Advanced Comprehensions
Overview
Comprehensions can include conditions to filter or modify values dynamically.
Filtering with Conditions
You can filter values in a list comprehension using a condition.
Consider the for loop below:
squares = []
for x in range(10):
if x % 2 == 0:
squares.append(x**2)
print(squares)
This can be rewritten as:
squares = [x**2 for x in range(10) if x % 2 == 0]
print(squares)
Note:
- The
%(modulo) operator checks ifxis even (x % 2 == 0). - Only even numbers are squared and added to the list.
Output:
[0, 4, 16, 36, 64]
The list comprehension only includes squares of even numbers, and it doesn't include anything for odd numbers.
Condition on Output
You can also apply a condition to the output instead of filtering.
values = [x**2 if x % 2 == 0 else 0 for x in range(10)]
print(values)
Output:
[0, 0, 4, 0, 16, 0, 36, 0, 64, 0]
The list comprehension includes both:
- squares of even numbers, and
- 0 for odd numbers.
Dictionary Comprehensions
Dict comprehensions create dictionaries using a similar syntax.
- The syntax uses
{}instead of[] - The key-value pair is separated by
:.
Consider the example below:
num_dict = {}
for x in range(1, 6):
num_dict[x] = -x
print(num_dict)
This can be written as:
num_dict = {x: -x for x in range(1, 6)}
print(num_dict)
Both will return the same output:
{1: -1, 2: -2, 3: -3, 4: -4, 5: -5}
Example: LoR Characters
In this example, we are working with a list of characters from The Lord of the Rings (LoR). We'll filter and modify the list using list comprehensions.
fellowship = ['frodo', 'samwise', 'merry', 'aragorn', 'legolas', 'boromir', 'gimli']
new_fellowship = [member for member in fellowship if len(member) >= 7]
print(new_fellowship)
Output:
'samwise', 'aragorn', 'legolas', 'boromir']
If you want to replace the characters whose names are shorter than 7 characters with an empty string (""), you can modify the list comprehension:
fellowship = ['frodo', 'samwise', 'merry', 'aragorn', 'legolas', 'boromir', 'gimli']
new_fellowship = [member if len(member) >= 7 else "" for member in fellowship]
print(new_fellowship)
Output:
['', 'samwise', '', 'aragorn', 'legolas', 'boromir', '']
Finally, if you want to create a dictionary where each character's name is a key, and the length of their name is the value, we can use a dictionary comprehension:
fellowship = ['frodo', 'samwise', 'merry', 'aragorn', 'legolas', 'boromir', 'gimli']
new_fellowship = {member: len(member) for member in fellowship}
print(new_fellowship)
Output:
{'frodo': 5, 'samwise': 7, 'merry': 5, 'aragorn': 7, 'legolas': 7, 'boromir': 7, 'gimli': 5}
Example: Time-stamped Data
In this example, we'll use a list comprehension to extract the time from Twitter data with timestamps.
- Extract 'created_at' column and assign it to
tweet_time. - Extract characters 12 to 19 from each timestamp in
tweet_time. - Assign the result to
tweet_clock_time.
Download the Twitter dataset here: tweets.csv
Solution:
import pandas as pd
df = pd.read_csv("tweets.csv")
tweet_time = df['created_at']
tweet_clock_time = [entry[11:19] for entry in tweet_time]
print(tweet_clock_time)
Output:
['23:40:17', '23:40:17', '23:40:17', '23:40:17', '23:40:17', '23:40:17',
'23:40:18', '23:40:17', '23:40:18', '23:40:18', '23:40:18', '23:40:17',
'23:40:18', '23:40:18', '23:40:17', '23:40:18', '23:40:18', '23:40:17',
'23:40:18', '23:40:17', '23:40:18', '23:40:18', '23:40:18', '23:40:18',
'23:40:17', '23:40:18', '23:40:18', '23:40:17', '23:40:18', '23:40:18',
............]
This code extracts only the time part of each timestamp from the 'created_at' column.
Example: Time-stamped Data with Conditions
In this example, we'll extract the time from time-stamped Twitter data, but only for timestamps where the seconds (entry[17:19]) are '19'.
- Extract the
created_atcolumn fromdfand assign it totweet_time. - Extract the time from each timestamp in
tweet_time(characters 12 to 19). - Add a condition to select only the timestamps where
entry[17:19] == '19'.
Download the Twitter dataset here: tweets.csv
Solution:
import pandas as pd
df = pd.read_csv("tweets.csv")
tweet_time = df['created_at']
tweet_clock_time = [entry[11:19] for entry in tweet_time if entry[17:19] == '19']
print(tweet_clock_time)
Output;
['23:40:19', '23:40:19', '23:40:19', '23:40:19', '23:40:19', '23:40:19', '23:40:19', '23:40:19',
'23:40:19', '23:40:19', '23:40:19', '23:40:19', '23:40:19', '23:40:19', '23:40:19', '23:40:19',
'23:40:19', '23:40:19', '23:40:19', '23:40:19', '23:40:19', '23:40:19', '23:40:19', '23:40:19', '
23:40:19', '23:40:19', '23:40:19', '23:40:19', '23:40:19', '23:40:19', '23:40:19', '23:40:19',
......................]