# UNEP-v1: 16-Element Metal and Alloy DFT Training Set for a Unified NeuroEvolution Potential, ColabFit Conversion

## Facts

- Provider: Keke Song, Zheyong Fan and coauthors; ColabFit conversion
- License: CC-BY-4.0 (https://zenodo.org/records/10081676)
- Size: 4 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/colabfit/UNEP_v1_2023_train
- Access: public, readable without an account
- DOI: https://doi.org/10.60732/23c88dd7
- Category: Chemistry & Materials
- Source: https://doi.org/10.1038/s41467-024-54554-x
- Page: https://scigantic.com/archive/37f47a99-ade1-53a4-b77c-5a465e0a3cd9
- Keywords: Chemistry & Materials, molecular dynamics, DFT, Parquet, Hugging Face

## About

The training set of UNEP-v1, the first unified neuroevolution potential (NEP) for 16 elemental metals and their alloys, implemented in the GPUMD molecular dynamics code (Song et al., Nature Communications 2024). It covers Ag, Al, Au, Cr, Cu, Mg, Mo, Ni, Pb, Pd, Pt, Ta, Ti, V, W and Zr, labeled with DFT energies, atomic forces and stresses. The ColabFit card counts 104,799 configurations and about 6.8 million atoms; the repository is about 653 MB (measured 2026-11-19). The test set is a separate ColabFit repository. 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. The original Zenodo record is CC-BY-4.0; cite Song et al., Nature Communications 2024.

## Load the data

```python
from datasets import load_dataset
ds = load_dataset("colabfit/UNEP_v1_2023_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/37f47a99-ade1-53a4-b77c-5a465e0a3cd9
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/37f47a99-ade1-53a4-b77c-5a465e0a3cd9
- Over MCP: `get_schema_card` with id `37f47a99-ade1-53a4-b77c-5a465e0a3cd9` 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)
- [Molecular dynamics datasets](https://scigantic.com/topics/molecular-dynamics)
