Source: 📖 Python Cookbook ch3.9 p98

Prefer numpy universal native functions over looping

The Numpy module comes with built-in 'universal functions', which are a set of standard mathematical functions commonly applied to arrays. Always prefer using these functions to affect the elements of an array over looping, as they can be hundreds of times faster.

Some of these universal functions are:

• np.sqrt(arr)
• np.cos(arr)

See also:

• Lists vs Numpy arrays