I have started using Deep Learning Toolbox and getting error "Error using trainNetwork (line 183) Number of observations in X and Y disagree."

I am using below code. It works perfectly fine when I have the 4 D image. When I change it to 3D it starts giving the error. Any help would be highly appreciated.

load('C:\Users\myROIFC_GroupTwo.mat');
load('C:\Users\myROIFC_GroupZero.mat');
trainingLabels = [zeros(30,1);2*ones(23,1)];
trainingLabels = categorical(trainingLabels);
trainingImages=[myROIfc_GroupZero;myROIfc_GroupTwo];
size(myROIfc_GroupZero)
size(trainingImages)
testImages = trainingImages;
trainingImagesNew = shiftdim(trainingImages,1);
size(trainingImagesNew)               
testImagesNew = shiftdim(testImages,1); 
trainingImagesNew=  (reshape(trainingImagesNew,116,116,53));
testImagesNew=  (reshape(testImagesNew,116,116,53));

    numTrainingImages=size(trainingImagesNew,3);
    idx = randperm(size(trainingImagesNew,3),floor(numTrainingImages/2));
    XValidation = trainingImagesNew(:,:,idx);
    trainingImagesNew(:,:,idx) = [];
    YValidation = trainingLabels(idx);
    trainingLabels(idx) = [];   
    
    opts = trainingOptions('sgdm', 'Plots', 'training-progress', ...
        'Momentum', 0.93, ...
        'InitialLearnRate', 0.01, ... %0.001
        'LearnRateSchedule', 'piecewise', ...
        'LearnRateDropFactor', 0.1, ... %0.1
        'LearnRateDropPeriod', 8, ... %8
        'L2Regularization', 0.004, ...
        'ValidationFrequency',15,...
        'MaxEpochs',80 , ... %40
        'MiniBatchSize', 24, ...
        'ValidationData',{XValidation,YValidation}, ...
        'Verbose', true);
    
   [height,width,numChannels, ~] = size(trainingImagesNew);

    imageSize = [height width numChannels];
    inputLayer = imageInputLayer(imageSize);

 numImageCategories = length(unique(trainingLabels));

 filterSize = [11 11];
    numFilters = 64; 
    
    middleLayers = [
        convolution2dLayer(filterSize,numFilters,'Padding',3) 
        reluLayer()     
        maxPooling2dLayer(3,'Stride',2)          
        convolution2dLayer(filterSize,numFilters,'Padding',3)  %2
        reluLayer()
        maxPooling2dLayer(3, 'Stride',2)        
        convolution2dLayer(filterSize,2 * numFilters,'Padding',2)
        reluLayer()
        maxPooling2dLayer(3,'Stride',2)        
        ]
    
    finalLayers = [
        fullyConnectedLayer(64)
        reluLayer
        fullyConnectedLayer(numImageCategories) 
        softmaxLayer
        classificationLayer
        ]
    

    layers = [
        inputLayer
        middleLayers
        finalLayers
        ]
    layers(2).Weights = 0.0001 * randn([filterSize numChannels numFilters]);
        tic
        GWI_RCNNv0 = trainNetwork(trainingImagesNew, trainingLabels, layers, opts);
        toc

Answer 1

[Srivardhan Gadila]

As you are using the following syntax of the trainNetwork function: net = trainNetwork(X,Y,layers,options), you can refer to the corresponding explanation of the format of the X & Y to understand the cause of the error.

X — Image or feature data (numeric array)

Image or feature data, specified as a numeric array. The size of the array depends on the type of input:

Input: Description

2-D images: A h-by-w-by-c-by-N numeric array, where h, w, and c are the height, width, and number of channels of the images, respectively, and N is the number of images.

3-D images: A h-by-w-by-d-by-c-by-N numeric array, where h, w, d, and c are the height, width, depth, and number of channels of the images, respectively, and N is the number of images.

Features: A N-by-numFeatures numeric array, where N is the number of observations and numFeatures is the number of features of the input data.

Y — Responses (categorical vector of labels | numeric array | cell array of categorical sequences | cell array of numeric sequences)

Responses, specified as a categorical vector of labels, a numeric array, a cell array of categorical sequences, or cell array of numeric sequences. The format of Y depends on the type of task. Responses must not contain NaNs.

Task: Format

Image or feature classification or Sequence-to-label classification: N-by-1 categorical vector of labels, where N is the number of observations.

Sequence-to-sequence classification: N-by-1 cell array of categorical sequences of labels, where N is the number of observations. Each sequence must have the same number of time steps as the corresponding predictor sequence.

For sequence-to-sequence classification tasks with one observation, sequences can also be a vector. In this case, Y must be a categorical sequence of labels.