How to create train data for 1d CNN

I try to reshape data for be the inpput of CNN started with reshape XTrain to 4d

but the error say "Error using categorical (line 358)

Could not find unique values in DATA using the UNIQUE function. Caused by:

Error using matlab.internal.math.uniqueCellstrHelper

Cell array input must be a cell array of character vectors."

my train data set is 2016 x 3 double

how I can get the input for CNN network (YTrain)

%% Reshaped input
XTrain = reshape(trainset,[1 2016 1 3]);

% Creat the labels 
YTrain = (trainset(:,3));
YTrain = num2cell(YTrain);
YTrain = categorical(YTrain); 

[KSSV]

What is your input and out put? What exactly is the targetted problem? You know for CNN inputs should be images/ matrices.

[Parichada Trairat]

my input consist of load demand, holidays and time sample

I want the load demand for output

the problem is prediction load demand

Answer 1

[Anshika Chaurasia]

Hi,

Refer to following documentation to know more about preparing data for 1D convolution:

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

Refer to following documentation for load forecasting:

Load Forecasting - MATLAB & Simulink (mathworks.com)

Electricity Load and Price Forecasting Webinar Case Study - File Exchange - MATLAB Central (mathworks.com)