DESCRIPTION

Package: conftree
Title: Subgroups with Trees using Conformal Prediction
Version: 0.0.1
Authors@R: 
    person("Holger", "Löwe", , "hbj.loewe@gmail.com", role = c("aut", "cre"))
Description: Detect subgroups in machine learning models with numeric outcomes (regression). This leverages CV+/Jackknife+ to simultaneously optimize inter-group heterogeneity and intra-group homogeneity. In this way, it is an extension of the r2p algorithm, see Lee et al. (2020) <arXiv:2006.07917v2>. This works with any of the learning algorithms for regression tasks available in 'tidymodels'.
License: MIT + file LICENSE
Encoding: UTF-8
Roxygen: list(markdown = TRUE)
RoxygenNote: 7.2.3
Depends:
    R (>= 3.6)
Imports:
    checkmate,
    ggparty,
    ggplot2,
    magrittr,
    parsnip,
    partykit,
    rlang,
    rsample
Suggests:
    ranger,
    tidymodels
URL: https://holgstr.github.io/conftree/