Python comprehensions

Seems like comprehensions is the quick way of using functional programming and readable code

Comprehensions

import pandas as pd

df=pd.read_csv("https://raw.githubusercontent.com/TheEconomist/big-mac-data/master/output-data/big-mac-raw-index.csv")

Note

old_list = [2,4,2,4,5,3,6]

new_list = [i*2 for i in old_list]

new_list

old_list = [2,4,2,4,5,3,6]
new_list =  [i*2 for i in old_list]
new_list

[4, 8, 4, 8, 10, 6, 12]

Add an if statement

Note

Adding the if-statement in the end of the comprehension

[i*2 for i in old_list if i%2==True]

The if-statement acts as a filter.

old_list = [2,4,2,4,5,3,6]
new_list =  [i*2 for i in old_list if i%2==True]
new_list

[10, 6]

If-statements that changes the calculation logic

old_list = [2,4,2,4,5,3,6]
new_list =  [i*2 if i > 3 else i * 3 for i in old_list if i%2==True]
new_list

[10, 9]

1st if statements adds logic, 2nd if statement filters the results.

Enumerate

Shorter and better than the for loop. Enumerate takes a string, finds the index and the value of it and adds it to a tuple. we can filter it to grab even indexed values in the string. Code below.

[char for idx,char in enumerate('abcde') if idx % 2 == 0]

['a', 'c', 'e']

Let’s say we want to uppercase the ‘aeuio’ chars. This is a 3-step process, first we add the for-loop for idx,char in enumerate(old_list), then we add the filter if idx % 2 == 0, finally we add char.upper() if char in ‘aeuio’ else char. Full code below.

old_list='abcde'
[char.upper() if char in 'aeuio' else char
for idx,char in enumerate(old_list)
if idx % 2 == 0]

['A', 'c', 'E']

Nested for loops

You may have multiple nested for loops, you can manage that with comprehensions too using code like this.

[(i,j) for i in range(5) for j in range(i)]

[(1, 0),
 (2, 0),
 (2, 1),
 (3, 0),
 (3, 1),
 (3, 2),
 (4, 0),
 (4, 1),
 (4, 2),
 (4, 3)]

Add a filter in between loop 1 and loop 2 in order to filter in loop 1. If you want to filter loop 2, you add the filter after loop 2.

Use comprehensions on sets, tuples or dicts

Example with dict.

{i: c for i, c in enumerate('abceabce')}

{0: 'a', 1: 'b', 2: 'c', 3: 'e', 4: 'a', 5: 'b', 6: 'c', 7: 'e'}

Warning

Careful with removing duplicate values. Switching index with key will drop duplicates.

Code of a switched index and key.

{c: i for i, c in enumerate('abceabce')}

{'a': 4, 'b': 5, 'c': 6, 'e': 7}

it will take the one in the end of the list.

Unpacking a comprehension

Here we will get a index for each pair of key and value, which is smooth way of unpacking the data that is inside.

arr = [('a', 1), ('b', 2), ('c', 2)]
[{key: value, 'i': idx} for idx, (key, value) in enumerate(arr)]

[{'a': 1, 'i': 0}, {'b': 2, 'i': 1}, {'c': 2, 'i': 2}]

Zipping

d = {'a': 1, 'b': 2, 'c':3}
d.keys()
d.values()

dict_values([1, 2, 3])

[(k,v) for k,v in d.items()]

[('a', 1), ('b', 2), ('c', 3)]

Let’s zip it!

[(a,b) for a, b in zip([1,2],[3,4])]

[(1, 3), (2, 4)]

You may add more arrays if needed.