Fast Partial Receiver Operating Characteristic (ROC) Test for Ecological Niche Modeling

Provides optimized 'C++' code for computing the partial Receiver Operating Characteristic (ROC) test used in niche and species distribution modeling. The implementation follows Peterson et al. (2008) . Parallelization via 'OpenMP' was implemented with assistance from the 'DeepSeek' Artificial Intelligence Assistant (< https://www.deepseek.com/>).

fpROC

Provides optimized ‘C++’ code for computing the partial Receiver Operating Characteristic (ROC) test used in niche and species distribution modeling. The implementation follows Peterson et al. (2008) doi:10.1016/j.ecolmodel.2007.11.008. Parallelization via ‘OpenMP’ was implemented with assistance from the ‘DeepSeek’ Artificial Intelligence Assistant (https://www.deepseek.com/).

Installation

You can install the development version of fpROC from GitHub with:

# install.packages("pak")
pak::pak("luismurao/fpROC")

Examples

The package can work with numerical vectors and terra SpatRaster objects.

An example using numerical data

set.seed(999)
# With numeric vectors
test_data <- rnorm(100)
pred_data <- rnorm(100)
result <- fpROC::auc_metrics(test_prediction = test_data, prediction = pred_data)

An example using terra SpatRaster objects.

set.seed(999)
# With SpatRaster
library(terra)
#> terra 1.8.54
r <- terra::rast(ncol=10, nrow=10)
values(r) <- rnorm(terra::ncell(r))
result <- fpROC::auc_metrics(test_prediction = test_data, prediction = r)

Acknowledgments

CONACYT Ciencia de Frontera CF-2023-I-1156. Laboratorio Nacional de Biología del Cambio Climático, SECIHTI, México. To PAPIIT-UNAM IA202824 and PAPIIT-UNAM IA203922.RGC-D thanks the Dirección General de Asuntos del Personal Académico (DGAPA) from the UNAM, and the Secretaría de Ciencia, Humanidades, Tecnología e Innovación for her postdoctoral scholarship.