Logo ROOT   6.16/01
Reference Guide
TMVACrossValidation.C File Reference
Tutorials » TMVA tutorials

Detailed Description

View in nbviewer Open in SWAN This macro provides an example of how to use TMVA for k-folds cross evaluation.

As input data is used a toy-MC sample consisting of two guassian distributions.

The output file "TMVA.root" can be analysed with the use of dedicated macros (simply say: root -l <macro.C>), which can be conveniently invoked through a GUI that will appear at the end of the run of this macro. Launch the GUI via the command:

root -l -e 'TMVA::TMVAGui("TMVA.root")'
e
#define e(i)
Definition: RSha256.hxx:103
l
auto * l
Definition: textangle.C:4

Cross Evaluation

Cross evaluation is a special case of k-folds cross validation where the splitting into k folds is computed deterministically. This ensures that the a given event will always end up in the same fold.

In addition all resulting classifiers are saved and can be applied to new data using MethodCrossValidation. One requirement for this to work is a splitting function that is evaluated for each event to determine into what fold it goes (for training/evaluation) or to what classifier (for application).

Split Expression

Cross evaluation uses a deterministic split to partition the data into folds called the split expression. The expression can be any valid TFormula as long as all parts used are defined.

For each event the split expression is evaluated to a number and the event is put in the fold corresponding to that number.

It is recommended to always use int([NumFolds]) at the end of the expression.

The split expression has access to all spectators and variables defined in the dataloader. Additionally, the number of folds in the split can be accessed with NumFolds (or numFolds).

Example

"int(fabs([eventID]))%int([NumFolds])"
DataSetInfo : [dataset] : Added class "Signal"
: Add Tree of type Signal with 1000 events
DataSetInfo : [dataset] : Added class "Background"
: Add Tree of type Background with 1000 events
: Evaluate method: BDTG
<HEADER> Factory : Booking method: BDTG_fold1
:
<HEADER> BDTG_fold1 : #events: (reweighted) sig: 499.5 bkg: 499.5
: #events: (unweighted) sig: 493 bkg: 506
: Training 100 Decision Trees ... patience please
: Elapsed time for training with 999 events: 0.0475 sec
<HEADER> BDTG_fold1 : [dataset] : Evaluation of BDTG_fold1 on training sample (999 events)
: Elapsed time for evaluation of 999 events: 0.0036 sec
: Creating xml weight file: dataset/weights/TMVACrossValidation_BDTG_fold1.weights.xml
: Creating standalone class: dataset/weights/TMVACrossValidation_BDTG_fold1.class.C
<HEADER> Factory : Test all methods
<HEADER> Factory : Test method: BDTG_fold1 for Classification performance
:
<HEADER> BDTG_fold1 : [dataset] : Evaluation of BDTG_fold1 on testing sample (999 events)
: Elapsed time for evaluation of 999 events: 0.00362 sec
<HEADER> Factory : Evaluate all methods
<HEADER> Factory : Evaluate classifier: BDTG_fold1
:
<HEADER> BDTG_fold1 : [dataset] : Loop over test events and fill histograms with classifier response...
:
<HEADER> Factory : Thank you for using TMVA!
: For citation information, please visit: http://tmva.sf.net/citeTMVA.html
<HEADER> Factory : Booking method: BDTG_fold2
:
<HEADER> BDTG_fold2 : #events: (reweighted) sig: 499.5 bkg: 499.5
: #events: (unweighted) sig: 506 bkg: 493
: Training 100 Decision Trees ... patience please
: Elapsed time for training with 999 events: 0.0491 sec
<HEADER> BDTG_fold2 : [dataset] : Evaluation of BDTG_fold2 on training sample (999 events)
: Elapsed time for evaluation of 999 events: 0.00348 sec
: Creating xml weight file: dataset/weights/TMVACrossValidation_BDTG_fold2.weights.xml
: Creating standalone class: dataset/weights/TMVACrossValidation_BDTG_fold2.class.C
<HEADER> Factory : Test all methods
<HEADER> Factory : Test method: BDTG_fold2 for Classification performance
:
<HEADER> BDTG_fold2 : [dataset] : Evaluation of BDTG_fold2 on testing sample (999 events)
: Elapsed time for evaluation of 999 events: 0.00352 sec
<HEADER> Factory : Evaluate all methods
<HEADER> Factory : Evaluate classifier: BDTG_fold2
:
<HEADER> BDTG_fold2 : [dataset] : Loop over test events and fill histograms with classifier response...
:
<HEADER> Factory : Thank you for using TMVA!
: For citation information, please visit: http://tmva.sf.net/citeTMVA.html
<HEADER> Factory : Booking method: BDTG
:
: Reading weightfile: dataset/weights/TMVACrossValidation_BDTG_fold1.weights.xml
: Reading weight file: dataset/weights/TMVACrossValidation_BDTG_fold1.weights.xml
: Reading weightfile: dataset/weights/TMVACrossValidation_BDTG_fold2.weights.xml
: Reading weight file: dataset/weights/TMVACrossValidation_BDTG_fold2.weights.xml
: Evaluate method: Fisher
<HEADER> Factory : Booking method: Fisher_fold1
:
<HEADER> Fisher_fold1 : Results for Fisher coefficients:
: -----------------------
: Variable: Coefficient:
: -----------------------
: x: +0.471
: y: +0.440
: (offset): +0.000
: -----------------------
: Elapsed time for training with 999 events: 0.000377 sec
<HEADER> Fisher_fold1 : [dataset] : Evaluation of Fisher_fold1 on training sample (999 events)
: Elapsed time for evaluation of 999 events: 7.2e-05 sec
: Creating xml weight file: dataset/weights/TMVACrossValidation_Fisher_fold1.weights.xml
: Creating standalone class: dataset/weights/TMVACrossValidation_Fisher_fold1.class.C
<HEADER> Factory : Test all methods
<HEADER> Factory : Test method: Fisher_fold1 for Classification performance
:
<HEADER> Fisher_fold1 : [dataset] : Evaluation of Fisher_fold1 on testing sample (999 events)
: Elapsed time for evaluation of 999 events: 0.000124 sec
<HEADER> Factory : Evaluate all methods
<HEADER> Factory : Evaluate classifier: Fisher_fold1
:
<HEADER> Fisher_fold1 : [dataset] : Loop over test events and fill histograms with classifier response...
:
<HEADER> Factory : Thank you for using TMVA!
: For citation information, please visit: http://tmva.sf.net/citeTMVA.html
<HEADER> Factory : Booking method: Fisher_fold2
:
<HEADER> Fisher_fold2 : Results for Fisher coefficients:
: -----------------------
: Variable: Coefficient:
: -----------------------
: x: +0.475
: y: +0.463
: (offset): +0.020
: -----------------------
: Elapsed time for training with 999 events: 0.000263 sec
<HEADER> Fisher_fold2 : [dataset] : Evaluation of Fisher_fold2 on training sample (999 events)
: Elapsed time for evaluation of 999 events: 6.99e-05 sec
: Creating xml weight file: dataset/weights/TMVACrossValidation_Fisher_fold2.weights.xml
: Creating standalone class: dataset/weights/TMVACrossValidation_Fisher_fold2.class.C
<HEADER> Factory : Test all methods
<HEADER> Factory : Test method: Fisher_fold2 for Classification performance
:
<HEADER> Fisher_fold2 : [dataset] : Evaluation of Fisher_fold2 on testing sample (999 events)
: Elapsed time for evaluation of 999 events: 9.01e-05 sec
<HEADER> Factory : Evaluate all methods
<HEADER> Factory : Evaluate classifier: Fisher_fold2
:
<HEADER> Fisher_fold2 : [dataset] : Loop over test events and fill histograms with classifier response...
:
<HEADER> Factory : Thank you for using TMVA!
: For citation information, please visit: http://tmva.sf.net/citeTMVA.html
<HEADER> Factory : Booking method: Fisher
:
: Reading weightfile: dataset/weights/TMVACrossValidation_Fisher_fold1.weights.xml
: Reading weight file: dataset/weights/TMVACrossValidation_Fisher_fold1.weights.xml
: Reading weightfile: dataset/weights/TMVACrossValidation_Fisher_fold2.weights.xml
: Reading weight file: dataset/weights/TMVACrossValidation_Fisher_fold2.weights.xml
<HEADER> Factory : [dataset] : Create Transformation "I" with events from all classes.
:
<HEADER> : Transformation, Variable selection :
: Input : variable 'x' <---> Output : variable 'x'
: Input : variable 'y' <---> Output : variable 'y'
<HEADER> TFHandler_Factory : Variable Mean RMS [ Min Max ]
: -----------------------------------------------------------
: x: -0.014284 1.4061 [ -4.1075 4.0969 ]
: y: -0.0066370 1.4204 [ -4.8520 4.0761 ]
: -----------------------------------------------------------
: Ranking input variables (method unspecific)...
<HEADER> IdTransformation : Ranking result (top variable is best ranked)
: --------------------------
: Rank : Variable : Separation
: --------------------------
: 1 : x : 5.429e-01
: 2 : y : 5.230e-01
: --------------------------
: Elapsed time for training with 1998 events: 4.77e-06 sec
<HEADER> BDTG : [dataset] : Evaluation of BDTG on training sample (1998 events)
: Elapsed time for evaluation of 1998 events: 0.00668 sec
: Creating xml weight file: dataset/weights/TMVACrossValidation_BDTG.weights.xml
: Creating standalone class: dataset/weights/TMVACrossValidation_BDTG.class.C
<WARNING> <WARNING> : MakeClassSpecificHeader not implemented for CrossValidation
<WARNING> <WARNING> : MakeClassSpecific not implemented for CrossValidation
: Elapsed time for training with 1998 events: 4.05e-06 sec
<HEADER> Fisher : [dataset] : Evaluation of Fisher on training sample (1998 events)
: Elapsed time for evaluation of 1998 events: 0.000491 sec
: Creating xml weight file: dataset/weights/TMVACrossValidation_Fisher.weights.xml
: Creating standalone class: dataset/weights/TMVACrossValidation_Fisher.class.C
<WARNING> <WARNING> : MakeClassSpecificHeader not implemented for CrossValidation
<WARNING> <WARNING> : MakeClassSpecific not implemented for CrossValidation
<HEADER> Factory : Test all methods
<HEADER> Factory : Test method: BDTG for Classification performance
:
<HEADER> BDTG : [dataset] : Evaluation of BDTG on testing sample (1998 events)
: Elapsed time for evaluation of 1998 events: 0.00662 sec
<HEADER> Factory : Test method: Fisher for Classification performance
:
<HEADER> Fisher : [dataset] : Evaluation of Fisher on testing sample (1998 events)
: Elapsed time for evaluation of 1998 events: 0.000473 sec
<HEADER> Factory : Evaluate all methods
<HEADER> Factory : Evaluate classifier: BDTG
:
<HEADER> BDTG : [dataset] : Loop over test events and fill histograms with classifier response...
:
<HEADER> TFHandler_BDTG : Variable Mean RMS [ Min Max ]
: -----------------------------------------------------------
: x: -0.014284 1.4061 [ -4.1075 4.0969 ]
: y: -0.0066370 1.4204 [ -4.8520 4.0761 ]
: -----------------------------------------------------------
<HEADER> Factory : Evaluate classifier: Fisher
:
<HEADER> Fisher : [dataset] : Loop over test events and fill histograms with classifier response...
:
<HEADER> TFHandler_Fisher : Variable Mean RMS [ Min Max ]
: -----------------------------------------------------------
: x: -0.014284 1.4061 [ -4.1075 4.0969 ]
: y: -0.0066370 1.4204 [ -4.8520 4.0761 ]
: -----------------------------------------------------------
:
: Evaluation results ranked by best signal efficiency and purity (area)
: -------------------------------------------------------------------------------------------------------------------
: DataSet MVA
: Name: Method: ROC-integ
: dataset Fisher : 0.971
: dataset BDTG : 0.967
: -------------------------------------------------------------------------------------------------------------------
:
: Testing efficiency compared to training efficiency (overtraining check)
: -------------------------------------------------------------------------------------------------------------------
: DataSet MVA Signal efficiency: from test sample (from training sample)
: Name: Method: @B=0.01 @B=0.10 @B=0.30
: -------------------------------------------------------------------------------------------------------------------
: dataset Fisher : 0.654 (0.654) 0.923 (0.923) 0.980 (0.980)
: dataset BDTG : 0.630 (0.630) 0.920 (0.920) 0.976 (0.976)
: -------------------------------------------------------------------------------------------------------------------
:
<HEADER> Dataset:dataset : Created tree 'TestTree' with 1998 events
:
<HEADER> Dataset:dataset : Created tree 'TrainTree' with 1998 events
:
<HEADER> Factory : Thank you for using TMVA!
: For citation information, please visit: http://tmva.sf.net/citeTMVA.html
: Evaluation done.
Summary for method BDT
Fold 0: ROC int: 0.96956, BkgEff@SigEff=0.3: 0.982
Fold 1: ROC int: 0.966535, BkgEff@SigEff=0.3: 0.97
Summary for method Fisher
Fold 0: ROC int: 0.971244, BkgEff@SigEff=0.3: 0.981
Fold 1: ROC int: 0.970957, BkgEff@SigEff=0.3: 0.981
==> Wrote root file: TMVA.root
==> TMVACrossValidation is done!
(int) 0
#include <cstdlib>
#include <iostream>
#include <map>
#include <string>
#include "TChain.h"
#include "TFile.h"
#include "TTree.h"
#include "TString.h"
#include "TObjString.h"
#include "TSystem.h"
#include "TROOT.h"
#include "TMVA/CrossValidation.h"
#include "TMVA/DataLoader.h"
#include "TMVA/Factory.h"
#include "TMVA/Tools.h"
#include "TMVA/TMVAGui.h"
// Helper function to load data into TTrees.
TTree *genTree(Int_t nPoints, Double_t offset, Double_t scale, UInt_t seed = 100)
{
TRandom3 rng(seed);
Float_t x = 0;
Float_t y = 0;
UInt_t eventID = 0;
TTree *data = new TTree();
data->Branch("x", &x, "x/F");
data->Branch("y", &y, "y/F");
data->Branch("eventID", &eventID, "eventID/I");
for (Int_t n = 0; n < nPoints; ++n) {
x = rng.Gaus(offset, scale);
y = rng.Gaus(offset, scale);
// For our simple example it is enough that the id's are uniformly
// distributed and independent of the data.
++eventID;
data->Fill();
}
// Important: Disconnects the tree from the memory locations of x and y.
data->ResetBranchAddresses();
return data;
}
int TMVACrossValidation()
{
// This loads the library
TMVA::Tools::Instance();
// --------------------------------------------------------------------------
// Load the data into TTrees. If you load data from file you can use a
// variant of
// ```
// TString filename = "/path/to/file";
// TFile * input = TFile::Open( filename );
// TTree * signalTree = (TTree*)input->Get("TreeName");
// ```
TTree *sigTree = genTree(1000, 1.0, 1.0, 100);
TTree *bkgTree = genTree(1000, -1.0, 1.0, 101);
// Create a ROOT output file where TMVA will store ntuples, histograms, etc.
TString outfileName("TMVA.root");
TFile *outputFile = TFile::Open(outfileName, "RECREATE");
// DataLoader definitions; We declare variables in the tree so that TMVA can
// find them. For more information see TMVAClassification tutorial.
TMVA::DataLoader *dataloader = new TMVA::DataLoader("dataset");
// Data variables
dataloader->AddVariable("x", 'F');
dataloader->AddVariable("y", 'F');
// Spectator used for split
dataloader->AddSpectator("eventID", 'I');
// Attaches the trees so they can be read from
dataloader->AddSignalTree(sigTree, 1.0);
dataloader->AddBackgroundTree(bkgTree, 1.0);
// The CV mechanism of TMVA splits up the training set into several folds.
// The test set is currently left unused. The `nTest_ClassName=1` assigns
// one event to the the test set for each class and puts the rest in the
// training set. A value of 0 is a special value and would split the
// datasets 50 / 50.
dataloader->PrepareTrainingAndTestTree("", "",
"nTest_Signal=1"
":nTest_Background=1"
":SplitMode=Random"
":NormMode=NumEvents"
":!V");
// --------------------------------------------------------------------------
//
// This sets up a CrossValidation class (which wraps a TMVA::Factory
// internally) for 2-fold cross validation.
//
// The split type can be "Random", "RandomStratified" or "Deterministic".
// For the last option, check the comment below. Random splitting randomises
// the order of events and distributes events as evenly as possible.
// RandomStratified applies the same logic but distributes events within a
// class as evenly as possible over the folds.
//
UInt_t numFolds = 2;
TString analysisType = "Classification";
TString splitType = "Random";
TString splitExpr = "";
//
// One can also use a custom splitting function for producing the folds.
// The example uses a dataset spectator `eventID`.
//
// The idea here is that eventID should be an event number that is integral,
// random and independent of the data, generated only once. This last
// property ensures that if a calibration is changed the same event will
// still be assigned the same fold.
//
// This can be used to use the cross validated classifiers in application,
// a technique that can simplify statistical analysis.
//
// If you want to run TMVACrossValidationApplication, make sure you have
// run this tutorial with the below line uncommented first.
//
// TString splitExpr = "int(fabs([eventID]))%int([NumFolds])";
TString cvOptions = Form("!V"
":!Silent"
":ModelPersistence"
":AnalysisType=%s"
":SplitType=%s"
":NumFolds=%i"
":SplitExpr=%s",
analysisType.Data(), splitType.Data(), numFolds,
splitExpr.Data());
TMVA::CrossValidation cv{"TMVACrossValidation", dataloader, outputFile, cvOptions};
// --------------------------------------------------------------------------
//
// Books a method to use for evaluation
//
cv.BookMethod(TMVA::Types::kBDT, "BDTG",
"!H:!V:NTrees=100:MinNodeSize=2.5%:BoostType=Grad"
":NegWeightTreatment=Pray:Shrinkage=0.10:nCuts=20"
":MaxDepth=2");
cv.BookMethod(TMVA::Types::kFisher, "Fisher",
"!H:!V:Fisher:VarTransform=None");
// --------------------------------------------------------------------------
//
// Train, test and evaluate the booked methods.
// Evaluates the booked methods once for each fold and aggregates the result
// in the specified output file.
//
cv.Evaluate();
// --------------------------------------------------------------------------
//
// Process some output programatically, printing the ROC score for each
// booked method.
//
size_t iMethod = 0;
for (auto && result : cv.GetResults()) {
std::cout << "Summary for method " << cv.GetMethods()[iMethod++].GetValue<TString>("MethodName")
<< std::endl;
for (UInt_t iFold = 0; iFold<cv.GetNumFolds(); ++iFold) {
std::cout << "\tFold " << iFold << ": "
<< "ROC int: " << result.GetROCValues()[iFold]
<< ", "
<< "BkgEff@SigEff=0.3: " << result.GetEff30Values()[iFold]
<< std::endl;
}
}
// --------------------------------------------------------------------------
//
// Save the output
//
outputFile->Close();
std::cout << "==> Wrote root file: " << outputFile->GetName() << std::endl;
std::cout << "==> TMVACrossValidation is done!" << std::endl;
// --------------------------------------------------------------------------
//
// Launch the GUI for the root macros
//
if (!gROOT->IsBatch()) {
// Draw cv-specific graphs
cv.GetResults()[0].DrawAvgROCCurve(kTRUE, "Avg ROC for BDTG");
cv.GetResults()[0].DrawAvgROCCurve(kTRUE, "Avg ROC for Fisher");
// You can also use the classical gui
TMVA::TMVAGui(outfileName);
}
return 0;
}
//
// This is used if the macro is compiled. If run through ROOT with
// `root -l -b -q MACRO.C` or similar it is unused.
//
int main(int argc, char **argv)
{
TMVACrossValidation();
}
Int_t
int Int_t
Definition: RtypesCore.h:41
UInt_t
unsigned int UInt_t
Definition: RtypesCore.h:42
Double_t
double Double_t
Definition: RtypesCore.h:55
Float_t
float Float_t
Definition: RtypesCore.h:53
kTRUE
const Bool_t kTRUE
Definition: RtypesCore.h:87
gROOT
#define gROOT
Definition: TROOT.h:410
Form
char * Form(const char *fmt,...)
TFile
A ROOT file is a suite of consecutive data records (TKey instances) with a well defined format.
Definition: TFile.h:48
TFile::Close
virtual void Close(Option_t *option="")
Close a file.
Definition: TFile.cxx:912
TFile::Open
static TFile * Open(const char *name, Option_t *option="", const char *ftitle="", Int_t compress=ROOT::RCompressionSetting::EDefaults::kUseGeneralPurpose, Int_t netopt=0)
Create / open a file.
Definition: TFile.cxx:3975
TMVA::CrossValidation
Class to perform cross validation, splitting the dataloader into folds.
Definition: CrossValidation.h:122
TMVA::Tools::Instance
static Tools & Instance()
Definition: Tools.cxx:75
TMVA::Types::kFisher
@ kFisher
Definition: Types.h:84
TMVA::Types::kBDT
@ kBDT
Definition: Types.h:88
TNamed::GetName
virtual const char * GetName() const
Returns name of object.
Definition: TNamed.h:47
TRandom3
Random number generator class based on M.
Definition: TRandom3.h:27
TString
Basic string class.
Definition: TString.h:131
TString::Data
const char * Data() const
Definition: TString.h:364
TTree
A TTree object has a header with a name and a title.
Definition: TTree.h:71
main
int main(int argc, char **argv)
Definition: histspeedtest.cxx:752
y
Double_t y[n]
Definition: legend1.C:17
x
Double_t x[n]
Definition: legend1.C:17
n
const Int_t n
Definition: legend1.C:16
TMVA::TMVAGui
void TMVAGui(const char *fName="TMVA.root", TString dataset="")
Author
Kim Albertsson (adapted from code originally by Andreas Hoecker)

Definition in file TMVACrossValidation.C.