xycmap

Bivariate colormap solutions.

This package makes it easy to create custom two-dimensional colormaps, apply them to your series, and add bivariate color legends to your plots.

Installation

pip install xycmap

Usage

Make a custom interpolated colormap by specifying four corner colors (see recognized formats here), and dimensions n:

import xycmap
corner_colors = ("lightgrey", "green", "blue", "red")
n = (5, 5)  # x, y
cmap = xycmap.custom_xycmap(corner_colors=corner_colors, n=n)

Or make a colormap by mixing two matplotlib colormaps, and specifying dimensions n:

import matplotlib.pyplot as plt
xcmap = plt.cm.rainbow
ycmap = plt.cm.Greys
n = (5, 5)  # x, y
cmap = xycmap.mean_xycmap(xcmap=xcmap, ycmap=ycmap, n=n)

With that in place, apply the colormap to two series that are numeric or categorical:

colors = xycmap.bivariate_color(sx=sx, sy=sy, cmap=cmap)

Note that you can apply limits to the axes, as well as pass custom bins for the axes (if numerical). See the docstring for details.

Then simply pass colors to your plot. To add a legend, create a new ax and run bivariate_legend() into the ax with the same parameters as bivariate_color(), e.g.:

cax = fig.add_axes([1, 0.25, 0.5, 0.5])
cax = xycmap.bivariate_legend(ax=cax, sx=sx, sy=sy, cmap=cmap)

Meta

Remco Bastiaan Jansen – r.b.jansen.uu@gmail.com - https://github.com/rbjansen

Distributed under the MIT license. See LICENSE for more information.