method_prepare: Method-specific data preparation

In crtests: Classification and Regression Tests

Description

Generic function for method-specific data preparation, if any is necessary

Usage

method_prepare(method, test, ...)

## Default S3 method:
method_prepare(method, test, ...)

## S3 method for class 'randomForest'
method_prepare(method, test, ...)

Arguments

method	The regression or classification method that needs specific data preparation.
test	The test being executed, whose data attribute is a list of the train and holdout data sets, that has already been prepared by prepare_data
...	Extra arguments to pass on to class methods

Value

A prepared data.frame

identity(data)

Methods (by class)

• default: Default function for method-specific data preparation. There is no default method-specific preparation, returns identity.

• randomForest: Random Forest specific data preparation. Calls group_levels on data, then relevels the holdout set so it has no levels not found in the training set (using prepare_data)

crtests documentation built on May 1, 2019, 9:09 p.m.