MATLAB: How to get the transfer function equation for the implemented neural network

neural network

I have implemented neural network of two layers and hidden layer size is 3. for both layers activation function is purelin. When I calculate theoretically output of neural network it is not matching with output provided by neural network of matalab. Plz let me know where i am doing wrong. Thanks

Best Answer

You have to obtain xsettings and tsettings from x and t.

Then apply them to x2 and t2.

Hope this helps.

Thank you for formally accepting my answer

Greg

Related Solutions

MATLAB: Could you please help me in Artificial neural network – supervised learning

1. It is very seldom that you will need

   a. That many inputs
   b. More than 1 hidden layer
   c. Anywhere near that many hidden nodes.

2. Typically, if you transform your variables to zero-mean/unit-variance via ZSCORE or MAPSTD, the coefficients of a linear model will indicate which variables can probably be ignored because they are either weakly correlated to the target OR are highly correlated with other variables.

Alternatives are

 a. Add squares and/or cross-products to the linear (in coefficients) model
 b. Use functions STEPWISE and/or STEPWISEFIT

3. PLEASE

   a. Do not post commands that assign default values.
   b. Include results of applying your code to an accessible data set so 
      that we know we are on the same page.
   c. Instead of posting your huge dataset, just pick one of the MATLAB  
      example sets
      help nndatasets
      doc  nndatasets

4. For the purpose of reproducibility, initialize the RNG before obtaining the random initial weights and random trn/val/tst data division.

5. I have posted many tutorials that emphasize minimizing the number of hidden nodes, H, to obtain better performance on non-training (validation, test and unseen) data.

6. Basically, you would like the number of unknown weights

 Nw = (I+1)*H+(H+1)*O

to be much less than the number of training equations

 Ntrneq = round(0.7*N*O)   % default approx.

A necessary condition is

 H <= Hub = floor((Ntrneq-O)/(I + O +1))

However H << Hub is preferable.

With I = 20, O = 1, N = 1000

 Ntrneq = Ntrn = 700
 Hub    = 45

7. My tutorials will explain how to perform a double loop search for

a. No. of hidden nodes
b. Initial RNG state (reproducible initial weights & datadivision).

8. For regression, search on subsets of

 greg fitnet tutorial Ntrials

Hope this helps.

Greg

MATLAB: How i can save the output of hidden layer of neural network

close all, clear all, clc, plt=0;

 delete h.mat
 delete LW.mat
 delete b2.mat
 delete tsettings.mat
 [ x, t ] = simplefit_dataset;
 net   = fitnet;
 rng(4151941) 
 [ net tr y0 ] = train( net, x, t);
 plt=plt+1,figure(plt)                                   % figure 1
 hold on
 plot(x,t,'b--','LineWidth',2)
 plot(x,y0,'r.','LineWidth',2)
 legend('TARGET','NN OUTPUT')
 xlabel('INPUT')
 ylabel('TARGETAND NN OUTPUT')
 title('SIMPLEFIT DATASET')

%Create tsettings, h, LW, b2

 [xn xsettings] = mapminmax(x);
 [tn tsettings]  = mapminmax(t);     % SAVE tsettings
 b1  = cell2mat(net.b(1))
 IW = cell2mat(net.IW)
 [ I N ]  = size(xn)
 B1  = b1*ones(1,N);
 % B1 = repmat(b,1,N);             % Alternate
 h    = tanh(B1+IW*xn);             % SAVE h
 LW = cell2mat(net.LW)           % SAVE LW
 b2 = cell2mat(net.b(2))            % SAVE b2
 clc
 whos h LW b2 tsettings
 dir
% save FILENAME ... is the command form of the syntax
% for convenient  saving from the command line.  With 
% command syntax, you do not need to  enclose strings in 
% single quotation marks.  Separate inputs with spaces
% instead of commas.  Do not use command syntax if 
% inputs such as FILENAME are variables.
 save h
 save LW
 save b2
 save tsettings
 dir
 whos N h LW b2 tsettings
 disp('BEFORE CLEARING SAVED HIDDEN VARIABLES')
 disp('ENTER TO CONTINUE')
 pause
 dir
 clear     h LW b2 tsettings
 whos N h LW b2 tsettings
 disp('AFTER CLEARING SAVED HIDDEN VARIABLES')
 disp('ENTER TO CONTINUE')
 pause
 load      h.mat
 load      LW b2 tsettings
 whos N h LW b2 tsettings
 disp('AFTER RELOADING SAVED HIDDEN VARIABLES')
 disp('ENTER TO CONTINUE')
 pause
 yn = b2 + LW*h;
 y  =  mapminmax('reverse',yn,tsettings);
 reloadingerror = max(abs(y-y0))
 break

% Hope this helps

Thank you for formally accepting my answer

Greg

Best Answer

Related Solutions

MATLAB: Could you please help me in Artificial neural network – supervised learning

MATLAB: How i can save the output of hidden layer of neural network

Related Question