Imputing missing values and creating dummy variables
After detecting missing values in the attrition
dataset and determining that they are missing completely at random (MCAR), you decide to use K Nearest Neighbors (KNN) imputation. While configuring your feature engineering recipe
, you decide to create dummy variables for all your nominal variables and update the role of the ...1
variable to "ID" so you can keep it in the dataset for reference, without affecting your model.
This exercise is part of the course
Feature Engineering in R
Exercise instructions
- Update the role of
...1
to "ID". - Impute values to all predictors where data are missing.
- Create dummy variables for all nominal predictors.
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) %>%
# Update the role of "...1" to "ID"
___(...1, new_role = "ID" ) %>%
# Impute values to all predictors where data are missing
step_impute_knn(___) %>%
# Create dummy variables for all nominal predictors
___(all_nominal_predictors())
lr_recipe