170_problem_census.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "PKZFwbX5AQxz"
   },
   "source": [
    "# American Census"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "rL6mvvJsAQx2"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import warnings\n",
    "\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.dummy import DummyClassifier\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "from sklearn.metrics import make_scorer\n",
    "\n",
    "from sklearn import preprocessing\n",
    "from sklearn.compose import ColumnTransformer\n",
    "from sklearn.impute import SimpleImputer\n",
    "from sklearn.preprocessing import OrdinalEncoder\n",
    "from sklearn.model_selection import StratifiedKFold\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "from sklearn.metrics import f1_score, accuracy_score\n",
    "\n",
    "from sklearn.model_selection import KFold\n",
    "from sklearn.model_selection import cross_validate"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "DGT9I97bAQyF"
   },
   "source": [
    "## Global config"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "Wf9kP0I6AQyH"
   },
   "outputs": [],
   "source": [
    "# Data config\n",
    "path_to_csv = \"adult.csv\"\n",
    "\n",
    "# No warnings\n",
    "def warn(*args, **kwargs):\n",
    "    pass\n",
    "warnings.warn = warn"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "iTifmGqGAQyR"
   },
   "source": [
    "## Load data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "v5BCUopTAQyT"
   },
   "outputs": [],
   "source": [
    "X = pd.read_csv(\"X_train.csv\")\n",
    "y = pd.read_csv(\"y_train.csv\")\n",
    "\n",
    "X_test = pd.read_csv(\"X_test.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "2eDWBDv4AQyW"
   },
   "source": [
    "## Helpers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "id": "uh0IEitoAQyX"
   },
   "outputs": [],
   "source": [
    "models = {\n",
    "    \"GaussianNB\": GaussianNB(),\n",
    "    \"DecisionTreeClassifier\": DecisionTreeClassifier(max_depth=None, min_samples_leaf=2, random_state=42),\n",
    "    \"KNeighborsClassifier\": KNeighborsClassifier(n_neighbors=3, weights=\"distance\"),  \n",
    "    \"SVM\": SVC(C=10, kernel=\"poly\",random_state=42),\n",
    "    \"LogisticRegression\": LogisticRegression(penalty=\"none\", C=10, random_state=42),\n",
    "}\n",
    "\n",
    "labelingColumns = np.array([3,4,5,6,7,8,9,10])\n",
    "def labeling(labeledColumn, colouringArray):\n",
    "  for x in range(labeledColumn.first_valid_index(), labeledColumn.first_valid_index() + len(labeledColumn)):\n",
    "    val = np.where(labeledColumn[x] == colouringArray)[0]\n",
    "    if (len(val) == 0):\n",
    "      val = -1\n",
    "    else:\n",
    "      val = val[0]\n",
    "    labeledColumn[x]=val\n",
    "\n",
    "def labelAll(X_train,labelingColumns):\n",
    "  for x in labelingColumns:\n",
    "    labelingColumn = X_train.iloc[:, x]\n",
    "    colouringArray = np.unique(getattr(X_train, X_train.columns[x]).tolist())\n",
    "    labeling(labelingColumn,colouringArray)\n",
    "    #new_labeling(labelingColumn)\n",
    "    print(labelingColumn)\n",
    "\n",
    "def imputeX(X_train):\n",
    "  global_imputer = SimpleImputer(strategy='most_frequent')\n",
    "  X_train[:] = global_imputer.fit_transform(X_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "h9NqZ_mXAQya"
   },
   "source": [
    "## Data exploration"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "LcJSU8MyAQyb"
   },
   "source": [
    "### Question 1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "XZ6AH4oAAQyh"
   },
   "source": [
    "### Question 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "id": "PHe9Z_G7GcoB"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sex_index = getattr(X, X.columns[9]).tolist()\n",
    "\n",
    "def sex_labeling(sex_index):\n",
    "    sex = np.array([' Female', ' Male'])\n",
    "\n",
    "    for x in range(0, len(sex_index)):\n",
    "        val = np.where(sex_index[x] == sex)[0]\n",
    "        if (len(val) == 0):\n",
    "            val = -1\n",
    "        else:\n",
    "            val = val[0]\n",
    "        sex_index[x] = val\n",
    "    return sex_index\n",
    "\n",
    "sex_index = np.array(sex_labeling(sex_index))\n",
    "sex_index0 = np.where(y.values[:,0] == 0)\n",
    "sex_index1 = np.where(y.values[:,0] == 1)\n",
    "\n",
    "sex_index0 = sex_index[sex_index0[0]]\n",
    "sex_index1 = sex_index[sex_index1[0]]\n",
    "\n",
    "plt.hist(sex_index0, label=\"Below 50k\")\n",
    "plt.hist(sex_index1, label=\"Above 50k\")\n",
    "plt.xlabel(X.columns[9])\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "race_index = getattr(X, X.columns[8]).tolist()\n",
    "\n",
    "def race_labeling(race_index):\n",
    "    race = np.array([' Amer-Indian-Eskimo', ' Asian-Pac-Islander', ' Black', ' Other', ' White'])\n",
    "\n",
    "    for x in range(0, len(race_index)):\n",
    "        val = np.where(race_index[x] == race)[0]\n",
    "        if (len(val) == 0):\n",
    "            val = -1\n",
    "        else:\n",
    "            val = val[0]\n",
    "        race_index[x] = val\n",
    "    return race_index\n",
    "\n",
    "race_index = np.array(race_labeling(race_index))\n",
    "race_index0 = np.where(y.values[:,0] == 0)\n",
    "race_index1 = np.where(y.values[:,0] == 1)\n",
    "\n",
    "race_index0 = race_index[race_index0[0]]\n",
    "race_index1 = race_index[race_index1[0]]\n",
    "\n",
    "plt.hist(race_index0, label=\"Below 50k\")\n",
    "plt.hist(race_index1, label=\"Above 50k\")\n",
    "plt.xlabel(X.columns[8])\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "jqyAkYZGAQym"
   },
   "source": [
    "## Data preparation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 870
    },
    "executionInfo": {
     "elapsed": 1452,
     "status": "ok",
     "timestamp": 1603871829569,
     "user": {
      "displayName": "Doruk Kırbeyi",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gg7lZvmFjsLSadU1G9V3NlTGJW8Mf6p62xQPrUlkw=s64",
      "userId": "12126647258254016890"
     },
     "user_tz": -180
    },
    "id": "aCRlguZ_O8mC",
    "outputId": "6bb8bc9b-a341-4f7d-d1a3-71cc96c58aee"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['age', 'education-num', 'hours-per-week', 'workclass', 'education',\n",
      "       'marital-status', 'occupation', 'relationship', 'race', 'sex',\n",
      "       'native-country'],\n",
      "      dtype='object')\n",
      "[17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40\n",
      " 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64\n",
      " 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 87 88 90]\n",
      "[ 1.  2.  3.  4.  5.  6.  7.  8.  9. 10. 11. 12. 13. 14. 15. 16. nan nan\n",
      " nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan\n",
      " nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan\n",
      " nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan\n",
      " nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan\n",
      " nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan\n",
      " nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan\n",
      " nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan\n",
      " nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan\n",
      " nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan\n",
      " nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan\n",
      " nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan\n",
      " nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan\n",
      " nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan\n",
      " nan nan nan nan]\n",
      "[ 1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24\n",
      " 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48\n",
      " 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 70 72 73 75\n",
      " 76 77 78 80 81 82 84 85 87 90 91 94 95 96 97 98 99]\n",
      "[' Federal-gov' ' Local-gov' ' Never-worked' ' Private' ' Self-emp-inc'\n",
      " ' Self-emp-not-inc' ' State-gov' ' Without-pay' 'nan']\n",
      "[' 10th' ' 11th' ' 12th' ' 1st-4th' ' 5th-6th' ' 7th-8th' ' 9th'\n",
      " ' Assoc-acdm' ' Assoc-voc' ' Bachelors' ' Doctorate' ' HS-grad'\n",
      " ' Masters' ' Preschool' ' Prof-school' ' Some-college']\n",
      "[' Divorced' ' Married-AF-spouse' ' Married-civ-spouse'\n",
      " ' Married-spouse-absent' ' Never-married' ' Separated' ' Widowed']\n",
      "[' Adm-clerical' ' Armed-Forces' ' Craft-repair' ' Exec-managerial'\n",
      " ' Farming-fishing' ' Handlers-cleaners' ' Machine-op-inspct'\n",
      " ' Other-service' ' Priv-house-serv' ' Prof-specialty' ' Protective-serv'\n",
      " ' Sales' ' Tech-support' ' Transport-moving' 'nan']\n",
      "[' Husband' ' Not-in-family' ' Other-relative' ' Own-child' ' Unmarried'\n",
      " ' Wife']\n",
      "[' Amer-Indian-Eskimo' ' Asian-Pac-Islander' ' Black' ' Other' ' White']\n",
      "[' Female' ' Male']\n",
      "[' Cambodia' ' Canada' ' China' ' Columbia' ' Cuba' ' Dominican-Republic'\n",
      " ' Ecuador' ' El-Salvador' ' England' ' France' ' Germany' ' Greece'\n",
      " ' Guatemala' ' Haiti' ' Honduras' ' Hong' ' Hungary' ' India' ' Iran'\n",
      " ' Ireland' ' Italy' ' Jamaica' ' Japan' ' Laos' ' Mexico' ' Nicaragua'\n",
      " ' Outlying-US(Guam-USVI-etc)' ' Peru' ' Philippines' ' Poland'\n",
      " ' Portugal' ' Puerto-Rico' ' Scotland' ' South' ' Taiwan' ' Thailand'\n",
      " ' Trinadad&Tobago' ' United-States' ' Vietnam' ' Yugoslavia' 'nan']\n"
     ]
    }
   ],
   "source": [
    "print(X.columns)\n",
    "print(np.unique(getattr(X, X.columns[0]).tolist()))\n",
    "print(np.unique(getattr(X, X.columns[1]).tolist()))\n",
    "print(np.unique(getattr(X, X.columns[2]).tolist()))\n",
    "print(np.unique(getattr(X, X.columns[3]).tolist()))\n",
    "print(np.unique(getattr(X, X.columns[4]).tolist()))\n",
    "print(np.unique(getattr(X, X.columns[5]).tolist()))\n",
    "print(np.unique(getattr(X, X.columns[6]).tolist()))\n",
    "print(np.unique(getattr(X, X.columns[7]).tolist()))\n",
    "print(np.unique(getattr(X, X.columns[8]).tolist()))\n",
    "print(np.unique(getattr(X, X.columns[9]).tolist()))\n",
    "print(np.unique(getattr(X, X.columns[10]).tolist()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ICUJgplrAQyn"
   },
   "source": [
    "### Question 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "id": "wzEcWw_nG-5w"
   },
   "outputs": [],
   "source": [
    "imputeX(X)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "gWUM1XiKAQyr"
   },
   "source": [
    "### Question 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "executionInfo": {
     "elapsed": 15145,
     "status": "ok",
     "timestamp": 1603873330083,
     "user": {
      "displayName": "Doruk Kırbeyi",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gg7lZvmFjsLSadU1G9V3NlTGJW8Mf6p62xQPrUlkw=s64",
      "userId": "12126647258254016890"
     },
     "user_tz": -180
    },
    "id": "Kr6jyvuf3pcS",
    "outputId": "e53421cc-b218-4507-8b18-11d8a818406b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0        3\n",
      "1        3\n",
      "2        3\n",
      "3        3\n",
      "4        0\n",
      "        ..\n",
      "16275    3\n",
      "16276    6\n",
      "16277    6\n",
      "16278    3\n",
      "16279    3\n",
      "Name: workclass, Length: 16280, dtype: object\n",
      "0        11\n",
      "1        15\n",
      "2         7\n",
      "3        15\n",
      "4        15\n",
      "         ..\n",
      "16275     8\n",
      "16276     9\n",
      "16277    12\n",
      "16278    11\n",
      "16279     1\n",
      "Name: education, Length: 16280, dtype: object\n",
      "0        2\n",
      "1        2\n",
      "2        4\n",
      "3        4\n",
      "4        2\n",
      "        ..\n",
      "16275    2\n",
      "16276    4\n",
      "16277    0\n",
      "16278    4\n",
      "16279    2\n",
      "Name: marital-status, Length: 16280, dtype: object\n",
      "0        11\n",
      "1         9\n",
      "2         9\n",
      "3         0\n",
      "4        13\n",
      "         ..\n",
      "16275    13\n",
      "16276     9\n",
      "16277     3\n",
      "16278     6\n",
      "16279    13\n",
      "Name: occupation, Length: 16280, dtype: object\n",
      "0        3\n",
      "1        0\n",
      "2        4\n",
      "3        1\n",
      "4        5\n",
      "        ..\n",
      "16275    0\n",
      "16276    1\n",
      "16277    4\n",
      "16278    1\n",
      "16279    0\n",
      "Name: relationship, Length: 16280, dtype: object\n",
      "0        1\n",
      "1        4\n",
      "2        2\n",
      "3        4\n",
      "4        4\n",
      "        ..\n",
      "16275    4\n",
      "16276    4\n",
      "16277    4\n",
      "16278    4\n",
      "16279    4\n",
      "Name: race, Length: 16280, dtype: object\n",
      "0        0\n",
      "1        1\n",
      "2        0\n",
      "3        0\n",
      "4        0\n",
      "        ..\n",
      "16275    1\n",
      "16276    0\n",
      "16277    0\n",
      "16278    1\n",
      "16279    1\n",
      "Name: sex, Length: 16280, dtype: object\n",
      "0        37\n",
      "1        37\n",
      "2        37\n",
      "3        37\n",
      "4        37\n",
      "         ..\n",
      "16275    37\n",
      "16276    37\n",
      "16277    37\n",
      "16278    37\n",
      "16279    37\n",
      "Name: native-country, Length: 16280, dtype: object\n"
     ]
    }
   ],
   "source": [
    "labelAll(X,labelingColumns)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "jGg8rgJjAQy5"
   },
   "source": [
    "### Question 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "id": "tiFjr6dX52Uw"
   },
   "outputs": [],
   "source": [
    "X = X.drop(X.columns[[8,9]], axis = 1)\n",
    "\n",
    "scaler = preprocessing.Normalizer()\n",
    "scalerX = scaler.fit(X)\n",
    "X = scalerX.transform(X)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Zju2MPdjAQzL"
   },
   "source": [
    "## Experiments"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "E6nBtEdpAQzM"
   },
   "source": [
    "### Question 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "id": "g4K3H5_8AQzN"
   },
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.143, random_state=42, shuffle=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "7FadAb9lAQzX"
   },
   "source": [
    "### Question 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 283
    },
    "executionInfo": {
     "elapsed": 77836,
     "status": "ok",
     "timestamp": 1603874045961,
     "user": {
      "displayName": "Doruk Kırbeyi",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gg7lZvmFjsLSadU1G9V3NlTGJW8Mf6p62xQPrUlkw=s64",
      "userId": "12126647258254016890"
     },
     "user_tz": -180
    },
    "id": "In_3nfSKd2an",
    "outputId": "4affdda2-e21d-4bed-bb8b-3f3bec5695d5"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GaussianNB\n",
      "- best_F1_score = 0.7823645403704786\n",
      "- best_accuracy_score = 0.7722974245588176\n",
      "DecisionTreeClassifier\n",
      "- best_F1_score = 0.7731488899248975\n",
      "- best_accuracy_score = 0.7789928583976236\n",
      "KNeighborsClassifier\n",
      "- best_F1_score = 0.7746232202935378\n",
      "- best_accuracy_score = 0.7757368054702534\n",
      "LogisticRegression\n",
      "- best_F1_score = 0.7747031025401246\n",
      "- best_accuracy_score = 0.7974200416582465\n",
      "SVM\n",
      "- best_F1_score = 0.7908767665565236\n",
      "- best_accuracy_score = 0.8074322062888895\n"
     ]
    }
   ],
   "source": [
    "random_state = 42\n",
    "scoring_method_f1 = make_scorer(lambda true_target, prediction: f1_score(true_target, prediction, average=\"weighted\"))\n",
    "scoring_method_accuracy = make_scorer(lambda true_target,  prediction: accuracy_score(true_target, prediction))\n",
    "\n",
    "model_parameters = {\n",
    "    \"GaussianNB\": {\n",
    "    \n",
    "    },\n",
    "    \"DecisionTreeClassifier\": {\n",
    "        'random_state': [42],\n",
    "        'max_depth': [None],\n",
    "        'min_samples_leaf':[2]\n",
    "    },\n",
    "    \"KNeighborsClassifier\": {\n",
    "        'n_neighbors': [3],\n",
    "        'weights': [\"distance\"]\n",
    "    },\n",
    "    \"LogisticRegression\": {\n",
    "        'C': [10],\n",
    "        'random_state': [42],\n",
    "        'penalty' : [\"none\"]\n",
    "\n",
    "    },\n",
    "    \"SVM\": {\n",
    "        'C': [10],\n",
    "        'random_state': [42],\n",
    "        'kernel': [\"poly\"]\n",
    "    },\n",
    "    \n",
    "}\n",
    "\n",
    "n_splits = 7\n",
    "\n",
    "fvals = []\n",
    "avals = []\n",
    "names = []\n",
    "\n",
    "for model_name, parameters in model_parameters.items():\n",
    "\n",
    "    names.append(model_name)\n",
    "    model = models[model_name]\n",
    "    \n",
    "    cv = StratifiedKFold(n_splits=n_splits, random_state=random_state, shuffle=True)\n",
    "    grid_search_F1 = GridSearchCV(model, parameters, cv=cv, verbose=False, scoring=scoring_method_f1).fit(X, np.ravel(y))\n",
    "    grid_search_accuracy = GridSearchCV(model, parameters, cv=cv, verbose=False, scoring=scoring_method_accuracy).fit(X, np.ravel(y))\n",
    "\n",
    "    best_F1_model = grid_search_F1.best_estimator_\n",
    "    best_F1_score = grid_search_F1.best_score_\n",
    "\n",
    "    best_accuracy_model = grid_search_accuracy.best_estimator_\n",
    "    best_accuracy_score = grid_search_accuracy.best_score_\n",
    "    \n",
    "    print(model_name)\n",
    "    print(\"- best_F1_score =\", best_F1_score)\n",
    "    print(\"- best_accuracy_score =\", best_accuracy_score)\n",
    "\n",
    "    fvals.append(best_F1_score)\n",
    "    avals.append(best_accuracy_score)\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "LFIS6LxvC7cP"
   },
   "source": [
    "Best Result with Default Hyper Parameters:\n",
    "\n",
    "split = 7\n",
    "\n",
    "SVM\n",
    "- best_F1_score = 0.7908767665565236\n",
    "- best_accuracy_score = 0.8074322062888895"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 387
    },
    "executionInfo": {
     "elapsed": 1125,
     "status": "ok",
     "timestamp": 1603874158153,
     "user": {
      "displayName": "Doruk Kırbeyi",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gg7lZvmFjsLSadU1G9V3NlTGJW8Mf6p62xQPrUlkw=s64",
      "userId": "12126647258254016890"
     },
     "user_tz": -180
    },
    "id": "EGviragdGfIM",
    "outputId": "e5dbfdf6-f475-4608-bd9a-e5aee6d3f869"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(14,6))\n",
    "kf = plt.subplot(1,1,1)\n",
    "\n",
    "kf.set_xlabel(\"K-Fold Classifier Type\")\n",
    "kf.set_ylabel(\"K-Fold Accuracy/F1 Score Value\")\n",
    "\n",
    "kf.plot(names, avals, label=\"Accuracy Score\", color=\"red\", marker='o')\n",
    "kf.plot(names, fvals, label=\"F1 Score\", color=\"black\", marker='o')\n",
    "\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 283
    },
    "executionInfo": {
     "elapsed": 5655,
     "status": "ok",
     "timestamp": 1603873352257,
     "user": {
      "displayName": "Doruk Kırbeyi",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gg7lZvmFjsLSadU1G9V3NlTGJW8Mf6p62xQPrUlkw=s64",
      "userId": "12126647258254016890"
     },
     "user_tz": -180
    },
    "id": "VSB9miWtAQzY",
    "outputId": "a1fdf9d5-15e6-4c0e-9e6f-fe8809bd7ffa"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GaussianNB\n",
      "   - accuracy_score 0.7891799055388579\n",
      "   - f1_score 0.7810217355184073\n",
      "DecisionTreeClassifier\n",
      "   - accuracy_score 0.798626019750966\n",
      "   - f1_score 0.8063880598843786\n",
      "KNeighborsClassifier\n",
      "   - accuracy_score 0.7900386431945041\n",
      "   - f1_score 0.7938230342897457\n",
      "SVM\n",
      "   - accuracy_score 0.8085015027908974\n",
      "   - f1_score 0.8260307052163429\n",
      "LogisticRegression\n",
      "   - accuracy_score 0.7977672820953199\n",
      "   - f1_score 0.8218590858788927\n"
     ]
    }
   ],
   "source": [
    "#***********\n",
    "#THIS METHOD IS WITHOUT KFOLD, AND THUS OVERFITS TO SHOW HIGHER RESULTS FOR THE PERFORMANCE METRICS, NO NEED TO RUN THIS ONE. IT IS ONLY HERE SO THE OVERFIT CAN BE OBSERVED AND COMPARED.\n",
    "#***********\n",
    "models = {\n",
    "    \"GaussianNB\": GaussianNB(),\n",
    "    \"DecisionTreeClassifier\": DecisionTreeClassifier(max_depth=None, min_samples_leaf=2, random_state=42),\n",
    "    \"KNeighborsClassifier\": KNeighborsClassifier(n_neighbors=3, weights=\"distance\"),  \n",
    "    \"SVM\": SVC(C=10, kernel=\"poly\",random_state=42),\n",
    "    \"LogisticRegression\": LogisticRegression(penalty=\"none\", C=10, random_state=42, max_iter = 10000),\n",
    "}\n",
    "\n",
    "for name, model in models.items():\n",
    "    model.fit(X_train,np.ravel(y_train,order='C'))\n",
    "    prediction = model.predict(X_test)\n",
    "    print(name)\n",
    "    f1_score_value = f1_score(prediction, y_test, average=\"weighted\")\n",
    "    accuracy = accuracy_score(prediction, y_test)\n",
    "    print(\"   - accuracy_score\", accuracy)\n",
    "    print(\"   - f1_score\", f1_score_value)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "5Ga6sCjHAQzg"
   },
   "source": [
    "### Question 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "id": "zrbbc8yB2m4l"
   },
   "outputs": [],
   "source": [
    "random_state = 42\n",
    "scoring_method_f1 = make_scorer(lambda true_target, prediction: f1_score(true_target, prediction, average=\"weighted\"))\n",
    "scoring_method_accuracy = make_scorer(lambda true_target,  prediction: accuracy_score(true_target, prediction))\n",
    "\n",
    "model_parameters = {\n",
    "    \"GaussianNB\": {\n",
    "    \n",
    "    },\n",
    "    \"DecisionTreeClassifier\": {\n",
    "        'random_state': [42],\n",
    "        'max_depth': [None, 2, 5, 10],\n",
    "        'min_samples_leaf':[1,2]\n",
    "    },\n",
    "    \"KNeighborsClassifier\": {\n",
    "        'n_neighbors': [1,3,5,7,9],\n",
    "        'weights': [\"uniform\", \"distance\"]\n",
    "    },\n",
    "    \"LogisticRegression\": {\n",
    "        'C': [0.3, 10],\n",
    "        'random_state': [42],\n",
    "        'max_iter': [10000],\n",
    "        'penalty': [\"l2\", \"none\"]\n",
    "    },\n",
    "    \"SVM\": {\n",
    "        'C': [0.3, 10],\n",
    "        'random_state': [42],\n",
    "        'kernel': [\"linear\", \"poly\"]\n",
    "    },\n",
    "    \n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 840
    },
    "executionInfo": {
     "elapsed": 201995,
     "status": "ok",
     "timestamp": 1603875588487,
     "user": {
      "displayName": "Doruk Kırbeyi",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gg7lZvmFjsLSadU1G9V3NlTGJW8Mf6p62xQPrUlkw=s64",
      "userId": "12126647258254016890"
     },
     "user_tz": -180
    },
    "id": "GkbwuCM524up",
    "outputId": "6ae3ce2a-6680-4178-b2a8-9636838565b0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model:  GaussianNB\n",
      " - best_F1_score = 0.7823645403704786\n",
      "    best F1 parameters:\n",
      " - best_accuracy_score = 0.7722974245588176\n",
      "   best accuracy parameters:\n",
      "Model:  DecisionTreeClassifier\n",
      " - best_F1_score = 0.8048323169950061\n",
      "    best F1 parameters:\n",
      "     - max_depth 10\n",
      "     - min_samples_leaf 2\n",
      "     - random_state 42\n",
      " - best_accuracy_score = 0.8089065965054093\n",
      "   best accuracy parameters:\n",
      "     - max_depth 10\n",
      "     - min_samples_leaf 2\n",
      "     - random_state 42\n",
      "Model:  KNeighborsClassifier\n",
      " - best_F1_score = 0.7918432286067414\n",
      "    best F1 parameters:\n",
      "     - n_neighbors 9\n",
      "     - weights uniform\n",
      " - best_accuracy_score = 0.7957615045574438\n",
      "   best accuracy parameters:\n",
      "     - n_neighbors 9\n",
      "     - weights uniform\n",
      "Model:  LogisticRegression\n",
      " - best_F1_score = 0.7747023780356728\n",
      "    best F1 parameters:\n",
      "     - C 0.3\n",
      "     - max_iter 10000\n",
      "     - penalty none\n",
      "     - random_state 42\n",
      " - best_accuracy_score = 0.7974200416582465\n",
      "   best accuracy parameters:\n",
      "     - C 0.3\n",
      "     - max_iter 10000\n",
      "     - penalty none\n",
      "     - random_state 42\n",
      "Model:  SVM\n",
      " - best_F1_score = 0.7908767665565236\n",
      "    best F1 parameters:\n",
      "     - C 10\n",
      "     - kernel poly\n",
      "     - random_state 42\n",
      " - best_accuracy_score = 0.8074322062888895\n",
      "   best accuracy parameters:\n",
      "     - C 10\n",
      "     - kernel poly\n",
      "     - random_state 42\n"
     ]
    }
   ],
   "source": [
    "n_splits = 7\n",
    "\n",
    "fvals_tuned = []\n",
    "avals_tuned = []\n",
    "names_tuned = []\n",
    "\n",
    "for model_name, parameters in model_parameters.items():\n",
    "\n",
    "    names_tuned.append(model_name)    \n",
    "    model = models[model_name]\n",
    "    \n",
    "    cv = StratifiedKFold(n_splits=n_splits, random_state=random_state, shuffle=True)\n",
    "    grid_search_F1 = GridSearchCV(model, parameters, cv=cv, verbose=False, scoring=scoring_method_f1).fit(X, np.ravel(y))\n",
    "    grid_search_accuracy = GridSearchCV(model, parameters, cv=cv, verbose=False, scoring=scoring_method_accuracy).fit(X, np.ravel(y))\n",
    "\n",
    "    best_F1_model = grid_search_F1.best_estimator_\n",
    "    best_F1_score = grid_search_F1.best_score_\n",
    "    best_F1_params = grid_search_F1.best_params_\n",
    "\n",
    "    best_accuracy_model = grid_search_accuracy.best_estimator_\n",
    "    best_accuracy_score = grid_search_accuracy.best_score_\n",
    "    best_accuracy_params = grid_search_accuracy.best_params_\n",
    "    \n",
    "    print(\"Model: \",model_name)\n",
    "    print(\" - best_F1_score =\", best_F1_score)\n",
    "    print(\"    best F1 parameters:\")\n",
    "    for k,v in best_F1_params.items():\n",
    "      print(\"     -\", k, v)\n",
    "    print(\" - best_accuracy_score =\", best_accuracy_score)\n",
    "    print(\"   best accuracy parameters:\")\n",
    "    for k,v in best_accuracy_params.items():\n",
    "        print(\"     -\", k, v)\n",
    "\n",
    "    fvals_tuned.append(best_F1_score)\n",
    "    avals_tuned.append(best_accuracy_score)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "IhP4tGuvAAph"
   },
   "source": [
    "Best Result with Hyper Parameter Tuning\n",
    "\n",
    "split = 7\n",
    "\n",
    "DecisionTreeClassifier\n",
    "- best_F1_score = 0.8048323169950061\n",
    "best F1 parameters:\n",
    "- max_depth 10\n",
    "- min_samples_leaf 2\n",
    "- random_state 42\n",
    "- best_accuracy_score = 0.8089065965054093\n",
    "best accuracy parameters:\n",
    "- max_depth 10\n",
    "- min_samples_leaf 2\n",
    "- random_state 42\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "z97L0WodAQzn"
   },
   "source": [
    "### Question 4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 667
    },
    "executionInfo": {
     "elapsed": 1341,
     "status": "ok",
     "timestamp": 1603876417608,
     "user": {
      "displayName": "Doruk Kırbeyi",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gg7lZvmFjsLSadU1G9V3NlTGJW8Mf6p62xQPrUlkw=s64",
      "userId": "12126647258254016890"
     },
     "user_tz": -180
    },
    "id": "qEE0eO3jAQzo",
    "outputId": "e92eb7f3-9602-4ebd-cf89-17df49c1d889"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2160x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2160x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(30,10))\n",
    "kf = plt.subplot(2,2,1)\n",
    "\n",
    "kf.set_xlabel(\"K-Fold Classifier Type\")\n",
    "kf.set_ylabel(\"K-Fold Accuracy/F1 Score Value With Default Hyper Parameters\")\n",
    "\n",
    "kf.plot(names, avals, label=\"Accuracy Score\", color=\"red\", marker='o')\n",
    "kf.plot(names, fvals, label=\"F1 Score\", color=\"black\", marker='o')\n",
    "\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "plt.figure(figsize=(30,10))\n",
    "kf = plt.subplot(2,2,2)\n",
    "\n",
    "kf.set_xlabel(\"K-Fold Classifier Type\")\n",
    "kf.set_ylabel(\"K-Fold Accuracy/F1 Score Value With Hyper Parameter Tuning\")\n",
    "\n",
    "kf.plot(names_tuned, avals_tuned, label=\"Accuracy Score\", color=\"red\", marker='o')\n",
    "kf.plot(names_tuned, fvals_tuned, label=\"F1 Score\", color=\"black\", marker='o')\n",
    "\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "EpAjTN1sAQzt"
   },
   "source": [
    "### Question 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "id": "rLCUvYRYAQzu"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0        3\n",
      "1        2\n",
      "2        2\n",
      "3        2\n",
      "4        2\n",
      "        ..\n",
      "14742    2\n",
      "14743    2\n",
      "14744    5\n",
      "14745    1\n",
      "14746    4\n",
      "Name: workclass, Length: 14747, dtype: object\n",
      "0        15\n",
      "1         2\n",
      "2        15\n",
      "3        11\n",
      "4        15\n",
      "         ..\n",
      "14742     7\n",
      "14743     7\n",
      "14744    11\n",
      "14745    11\n",
      "14746    15\n",
      "Name: education, Length: 14747, dtype: object\n",
      "0        6\n",
      "1        4\n",
      "2        2\n",
      "3        6\n",
      "4        2\n",
      "        ..\n",
      "14742    2\n",
      "14743    4\n",
      "14744    4\n",
      "14745    2\n",
      "14746    2\n",
      "Name: marital-status, Length: 14747, dtype: object\n",
      "0         3\n",
      "1         5\n",
      "2         2\n",
      "3         0\n",
      "4        11\n",
      "         ..\n",
      "14742     3\n",
      "14743     3\n",
      "14744     7\n",
      "14745     3\n",
      "14746     4\n",
      "Name: occupation, Length: 14747, dtype: object\n",
      "0        4\n",
      "1        3\n",
      "2        0\n",
      "3        1\n",
      "4        0\n",
      "        ..\n",
      "14742    0\n",
      "14743    1\n",
      "14744    1\n",
      "14745    0\n",
      "14746    0\n",
      "Name: relationship, Length: 14747, dtype: object\n",
      "0        4\n",
      "1        4\n",
      "2        3\n",
      "3        4\n",
      "4        4\n",
      "        ..\n",
      "14742    4\n",
      "14743    4\n",
      "14744    2\n",
      "14745    4\n",
      "14746    4\n",
      "Name: race, Length: 14747, dtype: object\n",
      "0        1\n",
      "1        1\n",
      "2        1\n",
      "3        0\n",
      "4        1\n",
      "        ..\n",
      "14742    1\n",
      "14743    1\n",
      "14744    1\n",
      "14745    1\n",
      "14746    1\n",
      "Name: sex, Length: 14747, dtype: object\n",
      "0        38\n",
      "1        38\n",
      "2        38\n",
      "3        38\n",
      "4        38\n",
      "         ..\n",
      "14742    38\n",
      "14743    38\n",
      "14744    38\n",
      "14745    38\n",
      "14746    38\n",
      "Name: native-country, Length: 14747, dtype: object\n"
     ]
    }
   ],
   "source": [
    "X_test = pd.read_csv(\"X_test.csv\")\n",
    "imputeX(X_test)\n",
    "labelAll(X_test,labelingColumns)\n",
    "X_test = X_test.drop(X_test.columns[[8,9]], axis = 1)\n",
    "scaler = preprocessing.Normalizer()\n",
    "scalerX_test = scaler.fit(X_test)\n",
    "X_test = scalerX_test.transform(X_test)\n",
    "\n",
    "model = DecisionTreeClassifier(max_depth=10, min_samples_leaf=2, random_state=42)\n",
    "model.fit(X,np.ravel(y,order='C'))\n",
    "prediction = model.predict(X_test)\n",
    "pd.DataFrame(prediction).to_csv(\"170_classes_problem_census.txt\", index=False, header=False)"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "GROUP_problem_census.ipynb",
   "provenance": [
    {
     "file_id": "16dEFc4nMnmt5TP40lkEWQpVd-qVjdv8y",
     "timestamp": 1603888449599
    },
    {
     "file_id": "1LqJOo-TozrrZMQnUoZ6e5ZlKRWD3oicr",
     "timestamp": 1603799164752
    },
    {
     "file_id": "17_8mdlMD107ItliZ9NmYO-T1JvPnn1As",
     "timestamp": 1603650689054
    },
    {
     "file_id": "1QohAbd9UUdYuQTjvURteDrSAa5J8HQYm",
     "timestamp": 1603630521278
    }
   ]
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}