From b4f2fd41e2021882a1c5870f2561bcb27f4b5c75 Mon Sep 17 00:00:00 2001
From: Arnaud Jaen <arnaud.jaen@c-s.fr>
Date: Wed, 3 Apr 2013 11:13:04 +0200
Subject: [PATCH] ENH: Add Neural Network machine learning model.
---
.../OpenCV/otbMachineLearningModelFactory.txx | 2 +
.../otbNeuralNetworkMachineLearningModel.h | 210 ++++++++++++++++++
.../otbNeuralNetworkMachineLearningModel.txx | 193 ++++++++++++++++
...NeuralNetworkMachineLearningModelFactory.h | 72 ++++++
...uralNetworkMachineLearningModelFactory.txx | 64 ++++++
Testing/Code/Learning/CMakeLists.txt | 19 +-
Testing/Code/Learning/otbLearningTests6.cxx | 3 +
.../otbMachineLearningModelCanRead.cxx | 28 +++
.../Learning/otbTrainMachineLearningModel.cxx | 67 ++++++
9 files changed, 657 insertions(+), 1 deletion(-)
create mode 100644 Code/UtilitiesAdapters/OpenCV/otbNeuralNetworkMachineLearningModel.h
create mode 100644 Code/UtilitiesAdapters/OpenCV/otbNeuralNetworkMachineLearningModel.txx
create mode 100644 Code/UtilitiesAdapters/OpenCV/otbNeuralNetworkMachineLearningModelFactory.h
create mode 100644 Code/UtilitiesAdapters/OpenCV/otbNeuralNetworkMachineLearningModelFactory.txx
diff --git a/Code/UtilitiesAdapters/OpenCV/otbMachineLearningModelFactory.txx b/Code/UtilitiesAdapters/OpenCV/otbMachineLearningModelFactory.txx
index cee47bf37f..1d8bf19db3 100644
--- a/Code/UtilitiesAdapters/OpenCV/otbMachineLearningModelFactory.txx
+++ b/Code/UtilitiesAdapters/OpenCV/otbMachineLearningModelFactory.txx
@@ -24,6 +24,7 @@
#include "otbSVMMachineLearningModelFactory.h"
#include "otbLibSVMMachineLearningModelFactory.h"
#include "otbBoostMachineLearningModelFactory.h"
+#include "otbNeuralNetworkMachineLearningModelFactory.h"
namespace otb
@@ -97,6 +98,7 @@ MachineLearningModelFactory<TInputValue,TOutputValue>
itk::ObjectFactoryBase::RegisterFactory(LibSVMMachineLearningModelFactory<TInputValue,TOutputValue>::New());
itk::ObjectFactoryBase::RegisterFactory(SVMMachineLearningModelFactory<TInputValue,TOutputValue>::New());
itk::ObjectFactoryBase::RegisterFactory(BoostMachineLearningModelFactory<TInputValue,TOutputValue>::New());
+ itk::ObjectFactoryBase::RegisterFactory(NeuralNetworkMachineLearningModelFactory<TInputValue,TOutputValue>::New());
firstTime = false;
}
diff --git a/Code/UtilitiesAdapters/OpenCV/otbNeuralNetworkMachineLearningModel.h b/Code/UtilitiesAdapters/OpenCV/otbNeuralNetworkMachineLearningModel.h
new file mode 100644
index 0000000000..029e7b32cc
--- /dev/null
+++ b/Code/UtilitiesAdapters/OpenCV/otbNeuralNetworkMachineLearningModel.h
@@ -0,0 +1,210 @@
+/*=========================================================================
+
+ Program: ORFEO Toolbox
+ Language: C++
+ Date: $Date$
+ Version: $Revision$
+
+
+ Copyright (c) Centre National d'Etudes Spatiales. All rights reserved.
+ See OTBCopyright.txt for details.
+
+
+ This software is distributed WITHOUT ANY WARRANTY; without even
+ the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR
+ PURPOSE. See the above copyright notices for more information.
+
+=========================================================================*/
+#ifndef __otbNeuralNetworkMachineLearningModel_h
+#define __otbNeuralNetworkMachineLearningModel_h
+
+#include "itkLightObject.h"
+#include "itkVariableLengthVector.h"
+#include "itkFixedArray.h"
+#include "itkListSample.h"
+#include "otbMachineLearningModel.h"
+
+class CvANN_MLP;
+
+namespace otb
+{
+template <class TInputValue, class TTargetValue>
+class ITK_EXPORT NeuralNetworkMachineLearningModel
+ : public MachineLearningModel <TInputValue, TTargetValue>
+{
+public:
+ /** Standard class typedefs. */
+ typedef NeuralNetworkMachineLearningModel Self;
+ typedef MachineLearningModel<TInputValue, TTargetValue> Superclass;
+ typedef itk::SmartPointer<Self> Pointer;
+ typedef itk::SmartPointer<const Self> ConstPointer;
+
+ // Input related typedefs
+ typedef TInputValue InputValueType;
+ typedef itk::VariableLengthVector<InputValueType> InputSampleType;
+ typedef itk::Statistics::ListSample<InputSampleType> InputListSampleType;
+
+ // Target related typedefs
+ typedef TTargetValue TargetValueType;
+ typedef itk::FixedArray<TargetValueType,1> TargetSampleType;
+ typedef itk::Statistics::ListSample<TargetSampleType> TargetListSampleType;
+
+ /** Run-time type information (and related methods). */
+ itkNewMacro(Self);
+ itkTypeMacro(NeuralNetworkMachineLearningModel, itk::MachineLearningModel);
+
+ /** Setters/Getters to the train method
+ * 2 methods are available:
+ * - CvANN_MLP_TrainParams::BACKPROP The back-propagation algorithm.
+ * - CvANN_MLP_TrainParams::RPROP The RPROP algorithm.
+ * Default is CvANN_MLP_TrainParams::RPROP.
+ * \see http://docs.opencv.org/modules/ml/doc/neural_networks.html
+ */
+ itkGetMacro(TrainMethod, int);
+ itkSetMacro(TrainMethod, int);
+
+ /**
+ * Set the number of neurons in each layer (including input and output layers).
+ * The number of neuron in the first layer (input layer) must be equal
+ * to the number of samples in the \c InputListSample
+ */
+ void SetLayerSizes (const std::vector<unsigned int> layers);
+
+
+ /** Setters/Getters to the neuron activation function
+ * 3 methods are available:
+ * - CvANN_MLP::IDENTITY
+ * - CvANN_MLP::SIGMOID_SYM
+ * - CvANN_MLP::GAUSSIAN
+ * Default is CvANN_MLP::SIGMOID_SYM
+ * \see http://docs.opencv.org/modules/ml/doc/neural_networks.html
+ */
+ itkGetMacro(ActivateFunction, int);
+ itkSetMacro(ActivateFunction, int);
+
+ /** Setters/Getters to the alpha parameter of the activation function
+ * Default is 1
+ * \see http://docs.opencv.org/modules/ml/doc/neural_networks.html
+ */
+ itkGetMacro(Alpha, double);
+ itkSetMacro(Alpha, double);
+
+ /** Setters/Getters to the beta parameter of the activation function
+ * Default is 1
+ * \see http://docs.opencv.org/modules/ml/doc/neural_networks.html
+ */
+ itkGetMacro(Beta, double);
+ itkSetMacro(Beta, double);
+
+ /** Strength of the weight gradient term in the BACKPROP method.
+ * The recommended value is about 0.1
+ * Default is 0.1
+ * \see http://docs.opencv.org/modules/ml/doc/neural_networks.html
+ */
+ itkGetMacro(BackPropDWScale, double);
+ itkSetMacro(BackPropDWScale, double);
+
+ /** Strength of the momentum term (the difference between weights on the 2 previous iterations).
+ * This parameter provides some inertia to smooth the random fluctuations of the weights.
+ * It can vary from 0 (the feature is disabled) to 1 and beyond.
+ * The value 0.1 or so is good enough
+ * Default is 0.1
+ * \see http://docs.opencv.org/modules/ml/doc/neural_networks.html
+ */
+ itkGetMacro(BackPropMomentScale, double);
+ itkSetMacro(BackPropMomentScale, double);
+
+ /** Initial value \Delta_0 of update-values \Delta_{ij} in RPROP method.
+ * Default is 0.1
+ * \see http://docs.opencv.org/modules/ml/doc/neural_networks.html
+ */
+ itkGetMacro(RegPropDW0, double);
+ itkSetMacro(RegPropDW0, double);
+
+ /** Update-values lower limit \Delta_{min} in RPROP method.
+ * It must be positive. Default is FLT_EPSILON
+ * \see http://docs.opencv.org/modules/ml/doc/neural_networks.html
+ */
+ itkGetMacro(RegPropDWMin, double);
+ itkSetMacro(RegPropDWMin, double);
+
+ /** Termination criteria.
+ * It can be CV_TERMCRIT_ITER or CV_TERMCRIT_EPS or CV_TERMCRIT_ITER+CV_TERMCRIT_EPS
+ * default is CV_TERMCRIT_ITER+CV_TERMCRIT_EPS.
+ * \see http://docs.opencv.org/modules/ml/doc/neural_networks.html
+ */
+ itkGetMacro(TermCriteriaType, int);
+ itkSetMacro(TermCriteriaType, int);
+
+ /** Maximum number of iteration used in the Termination criteria.
+ * default is 1000
+ * \see http://docs.opencv.org/modules/ml/doc/neural_networks.html
+ */
+ itkGetMacro(MaxIter, int);
+ itkSetMacro(MaxIter, int);
+
+ /** Epsilon value used in the Termination criteria.
+ * default is 0.01
+ * \see http://docs.opencv.org/modules/ml/doc/neural_networks.html
+ */
+ itkGetMacro(Epsilon, double);
+ itkSetMacro(Epsilon, double);
+
+ /** Train the machine learning model */
+ virtual void Train();
+
+ /** Predict values using the model */
+ virtual TargetSampleType Predict(const InputSampleType & input) const;
+
+ /** Save the model to file */
+ virtual void Save(const std::string & filename, const std::string & name="");
+
+ /** Load the model from file */
+ virtual void Load(const std::string & filename, const std::string & name="");
+
+ /** Determine the file type. Returns true if this ImageIO can read the
+ * file specified. */
+ virtual bool CanReadFile(const std::string &);
+
+ /** Determine the file type. Returns true if this ImageIO can write the
+ * file specified. */
+ virtual bool CanWriteFile(const std::string &);
+
+protected:
+ /** Constructor */
+ NeuralNetworkMachineLearningModel();
+
+ /** Destructor */
+ virtual ~NeuralNetworkMachineLearningModel();
+
+ /** PrintSelf method */
+ void PrintSelf(std::ostream& os, itk::Indent indent) const;
+
+private:
+ NeuralNetworkMachineLearningModel(const Self &); //purposely not implemented
+ void operator =(const Self&); //purposely not implemented
+
+ CvANN_MLP * m_ANNModel;
+ int m_TrainMethod;
+ int m_ActivateFunction;
+ std::vector<unsigned int> m_LayerSizes;
+ double m_Alpha;
+ double m_Beta;
+ double m_BackPropDWScale;
+ double m_BackPropMomentScale;
+ double m_RegPropDW0;
+ double m_RegPropDWMin;
+ int m_TermCriteriaType;
+ int m_MaxIter;
+ double m_Epsilon;
+
+
+
+};
+} // end namespace otb
+
+#ifndef OTB_MANUAL_INSTANTIATION
+#include "otbNeuralNetworkMachineLearningModel.txx"
+#endif
+
+#endif
diff --git a/Code/UtilitiesAdapters/OpenCV/otbNeuralNetworkMachineLearningModel.txx b/Code/UtilitiesAdapters/OpenCV/otbNeuralNetworkMachineLearningModel.txx
new file mode 100644
index 0000000000..4015d36f1a
--- /dev/null
+++ b/Code/UtilitiesAdapters/OpenCV/otbNeuralNetworkMachineLearningModel.txx
@@ -0,0 +1,193 @@
+/*=========================================================================
+
+ Program: ORFEO Toolbox
+ Language: C++
+ Date: $Date$
+ Version: $Revision$
+
+
+ Copyright (c) Centre National d'Etudes Spatiales. All rights reserved.
+ See OTBCopyright.txt for details.
+
+
+ This software is distributed WITHOUT ANY WARRANTY; without even
+ the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR
+ PURPOSE. See the above copyright notices for more information.
+
+=========================================================================*/
+#ifndef __otbNeuralNetworkMachineLearningModel_txx
+#define __otbNeuralNetworkMachineLearningModel_txx
+
+
+#include "otbNeuralNetworkMachineLearningModel.h"
+#include "otbOpenCVUtils.h"
+#include "itkMacro.h" // itkExceptionMacro
+#include <opencv2/opencv.hpp>
+
+namespace otb
+{
+
+template <class TInputValue, class TOutputValue>
+NeuralNetworkMachineLearningModel<TInputValue,TOutputValue>
+::NeuralNetworkMachineLearningModel() :
+ m_TrainMethod(CvANN_MLP_TrainParams::RPROP), m_ActivateFunction(CvANN_MLP::SIGMOID_SYM),
+ m_Alpha(0.), m_Beta(0.), m_BackPropDWScale(0.1), m_BackPropMomentScale(0.1),
+ m_RegPropDW0(0.1), m_RegPropDWMin(FLT_EPSILON), m_TermCriteriaType(CV_TERMCRIT_ITER + CV_TERMCRIT_EPS),
+ m_MaxIter(1000), m_Epsilon(0.01)
+{
+ m_ANNModel = new CvANN_MLP;
+ m_LayerSizes.clear();
+}
+
+
+template <class TInputValue, class TOutputValue>
+NeuralNetworkMachineLearningModel<TInputValue,TOutputValue>
+::~NeuralNetworkMachineLearningModel()
+{
+ delete m_ANNModel;
+}
+
+/** Train the machine learning model */
+template <class TInputValue, class TOutputValue>
+void
+NeuralNetworkMachineLearningModel<TInputValue,TOutputValue>
+::SetLayerSizes(const std::vector<unsigned int> layers)
+{
+ const unsigned int nbLayers = layers.size();
+ if (nbLayers < 3)
+ itkExceptionMacro(<< "Number of layers in the Neural Network must be >= 3")
+
+ m_LayerSizes = layers;
+}
+
+/** Train the machine learning model */
+template <class TInputValue, class TOutputValue>
+void
+NeuralNetworkMachineLearningModel<TInputValue,TOutputValue>
+::Train()
+{
+ //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);
+
+ cv::Mat labels;
+ otb::ListSampleToMat<TargetListSampleType>(this->GetTargetListSample(),labels);
+
+ 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,labels,cv::Mat(),cv::Mat(),params);
+}
+
+template <class TInputValue, class TOutputValue>
+typename NeuralNetworkMachineLearningModel<TInputValue,TOutputValue>
+::TargetSampleType
+NeuralNetworkMachineLearningModel<TInputValue,TOutputValue>
+::Predict(const InputSampleType & input) const
+{
+ //convert listsample to Mat
+ cv::Mat sample;
+
+ otb::SampleToMat<InputSampleType>(input,sample);
+
+ cv::Mat response(1, 1, CV_32FC1);
+ m_ANNModel->predict(sample,response);
+
+ TargetSampleType target;
+
+ target[0] = static_cast<TOutputValue>(response.at<float>(0,0));
+
+ return target;
+}
+
+template <class TInputValue, class TOutputValue>
+void
+NeuralNetworkMachineLearningModel<TInputValue,TOutputValue>
+::Save(const std::string & filename, const std::string & name)
+{
+ if (name == "")
+ m_ANNModel->save(filename.c_str(), 0);
+ else
+ m_ANNModel->save(filename.c_str(), name.c_str());
+}
+
+template <class TInputValue, class TOutputValue>
+void
+NeuralNetworkMachineLearningModel<TInputValue,TOutputValue>
+::Load(const std::string & filename, const std::string & name)
+{
+ if (name == "")
+ m_ANNModel->load(filename.c_str(), 0);
+ else
+ m_ANNModel->load(filename.c_str(), name.c_str());
+}
+
+template <class TInputValue, class TOutputValue>
+bool
+NeuralNetworkMachineLearningModel<TInputValue,TOutputValue>
+::CanReadFile(const std::string & file)
+{
+ std::ifstream ifs;
+ ifs.open(file.c_str());
+
+ if(!ifs)
+ {
+ std::cerr<<"Could not read file "<<file<<std::endl;
+ return false;
+ }
+
+ while (!ifs.eof())
+ {
+ std::string line;
+ std::getline(ifs, line);
+
+ if (line.find(CV_TYPE_NAME_ML_ANN_MLP) != std::string::npos)
+ {
+ std::cout<<"Reading a "<<CV_TYPE_NAME_ML_ANN_MLP<<" model !!!"<<std::endl;
+ return true;
+ }
+ }
+ ifs.close();
+ return false;
+}
+
+template <class TInputValue, class TOutputValue>
+bool
+NeuralNetworkMachineLearningModel<TInputValue,TOutputValue>
+::CanWriteFile(const std::string & file)
+{
+ return false;
+}
+
+template <class TInputValue, class TOutputValue>
+void
+NeuralNetworkMachineLearningModel<TInputValue,TOutputValue>
+::PrintSelf(std::ostream& os, itk::Indent indent) const
+{
+ // Call superclass implementation
+ Superclass::PrintSelf(os,indent);
+}
+
+} //end namespace otb
+
+#endif
diff --git a/Code/UtilitiesAdapters/OpenCV/otbNeuralNetworkMachineLearningModelFactory.h b/Code/UtilitiesAdapters/OpenCV/otbNeuralNetworkMachineLearningModelFactory.h
new file mode 100644
index 0000000000..494149756c
--- /dev/null
+++ b/Code/UtilitiesAdapters/OpenCV/otbNeuralNetworkMachineLearningModelFactory.h
@@ -0,0 +1,72 @@
+/*=========================================================================
+
+ Program: ORFEO Toolbox
+ Language: C++
+ Date: $Date$
+ Version: $Revision$
+
+
+ Copyright (c) Centre National d'Etudes Spatiales. All rights reserved.
+ See OTBCopyright.txt for details.
+
+
+ This software is distributed WITHOUT ANY WARRANTY; without even
+ the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR
+ PURPOSE. See the above copyright notices for more information.
+
+=========================================================================*/
+#ifndef __otbNeuralNetworkMachineLearningModelFactory_h
+#define __otbNeuralNetworkMachineLearningModelFactory_h
+
+#include "itkObjectFactoryBase.h"
+#include "itkImageIOBase.h"
+
+namespace otb
+{
+/** \class NeuralNetworkMachineLearningModelFactory
+ * \brief Creation d'un instance d'un objet SVMMachineLearningModel utilisant les object factory.
+ */
+template <class TInputValue, class TTargetValue>
+class ITK_EXPORT NeuralNetworkMachineLearningModelFactory : public itk::ObjectFactoryBase
+{
+public:
+ /** Standard class typedefs. */
+ typedef NeuralNetworkMachineLearningModelFactory Self;
+ typedef itk::ObjectFactoryBase Superclass;
+ typedef itk::SmartPointer<Self> Pointer;
+ typedef itk::SmartPointer<const Self> ConstPointer;
+
+ /** Class methods used to interface with the registered factories. */
+ virtual const char* GetITKSourceVersion(void) const;
+ virtual const char* GetDescription(void) const;
+
+ /** Method for class instantiation. */
+ itkFactorylessNewMacro(Self);
+
+ /** Run-time type information (and related methods). */
+ itkTypeMacro(NeuralNetworkMachineLearningModelFactory, itk::ObjectFactoryBase);
+
+ /** Register one factory of this type */
+ static void RegisterOneFactory(void)
+ {
+ NeuralNetworkMachineLearningModelFactory::Pointer Factory = NeuralNetworkMachineLearningModelFactory::New();
+ itk::ObjectFactoryBase::RegisterFactory(Factory);
+ }
+
+protected:
+ NeuralNetworkMachineLearningModelFactory();
+ virtual ~NeuralNetworkMachineLearningModelFactory();
+
+private:
+ NeuralNetworkMachineLearningModelFactory(const Self &); //purposely not implemented
+ void operator =(const Self&); //purposely not implemented
+
+};
+
+} // end namespace otb
+
+#ifndef OTB_MANUAL_INSTANTIATION
+#include "otbNeuralNetworkMachineLearningModelFactory.txx"
+#endif
+
+#endif
diff --git a/Code/UtilitiesAdapters/OpenCV/otbNeuralNetworkMachineLearningModelFactory.txx b/Code/UtilitiesAdapters/OpenCV/otbNeuralNetworkMachineLearningModelFactory.txx
new file mode 100644
index 0000000000..25a387d97d
--- /dev/null
+++ b/Code/UtilitiesAdapters/OpenCV/otbNeuralNetworkMachineLearningModelFactory.txx
@@ -0,0 +1,64 @@
+/*=========================================================================
+
+ Program: ORFEO Toolbox
+ Language: C++
+ Date: $Date$
+ Version: $Revision$
+
+
+ Copyright (c) Centre National d'Etudes Spatiales. All rights reserved.
+ See OTBCopyright.txt for details.
+
+
+ This software is distributed WITHOUT ANY WARRANTY; without even
+ the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR
+ PURPOSE. See the above copyright notices for more information.
+
+=========================================================================*/
+#include "otbNeuralNetworkMachineLearningModelFactory.h"
+
+#include "itkCreateObjectFunction.h"
+#include "otbNeuralNetworkMachineLearningModel.h"
+#include "itkVersion.h"
+
+namespace otb
+{
+
+template <class TInputValue, class TOutputValue>
+NeuralNetworkMachineLearningModelFactory<TInputValue,TOutputValue>
+::NeuralNetworkMachineLearningModelFactory()
+{
+
+ static std::string classOverride = std::string("otbMachineLearningModel");
+ static std::string subclass = std::string("otbNeuralNetworkMachineLearningModel");
+
+ this->RegisterOverride(classOverride.c_str(),
+ subclass.c_str(),
+ "Artificial Neural Network ML Model",
+ 1,
+ itk::CreateObjectFunction<NeuralNetworkMachineLearningModel<TInputValue,TOutputValue> >::New());
+}
+
+template <class TInputValue, class TOutputValue>
+NeuralNetworkMachineLearningModelFactory<TInputValue,TOutputValue>
+::~NeuralNetworkMachineLearningModelFactory()
+{
+}
+
+template <class TInputValue, class TOutputValue>
+const char*
+NeuralNetworkMachineLearningModelFactory<TInputValue,TOutputValue>
+::GetITKSourceVersion(void) const
+{
+ return ITK_SOURCE_VERSION;
+}
+
+template <class TInputValue, class TOutputValue>
+const char*
+NeuralNetworkMachineLearningModelFactory<TInputValue,TOutputValue>
+::GetDescription() const
+{
+ return "Artificial Neural Network machine learning model factory";
+}
+
+} // end namespace otb
diff --git a/Testing/Code/Learning/CMakeLists.txt b/Testing/Code/Learning/CMakeLists.txt
index 80e295b52e..686221a969 100644
--- a/Testing/Code/Learning/CMakeLists.txt
+++ b/Testing/Code/Learning/CMakeLists.txt
@@ -728,7 +728,6 @@ IF(OTB_USE_OPENCV)
ADD_TEST(leTuBoostMachineLearningModelNew ${LEARNING_TESTS6}
otbBoostMachineLearningModelNew)
-
ADD_TEST(leTvBoostMachineLearningModel ${LEARNING_TESTS6}
#--compare-ascii ${NOTOL}
#${BASELINE_FILES}/BoostLearningModel.txt
@@ -737,6 +736,19 @@ IF(OTB_USE_OPENCV)
${INPUTDATA}/letter.scale
${TEMP}/BoostMachineLearningModel.txt
)
+
+ ADD_TEST(leTuANNMachineLearningModelNew ${LEARNING_TESTS6}
+ otbANNMachineLearningModelNew)
+
+
+ ADD_TEST(leTvANNMachineLearningModel ${LEARNING_TESTS6}
+ #--compare-ascii ${NOTOL}
+ #${BASELINE_FILES}/ANNMachineLearningModel.txt
+ #${TEMP}/ANNMachineLearningModel.txt
+ otbANNMachineLearningModel
+ ${INPUTDATA}/letter.scale
+ ${TEMP}/ANNMachineLearningModel.txt
+ )
ADD_TEST(leTuImageClassificationFilterNew ${LEARNING_TESTS6}
otbImageClassificationFilterNew)
@@ -782,6 +794,11 @@ IF(OTB_USE_OPENCV)
otbBoostMachineLearningModelCanRead
${INPUTDATA}/boost_model.txt
)
+
+ ADD_TEST(leTuANNMachineLearningModelCanRead ${LEARNING_TESTS6}
+ otbNeuralNetworkMachineLearningModelCanRead
+ ${INPUTDATA}/NeuralNetworkMachineLearningModel.txt
+ )
ENDIF(OTB_USE_OPENCV)
diff --git a/Testing/Code/Learning/otbLearningTests6.cxx b/Testing/Code/Learning/otbLearningTests6.cxx
index 9d597f25ca..749102badb 100644
--- a/Testing/Code/Learning/otbLearningTests6.cxx
+++ b/Testing/Code/Learning/otbLearningTests6.cxx
@@ -34,10 +34,13 @@ void RegisterTests()
REGISTER_TEST(otbRandomForestsMachineLearningModel);
REGISTER_TEST(otbBoostMachineLearningModelNew);
REGISTER_TEST(otbBoostMachineLearningModel);
+ REGISTER_TEST(otbANNMachineLearningModelNew);
+ REGISTER_TEST(otbANNMachineLearningModel);
REGISTER_TEST(otbImageClassificationFilterNew);
REGISTER_TEST(otbImageClassificationFilter);
REGISTER_TEST(otbLibSVMMachineLearningModelCanRead);
REGISTER_TEST(otbSVMMachineLearningModelCanRead);
REGISTER_TEST(otbRandomForestsMachineLearningModelCanRead);
REGISTER_TEST(otbBoostMachineLearningModelCanRead);
+ REGISTER_TEST(otbNeuralNetworkMachineLearningModelCanRead);
}
diff --git a/Testing/Code/Learning/otbMachineLearningModelCanRead.cxx b/Testing/Code/Learning/otbMachineLearningModelCanRead.cxx
index 88b6537a34..5cb62adee1 100644
--- a/Testing/Code/Learning/otbMachineLearningModelCanRead.cxx
+++ b/Testing/Code/Learning/otbMachineLearningModelCanRead.cxx
@@ -21,6 +21,7 @@
#include "otbSVMMachineLearningModel.h"
#include "otbRandomForestsMachineLearningModel.h"
#include "otbBoostMachineLearningModel.h"
+#include "otbNeuralNetworkMachineLearningModel.h"
#include <iostream>
typedef otb::MachineLearningModel<float,short> MachineLearningModelType;
@@ -135,3 +136,30 @@ int otbBoostMachineLearningModelCanRead(int argc, char* argv[])
return EXIT_SUCCESS;
}
+int otbNeuralNetworkMachineLearningModelCanRead(int argc, char* argv[])
+{
+ if (argc != 2)
+ {
+ std::cerr << "Usage: " << argv[0]
+ << "<model>" << std::endl;
+ std::cerr << "Called here with " << argc << " arguments\n";
+ for (int i = 1; i < argc; ++i)
+ {
+ std::cerr << " - " << argv[i] << "\n";
+ }
+ return EXIT_FAILURE;
+ }
+ std::string filename(argv[1]);
+ typedef otb::NeuralNetworkMachineLearningModel<InputValueType, TargetValueType> ANNType;
+ ANNType::Pointer classifier = ANNType::New();
+ bool lCanRead = classifier->CanReadFile(filename);
+ if (lCanRead == false)
+ {
+ std::cerr << "Erreur otb::NeuralNetworkMachineLearningModel : impossible to open the file " << filename << "." << std::endl;
+ return EXIT_FAILURE;
+ }
+
+ return EXIT_SUCCESS;
+}
+
+
diff --git a/Testing/Code/Learning/otbTrainMachineLearningModel.cxx b/Testing/Code/Learning/otbTrainMachineLearningModel.cxx
index e19d0e6e8b..ba2e3f2968 100644
--- a/Testing/Code/Learning/otbTrainMachineLearningModel.cxx
+++ b/Testing/Code/Learning/otbTrainMachineLearningModel.cxx
@@ -26,6 +26,7 @@
#include "otbKNearestNeighborsMachineLearningModel.h"
#include "otbRandomForestsMachineLearningModel.h"
#include "otbBoostMachineLearningModel.h"
+#include "otbNeuralNetworkMachineLearningModel.h"
#include "otbConfusionMatrixCalculator.h"
@@ -412,5 +413,71 @@ int otbBoostMachineLearningModel(int argc, char * argv[])
return EXIT_SUCCESS;
}
+int otbANNMachineLearningModelNew(int argc, char * argv[])
+{
+ typedef otb::NeuralNetworkMachineLearningModel<InputValueType, TargetValueType> ANNType;
+ ANNType::Pointer classifier = ANNType::New();
+ return EXIT_SUCCESS;
+}
+
+int otbANNMachineLearningModel(int argc, char * argv[])
+{
+ if (argc != 3)
+ {
+ std::cout<<"Wrong number of arguments "<<std::endl;
+ std::cout<<"Usage : sample file, output file "<<std::endl;
+ return EXIT_FAILURE;
+ }
+
+
+ typedef otb::NeuralNetworkMachineLearningModel<InputValueType, TargetValueType> ANNType;
+ InputListSampleType::Pointer samples = InputListSampleType::New();
+ TargetListSampleType::Pointer labels = TargetListSampleType::New();
+ TargetListSampleType::Pointer predicted = TargetListSampleType::New();
+
+ if (!ReadDataFile(argv[1], samples, labels))
+ {
+ std::cout << "Failed to read samples file " << argv[1] << std::endl;
+ return EXIT_FAILURE;
+ }
+
+ std::cout<<"number of samples = "<<samples->Size()<<std::endl;
+
+ std::vector<unsigned int> layerSizes;
+ layerSizes.push_back(16);
+ layerSizes.push_back(25);
+ layerSizes.push_back(35);
+ layerSizes.push_back(45);
+ layerSizes.push_back(1);
+
+ ANNType::Pointer classifier = ANNType::New();
+ classifier->SetInputListSample(samples);
+ classifier->SetTargetListSample(labels);
+ classifier->SetLayerSizes(layerSizes);
+ classifier->SetTrainMethod(CvANN_MLP_TrainParams::BACKPROP);
+ classifier->SetBackPropDWScale(0.005);
+ classifier->SetBackPropMomentScale(0.005);
+ classifier->Train();
+
+ classifier->SetTargetListSample(predicted);
+ classifier->PredictAll();
+
+ ConfusionMatrixCalculatorType::Pointer cmCalculator = ConfusionMatrixCalculatorType::New();
+
+ cmCalculator->SetProducedLabels(predicted);
+ cmCalculator->SetReferenceLabels(labels);
+ cmCalculator->Compute();
+
+ std::cout << "Confusion matrix: " << std::endl;
+ std::cout << cmCalculator->GetConfusionMatrix() << std::endl;
+ std::cout << "Kappa: " << cmCalculator->GetKappaIndex() << std::endl;
+ std::cout << "Overall Accuracy: " << cmCalculator->GetOverallAccuracy() << std::endl;
+
+ classifier->Save(argv[2]);
+
+ return EXIT_SUCCESS;
+}
+
+
--
GitLab