NumPy sum()

import numpy as np

array1 = np.array([1, 2, 3, 4, 5])

# use sum() to calculate sum of array1 elements
result = np.sum(array1)

print(result)

# Output : 15

sum() Syntax

The syntax of sum() is:

numpy.sum(array, axis=None, dtype=None, out=None, keepdims=<no value>)

sum() Arguments

The sum() function takes following arguments:

• array - the input array
• axis (optional) - the axis along which the sum is calculated
• dtype (optional) - the data type of the returned sum
• out (optional) - the output array where the result will be stored
• keepdims (optional) - whether to preserve the input array's dimension (bool)

sum() Return Value

The sum() function returns the sum of array elements

Example 1: sum() With 2-D Array

The axis argument defines how we can find the sum of elements in a 2-D array.

• If axis = None, the array is flattened and the sum of the flattened array is returned.
• If axis = 0, the sum is calculated column-wise.
• If axis = 1, the sum is calculated row-wise.

import numpy as np

array = np.array([[10, 17, 25], 
  [15, 11, 22]])
                  
# return the sum of elements of the flattened array
result1 = np.sum(array)

print('The sum of flattened array: ', result1)

# return the column-wise sum
result2 = np.sum(array, axis = 0)

print('Column-wise sum (axis 0): ', result2)

# return the row-wise sum
result2 = np.sum(array, axis = 1)

print('Row-wise sum (axis 1): ', result2)

Output

The sum of flattened array:  100
Column-wise sum (axis 0):  [25 28 47]
Row-wise sum (axis 1):  [52 48]

Example 2: Use of out Argument in sum()

import numpy as np

array1 = np.array([[10, 17, 25], 
                 	    [15, 11, 22], 
                  	    [11, 19, 20]])

# create an empty array
array2= np.array([0, 0, 0])

# pass the 'out' argument to store the result in array2
np.sum(array1, axis = 0, out = array2)

print(array2)

Output

[36 47 67]

Here, after specifying out=array2, the result of sum of array1 along axis=0 is stored in the array2 array.

Example 3: sum() With keepdims

When keepdims = True, the dimensions of the resulting array matches the dimension of an input array.

import numpy as np

array1 = np.array([[10, 17, 25], 
            		    [15, 11, 22]])
print('Dimensions of original array: ', array1.ndim)

result = np.sum(array1, axis = 1)

print('\n Without keepdims: \n', result)
print('Dimensions of array: ', result.ndim) 

# set keepdims to True to retain the dimension of the input array
result = np.sum(array1, axis = 1, keepdims = True)

print('\n With keepdims: \n', result)
print('Dimensions of array: ', result.ndim)

Output

Dimensions of original array:  2

 Without keepdims: 
 [52 48]
Dimensions of array:  1

 With keepdims: 
 [[52]
 [48]]
Dimensions of array:  2

Without keepdims, the result is simply a one-dimensional array of indices.

With keepdims, the resulting array has the same number of dimensions as the input array.