numpy.ma.assert_array_equal#

ma.assert_array_equal(actual, desired, err_msg='', verbose=True, *, strict=False)[source]#

Raises an AssertionError if two masked arrays are not equal.

Given two array_like objects, check that their masks are equal and that their data is equal wherever it is not masked. Unlike numpy.testing.assert_array_equal, which ignores masked values, a value that is masked in only one of the objects is reported as a mismatch. Objects that are not masked arrays are treated as having no masked values. As in numpy.testing.assert_array_equal, NaNs in the same unmasked positions are considered equal, and a scalar is compared with every element of an array unless strict is set.

Added in version 2.6.0.

Parameters:

actualarray_like: The actual object to check.
desiredarray_like: The desired, expected object.
err_msgstr, optional: The error message to be printed in case of failure.
verbosebool, optional: If True, the conflicting values are appended to the error message.
strictbool, optional: If True, raise an AssertionError when either the shape or the data type of the array_like objects does not match, and disable the special handling of scalars.

Raises:

AssertionError: If the masks of actual and desired, or their unmasked values, are not equal.

See also

numpy.testing.assert_array_equal: Ignores values masked in either array.
numpy.ma.allequal

Examples

>>> import numpy as np

Masked values are not compared, so this does not raise:

>>> a = np.ma.array([1.0, 2.0, 3.0], mask=[False, True, False])
>>> b = np.ma.array([1.0, -9.0, 3.0], mask=[False, True, False])
>>> np.ma.assert_array_equal(a, b)

A value that is masked in only one of the objects is a mismatch:

>>> np.ma.assert_array_equal(a, [1.0, 2.0, 3.0])
Traceback (most recent call last):
    ...
AssertionError:
Arrays are not equal

Mismatched mask elements: 1 / 3 (33.3%)
Mask mismatch at index:
 [1]: -- (ACTUAL), 2.0 (DESIRED)
 ACTUAL: masked_array(data=[1.0, --, 3.0],
             mask=[False,  True, False],
       fill_value=1e+20)
 DESIRED: masked_array(data=[1., 2., 3.],
             mask=False,
       fill_value=1e+20)