Fig. 1From: Inductive transfer learning for molecular activity prediction: Next-Gen QSAR Models with MolPMoFiTScheme illustrating the MolPMoFiT Architecture: During the fine-tuning, learned weights are transferred between models. Vocab Size corresponds to the number of unique characters tokenized (see “Molecular representation” section) from SMILES in a data set. The stage of Task Specific Molecular Structure Prediction Model fine-tuning is optionalBack to article page