{
    "info": {
        "author": "Ethan Brooks",
        "author_email": "ethanbrooks@gmail.com",
        "bugtrack_url": null,
        "classifiers": [
            "Development Status :: 3 - Alpha",
            "Intended Audience :: Developers",
            "License :: OSI Approved :: MIT License",
            "Programming Language :: Python :: 2",
            "Programming Language :: Python :: 3",
            "Topic :: Scientific/Engineering :: Artificial Intelligence"
        ],
        "description": "Tensorflow Run Manager\n======================\n\nMachine learning engineers often run multiple versions of an algorithm concurrently. However, this can make keeping track of and reproducing runs difficult. This simple utility solves this problem by maintaining a database in human-readable YAML formal that tracks\n\n* A unique name assigned to each run.\n* A description of each run.\n* The exact command used for the run.\n* The date and time of the run.\n* The most recent commit before the run.\n\nInstallation\n------------\nThe only external prerequisites of this tool are ``tmux`` and ``git``. After that, ``pip install run-manager``.\n\nImportant paths and files\n-------------------------\nWhen you run ``runs new``, the utility automatically creates the following directory structure:\n\n.. code-block:: console\n\n  <Runs Directory>/\n      <Runs Database>\n      checkpoints/\n      tensorboard/<Run Name>/\n\nRuns Database\n~~~~~~~~~~~~~\nYAML file that stores historical information about Tensorflow runs.\n\nRun Name\n~~~~~~~~\nThis is a unique value that you assign to each run. The ``runs`` section explains how the program deals with collisions.\n\n``checkpoints`` directory\n~~~~~~~~~~~~~~~~~~~~~~~~~\nDirectory where model checkpoints are saved. Used in ``tf.train.Saver().save(sess, <checkpoints directory>/<Run Name>.ckpt)``.\n\n``tensorboard`` directory\n~~~~~~~~~~~~~~~~~~~~~~~~~\nDirectory where events are saved. Used in ``tf.summary.FileWriter(<tensorboard directory>/<Run Name>/)``.\n\nConfiguration\n-------------\nRuns can be extensively configured using command-line arguments, but the following values can also be configured in a ``.runsrc`` file:\n\n===================  ===============  ======================================================================================================================================================\nname                 default          description\n===================  ===============  ======================================================================================================================================================\n``runs-dir``         ``.runs/``       The name to use for your Runs Directory.\n``db-filename``      ``.runs.yml``    The name that you choose to save your runs database with.\n``tb-dir-flag``      ``--tb-dir``     The flag that gets passed to your program that specifies ``<tensorboard directory>/<Run Name>/``. If ``None``, no flag will be passed to your program.\n``save-path-flag``   ``--save-path``  The flag that gets passed to your program that specifies ``<checkpoints directory>/<Run Name>``. If ``None``, no flag will be passed to your program.\n``column-width``     ``30``           The default column width for the ``runs table`` command.\n``virtualenv-path``  ``None``         The path to your virtual environment directory, if you're using one. Used in the following command: ``Source <virtualenv-path>/bin/activate``.\n``extra-flags``      ``[]``           Flag, value pairs for extra, custom flags. The strings ``<runs-dir>`` and ``<run-name>`` will get replaced with the appropriate value.\n===================  ===============  ======================================================================================================================================================\n\nThe program expects to find the ``.runsrc`` in a parent of the current working directory. Unless specified otherwise, the ``.runs/`` directory will be adjacent to the ``.runsrc`` in the file structure.\n\nHere is an example ``.runsrc`` file:\n\n.. code-block:: yaml\n\n    runs-dir: .lstm-runs/\n    db-filename: lstm-runs.yml\n    tb-dir-flag: None\n    save-path-flag: -s\n    column-width: 10\n    virtualenv-path: /home/ethan/virtualenvs/baselines/\n    extra-flags:\n      - [--goal-log-dir, <runs-dir>/goal-logs/<run-name>.log]\n\nAssumptions\n-----------\nThis program tries to assume as little about your program as possible, while providing useful functionality. These assumptions are as follows:\n\n* You call the ``runs`` command from a directory whose parent contains the runs directory.\n* Your program lives in a Git repository.\n* The Git working tree is not dirty (if it is, the program will throw an informative error).\n* Your program accepts a ``--tb-dir`` flag, which your program uses in ``tf.train.Saver().save(sess, <tf-dir>)``, and a ``--save-path`` flag, which your program uses in ``tf.train.Saver().restore(sess, <save-path>)``. If your flags are different and you don't feel like changing them, you can specify the new flag names using command-line arguments (``--tb-dir-flag`` and ``--save-path-flag``) or in your ``.runsrc`` (see the `Configuration`_ section for more info). If you don't want to pass either flag to your program, set ``--tb-dir-flag`` or ``--save-path-flag`` (or the associated values in your ``.runsrc``) to `None`.\n\n\nSubcommands\n-----------\nFor detailed descriptions of each subcommand and its arguments, run\n\n.. code-block:: console\n\n  runs <subcommand> -h\n\n``new``\n~~~~~~~\nStart a new run and build the file structure (see `Important paths and files`_).\n\nIt will add an entry to the database keyed by name, with the following values:\n\n* command\n* commit\n* datetime\n* description\n* host\n\nFinally, it will execute the command in ``tmux``.\n\n.. code-block:: console\n\n    runs new 'run-name' 'python main.py' --description='Description of program'\n\n*Note:* the ``--tb-dir`` and ``--save-path`` flags will be automatically\nappended to this command argument, so do not include them in the ``<command>``\nargument.\n\n``delete``\n~~~~~~~~~~\nDelete all runs matching pattern. This command also deletes associated tensorboard and checkpoint files.\n\n.. code-block:: console\n\n  \u276f runs delete \"continuous.*\"\n  Delete the following runs?\n  continuous0\n  continuous1\n  continuous21509805012\n  continuous2\n  continuous11509804959\n  continuous3\n  continuous31509805040\n\n``list``\n~~~~~~~~\nList all runs matching pattern.\n\n.. code-block:: console\n\n  \u276f runs list --pattern=\"continuous.*\"\n  continuous21509805012\n  continuous0\n  continuous11509804959\n  continuous31509805040\n  continuous1\n  continuous2\n  continuous3\n\n``table``\n~~~~~~~~~\nDisplay entries in run-database in table form.\n\n.. code-block:: console\n\n  \u276f runs table\n  name                           command                            commit                             datetime                    description                          host\n  -----------------------------  ---------------------------------  ---------------------------------  --------------------------  ---------------------------------  ------\n  continuous2                    CUDA_VISIBLE_DEVICES=1 python ...  90c0ad704e54d5152d897a4e978cc7...  2017-11-03T13:46:48.633364  Run multiple runs to test stoc...    rldl3\n  continuous3                    CUDA_VISIBLE_DEVICES=1 python ...  90c0ad704e54d5152d897a4e978cc7...  2017-11-03T13:47:09.951233  Run multiple runs to test stoc...    _\n  continuous1                    CUDA_VISIBLE_DEVICES=1 python ...  90c0ad704e54d5152d897a4e978cc7...  2017-11-03T13:42:39.879031  Run multiple runs to test stoc...    _\n  house-cnn-no-current-pos       python train.py --timesteps-pe...  9fb9b5a                            2017-10-28T18:07:44.246089  This is the refactored CNN on ...    _\n  room-with-original-cnn         python run_custom.py --timeste...  8a5e1c2                            2017-10-28T17:09:49.971061  Test original cnn on room.mjcf       _\n  continuous11509804959          CUDA_VISIBLE_DEVICES=1 python ...  90c0ad704e54d5152d897a4e978cc7...  2017-11-04T10:15:59.373633  Run multiple runs to test stoc...    _\n  continuous31509805040          CUDA_VISIBLE_DEVICES=1 python ...  90c0ad704e54d5152d897a4e978cc7...  2017-11-04T10:17:20.286275  Run multiple runs to test stoc...    rldl4\n  room-cnn-no-current-pos        python train.py --timesteps-pe...  2873fbf                            2017-10-28T18:08:10.615461  This is the refactored CNN on ...    rldl4\n  continuous21509805012          CUDA_VISIBLE_DEVICES=1 python ...  90c0ad704e54d5152d897a4e978cc7...  2017-11-04T10:16:52.129656  Run multiple runs to test stoc...    _\n\n\nTo filter by regex, use ``--pattern`` flag.\n\n``lookup``\n~~~~~~~~~~\nLookup specific value associated with database entry.\n\n.. code-block:: console\n\n  \u276f runs lookup continuous0 commit\n  da6030dd973c810c330d9635eb8d9c2105bdfe2f\n\n``reproduce``\n~~~~~~~~~~~~~\nPrint out commands for reproducing run.\n\n.. code-block:: console\n\n  \u276f runs reproduce continuous0    \n  To reproduce:\n   git checkout da6030dd973c810c330d9635eb8d9c2105bdfe2f\n   runs new continuous0 'python run_custom.py --timesteps-per-batch=2048 --continuous-actions --neg-reward --use-cnn' --description='None'\n\n\n",
        "description_content_type": null,
        "docs_url": null,
        "download_url": "",
        "downloads": {
            "last_day": -1,
            "last_month": -1,
            "last_week": -1
        },
        "home_page": "https://github.com/lobachevzky/tf-run-manager",
        "keywords": "tensorflow utilities development",
        "license": "MIT",
        "maintainer": "",
        "maintainer_email": "",
        "name": "tf-run-manager",
        "package_url": "https://pypi.org/project/tf-run-manager/",
        "platform": "",
        "project_url": "https://pypi.org/project/tf-run-manager/",
        "project_urls": {
            "Homepage": "https://github.com/lobachevzky/tf-run-manager"
        },
        "release_url": "https://pypi.org/project/tf-run-manager/2.1.6/",
        "requires_dist": [
            "paramiko (==2.3.1)",
            "termcolor (==1.1.0)",
            "PyYAML (==3.12)",
            "tabulate (==0.8.1)"
        ],
        "requires_python": "",
        "summary": "A utility for tracking and reproducing Tensorflow runs.",
        "version": "2.1.6"
    },
    "last_serial": 3524920,
    "releases": {
        "1.0.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "b4c7878ac1254a25da1cb661322b7c4c",
                    "sha256": "ad6965a375ded7246cd7811a71e2d94c22c35a27a0964c4fbdcc44c6a1ec6580"
                },
                "downloads": -1,
                "filename": "tf_run_manager-1.0.0-py2.py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "b4c7878ac1254a25da1cb661322b7c4c",
                "packagetype": "bdist_wheel",
                "python_version": "py2.py3",
                "requires_python": null,
                "size": 6306,
                "upload_time": "2017-11-07T20:58:40",
                "url": "https://files.pythonhosted.org/packages/ae/b6/9015eee4cd72e3a0700bbf4d8c38167e412c338f431bd0fc68428f9a3f2f/tf_run_manager-1.0.0-py2.py3-none-any.whl"
            },
            {
                "comment_text": "",
                "digests": {
                    "md5": "72834a561d95406115f7cd80f8721269",
                    "sha256": "3f80046ce4f01177b7ad08b59ec1a37b15aa11651233cf52fcd27ed504de8971"
                },
                "downloads": -1,
                "filename": "tf-run-manager-1.0.0.tar.gz",
                "has_sig": false,
                "md5_digest": "72834a561d95406115f7cd80f8721269",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 3668,
                "upload_time": "2017-11-07T20:58:41",
                "url": "https://files.pythonhosted.org/packages/cb/ac/243223455ad5bcdecf691e98b99abb35ba63f6fb78976bbf761dac624ec9/tf-run-manager-1.0.0.tar.gz"
            }
        ],
        "1.0.1": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "da1e4c8e33cf7fc07234ec2fc17ea03b",
                    "sha256": "1925ae190ef28ec44026f5e53b4e73a068e651226fbd67f81826c6298bd3c4aa"
                },
                "downloads": -1,
                "filename": "tf_run_manager-1.0.1-py2.py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "da1e4c8e33cf7fc07234ec2fc17ea03b",
                "packagetype": "bdist_wheel",
                "python_version": "py2.py3",
                "requires_python": null,
                "size": 6249,
                "upload_time": "2017-11-08T15:59:41",
                "url": "https://files.pythonhosted.org/packages/4b/30/292c85b1249943318b328253ec0eb7a3349309be150c20925fbfa6ece674/tf_run_manager-1.0.1-py2.py3-none-any.whl"
            },
            {
                "comment_text": "",
                "digests": {
                    "md5": "99ccb559d0f113afe94a6a8f440b953b",
                    "sha256": "f3624ec1f81b8de2ec720740a7fb7fd5c6a5f9ae5624ed04ee9599543e350dcd"
                },
                "downloads": -1,
                "filename": "tf-run-manager-1.0.1.tar.gz",
                "has_sig": false,
                "md5_digest": "99ccb559d0f113afe94a6a8f440b953b",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 3653,
                "upload_time": "2017-11-08T15:59:44",
                "url": "https://files.pythonhosted.org/packages/ef/b4/b09ad3deaa5c18672ed10b87b783e3ca5c3aed2d47096c172d179d20a100/tf-run-manager-1.0.1.tar.gz"
            }
        ],
        "1.0.2": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "a679b7452ce4d9d92b03940e0891a5fb",
                    "sha256": "7c3e6f1be58f2db978be2646b1d4fae0ac247a1cdc39117be0cd835873b1fe95"
                },
                "downloads": -1,
                "filename": "tf_run_manager-1.0.2-py2.py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "a679b7452ce4d9d92b03940e0891a5fb",
                "packagetype": "bdist_wheel",
                "python_version": "py2.py3",
                "requires_python": null,
                "size": 7819,
                "upload_time": "2017-11-08T18:42:32",
                "url": "https://files.pythonhosted.org/packages/73/3b/d39ce06f76ad5e1dfcc741d78c23fef8d2c43b0ec4d34e957785c04503f4/tf_run_manager-1.0.2-py2.py3-none-any.whl"
            },
            {
                "comment_text": "",
                "digests": {
                    "md5": "41396db041455a62408ae60eab7569f4",
                    "sha256": "3a295c97a2676b8fe90a6defdbca06f465e59c36718f647449778191fb4a93c9"
                },
                "downloads": -1,
                "filename": "tf-run-manager-1.0.2.tar.gz",
                "has_sig": false,
                "md5_digest": "41396db041455a62408ae60eab7569f4",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 4562,
                "upload_time": "2017-11-08T18:42:33",
                "url": "https://files.pythonhosted.org/packages/e0/c1/da0374bd05a462bc42e058bd70c218731624470d31be6d132fb2414ec40a/tf-run-manager-1.0.2.tar.gz"
            }
        ],
        "1.0.3": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "2cb9ad796a5fd16f6c436e251b4e1bc1",
                    "sha256": "6fecc8345df84396642ebff93ee56177ad0e549624a9d683b21a132cc7101e24"
                },
                "downloads": -1,
                "filename": "tf_run_manager-1.0.3-py2.py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "2cb9ad796a5fd16f6c436e251b4e1bc1",
                "packagetype": "bdist_wheel",
                "python_version": "py2.py3",
                "requires_python": null,
                "size": 8159,
                "upload_time": "2017-11-08T22:20:21",
                "url": "https://files.pythonhosted.org/packages/19/b3/c4891d67ac77f2939d0ac3d666860e3b66f9981cf97002d08a75991d1233/tf_run_manager-1.0.3-py2.py3-none-any.whl"
            },
            {
                "comment_text": "",
                "digests": {
                    "md5": "6861796b4401044300948e7ebd395f40",
                    "sha256": "c3a4a8fe4c8a2d7e50d5338ba05863cf671215e3b451d88deca52ef6b27a42db"
                },
                "downloads": -1,
                "filename": "tf-run-manager-1.0.3.tar.gz",
                "has_sig": false,
                "md5_digest": "6861796b4401044300948e7ebd395f40",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 4827,
                "upload_time": "2017-11-08T22:20:22",
                "url": "https://files.pythonhosted.org/packages/74/14/2559227e03161a7b91cdca9c38c438416b17c7ba71a303b3e65e992f7013/tf-run-manager-1.0.3.tar.gz"
            }
        ],
        "1.0.4": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "4201319e013315f45b2afbf6694be408",
                    "sha256": "1de6e289b0a270382388d3fef001284c7ad1d0cf60533c526484bbb5bbae5f80"
                },
                "downloads": -1,
                "filename": "tf_run_manager-1.0.4-py2.py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "4201319e013315f45b2afbf6694be408",
                "packagetype": "bdist_wheel",
                "python_version": "py2.py3",
                "requires_python": null,
                "size": 8322,
                "upload_time": "2017-11-08T22:55:59",
                "url": "https://files.pythonhosted.org/packages/92/6f/339d4a574fe25d9a01bb33eb1b4910ac90c1953bea09a5689b89d4c55836/tf_run_manager-1.0.4-py2.py3-none-any.whl"
            },
            {
                "comment_text": "",
                "digests": {
                    "md5": "2ccce61d3de753d2106637115e7d1eea",
                    "sha256": "40099a5e4fa7c142622843b6bd10f94996d7ac8603024c607b9abb6d6cfccdd6"
                },
                "downloads": -1,
                "filename": "tf-run-manager-1.0.4.tar.gz",
                "has_sig": false,
                "md5_digest": "2ccce61d3de753d2106637115e7d1eea",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 4922,
                "upload_time": "2017-11-08T22:56:00",
                "url": "https://files.pythonhosted.org/packages/de/91/86056513b0363bc9760903a3bd7ca4889529aa12f04692828f5ec6361cae/tf-run-manager-1.0.4.tar.gz"
            }
        ],
        "1.0.5": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "7d98d78e24b568841240167aada95bfa",
                    "sha256": "4e48563367ef81e53f267dd309544501c7ca8c060c56079ba46dd423a9e4e25e"
                },
                "downloads": -1,
                "filename": "tf_run_manager-1.0.5-py2.py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "7d98d78e24b568841240167aada95bfa",
                "packagetype": "bdist_wheel",
                "python_version": "py2.py3",
                "requires_python": null,
                "size": 13169,
                "upload_time": "2017-12-01T23:04:08",
                "url": "https://files.pythonhosted.org/packages/18/8b/e127c3b6a7600b9ddf4397499643fdcfcf7c35f351011a1b839087808a73/tf_run_manager-1.0.5-py2.py3-none-any.whl"
            }
        ],
        "1.0.6": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "fd039f1634e8dfff7ffc672998fcd110",
                    "sha256": "2c15157cdef249d12c938aeb2f49a6043dd8aca6951f745d5691e463ac76c613"
                },
                "downloads": -1,
                "filename": "tf_run_manager-1.0.6-py2.py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "fd039f1634e8dfff7ffc672998fcd110",
                "packagetype": "bdist_wheel",
                "python_version": "py2.py3",
                "requires_python": null,
                "size": 13156,
                "upload_time": "2017-12-01T23:11:16",
                "url": "https://files.pythonhosted.org/packages/b7/68/915c90410a1710d730c30a1e5cb2d9612d9b91b0a753bf4aea5aebc9203d/tf_run_manager-1.0.6-py2.py3-none-any.whl"
            },
            {
                "comment_text": "",
                "digests": {
                    "md5": "c86e3512bec497050a2501dbd1108360",
                    "sha256": "d58c41936a8c1796e4986432709c132fb6afbc77bdbb2220b2f699fad8e409aa"
                },
                "downloads": -1,
                "filename": "tf-run-manager-1.0.6.tar.gz",
                "has_sig": false,
                "md5_digest": "c86e3512bec497050a2501dbd1108360",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 9475,
                "upload_time": "2017-12-01T23:11:17",
                "url": "https://files.pythonhosted.org/packages/3b/66/18807b043dbb90b55f78cf1f55b8cf8e8cb0880b4866ce61575737f2b4e1/tf-run-manager-1.0.6.tar.gz"
            }
        ],
        "1.1.6": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "a1bab6c3ec6a4b45772c3ba57d5df2d6",
                    "sha256": "5c481ec6ce1ebd48655e3c515c7f6c3a432cbdf522c2650e1fb562f090c1cf95"
                },
                "downloads": -1,
                "filename": "tf_run_manager-1.1.6-py2.py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "a1bab6c3ec6a4b45772c3ba57d5df2d6",
                "packagetype": "bdist_wheel",
                "python_version": "py2.py3",
                "requires_python": null,
                "size": 13608,
                "upload_time": "2018-01-22T21:20:49",
                "url": "https://files.pythonhosted.org/packages/cf/10/ff69265cf6c8f601ef84f865059c1558d520487cff1eb201541b5665f2a3/tf_run_manager-1.1.6-py2.py3-none-any.whl"
            }
        ],
        "2.1.6": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "c959c4738bb9d8e895e8a7f5effbb8b5",
                    "sha256": "8b786a64baab5c3ec87789c5fc93f4f9b1f25d9c1f4b0571511744922bccb965"
                },
                "downloads": -1,
                "filename": "tf_run_manager-2.1.6-py2.py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "c959c4738bb9d8e895e8a7f5effbb8b5",
                "packagetype": "bdist_wheel",
                "python_version": "py2.py3",
                "requires_python": null,
                "size": 21019,
                "upload_time": "2018-01-26T18:31:36",
                "url": "https://files.pythonhosted.org/packages/b2/b8/c45aa3ada9e88130526066671fc11a8524c23f3d8a6b8ee0b9cbc23ec219/tf_run_manager-2.1.6-py2.py3-none-any.whl"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "c959c4738bb9d8e895e8a7f5effbb8b5",
                "sha256": "8b786a64baab5c3ec87789c5fc93f4f9b1f25d9c1f4b0571511744922bccb965"
            },
            "downloads": -1,
            "filename": "tf_run_manager-2.1.6-py2.py3-none-any.whl",
            "has_sig": false,
            "md5_digest": "c959c4738bb9d8e895e8a7f5effbb8b5",
            "packagetype": "bdist_wheel",
            "python_version": "py2.py3",
            "requires_python": null,
            "size": 21019,
            "upload_time": "2018-01-26T18:31:36",
            "url": "https://files.pythonhosted.org/packages/b2/b8/c45aa3ada9e88130526066671fc11a8524c23f3d8a6b8ee0b9cbc23ec219/tf_run_manager-2.1.6-py2.py3-none-any.whl"
        }
    ]
}