***[Lintcode]Maximum Subarray Difference 最大子数组差

最新推荐文章于 2021-10-22 18:05:43 发布

青铁

最新推荐文章于 2021-10-22 18:05:43 发布

阅读量374

点赞数

分类专栏：算法文章标签： lintcode

本文链接：https://blog.csdn.net/jc69186918/article/details/53364865

版权

算法专栏收录该内容

191 篇文章 0 订阅

订阅专栏

Given an array with integers.

Find two non-overlapping subarrays A and B, which |SUM(A) - SUM(B)| is the largest.

Return the largest difference.

Example

For [1, 2, -3, 1], return 6.

分析：求四次最大/最小子字符串。分别为从左到右计算最大/最小连续子字符串，从右至左计算最大/最小连续子字符串。原理类似于动态规划，如

leftMin[i] = Math.min(leftMin[i - 1], Math.min(nums[i] + min, nums[i])); leftMin代表0到i处（包括）连续最小子字符串和。其等于index=i-1时leftMin的值，或者包含或者不包含nums[i]的sum。

最后遍历一次计算leftMin和rightMax 或者leftMax和rightMin的绝对值的最大值。注意错位求和。

public class Solution {
    /**
     * @param nums: A list of integers
     * @return: An integer indicate the value of maximum difference between two
     *          Subarrays
     */
    public int maxDiffSubArrays(int[] nums) {
        int[] leftMin = new int[nums.length];
        int[] leftMax = new int[nums.length];
        //i向前连续n项元素的和
        int min = 0, max = 0;
        for(int i = 0; i < nums.length; i++) {
            min = Math.min(nums[i] + min, nums[i]);
            max = Math.max(nums[i] + max, nums[i]);
            if(i == 0) {
                leftMin[i] = min;
                leftMax[i] = max;
            } else {
                leftMin[i] = Math.min(leftMin[i - 1], min);
                leftMax[i] = Math.max(leftMax[i - 1], max);
            }
        }
        int res = 0, endMin = 0, endMax = 0;
        min = 0;max = 0;
        for(int i = nums.length - 1; i > 0; i--) {
            min = Math.min(nums[i] + min, nums[i]);
            max = Math.max(nums[i] + max, nums[i]);
            if(i == nums.length - 1) {
                endMin = min;
                endMax = max;
            } else {
                endMin = Math.min(endMin, min);
                endMax = Math.max(endMax, max);
            }
            res = Math.max(Math.max(Math.abs(leftMax[i-1]-endMin),   
                Math.abs(endMax - leftMin[i-1])), res);  
        }
        return res;
    }

}