Numpy.ones(): Creating Arrays of Ones

Study Notes

Numpy.ones(): Creating Arrays of Ones

numpy.ones() is a function from the NumPy library that returns an array filled with ones. It's useful when you need to create an array with specific dimensions or shape. Here's how it works:

Array Creation

The most basic usage of numpy.ones() involves specifying the size of the resulting array. For example:

import numpy as np

## Create an array of ones with 5 elements
ones_array = np.ones((5,))

In this case, np.ones((5,)) creates an array of size 5 with all elements equal to 1. The (5,) is a tuple representing the shape of the array.

Scalar Creation

You can also use numpy.ones() to create a single scalar value. For instance:

import numpy as np

## Create a scalar value of 1
ones_scalar = np.ones(())

Here, np.ones(()) creates a single scalar value of 1. The empty tuple () denotes a scalar.

Array Resizing

You can use numpy.ones() with arrays to create a new array with the same shape but filled with ones. This can be useful when you need to make copies of existing arrays while preserving their structure:

import numpy as np

## Create a pre-existing array
array = np.array([1, 2])

## Use numpy.ones() to duplicate the array's shape
new_array = np.ones((len(array),)) * array

In this example, new_array is a copy of array, but all its elements are replaced with ones.

Memory Efficiency

Using numpy.ones() to fill whole rows or columns of matrices is more memory efficient than creating a new matrix and filling it with zeros followed by assigning zeros to specific values. For instance, if we want to add a constant 1 to every element in the first column of a matrix mat, we can do the following:

import numpy as np

## Fill the first column with ones
ones_column = np.ones((mat.shape, 1))
result = mat + ones_column

Here, ones_column contains ones along one dimension, so adding it to mat adds a constant value of 1 to every element in the first column of mat. This approach avoids the creation of potentially large temporary arrays created by other methods.

Description

numpy.ones() is a NumPy function used to create arrays filled with ones. Learn how to create arrays of specific dimensions, scalar values, and efficiently duplicate array shapes. Explore memory-efficient ways to use numpy.ones() for matrix operations.