compile {gRain}R Documentation

Compile a Bayesian network (aka a graphical independence network)

Description

Compiles a Bayesian network. This means creating a junction tree and establishing clique potentials.

Usage

compile(object, ...)
compile.grain(object, method = "mcwh", propagate = FALSE, root = NULL, smooth = 0, control = object$control, trace = 0, ...)

Arguments

object A grain object
method Triangulation method. Currently only "mcwh" is available
propagate If TRUE the network is also propagated meaning that the cliques of the junction tree are calibrated to each other.
root A set of variables which must be in the root of the junction tree
smooth If non-zero this value is added to all potentials (to avoid zeros in the joint distribution)
control Controlling the compilation process.
trace For debugging info. Do not use.
... Currently not used.

Value

A compiled Bayesian network; an object of class compgrain.

Author(s)

Søren Højsgaard, sorenh@agrsci.dk

See Also

grain, propagate


[Package gRain version 0.7.0 Index]