Tidymodels, Virtually

An Introduction to Machine Learning with Tidymodels

Welcome

Thank you for enrolling in this course! These are the materials for an introductory machine learning short course with tidymodels by Dr. Alison Presmanes Hill.

This four-hour pre-conference short course will provide a gentle introduction to machine learning with R using the modern suite of predictive modeling packages called tidymodels. We will build, evaluate, compare, and tune predictive models. Along the way, we’ll learn about key concepts in machine learning including overfitting, the holdout method, the bias-variance trade-off, ensembling, cross-validation, and feature engineering. Learners will gain knowledge about good predictive modeling practices, as well as hands-on experience using tidymodels packages like parsnip, rsample, recipes, yardstick, tune, and workflows.

The entire course will be recorded and made available to registered R/Medicine 2020 conference participants for replay. To protect the privacy of participants, no breakouts, video feeds, or chats will be recorded. We also request that you refrain from recording or screen-grabbing any part of the course.

Prework

Essential

• Please sign up for a free RStudio Cloud account here: https://rstudio.cloud.
• Install the latest version of Zoom.
• Join a pre-course technology check session - details are in the pre-course email.
• We want to provide a welcoming and supportive learning environment for everyone. To help us do so, please take a look at our code of conduct.

Optional

Please tune into class with a laptop that has the following installed:

On the day of the class, we’ll provide you with a link to an RStudio Cloud project that contains all of the course materials.

• A recent version of R (>=3.9.0), which is available for free at https://cran.r-project.org/

• A recent version of RStudio Desktop (>=1.3.0), available for free at https://www.rstudio.com/download (RStudio Desktop Open Source License)

• The R packages we will use, which you can install by connecting to the internet, opening RStudio, and running at the command line:

  install.packages(c("tidyverse", "tidymodels", 
                       "modeldata", "kknn", "rpart",
                       "rpart.plot", "rattle", 
                       "vip", "ranger", "partykit"))
    

We will use the software listed above only as an important backup should there be problems with the classroom server connection.

I look forward to meeting you,

Alison

Links

• RStudio Cloud workspace

Slides

• 00: Brief intro
• 01: Build a model
• 02: Build better workflows
• 03: Tune better models

Workshops

Upcoming and past offerings:

• R / Medicine 2020 Virtual Conference
• Becton, Dickinson and Company, Online
• Children’s Hospital of Philadelphia, Online
• rstudio::conf 2020, San Francisco, CA, January 27 & 28 Training Days

Acknowledgments

This is a website made with the distill package and a custom theme by Alison Hill & Desirée De Leon.