butcher (version 0.3.5)

axe-lm: Axing an lm.

Description

lm objects are created from the base stats package.

Usage

# S3 method for lm
axe_call(x, verbose = FALSE, ...)
# S3 method for lm
axe_env(x, verbose = FALSE, ...)
# S3 method for lm
axe_fitted(x, verbose = FALSE, ...)

Value

Axed lm object.

Arguments

x

A model object.

verbose

Print information each time an axe method is executed. Notes how much memory is released and what functions are disabled. Default is FALSE.

...

Any additional arguments related to axing.

Examples

# Load libraries
library(parsnip)
library(rsample)

# Load data
split <- initial_split(mtcars, prop = 9/10)
car_train <- training(split)

# Create model and fit
lm_fit <- linear_reg() %>%
  set_engine("lm") %>%
  fit(mpg ~ ., data = car_train)

out <- butcher(lm_fit, verbose = TRUE)

# Another lm object
wrapped_lm <- function() {
  some_junk_in_environment <- runif(1e6)
  fit <- lm(mpg ~ ., data = mtcars)
  return(fit)
}

# Remove junk
cleaned_lm <- axe_env(wrapped_lm(), verbose = TRUE)

# Check size
lobstr::obj_size(cleaned_lm)

# Compare environment in terms component
lobstr::obj_size(attr(wrapped_lm()$terms, ".Environment"))
lobstr::obj_size(attr(cleaned_lm$terms, ".Environment"))