Intermediate R: Machine Learning
This four week course introduces machine learning methods in R, specifically for analyses relevant to biomedical researchers. This course assumes attendees are familiar with basic R syntax, using packages, and basic data manipulation using tidyverse. The materials also assume a strong foundation in basic statistics as well as prior/concurrent participation in the fredhutch.io course Concepts in Machine Learning (or equivalent experience). At the end of this course, you will be able to apply basic principles of machine learning to research questions and will have established a foundation for further exploration of machine learning techniques. Please see each set of class materials for specific learning objectives. These materials are developed by fredhutch.io, the data and computational analysis training program at Fred Hutch. Each class in this course includes brief tutorials interspersed with challenge exercises.
Sessions of the course are periodically taught by instructors at Fred Hutch; each of the four classes is scheduled for two hours. The HackMD (interactive page used for sharing links and information) for instructor-led courses is here. The materials are also freely available for self-guided, work-at-your-own-pace study.
Required software: Software requirements for this course include:
The links above reference relevant sections of fredhutch.io’s Software page.
Materials for all lessons in this course include:
Solutions for exercises can be found in
For curriculum contributors and instructors
Please see the following resources for more information on:
- Teaching these materials.
instructors.mdincludes information for instructors to facilitate teaching each lesson, including additional options if a participant can’t install R, RStudio, or tidyverse on their computer.
hackmdio.mdis an archive of the interactive webpage used during lessons.
- Contributing to lessons.
Each lesson’s materials are described in R markdown (
.Rmd) and markdown (
.md) files in the GitHub repository. The former should be edited to make changes to the material and then knit to create the latter, which are then rendered in GitHub. The directories for each class hold figures for each lesson.