What value does trainNetwork expect from read(ds), when hasdata(ds)= false?

** Edit I figured out my issue. read(ds) should throw an error when hasdat(ds) = false. The error needs to include an error id with the mnemonic field 'noMoreData'. I am not exactly sure what component field should be, but it shouldn't be 'MATLAB' because my datastore script isn't part of a Matlab toolbox. I used the name of my datastore class as the component field. I was confused about the fields in matlab errors, so reading this article helped: https://www.mathworks.com/help/matlab/ref/error.html#bumt5nd-msgID. Here is the error message that worked for me:

error('myDatastore:noMoreData',['No more data to read.\nUse the reset ',...
                    'method to reset the datastore to the start of ' ,...
                    'the data. \nBefore calling the read method, ',...
                    'check if data is available to read ',...
                    'by using the hasdata method.']);

Original question:

I built a custom datastore, and want to use it with a neural network. However, I get an error using trainNetwork when it reaches the end of the first epoch. My read(ds) function returns a cell array, where the first 4 cells are double arrays and the last cell is a categorical. The error I get is "Input four expected a cell but received a double instead."

I developed the datastore based on this documentation, as well as looking at the code for arrayDatastore and tabularTextDatastore. The example for the custom datastore is:

function tf = hasdata(myds)
% Return true if more data is available.
tf = hasfile(myds.FileSet);
end
function [data,info] = read(myds)
% Read data and information about the extracted data.
if ~hasdata(myds)
    error(sprintf(['No more data to read.\nUse the reset ',...
        'method to reset the datastore to the start of ' ,...
        'the data. \nBefore calling the read method, ',...
        'check if data is available to read ',...
        'by using the hasdata method.']))
    
end

fileInfoTbl = nextfile(myds.FileSet);
data = MyFileReader(fileInfoTbl);
info.Size = size(data);
info.FileName = fileInfoTbl.FileName;
info.Offset = fileInfoTbl.Offset;

% Update CurrentFileIndex for tracking progress
if fileInfoTbl.Offset + fileInfoTbl.SplitSize >= ...
        fileInfoTbl.FileSize
    myds.CurrentFileIndex = myds.CurrentFileIndex + 1 ;
end


end 

The tabularTextDatastore readData function handles having no data like this:

% error early if no data available
if ~hasdata(ds)
    error(message('MATLAB:datastoreio:splittabledatastore:noMoreData'));
end

ArrayDatastore does this:

if ~arrds.hasdata()
    msgid = "MATLAB:io:datastore:array:read:NoMoreData";
    error(message(msgid));
end

I have tried each of these strategies in my read function (literally copy-pasting those pieces of code, but always get the same error. I can't seem to find the code underlying trainNetwork to see how it handles it. Does anyone know what trainNetwork expects from the read(ds) function when you've run out of data?