ggForestEffects: Dot plot of covariates effects

In moreparty: A Toolbox for Conditional Inference Trees and Random Forests

Dot plot of covariates effects

Description

Plots the effects (partial dependence or accumulated local effects) of the covariates of a supervised learning model in a single a dot plot.

Usage

ggForestEffects(dt, vline=0, xlabel="", ylabel="", main="")

Arguments

dt	data frame. Must have three columns : one with the names of the covariates (named "var"), one with the names of the categories of the covariates (named "cat"), one with the values of the effects (named "value"). Typically the result of GetAleData or GetPartialData functions.
vline	numeric. Coordinate on the x axis where a vertical line is added.
xlabel	character. Title of the x axis.
ylabel	character. Title of the y axis.
main	character. Title of the plot.

Note

There should be no duplicated categories. If it is the case, duplicated categories have to be renamed in dt prior to running ggForestEffects.

Author(s)

Nicolas Robette

References

Apley, D. W., Zhu J. "Visualizing the Effects of Predictor Variables in Black Box Supervised Learning Models". arXiv:1612.08468v2, 2019.

Molnar, Christoph. "Interpretable machine learning. A Guide for Making Black Box Models Explainable", 2019. https://christophm.github.io/interpretable-ml-book/.

See Also

GetAleData, GetPartialData

Examples

  ## Not run: 
  data(iris)
  iris2 = iris
  iris2$Species = factor(iris$Species == "versicolor")
  iris.cf = party::cforest(Species ~ ., data = iris2, controls = cforest_unbiased(mtry=2))
  ale <- GetAleData(iris.cf)
  ale$cat <- paste(ale$var,ale$cat,sep='_')  # to avoid duplicated categories
  ggForestEffects(ale)

## End(Not run)

moreparty documentation built on Nov. 22, 2023, 5:08 p.m.