Skip to main content

Table 2 The performance of heteroencoder in both the training and test sets

From: A de novo molecular generation method using latent vector based generative adversarial network

Dataset# compoundsValidity (%)Reconstruction error (%)
Training set974,1059918
Test set10,8239820
  1. Percent of valid SMILES strings generated by the decoder (validity), percent of molecules not reconstructed correctly from valid SMILES (reconstruction error)