Skip to main content
. 2018 Oct 15;8:15264. doi: 10.1038/s41598-018-33654-x

Figure 2.

Figure 2

Network architecture of Deep-RBPPred. Deep-RBPPred is a CNN network including 11 layers. Convolution Layer and Max Pooling Layer are designed to automatically process the feature. Local Response Normalization layer and Dropout Layer are designed to avoid over-fitting. Softmax layer is used to classify the protein with probability score. The batch size is assigned to 200. The learning rate is set to 0.0001.