查找集合列表的组合

Question

I'm using python 2.7. Given a list of sets, is there any quick and efficient way to generate the combinations? Input sets always have one item in each set and output sets all have length of 2

From:

[set(['item1']), set(['item2']), set(['item3'])]

To:

[set(['item1','item2']), set(['item2','item3']), set(['item3','item1'])]

Answer 1

Since the elements in your list of sets are all 1-element sets, the combinations that you are looking for are just the 2-element subsets of the union of those sets. You can obtain them like this:

>>> import itertools

>>> sets = [set(['item1']), set(['item2']), set(['item3'])]
>>> elements = set()
>>> for s in sets: elements.update(s)

thus

>>> elements
{'item1', 'item2', 'item3'}

And then, just this:

>>> pairs = [set(combo) for combo in itertools.combinations(elements,2)]
>>> pairs
[{'item1', 'item2'}, {'item1', 'item3'}, {'item2', 'item3'}]

Answer 2

As John pointed out itertools might be of help. Here's a quick example:

import itertools as it
sets = [set(range(0, 3)), set(range(2, 5)), set(range(4, 7))]
comb = list(it.combinations(sets, r=2))
comb

output: [({0, 1, 2}, {2, 3, 4}), ({0, 1, 2}, {4, 5, 6}), ({2, 3, 4}, {4, 5, 6})]

Then create an intersection in each iteration:

comb_sets = [a.intersection(b) for a, b in comb]
comb_sets

output: [{2}, set(), {4}]