Skip to main content
. 2021 Jul 30;19:4593–4602. doi: 10.1016/j.csbj.2021.07.032

Fig. 3.

Fig. 3

Schematic overview of the neural network architecture. From the last hidden layer, more specifically after the batch-normalization has been applied, the fingerprint is extracted. An additional linear transformation plus activation function are applied to the fingerprint before the network makes a prediction.