{
    "info": {
        "author": "CyberZHG",
        "author_email": "CyberZHG@users.noreply.github.com",
        "bugtrack_url": null,
        "classifiers": [
            "License :: OSI Approved :: MIT License",
            "Operating System :: OS Independent",
            "Programming Language :: Python :: 3.6"
        ],
        "description": "# Keras RAdam\n\n[![Travis](https://travis-ci.org/CyberZHG/keras-radam.svg)](https://travis-ci.org/CyberZHG/keras-radam)\n[![Coverage](https://coveralls.io/repos/github/CyberZHG/keras-radam/badge.svg?branch=master)](https://coveralls.io/github/CyberZHG/keras-radam)\n[![Version](https://img.shields.io/pypi/v/keras-rectified-adam.svg)](https://pypi.org/project/keras-rectified-adam/)\n![Downloads](https://img.shields.io/pypi/dm/keras-rectified-adam.svg)\n![License](https://img.shields.io/pypi/l/keras-rectified-adam.svg)\n\n![](https://img.shields.io/badge/keras-tensorflow-blue.svg)\n![](https://img.shields.io/badge/keras-theano-blue.svg)\n![](https://img.shields.io/badge/keras-tf.keras-blue.svg)\n![](https://img.shields.io/badge/keras-tf.keras/eager-blue.svg)\n![](https://img.shields.io/badge/keras-tf.keras/2.0_beta-blue.svg)\n\n\\[[\u4e2d\u6587](https://github.com/CyberZHG/keras-radam/blob/master/README.zh-CN.md)|[English](https://github.com/CyberZHG/keras-radam/blob/master/README.md)\\]\n\nUnofficial implementation of [RAdam](https://arxiv.org/pdf/1908.03265v1.pdf) in Keras and TensorFlow. \n\n## Install\n\n```bash\npip install keras-rectified-adam\n```\n\n## External Link\n\n- [tensorflow/addons:RectifiedAdam](https://github.com/tensorflow/addons/blob/master/tensorflow_addons/optimizers/rectified_adam.py)\n\n## Usage\n\n```python\nimport keras\nimport numpy as np\nfrom keras_radam import RAdam\n\n# Build toy model with RAdam optimizer\nmodel = keras.models.Sequential()\nmodel.add(keras.layers.Dense(input_shape=(17,), units=3))\nmodel.compile(RAdam(), loss='mse')\n\n# Generate toy data\nx = np.random.standard_normal((4096 * 30, 17))\nw = np.random.standard_normal((17, 3))\ny = np.dot(x, w)\n\n# Fit\nmodel.fit(x, y, epochs=5)\n```\n\n### TensorFlow without Keras\n\n```python\nfrom keras_radam.training import RAdamOptimizer\n\nRAdamOptimizer(learning_rate=1e-3)\n```\n\n### Use Warmup\n\n```python\nfrom keras_radam import RAdam\n\nRAdam(total_steps=10000, warmup_proportion=0.1, min_lr=1e-5)\n```\n\n## Q & A\n\n### About Correctness\n\nThe optimizer produces similar losses and weights to the official optimizer after 500 steps.\n\n### Use `tf.keras` or `tf-2.0`\n\nAdd `TF_KERAS=1` to environment variables to use `tensorflow.python.keras`.\n\n### Use `theano` Backend\n\nAdd `KERAS_BACKEND=theano` to environment variables to enable `theano` backend.",
        "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-radam",
        "keywords": "",
        "license": "MIT",
        "maintainer": "",
        "maintainer_email": "",
        "name": "tensorflow-radam",
        "package_url": "https://pypi.org/project/tensorflow-radam/",
        "platform": "",
        "project_url": "https://pypi.org/project/tensorflow-radam/",
        "project_urls": {
            "Homepage": "https://github.com/CyberZHG/keras-radam"
        },
        "release_url": "https://pypi.org/project/tensorflow-radam/0.15.0/",
        "requires_dist": null,
        "requires_python": "",
        "summary": "RAdam implemented in Keras & TensorFlow",
        "version": "0.15.0"
    },
    "last_serial": 5873057,
    "releases": {
        "0.10.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "cca81892c83bffa7d20bb5a4c6d9ce2f",
                    "sha256": "131ed7da4ed13dab936d2ae9ed9a46c52cf083b6f1bb6beff3d096c2fa38f845"
                },
                "downloads": -1,
                "filename": "tensorflow-radam-0.10.0.tar.gz",
                "has_sig": false,
                "md5_digest": "cca81892c83bffa7d20bb5a4c6d9ce2f",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 11321,
                "upload_time": "2019-08-27T15:15:50",
                "url": "https://files.pythonhosted.org/packages/46/06/f9056f46c044b61033564c79df17ec2ddd426ac4010a134c836345be978f/tensorflow-radam-0.10.0.tar.gz"
            }
        ],
        "0.11.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "437355acaccbe5e5b7b318e56626ad9d",
                    "sha256": "f22b3014530429dc5ba3594952bba18a0a57813b30ce104460f3990e90c43f76"
                },
                "downloads": -1,
                "filename": "tensorflow-radam-0.11.0.tar.gz",
                "has_sig": false,
                "md5_digest": "437355acaccbe5e5b7b318e56626ad9d",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 11343,
                "upload_time": "2019-08-29T15:21:32",
                "url": "https://files.pythonhosted.org/packages/9d/1e/52bc97bd3791f780ad86f79d81577fa51224f4a9c13c1831072c6f0fd285/tensorflow-radam-0.11.0.tar.gz"
            }
        ],
        "0.12.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "cf2c69446d453f8e84cc20ac832d5ee8",
                    "sha256": "076d7124a6f5ea7b74383517725dda5374bb6554ad1be616614efb120174c818"
                },
                "downloads": -1,
                "filename": "tensorflow-radam-0.12.0.tar.gz",
                "has_sig": false,
                "md5_digest": "cf2c69446d453f8e84cc20ac832d5ee8",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 11326,
                "upload_time": "2019-08-30T10:55:43",
                "url": "https://files.pythonhosted.org/packages/96/b2/8fa8e2bf2b3b49fe9c48ef3a386757b5a5f22272cc5697210c3ad6415154/tensorflow-radam-0.12.0.tar.gz"
            }
        ],
        "0.14.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "2b71f6c155a009416f8fe008b8671f27",
                    "sha256": "5e7ccc936b4fec7872faed99d0269db5196673e0114ce959a9428e70e46aad56"
                },
                "downloads": -1,
                "filename": "tensorflow-radam-0.14.0.tar.gz",
                "has_sig": false,
                "md5_digest": "2b71f6c155a009416f8fe008b8671f27",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 11555,
                "upload_time": "2019-09-19T02:34:49",
                "url": "https://files.pythonhosted.org/packages/4d/8e/b4507a46726ff3fef225ba2460a39020fcf4bc493a945c3010dd0955113c/tensorflow-radam-0.14.0.tar.gz"
            }
        ],
        "0.15.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "4426e97dde704d5fd7682ec759ccfe69",
                    "sha256": "ddb0a33850d4a98826139c6371b791427aaf237cfbcde225a0117fcafbdfc362"
                },
                "downloads": -1,
                "filename": "tensorflow-radam-0.15.0.tar.gz",
                "has_sig": false,
                "md5_digest": "4426e97dde704d5fd7682ec759ccfe69",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 11739,
                "upload_time": "2019-09-23T11:32:03",
                "url": "https://files.pythonhosted.org/packages/2d/40/3bb1e39bc19ddd1b74abf4fcb447b53727f6eaf20523f63b1b73f319feca/tensorflow-radam-0.15.0.tar.gz"
            }
        ],
        "0.8.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "cc898cd23d476911d4104f89ce88cfdf",
                    "sha256": "c5fb1111c96e545278e53fcad44a30d2e3d33eecb79ce9c06f58437d65746f7c"
                },
                "downloads": -1,
                "filename": "tensorflow-radam-0.8.0.tar.gz",
                "has_sig": false,
                "md5_digest": "cc898cd23d476911d4104f89ce88cfdf",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 11026,
                "upload_time": "2019-08-21T15:44:29",
                "url": "https://files.pythonhosted.org/packages/a2/fa/e77dfc3c03140d393f595c8df77d98b9c9e1437e93eda31082fc64b2fc0c/tensorflow-radam-0.8.0.tar.gz"
            }
        ],
        "0.9.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "12a11e7baebfe13244fbe0a68d4dd97f",
                    "sha256": "744055daee0c3948e6b00139e107623d8fb7d05506c70d2d04de3af670363994"
                },
                "downloads": -1,
                "filename": "tensorflow-radam-0.9.0.tar.gz",
                "has_sig": false,
                "md5_digest": "12a11e7baebfe13244fbe0a68d4dd97f",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 11266,
                "upload_time": "2019-08-21T17:43:00",
                "url": "https://files.pythonhosted.org/packages/9a/ef/27b74a3141bcb34941f48cc78b354d6499fdbbb04d200b78c7d55b4f0833/tensorflow-radam-0.9.0.tar.gz"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "4426e97dde704d5fd7682ec759ccfe69",
                "sha256": "ddb0a33850d4a98826139c6371b791427aaf237cfbcde225a0117fcafbdfc362"
            },
            "downloads": -1,
            "filename": "tensorflow-radam-0.15.0.tar.gz",
            "has_sig": false,
            "md5_digest": "4426e97dde704d5fd7682ec759ccfe69",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": null,
            "size": 11739,
            "upload_time": "2019-09-23T11:32:03",
            "url": "https://files.pythonhosted.org/packages/2d/40/3bb1e39bc19ddd1b74abf4fcb447b53727f6eaf20523f63b1b73f319feca/tensorflow-radam-0.15.0.tar.gz"
        }
    ]
}