BORG: Bounded Outcome Risk Guard for Model Evaluation
Automatically detects and enforces valid model evaluation by identifying information reuse between training and evaluation data. Guards against data leakage, look-ahead bias, and invalid cross-validation schemes that inflate performance estimates. Supports temporal, spatial, and grouped evaluation structures. Based on evaluation principles described in Roberts et al. (2017) <doi:10.1111/ecog.02881>, Kaufman et al. (2012) <doi:10.1145/2382577.2382579>, and Kapoor & Narayanan (2023) <doi:10.1016/j.patter.2023.100804>.
| Version: | 0.2.5 |
| Depends: | R (≥ 3.5) |
| Imports: | Rcpp, methods, stats, utils |
| LinkingTo: | Rcpp |
| Suggests: | caret, rsample, tidyselect, recipes, mlr3, sf, ranger, parsnip, workflows, xgboost, lightgbm, testthat (≥ 3.0.0), knitr, rmarkdown |
| Published: | 2026-03-20 |
| DOI: | 10.32614/CRAN.package.BORG (may not be active yet) |
| Author: | Gilles Colling 
[aut, cre, cph] |
| Maintainer: | Gilles Colling <gilles.colling051 at gmail.com> |
| BugReports: | https://github.com/gcol33/BORG/issues |
| License: | MIT + file LICENSE |
| URL: | https://github.com/gcol33/BORG, https://gillescolling.com/BORG/ |
| NeedsCompilation: | yes |
| Language: | en-US |
| Citation: | BORG citation info |
| Materials: | README, NEWS |
| CRAN checks: | BORG results |
Documentation:
| Reference manual: | BORG.html , BORG.pdf |
| Vignettes: |
Framework Integration (source, R code) Quick Start (source, R code) Risk Taxonomy (source, R code) |
Downloads:
| Package source: | BORG_0.2.5.tar.gz |
| Windows binaries: | r-devel: not available, r-release: not available, r-oldrel: not available |
| macOS binaries: | r-release (arm64): not available, r-oldrel (arm64): BORG_0.2.5.tgz, r-release (x86_64): BORG_0.2.5.tgz, r-oldrel (x86_64): BORG_0.2.5.tgz |
Linking:
Please use the canonical form https://CRAN.R-project.org/package=BORG to link to this page.