I wrote most of this package while at the University of Florida. A lot of this code is pedagogical where I'm trying to learn how to do things. It connects to DAKOTA, LAMMPS, and VASP. I am currently trying to import the code which I wrote (and I'm currently writing for pyflamestk) into pypospack, but writing it in a way which is more maintainable.
I pypospack is published though github, but I currently do not keep track of versioning. When the code base is more mature, I will implement versioning. The documentation here is temporary and meant as a supplement to the documentation in the code and the documentation contained within sphinx. Eventually, I'll post all this information onto a personal website, but this is what constitutes documentation at this time.
pypospack is provided free-of-charge and without warranty, as long as attribution is provided through the MIT license.
Python Potential Optimization Software Package (pypospack) is organized as a python library which can be used in the development of other applications. Rather than implementing calculations for molecular dynamics code and density functional theory code, it uses an external package as a calculator, so that any biases which are contained in the calculator are replicated when the potential is used. The external calculators supported are
This alleviates the problem of pypospack to have a rigorous verification and validation process. This software also makes extensive use of python packages administered by assuming an Anaconda distribution.
I've tried making the library as general as possible. It is written for Python 3 and I've limited the amount of dependencies to the Anaconda toolset. If there is significant interest is this library, I will integrate some other popular atomistic python libraries and interface to other software.
Development of this software would not be possible without funding from the University of Florida and the Department of Energy, particularly with some projects from Lawrence Berkeley Labs, Argonne National Lab, and SANDIA. Some of this work, I developed while at Idaho National Lab to automate work.
However, the software package is currently under development, and documentation within the code may be lacking at times. Developer documentation is predominantly provided through comments within the code. These inline comments are formatted using the Google Python Style Guide. A third party software, Sphinx, which supports the Google Python Style Guide through napoleon.
pypospack is currently under active development and the codebase isn't currently stable. The latest development branch can be found on github.
#!/bin/python at the top of python scripts use #!/bin/env pythonI am horrible at documenting my code. I think my code is somewhat reasonable in terms of readability, but I'll start writing better documentation as I start to clean up the code the development and write unit tests.
pyflamestk.base.py
pyflamestk.crystals.py
pyflamestk.dakota.py
pyflamestk.lammps.py
pyflamestk.potentials.py
pyflamestk.vasp.py
pyflamestk.phonon.py
pyflamestk.raman.py
I've started to do some testing of the software I'm developing. The testing framework is heavily influenced by "Chapter 13: Unit Testing" of the free online book "Dive into Python" by Mark Pilgrim.
vasp2lammps.py (Console application)
Converts a vasp file into a lammps file
/examples/buckingham/MgO_pareto_simulations
/examples/buckingham/MgO_pareto_postprocess
/examples/buckingham/MgO_pareto_iterate
This software is currently being developed jointly with both Lawrence Berkeley National Labs as well as Argonne National Labs. Current forks
/examples/buckingham/MgO_pareto_LBL
current development with: Oliver Ruebel (oruebel@lbl.gov)
Google Sites: https://sites.google.com/a/lbl.gov/sdav-chwm-pareto
Discusion: https://sites.google.com/a/lbl.gov/sdav-chwn-pareto/discussion
version controlled by: LBL
/examples/buckingham/MgO_pareto_ANL
current development with: Jeffrey Larson (jmlarson@anl.gov)
version controlled by: ANL
SVN password protected version control at: https://repo.anl-external.org