###############################################
#                Introduction                 #
###############################################

Automated Ab initio Modeling of Materials
Property Package (AMP2)

###############################################
#              File description               #
###############################################

Overall list of files in AMP2 is described 
below:
-----------------------------------------------
run.sh
README
AMP2_manual.pdf
config.yaml
src/
    amp2_input.py
    band.py
    cutoff.py
    dielectric.py
    dos.py
    effm.py
    genetic_algorithm.py
    genetic_operator.py
    get_result.py
    hse_gap.py
    input_conf.py
    kpoint.py
    magnetic_ordering.py
    main.py
    relax.py
    rerun_for_metal.py
    make_supercell.py
    mk_supercell.py
    module_AF.py
    module_GA.py
    module_amp2_input.py
    module_band.py
    module_converge.py
    module_dielectric.py
    module_dos.py
    module_effm.py
    module_hse.py
    module_log.py
    module_subr.py
    module_relax.py
    module_vasprun.py
    module_vector.py
    pot_table.yaml
    INCAR0
    KPOINTS_gamma
    U_table.yaml
    config_def.yaml
Examples/
    Si/
        POSCAR_Si
        config.yaml
    Ge/
        Ge.cif
        config.yaml
    NiO/
        POSCAR_NiO
        config.yaml
----------------------------------------------
Brief description of each file:
- run.sh: shell script file for execution
- README: brief description of code
- AMP2_manual.pdf: manual file for detail description
- config.yaml: configuration file for execution
- src/: path for source file of AMP2
- Examples/: path for examples

##############################################
#               Installation                 #
##############################################

AMP2 supprots Python 2.7 and 3. Currently, the
package is not compatible with lower version
than 2.7. AMP2 is utilizes Python modules in 
the following with link to each site.
  - numpy [https://www.numpy.org]
  - scipy [https://www.scipy.org]
  - spglib [https://atztogo.github.io/spglib]
  - PyYAML [https://pypi.org/project/PyYAML]
These modules should be pre-installed. In addition,
AMP2 needs gnuplot supporting 'pngcairo' and 
'pdfcairo' to draw various figures.

##############################################
#                Basic setup                 #
##############################################

AMP2 uses YAML style configuration file. All 
setting parameters used in AMP2 can be controlled
in 'config.yaml'. Before using AMP2, proper pathes
and mpi program command should be set to be suitable
for your system. Following commands are the
essential directories and programs to be set.

================ config.yaml =================
  Directory:  
    submit:		# the path of structure file
    src:		# the path of source directory
    pot_path_gga:	# the path of potential directory (PBE)
    pot_path_lda:	# the path of potential directory (LDA)
  Program:
    vasp_std:		# the path of vasp binary file (standard)
    vasp_gam:		# the path of vasp binary file (gamma-only)
    vasp_ncl:		# the path of vasp binary file (noncollinear)
    gnuplot:		# the path of gnuplot binary file
    mpi_command:	# mpi command (ex. mpirun, mpiexec)
==============================================

##############################################
#                Execution                   #
##############################################

You can execute AMP2 using Python command as following.

=============================================
python [src_path]/main.py [path for configuration file] [path for nodefile] [the number of cores]
=============================================

- [src_path] is the path for directory of source codes for AMP2.
- [path for configuration file] is the path for configuration file (config.yaml).
- [path for nodefile] is used to record the information of computing nodes such as
  PBS_nodefile in Portable Batch System {PBS} and HOSTNAME in Sun Grid Engine (SGE).
  In the PBS system, we recommand to use the command, "echo $PBS_nodefile > nodefile".
  Also, users can save an arbitrary text by writing in the nodefile.
- [the number of cores] is the number of cores to be used in parallel computing.

For the convenience, we provide the shell script file (run.sh) as following.

================= run.sh ===================
echo 'node information' > nodefile
NPROC=16 # The number of cores for parallel computing
### set path of config.yaml ###
conf=./config.yaml
################################
#### Do not change #############
src_path=`grep 'src_path' $conf | tr -s ' ' | cut -d " " -f 3`
################################
python $src_path/main.py $conf nodefile $NPROC >& stdout.x
==========================================

Before execution, you need to modify 'node information',
NPROC and conf. Then, you can execute AMP2 using shell
script as following.

==========================================
sh run.sh
==========================================

The shell script file can be easily integrated with 
job scheduler program such as portable batch system (PBS).


Installation, manual, and full details: https://amp2.readthedocs.io