How can use (if, end) inside the neural network script?

I have a script for neural network for 4 inputs and 1 target data (each one 104 times observation). when I run the network, the variety of performance in different runs are to much and it is sometime more and leas than 3, where my ideal performance is leas than 3. In order to solve this problem, I want to use a (if, then) inside the script. Although, I do not know this is applicable or not, but the network would not goes inside the loop. I would really appreciate to tell me how can I use (if, end) statement inside the script.

inputs = x;       
targets = t;      

[ I, N ]= size(x); 

performance=10;
if performance>3

% Create a Fitting Network
H = 10;
TF={'tansig','purelin'};
net = newff(x,t,H,TF);

net = init(net);
net = train(net,x,t);

net.divideFcn = 'dividerand';  % Divide data randomly
net.divideMode = 'sample';  % Divide up every sample
net.divideParam.trainRatio = 70/100;
net.divideParam.valRatio = 15/100;
net.divideParam.testRatio = 15/100;

net.trainparam.max_fail=10;

% Train the Network
[net,tr] = train(net,inputs,targets);

% Test the Network
outputs = net(inputs);
errors = gsubtract(targets,outputs);
performance = perform(net,targets,outputs);                                               

end

   W = outputs / [ ones(1,N) ; x ];
   y = W*[ ones(1,N) ;x ]; 

   view(net)

Answer 1

[Greg Heath]

Search the NEWSGROUP and ANSWERS with

 newff Ntrials

Hope this helps

Thank you for formally accepting my ANSWER

Greg

Answer 2

[Walter Roberson]

newff() is officially obsolete, and the documentation has been removed from current releases. You should try to convert it to more modern routines. See http://www.mathworks.com/matlabcentral/answers/77741-newff-create-a-feed-forward-backpropagation-network-obsoleted-in-r2010b-nnet-7-0

You might want to go back in and use nntool to rebuild the code.

You will probably want to train multiple times on different random subsets. I do not know what the best way of doing that is.