Learn R Programming

inlabru (version 2.3.1)

bru_mapper: Methods for bru_mapper objects

Description

Methods for bru_mapper objects

Usage

bru_mapper(...)

ibm_n(mapper, ...)

ibm_values(mapper, ...)

ibm_amatrix(mapper, input, ...)

ibm_valid_input(mapper, input, ...)

# S3 method for default bru_mapper( mapper, new_class = NULL, ibm_n = NULL, ibm_values = NULL, ibm_amatrix = NULL, ibm_valid_input = NULL, ... )

# S3 method for inla.mesh bru_mapper(mesh, ...)

# S3 method for inla.mesh.1d bru_mapper(mesh, indexed = NULL, ...)

bru_mapper_index(n = 1L, ...)

bru_mapper_linear(...)

bru_mapper_factor(values, factor_mapping, ...)

bru_mapper_offset(...)

bru_mapper_multi(mappers, ...)

Arguments

Arguments passed on to other methods

mapper

A mapper S3 object, normally inheriting from bru_mapper. For the default bru_mapper method, a list that will be converted to a bru_mapper object by adding class information and (optional) methods.

input

The values for which to produce validity information

new_class

If non-NULL, this is added at the front of the class definition

ibm_n

An ibm_n method function

ibm_values

An ibm_values method function

ibm_amatrix

An ibm_amatrix method function

ibm_valid_input

An ibm_valid_input method function

mesh

An inla.mesh.1d or inla.mesh.2d object to use as a mapper

indexed

logical; If TRUE, the ibm_values() output will be the integer indexing sequence for the latent variables (needed for spde models). If FALSE, the knot locations are returned (useful as an interpolator for rw2 models and similar). Default: NULL, to force user specification of this parameter

n

Size of a model for bru_mapper_index

values

Input values calculated by input_eval.bru_input()

factor_mapping

character; selects the type of factor mapping.

  • 'contrast' for leaving out the first factor level.

  • 'full' for keeping all levels.

mappers

A list of bru_mapper objects

Details

  • bru_mapper Generic mapper S3 constructor. See below for details of the default constructor that can be used to define new mappers in user code.

  • ibm_n Generic. Implementations must return the size of the latent vector being mapped to.

  • ibm_values Generic. Implementations must return a vector that is interpretable by an INLA::f(..., values = ...) specification, and has length ibm_n().

  • ibm_amatrix Generic. Implementations must return a (sparse) matrix of size NROW(input) (except for the bru_mapper_multi method) by ibm_n(mapper)

  • ibm_valid_input Generic. Implementations must return a logical vector of length NROW(input)

  • bru_mapper.default adds the "bru_mapper" class and new_class to an object. If provided, mapper method functions are added to an environment .envir in the object. The generic methods ibm_n, ibm_values, ibm_amatrix, and ibm_valid_input look for these functions first, and otherwise call UseMethod(). This is an alternative to using .S3method() to register the methods, e.g. .S3method("ibm_amatrix", "my_mapper_class", ibm_amatrix.my_mapper_class).

  • bru_mapper_multi constructs a kronecker product mapping

See Also

bru_mapper_methods for specific method implementations.

Examples

Run this code
# NOT RUN {
mapper <- bru_mapper_index(5)
ibm_amatrix(mapper, c(1,3,4,5,2))
# }

Run the code above in your browser using DataLab