Flatten a list of lists in one line in Python
Sometimes you need to flatten a list of lists. The old way would be to do this using a couple of loops one inside the other. While this works, it's clutter you can do without. This tip show how you can take a list of lists and flatten it in one line using list comprehension.
The loop way
#The list of lists
list_of_lists = [range(4), range(7)]
flattened_list = []
#flatten the lis
for x in list_of_lists:
for y in x:
flattened_list.append(y)
List comprehension way
#The list of lists
list_of_lists = [range(4), range(7)]
#flatten the lists
flattened_list = [y for x in list_of_lists for y in x]
This is not rocket science, but it's cleaner, and, I'm lead to believe, faster than the for loop method.
Of coarse itertools comes to the rescue,as stated below in the comments by <a href="https://coderwall.com/iromli ">iromli</a>
list(itertools.chain(*listoflists))
Which is faster than any of the above methods, and flattening lists of lists is exactly what it was designed to do.
Written by James Hurford
Related protips
13 Responses
Yeah that works too. Thanks. In fact the method you specify seems to be the fastest, followed by list comprehension, then distantly behind, the loop in a loop method.
Good way:
import itertools
flattened_list = list(itertools.chain(*list_of_lists))
;)
Say this on another site ....
flattenedlist = sum(listof_lists, [])
That, to me, looks anything but clean. Looking at it, I would honestly have no idea what it did until running it. It's like you're sacrificing readability for conciseness a la the perl syndrome.
I for one do not find that hard to follow at all. If you have trouble understanding it, I'm sorry, but I don't agree with the assertion is anything but clean. If you don't think it's clean, how about sharing your clean method, rather than complaining about mine
Useful info! Thanks
Thanks. very useful
Thanks for sharing
Мery well)
Thank you for this useful tip!
Thank you sharing this answer its very helpful
well informed and right information.
Very valuable post. Thanks!