11401 - Triangle Counting

[birlax]

geting TL please help i am storing all the solution in array a[1000000]
before reading any input and @ run time i directly print

a[input] which is the correct output what to do

[jurajz]

I used same method as you and have AC in 0.030. My precalculation is in O(n). If you have correct answer for n, then you can get correct answer for n+1 in O(1). Try to find it. If you used big int, it is not needed, int64 (in C/C++ long long) is enough.

One warning to this problem: The input is terminated with n less then 3, not with 0 (in lot of problems, input is terminated with 0). And surely it is not 0, I have one WA because of that

[andmej]

If you have correct answer for n, then you can get correct answer for n+1 in O(1). Try to find it.

Thanks for the tip. I was trying to brute-force every possible solution but it was too slow. I will think in a Dynamic Programming solution now.

Also, if a moderator reads this post I would suggest him to do two things:
1) Change the title of this thread to "11401 - Triangle counting" instead of "11401" alone.
2) Open a new subforum for volume CXIV. Clearly this problem is not from Volume CXIII.

I hope my comment is not an offense for anyone.

[arif_pasha]

Can anyone confirm the result for the following set:

Code: Select all

9
10
11
12
50
100
1000
10000
100000

I Get:

Code: Select all

34
50
70
95
9500
79625
82958750
83295837500
51903307670168

Thanks

[emotional blind]

arif_pasha,
All but last one is correct

input

Code: Select all

100000

output

Code: Select all

83329583375000

[arif_pasha]

Thanks.

For temporary variable i was using int instead of long long. that was overflowing.

[Brainless]

Hi,

I think that my algorithm is correct, but I still get WA.

Here is my source code :

Code: Select all

#include <iostream>
#include <map>
using namespace std;

typedef unsigned long ul;
typedef unsigned long long ull;
typedef long double ld;

int main()
{
	map<ul, ld> S;
	//ld S[1000001];
	ld R;
	S[3] = 0;
	S[4] = 1;
	ul b;
	for(ul n=5; n<=1000000; ++n)
	{
		b = n/2+1;
		R = n*(n-1)-(n+1)*(n-b)-b*(b-1);
		S[n] = S[n-1] + R;
	}
	ul n;
	cin >> n;
	ull res;
	while(n>=3)
	{
		if(n==0)
		{
			cout << 0 << endl;
		}
		else
		{
			res = (ull)S[n];
			cout << res << endl;
		}
		cin >> n;
	}
	
	return 0;
}

I need some help ! Thanks ...

[Brainless]

For n = 1 000 000, the output I get is :

2030481550920432

[Brainless]

Ok,ok ...

Now I got AC. The problem was on overflow of the value "b", which is not a double, but I unsigned long long ...

I submitted totally 8 times ...

[arif_pasha]

You can remove your code from the post. (don't spoil the fun to solve a problem )

[Obaida]

I don't know why this is WA!!!!!!!

Code: Select all

mistake fixed

[Jan]

Check the following line...

Code: Select all

f = i*(i-1)-(i+1)*(i-b)-b*(b-1);

Since i, b both are normal integers so, overflow will occur if i/b is too big. So, declare i and b as 64 bit integer.

[Obaida]

Thanks vai got ACC.

[Shahidul.CSE]

this is an interesting problem !!

[brianfry713]

My AC code runs each test case in constant time with O(max_n) pre-processing.
The output will fit in a long long unsigned.