soilcnn: Convolutional Neural Networks for Digital Soil Mapping

A framework, built on 'torch', for mapping a soil property from point observations and a stack of aligned rasters with a convolutional network that reads the neighbourhood of each point. It extracts the patches; resamples by spatial blocks or by k-fold nearest neighbour distance matching, after Linnenbrink et al. (2024) <doi:10.5194/gmd-17-5897-2024>; tunes and selects under a noise floor and the one-standard-error rule; refits the selected configuration under several seeds; and predicts median, mean and interval maps over rasters of any extent. The map's centre is the ensemble median over seeds; the mean corrects its back-transformation by smearing, after Duan (1983) <doi:10.1080/01621459.1983.10478017>. The intervals are conformal, calibrated three ways and each checked on the test set: on the cross-validated residuals; on a calibration set held out of every fold, as split conformal, after Lei et al. (2018) <doi:10.1080/01621459.2017.1307116>; or with every fold's model at each pixel, as CV+, after Barber et al. (2021) <doi:10.1214/20-AOS1965> – with points, or whole groups of them, weighing alike, after Dunn et al. (2023) <doi:10.1080/01621459.2022.2060112>. The area of applicability of Meyer and Pebesma (2021) <doi:10.1111/2041-210X.13650> is mapped beside them. What a model learned is read by permutation, by SHAP values from expected gradients, after Erion et al. (2021) <doi:10.1038/s42256-021-00343-w>, by SAGE, after Covert et al. (2020) <doi:10.48550/arXiv.2004.00668>, by accumulated local effects, after Apley and Zhu (2020) <doi:10.1111/rssb.12377>, and by refits without each variable. Random forest and multilayer perceptron baselines run under the same folds.

Version:	0.1.0
Depends:	R (≥ 4.5.0)
Imports:	callr, coro, dplyr, graphics, grDevices, janitor, matrixStats, parallel, ps, purrr, readr, stats, terra, tibble, tools, torch, utils
Suggests:	caret, CAST, FNN, knitr, pkgload, randomForest, ranger, rmarkdown, sf, testthat (≥ 3.0.0)
Published:	2026-10-21
DOI:	10.32614/CRAN.package.soilcnn (may not be active yet)
Author:	Cássio Marques Moquedace [aut, cre, cph], Clara Glória Oliveira Baldi [aut, cph]
Maintainer:	Cássio Marques Moquedace <cassiomoquedace at gmail.com>
BugReports:	https://github.com/moquedace/soilcnn/issues
License:	MIT + file LICENSE
URL:	https://github.com/moquedace/soilcnn, https://moquedace.github.io/soilcnn/
NeedsCompilation:	no
Language:	en-GB
Citation:	soilcnn citation info
Materials:	NEWS
CRAN checks:	soilcnn results

Documentation:

Reference manual:	soilcnn.html , soilcnn.pdf
Vignettes:	Getting started with soilcnn (source, R code)

Downloads:

Package source:	soilcnn_0.1.0.tar.gz
Windows binaries:	r-devel: not available, r-release: not available, r-oldrel: not available
macOS binaries:	r-release (arm64): soilcnn_0.1.0.tgz, r-oldrel (arm64): soilcnn_0.1.0.tgz, r-release (x86_64): soilcnn_0.1.0.tgz, r-oldrel (x86_64): soilcnn_0.1.0.tgz

Linking:

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