{
    "info": {
        "author": "ghcollin",
        "author_email": "",
        "bugtrack_url": null,
        "classifiers": [
            "Development Status :: 4 - Beta",
            "Intended Audience :: Developers",
            "Intended Audience :: Science/Research",
            "License :: OSI Approved :: MIT License",
            "Programming Language :: Python :: 2.7",
            "Programming Language :: Python :: 3",
            "Topic :: Scientific/Engineering :: Mathematics",
            "Topic :: Software Development :: Libraries :: Python Modules"
        ],
        "description": "`tftables <https://github.com/ghcollin/tftables>`_ allows convenient access to HDF5 files with Tensorflow.\nA class for reading batches of data out of arrays or tables is provided.\nA secondary class wraps both the primary reader and a Tensorflow FIFOQueue for straight-forward streaming \nof data from HDF5 files into Tensorflow operations.\n\nThe library is backed by `multitables <https://github.com/ghcollin/multitables>`_ for high-speed reading of HDF5\ndatasets. ``multitables`` is based on PyTables (``tables``), so this library can make use of any compression algorithms\nthat PyTables supports.\n\nLicence\n=======\n\nThis software is distributed under the MIT licence. \nSee the `LICENSE.txt <https://github.com/ghcollin/tftables/blob/master/LICENSE.txt>`_ file for details.\n\nInstallation\n============\n\n::\n\n    pip install tftables\n\nAlternatively, to install from HEAD, run\n\n::\n\n    pip install git+https://github.com/ghcollin/tftables.git\n\nYou can also `download <https://github.com/ghcollin/tftables/archive/master.zip>`_\nor `clone the repository <https://github.com/ghcollin/tftables>`_ and run\n\n::\n\n    python setup.py install\n\n``tftables`` depends on ``multitables``, ``numpy`` and ``tensorflow``. The package is compatible with the latest versions of python\n2 and 3.\n\nQuick start\n===========\n\nAn example of accessing a table in a HDF5 file.\n\n.. code:: python\n\n    import tftables\n    import tensorflow as tf\n\n    with tf.device('/cpu:0'):\n        # This function preprocesses the batches before they\n        # are loaded into the internal queue.\n        # You can cast data, or do one-hot transforms.\n        # If the dataset is a table, this function is required.\n        def input_transform(tbl_batch):\n            labels = tbl_batch['label']\n            data = tbl_batch['data']\n\n            truth = tf.to_float(tf.one_hot(labels, num_labels, 1, 0))\n            data_float = tf.to_float(data)\n\n            return truth, data_float\n\n        # Open the HDF5 file and create a loader for a dataset.\n        # The batch_size defines the length (in the outer dimension)\n        # of the elements (batches) returned by the reader.\n        # Takes a function as input that pre-processes the data.\n        loader = tftables.load_dataset(filename='path/to/h5_file.h5',\n                                       dataset_path='/internal/h5/path',\n                                       input_transform=input_transform,\n                                       batch_size=20)\n\n    # To get the data, we dequeue it from the loader.\n    # Tensorflow tensors are returned in the same order as input_transformation\n    truth_batch, data_batch = loader.dequeue()\n\n    # The placeholder can then be used in your network\n    result = my_network(truth_batch, data_batch)\n\n    with tf.Session() as sess:\n\n        # This context manager starts and stops the internal threads and\n        # processes used to read the data from disk and store it in the queue.\n        with loader.begin(sess):\n            for _ in range(num_iterations):\n                sess.run(result)\n\n\nIf the dataset is an array instead of a table. Then ``input_transform`` can be omitted\nif no pre-processing is required. If only a single pass through the dataset is desired,\nthen you should pass ``cyclic=False`` to ``load_dataset``.\n\n\nExamples\n========\n\nSee the `unit tests <https://github.com/ghcollin/tftables/blob/master/tftables_test.py>`_ for complete examples.\n\nExamples\n========\n\nSee the `how-to <http://tftables.readthedocs.io/en/latest/howto.html>`_ for more in-depth documentation, and the\n`unit tests <https://github.com/ghcollin/tftables/blob/master/tftables_test.py>`_ for complete examples.\n\nDocumentation\n=============\n\n`Online documentation <http://tftables.readthedocs.io/en/latest/>`_ is available.\nA `how to <http://tftables.readthedocs.io/en/latest/howto.html>`_ gives a basic overview of the library.\n\nOffline documentation can be built from the ``docs`` folder using ``sphinx``.\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/ghcollin/tftables",
        "keywords": "tensorflow HDF5",
        "license": "MIT",
        "maintainer": "",
        "maintainer_email": "",
        "name": "tftables",
        "package_url": "https://pypi.org/project/tftables/",
        "platform": "",
        "project_url": "https://pypi.org/project/tftables/",
        "project_urls": {
            "Homepage": "https://github.com/ghcollin/tftables"
        },
        "release_url": "https://pypi.org/project/tftables/1.1.2/",
        "requires_dist": [
            "multitables",
            "numpy (!=1.10.1)",
            "tensorflow"
        ],
        "requires_python": "",
        "summary": "Interface for reading HDF5 files into Tensorflow.",
        "version": "1.1.2"
    },
    "last_serial": 3586176,
    "releases": {
        "1.1.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "e5c922c3c75b135c8b7622d4447c8a71",
                    "sha256": "5c871f18b59c4ec69a63030be6bfffb20d797f5132beae08d183336dc3c50ecb"
                },
                "downloads": -1,
                "filename": "tftables-1.1.0-py2.py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "e5c922c3c75b135c8b7622d4447c8a71",
                "packagetype": "bdist_wheel",
                "python_version": "py2.py3",
                "requires_python": null,
                "size": 12622,
                "upload_time": "2017-03-07T23:09:30",
                "url": "https://files.pythonhosted.org/packages/e2/1b/959f74fa6138ccac4bd31a97becec502a76eecab44b252756f497342f3cb/tftables-1.1.0-py2.py3-none-any.whl"
            }
        ],
        "1.1.1": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "32891afbdb76e1833676f40fb1ca8007",
                    "sha256": "fe74647f928e79fec9ab3d88bec9be2c72825e098cd89bd2046549c535f34324"
                },
                "downloads": -1,
                "filename": "tftables-1.1.1-py2.py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "32891afbdb76e1833676f40fb1ca8007",
                "packagetype": "bdist_wheel",
                "python_version": "py2.py3",
                "requires_python": null,
                "size": 12729,
                "upload_time": "2017-03-08T07:50:11",
                "url": "https://files.pythonhosted.org/packages/68/b3/bd11c71ce1a73ac1baa0efc4e19434cd80c086d6c9a8b7b08d871fee1544/tftables-1.1.1-py2.py3-none-any.whl"
            }
        ],
        "1.1.2": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "dad740af62f20caa901bee1264534a13",
                    "sha256": "794aebe35618adbfe8e1f2e8adef1c71c85acabe9aa5112cb5c542d1a1bf98ef"
                },
                "downloads": -1,
                "filename": "tftables-1.1.2-py2.py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "dad740af62f20caa901bee1264534a13",
                "packagetype": "bdist_wheel",
                "python_version": "py2.py3",
                "requires_python": null,
                "size": 12420,
                "upload_time": "2018-02-16T03:17:56",
                "url": "https://files.pythonhosted.org/packages/e8/bc/7be1e26747cea5db241480ad84bafff93fbb6db4f502bc916bd89d8ecdcf/tftables-1.1.2-py2.py3-none-any.whl"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "dad740af62f20caa901bee1264534a13",
                "sha256": "794aebe35618adbfe8e1f2e8adef1c71c85acabe9aa5112cb5c542d1a1bf98ef"
            },
            "downloads": -1,
            "filename": "tftables-1.1.2-py2.py3-none-any.whl",
            "has_sig": false,
            "md5_digest": "dad740af62f20caa901bee1264534a13",
            "packagetype": "bdist_wheel",
            "python_version": "py2.py3",
            "requires_python": null,
            "size": 12420,
            "upload_time": "2018-02-16T03:17:56",
            "url": "https://files.pythonhosted.org/packages/e8/bc/7be1e26747cea5db241480ad84bafff93fbb6db4f502bc916bd89d8ecdcf/tftables-1.1.2-py2.py3-none-any.whl"
        }
    ]
}