CoolFace
Datasetpublic

materialyze/matpes

Dataset Summary Potential energy surface datasets with near-complete coverage of the periodic table are used to train foundation potentials (FPs), i.e., machine learning interatomic potentials (MLIPs) with near-complete coverage of the periodic table. MatPES is an initiative by the Materialyze Lab and the Materials Project to address critical deficiencies in such PES datasets for materials. Accuracy. MatPES is computed using static DFT calculations with stringent converegence… See the full description on the dataset page: https://huggingface.co/datasets/materialyze/matpes.

sourceHugging Facebsd-3-clauseupdated 3mo agoView on Hugging Face
6likes990downloads
Dataset Card

Dataset Description

Dataset Summary

Potential energy surface datasets with near-complete coverage of the periodic table are used to train foundation potentials (FPs), i.e., machine learning interatomic potentials (MLIPs) with near-complete coverage of the periodic table. MatPES is an initiative by the [Materialyze] Lab and the [Materials Project] to address critical deficiencies in such PES datasets for materials.

  1. 1.Accuracy. MatPES is computed using static DFT calculations with stringent converegence criteria. Please refer to the MatPESStaticSet in [pymatgen] for details.
  2. 2.Comprehensiveness. MatPES structures are sampled using a 2-stage version of DImensionality-Reduced Encoded Clusters with sTratified DIRECT sampling from a greatly expanded configuration of MD structures.
  3. 3.Quality. MatPES includes computed data from the PBE functional, as well as the high fidelity r2SCAN meta-GGA functional with improved description across diverse bonding and chemistries.

The initial v2025.1 release comprises ~400,000 structures from 300K MD simulations. The v2025.2 removes some duplicate structures and adds charge information. This dataset is much smaller than other PES datasets in the literature and yet achieves comparable or, in some cases, improved performance and reliability on trained FPs.

The dataset is provided as jsonl files to facilitate memory-efficient streaming. The original json files are retained for backwards compatibility.

MatPES is part of the MatML ecosystem, which includes the [MatGL] (Materials Graph Library) and [maml] (MAterials Machine Learning) packages, the [MatPES] (Materials Potential Energy Surface) dataset, and the [MatCalc] (Materials Calculator).

[Materialyze]: http://materialyze.ai [Materials Project]: https://materialsproject.org [M3GNet]: http://dx.doi.org/10.1038/s43588-022-00349-3 [CHGNet]: http://doi.org/10.1038/s42256-023-00716-3 [TensorNet]: https://arxiv.org/abs/2306.06482 [maml]: https://materialsvirtuallab.github.io/maml/ [MatGL]: https://matgl.ai [MatPES]: https://matpes.ai [MatCalc]: https://matcalc.ai