Seaborn is a Python visualization library for statistical plotting. It comes equipped with preset styles and color palettes so you can create complex, aesthetically pleasing charts with a few lines of code. It’s designed to work with NumPy and pandas data structures and to support statistical tasks completed in SciPy and statsmodels.
Seaborn is built on top of Python’s core visualization library matplotlib, but it’s meant to serve as a complement, not a replacement. In most cases, you’ll still use matplotlib for simple plotting, and you’ll need a knowledge of matplotlib to tweak Seaborn’s default plots.
- Data visualization with Seaborn (O’Reilly) - This tutorial gives a quick overview of the code needed to create statistical data visualizations such as histograms, pairplots, and factor plots.
- Data Visualization in Python: Advanced Functionality in Seaborn (Insight Data Labs) - This tutorial covers some of Seaborn’s most useful functions, such as conditional plotting and showing interactions between variables.
- Visualizing Google Forms Data with Seaborn (Chris Moffitt) - A great example of using Seaborn for business problems, this tutorial explores bar plots, time series graphs, and heatmaps, and clustermaps.
- Learning Python: Visualizing the NBA Draft (Savvas Tjortjoglou)- This NBA dataset makes for a fun time learning how to make line, bar, box, and violin plots.
- Official Seaborn documentation
- Plotting univariate and bivariate distributions
- Visualizing pairwise relationships
- Fitting linear models
- Exploring interactions between multiple variables
- Plotting wide-form data
- Making point plots, box plots, violin plots, and categorical scatter plots
For examples of the visualizations you can create with Seaborn, see this gallery.