# NequIP Benchmark Trajectories: LiPS, Li4P2O7 Melt-Quench and Formate on Cu, ColabFit Conversion

## Facts

- Provider: Simon Batzner, Albert Musaelian, Boris Kozinsky and coauthors (Harvard); ColabFit conversion
- License: CC-BY-4.0 (https://archive.materialscloud.org/api/records/ycbvx-knj69)
- Size: 6 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/colabfit/NequIP_NC_2022
- Access: public, readable without an account
- DOI: https://doi.org/10.24435/materialscloud:s0-5n
- Category: Chemistry & Materials
- Source: https://doi.org/10.1038/s41467-022-29939-5
- Page: https://scigantic.com/archive/2b90bf6a-9d06-56f1-b2b8-21b4ea37d0a0
- Keywords: Chemistry & Materials, DFT, quantum chemistry, interatomic potentials, graph learning, benchmark, Parquet, Hugging Face

## About

The materials benchmark trajectories released with NequIP, the E(3)-equivariant graph neural network interatomic potential: ab initio MD of the superionic conductor Li6.75P3S11 (LiPS), a Li4P2O7 (LiPO) glass melt-quench, and heterogeneous catalysis of formate decomposition on a copper surface, all computed with plane-wave DFT (VASP). ColabFit counts 56,822 configurations (about 7.6 million atoms) over C, Cu, H, Li, O, P and S with energies and atomic forces. These sets are widely reused to compare equivariant potentials (NequIP, Allegro, MACE) on lithium diffusion and surface reactions. This is the ColabFit Exchange conversion: configurations sit in Parquet files under co/ (positions, cell, species, computed properties and metadata per row) with dataset-level information in ds.parquet; read them with pyarrow, pandas, polars or Hugging Face datasets following the ColabFit schema at materials.colabfit.org/docs. Published on Materials Cloud under CC-BY-4.0; cite Batzner et al., Nature Communications 13, 2453 (2022).

## Load the data

```python
from datasets import load_dataset
ds = load_dataset("colabfit/NequIP_NC_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/2b90bf6a-9d06-56f1-b2b8-21b4ea37d0a0
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/2b90bf6a-9d06-56f1-b2b8-21b4ea37d0a0
- Over MCP: `get_schema_card` with id `2b90bf6a-9d06-56f1-b2b8-21b4ea37d0a0` at https://api.scigantic.com/mcp

## Related topics

- [Parquet datasets](https://scigantic.com/topics/parquet)
- [Benchmark datasets](https://scigantic.com/topics/benchmark)
- [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)
- [Quantum chemistry datasets](https://scigantic.com/topics/quantum-chemistry)
- [Graph learning datasets](https://scigantic.com/topics/graph-learning)
