How to create and plot a simple histogram with matplotlib and python ?

Example of how to create and plot a simple histogram with matplotlib and python from a dataset:

Plot a simple histogram using matplotlib

A simple histogram can be created with matplotlib using the function hist(), example:

Note: see for example Histograms vs. Bar Charts to understand the differences between the 2 plots.

import matplotlib.pyplot as plt

data = [1,2,2,3,3,3,4,4,5]

plt.hist(data)

plt.title('How to plot a simple histogram in matplotlib ?', fontsize=10)

plt.savefig("plot_simple_histogramme_matplotlib_01.png")

plt.show()

Histogram normalisation

To normalize the histogram, just add the option "normed", example histogram normalisation to 1

plt.hist(data, normed=1)

plt.title('How to plot a simple histogram in matplotlib ?', fontsize=10)

plt.savefig("plot_simple_histogramme_matplotlib_02.png")

plt.show()

Define the number of classes

By default the number of classes are determined by matplotlib, to define the number of classes just use the option "bins", example with bin = 2:

plt.hist(data, bins = 2)

plt.title('How to plot a simple histogram in matplotlib ?', fontsize=10)

plt.savefig("plot_simple_histogramme_matplotlib_03.png")

plt.show()

Define the number of classes and intervals

It is also possible to define the number of classes and intervals by providing a list to the option bins instead of just a number. For example with bin = [0,2,4,6] (with have 3 classes of interval [0,2[; [2,4[ et [4,6[)

plt.hist(data, bins = [0,2,4,6])

plt.title('How to plot a simple histogram in matplotlib ?', fontsize=10)

plt.savefig("plot_simple_histogramme_matplotlib_04.png")

plt.show()

It is interesting to compare with another histogram, example with another dataset:

data = [1,1,2,3,3,4,4,5,5]

plt.hist(data, bins = [0,2,4,6])

plt.title('How to plot a simple histogram in matplotlib ?', fontsize=10)

plt.savefig("plot_simple_histogramme_matplotlib_04.png")

plt.show()

Retrievals histogram parameters

In addition the function plt.hist() can return all the parameters of the histogram, example:

n, bins, patches = plt.hist(data, bins=[0,2,4,6])

print(n)
print(bins)
print(patches)

returns

[ 2.  3.  4.]
[0 2 4 6]
<a list of 3 Patch objects>

where n is the class frequencies, bins the intervals and patches the histogram formatting parameters (matplotlib.patches.Patch) for more information.

References