zoukankan      html  css  js  c++  java
  • 树(最小乘积生成树,克鲁斯卡尔算法):BOI timeismoney

    The NetLine company wants to offer broadband internet to N towns. For this, it suffices to construct
    a network of N-1 broadband links between the towns, with the property that a message can travel
    from any town to any other town on this network. NetLine has already identified all pairs of towns
    between which a direct link can be constructed. For each such possible link, they know the cost and
    the time it would take to construct the link.
    The company is interested in minimizing both the total amount of time (links are built one at a time)
    and the total amount of money spent to build the entire network. Since they couldn’t decide among
    the two criteria, they decided to use the following formula to evaluate the value of a network:
    SumTime = sum of times spent to construct the chosen links
    SumMoney = sum of the money spent to construct the chosen links
    V = SumTime * SumMoney
    Task
    Find a list of N-1 links to build, which minimizes the value V.
    Description of input
    The first line of input contains integers N – the number of towns and M – the number of pairs of
    towns which can be connected. The towns are numbered starting from 0 to N-1. Each of the next M
    lines contain four integers x, y, t and c – meaning town x can be connected to town y in time t and
    with cost c.
    Description of output
    In the first line of output print two numbers: the total time (SumTime) and total money (Sum-
    Money) used in the optimal solution (the one with minimal value V), separated by one space. The
    next N-1 lines describe the links to be constructed. Each line contains a pair of numbers (x,y) describing
    a link to be build (which must be among the possible links described in the input file). The
    pairs can be printed out in any order. When multiple solutions exist, you may print any of them.

    Constraints

    · 1 ≤ N ≤ 200
    · 1 ≤ M ≤ 10 000
    · 0 ≤ x,y ≤ N-1
    · 1 ≤ t,c ≤ 255
    · One test has M = N - 1
    · 40% of the tests will have for each possible link t = c
    Example
    timeismoney.in
    5 7
    0 1 161 79
    0 2 161 15
    0 3 13 153
    1 4 142 183
    2 4 236 80
    3 4 40 241
    2 1 65 92

    timeismoney.out

    279 501
    2 1
    0 3
    0 2
    3 4

      方案啥的很好解决,就不写了。

      和HNOI2014画框有类似的思想。

     1 #include <iostream>
     2 #include <cstring>
     3 #include <cstdio>
     4 using namespace std;
     5 const int maxn=210;
     6 const int maxm=10010;
     7 int fa[maxn],a[maxm],b[maxm];
     8 int u[maxm],v[maxm],n,m;
     9 
    10 struct Node{
    11     int x,y,z,id;
    12     Node(int a=0,int b=0,int c=0,int d=0){
    13         x=a;y=b;z=c;id=d;
    14     }
    15     friend bool operator <(Node a,Node b){
    16         return a.z<b.z;
    17     }
    18 }p[maxm];
    19 
    20 struct Point{
    21     int x,y;
    22     Point(int a=0,int b=0){
    23         x=a;y=b;
    24     }
    25     friend bool operator ==(Point a,Point b){
    26         return a.x==b.x&&a.y==b.y;
    27     }
    28 }lo,hi;
    29 
    30 int Find(int x){
    31     return fa[x]==x?x:fa[x]=Find(fa[x]);
    32 }
    33 
    34 Point Get_Ans(){
    35     sort(p+1,p+m+1);Point ret(0,0);
    36     for(int i=1;i<=n;i++)fa[i]=i;
    37     for(int i=1;i<=m;i++){
    38         int x=p[i].x,y=p[i].y;
    39         if(Find(x)!=Find(y)){
    40             ret.x+=a[p[i].id];
    41             ret.y+=b[p[i].id];
    42             fa[Find(y)]=Find(x);
    43         }
    44     }
    45     return ret;
    46 }
    47 
    48 Point Solve(Point l,Point r){
    49     for(int i=1;i<=m;i++)
    50         p[i]=Node(u[i],v[i],b[i]*(r.x-l.x)-a[i]*(r.y-l.y),i);
    51     Point mid=Get_Ans();
    52     if(mid==l||mid==r)return l.x*l.y<r.x*r.y?l:r;
    53     l=Solve(l,mid);r=Solve(mid,r);
    54     return l.x*l.y<r.x*r.y?l:r;    
    55 }
    56 
    57 int main(){
    58     freopen("timeismoney.in","r",stdin);
    59     freopen("timeismoney.out","w",stdout);
    60     scanf("%d%d",&n,&m);
    61     for(int i=1;i<=m;i++){
    62         scanf("%d%d",&u[i],&v[i]);u[i]+=1;v[i]+=1;
    63         scanf("%d%d",&a[i],&b[i]);
    64     }
    65     
    66     for(int i=1;i<=m;i++)p[i]=Node(u[i],v[i],a[i],i);lo=Get_Ans();
    67     for(int i=1;i<=m;i++)p[i]=Node(u[i],v[i],b[i],i);hi=Get_Ans();    
    68     Point ans=Solve(lo,hi);
    69     printf("%d %d
    ",ans.x,ans.y);    
    70     return 0;
    71 }
  • 相关阅读:
    R语言数据读入函数read.table
    R语言最小浮点数
    R语言randomForest包实现随机森林——iris数据集和kyphosis数据集
    R语言 rwordseg包的下载
    R语言AMORE包实现BP神经网络——German数据集
    R语言清除内存,无法分配大小为324.5 Mb的矢量
    R语言多重共现性的检测
    Machine Learning for hackers读书笔记(六)正则化:文本回归
    Machine Learning for hackers读书笔记(五)回归模型:预测网页访问量
    完整的Django入门指南学习笔记4
  • 原文地址:https://www.cnblogs.com/TenderRun/p/5700737.html
Copyright © 2011-2022 走看看