How can i train Stacked Autoencoders with cross-validation?

Hi to everyone

I m trying to implement a network with stacked autoencoders, same as the example with the wine

https://au.mathworks.com/help/deeplearning/ug/construct-deep-network-using-autoencoders.html

I want to perform cross validation but i have problem with the accelaration because the predict

function doesn't work on me.

In the end, i train the deep network on the wine data

deepnet = train(deepnet,X,T);

Any idea how can i perform cv in this example?