Trainer for binary SVMs natively supporting missing features. More...
#include <shark/Algorithms/Trainers/MissingFeatureSvmTrainer.h>
Inheritance diagram for shark::MissingFeatureSvmTrainer< InputType, CacheType >:
Public Member Functions | |
| MissingFeatureSvmTrainer (KernelType *kernel, double C, bool offset, bool unconstrained=false) | |
| Constructor. More... | |
| std::string | name () const |
| From INameable: return the class name. More... | |
| void | train (MissingFeaturesKernelExpansion< InputType > &svm, LabeledData< InputType, unsigned int > const &dataset) |
| void | setMaxIterations (std::size_t newIterations) |
| Public Member Functions inherited from shark::AbstractSvmTrainer< InputType, unsigned int, MissingFeaturesKernelExpansion< InputType > > | |
| AbstractSvmTrainer (KernelType *kernel, double C, bool offset, bool unconstrained=false) | |
| AbstractSvmTrainer (KernelType *kernel, double negativeC, double positiveC, bool offset, bool unconstrained=false) | |
| double | C () const |
| Return the value of the regularization parameter C. More... | |
| void | setC (double C) |
| Set the value of the regularization parameter C. More... | |
| RealVector const & | regularizationParameters () const |
| void | setRegularizationParameters (RealVector const ®ularizers) |
| Set the value of the regularization parameter C. More... | |
| KernelType * | kernel () |
| KernelType const * | kernel () const |
| void | setKernel (KernelType *kernel) |
| bool | isUnconstrained () const |
| bool | trainOffset () const |
| std::size_t | cacheSize () const |
| void | setCacheSize (std::size_t size) |
| RealVector | parameterVector () const |
| get the hyper-parameter vector More... | |
| void | setParameterVector (RealVector const &newParameters) |
| set the vector of hyper-parameters More... | |
| size_t | numberOfParameters () const |
| return the number of hyper-parameters More... | |
| Public Member Functions inherited from shark::AbstractTrainer< Model, LabelTypeT > | |
| virtual void | train (ModelType &model, DatasetType const &dataset)=0 |
| Core of the Trainer interface. More... | |
| Public Member Functions inherited from shark::INameable | |
| virtual | ~INameable () |
| Public Member Functions inherited from shark::ISerializable | |
| virtual | ~ISerializable () |
| Virtual d'tor. More... | |
| virtual void | read (InArchive &archive) |
| Read the component from the supplied archive. More... | |
| virtual void | write (OutArchive &archive) const |
| Write the component to the supplied archive. More... | |
| void | load (InArchive &archive, unsigned int version) |
| Versioned loading of components, calls read(...). More... | |
| void | save (OutArchive &archive, unsigned int version) const |
| Versioned storing of components, calls write(...). More... | |
| BOOST_SERIALIZATION_SPLIT_MEMBER () | |
| Public Member Functions inherited from shark::QpConfig | |
| QpConfig (bool precomputedFlag=false, bool sparsifyFlag=true) | |
| Constructor. More... | |
| QpStoppingCondition & | stoppingCondition () |
| Read/write access to the stopping condition. More... | |
| QpStoppingCondition const & | stoppingCondition () const |
| Read access to the stopping condition. More... | |
| QpSolutionProperties & | solutionProperties () |
| Access to the solution properties. More... | |
| bool & | precomputeKernel () |
| Flag for using a precomputed kernel matrix. More... | |
| bool const & | precomputeKernel () const |
| Flag for using a precomputed kernel matrix. More... | |
| bool & | sparsify () |
| Flag for sparsifying the model after training. More... | |
| bool const & | sparsify () const |
| Flag for sparsifying the model after training. More... | |
| bool & | shrinking () |
| Flag for shrinking in the decomposition solver. More... | |
| bool const & | shrinking () const |
| Flag for shrinking in the decomposition solver. More... | |
| bool & | s2do () |
| Flag for S2DO (instead of SMO) More... | |
| bool const & | s2do () const |
| Flag for S2DO (instead of SMO) More... | |
| unsigned int & | verbosity () |
| Verbosity level of the solver. More... | |
| unsigned int const & | verbosity () const |
| Verbosity level of the solver. More... | |
| unsigned long long const & | accessCount () const |
| Number of kernel accesses. More... | |
| void | setMinAccuracy (double a) |
| void | setMaxIterations (unsigned long long i) |
| void | setTargetValue (double v) |
| void | setMaxSeconds (double s) |
| Public Member Functions inherited from shark::IParameterizable<> | |
| virtual | ~IParameterizable () |
Protected Types | |
| typedef CacheType | QpFloatType |
| typedef AbstractKernelFunction< InputType > | KernelType |
Additional Inherited Members | |
| Public Types inherited from shark::AbstractSvmTrainer< InputType, unsigned int, MissingFeaturesKernelExpansion< InputType > > | |
| typedef AbstractKernelFunction< InputType > | KernelType |
| Public Types inherited from shark::AbstractTrainer< Model, LabelTypeT > | |
| typedef Model | ModelType |
| typedef ModelType::InputType | InputType |
| typedef LabelTypeT | LabelType |
| typedef LabeledData< InputType, LabelType > | DatasetType |
| Public Types inherited from shark::IParameterizable<> | |
| typedef RealVector | ParameterVectorType |
| Protected Attributes inherited from shark::AbstractSvmTrainer< InputType, unsigned int, MissingFeaturesKernelExpansion< InputType > > | |
| KernelType * | m_kernel |
| RealVector | m_regularizers |
| Vector of regularization parameters. More... | |
| bool | m_trainOffset |
| bool | m_unconstrained |
| Is log(C) stored internally as a parameter instead of C? If yes, then we get rid of the constraint C > 0 on the level of the parameter interface. More... | |
| std::size_t | m_cacheSize |
| Number of values in the kernel cache. The size of the cache in bytes is the size of one entry (4 for float, 8 for double) times this number. More... | |
| Protected Attributes inherited from shark::QpConfig | |
| QpStoppingCondition | m_stoppingcondition |
| conditions for when to stop the QP solver More... | |
| QpSolutionProperties | m_solutionproperties |
| properties of the approximate solution found by the solver More... | |
| bool | m_precomputedKernelMatrix |
| should the solver use a precomputed kernel matrix? More... | |
| bool | m_sparsify |
| should the trainer sparsify the model after training? More... | |
| bool | m_shrinking |
| should shrinking be used? More... | |
| bool | m_s2do |
| should S2DO be used instead of SMO? More... | |
| unsigned int | m_verbosity |
| verbosity level (currently unused) More... | |
| unsigned long long | m_accessCount |
| kernel access count More... | |
Detailed Description
template<class InputType, class CacheType = float>
class shark::MissingFeatureSvmTrainer< InputType, CacheType >
Trainer for binary SVMs natively supporting missing features.
This is a specialized variant of the standard binary C-SVM which can be used to learn from data with missing features, without the need for prior imputation of missing values. The key idea is that each example is considered as having an instance-specific margin value, which is computed in the lower-dimensional subspace for which all features of that example are present.
The resulting optimization problem has the drawback that it is not convex any more. However, a local minimum can be easily reached by an iterative wrapper algorithm around a standard C-SVM solver. In detail, example-wise weights \( s_i \) are incorporated into the quadratic term of the standard SVM optimization problem. These take initial values of 1, and are then iteratively updated according to the instance-specific margin values. After each such update, the SVM solver is again called, and so on. Usually, between 2 and 5 iterations have been shown to be sufficient for an acceptable level of convergence.
For details, see the paper:
Max-margin Classification of Data with Absent Features Gal Chechik, Geremy Heitz, Gal Elidan, Pieter Abbeel and Daphne Koller. JMLR 2008.
- Note
- Much of the code in this class is borrowed from the class CSvmTrainer
Definition at line 72 of file MissingFeatureSvmTrainer.h.
Member Typedef Documentation
◆ KernelType
|
protected |
Definition at line 77 of file MissingFeatureSvmTrainer.h.
◆ QpFloatType
|
protected |
Definition at line 76 of file MissingFeatureSvmTrainer.h.
Constructor & Destructor Documentation
◆ MissingFeatureSvmTrainer()
|
inline |
Constructor.
Definition at line 83 of file MissingFeatureSvmTrainer.h.
Member Function Documentation
◆ name()
|
inlinevirtual |
From INameable: return the class name.
Reimplemented from shark::INameable.
Definition at line 90 of file MissingFeatureSvmTrainer.h.
◆ setMaxIterations()
|
inline |
Definition at line 105 of file MissingFeatureSvmTrainer.h.
References shark::AbstractSvmTrainer< InputType, unsigned int, MissingFeaturesKernelExpansion< InputType > >::C(), shark::MissingFeaturesKernelExpansion< InputType >::computeNorm(), shark::LabeledData< InputT, LabelT >::element(), shark::LabeledData< InputT, LabelT >::inputs(), shark::LabeledData< InputT, LabelT >::labels(), shark::AbstractSvmTrainer< InputType, unsigned int, MissingFeaturesKernelExpansion< InputType > >::m_kernel, shark::QpConfig::m_shrinking, shark::LabeledData< InputT, LabelT >::numberOfElements(), SHARK_ASSERT, shark::QpConfig::solutionProperties(), shark::QpSolver< Problem, SelectionStrategy >::solve(), and shark::QpConfig::stoppingCondition().
◆ train()
|
inline |
Definition at line 93 of file MissingFeatureSvmTrainer.h.
References shark::KernelExpansion< InputType >::hasOffset(), shark::LabeledData< InputT, LabelT >::inputs(), shark::AbstractSvmTrainer< InputType, unsigned int, MissingFeaturesKernelExpansion< InputType > >::m_kernel, shark::AbstractSvmTrainer< InputType, unsigned int, MissingFeaturesKernelExpansion< InputType > >::m_trainOffset, shark::numberOfClasses(), shark::KernelExpansion< InputType >::setStructure(), and SHARK_RUNTIME_CHECK.
The documentation for this class was generated from the following file:
- include/shark/Algorithms/Trainers/MissingFeatureSvmTrainer.h