Skip to main content

Table 10 Comparison of training time (minutes) between ML-GAT, ML-GCN, GrAMME-SG and GrAMME-Fusion

From: Graph convolutional and attention models for entity classification in multilayer networks

Network ML-GAT ML-GCN GrAMME-SG GrAMME-Fusion
Balance 0.2935 0.1270 12.9228 12.7558
CKM-Social 0.2465 0.1275 3.0005 2.6893
DKPol 0.2626 0.1798 7.2329 6.2134
Leskovec-Ng 0.2632 0.1269 3.1790 2.4235
Vickers 0.2663 0.1255 0.3158 0.3242
Koumbia-2-mpx 0.2705 0.1727 52.3738 44.1908
  1. The training was performed on Google Colab with Tesla T4 GPU with the same hyperparameter setting described in "Experimental settings" section