Logo ROOT   6.08/07
Reference Guide
TMVAClassification_PDEFoam.class.C
Go to the documentation of this file.
1 // Class: ReadPDEFoam
2 // Automatically generated by MethodBase::MakeClass
3 //
4 
5 /* configuration options =====================================================
6 
7 #GEN -*-*-*-*-*-*-*-*-*-*-*- general info -*-*-*-*-*-*-*-*-*-*-*-
8 
9 Method : PDEFoam::PDEFoam
10 TMVA Release : 4.2.1 [262657]
11 ROOT Release : 6.08/07 [395271]
12 Creator : sftnight
13 Date : Thu May 31 21:35:42 2018
14 Host : Linux SFT-ubuntu-1710-1 4.13.0-31-generic #34-Ubuntu SMP Fri Jan 19 16:34:46 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux
15 Dir : /mnt/build/workspace/root-makedoc-v608/rootspi/rdoc/src/v6-08-00-patches/documentation/doxygen
16 Training events: 2000
17 Analysis type : [Classification]
18 
19 
20 #OPT -*-*-*-*-*-*-*-*-*-*-*-*- options -*-*-*-*-*-*-*-*-*-*-*-*-
21 
22 # Set by User:
23 V: "False" [Verbose output (short form of "VerbosityLevel" below - overrides the latter one)]
24 H: "False" [Print method-specific help message]
25 SigBgSeparate: "False" [Separate foams for signal and background]
26 TailCut: "1.000000e-03" [Fraction of outlier events that are excluded from the foam in each dimension]
27 VolFrac: "6.660000e-02" [Size of sampling box, used for density calculation during foam build-up (maximum value: 1.0 is equivalent to volume of entire foam)]
28 nActiveCells: "500" [Maximum number of active cells to be created by the foam]
29 nSampl: "2000" [Number of generated MC events per cell]
30 nBin: "5" [Number of bins in edge histograms]
31 Compress: "True" [Compress foam output file]
32 Nmin: "100" [Number of events in cell required to split cell]
33 Kernel: "None" [Kernel type used]
34 # Default:
35 VerbosityLevel: "Default" [Verbosity level]
36 VarTransform: "None" [List of variable transformations performed before training, e.g., "D_Background,P_Signal,G,N_AllClasses" for: "Decorrelation, PCA-transformation, Gaussianisation, Normalisation, each for the given class of events ('AllClasses' denotes all events of all classes, if no class indication is given, 'All' is assumed)"]
37 CreateMVAPdfs: "False" [Create PDFs for classifier outputs (signal and background)]
38 IgnoreNegWeightsInTraining: "False" [Events with negative weights are ignored in the training (but are included for testing and performance evaluation)]
39 MultiTargetRegression: "False" [Do regression with multiple targets]
40 MaxDepth: "0" [Maximum depth of cell tree (0=unlimited)]
41 FillFoamWithOrigWeights: "False" [Fill foam with original or boost weights]
42 UseYesNoCell: "False" [Return -1 or 1 for bkg or signal like events]
43 DTLogic: "None" [Use decision tree algorithm to split cells]
44 TargetSelection: "Mean" [Target selection method]
45 ##
46 
47 
48 #VAR -*-*-*-*-*-*-*-*-*-*-*-* variables *-*-*-*-*-*-*-*-*-*-*-*-
49 
50 NVar 4
51 var1+var2 myvar1 myvar1 myvar1 'F' [-8.14423561096,7.26972866058]
52 var1-var2 myvar2 myvar2 Expression 2 'F' [-3.96643972397,4.0258936882]
53 var3 var3 var3 Variable 3 units 'F' [-5.03730010986,4.27845287323]
54 var4 var4 var4 Variable 4 units 'F' [-5.95050764084,4.64035463333]
55 NSpec 2
56 var1*2 spec1 spec1 Spectator 1 units 'F' [-9.91655540466,8.67800140381]
57 var1*3 spec2 spec2 Spectator 2 units 'F' [-14.874833107,13.0170021057]
58 
59 
60 ============================================================================ */
61 
62 #include <vector>
63 #include <cmath>
64 #include <string>
65 #include <iostream>
66 
67 #ifndef IClassifierReader__def
68 #define IClassifierReader__def
69 
70 class IClassifierReader {
71 
72  public:
73 
74  // constructor
75  IClassifierReader() : fStatusIsClean( true ) {}
76  virtual ~IClassifierReader() {}
77 
78  // return classifier response
79  virtual double GetMvaValue( const std::vector<double>& inputValues ) const = 0;
80 
81  // returns classifier status
82  bool IsStatusClean() const { return fStatusIsClean; }
83 
84  protected:
85 
86  bool fStatusIsClean;
87 };
88 
89 #endif
90 
91 class ReadPDEFoam : public IClassifierReader {
92 
93  public:
94 
95  // constructor
96  ReadPDEFoam( std::vector<std::string>& theInputVars )
97  : IClassifierReader(),
98  fClassName( "ReadPDEFoam" ),
99  fNvars( 4 ),
100  fIsNormalised( false )
101  {
102  // the training input variables
103  const char* inputVars[] = { "var1+var2", "var1-var2", "var3", "var4" };
104 
105  // sanity checks
106  if (theInputVars.size() <= 0) {
107  std::cout << "Problem in class \"" << fClassName << "\": empty input vector" << std::endl;
108  fStatusIsClean = false;
109  }
110 
111  if (theInputVars.size() != fNvars) {
112  std::cout << "Problem in class \"" << fClassName << "\": mismatch in number of input values: "
113  << theInputVars.size() << " != " << fNvars << std::endl;
114  fStatusIsClean = false;
115  }
116 
117  // validate input variables
118  for (size_t ivar = 0; ivar < theInputVars.size(); ivar++) {
119  if (theInputVars[ivar] != inputVars[ivar]) {
120  std::cout << "Problem in class \"" << fClassName << "\": mismatch in input variable names" << std::endl
121  << " for variable [" << ivar << "]: " << theInputVars[ivar].c_str() << " != " << inputVars[ivar] << std::endl;
122  fStatusIsClean = false;
123  }
124  }
125 
126  // initialize min and max vectors (for normalisation)
127  fVmin[0] = 0;
128  fVmax[0] = 0;
129  fVmin[1] = 0;
130  fVmax[1] = 0;
131  fVmin[2] = 0;
132  fVmax[2] = 0;
133  fVmin[3] = 0;
134  fVmax[3] = 0;
135 
136  // initialize input variable types
137  fType[0] = 'F';
138  fType[1] = 'F';
139  fType[2] = 'F';
140  fType[3] = 'F';
141 
142  // initialize constants
143  Initialize();
144 
145  }
146 
147  // destructor
148  virtual ~ReadPDEFoam() {
149  Clear(); // method-specific
150  }
151 
152  // the classifier response
153  // "inputValues" is a vector of input values in the same order as the
154  // variables given to the constructor
155  double GetMvaValue( const std::vector<double>& inputValues ) const;
156 
157  private:
158 
159  // method-specific destructor
160  void Clear();
161 
162  // common member variables
163  const char* fClassName;
164 
165  const size_t fNvars;
166  size_t GetNvar() const { return fNvars; }
167  char GetType( int ivar ) const { return fType[ivar]; }
168 
169  // normalisation of input variables
170  const bool fIsNormalised;
171  bool IsNormalised() const { return fIsNormalised; }
172  double fVmin[4];
173  double fVmax[4];
174  double NormVariable( double x, double xmin, double xmax ) const {
175  // normalise to output range: [-1, 1]
176  return 2*(x - xmin)/(xmax - xmin) - 1.0;
177  }
178 
179  // type of input variable: 'F' or 'I'
180  char fType[4];
181 
182  // initialize internal variables
183  void Initialize();
184  double GetMvaValue__( const std::vector<double>& inputValues ) const;
185 
186  // private members (method specific)
187  inline double ReadPDEFoam::GetMvaValue( const std::vector<double>& inputValues ) const
188  {
189  // classifier response value
190  double retval = 0;
191 
192  // classifier response, sanity check first
193  if (!IsStatusClean()) {
194  std::cout << "Problem in class \"" << fClassName << "\": cannot return classifier response"
195  << " because status is dirty" << std::endl;
196  retval = 0;
197  }
198  else {
199  if (IsNormalised()) {
200  // normalise variables
201  std::vector<double> iV;
202  iV.reserve(inputValues.size());
203  int ivar = 0;
204  for (std::vector<double>::const_iterator varIt = inputValues.begin();
205  varIt != inputValues.end(); varIt++, ivar++) {
206  iV.push_back(NormVariable( *varIt, fVmin[ivar], fVmax[ivar] ));
207  }
208  retval = GetMvaValue__( iV );
209  }
210  else {
211  retval = GetMvaValue__( inputValues );
212  }
213  }
214 
215  return retval;
216  }
float xmin
Definition: THbookFile.cxx:93
Type GetType(const std::string &Name)
Definition: Systematics.cxx:34
Double_t x[n]
Definition: legend1.C:17
void Initialize(Bool_t useTMVAStyle=kTRUE)
Definition: tmvaglob.cxx:176
float xmax
Definition: THbookFile.cxx:93
PyObject * fType