From 64af665b6b1244a10b668bdfcadf6f3a75a6e4ac Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?C=C3=A9dric=20Traizet?= <traizetc@cesbio.cnes.fr>
Date: Wed, 10 May 2017 11:40:15 +0200
Subject: [PATCH] bug can't find appli
---
app/cbDimensionalityReduction.cxx | 2 +-
app/cbDimensionalityReductionTrainer.cxx | 141 +++++++++--------------
include/AutoencoderModel.h | 10 +-
include/AutoencoderModel.txx | 9 ++
include/DimensionalityReductionModel.h | 48 --------
include/cbLearningApplicationBaseDR.h | 4 +-
include/cbTrainAutoencoder.txx | 10 +-
include/encode_filter.h | 8 +-
8 files changed, 80 insertions(+), 152 deletions(-)
delete mode 100644 include/DimensionalityReductionModel.h
diff --git a/app/cbDimensionalityReduction.cxx b/app/cbDimensionalityReduction.cxx
index b195d053ba..9dfa3d6810 100644
--- a/app/cbDimensionalityReduction.cxx
+++ b/app/cbDimensionalityReduction.cxx
@@ -93,7 +93,7 @@ private:
FilterType::Pointer filter_dim_reduc;
ExtractROIFilterType::Pointer m_ExtractROIFilter;
-
+ //d
};
}
}
diff --git a/app/cbDimensionalityReductionTrainer.cxx b/app/cbDimensionalityReductionTrainer.cxx
index 9c7cc45142..98b88417f1 100644
--- a/app/cbDimensionalityReductionTrainer.cxx
+++ b/app/cbDimensionalityReductionTrainer.cxx
@@ -6,6 +6,9 @@
#include "itkVariableLengthVector.h"
+#include "otbShiftScaleSampleListFilter.h"
+#include "otbStatisticsXMLFileReader.h"
+
//#include "AutoencoderModel.h"
#include "otbSharkUtils.h"
@@ -29,39 +32,6 @@
#include "cbLearningApplicationBaseDR.h"
-template<class AutoencoderModel>
-AutoencoderModel trainAutoencoderModel(
- shark::UnlabeledData<shark::RealVector> const& data,//the data to train with
- std::size_t numHidden,//number of features in the autoencoder
- std::size_t iterations, //number of iterations to optimize
- double regularisation//strength of the regularisation
-){
- //create the model
- std::size_t inputs = dataDimension(data);
- AutoencoderModel model;
- model.setStructure(inputs, numHidden);
- initRandomUniform(model,-0.1*std::sqrt(1.0/inputs),0.1*std::sqrt(1.0/inputs));
- //create the objective function
-
- shark::LabeledData<shark::RealVector,shark::RealVector> trainSet(data,data);//labels identical to inputs
- shark::SquaredLoss<shark::RealVector> loss;
- shark::ErrorFunction error(trainSet, &model, &loss);
- shark::TwoNormRegularizer regularizer(error.numberOfVariables());
- error.setRegularizer(regularisation,®ularizer);
-
- shark::IRpropPlusFull optimizer;
- error.init();
- optimizer.init(error);
- std::cout<<"Optimizing model: "+model.name()<<std::endl;
- for(std::size_t i = 0; i != iterations; ++i){
- optimizer.step(error);
- std::cout<<i<<" "<<optimizer.solution().value<<std::endl;
- }
- //std::cout<<optimizer.solution().value<<std::endl;
- model.setParameterVector(optimizer.solution().point);
- return model;
-}
-
shark::Normalizer<shark::RealVector> trainNormalizer(const shark::UnlabeledData<shark::RealVector>& data)
{
bool removeMean = true;
@@ -88,11 +58,24 @@ public:
itkTypeMacro(CbDimensionalityReductionTrainer, otb::Application);
- typedef float ValueType;
- typedef itk::VariableLengthVector<ValueType> InputSampleType;
- typedef itk::Statistics::ListSample<InputSampleType> ListSampleType;
-
+ typedef Superclass::SampleType SampleType;
+ typedef Superclass::ListSampleType ListSampleType;
+ typedef Superclass::SampleImageType SampleImageType;
+
+ typedef double ValueType;
typedef itk::VariableLengthVector<ValueType> MeasurementType;
+
+ typedef otb::StatisticsXMLFileReader<SampleType> StatisticsReader;
+
+ typedef otb::Statistics::ShiftScaleSampleListFilter<ListSampleType, ListSampleType> ShiftScaleFilterType;
+
+
+
+ //typedef float ValueType;
+ //typedef itk::VariableLengthVector<ValueType> InputSampleType;
+ //typedef itk::Statistics::ListSample<InputSampleType> ListSampleType;
+
+ //typedef itk::VariableLengthVector<ValueType> MeasurementType;
typedef otb::MachineLearningModelFactory<ValueType, ValueType> ModelFactoryType;
@@ -105,16 +88,7 @@ private:
{
SetName("CbDimensionalityReductionTrainer");
SetDescription("Trainer for the dimensionality reduction algorithms used in the cbDimensionalityReduction application.");
- /*
- AddParameter(ParameterType_InputVectorData, "train", "Name of the input training vector data");
- SetParameterDescription("train","The vector data used for training.");
- AddParameter(ParameterType_StringList, "feat", "Field names to be calculated."); //
- SetParameterDescription("feat","List of field names in the input vector data used as features for training."); //
-
- AddParameter(ParameterType_Int, "k","target dimension");
- SetParameterDescription("k", "Dimension of the output feature vectors");
-*/
AddParameter(ParameterType_Group, "io", "Input and output data");
SetParameterDescription("io", "This group of parameters allows setting input and output data.");
@@ -124,6 +98,9 @@ private:
AddParameter(ParameterType_OutputFilename, "io.out", "Output model");
SetParameterDescription("io.out", "Output file containing the model estimated (.txt format).");
+ AddParameter(ParameterType_InputFilename, "io.stats", "Input XML image statistics file");
+ MandatoryOff("io.stats");
+ SetParameterDescription("io.stats", "XML file containing mean and variance of each feature.");
AddParameter(ParameterType_StringList, "feat", "Field names to be calculated."); //
SetParameterDescription("feat","List of field names in the input vector data used as features for training."); //
@@ -153,7 +130,7 @@ private:
void DoExecute()
{
- std::cout << "Appli !" << std::endl;
+ std::cout << "Appli Training!" << std::endl;
std::string shapefile = GetParameterString("io.vd");
@@ -175,51 +152,37 @@ private:
}
input->PushBack(mv);
}
- /*
- std::cout << input << std::endl;
- std::vector<shark::RealVector> features;
- otb::Shark::ListSampleToSharkVector<ListSampleType>( input, features);
- shark::Data<shark::RealVector> inputSamples = shark::createDataFromRange( features );
- std::size_t numHidden= GetParameterInt("k");
- std::size_t iterations = 100;
- double regularisation = 0;
-
-
- shark::Normalizer<shark::RealVector> normalizer = trainNormalizer(inputSamples);
- inputSamples = normalizer(inputSamples);
+// Statistics for shift/scale
+ MeasurementType meanMeasurementVector;
+ MeasurementType stddevMeasurementVector;
+ if (HasValue("io.stats") && IsParameterEnabled("io.stats"))
+ {
+ StatisticsReader::Pointer statisticsReader = StatisticsReader::New();
+ std::string XMLfile = GetParameterString("io.stats");
+ statisticsReader->SetFileName(XMLfile);
+ meanMeasurementVector = statisticsReader->GetStatisticVectorByName("mean");
+ stddevMeasurementVector = statisticsReader->GetStatisticVectorByName("stddev");
+ }
+ else
+ {
+ meanMeasurementVector.SetSize(nbFeatures);
+ meanMeasurementVector.Fill(0.);
+ stddevMeasurementVector.SetSize(nbFeatures);
+ stddevMeasurementVector.Fill(1.);
+ }
+
+ ShiftScaleFilterType::Pointer trainingShiftScaleFilter = ShiftScaleFilterType::New();
+ trainingShiftScaleFilter->SetInput(input);
+ trainingShiftScaleFilter->SetShifts(meanMeasurementVector);
+ trainingShiftScaleFilter->SetScales(stddevMeasurementVector);
+ trainingShiftScaleFilter->Update();
+
+ ListSampleType::Pointer trainingListSample= trainingShiftScaleFilter->GetOutput();
- std::cout << "normalizer trained and training set normalized" << std::endl;
-
- AutoencoderType net = trainAutoencoderModel<AutoencoderType>(inputSamples,numHidden,iterations,regularisation);
- std::cout << "autoencoder trained !!!!" << std::endl;
-
- // save the model to the file "net.model"
- std::ofstream ofs("net.model");
- shark::TextOutArchive oa(ofs);
- net.write(oa);
- ofs.close();
- // save the model to the file "net.model"
- std::ofstream norm_ofs("normalizer.model");
- boost::archive::polymorphic_text_oarchive onorm(norm_ofs);
- normalizer.write(onorm);
- norm_ofs.close();
- */
-
- std::cout << "Using a Machine learning model" << std::endl;
- /*
- AutoencoderModelType::Pointer dimredTrainer = AutoencoderModelType::New();
- dimredTrainer->SetNumberOfHiddenNeurons(5);
- dimredTrainer->SetNumberOfIterations(50);
- dimredTrainer->SetRegularization(0.1);
- dimredTrainer->SetInputListSample(input);
- dimredTrainer->Train();
- dimredTrainer->Save("net.model");
- std::cout << "ok" << std::endl;
- */
- this->Train(input,GetParameterString("io.out"));
-
+ this->Train(trainingListSample,GetParameterString("io.out"));
+ // d
}
diff --git a/include/AutoencoderModel.h b/include/AutoencoderModel.h
index 603f0dc4d2..e77e901ae1 100644
--- a/include/AutoencoderModel.h
+++ b/include/AutoencoderModel.h
@@ -51,10 +51,14 @@ public:
protected:
- AutoencoderModel(){};
-private:
+ AutoencoderModel();
+ //~AutoencoderModel() ITK_OVERRIDE;
+
virtual TargetSampleType DoPredict(const InputSampleType& input, ConfidenceValueType *quality=ITK_NULLPTR) const ITK_OVERRIDE;
-
+ virtual void DoPredictBatch(const InputListSampleType *, const unsigned int & startIndex, const unsigned int & size, TargetListSampleType *, ConfidenceListSampleType * = ITK_NULLPTR) const ITK_OVERRIDE;
+
+private:
+
AutoencoderType m_net;
unsigned int m_NumberOfHiddenNeurons;
unsigned int m_NumberOfIterations;
diff --git a/include/AutoencoderModel.txx b/include/AutoencoderModel.txx
index 50fc9940f9..72f9e2ef8f 100644
--- a/include/AutoencoderModel.txx
+++ b/include/AutoencoderModel.txx
@@ -14,6 +14,15 @@
namespace otb
{
+
+template <class TInputValue, class AutoencoderType>
+AutoencoderModel<TInputValue,AutoencoderType>::AutoencoderModel()
+{
+ //this->m_IsRegressionSupported = true;
+}
+
+
+
template <class TInputValue, class AutoencoderType>
void AutoencoderModel<TInputValue,AutoencoderType>::Train()
{
diff --git a/include/DimensionalityReductionModel.h b/include/DimensionalityReductionModel.h
deleted file mode 100644
index 05066e976b..0000000000
--- a/include/DimensionalityReductionModel.h
+++ /dev/null
@@ -1,48 +0,0 @@
-#ifndef DimensionalityReductionModel_h
-#define DimensionalityReductionModel_h
-#include "itkObject.h"
-#include "itkListSample.h"
-namespace otb
-{
-template <class TInputValue>
-class DimensionalityReductionModel: public itk::Object
-{
-
-public:
-
- typedef DimensionalityReductionModel Self;
- typedef itk::Object Superclass;
- typedef itk::SmartPointer<Self> Pointer;
- typedef itk::SmartPointer<const Self> ConstPointer;
-
- typedef TInputValue InputValueType;
- typedef itk::VariableLengthVector<InputValueType> InputSampleType;
- typedef itk::Statistics::ListSample<InputSampleType> InputListSampleType;
-
-
-
- itkSetObjectMacro(InputListSample,InputListSampleType);
- itkGetObjectMacro(InputListSample,InputListSampleType);
- itkGetConstObjectMacro(InputListSample,InputListSampleType);
-
- virtual void Save(const std::string & filename, const std::string & name="") = 0;
- virtual void Load(const std::string & filename, const std::string & name="") = 0;
-
- virtual void Train() = 0;
- virtual void Dimensionality_reduction() = 0;
-
-protected:
- DimensionalityReductionModel(){};
- typename InputListSampleType::Pointer m_InputListSample;
-
-
-};
-} // end namespace otb
-
-
-//#ifndef OTB_MANUAL_INSTANTIATION
-//#include "DimensionalityReductionModel.txx"
-//#endif
-
-
-#endif
diff --git a/include/cbLearningApplicationBaseDR.h b/include/cbLearningApplicationBaseDR.h
index fe42e2d875..ab359092d2 100644
--- a/include/cbLearningApplicationBaseDR.h
+++ b/include/cbLearningApplicationBaseDR.h
@@ -117,7 +117,7 @@ private:
#ifdef OTB_USE_SHARK
void InitAutoencoderParams();
template <class autoencoderchoice>
- void TrainAutoencoder(typename ListSampleType::Pointer trainingListSample, std::string modelPath){
+ void TrainAutoencoder(typename ListSampleType::Pointer trainingListSample, std::string modelPath);/*{
// typename AutoencoderModelType::Pointer dimredTrainer = AutoencoderModelType::New();
typename autoencoderchoice::Pointer dimredTrainer = autoencoderchoice::New();
dimredTrainer->SetNumberOfHiddenNeurons(GetParameterInt("model.autoencoder.nbneuron"));
@@ -126,7 +126,7 @@ private:
dimredTrainer->SetInputListSample(trainingListSample);
dimredTrainer->Train();
dimredTrainer->Save(modelPath);
-}; // !!!!!!!!!!!!!!!!! How to declare this method body in the .txx ? (double template...)
+}; // !!!!!!!!!!!!!!!!! How to declare this method body in the .txx ? (double template...) */
#endif
//@}
};
diff --git a/include/cbTrainAutoencoder.txx b/include/cbTrainAutoencoder.txx
index f28bad3297..feed6cf75e 100644
--- a/include/cbTrainAutoencoder.txx
+++ b/include/cbTrainAutoencoder.txx
@@ -44,13 +44,13 @@ cbLearningApplicationBaseDR<TInputValue,TOutputValue>
SetParameterDescription("model.autoencoder.normalizer",
"Strength of the L2 normalization used during training");
}
-/*
+
template <class TInputValue, class TOutputValue>
+template <typename autoencoderchoice>
void cbLearningApplicationBaseDR<TInputValue,TOutputValue>
-::template < autoencoderchoice> TrainAutoencoder(typename ListSampleType::Pointer trainingListSample,std::string modelPath)
+::TrainAutoencoder(typename ListSampleType::Pointer trainingListSample,std::string modelPath)
{
- // typename AutoencoderModelType::Pointer dimredTrainer = AutoencoderModelType::New();
- autoencoderchoice::Pointer dimredTrainer = autoencoderchoice::New();
+ typename autoencoderchoice::Pointer dimredTrainer = autoencoderchoice::New();
dimredTrainer->SetNumberOfHiddenNeurons(GetParameterInt("model.autoencoder.nbneuron"));
dimredTrainer->SetNumberOfIterations(GetParameterInt("model.autoencoder.nbiter"));
dimredTrainer->SetRegularization(GetParameterFloat("model.autoencoder.normalizer"));
@@ -58,7 +58,7 @@ void cbLearningApplicationBaseDR<TInputValue,TOutputValue>
dimredTrainer->Train();
dimredTrainer->Save(modelPath);
}
-*/
+
} //end namespace wrapper
} //end namespace otb
diff --git a/include/encode_filter.h b/include/encode_filter.h
index cd8f523505..903912be28 100644
--- a/include/encode_filter.h
+++ b/include/encode_filter.h
@@ -9,15 +9,15 @@ template< class TImage, class AutoencoderModel, class NormalizerModel>
class ITK_EXPORT EncodeFilter:public itk::ImageToImageFilter< TImage, TImage >
{
public:
- /** Standard class typedefs. */
+ // Standard class typedefs.
typedef EncodeFilter Self;
typedef itk::ImageToImageFilter< TImage, TImage > Superclass;
typedef itk::SmartPointer< Self > Pointer;
- /** Method for creation through the object factory. */
+ // Method for creation through the object factory.
itkNewMacro(Self);
- /** Run-time type information (and related methods). */
+ // Run-time type information (and related methods).
itkTypeMacro(EncodeFilter, ImageToImageFilter);
//void SetInputImage(const TImage* image);
@@ -33,7 +33,7 @@ class ITK_EXPORT EncodeFilter:public itk::ImageToImageFilter< TImage, TImage >
AutoencoderModel GetAutoencoderModel();
NormalizerModel GetNormalizerModel();
- /** Does the real work. */
+ // Does the real work.
virtual void GenerateOutputInformation();
virtual void BeforeThreadedGenerateData();
void ThreadedGenerateData(const typename TImage::RegionType &outputRegionForThread, unsigned int threadId) ITK_OVERRIDE;
--
GitLab