# HDNNP Water: DFT Energies and Forces of Water for Four Functionals (Morawietz and Behler), ColabFit Conversion

## Facts

- Provider: Tobias Morawietz and Jörg Behler; ColabFit conversion
- License: CC-BY-4.0 (https://zenodo.org/records/2634097)
- Size: 6 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/colabfit/HDNNP_H2O
- Access: public, readable without an account
- DOI: https://doi.org/10.60732/4f9f05e5
- Category: Chemistry & Materials
- Source: https://doi.org/10.1073/pnas.1602375113
- Page: https://scigantic.com/archive/109409c8-3c85-5ec3-9381-cb481929bb78
- Keywords: Chemistry & Materials, DFT, Parquet, Hugging Face

## About

Training data for the high-dimensional neural network potentials (HDNNP) of water that Morawietz, Behler and coworkers used to show how van der Waals interactions set the density maximum of liquid water (PNAS 2016). About 28,000 water configurations are split into four sets, one per exchange-correlation functional, each with DFT energies and atomic forces, so the same structures can be compared across functionals. The ColabFit card counts 14,537 unique configurations and about 1.52 million H and O atoms; the repository is about 258 MB (measured 2026-11-19). This is the ColabFit Exchange conversion: configurations are Parquet files under co/ (structure, positions, cell, computed properties and metadata per row), configuration sets under cs/ with their mapping in cs_co_map/, 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 Morawietz et al., PNAS 2016.

## Load the data

```python
from datasets import load_dataset
ds = load_dataset("colabfit/HDNNP_H2O")
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/109409c8-3c85-5ec3-9381-cb481929bb78
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/109409c8-3c85-5ec3-9381-cb481929bb78
- Over MCP: `get_schema_card` with id `109409c8-3c85-5ec3-9381-cb481929bb78` 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)
