My training data is in a cell array whilst my labels are a categorical array although I still get invalid training data error

I am trying to experiment with our CNN structure using a sequence input, with a dataset of 360x1 cell consisting of 64x769 arrays

64 = Channels

Dataset:

And this is our label data in a categorical array:

and this is the error I am getting

Error using trainNetwork (line 184)

Invalid training data. Responses must be a cell array of categorical response sequences.

Although I am using a categorical array it still outputs this error which is very confusing.

Answer 1

[Anshika Chaurasia]

Hi,

As mention in error: Responses should be cell array of categorical datatype. Hence, the response should like this:

For more information refer to following documentations:

Sequence-to-Sequence Classification Using 1-D Convolutions - MATLAB & Simulink (mathworks.com)

https://www.mathworks.com/help/deeplearning/ref/trainnetwork.html#mw_d0b3a2e4-09a0-42f9-a273-2bb25956fe66