Validation Error with trainbr is NaN

Question

gsourop 2017-3-24

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/331794-validation-error-with-trainbr-is-nan

回答： nazanin abz 2017-6-19

Hi everyone,

I would like to ask why do I get a vector of NaN values to the MSEval below when I apply the trainbr algorithm. However, when I apply the trainlm algorithm I get actual values.

a=randn(1,1000);
b= randn ( 1, 1000);
hiddenLayerSize= 1 ;
net = fitnet(hiddenLayerSize);
net.divideFcn = 'divideblock';
net.divideParam.trainRatio = 70/100;
net.divideParam.valRatio = 30/100;
net.divideParam.testRatio = 0/100;
net.trainFcn = 'trainbr';
net.trainParam.showWindow=false;
[net,tr] = train(net,a ,b);
outputs = net(a);
MSEval = tr.vperf

Thanks in advance.

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

nazanin abz 2017-6-19

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/331794-validation-error-with-trainbr-is-nan#answer_271107

it's not there by default anymore, you can use this command if you really need it. net.trainParam.max_fail = 6 However it's being said it's not necessary.