ENH: factorise duplicate code for classification and regression

Modules/Learning/Supervised/include/otbNeuralNetworkMachineLearningModel.h

@@ -193,6 +193,10 @@ private:
   NeuralNetworkMachineLearningModel(const Self &); //purposely not implemented
   void operator =(const Self&); //purposely not implemented
 
+  void CreateNetwork();
+  CvANN_MLP_TrainParams SetNetworkParameters();
+  void SetupNetworkAndTrain(cv::Mat& labels);
+
   CvANN_MLP * m_ANNModel;
   int m_TrainMethod;
   int m_ActivateFunction;

Modules/Learning/Supervised/include/otbNeuralNetworkMachineLearningModel.txx

@@ -130,9 +130,8 @@ void NeuralNetworkMachineLearningModel<TInputValue, TOutputValue>::LabelsToMat(c
     }
 }
 
-/** Train the machine learning model for classification*/
 template<class TInputValue, class TOutputValue>
-void NeuralNetworkMachineLearningModel<TInputValue, TOutputValue>::TrainClassification()
+void NeuralNetworkMachineLearningModel<TInputValue, TOutputValue>::CreateNetwork()
 {
   //Create the neural network
   const unsigned int nbLayers = m_LayerSizes.size();
@@ -147,14 +146,11 @@ void NeuralNetworkMachineLearningModel<TInputValue, TOutputValue>::TrainClassifi
     }
 
   m_ANNModel->create(layers, m_ActivateFunction, m_Alpha, m_Beta);
+}
 
-  //convert listsample to opencv matrix
-  cv::Mat samples;
-  otb::ListSampleToMat<InputListSampleType>(this->GetInputListSample(), samples);
-
-  cv::Mat matOutputANN;
-  LabelsToMat(this->GetTargetListSample(), matOutputANN);
-
+template<class TInputValue, class TOutputValue>
+CvANN_MLP_TrainParams NeuralNetworkMachineLearningModel<TInputValue, TOutputValue>::SetNetworkParameters()
+{
   CvANN_MLP_TrainParams params;
   params.train_method = m_TrainMethod;
   params.bp_dw_scale = m_BackPropDWScale;
@@ -163,9 +159,29 @@ void NeuralNetworkMachineLearningModel<TInputValue, TOutputValue>::TrainClassifi
   params.rp_dw_min = m_RegPropDWMin;
   CvTermCriteria term_crit = cvTermCriteria(m_TermCriteriaType, m_MaxIter, m_Epsilon);
   params.term_crit = term_crit;
+  return params;
+}
 
+template<class TInputValue, class TOutputValue>
+void NeuralNetworkMachineLearningModel<TInputValue, TOutputValue>::SetupNetworkAndTrain(cv::Mat& labels)
+{
+  //convert listsample to opencv matrix
+  cv::Mat samples;
+  otb::ListSampleToMat<InputListSampleType>(this->GetInputListSample(), samples);
+  this->CreateNetwork();
+  CvANN_MLP_TrainParams params = this->SetNetworkParameters();
   //train the Neural network model
-  m_ANNModel->train(samples, matOutputANN, cv::Mat(), cv::Mat(), params);
+  m_ANNModel->train(samples, labels, cv::Mat(), cv::Mat(), params);
+}
+
+/** Train the machine learning model for classification*/
+template<class TInputValue, class TOutputValue>
+void NeuralNetworkMachineLearningModel<TInputValue, TOutputValue>::TrainClassification()
+{
+  //Transform the targets into a matrix of labels
+  cv::Mat matOutputANN;
+  LabelsToMat(this->GetTargetListSample(), matOutputANN);
+  this->SetupNetworkAndTrain(matOutputANN);
 }
 
 template<class TInputValue, class TOutputValue>
@@ -217,38 +233,10 @@ typename NeuralNetworkMachineLearningModel<TInputValue, TOutputValue>::TargetSam
 template<class TInputValue, class TOutputValue>
 void NeuralNetworkMachineLearningModel<TInputValue, TOutputValue>::TrainRegression()
 {
-  //Create the neural network
-  const unsigned int nbLayers = m_LayerSizes.size();
-
-  if ( nbLayers == 0 )
-    itkExceptionMacro(<< "Number of layers in the Neural Network must be >= 3")
-
-      cv::Mat layers = cv::Mat(nbLayers, 1, CV_32SC1);
-  for (unsigned int i = 0; i < nbLayers; i++)
-    {
-    layers.row(i) = m_LayerSizes[i];
-    }
-
-  m_ANNModel->create(layers, m_ActivateFunction, m_Alpha, m_Beta);
-
-  //convert listsample to opencv matrix
-  cv::Mat samples;
-  otb::ListSampleToMat<InputListSampleType>(this->GetInputListSample(), samples);
-
+  //Transform the targets into an OpenCV matrix
   cv::Mat matOutputANN;
   otb::ListSampleToMat<TargetListSampleType>(this->GetTargetListSample(), matOutputANN);
-
-  CvANN_MLP_TrainParams params;
-  params.train_method = m_TrainMethod;
-  params.bp_dw_scale = m_BackPropDWScale;
-  params.bp_moment_scale = m_BackPropMomentScale;
-  params.rp_dw0 = m_RegPropDW0;
-  params.rp_dw_min = m_RegPropDWMin;
-  CvTermCriteria term_crit = cvTermCriteria(m_TermCriteriaType, m_MaxIter, m_Epsilon);
-  params.term_crit = term_crit;
-
-  //train the Neural network model
-  m_ANNModel->train(samples, matOutputANN, cv::Mat(), cv::Mat(), params);
+  this->SetupNetworkAndTrain(matOutputANN);
 }
 
 template<class TInputValue, class TOutputValue>