ML IWEIGHT: Difference between revisions
(Created page with "{{TAGDEF|ML_FF_IWEIGHT|[integer]|3}} Description: ---- == Related Tags and Sections == {{sc|ML_FF_IWEIGHT|Examples|Examples that use this tag}} ---- Category:INCARCa...") |
No edit summary |
||
| Line 1: | Line 1: | ||
{{TAGDEF|ML_FF_IWEIGHT|[integer]|3}} | {{TAGDEF|ML_FF_IWEIGHT|[integer]|3}} | ||
Description: | Description: Flag to control the weighting of training data in the machine learning force field method. | ||
---- | ---- | ||
The following cases for {{TAG|ML_FF_IWEIGHT}} are possible: | |||
*{{TAG|ML_FF_IWEIGHT}}=1: Unnormalized energy, force and stress training data are divided by the weights determined by the flags {{TAG|ML_FF_WTOTEN}}, {{TAG|ML_FF_WTIFOR}} and {{TAG|ML_FF_WTSIF}}, respectively. | |||
*{{TAG|ML_FF_IWEIGHT}}=2: The training data are normalized by using their averaged variances. The averaging is done over whole training data. The normalized energy, forces and stress tensor are divided by {{TAG|ML_FF_WTOTEN}}, {{TAG|ML_FF_WTIFOR}} and {{TAG|ML_FF_WTSIF}}, respectively. | |||
*{{TAG|ML_FF_IWEIGHT}}=3: Same as {{TAG|ML_FF_IWEIGHT}}=2 but training data is divided into individual systems. The energy, forces and stress tensor for a system are normalized only using the variance of the corresponding system. | |||
== Related Tags and Sections == | == Related Tags and Sections == | ||
{{TAG|ML_FF_LMLFF}}, {{TAG|ML_FF_WTOTEN}}, {{TAG|ML_FF_WTIFOR}}, {{TAG|ML_FF_WTSIF}} | |||
{{sc|ML_FF_IWEIGHT|Examples|Examples that use this tag}} | {{sc|ML_FF_IWEIGHT|Examples|Examples that use this tag}} | ||
---- | ---- | ||
[[Category:INCAR]][[Category:Machine Learning]][[Category:Machine Learned Force Fields]] | [[Category:INCAR]][[Category:Machine Learning]][[Category:Machine Learned Force Fields]][[Category:VASP6]] | ||
Revision as of 14:15, 13 May 2019
ML_FF_IWEIGHT = [integer]
Default: ML_FF_IWEIGHT = 3
Description: Flag to control the weighting of training data in the machine learning force field method.
The following cases for ML_FF_IWEIGHT are possible:
- ML_FF_IWEIGHT=1: Unnormalized energy, force and stress training data are divided by the weights determined by the flags ML_FF_WTOTEN, ML_FF_WTIFOR and ML_FF_WTSIF, respectively.
- ML_FF_IWEIGHT=2: The training data are normalized by using their averaged variances. The averaging is done over whole training data. The normalized energy, forces and stress tensor are divided by ML_FF_WTOTEN, ML_FF_WTIFOR and ML_FF_WTSIF, respectively.
- ML_FF_IWEIGHT=3: Same as ML_FF_IWEIGHT=2 but training data is divided into individual systems. The energy, forces and stress tensor for a system are normalized only using the variance of the corresponding system.
Related Tags and Sections
ML_FF_LMLFF, ML_FF_WTOTEN, ML_FF_WTIFOR, ML_FF_WTSIF