HDU 1043 & POJ 1077 Eight(康托展开+BFS | IDA*)
| Time Limit: 1000MS | Memory Limit: 65536K | |||
| Total Submissions: 30176 | Accepted: 13119 | Special Judge | ||
Description
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 x
where the only legal operation is to exchange 'x' with one of the tiles with which it shares an edge. As an example, the following sequence of moves solves a slightly scrambled puzzle:
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
5 6 7 8 5 6 7 8 5 6 7 8 5 6 7 8
9 x 10 12 9 10 x 12 9 10 11 12 9 10 11 12
13 14 11 15 13 14 11 15 13 14 x 15 13 14 15 x
r-> d-> r->
The letters in the previous row indicate which neighbor of the 'x' tile is swapped with the 'x' tile at each step; legal values are 'r','l','u' and 'd', for right, left, up, and down, respectively.
Not all puzzles can be solved; in 1870, a man named Sam Loyd was famous for distributing an unsolvable version of the puzzle, and
frustrating many people. In fact, all you have to do to make a regular puzzle into an unsolvable one is to swap two tiles (not counting the missing 'x' tile, of course).
In this problem, you will write a program for solving the less well-known 8-puzzle, composed of tiles on a three by three
arrangement.
Input
1 2 3 x 4 6 7 5 8
is described by this list:
1 2 3 x 4 6 7 5 8
Output
Sample Input
2 3 4 1 5 x 7 6 8
Sample Output
ullddrurdllurdruldr
Source
题目链接:POJ 1077
主要就是用康托展开来映射判重的问题,info::val就是康托展开hash值,info::step就是积累的状态。另外感觉这题剧毒,自己本来用int vis[]和char his[]想最后回溯记录答案从而代替速度比较慢的string,结果居然超时……TLE一晚上,要不是看了大牛的博客估计要一直T在这个坑点上。还有不知道为什么string的加号重载在C++编译器里会CE,换G++才过。相比单组输入的POJ,多组输入的HDU就友好多了,打个表就可以水过了,双广、A*神马的写起来麻烦就先不写了……
什么是康托展开?——康托展开介绍文章
POJ代码:
#include<iostream>
#include<algorithm>
#include<cstdlib>
#include<sstream>
#include<cstring>
#include<bitset>
#include<cstdio>
#include<string>
#include<deque>
#include<stack>
#include<cmath>
#include<queue>
#include<set>
#include<map>
using namespace std;
#define INF 0x3f3f3f3f
#define CLR(x,y) memset(x,y,sizeof(x))
#define LC(x) (x<<1)
#define RC(x) ((x<<1)+1)
#define MID(x,y) ((x+y)>>1)
typedef pair<int, int> pii;
typedef long long LL;
const double PI = acos(-1.0);
const int N = 362880 + 20;
int fact[10] = {1, 1, 2, 6, 24, 120, 720, 5040, 40320, 362880};
int direct[4][2] = {{ -1, 0}, {1, 0}, {0, -1}, {0, 1}};
char MOVE[5] = "udlr";
struct info
{
int s[9];
int indx;
string step;
int val;
};
info S, E;
int T;
int vis[N];
string ans;
int calcantor(int s[])
{
int r = 0;
for (int i = 0; i < 9; ++i)
{
int k = 0;
for (int j = i + 1; j < 9; ++j)
{
if (s[j] < s[i])
++k;
}
r = r + k * fact[8 - i];
}
return r;
}
bool check(const int &x, const int &y)
{
return (x >= 0 && x < 3 && y >= 0 && y < 3);
}
bool bfs()
{
CLR(vis, 0);
queue<info>Q;
Q.push(S);
vis[S.val] = 1;
info now, v;
while (!Q.empty())
{
now = Q.front();
Q.pop();
if (now.val == T)
{
ans = now.step;
return true;
}
for (int i = 0; i < 4; ++i)
{
int x = now.indx / 3;
int y = now.indx % 3;
x += direct[i][0];
y += direct[i][1];
if (check(x, y))
{
v = now;
v.indx = x * 3 + y;
v.s[now.indx] = v.s[v.indx];
v.s[v.indx] = 0;
v.val = calcantor(v.s);
if (!vis[v.val])
{
vis[v.val] = 1;
v.step = now.step + MOVE[i];
if (v.val == T)
{
ans = v.step;
return true;
}
Q.push(v);
}
}
}
}
return false;
}
int main(void)
{
char temp;
int i;
T = 46233;
while (cin >> temp)
{
if (temp == 'x')
{
S.s[0] = 0;
S.indx = 0;
}
else
S.s[0] = temp - '0';
for (i = 1; i < 9; ++i)
{
cin >> temp;
if (temp == 'x')
{
S.s[i] = 0;
S.indx = i;
}
else
S.s[i] = temp - '0';
}
S.val = calcantor(S.s);
puts(!bfs() ? "unsolvable" : ans.c_str());
}
return 0;
}
HDU代码:
#include<iostream>
#include<algorithm>
#include<cstdlib>
#include<sstream>
#include<cstring>
#include<bitset>
#include<cstdio>
#include<string>
#include<deque>
#include<stack>
#include<cmath>
#include<queue>
#include<set>
#include<map>
using namespace std;
#define INF 0x3f3f3f3f
#define CLR(x,y) memset(x,y,sizeof(x))
#define LC(x) (x<<1)
#define RC(x) ((x<<1)+1)
#define MID(x,y) ((x+y)>>1)
typedef pair<int, int> pii;
typedef long long LL;
const double PI = acos(-1.0);
const int N = 362880 + 10;
int fact[10] = {1, 1, 2, 6, 24, 120, 720, 5040, 40320, 362880};
struct info
{
int s[9];
int indx;
string path;
int val;
void cal()
{
val = 0;
for (int i = 0; i < 9; ++i)
{
int k = 0;
for (int j = i + 1; j < 9; ++j)
{
if (s[j] < s[i])
++k;
}
val = val + k * fact[8 - i];
}
}
};
info init = {{1, 2, 3, 4, 5, 6, 7, 8, 9}, 8, "", 0}, S;
int direct[4][2] = {{ -1, 0}, {1, 0}, {0, -1}, {0, 1}}; //ÏÂÉÏÓÒ×ó
char MOV[5] = "durl";
string ans[N];
int vis[N];
void bfs()
{
CLR(vis, 0);
queue<info>Q;
info now, v;
Q.push(init);
vis[init.val] = 1;
ans[init.val] = init.path;
while (!Q.empty())
{
now = Q.front();
Q.pop();
for (int i = 0; i < 4; ++i) //ÏÂÉÏÓÒ×ó
{
int x = now.indx / 3;
int y = now.indx % 3;
x += direct[i][0];
y += direct[i][1];
if (x >= 0 && x < 3 && y >= 0 && y < 3)
{
v = now;
v.indx = x * 3 + y;
v.path = MOV[i] + now.path;
swap(v.s[v.indx], v.s[now.indx]);
v.cal();
if (!vis[v.val])
{
vis[v.val] = 1;
ans[v.val] = v.path;
Q.push(v);
}
}
}
}
}
int main(void)
{
char temp[5], i;
bfs();
while (~scanf("%s", temp))
{
S.path = "";
if (temp[0] == 'x')
{
S.s[0] = 9;
S.indx = 0;
}
else
S.s[0] = temp[0] - '0';
for (i = 1; i < 9; ++i)
{
scanf("%s", temp);
if (temp[0] == 'x')
{
S.s[i] = 9;
S.indx = i;
}
else
S.s[i] = temp[0] - '0';
}
S.cal();
puts(!vis[S.val] ? "unsolvable" : ans[S.val].c_str());
}
return 0;
}
最近学了下IDA*,发现速度贼快,比哈希的不知道高到哪里去了,自己整理了一下一般写法,感觉还是比较模版的,可以参考上一篇IDA*的伪代码,IDA*快到如果用BFS要打表的HDU上的数据可以直接在线搜索过了,确实比较快,这里需要加一个防止走回路的剪枝(有效减少无用搜索),否则可能会超时,当然一开始要判断一下是否输入序列是无解的,可以暂时把x忽略掉,然后算7个数的逆序数,逆序数偶数才有解,奇数直接输出unsolvable,具体原理可以百度一下
代码:
#include <stdio.h>
#include <iostream>
#include <algorithm>
#include <cstdlib>
#include <sstream>
#include <numeric>
#include <cstring>
#include <bitset>
#include <string>
#include <deque>
#include <stack>
#include <cmath>
#include <queue>
#include <set>
#include <map>
using namespace std;
#define INF 0x3f3f3f3f
#define LC(x) (x<<1)
#define RC(x) ((x<<1)+1)
#define MID(x,y) ((x+y)>>1)
#define CLR(arr,val) memset(arr,val,sizeof(arr))
#define FAST_IO ios::sync_with_stdio(false);cin.tie(0);
typedef pair<int, int> pii;
typedef long long LL;
const double PI = acos(-1.0);
const int N = 9;
int arr[N], T[N] = {1, 2, 3, 4, 5, 6, 7, 8, 0};
int Maxdep, top, stdpos[N] = {8, 0, 1, 2, 3, 4, 5, 6, 7};
char Ops[1010]; int dis(int cur, int ori)
{
int x = cur / 3, y = cur % 3;
return abs(x - ori / 3) + abs(y - ori % 3);
}
int Need(int A[])
{
int ret = 0;
for (int i = 0; i < N; ++i)
if (A[i] != T[i])
ret += dis(i, stdpos[A[i]]);
return ret;
}
inline bool check(int x, int y)
{
return x >= 0 && x < 3 && y >= 0 && y < 3;
}
int dfs(int dep, int Arr[], char pre)
{
int need = Need(Arr);
if (need == 0)
return 1;
else if (need + dep > Maxdep)
return 0;
else
{
int x, y, idx, i;
int Temp[N];
for (int i = 0; i < N; ++i)
{
if (Arr[i] == 0)
{
idx = i;
break;
}
}
x = idx / 3, y = idx % 3;
if (pre != 'u' && check(x + 1, y))
{
for (i = 0; i < N; ++i)
Temp[i] = Arr[i];
swap(Temp[(x + 1) * 3 + y], Temp[idx]);
Ops[top++] = 'd';
if (dfs(dep + 1, Temp, 'd'))
return 1;
else
--top;
}
if (pre != 'l' && check(x, y + 1))
{
for (i = 0; i < N; ++i)
Temp[i] = Arr[i];
swap(Temp[x * 3 + y + 1], Temp[idx]);
Ops[top++] = 'r';
if (dfs(dep + 1, Temp, 'r'))
return 1;
else
--top;
}
if (pre != 'd' && check(x - 1, y))
{
for (i = 0; i < N; ++i)
Temp[i] = Arr[i];
swap(Temp[(x - 1) * 3 + y], Temp[idx]);
Ops[top++] = 'u';
if (dfs(dep + 1, Temp, 'u'))
return 1;
else
--top;
}
if (pre != 'r' && check(x, y - 1))
{
for (i = 0; i < N; ++i)
Temp[i] = Arr[i];
swap(Temp[x * 3 + y - 1], Temp[idx]);
Ops[top++] = 'l';
if (dfs(dep + 1, Temp, 'l'))
return 1;
else
--top;
}
}
return 0;
}
int main(void)
{
char n[3];
int i, j, x;
while (~scanf("%s", n))
{
if (n[0] == 'x')
x = 0;
else
x = n[0] - '0';
arr[0] = x;
for (i = 1; i < 9; ++i)
{
scanf("%s", n);
if (n[0] == 'x')
x = 0;
else
x = n[0] - '0';
arr[i] = x;
}
int inv = 0;
for (i = 0; i < N; ++i)
{
if (arr[i] == 0)
continue;
for (j = i + 1; j < N; ++j)
{
if (arr[j] == 0)
continue;
if (arr[i] > arr[j])
++inv;
}
}
if (inv & 1)
puts("unsolvable");
else
{
top = 0;
Maxdep = 1;
while (!dfs(0, arr, -1))
++Maxdep;
Ops[top] = '\0';
puts(Ops);
}
}
return 0;
}
HDU 1043 & POJ 1077 Eight(康托展开+BFS | IDA*)的更多相关文章
- Eight (HDU - 1043|POJ - 1077)(A* | 双向bfs+康拓展开)
The 15-puzzle has been around for over 100 years; even if you don't know it by that name, you've see ...
- HDU 1043 & POJ 1077 Eight(康托展开+BFS+预处理)
Eight Time Limit: 1000MS Memory Limit: 65536K Total Submissions: 30176 Accepted: 13119 Special ...
- HDU 1430 魔板(康托展开+BFS+预处理)
魔板 Time Limit: 10000/5000 MS (Java/Others) Memory Limit: 65536/32768 K (Java/Others) Total Submis ...
- Eight hdu 1043 poj 1077
Description The 15-puzzle has been around for over 100 years; even if you don't know it by that name ...
- HDU 1043 Eight (A* + HASH + 康托展开)
Eight Time Limit: 10000/5000 MS (Java/Others) Memory Limit: 65536/32768 K (Java/Others) Total Sub ...
- HDU 3567 Eight II 打表,康托展开,bfs,g++提交可过c++不可过 难度:3
http://acm.hdu.edu.cn/showproblem.php?pid=3567 相比Eight,似乎只是把目标状态由确定的改成不确定的,但是康托展开+曼哈顿为h值的A*和IDA*都不过, ...
- hdu 1043 Eight (八数码问题)【BFS】+【康拓展开】
<题目链接> 题目大意:给出一个3×3的矩阵(包含1-8数字和一个字母x),经过一些移动格子上的数后得到连续的1-8,最后一格是x,要求最小移动步数. 解题分析:本题用BFS来寻找路径,为 ...
- [HDOJ1043]Eight(康托展开 BFS 打表)
题目链接:http://acm.hdu.edu.cn/showproblem.php?pid=1043 八数码问题,因为固定了位置所以以目标位置开始搜索,把所有情况(相当于一个排列)都记录下来,用康托 ...
- hdu1043Eight (经典的八数码)(康托展开+BFS)
建议先学会用康托展开:http://blog.csdn.net/u010372095/article/details/9904497 Problem Description The 15-puzzle ...
随机推荐
- 用fast rcnn绘制loss曲线遇到的问题
运行fast rcnn的train,会进入ipython,要先exit退出才能继续运行程序 绘制图像时,用了命令: ./tools/train_net.py --gpu 0 --solver mode ...
- 复习C++_基础、函数、数组、字符串
程序的开发过程 程序 源程序:用源语言写的,有待翻译的程序: 目标程序:源程序通过翻译程序加工以后生成的机器语言程序: 可执行程序:连接目标程序以及库中的某些文件,生成的一个可执行文件,例如Windo ...
- sublime package control以及常用插件
一.package Control安装 1.sublime 3 import urllib.request,os; pf = 'Package Control.sublime-package'; ip ...
- CUDA:Supercomputing for the Masses (用于大量数据的超级计算)-第四节
了解和使用共享内存(1) Rob Farber 是西北太平洋国家实验室(Pacific Northwest National Laboratory)的高级科研人员.他在多个国家级的实验室进行大型并行运 ...
- java arraylist int[] 转换
double no=Double.valueOf("str");int num4=(int)no;double no1=Double.parseDouble("str&q ...
- axios获取后端数据
axios向后端请求数据时,一直获取不到数据, 后来改成这样写获取到了数据 不是一个this,有人说用箭头函数就可以了.
- 统计学基于SPSS贾俊平 授课笔记 发布作业 spss19cn 软件下载地址及破解包spss19_10039 下载地址
spss19cn软件下载地址及破解包spss19_10039 软件包下载地址一 http://www.33lc.com/soft/41991.html 软件包下载地址二 http://dl.pconl ...
- EventUtil处理js兼容性问题
var EventUtil={ addHandler:function(element,type,handler){ //添加事件 if(element.addEventListener){ elem ...
- webpack的配置处理
1.webpack对脚本的处理 1.Js用什么loader加载? 1>webpack 本身就支持js的加载, 2>通过babel-loader ES2015 加载js,再用 babel-p ...
- JZOJ 4743. 积木
Description Input Output Sample Input 38 7 63 9 41 10 5 Sample Output 18