ML ISTART: Difference between revisions

From VASP Wiki
No edit summary
No edit summary
Line 10: Line 10:


== Related Tags and Sections ==
== Related Tags and Sections ==
{{TAG|ML_LMLFF}}, {{TAG|ML_W1}}, {{TAG|ML_WTOTEN}}, {{TAG|ML_WTIFOR}}, {{TAG|ML_WTSIF}}
{{TAG|ML_LMLFF}}, {{TAG|ML_W1}}, {{TAG|ML_WTOTEN}}, {{TAG|ML_WTIFOR}}, {{TAG|ML_WTSIF}}, {{TAG|ML_IALGO_LINREG}}


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

Revision as of 15:52, 12 October 2021

ML_ISTART = [integer]
Default: ML_ISTART = 0 

Description: This tag decides if and how calculations are continued from existing data in machine learning force field method.


The following cases can be selected:

  • ML_ISTART=0: The training calculations are started from scratch without taking any existing data into considerations. Learning is executed.
  • ML_ISTART=1: The program reads the training data stored in the ML_AB file and restarts the training calculation. Learning is executed.
  • ML_ISTART=2: The program reads the force field parameters stored in the ML_FF file and executes calculations using the force field only. No learning is executed.
  • ML_ISTART=3: The program reads the training data stored in the ML_AB file and selects new local reference configurations from the training data.

Related Tags and Sections

ML_LMLFF, ML_W1, ML_WTOTEN, ML_WTIFOR, ML_WTSIF, ML_IALGO_LINREG

Examples that use this tag