Supplementary Material for MOSES: Combining Masked Oversampling with Ensemble Learning for Hierarchical Multi-label Classification in ChEBI

Supplementary material for _Cappellani et al.: MOSES: Combining Masked Oversampling with Ensemble Learning for Hierarchical Multi-label Classification in ChEBI_ (in submission) This contains the model checkpoints used during the evaluation and the ChEBI50 v241 dataset with the applied train/validation/test split. The checkpoints and dataset were created with [ChEB-AI](https://github.com/ChEB-AI/python-chebai/tree/feature/resampling3).