Python Partial Functions

Created with Sketch.

Python Partial Functions

Summary: In this tutorial, you’ll learn about Python partial functions and how to define partial functions using the partial function from the functools module.

Introduction to Python partial functions

The following example defines a function that multiplies two arguments:

def multiply(a, b):
return a*b

Code language: Python (python)

Sometimes, you just want to multiply an argument with a specified number e.g., 2. To do that, you can reuse the multiply function like this:

def double(a):
return multiply(a,2)

Code language: Python (python)

The double function returns the multiply function. It passed the number 2 to the second argument of the multiply function.

The following shows how to use the double function:

result = double(10)
print(result)

Code language: Python (python)

Put it all together:

def multiply(a, b):
return a*b

def double(a):
return multiply(a, 2)

result = double(10)
print(result) # 20

Code language: Python (python)

As you can see, the double function reduces the arguments of the multiply function.

The double function freezes the second argument of the multiply function, which results in a new function with a simpler signature.

In other words, double function reduces the complexity of the multiply function.

In Python, the double function is called a partial function.

In practice, you use partial functions when you want to reduce the number of arguments of a function to simplify the function’s signature.

Since you’ll create partial functions sometimes, Python provides you with the partial function from the functools standard module to help you define partial functions more easily.

Python partial function from functools module

The following shows the syntax of the partial function from the functools module:

functools.partial(fn, /, *args, **kwargs)

Code language: Python (python)

The partial function returns new partial object, which is a callable.

When you call the partial object, Python calls the fn function with the positional arguments args and keyword arguments kwargs.

The following example shows how to use the partial function to define the double function from the multiply function:

from functools import partial

def multiply(a, b):
return a*b

double = partial(multiply, b=2)

result = double(10)
print(result)

Code language: Python (python)

Output:

20

 

How it works.

  • First, import the partial function from the functools module.
  • Second, define the multiply function.
  • Third, return a partial object from the partial function and assign it to the double variable.

When you call the double, Python calls the multiply function where b argument defaults to 2.

If you pass more arguments to a partial object, Python appends them to the args argument.

When you pass additional keyword arguments to a partial object, Python extends and overrides the kwargs arguments.

Therefore, it’s possible to call the double like this:

double(10, b=3)

Code language: Python (python)

In this example, Python will call the multiply function where the value of the b argument is 3, not 2.

And you’ll see the following output:

30

 

Python partial functions and variables

Sometimes, you may want to use variables for creating partials. For example:

from functools import partial

def multiply(a, b):
return a*b

x = 2
f = partial(multiply, x)

result = f(10) # 20
print(result)

x = 3
result = f(10) # 20
print(result)

Code language: Python (python)

In this example, we change x to 3 and expect that f(10) would return 30 instead of 20.

However, f(10) returns 20 instead. It’s because Python evaluates the value of x in the following statement:

f = partial(multiply, x)

Code language: Python (python)

…but not after that, therefore, when x references to the new number (3), the partial function doesn’t change.

Summary

  • Use partial function from the functools module to create partial functions in Python.

Leave a Reply

Your email address will not be published. Required fields are marked *