根据元素顺序,找出集合中所有不相交的子集



假设我想在Python 2.7中实现一个解决方案。

我有一个字符串列表,例如a=[AA','BB','CC','DD']。

期望的输出将是a的一组不相交子集,例如a_1、a_2。。。A_N,使得

(A_1 U A_2 U ... U A_N) = A,
(A_1 ∩ A_2 ∩ ... ∩ A_N) = Ø, 

同时考虑A中元素的顺序(A_1,A_2,…,A_N不能包含A中的非相邻元素)。

对于A,这些将是:

A_1,A_2。。。A_N:

  • ['AA','BB','CC','DD']
  • ['AA'],['b','CC','DD']
  • ['AA','BB'],['CC','DD']
  • ['AA','BB','CC'],['DD']
  • ['AA'],['b'],[c'],[d'd']
  • ['AA','BB'],[CC'],[DD']
  • ['AA'],['b','CC'],[DD']
  • ['AA'],['b'],'CC','DD']

(希望我没有遗漏,但我想你明白了)

重点是效率——这意味着相对快速,不会浪费太多内存。我知道,对于一个更大的列表,组合的数量可能会激增,但我的列表永远不会超过5个元素。

我在这里找到了一个类似问题的答案,唯一的区别是我想要所有的子集,而他们只需要最大长度为2的子集。

该解决方案相当于找到整数的所有可能组合,这些整数的总和为n(输入列表的长度),然后将该解决方案重新映射到单词列表以找到其子集。

他们问题的伪代码:

push an empty list onto the stack;
while (the stack is not empty) {
  pop the top list off the stack;
  if (the sum of its entries is n)
    add it to the solution set;
  else if (the sum of its entries is less than n)
    add a 1 to a copy of the list and push it onto the stack;
    add a 2 to a copy of the list and push it onto the stack;
  }
}

此问题的伪代码(扩展):

push an empty list onto the stack;
while (the stack is not empty) {
  pop the top list off the stack;
  if (the sum of its entries is n)
    add it to the solution set;
  else if (the sum of its entries is less than n)
    for j = 1:n {
      add j to a copy of the list and push it onto the stack;
      }
  }
}

我的Python实现:

import copy
def generate_subsets(words):
    # get length of word list
    list_len = len(words)
    # initialize stack, subset_lens list
    stack = [[], ]
    subset_lens = []
    while stack:
        current_item = stack.pop(-1)
        if sum(current_item) == list_len:
            subset_lens.append(current_item)
        elif sum(current_item) < list_len:
            for j in range(1, list_len+1):
                new_item = copy.deepcopy(current_item)
                new_item.append(j)
                stack.append(new_item)
    # remap subset lengths to actual word subsets
    subsets = []
    for subset_len in subset_lens:
        subset = []
        starting_index = 0
        for index in subset_len:
            subset.append('_'.join(words[starting_index:starting_index+index]))
            starting_index+= index
        subsets.append(subset)
    return subsets

输入:

generate_subsets(['AA', 'BB', 'CC', 'DD'])

输出:

['AA_BB_CC_DD']
['AA_BB_CC', 'DD']
['AA_BB', 'CC_DD']
['AA_BB', 'CC', 'DD']
['AA', 'BB_CC_DD']
['AA', 'BB_CC', 'DD']
['AA', 'BB', 'CC_DD']
['AA', 'BB', 'CC', 'DD']

如果有人找到更有效的解决方案,我很乐意在回复/评论中看到它!

最新更新