3

In Python is there a simple way of adding the individual numbers of lists to the individual numbers of other lists? In my code I need to add about 10 long lists in a similar fashion to this:

listOne = [1,5,3,2,7]
listTwo = [6,2,4,8,5]
listThree = [3,2,9,1,1]

Therefore I want the result to be:

listSum = [10,9,16,11,13]

Thanks in advance

Adam
  • 75
  • 7
  • 2
    As a general rule it's always good to post your current code so we know what you've tried, what didn't work and why it didn't work. – Joe Smart Sep 03 '14 at 09:23

3 Answers3

7

Using zip, sum and list comprehension:

>>> lists = (listOne, listTwo, listThree)
>>> [sum(values) for values in zip(*lists)]
[10, 9, 16, 11, 13]
Vincent
  • 12,919
  • 1
  • 42
  • 64
  • 1
    very elegant solution, far better than what I was about to post! – Joe Smart Sep 03 '14 at 09:23
  • I am not sure but I am tying since question posted. I did this `map(sum, zip(*(listOne, listTwo, listThree)))` that outputs `[10, 9, 16, 11, 13]` .. Now I also use your code and out put is `[10, 9, 16, 11, 13]` but *not* `[10,10,9,11,13]` what I am missing?? – Grijesh Chauhan Sep 03 '14 at 09:32
  • @Grijesh, I didn't add the numbers in my question correctly, that was careless of me – Adam Sep 03 '14 at 09:35
  • @Adam No problem my answer and Vincent's answer is same. :) – Grijesh Chauhan Sep 03 '14 at 09:37
1

Alternatively, you can also use map and zip as follows:

>>> map(lambda x: sum(x), zip(listOne, listTwo, listThree))
[10, 9, 16, 11, 13]
skyuuka
  • 605
  • 1
  • 6
  • 22
0

Using numpy for vectorized operations is another option.

>>> import numpy as np
>>> (np.array(listOne) + np.array(listTwo) + np.array(listThree)).tolist()
[10, 9, 16, 11, 13]

Or more succinctly for many lists:

>>> lists = (listOne, listTwo, listThree)
>>> np.sum([np.array(l) for l in lists], axis=0).tolist()
[10, 9, 16, 11, 13]

Note: Each list will have to have the same dimension for this method to work. Otherwise, you will need to pad the arrays with the method described here: https://stackoverflow.com/a/40571482/5060792

For completeness: 

>>> listOne = [1,5,3,2,7]
>>> listTwo = [6,2,4,8,5]
>>> listThree = [3,2,9,1,1]
>>> listFour = [2,4,6,8,10,12,14]
>>> listFive = [1,3,5]

>>> l = [listOne, listTwo, listThree, listFour, listFive]

>>> def boolean_indexing(v, fillval=np.nan):
...     lens = np.array([len(item) for item in v])
...     mask = lens[:,None] > np.arange(lens.max())
...     out = np.full(mask.shape,fillval)
...     out[mask] = np.concatenate(v)
...     return out

>>> boolean_indexing(l,0)
array([[ 1,  5,  3,  2,  7,  0,  0],
       [ 6,  2,  4,  8,  5,  0,  0],
       [ 3,  2,  9,  1,  1,  0,  0],
       [ 2,  4,  6,  8, 10, 12, 14],
       [ 1,  3,  5,  0,  0,  0,  0]])


>>> [x.tolist() for x in boolean_indexing(l,0)]
[[1, 5, 3, 2, 7, 0, 0],
 [6, 2, 4, 8, 5, 0, 0],
 [3, 2, 9, 1, 1, 0, 0],
 [2, 4, 6, 8, 10, 12, 14],
 [1, 3, 5, 0, 0, 0, 0]]

>>> np.sum(boolean_indexing(l,0), axis=0).tolist()
[13, 16, 27, 19, 23, 12, 14]
Clay
  • 2,584
  • 1
  • 28
  • 63