深度神经网络LSTM序列分类应用

[XTrain,YTrain]=japaneseVowelsTrainData; XTrain(1:5) figure plot(XTrain{1}') xlabel("Time Step"); title("Training Observation 1") legend("Feature"+string(1:12),'Location','northeastoutside') %prepare data for padding 野割 %get the sequence lengths for each observation numObservations = numel(XTrain); for i=1:numObservations sequence=XTrain{i}; sequenceLengths(i)=size(sequence,2); end % sort by sequence lengths [sequenceLengths,idx]=sort(sequenceLengths); XTrain=XTrain(idx); YTrain=YTrain(idx); % view the sorted sequence in a bar chart figure bar(sequenceLengths) ylim([0 30]) xlabel("Sequence") ylabel("Length") title("Sorted Data") % minibatch miniBatchSize=27; %define LSTM network architecture inputSize=12; numHiddenUnits=100; numClasses=9; layers=[... sequenceInputLayer(inputSize) bilstmLayer(numHiddenUnits,'OutputMode','last') fullyConnectedLayer(numClasses) softmaxLayer classificationLayer ] % training options maxEpochs=100; miniBatchSize=27; options=trainingOptions('adam',... 'ExecutionEnvironment','cpu',... 'GradientThreshold',1,... 'MaxEpochs',maxEpochs,... 'MinibatchSize',miniBatchSize,... 'SequenceLength','longest',... 'Shuffle','never',... 'Verbose',0,... 'Plots','training-progress'); % train LSTM network net=trainNetwork(XTrain,YTrain,layers,options); % test LSTM network % load the test data [XTest,YTest]=japaneseVowelsTestData; XTest(1:3) % numObservationsTest = numel(XTest); for i=1:numObservationsTest sequence = XTest{i}; sequenceLengthsTest(i)=size(sequence,2); end [sequenceLengthsTest,idx]=sort(sequenceLengthsTest); XTest=XTest(idx); YTest=YTest(idx); % classify the test data miniBatchSize =27; YPred=classify(net,XTest,... 'MiniBatchSize',miniBatchSize,... 'SequenceLength','longest'); acc=sum(YPred==YTest)./numel(YTest)