OJ Board

I got WA several times..
Please suppose some test data

what should the test case be?
2
2 2000000
72 72

yes I agree that you shoule implent O(v^3) algo

since the number of edges is very big...

mine program is O(m*(n*d)^3)

using relabel-to-front

without bsearch

WA I'm thinking about the accuracy of my algo

If you know the actually relabel-to-front algo please send me a message about websit or ...

please suppose some useful I/O

I've tried to generate some but no use

How the O(n^3) method work?

please tell me

I'm running crazy

How was the correct solution's time complexity?

now I'm using is O((n^2)*m)

I did it because I thinked it should be close to time limit

Three pigs is not claimed that they should use different ways

I used a n^4 dp but got tle
I think it must has to be speed up
but i can't

here's my code:

#include<iostream>
#include<cstdio>
#include<cstring>
using namespace std;
char dp[301][301][3];
int dp1[301][301][3];
int s[3][3]={{1,1,0},{2,1,0},{0,2,2}};
char asd[301];
char zxc[301];
int qq,tt;
inline ...

when you make every three points set you have to check if they are one the same line(lines in input)

some words are reduplicate at the begining
these part shouldn't be done twice

tree can help you to solve

I don't know how to solve ths problem.
Just found a stubid greedy method
could someone who got ac explain it?

Ohh...yes

my guess is right
there's some test data for someone who's solving this problem

It is so strange.. :-?

Input consists of several test-cases. Each test-case starts with a line containing three numbers


Sample Input
4 5
1 2
2 3
3 4
4 1
1 3
0 0

I can't see the picture so I assumed the first line is n,m
which n represent number of cities
and m is number of roads

I used BFS ...

Maybe you can post your code

hmm...

WA shouldn't be the problem

I think you may have to check your algo

if the algo is correct there is hard to exist a data to result in WA