3. Generator functions

Avoid materializing large collections when you only need to iterate over them. Use yield to generate values on demand. Generator functions are often more memory-efficient.

Lazy generation also lets callers start consuming results before the full sequence exists, which can simplify streaming-style code. The key improvement is not just memory use but a cleaner separation between producing items and collecting them.

3.1. Don’t do this

def generate_sequential_numbers(n):
    nums = []
    for i in range(n):
        nums.append(i)
    return nums

sum(generate_sequential_numbers(10000000))

3.2. Do this

def generate_sequential_numbers(n):
    for i in range(n):
        yield i

sum(generate_sequential_numbers(10000000))

generate_sequential_numbers = lambda n: (i for i in range(n))

sum(generate_sequential_numbers(10000000))