May 14, 2025 Dataset Open Access

Fiedler, Lenz; Cangi, Attila

Citation Style Language JSON Export

{
  "publisher": "Rodare", 
  "issued": {
    "date-parts": [
      [
        2025, 
        5, 
        14
      ]
    ]
  }, 
  "DOI": "10.14278/rodare.3746", 
  "type": "dataset", 
  "abstract": "<p><strong>Boron data set for machine learning applications</strong></p>\n\n<p>This dataset contains DFT inputs, outputs, LDOS data and bispectrum descriptor vectors for an &alpha;-rhombohedral boron cell of 144 atoms at room temperature and ambient mass density. All simulations have been performed at an LDOS converged k-grid of 4x4x4 k-points.</p>\n\n<p>This dataset contains one .zip file for each of its five type of data (bispectrum descriptors, LDOS, DFT inputs, DFT outputs and trained models).</p>\n\n<p><em>Authors:</em></p>\n\n<p>- Fiedler, Lenz (HZDR / CASUS)<br>\n- Cangi, Attila (HZDR / CASUS)</p>\n\n<p>Affiliations<em>:</em></p>\n\n<p>HZDR - Helmholtz-Zentrum Dresden-Rossendorf<br>\nCASUS - Center for Advanced Systems Understanding</p>\n\n<p><em>Dataset description</em></p>\n\n<p>- Total size: 26 GB<br>\n- System: B144<br>\n- Temperature(s): 298K<br>\n- Mass density(ies): 2.483 gcc<br>\n- Crystal Structure: amorphous (material mp-160 in the materials project)<br>\n- Number of atomic snapshots: 15<br>\n- Contents:<br>\n&nbsp;&nbsp;&nbsp; - ideal crystal structure: no<br>\n&nbsp;&nbsp;&nbsp; - MD trajectory: no<br>\n&nbsp;&nbsp;&nbsp; - Atomic positions: no<br>\n&nbsp;&nbsp;&nbsp; - DFT inputs: yes<br>\n&nbsp;&nbsp;&nbsp; - DFT outputs (energies): yes<br>\n&nbsp;&nbsp;&nbsp; - SNAP vectors: yes<br>\n&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; - dimensions: 108x108x35x94 (last dimension: first three entries are x,y,z coordinates, data size is 91)<br>\n&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; - units: a.u.<br>\n&nbsp;&nbsp;&nbsp; - LDOS vectors: yes<br>\n&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; - dimensions: 108x108x35x241<br>\n&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; - units: 1/(eV*Angstrom^3)<br>\n&nbsp;&nbsp;&nbsp; - trained networks: yes</p>\n\n<p><br>\n<em>Dataset structure</em></p>\n\n<p>A .zip file is included for each for each of its five type of data:</p>\n\n<p>- ldos.zip: holds the LDOS vectors (one HDF5 file per snapshot)<br>\n- bispectrum.zip: holds the bispectrum fingerprint vectors&nbsp; (one HDF5 file per snapshot)<br>\n- dft_outputs: holds the outputs from the DFT calculations, i.e. energies and simulation parameters in a .json format (one per snapshot)<br>\n- dft_inputs: holds the inputs for the DFT calculations, in the form of a QE input file (one per snapshot)<br>\n- models: holds five trained NN models for the data set</p>", 
  "title": "Boron data set for machine learning applications", 
  "id": "3746", 
  "author": [
    {
      "family": "Fiedler, Lenz"
    }, 
    {
      "family": "Cangi, Attila"
    }
  ], 
  "version": "v1.0.0"
}