view article
Figure 2
The CNN, composed of an input layer, three pairs of convolutional and pooling layers, two fully connected layers, and an output layer. Each layer has a number of neurons that collect information from the previous layer. This information is converted into a specific value using an activation function to be transferred to the neurons in the next layer. The filter size was halved from one layer to the next.

IUCrJ
Volume 4| Part 4| July 2017| Pages 486-494
ISSN: 2052-2525
https://doi.org/10.1107/S205225251700714X