biomod2
and gstat
packages are added to the Suggests sectioncv_spatial
and spatialBlock
is increased to 100 to make the result matches with v2.1.4column
argumentextend
parameter is now added to spatialBlock
and the function now uses cv_spatial
internally.user_blocks
in cv_spatial
is restricted to random and predefined and systematic selection.raster
package dependencycv_spatial
function for square blocks now matches the one of version 2 function spatialBlock
(i.e. fold assignment starts from top-right corner; this is not the case for hexagon blocks)extend
parameter now.cv_spatial
cv_spatial
to reproducibility of earlier versionscv_nndm
function for large datasetscv_
cv_spatial
, cv_cluster
, cv_buffer
, and cv_nndm
cv_cluster
function generates blocks based on kmeans clustering. It now works on both environmental rasters and the spatial coordinates of sample pointscv_spatial_autocor
function now calculates the spatial autocorrelation range for both the response (i.e. the binary or continuous data) and a set of continuous raster covariatescv_plot
function allows for visualization of folds from all blocking strategies using ggplot facetsterra
package is now used for all raster processing and supports both stars
and raster
objects, as well as files on disk.cv_similarity
provides measures on possible extrapolation to testing foldssf
functions;foreach
to snowfall
;species
argument in spatialBlock
function accepts multi-class responses to find evenly distributed records in train and test folds;spatialAutoRange
function to accepts rasters with low number of pixels; #2maskBySpecies = FALSE
in spatialBlock
function is no longer supported;numLimit
argument in spatialBlock
function is only accepts numeric values, and 0 means searching for evenly distributed folds;speciesData
to spatialAutoRange
function;Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.