Designing output layers for the neural network model

Output layer design requires an understanding of the expected output. We have CSV data as input, and the output layer relies on the number of labels in the dataset. Output layers are the place where the actual prediction is formed based on the learning process that happened in the hidden layers.

In this recipe, we will design output layers for the neural network.

How to do it...

1. Determine the incoming/outgoing connections. Set the following:

incoming neurons = outgoing neurons from preceding hidden layer.
 outgoing neurons = number of labels
 

2. Configure the output layer for the neural network:

builder.layer(new OutputLayer.Builder(new LossMCXENT(weightsArray)).nIn(incomingConnectionCount...