inlabru-deprecated: Deprecated functions in inlabru

In inlabru: Bayesian Latent Gaussian Modelling using INLA and Extensions

Deprecated functions in inlabru

Description

These functions still attempt to do their job, but will be removed in a future version.

Usage

ipoints(
  samplers = NULL,
  domain = NULL,
  name = NULL,
  group = NULL,
  int.args = NULL,
  project = deprecated()
)

cprod(..., na.rm = NULL, .blockwise = FALSE)

integration_weight_aggregation(mesh, integ)

mesh_triangle_integration(mesh, tri_subset = NULL, nsub = NULL)

## Default S3 method:
bru_mapper(...)

bru_mapper_offset(...)

is.inside(mesh, loc, mesh.coords = NULL)

vertices.inla.mesh(...)

pixels(mesh, nx = 150, ny = 150, mask = TRUE)

Arguments

samplers	Description of the integration region boundary. In 1D, a length 2 vector or two-column matrix where each row describes an interval, or NULL In 2D either a SpatialPolygon or a SpatialLinesDataFrame with a weight column defining the width of the a transect line, and optionally further columns used by the group argument, or NULL. When domain is NULL, samplers may also be an inla.mesh.1d or inla.mesh object, that is then treated as a domain argument instead.
domain	Either when samplers is a 1D interval(s) definition only, domain can be a single integer for the number of integration points to place in each 1D interval, overriding int.args[["nsub1"]], and otherwise when samplers is NULL, domain can be a numeric vector of points, each given integration weight 1 (and no additional points are added in between), an inla.mesh.1d object for continuous 1D integration, or an inla.mesh.2d object for continuous 2D integration.
name	Character array stating the name of the domains dimension(s). If NULL, the names are taken from coordinate names from samplers for ⁠Spatial*⁠ objects, otherwise "x", "y", "z" for 2D regions and "x" for 1D regions
group	Column names of the samplers object (if applicable) for which the integration points are calculated independently and not merged when aggregating to mesh nodes.
int.args	List of arguments passed to bru_int_polygon. method: "stable" (to aggregate integration weights onto mesh nodes) or "direct" (to construct a within triangle/segment integration scheme without aggregating onto mesh nodes) nsub1, nsub2: integers controlling the number of internal integration points before aggregation. Points per triangle: (nsub2+1)^2. Points per knot segment: nsub1 poly_method: if set to "legacy", selects an old polygon integration method that doesn't handle holes. No longer supported, and will generate an error.
project	Deprecated in favour of int.args(method=...). If TRUE, aggregate the integration points to mesh vertices. Default: project = (identical(int.args$method, "stable"))
...	Usually passed on to other methods
na.rm	logical; if TRUE, the rows with weight NA from the non-overlapping full_join will be removed; if FALSE, set the undefined weights to NA. If NULL (default), act as TRUE, but warn if any elements needed removing.
.blockwise	logical; if FALSE, computes full tensor product integration. If TRUE, computes within-block tensor product integration (used internally by fm_int()). Default FALSE
mesh	An inla.mesh object
integ	list of loc, integration points, and weight, integration weights, or a SpatialPointsDataFrame. Only the coordinates and weight column are handled.
tri_subset	Optional triangle index vector for integration on a subset of the mesh triangles (Default NULL)
nsub	number of subdivision points along each triangle edge, giving (nsub + 1)^2 proto-integration points used to compute the vertex weights (default NULL=9, giving 100 integration points for each triangle)
loc	Points in space stored either as data.frame, a two-column matrix of x and y coordinates or a SpatialPoints object.
mesh.coords	Coordinate names of the mesh. Use only if loc is a data.frame with respective column names.
nx	Number of pixels in x direction
ny	Number of pixels in y direction
mask	If logical and TRUE, remove pixels that are outside the mesh. If mask is a Spatial object, only return pixels covered by this object.

Value

ipoints(): A data.frame, tibble, sf, or SpatialPointsDataFrame of 1D and 2D integration points, including a weight column and .block column.

A data.frame, sf, or SpatialPointsDataFrame of multidimensional integration points and their weights

• mesh_triangle_integration returns a list with elements loc and weight with integration points for the mesh

is.inside(): Single column matrix of Boolean values indicating if a point is inside the mesh.

SpatialPixelsDataFrame covering the mesh

Functions

• ipoints(): in favour of fmesher::fm_int()

• cprod(): (Blockwise) cross product of integration points.

  Calculates the groupwise cross product of integration points in different dimensions and multiplies their weights accordingly. If the object defining points in a particular dimension has no weights attached to it all weights are assumed to be 1.

  Legacy wrapper for fm_cprod()

• integration_weight_aggregation(): Aggregate integration weights onto mesh nodes

  Use fmesher::fm_vertex_projection() instead.

• mesh_triangle_integration(): Integration scheme for mesh triangle interiors

  Use fmesher::fm_int_mesh_2d_core() instead.

• bru_mapper(default): Calls bru_mapper_define, passing all arguments along. Mapper implementations should call bru_mapper_define() instead, and supply at least a new_class class name. Use of the bru_mapper.default method was deprecated from version 2.7.0, and removed in version 2.11.0

• bru_mapper_offset(): Creates a bru_mapper_const() mapper.

• is.inside(): Find out which points are inside a mesh. in favour of fm_is_within(). Replace is.inside(mesh, loc) with fm_is_within(loc, mesh).

• vertices.inla.mesh(): Extract vertex locations from an inla.mesh. Converts the vertices of an inla.mesh object into a SpatialPointsDataFrame. Deprecated in favour of fm_vertices()

• pixels(): Generate SpatialPixels covering an inla.mesh. in favour of fmesher::fm_pixels()

Author(s)

Finn Lindgren [e-mail]

See Also

fm_cprod()

fm_is_within()

inlabru documentation built on July 2, 2024, 1:07 a.m.