Codeforces 30C Shooting Gallery【dp】

最新推荐文章于 2022-08-29 19:55:25 发布

mengxiang000000

最新推荐文章于 2022-08-29 19:55:25 发布

阅读量564

点赞数

分类专栏： dp 文章标签： Codeforces 30C

本文链接：https://blog.csdn.net/mengxiang000000/article/details/53007032

版权

dp 专栏收录该内容

351 篇文章 2 订阅

订阅专栏

One warm and sunny day king Copa decided to visit the shooting gallery, located at the Central Park, and try to win the main prize — big pink plush panda. The king is not good at shooting, so he invited you to help him.

The shooting gallery is an infinite vertical plane with Cartesian coordinate system on it. The targets are points on this plane. Each target is described by it's coordinates x_i, and y_i, by the time of it's appearance t_i and by the number p_i, which gives the probability that Copa hits this target if he aims at it.

A target appears and disappears instantly, so Copa can hit the target only if at the moment t_i his gun sight aimed at (x_i, y_i). Speed of movement of the gun sight on the plane is equal to 1. Copa knows all the information about the targets beforehand (remember, he is a king!). He wants to play in the optimal way, which maximizes the expected value of the amount of hit targets. He can aim at any target at the moment 0.

Input

The first line contains integer n (1 ≤ n ≤ 1000) — amount of targets in the shooting gallery. Then n lines follow, each describing one target. Each description consists of four numbers x_i, y_i, t_i, p_i (where x_i, y_i, t_i — integers, - 1000 ≤ x_i, y_i ≤ 1000, 0 ≤ t_i ≤ 10⁹, real number p_i is given with no more than 6 digits after the decimal point, 0 ≤ p_i ≤ 1). No two targets may be at the same point.

Output

Output the maximum expected value of the amount of targets that was shot by the king. Your answer will be accepted if it differs from the correct answer by not more than 10^- 6.

Examples

Input

1
0 0 0 0.5

Output

0.5000000000

Input

2
0 0 0 0.6
5 0 5 0.7

Output

1.3000000000

题目大意：

一共有N个点有靶子，初始的时候可以让枪放在任意位子，然后给你N个靶子的信息，包括其坐标，出现的时间，以及打中的分数.

问最多能够获得多少积分。枪每一单位时间移动的速度为1个坐标单位。

思路：

1、考虑这样一个问题：我们首先一定是先打掉一个先出现的靶子之后再打其他后出现靶子，那么我们此时不妨设定dp【i】【j】表示打完第i个靶子之后再打第j个靶子能够得到的最高分数，其状态转移方程也很好想：dp【i】【j】=max（dp【i】【j】，dp【k】【i】+val【j】）；【1<=k<=n】暴力做法时间复杂度O（n^3）；显然需要对其进行优化。

2、继续考虑这样一个问题：我们首先一定是打掉一个先出现的靶子之后再打其他后出现的靶子，那么其一定有一个时间顺序，那么此时我们不妨将所有靶子出现的时间按照升序排序，那么我们此时设定到第i个靶子在第j个靶子之后打掉，那么此时一定j<i，那么我们此时可以设定dp【i】，表示打第i个出现的靶子的时候能够获得的最高分数。那么此时状态转移方程：

dp【i】=max（dp【i】，dp【j】+val【i】）【0<=j<i&&打完j之后移动枪的时间是充足的】；

此时时间复杂度O(N^2)

Ac代码：

#include<stdio.h>
#include<string.h>
#include<algorithm>
#include<math.h>
using namespace std;
struct node
{
    int x,y,t;
    double val;
}a[1515];
double dp[1515];
int cmp(node a,node b)
{
    return a.t<b.t;
}
int main()
{
    int n;
    while(~scanf("%d",&n))
    {
        for(int i=0;i<n;i++)
        {
            scanf("%d%d%d%lf",&a[i].x,&a[i].y,&a[i].t,&a[i].val);
        }
        sort(a,a+n,cmp);
        memset(dp,0,sizeof(dp));
        for(int i=0;i<n;i++)
        {
            dp[i]=a[i].val;
            for(int j=0;j<i;j++)
            {
                long long int t1=(a[i].t-a[j].t);
                t1*=(a[i].t-a[j].t);
                long long int tmp1=(a[i].x-a[j].x);
                long long int tmp2=(a[i].y-a[j].y);
                long long int t2=tmp1*tmp1+tmp2*tmp2;
                if(t1>=t2)
                {
                    dp[i]=max(dp[i],dp[j]+a[i].val);
                }
            }
        }
        double output=0;
        for(int i=0;i<n;i++)
        {
            output=max(output,dp[i]);
        }
        printf("%lf\n",output);
    }
}