Mol-BERT: An Effective Molecular Representation with BERT for Molecular Property Prediction

Table 1

The detailed description of selected benchmark datasets.

Dataset	Category	Compound	Tasks	Task type	Split method
BBBP	Physiology	2,053	1	Binary	Scaffold
Tox21	Physiology	8,014	12	Multilabel	Scaffold
SIDER	Physiology	1,427	27	Multilabel	Scaffold
ClinTox	Physiology	1,491	2	Multilabel	Scaffold