Learn R Programming

spatialRF (version 1.1.5)

Easy Spatial Modeling with Random Forest

Description

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 ) or the RFsp approach (Hengl et al. ). 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 ).

Copy Link

Version

Install

install.packages('spatialRF')

Monthly Downloads

439

Version

1.1.5

License

MIT + file LICENSE

Maintainer

Blas M. Benito

Last Published

December 19th, 2025

Functions in spatialRF (1.1.5)

pca_multithreshold

Compute Principal Component Analysis at multiple distance thresholds

Check if variable is binary with values 0 and 1

Compute Principal Component Analysis

plants_response

Response variable name for plant richness examples

moran_multithreshold

Moran's I test across multiple distance thresholds

optimization_function

Compute optimization scores for spatial predictor selection

Display sizes of objects in current R environment

Normality test of a numeric vector

Example fitted random forest model

Coordinates for plant richness data

plants_rf_spatial

Example fitted spatial random forest model

plot_evaluation

Visualize spatial cross-validation results

plot_importance

Visualize variable importance scores

plot_residuals_diagnostics

Plot residuals diagnostics

Plots a Moran's I test of model residuals

make_spatial_fold

Create spatially independent training and testing folds

get_predictions

Extract fitted predictions from model

mem_multithreshold

Compute Moran's Eigenvector Maps across multiple distance thresholds

plot_training_df_moran

Moran's I plots of a training data frame

Extract model residuals

plot_optimization

Optimization plot of a selection of spatial predictors

plot_response_curves

Plots the response curves of a model.

Plots a tuning object produced by rf_tuning()

print_performance

print_performance

Prints results of a Moran's I test

residuals_diagnostics

Normality test of a numeric vector

Random forest models with Moran's I test of the residuals

make_spatial_folds

Create multiple spatially independent training and testing folds

Compute Moran's Eigenvector Maps from distance matrix

prepare_importance_spatial

Prepares variable importance objects for spatial models

Evaluates random forest models with spatial cross-validation

Plant richness and predictors for American ecoregions

Compares models via spatial cross-validation

Fits spatial random forest models

Moran's I test for spatial autocorrelation

plot_response_surface

Plots the response surfaces of a random forest model

plants_distance

Distance matrix between ecoregion edges

rank_spatial_predictors

Ranks spatial predictors

plot_training_df

Scatterplots of a training data frame

Rescales a numeric vector into a new range

Contribution of each predictor to model transferability

plants_predictors

Predictor variable names for plant richness examples

statistical_mode

Statistical mode of a vector

Standard error of the mean of a numeric vector

Custom print method for random forest models

print_evaluation

Prints cross-validation results

Fits several random forest models on the same data

select_spatial_predictors_sequential

Sequential introduction of spatial predictors into a model

weights_from_distance_matrix

Transforms a distance matrix into a matrix of weights

Applies thinning to pairs of coordinates until reaching a given n

Tuning of random forest hyperparameters via spatial cross-validation

setup_parallel_execution

Setup parallel execution with automatic backend detection

select_spatial_predictors_recursive

Finds optimal combinations of spatial predictors

print_importance

Prints variable importance

root_mean_squared_error

RMSE and normalized RMSE

Applies thinning to pairs of coordinates

the_feature_engineer

Suggest variable interactions and composite features for random forest models

filter_spatial_predictors

Remove redundant spatial predictors

Extract evaluation metrics from cross-validated model

Area under the ROC curve

Multicollinearity reduction via Pearson correlation

Multicollinearity reduction via Variance Inflation Factor

beowulf_cluster

Create a Beowulf cluster for parallel computing

Extract Moran's I test results for model residuals

get_performance

Extract out-of-bag performance metrics from model

Generate case weights for imbalanced binary data

get_response_curves

Extract response curve data for plotting

get_spatial_predictors

Extract spatial predictors from spatial model

default_distance_thresholds

Default distance thresholds for spatial predictors

Convert VIF values to data frame

double_center_distance_matrix

Double-center a distance matrix

Extract variable importance from model

get_importance_local

Extract local variable importance from model