Error in NN training using differential evolution
显示 更早的评论
I want to train Neural network by using differential evolution algorithm code found here. I got this error when I use feedforwardnet(H,'traindiffevol') FCN does not return an info object. and when I use newff(x,t,'traindiffeverror) Layer sizes is not a row vector of positive integers. Note this network for classification and I am using k-fold cross validation. I am using MATLAB R2015B ON MAC OS.
close all, clear all, clc, plt=0;
tic
k = 10 % k-fold
%%%%%%%%%%%%%%%%%%%%%%%%%%
[x,t] = iris_dataset;
if iscell( iris_dataset) ;
x = cell2mat(x); % array
t = cell2mat(t);
end
[ I N ] = size(x) %[ 1 94 ]get row size with size() function get 2 dimension
[O N ] = size(t) % [ 1 94 ]
MSE00 = mean(var(t',1)) % 8.34 Biased Reference MSE00a is the MSE "a"djusted for the loss in estimation degrees of freedom caused by the bias of evaluating the MSE with the same data that was used to build the model.
MSE00a = mean(var(t',0)) % 8.43 Unbiased Reference
whos
%%%%%%%%%%%%%%%%%%%%%%%%%%%
rng('default') % Or substitute your lucky number
ind0 = randperm(N);
% ind0 = 1:N; % For debugging
M = floor(N/k) %
Ntrn = N-2*M % length(trnind)
Ntrneq = Ntrn*O % No. of training equations
H = 10;
Nw = (I+1)*H+(H+1)*O % No. of unknown weights
Ndof = Ntrneq-Nw % No. of estimation degrees of freedom
MSEgoal = 0.01*MSE00 %
MinGrad = MSEgoal/100 %
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Create a Pattern Recognition Network
% net = feedforwardnet(H,'traindiffevol');
net = newff(x,t,'traindiffevol');
net.trainParam.goal = MSEgoal;
net.trainParam.min_grad = MinGrad;
net.trainParam.epochs = 1000;
net.trainParam.show = 25;
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Input and Output Pre/Post-Processing Functions
%net.inputs{1}.processFcns = {'removeconstantrows','mapminmax'};
% net.outputs{2}.processFcns = {'removeconstantrows','mapminmax'};
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
k=10;
net.divideFcn = 'divideind';
cvFolds = crossvalind('Kfold', size(t,2), k); %# get indices of 10-fold CV
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
for i = 1:k %# for each fold
rngstate(i) = rng;
net = configure(net,x,t);
testIdx = (cvFolds == i); %# get indices of test instances
trainIdx = ~testIdx ; %# get indices training instances
trInd=find(trainIdx)
tstInd=find(testIdx)
net.divideParam.trainInd=trInd
net.divideParam.testInd=tstInd
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Choose a Performance Function
net.performFcn = 'mse'; % Mean squared error
% Train the Network
[net,tr] = train( net, x, t );
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%# test using test instances
outputs = net(x);
errors = gsubtract(t,outputs);
performance = perform(net,t,outputs)
trainTargets = t .* tr.trainMask{1};
testTargets = t .* tr.testMask{1};
trainPerformance = perform(net,trainTargets,outputs)
testPerformance = perform(net,testTargets,outputs)
test(k)=testPerformance;
%%%%%%%%%%%%%%%%%%%%%%%%%%
save net
stopcrit{i,1} = tr.stop;
bestepoch(i,1) = tr.best_epoch;
R2trn(i,1) = 1 - tr.best_perf/MSE00;
R2trna(i,1) = 1 - (Ntrneq/Ndof)* tr.best_perf/MSE00a;
R2val(i,1) = 1 - tr.best_vperf/MSE00;
R2tst(i,1) = 1 - tr.best_tperf/MSE00;
end
figure, plotconfusion(t,outputs)
accuracy=mean(test);
% View the Network
view(net)
stopcrit = stopcrit
result = [ bestepoch R2trn R2trna R2val R2tst]
minresult = min(result)
meanresult = mean(result)
medresult = median(result)
stdresult = std(result)
maxresult = max(result)
Elapsedtime = toc %3.87 sec
回答（0 个）
类别
在 帮助中心 和 File Exchange 中查找有关 Startup and Shutdown 的更多信息
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!
Translated by 
