Skip to content
Tutorials Center
My learning

Topics

Python Reference ยท Built-in Functions

Python filter() function

Reference for Python filter(): keep the items of an iterable for which a function returns true, the lazy iterator result, None as the function and comprehension alternatives.

filter() takes a function and an iterable and returns an iterator over the items for which the function returns a true value. Items that fail the test are skipped. Nothing is computed until you iterate over the result.

If you pass None instead of a function, filter() keeps the items that are themselves truthy. In everyday code a list comprehension with an if clause often reads better, but filter() shines when you already have a named predicate function or want to stay lazy.

Python
evens = filter(lambda n: n % 2 == 0, range(10))
print(list(evens))
# [0, 2, 4, 6, 8]

Try it

Python Predicates and None

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Python Lazy and one-shot

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Syntax

Python
filter(function, iterable)
Name Description
function Called with one item at a time; the item is kept if the result is true. Use None to keep truthy items.
iterable The source of items: a list, string, generator, file and so on.

Return value: a filter iterator yielding the kept items in their original order.

Raises: TypeError if iterable is not iterable or the function cannot be called with one argument. Exceptions raised by the function appear while you iterate.

Usage notes

  • It is an iterator, not a list. Python 2 returned a list; Python 3 does not. Use list(filter(...)) when you need one, and remember that an iterator can be consumed only once.
  • Truthiness decides. The function does not have to return True; any truthy value keeps the item, including non-empty strings and non-zero numbers.
  • filter(None, xs) drops 0, 0.0, "", None, False and empty containers. Be careful when zero is a valid value.
  • One iterable only. Unlike map(), filter() takes a single iterable.
  • Comprehensions vs filter(). [x for x in xs if cond(x)] is usually clearer for inline conditions. Prefer filter() when passing an existing function or composing lazy pipelines.
  • The opposite is itertools.filterfalse(), which keeps items where the function is false.
  • Original data is untouched. filter() never mutates its input.
  • Side effects run late. Because it is lazy, a predicate that prints or logs runs only when the result is consumed.