Risk Classification_updated.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.cm as cm\n",
    "import seaborn as sns\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn import tree\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import plot_confusion_matrix\n",
    "from sklearn.metrics import plot_roc_curve\n",
    "from sklearn.metrics import classification_report\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn.metrics import ConfusionMatrixDisplay\n",
    "from sklearn.metrics import roc_auc_score\n",
    "from sklearn import metrics\n",
    "from sklearn.metrics.cluster import contingency_matrix\n",
    "\n",
    "\n",
    "\n",
    "#constants\n",
    "random_state = 42\n",
    "plotColor = ['b','g','r','m','c', 'y']\n",
    "markers = ['+','o','*','^','v','>','<']\n",
    "\n",
    "#set up\n",
    "sns.set(style='white')\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>mortgage</th>\n",
       "      <th>loans</th>\n",
       "      <th>age</th>\n",
       "      <th>marital_status</th>\n",
       "      <th>income</th>\n",
       "      <th>risk</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>y</td>\n",
       "      <td>3</td>\n",
       "      <td>34</td>\n",
       "      <td>other</td>\n",
       "      <td>28060.70</td>\n",
       "      <td>bad loss</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>n</td>\n",
       "      <td>2</td>\n",
       "      <td>37</td>\n",
       "      <td>other</td>\n",
       "      <td>28009.34</td>\n",
       "      <td>bad loss</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>n</td>\n",
       "      <td>2</td>\n",
       "      <td>29</td>\n",
       "      <td>other</td>\n",
       "      <td>27614.60</td>\n",
       "      <td>bad loss</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>y</td>\n",
       "      <td>2</td>\n",
       "      <td>33</td>\n",
       "      <td>other</td>\n",
       "      <td>27287.18</td>\n",
       "      <td>bad loss</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>y</td>\n",
       "      <td>2</td>\n",
       "      <td>39</td>\n",
       "      <td>other</td>\n",
       "      <td>26954.06</td>\n",
       "      <td>bad loss</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  mortgage  loans  age marital_status    income      risk\n",
       "0        y      3   34          other  28060.70  bad loss\n",
       "1        n      2   37          other  28009.34  bad loss\n",
       "2        n      2   29          other  27614.60  bad loss\n",
       "3        y      2   33          other  27287.18  bad loss\n",
       "4        y      2   39          other  26954.06  bad loss"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv('ClassifyRisk.txt', sep=',')\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 246 entries, 0 to 245\n",
      "Data columns (total 6 columns):\n",
      " #   Column          Non-Null Count  Dtype  \n",
      "---  ------          --------------  -----  \n",
      " 0   mortgage        246 non-null    object \n",
      " 1   loans           246 non-null    int64  \n",
      " 2   age             246 non-null    int64  \n",
      " 3   marital_status  246 non-null    object \n",
      " 4   income          246 non-null    float64\n",
      " 5   risk            246 non-null    object \n",
      "dtypes: float64(1), int64(2), object(3)\n",
      "memory usage: 11.7+ KB\n"
     ]
    }
   ],
   "source": [
    "data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>loans</th>\n",
       "      <th>age</th>\n",
       "      <th>income</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>246.000000</td>\n",
       "      <td>246.000000</td>\n",
       "      <td>246.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1.308943</td>\n",
       "      <td>40.642276</td>\n",
       "      <td>38789.516748</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.843995</td>\n",
       "      <td>10.897017</td>\n",
       "      <td>13882.704336</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>15301.250000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>32.000000</td>\n",
       "      <td>26881.520000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>41.000000</td>\n",
       "      <td>37662.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2.000000</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>49398.227500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>3.000000</td>\n",
       "      <td>66.000000</td>\n",
       "      <td>78398.620000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            loans         age        income\n",
       "count  246.000000  246.000000    246.000000\n",
       "mean     1.308943   40.642276  38789.516748\n",
       "std      0.843995   10.897017  13882.704336\n",
       "min      0.000000   17.000000  15301.250000\n",
       "25%      1.000000   32.000000  26881.520000\n",
       "50%      1.000000   41.000000  37662.000000\n",
       "75%      2.000000   50.000000  49398.227500\n",
       "max      3.000000   66.000000  78398.620000"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "mortgage          False\n",
       "loans             False\n",
       "age               False\n",
       "marital_status    False\n",
       "income            False\n",
       "risk              False\n",
       "dtype: bool"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.isna().any()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "single     111\n",
       "married     78\n",
       "other       57\n",
       "Name: marital_status, dtype: int64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['marital_status'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "bad loss     123\n",
       "good risk    123\n",
       "Name: risk, dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['risk'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Encode the categorical values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>loans</th>\n",
       "      <th>age</th>\n",
       "      <th>income</th>\n",
       "      <th>risk</th>\n",
       "      <th>mortgage_y</th>\n",
       "      <th>marital_status_other</th>\n",
       "      <th>marital_status_single</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>34</td>\n",
       "      <td>28060.70</td>\n",
       "      <td>bad loss</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>37</td>\n",
       "      <td>28009.34</td>\n",
       "      <td>bad loss</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>29</td>\n",
       "      <td>27614.60</td>\n",
       "      <td>bad loss</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2</td>\n",
       "      <td>33</td>\n",
       "      <td>27287.18</td>\n",
       "      <td>bad loss</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2</td>\n",
       "      <td>39</td>\n",
       "      <td>26954.06</td>\n",
       "      <td>bad loss</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   loans  age    income      risk  mortgage_y  marital_status_other  \\\n",
       "0      3   34  28060.70  bad loss           1                     1   \n",
       "1      2   37  28009.34  bad loss           0                     1   \n",
       "2      2   29  27614.60  bad loss           0                     1   \n",
       "3      2   33  27287.18  bad loss           1                     1   \n",
       "4      2   39  26954.06  bad loss           1                     1   \n",
       "\n",
       "   marital_status_single  \n",
       "0                      0  \n",
       "1                      0  \n",
       "2                      0  \n",
       "3                      0  \n",
       "4                      0  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_encoded = pd.get_dummies(data, columns=['mortgage','marital_status'], drop_first=True)\n",
    "data_encoded.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 246 entries, 0 to 245\n",
      "Data columns (total 7 columns):\n",
      " #   Column                 Non-Null Count  Dtype  \n",
      "---  ------                 --------------  -----  \n",
      " 0   loans                  246 non-null    int64  \n",
      " 1   age                    246 non-null    int64  \n",
      " 2   income                 246 non-null    float64\n",
      " 3   risk                   246 non-null    object \n",
      " 4   mortgage_y             246 non-null    uint8  \n",
      " 5   marital_status_other   246 non-null    uint8  \n",
      " 6   marital_status_single  246 non-null    uint8  \n",
      "dtypes: float64(1), int64(2), object(1), uint8(3)\n",
      "memory usage: 8.5+ KB\n"
     ]
    }
   ],
   "source": [
    "data_encoded.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Formating Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>loans</th>\n",
       "      <th>age</th>\n",
       "      <th>income</th>\n",
       "      <th>mortgage_y</th>\n",
       "      <th>marital_status_other</th>\n",
       "      <th>marital_status_single</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>34</td>\n",
       "      <td>28060.70</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>37</td>\n",
       "      <td>28009.34</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>29</td>\n",
       "      <td>27614.60</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2</td>\n",
       "      <td>33</td>\n",
       "      <td>27287.18</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2</td>\n",
       "      <td>39</td>\n",
       "      <td>26954.06</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   loans  age    income  mortgage_y  marital_status_other  \\\n",
       "0      3   34  28060.70           1                     1   \n",
       "1      2   37  28009.34           0                     1   \n",
       "2      2   29  27614.60           0                     1   \n",
       "3      2   33  27287.18           1                     1   \n",
       "4      2   39  26954.06           1                     1   \n",
       "\n",
       "   marital_status_single  \n",
       "0                      0  \n",
       "1                      0  \n",
       "2                      0  \n",
       "3                      0  \n",
       "4                      0  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x = data_encoded.iloc[:,[0,1,2,4,5,6]]\n",
    "y = data_encoded.iloc[:,3]\n",
    "x.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = pd.get_dummies(y, columns=['risk'], drop_first=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Normalization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-12-90e6181a79a7>:4: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  x[['loans','age','income']] = sc.fit_transform(x[['loans','age','income']])\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\pandas\\core\\indexing.py:966: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  self.obj[item] = s\n"
     ]
    }
   ],
   "source": [
    "from sklearn.preprocessing import OrdinalEncoder\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "sc = StandardScaler()\n",
    "x[['loans','age','income']] = sc.fit_transform(x[['loans','age','income']])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>loans</th>\n",
       "      <th>age</th>\n",
       "      <th>income</th>\n",
       "      <th>mortgage_y</th>\n",
       "      <th>marital_status_other</th>\n",
       "      <th>marital_status_single</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2.007718</td>\n",
       "      <td>-0.610793</td>\n",
       "      <td>-0.774395</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.820462</td>\n",
       "      <td>-0.334927</td>\n",
       "      <td>-0.778102</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.820462</td>\n",
       "      <td>-1.070569</td>\n",
       "      <td>-0.806594</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.820462</td>\n",
       "      <td>-0.702748</td>\n",
       "      <td>-0.830226</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.820462</td>\n",
       "      <td>-0.151016</td>\n",
       "      <td>-0.854271</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      loans       age    income  mortgage_y  marital_status_other  \\\n",
       "0  2.007718 -0.610793 -0.774395           1                     1   \n",
       "1  0.820462 -0.334927 -0.778102           0                     1   \n",
       "2  0.820462 -1.070569 -0.806594           0                     1   \n",
       "3  0.820462 -0.702748 -0.830226           1                     1   \n",
       "4  0.820462 -0.151016 -0.854271           1                     1   \n",
       "\n",
       "   marital_status_single  \n",
       "0                      0  \n",
       "1                      0  \n",
       "2                      0  \n",
       "3                      0  \n",
       "4                      0  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(x, y, test_size = 0.6, random_state = 22,shuffle = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Decision Tree classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier(random_state=0)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_dt = DecisionTreeClassifier(random_state=0)\n",
    "clf_dt.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9256756756756757"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred=clf_dt.predict(X_test)\n",
    "from sklearn.metrics import accuracy_score\n",
    "accuracy_score(y_test, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn import tree\n",
    "plt.figure(figsize=(15, 10))\n",
    "tree.plot_tree(clf_dt, filled=True, rounded=True, \n",
    "          class_names=[\"bad loss\", \"good risk\"], feature_names=x.columns);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.92      0.92      0.92        74\n",
      "           1       0.92      0.92      0.92        74\n",
      "\n",
      "    accuracy                           0.92       148\n",
      "   macro avg       0.92      0.92      0.92       148\n",
      "weighted avg       0.92      0.92      0.92       148\n",
      "\n"
     ]
    },
    {
     "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 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "dtclf = DecisionTreeClassifier(random_state = 12)\n",
    "dtclf = dtclf.fit(X_train,y_train)\n",
    "y_pred = dtclf.predict(X_test)\n",
    "print(classification_report(y_test, y_pred))\n",
    "fpr, tpr, thresholds = metrics.roc_curve(y_test,y_pred,drop_intermediate=False)\n",
    "roc_auc = metrics.auc(fpr, tpr)\n",
    "display = metrics.RocCurveDisplay(fpr=fpr, tpr=tpr, roc_auc=roc_auc, estimator_name='Decision Tree Classifier')\n",
    "display.plot()  # doctest: +SKIP\n",
    "plt.plot([0, 1], [0, 1],'r--')\n",
    "plt.show() \n",
    "cm = confusion_matrix(y_test, y_pred)\n",
    "cm_display = ConfusionMatrixDisplay(cm,display_labels=[0,1]).plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Post Pruning with cost complexity pruning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "path = clf_dt.cost_complexity_pruning_path(X_train, y_train)\n",
    "ccp_alphas, impurities = path.ccp_alphas, path.impurities"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of nodes in the last tree is: 1 with ccp_alpha: 0.3687290969899666\n"
     ]
    }
   ],
   "source": [
    "clfs = []\n",
    "for ccp_alpha in ccp_alphas:\n",
    "    clf_dt = DecisionTreeClassifier(random_state=12, ccp_alpha=ccp_alpha)\n",
    "    clf_dt.fit(X_train, y_train)\n",
    "    clfs.append(clf_dt)\n",
    "print(\"Number of nodes in the last tree is: {} with ccp_alpha: {}\".format(\n",
    "      clfs[-1].tree_.node_count, ccp_alphas[-1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Accuracy vs alpha for training and testing sets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "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"
    }
   ],
   "source": [
    "train_scores = [clf_dt.score(X_train, y_train) for clf_dt in clfs]\n",
    "test_scores = [clf_dt.score(X_test, y_test) for clf_dt in clfs]\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "ax.set_xlabel(\"alpha\")\n",
    "ax.set_ylabel(\"accuracy\")\n",
    "ax.set_title(\"Accuracy vs alpha for training and testing sets\")\n",
    "ax.plot(ccp_alphas, train_scores, marker='o', label=\"train\",\n",
    "        drawstyle=\"steps-post\")\n",
    "ax.plot(ccp_alphas, test_scores, marker='o', label=\"test\",\n",
    "        drawstyle=\"steps-post\")\n",
    "ax.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier(ccp_alpha=0.016, random_state=0)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_dt = DecisionTreeClassifier(random_state=0, ccp_alpha=0.016)\n",
    "clf_dt.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9391891891891891"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred=clf_dt.predict(X_test)\n",
    "accuracy_score(y_test, y_pred)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(18, 5))\n",
    "tree.plot_tree(clf_dt, filled=True, rounded=True, \n",
    "          class_names=[\"bad loss\", \"good risk\"], feature_names=x.columns);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.92      0.96      0.94        74\n",
      "           1       0.96      0.92      0.94        74\n",
      "\n",
      "    accuracy                           0.94       148\n",
      "   macro avg       0.94      0.94      0.94       148\n",
      "weighted avg       0.94      0.94      0.94       148\n",
      "\n"
     ]
    },
    {
     "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": "iVBORw0KGgoAAAANSUhEUgAAATYAAAEKCAYAAACYBHl/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAaSklEQVR4nO3dfZQddX3H8fe9mxBWJAmxgU1YnkLkCwqGByNqSQGNtFRt2mKkBoH08KAHUCimihIPCUpPFUoRSsVao0hqDSbQYwm0SLSWJ9GqKQ+RrwETzJKkAskmIZBk997pHzOb3Fx2Z+YOe59mP69zBvbOzP3NbxP4nu9vfk+FIAgQEcmTYrMrICIy3BTYRCR3FNhEJHcU2EQkdxTYRCR3FNhEJHdGNbsC1cxsDDAd2ACUmlwdkTzqACYBP3P3nVkLMbMJwNiUt291900pyrwQuKzi1BHAHcC/ATcCncASd58fV06h1caxmdkpwIPNrofICDDD3R/K8kUzmzBu/+ClLdsKab+yGZiaJrhVPOOthAHtPcDDwKnAOmA5cJO73zfUd1suYyPM1Fh8c5muA5tdFUnrvJPf0uwqSEr9o3bxu0N+A9H/axmN3bKtwLdvKdE1Mf7GjS/AeZ/oOIAwu0sd2ICvAp8DpgCr3X0NgJktBmYDbRXYSgBdB8LBk5pdFUlrVP8+za6C1O51v+qZ+HtluroSHhJA2Pql28yqL/e6e2/1STObCXS6+/fM7CPsHYQ3AN1xz2zFwCYibaJMQDnxnt0Ge8W0EFgwyPmPEb5Tg7CTs/KdWWHvYl9LgU1EMgsIKBP/nj5g93u4GUBP1eXBsrV9CN+nzY1O9RB2dgzoAtbHPVOBTUQy66NMX0IHZN+ewNbj7mtTFPs24Nfuvj36/BhgZjYVWAPMARbFFaBxbCKSWZmAUsKRlNENYgoVmZ277yDM3pYBq4CngaVxBShjE5HMyikCV9I7uGrufidwZ9W5FcC0tGUosIlIZqUgoJTQFC01YaisApuIZBaQnJE1YwqAApuIZDbwHi3+nsZTYBORzPoD6EtIyfrVFBWRdlKikJiRlUg9n3TYKLCJSGblIDyS7mk0BTYRyaycImMrK2MTkXaipqiI5E5/UEycUtUfKLCJSBsJM7b4wKXhHiLSVspBQZ0HIpIv6ToPGk+BTUQyKwXFxLmgmisqIm2lTLGWFXQbRoFNRDLrC4rsSuj1DHtNG5u2KbCJSGZlCokDcMOMTYFNRNpEmWLicI9yqsWNhpcCm4hkVqJIKaEpmrSsUT0osIlIZmFTNH7rlHITug8U2EQks3JQSMzYyppSJSLtpC8YRV8Qn7H1BcrYRKSNhDMPkpqiescmIm2klKIpmnS9HhTYRCSzcObB8GZsZvZB4BpgP+B+d7/czGYCNwKdwBJ3nx9XhnaCF5HMykExmi869FFOeAdXycymALcBfwq8DTjRzM4EFgGzgGOA6dG5ISmwiUhmfUGRvqAj4agpzPwZYUbW4+59wNnAK8Bqd1/j7v3AYmB2XCFqiopIZuHMg6Sm6O7r3WZWfbnX3XsrPk8FdpnZ94FDgXuAp4ANFfdsALrjnqnAJiKZhQtNph7H9uAglxcCCyo+jwL+ADgNeBn4PvAqe082LZAwR0uBTUQyK6XI2CquzwB6qi73Vn3eCDzg7i8AmNndhM3OyvUsu4D1cc9UYBORzMopOgcqrve4+9qEIu8Bbjez8cA24ExgKXCVmU0F1gBzCDsThqTOAxHJrMyeDV2GOmqZd+DujwFfBh4CVgHPAV8F5gLLonNPEwa7ISljE5HM+gh7PuPvqW1Klbsv4rUZ2QpgWtoyFNhEJLMgRVM0qG24x7BQYBORzAYG4Sbd02gKbCKSWRlSLg3eWApsIpJZOUXGVsuUquGiwCYimdU4QLdhFNhEJLP+ILlXtD/hej0osIlIZun2PFDGJiJtRAtNikju6B2biOROQIoBuk2YuVnXwGZmc4D5wGjgJne/tZ7PayX3L5nAD+6cAMCunQWefaqT7658ijeOK3HbNZPpPnInHzjvpSbXUoZSLAZcccM6uo/cSblU4O/+6hA2PDem2dVqOX1BkVGJu1Q1PrDV7YlmdjBwHXAKcDxwsZm9pV7PazVnnL2J65c9w/XLnuHNx73CJV94nv5+uPqcKfzk/nHNrp4kOPmMrQBcOevNfPv6Lj62IHaVnBErCAq7V/gY6gia0BStZyidCfzQ3Te5+3bC2fgfquPzWtKv/7eT537dyR9/9CV2bO/g3E9t5L1nbW52tSTBo/8xjpv++hAADuzexeYX9NZmMGGvaPLRaPUMbJOpcTnfPPruzQdxzpUbAeg6dBdHn/hKk2skaZVLBebd9Fsu+eLzPLh8fLOr05IGekWTjkarZ2ArUuNyvnnz8pYO1j2zL8f//svNropkdMMVh3LBKUdzxfXrGNNZSv7CCJPUDE2zEGU91POJPcCkis+Jy/nmzRM/2Y8TZmxrdjUkg/eetYmzL/s/AHa+WiQoFyiXG595tLogGu4RdzTjHVs9Xxw8ACwws4nAduAs4OI6Pq/l9Dw7hq7DdjW7GpLBQ/eOY97fr+OGu56hY1TAbddMpm+nFpyu1h8U6UjIyPrzNAne3Z83s6uBHwH7AP/s7j+t1/Na0exLXhj0/LnzNja4JlKrna92cN3HD292NVpeOeoVTbqn0era1ePu3wG+U89niEjzBCTPPAg0V1RE2kma4RyaBC8ibUVzRUUkdxTYRCR3SuUi/eWEzVwSrteDApuIZFaPzVzM7EfAgUBfdOpjwP7AjUAnsMTd58eVocAmIpkNd1PUzArAUcBh7t4fnesEHDgVWAcsN7Mz3f2+ocpRYBORzIIUga3GmQcW/ft+M3sT8HXgCWC1u68BMLPFwGxAgU1Ehl+NGVu3mVVf7nX33orPBwArgE8QruP4X8CXqHFBDQU2EcmsFBQpJHUe7JmZ8OAglxcCCwY+uPujwKMDn83sG8C1wEMV30lcUEOBTUQyC1IM0K2YeTCDcHGMSpXZGmZ2CjDG3VdEpwrAWmpcUEOBTUQyKwcFCumboj3uvjahyPHAtWb2bsKm6PnAx4E7zWwqsAaYAyyKK0TLFYhIZkEQdg7EH+nLc/d7gOXAL4GfA4ui5ulcYBmwCniacEXuISljE5HMaszYUnH3zwOfrzq3ApiWtgwFNhHJbCArS7qn0RTYRCSzcrkACSsLN2PlYQU2Ecks7BHVskUikiMBKZqiCmwi0k7KQQG0bJGI5Ek43CP5nkZTYBOR7NJsr6eMTUTaSalcSJwrGqhXVETaSRAAaoqKSJ4EKToPNEBXRNpKQIrA1krDPcxsQtwX3X3T8FdHRNpNE1qaieIythcJ6zxYuA2AjrrUSETaRpBiShXlQsNztiEDm7trSSMRiZXmHRtBCwW2AWZWBK4EjiVch/wy4MvuXqpz3USkxaXpFW1GWzVN58H1wERgOuHClH9EuEzvJ+tYLxFpA2kztkZL09x8L+HqlTvcfQtwBvC+elZKRNrEQGBLOhosTWDrc/fdO8K4+06gv35VEpF2MTBXNOlotDRN0SfN7FKgw8JNAa8EVta3WiLSDoKA5F7RJgS2NBnb5cCJwEHAw8AbgSvqWSkRaRNByqPBEjM2d98KXNCAuohIm2nVzoM0wz0OBL5C2GHQB9wLfKpqW3oRGYladLhHmqbo14HfAO8A/gDYDHytnpUSkXZRSHk0VprOg8PdfVbF53lm9kS9KiQibaQcHUn3ZGBmNwC/5+5zzWwmcCPQCSxx9/lx302Tsa03syMqHtYNbMhWVRHJlTqNYzOz9wLnRz93AouAWcAxwHQzOzPu+3Gre/w7Yet4IrDSzB4ASsDpwOM111REcilpnFqtYS1aWeg64G8Id39/B7Da3ddE1xcDs4H7hiojrim6dIjzy2usp4jkVW2dB93hUNi99A7SEfk14GrgkOjzZPZuJW4AuuMeGbe6x+2DnTezAjA1rlARGSFqG+7x4CBXFwILBj6Y2YXAOndfYWZzo9NF9g6fBRLe3KUZ7vExwonw+1WcfgHoSvquiORcAIX0GdsMoKfqanW2djYwycxWAhMIJwQcRvgabEAXsD7ukWl6Ra8iHMN2NTAf+CAJaaCIjBApF5qM9Lj72rhb3X33AhtRxnYa8HFgtZlNBdYAcwg7E4aUpld0k7s/Rjg/9CB3vw44NcX3RCTvGjClyt13EK4wtAxYBTzN0H0AQLqMrc/MDgBWE/ZO/AAtCy4iUNeZB+7+LeBb0c8rCHtIU0kT2P4JuIewCbrSzP6MMGKKiLTkbi6JTVF3XwScEe1K9S7gC4Qv+ERkpGvRhSbjBuheWfW58uMlhNMbRGQEK7ToJPi4puhxMddaMPkUkYZrt8Dm7n/ZyIpUm/u+kxld3reZVZAafOO3mpDSLjZuLHDuOfsPS1ntmLGJiMRr14UmRUSGpIxNRHKnXQNbtBP8pwh3gr8M7QQvIpFCQPJCkq0Y2Nh7J/gC2gleRAa0aMZW607wW9FO8CISKQTpjkbTTvAikl1AipkHja+WdoIXkezauCmqneBFZFCt2hTVTvAiklnb9oqa2c2DnXd39YqKjHRt3BR9qeLYRrh6ribBi0hDVtDNIk1TdGHlZzP7W+D7dauRiLSNVp0EnyZj24u7bwMOrkNdRESGRZp3bLewJ+YWgJOAX9WzUiLSJlo0Y0szju3Fip8D4A7gX+pTHRFpK2UoJPSKBkm9pnWQJrAd6e7n1b0mItJ+WjRjS/OObZqZNX6lOBFpeQVSDNBtQr3SZGwbgKfM7CfAywMnNY5NROqRsZnZtcCHom9+w91vNLOZhBtIdQJL3H1+XBlDZmxmNib68VFgCfAce49pE5ERbrinVJnZqcB7gLcBbwc+YWbTgEXALOAYYLqZnRlXTlzG9ihwYvU4NhGR3cokT6mqofPA3X9sZqe7e7+ZHUwYo8YDq919DYCZLQZmA/cNVU5cYNN7NRGJlSoj23O9u2p/YoBed++tPOHufWa2EJgHfA+YTPhKbMAGoDvukXGBbV8zO4EhApy7/yKuYBEZAWp7x/bgIFcXAguqT7r7NWb2JeDfgaOqnlIgIQ+MC2xTgGUMHtiC6LqIjGS1BbYZQE/V1b2yNTM7GtjX3Ve6+ytmdhdhR0LlHitdwPq4R8YFtlXufkJClUVkBBsY7pFSj7uvTbhnCrDQzE4hDImzgK8B15vZVGANMIewM2FINc8VFRHZbZhX93D3e4HlwC+BnwOPuPt3CfddWQasAp4GlsaVE5ex/Xf66ojISFRIMaWqll5RAHdfQNV7N3dfAUxLW8aQgc3dL6+tOiIy4rTolCrtBC8imRVozXFhCmwikp0yNhHJnRQDdAMFNhFpK8rYRCRvCkFyr2hL7isqIjIkZWwikjdpJsErYxOR9qKMTUTyRhmbiOTPMC80OVwU2EQkszSre7TqZi4iIoPTOzYRyZvwHVt85NI7NhFpL8rYRCRv1CsqIrmTZqHJxIUo60CBTUSyU1NURPKoGU3NJApsIpKdMjYRyRt1HohI7hTKAYVywji2hOv1oMAmItmNxKaomY0FHgE+kGIH6Fz78EW/5eTTX2LU6IDl/zqJ+++a1OwqySCW/0M3Kx+YQGlXkdPP28Bhx73MHZ87kmJHwEFTdjD3y6spapvx3eox3MPMrgE+HH1c7u6fNrOZwI1AJ7DE3efHlVG3vyIzOxl4CDiqXs9oF8dN7+WY47cy75zj+cz505g4aWezqySDePrRcTz787F89q7H+fT3HmfT+jF8/6ZD+eDl6/jsXU/Qv6vA4ysmNLuarWeYdoEHiALYGcAJwPHASWb2EWARMAs4BphuZmfGlVPPjO0i4FLgjjo+oy2cdMpm1q7ej/m3rOIN+/Wz6IYpza6SDOKpH4/n4KO3c+tFx/Dqyx18+HNrKRQDtveOIghgx8sddIxuwmjTFlaHzoMNwKfcfReAmf2KMDla7e5ronOLgdnAfUMVUrfA5u4XRpWo1yPaxtjxfRw4eQcLLjmWgw7ewTW3PsXF7387rbnV7Mi1bdNoXnp+DJd/cxUvrtuXmy84hll/9VsWzz+Se245hM79Sxz9zi3NrmZrCYLk/fX2XO8eJB70unvvwAd3f2rgZzN7M2GT9BbCgDdgA9Ad90i9LWiArb2j+fnDE+jvK/L82jewa2eRcRP6ml0tqfLGA/o49tReRu0T0HXkq4weE/D1TxpXLX2C6370C9591u9Y8sUjml3NljLwji3piDwIrKk6rhisXDN7K/AD4K+B37B3o7ZAwvKVCmwNsOoXY3n7KZuAgAkTd7LvG0ps6x3d7GpJlTdP38qT/zWeIIDNG/dh1ytFJh62g879+wEYf9AuXtmigQSVBpqiSUdkBnBE1XFTdZlm9vvACuAqd78d6AEqe9u6gPVx9dLfUgP89Mdv4ti3b+GmJb+kUIR//MJUymU1Q1vNtJmb+fVj4/jiB6cRlAuc88VnGdNZ5muXHk1xVMCo0WXO/9Izza5ma6mtKdqTNDrCzA4B/g04291/GJ1+LLxkUwmzvDmEnQlDUmBrkEV/pw6DdjD76rWvOffZux9vfEXaRB06D+YB+wI3VryPuw2YCyyLrt0LLI0rpO6Bzd0Pr/czRKSJhnEArrtfDlw+xOVpactRxiYi2aXI2HI380BEcq4UQDEhcpU0V1RE2ohW9xCR/KmtV7RhFNhEJDNlbCKST1oaXETypFAKKCSkZAV1HohIOykEQYqd4BXYRKSdjMQVdEUk5wJS9Io2pCZ7UWATkczUKyoiOZRiHFsTUjYFNhHJrFAKKCQELvWKikh7UeeBiOSNhnuISP5orqiI5E5AwrYqqCkqIm0mRVNUGZuItJdymcSUrdz4TaYV2EQkuzQxq/FxTYFNRLILZx4k9Yo2qDIVFNhEJDv1iopI/mhKlYjkTSnFLPgMGZuZjQUeAT7g7mvNbCZwI9AJLHH3+XHfL9b8RBGRyMDMg6SjFmZ2MvAQcFT0uRNYBMwCjgGmm9mZcWUosIlIdgPv2JKO2lwEXAqsjz6/A1jt7mvcvR9YDMyOK0BNURHJLlXg2n2928yqL/a6e2/lCXe/EKDi3snAhopbNgDdcU9UYBOR7IKUy3sUAHhwkIsLgQUJBRSrHlIgYXScApuIZBfs/ke8MLDNAHqqrvS+5t7X6gEmVXzuYk8zdVAKbCKSXSnFlCrKA2/ze9x9bYanPAaYmU0F1gBzCDsThqTOAxHJLiinO14Hd98BzAWWAauAp4Glcd9RxiYi2aV9x5aBux9e8fMKYFra7yqwiUh2aXpFmzBZVIFNRLKrbbhHwyiwiUh2aTZMbgIFNhHJrlSCoBR/TyHheh0osIlIdmqKikj+qPNARPKmHIRHLAU2EWkjQRAQJAzADbSCroi0lVI5eReq1znzIAsFNhHJLkgR2AoKbCLSTrSZi4jkTVAuEyRkbIEyNhFpK2lmHmhfURFpK2mGe2gcm4i0lXKJoKQpVSKSJ2kWktRwDwA6APqLO5tdD6nBxo2FZldBUnrhhd1/Vx2vt6y+4i6CYnxTs7/Y93ofU7NWDGyTADYe8Hiz6yE1OPec/ZtdBandJODZjN/dCmz+XfczB6S8f3P0nYZoxcD2M8LdbDYAjW+ci+RfB2FQ+1nWAtx9U7S5ytiUX9nq7puyPq9WhWbM4xIRqSftUiUiuaPAJiK5o8AmIrmjwCYiuaPAJiK5o8AmIrmjwCYiudOKA3RzyczmAPOB0cBN7n5rk6skCcxsLPAI8AF3X9vk6kgNlLE1gJkdDFwHnAIcD1xsZm9pbq0kjpmdDDwEHNXsukjtFNgaYybwQ3ff5O7bgaXAh5pcJ4l3EXApsL7ZFZHaqSnaGJMJ574O2AC8o0l1kRTc/UIAM2t2VSQDZWyNUWTvBZILQOMXqRIZIRTYGqOHaDmmSBdq4ojUjZqijfEAsMDMJgLbgbOAi5tbJZH8UsbWAO7+PHA18CNgJfAdd/9pc2slkl9aj01EckcZm4jkjgKbiOSOApuI5I4Cm4jkjgKbiOSOxrG1CTM7nHAPyCcqTheAr7j7otdZ9j3AUnf/lpmtBE5z994h7h0H3O3u76nxGR8CLnP306rOnwb8g7sfm/D9AJjo7i/W8MxvAU+6+w211FXanwJbe3nV3Y8f+BCtGvKkmf2Puw/LDtOV5Q/hADTPVVqcAlsbc/fnzWw1cJSZnQhcAOwHbHH3083sAuASwlcOLxFmTE+b2WTgdsLJ+c8BBw6UWZkZmdlngfOBfmA1MBf4JtAZZXYnES7r8xXgTYQb8d48kEGa2bXAOdGzVyf9PmZ2FHArsD/hFLSVwNnuviO65Tozmx79PvPd/Z7oe4P+njX9YUqu6B1bGzOzdwFTgceiU28lbEaebmanEgalGe5+AvBl4O7ovluBn7j7W4FPAkcPUvafEAayd0XNxDXAZcBfsidzLBAuwXSVu58EnArMM7N3mtkswqljxwPvBsal+JUuAm5393dGv9cRwPsrrv/G3U8EPgrcbmYTE35PGaGUsbWXgUwJwr+7F4Fz3H1dtLzO4+6+Nbr+fsLg8EjF0jsHmNkEwvXh5gG4+zNm9sNBnjUT+J67b47uuxJ2v+sbcBRwJLCo4hmdwAnAW4C73H1b9L1FhEE0zmeA95nZp6OyJwNvrLh+W1SXJ81sFfAuwsU7h/o9ZYRSYGsvrya8A3u54ucO4A53/wyAmRUJA8VmwiWUChX39g9SVj8VSy2Z2XhgfNU9HYTN3sr3fgcBW4DrUzyj2r8S/jd5J7AcOLSqjFLFz0Wgj/jfU0YoNUXz6z+Bj5jZwHJJHwdWRD//B9HqImZ2KHD6IN9/APjzaN1/gAXAlYQBqsPMCoADr5rZR6OyDgGeJHz3dh8w28zGR8Hm3BR1/kPgWndfEn0+mTBwDZgbPedE9jTB435PGaGUseWUu99vZl8CfmBmZWAr8OfuHpjZpcA3zexXhGvFrRzk+/dG+zI8HDXxniJ8B/YK8NPo8wxgFvCVqPk4Gvi8uz8MYGbHAf9DmD39LzAxodqfA+42s+2EWd+PCQPYgClm9kvCTPIv3H0TEPd71vJHJjmi1T1EJHfUFBWR3FFgE5HcUWATkdxRYBOR3FFgE5HcUWATkdxRYBOR3FFgE5Hc+X9X53m2ncO7RAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(classification_report(y_test, y_pred))\n",
    "fpr, tpr, thresholds = metrics.roc_curve(y_test,y_pred,drop_intermediate=False)\n",
    "roc_auc = metrics.auc(fpr, tpr)\n",
    "display = metrics.RocCurveDisplay(fpr=fpr, tpr=tpr, roc_auc=roc_auc, estimator_name='Decision Tree Classifier')\n",
    "display.plot()  \n",
    "plt.plot([0, 1], [0, 1],'r--')\n",
    "plt.show() \n",
    "cm = confusion_matrix(y_test, y_pred)\n",
    "cm_display = ConfusionMatrixDisplay(cm,display_labels=[0,1]).plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: xgboost in c:\\programdata\\anaconda3\\lib\\site-packages (1.4.2)\n",
      "Requirement already satisfied: scipy in c:\\programdata\\anaconda3\\lib\\site-packages (from xgboost) (1.5.0)\n",
      "Requirement already satisfied: numpy in c:\\programdata\\anaconda3\\lib\\site-packages (from xgboost) (1.19.5)\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "pip install xgboost"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "XG BOOST"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "XGBClassifier(base_score=None, booster=None, colsample_bylevel=None,\n",
      "              colsample_bynode=None, colsample_bytree=None, gamma=None,\n",
      "              gpu_id=None, importance_type='gain', interaction_constraints=None,\n",
      "              learning_rate=None, max_delta_step=None, max_depth=None,\n",
      "              min_child_weight=None, missing=nan, monotone_constraints=None,\n",
      "              n_estimators=100, n_jobs=None, num_parallel_tree=None,\n",
      "              random_state=None, reg_alpha=None, reg_lambda=None,\n",
      "              scale_pos_weight=None, subsample=None, tree_method=None,\n",
      "              validate_parameters=None, verbosity=None)\n"
     ]
    }
   ],
   "source": [
    "from xgboost import XGBClassifier\n",
    "from sklearn.model_selection import cross_val_score, KFold\n",
    "\n",
    "xgbc = XGBClassifier()\n",
    "print(xgbc)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[09:57:20] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.4.0/src/learner.cc:1095: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\xgboost\\sklearn.py:1146: UserWarning: The use of label encoder in XGBClassifier is deprecated and will be removed in a future release. To remove this warning, do the following: 1) Pass option use_label_encoder=False when constructing XGBClassifier object; and 2) Encode your labels (y) as integers starting with 0, i.e. 0, 1, 2, ..., [num_class - 1].\n",
      "  warnings.warn(label_encoder_deprecation_msg, UserWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:63: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(*args, **kwargs)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "XGBClassifier(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
       "              colsample_bynode=1, colsample_bytree=1, gamma=0, gpu_id=-1,\n",
       "              importance_type='gain', interaction_constraints='',\n",
       "              learning_rate=0.300000012, max_delta_step=0, max_depth=6,\n",
       "              min_child_weight=1, missing=nan, monotone_constraints='()',\n",
       "              n_estimators=100, n_jobs=8, num_parallel_tree=1, random_state=0,\n",
       "              reg_alpha=0, reg_lambda=1, scale_pos_weight=1, subsample=1,\n",
       "              tree_method='exact', validate_parameters=1, verbosity=None)"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xgbc.fit(X_train,y_train)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\xgboost\\sklearn.py:1146: UserWarning: The use of label encoder in XGBClassifier is deprecated and will be removed in a future release. To remove this warning, do the following: 1) Pass option use_label_encoder=False when constructing XGBClassifier object; and 2) Encode your labels (y) as integers starting with 0, i.e. 0, 1, 2, ..., [num_class - 1].\n",
      "  warnings.warn(label_encoder_deprecation_msg, UserWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:63: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(*args, **kwargs)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\xgboost\\sklearn.py:1146: UserWarning: The use of label encoder in XGBClassifier is deprecated and will be removed in a future release. To remove this warning, do the following: 1) Pass option use_label_encoder=False when constructing XGBClassifier object; and 2) Encode your labels (y) as integers starting with 0, i.e. 0, 1, 2, ..., [num_class - 1].\n",
      "  warnings.warn(label_encoder_deprecation_msg, UserWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:63: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(*args, **kwargs)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\xgboost\\sklearn.py:1146: UserWarning: The use of label encoder in XGBClassifier is deprecated and will be removed in a future release. To remove this warning, do the following: 1) Pass option use_label_encoder=False when constructing XGBClassifier object; and 2) Encode your labels (y) as integers starting with 0, i.e. 0, 1, 2, ..., [num_class - 1].\n",
      "  warnings.warn(label_encoder_deprecation_msg, UserWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:63: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(*args, **kwargs)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\xgboost\\sklearn.py:1146: UserWarning: The use of label encoder in XGBClassifier is deprecated and will be removed in a future release. To remove this warning, do the following: 1) Pass option use_label_encoder=False when constructing XGBClassifier object; and 2) Encode your labels (y) as integers starting with 0, i.e. 0, 1, 2, ..., [num_class - 1].\n",
      "  warnings.warn(label_encoder_deprecation_msg, UserWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:63: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(*args, **kwargs)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\xgboost\\sklearn.py:1146: UserWarning: The use of label encoder in XGBClassifier is deprecated and will be removed in a future release. To remove this warning, do the following: 1) Pass option use_label_encoder=False when constructing XGBClassifier object; and 2) Encode your labels (y) as integers starting with 0, i.e. 0, 1, 2, ..., [num_class - 1].\n",
      "  warnings.warn(label_encoder_deprecation_msg, UserWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:63: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(*args, **kwargs)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[09:57:27] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.4.0/src/learner.cc:1095: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[09:57:27] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.4.0/src/learner.cc:1095: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[09:57:27] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.4.0/src/learner.cc:1095: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[09:57:27] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.4.0/src/learner.cc:1095: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[09:57:27] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.4.0/src/learner.cc:1095: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "Mean cross-validation score: 0.91\n"
     ]
    }
   ],
   "source": [
    "scores = cross_val_score(xgbc, X_train, y_train, cv=5)\n",
    "print(\"Mean cross-validation score: %.2f\" % scores.mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\xgboost\\sklearn.py:1146: UserWarning: The use of label encoder in XGBClassifier is deprecated and will be removed in a future release. To remove this warning, do the following: 1) Pass option use_label_encoder=False when constructing XGBClassifier object; and 2) Encode your labels (y) as integers starting with 0, i.e. 0, 1, 2, ..., [num_class - 1].\n",
      "  warnings.warn(label_encoder_deprecation_msg, UserWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:63: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(*args, **kwargs)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\xgboost\\sklearn.py:1146: UserWarning: The use of label encoder in XGBClassifier is deprecated and will be removed in a future release. To remove this warning, do the following: 1) Pass option use_label_encoder=False when constructing XGBClassifier object; and 2) Encode your labels (y) as integers starting with 0, i.e. 0, 1, 2, ..., [num_class - 1].\n",
      "  warnings.warn(label_encoder_deprecation_msg, UserWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:63: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(*args, **kwargs)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\xgboost\\sklearn.py:1146: UserWarning: The use of label encoder in XGBClassifier is deprecated and will be removed in a future release. To remove this warning, do the following: 1) Pass option use_label_encoder=False when constructing XGBClassifier object; and 2) Encode your labels (y) as integers starting with 0, i.e. 0, 1, 2, ..., [num_class - 1].\n",
      "  warnings.warn(label_encoder_deprecation_msg, UserWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:63: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(*args, **kwargs)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\xgboost\\sklearn.py:1146: UserWarning: The use of label encoder in XGBClassifier is deprecated and will be removed in a future release. To remove this warning, do the following: 1) Pass option use_label_encoder=False when constructing XGBClassifier object; and 2) Encode your labels (y) as integers starting with 0, i.e. 0, 1, 2, ..., [num_class - 1].\n",
      "  warnings.warn(label_encoder_deprecation_msg, UserWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:63: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(*args, **kwargs)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\xgboost\\sklearn.py:1146: UserWarning: The use of label encoder in XGBClassifier is deprecated and will be removed in a future release. To remove this warning, do the following: 1) Pass option use_label_encoder=False when constructing XGBClassifier object; and 2) Encode your labels (y) as integers starting with 0, i.e. 0, 1, 2, ..., [num_class - 1].\n",
      "  warnings.warn(label_encoder_deprecation_msg, UserWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:63: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(*args, **kwargs)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\xgboost\\sklearn.py:1146: UserWarning: The use of label encoder in XGBClassifier is deprecated and will be removed in a future release. To remove this warning, do the following: 1) Pass option use_label_encoder=False when constructing XGBClassifier object; and 2) Encode your labels (y) as integers starting with 0, i.e. 0, 1, 2, ..., [num_class - 1].\n",
      "  warnings.warn(label_encoder_deprecation_msg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[09:57:36] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.4.0/src/learner.cc:1095: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[09:57:36] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.4.0/src/learner.cc:1095: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[09:57:36] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.4.0/src/learner.cc:1095: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[09:57:36] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.4.0/src/learner.cc:1095: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[09:57:36] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.4.0/src/learner.cc:1095: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[09:57:36] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.4.0/src/learner.cc:1095: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[09:57:36] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.4.0/src/learner.cc:1095: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[09:57:36] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.4.0/src/learner.cc:1095: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[09:57:36] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.4.0/src/learner.cc:1095: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[09:57:36] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.4.0/src/learner.cc:1095: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:63: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(*args, **kwargs)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\xgboost\\sklearn.py:1146: UserWarning: The use of label encoder in XGBClassifier is deprecated and will be removed in a future release. To remove this warning, do the following: 1) Pass option use_label_encoder=False when constructing XGBClassifier object; and 2) Encode your labels (y) as integers starting with 0, i.e. 0, 1, 2, ..., [num_class - 1].\n",
      "  warnings.warn(label_encoder_deprecation_msg, UserWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:63: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(*args, **kwargs)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\xgboost\\sklearn.py:1146: UserWarning: The use of label encoder in XGBClassifier is deprecated and will be removed in a future release. To remove this warning, do the following: 1) Pass option use_label_encoder=False when constructing XGBClassifier object; and 2) Encode your labels (y) as integers starting with 0, i.e. 0, 1, 2, ..., [num_class - 1].\n",
      "  warnings.warn(label_encoder_deprecation_msg, UserWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:63: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(*args, **kwargs)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\xgboost\\sklearn.py:1146: UserWarning: The use of label encoder in XGBClassifier is deprecated and will be removed in a future release. To remove this warning, do the following: 1) Pass option use_label_encoder=False when constructing XGBClassifier object; and 2) Encode your labels (y) as integers starting with 0, i.e. 0, 1, 2, ..., [num_class - 1].\n",
      "  warnings.warn(label_encoder_deprecation_msg, UserWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:63: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(*args, **kwargs)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\xgboost\\sklearn.py:1146: UserWarning: The use of label encoder in XGBClassifier is deprecated and will be removed in a future release. To remove this warning, do the following: 1) Pass option use_label_encoder=False when constructing XGBClassifier object; and 2) Encode your labels (y) as integers starting with 0, i.e. 0, 1, 2, ..., [num_class - 1].\n",
      "  warnings.warn(label_encoder_deprecation_msg, UserWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:63: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(*args, **kwargs)\n"
     ]
    }
   ],
   "source": [
    "kfold = KFold(n_splits=10, shuffle=True)\n",
    "kf_cv_scores = cross_val_score(xgbc, X_train, y_train, cv=kfold )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "K-fold CV average score: 0.91\n"
     ]
    }
   ],
   "source": [
    "print(\"K-fold CV average score: %.2f\" % kf_cv_scores.mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[65  9]\n",
      " [ 2 72]]\n"
     ]
    }
   ],
   "source": [
    "ypred = xgbc.predict(X_test)\n",
    "cm = confusion_matrix(y_test,ypred)\n",
    "print(cm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.97      0.88      0.92        74\n",
      "           1       0.89      0.97      0.93        74\n",
      "\n",
      "    accuracy                           0.93       148\n",
      "   macro avg       0.93      0.93      0.93       148\n",
      "weighted avg       0.93      0.93      0.93       148\n",
      "\n"
     ]
    },
    {
     "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 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(classification_report(y_test, ypred))\n",
    "fpr, tpr, thresholds = metrics.roc_curve(y_test,ypred,drop_intermediate=False)\n",
    "roc_auc = metrics.auc(fpr, tpr)\n",
    "display = metrics.RocCurveDisplay(fpr=fpr, tpr=tpr, roc_auc=roc_auc, estimator_name='XGBClassifier')\n",
    "display.plot()  \n",
    "plt.plot([0, 1], [0, 1],'r--')\n",
    "plt.show() \n",
    "cm = confusion_matrix(y_test, ypred)\n",
    "cm_display = ConfusionMatrixDisplay(cm,display_labels=[0,1]).plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9121621621621622\n",
      "[[64 10]\n",
      " [ 3 71]]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:63: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(*args, **kwargs)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import BaggingClassifier\n",
    "bag_model=BaggingClassifier(base_estimator=clf_dt, n_estimators=100, bootstrap=True)\n",
    "bag_model=bag_model.fit(X_train,y_train)\n",
    "ytest_pred=bag_model.predict(X_test)\n",
    "print(bag_model.score(X_test, y_test))\n",
    "print(confusion_matrix(y_test, ytest_pred)) \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.96      0.86      0.91        74\n",
      "           1       0.88      0.96      0.92        74\n",
      "\n",
      "    accuracy                           0.91       148\n",
      "   macro avg       0.92      0.91      0.91       148\n",
      "weighted avg       0.92      0.91      0.91       148\n",
      "\n"
     ]
    },
    {
     "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 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(classification_report(y_test, ytest_pred))\n",
    "fpr, tpr, thresholds = metrics.roc_curve(y_test,ytest_pred,drop_intermediate=False)\n",
    "roc_auc = metrics.auc(fpr, tpr)\n",
    "display = metrics.RocCurveDisplay(fpr=fpr, tpr=tpr, roc_auc=roc_auc, estimator_name='Bagging_Classifier')\n",
    "display.plot()  \n",
    "plt.plot([0, 1], [0, 1],'r--')\n",
    "plt.show() \n",
    "cm = confusion_matrix(y_test, ytest_pred)\n",
    "cm_display = ConfusionMatrixDisplay(cm,display_labels=[0,1]).plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.9527027027027027\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-36-6c7d10eb2b3c>:3: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
      "  clf.fit(X_train,y_train)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "clf=RandomForestClassifier(n_estimators=100)\n",
    "clf.fit(X_train,y_train)\n",
    "y_pred_randomforest=clf.predict(X_test)\n",
    "print(\"Accuracy:\",metrics.accuracy_score(y_test, y_pred_randomforest))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.99      0.92      0.95        74\n",
      "           1       0.92      0.99      0.95        74\n",
      "\n",
      "    accuracy                           0.95       148\n",
      "   macro avg       0.95      0.95      0.95       148\n",
      "weighted avg       0.95      0.95      0.95       148\n",
      "\n"
     ]
    },
    {
     "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 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(classification_report(y_test, y_pred_randomforest))\n",
    "fpr, tpr, thresholds = metrics.roc_curve(y_test,y_pred_randomforest,drop_intermediate=False)\n",
    "roc_auc = metrics.auc(fpr, tpr)\n",
    "display = metrics.RocCurveDisplay(fpr=fpr, tpr=tpr, roc_auc=roc_auc, estimator_name='Random_Forest')\n",
    "display.plot()  \n",
    "plt.plot([0, 1], [0, 1],'r--')\n",
    "plt.show() \n",
    "cm = confusion_matrix(y_test, y_pred_randomforest)\n",
    "cm_display = ConfusionMatrixDisplay(cm,display_labels=[0,1]).plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.9121621621621622\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:63: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(*args, **kwargs)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "model = GradientBoostingClassifier()\n",
    "model.fit(X_train,y_train)\n",
    "y_pred_grad_boost=model.predict(X_test)\n",
    "print(\"Accuracy:\",metrics.accuracy_score(y_test, y_pred_grad_boost))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.94      0.88      0.91        74\n",
      "           1       0.89      0.95      0.92        74\n",
      "\n",
      "    accuracy                           0.91       148\n",
      "   macro avg       0.91      0.91      0.91       148\n",
      "weighted avg       0.91      0.91      0.91       148\n",
      "\n"
     ]
    },
    {
     "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 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(classification_report(y_test, y_pred_grad_boost))\n",
    "fpr, tpr, thresholds = metrics.roc_curve(y_test,y_pred_grad_boost,drop_intermediate=False)\n",
    "roc_auc = metrics.auc(fpr, tpr)\n",
    "display = metrics.RocCurveDisplay(fpr=fpr, tpr=tpr, roc_auc=roc_auc, estimator_name='Grad Boost')\n",
    "display.plot()  \n",
    "plt.plot([0, 1], [0, 1],'r--')\n",
    "plt.show() \n",
    "cm = confusion_matrix(y_test, y_pred_grad_boost)\n",
    "cm_display = ConfusionMatrixDisplay(cm,display_labels=[0,1]).plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "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": 4
}