How to use the reduce function on a list of tuples?

Question

I have this this list of tuple:

  a = [(1, 2), (1, 4), (1, 6)]

I would like to use the reduce function in order to get this result:

  (3, 12)

I tried:

  x = reduce(lambda x, y: x+y, a)

But i get an error...I want to add up all the elements in the first index of each tuple, then add up the second element.

Answer 1

If you want the output of a reduce to be a tuple, all the intermediate results should also be a tuple.

a = [(1, 2), (1, 4), (1, 6)]
print reduce(lambda x, y: (x[0] + y[0], x[1] + y[1]), a)

Output

(3, 12)

Edit: If you want to get (0, 0) when the list is empty

a = []
print reduce(lambda x, y: (x[0] + y[0], x[1] + y[1]), [(0, 0)] + a)

Output

(0, 0)

Edit 2: Reduce accepts default initializer as the last parameter, which is optional. By using that, the code becomes

a = []
print reduce(lambda x, y: (x[0] + y[0], x[1] + y[1]), a, (0, 0))

Answer 2

>>> a = [(1, 2), (1, 4), (1, 6)]
>>> map(sum, zip(*a))
[3, 12]

UPDATE

According to Raymond Hettinger,

zip-star trick abuses the stack to expensively compute a transpose.

Here's an alternative that does not use list comprehension.

>>> a = [(1, 2), (1, 4), (1, 6)]
>>> [sum(item[i] for item in a) for i in range(2)] # 2 == len(a[0])
[3, 12]
>>> a = []
>>> [sum(item[i] for item in a) for i in range(2)] # 2 == len(a[0])
[0, 0]

Answer 3

You were close. Just modify your code to unpack the input tuples first. Once you add the new values, just repack the result tuple:

>>> a = [(1, 2), (1, 4), (1, 6)]
>>> reduce(lambda (sx, sy), (x, y): (sx+x, sy+y), a)
(3, 12)

Answer 4

Try this:

>>> from numpy import asarray
>>> a = asarray([(1, 2), (1, 4), (1, 6)])
>>> reduce(lambda x,y:x+y, a)
array([ 3, 12])