Sets in Python

Introduction

So far, we've looked at the Object Oriented Programming (OOP) side of Python--turns out, we've been using objects for a long time without even realizing it.

A list, for instance, is an object, with its class-specific methods like append( ) or insert( ).

Today, we'll take a look at a similar object (/ data type) that operates in parallel to how lists do--sets.

What Are Sets?

Well, in math, you may know of sets to be a collection of distinct (non-repeating) values. This general notion carries over to Python, as well.

Sets:

• store unordered values

  • unlike lists, we can't index into them

• store completely distinct values

  • no dupes

• can store different data types

  • yes, unlike lists, sets can hold multiple types

• are mutable

  • unlike strings, we can add or remove elements from sets

  • however, since they're unordered, we can't change existing values

  • so this wouldn't work simply because indexes don't exist...

myset = set([0, 1.0, 'a', 2])
myset[0] = 1

Initializing Sets

So there are two ways to create or initialize a set.

The first is using the set( ) object itself:

def main():
	# notice how we pass values inside of square brackets to the object
	myset = set([0, 1.0, 'a', 2])
	print(myset)


main()

The second, and more shorthand way:

def main():
	myset = {0, 1.0, 'a', 2}
	print(myset)


main()

Both produce the exact same output, so it just comes down to preference:

{0, 1.0, 'a', 2}

Car Brands

Let's take an example in which sets might be the most practical choice of "storage," tracking all unique brands of cars.

# list of dictionaries storing car models and their brands
cars = [
	{"model": "Civic", "brand": "Honda"},
	{"model": "Camry", "brand": "Toyota"},
	{"model": "Model S", "brand": "Tesla"},
	{"model": "Accord", "brand": "Honda"}
]

# creating a list of distinct car brands
brands = []
for car in cars:
	if car["brand"] not in brands:
		brands.append(car["brand"])

So, using a list object, this technically works... but we can make some improvements using our set object.

# list of dictionaries storing car models and their brands
cars = [
	{"model": "Civic", "brand": "Honda"},
	{"model": "Camry", "brand": "Toyota"},
	{"model": "Model S", "brand": "Tesla"},
	{"model": "Accord", "brand": "Honda"}
]

# creating a set of car brands
brands = set()
for car in cars:
	brands.add(car["brand"])

Much cleaner. Using a set, we've automatically filtered out any duplicates by the very nature of the type itself.

And, like append( ) that we use with lists, sets use an instance method called add( ) (pretty self-explanatory.)

Final Thoughts

This was everything you need to know about sets--thanks for reading!