Package: spatialRF 1.1.5

spatialRF: Easy Spatial Modeling with Random Forest

Automatic generation and selection of spatial predictors for Random Forest models fitted to spatially structured data. Spatial predictors are constructed from a distance matrix among training samples using Moran's Eigenvector Maps (MEMs; Dray, Legendre, and Peres-Neto 2006 <doi:10.1016/j.ecolmodel.2006.02.015>) or the RFsp approach (Hengl et al. <doi:10.7717/peerj.5518>). These predictors are used alongside user-supplied explanatory variables in Random Forest models. The package provides functions for model fitting, multicollinearity reduction, interaction identification, hyperparameter tuning, evaluation via spatial cross-validation, and result visualization using partial dependence and interaction plots. Model fitting relies on the 'ranger' package (Wright and Ziegler 2017 <doi:10.18637/jss.v077.i01>).

Authors:Blas M. Benito [aut, cre, cph]

spatialRF_1.1.5.tar.gz
spatialRF_1.1.5.zip(r-4.7)spatialRF_1.1.5.zip(r-4.6)spatialRF_1.1.5.zip(r-4.5)
spatialRF_1.1.5.tgz(r-4.6-any)spatialRF_1.1.5.tgz(r-4.5-any)
spatialRF_1.1.5.tar.gz(r-4.7-any)spatialRF_1.1.5.tar.gz(r-4.6-any)
spatialRF_1.1.5.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
spatialRF/json (API)

# Install 'spatialRF' in R:

install.packages('spatialRF', repos = c('https://blasbenito.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/blasbenito/spatialrf/issues

Pkgdown/docs site:https://blasbenito.github.io

Datasets:

plants_df - Plant richness and predictors for American ecoregions
plants_distance - Distance matrix between ecoregion edges
plants_predictors - Predictor variable names for plant richness examples
plants_response - Response variable name for plant richness examples
plants_rf - Example fitted random forest model
plants_rf_spatial - Example fitted spatial random forest model
plants_xy - Coordinates for plant richness data

On CRAN:

random-forest spatial-analysis spatial-regression

6.42 score 124 stars 85 scripts 531 downloads 66 exports 52 dependencies

Last updated from:b7ab6291d6. Checks:9 OK. Indexed: yes.

Target	Result	Time
linux-devel-x86_64	OK	349
source / vignettes	OK	298
linux-release-x86_64	OK	355
macos-release-arm64	OK	187
macos-oldrel-arm64	OK	195
windows-devel	OK	269
windows-release	OK	263
windows-oldrel	OK	237
wasm-release	OK	226

Exports:.vif_to_df %>%auc auto_cor auto_vif beowulf_cluster case_weights default_distance_thresholds double_center_distance_matrix filter_spatial_predictors get_evaluation get_importance get_importance_local get_moran get_performance get_predictions get_residuals get_response_curves get_spatial_predictors is_binary make_spatial_fold make_spatial_folds mem mem_multithreshold moran moran_multithreshold objects_size optimization_function pca pca_multithreshold plot_evaluation plot_importance plot_moran plot_optimization plot_residuals_diagnostics plot_response_curves plot_response_surface plot_training_df plot_training_df_moran plot_tuning prepare_importance_spatial print_evaluation print_importance print_moran print_performance rank_spatial_predictors rescale_vector residuals_diagnostics residuals_test rf rf_compare rf_evaluate rf_importance rf_repeat rf_spatial rf_tuning root_mean_squared_error select_spatial_predictors_recursive select_spatial_predictors_sequential setup_parallel_execution standard_error statistical_mode the_feature_engineer thinning thinning_til_n weights_from_distance_matrix

Dependencies:assertthat base64enc cachem cli codetools commonmark cpp11 digest doParallel dplyr fansi farver fastmap foreach generics ggplot2 glue gridExtra gtable htmltools huxtable isoband iterators labeling lattice lifecycle magrittr Matrix memoise patchwork pillar pkgconfig purrr R6 ranger RColorBrewer Rcpp RcppEigen rlang S7 scales stringi stringr tibble tidyr tidyselect utf8 vctrs viridis viridisLite withr xml2

Help page	Topics
Convert VIF values to data frame	.vif_to_df
Area under the ROC curve	auc
Multicollinearity reduction via Pearson correlation	auto_cor
Multicollinearity reduction via Variance Inflation Factor	auto_vif
Create a Beowulf cluster for parallel computing	beowulf_cluster
Generate case weights for imbalanced binary data	case_weights
Default distance thresholds for spatial predictors	default_distance_thresholds
Double-center a distance matrix	double_center_distance_matrix
Remove redundant spatial predictors	filter_spatial_predictors
Extract evaluation metrics from cross-validated model	get_evaluation
Extract variable importance from model	get_importance
Extract local variable importance from model	get_importance_local
Extract Moran's I test results for model residuals	get_moran
Extract out-of-bag performance metrics from model	get_performance
Extract fitted predictions from model	get_predictions
Extract model residuals	get_residuals
Extract response curve data for plotting	get_response_curves
Extract spatial predictors from spatial model	get_spatial_predictors
Check if variable is binary with values 0 and 1	is_binary
Create spatially independent training and testing folds	make_spatial_fold
Create multiple spatially independent training and testing folds	make_spatial_folds
Compute Moran's Eigenvector Maps from distance matrix	mem
Compute Moran's Eigenvector Maps across multiple distance thresholds	mem_multithreshold
Moran's I test for spatial autocorrelation	moran
Moran's I test across multiple distance thresholds	moran_multithreshold
Display sizes of objects in current R environment	objects_size
Compute optimization scores for spatial predictor selection	optimization_function
Compute Principal Component Analysis	pca
Compute Principal Component Analysis at multiple distance thresholds	pca_multithreshold
Plant richness and predictors for American ecoregions	plants_df
Distance matrix between ecoregion edges	plants_distance
Predictor variable names for plant richness examples	plants_predictors
Response variable name for plant richness examples	plants_response
Example fitted random forest model	plants_rf
Example fitted spatial random forest model	plants_rf_spatial
Coordinates for plant richness data	plants_xy
Visualize spatial cross-validation results	plot_evaluation
Visualize variable importance scores	plot_importance
Plots a Moran's I test of model residuals	plot_moran
Optimization plot of a selection of spatial predictors	plot_optimization
Plot residuals diagnostics	plot_residuals_diagnostics
Plots the response curves of a model.	plot_response_curves
Plots the response surfaces of a random forest model	plot_response_surface
Scatterplots of a training data frame	plot_training_df
Moran's I plots of a training data frame	plot_training_df_moran
Plots a tuning object produced by 'rf_tuning()'	plot_tuning
Prepares variable importance objects for spatial models	prepare_importance_spatial
Prints cross-validation results	print_evaluation
Prints variable importance	print_importance
Prints results of a Moran's I test	print_moran
print_performance	print_performance
Custom print method for random forest models	print.rf
Ranks spatial predictors	rank_spatial_predictors
Rescales a numeric vector into a new range	rescale_vector
Normality test of a numeric vector	residuals_diagnostics
Normality test of a numeric vector	residuals_test
Random forest models with Moran's I test of the residuals	rf
Compares models via spatial cross-validation	rf_compare
Evaluates random forest models with spatial cross-validation	rf_evaluate
Contribution of each predictor to model transferability	rf_importance
Fits several random forest models on the same data	rf_repeat
Fits spatial random forest models	rf_spatial
Tuning of random forest hyperparameters via spatial cross-validation	rf_tuning
RMSE and normalized RMSE	root_mean_squared_error
Finds optimal combinations of spatial predictors	select_spatial_predictors_recursive
Sequential introduction of spatial predictors into a model	select_spatial_predictors_sequential
Setup parallel execution with automatic backend detection	setup_parallel_execution
Standard error of the mean of a numeric vector	standard_error
Statistical mode of a vector	statistical_mode
Suggest variable interactions and composite features for random forest models	the_feature_engineer
Applies thinning to pairs of coordinates	thinning
Applies thinning to pairs of coordinates until reaching a given n	thinning_til_n
Transforms a distance matrix into a matrix of weights	weights_from_distance_matrix

Package: spatialRF 1.1.5

spatialRF: Easy Spatial Modeling with Random Forest

Citation

Development and contributors

Readme and manuals

Help Manual

Usage by other packages (reverse dependencies)