What is the numpy.nansum() function in Numpy?

Overview

The numpy.nansum() function in NumPy is used to return the sum of elements in a given array over a given axis in such a way that the NaNs are treated as zeroes.

Syntax

numpy.nansum(a, axis=None, dtype=None, out=None)

Parameter

The numpy.nansum() function takes the following parameter values:

  • a (required): This is the input array containing numbers to be computed.
  • axis (optional): This is the axis along which the product is determined.
  • dtype (optional): This is the data-type of the output array.
  • out (optional): This is the alternate array where the result is placed.

Return value

The numpy.nansum() function returns an output array with the result.

Example

Let’s look at the code below:

import numpy as np
# creating an array
x = np.array([1, 2, np.nan, 2, np.nan])
# Implementing the nansum() function 
myarray = np.nansum(x, axis=0)
print(x)
print(myarray)

Explanation

  • Line 1: We import the numpy module.
  • Line 4: We create an array, x, using the array() method.
  • Line 7: We implement the np.nansum() function on the array. The result is assigned to a variable, myarray.
  • Line 9: We print the input array x.
  • Line 10: We print the variable myarray.

Note: The output 4 is obtained from the sum of the elements: 1+2+np.nan+2+np.nan = 1+2+0+2+0 = 4. This is because Nans values are treated as 0.

Free Resources

License: Creative Commons-Attribution-ShareAlike 4.0 (CC-BY-SA 4.0)