createInits: Create a list with initial values.

In cbdAmgen/bcgp0a: Bayesian Composite Gaussian Processes

Description

createInits returns a list that contains randomly generated initial values.

Usage

createInits(x, priors = createPriors(x), chains = 4)

Arguments

x	An n x d matrix containing the independent variables in the training set.
priors	A list that contains the parameter values for the priors.
chains	The number of Markov chains. The default is 4.

Details

This creates a list of length chains that contains randomly generated initial values. The intention is to specify initial values for the parameters for the Markov chains. The user can change the values as they like prior to inputting the list into bcgp.

Value

A list of length chains The elements of this list will be named lists, where each of these has the name of a parameter.

TODO

Decide whether to add options for "heteroscedastic" and "composite"

See Also

bcgp

Other preprocessing functions: createPriors

Examples

x <- matrix(runif(40), ncol= 4, nrow = 10)
createInits(x)
createInits(x, priors = createPriors(x, noise = TRUE), chains = 2)

cbdAmgen/bcgp0a documentation built on May 17, 2019, 10:01 a.m.