{
    "info": {
        "author": "Moriaki Saigusa",
        "author_email": "moriaki3193@gmail.com",
        "bugtrack_url": null,
        "classifiers": [
            "License :: OSI Approved :: MIT License",
            "Programming Language :: Python :: 3.7"
        ],
        "description": "# cdl2r\nCombination Dependent Learning to Rank (\u7d44\u5408\u305b\u4f9d\u5b58\u578b\u30e9\u30f3\u30ad\u30f3\u30b0\u5b66\u7fd2).\n\n## requirements\n- Python 3.6.x ~, 3.7.x ~\n\n## dependencies\n- NumPy\n- Pandas\n\n## installation\n```shell\n$ pip install cdl2r\n```\n\n## usage\n### 1. prepare your dataset\nThe dataset format is like SVM-rank one.\nThe difference is `eid` must be specified in a line.\nHere is a definition of a line.\n`|` symbol means `OR` (so `<str>|<int>` means the value must have either str or int type).\n\n```txt\n<line>     .=. <label> qid:<qid> eid:<eid> <features>#<comments>\n\n<label>    .=. <float>|<str as a class>\n<qid>      .=. <str>|<int>\n<eid>      .=. <str>|<int>\n<features> .=. <dim>:<value>\n<dim>      .=. <0 or Natural Number>\n<value>    .=. <float>\n<comments> .=. <Any text will do>\n```\n\nLet me show you an example.\n\n```txt\n0.5 qid:1 eid:x 1:0.1 2:-0.2 3:0.3#comment A\n0.0 qid:1 eid:y 1:-0.1 2:0.2 4:0.4\n-0.5 qid:1 eid:z 2:-0.2 3:0.3 4:-0.4#comment C\n0.5 qid:2 eid:y 1:0.1 2:-0.2 3:0.3\n0.0 qid:2 eid:z 1:-0.1 2:0.2 4:0.4\n-0.5 qid:2 eid:w 2:-0.2 3:0.3 4:-0.4#comment E\n```\n\n### 2. loading your dataset\n```python\nfrom cdl2r.dataset import load_data\n\n# loading dataset as a DataFrame object\ndata_path = '/path/to/dataset'\nn_dimensions = 10\ntrain = load_data(data_path, n_dimensions)\n# train.columns\n# >>> Index(['label', 'qid', 'eid', 'features'], dtype='object')\n```\n\n### 3. fitting the model\n```python\nfrom cdl2r.models import CDFMRegressor\n\n# define your model\nmodel = CDFMRegressor(n_factors=8, n_iterations=300, init_eta=1e-2)\n# fitting, printing out epoch losses if verbose is True\nmodel.fit(train, verbose=True)\n```\n\n### 4. save the model\n```python\nimport pickle\n\nwith open('/path/to/file.pkl', mode='wb') as fp:\n    pickle.dump(model, fp)\n```\n\n### 5. make prediction\n```python\n# loading test dataset\ntest = load_data(test_path, n_dimensions)\npred = model.predict(test)\n# pred.columns\n# >>> Index(['pred_label', 'qid', 'eid', 'features'], dtype='object')\n```\n\n## development\n### build Cython modules\n```shell\n$ python setup.py build_ext --inplace\n```\n\n### profiling\n```shell\n# decorate a method with `@profile` in a script where you want to profile.\n$ kernprof -l -v <script>.py\n```\n\n### pylint\n- max-line-length: 130\n- disable snake-case\n\n### release\n```shell\n# build\n$ python setup.py bdist_whell\n\n# testing upload\n$ twine upload --repository testpypi dist/<cdl2r-version-pkg>\n$ pip install --index-url https://test.pypi.org/simple/<cdl2r-version-pkg>\n\n# upload\n$ twine upload --repository pypi dist<cdl2r-version-pkg>\n```\n\n\n",
        "description_content_type": "",
        "docs_url": null,
        "download_url": "",
        "downloads": {
            "last_day": -1,
            "last_month": -1,
            "last_week": -1
        },
        "home_page": "https://github.com/moriaki3193/cdl2r",
        "keywords": "cdl2r L2R LETOR",
        "license": "Type license() to see the full license text",
        "maintainer": "",
        "maintainer_email": "",
        "name": "cdl2r",
        "package_url": "https://pypi.org/project/cdl2r/",
        "platform": "",
        "project_url": "https://pypi.org/project/cdl2r/",
        "project_urls": {
            "Homepage": "https://github.com/moriaki3193/cdl2r"
        },
        "release_url": "https://pypi.org/project/cdl2r/0.1.2/",
        "requires_dist": [
            "numpy",
            "pandas"
        ],
        "requires_python": "",
        "summary": "Combination Dependent Learning to Rank",
        "version": "0.1.2"
    },
    "last_serial": 4615255,
    "releases": {
        "0.1.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "e67c91ff00dc6f863b4d95422f3f143d",
                    "sha256": "895888134da92248ac00baa27acbb1d6a82af4ac69b9b1a931de44a0d06bc73c"
                },
                "downloads": -1,
                "filename": "cdl2r-0.1.0-py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "e67c91ff00dc6f863b4d95422f3f143d",
                "packagetype": "bdist_wheel",
                "python_version": "py3",
                "requires_python": null,
                "size": 17965,
                "upload_time": "2018-12-17T09:21:16",
                "url": "https://files.pythonhosted.org/packages/3d/25/04ae85f53effc5b4298cfac2fcc2b30cb747a861c2bb29706994d19b4192/cdl2r-0.1.0-py3-none-any.whl"
            }
        ],
        "0.1.1": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "2b43a236465824bb8bde907414470c57",
                    "sha256": "1f447bd4607c97d94d20b79ade4d19d14bb649a49a647eac04d0b84a461c2daf"
                },
                "downloads": -1,
                "filename": "cdl2r-0.1.1-py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "2b43a236465824bb8bde907414470c57",
                "packagetype": "bdist_wheel",
                "python_version": "py3",
                "requires_python": null,
                "size": 18426,
                "upload_time": "2018-12-18T12:59:42",
                "url": "https://files.pythonhosted.org/packages/61/3b/0dfc164c2f7c1ee5e03a4d0e7c5ef2909ce498998021fd41d4a505cc1cde/cdl2r-0.1.1-py3-none-any.whl"
            }
        ],
        "0.1.2": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "e5b42973da349b38225fa569f30ae794",
                    "sha256": "54ca7c0cab90838bb89a5344de16a41f3c6d760b591d07a1cc0ff51a2ce7e391"
                },
                "downloads": -1,
                "filename": "cdl2r-0.1.2-cp36-cp36m-macosx_10_14_x86_64.whl",
                "has_sig": false,
                "md5_digest": "e5b42973da349b38225fa569f30ae794",
                "packagetype": "bdist_wheel",
                "python_version": "cp36",
                "requires_python": null,
                "size": 35376,
                "upload_time": "2018-12-19T06:19:43",
                "url": "https://files.pythonhosted.org/packages/96/be/9c399c83d3667bd618d6f814911d2be229ca56f69ae7d77ea5ff9466d535/cdl2r-0.1.2-cp36-cp36m-macosx_10_14_x86_64.whl"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "e5b42973da349b38225fa569f30ae794",
                "sha256": "54ca7c0cab90838bb89a5344de16a41f3c6d760b591d07a1cc0ff51a2ce7e391"
            },
            "downloads": -1,
            "filename": "cdl2r-0.1.2-cp36-cp36m-macosx_10_14_x86_64.whl",
            "has_sig": false,
            "md5_digest": "e5b42973da349b38225fa569f30ae794",
            "packagetype": "bdist_wheel",
            "python_version": "cp36",
            "requires_python": null,
            "size": 35376,
            "upload_time": "2018-12-19T06:19:43",
            "url": "https://files.pythonhosted.org/packages/96/be/9c399c83d3667bd618d6f814911d2be229ca56f69ae7d77ea5ff9466d535/cdl2r-0.1.2-cp36-cp36m-macosx_10_14_x86_64.whl"
        }
    ]
}