Python - Maximum number of times element appear in list

Question

Given the following list my_list = [ ['red', 'yellow', 'green'], ['red', 'yellow', 'green'], ['red', 'green', 'red']], I want to find the element which appears at most. I have implemented a brute-force mathod together with a new list and a dictionary to hold the values for each element but the efficiency is not satisfactory. Which would be the best way to implement this search ? I am using Python 3.

What do you mean by "brute-force"? It's not like there is a magic algorithm out there that can count elements in a list without iterating over it. — DeepSpace, Jan 19 '18 at 13:46
I used 3 for loops. Copied all different elements in a new `list` and then used one `for` loop to keep track of each item in my new list and two more nested `for` loops for the given list. — , Jan 19 '18 at 13:49

score 3 · Accepted Answer · answered Jan 19 '18 at 13:49

3

my_list = [ ['red', 'yellow', 'green'], ['red', 'yellow', 'green'], ['red', 'green', 'red']]

new_list = [k for v in my_list for k in v]

from collections import Counter

c = Counter(new_list)

print c.most_common(1)
>>> [('red', 4)]

answered Jan 19 '18 at 13:49

Veera Balla Deva

790
6
19

1

For sake of memory, use a generator instead of `new_list`: `print(Counter(element for sublist in my_list for element in sublist).most_common(1))` – DeepSpace Jan 19 '18 at 14:01

Ajax1234 · Answer 2 · 2018-01-19T13:50:55.653

1

You can flatten the nested list and apply the max function:

my_list = [ ['red', 'yellow', 'green'], ['red', 'yellow', 'green'], ['red', 'green', 'red']]
new_list = [i for x in my_list for i in x]
max_val = max(new_list, key=lambda x:new_list.count(x))

Output:

'red'

edited Jan 19 '18 at 13:50

answered Jan 19 '18 at 13:44

Ajax1234

69,937
8
61
102

`new_list` is just a waste of memory. You can just keep a dictionary in memory and count as you go. – DeepSpace Jan 19 '18 at 13:45
@DeepSpace your solution would be more efficient, but how exactly would that save space? – Ajax1234 Jan 19 '18 at 13:46
waste of a `lambda` here – Chris_Rands Jan 19 '18 at 13:48
@Ajax1234 `new_list` is a waste of memory since it is an unnecessary flattened version of `my_list`. It can be turned into a generator at least. – DeepSpace Jan 19 '18 at 13:52
`from collections import Counter ; print(Counter(element for sublist in my_list for element in sublist).most_common(1)) ; # [('red', 4)]` – DeepSpace Jan 19 '18 at 13:56

RoadRunner · Answer 3 · 2018-01-19T13:57:04.593

You can flatten your list with itertools.chain.from_iterable, then use a collections.defaultdict to store the counts, and getting the max key is easy from there:

from collections import defaultdict
from operator import itemgetter
from itertools import chain

my_list = [['red', 'yellow', 'green'], ['red', 'yellow', 'green'], ['red', 'green', 'red']]

d = defaultdict(int)
for word in chain.from_iterable(my_list):
    d[word] += 1

print(max(d.items(), key = itemgetter(1))[0])
# red

Python - Maximum number of times element appear in list

3 Answers3