Skip to content

Tools for converting path sampling simulations to OPS format

License

Notifications You must be signed in to change notification settings

singraber/OPSPiggybacker

 
 

Repository files navigation

Codacy Badge Build Status Coverage Status Documentation Status

OPSPiggybackers

The goal of this little toy project is to create an OPS mover that creates a fake simulation with predetermined move behavior (what trial trajectories are generated; whether they are accepted). This could be useful either for testing, or for mapping output from another software package into formats that OPS can analyze.

The approach we use is to create a generic mover, which pretends to be a one-way shooting mover. Note that we can always add additional fake moves in a similar way: fake replica exchange, fake minus, etc. However, that requires a little more generalization.

One of the tricky bits is that it is possible that not all the information will be available. Of particular note, we might have a system that doesn't save all its frames. Figuring out how to handle that will be more difficult.

Installation

Simple installation

(TODO: this is whatever the E-CAM preferred approach is)

Developer installation

If you intend to develop, or stay up to date with the bleeding edge of development, you should use the following installation procedure:

  1. Install OpenPathSampling. All other required packages for OPSPiggybacker are also required packages for OPS.
  2. Clone this git repository.
  3. Change to the root directory of your clone, and use python setup.py develop or pip install -e .. You may need to preface that with sudo.

Roadmap

Current release: 0.1.0

Roadmap:

  • 0.1: Support for shooting moves with only one ensemble (e.g., TPS)
  • Coming soon: Support for shooting moves with multiple ensembles (e.g., TIS)
  • Coming later: Generalize mover input to handle arbitrary moves (not just shooting)

About

Tools for converting path sampling simulations to OPS format

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Python 97.3%
  • Shell 2.7%