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

# S3 method for class 'grain'
compile(
  object,
  propagate = FALSE,
  tug = NULL,
  root = NULL,
  control = object$control,
  details = 0,
  ...
)

Arguments

object

A grain object.

propagate

If TRUE the network is also propagated meaning that the cliques of the junction tree are calibrated to each other.

tug

A triangulated undirected graph.

root

A set of variables which must be in the root of the junction tree

control

Controlling the compilation process.

details

For debugging info. Do not use.

...

Currently not used.

Value

A compiled Bayesian network; an object of class grain.

References

Søren Højsgaard (2012). Graphical Independence Networks with the gRain Package for R. Journal of Statistical Software, 46(10), 1-26. https://www.jstatsoft.org/v46/i10/.

Author

Søren Højsgaard, sorenh@math.aau.dk