ML SIGV0: Difference between revisions

From VASP Wiki
No edit summary
No edit summary
 
(6 intermediate revisions by the same user not shown)
Line 2: Line 2:
{{TAGDEF|ML_SIGV0|[real]|1.0}}
{{TAGDEF|ML_SIGV0|[real]|1.0}}


Description: This flag sets the initial reversed and squared noise parameter <math>\frac{1}{\sigma_{\mathrm{v}}^{2}}</math> in the machine learning force field method.
Description: This flag sets the noise parameter <math>s_{\mathrm{v}}</math> (see [[Machine learning force field: Theory#Bayesian linear regression|here]] for definition) for the fitting in the machine learning force field method.
----
----
For details about the optimization of this regularization parameter see [[Machine learning force field: Theory#Bayesian error estimation|this section]].
If the regularization needs to be controlled manually, like e.g. in the fitting via singular value decomposition ({{TAG|ML_MODE}}=''REFIT'' or {{TAG|ML_IALGO_LINREG}}=4), the best is to keep this parameter constant at 1 and control the regularization via the precision parameter <math>s_{\mathrm{w}}</math> (see {{TAG|ML_SIGW0}}).
 
For the theory of this regularization parameter see [[Machine learning force field: Theory#Regression|this section]].


== Related tags and sections ==
== Related tags and sections ==
{{TAG|ML_LMLFF}}, {{TAG|ML_IREG}}, {{TAG|ML_SIGW0}}
{{TAG|ML_LMLFF}}, {{TAG|ML_IREG}}, {{TAG|ML_SIGW0}}, {{TAG|ML_IALGO_LINREG}}


{{sc|ML_SIGV0|Examples|Examples that use this tag}}
{{sc|ML_SIGV0|Examples|Examples that use this tag}}

Latest revision as of 16:05, 3 July 2023

ML_SIGV0 = [real]
Default: ML_SIGV0 = 1.0 

Description: This flag sets the noise parameter (see here for definition) for the fitting in the machine learning force field method.


If the regularization needs to be controlled manually, like e.g. in the fitting via singular value decomposition (ML_MODE=REFIT or ML_IALGO_LINREG=4), the best is to keep this parameter constant at 1 and control the regularization via the precision parameter (see ML_SIGW0).

For the theory of this regularization parameter see this section.

Related tags and sections

ML_LMLFF, ML_IREG, ML_SIGW0, ML_IALGO_LINREG

Examples that use this tag