Reproducing code example:

import numpy as np
a = np.array(['1999-12-15', 'NaT'], dtype='M8[ns]')

np.isnan(a)
np.nanmin(a)
np.nanmax(a)

Error message:

Each of those raises an:
TypeError: ufunc 'isnan' not supported for the input types, and the inputs could not be safely coerced to any supported types according to the casting rule ''safe''

Background

Numpy 1.18 returns NaT for np.min(a) (it did skip NaT before 1.18; #14717). This breaks xarray (pydata/xarray#3409) and pandas (pandas-dev/pandas#29058) - therefore it would be good to make np.isnan work with datetime.

Also: #9009, #1307, #5222

Numpy/Python version information:

1.18.0.dev0+3b5db53 3.7.3 | packaged by conda-forge | (default, Jul 1 2019, 21:52:21)
[GCC 7.3.0]