How can i use CNN?

I have a 3D feature set [10x2000x9, 10x2000x9,10x2000x9......................10x2000x9] and corrosponding ground truth in 4 class like [0,1,2,3]. Means for each 10x2000x9 their will be a ground truth from 0 to 3. How can i use CNN for this to classify in multiclass?

[KSSV]

You may go through the examples and pick the code and extend to your case.

Answer 1

[Srivardhan Gadila]

You can refer to Create Simple Deep Learning Network for Classification, Training a Model from Scratch, Get Started with Deep Learning Toolbox & Deep Learning Toolbox. Also the following code might give you some idea to get started quickly:

inputSize = [10 2000 9];
numSamples = 128;
numClasses = 4;

%% Generate random data for training the network.
trainData = randn([inputSize numSamples]);
trainLabels = categorical(randi([0 numClasses-1], numSamples,1));

%% Create a network.
layers = [
    imageInputLayer(inputSize,'Name','input')  
    convolution2dLayer(3,16,'Padding','same','Name','conv_1')
    batchNormalizationLayer('Name','BN_1')
    reluLayer('Name','relu_1')
    fullyConnectedLayer(10,'Name','fc1')
    fullyConnectedLayer(numClasses,'Name','fc2')
    softmaxLayer('Name','softmax')
    classificationLayer('Name','classOutput')];

lgraph = layerGraph(layers);

%% Define training options.
options = trainingOptions('adam', ...
    'InitialLearnRate',0.005, ...
    'LearnRateSchedule','piecewise',...
    'MaxEpochs',100, ...
    'MiniBatchSize',128, ...
    'Verbose',1, ...
    'Plots','training-progress');

%% Train the network.
net = trainNetwork(trainData,trainLabels,layers,options);