chandar-lab/ZINC_22
ZINC_22 Pretraining Dataset Dataset Description This dataset is derived from the ZINC-22 database (~70B synthesizable compounds as of Sept 2024) and was prepared for large-scale pretraining of molecular language models. We randomly sampled 1.5 billion molecules using a stratified heavy-atom count split (4–49 atoms) to ensure coverage of diverse chemical sizes.All molecules were deduplicated to remove repeats, canonicalized in SMILES format, and converted into… See the full description on the dataset page: https://huggingface.co/datasets/chandar-lab/ZINC_22.
| Author / organization | chandar-lab |
|---|---|
| Downloads | 1.6K (as reported by Hugging Face) |
| Likes | 1 |
| Created | 21 May 2024 |
| Last updated at source | 20 August 2025 |
Popularity, last 3 days
1,628 now (-403 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
Recent changes
- 16 Oct 2026 Description: ZINC_22 Pretraining Dataset Dataset Description This dataset is derived from the ZINC-22 database (~70B synthesizable compounds as of Sept 2024) and was prepared for large-scale pretraining of molecular language models. We randomly sampled → ZINC_22 Pretraining Dataset Dataset Description This dataset is derived from the ZINC-22 database (~70B synthesizable compounds as of Sept 2024) and was prepared for large-scale pretraining of molecular language models. We randomly sampled
Changes detected by Silytics between refreshes of the source.
More from chandar-lab
- chandar-lab/UR100PDataset
- chandar-lab/CoPePDataset
- chandar-lab/ZINC_250kDataset
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.