{
    "info": {
        "author": "Zhen Li, Tong Wang",
        "author_email": "tong-wang@uiowa.edu",
        "bugtrack_url": null,
        "classifiers": [
            "Development Status :: 3 - Alpha",
            "Intended Audience :: Developers",
            "License :: OSI Approved :: MIT License",
            "Programming Language :: Python :: 3.5",
            "Topic :: Software Development :: Build Tools"
        ],
        "description": "Bayesian Rule Set Mining\n\nFind the rule set from the data\nThe input data should follow the following format:\nX has to be a pandas DataFrame\nall the column names can not contain '_' or '<'\nand the column names can not be pure numbers\nThe categorical data should be represented in string\n(For example, gender needs to be 'male'/'female',\n or '0'/'1' to represent male and female respectively.)\nThe parser will only recognize this format of data.\nSo transform the data set first before using the\nfunctions.\ny hass to be a numpy.ndarray\n\nreference:\nWang, Tong, et al. \"Bayesian rule sets for interpretable classification.\" \nData Mining (ICDM), 2016 IEEE 16th International Conference on. IEEE, 2016.\n\nThe program is very picky on the input data format\nX needs to be a pandas DataFrame,\ny needs to be a nd.array\n\n    Parameters\n    ----------\n    max_rules : int, default 5000\n        Maximum number of rules when generating rules\n\n    max_iter : int, default 200\n        Maximun number of iteratations to find the rule set\n\n    chians : int, default 1\n        Number of chains that run in parallel\n\n    support : int, default 5\n        The support is the percentile threshold for the itemset\n        to be selected.\n\n    maxlen : int, default 3\n        The maximum number of items in a rule\n\n    #note need to replace all the alpha_1 to alpha_+\n    alpha_1 : float, default 100\n        alpha_+\n\n    beta_1 : float, default 1\n        beta_+\n\n    alpha_2 : float, default 100\n        alpha_-\n\n    beta_2 : float, default 1\n        beta_-\n\n    alpha_l : float array, shape (maxlen+1,)\n        default all elements to be 1\n\n    beta_l : float array, shape (maxlen+1,)\n        default corresponding patternSpace\n\n    level : int, default 4\n        Number of intervals to deal with numerical continous features\n\n    neg : boolean, default True\n        Negate the features\n\n    add_rules : list, default empty\n        User defined rules to add\n        it needs user to add numerical version of the rules\n\n    criteria : str, default 'precision'\n        When there are rules more than max_rules,\n        the criteria used to filter rules\n\n    greedy_initilization : boolean, default False\n        Wether start the rule set using a greedy\n        initilization (according to accuracy)\n\n    greedy_threshold : float, default 0.05\n        Threshold for the greedy algorithm\n        to find the starting rule set\n\n    propose_threshold : float, default 0.1\n        Threshold for a proposal to be accepted\n\n    method : str, default 'fpgrowth'\n        The method used to generate rules.\n        Can be 'fpgrowth' or 'forest'\n        Notice that if there are potentially many rules\n        then fpgrowth is not a good method as it will\n        have memory issue (because the rule screening is\n        after rule generations).\n\n\nSample usage\n\nfrom ruleset import *\n\ndf = pd.read_csv('data/adult.dat', header=None, sep=',', names=['age', 'workclass', 'fnlwgt', 'education', 'educationnum', 'matritalstatus', 'occupation', 'relationship', 'race', 'sex', 'capitalgain', 'capitalloss', 'hoursperweek', 'nativecountary', 'income'])\ny = (df['income'] == '>50K').as_matrix()\ndf.drop('income', axis=1, inplace=True)\nmodel = BayesianRuleSet(method='forest')\nmodel.fit(df, y)\n\n\n",
        "description_content_type": null,
        "docs_url": null,
        "download_url": "",
        "downloads": {
            "last_day": -1,
            "last_month": -1,
            "last_week": -1
        },
        "home_page": "https://github.com/zli37/bayesianRuleSet",
        "keywords": "data mining analysis",
        "license": "MIT",
        "maintainer": "",
        "maintainer_email": "",
        "name": "ruleset",
        "package_url": "https://pypi.org/project/ruleset/",
        "platform": "",
        "project_url": "https://pypi.org/project/ruleset/",
        "project_urls": {
            "Homepage": "https://github.com/zli37/bayesianRuleSet"
        },
        "release_url": "https://pypi.org/project/ruleset/1.0.1/",
        "requires_dist": [
            "numpy",
            "pandas",
            "scipy",
            "sklearn"
        ],
        "requires_python": ">=3",
        "summary": "Bayesian Rule Set Mining",
        "version": "1.0.1"
    },
    "last_serial": 3269188,
    "releases": {
        "1.0.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "7a2bb64836dd3949194c58649dffb00b",
                    "sha256": "1a68f757229a12fd62f277cb2860f0eba30332cd8c1af763cbc38c6187d82fba"
                },
                "downloads": -1,
                "filename": "ruleset-1.0.0-py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "7a2bb64836dd3949194c58649dffb00b",
                "packagetype": "bdist_wheel",
                "python_version": "py3",
                "requires_python": ">=3",
                "size": 15367,
                "upload_time": "2017-08-15T20:49:23",
                "url": "https://files.pythonhosted.org/packages/bd/6e/26642d8ef83dfccb0f6308dd928c9d4cdeaae895a67ef2dac079f736f510/ruleset-1.0.0-py3-none-any.whl"
            }
        ],
        "1.0.1": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "9c694701f513143797631723a823f119",
                    "sha256": "09ae4f86912094d7a652ff077ffd4dbfede7058154fb3b4d0d23a3735c661fed"
                },
                "downloads": -1,
                "filename": "ruleset-1.0.1-py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "9c694701f513143797631723a823f119",
                "packagetype": "bdist_wheel",
                "python_version": "py3",
                "requires_python": ">=3",
                "size": 15367,
                "upload_time": "2017-10-22T03:32:52",
                "url": "https://files.pythonhosted.org/packages/9f/3e/f1f88d114bd67c609756acb7ac6f4140c47581fdb30907b1ecf0a95ff32e/ruleset-1.0.1-py3-none-any.whl"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "9c694701f513143797631723a823f119",
                "sha256": "09ae4f86912094d7a652ff077ffd4dbfede7058154fb3b4d0d23a3735c661fed"
            },
            "downloads": -1,
            "filename": "ruleset-1.0.1-py3-none-any.whl",
            "has_sig": false,
            "md5_digest": "9c694701f513143797631723a823f119",
            "packagetype": "bdist_wheel",
            "python_version": "py3",
            "requires_python": ">=3",
            "size": 15367,
            "upload_time": "2017-10-22T03:32:52",
            "url": "https://files.pythonhosted.org/packages/9f/3e/f1f88d114bd67c609756acb7ac6f4140c47581fdb30907b1ecf0a95ff32e/ruleset-1.0.1-py3-none-any.whl"
        }
    ]
}