NUMPY Tutorial

🔹 NumPy Random Module

Generating random numbers is essential in simulations, games, testing, and many data science applications. NumPy provides a powerful random module with many functions for generating random numbers, arrays, and distributions.

1. Importing NumPy Random

import numpy as np
# You can use random functions with np.random

2. Generating Random Scalars

• np.random.rand(): Uniform distribution over [0, 1)
• np.random.randint(low, high): Random integer between low (inclusive) and high (exclusive)
• np.random.random(): Another way to get float in [0, 1)

# Random float between 0 and 1
print(np.random.rand())  # e.g., 0.374540

# Random integer between 10 and 20
print(np.random.randint(10, 20))  # e.g., 13

# Another random float [0,1)
print(np.random.random())  # e.g., 0.950714

3. Generating Random Arrays

You can generate arrays filled with random values:

# 1D array of 5 random floats [0,1)
arr1 = np.random.rand(5)
print(arr1)

# 2D array 3x4 with random floats
arr2 = np.random.rand(3, 4)
print(arr2)

# 1D array of 5 random integers between 0 and 10
arr3 = np.random.randint(0, 10, size=5)
print(arr3)

4. Random Samples from Distributions

NumPy lets you sample from many common probability distributions:

• np.random.normal(loc=0.0, scale=1.0, size=None): Normal (Gaussian) distribution with mean loc and std deviation scale.
• np.random.uniform(low=0.0, high=1.0, size=None): Uniform distribution over [low, high).
• np.random.choice(a, size=None, replace=True, p=None): Random sample from 1D array a.

# 5 samples from normal distribution (mean=0, std=1)
samples = np.random.normal(0, 1, 5)
print(samples)

# 5 samples from uniform distribution between 10 and 20
uniform_samples = np.random.uniform(10, 20, 5)
print(uniform_samples)

# Random choice from list ['a','b','c','d'], 4 picks, no replacement
choices = np.random.choice(['a', 'b', 'c', 'd'], 4, replace=False)
print(choices)

5. Setting Seed for Reproducibility

Random numbers are generated using a pseudo-random number generator. To get the same random numbers every run (helpful for debugging), you can set a seed:

np.random.seed(42)  # Fix the seed number
print(np.random.rand(3))  # Will always generate same output on every run

6. Summary of Useful Functions

7. Practical Use Cases

• Simulating random processes and experiments
• Creating random datasets for testing algorithms
• Random sampling and bootstrapping in statistics
• Random initialization in machine learning models