Python Memory Profiling - A Quick Demo

Python Memory Profiling

Quick Demo of Python Memory Profiling

README.md

Quick Demo of Python Memory Profiling

This is a demonstration of python memory profiling using the package memory_profiler. (psutil is a recommended optional package for optimizing the profiling).

The demonstration essentially builds a list of tuples in one function and a list of dicts in another. memory_profiler then gives the memory profile for both functions. It also doubles as a good demonstrator of why tuples are more memmory efficient than dicts.

How To Run:

1. make a virtualenvironment with virtualenv
2. install dependencies: pip install -r requirements.txt
3. run python main.py. This will make lists of 10000 items. To increase the number just run: python main.py <new number> e.g. python main.py 100000

main.py

import sys

from memory_profiler import profile


@profile
def difference_tuple_and_dicts():
    size = 10000
    if len(sys.argv) > 1:
        size = int(sys.argv[1])

    check_tuples(size)
    check_dicts(size)
    print("Done")


@profile
def check_dicts(size):
    a = []  # list of dicts
    for i in range(0, size):
        a.append({"name": "Ian", "profession": "engineer"})


@profile
def check_tuples(size):
    a = []  # list of tuples
    for i in range(0, size):
        a.append(tuple([i for i in range(0, 20)]))


if __name__ == "__main__":
    difference_tuple_and_dicts()

requirements.txt

memory_profiler
psutil