There is a newer version of this record available.

Software Open Access

ljvmiranda921/gym-lattice: Major Release (v0.1.0)

Lester James Validad Miranda


JSON-LD (schema.org) Export

{
  "description": "<p>Major Release (v0.1.0)</p>\n\n<p>Gym-lattice is an HP 2D Lattice Environment with a Gym-like API for the protein folding problem.</p>\n\n<p>This is a Python library that formulates Lau and Dill&#39;s (1989) hydrophobic-polar two-dimensional lattice model as a reinforcement learning problem. It follows OpenAI Gym&#39;s API, easing integration for reinforcement learning solutions.</p>\n\n<p>Features</p>\n\n<ul>\n\t<li>OpenAI Integration: uses Gym&#39;s API to ease compatibility to reinforcement learning solutions.</li>\n\t<li>Lattice 2D Environment: implements Dill and Lau&#39;s two-dimensional lattice as an RL problem.</li>\n\t<li>Command-line rendering environment: the method <code>render()</code> draws the chain on the command-line.</li>\n</ul>\n\n<p>Additionally, there is an option to set the penalty parameters for training the agent, this includes:</p>\n\n<ul>\n\t<li>Collision penalty (<code>collision_penalty</code>): accounts for the time whenever the agent decides to assign a molecule to an already-occupied space; and</li>\n\t<li>Trap penalty (<code>trap_penalty</code>): induces heavy deductions whenever the agent traps itself and cannot accomplish the task anymore.</li>\n</ul>\n\n<p>Tests</p>\n\n<ul>\n\t<li>Error-handling: all public-facing methods now use <code>Exceptions</code> instead of <code>asserts</code> when catching errors.</li>\n\t<li>Testing with pytest and tox: unit-testing is being done with <code>pytest</code> and <code>tox</code>.</li>\n</ul>", 
  "license": "", 
  "creator": [
    {
      "affiliation": "Waseda University", 
      "@id": "https://orcid.org/0000-0002-7872-6464", 
      "@type": "Person", 
      "name": "Lester James Validad Miranda"
    }
  ], 
  "url": "https://zenodo.org/record/1218143", 
  "codeRepository": "https://github.com/ljvmiranda921/gym-lattice/tree/v0.1.0", 
  "datePublished": "2018-04-14", 
  "version": "v0.1.0", 
  "keywords": [
    "reinforcement learning", 
    "protein folding problem", 
    "open source software"
  ], 
  "@context": "https://schema.org/", 
  "identifier": "https://doi.org/10.5281/zenodo.1218143", 
  "@id": "https://doi.org/10.5281/zenodo.1218143", 
  "@type": "SoftwareSourceCode", 
  "name": "ljvmiranda921/gym-lattice: Major Release (v0.1.0)"
}
57
1
views
downloads
All versions This version
Views 576
Downloads 10
Data volume 102.2 kB0 Bytes
Unique views 526
Unique downloads 10

Share

Cite as