给你一个整数数组 nums
和两个整数 firstLen
和 secondLen
,请你找出并返回两个非重叠 子数组 中元素的最大和
, 长度分别为 firstLen
和 secondLen
。
长度为 firstLen
的子数组可以出现在长为 secondLen
的子数组之前或之后,但二者必须是不重叠的。
子数组是数组的一个 连续 部分。
示例 1:
**输入:** nums = [0,6,5,2,2,5,1,9,4], firstLen = 1, secondLen = 2
**输出:** 20
**解释:** 子数组的一种选择中,[9] 长度为 1,[6,5] 长度为 2。
示例 2:
**输入:** nums = [3,8,1,3,2,1,8,9,0], firstLen = 3, secondLen = 2
**输出:** 29
**解释:** 子数组的一种选择中,[3,8,1] 长度为 3,[8,9] 长度为 2。
示例 3:
**输入:** nums = [2,1,5,6,0,9,5,0,3,8], firstLen = 4, secondLen = 3
**输出:** 31
**解释:** 子数组的一种选择中,[5,6,0,9] 长度为 4,[0,3,8] 长度为 3。
提示:
1 <= firstLen, secondLen <= 1000
2 <= firstLen + secondLen <= 1000
firstLen + secondLen <= nums.length <= 1000
0 <= nums[i] <= 1000
方法一:动态规划 + 滑动窗口
思路与算法
首先题目给出一个长度为 n 的数组 nums。现在我们需要返回两个长度分别为 firstLen 和 secondLen 的非重叠的子数组的最大和,firstLen} + \textit{secondLen} \le n,其中这两段子数组的前后顺序没有要求。
由于两段子数组的前后顺序没有区别,所以现在不妨设长度为 firstLen 的子数组在长度为 secondLen 的子数组前来计算此时的两段子数组的最大和。首先我们用 nums}[i,j] 来表示 nums}[i],\textit{nums}[i+1],\dots,\textit{nums}[j-1] 这一段子数组,并记 sum(\textit{nums}[l, r]) 表示子数组 nums}[l, r] 的和,dp}[i] 为 nums}[0,i + 1] 中长度为 firstLen 的最大子数组和,若不存在长度为 firstLen 的子数组则为 0。那么对于某一段长度为 secondLen 的子数组 nums}[j,j+\textit{secondLen}],0 \le j < j + \textit{secondLen} \le n,所以此时的两个数组的最大和为
dp[j-1]+sum(\textit{nums}[j, j + \textit{secondLen}])
又因为
dp}[i] = \max{\textit{dp}[i-1], sum(\textit{nums}[i+1-\textit{firstLen}, i+1])\
由于现在长度为 secondLen 在长度为 firstLen 的后面,所以用两个大小为 firstLen 和 secondLen 的滑动窗口分别从位置 0 和 firstLen 同时开始从左往右滑动,并在过程中维护窗口中的和。因为对于 \forall i < \textit{firstLen} - 1,有 dp}[i] = 0,并当 i = \textit{firstLen} - 1 时为初始第一个窗口的和。那么在两个窗口从左到右移动的过程中,通过移动第一个窗口来更新 dp 值,通过第二个窗口来计算此时的最大和,并记录移动过程中的最大值即可。
同理我们可以得到当 secondLen 的子数组在长度为 firstLen 的子数组前时,两段子数组的最大和,两种情况取较大值即为最终的答案。由于 dp}[i] 的求解只与 dp}[i-1] 有关,所以在实现的过程中我们可以通过「滚动数组」来进行空间优化。
代码
[sol1-C++]1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
| class Solution { public: int help(vector<int>& nums, int firstLen, int secondLen) { int suml = accumulate(nums.begin(), nums.begin() + firstLen, 0); int maxSumL = suml; int sumr = accumulate(nums.begin() + firstLen, nums.begin() + firstLen + secondLen, 0); int res = maxSumL + sumr; for (int i = firstLen + secondLen, j = firstLen; i < nums.size(); ++i, ++j) { suml += nums[j] - nums[j - firstLen]; maxSumL = max(maxSumL, suml); sumr += nums[i] - nums[i - secondLen]; res = max(res, maxSumL + sumr); } return res; }
int maxSumTwoNoOverlap(vector<int>& nums, int firstLen, int secondLen) { return max(help(nums, firstLen, secondLen), help(nums, secondLen, firstLen)); } };
|
[sol1-Java]1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
| class Solution { public int maxSumTwoNoOverlap(int[] nums, int firstLen, int secondLen) { return Math.max(help(nums, firstLen, secondLen), help(nums, secondLen, firstLen)); }
public int help(int[] nums, int firstLen, int secondLen) { int suml = 0; for (int i = 0; i < firstLen; ++i) { suml += nums[i]; } int maxSumL = suml; int sumr = 0; for (int i = firstLen; i < firstLen + secondLen; ++i) { sumr += nums[i]; } int res = maxSumL + sumr; for (int i = firstLen + secondLen, j = firstLen; i < nums.length; ++i, ++j) { suml += nums[j] - nums[j - firstLen]; maxSumL = Math.max(maxSumL, suml); sumr += nums[i] - nums[i - secondLen]; res = Math.max(res, maxSumL + sumr); } return res; } }
|
[sol1-Python3]1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
| class Solution: def maxSumTwoNoOverlap(self, nums: List[int], firstLen: int, secondLen: int) -> int: return max(self.help(nums, firstLen, secondLen), self.help(nums, secondLen, firstLen))
def help(self, nums, firstLen, secondLen): suml = 0 for i in range(0, firstLen): suml += nums[i] maxSumL = suml sumr = 0 for i in range(firstLen, firstLen + secondLen): sumr += nums[i] res = maxSumL + sumr j = firstLen for i in range(firstLen + secondLen, len(nums)): suml += nums[j] - nums[j - firstLen] maxSumL = max(maxSumL, suml) sumr += nums[i] - nums[i - secondLen] res = max(res, maxSumL + sumr) j += 1 return res
|
[sol1-Go]1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
| func help(nums []int, firstLen int, secondLen int) int { suml := 0 for i := 0; i < firstLen; i++ { suml += nums[i] } maxSumL := suml sumr := 0 for i := firstLen; i < firstLen+secondLen; i++ { sumr += nums[i] } res := maxSumL + sumr for i, j := firstLen+secondLen, firstLen; i < len(nums); i, j = i + 1, j + 1 { suml += nums[j] - nums[j - firstLen] maxSumL = max(maxSumL, suml) sumr += nums[i] - nums[i - secondLen] res = max(res, maxSumL + sumr) } return res }
func maxSumTwoNoOverlap(nums []int, firstLen int, secondLen int) int { return max(help(nums, firstLen, secondLen), help(nums, secondLen, firstLen)) }
func max(a int, b int) int { if a > b { return a } return b }
|
[sol1-JavaScript]1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
| var maxSumTwoNoOverlap = function(nums, firstLen, secondLen) { return Math.max(help(nums, firstLen, secondLen), help(nums, secondLen, firstLen)); };
function help(nums, firstLen, secondLen) { let suml = nums.slice(0, firstLen).reduce((acc, val) => acc + val, 0); let maxSumL = suml; let sumr = nums.slice(firstLen, firstLen + secondLen).reduce((acc, val) => acc + val, 0); let res = maxSumL + sumr; for (let i = firstLen + secondLen, j = firstLen; i < nums.length; i++, j++) { suml += nums[j] - nums[j - firstLen]; maxSumL = Math.max(maxSumL, suml); sumr += nums[i] - nums[i - secondLen]; res = Math.max(res, maxSumL + sumr); } return res; }
|
[sol1-C#]1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
| public class Solution { public int MaxSumTwoNoOverlap(int[] nums, int firstLen, int secondLen) { return Math.Max(Help(nums, firstLen, secondLen), Help(nums, secondLen, firstLen)); }
public int Help(int[] nums, int firstLen, int secondLen) { int suml = 0; for (int i = 0; i < firstLen; ++i) { suml += nums[i]; } int maxSumL = suml; int sumr = 0; for (int i = firstLen; i < firstLen + secondLen; ++i) { sumr += nums[i]; } int res = maxSumL + sumr; for (int i = firstLen + secondLen, j = firstLen; i < nums.Length; ++i, ++j) { suml += nums[j] - nums[j - firstLen]; maxSumL = Math.Max(maxSumL, suml); sumr += nums[i] - nums[i - secondLen]; res = Math.Max(res, maxSumL + sumr); } return res; } }
|
[sol1-C]1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
| inline int max(int a, int b) { return a > b ? a : b; }
int help(const int* nums, int numsSize, int firstLen, int secondLen) { int suml = 0; for (int i = 0; i < firstLen; i++) { suml += nums[i]; } int maxSumL = suml; int sumr = 0; for (int i = firstLen; i < firstLen + secondLen; i++) { sumr += nums[i]; } int res = maxSumL + sumr; for (int i = firstLen + secondLen, j = firstLen; i < numsSize; ++i, ++j) { suml += nums[j] - nums[j - firstLen]; maxSumL = max(maxSumL, suml); sumr += nums[i] - nums[i - secondLen]; res = max(res, maxSumL + sumr); } return res; }
int maxSumTwoNoOverlap(int* nums, int numsSize, int firstLen, int secondLen) { return max(help(nums, numsSize, firstLen, secondLen), help(nums, numsSize, secondLen, firstLen)); }
|
复杂度分析
- 时间复杂度:O(n),其中 n 为数组 nums 的长度。
- 空间复杂度:O(1),仅使用常量空间。