Tags: #public

---

generators are
faster than list comprehensions (at scale)?

in_nums = range(100)

def is_divisible_by_five(n):
    """
    :return bool:
    """
    if n % 5 == 0:
        return True
    else:
        return False

xyz = (i for i in in_nums if is_divisible_by_five(i)) # generator expression {: id="generator-expression" }
xyz = [i for i in in_nums if is_divisible_by_five(i)] # list comprehension {: id="list-comprehension" }

which is faster to iterate?

import timeit

print(timeit.timeit("""
in_nums = range(100)

def is_divisible_by_five(n):
    if n % 5 == 0:
        return True
    else:
        return False

xyz = (i for i in in_nums if is_divisible_by_five(i))
    """))
# 0.807376595000278 {: id="0.807376595000278" }

generator -> 0.807376595000278s

print(timeit.timeit("""
input_list = range(100)
def is_divisible_by_five(n):
    if n % 5 == 0:
        return True
    else:
        return False

xyz = [i for i in in_nums if is_divisible_by_five(i)]
    """))
# 17.095245823001278 {: id="17.095245823001278" }

list comprehension -> 17.095245823001278s

the generator in that consists of numbers if they are divisible by 5 in range(100)

vs. building the list into memory

the generator calculates the expression when next() is called ?

---

a generator function

def hello_world_gen():
    yield 'Hello'
    yield 'World'
    yield '!'

for e in hello_world_gen():
    print(e)

---

python