
|
Use["MLP"] = 0; // Recommended ANN
|
| 102 |
|
|
Use["MLPBFGS"] = 0; // Recommended ANN with optional training method
|
| 103 |
|
|
Use["MLPBNN"] = 0; // Recommended ANN with BFGS training method and bayesian regulator
|
| 104 |
|
|
Use["CFMlpANN"] = 0; // Depreciated ANN from ALEPH
|
| 105 |
|
|
Use["TMlpANN"] = 0; // ROOT's own ANN
|
| 106 |
|
|
//
|
| 107 |
|
|
// --- Support Vector Machine
|
| 108 |
|
|
Use["SVM"] = 0;
|
| 109 |
|
|
//
|
| 110 |
|
|
// --- Boosted Decision Trees
|
| 111 |
|
|
Use["BDT"] = 1; // uses Adaptive Boost
|
| 112 |
|
|
Use["BDTG"] = 0; // uses Gradient Boost
|
| 113 |
|
|
Use["BDTB"] = 0; // uses Bagging
|
| 114 |
|
|
Use["BDTD"] = 0; // decorrelation + Adaptive Boost
|
| 115 |
|
|
//
|
| 116 |
|
|
// --- Friedman's RuleFit method, ie, an optimised series of cuts ("rules")
|
| 117 |
|
|
Use["RuleFit"] = 0;
|
| 118 |
|
|
// ---------------------------------------------------------------
|
| 119 |
|
|
|
| 120 |
|
|
std::cout << std::endl << "==> Start TMVAClassification" << std::endl;
|
| 121 |
|
|
|
| 122 |
|
|
bool batchMode(false);
|
| 123 |
|
|
bool useDefaultMethods(true);
|
| 124 |
|
|
|
| 125 |
|
|
// // Select methods (don't look at this code - not of interest)
|
| 126 |
|
|
// for (int i=1; i<argc; i++) {
|
| 127 |
|
|
// std::string regMethod(argv[i]);
|
| 128 |
|
|
// if(regMethod=="-b" || regMethod=="--batch") {
|
| 129 |
|
|
// batchMode=true;
|
| 130 |
|
|
// continue;
|
| 131 |
|
|
// }
|
| 132 |
|
|
// if (Use.find(regMethod) == Use.end()) {
|
| 133 |
|
|
// std::cout << "Method \"" << regMethod << "\" not known in TMVA under this name. Choose among the following:" << std::endl;
|
| 134 |
|
|
// for (std::map<std::string,int>::iterator it = Use.begin(); it != Use.end(); it++) std::cout << it->first << " ";
|
| 135 |
|
|
// std::cout << std::endl;
|
| 136 |
|
|
// return 1;
|
| 137 |
|
|
// }
|
| 138 |
|
|
// useDefaultMethods = false;
|
| 139 |
|
|
// }
|
| 140 |
|
|
|
| 141 |
|
|
// if (!useDefaultMethods) {
|
| 142 |
|
|
// for (std::map<std::string,int>::iterator it = Use.begin(); it != Use.end(); it++) it->second = 0;
|
| 143 |
|
|
// for (int i=1; i<argc; i++) {
|
| 144 |
|
|
// std::string regMethod(argv[i]);
|
| 145 |
|
|
// if(regMethod=="-b" || regMethod=="--batch") continue;
|
| 146 |
|
|
// Use[regMethod] = 1;
|
| 147 |
|
|
// }
|
| 148 |
|
|
// }
|
| 149 |
|
|
|
| 150 |
|
|
// --------------------------------------------------------------------------------------------------
|
| 151 |
|
|
|
| 152 |
|
|
// --- Here the preparation phase begins
|
| 153 |
|
|
// string channel="Zmm";
|
| 154 |
|
|
|
| 155 |
|
|
// std::cout << "argc = " << argc << std::endl;
|
| 156 |
|
|
std::string channel;
|
| 157 |
|
|
if( argc > 1 )
|
| 158 |
|
|
channel=argv[1];
|
| 159 |
|
|
else{
|
| 160 |
|
|
std::cerr << "no channel selected" << std::endl;
|
| 161 |
|
|
return -1;
|
| 162 |
|
|
}
|
| 163 |
|
|
// Create a ROOT output file where TMVA will store ntuples, histograms, etc.
|
| 164 |
|
|
TString outfileName( "TMVA.root" );
|
| 165 |
|
|
outfileName=channel+outfileName;
|
| 166 |
|
|
TFile* outputFile = TFile::Open( outfileName, "RECREATE" );
|
| 167 |
|
|
|
| 168 |
|
|
// Create the factory object. Later you can choose the methods
|
| 169 |
|
|
// whose performance you'd like to investigate. The factory is
|
| 170 |
|
|
// the only TMVA object you have to interact with
|
| 171 |
|
|
//
|
| 172 |
|
|
// The first argument is the base of the name of all the
|
| 173 |
|
|
// weightfiles in the directory weight/
|
| 174 |
|
|
//
|
| 175 |
|
|
// The second argument is the output file for the training results
|
| 176 |
|
|
// All TMVA output can be suppressed by removing the "!" (not) in
|
| 177 |
|
|
// front of the "Silent" argument in the option string
|
| 178 |
|
|
TMVA::Factory *factory = new TMVA::Factory( "TMVAClassification", outputFile,
|
| 179 |
|
|
"!V:!Silent:Color:DrawProgressBar:Transformations=I;D;P;G,D:AnalysisType=Classification" );
|
| 180 |
|
|
|
| 181 |
|
|
// If you wish to modify default settings
|
| 182 |
|
|
// (please check "src/Config.h" to see all available global options)
|
| 183 |
|
|
// (TMVA::gConfig().GetVariablePlotting()).fTimesRMS = 8.0;
|
| 184 |
|
|
// (TMVA::gConfig().GetIONames()).fWeightFileDir = "myWeightDirectory";
|
| 185 |
|
|
(TMVA::gConfig().GetIONames()).fWeightFileExtension = channel+"_weight";
|
| 186 |
|
|
|
| 187 |
|
|
// Define the input variables that shall be used for the MVA training
|
| 188 |
|
|
// note that you may also use variable expressions, such as: "3*var1/var2*abs(var3)"
|
| 189 |
|
|
// [all types of expressions that can also be parsed by TTree::Draw( "expression" )]
|
| 190 |
|
|
|
| 191 |
|
|
float BR115 = 0.704;
|
| 192 |
|
|
float BR120 = 0.648;
|
| 193 |
|
|
float BR125 = 0.577;
|
| 194 |
|
|
float BR130 = 0.493;
|
| 195 |
|
|
float BR135 = 0.403;
|
| 196 |
|
|
|
| 197 |
|
|
// 115
|
| 198 |
|
|
float xSecWH = 0.7546*BR115;
|
| 199 |
|
|
float xSecZH = 0.4107*BR115;
|
| 200 |
|
|
float backgroundXsecWJets = 31314;
|
| 201 |
|
|
float backgroundXsecDY = 3048;
|
| 202 |
|
|
float backgroundXsecWZ = 18.3;
|
| 203 |
|
|
float backgroundXsecWW = 42.9;
|
| 204 |
|
|
float backgroundXsecZZ = 5.9;
|
| 205 |
|
|
float backgroundXsecTT = 165;
|
| 206 |
|
|
float backgroundXsecT_tchannel = 41.92;
|
| 207 |
|
|
float backgroundXsecT_tWDRchannel = 7.87;
|
| 208 |
|
|
float backgroundXsecTbar_tchannel = 22.65;
|
| 209 |
|
|
float backgroundXsecTbar_tWDRchannel = 7.87;
|
| 210 |
|
|
// float backgroundXsecTbar_tWDSchannel = 25;
|
| 211 |
|
|
|
| 212 |
|
|
//get Lumi from something serius
|
| 213 |
|
|
float lumi = 2047.0+8872.0;
|
| 214 |
|
|
|
| 215 |
|
|
float signalXsec;
|
| 216 |
|
|
std::vector<double> backgroundXsec;
|
| 217 |
|
|
std::vector<TFile*> myInputFile_signal;
|
| 218 |
|
|
std::vector<TFile*> myInputFile_background;
|
| 219 |
|
|
|
| 220 |
|
|
TCut mycuts = "";
|
| 221 |
|
|
TCut mycutb = "";
|
| 222 |
|
|
bool origCuts = true;
|
| 223 |
|
|
|
| 224 |
|
|
|
| 225 |
|
|
//signal for Z
|
| 226 |
|
|
if( channel == "Zmm" or channel == "Zee" or channel == "Znn" ){
|
| 227 |
|
|
myInputFile_signal.push_back(TFile::Open("MC_files/file_ZH_ZToLL_HToBB_M-115_7TeV-powheg-herwigppTreeFile.root"));
|
| 228 |
|
|
signalXsec = xSecZH;
|
| 229 |
|
|
}
|
| 230 |
|
|
//signal for W
|
| 231 |
|
|
if( channel == "Wm" or channel == "We" ){
|
| 232 |
|
|
myInputFile_signal.push_back(TFile::Open("MC_files/file_ZH_ZToLL_HToBB_M-115_7TeV-powheg-herwigppTreeFile.root"));
|
| 233 |
|
|
signalXsec = xSecWH;
|
| 234 |
|
|
}
|
| 235 |
|
|
//background
|
| 236 |
|
|
myInputFile_background.push_back(TFile::Open("MC_files/file_DYJetsToLL_TuneZ2_M-50_7TeV-madgraph-tauolaTreeFile.root"));
|
| 237 |
|
|
backgroundXsec.push_back(backgroundXsecDY);
|
| 238 |
|
|
myInputFile_background.push_back(TFile::Open("MC_files//file_WJetsToLNu_TuneZ2_7TeV-madgraph-tauolaTreeFile.root"));
|
| 239 |
|
|
backgroundXsec.push_back(backgroundXsecWJets);
|
| 240 |
|
|
myInputFile_background.push_back(TFile::Open("MC_files//file_TTJets_TuneZ2_7TeV-madgraph-tauolaTreeFile.root"));
|
| 241 |
|
|
backgroundXsec.push_back(backgroundXsecTT);
|
| 242 |
|
|
myInputFile_background.push_back(TFile::Open("MC_files//file_T_TuneZ2_t-channel_7TeV-powheg-tauolaTreeFile.root"));
|
| 243 |
|
|
backgroundXsec.push_back(backgroundXsecT_tchannel);
|
| 244 |
|
|
myInputFile_background.push_back(TFile::Open("MC_files//file_T_TuneZ2_tW-channel-DR_7TeV-powheg-tauolaTreeFile.root"));
|
| 245 |
|
|
backgroundXsec.push_back(backgroundXsecT_tWDRchannel);
|
| 246 |
|
|
myInputFile_background.push_back(TFile::Open("MC_files//file_Tbar_TuneZ2_t-channel_7TeV-powheg-tauolaTreeFile.root"));
|
| 247 |
|
|
backgroundXsec.push_back(backgroundXsecTbar_tchannel);
|
| 248 |
|
|
myInputFile_background.push_back(TFile::Open("MC_files//file_Tbar_TuneZ2_tW-channel-DR_7TeV-powheg-tauolaTreeFile.root"));
|
| 249 |
|
|
backgroundXsec.push_back(backgroundXsecTbar_tWDRchannel);
|
| 250 |
|
|
// myInputFile_background.push_back(TFile::Open("MC_files//file_Tbar_TuneZ2_tW-channel-DS_7TeV-powheg-tauolaTreeFile.root"));
|
| 251 |
|
|
// backgroundXsec.push_back(backgroundXsecTbar_tWDSchannel);
|
| 252 |
|
|
myInputFile_background.push_back(TFile::Open("MC_files///file_ZZ_TuneZ2_7TeV_pythia6_tauolaTreeFile.root"));
|
| 253 |
|
|
backgroundXsec.push_back(backgroundXsecZZ);
|
| 254 |
|
|
myInputFile_background.push_back(TFile::Open("MC_files/file_WW_TuneZ2_7TeV_pythia6_tauolaTreeFile.root"));
|
| 255 |
|
|
backgroundXsec.push_back(backgroundXsecWW);
|
| 256 |
|
|
//WZ still missing
|
| 257 |
|
|
// myInputFile_background.push_back(TFile::Open("MC_files/"));
|
| 258 |
|
|
// backgroundXsec.push_back(backgroundXsecWZ);
|
| 259 |
|
|
myInputFile_background.push_back(TFile::Open("MC_files//file_ZZ_TuneZ2_7TeV_pythia6_tauolaTreeFile.root"));
|
| 260 |
|
|
backgroundXsec.push_back(backgroundXsecZZ);
|
| 261 |
|
|
|
| 262 |
|
|
|
| 263 |
|
|
if(channel == "Zmm"){
|
| 264 |
|
|
factory->AddVariable( "bbMass", "bb mass" , "GeV/c2", 'F' );
|
| 265 |
|
|
factory->AddVariable( "VMass", "vector mass" , "GeV/c2", 'F' );
|
| 266 |
|
|
factory->AddVariable( "bbPt", "bb pt" , "GeV/c" , 'F' );
|
| 267 |
|
|
factory->AddVariable( "VPt", "vector pt" , "GeV/c" , 'F' );
|
| 268 |
|
|
factory->AddVariable( "btag1", "btag1" , "csv" , 'F' );
|
| 269 |
|
|
factory->AddVariable( "btag2", "btag2" , "csv" , 'F' );
|
| 270 |
|
|
factory->AddVariable( "DeltaPhiVH", "DeltaPhi(V,H)" , "" , 'F' );
|
| 271 |
|
|
factory->AddVariable( "DeltaEtabb", "DeltaEta(b,b)" , "" , 'F' );
|
| 272 |
|
|
|
| 273 |
|
|
//PRESELECTION
|
| 274 |
|
|
mycuts="(bPt1>20) && (bPt2>20) && (btag1>0.5) && (btag2>0.5) && (NaddJet<2) && (DeltaPhiVH>2.4)";
|
| 275 |
|
|
mycutb="(bPt1>20) && (bPt2>20) && (btag1>0.5) && (btag2>0.5) && (NaddJet<2) && (DeltaPhiVH>2.4)";
|
| 276 |
|
|
|
| 277 |
|
|
}
|
| 278 |
|
|
else if(channel == "Zee"){
|
| 279 |
|
|
factory->AddVariable( "bbMass", "bb mass" , "GeV/c2", 'F' );
|
| 280 |
|
|
factory->AddVariable( "VMass", "vector mass" , "GeV/c2", 'F' );
|
| 281 |
|
|
factory->AddVariable( "bbPt", "bb pt" , "GeV/c" , 'F' );
|
| 282 |
|
|
factory->AddVariable( "VPt", "vector pt" , "GeV/c" , 'F' );
|
| 283 |
|
|
factory->AddVariable( "btag1", "btag1" , "csv" , 'F' );
|
| 284 |
|
|
factory->AddVariable( "btag2", "btag2" , "csv" , 'F' );
|
| 285 |
|
|
factory->AddVariable( "DeltaPhiVH", "DeltaPhi(V,H)" , "" , 'F' );
|
| 286 |
|
|
factory->AddVariable( "DeltaEtabb", "DeltaEta(b,b)" , "" , 'F' );
|
| 287 |
|
|
|
| 288 |
|
|
//PRESELECTION
|
| 289 |
|
|
mycuts="bPt1>20 && bPt2>20 && btag1>0.5 && btag2>0.5 && NaddJet<2 && DeltaPhiVH>2.4";
|
| 290 |
|
|
mycutb="bPt1>20 && bPt2>20 && btag1>0.5 && btag2>0.5 && NaddJet<2 && DeltaPhiVH>2.4";
|
| 291 |
|
|
|
| 292 |
|
|
}
|
| 293 |
|
|
else if(channel == "Znn"){
|
| 294 |
|
|
factory->AddVariable( "bbMass", "bb mass" , "GeV/c2", 'F' );
|
| 295 |
|
|
factory->AddVariable( "bbPt", "bb pt" , "GeV/c" , 'F' );
|
| 296 |
|
|
factory->AddVariable( "pfMET", "met pt" , "GeV/c" , 'F' );
|
| 297 |
|
|
factory->AddVariable( "btag1", "btag1" , "csv" , 'F' );
|
| 298 |
|
|
factory->AddVariable( "btag2", "btag2" , "csv" , 'F' );
|
| 299 |
|
|
factory->AddVariable( "DeltaPhiVH", "DeltaPhi(V,H)" , "" , 'F' );
|
| 300 |
|
|
//NaddJet is imposed to be 0
|
| 301 |
|
|
// factory->AddVariable( "NaddJet", "NaddJet" , "" , 'F' );
|
| 302 |
|
|
|
| 303 |
|
|
//PRESELECTION
|
| 304 |
|
|
mycuts="bPt1>80 && bPt2>20 && bbPt>160 && btag1>0.5 && btag2>0.5 && NaddJet<1 && deltaPhipfMETjet1>0.5 && deltaPhipfMETjet2 && pfMETsig>5";
|
| 305 |
|
|
mycutb="bPt1>80 && bPt2>20 && bbPt>160 && btag1>0.5 && btag2>0.5 && NaddJet<1 && deltaPhipfMETjet1>0.5 && deltaPhipfMETjet2 && pfMETsig>5";
|
| 306 |
|
|
|
| 307 |
|
|
}
|
| 308 |
|
|
else if(channel == "We"){
|
| 309 |
|
|
factory->AddVariable( "bbMass", "bb mass" , "GeV/c2", 'F' );
|
| 310 |
|
|
factory->AddVariable( "bbPt", "bb pt" , "GeV/c" , 'F' );
|
| 311 |
|
|
factory->AddVariable( "VPt", "vector pt" , "GeV/c" , 'F' );
|
| 312 |
|
|
factory->AddVariable( "btag1", "btag1" , "csv" , 'F' );
|
| 313 |
|
|
factory->AddVariable( "btag2", "btag2" , "csv" , 'F' );
|
| 314 |
|
|
factory->AddVariable( "DeltaPhiVH", "DeltaPhi(V,H)" , "" , 'F' );
|
| 315 |
|
|
factory->AddVariable( "DeltaEtabb", "DeltaEta(b,b)" , "" , 'F' );
|
| 316 |
|
|
//NaddJet is imposed to be 0
|
| 317 |
|
|
// factory->AddVariable( "NaddJet", "NaddJet" , "" , 'F' );
|
| 318 |
|
|
|
| 319 |
|
|
//PRESELECTION
|
| 320 |
|
|
mycuts="bPt1>30 && bPt2>30 && bbPt>150 && VPt>150 && btag1>0.4 && btag2>0.4 && NaddJet<1 && pfMETsig>2";
|
| 321 |
|
|
mycutb="bPt1>30 && bPt2>30 && bbPt>150 && VPt>150 && btag1>0.4 && btag2>0.4 && NaddJet<1 && pfMETsig>2";
|
| 322 |
|
|
|
| 323 |
|
|
}
|
| 324 |
|
|
else if(channel == "Wm"){
|
| 325 |
|
|
factory->AddVariable( "bbMass", "bb mass" , "GeV/c2", 'F' );
|
| 326 |
|
|
factory->AddVariable( "bbPt", "bb pt" , "GeV/c" , 'F' );
|
| 327 |
|
|
factory->AddVariable( "VPt", "vector pt" , "GeV/c" , 'F' );
|
| 328 |
|
|
factory->AddVariable( "btag1", "btag1" , "csv" , 'F' );
|
| 329 |
|
|
factory->AddVariable( "btag2", "btag2" , "csv" , 'F' );
|
| 330 |
|
|
factory->AddVariable( "DeltaPhiVH", "DeltaPhi(V,H)" , "" , 'F' );
|
| 331 |
|
|
factory->AddVariable( "DeltaEtabb", "DeltaEta(b,b)" , "" , 'F' );
|
| 332 |
|
|
//NaddJet is imposed to be 0
|
| 333 |
|
|
// factory->AddVariable( "NaddJet", "NaddJet" , "" , 'F' );
|
| 334 |
|
|
|
| 335 |
|
|
//PRESELECTION
|
| 336 |
|
|
mycuts="bPt1>30 && bPt2>30 && bbPt>150 && VPt>150 && btag1>0.4 && btag2>0.4 && NaddJet<1";
|
| 337 |
|
|
mycutb="bPt1>30 && bPt2>30 && bbPt>150 && VPt>150 && btag1>0.4 && btag2>0.4 && NaddJet<1";
|
| 338 |
|
|
|
| 339 |
|
|
}
|
| 340 |
|
|
|
| 341 |
|
|
std::vector<Sample> samples_signal;
|
| 342 |
|
|
std::vector<Sample> samples_background;
|
| 343 |
|
|
|
| 344 |
|
|
std::string str_toReplace("TreeFile");
|
| 345 |
|
|
|
| 346 |
|
|
std::vector<TTree*> myInputTree_signal;
|
| 347 |
|
|
std::vector<TTree*> myInputTree_backgound;
|
| 348 |
|
|
for(unsigned int i = 0; i < myInputFile_signal.size(); ++i ){
|
| 349 |
|
|
std::string name = myInputFile_signal.at(0)->GetName();
|
| 350 |
|
|
samples_signal.push_back(Sample(signalXsec,"signal", name.replace(name.find("TreeFile"),str_toReplace.length() ,"_histos" ) , kRed,false));
|
| 351 |
|
|
TTree * tmpTree = (TTree*)myInputFile_signal.at(i)->Get("treeMVA");
|
| 352 |
|
|
factory->AddSignalTree( tmpTree , samples_signal.at(i).scale(lumi) );
|
| 353 |
|
|
}
|
| 354 |
|
|
for(unsigned int i = 0; i < myInputFile_background.size(); ++i ){
|
| 355 |
|
|
std::string name = myInputFile_background.at(i)->GetName();
|
| 356 |
|
|
samples_background.push_back(Sample(backgroundXsec.at(i),"background", name.replace(name.find("TreeFile"),str_toReplace.length() ,"_histos" ) ,i+10,false));
|
| 357 |
|
|
TTree * tmpbackTree = (TTree*)myInputFile_background.at(i)->Get("treeMVA");
|
| 358 |
|
|
factory->AddBackgroundTree( tmpbackTree, samples_background.at(i).scale(lumi) );
|
| 359 |
|
|
}
|
| 360 |
|
|
|
| 361 |
|
|
//from Michele and Matt
|
| 362 |
|
|
factory->PrepareTrainingAndTestTree( mycuts, mycutb,
|
| 363 |
|
|
"nTrain_Signal=0:nTrain_Background=0:SplitMode=Random:NormMode=NumEvents:!V" );
|
| 364 |
|
|
|
| 365 |
|
|
|
| 366 |
|
|
// ---- Book MVA methods
|
| 367 |
|
|
//
|
| 368 |
|
|
// Please lookup the various method configuration options in the corresponding cxx files, eg:
|
| 369 |
|
|
// src/MethoCuts.cxx, etc, or here: http://tmva.sourceforge.net/optionRef.html
|
| 370 |
|
|
// it is possible to preset ranges in the option string in which the cut optimisation should be done:
|
| 371 |
|
|
// "...:CutRangeMin[2]=-1:CutRangeMax[2]=1"...", where [2] is the third input variable
|
| 372 |
|
|
|
| 373 |
|
|
// Cut optimisation
|
| 374 |
|
|
if (Use["Cuts"])
|
| 375 |
|
|
factory->BookMethod( TMVA::Types::kCuts, "Cuts",
|
| 376 |
|
|
"!H:!V:FitMethod=MC:EffSel:SampleSize=200000:VarProp=FSmart" );
|
| 377 |
|
|
|
| 378 |
|
|
if (Use["CutsD"])
|
| 379 |
|
|
factory->BookMethod( TMVA::Types::kCuts, "CutsD",
|
| 380 |
|
|
"!H:!V:FitMethod=MC:EffSel:SampleSize=200000:VarProp=FSmart:VarTransform=Decorrelate" );
|
| 381 |
|
|
|
| 382 |
|
|
if (Use["CutsPCA"])
|
| 383 |
|
|
factory->BookMethod( TMVA::Types::kCuts, "CutsPCA",
|
| 384 |
|
|
"!H:!V:FitMethod=MC:EffSel:SampleSize=200000:VarProp=FSmart:VarTransform=PCA" );
|
| 385 |
|
|
|
| 386 |
|
|
if (Use["CutsGA"])
|
| 387 |
|
|
factory->BookMethod( TMVA::Types::kCuts, "CutsGA",
|
| 388 |
|
|
"H:!V:FitMethod=GA:CutRangeMin[0]=-10:CutRangeMax[0]=10:VarProp[1]=FMax:EffSel:Steps=30:Cycles=3:PopSize=400:SC_steps=10:SC_rate=5:SC_factor=0.95" );
|
| 389 |
|
|
|
| 390 |
|
|
if (Use["CutsSA"])
|
| 391 |
|
|
factory->BookMethod( TMVA::Types::kCuts, "CutsSA",
|
| 392 |
|
|
"!H:!V:FitMethod=SA:EffSel:MaxCalls=150000:KernelTemp=IncAdaptive:InitialTemp=1e+6:MinTemp=1e-6:Eps=1e-10:UseDefaultScale" );
|
| 393 |
|
|
|
| 394 |
|
|
// Likelihood ("naive Bayes estimator")
|
| 395 |
|
|
if (Use["Likelihood"])
|
| 396 |
|
|
factory->BookMethod( TMVA::Types::kLikelihood, "Likelihood",
|
| 397 |
|
|
"H:!V:!TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmoothBkg[1]=10:NSmooth=1:NAvEvtPerBin=50" );
|
| 398 |
|
|
|
| 399 |
|
|
// Decorrelated likelihood
|
| 400 |
|
|
if (Use["LikelihoodD"])
|
| 401 |
|
|
factory->BookMethod( TMVA::Types::kLikelihood, "LikelihoodD",
|
| 402 |
|
|
"!H:!V:TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmooth=5:NAvEvtPerBin=50:VarTransform=Decorrelate" );
|
| 403 |
|
|
|
| 404 |
|
|
// PCA-transformed likelihood
|
| 405 |
|
|
if (Use["LikelihoodPCA"])
|
| 406 |
|
|
factory->BookMethod( TMVA::Types::kLikelihood, "LikelihoodPCA",
|
| 407 |
|
|
"!H:!V:!TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmooth=5:NAvEvtPerBin=50:VarTransform=PCA" );
|
| 408 |
|
|
|
| 409 |
|
|
// Use a kernel density estimator to approximate the PDFs
|
| 410 |
|
|
if (Use["LikelihoodKDE"])
|
| 411 |
|
|
factory->BookMethod( TMVA::Types::kLikelihood, "LikelihoodKDE",
|
| 412 |
|
|
"!H:!V:!TransformOutput:PDFInterpol=KDE:KDEtype=Gauss:KDEiter=Adaptive:KDEFineFactor=0.3:KDEborder=None:NAvEvtPerBin=50" );
|
| 413 |
|
|
|
| 414 |
|
|
// Use a variable-dependent mix of splines and kernel density estimator
|
| 415 |
|
|
if (Use["LikelihoodMIX"])
|
| 416 |
|
|
factory->BookMethod( TMVA::Types::kLikelihood, "LikelihoodMIX",
|
| 417 |
|
|
"!H:!V:!TransformOutput:PDFInterpolSig[0]=KDE:PDFInterpolBkg[0]=KDE:PDFInterpolSig[1]=KDE:PDFInterpolBkg[1]=KDE:PDFInterpolSig[2]=Spline2:PDFInterpolBkg[2]=Spline2:PDFInterpolSig[3]=Spline2:PDFInterpolBkg[3]=Spline2:KDEtype=Gauss:KDEiter=Nonadaptive:KDEborder=None:NAvEvtPerBin=50" );
|
| 418 |
|
|
|
| 419 |
|
|
// Test the multi-dimensional probability density estimator
|
| 420 |
|
|
// here are the options strings for the MinMax and RMS methods, respectively:
|
| 421 |
|
|
// "!H:!V:VolumeRangeMode=MinMax:DeltaFrac=0.2:KernelEstimator=Gauss:GaussSigma=0.3" );
|
| 422 |
|
|
// "!H:!V:VolumeRangeMode=RMS:DeltaFrac=3:KernelEstimator=Gauss:GaussSigma=0.3" );
|
| 423 |
|
|
if (Use["PDERS"])
|
| 424 |
|
|
factory->BookMethod( TMVA::Types::kPDERS, "PDERS",
|
| 425 |
|
|
"!H:!V:NormTree=T:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600" );
|
| 426 |
|
|
|
| 427 |
|
|
if (Use["PDERSD"])
|
| 428 |
|
|
factory->BookMethod( TMVA::Types::kPDERS, "PDERSD",
|
| 429 |
|
|
"!H:!V:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600:VarTransform=Decorrelate" );
|
| 430 |
|
|
|
| 431 |
|
|
if (Use["PDERSPCA"])
|
| 432 |
|
|
factory->BookMethod( TMVA::Types::kPDERS, "PDERSPCA",
|
| 433 |
|
|
"!H:!V:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600:VarTransform=PCA" );
|
| 434 |
|
|
|
| 435 |
|
|
// Multi-dimensional likelihood estimator using self-adapting phase-space binning
|
| 436 |
|
|
if (Use["PDEFoam"])
|
| 437 |
|
|
factory->BookMethod( TMVA::Types::kPDEFoam, "PDEFoam",
|
| 438 |
|
|
"H:!V:SigBgSeparate=F:TailCut=0.001:VolFrac=0.0333:nActiveCells=500:nSampl=2000:nBin=5:Nmin=100:Kernel=None:Compress=T" );
|
| 439 |
|
|
|
| 440 |
|
|
if (Use["PDEFoamBoost"])
|
| 441 |
|
|
factory->BookMethod( TMVA::Types::kPDEFoam, "PDEFoamBoost",
|
| 442 |
|
|
"!H:!V:Boost_Num=30:Boost_Transform=linear:SigBgSeparate=F:MaxDepth=4:UseYesNoCell=T:DTLogic=MisClassificationError:FillFoamWithOrigWeights=F:TailCut=0:nActiveCells=500:nBin=20:Nmin=400:Kernel=None:Compress=T" );
|
| 443 |
|
|
|
| 444 |
|
|
// K-Nearest Neighbour classifier (KNN)
|
| 445 |
|
|
if (Use["KNN"])
|
| 446 |
|
|
factory->BookMethod( TMVA::Types::kKNN, "KNN",
|
| 447 |
|
|
"H:nkNN=20:ScaleFrac=0.8:SigmaFact=1.0:Kernel=Gaus:UseKernel=F:UseWeight=T:!Trim" );
|
| 448 |
|
|
|
| 449 |
|
|
// H-Matrix (chi2-squared) method
|
| 450 |
|
|
if (Use["HMatrix"])
|
| 451 |
|
|
factory->BookMethod( TMVA::Types::kHMatrix, "HMatrix", "!H:!V" );
|
| 452 |
|
|
|
| 453 |
|
|
// Linear discriminant (same as Fisher discriminant)
|
| 454 |
|
|
if (Use["LD"])
|
| 455 |
|
|
factory->BookMethod( TMVA::Types::kLD, "LD", "H:!V:VarTransform=None:CreateMVAPdfs:PDFInterpolMVAPdf=Spline2:NbinsMVAPdf=50:NsmoothMVAPdf=10" );
|
| 456 |
|
|
|
| 457 |
|
|
// Fisher discriminant (same as LD)
|
| 458 |
|
|
if (Use["Fisher"])
|
| 459 |
|
|
factory->BookMethod( TMVA::Types::kFisher, "Fisher", "H:!V:Fisher:CreateMVAPdfs:PDFInterpolMVAPdf=Spline2:NbinsMVAPdf=50:NsmoothMVAPdf=10" );
|
| 460 |
|
|
|
| 461 |
|
|
// Fisher with Gauss-transformed input variables
|
| 462 |
|
|
if (Use["FisherG"])
|
| 463 |
|
|
factory->BookMethod( TMVA::Types::kFisher, "FisherG", "H:!V:VarTransform=Gauss" );
|
| 464 |
|
|
|
| 465 |
|
|
// Composite classifier: ensemble (tree) of boosted Fisher classifiers
|
| 466 |
|
|
if (Use["BoostedFisher"])
|
| 467 |
|
|
factory->BookMethod( TMVA::Types::kFisher, "BoostedFisher", "H:!V:Boost_Num=20:Boost_Transform=log:Boost_Type=AdaBoost:Boost_AdaBoostBeta=0.2");
|
| 468 |
|
|
|
| 469 |
|
|
// Function discrimination analysis (FDA) -- test of various fitters - the recommended one is Minuit (or GA or SA)
|
| 470 |
|
|
if (Use["FDA_MC"])
|
| 471 |
|
|
factory->BookMethod( TMVA::Types::kFDA, "FDA_MC",
|
| 472 |
|
|
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1);(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=MC:SampleSize=100000:Sigma=0.1" );
|
| 473 |
|
|
|
| 474 |
|
|
if (Use["FDA_GA"]) // can also use Simulated Annealing (SA) algorithm (see Cuts_SA options])
|
| 475 |
|
|
factory->BookMethod( TMVA::Types::kFDA, "FDA_GA",
|
| 476 |
|
|
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1);(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=GA:PopSize=300:Cycles=3:Steps=20:Trim=True:SaveBestGen=1" );
|
| 477 |
|
|
|
| 478 |
|
|
if (Use["FDA_SA"]) // can also use Simulated Annealing (SA) algorithm (see Cuts_SA options])
|
| 479 |
|
|
factory->BookMethod( TMVA::Types::kFDA, "FDA_SA",
|
| 480 |
|
|
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1);(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=SA:MaxCalls=15000:KernelTemp=IncAdaptive:InitialTemp=1e+6:MinTemp=1e-6:Eps=1e-10:UseDefaultScale" );
|
| 481 |
|
|
|
| 482 |
|
|
if (Use["FDA_MT"])
|
| 483 |
|
|
factory->BookMethod( TMVA::Types::kFDA, "FDA_MT",
|
| 484 |
|
|
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1);(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=MINUIT:ErrorLevel=1:PrintLevel=-1:FitStrategy=2:UseImprove:UseMinos:SetBatch" );
|
| 485 |
|
|
|
| 486 |
|
|
if (Use["FDA_GAMT"])
|
| 487 |
|
|
factory->BookMethod( TMVA::Types::kFDA, "FDA_GAMT",
|
| 488 |
|
|
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1);(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=GA:Converger=MINUIT:ErrorLevel=1:PrintLevel=-1:FitStrategy=0:!UseImprove:!UseMinos:SetBatch:Cycles=1:PopSize=5:Steps=5:Trim" );
|
| 489 |
|
|
|
| 490 |
|
|
if (Use["FDA_MCMT"])
|
| 491 |
|
|
factory->BookMethod( TMVA::Types::kFDA, "FDA_MCMT",
|
| 492 |
|
|
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x2+(4)*x3:ParRanges=(-1,1);(-10,10);(-10,10);(-10,10);(-10,10):FitMethod=MC:Converger=MINUIT:ErrorLevel=1:PrintLevel=-1:FitStrategy=0:!UseImprove:!UseMinos:SetBatch:SampleSize=20" );
|
| 493 |
|
|
|
| 494 |
|
|
// TMVA ANN: MLP (recommended ANN) -- all ANNs in TMVA are Multilayer Perceptrons
|
| 495 |
|
|
if (Use["MLP"])
|
| 496 |
|
|
factory->BookMethod( TMVA::Types::kMLP, "MLP", "H:!V:NeuronType=tanh:VarTransform=N:NCycles=600:HiddenLayers=N+5:TestRate=5:!UseRegulator" );
|
| 497 |
|
|
|
| 498 |
|
|
if (Use["MLPBFGS"])
|
| 499 |
|
|
factory->BookMethod( TMVA::Types::kMLP, "MLPBFGS", "H:!V:NeuronType=tanh:VarTransform=N:NCycles=600:HiddenLayers=N+5:TestRate=5:TrainingMethod=BFGS:!UseRegulator" );
|
| 500 |
|
|
|
| 501 |
|
|
if (Use["MLPBNN"])
|
| 502 |
|
|
factory->BookMethod( TMVA::Types::kMLP, "MLPBNN", "H:!V:NeuronType=tanh:VarTransform=N:NCycles=600:HiddenLayers=N+5:TestRate=5:TrainingMethod=BFGS:UseRegulator" ); // BFGS training with bayesian regulators
|
| 503 |
|
|
|
| 504 |
|
|
// CF(Clermont-Ferrand)ANN
|
| 505 |
|
|
if (Use["CFMlpANN"])
|
| 506 |
|
|
factory->BookMethod( TMVA::Types::kCFMlpANN, "CFMlpANN", "!H:!V:NCycles=2000:HiddenLayers=N+1,N" ); // n_cycles:#nodes:#nodes:...
|
| 507 |
|
|
|
| 508 |
|
|
// Tmlp(Root)ANN
|
| 509 |
|
|
if (Use["TMlpANN"])
|
| 510 |
|
|
factory->BookMethod( TMVA::Types::kTMlpANN, "TMlpANN", "!H:!V:NCycles=200:HiddenLayers=N+1,N:LearningMethod=BFGS:ValidationFraction=0.3" ); // n_cycles:#nodes:#nodes:...
|
| 511 |
|
|
|
| 512 |
|
|
// Support Vector Machine
|
| 513 |
|
|
if (Use["SVM"])
|
| 514 |
|
|
factory->BookMethod( TMVA::Types::kSVM, "SVM", "Gamma=0.25:Tol=0.001:VarTransform=Norm" );
|
| 515 |
|
|
|
| 516 |
|
|
// Boosted Decision Trees
|
| 517 |
|
|
if (Use["BDTG"]) // Gradient Boost
|
| 518 |
|
|
factory->BookMethod( TMVA::Types::kBDT, "BDTG",
|
| 519 |
|
|
"!H:!V:NTrees=1000:BoostType=Grad:Shrinkage=0.10:UseBaggedGrad:GradBaggingFraction=0.5:nCuts=20:NNodesMax=5" );
|
| 520 |
|
|
|
| 521 |
|
|
if (Use["BDT"]) // Adaptive Boost
|
| 522 |
|
|
factory->BookMethod( TMVA::Types::kBDT, "BDT",
|
| 523 |
|
|
"!H:!V:NTrees=850:nEventsMin=150:MaxDepth=3:BoostType=AdaBoost:AdaBoostBeta=0.5:SeparationType=GiniIndex:nCuts=20:PruneMethod=NoPruning" );
|
| 524 |
|
|
|
| 525 |
|
|
if (Use["BDTB"]) // Bagging
|
| 526 |
|
|
factory->BookMethod( TMVA::Types::kBDT, "BDTB",
|
| 527 |
|
|
"!H:!V:NTrees=400:BoostType=Bagging:SeparationType=GiniIndex:nCuts=20:PruneMethod=NoPruning" );
|
| 528 |
|
|
|
| 529 |
|
|
if (Use["BDTD"]) // Decorrelation + Adaptive Boost
|
| 530 |
|
|
factory->BookMethod( TMVA::Types::kBDT, "BDTD",
|
| 531 |
|
|
"!H:!V:NTrees=400:nEventsMin=400:MaxDepth=3:BoostType=AdaBoost:SeparationType=GiniIndex:nCuts=20:PruneMethod=NoPruning:VarTransform=Decorrelate" );
|
| 532 |
|
|
|
| 533 |
|
|
// RuleFit -- TMVA implementation of Friedman's method
|
| 534 |
|
|
if (Use["RuleFit"])
|
| 535 |
|
|
factory->BookMethod( TMVA::Types::kRuleFit, "RuleFit",
|
| 536 |
|
|
"H:!V:RuleFitModule=RFTMVA:Model=ModRuleLinear:MinImp=0.001:RuleMinDist=0.001:NTrees=20:fEventsMin=0.01:fEventsMax=0.5:GDTau=-1.0:GDTauPrec=0.01:GDStep=0.01:GDNSteps=10000:GDErrScale=1.02" );
|
| 537 |
|
|
|
| 538 |
|
|
// For an example of the category classifier, see: TMVAClassificationCategory
|
| 539 |
|
|
|
| 540 |
|
|
// For an example of the category classifier usage, see: TMVAClassificationCategory
|
| 541 |
|
|
|
| 542 |
|
|
// --------------------------------------------------------------------------------------------------
|
| 543 |
|
|
|
| 544 |
|
|
// ---- Now you can optimize the setting (configuration) of the MVAs using the set of training events
|
| 545 |
|
|
|
| 546 |
|
|
// factory->OptimizeAllMethods("SigEffAt001","Scan");
|
| 547 |
|
|
// factory->OptimizeAllMethods("ROCIntegral","GA");
|
| 548 |
|
|
|
| 549 |
|
|
// --------------------------------------------------------------------------------------------------
|
| 550 |
|
|
|
| 551 |
|
|
// ---- Now you can tell the factory to train, test, and evaluate the MVAs
|
| 552 |
|
|
|
| 553 |
|
|
// Train MVAs using the set of training events
|
| 554 |
|
|
factory->TrainAllMethods();
|
| 555 |
|
|
|
| 556 |
|
|
// ---- Evaluate all MVAs using the set of test events
|
| 557 |
|
|
factory->TestAllMethods();
|
| 558 |
|
|
|
| 559 |
|
|
// ----- Evaluate and compare performance of all configured MVAs
|
| 560 |
|
|
factory->EvaluateAllMethods();
|
| 561 |
|
|
|
| 562 |
|
|
// --------------------------------------------------------------
|
| 563 |
|
|
|
| 564 |
|
|
// Save the output
|
| 565 |
|
|
outputFile->Close();
|
| 566 |
|
|
|
| 567 |
|
|
std::cout << "==> Wrote root file: " << outputFile->GetName() << std::endl
|
| 568 |
|
|
<< "==> TMVAClassification is done!" << std::endl
|
| 569 |
|
|
<< std::endl
|
| 570 |
|
|
<< "==> To view the results, launch the GUI: \"root -l ./TMVAGui.C\"" << std::endl
|
| 571 |
|
|
<< std::endl;
|
| 572 |
|
|
|
| 573 |
|
|
// Clean up
|
| 574 |
|
|
delete factory;
|
| 575 |
|
|
}
|
| 576 |
|
|
|
