ML CTIFOR: Difference between revisions
No edit summary |
mNo edit summary |
||
| Line 4: | Line 4: | ||
---- | ---- | ||
During learning, if a newly considered structure yields a Bayesian error in the force larger than {{TAG|ML_CTIFOR}}, a first principles calculation is performed, and the corresponding structure is added to the | During learning, if a newly considered structure yields a Bayesian error in the force larger than {{TAG|ML_CTIFOR}}, a first principles calculation is performed, and the corresponding structure is added to the data set of structures that are used when the force field is updated. The unit of {{TAG|ML_CTIFOR}} is eV/Angstrom. If the tag {{TAG|ML_ICRITERIA}}=''.TRUE.'' is set, {{TAG|ML_CTIFOR}} is automatically updated using past Bayesian error estimates for the forces. | ||
== Related Tags and Sections == | == Related Tags and Sections == | ||
Revision as of 20:27, 11 October 2021
ML_CTIFOR = [real]
Default: ML_CTIFOR =
Description: This flag sets the threshold for the Bayesian error estimation on the force in the machine learning force field method.
During learning, if a newly considered structure yields a Bayesian error in the force larger than ML_CTIFOR, a first principles calculation is performed, and the corresponding structure is added to the data set of structures that are used when the force field is updated. The unit of ML_CTIFOR is eV/Angstrom. If the tag ML_ICRITERIA=.TRUE. is set, ML_CTIFOR is automatically updated using past Bayesian error estimates for the forces.
Related Tags and Sections
ML_LMLFF, ML_ICRITERIA, ML_CDOUB