ML LSPARSDES: Difference between revisions

From VASP Wiki
No edit summary
No edit summary
Line 5: Line 5:
----
----


For details on the theory how angular descriptor are sparsified see [[Machine learning force field: Theory#Sparsification of angular descriptors|here]].
To use the [[Machine learning force field: Theory#Sparsification of angular descriptors|sparsification of angular descriptors set]] the following tags:
 
To use the sparsification of angular descriptors set the following tags:
*{{TAG|ML_LSPARSDES}}=''.TRUE.''.
*{{TAG|ML_LSPARSDES}}=''.TRUE.''.
*The number of the highest eigenvalues <math>k</math> to which the correlation is measured via the leverage scoring: {{TAG|ML_NRANK_SPARSDES}}.
*The number of the highest eigenvalues <math>k</math> to which the correlation is measured via the leverage scoring: {{TAG|ML_NRANK_SPARSDES}}.

Revision as of 12:44, 4 October 2023

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

Description: This tag specifies whether angular descriptor sparsification is enabled within the machine learning force field method.


To use the sparsification of angular descriptors set the following tags:

  • ML_LSPARSDES=.TRUE..
  • The number of the highest eigenvalues to which the correlation is measured via the leverage scoring: ML_NRANK_SPARSDES.
  • The ratio of the selected descriptors to the total number of descriptors: ML_RDES_SPARSDES. This tag controls the extent of the sparsification.

Related tags and articles

ML_LMLFF, ML_NRANK_SPARSDES, ML_RDES_SPARSDES

Examples that use this tag