title: Map, Reduce, Filter

Map, Reduce & Filter

Most engineers work with lists to process list of orders/users etc. Analyzing lists can get complex and cluttered quickly if using multiple for-loops and nested loops. Hence these above methods can streamline usage of list operations.

Map

If your task is to apply a specific method to each element of a list, map will come in handy. Say, you have a list of degree values and you’d like to convert all these values into list of values in Fahrenheit units.

Example Usage

inputs = [10,32,5,40,25]

def degreesToFahren(deg):
    fahrenheit = (9.0/5)*deg +32
    return fahrenheit

# The most common way of doing this    
result=[]
for i in inputs:
    iTofahren = degreesToFahren(i)
    result.append(iTofahren)
print(result)   # [50.0, 89.6, 41.0, 104.0, 77.0]

# Using Map
result = list(map(degreesToFahren,inputs))
print(result) # [50.0, 89.6, 41.0, 104.0, 77.0]    

As you might have noticed, using map is simply a one liner operation. Generally, if you have data = [a1,a2,...,an] and a function f(), then map(f,data):returns an iterator over f(a1),f(a2)...f(an). use list() to convert the iterator object into a python list.

Filter

Filter function removes data in a list that you need/don’t need , hence the name. Say, you want to filter a list based on values that you don’t need, eg values above 2.

Example Usage

data = [1.2,2.5,5.8,0.4,4.7,9.9]
result = list(filter(lambda x:x > 2,data))  
print(result) 

Output

[2.5, 5.8, 4.7, 9.9]

This is also a simple 1 liner similar to the map() function above. Refer to tutorial on lambda functions if you find this term unfamiliar.

Reduce

From creator of Python, Guido van Rossum
"Use functools.reduce if you really need it; however, 99% of the time an explicit for loop is more readable"

What it generally does is apply a function f() to elements of data in a list and use that result for the next value in the list.
Visually,

Data = [a₁,a₂,…,a_n]
function = f(x,y)

reduce(f,data):
Step 1: val₁ = f(a₁,a₂)
Step 2: val₂ = f(val₁,a₃)
Step 3: val₃ = f(val₂ ,a₄)
.
.
.
Step n-1: val_n-1 = f(val_n-2,a_n)

For example, you want to multiply all numbers in a list.

Example Usage

from functools import reduce

input = [1,2,3,4,5,6]
multiplier = lambda x,y:x*y
answer = reduce(multiplier,input) 
print(answer)

Output

720

However, the above could be computed using a simple for loop and usage of these methods are subject to preference.

More Information:

The official documentation for the above methods can be found at http://book.pythontips.com/en/latest/map_filter.html