Extract columns from list of tables in python

Question

I have a list of tables, where each table is a list of lists. I'd like to extract the columns from each table to get a lists of columns. An example should clarify:

input=[[[1,2,3],[4,5,6]],[[7,8,9],[10,11,12]],[[13,14,15],[16,17,18]]]
output=[[[1,4],[7,10],[13,16]],[[2,5],[8,11],[14,17]],[[3,6],[9,12],[15,18]]]

I know I can probably use several for loops, but it feels like there should be a nice one liner to do this.

I intend to assign the results to variables:

a,b,_=output

With a single table, I can extract the columns with:

>>>input=[[1,2,3],[4,5,6]]
>>>list(zip(*input))
[(1, 4), (2, 5), (3, 6)]

or for assignment:

>>>a,b,_=zip(*input)
>>>a
(1, 4)
>>>b
(2, 5)

But for a list of tables, I haven't been able to do it. Here are some of the things I've tried unsuccessfully:

>>>list(zip(*zip(*input)))
[([1, 2, 3], [4, 5, 6]), ([7, 8, 9], [10, 11, 12]), ([13, 14, 15], [16, 17, 18])]
>>>[list(zip(*inp)) for inp in input]
[[(1, 4), (2, 5), (3, 6)], [(7, 10), (8, 11), (9, 12)], [(13, 16), (14, 17), (15, 18)]]
>>>[[x,y,z] for x,y,z in [zip(*inp) for inp in input]]
#same as above
>>> [[x,y] for inp in input for x,y in zip(*inp)]
[[1, 4], [2, 5], [3, 6], [7, 10], [8, 11], [9, 12], [13, 16], [14, 17], [15, 18]]

And nested/unpacked assignment didn't work for me either:

>>>[[a,b,_]]=[zip(*inp) for inp in input]
ValueError: too many values to unpack (expected 1)
>>>[*a,*b,*_]=[[x,y] for x,y in [zip(*inp) for inp in input]]
SyntaxError: two starred expressions in assignment

Is there a one-liner to do what I'm trying to do?

Edit: Note that while the example is specifically 3 tables with 2 rows and 3 columns each, my actual use case has unknown numbers of tables and rows.

I ended up using this line in my code:

list(zip(*[zip(*inp) for inp in input]))

Prune · Accepted Answer · 2020-07-31 06:06:01Z

1

You got close with that last attempt. You need one more, well-chosen nesting level. EDIT: I added the final zip step to get the desired ordering. I also used that "star" notation to help show how to extend the concept.

given = [[[1,2,3],[4,5,6]],
         [[7,8,9],[10,11,12]],
         [[13,14,15],[16,17,18]]]

f1 = [[[a, b] for a, b in zip(list1, list2)] for list1, list2 in given]
print(f1)
f2 = list(zip(*f1))
print(f2)

Output (edited for readability)

[[[1, 4], [2, 5], [3, 6]], 
 [[7, 10], [8, 11], [9, 12]], 
 [[13, 16], [14, 17], [15, 18]]]

[([1, 4], [7, 10], [13, 16]),
 ([2, 5], [8, 11], [14, 17]),
 ([3, 6], [9, 12], [15, 18])]

The second one has tuples instead of lists at the middle level; is that okay? If not, can you fix it? (left as an exercise for the student ... )

edited Jul 31, 2020 at 6:06

answered Jul 30, 2020 at 0:35

Prune

78k14 gold badges63 silver badges83 bronze badges

Sign up to request clarification or add additional context in comments.

6 Comments

Kalev Maricq Over a year ago

Is there a way to make this work for columns of any length?

Prune Over a year ago

It already does work for the third dimension. If you want it to work for the first two, simply trade in the pair of variables, replacing with "star" expansion, such as *var. Is that enough to make it clear?

Kalev Maricq Over a year ago

The *var makes sense. Looking at this more though, it appears it does something slightly different than what I'm looking for. Applying this to my original input I get: [[[1, 4], [2, 5], [3, 6]], [[7, 10], [8, 11], [9, 12]], [[13, 16], [14, 17], [15, 18]]]

Prune Over a year ago

(1) As best I can see, the output you posted is identical to the desired output. (2) If you need to ask another question, please post it separately. Comments are not a good place to solve coding difficulties.

Kalev Maricq Over a year ago

I'm not sure what you're looking at, but my desired output is the one with [7,10] as the second pair.

|

Collectives™ on Stack Overflow

Extract columns from list of tables in python

1 Answer 1

6 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

6 Comments

Your Answer

Sign up or log in

Post as a guest

Related