Invalid validation data table. For networks with feature input, predictors must be numeric arrays, where each variable of the table corresponds to one feature.

Hello, I am new to matlab. I want to ask what to do if there is an invalid validation data table error. This is my code

filename = "Data1.txt";

tbl = readtable(filename,'TextType','String');

labelName = "output";

tbl = convertvars(tbl,labelName,'categorical');

head(tbl)

categoricalInputNames = ["class" "fractaldimension"];

tbl = convertvars(tbl,categoricalInputNames,'categorical');

for i = 1:numel(categoricalInputNames)

name = categoricalInputNames(i);

oh = onehotencode(tbl(:,name));

tbl = addvars(tbl,oh,'After',name);

tbl(:,name) = [];

end

tbl = splitvars(tbl);

head(tbl)

classNames = categories(tbl{:,labelName});

numObservations = size(tbl,1);

numObservationsTrain = floor(0.7*numObservations);

numObservationsValidation = floor(0.15*numObservations);

numObservationsTest = numObservations - numObservationsTrain - numObservationsValidation;

idx = randperm(numObservations);

idxTrain = idx(1:numObservationsTrain);

idxValidation = idx(numObservationsTrain+1:numObservationsTrain+numObservationsValidation);

idxTest = idx(numObservationsTrain+numObservationsValidation+1:end);

tblTrain = tbl(idxTrain,:);

tblValidation = tbl(idxValidation,:);

tblTest = tbl(idxTest,:);

numFeatures = size(tbl,2) - 1;

numClasses = numel(classNames);

layers = [

featureInputLayer(numFeatures,'Normalization', 'zscore')

fullyConnectedLayer(83)

batchNormalizationLayer

reluLayer

fullyConnectedLayer(numClasses)

softmaxLayer

classificationLayer];

miniBatchSize = 16;

options = trainingOptions('adam', ...

'MiniBatchSize',miniBatchSize, ...

'Shuffle','every-epoch', ...

'ValidationData',tblValidation, ...

'Plots','training-progress', ...

'Verbose',false);

net = trainNetwork(tblTrain,labelName,layers,options);

Invalid validation data table. For networks with feature input, predictors must be numeric arrays, where each

variable of the table corresponds to one feature.

Answer 1

[Rohit]

Hi Adib,

As mentioned in this documentation: https://www.mathworks.com/help/deeplearning/ref/trainingoptions.html , you need to specify the validation data as a datastore, table, or the cell array {predictors,responses}, where predictors contains the validation predictors and responses contains the validation responses.

So, you need to modify code as shown below to get rid of error and start the training.

options = trainingOptions('adam', ...
    'MiniBatchSize',miniBatchSize, ...
    'Shuffle','every-epoch', ...
    'ValidationData',{tblValidation,tblValidation(:,labelName)} ,... % passing validation date as cell array of predictors and responses
    'Plots','training-progress', ...
    'Verbose',false);