dufte 0.2.21

Clean matplotlib plots

Da kiekste, wa?

This package creates clean and beautiful plots that work on light and dark backgrounds. Inspired by the work of Edward Tufte.

To use, simply select the dufte style. Check out dufte.legend() and dufte.ylabel() for more duftness.

import matplotlib.pyplot as plt
import dufte
import numpy as np

plt.style.use(dufte.style)

np.random.seed(0)

x0 = np.linspace(0.0, 3.0, 100)
y0 = x0 / (x0 + 1)
y0 += 0.1 * np.random.rand(len(y0))
plt.plot(x0, y0, label="no balacing")

x1 = np.linspace(0.0, 3.0, 100)
y1 = 1.5 * x1 / (x1 + 1)
y1 += 0.1 * np.random.rand(len(y1))
plt.plot(x1, y1, label="CRV-27")

x2 = np.linspace(0.0, 3.0, 100)
y2 = 1.6 * x2 / (x2 + 1)
y2 += 0.1 * np.random.rand(len(y2))
plt.plot(x2, y2, label="CRV-27*")

dufte.ylabel("ylabel")
dufte.legend()

plt.show()

Further reading:

• Randal S. Olson's blog entry
• prettyplotlib
• Wikipedia: Chartjunk

Projects using dufte:

• perfplot
• stargraph

Background

The position $x_i$ of the line annotations is computed as the solution of a non-negative least-squares problem

\begin{align}
\frac{1}{2}\sum_i (x_i - t_i)^2 \to \min_x,\\
(x_i - x_j)^2 \ge a^2 \quad \forall i,j.
\end{align}

where $a$ is the minimum distance between two entries and $t_i$ is the target position.

Testing

To run the dufte unit tests, check out this repository and type

pytest

License

This software is published under the GPLv3 license.