Skip to content

Latest commit

 

History

History
59 lines (43 loc) · 2.13 KB

README.md

File metadata and controls

59 lines (43 loc) · 2.13 KB

OpenSim Moco paper

This repository holds the code and text for generating the manuscript on OpenSim Moco, a software toolkit for solving optimal control problems with OpenSim musculoskeletal models.

This manuscript was first released as a bioRxiv preprint: https://www.biorxiv.org/content/10.1101/839381v1.

To learn more about OpenSim Moco, visit https://simtk.org/projects/opensim-moco.

The code folder contains Python scripts to generate results for the manuscript.

The paper folder holds LaTeX source for generating a PDF of the manuscript.

The resources folder contains models and data used by the code.

The results and figures folders hold numerical results and figures used in the paper. You must run the Docker container to obtain the results and figures; we do not commit the results to the repository.

You can use Docker to reproduce the results and paper. First, install Docker on your computer. Then, run the following command:

# On Windows using Command Prompt (cmd).
docker run --volume %cd%:/output stanfordnmbl/mocopaper:ploscompbio1
# On Windows using PowerShell.
docker run --volume ${PWD}:/output stanfordnmbl/mocopaper:ploscompbio1
# On Mac/UNIX using bash.
docker run --volume $(pwd):/output stanfordnmbl/mocopaper:ploscompbio1

The container takes about 10 hours to run. The results, figures, and paper PDF will end up in your current directory.

If you want to run your own copy of the mocopaper repository instead of using the copy of mocopaper stored inside the container, use the following command instead:

docker run --volume <local-mocopaper-repo>:/mocopaper stanfordnmbl/mocopaper:ploscompbio1

The results are saved to the results and figures folders of <local-mocopaper-repo>, and the paper is saved to <local-mocopaper-repo>/paper/MocoPaper.pdf.

You can view the Docker container for generating this paper at https://hub.docker.com/repository/docker/stanfordnmbl/mocopaper.

It is possible to run these files on Windows and Mac, but we do not provide instructions.

Required Python packages

  • matplotlib
  • numpy
  • scipy
  • pillow
  • opencv-python