Transform input into a probability mass function (PMF)

Converts raster or vector inputs into a probability mass function where all values sum to 1. Raster inputs are normalised directly, while vector inputs are first rasterised using a supplied template.

Usage

transform_pmf(
  x,
  value_col = NULL,
  template = NULL,
  return_total = FALSE,
  snap = FALSE
)

Arguments

x

A `SpatRaster` containing values to normalise, or a `SpatVector` whose attribute values will be rasterised.

value_col

Character name of the attribute column to use when `x` is a `SpatVector`. Ignored for raster inputs.

template

`SpatRaster` providing geometry for rasterising vector inputs. Required when `x` is a `SpatVector`.

return_total

Logical; when `TRUE`, return both the PMF raster and the total sum of the input values. Default is `FALSE`.

snap

Logical; skip validation checks when `TRUE`. Default is `FALSE`.

Value

If `return_total = FALSE`, returns a `SpatRaster` whose cells sum to 1. If `return_total = TRUE`, returns a list with components:

pmf

`SpatRaster` of probabilities

total

Numeric total of the original values

Examples

if (FALSE) { # \dontrun{
pop <- terra::rast(u5pd)

# Basic PMF
pmf <- transform_pmf(pop)

# Include total for later reuse
result <- transform_pmf(pop, return_total = TRUE)
result$total
} # }