void TMVAClassification()
{
TMVA::Tools::Instance();
std::cout << "==> Start TMVAClassification" << std::endl;
TString outfileName( "TMVA.root" );
TFile* outputFile = TFile::Open( outfileName, "RECREATE" );
TMVA::Factory *factory = new TMVA::Factory( "TMVAClassification", outputFile,
"AnalysisType=Classification" );
factory->AddVariable("LifeTime", 'F');
factory->AddVariable("FlightDistance", 'F');
factory->AddVariable("pt", 'F');
TString fname = "../tau_data/training.root";
TFile *input = TFile::Open( fname );
TTree *tree = (TTree*)input->Get("data");
factory->AddTree(tree, "Signal", 1., "signal == 1", "Training");
factory->AddTree(tree, "Signal", 1., "signal == 1", "Test");
factory->AddTree(tree, "Background", 1., "signal == 0", "Training");
factory->AddTree(tree, "Background", 1., "signal == 0", "Test");
// gradient boosting training
factory->BookMethod(TMVA::Types::kBDT, "GBDT",
"NTrees=40:BoostType=Grad:Shrinkage=0.01:MaxDepth=7:UseNvars=6:nCuts=20:MinNodeSize=10");
factory->TrainAllMethods();
input->Close();
outputFile->Close();
delete factory;
}
void TMVAClassification( TString myMethodList = "" )
{
// The explicit loading of the shared libTMVA is done in TMVAlogon.C, defined in .rootrc
// if you use your private .rootrc, or run from a different directory, please copy the
// corresponding lines from .rootrc
// methods to be processed can be given as an argument; use format:
//
// mylinux~> root -l TMVAClassification.C\(\"myMethod1,myMethod2,myMethod3\"\)
//
// if you like to use a method via the plugin mechanism, we recommend using
//
// mylinux~> root -l TMVAClassification.C\(\"P_myMethod\"\)
// (an example is given for using the BDT as plugin (see below),
// but of course the real application is when you write your own
// method based)
//---------------------------------------------------------------
// This loads the library
TMVA::Tools::Instance();
// Default MVA methods to be trained + tested
std::map<std::string,int> Use;
// --- Cut optimisation
Use["Cuts"] = 1;
Use["CutsD"] = 1;
Use["CutsPCA"] = 0;
Use["CutsGA"] = 0;
Use["CutsSA"] = 0;
//
// --- 1-dimensional likelihood ("naive Bayes estimator")
Use["Likelihood"] = 1;
Use["LikelihoodD"] = 0; // the "D" extension indicates decorrelated input variables (see option strings)
Use["LikelihoodPCA"] = 1; // the "PCA" extension indicates PCA-transformed input variables (see option strings)
Use["LikelihoodKDE"] = 0;
Use["LikelihoodMIX"] = 0;
//
// --- Mutidimensional likelihood and Nearest-Neighbour methods
Use["PDERS"] = 1;
Use["PDERSD"] = 0;
Use["PDERSPCA"] = 0;
Use["PDEFoam"] = 0;
Use["PDEFoamBoost"] = 0; // uses generalised MVA method boosting
Use["KNN"] = 1; // k-nearest neighbour method
//
// --- Linear Discriminant Analysis
Use["LD"] = 1; // Linear Discriminant identical to Fisher
Use["Fisher"] = 0;
Use["FisherG"] = 0;
Use["BoostedFisher"] = 0; // uses generalised MVA method boosting
Use["HMatrix"] = 0;
//
// --- Function Discriminant analysis
Use["FDA_GA"] = 1; // minimisation of user-defined function using Genetics Algorithm
Use["FDA_SA"] = 0;
Use["FDA_MC"] = 0;
Use["FDA_MT"] = 0;
Use["FDA_GAMT"] = 0;
Use["FDA_MCMT"] = 0;
//
// --- Neural Networks (all are feed-forward Multilayer Perceptrons)
Use["MLP"] = 0; // Recommended ANN
Use["MLPBFGS"] = 0; // Recommended ANN with optional training method
Use["MLPBNN"] = 1; // Recommended ANN with BFGS training method and bayesian regulator
Use["CFMlpANN"] = 0; // Depreciated ANN from ALEPH
Use["TMlpANN"] = 0; // ROOT's own ANN
//
// --- Support Vector Machine
Use["SVM"] = 1;
//
// --- Boosted Decision Trees
Use["BDT"] = 1; // uses Adaptive Boost
Use["BDTG"] = 0; // uses Gradient Boost
Use["BDTB"] = 0; // uses Bagging
Use["BDTD"] = 0; // decorrelation + Adaptive Boost
Use["BDTF"] = 0; // allow usage of fisher discriminant for node splitting
//
// --- Friedman's RuleFit method, ie, an optimised series of cuts ("rules")
Use["RuleFit"] = 1;
// ---------------------------------------------------------------
(TMVA::gConfig().GetVariablePlotting()).fMaxNumOfAllowedVariablesForScatterPlots = 20;
std::cout << std::endl;
std::cout << "==> Start TMVAClassification" << std::endl;
// Select methods (don't look at this code - not of interest)
if (myMethodList != "") {
for (std::map<std::string,int>::iterator it = Use.begin(); it != Use.end(); it++) it->second = 0;
std::vector<TString> mlist = TMVA::gTools().SplitString( myMethodList, ',' );
for (UInt_t i=0; i<mlist.size(); i++) {
std::string regMethod(mlist[i]);
if (Use.find(regMethod) == Use.end()) {
std::cout << "Method \"" << regMethod << "\" not known in TMVA under this name. Choose among the following:" << std::endl;
for (std::map<std::string,int>::iterator it = Use.begin(); it != Use.end(); it++) std::cout << it->first << " ";
std::cout << std::endl;
return;
}
Use[regMethod] = 1;
}
}
// --------------------------------------------------------------------------------------------------
// --- Here the preparation phase begins
//FILEOUT
// Create a ROOT output file where TMVA will store ntuples, histograms, etc.
TString outfileName( "TMVA_2GeV_barrel2.root" );
TFile* outputFile = TFile::Open( outfileName, "RECREATE" );
// Create the factory object. Later you can choose the methods
// whose performance you'd like to investigate. The factory is
// the only TMVA object you have to interact with
//
// The first argument is the base of the name of all the
// weightfiles in the directory weight/
//
// The second argument is the output file for the training results
// All TMVA output can be suppressed by removing the "!" (not) in
// front of the "Silent" argument in the option string
// TMVA::Factory *factory = new TMVA::Factory( "TMVAClassification", outputFile,
// "!V:!Silent:Color:DrawProgressBar:Transformations=I;D;P;G,D:AnalysisType=Classification" );
TMVA::Factory* factory = new TMVA::Factory("TMVAMulticlass_2GeV_barrel2",outputFile,
"!V:!Silent:Color:!DrawProgressBar:Transformations=I;D;P;G,D:AnalysisType=multiclass" );
TFile* file = TFile::Open("/scratchfs/higgs/yudan/SingleParticle/Reco//all_2GeV/PID_2GeV_E30L_E10mm_H48L_H10mm_2GeV.root");
if (!file->IsOpen()) std::cout << "could not open file" <<std::endl;
TTree* treePi = (TTree*)file->Get("Pion");
if (treePi == 0) std::cout << "could not open tree" <<std::endl;
TTree* treeMu = (TTree*)file->Get("Muon");
if (treeMu == 0) std::cout << "could not open tree" <<std::endl;
TTree* treeE = (TTree*)file->Get("Electron");
if (treeE == 0) std::cout << "could not open tree" <<std::endl;
// If you wish to modify default settings
// (please check "src/Config.h" to see all available global options)
// (TMVA::gConfig().GetVariablePlotting()).fTimesRMS = 8.0;
// (TMVA::gConfig().GetIONames()).fWeightFileDir = "myWeightDirectory";
// Define the input variables that shall be used for the MVA training
// note that you may also use variable expressions, such as: "3*var1/var2*abs(var3)"
// [all types of expressions that can also be parsed by TTree::Draw( "expression" )]
// TCut mycuts = "NTrk==1 && MCPP[2]/MCPEn<0.7"; // for example: TCut mycuts = "abs(var1)<0.5 && abs(var2-0.5)<1";
factory->AddTree( treePi, TString("Pi"), 1. ,"abs(cosTheta)>0.3&&abs(cosTheta)<0.55");
factory->AddTree( treeMu, TString("Mu"), 1.,"abs(cosTheta)>0.3&&abs(cosTheta)<0.55");
factory->AddTree( treeE, TString("E"), 1.,"abs(cosTheta)>0.3&&abs(cosTheta)<0.55");
factory->AddVariable("EcalNHit","EcalNHit", "units", 'I');
factory->AddVariable("HcalNHit","HcalNHit", "units", 'I');
factory->AddVariable("NLEcal","NLEcal", "units", 'I');
factory->AddVariable("NLHcal","NLHcal", "units", 'I');
factory->AddVariable("maxDisHtoL","maxDisHtoL", "units", 'F');
factory->AddVariable("avDisHtoL","avDisHtoL", "units", 'F');
factory->AddVariable("EE := EcalEn/EClu","EcalEn", "units", 'F');
factory->AddVariable("graDepth","graDepth", "units", 'F');
factory->AddVariable("cluDepth","cluDepth", "units", 'F');
factory->AddVariable("minDepth","minDepth", "units", 'F');
factory->AddVariable("MaxDisHel","MaxDisHel", "units", 'F');
factory->AddVariable("FD_all","FD_all", "units", 'F');
factory->AddVariable("FD_ECAL","FD_ECAL", "units", 'F');
factory->AddVariable("FD_HCAL","FD_HCAL", "units", 'F');
factory->AddVariable("E_10 := EEClu_L10/EcalEn","EEClu_L10", "units", 'F');
factory->AddVariable("E_R := EEClu_R/EcalEn","EEClu_R", "units", 'F');
factory->AddVariable("E_r := EEClu_r/EcalEn","EEClu_r", "units", 'F');
factory->AddVariable("rms_Hcal","rms_Hcal", "units", 'F');
factory->AddVariable("av_NHH","av_NHH", "units", 'F');
factory->AddVariable("AL_Ecal","AL_Ecal", "units", 'I');
factory->AddVariable("FD_ECALF10","FD_ECALF10", "units", 'F');
factory->AddVariable("FD_ECALL20","FD_ECALL20", "units", 'F');
factory->AddVariable("NH_ECALF10","NH_ECALF10", "units", 'I');
factory->AddVariable("dEdx","dEdx", "units", 'F');
std::cout << "--- TMVAClassification : Using input file: " << std::endl;
// --- Register the training and test trees
// global event weights per tree (see below for setting event-wise weights)
Double_t signalWeight = 1.0;
Double_t backgroundWeight = 1.0;
// You can add an arbitrary number of signal or background trees
// To give different trees for training and testing, do as follows:
// factory->AddSignalTree( signalTrainingTree, signalTrainWeight, "Training" );
// factory->AddSignalTree( signalTestTree, signalTestWeight, "Test" );
// Use the following code instead of the above two or four lines to add signal and background
// training and test events "by hand"
// NOTE that in this case one should not give expressions (such as "var1+var2") in the input
// variable definition, but simply compute the expression before adding the event
//
// // --- begin ----------------------------------------------------------
// std::vector<Double_t> vars( 4 ); // vector has size of number of input variables
// Float_t treevars[4], weight;
//
// // Signal
// for (UInt_t ivar=0; ivar<4; ivar++) signal->SetBranchAddress( Form( "var%i", ivar+1 ), &(treevars[ivar]) );
// for (UInt_t i=0; i<signal->GetEntries(); i++) {
// signal->GetEntry(i);
// for (UInt_t ivar=0; ivar<4; ivar++) vars[ivar] = treevars[ivar];
// // add training and test events; here: first half is training, second is testing
// // note that the weight can also be event-wise
// if (i < signal->GetEntries()/2.0) factory->AddSignalTrainingEvent( vars, signalWeight );
// else factory->AddSignalTestEvent ( vars, signalWeight );
// }
//
// // Background (has event weights)
// background->SetBranchAddress( "weight", &weight );
// for (UInt_t ivar=0; ivar<4; ivar++) background->SetBranchAddress( Form( "var%i", ivar+1 ), &(treevars[ivar]) );
// for (UInt_t i=0; i<background->GetEntries(); i++) {
// background->GetEntry(i);
// for (UInt_t ivar=0; ivar<4; ivar++) vars[ivar] = treevars[ivar];
// // add training and test events; here: first half is training, second is testing
// // note that the weight can also be event-wise
// if (i < background->GetEntries()/2) factory->AddBackgroundTrainingEvent( vars, backgroundWeight*weight );
// else factory->AddBackgroundTestEvent ( vars, backgroundWeight*weight );
// }
// --- end ------------------------------------------------------------
//
// --- end of tree registration
// Set individual event weights (the variables must exist in the original TTree)
// for signal : factory->SetSignalWeightExpression ("weight1*weight2");
// for background: factory->SetBackgroundWeightExpression("weight1*weight2");
//factory->SetBackgroundWeightExpression( "weight" );
// Apply additional cuts on the signal and background samples (can be different)
// Tell the factory how to use the training and testing events
//
// If no numbers of events are given, half of the events in the tree are used
// for training, and the other half for testing:
// factory->PrepareTrainingAndTestTree( mycut, "SplitMode=random:!V" );
// To also specify the number of testing events, use:
// factory->PrepareTrainingAndTestTree( mycut,
// "NSigTrain=3000:NBkgTrain=3000:NSigTest=3000:NBkgTest=3000:SplitMode=Random:!V" );
factory->PrepareTrainingAndTestTree( "", "SplitMode=Random:NormMode=NumEvents:!V" );
// ---- Book MVA methods
//
// Please lookup the various method configuration options in the corresponding cxx files, eg:
// src/MethoCuts.cxx, etc, or here: http://tmva.sourceforge.net/optionRef.html
// it is possible to preset ranges in the option string in which the cut optimisation should be done:
// "...:CutRangeMin[2]=-1:CutRangeMax[2]=1"...", where [2] is the third input variable
// Cut optimisation
if (Use["Cuts"])
factory->BookMethod( TMVA::Types::kCuts, "Cuts",
"!H:!V:FitMethod=MC:EffSel:SampleSize=200000:VarProp=FSmart" );
if (Use["CutsD"])
factory->BookMethod( TMVA::Types::kCuts, "CutsD",
"!H:!V:FitMethod=MC:EffSel:SampleSize=200000:VarProp=FSmart:VarTransform=Decorrelate" );
if (Use["CutsPCA"])
factory->BookMethod( TMVA::Types::kCuts, "CutsPCA",
"!H:!V:FitMethod=MC:EffSel:SampleSize=200000:VarProp=FSmart:VarTransform=PCA" );
if (Use["CutsGA"])
factory->BookMethod( TMVA::Types::kCuts, "CutsGA",
"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" );
if (Use["CutsSA"])
factory->BookMethod( TMVA::Types::kCuts, "CutsSA",
"!H:!V:FitMethod=SA:EffSel:MaxCalls=150000:KernelTemp=IncAdaptive:InitialTemp=1e+6:MinTemp=1e-6:Eps=1e-10:UseDefaultScale" );
// Likelihood ("naive Bayes estimator")
if (Use["Likelihood"])
factory->BookMethod( TMVA::Types::kLikelihood, "Likelihood",
"H:!V:TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmoothBkg[1]=10:NSmooth=1:NAvEvtPerBin=50" );
// Decorrelated likelihood
if (Use["LikelihoodD"])
factory->BookMethod( TMVA::Types::kLikelihood, "LikelihoodD",
"!H:!V:TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmooth=5:NAvEvtPerBin=50:VarTransform=Decorrelate" );
// PCA-transformed likelihood
if (Use["LikelihoodPCA"])
factory->BookMethod( TMVA::Types::kLikelihood, "LikelihoodPCA",
"!H:!V:!TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmooth=5:NAvEvtPerBin=50:VarTransform=PCA" );
// Use a kernel density estimator to approximate the PDFs
if (Use["LikelihoodKDE"])
factory->BookMethod( TMVA::Types::kLikelihood, "LikelihoodKDE",
"!H:!V:!TransformOutput:PDFInterpol=KDE:KDEtype=Gauss:KDEiter=Adaptive:KDEFineFactor=0.3:KDEborder=None:NAvEvtPerBin=50" );
// Use a variable-dependent mix of splines and kernel density estimator
if (Use["LikelihoodMIX"])
factory->BookMethod( TMVA::Types::kLikelihood, "LikelihoodMIX",
"!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" );
// Test the multi-dimensional probability density estimator
// here are the options strings for the MinMax and RMS methods, respectively:
// "!H:!V:VolumeRangeMode=MinMax:DeltaFrac=0.2:KernelEstimator=Gauss:GaussSigma=0.3" );
// "!H:!V:VolumeRangeMode=RMS:DeltaFrac=3:KernelEstimator=Gauss:GaussSigma=0.3" );
if (Use["PDERS"])
factory->BookMethod( TMVA::Types::kPDERS, "PDERS",
"!H:!V:NormTree=T:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600" );
if (Use["PDERSD"])
factory->BookMethod( TMVA::Types::kPDERS, "PDERSD",
"!H:!V:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600:VarTransform=Decorrelate" );
if (Use["PDERSPCA"])
factory->BookMethod( TMVA::Types::kPDERS, "PDERSPCA",
"!H:!V:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600:VarTransform=PCA" );
// Multi-dimensional likelihood estimator using self-adapting phase-space binning
if (Use["PDEFoam"])
factory->BookMethod( TMVA::Types::kPDEFoam, "PDEFoam",
"!H:!V:SigBgSeparate=F:TailCut=0.001:VolFrac=0.0666:nActiveCells=500:nSampl=2000:nBin=5:Nmin=100:Kernel=None:Compress=T" );
if (Use["PDEFoamBoost"])
factory->BookMethod( TMVA::Types::kPDEFoam, "PDEFoamBoost",
"!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" );
// K-Nearest Neighbour classifier (KNN)
if (Use["KNN"])
factory->BookMethod( TMVA::Types::kKNN, "KNN",
"H:nkNN=20:ScaleFrac=0.8:SigmaFact=1.0:Kernel=Gaus:UseKernel=F:UseWeight=T:!Trim" );
// H-Matrix (chi2-squared) method
if (Use["HMatrix"])
factory->BookMethod( TMVA::Types::kHMatrix, "HMatrix", "!H:!V:VarTransform=None" );
// Linear discriminant (same as Fisher discriminant)
if (Use["LD"])
factory->BookMethod( TMVA::Types::kLD, "LD", "H:!V:VarTransform=None:CreateMVAPdfs:PDFInterpolMVAPdf=Spline2:NbinsMVAPdf=50:NsmoothMVAPdf=10" );
// Fisher discriminant (same as LD)
if (Use["Fisher"])
factory->BookMethod( TMVA::Types::kFisher, "Fisher", "H:!V:Fisher:VarTransform=None:CreateMVAPdfs:PDFInterpolMVAPdf=Spline2:NbinsMVAPdf=50:NsmoothMVAPdf=10" );
// Fisher with Gauss-transformed input variables
if (Use["FisherG"])
factory->BookMethod( TMVA::Types::kFisher, "FisherG", "H:!V:VarTransform=Gauss" );
// Composite classifier: ensemble (tree) of boosted Fisher classifiers
if (Use["BoostedFisher"])
factory->BookMethod( TMVA::Types::kFisher, "BoostedFisher",
"H:!V:Boost_Num=20:Boost_Transform=log:Boost_Type=AdaBoost:Boost_AdaBoostBeta=0.2:!Boost_DetailedMonitoring" );
// Function discrimination analysis (FDA) -- test of various fitters - the recommended one is Minuit (or GA or SA)
if (Use["FDA_MC"])
factory->BookMethod( TMVA::Types::kFDA, "FDA_MC",
"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" );
if (Use["FDA_GA"]) // can also use Simulated Annealing (SA) algorithm (see Cuts_SA options])
factory->BookMethod( TMVA::Types::kFDA, "FDA_GA",
"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" );
if (Use["FDA_SA"]) // can also use Simulated Annealing (SA) algorithm (see Cuts_SA options])
factory->BookMethod( TMVA::Types::kFDA, "FDA_SA",
"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" );
if (Use["FDA_MT"])
factory->BookMethod( TMVA::Types::kFDA, "FDA_MT",
"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" );
if (Use["FDA_GAMT"])
factory->BookMethod( TMVA::Types::kFDA, "FDA_GAMT",
"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" );
if (Use["FDA_MCMT"])
factory->BookMethod( TMVA::Types::kFDA, "FDA_MCMT",
"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" );
// TMVA ANN: MLP (recommended ANN) -- all ANNs in TMVA are Multilayer Perceptrons
if (Use["MLP"])
factory->BookMethod( TMVA::Types::kMLP, "MLP", "H:!V:NeuronType=tanh:VarTransform=N:NCycles=600:HiddenLayers=N+5:TestRate=5:!UseRegulator" );
if (Use["MLPBFGS"])
factory->BookMethod( TMVA::Types::kMLP, "MLPBFGS", "H:!V:NeuronType=tanh:VarTransform=N:NCycles=600:HiddenLayers=N+5:TestRate=5:TrainingMethod=BFGS:!UseRegulator" );
if (Use["MLPBNN"])
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
// CF(Clermont-Ferrand)ANN
if (Use["CFMlpANN"])
factory->BookMethod( TMVA::Types::kCFMlpANN, "CFMlpANN", "!H:!V:NCycles=2000:HiddenLayers=N+1,N" ); // n_cycles:#nodes:#nodes:...
// Tmlp(Root)ANN
if (Use["TMlpANN"])
factory->BookMethod( TMVA::Types::kTMlpANN, "TMlpANN", "!H:!V:NCycles=200:HiddenLayers=N+1,N:LearningMethod=BFGS:ValidationFraction=0.3" ); // n_cycles:#nodes:#nodes:...
// Support Vector Machine
if (Use["SVM"])
factory->BookMethod( TMVA::Types::kSVM, "SVM", "Gamma=0.25:Tol=0.001:VarTransform=Norm" );
// Boosted Decision Trees
if (Use["BDTG"]) // Gradient Boost
factory->BookMethod( TMVA::Types::kBDT, "BDTG",
"!H:!V:NTrees=1000:BoostType=Grad:Shrinkage=0.10:UseBaggedGrad:GradBaggingFraction=0.50:nCuts=20:NNodesMax=8" );
// "!H:!V:NTrees=1000:MinNodeSize=1.5%:BoostType=Grad:Shrinkage=0.10:UseBaggedGrad:GradBaggingFraction=0.5:nCuts=20:MaxDepth=2" );
if (Use["BDT"]) // Adaptive Boost
factory->BookMethod( TMVA::Types::kBDT, "BDT",
"!H:!V:NTrees=850:MinNodeSize=2.5%:MaxDepth=3:BoostType=AdaBoost:AdaBoostBeta=0.5:SeparationType=GiniIndex:nCuts=20" );
if (Use["BDTB"]) // Bagging
factory->BookMethod( TMVA::Types::kBDT, "BDTB",
"!H:!V:NTrees=400:BoostType=Bagging:SeparationType=GiniIndex:nCuts=20" );
if (Use["BDTD"]) // Decorrelation + Adaptive Boost
factory->BookMethod( TMVA::Types::kBDT, "BDTD",
"!H:!V:NTrees=400:MinNodeSize=5%:MaxDepth=3:BoostType=AdaBoost:SeparationType=GiniIndex:nCuts=20:VarTransform=Decorrelate" );
if (Use["BDTF"]) // Allow Using Fisher discriminant in node splitting for (strong) linearly correlated variables
factory->BookMethod( TMVA::Types::kBDT, "BDTMitFisher",
"!H:!V:NTrees=50:MinNodeSize=2.5%:UseFisherCuts:MaxDepth=3:BoostType=AdaBoost:AdaBoostBeta=0.5:SeparationType=GiniIndex:nCuts=20" );
// RuleFit -- TMVA implementation of Friedman's method
if (Use["RuleFit"])
factory->BookMethod( TMVA::Types::kRuleFit, "RuleFit",
"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" );
// For an example of the category classifier usage, see: TMVAClassificationCategory
// --------------------------------------------------------------------------------------------------
// ---- Now you can optimize the setting (configuration) of the MVAs using the set of training events
// factory->OptimizeAllMethods("SigEffAt001","Scan");
// factory->OptimizeAllMethods("ROCIntegral","FitGA");
// --------------------------------------------------------------------------------------------------
// ---- Now you can tell the factory to train, test, and evaluate the MVAs
// Train MVAs using the set of training events
factory->TrainAllMethods();
//
// // ---- Evaluate all MVAs using the set of test events
factory->TestAllMethods();
//
// // ----- Evaluate and compare performance of all configured MVAs
factory->EvaluateAllMethods();
// --------------------------------------------------------------
// Save the output
outputFile->Close();
std::cout << "==> Wrote root file: " << outputFile->GetName() << std::endl;
std::cout << "==> TMVAClassification is done!" << std::endl;
delete factory;
// Launch the GUI for the root macros
// if (!gROOT->IsBatch()) TMVAGui( outfileName );
}
int main(int argc, char** argv) {
if(argc != 2)
{
std::cerr << ">>>>> analysis.cpp::usage: " << argv[0] << " configFileName MVAconfigFileName" << std::endl ;
return 1;
}
// Parse the config file
parseConfigFile (argv[1]) ;
std::string treeName = gConfigParser -> readStringOption("Input::treeName");
std::string fileSamples = gConfigParser -> readStringOption("Input::fileSamples");
std::string inputDirectory = gConfigParser -> readStringOption("Input::inputDirectory");
std::string inputBeginningFile = "out_NtupleProducer_";
try {
inputBeginningFile = gConfigParser -> readStringOption("Input::inputBeginningFile");
}
catch (char const* exceptionString) {
std::cerr << " exception = " << exceptionString << std::endl;
}
std::cout << ">>>>> Input::inputBeginningFile " << inputBeginningFile << std::endl;
double LUMI = gConfigParser -> readDoubleOption("Options::Lumi");
std::vector<std::string> SignalName;
SignalName = gConfigParser -> readStringListOption("Options::SignalName");
for (int iSignalSample=0; iSignalSample<SignalName.size(); iSignalSample++) {
std::cout << " Signal[" << iSignalSample << "] = " << SignalName.at(iSignalSample) << std::endl;
}
std::string nameWeight = "1";
try {
nameWeight = gConfigParser -> readStringOption("Options::nameWeight");
}
catch (char const* exceptionString) {
std::cerr << " exception = " << exceptionString << std::endl;
}
std::cout << ">>>>> Input::nameWeight " << nameWeight << std::endl;
TTree *treeJetLepVect[200];
char *nameSample[1000];
char *nameHumanReadable[1000];
char* xsectionName[1000];
char nameFileIn[1000];
sprintf(nameFileIn,"%s",fileSamples.c_str());
int numberOfSamples = ReadFile(nameFileIn, nameSample, nameHumanReadable, xsectionName);
double Normalization[1000];
double xsection[1000];
for (int iSample=0; iSample<numberOfSamples; iSample++) {
xsection[iSample] = atof(xsectionName[iSample]);
}
for (int iSample=0; iSample<numberOfSamples; iSample++) {
char nameFile[20000];
sprintf(nameFile,"%s/%s%s.root",inputDirectory.c_str(),inputBeginningFile.c_str(),nameSample[iSample]);
TFile* f = new TFile(nameFile, "READ");
treeJetLepVect[iSample] = (TTree*) f->Get(treeName.c_str());
char nameTreeJetLep[100];
sprintf(nameTreeJetLep,"treeJetLep_%d",iSample);
treeJetLepVect[iSample]->SetName(nameTreeJetLep);
double XSection;
XSection = xsection[iSample];
Normalization[iSample] = XSection * LUMI / 1000.;
}
//==== cut
std::string CutFile = gConfigParser -> readStringOption("Selections::CutFile");
std::vector<std::string> vCut;
std::cout << " nCuts = " << ReadFileCut(CutFile, vCut) << std::endl;
std::string Cut;
if (vCut.size() != 0) {
Cut = vCut.at(0);
}
else {
Cut = "1";
}
//==== HiggsMass
std::string HiggsMass = gConfigParser -> readStringOption("Options::HiggsMass");
//==== list of methods
std::vector<std::string> vectorMyMethodList = gConfigParser -> readStringListOption("Options::MVAmethods");
TString myMethodList;
for (int iMVA = 0; iMVA < vectorMyMethodList.size(); iMVA++) {
if (iMVA == 0) myMethodList = Form ("%s",vectorMyMethodList.at(iMVA).c_str());
else myMethodList = Form ("%s,%s",myMethodList.Data(),vectorMyMethodList.at(iMVA).c_str());
}
//==== output
TString outfileName = gConfigParser -> readStringOption("Output::outFileName");
// This loads the library
TMVA::Tools::Instance();
// Default MVA methods to be trained + tested
std::map<std::string,int> Use;
Use["MLP"] = 1;
Use["BDTG"] = 1;
Use["FDA_GA"] = 0;
Use["PDEFoam"] = 0;
std::cout << std::endl;
std::cout << "==> Start TMVAClassification" << std::endl;
// Select methods (don't look at this code - not of interest)
if (myMethodList != "") {
for (std::map<std::string,int>::iterator it = Use.begin(); it != Use.end(); it++) it->second = 0;
std::vector<TString> mlist = TMVA::gTools().SplitString( myMethodList, ',' );
for (UInt_t i=0; i<mlist.size(); i++) {
std::string regMethod(mlist[i]);
if (Use.find(regMethod) == Use.end()) {
std::cout << "Method \"" << regMethod << "\" not known in TMVA under this name. Choose among the following:" << std::endl;
for (std::map<std::string,int>::iterator it = Use.begin(); it != Use.end(); it++) std::cout << it->first << " ";
std::cout << std::endl;
return 0;
}
Use[regMethod] = 1;
}
}
// --------------------------------------------------------------------------------------------------
// --- Here the preparation phase begins
// Create a new root output file
TFile* outputFile = TFile::Open( outfileName, "RECREATE" );
// TMVA::Factory *factory = new TMVA::Factory( "TMVAMulticlass", outputFile, "AnalysisType=multiclass:!V:!Silent:!V:Transformations=I;D" );
TMVA::Factory *factory = new TMVA::Factory( "TMVAMulticlass", outputFile, "!V:!Silent:Color:DrawProgressBar:Transformations=I;D;P;G,D:AnalysisType=multiclass" );
factory->AddVariable( "jetpt1" , 'F');
factory->AddVariable( "jetpt2" , 'F');
factory->AddVariable( "mjj" , 'F');
factory->AddVariable( "detajj" , 'F');
factory->AddVariable( "dphilljetjet" , 'F');
factory->AddVariable( "pt1" , 'F');
factory->AddVariable( "pt2" , 'F');
factory->AddVariable( "mll" , 'F');
factory->AddVariable( "dphill" , 'F');
factory->AddVariable( "mth" , 'F');
factory->AddVariable( "dphillmet" , 'F');
factory->AddVariable( "mpmet" , 'F');
factory->AddSpectator( "channel" , 'F');
for (int iSample=0; iSample<numberOfSamples; iSample++) {
int numEnt = treeJetLepVect[iSample]->GetEntries(Cut.c_str());
std::cout << " Sample = " << nameSample[iSample] << " ~ " << nameHumanReadable[iSample] << " --> " << numEnt << std::endl;
if (numEnt != 0) {
if (iSample == 0) factory->AddTree( treeJetLepVect[iSample], "Signal", Normalization[iSample] );
else if (iSample == 1) factory->AddTree( treeJetLepVect[iSample], "Background", Normalization[iSample] );
else factory->AddTree( treeJetLepVect[iSample], TString(nameHumanReadable[iSample]), Normalization[iSample] );
// factory->AddTree( treeJetLepVect[iSample], TString(nameHumanReadable[iSample]), Normalization[iSample] );
// factory->AddTree( treeJetLepVect[iSample], TString(nameHumanReadable[iSample]), Normalization[iSample] , nameWeight.c_str());
// factory->AddTree( treeJetLepVect[iSample], TString(nameHumanReadable[iSample]));
}
}
// for (int iSample=0; iSample<numberOfSamples; iSample++){
// int numEnt = treeJetLepVect[iSample]->GetEntries(Cut.c_str());
// std::cout << " Sample = " << nameSample[iSample] << " ~ " << nameHumanReadable[iSample] << " --> " << numEnt << std::endl;
// if (numEnt != 0) {
// bool isSig = false;
// for (std::vector<std::string>::const_iterator itSig = SignalName.begin(); itSig != SignalName.end(); itSig++){
// if (nameHumanReadable[iSample] == *itSig) isSig = true;
// }
// if (isSig) {
// factory->AddTree( treeJetLepVect[iSample], TString("Signal"), Normalization[iSample] ); //---> ci deve essere uno chiamato Signal!
// }
// else {
// factory->AddTree( treeJetLepVect[iSample], TString(nameHumanReadable[iSample]), Normalization[iSample] );
// }
// }
// }
//
// for (int iSample=0; iSample<numberOfSamples; iSample++){
// int numEnt = treeJetLepVect[iSample]->GetEntries(Cut.c_str());
// std::cout << " Sample = " << nameSample[iSample] << " ~ " << nameHumanReadable[iSample] << " --> " << numEnt << std::endl;
// if (numEnt != 0) {
// bool isSig = false;
// for (std::vector<std::string>::const_iterator itSig = SignalName.begin(); itSig != SignalName.end(); itSig++){
// if (nameHumanReadable[iSample] == *itSig) isSig = true;
// }
// if (isSig) {
// // factory->AddTree( treeJetLepVect[iSample], TString("Signal"), Normalization[iSample] ); //---> ci deve essere uno chiamato Signal!
// }
// else {
// factory->AddTree( treeJetLepVect[iSample], TString(nameHumanReadable[iSample]), Normalization[iSample] );
// }
// }
// }
std::cerr << " AAAAAAAAAAAAAAAAAAAAAAAAAAAAA " << std::endl;
TCut mycuts = Cut.c_str();
// factory->SetWeightExpression( nameWeight.c_str() );
// factory->SetBackgroundWeightExpression( nameWeight.c_str() );
// factory->SetSignalWeightExpression ( nameWeight.c_str() );
std::cerr << " BBBBBBBBBBBBBBBBBBBBBBBBBBBBB " << std::endl;
factory->PrepareTrainingAndTestTree( mycuts ,"SplitMode=Random:NormMode=None:!V");
// factory->PrepareTrainingAndTestTree( "" ,"SplitMode=Random:NormMode=None:!V");
std::cerr << " CCCCCCCCCCCCCCCCCCCCCCCCCCCCC " << std::endl;
// gradient boosted decision trees
// if (Use["BDTG"]) factory->BookMethod( TMVA::Types::kBDT, "BDTG", "!H:!V:NTrees=1000:BoostType=Grad:Shrinkage=0.10:UseBaggedGrad:GradBaggingFraction=0.50:nCuts=20:NNodesMax=8");
if (Use["BDTG"]) factory->BookMethod( TMVA::Types::kBDT, "BDTG", "!H:!V:NTrees=600:BoostType=Grad:Shrinkage=0.10:UseBaggedGrad:GradBaggingFraction=0.50:nCuts=20:NNodesMax=8");
// neural network
if (Use["MLP"]) factory->BookMethod( TMVA::Types::kMLP, "MLP", "!H:!V:NeuronType=tanh:NCycles=1000:HiddenLayers=N+5,5:TestRate=5:EstimatorType=MSE");
// functional discriminant with GA minimizer
if (Use["FDA_GA"]) factory->BookMethod( TMVA::Types::kFDA, "FDA_GA", "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" );
// PDE-Foam approach
if (Use["PDEFoam"]) factory->BookMethod( TMVA::Types::kPDEFoam, "PDEFoam", "!H:!V:TailCut=0.001:VolFrac=0.0666:nActiveCells=500:nSampl=2000:nBin=5:Nmin=100:Kernel=None:Compress=T" );
//==== Optimize parameters in MVA methods
// factory->OptimizeAllMethods();
// factory->OptimizeAllMethods("ROCIntegral","Scan");
//==== Train MVAs using the set of training events ====
factory->TrainAllMethods();
//==== Evaluate all MVAs using the set of test events ====
factory->TestAllMethods();
//==== Evaluate and compare performance of all configured MVAs ====
factory->EvaluateAllMethods();
// --------------------------------------------------------------
// Save the output
outputFile->Close();
std::cout << "==> Wrote root file: " << outputFile->GetName() << std::endl;
std::cout << "==> TMVAnalysis is done!" << std::endl;
delete factory;
//==== change position of weights file
std::string toDo;
toDo = "rm -r Weights-MVA-MultiClass/weights_" + HiggsMass + "_testVariables";
std::cerr << "toDo = " << toDo << std::endl;
system (toDo.c_str());
toDo = "mv weights Weights-MVA-MultiClass/weights_" + HiggsMass + "_testVariables";
std::cerr << "toDo = " << toDo << std::endl;
system (toDo.c_str());
// Launch the GUI for the root macros
// if (!gROOT->IsBatch()) TMVAGui( outfileName );
}
void TMVA_stop( TString signal_name = "T2tt", int train_region = 1, float x_parameter = 0.25)
{
// The explicit loading of the shared libTMVA is done in TMVAlogon.C, defined in .rootrc
// if you use your private .rootrc, or run from a different directory, please copy the
// corresponding lines from .rootrc
// methods to be processed can be given as an argument; use format:
//
// mylinux~> root -l TMVA_stop.C\(\"myMethod1,myMethod2,myMethod3\"\)
//
// if you like to use a method via the plugin mechanism, we recommend using
//
// mylinux~> root -l TMVA_stop.C\(\"P_myMethod\"\)
// (an example is given for using the BDT as plugin (see below),
// but of course the real application is when you write your own
// method based)
//-----------------------------------------------------
// define event selection (store in TCut sel)
//-----------------------------------------------------
TCut njets4("mini_njets>=4");
TCut met100("mini_met>=100");
TCut mt120("mini_mt>=120");
TCut nb1("mini_nb>=1");
TCut isotrk("mini_passisotrk==1");
TCut lep_pt30("mini_nlep>=1 && mini_lep1pt>30.0");
TCut sig("mini_sig==1");
TCut sel0 = njets4 + met100 + mt120 + nb1 + isotrk + lep_pt30 + sig;
cout << "Using selection : " << sel0.GetTitle() << endl;
cout << "Doing signal point : " << train_region << endl;
//-----------------------------------------------------
// choose which variables to include in MVA training
//-----------------------------------------------------
std::map<std::string,int> mvaVar;
mvaVar[ "met" ] = 1;
mvaVar[ "lep1pt" ] = 0;
mvaVar[ "mt2w" ] = 1;
mvaVar[ "htratiom" ] = 1;
mvaVar[ "chi2" ] = 1;
mvaVar[ "dphimjmin" ] = 1;
mvaVar[ "pt_b" ] = 0;
mvaVar[ "nb" ] = 0;
mvaVar[ "pt_J1" ] = 0;
mvaVar[ "pt_J2" ] = 0;
mvaVar[ "rand" ] = 0;
mvaVar[ "mt" ] = 0;
mvaVar[ "mt2bl" ] = 0;
mvaVar[ "mt2b" ] = 0;
mvaVar[ "lep1eta" ] = 0;
mvaVar[ "thrjetlm" ] = 0;
mvaVar[ "apljetlm" ] = 0;
mvaVar[ "sphjetlm" ] = 0;
mvaVar[ "cirjetlm" ] = 0;
mvaVar[ "chi2min" ] = 0;
mvaVar[ "chi2min_mt2b" ] = 0;
mvaVar[ "chi2min_mt2bl" ] = 0;
mvaVar[ "chi2min_mt2w" ] = 0;
mvaVar[ "mt2bmin" ] = 0;
mvaVar[ "mt2blmin" ] = 0;
mvaVar[ "mt2wmin_chi2" ] = 0;
mvaVar[ "mt2bmin_chi2" ] = 0;
mvaVar[ "mt2blmin_chi2" ] = 0;
mvaVar[ "mt2wmin_chi2prob" ] = 0;
mvaVar[ "mt2bmin_chi2prob" ] = 0;
mvaVar[ "mt2blmin_chi2prob" ] = 0;
mvaVar[ "htratiol" ] = 0;
mvaVar[ "dphimj1" ] = 0;
mvaVar[ "dphimj2" ] = 0;
mvaVar[ "metsig" ] = 0;
//---------------------------------
//choose bkg samples to include
//---------------------------------
cout << "Background trees: " << endl;
int n_backgrounds = 8;
TString backgrounds[] = {"ttdl_powheg", "ttsl_powheg", "w1to4jets", "tW_lep", "triboson", "diboson", "ttV", "DY1to4Jtot" };
TString bkgPath = "/nfs-3/userdata/stop/Train/V00-02-18__V00-03-00_4jetsMET100_bkg/";
TChain* chBackground = new TChain("t");
for (int i = 0; i < n_backgrounds; i++) {
TString backgroundChain = bkgPath + "/" + backgrounds[i] + ".root";
cout << " " << backgroundChain << endl;
chBackground ->Add(backgroundChain );
}
//---------------------------------
//choose signal sample to include
//---------------------------------
cout << "Signal trees: " << endl;
TString s_train_region = "";
s_train_region += train_region;
TString s_x_parameter = "";
s_x_parameter = Form("%.2f",x_parameter);
TString signalPath = "/nfs-3/userdata/stop/Train/";
TString signalVersion = "V00-02-18__V00-03-00_4jetsMET100_";
TChain *chSignal = new TChain("t");
TString base_name = signalPath + "/" + signalVersion + signal_name + "/" + signal_name + "_" + s_train_region;
if (signal_name == "T2bw") base_name = base_name + "_" + s_x_parameter;
TString signalChain = base_name + ".root" ;
cout << " " << signalChain << endl;
chSignal->Add(signalChain);
//-----------------------------------------------------
// choose backgrounds to include for multiple outputs
//-----------------------------------------------------
// bool doMultipleOutputs = false;
// TChain *chww = new TChain("Events");
// chww->Add(Form("%s/WWTo2L2Nu_PU_testFinal_baby.root",babyPath));
// chww->Add(Form("%s/GluGluToWWTo4L_PU_testFinal_baby.root",babyPath));
// TChain *chwjets = new TChain("Events");
// chwjets->Add(Form("%s/WJetsToLNu_PU_testFinal_baby.root",babyPath));
// TChain *chtt = new TChain("Events");
// chtt->Add(Form("%s/TTJets_PU_testFinal_baby.root",babyPath));
// std::map<std::string,int> includeBkg;
// includeBkg["ww"] = 1;
// includeBkg["wjets"] = 0;
// includeBkg["tt"] = 0;
//---------------------------------------------------------------
// This loads the library
TMVA::Tools::Instance();
// Default MVA methods to be trained + tested
std::map<std::string,int> Use;
// --- Cut optimisation
Use["Cuts"] = 0;
Use["CutsD"] = 0;
Use["CutsPCA"] = 0;
Use["CutsGA"] = 0;
Use["CutsSA"] = 0;
//
// --- 1-dimensional likelihood ("naive Bayes estimator")
Use["Likelihood"] = 0;
Use["LikelihoodD"] = 0; // the "D" extension indicates decorrelated input variables (see option strings)
Use["LikelihoodPCA"] = 0; // the "PCA" extension indicates PCA-transformed input variables (see option strings)
Use["LikelihoodKDE"] = 0;
Use["LikelihoodMIX"] = 0;
//
// --- Mutidimensional likelihood and Nearest-Neighbour methods
Use["PDERS"] = 0;
Use["PDERSD"] = 0;
Use["PDERSPCA"] = 0;
Use["PDEFoam"] = 0;
Use["PDEFoamBoost"] = 0; // uses generalised MVA method boosting
Use["KNN"] = 0; // k-nearest neighbour method
//
// --- Linear Discriminant Analysis
Use["LD"] = 0; // Linear Discriminant identical to Fisher
Use["Fisher"] = 0;
Use["FisherG"] = 0;
Use["BoostedFisher"] = 0; // uses generalised MVA method boosting
Use["HMatrix"] = 0;
//
// --- Function Discriminant analysis
Use["FDA_GA"] = 0; // minimisation of user-defined function using Genetics Algorithm
Use["FDA_SA"] = 0;
Use["FDA_MC"] = 0;
Use["FDA_MT"] = 0;
Use["FDA_GAMT"] = 0;
Use["FDA_MCMT"] = 0;
//
// --- Neural Networks (all are feed-forward Multilayer Perceptrons)
Use["MLP"] = 0; // Recommended ANN
Use["MLPBFGS"] = 0; // Recommended ANN with optional training method
Use["MLPBNN"] = 0; // Recommended ANN with BFGS training method and bayesian regulator
Use["CFMlpANN"] = 0; // Depreciated ANN from ALEPH
Use["TMlpANN"] = 0; // ROOT's own ANN
//
// --- Support Vector Machine
Use["SVM"] = 0;
//
// --- Boosted Decision Trees
Use["BDT"] = 1; // uses Adaptive Boost
Use["BDT1"] = 0; // uses Adaptive Boost
Use["BDTG"] = 0; // uses Gradient Boost
Use["BDTB"] = 0; // uses Bagging
Use["BDTD"] = 0; // decorrelation + Adaptive Boost
//
// --- Friedman's RuleFit method, ie, an optimised series of cuts ("rules")
Use["RuleFit"] = 0;
//
// --- multi-output MVA's
Use["multi_BDTG"] = 0;
Use["multi_MLP"] = 0;
Use["multi_FDA_GA"] = 0;
//
// ---------------------------------------------------------------
std::cout << std::endl;
std::cout << "==> Start TMVAClassification" << std::endl;
// --- Here the preparation phase begins
// Create a ROOT output file where TMVA will store ntuples, histograms, etc.
TString outfileName = "TMVA_" + signal_name + "_" + s_train_region;
if (signal_name == "T2bw") outfileName = outfileName +"_" + s_x_parameter;
outfileName += ".root";
TFile* outputFile = TFile::Open( outfileName, "RECREATE" );
TString classification_name = "classification_" + signal_name + "_" + s_train_region;
if (signal_name == "T2bw") classification_name = classification_name +"_" + s_x_parameter;
/*
TString multioutfileName( "TMVA_HWW_multi.root" );
TFile* multioutputFile;
if( doMultipleOutputs )
multioutputFile = TFile::Open( multioutfileName, "RECREATE" );
*/
// Create the factory object. Later you can choose the methods
// whose performance you'd like to investigate. The factory is
// the only TMVA object you have to interact with
//
// The first argument is the base of the name of all the
// weightfiles in the directory weight/
//
// The second argument is the output file for the training results
// All TMVA output can be suppressed by removing the "!" (not) in
// front of the "Silent" argument in the option string
TMVA::Factory *factory = new TMVA::Factory( classification_name, outputFile,
"!V:!Silent:Color:DrawProgressBar:Transformations=I;D;P;G,D:AnalysisType=Classification" );
/*
TMVA::Factory *multifactory;
if( doMultipleOutputs )
multifactory= new TMVA::Factory( "TMVAMulticlass", multioutputFile,
"!V:!Silent:Color:DrawProgressBar:Transformations=I;D;P;G,D:AnalysisType=multiclass" );
*/
// If you wish to modify default settings
// (please check "src/Config.h" to see all available global options)
// (TMVA::gConfig().GetVariablePlotting()).fTimesRMS = 8.0;
// (TMVA::gConfig().GetIONames()).fWeightFileDir = "myWeightDirectory";
// Define the input variables that shall be used for the MVA training
// note that you may also use variable expressions, such as: "3*var1/var2*abs(var3)"
// [all types of expressions that can also be parsed by TTree::Draw( "expression" )]
//factory->AddVariable( "myvar1 := var1+var2", 'F' );
//factory->AddVariable( "myvar2 := var1-var2", "Expression 2", "", 'F' );
//factory->AddVariable( "var3", "Variable 3", "units", 'F' );
//factory->AddVariable( "var4", "Variable 4", "units", 'F' );
//--------------------------------------------------------
// choose which variables to include in training
//--------------------------------------------------------
if( mvaVar[ "met" ] == 1 ) factory->AddVariable( "mini_met" , "E_{T}^{miss}" , "GeV", 'F' );
if( mvaVar[ "mt" ] == 1 ) factory->AddVariable( "mini_mt" , "M_{T}" , "GeV", 'F' );
if( mvaVar[ "mt2w" ] == 1 ) factory->AddVariable( "mini_mt2w" , "MT2W" , "GeV", 'F' );
if( mvaVar[ "mt2bl" ] == 1 ) factory->AddVariable( "mini_mt2bl" , "MT2bl" , "GeV", 'F' );
if( mvaVar[ "mt2b" ] == 1 ) factory->AddVariable( "mini_mt2b" , "MT2b" , "GeV", 'F' );
if( mvaVar[ "chi2" ] == 1 ) factory->AddVariable( "mini_chi2" , "chi2" , "" , 'F' );
if( mvaVar[ "lep1pt" ] == 1 ) factory->AddVariable( "mini_lep1pt" , "lepton pt" , "" , 'F' );
if( mvaVar[ "lep1eta" ] == 1 ) factory->AddVariable( "mini_lep1eta" , "lepton eta" , "" , 'F' );
if( mvaVar[ "thrjetlm" ] == 1 ) factory->AddVariable( "mini_thrjetlm" , "thrust" , "" , 'F' );
if( mvaVar[ "apljetlm" ] == 1 ) factory->AddVariable( "mini_apljetlm" , "aplanarity" , "" , 'F' );
if( mvaVar[ "sphjetlm" ] == 1 ) factory->AddVariable( "mini_sphjetlm" , "sphericity" , "" , 'F' );
if( mvaVar[ "cirjetlm" ] == 1 ) factory->AddVariable( "mini_cirjetlm" , "circularity" , "" , 'F' );
if( mvaVar[ "chi2min" ] == 1 ) factory->AddVariable( "mini_min(chi2min,100)" , "#chi^{2}_{min}" , "" , 'F' );
if( mvaVar[ "chi2minprob" ] == 1 ) factory->AddVariable( "mini_chi2minprob" , "Prob(#chi^{2}_{min})" , "" , 'F' );
if( mvaVar[ "chi2min_mt2b" ] == 1 ) factory->AddVariable( "mini_chi2min_mt2b" , "MT2b(#chi^{2}_{min})" , "" , 'F' );
if( mvaVar[ "chi2min_mt2bl" ] == 1 ) factory->AddVariable( "mini_chi2min_mt2bl" , "MT2bl(#chi^{2}_{min})" , "" , 'F' );
if( mvaVar[ "chi2min_mt2w" ] == 1 ) factory->AddVariable( "mini_chi2min_mt2w" , "MT2W(#chi^{2}_{min})" , "" , 'F' );
if( mvaVar[ "mt2bmin" ] == 1 ) factory->AddVariable( "mini_mt2bmin" , "MT2b_{min}" , "" , 'F' );
if( mvaVar[ "mt2blmin" ] == 1 ) factory->AddVariable( "mini_mt2blmin" , "MT2bl_{min}" , "" , 'F' );
if( mvaVar[ "mt2wmin" ] == 1 ) factory->AddVariable( "mini_mt2wmin" , "MT2W_{min}" , "" , 'F' );
if( mvaVar[ "mt2bmin_chi2" ] == 1 ) factory->AddVariable( "min(mt2bmin_chi2,100)" , "#chi^{2}(MT2b_{min})" , "" , 'F' );
if( mvaVar[ "mt2blmin_chi2" ] == 1 ) factory->AddVariable( "min(mt2blmin_chi2,100)" , "#chi^{2}(MT2bl_{min})" , "" , 'F' );
if( mvaVar[ "mt2wmin_chi2" ] == 1 ) factory->AddVariable( "min(mt2wmin_chi2,100)" , "#chi^{2}(MT2W_{min})" , "" , 'F' );
if( mvaVar[ "mt2bmin_chi2prob" ] == 1 ) factory->AddVariable( "mt2bmin_chi2prob" , "Prob(#chi^{2}(MT2b_{min}))" , "" , 'F' );
if( mvaVar[ "mt2blmin_chi2prob" ] == 1 ) factory->AddVariable( "mt2blmin_chi2prob" , "Prob(#chi^{2}(MT2bl_{min}))" , "" , 'F' );
if( mvaVar[ "mt2wmin_chi2prob" ] == 1 ) factory->AddVariable( "mt2wmin_chi2prob" , "Prob(#chi^{2}(MT2W_{min}))" , "" , 'F' );
if( mvaVar[ "htratiol" ] == 1 ) factory->AddVariable( "mini_htssl/(mini_htosl+mini_htssl)" , "H_{T}^{SSL}/H_{T}" , "" , 'F' );
if( mvaVar[ "htratiom" ] == 1 ) factory->AddVariable( "mini_htssm/(mini_htosm+mini_htssm)" , "H_{T}^{SSM}/H_{T}" , "" , 'F' );
if( mvaVar[ "dphimj1" ] == 1 ) factory->AddVariable( "mini_dphimj1" , "#Delta#phi(j1,E_{T}^{miss})", "" , 'F' );
if( mvaVar[ "dphimj2" ] == 1 ) factory->AddVariable( "mini_dphimj2" , "#Delta#phi(j2,E_{T}^{miss})", "" , 'F' );
if( mvaVar[ "dphimjmin" ] == 1 ) factory->AddVariable( "mini_dphimjmin" , "min(#Delta#phi(j_{1,2},E_{T}^{miss}))", "" , 'F' );
if( mvaVar[ "rand" ] == 1 ) factory->AddVariable( "mini_rand" , "random(0,1)" , "" , 'F' );
if( mvaVar[ "metsig" ] == 1 ) factory->AddVariable( "met/sqrt(htosl+htssl)" , "E_{T}^{miss}/#sqrt{H_{T}}" , "#sqrt{GeV}" , 'F' )
;
if( mvaVar[ "pt_b" ] == 1 ) factory->AddVariable( "mini_pt_b" , "P_T(b) GeV" , 'F' );
if( mvaVar[ "nb" ] == 1 ) factory->AddVariable( "mini_nb" , "P_T(b) GeV" , 'F' );
if( mvaVar[ "pt_J1" ] == 1 ) factory->AddVariable( "pt_J1" , "P_T(J1) GeV" , 'F' );
if( mvaVar[ "pt_J2" ] == 1 ) factory->AddVariable( "pt_J2" , "P_T(J2) GeV" , 'F' );
/*
if( doMultipleOutputs ){
if (mvaVar["lephard_pt"]) multifactory->AddVariable( "lephard_pt", "1st lepton pt", "GeV", 'F' );
if (mvaVar["lepsoft_pt"]) multifactory->AddVariable( "lepsoft_pt", "2nd lepton pt", "GeV", 'F' );
if (mvaVar["dil_dphi"]) multifactory->AddVariable( "dil_dphi", "dphi(ll)", "", 'F' );
if (mvaVar["dil_mass"]) multifactory->AddVariable( "dil_mass", "M(ll)", "GeV", 'F' );
if (mvaVar["event_type"]) multifactory->AddVariable( "event_type", "Dil Flavor Type", "", 'F' );
if (mvaVar["met_projpt"]) multifactory->AddVariable( "met_projpt", "Proj. MET", "GeV", 'F' );
if (mvaVar["met_pt"]) multifactory->AddVariable( "met_pt", "MET", "GeV", 'F' );
if (mvaVar["mt_lephardmet"]) multifactory->AddVariable( "mt_lephardmet", "MT(lep1,MET)", "GeV", 'F' );
if (mvaVar["mt_lepsoftmet"]) multifactory->AddVariable( "mt_lepsoftmet", "MT(lep2,MET)", "GeV", 'F' );
if (mvaVar["mthiggs"]) multifactory->AddVariable( "mthiggs", "MT(Higgs)", "GeV", 'F' );
if (mvaVar["dphi_lephardmet"]) multifactory->AddVariable( "dphi_lephardmet", "dphi(lep1,MET)", "GeV", 'F' );
if (mvaVar["dphi_lepsoftmet"]) multifactory->AddVariable( "dphi_lepsoftmet", "dphi(lep2,MET)", "GeV", 'F' );
if (mvaVar["lepsoft_fbrem"]) multifactory->AddVariable( "lepsoft_fbrem", "2nd lepton f_{brem}", "", 'F' );
if (mvaVar["lepsoft_eOverPIn"]) multifactory->AddVariable( "lepsoft_eOverPIn", "2nd lepton E/p", "", 'F' );
if (mvaVar["lepsoft_qdphi"]) multifactory->AddVariable( "lepsoft_q * lepsoft_dPhiIn", "2nd lepton q#times#Delta#phi", "", 'F' );
}
*/
// You can add so-called "Spectator variables", which are not used in the MVA training,
// but will appear in the final "TestTree" produced by TMVA. This TestTree will contain the
// input variables, the response values of all trained MVAs, and the spectator variables
//factory->AddSpectator( "spec1 := var1*2", "Spectator 1", "units", 'F' );
//factory->AddSpectator( "spec2 := var1*3", "Spectator 2", "units", 'F' );
// TTree* signalTrainingTree = (TTree*) chSignalTrain;
// TTree* signalTestTree = (TTree*) chSignalTest;
//
// TTree* bkgTrainingTree = (TTree*) chBkgTrain;
// TTree* bkgTestTree = (TTree*) chBkgTest;
// std::cout << "--- TMVAClassification : Using bkg input files: -------------------" << std::endl;
//
// TObjArray *listOfBkgFiles = chbackground->GetListOfFiles();
// TIter bkgFileIter(listOfBkgFiles);
// TChainElement* currentBkgFile = 0;
//
// while((currentBkgFile = (TChainElement*)bkgFileIter.Next())) {
// std::cout << currentBkgFile->GetTitle() << std::endl;
// }
//
// std::cout << "--- TMVAClassification : Using sig input files: -------------------" << std::endl;
//
// TObjArray *listOfSigFiles = chsignal->GetListOfFiles();
// TIter sigFileIter(listOfSigFiles);
// TChainElement* currentSigFile = 0;
//
// while((currentSigFile = (TChainElement*)sigFileIter.Next())) {
// std::cout << currentSigFile->GetTitle() << std::endl;
// }
// global event weights per tree (see below for setting event-wise weights)
Double_t signalWeight = 1.0;
Double_t backgroundWeight = 1.0;
// You can add an arbitrary number of signal or background trees
// factory->AddSignalTree ( chSignal, signalWeight );
// factory->AddBackgroundTree( chBackground, backgroundWeight );
factory->AddTree(chSignal, "Signal", signalWeight, sel0+"mini_rand < 0.5", "train");
factory->AddTree(chSignal, "Signal", signalWeight, sel0+"mini_rand >= 0.5", "test");
factory->AddTree(chBackground, "Background", backgroundWeight, sel0+"mini_rand < 0.5", "train");
factory->AddTree(chBackground, "Background", backgroundWeight, sel0+"mini_rand >= 0.5", "test");
// To give different trees for training and testing, do as follows:
//factory->AddSignalTree( signalTrainingTree, signalWeight, "Training" );
//factory->AddSignalTree( signalTestTree, signalWeight, "Test" );
//factory->AddBackgroundTree( bkgTrainingTree, backgroundWeight, "Training" );
//factory->AddBackgroundTree( bkgTestTree, backgroundWeight, "Test" );
// Use the following code instead of the above two or four lines to add signal and background
// training and test events "by hand"
// NOTE that in this case one should not give expressions (such as "var1+var2") in the input
// variable definition, but simply compute the expression before adding the event
//
// // --- begin ----------------------------------------------------------
// std::vector<Double_t> vars( 4 ); // vector has size of number of input variables
// Float_t treevars[4], weight;
//
// // Signal
// for (UInt_t ivar=0; ivar<4; ivar++) signal->SetBranchAddress( Form( "var%i", ivar+1 ), &(treevars[ivar]) );
// for (UInt_t i=0; i<signal->GetEntries(); i++) {
// signal->GetEntry(i);
// for (UInt_t ivar=0; ivar<4; ivar++) vars[ivar] = treevars[ivar];
// // add training and test events; here: first half is training, second is testing
// // note that the weight can also be event-wise
// if (i < signal->GetEntries()/2.0) factory->AddSignalTrainingEvent( vars, signalWeight );
// else factory->AddSignalTestEvent ( vars, signalWeight );
// }
//
// // Background (has event weights)
// background->SetBranchAddress( "weight", &weight );
// for (UInt_t ivar=0; ivar<4; ivar++) background->SetBranchAddress( Form( "var%i", ivar+1 ), &(treevars[ivar]) );
// for (UInt_t i=0; i<background->GetEntries(); i++) {
// background->GetEntry(i);
// for (UInt_t ivar=0; ivar<4; ivar++) vars[ivar] = treevars[ivar];
// // add training and test events; here: first half is training, second is testing
// // note that the weight can also be event-wise
// if (i < background->GetEntries()/2) factory->AddBackgroundTrainingEvent( vars, backgroundWeight*weight );
// else factory->AddBackgroundTestEvent ( vars, backgroundWeight*weight );
// }
// // --- end ------------------------------------------------------------
//
// --- end of tree registration
// Set individual event weights (the variables must exist in the original TTree)
factory->SetSignalWeightExpression ("mini_weight");
factory->SetBackgroundWeightExpression("mini_weight");
/*
if( doMultipleOutputs ){
multifactory->AddTree(signal,"Signal");
multifactory->SetSignalWeightExpression ("event_scale1fb");
multifactory->SetBackgroundWeightExpression("event_scale1fb");
multifactory->SetWeightExpression("event_scale1fb");
if( includeBkg["ww"] ){
TTree* ww = (TTree*) chww;
multifactory->AddTree(ww,"WW");
cout << "Added WW to multi-MVA" << endl;
}
if( includeBkg["wjets"] ){
TTree* wjets = (TTree*) chwjets;
multifactory->AddTree(wjets,"WJets");
cout << "Added W+jets to multi-MVA" << endl;
}
if( includeBkg["tt"] ){
TTree* tt = (TTree*) chtt;
multifactory->AddTree(tt,"tt");
cout << "Added ttbar multi-MVA" << endl;
}
}
*/
// Apply additional cuts on the signal and background samples (can be different)
TCut mycuts = sel0; // for example: TCut mycuts = "abs(var1)<0.5 && abs(var2-0.5)<1";
TCut mycutb = sel0; // for example: TCut mycutb = "abs(var1)<0.5";
// Tell the factory how to use the training and testing events
//
// If no numbers of events are given, half of the events in the tree are used
// for training, and the other half for testing:
// factory->PrepareTrainingAndTestTree( mycut, "SplitMode=random:!V" );
// To also specify the number of testing events, use:
// factory->PrepareTrainingAndTestTree( mycut,
// "NSigTrain=3000:NBkgTrain=3000:NSigTest=3000:NBkgTest=3000:SplitMode=Random:!V" );
//Use random splitting
// factory->PrepareTrainingAndTestTree( mycuts, mycutb,
// "nTrain_Signal=100000:nTrain_Background=0:SplitMode=Random:NormMode=NumEvents:!V" );
factory->PrepareTrainingAndTestTree( "", "",
"nTrain_Signal=0:nTrain_Background=0:NormMode=None:!V" );
// if( doMultipleOutputs ){
// multifactory->PrepareTrainingAndTestTree( mycuts, mycutb,
// "nTrain_Signal=0:nTrain_Background=0:SplitMode=Random:NormMode=NumEvents:!V" );
// }
//Use alternate splitting
//(this is preferable since its easier to track which events were used for training, but the job crashes! need to fix this...)
//factory->PrepareTrainingAndTestTree( mycuts, mycutb,
// "nTrain_Signal=0:nTrain_Background=0:SplitMode=Alternate:NormMode=NumEvents:!V" );
// ---- Book MVA methods
//
// Please lookup the various method configuration options in the corresponding cxx files, eg:
// src/MethoCuts.cxx, etc, or here: http://tmva.sourceforge.net/optionRef.html
// it is possible to preset ranges in the option string in which the cut optimisation should be done:
// "...:CutRangeMin[2]=-1:CutRangeMax[2]=1"...", where [2] is the third input variable
// Cut optimisation
if (Use["Cuts"])
factory->BookMethod( TMVA::Types::kCuts, "Cuts",
"!H:!V:FitMethod=MC:EffSel:SampleSize=200000:VarProp=FSmart" );
if (Use["CutsD"])
factory->BookMethod( TMVA::Types::kCuts, "CutsD",
"!H:!V:FitMethod=MC:EffSel:SampleSize=200000:VarProp=FSmart:VarTransform=Decorrelate" );
if (Use["CutsPCA"])
factory->BookMethod( TMVA::Types::kCuts, "CutsPCA",
"!H:!V:FitMethod=MC:EffSel:SampleSize=200000:VarProp=FSmart:VarTransform=PCA" );
if (Use["CutsGA"])
factory->BookMethod( TMVA::Types::kCuts, "CutsGA",
"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" );
if (Use["CutsSA"])
factory->BookMethod( TMVA::Types::kCuts, "CutsSA",
"!H:!V:FitMethod=SA:EffSel:MaxCalls=150000:KernelTemp=IncAdaptive:InitialTemp=1e+6:MinTemp=1e-6:Eps=1e-10:UseDefaultScale" );
// Likelihood ("naive Bayes estimator")
if (Use["Likelihood"])
factory->BookMethod( TMVA::Types::kLikelihood, "Likelihood",
"H:!V:!TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmoothBkg[1]=10:NSmooth=1:NAvEvtPerBin=50" );
// Decorrelated likelihood
if (Use["LikelihoodD"])
factory->BookMethod( TMVA::Types::kLikelihood, "LikelihoodD",
"!H:!V:TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmooth=5:NAvEvtPerBin=50:VarTransform=Decorrelate" );
// PCA-transformed likelihood
if (Use["LikelihoodPCA"])
factory->BookMethod( TMVA::Types::kLikelihood, "LikelihoodPCA",
"!H:!V:!TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmooth=5:NAvEvtPerBin=50:VarTransform=PCA" );
// Use a kernel density estimator to approximate the PDFs
if (Use["LikelihoodKDE"])
factory->BookMethod( TMVA::Types::kLikelihood, "LikelihoodKDE",
"!H:!V:!TransformOutput:PDFInterpol=KDE:KDEtype=Gauss:KDEiter=Adaptive:KDEFineFactor=0.3:KDEborder=None:NAvEvtPerBin=50" );
// Use a variable-dependent mix of splines and kernel density estimator
if (Use["LikelihoodMIX"])
factory->BookMethod( TMVA::Types::kLikelihood, "LikelihoodMIX",
"!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" );
// Test the multi-dimensional probability density estimator
// here are the options strings for the MinMax and RMS methods, respectively:
// "!H:!V:VolumeRangeMode=MinMax:DeltaFrac=0.2:KernelEstimator=Gauss:GaussSigma=0.3" );
// "!H:!V:VolumeRangeMode=RMS:DeltaFrac=3:KernelEstimator=Gauss:GaussSigma=0.3" );
if (Use["PDERS"])
factory->BookMethod( TMVA::Types::kPDERS, "PDERS",
"!H:!V:NormTree=T:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600" );
if (Use["PDERSD"])
factory->BookMethod( TMVA::Types::kPDERS, "PDERSD",
"!H:!V:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600:VarTransform=Decorrelate" );
if (Use["PDERSPCA"])
factory->BookMethod( TMVA::Types::kPDERS, "PDERSPCA",
"!H:!V:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600:VarTransform=PCA" );
// Multi-dimensional likelihood estimator using self-adapting phase-space binning
if (Use["PDEFoam"])
factory->BookMethod( TMVA::Types::kPDEFoam, "PDEFoam",
"H:!V:SigBgSeparate=F:TailCut=0.001:VolFrac=0.0333:nActiveCells=500:nSampl=2000:nBin=5:Nmin=100:Kernel=None:Compress=T" );
if (Use["PDEFoamBoost"])
factory->BookMethod( TMVA::Types::kPDEFoam, "PDEFoamBoost",
"!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" );
// K-Nearest Neighbour classifier (KNN)
if (Use["KNN"])
factory->BookMethod( TMVA::Types::kKNN, "KNN",
"H:nkNN=20:ScaleFrac=0.8:SigmaFact=1.0:Kernel=Gaus:UseKernel=F:UseWeight=T:!Trim" );
// H-Matrix (chi2-squared) method
if (Use["HMatrix"])
factory->BookMethod( TMVA::Types::kHMatrix, "HMatrix", "!H:!V" );
// Linear discriminant (same as Fisher discriminant)
if (Use["LD"])
factory->BookMethod( TMVA::Types::kLD, "LD", "H:!V:VarTransform=None:CreateMVAPdfs:PDFInterpolMVAPdf=Spline2:NbinsMVAPdf=50:NsmoothMVAPdf=10" );
// Fisher discriminant (same as LD)
if (Use["Fisher"])
factory->BookMethod( TMVA::Types::kFisher, "Fisher", "H:!V:Fisher:CreateMVAPdfs:PDFInterpolMVAPdf=Spline2:NbinsMVAPdf=50:NsmoothMVAPdf=10" );
// Fisher with Gauss-transformed input variables
if (Use["FisherG"])
factory->BookMethod( TMVA::Types::kFisher, "FisherG", "H:!V:VarTransform=Gauss" );
// Composite classifier: ensemble (tree) of boosted Fisher classifiers
if (Use["BoostedFisher"])
factory->BookMethod( TMVA::Types::kFisher, "BoostedFisher",
"H:!V:Boost_Num=20:Boost_Transform=log:Boost_Type=AdaBoost:Boost_AdaBoostBeta=0.2" );
// Function discrimination analysis (FDA) -- test of various fitters - the recommended one is Minuit (or GA or SA)
if (Use["FDA_MC"])
factory->BookMethod( TMVA::Types::kFDA, "FDA_MC",
"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" );
if (Use["FDA_GA"]) // can also use Simulated Annealing (SA) algorithm (see Cuts_SA options])
factory->BookMethod( TMVA::Types::kFDA, "FDA_GA",
"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" );
if (Use["FDA_SA"]) // can also use Simulated Annealing (SA) algorithm (see Cuts_SA options])
factory->BookMethod( TMVA::Types::kFDA, "FDA_SA",
"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" );
if (Use["FDA_MT"])
factory->BookMethod( TMVA::Types::kFDA, "FDA_MT",
"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" );
if (Use["FDA_GAMT"])
factory->BookMethod( TMVA::Types::kFDA, "FDA_GAMT",
"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" );
if (Use["FDA_MCMT"])
factory->BookMethod( TMVA::Types::kFDA, "FDA_MCMT",
"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" );
// TMVA ANN: MLP (recommended ANN) -- all ANNs in TMVA are Multilayer Perceptrons
if (Use["MLP"])
factory->BookMethod( TMVA::Types::kMLP, "MLP", "H:!V:NeuronType=tanh:VarTransform=N:NCycles=600:HiddenLayers=N+5:TestRate=5:!UseRegulator" );
// factory->BookMethod( TMVA::Types::kMLP, "MLP", "!H:!V:NeuronType=tanh:VarTransform=N:NCycles=1000:HiddenLayers=N+N:TestRate=5:!UseRegulator:LearningRate=0.2:DecayRate=0.001:BPMode=batch:BatchSize=500");
if (Use["MLPBFGS"])
factory->BookMethod( TMVA::Types::kMLP, "MLPBFGS", "H:!V:NeuronType=tanh:VarTransform=N:NCycles=600:HiddenLayers=N+5:TestRate=5:TrainingMethod=BFGS:!UseRegulator" );
if (Use["MLPBNN"])
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
// CF(Clermont-Ferrand)ANN
if (Use["CFMlpANN"])
factory->BookMethod( TMVA::Types::kCFMlpANN, "CFMlpANN", "!H:!V:NCycles=2000:HiddenLayers=N+1,N" ); // n_cycles:#nodes:#nodes:...
// Tmlp(Root)ANN
if (Use["TMlpANN"])
factory->BookMethod( TMVA::Types::kTMlpANN, "TMlpANN", "!H:!V:NCycles=200:HiddenLayers=N+1,N:LearningMethod=BFGS:ValidationFraction=0.3" ); // n_cycles:#nodes:#nodes:...
// Support Vector Machine
if (Use["SVM"])
factory->BookMethod( TMVA::Types::kSVM, "SVM", "Gamma=0.25:Tol=0.001:VarTransform=Norm" );
// Boosted Decision Trees
if (Use["BDTG"]) // Gradient Boost
factory->BookMethod( TMVA::Types::kBDT, "BDTG",
"!H:!V:NTrees=1000:BoostType=Grad:Shrinkage=0.10:UseBaggedGrad:GradBaggingFraction=0.5:nCuts=20:NNodesMax=5" );
if (Use["BDT"]) // Adaptive Boost
factory->BookMethod( TMVA::Types::kBDT, "BDT",
"!H:!V:NTrees=850:nEventsMin=150:MaxDepth=3:BoostType=AdaBoost:AdaBoostBeta=0.5:SeparationType=GiniIndex:nCuts=20:PruneMethod=NoPruning" );
if (Use["BDT1"]) // Adaptive Boost
factory->BookMethod( TMVA::Types::kBDT, "BDT1",
"!H:!V:NTrees=200:nEventsMin=300:MaxDepth=3:BoostType=AdaBoost:AdaBoostBeta=0.5:SeparationType=GiniIndex:nCuts=4:PruneMethod=NoPruning" );
if (Use["BDTB"]) // Bagging
factory->BookMethod( TMVA::Types::kBDT, "BDTB",
"!H:!V:NTrees=400:BoostType=Bagging:SeparationType=GiniIndex:nCuts=20:PruneMethod=NoPruning" );
if (Use["BDTD"]) // Decorrelation + Adaptive Boost
factory->BookMethod( TMVA::Types::kBDT, "BDTD",
"!H:!V:NTrees=400:nEventsMin=400:MaxDepth=3:BoostType=AdaBoost:SeparationType=GiniIndex:nCuts=20:PruneMethod=NoPruning:VarTransform=Decorrelate" );
// RuleFit -- TMVA implementation of Friedman's method
if (Use["RuleFit"])
factory->BookMethod( TMVA::Types::kRuleFit, "RuleFit",
"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" );
// if( doMultipleOutputs ){
// if (Use["multi_BDTG"]) // gradient boosted decision trees
// multifactory->BookMethod( TMVA::Types::kBDT, "BDTG", "!H:!V:NTrees=1000:BoostType=Grad:Shrinkage=0.10:UseBaggedGrad:GradBaggingFraction=0.50:nCuts=20:NNodesMax=8");
// if (Use["multi_MLP"]) // neural network
// multifactory->BookMethod( TMVA::Types::kMLP, "MLP", "!H:!V:NeuronType=tanh:NCycles=1000:HiddenLayers=N+5,5:TestRate=5:EstimatorType=MSE");
// if (Use["multi_FDA_GA"]) // functional discriminant with GA minimizer
// multifactory->BookMethod( TMVA::Types::kFDA, "FDA_GA", "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" );
// }
// For an example of the category classifier usage, see: TMVAClassificationCategory
// --------------------------------------------------------------------------------------------------
// ---- Now you can optimize the setting (configuration) of the MVAs using the set of training events
// factory->OptimizeAllMethods("SigEffAt001","Scan");
// factory->OptimizeAllMethods("ROCIntegral","GA");
// --------------------------------------------------------------------------------------------------
// ---- Now you can tell the factory to train, test, and evaluate the MVAs
// Train MVAs using the set of training events
factory->TrainAllMethods();
// ---- Evaluate all MVAs using the set of test events
factory->TestAllMethods();
// ----- Evaluate and compare performance of all configured MVAs
factory->EvaluateAllMethods();
// if( doMultipleOutputs ){
// // Train nulti-MVAs using the set of training events
// multifactory->TrainAllMethods();
// // ---- Evaluate all multi-MVAs using the set of test events
// multifactory->TestAllMethods();
// // ----- Evaluate and compare performance of all configured multi-MVAs
// multifactory->EvaluateAllMethods();
// }
// --------------------------------------------------------------
// Save the output
outputFile->Close();
//if( doMultipleOutputs ) multioutputFile->Close();
std::cout << "==> Wrote root file: " << outputFile->GetName() << std::endl;
std::cout << "==> TMVAClassification is done!" << std::endl;
delete factory;
// Launch the GUI for the root macros
if (!gROOT->IsBatch()) TMVAGui( outfileName );
}
void TMVAMulticlass(){
TString outfileName = "TMVAMulticlass.root";
TFile* outputFile = TFile::Open( outfileName, "RECREATE" );
TMVA::Factory *factory = new TMVA::Factory( "TMVAClassification", outputFile,
"!V:!Silent:Color:DrawProgressBar:Transformations=I;D;P;G,D:AnalysisType=multiclass" );
factory->AddVariable( "var0", 'F' );
factory->AddVariable( "var1", 'F' );
TFile *input(0);
TString fname = "./data.root";
if (!gSystem->AccessPathName( fname )) {
// first we try to find data.root in the local directory
std::cout << "--- TMVAMulticlass : Accessing " << fname << std::endl;
input = TFile::Open( fname );
}
else {
gROOT->LoadMacro( "./createData.C");
create_multiclassdata(20000);
cout << " created data.root for tests of the multiclass features"<<endl;
input = TFile::Open( fname );
}
if (!input) {
std::cout << "ERROR: could not open data file" << std::endl;
exit(1);
}
TTree *tree = (TTree*)input->Get("TreeR");
gROOT->cd( outfileName+TString(":/") );
factory->AddTree ( tree, "Signal1", 1. , "cls==0" );
factory->AddTree ( tree, "Signal2", 1. , "cls==1" );
factory->AddTree ( tree, "Background", 1., "cls==2" );
factory->PrepareTrainingAndTestTree( "", "SplitMode=Random:NormMode=NumEvents:!V" );
factory->BookMethod( TMVA::Types::kBDT, "BDT", "!H:!V:NTrees=1000:BoostType=Grad:Shrinkage=0.30:UseBaggedGrad:GradBaggingFraction=0.6:SeparationType=GiniIndex:nCuts=20:NNodesMax=5");
factory->BookMethod( TMVA::Types::kMLP, "MLP", "!H:!V:NeuronType=tanh:VarTransform=N:NCycles=100:HiddenLayers=N+5,3:TestRate=5"); // testing vartransforms
factory->BookMethod( TMVA::Types::kMLP, "MLP2", "!H:!V:NeuronType=tanh:NCycles=100:HiddenLayers=N+5,3:TestRate=5");
factory->BookMethod( TMVA::Types::kFDA, "FDA_GA",
"H:!V:Formula=(0)+(1)*x0+(2)*x1+(3)*x0*x1:ParRanges=(-1,1);(-10,10);(-10,10);(-10,10):FitMethod=GA:PopSize=300:Cycles=3:Steps=20:Trim=True:SaveBestGen=1" );
// Train MVAs using the set of training events
factory->TrainAllMethods();
// ---- Evaluate all MVAs using the set of test events
factory->TestAllMethods();
// ----- Evaluate and compare performance of all configured MVAs
factory->EvaluateAllMethods();
// --------------------------------------------------------------
// Save the output
outputFile->Close();
std::cout << "==> Wrote root file: " << outputFile->GetName() << std::endl;
std::cout << "==> TMVAClassification is done!" << std::endl;
delete factory;
// Launch the GUI for the root macros
if (!gROOT->IsBatch()) TMVAGui( outfileName );
}
int TMVAKaggleHiggs ( TString myMethodList = "" )
{
// The explicit loading of the shared libTMVA is done in TMVAlogon.C, defined in .rootrc
// if you use your private .rootrc, or run from a different directory, please copy the
// corresponding lines from .rootrc
// methods to be processed can be given as an argument; use format:
//
// mylinux~> root -l TMVAClassification.C\(\"myMethod1,myMethod2,myMethod3\"\)
//
// if you like to use a method via the plugin mechanism, we recommend using
//
// mylinux~> root -l TMVAClassification.C\(\"P_myMethod\"\)
// (an example is given for using the BDT as plugin (see below),
// but of course the real application is when you write your own
// method based)
//---------------------------------------------------------------
// This loads the library
TMVA::Tools::Instance();
// Default MVA methods to be trained + tested
std::map<std::string,int> Use;
// --- Cut optimisation
Use["Cuts"] = 0;
Use["CutsD"] = 0;
Use["CutsPCA"] = 0;
Use["CutsGA"] = 0;
Use["CutsSA"] = 0;
//
// --- 1-dimensional likelihood ("naive Bayes estimator")
Use["Likelihood"] = 0;
Use["LikelihoodD"] = 0; // the "D" extension indicates decorrelated input variables (see option strings)
Use["LikelihoodPCA"] = 0; // the "PCA" extension indicates PCA-transformed input variables (see option strings)
Use["LikelihoodKDE"] = 0;
Use["LikelihoodMIX"] = 0;
//
// --- Mutidimensional likelihood and Nearest-Neighbour methods
Use["PDERS"] = 0;
Use["PDERSD"] = 0;
Use["PDERSPCA"] = 0;
Use["PDEFoam"] = 0;
Use["PDEFoamBoost"] = 0; // uses generalised MVA method boosting
Use["KNN"] = 0; // k-nearest neighbour method
//
// --- Linear Discriminant Analysis
Use["LD"] = 0; // Linear Discriminant identical to Fisher
Use["Fisher"] = 0;
Use["FisherG"] = 0;
Use["BoostedFisher"] = 0; // uses generalised MVA method boosting
Use["HMatrix"] = 0;
//
// --- Function Discriminant analysis
Use["FDA_GA"] = 0; // minimisation of user-defined function using Genetics Algorithm
Use["FDA_SA"] = 0;
Use["FDA_MC"] = 0;
Use["FDA_MT"] = 0;
Use["FDA_GAMT"] = 0;
Use["FDA_MCMT"] = 0;
//
// --- Neural Networks (all are feed-forward Multilayer Perceptrons)
Use["MLP"] = 0; // Recommended ANN
Use["MLPBFGS"] = 0; // Recommended ANN with optional training method
Use["MLPBNN"] = 0; // Recommended ANN with BFGS training method and bayesian regulator
Use["CFMlpANN"] = 0; // Depreciated ANN from ALEPH
Use["TMlpANN"] = 0; // ROOT's own ANN
Use["NN"] = 1; // improved implementation of a NN
//
// --- Support Vector Machine
Use["SVM"] = 0;
//
// --- Boosted Decision Trees
Use["BDT"] = 1; // uses Adaptive Boost
Use["BDTG"] = 0; // uses Gradient Boost
Use["BDTB"] = 0; // uses Bagging
Use["BDTD"] = 0; // decorrelation + Adaptive Boost
Use["BDTF"] = 0; // allow usage of fisher discriminant for node splitting
//
// --- Friedman's RuleFit method, ie, an optimised series of cuts ("rules")
Use["RuleFit"] = 0;
// ---------------------------------------------------------------
std::cout << std::endl;
std::cout << "==> Start TMVAKaggleHiggs" << std::endl;
// Select methods (don't look at this code - not of interest)
if (myMethodList != "") {
for (std::map<std::string,int>::iterator it = Use.begin(); it != Use.end(); it++) it->second = 0;
std::vector<TString> mlist = TMVA::gTools().SplitString( myMethodList, ',' );
for (UInt_t i=0; i<mlist.size(); i++) {
std::string regMethod(mlist[i]);
if (Use.find(regMethod) == Use.end()) {
std::cout << "Method \"" << regMethod << "\" not known in TMVA under this name. Choose among the following:" << std::endl;
for (std::map<std::string,int>::iterator it = Use.begin(); it != Use.end(); it++) std::cout << it->first << " ";
std::cout << std::endl;
return 1;
}
Use[regMethod] = 1;
}
}
// --------------------------------------------------------------------------------------------------
// --- Here the preparation phase begins
// Read training and test data
// (it is also possible to use ASCII format as input -> see TMVA Users Guide)
TString fname = "./training.root";
if (gSystem->AccessPathName( fname )) // file does not exist in local directory
gSystem->Exec("curl -O http://root.cern.ch/files/tmva_class_example.root");
TFile *input = TFile::Open( fname );
std::cout << "--- TMVAClassification : Using input file: " << input->GetName() << std::endl;
// --- Register the training and test trees
TTree *tree = (TTree*)input->Get("data");
// Create a ROOT output file where TMVA will store ntuples, histograms, etc.
TString outfileName( "TMVA.root" );
TFile* outputFile = TFile::Open( outfileName, "RECREATE" );
// Create the factory object. Later you can choose the methods
// whose performance you'd like to investigate. The factory is
// the only TMVA object you have to interact with
//
// The first argument is the base of the name of all the
// weightfiles in the directory weight/
//
// The second argument is the output file for the training results
// All TMVA output can be suppressed by removing the "!" (not) in
// front of the "Silent" argument in the option string
TMVA::Factory *factory = new TMVA::Factory( "TMVAClassification", outputFile,
"!V:!Silent:Color:DrawProgressBar:Transformations=I:AnalysisType=Classification" );
// If you wish to modify default settings
// (please check "src/Config.h" to see all available global options)
// (TMVA::gConfig().GetVariablePlotting()).fTimesRMS = 8.0;
// (TMVA::gConfig().GetIONames()).fWeightFileDir = "myWeightDirectory";
// Define the input variables that shall be used for the MVA training
// note that you may also use variable expressions, such as: "3*var1/var2*abs(var3)"
// [all types of expressions that can also be parsed by TTree::Draw( "expression" )]
// factory->AddVariable( "myvar1 := var1+var2", 'F' );
// factory->AddVariable( "myvar2 := var1-var2", "Expression 2", "", 'F' );
// factory->AddVariable( "var3", "Variable 3", "units", 'F' );
// factory->AddVariable( "var4", "Variable 4", "units", 'F' );
TString limit ("-900.0");
TString replacementValue ("0.0");
std::vector<std::string> vars = {"DER_mass_MMC","DER_mass_transverse_met_lep","DER_mass_vis","DER_pt_h","DER_deltaeta_jet_jet","DER_mass_jet_jet","DER_prodeta_jet_jet","DER_deltar_tau_lep","DER_pt_tot","DER_sum_pt","DER_pt_ratio_lep_tau","DER_met_phi_centrality","DER_lep_eta_centrality","PRI_tau_pt","PRI_tau_eta","PRI_tau_phi","PRI_lep_pt","PRI_lep_eta","PRI_lep_phi","PRI_met","PRI_met_phi","PRI_met_sumet","PRI_jet_num","PRI_jet_leading_pt","PRI_jet_leading_eta","PRI_jet_leading_phi","PRI_jet_subleading_pt","PRI_jet_subleading_eta","PRI_jet_subleading_phi","PRI_jet_all_pt"};
for (std::vector<std::string>::iterator it = vars.begin (), itEnd = vars.end (); it != itEnd; ++it)
{
std::string s = *it;
TString current;
current.Form ("%s:=(%s<%s?%s:%s)",s.c_str (), s.c_str (), limit.Data (), replacementValue.Data (), s.c_str ());
factory->AddVariable (current, 'F');
}
// You can add so-called "Spectator variables", which are not used in the MVA training,
// but will appear in the final "TestTree" produced by TMVA. This TestTree will contain the
// input variables, the response values of all trained MVAs, and the spectator variables
// factory->AddSpectator( "spec1 := var1*2", "Spectator 1", "units", 'F' );
// factory->AddSpectator( "spec2 := var1*3", "Spectator 2", "units", 'F' );
// global event weights per tree (see below for setting event-wise weights)
Double_t weight = 1.0;
// Double_t backgroundWeight = 1.0;
// You can add an arbitrary number of signal or background trees
// factory->AddBackgroundTree( background, backgroundWeight );
factory->AddTree(tree, "Signal", 1., "Label == 1");
factory->AddTree(tree, "Background", 1., "Label == 0");
// To give different trees for training and testing, do as follows:
// factory->AddSignalTree( signalTrainingTree, signalTrainWeight, "Training" );
// factory->AddSignalTree( signalTestTree, signalTestWeight, "Test" );
// Use the following code instead of the above two or four lines to add signal and background
// training and test events "by hand"
// NOTE that in this case one should not give expressions (such as "var1+var2") in the input
// variable definition, but simply compute the expression before adding the event
//
// // --- begin ----------------------------------------------------------
// std::vector<Double_t> vars( 4 ); // vector has size of number of input variables
// Float_t treevars[4], weight;
//
// // Signal
// for (UInt_t ivar=0; ivar<4; ivar++) signal->SetBranchAddress( Form( "var%i", ivar+1 ), &(treevars[ivar]) );
// for (UInt_t i=0; i<signal->GetEntries(); i++) {
// signal->GetEntry(i);
// for (UInt_t ivar=0; ivar<4; ivar++) vars[ivar] = treevars[ivar];
// // add training and test events; here: first half is training, second is testing
// // note that the weight can also be event-wise
// if (i < signal->GetEntries()/2.0) factory->AddSignalTrainingEvent( vars, signalWeight );
// else factory->AddSignalTestEvent ( vars, signalWeight );
// }
//
// // Background (has event weights)
// background->SetBranchAddress( "weight", &weight );
// for (UInt_t ivar=0; ivar<4; ivar++) background->SetBranchAddress( Form( "var%i", ivar+1 ), &(treevars[ivar]) );
// for (UInt_t i=0; i<background->GetEntries(); i++) {
// background->GetEntry(i);
// for (UInt_t ivar=0; ivar<4; ivar++) vars[ivar] = treevars[ivar];
// // add training and test events; here: first half is training, second is testing
// // note that the weight can also be event-wise
// if (i < background->GetEntries()/2) factory->AddBackgroundTrainingEvent( vars, backgroundWeight*weight );
// else factory->AddBackgroundTestEvent ( vars, backgroundWeight*weight );
// }
// --- end ------------------------------------------------------------
//
// --- end of tree registration
// Set individual event weights (the variables must exist in the original TTree)
// for signal : factory->SetSignalWeightExpression ("weight1*weight2");
// for background: factory->SetBackgroundWeightExpression("weight1*weight2");
factory->SetSignalWeightExpression( "Weight" );
factory->SetBackgroundWeightExpression( "Weight" );
// Apply additional cuts on the signal and background samples (can be different)
TCut mycuts = "Label==1"; // for example: TCut mycuts = "abs(var1)<0.5 && abs(var2-0.5)<1";
TCut mycutb = "Label==0"; // for example: TCut mycutb = "abs(var1)<0.5";
// Tell the factory how to use the training and testing events
//
// If no numbers of events are given, half of the events in the tree are used
// for training, and the other half for testing:
// factory->PrepareTrainingAndTestTree( mycut, "SplitMode=random:!V" );
// To also specify the number of testing events, use:
// factory->PrepareTrainingAndTestTree( mycut,
// "NSigTrain=3000:NBkgTrain=3000:NSigTest=3000:NBkgTest=3000:SplitMode=Random:!V" );
factory->PrepareTrainingAndTestTree( mycuts, mycutb,
"nTrain_Signal=0:nTrain_Background=0:nTest_Signal=0:nTest_Background=0:SplitMode=Random:NormMode=NumEvents:!V" );
// factory->PrepareTrainingAndTestTree( mycuts, mycutb,
// "nTrain_Signal=5000:nTrain_Background=5000:nTest_Signal=5000:nTest_Background=5000:SplitMode=Random:NormMode=NumEvents:!V" );
// ---- Book MVA methods
//
// Please lookup the various method configuration options in the corresponding cxx files, eg:
// src/MethoCuts.cxx, etc, or here: http://tmva.sourceforge.net/optionRef.html
// it is possible to preset ranges in the option string in which the cut optimisation should be done:
// "...:CutRangeMin[2]=-1:CutRangeMax[2]=1"...", where [2] is the third input variable
// Cut optimisation
if (Use["Cuts"])
factory->BookMethod( TMVA::Types::kCuts, "Cuts",
"!H:!V:FitMethod=MC:EffSel:SampleSize=200000:VarProp=FSmart" );
if (Use["CutsD"])
factory->BookMethod( TMVA::Types::kCuts, "CutsD",
"!H:!V:FitMethod=MC:EffSel:SampleSize=200000:VarProp=FSmart:VarTransform=Decorrelate" );
if (Use["CutsPCA"])
factory->BookMethod( TMVA::Types::kCuts, "CutsPCA",
"!H:!V:FitMethod=MC:EffSel:SampleSize=200000:VarProp=FSmart:VarTransform=PCA" );
if (Use["CutsGA"])
factory->BookMethod( TMVA::Types::kCuts, "CutsGA",
"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" );
if (Use["CutsSA"])
factory->BookMethod( TMVA::Types::kCuts, "CutsSA",
"!H:!V:FitMethod=SA:EffSel:MaxCalls=150000:KernelTemp=IncAdaptive:InitialTemp=1e+6:MinTemp=1e-6:Eps=1e-10:UseDefaultScale" );
// Likelihood ("naive Bayes estimator")
if (Use["Likelihood"])
factory->BookMethod( TMVA::Types::kLikelihood, "Likelihood",
"H:!V:TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmoothBkg[1]=10:NSmooth=1:NAvEvtPerBin=50" );
// Decorrelated likelihood
if (Use["LikelihoodD"])
factory->BookMethod( TMVA::Types::kLikelihood, "LikelihoodD",
"!H:!V:TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmooth=5:NAvEvtPerBin=50:VarTransform=Decorrelate" );
// PCA-transformed likelihood
if (Use["LikelihoodPCA"])
factory->BookMethod( TMVA::Types::kLikelihood, "LikelihoodPCA",
"!H:!V:!TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmooth=5:NAvEvtPerBin=50:VarTransform=PCA" );
// Use a kernel density estimator to approximate the PDFs
if (Use["LikelihoodKDE"])
factory->BookMethod( TMVA::Types::kLikelihood, "LikelihoodKDE",
"!H:!V:!TransformOutput:PDFInterpol=KDE:KDEtype=Gauss:KDEiter=Adaptive:KDEFineFactor=0.3:KDEborder=None:NAvEvtPerBin=50" );
// Use a variable-dependent mix of splines and kernel density estimator
if (Use["LikelihoodMIX"])
factory->BookMethod( TMVA::Types::kLikelihood, "LikelihoodMIX",
"!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" );
// Test the multi-dimensional probability density estimator
// here are the options strings for the MinMax and RMS methods, respectively:
// "!H:!V:VolumeRangeMode=MinMax:DeltaFrac=0.2:KernelEstimator=Gauss:GaussSigma=0.3" );
// "!H:!V:VolumeRangeMode=RMS:DeltaFrac=3:KernelEstimator=Gauss:GaussSigma=0.3" );
if (Use["PDERS"])
factory->BookMethod( TMVA::Types::kPDERS, "PDERS",
"!H:!V:NormTree=T:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600" );
if (Use["PDERSD"])
factory->BookMethod( TMVA::Types::kPDERS, "PDERSD",
"!H:!V:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600:VarTransform=Decorrelate" );
if (Use["PDERSPCA"])
factory->BookMethod( TMVA::Types::kPDERS, "PDERSPCA",
"!H:!V:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=400:NEventsMax=600:VarTransform=PCA" );
// Multi-dimensional likelihood estimator using self-adapting phase-space binning
if (Use["PDEFoam"])
factory->BookMethod( TMVA::Types::kPDEFoam, "PDEFoam",
"!H:!V:SigBgSeparate=F:TailCut=0.001:VolFrac=0.0666:nActiveCells=500:nSampl=2000:nBin=5:Nmin=100:Kernel=None:Compress=T" );
if (Use["PDEFoamBoost"])
factory->BookMethod( TMVA::Types::kPDEFoam, "PDEFoamBoost",
"!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" );
// K-Nearest Neighbour classifier (KNN)
if (Use["KNN"])
factory->BookMethod( TMVA::Types::kKNN, "KNN",
"H:nkNN=20:ScaleFrac=0.8:SigmaFact=1.0:Kernel=Gaus:UseKernel=F:UseWeight=T:!Trim" );
// H-Matrix (chi2-squared) method
if (Use["HMatrix"])
factory->BookMethod( TMVA::Types::kHMatrix, "HMatrix", "!H:!V:VarTransform=None" );
// Linear discriminant (same as Fisher discriminant)
if (Use["LD"])
factory->BookMethod( TMVA::Types::kLD, "LD", "H:!V:VarTransform=None:CreateMVAPdfs:PDFInterpolMVAPdf=Spline2:NbinsMVAPdf=50:NsmoothMVAPdf=10" );
// Fisher discriminant (same as LD)
if (Use["Fisher"])
factory->BookMethod( TMVA::Types::kFisher, "Fisher", "H:!V:Fisher:VarTransform=None:CreateMVAPdfs:PDFInterpolMVAPdf=Spline2:NbinsMVAPdf=50:NsmoothMVAPdf=10" );
// Fisher with Gauss-transformed input variables
if (Use["FisherG"])
factory->BookMethod( TMVA::Types::kFisher, "FisherG", "H:!V:VarTransform=Gauss" );
// Composite classifier: ensemble (tree) of boosted Fisher classifiers
if (Use["BoostedFisher"])
factory->BookMethod( TMVA::Types::kFisher, "BoostedFisher",
"H:!V:Boost_Num=20:Boost_Transform=log:Boost_Type=AdaBoost:Boost_AdaBoostBeta=0.2:!Boost_DetailedMonitoring" );
// Function discrimination analysis (FDA) -- test of various fitters - the recommended one is Minuit (or GA or SA)
if (Use["FDA_MC"])
factory->BookMethod( TMVA::Types::kFDA, "FDA_MC",
"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" );
if (Use["FDA_GA"]) // can also use Simulated Annealing (SA) algorithm (see Cuts_SA options])
factory->BookMethod( TMVA::Types::kFDA, "FDA_GA",
"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" );
if (Use["FDA_SA"]) // can also use Simulated Annealing (SA) algorithm (see Cuts_SA options])
factory->BookMethod( TMVA::Types::kFDA, "FDA_SA",
"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" );
if (Use["FDA_MT"])
factory->BookMethod( TMVA::Types::kFDA, "FDA_MT",
"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" );
if (Use["FDA_GAMT"])
factory->BookMethod( TMVA::Types::kFDA, "FDA_GAMT",
"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" );
if (Use["FDA_MCMT"])
factory->BookMethod( TMVA::Types::kFDA, "FDA_MCMT",
"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" );
// TMVA ANN: MLP (recommended ANN) -- all ANNs in TMVA are Multilayer Perceptrons
if (Use["MLP"])
factory->BookMethod( TMVA::Types::kMLP, "MLP", "H:!V:NeuronType=tanh:VarTransform=N:NCycles=600:HiddenLayers=N+5:TestRate=5:!UseRegulator" );
if (Use["MLPBFGS"])
factory->BookMethod( TMVA::Types::kMLP, "MLPBFGS", "H:!V:NeuronType=tanh:VarTransform=N:NCycles=600:HiddenLayers=N+5:TestRate=5:TrainingMethod=BFGS:!UseRegulator" );
if (Use["MLPBNN"])
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
// improved neural network implementation
if (Use["NN"])
{
// TString layoutString ("Layout=TANH|(N+100)*2,LINEAR");
// TString layoutString ("Layout=SOFTSIGN|100,SOFTSIGN|50,SOFTSIGN|20,LINEAR");
// TString layoutString ("Layout=RELU|300,RELU|100,RELU|30,RELU|10,LINEAR");
// TString layoutString ("Layout=SOFTSIGN|50,SOFTSIGN|30,SOFTSIGN|20,SOFTSIGN|10,LINEAR");
// TString layoutString ("Layout=TANH|50,TANH|30,TANH|20,TANH|10,LINEAR");
// TString layoutString ("Layout=SOFTSIGN|50,SOFTSIGN|20,LINEAR");
TString layoutString ("Layout=SOFTSIGN|70,SOFTSIGN|30,LINEAR");
std::vector<TString> strategy;
strategy.push_back (TString ("LearningRate=1e-2,Momentum=0.9,Repetitions=1,ConvergenceSteps=70,BatchSize=120,TestRepetitions=7,WeightDecay=0.001,Regularization=NONE,DropConfig=0.5+0.5+0.5+0.5,DropRepetitions=1,Multithreading=True"));
strategy.push_back (TString ("LearningRate=1e-4,Momentum=0.5,Repetitions=1,ConvergenceSteps=70,BatchSize=80,TestRepetitions=7,WeightDecay=0.001,Regularization=L2,Multithreading=True,DropConfig=0.1+0.1+0.1+0.1,DropRepetitions=1"));
strategy.push_back (TString ("LearningRate=1e-5,Momentum=0.3,Repetitions=1,ConvergenceSteps=70,BatchSize=60,TestRepetitions=7,WeightDecay=0.0001,Regularization=L2,Multithreading=True"));
strategy.push_back (TString ("LearningRate=1e-6,Momentum=0.0,Repetitions=1,ConvergenceSteps=70,BatchSize=40,TestRepetitions=7,WeightDecay=0.0001,Regularization=NONE,Multithreading=True"));
// strategy.push_back (TString ("LearningRate=1e-6,Momentum=0.0,Repetitions=1,ConvergenceSteps=50,BatchSize=30,TestRepetitions=7,WeightDecay=0.0001,Regularization=NONE,Multithreading=True"));
TString trainingStrategyString ("TrainingStrategy=");
for (std::vector<TString>::const_iterator it = strategy.begin (), itEnd = strategy.end (); it != itEnd; ++it)
{
if (it != strategy.begin ())
trainingStrategyString += "|";
trainingStrategyString += *it;
}
// TString nnOptions ("!H:V:VarTransform=Normalize:ErrorStrategy=CROSSENTROPY");
TString nnOptions ("!H:V:ErrorStrategy=CROSSENTROPY:VarTransform=G:WeightInitialization=XAVIERUNIFORM");
// TString nnOptions ("!H:V:VarTransform=Normalize:ErrorStrategy=CHECKGRADIENTS");
nnOptions.Append (":"); nnOptions.Append (layoutString);
nnOptions.Append (":"); nnOptions.Append (trainingStrategyString);
factory->BookMethod( TMVA::Types::kNN, "NN", nnOptions ); // NN
}
// CF(Clermont-Ferrand)ANN
if (Use["CFMlpANN"])
factory->BookMethod( TMVA::Types::kCFMlpANN, "CFMlpANN", "!H:!V:NCycles=2000:HiddenLayers=N+1,N" ); // n_cycles:#nodes:#nodes:...
// Tmlp(Root)ANN
if (Use["TMlpANN"])
factory->BookMethod( TMVA::Types::kTMlpANN, "TMlpANN", "!H:!V:NCycles=200:HiddenLayers=N+1,N:LearningMethod=BFGS:ValidationFraction=0.3" ); // n_cycles:#nodes:#nodes:...
// Support Vector Machine
if (Use["SVM"])
factory->BookMethod( TMVA::Types::kSVM, "SVM", "Gamma=0.25:Tol=0.001:VarTransform=Norm" );
// Boosted Decision Trees
if (Use["BDTG"]) // Gradient Boost
factory->BookMethod( TMVA::Types::kBDT, "BDTG",
"!H:!V:NTrees=1000:MinNodeSize=2.5%:BoostType=Grad:Shrinkage=0.10:UseBaggedBoost:BaggedSampleFraction=0.5:nCuts=20:MaxDepth=2" );
if (Use["BDT"]) // Adaptive Boost
factory->BookMethod( TMVA::Types::kBDT, "BDT",
"!H:!V:NTrees=850:MinNodeSize=2.5%:MaxDepth=3:BoostType=AdaBoost:AdaBoostBeta=0.5:UseBaggedBoost:BaggedSampleFraction=0.5:SeparationType=GiniIndex:nCuts=20" );
if (Use["BDTB"]) // Bagging
factory->BookMethod( TMVA::Types::kBDT, "BDTB",
"!H:!V:NTrees=400:BoostType=Bagging:SeparationType=GiniIndex:nCuts=20" );
if (Use["BDTD"]) // Decorrelation + Adaptive Boost
factory->BookMethod( TMVA::Types::kBDT, "BDTD",
"!H:!V:NTrees=400:MinNodeSize=5%:MaxDepth=3:BoostType=AdaBoost:SeparationType=GiniIndex:nCuts=20:VarTransform=Decorrelate" );
if (Use["BDTF"]) // Allow Using Fisher discriminant in node splitting for (strong) linearly correlated variables
factory->BookMethod( TMVA::Types::kBDT, "BDTMitFisher",
"!H:!V:NTrees=50:MinNodeSize=2.5%:UseFisherCuts:MaxDepth=3:BoostType=AdaBoost:AdaBoostBeta=0.5:SeparationType=GiniIndex:nCuts=20" );
// RuleFit -- TMVA implementation of Friedman's method
if (Use["RuleFit"])
factory->BookMethod( TMVA::Types::kRuleFit, "RuleFit",
"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" );
// For an example of the category classifier usage, see: TMVAClassificationCategory
// --------------------------------------------------------------------------------------------------
// ---- Now you can optimize the setting (configuration) of the MVAs using the set of training events
// ---- STILL EXPERIMENTAL and only implemented for BDT's !
// factory->OptimizeAllMethods("SigEffAt001","Scan");
// factory->OptimizeAllMethods("ROCIntegral","FitGA");
// --------------------------------------------------------------------------------------------------
// ---- Now you can tell the factory to train, test, and evaluate the MVAs
// Train MVAs using the set of training events
factory->TrainAllMethods();
// ---- Evaluate all MVAs using the set of test events
factory->TestAllMethods();
// ----- Evaluate and compare performance of all configured MVAs
factory->EvaluateAllMethods();
// --------------------------------------------------------------
// Save the output
outputFile->Close();
std::cout << "==> Wrote root file: " << outputFile->GetName() << std::endl;
std::cout << "==> TMVAClassification is done!" << std::endl;
delete factory;
// Launch the GUI for the root macros
if (!gROOT->IsBatch()) TMVA::TMVAGui( outfileName );
return 0;
}
void TMVARegression( int optimIndex, int Cat=0, TString myMethodList = "" )
{
// The explicit loading of the shared libTMVA is done in TMVAlogon.C, defined in .rootrc
// if you use your private .rootrc, or run from a different directory, please copy the
// corresponding lines from .rootrc
// methods to be processed can be given as an argument; use format:
//
// mylinux~> root -l TMVARegression.C\(\"myMethod1,myMethod2,myMethod3\"\)
//
//---------------------------------------------------------------
// This loads the library
TMVA::Tools::Instance();
// Default MVA methods to be trained + tested
std::map<std::string,int> Use;
// --- Mutidimensional likelihood and Nearest-Neighbour methods
Use["PDERS"] = 0;
Use["PDEFoam"] = 0;
Use["KNN"] = 0;
//
// --- Linear Discriminant Analysis
Use["LD"] = 0;
//
// --- Function Discriminant analysis
Use["FDA_GA"] = 0;
Use["FDA_MC"] = 0;
Use["FDA_MT"] = 0;
Use["FDA_GAMT"] = 0;
//
// --- Neural Network
Use["MLP"] = 0;
//
// --- Support Vector Machine
Use["SVM"] = 0;
//
// --- Boosted Decision Trees
Use["BDT"] = 0;
Use["BDTG"] = 1;
// ---------------------------------------------------------------
std::cout << std::endl;
std::cout << "==> Start TMVARegression" << std::endl;
// Select methods (don't look at this code - not of interest)
if (myMethodList != "") {
for (std::map<std::string,int>::iterator it = Use.begin(); it != Use.end(); it++) it->second = 0;
std::vector<TString> mlist = gTools().SplitString( myMethodList, ',' );
for (UInt_t i=0; i<mlist.size(); i++) {
std::string regMethod(mlist[i]);
if (Use.find(regMethod) == Use.end()) {
std::cout << "Method \"" << regMethod << "\" not known in TMVA under this name. Choose among the following:" << std::endl;
for (std::map<std::string,int>::iterator it = Use.begin(); it != Use.end(); it++) std::cout << it->first << " ";
std::cout << std::endl;
return;
}
Use[regMethod] = 1;
}
}
// --------------------------------------------------------------------------------------------------
// --- Here the preparation phase begins
// Create a new root output file
TString outfileName( Form("TMVAoutput/TMVAReg_%i.root",optimIndex) );
TFile* outputFile = TFile::Open( outfileName, "RECREATE" );
// Create the factory object. Later you can choose the methods
// whose performance you'd like to investigate. The factory will
// then run the performance analysis for you.
//
// The first argument is the base of the name of all the
// weightfiles in the directory weight/
//
// The second argument is the output file for the training results
// All TMVA output can be suppressed by removing the "!" (not) in
// front of the "Silent" argument in the option string
TMVA::Factory *factory = new TMVA::Factory( Form("TMVARegression_%i_Cat%i",optimIndex,Cat), outputFile,
"!V:!Silent:!Color:!DrawProgressBar:Transformations=I:AnalysisType=Regression" );
// If you wish to modify default settings
// (please check "src/Config.h" to see all available global options)
// (TMVA::gConfig().GetVariablePlotting()).fTimesRMS = 8.0;
// (TMVA::gConfig().GetIONames()).fWeightFileDir = "myWeightDirectory";
// Define the input variables that shall be used for the MVA training
// note that you may also use variable expressions, such as: "3*var1/var2*abs(var3)"
// [all types of expressions that can also be parsed by TTree::Draw( "expression" )]
factory->AddVariable( "jet_eta", "jet_eta", "units", 'F' );
factory->AddVariable( "jet_emfrac", "jet_emfrac", "units", 'F' );
factory->AddVariable( "jet_hadfrac", "jet_hadfrac", "units", 'F' );
factory->AddVariable( "jet_nconstituents", "jet_nconst", "units", 'F' );
factory->AddVariable( "jet_vtx3dL", "jet_vtx3dL", "units", 'F' );
factory->AddVariable( "MET", "MET", "units", 'F' );
factory->AddVariable( "jet_dPhiMETJet", "jet_dPhiMETJet", "units", 'F' );
//factory->AddVariable( "hJet_vtxPt", "hJet_vtxPt", "units", 'F' );
//factory->AddVariable( "hJet_JECUnc", "hJet_JECUnc", "units", 'F' );
//factory->AddVariable( "hJet_ptLeadTrack", "hJet_ptLeadTrack", "units", 'F' );
//factory->AddVariable( "hJet_SoftLeptPtCut:=(hJet_SoftLeptIdlooseMu>0. || hJet_SoftLeptId95>0.) ? (hJet_SoftLeptPt) : (-99)", "hJet_SoftLeptPt", "units", 'F' );
//factory->AddVariable( "hJet_En", "hJet_En", "units", 'F' );
//factory->AddVariable( "hJet_Et", "hJet_Et", "units", 'F' );
//factory->AddVariable( "hJet_Mt", "hJet_Mt", "units", 'F' );
//factory->AddVariable( "hJet_nch", "hJet_nch", "units", 'F' );
//factory->AddVariable( "hJet_vtx3deL", "hJet_vtx3deL", "units", 'F' );
//factory->AddVariable( "hJet_vtxMass", "hJet_vtxMass", "units", 'F' );
//factory->AddVariable( "hJet_ptRaw", "hJet_ptRaw", "units", 'F' );
//factory->AddVariable( "hJet_EnRaw", "hJet_EnRaw", "units", 'F' );
//factory->AddVariable( "hJet_SoftLeptptRelCut:=(hJet_SoftLeptIdlooseMu>0. || hJet_SoftLeptId95>0.) ? (hJet_SoftLeptptRel) : (-99)", "hJet_SoftLeptptRel", "units", 'F' );
//factory->AddVariable( "hJet_SoftLeptdRCut:=(hJet_SoftLeptIdlooseMu>0. || hJet_SoftLeptId95>0.) ? (hJet_SoftLeptdR) : (-99)", "hJet_SoftLeptdR", "units", 'F' );
//factory->AddVariable( "rho25", "rho25", "units", 'F' );
//factory->AddVariable( "dPhiMETJet", "dPhiMETJet", "units", 'F' );
// You can add so-called "Spectator variables", which are not used in the MVA training,
// but will appear in the final "TestTree" produced by TMVA. This TestTree will contain the
// input variables, the response values of all trained MVAs, and the spectator variables
// Add the variable carrying the regression target
//factory->AddTarget( "jet_genPt" );
factory->AddTarget( "jet_genJetPt/jet_pt" );
// It is also possible to declare additional targets for multi-dimensional regression, ie:
// -- factory->AddTarget( "fvalue2" );
// BUT: this is currently ONLY implemented for MLP
// Read training and test data (see TMVAClassification for reading ASCII files)
// load the signal and background event samples from ROOT trees
/* TFile *input(0);
TString fname = "./tmva_reg_example.root";
if (!gSystem->AccessPathName( fname ))
input = TFile::Open( fname ); // check if file in local directory exists
else
input = TFile::Open( "http://root.cern.ch/files/tmva_reg_example.root" ); // if not: download from ROOT server
if (!input) {
std::cout << "ERROR: could not open data file" << std::endl;
exit(1);
}
std::cout << "--- TMVARegression : Using input file: " << input->GetName() << std::endl;
// --- Register the regression tree
TTree *regTree = (TTree*)input->Get("TreeR");
*/
TChain chainTraining("Events");
chainTraining.Add("TrainingFiles/training.root");
TTree *regTreeTraining = (TTree*) chainTraining;
TChain chainTesting("Events");
chainTesting.Add("TrainingFiles/testing.root");
TTree *regTreeTesting = (TTree*) chainTesting;
// global event weights per tree (see below for setting event-wise weights)
Double_t regWeight = 1.0;
// You can add an arbitrary number of regression trees
factory->AddTree( regTreeTraining, "Regression", regWeight, "", "training" );
factory->AddTree( regTreeTesting, "Regression", regWeight, "", "test" );
// Apply additional cuts on the signal and background samples (can be different)
TCut mycut = "hJet_pt[0]>20. && hJet_pt[1]>20. && fabs(hJet_eta[0])<2.5 && fabs(hJet_eta[1])<2.5 && hJet_csv[0]>0. && hJet_csv[1]>0. && hJet_ptLeadTrack[0]<1500. && hJet_ptLeadTrack[1]<1500. && hJet_genJetPt[0]>0. && hJet_genJetPt[1]>0. && hJet_puJetIdL[0]>0.0 && hJet_puJetIdL[1]>0.0";
TCut testingCut = "hJet_pt[0]>20. && hJet_pt[1]>20. && abs(hJet_eta[0])<2.5 && abs(hJet_eta[1])<2.5";
TCut mjjCut = "sqrt(pow(hJet_pt[0]*cos(hJet_phi[0])+hJet_pt[1]*cos(hJet_phi[1]),2)+pow(hJet_pt[0]*sin(hJet_phi[0])+hJet_pt[1]*sin(hJet_phi[1]),2)) < 110";
if(Cat==1) mjjCut = "sqrt(pow(hJet_pt[0]*cos(hJet_phi[0])+hJet_pt[1]*cos(hJet_phi[1]),2)+pow(hJet_pt[0]*sin(hJet_phi[0])+hJet_pt[1]*sin(hJet_phi[1]),2)) > 110";
TCut jetPtCut="jet_pt>90";
if(Cat==1) jetPtCut="jet_pt>90";
//TCut trainingCut = "hJet_pt[0]>20. && hJet_pt[1]>20. && abs(hJet_eta[0])<2.5 && abs(hJet_eta[1])<2.5 && hJet_genJetPt[0]>0. && hJet_genJetPt[1]>0. && hJet_csv[0]>0.0 && hJet_csv[1]>0.0 && hJet_ptLeadTrack[0]<1500. && hJet_ptLeadTrack[1]<1500.";
TCut trainingCut = "jet_pt>20. && abs(jet_eta)<2.5 && jet_genJetPt>0. && jet_dRJetGenJet < 0.4 && (jet_partonID)==5";
// for example: TCut mycut = "abs(var1)<0.5 && abs(var2-0.5)<1";
// for example: TCut mycut = "abs(var1)<0.5 && abs(var2-0.5)<1";
// tell the factory to use all remaining events in the trees after training for testing:
// factory->PrepareTrainingAndTestTree(mycut, "nTrain_Regression=600000:nTest_Regression=600000:SplitMode=Random:NormMode=NumEvents:!V");
//factory->PrepareTrainingAndTestTree(mycut, "nTrain_Regression=337500:nTest_Regression=337500:SplitMode=Random:NormMode=NumEvents:!V");
// factory->PrepareTrainingAndTestTree(mycut, "nTrain_Regression=158393:nTest_Regression=158393:SplitMode=Random:NormMode=NumEvents:!V");
//factory->PrepareTrainingAndTestTree(mycut, "nTrain_Regression=14197:nTest_Regression=14197:SplitMode=Random:NormMode=NumEvents:!V");
factory->PrepareTrainingAndTestTree(trainingCut+jetPtCut, "!V");
// If no numbers of events are given, half of the events in the tree are used
// for training, and the other half for testing:
// factory->PrepareTrainingAndTestTree( mycut, "SplitMode=random:!V" );
// ---- Book MVA methods
//
// please lookup the various method configuration options in the corresponding cxx files, eg:
// src/MethoCuts.cxx, etc, or here: http://tmva.sourceforge.net/optionRef.html
// it is possible to preset ranges in the option string in which the cut optimisation should be done:
// "...:CutRangeMin[2]=-1:CutRangeMax[2]=1"...", where [2] is the third input variable
// PDE - RS method
if (Use["PDERS"])
factory->BookMethod( TMVA::Types::kPDERS, "PDERS",
"!H:!V:NormTree=T:VolumeRangeMode=Adaptive:KernelEstimator=Gauss:GaussSigma=0.3:NEventsMin=40:NEventsMax=60:VarTransform=None" );
// And the options strings for the MinMax and RMS methods, respectively:
// "!H:!V:VolumeRangeMode=MinMax:DeltaFrac=0.2:KernelEstimator=Gauss:GaussSigma=0.3" );
// "!H:!V:VolumeRangeMode=RMS:DeltaFrac=3:KernelEstimator=Gauss:GaussSigma=0.3" );
if (Use["PDEFoam"])
factory->BookMethod( TMVA::Types::kPDEFoam, "PDEFoam",
"!H:!V:MultiTargetRegression=F:TargetSelection=Mpv:TailCut=0.001:VolFrac=0.0666:nActiveCells=500:nSampl=2000:nBin=5:Compress=T:Kernel=None:Nmin=10:VarTransform=None" );
// K-Nearest Neighbour classifier (KNN)
if (Use["KNN"])
factory->BookMethod( TMVA::Types::kKNN, "KNN",
"nkNN=20:ScaleFrac=0.8:SigmaFact=1.0:Kernel=Gaus:UseKernel=F:UseWeight=T:!Trim" );
// Linear discriminant
if (Use["LD"])
factory->BookMethod( TMVA::Types::kLD, "LD",
"!H:!V:VarTransform=None" );
// Function discrimination analysis (FDA) -- test of various fitters - the recommended one is Minuit (or GA or SA)
if (Use["FDA_MC"])
factory->BookMethod( TMVA::Types::kFDA, "FDA_MC",
"!H:!V:Formula=(0)+(1)*x0+(2)*x1:ParRanges=(-100,100);(-100,100);(-100,100):FitMethod=MC:SampleSize=100000:Sigma=0.1:VarTransform=D" );
if (Use["FDA_GA"]) // can also use Simulated Annealing (SA) algorithm (see Cuts_SA options) .. the formula of this example is good for parabolas
factory->BookMethod( TMVA::Types::kFDA, "FDA_GA",
"!H:!V:Formula=(0)+(1)*x0+(2)*x1:ParRanges=(-100,100);(-100,100);(-100,100):FitMethod=GA:PopSize=100:Cycles=3:Steps=30:Trim=True:SaveBestGen=1:VarTransform=Norm" );
if (Use["FDA_MT"])
factory->BookMethod( TMVA::Types::kFDA, "FDA_MT",
"!H:!V:Formula=(0)+(1)*x0+(2)*x1:ParRanges=(-100,100);(-100,100);(-100,100);(-10,10):FitMethod=MINUIT:ErrorLevel=1:PrintLevel=-1:FitStrategy=2:UseImprove:UseMinos:SetBatch" );
if (Use["FDA_GAMT"])
factory->BookMethod( TMVA::Types::kFDA, "FDA_GAMT",
"!H:!V:Formula=(0)+(1)*x0+(2)*x1:ParRanges=(-100,100);(-100,100);(-100,100):FitMethod=GA:Converger=MINUIT:ErrorLevel=1:PrintLevel=-1:FitStrategy=0:!UseImprove:!UseMinos:SetBatch:Cycles=1:PopSize=5:Steps=5:Trim" );
// Neural network (MLP)
if (Use["MLP"])
factory->BookMethod( TMVA::Types::kMLP, "MLP", "!H:!V:VarTransform=Norm:NeuronType=tanh:NCycles=20000:HiddenLayers=N+20:TestRate=6:TrainingMethod=BFGS:Sampling=0.3:SamplingEpoch=0.8:ConvergenceImprove=1e-6:ConvergenceTests=15:!UseRegulator" );
// Support Vector Machine
if (Use["SVM"])
factory->BookMethod( TMVA::Types::kSVM, "SVM", "Gamma=0.25:Tol=0.001:VarTransform=Norm" );
//100, 5
// Boosted Decision Tree //100,5 nCuts=-1
if (Use["BDT"])
factory->BookMethod( TMVA::Types::kBDT, "BDT",
"!H:!V:NTrees=100:nEventsMin=4:BoostType=AdaBoostR2:SeparationType=RegressionVariance:nCuts=-1:PruneMethod=CostComplexity:PruneStrength=30" );
const bool doScan1=0;
int ntreesArray[8] = {100,200,300,400,500,600,700,800};
float shrinkageArray[3] = {0.1,0.2,0.3};
float gradbaggingfracArray[3] = {0.7,0.8,0.9};
int maxdepthArray[3] = {2,3,4};
int nnodesmaxArray[3] = {5,10,15};
if(!doScan1) shrinkageArray[2]=1.0;
int nnodesmaxIndex=-1,maxdepthIndex=-1,gradbaggingfracIndex=-1,shrinkageIndex=-1,ntreesIndex=-1;
if(doScan1){
nnodesmaxIndex = optimIndex/216;
maxdepthIndex = (optimIndex-nnodesmaxIndex*216)/72;
gradbaggingfracIndex = (optimIndex-nnodesmaxIndex*216-maxdepthIndex*72)/24;
shrinkageIndex = (optimIndex-nnodesmaxIndex*216-maxdepthIndex*72-gradbaggingfracIndex*24)/8;
ntreesIndex = (optimIndex-nnodesmaxIndex*216-maxdepthIndex*72-gradbaggingfracIndex*24-shrinkageIndex*8);
}
else{
nnodesmaxIndex = optimIndex/72;
maxdepthIndex = (optimIndex-nnodesmaxIndex*72)/24;
shrinkageIndex = (optimIndex-nnodesmaxIndex*72-maxdepthIndex*24)/8;
ntreesIndex = (optimIndex-nnodesmaxIndex*72-maxdepthIndex*24-shrinkageIndex*8);
}
if (Use["BDTG"]){
if(doScan1)
factory->BookMethod( TMVA::Types::kBDT, "BDTG",
Form("!H:!V:NTrees=%i::BoostType=Grad:Shrinkage=%.1f:UseBaggedGrad:GradBaggingFraction=%.1f:nCuts=200:MaxDepth=%i:NNodesMax=%i",ntreesArray[ntreesIndex],shrinkageArray[shrinkageIndex],gradbaggingfracArray[gradbaggingfracIndex],maxdepthArray[maxdepthIndex],nnodesmaxArray[nnodesmaxIndex]) );
else
factory->BookMethod( TMVA::Types::kBDT, "BDTG",
Form("!H:!V:IgnoreNegWeights:NTrees=%i::BoostType=Grad:Shrinkage=%.1f:nCuts=200:MaxDepth=%i:NNodesMax=%i",ntreesArray[ntreesIndex],shrinkageArray[shrinkageIndex],maxdepthArray[maxdepthIndex],nnodesmaxArray[nnodesmaxIndex]) );
}
if(doScan1)
cout << Form("!H:!V:NTrees=%i::BoostType=Grad:Shrinkage=%.2f:UseBaggedGrad:GradBaggingFraction=%.2f:nCuts=200:MaxDepth=%i:NNodesMax=%i",ntreesArray[ntreesIndex],shrinkageArray[shrinkageIndex],gradbaggingfracArray[gradbaggingfracIndex],maxdepthArray[maxdepthIndex],nnodesmaxArray[nnodesmaxIndex])<<endl;
else
cout << Form("!H:!V:IgnoreNegWeights:NTrees=%i::BoostType=Grad:Shrinkage=%.1f:nCuts=200:MaxDepth=%i:NNodesMax=%i",ntreesArray[ntreesIndex],shrinkageArray[shrinkageIndex],maxdepthArray[maxdepthIndex],nnodesmaxArray[nnodesmaxIndex]) << endl;
// --------------------------------------------------------------------------------------------------
// ---- Now you can tell the factory to train, test, and evaluate the MVAs
// Train MVAs using the set of training events
factory->TrainAllMethods();
// ---- Evaluate all MVAs using the set of test events
factory->TestAllMethods();
// ----- Evaluate and compare performance of all configured MVAs
factory->EvaluateAllMethods();
// --------------------------------------------------------------
// Save the output
outputFile->Close();
std::cout << "==> Wrote root file: " << outputFile->GetName() << std::endl;
std::cout << "==> TMVARegression is done!" << std::endl;
delete factory;
// Launch the GUI for the root macros
if (!gROOT->IsBatch()) TMVARegGui( outfileName );
}
int main(int argc, char * argv[])
{
//Processing input options
int c;
std::string outFname;
outFname = std::string("QualityNaF.root");
// Open input files, get the trees
TChain *mc = InputFileReader("FileListNtuples_ext.txt","parametri_geo");
// Preparing options for the TMVA::Factory
std::string options(
"!V:"
"!Silent:"
"Color:"
"DrawProgressBar:"
"Transformations=I;D;P;G,D:"
"AnalysisType=Classification"
);
//Creating the factory
TFile * ldFile = new TFile(outFname.c_str(),"RECREATE");
TMVA::Factory * factory = new TMVA::Factory("QualityNaF", ldFile, options.c_str());
//Preparing variables
//general
/*factory->AddVariable("Chisquare", 'F');
factory->AddVariable("Layernonusati", 'I');
factory->AddVariable("NTofUsed", 'I');
factory->AddVariable("diffR", 'F');
factory->AddVariable("TOF_Up_Down", 'F');*/
//Tof
//factory->AddVariable("TOFchisq_s", 'F');
//factory->AddVariable("TOFchisq_t", 'F');
//RICH
factory->AddVariable("Richtotused", 'F');
factory->AddVariable("RichPhEl", 'F');
factory->AddVariable("RICHprob", 'F');
factory->AddVariable("RICHcollovertotal");
factory->AddVariable("RICHLipBetaConsistency");
factory->AddVariable("RICHTOFBetaConsistency");
factory->AddVariable("RICHChargeConsistency");
factory->AddVariable("RICHPmts");
factory->AddVariable("RICHgetExpected");
factory->AddVariable("tot_hyp_p_uncorr");
factory->AddVariable("Bad_ClusteringRICH");
factory->AddVariable("NSecondariesRICHrich");
//factory->AddVariable("HitHValldir");
//factory->AddVariable("HitHVallrefl");
//factory->AddVariable("HVBranchCheck:= (HitHValldir - HitHVoutdir) - (HitHVallrefl - HitHVoutrefl)");
factory->AddVariable("HitHVoutdir");
factory->AddVariable("HitHVoutrefl");
//Spectator Variables
factory->AddSpectator("R", 'F');
factory->AddSpectator("BetaRICH_new", 'F');
//Preselection cuts
std::string PreSelection = "qL1>0&&(joinCutmask&187)==187&&qL1<1.75&&R>0";
std::string ChargeCut = "qUtof>0.8&&qUtof<1.3&&qLtof>0.8&&qLtof<1.3";
std::string VelocityCut = /*"Beta<0.8";*/"((joinCutmask>>11))==1024&&BetaRICH_new>0&&BetaRICH_new<0.975";
std::string signalCut = /*"(R/Beta)*(1-Beta^2)^0.5>1.65&&GenMass>1&&GenMass<2";*/"(R/BetaRICH_new)*(1-BetaRICH_new^2)^0.5>0.5&&(R/BetaRICH_new)*(1-BetaRICH_new^2)^0.5<1.5";
std::string bkgndCut = /*"(R/Beta)*(1-Beta^2)^0.5>1.65&&GenMass>0&&GenMass<1";*/"(R/BetaRICH_new)*(1-BetaRICH_new^2)^0.5>3";
factory->AddTree(mc,"Signal" ,1,(PreSelection +"&&"+ ChargeCut + "&&" + VelocityCut + "&&"+ signalCut).c_str());
factory->AddTree(mc,"Background",1,(PreSelection +"&&"+ ChargeCut + "&&" + VelocityCut + "&&"+ bkgndCut).c_str());
// Preparing
std::string preselection = "";
std::string inputparams(
"SplitMode=Random:"
"NormMode=NumEvents:"
"!V"
);
factory->PrepareTrainingAndTestTree(preselection.c_str(),inputparams.c_str());
// Training
std::string trainparams ="!H:!V:MaxDepth=3";
factory->BookMethod(TMVA::Types::kBDT, "BDT", trainparams.c_str());
trainparams ="!H:!V";
factory->BookMethod(TMVA::Types::kLikelihood, "Likelihood", trainparams.c_str());
trainparams ="!H:!V:VarTransform=Decorrelate";
//factory->BookMethod(TMVA::Types::kLikelihood, "LikelihoodD", trainparams.c_str());
trainparams ="!H:!V";
//factory->BookMethod(TMVA::Types::kCuts, "Cuts", trainparams.c_str());
factory->TrainAllMethods();
factory->TestAllMethods();
factory->EvaluateAllMethods();
}
void TMVAMulticlass( TString myMethodList = "" )
{
// This loads the library
TMVA::Tools::Instance();
// to get access to the GUI and all tmva macros
TString tmva_dir(TString(gRootDir) + "/tmva");
if(gSystem->Getenv("TMVASYS"))
tmva_dir = TString(gSystem->Getenv("TMVASYS"));
gROOT->SetMacroPath(tmva_dir + "/test/:" + gROOT->GetMacroPath() );
gROOT->ProcessLine(".L TMVAMultiClassGui.C");
//---------------------------------------------------------------
// default MVA methods to be trained + tested
std::map<std::string,int> Use;
Use["MLP"] = 1;
Use["BDTG"] = 1;
Use["FDA_GA"] = 0;
Use["PDEFoam"] = 0;
//---------------------------------------------------------------
std::cout << std::endl;
std::cout << "==> Start TMVAMulticlass" << std::endl;
if (myMethodList != "") {
for (std::map<std::string,int>::iterator it = Use.begin(); it != Use.end(); it++) it->second = 0;
std::vector<TString> mlist = TMVA::gTools().SplitString( myMethodList, ',' );
for (UInt_t i=0; i<mlist.size(); i++) {
std::string regMethod(mlist[i]);
if (Use.find(regMethod) == Use.end()) {
std::cout << "Method \"" << regMethod << "\" not known in TMVA under this name. Choose among the following:" << std::endl;
for (std::map<std::string,int>::iterator it = Use.begin(); it != Use.end(); it++) std::cout << it->first << " ";
std::cout << std::endl;
return;
}
Use[regMethod] = 1;
}
}
// Create a new root output file.
TString outfileName = "TMVAMulticlass.root";
TFile* outputFile = TFile::Open( outfileName, "RECREATE" );
TMVA::Factory *factory = new TMVA::Factory( "TMVAMulticlass", outputFile,
"!V:!Silent:Color:DrawProgressBar:Transformations=I;D;P;G,D:AnalysisType=multiclass" );
factory->AddVariable( "var1", 'F' );
factory->AddVariable( "var2", "Variable 2", "", 'F' );
factory->AddVariable( "var3", "Variable 3", "units", 'F' );
factory->AddVariable( "var4", "Variable 4", "units", 'F' );
TFile *input(0);
TString fname = "./tmva_example_multiple_background.root";
if (!gSystem->AccessPathName( fname )) {
// first we try to find the file in the local directory
std::cout << "--- TMVAMulticlass : Accessing " << fname << std::endl;
input = TFile::Open( fname );
}
else {
cout << "Creating testdata...." << std::endl;
gROOT->ProcessLine(".L createData.C+");
gROOT->ProcessLine("create_MultipleBackground(2000)");
cout << " created tmva_example_multiple_background.root for tests of the multiclass features"<<endl;
input = TFile::Open( fname );
}
if (!input) {
std::cout << "ERROR: could not open data file" << std::endl;
exit(1);
}
TTree *signal = (TTree*)input->Get("TreeS");
TTree *background0 = (TTree*)input->Get("TreeB0");
TTree *background1 = (TTree*)input->Get("TreeB1");
TTree *background2 = (TTree*)input->Get("TreeB2");
gROOT->cd( outfileName+TString(":/") );
factory->AddTree (signal,"Signal");
factory->AddTree (background0,"bg0");
factory->AddTree (background1,"bg1");
factory->AddTree (background2,"bg2");
factory->PrepareTrainingAndTestTree( "", "SplitMode=Random:NormMode=NumEvents:!V" );
if (Use["BDTG"]) // gradient boosted decision trees
factory->BookMethod( TMVA::Types::kBDT, "BDTG", "!H:!V:NTrees=1000:BoostType=Grad:Shrinkage=0.10:UseBaggedBoost:BaggedSampleFraction=0.50:nCuts=20:MaxDepth=2");
if (Use["MLP"]) // neural network
factory->BookMethod( TMVA::Types::kMLP, "MLP", "!H:!V:NeuronType=tanh:NCycles=1000:HiddenLayers=N+5,5:TestRate=5:EstimatorType=MSE");
if (Use["FDA_GA"]) // functional discriminant with GA minimizer
factory->BookMethod( TMVA::Types::kFDA, "FDA_GA", "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" );
if (Use["PDEFoam"]) // PDE-Foam approach
factory->BookMethod( TMVA::Types::kPDEFoam, "PDEFoam", "!H:!V:TailCut=0.001:VolFrac=0.0666:nActiveCells=500:nSampl=2000:nBin=5:Nmin=100:Kernel=None:Compress=T" );
// Train MVAs using the set of training events
factory->TrainAllMethods();
// ---- Evaluate all MVAs using the set of test events
factory->TestAllMethods();
// ----- Evaluate and compare performance of all configured MVAs
factory->EvaluateAllMethods();
// --------------------------------------------------------------
// Save the output
outputFile->Close();
std::cout << "==> Wrote root file: " << outputFile->GetName() << std::endl;
std::cout << "==> TMVAClassification is done!" << std::endl;
delete factory;
// Launch the GUI for the root macros
if (!gROOT->IsBatch()) TMVAMultiClassGui( outfileName );
}
std::pair<TString,TString> TMVAClassification (
TString infilename,
AnalysisType analysisType = AnalysisType::DIRECT,
TString additionalRootFileName = "")
{
TMVA::Tools::Instance();
std::string tmstr (now ());
TString tmstmp (tmstr.c_str ());
std::cout << "==> Start TMVAClassification" << std::endl;
std::cout << "-------------------- open input file ---------------- " << std::endl;
TString fname = infilename; //pathToData + infilename + TString (".root");
if (analysisType != AnalysisType::TRANSFORMED)
fname = pathToData + infilename + TString (".root");
std::cout << "open file " << std::endl << fname.Data () << std::endl;
std::cout << "-------------------- get tree ---------------- " << std::endl;
TString treeName = "data";
if (analysisType == AnalysisType::TRANSFORMED)
treeName = "transformed";
std::cout << "-------------------- create tchain with treeName ---------------- " << std::endl;
std::cout << treeName << std::endl;
TChain* tree = new TChain (treeName);
std::cout << "add file" << std::endl;
std::cout << fname << std::endl;
tree->Add (fname);
TChain* treeFriend (NULL);
if (additionalRootFileName.Length () > 0)
{
std::cout << "-------------------- add additional input file ---------------- " << std::endl;
std::cout << additionalRootFileName << std::endl;
treeFriend = new TChain (treeName);
treeFriend->Add (additionalRootFileName);
tree->AddFriend (treeFriend,"p");
}
// tree->Draw ("mass:prediction");
// return std::make_pair(TString("hallo"),TString ("nix"));
TString outfileName;
if (analysisType == AnalysisType::BACKGROUND)
{
outfileName = TString ("BACK_" + infilename) + tmstmp + TString (".root");
}
else
outfileName += TString ( "TMVA__" ) + tmstmp + TString (".root");
std::cout << "-------------------- open output file ---------------- " << std::endl;
TFile* outputFile = TFile::Open( outfileName, "RECREATE" );
std::cout << "-------------------- prepare factory ---------------- " << std::endl;
TMVA::Factory *factory = new TMVA::Factory( "TMVAClassification", outputFile,
"AnalysisType=Classification:Transformations=I:!V" );
std::cout << "-------------------- add variables ---------------- " << std::endl;
for (auto varname : variableNames)
{
factory->AddVariable (varname.c_str (), 'F');
}
for (auto varname : spectatorNames)
{
factory->AddSpectator (varname.c_str (), 'F');
}
std::cout << "-------------------- add trees ---------------- " << std::endl;
TCut signalCut ("signal==1");
TCut backgroundCut ("signal==0");
if (analysisType == AnalysisType::TRANSFORMED)
{
signalCut = "(signal_original==1 && signal_in==0)";
backgroundCut = "(signal_original==0 && signal_in==0)";
}
if (analysisType == AnalysisType::BACKGROUND)
{
signalCut = TString("(signal==0) * (prediction > 0.7)");
backgroundCut = TString("(signal==0) * (prediction < 0.4)");
}
//tree->Draw ("prediction",signalCut);
//return std::make_pair(TString("hallo"),TString ("nix"));
factory->AddTree(tree, "Signal", 1.0, baseCut + signalCut, "TrainingTesting");
factory->AddTree(tree, "Background", 1.0, baseCut + backgroundCut, "TrainingTesting");
TCut mycuts = ""; // for example: TCut mycuts = "abs(var1)<0.5 && abs(var2-0.5)<1";
TCut mycutb = ""; // for example: TCut mycutb = "abs(var1)<0.5";
/* // Set individual event weights (the variables must exist in the original TTree) */
if (analysisType == AnalysisType::BACKGROUND)
{
factory->SetSignalWeightExpression ("prediction");
factory->SetBackgroundWeightExpression ("1");
}
std::cout << "-------------------- prepare ---------------- " << std::endl;
factory->PrepareTrainingAndTestTree( mycuts, mycutb,
"nTrain_Signal=0:nTrain_Background=0:nTest_Signal=0:nTest_Background=0:SplitMode=Random:NormMode=NumEvents:!V" );
TString methodName ("");
if (analysisType == AnalysisType::BACKGROUND)
methodName = TString ("TONBKG_") + tmstmp;
if (false)
{
// gradient boosting training
methodName += TString("GBDT");
factory->BookMethod(TMVA::Types::kBDT, methodName,
"NTrees=40:BoostType=Grad:Shrinkage=0.01:MaxDepth=7:UseNvars=6:nCuts=20:MinNodeSize=10");
}
if (false)
{
methodName += TString("Likelihood");
factory->BookMethod( TMVA::Types::kLikelihood, methodName,
"H:!V:TransformOutput:PDFInterpol=Spline2:NSmoothSig[0]=20:NSmoothBkg[0]=20:NSmoothBkg[1]=10:NSmooth=1:NAvEvtPerBin=50" );
}
if (false)
{
TString layoutString ("Layout=TANH|100,LINEAR");
TString training0 ("LearningRate=1e-1,Momentum=0.0,Repetitions=1,ConvergenceSteps=300,BatchSize=20,TestRepetitions=15,WeightDecay=0.001,Regularization=NONE,DropConfig=0.0+0.5+0.5+0.5,DropRepetitions=1,Multithreading=True");
TString training1 ("LearningRate=1e-2,Momentum=0.5,Repetitions=1,ConvergenceSteps=300,BatchSize=30,TestRepetitions=7,WeightDecay=0.001,Regularization=L2,Multithreading=True,DropConfig=0.0+0.1+0.1+0.1,DropRepetitions=1");
TString training2 ("LearningRate=1e-2,Momentum=0.3,Repetitions=1,ConvergenceSteps=300,BatchSize=40,TestRepetitions=7,WeightDecay=0.0001,Regularization=L2,Multithreading=True");
TString training3 ("LearningRate=1e-3,Momentum=0.1,Repetitions=1,ConvergenceSteps=200,BatchSize=70,TestRepetitions=7,WeightDecay=0.0001,Regularization=NONE,Multithreading=True");
TString trainingStrategyString ("TrainingStrategy=");
trainingStrategyString += training0 + "|" + training1 + "|" + training2 + "|" + training3;
TString nnOptions ("!H:V:ErrorStrategy=CROSSENTROPY:VarTransform=G:WeightInitialization=XAVIERUNIFORM");
nnOptions.Append (":"); nnOptions.Append (layoutString);
nnOptions.Append (":"); nnOptions.Append (trainingStrategyString);
methodName += TString("NNgauss");
factory->BookMethod( TMVA::Types::kNN, methodName, nnOptions ); // NN
}
if (false)
{
TString layoutString ("Layout=TANH|200,TANH|70,LINEAR");
TString training0 ("LearningRate=1e-2,Momentum=0.0,Repetitions=1,ConvergenceSteps=300,BatchSize=20,TestRepetitions=15,WeightDecay=0.001,Regularization=NONE,DropConfig=0.0+0.5+0.5+0.5,DropRepetitions=1,Multithreading=True");
TString training1 ("LearningRate=1e-3,Momentum=0.5,Repetitions=1,ConvergenceSteps=300,BatchSize=30,TestRepetitions=7,WeightDecay=0.001,Regularization=L2,Multithreading=True,DropConfig=0.0+0.1+0.1+0.1,DropRepetitions=1");
TString training2 ("LearningRate=1e-4,Momentum=0.3,Repetitions=1,ConvergenceSteps=300,BatchSize=40,TestRepetitions=7,WeightDecay=0.0001,Regularization=L2,Multithreading=True");
TString training3 ("LearningRate=1e-5,Momentum=0.1,Repetitions=1,ConvergenceSteps=200,BatchSize=70,TestRepetitions=7,WeightDecay=0.0001,Regularization=NONE,Multithreading=True");
TString trainingStrategyString ("TrainingStrategy=");
trainingStrategyString += training0 + "|" + training1 + "|" + training2 + "|" + training3;
// trainingStrategyString += training0 + "|" + training2 + "|" + training3;
// trainingStrategyString += training0 + "|" + training2;
// TString nnOptions ("!H:V:VarTransform=Normalize:ErrorStrategy=CROSSENTROPY");
TString nnOptions ("!H:V:ErrorStrategy=CROSSENTROPY:VarTransform=N:WeightInitialization=XAVIERUNIFORM");
// TString nnOptions ("!H:V:VarTransform=Normalize:ErrorStrategy=CHECKGRADIENTS");
nnOptions.Append (":"); nnOptions.Append (layoutString);
nnOptions.Append (":"); nnOptions.Append (trainingStrategyString);
methodName = TString("NNnormalized");
factory->BookMethod( TMVA::Types::kNN, methodName, nnOptions ); // NN
}
if (true)
{
TString layoutString ("Layout=TANH|100,TANH|50,LINEAR");
TString training0 ("LearningRate=1e-2,Momentum=0.0,Repetitions=1,ConvergenceSteps=100,BatchSize=20,TestRepetitions=7,WeightDecay=0.001,Regularization=NONE,DropConfig=0.0+0.5+0.5+0.5,DropRepetitions=1,Multithreading=True");
TString training1 ("LearningRate=1e-3,Momentum=0.0,Repetitions=1,ConvergenceSteps=20,BatchSize=30,TestRepetitions=7,WeightDecay=0.001,Regularization=L2,Multithreading=True,DropConfig=0.0+0.1+0.1+0.1,DropRepetitions=1");
TString training2 ("LearningRate=1e-4,Momentum=0.0,Repetitions=1,ConvergenceSteps=20,BatchSize=40,TestRepetitions=7,WeightDecay=0.0001,Regularization=L2,Multithreading=True");
TString training3 ("LearningRate=1e-5,Momentum=0.0,Repetitions=1,ConvergenceSteps=30,BatchSize=70,TestRepetitions=7,WeightDecay=0.0001,Regularization=NONE,Multithreading=True");
TString trainingStrategyString ("TrainingStrategy=");
trainingStrategyString += training0 + "|" + training1 + "|" + training2 + "|" + training3;
TString nnOptions ("!H:!V:ErrorStrategy=CROSSENTROPY:VarTransform=P+G:WeightInitialization=XAVIERUNIFORM");
nnOptions.Append (":"); nnOptions.Append (layoutString);
nnOptions.Append (":"); nnOptions.Append (trainingStrategyString);
methodName += TString("NNPG");
factory->BookMethod( TMVA::Types::kNN, methodName, nnOptions ); // NN
}
factory->TrainAllMethods();
// return std::make_pair(TString("hallo"),TString ("nix"));
factory->TestAllMethods();
factory->EvaluateAllMethods();
//input->Close();
outputFile->Close();
// TMVA::TMVAGui (outfileName);
delete factory;
delete tree;
switch (analysisType)
{
case AnalysisType::BACKGROUND:
std::cout << "DONE BACKGROUND" << std::endl;
break;
case AnalysisType::DIRECT:
std::cout << "DONE DIRECT" << std::endl;
break;
case AnalysisType::TRANSFORMED:
std::cout << "DONE TRANSFORMED" << std::endl;
break;
};
std::cout << "classification, return : " << outfileName << " , " << methodName << std::endl;
return std::make_pair (outfileName, methodName);
}