Boxplot Demo

Example boxplot code

import numpy as np
import matplotlib.pyplot as plt

# Fixing random state for reproducibility
np.random.seed(19680801)

# fake up some data
spread = np.random.rand(50) * 100
center = np.ones(25) * 50
flier_high = np.random.rand(10) * 100 + 100
flier_low = np.random.rand(10) * -100
data = np.concatenate((spread, center, flier_high, flier_low))

fig1, ax1 = plt.subplots()
ax1.set_title('Basic Plot')
ax1.boxplot(data)

Out:

{'whiskers': [<matplotlib.lines.Line2D object at 0x7ff0b0c59490>, <matplotlib.lines.Line2D object at 0x7ff0b0c591c0>], 'caps': [<matplotlib.lines.Line2D object at 0x7ff0b0c59670>, <matplotlib.lines.Line2D object at 0x7ff0b1d26eb0>], 'boxes': [<matplotlib.lines.Line2D object at 0x7ff0b1a78a90>], 'medians': [<matplotlib.lines.Line2D object at 0x7ff0b1d841f0>], 'fliers': [<matplotlib.lines.Line2D object at 0x7ff0b1d843a0>], 'means': []}

fig2, ax2 = plt.subplots()
ax2.set_title('Notched boxes')
ax2.boxplot(data, notch=True)

Out:

{'whiskers': [<matplotlib.lines.Line2D object at 0x7ff0b1d2ba30>, <matplotlib.lines.Line2D object at 0x7ff0b1d2b4f0>], 'caps': [<matplotlib.lines.Line2D object at 0x7ff0b1ccb490>, <matplotlib.lines.Line2D object at 0x7ff0b305a430>], 'boxes': [<matplotlib.lines.Line2D object at 0x7ff0b1448c70>], 'medians': [<matplotlib.lines.Line2D object at 0x7ff0b0c502b0>], 'fliers': [<matplotlib.lines.Line2D object at 0x7ff0b0c504c0>], 'means': []}

green_diamond = dict(markerfacecolor='g', marker='D')
fig3, ax3 = plt.subplots()
ax3.set_title('Changed Outlier Symbols')
ax3.boxplot(data, flierprops=green_diamond)

Out:

{'whiskers': [<matplotlib.lines.Line2D object at 0x7ff0b09fe340>, <matplotlib.lines.Line2D object at 0x7ff0b09fe5b0>], 'caps': [<matplotlib.lines.Line2D object at 0x7ff0b09fe760>, <matplotlib.lines.Line2D object at 0x7ff0b14476d0>], 'boxes': [<matplotlib.lines.Line2D object at 0x7ff0b09fe160>], 'medians': [<matplotlib.lines.Line2D object at 0x7ff0b14474f0>], 'fliers': [<matplotlib.lines.Line2D object at 0x7ff0b1447d00>], 'means': []}

fig4, ax4 = plt.subplots()
ax4.set_title('Hide Outlier Points')
ax4.boxplot(data, showfliers=False)

Out:

{'whiskers': [<matplotlib.lines.Line2D object at 0x7ff0b0bc18b0>, <matplotlib.lines.Line2D object at 0x7ff0b0cde2e0>], 'caps': [<matplotlib.lines.Line2D object at 0x7ff0b0cde640>, <matplotlib.lines.Line2D object at 0x7ff0b0cde9a0>], 'boxes': [<matplotlib.lines.Line2D object at 0x7ff0b0bc1670>], 'medians': [<matplotlib.lines.Line2D object at 0x7ff0b0cded00>], 'fliers': [], 'means': []}

red_square = dict(markerfacecolor='r', marker='s')
fig5, ax5 = plt.subplots()
ax5.set_title('Horizontal Boxes')
ax5.boxplot(data, vert=False, flierprops=red_square)

Out:

{'whiskers': [<matplotlib.lines.Line2D object at 0x7ff0b0b9f8e0>, <matplotlib.lines.Line2D object at 0x7ff0b0bc6280>], 'caps': [<matplotlib.lines.Line2D object at 0x7ff0b0bc65e0>, <matplotlib.lines.Line2D object at 0x7ff0b0bc6940>], 'boxes': [<matplotlib.lines.Line2D object at 0x7ff0b0b9fe20>], 'medians': [<matplotlib.lines.Line2D object at 0x7ff0b0bc6ca0>], 'fliers': [<matplotlib.lines.Line2D object at 0x7ff0b0b98040>], 'means': []}

fig6, ax6 = plt.subplots()
ax6.set_title('Shorter Whisker Length')
ax6.boxplot(data, flierprops=red_square, vert=False, whis=0.75)

Out:

{'whiskers': [<matplotlib.lines.Line2D object at 0x7ff0b0e9dfa0>, <matplotlib.lines.Line2D object at 0x7ff0b0a60130>], 'caps': [<matplotlib.lines.Line2D object at 0x7ff0b0a60b20>, <matplotlib.lines.Line2D object at 0x7ff0b0a60d90>], 'boxes': [<matplotlib.lines.Line2D object at 0x7ff0b0e9dc40>], 'medians': [<matplotlib.lines.Line2D object at 0x7ff0b0a606d0>], 'fliers': [<matplotlib.lines.Line2D object at 0x7ff0b0a6e0d0>], 'means': []}

Fake up some more data

spread = np.random.rand(50) * 100
center = np.ones(25) * 40
flier_high = np.random.rand(10) * 100 + 100
flier_low = np.random.rand(10) * -100
d2 = np.concatenate((spread, center, flier_high, flier_low))

Making a 2-D array only works if all the columns are the same length. If they are not, then use a list instead. This is actually more efficient because boxplot converts a 2-D array into a list of vectors internally anyway.

data = [data, d2, d2[::2]]
fig7, ax7 = plt.subplots()
ax7.set_title('Multiple Samples with Different sizes')
ax7.boxplot(data)

plt.show()

---

References

The use of the following functions, methods, classes and modules is shown in this example:

import matplotlib
matplotlib.axes.Axes.boxplot
matplotlib.pyplot.boxplot

Out:

<function boxplot at 0x7ff0b8634ca0>