set.options: Set and print options for network inference.

Description Usage Arguments Value See Also Examples

Description

Set and print options for network inference.

Usage

1
2
3
4
5
6
7
set.options(first.list = NULL, second.list = NULL, scale = TRUE,
  verbose = TRUE, mu = NULL, penalties = NULL, penalty.min = 0.01,
  penalty.max = NULL, n.penalties = 100, edges.max = Inf,
  symmetrization = c("AND", "OR"), initial.guess = NULL, max.it = NULL)

## S3 method for class 'internet3DOptions'
print(x, ...)

Arguments

first.list

list of edges supposed to be present (array with two columns and a number of rows corresponding to the number of edges with a prior)

second.list

list of edges supposed to be absent (array with two columns and a number of rows corresponding to the number of edges with a prior)

scale

logical. Should the data be scaled to unit variance prior the analysis

verbose

logical. Should messages be printed during the learning process

mu

positive number. Regularization parameter for the L2 penalty on edges with a prior assumption

penalties

a sequence of decreasing positive numbers to control the L1 regularization

penalty.min

if penalties=NULL, minimum value for the L1 regularization parameter

penalty.max

if penalties=NULL, maximum value for the L1 regularization parameter

n.penalties

if penalties=NULL, number of values to include in the sequence of positive numbers that control the L1 regularization

edges.max

maximum number of edges to stop the learning process

symmetrization

symmetrization rule, to be chosen between "AND" and "OR"

initial.guess

initial guess (if existing) for the solution of the optimization problem

max.it

maximum number of iterations allowed to solve the optimization problem

x

a internet3DOptions object

...

not used

Value

object of type internet3DOptions that can be used within the functions build.network or bootstrap.build

See Also

build.network, bootstrap.build

Examples

1
2

tuxette/internet3D documentation built on May 8, 2019, 11:59 p.m.