Return the maximum along an axis.
Parameters:  a : array_like
axis : int, optional
out : ndarray, optional


Returns:  amax : ndarray

See also
Notes
NaN values are propagated, that is if at least one item is nan, the corresponding max value will be nan as well. To ignore NaN values (matlab behavior), please use nanmax.
Examples
>>> a = np.arange(4).reshape((2,2))
>>> a
array([[0, 1],
[2, 3]])
>>> np.amax(a, axis=0)
array([2, 3])
>>> np.amax(a, axis=1)
array([1, 3])