# Open Catalyst 2022 (OC22) S2EF Out-of-Domain Validation Set, ColabFit Conversion

## Facts

- Provider: Meta FAIR and Carnegie Mellon University (Tran et al.); ColabFit conversion
- License: CC-BY-4.0 (https://raw.githubusercontent.com/facebookresearch/fairchem/main/docs/catalysts/datasets/oc22.md)
- Size: 4 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/colabfit/OC22-S2EF-Validation-out-of-domain
- Access: public, readable without an account
- Category: Chemistry & Materials
- Source: https://doi.org/10.1021/acscatal.2c05426
- Page: https://scigantic.com/archive/92006771-0667-5e00-92ae-1e393377a474
- Keywords: Chemistry & Materials, Parquet, Hugging Face

## About

Out-of-domain validation configurations for the Structure to Total Energy and Forces task of Open Catalyst 2022 (OC22), the Meta FAIR and Carnegie Mellon oxide electrocatalyst dataset that complements OC20, which had no oxide surfaces. Systems are adsorbates relevant to the oxygen evolution reaction (O, OH, OOH and related C/H/N/O species) on metal oxide slabs, computed with VASP using PBE+U and spin polarisation, with total energies and atomic forces; the out-of-domain split holds oxide compositions not present in training. ColabFit counts 457,249 configurations and about 36.9 million atoms over 52 elements. This is the ColabFit Exchange conversion: configurations sit in Parquet files under co/ (positions, cell, species, 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 schema at materials.colabfit.org/docs. Meta releases OC22 under CC-BY-4.0; cite Tran et al., ACS Catalysis 2023.

## Load the data

```python
from datasets import load_dataset
ds = load_dataset("colabfit/OC22-S2EF-Validation-out-of-domain")
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/92006771-0667-5e00-92ae-1e393377a474
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/92006771-0667-5e00-92ae-1e393377a474
- Over MCP: `get_schema_card` with id `92006771-0667-5e00-92ae-1e393377a474` 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)
