Pre-processed CrossDocked2020 dataset containing raw receptor PDB and ligand
SDF.gz files, organized for efficient loading.
Unique pairs: 2,533,288
Unique receptor PDB files: 24,533
Source types: cdonly, it0, it2, it2_redocked, it2_10p20n, mod_it2,
it2_completeset, it2_redocked_completeset, mod_it2_completeset
Fold splits: 3 folds (0, 1, 2) per source type category
(completeset categories only populate fold0)
Pose… See the full description on the dataset page: https://huggingface.co/datasets/kohbanye/crossdocked2020.