Get started learning Python with DataCamp's free Intro to Python tutorial. Learn Data Science by completing interactive coding challenges and watching videos by expert instructors. Start Now!

This site is generously supported by DataCamp. DataCamp offers online interactive Python Tutorials for Data Science. Join 11 million other learners and get started learning Python for data science today!

Good news! You can save 25% off your Datacamp annual subscription with the code LEARNPYTHON23ALE25 - Click here to redeem your discount

Lambda functions


Normally we define a function using the def keyword somewhere in the code and call it whenever we need to use it.

def sum(a,b):
    return a + b

a = 1
b = 2
c = sum(a,b)
print(c)

Now instead of defining the function somewhere and calling it, we can use python's lambda functions, which are inline functions defined at the same place we use it. So we don't need to declare a function somewhere and revisit the code just for a single time use.

They don't need to have a name, so they also called anonymous functions. We define a lambda function using the keyword lambda.

your_function_name = lambda inputs : output

So the above sum example using lambda function would be,

a = 1
b = 2
sum = lambda x,y : x + y
c = sum(a,b)
print(c)

Here we are assigning the lambda function to the variable sum, and upon giving the arguments i.e. a and b, it works like a normal function.

Exercise

Write a program using lambda functions to check if a number in the given list is odd. Print "True" if the number is odd or "False" if not for each element.

l = [2,4,7,3,14,19] for i in l: # your code here l = [2,4,7,3,14,19] for i in l: # your code here my_lambda = lambda x : (x % 2) == 1 print(my_lambda(i)) test_output_contains("False") test_output_contains("False") test_output_contains("True") test_output_contains("True") test_output_contains("False") test_output_contains("True") success_msg("Nice work!")

This site is generously supported by DataCamp. DataCamp offers online interactive Python Tutorials for Data Science. Join over a million other learners and get started learning Python for data science today!

Previous Tutorial Next Tutorial Take the Test
Copyright © codequickly.org . Read our Terms of Use and Privacy Policy