Numpy Array Indexing

Last Updated : 24 Jan, 2025

Array Indexing in NumPy is used to access or modify specific elements of an array. It allows to retrieve data from arrays by specifying the positions (indices) of elements. This is the important feature in NumPy that enables efficient data manipulation and analysis for numerical computing tasks.

Accessing 1D Array

In a 1D array, we can access elements using their zero-based index:

Python

import numpy as np  # Create a 1D NumPy array with five elements arr = np.array([10, 20, 30, 40, 50])  # Access and print the first element of the array print(arr[0])

Output

Table of Content

Access 2-D Array

To access elements, you specify the row index first and the column index second.

Python

import numpy as np   # Define a 2D array matrix = np.array([[1, 2, 3], [4, 5, 6], [7, 8, 9]])  # Access the element at row 1, column 2  print(matrix[1, 2])

Output

Accessing 3D Arrays

A 3D array can be visualized as a stack of 2D arrays. We need three indices:

Depth: Specifies the 2D slice.
Row: Specifies the row within the slice.
Column: Specifies the column within the row.

Python

import numpy as np  cube = np.array([[[1, 2, 3],                   [4, 5, 6],                   [7, 8, 9]],                                    [[10, 11, 12],                   [13, 14, 15],                   [16, 17, 18]]])  # Access the element at depth 1, row 2, column 0 print(cube[1, 2, 0])

Output

Slicing Arrays

We can extract a subset of elements using the start:stop:step syntax:

Python

import numpy as np  # Create a 1D NumPy array with six elements arr = np.array([0, 1, 2, 3, 4, 5])  # Use slicing to access a subset of the array print(arr[1:4])

Output

[1 2 3]

For multidimensional arrays, slicing can be applied to each dimension:

Python

import numpy as np  # Create a 2D NumPy array (matrix) with three rows and three columns matrix = np.array([[1, 2, 3], [4, 5, 6], [7, 8, 9]])  # Use slicing to extract a submatrix print(matrix[0:2, 1:3])

Output

[[2 3]  [5 6]]

Boolean Indexing

Boolean indexing allows we to filter elements based on conditions:

Python

import numpy as np  # Create a 1D NumPy array with five elements arr = np.array([10, 15, 20, 25, 30])  # Use boolean indexing to filter elements greater than 20 print(arr[arr > 20])

Output

[25 30]

Use logical operators like & (and), | (or), and ~ (not):

Python

import numpy as np   # Create a 1D NumPy array with five elements arr = np.array([10, 15, 20, 25, 30])  # Use boolean indexing with multiple conditions to filter elements print(arr[(arr > 10) & (arr < 30)])

Output

[15 20 25]

Ellipsis (...) in Indexing

The ellipsis (...) is a shorthand for selecting all dimensions not explicitly mentioned:

Python

import numpy as np  cube = np.random.rand(4, 4, 4)  # Selects the first "slice" along the last axis print(cube[..., 0])

Output

[[0.92937637 0.11283562 0.87685485 0.1698762 ]  [0.48812469 0.89827444 0.6408755  0.8818098 ]  [0.97334939 0.51590585 0.2744543  0.4618385 ]  [0.74940947 0.63351317 0.44032189 0.237636  ]]

Using np.newaxis

The np.newaxis keyword adds a new axis, increasing the array's dimensions:

Python

import numpy as np  # Create a 1D NumPy array with three elements arr = np.array([1, 2, 3])  # Add a new axis to convert the 1D array into a 2D column vector # arr[:, np.newaxis] inserts a new axis along the second dimension  print(arr[:, np.newaxis])

Output

[[1]  [2]  [3]]

Modifying Array Elements

NumPy allows direct modification of array elements using indexing:

Python

import numpy as np   # Create a 1D NumPy array with four elements arr = np.array([1, 2, 3, 4])  # Modify elements in the array using slicing arr[1:3] = 99  # Print the updated array to see the changes print(arr)

Output

[ 1 99 99  4]

NumPy Array Functions

kumar_satyam

Improve

Article Tags :

Practice Tags :

python

Numpy Array Indexing

Accessing 1D Array

Access 2-D Array

Accessing 3D Arrays

Slicing Arrays

Boolean Indexing

Ellipsis (...) in Indexing

Using np.newaxis

Modifying Array Elements

Similar Reads