Normalizing and log-transforming
You are handed a dataset, attrition_num
with numerical data about employees who left the company. Features include
Age
, DistanceFromHome
, and MonthlyRate
.
You want to use this data to build a model that can predict if an employee is likely to stay, denoted by Attrition
, a binary variable coded as a factor
. In preparation for modeling, you want to reduce possible skewness and prevent some variables from outweighing others due to variations in scale.
The attrition_num
data and the train
and test
splits are loaded for you.
This exercise is part of the course
Feature Engineering in R
Exercise instructions
- Normalize all numeric predictors.
- Log-transform all numeric features, with an offset of one.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
lr_model <- logistic_reg()
lr_recipe <-
recipe(Attrition~., data = train) %>%
# Normalize all numeric predictors
___(all_numeric_predictors()) %>%
# Log-transform all numeric features, with an offset of one
___(___, offset = ___)
lr_workflow <-
workflow() %>%
add_model(lr_model) %>%
add_recipe(lr_recipe)
lr_workflow