# Open Catalyst 2022 (OC22) IS2RE Training Set, ColabFit Conversion

## Facts

- Provider: Meta FAIR, Carnegie Mellon University, University of Toronto (Tran et al.); ColabFit conversion
- License: CC-BY-4.0 (https://raw.githubusercontent.com/facebookresearch/fairchem/main/docs/catalysts/datasets/oc22.md)
- Size: 96 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/colabfit/OC22-IS2RE-Train
- Access: public, readable without an account
- DOI: https://doi.org/10.60732/722bcab6
- Category: Chemistry & Materials
- Source: https://doi.org/10.1021/acscatal.2c05426
- Page: https://scigantic.com/archive/9bcb75b0-d5d8-562c-a76b-8b09e8c3608c
- Keywords: Chemistry & Materials, DFT, Parquet, Hugging Face

## About

Training trajectories for the initial structure to relaxed energy (IS2RE) task of Open Catalyst 2022 (OC22), the oxide electrocatalyst companion to OC20 from Meta FAIR, Carnegie Mellon and the University of Toronto. OC22 covers adsorbates on oxide surfaces relevant to the oxygen evolution reaction and was computed with DFT+U (PBE) in VASP. Configurations carry total energy and atomic forces. ColabFit's card counts 7,861,269 configurations and about 634 million atoms; the repository uses about 34 GB (measured 2026-11-14). The OC22 S2EF training split is held separately; this row adds the IS2RE relaxation set. This is the ColabFit Exchange conversion: configurations are Parquet files under co/ (structure, positions, cell, computed properties and metadata per row) with dataset-level information in ds.parquet; read them with pyarrow, pandas, polars or Hugging Face datasets, following the ColabFit configuration schema at materials.colabfit.org/docs. Upstream OC22 is licensed CC-BY-4.0 by Meta; cite Tran et al., ACS Catalysis 2023.

## Load the data

```python
from datasets import load_dataset
ds = load_dataset("colabfit/OC22-IS2RE-Train")
print(ds)
```

## Data structure

File types in sampled files: Parquet

Example file: `co/co_0.parquet`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/9bcb75b0-d5d8-562c-a76b-8b09e8c3608c
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/9bcb75b0-d5d8-562c-a76b-8b09e8c3608c
- Over MCP: `get_schema_card` with id `9bcb75b0-d5d8-562c-a76b-8b09e8c3608c` at https://api.scigantic.com/mcp

## Related topics

- [Parquet datasets](https://scigantic.com/topics/parquet)
- [Chemistry & Materials datasets](https://scigantic.com/topics/chemistry-and-materials)
- [DFT datasets](https://scigantic.com/topics/dft)
