Count sublists with the same values in a list

user_3960122 0
Viewed 73

How to count sublists with the same values (order doesn't matter) in a list?

I tried this:

from collections import Counter

Input = [
    [
        'Test123', 'heyhey123', 'another_unique_value',
    ],
    [
        'Test123', 'heyhey123', 'another_unique_value',
    ],
    [
        'heyhey123',
    ],
    [
        'Test123', 'heyhey123',
    ],
    [
        'another_unique_value', 'heyhey123', 'Test123'
    ]
]

Counter(str(e) for e in li)

Output:

Counter({
    "['Test123', 'heyhey123', 'another_unique_value']": 2},
    "['heyhey123']": 1},
    "['Test123', 'heyhey123']": 1},
    "['another_unique_value', 'heyhey123', 'Test123']": 1},
)

Obviously it takes the order from the values in the list in account. How do I count the sublists where the order doesn't matter?

The output I want is:

Counter({
    "['Test123', 'heyhey123', 'another_unique_value']": 3},
    "['heyhey123']": 1},
    "['Test123', 'heyhey123']": 1},
)
pandas counter python
2 Answers

You can replace

Counter(str(e) for e in li)

with

Counter(tuple(sorted(e)) for e in li)

Giving output:

Counter({('Test123', 'another_unique_value', 'heyhey123'): 3,
         ('heyhey123',): 1,
         ('Test123', 'heyhey123'): 1})

Another option would be to use set(e) to ignore the order of elements in the list, but this has the downside of ignoring repetitions - ['Test123', 'heyhey123', 'another_unique_value'] would be counted as the same as ['Test123', 'heyhey123', 'another_unique_value', 'another_unique_value'] - and in addition, when converting from the unhashable set to include in a Counter, an identical order is not guaranteed.

user_123661100

I think you're pretty close.

Counter(str(set(e)) for e in Input)
returns

Counter({"{'heyhey123', 'Test123', 'another_unique_value'}": 3,
     "{'heyhey123'}": 1,
     "{'heyhey123', 'Test123'}": 1})

which is almost identical to what you're looking for I believe :)

user_81886660
Related