numpy testing assert array NOT equal

Question

We have numpy.testing.assert_array_equal to assert that two arrays are equal.

But what is the best way to do numpy.testing.assert_array_not_equal, that is, to make sure that two arrays are NOT equal?

How unequal are the arrays? Must they differ in every element, or at least in one? — Mad Physicist, Jul 21 '16 at 14:39
At least one element should be unequal. That is, the content of the two arrays may not be the same. If one element differ, everything is OK. — Hallgeir Wilhelmsen, Jul 21 '16 at 17:13
Then @Eswcvlad's answer is probably the most elegant hack you will find. — Mad Physicist, Jul 21 '16 at 17:17

Eswcvlad · Accepted Answer · 2016-07-21T14:48:17.773

38

If you want to use specifically NumPy testing, then you can use numpy.testing.assert_array_equal together with numpy.testing.assert_raises for the opposite result. For example:

assert_raises(AssertionError, assert_array_equal, array_1, array_2)

Also there is numpy.testing.utils.assert_array_compare (it is used by numpy.testing.assert_array_equal), but I don't see it documented anywhere, so use with caution. This one will check that every element is different, so I guess this is not your use case:

import operator

assert_array_compare(operator.__ne__, array_1, array_2)

edited Jul 21 '16 at 14:48

answered Jul 21 '16 at 14:33

Eswcvlad

796
7
9

1

Neat hack for first method. Second one will only succeed if every element, not just some, is different as far as I can tell. Not sure if that is what OP wants. +1 either way. – Mad Physicist Jul 21 '16 at 14:41
Yes, you are right about the second one. Will edit the answer. – Eswcvlad Jul 21 '16 at 14:47
The first one does it for me! Thanks! – Hallgeir Wilhelmsen Jul 21 '16 at 17:24

Mark · Answer 2 · 2016-07-21T14:47:46.527

11

I don't think there is anything built directly into the NumPy testing framework but you could just use:

np.any(np.not_equal(a1,a2))

and assert true with the built in unittest framework or check with NumPy as assert_equal to True e.g.

np.testing.assert_equal(np.any(np.not_equal(a,a)), True)

edited Jul 21 '16 at 14:47

answered Jul 21 '16 at 14:25

Mark

768
3
7
23

You should probably just use `TestCase.assertTrue(np.any(…))` instead (or `nose.tools.assert_true(np.any(…))` if you are outside of a `TestCase`). – Robert Kern Jul 21 '16 at 14:58

score 7 · Answer 3 · answered Feb 28 '18 at 22:31

7

Not sure why this hasn't been posted, may be I didn't understand the question properly, but what about:

assert not np.array_equal(array1 , array2)

Any reason why you would like to keep it exclusively in the testing module of numpy?

answered Feb 28 '18 at 22:31

Jose Sergio Hleap

71
1
3

1

`assert_array_equal` handles NaN as values, not as uncomparable (always unequal to anything) like usual for floating point – azrdev May 07 '19 at 09:00

score 2 · Answer 4 · answered Jun 21 '20 at 20:38

2

Cleaner syntax to @Eswcvlad's answer:

import numpy as np

with np.testing.assert_raises(AssertionError):
    np.testing.assert_array_equal(expected, actual)

answered Jun 21 '20 at 20:38

Mikhail

8,692
8
56
82

score 0 · Answer 5 · answered Oct 01 '20 at 08:12

Perhaps you usually want to test if something is almost equal (considering decimal precision) and consequently want to test if something is NOT almost equal in some cases. Building on @Mikhail answer (and also using pytest.raises) this would give:

import numpy as np
import pytest
 
with pytest.raises(AssertionError):
   np.testing.assert_almost_equal(...)

score 0 · Answer 6 · answered Mar 01 '23 at 07:54

0

Improving the accepted answer, we can make assert_array_not_equal as follows

from numpy.testing import assert_array_equal, assert_raises


def assert_array_not_equal(x, y):
    return assert_raises(AssertionError, assert_array_equal, x, y)

answered Mar 01 '23 at 07:54

Muhammad Yasirroni

1,512
12
22

numpy testing assert array NOT equal

6 Answers6