Longest Consecutive Sequence

题目

Given an unsorted array of integers, find the length of the longest consecutive elements sequence.

For example,
Given [100, 4, 200, 1, 3, 2],
The longest consecutive elements sequence is [1, 2, 3, 4]. Return its length: 4.

Your algorithm should run in O(n) complexity.

思路一:先排序,再遍历

时间复杂度是:O(nlogn+n)

代码如下:

class Solution {
public:
    int longestConsecutive(vector<int> &num) {
        // Start typing your C/C++ solution below
        // DO NOT write int main() function
        if(num.size()<2)
            return num.size();
        sort(num.begin(),num.end());
        
        int max = 1;
        int tmp = 1;
        int i=1;
        while(i<num.size())
        {
            if(num[i]==num[i-1]+1)
            {
                tmp = tmp+1;
            }
            if(num[i]>num[i-1]+1)
            {
                max = tmp>max?tmp:max;
                tmp = 1;
            }  
            i = i+1;
        }
        max = tmp>max?tmp:max;       
        return max;
        
    }
};

此代码是完全可以通过 Juge Small  和 Juge Large 的。但是题目要求 是 Unsorted 和 O(N)。

思路二:Hash 思想

具体想法:

1、一边遍历vector,一边建立HashTable表。

2、对于当前的第i个元素,如果Map里面未出现,将其加入到Map中,并设置length=1:

(1)检测其左边的连续序列的左边界值 low:low = num[i]-1-mymap[num[i]-1]+1   这是已得到的当前元素的左连续序列的左边界元素,

          并更新该low元素和当前元素的lenth:mymap[num[i]] = mymap[low]+1;   mymap[low] = mymap[low]+1; (即更新该元素的左连续序列左右边界长度值length+1)

(2)检查其右边的连续序列的右边界值high,此时分两种情况:

          (A)如果当前元素没有左连续序列部分,即左边界为当前元素,则做类似于(1)的操作来处理右连续序列;

           (B)若有,则将当前元素插入到左连续序列 low = num[i]-mymap[num[i]]+1  和 右连续序列 high = num[i]+1+mymap[num[i]+1]-1 之间,

             并更新左边界low 和右边界high 的lenth长度:mymap[low] = mymap[low]+mymap[high];     mymap[high] = mymap[low];


实现代码:

class Solution {
public:
    int longestConsecutive(vector<int> &num) {
        // Start typing your C/C++ solution below
        // DO NOT write int main() function
        if(num.size()<2)
            return num.size();        
        int max = 1;
        unordered_map<long,int> mymap;
        int low,high;
        for(int i=0;i<num.size();i++)
        {
            if(mymap.find(num[i])!=mymap.end())
                continue ;
            mymap[num[i]]=1;
            if(mymap.find(num[i]-1)!=mymap.end())
            {
                low = num[i]-1-mymap[num[i]-1]+1;
                mymap[num[i]] = mymap[low]+1;
                mymap[low] = mymap[low]+1;
                max = mymap[low]>max?mymap[low]:max;
            }
            if(mymap.find(num[i]+1)!=mymap.end())
            {
                if(mymap[num[i]]==1)
                {
                    high = num[i]+1+mymap[num[i]+1]-1;
                    mymap[num[i]] = mymap[high]+1;
                    mymap[high] = mymap[high]+1;
                    max = mymap[high]>max?mymap[high]:max;
                }
                else
                {
                    low = num[i]-mymap[num[i]]+1;
                    high = num[i]+1+mymap[num[i]+1]-1;
                    mymap[low] = mymap[low]+mymap[high];
                    mymap[high] = mymap[low];
                    max = mymap[high]>max?mymap[high]:max;
                }                
            }
        }  
        return max;
        
    }
};

还可以进一步简化:因为如果 mymap[num[i]]==1 和 mymap[num[i]]>1 都有 low = num[i]-mymap[num[i]]+1 ,所以可以合并处理。

class Solution {  
public:  
    int longestConsecutive(vector<int> &num) {  
        // Start typing your C/C++ solution below  
        // DO NOT write int main() function  
        if(num.size()<2)  
            return num.size();          
        int max = 1;  
        unordered_map<long,int> mymap;  
        int low,high;  
        for(int i=0;i<num.size();i++)  
        {  
            if(mymap.find(num[i])!=mymap.end())  
                continue ;  
            mymap[num[i]]=1;  
            if(mymap.find(num[i]-1)!=mymap.end())  
            {  
                low = num[i]-1-mymap[num[i]-1]+1;  
                mymap[num[i]] = mymap[low]+1;  
                mymap[low] = mymap[low]+1;  
                max = mymap[low]>max?mymap[low]:max;  
            }  
            if(mymap.find(num[i]+1)!=mymap.end())  
            {  
                low = num[i]-mymap[num[i]]+1;  
                high = num[i]+1+mymap[num[i]+1]-1; 
                mymap[low] = mymap[low]+mymap[high];  
                mymap[high] = mymap[low];  
                max = mymap[high]>max?mymap[high]:max;            
            }  
        }    
        return max;            
    }  
};

Run Status: Accepted!
Program Runtime: 88 milli secs


最新:

public class Solution {
    public int longestConsecutive(int[] nums) {
        if(nums.length == 0){
            return 0;
        }
        Set<Integer> hash = new HashSet<>();
        int max = 0;
        for(int e : nums){
            hash.add(e);
        }
        for(int e : nums){
            int lenth = 1;
            int left = e-1;
            while(hash.contains(left)){
                lenth ++;
                hash.remove(left);   // if you have not this line, it will cause Time Limit Exceeded
                left --;
            }
            int right = e+1;
            while(hash.contains(right)){
                lenth ++;
                hash.remove(right);   // if you have not this line, it will cause Time Limit Exceeded
                right ++;
            }
            max = Math.max(max, lenth);
        }
        return max;
    }
}



<think>好的,我现在需要帮助用户理解他们提供的C++代码段。首先,我会仔细阅读代码,看看它做了什么。 代码中有两个部分:首先声明了一个`unordered_set<int> num_set;`,然后遍历`nums`中的每个元素,将它们插入到`num_set`中。作者是力扣官方题解,可能这段代码是用于解决最长连续序列问题的。 用户可能想知道这段代码的作用以及它在特定问题中的用途。需要解释`unordered_set`的特点,比如快速查找、去重等。然后循环部分,将`nums`中的每个元素插入到集合中,这样可以去重,并为后续的查找操作做准备。 接下来要考虑用户可能的背景。可能是刚开始学习C++或者算法,对STL容器不熟悉。需要用简单易懂的语言解释,避免使用过多术语,但又要准确。比如,解释`unordered_set`是基于哈希表实现的,插入和查找的平均时间复杂度是O(1),这样在处理大量数据时效率高。 然后,这段代码可能出现在最长连续序列的问题中,比如LeetCode 128题。那个问题需要找到一个未排序数组中数字连续的最长序列的长度。通常的解法是先利用集合去重,然后遍历每个数字,检查是否是序列的起点,再逐次查找下一个数字是否存在。因此,这个`num_set`的创建是为了后续的查找操作,提高效率。 用户可能没有说明具体的应用场景,但根据代码来源,可以推测是用于这个问题。需要指出这段代码的作用是预处理数据,去重并存储在哈希集合中,以便快速查询。 另外,需要注意代码中的细节,比如循环中使用`const int& num`来避免不必要的拷贝,提高效率。虽然对于基本类型int来说,用不用引用区别不大,但这也是一种良好的编程习惯。 最后,要确保解释清晰,结构合理,先介绍代码的结构,再解释每个部分的作用,最后联系到实际问题中的应用,帮助用户全面理解。</think>这段代码的作用是**将数组 `nums` 中的所有元素存入一个哈希集合(`unordered_set`)中,实现去重和快速查找**。以下是逐行解析: --- ### 代码解析 ```cpp unordered_set<int> num_set; // 1. 声明一个无序集合(基于哈希表实现) for (const int& num : nums) { // 2. 遍历数组 nums 中的每个元素 num num_set.insert(num); // 3. 将 num 插入集合(自动去重) } ``` --- ### 关键点说明 1. **`unordered_set` 的特性**: - **去重**:若 `nums` 中有重复元素,集合中只会保留一个。 - **快速查找**:基于哈希表实现,插入和查找的平均时间复杂度为 $O(1)$。 - **无序存储**:元素不按顺序存储(与 `set` 不同)。 2. **循环的作用**: - 将数组 `nums` 的每个元素存入集合,为后续操作(如“检查某个值是否存在”)做准备。 --- ### 典型应用场景(如 LeetCode 128. 最长连续序列) 在这类问题中,这段代码常用于: 1. **去重**:避免重复元素干扰连续序列的计算。 2. **快速查询**:后续通过 `num_set.count(num+1)` 等操作,快速判断某个值是否存在,从而统计连续序列长度。 --- ### 代码对比:为什么用 `unordered_set` 而不是其他容器? | 容器 | 特点 | |------------------|---------------------------------------------------------------------| | `vector` | 不去重,查找效率低($O(n)$) | | `set` | 自动排序,但插入和查找效率为 $O(\log n)$ | | `unordered_set` | 无序,但插入和查找效率最高($O(1)$),**最适合当前需求** | --- ### 示例说明 假设输入 `nums = [3, 1, 2, 2, 4]`,执行代码后: - `num_set` 中的内容为 `{1, 2, 3, 4}`(去重且无序存储)。 - 后续可通过 `num_set.find(3)` 快速判断数字 `3` 是否存在。 --- ### 总结 这段代码是**预处理数据的常见操作**,核心目的是通过哈希集合实现: 1. 去重 2. 为后续高频查找提供 $O(1)$ 时间复杂度的支持
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值