Preparing for evaluation

In order to measure the validate performance of your models you need compare the predicted values of life_expectancy for the observations from validate set to the actual values recorded. Here you will prepare both of these vectors for each partition.

Cet exercice fait partie du cours

<cours>Machine Learning in the Tidyverse</cours>

Voir le cours

Instructions de l’exercice

Extract the actual life_expectancy from the validate data frames and store these in the column validate_actual.
Predict the life_expectancy for each validate partition using the map2() and predict() functions in the column validate_predicted.

Exercice interactif pratique

Essayez cet exercice en complétant ce code d’exemple.

cv_prep_lm <- cv_models_lm %>% 
  mutate(
    # Extract the recorded life expectancy for the records in the validate data frames
    validate_actual = map(validate, ~.x$___),
    # Predict life expectancy for each validate set using its corresponding model
    validate_predicted = map2(.x = model, .y = validate, ~___(.x, .y))
  )

Modifier et exécuter le code