Compute and return np.ones(1) / 0:

• suppress all numpy floating-point errors while you do it
• the result is array([inf])

Output

[inf]