ROOT logo

class TMVA::MethodBDT: public TMVA::MethodBase

 Analysis of Boosted Decision Trees

 Boosted decision trees have been successfully used in High Energy
 Physics analysis for example by the MiniBooNE experiment
 (Yang-Roe-Zhu, physics/0508045). In Boosted Decision Trees, the
 selection is done on a majority vote on the result of several decision
 trees, which are all derived from the same training sample by
 supplying different event weights during the training.

 Decision trees:

 Successive decision nodes are used to categorize the
 events out of the sample as either signal or background. Each node
 uses only a single discriminating variable to decide if the event is
 signal-like ("goes right") or background-like ("goes left"). This
 forms a tree like structure with "baskets" at the end (leave nodes),
 and an event is classified as either signal or background according to
 whether the basket where it ends up has been classified signal or
 background during the training. Training of a decision tree is the
 process to define the "cut criteria" for each node. The training
 starts with the root node. Here one takes the full training event
 sample and selects the variable and corresponding cut value that gives
 the best separation between signal and background at this stage. Using
 this cut criterion, the sample is then divided into two subsamples, a
 signal-like (right) and a background-like (left) sample. Two new nodes
 are then created for each of the two sub-samples and they are
 constructed using the same mechanism as described for the root
 node. The devision is stopped once a certain node has reached either a
 minimum number of events, or a minimum or maximum signal purity. These
 leave nodes are then called "signal" or "background" if they contain
 more signal respective background events from the training sample.


 The idea behind the boosting is, that signal events from the training
 sample, that end up in a background node (and vice versa) are given a
 larger weight than events that are in the correct leave node. This
 results in a re-weighed training event sample, with which then a new
 decision tree can be developed. The boosting can be applied several
 times (typically 100-500 times) and one ends up with a set of decision
 trees (a forest).


 In this particular variant of the Boosted Decision Trees the boosting
 is not done on the basis of previous training results, but by a simple
 stochastic re-sampling of the initial training event sample.

 Random Trees:
 Similar to the "Random Forests" from Leo Breiman and Adele Cutler, it
 uses the bagging algorithm together and bases the determination of the
 best node-split during the training on a random subset of variables only
 which is individually chosen for each split.


 Applying an individual decision tree to a test event results in a
 classification of the event as either signal or background. For the
 boosted decision tree selection, an event is successively subjected to
 the whole set of decision trees and depending on how often it is
 classified as signal, a "likelihood" estimator is constructed for the
 event being signal or background. The value of this estimator is the
 one which is then used to select the events from an event sample, and
 the cut value on this estimator defines the efficiency and purity of
 the selection.

Function Members (Methods)

voidTObject::AbstractMethod(const char* method) const
voidTMVA::Configurable::AddOptionsXMLTo(void* parent) const
voidTMVA::MethodBase::AddOutput(TMVA::Types::ETreeType type, TMVA::Types::EAnalysisType analysisType)
virtual voidAddWeightsXMLTo(void* parent) const
virtual voidTObject::AppendPad(Option_t* option = "")
TDirectory*TMVA::MethodBase::BaseDir() const
Double_tBoost(vector<TMVA::Event*>, TMVA::DecisionTree* dt, Int_t iTree)
virtual voidTObject::Browse(TBrowser* b)
voidTMVA::Configurable::CheckForUnusedOptions() const
virtual voidTMVA::MethodBase::CheckSetup()
static TClass*Class()
virtual const char*TObject::ClassName() const
virtual voidTObject::Clear(Option_t* = "")
virtual TObject*TObject::Clone(const char* newname = "") const
virtual Int_tTObject::Compare(const TObject* obj) const
TMVA::ConfigurableTMVA::Configurable::Configurable(const TString& theOption = "")
virtual voidTObject::Copy(TObject& object) const
virtual const TMVA::Ranking*CreateRanking()
TMVA::DataSet*TMVA::MethodBase::Data() const
TMVA::DataSetInfo&TMVA::MethodBase::DataInfo() const
virtual voidDeclareOptions()
virtual voidTObject::Delete(Option_t* option = "")MENU
voidTMVA::MethodBase::DisableWriting(Bool_t setter)
virtual Int_tTObject::DistancetoPrimitive(Int_t px, Int_t py)
Bool_tTMVA::MethodBase::DoRegression() const
virtual voidTObject::Draw(Option_t* option = "")
virtual voidTObject::DrawClass() constMENU
virtual TObject*TObject::DrawClone(Option_t* option = "") constMENU
virtual voidTObject::Dump() constMENU
virtual voidTObject::Error(const char* method, const char* msgfmt) const
virtual voidTObject::Execute(const char* method, const char* params, Int_t* error = 0)
virtual voidTObject::Execute(TMethod* method, TObjArray* params, Int_t* error = 0)
virtual voidTObject::ExecuteEvent(Int_t event, Int_t px, Int_t py)
virtual voidTObject::Fatal(const char* method, const char* msgfmt) const
virtual TObject*TObject::FindObject(const char* name) const
virtual TObject*TObject::FindObject(const TObject* obj) const
TMVA::Types::EAnalysisTypeTMVA::MethodBase::GetAnalysisType() const
const vector<double>&GetBoostWeights() const
const char*TMVA::Configurable::GetConfigDescription() const
const char*TMVA::Configurable::GetConfigName() const
virtual Option_t*TObject::GetDrawOption() const
static Long_tTObject::GetDtorOnly()
virtual Double_tTMVA::MethodBase::GetEfficiency(const TString&, TMVA::Types::ETreeType, Double_t& err)
const TMVA::Event*TMVA::MethodBase::GetEvent() const
const TMVA::Event*TMVA::MethodBase::GetEvent(const TMVA::Event* ev) const
const TMVA::Event*TMVA::MethodBase::GetEvent(Long64_t ievt) const
const TMVA::Event*TMVA::MethodBase::GetEvent(Long64_t ievt, TMVA::Types::ETreeType type) const
const vector<TMVA::Event*>&TMVA::MethodBase::GetEventCollection(TMVA::Types::ETreeType type)
const vector<TMVA::DecisionTree*>&GetForest() const
virtual voidGetHelpMessage() const
virtual const char*TObject::GetIconName() const
const TString&TMVA::MethodBase::GetInputLabel(Int_t i) const
const TString&TMVA::MethodBase::GetInputTitle(Int_t i) const
const TString&TMVA::MethodBase::GetInputVar(Int_t i) const
const TString&TMVA::MethodBase::GetJobName() const
virtual Double_tTMVA::MethodBase::GetMaximumSignificance(Double_t SignalEvents, Double_t BackgroundEvents, Double_t& optimal_significance_value) const
Double_tTMVA::MethodBase::GetMean(Int_t ivar) const
const TString&TMVA::MethodBase::GetMethodName() const
TMVA::Types::EMVATMVA::MethodBase::GetMethodType() const
TStringTMVA::MethodBase::GetMethodTypeName() const
virtual Double_tGetMvaValue(Double_t* err = 0)
Double_tGetMvaValue(Double_t* err, UInt_t useNTrees)
virtual const char*TMVA::MethodBase::GetName() const
UInt_tTMVA::MethodBase::GetNEvents() const
UInt_tTMVA::MethodBase::GetNTargets() const
UInt_tTMVA::MethodBase::GetNvar() const
UInt_tTMVA::MethodBase::GetNVariables() const
virtual char*TObject::GetObjectInfo(Int_t px, Int_t py) const
static Bool_tTObject::GetObjectStat()
virtual Option_t*TObject::GetOption() const
const TString&TMVA::Configurable::GetOptions() const
virtual Double_tTMVA::MethodBase::GetProba(Double_t mvaVal, Double_t ap_sig)
const TStringTMVA::MethodBase::GetProbaName() const
virtual Double_tTMVA::MethodBase::GetRarity(Double_t mvaVal, TMVA::Types::ESBType reftype = Types::kBackground) const
virtual const vector<Float_t>&GetRegressionValues()
Double_tTMVA::MethodBase::GetRMS(Int_t ivar) const
virtual Double_tTMVA::MethodBase::GetROCIntegral(TMVA::PDF* pdfS = 0, TMVA::PDF* pdfB = 0) const
virtual Double_tTMVA::MethodBase::GetSeparation(TH1*, TH1*) const
virtual Double_tTMVA::MethodBase::GetSeparation(TMVA::PDF* pdfS = 0, TMVA::PDF* pdfB = 0) const
Double_tTMVA::MethodBase::GetSignalReferenceCut() const
virtual Double_tTMVA::MethodBase::GetSignificance() const
const TMVA::Event*TMVA::MethodBase::GetTestingEvent(Long64_t ievt) const
Double_tTMVA::MethodBase::GetTestTime() const
const TString&TMVA::MethodBase::GetTestvarName() const
virtual const char*TObject::GetTitle() const
virtual Double_tTMVA::MethodBase::GetTrainingEfficiency(const TString&)
const TMVA::Event*TMVA::MethodBase::GetTrainingEvent(Long64_t ievt) const
const vector<TMVA::Event*>&GetTrainingEvents() const
UInt_tTMVA::MethodBase::GetTrainingROOTVersionCode() const
TStringTMVA::MethodBase::GetTrainingROOTVersionString() const
UInt_tTMVA::MethodBase::GetTrainingTMVAVersionCode() const
TStringTMVA::MethodBase::GetTrainingTMVAVersionString() const
Double_tTMVA::MethodBase::GetTrainTime() const
const TMVA::TransformationHandler&TMVA::MethodBase::GetTransformationHandler() const
virtual UInt_tTObject::GetUniqueID() const
Double_tGetVariableImportance(UInt_t ivar)
TStringTMVA::MethodBase::GetWeightFileName() const
Double_tTMVA::MethodBase::GetXmax(Int_t ivar) const
Double_tTMVA::MethodBase::GetXmin(Int_t ivar) const
virtual Bool_tTObject::HandleTimer(TTimer* timer)
virtual Bool_tHasAnalysisType(TMVA::Types::EAnalysisType type, UInt_t numberClasses, UInt_t numberTargets)
virtual ULong_tTObject::Hash() const
Bool_tTMVA::MethodBase::HasMVAPdfs() const
virtual voidTObject::Info(const char* method, const char* msgfmt) const
virtual Bool_tTObject::InheritsFrom(const char* classname) const
virtual Bool_tTObject::InheritsFrom(const TClass* cl) const
virtual voidTObject::Inspect() constMENU
voidTObject::InvertBit(UInt_t f)
virtual TClass*IsA() const
virtual Bool_tTObject::IsEqual(const TObject* obj) const
virtual Bool_tTObject::IsFolder() const
Bool_tTObject::IsOnHeap() const
virtual Bool_tTMVA::MethodBase::IsSignalLike()
virtual Bool_tTObject::IsSortable() const
Bool_tTObject::IsZombie() const
virtual voidTObject::ls(Option_t* option = "") const
virtual voidTMVA::MethodBase::MakeClass(const TString& classFileName = TString("")) const
voidMakeClassInstantiateNode(TMVA::DecisionTreeNode* n, ostream& fout, const TString& className) const
virtual voidMakeClassSpecific(ostream&, const TString&) const
virtual voidMakeClassSpecificHeader(ostream&, const TString&) const
voidTObject::MayNotUse(const char* method) const
TDirectory*TMVA::MethodBase::MethodBaseDir() const
TMVA::MethodBDTMethodBDT(TMVA::DataSetInfo& theData, const TString& theWeightFile, TDirectory* theTargetDir = NULL)
TMVA::MethodBDTMethodBDT(const TString& jobName, const TString& methodTitle, TMVA::DataSetInfo& theData, const TString& theOption = "", TDirectory* theTargetDir = 0)
virtual Bool_tTMVA::MethodBase::MonitorBoost(TMVA::MethodBoost*)
virtual Bool_tTObject::Notify()
static voidTObject::operator delete(void* ptr)
static voidTObject::operator delete(void* ptr, void* vp)
static voidTObject::operator delete[](void* ptr)
static voidTObject::operator delete[](void* ptr, void* vp)
void*TObject::operator new(size_t sz)
void*TObject::operator new(size_t sz, void* vp)
void*TObject::operator new[](size_t sz)
void*TObject::operator new[](size_t sz, void* vp)
TMVA::IMethod&TMVA::IMethod::operator=(const TMVA::IMethod&)
virtual voidTObject::Paint(Option_t* option = "")
virtual voidTMVA::Configurable::ParseOptions()
virtual voidTObject::Pop()
virtual voidTObject::Print(Option_t* option = "") const
virtual voidTMVA::MethodBase::PrintHelpMessage() const
voidTMVA::Configurable::PrintOptions() const
virtual voidProcessOptions()
virtual Int_tTObject::Read(const char* name)
voidTMVA::Configurable::ReadOptionsFromStream(istream& istr)
voidTMVA::Configurable::ReadOptionsFromXML(void* node)
virtual voidReadWeightsFromStream(istream& istr)
virtual voidReadWeightsFromXML(void* parent)
virtual voidTObject::RecursiveRemove(TObject* obj)
voidTObject::ResetBit(UInt_t f)
virtual voidTObject::SaveAs(const char* filename = "", Option_t* option = "") constMENU
virtual voidTObject::SavePrimitive(basic_ostream<char,char_traits<char> >& out, Option_t* option = "")
virtual voidTMVA::MethodBase::SetAnalysisType(TMVA::Types::EAnalysisType type)
voidTMVA::MethodBase::SetBaseDir(TDirectory* methodDir)
voidTObject::SetBit(UInt_t f)
voidTObject::SetBit(UInt_t f, Bool_t set)
voidTMVA::Configurable::SetConfigDescription(const char* d)
voidTMVA::Configurable::SetConfigName(const char* n)
virtual voidTMVA::MethodBase::SetCurrentEvent(Long64_t ievt) const
virtual voidTObject::SetDrawOption(Option_t* option = "")MENU
static voidTObject::SetDtorOnly(void* obj)
voidTMVA::MethodBase::SetMethodBaseDir(TDirectory* methodDir)
voidTMVA::MethodBase::SetMethodDir(TDirectory* methodDir)
voidTMVA::Configurable::SetMsgType(TMVA::EMsgType t)
static voidTObject::SetObjectStat(Bool_t stat)
voidTMVA::Configurable::SetOptions(const TString& s)
voidTMVA::MethodBase::SetSignalReferenceCut(Double_t cut)
voidTMVA::MethodBase::SetTestTime(Double_t testTime)
voidTMVA::MethodBase::SetTestvarName(const TString& v = "")
voidTMVA::MethodBase::SetTrainTime(Double_t trainTime)
virtual voidTObject::SetUniqueID(UInt_t uid)
virtual voidShowMembers(TMemberInspector& insp, char* parent)
virtual voidStreamer(TBuffer& b)
voidStreamerNVirtual(TBuffer& b)
virtual voidTObject::SysError(const char* method, const char* msgfmt) const
Bool_tTObject::TestBit(UInt_t f) const
Int_tTObject::TestBits(UInt_t f) const
virtual voidTMVA::MethodBase::TestClassification()
virtual voidTMVA::MethodBase::TestRegression(Double_t& bias, Double_t& biasT, Double_t& dev, Double_t& devT, Double_t& rms, Double_t& rmsT, Double_t& mInf, Double_t& mInfT, Double_t& corr, TMVA::Types::ETreeType type)
Double_tTestTreeQuality(TMVA::DecisionTree* dt)
virtual voidTrain()
virtual voidTObject::UseCurrentStyle()
virtual voidTObject::Warning(const char* method, const char* msgfmt) const
virtual Int_tTObject::Write(const char* name = 0, Int_t option = 0, Int_t bufsize = 0)
virtual Int_tTObject::Write(const char* name = 0, Int_t option = 0, Int_t bufsize = 0) const
virtual voidTMVA::MethodBase::WriteEvaluationHistosToFile(TMVA::Types::ETreeType treetype)
virtual voidWriteMonitoringHistosToFile() const
voidTMVA::Configurable::WriteOptionsToStream(ostream& o, const TString& prefix) const
voidTMVA::MethodBase::WriteStateToFile() const
Double_tAdaBoost(vector<TMVA::Event*>, TMVA::DecisionTree* dt)
Double_tAdaBoostR2(vector<TMVA::Event*>, TMVA::DecisionTree* dt)
Double_tBagging(vector<TMVA::Event*>, Int_t iTree)
Double_tGetGradBoostMVA(TMVA::Event& e, UInt_t nTrees)
Double_tGetWeightedQuantile(vector<std::vector<Double_t> >& vec, const Double_t quantile, const Double_t SumOfWeights = 0.0)
Double_tGradBoost(vector<TMVA::Event*>, TMVA::DecisionTree* dt)
Double_tGradBoostRegression(vector<TMVA::Event*>, TMVA::DecisionTree* dt)
virtual voidInit()
Double_tRegBoost(vector<TMVA::Event*>, TMVA::DecisionTree* dt)
voidUpdateTargetsRegression(vector<TMVA::Event*>, Bool_t first = kFALSE)

Data Members

enum TMVA::MethodBase::EWeightFileType { kROOT
enum TMVA::MethodBase::ECutOrientation { kNegative
enum TObject::EStatusBits { kCanDelete
enum TObject::[unnamed] { kIsOnHeap
Bool_tTMVA::MethodBase::fSetupCompletedis method setup
const TMVA::Event*TMVA::MethodBase::fTmpEvent! temporary event when testing on a different DataSet than the own one
TMVA::Types::EAnalysisTypeTMVA::MethodBase::fAnalysisTypemethod-mode : true --> regression, false --> classification
UInt_tTMVA::MethodBase::fBackgroundClassindex of the Background-class
vector<TString>*TMVA::MethodBase::fInputVarsvector of input variables used in MVA
Int_tTMVA::MethodBase::fNbinsnumber of bins in representative histograms
Int_tTMVA::MethodBase::fNbinsHnumber of bins in evaluation histograms
TMVA::Ranking*TMVA::MethodBase::fRankingpointer to ranking object (created by derived classifiers)
vector<Float_t>*TMVA::MethodBase::fRegressionReturnValholds the return-value for the regression
UInt_tTMVA::MethodBase::fSignalClassindex of the Signal-class
Double_tfAdaBoostBetabeta parameter for AdaBoost algorithm
TStringfAdaBoostR2Lossloss type used in AdaBoostR2 (Linear,Quadratic or Exponential)
Bool_tfAutomaticuse user given prune strength or automatically determined one using a validation sample
Bool_tfBaggedGradBoostturn bagging in combination with grad boost on/off
TStringfBoostTypestring specifying the boost type
Double_tfBoostWeightntuple var: boost weight
vector<double>fBoostWeightsthe weights applied in the individual boosts
Double_tfErrorFractionntuple var: misclassification error fraction
vector<TMVA::Event*>fEventSamplethe training events
Double_tfFValidationEventsfraction of events to use for pruning
vector<TMVA::DecisionTree*,allocator<TMVA::DecisionTree*> >fForestthe collection of decision trees
Int_tfITreentuple var: ith tree
vector<double>fInitialWeightsthe initial event weights
UInt_tfMaxDepthmax depth
TTree*fMonitorNtuplemonitoring ntuple
Int_tfNCutsgrid used in cut applied in node splitting
UInt_tfNNodesMaxmax # of nodes
Int_tfNTreesnumber of decision trees requested
Bool_tfNoNegWeightsInTrainingignore negative event weights in the training
Int_tfNodeMinEventsmin number of events in node
Double_tfNodePurityLimitpurity limit for sig/bkg nodes
Bool_tfPruneBeforeBoostflag to prune before boosting
TMVA::DecisionTree::EPruneMethodfPruneMethodmethod used for prunig
TStringfPruneMethodSprune method option String
Double_tfPruneStrengtha parameter to set the "amount" of pruning..needs to be adjusted
Bool_tfRandomisedTreeschoose a random subset of possible cut variables at each node during training
vector<double>fRegResidualstemporary storage for regression residuals
Double_tfSampleFractionfraction of events used for bagged grad boost
Double_tfSampleSizeFractionrelative size of bagged event sample to original sample size
TMVA::SeparationBase*fSepTypethe separation used in node splitting
TStringfSepTypeSthe separation (option string) used in node splitting
Double_tfShrinkagelearning rate for gradient boost;
vector<TMVA::Event*>fSubSamplesubsample for bagged grad boost
Double_tfSumOfWeightstotal sum of all event weights
Double_tfTransitionPointbreak-down point for gradient regression
UInt_tfUseNTrainEventsnumber of randomly picked training events used in randomised (and bagged) trees
UInt_tfUseNvarsthe number of variables used in the randomised tree splitting
Bool_tfUseWeightedTreesuse average classification from the trees, or have the individual trees trees in the forest weighted (e.g. log(boostweight) from AdaBoost
Bool_tfUseYesNoLeafuse sig or bkg classification in leave nodes or sig/bkg
vector<TMVA::Event*>fValidationSamplethe Validation events
vector<Double_t>fVariableImportancethe relative importance of the different variables
static const Int_tfgDebugLeveldebug level determining some printout/control plots etc.

Class Charts

Inheritance Inherited Members Includes Libraries
Class Charts

Function documentation

MethodBDT(const TString& jobName, const TString& methodTitle, TMVA::DataSetInfo& theData, const TString& theOption = "", TDirectory* theTargetDir = 0)
 the standard constructor for the "boosted decision trees"
MethodBDT(TMVA::DataSetInfo& theData, const TString& theWeightFile, TDirectory* theTargetDir = NULL)
 constructor for calculating BDT-MVA using previously generated decision trees
 the result of the previous training (the decision trees) are read in via the
 weight file. Make sure the the variables correspond to the ones used in
 creating the "weight"-file
Bool_t HasAnalysisType(TMVA::Types::EAnalysisType type, UInt_t numberClasses, UInt_t numberTargets)
 BDT can handle classification with 2 classes and regression with one regression-target
void DeclareOptions()
 define the options (their key words) that can be set in the option string
 know options:
 nTrees        number of trees in the forest to be created
 BoostType     the boosting type for the trees in the forest (AdaBoost e.t.c..)
                  known: AdaBoost
                         AdaBoostR2 (Adaboost for regression)
 AdaBoostBeta     the boosting parameter, beta, for AdaBoost
 UseRandomisedTrees  choose at each node splitting a random set of variables
 UseNvars         use UseNvars variables in randomised trees
 UseNTrainEvents  number of training events used in randomised (and bagged) trees
 SeparationType   the separation criterion applied in the node splitting
                  known: GiniIndex
 nEventsMin:      the minimum number of events in a node (leaf criteria, stop splitting)
 nCuts:           the number of steps in the optimisation of the cut for a node (if < 0, then
                  step size is determined by the events)
 UseYesNoLeaf     decide if the classification is done simply by the node type, or the S/B
                  (from the training) in the leaf node
 NodePurityLimit  the minimum purity to classify a node as a signal node (used in pruning and boosting to determine
                  misclassification error rate)
 UseWeightedTrees use average classification from the trees, or have the individual trees
                  trees in the forest weighted (e.g. log(boostweight) from AdaBoost
 PruneMethod      The Pruning method:
                  known: NoPruning  // switch off pruning completely
 PruneStrength    a parameter to adjust the amount of pruning. Should be large enough such that overtraining is avoided.
 PruneBeforeBoost flag to prune the tree before applying boosting algorithm
 PruningValFraction   number of events to use for optimizing pruning (only if PruneStrength < 0, i.e. automatic pruning)
 IgnoreNegWeightsInTraining  Ignore negative weight events in the training.
 NNodesMax        maximum number of nodes allwed in the tree splitting, then it stops.
 MaxDepth         maximum depth of the decision tree allowed before further splitting is stopped
void DeclareCompatibilityOptions()
void ProcessOptions()
 the option string is decoded, for available options see "DeclareOptions"
void Init( void )
 common initialisation with defaults for the BDT-Method
~MethodBDT( void )
void InitEventSample( void )
 Write all Events from the Tree into a vector of Events, that are
 more easily manipulated. This method should never be called without
 existing trainingTree, as it the vector of events from the ROOT training tree
void Train()
 BDT training
void GetRandomSubSample()
 fills fEventSample with fSampleFraction*NEvents random training events
Double_t GetGradBoostMVA(TMVA::Event& e, UInt_t nTrees)
returns MVA value: -1 for background, 1 for signal
void UpdateTargets(vector<TMVA::Event*> )
Calculate residua for all events;
void UpdateTargetsRegression(vector<TMVA::Event*> , Bool_t first = kFALSE)
Calculate current residuals for all events and update targets for next iteration
Double_t GetWeightedQuantile(vector<std::vector<Double_t> >& vec, const Double_t quantile, const Double_t SumOfWeights = 0.0)
calculates the quantile of the distribution in vec[0] weighted with the values in vec[1]
Double_t GradBoost(vector<TMVA::Event*> , TMVA::DecisionTree* dt)
Calculate the desired response value for each region (line search)
Double_t GradBoostRegression(vector<TMVA::Event*> , TMVA::DecisionTree* dt)
 Implementation of M_TreeBoost using a Huber loss function as desribed by Friedman 1999
void InitGradBoost(vector<TMVA::Event*> )
 initialize targets for first tree
Double_t TestTreeQuality(TMVA::DecisionTree* dt)
 test the tree quality.. in terms of Miscalssification
Double_t Boost(vector<TMVA::Event*> , TMVA::DecisionTree* dt, Int_t iTree)
 apply the boosting alogrithim (the algorithm is selecte via the the "option" given
 in the constructor. The return value is the boosting weight
Double_t AdaBoost(vector<TMVA::Event*> , TMVA::DecisionTree* dt)
 the AdaBoost implementation.
 a new training sample is generated by weighting
 events that are misclassified by the decision tree. The weight
 applied is w = (1-err)/err or more general:
            w = ((1-err)/err)^beta
 where err is the fraction of misclassified events in the tree ( <0.5 assuming
 demanding the that previous selection was better than random guessing)
 and "beta" beeing a free parameter (standard: beta = 1) that modifies the
Double_t Bagging(vector<TMVA::Event*> , Int_t iTree)
 call it boot-strapping, re-sampling or whatever you like, in the end it is nothing
 else but applying "random" weights to each event.
Double_t RegBoost(vector<TMVA::Event*> , TMVA::DecisionTree* dt)
 a special boosting only for Regression ...
 maybe I'll implement it later...
Double_t AdaBoostR2(vector<TMVA::Event*> , TMVA::DecisionTree* dt)
 adaption of the AdaBoost to regression problems (see H.Drucker 1997)
void AddWeightsXMLTo(void* parent) const
 write weights to XML
void ReadWeightsFromXML(void* parent)
 reads the BDT from the xml file
void ReadWeightsFromStream(istream& istr)
 read the weights (BDT coefficients)
Double_t GetMvaValue(Double_t* err = 0)
Double_t GetMvaValue(Double_t* err, UInt_t useNTrees)
 Return the MVA value (range [-1;1]) that classifies the
 event according to the majority vote from the total number of
 decision trees.
const std::vector<Float_t> & GetRegressionValues()
 get the regression value generated by the BDTs
void WriteMonitoringHistosToFile( void )
 Here we could write some histograms created during the processing
 to the output file.
vector< Double_t > GetVariableImportance()
 Return the relative variable importance, normalized to all
 variables together having the importance 1. The importance in
 evaluated as the total separation-gain that this variable had in
 the decision trees (weighted by the number of events)
Double_t GetVariableImportance(UInt_t ivar)
 Returns the measure for the variable importance of variable "ivar"
 which is later used in GetVariableImportance() to calculate the
 relative variable importances.
const TMVA::Ranking* CreateRanking()
 Compute ranking of input variables
void GetHelpMessage() const
 Get help message text

 typical length of text line:
void MakeClassSpecific(ostream& , const TString& ) const
 make ROOT-independent C++ class for classifier response (classifier-specific implementation)
void MakeClassSpecificHeader(ostream& , const TString& ) const
 specific class header
void MakeClassInstantiateNode(TMVA::DecisionTreeNode* n, ostream& fout, const TString& className) const
 recursively descends a tree and writes the node instance to the output streem
const std::vector<TMVA::DecisionTree*>& GetForest() const
{ return fForest; }
const std::vector<TMVA::Event*>& GetTrainingEvents() const
{ return fEventSample; }
const std::vector<double>& GetBoostWeights() const
{ return fBoostWeights; }