Stop repeating yourself: functions and iteration in R

Description

Sooner or later most R users need to do the same thing many times, like run an analysis for each species, read in every file in a folder, fit a model to each treatment group. Writing functions is one of the more useful habits in day-to-day data analysis, and this workshop covers the two tools that handle repetition cleanly:

• Writing your own functions: wrapping repeated code into reusable functions with arguments, defaults, and return values
• Iterating with the purrr package: using map() and friends to apply a function across many inputs, replacing copy-paste or for-loops

We will work through practical examples together and you will learn how to take repetitive code of your own and refactor it into a function you apply across many inputs.

Slides in full screen     Download PDF slides

Resources

Workshop materials

The workshop repo has all demo scripts, exercises, and solutions:

github.com/selinaZitrone/functions-iteration-workshop

Writing functions

• R for Data Science on Functions: a practical introduction to functions
• Advanced R, Functions: a deeper dive once you’re comfortable

Iteration with purrr

• purrr website: full reference and articles
• purrr cheatsheet: the whole family on one page
• R for Data Science on Iteration: map() and other iteration iteration tools

More purrr tools:

• map2() / pmap(): iterate over two or more inputs at once
• safely() / possibly(): keep going when one element throws an error
• Parallel processing in purrr 1.1.0 and furrr: run a map across multiple CPU cores