I am trying to use a different data for my Validation and it is saying that: Training and validation responses must have the same categories. To view the categories of the res
4 visualizaciones (últimos 30 días)
Mostrar comentarios más antiguos
Ramon Miguel Legaspi
el 23 de Nov. de 2021
Respondida: Philip Brown
el 25 de Nov. de 2021
myfolder = 'C:\Users\Myname\Downloads\fall dataset\rgb';
dataDir = fullfile(myfolder);
imdir = fullfile(dataDir);
myfolder2 = 'C:\Users\Myname\Downloads\Validation';
dataDir2 = fullfile(myfolder2);
imdir2 = fullfile(dataDir2);
imds = imageDatastore(imdir, "IncludeSubfolders",true ,"LabelSource","foldernames");
imds2 = imageDatastore(imdir2,"IncludeSubfolders",true,"LabelSource","foldernames");
numTrainfiles =5172;
numValidfiles = 6598;
[imdsTrain] = splitEachLabel(imds,numTrainfiles,'randomized');
[imdsValidation] = splitEachLabel(imds2,numValidfiles,'randomized');
%definingarchitecture
inputSize = [ 240 320 3];
numClasses = numel(categories(imdsTrain.Labels));
numClasses2 = numel(categories(imdsValidation.Labels));
layers = [
imageInputLayer(inputSize)
convolution2dLayer(5,20)
batchNormalizationLayer
reluLayer
fullyConnectedLayer(numClasses)
softmaxLayer
classificationLayer];
%trainetwork
options = trainingOptions('sgdm', ...
'MaxEpochs',4, ...
'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);
0 comentarios
Respuesta aceptada
Philip Brown
el 25 de Nov. de 2021
It's likely that your training and validation folders contain different folder names, and those are being used as the class labels. For example, your training set has labels A, B, and C, but your validation set has labels A, B and D. This means your network never learns to classify into class D during training.
0 comentarios
Más respuestas (0)
Ver también
Categorías
Más información sobre Image Data Workflows en Help Center y File Exchange.
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!