In-2024-06-05

../../_images/mlp_dist27.png

The current structure dataset comprises 17485 structures. Procedures to generate structures and estimate MLPs are found in A. Seko, J. Appl. Phys. 133, 011101 (2023).

Predictions using Pareto optimal MLPs

../../_images/eqm_properties27.png

These properties are calculated for MLP equilibrium structures obtained by performing local structure optimizations from the DFT equilibrium structures. These DFT equilibrium structures are obtained by optimizing prototype structures that are included in ICSD. As a result, the structure type of the converged structure may sometimes differ from the one shown in the legend.

The other properties predicted using each Pareto optimal MLP are available from column Predictions in the following table.

Pareto optimals (on convex hull)

Name

Time

RMSE

Predictions

Files

polymlp-00002

0.029 / 0.003

13.337 / 0.0786

polymlp-00035

0.125 / 0.014

3.352 / 0.0252

predictions

polymlp.lammps polymlp.in

polymlp-00038

0.149 / 0.011

3.085 / 0.0227

predictions

polymlp.lammps polymlp.in

polymlp-00122

0.251 / 0.018

2.064 / 0.0235

predictions

polymlp.lammps polymlp.in

polymlp-00125

0.276 / 0.018

1.946 / 0.0219

predictions

polymlp.lammps polymlp.in

polymlp-00159

0.361 / 0.026

1.764 / 0.0193

predictions

polymlp.lammps polymlp.in

polymlp-00246

0.588 / 0.037

1.522 / 0.0193

predictions

polymlp.lammps polymlp.in

polymlp-00106

1.021 / 0.047

1.132 / 0.0173

predictions

polymlp.lammps polymlp.in

polymlp-00107

1.602 / 0.078

0.951 / 0.0159

predictions

polymlp.lammps polymlp.in

polymlp-00222

1.963 / 0.087

0.914 / 0.0156

predictions

polymlp.lammps polymlp.in

polymlp-00136

2.684 / 0.120

0.860 / 0.0143

predictions

polymlp.lammps polymlp.in

polymlp-00223

3.240 / 0.144

0.833 / 0.0146

predictions

polymlp.lammps polymlp.in

polymlp-00252

5.244 / 0.287

0.808 / 0.0136

predictions

polymlp.lammps polymlp.in

polymlp-00166

7.253 / 0.481

0.810 / 0.0134

predictions

polymlp.lammps polymlp.in

polymlp-00253

9.084 / 0.547

0.825 / 0.0134

predictions

polymlp.lammps polymlp.in

Units:

  • Time: [ms] (1core/36cores)

  • RMSE: [meV/atom]/[eV/ang.]

Column “Time” shows the time required to compute the energy and forces for 1 MD step and 1 atom, which is estimated from 10 runs for a large structure using a workstation with Intel(R) Xeon(R) CPU E5-2695 v4 @ 2.10GHz. Note that these MLPs should be carefully used for extreme structures. The MLPs often return meaningless values for them.

  • All Pareto optimal MLPs are available here.