Explain a random forest

Source: R/explain_forest.R

Explains a random forest in a html document using plots created by randomForestExplainer

explain_forest(forest, interactions = FALSE, data = NULL,
  vars = NULL, no_of_pred_plots = 3, pred_grid = 100,
  measures = NULL)

Arguments

forest

A randomForest object created with the option localImp = TRUE
interactions

Logical value: should variable interactions be considered (this may be time-consuming)
data

The data frame on which forest was trained - necessary if interactions = TRUE
vars

A character vector with variables with respect to which interactions will be considered if NULL then they will be selected using the important_variables() function
no_of_pred_plots

The number of most frequent interactions of numeric variables to plot predictions for
pred_grid

The number of points on the grid of plot_predict_interaction (decrease in case memory problems)
measures

A character vector specifying the importance measures to be used for plotting ggpairs

Value

A html document in your working directory

Examples

if (FALSE) {
forest <- randomForest::randomForest(Species ~ ., data = iris, localImp = TRUE)
explain_forest(forest, interactions = TRUE)
}