wolffd@0: wolffd@0: wolffd@0: wolffd@0: Netlab Reference Manual gmm wolffd@0: wolffd@0: wolffd@0: wolffd@0:

gmm wolffd@0:

wolffd@0:

wolffd@0: Purpose wolffd@0:

wolffd@0: Creates a Gaussian mixture model with specified architecture. wolffd@0: wolffd@0:

wolffd@0: Synopsis wolffd@0:

wolffd@0:

wolffd@0: mix = gmm(dim, ncentres, covartype)
wolffd@0: mix = gmm(dim, ncentres, covartype, ppca_dim)
wolffd@0:

wolffd@0: wolffd@0: wolffd@0:

wolffd@0: Description wolffd@0:

wolffd@0: wolffd@0: mix = gmm(dim, ncentres, covartype) takes wolffd@0: the dimension of the space dim, the number of centres in the wolffd@0: mixture model and the type of the mixture model, and returns a data wolffd@0: structure mix. wolffd@0: The mixture model type defines the covariance structure of each component wolffd@0: Gaussian: wolffd@0:

wolffd@0: 
wolffd@0:   'spherical' = single variance parameter for each component: stored as a vector
wolffd@0:   'diag' = diagonal matrix for each component: stored as rows of a matrix
wolffd@0:   'full' = full matrix for each component: stored as 3d array
wolffd@0:   'ppca' = probabilistic PCA: stored as principal components (in a 3d array
wolffd@0:     and associated variances and off-subspace noise
wolffd@0:

wolffd@0: wolffd@0: mix = gmm(dim, ncentres, covartype, ppca_dim) also sets the dimension of wolffd@0: the PPCA sub-spaces: the default value is one. wolffd@0: wolffd@0:

The priors are initialised to equal values summing to one, and the covariances wolffd@0: are all the identity matrix (or equivalent). The centres are wolffd@0: initialised randomly from a zero mean unit variance Gaussian. This makes use wolffd@0: of the MATLAB function randn and so the seed for the random weight wolffd@0: initialisation can be set using randn('state', s) where s is the wolffd@0: state value. wolffd@0: wolffd@0:

The fields in mix are wolffd@0:

wolffd@0:   
wolffd@0:   type = 'gmm'
wolffd@0:   nin = the dimension of the space
wolffd@0:   ncentres = number of mixture components
wolffd@0:   covartype = string for type of variance model
wolffd@0:   priors = mixing coefficients
wolffd@0:   centres = means of Gaussians: stored as rows of a matrix
wolffd@0:   covars = covariances of Gaussians
wolffd@0:

wolffd@0: wolffd@0: The additional fields for mixtures of PPCA are wolffd@0:

wolffd@0: 
wolffd@0:   U = principal component subspaces
wolffd@0:   lambda = in-space covariances: stored as rows of a matrix
wolffd@0:

wolffd@0: wolffd@0: The off-subspace noise is stored in covars. wolffd@0: wolffd@0:

wolffd@0: Example wolffd@0:

wolffd@0:

wolffd@0: 
wolffd@0: mix = gmm(2, 4, 'spherical');
wolffd@0:

wolffd@0: wolffd@0: This creates a Gaussian mixture model with 4 components in 2 dimensions. wolffd@0: The covariance structure is a spherical model. wolffd@0: wolffd@0:

wolffd@0: See Also wolffd@0:

wolffd@0: gmmpak, gmmunpak, gmmsamp, gmminit, gmmem, gmmactiv, gmmpost, gmmprob

wolffd@0: Pages: wolffd@0: Index wolffd@0:

wolffd@0: