Invalid training data. For classification tasks, responses must be a vector of categorical responses. For regression tasks, responses must be a vector, a matrix, or a 4-D arra

clc; clear all; close all;
load Projectdata.mat

% Split Data Glucose
GlucoseReadings_T = GlucoseReadings';
GlucoseReadings_train = GlucoseReadings_T;
train_GlucoseReadings = GlucoseReadings_train(1:84,:);
train_GR_output = GR_output(1:17);
%Split Data Insulin 
InsulinReadings_T = InsulinReadings';
InsulinReadings_train = InsulinReadings_T;
train_InsulinReadings = InsulinReadings_train(1:84,:);
train_INS_output = INS_output(1:17);
% Data Batch Glucose
GlucoseReadingsTrain=(reshape(train_GlucoseReadings', [1749,84]));
val_GlucoseReadings = GlucoseReadings_train(85:102,:);
val_GR_output = GR_output(85:102);
GlucoseReadingsVal=(reshape(val_GlucoseReadings', [1749,18]));
test_GlucoseReadings =GlucoseReadings_train(103:120,:);
test_GR_output = GR_output(103:120);
GlucoseReadingsTest=(reshape(test_GlucoseReadings', [1749,18]));
numFeatures = size(GlucoseReadings_T,2);
%Data Batch Insulin
InsulinReadingsTrain=(reshape(train_InsulinReadings', [1758,84]));
val_InsulinReadings = InsulinReadings_train(85:102,:);
val_INS_output = INS_output(85:102);
InsulinReadingsVal=(reshape(val_InsulinReadings', [1758,18]));
test_InsulinReadings = InsulinReadings_train(103:120,:);
test_INS_output = INS_output(103:120);
InsulinReadingsTest=(reshape(test_InsulinReadings', [1758,18]));
numFeatures1 = size(InsulinReadings_T,2);
% number of hidden units represent the size of the data
numHiddenUnits = 120;
%number of classes represent different patients normal,LIS,type2....
numClasses = length(categories(categorical(GR_output)));
numClasses1 = length(categories(categorical(INS_output)));
layers = [ ...
    sequenceInputLayer(numFeatures) 
    %dropoutLayer(0.5)
    instanceNormalizationLayer
    bilstmLayer(round(numHiddenUnits/2),'OutputMode','sequence')
    fullyConnectedLayer(numClasses)
    instanceNormalizationLayer
    softmaxLayer
    classificationLayer];
options = trainingOptions('adam', ...
    'MaxEpochs',100, ...
    'GradientThreshold',1, ...
    'Verbose',false, ...
    'ValidationData',{GlucoseReadingsVal, val_GR_output},...
    'ValidationData',{InsulinReadingsVal, val_INS_output},...
    'LearnRateDropFactor',0.2,...
    'LearnRateDropPeriod',5,...
    'Plots','training-progress');
% Train
whos

  Name                          Size                Bytes  Class                           Attributes

  GR_output                     1x120                7296  string                                    
  GlucoseReadings            1749x120             1679040  double                                    
  GlucoseReadingsTest        1749x18               251856  double                                    
  GlucoseReadingsTrain       1749x84              1175328  double                                    
  GlucoseReadingsVal         1749x18               251856  double                                    
  GlucoseReadings_T           120x1749            1679040  double                                    
  GlucoseReadings_train       120x1749            1679040  double                                    
  INS_output                    1x120                7296  string                                    
  InsulinReadings            1758x120             1687680  double                                    
  InsulinReadingsTest        1758x18               253152  double                                    
  InsulinReadingsTrain       1758x84              1181376  double                                    
  InsulinReadingsVal         1758x18               253152  double                                    
  InsulinReadings_T           120x1758            1687680  double                                    
  InsulinReadings_train       120x1758            1687680  double                                    
  layers                        7x1                  3332  nnet.cnn.layer.Layer                      
  numClasses                    1x1                     8  double                                    
  numClasses1                   1x1                     8  double                                    
  numFeatures                   1x1                     8  double                                    
  numFeatures1                  1x1                     8  double                                    
  numHiddenUnits                1x1                     8  double                                    
  options                       1x1                254865  nnet.cnn.TrainingOptionsADAM              
  test_GR_output                1x18                 1248  string                                    
  test_GlucoseReadings         18x1749             251856  double                                    
  test_INS_output               1x18                 1248  string                                    
  test_InsulinReadings         18x1758             253152  double                                    
  train_GR_output               1x17                 1048  string                                    
  train_GlucoseReadings        84x1749            1175328  double                                    
  train_INS_output              1x17                 1048  string                                    
  train_InsulinReadings        84x1758            1181376  double                                    
  val_GR_output                 1x18                 1248  string                                    
  val_GlucoseReadings          18x1749             251856  double                                    
  val_INS_output                1x18                 1248  string                                    
  val_InsulinReadings          18x1758             253152  double                                    

net = trainNetwork(GlucoseReadingsTrain,train_GR_output,layers,options);

Error using trainNetwork (line 184)
Invalid training data. For classification tasks, responses must be a vector of categorical responses. For regression tasks, responses must be a vector, a matrix, or a 4-D array of numeric responses which must not contain NaNs.

net1 = trainNetwork(InsulinReadingsTrain,train_INS_output,layers,options);
% Test
miniBatchSize = 27;
GR_outputPred = classify(net,GlucoseReadingsTest,...
    'MiniBatchSize',miniBatchSize,...
    'Environment','cpu');
acc = mean(GR_outputPred(:) == categorical(test_GR_output(:)))
acc1 = mean(INS_outputPred(:) == categorical(test_INS_output(:)))
figure
t = confusionchart(categorical(test_GR_output(:)),GR_outputPred(:));
figure
t1 = confusionchart(categorical(test_INS_output(:)),INS_outputPred(:));

[KSSV]

Please note that, you have to close/ acknowledge the already posted question and go for other question.

[Nathaniel Porter]

Sorry it was accepted hope that means closing it

Answer 1

[Walter Roberson]

net = trainNetwork(GlucoseReadingsTrain, categorical(train_GR_output), layers,options);

[Nathaniel Porter]

This however did not work it produced the error seen below:

Invalid training data. Sequence responses must have the same sequence length as the corresponding

predictors.

[Walter Roberson]

train_GR_output = GR_output(1:17);

That response should only be used with an input of size 17.

You should not be training on data only from one class: you should be training on data from all of your classes.

Answer 2

[yanqi liu]

clc; clear all; close all;
load Projectdata.mat

% Split Data Glucose
GR_output=categorical(GR_output);
INS_output=categorical(INS_output);
GlucoseReadings_T = GlucoseReadings';
GlucoseReadings_train = GlucoseReadings_T;
train_GlucoseReadings = GlucoseReadings_train(1:84,:);
train_GR_output = GR_output(1:84);
%Split Data Insulin 
InsulinReadings_T = InsulinReadings';
InsulinReadings_train = InsulinReadings_T;
train_InsulinReadings = InsulinReadings_train(1:84,:);
train_INS_output = INS_output(1:84);
% Data Batch Glucose
GlucoseReadingsTrain=(reshape(train_GlucoseReadings', [1749,84]));
val_GlucoseReadings = GlucoseReadings_train(85:102,:);
val_GR_output = GR_output(85:102);
GlucoseReadingsVal=(reshape(val_GlucoseReadings', [1749,18]));
test_GlucoseReadings =GlucoseReadings_train(103:120,:);
test_GR_output = GR_output(103:120);
GlucoseReadingsTest=(reshape(test_GlucoseReadings', [1749,18]));
numFeatures = size(GlucoseReadings_T,2);
%Data Batch Insulin
InsulinReadingsTrain=(reshape(train_InsulinReadings', [1758,84]));
val_InsulinReadings = InsulinReadings_train(85:102,:);
val_INS_output = INS_output(85:102);
InsulinReadingsVal=(reshape(val_InsulinReadings', [1758,18]));
test_InsulinReadings = InsulinReadings_train(103:120,:);
test_INS_output = INS_output(103:120);
InsulinReadingsTest=(reshape(test_InsulinReadings', [1758,18]));
numFeatures1 = size(InsulinReadings_T,2);
% number of hidden units represent the size of the data
numHiddenUnits = 120;
%number of classes represent different patients normal,LIS,type2....
numClasses = length(categories(categorical(GR_output)));
numClasses1 = length(categories(categorical(INS_output)));
layers = [ ...
    sequenceInputLayer(numFeatures) 
    dropoutLayer(0.5)
    %instanceNormalizationLayer
    bilstmLayer(round(numHiddenUnits/2),'OutputMode','sequence')
    fullyConnectedLayer(numClasses)
    %instanceNormalizationLayer
    dropoutLayer(0.5)
    softmaxLayer
    classificationLayer];
layers1 = [ ...
    sequenceInputLayer(numFeatures1) 
    dropoutLayer(0.5)
    %instanceNormalizationLayer
    bilstmLayer(round(numHiddenUnits/2),'OutputMode','sequence')
    fullyConnectedLayer(numClasses)
    %instanceNormalizationLayer
    dropoutLayer(0.5)
    softmaxLayer
    classificationLayer];
options = trainingOptions('adam', ...
    'MaxEpochs',100, ...
    'GradientThreshold',1, ...
    'Verbose',false, ...
    'ValidationData',{GlucoseReadingsVal, val_GR_output},...
    'LearnRateDropFactor',0.2,...
    'LearnRateDropPeriod',5,...
    'Plots','training-progress');
options1 = trainingOptions('adam', ...
    'MaxEpochs',100, ...
    'GradientThreshold',1, ...
    'Verbose',false, ...
    'ValidationData',{InsulinReadingsVal, val_INS_output},...
    'LearnRateDropFactor',0.2,...
    'LearnRateDropPeriod',5,...
    'Plots','training-progress');
% Train
% whos
net = trainNetwork(GlucoseReadingsTrain,train_GR_output,layers,options);
net1 = trainNetwork(InsulinReadingsTrain,train_INS_output,layers1,options1);
% Test
miniBatchSize = 27;
GR_outputPred = classify(net,GlucoseReadingsTest,...
    'MiniBatchSize',miniBatchSize,...
    'ExecutionEnvironment','cpu');
acc = mean(GR_outputPred(:) == categorical(test_GR_output(:)))

INS_outputPred = classify(net1,InsulinReadingsTest,...
    'MiniBatchSize',miniBatchSize,...
    'ExecutionEnvironment','cpu');
acc1 = mean(INS_outputPred(:) == categorical(test_INS_output(:)))

figure
t = confusionchart(categorical(test_GR_output(:)),GR_outputPred(:));
figure
t1 = confusionchart(categorical(test_INS_output(:)),INS_outputPred(:));

[Nathaniel Porter]

What layers do you recommned I can use to improve my training in terms of the gap between my validation and training line