imbalance_class_undersampling_oversampling_updated.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import warnings\n",
    "from imblearn.over_sampling import SMOTE\n",
    "from imblearn.pipeline import make_pipeline\n",
    "from pylab import rcParams\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import precision_score, recall_score, confusion_matrix\n",
    "from sklearn.metrics import f1_score, roc_auc_score, roc_curve\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.model_selection import GridSearchCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "np.random.seed(27)\n",
    "rcParams['figure.figsize'] = 10, 6\n",
    "warnings.filterwarnings('ignore')\n",
    "sns.set(style=\"darkgrid\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def generate_model_report(y_actual, y_predicted):\n",
    "    print(\"Accuracy = \" , accuracy_score(y_actual, y_predicted))\n",
    "    print(\"Precision = \" ,precision_score(y_actual, y_predicted))\n",
    "    print(\"Recall = \" ,recall_score(y_actual, y_predicted))\n",
    "    print(\"F1 Score = \" ,f1_score(y_actual, y_predicted))\n",
    "    pass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def generate_auc_roc_curve(clf, X_test):\n",
    "    y_pred_proba = clf.predict_proba(X_test)[:, 1]\n",
    "    fpr, tpr, thresholds = roc_curve(Y_test,  y_pred_proba)\n",
    "    auc = roc_auc_score(Y_test, y_pred_proba)\n",
    "    plt.plot(fpr,tpr,label=\"AUC ROC Curve with Area Under the curve =\"+str(auc))\n",
    "    plt.legend(loc=4)\n",
    "    plt.show()\n",
    "    pass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('data.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "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>Time</th>\n",
       "      <th>V1</th>\n",
       "      <th>V2</th>\n",
       "      <th>V3</th>\n",
       "      <th>V4</th>\n",
       "      <th>V5</th>\n",
       "      <th>V6</th>\n",
       "      <th>V7</th>\n",
       "      <th>V8</th>\n",
       "      <th>V9</th>\n",
       "      <th>...</th>\n",
       "      <th>V21</th>\n",
       "      <th>V22</th>\n",
       "      <th>V23</th>\n",
       "      <th>V24</th>\n",
       "      <th>V25</th>\n",
       "      <th>V26</th>\n",
       "      <th>V27</th>\n",
       "      <th>V28</th>\n",
       "      <th>Amount</th>\n",
       "      <th>Class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.359807</td>\n",
       "      <td>-0.072781</td>\n",
       "      <td>2.536347</td>\n",
       "      <td>1.378155</td>\n",
       "      <td>-0.338321</td>\n",
       "      <td>0.462388</td>\n",
       "      <td>0.239599</td>\n",
       "      <td>0.098698</td>\n",
       "      <td>0.363787</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.018307</td>\n",
       "      <td>0.277838</td>\n",
       "      <td>-0.110474</td>\n",
       "      <td>0.066928</td>\n",
       "      <td>0.128539</td>\n",
       "      <td>-0.189115</td>\n",
       "      <td>0.133558</td>\n",
       "      <td>-0.021053</td>\n",
       "      <td>149.62</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.191857</td>\n",
       "      <td>0.266151</td>\n",
       "      <td>0.166480</td>\n",
       "      <td>0.448154</td>\n",
       "      <td>0.060018</td>\n",
       "      <td>-0.082361</td>\n",
       "      <td>-0.078803</td>\n",
       "      <td>0.085102</td>\n",
       "      <td>-0.255425</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.225775</td>\n",
       "      <td>-0.638672</td>\n",
       "      <td>0.101288</td>\n",
       "      <td>-0.339846</td>\n",
       "      <td>0.167170</td>\n",
       "      <td>0.125895</td>\n",
       "      <td>-0.008983</td>\n",
       "      <td>0.014724</td>\n",
       "      <td>2.69</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>-1.358354</td>\n",
       "      <td>-1.340163</td>\n",
       "      <td>1.773209</td>\n",
       "      <td>0.379780</td>\n",
       "      <td>-0.503198</td>\n",
       "      <td>1.800499</td>\n",
       "      <td>0.791461</td>\n",
       "      <td>0.247676</td>\n",
       "      <td>-1.514654</td>\n",
       "      <td>...</td>\n",
       "      <td>0.247998</td>\n",
       "      <td>0.771679</td>\n",
       "      <td>0.909412</td>\n",
       "      <td>-0.689281</td>\n",
       "      <td>-0.327642</td>\n",
       "      <td>-0.139097</td>\n",
       "      <td>-0.055353</td>\n",
       "      <td>-0.059752</td>\n",
       "      <td>378.66</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.966272</td>\n",
       "      <td>-0.185226</td>\n",
       "      <td>1.792993</td>\n",
       "      <td>-0.863291</td>\n",
       "      <td>-0.010309</td>\n",
       "      <td>1.247203</td>\n",
       "      <td>0.237609</td>\n",
       "      <td>0.377436</td>\n",
       "      <td>-1.387024</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.108300</td>\n",
       "      <td>0.005274</td>\n",
       "      <td>-0.190321</td>\n",
       "      <td>-1.175575</td>\n",
       "      <td>0.647376</td>\n",
       "      <td>-0.221929</td>\n",
       "      <td>0.062723</td>\n",
       "      <td>0.061458</td>\n",
       "      <td>123.50</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2.0</td>\n",
       "      <td>-1.158233</td>\n",
       "      <td>0.877737</td>\n",
       "      <td>1.548718</td>\n",
       "      <td>0.403034</td>\n",
       "      <td>-0.407193</td>\n",
       "      <td>0.095921</td>\n",
       "      <td>0.592941</td>\n",
       "      <td>-0.270533</td>\n",
       "      <td>0.817739</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.009431</td>\n",
       "      <td>0.798278</td>\n",
       "      <td>-0.137458</td>\n",
       "      <td>0.141267</td>\n",
       "      <td>-0.206010</td>\n",
       "      <td>0.502292</td>\n",
       "      <td>0.219422</td>\n",
       "      <td>0.215153</td>\n",
       "      <td>69.99</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Time        V1        V2        V3        V4        V5        V6        V7  \\\n",
       "0   0.0 -1.359807 -0.072781  2.536347  1.378155 -0.338321  0.462388  0.239599   \n",
       "1   0.0  1.191857  0.266151  0.166480  0.448154  0.060018 -0.082361 -0.078803   \n",
       "2   1.0 -1.358354 -1.340163  1.773209  0.379780 -0.503198  1.800499  0.791461   \n",
       "3   1.0 -0.966272 -0.185226  1.792993 -0.863291 -0.010309  1.247203  0.237609   \n",
       "4   2.0 -1.158233  0.877737  1.548718  0.403034 -0.407193  0.095921  0.592941   \n",
       "\n",
       "         V8        V9  ...       V21       V22       V23       V24       V25  \\\n",
       "0  0.098698  0.363787  ... -0.018307  0.277838 -0.110474  0.066928  0.128539   \n",
       "1  0.085102 -0.255425  ... -0.225775 -0.638672  0.101288 -0.339846  0.167170   \n",
       "2  0.247676 -1.514654  ...  0.247998  0.771679  0.909412 -0.689281 -0.327642   \n",
       "3  0.377436 -1.387024  ... -0.108300  0.005274 -0.190321 -1.175575  0.647376   \n",
       "4 -0.270533  0.817739  ... -0.009431  0.798278 -0.137458  0.141267 -0.206010   \n",
       "\n",
       "        V26       V27       V28  Amount  Class  \n",
       "0 -0.189115  0.133558 -0.021053  149.62      0  \n",
       "1  0.125895 -0.008983  0.014724    2.69      0  \n",
       "2 -0.139097 -0.055353 -0.059752  378.66      0  \n",
       "3 -0.221929  0.062723  0.061458  123.50      0  \n",
       "4  0.502292  0.219422  0.215153   69.99      0  \n",
       "\n",
       "[5 rows x 31 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "target = 'Class'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df.loc[:, df.columns!=target]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "Y = df.loc[:, df.columns==target]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, \n",
    "                                                    test_size=0.33, \n",
    "                                                    random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0    284315\n",
      "1       492\n",
      "Name: Class, dtype: int64\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = sns.countplot(x=target, data=df)\n",
    "print(df[target].value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.1727485630620034"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "100* (492/float(df.shape[0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "99.82725143693798"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "100* (284315/float(df.shape[0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    190477\n",
       "1       343\n",
       "Name: Class, dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y_train[target].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf = LogisticRegression().fit(X_train, Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "Y_Test_Pred = clf.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "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>Actual</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Predicted</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>93800</td>\n",
       "      <td>78</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>38</td>\n",
       "      <td>71</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Actual         0   1\n",
       "Predicted           \n",
       "0          93800  78\n",
       "1             38  71"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(Y_Test_Pred, Y_test[target], rownames=['Predicted'], colnames=['Actual'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy =  0.9987657867577431\n",
      "Precision =  0.6513761467889908\n",
      "Recall =  0.47651006711409394\n",
      "F1 Score =  0.5503875968992248\n"
     ]
    }
   ],
   "source": [
    "generate_model_report(Y_test, Y_Test_Pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "generate_auc_roc_curve(clf, X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# class_weight='balanced'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"./class_weight_inp_image.png\">"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# https://scikit-learn.org/stable/modules/generated/sklearn.utils.class_weight.compute_class_weight.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0, 1]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unique_classes = list(df[target].unique())\n",
    "unique_classes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "out_dict = {}\n",
    "for classes in unique_classes:\n",
    "    out_dict[classes] = df.shape[0]/((df.loc[df[target] == classes].shape[0])\n",
    "                                     *len(unique_classes))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{0: 0.5008652375006595, 1: 289.4380081300813}"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "out_dict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(190820, 30) (190820, 1)\n"
     ]
    }
   ],
   "source": [
    "print (X_train.shape, Y_train.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf = LogisticRegression(class_weight='balanced').fit(X_train, Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.utils import class_weight"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([  0.50090037, 278.16326531])"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "class_weight.compute_class_weight('balanced', np.unique(Y_train), Y_train[target])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "Y_Test_Pred = clf.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "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>Actual</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Predicted</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>91548</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2290</td>\n",
       "      <td>136</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Actual         0    1\n",
       "Predicted            \n",
       "0          91548   13\n",
       "1           2290  136"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(Y_Test_Pred, Y_test[target], rownames=['Predicted'], colnames=['Actual'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy =  0.9754966112334684\n",
      "Precision =  0.05605935696619951\n",
      "Recall =  0.912751677852349\n",
      "F1 Score =  0.10563106796116506\n"
     ]
    }
   ],
   "source": [
    "generate_model_report(Y_test, Y_Test_Pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "generate_auc_roc_curve(clf, X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best parameters : {'class_weight': {0: 0.09736842105263158, 1: 0.9026315789473685}}\n"
     ]
    }
   ],
   "source": [
    "weights = np.linspace(0.05, 0.95, 20)\n",
    "gsc = GridSearchCV(\n",
    "    estimator=LogisticRegression(),\n",
    "    param_grid={\n",
    "        'class_weight': [{0: x, 1: 1.0-x} for x in weights]\n",
    "    },\n",
    "    scoring='f1',\n",
    "    cv=5\n",
    ")\n",
    "\n",
    "grid_result = gsc.fit(X_train, Y_train)\n",
    "print(\"Best parameters : %s\" % grid_result.best_params_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a2098efd0>"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_out = pd.DataFrame({'score': grid_result.cv_results_['mean_test_score'],\n",
    "                       'weight': weights })\n",
    "data_out.plot(x='weight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "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>score</th>\n",
       "      <th>weight</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.703298</td>\n",
       "      <td>0.050000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.770741</td>\n",
       "      <td>0.097368</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.763401</td>\n",
       "      <td>0.144737</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.753784</td>\n",
       "      <td>0.192105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.764725</td>\n",
       "      <td>0.239474</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.759793</td>\n",
       "      <td>0.286842</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.750428</td>\n",
       "      <td>0.334211</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.708962</td>\n",
       "      <td>0.381579</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.730055</td>\n",
       "      <td>0.428947</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.674311</td>\n",
       "      <td>0.476316</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.681596</td>\n",
       "      <td>0.523684</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.601097</td>\n",
       "      <td>0.571053</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.599889</td>\n",
       "      <td>0.618421</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.538431</td>\n",
       "      <td>0.665789</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.474159</td>\n",
       "      <td>0.713158</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.464639</td>\n",
       "      <td>0.760526</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.372763</td>\n",
       "      <td>0.807895</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>0.350569</td>\n",
       "      <td>0.855263</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>0.063391</td>\n",
       "      <td>0.902632</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>0.047618</td>\n",
       "      <td>0.950000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       score    weight\n",
       "0   0.703298  0.050000\n",
       "1   0.770741  0.097368\n",
       "2   0.763401  0.144737\n",
       "3   0.753784  0.192105\n",
       "4   0.764725  0.239474\n",
       "5   0.759793  0.286842\n",
       "6   0.750428  0.334211\n",
       "7   0.708962  0.381579\n",
       "8   0.730055  0.428947\n",
       "9   0.674311  0.476316\n",
       "10  0.681596  0.523684\n",
       "11  0.601097  0.571053\n",
       "12  0.599889  0.618421\n",
       "13  0.538431  0.665789\n",
       "14  0.474159  0.713158\n",
       "15  0.464639  0.760526\n",
       "16  0.372763  0.807895\n",
       "17  0.350569  0.855263\n",
       "18  0.063391  0.902632\n",
       "19  0.047618  0.950000"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_out"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf = LogisticRegression(**grid_result.best_params_).fit(X_train, Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "Y_Test_Pred = clf.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "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>Actual</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Predicted</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>93791</td>\n",
       "      <td>30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>47</td>\n",
       "      <td>119</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Actual         0    1\n",
       "Predicted            \n",
       "0          93791   30\n",
       "1             47  119"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(Y_Test_Pred, Y_test[target], rownames=['Predicted'], colnames=['Actual'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy =  0.9991807377616053\n",
      "Precision =  0.7168674698795181\n",
      "Recall =  0.7986577181208053\n",
      "F1 Score =  0.7555555555555555\n"
     ]
    }
   ],
   "source": [
    "generate_model_report(Y_test, Y_Test_Pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAl0AAAFuCAYAAAC7hK21AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJzt3XlYlPX+//HXLOIGpSAgaIaSB8l9aTtlx5TEFMLK8kRaZqHfo2nZctJKxFwS8+SWdsoSLfRYtnlUXDJPP9PSNk2NXNPUJBAQBVGBmfn9gU0iKIPCzQDPx3V1Xc3M577nPbyBefG+77k1ORwOhwAAAFChzJVdAAAAQE1A6AIAADAAoQsAAMAAhC4AAAADELoAAAAMQOgCAAAwAKELAADAAIQuAAAAAxC6AAAADEDoAgAAMAChCwAAwACELgAAAAMQugAAAAxgrewCznf8+CnZ7Y4K27+Pj6cyMnIqbP+4PPTF/dAT90Rf3A89cU8V3Rez2aSGDeuXeTu3Cl12u6NCQ9cfzwH3Q1/cDz1xT/TF/dAT9+SOfeHwIgAAgAEIXQAAAAYgdAEAABiA0AUAAGAAQhcAAIABCF0AAAAGIHQBAAAYgNAFAABggFJDV3x8vLp3766QkBDt2bOnxDU2m03jx49XWFiY7rzzTi1durTcCwUAAKjKSg1dPXr00KJFi9SkSZOLrlm+fLkOHTqktWvX6v3339fs2bN15MiRci0UAACgKis1dHXp0kUBAQGXXJOUlKT7779fZrNZ3t7eCgsL0+rVq8utSAAAgKquXP7txZSUFAUGBjpvBwQE6Pfffy+PXQMAUK6+2PabtvyUWi77quVhUX6erVz2hfLT+7bmanttw8ouoxi3+gevfXw8K/w5fH29Kvw5UHb0xf3QE/dEX67cD3vTdeRYjpo3ubpc9lfLw1Iu+0H5cseflXIJXQEBATp69KjatWsnqfjky1UZGTkV+q+C+/p66dix7ArbPy4PfXE/9MQ90ZfLc+Fk61Bajpr5eerp+9tf8b7piXuq6L6YzabLGhSVyyUjevXqpaVLl8putyszM1Pr1q1TeHh4eewaAIArsuWnVB1Ky3HebubnqZta+1diRaipSp10TZw4UWvXrlV6eroeffRRNWjQQCtXrlRMTIxGjhyptm3bKioqSj/++KN69uwpSRo+fLiuueaaCi8eAICSnD/d+mOy9fxDnSq5KtR0JofDUXHH88qIw4s1E31xP/TEPdEX18Uv+sEZtiTpptb+6tbh4pc+ulz0xD256+FFtzqRHgBw5crz03lVFdMtuCP+GSAAqGYuPIepJuK8LbgjJl0AyhVTlorj6jWhmPIA7olJF4ByxZSl8jHlAdwTky6gkrnjZOhKrrLNlKXicNI2ULUx6QIqWXWbDDFlAYCSMekCytHlTK3ccTLERAUAyh+TLqAcXc7UiskQANQMTLpQYxhx7pQ7Tq0AAO6BSRdqDCPOnWJqBQC4GCZduCzlOTW6kk/KlQVTKABAZWLShctSFT9xxxQKAFCZmHRVQ1Xt3CU+KQcAqAmYdFVDnLsEAID7YdJVRV1qmsW5SwAAuB8mXVXUpaZZTKEAAHA/TLoqUEWeW8U0CwCAqoVJVwWqyHOrmGYBAFC1MOkqg7JOrphGAQCAPzDpKoOyTq6YRgEAgD8w6XLBHxMuJlcAAOByMelywfmBi8kVAAC4HEy6SvHFtt+0+3CWQq5pwIQLAABcNiZdpfjjxHkmXAAA4Eow6SrB+Z9SPJSWo5BrGqhbhyaVXBUAAKjKmHSV4PxPKXIeFwAAKA9Mui6CTykCAIDyxKQLAADAAEy6znPh9bgAAADKC5Ou83A9LgAAUFGYdJ3D9bgAAEBFYtJ1DtfjAgAAFanGTrrOvxaXxPW4AABAxaqxk67zr8UlcT0uAABQsWrkpIvztwAAgNFq5KSL87cAAIDRalzoOn/KxflbAADAKDUudDHlAgAAlaHGhS5JTLkAAIDhamToAgAAMBqhCwAAwACELgAAAAPUqNC1+uuD2n04q7LLAAAANVCNCl3/b+sRSXxyEQAAGK9GhS6JTy4CAIDKUeNCFwAAQGUgdAEAABiA0AUAAGAAQhcAAIABCF0AAAAGsLqy6MCBAxo9erSysrLUoEEDxcfHKygoqMiajIwMjRkzRikpKcrPz9fNN9+sl156SVarS08BAABQrbk06Ro3bpyio6O1Zs0aRUdHKzY2ttiaf//73woODtby5cu1fPly/fTTT1q7dm25FwwAAFAVlRq6MjIylJycrIiICElSRESEkpOTlZmZWWSdyWTSqVOnZLfblZeXp/z8fPn7cxFSAAAAyYXQlZKSIn9/f1ksFkmSxWKRn5+fUlJSiqwbNmyYDhw4oNtuu835X+fOnSumagAAgCqm3E64Wr16tUJCQrRw4UKdOnVKMTExWr16tXr16uXyPnx8PMurnIuq5WGRr69XhT8PyoaeuB964p7oi/uhJ+7JHftSaugKCAhQamqqbDabLBaLbDab0tLSFBAQUGRdYmKiJk+eLLPZLC8vL3Xv3l1btmwpU+jKyMiR3e4o+6sog/w8m44dy67Q50DZ+Pp60RM3Q0/cE31xP/TEPVV0X8xm02UNiko9vOjj46PQ0FCtWLFCkrRixQqFhobK29u7yLqmTZtqw4YNkqS8vDx9/fXXatmyZZkLAgAAqI5c+vRiXFycEhMTFR4ersTERI0fP16SFBMTox07dkiSXnjhBX3//feKjIxU3759FRQUpAceeKDiKgcAAKhCXDqnKzg4WEuXLi12/7x585z/36xZMyUkJJRfZQAAANUIV6QHAAAwAKELAADAAIQuAAAAAxC6AAAADEDoAgAAMAChCwAAwACELgAAAAMQugAAAAxA6AIAADAAoQsAAMAAhC4AAAADELoAAAAMQOgCAAAwAKELAADAAIQuAAAAAxC6AAAADEDoAgAAMAChCwAAwACELgAAAAMQugAAAAxA6AIAADAAoQsAAMAAhC4AAAADELoAAAAMQOgCAAAwAKELAADAAIQuAAAAAxC6AAAADEDoAgAAMAChCwAAwACELgAAAAMQugAAAAxA6AIAADAAoQsAAMAAhC4AAAADELoAAAAMQOgCAAAwAKELAADAAIQuAAAAAxC6AAAADEDoAgAAMAChCwAAwACELgAAAAMQugAAAAxA6AIAADAAoQsAAMAAhC4AAAADELoAAAAMQOgCAAAwAKELAADAAIQuAAAAA7gUug4cOKD+/fsrPDxc/fv318GDB0tcl5SUpMjISEVERCgyMlLp6enlWSsAAECVZXVl0bhx4xQdHa2oqCgtW7ZMsbGxevfdd4us2bFjh15//XUtXLhQvr6+ys7OloeHR4UUDQAAUNWUOunKyMhQcnKyIiIiJEkRERFKTk5WZmZmkXULFizQ4MGD5evrK0ny8vJS7dq1K6BkAACAqqfU0JWSkiJ/f39ZLBZJksVikZ+fn1JSUoqs279/vw4fPqyHHnpI99xzj+bOnSuHw1ExVQMAAFQxLh1edIXNZtPu3buVkJCgvLw8Pf744woMDFTfvn1d3oePj2d5lXNRtTws8vX1qvDnQdnQE/dDT9wTfXE/9MQ9uWNfSg1dAQEBSk1Nlc1mk8Vikc1mU1pamgICAoqsCwwMVK9eveTh4SEPDw/16NFD27dvL1PoysjIkd1esdOx/Dybjh3LrtDnQNn4+nrREzdDT9wTfXE/9MQ9VXRfzGbTZQ2KSj286OPjo9DQUK1YsUKStGLFCoWGhsrb27vIuoiICG3cuFEOh0P5+fnavHmzWrVqVeaCAAAAqiOXLhkRFxenxMREhYeHKzExUePHj5ckxcTEaMeOHZKkPn36yMfHR71791bfvn113XXXqV+/fhVXOQAAQBVicrjR2e4VfXjxtaU/Kj/Ppucf6lRhz4GyYzzvfuiJe6Iv7oeeuKcqe3gRAAAAV47QBQAAYABCFwAAgAEIXQAAAAYgdAEAABiA0AUAAGAAQhcAAIABCF0AAAAGIHQBAAAYgNAFAABgAEIXAACAAQhdAAAABiB0AQAAGIDQBQAAYABCFwAAgAEIXQAAAAYgdAEAABiA0AUAAGAAQhcAAIABCF0AAAAGIHQBAAAYgNAFAABgAEIXAACAAQhdAAAABiB0AQAAGIDQBQAAYABCFwAAgAEIXQAAAAYgdAEAABiA0AUAAGCAGhO6vtj2m3buz6jsMgAAQA1VY0LXlp9SJUk3tfav5EoAAEBNVGNClyS1CfZRtw5NKrsMAABQA9Wo0AUAAFBZCF0AAAAGIHQBAAAYgNAFAABgAEIXAACAAQhdAAAABiB0AQAAGIDQBQAAYABCFwAAgAEIXQAAAAYgdAEAABiA0AUAAGAAQhcAAIABCF0AAAAGIHQBAAAYgNAFAABgAEIXAACAAVwKXQcOHFD//v0VHh6u/v376+DBgxdd+8svv6h9+/aKj48vrxoBAACqPJdC17hx4xQdHa01a9YoOjpasbGxJa6z2WwaN26cwsLCyrVIAACAqq7U0JWRkaHk5GRFRERIkiIiIpScnKzMzMxia9966y1169ZNQUFB5V4oAABAVVZq6EpJSZG/v78sFoskyWKxyM/PTykpKUXW7dq1Sxs3btSgQYMqpFAAAICqzFoeO8nPz9fYsWP1yiuvOMPZ5fDx8SyPckpUy6OwLl9frwp7Dlw++uJ+6Il7oi/uh564J3fsS6mhKyAgQKmpqbLZbLJYLLLZbEpLS1NAQIBzzbFjx3To0CENGTJEknTy5Ek5HA7l5ORowoQJLheTkZEju91xGS+jdPl5NtXysOjYsewK2T8un6+vF31xM/TEPdEX90NP3FNF98VsNl3WoKjU0OXj46PQ0FCtWLFCUVFRWrFihUJDQ+Xt7e1cExgYqC1btjhvz549W7m5uXr++efLXBAAAEB15NKnF+Pi4pSYmKjw8HAlJiZq/PjxkqSYmBjt2LGjQgsEAACoDlw6pys4OFhLly4tdv+8efNKXD9ixIgrqwoAAKCa4Yr0AAAABiB0AQAAGIDQBQAAYABCFwAAgAEIXQAAAAYgdAEAABiA0AUAAGAAQhcAAIABCF0AAAAGIHQBAAAYgNAFAABgAEIXAACAAQhdAAAABiB0AQAAGIDQBQAAYABCFwAAgAEIXQAAAAYgdAEAABiA0AUAAGAAQhcAAIABCF0AAAAGIHQBAAAYgNAFAABgAEIXAACAAQhdAAAABiB0AQAAGIDQBQAAYABCFwAAgAEIXQAAAAYgdAEAABiA0AUAAGAAQhcAAIABCF0AAAAGIHQBAAAYgNAFAABgAEIXAACAAQhdAAAABiB0AQAAGIDQBQAAYABCFwAAgAEIXQAAAAYgdAEAABiA0AUAAGAAQhcAAIABCF0AAAAGIHQBAAAYgNAFAABgAEIXAACAAQhdAAAABiB0AQAAGMDqyqIDBw5o9OjRysrKUoMGDRQfH6+goKAia+bMmaOkpCRZLBZZrVaNGjVKXbt2rYiaAQAAqhyXQte4ceMUHR2tqKgoLVu2TLGxsXr33XeLrGnXrp0GDx6sunXrateuXRowYIA2btyoOnXqVEjhAAAAVUmphxczMjKUnJysiIgISVJERISSk5OVmZlZZF3Xrl1Vt25dSVJISIgcDoeysrIqoGQAAICqp9TQlZKSIn9/f1ksFkmSxWKRn5+fUlJSLrrNp59+qmbNmqlx48blVykAAEAV5tLhxbL45ptvNHPmTM2fP7/M2/r4eJZ3OU61PApDo6+vV4U9By4ffXE/9MQ90Rf3Q0/ckzv2pdTQFRAQoNTUVNlsNlksFtlsNqWlpSkgIKDY2q1bt+q5557T3Llz1aJFizIXk5GRI7vdUebtXJGfZ1MtD4uOHcuukP3j8vn6etEXN0NP3BN9cT/0xD1VdF/MZtNlDYpKPbzo4+Oj0NBQrVixQpK0YsUKhYaGytvbu8i67du3a9SoUZo1a5Zat25d5kIAAACqM5eu0xUXF6fExESFh4crMTFR48ePlyTFxMRox44dkqTx48frzJkzio2NVVRUlKKiorR79+6KqxwAAKAKcemcruDgYC1durTY/fPmzXP+/0cffVR+VQEAAFQzXJEeAADAAIQuAAAAAxC6AAAADEDoAgAAMAChCwAAwACELgAAAAMQugAAAAxA6AIAADAAoQsAAMAAhC4AAAADELoAAAAMQOgCAAAwAKELAADAAIQuAAAAAxC6AAAADEDoAgAAMAChCwAAwACELgAAAAMQugAAAAxA6AIAADAAoQsAAMAAhC4AAAADELoAAAAMQOgCAAAwAKELAADAAIQuAAAAAxC6AAAADEDoAgAAMAChCwAAwACELgAAAAMQugAAAAxA6AIAADAAoQsAAMAAhC4AAAADELoAAAAMQOgCAAAwAKELAADAAIQuAAAAAxC6AAAADEDoAgAAMAChCwAAwACELgAAAAMQugAAAAxA6AIAADAAoQsAAMAA1souALgSNluBjh8/poKCvMoupVpJSzPLbrdXdhm4AH1xP/TEPZVXX8xmi+rW9ZSn59UymUxXvD9CF6q048ePqU6deqpfv3G5/ECgkNVqVkEBbyTuhr64H3rinsqjLw6HQzZbgbKzs3T8+DF5e/tdcV0cXkSVVlCQp/r1ryJwAQDKlclkktVaSw0a+Cgv70y57JPQhSqPwAUAqCgmk1mSo1z2xeFFoJydPHlSffv2UlTUfXryyWec97/zzps6ffq0nnjiKed9H330vnbt+lkvvhgnSTp06Fe98cZs7du3V3Xq1FbduvU0YMAg3X57tyLPkZJyVH//+z1q3jxYDoddVmstPfXUs2rbtr1zzYYNX2jBgnk6c+aMCgoK1LVrNw0dOlweHh6SpIKCAi1Y8LbWrVsrq9Uqh8Oum2++Vf/4xwhZrR5Fns/hcGjp0iX6738/keSQ3W5X+/YdNWzYk/Ly8irfL2A5efbZkRo16p9q0qSpkpKWq02bdmrW7FpJUlLScn311ZeaOHGqS/uy2Wy6774IhYZer1de+VdFli1JmjQpTq1aheq++/o773v99RmqX7+eHn10SJn2VdL33eX44YfvVFBQoBtvvFlS4ffg448P1MqVn1/RfquaQ4d+1aRJcTpx4oSuvvpqjRs3QYGBTYuty8hI16uvTlZKylEVFBTo4YcHKzy8t/Pxzz//TAsXvi2HwyGTyaQZM+bK29tHEybEav/+fc51+/fv1SuvTNNtt/3tvBoO6tFHH9I999zv7OuhQ79q6tRJysnJUX5+nrp3v1OPPTZUkrRmTZIWL35XBw8e0MiRTxf5vjp06KCmTZuiEyeyJElPPPGUbrjh5iKv5YcfvtNTTw3Tk08+49x28eL3tHz5Jzpy5LCmTHlNt97a1bl+584dmj37NZ05c1q1annouedeUEhIK0mS3W7XO++8qc8//0y1a3vIz89fr746s9R9Xmq7kh6bPn12kdewatUKTZoUp/j46c799usXKQ8PD3l41JYk/eMfI3TTTbdcrPXlgtAFlLPPPlul1q3bat26NRo2bKRq1arl0nbp6el64okhGjZspF55ZZrzvm+/3Vziek9PTy1YsFiS9MknH2rKlAlatOhDSdK2bT9o+vSpmjZtloKDr9PZs2c1aVKcXnstXqNHj5UkTZ48XmfPntH8+e+pXr36Kigo0MqV/1VeXp7q1CkauubNe0Pbtv2gWbPekLe3j+x2u7788gudPHmiTKHLZrPJYrG4vP5KTJs2y/n/SUnLdfXVDZyhq6w2b/5KjRr56scftykzM0Pe3j4lrjPy9VWkgoICWa1F3x62bv1ep0+fdoYud1FSrRVp2rRXdO+99ys8vLfWrElSfPxEzZz572LrZs+erlatrteUKa/p+PHjeuyxAerQoZP8/Rtr165kJSS8pZkz35CPTyPl5OQ4f0+MHfuycx979+7Rk0/+Qzfe+GcQsNlsmjp1srp27Vbk+d54Y5buuKOH7ruvv3JzczVw4AO65ZZbdf31bdSy5V8UFzdZiYkLitU5efLL6tv3PvXq1UeHDx/SyJH/p//852PVqVNHkpSbe0pvvDFbN9/81yLbdezYSbff3k1Tpkwocr/D4dBLL/1TcXGT1KFDJ/344za9/PJLSkxcKpPJpA8+WKxDh35VYuIHslqtyszMKHWfki653aUek6S0tFQtW/axWrduW2y/EyfGq0WL64rdX1Fc+k49cOCARo8eraysLDVo0EDx8fEKCgoqssZms2nixIn68ssvZTKZNGTIEN1///0VUTPg1lau/K+GDXtS7723QBs3/j/dcUeYS9t9/PEH6tixs3r16uO8r1GjRrrrrohSt+3YsbPmzJnhvD1//lt65JHBCg4u/GVSu3ZtPfvsaN13X4QGDXpc+fn52rDhf/r44yTVq1dfkmS1WhUVdW+xfefm5mrJkkVasGCRM2yYzWb97W/dJRWfGp1/OylpudatW6uGDRvowIEDevLJZzRjxqtKSFjs3P/gwQM0YsQodezYWatWrdDHHy+VzWaTp6ennn12tJo1CypSz5YtX+vDD5fo1Vdn6vjxTEVG9tTLL09R9+5hWrRooXJycjR06HD16xepqVOn6+efk7V798+aMWOa5s17Q8OHPylJOnXqlGJjx+iXX/bLy8tTEydOlY9PoxK/vitX/ld9+96rnTt3aPXqJEVHD3S+1vNf35gxY9WwoY9mzJiq1NTfdfbsWYWFhevhhwdLKpxWbdv2g/Lz89WgQQONGROrxo0DSu3vhd55500dOvSrTp3K0dGjv6lJk6aaMCFederUUU5OjqZMeVkHDx6Qn19jNWzYQA0bFvYtPz9fb701V9u2fa/8/AIFBwfrmWfGqF69epo0KU716tXT4cOHlZV1XPPnJzqfb//+fVq27GPZ7XZ999036tGjp8LCekqS3nxzjjZv3qQzZ85o9OhYtW/fQZL09dcb9e6783X2bJ5q1aqlESOeVps2xd/0jh1L04wZr+rIkcOSpLCwcA0c+KieeGKIHnxwoHMqcf7tJ54YorZt2ys5eac8PDzk7e2j4OCWeuCBByVJv/yyT88//4w++OBT5eae0uzZ07V//17l5eWpY8cuGjFi1GWF4+PHM7Vnzy5Nnz7HWev06VN1/PhxNWzYsMjaffv2qn//aElSw4YN1bLlX7R+/To9+OAAvf/+Yv397wOc32+enp4lPt/KlcvUs2cv53RakhITF+ivf+2q06dzdfr06fNWm5STkyNJOnv2jEwmkxo29JYkZ6gwm4ufUbRv3x7ddFNhoLrmmma66qqrtHnzJnXr1kNSYXiMjh6or77aWGS70NDWJdaclZWlnJxsdejQSZLUvn0HHTt2TLt371KrVqFasmSR5syZ5wzK5/8Bc7F9Srrkdpd6TJKmTp2kESOe1htvzFJlc+mcrnHjxik6Olpr1qxRdHS0YmNji61Zvny5Dh06pLVr1+r999/X7NmzdeTIkXIvGHBne/fu0cmTJ9W58w3q0ydSK1f+1+Vt9+zZpdat21zW827Y8D/16NHTeXv//r3F/qq76qqrFRjYVPv379OePbvVtGnhL9jSHDz4izw8ahULP67asWObBg8eqvnzE9W+fUedPn1a+/btlVT45vjHL+gff9yq9es/05w58zR/fqIefHCgXnnl5WL7a9++o376aacKCgr03XffqE2bdvr++28kSd9//606d76hyPo+fe5WSEionnrqWS1YsFg33HCTJOnnn5M1fPiTSkz8QEFBLfThh++XWP/x48f1ww/fqXv3O9W7d6SSkor29PzX17JliCZOjFW/fn/XvHnv6p13ErV581fOaeWAAYP09tvvauHC/ygsLPyK3gR27/5Z48ZN0qJFH6qgoEBr166SJCUkzFO9evWVmLhUsbETtHXrD85tFi1aqPr162vevHe1YMFi+fj46r33EpyP79y5Q5MmTS0SuCQpOPg6RUXdq169+mjBgsUaOHCQJOnEiRNq06adEhIW69FHY/Tvfxe+nt9+O6IFC97RtGmzNH9+op5//iXFxo4u8XW8/PJYtW7dVgsXLtHChUsUGXmPS6//l1/26V//mq1XX52p3r0jtXr1CudjK1cuV+/eETKZTJo9e7o6dOikefPeVULCYh0/nlniz2V2drYGDYou8b/x41+SJKWmpqpRIz9nYLNYLGrUyFdpaanF9hcS0krr1q2Vw+HQ0aO/aefO7UpNTZFU+DN19OhvGj48RoMHP6QFCwoPM54vPz9fn322Wn363O28b9++vfrmm83OMHe+J598RuvXf6a+fe9Sv36RevDBgQoICCz16xgSEqrPPlstSdq162cdOvSrfv+9sM6vv96k7Oxsl/9wlAoD5tVXN9CXX34hSdq4cYNyc08pNTVFp07l6MSJLK1fv04xMY9o6NBHnesu5VLblbbPTz75UM2bB1/0d+v48WP1yCN/17RpU5Sdne3y67xcpU66MjIylJycrISEwh/MiIgITZgwQZmZmfL29nauS0pK0v333y+z2Sxvb2+FhYVp9erVevzxxyuueuA8m3akaOP2lArZ923tAnRr29InEitXLlOvXn1kMpn0t7/doenTX9WxY2ny9fW76An/f9zvKON5mjk5ORo0KFpZWcdls9n05pt/vnmWvi/Xn6ysdV2obdsOatLkz3NeevXqo1WrlmvEiKfPvTlGymQyadOmDdq3b6+GDBl07nkdys4+WWx/derUUfPmLfTTTzv13XffaNCgxzV37kzl5+dr166fi5zXdint2rWXv39jSVLr1m307bdbSly3Zs1K3XprV9WrV1/t2nVQQYFNO3fucE5tzn99p0+f1tat3ysrK8u5fW7uKR08eFA33HCzNm/epI8/XqrTp3Nls9kuWltp3yuSdOONNzsP7V5/fRv99lvhH7lbt36np556TpLUoEED50RSkjZt2qBTp07piy/WS5Ly8/N03XUtnY9369ZDdevWvWhdF6pbt55zEtW6dVu9/nrhtHXLlq/1229HNHz4n+ef2Wy2Yodmc3NztXPndufk6I+aXXHnnb2ck4327TsqNzdX+/btVVBQc61bt8b587Bx4wb9/PNPWrJkkSTpzJkz8vPzL7YHJlrHAAASFklEQVQ/Ly8v5+H68vDEE6M0e/ZrGjQoWv7+jdWp0w2yWArrtdls2r9/r6ZPn6OCgnw988wI+fs3LjLV3rDhC/n7N1bLliGSCg+jxsdP1AsvjCtxSrds2ccKD++t6OiHlZ6erhEjhqhVq+tL/UPuhRfGafbs15SUtFxBQc3Vrl0HWa1WZWdn69//fl0zZsy55PYlmTx5mubOnaWEhHm6/vq2CgpqIYvFqoKCAuXn58vhsGvevIU6cuSwhg17XC1aXFfkd8SFLrWdp6dniY+1bNlSdru0fPkneuONd0rc75w58+Tv31h5eXmaNetfmj59qmJjix/aLE+lhq6UlBT5+/sXSfZ+fn5KSUkpErpSUlIUGPhnqg4ICNDvv/9epmJ8fEoesZaH3rc1lyT5+rrnSb813eX2JS3NLKu1cGBrsZhUUR9ktFhMzue5mMK/TNfIw8NDq1evlFR48dY1a1Zq0KDH5O3trdTUlCL7OXnyhLy9vWW1mhUaGqpdu5JLfZ7Ceszy9PRSYuIS2Ww2zZjxL8XFvaj589+VJLVs2VI///yTQkNDnducOHFCR48eUcuWLVVQUKAjRw4pNzfnotOuP+po2fI65eXl6ejRwyWeE1WrllUOh8O5vqAg/9xHrc0ym02qX79ekdcUERGpxx57RMOGjdDnn6/RvHkLZLWaZTJJkZFRGjLkH6W+/htuuFFbt36r5OSdGj36RS1Z4qPPP1+jli1bqn79P0ODxWI+t29TkR6azSbVrl3bedtqtcput5f4tV+1aoWOHz+ufv0iJRX+Zb1q1X/VoUP7Yq/PbC4MRgsWvCertei5fCkpRzV79nQlJLynwMAm2r79R8XGvlDiczZs2FDZ2SeLfa9cc801zq9rnTp1zqvfovz8vPN+FsxFajKb/3zt//znGHXpcmOx5zSZTPL0rHfR7z+z2VRkPxaLWR4eHs7btWpZZbPZztUn3XLLXzVu3KXfwP6s31zsea1Wq0ymP9fk5+c5e1hYa/0i2/TuHaE1a1aqU6fOat68uZo2bXLuEYemTn3tkm/qUuGka9iwmBIfa948WC+/PEmBgQFKT0+TyeSQxWKRzWZTevoxBQYGFKvf19dHL788yXl71KgRatGihaxWswICAtSjx52qV6+OpDq6/fZu2r07WZGRf061Vq36ryIjo5z7TU/P0NGjv+m55woPj+fkZMvhkE6fPqUxY8bqww+X6KOPlstqNatxYz916XKjduzYqvbt2zn3aTIV7aEkXXttM02b9uepCX//+31q0SJYv/76izIz0xUT84gk6cSJLG3a9KVycrL12GNDiuzzwt+PrVtfrzlzCs9zy8/PV+/eYQoObiEfH2/Vq1dPvXv3kdVqVlDQtQoJaaX9+/fo2mubXXSfl9que/ewEh/bs2e38vPzlZ6eroceKjzVKTMzQ1OmTNCwYU8oMrKvmjQpzCxWax316/eAnntu1CW+/83lkh/c6kT6jIwc2e3l87HMC7W9tqF8fb107FjFjw9RNlfSF7vd7rwA3s3XN9bN1zcuz9KKKO1Ce+vXr1ezZtcW+atq587tmjhxnAYMeFQdOnRWQsLbOno0RX5+/jp58oTWrVurUaP+qYICu/r27adHH31ISUlJ6tmzlyQpPf2Yvvpqo+6+u+ghF5vNLslxriaT/u//RujBB+/V//63Xl27dtMjjzyuCRNi1bp1O+eJ9PHxk3XHHWHy8yv8Gt166+165ZWJGjNmrOrVqy+bzaaPPvpAERFRuuoqT+fr9fCoowceiNbkyRM0YcIUNWzoLYfDoc8+W63WrdsqIKCJ9u3bq9zcwvNIPv98nby8Cre32x1yOBxFvnaNGvkrKKi5pk2bqmuvbSFf38YqKLDrllu6auLEcYqI6Cs/P3/ZbDbt3btHrVqF6kIdO96gCRPGqlmzIJlMFnXufIPefvtNRUb2LfJcNlvh90e9evV08mS287EL6yqpTklKTt6p7OxsLVu22jllOnYsTQMH9teIEc8U26527bpq376jFixI0KBBhVP+1NTfZbVadfJkjqxWq66+uqHy8gr00UdLL/p91aXLjZo58zXde29/XXXVVUpN/V1btnylhx8e5Py62u1F6//jdqdON2j58mVq3bqdTpzI0hdf/E933BGmggK7br31di1enKjQ0DaqXbuOcnNPKS0tTUFBzeVwFN3nherWrae0tDTn40W/B4ve7tz5Jr399lvas2evWrQIlqRzfwQUPWfHw6OO2rRpp8WLExUd/bAkOc8dDgxsop9+2qm//rWrDhz4RXv37pHNVrj/wotWFq21Z88+Gjp0kA4fPqS77op0PnbrrbdrwYIEPfvsaFksFmVlZSk395QCA5sUqaVu3fpFzjW8UEGBXVdd1UDXXfcXrV696tyJ9Kv0l7+0kpfX1cW+bidOZKl+fU9ZrVZ9//232r9/nyZMiFdBgV09evTS5s2bdOedd8lms+nbb79Rt249nPtIS0vVtm1bFRs70Xlfo0b+WrlynXP/538qtaDAroCAQG3atFF33RWh3NxT2rbtB/31r12L1FVSj48fz1SDBg1lMpmUlLRctWrVUseOXWQymbR8+WfOded/ovbCfV7Yi4yMdOf5agkJ76hDh04KCGh67rWHa9OmTYqKulfHj2dq7949uvbaFqXu81LblfRYixbBatr0WvXoEe7cx/nnBWZnn3KeO+pwOLRmzWpdd91fLvr9b7fbi7xPmc2myxoUlRq6AgIClJqa6vxUjs1mU1pamgICAoqtO3r0qNq1K0zVF06+gOouKWm5eva8q8h9bdq0k91u17ZtP6hDh04aOfJpjR79jOz2wjeO++7r75w6NGrkq9dff0tvvDFL8+a9obp165y7ZMQjpT537dq1FRPzDyUkzFPXrt3UqVMXPfXUs5o4MVZnz55Vfn6Buna9XUOHDndu89JL4zV//lsaPHigc1p18823Fjlp9w9Dhw7X++8v0ogRhR9Bdzgcateuo2655TY1adJUXbrcqIcf7q+AgEAFBQUpIyP9kvX27h2pCRNii3xSq0OHThoyZJhGj376XFjK1x13hJUYulq3bqMTJ7LUpUvh+VudO9+gN9+cU+x8rj/cffe9mjNnhv7zn/c0bNiTpX49/5CUtFxhYeFFDuv5+vrpL38J0f/+t67EbWJjJ2jWrNf08MOFH62vV6++xoyJVXDwdbrjjjANGNBf/v7+6tixs378cWuJ+7jhhpvVt++9GjFi6LnphFlPPfWcgoKalxr+Bw16XK+8Ml4DBtyvxo0DinzacMCAQXrnnTf1+OMPnzup2qTBg2MUFNS81K/F7bffoRdffE6DBkUXOZG+JNdc00yxsRM0ZcoEnT17VgUF+Wrbtn2JJ0rHxk7Qa6/Fa+DAB2Q2W3TnneEaMGCQHnroEY0dO1qbN3+l4ODrnIfZLqZx48YKCmqhrVu/V1zcZOf9Tz75jObOnaVBgx6UyWRSrVoeGjnymWKhy1XPPfeCJk4cp4SEt+Xl5aW4uD+nec8+O1KPP/5/atXqeiUn/6SZM6fJbDbr6qsbKD7+NecnAsPCemr37mQNGHC/TCazbrrpZkVERDn3s2rVCt16a1ddddXVLtf1wgtxmjHjVS1Zskg2W4F69OipW265VZL02WerNXfuLGVnn9SXX/4/JSYu1Guvva7mzVto48YNWrRooSSTmjRpqsmTp7l07cPFi9/V0qVLlJV1XJMnx8nDo7YSEz9Q/fqeWrbsY3322WrZ7Xa1ahWqMWP+PA986NDhmjx5vD78cIlMJpOGDh2ua68NKnWfl9qupMdK+1nJzMzQSy/9U3a7XTabXUFBzfXMMyWfd1ieTI4Lz94rwcCBA9WvXz9FRUVp2bJl+vDDD/Xee+8VWfPxxx9r5cqVmjdvnrKystS3b18tWrRI11xzjcvFVOSkS7qyiQoqzpX05ffff1Xjxpd3GQBcHP+0iXuiL+6Hnrin8u7Lhe81lzvpcunTi3FxcUpMTFR4eLgSExM1fvx4SVJMTIx27NghSYqKilLTpk3Vs2dPPfDAAxo+fHiZAhcAAEB15tKkyyhMumomJl3uh7/e3RN9cT/0xD1V6UkXAAAArgyhCwAAwACELlR5bnSEHABQzTgchZfmKQ+ELlRpVquHTp06SfACAJSrwmvw5SsrK10eHnXKZZ9udXFUoKwaNvTV8ePHlJOTVfpiuMxsNstu5+Rgd0Nf3A89cU/l1Rez2aK6dT3l6en6NdMuhdCFKs1isapRo9L/TUSUDZ/0dU/0xf3QE/fkrn3h8CIAAIABCF0AAAAGcKvDi2Zz+Xw6oLKfA2VHX9wPPXFP9MX90BP3VJF9udx9u9UV6QEAAKorDi8CAAAYgNAFAABgAEIXAACAAQhdAAAABiB0AQAAGIDQBQAAYABCFwAAgAEIXQAAAAYgdAEAABigWoauAwcOqH///goPD1f//v118ODBYmtsNpvGjx+vsLAw3XnnnVq6dKnxhdYgrvRkzpw56tOnj+6++27de++9+vLLL40vtAZxpSd/+OWXX9S+fXvFx8cbV2AN5WpfkpKSFBkZqYiICEVGRio9Pd3YQmsQV3qSkZGhIUOGKDIyUr169VJcXJwKCgqML7aGiI+PV/fu3RUSEqI9e/aUuMYt3+cd1dDAgQMdn376qcPhcDg+/fRTx8CBA4ut+eSTTxyDBw922Gw2R0ZGhqNr166Ow4cPG11qjeFKTzZs2ODIzc11OBwOx88//+zo3Lmz4/Tp04bWWZO40hOHw+EoKChwDBgwwPH00087pkyZYmSJNZIrfdm+fbvjrrvucqSlpTkcDofj5MmTjjNnzhhaZ03iSk8mTpzo/PnIy8tz9OvXz7Fy5UpD66xJvv32W8fRo0cdd9xxh2P37t0lrnHH9/lqN+nKyMhQcnKyIiIiJEkRERFKTk5WZmZmkXVJSUm6//77ZTab5e3trbCwMK1evboySq72XO1J165dVbduXUlSSEiIHA6HsrKyDK+3JnC1J5L01ltvqVu3bgoKCjK4yprH1b4sWLBAgwcPlq+vryTJy8tLtWvXNrzemsDVnphMJp06dUp2u115eXnKz8+Xv79/ZZRcI3Tp0kUBAQGXXOOO7/PVLnSlpKTI399fFotFkmSxWOTn56eUlJRi6wIDA523AwIC9Pvvvxtaa03hak/O9+mnn6pZs2Zq3LixUWXWKK72ZNeuXdq4caMGDRpUCVXWPK72Zf/+/Tp8+LAeeugh3XPPPZo7d64cDkdllFztudqTYcOG6cCBA7rtttuc/3Xu3LkySsY57vg+X+1CF6q+b775RjNnztS//vWvyi6lRsvPz9fYsWM1fvx45xsO3IPNZtPu3buVkJCg9957Txs2bNCyZcsqu6wabfXq1QoJCdHGjRu1YcMGfffdd5U+VYH7qXahKyAgQKmpqbLZbJIKfzmlpaUVG0MGBATo6NGjztspKSlMVSqIqz2RpK1bt+q5557TnDlz1KJFC6NLrTFc6cmxY8d06NAhDRkyRN27d9fChQv1wQcfaOzYsZVVdrXn6s9KYGCgevXqJQ8PD3l6eqpHjx7avn17ZZRc7bnak8TERN19990ym83y8vJS9+7dtWXLlsooGee44/t8tQtdPj4+Cg0N1YoVKyRJK1asUGhoqLy9vYus69Wrl5YuXSq73a7MzEytW7dO4eHhlVFytedqT7Zv365Ro0Zp1qxZat26dWWUWmO40pPAwEBt2bJF69ev1/r16/XII4/ogQce0IQJEyqr7GrP1Z+ViIgIbdy4UQ6HQ/n5+dq8ebNatWpVGSVXe672pGnTptqwYYMkKS8vT19//bVatmxpeL34kzu+z5sc1fBEgP3792v06NE6efKkrrrqKsXHx6tFixaKiYnRyJEj1bZtW9lsNr388svatGmTJCkmJkb9+/ev5MqrL1d6ct999+m3334rcvLp1KlTFRISUomVV1+u9OR8s2fPVm5urp5//vlKqrhmcKUvdrtd8fHx2rBhg8xms2677TY9//zzMpur3d/RbsGVnhw6dEjjxo1Tenq6bDabbrrpJr344ouyWq2VXX61NHHiRK1du1bp6elq2LChGjRooJUrV7r9+3y1DF0AAADuhj+LAAAADEDoAgAAMAChCwAAwACELgAAAAMQugAAAAxA6AIAADAAoQsAAMAAhC4AAAAD/H8zPDypf6XgZQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "generate_auc_roc_curve(clf, X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# SMOTE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{0: 190477, 1: 343}"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unique, count = np.unique(Y_train, return_counts=True)\n",
    "Y_train_dict_value_count = { k:v for (k,v) in zip(unique, count)}\n",
    "Y_train_dict_value_count"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "sm = SMOTE(random_state=12, ratio = 1.0)\n",
    "x_train_res, y_train_res = sm.fit_sample(X_train, Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{0: 190477, 1: 190477}"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unique, count = np.unique(y_train_res, return_counts=True)\n",
    "y_train_smote_value_count = { k:v for (k,v) in zip(unique, count)}\n",
    "y_train_smote_value_count"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf = LogisticRegression().fit(x_train_res, y_train_res)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "Y_Test_Pred = clf.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "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>Actual</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Predicted</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>92303</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1535</td>\n",
       "      <td>135</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Actual         0    1\n",
       "Predicted            \n",
       "0          92303   14\n",
       "1           1535  135"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(Y_Test_Pred, Y_test[target], rownames=['Predicted'], colnames=['Actual'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy =  0.9835189973081384\n",
      "Precision =  0.08083832335329341\n",
      "Recall =  0.9060402684563759\n",
      "F1 Score =  0.148433205057724\n"
     ]
    }
   ],
   "source": [
    "generate_model_report(Y_test, Y_Test_Pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "generate_auc_roc_curve(clf, X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.005     , 0.03222222, 0.05944444, 0.08666667, 0.11388889,\n",
       "       0.14111111, 0.16833333, 0.19555556, 0.22277778, 0.25      ])"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "weights = np.linspace(0.005, 0.25, 10)\n",
    "weights"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best parameters : {'smote__ratio': 0.005}\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a210a71d0>"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pipe = make_pipeline(\n",
    "    SMOTE(),\n",
    "    LogisticRegression()\n",
    ")\n",
    "\n",
    "weights = np.linspace(0.005, 0.25, 10)\n",
    "\n",
    "gsc = GridSearchCV(\n",
    "    estimator=pipe,\n",
    "    param_grid={\n",
    "        'smote__ratio': weights\n",
    "    },\n",
    "    scoring='f1',\n",
    "    cv=3\n",
    ")\n",
    "grid_result = gsc.fit(X_train, Y_train)\n",
    "\n",
    "print(\"Best parameters : %s\" % grid_result.best_params_)\n",
    "weight_f1_score_df = pd.DataFrame({ 'score': grid_result.cv_results_['mean_test_score'],\n",
    "                                   'weight': weights })\n",
    "weight_f1_score_df.plot(x='weight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "pipe = make_pipeline(\n",
    "    SMOTE(ratio=0.005),\n",
    "    LogisticRegression()\n",
    ")\n",
    "\n",
    "pipe.fit(X_train, Y_train)\n",
    "\n",
    "Y_Test_Pred = pipe.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "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>Actual</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Predicted</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>93798</td>\n",
       "      <td>41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>40</td>\n",
       "      <td>108</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Actual         0    1\n",
       "Predicted            \n",
       "0          93798   41\n",
       "1             40  108"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(Y_Test_Pred, Y_test[target], \n",
    "            rownames=['Predicted'], colnames=['Actual'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy =  0.9991381786842861\n",
      "Precision =  0.7297297297297297\n",
      "Recall =  0.7248322147651006\n",
      "F1 Score =  0.7272727272727273\n"
     ]
    }
   ],
   "source": [
    "generate_model_report(Y_test, Y_Test_Pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "generate_auc_roc_curve(clf, X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# UNDERSAMPLING"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "492\n"
     ]
    }
   ],
   "source": [
    "minority_class_len = len(df[df[target] == 1])\n",
    "print(minority_class_len)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Int64Index([     0,      1,      2,      3,      4,      5,      6,      7,\n",
      "                 8,      9,\n",
      "            ...\n",
      "            284797, 284798, 284799, 284800, 284801, 284802, 284803, 284804,\n",
      "            284805, 284806],\n",
      "           dtype='int64', length=284315)\n"
     ]
    }
   ],
   "source": [
    "majority_class_indices = df[df[target] == 0].index\n",
    "print(majority_class_indices)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "492\n"
     ]
    }
   ],
   "source": [
    "random_majority_indices = np.random.choice(majority_class_indices,\n",
    "                                           minority_class_len, \n",
    "                                           replace=False)\n",
    "print(len(random_majority_indices))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Int64Index([   541,    623,   4920,   6108,   6329,   6331,   6334,   6336,\n",
      "              6338,   6427,\n",
      "            ...\n",
      "            274382, 274475, 275992, 276071, 276864, 279863, 280143, 280149,\n",
      "            281144, 281674],\n",
      "           dtype='int64', length=492)\n"
     ]
    }
   ],
   "source": [
    "minority_class_indices = df[df[target] == 1].index\n",
    "print(minority_class_indices)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "under_sample_indices = np.concatenate([minority_class_indices,random_majority_indices])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "under_sample = df.loc[under_sample_indices]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a219d0438>"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(x=target, data=under_sample)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = under_sample.loc[:, df.columns!=target]\n",
    "Y = under_sample.loc[:, df.columns==target]\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.33, random_state=42)\n",
    "clf = LogisticRegression().fit(X_train, Y_train)\n",
    "Y_Test_Pred = clf.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy =  0.9323076923076923\n",
      "Precision =  0.9863013698630136\n",
      "Recall =  0.8780487804878049\n",
      "F1 Score =  0.9290322580645162\n"
     ]
    }
   ],
   "source": [
    "generate_model_report(Y_test, Y_Test_Pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "generate_auc_roc_curve(clf, X_test)"
   ]
  },
  {
   "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}