{
    "info": {
        "author": "CyberZHG",
        "author_email": "CyberZHG@gmail.com",
        "bugtrack_url": null,
        "classifiers": [
            "Operating System :: OS Independent",
            "Programming Language :: Python :: 2.7",
            "Programming Language :: Python :: 3.6"
        ],
        "description": "# Keras Ordered Neurons LSTM\n\n[![Travis](https://travis-ci.org/CyberZHG/keras-ordered-neurons.svg)](https://travis-ci.org/CyberZHG/keras-ordered-neurons)\n[![Coverage](https://coveralls.io/repos/github/CyberZHG/keras-ordered-neurons/badge.svg?branch=master)](https://coveralls.io/github/CyberZHG/keras-ordered-neurons)\n[![Version](https://img.shields.io/pypi/v/keras-ordered-neurons.svg)](https://pypi.org/project/keras-ordered-neurons/)\n![Downloads](https://img.shields.io/pypi/dm/keras-ordered-neurons.svg)\n[![996.ICU](https://img.shields.io/badge/license-Anti%20996-blue.svg)](https://996.icu) \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-cntk-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\n\\[[\u4e2d\u6587](https://github.com/CyberZHG/keras-ordered-neurons/blob/master/README.zh-CN.md)|[English](https://github.com/CyberZHG/keras-ordered-neurons/blob/master/README.md)\\]\n\nUnofficial implementation of [ON-LSTM](https://openreview.net/pdf?id=B1l6qiR5F7).\n\n## Install\n\n```bash\npip install keras-ordered-neurons\n```\n\n## Usage\n\n### Basic\n\nSame as `LSTM` except that an extra argument `chunk_size` should be given:\n\n```python\nfrom keras.models import Sequential\nfrom keras.layers import Embedding, Bidirectional, Dense\nfrom keras_ordered_neurons import ONLSTM\n\nmodel = Sequential()\nmodel.add(Embedding(input_shape=(None,), input_dim=10, output_dim=100))\nmodel.add(Bidirectional(ONLSTM(units=50, chunk_size=5)))\nmodel.add(Dense(units=2, activation='softmax'))\nmodel.compile(optimizer='adam', loss='sparse_categorical_crossentropy')\nmodel.summary()\n```\n\n### DropConnect\n\nSet `recurrent_dropconnect` to a non-zero value to enable drop-connect for recurrent weights:\n\n```python\nfrom keras_ordered_neurons import ONLSTM\n\nONLSTM(units=50, chunk_size=5, recurrent_dropconnect=0.2)\n```\n\n### Expected Split Points\n\nSet `return_splits` to `True` if you want to know the expected split points of master forget gate and master input gate.\n\n```python\nfrom keras.models import Model\nfrom keras.layers import Input, Embedding\nfrom keras_ordered_neurons import ONLSTM\n\ninputs = Input(shape=(None,))\nembed = Embedding(input_dim=10, output_dim=100)(inputs)\noutputs, splits = ONLSTM(units=50, chunk_size=5, return_sequences=True, return_splits=True)(embed)\nmodel = Model(inputs=inputs, outputs=splits)\nmodel.compile(optimizer='adam', loss='mse')\nmodel.summary(line_length=120)\n```\n\n### `tf.keras`\n\nAdd `TF_KERAS=1` to environment variables if you are using `tensorflow.python.keras`.",
        "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-ordered-neurons",
        "keywords": "",
        "license": "Anti 996",
        "maintainer": "",
        "maintainer_email": "",
        "name": "keras-on-lstm",
        "package_url": "https://pypi.org/project/keras-on-lstm/",
        "platform": "",
        "project_url": "https://pypi.org/project/keras-on-lstm/",
        "project_urls": {
            "Homepage": "https://github.com/CyberZHG/keras-ordered-neurons"
        },
        "release_url": "https://pypi.org/project/keras-on-lstm/0.8.0/",
        "requires_dist": null,
        "requires_python": "",
        "summary": "Unofficial implementation of ON-LSTM",
        "version": "0.8.0"
    },
    "last_serial": 5336670,
    "releases": {
        "0.8.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "2aec39346fe3c89a5f1587c42215e2db",
                    "sha256": "b42eac9836765e8a96c5e3f8a939fc7552ec4f6125efb438df273e0abe61eda5"
                },
                "downloads": -1,
                "filename": "keras-on-lstm-0.8.0.tar.gz",
                "has_sig": false,
                "md5_digest": "2aec39346fe3c89a5f1587c42215e2db",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 9752,
                "upload_time": "2019-05-30T10:23:00",
                "url": "https://files.pythonhosted.org/packages/8c/26/166451b98706b778d47146c46fe51ee5be3a3982663cb4bf44adeea95204/keras-on-lstm-0.8.0.tar.gz"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "2aec39346fe3c89a5f1587c42215e2db",
                "sha256": "b42eac9836765e8a96c5e3f8a939fc7552ec4f6125efb438df273e0abe61eda5"
            },
            "downloads": -1,
            "filename": "keras-on-lstm-0.8.0.tar.gz",
            "has_sig": false,
            "md5_digest": "2aec39346fe3c89a5f1587c42215e2db",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": null,
            "size": 9752,
            "upload_time": "2019-05-30T10:23:00",
            "url": "https://files.pythonhosted.org/packages/8c/26/166451b98706b778d47146c46fe51ee5be3a3982663cb4bf44adeea95204/keras-on-lstm-0.8.0.tar.gz"
        }
    ]
}