NPflow (version 0.9.0)

summary.DPMMclust: Summarizing Dirichlet Process Mixture Models

Description

Summary methods for DPMMclust objects.

Usage

## S3 method for class 'DPMMclust':
summary(object, burnin = 0, thin = 1, gs = NULL,
  lossFn = "F-measure", posterior_approx = FALSE, ...)

Arguments

object

a DPMMclust object.

burnin

integer giving the number of MCMC iterations to burn (defaults is half)

thin

integer giving the spacing at which MCMC iterations are kept. Default is 1, i.e. no thining.

gs

optionnal vector of length n containing the gold standard partition of the n observations to compare to the point estimate

lossFn

character string specifying the loss function to be used. Either "F-measure" or "Binder" (see Details). Default is "F-measure".

posterior_approx

logical flag wether a parametric approximation of the posterior should be computed. Default is FALSE

...

further arguments passed to or from other methods

Value

• a list:
  • burnin:
  {an integer passing along the burnin argument}
• thin:
  an integer passing along the thin argument
• lossFn:
  a character string passing along the lossFn argument
• point_estim:
• loss:
• index_estim:

Details

The cost of a point estimate partition is calculated using either a pairwise coincidence loss function (Binder), or 1-Fmeasure (F-measure).

See Also

similarityMat similarityMatC