ML REG: Difference between revisions
No edit summary |
No edit summary |
||
| Line 33: | Line 33: | ||
... ... | ... ... | ||
</pre> | </pre> | ||
The first column contains the ab-initio data. The second column contains the fitted data. | |||
*<code>Total energies (eV)</code>: The number of lines corresponds to the number of training structures (<code>The number of configurations</code> in the corresponding {{TAG|ML_ABN}} file). | |||
*<code>Forces (eV ang.^-1)</code>: | |||
---- | ---- | ||
[[Category:Files]][[Category:Machine-learned force fields]][[Category:Output Files]] | [[Category:Files]][[Category:Machine-learned force fields]][[Category:Output Files]] | ||
Revision as of 11:03, 24 October 2022
This file contains the regression results of the force field training compared to the ab-initio training data in the last training step.
A sample output should look like the following:
****************************************************************************************************
Results at 52 MD step.
====================================================================================================
Total energies (eV)
----------------------------------------------------------------------------------------------------
-16.991157 -17.523764
-17.130082 -17.637286
-17.327379 -17.802581
... ...
====================================================================================================
Forces (eV ang.^-1)
----------------------------------------------------------------------------------------------------
1.180622 0.723282
0.136526 0.018732
-0.033733 0.103882
0.888107 0.516528
-0.148552 -0.401311
0.159014 0.315306
0.689279 0.218246
0.718439 0.750706
... ...
====================================================================================================
Stress (kbar)
----------------------------------------------------------------------------------------------------
208.393665 178.750015
15.421420 8.376722
-5.264501 -3.403226
... ...
The first column contains the ab-initio data. The second column contains the fitted data.
Total energies (eV): The number of lines corresponds to the number of training structures (The number of configurationsin the corresponding ML_ABN file).Forces (eV ang.^-1):