NumPy ndarray.sum() Method

NumPy ndarray.sum()

The numpy.ndarray.sum() method calculates the sum of array elements along a specified axis. It can sum all elements in the array or along specific dimensions, and it supports additional parameters for customization.

Syntax

ndarray.sum(axis=None, dtype=None, out=None, keepdims=False, initial=0, where=True)

Parameters

Parameter	Type	Description
axis	None, int, or tuple of ints, optional	Axis or axes along which a sum is performed. If None, sums all elements.
dtype	dtype, optional	Specifies the data type of the output array and accumulator.
out	ndarray, optional	Alternative output array to store the result.
keepdims	bool, optional	If True, preserves the reduced dimensions as size one.
initial	scalar, optional	Starting value for the sum, useful for initializing reductions.
where	array_like of bool, optional	Specifies elements to include in the sum.

Return Value

Returns the sum of the array elements as a scalar if axis=None, or as an array if an axis is specified. The dtype of the result depends on the input and the dtype parameter.

---

Examples

1. Summing All Elements in ndarray

This example demonstrates summing all elements in a NumPy array.

import numpy as np

# Creating a 2D array
arr = np.array([[1, 2], 
                [3, 4]])

# Summing all elements
result = arr.sum()
print(result)  # Output: 10

Output:

10

2. Using the axis Parameter in ndarray.sum()

Summing elements along a specific axis.

import numpy as np

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

# Summing along axis 0 (columns)
result_axis0 = arr.sum(axis=0)
print(result_axis0)  # Output: [4 6]

# Summing along axis 1 (rows)
result_axis1 = arr.sum(axis=1)
print(result_axis1)  # Output: [3 7]

Output:

[4 6]
[3 7]

3. Changing Data Type with dtype

Setting the dtype to control precision.

import numpy as np

arr = np.array([1, 2, 3, 4], dtype=np.int8)

# Summing with a higher precision dtype
result = arr.sum(dtype=np.int32)
print(result)  # Output: 10 (stored as int32)

Output:

10

4. Keeping Dimensions with keepdims=True

Retaining reduced dimensions for broadcasting compatibility.

import numpy as np

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

# Summing along axis 1 with keepdims=True
result = arr.sum(axis=1, keepdims=True)
print(result)
# Output:
# [[3]
#  [7]]

Output:

[[3]
 [7]]

5. Using the initial Parameter in ndarray.sum()

Starting the sum with an initial value.

import numpy as np

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

# Summing with an initial value of 10
result = arr.sum(initial=10)
print(result)  # Output: 20

Output:

20

Sum = initial value + 1 + 2 + 3 + 4 = 10 + 1 + 2 + 3 + 4 = 20

6. Using the where Parameter in ndarray.sum()

Summing only selected elements using a condition.

import numpy as np

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

# Mask: Only sum even numbers
mask = (arr % 2 == 0)

result = arr.sum(where=mask)
print(result)  # Output: 6 (2 + 4)

Output:

6