在 numpy 数组的 python 列表中快速找到关节对

Question

I have a list of numpy arrays and want to get the indices of the joint pairs. Basically, if I have the following list of lists:

lst = [[1, 2, 3, 5], [1, 2, 9], [5, 6], [9]]

I want the following:

[[0, 1], [0, 2], [1, 3]]

Here's what I have so far, but it's slow:

out = out
for i, item in enumerate(lst):
    for j in range(i+1, len(lst)):
        if not set(item).isdisjoint(lst[j]):
            out.append([i, j])

I appreciate your help.

Answer 1

You could try something like this:

from itertools import combinations

lst = [[1, 2, 3, 5], [1, 2, 9], [5, 6], [9]]
lst_set = [set(e) for e in lst]

print([[x, y] for x, y in combinations(range(len(lst)), 2) if not lst_set[x].isdisjoint(lst_set[y])])

Output

[[0, 1], [0, 2], [1, 3]]