Last commit for doc/input-files/predict.x-input.org: 5874abaa643d4472a2aa9d1c5dbe454dadbd8d1f

Initial commit of the AENET code.

Bruno Mundim [2017-01-02 17:48:39]
Initial commit of the AENET code.
#+AUTHOR: Nongnuch Artrith
#+TITLE: Input file format for =predict.x=

* Alphabetic list of keywords

  All keywords are case insensitive and independent of the order.  Blank
  lines and lines starting with =!=, =#=, or =%= are ignored.

  - =debug= (optional) :: Activate debugging mode; additional output files
       will be created.
  - =files= (optional) :: Specifies a list of paths to input structures.
       This keyword may be used for batch processing of a larger number
       of structures.  The line following the keyword contains the
       number of input files =<NF>=, and each of the following =<NF>=
       lines contains a single file system path.  Alternatively, a
       single input structure may be passed to =predict.x= as command
       line argument.  The command line takes precedence over the list
       specified with the "files" keyword.
  - =forces= (optional) :: Activates evaluation of the atomic forces.
       Forces are also calculated, when the "relax" keyword is present.
  - =networks= (required) :: Specifies the ANN potential files for each
       chemical species.  On each of the =<NT>= lines following the
       keyword a chemical species =<T_i>= and the path to its
       corresponding ANN file is given.
  - =relax= (optional) :: Activate structural relaxation; this will
       automatically also activate the calculation of the atomic
       forces.  On the line following the =relax= keyword, several
       options can be specified.  See the example below.
  - =timing= (optional) :: Activate timing; additional output files will
       be created.
  - =types= (required) :: Specifies the number of different atomic species
       that may occur in structures and their chemical symbols.  The
       first line following the keyword specifies the number =<NT>= of
       different atom types; the following lines each contain one
       chemical symbol =<T_i>=.

* Input file template (predict.in)

#+BEGIN_EXAMPLE
TYPES
<NT>
<T_1>
<T_2>
...
<T_NT>

NETWORKS
<T_1>  <path/to/NN-1>
<T_2>  <path/to/NN-2>
...
<T_NT> <path/to/NN-NT>

FORCES

# or optimize coordinates:
#
# RELAX
# method=bfgs  F_conv=1.0d-2  E_conv=1.0d-6  steps=99
#
#    method: optimization method (currently only BFGS)
#    F_conv: convergence thershold for the forces
#    E_conv: convergence threshold for the energy
#    steps:  max. number of iterations

FILES
<NF>
<path/to/structure-1.xsf>
<path/to/structure-2.xsf>
...
<path/to/structure-NF.xsf>
#+END_EXAMPLE

* Input file example (predict.in) for TiO_{2}

#+BEGIN_EXAMPLE
TYPES
2
Ti
O

NETWORKS
  Ti Ti.10tw-10tw.ann
  O  O.10tw-10tw.ann

FORCES

FILES
10
structure0001.xsf
structure0002.xsf
structure0003.xsf
structure0004.xsf
structure0005.xsf
structure0006.xsf
structure0007.xsf
structure0008.xsf
structure0009.xsf
structure0010.xsf
#+END_EXAMPLE
ViewGit