Si-GAP-18

The complete dataset used to train the Si-GAP-18 model from Machine Learning a General-Purpose Interatomic Potential for Silicon. The CUR algorithm was used to select representative structures from a larger dataset. Energy and force labels were calculated using the PW91 exchange-correlation functional as implemented in CASTEP (see II.B: Database of the paper).

>>> from load_atoms import load_dataset
>>> load_dataset("Si-GAP-18")
Si-GAP-18:
    structures: 2,475
    atoms: 171,815
    species:
        Si: 100.00%
    properties:
        per atom: (forces)
        per structure: (config_type, cutoff, energy, nneightol)

License

This dataset is licensed under the CC BY-NC-SA 4.0 license.

Citation

If you use this dataset in your work, please cite the following:

@article{Bartok-18-12,
    title = {
        Machine Learning a General-Purpose Interatomic
        Potential for Silicon
    },
    author = {
        Bart{\'o}k, Albert P. and Kermode, James and Bernstein,
        Noam and Cs{\'a}nyi, G{\'a}bor
    },
    year = {2018},
    journal = {Physical Review X},
    volume = {8},
    number = {4},
    pages = {041048},
}

Properties

Per-atom:

Property	Units	Type	Description
forces	eV/Å	ndarray(N, 3)	force vectors (DFT)

Per-structure:

Property	Units	Type	Description
energy	eV	float64	total structure energy (DFT)
config_type		str	category of structure

Miscellaneous information

Si-GAP-18 is imported as an InMemoryAtomsDataset:

Importer script for Si-GAP-18

from __future__ import annotations

from pathlib import Path
from typing import Iterator

import ase.io
from ase import Atoms
from load_atoms.database.backend import BaseImporter, rename, unzip_file
from load_atoms.database.internet import FileDownload
from load_atoms.progress import Progress


class Importer(BaseImporter):
    @classmethod
    def files_to_download(cls) -> list[FileDownload]:
        return [
            FileDownload(
                url="https://zenodo.org/record/1250555/files/libAtoms/silicon-testing-framework-v1.0.zip",
                expected_hash="da0462802df1",
                local_name="zip-file.zip",
            )
        ]

    @classmethod
    def get_structures(
        cls, tmp_dir: Path, progress: Progress
    ) -> Iterator[Atoms]:
        contents_path = unzip_file(tmp_dir / "zip-file.zip", progress)

        for structure in ase.io.iread(
            contents_path
            / "libAtoms-silicon-testing-framework-fc252cb/models/GAP/gp_iter6_sparse9k.xml.xyz"  # noqa: E501
        ):
            yield rename(
                structure,
                {
                    "DFT_force": "forces",
                    "dft_force": "forces",
                    "DFT_energy": "energy",
                    "dft_energy": "energy",
                },
            )

DatabaseEntry for Si-GAP-18

name: Si-GAP-18
year: 2018
description: |
    The complete dataset used to train the `Si-GAP-18 <https://zenodo.org/records/1250555>`_ model
    from `Machine Learning a General-Purpose Interatomic Potential for Silicon <https://doi.org/10.1103/PhysRevX.8.041048>`_.
    The CUR algorithm was used to select representative structures from a larger dataset.
    Energy and force labels were calculated using the PW91 exchange-correlation functional as implemented in :code:`CASTEP`
    (see :code:`II.B: Database` of the paper).
category: Potential Fitting
minimum_load_atoms_version: 0.2
citation: |
    @article{Bartok-18-12,
        title = {
            Machine Learning a General-Purpose Interatomic
            Potential for Silicon
        },
        author = {
            Bart{\'o}k, Albert P. and Kermode, James and Bernstein,
            Noam and Cs{\'a}nyi, G{\'a}bor
        },
        year = {2018},
        journal = {Physical Review X},
        volume = {8},
        number = {4},
        pages = {041048},
    }
license: CC BY-NC-SA 4.0
representative_structure: 1283
per_atom_properties:
    forces:
        desc: force vectors (DFT)
        units: eV/Å
per_structure_properties:
    energy:
        desc: total structure energy (DFT)
        units: eV
    config_type:
        desc: category of structure


# TODO: remove after Dec 2024
# backwards compatability: unused as of 0.3.0
files:
     - url: https://zenodo.org/record/1250555/files/libAtoms/silicon-testing-framework-v1.0.zip
       hash: 97eb063f9655
processing:
     - UnZip
     - SelectFile:
           file: libAtoms-silicon-testing-framework-fc252cb/models/GAP/gp_iter6_sparse9k.xml.xyz
     - ReadASE
     - Rename:
           DFT_force: forces
           dft_force: forces
           DFT_energy: energy
           dft_energy: energy