Generating Random Data with NumPy

In this tutorial, we’ll explore how to generate random data using NumPy, covering basic random number generation, statistical distributions, and ensuring reproducibility through seed setting.

1. Basic Random Number Generation

NumPy provides simple functions for generating random numbers quickly. Two primary methods for basic random number generation are np.random.rand() and np.random.randint().

Example: np.random.rand()

This function generates random floating-point numbers uniformly distributed between 0 (inclusive) and 1 (exclusive).

Code Example:

import numpy as np

random_array = np.random.rand(3, 4)
print(random_array)

Explanation of parameters:

• The arguments (3, 4) represent the dimensions of the output array. The example above creates a 3-row by 4-column array of random numbers.

Example: np.random.randint()

Use this function when you need random integers between a specified range.

Code Example:

import numpy as np

random_integers = np.random.randint(10, 20, size=(5,))
print(random_integers)

Explanation of parameters:

• The first parameter (low=10) specifies the lowest integer value (inclusive).
• The second parameter (high=20) specifies the upper bound integer value (exclusive).
• The size=(5,) parameter defines the shape of the array, here generating a 1-dimensional array with 5 elements.

2. Statistical Distributions

NumPy offers methods to generate random numbers from specific statistical distributions. Two popular distributions are normal (Gaussian) and uniform distributions.

Example: np.random.normal()

Generate random numbers following a normal (Gaussian) distribution, defined by its mean and standard deviation.

Code Example:

import numpy as np

normal_data = np.random.normal(loc=0, scale=1, size=(1000,))
print(normal_data)

Explanation of parameters:

• loc specifies the mean (center) of the distribution (default is 0).
• scale specifies the standard deviation (spread) of the distribution (default is 1).
• size defines the shape of the output array.

Example: np.random.uniform()

Generate random numbers uniformly distributed between specified limits.

import numpy as np

uniform_data = np.random.uniform(low=-1, high=1, size=(10,))
print(uniform_data)

Explanation of parameters:

• low is the lower bound of the interval (inclusive).
• high is the upper bound of the interval (exclusive).
• size specifies the output shape.

3. Setting a Seed for Reproducibility

When generating random data, sometimes you’ll want to reproduce the same results for testing or analysis purposes. NumPy allows you to set a random seed using np.random.seed().

Example: np.random.seed()

import numpy as np

np.random.seed(42)  # Setting the seed for reproducibility
reproducible_data = np.random.rand(3)
print(reproducible_data)

Explanation of parameters:

• The argument to np.random.seed() is an integer value that initializes the random number generator. The same seed will always produce the same sequence of random numbers.

Summary

• Basic Functions:
  • np.random.rand() generates uniform floating-point numbers between 0 and 1.
  • np.random.randint() generates random integers within a specified range.
• Statistical Distributions:
  • np.random.normal() generates numbers from a normal distribution.
  • np.random.uniform() generates numbers from a uniform distribution.
• Reproducibility:
  • np.random.seed() ensures your random numbers can be replicated.