zoukankan      html  css  js  c++  java
  • hdu4179 限制最短路

    题意:
          这个题目估计读懂题意就ok了,关键是题意蛋疼,像我这样的英语渣渣活着可真难啊,题意大体是这样,给你n个点m条无向边,给你起点和终点,让你求从起点到终点的最短路径,其中有一些限制:
    (1) 所有的边的d必须小于等于题目给的D
    (2) 必须至少有一条边的d 等于题目给的 D
    下面说一下d的求法,对于每条边ab,如果a.z >= b.z也就是下坡,d = 0,否则d等于高度变化的绝对值*100 除以 线段在xoy面上的投影线段的长度,最后向下取证(不用考虑竖直上下的情况)。

    思路:

          题意懂了这个题目就好做了,直接判断建图,建两个图,一个是正向的一个是反向的,然后跑两边最短路,然后在枚举每一条d == D 的边取得最优的就行了。


    #include<stdio.h>
    #include<string.h>
    #include<queue>
    #include<math.h>
    
    #define N_node 11111
    #define N_edge 55555
    #define INF 1000000000
    
    using namespace std;
    
    typedef struct
    {
       int to ,next;
       double cost;
    }STAR;
    
    typedef struct
    {
       int a ,b;
    }EDGE;
    
    typedef struct
    {
       double x ,y ,z;
    }NODE;
    
    STAR E1[N_edge] ,E2[N_edge];
    EDGE edge[N_edge] ,ee[N_edge];
    NODE node[N_node];
    int list1[N_node] ,list2[N_node] ,tot;
    double dis1[N_node] ,dis2[N_node];
    
    void add(int a ,int b ,double c)
    {
       E1[++tot].to = b;
       E1[tot].cost = c;
       E1[tot].next = list1[a];
       list1[a] = tot;
       E2[tot].to = a;
       E2[tot].cost = c;
       E2[tot].next = list2[b];
       list2[b] = tot;
    }
    
    double get_dis(NODE a ,NODE b)
    {
       double x = (a.x - b.x) * (a.x - b.x);
       double y = (a.y - b.y) * (a.y - b.y);
       double z = (a.z - b.z) * (a.z - b.z);
       return sqrt(x + y + z);
    }
    
    int get_d(NODE a ,NODE b)
    {
        if(a.z >= b.z) return 0;
        double x = (a.x - b.x) * (a.x - b.x);
        double y = (a.y - b.y) * (a.y - b.y);
        double z = b.z - a.z;
        return int(z * 100 / sqrt(x + y));
    }
    
    void spfa(int s ,int n ,int list[] ,double s_x[] ,STAR E[])
    {
       int mark[N_node] = {0};
       for(int i = 0 ;i <= n ;i ++) s_x[i] = INF;
       s_x[s] = 0 ,mark[s] = 1;
       queue<int>q;
       q.push(s);
       while(!q.empty())
       {
          int xin ,tou;
          tou = q.front();
          q.pop();
          mark[tou] = 0;
          for(int k = list[tou] ;k ;k = E[k].next)
          {
             xin = E[k].to;
             if(s_x[xin] > s_x[tou] + E[k].cost)
             {
                s_x[xin] = s_x[tou] + E[k].cost;
                if(!mark[xin])
                {
                   mark[xin] = 1;
                   q.push(xin);
                }
             }
          }
       }
    }
    
    int main ()
    {
       int n ,m ,i ,s ,t ,d ,a ,b;
       while(~scanf("%d %d" ,&n ,&m) && n + m)
       {
          for(i = 1 ;i <= n ;i ++)
          scanf("%lf %lf %lf" ,&node[i].x ,&node[i].y ,&node[i].z);
          for(i = 1 ;i <= m ;i ++)
          scanf("%d %d" ,&ee[i].a ,&ee[i].b);
          scanf("%d %d %d" ,&s ,&t ,&d);
          memset(list1 ,0 ,sizeof(list1));
          memset(list2 ,0 ,sizeof(list2)) ,tot = 1;
          int edge_t = 0;
          for(i = 1 ;i <= m ;i ++)
          {
             a = ee[i].a ,b = ee[i].b;
             int d1 = get_d(node[a] ,node[b]);
             int d2 = get_d(node[b] ,node[a]);
             double dis = get_dis(node[a] ,node[b]);
             if(d1 <= d) add(a ,b ,dis);
             if(d2 <= d) add(b ,a ,dis);
             if(d1 == d) 
             {
                edge[++edge_t].a = a;
                edge[edge_t].b = b;
             }
             if(d2 == d)
             {
                edge[++edge_t].a = b;
                edge[edge_t].b = a;
             }
          }
          
          spfa(s ,n ,list1 ,dis1 ,E1);
          spfa(t ,n ,list2 ,dis2 ,E2);
          
          double ans = INF;
          for(i = 1 ;i <= edge_t ;i ++)
          {
             double now = dis1[edge[i].a] + get_dis(node[edge[i].a] ,node[edge[i].b]) + dis2[edge[i].b];
             if(ans > now) ans = now;
          }
          ans == INF ? puts("None") : printf("%.1lf
    " ,ans);
       }
       return 0;
    }
    

  • 相关阅读:
    K-means Algorithm
    Naive Bayes Algorithm
    Generalized Linear Models
    Support Vector Machine
    wordpress迁移
    Gaussian Discriminant Analysis
    Locally Weighted Regression
    Matlab绘图高级部分
    Supervised Learning-Regression
    html 字符串互转DOM
  • 原文地址:https://www.cnblogs.com/csnd/p/12062956.html
Copyright © 2011-2022 走看看