SpatialML: Spatial Machine Learning

Implements a spatial extension of the random forest algorithm (Georganos et al. (2019) <doi:10.1080/10106049.2019.1595177>). Provides a Geographically Weighted Random Forest regression and a routine to find the optimal bandwidth (Georganos and Kalogirou (2022) <doi:10.3390/ijgi11090471>). A lightweight cross-validation helper for tuning the 'mtry' parameter of a random forest and a generator of synthetic spatial test data are also included. The package depends on 'ranger' as its single random-forest back-end.

Version:	1.8.2
Depends:	R (≥ 4.1.0)
Imports:	ranger (≥ 0.15.1), stats, graphics, utils
Suggests:	knitr, markdown, testthat (≥ 3.0.0)
Published:	2026-07-06
DOI:	10.32614/CRAN.package.SpatialML
Author:	Stamatis Kalogirou [aut, cre, cph], Stefanos Georganos [aut]
Maintainer:	Stamatis Kalogirou <stamatis.science at gmail.com>
License:	GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
URL:	https://stamatisgeoai.eu/
NeedsCompilation:	no
Language:	en-GB
Materials:	NEWS
CRAN checks:	SpatialML results

Documentation:

Reference manual:	SpatialML.html , SpatialML.pdf
Vignettes:	Geographically Weighted Random Forest with SpatialML (source, R code)

Downloads:

Package source:	SpatialML_1.8.2.tar.gz
Windows binaries:	r-devel: SpatialML_1.8.2.zip, r-release: SpatialML_1.8.2.zip, r-oldrel: SpatialML_1.8.2.zip
macOS binaries:	r-release (arm64): SpatialML_1.8.2.tgz, r-oldrel (arm64): SpatialML_1.8.2.tgz, r-release (x86_64): SpatialML_1.8.2.tgz, r-oldrel (x86_64): SpatialML_1.8.2.tgz
Old sources:	SpatialML archive

Linking:

Please use the canonical form https://CRAN.R-project.org/package=SpatialML to link to this page.