Why are the results of forward and predict very different in deep learning?

Illustration
Vascellari - 2022-11-11T10:17:58+00:00
Question: Why are the results of forward and predict very different in deep learning?

When I use the "dlnetwork" type deep neural network model to make predictions, the results of the two functions are very different, except that using the predict function will freeze the batchNormalizationLayer and dropout layers.While forward does not freeze the parameters, he is the forward transfer function used in the training phase.     From the two pictures above, there are orders of magnitude difference in the output of the previous 10 results. Where does the problem appear?

Related Questions

  • How to create an attention layer for deep learning networks?
  • Nvidia NVLink in MATALB using multiple Geforce 2080 ti cards
  • How can I fit a Gumbel Maximum distribution using both
  • How to predict next value using time series?
  • How are the functions "train" and "trainNetwork" different underneath?

    • Expert Answer

    Profile picture of Kshitij Singh Kshitij Singh answered . 2025-11-20

    I ran into this exact problem, and I think I found a solution, I'll discover it when my model finishes training...
     
    As others said before, the problem occurs because batchNorms behave differently in forward() and predict(). But there is still a problem here: if you trained your model (forward), it should have converged to a solution that works well in inference (predict), but it doesn't. Something is wrong in the training too.
     
    What is wrong is that batchNorms don't update parameters the same way as other layers through (adam/rmsprop/sgdm)update functions. They update through the State property of the dlnetwork object. Consider the code:
     
    [gradients,loss] = dlfeval(@modelGradients,dlnet,dlX,Ylabel);
[dlnet,otherOutputs]=rmspropupdate(dlnet,gradients,otherInputs);
function [gradients,loss] = modelGradients(dlnet,dlX,Ylabel)
Y=forward(dlnet,dlX);
loss=myLoss(Y,Ylabel);
gradients=dlgradient(loss,dlnet.Learnables);
end
The code above is wrong if you have batchNorms, it won't update them. The batchNorms are updated through the State property returnet from forward and assigned to dlnet:
[gradients,state,loss] = dlfeval(@modelGradients,dlnet,dlX,Ylabel);
dlnet.State=state; % THIS!!!
[dlnet,otherOutputs]=rmspropupdate(dlnet,gradients,otherInputs);
function [gradients,state,loss] = modelGradients(dlnet,dlX,Ylabel)
[Y,state]=forward(dlnet,dlX); % THIS!!!
loss=myLoss(Y,Ylabel);
gradients=dlgradient(loss,dlnet.Learnables);
end
    Now that dlnet has a State property updated at every forward() call, the batchNorms are updated and your model should converge to a solution that works for predict().
     
    I would also like caling MathWorks attention that this detail is only present in documentation in ONE example of GAN networks (in spite of the omnipresence of batchNorm layers in deep learning models) and is never mentioned explicitly.


    Not satisfied with the answer ?? ASK NOW

    Get a Free Consultation or a Sample Assignment Review!

    Upload Assignment Talk to Expert