Counter and defaultdict reference

Both are dict subclasses — every dict method works. These pages list only what they add.

from collections import Counter, defaultdict

Counter

Construction

Form	Result
Counter(iterable)	tally each element — Counter('aab') → {'a': 2, 'b': 1}
Counter(mapping)	from existing counts — Counter({'a': 2})
Counter(**kwargs)	from keywords — Counter(a=2, b=1)
Counter()	empty

Methods and behaviour

Member	Does
c[key]	the count, or 0 if absent (no KeyError, no insertion)
c.most_common([n])	list of (item, count), highest first; all if n omitted
c.total()	sum of all counts (Python 3.10+)
c.elements()	iterator repeating each item by its count (ignores ≤ 0)
c.update(iterable_or_mapping)	add counts (not replace)
c.subtract(iterable_or_mapping)	subtract counts; may go negative
+c / -c	unary: keep positive / keep negative-flipped counts
len(c)	number of distinct items

Arithmetic operators (return a new Counter, dropping counts ≤ 0)

Operator	Meaning	Example on a=Counter(x=3,y=1), b=Counter(x=1,y=2)
a + b	add counts	{'x': 4, 'y': 3}
a - b	subtract, keep positives	{'x': 2}
a & b	intersection (min)	{'x': 1, 'y': 1}
a \| b	union (max)	{'x': 3, 'y': 2}

To keep zero/negative results, use the subtract method instead of -.

defaultdict

defaultdict(default_factory) — a dict that calls default_factory() (a zero-argument callable) to create a value for any missing key, on first access.

Common factories

Factory	Missing key becomes	Use for
list	[]	grouping items into buckets
int	0	counting / summing
set	set()	grouping unique items
dict	{}	nested mappings
lambda: defaultdict(int)	a nested defaultdict	multi-level structures
custom lambda: <value>	that value	any constant default

Behaviour

Aspect	Note
missing-key read	runs the factory and inserts the key (unlike Counter)
d.default_factory	the factory; None means "behave like a plain dict" (KeyError)
in / .get(key)	check without triggering the factory
dict(d)	a plain dict copy — no more auto-creation

defaultdict(None) has no factory, so missing keys raise KeyError like an ordinary dict.

Counter or defaultdict?

Want	Use
frequency counts + ranking	Counter
count without inserting on read	Counter (reads return 0 cleanly)
group items into lists/sets	defaultdict(list) / defaultdict(set)
accumulate sums or build nested structures	defaultdict(int) / nested defaultdict