Session Ready
Exercise

Evaluating base learners' performance

Before you proceed to create an ensemble model from the set of base learners, it is important to evaluate how they are doing individually on the classification task. Since they were all trained on identical resamples of the training data, it is also key to ensure their predictions are not highly correlated, as this could impact the efficacy of the entire ensemble model.

In this exercise, you will evaluate the performance of each base learner as well as their prediction correlations. The list of caret's train objects created through caretList() is available in your workspace as models.

Instructions 1/4
undefined XP
  • 1
  • 2
  • 3
  • 4
  • Using your models list and the appropriate function in the caret package, retrieve the classification results for each base learner in each resample. Save them as results, then summarize them.