Fig. 4From: GACNNMDA: a computational model for predicting potential human microbe-drug associations based on graph attention network and CNN-based classifierAnalysis of the impact of hyperparameters on performance of GACNNMDA. The subfigures from (a) to (c) show the AUC values of related values of the learning rate of GAT, the dropout of GAT and the learning rate of CNN, respectivelyBack to article page