Integrating uigetfile to CNN in general

I'm trying to integrate uigetfile function for the given CNN example I found so it doesn't have to be user-specific and it doesn't need to be adjusted by different users but when I tried it, I get errors. Here's the code I attempted with it:

[fileInput, pathInput] = uigetfile('*.png','MultiSelect','on');

[fileValid, pathValid] = uigetfile('*.png','MultiSelect','on');

numTrainfiles =numel(fileInput);

numValidfiles = numel(fileValid);

for k = 1:length(numTrainfiles)

baseFileNameInput{k} = fileInput{k};

fullFileName{k} = fullfile(baseFileNameInput);

imageArrayInput(k,:,:,:) = imread(fullFileName{k});

[~, height{k}, width{k}, colors{k}] = size(imageArrayInput);

end

for l = 1:length(numValidfiles)

baseFileNameValid{l} = fileValid{l};

fullFileName2{l} = fullfile(baseFileNameValid);

imageArrayValid(l,:,:,:) = imread(fullFileNameValid{l});

[~, height{l}, width{l}, colors{l}] = size(imageArrayValid);

end

imageAugmenter = imageDataAugmenter( ...

'RandRotation', [-20,20], ...

'RandXTranslation',[-3,3], ...

'RandYTranslation',[-3,3]);

imageSize = [height{k} width{k} colors{k}];

augimds = augmentedImageDatastore(imageSize,numTrainfiles,'DataAugmentation',imageAugmenter);

[imdsTrain] = splitEachLabel(imds,numTrainfiles,'randomized');

[imdsValidation] = splitEachLabel(imds2,numValidfiles,'randomized');

augimds = augmentedImageDatastore(imageSize,imdsTrain,'DataAugmentation',imageAugmenter);

%definingarchitecture

numClasses = numel(categories(imdsTrain.Labels));

numClasses2 = numel(categories(imdsValidation.Labels));

layers = [

imageInputLayer(imageSize)

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(numClasses)

softmaxLayer

classificationLayer];

options = trainingOptions('sgdm', ...

'MaxEpochs',6, ...

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

'MiniBatchSize',64,...

'ValidationData',imdsValidation, ...

'ValidationFrequency',30, ...

'Verbose',false, ...

'Plots','training-progress');

net = trainNetwork(imdsTrain,layers,options);

YPred = classify(net,imdsValidation);

yvalidation = imdsValidation.Labels;

accuracy = mean(Ypred == yvalidation);

And this is the error that I get:

"Error using imread>parse_inputs (line 504)

The file name or URL argument must be a character vector or string scalar.

Error in imread (line 342)

[source, fmt_s, extraArgs, was_cached_fmt_used] = parse_inputs(cached_fmt, varargin{:});

Error in cnn3 (line 9)

imageArrayInput(k,:,:,:) = imread(fullFileName{k});"

Hoping that anyone could answer this so that I could also integrate uigetfile for my own CNN

Answer 1

[Reshma Nerella]

Hi,

In this code

fullFileName{k} = fullfile(baseFileNameInput);
imageArrayInput(k,:,:,:) = imread(fullFileName{k});

fullFileName is a cell array of cell arrays.

Two possible fixes for the error:

1.Modify the fullFileName as cell array of character vectors

fullFileName = fullfile(baseFileNameInput);
imageArrayInput(k,:,:,:) = imread(fullFileName{k});

2.Refer the character vector properly

fullFileName{k} = fullfile(baseFileNameInput);
imageArrayInput(k,:,:,:) = imread(fullFileName{k}{1});

Hope this helps!