{
    "info": {
        "author": "Lucas Boppre Niehues",
        "author_email": "lucasboppre@gmail.com",
        "bugtrack_url": null,
        "classifiers": [],
        "description": "Bayesian\n========\n\n**bayesan** is a small Python utility to reason about probabilities.\nIt uses a Bayesian system to extract features, crunch belief updates and\nspew likelihoods back. You can use either the high-level functions to\nclassify instances with supervised learning, or update beliefs manually\nwith the ``Bayes`` class.\n\nIf you want to simply classify and move files into the most fitting folder, run\nthis program from the command line passing the root folder path as parameter.\n  \nHigh Level\n----------\n\n::\n\n  from bayesian import classify, classify_file\n\n  spams = [\"buy viagra\", \"dear recipient\", \"meet sexy singles\"] # etc\n  genuines = [\"let's meet tomorrow\", \"remember to buy milk\"]\n  message = \"remember the meeting tomorrow\"\n  # Classify as \"genuine\" because of the words \"remember\" and \"tomorrow\".\n  print classify(message, {'spam': spams, 'genuine': genuines})\n\n  # Decides if the person with those measures is male or female.\n  print classify_normal({'height': 6, 'weight': 130, 'foot size': 8},\n                        {'male': [{'height': 6, 'weight': 180, 'foot size': 12},\n                                  {'height': 5.92, 'weight': 190, 'foot size': 11},\n                                  {'height': 5.58, 'weight': 170, 'foot size': 12},\n                                  {'height': 5.92, 'weight': 165, 'foot size': 10}],\n                         'female': [{'height': 5, 'weight': 100, 'foot size': 6},\n                                    {'height': 5.5, 'weight': 150, 'foot size': 8},\n                                    {'height': 5.42, 'weight': 130, 'foot size': 7},\n                                    {'height': 5.75, 'weight': 150, 'foot size': 9}]})\n\n  # Classifies \"unknown_file\" as either a Python or Java file, considering\n  # you have directories with examples of each language.\n  print classify_file(\"unknown_file\", [\"java_files\", \"python_files\"])\n\n  # Classifies every file under \"folder\" as either a Python or Java file,\n  # considering you have subdirectories with examples of each language.\n  print classify_folder(\"folder\")\n\nLow Level\n-------------\n\n::\n\n  from bayesian import Bayes\n\n  print ' -- Spam Filter --'\n  # Database with number of sightings of each words in (genuine, spam)\n  # emails.\n  words_odds = {'buy': (5, 100), 'viagra': (1, 1000), 'meeting': (15, 2)}\n  # Emails to be analyzed.\n  emails = [\n            \"let's schedule a meeting for tomorrow\", # 100% genuine (meeting)\n            \"buy some viagra\", # 100% spam (buy, viagra)\n            \"buy coffee for the meeting\", # buy x meeting, should be genuine\n           ]\n\n  for email in emails:\n      # Start with priors of 90% chance being genuine, 10% spam.\n      # Probabilities are normalized automatically.\n      b = Bayes([('genuine', 90), ('spam', 10)])\n      # Update probabilities, using the words in the emails as events and the\n      # database of chances to figure out the change.\n      b.update_from_events(email.split(), words_odds)\n      # Print the email and if it's likely spam or not.\n      print email[:15] + '...', b.most_likely()\n      \n  print ''\n\n  print ' -- Spam Filter With Email Corpus -- '\n\n  # Email corpus. A hundred spam emails to buy products and with the word\n  # \"meeting\" thrown around. Genuine emails are about meetings and buying\n  # milk.\n  instances = {'spam': [\"buy viagra\", \"buy cialis\"] * 100 + [\"meeting love\"],\n               'genuine': [\"meeting tomorrow\", \"buy milk\"] * 100}\n\n  # Use str.split to extract features/events/words from the corpus and build\n  # the model.\n  model = bayesian.extract_events_odds(instances, str.split)\n  # Create a new Bayes instance with 10%/90% priors on emails being genuine.\n  b = Bayes({'spam': .9, 'genuine': .1})\n  # Update beliefs with features/events/words from an email.\n  b.update_from_events(\"buy coffee for meeting\".split(), model)\n  # Print the email and if it's likely spam or not.\n  print \"'buy coffee for meeting'\", ':', b\n\n  print ''\n\n  print ' -- Classic Cancer Test Problem --'\n  # 1% chance of having cancer.\n  b = Bayes([('not cancer', 0.99), ('cancer', 0.01)])\n  # Test positive, 9.6% false positives and 80% true positives\n  b.update((9.6, 80))\n  print b\n  print 'Most likely:', b.most_likely()\n\n  print ''\n\n  print ' -- Are You Cheating? -- '\n  results = ['heads', 'heads', 'tails', 'heads', 'heads']\n  events_odds = {'heads': {'honest': .5, 'cheating': .9},\n                 'tails': {'honest': .5, 'cheating': .1}}\n  b = Bayes({'cheating': .5, 'honest': .5})\n  b.update_from_events(results, events_odds)\n  print b\n\n\n  def b():\n      return Bayes((0.99, 0.01), labels=['not cancer', 'cancer'])\n\n  # Random equivalent examples, all achieve the same result.\n  b() * (9.6, 80)\n  (b() * (9.6, 80)).opposite().opposite()\n  b().update({'not cancer': 9.6, 'cancer': 80})\n  b().update((9.6, 80))\n  b().update_from_events(['pos'], {'pos': (9.6, 80)})\n  b().update_from_tests([True], [(9.6, 80)])\n  Bayes([('not cancer', 0.99), ('cancer', 0.01)]) * (9.6, 80)\n  Bayes({'not cancer': 0.99, 'cancer': 0.01}) * {'not cancer': 9.6,\n                                                 'cancer': 80}",
        "description_content_type": null,
        "docs_url": null,
        "download_url": "UNKNOWN",
        "downloads": {
            "last_day": -1,
            "last_month": -1,
            "last_week": -1
        },
        "home_page": "http://pypi.python.org/pypi/bayesian/",
        "keywords": null,
        "license": "LICENSE.txt",
        "maintainer": null,
        "maintainer_email": null,
        "name": "Bayesian",
        "package_url": "https://pypi.org/project/Bayesian/",
        "platform": "UNKNOWN",
        "project_url": "https://pypi.org/project/Bayesian/",
        "project_urls": {
            "Download": "UNKNOWN",
            "Homepage": "http://pypi.python.org/pypi/bayesian/"
        },
        "release_url": "https://pypi.org/project/Bayesian/0.3.1/",
        "requires_dist": null,
        "requires_python": null,
        "summary": "Library and utility module for Bayesian reasoning",
        "version": "0.3.1"
    },
    "last_serial": 1091648,
    "releases": {
        "0.1.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "12dd64c0c18e2813becbb0b6287cc0aa",
                    "sha256": "98ddcbb733495867448b587fb5ab8ff6128573b695fead9920fb12f6f58c0131"
                },
                "downloads": -1,
                "filename": "Bayesian-0.1.0.zip",
                "has_sig": false,
                "md5_digest": "12dd64c0c18e2813becbb0b6287cc0aa",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 8434,
                "upload_time": "2013-10-24T14:55:36",
                "url": "https://files.pythonhosted.org/packages/ca/89/5b80c421b20b4571b89d02aa0feba23885dfb230b8687d1334535c6734b6/Bayesian-0.1.0.zip"
            }
        ],
        "0.2.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "baf20a432000399b006e66716eee5e3b",
                    "sha256": "818ca53b2abdc602c8614aed605a859d88d2653b23ac5fc9d27ac97a85c5c600"
                },
                "downloads": -1,
                "filename": "Bayesian-0.2.0.zip",
                "has_sig": false,
                "md5_digest": "baf20a432000399b006e66716eee5e3b",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 5640,
                "upload_time": "2014-03-17T02:47:10",
                "url": "https://files.pythonhosted.org/packages/c1/a8/cff65e26098b00a2f45649a8be717eaa3ca08ce4fcb09507df69603a5de8/Bayesian-0.2.0.zip"
            }
        ],
        "0.2.1": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "1b290034ad886efc63c99382ff3d5a70",
                    "sha256": "bd846591b43d976e9bc9d12b488a5538095959e2c7ba3e2741f0e8ea2d785374"
                },
                "downloads": -1,
                "filename": "Bayesian-0.2.1.zip",
                "has_sig": false,
                "md5_digest": "1b290034ad886efc63c99382ff3d5a70",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 8827,
                "upload_time": "2014-04-05T03:14:51",
                "url": "https://files.pythonhosted.org/packages/18/77/a06309e7f3a00d5509ec2dc68684053bb939e5d9494da30eacdad03bc460/Bayesian-0.2.1.zip"
            }
        ],
        "0.3.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "bc2bb51ba80b921b36b3ea5a51c679bc",
                    "sha256": "6f2b816c674aa554fbe6d780f26a0ba224029940c54046258c4becbdc3d0ea16"
                },
                "downloads": -1,
                "filename": "Bayesian-0.3.0.zip",
                "has_sig": false,
                "md5_digest": "bc2bb51ba80b921b36b3ea5a51c679bc",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 10806,
                "upload_time": "2014-04-28T04:42:10",
                "url": "https://files.pythonhosted.org/packages/b1/e9/4e5cbb2fbe88f99d69c64a94a136bea7d2b242af60c24267a259101c0944/Bayesian-0.3.0.zip"
            }
        ],
        "0.3.1": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "d6ffb8409789dbac5cef54679d0bf205",
                    "sha256": "19c6299733abf320c3e747df59e9d9c6c7f9ecb67d0e3453824c990a9d269c48"
                },
                "downloads": -1,
                "filename": "Bayesian-0.3.1.zip",
                "has_sig": false,
                "md5_digest": "d6ffb8409789dbac5cef54679d0bf205",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 13644,
                "upload_time": "2014-05-14T04:42:43",
                "url": "https://files.pythonhosted.org/packages/be/47/581dda9408add9090dd3642df8a9335e8e5bfb31a74158857d1c4984cff3/Bayesian-0.3.1.zip"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "d6ffb8409789dbac5cef54679d0bf205",
                "sha256": "19c6299733abf320c3e747df59e9d9c6c7f9ecb67d0e3453824c990a9d269c48"
            },
            "downloads": -1,
            "filename": "Bayesian-0.3.1.zip",
            "has_sig": false,
            "md5_digest": "d6ffb8409789dbac5cef54679d0bf205",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": null,
            "size": 13644,
            "upload_time": "2014-05-14T04:42:43",
            "url": "https://files.pythonhosted.org/packages/be/47/581dda9408add9090dd3642df8a9335e8e5bfb31a74158857d1c4984cff3/Bayesian-0.3.1.zip"
        }
    ]
}