Зміст курсу
Optimization Techniques in Python
Optimization Techniques in Python
Leveraging map() and List Comprehensions
Whenever possible, it's more efficient to use Python's built-in functions and methods or library functions instead of writing algorithms from scratch. These are often highly optimized, designed to run faster, and capable of handling various edge cases.
Let’s begin with powerful tools like the map() function and list comprehensions, which can significantly speed up your code. Since list comprehensions are more optimized for performance and map() are implemented in C, they generally execute faster than manually written loops.
List Comprehensions
List comprehensions offer a more concise and often faster way to create lists compared to traditional for loops. Besides, they can make your code easier to read and understand.
Let's compare the performance of a for loop and a list comprehension by generating a list where each number from 1 to 10000000 is multiplied by 2:
import os os.system('wget https://codefinity-content-media-v2.s3.eu-west-1.amazonaws.com/courses/8d21890f-d960-4129-bc88-096e24211d53/section_1/chapter_3/decorators.py 2>/dev/null') from decorators import timeit_decorator numbers = list(range(1, 10000001)) @timeit_decorator(number=10) def list_comprehension(): return [x * 2 for x in numbers] @timeit_decorator(number=10) def for_loop(): result = [] for x in numbers: result.append(x * 2) return result list_1 = list_comprehension() list_2 = for_loop() print(list_1 == list_2)
Using map()
The map function applies a specified function to each item in an iterable (e.g., a list).
Besides being more concise than writing loops, it often runs faster because of C-level optimizations in its implementation, which reduce the overhead of repeated function calls.
While map can offer some performance benefits, especially when working with built-in functions, the speed difference compared to list comprehensions is often negligible. Both are highly optimized in Python, and for simple operations, the performance advantage of map is barely noticeable. In practice, the choice between the two should be based more on readability and specific use cases, such as memory efficiency with map.
Let's compare the performance of map, for loops, and list comprehensions when formatting 1000000 customer names to title case (e.g., 'john doe' -> 'John Doe'):
import os os.system('wget https://codefinity-content-media-v2.s3.eu-west-1.amazonaws.com/courses/8d21890f-d960-4129-bc88-096e24211d53/section_1/chapter_3/decorators.py 2>/dev/null') from decorators import timeit_decorator customer_names = ['john doe', 'jane smith', 'alex johnson', 'ivan smith'] * 250000 @timeit_decorator(number=50) def map_function(): return list(map(str.title, customer_names)) @timeit_decorator(number=50) def list_comprehension(): return [name.title() for name in customer_names] @timeit_decorator(number=50) def for_loop(): result = [] for name in customer_names: result.append(name.title()) return result formatted_with_map = map_function() formatted_with_comprehension = list_comprehension() formatted_with_for_loop = for_loop() print(formatted_with_map == formatted_with_comprehension == formatted_with_for_loop)
As expected, the for loop is the slowest approach of the three. In this example, map turned out to be faster than list comprehension, although in practice, performance can vary depending on the specific use case. Sometimes, list comprehensions may even outperform map, especially for simpler expressions.
Дякуємо за ваш відгук!
