Using `coefplot` • coefplot

library(coefplot)
#> Loading required package: ggplot2

`coefplot()`

coefplot provides S3 Methods for lm, glm, workflows, model_fit, rxGlm, rxLinMod, and rxLogit objects. These methods take the same arguments as the default coefplot() method.

`coefplot.lm()`

lm_model <- lm(mpg ~ wt + cyl*disp, data = mtcars)
coefplot(lm_model)

`coefplot.glm()`

glm_model <- 
  glm(
    mpg > 15 ~ wt + cyl*disp, 
    data = mtcars, 
    family = binomial(link = "logit")
  )
coefplot(glm_model)

coefplot(glm_model, trans = invlogit)

`coefplot.workflow()`

coefplot.workflow() plots the coefficients of a fitted workflows::workflow() object. Bceause this example is started with a workflows::workflow() call, it has the workflow class and thus uses coefplot.workflow().

workflows::workflow() |>
  workflows::add_formula(mpg ~ wt + cyl*disp) |>
  workflows::add_model(parsnip::linear_reg()) |>
  parsnip::fit(mtcars) |>
  coefplot()

`coefplot.model_fit()`

While coefplot.model_fit() passes directly from parsnip::fit() to coefplot() like in the example for coefplot.workflow(), because this example is contained entirely within parsnip calls, it has the class model_fit.

parsnip::linear_reg() |>
  parsnip::set_engine('lm') |>
  parsnip::fit(mpg ~ wt + cyl*disp, data = mtcars) |>
  coefplot()