Plotting Decision trees

Lucy D’Agostino McGowan

Plotting decision trees

There are several R packages that assist with tree plotting

• rpart.plot
• partykit
• rattle

Where to find out more about packages

1. Vignettes
2. Journal Article (R Journal is great)
3. Rstudio Community
4. StackOverflow
5. Twitter

Learn about partykit

Google partykit to find out how more about it (hint: use key words like rstats and decision tree plots)

03:00

rpart.plot

We’re going to focus on rpart.plot, but feel free to try the others!

install.packages("rpart.plot")

library(rpart.plot)

rpart.plot

tree_spec <- decision_tree(
  cost_complexity = 0.1,
  tree_depth = 10,
  mode = "regression") |>
  set_engine("rpart")

wf <- workflow() |>
  add_recipe(
    recipe(Salary ~ Hits + Years + PutOuts + RBI + Walks + Runs,
                   data = baseball)
  ) |>
  add_model(tree_spec)

model <- fit(wf, baseball)

rpart.plot(model$fit$fit$fit,
           roundint = FALSE)

rpart.plot

rpart.plot(model$fit$fit$fit, 
           roundint = FALSE)

Application Exercise

1. Open your application exercise
2. Install rpart.plot
3. Create a plot from your final decision tree

10:00