Package: midr
Type: Package
Title: Learning from Black-Box Models by Maximum Interpretation
        Decomposition
Version: 0.5.2
Authors@R: c(
    person("Ryoichi", "Asashiba", email = "ryoichi.asashiba@gmail.com",
           role = c("aut", "cre")),
    person("Hirokazu", "Iwasawa", role = c("aut")),
    person("Reiji", "Kozuma", role = c("ctb"))
  )
Description: 
  The goal of 'midr' is to provide a model-agnostic method for interpreting and explaining black-box predictive models by creating a globally interpretable surrogate model. The package implements 'Maximum Interpretation Decomposition' (MID), a functional decomposition technique that finds an optimal additive approximation of the original model. This approximation is achieved by minimizing the squared error between the predictions of the black-box model and the surrogate model. The theoretical foundations of MID are described in Iwasawa & Matsumori (2025) [Forthcoming], and the package itself is detailed in Asashiba et al. (2025) <doi:10.48550/arXiv.2506.08338>.
License: MIT + file LICENSE
Encoding: UTF-8
Imports: graphics, grDevices, Rcpp, RcppEigen, rlang, stats, utils
Suggests: datasets, ggplot2, khroma, knitr, RColorBrewer, rmarkdown,
        scales, shapviz, testthat, viridisLite
Config/testthat/edition: 3
RoxygenNote: 7.3.2
URL: https://github.com/ryo-asashi/midr,
        https://ryo-asashi.github.io/midr/
BugReports: https://github.com/ryo-asashi/midr/issues
LinkingTo: Rcpp, RcppEigen
NeedsCompilation: yes
Packaged: 2025-09-07 09:16:45 UTC; daysb
Author: Ryoichi Asashiba [aut, cre],
  Hirokazu Iwasawa [aut],
  Reiji Kozuma [ctb]
Maintainer: Ryoichi Asashiba <ryoichi.asashiba@gmail.com>
Repository: CRAN
Date/Publication: 2025-09-07 22:00:30 UTC
Built: R 4.4.1; aarch64-apple-darwin20; 2025-09-07 22:32:13 UTC; unix
Archs: midr.so.dSYM
