Skip to main content
. 2019 Dec 3;11:74. doi: 10.1186/s13321-019-0397-9

Table 2.

The performance of heteroencoder in both the training and test sets

Dataset # compounds Validity (%) Reconstruction error (%)
Training set 974,105 99 18
Test set 10,823 98 20

Percent of valid SMILES strings generated by the decoder (validity), percent of molecules not reconstructed correctly from valid SMILES (reconstruction error)