Could anyone help me how to solve the error stating Y must be a vector of categorical responses or a cell array of categorical response sequences.

In my code Y is {3x1} cell

where Y{1,1}=12x1 double

Y{2,1}=12x2 double

Y{3,1}=12x3 double

Could anyone help me how should I change Y to a cell array of categorical response sequences.

Answer 1

[Walter Roberson]

Y = cellfun(@categorical, Y, 'uniform', 0)

Answer 2

[jaah navi]

With the help of your command I can execute Y(in my code Y is YTrain) but when i tried to implement in my neural network code, error exists as "Error using nnet.internal.cnn.util.NetworkDataValidator/assertValidSequenceResponse (line 349)

Invalid training data. Responses must be a vector of categorical responses, or a cell array of categorical response

sequences" for the following lines.

YTrain = cellfun(@categorical, YTrain, 'uniform', 0)

net = trainNetwork(XTrain,YTrain,layers,options);

Could you please help me to solve it.

[Walter Roberson]

Sorry, No, that is not valid training information.

If XTrain is a numeric array, then YTrain must be a categorical vector with number of entries equal to the number of samples in XTrain.

If XTrain is a cell array of numeric arrays, then YTrain must be a cell array, and each entry must be a categorical vector with number of entries each to the number of samples in the corresponding XTrain cell.

YTrain corresponds to class labels: each training sample (no matter how many features it has) corresponds to exactly one class.

Your Y{2,1} is 12 x 2. That would imply that you have two classes for each of 12 samples, or else that you have 12 classes for each of two samples. Either way is not permitted.