ML LERR: Difference between revisions
No edit summary |
No edit summary |
||
| Line 1: | Line 1: | ||
{{DISPLAYTITLE:ML_LERR}} | {{DISPLAYTITLE:ML_LERR}} | ||
{{TAGDEF|ML_LERR|[logical]|.TRUE.}} | {{TAGDEF|ML_LERR|[logical]|.TRUE.}} | ||
{{NB|warning|This tag is deprecated and is replaced by {{TAG|ML_IERR}}.}} | |||
Description: Decides whether the Bayesian error estimates are calculated and written out or not. | Description: Decides whether the Bayesian error estimates are calculated and written out or not. | ||
| Line 7: | Line 8: | ||
== Related tags and articles == | == Related tags and articles == | ||
{{TAG|ML_LMLFF}}, {{TAG|ML_ISTART}} | {{TAG|ML_LMLFF}}, {{TAG|ML_IERR}}, {{TAG|ML_ISTART}} | ||
---- | ---- | ||
[[Category:INCAR tag]][[Category:Machine-learned force fields]] | [[Category:INCAR tag]][[Category:Machine-learned force fields]] | ||
Revision as of 08:18, 29 March 2023
ML_LERR = [logical]
Default: ML_LERR = .TRUE.
| Warning: This tag is deprecated and is replaced by ML_IERR. |
Description: Decides whether the Bayesian error estimates are calculated and written out or not.
The Bayesian error estimation is always required for training (ML_ISTART=0,1,3) but may be omitted for prediction runs, e.g., MD simulations with ML_ISTART=2. In this case, setting ML_LERR=.FALSE. can significantly speed up the run because the error estimation is comparable in computational cost to the actual prediction.
| Warning: Combine ML_LERR=.FALSE. only with ML_ISTART=2! Any other choice, e.g., ML_ISTART=1 with NSW=0 and ML_IALGO_LINREG=3, is untested and has no computational advantage. |