矩阵快速幂模板（持续更新）

经过几次比较，目前总结封装矩阵的模板

加上优化，用这个代码可以0MS过杭电上的一道题（以后可能还有更快的）

typedef long long ll;
const int MAXN = 10;
long long mod;
struct Matrix{
    long long mat[MAXN][MAXN];
    void Zero(){
        memset(mat, 0, sizeof(mat));
    }
    void Unit(){
        memset(mat, 0, sizeof(mat));
        for (int i = 0; i < MAXN; i++)
            mat[i][i] = 1;
    }
    void output(){
        for (int i = 0; i < MAXN; i++){
            for (int j = 0; j < MAXN; j++){
                printf("%d ", mat[i][j]);
            }
            printf("\n");
        }
    }
};

Matrix operator*(Matrix &a, Matrix &b){
    Matrix tmp;
    tmp.Zero();///初始化
    for (int k = 0; k < MAXN; k++){
        for (int i = 0; i < MAXN; i++){
            if (!a.mat[i][k])
                continue;
            for (int j = 0; j < MAXN; j++){
                tmp.mat[i][j] += a.mat[i][k] * b.mat[k][j] % mod;
                if ( tmp.mat[i][j] >= mod)
                    tmp.mat[i][j] -= mod;
            }

        }
    }
    return tmp;
}
Matrix operator ^(Matrix a, int k){
    Matrix tmp;///单位矩阵
    tmp.Unit();
    if(k<=1)
        return a;
    for (; k; k >>= 1){
        if (k & 1)
            tmp = tmp * a;
        a = a * a;
    }
    return tmp;
}

模板应用：

传送门：http://acm.split.hdu.edu.cn/showproblem.php?pid=4990

代码：（0MS）

#include<iostream>
#include<sstream>
#include<cstdio>
#include<cstring>
#include<string>
#include<algorithm>
#include<cmath>
#include<ctime>
#include<list>
#include<vector>
#include<set>
#include<map>
#include<stack>
#include<queue>
#pragma GCC optimize("02")
#define cl(a,b) memset(a,b,sizeof(a))
#define in freopen("F://1.txt","r",stdin)
#define out freopen("F://2.txt","w",stdout)
using namespace std;
typedef unsigned long long llu;
typedef long long ll;
const ll inf=(ll)1<<60;
const int maxn=1e4+7;
const int MAXN = 10;
ll mod;
struct Matrix{
    ll mat[MAXN][MAXN];
    void Zero(){
        //memset(mat, 0, sizeof(mat));
        cl(mat,0);
    }
    void Unit(){
        //memset(mat, 0, sizeof(mat));
        cl(mat,0);
        for (int i = 0; i < MAXN; i++)
            mat[i][i] = 1;
    }
    void output(){
        for (int i = 0; i < MAXN; i++){
            for (int j = 0; j < MAXN; j++){
                printf("%lld ", mat[i][j]);
            }
            printf("\n");
        }
    }
};

Matrix operator*(Matrix &a, Matrix &b){
    Matrix tmp;
    tmp.Zero();///初始化
    for (int k = 0; k < MAXN; k++){
        for (int i = 0; i < MAXN; i++){
            if (!a.mat[i][k])
                continue;
            for (int j = 0; j < MAXN; j++){
                tmp.mat[i][j] += a.mat[i][k] * b.mat[k][j] % mod;
                if ( tmp.mat[i][j] >= mod)
                    tmp.mat[i][j] -= mod;
            }

        }
    }
    return tmp;
}
Matrix operator ^(Matrix a, int k){
    Matrix tmp;///单位矩阵
    tmp.Unit();
    if(k<=1)
        return a;
    for (; k; k >>= 1){
        if (k & 1)
            tmp = tmp * a;
        a = a * a;
    }
    return tmp;
}
int main(){
    Matrix mt;
    mt.Zero();
    mt.mat[0][0] = 1;
    mt.mat[0][1] = 1;
    mt.mat[1][0] = 2;
    mt.mat[2][0] = 1;
    mt.mat[2][2] = 1;
    Matrix mt2;
    mt2.Zero();
    mt2.mat[0][0] = 2;
    mt2.mat[0][1] = 1;
    mt2.mat[0][2] = 1;
    ll n,m;
    while(~scanf("%lld%lld", &n, &m)){
        mod=m;
        if(n==1){
            printf("%lld\n", 1%m);
            continue;
        }
        else if(n==2){
            printf("%lld\n", 2%m);
            continue;
        }
        Matrix m3;
        m3 = mt ^ (n-2);
        Matrix ans = mt2 * m3;
        printf("%lld\n", ans.mat[0][0]);
    }
    return 0;
}