`pesmaker scf-setup`¶

scf-setup prepares VASP SCF calculation folders.

Normal users can let next run this stage.

Use¶

pesmaker scf-setup run.yaml

Minimal YAML¶

project: scf_run

labeling:
  engine: vasp
  output_dir: labeling
  input_dir: generated
  incar: templates/vasp/INCAR
  potcar_library: /path/to/VASP/potentials
  command: /path/to/vasp_std

jobs:
  submit_command: sbatch
  cores_cpu: 36
  sub_file: templates/sbatch/vasp_cpu_36.sh

Inputs¶

PESMaker looks for structures in this order:

labeling.input_manifest
labeling.input_dir
generation.output_dir
local generated/
runs/<project>/generated

When an input directory contains manifest.jsonl, PESMaker uses that manifest because it preserves exact frame indices for multi-frame files created by select. If no manifest is present, PESMaker recursively scans the directory for structure files and prepares one SCF folder per discovered structure.

Scanned input files can be named POSCAR, CONTCAR, or XDATCAR, or use one of these suffixes:

.cif
.extxyz
.poscar
.traj
.vasp
.xyz

For scanned multi-frame .xyz, .extxyz, .traj, or XDATCAR files, PESMaker prepares one SCF job per frame. This lets users point labeling.input_dir at a manually prepared folder even when the folder was not created by PESMaker and has no manifest.jsonl.

Outputs¶

labeling/
  labeling_manifest.jsonl
  ...
    POSCAR
    INCAR
    POTCAR
    submit.sh

VASP Resource Fields¶

jobs:
  cores_cpu: 36
  vasp_kpar: 2
  vasp_ncore: 6
  skip_completed: true
  check_scf_convergence: true

PESMaker writes KPAR or NCORE to INCAR only when jobs.vasp_kpar or jobs.vasp_ncore is explicitly set. If those keys are omitted, the INCAR template is left in control of VASP parallel tags.

When jobs.sub_file is provided, PESMaker does not scan and rewrite literal resource directives such as #SBATCH --ntasks or literal VASP command lines. It only replaces explicit placeholders such as {command}, {job_name}, {workdir}, {nodes}, {ntasks}, {cores_cpu}, {gpus}, {vasp_kpar}, and {vasp_ncore}. As a convenience, PESMaker does update literal #SBATCH --job-name=... and #SBATCH -J ... lines to the calculation folder name so queued jobs are easier to identify. For nested SCF folders, the queue name uses a compact parent/folder marker, for example mp-1186427_Pd_temp_300K/selected_000000 becomes mp1186427_Pd_sel000000. If the template does not contain {command}, the command line in the user script is left unchanged. If resource fields such as cores_cpu, nodes, or gpus are present, they affect the replacement value of {command} and the corresponding resource placeholders.

To let PESMaker add MPI ranks from the YAML, put {command} in the submit template where VASP should run. For example, with cores_cpu: 36 and command: /path/to/vasp_std, this template line:

{command}

is rendered as:

mpirun -np 36 /path/to/vasp_std

If the template instead contains a literal line such as mpirun /path/to/vasp_std, PESMaker keeps that line unchanged and does not add -np 36.

For CPU VASP jobs with cores_cpu or nodes set, PESMaker uses mpirun -np <nodes * cores_cpu> <labeling.command> when labeling.command is only the VASP executable. If the command starts with mpirun or mpiexec but does not include -np, -n, --np, or --ntasks, PESMaker inserts the same rank count. Existing rank-count options are kept unchanged.

For GPU VASP, request GPUs with jobs.gpus and usually omit vasp_kpar and vasp_ncore. When gpus is greater than zero, PESMaker does not add CPU VASP parallel tags to INCAR. If labeling.command is only the VASP executable, PESMaker runs it as mpirun -np <jobs.gpus> <labeling.command>. If the command starts with mpirun or mpiexec but does not include an explicit rank count, PESMaker inserts -np <jobs.gpus>. Commands that start with srun are kept unchanged.

project: 2D_Te_MD

labeling:
  engine: vasp
  output_dir: run_vasp_scf
  input_dir: selected
  incar: /home/a4s5d/LT/yixiu/MLP_structure/1.Te/1.Material_project_structure/INCAR
  potcar_library: /home/a4s5d/software/VASP/potentials
  command: /data/software/vasp6.4-gpu/bin/vasp_std

jobs:
  submit_command: sbatch
  cores_cpu: 6
  gpus: 1
  skip_completed: true
  check_scf_convergence: true
  sub_file: /home/a4s5d/LT/yixiu/MLP_structure/1.Te/1.Material_project_structure/sub_gpu.sh

This requests one GPU per prepared SCF folder. cores_cpu: 6 should match the CPU task count in the submit template, for example #SBATCH -n 6.

During SCF submission, skip_completed and check_scf_convergence both default to true. PESMaker skips only VASP folders that terminated normally without the electronic SCF nonconvergence marker.

Next Step¶

Preview or submit SCF jobs:

pesmaker submit run.yaml --dry-run
pesmaker submit run.yaml

With next, PESMaker writes the dry-run log and prints the real submit command.

pesmaker scf-setup¶