summary.DPMMclust: Summarizing Dirichlet Process Mixture Models
Description
Summary methods for DPMMclust objects.
Usage
"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).