Lambda function for data wrangling

Coming from a math background, to me, Lambda's represented the parameter in exponential distributions, eigenvalues in linear algebra, or in the context of AWS Lambda, a compute service for managing servers!

Whereas in programming, by definition, Lambda is an anonymous function. This means that unlike normal functions definitions, Lambdas do not have a name. Often, it's used as an argument for a higher-order function, and is lightweight in syntax, allowing your code to be less verbose, and easier to understand.

For me, I choose to write a Lambda function over a named function, if it's a simple function that will only be used a few times, passed as an argument.

Lambda functions in Python

Here's a simple function:

def hello(name):
    return f' Hello, my name is {name}'

And here is its lambda equivalent:

lambda name: f' Hello, my name is {name}'

The primary differences between named and lambda functions are:

1. lambda functions don't have a name

2. lambda functions are written on one line

3. lambda functions don't require a return. It's implied.
   

Syntax

lambda arguments: expression

Lambda functions can be used wherever function objects are required. A lambda function can take any number of arguments, but can only have one expression. They can also be used anywhere ordinary functions can!

Example 1

lambda is powerful when used in conjunction with a pandas data frame and apply. Given that we're in the Sweet Sixteen of March madness, let's use NCAA data.

import pandas as pd
from IPython.display import HTML

df = pd.DataFrame({
    'college': ['Duke', 'North Carolina', 'Virginia', 'Oregon'],
    'seed': [1, 1, 1, 12],
    'final_four_%': [55, 38, 53, 3]
    })

HTML(df.to_html(classes="table table-responsive table-striped table-bordered"))

Let's add a new column, upset, identifying whether or not a college is the favourite or underdog, using a lambda function.

With lambda , we're able to quickly spin up a function that looks up if a seed is a favourite or an underdog.

df["upset"] = df.seed.apply(lambda x, lookup = {1: "Favourite", 12: "Underdog"} : lookup[x])

# Sort by probability of reaching final four
HTML(df.sort_values("final_four_%",ascending=False).to_html(classes="table table-repsponsive table-striped table-bordered"))

Example 2

Here's a sorting example.

prospects = [
    {'name':'Zion Williamson', 'weight': 280, 'rank': 1}, 
    {'name': 'Ja Morant', 'weight': 175, 'rank':2},
    {'name': 'RJ Barrett', 'weight': 210, 'rank':3},
    {'name':'Bol Bol', 'weight': 235, 'rank':18}]

Let's figure out who the lightest prospect is, using sorted a Python method.

sorted(prospects, key= lambda x: x['weight'])

lightest_prospect = sorted(prospects, key= lambda x: x['weight'])[0]["name"]

print(lightest_prospect)

Ja Morant

With lambda in our toolbelt, we'll have a function to write elegant and beautiful code for data wrangling.