NumPy Sorting Arrays

Welcome to The Coding College, where coding concepts are made easy for everyone! In this article, we’ll cover sorting arrays in NumPy, a crucial operation for organizing data efficiently. Sorting arrays helps make data analysis and manipulation more systematic and accessible.

Why Sort Arrays in NumPy?

Sorting is essential for:

• Organizing data in ascending or descending order.
• Preparing datasets for algorithms that require sorted inputs.
• Improving data readability and interpretation.

NumPy provides the sort() function for sorting arrays in Python.

1. Sorting 1D Arrays

The sort() function arranges elements in ascending order by default.

Syntax

numpy.sort(array, axis=-1, kind=None, order=None)

• array: The array to be sorted.
• axis: Axis along which to sort. Default is -1 (last axis).
• kind: Sorting algorithm. Options:
  • 'quicksort' (default)
  • 'mergesort'
  • 'heapsort'
  • 'stable'
• order: For structured arrays, specifies the field to sort.

Example: Sort a 1D Array

import numpy as np

arr = np.array([50, 30, 20, 10, 40])
sorted_arr = np.sort(arr)
print(sorted_arr)

Output:

[10 20 30 40 50]

2. Sorting 2D Arrays

When working with multidimensional arrays, the axis parameter determines the sorting direction.

Example: Sort Along Rows (Default)

arr = np.array([[3, 2, 1], [6, 5, 4]])
sorted_arr = np.sort(arr)
print(sorted_arr)

Output:

[[1 2 3]
 [4 5 6]]

Example: Sort Along Columns (Axis = 0)

sorted_arr = np.sort(arr, axis=0)
print(sorted_arr)

Output:

[[3 2 1]
 [6 5 4]]

3. Using the argsort() Function

The argsort() function returns the indices that would sort the array.

Example: Get Sorted Indices of a 1D Array

arr = np.array([50, 30, 20, 10, 40])
indices = np.argsort(arr)
print(indices)

Output:

[3 2 1 4 0]

The indices [3, 2, 1, 4, 0] correspond to the sorted order [10, 20, 30, 40, 50].

Example: Sort Array Using Indices

sorted_arr = arr[indices]
print(sorted_arr)

Output:

[10 20 30 40 50]

4. Sorting in Descending Order

To sort in descending order, combine sort() with slicing.

Example: Sort 1D Array in Descending Order

arr = np.array([50, 30, 20, 10, 40])
sorted_desc = np.sort(arr)[::-1]
print(sorted_desc)

Output:

[50 40 30 20 10]

Example: Sort 2D Array in Descending Order Along Rows

arr = np.array([[3, 2, 1], [6, 5, 4]])
sorted_desc = np.sort(arr)[:, ::-1]
print(sorted_desc)

Output:

[[3 2 1]
 [6 5 4]]

5. Sorting Structured Arrays

For structured arrays (arrays with named fields), you can use the order parameter to sort by a specific field.

Example: Sort by a Field

data = np.array([(1, 'Alice', 50), (2, 'Bob', 20), (3, 'Charlie', 30)],
                dtype=[('id', 'i4'), ('name', 'U10'), ('score', 'i4')])
sorted_data = np.sort(data, order='score')
print(sorted_data)

Output:

[(2, 'Bob', 20) (3, 'Charlie', 30) (1, 'Alice', 50)]

Comparison of Sorting Algorithms

Practical Use Cases

1. Data Analysis: Organize data for better visualization.
2. Machine Learning: Preprocess datasets for algorithms requiring sorted input.
3. Rankings: Create ordered rankings for evaluations or scores.

Summary

Sorting arrays in NumPy is straightforward and highly customizable. Whether you need simple ascending sorting, custom ordering with indices, or efficient handling of structured data, NumPy’s sorting functions have you covered.

Explore more tutorials and tips on The Coding College and master Python programming with ease!