recipes (version 0.1.6)

update.step: Update a recipe step

Description

This step method for update() takes named arguments as ... who's values will replace the elements of the same name in the actual step.

Usage

# S3 method for step
update(object, ...)

Arguments

object

A recipe step.

...

Key-value pairs where the keys match up with names of elements in the step, and the values are the new values to update the step with.

Details

For a step to be updated, it must not already have been trained. Otherwise, conflicting information can arise between the data returned from juice() and the information in the step.

This function is mainly useful for finalizing varying parameters, but it can also be used on non-varying parameters as well (as seen in the example below).

Examples

# NOT RUN {
data(biomass)

biomass_tr <- biomass[biomass$dataset == "Training",]
biomass_te <- biomass[biomass$dataset == "Testing",]

# Create a recipe using step_bs() with degree = 3
rec <- recipe(
  HHV ~ carbon + hydrogen + oxygen + nitrogen + sulfur,
  data = biomass_tr
) %>%
  step_bs(carbon, hydrogen, degree = 3)

# Update the step to use degree = 4
rec2 <- rec
rec2$steps[[1]] <- update(rec2$steps[[1]], degree = 4)

# Prep both recipes
rec_prepped  <- prep(rec, training = biomass_tr)
rec2_prepped <- prep(rec2, training = biomass_tr)

# Juice both to see what changed
juice(rec_prepped)
juice(rec2_prepped)

# Cannot update a recipe step that has been trained!
# }
# NOT RUN {
update(rec_prepped$steps[[1]], degree = 4)
# }
# NOT RUN {
# }

Run the code above in your browser using DataCamp Workspace