Model Performance of Flashlight

Source: R/light_performance.R

Calculates performance of a flashlight with respect to one or more performance measure.

light_performance(x, ...)

# Default S3 method
light_performance(x, ...)

# S3 method for class 'flashlight'
light_performance(
  x,
  data = x$data,
  by = x$by,
  metrics = x$metrics,
  use_linkinv = FALSE,
  ...
)

# S3 method for class 'multiflashlight'
light_performance(x, ...)

Arguments

x

An object of class "flashlight" or "multiflashlight".

...

Arguments passed from or to other functions.

data

An optional data.frame.

by

An optional vector of column names used to additionally group the results. Will overwrite x$by.

metrics

An optional named list with metrics. Each metric takes at least four arguments: actual, predicted, case weights w and ....

use_linkinv

Should retransformation function be applied? Default is FALSE.

Value

An object of class "light_performance" with the following elements:

  • data: A tibble containing the results.

  • by Same as input by.

Details

The minimal required elements in the (multi-) flashlight are "y", "predict_function", "model", "data" and "metrics". The latter two can also directly be passed to light_performance(). Note that by default, no retransformation function is applied.

Methods (by class)

  • light_performance(default): Default method not implemented yet.

  • light_performance(flashlight): Model performance of flashlight object.

  • light_performance(multiflashlight): Model performance of multiflashlight object.

Examples

fit_part <- lm(Sepal.Length ~ Species + Petal.Length, data = iris)
fl_part <- flashlight(
  model = fit_part, label = "part", data = iris, y = "Sepal.Length"
)
plot(light_performance(fl_part, by = "Species"), fill = "chartreuse4")


# Second model
fit_full <- lm(Sepal.Length ~ ., data = iris)
fl_full <- flashlight(
  model = fit_full, label = "full", data = iris, y = "Sepal.Length"
)
fls <- multiflashlight(list(fl_part, fl_full))

plot(light_performance(fls, by = "Species"))

plot(light_performance(fls, by = "Species"), swap_dim = TRUE)