SHOGUN  6.1.3
CGaussianARDKernel Class Reference

## Detailed Description

Gaussian Kernel with Automatic Relevance Detection computed on CDotFeatures.

It is computed as

$k({\bf x},{\bf y})= \exp(-\frac{\Vert {\bf x}-{\bf y} \Vert}{2})$

There are three variants based on $$\Vert \cdot \Vert$$. The default case is $$\sum_{i=1}^{p}{{[\lambda \times ({\bf x_i}-{\bf y_i})] }^2}$$ where $$\lambda$$ is a positive scalar and $$p$$ is # of features. To use this case, please call set_scalar_weights( $$\lambda$$), where $$\lambda$$ is a positive scalar.

The second case is $$\sum_{i=1}^{p} {{[\lambda_i \times ({\bf x_i}-{\bf y_i})] }^2}$$ where $$\lambda$$ is a positive vector (we use $$\lambda$$ as a column vector) and $$p$$ is # of features. To use this case, please call set_vector_weights( $$\lambda$$), where $$\lambda$$ is a positive vector.

The last case is $$({\bf x}-{\bf y})^T \Lambda \Lambda^T ({\bf x}-{\bf y})$$ where $$\Lambda^T$$ is a $$d$$-by- $$p$$ upper triangular matrix with positive diagonal elements, $$p$$ is # of features and $$d \le p$$. To use this case, please call set_matrix_weights( $$\Lambda$$), where $$\Lambda$$ is a $$p$$-by- $$d$$ lower triangular matrix with positive diagonal elements. Note that only the lower triangular part of $$\Lambda$$ will be used

Indeed, the last case is more general than the first two cases. When $$\Lambda=\lambda I$$ is, the last case becomes the first case. When $$\Lambda=\textbf{diag}(\lambda)$$ is, the last case becomes the second case.

Definition at line 59 of file GaussianARDKernel.h.

Inheritance diagram for CGaussianARDKernel:
[legend]

## Public Types

typedef rxcpp::subjects::subject< ObservedValueSGSubject

typedef rxcpp::observable< ObservedValue, rxcpp::dynamic_observable< ObservedValue > > SGObservable

typedef rxcpp::subscriber< ObservedValue, rxcpp::observer< ObservedValue, void, void, void, void > > SGSubscriber

## Public Member Functions

CGaussianARDKernel ()

virtual ~CGaussianARDKernel ()

virtual EKernelType get_kernel_type ()

virtual const char * get_name () const

CGaussianARDKernel (int32_t size)

CGaussianARDKernel (CDotFeatures *l, CDotFeatures *r, int32_t size=10)

virtual bool init (CFeatures *l, CFeatures *r)

virtual SGMatrix< float64_tget_parameter_gradient (const TParameter *param, index_t index=-1)

virtual SGVector< float64_tget_parameter_gradient_diagonal (const TParameter *param, index_t index=-1)

virtual EFeatureClass get_feature_class ()

virtual EFeatureType get_feature_type ()

virtual SGMatrix< float64_tget_weights ()

virtual void set_scalar_weights (float64_t weight)

virtual void set_vector_weights (SGVector< float64_t > weights)

virtual void set_matrix_weights (SGMatrix< float64_t > weights)

float64_t kernel (int32_t idx_a, int32_t idx_b)

SGMatrix< float64_tget_kernel_matrix ()

template<class T >
SGMatrix< T > get_kernel_matrix ()

SGVector< float64_tget_kernel_diagonal (SGVector< float64_t > preallocated=SGVector< float64_t >())

virtual SGVector< float64_tget_kernel_col (int32_t j)

virtual SGVector< float64_tget_kernel_row (int32_t i)

void get_kernel_row (int32_t docnum, int32_t *active2dnum, float64_t *buffer, bool full_line=false)

virtual float64_t sum_symmetric_block (index_t block_begin, index_t block_size, bool no_diag=true)

virtual float64_t sum_block (index_t block_begin_row, index_t block_begin_col, index_t block_size_row, index_t block_size_col, bool no_diag=false)

virtual SGVector< float64_trow_wise_sum_symmetric_block (index_t block_begin, index_t block_size, bool no_diag=true)

virtual SGMatrix< float64_trow_wise_sum_squared_sum_symmetric_block (index_t block_begin, index_t block_size, bool no_diag=true)

virtual SGVector< float64_trow_col_wise_sum_block (index_t block_begin_row, index_t block_begin_col, index_t block_size_row, index_t block_size_col, bool no_diag=false)

virtual bool set_normalizer (CKernelNormalizer *normalizer)

virtual CKernelNormalizerget_normalizer ()

virtual bool init_normalizer ()

virtual void cleanup ()

void save (CFile *writer)

CFeaturesget_lhs ()

CFeaturesget_rhs ()

virtual int32_t get_num_vec_lhs ()

virtual int32_t get_num_vec_rhs ()

virtual bool has_features ()

bool get_lhs_equals_rhs ()

virtual void remove_lhs_and_rhs ()

virtual void remove_lhs ()

virtual void remove_rhs ()
takes all necessary steps if the rhs is removed from kernel More...

void set_cache_size (int32_t size)

int32_t get_cache_size ()

void cache_reset ()

int32_t get_max_elems_cache ()

int32_t get_activenum_cache ()

void cache_kernel_row (int32_t x)

void cache_multiple_kernel_rows (int32_t *key, int32_t varnum)

void kernel_cache_reset_lru ()

void kernel_cache_shrink (int32_t totdoc, int32_t num_shrink, int32_t *after)

void resize_kernel_cache (KERNELCACHE_IDX size, bool regression_hack=false)

void set_time (int32_t t)

int32_t kernel_cache_touch (int32_t cacheidx)

int32_t kernel_cache_check (int32_t cacheidx)

int32_t kernel_cache_space_available ()

void kernel_cache_init (int32_t size, bool regression_hack=false)

void kernel_cache_cleanup ()

void list_kernel ()

bool has_property (EKernelProperty p)

virtual void clear_normal ()

virtual void add_to_normal (int32_t vector_idx, float64_t weight)

EOptimizationType get_optimization_type ()

virtual void set_optimization_type (EOptimizationType t)

bool get_is_initialized ()

virtual bool init_optimization (int32_t count, int32_t *IDX, float64_t *weights)

virtual bool delete_optimization ()

bool init_optimization_svm (CSVM *svm)

virtual float64_t compute_optimized (int32_t vector_idx)

virtual void compute_batch (int32_t num_vec, int32_t *vec_idx, float64_t *target, int32_t num_suppvec, int32_t *IDX, float64_t *alphas, float64_t factor=1.0)

float64_t get_combined_kernel_weight ()

void set_combined_kernel_weight (float64_t nw)

virtual int32_t get_num_subkernels ()

virtual void compute_by_subkernel (int32_t vector_idx, float64_t *subkernel_contrib)

virtual const float64_tget_subkernel_weights (int32_t &num_weights)

virtual SGVector< float64_tget_subkernel_weights ()

virtual void set_subkernel_weights (SGVector< float64_t > weights)

int32_t ref ()

int32_t ref_count ()

int32_t unref ()

virtual CSGObjectshallow_copy () const

virtual CSGObjectdeep_copy () const

virtual bool is_generic (EPrimitiveType *generic) const

template<class T >
void set_generic ()

template<>
void set_generic ()

template<>
void set_generic ()

template<>
void set_generic ()

template<>
void set_generic ()

template<>
void set_generic ()

template<>
void set_generic ()

template<>
void set_generic ()

template<>
void set_generic ()

template<>
void set_generic ()

template<>
void set_generic ()

template<>
void set_generic ()

template<>
void set_generic ()

template<>
void set_generic ()

template<>
void set_generic ()

template<>
void set_generic ()

void unset_generic ()

virtual void print_serializable (const char *prefix="")

virtual bool save_serializable (CSerializableFile *file, const char *prefix="")

virtual bool load_serializable (CSerializableFile *file, const char *prefix="")

void set_global_io (SGIO *io)

SGIOget_global_io ()

void set_global_parallel (Parallel *parallel)

Parallelget_global_parallel ()

void set_global_version (Version *version)

Versionget_global_version ()

SGStringList< char > get_modelsel_names ()

void print_modsel_params ()

char * get_modsel_param_descr (const char *param_name)

index_t get_modsel_param_index (const char *param_name)

void build_gradient_parameter_dictionary (CMap< TParameter *, CSGObject *> *dict)

bool has (const std::string &name) const

template<typename T >
bool has (const Tag< T > &tag) const

template<typename T , typename U = void>
bool has (const std::string &name) const

template<typename T >
void set (const Tag< T > &_tag, const T &value)

template<typename T , typename U = void>
void set (const std::string &name, const T &value)

template<typename T >
get (const Tag< T > &_tag) const

template<typename T , typename U = void>
get (const std::string &name) const

SGObservableget_parameters_observable ()

void subscribe_to_parameters (ParameterObserverInterface *obs)

void list_observable_parameters ()

virtual void update_parameter_hash ()

virtual bool parameter_hash_changed ()

virtual bool equals (CSGObject *other, float64_t accuracy=0.0, bool tolerant=false)

virtual CSGObjectclone ()

## Static Public Member Functions

static CGaussianARDKernelobtain_from_generic (CKernel *kernel)

static CKernelobtain_from_generic (CSGObject *kernel)

## Public Attributes

SGIOio

Parallelparallel

Versionversion

Parameterm_parameters

Parameterm_model_selection_parameters

uint32_t m_hash

## Protected Member Functions

virtual float64_t distance (int32_t idx_a, int32_t idx_b)

virtual void precompute_squared ()

virtual SGVector< float64_tprecompute_squared_helper (CDotFeatures *df)

virtual float64_t compute_helper (SGVector< float64_t > avec, SGVector< float64_t >bvec)

virtual float64_t compute_gradient_helper (SGVector< float64_t > avec, SGVector< float64_t > bvec, float64_t scale, index_t index)

virtual float64_t get_parameter_gradient_helper (const TParameter *param, index_t index, int32_t idx_a, int32_t idx_b, SGVector< float64_t > avec, SGVector< float64_t > bvec)

virtual SGVector< float64_tget_feature_vector (int32_t idx, CFeatures *hs)

virtual float64_t compute (int32_t idx_a, int32_t idx_b)

virtual void set_weights (SGMatrix< float64_t > weights)

void lazy_update_weights ()

SGMatrix< float64_tget_weighted_vector (SGVector< float64_t > vec)

virtual SGMatrix< float64_tcompute_right_product (SGVector< float64_t >vec, float64_t &scalar_weight)

virtual void check_weight_gradient_index (index_t index)

void set_property (EKernelProperty p)

void unset_property (EKernelProperty p)

void set_is_initialized (bool p_init)

int32_t compute_row_start (int64_t offs, int32_t n, bool symmetric)

virtual void load_serializable_post () throw (ShogunException)

virtual void save_serializable_pre () throw (ShogunException)

virtual void save_serializable_post () throw (ShogunException)

virtual void register_params ()

virtual void load_serializable_pre () throw (ShogunException)

template<typename T >
void register_param (Tag< T > &_tag, const T &value)

template<typename T >
void register_param (const std::string &name, const T &value)

bool clone_parameters (CSGObject *other)

void observe (const ObservedValue value)

void register_observable_param (const std::string &name, const SG_OBS_VALUE_TYPE type, const std::string &description)

## Static Protected Member Functions

template<class T >
static void * get_kernel_matrix_helper (void *p)

## Protected Attributes

SGVector< float64_tm_sq_lhs

SGVector< float64_tm_sq_rhs

SGMatrix< float64_tm_weights_raw

SGVector< float64_tm_log_weights

index_t m_weights_rows

index_t m_weights_cols

EARDKernelType m_ARD_type

int32_t cache_size
cache_size in MB More...

KERNEL_CACHE kernel_cache
kernel cache More...

KERNELCACHE_ELEMkernel_matrix

CFeatureslhs
feature vectors to occur on left hand side More...

CFeaturesrhs
feature vectors to occur on right hand side More...

bool lhs_equals_rhs
lhs More...

int32_t num_lhs
number of feature vectors on left hand side More...

int32_t num_rhs
number of feature vectors on right hand side More...

float64_t combined_kernel_weight

bool optimization_initialized

EOptimizationType opt_type

uint64_t properties

CKernelNormalizernormalizer

## ◆ SGObservable

 inherited

Definition at line 130 of file SGObject.h.

## ◆ SGSubject

 inherited

Definition at line 127 of file SGObject.h.

## ◆ SGSubscriber

 typedef rxcpp::subscriber< ObservedValue, rxcpp::observer > SGSubscriber
inherited

Definition at line 133 of file SGObject.h.

## ◆ CGaussianARDKernel() [1/3]

 CGaussianARDKernel ( )

default constructor

Definition at line 19 of file GaussianARDKernel.cpp.

## ◆ ~CGaussianARDKernel()

 ~CGaussianARDKernel ( )
virtual

destructor

Definition at line 24 of file GaussianARDKernel.cpp.

## ◆ CGaussianARDKernel() [2/3]

 CGaussianARDKernel ( int32_t size )

constructor

Parameters
 size cache size width kernel width

Definition at line 60 of file GaussianARDKernel.cpp.

## ◆ CGaussianARDKernel() [3/3]

 CGaussianARDKernel ( CDotFeatures * l, CDotFeatures * r, int32_t size = 10 )

constructor

Parameters
 l features of left-hand side r features of right-hand side size cache size width kernel width

Definition at line 66 of file GaussianARDKernel.cpp.

## Member Function Documentation

 void add_to_normal ( int32_t vector_idx, float64_t weight )
virtualinherited

add vector*factor to 'virtual' normal vector

Parameters
 vector_idx index weight weight

Definition at line 830 of file Kernel.cpp.

 void build_gradient_parameter_dictionary ( CMap< TParameter *, CSGObject *> * dict )
inherited

Builds a dictionary of all parameters in SGObject as well of those of SGObjects that are parameters of this object. Dictionary maps parameters to the objects that own them.

Parameters
 dict dictionary of parameters to be built.

Definition at line 635 of file SGObject.cpp.

## ◆ cache_kernel_row()

 void cache_kernel_row ( int32_t x )
inherited

cache kernel row

Parameters
 x x

Definition at line 301 of file Kernel.cpp.

## ◆ cache_multiple_kernel_rows()

 void cache_multiple_kernel_rows ( int32_t * key, int32_t varnum )
inherited

cache multiple kernel rows

Parameters
 key key varnum

Definition at line 375 of file Kernel.cpp.

## ◆ cache_reset()

 void cache_reset ( )
inherited

cache reset

Definition at line 602 of file kernel/Kernel.h.

 void check_weight_gradient_index ( index_t index )
protectedvirtualinherited

check whether index of gradient wrt weights is valid

Parameters
 index the index of the element if parameter is a vector or matrix if the parameter is a matrix, index is the linearized index of the matrix (column-major)

Definition at line 258 of file ExponentialARDKernel.cpp.

## ◆ cleanup()

 void cleanup ( )
virtualinherited

clean up your kernel

base method only removes lhs and rhs overload to add further cleanup but make sure CKernel::cleanup() is called

Definition at line 172 of file Kernel.cpp.

## ◆ clear_normal()

 void clear_normal ( )
virtualinherited

for optimizable kernels, i.e. kernels where the weight vector can be computed explicitly (if it fits into memory)

Definition at line 835 of file Kernel.cpp.

## ◆ clone()

 CSGObject * clone ( )
virtualinherited

Creates a clone of the current object. This is done via recursively traversing all parameters, which corresponds to a deep copy. Calling equals on the cloned object always returns true although none of the memory of both objects overlaps.

Returns
an identical copy of the given object, which is disjoint in memory. NULL if the clone fails. Note that the returned object is SG_REF'ed

Definition at line 734 of file SGObject.cpp.

## ◆ clone_parameters()

 bool clone_parameters ( CSGObject * other )
protectedinherited

Definition at line 759 of file SGObject.cpp.

## ◆ compute()

 virtual float64_t compute ( int32_t idx_a, int32_t idx_b )
protectedvirtualinherited

compute kernel function for features a and b idx_{a,b} denote the index of the feature vectors in the corresponding feature object

Parameters
 idx_a index a idx_b index b
Returns
computed kernel function at indices a,b kernel(idx_a, idx_b)=exp(-distance(idx_a, idx_b))

Reimplemented from CDotKernel.

Definition at line 148 of file ExponentialARDKernel.h.

## ◆ compute_batch()

 void compute_batch ( int32_t num_vec, int32_t * vec_idx, float64_t * target, int32_t num_suppvec, int32_t * IDX, float64_t * alphas, float64_t factor = 1.0 )
virtualinherited

computes output for a batch of examples in an optimized fashion (favorable if kernel supports it, i.e. has KP_BATCHEVALUATION. to the outputvector target (of length num_vec elements) the output for the examples enumerated in vec_idx are added. therefore make sure that it is initialized with ZERO. the following num_suppvec, IDX, alphas arguments are the number of support vectors, their indices and weights

Reimplemented in CCombinedKernel, CWeightedDegreeStringKernel, and CWeightedDegreePositionStringKernel.

Definition at line 823 of file Kernel.cpp.

## ◆ compute_by_subkernel()

 void compute_by_subkernel ( int32_t vector_idx, float64_t * subkernel_contrib )
virtualinherited

compute by subkernel

Parameters
 vector_idx index subkernel_contrib subkernel contribution

Reimplemented in CCombinedKernel, CWeightedDegreeStringKernel, and CWeightedDegreePositionStringKernel.

Definition at line 845 of file Kernel.cpp.

 float64_t compute_gradient_helper ( SGVector< float64_t > avec, SGVector< float64_t > bvec, float64_t scale, index_t index )
protectedvirtual

helper function used to compute derivative with respect to weights

Parameters
 avec left feature vector bvec right feature vector scale scaling value index the linearized index of a weight matrix (column-major)
Returns
gradient with respect to parameter

Definition at line 140 of file GaussianARDKernel.cpp.

## ◆ compute_helper()

 float64_t compute_helper ( SGVector< float64_t > avec, SGVector< float64_t > bvec )
protectedvirtual

helper function used to compute kernel function for features avec and bvec

Parameters
 avec left feature vector bvec right feature vector
Returns
computed kernel value

Definition at line 118 of file GaussianARDKernel.cpp.

## ◆ compute_optimized()

 float64_t compute_optimized ( int32_t vector_idx )
virtualinherited

compute optimized

Parameters
 vector_idx index to compute
Returns
optimized value at given index

Definition at line 817 of file Kernel.cpp.

## ◆ compute_right_product()

 SGMatrix< float64_t > compute_right_product ( SGVector< float64_t > vec, float64_t & scalar_weight )
protectedvirtualinherited

helper function used to compute kernel value

The function is to compute for scalar weights: $$V=\textbf{vec}$$ and scalar_weight= $$\lambda$$ for vector weights: $$V=\lambda .* \textbf{vec}$$ for matrix weights: $$V=\Lambda^T * \textbf{vec}$$

Parameters
 vec feature vector scalar_weight set the scaling value, which is used in the scalar case (first case).
Returns
the result of $$V$$

Definition at line 239 of file ExponentialARDKernel.cpp.

## ◆ compute_row_start()

 int32_t compute_row_start ( int64_t offs, int32_t n, bool symmetric )
protectedinherited

compute row start offset for parallel kernel matrix computation

Parameters
 offs offset n number of columns symmetric whether matrix is symmetric

Definition at line 919 of file kernel/Kernel.h.

## ◆ deep_copy()

 CSGObject * deep_copy ( ) const
virtualinherited

A deep copy. All the instance variables will also be copied.

Definition at line 232 of file SGObject.cpp.

## ◆ delete_optimization()

 bool delete_optimization ( )
virtualinherited

delete optimization

Returns
if deleting was successful

Definition at line 811 of file Kernel.cpp.

## ◆ distance()

 float64_t distance ( int32_t idx_a, int32_t idx_b )
protectedvirtual

compute the distance between features a and b idx_{a,b} denote the index of the feature vectors in the corresponding feature object

Parameters
 idx_a index a idx_b index b
Returns
computed the distance

Note that in GaussianARDKernel, kernel(idx_a, idx_b)=exp(-distance(idx_a, idx_b))

Implements CExponentialARDKernel.

Definition at line 36 of file GaussianARDKernel.cpp.

## ◆ equals()

 bool equals ( CSGObject * other, float64_t accuracy = 0.0, bool tolerant = false )
virtualinherited

Recursively compares the current SGObject to another one. Compares all registered numerical parameters, recursion upon complex (SGObject) parameters. Does not compare pointers!

May be overwritten but please do with care! Should not be necessary in most cases.

Parameters
 other object to compare with accuracy accuracy to use for comparison (optional) tolerant allows linient check on float equality (within accuracy)
Returns
true if all parameters were equal, false if not

Definition at line 656 of file SGObject.cpp.

## ◆ get() [1/2]

 T get ( const Tag< T > & _tag ) const
inherited

Getter for a class parameter, identified by a Tag. Throws an exception if the class does not have such a parameter.

Parameters
 _tag name and type information of parameter
Returns
value of the parameter identified by the input tag

Definition at line 381 of file SGObject.h.

## ◆ get() [2/2]

 T get ( const std::string & name ) const
inherited

Getter for a class parameter, identified by a name. Throws an exception if the class does not have such a parameter.

Parameters
 name name of the parameter
Returns
value of the parameter corresponding to the input name and type

Definition at line 404 of file SGObject.h.

## ◆ get_activenum_cache()

 int32_t get_activenum_cache ( )
inherited

get activenum cache

Returns
activecnum cache

Definition at line 614 of file kernel/Kernel.h.

## ◆ get_cache_size()

 int32_t get_cache_size ( )
inherited

return the size of the kernel cache

Returns
size of kernel cache

Definition at line 598 of file kernel/Kernel.h.

## ◆ get_combined_kernel_weight()

 float64_t get_combined_kernel_weight ( )
inherited

get combined kernel weight

Returns
combined kernel weight

Definition at line 802 of file kernel/Kernel.h.

## ◆ get_feature_class()

 virtual EFeatureClass get_feature_class ( )
virtualinherited

return feature class the kernel can deal with

Returns
feature class DENSE

Reimplemented from CDotKernel.

Definition at line 93 of file ExponentialARDKernel.h.

## ◆ get_feature_type()

 virtual EFeatureType get_feature_type ( )
virtualinherited

return feature type the kernel can deal with

Returns
float64_t feature type

Reimplemented from CDotKernel.

Definition at line 99 of file ExponentialARDKernel.h.

## ◆ get_feature_vector()

 SGVector< float64_t > get_feature_vector ( int32_t idx, CFeatures * hs )
protectedvirtualinherited

get features vector given idx

Parameters
 idx index of CFeatures hs features
Returns
the features vector

Definition at line 53 of file ExponentialARDKernel.cpp.

## ◆ get_global_io()

 SGIO * get_global_io ( )
inherited

get the io object

Returns
io object

Definition at line 269 of file SGObject.cpp.

## ◆ get_global_parallel()

 Parallel * get_global_parallel ( )
inherited

get the parallel object

Returns
parallel object

Definition at line 311 of file SGObject.cpp.

## ◆ get_global_version()

 Version * get_global_version ( )
inherited

get the version object

Returns
version object

Definition at line 324 of file SGObject.cpp.

## ◆ get_is_initialized()

 bool get_is_initialized ( )
inherited

check if optimization is initialized

Returns
if optimization is initialized

Definition at line 753 of file kernel/Kernel.h.

## ◆ get_kernel_col()

 virtual SGVector get_kernel_col ( int32_t j )
virtualinherited

get column j

Returns
the jth column of the kernel matrix

Definition at line 262 of file kernel/Kernel.h.

## ◆ get_kernel_diagonal()

 SGVector get_kernel_diagonal ( SGVector< float64_t > preallocated = SGVector() )
inherited
Returns
Vector with diagonal elements of the kernel matrix. Note that left- and right-handside features must be set and of equal size
Parameters
 preallocated vector with space for results

Definition at line 230 of file kernel/Kernel.h.

## ◆ get_kernel_matrix() [1/2]

 template SGMatrix< float32_t > get_kernel_matrix< float32_t > ( )
inherited

get kernel matrix

Returns
computed kernel matrix (needs to be cleaned up)

Definition at line 219 of file kernel/Kernel.h.

## ◆ get_kernel_matrix() [2/2]

 SGMatrix< T > get_kernel_matrix ( )
inherited

get kernel matrix (templated)

Returns
the kernel matrix

Definition at line 1317 of file Kernel.cpp.

## ◆ get_kernel_matrix_helper()

 template void * get_kernel_matrix_helper< float32_t > ( void * p )
staticprotectedinherited

helper for computing the kernel matrix in a parallel way

Parameters

Definition at line 1266 of file Kernel.cpp.

## ◆ get_kernel_row() [1/2]

 virtual SGVector get_kernel_row ( int32_t i )
virtualinherited

get row i

Returns
the ith row of the kernel matrix

Definition at line 279 of file kernel/Kernel.h.

## ◆ get_kernel_row() [2/2]

 void get_kernel_row ( int32_t docnum, int32_t * active2dnum, float64_t * buffer, bool full_line = false )
inherited

get kernel row

Parameters
 docnum docnum active2dnum active2dnum buffer buffer full_line full line

Definition at line 237 of file Kernel.cpp.

## ◆ get_kernel_type()

 virtual EKernelType get_kernel_type ( )
virtual

return what type of kernel we are

Returns
kernel type GAUSSIANARD

Reimplemented from CExponentialARDKernel.

Reimplemented in CGaussianARDSparseKernel.

Definition at line 72 of file GaussianARDKernel.h.

## ◆ get_lhs()

 CFeatures* get_lhs ( )
inherited

get left-hand side of features used in kernel

Returns
features of left-hand side

Definition at line 504 of file kernel/Kernel.h.

## ◆ get_lhs_equals_rhs()

 bool get_lhs_equals_rhs ( )
inherited

test whether features on lhs and rhs are the same

Returns
true if features are the same

Definition at line 543 of file kernel/Kernel.h.

## ◆ get_max_elems_cache()

 int32_t get_max_elems_cache ( )
inherited

get maximum elements in cache

Returns
maximum elements in cache

Definition at line 608 of file kernel/Kernel.h.

## ◆ get_modelsel_names()

 SGStringList< char > get_modelsel_names ( )
inherited
Returns
vector of names of all parameters which are registered for model selection

Definition at line 536 of file SGObject.cpp.

## ◆ get_modsel_param_descr()

 char * get_modsel_param_descr ( const char * param_name )
inherited

Returns description of a given parameter string, if it exists. SG_ERROR otherwise

Parameters
 param_name name of the parameter
Returns
description of the parameter

Definition at line 560 of file SGObject.cpp.

## ◆ get_modsel_param_index()

 index_t get_modsel_param_index ( const char * param_name )
inherited

Returns index of model selection parameter with provided index

Parameters
 param_name name of model selection parameter
Returns
index of model selection parameter with provided name, -1 if there is no such

Definition at line 573 of file SGObject.cpp.

## ◆ get_name()

 virtual const char* get_name ( ) const
virtual

return the kernel's name

Returns
name GaussianARDKernel

Reimplemented from CExponentialARDKernel.

Reimplemented in CGaussianARDSparseKernel.

Definition at line 78 of file GaussianARDKernel.h.

## ◆ get_normalizer()

 CKernelNormalizer * get_normalizer ( )
virtualinherited

obtain the current kernel normalizer

Returns
the kernel normalizer

Definition at line 161 of file Kernel.cpp.

## ◆ get_num_subkernels()

 int32_t get_num_subkernels ( )
virtualinherited

get number of subkernels

Returns
number of subkernels

Reimplemented in CWeightedDegreeStringKernel, CWeightedDegreePositionStringKernel, CCombinedKernel, and CProductKernel.

Definition at line 840 of file Kernel.cpp.

## ◆ get_num_vec_lhs()

 virtual int32_t get_num_vec_lhs ( )
virtualinherited

get number of vectors of lhs features

Returns
number of vectors of left-hand side

Reimplemented in CCustomKernel.

Definition at line 516 of file kernel/Kernel.h.

## ◆ get_num_vec_rhs()

 virtual int32_t get_num_vec_rhs ( )
virtualinherited

get number of vectors of rhs features

Returns
number of vectors of right-hand side

Reimplemented in CCustomKernel.

Definition at line 525 of file kernel/Kernel.h.

## ◆ get_optimization_type()

 EOptimizationType get_optimization_type ( )
inherited

get optimization type

Returns
optimization type

Definition at line 741 of file kernel/Kernel.h.

 SGMatrix< float64_t > get_parameter_gradient ( const TParameter * param, index_t index = -1 )
virtual

return derivative with respect to specified parameter

Parameters
 param the parameter index the index of the element if parameter is a vector or matrix if the parameter is a matrix, index is the linearized index of the matrix (column-major)
Returns
gradient with respect to parameter

Implements CExponentialARDKernel.

Reimplemented in CGaussianARDSparseKernel.

Definition at line 268 of file GaussianARDKernel.cpp.

 SGVector< float64_t > get_parameter_gradient_diagonal ( const TParameter * param, index_t index = -1 )
virtual

return diagonal part of derivative with respect to specified parameter

Parameters
 param the parameter index the index of the element if parameter is a vector
Returns
diagonal part of gradient with respect to parameter

Reimplemented from CKernel.

Reimplemented in CGaussianARDSparseKernel.

Definition at line 202 of file GaussianARDKernel.cpp.

 float64_t get_parameter_gradient_helper ( const TParameter * param, index_t index, int32_t idx_a, int32_t idx_b, SGVector< float64_t > avec, SGVector< float64_t > bvec )
protectedvirtual

helper function to compute derivative with respect to specified parameter

Parameters
 param the parameter index the index of the element if parameter is a vector or matrix if the parameter is a matrix, index is the linearized index of the matrix (column-major) idx_a the row index of the gradient matrix idx_b the column index of the gradient matrix avec feature vector corresponding to idx_a bvec feature vector corresponding to idx_b
Returns
gradient at row idx_a and column idx_b with respect to parameter

Definition at line 249 of file GaussianARDKernel.cpp.

## ◆ get_parameters_observable()

 SGObservable* get_parameters_observable ( )
inherited

Get parameters observable

Returns
RxCpp observable

Definition at line 415 of file SGObject.h.

## ◆ get_rhs()

 CFeatures* get_rhs ( )
inherited

get right-hand side of features used in kernel

Returns
features of right-hand side

Definition at line 510 of file kernel/Kernel.h.

## ◆ get_subkernel_weights() [1/2]

 const float64_t * get_subkernel_weights ( int32_t & num_weights )
virtualinherited

get subkernel weights

Parameters
 num_weights number of weights will be stored here
Returns
subkernel weights

Reimplemented in CCombinedKernel, CWeightedDegreeStringKernel, and CWeightedDegreePositionStringKernel.

Definition at line 851 of file Kernel.cpp.

## ◆ get_subkernel_weights() [2/2]

 SGVector< float64_t > get_subkernel_weights ( )
virtualinherited

get subkernel weights (swig compatible)

Returns
subkernel weights

Reimplemented in CCombinedKernel.

Definition at line 857 of file Kernel.cpp.

## ◆ get_weighted_vector()

 SGMatrix< float64_t > get_weighted_vector ( SGVector< float64_t > vec )
protectedinherited

convert the m_log_weights in vector format to the matrix format in standard domain

Parameters
 vec weights in log domain in vector layout
Returns
weights in standard domain in matrix layout

Definition at line 207 of file ExponentialARDKernel.cpp.

## ◆ get_weights()

 SGMatrix< float64_t > get_weights ( )
virtualinherited

return current feature/dimension weights in matrix form Note that a vector weights is considered as a row vector (one-by-p matrix) where p is the number of features Note that a scalar weights is considered as a one-by-one matrix

Returns
weights in matrix form

Definition at line 114 of file ExponentialARDKernel.cpp.

## ◆ has() [1/3]

 bool has ( const std::string & name ) const
inherited

Checks if object has a class parameter identified by a name.

Parameters
 name name of the parameter
Returns
true if the parameter exists with the input name

Definition at line 304 of file SGObject.h.

## ◆ has() [2/3]

 bool has ( const Tag< T > & tag ) const
inherited

Checks if object has a class parameter identified by a Tag.

Parameters
 tag tag of the parameter containing name and type information
Returns
true if the parameter exists with the input tag

Definition at line 315 of file SGObject.h.

## ◆ has() [3/3]

 bool has ( const std::string & name ) const
inherited

Checks if a type exists for a class parameter identified by a name.

Parameters
 name name of the parameter
Returns
true if the parameter exists with the input name and type

Definition at line 326 of file SGObject.h.

## ◆ has_features()

 virtual bool has_features ( )
virtualinherited

test whether features have been assigned to lhs and rhs

Returns
true if features are assigned

Reimplemented in CCustomKernel, CCombinedKernel, and CProductKernel.

Definition at line 534 of file kernel/Kernel.h.

## ◆ has_property()

 bool has_property ( EKernelProperty p )
inherited

check if kernel has given property

Parameters
 p kernel property
Returns
if kernel has given property

Definition at line 723 of file kernel/Kernel.h.

## ◆ init()

 bool init ( CFeatures * l, CFeatures * r )
virtual

initialize kernel

Parameters
 l features of left-hand side r features of right-hand side
Returns
if initializing was successful

Reimplemented from CExponentialARDKernel.

Definition at line 73 of file GaussianARDKernel.cpp.

## ◆ init_normalizer()

 bool init_normalizer ( )
virtualinherited

initialize the current kernel normalizer

Returns
if init was successful

Definition at line 167 of file Kernel.cpp.

## ◆ init_optimization()

 bool init_optimization ( int32_t count, int32_t * IDX, float64_t * weights )
virtualinherited

initialize optimization

Parameters
 count count IDX index weights weights
Returns
if initializing was successful

Definition at line 804 of file Kernel.cpp.

## ◆ init_optimization_svm()

 bool init_optimization_svm ( CSVM * svm )
inherited

initialize optimization

Parameters
 svm svm model
Returns
if initializing was successful

Definition at line 887 of file Kernel.cpp.

## ◆ is_generic()

 bool is_generic ( EPrimitiveType * generic ) const
virtualinherited

If the SGSerializable is a class template then TRUE will be returned and GENERIC is set to the type of the generic.

Parameters
 generic set to the type of the generic if returning TRUE
Returns
TRUE if a class template.

Definition at line 330 of file SGObject.cpp.

## ◆ kernel()

 float64_t kernel ( int32_t idx_a, int32_t idx_b )
inherited

get kernel function for lhs feature vector a and rhs feature vector b

Parameters
 idx_a index of feature vector a idx_b index of feature vector b
Returns
computed kernel function

Definition at line 206 of file kernel/Kernel.h.

## ◆ kernel_cache_check()

 int32_t kernel_cache_check ( int32_t cacheidx )
inherited

check if row at given index is cached

Parameters
 cacheidx index in cache
Returns
if row at given index is cached

Definition at line 689 of file kernel/Kernel.h.

## ◆ kernel_cache_cleanup()

 void kernel_cache_cleanup ( )
inherited

cleanup kernel cache

Definition at line 543 of file Kernel.cpp.

## ◆ kernel_cache_init()

 void kernel_cache_init ( int32_t size, bool regression_hack = false )
inherited

initialize kernel cache

Parameters
 size size to initialize to regression_hack if hack for regression shall be applied

Definition at line 180 of file Kernel.cpp.

## ◆ kernel_cache_reset_lru()

 void kernel_cache_reset_lru ( )
inherited

kernel cache reset lru

Definition at line 530 of file Kernel.cpp.

## ◆ kernel_cache_shrink()

 void kernel_cache_shrink ( int32_t totdoc, int32_t num_shrink, int32_t * after )
inherited

kernel cache shrink

Parameters
 totdoc totdoc num_shrink number of shrink after after

Definition at line 471 of file Kernel.cpp.

## ◆ kernel_cache_space_available()

 int32_t kernel_cache_space_available ( )
inherited

check if there is room for one more row in kernel cache

Returns
if there is room for one more row in kernel cache

Definition at line 698 of file kernel/Kernel.h.

## ◆ kernel_cache_touch()

 int32_t kernel_cache_touch ( int32_t cacheidx )
inherited

update lru time of item at given index to avoid removal from cache

Parameters
 cacheidx index in cache
Returns
if updating was successful

Definition at line 674 of file kernel/Kernel.h.

## ◆ lazy_update_weights()

 void lazy_update_weights ( )
protectedinherited

convert the weights in log domain to standard domain when get_weights() is called

Definition at line 84 of file ExponentialARDKernel.cpp.

## ◆ list_kernel()

 void list_kernel ( )
inherited

list kernel

Definition at line 684 of file Kernel.cpp.

## ◆ list_observable_parameters()

 void list_observable_parameters ( )
inherited

Print to stdout a list of observable parameters

Definition at line 878 of file SGObject.cpp.

 void load ( CFile * loader )
inherited

load the kernel matrix

Parameters
 loader File object via which to load data

Definition at line 622 of file Kernel.cpp.

 bool load_serializable ( CSerializableFile * file, const char * prefix = "" )
virtualinherited

Load this object from file. If it will fail (returning FALSE) then this object will contain inconsistent data and should not be used!

Parameters
 file where to load from prefix prefix for members
Returns
TRUE if done, otherwise FALSE

Definition at line 403 of file SGObject.cpp.

 void load_serializable_post ( ) throw ( ShogunException )
protectedvirtualinherited

Can (optionally) be overridden to post-initialize some member variables which are not PARAMETER::ADD'ed. Make sure that at first the overridden method BASE_CLASS::LOAD_SERIALIZABLE_POST is called.

Exceptions
 ShogunException Will be thrown if an error occurres.

Reimplemented from CSGObject.

Definition at line 905 of file Kernel.cpp.

 void load_serializable_pre ( ) throw ( ShogunException )
protectedvirtualinherited

Can (optionally) be overridden to pre-initialize some member variables which are not PARAMETER::ADD'ed. Make sure that at first the overridden method BASE_CLASS::LOAD_SERIALIZABLE_PRE is called.

Exceptions
 ShogunException will be thrown if an error occurs.

Definition at line 455 of file SGObject.cpp.

## ◆ observe()

 void observe ( const ObservedValue value )
protectedinherited

Observe a parameter value and emit them to observer.

Parameters
 value Observed parameter's value

Definition at line 828 of file SGObject.cpp.

## ◆ obtain_from_generic() [1/2]

 CGaussianARDKernel * obtain_from_generic ( CKernel * kernel )
static
Parameters
 kernel is casted to CGaussianARDKernel, error if not possible is SG_REF'ed
Returns
casted CGaussianARDKernel object

Definition at line 106 of file GaussianARDKernel.cpp.

## ◆ obtain_from_generic() [2/2]

 CKernel * obtain_from_generic ( CSGObject * kernel )
staticinherited

Obtains a kernel from a generic SGObject with error checking. Note that if passing NULL, result will be NULL

Parameters
 kernel Object to cast to CKernel, is *not* SG_REFed
Returns
object casted to CKernel, NULL if not possible

Definition at line 873 of file Kernel.cpp.

## ◆ parameter_hash_changed()

 bool parameter_hash_changed ( )
virtualinherited
Returns
whether parameter combination has changed since last update

Definition at line 296 of file SGObject.cpp.

## ◆ precompute_squared()

 void precompute_squared ( )
protectedvirtual

helper function to compute quadratic terms in (a-b)^2 (== a^2+b^2-2ab)

Definition at line 93 of file GaussianARDKernel.cpp.

## ◆ precompute_squared_helper()

 SGVector< float64_t > precompute_squared_helper ( CDotFeatures * df )
protectedvirtual

helper function to compute quadratic terms in (a-b)^2 (== a^2+b^2-2ab)

Parameters
 buf buffer to store squared terms (will be allocated) df dot feature object based on which k(i,i) is computed

Definition at line 83 of file GaussianARDKernel.cpp.

## ◆ print_modsel_params()

 void print_modsel_params ( )
inherited

prints all parameter registered for model selection and their type

Definition at line 512 of file SGObject.cpp.

## ◆ print_serializable()

 void print_serializable ( const char * prefix = "" )
virtualinherited

prints registered parameters out

Parameters
 prefix prefix for members

Definition at line 342 of file SGObject.cpp.

## ◆ ref()

 int32_t ref ( )
inherited

increase reference counter

Returns
reference count

Definition at line 186 of file SGObject.cpp.

## ◆ ref_count()

 int32_t ref_count ( )
inherited

display reference counter

Returns
reference count

Definition at line 193 of file SGObject.cpp.

## ◆ register_observable_param()

 void register_observable_param ( const std::string & name, const SG_OBS_VALUE_TYPE type, const std::string & description )
protectedinherited

Register which params this object can emit.

Parameters
 name the param name type the param type description a user oriented description

Definition at line 871 of file SGObject.cpp.

## ◆ register_param() [1/2]

 void register_param ( Tag< T > & _tag, const T & value )
protectedinherited

Registers a class parameter which is identified by a tag. This enables the parameter to be modified by set() and retrieved by get(). Parameters can be registered in the constructor of the class.

Parameters
 _tag name and type information of parameter value value of the parameter

Definition at line 472 of file SGObject.h.

## ◆ register_param() [2/2]

 void register_param ( const std::string & name, const T & value )
protectedinherited

Registers a class parameter which is identified by a name. This enables the parameter to be modified by set() and retrieved by get(). Parameters can be registered in the constructor of the class.

Parameters
 name name of the parameter value value of the parameter along with type information

Definition at line 485 of file SGObject.h.

## ◆ register_params()

 void register_params ( )
protectedvirtualinherited

Separate the function of parameter registration This can be the first stage of a *general* framework for cross-validation or other parameter-based operations

Definition at line 928 of file Kernel.cpp.

## ◆ remove_lhs()

 void remove_lhs ( )
virtualinherited

remove lhs from kernel

Definition at line 655 of file Kernel.cpp.

## ◆ remove_lhs_and_rhs()

 void remove_lhs_and_rhs ( )
virtualinherited

remove lhs and rhs from kernel

Reimplemented in CCombinedKernel, and CProductKernel.

Definition at line 636 of file Kernel.cpp.

## ◆ remove_rhs()

 void remove_rhs ( )
virtualinherited

takes all necessary steps if the rhs is removed from kernel

remove rhs from kernel

Reimplemented in CCombinedKernel, CCommUlongStringKernel, and CProductKernel.

Definition at line 669 of file Kernel.cpp.

## ◆ resize_kernel_cache()

 void resize_kernel_cache ( KERNELCACHE_IDX size, bool regression_hack = false )
inherited

resize kernel cache

Parameters
 size new size regression_hack hack for regression

Definition at line 84 of file Kernel.cpp.

## ◆ row_col_wise_sum_block()

 SGVector< float64_t > row_col_wise_sum_block ( index_t block_begin_row, index_t block_begin_col, index_t block_size_row, index_t block_size_col, bool no_diag = false )
virtualinherited

Computes row-wise/col-wise sum of kernel values. This method is useful while computing statistical estimation of mean/variance over kernel values but the kernel matrix is too huge to be fit inside memory.

Parameters
 block_begin_row the row index at which the block starts block_begin_col the col index at which the block starts block_size_row the number of rows in the block block_size_col the number of cols in the block

For Example, block_begin_row 0, block_begin_col 4 and block_size_row 5, block_size_col 6 represents the block that starts at index (0,4) in the kernel matrix and goes upto (0+5-1,4+6-1) i.e. (4,9) both inclusive

Parameters
 no_diag if true (default is false), the diagonal elements are excluded from the row/col-wise sum, provided that block_size_row and block_size_col are same (i.e. the block is square). Otherwise, these are always added
Returns
a vector whose first block_size_row entries contain row-wise sum of kernel values computed as

$v[i]=\sum_{j}k(i+\text{block-begin-row}, j+\text{block-begin-col})$

and rest block_size_col entries col-wise sum of kernel values computed as

$v[\text{block-size-row}+j]=\sum_{i}k(i+\text{block-begin-row}, j+\text{block-begin-col})$

where $$i\in[0,\text{block-size-row}-1]$$ and $$j\in[0,\text{block-size-col}-1]$$

Reimplemented in CCustomKernel.

Definition at line 1212 of file Kernel.cpp.

## ◆ row_wise_sum_squared_sum_symmetric_block()

 SGMatrix< float64_t > row_wise_sum_squared_sum_symmetric_block ( index_t block_begin, index_t block_size, bool no_diag = true )
virtualinherited

Computes row-wise/col-wise sum and squared sum of kernel values from a symmetric part of the kernel matrix that always is supposed to contain the main upper diagonal. This method is useful while computing statistical estimation of mean/variance over kernel values but the kernel matrix is too huge to be fit inside memory.

Parameters
 block_begin the row and col index at which the block starts block_size the number of rows and cols in the block

For Example, block_begin 4 and block_size 5 represents the block that starts at index (4,4) in the kernel matrix and goes upto (4+5-1,4+5-1) i.e. (8,8) both inclusive

Parameters
 no_diag if true (default), the diagonal elements are excluded from the row/col-wise sum
Returns
a matrix whose first column contains the row-wise sum of kernel values computed as

$v_0[i]=\sum_{j}k(i+\text{block-begin}, j+\text{block-begin})$

and second column contains the row-wise sum of squared kernel values

$v_1[i]=\sum_{j}^k^2(i+\text{block-begin}, j+\text{block-begin})$

where $$i,j\in[0,\text{block-size}-1]$$

Reimplemented in CCustomKernel.

Definition at line 1153 of file Kernel.cpp.

## ◆ row_wise_sum_symmetric_block()

 SGVector< float64_t > row_wise_sum_symmetric_block ( index_t block_begin, index_t block_size, bool no_diag = true )
virtualinherited

Computes row-wise/col-wise sum from a symmetric part of the kernel matrix that always is supposed to contain the main upper diagonal. This method is useful while computing statistical estimation of mean/variance over kernel values but the kernel matrix is too huge to be fit inside memory.

Parameters
 block_begin the row and col index at which the block starts block_size the number of rows and cols in the block

For Example, block_begin 4 and block_size 5 represents the block that starts at index (4,4) in the kernel matrix and goes upto (4+5-1,4+5-1) i.e. (8,8) both inclusive

Parameters
 no_diag if true (default), the diagonal elements are excluded from the row/col-wise sum
Returns
vector containing row-wise sum computed as

$v[i]=\sum_{j}k(i+\text{block-begin}, j+\text{block-begin})$

where $$i,j\in[0,\text{block-size}-1]$$

Reimplemented in CCustomKernel.

Definition at line 1099 of file Kernel.cpp.

## ◆ save()

 void save ( CFile * writer )
inherited

save kernel matrix

Parameters
 writer File object via which to save data

Definition at line 628 of file Kernel.cpp.

## ◆ save_serializable()

 bool save_serializable ( CSerializableFile * file, const char * prefix = "" )
virtualinherited

Save this object to file.

Parameters
 file where to save the object; will be closed during returning if PREFIX is an empty string. prefix prefix for members
Returns
TRUE if done, otherwise FALSE

Definition at line 348 of file SGObject.cpp.

## ◆ save_serializable_post()

 void save_serializable_post ( ) throw ( ShogunException )
protectedvirtualinherited

Can (optionally) be overridden to post-initialize some member variables which are not PARAMETER::ADD'ed. Make sure that at first the overridden method BASE_CLASS::SAVE_SERIALIZABLE_POST is called.

Exceptions
 ShogunException Will be thrown if an error occurres.

Reimplemented from CSGObject.

Definition at line 920 of file Kernel.cpp.

## ◆ save_serializable_pre()

 void save_serializable_pre ( ) throw ( ShogunException )
protectedvirtualinherited

Can (optionally) be overridden to pre-initialize some member variables which are not PARAMETER::ADD'ed. Make sure that at first the overridden method BASE_CLASS::SAVE_SERIALIZABLE_PRE is called.

Exceptions
 ShogunException Will be thrown if an error occurres.

Reimplemented from CSGObject.

Definition at line 912 of file Kernel.cpp.

## ◆ set() [1/2]

 void set ( const Tag< T > & _tag, const T & value )
inherited

Setter for a class parameter, identified by a Tag. Throws an exception if the class does not have such a parameter.

Parameters
 _tag name and type information of parameter value value of the parameter

Definition at line 342 of file SGObject.h.

## ◆ set() [2/2]

 void set ( const std::string & name, const T & value )
inherited

Setter for a class parameter, identified by a name. Throws an exception if the class does not have such a parameter.

Parameters
 name name of the parameter value value of the parameter along with type information

Definition at line 368 of file SGObject.h.

## ◆ set_cache_size()

 void set_cache_size ( int32_t size )
inherited

set the size of the kernel cache

Parameters
 size of kernel cache

Definition at line 586 of file kernel/Kernel.h.

## ◆ set_combined_kernel_weight()

 void set_combined_kernel_weight ( float64_t nw )
inherited

set combined kernel weight

Parameters
 nw new combined kernel weight

Definition at line 808 of file kernel/Kernel.h.

## ◆ set_generic() [1/16]

 void set_generic ( )
inherited

Definition at line 73 of file SGObject.cpp.

## ◆ set_generic() [2/16]

 void set_generic ( )
inherited

Definition at line 78 of file SGObject.cpp.

## ◆ set_generic() [3/16]

 void set_generic ( )
inherited

Definition at line 83 of file SGObject.cpp.

## ◆ set_generic() [4/16]

 void set_generic ( )
inherited

Definition at line 88 of file SGObject.cpp.

## ◆ set_generic() [5/16]

 void set_generic ( )
inherited

Definition at line 93 of file SGObject.cpp.

## ◆ set_generic() [6/16]

 void set_generic ( )
inherited

Definition at line 98 of file SGObject.cpp.

## ◆ set_generic() [7/16]

 void set_generic ( )
inherited

Definition at line 103 of file SGObject.cpp.

## ◆ set_generic() [8/16]

 void set_generic ( )
inherited

Definition at line 108 of file SGObject.cpp.

## ◆ set_generic() [9/16]

 void set_generic ( )
inherited

Definition at line 113 of file SGObject.cpp.

## ◆ set_generic() [10/16]

 void set_generic ( )
inherited

Definition at line 118 of file SGObject.cpp.

## ◆ set_generic() [11/16]

 void set_generic ( )
inherited

Definition at line 123 of file SGObject.cpp.

## ◆ set_generic() [12/16]

 void set_generic ( )
inherited

Definition at line 128 of file SGObject.cpp.

## ◆ set_generic() [13/16]

 void set_generic ( )
inherited

Definition at line 133 of file SGObject.cpp.

## ◆ set_generic() [14/16]

 void set_generic ( )
inherited

Definition at line 138 of file SGObject.cpp.

## ◆ set_generic() [15/16]

 void set_generic ( )
inherited

Definition at line 143 of file SGObject.cpp.

## ◆ set_generic() [16/16]

 void set_generic ( )
inherited

set generic type to T

## ◆ set_global_io()

 void set_global_io ( SGIO * io )
inherited

set the io object

Parameters
 io io object to use

Definition at line 262 of file SGObject.cpp.

## ◆ set_global_parallel()

 void set_global_parallel ( Parallel * parallel )
inherited

set the parallel object

Parameters
 parallel parallel object to use

Definition at line 275 of file SGObject.cpp.

## ◆ set_global_version()

 void set_global_version ( Version * version )
inherited

set the version object

Parameters
 version version object to use

Definition at line 317 of file SGObject.cpp.

## ◆ set_is_initialized()

 void set_is_initialized ( bool p_init )
protectedinherited

set is initialized

Parameters
 p_init if optimization shall be set to initialized

Definition at line 899 of file kernel/Kernel.h.

## ◆ set_matrix_weights()

 void set_matrix_weights ( SGMatrix< float64_t > weights )
virtualinherited

setter for feature/dimension weights (matrix kernel)

Parameters
 weights a lower triangular matrix weight with positive diagonal elements

Definition at line 149 of file ExponentialARDKernel.cpp.

## ◆ set_normalizer()

 bool set_normalizer ( CKernelNormalizer * normalizer )
virtualinherited

set the current kernel normalizer

Returns
if successful

Reimplemented in CWeightedDegreeStringKernel.

Definition at line 149 of file Kernel.cpp.

## ◆ set_optimization_type()

 virtual void set_optimization_type ( EOptimizationType t )
virtualinherited

set optimization type

Parameters
 t optimization type to set

Reimplemented in CCombinedKernel.

Definition at line 747 of file kernel/Kernel.h.

## ◆ set_property()

 void set_property ( EKernelProperty p )
protectedinherited

set property

Parameters
 p kernel property to set

Definition at line 881 of file kernel/Kernel.h.

## ◆ set_scalar_weights()

 void set_scalar_weights ( float64_t weight )
virtualinherited

setter for feature/dimension weight (scalar kernel)

Parameters
 weight positive scalar weight

Definition at line 120 of file ExponentialARDKernel.cpp.

## ◆ set_subkernel_weights()

 void set_subkernel_weights ( SGVector< float64_t > weights )
virtualinherited

set subkernel weights

Parameters
 weights new subkernel weights

Reimplemented in CCombinedKernel, CWeightedDegreeStringKernel, and CWeightedDegreePositionStringKernel.

Definition at line 864 of file Kernel.cpp.

## ◆ set_time()

 void set_time ( int32_t t )
inherited

set the lru time

Parameters
 t the time to use

Definition at line 664 of file kernel/Kernel.h.

## ◆ set_vector_weights()

 void set_vector_weights ( SGVector< float64_t > weights )
virtualinherited

setter for feature/dimension weights (vector kernel)

Parameters
 weights positive vector weight

Definition at line 131 of file ExponentialARDKernel.cpp.

## ◆ set_weights()

 void set_weights ( SGMatrix< float64_t > weights )
protectedvirtualinherited

a general setter for feature/dimension weights (matrix kernel)

Parameters
 weights the weights can be scalar/vector/lower triangular matrix

Definition at line 67 of file ExponentialARDKernel.cpp.

## ◆ shallow_copy()

 CSGObject * shallow_copy ( ) const
virtualinherited

A shallow copy. All the SGObject instance variables will be simply assigned and SG_REF-ed.

Reimplemented in CGaussianKernel.

Definition at line 226 of file SGObject.cpp.

## ◆ subscribe_to_parameters()

 void subscribe_to_parameters ( ParameterObserverInterface * obs )
inherited

Subscribe a parameter observer to watch over params

Definition at line 811 of file SGObject.cpp.

## ◆ sum_block()

 float64_t sum_block ( index_t block_begin_row, index_t block_begin_col, index_t block_size_row, index_t block_size_col, bool no_diag = false )
virtualinherited

Computes sum of kernel values from a specified block. This method is useful while computing statistical estimation of mean/variance over kernel values but the kernel matrix is too huge to be fit inside memory.

Parameters
 block_begin_row the row index at which the block starts block_begin_col the col index at which the block starts block_size_row the number of rows in the block block_size_col the number of cols in the block

For example, block_begin_row 0, block_begin_col 4 and block_size_row 5, block_size_col 6 represents the block that starts at index (0,4) in the kernel matrix and goes upto (0+5-1,4+6-1) i.e. (4,9) both inclusive

Parameters
 no_diag if true (default is false), the diagonal elements are excluded from the sum, provided that block_size_row and block_size_col are same (i.e. the block is square). Otherwise, these are always added
Returns
sum of kernel values within the block computed as

$\sum_{i}\sum_{j}k(i+\text{block-begin-row}, j+\text{block-begin-col})$

where $$i\in[0,\text{block-size-row}-1]$$ and $$j\in[0,\text{block-size-col}-1]$$

Reimplemented in CCustomKernel.

Definition at line 1054 of file Kernel.cpp.

## ◆ sum_symmetric_block()

 float64_t sum_symmetric_block ( index_t block_begin, index_t block_size, bool no_diag = true )
virtualinherited

Computes sum from a symmetric part of the kernel matrix that always is supposed to contain the main upper diagonal. This method is useful while computing statistical estimation of mean/variance over kernel values but the kernel matrix is too huge to be fit inside memory.

Parameters
 block_begin the row and col index at which the block starts block_size the number of rows and cols in the block

For example, block_begin 4 and block_size 5 represents the block that starts at index (4,4) in the kernel matrix and goes upto (4+5-1,4+5-1) i.e. (8,8) both inclusive

Parameters
 no_diag if true (default), the diagonal elements are excluded from the sum
Returns
sum of kernel values within the block computed as

$\sum_{i}\sum_{j}k(i+\text{block-begin}, j+\text{block-begin})$

where $$i,j\in[0,\text{block-size}-1]$$

Reimplemented in CCustomKernel.

Definition at line 1003 of file Kernel.cpp.

## ◆ unref()

 int32_t unref ( )
inherited

decrement reference counter and deallocate object if refcount is zero before or after decrementing it

Returns
reference count

Definition at line 200 of file SGObject.cpp.

## ◆ unset_generic()

 void unset_generic ( )
inherited

unset generic type

this has to be called in classes specializing a template class

Definition at line 337 of file SGObject.cpp.

## ◆ unset_property()

 void unset_property ( EKernelProperty p )
protectedinherited

unset property

Parameters
 p kernel property to unset

Definition at line 890 of file kernel/Kernel.h.

## ◆ update_parameter_hash()

 void update_parameter_hash ( )
virtualinherited

Updates the hash of current parameter combination

Definition at line 282 of file SGObject.cpp.

## ◆ cache_size

 int32_t cache_size
protectedinherited

cache_size in MB

Definition at line 1047 of file kernel/Kernel.h.

## ◆ combined_kernel_weight

 float64_t combined_kernel_weight
protectedinherited

combined kernel weight

Definition at line 1072 of file kernel/Kernel.h.

## ◆ io

 SGIO* io
inherited

io

Definition at line 600 of file SGObject.h.

## ◆ kernel_cache

 KERNEL_CACHE kernel_cache
protectedinherited

kernel cache

Definition at line 1051 of file kernel/Kernel.h.

## ◆ kernel_matrix

 KERNELCACHE_ELEM* kernel_matrix
protectedinherited

this *COULD* store the whole kernel matrix usually not applicable / necessary to compute the whole matrix

Definition at line 1056 of file kernel/Kernel.h.

## ◆ lhs

 CFeatures* lhs
protectedinherited

feature vectors to occur on left hand side

Definition at line 1059 of file kernel/Kernel.h.

## ◆ lhs_equals_rhs

 bool lhs_equals_rhs
protectedinherited

lhs

Definition at line 1064 of file kernel/Kernel.h.

## ◆ m_ARD_type

 EARDKernelType m_ARD_type
protectedinherited

type of ARD kernel

Definition at line 118 of file ExponentialARDKernel.h.

inherited

parameters wrt which we can compute gradients

Definition at line 615 of file SGObject.h.

## ◆ m_hash

 uint32_t m_hash
inherited

Hash of parameter values

Definition at line 618 of file SGObject.h.

## ◆ m_log_weights

 SGVector m_log_weights
protectedinherited

feature weights in log domain in vector layout

Definition at line 109 of file ExponentialARDKernel.h.

## ◆ m_model_selection_parameters

 Parameter* m_model_selection_parameters
inherited

model selection parameters

Definition at line 612 of file SGObject.h.

## ◆ m_parameters

 Parameter* m_parameters
inherited

parameters

Definition at line 609 of file SGObject.h.

## ◆ m_sq_lhs

 SGVector m_sq_lhs
protected

squared left-hand side

Definition at line 165 of file GaussianARDKernel.h.

## ◆ m_sq_rhs

 SGVector m_sq_rhs
protected

squared right-hand side

Definition at line 167 of file GaussianARDKernel.h.

## ◆ m_weights_cols

 index_t m_weights_cols
protectedinherited

the number of columns of feature weights for vector layout

Definition at line 115 of file ExponentialARDKernel.h.

## ◆ m_weights_raw

 SGMatrix m_weights_raw
protectedinherited

feature weights in standard domain in the matrix layout, which is only used in get_weights()

Definition at line 106 of file ExponentialARDKernel.h.

## ◆ m_weights_rows

 index_t m_weights_rows
protectedinherited

the number of rows of feature weights for vector layout

Definition at line 112 of file ExponentialARDKernel.h.

## ◆ normalizer

 CKernelNormalizer* normalizer
protectedinherited

normalize the kernel(i,j) function based on this normalization function

Definition at line 1086 of file kernel/Kernel.h.

## ◆ num_lhs

 int32_t num_lhs
protectedinherited

number of feature vectors on left hand side

Definition at line 1067 of file kernel/Kernel.h.

## ◆ num_rhs

 int32_t num_rhs
protectedinherited

number of feature vectors on right hand side

Definition at line 1069 of file kernel/Kernel.h.

## ◆ opt_type

 EOptimizationType opt_type
protectedinherited

optimization type (currently FASTBUTMEMHUNGRY and SLOWBUTMEMEFFICIENT)

Definition at line 1079 of file kernel/Kernel.h.

## ◆ optimization_initialized

 bool optimization_initialized
protectedinherited

if optimization is initialized

Definition at line 1075 of file kernel/Kernel.h.

## ◆ parallel

 Parallel* parallel
inherited

parallel

Definition at line 603 of file SGObject.h.

## ◆ properties

 uint64_t properties
protectedinherited

kernel properties

Definition at line 1082 of file kernel/Kernel.h.

## ◆ rhs

 CFeatures* rhs
protectedinherited

feature vectors to occur on right hand side

Definition at line 1061 of file kernel/Kernel.h.

## ◆ version

 Version* version
inherited

version

Definition at line 606 of file SGObject.h.

The documentation for this class was generated from the following files:

SHOGUN Machine Learning Toolbox - Documentation