diff --git a/examples/homicide-dataset.csv b/examples/homicide-dataset.csv
new file mode 100644
index 00000000..b49933c1
--- /dev/null
+++ b/examples/homicide-dataset.csv
@@ -0,0 +1,1309 @@
+index,"resp","race"
+"1",0,"black"
+"2",0,"black"
+"3",0,"black"
+"4",0,"black"
+"5",0,"black"
+"6",0,"black"
+"7",0,"black"
+"8",0,"black"
+"9",0,"black"
+"10",0,"black"
+"11",0,"black"
+"12",0,"black"
+"13",0,"black"
+"14",0,"black"
+"15",0,"black"
+"16",0,"black"
+"17",0,"black"
+"18",0,"black"
+"19",0,"black"
+"20",0,"black"
+"21",0,"black"
+"22",0,"black"
+"23",0,"black"
+"24",0,"black"
+"25",0,"black"
+"26",0,"black"
+"27",0,"black"
+"28",0,"black"
+"29",0,"black"
+"30",0,"black"
+"31",0,"black"
+"32",0,"black"
+"33",0,"black"
+"34",0,"black"
+"35",0,"black"
+"36",0,"black"
+"37",0,"black"
+"38",0,"black"
+"39",0,"black"
+"40",0,"black"
+"41",0,"black"
+"42",0,"black"
+"43",0,"black"
+"44",0,"black"
+"45",0,"black"
+"46",0,"black"
+"47",0,"black"
+"48",0,"black"
+"49",0,"black"
+"50",0,"black"
+"51",0,"black"
+"52",0,"black"
+"53",0,"black"
+"54",0,"black"
+"55",0,"black"
+"56",0,"black"
+"57",0,"black"
+"58",0,"black"
+"59",0,"black"
+"60",0,"black"
+"61",0,"black"
+"62",0,"black"
+"63",0,"black"
+"64",0,"black"
+"65",0,"black"
+"66",0,"black"
+"67",0,"black"
+"68",0,"black"
+"69",0,"black"
+"70",0,"black"
+"71",0,"black"
+"72",0,"black"
+"73",0,"black"
+"74",0,"black"
+"75",0,"black"
+"76",0,"black"
+"77",0,"black"
+"78",0,"black"
+"79",0,"black"
+"80",0,"black"
+"81",0,"black"
+"82",0,"black"
+"83",0,"black"
+"84",0,"black"
+"85",0,"black"
+"86",0,"black"
+"87",0,"black"
+"88",0,"black"
+"89",0,"black"
+"90",0,"black"
+"91",0,"black"
+"92",0,"black"
+"93",0,"black"
+"94",0,"black"
+"95",0,"black"
+"96",0,"black"
+"97",0,"black"
+"98",0,"black"
+"99",0,"black"
+"100",0,"black"
+"101",0,"black"
+"102",0,"black"
+"103",0,"black"
+"104",0,"black"
+"105",0,"black"
+"106",0,"black"
+"107",0,"black"
+"108",0,"black"
+"109",0,"black"
+"110",0,"black"
+"111",0,"black"
+"112",0,"black"
+"113",0,"black"
+"114",0,"black"
+"115",0,"black"
+"116",0,"black"
+"117",0,"black"
+"118",0,"black"
+"119",0,"black"
+"120",1,"black"
+"121",1,"black"
+"122",1,"black"
+"123",1,"black"
+"124",1,"black"
+"125",1,"black"
+"126",1,"black"
+"127",1,"black"
+"128",1,"black"
+"129",1,"black"
+"130",1,"black"
+"131",1,"black"
+"132",1,"black"
+"133",1,"black"
+"134",1,"black"
+"135",1,"black"
+"136",2,"black"
+"137",2,"black"
+"138",2,"black"
+"139",2,"black"
+"140",2,"black"
+"141",2,"black"
+"142",2,"black"
+"143",2,"black"
+"144",2,"black"
+"145",2,"black"
+"146",2,"black"
+"147",2,"black"
+"148",3,"black"
+"149",3,"black"
+"150",3,"black"
+"151",3,"black"
+"152",3,"black"
+"153",3,"black"
+"154",3,"black"
+"155",4,"black"
+"156",4,"black"
+"157",4,"black"
+"158",5,"black"
+"159",5,"black"
+"160",0,"white"
+"161",0,"white"
+"162",0,"white"
+"163",0,"white"
+"164",0,"white"
+"165",0,"white"
+"166",0,"white"
+"167",0,"white"
+"168",0,"white"
+"169",0,"white"
+"170",0,"white"
+"171",0,"white"
+"172",0,"white"
+"173",0,"white"
+"174",0,"white"
+"175",0,"white"
+"176",0,"white"
+"177",0,"white"
+"178",0,"white"
+"179",0,"white"
+"180",0,"white"
+"181",0,"white"
+"182",0,"white"
+"183",0,"white"
+"184",0,"white"
+"185",0,"white"
+"186",0,"white"
+"187",0,"white"
+"188",0,"white"
+"189",0,"white"
+"190",0,"white"
+"191",0,"white"
+"192",0,"white"
+"193",0,"white"
+"194",0,"white"
+"195",0,"white"
+"196",0,"white"
+"197",0,"white"
+"198",0,"white"
+"199",0,"white"
+"200",0,"white"
+"201",0,"white"
+"202",0,"white"
+"203",0,"white"
+"204",0,"white"
+"205",0,"white"
+"206",0,"white"
+"207",0,"white"
+"208",0,"white"
+"209",0,"white"
+"210",0,"white"
+"211",0,"white"
+"212",0,"white"
+"213",0,"white"
+"214",0,"white"
+"215",0,"white"
+"216",0,"white"
+"217",0,"white"
+"218",0,"white"
+"219",0,"white"
+"220",0,"white"
+"221",0,"white"
+"222",0,"white"
+"223",0,"white"
+"224",0,"white"
+"225",0,"white"
+"226",0,"white"
+"227",0,"white"
+"228",0,"white"
+"229",0,"white"
+"230",0,"white"
+"231",0,"white"
+"232",0,"white"
+"233",0,"white"
+"234",0,"white"
+"235",0,"white"
+"236",0,"white"
+"237",0,"white"
+"238",0,"white"
+"239",0,"white"
+"240",0,"white"
+"241",0,"white"
+"242",0,"white"
+"243",0,"white"
+"244",0,"white"
+"245",0,"white"
+"246",0,"white"
+"247",0,"white"
+"248",0,"white"
+"249",0,"white"
+"250",0,"white"
+"251",0,"white"
+"252",0,"white"
+"253",0,"white"
+"254",0,"white"
+"255",0,"white"
+"256",0,"white"
+"257",0,"white"
+"258",0,"white"
+"259",0,"white"
+"260",0,"white"
+"261",0,"white"
+"262",0,"white"
+"263",0,"white"
+"264",0,"white"
+"265",0,"white"
+"266",0,"white"
+"267",0,"white"
+"268",0,"white"
+"269",0,"white"
+"270",0,"white"
+"271",0,"white"
+"272",0,"white"
+"273",0,"white"
+"274",0,"white"
+"275",0,"white"
+"276",0,"white"
+"277",0,"white"
+"278",0,"white"
+"279",0,"white"
+"280",0,"white"
+"281",0,"white"
+"282",0,"white"
+"283",0,"white"
+"284",0,"white"
+"285",0,"white"
+"286",0,"white"
+"287",0,"white"
+"288",0,"white"
+"289",0,"white"
+"290",0,"white"
+"291",0,"white"
+"292",0,"white"
+"293",0,"white"
+"294",0,"white"
+"295",0,"white"
+"296",0,"white"
+"297",0,"white"
+"298",0,"white"
+"299",0,"white"
+"300",0,"white"
+"301",0,"white"
+"302",0,"white"
+"303",0,"white"
+"304",0,"white"
+"305",0,"white"
+"306",0,"white"
+"307",0,"white"
+"308",0,"white"
+"309",0,"white"
+"310",0,"white"
+"311",0,"white"
+"312",0,"white"
+"313",0,"white"
+"314",0,"white"
+"315",0,"white"
+"316",0,"white"
+"317",0,"white"
+"318",0,"white"
+"319",0,"white"
+"320",0,"white"
+"321",0,"white"
+"322",0,"white"
+"323",0,"white"
+"324",0,"white"
+"325",0,"white"
+"326",0,"white"
+"327",0,"white"
+"328",0,"white"
+"329",0,"white"
+"330",0,"white"
+"331",0,"white"
+"332",0,"white"
+"333",0,"white"
+"334",0,"white"
+"335",0,"white"
+"336",0,"white"
+"337",0,"white"
+"338",0,"white"
+"339",0,"white"
+"340",0,"white"
+"341",0,"white"
+"342",0,"white"
+"343",0,"white"
+"344",0,"white"
+"345",0,"white"
+"346",0,"white"
+"347",0,"white"
+"348",0,"white"
+"349",0,"white"
+"350",0,"white"
+"351",0,"white"
+"352",0,"white"
+"353",0,"white"
+"354",0,"white"
+"355",0,"white"
+"356",0,"white"
+"357",0,"white"
+"358",0,"white"
+"359",0,"white"
+"360",0,"white"
+"361",0,"white"
+"362",0,"white"
+"363",0,"white"
+"364",0,"white"
+"365",0,"white"
+"366",0,"white"
+"367",0,"white"
+"368",0,"white"
+"369",0,"white"
+"370",0,"white"
+"371",0,"white"
+"372",0,"white"
+"373",0,"white"
+"374",0,"white"
+"375",0,"white"
+"376",0,"white"
+"377",0,"white"
+"378",0,"white"
+"379",0,"white"
+"380",0,"white"
+"381",0,"white"
+"382",0,"white"
+"383",0,"white"
+"384",0,"white"
+"385",0,"white"
+"386",0,"white"
+"387",0,"white"
+"388",0,"white"
+"389",0,"white"
+"390",0,"white"
+"391",0,"white"
+"392",0,"white"
+"393",0,"white"
+"394",0,"white"
+"395",0,"white"
+"396",0,"white"
+"397",0,"white"
+"398",0,"white"
+"399",0,"white"
+"400",0,"white"
+"401",0,"white"
+"402",0,"white"
+"403",0,"white"
+"404",0,"white"
+"405",0,"white"
+"406",0,"white"
+"407",0,"white"
+"408",0,"white"
+"409",0,"white"
+"410",0,"white"
+"411",0,"white"
+"412",0,"white"
+"413",0,"white"
+"414",0,"white"
+"415",0,"white"
+"416",0,"white"
+"417",0,"white"
+"418",0,"white"
+"419",0,"white"
+"420",0,"white"
+"421",0,"white"
+"422",0,"white"
+"423",0,"white"
+"424",0,"white"
+"425",0,"white"
+"426",0,"white"
+"427",0,"white"
+"428",0,"white"
+"429",0,"white"
+"430",0,"white"
+"431",0,"white"
+"432",0,"white"
+"433",0,"white"
+"434",0,"white"
+"435",0,"white"
+"436",0,"white"
+"437",0,"white"
+"438",0,"white"
+"439",0,"white"
+"440",0,"white"
+"441",0,"white"
+"442",0,"white"
+"443",0,"white"
+"444",0,"white"
+"445",0,"white"
+"446",0,"white"
+"447",0,"white"
+"448",0,"white"
+"449",0,"white"
+"450",0,"white"
+"451",0,"white"
+"452",0,"white"
+"453",0,"white"
+"454",0,"white"
+"455",0,"white"
+"456",0,"white"
+"457",0,"white"
+"458",0,"white"
+"459",0,"white"
+"460",0,"white"
+"461",0,"white"
+"462",0,"white"
+"463",0,"white"
+"464",0,"white"
+"465",0,"white"
+"466",0,"white"
+"467",0,"white"
+"468",0,"white"
+"469",0,"white"
+"470",0,"white"
+"471",0,"white"
+"472",0,"white"
+"473",0,"white"
+"474",0,"white"
+"475",0,"white"
+"476",0,"white"
+"477",0,"white"
+"478",0,"white"
+"479",0,"white"
+"480",0,"white"
+"481",0,"white"
+"482",0,"white"
+"483",0,"white"
+"484",0,"white"
+"485",0,"white"
+"486",0,"white"
+"487",0,"white"
+"488",0,"white"
+"489",0,"white"
+"490",0,"white"
+"491",0,"white"
+"492",0,"white"
+"493",0,"white"
+"494",0,"white"
+"495",0,"white"
+"496",0,"white"
+"497",0,"white"
+"498",0,"white"
+"499",0,"white"
+"500",0,"white"
+"501",0,"white"
+"502",0,"white"
+"503",0,"white"
+"504",0,"white"
+"505",0,"white"
+"506",0,"white"
+"507",0,"white"
+"508",0,"white"
+"509",0,"white"
+"510",0,"white"
+"511",0,"white"
+"512",0,"white"
+"513",0,"white"
+"514",0,"white"
+"515",0,"white"
+"516",0,"white"
+"517",0,"white"
+"518",0,"white"
+"519",0,"white"
+"520",0,"white"
+"521",0,"white"
+"522",0,"white"
+"523",0,"white"
+"524",0,"white"
+"525",0,"white"
+"526",0,"white"
+"527",0,"white"
+"528",0,"white"
+"529",0,"white"
+"530",0,"white"
+"531",0,"white"
+"532",0,"white"
+"533",0,"white"
+"534",0,"white"
+"535",0,"white"
+"536",0,"white"
+"537",0,"white"
+"538",0,"white"
+"539",0,"white"
+"540",0,"white"
+"541",0,"white"
+"542",0,"white"
+"543",0,"white"
+"544",0,"white"
+"545",0,"white"
+"546",0,"white"
+"547",0,"white"
+"548",0,"white"
+"549",0,"white"
+"550",0,"white"
+"551",0,"white"
+"552",0,"white"
+"553",0,"white"
+"554",0,"white"
+"555",0,"white"
+"556",0,"white"
+"557",0,"white"
+"558",0,"white"
+"559",0,"white"
+"560",0,"white"
+"561",0,"white"
+"562",0,"white"
+"563",0,"white"
+"564",0,"white"
+"565",0,"white"
+"566",0,"white"
+"567",0,"white"
+"568",0,"white"
+"569",0,"white"
+"570",0,"white"
+"571",0,"white"
+"572",0,"white"
+"573",0,"white"
+"574",0,"white"
+"575",0,"white"
+"576",0,"white"
+"577",0,"white"
+"578",0,"white"
+"579",0,"white"
+"580",0,"white"
+"581",0,"white"
+"582",0,"white"
+"583",0,"white"
+"584",0,"white"
+"585",0,"white"
+"586",0,"white"
+"587",0,"white"
+"588",0,"white"
+"589",0,"white"
+"590",0,"white"
+"591",0,"white"
+"592",0,"white"
+"593",0,"white"
+"594",0,"white"
+"595",0,"white"
+"596",0,"white"
+"597",0,"white"
+"598",0,"white"
+"599",0,"white"
+"600",0,"white"
+"601",0,"white"
+"602",0,"white"
+"603",0,"white"
+"604",0,"white"
+"605",0,"white"
+"606",0,"white"
+"607",0,"white"
+"608",0,"white"
+"609",0,"white"
+"610",0,"white"
+"611",0,"white"
+"612",0,"white"
+"613",0,"white"
+"614",0,"white"
+"615",0,"white"
+"616",0,"white"
+"617",0,"white"
+"618",0,"white"
+"619",0,"white"
+"620",0,"white"
+"621",0,"white"
+"622",0,"white"
+"623",0,"white"
+"624",0,"white"
+"625",0,"white"
+"626",0,"white"
+"627",0,"white"
+"628",0,"white"
+"629",0,"white"
+"630",0,"white"
+"631",0,"white"
+"632",0,"white"
+"633",0,"white"
+"634",0,"white"
+"635",0,"white"
+"636",0,"white"
+"637",0,"white"
+"638",0,"white"
+"639",0,"white"
+"640",0,"white"
+"641",0,"white"
+"642",0,"white"
+"643",0,"white"
+"644",0,"white"
+"645",0,"white"
+"646",0,"white"
+"647",0,"white"
+"648",0,"white"
+"649",0,"white"
+"650",0,"white"
+"651",0,"white"
+"652",0,"white"
+"653",0,"white"
+"654",0,"white"
+"655",0,"white"
+"656",0,"white"
+"657",0,"white"
+"658",0,"white"
+"659",0,"white"
+"660",0,"white"
+"661",0,"white"
+"662",0,"white"
+"663",0,"white"
+"664",0,"white"
+"665",0,"white"
+"666",0,"white"
+"667",0,"white"
+"668",0,"white"
+"669",0,"white"
+"670",0,"white"
+"671",0,"white"
+"672",0,"white"
+"673",0,"white"
+"674",0,"white"
+"675",0,"white"
+"676",0,"white"
+"677",0,"white"
+"678",0,"white"
+"679",0,"white"
+"680",0,"white"
+"681",0,"white"
+"682",0,"white"
+"683",0,"white"
+"684",0,"white"
+"685",0,"white"
+"686",0,"white"
+"687",0,"white"
+"688",0,"white"
+"689",0,"white"
+"690",0,"white"
+"691",0,"white"
+"692",0,"white"
+"693",0,"white"
+"694",0,"white"
+"695",0,"white"
+"696",0,"white"
+"697",0,"white"
+"698",0,"white"
+"699",0,"white"
+"700",0,"white"
+"701",0,"white"
+"702",0,"white"
+"703",0,"white"
+"704",0,"white"
+"705",0,"white"
+"706",0,"white"
+"707",0,"white"
+"708",0,"white"
+"709",0,"white"
+"710",0,"white"
+"711",0,"white"
+"712",0,"white"
+"713",0,"white"
+"714",0,"white"
+"715",0,"white"
+"716",0,"white"
+"717",0,"white"
+"718",0,"white"
+"719",0,"white"
+"720",0,"white"
+"721",0,"white"
+"722",0,"white"
+"723",0,"white"
+"724",0,"white"
+"725",0,"white"
+"726",0,"white"
+"727",0,"white"
+"728",0,"white"
+"729",0,"white"
+"730",0,"white"
+"731",0,"white"
+"732",0,"white"
+"733",0,"white"
+"734",0,"white"
+"735",0,"white"
+"736",0,"white"
+"737",0,"white"
+"738",0,"white"
+"739",0,"white"
+"740",0,"white"
+"741",0,"white"
+"742",0,"white"
+"743",0,"white"
+"744",0,"white"
+"745",0,"white"
+"746",0,"white"
+"747",0,"white"
+"748",0,"white"
+"749",0,"white"
+"750",0,"white"
+"751",0,"white"
+"752",0,"white"
+"753",0,"white"
+"754",0,"white"
+"755",0,"white"
+"756",0,"white"
+"757",0,"white"
+"758",0,"white"
+"759",0,"white"
+"760",0,"white"
+"761",0,"white"
+"762",0,"white"
+"763",0,"white"
+"764",0,"white"
+"765",0,"white"
+"766",0,"white"
+"767",0,"white"
+"768",0,"white"
+"769",0,"white"
+"770",0,"white"
+"771",0,"white"
+"772",0,"white"
+"773",0,"white"
+"774",0,"white"
+"775",0,"white"
+"776",0,"white"
+"777",0,"white"
+"778",0,"white"
+"779",0,"white"
+"780",0,"white"
+"781",0,"white"
+"782",0,"white"
+"783",0,"white"
+"784",0,"white"
+"785",0,"white"
+"786",0,"white"
+"787",0,"white"
+"788",0,"white"
+"789",0,"white"
+"790",0,"white"
+"791",0,"white"
+"792",0,"white"
+"793",0,"white"
+"794",0,"white"
+"795",0,"white"
+"796",0,"white"
+"797",0,"white"
+"798",0,"white"
+"799",0,"white"
+"800",0,"white"
+"801",0,"white"
+"802",0,"white"
+"803",0,"white"
+"804",0,"white"
+"805",0,"white"
+"806",0,"white"
+"807",0,"white"
+"808",0,"white"
+"809",0,"white"
+"810",0,"white"
+"811",0,"white"
+"812",0,"white"
+"813",0,"white"
+"814",0,"white"
+"815",0,"white"
+"816",0,"white"
+"817",0,"white"
+"818",0,"white"
+"819",0,"white"
+"820",0,"white"
+"821",0,"white"
+"822",0,"white"
+"823",0,"white"
+"824",0,"white"
+"825",0,"white"
+"826",0,"white"
+"827",0,"white"
+"828",0,"white"
+"829",0,"white"
+"830",0,"white"
+"831",0,"white"
+"832",0,"white"
+"833",0,"white"
+"834",0,"white"
+"835",0,"white"
+"836",0,"white"
+"837",0,"white"
+"838",0,"white"
+"839",0,"white"
+"840",0,"white"
+"841",0,"white"
+"842",0,"white"
+"843",0,"white"
+"844",0,"white"
+"845",0,"white"
+"846",0,"white"
+"847",0,"white"
+"848",0,"white"
+"849",0,"white"
+"850",0,"white"
+"851",0,"white"
+"852",0,"white"
+"853",0,"white"
+"854",0,"white"
+"855",0,"white"
+"856",0,"white"
+"857",0,"white"
+"858",0,"white"
+"859",0,"white"
+"860",0,"white"
+"861",0,"white"
+"862",0,"white"
+"863",0,"white"
+"864",0,"white"
+"865",0,"white"
+"866",0,"white"
+"867",0,"white"
+"868",0,"white"
+"869",0,"white"
+"870",0,"white"
+"871",0,"white"
+"872",0,"white"
+"873",0,"white"
+"874",0,"white"
+"875",0,"white"
+"876",0,"white"
+"877",0,"white"
+"878",0,"white"
+"879",0,"white"
+"880",0,"white"
+"881",0,"white"
+"882",0,"white"
+"883",0,"white"
+"884",0,"white"
+"885",0,"white"
+"886",0,"white"
+"887",0,"white"
+"888",0,"white"
+"889",0,"white"
+"890",0,"white"
+"891",0,"white"
+"892",0,"white"
+"893",0,"white"
+"894",0,"white"
+"895",0,"white"
+"896",0,"white"
+"897",0,"white"
+"898",0,"white"
+"899",0,"white"
+"900",0,"white"
+"901",0,"white"
+"902",0,"white"
+"903",0,"white"
+"904",0,"white"
+"905",0,"white"
+"906",0,"white"
+"907",0,"white"
+"908",0,"white"
+"909",0,"white"
+"910",0,"white"
+"911",0,"white"
+"912",0,"white"
+"913",0,"white"
+"914",0,"white"
+"915",0,"white"
+"916",0,"white"
+"917",0,"white"
+"918",0,"white"
+"919",0,"white"
+"920",0,"white"
+"921",0,"white"
+"922",0,"white"
+"923",0,"white"
+"924",0,"white"
+"925",0,"white"
+"926",0,"white"
+"927",0,"white"
+"928",0,"white"
+"929",0,"white"
+"930",0,"white"
+"931",0,"white"
+"932",0,"white"
+"933",0,"white"
+"934",0,"white"
+"935",0,"white"
+"936",0,"white"
+"937",0,"white"
+"938",0,"white"
+"939",0,"white"
+"940",0,"white"
+"941",0,"white"
+"942",0,"white"
+"943",0,"white"
+"944",0,"white"
+"945",0,"white"
+"946",0,"white"
+"947",0,"white"
+"948",0,"white"
+"949",0,"white"
+"950",0,"white"
+"951",0,"white"
+"952",0,"white"
+"953",0,"white"
+"954",0,"white"
+"955",0,"white"
+"956",0,"white"
+"957",0,"white"
+"958",0,"white"
+"959",0,"white"
+"960",0,"white"
+"961",0,"white"
+"962",0,"white"
+"963",0,"white"
+"964",0,"white"
+"965",0,"white"
+"966",0,"white"
+"967",0,"white"
+"968",0,"white"
+"969",0,"white"
+"970",0,"white"
+"971",0,"white"
+"972",0,"white"
+"973",0,"white"
+"974",0,"white"
+"975",0,"white"
+"976",0,"white"
+"977",0,"white"
+"978",0,"white"
+"979",0,"white"
+"980",0,"white"
+"981",0,"white"
+"982",0,"white"
+"983",0,"white"
+"984",0,"white"
+"985",0,"white"
+"986",0,"white"
+"987",0,"white"
+"988",0,"white"
+"989",0,"white"
+"990",0,"white"
+"991",0,"white"
+"992",0,"white"
+"993",0,"white"
+"994",0,"white"
+"995",0,"white"
+"996",0,"white"
+"997",0,"white"
+"998",0,"white"
+"999",0,"white"
+"1000",0,"white"
+"1001",0,"white"
+"1002",0,"white"
+"1003",0,"white"
+"1004",0,"white"
+"1005",0,"white"
+"1006",0,"white"
+"1007",0,"white"
+"1008",0,"white"
+"1009",0,"white"
+"1010",0,"white"
+"1011",0,"white"
+"1012",0,"white"
+"1013",0,"white"
+"1014",0,"white"
+"1015",0,"white"
+"1016",0,"white"
+"1017",0,"white"
+"1018",0,"white"
+"1019",0,"white"
+"1020",0,"white"
+"1021",0,"white"
+"1022",0,"white"
+"1023",0,"white"
+"1024",0,"white"
+"1025",0,"white"
+"1026",0,"white"
+"1027",0,"white"
+"1028",0,"white"
+"1029",0,"white"
+"1030",0,"white"
+"1031",0,"white"
+"1032",0,"white"
+"1033",0,"white"
+"1034",0,"white"
+"1035",0,"white"
+"1036",0,"white"
+"1037",0,"white"
+"1038",0,"white"
+"1039",0,"white"
+"1040",0,"white"
+"1041",0,"white"
+"1042",0,"white"
+"1043",0,"white"
+"1044",0,"white"
+"1045",0,"white"
+"1046",0,"white"
+"1047",0,"white"
+"1048",0,"white"
+"1049",0,"white"
+"1050",0,"white"
+"1051",0,"white"
+"1052",0,"white"
+"1053",0,"white"
+"1054",0,"white"
+"1055",0,"white"
+"1056",0,"white"
+"1057",0,"white"
+"1058",0,"white"
+"1059",0,"white"
+"1060",0,"white"
+"1061",0,"white"
+"1062",0,"white"
+"1063",0,"white"
+"1064",0,"white"
+"1065",0,"white"
+"1066",0,"white"
+"1067",0,"white"
+"1068",0,"white"
+"1069",0,"white"
+"1070",0,"white"
+"1071",0,"white"
+"1072",0,"white"
+"1073",0,"white"
+"1074",0,"white"
+"1075",0,"white"
+"1076",0,"white"
+"1077",0,"white"
+"1078",0,"white"
+"1079",0,"white"
+"1080",0,"white"
+"1081",0,"white"
+"1082",0,"white"
+"1083",0,"white"
+"1084",0,"white"
+"1085",0,"white"
+"1086",0,"white"
+"1087",0,"white"
+"1088",0,"white"
+"1089",0,"white"
+"1090",0,"white"
+"1091",0,"white"
+"1092",0,"white"
+"1093",0,"white"
+"1094",0,"white"
+"1095",0,"white"
+"1096",0,"white"
+"1097",0,"white"
+"1098",0,"white"
+"1099",0,"white"
+"1100",0,"white"
+"1101",0,"white"
+"1102",0,"white"
+"1103",0,"white"
+"1104",0,"white"
+"1105",0,"white"
+"1106",0,"white"
+"1107",0,"white"
+"1108",0,"white"
+"1109",0,"white"
+"1110",0,"white"
+"1111",0,"white"
+"1112",0,"white"
+"1113",0,"white"
+"1114",0,"white"
+"1115",0,"white"
+"1116",0,"white"
+"1117",0,"white"
+"1118",0,"white"
+"1119",0,"white"
+"1120",0,"white"
+"1121",0,"white"
+"1122",0,"white"
+"1123",0,"white"
+"1124",0,"white"
+"1125",0,"white"
+"1126",0,"white"
+"1127",0,"white"
+"1128",0,"white"
+"1129",0,"white"
+"1130",0,"white"
+"1131",0,"white"
+"1132",0,"white"
+"1133",0,"white"
+"1134",0,"white"
+"1135",0,"white"
+"1136",0,"white"
+"1137",0,"white"
+"1138",0,"white"
+"1139",0,"white"
+"1140",0,"white"
+"1141",0,"white"
+"1142",0,"white"
+"1143",0,"white"
+"1144",0,"white"
+"1145",0,"white"
+"1146",0,"white"
+"1147",0,"white"
+"1148",0,"white"
+"1149",0,"white"
+"1150",0,"white"
+"1151",0,"white"
+"1152",0,"white"
+"1153",0,"white"
+"1154",0,"white"
+"1155",0,"white"
+"1156",0,"white"
+"1157",0,"white"
+"1158",0,"white"
+"1159",0,"white"
+"1160",0,"white"
+"1161",0,"white"
+"1162",0,"white"
+"1163",0,"white"
+"1164",0,"white"
+"1165",0,"white"
+"1166",0,"white"
+"1167",0,"white"
+"1168",0,"white"
+"1169",0,"white"
+"1170",0,"white"
+"1171",0,"white"
+"1172",0,"white"
+"1173",0,"white"
+"1174",0,"white"
+"1175",0,"white"
+"1176",0,"white"
+"1177",0,"white"
+"1178",0,"white"
+"1179",0,"white"
+"1180",0,"white"
+"1181",0,"white"
+"1182",0,"white"
+"1183",0,"white"
+"1184",0,"white"
+"1185",0,"white"
+"1186",0,"white"
+"1187",0,"white"
+"1188",0,"white"
+"1189",0,"white"
+"1190",0,"white"
+"1191",0,"white"
+"1192",0,"white"
+"1193",0,"white"
+"1194",0,"white"
+"1195",0,"white"
+"1196",0,"white"
+"1197",0,"white"
+"1198",0,"white"
+"1199",0,"white"
+"1200",0,"white"
+"1201",0,"white"
+"1202",0,"white"
+"1203",0,"white"
+"1204",0,"white"
+"1205",0,"white"
+"1206",0,"white"
+"1207",0,"white"
+"1208",0,"white"
+"1209",0,"white"
+"1210",0,"white"
+"1211",0,"white"
+"1212",0,"white"
+"1213",0,"white"
+"1214",0,"white"
+"1215",0,"white"
+"1216",0,"white"
+"1217",0,"white"
+"1218",0,"white"
+"1219",0,"white"
+"1220",0,"white"
+"1221",0,"white"
+"1222",0,"white"
+"1223",0,"white"
+"1224",0,"white"
+"1225",0,"white"
+"1226",0,"white"
+"1227",0,"white"
+"1228",0,"white"
+"1229",0,"white"
+"1230",1,"white"
+"1231",1,"white"
+"1232",1,"white"
+"1233",1,"white"
+"1234",1,"white"
+"1235",1,"white"
+"1236",1,"white"
+"1237",1,"white"
+"1238",1,"white"
+"1239",1,"white"
+"1240",1,"white"
+"1241",1,"white"
+"1242",1,"white"
+"1243",1,"white"
+"1244",1,"white"
+"1245",1,"white"
+"1246",1,"white"
+"1247",1,"white"
+"1248",1,"white"
+"1249",1,"white"
+"1250",1,"white"
+"1251",1,"white"
+"1252",1,"white"
+"1253",1,"white"
+"1254",1,"white"
+"1255",1,"white"
+"1256",1,"white"
+"1257",1,"white"
+"1258",1,"white"
+"1259",1,"white"
+"1260",1,"white"
+"1261",1,"white"
+"1262",1,"white"
+"1263",1,"white"
+"1264",1,"white"
+"1265",1,"white"
+"1266",1,"white"
+"1267",1,"white"
+"1268",1,"white"
+"1269",1,"white"
+"1270",1,"white"
+"1271",1,"white"
+"1272",1,"white"
+"1273",1,"white"
+"1274",1,"white"
+"1275",1,"white"
+"1276",1,"white"
+"1277",1,"white"
+"1278",1,"white"
+"1279",1,"white"
+"1280",1,"white"
+"1281",1,"white"
+"1282",1,"white"
+"1283",1,"white"
+"1284",1,"white"
+"1285",1,"white"
+"1286",1,"white"
+"1287",1,"white"
+"1288",1,"white"
+"1289",1,"white"
+"1290",2,"white"
+"1291",2,"white"
+"1292",2,"white"
+"1293",2,"white"
+"1294",2,"white"
+"1295",2,"white"
+"1296",2,"white"
+"1297",2,"white"
+"1298",2,"white"
+"1299",2,"white"
+"1300",2,"white"
+"1301",2,"white"
+"1302",2,"white"
+"1303",2,"white"
+"1304",3,"white"
+"1305",3,"white"
+"1306",3,"white"
+"1307",3,"white"
+"1308",6,"white"
diff --git a/examples/plot_neg_binomial_homicide_victims.py b/examples/plot_neg_binomial_homicide_victims.py
new file mode 100644
index 00000000..e24ed363
--- /dev/null
+++ b/examples/plot_neg_binomial_homicide_victims.py
@@ -0,0 +1,114 @@
+# -*- coding: utf-8 -*-
+"""
+=======================================
+Number of Homicide victims
+=======================================
+
+This is an example of GLM with negative binomial distribution.
+We wrote this example taking inspiration from the R community
+below
+https://data.library.virginia.edu/getting-started-with-negative-binomial-regression-modeling/
+
+The data used are taken from a survey which asked people
+how many homicide victims they knew. The variables are "resp" and "race".
+The former indicates how many victims the respondent knew, the latter
+the ethnic group of the respondent (black or white).
+
+The nature of the empirical data suggests that we need to model
+count data (the number of homicide victims). In such scenarios,
+a common model we could use is the Poisson regression.
+
+However, if we inspect the dataset more closely, we will notice that
+the dataset is over-dispersed since the conditional mean exceeds the
+conditional variance. Basically, for each race (black or white,
+the variance is double the mean.
+
+We would need to apply another model which is the Negative Binomial regression.
+
+In this example, we will see how the Negative Binomial model will produce better
+results thanks to his dispersion parameter.
+
+"""
+
+########################################################
+# Author: Giovanni De Toni <giovanni.det@gmail.com>
+# License: MIT
+########################################################
+
+########################################################
+# Import libraries
+
+import pandas as pd
+import numpy as np
+from pyglmnet import GLM
+
+import matplotlib.pyplot as plt
+
+########################################################
+# Read and preprocess data
+df = pd.read_csv("./homicide-dataset.csv")[['resp', 'race']]
+
+########################################################
+# Histogram of type of program they are enrolled
+df.hist(column='resp', by=['race'])
+plt.show()
+
+# Print mean and standard deviation for each program enrolled.
+# We can see from here that the variance is higher that then mean for all
+# the levels, therefore hinting for over-dispersion.
+prog_mean = df.groupby('race').agg({'resp': ['mean', 'std']})
+print(prog_mean)
+
+########################################################
+# Feature
+# Model the race as a binary categorical feature
+df.race = pd.Categorical(df.race)
+df["race_code"] = df.race.cat.codes
+
+Xdsgn = df.drop(['race', 'resp'], axis=1)
+y = df['resp'].values
+
+########################################################
+# Fit the model using the Negative Binomial
+glm_nb = GLM(distr='neg-binomial',
+             alpha=0.0,
+             reg_lambda=0.0,
+             score_metric='pseudo_R2',
+             verbose=True,
+             learning_rate=1e-1,
+             max_iter=5000,
+             theta=0.20)
+glm_nb.fit(Xdsgn, y)
+print(glm_nb.beta0_, glm_nb.beta_)
+
+########################################################
+# Fit the model using the Poisson regression instead
+glm_poisson = GLM(distr='poisson',
+             alpha=0.0,
+             reg_lambda=0.0,
+             score_metric='pseudo_R2',
+             verbose=True,
+             max_iter=5000,
+             learning_rate=1e-1)
+glm_poisson.fit(Xdsgn, y)
+print(glm_poisson.beta0_, glm_poisson.beta_)
+
+########################################################
+# Plot convergence information for both negative binomial and poisson
+glm_nb.plot_convergence()
+glm_poisson.plot_convergence()
+plt.show()
+
+########################################################
+# Simulate the prediction given new data
+#
+# The Poisson model outputs the predicted mean (and therefore variance) of the distribution.
+# However, we can see from the exploratory analysis that the observed standard deviations are much larger.
+# The Negative Binomal generates the same mean, but we can use the dispersion parameter to
+# compute a more accurate estimate of the standard deviations (for both white and black classes).
+#
+pred_nb = np.array(glm_nb.predict([[0], [1]]))
+pred_poisson =  np.array(glm_poisson.predict([[0], [1]]))
+print("")
+print("NB Predicted means+std (black/white): {}, {}".format(pred_nb, np.sqrt(pred_nb+(pred_nb**2)*1/0.20)))
+print("Poisson Predicted means+std (black/white): {}, {}".format(pred_poisson, np.sqrt(pred_poisson)))
\ No newline at end of file
diff --git a/examples/plot_negative_binomial.py b/examples/plot_negative_binomial.py
new file mode 100644
index 00000000..93d95047
--- /dev/null
+++ b/examples/plot_negative_binomial.py
@@ -0,0 +1,108 @@
+# -*- coding: utf-8 -*-
+"""
+=======================================
+GLM with Negative Binomial Distribution
+=======================================
+
+This is an example of GLM with negative binomial distribution.
+We wrote this example taking inspiration from the R community
+below
+https://stats.idre.ucla.edu/r/dae/negative-binomial-regression/
+
+Here, we would like to predict the number of days absence of high school
+juniors at two schools from there type of program they are enrolled,
+and their math score.
+
+The nature of the empirical data suggests that we need to model
+count data (the number of days absent). In such scenarios, a common model
+we could use is the Poisson regression.
+
+However, if we inspect the dataset more closely, we will notice that
+the dataset is over-dispersed since the conditional mean exceeds the
+conditional variance. We would need to apply another model which is the
+Negative Binomial regression.
+
+The Negative Binomial regression can be seen as a mixture of Poisson
+regression in which the mean of the Poisson distribution can be seen
+as a random variable drawn from a Gamma distribution.
+
+This gives us an extra parameter which can be used to account for the over
+dispersion.
+
+In this example, we will apply both Negative Binomial regression and
+Poisson regression on the dataset.
+"""
+
+########################################################
+# Author: Titipat Achakulvisut <my.titipat@gmail.com>
+#         Giovanni De Toni <giovanni.det@gmail.com>
+# License: MIT
+########################################################
+
+########################################################
+# Import relevance libraries
+
+import pandas as pd
+from pyglmnet import GLM
+
+import matplotlib.pyplot as plt
+
+########################################################
+# Read and preprocess data
+df = pd.read_stata("https://stats.idre.ucla.edu/stat/stata/dae/nb_data.dta")
+
+########################################################
+# Change the program type to string (we don't need it)
+df['prog'].replace({1:"General", 2:"Academic", 3:"Vocational"}, inplace=True)
+
+########################################################
+# Histogram of type of program they are enrolled
+df.hist(column='daysabs', by=['prog'])
+plt.show()
+
+# Print mean and standard deviation for each program enrolled.
+# We can see from here that the variance is higher that then mean for all
+# the levels, therefore hinting for over-dispersion.
+prog_mean = df.groupby('prog').agg({'daysabs': ['mean', 'std']})
+print(prog_mean)
+
+########################################################
+# Feature
+X = df.drop('daysabs', axis=1)
+y = df['daysabs'].values
+
+########################################################
+# design matrix
+program_df = pd.get_dummies(df.prog)
+program_df_cleaned = program_df.drop('Vocational', axis=1)[["General", "Academic"]]
+Xdsgn = pd.concat((df['math'], program_df_cleaned), axis=1).values
+
+########################################################
+# Fit the model using the GLM
+glm_nb = GLM(distr='neg-binomial',
+             alpha=0.0,
+             reg_lambda=0.0,
+             score_metric='pseudo_R2',
+             verbose=True,
+             learning_rate=1e-6,
+             theta=1.032713156)
+glm_nb.fit(Xdsgn, y)
+print(glm_nb.beta0_, glm_nb.beta_)
+
+########################################################
+# Fit the model using the Poisson regression instead
+glm_poisson = GLM(distr='poisson',
+             alpha=0.0,
+             reg_lambda=0.0,
+             score_metric='pseudo_R2',
+             verbose=True,
+             learning_rate=1e-6)
+glm_poisson.fit(Xdsgn, y)
+print(glm_poisson.beta0_, glm_poisson.beta_)
+
+
+########################################################
+# Plot convergence information for both negative binomial and poisson
+glm_nb.plot_convergence()
+glm_poisson.plot_convergence()
+plt.show()
\ No newline at end of file