Learn Python by Building Data Science Applications

Using generators

Generators are not exactly data structures—they are functions. However, while normal functions compute their results and return them at once, generators can be stopped and resumed on the fly, resulting in an iterable-like behavior. In other words, you can loop over a generator, retrieving one value at a time. Unlike classic iterables, however, generators are lazy. They compute values once we ask for them, but not before we do. As a result of that, there are a few significant differences in their behavior as compared to iterables:

• First, generators use a fixed amount of memory. Even if you ask one to compute zillions of values, a generator will produce and store just one value every time you ask, which is great! In fact, ...