SHOGUN  4.1.0
CPCA Class Reference

Detailed Description

Preprocessor PCA performs principial component analysis on input feature vectors/matrices. When the init method in PCA is called with proper feature matrix X (with say N number of vectors and D feature dimension), a transformation matrix is computed and stored internally. This transformation matrix is then used to transform all D-dimensional feature vectors or feature matrices (with D feature dimensions) supplied via apply_to_feature_matrix or apply_to_feature_vector methods. This tranformation outputs the T-Dimensional approximation of all these input vectors and matrices (where T<=min(D,N)). The transformation matrix is essentially a DxT matrix, the columns of which correspond to the eigenvectors of the covariance matrix(XX') having top T eigenvalues.

This class provides 3 method options to compute the transformation matrix : EVD : Eigen Value Decomposition of Covariance Matrix ( $$XX^T$$) The covariance matrix $$XX^T$$ is first formed internally and then its eigenvectors and eigenvalues are computed. The time complexity of this method is $$~10D^3$$ and should be used when N > D.

SVD : Singular Value Decomposition of feature matrix X The transpose of feature matrix, $$X^T$$, is decomposed using SVD. $$X^T = UDV^T$$ The matrix V in this decomposition contains the required eigenvectors and the diagonal entries of the diagonal matrix D correspond to the non-negative eigenvalues. Eigenvalue, $$e_i$$, is derived from a diagonal element, $$d_i$$, using the formula $$e_i = \frac{\sqrt{d_i}}{N-1}$$. The time complexity of this method is $$~14DN^2$$ and should be used when N < D.

AUTO : This mode automagically chooses one of the above modes for the user based on whether N > D (chooses EVD) or N < D (chooses SVD).

This class provides 3 modes to determine the value of T :

FIXED_NUMBER : T is supplied by user directly using set_target_dims method

VARIANCE_EXPLAINED : The user supplies the fractional variance that he wants preserved in the target dimension T. From this supplied fractional variance (thresh), T is calculated as the smallest k such that the ratio of sum of largest k eigenvalues over total sum of all eigenvalues is greater than thresh.

THRESH : The user supplies a threshold. All eigenvectors with corresponding eigenvalue greater than the supplied threshold are chosen.

An option for whitening the transformation matrix is also given - do_whitening. Setting this option normalizes the eigenvectors (ie. the columns of transformation matrix) by dividing them with the square root of corresponding eigenvalues.

Note that vectors/matrices don't have to have zero mean as it is substracted within the class.

Definition at line 113 of file PCA.h.

Inheritance diagram for CPCA:
[legend]

Public Member Functions

CPCA (bool do_whitening=false, EPCAMode mode=FIXED_NUMBER, float64_t thresh=1e-6, EPCAMethod method=AUTO, EPCAMemoryMode mem_mode=MEM_REALLOCATE)

CPCA (EPCAMethod method, bool do_whitening=false, EPCAMemoryMode mem=MEM_REALLOCATE)

virtual ~CPCA ()

virtual bool init (CFeatures *features)

virtual void cleanup ()

virtual SGMatrix< float64_tapply_to_feature_matrix (CFeatures *features)

virtual SGVector< float64_tapply_to_feature_vector (SGVector< float64_t > vector)

SGMatrix< float64_tget_transformation_matrix ()

SGVector< float64_tget_eigenvalues ()

SGVector< float64_tget_mean ()

virtual const char * get_name () const

virtual EPreprocessorType get_type () const

EPCAMemoryMode get_memory_mode () const

void set_memory_mode (EPCAMemoryMode e)

void set_eigenvalue_zero_tolerance (float64_t eigenvalue_zero_tolerance=1e-15)

float64_t get_eigenvalue_zero_tolerance () const

void set_target_dim (int32_t dim)

int32_t get_target_dim () const

void set_distance (CDistance *distance)

CDistanceget_distance () const

void set_kernel (CKernel *kernel)

CKernelget_kernel () const

virtual CFeaturesapply (CFeatures *features)

virtual EFeatureClass get_feature_class ()
return that we are dense features (just fixed size matrices) More...

virtual EFeatureType get_feature_type ()
return feature type More...

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)

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 ()

Public Attributes

SGIOio

Parallelparallel

Versionversion

Parameterm_parameters

Parameterm_model_selection_parameters

uint32_t m_hash

Protected Member Functions

void init ()

virtual void load_serializable_pre () throw (ShogunException)

virtual void load_serializable_post () throw (ShogunException)

virtual void save_serializable_pre () throw (ShogunException)

virtual void save_serializable_post () throw (ShogunException)

Protected Attributes

SGMatrix< float64_tm_transformation_matrix

int32_t num_dim

int32_t num_old_dim

SGVector< float64_tm_mean_vector

SGVector< float64_tm_eigenvalues_vector

bool m_initialized

bool m_whitening

EPCAMode m_mode

float64_t m_thresh

EPCAMemoryMode m_mem_mode

EPCAMethod m_method

float64_t m_eigenvalue_zero_tolerance

int32_t m_target_dim

CDistancem_distance

CKernelm_kernel

CEmbeddingConverterm_converter

Constructor & Destructor Documentation

 CPCA ( bool do_whitening = false, EPCAMode mode = FIXED_NUMBER, float64_t thresh = 1e-6, EPCAMethod method = AUTO, EPCAMemoryMode mem_mode = MEM_REALLOCATE )

standard constructor

Parameters
 do_whitening normalize columns(eigenvectors) in transformation matrix mode mode of pca : FIXED_NUMBER/VARIANCE_EXPLAINED/THRESHOLD thresh threshold value for VARIANCE_EXPLAINED or THRESHOLD mode method Matrix decomposition method used : SVD/EVD/AUTO[default] mem_mode memory usage mode of PCA : MEM_REALLOCATE/MEM_IN_PLACE

Definition at line 26 of file PCA.cpp.

 CPCA ( EPCAMethod method, bool do_whitening = false, EPCAMemoryMode mem = MEM_REALLOCATE )

special constructor for FIXED_NUMBER mode

Parameters
 method Matrix decomposition method used : SVD/EVD/AUTO[default] do_whitening normalize columns(eigenvectors) in transformation matrix mem memory usage mode of PCA : MEM_REALLOCATE/MEM_IN_PLACE

Definition at line 37 of file PCA.cpp.

 ~CPCA ( )
virtual

destructor

Definition at line 80 of file PCA.cpp.

Member Function Documentation

 virtual CFeatures* apply ( CFeatures * features )
virtualinherited

generic interface for applying the preprocessor. used as a wrapper for apply_to_feature_matrix() method

Parameters
 features the dense input features
Returns
the result feature object after applying the preprocessor

Implements CPreprocessor.

 SGMatrix< float64_t > apply_to_feature_matrix ( CFeatures * features )
virtual

apply preprocessor to feature matrix

Parameters
 features features
Returns
processed feature matrix

Reimplemented from CDimensionReductionPreprocessor.

Definition at line 278 of file PCA.cpp.

 SGVector< float64_t > apply_to_feature_vector ( SGVector< float64_t > vector )
virtual

apply preprocessor to feature vector

Parameters
 vector feature vector
Returns
processed feature vector

Reimplemented from CDimensionReductionPreprocessor.

Definition at line 333 of file PCA.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 597 of file SGObject.cpp.

 void cleanup ( )
virtual

cleanup

Reimplemented from CDimensionReductionPreprocessor.

Definition at line 270 of file PCA.cpp.

 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 714 of file SGObject.cpp.

 CSGObject * deep_copy ( ) const
virtualinherited

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

Definition at line 198 of file SGObject.cpp.

 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 618 of file SGObject.cpp.

 CDistance * get_distance ( ) const
inherited

getter for distance

Returns
distance

Definition at line 88 of file DimensionReductionPreprocessor.cpp.

 float64_t get_eigenvalue_zero_tolerance ( ) const

get zero tolerance of eigenvalues during data whitening

Returns
zero tolerance value

Definition at line 380 of file PCA.cpp.

 SGVector< float64_t > get_eigenvalues ( )

get eigenvalues of PCA

Definition at line 355 of file PCA.cpp.

 virtual EFeatureClass get_feature_class ( )
virtualinherited

return that we are dense features (just fixed size matrices)

Implements CPreprocessor.

Reimplemented in CRandomFourierGaussPreproc.

 virtual EFeatureType get_feature_type ( )
virtualinherited

return feature type

Implements CPreprocessor.

Reimplemented in CRandomFourierGaussPreproc.

 SGIO * get_global_io ( )
inherited

get the io object

Returns
io object

Definition at line 235 of file SGObject.cpp.

 Parallel * get_global_parallel ( )
inherited

get the parallel object

Returns
parallel object

Definition at line 277 of file SGObject.cpp.

 Version * get_global_version ( )
inherited

get the version object

Returns
version object

Definition at line 290 of file SGObject.cpp.

 CKernel * get_kernel ( ) const
inherited

getter for kernel

Returns
kernel

Definition at line 101 of file DimensionReductionPreprocessor.cpp.

 SGVector< float64_t > get_mean ( )

get mean vector of original data

Definition at line 360 of file PCA.cpp.

 EPCAMemoryMode get_memory_mode ( ) const

return the PCA memory mode being used

Definition at line 365 of file PCA.cpp.

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

Definition at line 498 of file SGObject.cpp.

 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 522 of file SGObject.cpp.

 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 535 of file SGObject.cpp.

 virtual const char* get_name ( ) const
virtual
Returns
object name

Reimplemented from CDimensionReductionPreprocessor.

Definition at line 173 of file PCA.h.

 int32_t get_target_dim ( ) const
inherited

getter for target dimension

Returns
target dimension

Definition at line 76 of file DimensionReductionPreprocessor.cpp.

 SGMatrix< float64_t > get_transformation_matrix ( )

get transformation matrix, i.e. eigenvectors (potentially scaled if do_whitening is true)

Definition at line 350 of file PCA.cpp.

 virtual EPreprocessorType get_type ( ) const
virtual
Returns
a type of preprocessor

Reimplemented from CDimensionReductionPreprocessor.

Definition at line 176 of file PCA.h.

 bool init ( CFeatures * features )
virtual

initialize preprocessor from features

Parameters
 features

Reimplemented from CDimensionReductionPreprocessor.

Definition at line 84 of file PCA.cpp.

 void init ( )
protected

Definition at line 46 of file PCA.cpp.

 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 296 of file SGObject.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 369 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 occurs.

Definition at line 426 of file SGObject.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 421 of file SGObject.cpp.

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

Definition at line 262 of file SGObject.cpp.

 void print_modsel_params ( )
inherited

prints all parameter registered for model selection and their type

Definition at line 474 of file SGObject.cpp.

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

prints registered parameters out

Parameters
 prefix prefix for members

Definition at line 308 of file SGObject.cpp.

 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 314 of file SGObject.cpp.

 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 occurs.

Reimplemented in CKernel.

Definition at line 436 of file SGObject.cpp.

 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 occurs.

Definition at line 431 of file SGObject.cpp.

 void set_distance ( CDistance * distance )
inherited

setter for distance

Parameters
 distance distance to set

Definition at line 81 of file DimensionReductionPreprocessor.cpp.

 void set_eigenvalue_zero_tolerance ( float64_t eigenvalue_zero_tolerance = 1e-15 )

set zero tolerance of eigenvalues during data whitening

Parameters
 eigenvalue_zero_tolerance zero tolerance value

Definition at line 375 of file PCA.cpp.

 void set_generic ( )
inherited

Definition at line 41 of file SGObject.cpp.

 void set_generic ( )
inherited

Definition at line 46 of file SGObject.cpp.

 void set_generic ( )
inherited

Definition at line 51 of file SGObject.cpp.

 void set_generic ( )
inherited

Definition at line 56 of file SGObject.cpp.

 void set_generic ( )
inherited

Definition at line 61 of file SGObject.cpp.

 void set_generic ( )
inherited

Definition at line 66 of file SGObject.cpp.

 void set_generic ( )
inherited

Definition at line 71 of file SGObject.cpp.

 void set_generic ( )
inherited

Definition at line 76 of file SGObject.cpp.

 void set_generic ( )
inherited

Definition at line 81 of file SGObject.cpp.

 void set_generic ( )
inherited

Definition at line 86 of file SGObject.cpp.

 void set_generic ( )
inherited

Definition at line 91 of file SGObject.cpp.

 void set_generic ( )
inherited

Definition at line 96 of file SGObject.cpp.

 void set_generic ( )
inherited

Definition at line 101 of file SGObject.cpp.

 void set_generic ( )
inherited

Definition at line 106 of file SGObject.cpp.

 void set_generic ( )
inherited

Definition at line 111 of file SGObject.cpp.

 void set_generic ( )
inherited

set generic type to T

 void set_global_io ( SGIO * io )
inherited

set the io object

Parameters
 io io object to use

Definition at line 228 of file SGObject.cpp.

 void set_global_parallel ( Parallel * parallel )
inherited

set the parallel object

Parameters
 parallel parallel object to use

Definition at line 241 of file SGObject.cpp.

 void set_global_version ( Version * version )
inherited

set the version object

Parameters
 version version object to use

Definition at line 283 of file SGObject.cpp.

 void set_kernel ( CKernel * kernel )
inherited

setter for kernel

Parameters
 kernel kernel to set

Definition at line 94 of file DimensionReductionPreprocessor.cpp.

 void set_memory_mode ( EPCAMemoryMode e )

set PCA memory mode to be used

Parameters
 choice between MEM_REALLOCATE and MEM_IN_PLACE

Definition at line 370 of file PCA.cpp.

 void set_target_dim ( int32_t dim )
inherited

setter for target dimension

Parameters
 dim target dimension

Definition at line 70 of file DimensionReductionPreprocessor.cpp.

 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 192 of file SGObject.cpp.

 void unset_generic ( )
inherited

unset generic type

this has to be called in classes specializing a template class

Definition at line 303 of file SGObject.cpp.

 void update_parameter_hash ( )
virtualinherited

Updates the hash of current parameter combination

Definition at line 248 of file SGObject.cpp.

Member Data Documentation

 SGIO* io
inherited

io

Definition at line 369 of file SGObject.h.

 CEmbeddingConverter* m_converter
protectedinherited

embedding converter to be used

Definition at line 127 of file DimensionReductionPreprocessor.h.

 CDistance* m_distance
protectedinherited

distance to be used

Definition at line 121 of file DimensionReductionPreprocessor.h.

 float64_t m_eigenvalue_zero_tolerance
protected

eigenvalues within zero tolerance region are considered 0 while whitening to tackle numerical issues

Definition at line 228 of file PCA.h.

 SGVector m_eigenvalues_vector
protected

eigenvalues vector

Definition at line 211 of file PCA.h.

inherited

parameters wrt which we can compute gradients

Definition at line 384 of file SGObject.h.

 uint32_t m_hash
inherited

Hash of parameter values

Definition at line 387 of file SGObject.h.

 bool m_initialized
protected

initialized

Definition at line 213 of file PCA.h.

 CKernel* m_kernel
protectedinherited

kernel to be used

Definition at line 124 of file DimensionReductionPreprocessor.h.

 SGVector m_mean_vector
protected

mean vector

Definition at line 209 of file PCA.h.

 EPCAMemoryMode m_mem_mode
protected

PCA memory mode

Definition at line 221 of file PCA.h.

 EPCAMethod m_method
protected

PCA method

Definition at line 223 of file PCA.h.

 EPCAMode m_mode
protected

PCA mode

Definition at line 217 of file PCA.h.

 Parameter* m_model_selection_parameters
inherited

model selection parameters

Definition at line 381 of file SGObject.h.

 Parameter* m_parameters
inherited

parameters

Definition at line 378 of file SGObject.h.

 int32_t m_target_dim
protectedinherited

target dim of dimensionality reduction preprocessor

Definition at line 118 of file DimensionReductionPreprocessor.h.

 float64_t m_thresh
protected

thresh

Definition at line 219 of file PCA.h.

 SGMatrix m_transformation_matrix
protected

transformation matrix

Definition at line 203 of file PCA.h.

 bool m_whitening
protected

whitening

Definition at line 215 of file PCA.h.

 int32_t num_dim
protected

num dim

Definition at line 205 of file PCA.h.

 int32_t num_old_dim
protected

num old dim

Definition at line 207 of file PCA.h.

 Parallel* parallel
inherited

parallel

Definition at line 372 of file SGObject.h.

 Version* version
inherited

version

Definition at line 375 of file SGObject.h.

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

SHOGUN Machine Learning Toolbox - Documentation