Java Deep Learning Projects

Hidden layer 2

The number of input neurons is equal to the output of hidden layer 1. Then the number of outputs is an arbitrary value. Again we set a smaller value considering very few inputs and features:

 DenseLayer hidden_layer_2 = new DenseLayer.Builder()                .weightInit(WeightInit.XAVIER)                .activation(Activation.RELU)                .nIn(32).nOut(16)                .build();