Out-of-sample performance
In-sample performance provides insights about how well a model captures the data it is modeling. For predictive models, it's also important to check model performance on new, unseen data, the out-of-sample performance.
In this exercise, you will check the test set predictions of your model using MAE (mean absolute error).
Pre-loaded in your workspace again is the model that you built and used in the last exercises.
This exercise is part of the course
Machine Learning with Tree-Based Models in R
Exercise instructions
- Use
modelto predict the out-of-samplefinal_gradeand add your predictions tochocolate_testusingbind_cols(). - Calculate the mean absolute error using a
yardstickfunction.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Predict ratings on test set and add true grades
test_enriched <- predict(__, new_data = ___) %>%
bind_cols(___)
# Compute the mean absolute error using one single function
___(___,
___,
___)