How to store the randomly generated initial weights while using fitnet function of ANN toolbox?

While using "fitnet" function of ANN toolbox, the initial weights are generated by this function itself and that doesn’t appear anywhere. Can anyone suggest me how to store these randomly generated initial weights by fitnet function?

More specifically: "fitnet" function starts with Initial weights (at 0 epoch) to optimized weights (lets at 10th epochs) by using let’s say "traingdm". So I want to store Initial weights (at 0 epoch).

Thanks for your help....

Answer 1

[Greg Heath]

Cut and paste into the command line:

[x,t] = simplefit_dataset;
  net = fitnet(10);        % Inputs and Outputs are NOT defined here
  view(net)
  WB  = getwb(net)         %  Only net.b{1} = zeros(10,1)is defined
  rng(4151941)             % Initialize the RNG so that results can be duplicated
  net = configure(net,x,t);% Now inputs and outputs are defined and ready to train
  IW  = net.IW{1,1}
   b1 = net.b{1}
  LW  = net.LW{2,1}
   b2 = net.b{2}

Hope this helps.

Thank you for formally accepting my answer

Greg

[Vikas]

Thanks...