{
    "info": {
        "author": "Nikita Savelyev",
        "author_email": "n.a.savelyev@gmail.com",
        "bugtrack_url": null,
        "classifiers": [
            "Development Status :: 3 - Alpha",
            "Intended Audience :: Developers",
            "Intended Audience :: Science/Research",
            "License :: OSI Approved :: GNU General Public License v3 (GPLv3)",
            "Programming Language :: Python :: 2",
            "Programming Language :: Python :: 3",
            "Topic :: Scientific/Engineering",
            "Topic :: Software Development"
        ],
        "description": "Datapot\n=======\n\n|Build Status| *Open source tool for machine learning on semi-structured\ndata that creates numeric object-feature matrix from JSON. The idea of\nDatapot is to make the process of data preparation and feature\nextraction automatic, easy and effective.*\n\nUsage\n-----\n\n**Install Datapot:**\n\n.. code:: bash\n\n    $ git clone https://github.com/bashalex/datapot.git\n    $ cd datapot\n    $ pip install .\n\nTo **create a Datapot** object simply write the following:\n\n.. code:: python\n\n    >>> import datapot as dp \n    >>> data = dp.DataPot()\n\nDataPot has two main methods:\n^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n\n-  fit()\n-  transform()\n\nMethod ``fit(self, data, limit)`` goes through the first N objects (N =\nlimit), passes the possible features to Transformers. Each Transformer\nevaluates if a feature from current field or a number of fields can be\ncreated. As a result a dict of features and Transformers is created.\n\nTo apply ``fit()`` to JSON file:\n\n.. code:: python\n\n    >>> f = open('data/matches_test.jsonlines', 'r')\n    >>> data.fit(f, limit=100)\n    >>> data\n    DataPot class instance\n     - number of features without transformation: 806\n     - number of new features: 315\n    features to transform: \n        (u'players.0.gold_t', [ComplexTransformer])\n        (u'picks_bans.0.is_pick', [BoolToIntTransformer])\n        (u'players.0.kills_log.0.unit', [TfidfTransformer])\n        (u'players.1.xp_t', [ComplexTransformer])\n        (u'picks_bans.1.is_pick', [BoolToIntTransformer])\n        (u'players.1.kills_log.0.unit', [TfidfTransformer])\n        ...\n\nMethod ``transform(self, data, verbose)`` generates a pandas. DataFrame\nwith new features that were detected on the fit() call. If parameter\nverbose is true, progress description is printed during the feature\nextraction.\n\n.. code:: python\n\n    >>> df = data.transform(f, verbose=False)\n    fit transformers...OK\n    num of new features: 315\n\nExamples\n--------\n\nLook for `more examples <https://github.com/bashalex/datapot/tree/master/notebooks/>`__ of using Datapot with\ndifferent datasets and more Transformer specific.\n\nFeatures\n--------\n\nDatapot provides many ways of extracting features from JSON-s.\n\nData types that can be processed: - Boolean - Numerical array (transform\narray to their sum divided by average length of array in training set) -\nTime series (\u0441alculate descriptive statistical properties of a given\ntime series) - Timestamp (date, time, day of week, day of month etc.) -\nText (bag of words tf-idf, word2vec) - Categorial (one-hot encoding,\ndimension reduction)\n\nAuthors\n-------\n\n-  Alex Bash\n-  Yuriy Mokriy\n-  Nikita Savelyev\n-  Michal Rozenwald\n-  Peter Romov\n\nDatapot is a course work project of `the Faculty of Computer\nScience <https://cs.hse.ru/en/>`__ of `the Higher School of\nEconomics <https://www.hse.ru/en/>`__.\n\n.. |Build Status| image:: https://travis-ci.org/bashalex/datapot.svg?branch=master\n   :target: https://travis-ci.org/bashalex/datapot",
        "description_content_type": null,
        "docs_url": null,
        "download_url": "UNKNOWN",
        "downloads": {
            "last_day": -1,
            "last_month": -1,
            "last_week": -1
        },
        "home_page": "https://github.com/bashalex/datapot",
        "keywords": null,
        "license": "GNU v3.0",
        "maintainer": null,
        "maintainer_email": null,
        "name": "datapot",
        "package_url": "https://pypi.org/project/datapot/",
        "platform": "UNKNOWN",
        "project_url": "https://pypi.org/project/datapot/",
        "project_urls": {
            "Download": "UNKNOWN",
            "Homepage": "https://github.com/bashalex/datapot"
        },
        "release_url": "https://pypi.org/project/datapot/0.1.3/",
        "requires_dist": null,
        "requires_python": null,
        "summary": "Library for automatic feature extraction from JSON-datasets",
        "version": "0.1.3"
    },
    "last_serial": 2910738,
    "releases": {
        "0.1": [],
        "0.1.2": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "63eed56ddbe05e994ad9d92d26a6105b",
                    "sha256": "3eb99671d5979dc11e7c4dee6556aacc1939a585b809e2ae6483364dd6b48c09"
                },
                "downloads": -1,
                "filename": "datapot-0.1.2.tar.gz",
                "has_sig": false,
                "md5_digest": "63eed56ddbe05e994ad9d92d26a6105b",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 15243,
                "upload_time": "2017-05-30T20:09:15",
                "url": "https://files.pythonhosted.org/packages/95/00/820591223fcf5c1d4aaebe973d38cd8d58e982cc91669d8b22a5e077b5ee/datapot-0.1.2.tar.gz"
            }
        ],
        "0.1.3": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "e92a2b2a3aa36d3650c5e98123a1bd96",
                    "sha256": "06696991b3477e4af9ae8753621607ac9015e6361eb0dcb2ce431e06863158ba"
                },
                "downloads": -1,
                "filename": "datapot-0.1.3.tar.gz",
                "has_sig": false,
                "md5_digest": "e92a2b2a3aa36d3650c5e98123a1bd96",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 15247,
                "upload_time": "2017-05-30T20:17:19",
                "url": "https://files.pythonhosted.org/packages/3c/5b/639e1534777e934cbb4fdec8abcdfbeb5a17aee045b6d29b7207a637367d/datapot-0.1.3.tar.gz"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "e92a2b2a3aa36d3650c5e98123a1bd96",
                "sha256": "06696991b3477e4af9ae8753621607ac9015e6361eb0dcb2ce431e06863158ba"
            },
            "downloads": -1,
            "filename": "datapot-0.1.3.tar.gz",
            "has_sig": false,
            "md5_digest": "e92a2b2a3aa36d3650c5e98123a1bd96",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": null,
            "size": 15247,
            "upload_time": "2017-05-30T20:17:19",
            "url": "https://files.pythonhosted.org/packages/3c/5b/639e1534777e934cbb4fdec8abcdfbeb5a17aee045b6d29b7207a637367d/datapot-0.1.3.tar.gz"
        }
    ]
}