{ "info": { "author": "Samir Moustafa", "author_email": "samir.moustafa.97@gmail.com", "bugtrack_url": null, "classifiers": [ "Operating System :: OS Independent", "Programming Language :: Python :: 3", "Programming Language :: Python :: 3.5", "Programming Language :: Python :: 3.6" ], "description": "OpenAI gym Embedding world\n==========================\n\n
\n

\n
An eight-dimensional hypercube graph.
\n
\n\n[![Build Status](https://travis-ci.org/SamirMoustafa/gym-embedding-world.svg?branch=master)](https://travis-ci.org/SamirMoustafa/gym-embedding-world)\n\nTwo word embedding mapping compatible with [OpenAI gym](https://github.com/openai/gym>).\n\nRequirements:\n- Python 3.5+\n- OpenAI Gym\n- NumPy\n- Gensim\n\nInstall environment on anaconda\n-------------------------------\n\n $ conda env create -f gym-embedding-world/environment.yml\n $ source embedding-world\n $ pip install -e gym-embedding-world/.\n\nInstall environment on colab\n----------------------------\n\n !git clone \"https://github.com/SamirMoustafa/gym-embedding-world.git\"\n !pip install -e gym-embedding-world/.\n !mv gym-embedding-world gym-embedding-world-org\n !cp -r gym-embedding-world-org/embedding_world /content\n !ls embedding_world\n \nUsage\n-----\n\n $ python >>> import gym\n $ python >>> import embedding_world\n $ python >>> env = gym.make('embedding_world-v0')\n $ python >>> env.set_paths(embedding_from_file=\"... YOUR EMBEDDING PATH TO MAP FROM IT ...\",\n embedding_to_file =\"... YOUR EMBEDDING PATH TO MAP TO IT .....\")\n $ python >>> env.production_is_off()\n $ python >>> env.set_sentences('... YOUR SENTENCE TO TRANSLATE FROM IT ...', \n '... YOUR SENTENCE TO TRANSLATE TO IT .....')\n $ python >>> state, reward, done, info = env.step('dim(0)+1')\n\n``embedding_world-v0``\n----------------------\n\nEmbedding world is a simple environment based on OpenAI gym, that loads two-word embedding e.g. [Stanfrod GloVe](https://nlp.stanford.edu/projects/glove/) or [facebook fastText models](https://github.com/facebookresearch/fastText/blob/master/pretrained-vectors.md) with N-dimension and moves from one word(s) embedding-location to the other embedding using an agent actions such that actions that could be taken are `2N + 1` actions `{dimension(i)+1, dimension(i)-1}` \u222a ` {pickup}` \u2200 `i` in range from 1 to N\n\nwhich deterministically cause the corresponding state transitions\nbut actions that would take an agent of the grid leave a state unchanged.\nThe reward is negative for all transition until the goal is reached.\nThe terminal state(goal) is represented in a vector/s.\n\nThis environment has been built as part of a graduation project at [University of Alexandria, Department of Computer Science](http://sci.alexu.edu.eg/index.php/en/)\n\nPlease use this bibtex if you want to cite this repository in your publications:\n\n```\n@misc{embedding_world,\n author = {Samir Moustafa},\n title = {Embedding Environment for OpenAI Gym},\n year = {2019},\n publisher = {GitHub},\n journal = {GitHub repository},\n howpublished = {\\url{https://github.com/SamirMoustafa/gym-embedding-world}}\n}\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/SamirMoustafa/gym-embedding-world/", "keywords": "embedding_world", "license": "MIT", "maintainer": "", "maintainer_email": "", "name": "gym-embedding-world", "package_url": "https://pypi.org/project/gym-embedding-world/", "platform": "", "project_url": "https://pypi.org/project/gym-embedding-world/", "project_urls": { "Homepage": "https://github.com/SamirMoustafa/gym-embedding-world/" }, "release_url": "https://pypi.org/project/gym-embedding-world/0.0.4.0/", "requires_dist": null, "requires_python": ">=3.5,<3.7", "summary": "Two word embedding mapping compatible with OpenAI gym.", "version": "0.0.4.0" }, "last_serial": 4737471, "releases": { "0.0.3.2": [ { "comment_text": "", "digests": { "md5": "549fac14d2974ed87549d8948962e7e5", "sha256": "81f32debde3318337322425b03dd22e08f20935810bdcc0f793b2c87b25e800d" }, "downloads": -1, "filename": "gym-embedding_world-0.0.3.2.tar.gz", "has_sig": false, "md5_digest": "549fac14d2974ed87549d8948962e7e5", "packagetype": "sdist", "python_version": "source", "requires_python": ">=3.5,<3.7", "size": 7341, "upload_time": "2019-01-22T23:46:44", "url": "https://files.pythonhosted.org/packages/5e/f2/6e0eb4cef8bb34364e1ede0e58797817e2a842bbb36b3a23d5c2ef783e6d/gym-embedding_world-0.0.3.2.tar.gz" } ], "0.0.3.3": [ { "comment_text": "", "digests": { "md5": "4c4e3cdbae6f7f4e351ac39b2c6b7986", "sha256": "d966551bd98385516cc76a78cb48ae44919c26343247c3f13bb51082558c2ca0" }, "downloads": -1, "filename": "gym-embedding_world-0.0.3.3.tar.gz", "has_sig": false, "md5_digest": "4c4e3cdbae6f7f4e351ac39b2c6b7986", "packagetype": "sdist", "python_version": "source", "requires_python": ">=3.5,<3.7", "size": 8040, "upload_time": "2019-01-22T23:58:27", "url": "https://files.pythonhosted.org/packages/7a/bd/c0206f1ac9efb2bd94ae51a3c47160b3580fa819388dd536756c9a92541f/gym-embedding_world-0.0.3.3.tar.gz" } ], "0.0.3.4": [ { "comment_text": "", "digests": { "md5": "49a72b5405fb552be63cc2cbbf28110d", "sha256": "22a88a8a9f7051a706fa208252bf3a7cac1f79a09166beccb1517180c71cdf66" }, "downloads": -1, "filename": "gym-embedding_world-0.0.3.4.tar.gz", "has_sig": false, "md5_digest": "49a72b5405fb552be63cc2cbbf28110d", "packagetype": "sdist", "python_version": "source", "requires_python": ">=3.5,<3.7", "size": 8147, "upload_time": "2019-01-23T00:18:05", "url": "https://files.pythonhosted.org/packages/48/63/44f9291e66ea3b4677c605d6d63692981f001e960d73138e1b5fb66adebe/gym-embedding_world-0.0.3.4.tar.gz" } ], "0.0.4.0": [ { "comment_text": "", "digests": { "md5": "661f2ef926947d7bf2c98103999b4e90", "sha256": "34b0fd75c46482fce8175d8f02926456e7330aafb12b4e550f0b4f87bb4c695f" }, "downloads": -1, "filename": "gym-embedding_world-0.0.4.0.tar.gz", "has_sig": false, "md5_digest": "661f2ef926947d7bf2c98103999b4e90", "packagetype": "sdist", "python_version": "source", "requires_python": ">=3.5,<3.7", "size": 7864, "upload_time": "2019-01-24T21:52:17", "url": "https://files.pythonhosted.org/packages/46/98/ee2dcfcc88392b87f9a097c0203f411c5fef6a250e8a4f50f1b70ab98e55/gym-embedding_world-0.0.4.0.tar.gz" } ] }, "urls": [ { "comment_text": "", "digests": { "md5": "661f2ef926947d7bf2c98103999b4e90", "sha256": "34b0fd75c46482fce8175d8f02926456e7330aafb12b4e550f0b4f87bb4c695f" }, "downloads": -1, "filename": "gym-embedding_world-0.0.4.0.tar.gz", "has_sig": false, "md5_digest": "661f2ef926947d7bf2c98103999b4e90", "packagetype": "sdist", "python_version": "source", "requires_python": ">=3.5,<3.7", "size": 7864, "upload_time": "2019-01-24T21:52:17", "url": "https://files.pythonhosted.org/packages/46/98/ee2dcfcc88392b87f9a097c0203f411c5fef6a250e8a4f50f1b70ab98e55/gym-embedding_world-0.0.4.0.tar.gz" } ] }