ML SION1: Difference between revisions

Revision as of 19:54, 10 October 2021

ML_SION1 = [real]
Default: ML_SION1 = 0.333333

Description: This tag specifies the width [math]\displaystyle{ \sigma_\text{atom} }[/math] of the Gaussian functions used for broadening the atomic distributions for the radial descriptor [math]\displaystyle{ \rho^{(2)}_i(r) }[/math] within the machine learning force field method (see this section).

The unit of ML_SION1 is [math]\displaystyle{ \AA }[/math]. Background:{| class="vasp-dark-link-panel" style="border: 0px solid var(--vblue); --box-emph-color: var(--vblue); padding: 5px; color: var(--vdefault-text-nb); background: var(--vblue-bg)"

  | Tip: Our test calculations indicate that a ratio ML_SION2 / ML_SION1 = 1.5 results in an optimal training performance. Furthermore, a value of 0.5 was found to be a good default value for ML_SION2. Both findings together result in the above default value for ML_SION1.
  |}

Related Tags and Sections

ML_LMLFF, ML_SION2, ML_RCUT1, ML_RCUT2, ML_MRB1, ML_MRB2

Examples that use this tag

@@ Line 4: / Line 4: @@
 ----
-The unit of {{TAG|ML_SION1}} is in <math>\AA</math>.
+The unit of {{TAG|ML_SION1}} is <math>\AA</math>.
-Test calculations showed that a 1.5 smaller value for the broadening of the radial descriptor compared to the angular descriptor (see {{TAG|ML_SION2}}) gives optimal results.
+{{BOX|tip|Our test calculations indicate that a ratio {{TAG|ML_SION2}} / {{TAG|ML_SION1}} {{=}} 1.5 results in an optimal training performance. Furthermore, a value of 0.5 was found to be a good default value for {{TAG|ML_SION2}}. Both findings together result in the above default value for {{TAG|ML_SION1}}.|Background:}}
-The default value for {{TAG|ML_SION1}} is chosen such that {{TAG|ML_SION2}} becomes 0.5.
 == Related Tags and Sections ==