Dirichlet Process with spatial Weights class for spatially-aware Bayesian nonparametric clustering. More...

#include <DPW.hpp>

Inheritance diagram for DPW:

Public Member Functions
	DPW (Data &d, Params &p)
	Constructor for the Dirichlet Process with spatial Weights.

void	update_params () override
	Updates the parameters of the Dirichlet Process with spatial Weights.

Gibbs Sampling Methods
double	gibbs_prior_existing_cluster (int cls_idx, int obs_idx) const override
	Computes the log prior probability of assigning a data point to an existing cluster.

Eigen::VectorXd	gibbs_prior_existing_clusters (int obs_idx) const override
	Computes the log prior probabilities of assigning a data point to every existing cluster. This method is useful for Gibbs sampling over existing clusters. It returns a vector of log prior probabilities for all existing clusters.

double	gibbs_prior_new_cluster () const override
	Computes the log prior probability of assigning a data point to a new cluster.

Split-Merge Algorithm Methods
double	prior_ratio_split (int ci, int cj) const override
	Computes the prior ratio for a split operation in a spatially-aware split-merge MCMC algorithm.

double	prior_ratio_merge (int size_old_ci, int size_old_cj) const override
	Computes the prior ratio for a merge operation in a spatially-aware split-merge MCMC algorithm.

double	prior_ratio_shuffle (int size_old_ci, int size_old_cj, int ci, int cj) const override
	Computes the prior ratio for a shuffle operation in a spatially-aware split-merge MCMC algorithm.

Spatial Methods
int	get_neighbors_obs (int obs_idx, int cls_idx) const
	Returns the number of neighbors for a given observation in a specific cluster.

Eigen::VectorXi	get_neighbors_obs (int obs_idx) const
	Returns the number of neighbors for a given observation regardless of cluster membership. This method counts the total number of neighbors of an observation based on the adjacency matrix W.

int	get_neighbors_cls (int cls_idx, bool old_allo=false) const
	Returns the total number of neighbors for all observations in a given cluster.

Public Member Functions inherited from Process
	Process (Data &d, Params &p)
	Constructor initializing process with data and parameters.

void	set_old_allocations (const Eigen::VectorXi &new_allocations)
	Store current allocations for potential rollback.

void	set_idx_i (int i)
	Set index of first observation in split-merge pair.

void	set_idx_j (int j)
	Set index of second observation in split-merge pair.

virtual	~Process ()
	Virtual destructor for proper cleanup of derived classes.

Additional Inherited Members
Protected Attributes inherited from Process
Data &	data
	Reference to the data object containing observations and allocations.

Params &	params
	Reference to the parameters object containing process hyperparameters.

Eigen::VectorXi	old_allocations
	Storage for previous allocations to enable rollback in case of rejection.

int	idx_i
	Index of first observation involved in split-merge move.

int	idx_j
	Index of second observation involved in split-merge move.

const double	log_a = log(params.a)
	Precomputed logarithm of total mass parameter for efficiency.

Detailed Description

Dirichlet Process with spatial Weights class for spatially-aware Bayesian nonparametric clustering.

This class extends the Dirichlet Process to incorporate spatial information through an adjacency matrix W. It provides methods for Gibbs sampling and split-merge algorithms that account for spatial dependencies between observations in the clustering process.

Constructor & Destructor Documentation

◆ DPW()

DPW::DPW	(	Data &	d,
		Params &	p )

inline

Constructor for the Dirichlet Process with spatial Weights.

Parameters

d	Reference to the data object containing observations and cluster assignments.
p	Reference to the parameters object containing the adjacency matrix W and spatial coefficient.

Member Function Documentation

◆ get_neighbors_cls()

int DPW::get_neighbors_cls	(	int	cls_idx,
		bool	old_allo = false ) const

Returns the total number of neighbors for all observations in a given cluster.

This method computes the sum of all neighbor connections within a cluster, which is used in the spatial component of the prior calculations.

Parameters

cls_idx	The index of the cluster.
old_allo	If true, uses the old allocations for neighbor counting; otherwise, uses current allocations (default: false).

Returns: The total number of neighbors for the cluster.

Returns the total number of neighbors for all observations in a given cluster.

This method computes the sum of all neighbor connections within a cluster, which is used in the spatial component of the prior calculations.

Parameters

cls_idx	The index of the cluster.
old_allo	If true, uses the old allocations for neighbor counting; otherwise, uses current allocations.

Returns: The total number of neighbors for the cluster.

◆ get_neighbors_obs() [1/2]

Eigen::VectorXi DPW::get_neighbors_obs ( int obs_idx ) const

Returns the number of neighbors for a given observation regardless of cluster membership. This method counts the total number of neighbors of an observation based on the adjacency matrix W.

Parameters

obs_idx The index of the observation.

Returns: The total number of neighbors for the observation for all clusters.

Returns the number of neighbors for a given observation regardless of cluster membership. This method counts the total number of neighbors of an observation based on the adjacency matrix W.

Parameters

obs_idx The index of the observation.

Returns: The total number of neighbors for the observation for all clusters.

◆ get_neighbors_obs() [2/2]

int DPW::get_neighbors_obs	(	int	obs_idx,
		int	cls_idx ) const

Returns the number of neighbors for a given observation in a specific cluster.

This method counts the neighbors of an observation based on the adjacency matrix W, considering only neighbors that belong to the specified cluster.

Parameters

obs_idx	The index of the observation.
cls_idx	The index of the cluster to consider for neighbor counting.

Returns: The number of neighbors for the observation in the specified cluster.

Returns the number of neighbors for a given observation in a specific cluster.

This method counts the neighbors of an observation based on the adjacency matrix W, considering only neighbors that belong to the specified cluster.

Parameters

obs_idx	The index of the observation.
cls_idx	The index of the cluster to consider for neighbor counting.

Returns: The number of neighbors for the observation in the specified cluster.

◆ gibbs_prior_existing_cluster()

double DPW::gibbs_prior_existing_cluster	(	int	cls_idx,
		int	obs_idx ) const

nodiscardoverridevirtual

Computes the log prior probability of assigning a data point to an existing cluster.

This method incorporates spatial information by considering the number of neighbors in the target cluster when computing the prior probability.

Parameters

cls_idx	The index of the cluster.
obs_idx	The index of the observation to assign.

Returns: The log prior probability of assigning the data point to the existing cluster.

Computes the log prior probability of assigning a data point to an existing cluster.

This method incorporates spatial information by considering the number of neighbors in the target cluster when computing the prior probability.

Parameters

cls_idx	The index of the cluster.
obs_idx	The index of the observation to assign.

Returns: The log prior probability of assigning the data point to the existing cluster.

Implements Process.

◆ gibbs_prior_existing_clusters()

Eigen::VectorXd DPW::gibbs_prior_existing_clusters ( int obs_idx ) const

nodiscardoverridevirtual

Computes the log prior probabilities of assigning a data point to every existing cluster. This method is useful for Gibbs sampling over existing clusters. It returns a vector of log prior probabilities for all existing clusters.

Parameters

obs_idx The index of the observation to assign.

Returns: A vector of log prior probabilities for assigning the data point to each existing cluster.

Computes the log prior probabilities of assigning a data point to all existing clusters.

This method incorporates spatial information by considering the number of neighbors in each target cluster when computing the prior probabilities.

Parameters

obs_idx The index of the observation to assign.

Returns: A vector of log prior probabilities for assigning the data point to each existing cluster.

Implements Process.

◆ gibbs_prior_new_cluster()

double DPW::gibbs_prior_new_cluster ( ) const

nodiscardoverridevirtual

Computes the log prior probability of assigning a data point to a new cluster.

Returns: The log prior probability of assigning the data point to a new cluster.

Computes the log prior probability of assigning a data point to a new cluster.

Returns: The log prior probability of assigning the data point to a new cluster.

Implements Process.

◆ prior_ratio_merge()

double DPW::prior_ratio_merge	(	int	size_old_ci,
		int	size_old_cj ) const

nodiscardoverridevirtual

Computes the prior ratio for a merge operation in a spatially-aware split-merge MCMC algorithm.

This method accounts for both the Dirichlet Process prior and spatial dependencies when computing the acceptance ratio for merging clusters.

Parameters

size_old_ci	The size of the first cluster before the merge.
size_old_cj	The size of the second cluster before the merge.

Returns: The log prior ratio for the merge operation.

Computes the prior ratio for a merge operation in a spatially-aware split-merge MCMC algorithm.

This method accounts for both the Dirichlet Process prior and spatial dependencies when computing the acceptance ratio for merging clusters.

Parameters

size_old_ci	The size of the first cluster before the merge.
size_old_cj	The size of the second cluster before the merge.

Returns: The log prior ratio for the merge operation.

Implements Process.

◆ prior_ratio_shuffle()

double DPW::prior_ratio_shuffle	(	int	size_old_ci,
		int	size_old_cj,
		int	ci,
		int	cj ) const

nodiscardoverridevirtual

Computes the prior ratio for a shuffle operation in a spatially-aware split-merge MCMC algorithm.

This method accounts for both the Dirichlet Process prior and spatial dependencies when computing the acceptance ratio for shuffling observations between clusters.

Parameters

size_old_ci	The size of the first cluster before the shuffle.
size_old_cj	The size of the second cluster before the shuffle.
ci	The first cluster index involved in the shuffle.
cj	The second cluster index involved in the shuffle.

Returns: The log prior ratio for the shuffle operation.

Computes the prior ratio for a shuffle operation in a spatially-aware split-merge MCMC algorithm.

This method accounts for both the Dirichlet Process prior and spatial dependencies when computing the acceptance ratio for shuffling observations between clusters.

Parameters

size_old_ci	The size of the first cluster before the shuffle.
size_old_cj	The size of the second cluster before the shuffle.
ci	The first cluster index involved in the shuffle.
cj	The second cluster index involved in the shuffle.

Returns: The log prior ratio for the shuffle operation.

Implements Process.

◆ prior_ratio_split()

double DPW::prior_ratio_split	(	int	ci,
		int	cj ) const

nodiscardoverridevirtual

Computes the prior ratio for a split operation in a spatially-aware split-merge MCMC algorithm.

This method accounts for both the Dirichlet Process prior and spatial dependencies when computing the acceptance ratio for splitting clusters.

Parameters

ci	The first cluster index involved in the split.
cj	The second cluster index involved in the split.

Returns: The log prior ratio for the split operation.

Computes the prior ratio for a split operation in a spatially-aware split-merge MCMC algorithm.

This method accounts for both the Dirichlet Process prior and spatial dependencies when computing the acceptance ratio for splitting clusters.

Parameters

ci	The first cluster index involved in the split.
cj	The second cluster index involved in the split.

Returns: The log prior ratio for the split operation.

Implements Process.

◆ update_params()

void DPW::update_params ( )

inlineoverridevirtual

Updates the parameters of the Dirichlet Process with spatial Weights.

This is a null implementation as the DPW has no parameters to update beyond those handled by the base Process class.

Implements Process.

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

src/processes/DPW.hpp
src/processes/DPW.cpp

Public Member Functions

Additional Inherited Members

Detailed Description

Constructor & Destructor Documentation

◆ DPW()

Member Function Documentation

◆ get_neighbors_cls()

◆ get_neighbors_obs() [1/2]

◆ get_neighbors_obs() [2/2]

◆ gibbs_prior_existing_cluster()

◆ gibbs_prior_existing_clusters()

◆ gibbs_prior_new_cluster()

◆ prior_ratio_merge()

◆ prior_ratio_shuffle()

◆ prior_ratio_split()

◆ update_params()