beauty — make matplotlib beautiful

This package will make your matplotlib plots more beautiful by changing the default fonts and colors in matplotlib. It also includes various utility functions for performing common plot-beautifying tasks (such as removing axes, rescaling axes, etc.).

Most plotting functions (plot, hist, errorbar, bar) automatically change to use the new colors, but scatter and stem don't. You can use beauty.scatter and beauty.stem just like you would plt.scatter and plt.stem, respectively.

Installation

You can install beauty using pip:

pip install git+git://github.com/rameshvs/beauty.git

Customization

• You can set whether to use serif or sans-serif fonts using beauty.set_serif. By default, it uses sans serif fonts (False).

• You can set whether to use LaTeX for all rendering using beauty.set_tex. By default, it uses LaTeX for all rendering (True). This typically causes a slowdown in producing text (labels, titles, etc.), especially for the first figure of a session.

• You can customize other parameters like font sizes using beauty.set_font_sizes.

Example

import beauty
import matplotlib.pyplot as plt

plt.figure(figsize=(4,2))
plt.plot([4, 7, 13], label='Increasing')
plt.plot([7, 2, 1], label='Decreasing')
plt.plot([0, 6, 2], label='Up and down')
plt.legend(loc='upper left')

plt.figure(figsize=(3, 3))
beauty.scatter([4,7,13], [7,2,1])
plt.xlabel('$x$')
plt.ylabel('$y$')

More sample plots

For full demos (with code samples), see the wiki.

Here's an example of the normal distribution along with a histogram of samples from it:

Tips and tricks

• In order to get the most out of your plots (especially when using subplots), I recommend calling plt.tight_layout() to make matplotlib optimize your plot layout.

• Don't forget to use raw strings when embedding LaTeX in python strings! For example, plt.title(r'$\frac{x}{2}$') will work, but plt.title('$\frac{x}{2}$') won't (since Python treats \f in a non-raw string as a formfeed character).

Licensing

This package is released under the MIT license. See LICENSE.txt for more details.