How to Get the N-Largest Values of an Array using NumPy

The numpy.argsort() function can be used to find the n-largest values in an array. Instead of directly returning the sorted array, argsort() returns the indices that would sort the array. Using these indices with slicing, we can extract the top n largest elements efficiently.

In this example, we extract the 1 largest value from a NumPy array.

import numpy as np

arr = np.array([5, 1, 8, 3])
n = 1

idx = np.argsort(arr)
res = arr[idx][-n:]
print(n, "largest value:", res)

1 largest value: [8]

Explanation:

• np.argsort(arr) -> returns indices [1, 3, 0, 2]
• Sorting gives [1, 3, 5, 8]
• Last element -> [8]

Syntax

numpy.argsort(arr, axis=-1, kind='quicksort', order=None)

Parameters:

• arr: Input array.
• axis: Axis along which to sort. Default is -1 (last axis).
• kind: Sorting algorithm ('quicksort', 'mergesort', 'heapsort', 'stable').
• order: When sorting structured arrays, specifies field(s) to sort by.

Return Value:

• Returns an array of indices that would sort arr.
• Using these indices, we can reconstruct the sorted array or extract top n elements.

Examples

Example 1: This example extracts the 2 largest values from a NumPy array.

import numpy as np

arr = np.array([2, 0, 1, 5, 4, 1, 9])
print("Array:", arr)

idx = np.argsort(arr)
sorted_arr = arr[idx]

n = 2
res = sorted_arr[-n:]
print(n, "largest values:", res)

Array: [2 0 1 5 4 1 9]
2 largest values: [5 9]

Explanation:

• Sorted array -> [0, 1, 1, 2, 4, 5, 9].
• Last 2 elements -> [5, 9].

Example 2: This example extracts the top 3 values from a NumPy array.

import numpy as np

arr = np.array([2, 0, 1, 5, 4, 1, 9])
print("Array:", arr)

idx = np.argsort(arr)
sorted_arr = arr[idx]

n = 3
res = sorted_arr[-n:]
print(n, "largest values:", res)

Array: [2 0 1 5 4 1 9]
3 largest values: [4 5 9]

Explanation:

• Sorted array -> [0, 1, 1, 2, 4, 5, 9].
• Last 3 elements -> [4, 5, 9].

Example 3: This example shows how to flatten a 2D array and find the 2 largest values.

import numpy as np

arr = np.array([[3, 7, 1], [9, 2, 6]])
print("2D Array:\n", arr)

idx = np.argsort(arr, axis=None)
sorted_arr = arr.flatten()[idx]

n = 2
res = sorted_arr[-n:]
print(n, "largest values:", res)

2D Array:
 [[3 7 1]
 [9 2 6]]
2 largest values: [7 9]

Explanation:

• Flattened array -> [3, 7, 1, 9, 2, 6].
• Sorted array -> [1, 2, 3, 6, 7, 9].
• Last 2 values -> [7, 9].