Timeseries

⏱ Quick start

Making a simple line chart with matplotlib is pretty straightforward thanks to the plot() function.

If you provide only a series of values, it will consider that these values are ordered and will use values from 1 to n to create the X axis.🔥

For more control on the chart, see the dedicated section below.

The most basic histogram one can make with python and seaborn

# libraries
  import matplotlib.pyplot as plt
  import numpy as np

  # create data
  values=np.cumsum(np.random.randn(1000,1))

  # use the plot function
  plt.plot(values)
  

Area charts for timeseries

Area charts are very often used for timeseries visualization. They are particularly adapted when there is only 1 series to display. When several groups must be displayed, they can still be used using faceting.

The most basic area chart one can make with python and matplotlib

Area chart for timeseries with facting to display many groups

Line charts for timeseries

Line charts are probably the most common type of viz for timeseries. They are particularly adapted when several groups must be displayed to compare their evolution. But mind the spaghetti plot: too many groups make the figure unreadable.

Line chart with multiple groups

How to customize the matplotlib line plot appearance: width, stroke, color, style..

Line chart with multiple groups

Line chart and small multiple (variation)

How to deal with the date format. How to represent those dates properly on the X axis.

How to build a dual Y axis line chart to display 2 series with different units.

Heatmap for timeseries

A heatmap can be used to display some temporal data. Here is an example using matplotlib where the evolution of a temperature is displayed over the hour of the day (Y axis) and the day of the year (X axis) organized by month.

A heatmap for temporal data with Python and Matplotlib

From the web

A set of publication ready charts for timeseries. They are made with python and matplotlib and use a high level of customization.

The first example is a line chart showing how to add labels at the end of each series for better readability. The second shows how to use lollipop to display the evolution of a deviation around a trend.

How to add labels at the end of each line with matplotlib. Definitely better than using a legend!

A highly customized lollipop chart made with Python and Matplotlib to display a timeseries.

A multipanel connected scatterplot, with several groups displayed and one highlighted per panel.

Learn how to mimick The Economist's style with a figure combining both a line and an area chart.