Combining fancy and slice indexing in NumPy

Introduction

We use combining fancy and slice indexing to pick specific rows or columns and also select ranges of data at the same time. This helps us get exactly the data we want from a big table or array.

You want to select specific rows by their index but also want a continuous range of columns.

You need to extract certain elements from a matrix where rows are chosen by a list and columns by a slice.

You want to analyze parts of data by mixing exact positions and ranges.

You want to quickly grab multiple parts of data without looping.

Syntax

NumPy

array[fancy_index, slice]
# or
array[slice, fancy_index]

Fancy indexing means using a list or array of indices to pick specific elements.

Slice means selecting a continuous range using start:stop:step.

Examples

Select rows 0 and 2, and columns 1 to 2 (slice excludes stop index 3).

NumPy

import numpy as np
arr = np.arange(16).reshape(4,4)
rows = [0, 2]
cols = slice(1, 3)
result = arr[rows, cols]
print(result)

Select rows 2 to 4 and columns 1 and 3.

NumPy

arr = np.arange(20).reshape(5,4)
cols = [1, 3]
rows = slice(2, 5)
result = arr[rows, cols]
print(result)

Sample Program

This program creates a 5x5 grid of numbers. It picks rows 1, 3, and 4 using a list (fancy indexing) and columns 2 to 4 using a slice. Then it prints the original array and the selected part.

NumPy

import numpy as np

# Create a 5x5 array with numbers 0 to 24
arr = np.arange(25).reshape(5, 5)

# Fancy index: select rows 1, 3, and 4
rows = [1, 3, 4]

# Slice: select columns from 2 to 4 (2, 3, 4)
cols = slice(2, 5)

# Combine fancy indexing for rows and slice for columns
selected = arr[rows, cols]

print("Original array:\n", arr)
print("\nSelected rows and columns:\n", selected)

Important Notes

When combining fancy and slice indexing, the fancy index always applies to the first dimension you specify.

The result shape depends on the fancy index length and the slice length.

Mixing fancy and slice indexing is faster than looping over elements.

Summary

Fancy indexing uses lists or arrays of indices to pick specific rows or columns.

Slices select continuous ranges of rows or columns.

Combining them lets you pick exact rows and ranges of columns (or vice versa) easily.