ML IREG: Difference between revisions
No edit summary |
|||
| (One intermediate revision by the same user not shown) | |||
| Line 1: | Line 1: | ||
{{DISPLAYTITLE:ML_IREG}} | |||
{{TAGDEF|ML_IREG|[integer]|2}} | {{TAGDEF|ML_IREG|[integer]|2}} | ||
| Line 9: | Line 10: | ||
For the optimization of the noise parameter <math>\sigma_{\mathrm{v}}^{2}</math> see [[Machine learning force fields: Theory#Bayesian error estimation|this section]]. | For the optimization of the noise parameter <math>\sigma_{\mathrm{v}}^{2}</math> see [[Machine learning force fields: Theory#Bayesian error estimation|this section]]. | ||
== Related | == Related tags and articles == | ||
{{TAG|ML_LMLFF}}, {{TAG|ML_SIGV0}}, {{TAG|ML_SIGW0}} | {{TAG|ML_LMLFF}}, {{TAG|ML_SIGV0}}, {{TAG|ML_SIGW0}} | ||
| Line 15: | Line 16: | ||
---- | ---- | ||
[[Category:INCAR]][[Category:Machine | [[Category:INCAR tag]][[Category:Machine-learned force fields]] | ||
Latest revision as of 13:24, 8 April 2022
ML_IREG = [integer]
Default: ML_IREG = 2
Description: This tag specifies whether the regularization parameters are kept constant or not in the machine learning force field method.
The following cases are possible for this tag:
- ML_IREG=1: The (initial) precision (ML_SIGV0) and noise (ML_SIGW0) parameters are kept constant.
- ML_IREG=2: The parameters are optimized (default).
For the optimization of the noise parameter see this section.