After training a neural network with Batch Norm, at test time, to evaluate the neural network on a new example you should:

  1. After training a neural network with Batch Norm, at test time, to evaluate the neural network on a new example you should:

 Skip the step where you normalize using μ and 

    •  since a single test example cannot be normalized.

 

    •  If you implemented Batch Norm on mini-batches of (say) 256 examples, then to evaluate on one test example, duplicate that example 256 times so that you’re working with a mini-batch the same size as during training.

 

    •  Use the most recent mini-batch’s value of μ and  to perform the needed normalizations.

 

      •  Perform the needed normalizations, use μ and  estimated using an exponentially weighted average across mini-batches seen during training.

Get All Week Quiz Answer:

Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and Optimization Coursera Quiz Answer

Similar Posts