How can i create a deep neural network for a classificaton problem on glove foram data set

I create a deep neural network for a sentiment analysis classification on a glove format dataset but it doesnt work.it deletes al positive answers.and i cant set the output to1 because of two classes.neural network generates a 2D matrix of nans with two rows and number of columns is equal to observations.but i need just one row.codes are here: numClasses=numel(categories(YTrain)); inputSize_c = dimension_c; numResponses_c=1;numHiddenUnits_c=200; options_c = trainingOptions('adam', ... 'MaxEpochs',30, ... 'GradientThreshold',1, ... 'InitialLearnRate',0.005, ... 'LearnRateSchedule','piecewise', ... 'LearnRateDropPeriod',125, ... 'LearnRateDropFactor',0.2, ... 'Verbose',0, ... 'Plots','training-progress'); layers_c = [ ... sequenceInputLayer(inputSize_c) flattenLayer lstmLayer(numHiddenUnits_c,'outputMode','sequence') fullyConnectedLayer(numClasses) softmaxLayer classificationLayer];