What are the application of defining the Classes in Neural Network ?

Hi All

I was reading a comment in Neural Network to learn this topic more to develop my code

one thing I yet have not understood is that in the code we define classes in the code , what is that and why is it used ?

here is the comment :

%A quick way to see if any of the variables or classes appear to be different from the others is to standardize the inputs to zero-mean/unit-variance and compare the rows and columns of W0

%STANDARDIZATION (To compare Linear Model coefficients)

      zinputs = zscore(inputs')';
      W0z     = targets/[ones(1,N); zinputs];
      minmaxW0z  = minmax(W0z);   % [ 50 2 ]
      minmaxW0zp = minmax(W0z');  % [ 61 2 ]
      whos
      figure
      subplot(2,1,1)
      hold on
      plot(1:50,minmaxW0z(:,1),'bo')
      plot(1:50,minmaxW0z(:,2),'ro')
      subplot(2,1,2)
      hold on
      plot(1:61,minmaxW0zp(:,1),'bo')
      plot(1:61,minmaxW0zp(:,2),'ro')
%When the inputs are standardized, I see no significant differences between the weights associated with different classes or different variables

[Greg Heath]

What is this supposed to mean:

minmaxW0z = minmax(W0z); % [ 50 2 ]

minmaxW0zp = minmax(W0z'); % [ 61 2 ]

 ?

[farzad]

Dear Professor Heath

well Maybe I just did not use the right code , my major question is on the code you have written me , I don't precisely understand every line of it , the mapminmax is supposed to mean to normalize the data of input and target between -1 and +1 so we could say if the data belong to the same class , this is just one thing I have read , I think you have explained the same thing here , but in this code that is familiar :

close all, clear all, clc, plt = 0
[ x, t ]  = simpleclass_dataset;
[ I   N ] = size(x)   % [ 2 1000 ]
[ O N ] = size(t)    % [ 4 1000 ]
trueclass = vec2ind(t);
class1 = find(trueclass==1);
class2 = find(trueclass==2);
class3 = find(trueclass==3);
class4 = find(trueclass==4);
N1 = length(class1)    % 243
N2 = length(class2)    % 247
N3 = length(class3)    % 233
N4 = length(class4)    % 277
x1 =  x(:,class1);
x2 =  x(:,class2);
x3 =  x(:,class3);
x4 =  x(:,class4);
plt = plt + 1
hold on
plot(x1(1,:),x1(2,:),'ko')
plot(x2(1,:),x2(2,:),'bo')
plot(x3(1,:),x3(2,:),'ro')
plot(x4(1,:),x4(2,:),'go')
 Hub = -1+ceil( (0.7*N*O-O)/(I+O+1))  % 399
Hmax = 40   % Hmax << Hub 
dH    = 4   % Design ~10 candidate nets
Hmin  = 2   % I know 0 and 1 are too small
 rng(0)   % Allows duplicating the rsults
j=0
for h=Hmin:dH:Hmax
    j = j+1
    net            = newpr(x,t,h);
    [ net tr y  ]  = train( net, x, t );
    assignedclass  =  vec2ind(y);
    err = assignedclass~=trueclass;
    Nerr           = sum(err);
    PctErr(j,1)    = 100*Nerr/N;
end
h = (Hmin:dH:Hmax)';    
PctErr = PctErr;
results = [ h PctErr ]

here i just don't get it , why 4 classes ? cause my target contains 4 outputs ? and all my data are collected as class 2 , is it ok ?

Answer 1

[Greg Heath]

I don't think that you understand that MATLAB differentiates 4 basic types of nets

 Please read the command line documentation for

 help nndatasets

and

 doc nndatasets

 What type of data is simpleclass_dataset?

 How does it differ from simplefit_dataset?

 What type of net is fitnet?

 How does it differ from patternnet?

 Then read the online documentation at mathworks.com

[farzad]

Thank You Dear Professor Heath

that is right , the simpleclass has two columns of data , so for more than one input we have to classify them , and I also have searched the mathworks, but still knowing them , does not precisely tell me what is happening when you use vec2ind , my question was , for my data , all the data go to class 2 , does it make a problem ?

I also do not get this part :

 assignedclass  =  vec2ind(y);