python unique values in a list of tuples where one of the elements is an list

Question

Say I have an list list of tuples like this:

a = [(["I", "love", "apples"], "good"),
(["I", "love", "apples"], "good")),
(["I", "hate", "apples"], "bad")),
(["I", "dislike", "apples"], "bad")),]

I want to remove duplicated, but list(set(a))errors out:

TypeError: unhashable type: 'list'

Is there any pythonic way to remove duplicates from this type of list

While the above link is slightly different, the solution will work in this case too (with minor tweaks). — Cory Kramer, Oct 07 '14 at 17:42

score 2 · Accepted Answer · answered Oct 07 '14 at 17:44

Assuming you are creating the data, use tuples as the first item instead of lists

a = [
    (("I", "love", "apples"), "good"),
    (("I", "love", "apples"), "good"),
    (("I", "hate", "apples"), "bad"),
    (("I", "dislike", "apples"), "bad")
]

print len(set(a))

Output:

score 1 · Answer 2 · answered Oct 07 '14 at 17:43

1

lists are not hashable so you can not simple call set on a.

you can convert the inner lists to tuples, then call list(set()):

list(set([(tuple(sent), good_bad) for sent, good_bad in a]))

answered Oct 07 '14 at 17:43

Jeff Tsui

1,266
12
20

python unique values in a list of tuples where one of the elements is an list

2 Answers2