【问题标题】:Partitions of a set - Storing results in a series of nested lists集合的分区 - 将结果存储在一系列嵌套列表中
【发布时间】:2016-05-01 11:37:38
【问题描述】:

我有列出一组所有分区的代码。代码来自本站:Generating the Partitions of a Set

我不想将分区打印出来,而是将它们存储为列表。我想根据这个递归示例中返回的内容来建模我的结果:How to find all partitions of a set

我想要一个整数列表的列表。内层列表是包含在中间列表中的分区的子集,外层列表是包含所有分区的完整集合。

这是我的代码(从 C 转换为 Java,原始网站帖子中的 cmets 仍然存在):

import java.util.ArrayList;
import java.util.List;


public class PartitionApp {


  public static class PNR {
    static
    /*
    	printp
    		- print out the partitioning scheme s of n elements 
    		as: {1, 2, 4} {3}
    */
    ArrayList < ArrayList < ArrayList < Integer >>> outerList = new ArrayList < > ();
    public static void PNR(int[] s, int n) {
      /* Get the total number of partitions. In the example above, 2.*/

      int part_num = 1;
      int i;

      for (i = 0; i < n; ++i)
        if (s[i] > part_num) {

          part_num = s[i];
        }
        /* Print the p partitions. */

      int p;

      for (p = part_num; p >= 1; --p) {

        System.out.print("{");
        ArrayList < Integer > innerList = new ArrayList < > ();
        ArrayList < ArrayList < Integer >> middleList = new ArrayList < > ();
        /* If s[i] == p, then i + 1 is part of the pth partition. */
        for (i = 0; i < n; ++i) {
          if (s[i] == p) {
            innerList.add(i + 1);
            System.out.print(i + 1);
            System.out.print(",");
          }

        }
        middleList.add(innerList);
        outerList.add(middleList);

        System.out.print("} ");
      }

      System.out.print("\n");
      System.out.println(outerList);

    }

    /*
	next
		- given the partitioning scheme represented by s and m, generate
		the next

	Returns: 1, if a valid partitioning was found
		0, otherwise
*/
    static int next(int[] s, int[] m, int n) {
      /* Update s: 1 1 1 1 -> 2 1 1 1 -> 1 2 1 1 -> 2 2 1 1 -> 3 2 1 1 ->
	1 1 2 1 ... */
      /*int j;
	printf(" -> (");
	for (j = 0; j < n; ++j)
		printf("%d, ", s[j]);
	printf("\b\b)\n");*/
      int i = 0;
      ++s[i];
      while ((i < n - 1) && (s[i] > m[i + 1] + 1)) {
        s[i] = 1;
        ++i;
        ++s[i];
      }

      /* If i is has reached n-1 th element, then the last unique partitiong
	has been found*/
      if (i == n - 1)
        return 0;

      /* Because all the first i elements are now 1, s[i] (i + 1 th element)
	is the largest. So we update max by copying it to all the first i
	positions in m.*/
      if (s[i] > m[i])
        m[i] = s[i];
      for (int j = i - 1; j >= 0; --j) {
        m[j] = m[i];

      }


      /*	for (i = 0; i < n; ++i)
      		printf("%d ", m[i]);
      	getchar();*/
      return 1;
    }

    public static void main(String[] args) {
      int count = 0;
      int[] s = new int[16];
      /* s[i] is the number of the set in which the ith element
      			should go */
      int[] m = new int[16]; /* m[i] is the largest of the first i elements in s*/

      int n = 4;
      int i;
      /* The first way to partition a set is to put all the elements in the same
	   subset. */
      for (i = 0; i < n; ++i) {
        s[i] = 1;
        m[i] = 1;
      }

      /* Print the first partitioning. */
      PNR(s, n);

      /* Print the other partitioning schemes. */
      while (next(s, m, n) != 0) {
        PNR(s, n);
        count++;
      }
      count = count + 1;
      System.out.println("count = " + count);


      //	return 0;
    }

  }

}

我得到的 n=4 的结果如下所示(方括号替换为大括号以进行格式化):

{{{1, 2, 3, 4}}, {{1}}, {{2, 3, 4}}, {{2}}, {{1, 3, 4}}, {{ 1, 2}}, {{3, 4}}.....

没有“中间”分组。所有内部子集(应该是一组 n 元素的一部分)都作为列表包含在外部集合中。我没有正确设置内部、中间和外部列表,并且已经为此苦苦挣扎了一天。我希望有人能帮我看看我的错误。

谢谢, 丽贝卡

【问题讨论】:

    标签: java set combinatorics partition


    【解决方案1】:

    花了我一段时间,但我找到了解决方案!我所做的是从数组中取出所有可能的部分,然后对剩下的部分进行递归,然后将取出的部分添加为递归中返回的分区的一部分。然后进入一个包含所有可能分区的大数组。为了强制进行某种排序,我这样做是为了让我们取出的这部分总是采用第一个元素。这样你就不会得到像 [[1], [2, 3]] 和 [[2, 3], [1]] 这样的结果,它们基本上只是同一个分区。

    public static int[][][] getAllPartitions(int[] array) throws Exception {
        int[][][] res = new int[0][][];
        int n = 1;
        for (int i = 0; i < array.length; i++) {
            n *= 2;
        }
        for (int i = 1; i < n; i += 2) {
            boolean[] contains = new boolean[array.length];
            int length = 0;
            int k = i;
            for (int j = 0; j < array.length; j++) {
                contains[j] = k % 2 == 1;
                length += k % 2;
                k /= 2;
            }
            int[] firstPart = new int[length];
            int[] secondPart = new int[array.length - length];
            int p = 0;
            int q = 0;
            for (int j = 0; j < array.length; j++) {
                if (contains[j]) {
                    firstPart[p++] = array[j];
                } else {
                    secondPart[q++] = array[j];
                }
            }
            int[][][] partitions;
            if (length == array.length) {
                partitions = new int[][][] {{firstPart}};
            } else {
                partitions = getAllPartitions(secondPart);
                for (int j = 0; j < partitions.length; j++) {
                    int[][] partition = new int[partitions[j].length + 1][];
                    partition[0] = firstPart;
                    System.arraycopy(partitions[j], 0, partition, 1, partitions[j].length);
                    partitions[j] = partition;
                }
            }
            int[][][] newRes = new int[res.length + partitions.length][][];
            System.arraycopy(res, 0, newRes, 0, res.length);
            System.arraycopy(partitions, 0, newRes, res.length, partitions.length);
            res = newRes;
        }
        return res;
    }
    

    【讨论】:

    • 谢谢!顺序无关紧要;我只是希望能够识别由一定数量的子集组成的分区,然后测试这些子集以查看它们是否满足特定条件。感谢您的帮助 - 非常感谢。
    • 是的,我主要使用排序来防止重复。这只是一种从所有重复项中选择哪一个的方法。例如,没有排序的算法可以在其解决方案中同时输出 [[1,2],[3]] 和 [[3],[1,2]] ,这将是重复的。我的算法只会输出两者中的第一个,因为它是有序的。
    猜你喜欢
    • 1970-01-01
    • 1970-01-01
    • 1970-01-01
    • 1970-01-01
    • 1970-01-01
    • 1970-01-01
    • 1970-01-01
    • 1970-01-01
    • 1970-01-01
    相关资源
    最近更新 更多