importance: Compute variable importance for each resample

In jiho/joml: Utilities for Machine Learning

Compute variable importance for each resample

Description

Compute variable importance for each resample

Usage

importance(object, cores = 1, ...)

Arguments

object	an object output by xgb_fit(), which contains a model column.
cores	integer, number of cores to use for parallel computation.
...	passed to ⁠[xgboost::xgb.importance()]⁠.

Value

A data.frame with variables and their importance metric.

See Also

Other variable importance functions: plot_importance(), summarise_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.