How to Use Lambda Functions in Python (Pandas)

This short tutorial aims to show how you can use Lambda functions in Python, and especially in Pandas.

Introduction

The following is the basic structure of Lambda functions:

lambda bound_variable: function_or_expression

Lambda functions can have any number of arguments but only one expression, typically in one-line expression. The following is an example of adding the number 5 into variable X.

lambda x: x+5

It is the same as the following function.

def add(x):
    return(x+5)

Method 1: Use Lambda function as IIFEs

IIFEs stand for Immediately Invoked Function Expressions. An IIFE is a way to execute functions immediately when they are created. IIFEs are very useful because they don’t pollute the global object. The following is an example of using Lambda functions as IIFEs.

We append () to execute that function: (function)().

(lambda x: x+5)(10)
15

Method 2: Use Lambda with apply() in Pandas

We can use apply() to call a lambda function, which will be applied to every row or column of the dataframe and returns a modified version of the original dataframe. If axis = 0 in apply(), the lambda function will be applied to each column. In contrast, If axis = 1 in apply(), the lambda function will be applied to each row.

The following example applies a lambda function to each row to create a new column by calculating how old the cars are. The following shows the original data. 

import pandas as pd
car_data = {'Brand': ['Tesla', 'Tesla','Tesla','Ford'], 
     'Location': ['CA', 'CA','NY','MA'],
    'Year':[2019,2018,2020,2019]}
car_data=pd.DataFrame(data=car_data)
print(car_data)
    Brand Location  Year
0  Tesla       CA  2019
1  Tesla       CA  2018
2  Tesla       NY  2020
3   Ford       MA  2019

Next is the sample code for using apply() and lambda.

car_data["Year_old"]=car_data.apply(lambda x: 2022-x['Year'],axis=1)
print(car_data)
   Brand Location  Year   DateTime  Year_old
   Brand Location  Year  Year_old
0  Tesla       CA  2019         3
1  Tesla       CA  2018         4
2  Tesla       NY  2020         2
3   Ford       MA  2019         3

Method 3: Use filter() and list()

You can use the combination of filter(), list(), and lambda functions to find the elements within a column in Pandas. You can see that the lambda function and the dataframe column is connected with a comma ",".

list(filter(lambda x: x>=2019,car_data['Year']))
[2019, 2020, 2019]

Method 4: Use if-else in Lambda functions

You can use if else statements in lambda functions in Pandas. 

car_data['CA_or_not']=car_data['Location'].apply(lambda x: True if x=='CA' else False)
print(car_data)
   Brand Location  Year  Year_old  CA_or_not
0  Tesla       CA  2019         3       True
1  Tesla       CA  2018         4       True
2  Tesla       NY  2020         2      False
3   Ford       MA  2019         3      False

The following is another example.

car_data['EV_or_not']=car_data['Brand'].apply(lambda x: 'EV' if x=='Tesla' else 'Not Sure')
print(car_data)
   Brand Location  Year  Year_old  CA_or_not EV_or_not
0  Tesla       CA  2019         3       True        EV
1  Tesla       CA  2018         4       True        EV
2  Tesla       NY  2020         2      False        EV
3   Ford       MA  2019         3      False  Not Sure

Other Resource