Skip to main content

Table 3 Results of GuacaMol distribution-learning benchmarks

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

MetricSMILES-basedGraph-based
LSTMVAEAAEORGANGraphMCTSProposed
Validity0.9590.8700.8220.3791.0000.830
Uniqueness1.0000.9991.0000.8411.0000.944
Novelty0.9120.9740.9980.6870.9941.000
KLD0.9910.9820.8860.2670.5220.554
FCD0.9130.8630.5290.0000.0150.016