如何在 Java 中正确实现 Median-of-median 算法

如何解决如何在 Java 中正确实现 Median-of-median 算法

我正在尝试实现中值算法/BFPRT算法的中值。

main 中数组的检查输出为 5，但是我的代码甚至没有达到 一个适当的输出（输出-1，因为它没有达到i>k 或i

任何帮助将不胜感激。

我越是在意它，我的代码就越是乱七八糟。 请原谅乱码：

import java.util.Arrays;

public class getMedian {
    
    // Method for quickly returning a median
    public static int getMedian(int[] list) {
        Arrays.sort(list);

        if (list.length == 0) {
            System.out.println("Critical error,somehow passed empty list");
            return 5000;
        } else if (list.length <= 2) {
            return list[list.length / 2];
        } else {
            return list[(list.length - 1) / 2];
        }
    }

    // Method for getting the position of median from given list
    public static int findMedianPos(int[] arr) {
        int expectedMedianPos = 0;
        if (arr.length % 2 == 0) {
            expectedMedianPos = (arr.length / 2) - 1;
        } else {
            expectedMedianPos = (arr.length - 1) / 2;
        }
        return expectedMedianPos;

    }


    // Method for getting median of medians.
    static int medianFinder(int[] arr,int[] orArr,int[] helperArr,int lowerBound,int upperBound,int medPos) {
        int y = 0;
        if (arr.length % 5 != 0) {
            y = arr.length % 5;
        } else {
            y = 1;
        }
        int[] helper = new int[5];
        int[] helper2 = new int[y];
        int[] medians = new int[(int) Math.ceil(arr.length / 5.0)];
        int medianPointer = 0;
        if (arr.length <= 5) {
            return partition(orArr,helperArr,getMedian(arr),medPos);

        }
        for (int i = lowerBound; i < upperBound; i++) {
            helper[i % 5] = arr[i];
            helper2[i % helper2.length] = arr[i];
            if (i != lowerBound && (i + 1) % 5 == 0) {
                medianPointer = medianPointer + 1;

            }
            if (i == upperBound - 1 && (i + 1) % 5 != 0) {
                medians[medianPointer] = getMedian(helper2);
            }

        }

        return medianFinder(medians,orArr,medians.length,medPos);
    }

    public static int partition(int[] partitionedLists,int partitionElem,int medPos) {
        int b = 0;
        for (int i = 0; i < orArr.length; i++) {
            if (orArr[i] == partitionElem) {
                b = i; 
            }

        }
        int sizeL1 = 0;
        int sizeL2 = 0;
        for (int i = 0; i < partitionedLists.length; i++) {
            if (partitionedLists[i] < partitionedLists[b]) {
                sizeL1 = sizeL1 + 1;
            }
            if (partitionedLists[i] > partitionedLists[b]) {
                sizeL2 = sizeL2 + 1;
            }

        }

        int[] L1 = new int[sizeL1];
        int[] L2 = new int[sizeL2];
        int counterL1 = 0;
        int counterL2 = 0;
        for (int i = 0; i < partitionedLists.length; i++) {
            if (partitionedLists[i] < partitionedLists[b]) {
                L1[counterL1] = partitionedLists[i];
                counterL1 = counterL1 + 1;
            } else if (partitionedLists[i] > partitionedLists[b]) {
                L2[counterL2] = partitionedLists[i];
                counterL2 = counterL2 + 1;
            }

        }
        
        int k  = sizeL1 + 1;
        int i = medPos;
        if(i<k) {
            if(sizeL1 <= 5) {
                Arrays.sort(L1);
                return L1[medPos - 1];
            } else {
                medianFinder(L1,L1.length,i);
                
            }
        }
        if(i>k) {
            if(sizeL2 <= 5) {
                Arrays.sort(L2);
                return L2[medPos-1];
            } else {
                medianFinder(L2,L2.length,i);
            }
        }


        

        return -1;

    }

    public static void main(String[] args) {
        int[] numbers = {7,9,10,8,5,6,1,3,2,4};
        System.out.println(medianFinder(numbers,numbers,numbers.length,findMedianPos(numbers)+1));

    }
}