{
    "info": {
        "author": "Teodor Mihai Moldovan",
        "author_email": "moldovan@cs.berkeley.edu",
        "bugtrack_url": null,
        "classifiers": [
            "Development Status :: 2 - Pre-Alpha",
            "Intended Audience :: Developers",
            "Intended Audience :: Science/Research",
            "Topic :: Scientific/Engineering :: Artificial Intelligence"
        ],
        "description": "Description\n===========\n\ndpcluster is a package for grouping together (clustering) vectors. It automatically chooses the number of clusters that fits the data best. Specifically, it models the data as a Dirichlet Process mixture in the exponential family. For a tutorial see `\"Dirichlet Process\" by Y.W. Teh (2010) <http://www.gatsby.ucl.ac.uk/~ywteh/research/npbayes/dp.pdf>`_. Currently the only distribution implemented is the multivariate Gaussian with a Normal-Inverse-Wishart conjugate prior but extensions to other distributions are possible. \n\nTwo inference algorithms are implemented:\n\n* Variational inference as described in `\"Variational Inference for Dirichlet Process Mixtures\" by Blei et al. (2006) <http://ba.stat.cmu.edu/journal/2006/vol01/issue01/blei.pdf>`_. This is a batch algorithm that requires storing all data in memory.\n* An experimental on-line inference algorithm that requires only O(log(n)) memory where n is the total number of observations.\n\nTo install locally run::\n\n    python setup.py install --user\n\nUsage\n=====\n\nHere is a simple example to demonstrate clustering a number of random points in the plane::\n\n    >>> from dpcluster import *\n    >>> n = 10\n    >>> data = np.random.normal(size=2*n).reshape(-1,2)\n    >>> vdp = VDP(GaussianNIW(2))\n    >>> vdp.batch_learn(vdp.distr.sufficient_stats(data))\n    >>> plt.scatter(data[:,0],data[:,1])\n    >>> vdp.plot_clusters(slc=np.array([0,1]))\n    >>> plt.show()\n\nRunning this might produce 2-3 clusters depending on the randomly generated data. The adaptive nature of the Dirichlet Process mixture model becomes apparent when we increase the number of data points from ``n = 10`` to ``n = 500``. In this case the clustering algorithm will likely explain the data using only one cluster.\n\nToDo\n====\n\n* Implement more clustering algorithms e.g. based on Gibbs sampling, expectation propagation, stochastic gradient descent.\n* Implement more clustering distributions.\n* Re-implement algorithms to take advantage of multi-core or GPU computing.",
        "description_content_type": null,
        "docs_url": null,
        "download_url": "http://github.com/teodor-moldovan/dpcluster",
        "downloads": {
            "last_day": -1,
            "last_month": -1,
            "last_week": -1
        },
        "home_page": "http://dpcluster.readthedocs.org/",
        "keywords": null,
        "license": "UNKNOWN",
        "maintainer": null,
        "maintainer_email": null,
        "name": "dpcluster",
        "package_url": "https://pypi.org/project/dpcluster/",
        "platform": "UNKNOWN",
        "project_url": "https://pypi.org/project/dpcluster/",
        "project_urls": {
            "Download": "http://github.com/teodor-moldovan/dpcluster",
            "Homepage": "http://dpcluster.readthedocs.org/"
        },
        "release_url": "https://pypi.org/project/dpcluster/0.104/",
        "requires_dist": null,
        "requires_python": null,
        "summary": "dpcluster is a package for grouping together (clustering) vectors. It automatically chooses the number of clusters that fits the data best based on the underlying Dirichlet Process mixture model.",
        "version": "0.104"
    },
    "last_serial": 776479,
    "releases": {
        "0.1": [
            {
                "comment_text": "built for Linux-3.8.0-19-generic-x86_64-with-glibc2.7",
                "digests": {
                    "md5": "74a81630cc94a1573f2c6ac81e8eac43",
                    "sha256": "2f54ea478ffbe76fa39ae3e220ec7139414a00c3757bd0b306690ef1700dba68"
                },
                "downloads": -1,
                "filename": "dpcluster-0.1.linux-x86_64.tar.gz",
                "has_sig": false,
                "md5_digest": "74a81630cc94a1573f2c6ac81e8eac43",
                "packagetype": "bdist_dumb",
                "python_version": "any",
                "requires_python": null,
                "size": 21736,
                "upload_time": "2013-04-30T20:53:35",
                "url": "https://files.pythonhosted.org/packages/f1/93/6babad2fc4207fe1b87900b81ff30e361b899452931897f0035922135dee/dpcluster-0.1.linux-x86_64.tar.gz"
            },
            {
                "comment_text": "",
                "digests": {
                    "md5": "78a3e8732fc578ea09fb2a2261cb7d24",
                    "sha256": "9f5f6e6f83cc930a33fa8108e74bee5f0960ddeeb32bc2d06256bf434c179934"
                },
                "downloads": -1,
                "filename": "dpcluster-0.1.tar.gz",
                "has_sig": false,
                "md5_digest": "78a3e8732fc578ea09fb2a2261cb7d24",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 10216,
                "upload_time": "2013-04-30T20:54:55",
                "url": "https://files.pythonhosted.org/packages/f0/bb/dfb7a0dc245dfdd4a392573fbfc855a30405358d856d38214674df3979f1/dpcluster-0.1.tar.gz"
            }
        ],
        "0.101": [
            {
                "comment_text": "built for Linux-3.8.0-19-generic-x86_64-with-glibc2.7",
                "digests": {
                    "md5": "9018fd4ae8d4328292734248af1ad68f",
                    "sha256": "904363b0453da864b71b8ebe3f4a8018b19737b0d5eade68478222f13e3231df"
                },
                "downloads": -1,
                "filename": "dpcluster-0.101.linux-x86_64.tar.gz",
                "has_sig": false,
                "md5_digest": "9018fd4ae8d4328292734248af1ad68f",
                "packagetype": "bdist_dumb",
                "python_version": "any",
                "requires_python": null,
                "size": 21741,
                "upload_time": "2013-04-30T20:57:49",
                "url": "https://files.pythonhosted.org/packages/7c/c4/a7894ac8a2634585adbea7337090f38f7289218744cae852332fe12a596f/dpcluster-0.101.linux-x86_64.tar.gz"
            },
            {
                "comment_text": "",
                "digests": {
                    "md5": "171bf661cbf32c643278ad8e5b3cf740",
                    "sha256": "b2a1e99a27a8cb79c982ce5254d440af846ff7e475b63471239468bd35d31fee"
                },
                "downloads": -1,
                "filename": "dpcluster-0.101.tar.gz",
                "has_sig": false,
                "md5_digest": "171bf661cbf32c643278ad8e5b3cf740",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 10226,
                "upload_time": "2013-04-30T20:57:47",
                "url": "https://files.pythonhosted.org/packages/ac/81/020549a8453207180ce89b4f57cc69729a405d1e0b0bb15d7e331a129d75/dpcluster-0.101.tar.gz"
            }
        ],
        "0.102": [
            {
                "comment_text": "built for Linux-3.8.0-19-generic-x86_64-with-glibc2.7",
                "digests": {
                    "md5": "7fa18a8a3c6c1bf085686ea4438810a0",
                    "sha256": "18620fa0f8ec32412d16e8fb053d937a41d71a0df34c5e3a587812b7da7b9783"
                },
                "downloads": -1,
                "filename": "dpcluster-0.102.linux-x86_64.tar.gz",
                "has_sig": false,
                "md5_digest": "7fa18a8a3c6c1bf085686ea4438810a0",
                "packagetype": "bdist_dumb",
                "python_version": "any",
                "requires_python": null,
                "size": 21748,
                "upload_time": "2013-04-30T20:58:58",
                "url": "https://files.pythonhosted.org/packages/02/51/d56adc44172824d6cc1d1fce8f43856606e4e3514fe6fce37b70153fe9b7/dpcluster-0.102.linux-x86_64.tar.gz"
            },
            {
                "comment_text": "",
                "digests": {
                    "md5": "c7eda5908057404542735bf10cdd06d7",
                    "sha256": "2a24ca53bea14d055608bb817c29554f3aa0e8afbfb10a7b7baea2859e4cb8c2"
                },
                "downloads": -1,
                "filename": "dpcluster-0.102.tar.gz",
                "has_sig": false,
                "md5_digest": "c7eda5908057404542735bf10cdd06d7",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 10233,
                "upload_time": "2013-04-30T20:58:55",
                "url": "https://files.pythonhosted.org/packages/1d/dd/1ccd91e3e1344245dd39f46649b5565db6c4143899c7335b7010232a63e8/dpcluster-0.102.tar.gz"
            }
        ],
        "0.103": [
            {
                "comment_text": "built for Linux-3.8.0-19-generic-x86_64-with-glibc2.7",
                "digests": {
                    "md5": "92a2eb7f8d461769ba11b6a5052a0f33",
                    "sha256": "ae0607fc99f91a230933fcb9ce69d2348e5f9b1fc879b1b7c3c8224f7158f68a"
                },
                "downloads": -1,
                "filename": "dpcluster-0.103.linux-x86_64.tar.gz",
                "has_sig": false,
                "md5_digest": "92a2eb7f8d461769ba11b6a5052a0f33",
                "packagetype": "bdist_dumb",
                "python_version": "any",
                "requires_python": null,
                "size": 21748,
                "upload_time": "2013-05-02T18:35:11",
                "url": "https://files.pythonhosted.org/packages/8b/78/5205f7ce53ba0c8749f75809273982fb81421a506e14c1300dc63a6c620a/dpcluster-0.103.linux-x86_64.tar.gz"
            },
            {
                "comment_text": "",
                "digests": {
                    "md5": "d167dedfe7e0f30fb93ea0c58f16a66a",
                    "sha256": "9b526debe9c548a5012f2880c5095695789d8e7261bb05fea3dec0aee8c3b202"
                },
                "downloads": -1,
                "filename": "dpcluster-0.103.tar.gz",
                "has_sig": false,
                "md5_digest": "d167dedfe7e0f30fb93ea0c58f16a66a",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 10236,
                "upload_time": "2013-05-02T18:35:09",
                "url": "https://files.pythonhosted.org/packages/21/07/c3ce05cc95322fdc67193011137f123685b7a3ca47aa86d0e5b52fd0bfe5/dpcluster-0.103.tar.gz"
            }
        ],
        "0.104": [
            {
                "comment_text": "built for Linux-3.8.0-25-generic-x86_64-with-glibc2.7",
                "digests": {
                    "md5": "7f36cf0c54e2cbb72df1977628b42d7a",
                    "sha256": "6da1f8579697e35eea48330d6ad6bf2dfae76c016d02d99ae66d3decccb5720c"
                },
                "downloads": -1,
                "filename": "dpcluster-0.104.linux-x86_64.tar.gz",
                "has_sig": false,
                "md5_digest": "7f36cf0c54e2cbb72df1977628b42d7a",
                "packagetype": "bdist_dumb",
                "python_version": "any",
                "requires_python": null,
                "size": 34712,
                "upload_time": "2013-06-21T22:10:16",
                "url": "https://files.pythonhosted.org/packages/e7/e8/9b669738b8d722a98ba090b2dee2c74a0622e92068af835069ef21cb7588/dpcluster-0.104.linux-x86_64.tar.gz"
            },
            {
                "comment_text": "",
                "digests": {
                    "md5": "29e2f806300fc7c6d150cb0258a5b487",
                    "sha256": "b00d3e7b804e0ab6b80d59d3e9c895e0a3339b0144c3e9db955e65b24719cec4"
                },
                "downloads": -1,
                "filename": "dpcluster-0.104.tar.gz",
                "has_sig": false,
                "md5_digest": "29e2f806300fc7c6d150cb0258a5b487",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 14552,
                "upload_time": "2013-06-21T22:10:12",
                "url": "https://files.pythonhosted.org/packages/d6/c1/ef4c1ee0819cf4721adb2bacdfeeca7492e659f3157dfac47721d62a8cc9/dpcluster-0.104.tar.gz"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "built for Linux-3.8.0-25-generic-x86_64-with-glibc2.7",
            "digests": {
                "md5": "7f36cf0c54e2cbb72df1977628b42d7a",
                "sha256": "6da1f8579697e35eea48330d6ad6bf2dfae76c016d02d99ae66d3decccb5720c"
            },
            "downloads": -1,
            "filename": "dpcluster-0.104.linux-x86_64.tar.gz",
            "has_sig": false,
            "md5_digest": "7f36cf0c54e2cbb72df1977628b42d7a",
            "packagetype": "bdist_dumb",
            "python_version": "any",
            "requires_python": null,
            "size": 34712,
            "upload_time": "2013-06-21T22:10:16",
            "url": "https://files.pythonhosted.org/packages/e7/e8/9b669738b8d722a98ba090b2dee2c74a0622e92068af835069ef21cb7588/dpcluster-0.104.linux-x86_64.tar.gz"
        },
        {
            "comment_text": "",
            "digests": {
                "md5": "29e2f806300fc7c6d150cb0258a5b487",
                "sha256": "b00d3e7b804e0ab6b80d59d3e9c895e0a3339b0144c3e9db955e65b24719cec4"
            },
            "downloads": -1,
            "filename": "dpcluster-0.104.tar.gz",
            "has_sig": false,
            "md5_digest": "29e2f806300fc7c6d150cb0258a5b487",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": null,
            "size": 14552,
            "upload_time": "2013-06-21T22:10:12",
            "url": "https://files.pythonhosted.org/packages/d6/c1/ef4c1ee0819cf4721adb2bacdfeeca7492e659f3157dfac47721d62a8cc9/dpcluster-0.104.tar.gz"
        }
    ]
}