# Silver-Palladium (Ag-Pd) Alloy DFT Set Comparing MTP and SOAP-GAP Potentials (Rosenbrock et al. 2021), ColabFit Conversion

## Facts

- Provider: Conrad Rosenbrock, Konstantin Gubaev, Alexander Shapeev, Livia Partay, Noam Bernstein, Gabor Csanyi, Gus Hart (BYU); ColabFit conversion
- License: MIT (https://github.com/msg-byu/agpd/blob/master/LICENSE)
- Size: 6 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/colabfit/AgPd_NPJ_2021
- Access: public, readable without an account
- Category: Chemistry & Materials
- Source: https://doi.org/10.1038/s41524-020-00477-2
- Page: https://scigantic.com/archive/0b615264-0e2c-57bd-b0c0-2d33ee6d92e5
- Keywords: Chemistry & Materials, DFT, interatomic potentials, Parquet, Hugging Face

## About

VASP DFT energies, forces and virials for silver-palladium alloy structures, from a study that compared two machine-learned interatomic potentials for alloys, the Moment Tensor Potential (MTP) and the SOAP Gaussian Approximation Potential (SOAP-GAP), on the same actively learned Ag-Pd training data. The authors' repository holds the final GAP and MTP models with their training files, frozen-phonon configurations with phonopy output, and extra data: enumerated BCC and FCC configurations selected within the cluster expansion basis, a transition pathway and relaxed and unrelaxed seed structures. ColabFit's card counts 993 configurations and 7,260 atoms (Ag, Pd), labelled with energy, atomic forces and stress; about 1.7 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 msg-byu/agpd GitHub repository is released under the MIT License; cite Rosenbrock et al., npj Computational Materials (2021), doi 10.1038/s41524-020-00477-2.

## Load the data

```python
from datasets import load_dataset
ds = load_dataset("colabfit/AgPd_NPJ_2021")
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/0b615264-0e2c-57bd-b0c0-2d33ee6d92e5
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/0b615264-0e2c-57bd-b0c0-2d33ee6d92e5
- Over MCP: `get_schema_card` with id `0b615264-0e2c-57bd-b0c0-2d33ee6d92e5` 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)
