Python Data Science NumPy Random Zipf Data Distribution

Zipf's Law: states that in a collection the nth common term is 1/n times of the most common term.

E.g. 5th common word in english occurs nearly 1/5th times as the most used word.

It has the following two parameters.

1. a - distribution parameter.

2. size - The shape of the returned array.

Example 1: Draw out a sample for zipf distribution with distribution parameter of 2 with size 2x4.

Code

from numpy import random

x = random.zipf(a=2, size=(2, 4))

print(x)

the output will be

[[26  1 25  1]
 [ 1  1  2 33]]

Note: Every time the code is run the output may vary because of random generation.

Visualization of Zipf Distribution

Sample 100000 points but plotting only ones with value < 10 for more meaningful chart.

Example 2

Code

from numpy import random
import matplotlib.pyplot as plt
import seaborn as sns

x = random.zipf(a=2, size=100000)
sns.distplot(x[x<10], kde=False)

plt.show()

the output will be