Python Sets

Satellite · Data Types

Python Sets

Sets track unique elements and support fast membership checks. Use them for deduplication, tagging, and math-like operations.

Use set operationsConsider frozenset

Creation

items = {1, 2, 3}
empty = set()               # {} makes an empty dict, not a set
deduped = set(["a", "b", "a"])   # {'a', 'b'} — duplicates dropped

Two defining traits:

• Elements are unique — adding a duplicate is a no-op.
• Sets are unordered — they have no index, so items[0] raises TypeError.

Elements must be hashable (immutable), so you can store numbers, strings, and tuples, but not lists or dicts.

Adding and removing

s = {1, 2}
s.add(3)          # {1, 2, 3}
s.discard(9)      # no error if missing
s.remove(2)       # KeyError if missing
s.pop()           # remove and return an arbitrary element

Operations

a = {1, 2, 3}
b = {3, 4, 5}
a | b  # union            -> {1, 2, 3, 4, 5}
a & b  # intersection     -> {3}
a - b  # difference       -> {1, 2}
a ^ b  # symmetric diff   -> {1, 2, 4, 5}

Each operator has a method form (a.union(b), a.intersection(b)) that accepts any iterable, not just another set.

Membership checks are O(1) on average—far faster than scanning a list:

if email in unsubscribed:   # instant, even for millions of entries
    skip(email)

Subset and superset tests

{1, 2} <= {1, 2, 3}    # True — subset
{1, 2, 3} >= {1, 2}    # True — superset
{1, 2}.isdisjoint({3}) # True — no elements in common

Set comprehensions

squares = {n * n for n in range(6)}          # {0, 1, 4, 9, 16, 25}
first_letters = {name[0] for name in names}  # unique initials

Frozenset

frozenset is an immutable set. Because it is hashable, you can use it inside another set or as a dictionary key—something a regular set cannot do.

tags = frozenset({"python", "beginner"})
index = {tags: "article-1"}   # frozenset as a dict key

Next up in your learning path

Combine with dictionaries

Manage membership and counts together.

Companion

Use comprehensions

Build sets from loops quickly.

Reference

Compare tuples vs sets

Understand when immutability helps.

Companion

Frequently Asked Questions

How do I create an empty set in Python?

Use set(). The literal {} creates an empty dictionary, not a set, because dictionaries also use curly braces. Only non-empty sets like {1, 2} can use brace syntax.

empty = set()      # empty set
type({})           # <class 'dict'> — not a set!

Do Python sets keep their order?

No. Sets are unordered and have no index, so you cannot rely on element order or do set[0]. If you need order plus uniqueness, use dict.fromkeys(items) to deduplicate while preserving insertion order.

list(dict.fromkeys(["b", "a", "b", "c"]))   # ['b', 'a', 'c']

Why can I not add a list to a set?

Set elements must be hashable, and lists are mutable and therefore unhashable. Convert the list to a tuple first (or use a frozenset) if you need to store it in a set.

{[1, 2]}          # TypeError: unhashable type: 'list'
{(1, 2)}          # OK — tuples are hashable

When should I use a set instead of a list?

Use a set when you need unique elements or fast membership tests (in is O(1) for sets vs O(n) for lists). Use a list when order matters or duplicates are meaningful.