Skip to contents

FIGS Model Specification for Parsnip

Source: R/figs_tree.R
figs_tree.Rd

figs_tree() defines a Fast Interpretable Greedy-Tree Sums model for use with the parsnip and tidymodels ecosystem.

Usage

figs_tree(
  mode = "regression",
  engine = "figsr",
  max_splits = NULL,
  max_trees = NULL,
  min_n = NULL
)

Arguments

mode

A single character string for the prediction outcome mode: "regression" or "classification". Classification supports two-class outcomes only; a factor with more than two levels raises an error at fit time.

engine

A single character string for the computational engine. Only "figsr" is available.

max_splits

An integer for the maximum total splits across all trees. NULL (the default) leaves the engine default of 10 in place.

max_trees

An integer for the maximum number of trees. NULL (the default) leaves the number of trees unconstrained.

min_n

An integer for the minimum number of data points in a node to split. NULL (the default) leaves the engine default of 5 in place.

Value

A parsnip model specification object.

Examples

library(parsnip)
spec <- figs_tree(max_splits = 8) |>
  set_engine("figsr") |>
  set_mode("regression")
spec
#> figs tree Model Specification (regression)
#> 
#> Main Arguments:
#>   max_splits = 8
#> 
#> Computational engine: figsr 
#>