A Novel Molecular Representation Learning for Molecular Property Prediction with a Multiple SMILES-Based Augmentation

Table 1

The description of public molecular datasets, including data type, task type, and the number of compounds before and after augmentation.

Dataset	Data type	Task type	Compound	Compound after augmentation
ESOL	SMILES	Regression	1127	6762
Lipophilicity	SMILES	Regression	4200	25200
FreeSolv	SMILES	Regression	642	3852
HIV	SMILES	Classification	41127	69987
BACE	SMILES	Classification	1513	9078