{
    "info": {
        "author": "CyberZHG",
        "author_email": "CyberZHG@gmail.com",
        "bugtrack_url": null,
        "classifiers": [
            "License :: OSI Approved :: MIT License",
            "Operating System :: OS Independent",
            "Programming Language :: Python :: 2.7",
            "Programming Language :: Python :: 3.6"
        ],
        "description": "# Keras AdaBound\n\n[![Travis](https://travis-ci.org/CyberZHG/keras-adabound.svg)](https://travis-ci.org/CyberZHG/keras-adabound)\n[![Coverage](https://coveralls.io/repos/github/CyberZHG/keras-adabound/badge.svg?branch=master)](https://coveralls.io/github/CyberZHG/keras-adabound)\n\n[AdaBound](https://github.com/Luolc/AdaBound) optimizer in Keras.\n\n## Install\n\n```bash\npip install keras-adabound\n```\n\n## Usage\n\n### Use the optimizer\n\n```python\nfrom keras_adabound import AdaBound\n\nmodel.compile(optimizer=AdaBound(lr=1e-3, final_lr=0.1), loss=model_loss)\n```\n\n### Load with custom objects\n\n```python\nfrom keras_adabound import AdaBound\n\nmodel = keras.models.load_model(model_path, custom_objects={'AdaBound': AdaBound})\n```\n\n### About weight decay\n\nThe optimizer does not have an argument named `weight_decay` (as in the official repo) since it can be done by adding L2 regularizers to weights:\n\n```python\nimport keras\n\nregularizer = keras.regularizers.l2(WEIGHT_DECAY / 2)\nfor layer in model.layers:\n    for attr in ['kernel_regularizer', 'bias_regularizer']:\n        if hasattr(layer, attr) and layer.trainable:\n            setattr(layer, attr, regularizer)\n```",
        "description_content_type": "text/markdown",
        "docs_url": null,
        "download_url": "",
        "downloads": {
            "last_day": -1,
            "last_month": -1,
            "last_week": -1
        },
        "home_page": "https://github.com/CyberZHG/keras-adabound",
        "keywords": "",
        "license": "MIT",
        "maintainer": "",
        "maintainer_email": "",
        "name": "keras-adabound",
        "package_url": "https://pypi.org/project/keras-adabound/",
        "platform": "",
        "project_url": "https://pypi.org/project/keras-adabound/",
        "project_urls": {
            "Homepage": "https://github.com/CyberZHG/keras-adabound"
        },
        "release_url": "https://pypi.org/project/keras-adabound/0.5.0/",
        "requires_dist": null,
        "requires_python": "",
        "summary": "AdaBound optimizer in Keras",
        "version": "0.5.0"
    },
    "last_serial": 5260596,
    "releases": {
        "0.1.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "c555ba7d9478b36d8078a825e9bad8d9",
                    "sha256": "d24b89771cc9bb48aa826e2eeb4641102b7e9c57aeaf4cf5d3082096cf12411f"
                },
                "downloads": -1,
                "filename": "keras-adabound-0.1.0.tar.gz",
                "has_sig": false,
                "md5_digest": "c555ba7d9478b36d8078a825e9bad8d9",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 3602,
                "upload_time": "2019-02-28T12:33:12",
                "url": "https://files.pythonhosted.org/packages/29/b2/a135cbc4063a574f1ed3b0c2b3a76af08e706646120391a311fade863692/keras-adabound-0.1.0.tar.gz"
            }
        ],
        "0.2.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "b898f63fa537a8adf3ec4f5206224f06",
                    "sha256": "452c5f6b9884e56a1f6dd731701dbd322e446ad460b4f42fd76d9c277d9de4ec"
                },
                "downloads": -1,
                "filename": "keras-adabound-0.2.0.tar.gz",
                "has_sig": false,
                "md5_digest": "b898f63fa537a8adf3ec4f5206224f06",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 4511,
                "upload_time": "2019-02-28T13:03:12",
                "url": "https://files.pythonhosted.org/packages/47/51/d0cb538101020349a4e644724e857ebd8cc7a1c1bbaf2995778a3eef191b/keras-adabound-0.2.0.tar.gz"
            }
        ],
        "0.3.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "fff43d6c4f9494a0f9d0853e1098d905",
                    "sha256": "446e89971217ed6ba999d4a04e3681d48ef7148b2b375cdb85eded353b97d37e"
                },
                "downloads": -1,
                "filename": "keras-adabound-0.3.0.tar.gz",
                "has_sig": false,
                "md5_digest": "fff43d6c4f9494a0f9d0853e1098d905",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 4560,
                "upload_time": "2019-03-01T02:03:37",
                "url": "https://files.pythonhosted.org/packages/85/cb/48600f097952b3dd2da86b89634db1192818271bec9059e7a61957bbcb80/keras-adabound-0.3.0.tar.gz"
            }
        ],
        "0.4.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "6d7d5b17918740053c69daa84b161c00",
                    "sha256": "55c78c8495d5b52bf2bd5c3ef048520d81d14a991d9ae86d2a97e62df5e39977"
                },
                "downloads": -1,
                "filename": "keras-adabound-0.4.0.tar.gz",
                "has_sig": false,
                "md5_digest": "6d7d5b17918740053c69daa84b161c00",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 3913,
                "upload_time": "2019-03-04T16:29:31",
                "url": "https://files.pythonhosted.org/packages/2b/53/c7399e3d90a619bd932db8ff85bf9ed9fe38c17716687744fbaca2fd5e20/keras-adabound-0.4.0.tar.gz"
            }
        ],
        "0.4.1": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "5a132d367cecc10a10470d3d056a8d3a",
                    "sha256": "64f92020f221f3878db80ecf6fc8052a0da938061893de5f849cd71ee21a6ce2"
                },
                "downloads": -1,
                "filename": "keras-adabound-0.4.1.tar.gz",
                "has_sig": false,
                "md5_digest": "5a132d367cecc10a10470d3d056a8d3a",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 4376,
                "upload_time": "2019-03-12T08:28:29",
                "url": "https://files.pythonhosted.org/packages/9b/83/81b9e73aa089b0646feaacf911cd22cc4d5ea0374ef03609c35004b025e7/keras-adabound-0.4.1.tar.gz"
            }
        ],
        "0.5.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "740877d93fcf1fbfc1a4a9ae685d8e9b",
                    "sha256": "20ce742251cf0fdea3e874c401eb2b64f627a2642e2c177467e791976a8508a7"
                },
                "downloads": -1,
                "filename": "keras-adabound-0.5.0.tar.gz",
                "has_sig": false,
                "md5_digest": "740877d93fcf1fbfc1a4a9ae685d8e9b",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 4908,
                "upload_time": "2019-05-13T04:12:56",
                "url": "https://files.pythonhosted.org/packages/6f/4a/17e555ed3d8c5ed17425aeda909ed16b34dd20280b947e61dc9a03cd67dc/keras-adabound-0.5.0.tar.gz"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "740877d93fcf1fbfc1a4a9ae685d8e9b",
                "sha256": "20ce742251cf0fdea3e874c401eb2b64f627a2642e2c177467e791976a8508a7"
            },
            "downloads": -1,
            "filename": "keras-adabound-0.5.0.tar.gz",
            "has_sig": false,
            "md5_digest": "740877d93fcf1fbfc1a4a9ae685d8e9b",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": null,
            "size": 4908,
            "upload_time": "2019-05-13T04:12:56",
            "url": "https://files.pythonhosted.org/packages/6f/4a/17e555ed3d8c5ed17425aeda909ed16b34dd20280b947e61dc9a03cd67dc/keras-adabound-0.5.0.tar.gz"
        }
    ]
}