ML MHIS: Difference between revisions
mNo edit summary |
|||
| Line 2: | Line 2: | ||
{{TAGDEF|ML_MHIS|[integer]|10}} | {{TAGDEF|ML_MHIS|[integer]|10}} | ||
Description: This tag sets the number of estimated errors stored in memory to determine the threshold for the Bayesian error in the machine learning force field method. | Description: This tag sets the number of estimated errors stored in memory to determine the threshold for the Bayesian error in the machine learning force field method for {{TAG|ML_ICRITERIA}}=1. | ||
---- | ---- | ||
The | The use of this tag in combination with the learning algorithms is described here: [[Machine learning force field calculations: Basics#Threshold for error of forces|here]]. | ||
This flag is only used if {{TAG|ML_ICRITERIA}} | This flag is only used if {{TAG|ML_ICRITERIA}}=1. The ML code stores {{TAG|ML_MHIS}} Bayesian errors from previous training steps: immediately after a re-training of the ML-FF, the Bayesian errors of the forces are reevaluated for the current structure (that was also just added as training structure). The average and the maximum error of the forces is stored in the history. After {{TAG | ML_MHIS}} updates of the force field, the threshold {{TAG|ML_CTIFOR}} is first updated and is then updated after every further updated of the ML-FF. We recommend to read the section {{TAG|ML_ICRITERIA}} for further details. | ||
== Related tags and articles == | == Related tags and articles == | ||
{{TAG|ML_LMLFF}}, {{TAG|ML_ICRITERIA}}, {{TAG|ML_CTIFOR}}, {{TAG| | {{TAG|ML_LMLFF}}, {{TAG|ML_ICRITERIA}}, {{TAG|ML_CTIFOR}}, {{TAG|ML_CX}}, {{TAG|ML_CSLOPE}}, {{TAG|ML_CSIG}} | ||
{{sc|ML_MHIS|Examples|Examples that use this tag}} | {{sc|ML_MHIS|Examples|Examples that use this tag}} | ||
Revision as of 09:38, 17 September 2022
ML_MHIS = [integer]
Default: ML_MHIS = 10
Description: This tag sets the number of estimated errors stored in memory to determine the threshold for the Bayesian error in the machine learning force field method for ML_ICRITERIA=1.
The use of this tag in combination with the learning algorithms is described here: here.
This flag is only used if ML_ICRITERIA=1. The ML code stores ML_MHIS Bayesian errors from previous training steps: immediately after a re-training of the ML-FF, the Bayesian errors of the forces are reevaluated for the current structure (that was also just added as training structure). The average and the maximum error of the forces is stored in the history. After ML_MHIS updates of the force field, the threshold ML_CTIFOR is first updated and is then updated after every further updated of the ML-FF. We recommend to read the section ML_ICRITERIA for further details.
Related tags and articles
ML_LMLFF, ML_ICRITERIA, ML_CTIFOR, ML_CX, ML_CSLOPE, ML_CSIG