itpp::MOG_generic Class Reference
[Mixture of Gaussians (MOG)]

Generic Mixture of Gaussians (MOG) class. Used as a base for other MOG classes. More...

#include <mog_generic.h>

List of all members.

Public Member Functions

MOG_generic ()

Default constructor.

MOG_generic (const std::string &name_in)

Construct the MOG_generic object by loading the parameters from a model file.

MOG_generic (const int &K_in, const int &D_in, bool full_in=false)

construct a default model (all Gaussians have zero mean and unit variance for all dimensions)

MOG_generic (Array< vec > &means_in, bool full_in=false)

Construct a model using user supplied mean vectors.

MOG_generic (Array< vec > &means_in, Array< vec > &diag_covs_in, vec &weights_in)

Construct a model using user supplied parameters (diagonal covariance version).

MOG_generic (Array< vec > &means_in, Array< mat > &full_covs_in, vec &weights_in)

Construct a model using user supplied parameters (full covariance version).

virtual ~MOG_generic ()

Default destructor.

void init ()

Initialise the model to be empty.

void init (const int &K_in, const int &D_in, bool full_in=false)

initialise the model so that all Gaussians have zero mean and unit variance for all dimensions

void init (Array< vec > &means_in, bool full_in=false)

Initialise the model using user supplied mean vectors.

void init (Array< vec > &means_in, Array< vec > &diag_covs_in, vec &weights_in)

Initialise the model using user supplied parameters (diagonal covariance version).

void init (Array< vec > &means_in, Array< mat > &full_covs_in, vec &weights_in)

Initialise the model using user supplied parameters (full covariance version).

virtual void cleanup ()

Release memory used by the model. The model will be empty.

bool is_valid () const

Returns true if the model's parameters are valid.

bool is_full () const

Returns true if the model has full covariance matrices.

int get_K () const

Return the number of Gaussians.

int get_D () const

Return the dimensionality.

vec get_weights () const

Obtain a copy of the weight vector.

Array< vec > get_means () const

Obtain a copy of the array of mean vectors.

Array< vec > get_diag_covs () const

Obtain a copy of the array of diagonal covariance vectors.

Array< mat > get_full_covs () const

Obtain a copy of the array of full covariance matrices.

void set_means (Array< vec > &means_in)

Set the means of the model.

void set_diag_covs (Array< vec > &diag_covs_in)

Set the diagonal covariance vectors of the model.

void set_full_covs (Array< mat > &full_covs_in)

Set the full covariance matrices of the model.

void set_weights (vec &weights_in)

Set the weight vector of the model.

void set_means_zero ()

Set the means in the model to be zero.

void set_diag_covs_unity ()

Set the diagonal covariance vectors to be unity.

void set_full_covs_unity ()

Set the full covariance matrices to be unity.

void set_weights_uniform ()

Set all the weights to 1/K, where K is the number of Gaussians.

void set_checks (bool do_checks_in)

Enable/disable internal checks for likelihood functions.

void set_paranoid (bool paranoid_in)

Enable/disable paranoia about numerical stability.

virtual void load (const std::string &name_in)

Initialise the model by loading the parameters from a model file.

virtual void save (const std::string &name_in) const

Save the model's parameters to a model file.

virtual void join (const MOG_generic &B_in)

Mathematically join the model with a user supplied model.

virtual void convert_to_diag ()

Convert the model to use diagonal covariances.

virtual void convert_to_full ()

Convert the model to have full covariance matrices.

virtual double log_lhood_single_gaus (const vec &x_in, const int k)

calculate the log likelihood of vector x_in using only Gaussian k

virtual double log_lhood (const vec &x_in)

calculate the log likelihood of vector x_in

virtual double lhood (const vec &x_in)

calculate the likelihood of vector x_in

virtual double avg_log_lhood (const Array< vec > &X_in)

calculate the average log likelihood of an array of vectors X_in

Protected Member Functions

bool check_size (const vec &x_in) const

Check if vector x_in has the same dimensionality as the model.

bool check_size (const Array< vec > &X_in) const

Check if all vectors in Array X_in have the same dimensionality as the model.

bool check_array_uniformity (const Array< vec > &A) const

Check if all vectors in Array X_in have the same dimensionality.

void set_means_internal (Array< vec > &means_in)

ADD DOCUMENTATION HERE.

void set_diag_covs_internal (Array< vec > &diag_covs_in)

ADD DOCUMENTATION HERE.

void set_full_covs_internal (Array< mat > &full_covs_in)

ADD DOCUMENTATION HERE.

void set_weights_internal (vec &_weigths)

ADD DOCUMENTATION HERE.

void set_means_zero_internal ()

ADD DOCUMENTATION HERE.

void set_diag_covs_unity_internal ()

ADD DOCUMENTATION HERE.

void set_full_covs_unity_internal ()

ADD DOCUMENTATION HERE.

void set_weights_uniform_internal ()

ADD DOCUMENTATION HERE.

void convert_to_diag_internal ()

ADD DOCUMENTATION HERE.

void convert_to_full_internal ()

ADD DOCUMENTATION HERE.

virtual void setup_means ()

additional processing of mean vectors, done as the last step of mean initialisation

virtual void setup_covs ()

additional processing of covariance vectors/matrices, done as the last step of covariance initialisation

virtual void setup_weights ()

additional processing of the weight vector, done as the last step of weight initialisation

virtual void setup_misc ()

additional processing of miscellaneous parameters, done as the last step of overall initialisation

virtual double log_lhood_single_gaus_internal (const vec &x_in, const int k)

ADD DOCUMENTATION HERE.

virtual double log_lhood_internal (const vec &x_in)

ADD DOCUMENTATION HERE.

virtual double lhood_internal (const vec &x_in)

ADD DOCUMENTATION HERE.

Protected Attributes

bool do_checks

indicates whether checks on input data are done

bool valid

indicates whether the parameters are valid

bool full

indicates whether we are using full or diagonal covariance matrices

bool paranoid

indicates whether we are paranoid about numerical stability

int K

number of gaussians

int D

dimensionality

Array< vec > means

means

Array< vec > diag_covs

diagonal covariance matrices, stored as vectors

Array< mat > full_covs

full covariance matrices

vec weights

weights

double log_max_K

Pre-calcualted std::log(std::numeric_limits<double>::max() / K), where K is the number of Gaussians.

vec log_det_etc

Gaussian specific pre-calcualted constants.

vec log_weights

Pre-calculated log versions of the weights.

Array< mat > full_covs_inv

Pre-calcuated inverted version of each full covariance matrix.

Array< vec > diag_covs_inv_etc

Pre-calcuated inverted version of each diagonal covariance vector, where the covariance elements are first multiplied by two.

Detailed Description

Generic Mixture of Gaussians (MOG) class. Used as a base for other MOG classes.

Author:: Conrad Sanderson

Used for representing a statistical distribution as a convex combination of multi-variate Gaussian functions. Also known as a Gaussian Mixture Model. This class handles both full and diagonal covariance matrices, allows loading and saving of the MOG's parameters, as well as calculation of likelihoods. The parameters are set by the user or an optimisation algorithm (for example, see the MOG_diag_EM class). For speed and space reasons, diagonal and full covariance matrices are stored and handled separately.

Constructor & Destructor Documentation

itpp::MOG_generic::MOG_generic ( ) [inline]

Default constructor.

Note:: An empty model is created. The likelihood functions are not useable until the model's parameters are set

References init().

itpp::MOG_generic::MOG_generic ( const std::string & name_in ) [inline]

Construct the MOG_generic object by loading the parameters from a model file.

Parameters:

name_in

The model's filename

References load().

itpp::MOG_generic::MOG_generic	(	const int &	K_in,
		const int &	D_in,
		bool	full_in = `false`
	)			`[inline]`

construct a default model (all Gaussians have zero mean and unit variance for all dimensions)

Parameters:

	K_in	Number of Gaussians
	D_in	Dimensionality
	full_in	If true, use full covariance matrices; if false, use diagonal covariance matrices. Default = false.

References init().

itpp::MOG_generic::MOG_generic	(	Array< vec > &	means_in,
		bool	full_in = `false`
	)			`[inline]`

Construct a model using user supplied mean vectors.

Parameters:

	means_in	Array of mean vectors
	full_in	If true, use full covariance matrices; if false, use diagonal covariance matrices. Default = false.

Note:: The number of mean vectors specifies the number of Gaussians. The covariance matrices are set to the identity matrix. The weights for all Gaussians are the same, equal to 1/K, where K is the number of Gaussians

References init().

itpp::MOG_generic::MOG_generic	(	Array< vec > &	means_in,
		Array< vec > &	diag_covs_in,
		vec &	weights_in
	)			`[inline]`

Construct a model using user supplied parameters (diagonal covariance version).

Parameters:

	means_in	Array of mean vectors
	diag_covs_in	Array of vectors representing diagonal covariances
	weights_in	vector of weights

Note:: The number of mean vectors, covariance vectors and weights must be the same

References init().

itpp::MOG_generic::MOG_generic	(	Array< vec > &	means_in,
		Array< mat > &	full_covs_in,
		vec &	weights_in
	)			`[inline]`

Construct a model using user supplied parameters (full covariance version).

Parameters:

	means_in	Array of mean vectors
	full_covs_in	Array of full covariance matrices
	weights_in	vector of weights

Note:: The number of mean vectors, covariance matrices and weights must be the same

References init().

Member Function Documentation

void itpp::MOG_generic::cleanup ( ) [virtual]

Release memory used by the model. The model will be empty.

Note:: The likelihood functions are not useable until the model's parameters are re-initialised

Reimplemented in itpp::MOG_diag.

References D, diag_covs, diag_covs_inv_etc, do_checks, full_covs, full_covs_inv, K, log_det_etc, log_weights, means, itpp::Array< T >::set_size(), valid, and weights.

Referenced by init(), and ~MOG_generic().

void itpp::MOG_generic::convert_to_diag ( ) [virtual]

Convert the model to use diagonal covariances.

Note:: If the model is already diagonal, nothing is done. If the model has full covariance matrices, this results in irreversible information loss (in effect the off-diagonal covariance elements are now zero)

References convert_to_diag_internal(), and valid.

Referenced by itpp::MOG_diag::load(), and itpp::MOG_diag::MOG_diag().

void itpp::MOG_generic::convert_to_full ( ) [virtual]

Convert the model to have full covariance matrices.

Note:: If the model has full covariance matrices, nothing is done. If the model has diagonal covariances, the off-diagonal elements in the full covariance matrices are set to zero.

Reimplemented in itpp::MOG_diag.

References convert_to_full_internal(), and valid.

void itpp::MOG_generic::init	(	Array< vec > &	means_in,
		Array< mat > &	full_covs_in,
		vec &	weights_in
	)

Initialise the model using user supplied parameters (full covariance version).

Parameters:

	means_in	Array of mean vectors
	full_covs_in	Array of covariance matrices
	weights_in	vector of weights

Note:: The number of mean vectors, covariance matrices and weights must be the same

References check_array_uniformity(), D, do_checks, full, it_assert, K, paranoid, set_full_covs_internal(), set_means_internal(), set_weights_internal(), setup_misc(), itpp::Array< T >::size(), and valid.

void itpp::MOG_generic::init	(	Array< vec > &	means_in,
		Array< vec > &	diag_covs_in,
		vec &	weights_in
	)

Initialise the model using user supplied parameters (diagonal covariance version).

Parameters:

	means_in	Array of mean vectors
	diag_covs_in	Array of vectors representing diagonal covariances
	weights_in	vector of weights

Note:: The number of mean vectors, covariance vectors and weights must be the same

References check_array_uniformity(), D, do_checks, full, it_assert, K, paranoid, set_diag_covs_internal(), set_means_internal(), set_weights_internal(), setup_misc(), itpp::Array< T >::size(), and valid.

void itpp::MOG_generic::init	(	Array< vec > &	means_in,
		bool	full_in = `false`
	)

Initialise the model using user supplied mean vectors.

Parameters:

	means_in	Array of mean vectors
	full_in	If true, use full covariance matrices; if false, use diagonal covariance matrices. Default = false.

Note:: The number of mean vectors specifies the number of Gaussians. The covariance matrices are set to the identity matrix. The weights for all Gaussians are the same, equal to 1/K, where K is the number of Gaussians

References check_array_uniformity(), D, do_checks, full, it_assert, K, paranoid, set_diag_covs_unity_internal(), set_full_covs_unity_internal(), set_means(), set_weights_uniform_internal(), setup_misc(), itpp::Array< T >::size(), and valid.

void itpp::MOG_generic::init	(	const int &	K_in,
		const int &	D_in,
		bool	full_in = `false`
	)

initialise the model so that all Gaussians have zero mean and unit variance for all dimensions

Parameters:

	K_in	Number of Gaussians
	D_in	Dimensionality
	full_in	If true, use full covariance matrices; if false, use diagonal covariance matrices. Default = false.

References D, do_checks, full, it_assert, K, paranoid, set_diag_covs_unity_internal(), set_full_covs_unity_internal(), set_means_zero_internal(), set_weights_uniform_internal(), setup_misc(), and valid.

void itpp::MOG_generic::init ( )

Initialise the model to be empty.

Note:: The likelihood functions are not useable until the model's parameters are set

References cleanup().

Referenced by join(), load(), itpp::MOG_diag_EM_sup::ml(), itpp::MOG_diag::MOG_diag(), MOG_generic(), and itpp::MOG_diag_kmeans_sup::run().

void itpp::MOG_generic::join ( const MOG_generic & B_in ) [virtual]

Mathematically join the model with a user supplied model.

Parameters:

B_in

user supplied model

Note:: The Arrays of mean vectors and covariance vectors/matrices from the two models are simply concatenated, while the weights of the resultant model are a function of the original weights and numbers of Gaussians from both models. Specifically, $w_{new} = [ \alpha \cdot w_{A} ~~~ \beta \cdot w_{B} ]^T$ , where $w_{new}$ is the new weight vector, $w_{A}$ and $w_{B}$ are the weight vectors from model A and B, while $\alpha = K_A / (K_A + KB_in)$ and $\beta = 1-\alpha$ . In turn, and are the numbers of Gaussians in model A and B, respectively.

See On transforming statistical models... for more information.

References D, diag_covs, full, full_covs, get_D(), get_diag_covs(), get_full_covs(), get_K(), get_means(), get_weights(), init(), is_full(), is_valid(), it_assert, K, means, valid, and weights.

void itpp::MOG_generic::load ( const std::string & name_in ) [virtual]

Initialise the model by loading the parameters from a model file.

Parameters:

name_in

The model's filename

Reimplemented in itpp::MOG_diag.

References itpp::it_file::close(), itpp::exist(), itpp::it_file::exists(), init(), it_assert, and valid.

Referenced by MOG_generic().

void itpp::MOG_generic::save ( const std::string & name_in ) const [virtual]

Save the model's parameters to a model file.

Parameters:

name_in

The model's filename

References itpp::it_file::close(), diag_covs, full, full_covs, means, valid, and weights.

void itpp::MOG_generic::set_checks ( bool do_checks_in ) [inline]

Enable/disable internal checks for likelihood functions.

Parameters:

do_checks_in

If true, checks are enabled; if false, checks are disabled

Note:: Disabling checks will provide a speedup in the likelihood functions. Disable them only when you're happy that everything is working correctly.

References do_checks.

void itpp::MOG_generic::set_diag_covs ( Array< vec > & diag_covs_in )

Set the diagonal covariance vectors of the model.

Note:: The number of diagonal covariance vectors must match the number of Gaussians in the model

References set_diag_covs_internal(), and valid.

void itpp::MOG_generic::set_full_covs ( Array< mat > & full_covs_in )

Set the full covariance matrices of the model.

Note:: The number of covariance matrices must match the number of Gaussians in the model

References set_full_covs_internal(), and valid.

void itpp::MOG_generic::set_means ( Array< vec > & means_in )

Set the means of the model.

Note:: The number of means must match the number of Gaussians in the model

References set_means_internal(), and valid.

Referenced by init().

void itpp::MOG_generic::set_paranoid ( bool paranoid_in ) [inline]

Enable/disable paranoia about numerical stability.

Parameters:

paranoid_in

If true, calculate likelihoods using a safer, but slower method.

References paranoid.

void itpp::MOG_generic::set_weights ( vec & weights_in )

Set the weight vector of the model.

Note:: The number of elements in the weight vector must match the number of Gaussians in the model

References set_weights_internal(), and valid.

Member Data Documentation

vec itpp::MOG_generic::log_det_etc [protected]

Gaussian specific pre-calcualted constants.

Note:: Vector of pre-calculated $-\frac{D}{2}\log(2\pi) -\frac{1}{2}\log(|\Sigma|)$ for each Gaussian, where is the dimensionality and $|\Sigma|$ is the determinant for the Gaussian's covariance matrix $\Sigma$ .

Referenced by cleanup(), log_lhood_single_gaus_internal(), setup_covs(), and itpp::MOG_diag::setup_covs().

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


Public Member Functions
	MOG_generic ()
	Default constructor.
	MOG_generic (const std::string &name_in)
	Construct the MOG_generic object by loading the parameters from a model file.
	MOG_generic (const int &K_in, const int &D_in, bool full_in=false)
	construct a default model (all Gaussians have zero mean and unit variance for all dimensions)
	MOG_generic (Array< vec > &means_in, bool full_in=false)
	Construct a model using user supplied mean vectors.
	MOG_generic (Array< vec > &means_in, Array< vec > &diag_covs_in, vec &weights_in)
	Construct a model using user supplied parameters (diagonal covariance version).
	MOG_generic (Array< vec > &means_in, Array< mat > &full_covs_in, vec &weights_in)
	Construct a model using user supplied parameters (full covariance version).
virtual	~MOG_generic ()
	Default destructor.
void	init ()
	Initialise the model to be empty.
void	init (const int &K_in, const int &D_in, bool full_in=false)
	initialise the model so that all Gaussians have zero mean and unit variance for all dimensions
void	init (Array< vec > &means_in, bool full_in=false)
	Initialise the model using user supplied mean vectors.
void	init (Array< vec > &means_in, Array< vec > &diag_covs_in, vec &weights_in)
	Initialise the model using user supplied parameters (diagonal covariance version).
void	init (Array< vec > &means_in, Array< mat > &full_covs_in, vec &weights_in)
	Initialise the model using user supplied parameters (full covariance version).
virtual void	cleanup ()
	Release memory used by the model. The model will be empty.
bool	is_valid () const
	Returns true if the model's parameters are valid.
bool	is_full () const
	Returns true if the model has full covariance matrices.
int	get_K () const
	Return the number of Gaussians.
int	get_D () const
	Return the dimensionality.
vec	get_weights () const
	Obtain a copy of the weight vector.
Array< vec >	get_means () const
	Obtain a copy of the array of mean vectors.
Array< vec >	get_diag_covs () const
	Obtain a copy of the array of diagonal covariance vectors.
Array< mat >	get_full_covs () const
	Obtain a copy of the array of full covariance matrices.
void	set_means (Array< vec > &means_in)
	Set the means of the model.
void	set_diag_covs (Array< vec > &diag_covs_in)
	Set the diagonal covariance vectors of the model.
void	set_full_covs (Array< mat > &full_covs_in)
	Set the full covariance matrices of the model.
void	set_weights (vec &weights_in)
	Set the weight vector of the model.
void	set_means_zero ()
	Set the means in the model to be zero.
void	set_diag_covs_unity ()
	Set the diagonal covariance vectors to be unity.
void	set_full_covs_unity ()
	Set the full covariance matrices to be unity.
void	set_weights_uniform ()
	Set all the weights to 1/K, where K is the number of Gaussians.
void	set_checks (bool do_checks_in)
	Enable/disable internal checks for likelihood functions.
void	set_paranoid (bool paranoid_in)
	Enable/disable paranoia about numerical stability.
virtual void	load (const std::string &name_in)
	Initialise the model by loading the parameters from a model file.
virtual void	save (const std::string &name_in) const
	Save the model's parameters to a model file.
virtual void	join (const MOG_generic &B_in)
	Mathematically join the model with a user supplied model.
virtual void	convert_to_diag ()
	Convert the model to use diagonal covariances.
virtual void	convert_to_full ()
	Convert the model to have full covariance matrices.
virtual double	log_lhood_single_gaus (const vec &x_in, const int k)
	calculate the log likelihood of vector `x_in` using only Gaussian `k`
virtual double	log_lhood (const vec &x_in)
	calculate the log likelihood of vector `x_in`
virtual double	lhood (const vec &x_in)
	calculate the likelihood of vector `x_in`
virtual double	avg_log_lhood (const Array< vec > &X_in)
	calculate the average log likelihood of an array of vectors `X_in`
Protected Member Functions
bool	check_size (const vec &x_in) const
	Check if vector `x_in` has the same dimensionality as the model.
bool	check_size (const Array< vec > &X_in) const
	Check if all vectors in Array `X_in` have the same dimensionality as the model.
bool	check_array_uniformity (const Array< vec > &A) const
	Check if all vectors in Array `X_in` have the same dimensionality.
void	set_means_internal (Array< vec > &means_in)
	ADD DOCUMENTATION HERE.
void	set_diag_covs_internal (Array< vec > &diag_covs_in)
	ADD DOCUMENTATION HERE.
void	set_full_covs_internal (Array< mat > &full_covs_in)
	ADD DOCUMENTATION HERE.
void	set_weights_internal (vec &_weigths)
	ADD DOCUMENTATION HERE.
void	set_means_zero_internal ()
	ADD DOCUMENTATION HERE.
void	set_diag_covs_unity_internal ()
	ADD DOCUMENTATION HERE.
void	set_full_covs_unity_internal ()
	ADD DOCUMENTATION HERE.
void	set_weights_uniform_internal ()
	ADD DOCUMENTATION HERE.
void	convert_to_diag_internal ()
	ADD DOCUMENTATION HERE.
void	convert_to_full_internal ()
	ADD DOCUMENTATION HERE.
virtual void	setup_means ()
	additional processing of mean vectors, done as the last step of mean initialisation
virtual void	setup_covs ()
	additional processing of covariance vectors/matrices, done as the last step of covariance initialisation
virtual void	setup_weights ()
	additional processing of the weight vector, done as the last step of weight initialisation
virtual void	setup_misc ()
	additional processing of miscellaneous parameters, done as the last step of overall initialisation
virtual double	log_lhood_single_gaus_internal (const vec &x_in, const int k)
	ADD DOCUMENTATION HERE.
virtual double	log_lhood_internal (const vec &x_in)
	ADD DOCUMENTATION HERE.
virtual double	lhood_internal (const vec &x_in)
	ADD DOCUMENTATION HERE.
Protected Attributes
bool	do_checks
	indicates whether checks on input data are done
bool	valid
	indicates whether the parameters are valid
bool	full
	indicates whether we are using full or diagonal covariance matrices
bool	paranoid
	indicates whether we are paranoid about numerical stability
int	K
	number of gaussians
int	D
	dimensionality
Array< vec >	means
	means
Array< vec >	diag_covs
	diagonal covariance matrices, stored as vectors
Array< mat >	full_covs
	full covariance matrices
vec	weights
	weights
double	log_max_K
	Pre-calcualted std::log(std::numeric_limits<double>::max() / K), where K is the number of Gaussians.
vec	log_det_etc
	Gaussian specific pre-calcualted constants.
vec	log_weights
	Pre-calculated log versions of the weights.
Array< mat >	full_covs_inv
	Pre-calcuated inverted version of each full covariance matrix.
Array< vec >	diag_covs_inv_etc
	Pre-calcuated inverted version of each diagonal covariance vector, where the covariance elements are first multiplied by two.

itpp::MOG_generic Class Reference [Mixture of Gaussians (MOG)]

Public Member Functions

Protected Member Functions

Protected Attributes

Detailed Description

Constructor & Destructor Documentation

Member Function Documentation

Member Data Documentation

itpp::MOG_generic Class Reference
[Mixture of Gaussians (MOG)]