Top level interface for everything that holds parameters. More...
#include <shark/Core/IParameterizable.h>
Inheritance diagram for shark::IParameterizable:
Collaboration diagram for shark::IParameterizable:
Public Member Functions | |
| virtual | ~IParameterizable () |
| virtual RealVector | parameterVector () const |
| Return the parameter vector. More... | |
| virtual void | setParameterVector (RealVector const &newParameters) |
| Set the parameter vector. More... | |
| virtual std::size_t | numberOfParameters () const |
| Return the number of parameters. More... | |
Detailed Description
Top level interface for everything that holds parameters.
This interface is inherited by AbstractModel for unified access to the parameters of models, but also by objective functions and algorithms with hyper-parameters.
Definition at line 52 of file IParameterizable.h.
Constructor & Destructor Documentation
|
inlinevirtual |
Definition at line 54 of file IParameterizable.h.
Member Function Documentation
|
inlinevirtual |
Return the number of parameters.
Reimplemented in shark::ConcatenatedModel< InputType, OutputType >, shark::AbstractLinearSvmTrainer, shark::TruncatedExponentialLayer, shark::GaussianLayer, shark::BinaryLayer, shark::AbstractSvmTrainer< InputType, LabelType, Model >, shark::AbstractSvmTrainer< InputType, RealVector >, shark::AbstractSvmTrainer< InputType, unsigned int, MissingFeaturesKernelExpansion< InputType > >, shark::AbstractSvmTrainer< InputType, unsigned int >, shark::KernelExpansion< InputType >, shark::OneHotConverter, shark::Normalizer< DataType >, shark::WeightedSumKernel< InputType >, shark::ArgMaxConverter, shark::LinearModel< InputType, OutputType >, shark::RNNet, shark::detail::LinearModelWrapper< Matrix, InputType, OutputType >, shark::OneClassSvmTrainer< InputType, CacheType >, shark::PolynomialKernel< InputType >, shark::detail::ConcatenatedModelWrapper< InputType, IntermediateType, OutputType >, shark::CARTClassifier< LabelType >, shark::CARTClassifier< RealVector >, shark::EpsilonSvmTrainer< InputType, CacheType >, shark::ThresholdVectorConverter, shark::LassoRegression< InputVectorType >, shark::CMACMap, shark::ProductKernel< InputType >, shark::RBFNet, shark::ARDKernelUnconstrained< InputType >, shark::HierarchicalClustering< InputT >, shark::NearestNeighborRegression< InputType >, shark::NormalizedKernel< InputType >, shark::OneVersusOneClassifier< InputType >, shark::OnlineRNNet, shark::NearestNeighborClassifier< InputType >, shark::SoftNearestNeighborClassifier< InputType >, shark::RBM< VisibleLayerT, HiddenLayerT, RngT >, shark::FFNet< HiddenNeuron, OutputNeuron >, shark::GaussianRbfKernel< InputType >, shark::MonomialKernel< InputType >, shark::ThresholdConverter, shark::LDA, shark::LinearRegression, shark::DiscreteKernel, shark::Softmax, shark::detail::MklKernelWrapper< InputType, N >, shark::ScaledKernel< InputType >, shark::Centroids, shark::ClusteringModel< InputT, OutputT >, shark::ClusteringModel< InputT, RealVector >, shark::ClusteringModel< InputT, unsigned int >, shark::detail::SubrangeKernelWrapper< InputType >, shark::LinearClassifier, shark::SigmoidModel, and shark::LinearNorm.
Definition at line 69 of file IParameterizable.h.
References parameterVector().
Referenced by shark::ProductKernel< InputType >::addKernel(), shark::calculateKernelMatrixParameterDerivative(), shark::CSvmTrainer< InputType, CacheType >::computeBias(), shark::NegativeGaussianProcessEvidence< InputType, OutputType, LabelType >::eval(), shark::RadiusMarginQuotient< InputType, CacheType >::eval(), shark::DenoisingAutoencoderError< InputType, RngType >::evalDerivative(), shark::RadiusMarginQuotient< InputType, CacheType >::evalDerivative(), shark::NegativeGaussianProcessEvidence< InputType, OutputType, LabelType >::evalDerivative(), shark::SpanBoundCSvm< InputType >::evalDerivative(), shark::KernelTargetAlignment< InputType >::evalDerivative(), shark::detail::ParallelLossBasedErrorFunctionImpl< InputType, LabelType, OutputType >::evalDerivative(), shark::detail::LossBasedErrorFunctionImpl< InputType, LabelType, OutputType >::evalDerivativePointSet(), shark::initRandomNormal(), shark::initRandomUniform(), shark::detail::SubrangeKernelWrapper< InputType >::numberOfParameters(), shark::ClusteringModel< InputT, unsigned int >::numberOfParameters(), shark::ScaledKernel< InputType >::numberOfParameters(), shark::detail::MklKernelWrapper< InputType, N >::numberOfParameters(), shark::NormalizedKernel< InputType >::numberOfParameters(), shark::detail::ConcatenatedModelWrapper< InputType, IntermediateType, OutputType >::numberOfParameters(), shark::OneClassSvmTrainer< InputType, CacheType >::numberOfParameters(), shark::AbstractSvmTrainer< InputType, unsigned int >::numberOfParameters(), shark::LooErrorCSvm< InputType, CacheType >::numberOfVariables(), shark::detail::ErrorFunctionWrapper< InputType, LabelType, OutputType >::numberOfVariables(), shark::NegativeGaussianProcessEvidence< InputType, OutputType, LabelType >::numberOfVariables(), shark::DenoisingAutoencoderError< InputType, RngType >::numberOfVariables(), shark::CrossValidationError< ModelTypeT, LabelTypeT >::numberOfVariables(), shark::detail::NoisyErrorFunctionWrapper< InputType, LabelType, OutputType, RngType >::numberOfVariables(), shark::RadiusMarginQuotient< InputType, CacheType >::numberOfVariables(), shark::LooError< ModelTypeT, LabelType >::numberOfVariables(), shark::KernelTargetAlignment< InputType >::numberOfVariables(), shark::OneClassSvmTrainer< InputType, CacheType >::parameterVector(), shark::AbstractSvmTrainer< InputType, unsigned int >::parameterVector(), shark::OneClassSvmTrainer< InputType, CacheType >::setParameterVector(), shark::AbstractSvmTrainer< InputType, unsigned int >::setParameterVector(), shark::NegativeGaussianProcessEvidence< InputType, OutputType, LabelType >::setThreshold(), shark::detail::ConcatenatedModelWrapper< InputType, IntermediateType, OutputType >::weightedDerivatives(), and shark::detail::ConcatenatedModelWrapper< InputType, IntermediateType, OutputType >::weightedParameterDerivative().
|
inlinevirtual |
Return the parameter vector.
Reimplemented in shark::ConcatenatedModel< InputType, OutputType >, shark::AbstractLinearSvmTrainer, shark::TruncatedExponentialLayer, shark::GaussianLayer, shark::BinaryLayer, shark::AbstractSvmTrainer< InputType, LabelType, Model >, shark::AbstractSvmTrainer< InputType, RealVector >, shark::AbstractSvmTrainer< InputType, unsigned int, MissingFeaturesKernelExpansion< InputType > >, shark::AbstractSvmTrainer< InputType, unsigned int >, shark::KernelExpansion< InputType >, shark::OneHotConverter, shark::NBClassifier< InputType, OutputType >, shark::ArgMaxConverter, shark::Normalizer< DataType >, shark::LinearModel< InputType, OutputType >, shark::RNNet, shark::WeightedSumKernel< InputType >, shark::detail::LinearModelWrapper< Matrix, InputType, OutputType >, shark::CARTClassifier< LabelType >, shark::CARTClassifier< RealVector >, shark::FFNet< HiddenNeuron, OutputNeuron >, shark::ThresholdVectorConverter, shark::OneClassSvmTrainer< InputType, CacheType >, shark::detail::ConcatenatedModelWrapper< InputType, IntermediateType, OutputType >, shark::LassoRegression< InputVectorType >, shark::EpsilonSvmTrainer< InputType, CacheType >, shark::CMACMap, shark::ARDKernelUnconstrained< InputType >, shark::ProductKernel< InputType >, shark::RBM< VisibleLayerT, HiddenLayerT, RngT >, shark::HierarchicalClustering< InputT >, shark::PolynomialKernel< InputType >, shark::NormalizedKernel< InputType >, shark::RFClassifier, shark::NearestNeighborRegression< InputType >, shark::RBFNet, shark::OnlineRNNet, shark::ThresholdConverter, shark::MonomialKernel< InputType >, shark::NearestNeighborClassifier< InputType >, shark::SoftNearestNeighborClassifier< InputType >, shark::LinearRegression, shark::LDA, shark::OneVersusOneClassifier< InputType >, shark::DiscreteKernel, shark::GaussianRbfKernel< InputType >, shark::Centroids, shark::ScaledKernel< InputType >, shark::detail::MklKernelWrapper< InputType, N >, shark::Softmax, shark::ClusteringModel< InputT, OutputT >, shark::ClusteringModel< InputT, RealVector >, shark::ClusteringModel< InputT, unsigned int >, shark::LinearKernel< InputType >, shark::LinearClassifier, shark::detail::SubrangeKernelWrapper< InputType >, shark::KalmanFilter, shark::SigmoidModel, and shark::LinearNorm.
Definition at line 57 of file IParameterizable.h.
Referenced by numberOfParameters(), shark::detail::SubrangeKernelWrapper< InputType >::parameterVector(), shark::ClusteringModel< InputT, unsigned int >::parameterVector(), shark::detail::MklKernelWrapper< InputType, N >::parameterVector(), shark::ScaledKernel< InputType >::parameterVector(), shark::NormalizedKernel< InputType >::parameterVector(), shark::OneClassSvmTrainer< InputType, CacheType >::parameterVector(), shark::detail::ErrorFunctionWrapper< InputType, LabelType, OutputType >::proposeStartingPoint(), shark::DenoisingAutoencoderError< InputType, RngType >::proposeStartingPoint(), shark::detail::NoisyErrorFunctionWrapper< InputType, LabelType, OutputType, RngType >::proposeStartingPoint(), shark::SvmLogisticInterpretation< InputType >::proposeStartingPoint(), shark::AbstractKernelFunction< KernelInputType >::write(), and shark::AbstractModel< InputT, unsigned int >::write().
|
inlinevirtual |
Set the parameter vector.
Reimplemented in shark::ConcatenatedModel< InputType, OutputType >, shark::AbstractLinearSvmTrainer, shark::TruncatedExponentialLayer, shark::GaussianLayer, shark::BinaryLayer, shark::AbstractSvmTrainer< InputType, LabelType, Model >, shark::AbstractSvmTrainer< InputType, RealVector >, shark::AbstractSvmTrainer< InputType, unsigned int, MissingFeaturesKernelExpansion< InputType > >, shark::AbstractSvmTrainer< InputType, unsigned int >, shark::KernelExpansion< InputType >, shark::OneHotConverter, shark::NBClassifier< InputType, OutputType >, shark::Normalizer< DataType >, shark::WeightedSumKernel< InputType >, shark::ArgMaxConverter, shark::FFNet< HiddenNeuron, OutputNeuron >, shark::LinearModel< InputType, OutputType >, shark::RNNet, shark::detail::LinearModelWrapper< Matrix, InputType, OutputType >, shark::CARTClassifier< LabelType >, shark::CARTClassifier< RealVector >, shark::OneClassSvmTrainer< InputType, CacheType >, shark::ThresholdVectorConverter, shark::detail::ConcatenatedModelWrapper< InputType, IntermediateType, OutputType >, shark::EpsilonSvmTrainer< InputType, CacheType >, shark::LassoRegression< InputVectorType >, shark::PolynomialKernel< InputType >, shark::CMACMap, shark::RBM< VisibleLayerT, HiddenLayerT, RngT >, shark::ProductKernel< InputType >, shark::ARDKernelUnconstrained< InputType >, shark::HierarchicalClustering< InputT >, shark::NormalizedKernel< InputType >, shark::NearestNeighborRegression< InputType >, shark::RFClassifier, shark::RBFNet, shark::OnlineRNNet, shark::ThresholdConverter, shark::MonomialKernel< InputType >, shark::NearestNeighborClassifier< InputType >, shark::SoftNearestNeighborClassifier< InputType >, shark::OneVersusOneClassifier< InputType >, shark::LinearRegression, shark::LDA, shark::GaussianRbfKernel< InputType >, shark::DiscreteKernel, shark::Centroids, shark::detail::MklKernelWrapper< InputType, N >, shark::ScaledKernel< InputType >, shark::Softmax, shark::ClusteringModel< InputT, OutputT >, shark::ClusteringModel< InputT, RealVector >, shark::ClusteringModel< InputT, unsigned int >, shark::LinearKernel< InputType >, shark::LinearClassifier, shark::detail::SubrangeKernelWrapper< InputType >, shark::KalmanFilter, shark::SigmoidModel, and shark::LinearNorm.
Definition at line 63 of file IParameterizable.h.
References SHARK_ASSERT.
Referenced by shark::NegativeGaussianProcessEvidence< InputType, OutputType, LabelType >::eval(), shark::DenoisingAutoencoderError< InputType, RngType >::eval(), shark::detail::NoisyErrorFunctionWrapper< InputType, LabelType, OutputType, RngType >::eval(), shark::CrossValidationError< ModelTypeT, LabelTypeT >::eval(), shark::detail::CostBasedErrorFunctionImpl< InputType, LabelType, OutputType >::eval(), shark::RadiusMarginQuotient< InputType, CacheType >::eval(), shark::KernelTargetAlignment< InputType >::eval(), shark::SvmLogisticInterpretation< InputType >::eval(), shark::LooError< ModelTypeT, LabelType >::eval(), shark::detail::LossBasedErrorFunctionImpl< InputType, LabelType, OutputType >::eval(), shark::detail::ParallelLossBasedErrorFunctionImpl< InputType, LabelType, OutputType >::eval(), shark::detail::NoisyErrorFunctionWrapper< InputType, LabelType, OutputType, RngType >::evalDerivative(), shark::DenoisingAutoencoderError< InputType, RngType >::evalDerivative(), shark::RadiusMarginQuotient< InputType, CacheType >::evalDerivative(), shark::NegativeGaussianProcessEvidence< InputType, OutputType, LabelType >::evalDerivative(), shark::KernelTargetAlignment< InputType >::evalDerivative(), shark::detail::LossBasedErrorFunctionImpl< InputType, LabelType, OutputType >::evalDerivative(), shark::SvmLogisticInterpretation< InputType >::evalDerivative(), shark::detail::ParallelLossBasedErrorFunctionImpl< InputType, LabelType, OutputType >::evalDerivative(), shark::initRandomNormal(), shark::initRandomUniform(), shark::AbstractKernelFunction< KernelInputType >::read(), shark::AbstractModel< InputT, unsigned int >::read(), shark::detail::SubrangeKernelWrapper< InputType >::setParameterVector(), shark::ClusteringModel< InputT, unsigned int >::setParameterVector(), shark::detail::MklKernelWrapper< InputType, N >::setParameterVector(), shark::ScaledKernel< InputType >::setParameterVector(), shark::NormalizedKernel< InputType >::setParameterVector(), and shark::OneClassSvmTrainer< InputType, CacheType >::setParameterVector().
The documentation for this class was generated from the following file:
- include/shark/Core/IParameterizable.h