Skip to content

set() Function Complexity

The set() function creates sets from iterables or creates empty sets.

Complexity Analysis

Case Time Space Notes
Empty set O(1) O(1) set()
From iterable O(n) avg, O(n²) worst O(n) n = iterable length; worst case with hash collisions
From string O(n) O(n) n = string length
Duplicate removal O(n) O(n) Automatic deduplication
Shallow copy O(n) O(n) n = set size

Basic Usage

Element order in examples

Sets are unordered. The element order shown in comments below is illustrative; actual repr() output is not guaranteed and may differ between runs, Python versions, or hash-randomization settings.

Create Empty Set

# O(1)
s = set()  # set()

From List

# O(n) - where n = list length
s = set([1, 2, 3])           # {3, 2, 1}
s = set([1, 2, 2, 3, 3, 3])  # {2, 1, 3} - duplicates removed

From String

# O(n) - where n = string length
s = set("hello")      # {'e', 'h', 'l', 'o'}
s = set("aabbcc")     # {'c', 'b', 'a'} - each char once

From Other Iterables

# O(n) - where n = iterable length
s = set((1, 2, 3))          # {2, 1, 3} (from tuple)
s = set({1, 2, 3})          # {1, 3, 2} (from set - copy)
s = set(range(5))           # {0, 2, 1, 3, 4}
s = set(map(str, [1, 2]))   # {'1', '2'} (from generator)

From Dictionary

# O(n) - uses dict keys
d = {"a": 1, "b": 2, "c": 3}
s = set(d)  # {'b', 'a', 'c'} - only keys, not values

Complexity Details

Hashing and Insertion

# O(n) - hash each item and insert
# Each item:
# 1. Hash - O(1) average
# 2. Check for duplicates - O(1) average
# 3. Insert - O(1) average
# Total: O(n)

items = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
s = set(items)  # O(10)

Duplicate Removal

# O(n) - automatic deduplication
items = [1, 1, 2, 2, 3, 3]
s = set(items)  # O(6) - creates {2, 1, 3}

# Useful for removing duplicates
unique = set("mississippi")  # {'i', 'm', 's', 'p'}

From String

# O(n) - iterate through string
short = set("ab")     # O(2) - {'a', 'b'}
long = set("a" * 1000)  # O(1000) - {'a'} (one item!)

Common Patterns

Remove Duplicates

# O(n) - fastest way to remove duplicates
items = [1, 2, 2, 3, 3, 3, 4]
unique = set(items)  # O(7)
# {2, 1, 3, 4}

# Back to list if needed
unique_list = list(set(items))  # O(n)

Find Unique Items

# O(n) - check membership in set
items = [1, 2, 2, 3, 3, 3]
unique_count = len(set(items))  # O(n) - 3 unique items

Set Operations

# O(n) - after creating sets
s1 = set([1, 2, 3])
s2 = set([2, 3, 4])

intersection = s1 & s2      # {2, 3} - O(n)
union = s1 | s2            # {2, 1, 3, 4} - O(n)
difference = s1 - s2       # {1} - O(n)
symmetric_diff = s1 ^ s2   # {1, 4} - O(n)

Membership Testing

# O(1) - fast lookup
s = set([1, 2, 3, 4, 5])

if 3 in s:  # O(1)
    print("Found")

# vs list - O(n)
lst = [1, 2, 3, 4, 5]
if 3 in lst:  # O(n) - must scan
    print("Found")

Performance Patterns

vs List Creation

# O(n) - set creation
s = set([1, 2, 3, 4, 5])

# O(n) - list creation
lst = [1, 2, 3, 4, 5]

# Similar complexity, but set has better lookup: O(1) vs O(n)

Batch Operations

# O(n * m) - n items, m = avg hash time (usually O(1))
items = range(1000)
s = set(items)  # O(n)

# Then operations are fast
if 500 in s:    # O(1)
    pass

Edge Cases

Empty Set

# O(1)
s = set()       # set()
s = set([])     # set()
s = set("")     # set()

Single Item

# O(1)
s = set([1])    # {1}
s = set("a")    # {'a'}

All Duplicates

# O(n) - still must check all
items = [1, 1, 1, 1, 1]
s = set(items)  # O(5) - results in {1}

Unhashable Items

# O(n) - error for unhashable
try:
    s = set([[1, 2], [3, 4]])  # TypeError - lists not hashable
except TypeError:
    pass

# Use frozenset for nested immutable sets
s = set([frozenset([1, 2]), frozenset([3, 4])])  # OK

Set vs frozenset

# O(n) - both create from iterable
s = set([1, 2, 3])           # Mutable
fs = frozenset([1, 2, 3])    # Immutable

# frozenset can be in set, set cannot
s = {frozenset([1, 2]), frozenset([3, 4])}  # OK
# s = {{1, 2}, {3, 4}}  # TypeError

Memory Considerations

# O(n) - memory proportional to set size
small_set = set(range(10))      # 10 items
medium_set = set(range(10**4))  # 10,000 items
large_set = set(range(10**6))   # 1,000,000 items - uses significant memory

Set Literal vs set()

# Both O(n) but different syntax

# Literal - faster, no function call
s1 = {1, 2, 3, 4, 5}  # O(n)

# Function - more flexible
s2 = set([1, 2, 3, 4, 5])  # O(n)
s3 = set(range(5))  # O(n)

# Literal usually preferred for constants

Best Practices

✅ Do:

  • Use sets for membership testing (O(1) vs O(n) for lists)
  • Use sets to remove duplicates
  • Use set operations for intersection/union/difference
  • Use {} literal for constants, set() for iterables

❌ Avoid:

  • Using sets when order matters (they're unordered)
  • Putting unhashable items in sets (lists, dicts)
  • Creating sets of mutable objects (won't work)
  • Assuming set order is consistent

Version Notes

  • Python 2.x: set() available, {} creates dict
  • Python 3.x: set() and {} syntax available
  • All versions: Unordered collection with O(1) lookup