Setting up your workflow
You want to investigate if JobRole
alone can be a predictor for Attrition
. Given that JobRole
is a factor, you plan to use a Bayes encoder to represent it numerically in your model.
The embed
package and the corresponding test
and train
splits from the attrition
dataset are loaded in your workspace.
This exercise is part of the course
Feature Engineering in R
Exercise instructions
- Create recipe using the Bayes encoder.
- Bundle the model and recipe with a workflow.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
lr_model <- logistic_reg()
# Create recipe using the Bayes encoder
lr_recipe_glm <-
recipe(Attrition ~ JobRole, data = train) %>%
___(JobRole, outcome = ___(Attrition))
# Bundle with workflow
lr_workflow_glm <-
workflow() %>%
add_model(lr_model) %>%
add_recipe(___)
lr_workflow_glm