outliers12022.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "a3e6b7f0",
   "metadata": {},
   "source": [
    "## Data Mining and Machine Learning\n",
    "### Edgar Acuna\n",
    "### Novelty detection using Mahalanobis Distance, MCD, LOF, Isolation Forest and Clustering(kmeans, DBSCAN)\n",
    "#### Dataset Diabetes\n",
    "#### March 2022"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "c67ad736",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.ensemble import IsolationForest\n",
    "from sklearn.neighbors import LocalOutlierFactor\n",
    "from sklearn.metrics.pairwise import pairwise_distances\n",
    "from sklearn.preprocessing import StandardScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "1febf19e",
   "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>preg</th>\n",
       "      <th>plas</th>\n",
       "      <th>pres</th>\n",
       "      <th>skin</th>\n",
       "      <th>test</th>\n",
       "      <th>mass</th>\n",
       "      <th>pedi</th>\n",
       "      <th>age</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>6</td>\n",
       "      <td>148</td>\n",
       "      <td>72</td>\n",
       "      <td>35</td>\n",
       "      <td>0</td>\n",
       "      <td>33.6</td>\n",
       "      <td>0.627</td>\n",
       "      <td>50</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>85</td>\n",
       "      <td>66</td>\n",
       "      <td>29</td>\n",
       "      <td>0</td>\n",
       "      <td>26.6</td>\n",
       "      <td>0.351</td>\n",
       "      <td>31</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>8</td>\n",
       "      <td>183</td>\n",
       "      <td>64</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>23.3</td>\n",
       "      <td>0.672</td>\n",
       "      <td>32</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>89</td>\n",
       "      <td>66</td>\n",
       "      <td>23</td>\n",
       "      <td>94</td>\n",
       "      <td>28.1</td>\n",
       "      <td>0.167</td>\n",
       "      <td>21</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>137</td>\n",
       "      <td>40</td>\n",
       "      <td>35</td>\n",
       "      <td>168</td>\n",
       "      <td>43.1</td>\n",
       "      <td>2.288</td>\n",
       "      <td>33</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   preg  plas  pres  skin  test  mass   pedi  age  class\n",
       "0     6   148    72    35     0  33.6  0.627   50      2\n",
       "1     1    85    66    29     0  26.6  0.351   31      1\n",
       "2     8   183    64     0     0  23.3  0.672   32      2\n",
       "3     1    89    66    23    94  28.1  0.167   21      1\n",
       "4     0   137    40    35   168  43.1  2.288   33      2"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "url= \"https://academic.uprm.edu/eacuna/diabetes.dat\"\n",
    "names = ['preg', 'plas', 'pres', 'skin', 'test', 'mass', 'pedi', 'age', 'class']\n",
    "data = pd.read_table(url, names=names)\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a964b6f1",
   "metadata": {},
   "source": [
    "### Extracting all the datapoints form class 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "e35d8c59",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     preg  plas  pres  skin  test  mass   pedi  age\n",
      "1       1    85    66    29     0  26.6  0.351   31\n",
      "3       1    89    66    23    94  28.1  0.167   21\n",
      "5       5   116    74     0     0  25.6  0.201   30\n",
      "7      10   115     0     0     0  35.3  0.134   29\n",
      "10      4   110    92     0     0  37.6  0.191   30\n",
      "..    ...   ...   ...   ...   ...   ...    ...  ...\n",
      "762     9    89    62     0     0  22.5  0.142   33\n",
      "763    10   101    76    48   180  32.9  0.171   63\n",
      "764     2   122    70    27     0  36.8  0.340   27\n",
      "765     5   121    72    23   112  26.2  0.245   30\n",
      "767     1    93    70    31     0  30.4  0.315   23\n",
      "\n",
      "[500 rows x 8 columns]\n"
     ]
    }
   ],
   "source": [
    "yd=data['class']\n",
    "X1=data[data['class']==1]\n",
    "Xc1=X1.iloc[:,0:8]\n",
    "print(Xc1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1c88cb95",
   "metadata": {},
   "source": [
    "### 0. Exploring outliers using Boxplots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "074d8383",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "import seaborn as sns\n",
    "sns.boxplot(data=Xc1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a419b2a6",
   "metadata": {},
   "source": [
    "#### Feature test shows a large amount of outliers"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6347715d",
   "metadata": {},
   "source": [
    "### 1. Outlier Detection using Mahalanobis Distance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "072244ed",
   "metadata": {},
   "outputs": [],
   "source": [
    "#create function to calculate Mahalanobis distance\n",
    "def mahalanobis(x=None, data=None, cov=None):\n",
    "\n",
    "    x_mu = x - np.mean(data)\n",
    "    if not cov:\n",
    "        cov = np.cov(data.values.T)\n",
    "    inv_covmat = np.linalg.inv(cov)\n",
    "    left = np.dot(x_mu, inv_covmat)\n",
    "    mahal = np.dot(left, x_mu.T)\n",
    "    return mahal.diagonal()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "5426012b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x241cb7a4430>]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "mh=mahalanobis(Xc1,data=Xc1)\n",
    "plt.plot(mh)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "7a758fd6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x241cb80c0d0>]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "mh=mahalanobis(Xc1,data=Xc1)\n",
    "plt.plot(np.sort(mh),marker='.',linestyle='None',markersize = 2.0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "7ca32d34",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "sns.boxplot(x=mh)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "13998a38",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7      31.273760\n",
      "12     21.209561\n",
      "49     26.750454\n",
      "58     33.385928\n",
      "60     22.212416\n",
      "81     23.259208\n",
      "86     20.970823\n",
      "145    25.705911\n",
      "182    21.065892\n",
      "228    76.176431\n",
      "247    49.672987\n",
      "286    25.377663\n",
      "294    26.750191\n",
      "336    30.682114\n",
      "371    43.494126\n",
      "426    23.389467\n",
      "453    43.070890\n",
      "459    24.258342\n",
      "487    24.920184\n",
      "494    22.806980\n",
      "519    25.751790\n",
      "522    25.925746\n",
      "537    33.121450\n",
      "549    21.667486\n",
      "593    20.984161\n",
      "596    22.264786\n",
      "621    23.051348\n",
      "622    29.023577\n",
      "684    33.959666\n",
      "703    29.548576\n",
      "744    24.982187\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "mh1=pd.Series(mh,index=Xc1.index)\n",
    "outs= mh1[mh1 >20.0]\n",
    "#print(outs.index)\n",
    "print(outs)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9628e394",
   "metadata": {},
   "source": [
    "### 2. Minimum  Covariance Distance ( MCD)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "21c4b6e1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x241ccb42eb0>]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "from sklearn.covariance import EllipticEnvelope\n",
    "clf=EllipticEnvelope(contamination=0.25)\n",
    "clf.fit(Xc1)\n",
    "outmcd = clf.decision_function(Xc1)\n",
    "outmcd=pd.Series(outmcd,index=Xc1.index)\n",
    "plt.plot(outmcd)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "f9681363",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "sns.boxplot(x=outmcd)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "7916249b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7      -53.392126\n",
      "12     -46.568091\n",
      "30     -44.572234\n",
      "58     -55.810247\n",
      "67     -43.006858\n",
      "123    -91.499555\n",
      "148    -70.781273\n",
      "212    -43.927229\n",
      "222    -40.428223\n",
      "223    -52.804512\n",
      "228   -130.357450\n",
      "247    -74.285760\n",
      "263    -70.929257\n",
      "294   -133.298304\n",
      "336    -83.909630\n",
      "361    -65.883709\n",
      "362    -68.470700\n",
      "371    -57.119955\n",
      "453   -222.805214\n",
      "456    -95.875704\n",
      "459   -143.826508\n",
      "475    -76.022357\n",
      "479    -65.722634\n",
      "487    -81.913062\n",
      "489    -76.692458\n",
      "495    -74.498485\n",
      "509    -50.586686\n",
      "519    -68.824434\n",
      "537   -143.884049\n",
      "552    -67.598799\n",
      "596    -40.095379\n",
      "674    -71.022762\n",
      "684   -127.948121\n",
      "703    -70.860957\n",
      "763    -52.833853\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "outs1= outmcd[outmcd < -40.0]\n",
    "print(outs1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f712e3b5",
   "metadata": {},
   "source": [
    "### 3. Local Outlier Factor (LOF)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "4d6c88ce",
   "metadata": {},
   "outputs": [],
   "source": [
    "# fit the model for outlier detection (default)\n",
    "clf = LocalOutlierFactor(n_neighbors=20, contamination=0.05)\n",
    "# use fit_predict to compute the predicted labels of the training samples\n",
    "# (when LOF is used for outlier detection, the estimator has no predict,\n",
    "# decision_function and score_samples methods).\n",
    "y_pred = clf.fit_predict(Xc1)\n",
    "#print(np.where(y_pred==-1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "013a817f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x241ccc01ee0>]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "X_scores = clf.negative_outlier_factor_\n",
    "plt.plot(X_scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "60382699",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x241ccc5f940>]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "plt.plot(np.sort(X_scores),marker='.',linestyle='None',markersize = 2.0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "320adbc0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWAAAAD4CAYAAADSIzzWAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAAMFklEQVR4nO3dYWjc533A8d9jy8WyWRmxO4/aI5pRvSaQVzF7szeBOUMWS8deDAaDKOSFIKGOyat1iVgI0wJlr4LflLyTYWNvRtnCZC8xYezVYBKkTRu74mZkWod5qRzaBdutZf/3Qjojn+6kO0l3P93p84FQ6fTof7/nnvD16S+HlqqqAoDe25c9AMBeJcAASQQYIIkAAyQRYIAkQ50sPnr0aDUyMtKlUQAG0/z8/M+rqvpa4+MdBXhkZCTm5uZ2biqAPaCUcqPZ425BACQRYIAkAgyQRIABkggwQBIBBkgiwABJBBggiQADJBFggCQCDJBEgAGSCDBAEgEGSCLAAEkEGCCJAAMkEWCAJAIMkKSj/084gN3swoULUavVNl138+bNiIg4fvz4uq+Njo7GuXPndny2ZgQYGBi1Wi0+/tHVeHDoiQ3X7b/zi4iI+J9fDTU8frtrszUjwMBAeXDoibj7zfEN1wxfm42IWLeu/nivuAcMkESAAZIIMEASAQZIIsAASQQYIIkAAyQRYIAkAgyQRIABkggwQBIBBkgiwABJBBggiQADJBFggCQCDJBEgAGSCDBAEgEGSCLAAEkEGCCJAAMkEWCAJAIMkESAAZIIMEASAQZIIsAASQQYIIkAAyQRYIAkAgyQRIABkggwQBIBBkgiwABJBBggiQADJBFggCQCDJBEgIGeuXDhQly4cCF7jI50c+ahrlwVoIlarZY9Qse6ObN3wABJBBggiQADJBFggCQCDJBEgAGSCDBAEgEGSCLAAEkEGCCJAAMkEWCAJAIMkESAAZIIMEASAQZIIsAASQQYIIkAAyQRYIAkAgyQRIABkggwQBIBBkgiwABJBBggiQADJBFggCQCDJBEgAGSCDBAEgEGSCLAAEkEGCCJAAMkEWCAJAIMkESAAZL0JMBLS0vx2muvxdLSUi+eLs1e2Wddfb+1Wm1b++7V67abz6dxtqWlpXj11VfjlVde2XTeTs6h2XVrtVqMjY3Fc889Fy+//HLUarV1axq/b2lpKSYnJ2NsbCwmJyfXPXd9/UsvvRTj4+Px0UcfxdmzZ2NhYSHu37+/Ey/ZQOhJgGdmZuKTTz6Jixcv9uLp0uyVfdbV9zs9Pb2tfffqddvN59M428zMTHz66adx9erVTeft5ByaXXd6ejru3bsXERHXr1+P6enpdWsav29mZiYWFhbi3r17sbCwsO656+sXFxfjzp078c4778Tdu3fj7t27cevWrW2/XoOi6wFeWlqKy5cvR1VVcfny5V357mMn7JV91q3d7+Li4pb33avXbTefT+NstVotLl++/Ojrly5dajlvJ+dQX7v2uvPz87G4uPjYurWfX7p0ad08s7OzMTs7u+57Ws0fEbG8vPzo49u3b++q1z/TULefYGZmJh4+fBgREQ8ePIiLFy/G66+/3u2n7bm9ss+6tfut28q+e/W67ebzaZxtenr6sR/T79+/33LeTs5hZmZm3XXfeuutDWe7f/9+03laaTZ/o6qqYnJyMk6cOLHhc29FrVaLfb+utvz9++79Mmq1/4vz588/ds3h4eGdGG/98222oJQyWUqZK6XMff755x0/wZUrVx796be8vBwffvhh51P2gb2yz7q1+63byr579brt5vNpnK3+brKuqqqW83ZyDleuXFl33S+//HLD2da+s25Hs/mb+eKLL9q63qDb9B1wVVXvRcR7ERGnT5/u+I+WM2fOxOzsbCwvL8fQ0FA8//zzWxhz99sr+6xbu9+6rey7V6/bbj6fxtlOnDgRN27ceBSxUkrLeTs5hzNnzsT777//2HUPHz68YYRLKfHkk08+Ns9Gms3fzAsvvNCVn0DOnz8f89e3fo/54cGvxujJY/Huu+8+ds1u6fo94ImJidi3b+Vp9u/fHy+++GK3nzLFXtln3dr91m1l37163Xbz+TTONjU1FQcOHHj09QMHDrSct5NzmJiYWHfdt99+e8PZDhw40HSeoaHm792azd+olLKrXv9MXQ/wkSNHYmxsLEopMTY2FkeOHOn2U6bYK/usW7vfkZGRLe+7V6/bbj6fxtlGR0djbGzs0dfPnj3bct5OzqG+du11n3322RgZGXls3drPz549u26e8fHxGB8fX/c9reaPiMeC/cQTT+yq1z9TT/4a2sTERDzzzDMD/6feXtlnXX2/U1NT29p3r1633Xw+jbNNTEzE008/HU899dSm83ZyDs2uOzU1FQcPHoyIiJMnT8bU1NS6NY3fNzExEadOnYqDBw/GqVOn1j13ff3IyEgcOnQo3njjjRgeHo7h4eE4duzYtl+vQVHavbkesXIPeG5urovjAIOsfj917T3Wnb7+/PVbcfeb4xuuG7628tfoGtcNX5uNZ1vcA97OzKWU+aqqTjc+7j9FBkgiwABJBBggiQADJBFggCQCDJBEgAGSCDBAEgEGSCLAAEkEGCCJAAMkEWCAJAIMkESAAZIIMEASAQZIIsAASQQYIIkAAyQRYIAkAgyQRIABkggwQBIBBkgiwABJBBggiQADJBFggCQCDJBEgAGSCDBAEgEGSCLAAEkEGCCJAAMkEWCAJAIMkGQoewBg7xgdHc0eoWPdnFmAgZ45d+5c9ggd6+bMbkEAJBFggCQCDJBEgAGSCDBAEgEGSCLAAEkEGCCJAAMkEWCAJAIMkESAAZIIMEASAQZIIsAASQQYIIkAAyQRYIAkAgyQRIABkggwQBIBBkgiwABJBBggiQADJBFggCQCDJBEgAGSCDBAEgEGSCLAAEkEGCCJAAMkEWCAJAIMkESAAZIIMEASAQZIIsAASQQYIMlQ9gAAO2n/ndsxfG12kzVLERHr1u2/czsijnVrtHUEGBgYo6Ojba27eXM5IiKOH2+M7bG2r7ETBBgYGOfOncseoSPuAQMkEWCAJAIMkESAAZIIMEASAQZIIsAASQQYIIkAAyQRYIAkAgyQRIABkggwQBIBBkgiwABJBBggiQADJBFggCQCDJBEgAGSlKqq2l9cyucRcaN742zL0Yj4efYQXWJv/cne+lM39vZkVVVfa3ywowDvZqWUuaqqTmfP0Q321p/srT/1cm9uQQAkEWCAJIMU4PeyB+gie+tP9tafera3gbkHDNBvBukdMEBfEWCAJH0Z4FLK35RSflhK+biU8kEp5est1i2WUj5ZXTfX6zm3qoP9jZVSflJKqZVSvtPrObeilPJ3pZRrq/v7finlN1us67uz62Bv/Xhuf1ZK+XEp5WEppeVf0erTc2t3bzt/blVV9d0/EfHVNR+/FhHfa7FuMSKOZs/bjf1FxP6I+O+IOBkRX4mIH0TE09mzt7G3P4qIodWPvxsR3x2Us2tnb318bk9FxO9FxL9HxOkN1vXjuW26t26dW1++A66q6pdrPj0cEQP1m8Q29/f7EVGrqup6VVW/joh/jIg/6cV821FV1QdVVS2vfvqfEXEic56d1Obe+vXcrlZV9ZPsObqhzb115dz6MsAREaWUvy2l/DQi/iIi/rrFsioiPiilzJdSJns33fa1sb/jEfHTNZ//bPWxfvJyRFxq8bW+PbtVrfY2COe2kX4/t1a6cm5D271At5RSrkTEbzf50ptVVf1zVVVvRsSbpZS/iohvR8RbTdb+QVVVn5VSfisiPiylXKuq6j+6OHbbdmB/pcn37oqfBDbb2+qaNyNiOSL+vsVlduXZ7cDe+vrc2tC357bZJZo8tu1z27UBrqrqTJtL/yEi/jWaBLiqqs9W//d/Synfj5UfI9L/ZYjYkf39LCJ+Z83nJyLisx0Ybds221spZSIi/jgi/rBavcHW5Bq78ux2YG99e25tXqMvz60NXTm3vrwFUUr5xppPvxUR15qsOVxK+Y36x7HyC5If9WbC7WlnfxHxXxHxjVLK75ZSvhIRfx4R/9KL+bajlDIWEX8ZEd+qqupOizV9eXbt7C369Nza0a/n1qbunFv2byC3+FvLf4qVg/1hRLwfEcdXH/96RMyufnwyVn5T+YOI+HGs/KiRPvtO7W/18/GIWIiV3872xf4iohYr99I+Xv3ne4Nydu3srY/P7U9j5V3gryLiVkT82wCd26Z769a5+U+RAZL05S0IgEEgwABJBBggiQADJBFggCQCDJBEgAGS/D8OPiZmVRBpZAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(x=X_scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "55ced0dc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "18    -1.480470\n",
      "40    -1.738998\n",
      "62    -1.479756\n",
      "75    -2.620721\n",
      "86    -1.405982\n",
      "106   -1.756651\n",
      "145   -1.603657\n",
      "146   -1.472256\n",
      "153   -1.893882\n",
      "182   -2.477287\n",
      "211   -1.486654\n",
      "212   -1.402344\n",
      "228   -3.714689\n",
      "247   -3.183683\n",
      "260   -1.565333\n",
      "286   -2.246802\n",
      "342   -2.639874\n",
      "362   -1.608571\n",
      "392   -1.453543\n",
      "459   -1.842283\n",
      "486   -1.849507\n",
      "489   -1.607684\n",
      "537   -1.642067\n",
      "548   -1.572597\n",
      "549   -1.653318\n",
      "597   -1.596125\n",
      "622   -1.516020\n",
      "645   -1.603176\n",
      "672   -1.593961\n",
      "703   -1.416859\n",
      "728   -1.526423\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "lofs1=pd.Series(X_scores,index=Xc1.index)\n",
    "outs2= lofs1[lofs1 < -1.4]\n",
    "print(outs2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5deab755",
   "metadata": {},
   "source": [
    "### 4. Isolation Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "bd81d754",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x241ccd29700>]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAD4CAYAAADhNOGaAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAABmo0lEQVR4nO19eZgdVZn++9W9t7vT2UMWQhJIwLAvESKLyCo7KqiogAvqzCCOODO4/IwCjju4jI6OjgiKO6KjsggRBARZZEmQLSwhIQQSEkhCyNrpvtv5/VF1qr46dU4t99bt7tv3vM/TT99bt+psdc759u+QEAIWFhYWFp0LZ6gbYGFhYWExtLCEwMLCwqLDYQmBhYWFRYfDEgILCwuLDoclBBYWFhYdjuJQN6ARTJ48WcyePXuom2FhYWHRVnj44Yc3CCGmqNfbkhDMnj0bixcvHupmWFhYWLQViOgF3XWrGrKwsLDocFhCYGFhYdHhsITAwsLCosORCyEgolOIaCkRLSeiBZrf9yai+4logIg+pfy2koieIKJHicgq/i0sLCwGGU0bi4moAOAHAE4EsBrAIiK6UQjxFLttI4B/A3CmoZjjhBAbmm2LhYWFhUV25CERHApguRBihRCiDOBaAGfwG4QQ64QQiwBUcqjPwsLCwiJH5EEIZgBYxb6v9q6lhQDwFyJ6mIjON91EROcT0WIiWrx+/foGm2phYWFhoSIPQkCaa1lyWx8phDgYwKkAPkZER+tuEkJcKYSYL4SYP2VKJB5iWGLd1n78/bnhpfH627PrsWpj31A3w8LCYhghD0KwGsAs9n0mgDVpHxZCrPH+rwNwHVxV04jAWT+8H+de9eBQNyOE865+CEd9486hboaFhcUwQh6EYBGAuUQ0h4i6AJwN4MY0DxLRaCIaKz8DOAnAkhzaNCzwosd51+v28B8LC4vhi6YJgRCiCuBCALcCeBrA74QQTxLRBUR0AQAQ0c5EtBrAJwBcQkSriWgcgGkA7iWixwA8BOBmIcQtzbZpuKFcq/uf123tx3Prtw1hayyGAhu3l/HT+56HPRHQYjgil1xDQoiFABYq165gn1+GqzJSsQXAQXm0YTijXKujp1QAABz61TsAACsvP30om2QxyPjk7x7FnUvX4w2zJ2H/GeOHujkWFiHYyOJBQKVaT77JYkRj4/YyAKBq1YQWwxCWEAwCuGqoE/C7xauw5KXNQ92MYQW5/zs6H7sRhkqtju//dRn6K7WhbopFSlhCMAioVEcOF7ijnLy4/9/vH8db/ufeQWiNi2/f9iz+9FhqR7UhQd2zDTg0eJTgrqXrcMuSlwetPonfP7wa3/rLs/j+X5cPet0WjcESgkFAuTa0nNHmvkouhsr7n3sV+3z+lmG36f7f4lW4/elXhroZsah5IsEg0gF88KeLcMGvHh68Cj1UPAl48w6bSKBd0JYH07QbykMsEXzmD4/jlidfxoEzmzNSXn7LMwCAlRu259Gs3NBfqQ173bvwVUMjXzdUKrj8ZbXeWSrRvHHPsvU4ZLeJ6O1q/TZtJYIc8Oq2gdjNMQ8bwStb+vHUmi0NPbt+2wAAoFJrfLMUQvh6/1FdhYbLaQUGqnVUDWO8csP2YRHHMRSqocHC02u34Lu3L/O/S0Iw1AxQFqza2Ifl67YOdTN8rFi/De//yUO45LogrGrDtgFc9uenW+J+bglBDjjqG3fi2G/dZfy9EkMI9rzkz7jq7hUAgFO/ew9+fM8K7X2HX3YHTvvePcZy/vOGJfjnny/S/iY3yUITlsrVr+3w1RvNaJhuenwNvvinJxsvQIEQAv2Vmt82jqfWbMGx37oLVxnGdLCwcsN2LFvnLt6RaCx+94/ux3dufxb9lRq+fsszePFVlykaDhLBqo19xjXFcdQ37sQJ3747dbkvb+7Hn59Y20zTIli7eQdmL7gZtz31Cl7rc9VqKxiD+eSaLfjR31Zg/daBXOsFLCHIBX0JBtSywX1UCIFytY6vLnwagMtZfeXmp/3fX902wO6Nb8PP738Btz+9zv9eqdV9m4CUBJrhRjkXUm+CElx4zSP46X0rG35eRaUmUBd6t0wZ2b34hddyq68RcCaBRqBEICWuNZt24Id3PYfveUbiOAZosPCBqx/CV25+Gq9uG8DF1z2Bh3OaC9c8+AI+ds0/tAxIo3j0xU0AgN8/vMpn3ooe5/C3Z9fjvKsfAgDsvfPY3OqU6DhC8OCKV/HeHz/ge79s7qtg9oKb8btFqxKebBw61ZAQIlavfcuSl3HIV27HopUbs9dXrWPuxX/2dfp5cGZVplaKm/uDHTk7UHXfo7oghRC+SizvrfddV/wdF17zj5xLbV+UitImEH4Hzagi88JrfW78Rn+1jl8/+CL+9mw4c/GOcg0rGlC1bOwroy7yJXZyn+gqFvz5XCy4s/eX9wdnzk/o7cqtTomOIwTvufIB3Lf8VbyypR9AwDX+4oGVLatTJxHU6iK0uaq4d7k7YZ9Zm90uIDfHXz/wIoBgEzfp0dOgJjghcD9v2DaAXz3wQmjzH2x1fH/F7ZO6IH909wpcer2rX82bCV+08jXc9HijaoHoAA1Ua9jv87fghkdfaq5hSTV7arS80eXZBFTJOM1829JfaapNG7YNGCVuIJBWtvVXAQADSl1v+/69OP6//pa53k2e6iZPQjDgzeWuguMT1aLjju382RNzq0eHjiIEfMLJ5VjLYMTb3FfBfp+/JXNqaTlZONdarQtUYjh1OSkeWLERR33jr5nqk7XIDVvW06yxWP186fVLcMn1S/D46iB4LE9ROQ3kO1XrvWtpoCYbTgZancC0ZUcV28s1XHJ9fL7F/kqtKWL+v3c9h70vvSXWrbPKVIppUfIJQTV0Pc18O/ALf8E7f/j3TPVJCCEw/yu346LfPmq8R06Lrf1un3ewPWDd1n7fdmOCicjIMVT7+NSaLfjgTx+KJU4mDHjvtrvk+FK8lAjkDL7lP47KXG4adBQhWLcl0LnLTVL+T6O7vX/FBmwv13D1vStj76vXRWgxyUkhOXXAJQQ1rm5RNrIB75mbn1iLVRt3JLZNrR9wN53L//yM/zwnPFk9afj+oz76IjvfwGQ/qNTqLfF2kOOkqiX45q97tWs27cBP73s+9/YkQTc6wru6tb+q+TXA3pfegg/9TO8QkAYy1iJOynzdxX/Gx3/zSKZyuz3VkBpsmMQtyzn4ZIPecJL432ww2g5Ua9g24I6pHFvODK5+LX5dLXlpM/a85M+4QxOjYpIIPvvHx3HX0vV4qgFJXu4TXQXHl+KljUDOm90mjc5cbhp0FCF4ZWu//1nuV3IypvHmkBNn5sRRsfft/rmF+Nx1T/jf5WSRXD4A1GphiaAmVEKgF5c376j43I0JvncPBK7423NBOxiXktWlta5RDe0ywR2HlzYFC8okEXzuj0/gzQ2I4EkwSQR889cR+X/5xWJ88U9PYe3mbES2WegIZRYp6p5lydLo0pe3agn9nMnuJvK8wdVZShuq2uvOZ9bh1O/eE+H4JaREsD0jIdiwrTnvl6TYkZO/E3gBbfHWzNb+Kr5w45N4bXsZGxK8bx550TUs37U0eiLiph2u7UHl/OVcS/NOf/XAC5i94GZ/PcuyuouBakh6+gUMa2KxDaGjAsq4RCB8icD9Xkgxwms2uYRklwk9iff+5qHA+FyuuuK2NFwBrgGX2wjUiSM53Umju/yEZQBw0Bf/gqJDWP6104x1m9w8uRjLM6ImoV4XoQkvmzptXDcAl7v26zZIBNe3SP8tCaYqohMzEeverOQUByp1vOdH9+PDb5qDk/fbOVPdlVrd3wTTQg7PF//0JIQAvvC2/XJVpy15aTPe8j/34tMn7xX5bdbEXgBmQvAqm2dBe4UvhTz4/EZs669i3qwJmDWp17+nVHRHeIdCKPhGfd0jq3HAjPF43dTA42WVx1h1FRvjR5OYmZWvBpKqlAj+7KXc6K/UcODMCanq0W0NUiIo1+p4ddsAJvR2oeCQv3GnUa/98C6XSXttewVje0r+GisVHJ+IFr35JYtrFSHoKIlgHZcIvP9ZAn1e2uROrFExkX66CVCuCVx8/ZKQUapWF6ENIEIIPOlh+0CUC6vWhXExA8FmrLaEew+9/ycPaQPUHnp+I4647I6QHvnSG5bgk//3mP9dKGP2kreghQiruzjUjfq6R1bjpsebT1Uhx6lWr+Mn9z6P2QtuxuYdldCC0b1bKXJvG6jiwec3NpSKYYtB175+6wDOu/ohbOrTbazu/5/etxI/+/tKr+3B2Nz/3KuZ28EhpbNHPFdEDjkMqzfppSDpQDGuJ5jfP7k3UJ/d/9yr+PhvHsGZP7gv9Jw0Fm8fUCWCoF8XX7cEv/KcF9S2Th/vMla/fOAF/OmxNTjisjtw5d3PIQlxzhYqtihSdF0IrUTC34Wp9Gqt7hOWjdvLOOQrt+Orntu31CzU6gKzF9wc0gyokEyMfC8yFU130fEZwZJCWFpl7+ooQvBKjI3ASTESazd7hCSG2usMZOVqHdc8GF4E1boIic6qmCuNWgMGo9NxMQFs/gKJkQgeW7UJn9VM0u/fuRxrN/fjwRXBhvRrpe3q2ElO8qLfPorXf/k2Y7s4LvrtY7jwmmy6aB36vcVUrQv85iG3nXJDk9CtHcnJy+R4ae2jnNCbjK5X3bMCf3t2Pa7VuCQLzfbC332aDTArnnnZJfhSXWQi1nJ9TB7b7V97fPVmzJgwCvtOH4enPb23KjnIsdxRMauGytV6ZLxe8dbTtHEuIfj1Ay/gj/9YjbWb+/G1hc8Y+1Ot1dFfqeHLNz0V+W3py1u1mW+3KfYXhwgbtg1g/KgSZkwIVL06dZY6fbawsmRw161PupKGVA3JV3rNgy9ic59+nkibilzjvo2AEQIpEcjyWuX20FGEoJuJoHI9Z8kBI7lzkyT/yIuvaX32X9NwhrV6OI5A1elyDqan5OCS0/dJbJ9fFtuox3QH3J06yXs0IvkMT+21dnM/5n/ldvzgzmgGSdlU1SPj+kejHH5fuYov/Sm6YPNCIBEIX71Xq4uQXYCAiItiHuoIEyHwfcA1hichovYf/u7vXLq+qRiFoMagzFP++x7U68KXFE3qOykxTx4dEAIBd6zGdBeN/ZVjGfUacsdKxsxs2VHBU2u2+MROEg6poizX6ka9/ytb+rHwibXYPlDF6y7+M468/K+47pGouvHk/75bm/lWNcSTRwimjO0O2fz4+zUxB1zSk5u59O6Rc5AzDAd96S/4wNUP4WomXQFAn9f/H9+zAv2VWpgQeL/5xuIM+1Qj6ChCcNGJe+KK9x0CINgs5aJNM8CSSpv0f2//379jxfqoyuY1je61qsQRqAuAczBFx9Gmh5i94OaQukstq1oXIeNkhBBobASScKxYvw0btg3gm7cujdyjSgQmbxchBH58z/O4OoN3zt+Xb8CNGbKb+hJBTcBxGCFg9/zt2fXY+9Jb8PjqTf413SadBtxWkkQITHOK26qA6Lvnxto7n1kXssHocONja3CvZ0R2FI5UYkel5nt+mWwSsl0Tekv+NSHcsezpKvh6cXUumlRDcn5LSXTzjgpO+949+NrCZyBEIBHL0srVutHt8pyrHsC//vofvkpUZ8+Ig6oaKjiubn5ibwlnHRIcnqirX3U22MTeuyRmckykZkEltnc/ux5fUiQYecu1i1bhe3csC9RBhUAiGCxjcUcRAiDQ4fleQ75qKHmEZfBSnG3PVx8x6FJQVGv1kPSgLk6+aRccMm5cD66IRh5zDpPXraqtekrR17/NW8zc0KZClaa2DVS1nkzVenz0tA7n/vhB/FsG90X5Tqr1OqTd1pUIgntk3pbfP7zav5bWyKsSfb5RqBufhHyXazbtiGwsdSHwsqK60m3M0oPnQz9bhLd9P/5sh3/7zSN4308eBBBsFGq7d1RqEeZHRb+is/7Jvc/jjqfXgQjoLRV8Tlh1rDDHEYSD/TjhrItgLGW7ytW60QAskzrqGJ80UJkVhwg1IVB0HLxr/izfuF4JSQT6ceKqnn6Fc5eEOOu837yj4o8HUcB0/vrBF/Hd25f5bWlVipKOIwREYVErMBYDX7jxSbzjf+8zPeqL9HGveNtAdEPcrnG7c20EzFgsVEIQfC865OsKVegOijFNQjUYSScR+IE3MfmTJKGR//vKNbysIYDlar1hzjstBpj7aMFjxwaq9RA3LlWCDz0fEM00hODOpesw57MLsfTlICsl36hM7pRy/H987/P4zB8eD/0mBCIpDeTGzNVVnOvcsC099+sTAuX6jnKQmK9Sq+ONl90RiWSWtgN535dvego7KjUQEUZ1FXzduCoRyDpVhkduqlUmEUhU68GmL6d+uVYPuVhvY44Sckq/siWdy6lK7FRGxSFCvS58Dn5nz06R5hAp6ToKBOtEzj25v8QZshc+sRb/8FxTeXsG2HhwBuI7tz+Lx1ZvbulZFh3lPgowiQByM3O/F4h8Lw4TklRDQNQoBeg3VTfFRDiugINPBO6WpmJ7uRppj4njU1VDXQUHi1duxCtbBnD6gdPd9nuLTzX8cag2AgBYronQrNTqTWU8TYP+ahA1LYlOf6WmNarxAKJSIbld8gCex1dvwl5eoi/+XvgYrd28A+N6ShjdXQxJZAuVYCeBwIVR6qYl4RjTXcTGqrvJrNsyENIpp432lfNZvX0Hy9C6dnM/1mzuxxf/9BTOmDfDv0e2IxKTgXDqcZW41xhDwCE3Q7nhr2N++9VaoBqSz6sSwcZt5ZCNC4g6ApigMl+qRFBwCHUhfIZB5kvih0iZRnwTkwj6FIlATqu4CPB//XXUBlRwyJ9bAlE70raBaksj5DtQInD/y7leyyByqaKsDls17p46T4Sq4j7K1URCiNCCKDpk3Lj6ytFDWUyEoKwQm5oQOOuK+/ExZpzcponAVKHaCAA9IShXmycEW/orRjfTe5atxx//4ap7uLG4v1ILcU+ylXxc4iSClzbtwC1LXg75iq9+zVWVmVRDR1z2V7zrivsBhCUy1etLCIGHV7rcoC9ZiYAQSPzygZX437sCD6K0mgZTgsG+cqAaesFLEz2VeQfxZ1Vm1iHCKCY9FgomQqCohuqB2i7aTqFXDbHx0hm1dZKnDn2K2k5PCAJVTlfMGQrq1hAiBANhKUmWV8moGnqtr4zbnnIjmIUQkXlTrtZb5jEE5EQIiOgUIlpKRMuJaIHm972J6H4iGiCiT2V5Nm8EqiER+h+nKajU6iExNY450/n961xKa/V6aLLwTVXd2AsF8kVPFRu3lyOExuQVonIpXAyX4yAXjEkiIHIX9u8WrwrVq/PJHkipGto+UMVHfrk4pIKR+Owfn8CF1zyi/e39P3kIz77iEiDXRkB+vTpHOz7GcYTg/F8sxgW/ehh/fcbNV3TxdUvwpq/fiYFqLbRAd5SrWLe133chlGkF4hgFgYBD9o363vwYy/z31ZzzaYPO5FzTqYYCAu5emzouHBhZ8yWCcL4hIqA3TiLw7lU3L1mPTt1SrdX9ttaFQN2zJ3FOWDeOOhucDqpEoBqLieBJBO73Li8oLo3XEFdxyWjqokP46K8exh3enKkYjN4m3MA87oQIr03AJQStlAiaVg0RUQHADwCcCGA1gEVEdKMQgpvINwL4NwBnNvBsrpBD6UsE3njHDfIFvwxeMODqjt+8zzS8buqYyL061ZBWIqiFVUN881fvLzqOcUN1CQEjKIqkEapTuc65/k19FfRVgtwsJomg6BB+89Aq/OqBF0Ncos6Lo1xLN3n/+I/VuPXJVzBhVDS9rkwDsDHBS8S1EeglAokwITC3y5T7Zq9LbsEnT9zT/95XruHQr96hbYsJQoiI0Vb+H80kgk190QCoNOAumxw7KlWo03DKmG784M7luObBF3HfguN9glRT7FdA2J6kSnk1haD5bZY2CY1EUKsHUm9dBBswlwh06jBdkJ4OKkMWkQgorBrqKrj908cRKF5DfWWUCoRKTfhSUMEhX+VnKictVI0A4KqKWmkjyEMiOBTAciHECiFEGcC1AM7gNwgh1gkhFgFQLamJz+aNYGPyJn2KiD1OBAA3I+gJ39bnzdmm8STRTQo1joAvItXTpBjjNbRh20Co/LqIEoLxo0qhcj998l4Y3VXwvUQA4IZHX8KRl//Vj/Y0HbZTcMgvf4dCSFSUq3XfvzoO9y53XR93nxJNqCU5UZNhVqJSCzyFBlQxWgT3SBdSk/E96cD177O4ij4DsTRJZIDL7fkEQLjJCZ9c4wZAjWWEQOVg09CBuRcv9OdRxEZQrkdiVRwCvnnrUry0aUfofIxaXYTmhkOkSAThsfMJgaqiFHoCAbiqE64a0hECHT1Nej8SJo8uCddYHHgLSsagXK3jM79/HMd96y6jjWDzjgp28mIt5HpX53nWXF5dbD4KRNNll6v1YU8IZgDgIZSrvWu5PktE5xPRYiJavH59NAlUWqg2Aj/pXIO6bJVr0XkN6SZFRYkj4Byfen/BIeOGurW/GiIENU/M5pCTXN63x5QxmLfrBPRXgsnFTzcDzKohdRPw+6PpY3+llkoieNXzitFtoL3e5qgmNNOBDDYCjrd9/z7U6sIoESQdUhJWDenbFJfZVYBJozWBX9z/Ai77sxtFy7lulSPnY7Nxe9lPs83nX4UlMlQjmPvK1cj48nm2ndmaavXwuQVECNsIDBJBTeH8hXAPWNIyQsxYzHNZJalmNqUkBEmJGR1FNRQYi+v47eJVeH7DduayGX62Whfo7fYYFE/yUOd51jTU/H0JEd0D0krXjSIPQqBrXVpLSepnhRBXCiHmCyHmT5kyJXXjVDi+jcD9Lic/zxuTJR+7ygXpOBGdjvS8qx8KEY2waig8AYsxNgI3Rz1XDUXbJDdveZ9DQHexgP5KDZO8045eUoKWTENgMv7qFvuOci3RRsA3zX6NzWKMl9epTxH1f3n/SmOZqvuoOsvixPak1MQcSe6jOqhnUtzC1AmcGVHbyBmFc696AB/86SKUq9FIXJNE0F+pRQiUGhxXY8ZirqMmCnsNqXNAtk3X7wt+9TCe0KR8qNbrTCII2lIxMEe8nWnwueviz3ZwHFU15K6RNLp9IYL7pSpVneem1DCmucf7LaBPQjncjcWrAcxi32cCSBsa2syzDSGwEYS5GH6EXZYDXFSxN63XEBCOMOUvXi6K0d4mWHAcP/mUih2VWoh7qAkR4fyKikTgEKGn5KC/UvPVRmnTMZs4ad2Y9ZVroU1D1hV6jhnNuaFQlic3IFUiuPSGJyNlyY1uQJUIlKZVanVzUFXK07LGdBe16rNNfeXYxHFqokGePJAPLb+PCBBsCj3jGc5dXX54bpnmWl+5FpUIqnV/Q9vUV2Y2gnpYIgDFSgQmt1OJVzWOBDzXVq0utBy0jhCk5dGSUly7NoKA+HYxiUAFwU24J4+NFRB+bEqfEkcgoap2JJLOnAC85I0aoj3cJYJFAOYS0Rwi6gJwNoAbB+HZhqAGlOm4mCyGHtUQtk0jkpr0hWRY+LL+0Z74WYyJI9hRroVVQ/VoBlDpISM3XMcBeooFl2vxiu2vpOtzJonAC0aS4KkLJMrVus+FcS5UHQM1xbEOkpBEXDYVSnDPsg3avEhAet3u+FElLSH40M8WhTzMVPjphR03spVHGYclgqDNBS8KVkVfuYp9P3+rtv1xcQT8Xjm+m/sqbEMPzweH4uMIJAE2BVHp1HrVmvDdmbmNgCODYG7EcyxdO2diBNx2RyQCg2rqnKse8N2D63V2BoNJIjDMo7QSje5dtVIkaJoQCCGqAC4EcCuApwH8TgjxJBFdQEQXAAAR7UxEqwF8AsAlRLSaiMaZnm22TXFQQ/B1XEwWQqBOfp1qwVQe54LO/8XiyPVAIoiJLK7UQqonnlxMQk5SueE6ROguFdzFrlls3DCoIs5GoDIs3GURACboJIJasAnwDVyOa29X1EZg0sFLu8aP7l6BhU+8rL0HcAN6TDrctLrdCb16QqBL7c3Bo4jVucc5Pu57L10dVcjzMTiCxIjh+//79mURm0a5WvfHd/MOTgjqIWMxFGNxxEaQkLpCH/0eqIbUSFqJtJ5Sfjs09Rcc8u0A45lX2r3LNmDt5v6ojUCbayj8XcAlIF0Fx3dTVWMrVPdPiTSEoC5EpO9CtPbI1Vwii4UQCwEsVK5dwT6/DFftk+rZVsK3EXjftR4NmVRDqpuXxjBsKI/fy1PbBtywRwjI7DXUX6lFTjpTF4TkXuTmIlVDA5UaRE90Cozt0as9AHO67kpNoOQ4kRQMnJMcpyUE9eAEN64aYm0FwjYCk/51yUuNHXnIkZYQTOzt8oPMOIoOIU4pIedCV9GJjPFBsyb4+ZD4vCQvHYIKNWcRENiodHuomtagXKv7G/wmbiNQjcUIb6LqXJRtNdlGdBJSSDVkkAgyxmQZGa5iwUG5WsfE3pKvMnrIU/PIIETuNSQhJUk12FQIlziUChSKI+AwSZZxhKDoEKp1EfIs4xjuXkNthcBrKB+JIE0EoYlTMm1ovkTgie1EcSoZEeK4dHEE/iT3jcWEnpLrPqqL+lTD+jlMXEmlFnUV7WMZLwFgQm80TiCkGmLj4fvDe4uRSwRp9fgSWRjL1IRgdJfWVTjJ+0y+G11A27F7TsHfFxyPvaaNDb0X18MlWpbOriMN2DrffXVDK1frPrOxeUfF39DrIqoa4qfymY3F6Tc/bt+oi/Q2gjiYCJFU+0wcHZ1/ckykpMuLkNWrb9QnBCw3lDomjUgE0k4hEI3uBlorEXQcIVCzjzZrI4jLKZIE07nEkmuUqqG6ELGRsNzjSScRSLWSbKvrNeS4ahnNAhzbE+XcJcyEQEQWQzrVUJBfhi8e1fuFe+j0G8YtK3QBgWltBHtNG6M1SCal1JCbZZfmfRYLhF0mjEJ3yQlJBI4X/KRCrxqSR3dG+6HOVZ4CZBOzEVRVYzGRr0LSwXcfNUi+us3vXVfcjxe8DLfqUagSWbz36nVh9PiRDMpEjY1Kvi5+spiEbm+o1uq+aoivSTXozDSPdJkHJGR5QujVn8Pda6jN4A7nkjWbveMidaocFtSSwPFnUSOp+MM/9Of4qqqhutBvMOcetiuAsCeCegQmEIi/fv53Nol1RuKxGnWRhIkpqdaj5/f2lWuhxawzFnNixAmjH3Xq57HhEkHjxJdDd051WkIwf/ak1GVyyHfTrTkYR77jroIT2oQKRFqpUicRSJ21bmPVGSAlceivBHEE9ToiqqFQOUpTTAFlEjJt80Un7In3zJ8V+Z27j6rX08L1PguX8bnT9gYQcPsTNRKpZGykJKc9v4MNwNb+KuqeRBAKAot4ZOmZlThCEEgE+vTtrUpBDXQgIZD76TduWYrv3PZsJOweCCeeisvCCZjFYYk4DtGkhhiohgmBECKigzz9wOmYv9tEAOEYgHpdkz2SXP2jVA25BrQg+ErFuEYkAk1eob5y2FNF6z7Kcs7ojMXycT5WmVVDhus6NU5a1dABM8ZriWKSaqjKbAQqJBGJ/Eaup5OKtRqJoM+XCDSSrsYlkY99yEbAxkF95yrzVEtQDcm0zWfNn4mzD40SAp5ugiOJCeOo1MJpMaaO7cb5R+/hfXOvj9dJBH7W0BhCwLB5RwVCSIkgGBfVQcOk9o3zKOtKkgisjSA/cKr6xEubEyUC3VkCHEkHaDeSj18epD7aM+TV6iKifycE0Z7yTFpAH0fgEMFxKKQaku3STdh4G4H+eqUWVV/tKFdDXJ3ORjDAUg/zDZ7rj/l3IJk4qzCpGHR9KVfrqVJUj+4uhnzr48rkkJyeTtUnOVeVEGztr2oPQd+wPaqaipMIVNXQ2s39fqK8gWotZPQN+cF7ffr3N88FgAjzVEswFsv0I6UCafvNTyvjyCIRVJl0A4Q3TUkgJmklAvnf/cCb4Qd2MpHg5ifW+u3i70nn969DnETQzQ6K0rkLtzKje+cRAuW7Vg/IiEPcAS1Asj1BpwtWcd4Ru2mvc9WQzm2zp0sSgiAzp5rDCHA9fQpE/kInjzCYoKqGDt51QlCWgS0pK2cPEIVTHwN6iWCA+bfzRaImT2tGIjBtKKq0JjyjZXfR7D7LodvUkjRLkvHQSgSFsE97EnRHoMoxzLqx8ijlupIGWY7SRSfuiVP229koEZhU+v4RjIYjV+tCz5BksRE8sOLVEIPA56kk7GN7SpH6A9WQbEtQp24z/+atS5lEED0DXcIsEYTn7v87ZS9fTSjfe70eTROj9ilvdBwhiIq50QHnqqGkSMCkI+nSJF374hn7a6+P4aohTTmSI32BHSspU/pyOJ77aUUjEejQq0gE3O0zTk/J2zi2uxg6HhEIvKA4uO6fp+eQ4yqf5gtLGpUbPYBeQp0LVU9FodPf66Db1JJUhbJfus1eqie6NZKGDq9pEv1tizEWx8FVDXmcfS2chpqPU6EQtVekrapU1GfRrRm9htKVCwAX/Oof+Mk9z/vfeZt3meAeANRTctCjvNuCohrSBXaqU37zjornPtq8RHDSvtNwxB47AQjsRq7XkDUWtxS6BFISMvslX0RJhMDkJiahCwS7/B0HJDUTAPzEVjWNjQAIJwKb5LnG6SQCwNWFShVMwYmXCLoUosO9iOLE0xKTWvrKNdyzbEPoeMjZO0Wzi3LVm04iqMdIBN8860BzY1JA3cgrtbonEaRbFrqxMHnO+L+zgDJTe9JKBDr0xaiG4jBQDc7QrovwJszXjM5wrVOv6mA6clUYCYHwf+cw2d04YeRtnuERAiB6PKu8Tz0kHgjvAzsx19N1Wwf8gDK1rRImj0DVRlB0HJ+g+MZiYYojsBJBblDHkg/4tLHeuaVsAqjpgFUkqSl0OYJMUcIqfPfRun7y82AtedpUTSNWOuSmqKiyOII4iUDNm8JtBnHiKZcIDvYM2TzIa5cJo/Ctdx0UeoZLAdvKnBCEjcVcIpDuowfMGI8PHTnbP3g8K9QhqHgZMdNy5Lqx0OWaUusA9IRAvpNmJB0pYcWlwtZhgCUvrHlBTRK8mzI1BkfaQ3NKBYNEYDAWy1J12Xh1mDYuOHGNv5tZk3oBuPmHVEIg75ObLF876mlpe3iMYl/ZPQq1VAzq+It3uphEWmNxqej4BIV7DVljcYvBDT8CYVFevogsEkGS4bKkWdTcGPnBN842PhvYCPRxBDzsf/KYbv/eiI2A3Akv+0qkd3OcPr4H5x+9e2ixvm7qGBy/99RQWWrZEvy5S07fx2+P6X7APaIPcA3j/NaqIhGECIEnhfV2FfGfb90P/3rsHvji2/aL9CcrKrU6BjJIBI0cwyn7pdvspZSWtv44pN2cJQaYjaBWD6c44Juq41BE6klTlUPRdOrnHDoLh+8+Ca/1VXC5l4o7XK50FAhXIOfZnMmj8e75QcICvm75q/nYsa/DWw/aBe86ZBZ6SnrVkPzMiZyccwRX0tt5fBBUx12wdTBpClTVUMkJvI+415BWqrcSQX5Qba58wcgFWPY4w75yNRSspUOSRKDjgPjkO+sQbeYNAIFOXYjopkNEmMLOnJ08JlANRTdfVwKQHI6UEFS87aBd8LnT9gn9dvsnjsGk0WYbAb+XSzpBpGa0LRzS0D1GMVBXlIAy7pctDfhyURMR5s+eGOlPElSmObtqqAFCIOMIYjaRZm0fQGOEQD0wRwedRJBkFwECfTqfL5e940DsM32c8RnZBjVQTJbx7vmz8I2zDsJ3z57n90GCv5vxvSX8zzmvx8TRXZFgSVXtxYdtQJEIpMZAPhdHCExSjEoIigXHf988slgXQGglghyhRgBy90+pEqhU6zj3qgew7+dvTVYNJehi49wEgXiuUtoA6gYbAfduCUkECgdF3sZfYaohXb2SWzN5VrifYfytGCIK7meVs1En81Pe6Vyqy2qlVse9yzYEqaU1qiEu5sdFvpqgLjWpokjyGpJ9UJkKbbSwmqAtxn1UIg+JIGueHtd9NIgj4Js9qRJB3T1V7af3ucbZNCYC2V/V+y2OmMpyVTWLb+D1ijp0jhvcpx6mo8PX3h62z/H6icKqIanndzUHApPHdvvznxBvy1FtHvKdql5DxQL5BMCfE0p2APm7lQhyhLqR6SSCar2ORStfA5DGWBwvEeg2XK4aiiMEcgLUhUg0FE32bQSBfvhdnrThkLtpSc6t4OjrlYtU/iY3sTAhMEsEfHOT16Ua4X2H76p9/rn1bj7+0QohuOHRl/C+nzyIXz/4IgCXw5IcolQN8Q1T59OfBJXrlXnxu0vxy0L2QVWvveXA6ZF71U3d9xqK2eyzGIsbJRpnHTITJ+47zf8+UAkfcsMZWlX9V60LnHf1Q/jin57C1v5KKnuEP6cK5vmjQnLFql5djrv6HkwSAce+u4QlEF49P4YVCNQ78vjXUoF8hkUNKFOhGosLngooqhpyMGVMIGk4FD2YRqqArddQjojzGuKqIYmsqqH/u+AI/3ORRfByFDUbpoquQmBESsPdSYlAps0oOAHX70YWO+zkM71EICd2UVm0/F61PyHXQk4U5KlodTcH0VfOPCBy/zimDlJTX7+40XWJlTYYIQJ10fqt/ZjYWwoRx0bUKeqwViUhSChL1qqOhUxqdticIP2E2q64OALTM3H40fsPwVFzJ6e+n9ex/y7j/e/cRgCEg894Lx3Pa0hKYE+v3ZpKDSWnhiohxfE3sliVEPjRwJJR8f73G+II4sCJuUk1JBM5FhzHrytJNaTaNaSXkWpTLBYI0z3bw2t9ZTfTrBChdkgmx6qGcoTKWYclgkA1JJEkEah5b94wexI+cvTuANyJqnt53JPI5L3TVXR8gqHVFyrfpb3A5V7cSU2Mc+LVmFRDBUUikJu5Kj6H2sG+hyQd73NNiFDd/P6dxgQ2jtEpVDsvbuzDxdc9gafWbMHcaWNDv00YVcI7Dp6BWZNGGZ6OQt2/0qqG1Pw0EtKFlwfOqZt6nNeQ6Zk4TOztwkmMs08LQpg737BtAGWWJoQbXvmaKXpc857T3IR9SzTHUOoQSJmKsTZmd5MS27Z+1eUyTAhkGaFEfSkN+araqx4yFrubtiSQBQoYO4dI6whigkOB40hRWf/SCP3qtjIIUffRgBBY1VBuUIeSG7qkSiB8ToBZIvjR357Dig3Rw84df9KrFgkXcRy2RFfR0eY/UfH1dx6AvaaNDaWjkBIBD59X9fy6BehLBMr/uPaGjMVskUtiV6uLSN0SPAmdGsSm6/Inf/cofv3gi3hs9WZ/I/LLdQjffvc8HLLrxOiDJqhGz5qbwTJR3eL1QR1DmdSMb6JRiSDZRhD3G6AYOB3zedZJZejtTp7nHNuIQuoTL6Bsp9EuEX9ufXT+6+sLS5tB2XGqIfd/RDWkqCzlejO1OQ7q3Ayphrx9QBKYYoGtZ8qmwuNxB9PGhb2PJCHYsK3sHkuKsPuvtIU14KSWvn2tK3p4Ii6yeHRXwUuNEEy8uPzhl/35Ge1JWHKDKBYcrTwX8q4x6Bm7Ck5wdkKMMe49b9gVt150tL8YalIiYGopImDZumDBGo3FygKT7QxJE8qM4ZthUWP7UAkB52p4NsheRcevS861ib2LuVPHRn53600/pY3G4gR7g28wVIZQej5xjxEpVb31oF0AMK+hGGKTpNbghKJUcJBhP/JB0Ad3+S7UIWNnWH3CgxbTBq4VlE1bIo5zN9oIVInAkRJBso0g0i6Hf1bcRz1pn5/1zaXsNDmpJIiCsZ3O3FABBBLB9gEQKJJ0TsYLWWNxjuBj6Sa7CouTY7qKeJXlcNHlc0kCn6C6eR5SocRIBFJCkRMnziDqSw+e+6gqEXA4jkE1pHh2SA6GL1ZdWRJhkVdvB+HPhyWCcN/WbY0mVOMqAl1Ka7cu7WUtVKnjm39Zig3byokSgeSG1TEc60k1fHP0pSrvVmkjiCcE8e3uSmAk4pIGSrg6bp1EEEiW/F6/Pm+zlJJ0UooVCVOf4vrqEwJFKpdzXf1fa0AiCKmGiEIOBJKgyz2iyNYUIVlyU+uRhGBnhRCM7S5i9ymj8fV3HghQNA11Hl5kScjlqMp2QjS/TJiLGNNTxDWepwoArN82gK6ikylk3zfSQm/p122Sr5s6BssZ195VdDB1bA++d87rcaSXi2R0d8E3Nqn0g3PgVU81FHAv4XsdMngzKRxWsInpOXr1N5Okwx/h1XKJYKyyeekMkDxq13R2cpYgL/VQ+7ufXQ9Av0FKXPqWfX2dvDqXxo0qRtoWeGJ5xvMUNoIkzo+PbclxIgRtbE8xNt0x4M5L3VjJqk1qFschz3Af5paTYNJvx/X13699FEXHifQlcN+VUgYibUmrT1c94sJeQ+G8TY7DbQTpMwTI+yUB33lcmBAQEf76yWMBAJdevyTyPiUhsBJBC8EPfnco6sZYqYlQnpE04N46ugnJF7LcKP5wwRtDEbxy0rztoF18o6oUESf0lvyUwEHbPYlAqxqK6mX1cQRhY5befVTfV8DsFmtyP53ExnU3TR4iFTpOW0UWQlCv67nnOHXFB47YzU9ZoN530MwJ+I8T5uK/3n1QpJ1yz4hLQy2RtN5DbroFiqi41D7pukNEodxQEpLR4KohbumSc6K/kkwIzjl0ln8qmOm9JL2v3y1eZUzbIYchSCEdjIQu021cGbItvDvSRiD7WGTOH0R6+58JDpGfV0uVCDiIoqntJdNgvYZyhLp4y7XwhNdtDJMaJATR8DUXfCHLtTi+txQ6OvHr74wmVBvT7U7ua/75cOw+JWwsDSQCV/1QYJ5C6gQyEQK5kUuPn5KvGgo/GyqL/cY5YZNbKR8Qrt7RHWwfB5OhLumEMA4B4K5PHxsJMuJljB9VCqXhDrsbBs+QxyH+xwl7+tkuAe7hEsSoxLXfLSu+D/zZksehc6hpxHWcJFHUpx4IchVxSTkk0Xn9kR41q1+LnpImcdk7DsThu+/ktUF/T5rXpXoN+W1RVENShfP+w3cLEeM4hNNnuOpieUkSbVmuanfLkibbIcKqje5Y6Y5I5fepkdr+OrQSQX5QhzIsTgaLTKZsABogBAYOWMKkS5ecwCWn74MDZo6PPPfD9x6M9x62K/baOWoolXtDyFjMOPo7PnkMa5OeE1MDyfReQ+a+mu5TXVcluGqISxNxR2VKmCSCtG6DgLuQJ4/pxl4764kqAPzLUXPw+bcGeYzI0BeTG3BJkbLSqYbi2x2SKDUEZYySRkFLCEDYf8Z4LL7kBHzpjKB/UurisTS6fkqJ4Mk1WwBEVXv+s776JtyGN3rqziTCTeSmZYjL1uo7Sngb90eO2d2Pq0kChQi7a/9Q2xQEYjKJAJQpepsXudfOY/G50/bWBiASonYXabcZ9hIBEZ1CREuJaDkRLdD8TkT0Pe/3x4noYPbbSiJ6gogeJaLFebQnDuqiqCieBvLcUz6RklRDEY6bqYYk5eHrQOddAwSGRJO4PHvyaHz17QcYdLueasizERQLwaR1CNiDSRAOkXYB+vps6TWkfPdq0vYVcDfzf/NUVnyBhQ1ywbMyV9CoUiEkJaVRxZlsBI2cCKd6GqltDwk0IQ7SIPUwqDl2dO6j3373QXjwc28Oys3gNVQsUMTWoRLSuOImj+nG+w/fLZK0L5RaWkP81MjZtxwU3dT4/bxPz3z5FPziw4dGrpuwbaAaOj5V9lZlQuSGnYVzLih9q3kBkBxVjUTgULxbtwrepp3H9eD8o/fA9889OHIfUZDU71Mn7Ym7P30cuork/9YqNE0IiKgA4AcATgWwL4BziGhf5bZTAcz1/s4H8EPl9+OEEPOEEPObbU9ye8PfQ0EoTD/HE7pNGh3PXaiuj3IzYnQgnOXQwEVL17WGNjNmI+gr1zCqVNAuQvk9NvWF909KR6aNPNoXB584cU+svPz0ELkwSUgOEX7x4UNx+yePCW1uaSQwk0E3m0Tg/lfHO2wcN28sIeKe0nitcx/daUy34lse3+6Q15BGNTQupWoo+Ew4aT/XAH7CPu5/7k3Hn1YlAgA4ZLeJGFUySATew3xMe0oFX5JJs7cNVOuRyHNAIcRE/thmIQQhjzhvLFXpwz/HQ1G3pjn57r/fM89rE3DBMXvguL2mxG7oBPjM6KiuInbdqde/v4UCQS5eQ4cCWC6EWAEARHQtgDMAPMXuOQPAL4SrVHuAiCYQ0XQhxNoc6s8E/g6ECNsIHCc4xYtLBJzDcnWD4TK7ig62s5O2+OSS9YWSvhlUKJJbzLKZSXCOs69cxZjuotFGQI4aCOYuIrk4ZTuyBpSFN5fgs2O4XnAIR82dAgB4ZUtwEPtOKcR6k7E1i41AcnTRJHvBZ4I+TQgQDgwyvTK5cUp9su5gGrXNiRJBkRMCJ2IsViOjVTVdXUQ3lenjR2Hl5afj5sfX4vanX4lIyn5bfUIQzPcCkdFtl+vUdUhj3B+o1rWu0yEGwwnyWmVhnCOqIeUsBiCIT3ClbLkpk9FGMKpU8D2duH5/wal7p2hQMEckrxPkVkrXp0aQh2poBoBV7Ptq71raewSAvxDRw0R0fg7tiYVqvlVdziRXwW0EPMulbpGqqoUCm/y6BGUmd8xqrXGJQBKPuhDYNlBDb3fR6DVUoPDBNCY9tiQM6oIL1cv7wq6bVUP6ceBcru6QcRUmG4Fp7CaN7sLDl5yAg2ZN8K/5KgaVECjEzRSj9pUz9/cPBEpSDcm65HwLzamYMdWWafDU4m02lWdKjyHhezdxiSBEvKPR9/FSk14q1bXNhHK1Hj5HQMi2hOdShZ23kRYhIumQZ2MLb/A8a6+vbnWiAYkSvH75ftK2icBSWigBncNaNQS9xKKOUdw9RwohDoarPvoYER2trYTofCJaTESL169f33Bj1flfUVRDcqHyvOV8Euo6om4+gWqIT5ywCKpDzedQs7+WIKgG6BuoelHSek7CofBRlX5KCe+aHAO54RBrTtx5BFEVVPQefoeuDQD8gD6ZNkM3/00SgWlsHXIlDa6SEQaJQG2vabMa21PCew/bzWujgTAVpETgfpebDOdwoxKBtigfvO+kEVHVI0HD0pmcm3rIuVcOuY/y393/A0qCt7hxd58z/K65rgZQVWr1yMli6rMyPbbb3vQbZlhtGQRkqvUDahJJ0p4ipqKUMQaA2wjUaOzhLhGsBjCLfZ8JYE3ae4QQ8v86ANfBVTVFIIS4UggxXwgxf8qUKQ03Vl2wFSV17bfedRCO2H0nzJ4cLCYuauvepzrJucpBTkpOLEyLwhcJG3grknbUPBtBbxdTDSkLgwh6iaAgXRwV1ZCBo3e/h8sN1xOdwKYoZb65SXWdPO5S5++eRjXU21XA/jOki6R7nUsechlH89/wz/rocPVe0y2ynXJOyLHlx4yq0k0S52dKdvbew3bFM18+JXRGLxAe81P23xkAcObrVaHdhWR6dhgyecr+8N8LTtj54Mx5u+Cyd4SzzZrGUHdddSV2JYJgvEzG4oBz19elrz9chpv5U08IwtH60ZijSAPBovNTEwImEShxEsNdIlgEYC4RzSGiLgBnA7hRuedGAB/wvIcOB7BZCLGWiEYT0VgAIKLRAE4CsCSHNhmhTpKQjYDc7KG/Of9w9LDFxvW5Om5D3dil7vjfT5gbshH49xte6D+/aXeM7S7iTa/LTuiksbJWq2N7uYrR3QXj5FOT0Knpp+d7G/CHj5wTaXtEujBw+/y7ySbC93K+2X/t7fvjP9+6L/byMozG5UVSUWCbqhDRjSikStGoGNx7WZ8ofhGHPMR07ZE2Au+79GwZFaNuTFrv6jnYR+/pzpezDpmJnlIh0h/+bfcpo7Hy8tONJ4PJdvUxmxcvQP5eUVRHvMrPnLo3zjnUO39CcrNGiSF6XTV2l2v1yBGTbpnB57jo9zio0lJNRI3API5Arn8i4KPH7oHPnLI3DpgRdfWWKCbMDxU8jsBXK/sahtahaWOxEKJKRBcCuBVAAcDVQogniegC7/crACwEcBqA5QD6AHzIe3wagOu8F1cEcI0Q4pZm2xSHyAlloZwqeg41NIc1b0PdlI7dawoeufRETBzdhZseX+uVwcVYfdsOmDkeT3zx5KQuaMFPBOsbqGE0MxarKDikjW6W16aO68HKy08P2hsjEYSImtboaU46FyJG7CDwmRN78aEj5+CyhU8H7VJy/6WRCARE5OwA/lzd99IKl6XzsjJB/mTafEq+asiTCDRxBHGESF9muL277TQ6/L5iyksypvf4hCAI4uJrpkfjvcNjVtT6Ag7aQLg119XjJMupjMVmZiUOKpMjzx7gqDKJgFh/ekoFfPTYPXDLkrDPC39arqnUEgG4ZiBMRFoZUJZLriEhxEK4mz2/dgX7LAB8TPPcCgDpQgBzAtd315SXboqGTRr/CAdG5B9SIn8JqYZa8EKlCNpXrqFcq4dsBKqfuUN6F1aj3t2818dKBLoYCtMY6+p2YtqVJsWEEMEGLavlahWTsVi1e8QSAoQljkg7PSKj2gjSqtt0SMqDr84vU+yDDlJlxSUC/ohuQ1bHiJTfdG1Sb542rhsXHj8Xl16/JBIHUanVQ6o0nW2nUYlAnY/y/YztKfpnkZR1kcWsjDhTQWCcT9ceonDcAhB2PmkVOjqyWA3l5hOebzRJlDjOBc5XDaU8nrJRSEKzqc9lnV0bgb4e9YQySSiM6pYYg7Dqc8+h4wZNREFnBwgC26LtMhMt/aYqr4dsBIY4ApXwxb1+kx1GQs4j1UYQSs1RUOsPvr9h9sRImapqSEWcO2zSXNZt9PwRPSFQGafo+zZW672Do+dOwYGeiuW4vaaGbhmo6g8LMgX0ZdkwVdWQ3BN4riIeWSzfG++jalPgbqU8bXXKFhlVQzbFRI7gg6lmFA2H0uuHRvcqTNwpv79RjiUtpNgqz09wbQTx96e5BugXtu6ZiFEa0QnM7wlJBMVo3XESgTGgjF0WrG5fIuA2BEg3PWXjVL2GYgYyyU/edx/19oaqJno8ysFHy9eVaUJUQuW/xT6qJwTsnY3SBXYRKW1mnxM2MbmJOkQ4aNYE3LfgeN++IGE6PtTk1pwpslghJpL75+op6VBSdBzt+46TCCRRSLvmiZizhq8ashJB7uCDWVay/PHfwmmU+eYRRZyqJ8lvOy/IbJKbd3iul91miQCIqlCAdG53EY47hhPzdZtO9JraBh3hjROJjYSa3yy4Dt97LoVEoOqe496vX77hdz+gzPteYz7pvA4Ok/pMIukoS3Wzj7PxqNBu9Kw8k2rIxOioXK0KuYnKOmZMGBV535VaPUT8dCq9UP3amvRQx0baA3jupMBYjIjNCQg2e/le+B7h9y9lowjRwNJWqJJVdBwh4C9QTaGrns0aPBNfZirV0CC8zFKBAomgqxjLQWgJQYomRoPTgs8RjyKNfti0Kek4fNV9jiPN6VACIiKa61RDsV5DiH//Sa596rnTVcUQCOjiCOIJQZwEqj6fpjwO/SEowTN6f36zZ1mSaqim4Zj5vbW6e5C7jviZ5lKmFBOK9CL182OYnaLiS3GONt2DnEefY5HD7zt8V++3bGkvQjYCf+6G62kFOo4QcFRrMTYCxgaFOGXNy1Ct+2HEc0R5olhwmI0g6j7KD3ZvtD2mzR6IboakuR5y+TPoldVrcUn24uC6j4bbGVINifCCC9rIvhPF1pU0jEGKCfe7zlicVaefxFSo0lKSqil8L0W4/pCNwKgaMkgETvz8DzZKfRvvWbYBgJ4QhIhPgzaCsEdVsOlzg7XPMBCLI2D11X2JIBibr5x5AFZefnoDEgFFvIZkXarTR57oOEIQlghUbxomERhO2NIhlUQwCISgVHD8c325+6icyNf/65G47l/f6LaHdeqiE/cEEE60Z0KcakiFrxoyLPKkDS0wkiU2S4uQjcC7lqRiUOtzKEEi8Ber/vedvFQl8uyFqiL26+rnfEcWW07QJuV7jI1HB3Wz58/0GDhz9WwGtW6jjUCTKC7JLhJIcsHvcU4LKngesahqKPAaUmH2Goq6BKu/pT8xLSBGqn2llRJBxx1VGbYRxEkE6RePSV/Nnx0MQtBVIGzY5toIersKkU16pzHdfkI33p6zDpmJsw6ZmaqOiLE4ZgHqOPqwKB5fVyFhE0kD1UbAN5RPn7xXqJ1+vYoBPD6OgPz7dHjnwTNRIMLx+0zFT+9biWqtHuuuCiSrcpLsTbGELcU8jEgE/IQyjbW5mTiCgGOOX2+JqiFD/Trc8h9H4bhv3oWtA9WwkdlhNoKe6EFJofMIQjYC939JSScCwM+aqkZ7m0BEEalRDnmWtNdZ0eESQZyNgA9NtoUXKnOQVUOSuI3pLsa2ulHPpaRNk8PRLBr+Oe6ULiBY3I3SASFEhCOVG8r3z3093nGwnvip+up4r6HgPh1KBQdnH7orugteRG49evBJrE6/AdVQXPlpiKoaxas+ohrXiaJ2FbU+0xAGHDMvL3pzty7OJEQw2fP6qnxMHtON/bzUI2r2UXlWs+6kQpcQROek7IPOvnLgzAn47tnz8JUz909oVYCqn6XY/W4lghaAT5JY99FQHEF8mfz8gUh9UiJogqtNC67/7u0utsTfLC4dglEiMGwSOi5vb3b6muxOlgXA2yBYGwKJwP0Ql+FVbW8qY7Hxd1YQXBuBauiOzXWkoZVJ8zFCuHh5KaZEVDWktLdAoYh8NaBM9znJfTRpfXD3Yqkr548knQpoQjRfUcBIqXCTzrmfeQ1yJIJMs+EJe8Y8fV4nHYhYbiNl7FpIBzqQELA3qEoEJtVQ2H00+jpS6ckHyUYgoXPzywMR1VBIIghDfjf5eKv+8I9cemJoE2p2zPTG4uQkYAWlvXH36qQeDpUQ1eoi8m7UfprScJjuVxElLNk2yaS5U3Ic9CNYO+p5BDobganawH00vl1dhfj4hjyMxUQ8BbtGEnNIS9j8VCWNZItUQBR/Hkir0IGEIBjUeGNxfOppDjXwQ1dfI2cMZIXc5EZpEo81i6PmTsYxe07Buq0DoetpbASmjUiVCCYqJ5PlwQn5DLlCCOIN/FzMid88054exX9XT2CL0+lrjcUJO11cyop0qiGz1xCgiYR2zO6fSdNQpxrSoSuUPjz6e9aATZ48zi+DSQS6MXZCNoLgukxQoHqINQJC9KQ1fwyt++jgQJ0UEpyj1U7CWBtB8j15QS7Q0d3hhZzH/PnlPx2Gfz5q98iCDUfhqpytd49hk0gijn7mzgwrSy1RVd34aYEzqIbi9pUkjle9D4h6Z2WNI0iqK874nCbnTfSEs3B5qiSnBpTFBcupkK82iUAlHcDTrD0JQOJxlw7x9820BIZzLRoBURB0aN1HhwimAKekiNu4DS17rpHGIReozJPeihrVfqTLNcSuGVRuOsiym5II5Ph7M113xoIKVZUVt8CTVENBO4LPU5SjOGO9hmKkTBN0SRB1ZZsQMQYrvx84c0LouzpGWVRRgftofJuEACb2lvAOdo4C5w+yepjJ23gZ3GNHR8BMti6p3spD6idAIxE0L2kkwRICBiMnxj9q3nXciWLy9sFRDbl19Ha1TuOndiNuY9flGsoyDH7ZzYjailSSRjWkSjDxNoKwxGFsB7sjIhFovHBMv6VBM0nnktoDAF9/5wE4ad9pmDt1jF8+6ZdLolfV2w+egcljuvGe+bvqb/AwUK3hkc+fhG+/Zx723tn1+OnpiqbyTjtc+nUcbvdTXzoZh82ZxJ4hLWN37mFu21WVXyNwiPxDf6S9LA+GKLHeFpbdduCTg2fDDOcViT4Xt8n7XPEg2ghGa6I/80JEIohx2yNN3zMdI5iD/7TKsachBGomy7hXp9MZx90HRAlBxMnHIEGpv08f36OtK2sSOxVxxmvAjUe58gPzsceUMf79puyfSdzszIm9WHzJCdh1p97YNvVXAuP0d8+eh2vPPxxTxwb991VDGeVgrm5RE9D1dhVDzgumPn78+Nfhua+d5nsaNbVhs3LlediB9NI6UtBxxuI46Bbgp07aM3FqBZNQV2ayKiIvyNiHXtMRegrm7zYRb9xjp0x1RHMNxamGolxaJonAMGbTxiVHQKtlyJK6PDfEtBKBmj7BXH76jklCcPUH5+O3i1ZFyjf55AfXCIsuPkGb7gHQbOTcuyZFM1V1vKn73M/dlGokLwbo5P129j+P7i7i8N3D81a2Oe0y070vXR9M0mxIYUBuZHUefZUljCoVfMIi52qKI5IbhiUEDOrGI099umfZev+a7l3EHUcnL6mpjn/6wTdEzmZtFnKTSysR/P6jb8xcR5R7jd+03GfM6qP4uqIi8U8/+AYcu1f6ozxVYjR36ljsOW0MdovhQLMlLTO/ew7++zgvavX4vafh+L2nxdYvP5+wzzTsMWU0fnT3ChDFpwOJJXIpNqs4QhK6zsbWNGZJqqE0+MY7DzQSPYn4fF/poLPNhNRcCfYt//mmVJluGdPGdUeIkY0jGCSYJ3P87ErjiqhKBMftPVV3e1OQEoEaDJOnRBkXBRvhbCXHGDIkpq8r8BoKX4sjJqNU+wiF2zlrUi/+ctExsfWq3F7ae+PvC6sd0pYpPxcd8jeCpCqbVQ3FPa+7z2GBVirycJJIyrbK68lanzq3gvLc/6qhXX7T2hhy6KssQaf2aqW12NoIkGxoSnq98V5D4TpaCan/9o/JbEGVEWNxaNMM/6Y3FjciEQQLIMno/vbXz8AClg44LcfOYcqNpENarzD+a9Jc0KXkKBTI97BJ6ksaL6csz5s3efnf7FmVh7okzdpJa7SX0DHvOrfXUExHSPqN1iSPwW3GzVNWMYWpP62xeJAQFxDm/hD/vKr20T06GIRAGpPG56xy4ohy/WbVkPwpfGZB+nHQSQRpEq5dcMweANykcmndO7k6jRIWPEda1Ucop03CqtMR11CkexO5r/JUDcl3WSAylpvHtE86kQ3gzFxWiYCnygD7HC0vLFlp2pCLROCWsRPzQBoEgcASAoCnHdD/njS5UkkEg2As3l52D9uWKY9bgQjXH6NG0XLLmThz979JfI/DystPx8eOe53WYK3DE184WVtH0mtLbSPQPJNUpvsc+W1Kuw/E9bWhOAITf5TKRiA59cbnfyqJIKONQMcYhL2GZLn633X15OHvHxB+5rU4CExkxxMCorCIq72Hfda5cKU5qjJOasgLfWXX/3jCqOb9mU2IJJ2DeXGQfz29qiX0fMJiTVWG/z9h8+WbQAoDePBcuvJ5V5JVQ6x8JhGkPUnOdEAQLy8O0cSChnXBpK2kM6SbUZekicEpsLZkQUg1pDEGR2wEMXtFmqjtJHDiyut122ptBC0DFzuNhCDRWOyWodUbDqJEsG2g9RKBKqbv8KQQINr/wD4SXGskMyQnvlnHUY0sTvdM9Pmk8rNszlkkguCEL4dl3WyCu05jLI6ohvSQ1wsORdJSSORjLE5+eTqdfhx0t4UykWqk2UR7UQ6x/HxM/bZoJOO8kQshIKJTiGgpES0nogWa34mIvuf9/jgRHZz22VaD58Q3zVk+AbTuoz47ons62wRtBn0DrkQQtRHkN4PURHHyRDRApzbSLaYMhEDaCDTX0iJJ2ourF8igGsrUpvQSgfwYkggy1BUtuwFCYHhEvheHomcYwP+tedVQOokgysHH4ctn7I9T998ZR7B4BL2NAF65btk6KVdt53sPi4+SjgOXsoJrzaucktC0+ygRFQD8AMCJAFYDWERENwohnmK3nQpgrvd3GIAfAjgs5bMtRakQn2gKaM5LwxclB4ESSIlAEoI8OBQVXYoKQJ6R7NYXhs5Qm4VB1OlcsxOC7GOQFNAVvlfe2BihSaxfqhYbIIB+AFJG1Vyc+k97f5xEkAOrmabvWVNM7LpTL374vkNC10JuzsqhMNHEgNEyHYew9CunhLISZIVOipZ1D/cTyg4FsFwIsUIIUQZwLYAzlHvOAPAL4eIBABOIaHrKZ1sKLnaa31/8xpAmxUQrNmUVk70gI915q3lBVQ3FSgS+oTO41ghnzhdA1pxNcTpdc73B57RqnGzxEfG/69QRrkSQLkEbYOaM0zAkavvSSEVJEkEzMNkfQvUEFLnhenQZVE3zx1RLdzF6RGwWNOty3Sjy2DFmAFjFvq+Gy/Un3TMj5bMAACI6H8D5ALDrro2LXiq6Cg76PSNrUnQkoFeypDmqcjBUQ1e9/xA88PxGTOhtnbFYVQ1t7iv7n7OmoU6CievKAp3xLe0z7ueke73/GdqUTFyi5RcKBG+aplJ/mO5oJKDMhCCFdDR1td+OHDaxuKSO/j0NvGcVfG6pBxj5TgH+99YsaN2acQbBRpAHIdCNiNpk0z1pnnUvCnElgCsBYP78+bkNSTjdtB5Z0/6Gn3X/D4ZqaOq4HrztoF1aWocqEWxmEoE6gM3aCII87AGyGovTxhFwxOVPipYvOcfwjVd9YD5eeq1PX36GgDJZPrcRpEEae5cJaTbecJmEboNEQJEP2ZHKRtBgHAEHtxfKfcFkY2r1atalu2il11AehGA1gFns+0wAa1Le05Xi2ZaiVHD8t2oUp7lEoHkXcZOieYF1eIEvluP3nop/ftMcnPvjBwFE+yi/N5qGOvAaYtcatBFkU91wQpCwaSv/JU7cN5pDSG2T+fdo+WGvodjHvXsIkoTy29OloU4un8MhQo9BIvDRxB6WJcVEMwIIZwpViSCadqM1K7quUf/5DNEwlwgWAZhLRHMAvATgbADnKvfcCOBCIroWrupnsxBiLRGtT/FsS1EqOP4kNW0WSfr9nlIB73j9DLznDbMiv+n8kQcbeU4gLhFc/cE3hH6L5CHyDW7BNTkOsxPSDgN8Q9anCk4DnRdG2meA9AQ8y+ttJMVEVq8h0z1ZgrMkkuaPQ+n0+I0inUQg29J4O4oaicDkHtyq5eyr27jh2pcIWoemCYEQokpEFwK4FUABwNVCiCeJ6ALv9ysALARwGoDlAPoAfCju2WbblAV8AjfqNeQQ4dvvmRf77FDQgVbUqdoI4urz7SPKQv7Fhw/F3tPHJtal8xrKaixuxL0zi0QQJILLoHpK9Brin8l/xh+GNDYCE1OTRjWUceIkJQJ0K85UpFJ++jiCZhBWDclzK2QbwvOoVQZc3dGdQYqJ4a0aghBiIdzNnl+7gn0WAD6W9tnBRKngMH/oBglBzDwdTGPxYCCO81PHyaRfPXrPdGmkdXEE2Y3F+jbEweQ+qsvhJITmxgzl66AzVmeXCIK75s2agEdXbUpVN5AljqD5ALc0yBJH0Iy7aqkYVQ2ZjsBsVY/ldAq5jw7C5tHxaahLKQLKkhOPJW+Og+ECNhiISwAWGacmPTl0nFB2Y3H2DUJn3H7gs2/W5sSXm2GjNgh9/dHPWYLc+D3vOmQmLj59H/zs7ytT1e3WmU011OqNKpONoIktms9t/2xrb+KouYxaJxFEY5raQjXU7igVnWDwDRM62YXQfEMQRzAykE015KLRjULrNZRRFy3vzuQ1pNl0dzYcCxlw6VkkjqTf9TaCwPibQjXk/d9vl3FKGpXk9qV9X2mk3Tw2r2IG1VAzNImPk1QTFQuGclttI9ARguGeYqKd0RWyEejvyeLloWIojcU8PD4vdGVwKWnEdZND6zWUsSxpAMzylM5rxwTfy6NBiSPp9yCOwEG9Hr4WB9O8S/MuomcW6+/LohpqZgpmUg01Mdm52rOY4DXUKolA5zUUpGMf5jaCdka6pHPxZcRNCp2oN1g4Y94MPLZ6Mz590l65lVmKlQj0m06jXJpuAWSVLno9dU6W8PykIwk5GpEIsmUf5RKBrCsZgSSk1J1iHqobb7LXUHKZzWxhaaRAfxnnJBFIomA6q6T1NoIog2pVQy1EkRmLzTaCeKQL+c/SqnzQUyrga28/INcyY43Fyndfv91g57WqoQYJQX+lnr5eDUduQtLc0SFLriFuI8gUUGRQSeYZWdzSnYkhTe6eQDWUj41A1ukHqvk/qd/zhWR6QkGFgxBH0PGqoXTuo0kcXApxe4QYCboL5sAhU7KyRl37NGEEmYnKKJ8Q1FI/k+U8gkbE9WzZRwPOVKc/NpahKyz6Vd++FkzWZkpMlXSuATdhFSHtgFenqe5W5Q7TSwSSIRreSefaGl0FJ3GBNacakmWMDErAXexURLrYpI1Cd1Zr1k1KSgQ7MhCCkGomob5AIshPNaS1ETiEPae5sRczJo5KrMP3cjFcj29f4i3QFt4iDFaKCZ20qzsuFWihhK8JbtXZyvJGx6uG0nhUtKtqqBWIdx8NIzfVUBMrYFTJneJZVEMhr6GEe7NkBPXLz+B8EHgNOfinN83BwbtNxCG7TUysoxkbQerNVHktE3tLeI2lJW8GR+85BXc/u95tTwaX12bWmW5uF42EoDULuq5RDcmPVjXUQkxih0QnHdJtQqxEkLKMdkGa85n975qUulmgkwiyQkoEA42qhhKaPnWs61a63y7jUpefpF/WHe1ZcNwD4tMQAV6GOqfzdB/16/L+3/OZ4/GPS08M/dYoEf/Fhw/NdNKeH/mbk41ALTfQ3bvXWywQDMqJhhwdLxFw/3AyLNDkOILkekYGGUhaaMqm46fubayuPIxko7ubUw0lbSz7zxiP6z92JPbPQAiybLQ6r6E0MEkEqU4oa/CFjekuAt2G9jRQ5k0ffxOeWrMl1b15xBHoXKPV6HZZfMvdR1lTAonAuo+2DNPGBTPXfHh94xJBlnvaHSaJoNGAsmAR5qEaykII0quGADeFQxZkmQtBHEFGQmDgXLOc9pWELG+lkU1s5sRezJyYnJwQ4EdVZq7Ghy6C2awaaryeOMQGlLWmSreOFpbdFpg2ricxTUA+xmL3P1dFjTREbMXehUaJYJBiArjwuNdhpwbGrhH3UY5W6IKzEEZqUCKQb0NtfitUQ8MBeWQf1amGHAMz0qoh0hGCwbARdLxEsPM4phoy8H/JhCBdXaZ8NSMFxoCyht1HA07oUyfvhU+dnD0wrhGvIY5WrPdsCfDc/w2n346cGpdCem2F+2iLJeJGssyq0KmGTBJBqyAlJ937Hu4H07Q1JvZ2Jb7kPOIIAHO+mpGCiETg/W90X5HDOq6JM5ibJbytUOllGY8gjiCb8E6RDy6y+OSnrivm9jfvMw2Hzp6ET5y4Z6YysyIXY7HGNVomnasrkn2rEDiXBNfkHLASQYtw16eOheMQdp3Ui2XrthlfcnKSMPNvraTiQ4mDNHpxda+S49LoZtpdLOALb90Xx+w1taHnAaC3q7kp3oqFn2Wzal4i0JcXhzxVQ2O6i/jdBUfkVp4JASFovAwdsQ2EhMFZxzrVkOzb6O7WbdcdTQhmTx4NAPj1vxyGR17chJ6S4QButpxO2Gcabn/6ldDvaUTpkRJQBgDLvnqqdnM3qSGacYX74JFzGn4WyLapje0pYmt/tan68kbjXkN6DjlPx4ZWerFkhe587KzQB5S5lGCwuhoknQvaMmVsNz576t447YDpLau3owmBxNSxPTh5v52Nv/M1+P1zX4//W7wKl97wpPZ3FcNoreQGY1CZwTA51DRwxoRROPewXRPvu+2iY/DCq9tD14bc26tBicCU/ryRg2naAXmkmNAxa76NQLneqnXtq4aUJfaRY/ZoTYUeLCFIAzY/ekoFTB8fDvEfSdx+M9DJCMDQb6b3LTg+1X07j++J2HGG+tX6EkFm99FmvIYyVTUskIfXkA6yPD/it8URQVnySeWJNnzlg49IdGZEF24JAaDzGnL/tyOHKTHUr1ZWn9191Hs+QgjyUw2Z6hgK5JFiQoeh8hoa7D2lIyWC/WeMa8iFTyKqd00uYxislZbDZJgcDhtFoxhqIi8jo7Mavc3uo2mebb8XloexWFtuQRKCQTIWy3oHmXnqSEJw08ePynS/ujDU/Oip0gK339rKjLxzDQ0HDHXL37zPNPzho0dglwnJGUc5yECE8zxNbDiZv/I4j0AHU76rVvVdzWk0WLCqoRRQ34nqXdDG+1yuMKnQrGqocZQKDg7ZbVLm55rRZWftc6v15mmQR4oJCe4a7avklDiCVhuLBzvpXFOEgIgmEdFtRLTM+z/RcN8pRLSUiJYT0QJ2/QtE9BIRPer9ndZMe1oFlcsoFtJLBMPJxa7ViKjQUhxuPvzRno0PJILs7c/KVQ+HWJk8ziMAgGe+fAp+z+IeIknnWjwd6nWZdK6NCAGABQDuEELMBXCH9z0EIioA+AGAUwHsC+AcItqX3fIdIcQ8729hk+1pDZR3ooaiZzoxaiRD7aT3vZ1VQ+1KxEj53wqM9SK+u4tDnzYlrznWUyqE3KMlIchy5nUzGKq09c3aCM4AcKz3+ecA7gLwGeWeQwEsF0KsAAAiutZ77qkm6x40qO9EDUWPjyzuHERz348AG0Gbtt3kPpru2fB3E8e/4NR9MGNCL06JicEZLOQlEZjKHTyvIff/YDMgzUoE04QQawHA+6/LBTADwCr2fbV3TeJCInqciK42qZYAgIjOJ6LFRLR4/fr1TTY7GyKqIcVY3KZ7Re6IGotdtOqg78FAu77aQCJonWpoTHcRHz12j0FXY+gQxBHkW65c6yoxbJU6bKjcRxOXKBHdTkRLNH9npKxD1yM5ij8EsAeAeQDWAvgvUyFCiCuFEPOFEPOnTJmSsup8oHZAVQ21K9eYN0zuo+0sEbRr201eQ1me9b+3ATkM4gjybatkYvx08i0eC5ncbti5jwohTjD9RkSvENF0IcRaIpoOYJ3mttUAZrHvMwGs8cr2k/YQ0VUAbkrb8MFEsmoozlgsyxj+i6lZGNNQt3Hf27XpwVGV2aG+r+FgDE5CoBrKt9xAIhgcyLFuN/fRGwGc530+D8ANmnsWAZhLRHOIqAvA2d5z8IiHxNsBLGmyPS2BujDUXDuddHh9HKI+6+7/diYE7QpfNdSIRJBrSwYHwRxrjY1gsCiBGCKJoFlCcDmAE4loGYATve8gol2IaCEACCGqAC4EcCuApwH8TgghM7Z9g4ieIKLHARwH4KIm2zMoaCSgrBMQdRryuLQ2thEMB/13IwimZPb2z5rUiwm9Jew7fZxXwvAfg1ZJBCavoZYnnWsnryEhxKsA3qy5vgbAaez7QgAR11AhxPubqX+wkKQaintn7SBW54WRKBG0a8uDNNTZn+0pFfDo50/Ct/+yFE+t3dIWc7jluYbkhVZPCJt0bvgiWTVk4wgAcw6mtiYEbdr0INdQZ6BV2UcD91HVa6g1CM4jaFEFBlhCkALqO1EzQbbzRpcnTKqhdkxrLNHu77YTnBSA1sWsqJHFrcZQJZ1r4yU6eDB5w0h02sE0JkTGxdFfbye0a8ub8RpiheTSlsGAv3G2yEYwlCeUDQYsIUiBROKc5p210aJqFNEeehJBO/e9TZvejNeQjzbiYlolEajSf8tNBNJGMNziCCya42jbZyk1j5FpLB6ato/rKWLmxN6Gn28moKwd4Z9H0KJyVbQ6meRg2wgsIbDIDdFcQ971Nt6Mhsp79PEvnNzU874XTTNbYxu9uFa7j0oMlpqzrdxHLSw4TKkJ2vs8gvZsexsPeUNoVYoJde5+9tS9UanWcdK+rU20ZwlBGyKO6wpylHQeRoT7aI5l/e4jR6CrOEhmOcV/9LOn7o0XNvZlKmLCqBIAYLz3fzijVUdVqgkmd5kwCle8/5B8K9HAqoZGKNp4L0wNdcMPcg0NRWvyQZ5E7NA52U8aaxRqwoWPHLNH5jI+cMRuKBUdnPOGWck3DzEKLTIWD9XcHXZJ5yySMW6UHUYgxljczpSgTZvezAllEsWCg/cfvltOLWotfFflnMuV4/fBN87OueR09Q4W7A7WJFZefnrCHZ3jN2SyEbS1aqhNm65KBCMdrTqYBkizxvOHDSgboWiHxF3NQu1j4D46BI3JCe1KxJo5oawd0apcQ0MFm2JihKGNYnKahroIrbF46BBIBO3ag2wIjMUjo782stiibRHJNSQNeG0sErTrvtJxAWUjTiKwhMCiTWEyFrdziol2lWb8NNRD3I7BwkhQQ3JY99Fhim+96yAcMGN85ueCoypzbtCwhJqXpf3dR9sVLTqwa9iCiOBQ+xJuFdZ9dJjirENmhr7/8L0HY93WgdTPj4zpGQ+TRNDOett2bbrf7g6yURWckWMRse6jbYJTD5iefFOHQeXGJFPT1ikm2nRrke3uIDoAh6itmY6hhLURtBjtcMxfXogYi0eAaqhd2y73w07yWis41LYS3FDDEoIWY96siQCAPaaOGeKWtB4m99F25tLate0+IeggRuSMebvgiN13GupmtCWsaqjFOOfQWThij50wZ/LooW5KyxFRo9AIyD461A1oEL5qqHPoAC57x4FD3YSm8Y2zDsRvF60a9HotIWgxiKgjiACgSzHhoo3pQNuqGgKJwKKd8O75s/Du+YOf5K8p1RARTSKi24homfd/ouG+q4loHREtaeR5i/ZEq44PHEy0q2pIotUnaVmMDDRrI1gA4A4hxFwAd3jfdfgZgFOaeN6iDTASj6psV0gCZsmARRo0SwjOAPBz7/PPAZypu0kIcTeAjY0+b9EeULnn+bMn4rQDdkapYAnBYMMfcUsJLFKgWRvBNCHEWgAQQqwloqmtep6IzgdwPgDsuuuujbbXooVQbQFv3GMy3rjH5KFpTJO46eNvwt+eXT/UzWgYTgd6DVk0jkRCQES3A9Ad0Hlx/s0xQwhxJYArAWD+/Pl2dg9DtGvwlQ77zxiP/RtIKTJcIKWzen2IG2LRFkgkBEKIE0y/EdErRDTd4+anA1iXsf5mn7cYRrCmgOGDDswwYdEEmrUR3AjgPO/zeQBuGOTnLYYRLB0YPggiiy0psEhGs4TgcgAnEtEyACd630FEuxDRQnkTEf0GwP0A9iKi1UT0T3HPW7QpLCUYRrBeQxbp0ZSxWAjxKoA3a66vAXAa+35Oluct2hMjyUbQ7ujEXEMWjcPmGrLIDdZGMHwQvApLCSySYQmBRW6wdGD4wEoEFllgCYFFbmj3dAwjCZ14HoFF47CEwCI3tHNyuZEGKxFYZIElBBa5wRqLhw8cP9eQpQQWybCEwCI/WDowbPCmuW5qj90nj/wDkSyahz2PwCI3WBPB8MHZb5iFE/aZhilju4e6KRZtACsRWOQGSweGD4jIEgGL1LCEwCI3WK8hC4v2hCUEFrnBkgELi/aEJQQWucEKBBYW7QlLCCxygz2S0sKiPWEJgYWFhUWHwxICi9xgBQILi/aEJQQWucFGFltYtCcsIbDIDVYisLBoT1hCYJEbLB2wsGhPWEJgkRtsQJmFRXvCEgKL3GDJgIVFe8ISAovcYAUCC4v2hCUEFrnBqoYsLNoTlhBYWFhYdDiaIgRENImIbiOiZd7/iYb7riaidUS0RLn+BSJ6iYge9f5Oa6Y9FhYWFhbZ0axEsADAHUKIuQDu8L7r8DMApxh++44QYp73t7DJ9lhYWFhYZESzhOAMAD/3Pv8cwJm6m4QQdwPY2GRdFhYWFhYtQLOEYJoQYi0AeP+nNlDGhUT0uKc+0qqWAICIzieixUS0eP369Y2218LCwsJCQSIhIKLbiWiJ5u+MHOr/IYA9AMwDsBbAf5luFEJcKYSYL4SYP2XKlByqtrCwsLAAUhxeL4Q4wfQbEb1CRNOFEGuJaDqAdVkqF0K8wsq6CsBNWZ63sLCwsGgezaqGbgRwnvf5PAA3ZHnYIx4SbwewxHSvhYWFhUVr0CwhuBzAiUS0DMCJ3ncQ0S5E5HsAEdFvANwPYC8iWk1E/+T99A0ieoKIHgdwHICLmmyPhYWFhUVGJKqG4iCEeBXAmzXX1wA4jX0/x/D8+5up38LCwsKiedjIYgsLC4sOhyUEFhYWFh0OSwgsLCwsOhyWEFhYWFh0OCwhsLCwsOhwWEJgYWFh0eGwhMDCwsKiw9FUHIGFBQBc/7Ej8eSazUPdDAsLiwZhCYFF05g3awLmzZow1M2wsLBoEFY1ZGFhYdHhsITAwsLCosNhCYGFhYVFh8MSAgsLC4sOhyUEFhYWFh0OSwgsLCwsOhyWEFhYWFh0OCwhsLCwsOhwkBBiqNuQGUS0HsALDT4+GcCGHJvTDrB97gzYPncGmunzbkKIKerFtiQEzYCIFgsh5g91OwYTts+dAdvnzkAr+mxVQxYWFhYdDksILCwsLDocnUgIrhzqBgwBbJ87A7bPnYHc+9xxNgILCwsLizA6USKwsLCwsGCwhMDCwsKiw9FRhICITiGipUS0nIgWDHV78gIRXU1E64hoCbs2iYhuI6Jl3v+J7LfPemOwlIhOHppWNw4imkVEdxLR00T0JBH9u3d9JPe5h4geIqLHvD5/0bs+YvssQUQFInqEiG7yvo/oPhPRSiJ6gogeJaLF3rXW9lkI0RF/AAoAngOwO4AuAI8B2Heo25VT344GcDCAJezaNwAs8D4vAPB17/O+Xt+7AczxxqQw1H3I2N/pAA72Po8F8KzXr5HcZwIwxvtcAvAggMNHcp9Z3z8B4BoAN3nfR3SfAawEMFm51tI+d5JEcCiA5UKIFUKIMoBrAZwxxG3KBUKIuwFsVC6fAeDn3uefAziTXb9WCDEghHgewHK4Y9M2EEKsFUL8w/u8FcDTAGZgZPdZCCG2eV9L3p/ACO4zABDRTACnA/gxuzyi+2xAS/vcSYRgBoBV7Ptq79pIxTQhxFrA3TgBTPWuj6hxIKLZAF4Pl0Me0X32VCSPAlgH4DYhxIjvM4D/BvD/ANTZtZHeZwHgL0T0MBGd711raZ876fB60lzrRN/ZETMORDQGwB8A/IcQYguRrmvurZprbddnIUQNwDwimgDgOiLaP+b2tu8zEb0FwDohxMNEdGyaRzTX2qrPHo4UQqwhoqkAbiOiZ2LuzaXPnSQRrAYwi32fCWDNELVlMPAKEU0HAO//Ou/6iBgHIirBJQK/FkL80bs8ovssIYTYBOAuAKdgZPf5SABvI6KVcFW5xxPRrzCy+wwhxBrv/zoA18FV9bS0z51ECBYBmEtEc4ioC8DZAG4c4ja1EjcCOM/7fB6AG9j1s4mom4jmAJgL4KEhaF/DIJf1/wmAp4UQ32Y/jeQ+T/EkARDRKAAnAHgGI7jPQojPCiFmCiFmw12vfxVCvA8juM9ENJqIxsrPAE4CsASt7vNQW8gH2Rp/GlwPk+cAXDzU7cmxX78BsBZABS6H8E8AdgJwB4Bl3v9J7P6LvTFYCuDUoW5/A/19E1zx93EAj3p/p43wPh8I4BGvz0sAfN67PmL7rPT/WAReQyO2z3C9Gh/z/p6U+1Sr+2xTTFhYWFh0ODpJNWRhYWFhoYElBBYWFhYdDksILCwsLDoclhBYWFhYdDgsIbCwsLDocFhCYGFhYdHhsITAwsLCosPx/wHjynVc2HlY7AAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "model=IsolationForest(n_estimators=50, max_samples='auto', contamination=float(0.1),max_features=1.0)\n",
    "model.fit(Xc1)\n",
    "scores=model.decision_function(Xc1)\n",
    "plt.plot(scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "b524e608",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "sns.boxplot(x=scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "94dbf4a7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "49    -0.083432\n",
      "58    -0.085257\n",
      "60    -0.074452\n",
      "81    -0.090891\n",
      "228   -0.170106\n",
      "247   -0.098278\n",
      "371   -0.099601\n",
      "426   -0.070485\n",
      "453   -0.066546\n",
      "487   -0.118056\n",
      "494   -0.070990\n",
      "519   -0.056673\n",
      "522   -0.072073\n",
      "622   -0.090246\n",
      "673   -0.074576\n",
      "684   -0.069264\n",
      "744   -0.091888\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "scores=pd.Series(scores,index=Xc1.index)\n",
    "outs3= scores[scores < -0.05]\n",
    "print(outs3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3f31d107",
   "metadata": {},
   "source": [
    "### 5. Outlier detection using Kmeans"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "20970064",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.cluster import KMeans\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.metrics.pairwise import pairwise_distances\n",
    "from sklearn.neighbors import DistanceMetric"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "83c1172c",
   "metadata": {},
   "outputs": [],
   "source": [
    "scaler = StandardScaler()\n",
    "scaler.fit(Xc1)\n",
    "Xd= scaler.transform(Xc1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "4ef46318",
   "metadata": {},
   "outputs": [],
   "source": [
    "kmeans = KMeans(n_clusters=30, random_state=0).fit(Xd)\n",
    "clustlabels=kmeans.labels_\n",
    "#print(clustlabels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "ef116077",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Size of the formed clusters\n",
      "[[ 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20 21 22 23\n",
      "  24 25 26 27 28 29]\n",
      " [ 3 28 36 36 14  8 12 11  9 14 12  6 11 41  9 11  1 19 23 19 33 36 13  2\n",
      "  10 11 16  2 32 22]]\n"
     ]
    }
   ],
   "source": [
    "#Finding the size of the two clusters formed\n",
    "unique_elements, counts_elements = np.unique(clustlabels, return_counts=True)\n",
    "print(\"Size of the formed clusters\")\n",
    "print(np.asarray((unique_elements, counts_elements)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "3970a3a9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 75 182 342  49  60  81 426 494 522 228 336 453  58 622]\n"
     ]
    }
   ],
   "source": [
    "inst1=Xc1.index\n",
    "out11=inst1[np.where(clustlabels==0)]\n",
    "out12=inst1[np.where(clustlabels==11)]\n",
    "out13=inst1[np.where(clustlabels==16)]\n",
    "out14=inst1[np.where(clustlabels==23)]\n",
    "out15=inst1[np.where(clustlabels==27)]\n",
    "out1all=np.concatenate([out11,out12,out13,out14,out15])\n",
    "print(out1all)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d19922b6",
   "metadata": {},
   "source": [
    "### 6. Outlier detection using DBSCAN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "0eac7d1b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x241ce4951f0>]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "#Choosing the optimal epsilon\n",
    "k = 10\n",
    "# importing NearestNeighbors class \n",
    "from sklearn.neighbors import NearestNeighbors\n",
    "# initialize model\n",
    "neigh10 = NearestNeighbors(n_neighbors=10, n_jobs=-1)\n",
    "# train for getting nearest neighbour\n",
    "neigh10.fit(Xc1)\n",
    "dist10, ind10 = neigh10.kneighbors(Xc1)\n",
    "distances10 = [dist10[i][k-1] for i in range(len(dist10))]\n",
    "distances10 = np.sort(distances10, axis=0)\n",
    "plt.plot(distances10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "22d7dcdb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Int64Index([75, 153, 182, 228, 247, 248, 258, 286, 342, 392, 486, 519, 645,\n",
      "            710],\n",
      "           dtype='int64')\n"
     ]
    }
   ],
   "source": [
    "from sklearn.cluster import DBSCAN\n",
    "eps=50\n",
    "dbscan = DBSCAN(eps,min_samples=10).fit(Xc1)\n",
    "dbscanlabels=dbscan.labels_\n",
    "#print(dbscanlabels)\n",
    "inst=Xc1.index\n",
    "outd1=inst[np.where(dbscanlabels==-1)]\n",
    "print(outd1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c79c0d2a",
   "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}