{
    "info": {
        "author": "Richard Marko",
        "author_email": "srk@48.io",
        "bugtrack_url": null,
        "classifiers": [
            "Development Status :: 5 - Production/Stable",
            "Intended Audience :: Developers",
            "Operating System :: OS Independent",
            "Programming Language :: Python"
        ],
        "description": "python-mnist\n============\n\nSimple MNIST and EMNIST data parser written in pure Python.\n\nMNIST is a database of handwritten digits available on http://yann.lecun.com/exdb/mnist/.\nEMNIST is an extended MNIST database https://www.nist.gov/itl/iad/image-group/emnist-dataset.\n\nRequirements\n------------\n\n- Python 2 or Python 3\n\nUsage\n-----\n\n- ``git clone https://github.com/sorki/python-mnist``\n- ``cd python-mnist``\n- Get MNIST data::\n\n        ./get_data.sh\n\n- Check preview with::\n\n        PYTHONPATH=. ./bin/mnist_preview\n\n\nInstallation\n------------\n\nGet the package from PyPi::\n\n        pip install python-mnist\n\nor install with ``setup.py``::\n\n        python setup.py install\n\nCode sample::\n\n  from mnist import MNIST\n  mndata = MNIST('./dir_with_mnist_data_files')\n  images, labels = mndata.load_training()\n\nTo enable loading of gzip-ed files use::\n\n  mndata.gz = True\n\nLibrary tries to load files named `t10k-images-idx3-ubyte` `train-labels-idx1-ubyte` `train-images-idx3-ubyte` and `t10k-labels-idx1-ubyte`.\nIf loading throws an exception check if these names match.\n\nEMNIST\n------\n\n- Get EMNIST data::\n\n        ./get_emnist_data.sh\n\n- Check preview with::\n\n        PYTHONPATH=. ./bin/emnist_preview\n\nTo use EMNIST datasets you need to call::\n\n        mndata.select_emnist('digits')\n\nWhere `digits` is one of the available EMNIST datasets. You can choose from\n\n - balanced\n - byclass\n - bymerge\n - digits\n - letters\n - mnist\n\nEMNIST loader uses gziped files by default, this can be disabled by by setting::\n\n        mndata.gz = False\n\nYou also need to unpack EMNIST files as `get_emnist_data.sh` script won't do it for you.\nEMNIST loader also needs to mirror and rotate images so it is a bit slower (If this is an\nissue for you, you should repack the data to avoid mirroring and rotation on each load).\n\nNotes\n-----\n\nThis package doesn't use `numpy` by design as when I've tried to find a working implementation\nall of them were based on some archaic version of `numpy` and none of them worked. This loads\ndata files with `struct.unpack` instead.\n\nExample\n-------\n\n::\n        $ PYTHONPATH=. ./bin/mnist_preview\n        Showing num: 3\n\n        ............................\n        ............................\n        ............................\n        ............................\n        ............................\n        ............................\n        .............@@@@@..........\n        ..........@@@@@@@@@@........\n        .......@@@@@@......@@.......\n        .......@@@........@@@.......\n        .................@@.........\n        ................@@@.........\n        ...............@@@@@........\n        .............@@@............\n        .............@.......@......\n        .....................@......\n        .....................@@.....\n        ....................@@......\n        ...................@@@......\n        .................@@@@.......\n        ................@@@@........\n        ....@........@@@@@..........\n        ....@@@@@@@@@@@@............\n        ......@@@@@@................\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/sorki/python-mnist",
        "keywords": "",
        "license": "BSD",
        "maintainer": "",
        "maintainer_email": "",
        "name": "python-mnist",
        "package_url": "https://pypi.org/project/python-mnist/",
        "platform": "",
        "project_url": "https://pypi.org/project/python-mnist/",
        "project_urls": {
            "Homepage": "https://github.com/sorki/python-mnist"
        },
        "release_url": "https://pypi.org/project/python-mnist/0.6/",
        "requires_dist": null,
        "requires_python": "",
        "summary": "Simple MNIST and EMNIST data parser written in pure Python",
        "version": "0.6"
    },
    "last_serial": 4141095,
    "releases": {
        "0.2": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "21859b9c6440c409ba0c249f4d5ba0dc",
                    "sha256": "1e53dd1d4a000d425d599fec4b4599a4ac6e3217c2ba8a2886c53f32e186ec3f"
                },
                "downloads": -1,
                "filename": "python-mnist-0.2.tar.gz",
                "has_sig": false,
                "md5_digest": "21859b9c6440c409ba0c249f4d5ba0dc",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 4006,
                "upload_time": "2015-09-09T15:40:21",
                "url": "https://files.pythonhosted.org/packages/0c/0a/3be7199f989238d2396e0a39bd30e191e3cd496d7a144cd27bfb30f3462c/python-mnist-0.2.tar.gz"
            }
        ],
        "0.3": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "f8ba09f3fb2cd922f2b678e4b32b61ba",
                    "sha256": "30c6c87e3baecb826ff137907c76416431180a06dc07a45e916d77f8d160ff65"
                },
                "downloads": -1,
                "filename": "python-mnist-0.3.tar.gz",
                "has_sig": false,
                "md5_digest": "f8ba09f3fb2cd922f2b678e4b32b61ba",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 5665,
                "upload_time": "2015-09-09T15:45:55",
                "url": "https://files.pythonhosted.org/packages/c7/30/bb7a477d66f8a409c7b3ad37537e17e4f218faef8a05edff860cc2b2441e/python-mnist-0.3.tar.gz"
            }
        ],
        "0.4": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "59442f30965fabaf04923efbdc937f78",
                    "sha256": "c56da1a2548249589c19cb7f42a19a596aeeea882389e1c59874f0a08a1c6614"
                },
                "downloads": -1,
                "filename": "python-mnist-0.4.tar.gz",
                "has_sig": false,
                "md5_digest": "59442f30965fabaf04923efbdc937f78",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 9845,
                "upload_time": "2018-01-09T14:34:45",
                "url": "https://files.pythonhosted.org/packages/ca/08/715f8172fc60a5fcc046e193ddc6cd0fb6e1986e42660945294f0ad2cdc6/python-mnist-0.4.tar.gz"
            }
        ],
        "0.5": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "8eb347d5397434581d1ae95c77f461c9",
                    "sha256": "bc2bba70b51e18d4e1bce8936364dbfed02485e248e02e46e1e052da40c8ea9f"
                },
                "downloads": -1,
                "filename": "python-mnist-0.5.tar.gz",
                "has_sig": false,
                "md5_digest": "8eb347d5397434581d1ae95c77f461c9",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 9948,
                "upload_time": "2018-01-09T14:42:16",
                "url": "https://files.pythonhosted.org/packages/cd/e5/05e37dcf2ed8966a618cd348721e68a2bfad302438344acc16ed94e349fd/python-mnist-0.5.tar.gz"
            }
        ],
        "0.6": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "767213c553c2e25d15c042f05c78ec66",
                    "sha256": "5d59a44335eccb4b310efb2ebb76f44e8588a1732cfb4923f4a502b61d8b653a"
                },
                "downloads": -1,
                "filename": "python-mnist-0.6.tar.gz",
                "has_sig": false,
                "md5_digest": "767213c553c2e25d15c042f05c78ec66",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 10736,
                "upload_time": "2018-08-06T16:33:57",
                "url": "https://files.pythonhosted.org/packages/05/9c/f1c1e4d011b01ac436bba0ac6715b3f988bb7f8fec6f21f89cf820aa33e1/python-mnist-0.6.tar.gz"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "767213c553c2e25d15c042f05c78ec66",
                "sha256": "5d59a44335eccb4b310efb2ebb76f44e8588a1732cfb4923f4a502b61d8b653a"
            },
            "downloads": -1,
            "filename": "python-mnist-0.6.tar.gz",
            "has_sig": false,
            "md5_digest": "767213c553c2e25d15c042f05c78ec66",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": null,
            "size": 10736,
            "upload_time": "2018-08-06T16:33:57",
            "url": "https://files.pythonhosted.org/packages/05/9c/f1c1e4d011b01ac436bba0ac6715b3f988bb7f8fec6f21f89cf820aa33e1/python-mnist-0.6.tar.gz"
        }
    ]
}