Visualizations for 'mlr3'

Visualization package of the 'mlr3' ecosystem. It features plots for mlr3 objects such as tasks, learners, predictions, benchmark results, tuning instances and filters via the 'autoplot()' generic of 'ggplot2'. The package draws plots with the 'viridis' color palette and applies the minimal theme. Visualizations include barplots, boxplots, histograms, ROC curves, and Precision-Recall curves.

mlr3viz

Package website: release | dev

mlr3viz is the visualization package of the mlr3 ecosystem. It features plots for mlr3 objects such as tasks, learners, predictions, benchmark results, tuning instances and filters via the autoplot() generic of ggplot2. The package draws plots with the viridis color palette and applies the minimal theme. Visualizations include barplots, boxplots, histograms, ROC curves, and Precision-Recall curves.

The gallery features a showcase post of the plots in mlr3viz.

Installation

Install the last release from CRAN:

install.packages("mlr3viz")

Install the development version from GitHub:

# install.packages("pak")
pak::pak("mlr-org/mlr3viz")

Resources

The gallery features a showcase post of the visualization functions mlr3viz.

Short Demo

library(mlr3)
library(mlr3viz)

task = tsk("sonar")
learner = lrn("classif.rpart", predict_type = "prob")
rr = resample(task, learner, rsmp("cv", folds = 3), store_models = TRUE)

# Default plot for task
autoplot(task, type = "target")

# ROC curve for resample result
autoplot(rr, type = "roc")

For more example plots you can have a look at the pkgdown references of the respective functions.