Logo ROOT   6.10/09
Reference Guide
DataSetInfo.h
Go to the documentation of this file.
1 // // @(#)root/tmva $Id$
2 // Author: Andreas Hoecker, Peter Speckmayer, Joerg Stelzer, Helge Voss
3 
4 /**********************************************************************************
5  * Project: TMVA - a Root-integrated toolkit for multivariate data analysis *
6  * Package: TMVA *
7  * Class : DataSetInfo *
8  * Web : http://tmva.sourceforge.net *
9  * *
10  * Description: *
11  * Contains all the data information *
12  * *
13  * Authors (alphabetical): *
14  * Peter Speckmayer <speckmay@mail.cern.ch> - CERN, Switzerland *
15  * Joerg Stelzer <Joerg.Stelzer@cern.ch> - DESY, Germany *
16  * *
17  * Copyright (c) 2008-2011: *
18  * CERN, Switzerland *
19  * MPI-K Heidelberg, Germany *
20  * DESY Hamburg, Germany *
21  * *
22  * Redistribution and use in source and binary forms, with or without *
23  * modification, are permitted according to the terms listed in LICENSE *
24  * (http://tmva.sourceforge.net/LICENSE) *
25  **********************************************************************************/
26 
27 #ifndef ROOT_TMVA_DataSetInfo
28 #define ROOT_TMVA_DataSetInfo
29 
30 //////////////////////////////////////////////////////////////////////////
31 // //
32 // DataSetInfo //
33 // //
34 // Class that contains all the data information //
35 // //
36 //////////////////////////////////////////////////////////////////////////
37 
38 #include <iosfwd>
39 
40 #include "TObject.h"
41 #include "TString.h"
42 #include "TTree.h"
43 #include "TCut.h"
44 #include "TMatrixDfwd.h"
45 
46 #include "TMVA/Types.h"
47 #include "TMVA/VariableInfo.h"
48 #include "TMVA/ClassInfo.h"
49 #include "TMVA/Event.h"
50 
51 class TH2;
52 
53 namespace TMVA {
54 
55  class DataSet;
56  class VariableTransformBase;
57  class MsgLogger;
58  class DataSetManager;
59 
60  class DataSetInfo : public TObject {
61 
62  public:
63 
64  DataSetInfo(const TString& name = "Default");
65  virtual ~DataSetInfo();
66 
67  virtual const char* GetName() const { return fName.Data(); }
68 
69  // the data set
70  void ClearDataSet() const;
71  DataSet* GetDataSet() const;
72 
73  // ---
74  // the variable data
75  // ---
76  VariableInfo& AddVariable( const TString& expression, const TString& title = "", const TString& unit = "",
77  Double_t min = 0, Double_t max = 0, char varType='F',
78  Bool_t normalized = kTRUE, void* external = 0 );
79  VariableInfo& AddVariable( const VariableInfo& varInfo );
80 
81  VariableInfo& AddTarget ( const TString& expression, const TString& title, const TString& unit,
82  Double_t min, Double_t max, Bool_t normalized = kTRUE, void* external = 0 );
83  VariableInfo& AddTarget ( const VariableInfo& varInfo );
84 
85  VariableInfo& AddSpectator ( const TString& expression, const TString& title, const TString& unit,
86  Double_t min, Double_t max, char type = 'F', Bool_t normalized = kTRUE, void* external = 0 );
87  VariableInfo& AddSpectator ( const VariableInfo& varInfo );
88 
89  ClassInfo* AddClass ( const TString& className );
90 
91  // accessors
92 
93  // general
94  std::vector<VariableInfo>& GetVariableInfos() { return fVariables; }
95  const std::vector<VariableInfo>& GetVariableInfos() const { return fVariables; }
96  VariableInfo& GetVariableInfo( Int_t i ) { return fVariables.at(i); }
97  const VariableInfo& GetVariableInfo( Int_t i ) const { return fVariables.at(i); }
98 
99  std::vector<VariableInfo>& GetTargetInfos() { return fTargets; }
100  const std::vector<VariableInfo>& GetTargetInfos() const { return fTargets; }
101  VariableInfo& GetTargetInfo( Int_t i ) { return fTargets.at(i); }
102  const VariableInfo& GetTargetInfo( Int_t i ) const { return fTargets.at(i); }
103 
104  std::vector<VariableInfo>& GetSpectatorInfos() { return fSpectators; }
105  const std::vector<VariableInfo>& GetSpectatorInfos() const { return fSpectators; }
107  const VariableInfo& GetSpectatorInfo( Int_t i ) const { return fSpectators.at(i); }
108 
109 
110  UInt_t GetNVariables() const { return fVariables.size(); }
111  UInt_t GetNTargets() const { return fTargets.size(); }
112  UInt_t GetNSpectators(bool all=kTRUE) const;
113 
114  const TString& GetNormalization() const { return fNormalization; }
116 
117  void SetTrainingSumSignalWeights(Double_t trainingSumSignalWeights){fTrainingSumSignalWeights = trainingSumSignalWeights;}
118  void SetTrainingSumBackgrWeights(Double_t trainingSumBackgrWeights){fTrainingSumBackgrWeights = trainingSumBackgrWeights;}
119  void SetTestingSumSignalWeights (Double_t testingSumSignalWeights ){fTestingSumSignalWeights = testingSumSignalWeights ;}
120  void SetTestingSumBackgrWeights (Double_t testingSumBackgrWeights ){fTestingSumBackgrWeights = testingSumBackgrWeights ;}
121 
126 
127 
128 
129  // classification information
133  ClassInfo* GetClassInfo( Int_t clNum ) const;
134  ClassInfo* GetClassInfo( const TString& name ) const;
135  void PrintClasses() const;
136  UInt_t GetNClasses() const { return fClasses.size(); }
137  Bool_t IsSignal( const Event* ev ) const;
138  std::vector<Float_t>* GetTargetsForMulticlass( const Event* ev );
140 
141  // by variable
142  Int_t FindVarIndex( const TString& ) const;
143 
144  // weights
145  const TString GetWeightExpression(Int_t i) const { return GetClassInfo(i)->GetWeight(); }
146  void SetWeightExpression( const TString& exp, const TString& className = "" );
147 
148  // cuts
149  const TCut& GetCut (Int_t i) const { return GetClassInfo(i)->GetCut(); }
150  const TCut& GetCut ( const TString& className ) const { return GetClassInfo(className)->GetCut(); }
151  void SetCut ( const TCut& cut, const TString& className );
152  void AddCut ( const TCut& cut, const TString& className );
153  Bool_t HasCuts() const;
154 
155  std::vector<TString> GetListOfVariables() const;
156 
157  // correlation matrix
158  const TMatrixD* CorrelationMatrix ( const TString& className ) const;
159  void SetCorrelationMatrix ( const TString& className, TMatrixD* matrix );
160  void PrintCorrelationMatrix( const TString& className );
162  const TString& hName,
163  const TString& hTitle ) const;
164 
165  // options
167  const TString& GetSplitOptions() const { return fSplitOptions; }
168 
169  // root dir
171  TDirectory* GetRootDir() const { return fOwnRootDir; }
172 
173  void SetMsgType( EMsgType t ) const;
174 
176  private:
177 
179  void SetDataSetManager( DataSetManager* dsm ) { fDataSetManager = dsm; } // DSMTEST
180  friend class DataSetManager; // DSMTEST (datasetmanager test)
181 
182  DataSetInfo( const DataSetInfo& ) : TObject() {}
183 
184  void PrintCorrelationMatrix( TTree* theTree );
185 
186  TString fName; // name of the dataset info object
187 
188  mutable DataSet* fDataSet; // dataset, owned by this datasetinfo object
189  mutable Bool_t fNeedsRebuilding; // flag if rebuilding of dataset is needed (after change of cuts, vars, etc.)
190 
191  // expressions/formulas
192  std::vector<VariableInfo> fVariables; // list of variable expressions/internal names
193  std::vector<VariableInfo> fTargets; // list of targets expressions/internal names
194  std::vector<VariableInfo> fSpectators; // list of spectators expressions/internal names
195 
196  // the classes
197  mutable std::vector<ClassInfo*> fClasses; // name and other infos of the classes
198 
201 
206 
207 
208 
209  TDirectory* fOwnRootDir; // ROOT output dir
210  Bool_t fVerbose; // Verbosity
211 
212  UInt_t fSignalClass; // index of the class with the name signal
213 
214  std::vector<Float_t>* fTargetsForMulticlass;//-> all targets 0 except the one with index==classNumber
215 
216  mutable MsgLogger* fLogger; //! message logger
217  MsgLogger& Log() const { return *fLogger; }
218 
219  public:
220 
222  };
223 }
224 
225 #endif
const VariableInfo & GetTargetInfo(Int_t i) const
Definition: DataSetInfo.h:102
UInt_t GetNVariables() const
Definition: DataSetInfo.h:110
void SetTrainingSumBackgrWeights(Double_t trainingSumBackgrWeights)
Definition: DataSetInfo.h:118
VariableInfo & AddTarget(const TString &expression, const TString &title, const TString &unit, Double_t min, Double_t max, Bool_t normalized=kTRUE, void *external=0)
add a variable (can be a complex expression) to the set of variables used in the MV analysis ...
std::vector< VariableInfo > & GetSpectatorInfos()
Definition: DataSetInfo.h:104
const std::vector< VariableInfo > & GetVariableInfos() const
Definition: DataSetInfo.h:95
void SetCut(const TCut &cut, const TString &className)
set the cut for the classes
std::vector< TString > GetListOfVariables() const
returns list of variables
std::vector< VariableInfo > fTargets
Definition: DataSetInfo.h:193
const std::vector< VariableInfo > & GetTargetInfos() const
Definition: DataSetInfo.h:100
void SetTrainingSumSignalWeights(Double_t trainingSumSignalWeights)
Definition: DataSetInfo.h:117
Int_t GetTargetNameMaxLength() const
Basic string class.
Definition: TString.h:129
std::vector< ClassInfo * > fClasses
Definition: DataSetInfo.h:197
void SetTestingSumBackgrWeights(Double_t testingSumBackgrWeights)
Definition: DataSetInfo.h:120
int Int_t
Definition: RtypesCore.h:41
bool Bool_t
Definition: RtypesCore.h:59
void ClearDataSet() const
UInt_t GetNClasses() const
Definition: DataSetInfo.h:136
std::vector< VariableInfo > fSpectators
Definition: DataSetInfo.h:194
void AddCut(const TCut &cut, const TString &className)
set the cut for the classes
std::vector< VariableInfo > fVariables
Definition: DataSetInfo.h:192
void SetDataSetManager(DataSetManager *dsm)
Definition: DataSetInfo.h:179
TString fSplitOptions
Definition: DataSetInfo.h:200
Double_t GetTrainingSumSignalWeights()
VariableInfo & AddSpectator(const TString &expression, const TString &title, const TString &unit, Double_t min, Double_t max, char type='F', Bool_t normalized=kTRUE, void *external=0)
add a spectator (can be a complex expression) to the set of spectator variables used in the MV analys...
const std::vector< VariableInfo > & GetSpectatorInfos() const
Definition: DataSetInfo.h:105
Class that contains all the information of a class.
Definition: ClassInfo.h:49
const TString & GetNormalization() const
Definition: DataSetInfo.h:114
const VariableInfo & GetVariableInfo(Int_t i) const
Definition: DataSetInfo.h:97
#define ClassDef(name, id)
Definition: Rtypes.h:297
DataSet * fDataSet
Definition: DataSetInfo.h:188
TH2 * CreateCorrelationMatrixHist(const TMatrixD *m, const TString &hName, const TString &hTitle) const
virtual ~DataSetInfo()
destructor
Definition: DataSetInfo.cxx:88
Double_t fTrainingSumSignalWeights
Definition: DataSetInfo.h:202
void SetTestingSumSignalWeights(Double_t testingSumSignalWeights)
Definition: DataSetInfo.h:119
MsgLogger & Log() const
message logger
Definition: DataSetInfo.h:217
void PrintCorrelationMatrix(const TString &className)
calculates the correlation matrices for signal and background, prints them to standard output...
Class that contains all the data information.
Definition: DataSetInfo.h:60
Double_t fTestingSumSignalWeights
Definition: DataSetInfo.h:204
DataSetInfo(const DataSetInfo &)
Definition: DataSetInfo.h:182
TString fNormalization
Definition: DataSetInfo.h:199
std::vector< VariableInfo > & GetTargetInfos()
Definition: DataSetInfo.h:99
TDirectory * GetRootDir() const
Definition: DataSetInfo.h:171
DataSetInfo(const TString &name="Default")
constructor
Definition: DataSetInfo.cxx:61
A specialized string object used for TTree selections.
Definition: TCut.h:25
void SetCorrelationMatrix(const TString &className, TMatrixD *matrix)
Class that contains all the data information.
Definition: DataSet.h:69
void SetSplitOptions(const TString &so)
Definition: DataSetInfo.h:166
MsgLogger * fLogger
Definition: DataSetInfo.h:216
TDirectory * fOwnRootDir
Definition: DataSetInfo.h:209
Double_t fTrainingSumBackgrWeights
Definition: DataSetInfo.h:203
UInt_t GetNTargets() const
Definition: DataSetInfo.h:111
Double_t GetTestingSumSignalWeights()
const VariableInfo & GetSpectatorInfo(Int_t i) const
Definition: DataSetInfo.h:107
DataSetManager * GetDataSetManager()
Definition: DataSetInfo.h:175
Service class for 2-Dim histogram classes.
Definition: TH2.h:30
ClassInfo * GetClassInfo(Int_t clNum) const
const TMatrixD * CorrelationMatrix(const TString &className) const
VariableInfo & GetTargetInfo(Int_t i)
Definition: DataSetInfo.h:101
void SetWeightExpression(const TString &exp, const TString &className="")
set the weight expressions for the classes if class name is specified, set only for this class if cla...
unsigned int UInt_t
Definition: RtypesCore.h:42
TMarker * m
Definition: textangle.C:8
UInt_t GetNSpectators(bool all=kTRUE) const
void SetMsgType(EMsgType t) const
UInt_t GetSignalClassIndex()
Definition: DataSetInfo.h:139
void PrintClasses() const
Int_t FindVarIndex(const TString &) const
find variable by name
double Double_t
Definition: RtypesCore.h:55
Int_t GetVariableNameMaxLength() const
Class that contains all the data information.
Describe directory structure in memory.
Definition: TDirectory.h:34
Double_t GetTrainingSumBackgrWeights()
int type
Definition: TGX11.cxx:120
VariableInfo & GetSpectatorInfo(Int_t i)
Definition: DataSetInfo.h:106
VariableInfo & GetVariableInfo(Int_t i)
Definition: DataSetInfo.h:96
ClassInfo * AddClass(const TString &className)
Int_t GetClassNameMaxLength() const
virtual const char * GetName() const
Returns name of object.
Definition: DataSetInfo.h:67
ostringstream derivative to redirect and format output
Definition: MsgLogger.h:59
Mother of all ROOT objects.
Definition: TObject.h:37
Abstract ClassifierFactory template that handles arbitrary types.
std::vector< Float_t > * fTargetsForMulticlass
Definition: DataSetInfo.h:214
const TCut & GetCut() const
Definition: ClassInfo.h:64
TMVA::DataSetManager * fDataSetManager
Definition: DataSetInfo.h:178
const TString & GetSplitOptions() const
Definition: DataSetInfo.h:167
Bool_t HasCuts() const
const TCut & GetCut(Int_t i) const
Definition: DataSetInfo.h:149
VariableInfo & AddVariable(const TString &expression, const TString &title="", const TString &unit="", Double_t min=0, Double_t max=0, char varType='F', Bool_t normalized=kTRUE, void *external=0)
add a variable (can be a complex expression) to the set of variables used in the MV analysis ...
Bool_t IsSignal(const Event *ev) const
A TTree object has a header with a name and a title.
Definition: TTree.h:78
const TString & GetWeight() const
Definition: ClassInfo.h:63
Bool_t fNeedsRebuilding
Definition: DataSetInfo.h:189
Class for type info of MVA input variable.
Definition: VariableInfo.h:47
Double_t fTestingSumBackgrWeights
Definition: DataSetInfo.h:205
double exp(double)
const TString GetWeightExpression(Int_t i) const
Definition: DataSetInfo.h:145
const Bool_t kTRUE
Definition: RtypesCore.h:91
double norm(double *x, double *p)
Definition: unuranDistr.cxx:40
DataSet * GetDataSet() const
returns data set
void SetNormalization(const TString &norm)
Definition: DataSetInfo.h:115
void SetRootDir(TDirectory *d)
Definition: DataSetInfo.h:170
std::vector< VariableInfo > & GetVariableInfos()
Definition: DataSetInfo.h:94
Double_t GetTestingSumBackgrWeights()
const TCut & GetCut(const TString &className) const
Definition: DataSetInfo.h:150
std::vector< Float_t > * GetTargetsForMulticlass(const Event *ev)
const char * Data() const
Definition: TString.h:347