assert_array_equal does not compare NaNs as numbers when dtype is object

OS: macOS Sierra (10.12.5)
Python version: 2.7.12
numpy version: 1.13.0

I notice that numpy.testing.assert_array_equal should compare NaNs as numbers unlike elsewhere in numpy, which makes sense to me. I verify this with the following test:

>>> import numpy.testing as npt
>>> import numpy as np
>>> a = np.array([1, 2, np.nan])
>>> npt.assert_array_equal(a, a)

However, this does not work with arrays that have dtype object:

>>> b = np.array([1, 2, 'a', np.nan], dtype=object)
>>> npt.assert_array_equal(b, b)

AssertionError: 
Arrays are not equal

(mismatch 25.0%)
 x: array([1, 2, 'a', nan], dtype=object)
 y: array([1, 2, 'a', nan], dtype=object)

Is this the desired behaviour? The only workaround in my case is to cast the array as dtype str.

Parti da numpy.testing.assert_array_equal e riproduci i due esempi dell’issue, confrontando array di dtype regular e object-dtype contenenti NaN. Segui il percorso di confronto degli array di oggetti e aggiungi un test di regressione che mostri il comportamento di confronto previsto per NaN, quindi esegui i test NumPy Testing pertinenti.