datasauRus is a package storing the datasets from the paper Same Stats, Different Graphs: Generating Datasets with Varied Appearance and Identical Statistics through Simulated Annealing. It’s a useful package for:

  1. Having a dinosaur dataset
  2. Showing a dinosaur related variant of Anscombe’s Quartet

You can now get datasauRus on CRAN, though it might not be on all mirrors just yet.

install.packages("datasauRus")

Credit

This package wouldn’t exist without some nifty people:

Examples

We’ve already started playing with the datasauRus dataset…

Me

library(ggplot2)
library(datasauRus)
ggplot(datasaurus_dozen, aes(x=x, y=y, colour=dataset))+
  geom_point()+
  theme_void()+
  theme(legend.position = "none")+
  facet_wrap(~dataset, ncol=3)
datasauRus dozen!
datasauRus dozen!

Ramnath

library(ggplot2)
library(datasauRus)
library(gganimate)

p <- ggplot(datasaurus_dozen, aes(x = x, y = y, frame = dataset)) +
  geom_point() +
  theme(legend.position = "none")

gganimate(p, title_frame = FALSE)

In the package

Every dataset has an example associated with it. These can be found not just in the help files for each dataset but also in a standalone directory (inst/examples). We’d like to keep building on these so please feel free to add any visualisations you make with the datasets. If you haven’t worked on a package before and want some help, you can book some time with me to get your dataviz example into the package.

datasauRus now on CRAN
Tagged on:         

What do you think?