In grantmcdermott/parttree: Visualise simple decision tree partitions

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)

parttree

Visualize simple 2-D decision tree partitions in R. The parttree package is optimised to work with ggplot2, although it can be used to visualize tree partitions with base R graphics too.

Installation

This package is not yet on CRAN, but can be installed from GitHub with:

# install.packages("remotes")
remotes::install_github("grantmcdermott/parttree")

Quickstart

The parttree homepage includes an introductory vignette and detailed documentation. But here's a quickstart example using the "kyphosis" dataset that comes bundled with the rpart package. In this case, we are interested in predicting kyphosis recovery after spinal surgery, as a function of 1) the number of topmost vertebra that were operated, and 2) patient age. The key visualization layer below---provided by this package---is geom_partree().

library(rpart)     # For the dataset and fitting decisions trees
library(parttree)  # This package (will automatically load ggplot2 too)

fit = rpart(Kyphosis ~ Start + Age, data = kyphosis)

ggplot(kyphosis, aes(x = Start, y = Age)) +
  geom_parttree(data = fit, alpha = 0.1, aes(fill = Kyphosis)) + # <-- key layer
  geom_point(aes(col = Kyphosis)) +
  labs(
    x = "No. of topmost vertebra operated on", y = "Patient age (months)",
    caption = "Note: Points denote observations. Shading denotes model predictions."
    ) +
  theme_minimal()