João Bonifácio
  • Home
  • About
  • R Packages
  • Shiny Apps

R Packages

Open-source R packages published on CRAN and GitHub.

R Packages

Open-source R packages developed for statistical modeling, stream monitoring, and efficient data access.

deriva CRAN R Package

Tidy Drift Detection for Monitored Machine Learning Models

Detects concept drift and data drift in streams produced by deployed machine learning models, using a tidy interface that composes with the tidymodels ecosystem. Detectors are specified, fitted on a baseline period, and advanced over new batches of observations, returning tibbles annotated with warning and drift flags.

Installation

# Official CRAN release
install.packages("deriva")

# Development version from GitHub
# install.packages("remotes")
remotes::install_github("bonijoao/deriva")

CRAN Package Page → GitHub Repository

tidyprf CRAN R Package

Tidy Access to Brazilian Federal Highway Police (‘PRF’) Data

Downloads and reads Brazilian Federal Highway Police (‘PRF’) open data on traffic accidents by person, by occurrence, and traffic violations. Designed around tidyverse principles for fast and reproducible data analysis.

Installation

# Official CRAN release
install.packages("tidyprf")

# Development version from GitHub
# install.packages("remotes")
remotes::install_github("bonijoao/tidyprf")

CRAN Package Page → GitHub Repository

figsr GitHub R Package

Fast Interpretable Greedy-Tree Sums for Tree Ensembles

Implements FIGS, an additive tree-sum algorithm that greedily reduces residual impurity, deciding at each iteration whether to grow a new tree or deepen an existing one. It reaches accuracy comparable to random forests while keeping the fitted model readable as IF-THEN rules. Supports regression and binary classification, variable importance, bagging, and integrates with parsnip and tidymodels.

Installation

# Development version from GitHub
# install.packages("remotes")
remotes::install_github("bonijoao/figsr")

GitHub Repository →