I am happy to announce that a new tidymodels-centric version of my free, online, interactive course, Supervised Machine Learning: Case Studies in R, has been published! 🎉

This is at least the third version of this course I’ve built at this point 😁 but I believe it to be the best, in terms of how it communicates machine learning concepts and how useful to your real-world problems the demonstrated code will be. Huge thanks to my RStudio teammates such as Alison Hill and Max Kuhn for their feedback during the editing process! Similar to the last time I launched this course, it provides four case studies using data from the real world for you to practice your predictive modeling skills.

Supervised machine learning in R

One question we sometimes field from R users is about choosing to use tidymodels vs. caret. The original version of my course mostly used caret, and caret is a stable and broadly used framework for modeling and machine learning in R. Although new development for our team is focused on tidymodels and I personally am spending my time only on tidymodels projects, I know that there may be folks out there who want to learn from the original version of the course. The original, caret-centric version of the course will continue to be available at a different URL.

Revamping this course, built on the amazing framework created by Ines Montani, yet again has given me an opportunity to revisit the course development process and reflect on what I’ve learned. I have a post for the RStudio Education blog in the works, and will publish that soon.

The back-end code execution uses Binder. It can take a little while for Binder to spin up a new Docker container the first time you run a code exercise in a new session, so be patient when that happens! ⏳

Try it out

As the tidymodels ecosystem continues to develop, we are excited to have more resources for folks looking to learn, whether that’s articles on getting started or a course like mine. Contributions and comments on how to improve my course are welcome! Please file an issue or submit a pull request if you find something that could be fixed or improved.