Silica (SiO2) SCAN DFT Training Set for a Machine-Learned Interatomic Potential (Erhard et al. 2022), ColabFit Conversion
- Provider
- Linus C. Erhard, Jochen Rohrer, Karsten Albe, Volker L. Deringer; ColabFit conversion
- License
- CC-BY-4.0 (source)
- Size
- 6 files
- Format
- Parquet (in sampled files)
- Stored at
- Hugging Face (https://huggingface.co/datasets/colabfit/Silica_NPJCM_2022)
- Access
- Free and public. No account with the data provider is needed.
- DOI
- 10.60732/c2bee5fa
- Category
- Chemistry & Materials
- Source
- https://doi.org/10.1038/s41524-022-00768-w
About this dataset
Training data for the machine-learned interatomic potential (MLIP) for silica (SiO2) of Erhard, Rohrer, Albe and Deringer. Initial DFT runs compared LDA, PBE and PBEsol in GPAW/ASE with SCAN in VASP (PAW method); SCAN performed best and every structure was recomputed with SCAN at a 900 eV cutoff and 0.23 1/Angstrom k-spacing. Each configuration carries energy, atomic forces and the Cauchy stress. The ColabFit card counts 3,074 configurations and about 268,000 Si and O atoms; the repository is about 23 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 released under CC-BY-4.0; cite Erhard et al., npj Computational Materials 2022.
How to load this dataset in Python
from datasets import load_dataset
ds = load_dataset("colabfit/Silica_NPJCM_2022")
print(ds)Data structure
File types in sampled files: Parquet
Files
Example file: co/co_0.parquet
Analyze this dataset without downloading it
On Scigantic, Silica (SiO2) SCAN DFT Training Set for a Machine-Learned Interatomic Potential (Erhard et al. 2022), ColabFit Conversion opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.
This page is also available as Markdown and as JSON.
Related Chemistry & Materials datasets
- MAD (Massive Atomic Diversity) Training Split, ColabFit Conversion
- ANI-2x Conformers Recomputed at wB97M-V/def2-TZVPP, ColabFit Conversion
- Graphene Field Effect Transistor Urea Sensor
- Zirconium-Tin (Zr-Sn) Alloy DFT Training Set for MTP and MEAM Potentials (Mei et al. 2024), ColabFit Conversion
- ANI-1: 20 Million DFT Energies for Non-Equilibrium Small Molecules, ColabFit Conversion
- Solvated Protein Fragments (PhysNet Benchmark, revPBE-D3), ColabFit Conversion
- Acetylacetone (acac) Proton-Transfer Benchmark: 300 K MD Training Set with PBE-D3 Forces (BOTNet, Batatia et al.), ColabFit Conversion
- Plate reader instrument output with ASM: 140 exports from Agilent Gen5, Agilent Gen5 Image, BMG Labtech SMART Control, BMG Labtech MARS and 10 more