example2.asv

基于BP模型的神经网络模型

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
%
%  Data fitting DEMO of neural networks with matrix inputs.
%
%  Author: Povilas Daniu餴s, [e-mail]
%  http://ai.hacker.lt - lithuanian site about Artificial Intelligence.
%
%  TODO: weighted MSE minimization, maximal likelihood method, multiple
%  activation function support.
%  ----------------------------------------------------------------------

clear all


alpha = 0.9;       % inertia
eta = 0.005;       % inital learning rate
epsilon = 0.03;    % needed MSE
epsilon1 = 0.001;  % minimal descent (stopping criteria) - all iterations in this case
neurones = 20;    % su 20 neveikia :) 
n = 10;
numEpochs = 30;
earlyStop = 5;      
%[a,D] = textread('c:\laser1.txt','%s %f');
D =
D = (D - mean(D))/std(D);


for i=1:400
    data.training(i).mat = [ D(i+1), D(i+2) D(i+3); D(i+4),  D(i+5), D(i+6); D(i+7),  D(i+8), D(i+9); ]
    data.target(i) = D(i+10);
end

e = mNN_device(neurones,size(data.training(1).mat),alpha,eta,epsilon,epsilon1,earlyStop);
e_elm = ELM_train(e,data); 
s_elm = mNN_sim(e_elm,data);



plot(data.target,'r-'); hold on; plot(s_elm,'b-');
mse = sum((data.target - s_elm).^2) / length(data.target)

%hit_rate = sum(data.target .* s_elm > 0) / length(data.target)