insert.kernel.instance: Insert Kernel Instance into Model Tree

In mattdneal/gaussianProcess: Gaussian Process Fitting and Prediction

Description

Insert Kernel Instance into Model Tree

Usage

insert.kernel.instance(orig.model.tree, node, kernel.class.name, operation.id,
  hyper.params = NULL)

Arguments

orig.model.tree	a ModelTree object
node	the node at which to insert the kernel instance
kernel.class.name	a string indicating the class of kernel to insert
operation.id	a string indicating how to insert the kernel at node. Must be "*" or "+", or NULL if the model tree is currently empty.
hyper.params	the hyper parameters of the inserted kernel. Can be NULL to fit hyper parameters at a later time.

Value

a ModelTree object

Examples

mt <- create.model.tree.builtin()

mt <- insert.kernel.instance(mt, 1, "SE", NULL)
print(mt) # "SE.1"

mt <- insert.kernel.instance(mt, 1, "SE", "*")
print(mt) # "( SE.1 * SE.2 )"

mt <- insert.kernel.instance(mt, 3, "RQ", "+")
print(mt) # "( ( SE.1 * SE.2 ) + RQ.1 )"

mattdneal/gaussianProcess documentation built on May 21, 2019, 12:58 p.m.