Logo ROOT   6.14/05
Reference Guide
List of all members | Public Types | Public Member Functions | Public Attributes | Private Member Functions | Private Attributes | Friends | List of all members
TMVA::Factory Class Reference

This is the main MVA steering class.

It creates all MVA methods, and guides them through the training, testing and evaluation phases.

Definition at line 81 of file Factory.h.

Public Types

typedef std::vector< IMethod * > MVector
 
- Public Types inherited from TObject
enum  {
  kIsOnHeap = 0x01000000, kNotDeleted = 0x02000000, kZombie = 0x04000000, kInconsistent = 0x08000000,
  kBitMask = 0x00ffffff
}
 
enum  { kSingleKey = BIT(0), kOverwrite = BIT(1), kWriteDelete = BIT(2) }
 
enum  EDeprecatedStatusBits { kObjInCanvas = BIT(3) }
 
enum  EStatusBits {
  kCanDelete = BIT(0), kMustCleanup = BIT(3), kIsReferenced = BIT(4), kHasUUID = BIT(5),
  kCannotPick = BIT(6), kNoContextMenu = BIT(8), kInvalidObject = BIT(13)
}
 

Public Member Functions

 Factory (TString theJobName, TFile *theTargetFile, TString theOption="")
 Standard constructor. More...
 
 Factory (TString theJobName, TString theOption="")
 Constructor. More...
 
virtual ~Factory ()
 Destructor. More...
 
MethodBaseBookMethod (DataLoader *loader, TString theMethodName, TString methodTitle, TString theOption="")
 Book a classifier or regression method. More...
 
MethodBaseBookMethod (DataLoader *loader, Types::EMVA theMethod, TString methodTitle, TString theOption="")
 Books MVA method. More...
 
MethodBaseBookMethod (DataLoader *, TMVA::Types::EMVA, TString, TString, TMVA::Types::EMVA, TString)
 
void DeleteAllMethods (void)
 Delete methods. More...
 
void EvaluateAllMethods (void)
 Iterates over all MVAs that have been booked, and calls their evaluation methods. More...
 
void EvaluateAllVariables (DataLoader *loader, TString options="")
 Iterates over all MVA input variables and evaluates them. More...
 
TH1FEvaluateImportance (DataLoader *loader, VIType vitype, Types::EMVA theMethod, TString methodTitle, const char *theOption="")
 Evaluate Variable Importance. More...
 
IMethodGetMethod (const TString &datasetname, const TString &title) const
 Returns pointer to MVA that corresponds to given method title. More...
 
virtual const char * GetName () const
 Returns name of object. More...
 
TGraphGetROCCurve (DataLoader *loader, TString theMethodName, Bool_t setTitles=kTRUE, UInt_t iClass=0)
 Argument iClass specifies the class to generate the ROC curve in a multiclass setting. More...
 
TGraphGetROCCurve (TString datasetname, TString theMethodName, Bool_t setTitles=kTRUE, UInt_t iClass=0)
 Argument iClass specifies the class to generate the ROC curve in a multiclass setting. More...
 
TCanvasGetROCCurve (DataLoader *loader, UInt_t iClass=0)
 Draws ROC curves for all methods booked with the factory for a given class onto a canvas. More...
 
TCanvasGetROCCurve (TString datasetname, UInt_t iClass=0)
 Draws ROC curves for all methods booked with the factory for a given class. More...
 
TMultiGraphGetROCCurveAsMultiGraph (DataLoader *loader, UInt_t iClass)
 Generate a collection of graphs, for all methods for a given class. More...
 
TMultiGraphGetROCCurveAsMultiGraph (TString datasetname, UInt_t iClass)
 Generate a collection of graphs, for all methods for a given class. More...
 
Double_t GetROCIntegral (DataLoader *loader, TString theMethodName, UInt_t iClass=0)
 Calculate the integral of the ROC curve, also known as the area under curve (AUC), for a given method. More...
 
Double_t GetROCIntegral (TString datasetname, TString theMethodName, UInt_t iClass=0)
 Calculate the integral of the ROC curve, also known as the area under curve (AUC), for a given method. More...
 
Bool_t HasMethod (const TString &datasetname, const TString &title) const
 Checks whether a given method name is defined for a given dataset. More...
 
Bool_t IsModelPersistence ()
 
Bool_t IsSilentFile ()
 
virtual void MakeClass (const TString &datasetname, const TString &methodTitle="") const
 
std::map< TString, Double_tOptimizeAllMethods (TString fomType="ROCIntegral", TString fitType="FitGA")
 Iterates through all booked methods and sees if they use parameter tuning and if so. More...
 
void OptimizeAllMethodsForClassification (TString fomType="ROCIntegral", TString fitType="FitGA")
 
void OptimizeAllMethodsForRegression (TString fomType="ROCIntegral", TString fitType="FitGA")
 
void PrintHelpMessage (const TString &datasetname, const TString &methodTitle="") const
 Print predefined help message of classifier. More...
 
TDirectoryRootBaseDir ()
 
void SetVerbose (Bool_t v=kTRUE)
 
void TestAllMethods ()
 Evaluates all booked methods on the testing data and adds the output to the Results in the corresponiding DataSet. More...
 
void TrainAllMethods ()
 Iterates through all booked methods and calls training. More...
 
void TrainAllMethodsForClassification (void)
 
void TrainAllMethodsForRegression (void)
 
Bool_t Verbose (void) const
 
- Public Member Functions inherited from TMVA::Configurable
 Configurable (const TString &theOption="")
 constructor More...
 
virtual ~Configurable ()
 default destructor More...
 
void AddOptionsXMLTo (void *parent) const
 write options to XML file More...
 
template<class T >
void AddPreDefVal (const T &)
 
template<class T >
void AddPreDefVal (const TString &optname, const T &)
 
void CheckForUnusedOptions () const
 checks for unused options in option string More...
 
template<class T >
OptionBaseDeclareOptionRef (T &ref, const TString &name, const TString &desc="")
 
template<class T >
OptionBaseDeclareOptionRef (T *&ref, Int_t size, const TString &name, const TString &desc="")
 
template<class T >
TMVA::OptionBaseDeclareOptionRef (T &ref, const TString &name, const TString &desc)
 
template<class T >
TMVA::OptionBaseDeclareOptionRef (T *&ref, Int_t size, const TString &name, const TString &desc)
 
const char * GetConfigDescription () const
 
const char * GetConfigName () const
 
const TStringGetOptions () const
 
MsgLoggerLog () const
 
virtual void ParseOptions ()
 options parser More...
 
void PrintOptions () const
 prints out the options set in the options string and the defaults More...
 
void ReadOptionsFromStream (std::istream &istr)
 read option back from the weight file More...
 
void ReadOptionsFromXML (void *node)
 
void SetConfigDescription (const char *d)
 
void SetConfigName (const char *n)
 
void SetMsgType (EMsgType t)
 
void SetOptions (const TString &s)
 
void WriteOptionsToStream (std::ostream &o, const TString &prefix) const
 write options to output stream (e.g. in writing the MVA weight files More...
 
- Public Member Functions inherited from TNamed
 TNamed ()
 
 TNamed (const char *name, const char *title)
 
 TNamed (const TString &name, const TString &title)
 
 TNamed (const TNamed &named)
 TNamed copy ctor. More...
 
virtual ~TNamed ()
 TNamed destructor. More...
 
virtual void Clear (Option_t *option="")
 Set name and title to empty strings (""). More...
 
virtual TObjectClone (const char *newname="") const
 Make a clone of an object using the Streamer facility. More...
 
virtual Int_t Compare (const TObject *obj) const
 Compare two TNamed objects. More...
 
virtual void Copy (TObject &named) const
 Copy this to obj. More...
 
virtual void FillBuffer (char *&buffer)
 Encode TNamed into output buffer. More...
 
virtual const char * GetTitle () const
 Returns title of object. More...
 
virtual ULong_t Hash () const
 Return hash value for this object. More...
 
virtual Bool_t IsSortable () const
 
virtual void ls (Option_t *option="") const
 List TNamed name and title. More...
 
TNamedoperator= (const TNamed &rhs)
 TNamed assignment operator. More...
 
virtual void Print (Option_t *option="") const
 Print TNamed name and title. More...
 
virtual void SetName (const char *name)
 Set the name of the TNamed. More...
 
virtual void SetNameTitle (const char *name, const char *title)
 Set all the TNamed parameters (name and title). More...
 
virtual void SetTitle (const char *title="")
 Set the title of the TNamed. More...
 
virtual Int_t Sizeof () const
 Return size of the TNamed part of the TObject. More...
 
- Public Member Functions inherited from TObject
 TObject ()
 TObject constructor. More...
 
 TObject (const TObject &object)
 TObject copy ctor. More...
 
virtual ~TObject ()
 TObject destructor. More...
 
void AbstractMethod (const char *method) const
 Use this method to implement an "abstract" method that you don't want to leave purely abstract. More...
 
virtual void AppendPad (Option_t *option="")
 Append graphics object to current pad. More...
 
virtual void Browse (TBrowser *b)
 Browse object. May be overridden for another default action. More...
 
ULong_t CheckedHash ()
 Checked and record whether for this class has a consistent Hash/RecursiveRemove setup (*) and then return the regular Hash value for this object. More...
 
virtual const char * ClassName () const
 Returns name of class to which the object belongs. More...
 
virtual void Delete (Option_t *option="")
 Delete this object. More...
 
virtual Int_t DistancetoPrimitive (Int_t px, Int_t py)
 Computes distance from point (px,py) to the object. More...
 
virtual void Draw (Option_t *option="")
 Default Draw method for all objects. More...
 
virtual void DrawClass () const
 Draw class inheritance tree of the class to which this object belongs. More...
 
virtual TObjectDrawClone (Option_t *option="") const
 Draw a clone of this object in the current selected pad for instance with: gROOT->SetSelectedPad(gPad). More...
 
virtual void Dump () const
 Dump contents of object on stdout. More...
 
virtual void Error (const char *method, const char *msgfmt,...) const
 Issue error message. More...
 
virtual void Execute (const char *method, const char *params, Int_t *error=0)
 Execute method on this object with the given parameter string, e.g. More...
 
virtual void Execute (TMethod *method, TObjArray *params, Int_t *error=0)
 Execute method on this object with parameters stored in the TObjArray. More...
 
virtual void ExecuteEvent (Int_t event, Int_t px, Int_t py)
 Execute action corresponding to an event at (px,py). More...
 
virtual void Fatal (const char *method, const char *msgfmt,...) const
 Issue fatal error message. More...
 
virtual TObjectFindObject (const char *name) const
 Must be redefined in derived classes. More...
 
virtual TObjectFindObject (const TObject *obj) const
 Must be redefined in derived classes. More...
 
virtual Option_tGetDrawOption () const
 Get option used by the graphics system to draw this object. More...
 
virtual const char * GetIconName () const
 Returns mime type name of object. More...
 
virtual char * GetObjectInfo (Int_t px, Int_t py) const
 Returns string containing info about the object at position (px,py). More...
 
virtual Option_tGetOption () const
 
virtual UInt_t GetUniqueID () const
 Return the unique object id. More...
 
virtual Bool_t HandleTimer (TTimer *timer)
 Execute action in response of a timer timing out. More...
 
Bool_t HasInconsistentHash () const
 Return true is the type of this object is known to have an inconsistent setup for Hash and RecursiveRemove (i.e. More...
 
virtual void Info (const char *method, const char *msgfmt,...) const
 Issue info message. More...
 
virtual Bool_t InheritsFrom (const char *classname) const
 Returns kTRUE if object inherits from class "classname". More...
 
virtual Bool_t InheritsFrom (const TClass *cl) const
 Returns kTRUE if object inherits from TClass cl. More...
 
virtual void Inspect () const
 Dump contents of this object in a graphics canvas. More...
 
void InvertBit (UInt_t f)
 
virtual Bool_t IsEqual (const TObject *obj) const
 Default equal comparison (objects are equal if they have the same address in memory). More...
 
virtual Bool_t IsFolder () const
 Returns kTRUE in case object contains browsable objects (like containers or lists of other objects). More...
 
R__ALWAYS_INLINE Bool_t IsOnHeap () const
 
R__ALWAYS_INLINE Bool_t IsZombie () const
 
void MayNotUse (const char *method) const
 Use this method to signal that a method (defined in a base class) may not be called in a derived class (in principle against good design since a child class should not provide less functionality than its parent, however, sometimes it is necessary). More...
 
virtual Bool_t Notify ()
 This method must be overridden to handle object notification. More...
 
void Obsolete (const char *method, const char *asOfVers, const char *removedFromVers) const
 Use this method to declare a method obsolete. More...
 
void operator delete (void *ptr)
 Operator delete. More...
 
void operator delete[] (void *ptr)
 Operator delete []. More...
 
voidoperator new (size_t sz)
 
voidoperator new (size_t sz, void *vp)
 
voidoperator new[] (size_t sz)
 
voidoperator new[] (size_t sz, void *vp)
 
TObjectoperator= (const TObject &rhs)
 TObject assignment operator. More...
 
virtual void Paint (Option_t *option="")
 This method must be overridden if a class wants to paint itself. More...
 
virtual void Pop ()
 Pop on object drawn in a pad to the top of the display list. More...
 
virtual Int_t Read (const char *name)
 Read contents of object with specified name from the current directory. More...
 
virtual void RecursiveRemove (TObject *obj)
 Recursively remove this object from a list. More...
 
void ResetBit (UInt_t f)
 
virtual void SaveAs (const char *filename="", Option_t *option="") const
 Save this object in the file specified by filename. More...
 
virtual void SavePrimitive (std::ostream &out, Option_t *option="")
 Save a primitive as a C++ statement(s) on output stream "out". More...
 
void SetBit (UInt_t f, Bool_t set)
 Set or unset the user status bits as specified in f. More...
 
void SetBit (UInt_t f)
 
virtual void SetDrawOption (Option_t *option="")
 Set drawing option for object. More...
 
virtual void SetUniqueID (UInt_t uid)
 Set the unique object id. More...
 
virtual void SysError (const char *method, const char *msgfmt,...) const
 Issue system error message. More...
 
R__ALWAYS_INLINE Bool_t TestBit (UInt_t f) const
 
Int_t TestBits (UInt_t f) const
 
virtual void UseCurrentStyle ()
 Set current style settings in this object This function is called when either TCanvas::UseCurrentStyle or TROOT::ForceStyle have been invoked. More...
 
virtual void Warning (const char *method, const char *msgfmt,...) const
 Issue warning message. More...
 
virtual Int_t Write (const char *name=0, Int_t option=0, Int_t bufsize=0)
 Write this object to the current directory. More...
 
virtual Int_t Write (const char *name=0, Int_t option=0, Int_t bufsize=0) const
 Write this object to the current directory. More...
 

Public Attributes

std::map< TString, MVector * > fMethodsMap
 

Private Member Functions

MethodBaseBookMethodWeightfile (DataLoader *dataloader, TMVA::Types::EMVA methodType, const TString &weightfile)
 Adds an already constructed method to be managed by this factory. More...
 
TH1FEvaluateImportanceAll (DataLoader *loader, Types::EMVA theMethod, TString methodTitle, const char *theOption="")
 
TH1FEvaluateImportanceRandom (DataLoader *loader, UInt_t nseeds, Types::EMVA theMethod, TString methodTitle, const char *theOption="")
 
TH1FEvaluateImportanceShort (DataLoader *loader, Types::EMVA theMethod, TString methodTitle, const char *theOption="")
 
TH1FGetImportance (const int nbits, std::vector< Double_t > importances, std::vector< TString > varNames)
 
ROCCurveGetROC (DataLoader *loader, TString theMethodName, UInt_t iClass=0, Types::ETreeType type=Types::kTesting)
 Private method to generate a ROCCurve instance for a given method. More...
 
ROCCurveGetROC (TString datasetname, TString theMethodName, UInt_t iClass=0, Types::ETreeType type=Types::kTesting)
 Private method to generate a ROCCurve instance for a given method. More...
 
void Greetings ()
 Print welcome message. More...
 
void SetInputTreesFromEventAssignTrees ()
 
void WriteDataInformation (DataSetInfo &fDataSetInfo)
 

Private Attributes

Types::EAnalysisType fAnalysisType
 jobname, used as extension in weight file names More...
 
Bool_t fCorrelations
 verbosity level, controls granularity of logging More...
 
std::vector< TMVA::VariableTransformBase * > fDefaultTrfs
 ROOT output file. More...
 
TFilefgTargetFile
 
TString fJobName
 used in contructor wihtout file More...
 
Bool_t fModelPersistence
 the training type More...
 
TString fOptions
 list of transformations on default DataSet More...
 
Bool_t fROC
 enable to calculate corelations More...
 
Bool_t fSilentFile
 enable to calculate ROC values More...
 
TString fTransformations
 option string given by construction (presently only "V") More...
 
Bool_t fVerbose
 list of transformations to test More...
 
TString fVerboseLevel
 verbose mode More...
 

Friends

class CrossValidation
 

Additional Inherited Members

- Static Public Member Functions inherited from TObject
static Long_t GetDtorOnly ()
 Return destructor only flag. More...
 
static Bool_t GetObjectStat ()
 Get status of object stat flag. More...
 
static void SetDtorOnly (void *obj)
 Set destructor only flag. More...
 
static void SetObjectStat (Bool_t stat)
 Turn on/off tracking of objects in the TObjectTable. More...
 
- Protected Member Functions inherited from TMVA::Configurable
void EnableLooseOptions (Bool_t b=kTRUE)
 
const TStringGetReferenceFile () const
 
Bool_t LooseOptionCheckingEnabled () const
 
void ResetSetFlag ()
 resets the IsSet flag for all declare options to be called before options are read from stream More...
 
void WriteOptionsReferenceToFile ()
 write complete options to output stream More...
 
- Protected Member Functions inherited from TObject
virtual void DoError (int level, const char *location, const char *fmt, va_list va) const
 Interface to ErrorHandler (protected). More...
 
void MakeZombie ()
 
- Protected Attributes inherited from TMVA::Configurable
MsgLoggerfLogger
 
- Protected Attributes inherited from TNamed
TString fName
 
TString fTitle
 

#include <TMVA/Factory.h>

Inheritance diagram for TMVA::Factory:
[legend]

Member Typedef Documentation

◆ MVector

typedef std::vector<IMethod*> TMVA::Factory::MVector

Definition at line 85 of file Factory.h.

Constructor & Destructor Documentation

◆ Factory() [1/2]

TMVA::Factory::Factory ( TString  jobName,
TFile theTargetFile,
TString  theOption = "" 
)

Standard constructor.

  • jobname : this name will appear in all weight file names produced by the MVAs
  • theTargetFile : output ROOT file; the test tree and all evaluation plots will be stored here
  • theOption : option string; currently: "V" for verbose

Definition at line 119 of file Factory.cxx.

◆ Factory() [2/2]

TMVA::Factory::Factory ( TString  theJobName,
TString  theOption = "" 
)

Constructor.

Definition at line 203 of file Factory.cxx.

◆ ~Factory()

TMVA::Factory::~Factory ( void  )
virtual

Destructor.

Definition at line 312 of file Factory.cxx.

Member Function Documentation

◆ BookMethod() [1/3]

TMVA::MethodBase * TMVA::Factory::BookMethod ( TMVA::DataLoader loader,
TString  theMethodName,
TString  methodTitle,
TString  theOption = "" 
)

Book a classifier or regression method.

Definition at line 358 of file Factory.cxx.

◆ BookMethod() [2/3]

TMVA::MethodBase * TMVA::Factory::BookMethod ( TMVA::DataLoader loader,
Types::EMVA  theMethod,
TString  methodTitle,
TString  theOption = "" 
)

Books MVA method.

The option configuration string is custom for each MVA the TString field "theNameAppendix" serves to define (and distinguish) several instances of a given MVA, eg, when one wants to compare the performance of various configurations

Definition at line 490 of file Factory.cxx.

◆ BookMethod() [3/3]

MethodBase* TMVA::Factory::BookMethod ( DataLoader ,
TMVA::Types::EMVA  ,
TString  ,
TString  ,
TMVA::Types::EMVA  ,
TString   
)
inline

Definition at line 102 of file Factory.h.

◆ BookMethodWeightfile()

TMVA::MethodBase * TMVA::Factory::BookMethodWeightfile ( DataLoader loader,
TMVA::Types::EMVA  methodType,
const TString weightfile 
)
private

Adds an already constructed method to be managed by this factory.

Note
Private.
Know what you are doing when using this method. The method that you are loading could be trained already.

Definition at line 503 of file Factory.cxx.

◆ DeleteAllMethods()

void TMVA::Factory::DeleteAllMethods ( void  )

Delete methods.

Definition at line 330 of file Factory.cxx.

◆ EvaluateAllMethods()

void TMVA::Factory::EvaluateAllMethods ( void  )

Iterates over all MVAs that have been booked, and calls their evaluation methods.

Definition at line 1333 of file Factory.cxx.

◆ EvaluateAllVariables()

void TMVA::Factory::EvaluateAllVariables ( DataLoader loader,
TString  options = "" 
)

Iterates over all MVA input variables and evaluates them.

Definition at line 1318 of file Factory.cxx.

◆ EvaluateImportance()

TH1F * TMVA::Factory::EvaluateImportance ( DataLoader loader,
VIType  vitype,
Types::EMVA  theMethod,
TString  methodTitle,
const char *  theOption = "" 
)

Evaluate Variable Importance.

Definition at line 2148 of file Factory.cxx.

◆ EvaluateImportanceAll()

TH1F * TMVA::Factory::EvaluateImportanceAll ( DataLoader loader,
Types::EMVA  theMethod,
TString  methodTitle,
const char *  theOption = "" 
)
private

Definition at line 2171 of file Factory.cxx.

◆ EvaluateImportanceRandom()

TH1F * TMVA::Factory::EvaluateImportanceRandom ( DataLoader loader,
UInt_t  nseeds,
Types::EMVA  theMethod,
TString  methodTitle,
const char *  theOption = "" 
)
private

Definition at line 2372 of file Factory.cxx.

◆ EvaluateImportanceShort()

TH1F * TMVA::Factory::EvaluateImportanceShort ( DataLoader loader,
Types::EMVA  theMethod,
TString  methodTitle,
const char *  theOption = "" 
)
private

Definition at line 2267 of file Factory.cxx.

◆ GetImportance()

TH1F * TMVA::Factory::GetImportance ( const int  nbits,
std::vector< Double_t importances,
std::vector< TString varNames 
)
private

Definition at line 2484 of file Factory.cxx.

◆ GetMethod()

TMVA::IMethod * TMVA::Factory::GetMethod ( const TString datasetname,
const TString title 
) const

Returns pointer to MVA that corresponds to given method title.

Definition at line 561 of file Factory.cxx.

◆ GetName()

virtual const char* TMVA::Factory::GetName ( ) const
inlinevirtual

Returns name of object.

This default method returns the class name. Classes that give objects a name should override this method.

Reimplemented from TNamed.

Definition at line 97 of file Factory.h.

◆ GetROC() [1/2]

TMVA::ROCCurve * TMVA::Factory::GetROC ( TMVA::DataLoader loader,
TString  theMethodName,
UInt_t  iClass = 0,
Types::ETreeType  type = Types::kTesting 
)
private

Private method to generate a ROCCurve instance for a given method.

Handles the conversion from TMVA ResultSet to a format the ROCCurve class understands.

Note
You own the retured pointer.

Definition at line 743 of file Factory.cxx.

◆ GetROC() [2/2]

TMVA::ROCCurve * TMVA::Factory::GetROC ( TString  datasetname,
TString  theMethodName,
UInt_t  iClass = 0,
Types::ETreeType  type = Types::kTesting 
)
private

Private method to generate a ROCCurve instance for a given method.

Handles the conversion from TMVA ResultSet to a format the ROCCurve class understands.

Note
You own the retured pointer.

Definition at line 757 of file Factory.cxx.

◆ GetROCCurve() [1/4]

TGraph * TMVA::Factory::GetROCCurve ( DataLoader loader,
TString  theMethodName,
Bool_t  setTitles = kTRUE,
UInt_t  iClass = 0 
)

Argument iClass specifies the class to generate the ROC curve in a multiclass setting.

It is ignored for binary classification.

Returns a ROC graph for a given method, or nullptr on error.

Note: Evaluation of the given method must have been run prior to ROC generation through Factory::EvaluateAllMetods.

NOTE: The ROC curve is 1 vs. all where the given class is considered signal and the others considered background. This is ok in binary classification but in in multi class classification, the ROC surface is an N dimensional shape, where N is number of classes - 1.

Definition at line 903 of file Factory.cxx.

◆ GetROCCurve() [2/4]

TGraph * TMVA::Factory::GetROCCurve ( TString  datasetname,
TString  theMethodName,
Bool_t  setTitles = kTRUE,
UInt_t  iClass = 0 
)

Argument iClass specifies the class to generate the ROC curve in a multiclass setting.

It is ignored for binary classification.

Returns a ROC graph for a given method, or nullptr on error.

Note: Evaluation of the given method must have been run prior to ROC generation through Factory::EvaluateAllMetods.

NOTE: The ROC curve is 1 vs. all where the given class is considered signal and the others considered background. This is ok in binary classification but in in multi class classification, the ROC surface is an N dimensional shape, where N is number of classes - 1.

Definition at line 922 of file Factory.cxx.

◆ GetROCCurve() [3/4]

TCanvas * TMVA::Factory::GetROCCurve ( TMVA::DataLoader loader,
UInt_t  iClass = 0 
)

Draws ROC curves for all methods booked with the factory for a given class onto a canvas.

Argument iClass specifies the class to generate the ROC curve in a multiclass setting. It is ignored for binary classification.

NOTE: The ROC curve is 1 vs. all where the given class is considered signal and the others considered background. This is ok in binary classification but in in multi class classification, the ROC surface is an N dimensional shape, where N is number of classes - 1.

Definition at line 1040 of file Factory.cxx.

◆ GetROCCurve() [4/4]

TCanvas * TMVA::Factory::GetROCCurve ( TString  datasetname,
UInt_t  iClass = 0 
)

Draws ROC curves for all methods booked with the factory for a given class.

Argument iClass specifies the class to generate the ROC curve in a multiclass setting. It is ignored for binary classification.

NOTE: The ROC curve is 1 vs. all where the given class is considered signal and the others considered background. This is ok in binary classification but in in multi class classification, the ROC surface is an N dimensional shape, where N is number of classes - 1.

Definition at line 1056 of file Factory.cxx.

◆ GetROCCurveAsMultiGraph() [1/2]

TMultiGraph * TMVA::Factory::GetROCCurveAsMultiGraph ( DataLoader loader,
UInt_t  iClass 
)

Generate a collection of graphs, for all methods for a given class.

Suitable for comparing method performance.

Argument iClass specifies the class to generate the ROC curve in a multiclass setting. It is ignored for binary classification.

NOTE: The ROC curve is 1 vs. all where the given class is considered signal and the others considered background. This is ok in binary classification but in in multi class classification, the ROC surface is an N dimensional shape, where N is number of classes - 1.

Definition at line 972 of file Factory.cxx.

◆ GetROCCurveAsMultiGraph() [2/2]

TMultiGraph * TMVA::Factory::GetROCCurveAsMultiGraph ( TString  datasetname,
UInt_t  iClass 
)

Generate a collection of graphs, for all methods for a given class.

Suitable for comparing method performance.

Argument iClass specifies the class to generate the ROC curve in a multiclass setting. It is ignored for binary classification.

NOTE: The ROC curve is 1 vs. all where the given class is considered signal and the others considered background. This is ok in binary classification but in in multi class classification, the ROC surface is an N dimensional shape, where N is number of classes - 1.

Definition at line 989 of file Factory.cxx.

◆ GetROCIntegral() [1/2]

Double_t TMVA::Factory::GetROCIntegral ( TMVA::DataLoader loader,
TString  theMethodName,
UInt_t  iClass = 0 
)

Calculate the integral of the ROC curve, also known as the area under curve (AUC), for a given method.

Argument iClass specifies the class to generate the ROC curve in a multiclass setting. It is ignored for binary classification.

Definition at line 842 of file Factory.cxx.

◆ GetROCIntegral() [2/2]

Double_t TMVA::Factory::GetROCIntegral ( TString  datasetname,
TString  theMethodName,
UInt_t  iClass = 0 
)

Calculate the integral of the ROC curve, also known as the area under curve (AUC), for a given method.

Argument iClass specifies the class to generate the ROC curve in a multiclass setting. It is ignored for binary classification.

Definition at line 855 of file Factory.cxx.

◆ Greetings()

void TMVA::Factory::Greetings ( )
private

Print welcome message.

Options are: kLogoWelcomeMsg, kIsometricWelcomeMsg, kLeanWelcomeMsg

Definition at line 288 of file Factory.cxx.

◆ HasMethod()

Bool_t TMVA::Factory::HasMethod ( const TString datasetname,
const TString title 
) const

Checks whether a given method name is defined for a given dataset.

Definition at line 579 of file Factory.cxx.

◆ IsModelPersistence()

Bool_t TMVA::Factory::IsModelPersistence ( )

Definition at line 304 of file Factory.cxx.

◆ IsSilentFile()

Bool_t TMVA::Factory::IsSilentFile ( )

Definition at line 297 of file Factory.cxx.

◆ MakeClass()

void TMVA::Factory::MakeClass ( const TString datasetname,
const TString methodTitle = "" 
) const
virtual

Definition at line 1263 of file Factory.cxx.

◆ OptimizeAllMethods()

std::map< TString, Double_t > TMVA::Factory::OptimizeAllMethods ( TString  fomType = "ROCIntegral",
TString  fitType = "FitGA" 
)

Iterates through all booked methods and sees if they use parameter tuning and if so.

does just that i.e. calls "Method::Train()" for different parameter settings and keeps in mind the "optimal one"... and that's the one that will later on be used in the main training loop.

Definition at line 694 of file Factory.cxx.

◆ OptimizeAllMethodsForClassification()

void TMVA::Factory::OptimizeAllMethodsForClassification ( TString  fomType = "ROCIntegral",
TString  fitType = "FitGA" 
)
inline

Definition at line 110 of file Factory.h.

◆ OptimizeAllMethodsForRegression()

void TMVA::Factory::OptimizeAllMethodsForRegression ( TString  fomType = "ROCIntegral",
TString  fitType = "FitGA" 
)
inline

Definition at line 111 of file Factory.h.

◆ PrintHelpMessage()

void TMVA::Factory::PrintHelpMessage ( const TString datasetname,
const TString methodTitle = "" 
) const

Print predefined help message of classifier.

Iterate over methods and test.

Definition at line 1291 of file Factory.cxx.

◆ RootBaseDir()

TDirectory* TMVA::Factory::RootBaseDir ( )
inline

Definition at line 149 of file Factory.h.

◆ SetInputTreesFromEventAssignTrees()

void TMVA::Factory::SetInputTreesFromEventAssignTrees ( )
private

◆ SetVerbose()

void TMVA::Factory::SetVerbose ( Bool_t  v = kTRUE)

Definition at line 350 of file Factory.cxx.

◆ TestAllMethods()

void TMVA::Factory::TestAllMethods ( )

Evaluates all booked methods on the testing data and adds the output to the Results in the corresponiding DataSet.

Definition at line 1231 of file Factory.cxx.

◆ TrainAllMethods()

void TMVA::Factory::TrainAllMethods ( )

Iterates through all booked methods and calls training.

Definition at line 1093 of file Factory.cxx.

◆ TrainAllMethodsForClassification()

void TMVA::Factory::TrainAllMethodsForClassification ( void  )
inline

Definition at line 115 of file Factory.h.

◆ TrainAllMethodsForRegression()

void TMVA::Factory::TrainAllMethodsForRegression ( void  )
inline

Definition at line 116 of file Factory.h.

◆ Verbose()

Bool_t TMVA::Factory::Verbose ( void  ) const
inline

Definition at line 134 of file Factory.h.

◆ WriteDataInformation()

void TMVA::Factory::WriteDataInformation ( DataSetInfo fDataSetInfo)
private

Definition at line 596 of file Factory.cxx.

Friends And Related Function Documentation

◆ CrossValidation

friend class CrossValidation
friend

Definition at line 82 of file Factory.h.

Member Data Documentation

◆ fAnalysisType

Types::EAnalysisType TMVA::Factory::fAnalysisType
private

jobname, used as extension in weight file names

Definition at line 217 of file Factory.h.

◆ fCorrelations

Bool_t TMVA::Factory::fCorrelations
private

verbosity level, controls granularity of logging

Definition at line 211 of file Factory.h.

◆ fDefaultTrfs

std::vector<TMVA::VariableTransformBase*> TMVA::Factory::fDefaultTrfs
private

ROOT output file.

Definition at line 204 of file Factory.h.

◆ fgTargetFile

TFile* TMVA::Factory::fgTargetFile
private

Definition at line 201 of file Factory.h.

◆ fJobName

TString TMVA::Factory::fJobName
private

used in contructor wihtout file

Definition at line 215 of file Factory.h.

◆ fMethodsMap

std::map<TString,MVector*> TMVA::Factory::fMethodsMap

Definition at line 86 of file Factory.h.

◆ fModelPersistence

Bool_t TMVA::Factory::fModelPersistence
private

the training type

Definition at line 218 of file Factory.h.

◆ fOptions

TString TMVA::Factory::fOptions
private

list of transformations on default DataSet

Definition at line 207 of file Factory.h.

◆ fROC

Bool_t TMVA::Factory::fROC
private

enable to calculate corelations

Definition at line 212 of file Factory.h.

◆ fSilentFile

Bool_t TMVA::Factory::fSilentFile
private

enable to calculate ROC values

Definition at line 213 of file Factory.h.

◆ fTransformations

TString TMVA::Factory::fTransformations
private

option string given by construction (presently only "V")

Definition at line 208 of file Factory.h.

◆ fVerbose

Bool_t TMVA::Factory::fVerbose
private

list of transformations to test

Definition at line 209 of file Factory.h.

◆ fVerboseLevel

TString TMVA::Factory::fVerboseLevel
private

verbose mode

Definition at line 210 of file Factory.h.

Libraries for TMVA::Factory:
[legend]

The documentation for this class was generated from the following files: