{
    "info": {
        "author": "Daniel Watson",
        "author_email": "daniel.watson@nyu.edu",
        "bugtrack_url": null,
        "classifiers": [
            "Development Status :: 3 - Alpha",
            "Intended Audience :: Developers",
            "Programming Language :: Python :: 3.6"
        ],
        "description": "# TFBuild\n\nTensorFlow Build is an open source library for better science in deep\nlearning. The main features are (1) a growing repository of layers, models,\ndatasets, and full training and evaluation pipelines; and (2) abstractions for\nthe above that allow seamless prototyping, experimentation, and\nreproducibility of research.\n\n## Installation\n\nWe recommend installing with pip: `pip install tfbuild`.\n\n\n## Basics\n\nIMPORTANT: this section assumes basic familiarity with TensorFlow.\n\nA TensorFlow graph includes all operations that are part of training and\ninference routines. Almost always, this makes models messy, especially when\ndoing things such as: separating training and inference (and possibly other\nmodes), training models in a layer-wise fashion, composing multiple models,\ntraining multiple models jointly, and more. TFBuild solves this by defining\nabstractions that take care of all the issues above.\n\n### Models\n\nModels only define the mathematical operations done to the input(s) to create\none or more outputs (e.g. loss, discrete output from logits). These should be\ninput agnostic, meaning that we make no choice as to how we feed the inputs.\n\nModels can define trainable variables, but they should not define training\noperations! There are many reasons to opt for this choice. For example, we\ndon't care about defining optimizers in the graph during inference. This would\nalso complicate situations where we combine models and then train if both\ndefine training operations.\n\nTFBuild has a growing repository of these models. Since users are likely\ninterested only in a small subset of these, they are not downloaded by\ndefault. Instead, they can be uploaded and downloaded using our CLI:\n\n```bash\n# Under construction.\n```\n\n### Wrappers\n\nWrappers, as the name suggests, wrap models with functionality. This could be,\nfor example, training or inference. They return the same model instance, but\nchange their behavior by adding more ops to the computational graph and adding\nnew bound methods. For example:\n\n```python\nfrom tfbuild.wrappers import TrainingWrapper\n\n...\n\nmodel = TrainingWrapper(MyModel('my_model'))\nmodel.train(examples)  # not valid if not wrapped\n```\n\n### Input Pipelines\n\nUNDER CONSTRUCTION.\n\n\n## Recipes\n\nUNDER CONSTRUCTION.\n\n\n## Uploading to TF Build collections\n\nUNDER CONSTRUCTION.\n\n\n## Contributing\n\nSee the contribution guidelines in `CONTRIBUTING.md`.\n\n\n",
        "description_content_type": "",
        "docs_url": null,
        "download_url": "",
        "downloads": {
            "last_day": -1,
            "last_month": -1,
            "last_week": -1
        },
        "home_page": "http://www.danielwatson.me",
        "keywords": "",
        "license": "Apache-2.0",
        "maintainer": "",
        "maintainer_email": "",
        "name": "tfbuild",
        "package_url": "https://pypi.org/project/tfbuild/",
        "platform": "",
        "project_url": "https://pypi.org/project/tfbuild/",
        "project_urls": {
            "Homepage": "http://www.danielwatson.me"
        },
        "release_url": "https://pypi.org/project/tfbuild/0.1.2/",
        "requires_dist": [
            "absl-py (==0.4.1)",
            "asn1crypto (==0.24.0)",
            "astor (==0.7.1)",
            "certifi (==2018.10.15)",
            "cffi (==1.11.5)",
            "chardet (==3.0.4)",
            "cryptography (==2.3.1)",
            "gast (==0.2.0)",
            "grpcio (==1.15.0)",
            "idna (==2.7)",
            "Markdown (==2.6.11)",
            "numpy (==1.14.5)",
            "protobuf (==3.6.1)",
            "pycparser (==2.19)",
            "PyJWT (==1.6.4)",
            "requests (==2.20.0)",
            "six (==1.11.0)",
            "tensorboard (==1.10.0)",
            "tensorflow (==1.10.1)",
            "termcolor (==1.1.0)",
            "urllib3 (==1.24)",
            "Werkzeug (==0.14.1)"
        ],
        "requires_python": "",
        "summary": "",
        "version": "0.1.2"
    },
    "last_serial": 4475691,
    "releases": {
        "0.1": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "0a3e4912678681bf54eae41925784450",
                    "sha256": "0f032e34246f2d84a6d21bf23f494260f1174bf772b570efbcedca8718775dce"
                },
                "downloads": -1,
                "filename": "tfbuild-0.1.macosx-10.13-x86_64.tar.gz",
                "has_sig": false,
                "md5_digest": "0a3e4912678681bf54eae41925784450",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 19889,
                "upload_time": "2018-09-17T17:11:53",
                "url": "https://files.pythonhosted.org/packages/e1/13/e6538bf6c42e89985eef62275f2b0bcf585f24de60092b49fda6c18a8310/tfbuild-0.1.macosx-10.13-x86_64.tar.gz"
            },
            {
                "comment_text": "",
                "digests": {
                    "md5": "bd6b68296f18763d99fdafcead03b964",
                    "sha256": "8cb1606c1206a6c8e396581f8055237ed383e90ed46a0fa4a94f8329cd3045f6"
                },
                "downloads": -1,
                "filename": "tfbuild-0.1-py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "bd6b68296f18763d99fdafcead03b964",
                "packagetype": "bdist_wheel",
                "python_version": "py3",
                "requires_python": null,
                "size": 14184,
                "upload_time": "2018-09-17T17:11:51",
                "url": "https://files.pythonhosted.org/packages/1d/7e/8c910140a4b67a9decea7db754401a6f628ba2213ed09591d5c199f00d94/tfbuild-0.1-py3-none-any.whl"
            }
        ],
        "0.1.1": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "3637f5df90f52c7cdcaa3ec89296c268",
                    "sha256": "ba52bd13dd38b153966e1adbd68a7da50e622a05b86dfb2d947403756943aa10"
                },
                "downloads": -1,
                "filename": "tfbuild-0.1.1-py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "3637f5df90f52c7cdcaa3ec89296c268",
                "packagetype": "bdist_wheel",
                "python_version": "py3",
                "requires_python": null,
                "size": 29555,
                "upload_time": "2018-11-12T01:06:06",
                "url": "https://files.pythonhosted.org/packages/5f/9c/ec684d31188016b2c8719a2d2bcbdd01f0aea605ccc83faf47a7f4311051/tfbuild-0.1.1-py3-none-any.whl"
            }
        ],
        "0.1.2": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "d08d23b2fc203ac011c49ccae66852cd",
                    "sha256": "db8507328c792ff5a604ce7f95dc4cf6749ac333c05175a84ad9b5a1b7c4148d"
                },
                "downloads": -1,
                "filename": "tfbuild-0.1.2-py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "d08d23b2fc203ac011c49ccae66852cd",
                "packagetype": "bdist_wheel",
                "python_version": "py3",
                "requires_python": null,
                "size": 29553,
                "upload_time": "2018-11-12T01:13:34",
                "url": "https://files.pythonhosted.org/packages/82/dc/c1543c3024b2b7de954fe604f23f20757ec177ba5962a0b7898ffa692fc2/tfbuild-0.1.2-py3-none-any.whl"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "d08d23b2fc203ac011c49ccae66852cd",
                "sha256": "db8507328c792ff5a604ce7f95dc4cf6749ac333c05175a84ad9b5a1b7c4148d"
            },
            "downloads": -1,
            "filename": "tfbuild-0.1.2-py3-none-any.whl",
            "has_sig": false,
            "md5_digest": "d08d23b2fc203ac011c49ccae66852cd",
            "packagetype": "bdist_wheel",
            "python_version": "py3",
            "requires_python": null,
            "size": 29553,
            "upload_time": "2018-11-12T01:13:34",
            "url": "https://files.pythonhosted.org/packages/82/dc/c1543c3024b2b7de954fe604f23f20757ec177ba5962a0b7898ffa692fc2/tfbuild-0.1.2-py3-none-any.whl"
        }
    ]
}