How to Find the Longest Harmonious Subsequence?

  • 时间：2020-09-18 17:01:02
  • 分类：网络文摘
  • 阅读：127 次

We define a harmounious array as an array where the difference between its maximum value and its minimum value is exactly 1. Now, given an integer array, you need to find the length of its longest harmonious subsequence among all its possible subsequences.

Example 1:
Input: [1,3,2,2,5,2,3,7]
Output: 5
Explanation: The longest harmonious subsequence is [3,2,2,2,3].

Note: The length of the input array will not exceed 20,000.

Bruteforce All Subsequences

One solution is to bruteforce all the subsequences, which is exponential. The total subsequences is C(n, 1) + C(n, 2) + C(n, 3) + … C(n, n). We can use two nested loops, the outer loop is from 0 to (2^n)-1 where n is the size of the array, and the inner loop is from 0 to n-1.

By generating all the subsequences, we can increment the counter once we find the Harmonious subsequence where the min and max value is exactly one difference.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20

class Solution {
public:
    int findLHS(vector<int>& nums) {
        int ans = 0;
        for (int i = 0; i < (1 << nums.size()); ++ i) {
            int count = 0, minv = INT_MAX, maxv = INT_MIN;
            for (int j = 0; j < nums.size(); ++ j) {
                if ((i & (1 << j)) != 0) {
                    minv = min(minv, nums[j]);
                    maxv = max(maxv, nums[j]);
                    count ++;
                }
            }
            if ((minv != INT_MAX) && (maxv == minv + 1)) {
                ans = max(ans, count);
            }
        }        
        return ans;
    }
};
class Solution {
public:
    int findLHS(vector<int>& nums) {
        int ans = 0;
        for (int i = 0; i < (1 << nums.size()); ++ i) {
            int count = 0, minv = INT_MAX, maxv = INT_MIN;
            for (int j = 0; j < nums.size(); ++ j) {
                if ((i & (1 << j)) != 0) {
                    minv = min(minv, nums[j]);
                    maxv = max(maxv, nums[j]);
                    count ++;
                }
            }
            if ((minv != INT_MAX) && (maxv == minv + 1)) {
                ans = max(ans, count);
            }
        }        
        return ans;
    }
};

Apparently, this bruteforce is extremely inefficient.

Using a Hashmap

We can do one-pass scan to count the occurences for each number in a hash map. Then, by going through the numbers in the hash map, then the Harmonious subsequence can be checked if the next number is in the hash map – the length will be the sum of the counters for both numbers.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17

class Solution {
public:
    int findLHS(vector<int>& nums) {        
        unordered_map<int, int> count;
        for (const auto &n: nums) {
            count[n] ++;
        }
        int ans = 0;
        for (auto it = begin(count); it != end(count); ++ it) {
            int next = it->first + 1;
            if (count.find(next) != count.end()) {
                ans = max(ans, it->second + count[next]);
            }
        }
        return ans;
    }
};
class Solution {
public:
    int findLHS(vector<int>& nums) {        
        unordered_map<int, int> count;
        for (const auto &n: nums) {
            count[n] ++;
        }
        int ans = 0;
        for (auto it = begin(count); it != end(count); ++ it) {
            int next = it->first + 1;
            if (count.find(next) != count.end()) {
                ans = max(ans, it->second + count[next]);
            }
        }
        return ans;
    }
};

We can also do this one-pass, by checking the previous and next number in the hash map as we going through each number.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17

class Solution {
public:
    int findLHS(vector<int>& nums) {
        int ans = 0;
        unordered_map<int, int> data;
        for (const auto &n: nums) {
            data[n] ++;
            if (data.find(n + 1) != data.end()) {
                ans = max(ans, data[n] + data[n + 1]);
            }
            if (data.find(n - 1) != data.end()) {
                ans = max(ans, data[n] + data[n - 1]);
            }
        }        
        return ans;
    }
};
class Solution {
public:
    int findLHS(vector<int>& nums) {
        int ans = 0;
        unordered_map<int, int> data;
        for (const auto &n: nums) {
            data[n] ++;
            if (data.find(n + 1) != data.end()) {
                ans = max(ans, data[n] + data[n + 1]);
            }
            if (data.find(n - 1) != data.end()) {
                ans = max(ans, data[n] + data[n - 1]);
            }
        }        
        return ans;
    }
};

Both approaches are using hash map, thus O(N) space and O(N) time complexity.

–EOF (The Ultimate Computing & Technology Blog) —

推荐阅读：
论书的种类进化　 难忘七月之青塘小学实践　 数正方形个数的数学题：5X5方格共有36个格点　 数学题：小兔晴天可以采12个蘑菇　 数学题：老师说我像你这么大的时候你才刚刚1岁　 数学题：甲、丙两组人比乙组人数多2倍还多2人　 数学题：22名家长和老师陪同学生参加某次数学竞赛　 数学题：求X的长度是多少厘米　 正好可以把它平均切成2个相等的正方体　 数学题：求三角形ABC中阴影正方形的边长是多少厘米　
评论列表
添加评论