# Pure Magnesium (Mg) DFT Reference Set for Neural Network Potential Fitting (Yin, Stricker and Curtin 2020, Materials Cloud), ColabFit Conversion

## Facts

- Provider: Binglun Yin, Markus Stricker and W. A. Curtin; ColabFit conversion
- License: CC-BY-4.0 (https://archive.materialscloud.org/records/58xdx-v1x09)
- Size: 6 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/colabfit/pure_magnesium_DFT_PRM2020
- Access: public, readable without an account
- DOI: https://doi.org/10.24435/materialscloud:8f-1s
- Category: Chemistry & Materials
- Source: https://doi.org/10.1103/PhysRevMaterials.4.103602
- Page: https://scigantic.com/archive/7e992297-905c-5dcc-96da-27e70439c0fa
- Keywords: Chemistry & Materials, DFT, interatomic potentials, Parquet, Hugging Face

## About

VASP DFT calculations for pure hcp magnesium designed for fitting a Behler-Parrinello neural network potential for mechanics applications. The set follows a standard recipe for mechanics-oriented interatomic potentials and includes bulk properties, generalized stacking fault energies and stable stacking faults, decohesion and relaxed surfaces, dimers, corner and rod geometries, and vacancy formation. Configuration sets in the conversion follow these groups. ColabFit's card counts 405 configurations and 10,730 atoms (Mg), labelled with energy, atomic forces; the repository is about 0.8 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 10.24435/materialscloud:8f-1s) is released under CC-BY-4.0; cite Yin, Stricker and Curtin, Physical Review Materials 4, 103602 (2020).

## Load the data

```python
from datasets import load_dataset
ds = load_dataset("colabfit/pure_magnesium_DFT_PRM2020")
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/7e992297-905c-5dcc-96da-27e70439c0fa
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/7e992297-905c-5dcc-96da-27e70439c0fa
- Over MCP: `get_schema_card` with id `7e992297-905c-5dcc-96da-27e70439c0fa` 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)
