deal (version 1.1-2)

maketrylist: Create the full trylist

Description

For faster learning, a trylist is maintained.

Usage

maketrylist(initnw,data,prior=jointprior(network(data)),timetrace=FALSE)

Arguments

initnw

The network

data

Dataframe

prior

Joint prior

timetrace

For timing purposes

Value

• A list with one element per node in the network. Element i is a list of nodes. These nodes are copies of node i in the network, except that all possible parent configurations is present. And for each copy, the node is learned so that the parameter posteriors and log-likelihood contribution is calculated. These nodes can then be copied into networks where the node appears.

Details

For each node in the network, all possible parent configurations are created and learned. The result is called a trylist. This is very time-consuming, and a better choice is to maintain a trylist, while searching and indeed this is automatically done.

References

Further information about Deal can be found at: http://www.math.auc.dk/novo/deal.

See Also

reuselearn, networkfamily, heuristic

Examples

data(rats)
rats.nw <- network(rats)
rats.pr <- jointprior(rats.nw,12)
rats.nw <- learn(rats.nw,rats,rats.pr)$nw
rats.tr <- maketrylist(rats.nw,rats,rats.pr)

rats.hi <- heuristic(rats.nw,rats,rats.pr,trylist=rats.tr)$nw