CRF (version 0.4-3)

decode.cutset: Decoding method for graphs with a small cutset

Description

Computing the most likely configuration for CRF

Usage

decode.cutset(
  crf,
  cutset,
  engine = "default",
  start = apply(crf$node.pot, 1, which.max)
)

Arguments

crf

The CRF

cutset

A vector of nodes in the cutset

engine

The underlying engine for cutset decoding, possible values are "default", "none", "exact", "chain", and "tree".

start

An initial configuration, a good start will significantly reduce the seraching time

Value

This function will return the most likely configuration, which is a vector of length crf$n.nodes.

Details

Exact decoding for graphs with a small cutset using cutset conditioning

Examples

# NOT RUN {
library(CRF)
data(Small)
d <- decode.cutset(Small$crf, c(2))

# }