Haw can convert categories array to array of numeric or logical ?
Afficher commentaires plus anciens
I am creating a neural network in MATLAB,but i got this error
Error using nntraining.setup>setupPerWorker (line 96)
Targets T{1,1} is not numeric or logical.
outputFolder= fullfile(tempdir, '');
rootFolder = fullfile(outputFolder, '');
categories = {'2', '1', '0'};
imds = imageDatastore(fullfile(rootFolder, categories), 'LabelSource', 'foldernames');
numImageCategories = size(categories(imds.Labels),1); % get category labels
[trainingDS,validationDS] = splitEachLabel(imds,0.7,'randomize'); % generate training and validation set
LabelCnt = countEachLabel(imds); % load lable information
p=[];
t1=[];
[l,m]=size(trainingDS.Files);
for q=1:l
im1=readimage(trainingDS,q);
im=double( im1);
[r,c]=size(im)
p11=[];
for i=1:r
for j=1:c
p11=[p11;im(i,j,1)];
end
end
p=[p p11];
t1=[t1 trainingDS.Labels(q)];
t = categories(t);
end
% % % %% Multilayer backpropagation neural network
% % % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
net=newff(minmax(p), [4096 5 1], {'logsig','logsig','logsig'}, 'traingdx');
[net,tr,Y,E,Pf,Af]=train(net,p,t);
net
10 commentaires
Walter Roberson
le 5 Mar 2020
Try passing in categorical(t)
dziri halima
le 5 Mar 2020
dziri halima
le 5 Mar 2020
Walter Roberson
le 5 Mar 2020
Then do not use
categories = {'2', '1', '0'};
Use
categories = [2,1,0];
dziri halima
le 5 Mar 2020
Walter Roberson
le 5 Mar 2020
Sigh. Then
t = char(categories(t(:))) - '0';
dziri halima
le 5 Mar 2020
Walter Roberson
le 5 Mar 2020
Where do you assign to t to be able to use it in the line
t = categories(t);
? You have been building up t1 but not t.
dziri halima
le 5 Mar 2020
dziri halima
le 5 Mar 2020
Réponses (0)
Catégories
En savoir plus sur Matrix Indexing dans Centre d'aide et File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!