# Li-Mo-Ni-Ti Oxide (Li8Mo2Ni7Ti7O32) AIMD Structures for ANN Potentials with Taylor-Expanded Forces (Cooper et al. 2020, Materials Cloud), ColabFit Conversion

## Facts

- Provider: April M. Cooper, Johannes Kästner, Alexander Urban and Nongnuch Artrith; ColabFit conversion
- License: CC-BY-4.0 (https://archive.materialscloud.org/records/tcnxt-2mn21)
- Size: 4 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/colabfit/HO_LiMoNiTi_NPJCM_2020_LiMoNiTi_train
- Access: public, readable without an account
- DOI: https://doi.org/10.24435/materialscloud:2020.0037/v1
- Category: Chemistry & Materials
- Source: https://doi.org/10.1038/s41524-020-0323-8
- Page: https://scigantic.com/archive/2043dd1f-b6d0-5c58-91f8-c6c15d6d48da
- Keywords: Chemistry & Materials, molecular dynamics, quantum chemistry, Parquet, Hugging Face

## About

Training structures of the rock-salt transition-metal oxide Li8Mo2Ni7Ti7O32, from a study that adds atomic force information to artificial neural network (ANN) potential training through a Taylor expansion of the total energy, cutting the cost of training. Structures come from ab initio molecular dynamics (AIMD) in VASP with the SCAN meta-GGA functional; the original deposit stores them in XSF format for the aenet package and also holds water cluster and bulk water sets (revPBE+D3), which this row does not include. ColabFit's card counts 824 configurations and 46,144 atoms (Li, Mo, Ni, O, Ti), labelled with energy, atomic forces; the repository is about 6.7 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 original deposit (DOI 10.24435/materialscloud:2020.0037/v1) is released under CC-BY-4.0; cite Cooper et al., npj Computational Materials 6, 54 (2020), doi:10.1038/s41524-020-0323-8.

## Load the data

```python
from datasets import load_dataset
ds = load_dataset("colabfit/HO_LiMoNiTi_NPJCM_2020_LiMoNiTi_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/2043dd1f-b6d0-5c58-91f8-c6c15d6d48da
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/2043dd1f-b6d0-5c58-91f8-c6c15d6d48da
- Over MCP: `get_schema_card` with id `2043dd1f-b6d0-5c58-91f8-c6c15d6d48da` 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)
- [Molecular dynamics datasets](https://scigantic.com/topics/molecular-dynamics)
- [Quantum chemistry datasets](https://scigantic.com/topics/quantum-chemistry)
