• 51Nod1518 稳定多米诺覆盖 动态规划 插头dp 容斥原理


    原文链接https://www.cnblogs.com/zhouzhendong/p/51Nod1518.html

    题目传送门 - 51Nod1518

    题意

    51Nod真是个好OJ ,题意概括的真好,有助于博主偷懒不写题意概括。给51Nod 点赞!

    题解

      首先,我们忽略那个“稳定”的要求,求方案数。

      显然是一个插头dp裸题,我们可以在 $O(n^2cdot 2^n)$ 的时间复杂度中求出所有长宽的矩形区域的覆盖方案数。

      然后我们考虑容斥原理,奇加偶减。首先,枚举哪些相邻行之间有一条不穿过骨牌的直线,然后,用一个 $O(n)$ DP 来解决相邻列之间分割线的容斥。

      总的时间复杂度 $O(n^22^n)$ 。打出表之后,询问 $O(1)$ 。

    代码

    看着那些运行效率榜上15MS的代码我于是交了一份 0MS 的代码。正常的代码在这份代码之后。

    #include <bits/stdc++.h>
    int n,m,ans[17][17]={
    {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
    {0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
    {0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
    {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
    {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
    {0,0,0,0,0,0,6,0,108,0,1182,0,10338,0,79818,0,570342},
    {0,0,0,0,0,6,0,124,62,1646,1630,18120,25654,180288,317338,1684956,3416994},
    {0,0,0,0,0,0,124,0,13514,0,765182,0,32046702,0,136189727,0,378354090},
    {0,0,0,0,0,108,62,13514,25506,991186,3103578,57718190,238225406,965022920,388537910,937145938,315565230},
    {0,0,0,0,0,0,1646,0,991186,0,262834138,0,462717719,0,560132342,0,699538539},
    {0,0,0,0,0,1182,1630,765182,3103578,262834138,759280991,264577134,712492587,886997066,577689269,510014880,807555438},
    {0,0,0,0,0,0,18120,0,57718190,0,264577134,0,759141342,0,567660301,0,47051173},
    {0,0,0,0,0,10338,25654,32046702,238225406,462717719,712492587,759141342,398579168,83006813,821419653,942235780,558077885},
    {0,0,0,0,0,0,180288,0,965022920,0,886997066,0,83006813,0,690415372,0,620388364},
    {0,0,0,0,0,79818,317338,136189727,388537910,560132342,577689269,567660301,821419653,690415372,796514774,696587391,175421667},
    {0,0,0,0,0,0,1684956,0,937145938,0,510014880,0,942235780,0,696587391,0,856463275},
    {0,0,0,0,0,570342,3416994,378354090,315565230,699538539,807555438,47051173,558077885,620388364,175421667,856463275,341279366}
    };
    int main(){
    	while (~scanf("%d%d",&n,&m))
    		printf("%d
    ",ans[n][m]);
    	return 0;
    }
    

      

    正常的代码

    #include <bits/stdc++.h>
    using namespace std;
    int read(){
    	int x=0;
    	char ch=getchar();
    	while (!isdigit(ch))
    		ch=getchar();
    	while (isdigit(ch))
    		x=(x<<1)+(x<<3)+ch-48,ch=getchar();
    	return x;
    }
    const int N=17,S=1<<16,mod=1e9+7;
    int n,m,dp[2][S],tot[N][N],ans[N][N];
    int gbit(int v,int d){
    	return (v>>(d-1))&1;
    }
    void Solve_tot(int n,int m){
    	memset(dp,0,sizeof dp);
    	int T0=1,T1=0;
    	dp[T1][(1<<m)-1]=1;
    	for (int i=1;i<=n;i++){
    		for (int j=1;j<=m;j++){
    			T0^=1,T1^=1;
    			memset(dp[T1],0,sizeof dp[T1]);
    			for (int s=0;s<(1<<m);s++){
    				int v=dp[T0][s];
    				if (!v)
    					continue;
    				dp[T1][s^(1<<(j-1))]=(dp[T1][s^(1<<(j-1))]+v)%mod;
    				if (j>1&&!gbit(s,j-1)&&gbit(s,j)){
    					int _s=s^(1<<(j-2));
    					dp[T1][_s]=(dp[T1][_s]+v)%mod;
    				}
    			}
    		}
    		tot[i][m]=dp[T1][(1<<m)-1];
    	}
    }
    void Get_tot(int n){
    	for (int m=1;m<=16;m++)
    		Solve_tot(n,m);
    }
    int GetV(int n,int s,int len){
    	int v=1;
    	for (int i=1,j;i<=n;i=j){
    		for (j=i;j<n&&!((s>>j)&1);j++);
    		j++;
    		v=1LL*v*tot[j-i][len]%mod;
    	}
    	return v;
    }
    int cnt_1(int v){
    	int ans=0;
    	while (v)
    		ans+=v&1,v>>=1;
    	return ans;
    }
    void Solve_ans(int n,int m){
    	int dp[N],v[N];
    	for (int s=0;s<(1<<n);s++){
    		if (!(s&1))
    			continue;
    		memset(dp,0,sizeof dp);
    		for (int i=1;i<=m;i++)
    			v[i]=GetV(n,s,i);
    		dp[0]=1;
    		for (int i=1;i<=m;i++)
    			for (int j=0;j<i;j++)
    				dp[i]=(-1LL*dp[j]*v[i-j]+dp[i])%mod;
    		int f=(cnt_1(s)&1)?-1:1;
    		for (int i=1;i<=m;i++)
    			ans[n][i]=(ans[n][i]+f*dp[i])%mod;
    	}
    }
    void Get_ans(int m){
    	memset(ans,0,sizeof ans);
    	for (int n=1;n<=16;n++)
    		Solve_ans(n,m);
    	for (int i=1;i<=m;i++)
    		for (int j=1;j<=m;j++)
    			ans[i][j]=(ans[i][j]+mod)%mod;
    }
    int main(){
    	Get_tot(16);
    	Get_ans(16);
    	while (~scanf("%d%d",&n,&m))
    		printf("%d
    ",ans[n][m]);
    	return 0;
    }
    

      

  • 相关阅读:
    一、保护模式
    什么是docker swarm configs?及其在service中的使用?
    k8s将dockershim移除之后,如何继续使用docker?
    如何通过secrets保存敏感的数据?并且在service中使用?
    docker commit其实是干了这些事情?
    使用secret时,如何指定secret在容器中的名字?
    Docker, containerd, CRIO and runc之间的区别?
    如何在k8s工作节点上,查看容器对应的pod的名字?
    如何锁定swarm集群?
    docker swarm集群中,你必须知道的几种网络流量类型?
  • 原文地址:https://www.cnblogs.com/zhouzhendong/p/51Nod1518.html
Copyright © 2020-2023  润新知