>>> # The standard way to import NumPy:
>>> import numpy as np
>>> # Create a 2-D array, set every second element in
>>> # some rows and find max per row:
>>> x = np.arange(15, dtype=np.int64).reshape(3, 5)
>>> x[1:, ::2] = -99
>>> x
array([[ 0, 1, 2, 3, 4],
[-99, 6, -99, 8, -99],
[-99, 11, -99, 13, -99]])
>>> x.max(axis=1)
array([ 4, 8, 13])
>>> # Generate normally distributed random numbers:
>>> rng = np.random.default_rng()
>>> samples = rng.normal(size=2500)