Page 11 of 16
I'm in trouble
Posted: Mon Apr 02, 2007 10:41 pm
by raphaeljpb
Hi, this is my first thread. I can't recognize what's problem below. The code shows correct results at least $50, but I don't know the problem for bigger amounts.
Thanks in advance.
#include <stdio.h>
#include <stdlib.h>
#define SIZE 6001
int main() {
float quantidadeReal;
int quantidadeInteira = 0;
int tipos[] = { 1,2,4,10,20,40,100,200,400,1000, 2000 };
int i, j, numeroDeTipos = 11, c;
double * contagem;
while (1) {
scanf("%f", &quantidadeReal);
if (!quantidadeReal) break;
contagem = calloc(SIZE, sizeof(double));
quantidadeInteira = (int) (quantidadeReal * 100);
quantidadeInteira /= 5;
contagem[0] = 1.0;
for(i = 0; i < numeroDeTipos; i++) {
c = tipos;
for (j = c ; j <= quantidadeInteira; j++) {
contagem[j] += contagem[j-c];
}
}
printf("%7.2f %.0lf\n", quantidadeReal, contagem[quantidadeInteira]);
free(contagem);
}
return 0;
}
WA: Any test cases
Posted: Thu Jun 07, 2007 6:57 am
by gradientcurl
so far, never get to see any posted test cases. Here is a test output which I'm sure I've got it wrong somewhere:
Code: Select all
300.00 181490736388615
299.95 181000196059736
299.90 180510857366624
299.85 180510857366624
299.80 180510857366624
299.75 180022694940816
299.70 179535708782312
299.65 179535708782312
299.60 179535708782312
299.55 179049898891112
299.50 179049898891112
299.45 178565265267216
299.40 178081807910624
299.35 178081807910624
299.30 178081807910624
299.25 177599526821336
299.20 177118397097689
299.15 177118397097689
299.10 177118397097689
below is my code
Posted: Thu Jun 07, 2007 12:38 pm
by Jan
Check the set.
Set:
Hope it helps.
tks jan
Posted: Thu Jun 07, 2007 1:25 pm
by gradientcurl
tks jan. didnt take the double to integer conversion seriously.
help 147WA... i cant find the mistake
Posted: Thu Jun 07, 2007 4:16 pm
by shankar
i dont no whether my algorithm is right. for higher numbers it goes wrong i guess. so can some one helpme.
thanks
Code: Select all
#include<iostream>
#include<cstdio>
using namespace std;
unsigned long long a[6001][11]={0};int limit;
void dollar()
{int i,j,d[11]={1,2,4,10,20,40,100,200,400,1000,2000};
for(i=0;i<11;a[0][i]=a[1][i]=1,++i);
for(i=2;i<=6000;i++)
{ a[i][0]=1;
for(j=1;j<11;j++)
{
if(i<d[j])
{
a[i][j]=a[i][j-1];
}
else
{unsigned long long l=0;
while(l<=i)
{
a[i][j]+=a[i-l][j-1];l+=d[j];
}
}
}
}
}
int main()
{float g;int f,i,p,w;
limit=1;
for(i=0;i<6001;i++)
for(f=0;f<11;a[i][f++]=0);
dollar();
while(1)
{ scanf("%f",&g);
f=(int)(g/0.05);
printf("%d ",g);
if(f==0)
return 0;
cout<<a[f][10]<<endl;
// printf("%17ll\n",a[f][0]);
}
}
Posted: Sat Jun 09, 2007 1:31 am
by Jan
Check the set.
Set:Code: Select all
0.70 24
0.90 39
0.11 0
0.12 0
0.13 0
0.14 0
And your presentation is wrong, too.
Posted: Thu Jun 28, 2007 1:49 pm
by Itachi-Forsaken
I have a question regarding the output
Since the output is tabbed, I assume you have to use printf (for C/C++)
So what is the symbol after % for long long? Like 'd' is for integer
Also is there a way to properly convert a float to an integer? Or do you have to use a string and analyze it..
Thanks in advance
Posted: Thu Jun 28, 2007 1:59 pm
by Jan
'%lld' for long long.
Posted: Mon Aug 06, 2007 4:51 pm
by andysoft
Hello!
I am solving this prob and my algo works ok (at least i think so), but I cannot output large numbers. I used and use
unsigned long long type, and
printf("%lld\n",ans), but values for 300.00 input are either negative (-1600..00) or too small (like 2500000). Now I threw out the
printf and used
cout. I hope you may help me.
Code: Select all
#include <stdio.h>
#include <math.h>
#include <iostream>
int main(int argc, char* argv[]) {
unsigned long long a[6001];
int d[11] = {1,2,4,10,20,40,100,200,400,1000,2000};
float x;
int i,j,v,c;
for (i=1; i<=v; i++)
a[i] = 0;
a[0] = (unsigned long long)1;
for (i=0; i<11; i++) {
c = d[i];
for (j=c; j<=6000; j++)
a[j] += a[j-c];
}
while (1) {
scanf("%f",&x);
if (x==0.0) break;
v = (int)floor (x*20 + 0.1);
printf ("%6.2f",x);
std::cout.width(17);
std::cout << a[v] << std::endl;
}
return 0;
}
Debugger watches prove that my algo's right. Ans for 300.00 in watch is 181490736388615, but
how to output it??
Posted: Mon Aug 06, 2007 5:09 pm
by andysoft
Hi again!
1) Well, I re-wrote it in Pascal in just 2 minutes and got AC from the first time, BUT I still want to know the answer to my question about C/C++ output.
2) My prog ran during 0.346 seconds and I am 1383 / 1763 in the ranking. I would like to know how can I improve my code, if its structure is like this:
Code: Select all
1. Precompute all values from 1 to 6000 with usual for-loop: for each coin cost C do a[i] += a[i-C], a - 1..60000.
2. Read the input real number, multiply it by 25 and write the needed value from array a.
3. Types are: integer (int) everywhere except array a. It is int64 (long long)
Posted: Tue Aug 07, 2007 3:16 am
by emotional blind
All coin coin value is multiple of 5c
so just divide all coins by 5
and coins will be {2000c,1000c,400c,200c,100c,40c,20c,10c,4c,2c,1c}
Also divide your input by 5c
so you will require array of size 6000 only
thanks
Posted: Tue Aug 07, 2007 9:50 am
by andysoft
Yes, I have made it: divided everything needed by five.
Posted: Thu Aug 09, 2007 6:33 pm
by Kallol
where am I wrong ??
here is my code :
I checked almost all the input I got , I am getting WA . Is my output style wrong ??
Posted: Fri Aug 10, 2007 3:32 am
by mmonish
>>Kallol
Try this case
Try to avoid double calculation.It gives u WA.
Hope it helps.
Posted: Tue Aug 14, 2007 8:57 pm
by Kallol
I am really very much astonished with the behaviour of double in C.
double x = 5.10
x*=100;
printf("%0.lf",x);
and u will get 5.09 !!!
this is just horrible ...I thought , this floating point error is not there in double precision which was in the float data type. But it exists even in Long double .
Neway thanks for helping me with that tricky input. How could u guess and generate that input anyway ?? double gave me correcty answer in almost every othe cases
