我使用std :: pair来获取每个时间段的开始时间和结束时间,并将所有这些时间段放在一个数组中(比如类[no_of_time_periods])。 代码:
int getClassTimes(int N)
{
int subsets, startTime, endTime;
std::pair<int,int> classes[N];
for(int i=0;i<N;i++)
{
printf("Please Enter Class %d Start Time and End Time:",(i+1));
scanf("%d",&startTime);
scanf("%d",&endTime);
classes[i] = std::make_pair(startTime,endTime);
}
subsets = compute(classes,N);
return subsets;
}
我知道我可以使用以下条件检查两个时间段是否重叠:
if(!(classes[i].first<classes[j].second && classes[i].second>classes[j].first))
但我想检查两个以上的时间段。 例如:
Input:
No_Of_Time_Periods = 5
Time_Period 1 = 1 to 3
Time_Period 2 = 3 to 5
Time_Period 3 = 5 to 7
Time_Period 4 = 2 to 4
Time_Period 5 = 4 to 6
Calculation(Number of subsets of non-overlapping Time Periods):
**Note: If end time of one class is start time of other, they are non-overlapping.
Ex((1 to 3) and (3 to 5) are non-overlapping.)**
(1 to 3)(3 to 5)
(1 to 3)(3 to 5)(5 to 7)
(1 to 3)(4 to 6)
(1 to 3)(5 to 7)
(2 to 4)(4 to 6)
(2 to 4)(5 to 7)
(3 to 5)(5 to 7)
Output:
Total Number of subsets of non-overlapping classes:7
我找到了逻辑并且编写了代码。但是在在线评判(SPOJ)中提交时,它说“超出时间限制”。所以,显然我的代码没有得到很好的优化。如何使用c ++实现更好的性能?任何帮助,将不胜感激。提前谢谢!
答案 0 :(得分:1)
这太令人困惑了! 重叠测试应该是这样的:
if ((classes[i].second < classes[j].first) || (classes[i].first > classes[j].second))
printf("no overlap");
if (!((classes[i].second < classes[j].first) || (classes[i].first > classes[j].second)))
printf("overlap");
假设classes[n].first
始终小于classes[n].second
答案 1 :(得分:-1)
检查间隔树。您将获得对数的复杂性:http://en.wikipedia.org/wiki/Interval_tree