summarise_importance: Summarise variable importance across resamples

In jiho/joml: Utilities for Machine Learning

Summarise variable importance across resamples

Description

Summarise variable importance across resamples

Usage

summarise_importance(object)

Arguments

object

an object output by importance(), which contains an importance column.

Value

A data.frame with variables and the mean and sd of their importance metrics.

See Also

Other variable importance functions: importance(), plot_importance()

Examples

# fit a model on five bootstraps
m <- resample_boot(mtcars, 5) %>%
  xgb_fit(resp="mpg", expl=c("cyl", "hp", "qsec"),
    eta=0.1, max_depth=4, nrounds=20)
# compute variable importance for each model
m <- importance(m)
# look at importance for the first resample
m$importance[[1]]
# summarise across resamples
summarise_importance(m)
# plot the summarised importance
plot_importance(m)

jiho/joml documentation built on Dec. 6, 2023, 5:50 a.m.