neural network toolbox classify(net, x) function not working

11 visualizzazioni (ultimi 30 giorni)
matlab classify(net,x) generating undefined
overtime i run classify(net,x) on a trained neural network i get this output
<undefined>
<undefined>
<undefined>
<undefined>
my neural network is convolutional neural network that is trying to classify cancer images. Im using nfold cross validation andntrying to measure accuracy of each fold. For my second fold works fine but my first fold is generating this weird output. Im training my network 5 time for 5-fold cross validation and getting accuracy average over those 5
here is my code :
convnet11Fold1 = trainNetwork(Train1,layers,options)
save convnet11Fold1
YPred1 = predict(convnet11Fold1,nfoldTest1)
YTest1 = classify(convnet11Fold1,nfoldTest1)
TTest1 = nfoldTest1.Labels
accuracy1 = sum(YTest1 == TTest1)/numel(TTest1)
accuracyTotal(end+1)= accuracy1
totalT(end+1)= numel(TTest1)
totalY(end+1)= numel(YTest1)
convnet11Fold2 = trainNetwork(Train2,layers,options)
save convnet11Fold2
YTest2 = classify(convnet11Fold2,nfoldTest2)
TTest2 = nfoldTest2.Labels
accuracy2 = sum(YTest2 == TTest2)/numel(TTest2)
accuracyTotal(end+1)= accuracy2
totalT(end+1)= numel(TTest2)
totalY(end+1)= numel(YTest2)
  2 Commenti
hung ting
hung ting il 2 Apr 2018
Having the same trouble. Any suggestions?
Drew Grant
Drew Grant il 3 Dic 2020
Having the same problem too a year and a half later... help please!

Accedi per commentare.

Risposte (1)

Enrico Sanna
Enrico Sanna il 9 Set 2018
Having same problem on saved LSTM. Please help.

Categorie

Scopri di più su Deep Learning Toolbox in Help Center e File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by