We've talked a lot about different methods to validate the performance of your neural networks. So far, we've only had to deal with datasets that fit in memory. But this is almost never the case in production scenarios, since you need a lot of data to train a neural network. In this section, we'll discuss how to use the different metrics on out-of-memory datasets.
Measuring performance for out-of-memory datasets
Measuring performance when working with minibatch sources
When you use a minibatch data source, you need a slightly different setup for the loss and metric. Let's go back and review how you can set up training using a minibatch source and extend it with metrics to validate the model. First, we...
