ML LUSE NAMES: Difference between revisions

From VASP Wiki
(Created page with "{{DISPLAYTITLE:ML_LUSE_NAMES}} {{TAGDEF|ML_LUSE_NAMES|[logical]|.FALSE.}} Description: Decides whether trainings structures are additionally subdivided into groupes internally due to their structure name. ---- == Related tags and articles == {{TAG|ML_LMLFF}}, {{TAG|ML_WTOTEN}}, {{TAG|ML_WTIFOR}}, {{TAG|ML_WTSIF}}, {{TAG|ML_IWEIGHT}} {{sc|ML_IWEIGHT|Examples|Examples that use this tag}} ---- Category:INCAR tagCategory:Machine-learned force fields")
 
No edit summary
 
(2 intermediate revisions by the same user not shown)
Line 2: Line 2:
{{TAGDEF|ML_LUSE_NAMES|[logical]|.FALSE.}}
{{TAGDEF|ML_LUSE_NAMES|[logical]|.FALSE.}}


Description: Decides whether trainings structures are additionally subdivided into groupes internally due to their structure name.
Description: Decides whether training structures are additionally subdivided into groups internally due to their structure name.
----
----
This tag is important if the normalization via averages over subset standard deviations ({{TAG|ML_IWEIGHT}}=3) is employed.
By default ({{TAG|ML_LUSE_NAMES}}=''.FALSE.'') the division into subsets is based on the atom types and number of atoms per type. If two systems contain the same atom types and the same number of atoms per type then they are considered to be in the same subset. To further divide them into subsets set {{TAG|ML_LUSE_NAMES}}=''.TRUE.'' and choose different system names in the first line of the {{TAG|POSCAR}} file. This can be useful if training is performed for widely different materials, for instance, different phases with widely different energies. Without the finer subset assignment, the overall energy standard deviation might become large, reducing the weight of the energies too much of given subsets.


== Related tags and articles ==
== Related tags and articles ==

Latest revision as of 15:08, 29 August 2024

ML_LUSE_NAMES = [logical]
Default: ML_LUSE_NAMES = .FALSE. 

Description: Decides whether training structures are additionally subdivided into groups internally due to their structure name.


This tag is important if the normalization via averages over subset standard deviations (ML_IWEIGHT=3) is employed. By default (ML_LUSE_NAMES=.FALSE.) the division into subsets is based on the atom types and number of atoms per type. If two systems contain the same atom types and the same number of atoms per type then they are considered to be in the same subset. To further divide them into subsets set ML_LUSE_NAMES=.TRUE. and choose different system names in the first line of the POSCAR file. This can be useful if training is performed for widely different materials, for instance, different phases with widely different energies. Without the finer subset assignment, the overall energy standard deviation might become large, reducing the weight of the energies too much of given subsets.

Related tags and articles

ML_LMLFF, ML_WTOTEN, ML_WTIFOR, ML_WTSIF, ML_IWEIGHT

Examples that use this tag