.ones()

In NumPy, the .ones() function creates a new array of the given shape and type, filled with ones.

This function is particularly useful when there is a need to initialize an array with a placeholder value of 1, which might be multiplied by another value or used in various numerical computations later.

Syntax

numpy.ones(shape, dtype=None, order='C', *, like=None)

Parameters:

• shape (int or tuple of int): Defines the dimensions of the new array.
  • If an int, a 1-D array of that length is created.
  • If a tuple of ints, an array with those dimensions is created (e.g., (2, 3) for a 2x3 array).
• dtype (optional): The desired data type for the array elements. Defaults to float64. Examples: int32, bool_, float.
• order: ({‘C’, ‘F’}, optional): Controls whether the array is stored in row-major or column-major memory layout.
  • 'C' (default): Row-major (C-style)
  • 'F': Column-major (Fortran-style)
• like(array_like, optional): Reference object for creating arrays not strictly of NumPy type. If None, a NumPy array is returned.

Return value:

• out (ndarray): An array of ones of the given shape, dtype, and order.

Example

This example demonstrates the usage of the .ones() function:

import numpy as np

# Create a 1-D array with 5 ones (default dtype is float64)
arr1 = np.ones(5)
print("Array 1:\n", arr1)
print("Data type of Array 1:", arr1.dtype)
print("Shape of Array 1:", arr1.shape)

# Create a 2-D array (3 rows, 4 columns) of ones with integer type
arr2 = np.ones((3, 4), dtype=int)
print("\nArray 2:\n", arr2)
print("Data type of Array 2:", arr2.dtype)
print("Shape of Array 2:", arr2.shape)

# Create a 3-D array of ones
arr3 = np.ones((2, 3, 2))
print("\nArray 3:\n", arr3)
print("Data type of Array 3:", arr3.dtype)
print("Shape of Array 3:", arr3.shape)

Copy to clipboard
Copy to clipboard

Here is the output:

Array 1:
 [1. 1. 1. 1. 1.]
Data type of Array 1: float64
Shape of Array 1: (5,)

Array 2:
 [[1 1 1 1]
  [1 1 1 1]
  [1 1 1 1]]
Data type of Array 2: int64
Shape of Array 2: (3, 4)

Array 3:
 [[[1. 1.]
   [1. 1.]
   [1. 1.]]

  [[1. 1.]
   [1. 1.]
   [1. 1.]]]
Data type of Array 3: float64
Shape of Array 3: (2, 3, 2)

Copy to clipboard
Copy to clipboard

Codebyte Example: Add Bias Term to a Feature Matrix in Machine Learning

In many machine learning algorithms (like linear regression), there is a need to add a bias column of 1s to the input feature matrix. Here’s how numpy.ones() can be used:

Visit us
Visit us
Hide code
Code
Output
Hide output
Loading...
Copy to your clipboard
Copy Codebyte
Run