predict_raster: Make predictions for a whole raster

In tidysdm: Species Distribution Models with Tidymodels

Make predictions for a whole raster

Description

This function allows to use a raster as data to make predictions from a variety of tidymodels objects, such as simple_ensemble or stacks::stacks

Usage

predict_raster(object, raster, ...)

## Default S3 method:
predict_raster(object, raster, filename = "", n = 4, ...)

Arguments

object	the tidymodels object of interest
raster	the terra::SpatRaster or stars with the input data. It has to include levels with the same names as the variables used in object
...	parameters to be passed to the standard predict() function for the appropriate object type (e.g. metric_thresh, class_thresh or fun). See the documentation of the relevant predict() method for possible parameters (e.g. predict.simple_ensemble()).
filename	the name of the output file raster file; this is only needed to save large rasters, and can be left blank for rasters that can be kept in memory.
n	positive integer indicating how many copies the data may be in memory at any point in time (it defaults to 4). This is used to determine whether rasters can be processed in one go, or in chunks. If you get an out of memory error, increase n. See terra::writeStart() for more details.

Value

a terra::SpatRaster (or stars if that is the input) with the predictions

tidysdm documentation built on Dec. 14, 2025, 1:06 a.m.