Requests for technical support from the VASP group should be posted in the VASP-forum.

Difference between revisions of "Category:Machine Learning"

From Vaspwiki
Jump to navigationJump to search
(34 intermediate revisions by 2 users not shown)
Line 1: Line 1:
== Theoretical Background ==
+
This section describes the methodology used for force-field generation using machine learning. One first should checkout the theoretical background {{TAG|Machine learning force field: Theory}} together with the basic description how to run the machine learning calculations {{TAG|Machine learning force field calculations: Basics}}. Then gain some hands-on experience with the following tutorial: {{TAG|Liquid Si - MLFF}}.
*{{TAG|On-the-fly machine learning force field generation using Bayesian linear regression}}.
 
 
 
 
 
 
 
 
 
 
 
 
 
{{TAG|ML_FF_NDIM_SCALAPACK}}
 
 
 
{{TAG|ML_FF_LMLFF}}
 
 
 
{{TAG|ML_FF_MCONF}}
 
 
 
{{TAG|ML_FF_MCONF_NEW}}
 
 
 
{{TAG|ML_FF_ISAMPLE}}
 
 
 
{{TAG|ML_FF_IERR}}
 
 
 
! {{TAG|ML_FF_LMLPC}}
 
 
 
{{TAG|ML_FF_LMLMB}}
 
 
 
{{TAG|ML_FF_NWRITE}}
 
 
 
{{TAG|ML_FF_LBASIS_DISCARD}}
 
 
 
{{TAG|ML_FF_LCONF_DISCARD}}
 
 
 
{{TAG|ML_FF_ISTART}}
 
 
 
{{TAG|ML_FF_CDOUB}}
 
 
 
{{TAG|ML_FF_LCRITERIA}}
 
 
 
{{TAG|ML_FF_CSF}}
 
 
 
{{TAG|ML_FF_CTIFOR}}
 
 
 
{{TAG|ML_FF_CSIG}}
 
 
 
{{TAG|ML_FF_CSLOPE}}
 
 
 
{{TAG|ML_FF_NMDINT}}
 
 
 
{{TAG|ML_FF_MHIS}}
 
 
 
{{TAG|ML_FF_IWEIGHT}}
 
 
 
{{TAG|ML_FF_WTOTEN}}
 
 
 
{{TAG|ML_FF_WTIFOR}}
 
 
 
{{TAG|ML_FF_WTSIF}}
 
 
 
{{TAG|ML_FF_EATOM}}
 
 
 
{{TAG|ML_FF_LEATOM_MB}}
 
 
 
{{TAG|ML_FF_LHEAT_MB}}
 
 
 
{{TAG|ML_FF_ISCALE_TOTEN_MB}}
 
 
 
{{TAG|ML_FF_MB_MB}}
 
 
 
{{TAG|ML_FF_ISOAP1_MB}}
 
 
 
{{TAG|ML_FF_ISOAP2_MB}}
 
 
 
{{TAG|ML_FF_W1_MB}}
 
 
 
{{TAG|ML_FF_W2_MB}}
 
 
 
{{TAG|ML_FF_LNORM1_MB}}
 
 
 
{{TAG|ML_FF_LNORM2_MB}}
 
 
 
{{TAG|ML_FF_ICUT1_MB}}
 
 
 
{{TAG|ML_FF_ICUT2_MB}}
 
 
 
{{TAG|ML_FF_IBROAD1_MB}}
 
 
 
{{TAG|ML_FF_IBROAD2_MB}}
 
 
 
{{TAG|ML_FF_IREG_MB}}
 
 
 
{{TAG|ML_FF_SIGV0_MB}}
 
 
 
{{TAG|ML_FF_SIGW0_MB}}
 
 
 
{{TAG|ML_FF_MSPL1_MB}}
 
 
 
{{TAG|ML_FF_MSPL2_MB}}
 
 
 
{{TAG|ML_FF_NR1_MB}}
 
 
 
{{TAG|ML_FF_NR2_MB}}
 
 
 
{{TAG|ML_FF_MRB1_MB}}
 
 
 
{{TAG|ML_FF_MRB2_MB}}
 
 
 
{{TAG|ML_FF_RCUT1_MB}}
 
 
 
{{TAG|ML_FF_RCUT2_MB}}
 
 
 
{{TAG|ML_FF_SION1_MB}}
 
 
 
{{TAG|ML_FF_SION2_MB}}
 
 
 
{{TAG|ML_FF_NHYP1_MB}}
 
 
 
{{TAG|ML_FF_NHYP2_MB}}
 
 
 
{{TAG|ML_FF_LMAX2_MB}}
 
 
 
{{TAG|ML_FF_LAFILT2_MB}}
 
 
 
{{TAG|ML_FF_IAFILT2_MB}}
 
 
 
{{TAG|ML_FF_AFILT2_MB}}
 
 
 
{{TAG|ML_FF_LCOUPLE_MB}}
 
 
 
{{TAG|ML_FF_NATOM_COUPLED_MB}}
 
 
 
{{TAG|ML_FF_ICOUPLE_MB}}
 
 
 
{{TAG|ML_FF_RCOUPLE_MB}}
 
  
 +
== Input ==
  
 +
*Besides the usual input files ({{TAG|INCAR}}, {{TAG|POSCAR}}, etc.) the machine learning force field method requires the following input files:
 +
**{{TAG|ML_AB}} Ab initio data used to create the training data.
 +
**{{TAG|ML_FF}} File containing force field parameters.
  
 +
== Output ==
  
 +
*The machine learning force field method generates the following output files:
 +
**{{TAG|ML_LOGFILE}} Main output file for the machine learning force field method.
 +
**{{TAG|ML_ABN}} New abinitio data (used as {{TAG|ML_AB}} in the next run).
 +
**{{TAG|ML_REG}} Output file summarizing regression results.
 +
**{{TAG|ML_HIS}} Output file summarizing the histogram data.
 +
**{{TAG|ML_FFN}} File containing new force field parameters (used as {{TAG|ML_FF}} in the next run).
 +
**{{TAG|ML_EATOM}} Output file containing local atomic energies
 +
**{{TAG|ML_HEAT}} Output file including local heat flux
  
  
  
 +
All {{TAG|INCAR}} tags belonging to the machine learning force field method start with the prefix ''ML_FF_''. Input tags that are related to the many-body term end with ''_MB''.
  
 +
== Theoretical Background ==
 +
*{{TAG|Machine learning force field: Theory}}.
  
 +
== How to ==
 +
*{{TAG|Machine learning force field calculations: Basics}}.
 +
<!--*{{TAG|Machine learning force field calculations: Intermediate}}. -->
  
 +
== Tutorial ==
 +
*Basic tutorial to learn how to perform on-the-fly learning and how to control the accuracy of the force field: {{TAG|Liquid Si - MLFF}}.
  
 
----
 
----
[[Category:VASP|Machine Learning]][[Category:VASP6]]
+
[[Category:VASP|Machine Learning]][[Category:Alpha]]

Revision as of 11:10, 7 June 2021

This section describes the methodology used for force-field generation using machine learning. One first should checkout the theoretical background Machine learning force field: Theory together with the basic description how to run the machine learning calculations Machine learning force field calculations: Basics. Then gain some hands-on experience with the following tutorial: Liquid Si - MLFF.

Input

  • Besides the usual input files (INCAR, POSCAR, etc.) the machine learning force field method requires the following input files:
    • ML_AB Ab initio data used to create the training data.
    • ML_FF File containing force field parameters.

Output

  • The machine learning force field method generates the following output files:
    • ML_LOGFILE Main output file for the machine learning force field method.
    • ML_ABN New abinitio data (used as ML_AB in the next run).
    • ML_REG Output file summarizing regression results.
    • ML_HIS Output file summarizing the histogram data.
    • ML_FFN File containing new force field parameters (used as ML_FF in the next run).
    • ML_EATOM Output file containing local atomic energies
    • ML_HEAT Output file including local heat flux


All INCAR tags belonging to the machine learning force field method start with the prefix ML_FF_. Input tags that are related to the many-body term end with _MB.

Theoretical Background

How to

Tutorial

  • Basic tutorial to learn how to perform on-the-fly learning and how to control the accuracy of the force field: Liquid Si - MLFF.