Python Lambda, map() and filter()

A lambda is a tiny function without a name, and map() and filter() use functions to process whole lists at once.

Lambda Syntax

A lambda is written in one line:

lambda parameters: expression

It has no def, no name and no return. The value of the expression is returned automatically.

add_tax = lambda price: round(price * 1.18)
area = lambda w, h: w * h

print(add_tax(500))
print(area(12, 10))

590
120

These two lines work the same as normal functions. But giving a lambda a name like this is not the usual way to use it. Lambdas shine when you pass them into another function.

Lambda as a Sort Key

sorted(), min() and max() take a key function. It tells Python what to compare.

products = [
    {"name": "Mouse", "price": 1200},
    {"name": "Keyboard", "price": 3500},
    {"name": "USB cable", "price": 350},
]

by_price = sorted(products, key=lambda p: p["price"])
print([p["name"] for p in by_price])

print(max(products, key=lambda p: p["price"])["name"])

names = ["sara", "Ali", "bilal", "Zara"]
print(sorted(names))
print(sorted(names, key=lambda n: n.lower()))

['USB cable', 'Mouse', 'Keyboard']
Keyboard
['Ali', 'Zara', 'bilal', 'sara']
['Ali', 'bilal', 'sara', 'Zara']

Without a key, capital letters sort before small ones. With key=lambda n: n.lower() the sort ignores case.

map(): Change Every Item

map(function, collection) calls the function on every item. It returns a map object, so wrap it in list() to see the values.

prices_usd = [10, 25, 4]
rate = 280
prices_pkr = list(map(lambda d: d * rate, prices_usd))
print(prices_pkr)

typed = ["12", "7", "30"]
numbers = list(map(int, typed))
print(numbers, sum(numbers))

[2800, 7000, 1120]
[12, 7, 30] 49

You can pass any function to map(), not only a lambda. map(int, typed) turns every string into a number.

filter(): Keep Some Items

filter(function, collection) keeps only the items for which the function returns True.

ages = [15, 22, 17, 30, 12, 19]
adults = list(filter(lambda a: a >= 18, ages))
print(adults)

phones = ["03001234567", "12345", "03219876543", "0300"]
valid = list(filter(lambda p: len(p) == 11 and p.startswith("03"), phones))
print(valid)

[22, 30, 19]
['03001234567', '03219876543']

reduce(): Combine Into One Value

reduce() lives in the functools module. It takes the first two items, combines them, then combines the result with the next item, and so on.

from functools import reduce

numbers = [2, 3, 4, 5]
product = reduce(lambda a, b: a * b, numbers)
print(product)

120

The steps are: 2 x 3 = 6, then 6 x 4 = 24, then 24 x 5 = 120.

map/filter or a Comprehension?

Both styles give the same result. Many Python developers find comprehensions easier to read.

Limits of Lambda

• A lambda can hold only one expression. No loops, no if blocks, no multiple lines.
• It can use a short conditional expression: lambda m: "Pass" if m >= 50 else "Fail".
• If the logic grows, write a normal function with def and a clear name.

Exercise