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 0x7f6d580c38e0>, <matplotlib.lines.Line2D object at 0x7f6d53d11c60>], 'caps': [<matplotlib.lines.Line2D object at 0x7f6d53d110c0>, <matplotlib.lines.Line2D object at 0x7f6d52e8e1d0>], 'boxes': [<matplotlib.lines.Line2D object at 0x7f6d580c2b90>], 'medians': [<matplotlib.lines.Line2D object at 0x7f6d52e8f0a0>], 'fliers': [<matplotlib.lines.Line2D object at 0x7f6d53e13d00>], 'means': []}

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

Out:

{'whiskers': [<matplotlib.lines.Line2D object at 0x7f6d5840ce50>, <matplotlib.lines.Line2D object at 0x7f6d5840f700>], 'caps': [<matplotlib.lines.Line2D object at 0x7f6d5840ff70>, <matplotlib.lines.Line2D object at 0x7f6d5840d2a0>], 'boxes': [<matplotlib.lines.Line2D object at 0x7f6d5840e6e0>], 'medians': [<matplotlib.lines.Line2D object at 0x7f6d58151ae0>], 'fliers': [<matplotlib.lines.Line2D object at 0x7f6d581530a0>], '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 0x7f6d53d491b0>, <matplotlib.lines.Line2D object at 0x7f6d53d486d0>], 'caps': [<matplotlib.lines.Line2D object at 0x7f6d53d4ba60>, <matplotlib.lines.Line2D object at 0x7f6d53d481f0>], 'boxes': [<matplotlib.lines.Line2D object at 0x7f6d53d4ad40>], 'medians': [<matplotlib.lines.Line2D object at 0x7f6d53ea0af0>], 'fliers': [<matplotlib.lines.Line2D object at 0x7f6d53ea1c00>], 'means': []}

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

Out:

{'whiskers': [<matplotlib.lines.Line2D object at 0x7f6d536bee90>, <matplotlib.lines.Line2D object at 0x7f6d536bc610>], 'caps': [<matplotlib.lines.Line2D object at 0x7f6d536bd390>, <matplotlib.lines.Line2D object at 0x7f6d53a4e590>], 'boxes': [<matplotlib.lines.Line2D object at 0x7f6d536bc070>], 'medians': [<matplotlib.lines.Line2D object at 0x7f6d53a4d4b0>], '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 0x7f6d5367b730>, <matplotlib.lines.Line2D object at 0x7f6d5367a650>], 'caps': [<matplotlib.lines.Line2D object at 0x7f6d5367a8f0>, <matplotlib.lines.Line2D object at 0x7f6d5367a4a0>], 'boxes': [<matplotlib.lines.Line2D object at 0x7f6d53679840>], 'medians': [<matplotlib.lines.Line2D object at 0x7f6d53678ca0>], 'fliers': [<matplotlib.lines.Line2D object at 0x7f6d53679d20>], '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 0x7f6d535aba90>, <matplotlib.lines.Line2D object at 0x7f6d535a95d0>], 'caps': [<matplotlib.lines.Line2D object at 0x7f6d535c9300>, <matplotlib.lines.Line2D object at 0x7f6d535c9ae0>], 'boxes': [<matplotlib.lines.Line2D object at 0x7f6d535ab820>], 'medians': [<matplotlib.lines.Line2D object at 0x7f6d535cb940>], 'fliers': [<matplotlib.lines.Line2D object at 0x7f6d535ca410>], '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:

• matplotlib.axes.Axes.boxplot / matplotlib.pyplot.boxplot