autoplot.PredictionClust: Plots for Cluster Predictions

In mlr3viz: Visualizations for 'mlr3'

Plots for Cluster Predictions

Description

Visualizations for mlr3cluster::PredictionClust. The argument type controls what kind of plot is drawn. Possible choices are:

• "scatter" (default): scatterplot with correlation values and colored cluster assignments.

• "sil": Silhouette plot with mean silhouette value as the reference line. Requires package ggfortify.

• "pca": Perform PCA on data and color code cluster assignments. Inspired by and uses ggfortify::autoplot.kmeans.

Usage

## S3 method for class 'PredictionClust'
autoplot(
  object,
  task,
  row_ids = NULL,
  type = "scatter",
  theme = theme_minimal(),
  ...
)

Arguments

object	(mlr3cluster::PredictionClust).
task	(mlr3cluster::TaskClust).
row_ids	(integer()) Row ids to subset task data to ensure that only the data used to make predictions are shown in plots.
type	(character(1)): Type of the plot. See description.
theme	(ggplot2::theme()) The ggplot2::theme_minimal() is applied by default to all plots.
...	(ignored).

Value

ggplot2::ggplot().

References

Tang Y, Horikoshi M, Li W (2016). “ggfortify: Unified Interface to Visualize Statistical Result of Popular R Packages.” The R Journal, 8(2), 474–485. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.32614/RJ-2016-060")}.

Examples

if (requireNamespace("mlr3")) {
  library(mlr3)
  library(mlr3cluster)
  library(mlr3viz)

  task = tsk("usarrests")
  learner = lrn("clust.kmeans", centers = 3)
  object = learner$train(task)$predict(task)

  head(fortify(object))
  autoplot(object, task)
}

mlr3viz documentation built on July 1, 2024, 5:06 p.m.