hi, Everybody. Firstly Thanks for repyl. i wonder, is there any solution for nested loop in ANN. can you explain with a example? My project is about back propagation neural network this project. to run fast of this project, when i can change my class, will program run fast ?

hi, i wrote a code and it has 4 layers and difference count neuron in each layer, and also it has 5 input,output. i have read so many codes of neural network and they have run very fast but mine is very slow. finally, when i use 400 paterns for input and 100000 iterations ,i wait so much. can we solve this problem for me ? i wait your recommend. Best this code : Main Blocak

double err=10, mmse ;
int ss = data.GetLength(0);
snetwork(0);
initializes();
mmse = 0;
double mser=0.0;
for (i = 0; i <iterasyon; i++)

        {


          mser = 0.0;
          for (int xx = 0; xx <data.GetLength(0); xx++)
          { 
   

              ffw(xx);
              backpropagate(xx);
              mser += mse(nb\_cikis, xx);
          }
       
          mmse = mser / data.GetLength(0);
          if (mmse < 0.00001) break;
      } ;

       
        ffw\_egitimout();

backprobagate error:

void backpropagate(int r)
{
double sum;
for (int j = 0; j < networks.layers[networks.layers.Length - 1].noron.Length; j++)
{
networks.layers[networks.layers.Length - 1].noron[j].hata = nb_cikis[r,j] - networks.layers[networks.layers.Length - 1].noron[j].output;
networks.layers[networks.layers.Length - 1].noron[j].delta = networks.layers[networks.layers.Length - 1].noron[j].output * (1 - networks.layers[networks.layers.Length - 1].noron[j].output) * networks.layers[networks.layers.Length - 1].noron[j].hata;//son katman deltası hesaplanıyor...
}

    	//	find delta for hidden layers	
for(int i=networks.layers.Length-2;i>0;i--)
{
	for(int j=0;j<networks.layers\[i\].noron.Length;j++)
    {
		sum=0.0;
         for (int k = 0; k < networks.layers\[i\].noron\[j\].sbaglanti.Length; k++)
        {
		             
      sum+=  networks.layers\[laybul(networks.layers\[i\].noron\[j\].sbaglanti\[k\])\].noron\[nronbul(networks.layers\[i\].noron\[j\].sbaglanti\[k\])\].delta\*networks.layers\[laybul(networks.layers\[i\].noron\[j\].sbaglanti\[k\])\].noron\[nronbul(networks.layers\[i\].noron\[j\].sbaglanti\[k\])\].dentw\[k\];
			
		}
		networks.layers\[i\].noron\[j\].delta=networks.layers\[i\].noron\[j\].output\*(1-networks.layers\[i\].noron\[j\].output)\*sum;

//
}
}

        	//	apply momentum ( does n

:) i want to difference neuron in diffrence layers it means firs layer which has 4 neurons and second layer 7 neurons ...

Hi, Everybody i am writing ANN code with c# i wrote a class which has folowing structure; class neuron { //public dentrit[] dent = new dentrit[nron]; //public double[,] w = new double[lay,nron];// ilk katmanda w yok dikkat public double [] dentw = new double[nron]; public double bias,hata,delta ; public double[] input = new double[nron ]; public double output ; public int[] sbaglanti = new int[nron]; public int[] sbagk = new int[nron]; public int[] obaglanti = new int[nron]; public int numara; public int dents;//dentrit sayısı public int oid,sid; ~neuron() { } } class layer { public neuron[] noron = new neuron[nron]; public layer() { for (int index = 0; index < noron.Length; index++) { noron[index] = new neuron(); } } ~layer(){} } class net { public int ban, hen; public layer[] layers = new layer[lay]; public net() { for (int index = 0; index < layers.Length; index++) { layers[index] = new layer(); } } ~net(){} } net networks; in here, i want to change neuron count of layers and so how can i do it ? can you help me ? thanks

:) tanks you too, i used meumentum coefficient. May be you will begin study NN again? can you see your code ? and in there, can you write main block in your code? best.

Hi i am writing again. I want to test my code and is there any fucntion which are 5 inputs and 5 out? or square function example, 2 inputs x 2 outputs, 4 inputs x4 outputs. May be , you can recommend a function for my testing. in addition, i thank you so mouch for sharining your informations with me Best regards.

Hi, that is my actual daya set and i took my friend. Datas is measured patient at hospital. As a result Datas is Biological datas, My Friend clasified nntool in matlab. he used difference learnin method which is Levenberg-Marquart, i used gradient desent with momentum algortihm(gdm). so i hope i will classify with my soft. But test results is not good. 2.Yes I train each pair in input/output arrays in each one iteration

for i:0:i<100000;i++// iteration
for j:0:j<400;j++ // input/output array
{

calculate error;

  if error>0.000001
   { train network;
    update weigths;
   }
    else
    break;


 }

Firstly Thanks interest. my dataset is folowing is that:

data is 5x400 matrix
b_cikis is 5x400 matrix

[...
long data removed
...]

int iterasyon=100000,it=0;
while (it = 0.00001)//iteration start section
{
mmse = 0;
mae = 0;
for (int nf = 0; nf < 400; nf++)//dataset count*********
{
mmse += hesaplas(data, nf);
//mae += smean_abs_err;
trains(nf);

            }//nf

            mmse = 0.5 \* (mmse / 400 \* 5));
          

            if (it % 100 == 0)
            {
                yazs.WriteLine(mmse.ToString());
                yazs.Flush();
            }
            it++;

           

        }//it

modified on Wednesday, August 4, 2010 10:25 AM

Hi, Everbody. i have writen c# Artifical intelliget Neural Network. I have 500 count dataset which is normalized -1 and 1 and.i have used learning rate and momentum coefficient . i teaching my network with 400 count dataset. Network have 5 inputs and 5 outputs. my output is setuped as 00001/00010/00100/01000/10000 when my network teched with 10000 iterations and 400 datasets, my output have a problem. my network teached in last class of 10000 so when i tested my datasets , output only have been predicted as true class of 10000. as result my code can only learn last class and other classes can not predicted or poor predicected Example Test and Learning dataset 5 neuron input {{0.04418 ,0.18135 ,0.28143 ,0.34367 ,0.14938 ,}, {0.37418 ,0.35050 ,0.18184 ,0.07493 ,0.01855 ,}, {0.04540 ,0.17061 ,0.24506 ,0.32560 ,0.21333 ,}, {0.08494 ,0.25653 ,0.26922 ,0.26141 ,0.12790 ,}, {0.18135 ,0.36832 ,0.24799 ,0.15743 ,0.04491 ,}, {0.11887 ,0.29143 ,0.27581 ,0.22870 ,0.08518 ,}, {0.02758 ,0.26239 ,0.37247 ,0.29998 ,0.03759 ,}, {0.00781 ,0.15792 ,0.32145 ,0.41811 ,0.09470 ,}, {0.07811 ,0.31608 ,0.31901 ,0.24506 ,0.04174 ,}, {0.07664 ,0.32390 ,0.32829 ,0.21674 ,0.05443 ,}, .... 5 neuron output { {1 ,0 ,0 ,0 ,0 }, {1 ,0 ,0 ,0 ,0 }, {0 ,1 ,0 ,0 ,0 }, {0 ,1 ,0 ,0 ,0 }, {0 ,1 ,0 ,0 ,0 }, {0 ,0 ,1 ,0 ,0 }, {0 ,0 ,1 ,0 ,0 }, {0 ,0 ,0 ,1 ,0 }, {0 ,0 ,0 ,1 ,0 }, {0 ,0 ,0 ,0 ,1 }, {0 ,0 ,0 ,0 ,1 } ..... My trainin code blok is as folowing show

int iterasyon=100000,it=0;
while (it = 0.00001)//iteration start section
{
mmse = 0;
mae = 0;
for (int nf = 0; nf < 400; nf++)//dataset count*********
{
mmse += hesaplas(ndata, nf);
//mae += smean_abs_err;
trains(nf);

            }//nf

            mmse = 0.5 \* (mmse / 400 \* 5));
          

            if (it % 100 == 0)
            {
                yazs.WriteLine(mmse.ToString());
                yazs.Flush();
            }
            it++;

           

        }//it

modified on Tuesday, August 3, 2010 9:30 PM

hi ,i modified my code folowing that, but i get some error folowing code running ! double[,] a, b, c; a = new double[2000, 2000]; b=new double[2000,2000]; c=new double[2000,2000]; int s = 2000; Parallel.For(0, s, delegate(int i) { for (int j = 0; j < s; j++) { double v = 0; for (int k = 0; k < s; k++) { v += a[i, k] * b[k, j]; } c[i, j] = v; } }); bu my code isn't run :( Parallel.For(0, nrons, delegate(int k) { for (i = 0; i < nrons; i++) for (j = 0; j < nrons; j++) //G[i, j] = Math.Min(G[i, j], G[i, k] + G[k, j]); if (G[i, j]> (G[i, k] + G[k, j]) ) G[i, j] =(G[i, k] + G[k, j]) ; } ); why did i get error ?

i can't access parallel.for class in .net framework 3.5 sp1 and installed sp4 how can i do ?.

modified on Tuesday, May 4, 2010 4:32 AM

it is floyd marshall algorithm but it is run slowly how can i increase speed? i have quad pc and it use %25 cpu but i wait so much . for (int k = 0; k < 2000; k++) for (int i = 0; i < 2000; i++) for (int j = 0; j < 2000; j++) G[i, j] = Math.Min(G[i, j], G[i, k] + G[k, j]);

thanks for reply. i take notice of a point, when i look at cpu using, i observer my code is using cpu %25 and i have quad pc so i can't use %75 of my cpu performance is there this problem solution ? not: orneksay=dataset

hi, friends i wrote a code in c#.net and it's ok. but i have a problem which is running slow i hope thare is some suggestions my code is folow static void calis(int rer, int dizin) { while (it < iterasyon && mmse <= 0.00001) { mmse = 0; mae = 0; for (int nf = 0; nf < orneksay; nf++) { mmse += hesaplas(ndata, nf); trains(nf); }//nf mmse = 0.5 * (mmse / (orneksay * nron)); if (it % 100 == 0) { yazs.WriteLine(mmse.ToString()); yazs.Flush(); } it++; }//it dosyas.Close(); yazss.WriteLine(rer.ToString() + " " + mmse.ToString() + " " + it.ToString()); yazss.Flush(); dosyass.Close(); } } i call calis function and i see cpu usage %25 mey it improve its performs ? can you help me? regads.

modified on Thursday, April 1, 2010 2:53 AM

Hi, Everbodys. i am writing a code that is about calculating shortest paths in feed forward neural network i have like the following code with neural structure in code: nron=6 and lay=6; class neuron { public int[] nextconnect = new int[nron]; public int[] preconnect = new int[nron]; public int number; public int dents;//dentrit count public int nextid,preid; // next and pre connections counts ~neuron() { } } class layer { public neuron[] noron = new neuron[nron]; public layer() { for (int index = 0; index < noron.Length; index++) { noron[index] = new neuron(); } } ~layer(){} } class net { public double lr;//learning rate; public int ban, hen; public layer[] layers = new layer[lay]; public net() { for (int index = 0; index < layers.Length; index++) { layers[index] = new layer(); } } ~net(){} } net network; i have prepared using feed forward connection structure but i didn't development a algorithm that has strong code can u help me as algorithm or code ? regards.

[Message Deleted]

thanks friends. i solved with array.copy(); regards

i understand but if arrays is referans value, how can i solve my problem?

i have a class and i assing this class value then when i sorted this value or class value, both sorted example code folowing : int[] ggg=new int[5]; int[] ggg1=new int[5]; ggg = networks.layers[i].noron[j].sbaglanti; ggg1 = ggg; Array.Sort(ggg); how can i solve this problem ? regards