pymatgen.io.espresso
is a pymatgen
addon that adds support for Quantum ESPRESSO (QE). The goal of this package is to elevate QE to a first-class citizen in the pymatgen
-driven computational materials science ecosystem.
pymatgen.io.espresso
aims to provide classes whose public APIs are fully compatible with those from pymatgen.io.vasp
. The intention is to allow any pymatgen
-based VASP post-processing code to add QE support with little to no developer effort. Ideally, this should be as simple as going from this
from pymatgen.io.vasp.outputs import Vasprun
calc = Vasprun('vasprun.xml', **kwargs)
# Complicated code
to this
from pymatgen.io.espresso.outputs import PWxml
calc = PWxml('prefix.xml', **possibly_different_kwargs)
# Exact same complicated code, with no changes
without any additional changes to the actual post-processing code. Under the hood, pymatgen.io.espresso
automatically converts all the units, coordinates, and conventions used by QE to those used by VASP.
pymatgen.io.espresso
additionally provides utilities for parsing and creating Quantum ESPRESSO input files, and a converter from VASP inputs (INCAR
, KPOINTS
, and POSCAR
) to pw.x
inputs ("Caffeinator") is a work in progress.
Currently, the package is in pre-alpha and is not yet available on PyPi
. To install it:
pip install git+https://github.com/Griffin-Group/pymatgen-io-espresso
We have detailed documentation automatically generated from our doc strings, and some simple tutorials and examples are a work in progress.
Contributions are welcome! Please see the contributing guide for more information.
- Omar A. Ashour (@oashour): Creator and maintainer.
- Ella Banyas (@ebanyas): Sous-dev, bug fixes, caffeination module (WIP).
- Willis O'Leary (@wolearyc): Sous-dev, bug fixes.