plot.light_importance: Visualize Model Importance

Description

Minimal visualization of an object of class light_importance as geom_bar. The object returned is of class ggplot and can be further customized.

Usage

# S3 method for light_importance
plot(x, top_m = Inf, swap_dim = FALSE,
  facet_scales = "fixed", ...)

Arguments

An object of class light_importance.

top_m

Maximum number of important variables to be returned.

swap_dim

If multiflashlight and one "by" variable or single flashlight with two "by" variables, swap the role of dodge/fill variable and facet variable. If multiflashlight or one "by" variable, use facets instead of colors.

facet_scales

Scales argument passed to facet_wrap.

...

Further arguments passed to geom_bar.

Value

An object of class ggplot2.

Details

The plot is organized as a bar plot with variable names as x-aesthetic. Up to two additional dimensions (multiflashlight and one "by" variable or single flashlight with two "by" variables) can be visualized by facetting and dodge/fill. Set swap_dim = FALSE to revert the role of these two dimensions. One single additional dimension is visualized by a facet wrap, or - if swap_dim = FALSE - by dodge/fill.

Examples

Run this code

# NOT RUN {
fit_full <- lm(Sepal.Length ~ ., data = iris)
fit_part <- lm(Sepal.Length ~ Petal.Length, data = iris)
mod_full <- flashlight(model = fit_full, label = "full", data = iris, y = "Sepal.Length")
mod_part <- flashlight(model = fit_part, label = "part", data = iris, y = "Sepal.Length")
mods <- multiflashlight(list(mod_full, mod_part), by = "Species")

plot(light_importance(mod_full), fill = "darkred")
plot(light_importance(mod_full, variable_name = "v", label_name = "model",
  metric_name = "m", value_name = "drop"))
plot(light_importance(mod_full), top_m = 2)
plot(light_importance(mods))
plot(light_importance(mods), swap_dim = TRUE)
plot(light_importance(mods, by = NULL), fill = "darkgreen")
plot(light_importance(mods, by = NULL), swap_dim = TRUE)
# }