Input file

pytint uses a yaml file for specifying the input options. In this section, the various blocks of the input file is discussed. A complete sample input file is also provided in the Examples section.

main block

main block consists of the major options that the user has to provided. A sample block is shown below.

  temperature: [1000, 1400]
  pressure: [0]
  element: 'Cu'
  lattice: [FCC, LQD]
  nsims: 3
  • temperature: List of temperatures at which NEHI calculations have to be done. In case of regular use, one calculation will be started for each temperature. In case of --mode rs, one NEHI calculation will be done for the first temperature. After that, reversible scaling calculation will be done to extend the free energy upto the last temperature specified.

  • pressure: Pressure for the calculation. Currently only zero pressure (Helmholtz free energy) is supported.

  • element: The chemical symbol of the element used.

  • lattice: The lattices for which the free energy calculations have to be done. Supported lattices are BCC, FCC, HCP, DIA, SC and LQD.

  • nsims: The number of independent calculations to be carried out for finding the error in the estimated free energy.

md block

  timestep: 0.001
  pair_style: pace
  pair_coeff: "* * Cu.ace Cu"
  mass: 63.546
  tdamp: 0.1
  pdamp: 0.1
  nx: 5
  ny: 5
  nz: 5
  te: 25000
  ts: 50000
  • timestep: Timestep in ps for the md simulations

  • pair_style: Pair style used in LAMMPS. Supported pair styles are pace, eam and its variants, sw and snap.

  • pair_coeff: Pair coefficient command used in LAMMPS. The relative/full path of the input potential file is also specified here.

  • mass: Atomic mass of the element.

  • tdamp: Thermostat damping in units of time.

  • pdamp: Barostat damping in units of time.

  • nx: Number of units cells in the 001 direction.

  • ny: Number of units cells in the 010 direction.

  • nz: Number of units cells in the 100 direction.

  • te: Number of time steps for equilibration runs.

  • ts: Number of time steps for switching runs.

queue block

This block specifies the input parameters for submitting the job on a cluster

  scheduler: slurm
  cores: 40
  jobname: cu
  walltime: "23:50:00"
  queuename: shorttime
  memory: 3GB
    - anaconda/4
    - source .bashrc
    - conda activate py3
  #any other extra options
  # - "-j Y"
  • scheduler: The scheduler to be used for calculations. Supported options are local, slurm or sge.

  • cores: Number of cores to be used for the md runs.

  • jobname: Name of the job. Ignored for local.

  • walltime: Walltime for the job. Ignored for local.

  • queuename: Name of the submission queue. Ignored for local.

  • memory: Total memory requested per core. Ignored for local.

  • modules: Name of module that need to be loaded.

  • commands: Extra commands that will be run in the beginning of the submission script. If a conda environment is used, the activate statements will be here.

  • options: Further special options for the submission script.