Numpy split() Function

The Numpy split() Function divides an array into multiple subarrays along a specified axis. It can split the array into equal-sized subarrays if given an integer or at specified indices if given a list.

This function is particularly useful for breaking down data into smaller chunks for easier manipulation and analysis.

Syntax

The syntax for the Numpy split() function is as follows −

numpy.split(ary, indices_or_sections, axis=0)

Parameters

Below are the parameters of the Numpy split() Function −

• ary(array_like): The input array to be split.
• indices_or_sections(int or 1-D array): If an integer(N), the array will be divided into N equal sections. If an array, the values in the array indicate where to split the array.
• axis(int, optional): The axis along which to split, default is 0 i.e. split along rows.

Return Value

This Function returns a list of sub-arrays resulting from the split.

Example 1

Following is the example of the shows Numpy split() Function which helps in splitting an array of 9 elements into 3 equal sub-arrays −

import numpy as np

# Create an array
arr = np.arange(9)
print("Original array:")
print(arr)

# Split the array into 3 equal parts
result = np.split(arr, 3)
print("\nSplit array into 3 equal parts:")
for i, sub_array in enumerate(result):
    print(f"Sub-array {i+1}:")
    print(sub_array)

Output

Original array:
[0 1 2 3 4 5 6 7 8]

Split array into 3 equal parts:
Sub-array 1:
[0 1 2]
Sub-array 2:
[3 4 5]
Sub-array 3:
[6 7 8]

Example 2

Here in this example we show how to split a 2D array into 2 sub-arrays along the columns i.e. axis=1 −

import numpy as np

# Create a 2D array
arr = np.arange(16).reshape(4, 4)
print("Original 2D array:")
print(arr)

# Split the 2D array into 2 sub-arrays along columns
result = np.split(arr, 2, axis=1)
print("\nSplit 2D array into 2 sub-arrays along columns:")
for i, sub_array in enumerate(result):
    print(f"Sub-array {i+1}:")
    print(sub_array)

Output

Original 2D array:
[[ 0  1  2  3]
 [ 4  5  6  7]
 [ 8  9 10 11]
 [12 13 14 15]]

Split 2D array into 2 sub-arrays along columns:
Sub-array 1:
[[ 0  1]
 [ 4  5]
 [ 8  9]
 [12 13]]
Sub-array 2:
[[ 2  3]
 [ 6  7]
 [10 11]
 [14 15]]