pred_spec_auto: Automatic variable values for dependence

In aorsf: Accelerated Oblique Random Forests

Automatic variable values for dependence

Description

For partial dependence and individual conditional expectations, this function allows a variable to be considered without having to specify what values to set the variable at. The values used are based on quantiles for continuous variables (10th, 25th, 50th, 75th, and 90th) and unique categories for categorical variables.

Usage

pred_spec_auto(...)

Arguments

...	names of the variables to use. These can be in quotes or not in quotes (see examples).

Details

This function should only be used in the context of orsf_pd or orsf_ice functions.

Value

a character vector with the names

Examples

fit <- orsf(penguins_orsf, species ~., n_tree = 5)

orsf_pd_oob(fit, pred_spec_auto(flipper_length_mm))

aorsf documentation built on June 22, 2024, 10:31 a.m.