Python - Random

Overview

Estimated time: 20–30 minutes

The random module provides pseudo-random numbers for simulations and sampling. Learn the core APIs, reproducibility with seeds, and when to use secrets instead.

Learning Objectives

• Generate random numbers, choices, and shuffles.
• Use random.Random() instances for isolated RNG streams and reproducibility.
• Understand the difference between random (PRNG) and secrets (crypto-grade).

Prerequisites

• Basic Python lists and numbers

Examples

import random

random.seed(42)  # global seed
print(random.random())      # float in [0.0, 1.0)
print(random.randint(1, 6)) # inclusive
print(random.choice(['a','b','c']))
print(random.sample(range(10), k=3))

# Independent RNG instance
rng = random.Random(123)
print(rng.random())

items = [1,2,3,4]
random.shuffle(items)
print(items)

Expected Output (deterministic with seed): values will be stable for given seeds.

Common Pitfalls

• Security: do not use random for tokens/keys; use secrets.token_hex, etc.
• Seeding the global generator affects all users of random in your process; prefer per-instance RNGs in libraries.
• Shuffling in place mutates the list; copy first if you need the original.

Best Practices

• Use random.Random(seed) for reproducible experiments and tests.
• Document seeds alongside results to enable reproducibility.
• Use secrets for any security-relevant randomness.

Checks for Understanding

1. How to get a reproducible random sample of 5 integers in [0, 99]?
2. Which module should you use to generate a secure token for a password reset link?

Exercises

1. Simulate rolling two dice 10,000 times and estimate the probability distribution.
2. Write a function that returns a shuffled copy of a list without mutating the original.