Numpy: Maximal Value is NaN

Question

My question about Python is really trivial: What do I have to modify that the function max() returns a real value for any compiler?

import numpy as np
a = np.array([-1, 0, 1, np.nan])
# The maximal value is 1. It is not nan!
a.max()

There is the possibility to use the following code but it looks ugly to me:

a[np.logical_not(np.isnan(a))].max()

Answer 1

As was answered by @Divakar using np.nanmax(a) will solve your question.