Skip to content

Latest commit

 

History

History
106 lines (81 loc) · 3.06 KB

README.md

File metadata and controls

106 lines (81 loc) · 3.06 KB
Raw

aiida-ase

AiiDA plugin for ASE

Installation

  1. From PyPI
pip install aiida-ase
  1. From this repository (useful for development):
git clone https://github.com/aiidateam/aiida-ase
pip install -e aiida-ase

Usage

The main goal of this plugin is to be a wrap around for ASE.

To make it easy to setup the calculation generate a builder as follows

AseCalculation = CalculationFactory('ase.ase')
builder = AseCalculation.get_builder()

The main parameters for the builder that need to be specified are:

  1. Code
from aiida.orm import load_code
code = load_code('your-code-here@your-computer-here')
builder.code = code

NOTE: If using GPAW, there are two possibilities to set up the calculator a. Specify the Python executable with specific module loaded for GPAW b. Specify directly the GPAW executable. In this case a CMDLINE parameter will be needed (see below).

  1. Structure
builder.structure = structure
  1. k-points data
kpoints = KpointsData()
kpoints.set_kpoints_mesh([2,2,2])  # choose the right mesh here
builder.kpoints = kpoints
  1. Parameters

An example parameter set for GPAW is shown here in parts. See the examples folder for specific examples for other functionality (will be constantly updated).

Define a calculator for a PW calculation with GPAW. Here the name of the calculator is set to GPAW, args is the equivalent of arguments passed into the calculator used in ASE. Note that the @function functionality enables passing arguments to a function inside the calculators. In this example the equivalent ASE command is PW(300). Other arguments such as convergence and occupations can be added.

calculator = {
    'name': 'gpaw',
    'args': {
    'mode': {
        '@function': 'PW',
        'args': {'ecut': 300}
    },
    'convergence': {
        'energy': 1e-9
    },
    'occupations': {
        'name': 'fermi-dirac',
        'width':0.05
    }
}

Add here tags that will be written as atoms.get_xyz(), so for example the first item will be atoms.get_temperature().

atoms_getters = [
    'temperature',
    ['forces', {'apply_constraint': True}],
    ['masses', {}],
]

Some addition utility functions are:

  1. pre_lines: list of lines to added to the start of the python file
  2. post_lines: list of lines to added to the end of the python file
  3. extra_imports: list of extra imports as separated strings, for example ["numpy", "array"] will lead to from numpy import array

Note about choosing a code

  1. If using GPAW it is possible to run parallel calculations using /path/to/execut/gpaw python run_gpaw.py. Set up the code through AiiDA by adding in the gpaw executable. The add the python tag using the command line option
settings = {'CMDLINE': ['python']}
builder.settings = orm.Dict(settings)
  1. If the code you are interested in is present in this plugin registry it might make more sense to use that https://aiidateam.github.io/aiida-registry/

Documentation

The documentation for this package can be found on Read the Docs at http://aiida-ase.readthedocs.io/en/latest/