I want to add each element of list containing tuples. For example,
>>> list1
[(0, 1, 2), (3, 4, 5), (6, 7, 8)]
>>> list2
[(1, 2, 3), (4, 5, 6), (7, 8, 9)]
Answer should be
list3 = [(1,3,5),(7,9,11),(13,15,17)]
-
Please vote up and select an answer from one of the three below.Matt Alcock– Matt Alcock2012-02-22 22:47:53 +00:00Commented Feb 22, 2012 at 22:47
Add a comment
|
6 Answers
zip is your friend here.
result = []
for ta, tb in zip(list1, list2):
t =tuple(a+b for a, b in zip(ta, tb))
result.append(t)
print result
>> [(1,3,5),(7,9,11),(13,15,17)]
or more pythonic is:
result = [tuple(a+b for a, b in zip(ta, tb)) for ta, tb in zip(list1, list2)]
print result
>> [(1,3,5),(7,9,11),(13,15,17)]
result could simply be a generator by doing this:
result = (tuple(a+b for a, b in zip(ta, tb)) for ta, tb in zip(list1, list2))
2 Comments
Adam Parkin
I'm not entirely sure I'd consider the list comprehension there to be more Pythonic. It's definitely less readable. Using timeit I timed both snippets and the unrolled for loop was marginally faster as well (about 5%). Generally yeah, list comps should be preferred, but nested list comps tend to border on unreadable voodoo.
Matt Alcock
I agree Adam it's not readable the reason I mention it being more pythonic is that result could very easily then be a generator by changing the brackets. This then means the result is not allocated to mem or iterated through until it is needed. See modification above...
>>> list1 = [(0, 1, 2), (3, 4, 5), (6, 7, 8)]
>>> list2 = [(1, 2, 3), (4, 5, 6), (7, 8, 9)]
>>> [tuple(map(sum, zip(*i))) for i in zip(list1, list2)]
[(1, 3, 5), (7, 9, 11), (13, 15, 17)]
The nice thing about doing it this way is that it also works for an arbitrary number of lists
>>> list3 = [tuple(map(sum, zip(*i))) for i in zip(list1, list2)]
>>> [tuple(map(sum, zip(*i))) for i in zip(list1, list2, list3)]
[(2, 6, 10), (14, 18, 22), (26, 30, 34)]
3 Comments
DASADIYA CHAITANYA
if my list comes as dynamically then how to calculate this kind of things
John La Rooy
@DaSaDiYaChaiTAnYa, not sure what you mean. Can you give an example?
DASADIYA CHAITANYA
okey no problem i had resolved it from my side my question is above answer use the two different list like list1 and list2 but when there are the list comes from the dynamically then how to resolve this kind of things ..??
You can use a list comprehension, though it's a bit awkward:
list3 = [tuple(x + y for x, y in zip(t1, t2))
for t1, t2 in zip(list1, list2)]
If you are fine with a list of lists instead of a lists of tuples, an alternative is
list3 = [map(operator.add, *t) for t in zip(list1, list2)]
6 Comments
georg
Interesting how another poster considers the "awkward" code "more pythonic".
yurisich
If the OP has 11 rep, I'd call it awkward too! I thought the same thing, and consulted google's python style guide to help clarify. Honestly, it didn't help much.
Adam Parkin
Fun fact: the map() version above was ~50% faster than the list comp in my experiments (using timeit), and yet map is still a function that is frowned upon in the Python community.
John La Rooy
@Adam, I wouldn't say it's frowned upon, but there are only a relatively few specific cases where it still makes sense to use map over list comprehensions and it takes a fair while to learn when it should be used.
Rik Poggi
(+1) And the OP could wrap
tuple() around imap() if he really needs a tuple. |
>>> [tuple(sum(pair) for pair in zip(a,b)) for a,b in zip(list1,list2)]
[(1, 3, 5), (7, 9, 11), (13, 15, 17)]
Comments
Edit: The below is not at all needed. The map builtin automatically zips multiple iterables passed to it, so zip_with(func, iter1, iter2) can be replaced by map with the same arguments.
map(functools.partial(map, op.add), list1, list2)
This section isn't needed
Here's a snippet implementing something along the lines of Haskell's zipWith
def zip_with(func, xs, ys):
return [func(x, y) for (x, y) in zip(xs, ys)]
In [1]: def zip_with(func, xs, ys):
...: return [func(x, y) for (x, y) in zip(xs, ys)]
...:
In [2]: import operator as op
In [3]: zip_with(op.add, [1,2,3], [4,5,6])
Out[3]: [5, 7, 9]
In [4]: import functools
In [5]: zip_with(functools.partial(zip_with, op.add), list1, list2)
Out[5]: [[1, 3, 5], [7, 9, 11], [13, 15, 17]]
Comments
import operator
list3 = [tuple(map(operator.add, list1[i], list2[i]))
for i in range(len(list1))]
