Skip to main content

Table 1 Performance comparison for unconditional molecular graph generation

From: Efficient learning of non-autoregressive graph variational autoencoders for molecular graph generation

DatasetMetricGraphVAE [28]MolGAN [29]Proposed
A1A2Full
QM9Validity0.6100.9810.5500.9650.945
Uniqueness0.4090.1040.2930.2750.343
Novelty0.8500.9420.6120.7370.806
G-mean0.5960.4580.4620.5810.639
ZINCValidity0.1400.0170.0080.9260.919
Uniqueness0.3160.2010.9490.6140.762
Novelty1.0001.0001.0001.0001.000
G-mean0.3540.1510.1970.8280.888
  1. Best score for each metric is given in italic