{
    "info": {
        "author": "Alex Griffing",
        "author_email": "argriffi@ncsu.edu",
        "bugtrack_url": null,
        "classifiers": [
            "Development Status :: 3 - Alpha",
            "Environment :: Console",
            "Intended Audience :: Developers",
            "Intended Audience :: Science/Research",
            "License :: OSI Approved :: MIT License",
            "Natural Language :: English",
            "Operating System :: POSIX :: Linux",
            "Programming Language :: C",
            "Programming Language :: Python :: 2.6",
            "Programming Language :: Unix Shell",
            "Topic :: Scientific/Engineering :: Artificial Intelligence",
            "Topic :: Scientific/Engineering :: Bio-Informatics",
            "Topic :: Scientific/Engineering :: Information Analysis",
            "Topic :: Software Development :: Libraries :: Python Modules"
        ],
        "description": "About hmmus\n===========\n\nHmmus has some\nC implementations of HMM algorithms\nwith Python bindings,\nand it is meant to be useful under the following conditions:\n\n* The sequence of observations to be analyzed is so long\n  that it does not fit conveniently in RAM.\n* Likelihoods per hidden state per position have been precalculated.\n* Numerical stability is important, but is not so important\n  that error bounds on the output are required.\n* Speed is important.\n* The number of hidden states is small.\n* The matrix of probabilities of transitions between hidden states is dense.\n* Binary data files are acceptable as input and output.\n\nThis project would be especially useless in the following cases:\n\n* User friendly or pedagogically informative software is desired.\n* All of the data can fit in RAM and numerical stability is not an issue.\n* The hidden state transitions are defined by a large sparse graph.\n* The emission distributions are uncomplicated (e.g. finite or normal).\n* A variable number of observations are emitted per hidden state.\n* Silent states other than start and stop states are used.\n\n\nRequirements\n============\n\nOperating system requirements:\n\n* This project was developed using Ubuntu,\n  so it will probably work on Debian-based Linux distributions.\n* It might work with non-Debian-based Unix variants.\n* It probably will not work on Windows.\n\nMajor dependencies:\n\n* A recent version of Python-2.x_ (2.6+).\n* A C compiler which is not too different from gcc.\n\nPython package and module dependencies:\n\n* numpy_ (version 2.0+ to support the new-style buffer interface;\n  if this has not been released yet,\n  then use a development version from the subversion repository)\n* argparse_ (included in Python-2.7+ and in Python-3.2+)\n\n\nInstallation\n============\n\nSetting up virtualenv and pip\n-----------------------------\n\nA good way to install hmmus is with virtualenv_ and pip_.\nIf you are already using these programs and you've activated\na virtual environment, then you can ignore this section.\n\nThese programs have been packaged for Ubuntu and probably Debian,\nand can be installed from the Linux distribution package repository\nas follows::\n\n    $ sudo apt-get install python-virtualenv\n    $ sudo apt-get install python-pip\n\nAlternatively the development version can be downloaded::\n\n    $ hg clone http://bitbucket.org/ianb/virtualenv\n\nTo use a binary installation of virtualenv\nto create a virtual python environment::\n\n    $ virtualenv /path/to/myenv\n\nOr to use the source installation of virtualenv\nto create a virtual python environment::\n\n    $ /go/to/virtualenv.py --distribute --python=/go/to/python /go/to/myenv\n\nNow activate the virtual environment::\n\n    $ . /path/to/myenv/bin/activate\n\nInstalling required Python modules and packages\n-----------------------------------------------\n\nThe following packages and modules should be installed:\n\n* The ``numpy`` package should be installed\n  by ``sudo apt-get install python-numpy`` on Debian and Ubuntu.\n  Or to get a newer version, install from subversion.\n* The ``argparse`` module can be installed\n  by ``pip install argparse`` in the activated virtual environment.\n\nInstalling hmmus\n----------------\n\nThe easiest way to install hmmus is from the\npython package index pypi_ as follows::\n\n    $ pip install hmmus\n\nIf pypi is inaccessible for some reason,\nthen hmmus can alternatively be installed directly from its github_\nrepository as follows::\n\n    $ pip install git+git://github.com/argriffing/hmmus\n\nIf you are developing hmmus or have cloned the git repo\nas ``~/repos/hmmus`` for some other reason,\nhmmus can be installed from this local repository as follows::\n\n    $ pip install -e ~/repos/hmmus\n\n\nUninstalling hmmus\n------------------\n\nIt is easy to uninstall hmmus using pip::\n\n    $ pip uninstall hmmus\n\nIf this fails for some reason and you really want to get rid of hmmus,\nthen you can delete the virtual environment into which hmmus\nwas installed.\n\n\nDemo\n====\n\nIn its current incarnation\nhmmus provides some scripts for doing posterior decoding,\nusing unfriendly binary files for input and output.\nThe following commands create an empty directory\nand then fill it with some sample input files::\n\n    $ mkdir mydemo\n    $ cd mydemo\n    $ hmm-demo smith\n\nThis creates the files \n``distribution.bin``,\n``transitions.bin``, and\n``likelihoods.bin``\nfrom a numerical example in the paper\nhttp://www.cs.cmu.edu/~nasmith/papers/smith.tut04a.pdf\nwhich explains posterior decoding.\nThe first two binary files define the initial distribution\nand the transition matrix of the HMM.\nThe third binary file defines the sequence of\nlikelihoods at each position conditional on each hidden state.\n\nTo get the position specific posterior distributions of hidden states,\nrun these three commands::\n\n    $ hmm-forward\n    $ hmm-backward\n    $ hmm-posterior\n\nThis should create four more binary files in the ``mydemo`` directory,\nincluding one named ``posterior.bin`` which has the distributions of interest.\nTo look at this binary file, use the octal display utility with a format\nof 8-byte floating point numbers and a width of 24 bytes per row::\n\n    $ od --format=f8 --width=24 posterior.bin\n\nUntil better documentation is written,\ninformation about the usage of the hmmus-associated scripts can be found\nusing commands like this::\n\n    $ hmm-backward --help\n\n\nUsage\n=====\n\nFor now, the only interface to the\nposterior decoding is through the binary files.\n\n\n.. _Python-2.x: http://www.python.org\n.. _argparse: http://code.google.com/p/argparse\n.. _virtualenv: http://virtualenv.openplans.org\n.. _pip: http://pip.openplans.org\n.. _pypi: http://pypi.python.org\n.. _github: http://github.com\n.. _numpy: http://numpy.scipy.org",
        "description_content_type": null,
        "docs_url": null,
        "download_url": "http://pypi.python.org/packages/source/h/hmmus/hmmus-0.3.1.tar.gz",
        "downloads": {
            "last_day": -1,
            "last_month": -1,
            "last_week": -1
        },
        "home_page": "http://github.com/argriffing/hmmus",
        "keywords": null,
        "license": "MIT",
        "maintainer": null,
        "maintainer_email": null,
        "name": "hmmus",
        "package_url": "https://pypi.org/project/hmmus/",
        "platform": "Linux",
        "project_url": "https://pypi.org/project/hmmus/",
        "project_urls": {
            "Download": "http://pypi.python.org/packages/source/h/hmmus/hmmus-0.3.1.tar.gz",
            "Homepage": "http://github.com/argriffing/hmmus"
        },
        "release_url": "https://pypi.org/project/hmmus/0.3.1/",
        "requires_dist": null,
        "requires_python": null,
        "summary": "Posterior decoding with a hidden Markov model",
        "version": "0.3.1"
    },
    "last_serial": 792955,
    "releases": {
        "0.2.10": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "33167073637c849851dccba0118077f8",
                    "sha256": "93a76ff52f9817b64dafbd750608620cd76e0e9a138f7099e6f348365ba222b4"
                },
                "downloads": -1,
                "filename": "hmmus-0.2.10.tar.gz",
                "has_sig": false,
                "md5_digest": "33167073637c849851dccba0118077f8",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 14673,
                "upload_time": "2010-03-04T06:32:15",
                "url": "https://files.pythonhosted.org/packages/f6/cf/8035c918e99f44fdee2d083556b8dabbce2e53da25d03c917ce80a2373b3/hmmus-0.2.10.tar.gz"
            }
        ],
        "0.2.14": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "575f6efa6385b8e53bd0415633602af3",
                    "sha256": "d372a72c2bc03a9837215cc0269edcba4b9a7f03be39c713c6a49eb7d0300dde"
                },
                "downloads": -1,
                "filename": "hmmus-0.2.14.tar.gz",
                "has_sig": false,
                "md5_digest": "575f6efa6385b8e53bd0415633602af3",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 17246,
                "upload_time": "2010-04-29T17:03:02",
                "url": "https://files.pythonhosted.org/packages/ec/51/51e2528d0ff23ae0c6cdb812d6ba2e512301d9f497f4c6b13f510f6cf79f/hmmus-0.2.14.tar.gz"
            }
        ],
        "0.2.16": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "fc7da06ebda206dc0a4e3bccf621bff8",
                    "sha256": "91cbd9db70380fa32869d0894386ff5354fcedd89d9c4e8ed130893644755cd4"
                },
                "downloads": -1,
                "filename": "hmmus-0.2.16.tar.gz",
                "has_sig": false,
                "md5_digest": "fc7da06ebda206dc0a4e3bccf621bff8",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 18961,
                "upload_time": "2010-05-04T22:21:17",
                "url": "https://files.pythonhosted.org/packages/7e/91/b36cdc71133129124aefa027243d479a354a87df0d51ea21cb6c5a9fecdf/hmmus-0.2.16.tar.gz"
            }
        ],
        "0.2.2": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "995d4e701d54caf5fefa627ad18555ba",
                    "sha256": "268a544d78dd3fbd6ab0a51ea8230f41cedc6606d33749c3fe39559aa6d2389d"
                },
                "downloads": -1,
                "filename": "hmmus-0.2.2.tar.gz",
                "has_sig": false,
                "md5_digest": "995d4e701d54caf5fefa627ad18555ba",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 15377,
                "upload_time": "2010-03-01T16:31:01",
                "url": "https://files.pythonhosted.org/packages/75/00/809b9c035c15d6e2d2547ee77c311aa31df5d564eafb8134e1f64cae9199/hmmus-0.2.2.tar.gz"
            }
        ],
        "0.2.6": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "80ccd1ab721662e404e95eb17f56ae87",
                    "sha256": "8e91f3e74159df48712c8ca534e19c73082307f5c99cf277ed33854a1ecd75d9"
                },
                "downloads": -1,
                "filename": "hmmus-0.2.6.tar.gz",
                "has_sig": false,
                "md5_digest": "80ccd1ab721662e404e95eb17f56ae87",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 16404,
                "upload_time": "2010-03-02T02:02:18",
                "url": "https://files.pythonhosted.org/packages/37/7c/74913df1438693ff53de985ae6c8bc16adccdf478d0f93357fd427bcfa9b/hmmus-0.2.6.tar.gz"
            }
        ],
        "0.2.7": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "24d291988545b40e84b9fce68b06c100",
                    "sha256": "5ae01605ccf76be146c723ba4e94fdf0628c8286f4eb211d9832a5ae70949383"
                },
                "downloads": -1,
                "filename": "hmmus-0.2.7.tar.gz",
                "has_sig": false,
                "md5_digest": "24d291988545b40e84b9fce68b06c100",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 16465,
                "upload_time": "2010-03-02T19:51:10",
                "url": "https://files.pythonhosted.org/packages/d0/07/cea0308a3c55a17dd919f7dcc092817aee135358ed0c4fad6659c3839699/hmmus-0.2.7.tar.gz"
            }
        ],
        "0.3.1": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "610029b3494f9491baab35dc4cd14f7b",
                    "sha256": "2998c004c53828cf5684dc5106e9e5325a298a656258a9a67600b312f5551a25"
                },
                "downloads": -1,
                "filename": "hmmus-0.3.1.tar.gz",
                "has_sig": false,
                "md5_digest": "610029b3494f9491baab35dc4cd14f7b",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 34907,
                "upload_time": "2010-07-28T16:24:39",
                "url": "https://files.pythonhosted.org/packages/1d/96/8a6870bacc4471ef6002533fe95b363f4e2d014033f455a32df9ccc9ff63/hmmus-0.3.1.tar.gz"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "610029b3494f9491baab35dc4cd14f7b",
                "sha256": "2998c004c53828cf5684dc5106e9e5325a298a656258a9a67600b312f5551a25"
            },
            "downloads": -1,
            "filename": "hmmus-0.3.1.tar.gz",
            "has_sig": false,
            "md5_digest": "610029b3494f9491baab35dc4cd14f7b",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": null,
            "size": 34907,
            "upload_time": "2010-07-28T16:24:39",
            "url": "https://files.pythonhosted.org/packages/1d/96/8a6870bacc4471ef6002533fe95b363f4e2d014033f455a32df9ccc9ff63/hmmus-0.3.1.tar.gz"
        }
    ]
}