JoyPy is a one-function Python package based on matplotlib + pandas with a single purpose: drawing joyplots.
The code for JoyPy borrows from the code for kdes in pandas.plotting, and uses a couple
of utility functions therein. (This could be probably avoided with a little bit of
extra effort.)
Joyplots are stacked, partially overlapping density plots, simple as that. They are a nice way to plot data to visually compare distributions, especially those that change across one dimension (e.g., over time). Though hardly a new technique, they have become very popular lately thanks to the R package ggjoy (which is clearly much better developed/maintained than this one -- and I strongly suggest you use that if you can use R and ggplot.) Update: the ggjoy package has now been renamed ggridges.
If you don't know Joy Division, you are lucky: you can still listen to them for the first time! Here's a hint: google "Unknown Pleasures". This kind of plot is now also known as ridgeline plot, since the original name is controversial.
Take a look at this jupyter notebook for a couple of simple examples.
-
Python 3.5+
Compatibility with python 2.7 has been dropped with release 1.11. -
scipy >= 0.11
-
pandas >= 0.20
Not sure what are the oldest supported versions. As long as you have somewhat recent versions, you should be fine.
It's actually on PyPI, because why not:
pip install joypy
To install from github, run:
git clone git@github.com:sbebo/joypy.git
cd joypy
pip install .
Released under the MIT license.
This is just a sunday afternoon hack, so no guarantees! If you want to contribute or just copy/fork, feel free to.