[Python] Testing equality of lists which contain NaN

I have a small class for which I am writing tests. The result for my function contains nan which causes my test to fail because nan is not equal to any other nan. How do I write a proper test for this?

from math import nan
import unittest
import pandas as pd


class holder:
def __init__(self, source: list) -> None:
self.source = source
self.data = []

def calculate_rolling_average(self):
self.data = pd.Series(self.source).rolling(3).mean().to_list()


class test_value(unittest.TestCase):
def setUp(self):
self.hold = holder(source=[10, 20, 30, 40])

def test_rolling_simple_average_for_list(self):
expected_result = [nan, nan, 20.0, 30.0]
self.hold.calculate_rolling_average()
self.assertSequenceEqual(self.hold.data, expected_result)


unittest.main()



This results in:

>>> AssertionError: Sequences differ: [nan, nan, 20.0, 30.0] != [nan, nan, 20.0, 30.0]

First differing element 0:
nan
nan

[nan, nan, 20.0, 30.0]


I read How to assert if the two possibly NaN values are equal which is testing a singular value for numpy.nan so the same does not apply to this question.

Continue reading...