Fig. 5From: Accurate classification of membrane protein types based on sequence and evolutionary information using deep learningThe structure of autoencoder. It consists of an input layer, five hidden layers and an output layer. The input layer has 537 units; the first two layers and the last two layers of the five hidden layers have 128, 64, 64, and 128 units, respectively, the number of nodes in the middle layer of the hidden layers is an additional hyperparameter obtained by training; and the output layer has 537 unitsBack to article page