Skip to main content
. 2018 Jun 29;8:9879. doi: 10.1038/s41598-018-28222-2

Figure 4.

Figure 4

The structure of DNN regressor is determined by the grid search algorithm. It has 2 hidden layers having ‘Adam’ optimizer, ‘ReLU’ activation function with 0.1 learning rate and 1000 epochs for training iteration. The number of neurons is the same with the input features.