{
    "info": {
        "author": "Thiago P. Bueno",
        "author_email": "thiago.pbueno@gmail.com",
        "bugtrack_url": null,
        "classifiers": [
            "Development Status :: 3 - Alpha",
            "Environment :: Console",
            "Intended Audience :: Science/Research",
            "License :: OSI Approved :: GNU General Public License v3 (GPLv3)",
            "Natural Language :: English",
            "Operating System :: OS Independent",
            "Programming Language :: Python :: 3",
            "Topic :: Scientific/Engineering :: Artificial Intelligence"
        ],
        "description": "# pybayesbandit\n\nBayesian bandits in Python3.\n\n## Quickstart\n\n```text\n$ pip install pybayesbandit\n```\n\n## Usage\n\n```text\n$ pybayesbandit --help\nusage: pybayesbandit [-h] [-p PARAMS [PARAMS ...]] [-d MAXDEPTH] [-t TRIALS]\n                     [-C C] [-e EPISODES] [-hr HORIZON] [--plot] [-v]\n                     {random,ucb,thompson,vi,uct,rollout,aotree} {bernoulli}\n                     {total,simple}\n\nBayesian bandits in Python3.\n\npositional arguments:\n  {random,ucb,thompson,vi,uct,rollout,aotree}\n                        learner type\n  {bernoulli}           bandit type\n  {total,simple}        game setting\n\noptional arguments:\n  -h, --help            show this help message and exit\n  -p PARAMS [PARAMS ...], --params PARAMS [PARAMS ...]\n                        bandit parameters\n  -d MAXDEPTH, --maxdepth MAXDEPTH\n                        maximum number of timesteps in the tree lookahead\n                        (default=10)\n  -t TRIALS, --trials TRIALS\n                        number of trials in Monte-Carlo sampling (default=30)\n  -C C                  UCT exploration constant (default=2.0)\n  -e EPISODES, --episodes EPISODES\n                        number of simulation episodes (default=200)\n  -hr HORIZON, --horizon HORIZON\n                        number of timesteps in each episode (default=100)\n  --plot                plot cumulative regret\n  -v, --verbose         verbose mode\n```\n\n## Examples\n\n```text\n$ pybayesbandit ucb bernoulli total -p 0.5 0.8 0.3 -e 100 -hr 50 -v\n\nRunning pybayesbandit ...\n>> learner  = ucb\n>> bandit   = bernoulli([0.5, 0.8, 0.3])\n>> episodes = 100\n>> horizon  = 50\nDone in 0.257 sec.\n\nResults:\n>> Reward =  32.3900 \u00b1 3.4173\n>> Regret =   7.6530 \u00b1 1.7081\n```\n\n```text\n$ pybayesbandit thompson bernoulli total -p 0.5 0.8 0.3 -e 100 -hr 50 -v\n\nRunning pybayesbandit ...\n>> learner  = thompson\n>> bandit   = bernoulli([0.5, 0.8, 0.3])\n>> episodes = 100\n>> horizon  = 50\nDone in 0.297 sec.\n\nResults:\n>> Reward =  35.2200 \u00b1 3.8822\n>> Regret =   4.4560 \u00b1 2.6086\n```\n\n```text\n$ pybayesbandit uct bernoulli total -p 0.5 0.8 0.3 -e 100 -hr 50 --trials 15 --maxdepth 5 -v\n\nRunning pybayesbandit ...\n>> learner  = uct(trials=15, maxdepth=5, C=2.0)\n>> bandit   = bernoulli([0.5, 0.8, 0.3])\n>> episodes = 100\n>> horizon  = 50\nDone in 7.066 sec.\n\nResults:\n>> Reward =  36.2800 \u00b1 5.4828\n>> Regret =   3.4360 \u00b1 4.5856\n```\n\n## License\n\nCopyright (c) 2018 Thiago Pereira Bueno All Rights Reserved.\n\npybayesbandit is free software: you can redistribute it and/or modify it\nunder the terms of the GNU Lesser General Public License as published by\nthe Free Software Foundation, either version 3 of the License, or (at\nyour option) any later version.\n\npybayesbandit is distributed in the hope that it will be useful, but\nWITHOUT ANY WARRANTY; without even the implied warranty of\nMERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser\nGeneral Public License for more details.\n\nYou should have received a copy of the GNU Lesser General Public License\nalong with pybayesbandit. If not, see http://www.gnu.org/licenses/.",
        "description_content_type": "",
        "docs_url": null,
        "download_url": "",
        "downloads": {
            "last_day": -1,
            "last_month": -1,
            "last_week": -1
        },
        "home_page": "https://github.com/thiagopbueno/pybayesbandit",
        "keywords": "rl,bayesian,bandit-learning,belief-planning",
        "license": "GNU General Public License v3.0",
        "maintainer": "",
        "maintainer_email": "",
        "name": "pybayesbandit",
        "package_url": "https://pypi.org/project/pybayesbandit/",
        "platform": "",
        "project_url": "https://pypi.org/project/pybayesbandit/",
        "project_urls": {
            "Homepage": "https://github.com/thiagopbueno/pybayesbandit"
        },
        "release_url": "https://pypi.org/project/pybayesbandit/0.0.4/",
        "requires_dist": null,
        "requires_python": "",
        "summary": "Bayesian bandits in Python3.",
        "version": "0.0.4"
    },
    "last_serial": 4508893,
    "releases": {
        "0.0.2": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "4543fd6a0f8427674a3c39f0f77aa053",
                    "sha256": "eb1dc934281b2cc4677002053d2d9c21c33ab6e2460b0f9dad30eff368c6f1f6"
                },
                "downloads": -1,
                "filename": "pybayesbandit-0.0.2.tar.gz",
                "has_sig": false,
                "md5_digest": "4543fd6a0f8427674a3c39f0f77aa053",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 7809,
                "upload_time": "2018-11-11T16:51:42",
                "url": "https://files.pythonhosted.org/packages/ca/00/ba26bd8da75398c1412b31c108a30d5d622c62f4700b3c4e76dcb2001fb7/pybayesbandit-0.0.2.tar.gz"
            }
        ],
        "0.0.4": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "8a721189d78389320a161fb4972ba138",
                    "sha256": "4fd07382e1d3fb997f2c13a69a6dbad1b4f3d55025617437012298b8bc6fed08"
                },
                "downloads": -1,
                "filename": "pybayesbandit-0.0.4.tar.gz",
                "has_sig": false,
                "md5_digest": "8a721189d78389320a161fb4972ba138",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 9987,
                "upload_time": "2018-11-20T19:15:59",
                "url": "https://files.pythonhosted.org/packages/7b/e9/f8a51af72b57ef2802fdf8b5ad5ab2856f9272dd414429986204b04a18ba/pybayesbandit-0.0.4.tar.gz"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "8a721189d78389320a161fb4972ba138",
                "sha256": "4fd07382e1d3fb997f2c13a69a6dbad1b4f3d55025617437012298b8bc6fed08"
            },
            "downloads": -1,
            "filename": "pybayesbandit-0.0.4.tar.gz",
            "has_sig": false,
            "md5_digest": "8a721189d78389320a161fb4972ba138",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": null,
            "size": 9987,
            "upload_time": "2018-11-20T19:15:59",
            "url": "https://files.pythonhosted.org/packages/7b/e9/f8a51af72b57ef2802fdf8b5ad5ab2856f9272dd414429986204b04a18ba/pybayesbandit-0.0.4.tar.gz"
        }
    ]
}