Logo ROOT   6.08/07
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 #ifndef ROOT_TObject
41 #include "TObject.h"
42 #endif
43 #ifndef ROOT_TString
44 #include "TString.h"
45 #endif
46 #ifndef ROOT_TTree
47 #include "TTree.h"
48 #endif
49 #ifndef ROOT_TCut
50 #include "TCut.h"
51 #endif
52 #ifndef ROOT_TMatrixDfwd
53 #include "TMatrixDfwd.h"
54 #endif
55 
56 #ifndef ROOT_TMVA_Types
57 #include "TMVA/Types.h"
58 #endif
59 #ifndef ROOT_TMVA_VariableInfo
60 #include "TMVA/VariableInfo.h"
61 #endif
62 #ifndef ROOT_TMVA_ClassInfo
63 #include "TMVA/ClassInfo.h"
64 #endif
65 #ifndef ROOT_TMVA_Event
66 #include "TMVA/Event.h"
67 #endif
68 
69 class TH2;
70 
71 namespace TMVA {
72 
73  class DataSet;
74  class VariableTransformBase;
75  class MsgLogger;
76  class DataSetManager;
77 
78  class DataSetInfo : public TObject {
79 
80  public:
81 
82  DataSetInfo(const TString& name = "Default");
83  virtual ~DataSetInfo();
84 
85  virtual const char* GetName() const { return fName.Data(); }
86 
87  // the data set
88  void ClearDataSet() const;
89  DataSet* GetDataSet() const;
90 
91  // ---
92  // the variable data
93  // ---
94  VariableInfo& AddVariable( const TString& expression, const TString& title = "", const TString& unit = "",
95  Double_t min = 0, Double_t max = 0, char varType='F',
96  Bool_t normalized = kTRUE, void* external = 0 );
97  VariableInfo& AddVariable( const VariableInfo& varInfo );
98 
99  VariableInfo& AddTarget ( const TString& expression, const TString& title, const TString& unit,
100  Double_t min, Double_t max, Bool_t normalized = kTRUE, void* external = 0 );
101  VariableInfo& AddTarget ( const VariableInfo& varInfo );
102 
103  VariableInfo& AddSpectator ( const TString& expression, const TString& title, const TString& unit,
104  Double_t min, Double_t max, char type = 'F', Bool_t normalized = kTRUE, void* external = 0 );
105  VariableInfo& AddSpectator ( const VariableInfo& varInfo );
106 
107  ClassInfo* AddClass ( const TString& className );
108 
109  // accessors
110 
111  // general
112  std::vector<VariableInfo>& GetVariableInfos() { return fVariables; }
113  const std::vector<VariableInfo>& GetVariableInfos() const { return fVariables; }
115  const VariableInfo& GetVariableInfo( Int_t i ) const { return fVariables.at(i); }
116 
117  std::vector<VariableInfo>& GetTargetInfos() { return fTargets; }
118  const std::vector<VariableInfo>& GetTargetInfos() const { return fTargets; }
119  VariableInfo& GetTargetInfo( Int_t i ) { return fTargets.at(i); }
120  const VariableInfo& GetTargetInfo( Int_t i ) const { return fTargets.at(i); }
121 
122  std::vector<VariableInfo>& GetSpectatorInfos() { return fSpectators; }
123  const std::vector<VariableInfo>& GetSpectatorInfos() const { return fSpectators; }
125  const VariableInfo& GetSpectatorInfo( Int_t i ) const { return fSpectators.at(i); }
126 
127 
128  UInt_t GetNVariables() const { return fVariables.size(); }
129  UInt_t GetNTargets() const { return fTargets.size(); }
130  UInt_t GetNSpectators(bool all=kTRUE) const;
131 
132  const TString& GetNormalization() const { return fNormalization; }
134 
135  void SetTrainingSumSignalWeights(Double_t trainingSumSignalWeights){fTrainingSumSignalWeights = trainingSumSignalWeights;}
136  void SetTrainingSumBackgrWeights(Double_t trainingSumBackgrWeights){fTrainingSumBackgrWeights = trainingSumBackgrWeights;}
137  void SetTestingSumSignalWeights (Double_t testingSumSignalWeights ){fTestingSumSignalWeights = testingSumSignalWeights ;}
138  void SetTestingSumBackgrWeights (Double_t testingSumBackgrWeights ){fTestingSumBackgrWeights = testingSumBackgrWeights ;}
139 
144 
145 
146 
147  // classification information
151  ClassInfo* GetClassInfo( Int_t clNum ) const;
152  ClassInfo* GetClassInfo( const TString& name ) const;
153  void PrintClasses() const;
154  UInt_t GetNClasses() const { return fClasses.size(); }
155  Bool_t IsSignal( const Event* ev ) const;
156  std::vector<Float_t>* GetTargetsForMulticlass( const Event* ev );
158 
159  // by variable
160  Int_t FindVarIndex( const TString& ) const;
161 
162  // weights
163  const TString GetWeightExpression(Int_t i) const { return GetClassInfo(i)->GetWeight(); }
164  void SetWeightExpression( const TString& exp, const TString& className = "" );
165 
166  // cuts
167  const TCut& GetCut (Int_t i) const { return GetClassInfo(i)->GetCut(); }
168  const TCut& GetCut ( const TString& className ) const { return GetClassInfo(className)->GetCut(); }
169  void SetCut ( const TCut& cut, const TString& className );
170  void AddCut ( const TCut& cut, const TString& className );
171  Bool_t HasCuts() const;
172 
173  std::vector<TString> GetListOfVariables() const;
174 
175  // correlation matrix
176  const TMatrixD* CorrelationMatrix ( const TString& className ) const;
177  void SetCorrelationMatrix ( const TString& className, TMatrixD* matrix );
178  void PrintCorrelationMatrix( const TString& className );
180  const TString& hName,
181  const TString& hTitle ) const;
182 
183  // options
185  const TString& GetSplitOptions() const { return fSplitOptions; }
186 
187  // root dir
189  TDirectory* GetRootDir() const { return fOwnRootDir; }
190 
191  void SetMsgType( EMsgType t ) const;
192 
194  private:
195 
197  void SetDataSetManager( DataSetManager* dsm ) { fDataSetManager = dsm; } // DSMTEST
198  friend class DataSetManager; // DSMTEST (datasetmanager test)
199 
200  DataSetInfo( const DataSetInfo& ) : TObject() {}
201 
202  void PrintCorrelationMatrix( TTree* theTree );
203 
204  TString fName; // name of the dataset info object
205 
206  mutable DataSet* fDataSet; // dataset, owned by this datasetinfo object
207  mutable Bool_t fNeedsRebuilding; // flag if rebuilding of dataset is needed (after change of cuts, vars, etc.)
208 
209  // expressions/formulas
210  std::vector<VariableInfo> fVariables; // list of variable expressions/internal names
211  std::vector<VariableInfo> fTargets; // list of targets expressions/internal names
212  std::vector<VariableInfo> fSpectators; // list of spectators expressions/internal names
213 
214  // the classes
215  mutable std::vector<ClassInfo*> fClasses; // name and other infos of the classes
216 
219 
224 
225 
226 
227  TDirectory* fOwnRootDir; // ROOT output dir
228  Bool_t fVerbose; // Verbosity
229 
230  UInt_t fSignalClass; // index of the class with the name signal
231 
232  std::vector<Float_t>* fTargetsForMulticlass;//-> all targets 0 except the one with index==classNumber
233 
234  mutable MsgLogger* fLogger; //! message logger
235  MsgLogger& Log() const { return *fLogger; }
236 
237  public:
238 
240  };
241 }
242 
243 #endif
const VariableInfo & GetTargetInfo(Int_t i) const
Definition: DataSetInfo.h:120
UInt_t GetNVariables() const
Definition: DataSetInfo.h:128
void SetTrainingSumBackgrWeights(Double_t trainingSumBackgrWeights)
Definition: DataSetInfo.h:136
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:122
const std::vector< VariableInfo > & GetVariableInfos() const
Definition: DataSetInfo.h:113
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:211
const std::vector< VariableInfo > & GetTargetInfos() const
Definition: DataSetInfo.h:118
void SetTrainingSumSignalWeights(Double_t trainingSumSignalWeights)
Definition: DataSetInfo.h:135
Int_t GetTargetNameMaxLength() const
Basic string class.
Definition: TString.h:137
std::vector< ClassInfo * > fClasses
Definition: DataSetInfo.h:215
void SetTestingSumBackgrWeights(Double_t testingSumBackgrWeights)
Definition: DataSetInfo.h:138
int Int_t
Definition: RtypesCore.h:41
bool Bool_t
Definition: RtypesCore.h:59
void ClearDataSet() const
UInt_t GetNClasses() const
Definition: DataSetInfo.h:154
std::vector< VariableInfo > fSpectators
Definition: DataSetInfo.h:212
void AddCut(const TCut &cut, const TString &className)
set the cut for the classes
std::vector< VariableInfo > fVariables
Definition: DataSetInfo.h:210
void SetDataSetManager(DataSetManager *dsm)
Definition: DataSetInfo.h:197
TString fSplitOptions
Definition: DataSetInfo.h:218
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:123
const TString & GetNormalization() const
Definition: DataSetInfo.h:132
const VariableInfo & GetVariableInfo(Int_t i) const
Definition: DataSetInfo.h:115
#define ClassDef(name, id)
Definition: Rtypes.h:254
DataSet * fDataSet
Definition: DataSetInfo.h:206
TH2 * CreateCorrelationMatrixHist(const TMatrixD *m, const TString &hName, const TString &hTitle) const
virtual ~DataSetInfo()
destructor
Definition: DataSetInfo.cxx:94
Double_t fTrainingSumSignalWeights
Definition: DataSetInfo.h:220
void SetTestingSumSignalWeights(Double_t testingSumSignalWeights)
Definition: DataSetInfo.h:137
MsgLogger & Log() const
message logger
Definition: DataSetInfo.h:235
void PrintCorrelationMatrix(const TString &className)
calculates the correlation matrices for signal and background, prints them to standard output...
Double_t fTestingSumSignalWeights
Definition: DataSetInfo.h:222
DataSetInfo(const DataSetInfo &)
Definition: DataSetInfo.h:200
TString fNormalization
Definition: DataSetInfo.h:217
std::vector< VariableInfo > & GetTargetInfos()
Definition: DataSetInfo.h:117
TDirectory * GetRootDir() const
Definition: DataSetInfo.h:189
DataSetInfo(const TString &name="Default")
constructor
Definition: DataSetInfo.cxx:67
A specialized string object used for TTree selections.
Definition: TCut.h:27
void SetCorrelationMatrix(const TString &className, TMatrixD *matrix)
void SetSplitOptions(const TString &so)
Definition: DataSetInfo.h:184
MsgLogger * fLogger
Definition: DataSetInfo.h:234
TDirectory * fOwnRootDir
Definition: DataSetInfo.h:227
Double_t fTrainingSumBackgrWeights
Definition: DataSetInfo.h:221
UInt_t GetNTargets() const
Definition: DataSetInfo.h:129
Double_t GetTestingSumSignalWeights()
const VariableInfo & GetSpectatorInfo(Int_t i) const
Definition: DataSetInfo.h:125
DataSetManager * GetDataSetManager()
Definition: DataSetInfo.h:193
Service class for 2-Dim histogram classes.
Definition: TH2.h:36
ClassInfo * GetClassInfo(Int_t clNum) const
const TMatrixD * CorrelationMatrix(const TString &className) const
VariableInfo & GetTargetInfo(Int_t i)
Definition: DataSetInfo.h:119
EMsgType
Definition: Types.h:61
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:157
void PrintClasses() const
Int_t FindVarIndex(const TString &) const
find variable by name
double Double_t
Definition: RtypesCore.h:55
Int_t GetVariableNameMaxLength() const
Describe directory structure in memory.
Definition: TDirectory.h:44
Double_t GetTrainingSumBackgrWeights()
int type
Definition: TGX11.cxx:120
VariableInfo & GetSpectatorInfo(Int_t i)
Definition: DataSetInfo.h:124
VariableInfo & GetVariableInfo(Int_t i)
Definition: DataSetInfo.h:114
ClassInfo * AddClass(const TString &className)
Int_t GetClassNameMaxLength() const
virtual const char * GetName() const
Returns name of object.
Definition: DataSetInfo.h:85
Mother of all ROOT objects.
Definition: TObject.h:37
Abstract ClassifierFactory template that handles arbitrary types.
std::vector< Float_t > * fTargetsForMulticlass
Definition: DataSetInfo.h:232
const TCut & GetCut() const
Definition: ClassInfo.h:72
TMVA::DataSetManager * fDataSetManager
Definition: DataSetInfo.h:196
const TString & GetSplitOptions() const
Definition: DataSetInfo.h:185
Bool_t HasCuts() const
const TCut & GetCut(Int_t i) const
Definition: DataSetInfo.h:167
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:98
const TString & GetWeight() const
Definition: ClassInfo.h:71
Bool_t fNeedsRebuilding
Definition: DataSetInfo.h:207
Double_t fTestingSumBackgrWeights
Definition: DataSetInfo.h:223
double exp(double)
const Bool_t kTRUE
Definition: Rtypes.h:91
const TString GetWeightExpression(Int_t i) const
Definition: DataSetInfo.h:163
double norm(double *x, double *p)
Definition: unuranDistr.cxx:40
DataSet * GetDataSet() const
returns data set
void SetNormalization(const TString &norm)
Definition: DataSetInfo.h:133
void SetRootDir(TDirectory *d)
Definition: DataSetInfo.h:188
std::vector< VariableInfo > & GetVariableInfos()
Definition: DataSetInfo.h:112
Double_t GetTestingSumBackgrWeights()
char name[80]
Definition: TGX11.cxx:109
const TCut & GetCut(const TString &className) const
Definition: DataSetInfo.h:168
std::vector< Float_t > * GetTargetsForMulticlass(const Event *ev)
const char * Data() const
Definition: TString.h:349