Feature Elimination and Variable Importance in R with "caret" (2021)
Subscribe to RichardOnData here: / @richardondata
Patreon: / richardondata
GitHub: https://github.com/RichardOnData/YouT...
Part 1: • Preprocessing Data in R for ML with "...
"When Should You Use Random Forests?": • When Should You Use Random Forests?
In this video I continue with the multi-part tutorial series that began in the last video with the "caret" package. We will finish the process of applying transformations to the training set, and applying the same specified transformations to the testing set. We'll then get into algorithmically removing low information features via recursive feature elimination, training our first model, and then creating variable importance plots where applicable.
There are a few sources from which this tutorial draws influence and structure. The first is the GitHub documentation on "caret" from its creation, Max Kuhn. The second is a very well-written and comprehensive tutorial by author Selva Prabhakaran on Machine Learning Plus. Third is a helpful resource for dealing with class imbalance, as we often find with classification problems.
GitHub documentation from Max Kuhn: https://topepo.github.io/caret/
Tutorial by Selva Prabhakaran: https://www.machinelearningplus.com/m...
Tutorial on "caret" with class imbalances: https://shiring.github.io/machine_lea...
Watch video Feature Elimination and Variable Importance in R with "caret" (2021) online without registration, duration hours minute second in high quality. This video was added by user RichardOnData 12 January 2021, don't forget to share it with your friends and acquaintances, it has been viewed on our site 9,671 once and liked it 258 people.