Omi, Raymondo, Clay and Kimiko are on new adventure- in search of new Shen Gong Wu. But Evil
Boy Genius Jack Spicer is also there. Omi and Jack found the Shen Gong Wu at the same time so they
rushed for it but alas they touched it at the same time. Then what? It is time for “Xiaolin Showdown”.
Jack challenged Omi to play a game. The game is simple! There will be an N ∗ N board where
each cell in the board contains some number. They have to assign numbers to each row and column
separately so that w(i, j) ≤ row(i) + col(j) where w(i, j) is the number assigned to the cell located
at i-th row and j-th column, row(i) is the number assigned to i-th row and col(j) is the number

assigned to j-th column. That is simple isnt it? Well . . . the main part is that you have to minimize
1≤i≤n
(row(i) + col(j)).
Jack has taken his favorite “Monkey Stuff” and Omi has taken “Golden Tiger Claw”. With the help
of this “Golden Tiger Claw”, he can go anywhere in the world. He has come to you and seeking your
help. Jack is using his computer to solve this problem. So do it quick! Find the most optimal solution
for Omi so that you can also be part of history in saving the world from the darkness of evil.
Input
Input contains 15 test cases. Each case starts with N. Then there are N lines containing N numbers
each. All the numbers in input is positive integer within the limit 100 except N which can be at most
500.
Output
For each case in the first line there will be N numbers, the row assignments. In the next line there
will N column assignment. And at the last line the minimum sum should be given. If there are several
possible solutions give any.
Note: Be careful about the output format. You may get Wrong Answer if you don’t output properly.
Sample Input
2
1 1
1 1
Sample Output
1 1
0 0
2

【题意】

  给出一个n*n的矩阵(n<=500)给每一行x[i],每一列标号y[i],使得对任意a[i][j],x[i]+y[j]>=a[i][j]求行标与列标和最小

【分析】

  事实上和最佳匹配没什么关系,但是我们进行KM算法的时候,有w(i,j)<=row(i)+col(j),并且算出来的顶标之和是最小的,so。。。

代码如下:

 #include<cstdio>
#include<cstdlib>
#include<cstring>
#include<iostream>
#include<algorithm>
#include<queue>
#include<cmath>
using namespace std;
#define Maxn 510
#define Maxm 250010
#define INF 0xfffffff struct node
{
int x,y,c,next;
}t[Maxm];int len;
int first[Maxn]; int mymin(int x,int y) {return x<y?x:y;}
int mymax(int x,int y) {return x>y?x:y;} void ins(int x,int y,int c)
{
t[++len].x=x;t[len].y=y;t[len].c=c;
t[len].next=first[x];first[x]=len;
} int a[Maxn][Maxn];
int n; int lx[Maxn],ly[Maxn],match[Maxn],slack[Maxn];
bool visx[Maxn],visy[Maxn]; bool ffind(int x)
{
visx[x]=;
for(int i=first[x];i;i=t[i].next) if(!visy[t[i].y])
{
int y=t[i].y;
if(t[i].c==lx[x]+ly[y])
{
visy[y]=;
if(!match[y]||ffind(match[y]))
{
match[y]=x;
return ;
}
}
else slack[y]=mymin(slack[y],lx[x]+ly[y]-t[i].c);
}
return ;
} void solve()
{
memset(match,,sizeof(match));
memset(lx,,sizeof(lx));
memset(ly,,sizeof(ly));
for(int i=;i<=n;i++)
for(int j=first[i];j;j=t[j].next) lx[i]=mymax(lx[i],t[j].c); for(int i=;i<=n;i++)
{
for(int j=;j<=n;j++)
slack[j]=INF;
while()
{
memset(visx,,sizeof(visx));
memset(visy,,sizeof(visy));
if(ffind(i)) break;
int delta=INF;
for(int j=;j<=n;j++)
{
if(!visy[j])
{
delta=mymin(delta,slack[j]);
}
}
if(delta==INF) return;
for(int j=;j<=n;j++)
{
if(visx[j]) lx[j]-=delta;
if(visy[j]) ly[j]+=delta;
else slack[j]-=delta;
}
}
}
} int main()
{
while(scanf("%d",&n)!=EOF)
{
len=;
memset(first,,sizeof(first));
for(int i=;i<=n;i++)
for(int j=;j<=n;j++)
{
int x;
scanf("%d",&x);
ins(i,j,x);
}
solve();
for(int i=;i<=n;i++) printf("%d ",lx[i]);printf("\n");
for(int i=;i<=n;i++) printf("%d ",ly[i]);printf("\n");
int ans=;
for(int i=;i<=n;i++) ans+=lx[i]+ly[i];
printf("%d\n",ans);
}
return ;
}

[UVA 11383]

2016-10-27 15:13:52

【UVA 11383】 Golden Tiger Claw (KM算法副产物)的更多相关文章

  1. UVA 11383 - Golden Tiger Claw(二分图完美匹配扩展)

    UVA 11383 - Golden Tiger Claw 题目链接 题意:给定每列和每行的和,给定一个矩阵,要求每一个格子(x, y)的值小于row(i) + col(j),求一种方案,而且全部行列 ...

  2. UVA 11383 Golden Tiger Claw 金虎爪(KM算法)

    题意: 给一个n*n的矩阵,每个格子中有正整数w[i][j],试为每行和每列分别确定一个数字row[i]和col[i],使得任意格子w[i][j]<=row[i]+col[j]恒成立.先输row ...

  3. 【KM算法】UVA 11383 Golden Tiger Claw

    题目大意 给你一个\(n×n\)的矩阵G,每个位置有一个权,求两个一维数组\(row\)和\(col\),使\(row[i] + col[j]\ge G[i][j]\),并且\(∑row+∑col\) ...

  4. UVA11383 Golden Tiger Claw —— KM算法

    题目链接:https://vjudge.net/problem/UVA-11383 题解: 根据KM()算法,标杆满足:l(x) + l(y) >= w(x, y) . 当求完最大权匹配之后,所 ...

  5. UVA 11383 Golden Tiger Claw 题解

    题目 --> 题解 其实就是一个KM的板子 KM算法在进行中, 需要满足两个点的顶标值之和大于等于两点之间的边权, 所以进行一次KM即可. KM之后, 顶标之和就是最小的.因为如果不是最小的,就 ...

  6. UVA11383 Golden Tiger Claw KM算法

    题目链接:传送门 分析 这道题乍看上去没有思路,但是我们仔细一想就会发现这道题其实是一个二分图最大匹配的板子 我们可以把这道题想象成将男生和女生之间两两配对,使他们的好感度最大 我们把矩阵中的元素\( ...

  7. Uva - 11383 - Golden Tiger Claw

    题意:一个N*N的矩阵,第i行第j列的元素大小为w[i][j],每行求一个数row[i],每列求一个数col[j],使得row[i] + col[j] >= w[i][j],且所有的row[]与 ...

  8. UVA 11383 Golden Tiger Claw(最佳二分图完美匹配)

    题意:在一个N*N的方格中,各有一个整数w(i,j),现在要求给每行构造row(i),给每列构造col(j),使得任意w(i,j)<=row(i)+col(j),输出row(i)与col(j)之 ...

  9. uva11383 Golden Tiger Claw 深入理解km算法

    /** 题目: uva11383 Golden Tiger Claw 深入理解km算法 链接:https://vjudge.net/problem/UVA-11383 题意:lv 思路:lrj训练指南 ...

随机推荐

  1. Java对象与Json之间的转换

    使用Jackson的ObjectMapper对象的readValue和writeValueAsString方法可以进行转换. 对于简单基本类型或String类型的对象,使用上述方法可以满足. 但是如果 ...

  2. java Spring 基于注解的配置(一)

    注解引用:1.service.xml 配置注解模式 <?xml version="1.0" encoding="UTF-8"?> <beans ...

  3. 安卓获取Assets目录下的资源

    获取Assets目录下的资源 *:first-child { margin-top: 0 !important; } body>*:last-child { margin-bottom: 0 ! ...

  4. C#获取当前路径,获取当前路径的上一层路径

    C#获取当前路径的方法如下: (1)string path1 = System.Environment.CurrentDirectory; //C:\...\bin\Debug -获取和设置当前工作目 ...

  5. (一)初识Android

    第一节:手机操作系统简介 目前的主流智能操作系统有:Android , IOS , windows mobile ; Android 开源,属于谷歌公司,市场份额较大,前景广阔: IOS 属于苹果公司 ...

  6. react native android 开发,基础配置笔记。

    一.React-native-device-info https://github.com/rebeccahughes/react-native-device-info 二.修改App名称 三.定位权 ...

  7. Activiti安装

    1.Activiti下载与简介 1.1  简介 Activiti项目是一项新的基于Apache许可的开源BPM平台,从基础开始构建,旨在提供支持新的BPMN 2.0标准,包括支持对象管理组(OMG), ...

  8. javaee学习-Cookie使用范例

    Java中的javax.servlet.http.Cookie类用于创建一个Cookie Cookie类的主要方法 No. 方法 类型 描述 1 Cookie(String name, String  ...

  9. Nuage SDN

    Nuage推出纯软件解决方案虚拟化业务平台(VSP)由三部分组成:虚拟化业务目录(VSD).虚拟化业务控制器(VSC)和虚拟路由与交换(VRS). VSD是业务/IT策略引擎,可提供业务模板与分析,每 ...

  10. STL unique使用问题

    string strs[] = {"one","one","two","three","three" ...