In this recipe, we are going to look at some of the fuel efficiency metrics over time and in relation to other data points. To do so, we are going to have to replicate the functionality of two very popular R libraries, which are plyr and ggplot2, in Python. The split-apply-combine data analysis capabilities that are so handily covered by the plyr R library are handled equally well but in a slightly different fashion by pandas right out of the box. The data visualization abilities of ggplot2—an R library implementation of the grammar of graphics—are not handled as readily, as we shall see in this recipe.
If you've completed the previous recipe, you should have almost everything you need to continue. However, we are going to use a Python clone of the ggplot2 library for R, which is conveniently named ggplot. If you didn't complete the entire setup chapter and haven't yet installed the ggplot package, open up a terminal...