Count the frequency of a recurring list — inside a list of lists

Question

I have a list of lists in python and I need to find how many times each sub-list has occurred. Here is a sample,

from collections import Counter
list1 = [[ 1., 4., 2.5], [ 1., 2.66666667, 1.33333333], 
         [ 1., 2., 2.], [ 1., 2.66666667, 1.33333333], [ 1., 4., 2.5],
         [ 1., 2.66666667, 1.33333333]]   
c = Counter(x for x in iter(list1))
print c

I above code will work, if the elements of the list were hashable (say int), but in this case they are lists and I get an error

TypeError: unhashable type: 'list'

How can I count these lists so I get something like

[ 1., 2.66666667, 1.33333333], 3
[ 1., 4., 2.5], 2
[ 1., 2., 2.], 1

Answer 1

Just convert the lists to tuple:

>>> c = Counter(tuple(x) for x in iter(list1))
>>> c
Counter({(1.0, 2.66666667, 1.33333333): 3, (1.0, 4.0, 2.5): 2, (1.0, 2.0, 2.0): 1})

Remember to do the same for lookup:

>>> c[tuple(list1[0])]
2

Answer 2

Counter returns a dictionary like object which it's keys must be hashable. And since lists are not hashable you can convert them to tuple using map function:

>>> Counter(map(tuple, list1))
Counter({(1.0, 2.66666667, 1.33333333): 3, (1.0, 4.0, 2.5): 2, (1.0, 2.0, 2.0): 1})

Note that using map will perform slightly better than a generator expression because by passing a generator expression to Counter() python will get the values from generator function by itself, since using built-in function map has more performance in terms of execution time¹.

# Use generator expression
~ $ python -m timeit --setup "list1 = [[ 1., 4., 2.5], [ 1., 2.66666667, 1.33333333],[ 1., 2., 2.], [ 1., 2.66666667, 1.33333333], [ 1., 4., 2.5],[ 1., 2.66666667, 1.33333333]] ;from collections import Counter" "Counter(tuple(x) for x in iter(list1))"
100000 loops, best of 3: 9.86 usec per loop
# Use map
~ $ python -m timeit --setup "list1 = [[ 1., 4., 2.5], [ 1., 2.66666667, 1.33333333],[ 1., 2., 2.], [ 1., 2.66666667, 1.33333333], [ 1., 4., 2.5],[ 1., 2.66666667, 1.33333333]] ;from collections import Counter" "Counter(map(tuple, list1))"
100000 loops, best of 3: 7.92 usec per loop

From PEP 0289 -- Generator Expressions:

The semantics of a generator expression are equivalent to creating an anonymous generator function and calling it. For example:

g = (x**2 for x in range(10))
print g.next()

is equivalent to:

def __gen(exp):
    for x in exp:
        yield x**2
g = __gen(iter(range(10)))
print g.next()

---

_{Note that since generator expressions are better in terms of memory use, if you are dealing with large data you'd better use generator expression instead of map function.}

Answer 3

Try this

list1 = [[ 1., 4., 2.5], [ 1., 2.66666667, 1.33333333], 
         [ 1., 2., 2.], [ 1., 2.66666667, 1.33333333], [ 1., 4., 2.5],
         [ 1., 2.66666667, 1.33333333]]

counter = {}
for el in list1:
    el = str(el)     #This sorts your hashable part or use tuple(el)
    if el in counter:
        counter[el]+=1
    else:
        counter[el]=1

print(counter)

Should output

{'[1.0, 2.0, 2.0]': 1, '[1.0, 2.66666667, 1.33333333]': 3, '[1.0, 4.0, 2.5]': 2}