ML WTOTEN: Difference between revisions
No edit summary |
No edit summary |
||
| Line 6: | Line 6: | ||
---- | ---- | ||
For {{TAG|ML_IWEIGHT}}=2 and 3 (default), {{TAG|ML_WTOTEN}} | For {{TAG|ML_IWEIGHT}}=2 and 3 (default), {{TAG|ML_WTOTEN}} is unitless, and the data is multiplied by the weight. We recommend increasing {{TAG|ML_WTOTEN}} if you plan to apply the force field in a simulation where the accuracy of the total energy is most important. This puts a focus on the energy error and is desirable, for instance, for the computation of defect-formation energies. | ||
For {{TAG|ML_IWEIGHT}}=1, {{TAG|ML_WTOTEN}} has the unit of eV/atom and is used to divide the data by it. | For {{TAG|ML_IWEIGHT}}=1, {{TAG|ML_WTOTEN}} has the unit of eV/atom and is used to divide the data by it. | ||
Revision as of 08:06, 3 October 2023
ML_WTOTEN = [real]
| Default: ML_WTOTEN | = 0.005 | if ML_IWEIGHT=1 |
| = 1.0 | otherwise |
Description: Sets a weight for scaling the total energy in the training data.
For ML_IWEIGHT=2 and 3 (default), ML_WTOTEN is unitless, and the data is multiplied by the weight. We recommend increasing ML_WTOTEN if you plan to apply the force field in a simulation where the accuracy of the total energy is most important. This puts a focus on the energy error and is desirable, for instance, for the computation of defect-formation energies.
For ML_IWEIGHT=1, ML_WTOTEN has the unit of eV/atom and is used to divide the data by it.
Related tags and articles
ML_IWEIGHT, ML_WTIFOR, ML_WTSIF, ML_IALGO_LINREG, ML_LMLFF