Getting Started

mpltools provides tools for Matplotlib that make it easier to adjust the element styles, choose colors, make specialized plots, etc. For the most part, these tools are only loosely-connected in functionality, so the best way to get started is to look at the Example Gallery.


This package got its start by implementing plotting “styles”—essentially style sheets that are similar to matplotlibrc files.

There are a number of pre-defined styles located in mpltools/style/. For example, there’s a pre-defined stall called “ggplot”, which emulates the aesthetics of ggplot (a popular plotting package for R). To use this style, just add:

>>> from mpltools import style
>>> style.use('ggplot')

To list all available styles, use:

>>> print style.available

Defining your own style

Unfortunately, the syntax for an mplstyle file is slightly different than matplotlibrc files because mpltools uses ConfigObj to parse them. For example the first few lines of the “ggplot” style looks like:

patch.linewidth = 0.5
patch.facecolor = '#348ABD'  # blue
patch.edgecolor = '#EEEEEE'
patch.antialiased = True

Unlike matplotlibrc files, key/value pairs are separated by an equals sign and strings must be quoted.

You can specify styles in either an mplstyle file or a *.rc file located in ~/.mplstylelib/. In an mplstyle file, style names should be specified as sections. A simple mplstyle file might look like:


text.fontsize = 12
figure.dpi = 150


text.fontsize = 10 = 'serif'

Alternatively, a single style is specified in each *.rc file found in ~/.mplstylelib/, and the file name determines the style name. For example, a style file named ~/.mplstylelib/mystyle.rc would define mystyle.

Style priority

mpltools searches the current working directory and your home directory for mplstyle files. In addition, it looks in ~/.mplstylelib/ for *.rc files. If for some reason, you decide to define the same style in multiple places, the resolution order is

  1. ./mplstyle
  2. ~/.mplstyle
  3. ~/.mplstylelib/*.rc
  4. mpltools/style/*.rc

So, if you define ~/.mplstylelib/mystyle.rc and a section [mystyle] in ./mplstyle, then the later will update the former (redefined settings are overridden, but keys undefined in ./mplstyle remain).

plot2rst Sphinx extension

The plot2rst Sphinx extension provides a simple way to generate reStructuredText (rst) examples from python files. As the name suggests, there’s built-in handling of Matplotlib plots. Example python files will have their docstrings converted to rst, and python code will be placed in a Sphinx code-block. Check out the Example Gallery for details.

To generate your own examples, add 'mpltools.sphinx.plot2rst' to the list of extensions in your Sphinx configuration file. In addition, make sure the example directory(ies) in plot2rst_paths points to a directory with examples named plot_*.py and include an index.rst file. By default, the example path points to:

plot2rst_paths = ('../examples', 'auto_examples')

The first directory specifies the location of the python files, and the second directory specifies where to save the rst examples. Note that the paths are relative to the Sphinx source directory (where lives); using these defaults, I would define my example gallery as follows (this is a snippet from the mpltools directory structure):


When building the docs, plot2rst will generate the auto_examples directory, which will look something like:

               <generated images>

Note that python files are copied to the auto_examples directory (and later to the build directory) because a download link is added to the example.

If you’re wondering about all of the index.rst files in the examples directory, these are used for custom markup. They could be blank files, but more likely you’d want to add headers and possibly, descriptive text. For example, the doc/examples/index.rst file in mpltools just has:

.. _example-gallery:

Example Gallery

(the _example-gallery: markup is for Sphinx cross-referencing) and doc/examples/layout/index.rst has:

``layout`` module

Note: plot2rst was adapted from in scikits-image, which borrowed the implementation from scikit-learn.

Other tools

The remaining tools are just small functions I’ve found useful over the years. They are organized into the following modules:

Defines a light wrapper-class for working with matplotlib.animation
Add annotations to your plots (e.g. slope marker).
Color choice and custom colors (e.g. parameter-based color choice).
Alter visual layout of plots (e.g. figure size, crossed spines).
Specialty plotting functions (e.g. Hinton diagram).

See the Example Gallery for details.

Table Of Contents

Previous topic

mpltools: Tools for Matplotlib

Next topic