Error while trying to train network

I've been trying to make a system that will read images from an image store to classify images of solar panels as soiled or not.

I can't seem to be able to make the training work. The GUI starts but no iterations are done.

Code below:

infolder = 'C:\Users\Potato\Desktop\CDP\Solar_Panel_Soiling_Image_dataset\PanelImages';
imgFiles = dir([infolder,filesep,'\*.jpg']) ;
imds = imageDatastore(infolder,'IncludeSubfolders',true,'LabelSource','foldernames');
figure;
j= 0
for i = 1:20            
    j = j +1
    subplot(4,5,j);
    imshow(imds.Files{i}); % display the first 20 images
end
figure;
perm = randperm(10000,20);
for i = 1:20
    subplot(4,5,i);
    imshow(imds.Files{perm(i)});
end
labelCount = countEachLabel(imds)
img = readimage(imds,1);
size(img)

%% Training and Validation Sets
numTrainFiles = 5000;
[imdsTrain,imdsValidation] = splitEachLabel(imds,numTrainFiles,'randomize');

%% Network Archeticture
layers = [
    imageInputLayer([192 192 3])
    
    convolution2dLayer(3,8,'Padding','same')
    batchNormalizationLayer
    reluLayer
    
    maxPooling2dLayer(2,'Stride',2)
    
    convolution2dLayer(3,16,'Padding','same')
    batchNormalizationLayer
    reluLayer
    
    maxPooling2dLayer(2,'Stride',2)
    
    convolution2dLayer(3,32,'Padding','same')
    batchNormalizationLayer
    reluLayer
    
    fullyConnectedLayer(1)
    softmaxLayer
    classificationLayer];

%% Training Options
options = trainingOptions('sgdm', ...
    'InitialLearnRate',0.01, ...
    'MaxEpochs',4, ...
    'Shuffle','every-epoch', ...
    'ValidationData',imdsValidation, ...
    'ValidationFrequency',30, ...
    'Verbose',false, ...
    'Plots','training-progress');
net = trainNetwork(imdsTrain,layers,options);

%% Validation Set Classification and Accuracy
YPred = classify(net,imdsValidation);
YValidation = imdsValidation.Labels;
accuracy = sum(YPred == YValidation)/numel(YValidation)

Shows me this error:

Operation terminated by user during matlab.io.datastore.internal.fileset.CompressedFileSet/getFilesAsCellStr

In matlab.io.datastore.internal.fileset.ResolvedFileSet/getFiles (line 530)

files = getFilesAsCellStr(fs, ii);

In matlab.io.datastore.FileSet/getFiles (line 568)

files = fs.InternalFileSet.getFiles(ii);

In matlab.io.datastore.splitter.WholeFileCustomReadFileSetSplitter/getFilesAsCellStr (line 111)

filesAsCellStr = getFiles(splitter.Files, indices);

In matlab.io.datastore.ImageDatastore/nextFilesToRead (line 814)

files = imds.Splitter.getFilesAsCellStr(idxes);

In matlab.io.datastore.ImageDatastore/preFetchRead (line 603)

[filesToRead, idxes] = nextFilesToRead(imds);

In matlab.io.datastore.ImageDatastore/read (line 68)

[data, info] = preFetchRead(imds);

In nnet.internal.cnn.datastore.ImageDatastoreMiniBatchDatastore/read (line 39)

[data,info] = self.imds.read();

In nnet.internal.cnn.dispatcher.MiniBatchDatastoreDispatcher/readData (line 306)

[data,info] = read(this.Datastore);

In nnet.internal.cnn.dispatcher.MiniBatchDatastoreDispatcher/next (line 179)

[miniBatchData, miniBatchResponse] = readData(this);

In nnet.internal.cnn.util.ValidationStatelessPredictStrategy/computeLossAndMetric (line 57)

[X, T, idx] = data.next();

In nnet.internal.cnn.util.ValidationStatelessPredictStrategy/predictAndComputeMetrics (line 39)

metrics = computeLossAndMetric(this, net, data, "accuracy", "validationAccuracy");

In nnet.internal.cnn.util.ValidationReporter/addValidationToSummary (line 123)

metrics = this.PredictStrategy.predictAndComputeMetrics( net, this.Data );

In nnet.internal.cnn.util.ValidationReporter/computeIteration (line 74)

this.addValidationToSummary(summary, net);

In nnet.internal.cnn.util.VectorReporter/computeAndReport (line 68)

feval( method, this.Reporters{i}, varargin{:} );

In nnet.internal.cnn.util.VectorReporter/computeIteration (line 24)

computeAndReport( this, 'computeIteration', summary, network );

In nnet.internal.cnn.Trainer/train (line 148)

reporter.computeIteration( this.Summary, net );

In nnet.internal.cnn.trainNetwork.doTrainNetwork (line 108)

trainedNet = trainer.train(trainedNet, trainingDispatcher);

In trainNetwork (line 189)

[trainedNet, info] = nnet.internal.cnn.trainNetwork.doTrainNetwork(factory,varargin{:});

In test (line 68)

net = trainNetwork(imdsTrain,layers,options);

I've tried changing different variables to fix this with no luck. I'm a beginner so take it easy.

[Walter Roberson]

"Operation terminated by user" means that you used control-C to interrupt calculation.

[Khalifa Alsuwaidi]

It is just stuck in training iteration 1, nothing happens

[Matt J]

See if your datasore can read anything:

X=read(imdsTrain)

Answer 1

[Khalifa Alsuwaidi]

Thanks for your help guys. Fixed the folder tree and made some tweaks. It works!

infolder = 'C:\Users\Potato\Desktop\CDP\Solar_Panel_Soiling_Image_dataset\train';

imgFiles = dir([infolder,filesep,'\*.jpg']) ;

imds = imageDatastore(infolder,'IncludeSubfolders',true,'LabelSource','foldernames');

figure;

perm = randperm(22000,20);

for i = 1:20

subplot(4,5,i);

imshow(imds.Files{perm(i)});

end

labelCount = countEachLabel(imds)

img = readimage(imds,1);

size(img)

%% Training and Validation Sets

numTrainFiles = 5000;

[imdsTrain,imdsValidation] = splitEachLabel(imds,numTrainFiles,'randomize');

%% Network Archeticture

layers = [

imageInputLayer([192 192 3])

convolution2dLayer(3,8,'Padding','same')

batchNormalizationLayer

reluLayer

maxPooling2dLayer(2,'Stride',2)

convolution2dLayer(3,16,'Padding','same')

batchNormalizationLayer

reluLayer

maxPooling2dLayer(2,'Stride',2)

convolution2dLayer(3,32,'Padding','same')

batchNormalizationLayer

reluLayer

fullyConnectedLayer(2)

softmaxLayer

classificationLayer];

%% Training Options

options = trainingOptions('sgdm', ...

'InitialLearnRate',0.01, ...

'MaxEpochs',4, ...

'Shuffle','every-epoch', ...

'ValidationData',imdsValidation, ...

'ValidationFrequency',30, ...

'Verbose',false, ...

'Plots','training-progress');

net = trainNetwork(imdsTrain,layers,options);

%% Validation Set Classification and Accuracy

YPred = classify(net,imdsValidation);

YValidation = imdsValidation.Labels;

accuracy = sum(YPred == YValidation)/numel(YValidation)