Make sure you know about
- matplotlib color maps
- Palettable: many beautiful color palettes
Load the color module:
import plottools.colors as c
The color module provides a few color palettes as dictionaries that reference colors by standard color names.
c.color_palettes
is a dictionary referencing all color palettes provided by the module by their name.
For example
colors = c.color_palettes['muted']
colors['red']
returns the red color of the colors_muted palette. The returned
colors are valid matplotlib colors and can be passed on to the color,
facecolor, edgecolor, etc. arguments of the various matplotlib
plotting functions.
The following sections display the colors and their names of all provided palettes.
c.colors_plain
This palette is provided for completeness. The colors are made up of either full or half saturated RGB values, e.g. red is '#ff0000', green is '#00ff00', and orange is '#ff8000'.
c.colors_muted
This palette extends the colors_henninger palette.
c.colors_vivid
Vivid but not plain basic colors for a fresh look.
c.colors_tableau
This is matplotlib's tableau palette, the default in newer matplotlib
versions, also known as tab10.
c.colors_henninger
c.colors_scicomp
Nice red, orange and yellow, but blue and green need some improvement.
c.colors_unituebingen
c.colors_itten
c.colors_solarized
c.colors_material
Make colors lighter.
For 40% lightness of blue do
colors = c.color_palettes['muted']
lightblue = c.lighter(colors['blue'], 0.4)
Make colors darker.
For 40% darker blue do
colors = c.color_palettes['muted']
darkblue = c.darker(colors['blue'], 0.4)
Mix two colors.
For 30% transition between blue and orange do
colors = c.color_palettes['muted']
colors = c.color_palettes['muted']
color = c.gradient(colors['blue'], colors['orange'], 0.3)
If you want to use in your LaTeX document the same colors as in your plots,
then you can export matplotlib colors using the latex_colors() function.
Either for single colors:
colors = c.color_palettes['muted']
c.latex_colors(colors['red'], 'red')
writes to the console
\definecolor{red}{rgb}{0.753,0.153,0.090}
or for a whole palette:
c.latex_colors(c.color_palettes['vivid'])
writes to the console
\definecolor{red}{rgb}{0.843,0.063,0.000}
\definecolor{orange}{rgb}{1.000,0.565,0.000}
\definecolor{yellow}{rgb}{1.000,0.969,0.000}
...
Then copy the color definitions into you LaTeX preamble. Do not forget to
load the color or xcolor packages before:
\usepackage{xcolor}
You then can use the newly defined colors with the usual commands, like for example:
\textcolor{red}{Some text in my special red.}
Generate and register a color map from colors like this:
colors = c.color_palettes['muted']
cmcolors = [colors['red'],
c.lighter(colors['orange'], 0.85),
c.lighter(colors['yellow'], 0.2),
c.lighter(colors['lightblue'], 0.8),
colors['blue']]
cmvalues = [0.0, 0.25, 0.5, 0.8, 1.0]
c.colormap('RYB', cmcolors, cmvalues)
This is just a simple wrapper for
matplotlib.colors.LinearSegmentedColormap and matplotlib.cm import register_cmap.
The new colormap can then be used directly by its name for the cmap
arguments of imshow(), pcolormesh(), contourf(), etc.:
ax.imshow(image, cmap='RYB')
Retrieve a single color from a color map:
jet_red = c.cmap_color('jet', 0.0)
For displaying colors and color maps, four functions are provided:
plot_colors(): plot all colors of a palette and optionally some lighter and darker variants.
plot_complementary_colors(): plot complementary colors of a palette on top of each other.
plot_color_comparison(): plot matching colors of severals palettes on top of each other.
plot_colormap(): plot a color map and its luminance.
These functions are helpfull when creating new palettes. See their documentation for details on how to use them.