Lambdas (and Closures - Part II)

In Python, anonymous functions are declared with the lambda keyword.

How to use lambdas:

Refer to the following:

add_two_numbers = lambda x, y: x + y

print(add_two_numbers(10, 3))  # => 13
  • I am setting the variable add_two_numbers to be the lambda function
  • The variables after the lambda declaration are the function’s parameters: x, y
  • The resolution of the actions that follow the colon is its return value: x + y

Written as a traditional function, this would be:

def add_numbers(x, y):
  return x + y
  
add_two_numbers = add_numbers  # To follow the process above
print(add_two_numbers(10, 3))  # => 13

Because the return value of a lambda isn’t explicit, it’s useful to see how they are commonly used.

Lambdas as a boolean expressions

Boolean expressions are one of the most common cases of lambdas.

is_even = lambda x: x % 2 == 0
print(is_even(4))  # => True
print(is_even(3))  # => False
is_longer_than_5 = lambda s: len(s) > 5
name1 = "Reed"
name2 = "Reedworth"

print(is_longer_than_5(name1))  # => False
print(is_longer_than_5(name2))  # => True

Python Closures - Part II

I will refer to what I covered in my introduction to closures.

The inner functions we make in closures are better expressed with anonymous functions.

While the inner function is doing most of the work in the closure, we never find ourselves needing to refer to it by name. It ends up taking up a lot space, and it makes things more confusing.

Let’s clean up the closures we made in the introduction by replacing the inner functions with lambdas.

Navigate to the closure_practice directory (or anywhere you can play with some code).

subl exercise4.py  # Or the editor of your choice

Remember this code?

def times(x):
    def multiply(y):
        return x * y
        
    return multiply

To convert the inner function to a lambda function:

def times(x):
  return lambda y: x * y

Try this code out, and see if you can do the same with the exponent closure we made in Part I.

When you’re ready, comment that code out, and try to replace the inner function of the last example from Part I with a lambda:

def supply_x(x):
    def unnamed_function(func):  # Replace this function with a lambda
        return func(x)

    return unnamed_function

Look for lambdas

Lambda boolean expressions (themselves within closures, at times) are also common as:

  • Conditions in list comprehensions
  • Conditions in generators
  • The condition in these functional methods:

I’ll write about these in upcoming days, and you’ll see better how we put anonymous functions to use.


Further Reading:




Written on April 30, 2016