How to Generate Random Numbers in Python

If you generate random numbers for your job or academic courses, you will find this blog post useful.

Python random module

In this blog post, we will create random numbers in Python.

Randomness is found in everywhere, especially in machine learning. Cryptography is another area that uses randomness, in relation to ciphertext. Other areas using randomness are computer simulation, randomized design, slot machines and statistical sampling. A random process is a deterministic or indeterministic process for which what we do not know is greater than what we know. This delves into the Heisenberg's Uncertainty Principle, which is a fundamental concept in quantum physics.

Python uses the random module to generate random numbers. This is a built-in Python module. To use it, you have to import it using import random at the top of the Python program.

import python

The random module includes several functions. To view a list of functions in the random module, run this in the Python shell.

help(random)

You will see the description of the random module and a lot of info about it including description of the random.Random class and all methods in that class. There is also information about the classe random.SystemRandom and its methods.

Generate a random floating point number using random.random()

The random.random() function generates a float between 0 and 1.0, not including 1.0.

0.0 <= n < 1.0

Let us print a random floating point number less than 1.

Run this code in the Python shell or on VS Code and click on the Run button.

import random

print(random.random())

Your output will be something similar to this, less than 1.0, but most likely not this exact number.

0.44034633940339185

Generate a random integer between a and b using random.randint()

The random.randint(a, b) function returns a random integer which lies between a and b and inclusive of both.

a <= n <= b

Let us print a random number between 5 and 15.

import random

print(random.randint(5, 15))

OUTPUT:

12

Generate a random number from a range of integers using random.randrange()

The random.randrange(a, b) function generates a random number from a range of integers, we us.

If we want to generate a random integer between a and b, we can use random.randrange()

a <= n <= b

Let us print a random number between 12 and 25.

import random

print(random.randrange(12, 25))

This is identical to random.randint(), but with a difference. The third [and optional] parameter is the step.

Syntax:

random.randrange(start, stop[, step])

The step parameter increments the number by +step.

Let us print a random number among the multiples of 3 from 0 through 25.

The statement randrange(0, 25, 3) returns a range and is the equivalent of range(0, 25, 3). Let us see what we get when we apply list to it:

>>> list(range(0, 25, 3))
[0, 3, 6, 9, 12, 15, 18, 21, 24]

Type this into the Python shell:

import random

print(random.randrange(0, 25, 3))

OUTPUT:

18

Generate a list of 15 random integers from 10 through 50

If we want to create a list of 15 random numbers from 1 through 50, we can use randint() and combine it with list comprehension.

import random

random_numbers = [random.randint(1, 50) for _ in range(15)]
print(random_numbers)

OUTPUT:

[8, 4, 38, 36, 48, 39, 19, 37, 23, 20, 49, 45, 41, 8, 49]

Randomly shuffle a list of integers

The random.shuffle() function is used to randomly rearrange or shuffle the list of integers.

We will create a list of 20 random numbers from 40 through 75. We will randomly shuffle the list moons which contains [3, 5, 7, 2, 9, 4, 6, 2]

We will make it print the original list and the shuffled list.

import random

moons = [random.randrange(40, 75) for _ in range(20)]
print(moons)

random.shuffle(moons)
print(moons)

The shuffle() function does not return anything. It shuffles all the items in moons in-place. So, when you print the value of moons, you get a new order of the existing items.

OUTPUT:

[51, 65, 74, 71, 71, 49, 55, 40, 41, 71, 73, 67, 53, 44, 65, 56, 54, 49, 64, 62]
[71, 41, 51, 49, 62, 64, 40, 49, 71, 65, 56, 73, 55, 74, 67, 71, 53, 44, 54, 65]

Generate random floats using random.uniform()

The random.uniform(a, b) function returns a random float from a sequential list of numbers. This is very similar to random.randint().

a <= n <= b

Let us print a random float between 3 and 10.

import random

print(random.uniform(3, 10))

OUTPUT:

7.626657287029645

Select n different random items from without repitition

The random.sample(collection, k=n) function is used to select n different items from the list without repitition. Both parameters collection and k are mandatory.

For example, let us use random.sample() to randomly pick 3 numbers from this list.

grades = [90, 91, 68, 59, 12, 14, 99]

The code for that would be:

import random

grades = [90, 91, 68, 59, 12, 14, 99]
print(random.sample(grades, k=3))

OUTPUT:

[68, 59, 99]

If you do not enter a second parameter (for k), it will throw an error like this:

TypeError: Random.sample() missing 1 required positional argument: 'k'

Generate same sequence of numbers with random.seed

If you want to generate the same sequence of random numbers every time the program runs, set random.seed(n) where n can be a particular number.

import random

random.seed(10)
print(random.randint(10, 20))

random.seed(10)
print(random.randint(10, 20))

random.seed(10)
print(random.randint(10, 20))

OUTPUT:

19
19
19

As you can see, the output is exactly the same each time when you set random.seed() with value 10.

How secure is Python's random module?

It is not secure. That's why Python refers to randomly generated numbers as pseudo-random numbers if generated with the random module.

random uses the Mersenne Twister Number Generator, also known as PRNG (Pseudo Random Number Generator) that is fast. The problem is it is not truly random and can be guessed by a hacker or malicious actor.

If random.seed() is not called, the random number generator is seeded with random bits from the system time. This can be predictable.

Generate random string with os.urandom()

Instead of using the system time as seed, you can use os.urandom() or random.SystemRandom(). It is more unpredictable than the system time for getting the seed.

Also, os.urandom and random.SystemRandom() do not use the Mersenne Twister because they have to make system calls. Understandabily, they are slower, though more secure.

If you want to generate a random string of 16 bytes by making operating system calls, this is how you do it.

import os

print(os.urandom(16))

OUTPUT:

b'W\x9c\xde\xd0\xa3\x95\x07i\xff\x11J\xfb\xde\xca\xe4\xd8'

This is more secure than using system time.

Can I use the random module for security or cryptographic purposes?

No. The pseudo-random generators of the random module are meant for modeling and simulation. For generating cryptographically strong and random numbers to be used as passwords, account authentication, security tokens, etc, we should use the secret module.

You can read more about the secret module here.

How to replace random with secrets

It is easy to convert your existing code that uses the random module with the secrets module.

To find a random number between 0 and 10 using random:

import random

print(random.randint(10))

Now, the equivalent using secrets.

To find a random number between 0 and 10 using secrets:

import secrets

secrets.randbelow(10)

Conclusion

Let me know if this helped you in any way. If you would like more information about generating random numbers to be added, please email me or comment below. Thanks for reading.

Related Posts

If you have any questions, please contact me at arulbOsutkNiqlzziyties@gNqmaizl.bkcom. You can also post questions in our Facebook group. Thank you.