A panda paints a box plot. MidJourney 5
You have not made a picture since Monday.
That was deliberate. For two full days everything you produced was a table: grouped, aggregated, joined, stacked, pivoted and parsed. Some of those tables held real findings, and you read them the way you read a spreadsheet, one number at a time, taking my word for it that the ordering meant something.
Today is the tenth step, and it is the one all the others were for. You will learn the anatomy of a figure, so that you can control one instead of accepting whatever the defaults give you. Then you will learn a library that takes a whole table and the names of two columns and does the rest, and one argument in it, hue=, that will change how you look at data for the rest of your career.
This afternoon you pick your project.
Class materials
| Session | Session 1 | Session 2 |
|---|---|---|
| day 7 / morning | ๐ผ๏ธ The Anatomy of a Figure | ๐จ Name the Columns, Not the Colours |
| day 7 / afternoon | ๐ One Cloud of Points, Three Species | ๐ Project Kickoff |
End-of-day practice
Tonight you take two versions of the USDA plant hardiness map, eleven years apart, and find out what you can honestly say about the difference between them. Six figures, and only one of them is the one the evening was about.
Answer keys
Work the exercise first, then check it. The keys show the code, and more usefully the written answers, which are the half you cannot check on your own.