Generating duplicates with weights (python)

Question

I have the following list:

priority=['Emergency','High','Normal','Low']

I need to generate a list of 15000 values in the list above, with frequency of each of them being 10%,20%,30%,40%. Can somebody show me how to do it?

Answer 1

You can use random.random(), and assign indexes based on the range of the random number.

In [21]: priority=['Emergency','High','Normal','Low']

In [22]: idx = lambda x: 0 if x < 0.1 else (1 if x < 0.3 else (2 if x < 0.6 else 3))

In [23]: import random

In [24]: [priority[idx(random.random())] for _ in range(15000)]
Out[24]: 
['Normal',
 'High',
 'Emergency',
 'Emergency',
 'High',
 .
 .

In the above, I've relied on the order of the elements within your list to generate indexes, such that for 0 to 0.1 it is the first (10%), 0.1 to 0.3 its the second (0.3 - 0.1 will cover 20%), and so on.

Answer 2

Probably the easiest way is to use choice function of numpy.random.

import numpy as np

priority=['Emergency','High','Normal','Low']
b = np.random.choice(priority, 150, replace=True, p=[0.1, 0.2, 0.3, 0.4])

print(b)

Built-in random.choices does the same thing but it is not available <3.6.