标签:tac algo name tmp int bit print stack clu
\(n \le 20\) 考虑状态压缩\(dp\)。
考虑状态,\(dp(i,j,O)\)表示从\(i\)到\(j\)经过点集\(O\)的路径有多少。
\(dp(i,j,O \bigcup i)=\Sigma dp(i,p,O)\),\(j-p\)有一条边。
考虑内存,我们可以认定状态压缩串中\(lowbit(x)\)位是一条路的起点,这样我们直接省掉一维。空间限制卡进去了。
考虑答案怎么统计,就是\((\Sigma (dp(i,j,O)+dp(j,i,O))-m) \div2\)
然后就是神仙一般的代码,orz Van
#include<iostream>
#include<cstring>
#include<algorithm>
#include<cstdio>
#include<queue>
#include<bitset>
#include<vector>
#include<map>
#include<ctime>
#include<cstdlib>
#include<set>
#include<bitset>
#include<stack>
#include<list>
#include<cmath>
using namespace std;
#define RP(t,a,b) for(register ll (t)=(a),edd_=(b);t<=edd_;++t)
#define DRP(t,a,b) for(register ll (t)=(a),edd_=(b);t>=edd_;--t)
#define ERP(t,a) for(int t=head[a];t;t=e[t].nx)
#define Max(a,b) ((a)<(b)?(b):(a))
#define Min(a,b) ((a)<(b)?(a):(b))
#define TMP template<class ccf>
#define lef L,R,l,mid,pos<<1
#define rgt L,R,mid+1,r,pos<<1|1
#define midd register int mid=(l+r)>>1
#define chek if(R<l||r<L)return
#define all 1,n,1
#define pushup(x) seg[(x)]=seg[(x)<<1]+seg[(x)<<1|1]
#define lowbit(x) ((x)&(-(x)))
typedef long long ll;
TMP inline ccf qr(ccf k){
char c=getchar();
ccf x=0;
int q=1;
while(c<48||c>57)
q=c==45?-1:q,c=getchar();
while(c>=48&&c<=57)
x=x*10+c-48,c=getchar();
if(q==-1)
x=-x;
return x;
}
const int maxn=20;
const ll mod=1e9+7;
bool e[maxn][maxn];
ll dp[maxn][1<<maxn];
ll ans;
int n,m;
int main(){
#ifndef ONLINE_JUDGE
freopen("hamilton.in","r",stdin);
freopen("hamilton.out","w",stdout);
#endif
n=qr(1);
m=qr(1);
const int maxw=(1<<n)-1;int t1,t2;
RP(t,1,m){
t1=qr(1)-1;
t2=qr(1)-1;
e[t1][t2]=1;
e[t2][t1]=1;
//dp[t1][(1<<t1)|(1<<t2)]=1;
//dp[t2][(1<<t1)|(1<<t2)]=1;
}
RP(t,0,n-1)
dp[t][(1<<t)]=1;//lowbit(x)是一个路径的起点
RP(k,0,maxw){
RP(t,0,n-1){//起点
RP(i,0,n-1){//目标点
if(e[t][i]){
if(lowbit(k)>(1<<i))
continue;
if((1<<i)&k){
if((1<<i)==lowbit(k))
ans+=dp[t][k];
}
else
dp[i][k|(1<<i)]+=dp[t][k];
}
}
}
}
printf("%lld\n",((ans-m)>>1));
return 0;
}
标签:tac algo name tmp int bit print stack clu
原文地址:https://www.cnblogs.com/winlere/p/10324560.html