adamlilith/blockCV: Spatial and environmental blocking for k-fold cross-validation
The blockCV package creates spatially or environmentally separated training and testing folds for cross-validation to provide a robust error estimation in spatially structured environments. It also investigates the effective range of spatial autocorrelation in continuous raster covariates to find an initial realistic distance band to separate training and testing datasets spatially. The distance band can be used for spatial blocking or buffering to create folds.
Getting started
Browse package contents
Package details |
|
|---|---|
| Author | Roozbeh Valavi, Jane Elith, José Lahoz-Monfort and Gurutzeta Guillera-Arroita |
| Maintainer | Roozbeh Valavi <valavi.r@gmail.com> |
| License | GPL-3 |
| Version | 1.0.0 |
| URL | https://github.com/rvalavi/blockCV |
| Package repository | View on GitHub |
| Installation |
Install the latest version of this package by entering the following in R:
|
