ENH: change the interface of otb::MachineLearningModel to support regression

Modules/Core/Common/include/otbMachineLearningModel.h

@@ -94,10 +94,10 @@ public:
   //@}
 
   /** Train the machine learning model */
-  virtual void Train() = 0;
+  void Train();
 
   /** Predict values using the model */
-  virtual TargetSampleType Predict(const InputSampleType& input) const = 0;
+  TargetSampleType Predict(const InputSampleType& input) const;
 
   /** Classify all samples in InputListSample and fill TargetListSample with the associated label */
   void PredictAll();
@@ -134,6 +134,12 @@ public:
   itkGetObjectMacro(TargetListSample,TargetListSampleType);
   //@}
 
+  /**\name Classification vs regression mode accessors */
+  //@{
+  itkSetMacro(RegressionMode,bool);
+  itkGetMacro(RegressionMode,bool);
+  //@}
+
 protected:
   /** Constructor */
   MachineLearningModel();
@@ -150,6 +156,14 @@ protected:
   /** Target list sample */
   typename TargetListSampleType::Pointer m_TargetListSample;
 
+  /** Train the machine learning model */
+  virtual void TrainRegression() = 0;
+  virtual void TrainClassification() = 0;
+  /** Predict values using the model */
+  virtual TargetSampleType PredictRegression(const InputSampleType& input) const = 0;
+  virtual TargetSampleType PredictClassification(const InputSampleType& input) const = 0;
+
+  bool m_RegressionMode;
 private:
   MachineLearningModel(const Self &); //purposely not implemented
   void operator =(const Self&); //purposely not implemented

Modules/Core/Common/include/otbMachineLearningModel.txx

@@ -25,7 +25,7 @@ namespace otb
 
 template <class TInputValue, class TOutputValue>
 MachineLearningModel<TInputValue,TOutputValue>
-::MachineLearningModel()
+::MachineLearningModel() : m_RegressionMode(false)
 {}
 
 
@@ -34,6 +34,28 @@ MachineLearningModel<TInputValue,TOutputValue>
 ::~MachineLearningModel()
 {}
 
+template <class TInputValue, class TOutputValue>
+void
+MachineLearningModel<TInputValue,TOutputValue>
+::Train()
+{
+  if(m_RegressionMode)
+    return this->TrainRegression();
+  else
+    return this->TrainClassification();
+}
+template <class TInputValue, class TOutputValue>
+typename MachineLearningModel<TInputValue,TOutputValue>::TargetSampleType
+MachineLearningModel<TInputValue,TOutputValue>
+::Predict(const typename MachineLearningModel<TInputValue,TOutputValue>::InputSampleType& input) const
+{
+  if(m_RegressionMode)
+    return this->PredictRegression(input);
+  else
+    return this->PredictClassification(input);
+}
+
+
 template <class TInputValue, class TOutputValue>
 void
 MachineLearningModel<TInputValue,TOutputValue>