# Systematic Magnesium (Mg) Structure Set for Machine Learning Potentials and Defects: DFT (Poul et al. 2023, Edmond), ColabFit Conversion

## Facts

- Provider: Marvin Poul; ColabFit conversion
- License: CC0-1.0 (https://edmond.mpg.de/api/datasets/:persistentId/?persistentId=doi:10.17617/3.A3MB7Z)
- Size: 6 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/colabfit/Mg_edmonds_2022
- Access: public, readable without an account
- DOI: https://doi.org/10.17617/3.A3MB7Z
- Category: Chemistry & Materials
- Source: https://doi.org/10.1103/PhysRevB.107.104103
- Page: https://scigantic.com/archive/4def1bde-ff92-5915-944b-37ba6ea5a2fe
- Keywords: Chemistry & Materials, DFT, interatomic potentials, Parquet, Hugging Face

## About

DFT energies, forces and stresses for a systematically generated set of magnesium atomic structures, built to train machine learning interatomic potentials and test them on defects in Mg. The original deposit is the data for "Systematic Atomic Structure Datasets for Machine Learning Potentials: Application to Defects in Magnesium" on Edmond, the Max Planck research data repository. ColabFit's card counts 16,746 configurations and 78,239 atoms (Mg), labelled with energy, atomic forces, cauchy stress; the repository is about 22.9 MB. Files are the ColabFit Exchange conversion: Parquet files under co/ hold one row per configuration (atomic numbers, positions, cell, computed properties and metadata), ds.parquet holds dataset-level information, and cs/ groups configurations into named sets where defined; read them with pyarrow, pandas, polars or Hugging Face datasets. The original deposit (DOI 10.17617/3.A3MB7Z) is released under CC0-1.0; cite Poul et al., Physical Review B 107, 104103 (2023), doi:10.1103/PhysRevB.107.104103.

## Load the data

```python
from datasets import load_dataset
ds = load_dataset("colabfit/Mg_edmonds_2022")
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/4def1bde-ff92-5915-944b-37ba6ea5a2fe
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/4def1bde-ff92-5915-944b-37ba6ea5a2fe
- Over MCP: `get_schema_card` with id `4def1bde-ff92-5915-944b-37ba6ea5a2fe` 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)
- [Interatomic potentials datasets](https://scigantic.com/topics/interatomic-potentials)
