{ "info": { "author": "Ian Danforth", "author_email": "iandanforth@gmail.com", "bugtrack_url": null, "classifiers": [ "Programming Language :: Python", "Programming Language :: Python :: 3", "Programming Language :: Python :: 3.6", "Programming Language :: Python :: Implementation :: CPython", "Programming Language :: Python :: Implementation :: PyPy" ], "description": "\n# PyMuscle\n[](https://travis-ci.org/iandanforth/pymuscle)\n\nPyMuscle provides a motor unit based model of skeletal muscle. It simulates the\nrelationship between excitatory input and motor-unit output as well as fatigue \nover time.\n\nIt is compatible with [OpenAI Gym](https://gym.openai.com) environments and is \nintended to be useful for researchers in the machine learning community.\n\nPyMuscle can be used to enhance the realism of motor control for simulated \nagents. To get you started we provide a [toy example project](https://github.com/iandanforth/pymuscle/tree/master/examples) \nwhich uses PyMuscle in a simulation of arm curl and extension.\n\nOut of the box we provide a motor neuron pool model and a muscle fiber model\nbased on \"A motor unit-based model of muscle fatigue\" \n([Potvin and Fuglevand, 2017](http://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1005581)).\nIf you use this library as part of your research please cite that paper.\n\nWe hope to extend this model and support alternative models in the future.\n\n## More about PyMuscle\n\nMotor control in biological creatures is complex. PyMuscle allows you to capture\nsome of that complexity while remaining [performant](#performance). It provides \ngreater detail than sending torque values to simulated motors-as-joints but \nless detail (and computational cost) than a full biochemical model.\n\nPyMuscle is not tied to a particular physics library and can be used with a \nvariety of muscle body simulations. PyMuscle focuses on the relationship between \ncontrol signals (excitatory inputs to motor neurons) and per-motor-unit output.\n\nMotor unit output is dimensionless but can be interpreted as force. It can also\nbe used as a proxy for the contractile state of muscle bodies in the physics\nsim of your choice.\n\n# Background\n\n## Motor Units\n

