How to Loop the neural network training to choose the best performance?
조회 수: 1 (최근 30일)
이전 댓글 표시
Hi, I need some help on how to train a network for different value of Neurons, and save the MSE then choose the best MSE to select the best trained network.
I am using the fitnet as follows:
trainFcn = 'trainlm'; % Levenberg-Marquardt backpropagation.
% Create a Fitting Network & set number of neurons
hiddenLayerSize = 30;
net = fitnet(hiddenLayerSize,trainFcn);
[net,tr] = train(net,X,T);
testX = X(:,tr.testInd);
testT = T(:,tr.testInd);
testY = net(testX);
perf = mse(net,testT,testY)
댓글 수: 0
채택된 답변
KSSV
2018년 7월 31일
댓글 수: 3
KSSV
2018년 8월 1일
N = [ 5, 10, 15, 20, 25, 30, 35, 40. ];
NN = cell(length(N),1) ;
P = zeros(length(N),1) ;
for i = 1:length(N)
trainFcn = 'trainlm'; % Levenberg-Marquardt backpropagation.
% Create a Fitting Network & set number of neurons
hiddenLayerSize = N(i);
net = fitnet(hiddenLayerSize,trainFcn);
[net,tr] = train(net,X,T);
NN{i} = net ;
testX = X(:,tr.testInd);
testT = T(:,tr.testInd);
testY = net(testX);
perf = mse(net,testT,testY) ;
P(i) = perf ;
end
추가 답변 (0개)
참고 항목
카테고리
Help Center 및 File Exchange에서 Modeling and Prediction with NARX and Time-Delay Networks에 대해 자세히 알아보기
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!