I would expect every column to be centered on top of their bin number. Instead, the 1 and 2 bars are right of the number, the 3rd on to the left. Why is it not even consistent?
import matplotlib.pyplot as plt
degrees = [1, 1, 2, 2, 2, 2, 3]
plt.hist(degrees)
plt.show()
Short answer: It is not supposed to, use plt.bar() instead. For longer explanation, please read below.
The purpose of a histogram is to approximate the distribution of the data. For example
import numpy as np
plt.hist(np.random.normal(3, 7, 100))
which gives
Now, when you have much less data, which is integer valued, and call
plt.hist([1, 1, 2, 2, 2, 2, 3])
you also get approximation of the distribution of the data you provided. With the default parameters it looks like this:
The documentation of hist tells that
bins, it will default to 10.range, it will default to min and max of your dataTherefore, your data will be put inside 10 bins, with min at 1 and max at 3. These bins will be
In [45]: np.linspace(1,3, 11)
Out[45]: array([1. , 1.2, 1.4, 1.6, 1.8, 2. , 2.2, 2.4, 2.6, 2.8, 3. ])
Since you have only data inside bins 1.0 - 1.2, 2.0 - 2.2 and 2.8 - 3.0, you will see three bars centered at 1.1, 2.1 and 2.9.
If your data is integer (categorical) valued, like
degrees = [1, 1, 2, 2, 2, 2, 3]
and you want to know the relative sizes of these categories, you probably want to create a bar plot instead.
import matplotlib.pyplot as plt
from collections import Counter
degrees = [1, 1, 2, 2, 2, 2, 3]
counts = Counter(degrees)
plt.bar(counts.keys(), counts.values())
plt.show()
Because you didn't specify the bins. If you add the bins in:
degrees = [1, 1, 2, 2, 2, 2, 3]
plt.hist(degrees, bins=[1,2,3,4])
plt.show()
