This lesson requires a premium membership to access.
Premium membership includes unlimited access to all courses, quizzes, downloadable resources, and future content updates.
R’s ROCR package can be used for evaluating and visualizing the performance of classifiers / fitted models. It is helpful for estimating performance measures and plotting these measures over a range of cutoffs. (Note: the terms classifier and fitting model are used interchangeably)
The package features over 25 performance. The three important functions ‘prediction’, ‘performance’ and ‘plot' do most of the work. Let’s look at these functions and apply them to measure the performance of our model/classifier.
