Open ghost opened 1 year ago
hi,excuse me ,there are 445 kinds of drugs in the gold standard data set. How to find the network between drugs and diseases, the network between drugs and side effects, and the network of similarity between drugs and drugs for given different drugs? The folder of drugNets in the original paper shows 10 different heterogeneous networks. Could you share the codes for the construction of these networks?
Hi, calculation of AUC and AUPR metrics already exists in code, how did you calculate the recall@K for the model evaluation? (recall@K is reported in paper but there is no code in implementation)