hardhat (version 1.4.2)

run-mold: mold() according to a blueprint

Description

This is a developer facing function that is only used if you are creating your own blueprint subclass. It is called from mold() and dispatches off the S3 class of the blueprint. This gives you an opportunity to mold the data in a way that is specific to your blueprint.

run_mold() will be called with different arguments depending on the interface to mold() that is used:

• XY interface:

  • run_mold(blueprint, x = x, y = y)

• Formula interface:

  • run_mold(blueprint, data = data)

  • Additionally, the blueprint will have been updated to contain the formula.

• Recipe interface:

  • run_mold(blueprint, data = data)

  • Additionally, the blueprint will have been updated to contain the recipe.

If you write a blueprint subclass for new_xy_blueprint(), new_recipe_blueprint(), or new_formula_blueprint() then your run_mold() method signature must match whichever interface listed above will be used.

If you write a completely new blueprint inheriting only from new_blueprint() and write a new mold() method (because you aren't using an xy, formula, or recipe interface), then you will have full control over how run_mold() will be called.

Usage

run_mold(blueprint, ...)
# S3 method for default_formula_blueprint
run_mold(blueprint, ..., data, call = caller_env())
# S3 method for default_recipe_blueprint
run_mold(blueprint, ..., data, call = caller_env())
# S3 method for default_xy_blueprint
run_mold(blueprint, ..., x, y, call = caller_env())

Value

run_mold() methods return the object that is then immediately returned from mold(). See the return value section of mold() to understand what the structure of the return value should look like.

Arguments

blueprint

A preprocessing blueprint.

...

Not used. Required for extensibility.

data

A data frame or matrix containing the outcomes and predictors.

call

The call used for errors and warnings.

x

A data frame or matrix containing the predictors.

y

A data frame, matrix, or vector containing the outcomes.

Examples

bp <- default_xy_blueprint()

outcomes <- mtcars["mpg"]
predictors <- mtcars
predictors$mpg <- NULL

run_mold(bp, x = predictors, y = outcomes)